1 2022-12-20 Patrick Palka <ppalka@redhat.com>
4 * tree.cc (walk_tree_1) <case TREE_VEC>: Walk the elements
5 in forward instead of reverse order.
6 <case VECTOR_CST>: Likewise.
8 2022-12-20 Andrew MacLeod <amacleod@redhat.com>
10 PR tree-optimization/108139
11 * gimple-range-cache.cc (ranger_cache::fill_block_cache): Do not
12 use equivalences originating from PHIS.
14 2022-12-20 Richard Biener <rguenther@suse.de>
17 * doc/install.texi (GDC): Document GDC 9.4 or later is required
18 to build the D language frontend.
20 2022-12-20 Jiufu Guo <guojiufu@linux.ibm.com>
23 * config/rs6000/rs6000-protos.h (can_be_rotated_to_lowbits): New.
24 (can_be_rotated_to_positive_16bits): New.
25 (can_be_rotated_to_negative_15bits): New.
26 * config/rs6000/rs6000.cc (can_be_rotated_to_lowbits): New definition.
27 (can_be_rotated_to_positive_16bits): New definition.
28 (can_be_rotated_to_negative_15bits): New definition.
29 * config/rs6000/rs6000.md (*rotate_on_cmpdi): New define_insn_and_split.
32 2022-12-19 Jason Merrill <jason@redhat.com>
34 * sort.cc: Disable -Wconditionally-supported in
37 2022-12-19 Jason Merrill <jason@redhat.com>
40 * configure.ac (strict_warn): Add -Wconditionally-supported.
41 * configure: Regenerate.
43 2022-12-19 Richard Biener <rguenther@suse.de>
45 PR tree-optimization/108164
46 * tree-vect-loop-manip.cc (vect_update_ivs_after_vectorizer):
47 Perform vect_step_op_add update in the appropriate type.
49 2022-12-19 Richard Earnshaw <rearnsha@arm.com>
51 * config/arm/arm-c.cc (__ARM_FEATURE_CLZ): Fix definition of
52 preprocessor macro when target has CLZ in another ISA.
54 2022-12-19 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
56 * config/riscv/vector.md: Remove vste.
58 2022-12-19 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
60 * config.gcc: Add riscv-vsetvl.o.
61 * config/riscv/riscv-passes.def (INSERT_PASS_BEFORE): Add VSETVL PASS
63 * config/riscv/riscv-protos.h (make_pass_vsetvl): New function.
64 (enum avl_type): New enum.
65 (get_ta): New function.
67 (get_avl_type): Ditto.
68 (calculate_ratio): Ditto.
69 (enum tail_policy): New enum.
70 (enum mask_policy): Ditto.
71 * config/riscv/riscv-v.cc (calculate_ratio): New function.
72 (emit_pred_op): change the VLMAX mov codgen.
73 (get_ta): New function.
75 (enum tail_policy): Change enum.
76 (get_prefer_tail_policy): New function.
77 (enum mask_policy): Change enum.
78 (get_prefer_mask_policy): New function.
79 * config/riscv/t-riscv: Add riscv-vsetvl.o
80 * config/riscv/vector.md: Adjust attribute and pattern for VSETVL
82 (@vlmax_avl<mode>): Ditto.
83 (@vsetvl<mode>_no_side_effects): Delete.
84 (vsetvl_vtype_change_only): New MD pattern.
85 (@vsetvl_discard_result<mode>): Ditto.
86 * config/riscv/riscv-vsetvl.cc: New file.
87 * config/riscv/riscv-vsetvl.h: New file.
89 2022-12-19 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
91 * config/riscv/riscv-vector-switch.def (ENTRY): Correct attributes.
93 2022-12-19 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
96 * config/aarch64/aarch64-builtins.cc
97 (aarch64_expand_builtin_data_intrinsic): Handle NULL target.
99 2022-12-19 Jakub Jelinek <jakub@redhat.com>
101 * config/gnu-user.h (LIBHWASAN_EARLY_SPEC): Add libhwasan_preinit.o
102 to link spec if not -shared.
104 2022-12-19 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
106 * config/riscv/riscv-v.cc (emit_vlmax_vsetvl): Pass through VLMUL enum
107 instead of machine mode.
108 * config/riscv/riscv-vector-builtins-bases.cc: Ditto.
109 * config/riscv/riscv.cc (riscv_print_operand): Print LMUL by enum vlmul
110 instead of machine mode.
112 2022-12-19 Jiufu Guo <guojiufu@linux.ibm.com>
115 * config/rs6000/rs6000.cc (rs6000_emit_set_long_const): Add using
116 "li; x?oris" to build constant.
118 2022-12-19 liuhongt <hongtao.liu@intel.com>
121 * config/i386/gnu-user-common.h (GNU_USER_TARGET_MATHFILE_SPEC):
122 Don't add crtfastmath.o for -shared.
123 * doc/invoke.texi (-shared): Add related documentation.
125 2022-12-18 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
127 * config.gcc: Determine Solaris minor version.
128 Obsolete *-*-solaris2.11.[0-3]*.
129 * doc/install.texi (Specific, *-*-solaris2*): Document it.
131 2022-12-17 Segher Boessenkool <segher@kernel.crashing.org>
133 * config/rs6000/rs6000-logue.cc (rs6000_output_function_epilogue):
134 Handle GNU Rust for the tbtab lang field.
136 2022-12-16 John David Anglin <danglin@gcc.gnu.org>
138 * config/pa/pa.cc (pa_option_override): Disable -fstack-protector.
140 2022-12-16 Vladimir N. Makarov <vmakarov@redhat.com>
143 2022-12-15 Vladimir N. Makarov <vmakarov@redhat.com>
145 * ira-costs.cc: Include print-rtl.h.
146 (record_reg_classes, scan_one_insn): Add code to print debug info.
147 * ira.cc (ira_init_register_move_cost): Check that at least one hard
148 reg of the mode are in the class contents to calculate the
151 2022-12-16 Qing Zhao <qing.zhao@oracle.com>
153 * doc/invoke.texi: Document -Wstrict-flex-arrays option.
154 * gimple-array-bounds.cc (check_out_of_bounds_and_warn): Add two more
156 (array_bounds_checker::check_array_ref): Issue warnings for
157 -Wstrict-flex-arrays.
158 * opts.cc (finish_options): Issue warning for unsupported combination
159 of -Wstrict_flex_arrays and -fstrict-flex-array.
160 * tree-vrp.cc (execute_ranger_vrp): Enable the pass when
161 warn_strict_flex_array is true.
163 2022-12-16 Palmer Dabbelt <palmer@rivosinc.com>
165 * config/riscv/riscv.cc (riscv_option_override): Fix comment
168 2022-12-16 Palmer Dabbelt <palmer@rivosinc.com>
170 * doc/extend.texi (__builtin_riscv_pause): Imply
171 Xgnuzihintpausestate.
173 2022-12-16 Richard Biener <rguenther@suse.de>
176 * tree-inline.cc (remap_ssa_name): Do not unshare the
177 result from the decl_map.
179 2022-12-16 Richard Biener <rguenther@suse.de>
182 * tree-inline.cc (copy_bb): Remove handling of (foo *)&this->m
183 substitution which is done in remap_gimple_op_r via
186 2022-12-16 Jakub Jelinek <jakub@redhat.com>
188 PR rtl-optimization/106751
189 * loop-invariant.cc (move_invariant_reg): If preheader bb ends
190 with a JUMP_INSN, split the preheader edge and emit invariants
191 into the new preheader basic block.
193 2022-12-16 Richard Biener <rguenther@suse.de>
196 * tree-inline.cc (remap_gimple_stmt): Add stmts to the
197 sequence without updating them. Simplify x == x detection.
199 2022-12-16 Haochen Jiang <haochen.jiang@intel.com>
201 * config/i386/cmpccxaddintrin.h
202 (__cmpccxadd_epi32): Rename to _cmpccxadd_epi32.
203 (__cmpccxadd_epi64): Rename to _cmpccxadd_epi64.
205 2022-12-15 Vladimir N. Makarov <vmakarov@redhat.com>
207 * ira-costs.cc: Include print-rtl.h.
208 (record_reg_classes, scan_one_insn): Add code to print debug info.
209 * ira.cc (ira_init_register_move_cost): Check that at least one hard
210 reg of the mode are in the class contents to calculate the
213 2022-12-15 Siddhesh Poyarekar <siddhesh@gotplt.org>
215 * doc/extend.texi (__builtin_dynamic_object_size): Document
218 (Optimize calls to @code{__builtin_object_size}): Also mention
219 __builtin_dynamic_object_size.
221 2022-12-15 Siddhesh Poyarekar <siddhesh@gotplt.org>
224 * doc/invoke.texi (-fsanitize=object-size): Use
225 __builtin_dynamic_object_size instead of
226 __builtin_object_size.
228 2022-12-15 Richard Biener <rguenther@suse.de>
231 * tree-inline.cc (copy_edges_for_bb): Walk stmts backwards for
232 splitting the block to avoid quadratic behavior with setting
233 stmts BB on multliple splits.
235 2022-12-15 Spacetown <michael.foerderer@gmx.de>
237 PR gcov-profile/107537
238 * gcov.cc (output_branch_count): Add annotation '(fallthrough)'
239 or '(throw)' also to uncovered branches.
241 2022-12-15 Richard Biener <rguenther@suse.de>
244 * tree-inline.cc (copy_edges_for_bb): Do not update all
247 2022-12-15 Jakub Jelinek <jakub@redhat.com>
249 PR tree-optimization/108095
250 * tree-into-ssa.cc (maybe_register_def): Insert debug stmt
251 on all non-EH edges from asm goto if they have a single
252 predecessor rather than asserting there is at most one such edge.
253 Test whether there are no PHI nodes next to the single predecessor
256 2022-12-14 David Faust <david.faust@oracle.com>
259 * btfout.cc (get_section_name): New function.
260 (btf_collect_datasec): Use it here. Process functions, marking them
261 'extern' and generating DATASEC entries for them as appropriate. Move
262 creation of BTF_KIND_FUNC records to here...
263 (btf_dtd_emit_preprocess_cb): ... from here.
265 2022-12-14 David Faust <david.faust@oracle.com>
268 * btfout.cc (btf_collect_datasec): Correct size of void entries.
269 (btf_dvd_emit_preprocess_cb): Do not skip emitting variables which
271 (btf_init_postprocess): Create 'const void' type record if needed and
272 adjust variables to refer to it as appropriate.
274 2022-12-14 David Faust <david.faust@oracle.com>
277 * btfout.cc (btf_collect_datasec): Mark extern variables as such.
278 (btf_dvd_emit_preprocess_cb): Skip non-defining extern variable decl
279 if there is a defining decl for the same variable.
280 (btf_asm_varent): Accomodate 'extern' linkage.
282 2022-12-14 Martin Jambor <mjambor@suse.cz>
284 * ipa-sra.cc (create_parameter_descriptors): Consider the first
285 parameter of a method safe to dereference.
287 2022-12-14 Martin Jambor <mjambor@suse.cz>
289 * ipa-sra.cc (loaded_decls): Adjust comment.
290 (scan_expr_access): Also detect assignments of address of local
291 variables to a variable. Bail out early on SSA_NAMEs and
292 constants as an optimization.
294 2022-12-14 Gaius Mulley <gaiusmod2@gmail.com>
296 * configure.ac (HAVE_PYTHON): Test for Python3 added.
297 * dwarf2out.cc (gen_compile_unit_die): Check language_string
298 and language to DW_LANG_Modula2.
299 * doc/install.texi: Add m2 as a language.
300 (--disable-libgm2): Documented. Add make check-m2 to testing selected
302 * doc/sourcebuild.texi (GM2): New prerequisite item.
303 (Python3 modules) New item.
306 * configure: Rebuilt.
307 * config.in: Rebuilt.
308 * doc/gm2.texi: New file.
310 2022-12-14 Martin Jambor <mjambor@suse.cz>
312 * doc/invoke.texi (ipa-sra-ptrwrap-growth-factor): Fix the
314 * params.opt (ipa-sra-ptrwrap-growth-factor): Likewise.
316 2022-12-14 Tamar Christina <tamar.christina@arm.com>
319 * config/aarch64/aarch64.cc
320 (aarch64_vectorize_can_special_div_by_constant): Ensure input and output
323 2022-12-14 Martin Liska <mliska@suse.cz>
325 * doc/invoke.texi: Document ipa-sra-ptrwrap-growth-factor.
327 2022-12-14 Richard Biener <rguenther@suse.de>
329 PR tree-optimization/107617
330 * tree-ssa-sccvn.cc (vn_walk_cb_data::push_partial_def):
331 Handle negative pd.rhs_off.
332 (vn_reference_lookup_3): Properly provide pd.rhs_off
333 for .LEN_STORE on big-endian targets.
335 2022-12-13 Martin Jambor <mjambor@suse.cz>
337 * ipa-param-manipulation.cc
338 (ipa_param_body_adjustments::modify_expression): Bail out early if
339 there are no replacements.
340 (ipa_param_body_adjustments::modify_assignment): Likewise.
342 2022-12-13 Martin Jambor <mjambor@suse.cz>
344 * ipa-sra.cc (isra_param_desc): New fields safe_size,
345 conditionally_dereferenceable and safe_size_set.
346 (struct gensum_param_desc): New field conditionally_dereferenceable.
347 (struct isra_param_flow): Updated comment of field unit_size.
348 (ipa_sra_function_summaries::duplicate): Copy the new fields.
349 (isra_call_summary::dump): Dump unit_size when representing safe_size.
350 (dump_gensum_param_descriptor): Dump new flag.
351 (dump_isra_param_descriptor): Dump new fields.
352 (isra_analyze_call): Fill unit_size when it represents known safe
354 (check_gensum_access): Instead of disqualifying pointers which are not
355 always dereference, mark them as conditionally dereferencable if loads
357 (process_scan_results): Copy the conditionally_dereferenceable flag.
358 (isra_write_node_summary): Stream new fields, or assert they are not
360 (isra_read_node_info): Stream new fields.
361 (update_safe_size): New function.
362 (propagate_param_hints_accross_call): Propagate safe_sizes.
363 (propagate_hints_to_all_callees): New function.
364 (adjust_parameter_descriptions): Check conditionally_dereferenceable
365 candidates, rework dumping.
366 (ipa_sra_analysis): Move most of hint propagation for one node to
367 propagate_hints_to_all_callees. Add another loop to stabilize within
368 SCCs and another one to verify.
370 2022-12-13 Martin Jambor <mjambor@suse.cz>
372 * ipa-sra.cc (isra_param_desc): New field not_specially_constructed.
373 (struct isra_param_flow): New field constructed_for_calls.
374 (isra_call_summary::dump): Dump the new flag.
375 (loaded_decls): New variable.
376 (dump_isra_param_descriptor): New parameter hints, dump
377 not_specially_constructed if it is true.
378 (dump_isra_param_descriptors): New parameter hints, pass it to
379 dump_isra_param_descriptor.
380 (ipa_sra_function_summaries::duplicate): Duplicate new flag.
381 (create_parameter_descriptors): Adjust comment.
382 (get_gensum_param_desc): Bail out when decl2desc is NULL.
383 (scan_expr_access): Add loaded local variables to loaded_decls.
384 (scan_function): Survive if final_bbs is NULL.
385 (isra_analyze_call): Compute constructed_for_calls flag.
386 (process_scan_results): Be optimistic about size limits. Do not dump
387 computed param hints when dumpint IPA-SRA structures.
388 (isra_write_edge_summary): Stream constructed_for_calls.
389 (isra_read_edge_summary): Likewise.
390 (ipa_sra_dump_all_summaries): New parameter hints, pass it to
391 dump_isra_param_descriptor.
392 (flip_all_hints_pessimistic): New function.
393 (flip_all_param_hints_pessimistic): Likewise.
394 (propagate_param_hints): Likewise.
395 (disable_unavailable_parameters): Renamed to
396 adjust_parameter_descriptions. Expand size limits for parameters
397 which are specially contstructed by all callers. Check limits again.p
398 (ipa_sra_analysis): Pass required hints to ipa_sra_dump_all_summaries.
399 Add hint propagation.
400 (ipa_sra_summarize_function): Initialize and destory loaded_decls,
401 rearrange so that scan_function is called even when there are no
403 * params.opt (ipa-sra-ptrwrap-growth-factor): New parameter.
405 2022-12-13 Martin Jambor <mjambor@suse.cz>
407 * ipa-sra.cc (ipa_sra_analysis): Move top-down analysis before
408 bottom-up analysis. Replace FOR_EACH_VEC_ELT with C++11 iteration.
410 2022-12-13 Martin Jambor <mjambor@suse.cz>
413 * params.opt (ipa-sra-deref-prob-threshold): New parameter.
414 * doc/invoke.texi (ipa-sra-deref-prob-threshold): Document it.
415 * ipa-sra.cc (struct gensum_param_access): New field load_count.
416 (struct gensum_param_desc): New field safe_ref, adjusted comments.
417 (by_ref_count): Renamed to unsafe_by_ref_count, adjusted all uses.
418 (dump_gensum_access): Dump the new field.
419 (dump_gensum_param_descriptor): Likewise.
420 (create_parameter_descriptors): Set safe_ref field, move setting
421 by_ref forward. Only increment unsafe_by_ref_count for unsafe
423 (allocate_access): Initialize new field.
424 (mark_param_dereference): Adjust indentation. Only add data to
425 bb_dereferences for unsafe by_ref parameters.
426 (scan_expr_access): For loads, accumulate BB counts.
427 (dereference_probable_p): New function.
428 (check_gensum_access): Fix leading comment, add parameter FUN.
429 Check cumulative counts of loads for safe by_ref accesses instead
431 (process_scan_results): Do not propagate dereference distances for
432 safe by_ref parameters. Pass fun to check_gensum_access. Safe
433 by_ref params do not need the postdominance check.
435 2022-12-13 Martin Jambor <mjambor@suse.cz>
437 * ipa-cp.cc (clone_for_param_removal_p): New function.
438 (estimate_local_effects): Call it before considering cloning
439 just to remove unused parameters.
441 2022-12-13 Martin Jambor <mjambor@suse.cz>
444 * ipa-param-manipulation.h (class ipa_param_adjustments): Removed
445 member function get_updated_index_or_split.
446 (class ipa_param_body_adjustments): New overload of
447 register_replacement, new member function append_init_stmts, new
448 member m_split_agg_csts_inits.
449 * ipa-param-manipulation.cc: Include ipa-prop.h.
450 (ipa_param_adjustments::get_updated_index_or_split): Removed.
451 (ipa_param_body_adjustments::register_replacement): New overload, use
452 it from the older one.
453 (ipa_param_body_adjustments::common_initialization): Added the
454 capability to create replacements for conflicting IPA-CP discovered
456 (ipa_param_body_adjustments::ipa_param_body_adjustments): Construct
458 (ipa_param_body_adjustments::append_init_stmts): New function.
459 * ipa-sra.cc: Include ipa-prop.h.
460 (push_param_adjustments_for_index): Require IPA-CP transformation
461 summary as a parameter, do not create replacements which are known to
462 have constant values.
463 (process_isra_node_results): Find and pass to the above function the
464 IPA-CP transformation summary.
465 * ipa-prop.cc (adjust_agg_replacement_values): Remove the
466 functionality replacing IPA-SRA created scalar parameters with
467 constants. Simplify, do not require parameter descriptors, do not
469 (ipcp_transform_function): Simplify now that
470 adjust_agg_replacement_values does not change cfg. Move definition
471 and initialization of descriptors lower.
472 * tree-inline.cc (tree_function_versioning): Call append_init_stmts of
473 param_body_adjs, if there are any.
475 2022-12-13 Martin Jambor <mjambor@suse.cz>
477 * ipa-prop.cc (useful_ipcp_transformation_info_p): New function.
478 (write_ipcp_transformation_info): Added a parameter, simplified
479 given that is known not to be NULL.
480 (ipcp_write_transformation_summaries): Write out all useful
481 transformation summaries.
482 (read_ipcp_transformation_info): Simplify given that some info
484 (read_replacements_section): Remove assert.
485 * lto-cgraph.cc (add_node_to): Also set encode_body for clones.
486 * lto-streamer-out.cc (lto_output): Do not output virtual clones.
488 2022-12-13 Jakub Jelinek <jakub@redhat.com>
491 * config/i386/i386.md (*concat<half><mode>3_5, *concat<mode><dwi>3_6,
492 *concat<mode><dwi>3_7): Split alternative with =ro output constraint
493 into =r,o,o and use Wd input constraint for the last alternative which
494 is enabled for TARGET_64BIT. Reject ix86_endbr_immediate_operand
495 in the input constant.
497 2022-12-13 Tamar Christina <tamar.christina@arm.com>
499 * config/aarch64/aarch64.md (tbranch_<code><mode>3): Use gen_int_mode.
501 2022-12-13 Jakub Jelinek <jakub@redhat.com>
503 PR tree-optimization/108064
504 * tree-vect-patterns.cc (vect_recog_rotate_pattern): Pass uvectype
505 as 4th argument to append_pattern_def_seq for statements with lhs
508 2022-12-13 Richard Biener <rguenther@suse.de>
510 PR tree-optimization/105801
511 * tree-ssa-ccp.cc (likely_value): .DEFERRED_INIT produces
513 * doc/invoke.texi (ftrivial-auto-var-init): Explicitely
514 mention we treat variables without an initializer as
515 undefined also for optimization purposes.
517 2022-12-13 Tom Tromey <tom@tromey.com>
518 Mark Wielaard <mark@klomp.org>
519 Marc Poulhiès <dkm@kataplop.net>
521 * dwarf2out.cc (is_rust): New.
522 (base_type_die): Use DW_ATE_UTF for the Rust 'char' type.
523 (gen_compile_unit_die): Handle "GNU Rust".
525 2022-12-13 Richard Biener <rguenther@suse.de>
527 PR tree-optimization/108076
528 * tree-if-conv.cc (if_convertible_loop_p_1): Reject blocks
529 with non-local or forced labels that we later remove
532 2022-12-13 Haochen Gui <guihaoc@gcc.gnu.org>
534 * config/rs6000/rs6000.md (cbranchcc4): New expander.
536 2022-12-13 Haochen Gui <guihaoc@gcc.gnu.org>
538 * optabs.cc (prepare_cmp_insn): Return a NULL rtx other than
539 assertion failure when targets don't have cbranch optab or
540 predicate check fails.
542 2022-12-12 Wilco Dijkstra <wilco.dijkstra@arm.com>
544 * config/aarch64/aarch64.cc (aarch64_rtx_costs): Add correct costs
545 for 24-bit and 12-bit shifted immediate add/sub.
546 (TARGET_CONST_ANCHOR): Define.
547 * config/aarch64/predicates.md (aarch64_pluslong_immediate):
550 2022-12-12 Tamar Christina <tamar.christina@arm.com>
552 * match.pd: Add new rule.
554 2022-12-12 Tamar Christina <tamar.christina@arm.com>
556 * config/aarch64/aarch64.cc (aarch64_can_change_mode_class): Restrict
557 conversions between partial struct types properly.
559 2022-12-12 Tamar Christina <tamar.christina@arm.com>
561 * config/aarch64/aarch64.md (*tb<optab><mode>1): Rename to...
562 (*tb<optab><ALLI:mode><GPI:mode>1): ... this.
563 (tbranch_<code><mode>4): New.
564 * config/aarch64/iterators.md(ZEROM, zerom): New.
566 2022-12-12 Tamar Christina <tamar.christina@arm.com>
568 * dojump.cc (do_jump): Pass along value.
569 (do_jump_by_parts_greater_rtx): Likewise.
570 (do_jump_by_parts_zero_rtx): Likewise.
571 (do_jump_by_parts_equality_rtx): Likewise.
572 (do_compare_rtx_and_jump): Likewise.
573 (do_compare_and_jump): Likewise.
574 * dojump.h (do_compare_rtx_and_jump): New.
575 * optabs.cc (emit_cmp_and_jump_insn_1): Refactor to take optab to check.
576 (validate_test_and_branch): New.
577 (emit_cmp_and_jump_insns): Optiobally take a value, and when value is
578 supplied then check if it's suitable for tbranch.
579 * optabs.def (tbranch_eq$a4, tbranch_ne$a4): New.
580 * doc/md.texi (tbranch_@var{op}@var{mode}4): Document it.
581 * optabs.h (emit_cmp_and_jump_insns): New.
582 * tree.h (tree_zero_one_valued_p): New.
584 2022-12-12 Tamar Christina <tamar.christina@arm.com>
586 * config/aarch64/aarch64-simd.md (*aarch64_simd_movv2hf): New.
587 (mov<mode>, movmisalign<mode>, aarch64_dup_lane<mode>,
588 aarch64_store_lane0<mode>, aarch64_simd_vec_set<mode>,
589 @aarch64_simd_vec_copy_lane<mode>, vec_set<mode>,
590 reduc_<optab>_scal_<mode>, reduc_<fmaxmin>_scal_<mode>,
591 aarch64_reduc_<optab>_internal<mode>, aarch64_get_lane<mode>,
592 vec_init<mode><Vel>, vec_extract<mode><Vel>): Support V2HF.
593 (aarch64_simd_dupv2hf): New.
594 * config/aarch64/aarch64.cc (aarch64_classify_vector_mode):
596 * config/aarch64/iterators.md (VHSDF_P): New.
597 (V2F, VMOVE, nunits, Vtype, Vmtype, Vetype, stype, VEL,
598 Vel, q, vp): Add V2HF.
599 * config/arm/types.md (neon_fp_reduc_add_h): New.
601 2022-12-12 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
603 * config/aarch64/aarch64-c.cc (aarch64_update_cpp_builtins): Define
604 __ARM_FEATURE_PAUTH and __ARM_FEATURE_BTI when appropriate.
605 * config/aarch64/aarch64.h (TARGET_BTI): Define.
607 2022-12-12 Richard Biener <rguenther@suse.de>
609 * genmatch.cc (dt_simplify::gen): Revert last change.
610 * match.pd: Revert simplification of CONSTUCTOR leaf handling.
611 (&x cmp SSA_NAME): Handle ADDR_EXPR in SSA defs.
612 * fold-const.cc (split_address_to_core_and_offset): Handle
613 ADDR_EXPRs in SSA defs.
614 (address_compare): Likewise.
616 2022-12-12 Richard Biener <rguenther@suse.de>
618 PR tree-optimization/89317
619 * tree-ssa-ccp.cc (ccp_fold): Handle GIMPLE_COND via
620 gimple_fold_stmt_to_constant_1.
621 * match.pd (&a != &a + c): Apply to pointer_plus with non-ADDR_EXPR
624 2022-12-11 Richard Biener <rguenther@suse.de>
626 PR tree-optimization/89317
627 * match.pd ((p + b) - &p->c -> b - offsetof(c)): New patterns.
629 2022-12-11 Richard Biener <rguenther@suse.de>
631 * genmatch.cc (dt_node::gen_kids): Handle ADDR_EXPR in both
632 the GENERIC and GIMPLE op position.
633 (dt_simplify::gen): Capture both GENERIC and GIMPLE op
634 position for ADDR_EXPR and CONSTRUCTOR.
635 * match.pd: Simplify CONSTRUCTOR leaf handling.
637 2022-12-11 Richard Biener <rguenther@suse.de>
639 PR tree-optimization/106904
640 * tree.h (strip_zero_offset_components): Declare.
641 * tree.cc (strip_zero_offset_components): Define.
642 * tree-vect-data-refs.cc (vect_create_addr_base_for_vector_ref):
643 Strip zero offset components before building the address.
645 2022-12-10 Jakub Jelinek <jakub@redhat.com>
647 PR tree-optimization/107997
648 * tree-ssa-loop-ivopts.cc: Include cfganal.h.
649 (create_new_iv) <case IP_END>: If ip_end_pos bb is non-empty and ends
650 with a stmt which ends bb, instead of adding iv update after it split
651 the latch edge and insert iterator into the new latch bb.
653 2022-12-09 Jiufu Guo <guojiufu@linux.ibm.com>
655 * config/rs6000/rs6000.cc (rs6000_emit_set_const): Remove copy_rtx.
656 (rs6000_emit_set_long_const): Likewise.
658 2022-12-09 Martin Liska <mliska@suse.cz>
660 * config/i386/i386-builtins.cc (fold_builtin_cpu): Use same path
662 * doc/extend.texi: Fix "x86-64" use.
664 2022-12-09 Sebastian Huber <sebastian.huber@embedded-brains.de>
666 * config/rtems.h (SUBTARGET_CC1_SPEC): Rename to...
667 (OS_CC1_SPEC): ...this.
668 * gcc.cc (SUBTARGET_CC1_SPEC): Rename to...
669 (OS_CC1_SPEC): ...this.
671 2022-12-09 David Malcolm <dmalcolm@redhat.com>
673 * Makefile.in (ANALYZER_OBJS): Update for renaming of
674 analyzer/region-model-impl-calls.cc to analyzer/kf.cc.
676 2022-12-09 liuhongt <hongtao.liu@intel.com>
678 * doc/invoke.texi (x86 options): Document
679 -mlam={none,u48,u57}.
680 * config/i386/i386-opts.h (enum lam_type): New enum.
681 * config/i386/i386.cc (ix86_memtag_can_tag_addresses): New.
682 (ix86_memtag_set_tag): Ditto.
683 (ix86_memtag_extract_tag): Ditto.
684 (ix86_memtag_add_tag): Ditto.
685 (ix86_memtag_tag_size): Ditto.
686 (ix86_memtag_untagged_pointer): Ditto.
687 (TARGET_MEMTAG_CAN_TAG_ADDRESSES): New.
688 (TARGET_MEMTAG_ADD_TAG): Ditto.
689 (TARGET_MEMTAG_SET_TAG): Ditto.
690 (TARGET_MEMTAG_EXTRACT_TAG): Ditto.
691 (TARGET_MEMTAG_UNTAGGED_POINTER): Ditto.
692 (TARGET_MEMTAG_TAG_SIZE): Ditto.
693 (IX86_HWASAN_SHIFT): Ditto.
694 (IX86_HWASAN_TAG_SIZE): Ditto.
695 * config/i386/i386-expand.cc (ix86_expand_call): Untag code
697 * config/i386/i386-options.cc (ix86_option_override_internal):
698 Error when enable -mlam=[u48|u57] for 32-bit code.
699 * config/i386/i386.opt: Add -mlam=[none|u48|u57].
700 * config/i386/i386-protos.h (ix86_memtag_untagged_pointer):
702 (ix86_memtag_can_tag_addresses): Ditto.
704 2022-12-08 Marek Polacek <polacek@redhat.com>
706 * doc/invoke.texi (-fsanitize=address): Suggest options to improve
709 2022-12-08 Eugene Rozenfeld <erozen@microsoft.com>
712 * ipa-cp.cc (ipcp_propagate_stage): Fix profile count comparison
714 2022-12-08 David Faust <david.faust@oracle.com>
716 * config/bpf/bpf.md (bswap<mode>2): New define_insn.
718 2022-12-08 Sebastian Pop <spop@amazon.com>
721 * config/aarch64/aarch64-protos.h (aarch64_output_patchable_area):
723 * config/aarch64/aarch64.cc (aarch64_print_patchable_function_entry):
724 Emit an UNSPECV_PATCHABLE_AREA pseudo instruction.
725 (aarch64_output_patchable_area): New.
726 * config/aarch64/aarch64.md (UNSPECV_PATCHABLE_AREA): New.
727 (patchable_area): Define.
729 2022-12-08 Jakub Jelinek <jakub@redhat.com>
732 * cfgbuild.cc (find_bb_boundaries): If there are NOTEs in between
733 debug_insn (seen after flow_transfer_insn) and insn, move NOTEs
734 before all the DEBUG_INSNs and split after NOTEs. If there are
735 other insns like jump table data, clear debug_insn.
737 2022-12-08 Jakub Jelinek <jakub@redhat.com>
740 * config/i386/i386.md (HALF, half): New mode attributes.
741 (*concat<half><mode>3_5, *concat<mode><dwi>3_6,
742 *concat<mode><dwi>3_7): New define_insn_and_split patterns.
744 2022-12-08 Andrew MacLeod <amacleod@redhat.com>
746 PR tree-optimization/107985
748 (gimple_range_op_handler::gimple_range_op_handler): Check if type
749 of the operands is supported.
750 * gimple-range.cc (gimple_ranger::prefill_stmt_dependencies): Do
751 not assert if here is no range-op handler.
753 2022-12-08 Jiufu Guo <guojiufu@linux.ibm.com>
755 * config/rs6000/predicates.md: Use sext_hwi.
756 * config/rs6000/rs6000.cc (num_insns_constant_gpr): Likewise.
757 (darwin_rs6000_legitimate_lo_sum_const_p): Likewise.
758 (mem_operand_gpr): Likewise.
759 (mem_operand_ds_form): Likewise.
760 (rs6000_legitimize_address): Likewise.
761 (rs6000_emit_set_const): Likewise.
762 (rs6000_emit_set_long_const): Likewise.
763 (print_operand): Likewise.
764 (constant_generates_xxspltiw): Remove unnecessary expressions.
765 * config/rs6000/rs6000.md: Use sext_hwi.
767 2022-12-08 Richard Biener <rguenther@suse.de>
769 PR tree-optimization/107699
770 * match.pd (&a !=/== &a.b + c -> (&a - &a.b) !=/== c): New
773 2022-12-08 Jakub Jelinek <jakub@redhat.com>
775 * range-op-float.cc (frange_nextafter): For MODE_COMPOSITE_P from
776 denormal or zero, use real_nextafter on DFmode with conversions
778 (frange_arithmetic): For mode_composite, on top of rounding in the
779 right direction accept extra 1ulp error for PLUS/MINUS_EXPR, extra
780 2ulps error for MULT_EXPR and extra 3ulps error for RDIV_EXPR.
782 2022-12-08 Jakub Jelinek <jakub@redhat.com>
784 PR tree-optimization/107967
785 * range-op-float.cc (frange_arithmetic): Fix a thinko - if
786 inf is negative, use nextafter if !real_less (&result, &value)
787 rather than if real_less (&result, &value). If result is +-INF
788 while value is finite and -fno-rounding-math, don't do rounding
789 if !inexact or if result is significantly above max representable
790 value or below min representable value.
792 2022-12-07 Max Filippov <jcmvbkbc@gmail.com>
794 * config.gcc (xtensa*-*-*): Add xtensa-dynconfig.o to extra_objs.
795 * config/xtensa/t-xtensa (TM_H): Add xtensa-dynconfig.h.
796 (xtensa-dynconfig.o): New rule.
797 * config/xtensa/xtensa-dynconfig.c: New file.
798 * config/xtensa/xtensa-protos.h (xtensa_get_config_strings): New
800 * config/xtensa/xtensa.h (xtensa-config.h): Replace #include
801 with xtensa-dynconfig.h
802 (XCHAL_HAVE_MUL32_HIGH, XCHAL_HAVE_RELEASE_SYNC)
803 (XCHAL_HAVE_S32C1I, XCHAL_HAVE_THREADPTR)
804 (XCHAL_HAVE_FP_POSTINC): Drop definitions.
805 (TARGET_DIV32): Replace with __XCHAL_HAVE_DIV32.
806 (TARGET_CPU_CPP_BUILTINS): Add new 'builtin' variable and loop
807 through string array returned by the xtensa_get_config_strings
810 2022-12-07 Wilco Dijkstra <wilco.dijkstra@arm.com>
813 * config/aarch64/aarch64.cc (aarch64_expand_sve_const_vector):
814 Fix call to aarch64_move_imm to use SI/DI.
816 2022-12-07 Richard Biener <rguenther@suse.de>
819 * ipa-pure-const.cc (ipa_make_function_pure): Skip also
820 for functions already being const.
822 2022-12-07 Hongyu Wang <hongyu.wang@intel.com>
824 * config/i386/x86-tune.def (X86_TUNE_AVOID_256FMA_CHAINS): Add
825 m_SAPPHIRERAPIDS, m_ALDERLAKE and m_CORE_ATOM.
827 2022-12-07 Prathamesh Kulkarni <prathamesh.kulkarni@linaro.org>
830 * config/aarch64/aarch64-sve-builtins-base.cc: Use
831 gsi_replace_with_seq_vops to handle virtual operands, and gate
832 the transform on !flag_non_call_exceptions.
833 * gimple-fold.cc (gsi_replace_with_seq_vops): Make function non static.
834 * gimple-fold.h (gsi_replace_with_seq_vops): Declare.
836 2022-12-07 liuhongt <hongtao.liu@intel.com>
839 * config/i386/mmx.md (btruncv2sf2): Add TARGET_MMX_WITH_SSE to
842 2022-12-06 Wilco Dijkstra <wilco.dijkstra@arm.com>
844 * config/aarch64/aarch64.cc (aarch64_bitmask_imm): Use unsigned type.
845 (aarch64_is_mov_xn_imm): New function.
846 (aarch64_move_imm): Refactor, assert mode is SImode or DImode.
847 (aarch64_internal_mov_immediate): Assert mode is SImode or DImode.
848 Simplify special cases.
849 (aarch64_uimm12_shift): Simplify code.
850 (aarch64_clamp_to_uimm12_shift): Likewise.
851 (aarch64_movw_imm): Rename to aarch64_is_movz.
852 (aarch64_float_const_rtx_p): Pass either SImode or DImode to
853 aarch64_internal_mov_immediate.
854 (aarch64_rtx_costs): Likewise.
855 * config/aarch64/aarch64.md (movdi_aarch64): Merge 'N' and 'M'
856 constraints into single 'O'.
857 (mov<mode>_aarch64): Likewise.
858 * config/aarch64/aarch64-protos.h (aarch64_move_imm): Use unsigned.
859 (aarch64_bitmask_imm): Likewise.
860 (aarch64_uimm12_shift): Likewise.
861 (aarch64_is_mov_xn_imm): New prototype.
862 * config/aarch64/constraints.md: Add 'O' for 32/64-bit immediates,
863 limit 'N' to 64-bit only moves.
865 2022-12-06 Qing Zhao <qing.zhao@oracle.com>
867 * attribs.cc (strict_flex_array_level_of): New function.
868 * attribs.h (strict_flex_array_level_of): Prototype for new function.
869 * doc/invoke.texi: Update -Warray-bounds by specifying the impact from
870 -fstrict-flex-arrays. Also update -Warray-bounds=2 by eliminating its
871 impact on treating trailing arrays as flexible array members.
872 * gimple-array-bounds.cc (get_up_bounds_for_array_ref): New function.
873 (check_out_of_bounds_and_warn): New function.
874 (array_bounds_checker::check_array_ref): Update with call to the above
876 * tree.cc (array_ref_flexible_size_p): Add one new argument.
877 (component_ref_sam_type): New function.
878 (component_ref_size): Control with level of strict-flex-array.
879 * tree.h (array_ref_flexible_size_p): Update prototype.
880 (enum struct special_array_member): Add two new enum values.
881 (component_ref_sam_type): New prototype.
883 2022-12-06 David Malcolm <dmalcolm@redhat.com>
885 * doc/analyzer.texi: Drop out-of-date ideas for other checkers.
887 2022-12-06 David Malcolm <dmalcolm@redhat.com>
889 * Makefile.in (ANALYZER_OBJS): Add analyzer/call-details.o,
890 analyzer/kf-analyzer.o, and kf-lang-cp.o.
892 2022-12-06 Marcel Vollweiler <marcel@codesourcery.com>
894 * gimplify.cc (optimize_target_teams): Set initial num_teams_upper
895 to "-2" instead of "1" for non-existing num_teams clause in order to
896 disambiguate from the case of an existing num_teams clause with value 1.
898 2022-12-06 Andre Vieira <andre.simoesdiasvieira@arm.com>
901 * config/arm/mve.md (mve_vcmp<mve_cmp_op>q_n_<mode>,
902 @mve_vcmp<mve_cmp_op>q_n_f<mode>): Apply vec_duplicate to scalar
905 2022-12-06 Jakub Jelinek <jakub@redhat.com>
908 * config/i386/i386.md (cbranchbf4, cstorebf4): Guard expanders
909 with the same condition as cbranchsf4 or cstoresf4 expanders.
911 2022-12-06 Richard Biener <rguenther@suse.de>
913 PR tree-optimization/104475
914 * pointer-query.h (access_ref::ref_nullptr_p): New flag.
915 * pointer-query.cc (access_ref::access_ref): Initialize
917 (compute_objsize_r): Set ref_nullptr_p if we treat it that way.
918 (access_ref::inform_access): If ref was treated as nullptr
919 based, indicate that.
921 2022-12-06 Jakub Jelinek <jakub@redhat.com>
923 PR tree-optimization/107972
924 * range-op-float.cc (frange_drop_infs): New function.
925 (float_binary_op_range_finish): Add DIV_OP2 argument. If DIV_OP2 is
926 false and lhs is finite or if DIV_OP2 is true and lhs is non-zero and
927 not NAN, r must be finite too.
928 (foperator_div::op2_range): Pass true to DIV_OP2 of
929 float_binary_op_range_finish.
931 2022-12-06 Jakub Jelinek <jakub@redhat.com>
933 PR tree-optimization/107975
934 * range-op-float.cc (foperator_mult::op1_range,
935 foperator_div::op1_range, foperator_div::op2_range): Just
936 return float_binary_op_range_finish result if lhs is known
937 NAN, or the other operand is known NAN or UNDEFINED.
939 2022-12-06 Prathamesh Kulkarni <prathamesh.kulkarni@linaro.org>
941 * config/aarch64/aarch64.cc (aarch64_expand_vector_init): Use dup
942 and zip1 for interleaving elements in vector initializer.
944 2022-12-05 Richard Biener <rguenther@suse.de>
947 * tree-into-ssa.cc (rewrite_update_phi_arguments): Only
948 update the argument when the reaching definition is different
949 from the current argument. Keep an existing argument
952 2022-12-05 Richard Biener <rguenther@suse.de>
954 PR tree-optimization/106868
955 * gimple-ssa-warn-access.cc (pass_waccess::gimple_call_return_arg_ref):
956 Inline into single user ...
957 (pass_waccess::check_dangling_uses): ... here and adjust the
958 call and the PHI case to require that ref.aref is the address
961 2022-12-05 Jakub Jelinek <jakub@redhat.com>
964 * match.pd (cmp @0 REAL_CST@1): Don't optimize x cmp NaN
965 or NaN cmp x to false/true for cmp >/>=/</<= if -ftrapping-math.
967 2022-12-05 Jakub Jelinek <jakub@redhat.com>
969 PR tree-optimization/107879
970 * range-op-float.cc (foperator_mult::op1_range): If both
971 lhs and op2 ranges contain zero or both ranges contain
972 some infinity, set r range to zero_to_inf_range depending on
974 (foperator_div::op2_range): Similarly for lhs and op1 ranges.
975 (foperator_div::op1_range): If lhs range contains zero and op2
976 range contains some infinity or vice versa, set r range to
977 zero_to_inf_range depending on signbit_known_p.
978 (foperator_div::rv_fold): Fix up condition for returning known NAN.
980 2022-12-05 Richard Biener <rguenther@suse.de>
982 PR tree-optimization/107833
983 PR tree-optimization/107839
984 * cfghooks.cc: Include tree.h.
985 * tree-ssa-loop-im.cc (movement_possibility): Wrap and
986 make stmts using any ssa_name_maybe_undef_p operand
987 to preserve execution.
988 (loop_invariant_motion_in_fun): Call mark_ssa_maybe_undefs
989 to init maybe-undefined status.
990 * tree-ssa-loop-ivopts.cc (ssa_name_maybe_undef_p,
991 ssa_name_set_maybe_undef, ssa_name_any_use_dominates_bb_p,
992 mark_ssa_maybe_undefs): Move ...
993 * tree-ssa.cc: ... here.
994 * tree-ssa.h (ssa_name_any_use_dominates_bb_p,
995 mark_ssa_maybe_undefs): Declare.
996 (ssa_name_maybe_undef_p, ssa_name_set_maybe_undef): Define.
998 2022-12-05 Andrew Pinski <pinskia@gmail.com>
1000 PR tree-optimization/107956
1001 * tree-vect-patterns.cc (vect_recog_mask_conversion_pattern):
1002 Check for NULL LHS on masked loads.
1004 2022-12-05 Kewen Lin <linkw@linux.ibm.com>
1005 Richard Sandiford <richard.sandiford@arm.com>
1007 PR tree-optimization/107412
1008 * gimple-fold.cc (gimple_fold_partial_load_store_mem_ref): Use
1009 untruncated type for the length, and avoid to_constant and tree
1010 arithmetic for subtraction.
1012 2022-12-02 Bernhard Reutner-Fischer <rep.dot.nop@gmail.com>
1013 Jason Merrill <jason@redhat.com>
1015 * function.cc (init_function_start): Use DECL_RESULT location
1016 for -Waggregate-return warning.
1018 2022-12-02 Andrew MacLeod <amacleod@redhat.com>
1020 * fold-const.cc (fold_unary_loc): Check TREE_TYPE of node.
1021 (tree_invalid_nonnegative_warnv_p): Likewise.
1023 2022-12-02 Jason Merrill <jason@redhat.com>
1025 * gcc.cc (validate_switches): Reset suffix/starred on loop.
1027 2022-12-02 Vladimir N. Makarov <vmakarov@redhat.com>
1029 * lra-constraints.cc (curr_insn_transform): Check available hard
1030 regs for pseudo and its subreg to decide what to reload.
1032 2022-12-02 liuhongt <hongtao.liu@intel.com>
1034 * config/i386/i386-expand.cc
1035 (ix86_expand_fast_convert_bf_to_sf): Use extendbfsf2_1 for
1036 nonimmediate operand.
1038 2022-12-02 Martin Liska <mliska@suse.cz>
1040 * configure: Regenerate.
1042 2022-12-02 Jakub Jelinek <jakub@redhat.com>
1045 * config/i386/i386-expand.cc (ix86_vector_duplicate_value): Save/restore
1046 recog_data around recog_memoized calls.
1048 2022-12-02 Michael Collison <collison@rivosinc.com>
1050 * match.pd ((x & 0x1) == 0) ? y : z <op> y
1051 -> (-(typeof(y))(x & 0x1) & z) <op> y.
1053 2022-12-02 Richard Biener <rguenther@suse.de>
1055 PR tree-optimization/107946
1056 * params.opt (-param=max-unswitch-depth=): New.
1057 * doc/invoke.texi (--param=max-unswitch-depth): Document.
1058 * tree-ssa-loop-unswitch.cc (init_loop_unswitch_info): Honor
1059 --param=max-unswitch-depth
1061 2022-12-02 Eric Gallager <egallager@gcc.gnu.org>
1064 * configure: Regenerate.
1065 * configure.ac: Document --with-dwarf2 flag as also
1066 applying to later DWARF standards.
1067 * doc/install.texi: Likewise.
1069 2022-12-02 liuhongt <hongtao.liu@intel.com>
1072 * config/i386/i386.md (extendbfsf2_1): Change type from
1073 sseishft to sseishft1.
1075 2022-12-01 Alex Coplan <alex.coplan@arm.com>
1077 * varasm.cc (assemble_variable): Fix type confusion bug when
1078 checking for ".vtable_map_vars" section.
1080 2022-12-01 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
1082 * config/riscv/riscv-v.cc (emit_pred_op): Adapt for mask mode.
1083 * config/riscv/vector.md: Remove Tail && make policy operand for mask mode mov.
1085 2022-12-01 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
1087 * config/riscv/riscv-protos.h (enum vlmul_type): New enum.
1088 (get_vlmul): New function.
1090 * config/riscv/riscv-v.cc (struct mode_vtype_group): New struct.
1091 (ENTRY): Adapt for attributes.
1092 (enum vlmul_type): New enum.
1093 (get_vlmul): New function.
1094 (get_ratio): New function.
1095 * config/riscv/riscv-vector-switch.def (ENTRY): Adapt for attributes.
1096 * config/riscv/riscv.cc (ENTRY): Ditto.
1097 * config/riscv/vector.md (false,true): Add attributes.
1099 2022-12-01 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
1101 * config/riscv/constraints.md (Wdm): New constraint.
1102 * config/riscv/predicates.md (direct_broadcast_operand): New predicate.
1103 * config/riscv/riscv-protos.h (RVV_VLMAX): New macro.
1104 (emit_pred_op): Refine function.
1105 * config/riscv/riscv-selftests.cc (run_const_vector_selftests): New function.
1106 (run_broadcast_selftests): Ditto.
1107 (BROADCAST_TEST): New tests.
1108 (riscv_run_selftests): More tests.
1109 * config/riscv/riscv-v.cc (emit_pred_move): Refine function.
1110 (emit_vlmax_vsetvl): Ditto.
1111 (emit_pred_op): Ditto.
1112 (expand_const_vector): New function.
1113 (legitimize_move): Add constant vector support.
1114 * config/riscv/riscv.cc (riscv_print_operand): New asm print rule for const vector.
1115 * config/riscv/riscv.h (X0_REGNUM): New macro.
1116 * config/riscv/vector-iterators.md: New attribute.
1117 * config/riscv/vector.md (vec_duplicate<mode>): New pattern.
1118 (@pred_broadcast<mode>): New pattern.
1120 2022-12-01 Paul-Antoine Arras <pa@codesourcery.com>
1122 * config/gcn/gcn-opts.h (TARGET_FIJI): -march=fiji.
1123 (TARGET_VEGA10): -march=gfx900.
1124 (TARGET_VEGA20): -march=gfx906.
1125 (TARGET_GFX908): -march=gfx908.
1126 (TARGET_GFX90a): -march=gfx90a.
1127 * config/gcn/gcn.h (TARGET_CPU_CPP_BUILTINS): Define a builtin that
1128 uniquely maps to '-march'.
1130 2022-12-01 Richard Biener <rguenther@suse.de>
1132 PR tree-optimization/107937
1133 * gimple-predicate-analysis.h (predicate::is_true): New.
1134 (predicate::is_false): Likewise.
1135 (predicate::empty_val): Likewise.
1136 (uninit_analysis::uninit_analysis): Properly initialize
1138 * gimple-predicate-analysis.cc (simplify_1b): Indicate
1139 whether the chain became empty.
1140 (predicate::simplify): Release emptied chain before removing it.
1141 (predicate::normalize): Replace temporary object with assertion.
1142 (uninit_analysis::is_use_guarded): Deal with predicates
1143 that simplify to true/false.
1145 2022-12-01 Richard Biener <rguenther@suse.de>
1147 PR tree-optimization/107935
1148 * tree-ssa-sccvn.cc (visit_phi): Honor forced VARYING on
1151 2022-12-01 Jakub Jelinek <jakub@redhat.com>
1154 * config/i386/i386.md (*concat<mode><dwi>3_1, *concat<mode><dwi>3_2):
1155 For operands which are zero_extend arguments allow memory if
1156 output operand is a register.
1157 (*concat<mode><dwi>3_3, *concat<mode><dwi>3_4): Likewise. If
1158 both input operands are memory, use early clobber on output operand.
1159 * config/i386/i386-expand.cc (split_double_concat): Deal with corner
1160 cases where one input is memory and the other is not and the address
1161 of the memory input uses a register we'd overwrite before loading
1162 the memory into a register.
1164 2022-12-01 Haochen Gui <guihaoc@gcc.gnu.org>
1166 * config/rs6000/rs6000-call.cc (swap_endian_selector_for_mode):
1167 Corrects comments of this function and make them clear.
1169 2022-12-01 liuhongt <hongtao.liu@intel.com>
1172 * config/i386/i386-expand.cc (ix86_expand_vec_set_builtin):
1173 Convert op1 to target mode whenever mode mismatch.
1175 2022-12-01 David Malcolm <dmalcolm@redhat.com>
1179 (diagnostic_path::get_first_event_in_a_function): New decl.
1180 * diagnostic.cc (diagnostic_path::get_first_event_in_a_function):
1182 (diagnostic_path::interprocedural_p): Ignore leading events that
1183 are outside of any function.
1185 2022-12-01 David Malcolm <dmalcolm@redhat.com>
1187 * Makefile.in (ANALYZER_OBJS): Add analyzer/bounds-checking.o.
1189 2022-12-01 Haochen Gui <guihaoc@gcc.gnu.org>
1192 * config/rs6000/rs6000-call.cc (swap_endian_selector_for_mode):
1193 Generate permute index directly for little endian targets.
1194 * config/rs6000/vsx.md (revb_<mode>): Call vprem directly with
1195 corresponding permute indexes.
1197 2022-11-30 John David Anglin <danglin@gcc.gnu.org>
1199 * config/pa/pa.md (addvdi3): Force operand 2 to a register.
1200 Remove "addi,tsv,*" instruction from unamed pattern.
1201 (subvdi3): Force operand 1 to a register.
1202 Remove "subi,tsv" instruction from from unamed pattern.
1204 2022-11-30 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
1206 * config/aarch64/aarch64.md (aarch64_cpymemdi): Specify clobber of CC reg.
1207 (*aarch64_cpymemdi): Likewise.
1208 (aarch64_movmemdi): Likewise.
1209 (aarch64_setmemdi): Likewise.
1210 (*aarch64_setmemdi): Likewise.
1212 2022-11-30 Martin Liska <mliska@suse.cz>
1214 * tree-switch-conversion.cc (bit_test_cluster::emit): Remove
1215 dead variable bt_range.
1217 2022-11-30 Iskander Shakirzyanov <iskander@ispras.ru>
1218 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
1221 * common.opt (Warray-bounds): Turn into alias of
1223 * builtins.cc (c_strlen): Use OPT_Warray_bounds_
1224 instead of OPT_Warray_bounds.
1225 * diagnostic-spec.cc (nowarn_spec_t::nowarn_spec_t): Ditto.
1226 * gimple-array-bounds.cc (array_bounds_checker::check_array_ref,
1227 array_bounds_checker::check_mem_ref,
1228 array_bounds_checker::check_addr_expr,
1229 array_bounds_checker::check_array_bounds): Ditto.
1230 * gimple-ssa-warn-restrict.cc (maybe_diag_access_bounds): Ditto.
1232 2022-11-30 Martin Liska <mliska@suse.cz>
1234 PR tree-optimization/101301
1235 PR tree-optimization/103680
1236 * tree-switch-conversion.cc (bit_test_cluster::emit):
1237 Handle correctly remaining probability.
1238 (switch_decision_tree::try_switch_expansion): Fix BB's count
1239 where a cluster expansion happens.
1240 (switch_decision_tree::emit_cmp_and_jump_insns): Fill up also
1242 (switch_decision_tree::do_jump_if_equal): Likewise.
1243 (switch_decision_tree::emit_case_nodes): Handle special case
1244 for BT expansion which can also fallback to a default BB.
1245 * tree-switch-conversion.h (cluster::cluster): Add
1246 m_default_prob probability.
1248 2022-11-30 Richard Biener <rguenther@suse.de>
1250 PR tree-optimization/107919
1251 * gimple-predicate-analysis.cc (simplify_1): Rename to ...
1252 (simplify_1a): .. this.
1254 (predicate::simplify): Call both simplify_1a and simplify_1b.
1256 2022-11-30 Richard Biener <rguenther@suse.de>
1258 * tree-ssa-uninit.cc (find_uninit_use): Dump the edge for a
1261 2022-11-30 Richard Biener <rguenther@suse.de>
1263 PR tree-optimization/107919
1264 * gimple-predicate-analysis.cc (predicate::simplify_2):
1265 Handle predicates of arbitrary length.
1267 2022-11-30 Jakub Jelinek <jakub@redhat.com>
1269 PR tree-optimization/107835
1270 * tree-chrec.cc (chrec_apply): Don't handle "{a, +, a} (x-1)"
1271 as "a*x" if type is a pointer type.
1273 2022-11-30 Paul-Antoine Arras <pa@codesourcery.com>
1275 * config/gcn/gcn.cc (gcn_omp_device_kind_arch_isa): Add gfx803.
1276 * config/gcn/t-omp-device: Add gfx803.
1278 2022-11-30 Lulu Cheng <chenglulu@loongson.cn>
1280 * config/loongarch/linux.h (STACK_CHECK_MOVING_SP):
1281 Define this macro to 1.
1282 * config/loongarch/loongarch.cc (STACK_CLASH_PROTECTION_GUARD_SIZE):
1284 (loongarch_first_stack_step): Return the size of the first drop stack
1285 according to whether stack checking is performed.
1286 (loongarch_emit_probe_stack_range): Adjust the method of stack checking in prologue.
1287 (loongarch_output_probe_stack_range): Delete useless code.
1288 (loongarch_expand_prologue): Adjust the method of stack checking in prologue.
1289 (loongarch_option_override_internal): Enforce that interval is the same
1290 size as size so the mid-end does the right thing.
1291 * config/loongarch/loongarch.h (STACK_CLASH_MAX_UNROLL_PAGES):
1292 New macro decide whether to loop stack detection.
1294 2022-11-30 David Malcolm <dmalcolm@redhat.com>
1297 * doc/invoke.texi (Static Analyzer Options): Add isatty, ferror,
1298 fileno, and getc to the list of functions known to the analyzer.
1300 2022-11-29 Richard Biener <rguenther@suse.de>
1302 PR tree-optimization/107852
1303 * tree-ssa-sccvn.cc (visit_phi): Use equivalences recorded
1304 as predicated values to elide more redundant PHIs.
1306 2022-11-29 Richard Biener <rguenther@suse.de>
1308 PR tree-optimization/106995
1309 * tree-if-conv.cc (pass_if_conversion::execute): Also redirect the
1310 versioning condition to the original loop if this very loop
1311 vanished during CFG cleanup.
1313 2022-11-29 Richard Biener <rguenther@suse.de>
1315 PR tree-optimization/107898
1316 * gimple-ssa-warn-alloca.cc (alloca_call_type): Check
1317 the type of the alloca argument is compatible with size_t
1318 before querying ranges.
1320 2022-11-29 Richard Biener <rguenther@suse.de>
1323 * multiple_target.cc (pass_target_clone::gate): Disable
1326 2022-11-29 Martin Liska <mliska@suse.cz>
1328 * configure: Regenerate.
1330 2022-11-29 YunQiang Su <yunqiang.su@cipunited.com>
1332 * configure.ac: add description for
1333 AC_DEFINE(ENABLE_MULTIARCH, 1)
1335 2022-11-28 Andrew Pinski <apinski@marvell.com>
1337 * match.pd ((A / (1 << B)) -> (A >> B).):
1340 2022-11-28 Sinan <sinan.lin@linux.alibaba.com>
1342 * config/riscv/riscv.cc (riscv_build_integer): Improve some cases
1343 of loading 64bit constants for rv32.
1345 2022-11-28 Maciej W. Rozycki <macro@embecosm.com>
1347 * config/riscv/riscv.cc (riscv_emit_int_order_test): Use EQ 0
1348 rather that XOR 1 for LE and LEU operations.
1350 2022-11-28 Richard Biener <rguenther@suse.de>
1352 PR tree-optimization/107896
1353 * tree-vect-stmts.cc (supportable_widening_operation):
1354 Handle non-vector mode intermediate mode.
1356 2022-11-28 Frolov Daniil <frolov.da@phystech.edu>
1358 * gimple-ssa-sprintf.cc (fmtresult::type_max_digits): Handle
1360 (tree_digits): Likewise.
1361 (format_integer): Likewise.
1362 (parse_directive): Add cases for %b and %B directives.
1364 2022-11-28 Fei Gao <gaofei@eswincomputing.com>
1366 * config/riscv/riscv.cc (riscv_first_stack_step): Fix computation
1367 of MIN_FIRST_STEP to cover FP save area too.
1369 2022-11-28 Richard Biener <rguenther@suse.de>
1371 PR tree-optimization/107493
1372 * tree-scalar-evolution.cc (scev_dfs::follow_ssa_edge_expr):
1373 Only handle no-op and sign-changing conversions.
1375 2022-11-28 Tobias Burnus <tobias@codesourcery.com>
1377 * config/gcn/gcn.cc (gcn_expand_builtin_1): Work on s1 instead
1378 of s[0:1] and use USE to prevent removal of setting that register.
1379 * config/gcn/gcn.md (prologue_use_di): Remove.
1381 2022-11-28 Yuri Gribov <y.gribov@samsung.com>
1384 * sanopt.cc: Do not optimize out checks for non-SSA addresses.
1386 2022-11-28 Jakub Jelinek <jakub@redhat.com>
1389 * config/i386/i386.opt (x_ix86_abi): Remove TargetSave.
1390 (ix86_abi): Replace it with TargetVariable.
1391 * config/i386/i386-options.cc (ix86_function_specific_save,
1392 ix86_function_specific_restore): Don't save and restore x_ix86_abi.
1394 2022-11-28 Andrea Corallo <andrea.corallo@arm.com>
1396 * config/arm/mve.md (mve_vrmlaldavhq_<supf>v4si,
1397 mve_vrmlaldavhaq_<supf>v4si): Fix spacing vs tabs.
1399 2022-11-28 Andrea Corallo <andrea.corallo@arm.com>
1401 * config/arm/mve.md (mve_vmlaldavaq_<supf><mode>)
1402 (mve_vmlaldavaxq_s<mode>, mve_vmlaldavaxq_p_<supf><mode>): Fix
1405 2022-11-28 Andrea Corallo <andrea.corallo@arm.com>
1407 * config/arm/mve.md (mve_vsubq_n_f<mode>): Fix spacing.
1409 2022-11-28 Andrea Corallo <andrea.corallo@arm.com>
1411 * config/arm/mve.md (mve_vaddlvq_p_<supf>v4si)
1412 (mve_vaddq_n_<supf><mode>, mve_vaddvaq_<supf><mode>)
1413 (mve_vaddlvaq_<supf>v4si, mve_vaddq_n_f<mode>)
1414 (mve_vaddlvaq_p_<supf>v4si, mve_vaddq<mode>, mve_vaddq_f<mode>):
1417 2022-11-28 Stam Markianos-Wright <stam.markianos-wright@arm.com>
1419 * config/arm/arm_mve.h (__arm_vsubq_x FP): New overloads.
1420 (__arm_vsubq_x Integer): New.
1422 2022-11-28 Stam Markianos-Wright <stam.markianos-wright@arm.com>
1425 * config/arm/arm_mve.h (__ARM_mve_typeid): Add float types.
1427 2022-11-28 Stam Markianos-Wright <stam.markianos-wright@arm.com>
1430 * config/arm/arm_mve.h (__arm_vaddq): Fix Overloading.
1431 (__arm_vmulq): Likewise.
1432 (__arm_vcmpeqq): Likewise.
1433 (__arm_vcmpneq): Likewise.
1434 (__arm_vmaxnmavq): Likewise.
1435 (__arm_vmaxnmvq): Likewise.
1436 (__arm_vminnmavq): Likewise.
1437 (__arm_vsubq): Likewise.
1438 (__arm_vminnmvq): Likewise.
1439 (__arm_vrshlq): Likewise.
1440 (__arm_vqsubq): Likewise.
1441 (__arm_vqdmulltq): Likewise.
1442 (__arm_vqdmullbq): Likewise.
1443 (__arm_vqdmulhq): Likewise.
1444 (__arm_vqaddq): Likewise.
1445 (__arm_vhaddq): Likewise.
1446 (__arm_vhsubq): Likewise.
1447 (__arm_vqdmlashq): Likewise.
1448 (__arm_vqrdmlahq): Likewise.
1449 (__arm_vmlasq): Likewise.
1450 (__arm_vqdmlahq): Likewise.
1451 (__arm_vmaxnmavq_p): Likewise.
1452 (__arm_vmaxnmvq_p): Likewise.
1453 (__arm_vminnmavq_p): Likewise.
1454 (__arm_vminnmvq_p): Likewise.
1455 (__arm_vfmasq_m): Likewise.
1456 (__arm_vsetq_lane): Likewise.
1457 (__arm_vcmpneq_m): Likewise.
1458 (__arm_vhaddq_x): Likewise.
1459 (__arm_vhsubq_x): Likewise.
1460 (__arm_vqrdmlashq_m): Likewise.
1461 (__arm_vqdmlashq_m): Likewise.
1462 (__arm_vmlaldavaxq_p): Likewise.
1463 (__arm_vmlasq_m): Likewise.
1464 (__arm_vqdmulhq_m): Likewise.
1465 (__arm_vqdmulltq_m): Likewise.
1466 (__arm_viwdupq_m): Likewise.
1467 (__arm_viwdupq_u16): Likewise.
1468 (__arm_viwdupq_u32): Likewise.
1469 (__arm_viwdupq_u8): Likewise.
1470 (__arm_vdwdupq_m): Likewise.
1471 (__arm_vdwdupq_u16): Likewise.
1472 (__arm_vdwdupq_u32): Likewise.
1473 (__arm_vdwdupq_u8): Likewise.
1474 (__arm_vaddlvaq): Likewise.
1475 (__arm_vaddlvaq_p): Likewise.
1476 (__arm_vaddvaq): Likewise.
1477 (__arm_vaddvaq_p): Likewise.
1478 (__arm_vcmphiq_m): Likewise.
1479 (__arm_vmladavaq_p): Likewise.
1480 (__arm_vmladavaxq): Likewise.
1481 (__arm_vmlaldavaxq): Likewise.
1482 (__arm_vrmlaldavhaq_p): Likewise.
1484 2022-11-28 Stam Markianos-Wright <stam.markianos-wright@arm.com>
1487 * config/arm/arm_mve.h (__arm_vaddq_m_n_s8): Change types.
1488 (__arm_vaddq_m_n_s32): Likewise.
1489 (__arm_vaddq_m_n_s16): Likewise.
1490 (__arm_vaddq_m_n_u8): Likewise.
1491 (__arm_vaddq_m_n_u32): Likewise.
1492 (__arm_vaddq_m_n_u16): Likewise.
1493 (__arm_vaddq_m): Fix Overloading.
1494 (__ARM_mve_coerce3): New.
1496 2022-11-28 Andrea Corallo <andrea.corallo@arm.com>
1498 * config/arm/mve.md (mve_vabsq_f<mode>): Fix spacing.
1500 2022-11-28 Andrea Corallo <andrea.corallo@arm.com>
1502 * config/arm/mve.md (@mve_vcmp<mve_cmp_op>q_<mode>): Fix
1504 * config/arm/arm_mve.h (__arm_vcmpgtq_m, __arm_vcmpleq_m)
1505 (__arm_vcmpltq_m, __arm_vcmpneq_m): Add missing defines.
1507 2022-11-28 Andrea Corallo <andrea.corallo@arm.com>
1509 * config/arm/mve.md (mve_vdupq_n_f<mode>)
1510 (mve_vdupq_n_<supf><mode>, mve_vdupq_m_n_<supf><mode>)
1511 (mve_vdupq_m_n_f<mode>): Fix spacing.
1513 2022-11-28 Andrea Corallo <andrea.corallo@arm.com>
1515 * config/arm/mve.md (mve_vdwdupq_m_wb_u<mode>_insn): Fix spacing.
1517 2022-11-28 Andrea Corallo <andrea.corallo@arm.com>
1519 * config/arm/mve.md (mve_vddupq_u<mode>_insn): Fix 'vddup.u'
1521 (mve_vddupq_m_wb_u<mode>_insn): Likewise.
1523 2022-11-28 Andrea Corallo <andrea.corallo@arm.com>
1525 * config/arm/vfp.md (*thumb2_movhi_vfp, *thumb2_movhi_fp16): Fix
1526 'vmsr' spacing and reg capitalization.
1528 2022-11-28 Richard Biener <rguenther@suse.de>
1530 PR tree-optimization/107876
1531 * tree-ssa-loop-unswitch.cc (clean_up_after_unswitching): Wipe
1532 dominator info if we removed an edge.
1534 2022-11-28 Richard Biener <rguenther@suse.de>
1536 PR tree-optimization/107867
1537 * tree-ssa-forwprop.cc (pass_forwprop::execute): Handle
1538 abnormal cleanup after substitution.
1540 2022-11-28 Lulu Cheng <chenglulu@loongson.cn>
1542 * config/loongarch/loongarch.cc (enum loongarch_load_imm_method):
1543 Remove the member METHOD_INSV that is not currently used.
1544 (struct loongarch_integer_op): Define a new member curr_value,
1545 that records the value of the number stored in the destination
1546 register immediately after the current instruction has run.
1547 (loongarch_build_integer): Assign a value to the curr_value member variable.
1548 (loongarch_move_integer): Adds information for the immediate load instruction.
1549 * config/loongarch/loongarch.md (*movdi_32bit): Redefine as define_insn_and_split.
1550 (*movdi_64bit): Likewise.
1551 (*movsi_internal): Likewise.
1552 (*movhi_internal): Likewise.
1553 * config/loongarch/predicates.md: Return true as long as it is CONST_INT, ensure
1554 that the immediate number is not optimized by decomposition during expand
1557 2022-11-28 liuhongt <hongtao.liu@intel.com>
1560 * config/i386/avx512bf16intrin.h (_mm_cvtsbh_ss): Refined.
1561 * config/i386/i386-builtin-types.def (FLOAT_FTYPE_BFLOAT16):
1563 * config/i386/i386-builtin.def (BDESC): New builtin.
1564 * config/i386/i386-expand.cc (ix86_expand_args_builtin):
1566 * config/i386/i386.md (extendbfsf2): New expander.
1567 (extendbfsf2_1): New define_insn.
1568 (truncsfbf2): Ditto.
1570 2022-11-26 Andrew Pinski <apinski@marvell.com>
1572 PR tree-optimization/103356
1573 * match.pd: ((~a) == b -> a ^ b): New pattern.
1575 2022-11-25 Sandra Loosemore <sandra@codesourcery.com>
1577 * common.opt (fopenmp-target-simd-clone): New option.
1578 (target_simd_clone_device): New enum to go with it.
1579 * doc/invoke.texi (-fopenmp-target-simd-clone): Document.
1580 * flag-types.h (enum omp_target_simd_clone_device_kind): New.
1581 * omp-simd-clone.cc (auto_simd_fail): New function.
1582 (auto_simd_check_stmt): New function.
1583 (plausible_type_for_simd_clone): New function.
1584 (ok_for_auto_simd_clone): New function.
1585 (simd_clone_create): Add force_local argument, make the symbol
1586 have internal linkage if it is true.
1587 (expand_simd_clones): Also check for cloneable functions with
1588 "omp declare target". Pass explicit_p argument to
1589 simd_clone.compute_vecsize_and_simdlen target hook.
1590 * opts.cc (default_options_table): Add -fopenmp-target-simd-clone.
1591 * target.def (TARGET_SIMD_CLONE_COMPUTE_VECSIZE_AND_SIMDLEN):
1592 Add bool explicit_p argument.
1593 * doc/tm.texi: Regenerated.
1594 * config/aarch64/aarch64.cc
1595 (aarch64_simd_clone_compute_vecsize_and_simdlen): Update.
1597 (gcn_simd_clone_compute_vecsize_and_simdlen): Update.
1598 * config/i386/i386.cc
1599 (ix86_simd_clone_compute_vecsize_and_simdlen): Update.
1601 2022-11-25 Tamar Christina <tamar.christina@arm.com>
1604 * config/aarch64/aarch64.cc
1605 (aarch64_vectorize_can_special_div_by_constant): Check validity during
1606 codegen phase as well.
1608 2022-11-25 Eric Botcazou <ebotcazou@adacore.com>
1610 * range-op.cc (operator_bitwise_xor::op1_range): Fix thinko.
1612 2022-11-25 Richard Biener <rguenther@suse.de>
1614 PR tree-optimization/107865
1615 * tree-cfg.cc (move_sese_region_to_fn): Free the number of
1616 iterations of moved loops.
1618 2022-11-25 Kewen.Lin <linkw@gcc.gnu.org>
1621 * config/rs6000/rs6000.cc (rs6000_print_patchable_function_entry):
1622 Adjust to call function default_print_patchable_function_entry.
1623 * targhooks.cc (default_print_patchable_function_entry_1): Remove and
1624 move the flags preparation ...
1625 (default_print_patchable_function_entry): ... here, adjust to use
1626 current_function_funcdef_no for label no.
1627 * targhooks.h (default_print_patchable_function_entry_1): Remove.
1628 * varasm.cc (default_elf_asm_named_section): Adjust code for
1629 __patchable_function_entries section support with LPFE label.
1631 2022-11-24 Wilco Dijkstra <wilco.dijkstra@arm.com>
1633 PR tree-optimization/107413
1634 * config/aarch64/aarch64.cc (struct tune_params): Add
1635 fma_reassoc_width to all CPU tuning structures.
1636 (aarch64_reassociation_width): Use fma_reassoc_width.
1637 * config/aarch64/aarch64-protos.h (struct tune_params): Add
1640 2022-11-24 Jakub Jelinek <jakub@redhat.com>
1642 PR middle-end/107317
1643 * asan.cc: Include diagnostic-core.h.
1644 (asan_emit_stack_protection): Return NULL early if seen_error ().
1646 2022-11-24 Aldy Hernandez <aldyh@redhat.com>
1648 * vr-values.cc (simplify_using_ranges::compare_names): Remove.
1649 (vrp_evaluate_conditional_warnv_with_ops): Remove call to
1651 (simplify_using_ranges::vrp_visit_cond_stmt): Remove use_equiv_p
1652 argument to vrp_evaluate_conditional_warnv_with_ops.
1653 * vr-values.h (class simplify_using_ranges): Remove
1655 Remove use_equiv_p to vrp_evaluate_conditional_warnv_with_ops.
1657 2022-11-24 Aldy Hernandez <aldyh@redhat.com>
1659 * tree-vrp.cc (overflow_comparison_p_1): Remove follow_assert_exprs.
1660 (overflow_comparison_p): Remove use_equiv_p.
1661 * tree-vrp.h (overflow_comparison_p): Same.
1662 * vr-values.cc (vrp_evaluate_conditional_warnv_with_ops): Remove
1663 use_equiv_p argument to overflow_comparison_p.
1665 2022-11-24 Aldy Hernandez <aldyh@redhat.com>
1667 * doc/gimple.texi: Remove ASSERT_EXPR references.
1668 * fold-const.cc (tree_expr_nonzero_warnv_p): Same.
1669 (fold_binary_loc): Same.
1670 (tree_expr_nonnegative_warnv_p): Same.
1671 * gimple-array-bounds.cc (get_base_decl): Same.
1672 * gimple-pretty-print.cc (dump_unary_rhs): Same.
1673 * gimple.cc (get_gimple_rhs_num_ops): Same.
1674 * pointer-query.cc (handle_ssa_name): Same.
1675 * tree-cfg.cc (verify_gimple_assign_single): Same.
1676 * tree-pretty-print.cc (dump_generic_node): Same.
1677 * tree-scalar-evolution.cc (scev_dfs::follow_ssa_edge_expr):Same.
1678 (interpret_rhs_expr): Same.
1679 * tree-ssa-operands.cc (operands_scanner::get_expr_operands): Same.
1680 * tree-ssa-propagate.cc
1681 (substitute_and_fold_dom_walker::before_dom_children): Same.
1682 * tree-ssa-threadedge.cc: Same.
1683 * tree-vrp.cc (overflow_comparison_p): Same.
1684 * tree.def (ASSERT_EXPR): Add note.
1685 * tree.h (ASSERT_EXPR_VAR): Remove.
1686 (ASSERT_EXPR_COND): Remove.
1687 * vr-values.cc (simplify_using_ranges::vrp_visit_cond_stmt):
1690 2022-11-24 Aldy Hernandez <aldyh@redhat.com>
1692 * Makefile.in: Remove value-range-equiv.o
1693 * gimple-array-bounds.cc
1694 (array_bounds_checker::array_bounds_checker): Remove comment.
1695 * tree-vrp.cc (supported_types_p): Remove use of value_range_equiv.
1696 * value-query.cc (class equiv_allocator): Same.
1697 (range_query::allocate_value_range_equiv): Remove.
1698 (range_query::free_value_range_equiv): Remove.
1699 (range_query::get_value_range): Remove.
1700 * value-query.h (class range_query): Remove get_value_range.
1701 Remove allocate_value_range_equiv.
1702 Remove free_value_range_equiv.
1703 * vr-values.cc (compare_ranges): Replace value_range_equiv with
1705 (simplify_using_ranges::get_vr_for_comparison): Same.
1706 (simplify_using_ranges::compare_names): Same.
1707 * vr-values.h: Remove value_range_equiv references.
1708 * value-range-equiv.cc: Removed.
1709 * value-range-equiv.h: Removed.
1711 2022-11-24 Aldy Hernandez <aldyh@redhat.com>
1713 * doc/invoke.texi: Remove docs for max-vrp-switch-assertions,
1714 vrp1-mode, and vrp2-mode.
1716 * range-op.cc (masked_increment): Move from tree-vrp.cc.
1717 * tree-vrp.cc (class live_names): Remove.
1718 (live_names::init_bitmap_if_needed): Remove.
1719 (live_names::block_has_live_names_p): Remove.
1720 (live_names::clear_block): Remove.
1721 (live_names::merge): Remove.
1722 (live_names::set): Remove.
1723 (live_names::clear): Remove.
1724 (live_names::live_names): Remove.
1725 (live_names::~live_names): Remove.
1726 (live_names::live_on_block_p): Remove.
1727 (live_names::live_on_edge_p): Remove.
1728 (get_single_symbol): Make static.
1729 (build_symbolic_expr): Remove.
1730 (adjust_symbolic_bound): Remove.
1731 (combine_bound): Remove.
1732 (set_value_range_with_overflow): Remove.
1733 (extract_range_from_pointer_plus_expr): Remove.
1734 (extract_range_from_plus_minus_expr): Remove.
1735 (drop_undefines_to_varying): Remove.
1736 (range_fold_binary_symbolics_p): Remove.
1737 (range_fold_unary_symbolics_p): Remove.
1738 (range_fold_binary_expr): Remove.
1739 (infer_value_range): Remove.
1740 (dump_assert_info): Remove.
1741 (dump_asserts_info): Remove.
1742 (add_assert_info): Remove.
1743 (extract_code_and_val_from_cond_with_ops): Remove.
1744 (masked_increment): Move to range-op.cc.
1745 (register_edge_assert_for_2): Remove.
1746 (find_case_label_index): Remove.
1747 (find_case_label_range): Remove.
1748 (register_edge_assert_for_1): Remove.
1749 (is_masked_range_test): Remove.
1750 (register_edge_assert_for): Remove.
1751 (stmt_interesting_for_vrp): Remove.
1752 (struct case_info): Remove.
1753 (struct assert_locus): Remove.
1754 (class vrp_asserts): Remove.
1755 (vrp_asserts::build_assert_expr_for): Remove.
1756 (vrp_asserts::dump): Remove.
1757 (vrp_asserts::register_new_assert_for): Remove.
1758 (vrp_asserts::finish_register_edge_assert_for): Remove.
1759 (vrp_asserts::find_conditional_asserts): Remove.
1760 (vrp_asserts::compare_case_labels): Remove.
1761 (vrp_asserts::find_switch_asserts): Remove.
1762 (vrp_asserts::find_assert_locations_in_bb): Remove.
1763 (vrp_asserts::find_assert_locations): Remove.
1764 (vrp_asserts::process_assert_insertions_for): Remove.
1765 (vrp_asserts::compare_assert_loc): Remove.
1766 (vrp_asserts::process_assert_insertions): Remove.
1767 (vrp_asserts::insert_range_assertions): Remove.
1768 (vrp_asserts::all_imm_uses_in_stmt_or_feed_cond): Remove.
1769 (vrp_asserts::remove_range_assertions): Remove.
1770 (class vrp_prop): Remove.
1771 (vrp_prop::initialize): Remove.
1772 (enum ssa_prop_result): Remove.
1773 (vrp_prop::visit_stmt): Remove.
1774 (vrp_prop::visit_phi): Remove.
1775 (vrp_prop::finalize): Remove.
1776 (class vrp_folder): Remove.
1777 (vrp_folder::fold_predicate_in): Remove.
1778 (vrp_folder::fold_stmt): Remove.
1779 (vrp_folder::simplify_casted_conds): Remove.
1780 (execute_vrp): Remove.
1781 * tree-vrp.h (struct assert_info): Remove.
1782 (register_edge_assert_for): Remove.
1783 (stmt_interesting_for_vrp): Remove.
1784 (infer_value_range): Remove.
1785 (get_single_symbol): Remove.
1786 (masked_increment): Remove.
1787 (execute_ranger_vrp): Remove.
1788 * vr-values.cc (set_value_range_to_nonnegative): Remove.
1789 (set_value_range_to_truthvalue): Remove.
1790 (vr_values::get_lattice_entry): Remove.
1791 (vr_values::get_value_range): Remove.
1792 (vr_values::range_of_expr): Remove.
1793 (vr_values::value_of_expr): Remove.
1794 (vr_values::value_on_edge): Remove.
1795 (vr_values::value_of_stmt): Remove.
1796 (vr_values::set_def_to_varying): Remove.
1797 (vr_values::set_defs_to_varying): Remove.
1798 (vr_values::update_value_range): Remove.
1799 (symbolic_range_based_on_p): Remove.
1800 (gimple_assign_nonzero_p): Remove.
1801 (gimple_stmt_nonzero_p): Remove.
1802 (vr_values::vrp_stmt_computes_nonzero): Remove.
1803 (vr_values::op_with_constant_singleton_value_range): Remove.
1804 (vr_values::extract_range_for_var_from_comparison_expr): Remove.
1805 (vr_values::extract_range_from_assert): Remove.
1806 (vr_values::extract_range_from_ssa_name): Remove.
1807 (vr_values::extract_range_from_binary_expr): Remove.
1808 (vr_values::extract_range_from_unary_expr): Remove.
1809 (vr_values::extract_range_from_cond_expr): Remove.
1810 (vr_values::extract_range_from_comparison): Remove.
1811 (vr_values::extract_range_from_ubsan_builtin): Remove.
1812 (vr_values::extract_range_basic): Remove.
1813 (vr_values::extract_range_from_assignment): Remove.
1814 (vr_values::adjust_range_with_scev): Remove.
1815 (vr_values::dump): Remove.
1816 (vr_values::vr_values): Remove.
1817 (vr_values::~vr_values): Remove.
1818 (vrp_valueize): Remove.
1819 (vrp_valueize_1): Remove.
1820 (get_output_for_vrp): Remove.
1821 (vr_values::vrp_visit_assignment_or_call): Remove.
1822 (simplify_using_ranges::vrp_evaluate_conditional): Remove.
1823 (vr_values::vrp_visit_switch_stmt): Remove.
1824 (vr_values::extract_range_from_stmt): Remove.
1825 (vr_values::extract_range_from_phi_node): Remove.
1826 (simplify_using_ranges::fold_cond): Add FIXME note.
1827 (vr_values::set_vr_value): Remove.
1828 (vr_values::swap_vr_value): Remove.
1829 * vr-values.h (class vr_values): Remove.
1830 (get_output_for_vrp): Remove.
1832 2022-11-24 Hongyu Wang <hongyu.wang@intel.com>
1835 * common/config/i386/i386-common.cc (ix86_optimization_table):
1836 Enable loop unroll O2, disable -fweb and -frename-registers
1838 * config/i386/i386-options.cc
1839 (ix86_override_options_after_change):
1840 Disable small loop unroll when funroll-loops enabled, reset
1841 cunroll_grow_size when it is not explicitly enabled.
1842 (ix86_option_override_internal): Call
1843 ix86_override_options_after_change instead of calling
1844 ix86_recompute_optlev_based_flags and ix86_default_align
1846 * config/i386/i386.cc (ix86_loop_unroll_adjust): Adjust unroll
1847 factor if -munroll-only-small-loops enabled.
1848 * loop-init.cc (pass_rtl_unroll_loops::gate): Do not enable
1849 loop unrolling for -O2-speed.
1850 (pass_rtl_unroll_loops::execute): Rmove
1851 targetm.loop_unroll_adjust check.
1853 2022-11-23 Ramana Radhakrishnan <ramana.gcc@gmail.com>
1855 * config/arm/types.md: Update comment.
1856 (is_neon_type): Add neon_fcmla, neon_fcadd.
1858 2022-11-23 Jonathan Wakely <jwakely@redhat.com>
1860 * doc/invoke.texi (C++ Dialect Options): Recommend using
1861 -Wdelete-non-virtual-dtor instead of -Wnon-virtual-dtor.
1863 2022-11-23 Jakub Jelinek <jakub@redhat.com>
1866 * diagnostic.cc (test_diagnostic_get_location_text): Test
1867 special_fname_builtin () rather than "<built-in>" and expect
1868 special_fname_builtin () concatenated with ":" for it.
1870 2022-11-23 Martin Liska <mliska@suse.cz>
1872 * Makefile.in: Set DO_LINK_MUTEX.
1874 2022-11-23 Lulu Cheng <chenglulu@loongson.cn>
1875 xujiahao <xujiahao@loongson.cn>
1877 * config/loongarch/constraints.md (ZD): New constraint.
1878 * config/loongarch/loongarch-def.c: Initial number of parallel prefetch.
1879 * config/loongarch/loongarch-tune.h (struct loongarch_cache):
1880 Define number of parallel prefetch.
1881 * config/loongarch/loongarch.cc (loongarch_option_override_internal):
1882 Set up parameters to be used in prefetching algorithm.
1883 * config/loongarch/loongarch.md (prefetch): New template.
1885 2022-11-23 Marek Polacek <polacek@redhat.com>
1888 2022-11-23 Marek Polacek <polacek@redhat.com>
1890 * Makefile.in: Set LD_PICFLAG. Use it. Set enable_host_pie.
1891 Remove NO_PIE_CFLAGS and NO_PIE_FLAG. Pass LD_PICFLAG to
1892 ALL_LINKERFLAGS. Use the "pic" build of libiberty if --enable-host-pie.
1893 * configure.ac (--enable-host-shared): Don't set PICFLAG here.
1894 (--enable-host-pie): New check. Set PICFLAG and LD_PICFLAG after this
1896 * configure: Regenerate.
1897 * doc/install.texi: Document --enable-host-pie.
1899 2022-11-23 Marek Polacek <polacek@redhat.com>
1902 2022-11-23 Marek Polacek <polacek@redhat.com>
1904 * configure.ac (--enable-host-bind-now): New check. Add
1905 -Wl,-z,now to LD_PICFLAG if --enable-host-bind-now.
1906 * configure: Regenerate.
1907 * doc/install.texi: Document --enable-host-bind-now.
1909 2022-11-23 Marek Polacek <polacek@redhat.com>
1911 * configure.ac (--enable-host-bind-now): New check. Add
1912 -Wl,-z,now to LD_PICFLAG if --enable-host-bind-now.
1913 * configure: Regenerate.
1914 * doc/install.texi: Document --enable-host-bind-now.
1916 2022-11-23 Marek Polacek <polacek@redhat.com>
1918 * Makefile.in: Set LD_PICFLAG. Use it. Set enable_host_pie.
1919 Remove NO_PIE_CFLAGS and NO_PIE_FLAG. Pass LD_PICFLAG to
1920 ALL_LINKERFLAGS. Use the "pic" build of libiberty if --enable-host-pie.
1921 * configure.ac (--enable-host-shared): Don't set PICFLAG here.
1922 (--enable-host-pie): New check. Set PICFLAG and LD_PICFLAG after this
1924 * configure: Regenerate.
1925 * doc/install.texi: Document --enable-host-pie.
1927 2022-11-22 Jeff Law <jlaw@ventanamicro.com>
1929 * tree-ssa-dom.cc (record_edge_info): Fix thinko in last commit.
1931 2022-11-22 Jeff Law <jlaw@ventanamicro.com>
1933 * tree-ssa-dom.cc (record_edge_info): Fix comment typos.
1935 2022-11-22 Jeff Law <jeffreyalaw@gmail.com>
1937 * config/riscv/riscv.cc (riscv_get_separate_components): Do not
1938 do shrink-wrapping for a frame with a variable size.
1940 2022-11-22 Siddhesh Poyarekar <siddhesh@gotplt.org>
1942 * tree-object-size.cc (todo): New variable.
1943 (object_sizes_execute): Use it.
1944 (strdup_object_size): New function.
1945 (call_object_size): Use it.
1947 2022-11-22 Eric Botcazou <ebotcazou@adacore.com>
1949 * tree-loop-distribution.cc (loop_distribution::classify_builtin_ldst):
1950 Bail out if source and destination do not have the same storage order.
1952 2022-11-22 Martin Jambor <mjambor@suse.cz>
1954 * ipa-sra.cc (ipa_sra_dump_all_summaries): Dump edge summaries even
1957 2022-11-22 Martin Jambor <mjambor@suse.cz>
1959 * ipa-cp.cc (push_agg_values_from_edge): Do not consider constants
1960 in unused aggregate parameters.
1962 2022-11-22 Martin Jambor <mjambor@suse.cz>
1964 * ipa-param-manipulation.h (ipa_param_body_adjustments): Removed
1965 member m_split_modifications_p.
1966 * ipa-param-manipulation.cc
1967 (ipa_param_body_adjustments::common_initialization): Do not set
1968 m_split_modifications_p.
1969 (ipa_param_body_adjustments::ipa_param_body_adjustments): Remove
1970 initializations of m_split_modifications_p.
1971 (ipa_param_body_adjustments::modify_call_stmt): Check that
1972 m_replacements is empty instead of m_split_modifications_p.
1974 2022-11-22 Martin Jambor <mjambor@suse.cz>
1977 * ipa-cp.cc (push_agg_values_from_edge): New parameter
1978 optimize_self_recursion, use it to decide whether to pass interim to
1979 the helper function.
1980 (find_aggregate_values_for_callers_subset): Pass true in the new
1981 parameter of push_agg_values_from_edge.
1982 (cgraph_edge_brings_all_agg_vals_for_node): Pass false in the new
1983 parameter of push_agg_values_from_edge.
1985 2022-11-22 Tobias Burnus <tobias@codesourcery.com>
1986 Andrew Stubbs <ams@codesourcery.com>
1988 * config/gcn/gcn-builtins.def (FIRST_CALL_THIS_THREAD_P,
1989 GET_STACK_LIMIT): Add new builtins.
1990 * config/gcn/gcn.cc (gcn_expand_builtin_1): Expand them.
1991 * config/gcn/gcn.md (prologue_use): Add "register_operand" as
1992 arg to match_operand.
1993 (prologue_use_di): New; DI insn_and_split variant of the former.
1995 2022-11-22 Richard Biener <rguenther@suse.de>
1997 PR tree-optimization/107803
1998 * tree-ssa-propagate.cc (substitute_and_fold_dom_walker): Add
1999 need_ab_cleanup member.
2000 (substitute_and_fold_dom_walker::before_dom_children): When
2001 a stmt can no longer transfer control flow abnormally set
2003 (substitute_and_fold_engine::substitute_and_fold): Cleanup
2004 abnormal control flow.
2006 2022-11-22 Richard Biener <rguenther@suse.de>
2008 PR tree-optimization/107672
2009 * tree-vect-stmts.cc (supportable_widening_operation): Avoid
2010 type_for_mode on vector modes.
2012 2022-11-22 Richard Biener <rguenther@suse.de>
2014 PR tree-optimization/107766
2015 * tree-vect-slp-patterns.cc (complex_mul_pattern::matches):
2016 Use *node to check for FP vector types.
2018 2022-11-22 liuhongt <hongtao.liu@intel.com>
2020 * config/i386/mmx.md (*mov<mode>_internal): Add
2021 ix86_hard_reg_move_ok to condition.
2023 2022-11-22 liuhongt <hongtao.liu@intel.com>
2025 * config/i386/i386.cc (ix86_can_change_mode_class): Also guard
2027 (ix86_hard_regno_mode_ok): Remove VALID_AVX512FP16_SCALAR_MODE
2028 * config/i386/i386.h (VALID_AVX512FP16_SCALAR_MODE): Merged to
2030 (VALID_AVX512F_SCALAR_MODE): .. this, also add HImode.
2031 (VALID_SSE_REG_MODE): Add DI/HImode.
2033 2022-11-21 Sebastian Huber <sebastian.huber@embedded-brains.de>
2035 * config/rtems.h (SUBTARGET_CC1_SPEC): Undef and define.
2037 2022-11-21 Sebastian Huber <sebastian.huber@embedded-brains.de>
2039 * gcc.cc (SUBTARGET_CC1_SPEC): Define if not defined.
2040 (cc1_spec): Append SUBTARGET_CC1_SPEC.
2042 2022-11-21 Dimitar Dimitrov <dimitar@dinux.eu>
2044 * doc/sourcebuild.texi (sockets): Document new check.
2046 2022-11-21 Andrew Carlotti <andrew.carlotti@arm.com>
2048 * tree-ssa-loop.h (tree_niter_desc): Update comment.
2050 2022-11-21 Andrew Carlotti <andrew.carlotti@arm.com>
2052 * tree-ssa-loop-niter.cc
2053 (number_of_iterations_exit_assumptions): Modify to call...
2054 (number_of_iterations_bitcount): ...this new function.
2055 (number_of_iterations_popcount): Now called by the above.
2056 Refactor, and extract popcount expression builder to...
2057 (build_popcount_expr): this new function.
2059 2022-11-21 Andrew Carlotti <andrew.carlotti@arm.com>
2061 * tree-ssa-loop-niter.cc (ssa_defined_by_minus_one_stmt_p): Move
2062 (number_of_iterations_popcount): Move, and remove separate prototype.
2064 2022-11-21 Andrew Carlotti <andrew.carlotti@arm.com>
2066 * tree-ssa-loop-niter.cc (number_of_iterations_exit_assumptions):
2067 Move at_stmt assignment.
2069 2022-11-21 Christophe Lyon <christophe.lyon@arm.com>
2071 * genmultilib: Fix options and dirnames/osdirnames sanity check.
2073 2022-11-21 Philipp Tomsich <philipp.tomsich@vrull.eu>
2076 * config/riscv/riscv.md
2077 (*branch<ANYI:mode>_shiftedarith_equals_zero): Wrap ANYI
2078 in a subreg, as our branch instructions only supports X.
2080 2022-11-21 Jakub Jelinek <jakub@redhat.com>
2083 * config/i386/avx512fp16intrin.h (_mm512_castph512_ph128,
2084 _mm512_castph512_ph256, _mm512_castph128_ph512,
2085 _mm512_castph256_ph512, _mm512_set1_pch): Uglify names of local
2086 variables and union members.
2087 * config/i386/avx512fp16vlintrin.h (_mm256_castph256_ph128,
2088 _mm256_castph128_ph256, _mm256_set1_pch, _mm_set1_pch): Likewise.
2089 * config/i386/smmintrin.h (_mm_extract_ps): Likewise.
2091 2022-11-21 Christophe Lyon <christophe.lyon@arm.com>
2093 * genmultilib: Add sanity check.
2095 2022-11-21 Eric Botcazou <ebotcazou@adacore.com>
2097 * config/arm/arm.cc (arm_attribute_table) <cmse_nonsecure_call>:
2098 Change decl_required field to false.
2099 (arm_handle_cmse_nonsecure_call): Deal with a TYPE node.
2101 2022-11-21 Andrew Pinski <apinski@marvell.com>
2103 * gimplify.cc (gimplify_modify_expr): If
2104 either *from_p or *to_p were error_operand
2107 2022-11-21 liuhongt <hongtao.liu@intel.com>
2109 * config/i386/i386-builtins.cc (def_builtin): Handle "shared"
2110 avx512bf16vl-avxneconvert builtins.
2112 2022-11-20 Jakub Jelinek <jakub@redhat.com>
2115 * reg-stack.cc (next_flags_user): Add DEBUG_SEEN argument.
2116 If >= 0 and a DEBUG_INSN would be otherwise returned, set
2117 DEBUG_SEEN to 1 and ignore it.
2118 (swap_rtx_condition): Add DEBUG_SEEN argument. In >= 0
2119 mode only set DEBUG_SEEN to 1 if problematic DEBUG_ISNSs
2120 were seen and revert all changes on success in that case.
2121 Don't try to recog_memoized DEBUG_INSNs.
2122 (compare_for_stack_reg): Adjust swap_rtx_condition caller.
2123 If it returns true and debug_seen is 1, call swap_rtx_condition
2124 again with debug_seen -1.
2126 2022-11-20 Jeff Law <jlaw@ventanamicro.com>
2129 * config/mn10300/mn10300.cc (mn10300_print_operand): Remove
2130 extraneous semicolon.
2131 * config/nvptx/nvptx.cc (nvptx_goacc_reduction_fini): Likewise.
2133 2022-11-19 Andrew Pinski <apinski@marvell.com>
2136 * tree-core.h (tree_code_type): Constexprify
2137 by including all-tree.def.
2138 (tree_code_length): Likewise.
2139 * tree.cc (tree_code_type): Remove.
2140 (tree_code_length): Remove.
2142 2022-11-19 Thomas Schwinge <thomas@codesourcery.com>
2144 * config/nvptx/nvptx.h (STARTFILE_SPEC): Fix 'crt0.o' for
2147 2022-11-19 Jonathan Wakely <jwakely@redhat.com>
2148 LIU Hao <lh_mouse@126.com>
2150 * system.h [INCLUDE_MUTEX]: Include header for std::mutex.
2152 2022-11-19 Jakub Jelinek <jakub@redhat.com>
2155 * config/i386/i386-protos.h (ix86_expand_fast_convert_bf_to_sf):
2157 * config/i386/i386-expand.cc (ix86_expand_fast_convert_bf_to_sf): New
2159 * config/i386/i386.md (cbranchbf4, cstorebf4): Use it.
2161 2022-11-19 Jeff Chapman II <jchapman@lock3software.com>
2162 Andrew Sutton <asutton@lock3software.com>
2163 Andrew Marmaduke <amarmaduke@lock3software.com>
2164 Michael Lopez <mlopez@lock3software.com>
2165 Jason Merrill <jason@redhat.com>
2167 * doc/invoke.texi: Document contracts flags.
2169 2022-11-18 Philipp Tomsich <philipp.tomsich@vrull.eu>
2171 * config/riscv/bitmanip.md (*minmax): Additional pattern for
2172 min/max against constants that are extension-invariant.
2173 * config/riscv/iterators.md (minmax_optab): Add an iterator
2174 that has only min and max rtl.
2176 2022-11-18 Philipp Tomsich <philipp.tomsich@vrull.eu>
2178 * config/riscv/bitmanip.md
2179 (*branch<X:mode>_mask_twobits_equals_singlebit):
2180 Handle "if ((a & T) == C)" using Zbs, when T has 2 bits set and C
2181 has one of these tow bits set.
2182 * config/riscv/predicates.md (const_twobits_not_arith_operand):
2185 2022-11-18 Philipp Tomsich <philipp.tomsich@vrull.eu>
2187 * config/riscv/bitmanip.md (*<or_optab>i<mode>_extrabit):
2188 New pattern for binvi+binvi/xori and bseti+bseti/ori
2189 (*andi<mode>_extrabit): New pattern for bclri+bclri/andi
2190 * config/riscv/iterators.md (any_or): Match or and ior
2191 * config/riscv/predicates.md (const_twobits_operand):
2193 (uimm_extra_bit_operand): New predicate.
2194 (uimm_extra_bit_or_twobits): New predicate.
2195 (not_uimm_extra_bit_operand): New predicate.
2196 (not_uimm_extra_bit_or_nottwobits): New predicate.
2197 * config/riscv/riscv.h (UIMM_EXTRA_BIT_OPERAND):
2198 Helper for the uimm_extra_bit_operand and
2199 not_uimm_extra_bit_operand predicates.
2201 2022-11-18 Philipp Tomsich <philipp.tomsich@vrull.eu>
2203 * config/riscv/bitmanip.md: Handle corner-cases for combine
2204 when chaining slli(.uw)? + addw
2205 * config/riscv/riscv-protos.h (riscv_shamt_matches_mask_p):
2207 * config/riscv/riscv.cc (riscv_shamt_matches_mask_p):
2208 Helper for evaluating the relationship between two operands.
2210 2022-11-18 Philipp Tomsich <philipp.tomsich@vrull.eu>
2212 * config/riscv/bitmanip.md: Add a define_split to optimize
2213 slliw + addiw + divw into sh[123]add + divw.
2215 2022-11-18 Philipp Tomsich <philipp.tomsich@vrull.eu>
2217 * config/riscv/predicates.md (shifted_const_arith_operand): New predicate.
2218 (uimm_extra_bit_operand): New predicate.
2219 * config/riscv/riscv.md (*branch<ANYI:mode>_shiftedarith_equals_zero):
2221 (*branch<ANYI:mode>_shiftedmask_equals_zero): New pattern.
2223 2022-11-18 Philipp Tomsich <philipp.tomsich@vrull.eu>
2225 * config/riscv/bitmanip.md (*bsetidisi): New pattern.
2227 2022-11-18 Richard Purdie <richard.purdie@linuxfoundation.org>
2229 * doc/invoke.texi: Document prefix-maps don't affect directives.
2231 2022-11-18 Andrew Pinski <apinski@marvell.com>
2233 PR middle-end/107705
2234 * function.cc (aggregate_value_p): Return 0 if
2235 the function type was an error operand.
2237 2022-11-18 Andrew Pinski <apinski@marvell.com>
2242 * gimplify.cc (gimplify_compound_lval): Return GS_ERROR
2243 if gimplify_expr had return GS_ERROR.
2244 (gimplify_call_expr): Likewise.
2246 2022-11-18 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
2248 * config/aarch64/atomics.md (*aarch64_atomic_load<ALLX:mode>_rcpc_sext):
2249 Use <GPI:w> for destination format.
2250 * config/aarch64/iterators.md (w_sz): Delete.
2252 2022-11-18 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
2254 * config/aarch64/aarch64.h (TARGET_RCPC2): Define.
2255 * config/aarch64/atomics.md (*aarch64_atomic_load<ALLX:mode>_rcpc_zext):
2256 Adjust output template.
2257 (*aarch64_atomic_load<ALLX:mode>_rcpc_sex): Guard on TARGET_RCPC2.
2258 Adjust output template.
2259 * config/aarch64/iterators.md (w_sz): New mode attr.
2261 2022-11-18 Richard Biener <rguenther@suse.de>
2263 PR tree-optimization/107647
2264 * tree-vect-slp-patterns.cc (addsub_pattern::recognize): Only
2265 allow FMA generation with -ffp-contract=fast for FP types.
2266 (complex_mul_pattern::matches): Likewise.
2268 2022-11-18 Jinyang He <hejinyang@loongson.cn>
2271 * config/loongarch/sync.md
2272 (atomic_cas_value_exchange_7_<mode>): New define_insn.
2273 (atomic_exchange): Use atomic_cas_value_exchange_7_si instead of
2274 atomic_cas_value_cmp_and_7_si.
2276 2022-11-17 Andrew Pinski <apinski@marvell.com>
2278 PR middle-end/107734
2279 * match.pd (perm + vector op pattern): Clear the sbitmap before
2282 2022-11-17 Aldy Hernandez <aldyh@redhat.com>
2284 PR tree-optimization/107732
2285 * range-op-float.cc (foperator_abs::op1_range): Early exit when
2286 result is undefined.
2288 2022-11-17 Philipp Tomsich <philipp.tomsich@vrull.eu>
2290 * config/riscv/bitmanip.md (*bclri<mode>_nottwobits): New pattern.
2291 (*bclridisi_nottwobits): New pattern, handling the sign-bit.
2292 * config/riscv/predicates.md (const_nottwobits_operand):
2295 2022-11-17 Philipp Tomsich <philipp.tomsich@vrull.eu>
2297 * config/riscv/bitmanip.md: Add a splitter to generate
2298 polarity-reversed masks from a set bit using bexti + addi.
2300 2022-11-17 mtsamis <manolis.tsamis@vrull.eu>
2302 * config/riscv/riscv.cc (struct machine_function): Add array to store
2303 register wrapping information.
2304 (riscv_for_each_saved_reg): Skip registers that are wrapped separetely.
2305 (riscv_get_separate_components): New function.
2306 (riscv_components_for_bb): Likewise.
2307 (riscv_disqualify_components): Likewise.
2308 (riscv_process_components): Likewise.
2309 (riscv_emit_prologue_components): Likewise.
2310 (riscv_emit_epilogue_components): Likewise.
2311 (riscv_set_handled_components): Likewise.
2312 (TARGET_SHRINK_WRAP_GET_SEPARATE_COMPONENTS): Define.
2313 (TARGET_SHRINK_WRAP_COMPONENTS_FOR_BB): Likewise.
2314 (TARGET_SHRINK_WRAP_DISQUALIFY_COMPONENTS): Likewise.
2315 (TARGET_SHRINK_WRAP_EMIT_PROLOGUE_COMPONENTS): Likewise.
2316 (TARGET_SHRINK_WRAP_EMIT_EPILOGUE_COMPONENTS): Likewise.
2317 (TARGET_SHRINK_WRAP_SET_HANDLED_COMPONENTS): Likewise.
2319 2022-11-17 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
2321 * config/aarch64/atomics.md (*aarch64_atomic_load<ALLX:mode>_rcpc_zext):
2322 Add mode size check to condition.
2323 (*aarch64_atomic_load<ALLX:mode>_rcpc_sext): Likewise.
2325 2022-11-17 Aldy Hernandez <aldyh@redhat.com>
2327 PR tree-optimization/68097
2328 * gimple-fold.cc (gimple_stmt_nonnegative_warnv_p): Call
2329 range_of_stmt for floats.
2331 2022-11-17 Lili Cui <lili.cui@intel.com>
2333 * config/i386/x86-tune.def
2334 (X86_TUNE_AVX256_MOVE_BY_PIECES): Add alderlake.
2335 (X86_TUNE_AVX256_STORE_BY_PIECES): Ditto.
2337 2022-11-17 Tamar Christina <tamar.christina@arm.com>
2339 PR tree-optimization/107717
2340 * match.pd: Ensure same SSA_NAME.
2342 2022-11-17 Tamar Christina <tamar.christina@arm.com>
2344 * match.pd: Replace GET_MODE_WIDER_MODE with
2347 2022-11-17 Aldy Hernandez <aldyh@redhat.com>
2349 * range-op-float.cc (range_operator_float::fold_range): Make check
2350 for maybe_isnan more readable.
2352 2022-11-17 Kewen Lin <linkw@linux.ibm.com>
2354 * gimple-fold.cc (gimple_fold_partial_load_store_mem_ref): Use
2355 tree_fits_shwi_p rather than tree_fits_uhwi_p as bias is signed.
2357 2022-11-17 Jia-Wei Chen <jiawei@iscas.ac.cn>
2359 * config/riscv/riscv.cc (riscv_expand_epilogue):
2360 Do not emit useless add sp, sp, 0 instrutions.
2362 2022-11-17 Jeff Law <jlaw@ventanamicro.com>
2364 * config/sh/sh_treg_combine.cc: Include tm-preds.h.
2365 (t_reg_operand): Remove bogus prototype.
2366 (negt_reg_operand): Likewise.
2368 2022-11-16 Marek Polacek <polacek@redhat.com>
2371 * doc/invoke.texi: Document -Winvalid-constexpr.
2373 2022-11-16 Philipp Tomsich <philipp.tomsich@vrull.eu>
2375 * config/riscv/bitmanip.md: Add split covering
2376 "(a & (1 << BIT_NO)) ? 0 : 1".
2378 2022-11-16 Philipp Tomsich <philipp.tomsich@vrull.eu>
2380 * config/riscv/bitmanip.md (*bext<mode>): Add an additional
2381 pattern that allows the 3rd argument to zero_extract to be
2382 an Xmode register operand.
2384 2022-11-16 Richard Biener <rguenther@suse.de>
2386 PR tree-optimization/107686
2387 * tree-ssa-forwprop.cc (optimize_vector_load): Restrict
2388 VEC_UNPACK support to integral typed bitfield refs.
2390 2022-11-16 Richard Biener <rguenther@suse.de>
2392 PR middle-end/107679
2393 * tree-into-ssa.cc (maybe_register_def): Use new temporary if
2394 we require an uninitialized value for a parameter decl.
2396 2022-11-16 Alexander Monakov <amonakov@ispras.ru>
2399 * doc/invoke.texi (-mrelax-cmpxchg-loop): Reword description.
2401 2022-11-16 Alexander Monakov <amonakov@ispras.ru>
2404 * config/i386/znver.md: (znver1_fp_op_mul): Correct cycles in
2406 (znver1_fp_op_mul_load): Ditto.
2407 (znver1_mmx_mul): Ditto.
2408 (znver1_mmx_load): Ditto.
2409 (znver1_ssemul_ss_ps): Ditto.
2410 (znver1_ssemul_ss_ps_load): Ditto.
2411 (znver1_ssemul_avx256_ps): Ditto.
2412 (znver1_ssemul_avx256_ps_load): Ditto.
2413 (znver1_ssemul_sd_pd): Ditto.
2414 (znver1_ssemul_sd_pd_load): Ditto.
2415 (znver2_ssemul_sd_pd): Ditto.
2416 (znver2_ssemul_sd_pd_load): Ditto.
2417 (znver1_ssemul_avx256_pd): Ditto.
2418 (znver1_ssemul_avx256_pd_load): Ditto.
2419 (znver1_sseimul): Ditto.
2420 (znver1_sseimul_avx256): Ditto.
2421 (znver1_sseimul_load): Ditto.
2422 (znver1_sseimul_avx256_load): Ditto.
2423 (znver1_sseimul_di): Ditto.
2424 (znver1_sseimul_load_di): Ditto.
2426 2022-11-16 Alexander Monakov <amonakov@ispras.ru>
2429 * config/i386/znver.md (znver1_fdiv): New automaton.
2430 (znver1-fdiv): New unit.
2431 (znver1_fp_op_div): Correct unit and cycles in the reservation.
2432 (znver1_fp_op_div_load): Ditto.
2433 (znver1_fp_op_idiv_load): Ditto.
2434 (znver2_fp_op_idiv_load): Ditto.
2435 (znver1_ssediv_ss_ps): Ditto.
2436 (znver1_ssediv_ss_ps_load): Ditto.
2437 (znver1_ssediv_sd_pd): Ditto.
2438 (znver1_ssediv_sd_pd_load): Ditto.
2439 (znver1_ssediv_avx256_ps): Ditto.
2440 (znver1_ssediv_avx256_ps_load): Ditto.
2441 (znver1_ssediv_avx256_pd): Ditto.
2442 (znver1_ssediv_avx256_pd_load): Ditto.
2444 2022-11-16 Tobias Burnus <tobias@codesourcery.com>
2445 Andrew Stubbs <ams@codesourcery.com>
2447 * config/gcn/gcn-builtins.def (KERNARG_PTR): Add.
2448 * config/gcn/gcn.cc (gcn_init_builtin_types): Change siptr_type_node,
2449 sfptr_type_node and voidptr_type_node from FLAT to ADDR_SPACE_DEFAULT.
2450 (gcn_expand_builtin_1): Handle GCN_BUILTIN_KERNARG_PTR.
2451 (gcn_oacc_dim_size): Return in ADDR_SPACE_FLAT.
2453 2022-11-16 David Malcolm <dmalcolm@redhat.com>
2455 * Makefile.in (ANALYZER_OBJS): Add analyzer/checker-event.o.
2457 2022-11-16 Max Filippov <jcmvbkbc@gmail.com>
2460 * config/m68k/predicates.md (symbolic_operand): Return false
2461 when UNSPEC is under the CONST node.
2463 2022-11-16 Christoph Müllner <christoph.muellner@vrull.eu>
2465 * doc/invoke.texi: Fix PRU/RISC-V option list formatting.
2467 2022-11-16 Tobias Burnus <tobias@codesourcery.com>
2469 * config/nvptx/mkoffload.cc (process): Recognize '$nohost$...'
2470 besides tailing '$nohost' as being for reverse offload.
2472 2022-11-16 Jakub Jelinek <jakub@redhat.com>
2474 PR tree-optimization/107668
2475 * range-op-float.cc (float_binary_op_range_finish): Set VARYING
2476 also when r is UNDEFINED.
2478 2022-11-16 Kewen Lin <linkw@linux.ibm.com>
2480 PR rtl-optimization/90259
2481 * function.cc (rest_of_handle_thread_prologue_and_epilogue): Add
2482 parameter fun, and call function purge_all_dead_edges.
2483 (pass_thread_prologue_and_epilogue::execute): Name unamed parameter
2484 as fun, and use it for rest_of_handle_thread_prologue_and_epilogue.
2486 2022-11-15 Philipp Tomsich <philipp.tomsich@vrull.eu>
2488 * config/riscv/riscv-builtins.cc (struct riscv_builtin_description):
2489 add the pause machine-dependent builtin with no result and no
2490 arguments; mark it as always present (pause is a true hint
2491 that encodes into a fence-insn, if not supported with the new
2493 * config/riscv/riscv-ftypes.def: Add type for void -> void.
2494 * config/riscv/riscv.md (riscv_pause): Add risc_pause and
2496 * doc/extend.texi: Document __builtin_riscv_pause.
2497 * optabs.cc (maybe_gen_insn): Allow nops == 0 (void -> void).
2499 2022-11-15 David Faust <david.faust@oracle.com>
2501 * config/bpf/bpf.cc (bpf_expand_builtin): Avoid use of uninitialized
2502 variable in error case.
2504 2022-11-15 David Malcolm <dmalcolm@redhat.com>
2507 * doc/invoke.texi (Static Analyzer Options): Add
2508 -Wanalyzer-fd-phase-mismatch and -Wanalyzer-fd-type-mismatch. Add
2509 "socket", "bind", "listen", "accept", and "connect" to the list of
2510 functions known to the analyzer.
2512 2022-11-15 David Malcolm <dmalcolm@redhat.com>
2515 * Makefile.in (ANALYZER_OBJS): Add analyzer/analyzer-language.o.
2516 (GTFILES): Add analyzer/analyzer-language.cc.
2517 * doc/analyzer.texi: Document __analyzer_dump_named_constant.
2519 2022-11-15 Lewis Hyatt <lhyatt@gmail.com>
2521 * json.h (class string): Add M_LEN member to store the length of
2522 the data. Add constructor taking an explicit length.
2523 * json.cc (string::string): Implement the new constructor.
2524 (string::print): Support printing strings that are not
2525 null-terminated. Escape embdedded null bytes on output.
2526 (test_writing_strings): Test the new null-byte-related features of
2529 2022-11-15 Lewis Hyatt <lhyatt@gmail.com>
2531 * diagnostic.cc (diagnostic_get_location_text): Use
2532 special_fname_builtin () rather than a hardcoded string (which was
2533 also incorrectly left untranslated previously.)
2534 * input.cc (special_fname_builtin): New function.
2535 (expand_location_1): Use special_fname_builtin () rather than a
2537 (test_builtins): Likewise.
2538 * input.h (special_fname_builtin): Declare.
2540 2022-11-15 Lewis Hyatt <lhyatt@gmail.com>
2542 * tree-diagnostic.cc (maybe_unwind_expanded_macro_loc): Handle ad-hoc
2543 location in return value of linemap_resolve_location().
2545 2022-11-15 Jonathan Wakely <jwakely@redhat.com>
2547 * doc/cpp.texi (Pragmas): Use @item and @itemx for region
2550 2022-11-15 Maciej W. Rozycki <macro@embecosm.com>
2552 * ira-color.cc (assign_hard_reg): Remove duplicate `memset' over
2555 2022-11-15 Andre Vieira <andre.simoesdiasvieira@arm.com>
2557 * config/aarch64/atomics.md
2558 (*aarch64_atomic_load<ALLX:mode>_rcpc_zext): New pattern.
2559 (*aarch64_atomic_load<ALLX:mode>_rcpc_sext): New pattern.
2561 2022-11-15 Andre Vieira <andre.simoesdiasvieira@arm.com>
2562 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
2564 * config/aarch64/aarch64.h (AARCH64_ISA_RCPC): New Macro.
2565 (TARGET_RCPC): New Macro.
2566 * config/aarch64/atomics.md (atomic_load<mode>): Change into an expand.
2567 (aarch64_atomic_load<mode>_rcpc): New define_insn for ldapr.
2568 (aarch64_atomic_load<mode>): Rename of old define_insn for ldar.
2569 * config/aarch64/iterators.md (UNSPEC_LDAP): New unspec enum value.
2570 * doc/invoke.texi (rcpc): Ammend documentation to mention the effects
2573 2022-11-15 Hongyu Wang <hongyu.wang@intel.com>
2576 * match.pd: New perm + vector op patterns for int and fp vector.
2578 2022-11-15 Andrew Pinski <apinski@marvell.com>
2580 * doc/invoke.texi: Remove Score option section.
2582 2022-11-15 Andrew Pinski <apinski@marvell.com>
2584 * doc/extend.texi: Remove picoChip builtin section.
2585 * doc/invoke.texi: Remove picoChip option section.
2587 2022-11-15 Andrew Pinski <apinski@marvell.com>
2589 * doc/extend.texi: Remove MeP documentation.
2590 * doc/invoke.texi: Remove MeP Options documentation.
2592 2022-11-15 Andrew Pinski <apinski@marvell.com>
2594 * doc/invoke.texi: Fix opindex for mcall-aixdesc and mcall-openbsd.
2596 2022-11-14 Philipp Tomsich <philipp.tomsich@vrull.eu>
2599 2022-11-14 Philipp Tomsich <philipp.tomsich@vrull.eu>
2601 * config/riscv/riscv-cores.def (RISCV_TUNE): Add ventana-vt1.
2602 (RISCV_CORE): Ditto.
2603 * config/riscv/riscv-opts.h (enum riscv_microarchitecture_type): Ditto.
2604 * config/riscv/riscv.cc: Add tune_info for ventana-vt1.
2605 * doc/invoke.texi: Document -mcpu= and -mtune with ventana-vt1.
2607 2022-11-14 Philipp Tomsich <philipp.tomsich@vrull.eu>
2610 2022-11-14 Philipp Tomsich <philipp.tomsich@vrull.eu>
2612 * config/riscv/riscv.cc (enum riscv_fusion_pairs): Add symbolic
2613 constants to identify supported fusion patterns.
2614 (struct riscv_tune_param): Add fusible_op field.
2615 (riscv_macro_fusion_p): Implement.
2616 (riscv_fusion_enabled_p): Implement.
2617 (riscv_macro_fusion_pair_p): Implement and recognize fusible
2618 idioms for Ventana VT1.
2619 (TARGET_SCHED_MACRO_FUSION_P): Point to riscv_macro_fusion_p.
2620 (TARGET_SCHED_MACRO_FUSION_PAIR_P): Point to
2621 riscv_macro_fusion_pair_p.
2623 2022-11-14 Maciej W. Rozycki <macro@embecosm.com>
2625 * ira-build.cc (create_insn_allocnos): Fix documentation.
2627 2022-11-14 Tamar Christina <tamar.christina@arm.com>
2629 * match.pd: Remove returns.
2631 2022-11-14 Philipp Tomsich <philipp.tomsich@vrull.eu>
2633 * config/riscv/bitmanip.md (orcb<mode>2): Add orc.b as an
2635 * config/riscv/riscv.md: Add UNSPEC_ORC_B.
2637 2022-11-14 Tamar Christina <tamar.christina@arm.com>
2639 * target.def: Fix typo.
2641 2022-11-14 Philipp Tomsich <philipp.tomsich@vrull.eu>
2643 * config/riscv/riscv.cc (enum riscv_fusion_pairs): Add symbolic
2644 constants to identify supported fusion patterns.
2645 (struct riscv_tune_param): Add fusible_op field.
2646 (riscv_macro_fusion_p): Implement.
2647 (riscv_fusion_enabled_p): Implement.
2648 (riscv_macro_fusion_pair_p): Implement and recognize fusible
2649 idioms for Ventana VT1.
2650 (TARGET_SCHED_MACRO_FUSION_P): Point to riscv_macro_fusion_p.
2651 (TARGET_SCHED_MACRO_FUSION_PAIR_P): Point to
2652 riscv_macro_fusion_pair_p.
2654 2022-11-14 Philipp Tomsich <philipp.tomsich@vrull.eu>
2656 * config/riscv/riscv-cores.def (RISCV_TUNE): Add ventana-vt1.
2657 (RISCV_CORE): Ditto.
2658 * config/riscv/riscv-opts.h (enum riscv_microarchitecture_type): Ditto.
2659 * config/riscv/riscv.cc: Add tune_info for ventana-vt1.
2660 * doc/invoke.texi: Document -mcpu= and -mtune with ventana-vt1.
2662 2022-11-14 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
2664 * config/aarch64/aarch64-option-extensions.def (cssc): Define.
2665 * config/aarch64/aarch64.h (AARCH64_ISA_CSSC): Define.
2666 (TARGET_CSSC): Likewise.
2667 * config/aarch64/aarch64.md (*aarch64_abs<mode>2_cssc_ins): New define_insn.
2668 (abs<mode>2): Adjust for the above.
2669 (aarch64_umax<mode>3_insn): New define_insn.
2670 (umax<mode>3): Adjust for the above.
2671 (*aarch64_popcount<mode>2_cssc_insn): New define_insn.
2672 (popcount<mode>2): Adjust for the above.
2673 (<optab><mode>3): New define_insn.
2674 * config/aarch64/constraints.md (Usm): Define.
2676 * doc/invoke.texi (AArch64 options): Document +cssc.
2677 * config/aarch64/iterators.md (MAXMIN_NOUMAX): New code iterator.
2678 * config/aarch64/predicates.md (aarch64_sminmax_immediate): Define.
2679 (aarch64_sminmax_operand): Likewise.
2680 (aarch64_uminmax_immediate): Likewise.
2681 (aarch64_uminmax_operand): Likewise.
2683 2022-11-14 Tamar Christina <tamar.christina@arm.com>
2685 * config/aarch64/aarch64-sve2.md (@aarch64_bitmask_udiv<mode>3): New.
2687 2022-11-14 Tamar Christina <tamar.christina@arm.com>
2689 * config/aarch64/aarch64-simd.md (@aarch64_bitmask_udiv<mode>3): New.
2690 * config/aarch64/aarch64.cc (aarch64_vectorize_can_special_div_by_constant): New.
2692 2022-11-14 Tamar Christina <tamar.christina@arm.com>
2694 * expmed.h (expand_divmod): Pass tree operands down in addition to RTX.
2695 * expmed.cc (expand_divmod): Likewise.
2696 * explow.cc (round_push, align_dynamic_address): Likewise.
2697 * expr.cc (force_operand, expand_expr_divmod): Likewise.
2698 * optabs.cc (expand_doubleword_mod, expand_doubleword_divmod):
2700 * target.h: Include tree-core.
2701 * target.def (can_special_div_by_const): New.
2702 * targhooks.cc (default_can_special_div_by_const): New.
2703 * targhooks.h (default_can_special_div_by_const): New.
2704 * tree-vect-generic.cc (expand_vector_operation): Use it.
2705 * doc/tm.texi.in: Document it.
2706 * doc/tm.texi: Regenerate.
2707 * tree-vect-patterns.cc (vect_recog_divmod_pattern): Check for support.
2708 * tree-vect-stmts.cc (vectorizable_operation): Likewise.
2710 2022-11-14 Tamar Christina <tamar.christina@arm.com>
2712 * generic-match-head.cc: Include langooks.
2713 * gimple-match-head.cc: Likewise.
2714 * match.pd: Add fneg/fadd rule.
2716 2022-11-14 Srinath Parvathaneni <srinath.parvathaneni@arm.com>
2718 * config/arm/arm-cpus.in (cortex-x1c): Define new CPU.
2719 * config/arm/arm-tables.opt: Regenerate.
2720 * config/arm/arm-tune.md: Likewise.
2721 * doc/invoke.texi: Document Cortex-X1C CPU.
2723 2022-11-14 Srinath Parvathaneni <srinath.parvathaneni@arm.com>
2725 * config/aarch64/aarch64-cores.def (AARCH64_CORE): Add Cortex-X3
2727 * config/aarch64/aarch64-tune.md: Regenerate.
2728 * doc/invoke.texi: Document Cortex-X3 CPU.
2730 2022-11-14 Andrew Pinski <apinski@marvell.com>
2732 * doc/invoke.texi: Fix @opindex
2735 2022-11-14 Andrew Pinski <apinski@marvell.com>
2737 * doc/invoke.texi: Remove the front - from
2740 2022-11-14 Philipp Tomsich <philipp.tomsich@vrull.eu>
2742 * config/aarch64/aarch64-cores.def (AARCH64_CORE): Add ampere1a.
2743 * config/aarch64/aarch64-cost-tables.h: Add ampere1a_extra_costs.
2744 * config/aarch64/aarch64-fusion-pairs.def (AARCH64_FUSION_PAIR):
2745 Define a new fusion pair for A+B+1/A-B-1 (i.e., add/subtract two
2746 registers and then +1/-1).
2747 * config/aarch64/aarch64-tune.md: Regenerate.
2748 * config/aarch64/aarch64.cc (aarch_macro_fusion_pair_p): Implement
2749 idiom-matcher for the new fusion pair.
2750 * doc/invoke.texi: Add ampere1a.
2752 2022-11-14 Richard Biener <rguenther@suse.de>
2754 * match.pd: Remove duplicates.
2756 2022-11-14 Martin Liska <mliska@suse.cz>
2758 * doc/contrib.texi: Port from Sphinx.
2759 * doc/cpp.texi: Likewise.
2760 * doc/install.texi: Likewise.
2761 * doc/invoke.texi: Likewise.
2763 2022-11-14 Martin Liska <mliska@suse.cz>
2766 2022-11-14 Martin Liska <mliska@suse.cz>
2768 * doc/cpp/character-sets.rst: New file.
2769 * doc/cpp/conditional-syntax.rst: New file.
2770 * doc/cpp/conditional-uses.rst: New file.
2771 * doc/cpp/conditionals.rst: New file.
2772 * doc/cpp/conf.py: New file.
2773 * doc/cpp/copyright.rst: New file.
2774 * doc/cpp/deleted-code.rst: New file.
2775 * doc/cpp/diagnostics.rst: New file.
2776 * doc/cpp/environment-variables.rst: New file.
2777 * doc/cpp/gnu-free-documentation-license.rst: New file.
2778 * doc/cpp/header-files.rst: New file.
2779 * doc/cpp/header-files/alternatives-to-wrapper-ifndef.rst: New file.
2780 * doc/cpp/header-files/computed-includes.rst: New file.
2781 * doc/cpp/header-files/include-operation.rst: New file.
2782 * doc/cpp/header-files/include-syntax.rst: New file.
2783 * doc/cpp/header-files/once-only-headers.rst: New file.
2784 * doc/cpp/header-files/search-path.rst: New file.
2785 * doc/cpp/header-files/system-headers.rst: New file.
2786 * doc/cpp/header-files/wrapper-headers.rst: New file.
2787 * doc/cpp/implementation-defined-behavior.rst: New file.
2788 * doc/cpp/implementation-details.rst: New file.
2789 * doc/cpp/implementation-limits.rst: New file.
2790 * doc/cpp/index.rst: New file.
2791 * doc/cpp/indices-and-tables.rst: New file.
2792 * doc/cpp/initial-processing.rst: New file.
2793 * doc/cpp/invocation.rst: New file.
2794 * doc/cpp/line-control.rst: New file.
2795 * doc/cpp/macros.rst: New file.
2796 * doc/cpp/macros/concatenation.rst: New file.
2797 * doc/cpp/macros/directives-within-macro-arguments.rst: New file.
2798 * doc/cpp/macros/function-like-macros.rst: New file.
2799 * doc/cpp/macros/macro-arguments.rst: New file.
2800 * doc/cpp/macros/macro-pitfalls.rst: New file.
2801 * doc/cpp/macros/object-like-macros.rst: New file.
2802 * doc/cpp/macros/predefined-macros.rst: New file.
2803 * doc/cpp/macros/stringizing.rst: New file.
2804 * doc/cpp/macros/undefining-and-redefining-macros.rst: New file.
2805 * doc/cpp/macros/variadic-macros.rst: New file.
2806 * doc/cpp/obsolete-features.rst: New file.
2807 * doc/cpp/other-directives.rst: New file.
2808 * doc/cpp/overview.rst: New file.
2809 * doc/cpp/pragmas.rst: New file.
2810 * doc/cpp/preprocessor-output.rst: New file.
2811 * doc/cpp/the-preprocessing-language.rst: New file.
2812 * doc/cpp/tokenization.rst: New file.
2813 * doc/cpp/traditional-lexical-analysis.rst: New file.
2814 * doc/cpp/traditional-macros.rst: New file.
2815 * doc/cpp/traditional-miscellany.rst: New file.
2816 * doc/cpp/traditional-mode.rst: New file.
2817 * doc/cpp/traditional-warnings.rst: New file.
2818 * doc/cppinternals/conf.py: New file.
2819 * doc/cppinternals/copyright.rst: New file.
2820 * doc/cppinternals/cppinternals.rst: New file.
2821 * doc/cppinternals/cpplib.rst: New file.
2822 * doc/cppinternals/files.rst: New file.
2823 * doc/cppinternals/index.rst: New file.
2824 * doc/cppinternals/indices-and-tables.rst: New file.
2825 * doc/cppinternals/internal-representation-of-macros.rst: New file.
2826 * doc/cppinternals/just-which-line-number-anyway.rst: New file.
2827 * doc/cppinternals/lexing-a-line.rst: New file.
2828 * doc/cppinternals/lexing-a-token.rst: New file.
2829 * doc/cppinternals/looking-for-a-function-like-macros-opening-parenthesis.rst: New file.
2830 * doc/cppinternals/macro-expansion-overview.rst: New file.
2831 * doc/cppinternals/marking-tokens-ineligible-for-future-expansion.rst: New file.
2832 * doc/cppinternals/multiple-include-optimization.rst: New file.
2833 * doc/cppinternals/overview.rst: New file.
2834 * doc/cppinternals/representation-of-line-numbers.rst: New file.
2835 * doc/cppinternals/scanning-the-replacement-list-for-macros-to-expand.rst: New file.
2836 * doc/gcc/binary-compatibility.rst: New file.
2837 * doc/gcc/c++-implementation-defined-behavior.rst: New file.
2838 * doc/gcc/c-implementation-defined-behavior.rst: New file.
2839 * doc/gcc/c-implementation-defined-behavior/architecture.rst: New file.
2840 * doc/gcc/c-implementation-defined-behavior/arrays-and-pointers.rst: New file.
2841 * doc/gcc/c-implementation-defined-behavior/characters.rst: New file.
2842 * doc/gcc/c-implementation-defined-behavior/declarators.rst: New file.
2843 * doc/gcc/c-implementation-defined-behavior/environment.rst: New file.
2844 * doc/gcc/c-implementation-defined-behavior/floating-point.rst: New file.
2845 * doc/gcc/c-implementation-defined-behavior/hints.rst: New file.
2846 * doc/gcc/c-implementation-defined-behavior/identifiers.rst: New file.
2847 * doc/gcc/c-implementation-defined-behavior/integers.rst: New file.
2848 * doc/gcc/c-implementation-defined-behavior/library-functions.rst: New file.
2849 * doc/gcc/c-implementation-defined-behavior/locale-specific-behavior.rst: New file.
2850 * doc/gcc/c-implementation-defined-behavior/preprocessing-directives.rst: New file.
2851 * doc/gcc/c-implementation-defined-behavior/qualifiers.rst: New file.
2852 * doc/gcc/c-implementation-defined-behavior/statements.rst: New file.
2853 * doc/gcc/c-implementation-defined-behavior/structures-unions-enumerations-and-bit-fields.rst:
2855 * doc/gcc/c-implementation-defined-behavior/translation.rst: New file.
2856 * doc/gcc/conditionally-supported-behavior.rst: New file.
2857 * doc/gcc/conf.py: New file.
2858 * doc/gcc/contributing-to-gcc-development.rst: New file.
2859 * doc/gcc/contributors-to-gcc.rst: New file.
2860 * doc/gcc/copyright.rst: New file.
2861 * doc/gcc/exception-handling.rst: New file.
2862 * doc/gcc/extensions-to-the-c++-language.rst: New file.
2863 * doc/gcc/extensions-to-the-c++-language/backwards-compatibility.rst: New file.
2864 * doc/gcc/extensions-to-the-c++-language/c++-concepts.rst: New file.
2865 * doc/gcc/extensions-to-the-c++-language/c++-interface-and-implementation-pragmas.rst:
2867 * doc/gcc/extensions-to-the-c++-language/c++-specific-variable-function-and-type-attributes.rst:
2869 * doc/gcc/extensions-to-the-c++-language/deprecated-features.rst: New file.
2870 * doc/gcc/extensions-to-the-c++-language/extracting-the-function-pointer-from-a-bound-pointer-to-member-function.rst:
2872 * doc/gcc/extensions-to-the-c++-language/function-multiversioning.rst: New file.
2873 * doc/gcc/extensions-to-the-c++-language/restricting-pointer-aliasing.rst: New file.
2874 * doc/gcc/extensions-to-the-c++-language/type-traits.rst: New file.
2875 * doc/gcc/extensions-to-the-c++-language/vague-linkage.rst: New file.
2876 * doc/gcc/extensions-to-the-c++-language/when-is-a-volatile-c++-object-accessed.rst:
2878 * doc/gcc/extensions-to-the-c++-language/wheres-the-template.rst: New file.
2879 * doc/gcc/extensions-to-the-c-language-family.rst: New file.
2880 * doc/gcc/extensions-to-the-c-language-family/128-bit-integers.rst: New file.
2881 * doc/gcc/extensions-to-the-c-language-family/additional-floating-types.rst: New file.
2882 * doc/gcc/extensions-to-the-c-language-family/alternate-keywords.rst: New file.
2883 * doc/gcc/extensions-to-the-c-language-family/an-inline-function-is-as-fast-as-a-macro.rst:
2885 * doc/gcc/extensions-to-the-c-language-family/arithmetic-on-void-and-function-pointers.rst:
2887 * doc/gcc/extensions-to-the-c-language-family/arrays-of-length-zero.rst: New file.
2888 * doc/gcc/extensions-to-the-c-language-family/arrays-of-variable-length.rst: New file.
2889 * doc/gcc/extensions-to-the-c-language-family/attribute-syntax.rst: New file.
2890 * doc/gcc/extensions-to-the-c-language-family/binary-constants-using-the-0b-prefix.rst:
2892 * doc/gcc/extensions-to-the-c-language-family/built-in-functions-for-memory-model-aware-atomic-operations.rst:
2894 * doc/gcc/extensions-to-the-c-language-family/built-in-functions-to-perform-arithmetic-with-overflow-checking.rst:
2896 * doc/gcc/extensions-to-the-c-language-family/c++-style-comments.rst: New file.
2897 * doc/gcc/extensions-to-the-c-language-family/case-ranges.rst: New file.
2898 * doc/gcc/extensions-to-the-c-language-family/cast-to-a-union-type.rst: New file.
2899 * doc/gcc/extensions-to-the-c-language-family/complex-numbers.rst: New file.
2900 * doc/gcc/extensions-to-the-c-language-family/compound-literals.rst: New file.
2901 * doc/gcc/extensions-to-the-c-language-family/conditionals-with-omitted-operands.rst:
2903 * doc/gcc/extensions-to-the-c-language-family/constructing-function-calls.rst: New file.
2904 * doc/gcc/extensions-to-the-c-language-family/decimal-floating-types.rst: New file.
2905 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions.rst:
2907 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/aarch64-function-attributes.rst:
2909 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/amd-gcn-function-attributes.rst:
2911 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/arc-function-attributes.rst:
2913 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/arm-function-attributes.rst:
2915 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/avr-function-attributes.rst:
2917 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/blackfin-function-attributes.rst:
2919 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/bpf-function-attributes.rst:
2921 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/c-sky-function-attributes.rst:
2923 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/common-function-attributes.rst:
2925 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/epiphany-function-attributes.rst:
2927 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/h8-300-function-attributes.rst:
2929 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/ia-64-function-attributes.rst:
2931 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/m32c-function-attributes.rst:
2933 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/m32r-d-function-attributes.rst:
2935 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/m68k-function-attributes.rst:
2937 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/mcore-function-attributes.rst:
2939 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/mep-function-attributes.rst:
2941 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/microblaze-function-attributes.rst:
2943 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/microsoft-windows-function-attributes.rst:
2945 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/mips-function-attributes.rst:
2947 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/msp430-function-attributes.rst:
2949 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/nds32-function-attributes.rst:
2951 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/nios-ii-function-attributes.rst:
2953 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/nvidia-ptx-function-attributes.rst:
2955 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/powerpc-function-attributes.rst:
2957 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/risc-v-function-attributes.rst:
2959 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/rl78-function-attributes.rst:
2961 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/rx-function-attributes.rst:
2963 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/s-390-function-attributes.rst:
2965 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/sh-function-attributes.rst:
2967 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/symbian-os-function-attributes.rst:
2969 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/v850-function-attributes.rst:
2971 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/visium-function-attributes.rst:
2973 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/x86-function-attributes.rst:
2975 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/xstormy16-function-attributes.rst:
2977 * doc/gcc/extensions-to-the-c-language-family/designated-initializers.rst: New file.
2978 * doc/gcc/extensions-to-the-c-language-family/determining-the-alignment-of-functions-types-or-variables.rst:
2980 * doc/gcc/extensions-to-the-c-language-family/dollar-signs-in-identifier-names.rst:
2982 * doc/gcc/extensions-to-the-c-language-family/double-word-integers.rst: New file.
2983 * doc/gcc/extensions-to-the-c-language-family/enumerator-attributes.rst: New file.
2984 * doc/gcc/extensions-to-the-c-language-family/fixed-point-types.rst: New file.
2985 * doc/gcc/extensions-to-the-c-language-family/format-checks-specific-to-particular-target-machines.rst:
2987 * doc/gcc/extensions-to-the-c-language-family/function-names-as-strings.rst: New file.
2988 * doc/gcc/extensions-to-the-c-language-family/getting-the-return-or-frame-address-of-a-function.rst:
2990 * doc/gcc/extensions-to-the-c-language-family/half-precision-floating-point.rst: New file.
2991 * doc/gcc/extensions-to-the-c-language-family/hex-floats.rst: New file.
2992 * doc/gcc/extensions-to-the-c-language-family/how-to-use-inline-assembly-language-in-c-code.rst:
2994 * doc/gcc/extensions-to-the-c-language-family/incomplete-enum-types.rst: New file.
2995 * doc/gcc/extensions-to-the-c-language-family/label-attributes.rst: New file.
2996 * doc/gcc/extensions-to-the-c-language-family/labels-as-values.rst: New file.
2997 * doc/gcc/extensions-to-the-c-language-family/legacy-sync-built-in-functions-for-atomic-memory-access.rst:
2999 * doc/gcc/extensions-to-the-c-language-family/locally-declared-labels.rst: New file.
3000 * doc/gcc/extensions-to-the-c-language-family/macros-with-a-variable-number-of-arguments.rst:
3002 * doc/gcc/extensions-to-the-c-language-family/mixed-declarations-labels-and-code.rst:
3004 * doc/gcc/extensions-to-the-c-language-family/named-address-spaces.rst: New file.
3005 * doc/gcc/extensions-to-the-c-language-family/nested-functions.rst: New file.
3006 * doc/gcc/extensions-to-the-c-language-family/non-constant-initializers.rst: New file.
3007 * doc/gcc/extensions-to-the-c-language-family/non-lvalue-arrays-may-have-subscripts.rst:
3009 * doc/gcc/extensions-to-the-c-language-family/nonlocal-gotos.rst: New file.
3010 * doc/gcc/extensions-to-the-c-language-family/object-size-checking-built-in-functions.rst:
3012 * doc/gcc/extensions-to-the-c-language-family/other-built-in-functions-provided-by-gcc.rst:
3014 * doc/gcc/extensions-to-the-c-language-family/pointer-arguments-in-variadic-functions.rst:
3016 * doc/gcc/extensions-to-the-c-language-family/pointers-to-arrays-with-qualifiers-work-as-expected.rst:
3018 * doc/gcc/extensions-to-the-c-language-family/pragmas-accepted-by-gcc.rst: New file.
3019 * doc/gcc/extensions-to-the-c-language-family/prototypes-and-old-style-function-definitions.rst:
3021 * doc/gcc/extensions-to-the-c-language-family/referring-to-a-type-with-typeof.rst: New file.
3022 * doc/gcc/extensions-to-the-c-language-family/slightly-looser-rules-for-escaped-newlines.rst:
3024 * doc/gcc/extensions-to-the-c-language-family/specifying-attributes-of-types.rst: New file.
3025 * doc/gcc/extensions-to-the-c-language-family/specifying-attributes-of-variables.rst:
3027 * doc/gcc/extensions-to-the-c-language-family/statement-attributes.rst: New file.
3028 * doc/gcc/extensions-to-the-c-language-family/statements-and-declarations-in-expressions.rst:
3030 * doc/gcc/extensions-to-the-c-language-family/structures-with-no-members.rst: New file.
3031 * doc/gcc/extensions-to-the-c-language-family/support-for-offsetof.rst: New file.
3032 * doc/gcc/extensions-to-the-c-language-family/target-builtins.rst: New file.
3033 * doc/gcc/extensions-to-the-c-language-family/target-builtins/aarch64-built-in-functions.rst:
3035 * doc/gcc/extensions-to-the-c-language-family/target-builtins/alpha-built-in-functions.rst:
3037 * doc/gcc/extensions-to-the-c-language-family/target-builtins/altera-nios-ii-built-in-functions.rst:
3039 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arc-built-in-functions.rst:
3041 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arc-simd-built-in-functions.rst:
3043 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arm-armv8-m-security-extensions.rst:
3045 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arm-c-language-extensions-acle.rst:
3047 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arm-floating-point-status-and-control-intrinsics.rst:
3049 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arm-iwmmxt-built-in-functions.rst:
3051 * doc/gcc/extensions-to-the-c-language-family/target-builtins/avr-built-in-functions.rst:
3053 * doc/gcc/extensions-to-the-c-language-family/target-builtins/basic-powerpc-built-in-functions.rst:
3055 * doc/gcc/extensions-to-the-c-language-family/target-builtins/blackfin-built-in-functions.rst:
3057 * doc/gcc/extensions-to-the-c-language-family/target-builtins/bpf-built-in-functions.rst:
3059 * doc/gcc/extensions-to-the-c-language-family/target-builtins/fr-v-built-in-functions.rst:
3061 * doc/gcc/extensions-to-the-c-language-family/target-builtins/mips-dsp-built-in-functions.rst:
3063 * doc/gcc/extensions-to-the-c-language-family/target-builtins/mips-loongson-built-in-functions.rst:
3065 * doc/gcc/extensions-to-the-c-language-family/target-builtins/mips-paired-single-support.rst:
3067 * doc/gcc/extensions-to-the-c-language-family/target-builtins/mips-simd-architecture-msa-support.rst:
3069 * doc/gcc/extensions-to-the-c-language-family/target-builtins/msp430-built-in-functions.rst:
3071 * doc/gcc/extensions-to-the-c-language-family/target-builtins/nds32-built-in-functions.rst:
3073 * doc/gcc/extensions-to-the-c-language-family/target-builtins/other-mips-built-in-functions.rst:
3075 * doc/gcc/extensions-to-the-c-language-family/target-builtins/picochip-built-in-functions.rst:
3077 * doc/gcc/extensions-to-the-c-language-family/target-builtins/powerpc-altivec-vsx-built-in-functions.rst:
3079 * doc/gcc/extensions-to-the-c-language-family/target-builtins/powerpc-atomic-memory-operation-functions.rst:
3081 * doc/gcc/extensions-to-the-c-language-family/target-builtins/powerpc-hardware-transactional-memory-built-in-functions.rst:
3083 * doc/gcc/extensions-to-the-c-language-family/target-builtins/powerpc-matrix-multiply-assist-built-in-functions.rst:
3085 * doc/gcc/extensions-to-the-c-language-family/target-builtins/pru-built-in-functions.rst:
3087 * doc/gcc/extensions-to-the-c-language-family/target-builtins/risc-v-built-in-functions.rst:
3089 * doc/gcc/extensions-to-the-c-language-family/target-builtins/rx-built-in-functions.rst:
3091 * doc/gcc/extensions-to-the-c-language-family/target-builtins/s-390-system-z-built-in-functions.rst:
3093 * doc/gcc/extensions-to-the-c-language-family/target-builtins/sh-built-in-functions.rst:
3095 * doc/gcc/extensions-to-the-c-language-family/target-builtins/sparc-vis-built-in-functions.rst:
3097 * doc/gcc/extensions-to-the-c-language-family/target-builtins/ti-c6x-built-in-functions.rst:
3099 * doc/gcc/extensions-to-the-c-language-family/target-builtins/x86-built-in-functions.rst:
3101 * doc/gcc/extensions-to-the-c-language-family/target-builtins/x86-control-flow-protection-intrinsics.rst:
3103 * doc/gcc/extensions-to-the-c-language-family/target-builtins/x86-transactional-memory-intrinsics.rst:
3105 * doc/gcc/extensions-to-the-c-language-family/the-character-esc-in-constants.rst: New file.
3106 * doc/gcc/extensions-to-the-c-language-family/thread-local-storage.rst: New file.
3107 * doc/gcc/extensions-to-the-c-language-family/unnamed-structure-and-union-fields.rst:
3109 * doc/gcc/extensions-to-the-c-language-family/using-vector-instructions-through-built-in-functions.rst:
3111 * doc/gcc/extensions-to-the-c-language-family/when-is-a-volatile-object-accessed.rst:
3113 * doc/gcc/extensions-to-the-c-language-family/x86-specific-memory-model-extensions-for-transactional-memory.rst:
3115 * doc/gcc/funding.rst: New file.
3116 * doc/gcc/gcc-command-options.rst: New file.
3117 * doc/gcc/gcc-command-options/c++-modules.rst: New file.
3118 * doc/gcc/gcc-command-options/compiling-c++-programs.rst: New file.
3119 * doc/gcc/gcc-command-options/description.rst: New file.
3120 * doc/gcc/gcc-command-options/environment-variables-affecting-gcc.rst: New file.
3121 * doc/gcc/gcc-command-options/gcc-developer-options.rst: New file.
3122 * doc/gcc/gcc-command-options/machine-dependent-options.rst: New file.
3123 * doc/gcc/gcc-command-options/machine-dependent-options/aarch64-options.rst: New file.
3124 * doc/gcc/gcc-command-options/machine-dependent-options/adapteva-epiphany-options.rst:
3126 * doc/gcc/gcc-command-options/machine-dependent-options/amd-gcn-options.rst: New file.
3127 * doc/gcc/gcc-command-options/machine-dependent-options/arc-options.rst: New file.
3128 * doc/gcc/gcc-command-options/machine-dependent-options/arm-options.rst: New file.
3129 * doc/gcc/gcc-command-options/machine-dependent-options/avr-mmcu.rst: New file.
3130 * doc/gcc/gcc-command-options/machine-dependent-options/avr-options.rst: New file.
3131 * doc/gcc/gcc-command-options/machine-dependent-options/blackfin-options.rst: New file.
3132 * doc/gcc/gcc-command-options/machine-dependent-options/c-sky-options.rst: New file.
3133 * doc/gcc/gcc-command-options/machine-dependent-options/c6x-options.rst: New file.
3134 * doc/gcc/gcc-command-options/machine-dependent-options/cris-options.rst: New file.
3135 * doc/gcc/gcc-command-options/machine-dependent-options/darwin-options.rst: New file.
3136 * doc/gcc/gcc-command-options/machine-dependent-options/dec-alpha-options.rst: New file.
3137 * doc/gcc/gcc-command-options/machine-dependent-options/ebpf-options.rst: New file.
3138 * doc/gcc/gcc-command-options/machine-dependent-options/fr30-options.rst: New file.
3139 * doc/gcc/gcc-command-options/machine-dependent-options/frv-options.rst: New file.
3140 * doc/gcc/gcc-command-options/machine-dependent-options/ft32-options.rst: New file.
3141 * doc/gcc/gcc-command-options/machine-dependent-options/gnu-linux-options.rst: New file.
3142 * doc/gcc/gcc-command-options/machine-dependent-options/h8-300-options.rst: New file.
3143 * doc/gcc/gcc-command-options/machine-dependent-options/hppa-options.rst: New file.
3144 * doc/gcc/gcc-command-options/machine-dependent-options/ia-64-options.rst: New file.
3145 * doc/gcc/gcc-command-options/machine-dependent-options/ibm-rs-6000-and-powerpc-options.rst:
3147 * doc/gcc/gcc-command-options/machine-dependent-options/lm32-options.rst: New file.
3148 * doc/gcc/gcc-command-options/machine-dependent-options/loongarch-options.rst: New file.
3149 * doc/gcc/gcc-command-options/machine-dependent-options/m32c-options.rst: New file.
3150 * doc/gcc/gcc-command-options/machine-dependent-options/m32r-d-options.rst: New file.
3151 * doc/gcc/gcc-command-options/machine-dependent-options/m680x0-options.rst: New file.
3152 * doc/gcc/gcc-command-options/machine-dependent-options/mcore-options.rst: New file.
3153 * doc/gcc/gcc-command-options/machine-dependent-options/mep-options.rst: New file.
3154 * doc/gcc/gcc-command-options/machine-dependent-options/microblaze-options.rst: New file.
3155 * doc/gcc/gcc-command-options/machine-dependent-options/mips-options.rst: New file.
3156 * doc/gcc/gcc-command-options/machine-dependent-options/mmix-options.rst: New file.
3157 * doc/gcc/gcc-command-options/machine-dependent-options/mn10300-options.rst: New file.
3158 * doc/gcc/gcc-command-options/machine-dependent-options/moxie-options.rst: New file.
3159 * doc/gcc/gcc-command-options/machine-dependent-options/msp430-options.rst: New file.
3160 * doc/gcc/gcc-command-options/machine-dependent-options/nds32-options.rst: New file.
3161 * doc/gcc/gcc-command-options/machine-dependent-options/nios-ii-options.rst: New file.
3162 * doc/gcc/gcc-command-options/machine-dependent-options/nvidia-ptx-options.rst: New file.
3163 * doc/gcc/gcc-command-options/machine-dependent-options/openrisc-options.rst: New file.
3164 * doc/gcc/gcc-command-options/machine-dependent-options/options-for-system-v.rst: New file.
3165 * doc/gcc/gcc-command-options/machine-dependent-options/pdp-11-options.rst: New file.
3166 * doc/gcc/gcc-command-options/machine-dependent-options/picochip-options.rst: New file.
3167 * doc/gcc/gcc-command-options/machine-dependent-options/powerpc-options.rst: New file.
3168 * doc/gcc/gcc-command-options/machine-dependent-options/pru-options.rst: New file.
3169 * doc/gcc/gcc-command-options/machine-dependent-options/risc-v-options.rst: New file.
3170 * doc/gcc/gcc-command-options/machine-dependent-options/rl78-options.rst: New file.
3171 * doc/gcc/gcc-command-options/machine-dependent-options/rx-options.rst: New file.
3172 * doc/gcc/gcc-command-options/machine-dependent-options/s-390-and-zseries-options.rst:
3174 * doc/gcc/gcc-command-options/machine-dependent-options/score-options.rst: New file.
3175 * doc/gcc/gcc-command-options/machine-dependent-options/sh-options.rst: New file.
3176 * doc/gcc/gcc-command-options/machine-dependent-options/solaris-2-options.rst: New file.
3177 * doc/gcc/gcc-command-options/machine-dependent-options/sparc-options.rst: New file.
3178 * doc/gcc/gcc-command-options/machine-dependent-options/v850-options.rst: New file.
3179 * doc/gcc/gcc-command-options/machine-dependent-options/vax-options.rst: New file.
3180 * doc/gcc/gcc-command-options/machine-dependent-options/visium-options.rst: New file.
3181 * doc/gcc/gcc-command-options/machine-dependent-options/vms-options.rst: New file.
3182 * doc/gcc/gcc-command-options/machine-dependent-options/vxworks-options.rst: New file.
3183 * doc/gcc/gcc-command-options/machine-dependent-options/x86-options.rst: New file.
3184 * doc/gcc/gcc-command-options/machine-dependent-options/x86-windows-options.rst: New file.
3185 * doc/gcc/gcc-command-options/machine-dependent-options/xstormy16-options.rst: New file.
3186 * doc/gcc/gcc-command-options/machine-dependent-options/xtensa-options.rst: New file.
3187 * doc/gcc/gcc-command-options/machine-dependent-options/zseries-options.rst: New file.
3188 * doc/gcc/gcc-command-options/option-summary.rst: New file.
3189 * doc/gcc/gcc-command-options/options-controlling-c++-dialect.rst: New file.
3190 * doc/gcc/gcc-command-options/options-controlling-c-dialect.rst: New file.
3191 * doc/gcc/gcc-command-options/options-controlling-objective-c-and-objective-c++-dialects.rst:
3193 * doc/gcc/gcc-command-options/options-controlling-the-kind-of-output.rst: New file.
3194 * doc/gcc/gcc-command-options/options-controlling-the-preprocessor.rst: New file.
3195 * doc/gcc/gcc-command-options/options-for-code-generation-conventions.rst: New file.
3196 * doc/gcc/gcc-command-options/options-for-debugging-your-program.rst: New file.
3197 * doc/gcc/gcc-command-options/options-for-directory-search.rst: New file.
3198 * doc/gcc/gcc-command-options/options-for-linking.rst: New file.
3199 * doc/gcc/gcc-command-options/options-that-control-optimization.rst: New file.
3200 * doc/gcc/gcc-command-options/options-that-control-static-analysis.rst: New file.
3201 * doc/gcc/gcc-command-options/options-to-control-diagnostic-messages-formatting.rst:
3203 * doc/gcc/gcc-command-options/options-to-request-or-suppress-warnings.rst: New file.
3204 * doc/gcc/gcc-command-options/passing-options-to-the-assembler.rst: New file.
3205 * doc/gcc/gcc-command-options/program-instrumentation-options.rst: New file.
3206 * doc/gcc/gcc-command-options/specifying-subprocesses-and-the-switches-to-pass-to-them.rst:
3208 * doc/gcc/gcc-command-options/using-precompiled-headers.rst: New file.
3209 * doc/gcc/gcc.rst: New file.
3210 * doc/gcc/gcov-dump.rst: New file.
3211 * doc/gcc/gcov-tool.rst: New file.
3212 * doc/gcc/gcov.rst: New file.
3213 * doc/gcc/gcov/brief-description-of-gcov-data-files.rst: New file.
3214 * doc/gcc/gcov/data-file-relocation-to-support-cross-profiling.rst: New file.
3215 * doc/gcc/gcov/introduction-to-gcov.rst: New file.
3216 * doc/gcc/gcov/invoking-gcov.rst: New file.
3217 * doc/gcc/gcov/profiling-and-test-coverage-in-freestanding-environments.rst: New file.
3218 * doc/gcc/gcov/using-gcov-with-gcc-optimization.rst: New file.
3219 * doc/gcc/general-public-license-3.rst: New file.
3220 * doc/gcc/gnu-free-documentation-license.rst: New file.
3221 * doc/gcc/gnu-objective-c-features.rst: New file.
3222 * doc/gcc/gnu-objective-c-features/compatibilityalias.rst: New file.
3223 * doc/gcc/gnu-objective-c-features/constant-string-objects.rst: New file.
3224 * doc/gcc/gnu-objective-c-features/exceptions.rst: New file.
3225 * doc/gcc/gnu-objective-c-features/fast-enumeration.rst: New file.
3226 * doc/gcc/gnu-objective-c-features/garbage-collection.rst: New file.
3227 * doc/gcc/gnu-objective-c-features/gnu-objective-c-runtime-api.rst: New file.
3228 * doc/gcc/gnu-objective-c-features/load-executing-code-before-main.rst: New file.
3229 * doc/gcc/gnu-objective-c-features/messaging-with-the-gnu-objective-c-runtime.rst: New file.
3230 * doc/gcc/gnu-objective-c-features/synchronization.rst: New file.
3231 * doc/gcc/gnu-objective-c-features/type-encoding.rst: New file.
3232 * doc/gcc/gnu.rst: New file.
3233 * doc/gcc/have-you-found-a-bug.rst: New file.
3234 * doc/gcc/how-and-where-to-report-bugs.rst: New file.
3235 * doc/gcc/how-to-get-help-with-gcc.rst: New file.
3236 * doc/gcc/index.rst: New file.
3237 * doc/gcc/indices-and-tables.rst: New file.
3238 * doc/gcc/known-causes-of-trouble-with-gcc.rst: New file.
3239 * doc/gcc/known-causes-of-trouble-with-gcc/actual-bugs-we-havent-fixed-yet.rst: New file.
3240 * doc/gcc/known-causes-of-trouble-with-gcc/certain-changes-we-dont-want-to-make.rst:
3242 * doc/gcc/known-causes-of-trouble-with-gcc/common-misunderstandings-with-gnu-c.rst:
3244 * doc/gcc/known-causes-of-trouble-with-gcc/disappointments-and-misunderstandings.rst:
3246 * doc/gcc/known-causes-of-trouble-with-gcc/fixed-header-files.rst: New file.
3247 * doc/gcc/known-causes-of-trouble-with-gcc/incompatibilities-of-gcc.rst: New file.
3248 * doc/gcc/known-causes-of-trouble-with-gcc/interoperation.rst: New file.
3249 * doc/gcc/known-causes-of-trouble-with-gcc/standard-libraries.rst: New file.
3250 * doc/gcc/known-causes-of-trouble-with-gcc/warning-messages-and-error-messages.rst:
3252 * doc/gcc/language-standards-supported-by-gcc.rst: New file.
3253 * doc/gcc/language-standards-supported-by-gcc/c++-language.rst: New file.
3254 * doc/gcc/language-standards-supported-by-gcc/c-language.rst: New file.
3255 * doc/gcc/language-standards-supported-by-gcc/d-language.rst: New file.
3256 * doc/gcc/language-standards-supported-by-gcc/go-language.rst: New file.
3257 * doc/gcc/language-standards-supported-by-gcc/objective-c-and-objective-c++-languages.rst:
3259 * doc/gcc/language-standards-supported-by-gcc/references-for-other-languages.rst: New file.
3260 * doc/gcc/lto-dump.rst: New file.
3261 * doc/gcc/programming-languages-supported-by-gcc.rst: New file.
3262 * doc/gcc/reporting-bugs.rst: New file.
3263 * doc/gccint/analysis-and-optimization-of-gimple-tuples.rst: New file.
3264 * doc/gccint/analysis-and-optimization-of-gimple-tuples/alias-analysis.rst: New file.
3265 * doc/gccint/analysis-and-optimization-of-gimple-tuples/annotations.rst: New file.
3266 * doc/gccint/analysis-and-optimization-of-gimple-tuples/memory-model.rst: New file.
3267 * doc/gccint/analysis-and-optimization-of-gimple-tuples/ssa-operands.rst: New file.
3268 * doc/gccint/analysis-and-optimization-of-gimple-tuples/static-single-assignment.rst:
3270 * doc/gccint/analysis-and-representation-of-loops.rst: New file.
3271 * doc/gccint/analysis-and-representation-of-loops/data-dependency-analysis.rst: New file.
3272 * doc/gccint/analysis-and-representation-of-loops/iv-analysis-on-rtl.rst: New file.
3273 * doc/gccint/analysis-and-representation-of-loops/loop-closed-ssa-form.rst: New file.
3274 * doc/gccint/analysis-and-representation-of-loops/loop-manipulation.rst: New file.
3275 * doc/gccint/analysis-and-representation-of-loops/loop-querying.rst: New file.
3276 * doc/gccint/analysis-and-representation-of-loops/loop-representation.rst: New file.
3277 * doc/gccint/analysis-and-representation-of-loops/number-of-iterations-analysis.rst:
3279 * doc/gccint/analysis-and-representation-of-loops/scalar-evolutions.rst: New file.
3280 * doc/gccint/analyzer-internals.rst: New file.
3281 * doc/gccint/collect2.rst: New file.
3282 * doc/gccint/conf.py: New file.
3283 * doc/gccint/contributing-to-gcc-development.rst: New file.
3284 * doc/gccint/contributors-to-gcc.rst: New file.
3285 * doc/gccint/control-flow-graph.rst: New file.
3286 * doc/gccint/control-flow-graph/basic-blocks.rst: New file.
3287 * doc/gccint/control-flow-graph/edges.rst: New file.
3288 * doc/gccint/control-flow-graph/liveness-information.rst: New file.
3289 * doc/gccint/control-flow-graph/maintaining-the-cfg.rst: New file.
3290 * doc/gccint/control-flow-graph/profile-information.rst: New file.
3291 * doc/gccint/copyright.rst: New file.
3292 * doc/gccint/debugging-the-analyzer.rst: New file.
3293 * doc/gccint/funding.rst: New file.
3294 * doc/gccint/gcc-and-portability.rst: New file.
3295 * doc/gccint/general-public-license-3.rst: New file.
3296 * doc/gccint/generic.rst: New file.
3297 * doc/gccint/generic/attributes-in-trees.rst: New file.
3298 * doc/gccint/generic/c-and-c++-trees.rst: New file.
3299 * doc/gccint/generic/declarations.rst: New file.
3300 * doc/gccint/generic/deficiencies.rst: New file.
3301 * doc/gccint/generic/expressions.rst: New file.
3302 * doc/gccint/generic/functions.rst: New file.
3303 * doc/gccint/generic/language-dependent-trees.rst: New file.
3304 * doc/gccint/generic/overview.rst: New file.
3305 * doc/gccint/generic/statements.rst: New file.
3306 * doc/gccint/generic/types.rst: New file.
3307 * doc/gccint/gimple-api.rst: New file.
3308 * doc/gccint/gimple.rst: New file.
3309 * doc/gccint/gimple/adding-a-new-gimple-statement-code.rst: New file.
3310 * doc/gccint/gimple/class-hierarchy-of-gimple-statements.rst: New file.
3311 * doc/gccint/gimple/exception-handling.rst: New file.
3312 * doc/gccint/gimple/gimple-instruction-set.rst: New file.
3313 * doc/gccint/gimple/gimple-sequences.rst: New file.
3314 * doc/gccint/gimple/manipulating-gimple-statements.rst: New file.
3315 * doc/gccint/gimple/operands.rst: New file.
3316 * doc/gccint/gimple/sequence-iterators.rst: New file.
3317 * doc/gccint/gimple/statement-and-operand-traversals.rst: New file.
3318 * doc/gccint/gimple/temporaries.rst: New file.
3319 * doc/gccint/gimple/tuple-representation.rst: New file.
3320 * doc/gccint/gimple/tuple-specific-accessors.rst: New file.
3321 * doc/gccint/gimple/tuple-specific-accessors/gimpleasm.rst: New file.
3322 * doc/gccint/gimple/tuple-specific-accessors/gimpleassign.rst: New file.
3323 * doc/gccint/gimple/tuple-specific-accessors/gimplebind.rst: New file.
3324 * doc/gccint/gimple/tuple-specific-accessors/gimplecall.rst: New file.
3325 * doc/gccint/gimple/tuple-specific-accessors/gimplecatch.rst: New file.
3326 * doc/gccint/gimple/tuple-specific-accessors/gimplecond.rst: New file.
3327 * doc/gccint/gimple/tuple-specific-accessors/gimpledebug.rst: New file.
3328 * doc/gccint/gimple/tuple-specific-accessors/gimpleehfilter.rst: New file.
3329 * doc/gccint/gimple/tuple-specific-accessors/gimplegoto.rst: New file.
3330 * doc/gccint/gimple/tuple-specific-accessors/gimplelabel.rst: New file.
3331 * doc/gccint/gimple/tuple-specific-accessors/gimplenop.rst: New file.
3332 * doc/gccint/gimple/tuple-specific-accessors/gimpleompatomicload.rst: New file.
3333 * doc/gccint/gimple/tuple-specific-accessors/gimpleompatomicstore.rst: New file.
3334 * doc/gccint/gimple/tuple-specific-accessors/gimpleompcontinue.rst: New file.
3335 * doc/gccint/gimple/tuple-specific-accessors/gimpleompcritical.rst: New file.
3336 * doc/gccint/gimple/tuple-specific-accessors/gimpleompfor.rst: New file.
3337 * doc/gccint/gimple/tuple-specific-accessors/gimpleompmaster.rst: New file.
3338 * doc/gccint/gimple/tuple-specific-accessors/gimpleompordered.rst: New file.
3339 * doc/gccint/gimple/tuple-specific-accessors/gimpleompparallel.rst: New file.
3340 * doc/gccint/gimple/tuple-specific-accessors/gimpleompreturn.rst: New file.
3341 * doc/gccint/gimple/tuple-specific-accessors/gimpleompsection.rst: New file.
3342 * doc/gccint/gimple/tuple-specific-accessors/gimpleompsections.rst: New file.
3343 * doc/gccint/gimple/tuple-specific-accessors/gimpleompsingle.rst: New file.
3344 * doc/gccint/gimple/tuple-specific-accessors/gimplephi.rst: New file.
3345 * doc/gccint/gimple/tuple-specific-accessors/gimpleresx.rst: New file.
3346 * doc/gccint/gimple/tuple-specific-accessors/gimplereturn.rst: New file.
3347 * doc/gccint/gimple/tuple-specific-accessors/gimpleswitch.rst: New file.
3348 * doc/gccint/gimple/tuple-specific-accessors/gimpletry.rst: New file.
3349 * doc/gccint/gimple/tuple-specific-accessors/gimplewithcleanupexpr.rst: New file.
3350 * doc/gccint/gnu-free-documentation-license.rst: New file.
3351 * doc/gccint/guidelines-for-diagnostics.rst: New file.
3352 * doc/gccint/guidelines-for-options.rst: New file.
3353 * doc/gccint/host-common.rst: New file.
3354 * doc/gccint/host-configuration.rst: New file.
3355 * doc/gccint/host-filesystem.rst: New file.
3356 * doc/gccint/host-makefile-fragments.rst: New file.
3357 * doc/gccint/host-misc.rst: New file.
3358 * doc/gccint/index.rst: New file.
3359 * doc/gccint/indices-and-tables.rst: New file.
3360 * doc/gccint/interfacing-to-gcc-output.rst: New file.
3361 * doc/gccint/introduction.rst: New file.
3362 * doc/gccint/language-front-ends-in-gcc.rst: New file.
3363 * doc/gccint/link-time-optimization.rst: New file.
3364 * doc/gccint/link-time-optimization/design-overview.rst: New file.
3365 * doc/gccint/link-time-optimization/internal-flags-controlling-lto1.rst: New file.
3366 * doc/gccint/link-time-optimization/lto-file-sections.rst: New file.
3367 * doc/gccint/link-time-optimization/using-summary-information-in-ipa-passes.rst: New file.
3368 * doc/gccint/link-time-optimization/whole-program-assumptions-linker-plugin-and-symbol-visibilities.rst:
3370 * doc/gccint/machine-descriptions.rst: New file.
3371 * doc/gccint/machine-descriptions/c-statements-for-assembler-output.rst: New file.
3372 * doc/gccint/machine-descriptions/canonicalization-of-instructions.rst: New file.
3373 * doc/gccint/machine-descriptions/conditional-execution.rst: New file.
3374 * doc/gccint/machine-descriptions/constant-definitions.rst: New file.
3375 * doc/gccint/machine-descriptions/defining-how-to-split-instructions.rst: New file.
3376 * doc/gccint/machine-descriptions/defining-jump-instruction-patterns.rst: New file.
3377 * doc/gccint/machine-descriptions/defining-looping-instruction-patterns.rst: New file.
3378 * doc/gccint/machine-descriptions/defining-rtl-sequences-for-code-generation.rst: New file.
3379 * doc/gccint/machine-descriptions/everything-about-instruction-patterns.rst: New file.
3380 * doc/gccint/machine-descriptions/example-of-defineinsn.rst: New file.
3381 * doc/gccint/machine-descriptions/including-patterns-in-machine-descriptions.rst: New file.
3382 * doc/gccint/machine-descriptions/instruction-attributes.rst: New file.
3383 * doc/gccint/machine-descriptions/interdependence-of-patterns.rst: New file.
3384 * doc/gccint/machine-descriptions/iterators.rst: New file.
3385 * doc/gccint/machine-descriptions/machine-specific-peephole-optimizers.rst: New file.
3386 * doc/gccint/machine-descriptions/operand-constraints.rst: New file.
3387 * doc/gccint/machine-descriptions/output-templates-and-operand-substitution.rst: New file.
3388 * doc/gccint/machine-descriptions/overview-of-how-the-machine-description-is-used.rst:
3390 * doc/gccint/machine-descriptions/predicates.rst: New file.
3391 * doc/gccint/machine-descriptions/rtl-template.rst: New file.
3392 * doc/gccint/machine-descriptions/rtl-templates-transformations.rst: New file.
3393 * doc/gccint/machine-descriptions/standard-pattern-names-for-generation.rst: New file.
3394 * doc/gccint/machine-descriptions/when-the-order-of-patterns-matters.rst: New file.
3395 * doc/gccint/makefile-fragments.rst: New file.
3396 * doc/gccint/match-and-simplify.rst: New file.
3397 * doc/gccint/memory-management-and-type-information.rst: New file.
3398 * doc/gccint/memory-management-and-type-information/how-to-invoke-the-garbage-collector.rst:
3400 * doc/gccint/memory-management-and-type-information/marking-roots-for-the-garbage-collector.rst:
3402 * doc/gccint/memory-management-and-type-information/source-files-containing-type-information.rst:
3404 * doc/gccint/memory-management-and-type-information/support-for-inheritance.rst: New file.
3405 * doc/gccint/memory-management-and-type-information/support-for-user-provided-gc-marking-routines.rst:
3407 * doc/gccint/memory-management-and-type-information/the-inside-of-a-gty.rst: New file.
3408 * doc/gccint/memory-management-and-type-information/troubleshooting-the-garbage-collector.rst:
3410 * doc/gccint/option-file-format.rst: New file.
3411 * doc/gccint/option-properties.rst: New file.
3412 * doc/gccint/option-specification-files.rst: New file.
3413 * doc/gccint/passes-and-files-of-the-compiler.rst: New file.
3414 * doc/gccint/passes-and-files-of-the-compiler/gimplification-pass.rst: New file.
3415 * doc/gccint/passes-and-files-of-the-compiler/inter-procedural-optimization-passes.rst:
3417 * doc/gccint/passes-and-files-of-the-compiler/optimization-info.rst: New file.
3418 * doc/gccint/passes-and-files-of-the-compiler/parsing-pass.rst: New file.
3419 * doc/gccint/passes-and-files-of-the-compiler/pass-manager.rst: New file.
3420 * doc/gccint/passes-and-files-of-the-compiler/rtl-passes.rst: New file.
3421 * doc/gccint/passes-and-files-of-the-compiler/tree-ssa-passes.rst: New file.
3422 * doc/gccint/plugins.rst: New file.
3423 * doc/gccint/plugins/building-gcc-plugins.rst: New file.
3424 * doc/gccint/plugins/controlling-which-passes-are-being-run.rst: New file.
3425 * doc/gccint/plugins/giving-information-about-a-plugin.rst: New file.
3426 * doc/gccint/plugins/interacting-with-the-gcc-garbage-collector.rst: New file.
3427 * doc/gccint/plugins/interacting-with-the-pass-manager.rst: New file.
3428 * doc/gccint/plugins/keeping-track-of-available-passes.rst: New file.
3429 * doc/gccint/plugins/loading-plugins.rst: New file.
3430 * doc/gccint/plugins/plugin-api.rst: New file.
3431 * doc/gccint/plugins/recording-information-about-pass-execution.rst: New file.
3432 * doc/gccint/plugins/registering-custom-attributes-or-pragmas.rst: New file.
3433 * doc/gccint/rtl-representation.rst: New file.
3434 * doc/gccint/rtl-representation/access-to-operands.rst: New file.
3435 * doc/gccint/rtl-representation/access-to-special-operands.rst: New file.
3436 * doc/gccint/rtl-representation/assembler-instructions-as-expressions.rst: New file.
3437 * doc/gccint/rtl-representation/bit-fields.rst: New file.
3438 * doc/gccint/rtl-representation/comparison-operations.rst: New file.
3439 * doc/gccint/rtl-representation/constant-expression-types.rst: New file.
3440 * doc/gccint/rtl-representation/conversions.rst: New file.
3441 * doc/gccint/rtl-representation/declarations.rst: New file.
3442 * doc/gccint/rtl-representation/embedded-side-effects-on-addresses.rst: New file.
3443 * doc/gccint/rtl-representation/flags-in-an-rtl-expression.rst: New file.
3444 * doc/gccint/rtl-representation/insns.rst: New file.
3445 * doc/gccint/rtl-representation/machine-modes.rst: New file.
3446 * doc/gccint/rtl-representation/on-the-side-ssa-form-for-rtl.rst: New file.
3447 * doc/gccint/rtl-representation/reading-rtl.rst: New file.
3448 * doc/gccint/rtl-representation/registers-and-memory.rst: New file.
3449 * doc/gccint/rtl-representation/rtl-classes-and-formats.rst: New file.
3450 * doc/gccint/rtl-representation/rtl-expressions-for-arithmetic.rst: New file.
3451 * doc/gccint/rtl-representation/rtl-object-types.rst: New file.
3452 * doc/gccint/rtl-representation/rtl-representation-of-function-call-insns.rst: New file.
3453 * doc/gccint/rtl-representation/side-effect-expressions.rst: New file.
3454 * doc/gccint/rtl-representation/structure-sharing-assumptions.rst: New file.
3455 * doc/gccint/rtl-representation/variable-location-debug-information-in-rtl.rst: New file.
3456 * doc/gccint/rtl-representation/vector-operations.rst: New file.
3457 * doc/gccint/sizes-and-offsets-as-runtime-invariants.rst: New file.
3458 * doc/gccint/sizes-and-offsets-as-runtime-invariants/alignment-of-polyints.rst: New file.
3459 * doc/gccint/sizes-and-offsets-as-runtime-invariants/arithmetic-on-polyints.rst: New file.
3460 * doc/gccint/sizes-and-offsets-as-runtime-invariants/comparisons-involving-polyint.rst:
3462 * doc/gccint/sizes-and-offsets-as-runtime-invariants/computing-bounds-on-polyints.rst:
3464 * doc/gccint/sizes-and-offsets-as-runtime-invariants/consequences-of-using-polyint.rst:
3466 * doc/gccint/sizes-and-offsets-as-runtime-invariants/converting-polyints.rst: New file.
3467 * doc/gccint/sizes-and-offsets-as-runtime-invariants/guidelines-for-using-polyint.rst:
3469 * doc/gccint/sizes-and-offsets-as-runtime-invariants/miscellaneous-polyint-routines.rst:
3471 * doc/gccint/sizes-and-offsets-as-runtime-invariants/overview-of-polyint.rst: New file.
3472 * doc/gccint/source-tree-structure-and-build-system.rst: New file.
3473 * doc/gccint/source-tree-structure-and-build-system/configure-terms-and-history.rst:
3475 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory.rst: New file.
3476 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/anatomy-of-a-language-front-end.rst:
3478 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/anatomy-of-a-target-back-end.rst:
3480 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/build-system-in-the-gcc-directory.rst:
3482 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/building-documentation.rst:
3484 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/configuration-in-the-gcc-directory.rst:
3486 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/headers-installed-by-gcc.rst:
3488 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/library-source-files-and-headers-under-the-gcc-directory.rst:
3490 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/makefile-targets.rst:
3492 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/subdirectories-of-gcc.rst:
3494 * doc/gccint/source-tree-structure-and-build-system/top-level-source-directory.rst:
3496 * doc/gccint/standard-header-file-directories.rst: New file.
3497 * doc/gccint/static-analyzer.rst: New file.
3498 * doc/gccint/target-macros.rst: New file.
3499 * doc/gccint/target-macros/adding-support-for-named-address-spaces.rst: New file.
3500 * doc/gccint/target-macros/addressing-modes.rst: New file.
3501 * doc/gccint/target-macros/adjusting-the-instruction-scheduler.rst: New file.
3502 * doc/gccint/target-macros/anchored-addresses.rst: New file.
3503 * doc/gccint/target-macros/c++-abi-parameters.rst: New file.
3504 * doc/gccint/target-macros/condition-code-status.rst: New file.
3505 * doc/gccint/target-macros/controlling-debugging-information-format.rst: New file.
3506 * doc/gccint/target-macros/controlling-the-compilation-driver-gcc.rst: New file.
3507 * doc/gccint/target-macros/cross-compilation-and-floating-point.rst: New file.
3508 * doc/gccint/target-macros/d-abi-parameters.rst: New file.
3509 * doc/gccint/target-macros/defining-coprocessor-specifics-for-mips-targets.rst: New file.
3510 * doc/gccint/target-macros/defining-data-structures-for-per-function-information.rst:
3512 * doc/gccint/target-macros/defining-target-specific-uses-of-attribute.rst: New file.
3513 * doc/gccint/target-macros/defining-the-output-assembler-language.rst: New file.
3514 * doc/gccint/target-macros/defining-the-output-assembler-language/assembler-commands-for-alignment.rst:
3516 * doc/gccint/target-macros/defining-the-output-assembler-language/assembler-commands-for-exception-regions.rst:
3518 * doc/gccint/target-macros/defining-the-output-assembler-language/how-initialization-functions-are-handled.rst:
3520 * doc/gccint/target-macros/defining-the-output-assembler-language/macros-controlling-initialization-routines.rst:
3522 * doc/gccint/target-macros/defining-the-output-assembler-language/output-and-generation-of-labels.rst:
3524 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-assembler-instructions.rst:
3526 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-data.rst:
3528 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-dispatch-tables.rst:
3530 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-uninitialized-variables.rst:
3532 * doc/gccint/target-macros/defining-the-output-assembler-language/the-overall-framework-of-an-assembler-file.rst:
3534 * doc/gccint/target-macros/describing-relative-costs-of-operations.rst: New file.
3535 * doc/gccint/target-macros/dividing-the-output-into-sections-texts-data.rst: New file.
3536 * doc/gccint/target-macros/emulating-tls.rst: New file.
3537 * doc/gccint/target-macros/implementing-the-varargs-macros.rst: New file.
3538 * doc/gccint/target-macros/implicit-calls-to-library-routines.rst: New file.
3539 * doc/gccint/target-macros/layout-of-source-language-data-types.rst: New file.
3540 * doc/gccint/target-macros/miscellaneous-parameters.rst: New file.
3541 * doc/gccint/target-macros/mode-switching-instructions.rst: New file.
3542 * doc/gccint/target-macros/parameters-for-precompiled-header-validity-checking.rst:
3544 * doc/gccint/target-macros/position-independent-code.rst: New file.
3545 * doc/gccint/target-macros/register-classes.rst: New file.
3546 * doc/gccint/target-macros/register-usage.rst: New file.
3547 * doc/gccint/target-macros/run-time-target-specification.rst: New file.
3548 * doc/gccint/target-macros/stack-layout-and-calling-conventions.rst: New file.
3549 * doc/gccint/target-macros/stack-layout-and-calling-conventions/basic-stack-layout.rst:
3551 * doc/gccint/target-macros/stack-layout-and-calling-conventions/caller-saves-register-allocation.rst:
3553 * doc/gccint/target-macros/stack-layout-and-calling-conventions/eliminating-frame-pointer-and-arg-pointer.rst:
3555 * doc/gccint/target-macros/stack-layout-and-calling-conventions/exception-handling-support.rst:
3557 * doc/gccint/target-macros/stack-layout-and-calling-conventions/function-entry-and-exit.rst:
3559 * doc/gccint/target-macros/stack-layout-and-calling-conventions/generating-code-for-profiling.rst:
3561 * doc/gccint/target-macros/stack-layout-and-calling-conventions/how-large-values-are-returned.rst:
3563 * doc/gccint/target-macros/stack-layout-and-calling-conventions/how-scalar-function-values-are-returned.rst:
3565 * doc/gccint/target-macros/stack-layout-and-calling-conventions/miscellaneous-register-hooks.rst:
3567 * doc/gccint/target-macros/stack-layout-and-calling-conventions/passing-arguments-in-registers.rst:
3569 * doc/gccint/target-macros/stack-layout-and-calling-conventions/passing-function-arguments-on-the-stack.rst:
3571 * doc/gccint/target-macros/stack-layout-and-calling-conventions/permitting-tail-calls.rst:
3573 * doc/gccint/target-macros/stack-layout-and-calling-conventions/registers-that-address-the-stack-frame.rst:
3575 * doc/gccint/target-macros/stack-layout-and-calling-conventions/shrink-wrapping-separate-components.rst:
3577 * doc/gccint/target-macros/stack-layout-and-calling-conventions/specifying-how-stack-checking-is-done.rst:
3579 * doc/gccint/target-macros/stack-layout-and-calling-conventions/stack-smashing-protection.rst:
3581 * doc/gccint/target-macros/storage-layout.rst: New file.
3582 * doc/gccint/target-macros/support-for-nested-functions.rst: New file.
3583 * doc/gccint/target-macros/the-global-targetm-variable.rst: New file.
3584 * doc/gccint/target-makefile-fragments.rst: New file.
3585 * doc/gccint/testsuites.rst: New test.
3586 * doc/gccint/testsuites/ada-language-testsuites.rst: New test.
3587 * doc/gccint/testsuites/c-language-testsuites.rst: New test.
3588 * doc/gccint/testsuites/directives-used-within-dejagnu-tests.rst: New test.
3589 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/commands-for-use-in-dg-final.rst:
3591 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/features-for-dg-add-options.rst:
3593 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/keywords-describing-target-attributes.rst:
3595 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/selecting-targets-to-which-a-test-applies.rst:
3597 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/syntax-and-descriptions-of-test-directives.rst:
3599 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/variants-of-dg-require-support.rst:
3601 * doc/gccint/testsuites/idioms-used-in-testsuite-code.rst: New test.
3602 * doc/gccint/testsuites/support-for-testing-binary-compatibility.rst: New test.
3603 * doc/gccint/testsuites/support-for-testing-gcov.rst: New test.
3604 * doc/gccint/testsuites/support-for-testing-gimple-passes.rst: New test.
3605 * doc/gccint/testsuites/support-for-testing-link-time-optimizations.rst: New test.
3606 * doc/gccint/testsuites/support-for-testing-profile-directed-optimizations.rst: New test.
3607 * doc/gccint/testsuites/support-for-testing-rtl-passes.rst: New test.
3608 * doc/gccint/testsuites/support-for-torture-testing-using-multiple-options.rst: New test.
3609 * doc/gccint/the-gcc-low-level-runtime-library.rst: New file.
3610 * doc/gccint/the-gcc-low-level-runtime-library/language-independent-routines-for-exception-handling.rst:
3612 * doc/gccint/the-gcc-low-level-runtime-library/miscellaneous-runtime-library-routines.rst:
3614 * doc/gccint/the-gcc-low-level-runtime-library/routines-for-decimal-floating-point-emulation.rst:
3616 * doc/gccint/the-gcc-low-level-runtime-library/routines-for-fixed-point-fractional-emulation.rst:
3618 * doc/gccint/the-gcc-low-level-runtime-library/routines-for-floating-point-emulation.rst:
3620 * doc/gccint/the-gcc-low-level-runtime-library/routines-for-integer-arithmetic.rst:
3622 * doc/gccint/the-language.rst: New file.
3623 * doc/gccint/user-experience-guidelines.rst: New file.
3624 * doc/install/binaries.rst: New file.
3625 * doc/install/building.rst: New file.
3626 * doc/install/building/building-a-cross-compiler.rst: New file.
3627 * doc/install/building/building-a-native-compiler.rst: New file.
3628 * doc/install/building/building-in-parallel.rst: New file.
3629 * doc/install/building/building-the-ada-compiler.rst: New file.
3630 * doc/install/building/building-the-d-compiler.rst: New file.
3631 * doc/install/building/building-with-profile-feedback.rst: New file.
3632 * doc/install/conf.py: New file.
3633 * doc/install/configuration.rst: New file.
3634 * doc/install/copyright.rst: New file.
3635 * doc/install/downloading-gcc.rst: New file.
3636 * doc/install/final-installation.rst: New file.
3637 * doc/install/gnu-free-documentation-license.rst: New file.
3638 * doc/install/host-target-specific-installation-notes-for-gcc.rst: New file.
3639 * doc/install/how-can-you-run-the-testsuite-on-selected-tests.rst: New test.
3640 * doc/install/how-to-interpret-test-results.rst: New file.
3641 * doc/install/index.rst: New file.
3642 * doc/install/indices-and-tables.rst: New file.
3643 * doc/install/installing-gcc.rst: New file.
3644 * doc/install/passing-options-and-running-multiple-testsuites.rst: New test.
3645 * doc/install/prerequisites.rst: New file.
3646 * doc/install/submitting-test-results.rst: New file.
3647 * doc/install/testing.rst: New file.
3649 2022-11-14 Martin Liska <mliska@suse.cz>
3651 * doc/gcc/gcc-command-options/options-controlling-c++-dialect.rst: Revert.
3653 2022-11-14 Martin Liska <mliska@suse.cz>
3656 2022-11-14 Martin Liska <mliska@suse.cz>
3658 * common/common-target.def: Port to RST.
3659 * target.def: Port to RST.
3661 2022-11-14 Martin Liska <mliska@suse.cz>
3664 2022-11-09 Martin Liska <mliska@suse.cz>
3666 * doc/gccint/target-macros/adding-support-for-named-address-spaces.rst: Use tm.rst.in file.
3667 * doc/gccint/target-macros/addressing-modes.rst: Likewise.
3668 * doc/gccint/target-macros/adjusting-the-instruction-scheduler.rst: Likewise.
3669 * doc/gccint/target-macros/anchored-addresses.rst: Likewise.
3670 * doc/gccint/target-macros/c++-abi-parameters.rst: Likewise.
3671 * doc/gccint/target-macros/condition-code-status.rst: Likewise.
3672 * doc/gccint/target-macros/controlling-debugging-information-format.rst: Likewise.
3673 * doc/gccint/target-macros/controlling-the-compilation-driver-gcc.rst: Likewise.
3674 * doc/gccint/target-macros/d-abi-parameters.rst: Likewise.
3675 * doc/gccint/target-macros/defining-target-specific-uses-of-attribute.rst: Likewise.
3676 * doc/gccint/target-macros/defining-the-output-assembler-language/assembler-commands-for-exception-regions.rst:
3678 * doc/gccint/target-macros/defining-the-output-assembler-language/macros-controlling-initialization-routines.rst:
3680 * doc/gccint/target-macros/defining-the-output-assembler-language/output-and-generation-of-labels.rst:
3682 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-assembler-instructions.rst:
3684 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-data.rst:
3686 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-dispatch-tables.rst:
3688 * doc/gccint/target-macros/defining-the-output-assembler-language/the-overall-framework-of-an-assembler-file.rst:
3690 * doc/gccint/target-macros/describing-relative-costs-of-operations.rst: Likewise.
3691 * doc/gccint/target-macros/dividing-the-output-into-sections-texts-data.rst: Likewise.
3692 * doc/gccint/target-macros/emulating-tls.rst: Likewise.
3693 * doc/gccint/target-macros/implementing-the-varargs-macros.rst: Likewise.
3694 * doc/gccint/target-macros/implicit-calls-to-library-routines.rst: Likewise.
3695 * doc/gccint/target-macros/layout-of-source-language-data-types.rst: Likewise.
3696 * doc/gccint/target-macros/miscellaneous-parameters.rst: Likewise.
3697 * doc/gccint/target-macros/mode-switching-instructions.rst: Likewise.
3698 * doc/gccint/target-macros/parameters-for-precompiled-header-validity-checking.rst:
3700 * doc/gccint/target-macros/register-classes.rst: Likewise.
3701 * doc/gccint/target-macros/register-usage.rst: Likewise.
3702 * doc/gccint/target-macros/run-time-target-specification.rst: Likewise.
3703 * doc/gccint/target-macros/stack-layout-and-calling-conventions/basic-stack-layout.rst:
3705 * doc/gccint/target-macros/stack-layout-and-calling-conventions/eliminating-frame-pointer-and-arg-pointer.rst:
3707 * doc/gccint/target-macros/stack-layout-and-calling-conventions/function-entry-and-exit.rst:
3709 * doc/gccint/target-macros/stack-layout-and-calling-conventions/generating-code-for-profiling.rst:
3711 * doc/gccint/target-macros/stack-layout-and-calling-conventions/how-large-values-are-returned.rst:
3713 * doc/gccint/target-macros/stack-layout-and-calling-conventions/how-scalar-function-values-are-returned.rst:
3715 * doc/gccint/target-macros/stack-layout-and-calling-conventions/miscellaneous-register-hooks.rst:
3717 * doc/gccint/target-macros/stack-layout-and-calling-conventions/passing-arguments-in-registers.rst:
3719 * doc/gccint/target-macros/stack-layout-and-calling-conventions/passing-function-arguments-on-the-stack.rst:
3721 * doc/gccint/target-macros/stack-layout-and-calling-conventions/permitting-tail-calls.rst:
3723 * doc/gccint/target-macros/stack-layout-and-calling-conventions/registers-that-address-the-stack-frame.rst:
3725 * doc/gccint/target-macros/stack-layout-and-calling-conventions/shrink-wrapping-separate-components.rst:
3727 * doc/gccint/target-macros/stack-layout-and-calling-conventions/specifying-how-stack-checking-is-done.rst:
3729 * doc/gccint/target-macros/stack-layout-and-calling-conventions/stack-smashing-protection.rst:
3731 * doc/gccint/target-macros/storage-layout.rst: Likewise.
3732 * doc/gccint/target-macros/support-for-nested-functions.rst: Likewise.
3734 2022-11-14 Martin Liska <mliska@suse.cz>
3737 2022-11-14 Martin Liska <mliska@suse.cz>
3739 * Makefile.in: Support Sphinx based documentation.
3740 * configure: Regenerate.
3741 * configure.ac: Support Sphinx based documentation.
3742 * genhooks.cc (struct s_hook): Emit tm.rst.in file.
3744 2022-11-14 Martin Liska <mliska@suse.cz>
3747 2022-11-09 Martin Liska <mliska@suse.cz>
3749 * doc/gccint/target-macros/tm.rst.in: New file.
3751 2022-11-14 Martin Liska <mliska@suse.cz>
3754 2022-11-14 Martin Liska <mliska@suse.cz>
3756 * doc/analyzer.texi: Removed.
3757 * doc/avr-mmcu.texi: Removed.
3758 * doc/bugreport.texi: Removed.
3759 * doc/cfg.texi: Removed.
3760 * doc/collect2.texi: Removed.
3761 * doc/compat.texi: Removed.
3762 * doc/configfiles.texi: Removed.
3763 * doc/configterms.texi: Removed.
3764 * doc/contrib.texi: Removed.
3765 * doc/contribute.texi: Removed.
3766 * doc/cpp.texi: Removed.
3767 * doc/cppdiropts.texi: Removed.
3768 * doc/cppenv.texi: Removed.
3769 * doc/cppinternals.texi: Removed.
3770 * doc/cppopts.texi: Removed.
3771 * doc/cppwarnopts.texi: Removed.
3772 * doc/extend.texi: Removed.
3773 * doc/fragments.texi: Removed.
3774 * doc/frontends.texi: Removed.
3775 * doc/gcc.texi: Removed.
3776 * doc/gccint.texi: Removed.
3777 * doc/gcov-dump.texi: Removed.
3778 * doc/gcov-tool.texi: Removed.
3779 * doc/gcov.texi: Removed.
3780 * doc/generic.texi: Removed.
3781 * doc/gimple.texi: Removed.
3782 * doc/gnu.texi: Removed.
3783 * doc/gty.texi: Removed.
3784 * doc/headerdirs.texi: Removed.
3785 * doc/hostconfig.texi: Removed.
3786 * doc/implement-c.texi: Removed.
3787 * doc/implement-cxx.texi: Removed.
3788 * doc/include/fdl.texi: Removed.
3789 * doc/include/funding.texi: Removed.
3790 * doc/include/gcc-common.texi: Removed.
3791 * doc/include/gpl_v3.texi: Removed.
3792 * doc/install.texi: Removed.
3793 * doc/interface.texi: Removed.
3794 * doc/invoke.texi: Removed.
3795 * doc/languages.texi: Removed.
3796 * doc/libgcc.texi: Removed.
3797 * doc/loop.texi: Removed.
3798 * doc/lto-dump.texi: Removed.
3799 * doc/lto.texi: Removed.
3800 * doc/makefile.texi: Removed.
3801 * doc/match-and-simplify.texi: Removed.
3802 * doc/md.texi: Removed.
3803 * doc/objc.texi: Removed.
3804 * doc/optinfo.texi: Removed.
3805 * doc/options.texi: Removed.
3806 * doc/passes.texi: Removed.
3807 * doc/plugins.texi: Removed.
3808 * doc/poly-int.texi: Removed.
3809 * doc/portability.texi: Removed.
3810 * doc/rtl.texi: Removed.
3811 * doc/service.texi: Removed.
3812 * doc/sourcebuild.texi: Removed.
3813 * doc/standards.texi: Removed.
3814 * doc/tm.texi: Removed.
3815 * doc/tree-ssa.texi: Removed.
3816 * doc/trouble.texi: Removed.
3817 * doc/ux.texi: Removed.
3818 * doc/tm.texi.in: Removed.
3820 2022-11-14 Martin Liska <mliska@suse.cz>
3823 2022-11-09 Martin Liska <mliska@suse.cz>
3825 * target.def: Use proper lexers for target macros.
3826 * doc/gccint/target-macros/tm.rst.in: Re-generate.
3828 2022-11-14 Martin Liska <mliska@suse.cz>
3831 2022-11-09 Martin Liska <mliska@suse.cz>
3833 * Makefile.in: Support --with-sphinx-build.
3835 * configure: Regenerate.
3837 2022-11-14 Martin Liska <mliska@suse.cz>
3840 2022-11-14 Martin Liska <mliska@suse.cz>
3842 * doc/gcc/gcc-command-options/options-controlling-the-kind-of-output.rst:
3843 Fix cross manual refs.
3844 * doc/gcc/language-standards-supported-by-gcc/references-for-other-languages.rst:
3847 2022-11-14 Martin Liska <mliska@suse.cz>
3850 2022-11-09 Martin Liska <mliska@suse.cz>
3852 * doc/gccint/target-macros/defining-the-output-assembler-language/assembler-commands-for-exception-regions.rst:
3853 Do not include tm.rst.in.
3854 * doc/gccint/target-macros/defining-the-output-assembler-language/the-overall-framework-of-an-assembler-file.rst:
3856 * doc/gccint/target-macros/run-time-target-specification.rst:
3859 2022-11-14 Martin Liska <mliska@suse.cz>
3862 2022-11-09 Martin Liska <mliska@suse.cz>
3864 * doc/gccint/target-macros/stack-layout-and-calling-conventions/passing-arguments-in-registers.rst:
3866 * doc/gccint/target-macros/storage-layout.rst: Likewise.
3867 * doc/gcc/gcc-command-options/machine-dependent-options/x86-options.rst: Sync with master.
3868 * doc/gcc/gcc-command-options/option-summary.rst: Likewise.
3870 2022-11-14 Martin Liska <mliska@suse.cz>
3873 2022-11-09 Martin Liska <mliska@suse.cz>
3875 * Makefile.in: Support installation if sphinx-build is missing.
3877 2022-11-14 Martin Liska <mliska@suse.cz>
3880 2022-11-09 Martin Liska <mliska@suse.cz>
3882 * Makefile.in: Add vpath default for .rst files.
3883 * config/avr/avr-devices.cc: For port RST.
3884 * config/avr/avr-mcus.def: Update path.
3885 * config/avr/gen-avr-mmcu-texi.cc: Moved to...
3886 * config/avr/gen-avr-mmcu-rst.cc: ...here.
3887 * config/avr/t-avr: Update to rst.
3888 * doc/gcc/gcc-command-options/machine-dependent-options/avr-mmcu.rst: Re-generate.
3890 2022-11-14 Martin Liska <mliska@suse.cz>
3893 2022-11-09 Martin Liska <mliska@suse.cz>
3895 * opts.cc (get_option_html_page): Port to sphinx URLs.
3896 (get_option_url): Likewise.
3897 (test_get_option_html_page): Likewise.
3899 2022-11-14 Martin Liska <mliska@suse.cz>
3902 2022-11-14 Martin Liska <mliska@suse.cz>
3904 * doc/cpp/character-sets.rst:
3905 Add trailing newline.
3906 * doc/cpp/conditional-syntax.rst:
3907 Add trailing newline.
3908 * doc/cpp/conditional-uses.rst:
3909 Add trailing newline.
3910 * doc/cpp/conditionals.rst:
3911 Add trailing newline.
3912 * doc/cpp/copyright.rst:
3913 Add trailing newline.
3914 * doc/cpp/deleted-code.rst:
3915 Add trailing newline.
3916 * doc/cpp/diagnostics.rst:
3917 Add trailing newline.
3918 * doc/cpp/environment-variables.rst:
3919 Add trailing newline.
3920 * doc/cpp/gnu-free-documentation-license.rst:
3921 Add trailing newline.
3922 * doc/cpp/header-files.rst:
3923 Add trailing newline.
3924 * doc/cpp/header-files/alternatives-to-wrapper-ifndef.rst:
3925 Add trailing newline.
3926 * doc/cpp/header-files/computed-includes.rst:
3927 Add trailing newline.
3928 * doc/cpp/header-files/include-operation.rst:
3929 Add trailing newline.
3930 * doc/cpp/header-files/include-syntax.rst:
3931 Add trailing newline.
3932 * doc/cpp/header-files/once-only-headers.rst:
3933 Add trailing newline.
3934 * doc/cpp/header-files/search-path.rst:
3935 Add trailing newline.
3936 * doc/cpp/header-files/system-headers.rst:
3937 Add trailing newline.
3938 * doc/cpp/header-files/wrapper-headers.rst:
3939 Add trailing newline.
3940 * doc/cpp/implementation-defined-behavior.rst:
3941 Add trailing newline.
3942 * doc/cpp/implementation-details.rst:
3943 Add trailing newline.
3944 * doc/cpp/implementation-limits.rst:
3945 Add trailing newline.
3946 * doc/cpp/index.rst:
3947 Add trailing newline.
3948 * doc/cpp/indices-and-tables.rst:
3949 Add trailing newline.
3950 * doc/cpp/initial-processing.rst:
3951 Add trailing newline.
3952 * doc/cpp/invocation.rst:
3953 Add trailing newline.
3954 * doc/cpp/line-control.rst:
3955 Add trailing newline.
3956 * doc/cpp/macros.rst:
3957 Add trailing newline.
3958 * doc/cpp/macros/concatenation.rst:
3959 Add trailing newline.
3960 * doc/cpp/macros/directives-within-macro-arguments.rst:
3961 Add trailing newline.
3962 * doc/cpp/macros/function-like-macros.rst:
3963 Add trailing newline.
3964 * doc/cpp/macros/macro-arguments.rst:
3965 Add trailing newline.
3966 * doc/cpp/macros/macro-pitfalls.rst:
3967 Add trailing newline.
3968 * doc/cpp/macros/object-like-macros.rst:
3969 Add trailing newline.
3970 * doc/cpp/macros/predefined-macros.rst:
3971 Add trailing newline.
3972 * doc/cpp/macros/stringizing.rst:
3973 Add trailing newline.
3974 * doc/cpp/macros/undefining-and-redefining-macros.rst:
3975 Add trailing newline.
3976 * doc/cpp/macros/variadic-macros.rst:
3977 Add trailing newline.
3978 * doc/cpp/obsolete-features.rst:
3979 Add trailing newline.
3980 * doc/cpp/other-directives.rst:
3981 Add trailing newline.
3982 * doc/cpp/overview.rst:
3983 Add trailing newline.
3984 * doc/cpp/pragmas.rst:
3985 Add trailing newline.
3986 * doc/cpp/preprocessor-output.rst:
3987 Add trailing newline.
3988 * doc/cpp/the-preprocessing-language.rst:
3989 Add trailing newline.
3990 * doc/cpp/tokenization.rst:
3991 Add trailing newline.
3992 * doc/cpp/traditional-lexical-analysis.rst:
3993 Add trailing newline.
3994 * doc/cpp/traditional-macros.rst:
3995 Add trailing newline.
3996 * doc/cpp/traditional-miscellany.rst:
3997 Add trailing newline.
3998 * doc/cpp/traditional-mode.rst:
3999 Add trailing newline.
4000 * doc/cpp/traditional-warnings.rst:
4001 Add trailing newline.
4002 * doc/cppinternals/copyright.rst:
4003 Add trailing newline.
4004 * doc/cppinternals/cppinternals.rst:
4005 Add trailing newline.
4006 * doc/cppinternals/cpplib.rst:
4007 Add trailing newline.
4008 * doc/cppinternals/files.rst:
4009 Add trailing newline.
4010 * doc/cppinternals/index.rst:
4011 Add trailing newline.
4012 * doc/cppinternals/indices-and-tables.rst:
4013 Add trailing newline.
4014 * doc/cppinternals/internal-representation-of-macros.rst:
4015 Add trailing newline.
4016 * doc/cppinternals/just-which-line-number-anyway.rst:
4017 Add trailing newline.
4018 * doc/cppinternals/lexing-a-line.rst:
4019 Add trailing newline.
4020 * doc/cppinternals/lexing-a-token.rst:
4021 Add trailing newline.
4022 * doc/cppinternals/looking-for-a-function-like-macros-opening-parenthesis.rst:
4023 Add trailing newline.
4024 * doc/cppinternals/macro-expansion-overview.rst:
4025 Add trailing newline.
4026 * doc/cppinternals/marking-tokens-ineligible-for-future-expansion.rst:
4027 Add trailing newline.
4028 * doc/cppinternals/multiple-include-optimization.rst:
4029 Add trailing newline.
4030 * doc/cppinternals/overview.rst:
4031 Add trailing newline.
4032 * doc/cppinternals/representation-of-line-numbers.rst:
4033 Add trailing newline.
4034 * doc/cppinternals/scanning-the-replacement-list-for-macros-to-expand.rst:
4035 Add trailing newline.
4036 * doc/gcc/binary-compatibility.rst:
4037 Add trailing newline.
4038 * doc/gcc/c++-implementation-defined-behavior.rst:
4039 Add trailing newline.
4040 * doc/gcc/c-implementation-defined-behavior.rst:
4041 Add trailing newline.
4042 * doc/gcc/c-implementation-defined-behavior/architecture.rst:
4043 Add trailing newline.
4044 * doc/gcc/c-implementation-defined-behavior/arrays-and-pointers.rst:
4045 Add trailing newline.
4046 * doc/gcc/c-implementation-defined-behavior/characters.rst:
4047 Add trailing newline.
4048 * doc/gcc/c-implementation-defined-behavior/declarators.rst:
4049 Add trailing newline.
4050 * doc/gcc/c-implementation-defined-behavior/environment.rst:
4051 Add trailing newline.
4052 * doc/gcc/c-implementation-defined-behavior/floating-point.rst:
4053 Add trailing newline.
4054 * doc/gcc/c-implementation-defined-behavior/hints.rst:
4055 Add trailing newline.
4056 * doc/gcc/c-implementation-defined-behavior/identifiers.rst:
4057 Add trailing newline.
4058 * doc/gcc/c-implementation-defined-behavior/integers.rst:
4059 Add trailing newline.
4060 * doc/gcc/c-implementation-defined-behavior/library-functions.rst:
4061 Add trailing newline.
4062 * doc/gcc/c-implementation-defined-behavior/locale-specific-behavior.rst:
4063 Add trailing newline.
4064 * doc/gcc/c-implementation-defined-behavior/preprocessing-directives.rst:
4065 Add trailing newline.
4066 * doc/gcc/c-implementation-defined-behavior/qualifiers.rst:
4067 Add trailing newline.
4068 * doc/gcc/c-implementation-defined-behavior/statements.rst:
4069 Add trailing newline.
4070 * doc/gcc/c-implementation-defined-behavior/structures-unions-enumerations-and-bit-fields.rst:
4071 Add trailing newline.
4072 * doc/gcc/c-implementation-defined-behavior/translation.rst:
4073 Add trailing newline.
4074 * doc/gcc/conditionally-supported-behavior.rst:
4075 Add trailing newline.
4076 * doc/gcc/contributing-to-gcc-development.rst:
4077 Add trailing newline.
4078 * doc/gcc/contributors-to-gcc.rst:
4079 Add trailing newline.
4080 * doc/gcc/copyright.rst:
4081 Add trailing newline.
4082 * doc/gcc/exception-handling.rst:
4083 Add trailing newline.
4084 * doc/gcc/extensions-to-the-c++-language.rst:
4085 Add trailing newline.
4086 * doc/gcc/extensions-to-the-c++-language/backwards-compatibility.rst:
4087 Add trailing newline.
4088 * doc/gcc/extensions-to-the-c++-language/c++-concepts.rst:
4089 Add trailing newline.
4090 * doc/gcc/extensions-to-the-c++-language/c++-interface-and-implementation-pragmas.rst:
4091 Add trailing newline.
4092 * doc/gcc/extensions-to-the-c++-language/c++-specific-variable-function-and-type-attributes.rst:
4093 Add trailing newline.
4094 * doc/gcc/extensions-to-the-c++-language/deprecated-features.rst:
4095 Add trailing newline.
4096 * doc/gcc/extensions-to-the-c++-language/extracting-the-function-pointer-from-a-bound-pointer-to-member-function.rst:
4097 Add trailing newline.
4098 * doc/gcc/extensions-to-the-c++-language/function-multiversioning.rst:
4099 Add trailing newline.
4100 * doc/gcc/extensions-to-the-c++-language/restricting-pointer-aliasing.rst:
4101 Add trailing newline.
4102 * doc/gcc/extensions-to-the-c++-language/type-traits.rst:
4103 Add trailing newline.
4104 * doc/gcc/extensions-to-the-c++-language/vague-linkage.rst:
4105 Add trailing newline.
4106 * doc/gcc/extensions-to-the-c++-language/when-is-a-volatile-c++-object-accessed.rst:
4107 Add trailing newline.
4108 * doc/gcc/extensions-to-the-c++-language/wheres-the-template.rst:
4109 Add trailing newline.
4110 * doc/gcc/extensions-to-the-c-language-family.rst:
4111 Add trailing newline.
4112 * doc/gcc/extensions-to-the-c-language-family/128-bit-integers.rst:
4113 Add trailing newline.
4114 * doc/gcc/extensions-to-the-c-language-family/additional-floating-types.rst:
4115 Add trailing newline.
4116 * doc/gcc/extensions-to-the-c-language-family/alternate-keywords.rst:
4117 Add trailing newline.
4118 * doc/gcc/extensions-to-the-c-language-family/an-inline-function-is-as-fast-as-a-macro.rst:
4119 Add trailing newline.
4120 * doc/gcc/extensions-to-the-c-language-family/arithmetic-on-void-and-function-pointers.rst:
4121 Add trailing newline.
4122 * doc/gcc/extensions-to-the-c-language-family/arrays-of-length-zero.rst:
4123 Add trailing newline.
4124 * doc/gcc/extensions-to-the-c-language-family/arrays-of-variable-length.rst:
4125 Add trailing newline.
4126 * doc/gcc/extensions-to-the-c-language-family/attribute-syntax.rst:
4127 Add trailing newline.
4128 * doc/gcc/extensions-to-the-c-language-family/binary-constants-using-the-0b-prefix.rst:
4129 Add trailing newline.
4130 * doc/gcc/extensions-to-the-c-language-family/built-in-functions-for-memory-model-aware-atomic-operations.rst:
4131 Add trailing newline.
4132 * doc/gcc/extensions-to-the-c-language-family/built-in-functions-to-perform-arithmetic-with-overflow-checking.rst:
4133 Add trailing newline.
4134 * doc/gcc/extensions-to-the-c-language-family/c++-style-comments.rst:
4135 Add trailing newline.
4136 * doc/gcc/extensions-to-the-c-language-family/case-ranges.rst:
4137 Add trailing newline.
4138 * doc/gcc/extensions-to-the-c-language-family/cast-to-a-union-type.rst:
4139 Add trailing newline.
4140 * doc/gcc/extensions-to-the-c-language-family/complex-numbers.rst:
4141 Add trailing newline.
4142 * doc/gcc/extensions-to-the-c-language-family/compound-literals.rst:
4143 Add trailing newline.
4144 * doc/gcc/extensions-to-the-c-language-family/conditionals-with-omitted-operands.rst:
4145 Add trailing newline.
4146 * doc/gcc/extensions-to-the-c-language-family/constructing-function-calls.rst:
4147 Add trailing newline.
4148 * doc/gcc/extensions-to-the-c-language-family/decimal-floating-types.rst:
4149 Add trailing newline.
4150 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions.rst:
4151 Add trailing newline.
4152 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/aarch64-function-attributes.rst:
4153 Add trailing newline.
4154 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/amd-gcn-function-attributes.rst:
4155 Add trailing newline.
4156 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/arc-function-attributes.rst:
4157 Add trailing newline.
4158 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/arm-function-attributes.rst:
4159 Add trailing newline.
4160 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/avr-function-attributes.rst:
4161 Add trailing newline.
4162 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/blackfin-function-attributes.rst:
4163 Add trailing newline.
4164 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/bpf-function-attributes.rst:
4165 Add trailing newline.
4166 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/c-sky-function-attributes.rst:
4167 Add trailing newline.
4168 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/common-function-attributes.rst:
4169 Add trailing newline.
4170 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/epiphany-function-attributes.rst:
4171 Add trailing newline.
4172 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/h8-300-function-attributes.rst:
4173 Add trailing newline.
4174 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/ia-64-function-attributes.rst:
4175 Add trailing newline.
4176 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/m32c-function-attributes.rst:
4177 Add trailing newline.
4178 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/m32r-d-function-attributes.rst:
4179 Add trailing newline.
4180 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/m68k-function-attributes.rst:
4181 Add trailing newline.
4182 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/mcore-function-attributes.rst:
4183 Add trailing newline.
4184 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/mep-function-attributes.rst:
4185 Add trailing newline.
4186 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/microblaze-function-attributes.rst:
4187 Add trailing newline.
4188 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/microsoft-windows-function-attributes.rst:
4189 Add trailing newline.
4190 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/mips-function-attributes.rst:
4191 Add trailing newline.
4192 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/msp430-function-attributes.rst:
4193 Add trailing newline.
4194 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/nds32-function-attributes.rst:
4195 Add trailing newline.
4196 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/nios-ii-function-attributes.rst:
4197 Add trailing newline.
4198 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/nvidia-ptx-function-attributes.rst:
4199 Add trailing newline.
4200 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/powerpc-function-attributes.rst:
4201 Add trailing newline.
4202 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/risc-v-function-attributes.rst:
4203 Add trailing newline.
4204 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/rl78-function-attributes.rst:
4205 Add trailing newline.
4206 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/rx-function-attributes.rst:
4207 Add trailing newline.
4208 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/s-390-function-attributes.rst:
4209 Add trailing newline.
4210 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/sh-function-attributes.rst:
4211 Add trailing newline.
4212 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/symbian-os-function-attributes.rst:
4213 Add trailing newline.
4214 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/v850-function-attributes.rst:
4215 Add trailing newline.
4216 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/visium-function-attributes.rst:
4217 Add trailing newline.
4218 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/x86-function-attributes.rst:
4219 Add trailing newline.
4220 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/xstormy16-function-attributes.rst:
4221 Add trailing newline.
4222 * doc/gcc/extensions-to-the-c-language-family/designated-initializers.rst:
4223 Add trailing newline.
4224 * doc/gcc/extensions-to-the-c-language-family/determining-the-alignment-of-functions-types-or-variables.rst:
4225 Add trailing newline.
4226 * doc/gcc/extensions-to-the-c-language-family/dollar-signs-in-identifier-names.rst:
4227 Add trailing newline.
4228 * doc/gcc/extensions-to-the-c-language-family/double-word-integers.rst:
4229 Add trailing newline.
4230 * doc/gcc/extensions-to-the-c-language-family/enumerator-attributes.rst:
4231 Add trailing newline.
4232 * doc/gcc/extensions-to-the-c-language-family/fixed-point-types.rst:
4233 Add trailing newline.
4234 * doc/gcc/extensions-to-the-c-language-family/format-checks-specific-to-particular-target-machines.rst:
4235 Add trailing newline.
4236 * doc/gcc/extensions-to-the-c-language-family/function-names-as-strings.rst:
4237 Add trailing newline.
4238 * doc/gcc/extensions-to-the-c-language-family/getting-the-return-or-frame-address-of-a-function.rst:
4239 Add trailing newline.
4240 * doc/gcc/extensions-to-the-c-language-family/half-precision-floating-point.rst:
4241 Add trailing newline.
4242 * doc/gcc/extensions-to-the-c-language-family/hex-floats.rst:
4243 Add trailing newline.
4244 * doc/gcc/extensions-to-the-c-language-family/how-to-use-inline-assembly-language-in-c-code.rst:
4245 Add trailing newline.
4246 * doc/gcc/extensions-to-the-c-language-family/incomplete-enum-types.rst:
4247 Add trailing newline.
4248 * doc/gcc/extensions-to-the-c-language-family/label-attributes.rst:
4249 Add trailing newline.
4250 * doc/gcc/extensions-to-the-c-language-family/labels-as-values.rst:
4251 Add trailing newline.
4252 * doc/gcc/extensions-to-the-c-language-family/legacy-sync-built-in-functions-for-atomic-memory-access.rst:
4253 Add trailing newline.
4254 * doc/gcc/extensions-to-the-c-language-family/locally-declared-labels.rst:
4255 Add trailing newline.
4256 * doc/gcc/extensions-to-the-c-language-family/macros-with-a-variable-number-of-arguments.rst:
4257 Add trailing newline.
4258 * doc/gcc/extensions-to-the-c-language-family/mixed-declarations-labels-and-code.rst:
4259 Add trailing newline.
4260 * doc/gcc/extensions-to-the-c-language-family/named-address-spaces.rst:
4261 Add trailing newline.
4262 * doc/gcc/extensions-to-the-c-language-family/nested-functions.rst:
4263 Add trailing newline.
4264 * doc/gcc/extensions-to-the-c-language-family/non-constant-initializers.rst:
4265 Add trailing newline.
4266 * doc/gcc/extensions-to-the-c-language-family/non-lvalue-arrays-may-have-subscripts.rst:
4267 Add trailing newline.
4268 * doc/gcc/extensions-to-the-c-language-family/nonlocal-gotos.rst:
4269 Add trailing newline.
4270 * doc/gcc/extensions-to-the-c-language-family/object-size-checking-built-in-functions.rst:
4271 Add trailing newline.
4272 * doc/gcc/extensions-to-the-c-language-family/other-built-in-functions-provided-by-gcc.rst:
4273 Add trailing newline.
4274 * doc/gcc/extensions-to-the-c-language-family/pointer-arguments-in-variadic-functions.rst:
4275 Add trailing newline.
4276 * doc/gcc/extensions-to-the-c-language-family/pointers-to-arrays-with-qualifiers-work-as-expected.rst:
4277 Add trailing newline.
4278 * doc/gcc/extensions-to-the-c-language-family/pragmas-accepted-by-gcc.rst:
4279 Add trailing newline.
4280 * doc/gcc/extensions-to-the-c-language-family/prototypes-and-old-style-function-definitions.rst:
4281 Add trailing newline.
4282 * doc/gcc/extensions-to-the-c-language-family/referring-to-a-type-with-typeof.rst:
4283 Add trailing newline.
4284 * doc/gcc/extensions-to-the-c-language-family/slightly-looser-rules-for-escaped-newlines.rst:
4285 Add trailing newline.
4286 * doc/gcc/extensions-to-the-c-language-family/specifying-attributes-of-types.rst:
4287 Add trailing newline.
4288 * doc/gcc/extensions-to-the-c-language-family/specifying-attributes-of-variables.rst:
4289 Add trailing newline.
4290 * doc/gcc/extensions-to-the-c-language-family/statement-attributes.rst:
4291 Add trailing newline.
4292 * doc/gcc/extensions-to-the-c-language-family/statements-and-declarations-in-expressions.rst:
4293 Add trailing newline.
4294 * doc/gcc/extensions-to-the-c-language-family/structures-with-no-members.rst:
4295 Add trailing newline.
4296 * doc/gcc/extensions-to-the-c-language-family/support-for-offsetof.rst:
4297 Add trailing newline.
4298 * doc/gcc/extensions-to-the-c-language-family/target-builtins.rst:
4299 Add trailing newline.
4300 * doc/gcc/extensions-to-the-c-language-family/target-builtins/aarch64-built-in-functions.rst:
4301 Add trailing newline.
4302 * doc/gcc/extensions-to-the-c-language-family/target-builtins/alpha-built-in-functions.rst:
4303 Add trailing newline.
4304 * doc/gcc/extensions-to-the-c-language-family/target-builtins/altera-nios-ii-built-in-functions.rst:
4305 Add trailing newline.
4306 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arc-built-in-functions.rst:
4307 Add trailing newline.
4308 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arc-simd-built-in-functions.rst:
4309 Add trailing newline.
4310 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arm-armv8-m-security-extensions.rst:
4311 Add trailing newline.
4312 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arm-c-language-extensions-acle.rst:
4313 Add trailing newline.
4314 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arm-floating-point-status-and-control-intrinsics.rst:
4315 Add trailing newline.
4316 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arm-iwmmxt-built-in-functions.rst:
4317 Add trailing newline.
4318 * doc/gcc/extensions-to-the-c-language-family/target-builtins/avr-built-in-functions.rst:
4319 Add trailing newline.
4320 * doc/gcc/extensions-to-the-c-language-family/target-builtins/basic-powerpc-built-in-functions.rst:
4321 Add trailing newline.
4322 * doc/gcc/extensions-to-the-c-language-family/target-builtins/blackfin-built-in-functions.rst:
4323 Add trailing newline.
4324 * doc/gcc/extensions-to-the-c-language-family/target-builtins/bpf-built-in-functions.rst:
4325 Add trailing newline.
4326 * doc/gcc/extensions-to-the-c-language-family/target-builtins/fr-v-built-in-functions.rst:
4327 Add trailing newline.
4328 * doc/gcc/extensions-to-the-c-language-family/target-builtins/mips-dsp-built-in-functions.rst:
4329 Add trailing newline.
4330 * doc/gcc/extensions-to-the-c-language-family/target-builtins/mips-loongson-built-in-functions.rst:
4331 Add trailing newline.
4332 * doc/gcc/extensions-to-the-c-language-family/target-builtins/mips-paired-single-support.rst:
4333 Add trailing newline.
4334 * doc/gcc/extensions-to-the-c-language-family/target-builtins/mips-simd-architecture-msa-support.rst:
4335 Add trailing newline.
4336 * doc/gcc/extensions-to-the-c-language-family/target-builtins/msp430-built-in-functions.rst:
4337 Add trailing newline.
4338 * doc/gcc/extensions-to-the-c-language-family/target-builtins/nds32-built-in-functions.rst:
4339 Add trailing newline.
4340 * doc/gcc/extensions-to-the-c-language-family/target-builtins/other-mips-built-in-functions.rst:
4341 Add trailing newline.
4342 * doc/gcc/extensions-to-the-c-language-family/target-builtins/picochip-built-in-functions.rst:
4343 Add trailing newline.
4344 * doc/gcc/extensions-to-the-c-language-family/target-builtins/powerpc-altivec-vsx-built-in-functions.rst:
4345 Add trailing newline.
4346 * doc/gcc/extensions-to-the-c-language-family/target-builtins/powerpc-atomic-memory-operation-functions.rst:
4347 Add trailing newline.
4348 * doc/gcc/extensions-to-the-c-language-family/target-builtins/powerpc-hardware-transactional-memory-built-in-functions.rst:
4349 Add trailing newline.
4350 * doc/gcc/extensions-to-the-c-language-family/target-builtins/powerpc-matrix-multiply-assist-built-in-functions.rst:
4351 Add trailing newline.
4352 * doc/gcc/extensions-to-the-c-language-family/target-builtins/pru-built-in-functions.rst:
4353 Add trailing newline.
4354 * doc/gcc/extensions-to-the-c-language-family/target-builtins/risc-v-built-in-functions.rst:
4355 Add trailing newline.
4356 * doc/gcc/extensions-to-the-c-language-family/target-builtins/rx-built-in-functions.rst:
4357 Add trailing newline.
4358 * doc/gcc/extensions-to-the-c-language-family/target-builtins/s-390-system-z-built-in-functions.rst:
4359 Add trailing newline.
4360 * doc/gcc/extensions-to-the-c-language-family/target-builtins/sh-built-in-functions.rst:
4361 Add trailing newline.
4362 * doc/gcc/extensions-to-the-c-language-family/target-builtins/sparc-vis-built-in-functions.rst:
4363 Add trailing newline.
4364 * doc/gcc/extensions-to-the-c-language-family/target-builtins/ti-c6x-built-in-functions.rst:
4365 Add trailing newline.
4366 * doc/gcc/extensions-to-the-c-language-family/target-builtins/x86-built-in-functions.rst:
4367 Add trailing newline.
4368 * doc/gcc/extensions-to-the-c-language-family/target-builtins/x86-control-flow-protection-intrinsics.rst:
4369 Add trailing newline.
4370 * doc/gcc/extensions-to-the-c-language-family/target-builtins/x86-transactional-memory-intrinsics.rst:
4371 Add trailing newline.
4372 * doc/gcc/extensions-to-the-c-language-family/the-character-esc-in-constants.rst:
4373 Add trailing newline.
4374 * doc/gcc/extensions-to-the-c-language-family/thread-local-storage.rst:
4375 Add trailing newline.
4376 * doc/gcc/extensions-to-the-c-language-family/unnamed-structure-and-union-fields.rst:
4377 Add trailing newline.
4378 * doc/gcc/extensions-to-the-c-language-family/using-vector-instructions-through-built-in-functions.rst:
4379 Add trailing newline.
4380 * doc/gcc/extensions-to-the-c-language-family/when-is-a-volatile-object-accessed.rst:
4381 Add trailing newline.
4382 * doc/gcc/extensions-to-the-c-language-family/x86-specific-memory-model-extensions-for-transactional-memory.rst:
4383 Add trailing newline.
4384 * doc/gcc/funding.rst:
4385 Add trailing newline.
4386 * doc/gcc/gcc-command-options.rst:
4387 Add trailing newline.
4388 * doc/gcc/gcc-command-options/compiling-c++-programs.rst:
4389 Add trailing newline.
4390 * doc/gcc/gcc-command-options/description.rst:
4391 Add trailing newline.
4392 * doc/gcc/gcc-command-options/environment-variables-affecting-gcc.rst:
4393 Add trailing newline.
4394 * doc/gcc/gcc-command-options/gcc-developer-options.rst:
4395 Add trailing newline.
4396 * doc/gcc/gcc-command-options/machine-dependent-options.rst:
4397 Add trailing newline.
4398 * doc/gcc/gcc-command-options/machine-dependent-options/aarch64-options.rst:
4399 Add trailing newline.
4400 * doc/gcc/gcc-command-options/machine-dependent-options/adapteva-epiphany-options.rst:
4401 Add trailing newline.
4402 * doc/gcc/gcc-command-options/machine-dependent-options/amd-gcn-options.rst:
4403 Add trailing newline.
4404 * doc/gcc/gcc-command-options/machine-dependent-options/arc-options.rst:
4405 Add trailing newline.
4406 * doc/gcc/gcc-command-options/machine-dependent-options/arm-options.rst:
4407 Add trailing newline.
4408 * doc/gcc/gcc-command-options/machine-dependent-options/avr-options.rst:
4409 Add trailing newline.
4410 * doc/gcc/gcc-command-options/machine-dependent-options/blackfin-options.rst:
4411 Add trailing newline.
4412 * doc/gcc/gcc-command-options/machine-dependent-options/c-sky-options.rst:
4413 Add trailing newline.
4414 * doc/gcc/gcc-command-options/machine-dependent-options/c6x-options.rst:
4415 Add trailing newline.
4416 * doc/gcc/gcc-command-options/machine-dependent-options/cris-options.rst:
4417 Add trailing newline.
4418 * doc/gcc/gcc-command-options/machine-dependent-options/darwin-options.rst:
4419 Add trailing newline.
4420 * doc/gcc/gcc-command-options/machine-dependent-options/dec-alpha-options.rst:
4421 Add trailing newline.
4422 * doc/gcc/gcc-command-options/machine-dependent-options/ebpf-options.rst:
4423 Add trailing newline.
4424 * doc/gcc/gcc-command-options/machine-dependent-options/fr30-options.rst:
4425 Add trailing newline.
4426 * doc/gcc/gcc-command-options/machine-dependent-options/frv-options.rst:
4427 Add trailing newline.
4428 * doc/gcc/gcc-command-options/machine-dependent-options/ft32-options.rst:
4429 Add trailing newline.
4430 * doc/gcc/gcc-command-options/machine-dependent-options/gnu-linux-options.rst:
4431 Add trailing newline.
4432 * doc/gcc/gcc-command-options/machine-dependent-options/h8-300-options.rst:
4433 Add trailing newline.
4434 * doc/gcc/gcc-command-options/machine-dependent-options/hppa-options.rst:
4435 Add trailing newline.
4436 * doc/gcc/gcc-command-options/machine-dependent-options/ia-64-options.rst:
4437 Add trailing newline.
4438 * doc/gcc/gcc-command-options/machine-dependent-options/ibm-rs-6000-and-powerpc-options.rst:
4439 Add trailing newline.
4440 * doc/gcc/gcc-command-options/machine-dependent-options/lm32-options.rst:
4441 Add trailing newline.
4442 * doc/gcc/gcc-command-options/machine-dependent-options/loongarch-options.rst:
4443 Add trailing newline.
4444 * doc/gcc/gcc-command-options/machine-dependent-options/m32c-options.rst:
4445 Add trailing newline.
4446 * doc/gcc/gcc-command-options/machine-dependent-options/m32r-d-options.rst:
4447 Add trailing newline.
4448 * doc/gcc/gcc-command-options/machine-dependent-options/m680x0-options.rst:
4449 Add trailing newline.
4450 * doc/gcc/gcc-command-options/machine-dependent-options/mcore-options.rst:
4451 Add trailing newline.
4452 * doc/gcc/gcc-command-options/machine-dependent-options/mep-options.rst:
4453 Add trailing newline.
4454 * doc/gcc/gcc-command-options/machine-dependent-options/microblaze-options.rst:
4455 Add trailing newline.
4456 * doc/gcc/gcc-command-options/machine-dependent-options/mips-options.rst:
4457 Add trailing newline.
4458 * doc/gcc/gcc-command-options/machine-dependent-options/mmix-options.rst:
4459 Add trailing newline.
4460 * doc/gcc/gcc-command-options/machine-dependent-options/mn10300-options.rst:
4461 Add trailing newline.
4462 * doc/gcc/gcc-command-options/machine-dependent-options/moxie-options.rst:
4463 Add trailing newline.
4464 * doc/gcc/gcc-command-options/machine-dependent-options/msp430-options.rst:
4465 Add trailing newline.
4466 * doc/gcc/gcc-command-options/machine-dependent-options/nds32-options.rst:
4467 Add trailing newline.
4468 * doc/gcc/gcc-command-options/machine-dependent-options/nios-ii-options.rst:
4469 Add trailing newline.
4470 * doc/gcc/gcc-command-options/machine-dependent-options/nvidia-ptx-options.rst:
4471 Add trailing newline.
4472 * doc/gcc/gcc-command-options/machine-dependent-options/openrisc-options.rst:
4473 Add trailing newline.
4474 * doc/gcc/gcc-command-options/machine-dependent-options/options-for-system-v.rst:
4475 Add trailing newline.
4476 * doc/gcc/gcc-command-options/machine-dependent-options/pdp-11-options.rst:
4477 Add trailing newline.
4478 * doc/gcc/gcc-command-options/machine-dependent-options/picochip-options.rst:
4479 Add trailing newline.
4480 * doc/gcc/gcc-command-options/machine-dependent-options/powerpc-options.rst:
4481 Add trailing newline.
4482 * doc/gcc/gcc-command-options/machine-dependent-options/pru-options.rst:
4483 Add trailing newline.
4484 * doc/gcc/gcc-command-options/machine-dependent-options/risc-v-options.rst:
4485 Add trailing newline.
4486 * doc/gcc/gcc-command-options/machine-dependent-options/rl78-options.rst:
4487 Add trailing newline.
4488 * doc/gcc/gcc-command-options/machine-dependent-options/rx-options.rst:
4489 Add trailing newline.
4490 * doc/gcc/gcc-command-options/machine-dependent-options/s-390-and-zseries-options.rst:
4491 Add trailing newline.
4492 * doc/gcc/gcc-command-options/machine-dependent-options/score-options.rst:
4493 Add trailing newline.
4494 * doc/gcc/gcc-command-options/machine-dependent-options/sh-options.rst:
4495 Add trailing newline.
4496 * doc/gcc/gcc-command-options/machine-dependent-options/solaris-2-options.rst:
4497 Add trailing newline.
4498 * doc/gcc/gcc-command-options/machine-dependent-options/sparc-options.rst:
4499 Add trailing newline.
4500 * doc/gcc/gcc-command-options/machine-dependent-options/v850-options.rst:
4501 Add trailing newline.
4502 * doc/gcc/gcc-command-options/machine-dependent-options/vax-options.rst:
4503 Add trailing newline.
4504 * doc/gcc/gcc-command-options/machine-dependent-options/visium-options.rst:
4505 Add trailing newline.
4506 * doc/gcc/gcc-command-options/machine-dependent-options/vms-options.rst:
4507 Add trailing newline.
4508 * doc/gcc/gcc-command-options/machine-dependent-options/vxworks-options.rst:
4509 Add trailing newline.
4510 * doc/gcc/gcc-command-options/machine-dependent-options/x86-windows-options.rst:
4511 Add trailing newline.
4512 * doc/gcc/gcc-command-options/machine-dependent-options/xstormy16-options.rst:
4513 Add trailing newline.
4514 * doc/gcc/gcc-command-options/machine-dependent-options/xtensa-options.rst:
4515 Add trailing newline.
4516 * doc/gcc/gcc-command-options/machine-dependent-options/zseries-options.rst:
4517 Add trailing newline.
4518 * doc/gcc/gcc-command-options/options-controlling-c++-dialect.rst:
4519 Add trailing newline.
4520 * doc/gcc/gcc-command-options/options-controlling-c-dialect.rst:
4521 Add trailing newline.
4522 * doc/gcc/gcc-command-options/options-controlling-objective-c-and-objective-c++-dialects.rst:
4523 Add trailing newline.
4524 * doc/gcc/gcc-command-options/options-controlling-the-kind-of-output.rst:
4525 Add trailing newline.
4526 * doc/gcc/gcc-command-options/options-controlling-the-preprocessor.rst:
4527 Add trailing newline.
4528 * doc/gcc/gcc-command-options/options-for-code-generation-conventions.rst:
4529 Add trailing newline.
4530 * doc/gcc/gcc-command-options/options-for-debugging-your-program.rst:
4531 Add trailing newline.
4532 * doc/gcc/gcc-command-options/options-for-directory-search.rst:
4533 Add trailing newline.
4534 * doc/gcc/gcc-command-options/options-for-linking.rst:
4535 Add trailing newline.
4536 * doc/gcc/gcc-command-options/options-that-control-optimization.rst:
4537 Add trailing newline.
4538 * doc/gcc/gcc-command-options/options-that-control-static-analysis.rst:
4539 Add trailing newline.
4540 * doc/gcc/gcc-command-options/options-to-control-diagnostic-messages-formatting.rst:
4541 Add trailing newline.
4542 * doc/gcc/gcc-command-options/options-to-request-or-suppress-warnings.rst:
4543 Add trailing newline.
4544 * doc/gcc/gcc-command-options/passing-options-to-the-assembler.rst:
4545 Add trailing newline.
4546 * doc/gcc/gcc-command-options/program-instrumentation-options.rst:
4547 Add trailing newline.
4548 * doc/gcc/gcc-command-options/specifying-subprocesses-and-the-switches-to-pass-to-them.rst:
4549 Add trailing newline.
4550 * doc/gcc/gcc-command-options/using-precompiled-headers.rst:
4551 Add trailing newline.
4553 Add trailing newline.
4554 * doc/gcc/gcov-dump.rst:
4555 Add trailing newline.
4556 * doc/gcc/gcov-tool.rst:
4557 Add trailing newline.
4559 Add trailing newline.
4560 * doc/gcc/gcov/brief-description-of-gcov-data-files.rst:
4561 Add trailing newline.
4562 * doc/gcc/gcov/data-file-relocation-to-support-cross-profiling.rst:
4563 Add trailing newline.
4564 * doc/gcc/gcov/introduction-to-gcov.rst:
4565 Add trailing newline.
4566 * doc/gcc/gcov/invoking-gcov.rst:
4567 Add trailing newline.
4568 * doc/gcc/gcov/profiling-and-test-coverage-in-freestanding-environments.rst:
4569 Add trailing newline.
4570 * doc/gcc/gcov/using-gcov-with-gcc-optimization.rst:
4571 Add trailing newline.
4572 * doc/gcc/general-public-license-3.rst:
4573 Add trailing newline.
4574 * doc/gcc/gnu-free-documentation-license.rst:
4575 Add trailing newline.
4576 * doc/gcc/gnu-objective-c-features.rst:
4577 Add trailing newline.
4578 * doc/gcc/gnu-objective-c-features/compatibilityalias.rst:
4579 Add trailing newline.
4580 * doc/gcc/gnu-objective-c-features/constant-string-objects.rst:
4581 Add trailing newline.
4582 * doc/gcc/gnu-objective-c-features/exceptions.rst:
4583 Add trailing newline.
4584 * doc/gcc/gnu-objective-c-features/fast-enumeration.rst:
4585 Add trailing newline.
4586 * doc/gcc/gnu-objective-c-features/garbage-collection.rst:
4587 Add trailing newline.
4588 * doc/gcc/gnu-objective-c-features/gnu-objective-c-runtime-api.rst:
4589 Add trailing newline.
4590 * doc/gcc/gnu-objective-c-features/load-executing-code-before-main.rst:
4591 Add trailing newline.
4592 * doc/gcc/gnu-objective-c-features/messaging-with-the-gnu-objective-c-runtime.rst:
4593 Add trailing newline.
4594 * doc/gcc/gnu-objective-c-features/synchronization.rst:
4595 Add trailing newline.
4596 * doc/gcc/gnu-objective-c-features/type-encoding.rst:
4597 Add trailing newline.
4599 Add trailing newline.
4600 * doc/gcc/have-you-found-a-bug.rst:
4601 Add trailing newline.
4602 * doc/gcc/how-and-where-to-report-bugs.rst:
4603 Add trailing newline.
4604 * doc/gcc/how-to-get-help-with-gcc.rst:
4605 Add trailing newline.
4606 * doc/gcc/index.rst:
4607 Add trailing newline.
4608 * doc/gcc/indices-and-tables.rst:
4609 Add trailing newline.
4610 * doc/gcc/known-causes-of-trouble-with-gcc.rst:
4611 Add trailing newline.
4612 * doc/gcc/known-causes-of-trouble-with-gcc/actual-bugs-we-havent-fixed-yet.rst:
4613 Add trailing newline.
4614 * doc/gcc/known-causes-of-trouble-with-gcc/certain-changes-we-dont-want-to-make.rst:
4615 Add trailing newline.
4616 * doc/gcc/known-causes-of-trouble-with-gcc/common-misunderstandings-with-gnu-c.rst:
4617 Add trailing newline.
4618 * doc/gcc/known-causes-of-trouble-with-gcc/disappointments-and-misunderstandings.rst:
4619 Add trailing newline.
4620 * doc/gcc/known-causes-of-trouble-with-gcc/fixed-header-files.rst:
4621 Add trailing newline.
4622 * doc/gcc/known-causes-of-trouble-with-gcc/incompatibilities-of-gcc.rst:
4623 Add trailing newline.
4624 * doc/gcc/known-causes-of-trouble-with-gcc/interoperation.rst:
4625 Add trailing newline.
4626 * doc/gcc/known-causes-of-trouble-with-gcc/standard-libraries.rst:
4627 Add trailing newline.
4628 * doc/gcc/known-causes-of-trouble-with-gcc/warning-messages-and-error-messages.rst:
4629 Add trailing newline.
4630 * doc/gcc/language-standards-supported-by-gcc.rst:
4631 Add trailing newline.
4632 * doc/gcc/language-standards-supported-by-gcc/c++-language.rst:
4633 Add trailing newline.
4634 * doc/gcc/language-standards-supported-by-gcc/c-language.rst:
4635 Add trailing newline.
4636 * doc/gcc/language-standards-supported-by-gcc/d-language.rst:
4637 Add trailing newline.
4638 * doc/gcc/language-standards-supported-by-gcc/go-language.rst:
4639 Add trailing newline.
4640 * doc/gcc/language-standards-supported-by-gcc/objective-c-and-objective-c++-languages.rst:
4641 Add trailing newline.
4642 * doc/gcc/lto-dump.rst:
4643 Add trailing newline.
4644 * doc/gcc/programming-languages-supported-by-gcc.rst:
4645 Add trailing newline.
4646 * doc/gcc/reporting-bugs.rst:
4647 Add trailing newline.
4648 * doc/gccint/analysis-and-optimization-of-gimple-tuples.rst:
4649 Add trailing newline.
4650 * doc/gccint/analysis-and-optimization-of-gimple-tuples/alias-analysis.rst:
4651 Add trailing newline.
4652 * doc/gccint/analysis-and-optimization-of-gimple-tuples/annotations.rst:
4653 Add trailing newline.
4654 * doc/gccint/analysis-and-optimization-of-gimple-tuples/memory-model.rst:
4655 Add trailing newline.
4656 * doc/gccint/analysis-and-optimization-of-gimple-tuples/ssa-operands.rst:
4657 Add trailing newline.
4658 * doc/gccint/analysis-and-optimization-of-gimple-tuples/static-single-assignment.rst:
4659 Add trailing newline.
4660 * doc/gccint/analysis-and-representation-of-loops.rst:
4661 Add trailing newline.
4662 * doc/gccint/analysis-and-representation-of-loops/data-dependency-analysis.rst:
4663 Add trailing newline.
4664 * doc/gccint/analysis-and-representation-of-loops/iv-analysis-on-rtl.rst:
4665 Add trailing newline.
4666 * doc/gccint/analysis-and-representation-of-loops/loop-closed-ssa-form.rst:
4667 Add trailing newline.
4668 * doc/gccint/analysis-and-representation-of-loops/loop-manipulation.rst:
4669 Add trailing newline.
4670 * doc/gccint/analysis-and-representation-of-loops/loop-querying.rst:
4671 Add trailing newline.
4672 * doc/gccint/analysis-and-representation-of-loops/loop-representation.rst:
4673 Add trailing newline.
4674 * doc/gccint/analysis-and-representation-of-loops/number-of-iterations-analysis.rst:
4675 Add trailing newline.
4676 * doc/gccint/analysis-and-representation-of-loops/scalar-evolutions.rst:
4677 Add trailing newline.
4678 * doc/gccint/analyzer-internals.rst:
4679 Add trailing newline.
4680 * doc/gccint/collect2.rst:
4681 Add trailing newline.
4682 * doc/gccint/contributing-to-gcc-development.rst:
4683 Add trailing newline.
4684 * doc/gccint/contributors-to-gcc.rst:
4685 Add trailing newline.
4686 * doc/gccint/control-flow-graph.rst:
4687 Add trailing newline.
4688 * doc/gccint/control-flow-graph/basic-blocks.rst:
4689 Add trailing newline.
4690 * doc/gccint/control-flow-graph/edges.rst:
4691 Add trailing newline.
4692 * doc/gccint/control-flow-graph/liveness-information.rst:
4693 Add trailing newline.
4694 * doc/gccint/control-flow-graph/maintaining-the-cfg.rst:
4695 Add trailing newline.
4696 * doc/gccint/control-flow-graph/profile-information.rst:
4697 Add trailing newline.
4698 * doc/gccint/copyright.rst:
4699 Add trailing newline.
4700 * doc/gccint/debugging-the-analyzer.rst:
4701 Add trailing newline.
4702 * doc/gccint/funding.rst:
4703 Add trailing newline.
4704 * doc/gccint/gcc-and-portability.rst:
4705 Add trailing newline.
4706 * doc/gccint/general-public-license-3.rst:
4707 Add trailing newline.
4708 * doc/gccint/generic.rst:
4709 Add trailing newline.
4710 * doc/gccint/generic/attributes-in-trees.rst:
4711 Add trailing newline.
4712 * doc/gccint/generic/c-and-c++-trees.rst:
4713 Add trailing newline.
4714 * doc/gccint/generic/declarations.rst:
4715 Add trailing newline.
4716 * doc/gccint/generic/deficiencies.rst:
4717 Add trailing newline.
4718 * doc/gccint/generic/expressions.rst:
4719 Add trailing newline.
4720 * doc/gccint/generic/functions.rst:
4721 Add trailing newline.
4722 * doc/gccint/generic/language-dependent-trees.rst:
4723 Add trailing newline.
4724 * doc/gccint/generic/overview.rst:
4725 Add trailing newline.
4726 * doc/gccint/generic/statements.rst:
4727 Add trailing newline.
4728 * doc/gccint/generic/types.rst:
4729 Add trailing newline.
4730 * doc/gccint/gimple-api.rst:
4731 Add trailing newline.
4732 * doc/gccint/gimple.rst:
4733 Add trailing newline.
4734 * doc/gccint/gimple/adding-a-new-gimple-statement-code.rst:
4735 Add trailing newline.
4736 * doc/gccint/gimple/class-hierarchy-of-gimple-statements.rst:
4737 Add trailing newline.
4738 * doc/gccint/gimple/exception-handling.rst:
4739 Add trailing newline.
4740 * doc/gccint/gimple/gimple-instruction-set.rst:
4741 Add trailing newline.
4742 * doc/gccint/gimple/gimple-sequences.rst:
4743 Add trailing newline.
4744 * doc/gccint/gimple/manipulating-gimple-statements.rst:
4745 Add trailing newline.
4746 * doc/gccint/gimple/operands.rst:
4747 Add trailing newline.
4748 * doc/gccint/gimple/sequence-iterators.rst:
4749 Add trailing newline.
4750 * doc/gccint/gimple/statement-and-operand-traversals.rst:
4751 Add trailing newline.
4752 * doc/gccint/gimple/temporaries.rst:
4753 Add trailing newline.
4754 * doc/gccint/gimple/tuple-representation.rst:
4755 Add trailing newline.
4756 * doc/gccint/gimple/tuple-specific-accessors.rst:
4757 Add trailing newline.
4758 * doc/gccint/gimple/tuple-specific-accessors/gimpleasm.rst:
4759 Add trailing newline.
4760 * doc/gccint/gimple/tuple-specific-accessors/gimpleassign.rst:
4761 Add trailing newline.
4762 * doc/gccint/gimple/tuple-specific-accessors/gimplebind.rst:
4763 Add trailing newline.
4764 * doc/gccint/gimple/tuple-specific-accessors/gimplecall.rst:
4765 Add trailing newline.
4766 * doc/gccint/gimple/tuple-specific-accessors/gimplecatch.rst:
4767 Add trailing newline.
4768 * doc/gccint/gimple/tuple-specific-accessors/gimplecond.rst:
4769 Add trailing newline.
4770 * doc/gccint/gimple/tuple-specific-accessors/gimpledebug.rst:
4771 Add trailing newline.
4772 * doc/gccint/gimple/tuple-specific-accessors/gimpleehfilter.rst:
4773 Add trailing newline.
4774 * doc/gccint/gimple/tuple-specific-accessors/gimplegoto.rst:
4775 Add trailing newline.
4776 * doc/gccint/gimple/tuple-specific-accessors/gimplelabel.rst:
4777 Add trailing newline.
4778 * doc/gccint/gimple/tuple-specific-accessors/gimplenop.rst:
4779 Add trailing newline.
4780 * doc/gccint/gimple/tuple-specific-accessors/gimpleompatomicload.rst:
4781 Add trailing newline.
4782 * doc/gccint/gimple/tuple-specific-accessors/gimpleompatomicstore.rst:
4783 Add trailing newline.
4784 * doc/gccint/gimple/tuple-specific-accessors/gimpleompcontinue.rst:
4785 Add trailing newline.
4786 * doc/gccint/gimple/tuple-specific-accessors/gimpleompcritical.rst:
4787 Add trailing newline.
4788 * doc/gccint/gimple/tuple-specific-accessors/gimpleompfor.rst:
4789 Add trailing newline.
4790 * doc/gccint/gimple/tuple-specific-accessors/gimpleompmaster.rst:
4791 Add trailing newline.
4792 * doc/gccint/gimple/tuple-specific-accessors/gimpleompordered.rst:
4793 Add trailing newline.
4794 * doc/gccint/gimple/tuple-specific-accessors/gimpleompparallel.rst:
4795 Add trailing newline.
4796 * doc/gccint/gimple/tuple-specific-accessors/gimpleompreturn.rst:
4797 Add trailing newline.
4798 * doc/gccint/gimple/tuple-specific-accessors/gimpleompsection.rst:
4799 Add trailing newline.
4800 * doc/gccint/gimple/tuple-specific-accessors/gimpleompsections.rst:
4801 Add trailing newline.
4802 * doc/gccint/gimple/tuple-specific-accessors/gimpleompsingle.rst:
4803 Add trailing newline.
4804 * doc/gccint/gimple/tuple-specific-accessors/gimplephi.rst:
4805 Add trailing newline.
4806 * doc/gccint/gimple/tuple-specific-accessors/gimpleresx.rst:
4807 Add trailing newline.
4808 * doc/gccint/gimple/tuple-specific-accessors/gimplereturn.rst:
4809 Add trailing newline.
4810 * doc/gccint/gimple/tuple-specific-accessors/gimpleswitch.rst:
4811 Add trailing newline.
4812 * doc/gccint/gimple/tuple-specific-accessors/gimpletry.rst:
4813 Add trailing newline.
4814 * doc/gccint/gimple/tuple-specific-accessors/gimplewithcleanupexpr.rst:
4815 Add trailing newline.
4816 * doc/gccint/gnu-free-documentation-license.rst:
4817 Add trailing newline.
4818 * doc/gccint/guidelines-for-diagnostics.rst:
4819 Add trailing newline.
4820 * doc/gccint/guidelines-for-options.rst:
4821 Add trailing newline.
4822 * doc/gccint/host-common.rst:
4823 Add trailing newline.
4824 * doc/gccint/host-configuration.rst:
4825 Add trailing newline.
4826 * doc/gccint/host-filesystem.rst:
4827 Add trailing newline.
4828 * doc/gccint/host-makefile-fragments.rst:
4829 Add trailing newline.
4830 * doc/gccint/host-misc.rst:
4831 Add trailing newline.
4832 * doc/gccint/index.rst:
4833 Add trailing newline.
4834 * doc/gccint/indices-and-tables.rst:
4835 Add trailing newline.
4836 * doc/gccint/interfacing-to-gcc-output.rst:
4837 Add trailing newline.
4838 * doc/gccint/introduction.rst:
4839 Add trailing newline.
4840 * doc/gccint/language-front-ends-in-gcc.rst:
4841 Add trailing newline.
4842 * doc/gccint/link-time-optimization.rst:
4843 Add trailing newline.
4844 * doc/gccint/link-time-optimization/design-overview.rst:
4845 Add trailing newline.
4846 * doc/gccint/link-time-optimization/internal-flags-controlling-lto1.rst:
4847 Add trailing newline.
4848 * doc/gccint/link-time-optimization/lto-file-sections.rst:
4849 Add trailing newline.
4850 * doc/gccint/link-time-optimization/using-summary-information-in-ipa-passes.rst:
4851 Add trailing newline.
4852 * doc/gccint/link-time-optimization/whole-program-assumptions-linker-plugin-and-symbol-visibilities.rst:
4853 Add trailing newline.
4854 * doc/gccint/machine-descriptions.rst:
4855 Add trailing newline.
4856 * doc/gccint/machine-descriptions/c-statements-for-assembler-output.rst:
4857 Add trailing newline.
4858 * doc/gccint/machine-descriptions/canonicalization-of-instructions.rst:
4859 Add trailing newline.
4860 * doc/gccint/machine-descriptions/conditional-execution.rst:
4861 Add trailing newline.
4862 * doc/gccint/machine-descriptions/constant-definitions.rst:
4863 Add trailing newline.
4864 * doc/gccint/machine-descriptions/defining-how-to-split-instructions.rst:
4865 Add trailing newline.
4866 * doc/gccint/machine-descriptions/defining-jump-instruction-patterns.rst:
4867 Add trailing newline.
4868 * doc/gccint/machine-descriptions/defining-looping-instruction-patterns.rst:
4869 Add trailing newline.
4870 * doc/gccint/machine-descriptions/defining-rtl-sequences-for-code-generation.rst:
4871 Add trailing newline.
4872 * doc/gccint/machine-descriptions/everything-about-instruction-patterns.rst:
4873 Add trailing newline.
4874 * doc/gccint/machine-descriptions/example-of-defineinsn.rst:
4875 Add trailing newline.
4876 * doc/gccint/machine-descriptions/including-patterns-in-machine-descriptions.rst:
4877 Add trailing newline.
4878 * doc/gccint/machine-descriptions/instruction-attributes.rst:
4879 Add trailing newline.
4880 * doc/gccint/machine-descriptions/interdependence-of-patterns.rst:
4881 Add trailing newline.
4882 * doc/gccint/machine-descriptions/iterators.rst:
4883 Add trailing newline.
4884 * doc/gccint/machine-descriptions/machine-specific-peephole-optimizers.rst:
4885 Add trailing newline.
4886 * doc/gccint/machine-descriptions/operand-constraints.rst:
4887 Add trailing newline.
4888 * doc/gccint/machine-descriptions/output-templates-and-operand-substitution.rst:
4889 Add trailing newline.
4890 * doc/gccint/machine-descriptions/overview-of-how-the-machine-description-is-used.rst:
4891 Add trailing newline.
4892 * doc/gccint/machine-descriptions/predicates.rst:
4893 Add trailing newline.
4894 * doc/gccint/machine-descriptions/rtl-template.rst:
4895 Add trailing newline.
4896 * doc/gccint/machine-descriptions/rtl-templates-transformations.rst:
4897 Add trailing newline.
4898 * doc/gccint/machine-descriptions/standard-pattern-names-for-generation.rst:
4899 Add trailing newline.
4900 * doc/gccint/machine-descriptions/when-the-order-of-patterns-matters.rst:
4901 Add trailing newline.
4902 * doc/gccint/makefile-fragments.rst:
4903 Add trailing newline.
4904 * doc/gccint/match-and-simplify.rst:
4905 Add trailing newline.
4906 * doc/gccint/memory-management-and-type-information.rst:
4907 Add trailing newline.
4908 * doc/gccint/memory-management-and-type-information/how-to-invoke-the-garbage-collector.rst:
4909 Add trailing newline.
4910 * doc/gccint/memory-management-and-type-information/marking-roots-for-the-garbage-collector.rst:
4911 Add trailing newline.
4912 * doc/gccint/memory-management-and-type-information/source-files-containing-type-information.rst:
4913 Add trailing newline.
4914 * doc/gccint/memory-management-and-type-information/support-for-inheritance.rst:
4915 Add trailing newline.
4916 * doc/gccint/memory-management-and-type-information/support-for-user-provided-gc-marking-routines.rst:
4917 Add trailing newline.
4918 * doc/gccint/memory-management-and-type-information/the-inside-of-a-gty.rst:
4919 Add trailing newline.
4920 * doc/gccint/memory-management-and-type-information/troubleshooting-the-garbage-collector.rst:
4921 Add trailing newline.
4922 * doc/gccint/option-file-format.rst:
4923 Add trailing newline.
4924 * doc/gccint/option-properties.rst:
4925 Add trailing newline.
4926 * doc/gccint/option-specification-files.rst:
4927 Add trailing newline.
4928 * doc/gccint/passes-and-files-of-the-compiler.rst:
4929 Add trailing newline.
4930 * doc/gccint/passes-and-files-of-the-compiler/gimplification-pass.rst:
4931 Add trailing newline.
4932 * doc/gccint/passes-and-files-of-the-compiler/inter-procedural-optimization-passes.rst:
4933 Add trailing newline.
4934 * doc/gccint/passes-and-files-of-the-compiler/optimization-info.rst:
4935 Add trailing newline.
4936 * doc/gccint/passes-and-files-of-the-compiler/parsing-pass.rst:
4937 Add trailing newline.
4938 * doc/gccint/passes-and-files-of-the-compiler/pass-manager.rst:
4939 Add trailing newline.
4940 * doc/gccint/passes-and-files-of-the-compiler/rtl-passes.rst:
4941 Add trailing newline.
4942 * doc/gccint/passes-and-files-of-the-compiler/tree-ssa-passes.rst:
4943 Add trailing newline.
4944 * doc/gccint/plugins.rst:
4945 Add trailing newline.
4946 * doc/gccint/plugins/building-gcc-plugins.rst:
4947 Add trailing newline.
4948 * doc/gccint/plugins/controlling-which-passes-are-being-run.rst:
4949 Add trailing newline.
4950 * doc/gccint/plugins/giving-information-about-a-plugin.rst:
4951 Add trailing newline.
4952 * doc/gccint/plugins/interacting-with-the-gcc-garbage-collector.rst:
4953 Add trailing newline.
4954 * doc/gccint/plugins/interacting-with-the-pass-manager.rst:
4955 Add trailing newline.
4956 * doc/gccint/plugins/keeping-track-of-available-passes.rst:
4957 Add trailing newline.
4958 * doc/gccint/plugins/loading-plugins.rst:
4959 Add trailing newline.
4960 * doc/gccint/plugins/plugin-api.rst:
4961 Add trailing newline.
4962 * doc/gccint/plugins/recording-information-about-pass-execution.rst:
4963 Add trailing newline.
4964 * doc/gccint/plugins/registering-custom-attributes-or-pragmas.rst:
4965 Add trailing newline.
4966 * doc/gccint/rtl-representation.rst:
4967 Add trailing newline.
4968 * doc/gccint/rtl-representation/access-to-operands.rst:
4969 Add trailing newline.
4970 * doc/gccint/rtl-representation/access-to-special-operands.rst:
4971 Add trailing newline.
4972 * doc/gccint/rtl-representation/assembler-instructions-as-expressions.rst:
4973 Add trailing newline.
4974 * doc/gccint/rtl-representation/bit-fields.rst:
4975 Add trailing newline.
4976 * doc/gccint/rtl-representation/comparison-operations.rst:
4977 Add trailing newline.
4978 * doc/gccint/rtl-representation/constant-expression-types.rst:
4979 Add trailing newline.
4980 * doc/gccint/rtl-representation/conversions.rst:
4981 Add trailing newline.
4982 * doc/gccint/rtl-representation/declarations.rst:
4983 Add trailing newline.
4984 * doc/gccint/rtl-representation/embedded-side-effects-on-addresses.rst:
4985 Add trailing newline.
4986 * doc/gccint/rtl-representation/flags-in-an-rtl-expression.rst:
4987 Add trailing newline.
4988 * doc/gccint/rtl-representation/insns.rst:
4989 Add trailing newline.
4990 * doc/gccint/rtl-representation/machine-modes.rst:
4991 Add trailing newline.
4992 * doc/gccint/rtl-representation/on-the-side-ssa-form-for-rtl.rst:
4993 Add trailing newline.
4994 * doc/gccint/rtl-representation/reading-rtl.rst:
4995 Add trailing newline.
4996 * doc/gccint/rtl-representation/registers-and-memory.rst:
4997 Add trailing newline.
4998 * doc/gccint/rtl-representation/rtl-classes-and-formats.rst:
4999 Add trailing newline.
5000 * doc/gccint/rtl-representation/rtl-expressions-for-arithmetic.rst:
5001 Add trailing newline.
5002 * doc/gccint/rtl-representation/rtl-object-types.rst:
5003 Add trailing newline.
5004 * doc/gccint/rtl-representation/rtl-representation-of-function-call-insns.rst:
5005 Add trailing newline.
5006 * doc/gccint/rtl-representation/side-effect-expressions.rst:
5007 Add trailing newline.
5008 * doc/gccint/rtl-representation/structure-sharing-assumptions.rst:
5009 Add trailing newline.
5010 * doc/gccint/rtl-representation/variable-location-debug-information-in-rtl.rst:
5011 Add trailing newline.
5012 * doc/gccint/rtl-representation/vector-operations.rst:
5013 Add trailing newline.
5014 * doc/gccint/sizes-and-offsets-as-runtime-invariants.rst:
5015 Add trailing newline.
5016 * doc/gccint/sizes-and-offsets-as-runtime-invariants/alignment-of-polyints.rst:
5017 Add trailing newline.
5018 * doc/gccint/sizes-and-offsets-as-runtime-invariants/arithmetic-on-polyints.rst:
5019 Add trailing newline.
5020 * doc/gccint/sizes-and-offsets-as-runtime-invariants/comparisons-involving-polyint.rst:
5021 Add trailing newline.
5022 * doc/gccint/sizes-and-offsets-as-runtime-invariants/computing-bounds-on-polyints.rst:
5023 Add trailing newline.
5024 * doc/gccint/sizes-and-offsets-as-runtime-invariants/consequences-of-using-polyint.rst:
5025 Add trailing newline.
5026 * doc/gccint/sizes-and-offsets-as-runtime-invariants/converting-polyints.rst:
5027 Add trailing newline.
5028 * doc/gccint/sizes-and-offsets-as-runtime-invariants/guidelines-for-using-polyint.rst:
5029 Add trailing newline.
5030 * doc/gccint/sizes-and-offsets-as-runtime-invariants/miscellaneous-polyint-routines.rst:
5031 Add trailing newline.
5032 * doc/gccint/sizes-and-offsets-as-runtime-invariants/overview-of-polyint.rst:
5033 Add trailing newline.
5034 * doc/gccint/source-tree-structure-and-build-system.rst:
5035 Add trailing newline.
5036 * doc/gccint/source-tree-structure-and-build-system/configure-terms-and-history.rst:
5037 Add trailing newline.
5038 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory.rst:
5039 Add trailing newline.
5040 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/anatomy-of-a-language-front-end.rst:
5041 Add trailing newline.
5042 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/anatomy-of-a-target-back-end.rst:
5043 Add trailing newline.
5044 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/build-system-in-the-gcc-directory.rst:
5045 Add trailing newline.
5046 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/configuration-in-the-gcc-directory.rst:
5047 Add trailing newline.
5048 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/headers-installed-by-gcc.rst:
5049 Add trailing newline.
5050 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/library-source-files-and-headers-under-the-gcc-directory.rst:
5051 Add trailing newline.
5052 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/makefile-targets.rst:
5053 Add trailing newline.
5054 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/subdirectories-of-gcc.rst:
5055 Add trailing newline.
5056 * doc/gccint/source-tree-structure-and-build-system/top-level-source-directory.rst:
5057 Add trailing newline.
5058 * doc/gccint/standard-header-file-directories.rst:
5059 Add trailing newline.
5060 * doc/gccint/static-analyzer.rst:
5061 Add trailing newline.
5062 * doc/gccint/target-macros.rst:
5063 Add trailing newline.
5064 * doc/gccint/target-macros/controlling-debugging-information-format.rst:
5065 Add trailing newline.
5066 * doc/gccint/target-macros/controlling-the-compilation-driver-gcc.rst:
5067 Add trailing newline.
5068 * doc/gccint/target-macros/cross-compilation-and-floating-point.rst:
5069 Add trailing newline.
5070 * doc/gccint/target-macros/defining-coprocessor-specifics-for-mips-targets.rst:
5071 Add trailing newline.
5072 * doc/gccint/target-macros/defining-data-structures-for-per-function-information.rst:
5073 Add trailing newline.
5074 * doc/gccint/target-macros/defining-the-output-assembler-language.rst:
5075 Add trailing newline.
5076 * doc/gccint/target-macros/defining-the-output-assembler-language/assembler-commands-for-alignment.rst:
5077 Add trailing newline.
5078 * doc/gccint/target-macros/defining-the-output-assembler-language/how-initialization-functions-are-handled.rst:
5079 Add trailing newline.
5080 * doc/gccint/target-macros/defining-the-output-assembler-language/macros-controlling-initialization-routines.rst:
5081 Add trailing newline.
5082 * doc/gccint/target-macros/defining-the-output-assembler-language/output-and-generation-of-labels.rst:
5083 Add trailing newline.
5084 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-assembler-instructions.rst:
5085 Add trailing newline.
5086 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-data.rst:
5087 Add trailing newline.
5088 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-uninitialized-variables.rst:
5089 Add trailing newline.
5090 * doc/gccint/target-macros/implicit-calls-to-library-routines.rst:
5091 Add trailing newline.
5092 * doc/gccint/target-macros/layout-of-source-language-data-types.rst:
5093 Add trailing newline.
5094 * doc/gccint/target-macros/position-independent-code.rst:
5095 Add trailing newline.
5096 * doc/gccint/target-macros/register-usage.rst:
5097 Add trailing newline.
5098 * doc/gccint/target-macros/stack-layout-and-calling-conventions.rst:
5099 Add trailing newline.
5100 * doc/gccint/target-macros/stack-layout-and-calling-conventions/basic-stack-layout.rst:
5101 Add trailing newline.
5102 * doc/gccint/target-macros/stack-layout-and-calling-conventions/caller-saves-register-allocation.rst:
5103 Add trailing newline.
5104 * doc/gccint/target-macros/stack-layout-and-calling-conventions/exception-handling-support.rst:
5105 Add trailing newline.
5106 * doc/gccint/target-macros/stack-layout-and-calling-conventions/passing-function-arguments-on-the-stack.rst:
5107 Add trailing newline.
5108 * doc/gccint/target-macros/stack-layout-and-calling-conventions/registers-that-address-the-stack-frame.rst:
5109 Add trailing newline.
5110 * doc/gccint/target-macros/support-for-nested-functions.rst:
5111 Add trailing newline.
5112 * doc/gccint/target-macros/the-global-targetm-variable.rst:
5113 Add trailing newline.
5114 * doc/gccint/target-makefile-fragments.rst:
5115 Add trailing newline.
5116 * doc/gccint/testsuites.rst:
5117 Add trailing newline.
5118 * doc/gccint/testsuites/ada-language-testsuites.rst:
5119 Add trailing newline.
5120 * doc/gccint/testsuites/c-language-testsuites.rst:
5121 Add trailing newline.
5122 * doc/gccint/testsuites/directives-used-within-dejagnu-tests.rst:
5123 Add trailing newline.
5124 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/commands-for-use-in-dg-final.rst:
5125 Add trailing newline.
5126 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/features-for-dg-add-options.rst:
5127 Add trailing newline.
5128 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/keywords-describing-target-attributes.rst:
5129 Add trailing newline.
5130 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/selecting-targets-to-which-a-test-applies.rst:
5131 Add trailing newline.
5132 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/syntax-and-descriptions-of-test-directives.rst:
5133 Add trailing newline.
5134 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/variants-of-dg-require-support.rst:
5135 Add trailing newline.
5136 * doc/gccint/testsuites/idioms-used-in-testsuite-code.rst:
5137 Add trailing newline.
5138 * doc/gccint/testsuites/support-for-testing-binary-compatibility.rst:
5139 Add trailing newline.
5140 * doc/gccint/testsuites/support-for-testing-gcov.rst:
5141 Add trailing newline.
5142 * doc/gccint/testsuites/support-for-testing-gimple-passes.rst:
5143 Add trailing newline.
5144 * doc/gccint/testsuites/support-for-testing-link-time-optimizations.rst:
5145 Add trailing newline.
5146 * doc/gccint/testsuites/support-for-testing-profile-directed-optimizations.rst:
5147 Add trailing newline.
5148 * doc/gccint/testsuites/support-for-testing-rtl-passes.rst:
5149 Add trailing newline.
5150 * doc/gccint/testsuites/support-for-torture-testing-using-multiple-options.rst:
5151 Add trailing newline.
5152 * doc/gccint/the-gcc-low-level-runtime-library.rst:
5153 Add trailing newline.
5154 * doc/gccint/the-gcc-low-level-runtime-library/language-independent-routines-for-exception-handling.rst:
5155 Add trailing newline.
5156 * doc/gccint/the-gcc-low-level-runtime-library/miscellaneous-runtime-library-routines.rst:
5157 Add trailing newline.
5158 * doc/gccint/the-gcc-low-level-runtime-library/routines-for-decimal-floating-point-emulation.rst:
5159 Add trailing newline.
5160 * doc/gccint/the-gcc-low-level-runtime-library/routines-for-fixed-point-fractional-emulation.rst:
5161 Add trailing newline.
5162 * doc/gccint/the-gcc-low-level-runtime-library/routines-for-floating-point-emulation.rst:
5163 Add trailing newline.
5164 * doc/gccint/the-gcc-low-level-runtime-library/routines-for-integer-arithmetic.rst:
5165 Add trailing newline.
5166 * doc/gccint/the-language.rst:
5167 Add trailing newline.
5168 * doc/gccint/user-experience-guidelines.rst:
5169 Add trailing newline.
5170 * doc/install/binaries.rst:
5171 Add trailing newline.
5172 * doc/install/building.rst:
5173 Add trailing newline.
5174 * doc/install/building/building-a-cross-compiler.rst:
5175 Add trailing newline.
5176 * doc/install/building/building-a-native-compiler.rst:
5177 Add trailing newline.
5178 * doc/install/building/building-in-parallel.rst:
5179 Add trailing newline.
5180 * doc/install/building/building-the-ada-compiler.rst:
5181 Add trailing newline.
5182 * doc/install/building/building-the-d-compiler.rst:
5183 Add trailing newline.
5184 * doc/install/building/building-with-profile-feedback.rst:
5185 Add trailing newline.
5186 * doc/install/configuration.rst:
5187 Add trailing newline.
5188 * doc/install/copyright.rst:
5189 Add trailing newline.
5190 * doc/install/downloading-gcc.rst:
5191 Add trailing newline.
5192 * doc/install/final-installation.rst:
5193 Add trailing newline.
5194 * doc/install/gnu-free-documentation-license.rst:
5195 Add trailing newline.
5196 * doc/install/host-target-specific-installation-notes-for-gcc.rst:
5197 Add trailing newline.
5198 * doc/install/how-can-you-run-the-testsuite-on-selected-tests.rst:
5199 Add trailing newline.
5200 * doc/install/how-to-interpret-test-results.rst:
5201 Add trailing newline.
5202 * doc/install/index.rst:
5203 Add trailing newline.
5204 * doc/install/indices-and-tables.rst:
5205 Add trailing newline.
5206 * doc/install/installing-gcc.rst:
5207 Add trailing newline.
5208 * doc/install/passing-options-and-running-multiple-testsuites.rst:
5209 Add trailing newline.
5210 * doc/install/prerequisites.rst:
5211 Add trailing newline.
5212 * doc/install/submitting-test-results.rst:
5213 Add trailing newline.
5214 * doc/install/testing.rst:
5215 Add trailing newline.
5217 2022-11-14 Martin Liska <mliska@suse.cz>
5220 2022-11-14 Martin Liska <mliska@suse.cz>
5222 * Makefile.in: Use new install URL.
5223 * doc/gcc/gcc-command-options/machine-dependent-options/avr-options.rst:
5224 Use intersphinx link.
5225 * doc/gcc/gcc-command-options/options-to-control-diagnostic-messages-formatting.rst:
5227 * doc/gccint/source-tree-structure-and-build-system.rst: Use
5229 * doc/install/host-target-specific-installation-notes-for-gcc.rst: Likewise.
5230 * doc/install/installing-gcc.rst: Likewise.
5232 2022-11-14 Martin Liska <mliska@suse.cz>
5235 2022-11-09 Martin Liska <mliska@suse.cz>
5236 Sinan <sinan.lin@linux.alibaba.com>
5238 * doc/gccint/analysis-and-optimization-of-gimple-tuples/ssa-operands.rst:
5239 Add missing variable name.
5241 2022-11-14 Martin Liska <mliska@suse.cz>
5244 2022-11-14 Martin Liska <mliska@suse.cz>
5246 * doc/cpp/conf.py: Add newline at last line.
5247 * doc/cppinternals/conf.py: Add newline at last line.
5248 * doc/gcc/conf.py: Add newline at last line.
5249 * doc/gccint/conf.py: Add newline at last line.
5250 * doc/install/conf.py: Add newline at last line.
5252 2022-11-14 Martin Liska <mliska@suse.cz>
5254 * doc/gcc/gcc-command-options/options-that-control-static-analysis.rst:
5257 2022-11-14 Martin Liska <mliska@suse.cz>
5259 * doc/cpp/pragmas.rst: Revert.
5261 2022-11-14 Martin Liska <mliska@suse.cz>
5263 * doc/gcc/gcc-command-options/machine-dependent-options/aarch64-options.rst: Revert.
5265 2022-11-14 Martin Liska <mliska@suse.cz>
5267 * doc/gcc/gcc-command-options/machine-dependent-options/aarch64-options.rst: Revert.
5269 2022-11-14 Martin Liska <mliska@suse.cz>
5271 * doc/gcc/gcc-command-options/options-that-control-static-analysis.rst: Revert.
5272 * doc/gcc/gcc-command-options/options-to-request-or-suppress-warnings.rst: Revert.
5274 2022-11-14 Martin Liska <mliska@suse.cz>
5277 2022-11-13 Jonathan Wakely <jwakely@redhat.com>
5279 * doc/install/testing.rst: Remove anachronism about separate
5282 2022-11-14 Martin Liska <mliska@suse.cz>
5285 2022-11-13 Martin Liska <mliska@suse.cz>
5288 * configure: Regenerate.
5289 * configure.ac: Always set sphinx-build.
5291 2022-11-14 Martin Liska <mliska@suse.cz>
5293 * doc/gcc/gcc-command-options/option-summary.rst: Revert.
5294 * doc/gcc/gcc-command-options/options-that-control-static-analysis.rst: Revert.
5296 2022-11-14 Haochen Jiang <haochen.jiang@intel.com>
5298 * common/config/i386/i386-common.cc
5299 (OPTION_MASK_ISA2_AMX_INT8_SET): Add AMX-TILE dependency.
5300 (OPTION_MASK_ISA2_AMX_BF16_SET): Ditto.
5301 (OPTION_MASK_ISA2_AMX_FP16_SET): Ditto.
5302 (OPTION_MASK_ISA2_AMX_TILE_UNSET): Disable AMX_{INT8,
5303 BF16, FP16} when disable AMX_TILE.
5305 2022-11-14 Martin Liska <mliska@suse.cz>
5307 * doc/gcc/gcc-command-options/machine-dependent-options/x86-options.rst:
5309 * doc/gcc/gcc-command-options/option-summary.rst:
5312 2022-11-14 Hongyu Wang <hongyu.wang@intel.com>
5314 * common/config/i386/i386-common.cc (ix86_optimization_table):
5315 Enable small loop unroll at O2 by default.
5316 * config/i386/i386.cc (ix86_loop_unroll_adjust): Adjust unroll
5317 factor if -munroll-only-small-loops enabled and -funroll-loops/
5318 -funroll-all-loops are disabled.
5319 * config/i386/i386.h (struct processor_costs): Add 2 field
5320 small_unroll_ninsns and small_unroll_factor.
5321 * config/i386/i386.opt: Add -munroll-only-small-loops.
5322 * doc/gcc/gcc-command-options/machine-dependent-options/x86-options.rst:
5323 Document -munroll-only-small-loops.
5324 * doc/gcc/gcc-command-options/option-summary.rst: Likewise.
5325 * loop-init.cc (pass_rtl_unroll_loops::gate): Enable rtl
5326 loop unrolling for -O2-speed and above if target hook
5327 loop_unroll_adjust exists.
5328 (pass_rtl_unroll_loops::execute): Set UAP_UNROLL flag
5329 when target hook loop_unroll_adjust exists.
5330 * config/i386/x86-tune-costs.h: Update all processor costs
5331 with small_unroll_ninsns = 4 and small_unroll_factor = 2.
5333 2022-11-13 David Malcolm <dmalcolm@redhat.com>
5336 * doc/gcc/gcc-command-options/option-summary.rst: Add
5337 -Wno-analyzer-tainted-assertion.
5338 * doc/gcc/gcc-command-options/options-that-control-static-analysis.rst:
5339 Add -Wno-analyzer-tainted-assertion.
5341 2022-11-13 Philipp Tomsich <philipp.tomsich@vrull.eu>
5343 * config/riscv/bitmanip.md: New define_split.
5345 2022-11-13 Philipp Tomsich <philipp.tomsich@vrull.eu>
5347 * config/riscv/riscv.cc (riscv_rtx_costs): Recognize shNadd,
5348 if expressed as a plus and multiplication with a power-of-2.
5349 Split costing for MINUS from PLUS.
5351 2022-11-13 Martin Liska <mliska@suse.cz>
5354 * configure: Regenerate.
5355 * configure.ac: Always set sphinx-build.
5357 2022-11-13 Joseph Myers <joseph@codesourcery.com>
5359 * ginclude/float.h [__STDC_VERSION__ > 201710L]
5360 (__STDC_VERSION_FLOAT_H__): New macro.
5361 * ginclude/stdarg.h [__STDC_VERSION__ > 201710L]
5362 (__STDC_VERSION_STDARG_H__): New macro.
5363 * ginclude/stdatomic.h [__STDC_VERSION__ > 201710L]
5364 (__STDC_VERSION_STDATOMIC_H__): New macro.
5365 * ginclude/stddef.h [__STDC_VERSION__ > 201710L]
5366 (__STDC_VERSION_STDDEF_H__): New macro.
5367 * ginclude/stdint-gcc.h [__STDC_VERSION__ > 201710L]
5368 (__STDC_VERSION_STDINT_H__): New macro.
5369 * glimits.h [__STDC_VERSION__ > 201710L]
5370 (__STDC_VERSION_LIMITS_H__): New macro.
5372 2022-11-13 Jonathan Wakely <jwakely@redhat.com>
5374 * doc/install/testing.rst: Remove anachronism about separate
5377 2022-11-12 Joseph Myers <joseph@codesourcery.com>
5379 * dfp.cc (decimal_from_binary): Convert a canonical NaN to a
5382 2022-11-12 Aldy Hernandez <aldyh@redhat.com>
5384 * value-range.cc (range_tests_signbit): Move to set from here...
5385 (range_tests_signed_zeros): ...to here.
5387 2022-11-12 Aldy Hernandez <aldyh@redhat.com>
5389 * range-op-float.cc (build_lt): Adjust with frange_nextafter
5390 instead of default to a closed range.
5393 2022-11-12 Aldy Hernandez <aldyh@redhat.com>
5394 Jakub Jelinek <jakub@redhat.com>
5396 * range-op-float.cc (float_binary_op_range_finish): New function.
5397 (foperator_plus::op1_range): New.
5398 (foperator_plus::op2_range): New.
5399 (foperator_minus::op1_range): New.
5400 (foperator_minus::op2_range): New.
5401 (foperator_mult::op1_range): New.
5402 (foperator_mult::op2_range): New.
5403 (foperator_div::op1_range): New.
5404 (foperator_div::op2_range): New.
5406 2022-11-12 Jakub Jelinek <jakub@redhat.com>
5408 PR tree-optimization/107569
5409 * range-op-float.cc (zero_p, contains_p, singleton_inf_p,
5410 signbit_known_p, zero_range, inf_range, zero_to_inf_range): New
5412 (foperator_mult_div_base): New class.
5413 (foperator_mult, foperator_div): Derive from that and use
5414 protected static method from it as well as above new functions
5415 to simplify the code.
5417 2022-11-12 Jakub Jelinek <jakub@redhat.com>
5419 PR tree-optimization/107569
5420 * range-op-float.cc (foperator_div): New class.
5421 (floating_op_table::floating_op_table): Use foperator_div
5424 2022-11-12 Jakub Jelinek <jakub@redhat.com>
5426 PR tree-optimization/107569
5427 PR tree-optimization/107591
5428 * range-op.h (range_operator_float::rv_fold): Add relation_kind
5430 * range-op-float.cc (range_operator_float::fold_range): Name
5431 last argument trio and pass trio.op1_op2 () as last argument to
5433 (range_operator_float::rv_fold): Add relation_kind argument.
5434 (foperator_plus::rv_fold, foperator_minus::rv_fold): Likewise.
5435 (foperator_mult): New class.
5436 (floating_op_table::floating_op_table): Use foperator_mult for
5439 2022-11-12 Xi Ruoyao <xry111@xry111.site>
5441 * config/loongarch/loongarch.md (UNSPEC_FLOGB): New unspec.
5443 (logb_non_negative<mode>2): New instruction template.
5444 (logb<mode>2): New define_expand.
5446 2022-11-12 Xi Ruoyao <xry111@xry111.site>
5448 * config/loongarch/loongarch.md (UNSPEC_FSCALEB): New unspec.
5449 (type): Add fscaleb.
5450 (IMODE): New mode attr.
5451 (ldexp<mode>3): New instruction template.
5453 2022-11-12 Xi Ruoyao <xry111@xry111.site>
5455 * config/loongarch/loongarch.md (UNSPEC_FTINT): New unspec.
5456 (UNSPEC_FTINTRM): Likewise.
5457 (UNSPEC_FTINTRP): Likewise.
5458 (LRINT): New define_int_iterator.
5459 (lrint_pattern): New define_int_attr.
5460 (lrint_submenmonic): Likewise.
5461 (lrint_allow_inexact): Likewise.
5462 (ANYFI): New define_mode_iterator.
5463 (lrint<ANYF><ANYFI>): New instruction template.
5465 2022-11-12 Xi Ruoyao <xry111@xry111.site>
5467 * config/loongarch/loongarch.md (frint_<fmt>): Rename to ..
5468 (rint<mode>2): .. this.
5470 2022-11-11 David Malcolm <dmalcolm@redhat.com>
5473 * doc/gcc/gcc-command-options/options-that-control-static-analysis.rst:
5474 Add -Wanalyzer-infinite-recursion.
5475 * doc/gcc/gcc-command-options/options-to-request-or-suppress-warnings.rst
5476 (-Winfinite-recursion): Mention -Wanalyzer-infinite-recursion.
5478 2022-11-11 David Malcolm <dmalcolm@redhat.com>
5481 * Makefile.in (ANALYZER_OBJS): Add analyzer/infinite-recursion.o.
5483 2022-11-11 Andrew MacLeod <amacleod@redhat.com>
5485 PR tree-optimization/107523
5486 * gimple-range.cc (gimple_ranger::update_stmt): Use fur_stmt
5487 rather than fur_depend.
5489 2022-11-11 Andrew MacLeod <amacleod@redhat.com>
5491 * tree-vrp.cc (rvrp_folder::rvrp_folder): Init m_last_bb_stmt.
5492 (rvrp_folder::pre_fold_bb): Set m_last_bb_stmt.
5493 (rvrp_folder::pre_fold_stmt): Check for transitive inferred ranges.
5494 (rvrp_folder::fold_stmt): Check in pre_fold_stmt instead.
5496 2022-11-11 Srinath Parvathaneni <srinath.parvathaneni@arm.com>
5498 * config/aarch64/aarch64-cores.def (AARCH64_CORE): Add Cortex-X1C
5500 * config/aarch64/aarch64-tune.md: Regenerate.
5501 * doc/gcc/gcc-command-options/machine-dependent-options/aarch64-options.rst:
5502 Document Cortex-X1C CPU.
5504 2022-11-11 Srinath Parvathaneni <srinath.parvathaneni@arm.com>
5506 * config/aarch64/aarch64-cores.def (AARCH64_CORE): Add Cortex-A715
5508 * config/aarch64/aarch64-tune.md: Regenerate.
5509 * doc/gcc/gcc-command-options/machine-dependent-options/aarch64-options.rst:
5510 Document Cortex-A715 CPU.
5512 2022-11-11 Richard Biener <rguenther@suse.de>
5513 Nikita Voronov <nik_1357@mail.ru>
5515 PR tree-optimization/107554
5516 * tree-ssa-strlen.cc (strlen_pass::count_nonzero_bytes):
5517 Use unsigned HOST_WIDE_INT type for the strlen.
5519 2022-11-11 Richard Biener <rguenther@suse.de>
5521 PR tree-optimization/105142
5522 * gimple-fold.cc (fosa_unwind): New global.
5523 (follow_outer_ssa_edges): When the SSA definition to follow
5524 is does not dominate fosa_bb, temporarily clear flow-sensitive
5525 info. Make sure to not expand stmts with not defined overflow.
5526 (maybe_fold_comparisons_from_match_pd): Set up unwind stack
5527 for follow_outer_ssa_edges and unwind flow-sensitive info
5528 clearing after matching.
5530 2022-11-11 Aldy Hernandez <aldyh@redhat.com>
5532 * range-op.cc (operator_mult::fold_range): Remove.
5533 (operator_div::fold_range): Remove.
5534 (operator_bitwise_and): Remove.
5536 2022-11-11 Aldy Hernandez <aldyh@redhat.com>
5538 * range-op.cc (update_known_bitmask): Avoid unnecessary intersection.
5540 2022-11-11 Aldy Hernandez <aldyh@redhat.com>
5542 * range-op.cc (range_operator::fold_range): Call
5543 update_known_bitmask.
5544 (operator_bitwise_and::fold_range): Avoid setting nonzero bits
5545 when range is undefined.
5547 2022-11-11 Aldy Hernandez <aldyh@redhat.com>
5549 * range-op.cc (class operator_div): Remove tree code.
5550 (operator_div::wi_op_overflows): Handle EXACT_DIV_EXPR as
5553 2022-11-11 Aldy Hernandez <aldyh@redhat.com>
5555 * range-op.cc: (range_op_table::set): Set m_code.
5556 (integral_table::integral_table): Handle shared entries.
5557 (pointer_table::pointer_table): Same.
5558 * range-op.h (class range_operator): Add m_code.
5560 2022-11-11 Richard Biener <rguenther@suse.de>
5562 PR tree-optimization/107618
5563 * tree-ssa-copy.cc (stmt_may_generate_copy): Simulate all
5564 assignments with a single SSA use.
5565 (copy_prop_visit_assignment): Use gimple_fold_stmt_to_constant_1
5566 to perform simple constant folding.
5567 (copy_prop::visit_stmt): Visit all assignments.
5569 2022-11-11 Richard Biener <rguenther@suse.de>
5571 PR tree-optimization/84646
5572 * tree-ssa-dce.cc (pass_dce::set_pass_param): Add param
5573 wheter to run update-address-taken.
5574 (pass_dce::execute): Honor it.
5575 * passes.def: Exchange last DCE and CD-DCE invocations.
5576 Swap pass_tail_calls and the last DCE.
5578 2022-11-11 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
5579 Monk Chiang <monk.chiang@sifive.com>
5581 * config/riscv/riscv-v.cc (emit_pred_move): Adjust for scalable register spilling.
5582 (legitimize_move): Ditto.
5583 * config/riscv/riscv.cc (riscv_v_adjust_scalable_frame): New function.
5584 (riscv_first_stack_step): Adjust for scalable register spilling.
5585 (riscv_expand_prologue): Ditto.
5586 (riscv_expand_epilogue): Ditto.
5587 (riscv_dwarf_poly_indeterminate_value): New function.
5588 (TARGET_DWARF_POLY_INDETERMINATE_VALUE): New target hook support for register spilling.
5589 * config/riscv/riscv.h (RISCV_DWARF_VLENB): New macro.
5590 (RISCV_PROLOGUE_TEMP2_REGNUM): Ditto.
5591 (RISCV_PROLOGUE_TEMP2): Ditto.
5592 * config/riscv/vector-iterators.md: New iterators.
5593 * config/riscv/vector.md (*mov<mode>): Fix it for register spilling.
5594 (*mov<mode>_whole): New pattern.
5595 (*mov<mode>_fract): New pattern.
5596 (@pred_mov<mode>): Fix it for register spilling.
5598 2022-11-11 Jonathan Wakely <jwakely@redhat.com>
5601 * doc/cpp/pragmas.rst (Pragmas): Document region pragmas.
5603 2022-11-11 Haochen Jiang <haochen.jiang@intel.com>
5605 * config/i386/i386-builtin.def (BDESC): Add
5606 OPTION_MASK_ISA2_PREFETCHI for prefetchi builtin.
5607 * config/i386/i386-expand.cc (ix86_expand_builtin):
5608 Add ISA check before emit_insn.
5609 * config/i386/prfchiintrin.h: Add target for intrin.
5611 2022-11-10 David Malcolm <dmalcolm@redhat.com>
5614 * tristate.h (tristate::is_unknown): New.
5616 2022-11-10 David Malcolm <dmalcolm@redhat.com>
5618 * doc/gcc/gcc-command-options/options-that-control-static-analysis.rst:
5619 Add -Wanalyzer-deref-before-check.
5621 2022-11-10 Marek Polacek <polacek@redhat.com>
5623 * doc/gcc/gcc-command-options/options-controlling-c++-dialect.rst:
5624 Extend the description of -Wdangling-reference.
5626 2022-11-10 Andrew Pinski <apinski@marvell.com>
5628 * config/bfin/bfin.h (SLOW_SHORT_ACCESS): Delete.
5629 * config/i386/i386.h (SLOW_SHORT_ACCESS): Delete.
5630 * system.h: Poison SLOW_SHORT_ACCESS
5632 2022-11-10 Aldy Hernandez <aldyh@redhat.com>
5634 * value-range.cc (frange::set_nonnegative): Remove NAN sign handling.
5635 (range_tests_signed_zeros): Adjust test.
5637 2022-11-10 Richard Biener <rguenther@suse.de>
5639 PR tree-optimization/84646
5640 * tree-ssa-forwprop.cc (pass_forwprop::execute): Improve
5641 copy propagation across PHIs.
5643 2022-11-10 Philipp Tomsich <philipp.tomsich@vrull.eu>
5645 * config/riscv/riscv-cores.def (RISCV_TUNE): Update
5646 sifive-7-series to point to the sifive_7 pipeline description.
5648 2022-11-10 Richard Biener <rguenther@suse.de>
5650 PR tree-optimization/84646
5651 * tree-ssa-ccp.cc (set_lattice_value): Make sure we
5652 allow a const -> copy transition and avoid using meet
5655 2022-11-10 Martin Liska <mliska@suse.cz>
5657 * doc/cpp/conf.py: Add newline at last line.
5658 * doc/cppinternals/conf.py: Add newline at last line.
5659 * doc/gcc/conf.py: Add newline at last line.
5660 * doc/gccint/conf.py: Add newline at last line.
5661 * doc/install/conf.py: Add newline at last line.
5663 2022-11-10 Richard Biener <rguenther@suse.de>
5665 * tree-ssa-loop-unswitch.cc (init_loop_unswitch_info): First collect
5666 candidates and determine the outermost loop to unswitch.
5667 (tree_ssa_unswitch_loops): First perform all guard hoisting,
5668 then perform unswitching on innermost loop predicates.
5669 (find_unswitching_predicates_for_bb): Keep track of the
5670 outermost loop to unswitch.
5671 (evaluate_bbs): Adjust exit test.
5672 (tree_unswitch_single_loop): Dump whether we unswitched an outer
5674 (tree_unswitch_loop): Remove assert we unswitch only innermost
5677 2022-11-10 Jakub Jelinek <jakub@redhat.com>
5681 * config/i386/i386-expand.cc (ix86_expand_int_sse_cmp): Use
5682 gen_int_mode rather than GEN_INT.
5684 2022-11-09 Martin Liska <mliska@suse.cz>
5685 Sinan <sinan.lin@linux.alibaba.com>
5687 * doc/gccint/analysis-and-optimization-of-gimple-tuples/ssa-operands.rst:
5688 Add missing variable name.
5690 2022-11-09 Aldy Hernandez <aldyh@redhat.com>
5692 * value-range-storage.cc (frange_storage_slot::get_frange): Clear
5694 * value-range.cc (range_tests_floats): New test.
5696 2022-11-09 Aldy Hernandez <aldyh@redhat.com>
5698 * range-op-float.cc (class foperator_plus): Remove op[12]_range.
5699 (class foperator_minus): Same.
5701 2022-11-09 Qing Zhao <qing.zhao@oracle.com>
5703 * gimple-array-bounds.cc (trailing_array): Replace
5704 array_at_struct_end_p with new name and update comments.
5705 * gimple-fold.cc (get_range_strlen_tree): Likewise.
5706 * gimple-ssa-warn-restrict.cc (builtin_memref::builtin_memref):
5708 * graphite-sese-to-poly.cc (bounds_are_valid): Likewise.
5709 * tree-if-conv.cc (idx_within_array_bound): Likewise.
5710 * tree-object-size.cc (addr_object_size): Likewise.
5711 * tree-ssa-alias.cc (component_ref_to_zero_sized_trailing_array_p):
5713 (stmt_kills_ref_p): Likewise.
5714 * tree-ssa-loop-niter.cc (idx_infer_loop_bounds): Likewise.
5715 * tree-ssa-strlen.cc (maybe_set_strlen_range): Likewise.
5716 * tree.cc (array_at_struct_end_p): Rename to ...
5717 (array_ref_flexible_size_p): ... this.
5718 (component_ref_size): Replace array_at_struct_end_p with new name.
5719 * tree.h (array_at_struct_end_p): Rename to ...
5720 (array_ref_flexible_size_p): ... this.
5722 2022-11-09 Martin Liska <mliska@suse.cz>
5724 * Makefile.in: Use new install URL.
5725 * doc/gcc/gcc-command-options/machine-dependent-options/avr-options.rst:
5726 Use intersphinx link.
5727 * doc/gcc/gcc-command-options/options-to-control-diagnostic-messages-formatting.rst:
5729 * doc/gccint/source-tree-structure-and-build-system.rst: Use
5731 * doc/install/host-target-specific-installation-notes-for-gcc.rst: Likewise.
5732 * doc/install/installing-gcc.rst: Likewise.
5734 2022-11-09 Richard Biener <rguenther@suse.de>
5736 PR tree-optimization/84646
5737 * tree-ssa-threadbackward.cc (back_threader::maybe_register_path):
5738 Remove premature cycle rejection.
5740 2022-11-09 Martin Liska <mliska@suse.cz>
5742 * doc/cpp/character-sets.rst:
5743 Add trailing newline.
5744 * doc/cpp/conditional-syntax.rst:
5745 Add trailing newline.
5746 * doc/cpp/conditional-uses.rst:
5747 Add trailing newline.
5748 * doc/cpp/conditionals.rst:
5749 Add trailing newline.
5750 * doc/cpp/copyright.rst:
5751 Add trailing newline.
5752 * doc/cpp/deleted-code.rst:
5753 Add trailing newline.
5754 * doc/cpp/diagnostics.rst:
5755 Add trailing newline.
5756 * doc/cpp/environment-variables.rst:
5757 Add trailing newline.
5758 * doc/cpp/gnu-free-documentation-license.rst:
5759 Add trailing newline.
5760 * doc/cpp/header-files.rst:
5761 Add trailing newline.
5762 * doc/cpp/header-files/alternatives-to-wrapper-ifndef.rst:
5763 Add trailing newline.
5764 * doc/cpp/header-files/computed-includes.rst:
5765 Add trailing newline.
5766 * doc/cpp/header-files/include-operation.rst:
5767 Add trailing newline.
5768 * doc/cpp/header-files/include-syntax.rst:
5769 Add trailing newline.
5770 * doc/cpp/header-files/once-only-headers.rst:
5771 Add trailing newline.
5772 * doc/cpp/header-files/search-path.rst:
5773 Add trailing newline.
5774 * doc/cpp/header-files/system-headers.rst:
5775 Add trailing newline.
5776 * doc/cpp/header-files/wrapper-headers.rst:
5777 Add trailing newline.
5778 * doc/cpp/implementation-defined-behavior.rst:
5779 Add trailing newline.
5780 * doc/cpp/implementation-details.rst:
5781 Add trailing newline.
5782 * doc/cpp/implementation-limits.rst:
5783 Add trailing newline.
5784 * doc/cpp/index.rst:
5785 Add trailing newline.
5786 * doc/cpp/indices-and-tables.rst:
5787 Add trailing newline.
5788 * doc/cpp/initial-processing.rst:
5789 Add trailing newline.
5790 * doc/cpp/invocation.rst:
5791 Add trailing newline.
5792 * doc/cpp/line-control.rst:
5793 Add trailing newline.
5794 * doc/cpp/macros.rst:
5795 Add trailing newline.
5796 * doc/cpp/macros/concatenation.rst:
5797 Add trailing newline.
5798 * doc/cpp/macros/directives-within-macro-arguments.rst:
5799 Add trailing newline.
5800 * doc/cpp/macros/function-like-macros.rst:
5801 Add trailing newline.
5802 * doc/cpp/macros/macro-arguments.rst:
5803 Add trailing newline.
5804 * doc/cpp/macros/macro-pitfalls.rst:
5805 Add trailing newline.
5806 * doc/cpp/macros/object-like-macros.rst:
5807 Add trailing newline.
5808 * doc/cpp/macros/predefined-macros.rst:
5809 Add trailing newline.
5810 * doc/cpp/macros/stringizing.rst:
5811 Add trailing newline.
5812 * doc/cpp/macros/undefining-and-redefining-macros.rst:
5813 Add trailing newline.
5814 * doc/cpp/macros/variadic-macros.rst:
5815 Add trailing newline.
5816 * doc/cpp/obsolete-features.rst:
5817 Add trailing newline.
5818 * doc/cpp/other-directives.rst:
5819 Add trailing newline.
5820 * doc/cpp/overview.rst:
5821 Add trailing newline.
5822 * doc/cpp/pragmas.rst:
5823 Add trailing newline.
5824 * doc/cpp/preprocessor-output.rst:
5825 Add trailing newline.
5826 * doc/cpp/the-preprocessing-language.rst:
5827 Add trailing newline.
5828 * doc/cpp/tokenization.rst:
5829 Add trailing newline.
5830 * doc/cpp/traditional-lexical-analysis.rst:
5831 Add trailing newline.
5832 * doc/cpp/traditional-macros.rst:
5833 Add trailing newline.
5834 * doc/cpp/traditional-miscellany.rst:
5835 Add trailing newline.
5836 * doc/cpp/traditional-mode.rst:
5837 Add trailing newline.
5838 * doc/cpp/traditional-warnings.rst:
5839 Add trailing newline.
5840 * doc/cppinternals/copyright.rst:
5841 Add trailing newline.
5842 * doc/cppinternals/cppinternals.rst:
5843 Add trailing newline.
5844 * doc/cppinternals/cpplib.rst:
5845 Add trailing newline.
5846 * doc/cppinternals/files.rst:
5847 Add trailing newline.
5848 * doc/cppinternals/index.rst:
5849 Add trailing newline.
5850 * doc/cppinternals/indices-and-tables.rst:
5851 Add trailing newline.
5852 * doc/cppinternals/internal-representation-of-macros.rst:
5853 Add trailing newline.
5854 * doc/cppinternals/just-which-line-number-anyway.rst:
5855 Add trailing newline.
5856 * doc/cppinternals/lexing-a-line.rst:
5857 Add trailing newline.
5858 * doc/cppinternals/lexing-a-token.rst:
5859 Add trailing newline.
5860 * doc/cppinternals/looking-for-a-function-like-macros-opening-parenthesis.rst:
5861 Add trailing newline.
5862 * doc/cppinternals/macro-expansion-overview.rst:
5863 Add trailing newline.
5864 * doc/cppinternals/marking-tokens-ineligible-for-future-expansion.rst:
5865 Add trailing newline.
5866 * doc/cppinternals/multiple-include-optimization.rst:
5867 Add trailing newline.
5868 * doc/cppinternals/overview.rst:
5869 Add trailing newline.
5870 * doc/cppinternals/representation-of-line-numbers.rst:
5871 Add trailing newline.
5872 * doc/cppinternals/scanning-the-replacement-list-for-macros-to-expand.rst:
5873 Add trailing newline.
5874 * doc/gcc/binary-compatibility.rst:
5875 Add trailing newline.
5876 * doc/gcc/c++-implementation-defined-behavior.rst:
5877 Add trailing newline.
5878 * doc/gcc/c-implementation-defined-behavior.rst:
5879 Add trailing newline.
5880 * doc/gcc/c-implementation-defined-behavior/architecture.rst:
5881 Add trailing newline.
5882 * doc/gcc/c-implementation-defined-behavior/arrays-and-pointers.rst:
5883 Add trailing newline.
5884 * doc/gcc/c-implementation-defined-behavior/characters.rst:
5885 Add trailing newline.
5886 * doc/gcc/c-implementation-defined-behavior/declarators.rst:
5887 Add trailing newline.
5888 * doc/gcc/c-implementation-defined-behavior/environment.rst:
5889 Add trailing newline.
5890 * doc/gcc/c-implementation-defined-behavior/floating-point.rst:
5891 Add trailing newline.
5892 * doc/gcc/c-implementation-defined-behavior/hints.rst:
5893 Add trailing newline.
5894 * doc/gcc/c-implementation-defined-behavior/identifiers.rst:
5895 Add trailing newline.
5896 * doc/gcc/c-implementation-defined-behavior/integers.rst:
5897 Add trailing newline.
5898 * doc/gcc/c-implementation-defined-behavior/library-functions.rst:
5899 Add trailing newline.
5900 * doc/gcc/c-implementation-defined-behavior/locale-specific-behavior.rst:
5901 Add trailing newline.
5902 * doc/gcc/c-implementation-defined-behavior/preprocessing-directives.rst:
5903 Add trailing newline.
5904 * doc/gcc/c-implementation-defined-behavior/qualifiers.rst:
5905 Add trailing newline.
5906 * doc/gcc/c-implementation-defined-behavior/statements.rst:
5907 Add trailing newline.
5908 * doc/gcc/c-implementation-defined-behavior/structures-unions-enumerations-and-bit-fields.rst:
5909 Add trailing newline.
5910 * doc/gcc/c-implementation-defined-behavior/translation.rst:
5911 Add trailing newline.
5912 * doc/gcc/conditionally-supported-behavior.rst:
5913 Add trailing newline.
5914 * doc/gcc/contributing-to-gcc-development.rst:
5915 Add trailing newline.
5916 * doc/gcc/contributors-to-gcc.rst:
5917 Add trailing newline.
5918 * doc/gcc/copyright.rst:
5919 Add trailing newline.
5920 * doc/gcc/exception-handling.rst:
5921 Add trailing newline.
5922 * doc/gcc/extensions-to-the-c++-language.rst:
5923 Add trailing newline.
5924 * doc/gcc/extensions-to-the-c++-language/backwards-compatibility.rst:
5925 Add trailing newline.
5926 * doc/gcc/extensions-to-the-c++-language/c++-concepts.rst:
5927 Add trailing newline.
5928 * doc/gcc/extensions-to-the-c++-language/c++-interface-and-implementation-pragmas.rst:
5929 Add trailing newline.
5930 * doc/gcc/extensions-to-the-c++-language/c++-specific-variable-function-and-type-attributes.rst:
5931 Add trailing newline.
5932 * doc/gcc/extensions-to-the-c++-language/deprecated-features.rst:
5933 Add trailing newline.
5934 * doc/gcc/extensions-to-the-c++-language/extracting-the-function-pointer-from-a-bound-pointer-to-member-function.rst:
5935 Add trailing newline.
5936 * doc/gcc/extensions-to-the-c++-language/function-multiversioning.rst:
5937 Add trailing newline.
5938 * doc/gcc/extensions-to-the-c++-language/restricting-pointer-aliasing.rst:
5939 Add trailing newline.
5940 * doc/gcc/extensions-to-the-c++-language/type-traits.rst:
5941 Add trailing newline.
5942 * doc/gcc/extensions-to-the-c++-language/vague-linkage.rst:
5943 Add trailing newline.
5944 * doc/gcc/extensions-to-the-c++-language/when-is-a-volatile-c++-object-accessed.rst:
5945 Add trailing newline.
5946 * doc/gcc/extensions-to-the-c++-language/wheres-the-template.rst:
5947 Add trailing newline.
5948 * doc/gcc/extensions-to-the-c-language-family.rst:
5949 Add trailing newline.
5950 * doc/gcc/extensions-to-the-c-language-family/128-bit-integers.rst:
5951 Add trailing newline.
5952 * doc/gcc/extensions-to-the-c-language-family/additional-floating-types.rst:
5953 Add trailing newline.
5954 * doc/gcc/extensions-to-the-c-language-family/alternate-keywords.rst:
5955 Add trailing newline.
5956 * doc/gcc/extensions-to-the-c-language-family/an-inline-function-is-as-fast-as-a-macro.rst:
5957 Add trailing newline.
5958 * doc/gcc/extensions-to-the-c-language-family/arithmetic-on-void-and-function-pointers.rst:
5959 Add trailing newline.
5960 * doc/gcc/extensions-to-the-c-language-family/arrays-of-length-zero.rst:
5961 Add trailing newline.
5962 * doc/gcc/extensions-to-the-c-language-family/arrays-of-variable-length.rst:
5963 Add trailing newline.
5964 * doc/gcc/extensions-to-the-c-language-family/attribute-syntax.rst:
5965 Add trailing newline.
5966 * doc/gcc/extensions-to-the-c-language-family/binary-constants-using-the-0b-prefix.rst:
5967 Add trailing newline.
5968 * doc/gcc/extensions-to-the-c-language-family/built-in-functions-for-memory-model-aware-atomic-operations.rst:
5969 Add trailing newline.
5970 * doc/gcc/extensions-to-the-c-language-family/built-in-functions-to-perform-arithmetic-with-overflow-checking.rst:
5971 Add trailing newline.
5972 * doc/gcc/extensions-to-the-c-language-family/c++-style-comments.rst:
5973 Add trailing newline.
5974 * doc/gcc/extensions-to-the-c-language-family/case-ranges.rst:
5975 Add trailing newline.
5976 * doc/gcc/extensions-to-the-c-language-family/cast-to-a-union-type.rst:
5977 Add trailing newline.
5978 * doc/gcc/extensions-to-the-c-language-family/complex-numbers.rst:
5979 Add trailing newline.
5980 * doc/gcc/extensions-to-the-c-language-family/compound-literals.rst:
5981 Add trailing newline.
5982 * doc/gcc/extensions-to-the-c-language-family/conditionals-with-omitted-operands.rst:
5983 Add trailing newline.
5984 * doc/gcc/extensions-to-the-c-language-family/constructing-function-calls.rst:
5985 Add trailing newline.
5986 * doc/gcc/extensions-to-the-c-language-family/decimal-floating-types.rst:
5987 Add trailing newline.
5988 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions.rst:
5989 Add trailing newline.
5990 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/aarch64-function-attributes.rst:
5991 Add trailing newline.
5992 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/amd-gcn-function-attributes.rst:
5993 Add trailing newline.
5994 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/arc-function-attributes.rst:
5995 Add trailing newline.
5996 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/arm-function-attributes.rst:
5997 Add trailing newline.
5998 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/avr-function-attributes.rst:
5999 Add trailing newline.
6000 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/blackfin-function-attributes.rst:
6001 Add trailing newline.
6002 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/bpf-function-attributes.rst:
6003 Add trailing newline.
6004 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/c-sky-function-attributes.rst:
6005 Add trailing newline.
6006 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/common-function-attributes.rst:
6007 Add trailing newline.
6008 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/epiphany-function-attributes.rst:
6009 Add trailing newline.
6010 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/h8-300-function-attributes.rst:
6011 Add trailing newline.
6012 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/ia-64-function-attributes.rst:
6013 Add trailing newline.
6014 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/m32c-function-attributes.rst:
6015 Add trailing newline.
6016 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/m32r-d-function-attributes.rst:
6017 Add trailing newline.
6018 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/m68k-function-attributes.rst:
6019 Add trailing newline.
6020 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/mcore-function-attributes.rst:
6021 Add trailing newline.
6022 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/mep-function-attributes.rst:
6023 Add trailing newline.
6024 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/microblaze-function-attributes.rst:
6025 Add trailing newline.
6026 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/microsoft-windows-function-attributes.rst:
6027 Add trailing newline.
6028 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/mips-function-attributes.rst:
6029 Add trailing newline.
6030 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/msp430-function-attributes.rst:
6031 Add trailing newline.
6032 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/nds32-function-attributes.rst:
6033 Add trailing newline.
6034 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/nios-ii-function-attributes.rst:
6035 Add trailing newline.
6036 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/nvidia-ptx-function-attributes.rst:
6037 Add trailing newline.
6038 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/powerpc-function-attributes.rst:
6039 Add trailing newline.
6040 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/risc-v-function-attributes.rst:
6041 Add trailing newline.
6042 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/rl78-function-attributes.rst:
6043 Add trailing newline.
6044 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/rx-function-attributes.rst:
6045 Add trailing newline.
6046 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/s-390-function-attributes.rst:
6047 Add trailing newline.
6048 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/sh-function-attributes.rst:
6049 Add trailing newline.
6050 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/symbian-os-function-attributes.rst:
6051 Add trailing newline.
6052 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/v850-function-attributes.rst:
6053 Add trailing newline.
6054 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/visium-function-attributes.rst:
6055 Add trailing newline.
6056 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/x86-function-attributes.rst:
6057 Add trailing newline.
6058 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/xstormy16-function-attributes.rst:
6059 Add trailing newline.
6060 * doc/gcc/extensions-to-the-c-language-family/designated-initializers.rst:
6061 Add trailing newline.
6062 * doc/gcc/extensions-to-the-c-language-family/determining-the-alignment-of-functions-types-or-variables.rst:
6063 Add trailing newline.
6064 * doc/gcc/extensions-to-the-c-language-family/dollar-signs-in-identifier-names.rst:
6065 Add trailing newline.
6066 * doc/gcc/extensions-to-the-c-language-family/double-word-integers.rst:
6067 Add trailing newline.
6068 * doc/gcc/extensions-to-the-c-language-family/enumerator-attributes.rst:
6069 Add trailing newline.
6070 * doc/gcc/extensions-to-the-c-language-family/fixed-point-types.rst:
6071 Add trailing newline.
6072 * doc/gcc/extensions-to-the-c-language-family/format-checks-specific-to-particular-target-machines.rst:
6073 Add trailing newline.
6074 * doc/gcc/extensions-to-the-c-language-family/function-names-as-strings.rst:
6075 Add trailing newline.
6076 * doc/gcc/extensions-to-the-c-language-family/getting-the-return-or-frame-address-of-a-function.rst:
6077 Add trailing newline.
6078 * doc/gcc/extensions-to-the-c-language-family/half-precision-floating-point.rst:
6079 Add trailing newline.
6080 * doc/gcc/extensions-to-the-c-language-family/hex-floats.rst:
6081 Add trailing newline.
6082 * doc/gcc/extensions-to-the-c-language-family/how-to-use-inline-assembly-language-in-c-code.rst:
6083 Add trailing newline.
6084 * doc/gcc/extensions-to-the-c-language-family/incomplete-enum-types.rst:
6085 Add trailing newline.
6086 * doc/gcc/extensions-to-the-c-language-family/label-attributes.rst:
6087 Add trailing newline.
6088 * doc/gcc/extensions-to-the-c-language-family/labels-as-values.rst:
6089 Add trailing newline.
6090 * doc/gcc/extensions-to-the-c-language-family/legacy-sync-built-in-functions-for-atomic-memory-access.rst:
6091 Add trailing newline.
6092 * doc/gcc/extensions-to-the-c-language-family/locally-declared-labels.rst:
6093 Add trailing newline.
6094 * doc/gcc/extensions-to-the-c-language-family/macros-with-a-variable-number-of-arguments.rst:
6095 Add trailing newline.
6096 * doc/gcc/extensions-to-the-c-language-family/mixed-declarations-labels-and-code.rst:
6097 Add trailing newline.
6098 * doc/gcc/extensions-to-the-c-language-family/named-address-spaces.rst:
6099 Add trailing newline.
6100 * doc/gcc/extensions-to-the-c-language-family/nested-functions.rst:
6101 Add trailing newline.
6102 * doc/gcc/extensions-to-the-c-language-family/non-constant-initializers.rst:
6103 Add trailing newline.
6104 * doc/gcc/extensions-to-the-c-language-family/non-lvalue-arrays-may-have-subscripts.rst:
6105 Add trailing newline.
6106 * doc/gcc/extensions-to-the-c-language-family/nonlocal-gotos.rst:
6107 Add trailing newline.
6108 * doc/gcc/extensions-to-the-c-language-family/object-size-checking-built-in-functions.rst:
6109 Add trailing newline.
6110 * doc/gcc/extensions-to-the-c-language-family/other-built-in-functions-provided-by-gcc.rst:
6111 Add trailing newline.
6112 * doc/gcc/extensions-to-the-c-language-family/pointer-arguments-in-variadic-functions.rst:
6113 Add trailing newline.
6114 * doc/gcc/extensions-to-the-c-language-family/pointers-to-arrays-with-qualifiers-work-as-expected.rst:
6115 Add trailing newline.
6116 * doc/gcc/extensions-to-the-c-language-family/pragmas-accepted-by-gcc.rst:
6117 Add trailing newline.
6118 * doc/gcc/extensions-to-the-c-language-family/prototypes-and-old-style-function-definitions.rst:
6119 Add trailing newline.
6120 * doc/gcc/extensions-to-the-c-language-family/referring-to-a-type-with-typeof.rst:
6121 Add trailing newline.
6122 * doc/gcc/extensions-to-the-c-language-family/slightly-looser-rules-for-escaped-newlines.rst:
6123 Add trailing newline.
6124 * doc/gcc/extensions-to-the-c-language-family/specifying-attributes-of-types.rst:
6125 Add trailing newline.
6126 * doc/gcc/extensions-to-the-c-language-family/specifying-attributes-of-variables.rst:
6127 Add trailing newline.
6128 * doc/gcc/extensions-to-the-c-language-family/statement-attributes.rst:
6129 Add trailing newline.
6130 * doc/gcc/extensions-to-the-c-language-family/statements-and-declarations-in-expressions.rst:
6131 Add trailing newline.
6132 * doc/gcc/extensions-to-the-c-language-family/structures-with-no-members.rst:
6133 Add trailing newline.
6134 * doc/gcc/extensions-to-the-c-language-family/support-for-offsetof.rst:
6135 Add trailing newline.
6136 * doc/gcc/extensions-to-the-c-language-family/target-builtins.rst:
6137 Add trailing newline.
6138 * doc/gcc/extensions-to-the-c-language-family/target-builtins/aarch64-built-in-functions.rst:
6139 Add trailing newline.
6140 * doc/gcc/extensions-to-the-c-language-family/target-builtins/alpha-built-in-functions.rst:
6141 Add trailing newline.
6142 * doc/gcc/extensions-to-the-c-language-family/target-builtins/altera-nios-ii-built-in-functions.rst:
6143 Add trailing newline.
6144 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arc-built-in-functions.rst:
6145 Add trailing newline.
6146 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arc-simd-built-in-functions.rst:
6147 Add trailing newline.
6148 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arm-armv8-m-security-extensions.rst:
6149 Add trailing newline.
6150 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arm-c-language-extensions-acle.rst:
6151 Add trailing newline.
6152 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arm-floating-point-status-and-control-intrinsics.rst:
6153 Add trailing newline.
6154 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arm-iwmmxt-built-in-functions.rst:
6155 Add trailing newline.
6156 * doc/gcc/extensions-to-the-c-language-family/target-builtins/avr-built-in-functions.rst:
6157 Add trailing newline.
6158 * doc/gcc/extensions-to-the-c-language-family/target-builtins/basic-powerpc-built-in-functions.rst:
6159 Add trailing newline.
6160 * doc/gcc/extensions-to-the-c-language-family/target-builtins/blackfin-built-in-functions.rst:
6161 Add trailing newline.
6162 * doc/gcc/extensions-to-the-c-language-family/target-builtins/bpf-built-in-functions.rst:
6163 Add trailing newline.
6164 * doc/gcc/extensions-to-the-c-language-family/target-builtins/fr-v-built-in-functions.rst:
6165 Add trailing newline.
6166 * doc/gcc/extensions-to-the-c-language-family/target-builtins/mips-dsp-built-in-functions.rst:
6167 Add trailing newline.
6168 * doc/gcc/extensions-to-the-c-language-family/target-builtins/mips-loongson-built-in-functions.rst:
6169 Add trailing newline.
6170 * doc/gcc/extensions-to-the-c-language-family/target-builtins/mips-paired-single-support.rst:
6171 Add trailing newline.
6172 * doc/gcc/extensions-to-the-c-language-family/target-builtins/mips-simd-architecture-msa-support.rst:
6173 Add trailing newline.
6174 * doc/gcc/extensions-to-the-c-language-family/target-builtins/msp430-built-in-functions.rst:
6175 Add trailing newline.
6176 * doc/gcc/extensions-to-the-c-language-family/target-builtins/nds32-built-in-functions.rst:
6177 Add trailing newline.
6178 * doc/gcc/extensions-to-the-c-language-family/target-builtins/other-mips-built-in-functions.rst:
6179 Add trailing newline.
6180 * doc/gcc/extensions-to-the-c-language-family/target-builtins/picochip-built-in-functions.rst:
6181 Add trailing newline.
6182 * doc/gcc/extensions-to-the-c-language-family/target-builtins/powerpc-altivec-vsx-built-in-functions.rst:
6183 Add trailing newline.
6184 * doc/gcc/extensions-to-the-c-language-family/target-builtins/powerpc-atomic-memory-operation-functions.rst:
6185 Add trailing newline.
6186 * doc/gcc/extensions-to-the-c-language-family/target-builtins/powerpc-hardware-transactional-memory-built-in-functions.rst:
6187 Add trailing newline.
6188 * doc/gcc/extensions-to-the-c-language-family/target-builtins/powerpc-matrix-multiply-assist-built-in-functions.rst:
6189 Add trailing newline.
6190 * doc/gcc/extensions-to-the-c-language-family/target-builtins/pru-built-in-functions.rst:
6191 Add trailing newline.
6192 * doc/gcc/extensions-to-the-c-language-family/target-builtins/risc-v-built-in-functions.rst:
6193 Add trailing newline.
6194 * doc/gcc/extensions-to-the-c-language-family/target-builtins/rx-built-in-functions.rst:
6195 Add trailing newline.
6196 * doc/gcc/extensions-to-the-c-language-family/target-builtins/s-390-system-z-built-in-functions.rst:
6197 Add trailing newline.
6198 * doc/gcc/extensions-to-the-c-language-family/target-builtins/sh-built-in-functions.rst:
6199 Add trailing newline.
6200 * doc/gcc/extensions-to-the-c-language-family/target-builtins/sparc-vis-built-in-functions.rst:
6201 Add trailing newline.
6202 * doc/gcc/extensions-to-the-c-language-family/target-builtins/ti-c6x-built-in-functions.rst:
6203 Add trailing newline.
6204 * doc/gcc/extensions-to-the-c-language-family/target-builtins/x86-built-in-functions.rst:
6205 Add trailing newline.
6206 * doc/gcc/extensions-to-the-c-language-family/target-builtins/x86-control-flow-protection-intrinsics.rst:
6207 Add trailing newline.
6208 * doc/gcc/extensions-to-the-c-language-family/target-builtins/x86-transactional-memory-intrinsics.rst:
6209 Add trailing newline.
6210 * doc/gcc/extensions-to-the-c-language-family/the-character-esc-in-constants.rst:
6211 Add trailing newline.
6212 * doc/gcc/extensions-to-the-c-language-family/thread-local-storage.rst:
6213 Add trailing newline.
6214 * doc/gcc/extensions-to-the-c-language-family/unnamed-structure-and-union-fields.rst:
6215 Add trailing newline.
6216 * doc/gcc/extensions-to-the-c-language-family/using-vector-instructions-through-built-in-functions.rst:
6217 Add trailing newline.
6218 * doc/gcc/extensions-to-the-c-language-family/when-is-a-volatile-object-accessed.rst:
6219 Add trailing newline.
6220 * doc/gcc/extensions-to-the-c-language-family/x86-specific-memory-model-extensions-for-transactional-memory.rst:
6221 Add trailing newline.
6222 * doc/gcc/funding.rst:
6223 Add trailing newline.
6224 * doc/gcc/gcc-command-options.rst:
6225 Add trailing newline.
6226 * doc/gcc/gcc-command-options/compiling-c++-programs.rst:
6227 Add trailing newline.
6228 * doc/gcc/gcc-command-options/description.rst:
6229 Add trailing newline.
6230 * doc/gcc/gcc-command-options/environment-variables-affecting-gcc.rst:
6231 Add trailing newline.
6232 * doc/gcc/gcc-command-options/gcc-developer-options.rst:
6233 Add trailing newline.
6234 * doc/gcc/gcc-command-options/machine-dependent-options.rst:
6235 Add trailing newline.
6236 * doc/gcc/gcc-command-options/machine-dependent-options/aarch64-options.rst:
6237 Add trailing newline.
6238 * doc/gcc/gcc-command-options/machine-dependent-options/adapteva-epiphany-options.rst:
6239 Add trailing newline.
6240 * doc/gcc/gcc-command-options/machine-dependent-options/amd-gcn-options.rst:
6241 Add trailing newline.
6242 * doc/gcc/gcc-command-options/machine-dependent-options/arc-options.rst:
6243 Add trailing newline.
6244 * doc/gcc/gcc-command-options/machine-dependent-options/arm-options.rst:
6245 Add trailing newline.
6246 * doc/gcc/gcc-command-options/machine-dependent-options/avr-options.rst:
6247 Add trailing newline.
6248 * doc/gcc/gcc-command-options/machine-dependent-options/blackfin-options.rst:
6249 Add trailing newline.
6250 * doc/gcc/gcc-command-options/machine-dependent-options/c-sky-options.rst:
6251 Add trailing newline.
6252 * doc/gcc/gcc-command-options/machine-dependent-options/c6x-options.rst:
6253 Add trailing newline.
6254 * doc/gcc/gcc-command-options/machine-dependent-options/cris-options.rst:
6255 Add trailing newline.
6256 * doc/gcc/gcc-command-options/machine-dependent-options/darwin-options.rst:
6257 Add trailing newline.
6258 * doc/gcc/gcc-command-options/machine-dependent-options/dec-alpha-options.rst:
6259 Add trailing newline.
6260 * doc/gcc/gcc-command-options/machine-dependent-options/ebpf-options.rst:
6261 Add trailing newline.
6262 * doc/gcc/gcc-command-options/machine-dependent-options/fr30-options.rst:
6263 Add trailing newline.
6264 * doc/gcc/gcc-command-options/machine-dependent-options/frv-options.rst:
6265 Add trailing newline.
6266 * doc/gcc/gcc-command-options/machine-dependent-options/ft32-options.rst:
6267 Add trailing newline.
6268 * doc/gcc/gcc-command-options/machine-dependent-options/gnu-linux-options.rst:
6269 Add trailing newline.
6270 * doc/gcc/gcc-command-options/machine-dependent-options/h8-300-options.rst:
6271 Add trailing newline.
6272 * doc/gcc/gcc-command-options/machine-dependent-options/hppa-options.rst:
6273 Add trailing newline.
6274 * doc/gcc/gcc-command-options/machine-dependent-options/ia-64-options.rst:
6275 Add trailing newline.
6276 * doc/gcc/gcc-command-options/machine-dependent-options/ibm-rs-6000-and-powerpc-options.rst:
6277 Add trailing newline.
6278 * doc/gcc/gcc-command-options/machine-dependent-options/lm32-options.rst:
6279 Add trailing newline.
6280 * doc/gcc/gcc-command-options/machine-dependent-options/loongarch-options.rst:
6281 Add trailing newline.
6282 * doc/gcc/gcc-command-options/machine-dependent-options/m32c-options.rst:
6283 Add trailing newline.
6284 * doc/gcc/gcc-command-options/machine-dependent-options/m32r-d-options.rst:
6285 Add trailing newline.
6286 * doc/gcc/gcc-command-options/machine-dependent-options/m680x0-options.rst:
6287 Add trailing newline.
6288 * doc/gcc/gcc-command-options/machine-dependent-options/mcore-options.rst:
6289 Add trailing newline.
6290 * doc/gcc/gcc-command-options/machine-dependent-options/mep-options.rst:
6291 Add trailing newline.
6292 * doc/gcc/gcc-command-options/machine-dependent-options/microblaze-options.rst:
6293 Add trailing newline.
6294 * doc/gcc/gcc-command-options/machine-dependent-options/mips-options.rst:
6295 Add trailing newline.
6296 * doc/gcc/gcc-command-options/machine-dependent-options/mmix-options.rst:
6297 Add trailing newline.
6298 * doc/gcc/gcc-command-options/machine-dependent-options/mn10300-options.rst:
6299 Add trailing newline.
6300 * doc/gcc/gcc-command-options/machine-dependent-options/moxie-options.rst:
6301 Add trailing newline.
6302 * doc/gcc/gcc-command-options/machine-dependent-options/msp430-options.rst:
6303 Add trailing newline.
6304 * doc/gcc/gcc-command-options/machine-dependent-options/nds32-options.rst:
6305 Add trailing newline.
6306 * doc/gcc/gcc-command-options/machine-dependent-options/nios-ii-options.rst:
6307 Add trailing newline.
6308 * doc/gcc/gcc-command-options/machine-dependent-options/nvidia-ptx-options.rst:
6309 Add trailing newline.
6310 * doc/gcc/gcc-command-options/machine-dependent-options/openrisc-options.rst:
6311 Add trailing newline.
6312 * doc/gcc/gcc-command-options/machine-dependent-options/options-for-system-v.rst:
6313 Add trailing newline.
6314 * doc/gcc/gcc-command-options/machine-dependent-options/pdp-11-options.rst:
6315 Add trailing newline.
6316 * doc/gcc/gcc-command-options/machine-dependent-options/picochip-options.rst:
6317 Add trailing newline.
6318 * doc/gcc/gcc-command-options/machine-dependent-options/powerpc-options.rst:
6319 Add trailing newline.
6320 * doc/gcc/gcc-command-options/machine-dependent-options/pru-options.rst:
6321 Add trailing newline.
6322 * doc/gcc/gcc-command-options/machine-dependent-options/risc-v-options.rst:
6323 Add trailing newline.
6324 * doc/gcc/gcc-command-options/machine-dependent-options/rl78-options.rst:
6325 Add trailing newline.
6326 * doc/gcc/gcc-command-options/machine-dependent-options/rx-options.rst:
6327 Add trailing newline.
6328 * doc/gcc/gcc-command-options/machine-dependent-options/s-390-and-zseries-options.rst:
6329 Add trailing newline.
6330 * doc/gcc/gcc-command-options/machine-dependent-options/score-options.rst:
6331 Add trailing newline.
6332 * doc/gcc/gcc-command-options/machine-dependent-options/sh-options.rst:
6333 Add trailing newline.
6334 * doc/gcc/gcc-command-options/machine-dependent-options/solaris-2-options.rst:
6335 Add trailing newline.
6336 * doc/gcc/gcc-command-options/machine-dependent-options/sparc-options.rst:
6337 Add trailing newline.
6338 * doc/gcc/gcc-command-options/machine-dependent-options/v850-options.rst:
6339 Add trailing newline.
6340 * doc/gcc/gcc-command-options/machine-dependent-options/vax-options.rst:
6341 Add trailing newline.
6342 * doc/gcc/gcc-command-options/machine-dependent-options/visium-options.rst:
6343 Add trailing newline.
6344 * doc/gcc/gcc-command-options/machine-dependent-options/vms-options.rst:
6345 Add trailing newline.
6346 * doc/gcc/gcc-command-options/machine-dependent-options/vxworks-options.rst:
6347 Add trailing newline.
6348 * doc/gcc/gcc-command-options/machine-dependent-options/x86-windows-options.rst:
6349 Add trailing newline.
6350 * doc/gcc/gcc-command-options/machine-dependent-options/xstormy16-options.rst:
6351 Add trailing newline.
6352 * doc/gcc/gcc-command-options/machine-dependent-options/xtensa-options.rst:
6353 Add trailing newline.
6354 * doc/gcc/gcc-command-options/machine-dependent-options/zseries-options.rst:
6355 Add trailing newline.
6356 * doc/gcc/gcc-command-options/options-controlling-c++-dialect.rst:
6357 Add trailing newline.
6358 * doc/gcc/gcc-command-options/options-controlling-c-dialect.rst:
6359 Add trailing newline.
6360 * doc/gcc/gcc-command-options/options-controlling-objective-c-and-objective-c++-dialects.rst:
6361 Add trailing newline.
6362 * doc/gcc/gcc-command-options/options-controlling-the-kind-of-output.rst:
6363 Add trailing newline.
6364 * doc/gcc/gcc-command-options/options-controlling-the-preprocessor.rst:
6365 Add trailing newline.
6366 * doc/gcc/gcc-command-options/options-for-code-generation-conventions.rst:
6367 Add trailing newline.
6368 * doc/gcc/gcc-command-options/options-for-debugging-your-program.rst:
6369 Add trailing newline.
6370 * doc/gcc/gcc-command-options/options-for-directory-search.rst:
6371 Add trailing newline.
6372 * doc/gcc/gcc-command-options/options-for-linking.rst:
6373 Add trailing newline.
6374 * doc/gcc/gcc-command-options/options-that-control-optimization.rst:
6375 Add trailing newline.
6376 * doc/gcc/gcc-command-options/options-that-control-static-analysis.rst:
6377 Add trailing newline.
6378 * doc/gcc/gcc-command-options/options-to-control-diagnostic-messages-formatting.rst:
6379 Add trailing newline.
6380 * doc/gcc/gcc-command-options/options-to-request-or-suppress-warnings.rst:
6381 Add trailing newline.
6382 * doc/gcc/gcc-command-options/passing-options-to-the-assembler.rst:
6383 Add trailing newline.
6384 * doc/gcc/gcc-command-options/program-instrumentation-options.rst:
6385 Add trailing newline.
6386 * doc/gcc/gcc-command-options/specifying-subprocesses-and-the-switches-to-pass-to-them.rst:
6387 Add trailing newline.
6388 * doc/gcc/gcc-command-options/using-precompiled-headers.rst:
6389 Add trailing newline.
6391 Add trailing newline.
6392 * doc/gcc/gcov-dump.rst:
6393 Add trailing newline.
6394 * doc/gcc/gcov-tool.rst:
6395 Add trailing newline.
6397 Add trailing newline.
6398 * doc/gcc/gcov/brief-description-of-gcov-data-files.rst:
6399 Add trailing newline.
6400 * doc/gcc/gcov/data-file-relocation-to-support-cross-profiling.rst:
6401 Add trailing newline.
6402 * doc/gcc/gcov/introduction-to-gcov.rst:
6403 Add trailing newline.
6404 * doc/gcc/gcov/invoking-gcov.rst:
6405 Add trailing newline.
6406 * doc/gcc/gcov/profiling-and-test-coverage-in-freestanding-environments.rst:
6407 Add trailing newline.
6408 * doc/gcc/gcov/using-gcov-with-gcc-optimization.rst:
6409 Add trailing newline.
6410 * doc/gcc/general-public-license-3.rst:
6411 Add trailing newline.
6412 * doc/gcc/gnu-free-documentation-license.rst:
6413 Add trailing newline.
6414 * doc/gcc/gnu-objective-c-features.rst:
6415 Add trailing newline.
6416 * doc/gcc/gnu-objective-c-features/compatibilityalias.rst:
6417 Add trailing newline.
6418 * doc/gcc/gnu-objective-c-features/constant-string-objects.rst:
6419 Add trailing newline.
6420 * doc/gcc/gnu-objective-c-features/exceptions.rst:
6421 Add trailing newline.
6422 * doc/gcc/gnu-objective-c-features/fast-enumeration.rst:
6423 Add trailing newline.
6424 * doc/gcc/gnu-objective-c-features/garbage-collection.rst:
6425 Add trailing newline.
6426 * doc/gcc/gnu-objective-c-features/gnu-objective-c-runtime-api.rst:
6427 Add trailing newline.
6428 * doc/gcc/gnu-objective-c-features/load-executing-code-before-main.rst:
6429 Add trailing newline.
6430 * doc/gcc/gnu-objective-c-features/messaging-with-the-gnu-objective-c-runtime.rst:
6431 Add trailing newline.
6432 * doc/gcc/gnu-objective-c-features/synchronization.rst:
6433 Add trailing newline.
6434 * doc/gcc/gnu-objective-c-features/type-encoding.rst:
6435 Add trailing newline.
6437 Add trailing newline.
6438 * doc/gcc/have-you-found-a-bug.rst:
6439 Add trailing newline.
6440 * doc/gcc/how-and-where-to-report-bugs.rst:
6441 Add trailing newline.
6442 * doc/gcc/how-to-get-help-with-gcc.rst:
6443 Add trailing newline.
6444 * doc/gcc/index.rst:
6445 Add trailing newline.
6446 * doc/gcc/indices-and-tables.rst:
6447 Add trailing newline.
6448 * doc/gcc/known-causes-of-trouble-with-gcc.rst:
6449 Add trailing newline.
6450 * doc/gcc/known-causes-of-trouble-with-gcc/actual-bugs-we-havent-fixed-yet.rst:
6451 Add trailing newline.
6452 * doc/gcc/known-causes-of-trouble-with-gcc/certain-changes-we-dont-want-to-make.rst:
6453 Add trailing newline.
6454 * doc/gcc/known-causes-of-trouble-with-gcc/common-misunderstandings-with-gnu-c.rst:
6455 Add trailing newline.
6456 * doc/gcc/known-causes-of-trouble-with-gcc/disappointments-and-misunderstandings.rst:
6457 Add trailing newline.
6458 * doc/gcc/known-causes-of-trouble-with-gcc/fixed-header-files.rst:
6459 Add trailing newline.
6460 * doc/gcc/known-causes-of-trouble-with-gcc/incompatibilities-of-gcc.rst:
6461 Add trailing newline.
6462 * doc/gcc/known-causes-of-trouble-with-gcc/interoperation.rst:
6463 Add trailing newline.
6464 * doc/gcc/known-causes-of-trouble-with-gcc/standard-libraries.rst:
6465 Add trailing newline.
6466 * doc/gcc/known-causes-of-trouble-with-gcc/warning-messages-and-error-messages.rst:
6467 Add trailing newline.
6468 * doc/gcc/language-standards-supported-by-gcc.rst:
6469 Add trailing newline.
6470 * doc/gcc/language-standards-supported-by-gcc/c++-language.rst:
6471 Add trailing newline.
6472 * doc/gcc/language-standards-supported-by-gcc/c-language.rst:
6473 Add trailing newline.
6474 * doc/gcc/language-standards-supported-by-gcc/d-language.rst:
6475 Add trailing newline.
6476 * doc/gcc/language-standards-supported-by-gcc/go-language.rst:
6477 Add trailing newline.
6478 * doc/gcc/language-standards-supported-by-gcc/objective-c-and-objective-c++-languages.rst:
6479 Add trailing newline.
6480 * doc/gcc/lto-dump.rst:
6481 Add trailing newline.
6482 * doc/gcc/programming-languages-supported-by-gcc.rst:
6483 Add trailing newline.
6484 * doc/gcc/reporting-bugs.rst:
6485 Add trailing newline.
6486 * doc/gccint/analysis-and-optimization-of-gimple-tuples.rst:
6487 Add trailing newline.
6488 * doc/gccint/analysis-and-optimization-of-gimple-tuples/alias-analysis.rst:
6489 Add trailing newline.
6490 * doc/gccint/analysis-and-optimization-of-gimple-tuples/annotations.rst:
6491 Add trailing newline.
6492 * doc/gccint/analysis-and-optimization-of-gimple-tuples/memory-model.rst:
6493 Add trailing newline.
6494 * doc/gccint/analysis-and-optimization-of-gimple-tuples/ssa-operands.rst:
6495 Add trailing newline.
6496 * doc/gccint/analysis-and-optimization-of-gimple-tuples/static-single-assignment.rst:
6497 Add trailing newline.
6498 * doc/gccint/analysis-and-representation-of-loops.rst:
6499 Add trailing newline.
6500 * doc/gccint/analysis-and-representation-of-loops/data-dependency-analysis.rst:
6501 Add trailing newline.
6502 * doc/gccint/analysis-and-representation-of-loops/iv-analysis-on-rtl.rst:
6503 Add trailing newline.
6504 * doc/gccint/analysis-and-representation-of-loops/loop-closed-ssa-form.rst:
6505 Add trailing newline.
6506 * doc/gccint/analysis-and-representation-of-loops/loop-manipulation.rst:
6507 Add trailing newline.
6508 * doc/gccint/analysis-and-representation-of-loops/loop-querying.rst:
6509 Add trailing newline.
6510 * doc/gccint/analysis-and-representation-of-loops/loop-representation.rst:
6511 Add trailing newline.
6512 * doc/gccint/analysis-and-representation-of-loops/number-of-iterations-analysis.rst:
6513 Add trailing newline.
6514 * doc/gccint/analysis-and-representation-of-loops/scalar-evolutions.rst:
6515 Add trailing newline.
6516 * doc/gccint/analyzer-internals.rst:
6517 Add trailing newline.
6518 * doc/gccint/collect2.rst:
6519 Add trailing newline.
6520 * doc/gccint/contributing-to-gcc-development.rst:
6521 Add trailing newline.
6522 * doc/gccint/contributors-to-gcc.rst:
6523 Add trailing newline.
6524 * doc/gccint/control-flow-graph.rst:
6525 Add trailing newline.
6526 * doc/gccint/control-flow-graph/basic-blocks.rst:
6527 Add trailing newline.
6528 * doc/gccint/control-flow-graph/edges.rst:
6529 Add trailing newline.
6530 * doc/gccint/control-flow-graph/liveness-information.rst:
6531 Add trailing newline.
6532 * doc/gccint/control-flow-graph/maintaining-the-cfg.rst:
6533 Add trailing newline.
6534 * doc/gccint/control-flow-graph/profile-information.rst:
6535 Add trailing newline.
6536 * doc/gccint/copyright.rst:
6537 Add trailing newline.
6538 * doc/gccint/debugging-the-analyzer.rst:
6539 Add trailing newline.
6540 * doc/gccint/funding.rst:
6541 Add trailing newline.
6542 * doc/gccint/gcc-and-portability.rst:
6543 Add trailing newline.
6544 * doc/gccint/general-public-license-3.rst:
6545 Add trailing newline.
6546 * doc/gccint/generic.rst:
6547 Add trailing newline.
6548 * doc/gccint/generic/attributes-in-trees.rst:
6549 Add trailing newline.
6550 * doc/gccint/generic/c-and-c++-trees.rst:
6551 Add trailing newline.
6552 * doc/gccint/generic/declarations.rst:
6553 Add trailing newline.
6554 * doc/gccint/generic/deficiencies.rst:
6555 Add trailing newline.
6556 * doc/gccint/generic/expressions.rst:
6557 Add trailing newline.
6558 * doc/gccint/generic/functions.rst:
6559 Add trailing newline.
6560 * doc/gccint/generic/language-dependent-trees.rst:
6561 Add trailing newline.
6562 * doc/gccint/generic/overview.rst:
6563 Add trailing newline.
6564 * doc/gccint/generic/statements.rst:
6565 Add trailing newline.
6566 * doc/gccint/generic/types.rst:
6567 Add trailing newline.
6568 * doc/gccint/gimple-api.rst:
6569 Add trailing newline.
6570 * doc/gccint/gimple.rst:
6571 Add trailing newline.
6572 * doc/gccint/gimple/adding-a-new-gimple-statement-code.rst:
6573 Add trailing newline.
6574 * doc/gccint/gimple/class-hierarchy-of-gimple-statements.rst:
6575 Add trailing newline.
6576 * doc/gccint/gimple/exception-handling.rst:
6577 Add trailing newline.
6578 * doc/gccint/gimple/gimple-instruction-set.rst:
6579 Add trailing newline.
6580 * doc/gccint/gimple/gimple-sequences.rst:
6581 Add trailing newline.
6582 * doc/gccint/gimple/manipulating-gimple-statements.rst:
6583 Add trailing newline.
6584 * doc/gccint/gimple/operands.rst:
6585 Add trailing newline.
6586 * doc/gccint/gimple/sequence-iterators.rst:
6587 Add trailing newline.
6588 * doc/gccint/gimple/statement-and-operand-traversals.rst:
6589 Add trailing newline.
6590 * doc/gccint/gimple/temporaries.rst:
6591 Add trailing newline.
6592 * doc/gccint/gimple/tuple-representation.rst:
6593 Add trailing newline.
6594 * doc/gccint/gimple/tuple-specific-accessors.rst:
6595 Add trailing newline.
6596 * doc/gccint/gimple/tuple-specific-accessors/gimpleasm.rst:
6597 Add trailing newline.
6598 * doc/gccint/gimple/tuple-specific-accessors/gimpleassign.rst:
6599 Add trailing newline.
6600 * doc/gccint/gimple/tuple-specific-accessors/gimplebind.rst:
6601 Add trailing newline.
6602 * doc/gccint/gimple/tuple-specific-accessors/gimplecall.rst:
6603 Add trailing newline.
6604 * doc/gccint/gimple/tuple-specific-accessors/gimplecatch.rst:
6605 Add trailing newline.
6606 * doc/gccint/gimple/tuple-specific-accessors/gimplecond.rst:
6607 Add trailing newline.
6608 * doc/gccint/gimple/tuple-specific-accessors/gimpledebug.rst:
6609 Add trailing newline.
6610 * doc/gccint/gimple/tuple-specific-accessors/gimpleehfilter.rst:
6611 Add trailing newline.
6612 * doc/gccint/gimple/tuple-specific-accessors/gimplegoto.rst:
6613 Add trailing newline.
6614 * doc/gccint/gimple/tuple-specific-accessors/gimplelabel.rst:
6615 Add trailing newline.
6616 * doc/gccint/gimple/tuple-specific-accessors/gimplenop.rst:
6617 Add trailing newline.
6618 * doc/gccint/gimple/tuple-specific-accessors/gimpleompatomicload.rst:
6619 Add trailing newline.
6620 * doc/gccint/gimple/tuple-specific-accessors/gimpleompatomicstore.rst:
6621 Add trailing newline.
6622 * doc/gccint/gimple/tuple-specific-accessors/gimpleompcontinue.rst:
6623 Add trailing newline.
6624 * doc/gccint/gimple/tuple-specific-accessors/gimpleompcritical.rst:
6625 Add trailing newline.
6626 * doc/gccint/gimple/tuple-specific-accessors/gimpleompfor.rst:
6627 Add trailing newline.
6628 * doc/gccint/gimple/tuple-specific-accessors/gimpleompmaster.rst:
6629 Add trailing newline.
6630 * doc/gccint/gimple/tuple-specific-accessors/gimpleompordered.rst:
6631 Add trailing newline.
6632 * doc/gccint/gimple/tuple-specific-accessors/gimpleompparallel.rst:
6633 Add trailing newline.
6634 * doc/gccint/gimple/tuple-specific-accessors/gimpleompreturn.rst:
6635 Add trailing newline.
6636 * doc/gccint/gimple/tuple-specific-accessors/gimpleompsection.rst:
6637 Add trailing newline.
6638 * doc/gccint/gimple/tuple-specific-accessors/gimpleompsections.rst:
6639 Add trailing newline.
6640 * doc/gccint/gimple/tuple-specific-accessors/gimpleompsingle.rst:
6641 Add trailing newline.
6642 * doc/gccint/gimple/tuple-specific-accessors/gimplephi.rst:
6643 Add trailing newline.
6644 * doc/gccint/gimple/tuple-specific-accessors/gimpleresx.rst:
6645 Add trailing newline.
6646 * doc/gccint/gimple/tuple-specific-accessors/gimplereturn.rst:
6647 Add trailing newline.
6648 * doc/gccint/gimple/tuple-specific-accessors/gimpleswitch.rst:
6649 Add trailing newline.
6650 * doc/gccint/gimple/tuple-specific-accessors/gimpletry.rst:
6651 Add trailing newline.
6652 * doc/gccint/gimple/tuple-specific-accessors/gimplewithcleanupexpr.rst:
6653 Add trailing newline.
6654 * doc/gccint/gnu-free-documentation-license.rst:
6655 Add trailing newline.
6656 * doc/gccint/guidelines-for-diagnostics.rst:
6657 Add trailing newline.
6658 * doc/gccint/guidelines-for-options.rst:
6659 Add trailing newline.
6660 * doc/gccint/host-common.rst:
6661 Add trailing newline.
6662 * doc/gccint/host-configuration.rst:
6663 Add trailing newline.
6664 * doc/gccint/host-filesystem.rst:
6665 Add trailing newline.
6666 * doc/gccint/host-makefile-fragments.rst:
6667 Add trailing newline.
6668 * doc/gccint/host-misc.rst:
6669 Add trailing newline.
6670 * doc/gccint/index.rst:
6671 Add trailing newline.
6672 * doc/gccint/indices-and-tables.rst:
6673 Add trailing newline.
6674 * doc/gccint/interfacing-to-gcc-output.rst:
6675 Add trailing newline.
6676 * doc/gccint/introduction.rst:
6677 Add trailing newline.
6678 * doc/gccint/language-front-ends-in-gcc.rst:
6679 Add trailing newline.
6680 * doc/gccint/link-time-optimization.rst:
6681 Add trailing newline.
6682 * doc/gccint/link-time-optimization/design-overview.rst:
6683 Add trailing newline.
6684 * doc/gccint/link-time-optimization/internal-flags-controlling-lto1.rst:
6685 Add trailing newline.
6686 * doc/gccint/link-time-optimization/lto-file-sections.rst:
6687 Add trailing newline.
6688 * doc/gccint/link-time-optimization/using-summary-information-in-ipa-passes.rst:
6689 Add trailing newline.
6690 * doc/gccint/link-time-optimization/whole-program-assumptions-linker-plugin-and-symbol-visibilities.rst:
6691 Add trailing newline.
6692 * doc/gccint/machine-descriptions.rst:
6693 Add trailing newline.
6694 * doc/gccint/machine-descriptions/c-statements-for-assembler-output.rst:
6695 Add trailing newline.
6696 * doc/gccint/machine-descriptions/canonicalization-of-instructions.rst:
6697 Add trailing newline.
6698 * doc/gccint/machine-descriptions/conditional-execution.rst:
6699 Add trailing newline.
6700 * doc/gccint/machine-descriptions/constant-definitions.rst:
6701 Add trailing newline.
6702 * doc/gccint/machine-descriptions/defining-how-to-split-instructions.rst:
6703 Add trailing newline.
6704 * doc/gccint/machine-descriptions/defining-jump-instruction-patterns.rst:
6705 Add trailing newline.
6706 * doc/gccint/machine-descriptions/defining-looping-instruction-patterns.rst:
6707 Add trailing newline.
6708 * doc/gccint/machine-descriptions/defining-rtl-sequences-for-code-generation.rst:
6709 Add trailing newline.
6710 * doc/gccint/machine-descriptions/everything-about-instruction-patterns.rst:
6711 Add trailing newline.
6712 * doc/gccint/machine-descriptions/example-of-defineinsn.rst:
6713 Add trailing newline.
6714 * doc/gccint/machine-descriptions/including-patterns-in-machine-descriptions.rst:
6715 Add trailing newline.
6716 * doc/gccint/machine-descriptions/instruction-attributes.rst:
6717 Add trailing newline.
6718 * doc/gccint/machine-descriptions/interdependence-of-patterns.rst:
6719 Add trailing newline.
6720 * doc/gccint/machine-descriptions/iterators.rst:
6721 Add trailing newline.
6722 * doc/gccint/machine-descriptions/machine-specific-peephole-optimizers.rst:
6723 Add trailing newline.
6724 * doc/gccint/machine-descriptions/operand-constraints.rst:
6725 Add trailing newline.
6726 * doc/gccint/machine-descriptions/output-templates-and-operand-substitution.rst:
6727 Add trailing newline.
6728 * doc/gccint/machine-descriptions/overview-of-how-the-machine-description-is-used.rst:
6729 Add trailing newline.
6730 * doc/gccint/machine-descriptions/predicates.rst:
6731 Add trailing newline.
6732 * doc/gccint/machine-descriptions/rtl-template.rst:
6733 Add trailing newline.
6734 * doc/gccint/machine-descriptions/rtl-templates-transformations.rst:
6735 Add trailing newline.
6736 * doc/gccint/machine-descriptions/standard-pattern-names-for-generation.rst:
6737 Add trailing newline.
6738 * doc/gccint/machine-descriptions/when-the-order-of-patterns-matters.rst:
6739 Add trailing newline.
6740 * doc/gccint/makefile-fragments.rst:
6741 Add trailing newline.
6742 * doc/gccint/match-and-simplify.rst:
6743 Add trailing newline.
6744 * doc/gccint/memory-management-and-type-information.rst:
6745 Add trailing newline.
6746 * doc/gccint/memory-management-and-type-information/how-to-invoke-the-garbage-collector.rst:
6747 Add trailing newline.
6748 * doc/gccint/memory-management-and-type-information/marking-roots-for-the-garbage-collector.rst:
6749 Add trailing newline.
6750 * doc/gccint/memory-management-and-type-information/source-files-containing-type-information.rst:
6751 Add trailing newline.
6752 * doc/gccint/memory-management-and-type-information/support-for-inheritance.rst:
6753 Add trailing newline.
6754 * doc/gccint/memory-management-and-type-information/support-for-user-provided-gc-marking-routines.rst:
6755 Add trailing newline.
6756 * doc/gccint/memory-management-and-type-information/the-inside-of-a-gty.rst:
6757 Add trailing newline.
6758 * doc/gccint/memory-management-and-type-information/troubleshooting-the-garbage-collector.rst:
6759 Add trailing newline.
6760 * doc/gccint/option-file-format.rst:
6761 Add trailing newline.
6762 * doc/gccint/option-properties.rst:
6763 Add trailing newline.
6764 * doc/gccint/option-specification-files.rst:
6765 Add trailing newline.
6766 * doc/gccint/passes-and-files-of-the-compiler.rst:
6767 Add trailing newline.
6768 * doc/gccint/passes-and-files-of-the-compiler/gimplification-pass.rst:
6769 Add trailing newline.
6770 * doc/gccint/passes-and-files-of-the-compiler/inter-procedural-optimization-passes.rst:
6771 Add trailing newline.
6772 * doc/gccint/passes-and-files-of-the-compiler/optimization-info.rst:
6773 Add trailing newline.
6774 * doc/gccint/passes-and-files-of-the-compiler/parsing-pass.rst:
6775 Add trailing newline.
6776 * doc/gccint/passes-and-files-of-the-compiler/pass-manager.rst:
6777 Add trailing newline.
6778 * doc/gccint/passes-and-files-of-the-compiler/rtl-passes.rst:
6779 Add trailing newline.
6780 * doc/gccint/passes-and-files-of-the-compiler/tree-ssa-passes.rst:
6781 Add trailing newline.
6782 * doc/gccint/plugins.rst:
6783 Add trailing newline.
6784 * doc/gccint/plugins/building-gcc-plugins.rst:
6785 Add trailing newline.
6786 * doc/gccint/plugins/controlling-which-passes-are-being-run.rst:
6787 Add trailing newline.
6788 * doc/gccint/plugins/giving-information-about-a-plugin.rst:
6789 Add trailing newline.
6790 * doc/gccint/plugins/interacting-with-the-gcc-garbage-collector.rst:
6791 Add trailing newline.
6792 * doc/gccint/plugins/interacting-with-the-pass-manager.rst:
6793 Add trailing newline.
6794 * doc/gccint/plugins/keeping-track-of-available-passes.rst:
6795 Add trailing newline.
6796 * doc/gccint/plugins/loading-plugins.rst:
6797 Add trailing newline.
6798 * doc/gccint/plugins/plugin-api.rst:
6799 Add trailing newline.
6800 * doc/gccint/plugins/recording-information-about-pass-execution.rst:
6801 Add trailing newline.
6802 * doc/gccint/plugins/registering-custom-attributes-or-pragmas.rst:
6803 Add trailing newline.
6804 * doc/gccint/rtl-representation.rst:
6805 Add trailing newline.
6806 * doc/gccint/rtl-representation/access-to-operands.rst:
6807 Add trailing newline.
6808 * doc/gccint/rtl-representation/access-to-special-operands.rst:
6809 Add trailing newline.
6810 * doc/gccint/rtl-representation/assembler-instructions-as-expressions.rst:
6811 Add trailing newline.
6812 * doc/gccint/rtl-representation/bit-fields.rst:
6813 Add trailing newline.
6814 * doc/gccint/rtl-representation/comparison-operations.rst:
6815 Add trailing newline.
6816 * doc/gccint/rtl-representation/constant-expression-types.rst:
6817 Add trailing newline.
6818 * doc/gccint/rtl-representation/conversions.rst:
6819 Add trailing newline.
6820 * doc/gccint/rtl-representation/declarations.rst:
6821 Add trailing newline.
6822 * doc/gccint/rtl-representation/embedded-side-effects-on-addresses.rst:
6823 Add trailing newline.
6824 * doc/gccint/rtl-representation/flags-in-an-rtl-expression.rst:
6825 Add trailing newline.
6826 * doc/gccint/rtl-representation/insns.rst:
6827 Add trailing newline.
6828 * doc/gccint/rtl-representation/machine-modes.rst:
6829 Add trailing newline.
6830 * doc/gccint/rtl-representation/on-the-side-ssa-form-for-rtl.rst:
6831 Add trailing newline.
6832 * doc/gccint/rtl-representation/reading-rtl.rst:
6833 Add trailing newline.
6834 * doc/gccint/rtl-representation/registers-and-memory.rst:
6835 Add trailing newline.
6836 * doc/gccint/rtl-representation/rtl-classes-and-formats.rst:
6837 Add trailing newline.
6838 * doc/gccint/rtl-representation/rtl-expressions-for-arithmetic.rst:
6839 Add trailing newline.
6840 * doc/gccint/rtl-representation/rtl-object-types.rst:
6841 Add trailing newline.
6842 * doc/gccint/rtl-representation/rtl-representation-of-function-call-insns.rst:
6843 Add trailing newline.
6844 * doc/gccint/rtl-representation/side-effect-expressions.rst:
6845 Add trailing newline.
6846 * doc/gccint/rtl-representation/structure-sharing-assumptions.rst:
6847 Add trailing newline.
6848 * doc/gccint/rtl-representation/variable-location-debug-information-in-rtl.rst:
6849 Add trailing newline.
6850 * doc/gccint/rtl-representation/vector-operations.rst:
6851 Add trailing newline.
6852 * doc/gccint/sizes-and-offsets-as-runtime-invariants.rst:
6853 Add trailing newline.
6854 * doc/gccint/sizes-and-offsets-as-runtime-invariants/alignment-of-polyints.rst:
6855 Add trailing newline.
6856 * doc/gccint/sizes-and-offsets-as-runtime-invariants/arithmetic-on-polyints.rst:
6857 Add trailing newline.
6858 * doc/gccint/sizes-and-offsets-as-runtime-invariants/comparisons-involving-polyint.rst:
6859 Add trailing newline.
6860 * doc/gccint/sizes-and-offsets-as-runtime-invariants/computing-bounds-on-polyints.rst:
6861 Add trailing newline.
6862 * doc/gccint/sizes-and-offsets-as-runtime-invariants/consequences-of-using-polyint.rst:
6863 Add trailing newline.
6864 * doc/gccint/sizes-and-offsets-as-runtime-invariants/converting-polyints.rst:
6865 Add trailing newline.
6866 * doc/gccint/sizes-and-offsets-as-runtime-invariants/guidelines-for-using-polyint.rst:
6867 Add trailing newline.
6868 * doc/gccint/sizes-and-offsets-as-runtime-invariants/miscellaneous-polyint-routines.rst:
6869 Add trailing newline.
6870 * doc/gccint/sizes-and-offsets-as-runtime-invariants/overview-of-polyint.rst:
6871 Add trailing newline.
6872 * doc/gccint/source-tree-structure-and-build-system.rst:
6873 Add trailing newline.
6874 * doc/gccint/source-tree-structure-and-build-system/configure-terms-and-history.rst:
6875 Add trailing newline.
6876 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory.rst:
6877 Add trailing newline.
6878 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/anatomy-of-a-language-front-end.rst:
6879 Add trailing newline.
6880 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/anatomy-of-a-target-back-end.rst:
6881 Add trailing newline.
6882 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/build-system-in-the-gcc-directory.rst:
6883 Add trailing newline.
6884 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/configuration-in-the-gcc-directory.rst:
6885 Add trailing newline.
6886 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/headers-installed-by-gcc.rst:
6887 Add trailing newline.
6888 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/library-source-files-and-headers-under-the-gcc-directory.rst:
6889 Add trailing newline.
6890 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/makefile-targets.rst:
6891 Add trailing newline.
6892 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/subdirectories-of-gcc.rst:
6893 Add trailing newline.
6894 * doc/gccint/source-tree-structure-and-build-system/top-level-source-directory.rst:
6895 Add trailing newline.
6896 * doc/gccint/standard-header-file-directories.rst:
6897 Add trailing newline.
6898 * doc/gccint/static-analyzer.rst:
6899 Add trailing newline.
6900 * doc/gccint/target-macros.rst:
6901 Add trailing newline.
6902 * doc/gccint/target-macros/controlling-debugging-information-format.rst:
6903 Add trailing newline.
6904 * doc/gccint/target-macros/controlling-the-compilation-driver-gcc.rst:
6905 Add trailing newline.
6906 * doc/gccint/target-macros/cross-compilation-and-floating-point.rst:
6907 Add trailing newline.
6908 * doc/gccint/target-macros/defining-coprocessor-specifics-for-mips-targets.rst:
6909 Add trailing newline.
6910 * doc/gccint/target-macros/defining-data-structures-for-per-function-information.rst:
6911 Add trailing newline.
6912 * doc/gccint/target-macros/defining-the-output-assembler-language.rst:
6913 Add trailing newline.
6914 * doc/gccint/target-macros/defining-the-output-assembler-language/assembler-commands-for-alignment.rst:
6915 Add trailing newline.
6916 * doc/gccint/target-macros/defining-the-output-assembler-language/how-initialization-functions-are-handled.rst:
6917 Add trailing newline.
6918 * doc/gccint/target-macros/defining-the-output-assembler-language/macros-controlling-initialization-routines.rst:
6919 Add trailing newline.
6920 * doc/gccint/target-macros/defining-the-output-assembler-language/output-and-generation-of-labels.rst:
6921 Add trailing newline.
6922 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-assembler-instructions.rst:
6923 Add trailing newline.
6924 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-data.rst:
6925 Add trailing newline.
6926 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-uninitialized-variables.rst:
6927 Add trailing newline.
6928 * doc/gccint/target-macros/implicit-calls-to-library-routines.rst:
6929 Add trailing newline.
6930 * doc/gccint/target-macros/layout-of-source-language-data-types.rst:
6931 Add trailing newline.
6932 * doc/gccint/target-macros/position-independent-code.rst:
6933 Add trailing newline.
6934 * doc/gccint/target-macros/register-usage.rst:
6935 Add trailing newline.
6936 * doc/gccint/target-macros/stack-layout-and-calling-conventions.rst:
6937 Add trailing newline.
6938 * doc/gccint/target-macros/stack-layout-and-calling-conventions/basic-stack-layout.rst:
6939 Add trailing newline.
6940 * doc/gccint/target-macros/stack-layout-and-calling-conventions/caller-saves-register-allocation.rst:
6941 Add trailing newline.
6942 * doc/gccint/target-macros/stack-layout-and-calling-conventions/exception-handling-support.rst:
6943 Add trailing newline.
6944 * doc/gccint/target-macros/stack-layout-and-calling-conventions/passing-function-arguments-on-the-stack.rst:
6945 Add trailing newline.
6946 * doc/gccint/target-macros/stack-layout-and-calling-conventions/registers-that-address-the-stack-frame.rst:
6947 Add trailing newline.
6948 * doc/gccint/target-macros/support-for-nested-functions.rst:
6949 Add trailing newline.
6950 * doc/gccint/target-macros/the-global-targetm-variable.rst:
6951 Add trailing newline.
6952 * doc/gccint/target-makefile-fragments.rst:
6953 Add trailing newline.
6954 * doc/gccint/testsuites.rst:
6955 Add trailing newline.
6956 * doc/gccint/testsuites/ada-language-testsuites.rst:
6957 Add trailing newline.
6958 * doc/gccint/testsuites/c-language-testsuites.rst:
6959 Add trailing newline.
6960 * doc/gccint/testsuites/directives-used-within-dejagnu-tests.rst:
6961 Add trailing newline.
6962 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/commands-for-use-in-dg-final.rst:
6963 Add trailing newline.
6964 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/features-for-dg-add-options.rst:
6965 Add trailing newline.
6966 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/keywords-describing-target-attributes.rst:
6967 Add trailing newline.
6968 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/selecting-targets-to-which-a-test-applies.rst:
6969 Add trailing newline.
6970 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/syntax-and-descriptions-of-test-directives.rst:
6971 Add trailing newline.
6972 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/variants-of-dg-require-support.rst:
6973 Add trailing newline.
6974 * doc/gccint/testsuites/idioms-used-in-testsuite-code.rst:
6975 Add trailing newline.
6976 * doc/gccint/testsuites/support-for-testing-binary-compatibility.rst:
6977 Add trailing newline.
6978 * doc/gccint/testsuites/support-for-testing-gcov.rst:
6979 Add trailing newline.
6980 * doc/gccint/testsuites/support-for-testing-gimple-passes.rst:
6981 Add trailing newline.
6982 * doc/gccint/testsuites/support-for-testing-link-time-optimizations.rst:
6983 Add trailing newline.
6984 * doc/gccint/testsuites/support-for-testing-profile-directed-optimizations.rst:
6985 Add trailing newline.
6986 * doc/gccint/testsuites/support-for-testing-rtl-passes.rst:
6987 Add trailing newline.
6988 * doc/gccint/testsuites/support-for-torture-testing-using-multiple-options.rst:
6989 Add trailing newline.
6990 * doc/gccint/the-gcc-low-level-runtime-library.rst:
6991 Add trailing newline.
6992 * doc/gccint/the-gcc-low-level-runtime-library/language-independent-routines-for-exception-handling.rst:
6993 Add trailing newline.
6994 * doc/gccint/the-gcc-low-level-runtime-library/miscellaneous-runtime-library-routines.rst:
6995 Add trailing newline.
6996 * doc/gccint/the-gcc-low-level-runtime-library/routines-for-decimal-floating-point-emulation.rst:
6997 Add trailing newline.
6998 * doc/gccint/the-gcc-low-level-runtime-library/routines-for-fixed-point-fractional-emulation.rst:
6999 Add trailing newline.
7000 * doc/gccint/the-gcc-low-level-runtime-library/routines-for-floating-point-emulation.rst:
7001 Add trailing newline.
7002 * doc/gccint/the-gcc-low-level-runtime-library/routines-for-integer-arithmetic.rst:
7003 Add trailing newline.
7004 * doc/gccint/the-language.rst:
7005 Add trailing newline.
7006 * doc/gccint/user-experience-guidelines.rst:
7007 Add trailing newline.
7008 * doc/install/binaries.rst:
7009 Add trailing newline.
7010 * doc/install/building.rst:
7011 Add trailing newline.
7012 * doc/install/building/building-a-cross-compiler.rst:
7013 Add trailing newline.
7014 * doc/install/building/building-a-native-compiler.rst:
7015 Add trailing newline.
7016 * doc/install/building/building-in-parallel.rst:
7017 Add trailing newline.
7018 * doc/install/building/building-the-ada-compiler.rst:
7019 Add trailing newline.
7020 * doc/install/building/building-the-d-compiler.rst:
7021 Add trailing newline.
7022 * doc/install/building/building-with-profile-feedback.rst:
7023 Add trailing newline.
7024 * doc/install/configuration.rst:
7025 Add trailing newline.
7026 * doc/install/copyright.rst:
7027 Add trailing newline.
7028 * doc/install/downloading-gcc.rst:
7029 Add trailing newline.
7030 * doc/install/final-installation.rst:
7031 Add trailing newline.
7032 * doc/install/gnu-free-documentation-license.rst:
7033 Add trailing newline.
7034 * doc/install/host-target-specific-installation-notes-for-gcc.rst:
7035 Add trailing newline.
7036 * doc/install/how-can-you-run-the-testsuite-on-selected-tests.rst:
7037 Add trailing newline.
7038 * doc/install/how-to-interpret-test-results.rst:
7039 Add trailing newline.
7040 * doc/install/index.rst:
7041 Add trailing newline.
7042 * doc/install/indices-and-tables.rst:
7043 Add trailing newline.
7044 * doc/install/installing-gcc.rst:
7045 Add trailing newline.
7046 * doc/install/passing-options-and-running-multiple-testsuites.rst:
7047 Add trailing newline.
7048 * doc/install/prerequisites.rst:
7049 Add trailing newline.
7050 * doc/install/submitting-test-results.rst:
7051 Add trailing newline.
7052 * doc/install/testing.rst:
7053 Add trailing newline.
7055 2022-11-09 Martin Liska <mliska@suse.cz>
7057 * opts.cc (get_option_html_page): Port to sphinx URLs.
7058 (get_option_url): Likewise.
7059 (test_get_option_html_page): Likewise.
7061 2022-11-09 Jakub Jelinek <jakub@redhat.com>
7063 PR tree-optimization/107569
7064 * range-op-float.cc (foperator_abs::op1_range): Clear NaNs
7065 from the negatives frange before unioning it into r.
7067 2022-11-09 Martin Liska <mliska@suse.cz>
7069 * Makefile.in: Add vpath default for .rst files.
7070 * config/avr/avr-devices.cc: For port RST.
7071 * config/avr/avr-mcus.def: Update path.
7072 * config/avr/gen-avr-mmcu-texi.cc: Moved to...
7073 * config/avr/gen-avr-mmcu-rst.cc: ...here.
7074 * config/avr/t-avr: Update to rst.
7075 * doc/gcc/gcc-command-options/machine-dependent-options/avr-mmcu.rst: Re-generate.
7077 2022-11-09 Philipp Tomsich <philipp.tomsich@vrull.eu>
7079 * config/riscv/riscv.cc (riscv_rtx_costs): Add BSWAP.
7081 2022-11-09 Xi Ruoyao <xry111@xry111.site>
7083 * config/loongarch/loongarch.cc (loongarch_emit_int_compare):
7084 Avoid signed overflow.
7086 2022-11-09 Aldy Hernandez <aldyh@redhat.com>
7088 * range-op-float.cc (foperator_plus::op1_range): New.
7089 (foperator_plus::op2_range): New.
7090 (foperator_minus::op1_range): New.
7091 (foperator_minus::op2_range): New.
7093 2022-11-09 Martin Liska <mliska@suse.cz>
7095 * Makefile.in: Support installation if sphinx-build is missing.
7097 2022-11-09 Martin Liska <mliska@suse.cz>
7099 * doc/gccint/target-macros/stack-layout-and-calling-conventions/passing-arguments-in-registers.rst:
7101 * doc/gccint/target-macros/storage-layout.rst: Likewise.
7102 * doc/gcc/gcc-command-options/machine-dependent-options/x86-options.rst: Sync with master.
7103 * doc/gcc/gcc-command-options/option-summary.rst: Likewise.
7105 2022-11-09 Martin Liska <mliska@suse.cz>
7107 * doc/gccint/target-macros/defining-the-output-assembler-language/assembler-commands-for-exception-regions.rst:
7108 Do not include tm.rst.in.
7109 * doc/gccint/target-macros/defining-the-output-assembler-language/the-overall-framework-of-an-assembler-file.rst:
7111 * doc/gccint/target-macros/run-time-target-specification.rst:
7114 2022-11-09 Martin Liska <mliska@suse.cz>
7116 * doc/gcc/gcc-command-options/options-controlling-the-kind-of-output.rst:
7117 Fix cross manual refs.
7118 * doc/gcc/language-standards-supported-by-gcc/references-for-other-languages.rst:
7121 2022-11-09 Martin Liska <mliska@suse.cz>
7123 * Makefile.in: Support --with-sphinx-build.
7125 * configure: Regenerate.
7127 2022-11-09 Martin Liska <mliska@suse.cz>
7129 * target.def: Use proper lexers for target macros.
7130 * doc/gccint/target-macros/tm.rst.in: Re-generate.
7132 2022-11-09 Martin Liska <mliska@suse.cz>
7134 * doc/analyzer.texi: Removed.
7135 * doc/avr-mmcu.texi: Removed.
7136 * doc/bugreport.texi: Removed.
7137 * doc/cfg.texi: Removed.
7138 * doc/collect2.texi: Removed.
7139 * doc/compat.texi: Removed.
7140 * doc/configfiles.texi: Removed.
7141 * doc/configterms.texi: Removed.
7142 * doc/contrib.texi: Removed.
7143 * doc/contribute.texi: Removed.
7144 * doc/cpp.texi: Removed.
7145 * doc/cppdiropts.texi: Removed.
7146 * doc/cppenv.texi: Removed.
7147 * doc/cppinternals.texi: Removed.
7148 * doc/cppopts.texi: Removed.
7149 * doc/cppwarnopts.texi: Removed.
7150 * doc/extend.texi: Removed.
7151 * doc/fragments.texi: Removed.
7152 * doc/frontends.texi: Removed.
7153 * doc/gcc.texi: Removed.
7154 * doc/gccint.texi: Removed.
7155 * doc/gcov-dump.texi: Removed.
7156 * doc/gcov-tool.texi: Removed.
7157 * doc/gcov.texi: Removed.
7158 * doc/generic.texi: Removed.
7159 * doc/gimple.texi: Removed.
7160 * doc/gnu.texi: Removed.
7161 * doc/gty.texi: Removed.
7162 * doc/headerdirs.texi: Removed.
7163 * doc/hostconfig.texi: Removed.
7164 * doc/implement-c.texi: Removed.
7165 * doc/implement-cxx.texi: Removed.
7166 * doc/include/fdl.texi: Removed.
7167 * doc/include/funding.texi: Removed.
7168 * doc/include/gcc-common.texi: Removed.
7169 * doc/include/gpl_v3.texi: Removed.
7170 * doc/install.texi: Removed.
7171 * doc/interface.texi: Removed.
7172 * doc/invoke.texi: Removed.
7173 * doc/languages.texi: Removed.
7174 * doc/libgcc.texi: Removed.
7175 * doc/loop.texi: Removed.
7176 * doc/lto-dump.texi: Removed.
7177 * doc/lto.texi: Removed.
7178 * doc/makefile.texi: Removed.
7179 * doc/match-and-simplify.texi: Removed.
7180 * doc/md.texi: Removed.
7181 * doc/objc.texi: Removed.
7182 * doc/optinfo.texi: Removed.
7183 * doc/options.texi: Removed.
7184 * doc/passes.texi: Removed.
7185 * doc/plugins.texi: Removed.
7186 * doc/poly-int.texi: Removed.
7187 * doc/portability.texi: Removed.
7188 * doc/rtl.texi: Removed.
7189 * doc/service.texi: Removed.
7190 * doc/sourcebuild.texi: Removed.
7191 * doc/standards.texi: Removed.
7192 * doc/tm.texi: Removed.
7193 * doc/tree-ssa.texi: Removed.
7194 * doc/trouble.texi: Removed.
7195 * doc/ux.texi: Removed.
7196 * doc/tm.texi.in: Removed.
7198 2022-11-09 Martin Liska <mliska@suse.cz>
7200 * doc/gccint/target-macros/tm.rst.in: New file.
7202 2022-11-09 Martin Liska <mliska@suse.cz>
7204 * Makefile.in: Support Sphinx based documentation.
7205 * configure: Regenerate.
7206 * configure.ac: Support Sphinx based documentation.
7207 * genhooks.cc (struct s_hook): Emit tm.rst.in file.
7209 2022-11-09 Martin Liska <mliska@suse.cz>
7211 * doc/gccint/target-macros/adding-support-for-named-address-spaces.rst: Use tm.rst.in file.
7212 * doc/gccint/target-macros/addressing-modes.rst: Likewise.
7213 * doc/gccint/target-macros/adjusting-the-instruction-scheduler.rst: Likewise.
7214 * doc/gccint/target-macros/anchored-addresses.rst: Likewise.
7215 * doc/gccint/target-macros/c++-abi-parameters.rst: Likewise.
7216 * doc/gccint/target-macros/condition-code-status.rst: Likewise.
7217 * doc/gccint/target-macros/controlling-debugging-information-format.rst: Likewise.
7218 * doc/gccint/target-macros/controlling-the-compilation-driver-gcc.rst: Likewise.
7219 * doc/gccint/target-macros/d-abi-parameters.rst: Likewise.
7220 * doc/gccint/target-macros/defining-target-specific-uses-of-attribute.rst: Likewise.
7221 * doc/gccint/target-macros/defining-the-output-assembler-language/assembler-commands-for-exception-regions.rst:
7223 * doc/gccint/target-macros/defining-the-output-assembler-language/macros-controlling-initialization-routines.rst:
7225 * doc/gccint/target-macros/defining-the-output-assembler-language/output-and-generation-of-labels.rst:
7227 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-assembler-instructions.rst:
7229 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-data.rst:
7231 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-dispatch-tables.rst:
7233 * doc/gccint/target-macros/defining-the-output-assembler-language/the-overall-framework-of-an-assembler-file.rst:
7235 * doc/gccint/target-macros/describing-relative-costs-of-operations.rst: Likewise.
7236 * doc/gccint/target-macros/dividing-the-output-into-sections-texts-data.rst: Likewise.
7237 * doc/gccint/target-macros/emulating-tls.rst: Likewise.
7238 * doc/gccint/target-macros/implementing-the-varargs-macros.rst: Likewise.
7239 * doc/gccint/target-macros/implicit-calls-to-library-routines.rst: Likewise.
7240 * doc/gccint/target-macros/layout-of-source-language-data-types.rst: Likewise.
7241 * doc/gccint/target-macros/miscellaneous-parameters.rst: Likewise.
7242 * doc/gccint/target-macros/mode-switching-instructions.rst: Likewise.
7243 * doc/gccint/target-macros/parameters-for-precompiled-header-validity-checking.rst:
7245 * doc/gccint/target-macros/register-classes.rst: Likewise.
7246 * doc/gccint/target-macros/register-usage.rst: Likewise.
7247 * doc/gccint/target-macros/run-time-target-specification.rst: Likewise.
7248 * doc/gccint/target-macros/stack-layout-and-calling-conventions/basic-stack-layout.rst:
7250 * doc/gccint/target-macros/stack-layout-and-calling-conventions/eliminating-frame-pointer-and-arg-pointer.rst:
7252 * doc/gccint/target-macros/stack-layout-and-calling-conventions/function-entry-and-exit.rst:
7254 * doc/gccint/target-macros/stack-layout-and-calling-conventions/generating-code-for-profiling.rst:
7256 * doc/gccint/target-macros/stack-layout-and-calling-conventions/how-large-values-are-returned.rst:
7258 * doc/gccint/target-macros/stack-layout-and-calling-conventions/how-scalar-function-values-are-returned.rst:
7260 * doc/gccint/target-macros/stack-layout-and-calling-conventions/miscellaneous-register-hooks.rst:
7262 * doc/gccint/target-macros/stack-layout-and-calling-conventions/passing-arguments-in-registers.rst:
7264 * doc/gccint/target-macros/stack-layout-and-calling-conventions/passing-function-arguments-on-the-stack.rst:
7266 * doc/gccint/target-macros/stack-layout-and-calling-conventions/permitting-tail-calls.rst:
7268 * doc/gccint/target-macros/stack-layout-and-calling-conventions/registers-that-address-the-stack-frame.rst:
7270 * doc/gccint/target-macros/stack-layout-and-calling-conventions/shrink-wrapping-separate-components.rst:
7272 * doc/gccint/target-macros/stack-layout-and-calling-conventions/specifying-how-stack-checking-is-done.rst:
7274 * doc/gccint/target-macros/stack-layout-and-calling-conventions/stack-smashing-protection.rst:
7276 * doc/gccint/target-macros/storage-layout.rst: Likewise.
7277 * doc/gccint/target-macros/support-for-nested-functions.rst: Likewise.
7279 2022-11-09 Martin Liska <mliska@suse.cz>
7281 * common/common-target.def: Port to RST.
7282 * target.def: Port to RST.
7284 2022-11-09 Martin Liska <mliska@suse.cz>
7286 * doc/cpp/character-sets.rst: New file.
7287 * doc/cpp/conditional-syntax.rst: New file.
7288 * doc/cpp/conditional-uses.rst: New file.
7289 * doc/cpp/conditionals.rst: New file.
7290 * doc/cpp/conf.py: New file.
7291 * doc/cpp/copyright.rst: New file.
7292 * doc/cpp/deleted-code.rst: New file.
7293 * doc/cpp/diagnostics.rst: New file.
7294 * doc/cpp/environment-variables.rst: New file.
7295 * doc/cpp/gnu-free-documentation-license.rst: New file.
7296 * doc/cpp/header-files.rst: New file.
7297 * doc/cpp/header-files/alternatives-to-wrapper-ifndef.rst: New file.
7298 * doc/cpp/header-files/computed-includes.rst: New file.
7299 * doc/cpp/header-files/include-operation.rst: New file.
7300 * doc/cpp/header-files/include-syntax.rst: New file.
7301 * doc/cpp/header-files/once-only-headers.rst: New file.
7302 * doc/cpp/header-files/search-path.rst: New file.
7303 * doc/cpp/header-files/system-headers.rst: New file.
7304 * doc/cpp/header-files/wrapper-headers.rst: New file.
7305 * doc/cpp/implementation-defined-behavior.rst: New file.
7306 * doc/cpp/implementation-details.rst: New file.
7307 * doc/cpp/implementation-limits.rst: New file.
7308 * doc/cpp/index.rst: New file.
7309 * doc/cpp/indices-and-tables.rst: New file.
7310 * doc/cpp/initial-processing.rst: New file.
7311 * doc/cpp/invocation.rst: New file.
7312 * doc/cpp/line-control.rst: New file.
7313 * doc/cpp/macros.rst: New file.
7314 * doc/cpp/macros/concatenation.rst: New file.
7315 * doc/cpp/macros/directives-within-macro-arguments.rst: New file.
7316 * doc/cpp/macros/function-like-macros.rst: New file.
7317 * doc/cpp/macros/macro-arguments.rst: New file.
7318 * doc/cpp/macros/macro-pitfalls.rst: New file.
7319 * doc/cpp/macros/object-like-macros.rst: New file.
7320 * doc/cpp/macros/predefined-macros.rst: New file.
7321 * doc/cpp/macros/stringizing.rst: New file.
7322 * doc/cpp/macros/undefining-and-redefining-macros.rst: New file.
7323 * doc/cpp/macros/variadic-macros.rst: New file.
7324 * doc/cpp/obsolete-features.rst: New file.
7325 * doc/cpp/other-directives.rst: New file.
7326 * doc/cpp/overview.rst: New file.
7327 * doc/cpp/pragmas.rst: New file.
7328 * doc/cpp/preprocessor-output.rst: New file.
7329 * doc/cpp/the-preprocessing-language.rst: New file.
7330 * doc/cpp/tokenization.rst: New file.
7331 * doc/cpp/traditional-lexical-analysis.rst: New file.
7332 * doc/cpp/traditional-macros.rst: New file.
7333 * doc/cpp/traditional-miscellany.rst: New file.
7334 * doc/cpp/traditional-mode.rst: New file.
7335 * doc/cpp/traditional-warnings.rst: New file.
7336 * doc/cppinternals/conf.py: New file.
7337 * doc/cppinternals/copyright.rst: New file.
7338 * doc/cppinternals/cppinternals.rst: New file.
7339 * doc/cppinternals/cpplib.rst: New file.
7340 * doc/cppinternals/files.rst: New file.
7341 * doc/cppinternals/index.rst: New file.
7342 * doc/cppinternals/indices-and-tables.rst: New file.
7343 * doc/cppinternals/internal-representation-of-macros.rst: New file.
7344 * doc/cppinternals/just-which-line-number-anyway.rst: New file.
7345 * doc/cppinternals/lexing-a-line.rst: New file.
7346 * doc/cppinternals/lexing-a-token.rst: New file.
7347 * doc/cppinternals/looking-for-a-function-like-macros-opening-parenthesis.rst: New file.
7348 * doc/cppinternals/macro-expansion-overview.rst: New file.
7349 * doc/cppinternals/marking-tokens-ineligible-for-future-expansion.rst: New file.
7350 * doc/cppinternals/multiple-include-optimization.rst: New file.
7351 * doc/cppinternals/overview.rst: New file.
7352 * doc/cppinternals/representation-of-line-numbers.rst: New file.
7353 * doc/cppinternals/scanning-the-replacement-list-for-macros-to-expand.rst: New file.
7354 * doc/gcc/binary-compatibility.rst: New file.
7355 * doc/gcc/c++-implementation-defined-behavior.rst: New file.
7356 * doc/gcc/c-implementation-defined-behavior.rst: New file.
7357 * doc/gcc/c-implementation-defined-behavior/architecture.rst: New file.
7358 * doc/gcc/c-implementation-defined-behavior/arrays-and-pointers.rst: New file.
7359 * doc/gcc/c-implementation-defined-behavior/characters.rst: New file.
7360 * doc/gcc/c-implementation-defined-behavior/declarators.rst: New file.
7361 * doc/gcc/c-implementation-defined-behavior/environment.rst: New file.
7362 * doc/gcc/c-implementation-defined-behavior/floating-point.rst: New file.
7363 * doc/gcc/c-implementation-defined-behavior/hints.rst: New file.
7364 * doc/gcc/c-implementation-defined-behavior/identifiers.rst: New file.
7365 * doc/gcc/c-implementation-defined-behavior/integers.rst: New file.
7366 * doc/gcc/c-implementation-defined-behavior/library-functions.rst: New file.
7367 * doc/gcc/c-implementation-defined-behavior/locale-specific-behavior.rst: New file.
7368 * doc/gcc/c-implementation-defined-behavior/preprocessing-directives.rst: New file.
7369 * doc/gcc/c-implementation-defined-behavior/qualifiers.rst: New file.
7370 * doc/gcc/c-implementation-defined-behavior/statements.rst: New file.
7371 * doc/gcc/c-implementation-defined-behavior/structures-unions-enumerations-and-bit-fields.rst:
7373 * doc/gcc/c-implementation-defined-behavior/translation.rst: New file.
7374 * doc/gcc/conditionally-supported-behavior.rst: New file.
7375 * doc/gcc/conf.py: New file.
7376 * doc/gcc/contributing-to-gcc-development.rst: New file.
7377 * doc/gcc/contributors-to-gcc.rst: New file.
7378 * doc/gcc/copyright.rst: New file.
7379 * doc/gcc/exception-handling.rst: New file.
7380 * doc/gcc/extensions-to-the-c++-language.rst: New file.
7381 * doc/gcc/extensions-to-the-c++-language/backwards-compatibility.rst: New file.
7382 * doc/gcc/extensions-to-the-c++-language/c++-concepts.rst: New file.
7383 * doc/gcc/extensions-to-the-c++-language/c++-interface-and-implementation-pragmas.rst:
7385 * doc/gcc/extensions-to-the-c++-language/c++-specific-variable-function-and-type-attributes.rst:
7387 * doc/gcc/extensions-to-the-c++-language/deprecated-features.rst: New file.
7388 * doc/gcc/extensions-to-the-c++-language/extracting-the-function-pointer-from-a-bound-pointer-to-member-function.rst:
7390 * doc/gcc/extensions-to-the-c++-language/function-multiversioning.rst: New file.
7391 * doc/gcc/extensions-to-the-c++-language/restricting-pointer-aliasing.rst: New file.
7392 * doc/gcc/extensions-to-the-c++-language/type-traits.rst: New file.
7393 * doc/gcc/extensions-to-the-c++-language/vague-linkage.rst: New file.
7394 * doc/gcc/extensions-to-the-c++-language/when-is-a-volatile-c++-object-accessed.rst:
7396 * doc/gcc/extensions-to-the-c++-language/wheres-the-template.rst: New file.
7397 * doc/gcc/extensions-to-the-c-language-family.rst: New file.
7398 * doc/gcc/extensions-to-the-c-language-family/128-bit-integers.rst: New file.
7399 * doc/gcc/extensions-to-the-c-language-family/additional-floating-types.rst: New file.
7400 * doc/gcc/extensions-to-the-c-language-family/alternate-keywords.rst: New file.
7401 * doc/gcc/extensions-to-the-c-language-family/an-inline-function-is-as-fast-as-a-macro.rst:
7403 * doc/gcc/extensions-to-the-c-language-family/arithmetic-on-void-and-function-pointers.rst:
7405 * doc/gcc/extensions-to-the-c-language-family/arrays-of-length-zero.rst: New file.
7406 * doc/gcc/extensions-to-the-c-language-family/arrays-of-variable-length.rst: New file.
7407 * doc/gcc/extensions-to-the-c-language-family/attribute-syntax.rst: New file.
7408 * doc/gcc/extensions-to-the-c-language-family/binary-constants-using-the-0b-prefix.rst:
7410 * doc/gcc/extensions-to-the-c-language-family/built-in-functions-for-memory-model-aware-atomic-operations.rst:
7412 * doc/gcc/extensions-to-the-c-language-family/built-in-functions-to-perform-arithmetic-with-overflow-checking.rst:
7414 * doc/gcc/extensions-to-the-c-language-family/c++-style-comments.rst: New file.
7415 * doc/gcc/extensions-to-the-c-language-family/case-ranges.rst: New file.
7416 * doc/gcc/extensions-to-the-c-language-family/cast-to-a-union-type.rst: New file.
7417 * doc/gcc/extensions-to-the-c-language-family/complex-numbers.rst: New file.
7418 * doc/gcc/extensions-to-the-c-language-family/compound-literals.rst: New file.
7419 * doc/gcc/extensions-to-the-c-language-family/conditionals-with-omitted-operands.rst:
7421 * doc/gcc/extensions-to-the-c-language-family/constructing-function-calls.rst: New file.
7422 * doc/gcc/extensions-to-the-c-language-family/decimal-floating-types.rst: New file.
7423 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions.rst:
7425 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/aarch64-function-attributes.rst:
7427 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/amd-gcn-function-attributes.rst:
7429 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/arc-function-attributes.rst:
7431 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/arm-function-attributes.rst:
7433 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/avr-function-attributes.rst:
7435 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/blackfin-function-attributes.rst:
7437 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/bpf-function-attributes.rst:
7439 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/c-sky-function-attributes.rst:
7441 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/common-function-attributes.rst:
7443 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/epiphany-function-attributes.rst:
7445 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/h8-300-function-attributes.rst:
7447 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/ia-64-function-attributes.rst:
7449 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/m32c-function-attributes.rst:
7451 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/m32r-d-function-attributes.rst:
7453 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/m68k-function-attributes.rst:
7455 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/mcore-function-attributes.rst:
7457 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/mep-function-attributes.rst:
7459 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/microblaze-function-attributes.rst:
7461 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/microsoft-windows-function-attributes.rst:
7463 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/mips-function-attributes.rst:
7465 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/msp430-function-attributes.rst:
7467 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/nds32-function-attributes.rst:
7469 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/nios-ii-function-attributes.rst:
7471 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/nvidia-ptx-function-attributes.rst:
7473 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/powerpc-function-attributes.rst:
7475 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/risc-v-function-attributes.rst:
7477 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/rl78-function-attributes.rst:
7479 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/rx-function-attributes.rst:
7481 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/s-390-function-attributes.rst:
7483 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/sh-function-attributes.rst:
7485 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/symbian-os-function-attributes.rst:
7487 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/v850-function-attributes.rst:
7489 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/visium-function-attributes.rst:
7491 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/x86-function-attributes.rst:
7493 * doc/gcc/extensions-to-the-c-language-family/declaring-attributes-of-functions/xstormy16-function-attributes.rst:
7495 * doc/gcc/extensions-to-the-c-language-family/designated-initializers.rst: New file.
7496 * doc/gcc/extensions-to-the-c-language-family/determining-the-alignment-of-functions-types-or-variables.rst:
7498 * doc/gcc/extensions-to-the-c-language-family/dollar-signs-in-identifier-names.rst:
7500 * doc/gcc/extensions-to-the-c-language-family/double-word-integers.rst: New file.
7501 * doc/gcc/extensions-to-the-c-language-family/enumerator-attributes.rst: New file.
7502 * doc/gcc/extensions-to-the-c-language-family/fixed-point-types.rst: New file.
7503 * doc/gcc/extensions-to-the-c-language-family/format-checks-specific-to-particular-target-machines.rst:
7505 * doc/gcc/extensions-to-the-c-language-family/function-names-as-strings.rst: New file.
7506 * doc/gcc/extensions-to-the-c-language-family/getting-the-return-or-frame-address-of-a-function.rst:
7508 * doc/gcc/extensions-to-the-c-language-family/half-precision-floating-point.rst: New file.
7509 * doc/gcc/extensions-to-the-c-language-family/hex-floats.rst: New file.
7510 * doc/gcc/extensions-to-the-c-language-family/how-to-use-inline-assembly-language-in-c-code.rst:
7512 * doc/gcc/extensions-to-the-c-language-family/incomplete-enum-types.rst: New file.
7513 * doc/gcc/extensions-to-the-c-language-family/label-attributes.rst: New file.
7514 * doc/gcc/extensions-to-the-c-language-family/labels-as-values.rst: New file.
7515 * doc/gcc/extensions-to-the-c-language-family/legacy-sync-built-in-functions-for-atomic-memory-access.rst:
7517 * doc/gcc/extensions-to-the-c-language-family/locally-declared-labels.rst: New file.
7518 * doc/gcc/extensions-to-the-c-language-family/macros-with-a-variable-number-of-arguments.rst:
7520 * doc/gcc/extensions-to-the-c-language-family/mixed-declarations-labels-and-code.rst:
7522 * doc/gcc/extensions-to-the-c-language-family/named-address-spaces.rst: New file.
7523 * doc/gcc/extensions-to-the-c-language-family/nested-functions.rst: New file.
7524 * doc/gcc/extensions-to-the-c-language-family/non-constant-initializers.rst: New file.
7525 * doc/gcc/extensions-to-the-c-language-family/non-lvalue-arrays-may-have-subscripts.rst:
7527 * doc/gcc/extensions-to-the-c-language-family/nonlocal-gotos.rst: New file.
7528 * doc/gcc/extensions-to-the-c-language-family/object-size-checking-built-in-functions.rst:
7530 * doc/gcc/extensions-to-the-c-language-family/other-built-in-functions-provided-by-gcc.rst:
7532 * doc/gcc/extensions-to-the-c-language-family/pointer-arguments-in-variadic-functions.rst:
7534 * doc/gcc/extensions-to-the-c-language-family/pointers-to-arrays-with-qualifiers-work-as-expected.rst:
7536 * doc/gcc/extensions-to-the-c-language-family/pragmas-accepted-by-gcc.rst: New file.
7537 * doc/gcc/extensions-to-the-c-language-family/prototypes-and-old-style-function-definitions.rst:
7539 * doc/gcc/extensions-to-the-c-language-family/referring-to-a-type-with-typeof.rst: New file.
7540 * doc/gcc/extensions-to-the-c-language-family/slightly-looser-rules-for-escaped-newlines.rst:
7542 * doc/gcc/extensions-to-the-c-language-family/specifying-attributes-of-types.rst: New file.
7543 * doc/gcc/extensions-to-the-c-language-family/specifying-attributes-of-variables.rst:
7545 * doc/gcc/extensions-to-the-c-language-family/statement-attributes.rst: New file.
7546 * doc/gcc/extensions-to-the-c-language-family/statements-and-declarations-in-expressions.rst:
7548 * doc/gcc/extensions-to-the-c-language-family/structures-with-no-members.rst: New file.
7549 * doc/gcc/extensions-to-the-c-language-family/support-for-offsetof.rst: New file.
7550 * doc/gcc/extensions-to-the-c-language-family/target-builtins.rst: New file.
7551 * doc/gcc/extensions-to-the-c-language-family/target-builtins/aarch64-built-in-functions.rst:
7553 * doc/gcc/extensions-to-the-c-language-family/target-builtins/alpha-built-in-functions.rst:
7555 * doc/gcc/extensions-to-the-c-language-family/target-builtins/altera-nios-ii-built-in-functions.rst:
7557 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arc-built-in-functions.rst:
7559 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arc-simd-built-in-functions.rst:
7561 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arm-armv8-m-security-extensions.rst:
7563 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arm-c-language-extensions-acle.rst:
7565 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arm-floating-point-status-and-control-intrinsics.rst:
7567 * doc/gcc/extensions-to-the-c-language-family/target-builtins/arm-iwmmxt-built-in-functions.rst:
7569 * doc/gcc/extensions-to-the-c-language-family/target-builtins/avr-built-in-functions.rst:
7571 * doc/gcc/extensions-to-the-c-language-family/target-builtins/basic-powerpc-built-in-functions.rst:
7573 * doc/gcc/extensions-to-the-c-language-family/target-builtins/blackfin-built-in-functions.rst:
7575 * doc/gcc/extensions-to-the-c-language-family/target-builtins/bpf-built-in-functions.rst:
7577 * doc/gcc/extensions-to-the-c-language-family/target-builtins/fr-v-built-in-functions.rst:
7579 * doc/gcc/extensions-to-the-c-language-family/target-builtins/mips-dsp-built-in-functions.rst:
7581 * doc/gcc/extensions-to-the-c-language-family/target-builtins/mips-loongson-built-in-functions.rst:
7583 * doc/gcc/extensions-to-the-c-language-family/target-builtins/mips-paired-single-support.rst:
7585 * doc/gcc/extensions-to-the-c-language-family/target-builtins/mips-simd-architecture-msa-support.rst:
7587 * doc/gcc/extensions-to-the-c-language-family/target-builtins/msp430-built-in-functions.rst:
7589 * doc/gcc/extensions-to-the-c-language-family/target-builtins/nds32-built-in-functions.rst:
7591 * doc/gcc/extensions-to-the-c-language-family/target-builtins/other-mips-built-in-functions.rst:
7593 * doc/gcc/extensions-to-the-c-language-family/target-builtins/picochip-built-in-functions.rst:
7595 * doc/gcc/extensions-to-the-c-language-family/target-builtins/powerpc-altivec-vsx-built-in-functions.rst:
7597 * doc/gcc/extensions-to-the-c-language-family/target-builtins/powerpc-atomic-memory-operation-functions.rst:
7599 * doc/gcc/extensions-to-the-c-language-family/target-builtins/powerpc-hardware-transactional-memory-built-in-functions.rst:
7601 * doc/gcc/extensions-to-the-c-language-family/target-builtins/powerpc-matrix-multiply-assist-built-in-functions.rst:
7603 * doc/gcc/extensions-to-the-c-language-family/target-builtins/pru-built-in-functions.rst:
7605 * doc/gcc/extensions-to-the-c-language-family/target-builtins/risc-v-built-in-functions.rst:
7607 * doc/gcc/extensions-to-the-c-language-family/target-builtins/rx-built-in-functions.rst:
7609 * doc/gcc/extensions-to-the-c-language-family/target-builtins/s-390-system-z-built-in-functions.rst:
7611 * doc/gcc/extensions-to-the-c-language-family/target-builtins/sh-built-in-functions.rst:
7613 * doc/gcc/extensions-to-the-c-language-family/target-builtins/sparc-vis-built-in-functions.rst:
7615 * doc/gcc/extensions-to-the-c-language-family/target-builtins/ti-c6x-built-in-functions.rst:
7617 * doc/gcc/extensions-to-the-c-language-family/target-builtins/x86-built-in-functions.rst:
7619 * doc/gcc/extensions-to-the-c-language-family/target-builtins/x86-control-flow-protection-intrinsics.rst:
7621 * doc/gcc/extensions-to-the-c-language-family/target-builtins/x86-transactional-memory-intrinsics.rst:
7623 * doc/gcc/extensions-to-the-c-language-family/the-character-esc-in-constants.rst: New file.
7624 * doc/gcc/extensions-to-the-c-language-family/thread-local-storage.rst: New file.
7625 * doc/gcc/extensions-to-the-c-language-family/unnamed-structure-and-union-fields.rst:
7627 * doc/gcc/extensions-to-the-c-language-family/using-vector-instructions-through-built-in-functions.rst:
7629 * doc/gcc/extensions-to-the-c-language-family/when-is-a-volatile-object-accessed.rst:
7631 * doc/gcc/extensions-to-the-c-language-family/x86-specific-memory-model-extensions-for-transactional-memory.rst:
7633 * doc/gcc/funding.rst: New file.
7634 * doc/gcc/gcc-command-options.rst: New file.
7635 * doc/gcc/gcc-command-options/c++-modules.rst: New file.
7636 * doc/gcc/gcc-command-options/compiling-c++-programs.rst: New file.
7637 * doc/gcc/gcc-command-options/description.rst: New file.
7638 * doc/gcc/gcc-command-options/environment-variables-affecting-gcc.rst: New file.
7639 * doc/gcc/gcc-command-options/gcc-developer-options.rst: New file.
7640 * doc/gcc/gcc-command-options/machine-dependent-options.rst: New file.
7641 * doc/gcc/gcc-command-options/machine-dependent-options/aarch64-options.rst: New file.
7642 * doc/gcc/gcc-command-options/machine-dependent-options/adapteva-epiphany-options.rst:
7644 * doc/gcc/gcc-command-options/machine-dependent-options/amd-gcn-options.rst: New file.
7645 * doc/gcc/gcc-command-options/machine-dependent-options/arc-options.rst: New file.
7646 * doc/gcc/gcc-command-options/machine-dependent-options/arm-options.rst: New file.
7647 * doc/gcc/gcc-command-options/machine-dependent-options/avr-mmcu.rst: New file.
7648 * doc/gcc/gcc-command-options/machine-dependent-options/avr-options.rst: New file.
7649 * doc/gcc/gcc-command-options/machine-dependent-options/blackfin-options.rst: New file.
7650 * doc/gcc/gcc-command-options/machine-dependent-options/c-sky-options.rst: New file.
7651 * doc/gcc/gcc-command-options/machine-dependent-options/c6x-options.rst: New file.
7652 * doc/gcc/gcc-command-options/machine-dependent-options/cris-options.rst: New file.
7653 * doc/gcc/gcc-command-options/machine-dependent-options/darwin-options.rst: New file.
7654 * doc/gcc/gcc-command-options/machine-dependent-options/dec-alpha-options.rst: New file.
7655 * doc/gcc/gcc-command-options/machine-dependent-options/ebpf-options.rst: New file.
7656 * doc/gcc/gcc-command-options/machine-dependent-options/fr30-options.rst: New file.
7657 * doc/gcc/gcc-command-options/machine-dependent-options/frv-options.rst: New file.
7658 * doc/gcc/gcc-command-options/machine-dependent-options/ft32-options.rst: New file.
7659 * doc/gcc/gcc-command-options/machine-dependent-options/gnu-linux-options.rst: New file.
7660 * doc/gcc/gcc-command-options/machine-dependent-options/h8-300-options.rst: New file.
7661 * doc/gcc/gcc-command-options/machine-dependent-options/hppa-options.rst: New file.
7662 * doc/gcc/gcc-command-options/machine-dependent-options/ia-64-options.rst: New file.
7663 * doc/gcc/gcc-command-options/machine-dependent-options/ibm-rs-6000-and-powerpc-options.rst:
7665 * doc/gcc/gcc-command-options/machine-dependent-options/lm32-options.rst: New file.
7666 * doc/gcc/gcc-command-options/machine-dependent-options/loongarch-options.rst: New file.
7667 * doc/gcc/gcc-command-options/machine-dependent-options/m32c-options.rst: New file.
7668 * doc/gcc/gcc-command-options/machine-dependent-options/m32r-d-options.rst: New file.
7669 * doc/gcc/gcc-command-options/machine-dependent-options/m680x0-options.rst: New file.
7670 * doc/gcc/gcc-command-options/machine-dependent-options/mcore-options.rst: New file.
7671 * doc/gcc/gcc-command-options/machine-dependent-options/mep-options.rst: New file.
7672 * doc/gcc/gcc-command-options/machine-dependent-options/microblaze-options.rst: New file.
7673 * doc/gcc/gcc-command-options/machine-dependent-options/mips-options.rst: New file.
7674 * doc/gcc/gcc-command-options/machine-dependent-options/mmix-options.rst: New file.
7675 * doc/gcc/gcc-command-options/machine-dependent-options/mn10300-options.rst: New file.
7676 * doc/gcc/gcc-command-options/machine-dependent-options/moxie-options.rst: New file.
7677 * doc/gcc/gcc-command-options/machine-dependent-options/msp430-options.rst: New file.
7678 * doc/gcc/gcc-command-options/machine-dependent-options/nds32-options.rst: New file.
7679 * doc/gcc/gcc-command-options/machine-dependent-options/nios-ii-options.rst: New file.
7680 * doc/gcc/gcc-command-options/machine-dependent-options/nvidia-ptx-options.rst: New file.
7681 * doc/gcc/gcc-command-options/machine-dependent-options/openrisc-options.rst: New file.
7682 * doc/gcc/gcc-command-options/machine-dependent-options/options-for-system-v.rst: New file.
7683 * doc/gcc/gcc-command-options/machine-dependent-options/pdp-11-options.rst: New file.
7684 * doc/gcc/gcc-command-options/machine-dependent-options/picochip-options.rst: New file.
7685 * doc/gcc/gcc-command-options/machine-dependent-options/powerpc-options.rst: New file.
7686 * doc/gcc/gcc-command-options/machine-dependent-options/pru-options.rst: New file.
7687 * doc/gcc/gcc-command-options/machine-dependent-options/risc-v-options.rst: New file.
7688 * doc/gcc/gcc-command-options/machine-dependent-options/rl78-options.rst: New file.
7689 * doc/gcc/gcc-command-options/machine-dependent-options/rx-options.rst: New file.
7690 * doc/gcc/gcc-command-options/machine-dependent-options/s-390-and-zseries-options.rst:
7692 * doc/gcc/gcc-command-options/machine-dependent-options/score-options.rst: New file.
7693 * doc/gcc/gcc-command-options/machine-dependent-options/sh-options.rst: New file.
7694 * doc/gcc/gcc-command-options/machine-dependent-options/solaris-2-options.rst: New file.
7695 * doc/gcc/gcc-command-options/machine-dependent-options/sparc-options.rst: New file.
7696 * doc/gcc/gcc-command-options/machine-dependent-options/v850-options.rst: New file.
7697 * doc/gcc/gcc-command-options/machine-dependent-options/vax-options.rst: New file.
7698 * doc/gcc/gcc-command-options/machine-dependent-options/visium-options.rst: New file.
7699 * doc/gcc/gcc-command-options/machine-dependent-options/vms-options.rst: New file.
7700 * doc/gcc/gcc-command-options/machine-dependent-options/vxworks-options.rst: New file.
7701 * doc/gcc/gcc-command-options/machine-dependent-options/x86-options.rst: New file.
7702 * doc/gcc/gcc-command-options/machine-dependent-options/x86-windows-options.rst: New file.
7703 * doc/gcc/gcc-command-options/machine-dependent-options/xstormy16-options.rst: New file.
7704 * doc/gcc/gcc-command-options/machine-dependent-options/xtensa-options.rst: New file.
7705 * doc/gcc/gcc-command-options/machine-dependent-options/zseries-options.rst: New file.
7706 * doc/gcc/gcc-command-options/option-summary.rst: New file.
7707 * doc/gcc/gcc-command-options/options-controlling-c++-dialect.rst: New file.
7708 * doc/gcc/gcc-command-options/options-controlling-c-dialect.rst: New file.
7709 * doc/gcc/gcc-command-options/options-controlling-objective-c-and-objective-c++-dialects.rst:
7711 * doc/gcc/gcc-command-options/options-controlling-the-kind-of-output.rst: New file.
7712 * doc/gcc/gcc-command-options/options-controlling-the-preprocessor.rst: New file.
7713 * doc/gcc/gcc-command-options/options-for-code-generation-conventions.rst: New file.
7714 * doc/gcc/gcc-command-options/options-for-debugging-your-program.rst: New file.
7715 * doc/gcc/gcc-command-options/options-for-directory-search.rst: New file.
7716 * doc/gcc/gcc-command-options/options-for-linking.rst: New file.
7717 * doc/gcc/gcc-command-options/options-that-control-optimization.rst: New file.
7718 * doc/gcc/gcc-command-options/options-that-control-static-analysis.rst: New file.
7719 * doc/gcc/gcc-command-options/options-to-control-diagnostic-messages-formatting.rst:
7721 * doc/gcc/gcc-command-options/options-to-request-or-suppress-warnings.rst: New file.
7722 * doc/gcc/gcc-command-options/passing-options-to-the-assembler.rst: New file.
7723 * doc/gcc/gcc-command-options/program-instrumentation-options.rst: New file.
7724 * doc/gcc/gcc-command-options/specifying-subprocesses-and-the-switches-to-pass-to-them.rst:
7726 * doc/gcc/gcc-command-options/using-precompiled-headers.rst: New file.
7727 * doc/gcc/gcc.rst: New file.
7728 * doc/gcc/gcov-dump.rst: New file.
7729 * doc/gcc/gcov-tool.rst: New file.
7730 * doc/gcc/gcov.rst: New file.
7731 * doc/gcc/gcov/brief-description-of-gcov-data-files.rst: New file.
7732 * doc/gcc/gcov/data-file-relocation-to-support-cross-profiling.rst: New file.
7733 * doc/gcc/gcov/introduction-to-gcov.rst: New file.
7734 * doc/gcc/gcov/invoking-gcov.rst: New file.
7735 * doc/gcc/gcov/profiling-and-test-coverage-in-freestanding-environments.rst: New file.
7736 * doc/gcc/gcov/using-gcov-with-gcc-optimization.rst: New file.
7737 * doc/gcc/general-public-license-3.rst: New file.
7738 * doc/gcc/gnu-free-documentation-license.rst: New file.
7739 * doc/gcc/gnu-objective-c-features.rst: New file.
7740 * doc/gcc/gnu-objective-c-features/compatibilityalias.rst: New file.
7741 * doc/gcc/gnu-objective-c-features/constant-string-objects.rst: New file.
7742 * doc/gcc/gnu-objective-c-features/exceptions.rst: New file.
7743 * doc/gcc/gnu-objective-c-features/fast-enumeration.rst: New file.
7744 * doc/gcc/gnu-objective-c-features/garbage-collection.rst: New file.
7745 * doc/gcc/gnu-objective-c-features/gnu-objective-c-runtime-api.rst: New file.
7746 * doc/gcc/gnu-objective-c-features/load-executing-code-before-main.rst: New file.
7747 * doc/gcc/gnu-objective-c-features/messaging-with-the-gnu-objective-c-runtime.rst: New file.
7748 * doc/gcc/gnu-objective-c-features/synchronization.rst: New file.
7749 * doc/gcc/gnu-objective-c-features/type-encoding.rst: New file.
7750 * doc/gcc/gnu.rst: New file.
7751 * doc/gcc/have-you-found-a-bug.rst: New file.
7752 * doc/gcc/how-and-where-to-report-bugs.rst: New file.
7753 * doc/gcc/how-to-get-help-with-gcc.rst: New file.
7754 * doc/gcc/index.rst: New file.
7755 * doc/gcc/indices-and-tables.rst: New file.
7756 * doc/gcc/known-causes-of-trouble-with-gcc.rst: New file.
7757 * doc/gcc/known-causes-of-trouble-with-gcc/actual-bugs-we-havent-fixed-yet.rst: New file.
7758 * doc/gcc/known-causes-of-trouble-with-gcc/certain-changes-we-dont-want-to-make.rst:
7760 * doc/gcc/known-causes-of-trouble-with-gcc/common-misunderstandings-with-gnu-c.rst:
7762 * doc/gcc/known-causes-of-trouble-with-gcc/disappointments-and-misunderstandings.rst:
7764 * doc/gcc/known-causes-of-trouble-with-gcc/fixed-header-files.rst: New file.
7765 * doc/gcc/known-causes-of-trouble-with-gcc/incompatibilities-of-gcc.rst: New file.
7766 * doc/gcc/known-causes-of-trouble-with-gcc/interoperation.rst: New file.
7767 * doc/gcc/known-causes-of-trouble-with-gcc/standard-libraries.rst: New file.
7768 * doc/gcc/known-causes-of-trouble-with-gcc/warning-messages-and-error-messages.rst:
7770 * doc/gcc/language-standards-supported-by-gcc.rst: New file.
7771 * doc/gcc/language-standards-supported-by-gcc/c++-language.rst: New file.
7772 * doc/gcc/language-standards-supported-by-gcc/c-language.rst: New file.
7773 * doc/gcc/language-standards-supported-by-gcc/d-language.rst: New file.
7774 * doc/gcc/language-standards-supported-by-gcc/go-language.rst: New file.
7775 * doc/gcc/language-standards-supported-by-gcc/objective-c-and-objective-c++-languages.rst:
7777 * doc/gcc/language-standards-supported-by-gcc/references-for-other-languages.rst: New file.
7778 * doc/gcc/lto-dump.rst: New file.
7779 * doc/gcc/programming-languages-supported-by-gcc.rst: New file.
7780 * doc/gcc/reporting-bugs.rst: New file.
7781 * doc/gccint/analysis-and-optimization-of-gimple-tuples.rst: New file.
7782 * doc/gccint/analysis-and-optimization-of-gimple-tuples/alias-analysis.rst: New file.
7783 * doc/gccint/analysis-and-optimization-of-gimple-tuples/annotations.rst: New file.
7784 * doc/gccint/analysis-and-optimization-of-gimple-tuples/memory-model.rst: New file.
7785 * doc/gccint/analysis-and-optimization-of-gimple-tuples/ssa-operands.rst: New file.
7786 * doc/gccint/analysis-and-optimization-of-gimple-tuples/static-single-assignment.rst:
7788 * doc/gccint/analysis-and-representation-of-loops.rst: New file.
7789 * doc/gccint/analysis-and-representation-of-loops/data-dependency-analysis.rst: New file.
7790 * doc/gccint/analysis-and-representation-of-loops/iv-analysis-on-rtl.rst: New file.
7791 * doc/gccint/analysis-and-representation-of-loops/loop-closed-ssa-form.rst: New file.
7792 * doc/gccint/analysis-and-representation-of-loops/loop-manipulation.rst: New file.
7793 * doc/gccint/analysis-and-representation-of-loops/loop-querying.rst: New file.
7794 * doc/gccint/analysis-and-representation-of-loops/loop-representation.rst: New file.
7795 * doc/gccint/analysis-and-representation-of-loops/number-of-iterations-analysis.rst:
7797 * doc/gccint/analysis-and-representation-of-loops/scalar-evolutions.rst: New file.
7798 * doc/gccint/analyzer-internals.rst: New file.
7799 * doc/gccint/collect2.rst: New file.
7800 * doc/gccint/conf.py: New file.
7801 * doc/gccint/contributing-to-gcc-development.rst: New file.
7802 * doc/gccint/contributors-to-gcc.rst: New file.
7803 * doc/gccint/control-flow-graph.rst: New file.
7804 * doc/gccint/control-flow-graph/basic-blocks.rst: New file.
7805 * doc/gccint/control-flow-graph/edges.rst: New file.
7806 * doc/gccint/control-flow-graph/liveness-information.rst: New file.
7807 * doc/gccint/control-flow-graph/maintaining-the-cfg.rst: New file.
7808 * doc/gccint/control-flow-graph/profile-information.rst: New file.
7809 * doc/gccint/copyright.rst: New file.
7810 * doc/gccint/debugging-the-analyzer.rst: New file.
7811 * doc/gccint/funding.rst: New file.
7812 * doc/gccint/gcc-and-portability.rst: New file.
7813 * doc/gccint/general-public-license-3.rst: New file.
7814 * doc/gccint/generic.rst: New file.
7815 * doc/gccint/generic/attributes-in-trees.rst: New file.
7816 * doc/gccint/generic/c-and-c++-trees.rst: New file.
7817 * doc/gccint/generic/declarations.rst: New file.
7818 * doc/gccint/generic/deficiencies.rst: New file.
7819 * doc/gccint/generic/expressions.rst: New file.
7820 * doc/gccint/generic/functions.rst: New file.
7821 * doc/gccint/generic/language-dependent-trees.rst: New file.
7822 * doc/gccint/generic/overview.rst: New file.
7823 * doc/gccint/generic/statements.rst: New file.
7824 * doc/gccint/generic/types.rst: New file.
7825 * doc/gccint/gimple-api.rst: New file.
7826 * doc/gccint/gimple.rst: New file.
7827 * doc/gccint/gimple/adding-a-new-gimple-statement-code.rst: New file.
7828 * doc/gccint/gimple/class-hierarchy-of-gimple-statements.rst: New file.
7829 * doc/gccint/gimple/exception-handling.rst: New file.
7830 * doc/gccint/gimple/gimple-instruction-set.rst: New file.
7831 * doc/gccint/gimple/gimple-sequences.rst: New file.
7832 * doc/gccint/gimple/manipulating-gimple-statements.rst: New file.
7833 * doc/gccint/gimple/operands.rst: New file.
7834 * doc/gccint/gimple/sequence-iterators.rst: New file.
7835 * doc/gccint/gimple/statement-and-operand-traversals.rst: New file.
7836 * doc/gccint/gimple/temporaries.rst: New file.
7837 * doc/gccint/gimple/tuple-representation.rst: New file.
7838 * doc/gccint/gimple/tuple-specific-accessors.rst: New file.
7839 * doc/gccint/gimple/tuple-specific-accessors/gimpleasm.rst: New file.
7840 * doc/gccint/gimple/tuple-specific-accessors/gimpleassign.rst: New file.
7841 * doc/gccint/gimple/tuple-specific-accessors/gimplebind.rst: New file.
7842 * doc/gccint/gimple/tuple-specific-accessors/gimplecall.rst: New file.
7843 * doc/gccint/gimple/tuple-specific-accessors/gimplecatch.rst: New file.
7844 * doc/gccint/gimple/tuple-specific-accessors/gimplecond.rst: New file.
7845 * doc/gccint/gimple/tuple-specific-accessors/gimpledebug.rst: New file.
7846 * doc/gccint/gimple/tuple-specific-accessors/gimpleehfilter.rst: New file.
7847 * doc/gccint/gimple/tuple-specific-accessors/gimplegoto.rst: New file.
7848 * doc/gccint/gimple/tuple-specific-accessors/gimplelabel.rst: New file.
7849 * doc/gccint/gimple/tuple-specific-accessors/gimplenop.rst: New file.
7850 * doc/gccint/gimple/tuple-specific-accessors/gimpleompatomicload.rst: New file.
7851 * doc/gccint/gimple/tuple-specific-accessors/gimpleompatomicstore.rst: New file.
7852 * doc/gccint/gimple/tuple-specific-accessors/gimpleompcontinue.rst: New file.
7853 * doc/gccint/gimple/tuple-specific-accessors/gimpleompcritical.rst: New file.
7854 * doc/gccint/gimple/tuple-specific-accessors/gimpleompfor.rst: New file.
7855 * doc/gccint/gimple/tuple-specific-accessors/gimpleompmaster.rst: New file.
7856 * doc/gccint/gimple/tuple-specific-accessors/gimpleompordered.rst: New file.
7857 * doc/gccint/gimple/tuple-specific-accessors/gimpleompparallel.rst: New file.
7858 * doc/gccint/gimple/tuple-specific-accessors/gimpleompreturn.rst: New file.
7859 * doc/gccint/gimple/tuple-specific-accessors/gimpleompsection.rst: New file.
7860 * doc/gccint/gimple/tuple-specific-accessors/gimpleompsections.rst: New file.
7861 * doc/gccint/gimple/tuple-specific-accessors/gimpleompsingle.rst: New file.
7862 * doc/gccint/gimple/tuple-specific-accessors/gimplephi.rst: New file.
7863 * doc/gccint/gimple/tuple-specific-accessors/gimpleresx.rst: New file.
7864 * doc/gccint/gimple/tuple-specific-accessors/gimplereturn.rst: New file.
7865 * doc/gccint/gimple/tuple-specific-accessors/gimpleswitch.rst: New file.
7866 * doc/gccint/gimple/tuple-specific-accessors/gimpletry.rst: New file.
7867 * doc/gccint/gimple/tuple-specific-accessors/gimplewithcleanupexpr.rst: New file.
7868 * doc/gccint/gnu-free-documentation-license.rst: New file.
7869 * doc/gccint/guidelines-for-diagnostics.rst: New file.
7870 * doc/gccint/guidelines-for-options.rst: New file.
7871 * doc/gccint/host-common.rst: New file.
7872 * doc/gccint/host-configuration.rst: New file.
7873 * doc/gccint/host-filesystem.rst: New file.
7874 * doc/gccint/host-makefile-fragments.rst: New file.
7875 * doc/gccint/host-misc.rst: New file.
7876 * doc/gccint/index.rst: New file.
7877 * doc/gccint/indices-and-tables.rst: New file.
7878 * doc/gccint/interfacing-to-gcc-output.rst: New file.
7879 * doc/gccint/introduction.rst: New file.
7880 * doc/gccint/language-front-ends-in-gcc.rst: New file.
7881 * doc/gccint/link-time-optimization.rst: New file.
7882 * doc/gccint/link-time-optimization/design-overview.rst: New file.
7883 * doc/gccint/link-time-optimization/internal-flags-controlling-lto1.rst: New file.
7884 * doc/gccint/link-time-optimization/lto-file-sections.rst: New file.
7885 * doc/gccint/link-time-optimization/using-summary-information-in-ipa-passes.rst: New file.
7886 * doc/gccint/link-time-optimization/whole-program-assumptions-linker-plugin-and-symbol-visibilities.rst:
7888 * doc/gccint/machine-descriptions.rst: New file.
7889 * doc/gccint/machine-descriptions/c-statements-for-assembler-output.rst: New file.
7890 * doc/gccint/machine-descriptions/canonicalization-of-instructions.rst: New file.
7891 * doc/gccint/machine-descriptions/conditional-execution.rst: New file.
7892 * doc/gccint/machine-descriptions/constant-definitions.rst: New file.
7893 * doc/gccint/machine-descriptions/defining-how-to-split-instructions.rst: New file.
7894 * doc/gccint/machine-descriptions/defining-jump-instruction-patterns.rst: New file.
7895 * doc/gccint/machine-descriptions/defining-looping-instruction-patterns.rst: New file.
7896 * doc/gccint/machine-descriptions/defining-rtl-sequences-for-code-generation.rst: New file.
7897 * doc/gccint/machine-descriptions/everything-about-instruction-patterns.rst: New file.
7898 * doc/gccint/machine-descriptions/example-of-defineinsn.rst: New file.
7899 * doc/gccint/machine-descriptions/including-patterns-in-machine-descriptions.rst: New file.
7900 * doc/gccint/machine-descriptions/instruction-attributes.rst: New file.
7901 * doc/gccint/machine-descriptions/interdependence-of-patterns.rst: New file.
7902 * doc/gccint/machine-descriptions/iterators.rst: New file.
7903 * doc/gccint/machine-descriptions/machine-specific-peephole-optimizers.rst: New file.
7904 * doc/gccint/machine-descriptions/operand-constraints.rst: New file.
7905 * doc/gccint/machine-descriptions/output-templates-and-operand-substitution.rst: New file.
7906 * doc/gccint/machine-descriptions/overview-of-how-the-machine-description-is-used.rst:
7908 * doc/gccint/machine-descriptions/predicates.rst: New file.
7909 * doc/gccint/machine-descriptions/rtl-template.rst: New file.
7910 * doc/gccint/machine-descriptions/rtl-templates-transformations.rst: New file.
7911 * doc/gccint/machine-descriptions/standard-pattern-names-for-generation.rst: New file.
7912 * doc/gccint/machine-descriptions/when-the-order-of-patterns-matters.rst: New file.
7913 * doc/gccint/makefile-fragments.rst: New file.
7914 * doc/gccint/match-and-simplify.rst: New file.
7915 * doc/gccint/memory-management-and-type-information.rst: New file.
7916 * doc/gccint/memory-management-and-type-information/how-to-invoke-the-garbage-collector.rst:
7918 * doc/gccint/memory-management-and-type-information/marking-roots-for-the-garbage-collector.rst:
7920 * doc/gccint/memory-management-and-type-information/source-files-containing-type-information.rst:
7922 * doc/gccint/memory-management-and-type-information/support-for-inheritance.rst: New file.
7923 * doc/gccint/memory-management-and-type-information/support-for-user-provided-gc-marking-routines.rst:
7925 * doc/gccint/memory-management-and-type-information/the-inside-of-a-gty.rst: New file.
7926 * doc/gccint/memory-management-and-type-information/troubleshooting-the-garbage-collector.rst:
7928 * doc/gccint/option-file-format.rst: New file.
7929 * doc/gccint/option-properties.rst: New file.
7930 * doc/gccint/option-specification-files.rst: New file.
7931 * doc/gccint/passes-and-files-of-the-compiler.rst: New file.
7932 * doc/gccint/passes-and-files-of-the-compiler/gimplification-pass.rst: New file.
7933 * doc/gccint/passes-and-files-of-the-compiler/inter-procedural-optimization-passes.rst:
7935 * doc/gccint/passes-and-files-of-the-compiler/optimization-info.rst: New file.
7936 * doc/gccint/passes-and-files-of-the-compiler/parsing-pass.rst: New file.
7937 * doc/gccint/passes-and-files-of-the-compiler/pass-manager.rst: New file.
7938 * doc/gccint/passes-and-files-of-the-compiler/rtl-passes.rst: New file.
7939 * doc/gccint/passes-and-files-of-the-compiler/tree-ssa-passes.rst: New file.
7940 * doc/gccint/plugins.rst: New file.
7941 * doc/gccint/plugins/building-gcc-plugins.rst: New file.
7942 * doc/gccint/plugins/controlling-which-passes-are-being-run.rst: New file.
7943 * doc/gccint/plugins/giving-information-about-a-plugin.rst: New file.
7944 * doc/gccint/plugins/interacting-with-the-gcc-garbage-collector.rst: New file.
7945 * doc/gccint/plugins/interacting-with-the-pass-manager.rst: New file.
7946 * doc/gccint/plugins/keeping-track-of-available-passes.rst: New file.
7947 * doc/gccint/plugins/loading-plugins.rst: New file.
7948 * doc/gccint/plugins/plugin-api.rst: New file.
7949 * doc/gccint/plugins/recording-information-about-pass-execution.rst: New file.
7950 * doc/gccint/plugins/registering-custom-attributes-or-pragmas.rst: New file.
7951 * doc/gccint/rtl-representation.rst: New file.
7952 * doc/gccint/rtl-representation/access-to-operands.rst: New file.
7953 * doc/gccint/rtl-representation/access-to-special-operands.rst: New file.
7954 * doc/gccint/rtl-representation/assembler-instructions-as-expressions.rst: New file.
7955 * doc/gccint/rtl-representation/bit-fields.rst: New file.
7956 * doc/gccint/rtl-representation/comparison-operations.rst: New file.
7957 * doc/gccint/rtl-representation/constant-expression-types.rst: New file.
7958 * doc/gccint/rtl-representation/conversions.rst: New file.
7959 * doc/gccint/rtl-representation/declarations.rst: New file.
7960 * doc/gccint/rtl-representation/embedded-side-effects-on-addresses.rst: New file.
7961 * doc/gccint/rtl-representation/flags-in-an-rtl-expression.rst: New file.
7962 * doc/gccint/rtl-representation/insns.rst: New file.
7963 * doc/gccint/rtl-representation/machine-modes.rst: New file.
7964 * doc/gccint/rtl-representation/on-the-side-ssa-form-for-rtl.rst: New file.
7965 * doc/gccint/rtl-representation/reading-rtl.rst: New file.
7966 * doc/gccint/rtl-representation/registers-and-memory.rst: New file.
7967 * doc/gccint/rtl-representation/rtl-classes-and-formats.rst: New file.
7968 * doc/gccint/rtl-representation/rtl-expressions-for-arithmetic.rst: New file.
7969 * doc/gccint/rtl-representation/rtl-object-types.rst: New file.
7970 * doc/gccint/rtl-representation/rtl-representation-of-function-call-insns.rst: New file.
7971 * doc/gccint/rtl-representation/side-effect-expressions.rst: New file.
7972 * doc/gccint/rtl-representation/structure-sharing-assumptions.rst: New file.
7973 * doc/gccint/rtl-representation/variable-location-debug-information-in-rtl.rst: New file.
7974 * doc/gccint/rtl-representation/vector-operations.rst: New file.
7975 * doc/gccint/sizes-and-offsets-as-runtime-invariants.rst: New file.
7976 * doc/gccint/sizes-and-offsets-as-runtime-invariants/alignment-of-polyints.rst: New file.
7977 * doc/gccint/sizes-and-offsets-as-runtime-invariants/arithmetic-on-polyints.rst: New file.
7978 * doc/gccint/sizes-and-offsets-as-runtime-invariants/comparisons-involving-polyint.rst:
7980 * doc/gccint/sizes-and-offsets-as-runtime-invariants/computing-bounds-on-polyints.rst:
7982 * doc/gccint/sizes-and-offsets-as-runtime-invariants/consequences-of-using-polyint.rst:
7984 * doc/gccint/sizes-and-offsets-as-runtime-invariants/converting-polyints.rst: New file.
7985 * doc/gccint/sizes-and-offsets-as-runtime-invariants/guidelines-for-using-polyint.rst:
7987 * doc/gccint/sizes-and-offsets-as-runtime-invariants/miscellaneous-polyint-routines.rst:
7989 * doc/gccint/sizes-and-offsets-as-runtime-invariants/overview-of-polyint.rst: New file.
7990 * doc/gccint/source-tree-structure-and-build-system.rst: New file.
7991 * doc/gccint/source-tree-structure-and-build-system/configure-terms-and-history.rst:
7993 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory.rst: New file.
7994 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/anatomy-of-a-language-front-end.rst:
7996 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/anatomy-of-a-target-back-end.rst:
7998 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/build-system-in-the-gcc-directory.rst:
8000 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/building-documentation.rst:
8002 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/configuration-in-the-gcc-directory.rst:
8004 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/headers-installed-by-gcc.rst:
8006 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/library-source-files-and-headers-under-the-gcc-directory.rst:
8008 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/makefile-targets.rst:
8010 * doc/gccint/source-tree-structure-and-build-system/the-gcc-subdirectory/subdirectories-of-gcc.rst:
8012 * doc/gccint/source-tree-structure-and-build-system/top-level-source-directory.rst:
8014 * doc/gccint/standard-header-file-directories.rst: New file.
8015 * doc/gccint/static-analyzer.rst: New file.
8016 * doc/gccint/target-macros.rst: New file.
8017 * doc/gccint/target-macros/adding-support-for-named-address-spaces.rst: New file.
8018 * doc/gccint/target-macros/addressing-modes.rst: New file.
8019 * doc/gccint/target-macros/adjusting-the-instruction-scheduler.rst: New file.
8020 * doc/gccint/target-macros/anchored-addresses.rst: New file.
8021 * doc/gccint/target-macros/c++-abi-parameters.rst: New file.
8022 * doc/gccint/target-macros/condition-code-status.rst: New file.
8023 * doc/gccint/target-macros/controlling-debugging-information-format.rst: New file.
8024 * doc/gccint/target-macros/controlling-the-compilation-driver-gcc.rst: New file.
8025 * doc/gccint/target-macros/cross-compilation-and-floating-point.rst: New file.
8026 * doc/gccint/target-macros/d-abi-parameters.rst: New file.
8027 * doc/gccint/target-macros/defining-coprocessor-specifics-for-mips-targets.rst: New file.
8028 * doc/gccint/target-macros/defining-data-structures-for-per-function-information.rst:
8030 * doc/gccint/target-macros/defining-target-specific-uses-of-attribute.rst: New file.
8031 * doc/gccint/target-macros/defining-the-output-assembler-language.rst: New file.
8032 * doc/gccint/target-macros/defining-the-output-assembler-language/assembler-commands-for-alignment.rst:
8034 * doc/gccint/target-macros/defining-the-output-assembler-language/assembler-commands-for-exception-regions.rst:
8036 * doc/gccint/target-macros/defining-the-output-assembler-language/how-initialization-functions-are-handled.rst:
8038 * doc/gccint/target-macros/defining-the-output-assembler-language/macros-controlling-initialization-routines.rst:
8040 * doc/gccint/target-macros/defining-the-output-assembler-language/output-and-generation-of-labels.rst:
8042 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-assembler-instructions.rst:
8044 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-data.rst:
8046 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-dispatch-tables.rst:
8048 * doc/gccint/target-macros/defining-the-output-assembler-language/output-of-uninitialized-variables.rst:
8050 * doc/gccint/target-macros/defining-the-output-assembler-language/the-overall-framework-of-an-assembler-file.rst:
8052 * doc/gccint/target-macros/describing-relative-costs-of-operations.rst: New file.
8053 * doc/gccint/target-macros/dividing-the-output-into-sections-texts-data.rst: New file.
8054 * doc/gccint/target-macros/emulating-tls.rst: New file.
8055 * doc/gccint/target-macros/implementing-the-varargs-macros.rst: New file.
8056 * doc/gccint/target-macros/implicit-calls-to-library-routines.rst: New file.
8057 * doc/gccint/target-macros/layout-of-source-language-data-types.rst: New file.
8058 * doc/gccint/target-macros/miscellaneous-parameters.rst: New file.
8059 * doc/gccint/target-macros/mode-switching-instructions.rst: New file.
8060 * doc/gccint/target-macros/parameters-for-precompiled-header-validity-checking.rst:
8062 * doc/gccint/target-macros/position-independent-code.rst: New file.
8063 * doc/gccint/target-macros/register-classes.rst: New file.
8064 * doc/gccint/target-macros/register-usage.rst: New file.
8065 * doc/gccint/target-macros/run-time-target-specification.rst: New file.
8066 * doc/gccint/target-macros/stack-layout-and-calling-conventions.rst: New file.
8067 * doc/gccint/target-macros/stack-layout-and-calling-conventions/basic-stack-layout.rst:
8069 * doc/gccint/target-macros/stack-layout-and-calling-conventions/caller-saves-register-allocation.rst:
8071 * doc/gccint/target-macros/stack-layout-and-calling-conventions/eliminating-frame-pointer-and-arg-pointer.rst:
8073 * doc/gccint/target-macros/stack-layout-and-calling-conventions/exception-handling-support.rst:
8075 * doc/gccint/target-macros/stack-layout-and-calling-conventions/function-entry-and-exit.rst:
8077 * doc/gccint/target-macros/stack-layout-and-calling-conventions/generating-code-for-profiling.rst:
8079 * doc/gccint/target-macros/stack-layout-and-calling-conventions/how-large-values-are-returned.rst:
8081 * doc/gccint/target-macros/stack-layout-and-calling-conventions/how-scalar-function-values-are-returned.rst:
8083 * doc/gccint/target-macros/stack-layout-and-calling-conventions/miscellaneous-register-hooks.rst:
8085 * doc/gccint/target-macros/stack-layout-and-calling-conventions/passing-arguments-in-registers.rst:
8087 * doc/gccint/target-macros/stack-layout-and-calling-conventions/passing-function-arguments-on-the-stack.rst:
8089 * doc/gccint/target-macros/stack-layout-and-calling-conventions/permitting-tail-calls.rst:
8091 * doc/gccint/target-macros/stack-layout-and-calling-conventions/registers-that-address-the-stack-frame.rst:
8093 * doc/gccint/target-macros/stack-layout-and-calling-conventions/shrink-wrapping-separate-components.rst:
8095 * doc/gccint/target-macros/stack-layout-and-calling-conventions/specifying-how-stack-checking-is-done.rst:
8097 * doc/gccint/target-macros/stack-layout-and-calling-conventions/stack-smashing-protection.rst:
8099 * doc/gccint/target-macros/storage-layout.rst: New file.
8100 * doc/gccint/target-macros/support-for-nested-functions.rst: New file.
8101 * doc/gccint/target-macros/the-global-targetm-variable.rst: New file.
8102 * doc/gccint/target-makefile-fragments.rst: New file.
8103 * doc/gccint/testsuites.rst: New test.
8104 * doc/gccint/testsuites/ada-language-testsuites.rst: New test.
8105 * doc/gccint/testsuites/c-language-testsuites.rst: New test.
8106 * doc/gccint/testsuites/directives-used-within-dejagnu-tests.rst: New test.
8107 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/commands-for-use-in-dg-final.rst:
8109 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/features-for-dg-add-options.rst:
8111 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/keywords-describing-target-attributes.rst:
8113 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/selecting-targets-to-which-a-test-applies.rst:
8115 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/syntax-and-descriptions-of-test-directives.rst:
8117 * doc/gccint/testsuites/directives-used-within-dejagnu-tests/variants-of-dg-require-support.rst:
8119 * doc/gccint/testsuites/idioms-used-in-testsuite-code.rst: New test.
8120 * doc/gccint/testsuites/support-for-testing-binary-compatibility.rst: New test.
8121 * doc/gccint/testsuites/support-for-testing-gcov.rst: New test.
8122 * doc/gccint/testsuites/support-for-testing-gimple-passes.rst: New test.
8123 * doc/gccint/testsuites/support-for-testing-link-time-optimizations.rst: New test.
8124 * doc/gccint/testsuites/support-for-testing-profile-directed-optimizations.rst: New test.
8125 * doc/gccint/testsuites/support-for-testing-rtl-passes.rst: New test.
8126 * doc/gccint/testsuites/support-for-torture-testing-using-multiple-options.rst: New test.
8127 * doc/gccint/the-gcc-low-level-runtime-library.rst: New file.
8128 * doc/gccint/the-gcc-low-level-runtime-library/language-independent-routines-for-exception-handling.rst:
8130 * doc/gccint/the-gcc-low-level-runtime-library/miscellaneous-runtime-library-routines.rst:
8132 * doc/gccint/the-gcc-low-level-runtime-library/routines-for-decimal-floating-point-emulation.rst:
8134 * doc/gccint/the-gcc-low-level-runtime-library/routines-for-fixed-point-fractional-emulation.rst:
8136 * doc/gccint/the-gcc-low-level-runtime-library/routines-for-floating-point-emulation.rst:
8138 * doc/gccint/the-gcc-low-level-runtime-library/routines-for-integer-arithmetic.rst:
8140 * doc/gccint/the-language.rst: New file.
8141 * doc/gccint/user-experience-guidelines.rst: New file.
8142 * doc/install/binaries.rst: New file.
8143 * doc/install/building.rst: New file.
8144 * doc/install/building/building-a-cross-compiler.rst: New file.
8145 * doc/install/building/building-a-native-compiler.rst: New file.
8146 * doc/install/building/building-in-parallel.rst: New file.
8147 * doc/install/building/building-the-ada-compiler.rst: New file.
8148 * doc/install/building/building-the-d-compiler.rst: New file.
8149 * doc/install/building/building-with-profile-feedback.rst: New file.
8150 * doc/install/conf.py: New file.
8151 * doc/install/configuration.rst: New file.
8152 * doc/install/copyright.rst: New file.
8153 * doc/install/downloading-gcc.rst: New file.
8154 * doc/install/final-installation.rst: New file.
8155 * doc/install/gnu-free-documentation-license.rst: New file.
8156 * doc/install/host-target-specific-installation-notes-for-gcc.rst: New file.
8157 * doc/install/how-can-you-run-the-testsuite-on-selected-tests.rst: New test.
8158 * doc/install/how-to-interpret-test-results.rst: New file.
8159 * doc/install/index.rst: New file.
8160 * doc/install/indices-and-tables.rst: New file.
8161 * doc/install/installing-gcc.rst: New file.
8162 * doc/install/passing-options-and-running-multiple-testsuites.rst: New test.
8163 * doc/install/prerequisites.rst: New file.
8164 * doc/install/submitting-test-results.rst: New file.
8165 * doc/install/testing.rst: New file.
8167 2022-11-09 Aldy Hernandez <aldyh@redhat.com>
8169 * range-op-float.cc (class foperator_minus): New.
8170 (floating_op_table::floating_op_table): Add MINUS_EXPR entry.
8172 2022-11-09 Aldy Hernandez <aldyh@redhat.com>
8174 * range-op-float.cc (range_operator_float::fold_range): Abstract
8175 out from foperator_plus.
8176 (range_operator_float::rv_fold): New.
8177 (foperator_plus::fold_range): Remove.
8178 (foperator_plus::rv_fold): New.
8179 (propagate_nans): Remove.
8180 * range-op.h (class range_operator_float): Add rv_fold.
8182 2022-11-09 Aldy Hernandez <aldyh@redhat.com>
8184 * range-op-float.cc (foperator_plus::fold_range): Set NAN for
8185 addition of different signed infinities.
8186 (range_op_float_tests): New test.
8188 2022-11-09 liuhongt <hongtao.liu@intel.com>
8191 * config/i386/sse.md (avx512f_movddup512<mask_name>): Change
8192 type from sselog to sselog1.
8193 (avx_movddup256<mask_name>): Ditto.
8195 2022-11-08 David Faust <david.faust@oracle.com>
8197 * config/bpf/bpf.cc (struct core_walk_data): Add field `which'...
8198 (bpf_resolve_overloaded_builtin): ... set it here. Use values of enum
8199 bpf_builtins for error checks.
8200 (bpf_core_walk): Use values of enum bpf_builtins.
8201 (bpf_core_newdecl): Likewise.
8202 (bpf_expand_builtin): Likewise.
8204 2022-11-08 Aldy Hernandez <aldyh@redhat.com>
8206 PR tree-optimization/24021
8207 * range-op-float.cc (propagate_nans): New.
8208 (frange_nextafter): New.
8209 (frange_arithmetic): New.
8210 (class foperator_plus): New.
8211 (floating_op_table::floating_op_table): Add PLUS_EXPR entry.
8213 2022-11-08 Aldy Hernandez <aldyh@redhat.com>
8215 * real.cc (real_isdenormal): New.
8216 (encode_ieee_single): Call real_isdenormal.
8217 (encode_ieee_double): Same.
8218 (encode_ieee_extended): Same.
8219 (encode_ieee_quad): Same.
8220 (encode_ieee_half): Same.
8221 (encode_arm_bfloat_half): Same.
8222 * real.h (real_isdenormal): Add mode argument. Rewrite for
8224 * value-range.cc (frange::flush_denormals_to_zero): Pass mode to
8227 2022-11-08 Richard Biener <rguenther@suse.de>
8229 PR tree-optimization/107389
8230 * gimple-low.cc (lower_builtin_assume_aligned): New.
8231 (lower_stmt): Call it.
8233 2022-11-08 Aldy Hernandez <aldyh@redhat.com>
8235 * range-op.cc (operator_div::fold_range): Call
8236 update_known_bitmask.
8237 * tree-ssa-ccp.cc (bit_value_binop): Handle divisions by powers of
8240 2022-11-08 Kwok Cheung Yeung <kcy@codesourcery.com>
8242 * config/gcn/gcn.cc (gcn_expand_builtin_1): Expand first argument
8243 of GCN_BUILTIN_LDEXPV to V64DFmode.
8245 2022-11-08 Kwok Cheung Yeung <kcy@codesourcery.com>
8247 * config/gcn/gcn-builtins.def (FABSV, FLOORVF, FLOORV): New builtins.
8248 * config/gcn/gcn.cc (gcn_expand_builtin_1): Expand GCN_BUILTIN_FABSV,
8249 GCN_BUILTIN_FLOORVF and GCN_BUILTIN_FLOORV.
8251 2022-11-08 Aldy Hernandez <aldyh@redhat.com>
8253 * range-op.cc (irange_to_masked_value): New.
8254 (update_known_bitmask): New.
8255 (operator_mult::fold_range): Call update_known_bitmask.
8257 2022-11-08 Jakub Jelinek <jakub@redhat.com>
8259 PR tree-optimization/107547
8260 * tree-call-cdce.cc (get_no_error_domain): Handle CASE_FLT_FN_FLOATN_NX
8261 of BUILT_IN_{ACOS,ASIN,ACOSH,ATANH,LOG,LOG2,LOG10,LOG1P}. Handle
8262 BUILT_IN_{COSH,SINH,EXP,EXPM1,EXP2}F{16,32,64,128}.
8264 2022-11-08 Jakub Jelinek <jakub@redhat.com>
8267 * config/i386/predicates.md (vector_or_const_vector_operand): New
8269 * config/i386/sse.md (vec_cmp<mode><sseintvecmodelower>,
8270 vec_cmpv2div2di, vec_cmpu<mode><sseintvecmodelower>,
8271 vec_cmpuv2div2di): Use nonimmediate_or_const_vector_operand
8272 predicate instead of nonimmediate_operand and
8273 vector_or_const_vector_operand instead of vector_operand.
8274 * config/i386/i386-expand.cc (ix86_expand_int_sse_cmp): For
8275 LE/LEU or GE/GEU with CONST_VECTOR cop1 try to transform those
8276 into LE/LEU or GT/GTU with larger or smaller by one cop1 if
8277 there is no wrap-around. Force CONST_VECTOR cop0 or cop1 into
8278 REG. Formatting fix.
8280 2022-11-08 Max Filippov <jcmvbkbc@gmail.com>
8282 PR rtl-optimization/107482
8283 * ira-color.cc (assign_hard_reg): Only call
8284 update_costs_from_copies when retry_p is false.
8286 2022-11-08 konglin1 <lingling.kong@intel.com>
8289 2022-11-07 konglin1 <lingling.kong@intel.com>
8291 * config/i386/i386.opt:Add -mprefer-remote-atomic.
8292 * config/i386/sync.md (atomic_<plus_logic><mode>):
8294 (atomic_add<mode>): Rename to below one.
8295 (atomic_add<mode>_1): To this.
8296 (atomic_<logic><mode>): Ditto.
8297 (atomic_<logic><mode>_1): Ditto.
8298 * doc/invoke.texi: Add -mprefer-remote-atomic.
8300 2022-11-08 Haochen Jiang <haochen.jiang@intel.com>
8302 * config/i386/i386-options.cc (m_CORE_ATOM): New.
8303 * config/i386/x86-tune.def
8304 (X86_TUNE_SCHEDULE): Initial tune for CORE_ATOM.
8305 (X86_TUNE_PARTIAL_REG_DEPENDENCY): Ditto.
8306 (X86_TUNE_SSE_PARTIAL_REG_DEPENDENCY): Ditto.
8307 (X86_TUNE_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY): Ditto.
8308 (X86_TUNE_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY): Ditto.
8309 (X86_TUNE_DEST_FALSE_DEP_FOR_GLC): Ditto.
8310 (X86_TUNE_MEMORY_MISMATCH_STALL): Ditto.
8311 (X86_TUNE_USE_LEAVE): Ditto.
8312 (X86_TUNE_PUSH_MEMORY): Ditto.
8313 (X86_TUNE_USE_INCDEC): Ditto.
8314 (X86_TUNE_INTEGER_DFMODE_MOVES): Ditto.
8315 (X86_TUNE_PREFER_KNOWN_REP_MOVSB_STOSB): Ditto.
8316 (X86_TUNE_MISALIGNED_MOVE_STRING_PRO_EPILOGUES): Ditto.
8317 (X86_TUNE_USE_SAHF): Ditto.
8318 (X86_TUNE_USE_BT): Ditto.
8319 (X86_TUNE_AVOID_FALSE_DEP_FOR_BMI): Ditto.
8320 (X86_TUNE_ONE_IF_CONV_INSN): Ditto.
8321 (X86_TUNE_AVOID_MFENCE): Ditto.
8322 (X86_TUNE_USE_SIMODE_FIOP): Ditto.
8323 (X86_TUNE_EXT_80387_CONSTANTS): Ditto.
8324 (X86_TUNE_SSE_UNALIGNED_LOAD_OPTIMAL): Ditto.
8325 (X86_TUNE_SSE_UNALIGNED_STORE_OPTIMAL): Ditto.
8326 (X86_TUNE_SSE_TYPELESS_STORES): Ditto.
8327 (X86_TUNE_SSE_LOAD0_BY_PXOR): Ditto.
8328 (X86_TUNE_AVOID_4BYTE_PREFIXES): Ditto.
8329 (X86_TUNE_USE_GATHER_2PARTS): Ditto.
8330 (X86_TUNE_USE_GATHER_4PARTS): Ditto.
8331 (X86_TUNE_USE_GATHER): Ditto.
8333 2022-11-08 Andrew MacLeod <amacleod@redhat.com>
8335 PR tree-optimization/104530
8336 * gimple-range-cache.cc (ranger_cache::register_inferred_value):
8338 (ranger_cache::apply_inferred_ranges): Move setting cache to
8340 * gimple-range-cache.h (register_inferred_value): New prototype.
8341 * gimple-range-infer.cc (infer_range_manager::has_range_p): New.
8342 * gimple-range-infer.h (has_range_p): New prototype.
8343 * gimple-range.cc (register_transitive_inferred_ranges): New.
8344 * gimple-range.h (register_transitive_inferred_ranges): New proto.
8345 * tree-vrp.cc (rvrp_folder::fold_stmt): Check for transitive inferred
8346 ranges at the end of the block before folding final stmt.
8348 2022-11-07 David Faust <david.faust@oracle.com>
8350 * config/bpf/bpf.cc (handle_attr_preserve): Use maybe_make_core_relo().
8352 2022-11-07 Aldy Hernandez <aldyh@redhat.com>
8354 PR tree-optimization/55157
8355 * range-op.cc (operator_mult::wi_fold): Optimize multiplications
8358 2022-11-07 H.J. Lu <hjl.tools@gmail.com>
8360 PR middle-end/102566
8361 * tree-ssa-ccp.cc (optimize_atomic_bit_test_and): Also handle
8362 if (_5 < 0) and if (_5 >= 0).
8364 2022-11-07 Richard Purdie <richard.purdie@linuxfoundation.org>
8366 * file-prefix-map.cc (remap_filename): Handle NULL filenames.
8368 2022-11-07 Alexander Monakov <amonakov@ispras.ru>
8370 PR tree-optimization/107505
8371 * tree-ssa-sink.cc (statement_sink_location): Additionally
8372 reject ECF_RETURNS_TWICE calls.
8374 2022-11-07 Aldy Hernandez <aldyh@redhat.com>
8376 PR tree-optimization/107541
8377 * range-op.cc (operator_div::fold_range): Restrict power of 2
8378 optimization to positive numbers.
8380 2022-11-07 Richard Biener <rguenther@suse.de>
8382 * tree-ssa-loop-unswitch.cc (unswitch_predicate::count): New.
8383 (unswitch_predicate::unswitch_predicate): Initialize count.
8384 (init_loop_unswitch_info): First collect candidates and
8385 determine the outermost loop to unswitch.
8386 (tree_ssa_unswitch_loops): First perform all guard hoisting,
8387 then perform unswitching on innermost loop predicates.
8388 (find_unswitching_predicates_for_bb): Keep track of the
8389 most profitable predicate to unswitch on.
8390 (tree_unswitch_single_loop): Unswitch given predicate if
8393 2022-11-07 Martin Liska <mliska@suse.cz>
8394 Gerald Pfeifer <gerald@pfeifer.com>
8396 * doc/invoke.texi: Improve wording.
8398 2022-11-07 Martin Liska <mliska@suse.cz>
8400 * range-op.cc: Add final override keywords.
8402 2022-11-07 Kewen Lin <linkw@linux.ibm.com>
8404 PR tree-optimization/107412
8405 * gimple-fold.cc (gimple_fold_mask_load_store_mem_ref): Rename to ...
8406 (gimple_fold_partial_load_store_mem_ref): ... this, add one parameter
8407 mask_p indicating it's for mask or length, and add some handlings for
8408 IFN LEN_{LOAD,STORE}.
8409 (gimple_fold_mask_load): Rename to ...
8410 (gimple_fold_partial_load): ... this, add one parameter mask_p.
8411 (gimple_fold_mask_store): Rename to ...
8412 (gimple_fold_partial_store): ... this, add one parameter mask_p.
8413 (gimple_fold_call): Add the handlings for IFN LEN_{LOAD,STORE},
8414 and adjust calls on gimple_fold_mask_load_store_mem_ref to
8415 gimple_fold_partial_load_store_mem_ref.
8417 2022-11-07 Hu, Lin1 <lin1.hu@intel.com>
8419 * common/config/i386/cpuinfo.h
8420 (get_intel_cpu): Handle Grand Ridge.
8421 * common/config/i386/i386-common.cc
8422 (processor_names): Add grandridge.
8423 (processor_alias_table): Ditto.
8424 * common/config/i386/i386-cpuinfo.h:
8425 (enum processor_types): Add INTEL_GRANDRIDGE.
8426 * config.gcc: Add -march=grandridge.
8427 * config/i386/driver-i386.cc (host_detect_local_cpu):
8429 * config/i386/i386-c.cc (ix86_target_macros_internal):
8431 * config/i386/i386-options.cc (m_GRANDRIDGE): New define.
8432 (processor_cost_table): Add grandridge.
8433 * config/i386/i386.h (enum processor_type):
8434 Add PROCESSOR_GRANDRIDGE.
8435 (PTA_GRANDRIDGE): Ditto.
8436 * doc/extend.texi: Add grandridge.
8437 * doc/invoke.texi: Ditto.
8439 2022-11-07 konglin1 <lingling.kong@intel.com>
8441 * config/i386/i386.opt:Add -mprefer-remote-atomic.
8442 * config/i386/sync.md (atomic_<plus_logic><mode>):
8444 (atomic_add<mode>): Rename to below one.
8445 (atomic_add<mode>_1): To this.
8446 (atomic_<logic><mode>): Ditto.
8447 (atomic_<logic><mode>_1): Ditto.
8448 * doc/invoke.texi: Add -mprefer-remote-atomic.
8450 2022-11-07 konglin1 <lingling.kong@intel.com>
8452 * common/config/i386/cpuinfo.h (get_available_features):
8454 * common/config/i386/i386-common.cc (OPTION_MASK_ISA2_RAOINT_SET,
8455 OPTION_MASK_ISA2_RAOINT_UNSET): New.
8456 (ix86_handle_option): Handle -mraoint.
8457 * common/config/i386/i386-cpuinfo.h (enum processor_features):
8459 * common/config/i386/i386-isas.h: Add ISA_NAME_TABLE_ENTRY for
8461 * config.gcc: Add raointintrin.h
8462 * config/i386/cpuid.h (bit_RAOINT): New.
8463 * config/i386/i386-builtin.def (BDESC): Add new builtins.
8464 * config/i386/i386-c.cc (ix86_target_macros_internal): Define
8466 * config/i386/i386-isa.def (RAOINT): Add DEF_PTA(RAOINT).
8467 * config/i386/i386-options.cc (ix86_valid_target_attribute_inner_p):
8469 * config/i386/sync.md (rao_a<raointop><mode>): New define insn.
8470 * config/i386/i386.opt: Add option -mraoint.
8471 * config/i386/x86gprintrin.h: Include raointintrin.h.
8472 * doc/extend.texi: Document raoint.
8473 * doc/invoke.texi: Document -mraoint.
8474 * doc/sourcebuild.texi: Document target raoint.
8475 * config/i386/raointintrin.h: New file.
8477 2022-11-07 Haochen Jiang <haochen.jiang@intel.com>
8479 * common/config/i386/cpuinfo.h
8480 (get_intel_cpu): Handle Granite Rapids.
8481 * common/config/i386/i386-common.cc:
8482 (processor_names): Add graniterapids.
8483 (processor_alias_table): Ditto.
8484 * common/config/i386/i386-cpuinfo.h
8485 (enum processor_subtypes): Add INTEL_GRANTIERAPIDS.
8486 * config.gcc: Add -march=graniterapids.
8487 * config/i386/driver-i386.cc (host_detect_local_cpu):
8488 Handle graniterapids.
8489 * config/i386/i386-c.cc (ix86_target_macros_internal):
8491 * config/i386/i386-options.cc (m_GRANITERAPIDS): New.
8492 (processor_cost_table): Add graniterapids.
8493 * config/i386/i386.h (enum processor_type):
8494 Add PROCESSOR_GRANITERAPIDS.
8495 (PTA_GRANITERAPIDS): Ditto.
8496 * doc/extend.texi: Add graniterapids.
8497 * doc/invoke.texi: Ditto.
8499 2022-11-07 Haochen Jiang <haochen.jiang@intel.com>
8500 Hongtao Liu <hongtao.liu@intel.com>
8502 * common/config/i386/cpuinfo.h (get_available_features):
8504 * common/config/i386/i386-common.cc
8505 (OPTION_MASK_ISA2_PREFETCHI_SET,
8506 OPTION_MASK_ISA2_PREFETCHI_UNSET): New.
8507 (ix86_handle_option): Handle -mprefetchi.
8508 * common/config/i386/i386-cpuinfo.h
8509 (enum processor_features): Add FEATURE_PREFETCHI.
8510 * common/config/i386/i386-isas.h: Add ISA_NAME_TABLE_ENTRY
8512 * config.gcc: Add prfchiintrin.h.
8513 * config/i386/cpuid.h (bit_PREFETCHI): New.
8514 * config/i386/i386-builtin-types.def:
8515 Add DEF_FUNCTION_TYPE (VOID, PCVOID, INT)
8516 and DEF_FUNCTION_TYPE (VOID, PCVOID, INT, INT, INT).
8517 * config/i386/i386-builtin.def (BDESC): Add new builtins.
8518 * config/i386/i386-c.cc (ix86_target_macros_internal):
8519 Define __PREFETCHI__.
8520 * config/i386/i386-expand.cc: Handle new builtins.
8521 * config/i386/i386-isa.def (PREFETCHI):
8522 Add DEF_PTA(PREFETCHI).
8523 * config/i386/i386-options.cc
8524 (ix86_valid_target_attribute_inner_p): Handle prefetchi.
8525 * config/i386/i386.md (prefetchi): New define_insn.
8526 * config/i386/i386.opt: Add option -mprefetchi.
8527 * config/i386/predicates.md (local_func_symbolic_operand):
8529 * config/i386/x86gprintrin.h: Include prfchiintrin.h.
8530 * config/i386/xmmintrin.h (enum _mm_hint): New enum for
8532 (_mm_prefetch): Handle the highest bit of enum.
8533 * doc/extend.texi: Document prefetchi.
8534 * doc/invoke.texi: Document -mprefetchi.
8535 * doc/sourcebuild.texi: Document target prefetchi.
8536 * config/i386/prfchiintrin.h: New file.
8538 2022-11-06 Uroš Bizjak <ubizjak@gmail.com>
8540 * optabs.cc (can_vec_set_var_idx_p): Use operand[2]
8541 mode when checking vec_set_optab.
8543 2022-11-06 Xi Ruoyao <xry111@xry111.site>
8545 * config/loongarch/loongarch.md (UNSPEC_FCOPYSIGN): New unspec.
8546 (type): Add fcopysign.
8547 (copysign<mode>3): New instruction template.
8549 2022-11-06 Bernhard Reutner-Fischer <aldot@gcc.gnu.org>
8551 * multiple_target.cc (expand_target_clones): Free memory.
8553 2022-11-06 Bernhard Reutner-Fischer <aldot@gcc.gnu.org>
8555 * cgraph.cc (cgraph_node::make_local): Remove redundant set_section.
8556 * multiple_target.cc (create_dispatcher_calls): Likewise.
8558 2022-11-05 Jonathan Wakely <jwakely@redhat.com>
8561 * doc/cppopts.texi: Document -fwide-exec-charset defaults
8564 2022-11-05 Alexandre Oliva <oliva@adacore.com>
8566 * common.opt (fmultiflags): New.
8567 * doc/invoke.texi: Document it.
8568 * gcc.cc (driver_self_specs): Discard it.
8569 * opts.cc (common_handle_option): Ignore it in the driver.
8571 2022-11-04 Jeff Chapman II <jchapman@lock3software.com>
8572 Jason Merrill <jason@redhat.com>
8574 * input.cc (get_source_text_between): New fn.
8575 * input.h (get_source_text_between): Declare.
8577 2022-11-04 Aldy Hernandez <aldyh@redhat.com>
8579 PR tree-optimization/107342
8580 * range-op.cc (operator_mult::fold_range): New.
8581 (operator_div::fold_range): New.
8583 2022-11-04 Eric Botcazou <ebotcazou@adacore.com>
8585 * fold-const.cc (operand_compare::operand_equal_p) <COMPONENT_REF>:
8586 Do not take into account operand 2.
8587 (operand_compare::hash_operand) <COMPONENT_REF>: Likewise.
8589 2022-11-04 Eric Botcazou <ebotcazou@adacore.com>
8591 * expr.cc (emit_group_store): Do not use subword paradoxical subregs
8593 2022-11-04 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
8595 * config/aarch64/aarch64-sve2.md: Fix typo in Cryptographic
8598 2022-11-04 Thomas Schwinge <thomas@codesourcery.com>
8600 * config.gcc [target *-intelmic-* | *-intelmicemul-*]: Remove.
8601 * config/i386/i386-options.cc (ix86_omp_device_kind_arch_isa)
8602 [ACCEL_COMPILER]: Remove.
8603 * config/i386/intelmic-mkoffload.cc: Remove.
8604 * config/i386/intelmic-offload.h: Likewise.
8605 * config/i386/t-intelmic: Likewise.
8606 * config/i386/t-omp-device: Likewise.
8607 * configure.ac [target *-intelmic-* | *-intelmicemul-*]: Remove.
8608 * configure: Regenerate.
8609 * doc/install.texi (--enable-offload-targets=[...]): Update.
8610 * doc/sourcebuild.texi: Remove 'liboffloadmic' documentation.
8612 2022-11-04 Thomas Schwinge <thomas@codesourcery.com>
8614 * Makefile.in (OBJS): Remove 'dbxout.o'.
8615 * config/nvptx/nvptx.cc: Don't '#include "dbxout.h"'.
8616 * dbxout.cc: Remove.
8617 * dbxout.h: Likewise.
8618 * target-def.h (TARGET_ASM_CONSTRUCTOR, TARGET_ASM_DESTRUCTOR):
8619 Default to 'default_asm_out_constructor',
8620 'default_asm_out_destructor'.
8621 * targhooks.cc (default_asm_out_constructor)
8622 (default_asm_out_destructor): New.
8623 * targhooks.h (default_asm_out_constructor)
8624 (default_asm_out_destructor): Declare.
8626 2022-11-04 Thomas Schwinge <thomas@codesourcery.com>
8628 * Makefile.in (OBJS): Add 'dbxout.o'.
8629 * config/nvptx/nvptx.cc: '#include "dbxout.h"'.
8631 * dbxout.h: Likewise.
8632 * target-def.h (TARGET_ASM_CONSTRUCTOR, TARGET_ASM_DESTRUCTOR):
8633 Default to 'default_stabs_asm_out_constructor',
8634 'default_stabs_asm_out_destructor'.
8636 2022-11-04 Hongyu Wang <hongyu.wang@intel.com>
8637 Haochen Jiang <haochen.jiang@intel.com>
8639 * common/config/i386/cpuinfo.h (get_available_features): Detect
8641 * common/config/i386/i386-common.cc (OPTION_MASK_ISA2_AMX_FP16_SET,
8642 OPTION_MASK_ISA2_AMX_FP16_UNSET): New macros.
8643 (ix86_handle_option): Handle -mamx-fp16.
8644 * common/config/i386/i386-cpuinfo.h (enum processor_features):
8645 Add FEATURE_AMX_FP16.
8646 * common/config/i386/i386-isas.h: Add ISA_NAME_TABLE_ENTRY for
8648 * config.gcc: Add amxfp16intrin.h.
8649 * config/i386/cpuid.h (bit_AMX_FP16): New.
8650 * config/i386/i386-c.cc (ix86_target_macros_internal): Define
8652 * config/i386/i386-isa.def: Add DEF_PTA for AMX_FP16.
8653 * config/i386/i386-options.cc (isa2_opts): Add -mamx-fp16.
8654 (ix86_valid_target_attribute_inner_p): Add new ATTR.
8655 (ix86_option_override_internal): Handle AMX-FP16.
8656 * config/i386/i386.opt: Add -mamx-fp16.
8657 * config/i386/immintrin.h: Include amxfp16intrin.h.
8658 * doc/extend.texi: Document -mamx-fp16.
8659 * doc/invoke.texi: Document amx-fp16.
8660 * doc/sourcebuild.texi: Document amx_fp16.
8661 * config/i386/amxfp16intrin.h: New file.
8663 2022-11-04 Haochen Jiang <haochen.jiang@intel.com>
8665 * common/config/i386/cpuinfo.h (get_intel_cpu):
8667 * common/config/i386/i386-common.cc
8668 (processor_names): Add Sierra Forest.
8669 (processor_alias_table): Ditto.
8670 * common/config/i386/i386-cpuinfo.h
8671 (enum processor_types): Add INTEL_SIERRAFOREST.
8672 * config.gcc: Add -march=sierraforest.
8673 * config/i386/driver-i386.cc (host_detect_local_cpu):
8674 Handle Sierra Forest.
8675 * config/i386/i386-c.cc (ix86_target_macros_internal):
8677 * config/i386/i386-options.cc (m_SIERRAFOREST): New define.
8678 (processor_cost_table): Add sierra forest.
8679 * config/i386/i386.h (enum processor_type):
8680 Add PROCESSOR_SIERRA_FOREST.
8681 (PTA_SIERRAFOREST): Ditto.
8682 * doc/extend.texi: Add sierra forest.
8683 * doc/invoke.texi: Ditto.
8685 2022-11-04 Haochen Jiang <haochen.jiang@intel.com>
8687 * common/config/i386/cpuinfo.h (get_available_features):
8689 * common/config/i386/i386-common.cc
8690 (OPTION_MASK_ISA2_CMPCCXADD_SET,
8691 OPTION_MASK_ISA2_CMPCCXADD_UNSET): New.
8692 (ix86_handle_option): Handle -mcmpccxadd.
8693 * common/config/i386/i386-cpuinfo.h (enum processor_features):
8694 Add FEATURE_CMPCCXADD.
8695 * common/config/i386/i386-isas.h: Add ISA_NAME_TABLE_ENTRY for
8697 * config.gcc: Add cmpccxaddintrin.h.
8698 * config/i386/cpuid.h (bit_CMPCCXADD): New.
8699 * config/i386/i386-builtin-types.def:
8700 Add DEF_FUNCTION_TYPE(INT, PINT, INT, INT, INT)
8701 and DEF_FUNCTION_TYPE(LONGLONG, PLONGLONG, LONGLONG, LONGLONG, INT).
8702 * config/i386/i386-builtin.def (BDESC): Add new builtins.
8703 * config/i386/i386-c.cc (ix86_target_macros_internal): Define
8705 * config/i386/i386-expand.cc (ix86_expand_special_args_builtin):
8706 Add new parameter to indicate constant position.
8707 Handle INT_FTYPE_PINT_INT_INT_INT
8708 and LONGLONG_FTYPE_PLONGLONG_LONGLONG_LONGLONG_INT.
8709 * config/i386/i386-isa.def (CMPCCXADD): Add DEF_PTA(CMPCCXADD).
8710 * config/i386/i386-options.cc (isa2_opts): Add -mcmpccxadd.
8711 (ix86_valid_target_attribute_inner_p): Handle cmpccxadd.
8712 * config/i386/i386.opt: Add option -mcmpccxadd.
8713 * config/i386/sync.md (cmpccxadd_<mode>): New define insn.
8714 * config/i386/x86gprintrin.h: Include cmpccxaddintrin.h.
8715 * doc/extend.texi: Document cmpccxadd.
8716 * doc/invoke.texi: Document -mcmpccxadd.
8717 * doc/sourcebuild.texi: Document target cmpccxadd.
8718 * config/i386/cmpccxaddintrin.h: New file.
8720 2022-11-03 Jason Merrill <jason@redhat.com>
8722 * doc/invoke.texi: -fconcepts no longer implies
8723 -fconcepts-ts before C++20.
8725 2022-11-03 Kwok Cheung Yeung <kcy@codesourcery.com>
8727 * config/gcn/gcn-valu.md (math_unop_insn): New attribute.
8728 (<math_unop><mode>2, <math_unop><mode>2<exec>, <math_unop><mode>2,
8729 <math_unop><mode>2<exec>, *<math_unop><mode>2_insn,
8730 *<math_unop><mode>2<exec>_insn): Use math_unop_insn to generate
8733 2022-11-03 Andrew MacLeod <amacleod@redhat.com>
8735 * gimple-range.cc (gimple_ranger::update_stmt): New.
8736 * gimple-range.h (gimple_ranger::update_stmt): New prototype.
8737 * tree-ssa-operands.cc (update_stmt_operands): Notify range
8738 query that stmt has changed.
8739 * value-query.h (range_query::update_stmt): New.
8741 2022-11-03 David Malcolm <dmalcolm@redhat.com>
8743 * make-unique.h: New file.
8745 2022-11-03 Uroš Bizjak <ubizjak@gmail.com>
8748 * config/i386/i386.md (eliminate reg-reg move by inverting the
8749 condition of a cmove #2 peephole2): Check if eliminated move
8750 initialized a register, used in the moved instruction.
8752 2022-11-03 Andrew Stubbs <ams@codesourcery.com>
8755 * config/gcn/gcn.cc (gcn_expand_reduc_scalar): Remove duplicate
8756 UNSPEC_SMIN_DPP_SHR conditionals.
8758 2022-11-02 Christoph Müllner <christoph.muellner@vrull.eu>
8760 * common/config/riscv/riscv-common.cc: Add zawrs extension.
8761 * config/riscv/riscv-opts.h (MASK_ZAWRS): New.
8762 (TARGET_ZAWRS): New.
8763 * config/riscv/riscv.opt: New.
8765 2022-11-02 Rasmus Villemoes <rv@rasmusvillemoes.dk>
8768 * gcc.cc (ASM_MAP): Honour -ffile-prefix-map.
8770 2022-11-02 Aldy Hernandez <aldyh@redhat.com>
8772 * value-range.cc (frange::contains_p): Fix signed zero handling.
8773 (range_tests_signed_zeros): New test.
8775 2022-11-02 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
8777 * builtins.cc (mathfn_built_in_2): #undef HUGE_VAL, NAN.
8779 2022-11-02 Xionghu Luo <xionghuluo@tencent.com>
8782 * config/rs6000/altivec.md: (*altivec_vrl<VI_char>): Named to...
8783 (altivec_vrl<VI_char>): ...this.
8784 * config/rs6000/vsx.md (revb_<mode>): Call vspltish and vrlh when
8785 target is Power8 and mode is V8HI.
8787 2022-11-01 David Seifert <soap@gentoo.org>
8789 * configure.ac: Add AC_CACHE_CHECK(..., gcc_cv_sys_sdt_h).
8790 * configure: Regenerate.
8792 2022-11-01 Richard Purdie <richard.purdie@linuxfoundation.org>
8794 * file-prefix-map.cc (remap_filename): Allow remapping of relative paths.
8796 2022-11-01 Aldy Hernandez <aldyh@redhat.com>
8798 PR tree-optimization/107490
8799 * range-op-float.cc (foperator_unordered_lt::op1_range): Handle
8801 (foperator_unordered_lt::op2_range): Same.
8802 (foperator_unordered_le::op1_range): Same.
8803 (foperator_unordered_le::op2_range): Same.
8804 (foperator_unordered_gt::op1_range): Same.
8805 (foperator_unordered_gt::op2_range): Same.
8806 (foperator_unordered_ge::op1_range): Same.
8807 (foperator_unordered_ge::op2_range): Same.
8809 2022-11-01 Andrew MacLeod <amacleod@redhat.com>
8811 PR tree-optimization/107497
8812 * tree-vrp.cc (remove_unreachable::remove_and_update_globals):
8813 Check that ssa-name still exists before accessing it.
8815 2022-11-01 Andrew MacLeod <amacleod@redhat.com>
8817 * params.opt (param_vrp1_mode): Make ranger default.
8819 2022-11-01 Andrew MacLeod <amacleod@redhat.com>
8821 * tree-vrp.cc (class remove_unreachable): New.
8822 (remove_unreachable::maybe_register_block): New.
8823 (remove_unreachable::remove_and_update_globals): New.
8824 (rvrp_folder::rvrp_folder): Initialize m_unreachable.
8825 (rvrp_folder::post_fold_bb): Maybe register unreachable block.
8826 (rvrp_folder::m_unreachable): New member.
8827 (execute_ranger_vrp): Add final_pass flag, remove unreachables.
8829 2022-11-01 Andrew MacLeod <amacleod@redhat.com>
8831 * gimple-range-cache.cc (ranger_cache::fill_block_cache): Allow
8832 exit block to be specified.
8833 (ranger_cache::range_from_dom): If exit block is specified, use
8834 the immediate predecessor instead of the dominator to start.
8835 * gimple-range.cc (gimple_ranger::range_on_exit): Allow query
8838 2022-11-01 Andrew MacLeod <amacleod@redhat.com>
8840 * value-range.cc (irange::intersect_nonzero_bits): If new
8841 non-zero mask is the same as original, flag no change.
8843 2022-11-01 Alexander Monakov <amonakov@ispras.ru>
8846 * config/i386/znver.md (znver1_idiv): New automaton.
8847 (znver1-idiv): New unit.
8848 (znver1_idiv_DI): Correct unit and cycles in the reservation.
8849 (znver1_idiv_SI): Ditto.
8850 (znver1_idiv_HI): Ditto.
8851 (znver1_idiv_QI): Ditto.
8852 (znver1_idiv_mem_DI): Ditto.
8853 (znver1_idiv_mem_SI): Ditto.
8854 (znver1_idiv_mem_HI): Ditto.
8855 (znver1_idiv_mem_QI): Ditto.
8856 (znver3_idiv_DI): Ditto.
8857 (znver3_idiv_SI): Ditto.
8858 (znver3_idiv_HI): Ditto.
8859 (znver3_idiv_QI): Ditto.
8860 (znver3_idiv_mem_DI): Ditto.
8861 (znver3_idiv_mem_SI): Ditto.
8862 (znver3_idiv_mem_HI): Ditto.
8863 (znver3_idiv_mem_QI): Ditto.
8865 2022-11-01 liuhongt <hongtao.liu@intel.com>
8868 * config/i386/sse.md (*vec_interleave_highv2df): Remove
8870 (*vec_interleave_lowv2df): Ditto.
8871 (vec_concatv2df): Ditto.
8872 (*avx512f_unpcklpd512<mask_name>): Ditto and renamed to ..
8873 (avx512f_unpcklpd512<mask_name>): .. this.
8874 (avx512f_movddup512<mask_name>): Change to define_insn.
8875 (avx_movddup256<mask_name>): Ditto.
8876 (*avx_unpcklpd256<mask_name>): Remove constraint 1 and renamed
8878 (avx_unpcklpd256<mask_name>): .. this.
8879 * config/i386/i386.cc (ix86_vec_interleave_v2df_operator_ok):
8880 Disallow MEM_P (op1) && MEM_P (op2).
8882 2022-11-01 liuhongt <hongtao.liu@intel.com>
8885 * config/i386/i386.md (*x86_64_shld_1): Rename to ..
8886 (x86_64_shld_1): .. this.
8887 (*x86_shld_1): Rename to ..
8888 (x86_shld_1): .. this.
8889 (*x86_64_shrd_1): Rename to ..
8890 (x86_64_shrd_1): .. this.
8891 (*x86_shrd_1): Rename to ..
8892 (x86_shrd_1): .. this.
8893 (*x86_64_shld_shrd_1_nozext): New pre_reload splitter.
8894 (*x86_shld_shrd_1_nozext): Ditto.
8895 (*x86_64_shrd_shld_1_nozext): Ditto.
8896 (*x86_shrd_shld_1_nozext): Ditto.
8898 2022-11-01 Cui,Lili <lili.cui@intel.com>
8900 * ipa-inline-analysis.cc (do_estimate_edge_time): Add function attribute
8901 judgement for INLINE_HINT_known_hot hint.
8903 2022-10-31 Guillermo E. Martinez <guillermo.e.martinez@oracle.com>
8905 * btfout.cc (btf_calc_num_vbytes): Compute enumeration size depending of
8906 enumerator type btf_enum{,64}.
8907 (btf_asm_type): Update btf_kflag according to enumeration type sign
8908 using dtd_enum_unsigned field for both: BTF_KIND_ENUM{,64}.
8909 (btf_asm_enum_const): New argument to represent the size of
8910 the BTF enum type, writing the enumerator constant value for
8911 32 bits, if it's 64 bits then explicitly writes lower 32-bits
8912 value and higher 32-bits value.
8913 (output_asm_btf_enum_list): Add enumeration size argument.
8914 * ctfc.cc (ctf_add_enum): New argument to represent CTF enum
8916 (ctf_add_generic): Use of ei_{name. size, unsigned} to build the
8917 dtd structure containing enumeration information.
8918 (ctf_add_enumerator): Update comment mention support for BTF
8919 enumeration in 64-bits.
8920 * dwarf2ctf.cc (gen_ctf_enumeration_type): Extract signedness
8921 for enumeration type and use it in ctf_add_enum.
8922 * ctfc.h (ctf_dmdef): Update dmd_value to HOST_WIDE_INT to allow
8923 use 32/64 bits enumerators.
8925 (ctf_dtdef): New field to describe enum signedness.
8927 2022-10-31 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
8929 * config/riscv/riscv-vector-builtins-bases.cc: Change constexpr back to CONSTEXPR.
8930 * config/riscv/riscv-vector-builtins-shapes.cc (SHAPE): Ditto.
8931 * config/riscv/riscv-vector-builtins.cc (struct registered_function_hasher): Ditto.
8932 * config/riscv/riscv-vector-builtins.h (struct rvv_arg_type_info): Ditto.
8934 2022-10-31 Andrew Stubbs <ams@codesourcery.com>
8936 * config/gcn/gcn-valu.md (fminmaxop): New iterator.
8937 (<fexpander><mode>3): New define_expand.
8938 (<fexpander><mode>3<exec>): Likewise.
8939 (reduc_<fexpander>_scal_<mode>): Likewise.
8940 * config/gcn/gcn.md (fexpander): New attribute.
8942 2022-10-31 Andrew Stubbs <ams@codesourcery.com>
8944 * config/gcn/gcn-valu.md (V64_SI): Delete iterator.
8946 (V64_1REG): Likewise.
8947 (V64_INT_1REG): Likewise.
8948 (V64_2REG): Likewise.
8949 (V64_ALL): Likewise.
8951 (reduc_<reduc_op>_scal_<mode>): Use V_ALL. Use gen_vec_extract.
8952 (fold_left_plus_<mode>): Use V_FP.
8953 (*<reduc_op>_dpp_shr_<mode>): Use V_1REG.
8954 (*<reduc_op>_dpp_shr_<mode>): Use V_DI.
8955 (*plus_carry_dpp_shr_<mode>): Use V_INT_1REG.
8956 (*plus_carry_in_dpp_shr_<mode>): Use V_SI.
8957 (*plus_carry_dpp_shr_<mode>): Use V_DI.
8958 (mov_from_lane63_<mode>): Delete.
8959 (mov_from_lane63_<mode>): Delete.
8960 * config/gcn/gcn.cc (gcn_expand_reduc_scalar): Support partial vectors.
8961 * config/gcn/gcn.md (unspec): Remove UNSPEC_MOV_FROM_LANE63.
8963 2022-10-31 Andrew Stubbs <ams@codesourcery.com>
8965 * config/gcn/gcn.cc (gcn_simd_clone_compute_vecsize_and_simdlen):
8966 Set base_type as ARG_UNUSED.
8968 2022-10-31 Jakub Jelinek <jakub@redhat.com>
8970 * builtin-types.def (BT_COMPLEX_FLOAT16, BT_COMPLEX_FLOAT32,
8971 BT_COMPLEX_FLOAT64, BT_COMPLEX_FLOAT128, BT_COMPLEX_FLOAT32X,
8972 BT_COMPLEX_FLOAT64X, BT_COMPLEX_FLOAT128X,
8973 BT_FN_COMPLEX_FLOAT16_COMPLEX_FLOAT16,
8974 BT_FN_COMPLEX_FLOAT32_COMPLEX_FLOAT32,
8975 BT_FN_COMPLEX_FLOAT64_COMPLEX_FLOAT64,
8976 BT_FN_COMPLEX_FLOAT128_COMPLEX_FLOAT128,
8977 BT_FN_COMPLEX_FLOAT32X_COMPLEX_FLOAT32X,
8978 BT_FN_COMPLEX_FLOAT64X_COMPLEX_FLOAT64X,
8979 BT_FN_COMPLEX_FLOAT128X_COMPLEX_FLOAT128X,
8980 BT_FN_FLOAT16_COMPLEX_FLOAT16, BT_FN_FLOAT32_COMPLEX_FLOAT32,
8981 BT_FN_FLOAT64_COMPLEX_FLOAT64, BT_FN_FLOAT128_COMPLEX_FLOAT128,
8982 BT_FN_FLOAT32X_COMPLEX_FLOAT32X, BT_FN_FLOAT64X_COMPLEX_FLOAT64X,
8983 BT_FN_FLOAT128X_COMPLEX_FLOAT128X,
8984 BT_FN_COMPLEX_FLOAT16_COMPLEX_FLOAT16_COMPLEX_FLOAT16,
8985 BT_FN_COMPLEX_FLOAT32_COMPLEX_FLOAT32_COMPLEX_FLOAT32,
8986 BT_FN_COMPLEX_FLOAT64_COMPLEX_FLOAT64_COMPLEX_FLOAT64,
8987 BT_FN_COMPLEX_FLOAT128_COMPLEX_FLOAT128_COMPLEX_FLOAT128,
8988 BT_FN_COMPLEX_FLOAT32X_COMPLEX_FLOAT32X_COMPLEX_FLOAT32X,
8989 BT_FN_COMPLEX_FLOAT64X_COMPLEX_FLOAT64X_COMPLEX_FLOAT64X,
8990 BT_FN_COMPLEX_FLOAT128X_COMPLEX_FLOAT128X_COMPLEX_FLOAT128X): New.
8991 * builtins.def (CABS_TYPE, CACOSH_TYPE, CARG_TYPE, CASINH_TYPE,
8992 CPOW_TYPE, CPROJ_TYPE): Define and undefine later.
8993 (BUILT_IN_CABS, BUILT_IN_CACOSH, BUILT_IN_CACOS, BUILT_IN_CARG,
8994 BUILT_IN_CASINH, BUILT_IN_CASIN, BUILT_IN_CATANH, BUILT_IN_CATAN,
8995 BUILT_IN_CCOSH, BUILT_IN_CCOS, BUILT_IN_CEXP, BUILT_IN_CLOG,
8996 BUILT_IN_CPOW, BUILT_IN_CPROJ, BUILT_IN_CSINH, BUILT_IN_CSIN,
8997 BUILT_IN_CSQRT, BUILT_IN_CTANH, BUILT_IN_CTAN): Add
8998 DEF_EXT_LIB_FLOATN_NX_BUILTINS.
8999 * fold-const-call.cc (fold_const_call_sc, fold_const_call_cc,
9000 fold_const_call_ccc): Add various CASE_CFN_*_FN: cases when
9001 CASE_CFN_* is present.
9002 * gimple-ssa-backprop.cc (backprop::process_builtin_call_use):
9004 * builtins.cc (expand_builtin, fold_builtin_1): Likewise.
9005 * fold-const.cc (negate_mathfn_p, tree_expr_finite_p,
9006 tree_expr_maybe_signaling_nan_p, tree_expr_maybe_nan_p,
9007 tree_expr_maybe_real_minus_zero_p, tree_call_nonnegative_warnv_p):
9010 2022-10-31 Jakub Jelinek <jakub@redhat.com>
9012 * builtin-types.def (BT_FN_BFLOAT16_BFLOAT16_BFLOAT16): New.
9013 * builtins.def (BUILT_IN_NEXTAFTERF16B): New builtin.
9014 * fold-const-call.cc (fold_const_call_sss): Handle
9015 CFN_BUILT_IN_NEXTAFTERF16B.
9017 2022-10-31 Jakub Jelinek <jakub@redhat.com>
9019 * builtin-types.def (BT_FLOAT16_PTR, BT_FLOAT32_PTR, BT_FLOAT64_PTR,
9020 BT_FLOAT128_PTR, BT_FLOAT32X_PTR, BT_FLOAT64X_PTR, BT_FLOAT128X_PTR):
9021 New DEF_PRIMITIVE_TYPE.
9022 (BT_FN_INT_FLOAT16, BT_FN_INT_FLOAT32, BT_FN_INT_FLOAT64,
9023 BT_FN_INT_FLOAT128, BT_FN_INT_FLOAT32X, BT_FN_INT_FLOAT64X,
9024 BT_FN_INT_FLOAT128X, BT_FN_LONG_FLOAT16, BT_FN_LONG_FLOAT32,
9025 BT_FN_LONG_FLOAT64, BT_FN_LONG_FLOAT128, BT_FN_LONG_FLOAT32X,
9026 BT_FN_LONG_FLOAT64X, BT_FN_LONG_FLOAT128X, BT_FN_LONGLONG_FLOAT16,
9027 BT_FN_LONGLONG_FLOAT32, BT_FN_LONGLONG_FLOAT64,
9028 BT_FN_LONGLONG_FLOAT128, BT_FN_LONGLONG_FLOAT32X,
9029 BT_FN_LONGLONG_FLOAT64X, BT_FN_LONGLONG_FLOAT128X): New
9030 DEF_FUNCTION_TYPE_1.
9031 (BT_FN_FLOAT16_FLOAT16_FLOAT16PTR, BT_FN_FLOAT32_FLOAT32_FLOAT32PTR,
9032 BT_FN_FLOAT64_FLOAT64_FLOAT64PTR, BT_FN_FLOAT128_FLOAT128_FLOAT128PTR,
9033 BT_FN_FLOAT32X_FLOAT32X_FLOAT32XPTR,
9034 BT_FN_FLOAT64X_FLOAT64X_FLOAT64XPTR,
9035 BT_FN_FLOAT128X_FLOAT128X_FLOAT128XPTR, BT_FN_FLOAT16_FLOAT16_INT,
9036 BT_FN_FLOAT32_FLOAT32_INT, BT_FN_FLOAT64_FLOAT64_INT,
9037 BT_FN_FLOAT128_FLOAT128_INT, BT_FN_FLOAT32X_FLOAT32X_INT,
9038 BT_FN_FLOAT64X_FLOAT64X_INT, BT_FN_FLOAT128X_FLOAT128X_INT,
9039 BT_FN_FLOAT16_FLOAT16_INTPTR, BT_FN_FLOAT32_FLOAT32_INTPTR,
9040 BT_FN_FLOAT64_FLOAT64_INTPTR, BT_FN_FLOAT128_FLOAT128_INTPTR,
9041 BT_FN_FLOAT32X_FLOAT32X_INTPTR, BT_FN_FLOAT64X_FLOAT64X_INTPTR,
9042 BT_FN_FLOAT128X_FLOAT128X_INTPTR, BT_FN_FLOAT16_FLOAT16_LONG,
9043 BT_FN_FLOAT32_FLOAT32_LONG, BT_FN_FLOAT64_FLOAT64_LONG,
9044 BT_FN_FLOAT128_FLOAT128_LONG, BT_FN_FLOAT32X_FLOAT32X_LONG,
9045 BT_FN_FLOAT64X_FLOAT64X_LONG, BT_FN_FLOAT128X_FLOAT128X_LONG): New
9046 DEF_FUNCTION_TYPE_2.
9047 (BT_FN_FLOAT16_FLOAT16_FLOAT16_INTPTR,
9048 BT_FN_FLOAT32_FLOAT32_FLOAT32_INTPTR,
9049 BT_FN_FLOAT64_FLOAT64_FLOAT64_INTPTR,
9050 BT_FN_FLOAT128_FLOAT128_FLOAT128_INTPTR,
9051 BT_FN_FLOAT32X_FLOAT32X_FLOAT32X_INTPTR,
9052 BT_FN_FLOAT64X_FLOAT64X_FLOAT64X_INTPTR,
9053 BT_FN_FLOAT128X_FLOAT128X_FLOAT128X_INTPTR): New DEF_FUNCTION_TYPE_3.
9054 * builtins.def (ACOSH_TYPE, ATAN2_TYPE, ATANH_TYPE, COSH_TYPE,
9055 FDIM_TYPE, HUGE_VAL_TYPE, HYPOT_TYPE, ILOGB_TYPE, LDEXP_TYPE,
9056 LGAMMA_TYPE, LLRINT_TYPE, LOG10_TYPE, LRINT_TYPE, MODF_TYPE,
9057 NEXTAFTER_TYPE, REMQUO_TYPE, SCALBLN_TYPE, SCALBN_TYPE, SINH_TYPE):
9058 Define and undefine later.
9059 (FMIN_TYPE, SQRT_TYPE): Undefine at a later line.
9060 (INF_TYPE): Define at a later line.
9061 (BUILT_IN_ACOSH, BUILT_IN_ACOS, BUILT_IN_ASINH, BUILT_IN_ASIN,
9062 BUILT_IN_ATAN2, BUILT_IN_ATANH, BUILT_IN_ATAN, BUILT_IN_CBRT,
9063 BUILT_IN_COSH, BUILT_IN_COS, BUILT_IN_ERFC, BUILT_IN_ERF,
9064 BUILT_IN_EXP2, BUILT_IN_EXP, BUILT_IN_EXPM1, BUILT_IN_FDIM,
9065 BUILT_IN_FMOD, BUILT_IN_FREXP, BUILT_IN_HYPOT, BUILT_IN_ILOGB,
9066 BUILT_IN_LDEXP, BUILT_IN_LGAMMA, BUILT_IN_LLRINT, BUILT_IN_LLROUND,
9067 BUILT_IN_LOG10, BUILT_IN_LOG1P, BUILT_IN_LOG2, BUILT_IN_LOGB,
9068 BUILT_IN_LOG, BUILT_IN_LRINT, BUILT_IN_LROUND, BUILT_IN_MODF,
9069 BUILT_IN_NEXTAFTER, BUILT_IN_POW, BUILT_IN_REMAINDER, BUILT_IN_REMQUO,
9070 BUILT_IN_SCALBLN, BUILT_IN_SCALBN, BUILT_IN_SINH, BUILT_IN_SIN,
9071 BUILT_IN_TANH, BUILT_IN_TAN, BUILT_IN_TGAMMA): Add
9072 DEF_EXT_LIB_FLOATN_NX_BUILTINS.
9073 (BUILT_IN_HUGE_VAL): Use HUGE_VAL_TYPE instead of INF_TYPE in
9074 DEF_GCC_FLOATN_NX_BUILTINS.
9075 * fold-const-call.cc (fold_const_call_ss): Add various CASE_CFN_*_FN:
9076 cases when CASE_CFN_* is present.
9077 (fold_const_call_sss): Likewise.
9078 * builtins.cc (mathfn_built_in_2): Use CASE_MATHFN_FLOATN instead of
9079 CASE_MATHFN for various builtins in SEQ_OF_CASE_MATHFN macro.
9080 (builtin_with_linkage_p): Add CASE_FLT_FN_FLOATN_NX for various
9081 builtins next to CASE_FLT_FN.
9082 * fold-const.cc (tree_call_nonnegative_warnv_p): Add CASE_CFN_*_FN:
9083 next to CASE_CFN_*: for various builtins.
9084 * tree-call-cdce.cc (can_test_argument_range): Add
9085 CASE_FLT_FN_FLOATN_NX next to CASE_FLT_FN for various builtins.
9086 (edom_only_function): Likewise.
9088 2022-10-31 konglin1 <lingling.kong@intel.com>
9090 * common/config/i386/i386-common.cc
9091 (OPTION_MASK_ISA2_AVXNECONVERT_SET,
9092 OPTION_MASK_ISA2_AVXNECONVERT_UNSET): New.
9093 (ix86_handle_option): Handle -mavxneconvert, unset
9094 avxneconvert when avx2 is disabled.
9095 * common/config/i386/i386-cpuinfo.h (processor_types): Add
9096 FEATURE_AVXNECONVERT.
9097 * common/config/i386/i386-isas.h: Add ISA_NAME_TABLE_ENTRY for
9099 * common/config/i386/cpuinfo.h (get_available_features):
9100 Detect avxneconvert.
9101 * config.gcc: Add avxneconvertintrin.h
9102 * config/i386/avxneconvertintrin.h: New.
9103 * config/i386/avx512bf16vlintrin.h (_mm256_cvtneps_pbh):
9104 Unified builtin with avxneconvert.
9105 (_mm_cvtneps_pbh): Ditto.
9106 * config/i386/cpuid.h (bit_AVXNECONVERT): New.
9107 * config/i386/i386-builtin-types.def: Add
9108 DEF_POINTER_TYPE (PCV8HF, V8HF, CONST),
9109 DEF_POINTER_TYPE (PCV8BF, V8BF, CONST),
9110 DEF_POINTER_TYPE (PCV16HF, V16HF, CONST),
9111 DEF_POINTER_TYPE (PCV16BF, V16BF, CONST),
9112 DEF_FUNCTION_TYPE (V4SF, PCBFLOAT16),
9113 DEF_FUNCTION_TYPE (V4SF, PCFLOAT16),
9114 DEF_FUNCTION_TYPE (V8SF, PCBFLOAT16),
9115 DEF_FUNCTION_TYPE (V8SF, PCFLOAT16),
9116 DEF_FUNCTION_TYPE (V4SF, PCV8BF),
9117 DEF_FUNCTION_TYPE (V4SF, PCV8HF),
9118 DEF_FUNCTION_TYPE (V8SF, PCV16HF),
9119 DEF_FUNCTION_TYPE (V8SF, PCV16BF),
9120 * config/i386/i386-builtin.def: Add new builtins.
9121 * config/i386/i386-c.cc (ix86_target_macros_internal): Define
9123 * config/i386/i386-expand.cc (ix86_expand_special_args_builtin):
9124 Handle V4SF_FTYPE_PCBFLOAT16,V8SF_FTYPE_PCBFLOAT16, V4SF_FTYPE_PCFLOAT16,
9125 V8SF_FTYPE_PCFLOAT16,V4SF_FTYPE_PCV8BF,
9126 V4SF_FTYPE_PCV8HF,V8SF_FTYPE_PCV16BF,V8SF_FTYPE_PCV16HF.
9127 * config/i386/i386-isa.def : Add DEF_PTA(AVXNECONVERT) New.
9128 * config/i386/i386-options.cc (isa2_opts): Add -mavxneconvert.
9129 (ix86_valid_target_attribute_inner_p): Handle avxneconvert.
9130 * config/i386/i386.md: Add attr avx512bf16vl and avxneconvert.
9131 * config/i386/i386.opt: Add option -mavxneconvert.
9132 * config/i386/immintrin.h: Inculde avxneconvertintrin.h.
9133 * config/i386/sse.md (vbcstnebf162ps_<mode>): New define_insn.
9134 (vbcstnesh2ps_<mode>): Ditto.
9135 (vcvtnee<bf16_ph>2ps_<mode>):Ditto.
9136 (vcvtneo<bf16_ph>2ps_<mode>):Ditto.
9137 (vcvtneps2bf16_v4sf): Ditto.
9138 (*vcvtneps2bf16_v4sf): Ditto.
9139 (vcvtneps2bf16_v8sf): Ditto.
9140 * doc/invoke.texi: Document -mavxneconvert.
9141 * doc/extend.texi: Document avxneconvert.
9142 * doc/sourcebuild.texi: Document target avxneconvert.
9144 2022-10-31 konglin1 <lingling.kong@intel.com>
9146 * config/i386/avx512bf16intrin.h (__attribute__): Change short to bf16.
9147 (_mm_cvtsbh_ss): Ditto.
9148 (_mm512_cvtne2ps_pbh): Ditto.
9149 (_mm512_mask_cvtne2ps_pbh): Ditto.
9150 (_mm512_maskz_cvtne2ps_pbh): Ditto.
9151 * config/i386/avx512bf16vlintrin.h (__attribute__): Ditto.
9152 (_mm256_cvtne2ps_pbh): Ditto.
9153 (_mm256_mask_cvtne2ps_pbh): Ditto.
9154 (_mm256_maskz_cvtne2ps_pbh): Ditto.
9155 (_mm_cvtne2ps_pbh): Ditto.
9156 (_mm_mask_cvtne2ps_pbh): Ditto.
9157 (_mm_maskz_cvtne2ps_pbh): Ditto.
9158 (_mm_cvtness_sbh): Ditto.
9159 * config/i386/i386-builtin-types.def (V8BF): Add new
9160 DEF_VECTOR_TYPE for BFmode.
9163 * config/i386/i386-builtin.def (BDESC): Fixed builtins.
9164 * config/i386/i386-expand.cc (ix86_expand_args_builtin): Changed
9165 avx512bf16 ix86_builtin_func_type included HI to BF.
9166 * config/i386/immintrin.h: Add SSE2 depend for avx512bf16.
9167 * config/i386/sse.md (TARGET_AVX512VL): Changed HI vector to BF
9169 (avx512f_cvtneps2bf16_v4sf): New define_expand.
9170 (*avx512f_cvtneps2bf16_v4sf): New define_insn.
9171 (avx512f_cvtneps2bf16_v4sf_maskz):Ditto.
9172 (avx512f_cvtneps2bf16_v4sf_mask): Ditto.
9173 (avx512f_cvtneps2bf16_v4sf_mask_1): Ditto.
9175 2022-10-31 liuhongt <hongtao.liu@intel.com>
9178 * config/i386/i386-modes.def (VECTOR_MODE): Support V2BFmode.
9179 * config/i386/i386.cc (classify_argument): Handle V4BFmode and
9181 (ix86_convert_const_vector_to_integer): Ditto.
9182 * config/i386/i386.h (VALID_AVX512FP16_REG_MODE): Remove
9184 (VALID_SSE2_REG_MODE): Add V4BFmode and V2BFmode.
9185 (VALID_MMX_REG_MODE): Add V4BFmode.
9186 * config/i386/i386.md (mode): Add V4BF and V2BF.
9188 * config/i386/mmx.md (MMXMODE) Add V4BF.
9190 (V_16_32_64): Add V4BF and V2BF.
9191 (mmxinsnmode): Add V4BF and V2BF.
9192 (*mov<mode>_internal): Hanlde V4BFmode and V2BFmode.
9194 2022-10-29 Eric Botcazou <ebotcazou@adacore.com>
9196 * configure.ac (sjlj-exceptions): Restore dropped line.
9197 * configure: Regenerate.
9199 2022-10-29 Eric Botcazou <ebotcazou@adacore.com>
9201 * alias.cc (init_alias_analysis): Do not record sets to the hard
9202 frame pointer if the frame pointer has not been eliminated.
9204 2022-10-29 Iain Buclaw <ibuclaw@gdcproject.org>
9206 * config/darwin-d.cc (TARGET_D_MINFO_START_NAME): Rename to ...
9207 (TARGET_D_MINFO_SECTION_START): ...this.
9208 (TARGET_D_MINFO_END_NAME): Rename to ...
9209 (TARGET_D_MINFO_SECTION_END): ... this.
9210 * config/elfos.h (TARGET_D_MINFO_SECTION): Remove.
9211 (TARGET_D_MINFO_START_NAME): Remove.
9212 (TARGET_D_MINFO_END_NAME): Remove.
9213 * config/i386/cygwin-d.cc (TARGET_D_MINFO_SECTION): Remove.
9214 (TARGET_D_MINFO_START_NAME): Remove.
9215 (TARGET_D_MINFO_END_NAME): Remove.
9216 * config/i386/winnt-d.cc (TARGET_D_MINFO_SECTION): Remove.
9217 (TARGET_D_MINFO_START_NAME): Remove.
9218 (TARGET_D_MINFO_END_NAME): Remove.
9219 * doc/tm.texi: Regenerate.
9220 * doc/tm.texi.in (TARGET_D_MINFO_START_NAME): Rename to ...
9221 (TARGET_D_MINFO_SECTION_START): ...this.
9222 (TARGET_D_MINFO_END_NAME): Rename to ...
9223 (TARGET_D_MINFO_SECTION_END): ...this.
9225 2022-10-29 Iain Buclaw <ibuclaw@gdcproject.org>
9227 * config.gcc: Split out glibc-d.o into linux-d.o, kfreebsd-d.o,
9228 kopensolaris-d.o, and gnu-d.o. Split out cygwin-d.o from winnt-d.o.
9229 * config/arm/linux-eabi.h (EXTRA_TARGET_D_OS_VERSIONS): Remove.
9230 * config/gnu.h (GNU_USER_TARGET_D_OS_VERSIONS): Remove.
9231 * config/i386/cygwin.h (EXTRA_TARGET_D_OS_VERSIONS): Remove.
9232 * config/i386/linux-common.h (EXTRA_TARGET_D_OS_VERSIONS): Remove.
9233 * config/i386/mingw32.h (EXTRA_TARGET_D_OS_VERSIONS): Remove.
9234 * config/i386/t-cygming: Add cygwin-d.o.
9235 * config/i386/winnt-d.cc (winnt_d_os_builtins): Only add
9236 MinGW-specific version condition.
9237 * config/kfreebsd-gnu.h (GNU_USER_TARGET_D_OS_VERSIONS): Remove.
9238 * config/kopensolaris-gnu.h (GNU_USER_TARGET_D_OS_VERSIONS): Remove.
9239 * config/linux-android.h (ANDROID_TARGET_D_OS_VERSIONS): Remove.
9240 * config/linux.h (GNU_USER_TARGET_D_OS_VERSIONS): Remove.
9241 * config/mips/linux-common.h (EXTRA_TARGET_D_OS_VERSIONS): Remove.
9242 * config/t-glibc: Remove glibc-d.o, add gnu-d.o, kfreebsd-d.o,
9244 * config/t-linux: Add linux-d.o.
9245 * config/glibc-d.cc: Remove file.
9246 * config/gnu-d.cc: New file.
9247 * config/i386/cygwin-d.cc: New file.
9248 * config/kfreebsd-d.cc: New file.
9249 * config/kopensolaris-d.cc: New file.
9250 * config/linux-d.cc: New file.
9252 2022-10-29 Jeff Law <jeffreyalaw@gmail.com>
9254 * config/h8300/h8300.cc (pre_incdec_with_reg): Make reg argument
9256 * config/h8300/h8300-protos.h (pre_incdec_with_reg): Adjust prototype.
9258 2022-10-28 Joseph Myers <joseph@codesourcery.com>
9260 * config/aarch64/aarch64.cc (aarch64_setup_incoming_varargs):
9261 Check TYPE_NO_NAMED_ARGS_STDARG_P.
9262 * config/alpha/alpha.cc (alpha_setup_incoming_varargs): Likewise.
9263 * config/arc/arc.cc (arc_setup_incoming_varargs): Likewise.
9264 * config/arm/arm.cc (arm_setup_incoming_varargs): Likewise.
9265 * config/csky/csky.cc (csky_setup_incoming_varargs): Likewise.
9266 * config/epiphany/epiphany.cc (epiphany_setup_incoming_varargs):
9268 * config/fr30/fr30.cc (fr30_setup_incoming_varargs): Likewise.
9269 * config/frv/frv.cc (frv_setup_incoming_varargs): Likewise.
9270 * config/ft32/ft32.cc (ft32_setup_incoming_varargs): Likewise.
9271 * config/i386/i386.cc (ix86_setup_incoming_varargs): Likewise.
9272 * config/ia64/ia64.cc (ia64_setup_incoming_varargs): Likewise.
9273 * config/loongarch/loongarch.cc
9274 (loongarch_setup_incoming_varargs): Likewise.
9275 * config/m32r/m32r.cc (m32r_setup_incoming_varargs): Likewise.
9276 * config/mcore/mcore.cc (mcore_setup_incoming_varargs): Likewise.
9277 * config/mips/mips.cc (mips_setup_incoming_varargs): Likewise.
9278 * config/mmix/mmix.cc (mmix_setup_incoming_varargs): Likewise.
9279 * config/nds32/nds32.cc (nds32_setup_incoming_varargs): Likewise.
9280 * config/nios2/nios2.cc (nios2_setup_incoming_varargs): Likewise.
9281 * config/riscv/riscv.cc (riscv_setup_incoming_varargs): Likewise.
9282 * config/rs6000/rs6000-call.cc (setup_incoming_varargs): Likewise.
9283 * config/sh/sh.cc (sh_setup_incoming_varargs): Likewise.
9284 * config/visium/visium.cc (visium_setup_incoming_varargs):
9286 * config/vms/vms-c.cc (vms_c_common_override_options): Do not set
9287 flag_allow_parameterless_variadic_functions.
9288 * doc/invoke.texi (-fallow-parameterless-variadic-functions): Do
9289 not document option.
9290 * function.cc (assign_parms): Call assign_parms_setup_varargs for
9291 TYPE_NO_NAMED_ARGS_STDARG_P case.
9292 * ginclude/stdarg.h [__STDC_VERSION__ > 201710L] (va_start): Make
9293 variadic macro. Pass second argument of 0 to __builtin_va_start.
9294 * target.def (setup_incoming_varargs): Update documentation.
9295 * doc/tm.texi: Regenerate.
9296 * tree-core.h (struct tree_type_common): Add
9297 no_named_args_stdarg_p.
9298 * tree-streamer-in.cc (unpack_ts_type_common_value_fields): Unpack
9299 TYPE_NO_NAMED_ARGS_STDARG_P.
9300 * tree-streamer-out.cc (pack_ts_type_common_value_fields): Pack
9301 TYPE_NO_NAMED_ARGS_STDARG_P.
9302 * tree.cc (type_cache_hasher::equal): Compare
9303 TYPE_NO_NAMED_ARGS_STDARG_P.
9304 (build_function_type): Add argument no_named_args_stdarg_p.
9305 (build_function_type_list_1, build_function_type_array_1)
9306 (reconstruct_complex_type): Update calls to build_function_type.
9307 (stdarg_p, prototype_p): Return true for (...) functions.
9308 (gimple_canonical_types_compatible_p): Compare
9309 TYPE_NO_NAMED_ARGS_STDARG_P.
9310 * tree.h (TYPE_NO_NAMED_ARGS_STDARG_P): New.
9311 (build_function_type): Update prototype.
9313 2022-10-28 Andre Vieira <andre.simoesdiasvieira@arm.com>
9315 PR tree-optimization/107346
9316 * tree-vect-data-refs.cc (vect_check_gather_scatter): Reject offsets
9317 that aren't multiples of BITS_PER_UNIT.
9319 2022-10-28 Richard Biener <rguenther@suse.de>
9321 PR tree-optimization/107407
9322 * tree-ssa-dse.cc (dse_classify_store): Perform backedge
9323 varying index check when collecting PHI uses rather than
9324 after optimizing processing of the candidate defs.
9326 2022-10-28 Richard Biener <rguenther@suse.de>
9328 PR tree-optimization/107447
9329 * tree-ssa-loop-im.cc (determine_max_movement): Do not
9330 hoist returns-twice calls.
9332 2022-10-28 Richard Biener <rguenther@suse.de>
9334 PR tree-optimization/107435
9335 * tree-vect-loop.cc (vectorizable_recurr): Convert initial
9336 value to vector component type.
9338 2022-10-28 Julian Brown <julian@codesourcery.com>
9339 Thomas Schwinge <thomas@codesourcery.com>
9342 * omp-low.cc (oacc_privatization_candidate_p): Artificial vars are not
9343 privatization candidates.
9345 2022-10-28 Martin Liska <mliska@suse.cz>
9348 * doc/invoke.texi: Document sanitizers can trigger warnings.
9350 2022-10-28 Thomas Schwinge <thomas@codesourcery.com>
9352 * doc/makefile.texi (Makefile Targets): Document
9353 'distclean-stage[N]'.
9355 2022-10-28 Aldy Hernandez <aldyh@redhat.com>
9357 * value-range.cc (range_tests_floats): Use HONOR_INFINITIES.
9359 2022-10-27 Eric Botcazou <ebotcazou@adacore.com>
9361 * config/aarch64/aarch64.h (DONT_USE_BUILTIN_SETJMP): Delete.
9363 2022-10-27 H.J. Lu <hjl.tools@gmail.com>
9366 * config/i386/i386.md (UNSPEC_CC_NE): New.
9367 Replace ne:CCC/ne:CCO with UNSPEC_CC_NE in neg patterns.
9369 2022-10-27 Andrew Pinski <apinski@marvell.com>
9371 * tree-ssa-phiopt.cc: Include tree-ssa-dce.h
9372 (replace_phi_edge_with_variable):
9373 New argument, dce_ssa_names. Call simple_dce_from_worklist.
9374 (match_simplify_replacement): If we inserted a sequence,
9375 mark the lhs of the new sequence to be possible dce.
9376 Always move the statement and mark the lhs (if it is a name)
9377 as possible to remove.
9379 2022-10-27 Richard Sandiford <richard.sandiford@arm.com>
9381 * config/aarch64/aarch64-protos.h: Replace constexpr with
9383 * config/aarch64/aarch64-sve-builtins-base.cc: Likewise.
9384 * config/aarch64/aarch64-sve-builtins-functions.h: Likewise.
9385 * config/aarch64/aarch64-sve-builtins-shapes.cc: Likewise.
9386 * config/aarch64/aarch64-sve-builtins-sve2.cc: Likewise.
9387 * config/aarch64/aarch64-sve-builtins.cc: Likewise.
9388 * config/aarch64/aarch64.cc: Likewise.
9389 * config/aarch64/driver-aarch64.cc: Likewise
9391 2022-10-27 Aldy Hernandez <aldyh@redhat.com>
9393 PR tree-optimization/107394
9394 * value-range-storage.cc (frange_storage_slot::get_frange): Use
9397 2022-10-27 Thomas Schwinge <thomas@codesourcery.com>
9399 * optc-save-gen.awk: Clarify 'Init' option property usage for
9400 streaming optimization.
9402 2022-10-27 Torbjörn SVENSSON <torbjorn.svensson@foss.st.com>
9403 Yvan ROUX <yvan.roux@foss.st.com>
9405 * ira.cc: Resize array after reg number increased.
9407 2022-10-27 Jiawei <jiawei@iscas.ac.cn>
9408 Sinan Lin <sinan@isrc.iscas.ac.cn>
9410 * config/riscv/constraints.md (TARGET_ZFINX ? GR_REGS): Set GPRS
9411 use while Zfinx is enable.
9412 * config/riscv/riscv.cc (riscv_hard_regno_mode_ok): Limit odd
9413 registers use when Zdinx enable in RV32 cases.
9414 (riscv_option_override): New target enable MASK_FDIV.
9415 (riscv_libgcc_floating_mode_supported_p): New error info when
9416 use incompatible arch&abi.
9417 (riscv_excess_precision): New target enable FLOAT16.
9419 2022-10-27 Jiawei <jiawei@iscas.ac.cn>
9421 * config/riscv/iterators.md (TARGET_ZFINX):New target.
9422 (TARGET_ZDINX): Ditto.
9423 (TARGET_ZHINX): Ditto.
9424 * config/riscv/riscv-builtins.cc (AVAIL): Ditto.
9425 (riscv_atomic_assign_expand_fenv): Ditto.
9426 * config/riscv/riscv-c.cc (riscv_cpu_cpp_builtins): Ditto.
9427 * config/riscv/riscv.md: Ditto.
9429 2022-10-27 Jiawei <jiawei@iscas.ac.cn>
9430 Sinan Lin <sinan@isrc.iscas.ac.cn>
9432 * common/config/riscv/riscv-common.cc: New extensions.
9433 * config/riscv/arch-canonicalize: New imply relations.
9434 * config/riscv/riscv-opts.h (MASK_ZFINX): New mask.
9435 (MASK_ZDINX): Ditto.
9436 (MASK_ZHINX): Ditto.
9437 (MASK_ZHINXMIN): Ditto.
9438 (TARGET_ZFINX): New target.
9439 (TARGET_ZDINX): Ditto.
9440 (TARGET_ZHINX): Ditto.
9441 (TARGET_ZHINXMIN): Ditto.
9442 * config/riscv/riscv.opt: New target variable.
9444 2022-10-26 David Faust <david.faust@oracle.com>
9446 * config/bpf/bpf.cc: Support __builtin_preserve_field_info.
9447 (enum bpf_builtins): Add new builtin.
9448 (bpf_init_builtins): Likewise.
9449 (bpf_core_field_info): New function.
9450 (bpf_expand_builtin): Accomodate new builtin. Refactor adding new
9452 (maybe_make_core_relo): ... here. New function.
9453 (bpf_resolve_overloaded_builtin): Accomodate new builtin.
9454 (bpf_core_newdecl): Likewise.
9455 (bpf_core_walk): Likewise.
9456 (bpf_core_is_maybe_aggregate_access): Improve logic.
9457 (struct core_walk_data): New.
9458 * config/bpf/coreout.cc (bpf_core_reloc_add): Allow adding different
9460 * config/bpf/coreout.h: Analogous change.
9461 * doc/extend.texi: Document BPF __builtin_preserve_field_info.
9463 2022-10-26 Marek Polacek <polacek@redhat.com>
9466 * doc/invoke.texi: Document -Wdangling-reference.
9468 2022-10-26 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
9470 * config/xtensa/xtensa.md (movdi):
9471 Copy operands[0...1] to ops[0...3] and then use the latter before
9472 calling xtensa_split_DI_reg_imm() and emitting insns.
9474 2022-10-26 Alexander Monakov <amonakov@ispras.ru>
9477 * ipa-visibility.cc (function_and_variable_visibility):
9478 Conditionally upgrade TLS model instead of asserting.
9480 2022-10-26 Andrew MacLeod <amacleod@redhat.com>
9482 * gimple-range-fold.cc (fold_using_range::fold_stmt): Check if
9483 stmt is non-negative and adjust the range.
9485 2022-10-26 Martin Liska <mliska@suse.cz>
9487 * common/config/i386/cpuinfo.h (has_cpu_feature): Add comment.
9488 (reset_cpu_feature): New.
9489 (get_zhaoxin_cpu): Use reset_cpu_feature.
9491 2022-10-26 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
9493 * config/riscv/riscv.cc (riscv_expand_epilogue): Fix statement.
9495 2022-10-26 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
9498 * config/riscv/riscv-modes.def (VECTOR_BOOL_MODE): Set to minimum size.
9499 (ADJUST_NUNITS): Adjust according to -march.
9500 (ADJUST_BYTESIZE): Ditto.
9501 * config/riscv/riscv-protos.h (riscv_v_ext_enabled_vector_mode_p):
9503 (riscv_v_ext_vector_mode_p): Change function implementation.
9504 * config/riscv/riscv-vector-builtins.cc (rvv_switcher::rvv_switcher):
9505 Change to riscv_v_ext_vector_mode_p.
9506 (register_builtin_type): Ditto.
9507 * config/riscv/riscv.cc (riscv_v_ext_vector_mode_p): Change to enabled
9510 (riscv_v_ext_enabled_vector_mode_p): Remove.
9511 (riscv_v_adjust_nunits): New function.
9512 (riscv_vector_mode_supported_p): Use riscv_v_ext_vector_mode_p instead.
9513 * config/riscv/riscv.h (riscv_v_adjust_nunits): New function.
9515 2022-10-26 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
9517 * config.gcc (riscv*): Add riscv-v.o to extra_objs.
9518 * config/riscv/constraints.md (vu): New constraint.
9522 * config/riscv/predicates.md (vector_length_operand): New.
9523 (reg_or_mem_operand): Ditto.
9524 (vector_move_operand): Ditto.
9525 (vector_mask_operand): Ditto.
9526 (vector_merge_operand): Ditto.
9527 * config/riscv/riscv-protos.h (riscv_regmode_natural_size) New.
9528 (riscv_vector::const_vec_all_same_in_range_p): Ditto.
9529 (riscv_vector::legitimize_move): Ditto.
9530 (tail_policy): Ditto.
9531 (mask_policy): Ditto.
9532 * config/riscv/riscv-v.cc: New.
9533 * config/riscv/riscv-vector-builtins-bases.cc
9534 (vsetvl::expand): Refactor how LMUL encoding.
9535 * config/riscv/riscv.cc (riscv_print_operand): Update how LMUL
9536 print and mask operand print.
9537 (riscv_regmode_natural_size): New.
9538 * config/riscv/riscv.h (REGMODE_NATURAL_SIZE): New.
9539 * config/riscv/riscv.md (mode): Add vector modes.
9540 * config/riscv/t-riscv (riscv-v.o) New.
9541 * config/riscv/vector-iterators.md: New.
9542 * config/riscv/vector.md (vundefined<mode>): New.
9545 (@vsetvl<mode>_no_side_effects): New.
9546 (@pred_mov<mode>): New.
9548 2022-10-26 Monk Chiang <monk.chiang@sifive.com>
9550 * common/config/riscv/riscv-common.cc (riscv_ext_version_table):
9551 Add svinval and svnapot extension.
9552 (riscv_ext_flag_table): Ditto.
9553 * config/riscv/riscv-opts.h (MASK_SVINVAL): New.
9554 (MASK_SVNAPOT): Ditto.
9555 (TARGET_SVINVAL): Ditto.
9556 (TARGET_SVNAPOT): Ditto.
9557 * config/riscv/riscv.opt (riscv_sv_subext): New.
9559 2022-10-26 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
9561 * config/riscv/riscv-modes.def: Adjust table indentation in commnet.
9563 2022-10-26 Martin Liska <mliska@suse.cz>
9565 * configure: Regenerate.
9567 2022-10-26 Aldy Hernandez <aldyh@redhat.com>
9569 * value-range.cc (frange::set): Use HONOR_*.
9570 (frange::verify_range): Same.
9571 * value-range.h (frange_val_min): Same.
9572 (frange_val_max): Same.
9574 2022-10-26 Jiufu Guo <guojiufu@linux.ibm.com>
9577 * config/rs6000/rs6000.cc (rs6000_cannot_force_const_mem): Return true
9580 2022-10-26 Kito Cheng <kito.cheng@sifive.com>
9582 * common/config/riscv/riscv-common.cc (riscv_ext_version_table):
9584 (riscv_supported_std_ext): Ditto.
9585 (multi_letter_subset_rank): Remove `h`.
9586 (riscv_subset_list::parse_std_ext): Handle `h` as single letter
9588 (riscv_subset_list::parse): Ditto.
9590 2022-10-25 Eugene Rozenfeld <erozen@microsoft.com>
9592 * auto-profile.cc (get_combined_location): Include discriminator in the
9593 returned combined location.
9594 (read_function_instance): Read discriminators from profiles.
9596 2022-10-25 H.J. Lu <hjl.tools@gmail.com>
9599 * expr.cc (get_inner_reference): Always use TYPE_MODE for vector
9600 field with vector raw mode.
9602 2022-10-25 Segher Boessenkool <segher@kernel.crashing.org>
9604 * config/rs6000/rs6000.md (CCEITHER): Delete.
9607 (isel_<un>signed_<GPR:mode>): Rename to...
9608 (isel_<CCANY:mode>_<GPR:mode>): ... this. Adjust.
9609 (*isel_reversed_<un>signed_<GPR:mode>): Rename to...
9610 (*isel_reversed_<CCANY:mode>_<GPR:mode>): ... this. Adjust.
9611 (setbc_<un>signed_<GPR:mode>): Rename to...
9612 (setbc_<CCANY:mode>_<GPR:mode>C): ... this. Adjust."
9613 (*setbcr_<un>signed_<GPR:mode>): Rename to ...
9614 (*setbcr_<CCANY:mode>_<GPR:mode>): ... this. Adjust.
9615 (*setnbc_<un>signed_<GPR:mode>): Rename to ...
9616 (*setnbc_<CCANY:mode>_<GPR:mode>): ... this. Adjust.
9617 (*setnbcr_<un>signed_<GPR:mode>): Rename to ...
9618 (*setnbcr_<CCANY:mode>_<GPR:mode>): ... this. Adjust.
9619 (eq<mode>3 for GPR): Adjust.
9620 (ne<mode>3 for GPR): Adjust.
9621 * config/rs6000/rs6000-string.cc (do_isel): Adjust.
9622 * config/rs6000/rs6000.cc (rs6000_emit_int_cmove): Adjust.
9624 2022-10-25 Richard Biener <rguenther@suse.de>
9626 PR tree-optimization/107176
9627 PR tree-optimization/66375
9628 PR tree-optimization/42512
9629 * tree-scalar-evolution.cc (follow_ssa_edge_expr): Revert
9630 the PR66375 fix, do not not associate PLUS_EXPR to be able
9631 to use tail-recursion.
9632 (follow_ssa_edge_binary): Likewise.
9633 (interpret_loop_phi): Revert PR42512 fix, do not throw
9634 away analyze_evolution_in_loop result after the fact.
9635 (follow_ssa_edge_expr): When reaching halting_phi initalize
9636 the evolution to the symbolic value of the PHI result.
9637 (add_to_evolution_1): When adding the first evolution verify
9638 we can handle the expression wrapping the symbolic evolution
9639 and replace that in full using the initial condition.
9640 (class scev_dfs): New, contains ...
9641 (follow_ssa_edge_expr, follow_ssa_edge_binary,
9642 follow_ssa_edge_in_condition_phi_branch,
9643 follow_ssa_edge_in_condition_phi,
9644 follow_ssa_edge_inner_loop_phi,
9645 add_to_evolution, add_to_evolution_1): ... these with
9646 loop and halting_phi arguments in class data.
9647 (scev_dfs::get_ev): New toplevel DFS entry, start with
9648 a chrec_dont_know evolution.
9649 (analyze_evolution_in_loop): Use scev_dfs.
9651 2022-10-25 Eric Botcazou <ebotcazou@adacore.com>
9653 * profile.cc (branch_prob): Be prepared for ignored functions with
9654 DECL_SOURCE_LOCATION set to UNKNOWN_LOCATION.
9656 2022-10-25 Richard Biener <rguenther@suse.de>
9658 * tree-scalar-evolution.cc (follow_ssa_edge_expr): Move
9659 STRIP_USELESS_TYPE_CONVERSIONS to where it matters.
9661 2022-10-25 Tejas Joshi <TejasSanjay.Joshi@amd.com>
9663 * common/config/i386/i386-common.cc (processor_alias_table): Use
9664 CPU_ZNVER3 for znver4.
9665 * config/i386/znver.md: Remove znver4 reservations.
9667 2022-10-25 Jakub Jelinek <jakub@redhat.com>
9669 * gimplify.cc (gimple_boolify): Fix comment typos, prduce -> produce
9672 2022-10-25 Jakub Jelinek <jakub@redhat.com>
9674 PR tree-optimization/107368
9675 * gimplify.cc (gimplify_call_expr): For complex IFN_ASSUME
9676 conditions call gimple_boolify on the condition.
9678 2022-10-25 YunQiang Su <yunqiang.su@cipunited.com>
9680 * config.gcc: add -with-compact-branches=policy build option.
9681 * doc/install.texi: Likewise.
9682 * config/mips/mips.h: Likewise.
9684 2022-10-25 YunQiang Su <yunqiang.su@cipunited.com>
9686 * config/mips/mips.cc (mips_option_override): not trigger error
9687 for compact-branches=always for pre-R6.
9688 * config/mips/mips.h (TARGET_RTP_PIC): not trigger error for
9689 compact-branches=always for pre-R6.
9690 (TARGET_CB_NEVER): Likewise.
9691 (TARGET_CB_ALWAYS): Likewise.
9692 (struct mips_cpu_info): define macros for compact branch policy.
9693 * doc/invoke.texi: Document "always" with pre-R6.
9695 2022-10-25 Jakub Jelinek <jakub@redhat.com>
9697 PR tree-optimization/107369
9698 * gimplify.cc (gimplify_call_expr): If seen_error, handle complex
9699 IFN_ASSUME the same as for -O0.
9701 2022-10-25 YunQiang Su <yunqiang.su@cipunited.com>
9703 * configure.ac: AC_DEFINE(ENABLE_MULTIARCH, 1)
9704 * configure: Regenerated.
9705 * config.in: Regenerated.
9706 * config/mips/mips.h: don't define STANDARD_STARTFILE_PREFIX_1
9707 if ENABLE_MULTIARCH is defined.
9708 * config/mips/t-linux64: define correct multiarch path when
9709 multiarch is enabled.
9711 2022-10-25 Richard Biener <rguenther@suse.de>
9713 PR tree-optimization/100756
9714 * tree-ssa-loop-niter.cc (expand_simple_operations): Also
9715 expand multiplications by invariants.
9717 2022-10-25 Kewen Lin <linkw@linux.ibm.com>
9719 PR tree-optimization/107338
9720 * tree-vect-patterns.cc (vect_recog_bitfield_ref_pattern): Move
9721 shfit_n calculation before the adjustments for widening loads.
9723 2022-10-25 Martin Liska <mliska@suse.cz>
9725 * common/config/riscv/riscv-common.cc
9726 (riscv_get_valid_option_values): Get out of ifdef.
9728 2022-10-25 Martin Liska <mliska@suse.cz>
9731 * common/config/i386/i386-cpuinfo.h (enum processor_vendor):
9732 Fix pedantic warning.
9734 2022-10-24 Martin Liska <mliska@suse.cz>
9737 * diagnostic-format-sarif.cc
9738 (sarif_builder::maybe_make_physical_location_object): Gracefully
9739 reject locations with NULL filename.
9741 2022-10-24 David Malcolm <dmalcolm@redhat.com>
9744 * doc/invoke.texi (Static Analyzer Options): Add "pipe" and
9745 "pipe2" to the list of functions the analyzer has hardcoded
9748 2022-10-24 Jason Merrill <jason@redhat.com>
9750 * tree.h (build_string_literal): New one-argument overloads that
9751 take tree (identifier) and const char *.
9752 * builtins.cc (fold_builtin_FILE)
9753 (fold_builtin_FUNCTION)
9754 * gimplify.cc (gimple_add_init_for_auto_var)
9755 * vtable-verify.cc (verify_bb_vtables): Simplify calls.
9757 2022-10-24 Martin Liska <mliska@suse.cz>
9760 * common/config/i386/i386-cpuinfo.h (enum processor_vendor):
9761 Reorder enum values as BUILTIN_VENDOR_MAX should not point
9762 in the middle of the valid enum values.
9764 2022-10-24 Marek Polacek <polacek@redhat.com>
9767 * tree.cc (maybe_wrap_with_location): Don't create a location wrapper
9768 when the type is erroneous.
9770 2022-10-24 Wilco Dijkstra <wdijkstr@arm.com>
9773 * config/aarch64/aarch64.cc (aarch64_internal_mov_immediate)
9774 Add support for a bitmask immediate with 2 MOVKs.
9775 (aarch64_check_bitmask): New function after refactorization.
9776 (aarch64_bitmask_imm): Simplify replication of small modes.
9777 Split function into 64-bit only version for efficiency.
9778 (aarch64_move_imm): Move near other immediate functions.
9779 (aarch64_uimm12_shift): Likewise.
9780 (aarch64_clamp_to_uimm12_shift): Likewise.
9781 (aarch64_movk_shift): Likewise.
9782 (aarch64_replicate_bitmask_imm): Likewise.
9783 (aarch64_and_split_imm1): Likewise.
9784 (aarch64_and_split_imm2): Likewise.
9785 (aarch64_and_bitmask_imm): Likewise.
9786 (aarch64_movw_imm): Likewise.
9788 2022-10-24 Aldy Hernandez <aldyh@redhat.com>
9790 PR tree-optimization/107355
9791 * range-op-float.cc (foperator_abs::op1_range): Handle NAN.
9793 2022-10-24 Tobias Burnus <tobias@codesourcery.com>
9795 PR middle-end/107236
9796 * omp-expand.cc (expand_omp_target): Set calls_declare_variant_alt
9797 in DECL_CONTEXT and not to cfun->decl.
9798 * cgraphclones.cc (cgraph_node::create_clone): Copy also the
9799 node's calls_declare_variant_alt value.
9801 2022-10-24 Kito Cheng <kito.cheng@sifive.com>
9803 * common/config/riscv/riscv-common.cc (riscv_tunes): New.
9804 (riscv_get_valid_option_values): New.
9805 (TARGET_GET_VALID_OPTION_VALUES): New.
9806 * config/riscv/riscv-cores.def (RISCV_TUNE): New, define options
9808 (RISCV_CORE): Fix comment.
9809 * config/riscv/riscv.cc (riscv_tune_info_table): Move definition to
9812 2022-10-24 Aldy Hernandez <aldyh@redhat.com>
9814 PR tree-optimization/107365
9815 * value-range.cc (frange::verify_range): Predicate NAN check in
9816 VARYING range on HONOR_NANS instead of flag_finite_math_only.
9817 (range_tests_floats): Same.
9818 (range_tests_floats_various): New.
9819 (range_tests): Call range_tests_floats_various.
9821 2022-10-24 Torbjörn SVENSSON <torbjorn.svensson@foss.st.com>
9822 Yvan ROUX <yvan.roux@foss.st.com>
9824 * lto-wrapper.cc: Quote paths in makefile.
9826 2022-10-24 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
9828 * config/riscv/riscv.cc (riscv_legitimize_move): Support (set (mem) (const_poly_int)).
9830 2022-10-24 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
9832 * config/riscv/riscv-vector-builtins-bases.cc: Replace CONSTEXPR
9833 with constexpr throughout.
9834 * config/riscv/riscv-vector-builtins-shapes.cc (SHAPE): Likewise.
9835 * config/riscv/riscv-vector-builtins.cc
9836 (struct registered_function_hasher): Likewise.
9837 * config/riscv/riscv-vector-builtins.h (struct rvv_arg_type_info):
9840 2022-10-24 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
9842 * config/riscv/riscv-vector-switch.def (ENTRY): Remove unused TI/TF vector modes.
9844 2022-10-24 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
9846 * config/riscv/riscv.h (REG_CLASS_CONTENTS): Fix ALL_REGS.
9848 2022-10-22 Michael Eager <eager@eagercon.com>
9850 * config/microblaze/microblaze.cc
9851 (microblaze_legitimize_address): Initialize 'reg' to NULL, check for NULL.
9852 (microblaze_address_insns): Replace abort() with gcc_unreachable().
9853 (print_operand_address): Same.
9854 (microblaze_expand_move): Initialize 'p1' to NULL, check for NULL.
9855 (get_branch_target): Replace abort() with gcc_unreachable().
9857 2022-10-22 Aldy Hernandez <aldyh@redhat.com>
9859 * value-range.cc (range_tests_floats): Predicate [-Inf, +Inf] test
9860 with !flag_finite_math_only.
9862 2022-10-22 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
9864 * config/xtensa/xtensa.cc (xtensa_conditional_register_usage):
9865 Remove register A0 from FIXED_REGS if the CALL0 ABI.
9866 (xtensa_expand_epilogue): Change to emit '(use (reg:SI A0_REG))'
9867 unconditionally after restoring callee-saved registers for
9868 sibling-call functions, in order to prevent misleading that
9869 register A0 is free to use.
9871 2022-10-21 Jakub Jelinek <jakub@redhat.com>
9874 * config/i386/i386-expand.cc (ix86_prepare_fp_compare_args): For
9875 BFmode comparisons promote arguments to SFmode and recurse.
9876 (ix86_expand_int_movcc, ix86_expand_fp_movcc): Return false early
9877 if comparison operands are BFmode and operands[1] is not
9878 ix86_fp_comparison_operator.
9880 2022-10-21 Tejas Joshi <TejasSanjay.Joshi@amd.com>
9882 * common/config/i386/cpuinfo.h (get_amd_cpu): Recognize znver4.
9883 * common/config/i386/i386-common.cc (processor_names): Add znver4.
9884 (processor_alias_table): Add znver4 and modularize old znvers.
9885 * common/config/i386/i386-cpuinfo.h (processor_subtypes):
9887 * config.gcc (x86_64-*-* |...): Likewise.
9888 * config/i386/driver-i386.cc (host_detect_local_cpu): Let
9889 -march=native recognize znver4 cpus.
9890 * config/i386/i386-c.cc (ix86_target_macros_internal): Add znver4.
9891 * config/i386/i386-options.cc (m_ZNVER4): New definition.
9892 (m_ZNVER): Include m_ZNVER4.
9893 (processor_cost_table): Add znver4.
9894 * config/i386/i386.cc (ix86_reassociation_width): Likewise.
9895 * config/i386/i386.h (processor_type): Add PROCESSOR_ZNVER4.
9896 (PTA_ZNVER1): New definition.
9897 (PTA_ZNVER2): Likewise.
9898 (PTA_ZNVER3): Likewise.
9899 (PTA_ZNVER4): Likewise.
9900 * config/i386/i386.md (define_attr "cpu"): Add znver4 and rename
9902 * config/i386/x86-tune-costs.h (znver4_cost): New definition.
9903 * config/i386/x86-tune-sched.cc (ix86_issue_rate): Add znver4.
9904 (ix86_adjust_cost): Likewise.
9905 * config/i386/znver1.md: Rename to znver.md.
9906 * config/i386/znver.md: Add new reservations for znver4.
9907 * doc/extend.texi: Add details about znver4.
9908 * doc/invoke.texi: Likewise.
9910 2022-10-21 Richard Biener <rguenther@suse.de>
9912 PR tree-optimization/107323
9913 * tree-loop-distribution.cc (pg_unmark_merged_alias_ddrs):
9915 (loop_distribution::break_alias_scc_partitions): Revert
9916 postorder save/restore from the PR94125 fix. Instead
9917 make sure to not ignore edges from SCCs we are going to
9920 2022-10-21 Monk Chiang <monk.chiang@sifive.com>
9922 * config/riscv/riscv.md: Add atomic type attribute.
9923 * config/riscv/sync.md: Add atomic type for atomic instructions.
9925 2022-10-21 Jakub Jelinek <jakub@redhat.com>
9927 PR tree-optimization/54346
9928 * match.pd ((vec_perm (vec_perm@0 @1 @2 VECTOR_CST) @0 VECTOR_CST)):
9929 Optimize nested VEC_PERM_EXPRs even if target can't handle the
9930 new one provided we don't increase number of VEC_PERM_EXPRs the
9931 target can't handle.
9933 2022-10-21 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
9935 * config.gcc: Add riscv-vector-builtins-bases.o and riscv-vector-builtins-shapes.o
9936 * config/riscv/riscv-vector-builtins.cc (DEF_RVV_I_OPS): New macro.
9937 (DEF_RVV_FUNCTION): Ditto.
9938 (handle_pragma_vector): Add intrinsic framework.
9939 * config/riscv/riscv.cc (riscv_print_operand): Add operand print for vsetvl/vsetvlmax.
9940 * config/riscv/riscv.md: include vector.md.
9941 * config/riscv/t-riscv: Add riscv-vector-builtins-bases.o and riscv-vector-builtins-shapes.o
9942 * config/riscv/riscv-vector-builtins-bases.cc: New file.
9943 * config/riscv/riscv-vector-builtins-bases.h: New file.
9944 * config/riscv/riscv-vector-builtins-functions.def: New file.
9945 * config/riscv/riscv-vector-builtins-shapes.cc: New file.
9946 * config/riscv/riscv-vector-builtins-shapes.h: New file.
9947 * config/riscv/riscv-vector-builtins-types.def: New file.
9948 * config/riscv/vector.md: New file.
9950 2022-10-21 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
9952 * config.gcc: Add gt files since function_instance is GTY ((user)).
9953 * config/riscv/riscv-builtins.cc (riscv_init_builtins): Add RVV intrinsic framework.
9954 (riscv_builtin_decl): Ditto.
9955 (riscv_expand_builtin): Ditto.
9956 * config/riscv/riscv-protos.h (builtin_decl): New function.
9957 (expand_builtin): Ditto.
9958 (enum riscv_builtin_class): New enum to classify RVV intrinsic and RISC-V general built-in.
9959 * config/riscv/riscv-vector-builtins.cc (class GTY): New declaration.
9960 (struct registered_function_hasher): New struct.
9961 (DEF_RVV_OP_TYPE): New macro.
9962 (DEF_RVV_TYPE): Ditto.
9963 (DEF_RVV_PRED_TYPE): Ditto.
9964 (GTY): New declaration.
9965 (add_attribute): New function.
9966 (check_required_extensions): Ditto.
9967 (rvv_arg_type_info::get_tree_type): Ditto.
9968 (function_instance::function_instance): Ditto.
9969 (function_instance::operator==): Ditto.
9970 (function_instance::any_type_float_p): Ditto.
9971 (function_instance::get_return_type): Ditto.
9972 (function_instance::get_arg_type): Ditto.
9973 (function_instance::hash): Ditto.
9974 (function_instance::call_properties): Ditto.
9975 (function_instance::reads_global_state_p): Ditto.
9976 (function_instance::modifies_global_state_p): Ditto.
9977 (function_instance::could_trap_p): Ditto.
9978 (function_builder::function_builder): Ditto.
9979 (function_builder::~function_builder): Ditto.
9980 (function_builder::allocate_argument_types): Ditto.
9981 (function_builder::register_function_group): Ditto.
9982 (function_builder::append_name): Ditto.
9983 (function_builder::finish_name): Ditto.
9984 (function_builder::get_attributes): Ditto.
9985 (function_builder::add_function): Ditto.
9986 (function_builder::add_unique_function): Ditto.
9987 (function_call_info::function_call_info): Ditto.
9988 (function_expander::function_expander): Ditto.
9989 (function_expander::add_input_operand): Ditto.
9990 (function_expander::generate_insn): Ditto.
9991 (registered_function_hasher::hash): Ditto.
9992 (registered_function_hasher::equal): Ditto.
9993 (builtin_decl): Ditto.
9994 (expand_builtin): Ditto.
9995 (gt_ggc_mx): Define for using GCC garbage collect.
9996 (gt_pch_nx): Define for using GCC garbage collect.
9997 * config/riscv/riscv-vector-builtins.def (DEF_RVV_OP_TYPE): New macro.
9998 (DEF_RVV_PRED_TYPE): Ditto.
9999 (vbool64_t): Add suffix.
10000 (vbool32_t): Ditto.
10001 (vbool16_t): Ditto.
10006 (vint8mf8_t): Ditto.
10007 (vuint8mf8_t): Ditto.
10008 (vint8mf4_t): Ditto.
10009 (vuint8mf4_t): Ditto.
10010 (vint8mf2_t): Ditto.
10011 (vuint8mf2_t): Ditto.
10012 (vint8m1_t): Ditto.
10013 (vuint8m1_t): Ditto.
10014 (vint8m2_t): Ditto.
10015 (vuint8m2_t): Ditto.
10016 (vint8m4_t): Ditto.
10017 (vuint8m4_t): Ditto.
10018 (vint8m8_t): Ditto.
10019 (vuint8m8_t): Ditto.
10020 (vint16mf4_t): Ditto.
10021 (vuint16mf4_t): Ditto.
10022 (vint16mf2_t): Ditto.
10023 (vuint16mf2_t): Ditto.
10024 (vint16m1_t): Ditto.
10025 (vuint16m1_t): Ditto.
10026 (vint16m2_t): Ditto.
10027 (vuint16m2_t): Ditto.
10028 (vint16m4_t): Ditto.
10029 (vuint16m4_t): Ditto.
10030 (vint16m8_t): Ditto.
10031 (vuint16m8_t): Ditto.
10032 (vint32mf2_t): Ditto.
10033 (vuint32mf2_t): Ditto.
10034 (vint32m1_t): Ditto.
10035 (vuint32m1_t): Ditto.
10036 (vint32m2_t): Ditto.
10037 (vuint32m2_t): Ditto.
10038 (vint32m4_t): Ditto.
10039 (vuint32m4_t): Ditto.
10040 (vint32m8_t): Ditto.
10041 (vuint32m8_t): Ditto.
10042 (vint64m1_t): Ditto.
10043 (vuint64m1_t): Ditto.
10044 (vint64m2_t): Ditto.
10045 (vuint64m2_t): Ditto.
10046 (vint64m4_t): Ditto.
10047 (vuint64m4_t): Ditto.
10048 (vint64m8_t): Ditto.
10049 (vuint64m8_t): Ditto.
10050 (vfloat32mf2_t): Ditto.
10051 (vfloat32m1_t): Ditto.
10052 (vfloat32m2_t): Ditto.
10053 (vfloat32m4_t): Ditto.
10054 (vfloat32m8_t): Ditto.
10055 (vfloat64m1_t): Ditto.
10056 (vfloat64m2_t): Ditto.
10057 (vfloat64m4_t): Ditto.
10058 (vfloat64m8_t): Ditto.
10091 * config/riscv/riscv-vector-builtins.h (GCC_RISCV_VECTOR_BUILTINS_H): New macro.
10092 (RVV_REQUIRE_RV64BIT): Ditto.
10093 (RVV_REQUIRE_ZVE64): Ditto.
10094 (RVV_REQUIRE_ELEN_FP_32): Ditto.
10095 (RVV_REQUIRE_ELEN_FP_64): Ditto.
10096 (enum operand_type_index): New enum.
10097 (DEF_RVV_OP_TYPE): New macro.
10098 (enum predication_type_index): New enum.
10099 (DEF_RVV_PRED_TYPE): New macro.
10100 (enum rvv_base_type): New enum.
10101 (struct rvv_builtin_suffixes): New struct.
10102 (struct rvv_arg_type_info): Ditto.
10103 (struct rvv_type_info): Ditto.
10104 (struct rvv_op_info): Ditto.
10105 (class registered_function): New class.
10106 (class function_base): Ditto.
10107 (class function_shape): Ditto.
10108 (struct function_group_info): New struct.
10109 (class GTY): New class.
10110 (class function_builder): Ditto.
10111 (class function_call_info): Ditto.
10112 (function_call_info::function_returns_void_p): New function.
10113 (class function_expander): New class.
10114 (function_instance::operator!=): New function.
10115 (function_expander::expand): Ditto.
10116 (function_expander::add_input_operand): Ditto.
10117 (function_base::call_properties): Ditto.
10119 2022-10-21 Haochen Jiang <haochen.jiang@intel.com>
10121 * config/i386/sse.md (ssedvecmode): Rename from VI1SI.
10122 (ssedvecmodelower): Rename from vi1si.
10123 (sdot_prod<mode>): New define_expand.
10124 (udot_prod<mode>): Ditto.
10126 2022-10-21 Kong Lingling <lingling.kong@intel.com>
10127 Hongyu Wang <hongyu.wang@intel.com>
10128 Haochen Jiang <haochen.jiang@intel.com>
10130 * common/config/i386/cpuinfo.h (get_available_features): Detect
10132 * common/config/i386/i386-common.cc
10133 (OPTION_MASK_ISA2_AVXVNNIINT8_SET): New.
10134 (OPTION_MASK_ISA2_AVXVNNIINT8_UNSET): Ditto.
10135 (ix86_handle_option): Handle -mavxvnniint8.
10136 * common/config/i386/i386-cpuinfo.h (enum processor_features):
10137 Add FEATURE_AVXVNNIINT8.
10138 * common/config/i386/i386-isas.h: Add ISA_NAME_TABLE_ENTRY for
10140 * config.gcc: Add avxvnniint8intrin.h.
10141 * config/i386/avxvnniint8intrin.h: New file.
10142 * config/i386/cpuid.h (bit_AVXVNNIINT8): New.
10143 * config/i386/i386-builtin.def: Add new builtins.
10144 * config/i386/i386-c.cc (ix86_target_macros_internal): Define
10146 * config/i386/i386-options.cc (isa2_opts): Add -mavxvnniint8.
10147 (ix86_valid_target_attribute_inner_p): Handle avxvnniint8.
10148 * config/i386/i386-isa.def: Add DEF_PTA(AVXVNNIINT8) New..
10149 * config/i386/i386.opt: Add option -mavxvnniint8.
10150 * config/i386/immintrin.h: Include avxvnniint8intrin.h.
10151 * config/i386/sse.md (UNSPEC_VPMADDUBSWACCD
10152 UNSPEC_VPMADDUBSWACCSSD,UNSPEC_VPMADDWDACCD,
10153 UNSPEC_VPMADDWDACCSSD): Rename according to new style.
10154 (vpdp<vpdotprodtype>_<mode>): New define_insn.
10155 * doc/extend.texi: Document avxvnniint8.
10156 * doc/invoke.texi: Document -mavxvnniint8.
10157 * doc/sourcebuild.texi: Document target avxvnniint8.
10159 2022-10-21 Hongyu Wang <hongyu.wang@intel.com>
10161 * common/config/i386/i386-common.cc
10162 (OPTION_MASK_ISA_AVXIFMA_SET, OPTION_MASK_ISA2_AVXIFMA_UNSET,
10163 OPTION_MASK_ISA2_AVX2_UNSET): New macro.
10164 (ix86_handle_option): Handle -mavxifma.
10165 * common/config/i386/i386-cpuinfo.h (processor_types): Add
10167 * common/config/i386/i386-isas.h: Add ISA_NAME_TABLE_ENTRY for
10169 * common/config/i386/cpuinfo.h (get_available_features):
10171 * config.gcc: Add avxifmaintrin.h
10172 * config/i386/avx512ifmavlintrin.h: (_mm_madd52lo_epu64): Change
10174 (_mm_madd52hi_epu64): Likewise.
10175 (_mm256_madd52lo_epu64): Likewise.
10176 (_mm256_madd52hi_epu64): Likewise.
10177 * config/i386/avxifmaintrin.h: New header.
10178 * config/i386/cpuid.h (bit_AVXIFMA): New.
10179 * config/i386/i386-builtin.def: Add new builtins, and correct
10180 pattern names for AVX512IFMA.
10181 * config/i386/i386-builtins.cc (def_builtin): Handle AVX-IFMA
10182 builtins like AVX-VNNI.
10183 * config/i386/i386-c.cc (ix86_target_macros_internal): Define
10185 * config/i386/i386-expand.cc (ix86_check_builtin_isa_match):
10186 Relax ISA masks for AVXIFMA.
10187 * config/i386/i386-isa.def: Add AVXIFMA.
10188 * config/i386/i386-options.cc (isa2_opts): Add -mavxifma.
10189 (ix86_valid_target_attribute_inner_p): Handle avxifma.
10190 * config/i386/i386.md (isa): Add attr avxifma and avxifmavl.
10191 * config/i386/i386.opt: Add option -mavxifma.
10192 * config/i386/immintrin.h: Inculde avxifmaintrin.h.
10193 * config/i386/sse.md (avx_vpmadd52<vpmadd52type>_<mode>):
10195 (vpamdd52<vpmadd52type><mode><sd_maskz_name>): Remove.
10196 (vpamdd52huq<mode>_maskz): Rename to ...
10197 (vpmadd52huq<mode>_maskz): ... this.
10198 (vpamdd52luq<mode>_maskz): Rename to ...
10199 (vpmadd52luq<mode>_maskz): ... this.
10200 (vpmadd52<vpmadd52type><mode>): New define_insn.
10201 (vpmadd52<vpmadd52type>v8di): Likewise.
10202 (vpmadd52<vpmadd52type><mode>_maskz_1): Likewise.
10203 (vpamdd52<vpmadd52type><mode>_mask): Rename to ...
10204 (vpmadd52<vpmadd52type><mode>_mask): ... this.
10205 * doc/invoke.texi: Document -mavxifma.
10206 * doc/extend.texi: Document avxifma.
10207 * doc/sourcebuild.texi: Document target avxifma.
10209 2022-10-20 Aldy Hernandez <aldyh@redhat.com>
10212 * value-query.cc (get_range_global): Handle non integer ranges for
10213 default def SSA names.
10215 2022-10-20 Aldy Hernandez <aldyh@redhat.com>
10217 * range-op-float.cc (foperator_unordered_lt::op1_range): New.
10218 (foperator_unordered_lt::op2_range): New.
10220 2022-10-20 Artem Klimov <jakmobius@gmail.com>
10221 Alexander Monakov <amonakov@gcc.gnu.org>
10223 PR middle-end/99619
10224 * ipa-visibility.cc (function_and_variable_visibility): Promote
10225 TLS access model afer visibility optimizations.
10226 * varasm.cc (have_optimized_refs): New helper.
10227 (optimize_dyn_tls_for_decl_p): New helper. Use it ...
10228 (decl_default_tls_model): ... here in place of 'optimize' check.
10230 2022-10-20 Aldy Hernandez <aldyh@redhat.com>
10232 * range-op-float.cc (foperator_unordered_le::op1_range): Adjust
10233 false side with a NAN operand.
10234 (foperator_unordered_le::op2_range): Same.
10235 (foperator_unordered_gt::op1_range): Same.
10236 (foperator_unordered_gt::op2_range): Same.
10237 (foperator_unordered_ge::op1_range): Same.
10238 (foperator_unordered_ge::op2_range): Same.
10239 (foperator_unordered_equal::op1_range): Same.
10241 2022-10-20 Andre Vieira <andre.simoesdiasvieira@arm.com>
10243 PR tree-optimization/107326
10244 * tree-vect-patterns.cc (vect_recog_bitfield_ref_pattern): Change
10245 vectype when widening container.
10247 2022-10-20 Aldy Hernandez <aldyh@redhat.com>
10249 * value-range.h (frange::set_varying): Do not set NAN flags for
10251 * value-range.cc (frange::normalize_kind): Adjust for no NAN when
10253 (frange::verify_range): Same.
10254 * range-op-float.cc (maybe_isnan): Remove flag_finite_math_only check.
10256 2022-10-20 Aldy Hernandez <aldyh@redhat.com>
10258 * range-op-float.cc (finite_operand_p): Remove.
10259 (finite_operands_p): Rename to...
10260 (maybe_isnan): ...this.
10261 (frelop_early_resolve): Use maybe_isnan instead of finite_operands_p.
10262 (foperator_equal::fold_range): Same.
10263 (foperator_equal::op1_range): Same.
10264 (foperator_not_equal::fold_range): Same.
10265 (foperator_lt::fold_range): Same.
10266 (foperator_le::fold_range): Same.
10267 (foperator_gt::fold_range): Same.
10268 (foperator_ge::fold_range): Same.
10270 2022-10-20 Jakub Jelinek <jakub@redhat.com>
10272 * passes.cc (pass_manager::register_pass): Fix a comment
10273 typo - copmilation -> compilation.
10275 2022-10-20 Richard Biener <rguenther@suse.de>
10277 * tree-vect-loop.cc (vect_phi_first_order_recurrence_p):
10278 Disallow latch PHI defs.
10279 (vectorizable_recurr): Revert previous change.
10281 2022-10-20 Julian Brown <julian@codesourcery.com>
10284 * config/gcn/gcn.cc (gcn_detect_incoming_pointer_arg): Any pointer
10285 argument forces FLAT addressing mode, not just
10286 pointer-to-non-aggregate.
10288 2022-10-20 Thomas Schwinge <thomas@codesourcery.com>
10290 * configure.ac (AC_CONFIG_MACRO_DIRS): Instantiate.
10291 * configure: Regenerate.
10293 2022-10-20 Richard Sandiford <richard.sandiford@arm.com>
10295 * config/aarch64/aarch64-sve-builtins.h
10296 (gimple_folder::fold_to_cstu): New member function.
10297 * config/aarch64/aarch64-sve-builtins.cc
10298 (gimple_folder::fold_to_cstu): Define.
10299 * config/aarch64/aarch64-sve-builtins-base.cc
10300 (svcnt_bhwd_impl::fold): Use it.
10302 2022-10-20 Richard Sandiford <richard.sandiford@arm.com>
10304 * config/aarch64/aarch64-sve-builtins-functions.h (quiet)
10305 (rtx_code_function, rtx_code_function_rotated, unspec_based_function)
10306 (unspec_based_function_rotated, unspec_based_function_exact_insn)
10307 (unspec_based_fused_function, unspec_based_fused_lane_function):
10308 Replace constructors with using directives.
10309 * config/aarch64/aarch64-sve-builtins-base.cc (svcnt_bhwd_pat_impl)
10310 (svcreate_impl, svdotprod_lane_impl, svget_impl, svld1_extend_impl)
10311 (svld1_gather_extend_impl, svld234_impl, svldff1_gather_extend)
10312 (svset_impl, svst1_scatter_truncate_impl, svst1_truncate_impl)
10313 (svst234_impl, svundef_impl): Likewise.
10314 * config/aarch64/aarch64-sve-builtins-sve2.cc
10315 (svldnt1_gather_extend_impl, svmovl_lb_impl): Likewise.
10316 (svstnt1_scatter_truncate_impl): Likewise.
10318 2022-10-20 Richard Sandiford <richard.sandiford@arm.com>
10320 * config/aarch64/aarch64-sve-builtins-base.cc: Replace CONSTEXPR
10321 with constexpr throughout.
10322 * config/aarch64/aarch64-sve-builtins-functions.h: Likewise.
10323 * config/aarch64/aarch64-sve-builtins-shapes.cc: Likewise.
10324 * config/aarch64/aarch64-sve-builtins-sve2.cc: Likewise.
10325 * config/aarch64/aarch64-sve-builtins.cc: Likewise.
10327 2022-10-20 Richard Sandiford <richard.sandiford@arm.com>
10329 * config/aarch64/aarch64-sve.md (*aarch64_brk<brk_op>_cc): Remove
10330 merging alternative.
10331 (*aarch64_brk<brk_op>_ptest): Likewise.
10333 2022-10-20 Richard Sandiford <richard.sandiford@arm.com>
10335 * config/aarch64/iterators.md (SVE_BRKP): New iterator.
10336 * config/aarch64/aarch64-sve.md (*aarch64_brkn_cc): New pattern.
10337 (*aarch64_brkn_ptest): Likewise.
10338 (*aarch64_brk<brk_op>_cc): Restrict to SVE_BRKP.
10339 (*aarch64_brk<brk_op>_ptest): Likewise.
10341 2022-10-20 Richard Biener <rguenther@suse.de>
10345 * tree-cfg.h (verify_gimple_in_seq): Add parameter to
10346 indicate whether to emit an ICE. Add return value.
10347 (verify_gimple_in_cfg): Likewise.
10348 * tree-cfg.cc (verify_gimple_in_seq): Likewise.
10349 (verify_gimple_in_cfg): Likewise.
10351 2022-10-20 Richard Biener <rguenther@suse.de>
10353 PR tree-optimization/107240
10354 * tree-vect-patterns.cc (vect_recog_bit_insert_pattern): Attempt to
10355 simplify shifted value first.
10357 2022-10-20 Andrew MacLeod <amacleod@redhat.com>
10359 * gimple-range-gori.h (compute_operand_range): Make public.
10360 * gimple-range-infer.cc (gimple_infer_range::check_assume_func): New.
10361 (gimple_infer_range::gimple_infer_range): Check for assume calls.
10362 * gimple-range-infer.h (check_assume_func): Add prototype.
10363 * gimple-range.cc (assume_query::assume_range_p): New.
10364 (assume_query::range_of_expr): New.
10365 (assume_query::assume_query): New.
10366 (assume_query::calculate_op): New.
10367 (assume_query::calculate_phi): New.
10368 (assume_query::check_taken_edge): New.
10369 (assume_query::calculate_stmt): New.
10370 (assume_query::dump): New.
10371 * gimple-range.h (class assume_query): New.
10372 * tree-vrp.cc (pass_assumptions::execute): Add processing.
10374 2022-10-19 Aldy Hernandez <aldyh@redhat.com>
10376 * range-op-float.cc (build_le): Document result.
10379 (foperator_ge::op2_range): Check result of build_*.
10380 (foperator_unordered_le::op1_range): Same.
10381 (foperator_unordered_le::op2_range): Same.
10382 (foperator_unordered_gt::op1_range): Same.
10383 (foperator_unordered_gt::op2_range): Same.
10384 (foperator_unordered_ge::op1_range): Same.
10385 (foperator_unordered_ge::op2_range): Same.
10387 2022-10-19 Marek Polacek <polacek@redhat.com>
10390 * doc/invoke.texi: Update documentation of -Wuseless-cast.
10392 2022-10-19 Andrew MacLeod <amacleod@redhat.com>
10394 * gimple-range-cache.cc (ranger_cache::range_from_dom): Use
10395 Value_Range not int_range_max.
10397 2022-10-19 Aldy Hernandez <aldyh@redhat.com>
10399 PR tree-optimization/107312
10400 * range.h (range_true_and_false): Special case 1-bit signed types.
10401 * value-range.cc (range_tests_misc): New test.
10403 2022-10-19 LIU Hao <lh_mouse@126.com>
10405 * config/i386/mingw-mcfgthread.h: New file
10406 * config/i386/mingw32.h: Add builtin macro and default libraries
10407 for mcfgthread when thread model is `mcf`
10408 * config.gcc: Include 'i386/mingw-mcfgthread.h' when thread model
10410 * configure.ac: Recognize `mcf` as a valid thread model
10411 * config.in: Regenerate
10412 * configure: Regenerate
10414 2022-10-19 Lewis Hyatt <lhyatt@gmail.com>
10416 * gengtype.cc (output_escaped_param): Add missing const.
10417 (get_string_option): Add missing check for option type.
10418 (walk_type): Support new "string_length" GTY option.
10419 (write_types_process_field): Likewise.
10420 * ggc-common.cc (gt_pch_note_object): Add optional length argument.
10421 * ggc.h (gt_pch_note_object): Adjust prototype for new argument.
10422 (gt_pch_n_S2): Declare...
10423 * stringpool.cc (gt_pch_n_S2): ...new function.
10424 * doc/gty.texi: Document new GTY((string_length)) option.
10426 2022-10-19 Martin Liska <mliska@suse.cz>
10428 * doc/extend.texi: Remove useless @tie{} directives.
10430 2022-10-19 Martin Jambor <mjambor@suse.cz>
10432 PR tree-optimization/107206
10433 * tree-sra.cc (struct access): New field grp_result_of_prop_from_lhs.
10434 (analyze_access_subtree): Do not create replacements for accesses with
10435 this flag when not toally scalarizing.
10436 (propagate_subaccesses_from_lhs): Set the new flag.
10438 2022-10-19 Stefan Schulze Frielinghaus <stefansf@linux.ibm.com>
10441 * config/s390/s390.cc (s390_call_saved_register_used): For a
10442 parameter with BLKmode fix determining number of consecutive
10445 2022-10-19 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
10447 * config/xtensa/xtensa-protos.h
10448 (xtensa_split1_finished_p, xtensa_split_DI_reg_imm): New prototypes.
10449 * config/xtensa/xtensa.cc
10450 (xtensa_split1_finished_p, xtensa_split_DI_reg_imm, xtensa_lra_p):
10452 (TARGET_LRA_P): Replace the dummy hook with xtensa_lra_p.
10453 (xt_true_regnum): Rework.
10454 * config/xtensa/xtensa.h (CALL_REALLY_USED_REGISTERS):
10455 Switch from CALL_USED_REGISTERS, and revise the comment.
10456 * config/xtensa/constraints.md (Y):
10457 Use !xtensa_split1_finished_p() instead of can_create_pseudo_p().
10458 * config/xtensa/predicates.md (move_operand): Ditto.
10459 * config/xtensa/xtensa.md: Add two new split patterns:
10460 - splits DImode immediate load into two SImode ones
10461 - puts out-of-constraint SImode constants into the constant pool
10462 * config/xtensa/xtensa.opt (-mlra): New target-specific option
10463 for testing purpose.
10465 2022-10-19 Robin Dapp <rdapp@linux.ibm.com>
10467 * config/s390/s390.md: Move reload_completed and check operands for REG_P.
10469 2022-10-19 Jakub Jelinek <jakub@redhat.com>
10471 PR middle-end/107262
10472 * expr.cc (convert_mode_scalar): For BFmode -> SFmode conversions
10473 of constants, use simplify_unary_operation if fromi has VOIDmode
10474 instead of recursive convert_mode_scalar.
10476 2022-10-19 Jakub Jelinek <jakub@redhat.com>
10478 PR tree-optimization/106990
10479 * match.pd ((~X - ~Y) -> Y - X, -x & 1 -> x & 1): Guard with
10480 !TYPE_OVERFLOW_SANITIZED (type).
10482 2022-10-19 Jakub Jelinek <jakub@redhat.com>
10484 * config/i386/i386-builtins.cc (ix86_bf16_ptr_type_node): Remove.
10485 (ix86_bf16_type_node): New variable.
10486 (ix86_register_bf16_builtin_type): If bfloat16_type_node is NULL
10487 from generic code, set only ix86_bf16_type_node to a new REAL_TYPE
10488 rather than bfloat16_type_node, otherwise set ix86_bf16_type_node
10489 to bfloat16_type_node. Register __bf16 on ix86_bf16_type_node
10490 rather than bfloat16_type_node. Don't initialize unused
10491 ix86_bf16_ptr_type_node.
10492 * config/i386/i386-builtin-types.def (BFLOAT16): Use
10493 ix86_bf16_type_node rather than bfloat16_type_node.
10495 2022-10-19 Richard Biener <rguenther@suse.de>
10497 PR tree-optimization/106781
10498 * cgraph.cc (cgraph_edge::redirect_call_stmt_to_callee): Copy
10499 LHS removal from fixup_noreturn_call.
10501 2022-10-19 liuhongt <hongtao.liu@intel.com>
10504 * config/i386/i386-expand.cc (ix86_vec_perm_index_canon): New.
10505 (expand_vec_perm_shufps_shufps): Call
10506 ix86_vec_perm_index_canon
10508 2022-10-18 Martin Jambor <mjambor@suse.cz>
10510 * ipa-prop.h (ipa_agg_value): Remove type.
10511 (ipa_agg_value_set): Likewise.
10512 (ipa_copy_agg_values): Remove function.
10513 (ipa_release_agg_values): Likewise.
10514 (ipa_auto_call_arg_values) Add a forward declaration.
10515 (ipa_call_arg_values): Likewise.
10516 (class ipa_argagg_value_list): New constructors, added member function
10518 (class ipa_auto_call_arg_values): Removed the destructor and member
10519 function safe_aggval_at. Use ipa_argagg_values for m_known_aggs.
10520 (class ipa_call_arg_values): Removed member function safe_aggval_at.
10521 Use ipa_argagg_values for m_known_aggs.
10522 (ipa_get_indirect_edge_target): Removed declaration.
10523 (ipa_find_agg_cst_for_param): Likewise.
10524 (ipa_find_agg_cst_from_init): New declaration.
10525 (ipa_agg_value_from_jfunc): Likewise.
10526 (ipa_agg_value_set_from_jfunc): Removed declaration.
10527 (ipa_push_agg_values_from_jfunc): New declaration.
10528 * ipa-cp.cc (ipa_agg_value_from_node): Renamed to
10529 ipa_agg_value_from_jfunc, made public.
10530 (ipa_agg_value_set_from_jfunc): Removed.
10531 (ipa_push_agg_values_from_jfunc): New function.
10532 (ipa_get_indirect_edge_target_1): Removed known_aggs parameter, use
10533 avs for this purpose too.
10534 (ipa_get_indirect_edge_target): Removed the overload working on
10535 ipa_auto_call_arg_values, use ipa_argagg_value_list in the remaining
10537 (devirtualization_time_bonus): Use ipa_argagg_value_list and
10538 ipa_get_indirect_edge_target_1 instead of
10539 ipa_get_indirect_edge_target.
10540 (context_independent_aggregate_values): Removed function.
10541 (gather_context_independent_values): Work on ipa_argagg_value_list.
10542 (estimate_local_effects): Likewise, define some iterator variables
10543 only in the construct where necessary.
10544 (ipcp_discover_new_direct_edges): Adjust the call to
10545 ipa_get_indirect_edge_target_1.
10546 (push_agg_values_for_index_from_edge): Adjust the call
10547 ipa_agg_value_from_node which has been renamed to
10548 ipa_agg_value_from_jfunc.
10549 * ipa-fnsummary.cc (evaluate_conditions_for_known_args): Work on
10550 ipa_argagg_value_list.
10551 (evaluate_properties_for_edge): Replace manual filling in aggregate
10552 values with call to ipa_push_agg_values_from_jfunc.
10553 (estimate_calls_size_and_time): Work on ipa_argagg_value_list.
10554 (ipa_cached_call_context::duplicate_from): Likewise.
10555 (ipa_cached_call_context::release): Likewise.
10556 (ipa_call_context::equal_to): Likewise.
10557 * ipa-prop.cc (ipa_find_agg_cst_from_init): Make public.
10558 (ipa_find_agg_cst_for_param): Removed function.
10559 (ipa_find_agg_cst_from_jfunc_items): New function.
10560 (try_make_edge_direct_simple_call): Replace calls to
10561 ipa_agg_value_set_from_jfunc and ipa_find_agg_cst_for_param with
10562 ipa_find_agg_cst_from_init and ipa_find_agg_cst_from_jfunc_items.
10563 (try_make_edge_direct_virtual_call): Replace calls to
10564 ipa_agg_value_set_from_jfunc and ipa_find_agg_cst_for_param with
10565 simple query of constant jump function and a call to
10566 ipa_find_agg_cst_from_jfunc_items.
10567 (ipa_auto_call_arg_values::~ipa_auto_call_arg_values): Removed.
10569 2022-10-18 Martin Jambor <mjambor@suse.cz>
10571 * ipa-prop.h (IPA_PROP_ARG_INDEX_LIMIT_BITS): New.
10572 (ipcp_transformation): Added forward declaration.
10573 (ipa_argagg_value): New type.
10574 (ipa_argagg_value_list): New type.
10575 (ipa_agg_replacement_value): Removed type.
10576 (ipcp_transformation): Switch from using ipa_agg_replacement_value
10577 to ipa_argagg_value_list.
10578 (ipa_get_agg_replacements_for_node): Removed.
10579 (ipa_dump_agg_replacement_values): Removed declaration.
10580 * ipa-cp.cc: Define INCLUDE_ALGORITHM.
10581 (values_equal_for_ipcp_p): Moved up in the file.
10582 (ipa_argagg_value_list::dump): New function.
10583 (ipa_argagg_value_list::debug): Likewise.
10584 (ipa_argagg_value_list::get_elt): Likewise.
10585 (ipa_argagg_value_list::get_elt_for_index): Likewise.
10586 (ipa_argagg_value_list::get_value): New overloaded functions.
10587 (ipa_argagg_value_list::superset_of_p): New function.
10588 (new ipa_argagg_value_list::push_adjusted_values): Likewise.
10589 (push_agg_values_from_plats): Likewise.
10590 (intersect_argaggs_with): Likewise.
10591 (get_clone_agg_value): Removed.
10592 (ipa_agg_value_from_node): Make last parameter const, use
10593 ipa_argagg_value_list to search values coming from clones.
10594 (ipa_get_indirect_edge_target_1): Use ipa_argagg_value_list to search
10595 values coming from clones.
10596 (ipcp_discover_new_direct_edges): Pass around a vector of
10597 ipa_argagg_values rather than a link list of replacement values.
10598 (cgraph_edge_brings_value_p): Use ipa_argagg_value_list to search
10599 values coming from clones.
10600 (create_specialized_node): Work with a vector of ipa_argagg_values
10601 rather than a link list of replacement values.
10602 (self_recursive_agg_pass_through_p): Make the pointer parameters
10604 (copy_plats_to_inter): Removed.
10605 (intersect_with_plats): Likewise.
10606 (agg_replacements_to_vector): Likewise.
10607 (intersect_with_agg_replacements): Likewise.
10608 (intersect_aggregates_with_edge): Likewise.
10609 (push_agg_values_for_index_from_edge): Likewise.
10610 (push_agg_values_from_edge): Likewise.
10611 (find_aggregate_values_for_callers_subset): Rewrite.
10612 (cgraph_edge_brings_all_agg_vals_for_node): Likewise.
10613 (ipcp_val_agg_replacement_ok_p): Use ipa_argagg_value_list to search
10615 (decide_about_value): Work with a vector of ipa_argagg_values rather
10616 than a link list of replacement values.
10617 (decide_whether_version_node): Likewise.
10618 (ipa_analyze_node): Check number of parameters, assert that there
10619 are no descriptors when bailing out.
10620 * ipa-prop.cc (ipa_set_node_agg_value_chain): Switch to a vector of
10622 (ipa_node_params_t::duplicate): Removed superfluous handling of
10623 ipa_agg_replacement_values. Name of src parameter removed because
10624 it is no longer used.
10625 (ipcp_transformation_t::duplicate): Replaced duplication of
10626 ipa_agg_replacement_values with copying vector m_agg_values.
10627 (ipa_dump_agg_replacement_values): Removed.
10628 (write_ipcp_transformation_info): Stream the new data-structure
10629 instead of the old.
10630 (read_ipcp_transformation_info): Likewise.
10631 (adjust_agg_replacement_values): Work with ipa_argagg_values instead
10632 of linked lists of ipa_agg_replacement_values, copy the items and
10633 truncate the vector as necessary to keep it sorted instead of marking
10634 items as invalid. Return one bool if CFG should be updated.
10635 (ipcp_modif_dom_walker): Store ipcp_transformation instead of
10636 linked list of ipa_agg_replacement_values.
10637 (ipcp_modif_dom_walker::before_dom_children): Use
10638 ipa_argagg_value_list instead of walking a list of
10639 ipa_agg_replacement_values.
10640 (ipcp_transform_function): Switch to the new data structure, adjust
10643 2022-10-18 Richard Biener <rguenther@suse.de>
10645 PR tree-optimization/107302
10646 * tree-vect-loop.cc (vectorizable_recurrence): Fix vec_perm
10647 placement for a PHI latch def.
10649 2022-10-18 Andre Vieira <andre.simoesdiasvieira@arm.com>
10651 PR tree-optimization/107275
10652 * tree-if-conv.cc (if_convertible_loop_p_1): Move
10653 find_data_references_in_loop call from here...
10654 (if_convertible_loop_p): And move data-reference vector initialization
10656 (tree_if_conversion):... to here.
10658 2022-10-18 Jakub Jelinek <jakub@redhat.com>
10661 * gimple.def (GIMPLE_ASSUME): New statement kind.
10662 * gimple.h (struct gimple_statement_assume): New type.
10663 (is_a_helper <gimple_statement_assume *>::test,
10664 is_a_helper <const gimple_statement_assume *>::test): New.
10665 (gimple_build_assume): Declare.
10666 (gimple_has_substatements): Return true for GIMPLE_ASSUME.
10667 (gimple_assume_guard, gimple_assume_set_guard,
10668 gimple_assume_guard_ptr, gimple_assume_body_ptr, gimple_assume_body):
10669 New inline functions.
10670 * gsstruct.def (GSS_ASSUME): New.
10671 * gimple.cc (gimple_build_assume): New function.
10672 (gimple_copy): Handle GIMPLE_ASSUME.
10673 * gimple-pretty-print.cc (dump_gimple_assume): New function.
10674 (pp_gimple_stmt_1): Handle GIMPLE_ASSUME.
10675 * gimple-walk.cc (walk_gimple_op): Handle GIMPLE_ASSUME.
10676 * omp-low.cc (WALK_SUBSTMTS): Likewise.
10677 (lower_omp_1): Likewise.
10678 * omp-oacc-kernels-decompose.cc (adjust_region_code_walk_stmt_fn):
10680 * tree-cfg.cc (verify_gimple_stmt, verify_gimple_in_seq_2): Likewise.
10681 * function.h (struct function): Add assume_function bitfield.
10682 * gimplify.cc (gimplify_call_expr): If the assumption isn't
10683 simple enough, expand it into GIMPLE_ASSUME wrapped block or
10685 * gimple-low.cc: Include attribs.h.
10686 (create_assumption_fn): New function.
10687 (struct lower_assumption_data): New type.
10688 (find_assumption_locals_r, assumption_copy_decl,
10689 adjust_assumption_stmt_r, adjust_assumption_stmt_op,
10690 lower_assumption): New functions.
10691 (lower_stmt): Handle GIMPLE_ASSUME.
10692 * tree-ssa-ccp.cc (pass_fold_builtins::execute): Remove
10694 * lto-streamer-out.cc (output_struct_function_base): Pack
10695 assume_function bit.
10696 * lto-streamer-in.cc (input_struct_function_base): And unpack it.
10697 * cgraphunit.cc (cgraph_node::expand): Don't verify assume_function
10698 has TREE_ASM_WRITTEN set and don't release its body.
10699 (symbol_table::compile): Allow assume functions not to have released
10701 * internal-fn.cc (expand_ASSUME): Remove gcc_unreachable.
10702 * passes.cc (execute_one_pass): For TODO_discard_function don't
10703 release body of assume functions.
10704 * cgraph.cc (cgraph_node::verify_node): Don't verify cgraph nodes
10705 of PROP_assumptions_done functions.
10706 * tree-pass.h (PROP_assumptions_done): Define.
10707 (TODO_discard_function): Adjust comment.
10708 (make_pass_assumptions): Declare.
10709 * passes.def (pass_assumptions): Add.
10710 * timevar.def (TV_TREE_ASSUMPTIONS): New.
10711 * tree-inline.cc (remap_gimple_stmt): Handle GIMPLE_ASSUME.
10712 * tree-vrp.cc (pass_data_assumptions): New variable.
10713 (pass_assumptions): New class.
10714 (make_pass_assumptions): New function.
10716 2022-10-18 Richard Biener <rguenther@suse.de>
10718 PR tree-optimization/107301
10719 * gimple-ssa-isolate-paths.cc (handle_return_addr_local_phi_arg):
10720 Check whether we can duplicate the block.
10721 (find_implicit_erroneous_behavior): Likewise.
10723 2022-10-18 Andrew MacLeod <amacleod@redhat.com>
10725 PR tree-optimization/107273
10726 * value-relation.cc (equiv_oracle::add_partial_equiv): Merge
10727 instead of copying precison of each member.
10729 2022-10-17 Jeff Law <jeffreyalaw@gmail.com>
10732 * config/h8300/combiner.md: Replace '<' preincment constraint with
10733 ZA/Z1..ZH/Z7 combinations.
10734 * config/h8300/movepush.md: Similarly
10736 2022-10-17 Jeff Law <jeffreyalaw@gmail.com>
10738 * config/h8300/constraints.md (Za..Zh): New constraints for
10739 autoinc addresses using a specific register.
10740 * config/h8300/h8300.cc (pre_incdec_with_reg): New function.
10741 * config/h8300/h8300-protos.h (pre_incdec_with_reg): Add prototype.
10743 2022-10-17 Jeff Law <jlaw@ventanamicro.com>
10745 * config/i386/cet.c: Remove accidental commit.
10746 * config/i386/driver-mingw32.c: Likewise.
10747 * config/i386/i386-builtins.c: Likewise.
10748 * config/i386/i386-d.c: Likewise.
10749 * config/i386/i386-expand.c: Likewise.
10750 * config/i386/i386-features.c: Likewise.
10751 * config/i386/i386-options.c: Likewise.
10752 * config/i386/t-cet: Likewise.
10753 * config/i386/x86-tune-sched-atom.c: Likewise.
10754 * config/i386/x86-tune-sched-bd.c: Likewise.
10755 * config/i386/x86-tune-sched-core.c: Likewise.
10756 * config/i386/x86-tune-sched.c: Likewise.
10758 2022-10-17 Jeff Law <jeffreyalaw@gmail.com>
10760 * common/config/h8300/h8300-common.cc (h8300_option_optimization_table):
10761 Enable redundant extension elimination at -O2 and above.
10762 * config/i386/cet.c: New file.
10763 * config/i386/driver-mingw32.c: New file.
10764 * config/i386/i386-builtins.c: New file.
10765 * config/i386/i386-d.c: New file.
10766 * config/i386/i386-expand.c: New file.
10767 * config/i386/i386-features.c: New file.
10768 * config/i386/i386-options.c: New file.
10769 * config/i386/t-cet: New file.
10770 * config/i386/x86-tune-sched-atom.c: New file.
10771 * config/i386/x86-tune-sched-bd.c: New file.
10772 * config/i386/x86-tune-sched-core.c: New file.
10773 * config/i386/x86-tune-sched.c: New file.
10775 2022-10-17 Jeff Law <jeffreyalaw@gmail.com>
10777 * config/h8300/extensions.md (CCZN setting zero extended load): Add
10780 2022-10-17 Aldy Hernandez <aldyh@redhat.com>
10782 PR tree-optimization/107293
10784 (dom_opt_dom_walker::set_global_ranges_from_unreachable_edges):
10785 Check that condition post-dominates the definition point.
10787 2022-10-17 Aldy Hernandez <aldyh@redhat.com>
10789 PR tree-optimization/107286
10790 * value-range.cc (range_tests_floats): Do not test for -Inf when
10791 flag_finite_math_only.
10793 2022-10-17 Andrew MacLeod <amacleod@redhat.com>
10795 * gimple-range-fold.cc (fold_using_range::range_of_range_op):
10796 Provide relation_trio class.
10797 * gimple-range-gori.cc (gori_compute::refine_using_relation):
10798 Provide relation_trio class.
10799 (gori_compute::refine_using_relation): Ditto.
10800 (gori_compute::compute_operand1_range): Provide lhs_op2 and
10801 op1_op2 relations via relation_trio class.
10802 (gori_compute::compute_operand2_range): Ditto.
10803 * gimple-range-op.cc (gimple_range_op_handler::calc_op1): Use
10804 relation_trio instead of relation_kind.
10805 (gimple_range_op_handler::calc_op2): Ditto.
10806 (*::fold_range): Ditto.
10807 * gimple-range-op.h (gimple_range_op::calc_op1): Adjust prototypes.
10808 (gimple_range_op::calc_op2): Adjust prototypes.
10809 * range-op-float.cc (*::fold_range): Use relation_trio instead of
10811 (*::op1_range): Ditto.
10812 (*::op2_range): Ditto.
10813 * range-op.cc (*::fold_range): Use relation_trio instead of
10815 (*::op1_range): Ditto.
10816 (*::op2_range): Ditto.
10817 * range-op.h (class range_operator): Adjust prototypes.
10818 (class range_operator_float): Ditto.
10819 (class range_op_handler): Adjust prototypes.
10820 (relop_early_resolve): Pickup op1_op2 relation from relation_trio.
10821 * value-relation.cc (VREL_LAST): Adjust use to be one past the end of
10823 (relation_oracle::validate_relation): Use relation_trio in call
10825 * value-relation.h (enum relation_kind_t): Add VREL_LAST as
10827 (class relation_trio): New.
10828 (TRIO_VARYING, TRIO_SHIFT, TRIO_MASK): New.
10830 2022-10-17 Andrew MacLeod <amacleod@redhat.com>
10832 * range-op-float.cc (foperator_not_equal::op1_range): Check for
10833 VREL_EQ after singleton.
10834 (foperator_unordered::op1_range): Set VARYING before calling
10836 (foperator_ordered::op1_range): Set rather than clear NAN if both
10837 operands are the same.
10839 2022-10-17 Andrew MacLeod <amacleod@redhat.com>
10841 * value-relation.cc (value_relation::dump): Change message.
10842 * value-relation.h (value_relation::set_relation): If op1 is the
10843 same as op2 do not create a relation.
10845 2022-10-17 Thomas Schwinge <thomas@codesourcery.com>
10847 * config/gcn/gcn.cc (VnMODE): Use 'case E_QImode:' instead of
10848 'case QImode:', etc.
10850 2022-10-17 Thomas Schwinge <thomas@codesourcery.com>
10852 * gimple-expr.cc (mark_addressable_2): Tag as 'static'.
10854 2022-10-17 Richard Biener <rguenther@suse.de>
10855 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
10857 PR tree-optimization/99409
10858 PR tree-optimization/99394
10859 * tree-vectorizer.h (vect_def_type::vect_first_order_recurrence): Add.
10860 (stmt_vec_info_type::recurr_info_type): Likewise.
10861 (vectorizable_recurr): New function.
10862 * tree-vect-loop.cc (vect_phi_first_order_recurrence_p): New
10864 (vect_analyze_scalar_cycles_1): Look for first order
10866 (vect_analyze_loop_operations): Handle them.
10867 (vect_transform_loop): Likewise.
10868 (vectorizable_recurr): New function.
10869 (maybe_set_vectorized_backedge_value): Handle the backedge value
10870 setting in the first order recurrence PHI and the permutes.
10871 * tree-vect-stmts.cc (vect_analyze_stmt): Handle first order
10873 (vect_transform_stmt): Likewise.
10874 (vect_is_simple_use): Likewise.
10875 (vect_is_simple_use): Likewise.
10876 * tree-vect-slp.cc (vect_get_and_check_slp_defs): Likewise.
10877 (vect_build_slp_tree_2): Likewise.
10878 (vect_schedule_scc): Handle the backedge value setting in the
10879 first order recurrence PHI and the permutes.
10881 2022-10-17 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
10883 * config/riscv/t-riscv: Change Tab into 2 space.
10885 2022-10-17 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
10887 * config/riscv/riscv-vector-builtins.cc (mangle_builtin_type): Move down the function.
10889 2022-10-17 Arsen Arsenovic <arsen@aarsen.me>
10891 * config.gcc: --target=*-elf --without-{newlib,headers} should
10894 2022-10-17 Hu, Lin1 <lin1.hu@intel.com>
10896 * common/config/i386/cpuinfo.h:
10897 (get_intel_cpu): Handle Meteorlake.
10898 * common/config/i386/i386-common.cc:
10899 (processor_alias_table): Add Meteorlake.
10901 2022-10-17 Haochen Jiang <haochen.jiang@intel.com>
10903 * common/config/i386/cpuinfo.h:
10904 (get_intel_cpu): Handle Raptorlake.
10905 * common/config/i386/i386-common.cc:
10906 (processor_alias_table): Add Raptorlake.
10908 2022-10-16 Jeff Law <jeffreyalaw@gmail.com>
10910 * config/h8300/constraints.md (Z0..Z7): New register
10912 * config/h8300/h8300.h (reg_class): Add new classes.
10913 (REG_CLASS_NAMES): Similarly.
10914 (REG_CLASS_CONTENTS): Similarly.
10916 2022-10-16 Jeff Law <jeffreyalaw@gmail.com>
10918 * config/h8300/constraints.md (Zz constraint): Renamed
10920 * config/h8300/movepush.md (movqi_h8sx, movhi_h8sx): Adjust
10921 constraint to use Zz instead of Z.
10923 2022-10-16 Jeff Law <jeffreyalaw@gmail.com>
10925 * config/h8300/h8300.cc (h8300_register_move_cost): Fix typo.
10927 2022-10-14 Aldy Hernandez <aldyh@redhat.com>
10929 * value-range.cc (frange::set): Implement distinction between
10930 HONOR_SIGNED_ZEROS and MODE_HAS_SIGNED_ZEROS.
10932 2022-10-14 Aldy Hernandez <aldyh@redhat.com>
10934 * gimple-range-op.cc (class cfn_copysign): New.
10935 (gimple_range_op_handler::maybe_builtin_call): Add
10936 CFN_BUILT_IN_COPYSIGN*.
10938 2022-10-14 Aldy Hernandez <aldyh@redhat.com>
10940 * real.h (real_isdenormal): Check rvc_normal.
10941 * value-range.cc (range_tests_floats): New test.
10943 2022-10-14 Aldy Hernandez <aldyh@redhat.com>
10945 * gimple-range-op.cc
10946 (gimple_range_op_handler::maybe_builtin_call): Replace
10947 CFN_BUILTIN_SIGNBIT* cases with CASE_FLT_FN.
10949 2022-10-14 Aldy Hernandez <aldyh@redhat.com>
10951 * value-range.cc (frange::set): Normalize ranges for both bounds.
10953 2022-10-14 Aldy Hernandez <aldyh@redhat.com>
10955 * value-range.cc (frange::set): Drop -0.0 for !HONOR_SIGNED_ZEROS.
10957 2022-10-14 Aldy Hernandez <aldyh@redhat.com>
10959 * gimple-range-op.cc
10960 (gimple_range_op_handler::maybe_builtin_call): Add
10961 CFN_BUILT_IN_SIGNBIT[FL]* entries.
10963 2022-10-14 Richard Biener <rguenther@suse.de>
10965 PR tree-optimization/107254
10966 * tree-vect-slp.cc (vect_slp_analyze_node_operations_1):
10967 For permutes also analyze live lanes.
10968 (vect_schedule_slp_node): For permutes also code generate
10969 live lane extracts.
10971 2022-10-14 Eric Botcazou <ebotcazou@adacore.com>
10974 * config/sparc/sparc.cc (sparc_expand_prologue): Emit a frame
10975 blockage for leaf functions.
10976 (sparc_flat_expand_prologue): Emit frame instead of full blockage.
10977 (sparc_expand_epilogue): Emit a frame blockage for leaf functions.
10978 (sparc_flat_expand_epilogue): Emit frame instead of full blockage.
10980 2022-10-14 Jakub Jelinek <jakub@redhat.com>
10982 * tree-core.h (enum tree_index): Add TI_BFLOAT16_TYPE.
10983 * tree.h (bfloat16_type_node): Define.
10984 * tree.cc (excess_precision_type): Promote bfloat16_type_mode
10985 like float16_type_mode.
10986 (build_common_tree_nodes): Initialize bfloat16_type_node if
10987 BFmode is supported.
10988 * expmed.h (maybe_expand_shift): Declare.
10989 * expmed.cc (maybe_expand_shift): No longer static.
10990 * expr.cc (convert_mode_scalar): Don't ICE on BF -> HF or HF -> BF
10991 conversions. If there is no optab, handle BF -> {DF,XF,TF,HF}
10992 conversions as separate BF -> SF -> {DF,XF,TF,HF} conversions, add
10993 -ffast-math generic implementation for BF -> SF and SF -> BF
10995 * builtin-types.def (BT_BFLOAT16, BT_FN_BFLOAT16_CONST_STRING): New.
10996 * builtins.def (BUILT_IN_NANSF16B): New builtin.
10997 * fold-const-call.cc (fold_const_call): Handle CFN_BUILT_IN_NANSF16B.
10998 * config/i386/i386.cc (classify_argument): Handle E_BCmode.
10999 (ix86_libgcc_floating_mode_supported_p): Also return true for BFmode
11001 (ix86_mangle_type): Mangle BFmode as DF16b.
11002 (ix86_invalid_conversion, ix86_invalid_unary_op,
11003 ix86_invalid_binary_op): Remove.
11004 (TARGET_INVALID_CONVERSION, TARGET_INVALID_UNARY_OP,
11005 TARGET_INVALID_BINARY_OP): Don't redefine.
11006 * config/i386/i386-builtins.cc (ix86_bf16_type_node): Remove.
11007 (ix86_register_bf16_builtin_type): Use bfloat16_type_node rather than
11008 ix86_bf16_type_node, only create it if still NULL.
11009 * config/i386/i386-builtin-types.def (BFLOAT16): Likewise.
11010 * config/i386/i386.md (cbranchbf4, cstorebf4): New expanders.
11012 2022-10-14 Jakub Jelinek <jakub@redhat.com>
11016 * doc/invoke.texi (-fexcess-precision=standard): Mention that the
11017 option now also works in C++.
11019 2022-10-13 Eric Botcazou <ebotcazou@adacore.com>
11021 * gimple-ssa-warn-access.cc (pass_waccess::check_call): Return
11022 early for calls made from thunks.
11024 2022-10-13 Eric Botcazou <ebotcazou@adacore.com>
11026 * expr.cc (emit_group_stote): Fix handling of modes of different
11027 sizes for big-endian targets in latest change and add commentary.
11029 2022-10-13 Martin Liska <mliska@suse.cz>
11031 * output.h (assemble_vtv_preinit_initializer): Remove.
11032 * varasm.cc (assemble_vtv_preinit_initializer): Remove.
11034 2022-10-13 Andrew MacLeod <amacleod@redhat.com>
11036 PR tree-optimization/102540
11037 PR tree-optimization/102872
11038 * gimple-range-cache.cc (ranger_cache::fill_block_cache):
11039 Handle partial equivs.
11040 (ranger_cache::range_from_dom): Cleanup dump output.
11042 2022-10-13 Andrew MacLeod <amacleod@redhat.com>
11044 * range-op.cc (operator_cast::lhs_op1_relation): New.
11045 (operator_bitwise_and::lhs_op1_relation): New.
11047 2022-10-13 Andrew MacLeod <amacleod@redhat.com>
11049 * gimple-range-cache.cc (ranger_cache::fill_block_cache): Use
11051 * value-relation.cc
11052 (equiv_relation_iterator::equiv_relation_iterator): New.
11053 (equiv_relation_iterator::next): New.
11054 (equiv_relation_iterator::get_name): New.
11055 * value-relation.h (class relation_oracle): Privatize some methods.
11056 (class equiv_relation_iterator): New.
11057 (FOR_EACH_EQUIVALENCE): New.
11058 (FOR_EACH_PARTIAL_EQUIV): New.
11059 (FOR_EACH_PARTIAL_AND_FULL_EQUIV): New.
11061 2022-10-13 Andrew MacLeod <amacleod@redhat.com>
11063 * value-relation.cc (equiv_chain::dump): Don't print empty
11065 (equiv_oracle::equiv_oracle): Allocate a partial equiv table.
11066 (equiv_oracle::~equiv_oracle): Release the partial equiv table.
11067 (equiv_oracle::add_partial_equiv): New.
11068 (equiv_oracle::partial_equiv_set): New.
11069 (equiv_oracle::partial_equiv): New.
11070 (equiv_oracle::query_relation): Check for partial equivs too.
11071 (equiv_oracle::dump): Also dump partial equivs.
11072 (dom_oracle::register_relation): Handle partial equivs.
11073 (dom_oracle::query_relation): Check for partial equivs.
11074 * value-relation.h (enum relation_kind_t): Add partial equivs.
11075 (relation_partial_equiv_p): New.
11076 (relation_equiv_p): New.
11077 (class pe_slice): New.
11078 (class equiv_oracle): Add prototypes.
11083 2022-10-13 Richard Biener <rguenther@suse.de>
11085 PR tree-optimization/107247
11086 * tree-vect-loop.cc (vect_create_epilog_for_reduction):
11087 Reduce multi vector SLP reduction accumulators. Check
11088 the adjusted number of accumulator vectors against
11089 one for the re-use in the epilogue.
11091 2022-10-13 Jakub Jelinek <jakub@redhat.com>
11093 * genmodes.cc (emit_mode_wider): Emit previous content of
11094 mode_wider array into mode_next array and for mode_wider
11095 emit always VOIDmode for !CLASS_HAS_WIDER_MODES_P classes,
11096 otherwise skip through modes with the same precision.
11097 * machmode.h (mode_next): Declare.
11098 (GET_MODE_NEXT_MODE): New inline function.
11099 (mode_iterator::get_next, mode_iterator::get_known_next): New
11100 function templates.
11101 (FOR_EACH_MODE_IN_CLASS): Use get_next instead of get_wider.
11102 (FOR_EACH_MODE): Use get_known_next instead of get_known_wider.
11103 (FOR_EACH_MODE_FROM): Use get_next instead of get_wider.
11104 (FOR_EACH_WIDER_MODE_FROM): Define.
11105 (FOR_EACH_NEXT_MODE): Define.
11106 * expmed.cc (emit_store_flag_1): Use FOR_EACH_WIDER_MODE_FROM
11107 instead of FOR_EACH_MODE_FROM.
11108 * optabs.cc (prepare_cmp_insn): Likewise. Remove redundant
11109 !CLASS_HAS_WIDER_MODES_P check.
11110 (prepare_float_lib_cmp): Use FOR_EACH_WIDER_MODE_FROM instead of
11111 FOR_EACH_MODE_FROM.
11112 * config/i386/i386-expand.cc (get_mode_wider_vector): Use
11113 GET_MODE_NEXT_MODE instead of GET_MODE_WIDER_MODE.
11115 2022-10-13 Wilco Dijkstra <wdijkstr@arm.com>
11118 * config/aarch64/aarch64.cc (aarch64_select_cc_mode): Allow
11119 GT/LE for merging compare with zero into AND.
11120 (aarch64_get_condition_code_1): Add CC_NZVmode support.
11121 * config/aarch64/aarch64-modes.def: Add CC_NZV.
11122 * config/aarch64/aarch64.md: Use CC_NZV in cmp+and patterns.
11124 2022-10-13 Richard Biener <rguenther@suse.de>
11126 PR tree-optimization/107160
11127 * tree-vect-loop.cc (vect_create_epilog_for_reduction):
11128 Do not register accumulator if we failed to reduce it
11129 to a single vector.
11131 2022-10-13 Aldy Hernandez <aldyh@redhat.com>
11133 * range-op-float.cc (range_operator_float::op1_op2_relation): New.
11134 (class foperator_equal): Add using.
11135 (class foperator_not_equal): Same.
11136 (class foperator_lt): Same.
11137 (class foperator_le): Same.
11138 (class foperator_gt): Same.
11139 (class foperator_ge): Same.
11140 * range-op.cc (range_op_handler::op1_op2_relation): New.
11141 * range-op.h (range_operator_float::op1_op2_relation): New.
11143 2022-10-13 Richard Biener <rguenther@suse.de>
11145 * genmatch.cc (parser::parse_c_expr): Diagnose 'return'.
11146 * match.pd: Replace 'return' statements in with expressions
11147 with appropriate variants.
11149 2022-10-13 Andre Vieira <andre.simoesdiasvieira@arm.com>
11151 PR tree-optimization/107229
11152 * tree-if-conv.cc (get_bitfield_rep): Fix bitposition calculation.
11154 2022-10-13 Andre Vieira <andre.simoesdiasvieira@arm.com>
11156 PR tree-optimization/107226
11157 * tree-vect-patterns.cc (vect_recog_bitfield_ref_pattern): Reject
11158 BITFIELD_REF's with non integral typed first operands.
11160 2022-10-13 Joseph Myers <joseph@codesourcery.com>
11162 * ginclude/float.h (FLT_IS_IEC_60559, DBL_IS_IEC_60559)
11163 (LDBL_IS_IEC_60559): Update comment.
11165 2022-10-12 Aldy Hernandez <aldyh@redhat.com>
11167 * range-op-float.cc (class foperator_negate): New.
11168 (floating_op_table::floating_op_table): Add NEGATE_EXPR
11169 (range_op_float_tests): Add negate tests.
11171 2022-10-12 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
11173 * config/riscv/riscv-vector-builtins.h: Remove unused macro.
11175 2022-10-12 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
11177 * config/riscv/riscv-vector-builtins.cc (DEF_RVV_TYPE): Apply
11179 (add_vector_type_attribute): Ditto.
11180 * config/riscv/riscv-vector-builtins.def (DEF_RVV_TYPE): Apply
11182 * config/riscv/riscv-vector-builtins.h (DEF_RVV_TYPE): Apply
11185 2022-10-12 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
11187 * config/riscv/riscv-vector-builtins.cc (builtin_types): Redefine
11189 (build_const_pointer): New function.
11190 (register_builtin_type): Ditto.
11191 (DEF_RVV_TYPE): Simplify macro.
11192 (register_vector_type): Refine implementation.
11193 * config/riscv/riscv-vector-builtins.h (rvv_builtin_types_t): New.
11195 2022-10-12 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
11197 * config/riscv/riscv-vector-builtins.h (class rvv_switcher): Move to
11199 * config/riscv/riscv-vector-builtins.cc (class rvv_switcher):
11202 2022-10-12 Cui,Lili <lili.cui@intel.com>
11204 * config/i386/driver-i386.cc (host_detect_local_cpu):
11205 Move sapphirerapids out of AVX512_VP2INTERSECT.
11206 * config/i386/i386.h: Remove AVX512_VP2INTERSECT from PTA_SAPPHIRERAPIDS
11207 * doc/invoke.texi: Remove AVX512_VP2INTERSECT from SAPPHIRERAPIDS
11209 2022-10-12 Martin Liska <mliska@suse.cz>
11211 * gcov-io.cc (gcov_write_summary): Rename to ...
11212 (gcov_write_object_summary): ... this.
11213 * gcov-io.h (GCOV_TAG_OBJECT_SUMMARY_LENGTH): Rename from ...
11214 (GCOV_TAG_SUMMARY_LENGTH): ... this.
11216 2022-10-12 Martin Liska <mliska@suse.cz>
11218 * configure: Regenerate.
11220 2022-10-12 Aldy Hernandez <aldyh@redhat.com>
11222 * range-op-float.cc (frange_float): New.
11223 (range_op_float_tests): New.
11224 * range-op.cc (range_op_tests): Call range_op_float_tests.
11226 2022-10-12 Aldy Hernandez <aldyh@redhat.com>
11228 * value-range.h (frange::nan_signbit_p): New.
11230 2022-10-12 Aldy Hernandez <aldyh@redhat.com>
11232 * value-range.cc (frange::set_nonnegative): Pass bool to
11234 * value-range.h: Disallow conversion to bool in update_nan().
11236 2022-10-12 Aldy Hernandez <aldyh@redhat.com>
11238 * value-range.h (frange::frange): Add constructor taking type.
11240 2022-10-12 Aldy Hernandez <aldyh@redhat.com>
11242 * range-op-float.cc: Add relation_kind = VREL_VARYING to all
11245 2022-10-11 Aldy Hernandez <aldyh@redhat.com>
11247 * gimple-range-gori.cc (gori_compute::logical_combine): Avoid
11248 calling tracer.trailer().
11250 2022-10-11 Jakub Jelinek <jakub@redhat.com>
11253 * config/i386/i386.md (*notxor<mode>_1): Use MASK_REG_P (x) instead of
11254 MASK_REGNO_P (REGNO (x)).
11256 2022-10-11 Aldy Hernandez <aldyh@redhat.com>
11258 * range-op-float.cc (class foperator_abs): New.
11259 (floating_op_table::floating_op_table): Add ABS_EXPR entry.
11261 2022-10-11 Aldy Hernandez <aldyh@redhat.com>
11263 * range-op-float.cc (foperator_unordered_le::op1_range): New.
11264 (foperator_unordered_le::op2_range): New.
11265 (foperator_unordered_gt::op1_range): New.
11266 (foperator_unordered_gt::op2_range): New.
11267 (foperator_unordered_ge::op1_range): New.
11268 (foperator_unordered_ge::op2_range): New.
11269 (foperator_unordered_equal::op1_range): New.
11271 2022-10-11 Aldy Hernandez <aldyh@redhat.com>
11273 * range-op-float.cc (class foperator_unordered_lt): New.
11274 (class foperator_relop_unknown): Remove
11275 (class foperator_unordered_le): New.
11276 (class foperator_unordered_gt): New.
11277 (class foperator_unordered_ge): New.
11278 (class foperator_unordered_equal): New.
11279 (floating_op_table::floating_op_table): Replace all UN_EXPR
11280 entries with their appropriate fop_unordered_* counterpart.
11282 2022-10-11 Aldy Hernandez <aldyh@redhat.com>
11284 * range-op.cc (operator_equal::op1_range): Move BRS_TRUE case up.
11285 (operator_lt::op2_range): Same.
11286 (operator_le::op2_range): Same.
11287 (operator_gt::op2_range): Same.
11288 (operator_ge::op2_range): Same.
11290 2022-10-11 Richard Biener <rguenther@suse.de>
11292 PR tree-optimization/107212
11293 * tree-vect-loop.cc (vectorizable_reduction): Make sure to
11294 set STMT_VINFO_REDUC_DEF for all live lanes in a SLP
11296 (vectorizable_live_operation): Do not pun to the SLP
11297 node representative for reduction epilogue generation.
11299 2022-10-11 Andrew Stubbs <ams@codesourcery.com>
11301 * config/gcn/gcn-valu.md (neg<mode>2): New define_expand.
11303 2022-10-11 Andrew Stubbs <ams@codesourcery.com>
11305 * config/gcn/gcn-valu.md (vec_init<V_ALL:mode><V_ALL_ALT:mode>): New.
11306 * config/gcn/gcn.cc (GEN_VN): Add andvNsi3, subvNsi3.
11307 (GEN_VNM): Add gathervNm_expr.
11308 (GEN_VN_NOEXEC): Add vec_seriesvNsi.
11309 (gcn_expand_vector_init): Add initialization of vectors from smaller
11312 2022-10-11 Andrew Stubbs <ams@codesourcery.com>
11314 * config/gcn/gcn-protos.h (get_exec): Add prototypes for two variants.
11315 * config/gcn/gcn-valu.md
11316 (vec_extract<V_ALL:mode><V_ALL_ALT:mode>): New define_expand.
11317 * config/gcn/gcn.cc (get_exec): Export the existing function. Add a
11318 new overload variant.
11320 2022-10-11 Andrew Stubbs <ams@codesourcery.com>
11322 * config/gcn/gcn-valu.md
11323 (<cvt_name><VCVT_MODE:mode><VCVT_FMODE:mode>2<exec>): Use MODE_VF.
11324 (<cvt_name><VCVT_FMODE:mode><VCVT_IMODE:mode>2<exec>): Likewise.
11325 * config/gcn/gcn.h (MODE_VF): New macro.
11327 2022-10-11 Andrew Stubbs <ams@codesourcery.com>
11329 * config/gcn/gcn-modes.def (VECTOR_MODE): Add new modes
11330 V32QI, V32HI, V32SI, V32DI, V32TI, V32HF, V32SF, V32DF,
11331 V16QI, V16HI, V16SI, V16DI, V16TI, V16HF, V16SF, V16DF,
11332 V8QI, V8HI, V8SI, V8DI, V8TI, V8HF, V8SF, V8DF,
11333 V4QI, V4HI, V4SI, V4DI, V4TI, V4HF, V4SF, V4DF,
11334 V2QI, V2HI, V2SI, V2DI, V2TI, V2HF, V2SF, V2DF.
11335 (ADJUST_ALIGNMENT): Likewise.
11336 * config/gcn/gcn-protos.h (gcn_full_exec): Delete.
11337 (gcn_full_exec_reg): Delete.
11338 (gcn_scalar_exec): Delete.
11339 (gcn_scalar_exec_reg): Delete.
11340 (vgpr_1reg_mode_p): Use inner mode to identify vector registers.
11341 (vgpr_2reg_mode_p): Likewise.
11342 (vgpr_vector_mode_p): Use VECTOR_MODE_P.
11343 * config/gcn/gcn-valu.md (V_QI, V_HI, V_HF, V_SI, V_SF, V_DI, V_DF,
11344 V_QIHI, V_1REG, V_INT_1REG, V_INT_1REG_ALT, V_FP_1REG, V_2REG, V_noQI,
11345 V_noHI, V_INT_noQI, V_INT_noHI, V_ALL, V_ALL_ALT, V_INT, V_FP):
11346 Add additional vector modes.
11347 (V64_SI, V64_DI, V64_ALL, V64_FP): New iterators.
11348 (scalar_mode, SCALAR_MODE, vnsi, VnSI, vndi, VnDI, sdwa):
11349 Add additional vector mode mappings.
11350 (mov<mode>): Implement vector length conversions.
11351 (ldexp<mode>3<exec>): Use VnSI.
11352 (frexp<mode>_exp2<exec>): Likewise.
11353 (VCVT_MODE, VCVT_FMODE, VCVT_IMODE): Add additional vector modes.
11354 (reduc_<reduc_op>_scal_<mode>): Use V64_ALL.
11355 (fold_left_plus_<mode>): Use V64_FP.
11356 (*<reduc_op>_dpp_shr_<mode>): Use V64_1REG.
11357 (*<reduc_op>_dpp_shr_<mode>): Use V64_DI.
11358 (*plus_carry_dpp_shr_<mode>): Use V64_INT_1REG.
11359 (*plus_carry_in_dpp_shr_<mode>): Use V64_SI.
11360 (*plus_carry_dpp_shr_<mode>): Use V64_DI.
11361 (mov_from_lane63_<mode>): Use V64_2REG.
11362 * config/gcn/gcn.cc (VnMODE): New function.
11363 (gcn_can_change_mode_class): Support multiple vector sizes.
11364 (gcn_modes_tieable_p): Likewise.
11365 (gcn_operand_part): Likewise.
11366 (gcn_scalar_exec): Delete function.
11367 (gcn_scalar_exec_reg): Delete function.
11368 (gcn_full_exec): Delete function.
11369 (gcn_full_exec_reg): Delete function.
11370 (gcn_inline_fp_constant_p): Support multiple vector sizes.
11371 (gcn_fp_constant_p): Likewise.
11373 (GEN_VN_NOEXEC): New macro.
11374 (GEN_VNM_NOEXEC): New macro.
11375 (GEN_VN): New macro.
11376 (GEN_VNM): New macro.
11377 (GET_VN_FN): New macro.
11378 (CODE_FOR): New macro.
11379 (CODE_FOR_OP): New macro.
11380 (gen_mov_with_exec): Delete function.
11381 (gen_duplicate_load): Delete function.
11382 (gcn_expand_vector_init): Support multiple vector sizes.
11383 (strided_constant): Likewise.
11384 (gcn_addr_space_legitimize_address): Likewise.
11385 (gcn_expand_scalar_to_vector_address): Likewise.
11386 (gcn_expand_scaled_offsets): Likewise.
11387 (gcn_secondary_reload): Likewise.
11388 (gcn_valid_cvt_p): Likewise.
11389 (gcn_expand_builtin_1): Likewise.
11390 (gcn_make_vec_perm_address): Likewise.
11391 (gcn_vectorize_vec_perm_const): Likewise.
11392 (gcn_vector_mode_supported_p): Likewise.
11393 (gcn_autovectorize_vector_modes): New hook.
11394 (gcn_related_vector_mode): Support multiple vector sizes.
11395 (gcn_expand_dpp_shr_insn): Add FIXME comment.
11396 (gcn_md_reorg): Support multiple vector sizes.
11397 (print_reg): Likewise.
11398 (print_operand): Likewise.
11399 (TARGET_VECTORIZE_AUTOVECTORIZE_VECTOR_MODES): New hook.
11401 2022-10-11 Andre Vieira <andre.simoesdiasvieira@arm.com>
11403 * tree-if-conv.cc (if_convertible_loop_p_1): Move ordering of loop bb's from
11405 (tree_if_conversion): ... to here. Also call bitfield lowering when
11407 (version_loop_for_if_conversion): Adapt to enable loop versioning when we only
11408 need to lower bitfields.
11409 (ifcvt_split_critical_edges): Relax condition of expected loop form as this is
11411 (get_bitfield_rep): New function.
11412 (lower_bitfield): Likewise.
11413 (bitfields_to_lower_p): Likewise.
11414 (need_to_lower_bitfields): New global boolean.
11415 (need_to_ifcvt): Likewise.
11416 * tree-vect-data-refs.cc (vect_find_stmt_data_reference): Improve diagnostic
11418 * tree-vect-patterns.cc (vect_recog_temp_ssa_var): Add default value for last
11420 (vect_recog_bitfield_ref_pattern): New.
11421 (vect_recog_bit_insert_pattern): New.
11423 2022-10-11 liuhongt <hongtao.liu@intel.com>
11426 * config/i386/i386.md (*notxor<mode>_1): New post_reload
11427 define_insn_and_split.
11428 (*notxorqi_1): Ditto.
11430 2022-10-11 Aldy Hernandez <aldyh@redhat.com>
11432 PR tree-optimization/107195
11433 * value-range.cc (irange::set_range_from_nonzero_bits): Set range
11434 to [0,0] when nonzero mask is 0.
11436 2022-10-11 Olivier Hainque <hainque@adacore.com>
11438 * configure: Regenerate.
11440 2022-10-11 Olivier Hainque <hainque@adacore.com>
11442 * config.gcc (*vxworks*): Add t-slibgcc fragment
11445 2022-10-11 Olivier Hainque <hainque@adacore.com>
11447 * config/vxworks.h (VX_LGCC_EH_SO0, VX_LGCC_EH_SO1): New
11449 (VXWORKS_LIBGCC_SPEC): Use them and document.
11451 2022-10-11 Martin Liska <mliska@suse.cz>
11453 * gimple-range-op.cc: Add override keyword.
11455 2022-10-11 Eugene Rozenfeld <erozen@microsoft.com>
11458 * tree-cfg.cc (assign_discriminators): Move declaration of cur_locus_e
11461 2022-10-11 Liwei Xu <liwei.xu@intel.com>
11462 liuhongt <hongtao.liu@intel.com>
11464 PR tree-optimization/54346
11465 * match.pd: Merge the index of VCST then generates the new vec_perm.
11467 2022-10-11 Jeff Law <jeffreyalaw@gmail.com>
11469 PR rtl-optimization/107182
11470 * cfgrtl.cc (fixup_reorder_chain): When optimizing a jump to a
11471 return, clear EDGE_CROSSING on the appropriate edge.
11473 2022-10-11 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
11475 * config/riscv/riscv-vector-builtins.cc (struct vector_type_info): Move
11476 from config/riscv/riscv-vector-builtins.h.
11477 (DEF_RVV_TYPE): Change USER_NAME to NAME.
11478 (register_vector_type): Change user_name to name.
11479 * config/riscv/riscv-vector-builtins.def (DEF_RVV_TYPE): Change
11481 * config/riscv/riscv-vector-builtins.h (struct vector_type_info): Move
11482 to riscv-vector-builtins.cc.
11483 (DEF_RVV_TYPE): Change USER_NAME to NAME.
11485 2022-10-11 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
11487 * config/riscv/riscv.md: Add vsetvl instruction type.
11489 2022-10-10 Nathan Sidwell <nathan@acm.org>
11491 * common.opt (-fabi-version=): Document 18.
11492 * doc/invoke.texi (-fabi-version): Document 18.
11494 2022-10-10 Andrea Corallo <andrea.corallo@arm.com>
11497 * toplev.cc (toplev::main): Don't run self tests in case of
11500 2022-10-10 Kito Cheng <kito.cheng@sifive.com>
11502 * config/riscv/riscv-c.cc: Add newline to the end of file.
11504 2022-10-10 Aldy Hernandez <aldyh@redhat.com>
11506 * range-op-float.cc (class foperator_identity): Make members public.
11507 (class foperator_equal): Same.
11508 (class foperator_not_equal): Same.
11509 (class foperator_lt): Same.
11510 (class foperator_le): Same.
11511 (class foperator_gt): Same.
11512 (class foperator_ge): Same.
11513 (class foperator_unordered): Same.
11514 (class foperator_ordered): Same.
11516 2022-10-10 Aldy Hernandez <aldyh@redhat.com>
11518 * value-range.h (frange::maybe_isnan): New.
11520 2022-10-10 Aldy Hernandez <aldyh@redhat.com>
11522 * range-op-float.cc (foperator_not_equal::op1_range): Set NAN on
11523 TRUE side for x != x.
11525 2022-10-10 Aldy Hernandez <aldyh@redhat.com>
11527 * range-op-float.cc (foperator_unordered::op1_range): Set NAN when
11528 operands are equal and result is TRUE.
11530 2022-10-10 Aldy Hernandez <aldyh@redhat.com>
11532 * range.h (range_true): Return int_range<2>.
11533 (range_false): Same.
11534 (range_true_and_false): Same.
11536 2022-10-10 Aldy Hernandez <aldyh@redhat.com>
11538 * gimple-range-op.cc: Add op1_range entry for __builtin_signbit.
11540 2022-10-10 liuhongt <hongtao.liu@intel.com>
11543 * config/i386/i386.md (lrint<MODEF:mode><SWI48:mode>2): Swap
11544 predicate of operands[0] and operands[1].
11546 2022-10-10 Claudiu Zissulescu <claziss@gmail.com>
11548 * common/config/arc/arc-common.cc (arc_option_optimization_table):
11549 Remove Rcq and Rcw options.
11550 * config/arc/arc.opt (mRcq): Ignore option, preserve it for
11551 backwards compatibility.
11553 * doc/invoke.texi (mRcw, mRcq): Update document.
11555 2022-10-10 Claudiu Zissulescu <claziss@gmail.com>
11557 * config/arc/arc.cc (arc_check_short_reg_p): New function.
11558 (arc_address_cost): Replace satisfies_constraint_Rcq with the
11559 above new function.
11560 (arc_output_addsi): Likewise.
11561 (split_addsi): Likewise.
11562 (split_subsi): Likewise.
11563 * config/arc/arc.md (movqi_insn): Remove Rcq constraint.
11564 (movhi_insn): Likewise.
11565 (movsi_insn): Likewise.
11566 (tst_movb): Likewise.
11568 (tst_bitfield): Likewise.
11569 (abssi2): Likewise.
11570 (addsi3_mixed): Likewise.
11571 (mulhisi3_reg): Likewise.
11572 (umulhisi3_reg): Likewise.
11573 (mulsi_600): Likewise.
11575 (subsi3_insn): Likewise.
11576 (bicsi3_insn): Likewise.
11577 (xorsi3): Likewise.
11578 (negsi2): Likewise.
11579 (one_cmplsi2): Likewise.
11580 (lshrsi3_insn): Likewise.
11581 (cmpsi_cc_insn_mixed): Likewise.
11582 (cmpsi_cc_zn_insn): Likewise.
11584 (cmpsi_cc_z_insn): Likewise.
11585 (cmpsi_cc_c_insn): Likewise.
11586 (indirect_jump): Likewise.
11587 (casesi_jump): Likewise.
11588 (call_i): Likewise.
11589 (call_value_i): Likewise.
11591 (abssf2): Likewise.
11592 (ashlsi2_cnt1): Likewise.
11593 (lshrsi3_cnt1): Likewise.
11594 (ashrsi3_cnt1): Likewise.
11595 * config/arc/constraints.md (Rcq): Remove.
11597 2022-10-10 Claudiu Zissulescu <claziss@gmail.com>
11599 * config/arc/arc.md (smaxsi3): Remove Rcw.
11600 (sminsi3): Likewise.
11601 (addsi3_mixed): Likewise.
11602 (add_f_2): Likewise.
11603 (subsi3_insn): Likewise.
11609 (bset_insn): Likewise.
11610 (bxor_insn): Likewise.
11611 (bclr_insn): Likewise.
11612 (bmsk_insn): Likewise.
11613 (bicsi3_insn): Likewise.
11614 (xorsi3): Likewise.
11615 (negsi2): Likewise.
11616 (lshrsi3_insn): Likewise.
11617 (abssf2): Likewise.
11618 (negsf2): Likewise.
11619 * config/arc/constraints.md(Rcw): Remove it.
11621 2022-10-10 Claudiu Zissulescu <claziss@gmail.com>
11623 * config/arc/arc.md(mulsi3_700): Remove Rcr.
11624 (mulsi3_highpart): Likewise.
11625 (umulsi3_highpart_i): Likewise.
11626 (umulsi3_highpart_int): Likewise.
11629 * config/arc/constraints.md (Rcr): Remove it.
11631 2022-10-10 Claudiu Zissulescu <claziss@gmail.com>
11633 * config/arc/arc.cc (arc_save_callee_enter): Use negative offsets.
11635 2022-10-09 Dimitar Dimitrov <dimitar@dinux.eu>
11638 * config/pru/pru-protos.h (pru_noteq_condition): New
11639 function declaration.
11640 * config/pru/pru.cc (pru_noteq_condition): New function.
11641 * config/pru/pru.md (cbranchdi4): Define new pattern.
11643 2022-10-09 Dimitar Dimitrov <dimitar@dinux.eu>
11645 * config/pru/pru.md (lshrdi3): New expand pattern.
11648 2022-10-09 YunQiang Su <yunqiang.su@cipunited.com>
11650 * config.gcc: set with_arch to default_mips_arch if no defined.
11651 * config/mips/driver-native.cc (host_detect_local_cpu):
11652 try getauxval(AT_BASE_PLATFORM) and _MIPS_ARCH, too.
11653 pass -mnan=2008 if __mips_nan2008__ is defined.
11654 * config.in: define HAVE_SYS_AUXV_H and HAVE_GETAUXVAL.
11655 * configure.ac: detect sys/auxv.h and getauxval.
11656 * configure: regenerated.
11658 2022-10-07 Eugene Rozenfeld <erozen@microsoft.com>
11660 * tree-cfg.cc (assign_discriminators): Set discriminators for call stmts
11661 on the same line within the same basic block.
11663 2022-10-07 Qing Zhao <qing.zhao@oracle.com>
11665 PR tree-optimization/101836
11666 * tree-object-size.cc (addr_object_size): Use array_at_struct_end_p
11667 to determine a flexible array member reference.
11669 2022-10-07 Qing Zhao <qing.zhao@oracle.com>
11671 * doc/extend.texi: Document strict_flex_array attribute.
11672 * doc/invoke.texi: Document -fstrict-flex-arrays[=n] option.
11673 * print-tree.cc (print_node): Print new bit decl_not_flexarray.
11674 * tree-core.h (struct tree_decl_common): New bit field
11675 decl_not_flexarray.
11676 * tree-streamer-in.cc (unpack_ts_decl_common_value_fields): Stream
11677 in new bit decl_not_flexarray.
11678 * tree-streamer-out.cc (pack_ts_decl_common_value_fields): Stream
11679 out new bit decl_not_flexarray.
11680 * tree.cc (array_at_struct_end_p): Update it with the new bit field
11681 decl_not_flexarray.
11682 * tree.h (DECL_NOT_FLEXARRAY): New flag.
11684 2022-10-07 Olivier Hainque <hainque@adacore.com>
11686 * config/vxworks/_vxworks-versions.h: Use OS specific
11687 paths in #include of version.h.
11689 2022-10-07 Martin Liska <mliska@suse.cz>
11691 * opts.cc (finish_options): Print sorry message only
11692 for -flive-patching=inline-only-static.
11694 2022-10-07 Jason Merrill <jason@redhat.com>
11696 * gimplify.cc (gimplify_modify_expr_rhs): Don't optimize
11697 x = *(A*)&<expr> to x = <expr> for a TREE_ADDRESSABLE type.
11699 2022-10-07 Richard Biener <rguenther@suse.de>
11701 PR tree-optimization/107153
11702 * tree-cfg.cc (gimple_duplicate_sese_tail): Do not update
11704 * tree-parloops.cc (gen_parallel_loop): Update SSA form
11705 after to-exit-first transform, no PHI insertion is necessary.
11706 (pass_parallelize_loops::execute): Force re-write of the
11707 virtual operand SSA web.
11709 2022-10-07 Jonathan Wakely <jwakely@redhat.com>
11711 * value-range.cc (irange::irange_contains_p): Fix comment typo.
11713 2022-10-07 Eric Botcazou <ebotcazou@adacore.com>
11715 * function.cc (thread_prologue_and_epilogue_insns): Update only
11716 entry and exit blocks when not optimizing. Remove dead statement.
11718 2022-10-07 Aldy Hernandez <aldyh@redhat.com>
11720 * value-range.cc (irange::irange_set): Convert nonzero mask to
11722 (irange::irange_set_anti_range): Same.
11723 (irange::set): Same.
11724 (irange::verify_range): Same.
11725 (irange::contains_p): Same.
11726 (irange::invert): Same.
11727 (irange::set_range_from_nonzero_bits): Same.
11728 (irange::set_nonzero_bits): Same.
11729 (mask_to_wi): Same.
11730 (irange::intersect_nonzero_bits): Same.
11731 (irange::union_nonzero_bits): Same.
11732 * value-range.h (irange::varying_compatible_p): Same.
11735 (irange::set_undefined): Same.
11736 (irange::set_varying): Same.
11738 2022-10-07 Martin Liska <mliska@suse.cz>
11740 * config/i386/i386-protos.h (ix86_binary_operator_ok): Add array
11741 size to function parameter.
11742 (ix86_unary_operator_ok): Likewise.
11744 2022-10-07 Martin Liska <mliska@suse.cz>
11746 * auto-profile.cc (get_inline_stack): Remove unused variable.
11748 2022-10-07 Jakub Jelinek <jakub@redhat.com>
11750 * ipa-prop.h (ipa_constant_data): Fix comment typo.
11751 * value-range.cc (irange::irange_contains_p): Likewise.
11752 * value-relation.cc (dom_oracle::set_one_relation): Likewise.
11753 * gimple-predicate-analysis.cc (predicate::simplify_4): Likewise.
11754 * tree-inline.cc (remap_ssa_name): Likewise.
11756 2022-10-07 Jakub Jelinek <jakub@redhat.com>
11758 * attribs.h (is_attribute_namespace_p): New inline function.
11759 (lookup_attribute): Document meaning of ATTR_NS equal to "".
11760 * attribs.cc (remove_attribute): Use is_attribute_namespace_p.
11761 (private_lookup_attribute): For ATTR_NS "" match either standard
11762 attribute or "gnu" namespace one.
11764 2022-10-07 Olivier Hainque <hainque@adacore.com>
11766 * config/vxworks.h (DWARF_VERSION_DEFAULT): Adjust from
11767 4 to 3 for VxWorks >= 7.
11769 2022-10-07 Olivier Hainque <hainque@adacore.com>
11771 * defaults.h (DWARF_DEFAULT_VERSION): Define if not
11773 * common.opt (gdwarf-): Use it.
11774 * doc/tm.texi.in (DWARF_DEFAULT_VERSION): Document.
11775 * doc/tm.texi: Update accordingly.
11776 * config/vxworks.h (DWARF_DEFAULT_VERSION): Redefine.
11777 * config/vxworks.cc: Remove code setting dwarf_version, now
11778 handled by the DWARF_DEFAULT_VERSION redefinition.
11780 2022-10-07 Olivier Hainque <hainque@adacore.com>
11782 * ginclude/stddef.h: #undef offsetof before #define.
11784 2022-10-06 Aldy Hernandez <aldyh@redhat.com>
11786 PR tree-optimization/107170
11787 * vr-values.cc (vr_values::range_of_expr): Do not die on
11790 2022-10-06 Joseph Myers <joseph@codesourcery.com>
11792 * doc/invoke.texi (-fno-asm): Update description of effects on
11795 2022-10-06 Eric Botcazou <ebotcazou@adacore.com>
11797 * gimple-loop-jam.cc (tree_loop_unroll_and_jam): Bail out for a self
11798 dependency that is a write-after-write if the access function is not
11799 affine or constant.
11801 2022-10-06 Eric Botcazou <ebotcazou@adacore.com>
11803 * df-scan.cc (df_ref_create_structure): Minor cleanup.
11805 2022-10-06 Richard Biener <rguenther@suse.de>
11807 PR middle-end/107115
11808 * expr.cc (store_expr): Check mems_same_for_tbaa_p before
11809 eliding a seemingly redundant store.
11811 2022-10-06 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
11813 * config/aarch64/atomics.md (*atomic_load<ALLX:mode>_zext<SD_HSDI:mode>):
11816 2022-10-06 Philipp Tomsich <philipp.tomsich@vrull.eu>
11818 * config/aarch64/aarch64-cores.def (AARCH64_CORE): Update
11819 Ampere-1 core entry.
11821 2022-10-06 Philipp Tomsich <philipp.tomsich@vrull.eu>
11823 * config/aarch64/driver-aarch64.cc (readline): Fix off-by-one.
11825 2022-10-06 Richard Biener <rguenther@suse.de>
11827 PR tree-optimization/107107
11828 * tree-ssa-sccvn.cc (visit_reference_op_store): Do not
11829 affect value-numbering when doing the tail merging
11830 MODIFY_EXPR lookup.
11832 2022-10-06 Claudiu Zissulescu <claziss@synopsys.com>
11834 * config/arc/linux.h (LINK_SPEC): Remove max-page-size and
11837 2022-10-06 Jakub Jelinek <jakub@redhat.com>
11840 * internal-fn.def (ASSUME): New internal function.
11841 * internal-fn.h (expand_ASSUME): Declare.
11842 * internal-fn.cc (expand_ASSUME): Define.
11843 * gimplify.cc (gimplify_call_expr): Gimplify IFN_ASSUME.
11844 * fold-const.h (simple_condition_p): Declare.
11845 * fold-const.cc (simple_operand_p_2): Rename to ...
11846 (simple_condition_p): ... this. Remove forward declaration.
11847 No longer static. Adjust function comment and fix a typo in it.
11848 Adjust recursive call.
11849 (simple_operand_p): Adjust function comment.
11850 (fold_truth_andor): Adjust simple_operand_p_2 callers to call
11851 simple_condition_p.
11852 * doc/extend.texi: Document assume attribute. Move fallthrough
11853 attribute example to its section.
11855 2022-10-06 Stefan Schulze Frielinghaus <stefansf@linux.ibm.com>
11857 PR rtl-optimization/107088
11858 * cselib.cc (new_cselib_val): Skip BImode while keeping track of
11859 subvalue relations.
11861 2022-10-06 Aldy Hernandez <aldyh@redhat.com>
11863 * value-range.cc (frange::set): Call set_nan unconditionally.
11864 (range_tests_nan): Adjust tests.
11865 (range_tests_signed_zeros): Same.
11866 (range_tests_floats): Same.
11867 * value-range.h (frange::update_nan): Guard with HONOR_NANS.
11868 (frange::set_nan): Set undefined if !HONOR_NANS.
11870 2022-10-06 Aldy Hernandez <aldyh@redhat.com>
11872 * range-op-float.cc (foperator_lt::fold_range): Remove extra check
11873 to finite_operands_p.
11874 (foperator_le::fold_range): Same.
11875 (foperator_gt::fold_range): Same.
11876 (foperator_ge::fold_range): Same.
11878 2022-10-06 Aldy Hernandez <aldyh@redhat.com>
11880 * value-range-pretty-print.cc (vrange_printer::print_real_value):
11881 Avoid printing INF and NAN twice.
11883 2022-10-05 Segher Boessenkool <segher@kernel.crashing.org>
11885 * config/rs6000/constraints.md (wD): Delete.
11886 * doc/md.texi (Machine Constraints): Adjust.
11888 2022-10-05 Segher Boessenkool <segher@kernel.crashing.org>
11890 * config/rs6000/vsx.md (vsx_extract_<mode>): Replace define_insn by a
11891 define_expand. Split the contents to...
11892 (*vsx_extract_<mode>_0): ... this. Rewrite.
11893 (*vsx_extract_<mode>_1): ... and this. Rewrite.
11895 2022-10-05 Segher Boessenkool <segher@kernel.crashing.org>
11897 * config/rs6000/vsx.md (*vsx_extract_<mode>_store): Use "n" instead of
11900 2022-10-05 David Malcolm <dmalcolm@redhat.com>
11903 * doc/analyzer.texi (__analyzer_get_unknown_ptr): Document.
11905 2022-10-05 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
11907 * config.gcc: Add riscv_vector.h.
11908 * config/riscv/riscv-builtins.cc: Add RVV builtin types support.
11909 * config/riscv/riscv-c.cc (riscv_pragma_intrinsic): New function.
11910 (riscv_register_pragmas): Ditto.
11911 * config/riscv/riscv-protos.h (riscv_register_pragmas): Ditto.
11912 (init_builtins): Move declaration from riscv-vector-builtins.h to riscv-protos.h.
11913 (mangle_builtin_type): Ditto.
11914 (verify_type_context): Ditto.
11915 (handle_pragma_vector): New function.
11916 * config/riscv/riscv-vector-builtins.cc (GTY): New variable.
11917 (register_vector_type): New function.
11918 (init_builtins): Add RVV builtin types support.
11919 (handle_pragma_vector): New function.
11920 * config/riscv/riscv-vector-builtins.h (GCC_RISCV_V_BUILTINS_H): Change
11921 name according to file name.
11922 (GCC_RISCV_VECTOR_BUILTINS_H): Ditto.
11923 (init_builtins): Remove declaration in riscv-vector-builtins.h.
11924 (mangle_builtin_type): Ditto.
11925 (verify_type_context): Ditto.
11926 * config/riscv/riscv.cc: Adjust for RVV builtin types support.
11927 * config/riscv/riscv.h (REGISTER_TARGET_PRAGMAS): New macro.
11928 * config/riscv/t-riscv: Remove redundant file including.
11929 * config/riscv/riscv_vector.h: New file.
11931 2022-10-05 Aldy Hernandez <aldyh@redhat.com>
11933 * range-op.cc (operator_cast::fold_range): Handle truncating casts
11936 2022-10-05 Martin Liska <mliska@suse.cz>
11939 * attribs.h (lookup_attribute_by_prefix): Support the attribute
11940 starting with underscore (_Noreturn).
11942 2022-10-05 Aldy Hernandez <aldyh@redhat.com>
11944 PR tree-optimization/107052
11945 * gimple-range-op.cc (cfn_popcount::fold_range): Take into account
11948 2022-10-05 Aldy Hernandez <aldyh@redhat.com>
11950 PR tree-optimization/107052
11951 * range-op.cc (operator_cast::fold_range): Set nonzero mask.
11953 2022-10-05 Eric Botcazou <ebotcazou@adacore.com>
11955 PR tree-optimization/106698
11956 * pointer-query.cc (handle_array_ref): Fix handling of low bound.
11958 2022-10-05 Vineet Gupta <vineetg@rivosinc.com>
11960 * config/riscv/riscv-c.cc (riscv_cpu_cpp_builtins):
11961 Remove __riscv_cmodel_pic, that deprecated in last version.
11963 2022-10-05 David Malcolm <dmalcolm@redhat.com>
11966 * Makefile.in (ANALYZER_OBJS): Add analyzer/call-summary.o.
11968 2022-10-04 Jakub Jelinek <jakub@redhat.com>
11970 * attribs.h (remove_attribute): Declare overload with additional
11972 (private_lookup_attribute): Declare overload with additional
11973 attr_ns and attr_ns_len arguments.
11974 (lookup_attribute): New overload with additional attr_ns argument.
11975 * attribs.cc (remove_attribute): New overload with additional
11977 (private_lookup_attribute): New overload with additional
11978 attr_ns and attr_ns_len arguments.
11980 2022-10-04 Jakub Jelinek <jakub@redhat.com>
11982 * attribs.cc (handle_ignored_attributes_option, decl_attributes,
11983 common_function_versions): Use auto_diagnostic_group.
11985 2022-10-04 Aldy Hernandez <aldyh@redhat.com>
11987 * value-range.cc (irange::set_nonzero_bits): Remove assert.
11989 2022-10-04 Richard Sandiford <richard.sandiford@arm.com>
11991 * config/aarch64/aarch64.h (AARCH64_ISA_RCPC): New macro.
11992 * config/aarch64/aarch64-arches.def (armv8.3-a): Include RCPC.
11993 * config/aarch64/aarch64-cores.def (thunderx3t110, zeus, neoverse-v1)
11994 (neoverse-512tvb, saphira): Remove RCPC from these Armv8.3-A+ cores.
11995 * config/aarch64/aarch64-c.cc (aarch64_update_cpp_builtins): Define
11996 __ARM_FEATURE_RCPC when appropriate.
11998 2022-10-04 Tobias Burnus <tobias@codesourcery.com>
12000 * doc/invoke.texi (-fopenmp): Mention C++ attribut syntax.
12001 (-fopenmp-simd): Likewise; update permitted directives.
12003 2022-10-04 Tobias Burnus <tobias@codesourcery.com>
12005 * doc/install.texi (Specific): Add missing items to bullet list.
12006 (amdgcn): Update LLVM requirements, use version not date for newlib.
12007 (nvptx): Use version not git hash for newlib.
12009 2022-10-04 Aldy Hernandez <aldyh@redhat.com>
12011 * value-range-storage.cc (irange_storage_slot::set_irange): Remove
12013 * value-range.cc (irange::irange_set): Adjust for nonzero mask
12015 (irange::irange_set_anti_range): Same.
12016 (irange::set): Same.
12017 (irange::verify_range): Same.
12018 (irange::legacy_equal_p): Same.
12019 (irange::operator==): Same.
12020 (irange::contains_p): Same.
12021 (irange::legacy_intersect): Same.
12022 (irange::legacy_union): Same.
12023 (irange::irange_single_pair_union): Call union_nonzero_bits.
12024 (irange::irange_union): Same.
12025 (irange::irange_intersect): Call intersect_nonzero_bits.
12026 (irange::intersect): Adjust for nonzero mask being a wide int.
12027 (irange::invert): Same.
12028 (irange::set_nonzero_bits): Same.
12029 (irange::get_nonzero_bits_from_range): New.
12030 (irange::set_range_from_nonzero_bits): New.
12031 (irange::get_nonzero_bits): Adjust for nonzero mask being a wide
12033 (irange::intersect_nonzero_bits): Same.
12034 (irange::union_nonzero_bits): Same.
12035 (range_tests_nonzero_bits): Remove test.
12036 * value-range.h (irange::varying_compatible_p): Adjust for nonzero
12037 mask being a wide int.
12040 (irange::set_undefined): Same.
12041 (irange::set_varying): Same.
12042 (irange::normalize_kind): Same.
12044 2022-10-04 Aldy Hernandez <aldyh@redhat.com>
12046 PR tree-optimization/107130
12047 * gimple-range-op.cc (class cfn_popcount): Call op_cfn_ffs.
12048 (class cfn_ffs): New.
12049 (gimple_range_op_handler::maybe_builtin_call): Separate out
12050 CASE_CFN_FFS into its own case.
12052 2022-10-03 Sergei Trofimovich <siarheit@google.com>
12054 * config/i386/t-i386: Add build-time dependencies against
12055 i386-builtin-types.inc to i386-builtins.o, i386-expand.o,
12058 2022-10-03 Andrew Stubbs <ams@codesourcery.com>
12060 * config/gcn/gcn-valu.md (while_ultsidi): Limit mask length using
12062 * doc/md.texi (while_ult): Document new operand 3 usage.
12063 * internal-fn.cc (expand_while_optab_fn): Set operand 3 when lhs_type
12064 maps to a non-vector mode.
12066 2022-10-03 Andrew MacLeod <amacleod@redhat.com>
12068 PR tree-optimization/107109
12069 * range-op.cc (adjust_op1_for_overflow): Don't process undefined.
12071 2022-10-03 Christophe Lyon <christophe.lyon@arm.com>
12073 * config/arm/mve.md (mve_vrev64q_m_<supf><mode>): Add early
12075 (mve_vrev64q_m_f<mode>): Likewise.
12077 2022-10-03 Aldy Hernandez <aldyh@redhat.com>
12079 * value-range.cc (irange::set_nonzero_bits): Do not pessimize range.
12080 (range_tests_nonzero_bits): New test.
12082 2022-10-03 Aldy Hernandez <aldyh@redhat.com>
12084 * value-range.cc (irange::operator==): Early bail on m_num_ranges
12087 2022-10-03 Aldy Hernandez <aldyh@redhat.com>
12089 * value-range.cc (irange::legacy_equal_p): Remove nonozero mask
12090 check when comparing VR_VARYING ranges.
12092 2022-10-03 Aldy Hernandez <aldyh@redhat.com>
12094 * ipa-prop.cc (struct ipa_vr_ggc_hash_traits): Do not compare
12095 incompatible ranges in ipa-prop.
12097 2022-10-02 Jakub Jelinek <jakub@redhat.com>
12099 PR tree-optimization/107121
12100 * tree-cfg.cc (verify_gimple_call): Fix a typo in diagnostics,
12101 DEFFERED_INIT -> DEFERRED_INIT.
12103 2022-10-02 Marc Poulhiès <poulhies@adacore.com>
12105 * config/vxworks-driver.cc: New.
12106 * config.gcc (*vxworks*): Add vxworks-driver.o in extra_gcc_objs.
12107 * config/t-vxworks: Add vxworks-driver.o.
12108 * config/vxworks.h (GCC_DRIVER_HOST_INITIALIZATION): New.
12110 2022-10-02 Olivier Hainque <hainque@adacore.com>
12112 * config/vxworks.h (VX_CRTBEGIN_SPEC, VX_CRTEND_SPEC): If
12113 HAVE_INITFINI_ARRAY_SUPPORT, pick crtstuff objects regardless
12114 of ENABLE_SHARED_LIBGCC.
12116 2022-10-01 Stafford Horne <shorne@gmail.com>
12118 * config/or1k/or1k.cc (TARGET_HAVE_TLS): Only define if
12119 HAVE_AS_TLS is defined.
12121 2022-10-01 Julian Brown <julian@codesourcery.com>
12123 * gimplify.cc (omp_group_base): Fix IF_PRESENT (no_create)
12126 2022-10-01 Jeff Law <jeffreyalaw@gmail.com>
12128 * config/h8300/h8300.md (HSI2): New iterator.
12129 (eqne_invert): Similarly.
12130 * config/h8300/testcompare.md (testhi_upper_z): New pattern.
12131 (cmpqi_z, cmphi_z, cmpsi_z): Likewise.
12132 (store_z_qi, store_z_i_qi, store_z_hi, store_z_hi_sb): New
12133 define_insn_and_splits and/or define_insns.
12134 (store_z_hi_neg, store_z_hi_and, store_z_<mode>): Likewise.
12135 (store_z_<mode>_neg, store_z_<mode>_and, store_z): Likewise.
12137 2022-10-01 Jeff Law <jeffreyalaw@gmail.com>
12139 * tree-ssa-dom.cc (record_edge_info): Install correct version of
12142 2022-10-01 Eugene Rozenfeld <erozen@microsoft.com>
12144 * dwarf2out.cc (add_call_src_coords_attributes): Emit discriminators for inlined call sites.
12146 2022-09-30 Jeff Law <jeffreyalaw@gmail.com>
12148 * tree-ssa-dom.cc (single_block_loop_p): New function.
12149 (record_edge_info): Also record equivalences for the outgoing
12150 edge of a single block loop where the condition is an invariant.
12152 2022-09-30 Jeff Law <jeffreyalaw@gmail.com>
12154 * tree-ssa-dom.cc (free_dom_edge_info): Clear e->aux too.
12155 (free_all_edge_infos): Do not clear e->aux here.
12157 2022-09-30 H.J. Lu <hjl.tools@gmail.com>
12159 * target.def (TARGET_C_EXCESS_PRECISION): Document
12160 -fexcess-precision=16.
12162 2022-09-30 Palmer Dabbelt <palmer@rivosinc.com>
12164 * doc/tm.texi (TARGET_C_EXCESS_PRECISION): Add 16.
12166 2022-09-30 Palmer Dabbelt <palmer@rivosinc.com>
12169 * config/riscv/riscv.cc (riscv_excess_precision): Add support
12170 for EXCESS_PRECISION_TYPE_FLOAT16.
12172 2022-09-30 Jakub Jelinek <jakub@redhat.com>
12175 * config/arm/arm.cc (arm_mangle_type): Mangle just __fp16 as Dh
12176 and _Float16 as DF16_.
12177 * config/csky/csky.cc (csky_init_builtins): Fix a comment typo.
12178 (csky_mangle_type): Mangle __fp16 as Dh and _Float16 as DF16_
12179 rather than mangling __fp16 as __fp16.
12181 2022-09-30 Lewis Hyatt <lhyatt@gmail.com>
12183 PR preprocessor/69543
12184 * tree-ssa-uninit.cc (warn_uninit): Stop stripping macro tracking
12185 information away from the diagnostic location.
12186 (maybe_warn_read_write_only): Likewise.
12187 (maybe_warn_operand): Likewise.
12189 2022-09-30 Jakub Jelinek <jakub@redhat.com>
12192 * config/aarch64/aarch64.cc (aarch64_mangle_type): Mangle just __fp16
12193 as Dh and _Float16 as DF16_.
12195 2022-09-30 Jakub Jelinek <jakub@redhat.com>
12198 * config/i386/i386.cc (ix86_mangle_type): Always return NULL
12199 for float128_type_node or float64x_type_node, don't check
12200 float128t_type_node later on.
12201 * config/ia64/ia64.cc (ia64_mangle_type): Always return NULL
12202 for float128_type_node or float64x_type_node.
12203 * config/rs6000/rs6000.cc (rs6000_mangle_type): Likewise.
12204 Don't check float128_type_node later on.
12205 * config/s390/s390.cc (s390_mangle_type): Don't use
12206 TYPE_MAIN_VARIANT on type which was set to TYPE_MAIN_VARIANT
12207 a few lines earlier.
12209 2022-09-30 Torbjörn SVENSSON <torbjorn.svensson@foss.st.com>
12210 Yvan ROUX <yvan.roux@foss.st.com>
12212 * doc/sourcebuild.texi: Document new vma_equals_lma effective
12215 2022-09-30 Jiawei <jiawei@iscas.ac.cn>
12217 * config/riscv/riscv.cc (riscv_file_start): New .option.
12218 * config/riscv/riscv.opt: New options.
12219 * doc/invoke.texi: New definations.
12221 2022-09-30 Kewen Lin <linkw@linux.ibm.com>
12225 * doc/invoke.texi (option -fpatchable-function-entry): Adjust the
12226 documentation for PowerPC ELFv2 ABI dual entry points.
12227 * config/rs6000/rs6000-internal.h
12228 (rs6000_print_patchable_function_entry): New function declaration.
12229 * config/rs6000/rs6000-logue.cc (rs6000_output_function_prologue):
12230 Support patchable-function-entry by emitting nops before and after
12231 local entry for the function that needs global entry.
12232 * config/rs6000/rs6000.cc (rs6000_print_patchable_function_entry): Skip
12233 the function that needs global entry till global entry has been
12235 * config/rs6000/rs6000.h (struct machine_function): New bool member
12236 global_entry_emitted.
12238 2022-09-30 Richard Biener <rguenther@suse.de>
12240 PR tree-optimization/107095
12241 * tree-ssa-dse.cc (initialize_ao_ref_for_dse): Use data arg
12242 for .MASK_STORE size.
12244 2022-09-29 Andrew MacLeod <amacleod@redhat.com>
12246 * range-op.cc (plus_minus_ranges): New.
12247 (adjust_op1_for_overflow): New.
12248 (operator_plus::op1_range): Use new adjustment.
12249 (operator_plus::op2_range): Ditto.
12250 (operator_minus::op1_range): Ditto.
12251 * value-relation.h (relation_lt_le_gt_ge_p): New.
12253 2022-09-29 Andrew MacLeod <amacleod@redhat.com>
12255 * gimple-range-gori.cc (gori_compute::refine_using_relation): New.
12256 (gori_compute::compute_operand1_range): Invoke
12257 refine_using_relation when applicable.
12258 (gori_compute::compute_operand2_range): Ditto.
12259 * gimple-range-gori.h (class gori_compute): Adjust prototypes.
12261 2022-09-29 Andrew MacLeod <amacleod@redhat.com>
12263 * gimple-range-gori.cc (ori_compute::compute_operand_range):
12264 Create a relation record and pass it along when possible.
12265 (gori_compute::compute_operand1_range): Pass relation along.
12266 (gori_compute::compute_operand2_range): Ditto.
12267 (gori_compute::compute_operand1_and_operand2_range): Ditto.
12268 * gimple-range-gori.h (class gori_compute): Adjust prototypes.
12269 * gimple-range-op.cc (gimple_range_op_handler::calc_op1): Pass
12270 relation to op1_range call.
12271 (gimple_range_op_handler::calc_op2): Pass relation to op2_range call.
12272 * gimple-range-op.h (class gimple_range_op_handler): Adjust
12275 2022-09-29 Andrew MacLeod <amacleod@redhat.com>
12277 * value-relation.cc (class value_relation): Move to .h file.
12278 (value_relation::set_relation): Ditto.
12279 (value_relation::value_relation): ditto.
12280 * value-relation.h (class value_relation): Move from .cc file.
12281 (value_relation::set_relation): Ditto
12282 (value_relation::value_relation): Ditto.
12284 2022-09-29 Andrew MacLeod <amacleod@redhat.com>
12286 * range-op.cc (operator_minus::op2_range): Check for undefined.
12287 (operator_mult::op1_range): Ditto.
12288 (operator_exact_divide::op1_range): Ditto.
12289 (operator_lshift::op1_range): Ditto.
12290 (operator_rshift::op1_range): Ditto.
12291 (operator_cast::op1_range): Ditto.
12292 (operator_bitwise_and::op1_range): Ditto.
12293 (operator_bitwise_or::op1_range): Ditto.
12294 (operator_trunc_mod::op1_range): Ditto.
12295 (operator_trunc_mod::op2_range): Ditto.
12296 (operator_bitwise_not::op1_range): Ditto.
12297 (pointer_or_operator::op1_range): Ditto.
12298 (range_op_handler::op1_range): Ditto.
12299 (range_op_handler::op2_range): Ditto.
12301 2022-09-29 Andrew Stubbs <ams@codesourcery.com>
12303 * config/gcn/gcn.cc (gcn_simd_clone_compute_vecsize_and_simdlen):
12304 Remove unused elt_bits variable.
12306 2022-09-29 Olivier Hainque <hainque@adacore.com>
12308 * config/vxworks.h: Add comment on our use of
12309 HAVE_INITFINI_ARRAY_SUPPORT.
12311 2022-09-29 Olivier Hainque <hainque@adacore.com>
12313 * config/aarch64/t-aarch64-vxworks: Request multilib
12314 variants for mcmodel=large.
12316 2022-09-29 Olivier Hainque <hainque@adacore.com>
12318 * config/rs6000/vxworks.h (TARGET_FLOAT128_ENABLE_TYPE): Remove
12321 2022-09-29 Olivier Hainque <hainque@adacore.com>
12323 * config/vx-common.h (DWARF2_UNWIND_INFO): #define to 0
12324 when ARM_UNWIND_INFO is set.
12326 2022-09-29 Julian Brown <julian@codesourcery.com>
12328 PR middle-end/107028
12329 * gimplify.cc (omp_check_mapping_compatibility,
12330 oacc_resolve_clause_dependencies): New functions.
12331 (omp_accumulate_sibling_list): Remove redundant duplicate clause
12332 detection for OpenACC.
12333 (build_struct_sibling_lists): Skip deleted groups. Don't build sibling
12334 list for struct variables that are fully mapped on the same directive
12336 (gimplify_scan_omp_clauses): Call oacc_resolve_clause_dependencies.
12338 2022-09-29 Jose E. Marchesi <jose.marchesi@oracle.com>
12340 PR middle-end/25521
12341 * varasm.cc (categorize_decl_for_section): Place `const volatile'
12342 objects in read-only sections.
12343 (default_select_section): Likewise.
12345 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12347 * tree-data-ref.cc (dr_may_alias_p): Use to_poly_widest instead
12350 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12352 * config/aarch64/aarch64.h (TARGET_CRYPTO, TARGET_SHA3, TARGET_SM4)
12353 (TARGET_DOTPROD): Don't depend on TARGET_SIMD.
12354 (TARGET_AES, TARGET_SHA2): Likewise. Remove TARGET_CRYPTO test.
12355 (TARGET_FP_F16INST): Don't depend on TARGET_FLOAT.
12356 (TARGET_SVE2, TARGET_SVE_F32MM, TARGET_SVE_F64MM): Don't depend
12358 (TARGET_SVE2_AES, TARGET_SVE2_BITPERM, TARGET_SVE2_SHA3)
12359 (TARGET_SVE2_SM4): Don't depend on TARGET_SVE2.
12360 (TARGET_F32MM, TARGET_F64MM): Delete.
12361 * config/aarch64/aarch64-c.cc (aarch64_update_cpp_builtins): Guard
12362 float macros with just TARGET_FLOAT rather than TARGET_FLOAT
12364 * config/aarch64/aarch64-simd.md (copysign<mode>3): Depend
12365 only on TARGET_SIMD, rather than TARGET_FLOAT && TARGET_SIMD.
12366 (aarch64_crypto_aes<aes_op>v16qi): Depend only on TARGET_AES,
12367 rather than TARGET_SIMD && TARGET_AES.
12368 (aarch64_crypto_aes<aesmc_op>v16qi): Likewise.
12369 (*aarch64_crypto_aese_fused): Likewise.
12370 (*aarch64_crypto_aesd_fused): Likewise.
12371 (aarch64_crypto_pmulldi): Likewise.
12372 (aarch64_crypto_pmullv2di): Likewise.
12373 (aarch64_crypto_sha1hsi): Likewise TARGET_SHA2.
12374 (aarch64_crypto_sha1hv4si): Likewise.
12375 (aarch64_be_crypto_sha1hv4si): Likewise.
12376 (aarch64_crypto_sha1su1v4si): Likewise.
12377 (aarch64_crypto_sha1<sha1_op>v4si): Likewise.
12378 (aarch64_crypto_sha1su0v4si): Likewise.
12379 (aarch64_crypto_sha256h<sha256_op>v4si): Likewise.
12380 (aarch64_crypto_sha256su0v4si): Likewise.
12381 (aarch64_crypto_sha256su1v4si): Likewise.
12382 (aarch64_crypto_sha512h<sha512_op>qv2di): Likewise TARGET_SHA3.
12383 (aarch64_crypto_sha512su0qv2di): Likewise.
12384 (aarch64_crypto_sha512su1qv2di, eor3q<mode>4): Likewise.
12385 (aarch64_rax1qv2di, aarch64_xarqv2di, bcaxq<mode>4): Likewise.
12386 (aarch64_sm3ss1qv4si): Likewise TARGET_SM4.
12387 (aarch64_sm3tt<sm3tt_op>qv4si): Likewise.
12388 (aarch64_sm3partw<sm3part_op>qv4si): Likewise.
12389 (aarch64_sm4eqv4si, aarch64_sm4ekeyqv4si): Likewise.
12390 * config/aarch64/aarch64.md (<FLOATUORS:optab>dihf2)
12391 (copysign<GPF:mode>3, copysign<GPF:mode>3_insn)
12392 (xorsign<mode>3): Remove redundant TARGET_FLOAT condition.
12394 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12396 * config/aarch64/aarch64.opt (aarch64_asm_isa_flags): New variable.
12397 * config/aarch64/aarch64.h (aarch64_asm_isa_flags)
12398 (aarch64_isa_flags): Redefine as read-only macros.
12399 (TARGET_SIMD, TARGET_FLOAT, TARGET_SVE): Don't depend on
12400 !TARGET_GENERAL_REGS_ONLY.
12401 * common/config/aarch64/aarch64-common.cc
12402 (aarch64_set_asm_isa_flags): New function.
12403 (aarch64_handle_option): Call it when updating -mgeneral-regs.
12404 * config/aarch64/aarch64-protos.h (aarch64_simd_switcher): Replace
12405 m_old_isa_flags with m_old_asm_isa_flags.
12406 (aarch64_set_asm_isa_flags): Declare.
12407 * config/aarch64/aarch64-builtins.cc
12408 (aarch64_simd_switcher::aarch64_simd_switcher)
12409 (aarch64_simd_switcher::~aarch64_simd_switcher): Save and restore
12410 aarch64_asm_isa_flags instead of aarch64_isa_flags.
12411 * config/aarch64/aarch64-sve-builtins.cc
12412 (check_required_extensions): Use aarch64_asm_isa_flags instead
12413 of aarch64_isa_flags.
12414 * config/aarch64/aarch64.cc (aarch64_set_asm_isa_flags): New function.
12415 (aarch64_override_options, aarch64_handle_attr_arch)
12416 (aarch64_handle_attr_cpu, aarch64_handle_attr_isa_flags): Use
12417 aarch64_set_asm_isa_flags to set the ISA flags.
12418 (aarch64_option_print, aarch64_declare_function_name)
12419 (aarch64_start_file): Use aarch64_asm_isa_flags instead
12420 of aarch64_isa_flags.
12421 (aarch64_can_inline_p): Check aarch64_asm_isa_flags as well as
12424 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12426 * common/config/aarch64/aarch64-common.cc (all_extensions):
12427 Include the feature flag in flags_on and flags_off.
12428 (aarch64_parse_extension): Update accordingly.
12429 (aarch64_get_extension_string_for_isa_flags): Likewise.
12431 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12433 * common/config/aarch64/aarch64-common.cc: Use aarch64_feature_flags
12434 for feature flags throughout.
12435 * config/aarch64/aarch64-protos.h: Likewise.
12436 * config/aarch64/aarch64-sve-builtins.h: Likewise.
12437 * config/aarch64/aarch64-sve-builtins.cc: Likewise.
12438 * config/aarch64/aarch64.cc: Likewise.
12439 * config/aarch64/aarch64.opt: Likewise.
12440 * config/aarch64/driver-aarch64.cc: Likewise.
12442 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12444 * common/config/aarch64/aarch64-common.cc (aarch64_option_extension)
12445 (processor_name_to_arch, arch_to_arch_name): Remove const from
12447 (all_extensions, all_cores, all_architectures): Make a constexpr.
12448 * config/aarch64/aarch64.cc (processor): Remove const from
12450 (all_architectures): Make a constexpr.
12451 * config/aarch64/driver-aarch64.cc (aarch64_core_data)
12452 (aarch64_arch_driver_info): Remove const from member variables.
12453 (aarch64_cpu_data, aarch64_arches): Make a constexpr.
12454 (get_arch_from_id): Return a pointer to const.
12455 (host_detect_local_cpu): Update accordingly.
12457 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12459 * common/config/aarch64/aarch64-common.cc (processor_name_to_arch)
12460 (arch_to_arch_name): Use const char * instead of std::string.
12462 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12464 * common/config/aarch64/aarch64-common.cc
12465 (TARGET_OPTION_INIT_STRUCT): Delete.
12466 (aarch64_option_extension): Remove is_synthetic_flag.
12467 (all_extensions): Update accordingly.
12468 (all_extensions_by_on, opt_ext, opt_ext_cmp): Delete.
12469 (aarch64_option_init_struct, aarch64_contains_opt): Delete.
12470 (aarch64_get_extension_string_for_isa_flags): Rewrite to use
12471 all_extensions instead of all_extensions_on.
12473 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12475 * config/aarch64/aarch64-option-extensions.def: Switch to a new format.
12476 * config/aarch64/aarch64-cores.def: Use the same format to specify
12478 * config/aarch64/aarch64-arches.def: Likewise, moving that information
12480 * config/aarch64/aarch64-opts.h (aarch64_feature_flags): New typedef.
12481 * config/aarch64/aarch64.h (aarch64_feature): New class enum.
12482 Turn AARCH64_FL_* macros into constexprs, getting the definitions
12483 from aarch64-option-extensions.def. Remove AARCH64_FL_FOR_* macros.
12484 * common/config/aarch64/aarch64-common.cc: Include
12485 aarch64-feature-deps.h.
12486 (all_extensions): Update for new .def format.
12487 (all_extensions_by_on, all_cores, all_architectures): Likewise.
12488 * config/aarch64/driver-aarch64.cc: Include aarch64-feature-deps.h.
12489 (aarch64_extensions): Update for new .def format.
12490 (aarch64_cpu_data, aarch64_arches): Likewise.
12491 * config/aarch64/aarch64.cc: Include aarch64-feature-deps.h.
12492 (all_architectures, all_cores): Update for new .def format.
12493 * config/aarch64/aarch64-sve-builtins.cc
12494 (check_required_extensions): Likewise.
12495 * config/aarch64/aarch64-feature-deps.h: New file.
12497 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12499 * config/aarch64/aarch64-option-extensions.def: Move crypto
12502 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12504 * config/aarch64/aarch64-option-extensions.def (dotprod): Depend
12505 on fp as well as simd.
12507 (aes): Likewise. Make +noaes disable crypto.
12508 (sha2): Likewise +nosha2. Also make +nosha2 disable sha3 and
12510 (sve2-sha3): Depend on sha2 as well as sha3.
12512 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12515 * config/aarch64/aarch64.h (oAARCH64_FL_RCPC8_4): Delete.
12516 (AARCH64_FL_FOR_V8_4A): Update accordingly.
12517 (AARCH64_ISA_RCPC8_4): Use AARCH64_FL_V8_4A directly.
12518 * config/aarch64/aarch64-cores.def (thunderx3t110): Remove
12519 AARCH64_FL_RCPC8_4.
12521 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12523 * config/aarch64/aarch64-cores.def: Remove AARCH64_FL_FOR_<ARCH>
12524 from the flags field.
12525 * common/config/aarch64/aarch64-common.cc (all_cores): Add it
12527 * config/aarch64/aarch64.cc (all_cores): Likewise.
12528 * config/aarch64/driver-aarch64.cc (all_cores): Likewise.
12530 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12532 * config.gcc: Remove dead aarch64-option-extensions.def code.
12533 * config/aarch64/aarch64-arches.def: Update comment.
12535 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12537 * config/aarch64/aarch64-arches.def: Add a leading "V" to the
12539 * config/aarch64/aarch64-cores.def: Update accordingly.
12540 * common/config/aarch64/aarch64-common.cc (all_cores): Likewise.
12541 * config/aarch64/aarch64.cc (all_cores): Likewise.
12542 * config/aarch64/driver-aarch64.cc (aarch64_arches): Skip the
12545 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12547 * config/aarch64/aarch64.h (AARCH64_FL_FOR_ARCH8): Rename to...
12548 (AARCH64_FL_FOR_V8A): ...this.
12549 (AARCH64_FL_FOR_ARCH8_1): Rename to...
12550 (AARCH64_FL_FOR_V8_1A): ...this.
12551 (AARCH64_FL_FOR_ARCH8_2): Rename to...
12552 (AARCH64_FL_FOR_V8_2A): ...this.
12553 (AARCH64_FL_FOR_ARCH8_3): Rename to...
12554 (AARCH64_FL_FOR_V8_3A): ...this.
12555 (AARCH64_FL_FOR_ARCH8_4): Rename to...
12556 (AARCH64_FL_FOR_V8_4A): ...this.
12557 (AARCH64_FL_FOR_ARCH8_5): Rename to...
12558 (AARCH64_FL_FOR_V8_5A): ...this.
12559 (AARCH64_FL_FOR_ARCH8_6): Rename to...
12560 (AARCH64_FL_FOR_V8_6A): ...this.
12561 (AARCH64_FL_FOR_ARCH8_7): Rename to...
12562 (AARCH64_FL_FOR_V8_7A): ...this.
12563 (AARCH64_FL_FOR_ARCH8_8): Rename to...
12564 (AARCH64_FL_FOR_V8_8A): ...this.
12565 (AARCH64_FL_FOR_ARCH8_R): Rename to...
12566 (AARCH64_FL_FOR_V8R): ...this.
12567 (AARCH64_FL_FOR_ARCH9): Rename to...
12568 (AARCH64_FL_FOR_V9A): ...this.
12569 (AARCH64_FL_FOR_ARCH9_1): Rename to...
12570 (AARCH64_FL_FOR_V9_1A): ...this.
12571 (AARCH64_FL_FOR_ARCH9_2): Rename to...
12572 (AARCH64_FL_FOR_V9_2A): ...this.
12573 (AARCH64_FL_FOR_ARCH9_3): Rename to...
12574 (AARCH64_FL_FOR_V9_3A): ...this.
12575 * common/config/aarch64/aarch64-common.cc (all_cores): Update
12577 * config/aarch64/aarch64-arches.def: Likewise.
12578 * config/aarch64/aarch64-cores.def: Likewise.
12579 * config/aarch64/aarch64.cc (all_cores): Likewise.
12581 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12583 * config/aarch64/aarch64.h (AARCH64_FL_V8_1, AARCH64_FL_V8_2)
12584 (AARCH64_FL_V8_3, AARCH64_FL_V8_4, AARCH64_FL_V8_5, AARCH64_FL_V8_6)
12585 (AARCH64_FL_V9, AARCH64_FL_V8_7, AARCH64_FL_V8_8, AARCH64_FL_V9_1)
12586 (AARCH64_FL_V9_2, AARCH64_FL_V9_3): Add "A" to the end of the name.
12587 (AARCH64_FL_V8_R): Rename to AARCH64_FL_V8R.
12588 (AARCH64_FL_FOR_ARCH8_1, AARCH64_FL_FOR_ARCH8_2): Update accordingly.
12589 (AARCH64_FL_FOR_ARCH8_3, AARCH64_FL_FOR_ARCH8_4): Likewise.
12590 (AARCH64_FL_FOR_ARCH8_5, AARCH64_FL_FOR_ARCH8_6): Likewise.
12591 (AARCH64_FL_FOR_ARCH8_7, AARCH64_FL_FOR_ARCH8_8): Likewise.
12592 (AARCH64_FL_FOR_ARCH8_R, AARCH64_FL_FOR_ARCH9): Likewise.
12593 (AARCH64_FL_FOR_ARCH9_1, AARCH64_FL_FOR_ARCH9_2): Likewise.
12594 (AARCH64_FL_FOR_ARCH9_3, AARCH64_ISA_V8_2A, AARCH64_ISA_V8_3A)
12595 (AARCH64_ISA_V8_4A, AARCH64_ISA_V8_5A, AARCH64_ISA_V8_6A): Likewise.
12596 (AARCH64_ISA_V8R, AARCH64_ISA_V9A, AARCH64_ISA_V9_1A): Likewise.
12597 (AARCH64_ISA_V9_2A, AARCH64_ISA_V9_3A): Likewise.
12599 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12601 * config/aarch64/aarch64.h (AARCH64_ISA_V8_2, AARCH64_ISA_V8_3)
12602 (AARCH64_ISA_V8_4, AARCH64_ISA_V8_5, AARCH64_ISA_V8_6)
12603 (AARCH64_ISA_V9, AARCH64_ISA_V9_1, AARCH64_ISA_V9_2)
12604 (AARCH64_ISA_V9_3): Add "A" to the end of the name.
12605 (AARCH64_ISA_V8_R): Rename to AARCH64_ISA_V8R.
12606 (TARGET_ARMV8_3, TARGET_JSCVT, TARGET_FRINT, TARGET_MEMTAG): Update
12608 * common/config/aarch64/aarch64-common.cc
12609 (aarch64_get_extension_string_for_isa_flags): Likewise.
12610 * config/aarch64/aarch64-c.cc
12611 (aarch64_define_unconditional_macros): Likewise.
12613 2022-09-29 Richard Sandiford <richard.sandiford@arm.com>
12615 * Makefile.in (GTFILES): Add OPTIONS_H_EXTRA.
12617 2022-09-29 Jakub Jelinek <jakub@redhat.com>
12619 PR bootstrap/107059
12620 * cppdefault.cc (cpp_include_defaults): If SYSROOT_HEADERS_SUFFIX_SPEC
12621 isn't defined, add FIXED_INCLUDE_DIR entry with multilib flag 2
12622 before FIXED_INCLUDE_DIR entry with multilib flag 0.
12623 * gcc.cc (do_spec_1): If multiarch_dir, add
12624 include-fixed/multiarch_dir paths before include-fixed paths.
12626 2022-09-29 Martin Liska <mliska@suse.cz>
12629 * common.opt: Add -gz=zstd value.
12630 * configure.ac: Detect --compress-debug-sections=zstd
12631 for both linker and assembler.
12632 * configure: Regenerate.
12633 * gcc.cc (LINK_COMPRESS_DEBUG_SPEC): Handle -gz=zstd.
12634 (ASM_COMPRESS_DEBUG_SPEC): Likewise.
12636 2022-09-29 Richard Biener <rguenther@suse.de>
12638 PR tree-optimization/105646
12639 * tree-ssa-uninit.cc (warn_uninitialized_vars): Pre-compute
12640 the set of fallthru reachable blocks from function entry
12641 and use that to determine wlims.always_executed.
12643 2022-09-29 liuhongt <hongtao.liu@intel.com>
12645 PR tree-optimization/107055
12646 * tree-vect-loop-manip.cc (vect_can_advance_ivs_p): Check for
12647 nonlinear induction variables.
12648 * tree-vect-loop.cc (vect_can_peel_nonlinear_iv_p): New
12650 (vectorizable_nonlinear_induction): Put part codes into
12651 vect_can_peel_nonlinear_iv_p.
12652 * tree-vectorizer.h (vect_can_peel_nonlinear_iv_p): Declare.
12654 2022-09-28 Eugene Rozenfeld <erozen@microsoft.com>
12656 * basic-block.h: Remove discriminator from basic blocks.
12657 * cfghooks.cc (split_block_1): Remove discriminator from basic blocks.
12658 * final.cc (final_start_function_1): Switch from per-bb to per statement
12660 (final_scan_insn_1): Don't keep track of basic block discriminators.
12661 (compute_discriminator): Switch from basic block discriminators to
12662 instruction discriminators.
12663 (insn_discriminator): New function to return instruction discriminator.
12664 (notice_source_line): Use insn_discriminator.
12665 * gimple-pretty-print.cc (dump_gimple_bb_header): Remove dumping of
12666 basic block discriminators.
12667 * gimple-streamer-in.cc (input_bb): Remove reading of basic block
12669 * gimple-streamer-out.cc (output_bb): Remove writing of basic block
12671 * input.cc (make_location): Pass 0 discriminator to COMBINE_LOCATION_DATA.
12672 (location_with_discriminator): New function to combine locus with
12674 (has_discriminator): New function to check if a location has a discriminator.
12675 (get_discriminator_from_loc): New function to get the discriminator
12677 * input.h: Declarations of new functions.
12678 * lto-streamer-in.cc (cmp_loc): Use discriminators in location comparison.
12679 (apply_location_cache): Keep track of current discriminator.
12680 (input_location_and_block): Read discriminator from stream.
12681 * lto-streamer-out.cc (clear_line_info): Set current discriminator to
12683 (lto_output_location_1): Write discriminator to stream.
12684 * lto-streamer.h: Add discriminator to cached_location.
12685 Add current_discr to lto_location_cache.
12686 Add current_discr to output_block.
12687 * print-rtl.cc (print_rtx_operand_code_i): Print discriminator.
12688 * rtl.h: Add extern declaration of insn_discriminator.
12689 * tree-cfg.cc (assign_discriminator): New function to assign a unique
12690 discriminator value to all statements in a basic block that have the given
12692 (assign_discriminators): Assign discriminators to statement locations.
12693 * tree-pretty-print.cc (dump_location): Dump discriminators.
12694 * tree.cc (set_block): Preserve discriminator when setting block.
12695 (set_source_range): Preserve discriminator when setting source range.
12697 2022-09-28 H.J. Lu <hjl.tools@gmail.com>
12700 * config/i386/predicates.md (encodekey128_operation): Check
12701 XMM4-XMM6 as clobbered.
12702 (encodekey256_operation): Likewise.
12703 * config/i386/sse.md (encodekey128u32): Clobber XMM4-XMM6.
12704 (encodekey256u32): Likewise.
12706 2022-09-28 Ju-Zhe Zhong <juzhe.zhong@rivai.ai>
12708 * config.gcc: Add riscv-vector-builtins.o.
12709 * config/riscv/riscv-builtins.cc (riscv_init_builtins): Add RVV builtin function.
12710 * config/riscv/riscv-protos.h (riscv_v_ext_enabled_vector_mode_p): New function.
12711 * config/riscv/riscv.cc (ENTRY): New macro.
12712 (riscv_v_ext_enabled_vector_mode_p): New function.
12713 (riscv_mangle_type): Add RVV mangle.
12714 (riscv_vector_mode_supported_p): Adjust RVV machine mode.
12715 (riscv_verify_type_context): Add context check for RVV.
12716 (riscv_vector_alignment): Add RVV alignment target hook support.
12717 (TARGET_VECTOR_MODE_SUPPORTED_P): New target hook support.
12718 (TARGET_VERIFY_TYPE_CONTEXT): Ditto.
12719 (TARGET_VECTOR_ALIGNMENT): Ditto.
12720 * config/riscv/t-riscv: Add riscv-vector-builtins.o
12721 * config/riscv/riscv-vector-builtins.cc: New file.
12722 * config/riscv/riscv-vector-builtins.def: New file.
12723 * config/riscv/riscv-vector-builtins.h: New file.
12724 * config/riscv/riscv-vector-switch.def: New file.
12726 2022-09-28 Stefan Schulze Frielinghaus <stefansf@linux.ibm.com>
12728 * var-tracking.cc (vt_add_function_parameter): Add entry values
12729 up to maximal register mode.
12731 2022-09-28 Stefan Schulze Frielinghaus <stefansf@linux.ibm.com>
12733 * cselib.cc (new_cselib_val): Keep track of further subvalue
12736 2022-09-28 Andrea Corallo <andrea.corallo@arm.com>
12738 * config/arm/arm-c.cc (arm_cpu_builtins): Define
12739 __ARM_FEATURE_AES and __ARM_FEATURE_SHA2.
12741 2022-09-28 Xi Ruoyao <xry111@xry111.site>
12743 PR tree-optimization/105414
12744 * config/loongarch/loongarch.md (UNSPEC_FMAX): New unspec.
12745 (UNSPEC_FMIN): Likewise.
12746 (fmax<mode>3): Use UNSPEC_FMAX instead of smax.
12747 (fmin<mode>3): Use UNSPEC_FMIN instead of smin.
12749 2022-09-28 Lulu Cheng <chenglulu@loongson.cn>
12751 * config/loongarch/loongarch.cc (loongarch_asan_shadow_offset):
12752 Fixed typo in "asan_mapping.h".
12754 2022-09-28 H.J. Lu <hjl.tools@gmail.com>
12756 PR middle-end/58245
12757 * calls.cc: Include "tree-eh.h".
12758 (expand_call): Check stack canary before throwing exception.
12760 2022-09-27 Eugene Rozenfeld <erozen@microsoft.com>
12762 * ipa-cp.cc (good_cloning_opportunity_p): Fix profile count comparison.
12764 2022-09-27 Kim Kuparinen <kim.kuparinen@rightware.com>
12766 * doc/invoke.texi: Update ABI version info.
12768 2022-09-27 Aldy Hernandez <aldyh@redhat.com>
12770 * gimple-range-op.cc (cfn_popcount): Calculate the popcount of a
12773 2022-09-27 Aldy Hernandez <aldyh@redhat.com>
12775 * value-range.cc (irange::set_nonzero_bits): Set range when known.
12777 2022-09-27 Aldy Hernandez <aldyh@redhat.com>
12779 * value-range.h (irange::set): New version taking wide_int_ref.
12781 2022-09-27 Jakub Jelinek <jakub@redhat.com>
12783 PR tree-optimization/107029
12784 * tree-ssa-reassoc.cc (optimize_range_tests_cmp_bitwise): Treat
12785 OFFSET_TYPE like POINTER_TYPE, except that OFFSET_TYPE may be
12786 signed and so can trigger even the (b % 4) == 3 case.
12788 2022-09-27 Jeff Law <jeffreyalaw@gmail.com>
12790 * cfgrtl.cc (fixup_reorder_chain): Verify that simple_return
12791 and return are available before trying to use them.
12793 2022-09-27 Jakub Jelinek <jakub@redhat.com>
12797 * tree-core.h (enum tree_index): Add TI_FLOAT128T_TYPE
12799 * tree.h (float128t_type_node): Define.
12800 * tree.cc (build_common_tree_nodes): Initialize float128t_type_node.
12801 * builtins.def (DEF_FLOATN_BUILTIN): Adjust comment now that
12802 _Float<N> is supported in C++ too.
12803 * config/i386/i386.cc (ix86_mangle_type): Only mangle as "g"
12804 float128t_type_node.
12805 * config/i386/i386-builtins.cc (ix86_init_builtin_types): Use
12806 float128t_type_node for __float128 instead of float128_type_node
12807 and create it if NULL.
12808 * config/i386/avx512fp16intrin.h (_mm_setzero_ph, _mm256_setzero_ph,
12809 _mm512_setzero_ph, _mm_set_sh, _mm_load_sh): Use 0.0f16 instead of
12811 * config/ia64/ia64.cc (ia64_init_builtins): Use
12812 float128t_type_node for __float128 instead of float128_type_node
12813 and create it if NULL.
12814 * config/rs6000/rs6000-c.cc (is_float128_p): Also return true
12815 for float128t_type_node if non-NULL.
12816 * config/rs6000/rs6000.cc (rs6000_mangle_type): Don't mangle
12817 float128_type_node as "u9__ieee128".
12818 * config/rs6000/rs6000-builtin.cc (rs6000_init_builtins): Use
12819 float128t_type_node for __float128 instead of float128_type_node
12820 and create it if NULL.
12822 2022-09-26 Martin Liska <mliska@suse.cz>
12824 * doc/invoke.texi: Add missing dash for
12825 Wanalyzer-exposure-through-uninit-copy.
12827 2022-09-26 Aldy Hernandez <aldyh@redhat.com>
12829 PR tree-optimization/107009
12830 * range-op.cc (operator_bitwise_and::op1_range): Optimize 0 = x & MASK.
12831 (range_op_bitwise_and_tests): New test.
12833 2022-09-26 Aldy Hernandez <aldyh@redhat.com>
12835 PR tree-optimization/107009
12837 (dom_opt_dom_walker::set_global_ranges_from_unreachable_edges):
12838 Iterate over exports.
12840 2022-09-26 Thomas Schwinge <thomas@codesourcery.com>
12842 * config.gcc (with_arch) [nvptx]: Allow '--with-arch' to override
12844 * config/nvptx/gen-multilib-matches.sh: New.
12845 * config/nvptx/t-nvptx (MULTILIB_OPTIONS, MULTILIB_MATCHES)
12846 (MULTILIB_EXCEPTIONS): Handle this.
12847 * doc/install.texi (Specific) <nvptx-*-none>: Document this.
12848 * doc/invoke.texi (Nvidia PTX Options): Likewise.
12850 2022-09-26 Thomas Schwinge <thomas@codesourcery.com>
12852 * config.gcc (TM_MULTILIB_CONFIG) [nvptx]: Set to '$with_arch'.
12853 * config/nvptx/t-nvptx (MULTILIB_OPTIONS, MULTILIB_MATCHES)
12854 (MULTILIB_EXCEPTIONS): Handle it.
12856 2022-09-26 Thomas Schwinge <thomas@codesourcery.com>
12858 * config.gcc (with_arch) [nvptx]: Set to 'sm_30'.
12859 * config/nvptx/nvptx.cc (nvptx_option_override): Assert that
12861 * config/nvptx/nvptx.h (OPTION_DEFAULT_SPECS): Define.
12862 * config/nvptx/nvptx.opt (misa=): Remove 'Init'.
12864 2022-09-26 Thomas Schwinge <thomas@codesourcery.com>
12866 * config/nvptx/nvptx.h (ASM_SPEC): Define.
12868 2022-09-26 Jeff Law <jeffreyalaw@gmail.com>
12870 * cfgcleanup.cc (bb_is_just_return): No longer static.
12871 * cfgcleanup.h (bb_is_just_return): Add prototype.
12872 * cfgrtl.cc (fixup_reorder_chain): Do not create an
12873 unconditional jump to a return block. Conditionally
12874 remove unreachable blocks.
12876 2022-09-26 Tobias Burnus <tobias@codesourcery.com>
12878 PR middle-end/106982
12879 * omp-low.cc (lower_oacc_reductions): Add some unshare_expr.
12881 2022-09-26 Martin Liska <mliska@suse.cz>
12883 * config/s390/s390.cc (s390_rtx_costs): Remove dest variable
12886 2022-09-26 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
12888 * config/aarch64/aarch64-arches.def (armv9.1-a): Define.
12889 (armv9.2-a): Likewise.
12890 (armv9.3-a): Likewise.
12891 * config/aarch64/aarch64.h (AARCH64_FL_V9_1): Likewise.
12892 (AARCH64_FL_V9_2): Likewise.
12893 (AARCH64_FL_V9_3): Likewise.
12894 (AARCH64_FL_FOR_ARCH9_1): Likewise.
12895 (AARCH64_FL_FOR_ARCH9_2): Likewise.
12896 (AARCH64_FL_FOR_ARCH9_3): Likewise.
12897 (AARCH64_ISA_V9_1): Likewise.
12898 (AARCH64_ISA_V9_2): Likewise.
12899 (AARCH64_ISA_V9_3): Likewise.
12900 * doc/invoke.texi (AArch64 Options): Document armv9.1-a, armv9.2-a,
12901 armv9.3-a values to -march.
12903 2022-09-26 Martin Liska <mliska@suse.cz>
12905 * value-range.cc (tree_compare): Remove unused function.
12907 2022-09-26 Kewen Lin <linkw@linux.ibm.com>
12910 * config/rs6000/rs6000-logue.cc (rs6000_emit_epilogue): Update the
12911 condition for adding REG_CFA_DEF_CFA reg note with
12912 frame_pointer_needed_indeed.
12914 2022-09-26 Kewen Lin <linkw@linux.ibm.com>
12917 * config/rs6000/vector.md (vec_shr_<mode>): Replace condition
12918 TARGET_ALTIVEC with VECTOR_UNIT_ALTIVEC_OR_VSX_P.
12920 2022-09-26 Hongtao Liu <hongtao.liu@intel.com>
12921 Liwei Xu <liwei.xu@intel.com>
12924 * config/i386/i386-expand.cc (expand_vec_perm_shufps_shufps):
12926 (ix86_expand_vec_perm_const_1): Insert
12927 expand_vec_perm_shufps_shufps at the end of 2-instruction
12930 2022-09-25 Torbjörn SVENSSON <torbjorn.svensson@foss.st.com>
12932 * doc/sourcebuild.texi: Fix chapter level.
12934 2022-09-24 Jakub Jelinek <jakub@redhat.com>
12937 * omp-low.cc (lower_omp_taskgroup): Don't add GOMP_RETURN statement
12939 * omp-expand.cc (build_omp_regions_1): Clarify GF_OMP_TARGET_KIND_DATA
12940 is not stand-alone directive. For GIMPLE_OMP_TASKGROUP, also don't
12942 (omp_make_gimple_edges) <case GIMPLE_OMP_TASKGROUP>: Reset
12943 cur_region back after new_omp_region.
12945 2022-09-23 Vineet Gupta <vineetg@rivosinc.com>
12947 * config/riscv/riscv.h (LOCAL_SYM_P): New.
12948 (USE_LOAD_ADDRESS_MACRO): Simplify by calling LOCAL_SYM_P.
12950 2022-09-23 zhongjuzhe <juzhe.zhong@rivai.ai>
12952 * config/riscv/riscv-modes.def (VECTOR_BOOL_MODE): Add RVV mask modes.
12953 (ADJUST_NUNITS): Adjust nunits using riscv_vector_chunks.
12954 (ADJUST_ALIGNMENT): Adjust alignment.
12955 (ADJUST_BYTESIZE): Adjust bytesize using riscv_vector_chunks.
12956 (RVV_MODES): New macro.
12957 (VECTOR_MODE_WITH_PREFIX): Add RVV vector modes.
12958 (VECTOR_MODES_WITH_PREFIX): Add RVV vector modes.
12960 2022-09-23 zhongjuzhe <juzhe.zhong@rivai.ai>
12962 * common/config/riscv/riscv-common.cc: Change "static void" to "void".
12963 * config.gcc: Add riscv-selftests.o
12964 * config/riscv/predicates.md: Allow const_poly_int.
12965 * config/riscv/riscv-protos.h (riscv_reinit): New function.
12966 (riscv_parse_arch_string): change as exten function.
12967 (riscv_run_selftests): New function.
12968 * config/riscv/riscv.cc (riscv_cannot_force_const_mem): Don't allow poly
12970 (riscv_report_v_required): New function.
12971 (riscv_expand_op): New function.
12972 (riscv_expand_mult_with_const_int): New function.
12973 (riscv_legitimize_poly_move): Ditto.
12974 (riscv_legitimize_move): New function.
12975 (riscv_hard_regno_mode_ok): Add VL/VTYPE register allocation and fix
12977 (riscv_convert_vector_bits): Fix riscv_vector_chunks configuration for
12979 (riscv_reinit): New function.
12980 (TARGET_RUN_TARGET_SELFTESTS): New target hook support.
12981 * config/riscv/t-riscv: Add riscv-selftests.o.
12982 * config/riscv/riscv-selftests.cc: New file.
12984 2022-09-23 Richard Biener <rguenther@suse.de>
12986 PR tree-optimization/106922
12987 * tree-ssa-sccvn.cc (vn_reference_lookup_3): Allow
12988 an arbitrary number of same valued skipped stores.
12990 2022-09-23 Aldy Hernandez <aldyh@redhat.com>
12992 * value-range.cc (frange::set): Swap setters such that the one
12993 accepting REAL_VALUE_TYPE does all the work.
12995 2022-09-23 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
12997 * config/aarch64/aarch64-cores.def (neoverse-v2): New entry.
12998 (demeter): Update tunings to neoversev2.
12999 * config/aarch64/aarch64-tune.md: Regenerate.
13000 * config/aarch64/aarch64.cc (demeter_addrcost_table): Rename to
13001 neoversev2_addrcost_table.
13002 (demeter_regmove_cost): Rename to neoversev2_addrcost_table.
13003 (demeter_advsimd_vector_cost): Rename to neoversev2_advsimd_vector_cost.
13004 (demeter_sve_vector_cost): Rename to neoversev2_sve_vector_cost.
13005 (demeter_scalar_issue_info): Rename to neoversev2_scalar_issue_info.
13006 (demeter_advsimd_issue_info): Rename to neoversev2_advsimd_issue_info.
13007 (demeter_sve_issue_info): Rename to neoversev2_sve_issue_info.
13008 (demeter_vec_issue_info): Rename to neoversev2_vec_issue_info.
13009 Update references to above.
13010 (demeter_vector_cost): Rename to neoversev2_vector_cost.
13011 (demeter_tunings): Rename to neoversev2_tunings.
13012 (aarch64_vec_op_count::rename_cycles_per_iter): Use
13013 neoversev2_sve_issue_info instead of demeter_sve_issue_info.
13014 * doc/invoke.texi (AArch64 Options): Document neoverse-v2.
13016 2022-09-23 Aldy Hernandez <aldyh@redhat.com>
13018 * range-op-float.cc (build_le): Use vrp_val_*.
13022 * value-range.cc (frange::set): Chop ranges outside of the
13023 representable numbers for -ffinite-math-only.
13024 (frange::normalize_kind): Use vrp_val*.
13025 (frange::verify_range): Same.
13026 (frange::set_nonnegative): Same.
13027 (range_tests_floats): Remove tests that depend on -INF and +INF.
13028 * value-range.h (real_max_representable): Add prototype.
13029 (real_min_representable): Same.
13030 (vrp_val_max): Set max representable number for
13031 -ffinite-math-only.
13032 (vrp_val_min): Same but for min.
13033 (frange::set_varying): Use vrp_val*.
13035 2022-09-23 Aldy Hernandez <aldyh@redhat.com>
13037 * real.cc (debug): New.
13039 2022-09-23 Aldy Hernandez <aldyh@redhat.com>
13041 * value-range-pretty-print.cc (vrange_printer::print_real_value): New.
13042 (vrange_printer::visit): Call print_real_value.
13043 * value-range-pretty-print.h: New print_real_value.
13045 2022-09-23 Martin Liska <mliska@suse.cz>
13047 * common.opt: Update -flto-compression-level documentation.
13048 * opts.cc (print_filtered_help): Do not append range to an
13049 option that uses \t syntax.
13051 2022-09-23 Jakub Jelinek <jakub@redhat.com>
13053 * attribs.cc (decl_attributes): Improve diagnostics, instead of
13054 saying expected between 1 and 1, found 2 just say expected 1, found 2.
13056 2022-09-23 Hu, Lin1 <lin1.hu@intel.com>
13059 * config/i386/constraints.md (BH): New define_constraint.
13060 * config/i386/i386.cc (standard_sse_constant_p): Add return
13061 3/4 when operand matches new predicate.
13062 (standard_sse_constant_opcode): Add new alternative branch to
13064 * config/i386/predicates.md
13065 (vector_all_ones_zero_extend_half_operand): New define_predicate.
13066 (vector_all_ones_zero_extend_quarter_operand): Ditto.
13067 * config/i386/sse.md: Add constraint to insn "mov<mode>_internal".
13069 2022-09-22 Andrew MacLeod <amacleod@redhat.com>
13071 * gimple-range-fold.cc (range_of_range_op): Handle no operands.
13072 (range_of_call): Do not check for builtins.
13073 (fold_using_range::range_of_builtin_call): Delete.
13074 (fold_using_range::range_of_builtin_int_call): Delete.
13075 * gimple-range-fold.h: Adjust prototypes.
13076 * gimple-range-op.cc (class cfn_parity): New.
13077 (gimple_range_op_handler::maybe_builtin_call): Set arguments.
13079 2022-09-22 Andrew MacLeod <amacleod@redhat.com>
13081 * gimple-range-fold.cc (range_of_builtin_int_call): Remove case
13082 for CFN_GOACC_DIM_*.
13083 * gimple-range-op.cc (class cfn_goacc_dim): New.
13084 (gimple_range_op_handler::maybe_builtin_call): Set arguments.
13086 2022-09-22 Andrew MacLeod <amacleod@redhat.com>
13088 * gimple-range-fold.cc (range_of_builtin_int_call): Remove case
13089 for CFN_BUILT_IN_STRLEN.
13090 * gimple-range-op.cc (class cfn_strlen): New.
13091 (gimple_range_op_handler::maybe_builtin_call): Set arguments.
13093 2022-09-22 Andrew MacLeod <amacleod@redhat.com>
13095 * gimple-range-fold.cc (range_of_builtin_ubsan_call): Delete.
13096 (range_of_builtin_int_call): Remove cases for
13097 CFN_BUILT_IN_UBSAN_CHECK.
13098 * gimple-range-op.cc (class cfn_ubsan): New.
13099 (gimple_range_op_handler::maybe_builtin_call): Set arguments.
13101 2022-09-22 Andrew MacLeod <amacleod@redhat.com>
13103 * gimple-range-fold.cc (range_of_builtin_int_call): Remove case
13104 for CFN_BUILT_IN_CLRSB.
13105 * gimple-range-op.cc (class cfn_clrsb): New.
13106 (gimple_range_op_handler::maybe_builtin_call): Set arguments.
13108 2022-09-22 Andrew MacLeod <amacleod@redhat.com>
13110 * gimple-range-fold.cc (range_of_builtin_int_call): Remove case
13112 * gimple-range-op.cc (class cfn_ctz): New.
13113 (gimple_range_op_handler::maybe_builtin_call): Set arguments.
13115 2022-09-22 Andrew MacLeod <amacleod@redhat.com>
13117 * gimple-range-fold.cc (range_of_builtin_int_call): Remove case
13119 * gimple-range-op.cc (class cfn_clz): New.
13120 (gimple_range_op_handler::maybe_builtin_call): Set arguments.
13122 2022-09-22 Andrew MacLeod <amacleod@redhat.com>
13124 * gimple-range-fold.cc (range_of_builtin_int_call): Remove case
13125 for CFN_FFS and CFN_POPCOUNT.
13126 * gimple-range-op.cc (class cfn_pocount): New.
13127 (gimple_range_op_handler::maybe_builtin_call): Set arguments.
13129 2022-09-22 Andrew MacLeod <amacleod@redhat.com>
13131 * gimple-range-fold.cc (get_letter_range): Move to new class.
13132 (range_of_builtin_int_call): Remove case for CFN_BUILT_IN_TOUPPER
13133 and CFN_BUILT_IN_TOLOWER.
13134 * gimple-range-op.cc (class cfn_toupper_tolower): New.
13135 (gimple_range_op_handler::maybe_builtin_call): Set arguments.
13137 2022-09-22 Andrew MacLeod <amacleod@redhat.com>
13139 * gimple-range-fold.cc (range_of_builtin_int_call): Remove case
13140 for CFN_BUILT_IN_SIGNBIT.
13141 * gimple-range-op.cc (class cfn_signbit): New.
13142 (gimple_range_op_handler::maybe_builtin_call): Set arguments.
13144 2022-09-22 Andrew MacLeod <amacleod@redhat.com>
13146 * gimple-range-fold.cc
13147 (fold_using_range::range_of_builtin_int_call): Remove case for
13148 CFN_BUILT_IN_CONSTANT_P.
13149 * gimple-range-op.cc (gimple_range_op_handler::supported_p):
13150 Check if a call also creates a range-op object.
13151 (gimple_range_op_handler): Also check builtin calls.
13152 (class cfn_constant_float_p): New. Float CFN_BUILT_IN_CONSTANT_P.
13153 (class cfn_constant_p): New. Integral CFN_BUILT_IN_CONSTANT_P.
13154 (gimple_range_op_handler::maybe_builtin_call): Set arguments and
13155 handler for supported built-in calls.
13156 * gimple-range-op.h (maybe_builtin_call): New prototype.
13158 2022-09-22 Andrew MacLeod <amacleod@redhat.com>
13160 * gimple-range-fold.cc (range_of_range_op): Set result to
13161 VARYING if the call to fold_range fails.
13162 * tree-data-ref.cc (compute_distributive_range): Ditto.
13163 * tree-vrp.cc (range_fold_binary_expr): Ditto.
13164 (range_fold_unary_expr): Ditto.
13165 * value-query.cc (range_query::get_tree_range): Ditto.
13167 2022-09-22 Andrew MacLeod <amacleod@redhat.com>
13169 * range-op-float.cc (range_operator_float::fold_range): New base
13170 method for "int = float op int".
13171 * range-op.cc (range_op_handler::fold_range): New case.
13172 * range-op.h: Update prototypes.
13174 2022-09-22 Andrew MacLeod <amacleod@redhat.com>
13176 * gimple-range-op.cc (gimple_range_op_handler::calc_op1): Use
13177 operand 1 for second range if there is no operand 2.
13179 2022-09-22 Andrew MacLeod <amacleod@redhat.com>
13181 * Makefile.in (OBJS): Add gimple-range-op.o.
13182 * gimple-range-edge.cc (gimple_outgoing_range_stmt_p): Use
13183 gimple_range_op_handler.
13184 * gimple-range-fold.cc (gimple_range_base_of_assignment): Move
13185 to a method in gimple_range_op_handler.
13186 (gimple_range_operand1): Ditto.
13187 (gimple_range_operand2): Ditto.
13188 (fold_using_range::fold_stmt): Use gimple_range_op_handler.
13189 (fold_using_range::range_of_range_op): Ditto.
13190 (fold_using_range::relation_fold_and_or): Ditto.
13191 (fur_source::register_outgoing_edges): Ditto.
13192 (gimple_range_ssa_names): Relocate to gimple-range-op.cc.
13193 * gimple-range-fold.h: Adjust prototypes.
13194 * gimple-range-gori.cc (gimple_range_calc_op1): Move
13195 to a method in gimple_range_op_handler.
13196 (gimple_range_calc_op2): Ditto.
13197 (gori_compute::compute_operand_range): Use
13198 gimple_range_op_handler.
13199 (gori_compute::compute_logical_operands): Ditto.
13200 (compute_operand1_range): Ditto.
13201 (gori_compute::compute_operand2_range): Ditto.
13202 (gori_compute::compute_operand1_and_operand2_range): Ditto.
13203 * gimple-range-gori.h: Adjust protoypes.
13204 * gimple-range-op.cc: New. Supply gimple_range_op_handler methods.
13205 * gimple-range-op.h: New. Supply gimple_range_op_handler class.
13206 * gimple-range.cc (gimple_ranger::prefill_name): Use
13207 gimple_range_op_handler.
13208 (gimple_ranger::prefill_stmt_dependencies): Ditto.
13209 * gimple-range.h: Include gimple-range-op.h.
13210 * range-op.cc (range_op_handler::range_op_handler): Adjust and
13211 remove gimple * parameter option.
13212 * range-op.h: Adjust prototypes.
13214 2022-09-22 Andrew MacLeod <amacleod@redhat.com>
13216 * range-op.cc (range_op_handler::set_op_handler): Set new fields.
13217 (ange_op_handler::range_op_handler): Likewise.
13218 (range_op_handler::operator bool): Remove.
13219 (range_op_handler::fold_range): Use appropriate handler.
13220 (range_op_handler::op1_range): Likewise.
13221 (range_op_handler::op2_range): Likewise.
13222 (range_op_handler::lhs_op1_relation): Likewise.
13223 (range_op_handler::lhs_op2_relation): Likewise.
13224 (range_op_handler::op1_op2_relation): Likewise.
13225 * range-op.h (class range_op_handler): Store handler pointers.
13226 (range_op_handler:: operator bool): Inline.
13228 2022-09-22 Andrew MacLeod <amacleod@redhat.com>
13230 * tree-ssa-threadbackward.cc
13231 (back_threader::find_paths_to_names): Replace sequence with
13232 a call to gimple_range_ssa_names.
13234 2022-09-22 Martin Liska <mliska@suse.cz>
13235 Fangrui Song <i@maskray.me>
13237 * configure: Regenerate.
13238 * configure.ac: Simplify to gcc_cv_ld_compress_debug={0,1}
13239 and gcc_cv_as_compress_debug={0,1}.
13240 * doc/invoke.texi: Document the removal.
13241 * gcc.cc (LINK_COMPRESS_DEBUG_SPEC): Simplify and ignore
13243 (ASM_COMPRESS_DEBUG_SPEC): Likewise.
13245 2022-09-22 Richard Biener <rguenther@suse.de>
13247 PR tree-optimization/106922
13248 * tree-ssa-sccvn.cc (vn_walk_cb_data::same_val): New member.
13249 (vn_walk_cb_data::finish): Perform delayed verification of
13250 a skipped may-alias.
13251 (vn_reference_lookup_pieces): Likewise.
13252 (vn_reference_lookup): Likewise.
13253 (vn_reference_lookup_3): When skipping stores of the same
13254 value also handle constant stores that are more than a
13255 single VDEF away by delaying the verification.
13257 2022-09-22 Max Filippov <jcmvbkbc@gmail.com>
13259 * config/xtensa/xtensa.cc (TARGET_MAX_ANCHOR_OFFSET): New
13262 2022-09-22 Max Filippov <jcmvbkbc@gmail.com>
13264 * config/xtensa/xtensa.cc (xtensa_can_output_mi_thunk)
13265 (xtensa_output_mi_thunk): New functions.
13266 (TARGET_ASM_CAN_OUTPUT_MI_THUNK)
13267 (TARGET_ASM_OUTPUT_MI_THUNK): New macro definitions.
13268 (xtensa_prepare_expand_call): Use fixed register a8 as temporary
13269 when called with reload_completed set to 1.
13271 2022-09-22 Richard Biener <rguenther@suse.de>
13273 PR tree-optimization/99407
13274 * tree-ssa-dse.cc (dse_stmt_to_dr_map): New global.
13275 (dse_classify_store): Use data-ref analysis to disambiguate more uses.
13276 (pass_dse::use_dr_analysis_p): New pass parameter.
13277 (pass_dse::set_pass_param): Implement.
13278 (pass_dse::execute): Allocate and deallocate dse_stmt_to_dr_map.
13279 * passes.def: Allow DR analysis for the DSE pass before loop.
13281 2022-09-22 Richard Biener <rguenther@suse.de>
13283 * tree-ssa-sccvn.cc (can_track_predicate_on_edge): New
13284 function split out from ...
13285 (vn_nary_op_insert_pieces_predicated): ... here.
13287 2022-09-22 liuhongt <hongtao.liu@intel.com>
13290 * config/i386/mmx.md (floorv2sf2): Fix typo, use
13291 register_operand instead of vector_operand for operands[1].
13293 2022-09-21 Aldy Hernandez <aldyh@redhat.com>
13295 PR tree-optimization/106967
13296 * value-range.cc (frange::set): Set known NANs to undefined for
13297 flag_finite_math_only.
13299 2022-09-21 Aldy Hernandez <aldyh@redhat.com>
13301 * value-range.cc (frange::combine_zeros): Call set_undefined.
13302 (frange::intersect_nans): Same.
13303 (frange::intersect): Same.
13304 (frange::verify_range): Undefined ranges do not have a type.
13305 * value-range.h (frange::set_undefined): Clear NAN flags and type.
13307 2022-09-21 Richard Biener <rguenther@suse.de>
13309 PR tree-optimization/106984
13310 * tsan.cc (instrument_builtin_call): Build the COND_EXPR condition in
13311 a separate statement.
13313 2022-09-21 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
13315 * config/aarch64/aarch64.h (HAVE_LOCAL_CPU_DETECT,
13316 EXTRA_SPEC_FUNCTIONS, MCPU_MTUNE_NATIVE_SPECS): Move definitions up before
13317 OPTION_DEFAULT_SPECS.
13318 (MCPU_MTUNE_NATIVE_SPECS): Pass "cpu" to
13319 local_cpu_detect when rewriting -march=native and no -mcpu or -mtune
13321 (CONFIG_TUNE_SPEC): Define.
13322 (OPTION_DEFAULT_SPECS): Use CONFIG_TUNE_SPEC for "tune".
13324 2022-09-21 Aldy Hernandez <aldyh@redhat.com>
13326 PR tree-optimization/106967
13327 * range-op-float.cc (foperator_equal::fold_range): Adjust for NAN.
13328 (foperator_equal::op1_range): Same.
13329 (foperator_not_equal::fold_range): Same.
13330 (foperator_not_equal::op1_range): Same.
13331 (foperator_lt::fold_range): Same.
13332 (foperator_lt::op1_range): Same.
13333 (foperator_lt::op2_range): Same.
13334 (foperator_le::fold_range): Same.
13335 (foperator_le::op1_range): Same.
13336 (foperator_le::op2_range): Same.
13337 (foperator_gt::fold_range): Same.
13338 (foperator_gt::op1_range): Same.
13339 (foperator_gt::op2_range): Same.
13340 (foperator_ge::fold_range): Same.
13341 (foperator_ge::op1_range): Same.
13342 (foperator_ge::op2_range): Same.
13343 (foperator_unordered::op1_range): Same.
13344 (foperator_ordered::fold_range): Same.
13345 (foperator_ordered::op1_range): Same.
13346 (build_le): Assert that we don't have a NAN.
13351 2022-09-21 liuhongt <hongtao.liu@intel.com>
13353 PR tree-optimization/106963
13354 * tree-vect-loop.cc (vect_create_nonlinear_iv_init): Use
13355 vec_gen_perm_mask_any instead of vec_gen_perm_mask_check.
13357 2022-09-20 Aldy Hernandez <aldyh@redhat.com>
13359 * value-range.h (frange::maybe_isnan): Return false for
13362 2022-09-20 Aldy Hernandez <aldyh@redhat.com>
13364 * value-range.cc (frange::set_nonnegative): Set +NAN.
13365 (range_tests_signed_zeros): New test.
13366 * value-range.h (frange::update_nan): New overload to set NAN sign.
13368 2022-09-20 Richard Sandiford <richard.sandiford@arm.com>
13371 * config/aarch64/aarch64-sve-builtins.cc (scalar_types)
13372 (acle_vector_types, acle_svpattern, acle_svprfop): Add GTY
13373 markup to (new) extern declarations instead of to the main
13376 2022-09-20 Richard Sandiford <richard.sandiford@arm.com>
13378 PR tree-optimization/106794
13379 PR tree-optimization/106914
13380 * tree-vect-slp.cc (vect_optimize_slp_pass::internal_node_cost):
13381 Only consider loads that already have a permutation.
13382 (vect_optimize_slp_pass::start_choosing_layouts): Assert that
13383 loads with permutations are leaf nodes. Prevent any kind of grouped
13384 access from changing layout if it doesn't have a load permutation.
13386 2022-09-20 Richard Sandiford <richard.sandiford@arm.com>
13388 * tree-vect-data-refs.cc (vect_check_gather_scatter): Restrict
13389 early-out optimisation to SSA_NAMEs.
13391 2022-09-20 Martin Liska <mliska@suse.cz>
13393 * ctfc.cc (ctf_add_string): Replace "the the" with "the".
13394 * doc/md.texi: Likewise.
13395 * gimple-range-infer.cc (non_null_loadstore): Likewise.
13397 2022-09-20 liuhongt <hongtao.liu@intel.com>
13400 * config/i386/mmx.md (nearbyintv2sf2): New expander.
13401 (rintv2sf2): Ditto.
13402 (ceilv2sf2): Ditto.
13403 (lceilv2sfv2si2): Ditto.
13404 (floorv2sf2): Ditto.
13405 (lfloorv2sfv2si2): Ditto.
13406 (btruncv2sf2): Ditto.
13407 (lrintv2sfv2si2): Ditto.
13408 (roundv2sf2): Ditto.
13409 (lroundv2sfv2si2): Ditto.
13410 (*mmx_roundv2sf2): New define_insn.
13412 2022-09-20 konglin1 <lingling.kong@intel.com>
13414 PR middle-end/105735
13415 * tree-scalar-evolution.cc
13416 (analyze_and_compute_bitop_with_inv_effect): New function.
13417 (final_value_replacement_loop): Enhanced to handle bitop
13418 with inv induction.
13420 2022-09-20 Xi Ruoyao <xry111@xry111.site>
13422 * config/loongarch/gnu-user.h (GNU_USER_TARGET_LINK_SPEC): For
13423 -static-pie, pass -static -pie --no-dynamic-linker -z text to
13424 the linker, and do not pass --dynamic-linker.
13426 2022-09-20 Aldy Hernandez <aldyh@redhat.com>
13428 * value-range.cc (frange::flush_denormals_to_zero): New.
13429 (frange::set): Call flush_denormals_to_zero.
13430 * value-range.h (class frange): Add flush_denormals_to_zero.
13432 2022-09-20 liuhongt <hongtao.liu@intel.com>
13434 * config/i386/x86-tune-sched.cc (ix86_issue_rate): Adjust for
13435 latest Intel processors.
13437 2022-09-20 konglin1 <lingling.kong@intel.com>
13440 * config/i386/i386-expand.cc (ix86_expand_vector_init_duplicate):
13441 Fixed V16BF mode case.
13443 2022-09-19 Torbjörn SVENSSON <torbjorn.svensson@foss.st.com>
13445 * targhooks.cc (default_zero_call_used_regs): Improve sorry
13448 2022-09-18 Julian Brown <julian@codesourcery.com>
13450 * gimplify.cc (omp_segregate_mapping_groups): Update comment.
13451 (gimplify_adjust_omp_clauses): Move ATTACH and
13452 ATTACH_ZERO_LENGTH_ARRAY_SECTION nodes to the end of the clause list
13453 for offloaded OpenMP regions.
13455 2022-09-18 Jakub Jelinek <jakub@redhat.com>
13457 PR middle-end/106831
13458 * value-range.cc (frange::singleton_p): Avoid propagating long
13459 doubles that may have multiple representations.
13461 2022-09-18 Aldy Hernandez <aldyh@redhat.com>
13463 * range-op-float.cc (frange_add_zeros): Replace set_signbit with
13465 * value-query.cc (range_query::get_tree_range): Remove set_signbit
13467 * value-range-pretty-print.cc (vrange_printer::print_frange_prop):
13469 (vrange_printer::print_frange_nan): New.
13470 * value-range-pretty-print.h (print_frange_prop): Remove.
13471 (print_frange_nan): New.
13472 * value-range-storage.cc (frange_storage_slot::set_frange): Set
13473 kind and NAN fields.
13474 (frange_storage_slot::get_frange): Restore kind and NAN fields.
13475 * value-range-storage.h (class frange_storage_slot): Add kind and
13477 * value-range.cc (frange::update_nan): Remove.
13478 (frange::set_signbit): Remove.
13479 (frange::set): Adjust for NAN fields.
13480 (frange::normalize_kind): Remove m_props.
13481 (frange::combine_zeros): New.
13482 (frange::union_nans): New.
13483 (frange::union_): Handle new NAN fields.
13484 (frange::intersect_nans): New.
13485 (frange::intersect): Handle new NAN fields.
13486 (frange::operator=): Same.
13487 (frange::operator==): Same.
13488 (frange::contains_p): Same.
13489 (frange::singleton_p): Remove special case for signed zeros.
13490 (frange::verify_range): Adjust for new NAN fields.
13491 (frange::set_zero): Handle signed zeros.
13492 (frange::set_nonnegative): Same.
13493 (range_tests_nan): Adjust tests.
13494 (range_tests_signed_zeros): Same.
13495 (range_tests_signbit): Same.
13496 (range_tests_floats): Same.
13497 * value-range.h (class fp_prop): Remove.
13498 (FP_PROP_ACCESSOR): Remove.
13499 (class frange_props): Remove
13500 (frange::lower_bound): NANs don't have endpoints.
13501 (frange::upper_bound): Same.
13502 (frange_props::operator==): Remove.
13503 (frange_props::union_): Remove.
13504 (frange_props::intersect): Remove.
13505 (frange::update_nan): New.
13506 (frange::clear_nan): New.
13507 (frange::undefined_p): New.
13508 (frange::set_nan): New.
13509 (frange::known_finite): Adjust for new NAN representation.
13510 (frange::maybe_isnan): Same.
13511 (frange::known_isnan): Same.
13512 (frange::signbit_p): Same.
13513 * gimple-range-fold.cc (range_of_builtin_int_call): Rename
13514 known_signbit_p into signbit_p.
13516 2022-09-17 Jan-Benedict Glaw <jbglaw@lug-owl.de>
13518 * config/csky/csky.h (FUNCTION_ARG_REGNO_P): Cast REGNO to (int)
13519 to prevent warning.
13521 2022-09-17 Jakub Jelinek <jakub@redhat.com>
13523 PR tree-optimization/106958
13524 * tree-ssa-reassoc.cc (optimize_range_tests_cmp_bitwise): If
13525 id >= l, cast op to type1, otherwise to pointer_sized_int_node.
13526 If type has pointer type, cast exp to pointer_sized_int_node
13529 2022-09-16 liuhongt <hongtao.liu@intel.com>
13532 * config/i386/i386-builtins.cc
13533 (ix86_builtin_vectorized_function): Modernized with
13534 corresponding expanders.
13535 * config/i386/sse.md (lrint<mode><sseintvecmodelower>2): New
13537 (floor<mode>2): Ditto.
13538 (lfloor<mode><sseintvecmodelower>2): Ditto.
13539 (ceil<mode>2): Ditto.
13540 (lceil<mode><sseintvecmodelower>2): Ditto.
13541 (btrunc<mode>2): Ditto.
13542 (lround<mode><sseintvecmodelower>2): Ditto.
13543 (exp2<mode>2): Ditto.
13545 2022-09-15 Joseph Myers <joseph@codesourcery.com>
13547 * ginclude/float.h (INFINITY): Define only if
13548 [__FLT_HAS_INFINITY__].
13550 2022-09-15 Richard Biener <rguenther@suse.de>
13552 PR tree-optimization/106922
13553 * tree-ssa-pre.cc (translate_vuse_through_block): Only
13554 keep the VUSE if its def dominates PHIBLOCK.
13555 (prune_clobbered_mems): Rewrite logic so we check whether
13556 a value dies in a block when the VUSE def doesn't dominate it.
13558 2022-09-15 Richard Biener <rguenther@suse.de>
13560 * tree.cc (build_common_tree_nodes): Initialize void_list_node
13563 2022-09-15 Jiufu Guo <guojiufu@linux.ibm.com>
13566 * config/rs6000/rs6000.cc (rs6000_emit_set_long_const): Use pli.
13568 2022-09-14 Aldy Hernandez <aldyh@redhat.com>
13570 * range-op-float.cc (frange_add_zeros): New.
13571 (build_le): Call frange_add_zeros.
13573 (foperator_equal::op1_range): Same.
13574 (foperator_not_equal::op1_range): Same.
13576 2022-09-14 Aldy Hernandez <aldyh@redhat.com>
13578 * range-op-float.cc (build_le): Accept frange instead of number.
13582 (foperator_lt::op1_range): Pass full range to build_*.
13583 (foperator_lt::op2_range): Same.
13584 (foperator_le::op1_range): Same.
13585 (foperator_le::op2_range): Same.
13586 (foperator_gt::op1_range): Same.
13587 (foperator_gt::op2_range): Same.
13588 (foperator_ge::op1_range): Same.
13589 (foperator_ge::op2_range): Same.
13591 2022-09-14 Aldy Hernandez <aldyh@redhat.com>
13593 * value-range.cc (frange::set): Use set_nan.
13594 * value-range.h (frange::set_nan): Inline code originally in
13597 2022-09-14 Aldy Hernandez <aldyh@redhat.com>
13599 * range-op-float.cc (frange_set_nan): Remove.
13600 (build_lt): Use set_nan, update_nan, clear_nan.
13602 (foperator_equal::op1_range): Same.
13603 (foperator_not_equal::op1_range): Same.
13604 (foperator_lt::op1_range): Same.
13605 (foperator_lt::op2_range): Same.
13606 (foperator_le::op1_range): Same.
13607 (foperator_le::op2_range): Same.
13608 (foperator_gt::op1_range): Same.
13609 (foperator_gt::op2_range): Same.
13610 (foperator_ge::op1_range): Same.
13611 (foperator_ge::op2_range): Same.
13612 (foperator_unordered::op1_range): Same.
13613 (foperator_ordered::op1_range): Same.
13614 * value-query.cc (range_query::get_tree_range): Same.
13615 * value-range.cc (frange::set_nan): Same.
13616 (frange::update_nan): Same.
13617 (frange::union_): Same.
13618 (frange::intersect): Same.
13619 (range_tests_nan): Same.
13620 (range_tests_signed_zeros): Same.
13621 (range_tests_signbit): Same.
13622 (range_tests_floats): Same.
13623 * value-range.h (class frange): Add update_nan and clear_nan.
13624 (frange::set_nan): New.
13626 2022-09-14 Aldy Hernandez <aldyh@redhat.com>
13628 * value-query.cc (range_query::get_tree_range): Remove check for overflow.
13629 * value-range-pretty-print.cc (vrange_printer::visit): Move read
13630 of type until after undefined_p is checked.
13631 * value-range.cc (frange::set): Remove asserts for REAL_CST.
13632 (frange::contains_p): Tidy up.
13633 (range_tests_nan): Add comment.
13634 * value-range.h (frange::type): Check for undefined_p.
13635 (frange::set_undefined): Remove set of endpoints.
13637 2022-09-14 Julian Brown <julian@codesourcery.com>
13639 * gimplify.cc (omp_group_last): Allow GOMP_MAP_ATTACH_DETACH after
13640 GOMP_MAP_STRUCT (for reindexing).
13641 (omp_gather_mapping_groups): Reimplement using...
13642 (omp_gather_mapping_groups_1): This new function. Stop processing at
13644 (omp_group_base): Allow GOMP_MAP_TO_PSET without any following node.
13645 (omp_index_mapping_groups): Reimplement using...
13646 (omp_index_mapping_groups_1): This new function. Handle
13648 (omp_reindex_mapping_groups, omp_mapped_by_containing_struct): New
13650 (omp_tsort_mapping_groups_1): Adjust handling of base group being the
13651 same as current group. Use omp_mapped_by_containing_struct.
13652 (omp_build_struct_sibling_lists): Use omp_mapped_by_containing_struct
13653 and omp_reindex_mapping_groups. Robustify group deletion for reordered
13655 (gimplify_scan_omp_clauses): Update calls to
13656 omp_build_struct_sibling_lists.
13658 2022-09-14 Julian Brown <julian@codesourcery.com>
13660 * gimplify.cc (gimplify_omp_var_data): Remove GOVD_MAP_HAS_ATTACHMENTS.
13661 (GOMP_FIRSTPRIVATE_IMPLICIT): Renumber.
13662 (insert_struct_comp_map): Refactor function into...
13663 (build_omp_struct_comp_nodes): This new function. Remove list handling
13664 and improve self-documentation.
13665 (extract_base_bit_offset): Remove BASE_REF, OFFSETP parameters. Move
13666 code to strip outer parts of address out of function, but strip no-op
13668 (omp_mapping_group): Add DELETED field for use during reindexing.
13669 (omp_strip_components_and_deref, omp_strip_indirections): New functions.
13670 (omp_group_last, omp_group_base): Add GOMP_MAP_STRUCT handling.
13671 (omp_gather_mapping_groups): Initialise DELETED field for new groups.
13672 (omp_index_mapping_groups): Notice DELETED groups when (re)indexing.
13673 (omp_siblist_insert_node_after, omp_siblist_move_node_after,
13674 omp_siblist_move_nodes_after, omp_siblist_move_concat_nodes_after): New
13676 (omp_accumulate_sibling_list): New function to build up GOMP_MAP_STRUCT
13677 node groups for sibling lists. Outlined from gimplify_scan_omp_clauses.
13678 (omp_build_struct_sibling_lists): New function.
13679 (gimplify_scan_omp_clauses): Remove struct_map_to_clause,
13680 struct_seen_clause, struct_deref_set. Call
13681 omp_build_struct_sibling_lists as pre-pass instead of handling sibling
13682 lists in the function's main processing loop.
13683 (gimplify_adjust_omp_clauses_1): Remove GOVD_MAP_HAS_ATTACHMENTS
13684 handling, unused now.
13685 * omp-low.cc (scan_sharing_clauses): Handle pointer-type indirect
13686 struct references, and references to pointers to structs also.
13688 2022-09-14 Richard Biener <rguenther@suse.de>
13690 PR tree-optimization/106938
13691 * tree-cfg.cc (execute_fixup_cfg): Purge dead abnormal
13692 edges for all last stmts in a block. Do EH cleanup
13693 only on the last stmt in a block.
13695 2022-09-14 Aldy Hernandez <aldyh@redhat.com>
13697 PR tree-optimization/106936
13698 * value-query.cc (range_query::get_value_range): Remove assert.
13700 2022-09-14 Jan-Benedict Glaw <jbglaw@lug-owl.de>
13702 * config/mips/mips.cc (mips_option_override): Drop unused variable.
13704 2022-09-14 Julian Brown <julian@codesourcery.com>
13706 * gimplify.cc (is_or_contains_p, omp_target_reorder_clauses): Delete
13708 (omp_tsort_mark): Add enum.
13709 (omp_mapping_group): Add struct.
13710 (debug_mapping_group, omp_get_base_pointer, omp_get_attachment,
13711 omp_group_last, omp_gather_mapping_groups, omp_group_base,
13712 omp_index_mapping_groups, omp_containing_struct,
13713 omp_tsort_mapping_groups_1, omp_tsort_mapping_groups,
13714 omp_segregate_mapping_groups, omp_reorder_mapping_groups): New
13716 (gimplify_scan_omp_clauses): Call above functions instead of
13717 omp_target_reorder_clauses, unless we've seen an error.
13718 * omp-low.cc (scan_sharing_clauses): Avoid strict test if we haven't
13719 sorted mapping groups.
13721 2022-09-14 Jakub Jelinek <jakub@redhat.com>
13723 PR tree-optimization/106878
13724 * tree-cfg.cc (verify_gimple_assign_binary): Disallow pointer,
13725 reference or OFFSET_TYPE BIT_IOR_EXPR, BIT_XOR_EXPR or, unless
13726 the second argument is INTEGER_CST, BIT_AND_EXPR.
13727 * match.pd ((type) X op CST -> (type) (X op ((type-x) CST)),
13728 (type) (((type2) X) op Y) -> (X op (type) Y)): Punt for
13729 POINTER_TYPE_P or OFFSET_TYPE.
13730 * tree-ssa-reassoc.cc (optimize_range_tests_cmp_bitwise): For
13731 pointers cast them to pointer sized integers first.
13733 2022-09-14 Richard Biener <rguenther@suse.de>
13735 PR tree-optimization/106934
13736 * tree-ssa.cc (non_rewritable_mem_ref_base): Avoid BIT_FIELD_REFs
13738 (maybe_rewrite_mem_ref_base): Likewise.
13740 2022-09-14 liuhongt <hongtao.liu@intel.com>
13742 PR tree-optimization/106905
13743 * tree-vect-loop.cc (vectorizable_nonlinear_induction): Return
13744 false when !vect_use_loop_mask_for_alignment_p (loop_vinfo) &&
13745 LOOP_VINFO_PEELING_FOR_ALIGNMENT (loop_vinfo) < 0.
13747 2022-09-13 Roger Sayle <roger@nextmovesoftware.com>
13750 * reg-stack.cc (move_for_stack_reg): Check for any_malformed_asm
13753 2022-09-13 Max Filippov <jcmvbkbc@gmail.com>
13756 2022-09-12 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
13758 * config/xtensa/xtensa.cc (TARGET_CONSTANT_OK_FOR_CPROP_P):
13759 New macro definition.
13760 (xtensa_constant_ok_for_cprop_p):
13761 Implement the hook as mentioned above.
13763 2022-09-13 Kewen Lin <linkw@linux.ibm.com>
13766 * config/rs6000/rs6000-c.cc (altivec_resolve_overloaded_builtin): Fix
13767 the equality check for argument number, and move this hunk ahead.
13769 2022-09-13 Kewen.Lin <linkw@gcc.gnu.org>
13772 * config/rs6000/rs6000-builtin.cc (rs6000_gimple_fold_builtin): Add
13773 the handling for unresolved overloaded builtin function.
13774 (rs6000_expand_builtin): Likewise.
13776 2022-09-13 Kewen Lin <linkw@linux.ibm.com>
13778 * config/rs6000/rs6000.cc (class rs6000_cost_data): Add new members
13779 m_nstores, m_reduc_factor, m_gather_load and member function
13780 determine_suggested_unroll_factor.
13781 (rs6000_cost_data::update_target_cost_per_stmt): Update for m_nstores,
13782 m_reduc_factor and m_gather_load.
13783 (rs6000_cost_data::determine_suggested_unroll_factor): New function.
13784 (rs6000_cost_data::finish_cost): Use determine_suggested_unroll_factor.
13785 * config/rs6000/rs6000.opt (rs6000-vect-unroll-limit): New parameter.
13786 (rs6000-vect-unroll-issue): Likewise.
13787 (rs6000-vect-unroll-reduc-threshold): Likewise.
13788 * doc/invoke.texi (rs6000-vect-unroll-limit): Document new parameter.
13790 2022-09-13 Richard Biener <rguenther@suse.de>
13792 PR middle-end/106909
13793 * gimple-fold.cc (gimple_fold_call): Clear the ctrl-altering
13794 flag of a unreachable call.
13796 2022-09-13 Richard Biener <rguenther@suse.de>
13798 PR tree-optimization/106913
13799 * tree-ssa-uninit.cc (warn_uninitialized_vars): Do not set
13800 ft_reachable on EXIT_BLOCK.
13802 2022-09-13 Richard Sandiford <richard.sandiford@arm.com>
13804 * config/aarch64/aarch64.cc (aarch64_classify_vector_mode): Use
13805 TARGET_FLOAT instead of TARGET_SIMD.
13806 (aarch64_vectorize_related_mode): Restrict ADVSIMD handling to
13808 (aarch64_hard_regno_mode_ok): Don't allow tuples of 2 64-bit vectors
13810 (aarch64_classify_address): Treat little-endian structure moves
13811 like big-endian for TARGET_FLOAT && !TARGET_SIMD.
13812 (aarch64_secondary_memory_needed): New function.
13813 (aarch64_secondary_reload): Handle 128-bit Advanced SIMD vectors
13814 in the same way as TF, TI and TD.
13815 (aarch64_rtx_mult_cost): Restrict ADVSIMD handling to TARGET_SIMD.
13816 (aarch64_rtx_costs): Likewise.
13817 (aarch64_register_move_cost): Treat a pair of 64-bit vectors
13818 separately from a single 128-bit vector. Handle the cost implied
13819 by aarch64_secondary_memory_needed.
13820 (aarch64_simd_valid_immediate): Restrict ADVSIMD handling to
13822 (aarch64_expand_vec_perm_const_1): Likewise.
13823 (TARGET_SECONDARY_MEMORY_NEEDED): New macro.
13824 * config/aarch64/iterators.md (VTX): New iterator.
13825 * config/aarch64/aarch64.md (arches): Add fp_q as a synonym of simd.
13826 (arch_enabled): Adjust accordingly.
13827 (@aarch64_reload_mov<TX:mode>): Extend to...
13828 (@aarch64_reload_mov<VTX:mode>): ...this.
13829 * config/aarch64/aarch64-simd.md (mov<mode>): Require TARGET_FLOAT
13830 rather than TARGET_SIMD.
13831 (movmisalign<mode>): Likewise.
13832 (load_pair<DREG:mode><DREG2:mode>): Likewise.
13833 (vec_store_pair<DREG:mode><DREG2:mode>): Likewise.
13834 (load_pair<VQ:mode><VQ2:mode>): Likewise.
13835 (vec_store_pair<VQ:mode><VQ2:mode>): Likewise.
13836 (@aarch64_split_simd_mov<mode>): Likewise.
13837 (aarch64_get_low<mode>): Likewise.
13838 (aarch64_get_high<mode>): Likewise.
13839 (aarch64_get_half<mode>): Likewise. Canonicalize to a move for
13841 (*aarch64_simd_mov<VDMOV:mode>): Require TARGET_FLOAT rather than
13842 TARGET_SIMD. Use different w<-w and r<-w instructions for
13843 !TARGET_SIMD. Disable immediate moves for !TARGET_SIMD but
13844 add an alternative specifically for w<-Z.
13845 (*aarch64_simd_mov<VQMOV:mode>): Require TARGET_FLOAT rather than
13846 TARGET_SIMD. Likewise for the associated define_splits. Disable
13847 FPR moves and immediate moves for !TARGET_SIMD but add an alternative
13848 specifically for w<-Z.
13849 (aarch64_simd_mov_from_<mode>high): Require TARGET_FLOAT rather than
13850 TARGET_SIMD. Restrict the existing alternatives to TARGET_SIMD
13851 but add a new r<-w one for !TARGET_SIMD.
13852 (*aarch64_get_high<mode>): New pattern.
13853 (load_pair_lanes<mode>): Require TARGET_FLOAT rather than TARGET_SIMD.
13854 (store_pair_lanes<mode>): Likewise.
13855 (*aarch64_combine_internal<mode>): Likewise. Restrict existing
13856 w<-w, w<-r and w<-m alternatives to TARGET_SIMD but add a new w<-r
13857 alternative for !TARGET_SIMD.
13858 (*aarch64_combine_internal_be<mode>): Likewise.
13859 (aarch64_combinez<mode>): Require TARGET_FLOAT rather than TARGET_SIMD.
13860 Remove bogus arch attribute.
13861 (*aarch64_combinez_be<mode>): Likewise.
13862 (@aarch64_vec_concat<mode>): Require TARGET_FLOAT rather than
13864 (aarch64_combine<mode>): Likewise.
13865 (aarch64_rev_reglist<mode>): Likewise.
13866 (mov<mode>): Likewise.
13867 (*aarch64_be_mov<VSTRUCT_2D:mode>): Extend to TARGET_FLOAT &&
13868 !TARGET_SIMD, regardless of endianness. Extend associated
13869 define_splits in the same way, both for this pattern and the
13871 (*aarch64_be_mov<VSTRUCT_2Qmode>): Likewise. Restrict w<-w
13872 alternative to TARGET_SIMD.
13873 (*aarch64_be_movoi): Likewise.
13874 (*aarch64_be_movci): Likewise.
13875 (*aarch64_be_movxi): Likewise.
13876 (*aarch64_be_mov<VSTRUCT_4QD:mode>): Extend to TARGET_FLOAT
13877 && !TARGET_SIMD, regardless of endianness. Restrict w<-w alternative
13878 to TARGET_SIMD for tuples of 128-bit vectors.
13879 (*aarch64_be_mov<VSTRUCT_4QD:mode>): Likewise.
13880 * config/aarch64/aarch64-ldpstp.md: Remove TARGET_SIMD condition
13881 from DREG STP peephole. Change TARGET_SIMD to TARGET_FLOAT in
13882 the VQ and VP_2E LDP and STP peepholes.
13884 2022-09-13 Richard Sandiford <richard.sandiford@arm.com>
13886 * config/aarch64/aarch64-simd.md (movv8di): Remove TARGET_SIMD
13887 condition. Likewise for the related define_split. Tweak formatting.
13889 2022-09-12 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
13891 * config/xtensa/xtensa.cc (TARGET_CONSTANT_OK_FOR_CPROP_P):
13892 New macro definition.
13893 (xtensa_constant_ok_for_cprop_p):
13894 Implement the hook as mentioned above.
13896 2022-09-12 Joseph Myers <joseph@codesourcery.com>
13898 * ginclude/stdatomic.h [defined __STDC_VERSION__ &&
13899 __STDC_VERSION__ > 201710L] (ATOMIC_VAR_INIT): Do not define.
13901 2022-09-12 Tobias Burnus <tobias@codesourcery.com>
13903 * config/nvptx/mkoffload.cc (process): Replace a fatal_error by
13904 a warning + not enabling offloading if -misa=sm_30 prevents
13906 (main): Use tool_name as progname for diagnostic.
13907 * config/gcn/mkoffload.cc (main): Likewise.
13909 2022-09-12 Aldy Hernandez <aldyh@redhat.com>
13911 * value-range.cc (frange::set_signbit): Avoid changing sign when
13912 already in the correct sign.
13914 2022-09-12 Max Filippov <jcmvbkbc@gmail.com>
13916 * config/xtensa/xtensa.cc (xtensa_function_value_regno_p):
13917 Recognize all 4 return registers.
13918 * config/xtensa/xtensa.h (GP_RETURN_REG_COUNT): New definition.
13919 * config/xtensa/xtensa.md (untyped_call): New pattern.
13921 2022-09-12 Jonathan Wakely <jwakely@redhat.com>
13923 * doc/extend.texi (Floating Types): Fix "_float128" typo.
13925 2022-09-10 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
13927 * config/xtensa/xtensa.cc (xtensa_constantsynth):
13928 Add new pattern for the abovementioned case.
13930 2022-09-10 Akari Takahashi <akaritakahashioss@gmail.com>
13931 Segher Boessenkool <segher@kernel.crashing.org>
13933 * config/rs6000/rs6000.cc (get_memref_parts): Regularize some code.
13935 2022-09-09 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
13937 * config/xtensa/xtensa.md: Rewrite the split pattern that performs
13938 the abovementioned process so that insns that overwrite clobbered
13939 register no longer need to be contiguous.
13940 (DSC): Remove as no longer needed.
13942 2022-09-09 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
13944 * config/xtensa/xtensa.cc (machine_function): New boolean member as
13945 a flag that controls whether to emit the insns for stack pointer
13946 adjustment inside of the pro/epilogue.
13947 (xtensa_emit_adjust_stack_ptr): New function to share the common
13948 codes and to emit insns if not inhibited.
13949 (xtensa_expand_epilogue): Change to use the function mentioned
13950 above when using the CALL0 ABI.
13951 (xtensa_expand_prologue): Ditto.
13952 And also change to set the inhibit flag used by
13953 xtensa_emit_adjust_stack_ptr() to true if the stack pointer is only
13954 used for its own adjustment.
13956 2022-09-09 David Malcolm <dmalcolm@redhat.com>
13958 * doc/invoke.texi (Static Analyzer Options): Add
13959 -Wanalyzer-exposure-through-uninit-copy.
13961 2022-09-09 David Malcolm <dmalcolm@redhat.com>
13963 * Makefile.in (ANALYZER_OBJS): Add
13964 analyzer/known-function-manager.o.
13966 2022-09-09 Tobias Burnus <tobias@codesourcery.com>
13968 * config/nvptx/mkoffload.cc (struct id_map): Add 'dim' member.
13969 (record_id): Store func name without quotes, store dim separately.
13970 (process): For GOMP_REQUIRES_REVERSE_OFFLOAD, check that -march is
13971 at least sm_35, create '$offload_func_table' global array and init
13972 with reverse-offload function addresses.
13973 * config/nvptx/nvptx.cc (write_fn_proto_1, write_fn_proto): New
13974 force_public attribute to force .visible.
13975 (nvptx_declare_function_name): For "omp target
13976 device_ancestor_nohost" attribut, force .visible/TREE_PUBLIC.
13978 2022-09-09 Tobias Burnus <tobias@codesourcery.com>
13980 * config/gcn/mkoffload.cc (process_asm): Create .offload_func_table,
13981 similar to pre-existing .offload_var_table.
13983 2022-09-09 Joseph Myers <joseph@codesourcery.com>
13985 * ginclude/stddef.h [__STDC_VERSION__ > 201710L] (unreachable):
13988 2022-09-09 Kewen Lin <linkw@linux.ibm.com>
13990 PR middle-end/106833
13991 * tree.cc (verify_opaque_type): New function.
13992 (verify_type): Call verify_opaque_type for OPAQUE_TYPE.
13994 2022-09-09 Kwok Cheung Yeung <kcy@codesourcery.com>
13996 * config/gcn/gcn-builtins.def (FABSVF, LDEXPVF, LDEXPV, FREXPVF_EXP,
13997 FREXPVF_MANT, FREXPV_EXP, FREXPV_MANT): Add new builtins.
13998 * config/gcn/gcn-protos.h (gcn_dconst1over2pi): New prototype.
13999 * config/gcn/gcn-valu.md (MATH_UNOP_1OR2REG, MATH_UNOP_1REG,
14000 MATH_UNOP_TRIG): New iterators.
14001 (math_unop): New attributes.
14002 (<math_unop><mode>2, <math_unop><mode>2<exec>,
14003 <math_unop><mode>2, <math_unop><mode>2<exec>,
14004 *<math_unop><mode>2_insn, *<math_unop><mode>2<exec>_insn,
14005 ldexp<mode>3, ldexp<mode>3<exec>,
14006 frexp<mode>_exp2, frexp<mode>_mant2,
14007 frexp<mode>_exp2<exec>, frexp<mode>_mant2<exec>): New instructions.
14008 (<math_unop><mode>2, <math_unop><mode>2<exec>): New expanders.
14009 * config/gcn/gcn.cc (init_ext_gcn_constants): Update definition of
14011 (gcn_dconst1over2pi): New.
14012 (gcn_builtin_type_index): Add entry for v64df type.
14013 (v64df_type_node): New.
14014 (gcn_init_builtin_types): Initialize v64df_type_node.
14015 (gcn_expand_builtin_1): Expand new builtins to instructions.
14016 (print_operand): Fix assembler output for 1/(2*PI) constant.
14017 * config/gcn/gcn.md (unspec): Add new entries.
14019 2022-09-09 Richard Biener <rguenther@suse.de>
14021 PR tree-optimization/106722
14022 * tree-predcom.cc (ref_at_iteration): Do not associate the
14023 constant part of the offset into the MEM_REF offset
14024 operand, across a non-zero offset.
14026 2022-09-09 Kito Cheng <kito.cheng@sifive.com>
14028 * common/config/riscv/riscv-common.cc (RISCV_USE_CUSTOMISED_MULTI_LIB):
14029 Move forward for cover all all necessary functions for suppress
14030 unused function warnings.
14031 (riscv_multi_lib_check): Move forward, and tweak message to suppress
14032 -Werror=format-diag warning.
14034 2022-09-09 Richard Biener <rguenther@suse.de>
14036 PR tree-optimization/106881
14037 * gimple-predicate-analysis.cc (simple_control_dep_chain):
14038 Add only non-fallthru edges and avoid the same set of edges
14039 as compute_control_dep_chain_pdom does.
14041 2022-09-09 Jan-Benedict Glaw <jbglaw@lug-owl.de>
14043 * doc/tm.texi.in (TARGET_OVERRIDES_FORMAT_ATTRIBUTES): Document requirement
14044 of TARGET_OVERRIDES_FORMAT_ATTRIBUTES_COUNT being defined as well.
14045 * doc/tm.texi: Regenerate.
14047 2022-09-09 Martin Liska <mliska@suse.cz>
14049 * optc-save-gen.awk: Always compare array option values with memcmp.
14051 2022-09-08 Jonathan Wakely <jwakely@redhat.com>
14054 * doc/extend.texi (Type Traits): Fix requirements. Document
14055 __is_aggregate and __is_final.
14057 2022-09-08 Tim Lange <mail@tim-lange.me>
14061 State that the checker also reasons about symbolic values.
14063 2022-09-08 Richard Sandiford <richard.sandiford@arm.com>
14065 PR tree-optimization/106886
14066 * tree-vect-slp.cc (vect_optimize_slp_pass::get_result_with_layout):
14067 Fix copying of scalar stmts.
14069 2022-09-08 Chung-Lin Tang <cltang@codesourcery.com>
14071 * config/nios2/linux.h (MUSL_DYNAMIC_LINKER): Add #undef before #define.
14073 2022-09-08 Richard Biener <rguenther@suse.de>
14075 PR middle-end/106870
14076 * gimple-harden-conditionals.cc (insert_check_and_trap):
14077 Set the control-altering flag on the built IFN_TRAP.
14078 * gimple.cc (gimple_build_builtin_unreachable): Likewise.
14079 * tree-cfg.cc (handle_abnormal_edges): Set the control-altering
14080 flag on the .ABNORMAL_DISPATCHER call.
14081 * tree-cfgcleanup.cc (cleanup_call_ctrl_altering_flag): Avoid
14082 resetting the control altering flag for ECF_NORETURN calls.
14083 (cleanup_control_flow_bb): Set the control altering flag on
14084 discovered noreturn calls.
14085 * symtab-thunks.cc (expand_thunk): Set the control altering
14086 flag for the noreturn tailcall case.
14087 * tree-eh.cc (lower_resx): Likewisw for trap and unwind_resume
14090 2022-09-08 Jakub Jelinek <jakub@redhat.com>
14092 * omp-expand.cc (expand_omp_ordered_sink): Add CONT_BB argument.
14093 Add doacross(sink:omp_cur_iteration-1) support.
14094 (expand_omp_ordered_source_sink): Clear counts[fd->ordered + 1].
14095 Adjust expand_omp_ordered_sink caller.
14096 (expand_omp_for_ordered_loops): If counts[fd->ordered + 1] is
14097 non-NULL, set that variable to true at the start of outermost
14098 non-collapsed loop and set it to false at the end of innermost
14100 (expand_omp_for_generic): If fd->ordered, allocate
14101 1 + (fd->ordered - fd->collapse) further elements in counts array.
14102 Copy to counts + 2 + fd->ordered the counts of fd->collapse ..
14103 fd->ordered - 1 loop if any.
14105 2022-09-08 Richard Biener <rguenther@suse.de>
14107 PR tree-optimization/106881
14108 * gimple-predicate-analysis.cc (compute_control_dep_chain_pdom):
14109 Add only non-fallthru edges and avoid the same set of edges
14110 as the caller does.
14112 2022-09-08 Aldy Hernandez <aldyh@redhat.com>
14114 * gimple-range-fold.cc
14115 (fold_using_range::range_of_builtin_int_call): Use fpclassify like API.
14116 * range-op-float.cc (finite_operand_p): Same.
14117 (finite_operands_p): Same.
14118 (foperator_lt::fold_range): Same.
14119 (foperator_le::fold_range): Same.
14120 (foperator_gt::fold_range): Same.
14121 (foperator_ge::fold_range): Same.
14122 (foperator_unordered::fold_range): Same.
14123 (foperator_unordered::op1_range): Same.
14124 (foperator_ordered::fold_range): Same.
14125 * value-range.cc (frange::set_nan): Same.
14126 (frange::set_signbit): Same.
14127 (frange::union_): Same.
14128 (frange::intersect): Same.
14129 (frange::operator==): Same.
14130 (frange::singleton_p): Same.
14131 (frange::verify_range): Same.
14132 (range_tests_nan): Same.
14133 (range_tests_floats): Same.
14134 * value-range.h(frange::known_finite): New.
14135 (frange::maybe_inf): New.
14136 (frange::known_inf): New.
14137 (frange::maybe_nan): New.
14138 (frange::known_nan): New.
14139 (frange::known_signbit): New.
14141 2022-09-08 Iain Buclaw <ibuclaw@gdcproject.org>
14143 * config/darwin-d.cc: Include tm.h.
14144 * config/dragonfly-d.cc: Likewise.
14145 * config/freebsd-d.cc: Remove memmodel.h.
14146 * config/glibc-d.cc: Likewise.
14147 * config/netbsd-d.cc: Include tm.h.
14148 * config/openbsd-d.cc: Likewise.
14149 * config/sol2-d.cc: Likewise.
14151 2022-09-08 Christophe Lyon <christophe.lyon@arm.com>
14153 * config/arm/mve.md (mve_vqshluq_n_s<mode>): Use
14154 MVE_pred/MVE_constraint instead of mve_imm_7/Ra.
14155 (mve_vqshluq_m_n_s<mode>): Likewise.
14156 (mve_vqrshrnbq_n_<supf><mode>): Use MVE_pred3/MVE_constraint3
14157 instead of mve_imm_8/Rb.
14158 (mve_vqrshrunbq_n_s<mode>): Likewise.
14159 (mve_vqrshrntq_n_<supf><mode>): Likewise.
14160 (mve_vqrshruntq_n_s<mode>): Likewise.
14161 (mve_vrshrnbq_n_<supf><mode>): Likewise.
14162 (mve_vrshrntq_n_<supf><mode>): Likewise.
14163 (mve_vqrshrnbq_m_n_<supf><mode>): Likewise.
14164 (mve_vqrshrntq_m_n_<supf><mode>): Likewise.
14165 (mve_vrshrnbq_m_n_<supf><mode>): Likewise.
14166 (mve_vrshrntq_m_n_<supf><mode>): Likewise.
14167 (mve_vqrshrunbq_m_n_s<mode>): Likewise.
14168 (mve_vsriq_n_<supf><mode): Use MVE_pred2/MVE_constraint2 instead
14169 of mve_imm_selective_upto_8/Rg.
14170 (mve_vsriq_m_n_<supf><mode>): Likewise.
14172 2022-09-08 Jiufu Guo <guojiufu@linux.ibm.com>
14174 * config/rs6000/rs6000.md (splitter for set to and_mask constants):
14175 Use int_reg_operand (instead of int_reg_operand_not_pseudo).
14176 (splitter for multi-insn constant loads): Ditto.
14178 2022-09-08 Sebastian Huber <sebastian.huber@embedded-brains.de>
14180 * config/rs6000/rtems.h (CPP_OS_DEFAULT_SPEC): Define __PPC_VRSAVE__ if
14181 -mvrsave is present.
14182 * config/rs6000/t-rtems: Add -mvrsave multilib variants for
14185 2022-09-07 Martin Liska <mliska@suse.cz>
14187 * configure.ac: Restore detection of HAVE_XCOFF_DWARF_EXTRAS.
14188 * config/rs6000/rs6000.cc (HAVE_XCOFF_DWARF_EXTRAS): Reset it.
14189 * configure: Regenerate.
14190 * config.in: Regenerate.
14192 2022-09-07 Surya Kumari Jangala <jskumari@linux.ibm.com>
14194 PR rtl-optimization/105586
14195 * sched-rgn.cc (save_state_for_fallthru_edge): New function.
14196 (schedule_region): Use it for all blocks.
14198 2022-09-07 Joseph Myers <joseph@codesourcery.com>
14200 * ginclude/stdalign.h [defined __STDC_VERSION__ &&
14201 __STDC_VERSION__ > 201710L]: Disable all content.
14202 * ginclude/stdbool.h [defined __STDC_VERSION__ && __STDC_VERSION__
14203 > 201710L] (bool, true, false): Do not define.
14205 2022-09-07 Martin Liska <mliska@suse.cz>
14207 PR bootstrap/106855
14208 * collect2.cc (scan_prog_file): Restore if XCOFF_DEBUGGING_INFO.
14209 * config/rs6000/rs6000.cc (rs6000_option_override_internal):
14210 Restore usage of XCOFF_DEBUGGING_INFO.
14211 * config/rs6000/xcoff.h (XCOFF_DEBUGGING_INFO): Restore.
14212 * dwarf2asm.cc (XCOFF_DEBUGGING_INFO): Restore support for
14213 XCOFF_DEBUGGING_INFO.
14214 (dw2_asm_output_nstring): Likewise.
14215 (USE_LINKONCE_INDIRECT): Likewise.
14216 * dwarf2out.cc (XCOFF_DEBUGGING_INFO): Likewise.
14217 (HAVE_XCOFF_DWARF_EXTRAS): Likewise.
14218 (output_fde): Likewise.
14219 (output_call_frame_info): Likewise.
14220 (have_macinfo): Likewise.
14221 (add_AT_loc_list): Likewise.
14222 (add_AT_view_list): Likewise.
14223 (output_compilation_unit_header): Likewise.
14224 (output_pubnames): Likewise.
14225 (output_aranges): Likewise.
14226 (output_line_info): Likewise.
14227 (output_macinfo): Likewise.
14228 (dwarf2out_finish): Likewise.
14229 (dwarf2out_early_finish): Likewise.
14231 2022-09-07 Richard Biener <rguenther@suse.de>
14233 PR tree-optimization/106866
14234 * tree-ssa-dce.cc (eliminate_unnecessary_stmts): When
14235 we changed cfun->calls_setjmp make sure to purge all
14236 abnormal call edges.
14238 2022-09-07 Xianmiao Qu <cooper.qu@linux.alibaba.com>
14240 * config/csky/csky.cc (csky_emit_compare_float): Fix the expanding of
14241 float LE comparing with zero for fpuv3.
14242 * config/csky/csky.h (TARGET_SUPPORT_FPV2): New, true if any fpuv2
14243 features are enabled.
14245 2022-09-07 Richard Biener <rguenther@suse.de>
14247 PR tree-optimization/106860
14248 * tree-ssa-loop-split.cc (split_loop): Find the exit to
14249 latch edge from the loop exit edge instead of from the
14250 latch. Verify we're going to find it.
14252 2022-09-07 Richard Biener <rguenther@suse.de>
14254 * gimple-predicate-analysis.cc (dfs_mark_dominating_region):
14255 Adjust to take the region exit source as argument.
14256 (uninit_analysis::init_from_phi_def): Adjust.
14257 (uninit_analysis::init_use_preds): Mark the dominating region
14258 before computing control dependences.
14260 2022-09-07 Richard Sandiford <richard.sandiford@arm.com>
14262 * config/aarch64/aarch64.md (*mov<SHORT:mode>_aarch64): Extend
14263 w<-w, r<-w and w<-r alternatives to !simd, using 32-bit moves
14264 in that case. Extend w<-r to w<-Z.
14265 (*mov<HFBF:mode>_aarch64): Likewise, but with Y instead of Z.
14266 (*movti_aarch64): Use an FMOV from XZR for w<-Z if MOVI is not
14268 (define_split): Do not apply the floating-point immediate-to-register
14269 split to zeros, even if MOVI is not available.
14271 2022-09-07 Richard Sandiford <richard.sandiford@arm.com>
14273 * config/aarch64/aarch64.cc (aarch64_conditional_register_usage):
14274 Disallow use of FPRs in register asms for !TARGET_FLOAT.
14276 2022-09-07 Lulu Cheng <chenglulu@loongson.cn>
14279 * config/loongarch/loongarch.cc (loongarch_asan_shadow_offset): New.
14280 (TARGET_ASAN_SHADOW_OFFSET): New.
14282 2022-09-07 Jakub Jelinek <jakub@redhat.com>
14284 * doc/invoke.texi (-Wno-unicode): Document.
14286 2022-09-07 Jiufu Guo <guojiufu@linux.ibm.com>
14288 * config/rs6000/rs6000.md (const_scalar_int splitter): Remove.
14290 2022-09-07 Jiufu Guo <guojiufu@linux.ibm.com>
14292 * config/rs6000/rs6000.md: (constant splitters): Use "(pc)" as the
14295 2022-09-07 liuhongt <hongtao.liu@intel.com>
14297 PR tree-optimization/103144
14298 * tree-vect-loop.cc (vect_is_nonlinear_iv_evolution): New function.
14299 (vect_analyze_scalar_cycles_1): Detect nonlinear iv by upper function.
14300 (vect_create_nonlinear_iv_init): New function.
14301 (vect_peel_nonlinear_iv_init): Ditto.
14302 (vect_create_nonlinear_iv_step): Ditto
14303 (vect_create_nonlinear_iv_vec_step): Ditto
14304 (vect_update_nonlinear_iv): Ditto
14305 (vectorizable_nonlinear_induction): Ditto.
14306 (vectorizable_induction): Call
14307 vectorizable_nonlinear_induction when induction_type is not
14309 * tree-vect-loop-manip.cc (vect_update_ivs_after_vectorizer):
14310 Update nonlinear iv for epilogue loop.
14311 * tree-vectorizer.h (enum vect_induction_op_type): New enum.
14312 (STMT_VINFO_LOOP_PHI_EVOLUTION_TYPE): New Macro.
14314 2022-09-06 Richard Biener <rguenther@suse.de>
14316 PR tree-optimization/106754
14317 * gimple-predicate-analysis.cc (compute_control_dep_chain_pdom):
14318 New function, split out from compute_control_dep_chain. Handle
14319 loop-exit like conditions here by pushing to the control vector.
14320 (compute_control_dep_chain): Adjust and streamline dumping.
14321 In the wrapper perform a post-dominator walk as well.
14322 (uninit_analysis::init_use_preds): Remove premature early exit.
14324 2022-09-06 Max Filippov <jcmvbkbc@gmail.com>
14326 * config/xtensa/linux.h (LINK_SPEC): Add static-pie.
14328 2022-09-06 Aldy Hernandez <aldyh@redhat.com>
14330 * range-op-float.cc (build_le): Handle NANs and going past infinity.
14334 (foperator_lt::op1_range): Avoid adjustments to range if build_*
14336 (foperator_lt::op2_range): Same.
14337 (foperator_le::op1_range): Same.
14338 (foperator_le::op2_range): Same.
14339 (foperator_gt::op1_range): Same.
14340 (foperator_gt::op2_range): Same.
14342 2022-09-06 Richard Biener <rguenther@suse.de>
14344 * gimple-predicate-analysis.cc (compute_control_dep_chain):
14345 Add output flag to indicate whether we possibly have dropped
14346 any chains. Return whether the info is complete from the
14348 (uninit_analysis::init_use_preds): Adjust accordingly, with
14349 a workaround for PR106754.
14350 (uninit_analysis::init_from_phi_def): Properly guard the
14351 case where we complete an empty chain.
14353 2022-09-06 Jan-Benedict Glaw <jbglaw@lug-owl.de>
14355 * config/msp430/msp430.cc (msp430_single_op_cost): Document unused argument.
14357 2022-09-06 Prathamesh Kulkarni <prathamesh.kulkarni@linaro.org>
14359 * tree-ssa-forwprop.cc (simplify_permutation): Set res_type to a vector
14360 type with same element type as arg0, and length as op2.
14362 2022-09-06 Richard Biener <rguenther@suse.de>
14364 PR tree-optimization/106844
14365 * gimple-predicate-analysis.cc (compute_control_dep_chain):
14366 Return whether we found a chain.
14368 2022-09-06 Richard Biener <rguenther@suse.de>
14370 PR tree-optimization/106841
14371 * tree-vect-slp.cc (vect_detect_hybrid_slp): Also process
14372 scatter/gather offset.
14374 2022-09-06 Jan-Benedict Glaw <jbglaw@lug-owl.de>
14376 * config/bpf/bpf.h (REGNO_REG_CLASS): Reference arguments as (void).
14378 2022-09-06 Jakub Jelinek <jakub@redhat.com>
14380 * gimple.h (enum gf_mask): Add GF_OMP_ORDERED_STANDALONE enumerator.
14381 (gimple_omp_subcode): Use GIMPLE_OMP_ORDERED instead of
14382 GIMPLE_OMP_TEAMS as upper bound.
14383 (gimple_omp_ordered_standalone_p, gimple_omp_ordered_standalone): New
14385 * gimplify.cc (find_standalone_omp_ordered): Look for OMP_ORDERED with
14386 NULL OMP_ORDERED_BODY rather than with OMP_DOACROSS clause.
14387 (gimplify_expr): Call gimple_omp_ordered_standalone for OMP_ORDERED
14388 with NULL OMP_ORDERED_BODY.
14389 * omp-low.cc (check_omp_nesting_restrictions): Use
14390 gimple_omp_ordered_standalone_p test instead of
14391 omp_find_clause (..., OMP_CLAUSE_DOACROSS).
14392 (lower_omp_ordered): Likewise.
14393 * omp-expand.cc (expand_omp, build_omp_regions_1,
14394 omp_make_gimple_edges): Likewise.
14396 2022-09-06 Xianmiao Qu <cooper.qu@linux.alibaba.com>
14398 * config/csky/csky.md (cskyv2_addcc): Fix missing operand.
14399 (cskyv2_addcc_invert): Likewise.
14401 2022-09-06 Jose E. Marchesi <jose.marchesi@oracle.com>
14403 * config/bpf/bpf.cc (bpf_expand_prologue): Remove unused automatic
14405 (bpf_expand_epilogue): Likewise.
14407 2022-09-06 liuhongt <hongtao.liu@intel.com>
14409 * config/i386/i386-builtin.def (IX86_BUILTIN_CVTPS2PH512):
14410 Map to CODE_FOR_avx512f_vcvtps2ph512_mask_sae.
14411 * config/i386/sse.md (<mask_codefor>avx512f_vcvtps2ph512<mask_name>): Extend to ..
14412 (<mask_codefor>avx512f_vcvtps2ph512<mask_name><round_saeonly_name>): .. this.
14413 (avx512f_vcvtps2ph512_mask_sae): New expander
14415 2022-09-05 Aldy Hernandez <aldyh@redhat.com>
14417 * gimple-range-fold.cc
14418 (fold_using_range::range_of_builtin_int_call): Ignore sign bit
14419 when there's the possibility of a NAN.
14421 2022-09-05 Aldy Hernandez <aldyh@redhat.com>
14423 * gimple-range-fold.cc
14424 (fold_using_range::range_of_builtin_int_call): Fold a set signbit
14425 in __builtin_signbit to nonzero.
14427 2022-09-05 Alexander Monakov <amonakov@ispras.ru>
14430 * config/i386/i386.md (sse4_2_crc32di): Model that only low 32
14431 bits of operand 0 are consumed, and the result is zero-extended
14434 2022-09-05 Aldy Hernandez <aldyh@redhat.com>
14436 PR middle-end/106831
14437 * value-range.h (frange::supports_p): Disable decimal floats.
14438 * range-op-float.cc (frange_drop_inf): Remove DECIMAL_FLOAT_MODE_P
14440 (frange_drop_ninf): Same.
14442 2022-09-05 Aldy Hernandez <aldyh@redhat.com>
14444 PR middle-end/106824
14445 * value-range.cc (frange::set_nan): Set undefined when updating a
14448 2022-09-05 Kito Cheng <kito.cheng@sifive.com>
14450 * config/riscv/riscv.cc (riscv_option_override): Fix wrong
14451 condition for MASK_DIV and simplify incompatible checking.
14452 * config/riscv/riscv.md (muldi3): Adding parentheses.
14454 2022-09-05 Kyrylo Tkachov <kyrylo.tkachov@arm.com>
14456 * config/aarch64/aarch64.cc (aarch64_validate_march): Check if invalid arch
14457 string is a valid -mcpu string and emit hint.
14459 2022-09-05 LiaoShihua <shihua@iscas.ac.cn>
14461 * common/config/riscv/riscv-common.cc: Add Zmmul.
14462 * config/riscv/riscv-opts.h (MASK_ZMMUL): New.
14463 (TARGET_ZMMUL): Ditto.
14464 * config/riscv/riscv.cc (riscv_option_override):Ditto.
14465 * config/riscv/riscv.md: Add Zmmul
14466 * config/riscv/riscv.opt: Ditto.
14468 2022-09-05 Richard Biener <rguenther@suse.de>
14470 * tree-cfg.h (get_cases_for_edge): Declare.
14471 * tree-cfg.cc (get_cases_for_edge): Export.
14472 * tree-ssa-uninit.cc (execute_late_warn_uninitialized):
14473 Start and end recording case labels.
14474 * gimple-predicate-analysis.cc (MAX_SWITCH_CASES): Remove.
14475 (predicate::init_from_control_deps): Use get_cases_for_edge.
14477 2022-09-05 Richard Biener <rguenther@suse.de>
14479 * gimple-predicate-analysis.cc (MAX_POSTDOM_CHECK): Remove.
14480 (compute_control_dep_chain): Move uninit-control-dep-attempts
14481 checking where it also counts the post-dominator check
14484 2022-09-05 Richard Biener <rguenther@suse.de>
14486 * gimple-predicate-analysis.h (predicate::debug): New.
14487 (predicate::dump): Add FILE * argument, add base overload.
14488 * gimple-predicate-analysis.cc (debug): New.
14489 (dump_pred_info): Add FILE * argument.
14490 (dump_pred_chain): Likewise.
14491 (predicate::dump): Split out preamble into overload. Add
14493 (predicate::debug): New.
14494 (predicate::simplify): Adjust.
14495 (predicate::normalize): Likewise.
14496 (predicate::init_from_control_deps): Likewise.
14498 2022-09-05 Richard Sandiford <richard.sandiford@arm.com>
14500 * config/aarch64/aarch64-protos.h
14501 (aarch64_init_simd_builtins): Remove prototype.
14502 * config/aarch64/aarch64-builtins.cc
14503 (aarch64_simd_builtins_initialized_p): Delete.
14504 (aarch64_init_simd_builtins): Make static. Remove protection
14505 against multiple calls.
14506 * config/aarch64/aarch64-c.cc (aarch64_pragma_target_parse): Remove
14507 lazy SIMD builtin initialization.
14508 * config/aarch64/aarch64.cc
14509 (aarch64_option_valid_attribute_p): Likewise.
14511 2022-09-05 Xi Ruoyao <xry111@xry111.site>
14513 * config/loongarch/genopts/loongarch.opt.in: Add
14514 -mdirect-extern-access option.
14515 * config/loongarch/loongarch.opt: Regenerate.
14516 * config/loongarch/loongarch.cc
14517 (loongarch_symbol_binds_local_p): Return true if
14518 TARGET_DIRECT_EXTERN_ACCESS.
14519 (loongarch_option_override_internal): Complain if
14520 -mdirect-extern-access is used with -fPIC or -fpic.
14521 * doc/invoke.texi: Document -mdirect-extern-access for
14524 2022-09-05 Martin Liska <mliska@suse.cz>
14526 * doc/tm.texi.in: Rename DEBUGGER_REGISTER_NUMBER to
14528 * doc/tm.texi: Regenerate.
14530 2022-09-05 Martin Liska <mliska@suse.cz>
14532 * config/aarch64/aarch64-protos.h (aarch64_dbx_regno):
14533 Rename DBX_REGISTER_NUMBER to DEBUGGER_REGNO.
14534 (aarch64_debugger_regno): Likewise.
14535 * config/aarch64/aarch64.cc (aarch64_dbx_regno): Likewise.
14536 (aarch64_debugger_regno): Likewise.
14537 * config/aarch64/aarch64.h (DBX_REGISTER_NUMBER): Likewise.
14538 (DEBUGGER_REGNO): Likewise.
14539 (DWARF_FRAME_REGNUM): Likewise.
14540 * config/alpha/alpha.h (DWARF_FRAME_REGNUM): Likewise.
14541 * config/arc/arc.cc (arc_init_reg_tables): Likewise.
14542 * config/arc/arc.h (DBX_REGISTER_NUMBER): Likewise.
14543 (DEBUGGER_REGNO): Likewise.
14544 * config/arm/arm-protos.h (arm_dbx_regno): Likewise.
14545 (arm_debugger_regno): Likewise.
14546 * config/arm/arm.cc (arm_dbx_regno): Likewise.
14547 (arm_debugger_regno): Likewise.
14548 * config/arm/arm.h (DBX_REGISTER_NUMBER): Likewise.
14549 (DEBUGGER_REGNO): Likewise.
14550 * config/bfin/bfin.h (DBX_REGISTER_NUMBER): Likewise.
14551 (DEBUGGER_REGNO): Likewise.
14552 * config/c6x/c6x.cc: Likewise.
14553 * config/c6x/c6x.h (DBX_REGISTER_NUMBER): Likewise.
14554 (DEBUGGER_REGNO): Likewise.
14555 * config/cris/cris.h (enum reg_class): Likewise.
14556 (DBX_REGISTER_NUMBER): Likewise.
14557 (DEBUGGER_REGNO): Likewise.
14558 * config/csky/csky.cc (enum reg_class): Likewise.
14559 * config/csky/csky.h (DWARF_FRAME_REGNUM): Likewise.
14560 (DBX_REGISTER_NUMBER): Likewise.
14561 (DEBUGGER_REGNO): Likewise.
14562 * config/frv/frv.h (DBX_REGISTER_NUMBER): Likewise.
14563 (DEBUGGER_REGNO): Likewise.
14564 * config/gcn/gcn-hsa.h (DBX_REGISTER_NUMBER): Likewise.
14565 (DEBUGGER_REGNO): Likewise.
14566 * config/gcn/gcn.cc (print_operand): Likewise.
14567 * config/i386/bsd.h (ASM_QUAD): Likewise.
14568 * config/i386/cygming.h (DBX_REGISTER_NUMBER): Likewise.
14569 (DEBUGGER_REGNO): Likewise.
14570 (DWARF_FRAME_REGNUM): Likewise.
14571 * config/i386/darwin.h (DBX_REGISTER_NUMBER): Likewise.
14572 (DEBUGGER_REGNO): Likewise.
14573 * config/i386/djgpp.h (DBX_REGISTER_NUMBER): Likewise.
14574 (DEBUGGER_REGNO): Likewise.
14575 * config/i386/dragonfly.h (DBX_REGISTER_NUMBER): Likewise.
14576 (DEBUGGER_REGNO): Likewise.
14577 * config/i386/freebsd.h (DBX_REGISTER_NUMBER): Likewise.
14578 (DEBUGGER_REGNO): Likewise.
14579 * config/i386/gas.h: Likewise.
14580 * config/i386/gnu-user.h (DBX_REGISTER_NUMBER): Likewise.
14581 (DEBUGGER_REGNO): Likewise.
14582 * config/i386/i386.cc (enum reg_class): Likewise.
14583 * config/i386/i386.h (DBX_REGISTER_NUMBER): Likewise.
14584 (DEBUGGER_REGNO): Likewise.
14585 * config/i386/i386elf.h (DBX_REGISTER_NUMBER): Likewise.
14586 (DEBUGGER_REGNO): Likewise.
14587 * config/i386/iamcu.h (DBX_REGISTER_NUMBER): Likewise.
14588 (DEBUGGER_REGNO): Likewise.
14589 * config/i386/lynx.h (DBX_REGISTER_NUMBER): Likewise.
14590 (DEBUGGER_REGNO): Likewise.
14591 * config/i386/netbsd-elf.h (DBX_REGISTER_NUMBER): Likewise.
14592 (DEBUGGER_REGNO): Likewise.
14593 * config/i386/nto.h (DBX_REGISTER_NUMBER): Likewise.
14594 (DEBUGGER_REGNO): Likewise.
14595 * config/i386/openbsdelf.h (DBX_REGISTER_NUMBER): Likewise.
14596 (DEBUGGER_REGNO): Likewise.
14597 * config/i386/sysv4.h (DBX_REGISTER_NUMBER): Likewise.
14598 (DEBUGGER_REGNO): Likewise.
14599 * config/i386/vxworks.h (DBX_REGISTER_NUMBER): Likewise.
14600 (DEBUGGER_REGNO): Likewise.
14601 * config/i386/x86-64.h (DBX_REGISTER_NUMBER): Likewise.
14602 (DEBUGGER_REGNO): Likewise.
14603 * config/ia64/ia64-protos.h (ia64_dbx_regno): Likewise.
14604 (ia64_debugger_regno): Likewise.
14605 * config/ia64/ia64.cc (ia64_output_function_prologue): Likewise.
14606 (ia64_dbx_regno): Likewise.
14607 (ia64_debugger_regno): Likewise.
14608 (process_cfa_adjust_cfa): Likewise.
14609 (process_cfa_register): Likewise.
14610 (ia64_asm_unwind_emit): Likewise.
14611 * config/ia64/ia64.h: Likewise.
14612 * config/ia64/sysv4.h (DBX_REGISTER_NUMBER): Likewise.
14613 (DEBUGGER_REGNO): Likewise.
14614 * config/lm32/lm32.h (DBX_REGISTER_NUMBER): Likewise.
14615 (DEBUGGER_REGNO): Likewise.
14616 * config/m32c/m32c.cc (m32c_eh_return_stackadj_rtx): Likewise.
14617 * config/m32c/m32c.h (DBX_REGISTER_NUMBER): Likewise.
14618 (DEBUGGER_REGNO): Likewise.
14619 * config/m68k/linux.h (DBX_REGISTER_NUMBER): Likewise.
14620 (DEBUGGER_REGNO): Likewise.
14621 * config/m68k/m68k.h (DBX_REGISTER_NUMBER): Likewise.
14622 (DEBUGGER_REGNO): Likewise.
14623 (__transfer_from_trampoline): Likewise.
14624 * config/m68k/m68kelf.h (DBX_REGISTER_NUMBER): Likewise.
14625 (DEBUGGER_REGNO): Likewise.
14626 * config/m68k/netbsd-elf.h (M68K_STATIC_CHAIN_REG_NAME): Likewise.
14627 (DBX_REGISTER_NUMBER): Likewise.
14628 (DEBUGGER_REGNO): Likewise.
14629 * config/microblaze/microblaze.cc (microblaze_option_override): Likewise.
14630 * config/microblaze/microblaze.h (DBX_REGISTER_NUMBER): Likewise.
14631 (DEBUGGER_REGNO): Likewise.
14632 (GP_DBX_FIRST): Likewise.
14633 (GP_DEBUGGER_FIRST): Likewise.
14634 * config/mips/vxworks.h (DBX_REGISTER_NUMBER): Likewise.
14635 (DEBUGGER_REGNO): Likewise.
14636 * config/mmix/mmix-protos.h (mmix_dbx_regno): Likewise.
14637 (mmix_debugger_regno): Likewise.
14638 * config/mmix/mmix.cc (mmix_dbx_regno): Likewise.
14639 (mmix_debugger_regno): Likewise.
14640 * config/mmix/mmix.h (DBX_REGISTER_NUMBER): Likewise.
14641 (DEBUGGER_REGNO): Likewise.
14642 * config/nds32/nds32-protos.h (nds32_dbx_regno): Likewise.
14643 (nds32_debugger_regno): Likewise.
14644 * config/nds32/nds32.cc (nds32_dbx_regno): Likewise.
14645 (nds32_debugger_regno): Likewise.
14646 (nds32_use_blocks_for_constant_p): Likewise.
14647 * config/nds32/nds32.h (DBX_REGISTER_NUMBER): Likewise.
14648 (DEBUGGER_REGNO): Likewise.
14649 * config/nvptx/nvptx.h (DBX_REGISTER_NUMBER): Likewise.
14650 (DEBUGGER_REGNO): Likewise.
14651 * config/or1k/or1k.h (DBX_REGISTER_NUMBER): Likewise.
14652 (DEBUGGER_REGNO): Likewise.
14653 * config/pa/pa32-regs.h (DBX_REGISTER_NUMBER): Likewise.
14654 (DEBUGGER_REGNO): Likewise.
14655 * config/pa/pa64-regs.h (DBX_REGISTER_NUMBER): Likewise.
14656 (DEBUGGER_REGNO): Likewise.
14657 * config/rs6000/rs6000-protos.h (rs6000_dbx_regno): Likewise.
14658 (rs6000_debugger_regno): Likewise.
14659 * config/rs6000/rs6000.cc (rs6000_dbx_regno): Likewise.
14660 (rs6000_debugger_regno): Likewise.
14661 * config/rs6000/rs6000.h (DBX_REGISTER_NUMBER): Likewise.
14662 (DEBUGGER_REGNO): Likewise.
14663 (DWARF2_FRAME_REG_OUT): Likewise.
14664 * config/s390/s390.h (DBX_REGISTER_NUMBER): Likewise.
14665 (DEBUGGER_REGNO): Likewise.
14666 * config/sh/elf.h (DBX_REGISTER_NUMBER): Likewise.
14667 (DEBUGGER_REGNO): Likewise.
14668 * config/sh/linux.h (DBX_REGISTER_NUMBER): Likewise.
14669 (DEBUGGER_REGNO): Likewise.
14670 * config/sh/sh.h (DBX_REGISTER_NUMBER): Likewise.
14671 (DEBUGGER_REGNO): Likewise.
14672 (SH_DBX_REGISTER_NUMBER): Likewise.
14673 (SH_DEBUGGER_REGNO): Likewise.
14674 * config/visium/visium.h (DBX_REGISTER_NUMBER): Likewise.
14675 (DEBUGGER_REGNO): Likewise.
14676 * config/xtensa/elf.h (DBX_REGISTER_NUMBER): Likewise.
14677 (DEBUGGER_REGNO): Likewise.
14678 * config/xtensa/linux.h (DBX_REGISTER_NUMBER): Likewise.
14679 (DEBUGGER_REGNO): Likewise.
14680 * config/xtensa/uclinux.h (DBX_REGISTER_NUMBER): Likewise.
14681 (DEBUGGER_REGNO): Likewise.
14682 * config/xtensa/xtensa-protos.h (xtensa_dbx_regno): Likewise.
14683 (xtensa_debugger_regno): Likewise.
14684 * config/xtensa/xtensa.cc (xtensa_dbx_regno): Likewise.
14685 (xtensa_debugger_regno): Likewise.
14686 * config/xtensa/xtensa.h (DBX_REGISTER_NUMBER): Likewise.
14687 (DEBUGGER_REGNO): Likewise.
14688 * defaults.h (DBX_REGISTER_NUMBER): Likewise.
14689 (DEBUGGER_REGNO): Likewise.
14690 (DWARF_FRAME_REGNUM): Likewise.
14691 * doc/tm.texi: Likewise.
14692 * doc/tm.texi.in: Likewise.
14693 * dwarf2out.cc (dbx_reg_number): Likewise.
14694 (debugger_reg_number): Likewise.
14695 (reg_loc_descriptor): Likewise.
14696 (multiple_reg_loc_descriptor): Likewise.
14697 (mem_loc_descriptor): Likewise.
14698 * except.cc: Likewise.
14700 2022-09-05 konglin1 <lingling.kong@intel.com>
14703 * config/i386/i386-expand.cc (ix86_expand_vector_init_duplicate):
14705 (expand_vec_perm_broadcast_1): Ditto.
14706 * config/i386/sse.md (avx512fmaskhalfmode): Add BF vector mode.
14707 (vec_set<mode>_0): Add @ to it.
14708 (@vec_set<mode>_0): Ditto.
14709 (vec_interleave_high<mode><mask_name>): Ditto.
14710 (@vec_interleave_high<mode><mask_name>): Ditto.
14711 (vec_interleave_low<mode><mask_name>): Ditto.
14712 (@vec_interleave_low<mode><mask_name>): Ditto.
14713 * config/i386/subst.md (SUBST_V): Add BF vector mode.
14715 2022-09-05 Aldy Hernandez <aldyh@redhat.com>
14717 * value-range.cc (early_nan_resolve): Remove.
14718 (frange::intersect): Handle NANs.
14720 2022-09-04 Aldy Hernandez <aldyh@redhat.com>
14722 * value-range.cc (frange::union_): Do not drop properties when
14723 unioning a NAN with something else.
14724 (range_tests_signed_zeros): Add tests.
14726 2022-09-04 Aldy Hernandez <aldyh@redhat.com>
14728 * real.cc (encode_ieee_single): Use real_isdenormal.
14729 (encode_ieee_double): Same.
14730 (encode_ieee_extended): Same.
14731 (encode_ieee_quad): Same.
14732 (encode_ieee_half): Same.
14733 (encode_arm_bfloat_half): Same.
14734 * real.h (real_isdenormal): New.
14736 2022-09-03 Aldy Hernandez <aldyh@redhat.com>
14738 * value-range.cc (frange::singleton_p): Move NAN check to the top.
14740 2022-09-03 Jan-Benedict Glaw <jbglaw@lug-owl.de>
14742 * config/nvptx/nvptx.h (ASM_OUTPUT_DEF): Reference macro arguments.
14744 2022-09-03 Jakub Jelinek <jakub@redhat.com>
14746 * tree-core.h (enum omp_clause_code): Add OMP_CLAUSE_DOACROSS.
14747 (enum omp_clause_depend_kind): Remove OMP_CLAUSE_DEPEND_SOURCE
14748 and OMP_CLAUSE_DEPEND_SINK, add OMP_CLAUSE_DEPEND_INVALID.
14749 (enum omp_clause_doacross_kind): New type.
14750 (struct tree_omp_clause): Add subcode.doacross_kind member.
14751 * tree.h (OMP_CLAUSE_DEPEND_SINK_NEGATIVE): Remove.
14752 (OMP_CLAUSE_DOACROSS_KIND): Define.
14753 (OMP_CLAUSE_DOACROSS_SINK_NEGATIVE): Define.
14754 (OMP_CLAUSE_DOACROSS_DEPEND): Define.
14755 (OMP_CLAUSE_ORDERED_DOACROSS): Define.
14756 * tree.cc (omp_clause_num_ops, omp_clause_code_name): Add
14757 OMP_CLAUSE_DOACROSS entries.
14758 * tree-nested.cc (convert_nonlocal_omp_clauses,
14759 convert_local_omp_clauses): Handle OMP_CLAUSE_DOACROSS.
14760 * tree-pretty-print.cc (dump_omp_clause): Don't handle
14761 OMP_CLAUSE_DEPEND_SOURCE and OMP_CLAUSE_DEPEND_SINK. Handle
14762 OMP_CLAUSE_DOACROSS.
14763 * gimplify.cc (gimplify_omp_depend): Don't handle
14764 OMP_CLAUSE_DEPEND_SOURCE and OMP_CLAUSE_DEPEND_SINK.
14765 (gimplify_scan_omp_clauses): Likewise. Handle OMP_CLAUSE_DOACROSS.
14766 (gimplify_adjust_omp_clauses): Handle OMP_CLAUSE_DOACROSS.
14767 (find_standalone_omp_ordered): New function.
14768 (gimplify_omp_for): When OMP_CLAUSE_ORDERED is present, search
14769 body for OMP_ORDERED with OMP_CLAUSE_DOACROSS and if found,
14770 set OMP_CLAUSE_ORDERED_DOACROSS.
14771 (gimplify_omp_ordered): Don't handle OMP_CLAUSE_DEPEND_SINK or
14772 OMP_CLAUSE_DEPEND_SOURCE, instead check OMP_CLAUSE_DOACROSS, adjust
14773 diagnostics that presence or absence of ordered clause parameter
14774 is irrelevant. Handle doacross(sink:omp_cur_iteration-1). Use
14775 actual user name of the clause - doacross or depend - in diagnostics.
14776 * omp-general.cc (omp_extract_for_data): Don't set fd->ordered
14777 if !OMP_CLAUSE_ORDERED_DOACROSS (t). If
14778 OMP_CLAUSE_ORDERED_DOACROSS (t) but !OMP_CLAUSE_ORDERED_EXPR (t),
14779 set fd->ordered to -1 and set it after the loop in that case to
14781 * omp-low.cc (check_omp_nesting_restrictions): Don't handle
14782 OMP_CLAUSE_DEPEND_SOURCE nor OMP_CLAUSE_DEPEND_SINK, instead check
14783 OMP_CLAUSE_DOACROSS. Use actual user name of the clause - doacross
14784 or depend - in diagnostics. Diagnose mixing of stand-alone and
14785 block associated ordered constructs binding to the same loop.
14786 (lower_omp_ordered_clauses): Don't handle OMP_CLAUSE_DEPEND_SINK,
14787 instead handle OMP_CLAUSE_DOACROSS.
14788 (lower_omp_ordered): Look for OMP_CLAUSE_DOACROSS instead of
14790 (lower_depend_clauses): Don't handle OMP_CLAUSE_DEPEND_SOURCE and
14791 OMP_CLAUSE_DEPEND_SINK.
14792 * omp-expand.cc (expand_omp_ordered_sink): Emit a sorry for
14793 doacross(sink:omp_cur_iteration-1).
14794 (expand_omp_ordered_source_sink): Use
14795 OMP_CLAUSE_DOACROSS_SINK_NEGATIVE instead of
14796 OMP_CLAUSE_DEPEND_SINK_NEGATIVE. Use actual user name of the clause
14797 - doacross or depend - in diagnostics.
14798 (expand_omp): Look for OMP_CLAUSE_DOACROSS clause instead of
14800 (build_omp_regions_1): Likewise.
14801 (omp_make_gimple_edges): Likewise.
14802 * lto-streamer-out.cc (hash_tree): Handle OMP_CLAUSE_DOACROSS.
14803 * tree-streamer-in.cc (unpack_ts_omp_clause_value_fields): Likewise.
14804 * tree-streamer-out.cc (pack_ts_omp_clause_value_fields): Likewise.
14806 2022-09-02 David Malcolm <dmalcolm@redhat.com>
14809 * doc/invoke.texi (Warning Options): Add -Wxor-used-as-pow.
14811 2022-09-02 Iain Buclaw <ibuclaw@gdcproject.org>
14814 * config.gcc: Set tm_d_file to ${cpu_type}/${cpu_type}-d.h.
14815 * config/aarch64/aarch64-d.cc: Include tm_d.h.
14816 * config/aarch64/aarch64-protos.h (aarch64_d_target_versions): Move to
14817 config/aarch64/aarch64-d.h.
14818 (aarch64_d_register_target_info): Likewise.
14819 * config/aarch64/aarch64.h (TARGET_D_CPU_VERSIONS): Likewise.
14820 (TARGET_D_REGISTER_CPU_TARGET_INFO): Likewise.
14821 * config/arm/arm-d.cc: Include tm_d.h and arm-protos.h instead of
14823 * config/arm/arm-protos.h (arm_d_target_versions): Move to
14824 config/arm/arm-d.h.
14825 (arm_d_register_target_info): Likewise.
14826 * config/arm/arm.h (TARGET_D_CPU_VERSIONS): Likewise.
14827 (TARGET_D_REGISTER_CPU_TARGET_INFO): Likewise.
14828 * config/default-d.cc: Remove memmodel.h include.
14829 * config/freebsd-d.cc: Include tm_d.h instead of tm_p.h.
14830 * config/glibc-d.cc: Likewise.
14831 * config/i386/i386-d.cc: Include tm_d.h.
14832 * config/i386/i386-protos.h (ix86_d_target_versions): Move to
14833 config/i386/i386-d.h.
14834 (ix86_d_register_target_info): Likewise.
14835 (ix86_d_has_stdcall_convention): Likewise.
14836 * config/i386/i386.h (TARGET_D_CPU_VERSIONS): Likewise.
14837 (TARGET_D_REGISTER_CPU_TARGET_INFO): Likewise.
14838 (TARGET_D_HAS_STDCALL_CONVENTION): Likewise.
14839 * config/i386/winnt-d.cc: Include tm_d.h instead of tm_p.h.
14840 * config/mips/mips-d.cc: Include tm_d.h.
14841 * config/mips/mips-protos.h (mips_d_target_versions): Move to
14842 config/mips/mips-d.h.
14843 (mips_d_register_target_info): Likewise.
14844 * config/mips/mips.h (TARGET_D_CPU_VERSIONS): Likewise.
14845 (TARGET_D_REGISTER_CPU_TARGET_INFO): Likewise.
14846 * config/netbsd-d.cc: Include tm_d.h instead of tm.h and memmodel.h.
14847 * config/openbsd-d.cc: Likewise.
14848 * config/pa/pa-d.cc: Include tm_d.h.
14849 * config/pa/pa-protos.h (pa_d_target_versions): Move to
14851 (pa_d_register_target_info): Likewise.
14852 * config/pa/pa.h (TARGET_D_CPU_VERSIONS): Likewise.
14853 (TARGET_D_REGISTER_CPU_TARGET_INFO): Likewise.
14854 * config/riscv/riscv-d.cc: Include tm_d.h.
14855 * config/riscv/riscv-protos.h (riscv_d_target_versions): Move to
14856 config/riscv/riscv-d.h.
14857 (riscv_d_register_target_info): Likewise.
14858 * config/riscv/riscv.h (TARGET_D_CPU_VERSIONS): Likewise.
14859 (TARGET_D_REGISTER_CPU_TARGET_INFO): Likewise.
14860 * config/rs6000/rs6000-d.cc: Include tm_d.h.
14861 * config/rs6000/rs6000-protos.h (rs6000_d_target_versions): Move to
14862 config/rs6000/rs6000-d.h.
14863 (rs6000_d_register_target_info): Likewise.
14864 * config/rs6000/rs6000.h (TARGET_D_CPU_VERSIONS) Likewise.:
14865 (TARGET_D_REGISTER_CPU_TARGET_INFO) Likewise.:
14866 * config/s390/s390-d.cc: Include tm_d.h.
14867 * config/s390/s390-protos.h (s390_d_target_versions): Move to
14868 config/s390/s390-d.h.
14869 (s390_d_register_target_info): Likewise.
14870 * config/s390/s390.h (TARGET_D_CPU_VERSIONS): Likewise.
14871 (TARGET_D_REGISTER_CPU_TARGET_INFO): Likewise.
14872 * config/sol2-d.cc: Include tm_d.h instead of tm.h and memmodel.h.
14873 * config/sparc/sparc-d.cc: Include tm_d.h.
14874 * config/sparc/sparc-protos.h (sparc_d_target_versions): Move to
14875 config/sparc/sparc-d.h.
14876 (sparc_d_register_target_info): Likewise.
14877 * config/sparc/sparc.h (TARGET_D_CPU_VERSIONS): Likewise.
14878 (TARGET_D_REGISTER_CPU_TARGET_INFO): Likewise.
14879 * configure: Regenerate.
14880 * configure.ac (tm_d_file): Remove defaults.h.
14881 (tm_d_include_list): Remove options.h and insn-constants.h.
14882 * config/aarch64/aarch64-d.h: New file.
14883 * config/arm/arm-d.h: New file.
14884 * config/i386/i386-d.h: New file.
14885 * config/mips/mips-d.h: New file.
14886 * config/pa/pa-d.h: New file.
14887 * config/riscv/riscv-d.h: New file.
14888 * config/rs6000/rs6000-d.h: New file.
14889 * config/s390/s390-d.h: New file.
14890 * config/sparc/sparc-d.h: New file.
14892 2022-09-02 Aldy Hernandez <aldyh@redhat.com>
14894 * builtins.cc (fold_builtin_inf): Convert use of real_info to dconstinf.
14895 (fold_builtin_fpclassify): Same.
14896 * fold-const-call.cc (fold_const_call_cc): Same.
14898 * omp-low.cc (omp_reduction_init_op): Same.
14899 * realmpfr.cc (real_from_mpfr): Same.
14900 * tree.cc (build_complex_inf): Same.
14902 2022-09-02 Peter Bergner <bergner@linux.ibm.com>
14904 * config/rs6000/rs6000-builtin.cc (rs6000_gimple_fold_mma_builtin): Use
14905 NOP_EXPR for MMA pointer casting.
14907 2022-09-02 Richard Sandiford <richard.sandiford@arm.com>
14909 * tree-vect-slp.cc (vect_optimize_slp_pass::internal_node_cost):
14910 Reduce the fallback cost to 1. Only use it if the number of
14911 input lanes is equal to the number of output lanes.
14913 2022-09-02 Richard Sandiford <richard.sandiford@arm.com>
14915 PR tree-optimization/106787
14916 * tree-vect-slp.cc (vect_map_to_instance): New function, split out
14918 (vect_bb_partition_graph_r): ...here. Replace the visited set
14919 with a map from nodes to instances. Ensure that a node only
14920 appears in one partition.
14921 (vect_bb_partition_graph): Update accordingly.
14923 2022-09-02 Richard Biener <rguenther@suse.de>
14925 * timevar.def (TV_TREE_RPO_VN): New.
14926 * tree-ssa-sccvn.h (do_rpo_vn): Remove one overload.
14927 * tree-ssa-sccvn.cc (do_rpo_vn_1): Rename the worker.
14928 (do_rpo_vn): Unify the public API, track with TV_TREE_RPO_VN.
14929 (pass_fre::execute): Adjust.
14930 * tree-ssa-uninit.cc (execute_early_warn_uninitialized): Adjust.
14932 2022-09-02 Richard Biener <rguenther@suse.de>
14934 PR tree-optimization/106809
14935 * tree-ssa-sccvn.cc (dominaged_by_p_w_unex): Check we have
14936 more than one successor before doing extra work.
14938 2022-09-02 Kito Cheng <kito.cheng@sifive.com>
14940 * common/config/riscv/riscv-common.cc: Include <vector>.
14941 (struct riscv_multi_lib_info_t): New.
14942 (riscv_subset_list::match_score): Ditto.
14943 (find_last_appear_switch): Ditto.
14944 (prefixed_with): Ditto.
14945 (struct multi_lib_info_t): Ditto.
14946 (riscv_current_arch_str): Ditto.
14947 (riscv_current_abi_str): Ditto.
14948 (riscv_multi_lib_info_t::parse): Ditto.
14949 (riscv_check_cond): Ditto.
14950 (riscv_check_conds): Ditto.
14951 (riscv_compute_multilib): Ditto.
14952 (TARGET_COMPUTE_MULTILIB): Defined.
14953 * config/riscv/elf.h (LIB_SPEC): Call riscv_multi_lib_check if
14955 (RISCV_USE_CUSTOMISED_MULTI_LIB): New.
14956 * config/riscv/riscv.h (riscv_multi_lib_check): New.
14957 (EXTRA_SPEC_FUNCTIONS): Add riscv_multi_lib_check.
14958 * config/riscv/riscv-subset.h (riscv_subset_list::match_score): New.
14960 2022-09-02 Kito Cheng <kito.cheng@sifive.com>
14962 * common/common-target.def (compute_multilib): New.
14963 * common/common-targhooks.h (default_compute_multilib): New.
14964 * common/common-targhooks.cc (default_compute_multilib): New.
14965 * doc/tm.texi.in (TARGET_COMPUTE_MULTILIB): New.
14966 * doc/tm.texi: Regen.
14967 * gcc.cc: Include common/common-target.h.
14968 (set_multilib_dir) Call targetm_common.compute_multilib.
14969 (SWITCH_LIVE): Move to opts.h.
14970 (SWITCH_FALSE): Ditto.
14971 (SWITCH_IGNORE): Ditto.
14972 (SWITCH_IGNORE_PERMANENTLY): Ditto.
14973 (SWITCH_KEEP_FOR_GCC): Ditto.
14974 (struct switchstr): Ditto.
14975 * opts.h (SWITCH_LIVE): Move from gcc.c.
14976 (SWITCH_FALSE): Ditto.
14977 (SWITCH_IGNORE): Ditto.
14978 (SWITCH_IGNORE_PERMANENTLY): Ditto.
14979 (SWITCH_KEEP_FOR_GCC): Ditto.
14980 (struct switchstr): Ditto.
14982 2022-09-02 Martin Liska <mliska@suse.cz>
14984 * config/pdp11/pdp11.h (PREFERRED_DEBUGGING_TYPE): Disable
14987 2022-09-02 Martin Liska <mliska@suse.cz>
14989 * Makefile.in: Remove -gstabs option support, DBX-related
14990 macros and DBX debugging info support.
14991 * collect2.cc (scan_prog_file): Likewise.
14992 * common.opt: Likewise.
14993 * config.gcc: Likewise.
14994 * config.in: Likewise.
14995 * config/aarch64/aarch64-elf.h (DBX_DEBUGGING_INFO): Likewise.
14996 * config/alpha/alpha.cc: Likewise.
14997 * config/alpha/elf.h (ASM_SPEC): Likewise.
14998 * config/arc/arc.h (DBX_DEBUGGING_INFO): Likewise.
14999 (DBX_CONTIN_LENGTH): Likewise.
15000 * config/arm/aout.h (DBX_DEBUGGING_INFO): Likewise.
15001 (DBX_CONTIN_LENGTH): Likewise.
15002 * config/arm/netbsd-elf.h (DBX_CONTIN_LENGTH): Likewise.
15003 * config/darwin.h (DSYMUTIL_SPEC): Likewise.
15004 (ASM_DEBUG_SPEC): Likewise.
15005 (DBX_DEBUGGING_INFO): Likewise.
15006 (DBX_USE_BINCL): Likewise.
15007 (DBX_CONTIN_LENGTH): Likewise.
15008 (DBX_OUTPUT_NULL_N_SO_AT_MAIN_SOURCE_FILE_END): Likewise.
15009 * config/epiphany/epiphany.h (DBX_DEBUGGING_INFO): Likewise.
15010 (DBX_CONTIN_LENGTH): Likewise.
15011 * config/i386/bsd.h (DBX_NO_XREFS): Likewise.
15012 (DBX_CONTIN_LENGTH): Likewise.
15013 * config/i386/gas.h (DBX_NO_XREFS): Likewise.
15014 (DBX_CONTIN_LENGTH): Likewise.
15015 * config/ia64/ia64.h: Likewise.
15016 * config/ia64/sysv4.h (DBX_DEBUGGING_INFO): Likewise.
15017 * config/m68k/linux.h (DBX_CONTIN_LENGTH): Likewise.
15018 * config/m68k/openbsd.h (DBX_DEBUGGING_INFO): Likewise.
15019 (DBX_CONTIN_LENGTH): Likewise.
15020 (DBX_CONTIN_CHAR): Likewise.
15021 * config/mips/mips.cc (mips_output_filename): Likewise.
15022 (mips_option_override): Likewise.
15023 * config/mips/mips.h (SUBTARGET_ASM_DEBUGGING_SPEC): Likewise.
15024 (DBX_DEBUGGING_INFO): Likewise.
15025 (DBX_CONTIN_LENGTH): Likewise.
15026 (DBX_REGISTER_NUMBER): Likewise.
15027 (GP_DBX_FIRST): Likewise.
15028 (FP_DBX_FIRST): Likewise.
15029 (MD_DBX_FIRST): Likewise.
15030 * config/nvptx/nvptx.cc: Likewise.
15031 * config/openbsd.h (DBX_NO_XREFS): Likewise.
15032 * config/pa/pa-64.h (DBX_DEBUGGING_INFO): Likewise.
15033 * config/pa/pa.h (ASSEMBLER_DIALECT): Likewise.
15034 (DBX_CONTIN_LENGTH): Likewise.
15035 * config/pa/som.h (PREFERRED_DEBUGGING_TYPE): Likewise.
15036 (DBX_USE_BINCL): Likewise.
15037 (DBX_LINES_FUNCTION_RELATIVE): Likewise.
15038 (DBX_OUTPUT_NULL_N_SO_AT_MAIN_SOURCE_FILE_END): Likewise.
15039 * config/pdp11/pdp11.cc: Likewise.
15040 * config/pdp11/pdp11.h (DBX_DEBUGGING_INFO): Likewise.
15041 (PREFERRED_DEBUGGING_TYPE): Likewise.
15042 (DBX_CONTIN_LENGTH): Likewise.
15043 * config/rs6000/rs6000-builtin.cc: Likewise.
15044 * config/rs6000/rs6000-call.cc: Likewise.
15045 * config/rs6000/rs6000-logue.cc (defined): Likewise.
15046 * config/rs6000/rs6000.cc (rs6000_option_override_internal): Likewise.
15047 (HAVE_XCOFF_DWARF_EXTRAS): Likewise.
15048 (rs6000_xcoff_declare_function_name): Likewise.
15049 * config/rs6000/sysv4.h (DBX_DEBUGGING_INFO): Likewise.
15050 (DBX_FUNCTION_FIRST): Likewise.
15051 * config/rs6000/xcoff.h (XCOFF_DEBUGGING_INFO): Likewise.
15052 * config/rx/rx.h (DBX_DEBUGGING_INFO): Likewise.
15053 * config/sh/elf.h (DBX_LINES_FUNCTION_RELATIVE): Likewise.
15054 (DBX_OUTPUT_NULL_N_SO_AT_MAIN_SOURCE_FILE_END): Likewise.
15055 * config/sol2.h (NO_DBX_BNSYM_ENSYM): Likewise.
15056 * config/sparc/freebsd.h (DBX_CONTIN_CHAR): Likewise.
15057 * config/sparc/netbsd-elf.h (DBX_CONTIN_CHAR): Likewise.
15058 * config/sparc/sparc.h (DBX_CONTIN_LENGTH): Likewise.
15059 * config/vax/vax.cc (vax_file_start): Likewise.
15060 * config/vax/vax.h (DBX_DEBUGGING_INFO): Likewise.
15061 (DBX_CONTIN_LENGTH): Likewise.
15062 (DBX_CONTIN_CHAR): Likewise.
15063 (DBX_NO_XREFS): Likewise.
15064 (DBX_STATIC_STAB_DATA_SECTION): Likewise.
15065 * config/vx-common.h (DBX_DEBUGGING_INFO): Likewise.
15066 (XCOFF_DEBUGGING_INFO): Likewise.
15067 * configure: Regenerate. Likewise.
15068 * configure.ac: Likewise.
15069 * debug.h: Likewise.
15070 * doc/install.texi: Likewise.
15071 * doc/invoke.texi: Likewise.
15072 * doc/passes.texi: Likewise.
15073 * doc/tm.texi: Likewise.
15074 * doc/tm.texi.in: Likewise.
15075 * dwarf2asm.cc (XCOFF_DEBUGGING_INFO): Likewise.
15076 (dw2_asm_output_nstring): Likewise.
15077 (USE_LINKONCE_INDIRECT): Likewise.
15078 * dwarf2out.cc (XCOFF_DEBUGGING_INFO): Likewise.
15079 (HAVE_XCOFF_DWARF_EXTRAS): Likewise.
15080 (output_fde): Likewise.
15081 (output_call_frame_info): Likewise.
15082 (have_macinfo): Likewise.
15083 (add_AT_loc_list): Likewise.
15084 (add_AT_view_list): Likewise.
15085 (output_compilation_unit_header): Likewise.
15086 (output_pubnames): Likewise.
15087 (output_aranges): Likewise.
15088 (output_line_info): Likewise.
15089 (output_macinfo): Likewise.
15090 (dwarf2out_finish): Likewise.
15091 (dwarf2out_early_finish): Likewise.
15092 * final.cc (final_scan_insn_1): Likewise.
15093 (rest_of_handle_final): Likewise.
15094 * flag-types.h (enum debug_info_type): Likewise.
15095 (DBX_DEBUG): Likewise.
15096 (XCOFF_DEBUG): Likewise.
15097 * function.cc (defined): Likewise.
15098 * gcc.cc (defined): Likewise.
15099 (ASM_DEBUG_SPEC): Likewise.
15100 (ASM_DEBUG_OPTION_SPEC): Likewise.
15101 * opts.cc (common_handle_option): Likewise.
15102 (set_debug_level): Likewise.
15103 * system.h (fancy_abort): Likewise.
15104 * target-def.h (TARGET_ASM_CONSTRUCTOR): Likewise.
15105 (TARGET_ASM_DESTRUCTOR): Likewise.
15106 * toplev.cc (defined): Likewise.
15107 * varasm.cc: Likewise.
15108 * config/dbxcoff.h: Removed.
15109 * config/dbxelf.h: Removed.
15110 * dbxout.cc: Removed.
15111 * dbxout.h: Removed.
15112 * gstab.h: Removed.
15113 * stab.def: Removed.
15114 * xcoffout.cc: Removed.
15115 * xcoffout.h: Removed.
15117 2022-09-02 Simon Rainer <gcc.gnu@vvalter.com>
15120 * config/i386/i386-features.cc (ix86_get_function_versions_dispatcher):
15121 Set TREE_NOTHROW correctly for dispatcher declaration.
15122 * config/rs6000/rs6000.cc (rs6000_get_function_versions_dispatcher):
15125 2022-09-01 Tamar Christina <tamar.christina@arm.com>
15128 * config/aarch64/aarch64.cc
15129 (aarch64_vector_costs::prefer_unrolled_loop): Replace %u with
15130 HOST_WIDE_INT_PRINT_UNSIGNED.
15132 2022-09-01 Richard Sandiford <richard.sandiford@arm.com>
15134 * tree-vect-slp.cc (vect_build_slp_tree_2): When building a
15135 VEC_PERM_EXPR of an existing vector, set the SLP_TREE_LANES
15136 to the number of vector elements, if that's a known constant.
15137 (vect_optimize_slp_pass::is_compatible_layout): Remove associated
15138 comment about zero SLP_TREE_LANES.
15139 (vect_optimize_slp_pass::start_choosing_layouts): Iterate over
15140 all partition members when looking for potential layouts.
15141 Handle existing permutes of fixed-length vectors.
15143 2022-09-01 Uroš Bizjak <ubizjak@gmail.com>
15146 * config/i386/i386.md (moves to/from AX_REG into xchg peephole2):
15147 Do not convert a move pattern where both operands are AX_REG.
15149 2022-09-01 Aldy Hernandez <aldyh@redhat.com>
15151 * range-op-float.cc (build_le): Convert to dconst*inf.
15153 * value-range.cc (frange::set_signbit): Same.
15154 (frange::normalize_kind): Same.
15155 (range_tests_floats): Same.
15156 * value-range.h (vrp_val_max): Same.
15157 (vrp_val_min): Same.
15158 (frange::set_varying): Same.
15160 2022-09-01 Aldy Hernandez <aldyh@redhat.com>
15162 * emit-rtl.cc (init_emit_once): Initialize dconstinf and
15164 * real.h: Add dconstinf and dconstninf.
15166 2022-09-01 Richard Biener <rguenther@suse.de>
15168 * gimple-predicate-analysis.cc (compute_control_dep_chain):
15169 Remove cycle detection, instead avoid walking backedges.
15171 2022-09-01 Richard Biener <rguenther@suse.de>
15173 * gimple-predicate-analysis.cc (compute_control_dep_chain):
15174 New wrapping overload.
15175 (uninit_analysis::init_use_preds): Simplify.
15176 (uninit_analysis::init_from_phi_def): Likewise.
15178 2022-09-01 Aldy Hernandez <aldyh@redhat.com>
15180 * gimple-range-fold.cc
15181 (fold_using_range::range_of_builtin_int_call): Add case for
15182 CFN_BUILT_IN_SIGNBIT.
15184 2022-09-01 Aldy Hernandez <aldyh@redhat.com>
15186 * range-op-float.cc (foperator_equal::op1_range): Do not copy sign
15188 (foperator_not_equal::op1_range): Same.
15189 * value-query.cc (range_query::get_tree_range): Set sign bit.
15190 * value-range-pretty-print.cc (vrange_printer::visit): Dump sign bit.
15191 * value-range.cc (frange::set_signbit): New.
15192 (frange::set): Adjust for sign bit.
15193 (frange::normalize_kind): Same.
15194 (frange::union_): Remove useless comment.
15195 (frange::intersect): Same.
15196 (frange::contains_p): Adjust for sign bit.
15197 (frange::singleton_p): Same.
15198 (frange::verify_range): Same.
15199 (range_tests_signbit): New tests.
15200 (range_tests_floats): Call range_tests_signbit.
15201 * value-range.h (class frange_props): Add signbit
15202 (class frange): Same.
15204 2022-09-01 Jakub Jelinek <jakub@redhat.com>
15207 * dumpfile.h (dump_printf_loc): Use ATTRIBUTE_GCC_DUMP_PRINTF (3, 4)
15208 instead of ATTRIBUTE_GCC_DUMP_PRINTF (3, 0).
15209 * tree-parloops.cc (parloops_is_slp_reduction): Cast pointers to
15210 derived types of gimple to gimple * to avoid -Wformat warnings.
15211 * tree-vect-loop-manip.cc (vect_set_loop_condition,
15212 vect_update_ivs_after_vectorizer): Likewise.
15213 * tree-vect-stmts.cc (vectorizable_load): Likewise.
15214 * tree-vect-patterns.cc (vect_split_statement,
15215 vect_recog_mulhs_pattern, vect_recog_average_pattern,
15216 vect_determine_precisions_from_range,
15217 vect_determine_precisions_from_users): Likewise.
15218 * gimple-loop-versioning.cc
15219 (loop_versioning::analyze_term_using_scevs): Likewise.
15220 * tree-vect-slp.cc (vect_build_slp_tree_1): Likewise.
15221 (vect_build_slp_tree): Cast slp_tree to void * to avoid
15223 (optimize_load_redistribution_1, vect_match_slp_patterns,
15224 vect_build_slp_instance, vect_optimize_slp_pass::materialize,
15225 vect_optimize_slp_pass::dump, vect_slp_convert_to_external,
15226 vect_slp_analyze_node_operations, vect_bb_partition_graph): Likewise.
15227 (vect_print_slp_tree): Likewise. Also use
15228 HOST_WIDE_INT_PRINT_UNSIGNED instead of %u.
15229 * tree-vect-loop.cc (vect_determine_vectorization_factor,
15230 vect_analyze_scalar_cycles_1, vect_analyze_loop_operations,
15231 vectorizable_induction, vect_transform_loop): Cast pointers to derived
15232 types of gimple to gimple * to avoid -Wformat warnings.
15233 (vect_analyze_loop_2): Cast slp_tree to void * to avoid
15235 (vect_estimate_min_profitable_iters): Use HOST_WIDE_INT_PRINT_UNSIGNED
15237 * tree-vect-slp-patterns.cc (vect_pattern_validate_optab): Use %G
15238 instead of %T and STMT_VINFO_STMT (SLP_TREE_REPRESENTATIVE (node))
15239 instead of SLP_TREE_DEF_TYPE (node).
15241 2022-09-01 Jakub Jelinek <jakub@redhat.com>
15244 * doc/invoke.texi (-Winvalid-utf8): Document it.
15246 2022-09-01 Aldy Hernandez <aldyh@redhat.com>
15248 PR tree-optimization/106785
15249 * value-range.cc (range_tests_nan): Adjust tests for !HONOR_NANS.
15250 (range_tests_floats): Same.
15252 2022-09-01 konglin1 <lingling.kong@intel.com>
15254 * tree-if-conv.cc (is_cond_scalar_reduction): Add MULT_EXPR
15257 2022-09-01 Peter Bergner <bergner@linux.ibm.com>
15260 * config/rs6000/rs6000-builtin.cc (rs6000_gimple_fold_mma_builtin):
15261 Enforce the use of a valid MMA pointer type.
15263 2022-09-01 zhongjuzhe <juzhe.zhong@rivai.ai>
15265 * config/riscv/riscv.cc (riscv_conditional_register_usage): Add vector
15268 2022-09-01 zhongjuzhe <juzhe.zhong@rivai.ai>
15270 * config/riscv/riscv.cc (riscv_const_insns): Add cost of poly_int.
15271 (riscv_output_move): Add csrr vlenb assembly.
15272 * config/riscv/riscv.md (move_type): Add csrr vlenb type.
15273 (ext): New attribute.
15274 (ext_enabled): Ditto.
15277 2022-09-01 zhongjuzhe <juzhe.zhong@rivai.ai>
15279 * config/riscv/constraints.md (TARGET_VECTOR ? V_REGS : NO_REGS): Add
15281 (TARGET_VECTOR ? VD_REGS : NO_REGS): Add "vd" constraint.
15282 (TARGET_VECTOR ? VM_REGS : NO_REGS): Add "vm" constraint.
15283 (vp): Add poly constraint.
15285 2022-09-01 zhongjuzhe <juzhe.zhong@rivai.ai>
15287 * config/riscv/riscv.h (enum reg_class): Change vype to vtype.
15289 2022-09-01 zhongjuzhe <juzhe.zhong@rivai.ai>
15291 * config/riscv/riscv.cc (riscv_convert_vector_bits): Change
15292 configuration according to TARGET_MIN_VLEN.
15293 * config/riscv/riscv.h (UNITS_PER_FP_REG): Fix comment.
15295 2022-08-31 Martin Liska <mliska@suse.cz>
15297 * config.build: Remove deprecated ports.
15298 * config.gcc: Likewise.
15299 * config.host: Likewise.
15300 * configure.ac: Likewise.
15301 * configure: Regenerate.
15302 * config/pa/pa-hpux10.h: Removed.
15303 * config/pa/pa-hpux10.opt: Removed.
15304 * config/pa/t-dce-thr: Removed.
15306 2022-08-31 Martin Liska <mliska@suse.cz>
15308 * Makefile.in: Always install limits.h and syslimits.h to
15310 * configure.ac: Assign STMP_FIXINC blank if
15311 --disable-fixincludes is used.
15312 * configure: Regenerate.
15314 2022-08-31 Richard Biener <rguenther@suse.de>
15316 * gimple-predicate-analysis.cc (predicate::init_from_control_deps):
15317 Assert the guard_bb isn't empty and has more than one successor.
15318 Drop appropriate parts of the predicate when an edge fails to
15319 register a predicate.
15320 (predicate::dump): Dump empty predicate as TRUE.
15322 2022-08-31 Richard Biener <rguenther@suse.de>
15324 PR tree-optimization/90994
15325 * gimple-predicate-analysis.cc (predicate::init_from_control_deps):
15326 Ignore exceptional control flow and skip the edge for the purpose of
15327 predicate generation also for non-calls.
15329 2022-08-31 Aldy Hernandez <aldyh@redhat.com>
15331 * value-range-storage.cc (frange_storage_slot::set_frange): Save
15333 (frange_storage_slot::get_frange): Restore endpoints.
15334 * value-range-storage.h (class frange_storage_slot): Add endpoint
15337 2022-08-31 Martin Liska <mliska@suse.cz>
15339 PR tree-optimization/106789
15340 * range-op-float.cc (default_frelop_fold_range): Remove the
15343 2022-08-31 Martin Liska <mliska@suse.cz>
15345 * value-range.h: Add more override keywords.
15347 2022-08-31 Martin Liska <mliska@suse.cz>
15349 * value-range.h: Add override.
15351 2022-08-31 Richard Biener <rguenther@suse.de>
15353 PR tree-optimization/65244
15354 * gimple-predicate-analysis.h (predicate::init_from_control_deps):
15355 Add argument to specify whether the predicate is for the USE.
15356 * gimple-predicate-analysis.cc (predicate::init_from_control_deps):
15357 Also include predicates effective fallthru control edges when
15358 the predicate is for the USE.
15360 2022-08-31 Richard Biener <rguenther@suse.de>
15362 PR tree-optimization/73550
15363 * gimple-predicate-analysis.cc (predicate::init_from_control_deps):
15364 Sanitize debug dumping. Handle case labels with a CASE_HIGH.
15365 (predicate::dump): Adjust for better readability.
15367 2022-08-31 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
15368 Jakub Jelinek <jakub@redhat.com>
15370 * omp-simd-clone.cc (simd_clone_adjust_return_type,
15371 simd_clone_adjust_argument_types): Use known_eq (veclen, 0U)
15372 instead of known_eq (veclen, 0) to avoid -Wsign-compare warnings.
15374 2022-08-31 Richard Sandiford <richard.sandiford@arm.com>
15376 * tree-vect-slp.cc (vect_optimize_slp_pass::dump): Remove bogus
15379 2022-08-31 zhongjuzhe <juzhe.zhong@rivai.ai>
15381 * tree-vect-loop-manip.cc (vect_gen_vector_loop_niters): Simply
15382 initialize const_vf to 0.
15384 2022-08-31 Martin Liska <mliska@suse.cz>
15386 * config.gcc: Remove cr16.
15388 2022-08-30 Martin Jambor <mjambor@suse.cz>
15390 * vec.h (array_slice): Add constructors for non-const reference to
15391 heap vector and pointers to heap vectors.
15393 2022-08-30 Aldy Hernandez <aldyh@redhat.com>
15395 * value-range.cc (early_nan_resolve): Change comment.
15396 (frange::union_): Handle union when one side is a NAN.
15397 (range_tests_nan): Add tests for NAN union.
15399 2022-08-30 Andrew Stubbs <ams@codesourcery.com>
15401 * config/gcn/gcn.cc (gcn_simd_clone_compute_vecsize_and_simdlen): New.
15402 (gcn_simd_clone_adjust): New.
15403 (gcn_simd_clone_usable): New.
15404 (TARGET_SIMD_CLONE_ADJUST): New.
15405 (TARGET_SIMD_CLONE_COMPUTE_VECSIZE_AND_SIMDLEN): New.
15406 (TARGET_SIMD_CLONE_USABLE): New.
15408 2022-08-30 Andrew Stubbs <ams@codesourcery.com>
15410 * doc/tm.texi: Regenerate.
15411 * omp-simd-clone.cc (simd_clone_adjust_return_type): Allow zero
15413 (simd_clone_adjust_argument_types): Likewise.
15414 * target.def (compute_vecsize_and_simdlen): Document the new
15415 vecsize_int and vecsize_float semantics.
15417 2022-08-30 Richard Sandiford <richard.sandiford@arm.com>
15419 * expmed.cc (store_bit_field_1): Fix byte offset calculation
15420 for undefined structures.
15422 2022-08-30 Richard Sandiford <richard.sandiford@arm.com>
15424 * params.opt (-param=vect-max-layout-candidates=): New parameter.
15425 * doc/invoke.texi (vect-max-layout-candidates): Document it.
15426 * tree-vectorizer.h (auto_lane_permutation_t): New typedef.
15427 (auto_load_permutation_t): Likewise.
15428 * tree-vect-slp.cc (vect_slp_node_weight): New function.
15429 (slpg_layout_cost): New class.
15430 (slpg_vertex): Replace perm_in and perm_out with partition,
15431 out_degree, weight and out_weight.
15432 (slpg_partition_info, slpg_partition_layout_costs): New classes.
15433 (vect_optimize_slp_pass): Likewise, cannibalizing some part of
15434 the previous vect_optimize_slp.
15435 (vect_optimize_slp): Use it.
15437 2022-08-30 Richard Sandiford <richard.sandiford@arm.com>
15439 * hash-traits.h (vec_hash_base): New class.
15440 (vec_free_hash_base): Likewise.
15442 2022-08-30 Richard Sandiford <richard.sandiford@arm.com>
15444 * hash-traits.h (int_hash_base): New struct, split out from...
15445 (int_hash): ...this class, which now inherits from int_hash_base.
15446 * hash-map-traits.h (unbounded_hashmap_traits): Take a template
15447 parameter for the key that provides hash and equality functions.
15448 (unbounded_int_hashmap_traits): Turn into a type alias of
15449 unbounded_hashmap_traits.
15451 2022-08-30 Richard Sandiford <richard.sandiford@arm.com>
15453 * graphds.cc (graphds_scc): Add a pass-back parameter for the
15455 * graphds.h (graphds_scc): Update prototype accordingly.
15457 2022-08-30 Richard Sandiford <richard.sandiford@arm.com>
15459 * tree-vect-slp.cc (vect_transform_slp_perm_load_1): Split out from...
15460 (vect_transform_slp_perm_load): ...here. Use SLP_TREE_VECTYPE instead
15461 of STMT_VINFO_VECTYPE.
15463 2022-08-30 Richard Sandiford <richard.sandiford@arm.com>
15465 * tree-vect-slp.cc (vectorizable_slp_permutation_1): Split out from...
15466 (vectorizable_slp_permutation): ...here.
15468 2022-08-30 Richard Sandiford <richard.sandiford@arm.com>
15470 * tree-vect-stmts.cc (get_related_vectype_for_scalar_type): Check
15471 that the requested number of units is interoperable with the requested
15474 2022-08-30 Martin Liska <mliska@suse.cz>
15476 * config.gcc: Remove the port.
15477 * config/m32c/rtems.h: Removed.
15479 2022-08-30 Richard Biener <rguenther@suse.de>
15481 PR tree-optimization/73550
15482 * gimple-predicate-analysis.cc (predicate::init_from_control_deps):
15483 Do not apply MAX_NUM_CHAINS again.
15485 2022-08-30 Richard Biener <rguenther@suse.de>
15487 * gimple-predicate-analysis.cc (format_edge_vec): Dump
15488 both source and destination.
15489 (dump_dep_chains): Remove.
15490 (uninit_analysis::init_use_preds): Remove redundant
15493 2022-08-30 Aldy Hernandez <aldyh@redhat.com>
15495 * value-range-storage.cc (frange_storage_slot::get_frange): Use
15497 * value-range.cc (frange::set_nan): New.
15498 (frange_nan): Move to header file.
15499 (range_tests_nan): Adjust frange_nan callers to pass type.
15501 * value-range.h (FRANGE_PROP_ACCESSOR): Remove.
15504 2022-08-30 Richard Biener <rguenther@suse.de>
15506 PR tree-optimization/67196
15507 * gimple-predicate-analysis.cc (uninit_analysis::is_use_guarded):
15508 Simplify and normalize use prediates before first use.
15510 2022-08-30 Richard Biener <rguenther@suse.de>
15512 * gimple-predicate-analysis.cc (dump_pred_chain): Fix
15513 parentizing and AND prepending.
15514 (predicate::dump): Do not dump the GENERIC expanded
15515 predicate, properly parentize and prepend ORs to the
15516 piecewise predicate dump.
15517 (build_pred_expr): Remove.
15519 2022-08-30 Aldy Hernandez <aldyh@redhat.com>
15521 * range-op-float.cc (finite_operand_p): New.
15526 (foperator_equal::fold_range): New implementation with endpoints.
15527 (foperator_equal::op1_range): Same.
15528 (foperator_not_equal::fold_range): Same.
15529 (foperator_not_equal::op1_range): Same.
15530 (foperator_lt::fold_range): Same.
15531 (foperator_lt::op1_range): Same.
15532 (foperator_lt::op2_range): Same.
15533 (foperator_le::fold_range): Same.
15534 (foperator_le::op1_range): Same.
15535 (foperator_le::op2_range): Same.
15536 (foperator_gt::fold_range): Same.
15537 (foperator_gt::op1_range): Same.
15538 (foperator_gt::op2_range): Same.
15539 (foperator_ge::fold_range): Same.
15540 (foperator_ge::op1_range): Same.
15541 (foperator_ge::op2_range): Same.
15543 2022-08-30 Aldy Hernandez <aldyh@redhat.com>
15545 * range-op-float.cc (frange_set_nan): New.
15546 (frange_drop_inf): New.
15547 (frange_drop_ninf): New.
15548 (foperator_equal::op1_range): Adjust for endpoints.
15549 (foperator_lt::op1_range): Same.
15550 (foperator_lt::op2_range): Same.
15551 (foperator_gt::op1_range): Same.
15552 (foperator_gt::op2_range): Same.
15553 (foperator_unordered::op1_range): Same.
15554 * value-query.cc (range_query::get_tree_range): Same.
15555 * value-range-pretty-print.cc (vrange_printer::visit): Same.
15556 * value-range-storage.cc (frange_storage_slot::get_frange): Same.
15557 * value-range.cc (frange::set): Same.
15558 (frange::normalize_kind): Same.
15559 (frange::union_): Same.
15560 (frange::intersect): Same.
15561 (frange::operator=): Same.
15562 (early_nan_resolve): New.
15563 (frange::contains_p): New.
15564 (frange::singleton_p): New.
15565 (frange::set_nonzero): New.
15566 (frange::nonzero_p): New.
15567 (frange::set_zero): New.
15568 (frange::zero_p): New.
15569 (frange::set_nonnegative): New.
15570 (frange_float): New.
15572 (range_tests_nan): New.
15573 (range_tests_signed_zeros): New.
15574 (range_tests_floats): New.
15575 (range_tests): New.
15576 * value-range.h (frange::lower_bound): New.
15577 (frange::upper_bound): New.
15578 (vrp_val_min): Use real_inf with a sign instead of negating inf.
15579 (frange::frange): New.
15580 (frange::set_varying): Adjust for endpoints.
15581 (real_max_representable): New.
15582 (real_min_representable): New.
15584 2022-08-30 Aldy Hernandez <aldyh@redhat.com>
15586 * match.pd ((cmp @0 zerop) real_zerop (negate@1 @0)): Add variant
15589 2022-08-30 Martin Liska <mliska@suse.cz>
15591 * config/s390/s390.cc (s390_rtx_costs): Use proper type as
15594 2022-08-30 Richard Biener <rguenther@suse.de>
15596 * tree-ssa-uninit.cc (warn_uninitialized_vars): Pre-compute
15597 the set of fallthru reachable blocks from function entry
15598 and use that to determine wlims.always_executed.
15600 2022-08-30 Richard Biener <rguenther@suse.de>
15602 PR tree-optimization/56654
15603 * tree-ssa-uninit.cc (cand_cmp): New.
15604 (find_uninit_use): First process all PHIs and collect candidate
15605 stmts, then sort those after RPO.
15606 (warn_uninitialized_phi): Pass on bb_to_rpo.
15607 (execute_late_warn_uninitialized): Compute and pass on
15608 reverse lookup of RPO number from basic block index.
15610 2022-08-30 Richard Biener <rguenther@suse.de>
15612 * gimple-predicate-analysis.h (uninit_analysis::operator()):
15614 * gimple-predicate-analysis.cc
15615 (uninit_analysis::collect_phi_def_edges): Use phi_arg_set,
15617 * tree-ssa-uninit.cc (defined_args): New global.
15618 (compute_uninit_opnds_pos): Mask with the recorded set
15619 of guarded maybe-uninitialized uses.
15620 (uninit_undef_val_t::operator()): Remove.
15621 (find_uninit_use): Process all PHI uses, recording the
15622 guarded ones and marking the PHI result as uninitialized
15624 (warn_uninitialized_phi): Adjust.
15625 (execute_late_warn_uninitialized): Get rid of the PHI worklist
15626 and instead walk the function in RPO order.
15627 * spellcheck.h (best_match::m_best_candidate_len): Initialize.
15629 2022-08-30 Tamar Christina <tamar.christina@arm.com>
15631 PR tree-optimization/106744
15632 * tree-ssa-phiopt.cc (minmax_replacement): Correct arguments.
15634 2022-08-30 Tamar Christina <tamar.christina@arm.com>
15636 * expmed.cc (store_bit_field_1): Initialize regnum to 0.
15638 2022-08-29 David Faust <david.faust@oracle.com>
15641 * config/bpf/coreout.cc (bpf_core_get_sou_member_index): Fix
15642 computation of index for anonymous members.
15644 2022-08-29 Jose E. Marchesi <jose.marchesi@oracle.com>
15646 * config/bpf/bpf.cc (bpf_target_macros): Define __bpf__ as a
15649 2022-08-29 H.J. Lu <hjl.tools@gmail.com>
15652 * config/i386/i386-expand.cc
15653 (ix86_avx256_split_vector_move_misalign): Handle E_V16BFmode.
15654 * config/i386/sse.md (V_256H): Add V16BF.
15656 2022-08-29 Robin Dapp <rdapp@linux.ibm.com>
15658 * config/s390/s390.cc (s390_address_cost): Declare.
15659 (s390_hard_regno_nregs): Declare.
15660 (s390_rtx_costs): Add handling for REG and MEM in SET.
15662 2022-08-29 Robin Dapp <rdapp@linux.ibm.com>
15664 * config/s390/s390.cc (expand_perm_with_vpdi): Recognize swap pattern.
15665 (is_reverse_perm_mask): New function.
15666 (expand_perm_with_rot): Recognize reverse pattern.
15667 (expand_perm_with_vstbrq): New function.
15668 (expand_perm_with_vster): Use vler/vster for element reversal on z15.
15669 (vectorize_vec_perm_const_1): Use.
15670 (s390_vectorize_vec_perm_const): Add expand functions.
15671 * config/s390/vx-builtins.md: Prefer vster over vler.
15673 2022-08-29 Robin Dapp <rdapp@linux.ibm.com>
15675 * config/s390/s390.md: Remove UNSPEC_VEC_EXTRACT.
15676 * config/s390/vector.md: Rewrite patterns to use vec_select.
15677 * config/s390/vx-builtins.md (vec_scatter_element<V_HW_2:mode>_SI):
15680 2022-08-29 Robin Dapp <rdapp@linux.ibm.com>
15683 * config/s390/vector.md (@vpdi4_2<mode>): New pattern.
15684 (rotl<mode>3_di): New pattern.
15685 * config/s390/vx-builtins.md: Use vpdi and verll for reversing
15688 2022-08-29 Robin Dapp <rdapp@linux.ibm.com>
15690 * config/s390/s390.cc (s390_issue_rate): Add z15.
15692 2022-08-29 Robin Dapp <rdapp@linux.ibm.com>
15694 * common/config/s390/s390-common.cc: Enable -funroll-loops and
15695 -munroll-only-small-loops for OPT_LEVELS_2_PLUS_SPEED_ONLY.
15696 * config/s390/s390.cc (s390_loop_unroll_adjust): Do not unroll
15697 loops larger than 12 instructions.
15698 (s390_override_options_after_change): Set unroll options.
15699 (s390_option_override_internal): Likewise.
15700 * config/s390/s390.opt: Document munroll-only-small-loops.
15702 2022-08-29 Richard Biener <rguenther@suse.de>
15704 * gimple-predicate-analysis.cc (is_loop_exit,
15705 find_control_equiv_block): Inline into single caller ...
15706 (uninit_analysis::init_use_preds): ... here and refactor.
15708 2022-08-29 Richard Biener <rguenther@suse.de>
15710 * gimple-predicate-analysis.cc (compute_control_dep_chain):
15711 Inline is_loop_exit and refactor, add comment about
15714 2022-08-29 Kito Cheng <kito.cheng@sifive.com>
15716 * config/riscv/riscv.cc (riscv_frame_info): Introduce `reset(void)`;
15717 (riscv_frame_info::reset(void)): New.
15718 (riscv_compute_frame_info): Use riscv_frame_info::reset instead
15719 of memset when clean frame.
15721 2022-08-29 zhongjuzhe <juzhe.zhong@rivai.ai>
15723 * config/riscv/riscv.cc (riscv_v_ext_vector_mode_p): New function.
15724 (riscv_classify_address): Disallow PLUS/LO_SUM/CONST_INT address types for RVV.
15725 (riscv_address_insns): Add RVV modes condition.
15726 (riscv_binary_cost): Ditto.
15727 (riscv_rtx_costs): Adjust cost for RVV.
15728 (riscv_secondary_memory_needed): Add RVV modes condition.
15729 (riscv_hard_regno_nregs): Add RVV register allocation.
15730 (riscv_hard_regno_mode_ok): Add RVV register allocation.
15731 (riscv_class_max_nregs): Add RVV register allocation.
15732 * config/riscv/riscv.h (DWARF_FRAME_REGNUM): Add VL/VTYPE and vector registers in Dwarf.
15733 (UNITS_PER_V_REG): New macro.
15734 (FIRST_PSEUDO_REGISTER): Adjust first pseudo num for RVV.
15735 (V_REG_FIRST): New macro.
15736 (V_REG_LAST): Ditto.
15737 (V_REG_NUM): Ditto.
15740 (VTYPE_REG_P): Ditto.
15741 (RISCV_DWARF_VL): Ditto.
15742 (RISCV_DWARF_VTYPE): Ditto.
15743 (enum reg_class): Add RVV register types.
15744 (REG_CLASS_CONTENTS): Add RVV register types.
15745 * config/riscv/riscv.md: Add VL/VTYPE register number constants.
15747 2022-08-29 zhongjuzhe <juzhe.zhong@rivai.ai>
15749 * config/riscv/riscv.md: Add new type for vector instructions.
15751 2022-08-28 Peter Bergner <bergner@linux.ibm.com>
15754 * config/rs6000/rs6000.cc (rs6000_invalid_conversion): Remove handling
15755 of MMA pointer conversions.
15757 2022-08-27 Xi Ruoyao <xry111@xry111.site>
15759 * config/i386/gcc-auto-profile: Regenerate.
15761 2022-08-26 Aldy Hernandez <aldyh@redhat.com>
15763 * real.cc (real_iszero): New.
15764 * real.h (real_iszero): New.
15766 2022-08-26 Aldy Hernandez <aldyh@redhat.com>
15768 * real.cc (real_isinf): New overload.
15769 (real_inf): Add sign argument.
15770 * real.h (real_isinf): New overload.
15771 (real_inf): Add sign argument.
15773 2022-08-26 Marek Polacek <polacek@redhat.com>
15776 * doc/invoke.texi: Document -Wself-move.
15778 2022-08-26 Aldy Hernandez <aldyh@redhat.com>
15780 * value-range.cc (vrange::set): Set varying.
15781 (vrange::set_nonzero): Same.
15782 (vrange::set_zero): Same.
15783 (vrange::set_nonnegative): Same.
15785 2022-08-26 Aldy Hernandez <aldyh@redhat.com>
15787 * range-op-float.cc (foperator_equal::op1_range): Do not blindly
15788 copy op2 range when honoring signed zeros.
15790 2022-08-26 Aldy Hernandez <aldyh@redhat.com>
15792 * tree-ssa-threadbackward.cc (possibly_profitable_path_p): Always
15794 (profitable_path_p): Same.
15796 2022-08-26 Richard Biener <rguenther@suse.de>
15798 * gimple-predicate-analysis.h
15799 (uninit_analysis::use_cannot_happen): Remove.
15800 * gimple-predicate-analysis.cc (can_be_invalidated_p): Remove.
15801 (uninit_analysis::use_cannot_happen): Likewise.
15802 (uninit_analysis::is_use_guarded): Do not call
15804 (dump_predicates): Remove.
15805 (simple_control_dep_chain): Remove edge overload.
15807 2022-08-26 Tobias Burnus <tobias@codesourcery.com>
15809 * internal-fn.cc (expand_GOMP_TARGET_REV): New.
15810 * internal-fn.def (GOMP_TARGET_REV): New.
15811 * lto-cgraph.cc (lto_output_node, verify_node_partition): Mark
15812 'omp target device_ancestor_host' as in_other_partition and don't
15814 * omp-low.cc (create_omp_child_function): Mark as 'noclone'.
15815 * omp-expand.cc (expand_omp_target): For reverse offload, remove
15816 sorry, use device = GOMP_DEVICE_HOST_FALLBACK and create
15817 empty-body nohost function.
15818 * omp-offload.cc (execute_omp_device_lower): Handle
15819 IFN_GOMP_TARGET_REV.
15820 (pass_omp_target_link::execute): For ACCEL_COMPILER, don't
15821 nullify fn argument for reverse offload
15823 2022-08-26 Jakub Jelinek <jakub@redhat.com>
15825 * builtins.def (BUILT_IN_ISSIGNALING): New built-in.
15826 * builtins.cc (expand_builtin_issignaling): New function.
15827 (expand_builtin_signbit): Don't overwrite target.
15828 (expand_builtin): Handle BUILT_IN_ISSIGNALING.
15829 (fold_builtin_classify): Likewise.
15830 (fold_builtin_1): Likewise.
15831 * optabs.def (issignaling_optab): New.
15832 * fold-const-call.cc (fold_const_call_ss): Handle
15833 BUILT_IN_ISSIGNALING.
15834 * config/i386/i386.md (issignalingxf2): New expander.
15835 * doc/extend.texi (__builtin_issignaling): Document.
15836 (__builtin_isinf, __builtin_isnan): Clarify behavior with
15837 -ffinite-math-only.
15838 * doc/md.texi (issignaling<mode>2): Likewise.
15840 2022-08-26 Jakub Jelinek <jakub@redhat.com>
15842 PR tree-optimization/106099
15843 * internal-fn.def (TRAP): Add ECF_LOOPING_CONST_OR_PURE flag.
15844 * tree-cfg.cc (execute_fixup_cfg): Add IFN_TRAP instead of
15845 __builtin_trap to avoid the need of vops.
15847 2022-08-26 Richard Biener <rguenther@suse.de>
15849 * gimple-predicate-analysis.cc (dfs_mark_dominating_region):
15851 (compute_control_dep_chain): Adjust to honor marked region
15853 (uninit_analysis::init_from_phi_def): Pre-mark the dominating
15854 region to improve compute_control_dep_chain walking.
15855 * vec.h (vec<T, va_heap, vl_ptr>::allocated): Add forwarder.
15857 2022-08-26 Richard Biener <rguenther@suse.de>
15859 * gimple-predicate-analysis.cc
15860 (uninit_analysis::collect_phi_def_edges): Only expand a
15861 PHI def edge when it is possibly undefined.
15863 2022-08-26 Martin Liska <mliska@suse.cz>
15865 * doc/extend.texi: Remove cr16 related stuff.
15866 * doc/install.texi: Likewise.
15867 * doc/invoke.texi: Likewise.
15868 * doc/md.texi: Likewise.
15869 * function-tests.cc (test_expansion_to_rtl): Likewise.
15870 * common/config/cr16/cr16-common.cc: Removed.
15871 * config/cr16/constraints.md: Removed.
15872 * config/cr16/cr16-protos.h: Removed.
15873 * config/cr16/cr16.cc: Removed.
15874 * config/cr16/cr16.h: Removed.
15875 * config/cr16/cr16.md: Removed.
15876 * config/cr16/cr16.opt: Removed.
15877 * config/cr16/predicates.md: Removed.
15878 * config/cr16/t-cr16: Removed.
15880 2022-08-26 liuhongt <hongtao.liu@intel.com>
15883 * config/i386/i386-builtin.def (BDESC): Add
15884 CODE_FOR_avx_blendvpd256/CODE_FOR_avx_blendvps256 to
15885 corresponding builtins.
15886 * config/i386/i386.cc (ix86_gimple_fold_builtin):
15887 Don't fold IX86_BUILTIN_PBLENDVB256, IX86_BUILTIN_BLENDVPS256,
15888 IX86_BUILTIN_BLENDVPD256 w/o TARGET_AVX2.
15890 2022-08-25 Marek Polacek <polacek@redhat.com>
15892 * ginclude/stddef.h: Define nullptr_t.
15894 2022-08-25 Joseph Myers <joseph@codesourcery.com>
15896 * gimplify.cc (gimplify_modify_expr): Convert initialization from
15897 a variable-size CONSTRUCTOR to memset before call to
15898 gimplify_modify_expr_rhs.
15900 2022-08-25 Jason Merrill <jason@redhat.com>
15902 * dwarf2out.cc (base_type_die): Also use DW_ATE_UTF for char8_t.
15904 2022-08-25 Andreas Krebbel <krebbel@linux.ibm.com>
15907 * config/s390/predicates.md (subreg_register_operand): New
15909 * config/s390/s390-protos.h (s390_gen_lowpart_subreg): New
15910 function prototype.
15911 * config/s390/s390.cc (s390_gen_lowpart_subreg): New function.
15912 (s390_expand_insv): Use s390_gen_lowpart_subreg instead of
15914 * config/s390/s390.md ("*get_tp_64", "*zero_extendhisi2_31")
15915 ("*zero_extendqisi2_31", "*zero_extendqihi2_31"): Likewise.
15916 ("movstrictqi", "movstricthi", "movstrictsi"): Use the
15917 subreg_register_operand predicate instead of register_operand.
15919 2022-08-25 Xi Ruoyao <xry111@xry111.site>
15921 * config/loongarch/loongarch-protos.h (loongarch_symbol_type):
15922 Add SYMBOL_PCREL64 and change the description for SYMBOL_PCREL.
15923 * config/loongarch/loongarch.cc (loongarch_attribute_table):
15924 New attribute table.
15925 (TARGET_ATTRIBUTE_TABLE): Define the target hook.
15926 (loongarch_handle_model_attribute): New static function.
15927 (loongarch_classify_symbol): Take TARGET_CMODEL_EXTREME and the
15928 model attribute of SYMBOL_REF_DECL into account returning
15929 SYMBOL_PCREL or SYMBOL_PCREL64.
15930 (loongarch_use_anchors_for_symbol_p): New static function.
15931 (TARGET_USE_ANCHORS_FOR_SYMBOL_P): Define the target hook.
15932 (loongarch_symbol_extreme_p): New static function.
15933 (loongarch_symbolic_constant_p): Handle SYMBOL_PCREL64.
15934 (loongarch_symbol_insns): Likewise.
15935 (loongarch_split_symbol_type): Likewise.
15936 (loongarch_split_symbol): Check SYMBOL_PCREL64 instead of
15937 TARGET_CMODEL_EXTREME for PC-relative addressing.
15938 (loongarch_print_operand_reloc): Likewise.
15939 * doc/extend.texi (Variable Attributes): Document new
15940 LoongArch specific attribute.
15942 2022-08-25 Xi Ruoyao <xry111@xry111.site>
15944 * config/loongarch/loongarch.cc (loongarch_classify_symbol):
15945 Return early if the rtx is not SYMBOL_REF.
15947 2022-08-25 Richard Biener <rguenther@suse.de>
15949 PR tree-optimization/106737
15950 * tree-parloops.cc (transform_to_exit_first_loop_alt): Do not
15953 2022-08-25 Chenghua Xu <xuchenghua@loongson.cn>
15956 * config/loongarch/loongarch.cc (loongarch_build_integer):
15958 * config/loongarch/loongarch.h (IMM_REACH): Likewise.
15959 (HWIT_1U): New Defined.
15960 (LU12I_OPERAND): Use HOST_WIDE_INT.
15961 (LU32I_OPERAND): Likewise.
15962 (LU52I_OPERAND): Likewise.
15963 (HWIT_UC_0xFFF): Likwise.
15965 2022-08-24 Andrew Pinski <apinski@marvell.com>
15969 * config/riscv/bitmanip.md (*shNadduw): Use n constraint
15971 (*slliuw): Likewise.
15972 (*bexti): Likewise. Also add a check for operands[2] to be less
15973 than the mode bitsize.
15975 2022-08-24 Andrew Pinski <apinski@marvell.com>
15977 * config/riscv/constraints.md (DbS): New constraint.
15978 (DnS): New constraint.
15979 * config/riscv/bitmanip.md (*bset<mode>_1_mask): Use new constraint.
15980 (*bclr<mode>): Likewise.
15981 (*binvi<mode>): Likewise.
15983 2022-08-24 Andrew Pinski <apinski@marvell.com>
15986 * config/riscv/predicates.md (single_bit_mask_operand):
15987 Use SINGLE_BIT_MASK_OPERAND instead of directly calling pow2p_hwi.
15988 (not_single_bit_mask_operand): Likewise.
15989 * config/riscv/riscv.cc (riscv_build_integer_1): Don't special case
15990 1<<31 for 32bits as it is already handled.
15991 Call trunc_int_for_mode on the upper part after the subtraction.
15992 (riscv_move_integer): Call trunc_int_for_mode before generating
15993 the integer just make sure the constant has been sign extended
15995 (riscv_emit_int_compare): Call trunc_int_for_mode after doing the
15996 addition for the new rhs.
15997 * config/riscv/riscv.h (SINGLE_BIT_MASK_OPERAND): If !TARGET64BIT,
15998 then mask off the upper 32bits of the HWI as it will be sign extended.
16000 2022-08-24 Andrew Pinski <apinski@marvell.com>
16002 * config/riscv/constraints.md (DsS): New constraint.
16003 (DsD): New constraint.
16004 * config/riscv/iterators.md (shiftm1c): New iterator.
16005 * config/riscv/bitmanip.md (*bset<mode>_mask):
16007 (*bset<mode>_1_mask): Likewise.
16009 2022-08-24 Andrew Pinski <apinski@marvell.com>
16011 * config/riscv/constraints.md (Ds3): New constraint.
16012 * config/riscv/predicates.md (imm123_operand): New predicate.
16013 * config/riscv/bitmanip.md (*shNadd): Use Ds3 and imm123_operand.
16014 (*shNadduw): Likewise.
16016 2022-08-24 Andrew Pinski <apinski@marvell.com>
16018 * config/riscv/riscv.cc (riscv_print_operand):
16020 (riscv_print_operand_punct_valid_p): New function
16021 (TARGET_PRINT_OPERAND_PUNCT_VALID_P): Define.
16022 * config/riscv/bitmanip.md (<bitmanip_optab>si2/clz_ctz_pcnt):
16023 Use %~ instead of conditional the pattern on TARGET_64BIT.
16024 (rotrsi3): Likewise.
16025 (rotlsi3): Likewise.
16026 * config/riscv/riscv.md: Add ~ to the list of modifiers.
16027 (addsi3): Use %~ instead of conditional the pattern on TARGET_64BIT.
16028 (subsi3): Likewise.
16029 (negsi2): Likewise.
16030 (mulsi3): Likewise.
16031 (optab>si3/any_div): Likewise.
16032 (*add<mode>hi3): Likewise.
16033 (<optab>si3/any_shift): Likewise.
16035 2022-08-24 Andrew Pinski <apinski@marvell.com>
16037 * config/riscv/riscv.cc (riscv_print_operand): Make a mention to
16038 keep the list in riscv.md in sync with this list.
16039 * config/riscv/riscv.md: Add list of modifiers as comments.
16041 2022-08-24 Andrew Pinski <apinski@marvell.com>
16043 * config/riscv/sync.md (any_atomic, atomic_optab): Move to ...
16044 * config/riscv/iterators.md: Here.
16046 2022-08-24 Andrew Pinski <apinski@marvell.com>
16048 * config/riscv/bitmanip.md
16049 (bitmanip_bitwise, bitmanip_minmax, clz_ctz_pcna,
16050 tbitmanip_optab, bitmanip_insn, shiftm1): Move to ...
16051 * config/riscv/iterators.md: Here.
16053 2022-08-24 Andrew Pinski <apinski@marvell.com>
16055 * config/riscv/riscv.md (GPR): Move to new file.
16056 (P, X, BR): Likewise.
16057 (MOVE32, MOVE64, SHORT): Likewise.
16058 (HISI, SUPERQI, SUBX): Likewise.
16059 (ANYI, ANYF, SOFTF): Likewise.
16060 (size, load, default_load): Likewise.
16061 (softload, store, softstore): Likewise.
16062 (reg, fmt, ifmt, amo): Likewise.
16063 (UNITMODE, HALFMODE): Likewise.
16064 (RINT, rint_pattern, rint_rm): Likewise.
16065 (QUIET_COMPARISON, quiet_pattern, QUIET_PATTERN): Likewise.
16066 (any_extend, any_shiftrt, any_shift): Likewise.
16067 (any_bitwise): Likewise.
16068 (any_div, any_mod): Likewise.
16069 (any_gt, any_ge, any_lt, any_le): Likewise.
16071 (optab, insn): Likewise.
16072 * config/riscv/iterators.md: New file.
16074 2022-08-24 Andrew Pinski <apinski@marvell.com>
16077 * config/riscv/bitmanip.md (bswaphi2): New pattern.
16079 2022-08-24 Andrew Pinski <apinski@marvell.com>
16082 * config/riscv/bitmanip.md (bswap<mode>2): Remove
16083 condition on TARGET_64BIT as X is already conditional there.
16085 2022-08-24 Joseph Myers <joseph@codesourcery.com>
16087 * tree.cc (build_real): Give DFP dconst0 the minimum quantum
16088 exponent for the type.
16090 2022-08-24 Jose E. Marchesi <jose.marchesi@oracle.com>
16093 * config/bpf/bpf.cc (bpf_legitimate_address_p): Recognize integer
16094 constants as legitimate addresses for functions.
16095 (bpf_small_register_classes_for_mode_p): Define target hook.
16097 2022-08-24 Richard Biener <rguenther@suse.de>
16099 * gimple-predicate-analysis.cc: Move predicate normalization
16100 after the comment documenting it.
16102 2022-08-24 Richard Biener <rguenther@suse.de>
16104 * gimple-predicate-analysis.h (predicate): Split out
16105 non-predicate related functionality into ..
16106 (uninit_analysis): .. this new class.
16107 * gimple-predicate-analysis.cc: Refactor into two classes.
16108 * tree-ssa-uninit.cc (find_uninit_use): Use uninit_analysis.
16110 2022-08-24 Richard Biener <rguenther@suse.de>
16112 * gimple-predicate-analysis.cc (predicate::use_cannot_happen):
16113 Do simple_control_dep_chain only up to cd_root, add the PHI
16114 operand edge to the chains like init_from_phi_def does.
16115 (predicate::is_use_guarded): Speedup early out, avoid half-way
16116 initializing the PHI def predicate.
16118 2022-08-24 Jakub Jelinek <jakub@redhat.com>
16121 * config/i386/sse.md (shuffletype): Add V32BF, V16BF and V8BF entries.
16122 Change V32HF, V16HF and V8HF entries from "f" to "i".
16123 (iptr): Add V32BF, V16BF, V8BF and BF entries.
16124 (i128vldq): Add V16HF and V16BF entries.
16125 (avx512er_vmrcp28<mode><mask_name><round_saeonly_name>): Fix typo,
16126 mask_opernad3 -> mask_operand3.
16128 2022-08-24 Martin Liska <mliska@suse.cz>
16129 Jørgen Kvalsvik <j@lambda.is>
16131 * gcov.cc (add_line_counts): Add group functions to coverage
16133 (accumulate_line_counts): Similarly for files.
16135 2022-08-24 Lulu Cheng <chenglulu@loongson.cn>
16137 * config/loongarch/genopts/loongarch-strings: Support code model medium.
16138 * config/loongarch/genopts/loongarch.opt.in: Likewise.
16139 * config/loongarch/loongarch-def.c: Likewise.
16140 * config/loongarch/loongarch-def.h (CMODEL_LARGE): Likewise.
16141 (CMODEL_EXTREME): Likewise.
16142 (N_CMODEL_TYPES): Likewise.
16143 (CMODEL_MEDIUM): Likewise.
16144 * config/loongarch/loongarch-opts.cc: Likewise.
16145 * config/loongarch/loongarch-opts.h (TARGET_CMODEL_MEDIUM): Likewise.
16146 * config/loongarch/loongarch-str.h (STR_CMODEL_MEDIUM): Likewise.
16147 * config/loongarch/loongarch.cc (loongarch_call_tls_get_addr):
16148 Tls symbol Loading support medium mode.
16149 (loongarch_legitimize_call_address): When medium mode, make a symbolic
16150 jump with two instructions.
16151 (loongarch_option_override_internal): Support medium.
16152 * config/loongarch/loongarch.md (@pcalau12i<mode>): New template.
16153 (@sibcall_internal_1<mode>): New function call templates added to support
16155 (@sibcall_value_internal_1<mode>): Likewise.
16156 (@sibcall_value_multiple_internal_1<mode>): Likewise.
16157 (@call_internal_1<mode>): Likewise.
16158 (@call_value_internal_1<mode>): Likewise.
16159 (@call_value_multiple_internal_1<mode>): Likewise.
16160 * config/loongarch/loongarch.opt: Support medium.
16161 * config/loongarch/predicates.md: Add processing about medium mode.
16162 * doc/invoke.texi: Document for '-mcmodel=medium'.
16164 2022-08-24 Richard Biener <rguenther@suse.de>
16166 * gimple-predicate-analysis.cc (predicate::use_cannot_happen):
16167 Start the compute_control_dep_chain walk from the immediate
16168 dominator of the PHI.
16170 2022-08-23 H.J. Lu <hjl.tools@gmail.com>
16173 * config/i386/amxtileintrin.h (_tile_loadd_internal): Cast to
16175 (_tile_stream_loadd_internal): Likewise.
16176 (_tile_stored_internal): Likewise.
16178 2022-08-23 Richard Biener <rguenther@suse.de>
16180 PR tree-optimization/106722
16181 * gimple-predicate-analysis.h (MAX_NUM_CHAINS, MAX_CHAIN_LEN,
16182 MAX_POSTDOM_CHECK, MAX_SWITCH_CASES): Move ...
16183 * gimple-predicate-analysis.cc: ... here and document.
16184 (simple_control_dep_chain): New function, factored from
16185 predicate::use_cannot_happen.
16186 (predicate::use_cannot_happen): Adjust.
16187 (predicate::predicate): Use simple_control_dep_chain as fallback.
16189 2022-08-23 Aldy Hernandez <aldyh@redhat.com>
16191 * range-op-float.cc (foperator_equal::op1_range): Set range to
16194 2022-08-23 Richard Biener <rguenther@suse.de>
16196 * gimple-predicate-analysis.cc (is_loop_exit): Split out
16198 (is_non_loop_exit_postdominating): ... here. Remove after
16200 (find_control_equiv_block): ... here.
16201 (compute_control_dep_chain): ... and here.
16202 (predicate::is_use_guarded): Do not excempt loop exits
16203 from short-cutting the case of the use post-dominating the
16206 2022-08-23 Andrew MacLeod <amacleod@redhat.com>
16208 PR tree-optimization/106687
16209 * range-op.cc (operator_minus::lhs_op1_relation): Return VREL_LE
16210 for the VREL_GT case as well.
16212 2022-08-22 Dimitar Dimitrov <dimitar@dinux.eu>
16214 * config/pru/pru.md (pru_<code>di3): New alternative for
16215 two operands but without earlyclobber.
16217 2022-08-22 Dimitar Dimitrov <dimitar@dinux.eu>
16219 * config/pru/pru.md (prumov<mode>, mov<mode>): Add
16220 variants for loading -1 consts.
16222 2022-08-22 Dimitar Dimitrov <dimitar@dinux.eu>
16225 * config/pru/constraints.md (Um): New constraint for -1.
16226 (Uf): New constraint for IOR fill-bytes constants.
16227 (Uz): New constraint for AND zero-bytes constants.
16228 * config/pru/predicates.md (const_fillbytes_operand): New
16229 predicate for IOR fill-bytes constants.
16230 (const_zerobytes_operand): New predicate for AND zero-bytes
16232 * config/pru/pru-protos.h (pru_output_sign_extend): Remove.
16233 (struct pru_byterange): New struct to describe a byte range.
16234 (pru_calc_byterange): New declaration.
16235 * config/pru/pru.cc (pru_rtx_costs): Add penalty for
16236 64-bit zero-extend.
16237 (pru_output_sign_extend): Remove.
16238 (pru_calc_byterange): New helper function to extract byte
16239 range info from a constant.
16240 (pru_print_operand): Remove 'y' and 'z' print modifiers.
16241 * config/pru/pru.md (zero_extendqidi2): New pattern.
16242 (zero_extendhidi2): New pattern.
16243 (zero_extendsidi2): New pattern.
16244 (extend<EQS0:mode><EQD:mode>2): Rewrite as an expand.
16245 (@pru_ior_fillbytes<mode>): New pattern.
16246 (@pru_and_zerobytes<mode>): New pattern.
16247 (<code>di3): Rewrite as an expand and handle ZERO and FILL
16249 (pru_<code>di3): New name for <code>di3.
16250 (@cbranch_qbbx_const_<BIT_TEST:code><HIDI:mode>): New pattern to
16251 handle bit-test for 64-bit registers.
16253 2022-08-22 Richard Biener <rguenther@suse.de>
16255 * gimple-predicate-analysis.h (predicate::m_use_expr): Remove.
16256 (predicate::def_expr): Likewise.
16257 (predicate::use_expr): Likewise.
16258 (predicate::expr): Likewise.
16259 * gimple-predicate-analysis.cc (predicate::def_expr): Remove.
16260 (predicate::use_expr): Likewise.
16261 (predicate::expr): Likewise.
16262 (predicate::is_use_guarded): Do not build m_use_expr.
16264 2022-08-22 Martin Liska <mliska@suse.cz>
16267 * configure.ac: Detect O_NONBLOCK flag for open.
16268 * config.in: Regenerate.
16269 * configure: Regenerate.
16270 * opts-common.cc (jobserver_info::connect): Set is_connected
16271 properly based on O_NONBLOCK.
16272 * opts-jobserver.h (struct jobserver_info): Add is_connected
16275 2022-08-22 zhongjuzhe <juzhe.zhong@rivai.ai>
16277 * simplify-rtx.cc (test_vector_subregs_fore_back): Make first value
16278 and repeat value different.
16280 2022-08-22 Tobias Burnus <tobias@codesourcery.com>
16283 * lto-wrapper.cc (free_array_of_ptrs): Move before tool_cleanup.
16284 (tool_cleanup): Unlink offload_names.
16285 (compile_offload_image): Take filename argument to set it early.
16286 (compile_images_for_offload_targets): Update call; set
16287 offload_names to NULL after freeing the array.
16289 2022-08-22 Richard Biener <rguenther@suse.de>
16291 PR tree-optimization/105937
16292 * tree-ssa-uninit.cc (find_uninit_use): Do not queue PHIs
16294 (execute_late_warn_uninitialized): Mark backedges.
16296 2022-08-22 Richard Biener <rguenther@suse.de>
16298 * gimple-predicate-analysis.cc (predicate::use_cannot_happen):
16299 If the use is guarded with multiple predicate paths compute
16300 the predicates intersection before going forward. When
16301 compute_control_dep_chain wasn't able to come up with at
16302 least one path from function entry to the PHI edge compute
16303 a conservative sparse path instead.
16305 2022-08-20 Lulu Cheng <chenglulu@loongson.cn>
16307 * config/loongarch/loongarch-opts.cc: Allow cmodel to be extreme.
16308 * config/loongarch/loongarch.cc (loongarch_call_tls_get_addr):
16309 Add extreme support for TLS GD and LD types.
16310 (loongarch_legitimize_tls_address): Add extreme support for TLS LE
16312 (loongarch_split_symbol): When compiling with -mcmodel=extreme,
16313 the symbol address will be obtained through five instructions.
16314 (loongarch_print_operand_reloc): Add support.
16315 (loongarch_print_operand): Add support.
16316 (loongarch_print_operand_address): Add support.
16317 (loongarch_option_override_internal): Set '-mcmodel=extreme' option
16318 incompatible with '-mno-explicit-relocs'.
16319 * config/loongarch/loongarch.md (@lui_l_hi20<mode>):
16320 Loads bits 12-31 of data into registers.
16321 (lui_h_lo20): Load bits 32-51 of the data and spell bits 0-31 of
16322 the source register.
16323 (lui_h_hi12): Load bits 52-63 of the data and spell bits 0-51 of
16324 the source register.
16325 * config/loongarch/predicates.md: Symbols need to be decomposed
16326 when defining the macro TARGET_CMODEL_EXTREME
16327 * doc/invoke.texi: Modify the description information of cmodel in the document.
16328 Document -W[no-]extreme-plt.
16330 2022-08-19 Tobias Burnus <tobias@codesourcery.com>
16332 * config/gcn/mkoffload.cc (main): Add omp_requires_file and dbgobj to
16334 * config/i386/intelmic-mkoffload.cc (prepare_target_image): Add
16335 omp_requires_file to temp_files.
16336 * config/nvptx/mkoffload.cc (omp_requires_file): New global static var.
16337 (main): Remove local omp_requires_file var.
16338 (tool_cleanup): Handle omp_requires_file.
16340 2022-08-19 Aldy Hernandez <aldyh@redhat.com>
16342 * gimple-range-path.cc (path_range_query::path_range_query):
16343 Remove constructor that takes edge.
16344 * gimple-range-path.h (class path_range_query): Same.
16345 * tree-ssa-loop-ch.cc (edge_range_query): New.
16346 (entry_loop_condition_is_static): Call edge_range_query.
16348 2022-08-18 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
16350 * config/xtensa/xtensa.h
16351 (enum reg_class, REG_CLASS_NAMES, REG_CLASS_CONTENTS):
16352 Add new register class "ISC_REGS".
16353 * config/xtensa/constraints.md (c): Add new register constraint.
16354 * config/xtensa/xtensa.md (define_constants): Remove "A11_REG".
16355 (sibcall_internal, sibcall_value_internal):
16356 Change to use the new register constraint, and remove two split
16357 patterns for fixups that are no longer needed.
16359 2022-08-18 Maciej W. Rozycki <macro@embecosm.com>
16361 * config/riscv/riscv.md (*mov<GPR:mode><X:mode>cc): Fix output
16362 pattern formatting.
16364 2022-08-18 Tim Lange <mail@tim-lange.me>
16367 * doc/invoke.texi: Add Wanalyzer-imprecise-fp-arithmetic.
16369 2022-08-18 Aldy Hernandez <aldyh@redhat.com>
16371 * gimple-range-path.cc (path_range_query::path_range_query): Add
16372 various constructors to take a path.
16373 (path_range_query::~path_range_query): Remove m_alloced_ranger.
16374 (path_range_query::range_on_path_entry): Adjust for m_ranger being
16376 (path_range_query::set_path): Rename to...
16377 (path_range_query::reset_path): ...this and call compute_ranges.
16378 (path_range_query::ssa_range_in_phi): Adjust for m_ranger
16380 (path_range_query::range_defined_in_block): Same.
16381 (path_range_query::compute_ranges_in_block): Same.
16382 (path_range_query::adjust_for_non_null_uses): Same.
16383 (path_range_query::compute_exit_dependencies): Use m_path instead
16385 (path_range_query::compute_ranges): Remove path argument.
16386 (path_range_query::range_of_stmt): Adjust for m_ranger reference.
16387 (path_range_query::compute_outgoing_relations): Same.
16388 * gimple-range-path.h (class path_range_query): Add various
16390 Make compute_ranges and compute_exit_dependencies private.
16391 Rename set_path to reset_path.
16392 Make m_ranger a reference.
16393 Remove m_alloced_ranger.
16394 * tree-ssa-dom.cc (pass_dominator::execute): Adjust constructor to
16396 * tree-ssa-loop-ch.cc (entry_loop_condition_is_static): Take a
16397 ranger and instantiate a new path_range_query every time.
16398 (ch_base::copy_headers): Pass ranger instead of path_range_query.
16399 * tree-ssa-threadbackward.cc (class back_threader): Remove m_solver.
16400 (back_threader::~back_threader): Remove m_solver.
16401 (back_threader::find_taken_edge_switch): Adjust for m_ranger
16403 (back_threader::find_taken_edge_cond): Same.
16404 (back_threader::dump): Remove m_solver.
16405 (back_threader::back_threader): Move verify_marked_backedges
16406 here from the path_range_query constructor.
16407 * tree-ssa-threadedge.cc (hybrid_jt_simplifier::simplify): Move
16408 some code from compute_ranges_from_state here.
16409 (hybrid_jt_simplifier::compute_ranges_from_state): Rename...
16410 (hybrid_jt_simplifier::compute_exit_dependencies): ...to this.
16411 * tree-ssa-threadedge.h (class hybrid_jt_simplifier): Rename
16412 compute_ranges_from_state to compute_exit_dependencies.
16415 2022-08-18 Richard Biener <rguenther@suse.de>
16417 PR middle-end/106617
16418 * match.pd ((a ? b : c) > d -> a ? (b > d) : (c > d)): Fix
16419 guard, disable on GENERIC to not cause quadratic behavior
16420 with the fold-const.cc implementation and the use of !
16422 2022-08-18 Andrew Pinski <apinski@marvell.com>
16424 PR gcov-profile/106659
16425 * gcov-dump.cc (INCLUDE_VECTOR): Include vector.h with
16428 2022-08-18 konglin1 <lingling.kong@intel.com>
16430 * config/i386/i386-expand.cc (ix86_expand_sse_movcc): Handle vector
16432 (ix86_expand_vector_init_duplicate): Support vector BFmode.
16433 (ix86_expand_vector_init_one_nonzero): Ditto.
16434 (ix86_expand_vector_init_one_var): Ditto.
16435 (ix86_expand_vector_init_concat): Ditto.
16436 (ix86_expand_vector_init_interleave): Ditto.
16437 (ix86_expand_vector_init_general): Ditto.
16438 (ix86_expand_vector_init): Ditto.
16439 (ix86_expand_vector_set_var): Ditto.
16440 (ix86_expand_vector_set): Ditto.
16441 (ix86_expand_vector_extract): Ditto.
16442 * config/i386/i386.cc (classify_argument): Add BF vector modes.
16443 (function_arg_64): Ditto.
16444 (ix86_gimplify_va_arg): Ditto.
16445 (ix86_get_ssemov): Ditto.
16446 * config/i386/i386.h (VALID_AVX256_REG_MODE): Add BF vector modes.
16447 (VALID_AVX512F_REG_MODE): Ditto.
16448 (host_detect_local_cpu): Ditto.
16449 (VALID_SSE2_REG_MODE): Ditto.
16450 * config/i386/i386.md: Add BF vector modes.
16451 (MODE_SIZE): Ditto.
16452 (ssemodesuffix): Add bf suffix for BF vector modes.
16453 (ssevecmode): Ditto.
16454 * config/i386/sse.md (VMOVE): Adjust for BF vector modes.
16455 (VI12HFBF_AVX512VL): Ditto.
16456 (V_256_512): Ditto.
16457 (VF_AVX512HFBF16): Ditto.
16458 (VF_AVX512BWHFBF16): Ditto.
16461 (VIHFBF_256): Ditto.
16462 (VIHFBF_AVX512BW): Ditto.
16463 (VI2F_256_512):Ditto.
16467 (sseinsnmode): Ditto.
16468 (sseconstm1): Ditto.
16469 (sseintmodesuffix): New mode_attr.
16470 (avx512fmaskmode): Ditto.
16471 (avx512fmaskmodelower): Ditto.
16472 (ssedoublevecmode): Ditto.
16473 (ssehalfvecmode): Ditto.
16474 (ssehalfvecmodelower): Ditto.
16475 (ssescalarmode): Add vector BFmode mapping.
16476 (ssescalarmodelower): Ditto.
16477 (ssexmmmode): Ditto.
16478 (ternlogsuffix): Ditto.
16479 (ssescalarsize): Ditto.
16480 (sseintprefix): Ditto.
16483 (bcstscalarsuff): Ditto.
16484 (<avx512>_blendm<mode>): New define_insn for BFmode.
16485 (<avx512>_store<mode>_mask): Ditto.
16486 (vcond_mask_<mode><avx512fmaskmodelower>): Ditto.
16487 (vec_set<mode>_0): New define_insn for BF vector set.
16488 (V8BFH_128): New mode_iterator for BFmode.
16489 (avx512fp16_mov<mode>): Ditto.
16490 (vec_set<mode>): New define_insn for BF vector set.
16491 (@vec_extract_hi_<mode>): Ditto.
16492 (@vec_extract_lo_<mode>): Ditto.
16493 (vec_set_hi_<mode>): Ditto.
16494 (vec_set_lo_<mode>): Ditto.
16495 (*vec_extract<mode>_0): New define_insn_and_split for BF
16497 (*vec_extract<mode>): New define_insn.
16498 (VEC_EXTRACT_MODE): Add BF vector modes.
16499 (PINSR_MODE): Add V8BF.
16501 (pinsr_evex_isa): Ditto.
16502 (<sse2p4_1>_pinsr<ssemodesuffix>): Adjust to support
16503 insert for V8BFmode.
16504 (pbroadcast_evex_isa): Add BF vector modes.
16505 (AVX2_VEC_DUP_MODE): Ditto.
16506 (VEC_INIT_MODE): Ditto.
16507 (VEC_INIT_HALF_MODE): Ditto.
16508 (avx2_pbroadcast<mode>): Adjust to support BF vector mode
16510 (avx2_pbroadcast<mode>_1): Ditto.
16511 (<avx512>_vec_dup<mode>_1): Ditto.
16512 (<mask_codefor><avx512>_vec_dup_gpr<mode><mask_name>):
16515 2022-08-18 Martin Liska <mliska@suse.cz>
16517 * configure: Regenerate.
16519 2022-08-18 Haochen Gui <guihaoc@gcc.gnu.org>
16522 * config/rs6000/rs6000.md (<u>maddditi4): New pattern for multiply-add.
16523 (<u>madddi4_highpart): New.
16524 (<u>madddi4_highpart_le): New.
16526 2022-08-18 Aldy Hernandez <aldyh@redhat.com>
16528 * gimple-range-path.cc
16529 (path_range_query::compute_exit_dependencies): Use
16530 gimple_range_ssa_names.
16532 2022-08-18 zhongjuzhe <juzhe.zhong@rivai.ai>
16534 * config/riscv/predicates.md: Adjust runtime invariant.
16535 * config/riscv/riscv-modes.def (MAX_BITSIZE_MODE_ANY_MODE): New.
16536 (NUM_POLY_INT_COEFFS): New.
16537 * config/riscv/riscv-protos.h (riscv_initial_elimination_offset):Adjust
16539 * config/riscv/riscv-sr.cc (riscv_remove_unneeded_save_restore_calls):
16540 Adjust runtime invariant.
16541 * config/riscv/riscv.cc (struct riscv_frame_info): Adjust runtime
16543 (enum riscv_microarchitecture_type): Ditto.
16544 (riscv_valid_offset_p): Ditto.
16545 (riscv_valid_lo_sum_p): Ditto.
16546 (riscv_address_insns): Ditto.
16547 (riscv_load_store_insns): Ditto.
16548 (riscv_legitimize_move): Ditto.
16549 (riscv_binary_cost): Ditto.
16550 (riscv_rtx_costs): Ditto.
16551 (riscv_output_move): Ditto.
16552 (riscv_extend_comparands): Ditto.
16553 (riscv_flatten_aggregate_field): Ditto.
16554 (riscv_get_arg_info): Ditto.
16555 (riscv_pass_by_reference): Ditto.
16556 (riscv_elf_select_rtx_section): Ditto.
16557 (riscv_stack_align): Ditto.
16558 (riscv_compute_frame_info): Ditto.
16559 (riscv_initial_elimination_offset): Ditto.
16560 (riscv_set_return_address): Ditto.
16561 (riscv_for_each_saved_reg): Ditto.
16562 (riscv_first_stack_step): Ditto.
16563 (riscv_expand_prologue): Ditto.
16564 (riscv_expand_epilogue): Ditto.
16565 (riscv_can_use_return_insn): Ditto.
16566 (riscv_secondary_memory_needed): Ditto.
16567 (riscv_hard_regno_nregs): Ditto.
16568 (riscv_convert_vector_bits): New.
16569 (riscv_option_override): Adjust runtime invariant.
16570 (riscv_promote_function_mode): Ditto.
16571 * config/riscv/riscv.h (POLY_SMALL_OPERAND_P): New.
16572 (BITS_PER_RISCV_VECTOR): New.
16573 (BYTES_PER_RISCV_VECTOR): New.
16574 * config/riscv/riscv.md: Adjust runtime invariant.
16576 2022-08-18 Lulu Cheng <chenglulu@loongson.cn>
16578 * config/loongarch/loongarch.cc (loongarch_call_tls_get_addr):
16579 Get __tls_get_addr address through got table when disable plt.
16581 2022-08-18 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
16583 * config/xtensa/xtensa.cc (xtensa_expand_prologue):
16584 Use an "addmi" machine instruction for updating the stack pointer
16585 rather than addition/subtraction via hard register A9, if the amount
16586 of change satisfies the literal value conditions of that instruction
16587 when the CALL0 ABI is used.
16588 (xtensa_expand_epilogue): Ditto.
16589 And also inhibit the stack pointer addition of constant zero.
16591 2022-08-17 Roger Sayle <roger@nextmovesoftware.com>
16594 * config/i386/i386-features.cc
16595 (timde_scalar_chain::compute_convert_gain): Replace incorrect use
16596 of XINT with INTVAL (XEXP (src, 1)).
16598 2022-08-17 Aldy Hernandez <aldyh@redhat.com>
16600 * gimple-range-path.cc
16601 (path_range_query::compute_ranges_in_block): Remove
16602 set_root_oracle call.
16603 (path_range_query::compute_ranges): Pass ranger oracle to
16605 * value-relation.cc (path_oracle::reset_path): Set root oracle.
16606 * value-relation.h (path_oracle::reset_path): Add root oracle
16609 2022-08-17 Marek Polacek <polacek@redhat.com>
16612 * diagnostic-spec.cc (nowarn_spec_t::nowarn_spec_t): Handle
16613 OPT_Wpessimizing_move and OPT_Wredundant_move.
16614 * diagnostic-spec.h (nowarn_spec_t): Add NW_REDUNDANT enumerator.
16616 2022-08-17 Francois-Xavier Coudert <fxcoudert@gcc.gnu.org>
16617 Jakub Jelinek <jakub@redhat.com>
16620 * common.opt (static-libquadmath): New option.
16621 * gcc.cc (driver_handle_option): Always accept -static-libquadmath.
16622 * config/darwin.h (LINK_SPEC): Handle -static-libquadmath.
16624 2022-08-17 Tobias Burnus <tobias@codesourcery.com>
16626 * lto-cgraph.cc (input_offload_tables): Improve requires diagnostic
16627 when filenames come out identically.
16629 2022-08-17 Tobias Burnus <tobias@codesourcery.com>
16631 PR middle-end/106548
16632 * omp-low.cc (lower_rec_input_clauses): Use build_outer_var_ref
16633 for 'simd' linear-step values that are variable.
16635 2022-08-17 Tobias Burnus <tobias@codesourcery.com>
16636 Chung-Lin Tang <cltang@codesourcery.com>
16639 * gimplify.cc (omp_notice_variable): Call omp_mappable_type
16640 instead of removed langhook.
16641 * omp-general.h (omp_mappable_type): New prototype.
16642 * omp-general.cc (omp_mappable_type): New; moved from ...
16643 * langhooks.cc (lhd_omp_mappable_type): ... here.
16644 * langhooks-def.h (lhd_omp_mappable_type,
16645 LANG_HOOKS_OMP_MAPPABLE_TYPE): Remove.
16646 (LANG_HOOKS_FOR_TYPES_INITIALIZER): Remote the latter.
16647 * langhooks.h (struct lang_hooks_for_types): Remove
16650 2022-08-17 Christophe Lyon <christophe.lyon@arm.com>
16652 * config.gcc (arm): Define with_float to hard if target name ends
16655 2022-08-17 Richard Biener <rguenther@suse.de>
16657 * tree-ssa-threadbackward.cc
16658 (back_threader_profitability): Split profitable_path_p
16659 into possibly_profitable_path_p and itself, keep state
16661 (back_threader::m_profit): Remove.
16662 (back_threader::find_paths): Likewise.
16663 (back_threader::maybe_register_path): Take profitability
16664 instance as parameter.
16665 (back_threader::find_paths_to_names): Likewise. Use
16666 possibly_profitable_path_p and avoid the path range query
16667 when the path is currently too large.
16668 (back_threader::find_paths): Fold into ...
16669 (back_threader::maybe_thread_block): ... this.
16670 (get_gimple_control_stmt): Remove.
16671 (back_threader_profitability::possibly_profitable_path_p):
16672 Split out from profitable_path_p, do early profitability
16674 (back_threader_profitability::profitable_path_p): Do final
16675 profitability path after the taken edge has been determined.
16677 2022-08-17 Xi Ruoyao <xry111@xry111.site>
16679 * config/loongarch/loongarch.md (fmax<mode>3): New RTL pattern.
16680 (fmin<mode>3): Likewise.
16682 2022-08-17 Andrew MacLeod <amacleod@redhat.com>
16684 * gimple-range-fold.cc (gimple_range_ssa_names): New.
16685 * gimple-range-fold.h (gimple_range_ssa_names): New prototype.
16686 * gimple-range-gori.cc (range_def_chain::get_def_chain): Move
16687 code to new routine.
16689 2022-08-16 Martin Liska <mliska@suse.cz>
16692 * doc/install.texi: Remove link to www.bullfreeware.com
16694 2022-08-16 Kito Cheng <kito.cheng@sifive.com>
16696 * common/config/riscv/riscv-common.cc (riscv_implied_info): Add
16698 (riscv_ext_version_table): Ditto.
16699 (riscv_ext_flag_table): Ditto.
16700 * config/riscv/riscv-opts.h (MASK_ZFHMIN): New.
16702 (TARGET_ZFHMIN): Ditto.
16703 (TARGET_ZFH): Ditto.
16704 * config/riscv/riscv.cc (riscv_output_move): Handle HFmode move
16705 for zfh and zfhmin.
16706 (riscv_emit_float_compare): Handle HFmode.
16707 * config/riscv/riscv.md (ANYF): Add HF.
16712 (truncdfhf2): Ditto.
16713 (extendhfsf2): Ditto.
16714 (extendhfdf2): Ditto.
16715 (*movhf_hardfloat): Ditto.
16716 (*movhf_softfloat): Make sure not ZFHMIN.
16717 * config/riscv/riscv.opt (riscv_zf_subext): New.
16719 2022-08-16 Kito Cheng <kito.cheng@sifive.com>
16721 * config/riscv/riscv-builtins.cc: include stringpool.h
16722 (riscv_float16_type_node): New.
16723 (riscv_init_builtin_types): Ditto.
16724 (riscv_init_builtins): Call riscv_init_builtin_types.
16725 * config/riscv/riscv-modes.def (HF): New.
16726 * config/riscv/riscv.cc (riscv_output_move): Handle HFmode.
16727 (riscv_mangle_type): New.
16728 (riscv_scalar_mode_supported_p): Ditto.
16729 (riscv_libgcc_floating_mode_supported_p): Ditto.
16730 (riscv_excess_precision): Ditto.
16731 (riscv_floatn_mode): Ditto.
16732 (riscv_init_libfuncs): Ditto.
16733 (TARGET_MANGLE_TYPE): Ditto.
16734 (TARGET_SCALAR_MODE_SUPPORTED_P): Ditto.
16735 (TARGET_LIBGCC_FLOATING_MODE_SUPPORTED_P): Ditto.
16736 (TARGET_INIT_LIBFUNCS): Ditto.
16737 (TARGET_C_EXCESS_PRECISION): Ditto.
16738 (TARGET_FLOATN_MODE): Ditto.
16739 * config/riscv/riscv.md (mode): Add HF.
16740 (softload): Add HF.
16741 (softstore): Ditto.
16745 (*movhf_softfloat): New.
16747 2022-08-16 Richard Biener <rguenther@suse.de>
16749 * tree-ssa-threadbackward.cc (back_threader::find_paths_to_names):
16750 Do not walk further if we are leaving the current loop.
16752 2022-08-16 Sergei Trofimovich <siarheit@google.com>
16755 * gcc.cc (driver::detect_jobserver): Allocate storage xputenv()
16756 argument using xstrdup().
16758 2022-08-16 Aldy Hernandez <aldyh@redhat.com>
16760 * gimple-range-path.cc (path_range_query::import_p): Rename to...
16761 (path_range_query::exit_dependency_p): ...this.
16762 (path_range_query::dump): Rename imports to exit dependencies.
16763 (path_range_query::compute_ranges_in_phis): Same.
16764 (path_range_query::compute_ranges_in_block): Same.
16765 (path_range_query::adjust_for_non_null_uses): Same.
16766 (path_range_query::compute_ranges): Same.
16767 (path_range_query::compute_phi_relations): Same.
16768 (path_range_query::add_to_imports): Rename to...
16769 (path_range_query::add_to_exit_dependencies): ...this.
16770 (path_range_query::compute_imports): Rename to...
16771 (path_range_query::compute_exit_dependencies): ...this.
16772 * gimple-range-path.h (class path_range_query): Rename imports to
16775 2022-08-16 Martin Liska <mliska@suse.cz>
16777 * value-range-storage.h (class obstack_vrange_allocator): Mark
16778 the class as final.
16779 (class ggc_vrange_allocator): Likewise.
16781 2022-08-16 Martin Liska <mliska@suse.cz>
16783 * value-range-equiv.h (class value_range_equiv): Add virtual
16785 * value-range.h: Likewise.
16787 2022-08-16 Richard Biener <rguenther@suse.de>
16789 PR middle-end/106630
16790 * match.pd ((T)(x * CST) -> (T)x * CST): Restrict to
16791 narrowing conversions.
16793 2022-08-16 Martin Liska <mliska@suse.cz>
16795 * value-range-equiv.h (class value_range_equiv):
16797 2022-08-16 Martin Liska <mliska@suse.cz>
16799 * config/i386/i386-features.h (class general_scalar_chain): Add
16800 final override for a method.
16801 (class timode_scalar_chain): Likewise.
16803 2022-08-16 Richard Biener <rguenther@suse.de>
16805 * doc/invoke.texi (max-jump-thread-paths): Adjust.
16807 2022-08-16 Martin Liska <mliska@suse.cz>
16809 * opts-common.cc (jobserver_info::connect): Open fifo
16810 in non-blocking mode.
16812 2022-08-16 Kewen.Lin <linkw@gcc.gnu.org>
16815 * config/rs6000/mma.md (define_expand movoo): Move TARGET_MMA condition
16816 check to preparation statements and add handlings for !TARGET_MMA.
16817 (define_expand movxo): Likewise.
16819 2022-08-16 Kewen Lin <linkw@linux.ibm.com>
16821 PR tree-optimization/106322
16822 * tree-vect-stmts.cc (vectorizable_call): Don't allow
16823 vect_emulated_vector_p type for both vectype_in and vectype_out.
16825 2022-08-16 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
16827 * common/config/xtensa/xtensa-common.cc
16828 (xtensa_option_optimization_table): Add OPT_fsplit_wide_types_early
16829 for OPT_LEVELS_ALL in order to restore pre-GCC10 behavior.
16831 2022-08-15 Andrew MacLeod <amacleod@redhat.com>
16833 PR tree-optimization/106621
16834 * value-range.cc (irange::set): Check for POLY_INT_CST early.
16836 2022-08-15 Roger Sayle <roger@nextmovesoftware.com>
16838 * config/i386/i386-features.cc
16839 (timode_scalar_chain::compute_convert_gain): Provide costs for
16840 shifts and rotates.
16841 (timode_scalar_chain::convert_insn): Handle ASHIFTRT, ROTATERT
16842 and ROTATE just like existing ASHIFT and LSHIFTRT cases.
16843 (timode_scalar_to_vector_candidate_p): Handle all shifts and
16844 rotates by integer constants between 0 and 127.
16846 2022-08-15 Roger Sayle <roger@nextmovesoftware.com>
16848 * config/i386/i386-features.cc
16849 (timode_scalar_chain::compute_convert_gain): Provide gains for
16850 comparisons against 0/-1, including "*testti" patterns.
16852 2022-08-15 Roger Sayle <roger@nextmovesoftware.com>
16854 PR tree-optimization/64992
16855 PR tree-optimization/98956
16856 * match.pd (ne (lshift @0 @1) 0): Simplify (X << C) != 0 to X
16857 when X is zero_one_valued_p and the shift constant C is valid.
16858 (eq (lshift @0 @1) 0): Likewise, simplify (X << C) == 0 to !X
16859 when X is zero_one_valued_p and the shift constant C is valid.
16861 2022-08-15 Roger Sayle <roger@nextmovesoftware.com>
16862 Richard Biener <rguenther@suse.de>
16864 PR tree-optimization/71343
16865 * match.pd (op (lshift @0 @1) (lshift @2 @1)): Optimize the
16866 expression (X<<C) + (Y<<C) to (X+Y)<<C for multiple operators.
16867 (op (rshift @0 @1) (rshift @2 @1)): Likewise, simplify (X>>C)^(Y>>C)
16868 to (X^Y)>>C for binary logical operators, AND, IOR and XOR.
16870 2022-08-15 Richard Biener <rguenther@suse.de>
16872 * gimple-range-path.cc (range_on_path_entry): Just
16873 call range_on_entry.
16875 2022-08-15 Jakub Jelinek <jakub@redhat.com>
16877 PR rtl-optimization/106590
16878 * ifcvt.cc (check_for_cc_cmp_clobbers): New function.
16879 (noce_convert_multiple_sets_1): If SEQ sets or clobbers any regs
16880 mentioned in cc_cmp or rev_cc_cmp, don't consider seq2 for any
16881 further conditional moves.
16883 2022-08-15 konglin1 <lingling.kong@intel.com>
16885 * config/i386/i386-builtin-types.def (BFLOAT16): New primitive type.
16886 * config/i386/i386-builtins.cc : Support __bf16 type for i386 backend.
16887 (ix86_register_bf16_builtin_type): New function.
16888 (ix86_bf16_type_node): New.
16889 (ix86_bf16_ptr_type_node): Ditto.
16890 (ix86_init_builtin_types): Add ix86_register_bf16_builtin_type function call.
16891 * config/i386/i386-modes.def (FLOAT_MODE): Add BFmode.
16892 (ADJUST_FLOAT_FORMAT): Ditto.
16893 * config/i386/i386.cc (classify_argument): Handle BFmode.
16894 (construct_container): Ditto.
16895 (function_value_32): Return __bf16 by %xmm0.
16896 (function_value_64): Return __bf16 by SSE register.
16897 (ix86_output_ssemov): Handle BFmode.
16898 (ix86_legitimate_constant_p): Disable BFmode constant double.
16899 (ix86_secondary_reload): Require gpr as intermediate register
16900 to store __bf16 from sse register when sse4 is not available.
16901 (ix86_scalar_mode_supported_p): Enable __bf16 under sse2.
16902 (ix86_mangle_type): Add manlging for __bf16 type.
16903 (ix86_invalid_conversion): New function for target hook.
16904 (ix86_invalid_unary_op): Ditto.
16905 (ix86_invalid_binary_op): Ditto.
16906 (TARGET_INVALID_CONVERSION): New define for target hook.
16907 (TARGET_INVALID_UNARY_OP): Ditto.
16908 (TARGET_INVALID_BINARY_OP): Ditto.
16909 * config/i386/i386.h (host_detect_local_cpu): Add BFmode.
16910 * config/i386/i386.md ("mode"): Add BFmode.
16911 (MODE_SIZE): Ditto.
16912 (X87MODEFH): Ditto.
16913 (HFBF): Add new define_mode_iterator.
16914 (*pushhf_rex64): Change for BFmode.
16915 (*push<mode>_rex64): Ditto.
16917 (*push<mode>): Ditto.
16919 (hfbfconstf): Add new define_mode_attr.
16920 (*mov<mode>_internal): Add BFmode.
16922 2022-08-13 Roger Sayle <roger@nextmovesoftware.com>
16923 Uroš Bizjak <ubizjak@gmail.com>
16925 * config/i386/predicates.md (const_0_to_255_not_mul_8_operand):
16926 New predicate for values between 0/1 and 255, not multiples of 8.
16927 * config/i386/sse.md (ashlv1ti3): Delay lowering of logical left
16928 shifts by constant bit counts.
16929 (*ashlvti3_internal): New define_insn_and_split that lowers
16930 logical left shifts by constant bit counts, that aren't multiples
16931 of 8, before reload.
16932 (lshrv1ti3): Delay lowering of logical right shifts by constant.
16933 (*lshrv1ti3_internal): New define_insn_and_split that lowers
16934 logical right shifts by constant bit counts, that aren't multiples
16935 of 8, before reload.
16936 (ashrv1ti3):: Delay lowering of arithmetic right shifts by
16937 constant bit counts.
16938 (*ashrv1ti3_internal): New define_insn_and_split that lowers
16939 arithmetic right shifts by constant bit counts before reload.
16940 (rotlv1ti3): Delay lowering of rotate left by constant.
16941 (*rotlv1ti3_internal): New define_insn_and_split that lowers
16942 rotate left by constant bits counts before reload.
16943 (rotrv1ti3): Delay lowering of rotate right by constant.
16944 (*rotrv1ti3_internal): New define_insn_and_split that lowers
16945 rotate right by constant bits counts before reload.
16947 2022-08-12 Srinath Parvathaneni <srinath.parvathaneni@arm.com>
16949 * doc/invoke.texi (Arm Options): Document -mcpu=cortex-m55 options.
16951 2022-08-12 Jan Hubicka <hubicka@ucw.cz>
16953 PR middle-end/106057
16954 * ipa-devirt.cc (type_or_derived_type_possibly_instantiated_p): New
16956 (possible_polymorphic_call_targets): Use it.
16958 2022-08-12 Andrew Carlotti <andrew.carlotti@arm.com>
16960 * tree-ssa-loop.h: Improve comment
16962 2022-08-12 Jakub Jelinek <jakub@redhat.com>
16964 PR tree-optimization/106506
16965 * tree-ssa-phiopt.cc (spaceship_replacement): Don't punt for
16966 is_cast or orig_use_lhs cases if phi_bb has 3 predecessors.
16968 2022-08-12 Richard Biener <rguenther@suse.de>
16970 PR tree-optimization/106593
16971 * tree-ssa-threadbackward.cc (back_threader::find_paths):
16972 If the imports from the conditional do not satisfy
16973 gimple_range_ssa_p don't try to thread anything.
16975 2022-08-12 Tamar Christina <tamar.christina@arm.com>
16978 * config/aarch64/aarch64-sve.md (*fcmuo<mode>_nor_combine,
16979 *fcmuo<mode>_bic_combine): Don't accept comparisons against zero.
16981 2022-08-12 Tim Lange <mail@tim-lange.me>
16984 * doc/invoke.texi: Add Wanalyzer-out-of-bounds.
16986 2022-08-12 Andrew Pinski <apinski@marvell.com>
16988 * config/aarch64/aarch64.md: Remove comment
16989 about MD_INCLUDES as it is out of date and not needed.
16991 2022-08-11 Richard Biener <rguenther@suse.de>
16993 * gimple-range-path.cc (path_range_query::compute_imports):
16994 Restrict walking SSA defs to blocks inside the path. Track
16995 the same operands as range_def_chain::get_def_chain does.
16997 2022-08-11 Richard Biener <rguenther@suse.de>
16999 PR tree-optimization/106514
17000 * tree-ssa-threadbackward.cc (back_threader::find_paths_to_names):
17001 Compute and unwind both m_imports and interesting on the fly during
17003 (back_threader::find_paths): Compute the original m_imports
17004 from just the SSA uses of the exit conditional. Drop
17005 handling single_succ_to_potentially_threadable_block.
17006 * gimple-range-path.cc (path_range_query::ssa_range_in_phi): Handle
17007 constant PHI arguments without crashing. Use PHI_ARG_DEF_FROM_EDGE.
17009 2022-08-11 Richard Biener <rguenther@suse.de>
17011 * gimple-range-path.h (path_range_query::compute_imports):
17012 Take path as argument, not the exit block.
17013 * gimple-range-path.cc (path_range_query::compute_imports):
17014 Likewise, and adjust, avoiding possibly stale m_path.
17015 (path_range_query::compute_outgoing_relations): Register
17016 relations for all conditionals.
17017 * tree-ssa-threadbackward.cc (back_threader::find_paths):
17020 2022-08-11 Kewen Lin <linkw@linux.ibm.com>
17022 * config/rs6000/rs6000-builtin.cc (rs6000_init_builtins): Fix the
17023 oversight on ENB_CELL by simplifying with rs6000_builtin_is_supported.
17024 (rs6000_expand_builtin): Simplify with rs6000_builtin_is_supported.
17026 2022-08-11 Kewen Lin <linkw@linux.ibm.com>
17028 * config/rs6000/rs6000-internal.h (rs6000_global_entry_point_needed_p):
17029 Remove function declaration.
17031 2022-08-10 Richard Biener <rguenther@suse.de>
17033 PR tree-optimization/106513
17034 * gimple-ssa-store-merging.cc (do_shift_rotate): Use uint64_t
17037 2022-08-10 Martin Liska <mliska@suse.cz>
17040 * opts-jobserver.h (struct jobserver_info): Add pipefd.
17041 (jobserver_info::connect): New.
17042 (jobserver_info::disconnect): Likewise.
17043 (jobserver_info::get_token): Likewise.
17044 (jobserver_info::return_token): Likewise.
17045 * opts-common.cc: Implement the new functions.
17047 2022-08-10 Martin Liska <mliska@suse.cz>
17049 * opts-jobserver.h: Add one member.
17050 * opts-common.cc (jobserver_info::jobserver_info): Parse FIFO
17051 format of --jobserver-auth.
17053 2022-08-10 Martin Liska <mliska@suse.cz>
17055 * gcc.cc (driver::detect_jobserver): Remove and move to
17057 * lto-wrapper.cc (jobserver_active_p): Likewise.
17058 (run_gcc): Likewise.
17059 * opts-jobserver.h: New file.
17060 * opts-common.cc (jobserver_info::jobserver_info): New function.
17062 2022-08-09 Roger Sayle <roger@nextmovesoftware.com>
17064 * config/i386/i386-features.cc (scalar_chain::convert_compare):
17065 Create new pseudos only when/if needed. Add support for TEST,
17066 i.e. (COMPARE (AND x y) (const_int 0)), using UNSPEC_PTEST.
17067 When broadcasting V2DImode and V4SImode use new pseudo register.
17068 (timode_scalar_chain::convert_op): Do nothing if operand is
17069 already V1TImode. Avoid generating useless SUBREG conversions,
17070 i.e. (SUBREG:V1TImode (REG:V1TImode) 0). Handle CONST_WIDE_INT
17071 in addition to CONST_INT by using CONST_SCALAR_INT_P.
17072 (convertible_comparison_p): Use CONST_SCALAR_INT_P to match both
17073 CONST_WIDE_INT and CONST_INT. Recognize new *testti_doubleword
17074 pattern as an STV candidate.
17075 (timode_scalar_to_vector_candidate_p): Allow CONST_SCALAR_INT_P
17076 operands in binary logic operations.
17077 * config/i386/i386.cc (ix86_rtx_costs) <case UNSPEC>: Add costs
17078 for UNSPEC_PTEST; a PTEST that performs an AND has the same cost
17079 as regular PTEST, i.e. cost->sse_op.
17080 * config/i386/i386.md (*testti_doubleword): New pre-reload
17081 define_insn_and_split that recognizes comparison of TI mode AND
17083 * config/i386/sse.md (*ptest<mode>_and): New pre-reload
17084 define_insn_and_split that recognizes UNSPEC_PTEST of identical
17087 2022-08-09 Roger Sayle <roger@nextmovesoftware.com>
17088 Richard Biener <rguenther@suse.de>
17090 PR middle-end/21137
17091 PR tree-optimization/98954
17092 * fold-const.cc (fold_binary_loc): Remove optimizations to
17093 optimize ((X >> C1) & C2) ==/!= 0.
17094 * match.pd (cmp (bit_and (lshift @0 @1) @2) @3): Remove wi::ctz
17095 check, and handle all values of INTEGER_CSTs @2 and @3.
17096 (cmp (bit_and (rshift @0 @1) @2) @3): Likewise, remove wi::clz
17097 checks, and handle all values of INTEGER_CSTs @2 and @3.
17099 2022-08-09 David Malcolm <dmalcolm@redhat.com>
17101 * doc/invoke.texi (Static Analyzer Options): Add notes on which
17102 functions the analyzer has hardcoded knowledge of.
17104 2022-08-09 Andrew Stubbs <ams@codesourcery.com>
17106 * config/gcn/gcn.cc (gcn_function_value): Allow vector return values.
17107 (num_arg_regs): Allow vector arguments.
17108 (gcn_function_arg): Likewise.
17109 (gcn_function_arg_advance): Likewise.
17110 (gcn_arg_partial_bytes): Likewise.
17111 (gcn_return_in_memory): Likewise.
17112 (gcn_expand_epilogue): Get return value from v8.
17113 * config/gcn/gcn.h (RETURN_VALUE_REG): Set to v8.
17114 (FIRST_PARM_REG): USE FIRST_SGPR_REG for clarity.
17115 (FIRST_VPARM_REG): New.
17116 (FUNCTION_ARG_REGNO_P): Allow vector parameters.
17117 (struct gcn_args): Add vnum field.
17118 (LIBCALL_VALUE): All vector return values.
17119 * config/gcn/gcn.md (gcn_call_value): Add vector constraints.
17120 (gcn_call_value_indirect): Likewise.
17122 2022-08-09 Richard Biener <rguenther@suse.de>
17124 * omp-expand.cc (expand_omp_atomic_load): Emit GIMPLE
17125 directly. Avoid update_ssa when in SSA form.
17126 (expand_omp_atomic_store): Likewise.
17127 (expand_omp_atomic_fetch_op): Avoid update_ssa when in SSA
17129 (expand_omp_atomic_pipeline): Likewise.
17130 (expand_omp_atomic_mutex): Likewise.
17131 * tree-parloops.cc (gen_parallel_loop): Use
17132 TODO_update_ssa_no_phi after loop_version.
17134 2022-08-09 Richard Biener <rguenther@suse.de>
17136 * doc/invoke.texi (max-fsm-thread-length): Remove.
17137 * params.opt (max-fsm-thread-length): Likewise.
17138 * tree-ssa-threadbackward.cc
17139 (back_threader_profitability::profitable_path_p): Do not
17140 check max-fsm-thread-length.
17142 2022-08-09 Richard Biener <rguenther@suse.de>
17144 PR tree-optimization/106514
17145 * params.opt (max-jump-thread-paths): New.
17146 * doc/invoke.texi (max-jump-thread-paths): Document.
17147 * tree-ssa-threadbackward.cc (back_threader::find_paths_to_names):
17148 Honor max-jump-thread-paths, take overall_path argument.
17149 (back_threader::find_paths): Pass 1 as initial overall_path.
17151 2022-08-09 Tobias Burnus <tobias@codesourcery.com>
17153 PR middle-end/106492
17154 * omp-low.cc (lower_rec_input_clauses): Add missing folding
17155 to data type of linear-clause list item.
17157 2022-08-08 Andrew MacLeod <amacleod@redhat.com>
17159 PR tree-optimization/106556
17160 * gimple-range-gori.cc (gori_compute::condexpr_adjust): Use the
17161 type of the cond_expr operands being evaluted.
17163 2022-08-08 Tom Honermann <tom@honermann.net>
17165 * ginclude/stdatomic.h (atomic_char8_t,
17166 ATOMIC_CHAR8_T_LOCK_FREE): New typedef and macro.
17168 2022-08-08 Andrew Pinski <apinski@marvell.com>
17170 PR middle-end/103645
17171 * gimplify.cc (gimplify_init_constructor): Don't build/add
17172 gimple assignment of an empty type.
17174 2022-08-08 Richard Biener <rguenther@suse.de>
17178 * dwarf2out.cc (dwarf2out_register_external_die): Restore
17180 * lto-streamer-in.cc (lto_read_tree_1): Use lto_input_tree_1
17181 to input DECL_INITIAL, avoiding to commit drefs.
17183 2022-08-07 Roger Sayle <roger@nextmovesoftware.com>
17185 * config/i386/i386.md (*cmp<dwi>_doubleword): Change predicate
17186 for x86_64_hilo_general_operand to general operand. Call
17187 force_reg on parts that are not x86_64_immediate_operand.
17189 2022-08-05 David Malcolm <dmalcolm@redhat.com>
17192 * doc/invoke.texi: Add -Wanalyzer-jump-through-null.
17194 2022-08-05 Roger Sayle <roger@nextmovesoftware.com>
17196 * expmed.cc (emit_store_flag_1): Move code to expand double word
17197 equality and inequality against zero or -1, using word operations,
17198 to after trying to use the backend's cstore<mode>4 optab/expander.
17200 2022-08-05 Tamar Christina <tamar.christina@arm.com>
17202 PR middle-end/106534
17203 * tree-ssa-phiopt.cc (tree_ssa_phiopt_worker): Guard the
17204 value_replacement and store_elim from diamonds.
17206 2022-08-05 Richard Biener <rguenther@suse.de>
17208 * tree-ssa-threadbackward.cc (back_threader::maybe_register_path):
17209 Check whether the registry register_path rejected the path.
17210 (back_threader_registry::register_path): Return whether
17211 register_jump_thread succeeded.
17213 2022-08-05 Aldy Hernandez <aldyh@redhat.com>
17215 PR tree-optimization/106514
17216 * value-range.cc (unsupported_range::unsupported_range): Move...
17217 * value-range.h (unsupported_range::unsupported_range): ...here.
17218 (unsupported_range::set_undefined): New.
17220 2022-08-05 Richard Biener <rguenther@suse.de>
17222 PR tree-optimization/106533
17223 * tree-loop-distribution.cc (loop_distribution::execute): Continue
17224 analyzing the inner loops when find_seed_stmts_for_distribution
17227 2022-08-05 Andrew Pinski <apinski@marvell.com>
17229 * config/riscv/predicates.md (splittable_const_int_operand):
17230 Remove the check for TARGET_64BIT for single bit const values.
17232 2022-08-04 Andrew MacLeod <amacleod@redhat.com>
17234 PR tree-optimization/106514
17235 * gimple-range-path.cc (path_range_query::compute_ranges_in_block):
17236 Use EXECUTE_IF_AND_IN_BITMAP to loop over 2 bitmaps.
17238 2022-08-04 Tamar Christina <tamar.christina@arm.com>
17240 * match.pd: New bit_not rule.
17242 2022-08-04 Tamar Christina <tamar.christina@arm.com>
17244 PR middle-end/106519
17245 * tree-ssa-phiopt.cc (tree_ssa_phiopt_worker): Check final phi edge for
17248 2022-08-04 Sam Feifer <sfeifer@redhat.com>
17250 PR tree-optimization/106243
17251 * match.pd (-x & 1): New simplification.
17253 2022-08-04 Richard Biener <rguenther@suse.de>
17255 PR tree-optimization/106521
17256 * gimple-loop-jam.cc (tree_loop_unroll_and_jam): Perform
17257 CFG cleanup manually before rewriting into LC SSA.
17259 2022-08-04 Richard Biener <rguenther@suse.de>
17261 * tree-ssa-threadbackward.cc (populate_worklist): Remove.
17262 (back_threader::resolve_phi): Likewise.
17263 (back_threader::find_paths_to_names): Rewrite greedy search.
17265 2022-08-04 Ilya Leoshkevich <iii@linux.ibm.com>
17267 * config/s390/vector.md (V_HW_FT): New iterator.
17268 * config/s390/vx-builtins.md (vsel<mode>): Use V_HW_FT instead
17271 2022-08-03 Michael Meissner <meissner@linux.ibm.com>
17273 * config/rs6000/rs6000.cc (rs6000_option_override_internal): Remove code
17274 setting -mblock-ops-vector-pair.
17276 2022-08-03 Andrew MacLeod <amacleod@redhat.com>
17278 PR tree-optimization/106514
17279 * value-relation.cc (path_oracle::killing_def) Do not walk the
17280 equivalence set clearing bits.
17282 2022-08-03 Tamar Christina <tamar.christina@arm.com>
17284 * tree-ssa-phiopt.cc (minmax_replacement): Optionally search for the phi
17285 sequence of a three-way conditional.
17286 (replace_phi_edge_with_variable): Support diamonds.
17287 (tree_ssa_phiopt_worker): Detect diamond phi structure for three-way
17289 (strip_bit_not, invert_minmax_code): New.
17291 2022-08-03 Richard Earnshaw <rearnsha@arm.com>
17293 PR rtl-optimization/106187
17294 * alias.h (mems_same_for_tbaa_p): Declare.
17295 * alias.cc (mems_same_for_tbaa_p): New function.
17296 * dse.cc (record_store): Use it instead of open-coding
17298 * cselib.h (cselib_redundant_set_p): Declare.
17299 * cselib.cc: Include alias.h
17300 (cselib_redundant_set_p): New function.
17301 * cfgcleanup.cc: (mark_effect): Use cselib_redundant_set_p instead
17302 of rtx_equal_for_cselib_p.
17303 * postreload.cc (reload_cse_simplify): Use cselib_redundant_set_p.
17304 (reload_cse_noop_set_p): Delete.
17306 2022-08-03 Martin Liska <mliska@suse.cz>
17308 * doc/gcov-dump.texi: Document the new option.
17309 * gcov-dump.cc (main): Parse the new option.
17310 (print_usage): Show the option.
17311 (tag_counters): Sort key:value pairs of TOP N counter.
17313 2022-08-03 Martin Liska <mliska@suse.cz>
17315 * profile.cc (compute_branch_probabilities): Do not collect
17316 stats unless TDF_DETAILS.
17318 2022-08-03 Roger Sayle <roger@nextmovesoftware.com>
17319 Uroš Bizjak <ubizjak@gmail.com>
17322 * config/i386/i386.md (peephole2): New peephole2 to convert
17323 SWI48 moves to/from %rax/%eax where the src is dead to xchg,
17324 when optimizing for minimal size with -Oz.
17326 2022-08-03 Roger Sayle <roger@nextmovesoftware.com>
17328 * config/i386/i386.md (*cmp<dwi>_doubleword): Add a special case
17329 to split comparisons against -1 using AND and CMP -1 instructions.
17331 2022-08-03 Roger Sayle <roger@nextmovesoftware.com>
17333 * config/i386/i386-features.cc (compute_convert_gain): Add gain
17334 for converting suitable TImode shift to a V1TImode shift.
17335 (timode_scalar_chain::convert_insn): Add support for converting
17336 suitable ASHIFT and LSHIFTRT.
17337 (timode_scalar_to_vector_candidate_p): Consider logical shifts
17338 by integer constants that are multiples of 8 to be candidates.
17340 2022-08-03 Roger Sayle <roger@nextmovesoftware.com>
17341 Segher Boessenkool <segher@kernel.crashing.org>
17342 Richard Sandiford <richard.sandiford@arm.com>
17344 * simplify-rtx.cc (simplify_unary_operation_1) <ABS>: Add
17345 optimizations for CLRSB, PARITY, POPCOUNT, SS_ABS and LSHIFTRT
17346 that are all positive to complement the existing FFS and
17347 idempotent ABS simplifications.
17348 <SIGN_EXTEND>: Canonicalize SIGN_EXTEND to ZERO_EXTEND when
17349 val_signbit_known_clear_p is true of the operand.
17350 Simplify sign extensions of SUBREG truncations of operands
17351 that are already suitably (zero) extended.
17352 <ZERO_EXTEND>: Simplify zero extensions of SUBREG truncations
17353 of operands that are already suitably zero extended.
17355 2022-08-02 Andrew MacLeod <amacleod@redhat.com>
17357 PR tree-optimization/106510
17358 * gimple-range-fold.cc (fur_source::register_outgoing_edges):
17359 Check for unsupported statements early.
17361 2022-08-02 Andrew MacLeod <amacleod@redhat.com>
17363 PR tree-optimization/106474
17364 * gimple-range-cache.cc (ranger_cache::fill_block_cache): Query
17365 range of equivalences that may contribute to the range.
17367 2022-08-02 Jose E. Marchesi <jose.marchesi@oracle.com>
17369 * btfout.cc (output_asm_btf_vlen_bytes): Do not use the CHAR
17370 encoding bit in BTF.
17372 2022-08-02 Aldy Hernandez <aldyh@redhat.com>
17374 * gimple-range-fold.cc (fold_using_range::range_of_phi): Remove
17376 (tree_lower_bound): New.
17377 (tree_upper_bound): New.
17378 (fold_using_range::range_of_ssa_name_with_loop_info): Convert to
17380 * gimple-range-fold.h (range_of_ssa_name_with_loop_info): Change
17381 argument to vrange.
17383 2022-08-02 Richard Biener <rguenther@suse.de>
17385 * tree-ssa-threadbackward.cc
17386 (back_threader_profitability::profitable_path_p): Apply
17387 size constraints to all paths again.
17389 2022-08-02 Aldy Hernandez <aldyh@redhat.com>
17391 * range-op-float.cc (finite_operands_p): New.
17392 (frelop_early_resolve): New.
17393 (default_frelop_fold_range): New.
17394 (class foperator_equal): New.
17395 (class foperator_not_equal): New.
17396 (class foperator_lt): New.
17397 (class foperator_le): New.
17398 (class foperator_gt): New.
17399 (class foperator_ge): New.
17400 (class foperator_unordered): New.
17401 (class foperator_ordered): New.
17402 (class foperator_relop_unknown): New.
17403 (floating_op_table::floating_op_table): Add above classes to
17405 * value-range.h (frange::supports_p): Enable.
17407 2022-08-02 Aldy Hernandez <aldyh@redhat.com>
17409 * tree-core.h (struct tree_ssa_name): Add frange_info and
17410 reshuffle the rest.
17411 * value-range-storage.cc (vrange_storage::alloc_slot): Add case
17413 (vrange_storage::set_vrange): Same.
17414 (vrange_storage::get_vrange): Same.
17415 (vrange_storage::fits_p): Same.
17416 (frange_storage_slot::alloc_slot): New.
17417 (frange_storage_slot::set_frange): New.
17418 (frange_storage_slot::get_frange): New.
17419 (frange_storage_slot::fits_p): New.
17420 * value-range-storage.h (class frange_storage_slot): New.
17422 2022-08-02 Aldy Hernandez <aldyh@redhat.com>
17424 * ipa-prop.cc (ipa_compute_jump_functions_for_edge): Limit ranger
17425 query to integrals.
17427 2022-08-02 Aldy Hernandez <aldyh@redhat.com>
17429 * value-range.cc (frange::set): Initialize m_props and cleanup.
17431 2022-08-02 Richard Biener <rguenther@suse.de>
17433 PR tree-optimization/106497
17434 * tree-ssa-threadupdate.cc (fwd_jt_path_registry::update_cfg):
17435 Also verify we can copy EDGE_COPY_SRC_JOINER_BLOCK.
17437 2022-08-02 Martin Liska <mliska@suse.cz>
17439 * profile.cc (compute_branch_probabilities): Dump details only
17441 * symtab.cc (symtab_node::dump_base): Do not dump pointer unless
17442 TDF_ADDRESS is used, it makes comparison harder.
17444 2022-08-02 Richard Biener <rguenther@suse.de>
17446 PR tree-optimization/106498
17447 * omp-expand.cc (expand_omp_taskreg): Do not perform virtual
17449 (expand_omp_for): Or here.
17450 (execute_expand_omp): Instead schedule it here together
17451 with CFG cleanup via TODO.
17453 2022-08-02 Richard Biener <rguenther@suse.de>
17456 * dwarf2out.cc (dwarf2out_register_external_die): Adjust
17459 2022-08-02 Richard Biener <rguenther@suse.de>
17461 PR tree-optimization/106495
17462 * tree-ssa-threadbackward.cc
17463 (back_threader_profitability::profitable_path_p): If known_edge
17464 is probably never executed avoid threading.
17466 2022-08-01 David Malcolm <dmalcolm@redhat.com>
17468 * doc/invoke.texi (-Wanalyzer-putenv-of-auto-var): Fix copy&paste
17471 2022-08-01 Roger Sayle <roger@nextmovesoftware.com>
17472 Uroš Bizjak <ubizjak@gmail.com>
17475 * config/i386/i386-features.cc (timode_scalar_chain::convert_insn):
17476 Convert a CONST_SCALAR_INT_P in a REG_EQUAL note into a V1TImode
17479 2022-08-01 H.J. Lu <hjl.tools@gmail.com>
17482 * config/i386/i386.cc (ix86_ifunc_ref_local_ok): New.
17483 (TARGET_IFUNC_REF_LOCAL_OK): Use it.
17485 2022-08-01 Jose E. Marchesi <jose.marchesi@oracle.com>
17488 * ctfc.h (struct ctf_dtdef): Add field linkage.
17489 * ctfc.cc (ctf_add_function): Set ctti_linkage.
17490 * dwarf2ctf.cc (gen_ctf_function_type): Pass a linkage for
17491 function types and subprograms.
17492 * btfout.cc (btf_asm_func_type): Emit linkage information for the
17494 (btf_dtd_emit_preprocess_cb): Propagate the linkage information
17497 2022-08-01 Andrew Stubbs <ams@codesourcery.com>
17498 Jakub Jelinek <jakub@redhat.com>
17500 * omp-simd-clone.cc (simd_clone_adjust): Convert shift_cnt to match
17503 2022-08-01 Sam Feifer <sfeifer@redhat.com>
17505 PR tree-optimization/104992
17506 * match.pd (x / y * y == x): New simplification.
17508 2022-08-01 Aldy Hernandez <aldyh@redhat.com>
17510 * value-range.cc (tree_compare): New.
17511 (frange::set): Make more general.
17512 (frange::normalize_kind): Cleanup and return bool.
17513 (frange::union_): Use normalize_kind return value.
17514 (frange::intersect): Same.
17515 (frange::verify_range): Remove unnecessary else.
17516 * value-range.h (vrp_val_max): Move before frange class.
17517 (vrp_val_min): Same.
17518 (frange::frange): Remove set to m_type.
17520 2022-08-01 Aldy Hernandez <aldyh@redhat.com>
17522 * value-range.cc (vrange::supports_type_p): Use const_tree.
17523 (irange::supports_type_p): Same.
17524 (frange::supports_type_p): Same.
17525 * value-range.h (Value_Range::supports_type_p): Same.
17526 (irange::supports_p): Same.
17528 2022-08-01 Aldy Hernandez <aldyh@redhat.com>
17530 * gimple-range-fold.cc (fold_using_range::range_of_phi): Only
17531 query SCEV for integers.
17532 (fold_using_range::range_of_ssa_name_with_loop_info): Remove
17535 2022-07-31 Roger Sayle <roger@nextmovesoftware.com>
17537 * config/i386/i386.md (define_expand <any_rotate>ti3): For
17538 rotations by 64 bits use new rot[lr]64ti2_doubleword pattern.
17539 (rot[lr]64ti2_doubleword): New post-reload splitter.
17541 2022-07-31 Roger Sayle <roger@nextmovesoftware.com>
17542 H.J. Lu <hjl.tools@gmail.com>
17545 * config/i386/i386-features.cc (timode_check_non_convertible_regs):
17546 Do nothing if REGNO is set in the REGS bitmap, or is a hard reg.
17547 (timode_remove_non_convertible_regs): Update comment.
17548 Call timode_check_non_convertible_reg on all TImode register
17549 DEFs and USEs in each instruction.
17551 2022-07-30 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
17553 * config/xtensa/xtensa.md: Change hard register number used in
17554 the split patterns for indirect sibling call fixups from 10 to 11,
17555 the last free one for the CALL0 ABI.
17557 2022-07-30 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
17559 * config/xtensa/xtensa.cc (xtensa_rtx_costs):
17560 Add new case for IF_THEN_ELSE.
17562 2022-07-29 Andrew Stubbs <ams@codesourcery.com>
17564 * config/gcn/gcn-valu.md (V_INT_noHI): New iterator.
17565 (<expander><mode>3<exec>): Use V_INT_noHI.
17566 (v<expander><mode>3<exec>): Likewise.
17568 2022-07-29 Andrew Stubbs <ams@codesourcery.com>
17570 * config/gcn/gcn.md (one_cmpldi2): New.
17572 2022-07-29 Richard Biener <rguenther@suse.de>
17574 PR tree-optimization/105679
17575 * tree-ssa-threadbackward.cc
17576 (back_threader_profitability::profitable_path_p): Avoid threading
17577 when the entry edge is probably never executed.
17579 2022-07-29 Richard Biener <rguenther@suse.de>
17581 PR tree-optimization/106422
17582 * tree-ssa-threadupdate.cc (fwd_jt_path_registry::update_cfg):
17583 Check whether we can copy thread blocks and cancel the thread if not.
17585 2022-07-29 Jakub Jelinek <jakub@redhat.com>
17587 PR middle-end/106449
17588 * omp-expand.cc (expand_omp_simd): Fix up handling of pointer
17589 iterators in non-rectangular simd loops. Unshare fd->loops[i].n2
17590 or n2 before regimplifying it inside of a condition.
17592 2022-07-29 Jakub Jelinek <jakub@redhat.com>
17594 * omp-expand.cc (expand_omp_for_init_counts, expand_omp_for_init_vars,
17595 extract_omp_for_update_vars, expand_omp_for_ordered_loops,
17596 expand_omp_simd): Don't fold_convert second argument to
17597 fold_build_pointer_plus to sizetype.
17599 2022-07-29 Lulu Cheng <chenglulu@loongson.cn>
17601 * config.in: Regenerate.
17602 * config/loongarch/loongarch.h (ASM_PREFERRED_EH_DATA_FORMAT):
17603 Select the value of the macro definition according to whether
17604 HAVE_AS_EH_FRAME_PCREL_ENCODING_SUPPORT is defined.
17605 * configure: Regenerate.
17606 * configure.ac: Reinstate HAVE_AS_EH_FRAME_PCREL_ENCODING_SUPPORT test.
17608 2022-07-29 Richard Biener <rguenther@suse.de>
17610 * gimple-ssa-warn-restrict.cc (builtin_memref::set_base_and_offset):
17611 Use CONVERT_EXPR_CODE_P.
17613 2022-07-29 Richard Biener <rguenther@suse.de>
17615 * tree-vect-patterns.cc (vect_recog_bool_pattern): Use
17616 get_vectype_for_scalar_type instead of
17617 vect_get_vector_types_for_stmt.
17619 2022-07-28 David Malcolm <dmalcolm@redhat.com>
17622 * doc/invoke.texi: Add -Wanalyzer-putenv-of-auto-var.
17624 2022-07-28 David Malcolm <dmalcolm@redhat.com>
17626 * doc/invoke.texi (-fdiagnostics-show-cwe): Use uref rather than
17628 (Static Analyzer Options): Likewise. Add urefs for all of the
17629 warnings that have associated CWE identifiers.
17631 2022-07-28 Maciej W. Rozycki <macro@embecosm.com>
17633 * doc/implement-c.texi (Floating point implementation): Mention
17634 `-fno-trapping-math' in the context of FENV_ACCESS pragma.
17635 * doc/invoke.texi (Optimize Options): Clarify FENV_ACCESS pragma
17636 implication in the descriptions of `-fno-trapping-math' and
17639 2022-07-28 Maciej W. Rozycki <macro@embecosm.com>
17641 * config/riscv/riscv.md (UNSPECV_FSNVSNAN): New constant.
17642 (QUIET_PATTERN): New int attribute.
17643 (f<quiet_pattern>_quiet<ANYF:mode><X:mode>4): Emit the intended
17644 RTL insns entirely within the preparation statements.
17645 (*f<quiet_pattern>_quiet<ANYF:mode><X:mode>4_default)
17646 (*f<quiet_pattern>_quiet<ANYF:mode><X:mode>4_snan): Remove
17648 (*riscv_fsnvsnan<mode>2): New insn.
17650 2022-07-28 Richard Biener <rguenther@suse.de>
17652 PR middle-end/106457
17653 * tree.cc (array_at_struct_end_p): Handle array objects
17656 2022-07-28 Jakub Jelinek <jakub@redhat.com>
17658 PR tree-optimization/106099
17659 * internal-fn.def (TRAP): New internal fn.
17660 * internal-fn.h (expand_TRAP): Declare.
17661 * internal-fn.cc (expand_TRAP): Define.
17662 * gimple.cc (gimple_build_builtin_unreachable): For BUILT_IN_TRAP,
17663 use internal fn rather than builtin.
17665 2022-07-27 Andrew Carlotti <andrew.carlotti@arm.com>
17667 * doc/loop.texi: Refer to LOOPS_HAVE_RECORDED_EXITS instead.
17669 2022-07-27 WANG Xuerui <i@xen0n.name>
17671 * doc/invoke.texi: Document -m[no-]explicit-relocs for
17674 2022-07-27 Maciej W. Rozycki <macro@embecosm.com>
17676 * config/riscv/riscv.md (stack_protect_set_<mode>): Remove
17677 duplicate backslashes.
17679 2022-07-27 Maciej W. Rozycki <macro@embecosm.com>
17681 * config/riscv/riscv.cc (riscv_rtx_costs) <IF_THEN_ELSE>: New
17684 2022-07-27 Jakub Jelinek <jakub@redhat.com>
17687 * cgraphunit.cc (cgraph_node::assemble_thunks_and_aliases): Don't
17688 output asm thunks for -dx.
17690 2022-07-27 Jakub Jelinek <jakub@redhat.com>
17692 PR middle-end/106332
17693 * opts-common.cc (candidates_list_and_hint): Add gcc_assert
17694 that candidates is not an empty vector.
17696 2022-07-27 Xi Ruoyao <xry111@xry111.site>
17698 * configure.ac (HAVE_AS_EXPLICIT_RELOCS): Define to 1 if the
17699 assembler supports explicit relocation for LoongArch.
17700 * configure: Regenerate.
17701 * config/loongarch/loongarch-opts.h (HAVE_AS_EXPLICIT_RELOCS):
17702 Define to 0 if not defined.
17703 * config/loongarch/genopts/loongarch.opt.in
17704 (TARGET_EXPLICIT_RELOCS): Default to HAVE_AS_EXPLICIT_RELOCS.
17705 * config/loongarch/loongarch.opt: Regenerate.
17707 2022-07-26 Peter Bergner <bergner@linux.ibm.com>
17710 * expr.cc (count_type_elements): Handle OPAQUE_TYPE.
17712 2022-07-26 Lulu Cheng <chenglulu@loongson.cn>
17714 * config/loongarch/loongarch-opts.cc: Modify the output message string
17717 2022-07-26 Martin Liska <mliska@suse.cz>
17719 * doc/tm.texi.in: Fix placement of defmac.
17720 * doc/tm.texi: Copy.
17722 2022-07-26 Martin Liska <mliska@suse.cz>
17724 * doc/tm.texi.in: Fix cross @defmac and @hook.
17725 * doc/tm.texi: Copy.
17727 2022-07-26 Aldy Hernandez <aldyh@redhat.com>
17729 PR tree-optimization/106444
17730 * value-range-pretty-print.cc (vrange_printer::visit): Handle
17732 (vrange_printer::print_irange_bound): Work on wide_int's.
17733 * value-range-pretty-print.h (print_irange_bound): Same.
17734 * value-range.cc (irange::get_nonzero_bits): Handle legacy ranges.
17736 2022-07-26 Richard Biener <rguenther@suse.de>
17738 * tree-ssa-alias.cc (ptr_derefs_may_alias_p): If ptr1
17739 points to a constant continue checking ptr2.
17741 2022-07-26 Andrew Carlotti <Andrew.Carlotti@arm.com>
17743 * config/aarch64/aarch64-builtins.cc
17744 (MODE_d_bf16, MODE_d_f16, MODE_d_f32, MODE_d_f64, MODE_d_s8)
17745 (MODE_d_s16, MODE_d_s32, MODE_d_s64, MODE_d_u8, MODE_d_u16)
17746 (MODE_d_u32, MODE_d_u64, MODE_d_p8, MODE_d_p16, MODE_d_p64)
17747 (MODE_q_bf16, MODE_q_f16, MODE_q_f32, MODE_q_f64, MODE_q_s8)
17748 (MODE_q_s16, MODE_q_s32, MODE_q_s64, MODE_q_u8, MODE_q_u16)
17749 (MODE_q_u32, MODE_q_u64, MODE_q_p8, MODE_q_p16, MODE_q_p64)
17750 (MODE_q_p128): Define macro to map to corresponding mode name.
17751 (QUAL_bf16, QUAL_f16, QUAL_f32, QUAL_f64, QUAL_s8, QUAL_s16)
17752 (QUAL_s32, QUAL_s64, QUAL_u8, QUAL_u16, QUAL_u32, QUAL_u64)
17753 (QUAL_p8, QUAL_p16, QUAL_p64, QUAL_p128): Define macro to map to
17754 corresponding qualifier name.
17755 (LENGTH_d, LENGTH_q): Define macro to map to "" or "q" suffix.
17756 (SIMD_INTR_MODE, SIMD_INTR_QUAL, SIMD_INTR_LENGTH_CHAR): Macro
17757 functions for the above mappings
17758 (VREINTERPRET_BUILTIN2, VREINTERPRET_BUILTINS1, VREINTERPRET_BUILTINS)
17759 (VREINTERPRETQ_BUILTIN2, VREINTERPRETQ_BUILTINS1)
17760 (VREINTERPRETQ_BUILTINS, VREINTERPRET_BUILTIN)
17761 (AARCH64_SIMD_VREINTERPRET_BUILTINS): New macros to create definitions
17762 for all vreinterpret intrinsics
17763 (enum aarch64_builtins): Add vreinterpret function codes
17764 (aarch64_init_simd_intrinsics): New
17765 (handle_arm_neon_h): Improved comment.
17766 (aarch64_general_fold_builtin): Fold vreinterpret calls
17767 * config/aarch64/arm_neon.h
17768 (vreinterpret_p8_f16, vreinterpret_p8_f64, vreinterpret_p8_s8)
17769 (vreinterpret_p8_s16, vreinterpret_p8_s32, vreinterpret_p8_s64)
17770 (vreinterpret_p8_f32, vreinterpret_p8_u8, vreinterpret_p8_u16)
17771 (vreinterpret_p8_u32, vreinterpret_p8_u64, vreinterpret_p8_p16)
17772 (vreinterpret_p8_p64, vreinterpretq_p8_f64, vreinterpretq_p8_s8)
17773 (vreinterpretq_p8_s16, vreinterpretq_p8_s32, vreinterpretq_p8_s64)
17774 (vreinterpretq_p8_f16, vreinterpretq_p8_f32, vreinterpretq_p8_u8)
17775 (vreinterpretq_p8_u16, vreinterpretq_p8_u32, vreinterpretq_p8_u64)
17776 (vreinterpretq_p8_p16, vreinterpretq_p8_p64, vreinterpretq_p8_p128)
17777 (vreinterpret_p16_f16, vreinterpret_p16_f64, vreinterpret_p16_s8)
17778 (vreinterpret_p16_s16, vreinterpret_p16_s32, vreinterpret_p16_s64)
17779 (vreinterpret_p16_f32, vreinterpret_p16_u8, vreinterpret_p16_u16)
17780 (vreinterpret_p16_u32, vreinterpret_p16_u64, vreinterpret_p16_p8)
17781 (vreinterpret_p16_p64, vreinterpretq_p16_f64, vreinterpretq_p16_s8)
17782 (vreinterpretq_p16_s16, vreinterpretq_p16_s32, vreinterpretq_p16_s64)
17783 (vreinterpretq_p16_f16, vreinterpretq_p16_f32, vreinterpretq_p16_u8)
17784 (vreinterpretq_p16_u16, vreinterpretq_p16_u32, vreinterpretq_p16_u64)
17785 (vreinterpretq_p16_p8, vreinterpretq_p16_p64, vreinterpretq_p16_p128)
17786 (vreinterpret_p64_f16, vreinterpret_p64_f64, vreinterpret_p64_s8)
17787 (vreinterpret_p64_s16, vreinterpret_p64_s32, vreinterpret_p64_s64)
17788 (vreinterpret_p64_f32, vreinterpret_p64_u8, vreinterpret_p64_u16)
17789 (vreinterpret_p64_u32, vreinterpret_p64_u64, vreinterpret_p64_p8)
17790 (vreinterpret_p64_p16, vreinterpretq_p64_f64, vreinterpretq_p64_s8)
17791 (vreinterpretq_p64_s16, vreinterpretq_p64_s32, vreinterpretq_p64_s64)
17792 (vreinterpretq_p64_f16, vreinterpretq_p64_f32, vreinterpretq_p64_p128)
17793 (vreinterpretq_p64_u8, vreinterpretq_p64_u16, vreinterpretq_p64_p16)
17794 (vreinterpretq_p64_u32, vreinterpretq_p64_u64, vreinterpretq_p64_p8)
17795 (vreinterpretq_p128_p8, vreinterpretq_p128_p16, vreinterpretq_p128_f16)
17796 (vreinterpretq_p128_f32, vreinterpretq_p128_p64, vreinterpretq_p128_s64)
17797 (vreinterpretq_p128_u64, vreinterpretq_p128_s8, vreinterpretq_p128_s16)
17798 (vreinterpretq_p128_s32, vreinterpretq_p128_u8, vreinterpretq_p128_u16)
17799 (vreinterpretq_p128_u32, vreinterpret_f16_f64, vreinterpret_f16_s8)
17800 (vreinterpret_f16_s16, vreinterpret_f16_s32, vreinterpret_f16_s64)
17801 (vreinterpret_f16_f32, vreinterpret_f16_u8, vreinterpret_f16_u16)
17802 (vreinterpret_f16_u32, vreinterpret_f16_u64, vreinterpret_f16_p8)
17803 (vreinterpret_f16_p16, vreinterpret_f16_p64, vreinterpretq_f16_f64)
17804 (vreinterpretq_f16_s8, vreinterpretq_f16_s16, vreinterpretq_f16_s32)
17805 (vreinterpretq_f16_s64, vreinterpretq_f16_f32, vreinterpretq_f16_u8)
17806 (vreinterpretq_f16_u16, vreinterpretq_f16_u32, vreinterpretq_f16_u64)
17807 (vreinterpretq_f16_p8, vreinterpretq_f16_p128, vreinterpretq_f16_p16)
17808 (vreinterpretq_f16_p64, vreinterpret_f32_f16, vreinterpret_f32_f64)
17809 (vreinterpret_f32_s8, vreinterpret_f32_s16, vreinterpret_f32_s32)
17810 (vreinterpret_f32_s64, vreinterpret_f32_u8, vreinterpret_f32_u16)
17811 (vreinterpret_f32_u32, vreinterpret_f32_u64, vreinterpret_f32_p8)
17812 (vreinterpret_f32_p16, vreinterpret_f32_p64, vreinterpretq_f32_f16)
17813 (vreinterpretq_f32_f64, vreinterpretq_f32_s8, vreinterpretq_f32_s16)
17814 (vreinterpretq_f32_s32, vreinterpretq_f32_s64, vreinterpretq_f32_u8)
17815 (vreinterpretq_f32_u16, vreinterpretq_f32_u32, vreinterpretq_f32_u64)
17816 (vreinterpretq_f32_p8, vreinterpretq_f32_p16, vreinterpretq_f32_p64)
17817 (vreinterpretq_f32_p128, vreinterpret_f64_f16, vreinterpret_f64_f32)
17818 (vreinterpret_f64_p8, vreinterpret_f64_p16, vreinterpret_f64_p64)
17819 (vreinterpret_f64_s8, vreinterpret_f64_s16, vreinterpret_f64_s32)
17820 (vreinterpret_f64_s64, vreinterpret_f64_u8, vreinterpret_f64_u16)
17821 (vreinterpret_f64_u32, vreinterpret_f64_u64, vreinterpretq_f64_f16)
17822 (vreinterpretq_f64_f32, vreinterpretq_f64_p8, vreinterpretq_f64_p16)
17823 (vreinterpretq_f64_p64, vreinterpretq_f64_s8, vreinterpretq_f64_s16)
17824 (vreinterpretq_f64_s32, vreinterpretq_f64_s64, vreinterpretq_f64_u8)
17825 (vreinterpretq_f64_u16, vreinterpretq_f64_u32, vreinterpretq_f64_u64)
17826 (vreinterpret_s64_f16, vreinterpret_s64_f64, vreinterpret_s64_s8)
17827 (vreinterpret_s64_s16, vreinterpret_s64_s32, vreinterpret_s64_f32)
17828 (vreinterpret_s64_u8, vreinterpret_s64_u16, vreinterpret_s64_u32)
17829 (vreinterpret_s64_u64, vreinterpret_s64_p8, vreinterpret_s64_p16)
17830 (vreinterpret_s64_p64, vreinterpretq_s64_f64, vreinterpretq_s64_s8)
17831 (vreinterpretq_s64_s16, vreinterpretq_s64_s32, vreinterpretq_s64_f16)
17832 (vreinterpretq_s64_f32, vreinterpretq_s64_u8, vreinterpretq_s64_u16)
17833 (vreinterpretq_s64_u32, vreinterpretq_s64_u64, vreinterpretq_s64_p8)
17834 (vreinterpretq_s64_p16, vreinterpretq_s64_p64, vreinterpretq_s64_p128)
17835 (vreinterpret_u64_f16, vreinterpret_u64_f64, vreinterpret_u64_s8)
17836 (vreinterpret_u64_s16, vreinterpret_u64_s32, vreinterpret_u64_s64)
17837 (vreinterpret_u64_f32, vreinterpret_u64_u8, vreinterpret_u64_u16)
17838 (vreinterpret_u64_u32, vreinterpret_u64_p8, vreinterpret_u64_p16)
17839 (vreinterpret_u64_p64, vreinterpretq_u64_f64, vreinterpretq_u64_s8)
17840 (vreinterpretq_u64_s16, vreinterpretq_u64_s32, vreinterpretq_u64_s64)
17841 (vreinterpretq_u64_f16, vreinterpretq_u64_f32, vreinterpretq_u64_u8)
17842 (vreinterpretq_u64_u16, vreinterpretq_u64_u32, vreinterpretq_u64_p8)
17843 (vreinterpretq_u64_p16, vreinterpretq_u64_p64, vreinterpretq_u64_p128)
17844 (vreinterpret_s8_f16, vreinterpret_s8_f64, vreinterpret_s8_s16)
17845 (vreinterpret_s8_s32, vreinterpret_s8_s64, vreinterpret_s8_f32)
17846 (vreinterpret_s8_u8, vreinterpret_s8_u16, vreinterpret_s8_u32)
17847 (vreinterpret_s8_u64, vreinterpret_s8_p8, vreinterpret_s8_p16)
17848 (vreinterpret_s8_p64, vreinterpretq_s8_f64, vreinterpretq_s8_s16)
17849 (vreinterpretq_s8_s32, vreinterpretq_s8_s64, vreinterpretq_s8_f16)
17850 (vreinterpretq_s8_f32, vreinterpretq_s8_u8, vreinterpretq_s8_u16)
17851 (vreinterpretq_s8_u32, vreinterpretq_s8_u64, vreinterpretq_s8_p8)
17852 (vreinterpretq_s8_p16, vreinterpretq_s8_p64, vreinterpretq_s8_p128)
17853 (vreinterpret_s16_f16, vreinterpret_s16_f64, vreinterpret_s16_s8)
17854 (vreinterpret_s16_s32, vreinterpret_s16_s64, vreinterpret_s16_f32)
17855 (vreinterpret_s16_u8, vreinterpret_s16_u16, vreinterpret_s16_u32)
17856 (vreinterpret_s16_u64, vreinterpret_s16_p8, vreinterpret_s16_p16)
17857 (vreinterpret_s16_p64, vreinterpretq_s16_f64, vreinterpretq_s16_s8)
17858 (vreinterpretq_s16_s32, vreinterpretq_s16_s64, vreinterpretq_s16_f16)
17859 (vreinterpretq_s16_f32, vreinterpretq_s16_u8, vreinterpretq_s16_u16)
17860 (vreinterpretq_s16_u32, vreinterpretq_s16_u64, vreinterpretq_s16_p8)
17861 (vreinterpretq_s16_p16, vreinterpretq_s16_p64, vreinterpretq_s16_p128)
17862 (vreinterpret_s32_f16, vreinterpret_s32_f64, vreinterpret_s32_s8)
17863 (vreinterpret_s32_s16, vreinterpret_s32_s64, vreinterpret_s32_f32)
17864 (vreinterpret_s32_u8, vreinterpret_s32_u16, vreinterpret_s32_u32)
17865 (vreinterpret_s32_u64, vreinterpret_s32_p8, vreinterpret_s32_p16)
17866 (vreinterpret_s32_p64, vreinterpretq_s32_f64, vreinterpretq_s32_s8)
17867 (vreinterpretq_s32_s16, vreinterpretq_s32_s64, vreinterpretq_s32_f16)
17868 (vreinterpretq_s32_f32, vreinterpretq_s32_u8, vreinterpretq_s32_u16)
17869 (vreinterpretq_s32_u32, vreinterpretq_s32_u64, vreinterpretq_s32_p8)
17870 (vreinterpretq_s32_p16, vreinterpretq_s32_p64, vreinterpretq_s32_p128)
17871 (vreinterpret_u8_f16, vreinterpret_u8_f64, vreinterpret_u8_s8)
17872 (vreinterpret_u8_s16, vreinterpret_u8_s32, vreinterpret_u8_s64)
17873 (vreinterpret_u8_f32, vreinterpret_u8_u16, vreinterpret_u8_u32)
17874 (vreinterpret_u8_u64, vreinterpret_u8_p8, vreinterpret_u8_p16)
17875 (vreinterpret_u8_p64, vreinterpretq_u8_f64, vreinterpretq_u8_s8)
17876 (vreinterpretq_u8_s16, vreinterpretq_u8_s32, vreinterpretq_u8_s64)
17877 (vreinterpretq_u8_f16, vreinterpretq_u8_f32, vreinterpretq_u8_u16)
17878 (vreinterpretq_u8_u32, vreinterpretq_u8_u64, vreinterpretq_u8_p8)
17879 (vreinterpretq_u8_p16, vreinterpretq_u8_p64, vreinterpretq_u8_p128)
17880 (vreinterpret_u16_f16, vreinterpret_u16_f64, vreinterpret_u16_s8)
17881 (vreinterpret_u16_s16, vreinterpret_u16_s32, vreinterpret_u16_s64)
17882 (vreinterpret_u16_f32, vreinterpret_u16_u8, vreinterpret_u16_u32)
17883 (vreinterpret_u16_u64, vreinterpret_u16_p8, vreinterpret_u16_p16)
17884 (vreinterpret_u16_p64, vreinterpretq_u16_f64, vreinterpretq_u16_s8)
17885 (vreinterpretq_u16_s16, vreinterpretq_u16_s32, vreinterpretq_u16_s64)
17886 (vreinterpretq_u16_f16, vreinterpretq_u16_f32, vreinterpretq_u16_u8)
17887 (vreinterpretq_u16_u32, vreinterpretq_u16_u64, vreinterpretq_u16_p8)
17888 (vreinterpretq_u16_p16, vreinterpretq_u16_p64, vreinterpretq_u16_p128)
17889 (vreinterpret_u32_f16, vreinterpret_u32_f64, vreinterpret_u32_s8)
17890 (vreinterpret_u32_s16, vreinterpret_u32_s32, vreinterpret_u32_s64)
17891 (vreinterpret_u32_f32, vreinterpret_u32_u8, vreinterpret_u32_u16)
17892 (vreinterpret_u32_u64, vreinterpret_u32_p8, vreinterpret_u32_p16)
17893 (vreinterpret_u32_p64, vreinterpretq_u32_f64, vreinterpretq_u32_s8)
17894 (vreinterpretq_u32_s16, vreinterpretq_u32_s32, vreinterpretq_u32_s64)
17895 (vreinterpretq_u32_f16, vreinterpretq_u32_f32, vreinterpretq_u32_u8)
17896 (vreinterpretq_u32_u16, vreinterpretq_u32_u64, vreinterpretq_u32_p8)
17897 (vreinterpretq_u32_p16, vreinterpretq_u32_p64, vreinterpretq_u32_p128)
17898 (vreinterpretq_f64_p128, vreinterpretq_p128_f64, vreinterpret_bf16_u8)
17899 (vreinterpret_bf16_u16, vreinterpret_bf16_u32, vreinterpret_bf16_u64)
17900 (vreinterpret_bf16_s8, vreinterpret_bf16_s16, vreinterpret_bf16_s32)
17901 (vreinterpret_bf16_s64, vreinterpret_bf16_p8, vreinterpret_bf16_p16)
17902 (vreinterpret_bf16_p64, vreinterpret_bf16_f16, vreinterpret_bf16_f32)
17903 (vreinterpret_bf16_f64, vreinterpretq_bf16_u8, vreinterpretq_bf16_u16)
17904 (vreinterpretq_bf16_u32, vreinterpretq_bf16_u64, vreinterpretq_bf16_s8)
17905 (vreinterpretq_bf16_s16, vreinterpretq_bf16_s32, vreinterpretq_bf16_s64)
17906 (vreinterpretq_bf16_p8, vreinterpretq_bf16_p16, vreinterpretq_bf16_p64)
17907 (vreinterpretq_bf16_p128, vreinterpretq_bf16_f16)
17908 (vreinterpretq_bf16_f32, vreinterpretq_bf16_f64, vreinterpret_s8_bf16)
17909 (vreinterpret_s16_bf16, vreinterpret_s32_bf16, vreinterpret_s64_bf16)
17910 (vreinterpret_u8_bf16, vreinterpret_u16_bf16, vreinterpret_u32_bf16)
17911 (vreinterpret_u64_bf16, vreinterpret_f16_bf16, vreinterpret_f32_bf16)
17912 (vreinterpret_f64_bf16, vreinterpret_p8_bf16, vreinterpret_p16_bf16)
17913 (vreinterpret_p64_bf16, vreinterpretq_s8_bf16, vreinterpretq_s16_bf16)
17914 (vreinterpretq_s32_bf16, vreinterpretq_s64_bf16, vreinterpretq_u8_bf16)
17915 (vreinterpretq_u16_bf16, vreinterpretq_u32_bf16, vreinterpretq_u64_bf16)
17916 (vreinterpretq_f16_bf16, vreinterpretq_f32_bf16, vreinterpretq_f64_bf16)
17917 (vreinterpretq_p8_bf16, vreinterpretq_p16_bf16, vreinterpretq_p64_bf16)
17918 (vreinterpretq_p128_bf16): Delete
17920 2022-07-26 Andrew Carlotti <andrew.carlotti@arm.com>
17922 * config/aarch64/aarch64-builtins.cc
17923 (aarch64_simd_builtin_std_type): Rename to...
17924 (aarch64_int_or_fp_type): ...this, and allow irrelevant qualifiers.
17925 (aarch64_lookup_simd_builtin_type): Rename to...
17926 (aarch64_simd_builtin_type): ...this. Add const/pointer
17927 support, and extract table lookup to...
17928 (aarch64_lookup_simd_type_in_table): ...this function.
17929 (aarch64_init_crc32_builtins): Update to use aarch64_simd_builtin_type.
17930 (aarch64_init_fcmla_laneq_builtins): Ditto.
17931 (aarch64_init_simd_builtin_functions): Ditto.
17933 2022-07-26 Andrew Carlotti <andrew.carlotti@arm.com>
17935 * config/aarch64/aarch64-builtins.cc
17936 (aarch64_general_gimple_fold_builtin): Add combine.
17938 2022-07-26 Richard Biener <rguenther@suse.de>
17940 PR tree-optimization/106189
17941 * gimple-array-bounds.cc (array_bounds_checker::check_mem_ref):
17942 Divide using offset_ints.
17944 2022-07-26 Lulu Cheng <chenglulu@loongson.cn>
17946 * common/config/loongarch/loongarch-common.cc:
17947 Enable '-fsection-anchors' when O1 and more advanced optimization.
17948 * config/loongarch/genopts/loongarch.opt.in: Add new option
17949 '-mexplicit-relocs', and enable by default.
17950 * config/loongarch/loongarch-protos.h (loongarch_split_move_insn_p):
17951 Delete function declaration.
17952 (loongarch_split_move_insn): Delete function declaration.
17953 (loongarch_split_symbol_type): Add function declaration.
17954 * config/loongarch/loongarch.cc (enum loongarch_address_type):
17955 Add new address type 'ADDRESS_LO_SUM'.
17956 (loongarch_classify_symbolic_expression): New function definitions.
17957 Classify the base of symbolic expression X, given that X appears in
17959 (loongarch_symbol_insns): Add a judgment condition TARGET_EXPLICIT_RELOCS.
17960 (loongarch_split_symbol_type): New function definitions.
17961 Determines whether the symbol load should be split into two instructions.
17962 (loongarch_valid_lo_sum_p): New function definitions.
17963 Return true if a LO_SUM can address a value of mode MODE when the LO_SUM
17964 symbol has type SYMBOL_TYPE.
17965 (loongarch_classify_address): Add handling of 'LO_SUM'.
17966 (loongarch_address_insns): Add handling of 'ADDRESS_LO_SUM'.
17967 (loongarch_signed_immediate_p): Sort code.
17968 (loongarch_12bit_offset_address_p): Return true if address type is ADDRESS_LO_SUM.
17969 (loongarch_const_insns): Add handling of 'HIGH'.
17970 (loongarch_split_move_insn_p): Add the static attribute to the function.
17971 (loongarch_emit_set): New function definitions.
17972 (loongarch_call_tls_get_addr): Add symbol handling when defining TARGET_EXPLICIT_RELOCS.
17973 (loongarch_legitimize_tls_address): Add symbol handling when defining the
17974 TARGET_EXPLICIT_RELOCS macro.
17975 (loongarch_split_symbol): New function definitions. Split symbol.
17976 (loongarch_legitimize_address): Add codes see if the address can split into a high part
17978 (loongarch_legitimize_const_move): Add codes split moves of symbolic constants into
17980 (loongarch_split_move_insn): Delete function definitions.
17981 (loongarch_output_move): Add support for HIGH and LO_SUM.
17982 (loongarch_print_operand_reloc): New function definitions.
17983 Print symbolic operand OP, which is part of a HIGH or LO_SUM in context CONTEXT.
17984 (loongarch_memmodel_needs_release_fence): Sort code.
17985 (loongarch_print_operand): Rearrange alphabetical order and add H and L to support HIGH
17987 (loongarch_print_operand_address): Add handling of 'ADDRESS_LO_SUM'.
17988 (TARGET_MIN_ANCHOR_OFFSET): Define macro to -IMM_REACH/2.
17989 (TARGET_MAX_ANCHOR_OFFSET): Define macro to IMM_REACH/2-1.
17990 * config/loongarch/loongarch.md (movti): Delete the template.
17991 (*movti): Delete the template.
17992 (movtf): Delete the template.
17993 (*movtf): Delete the template.
17994 (*low<mode>): New template of normal symbol low address.
17995 (@tls_low<mode>): New template of tls symbol low address.
17996 (@ld_from_got<mode>): New template load address from got table.
17997 (@ori_l_lo12<mode>): New template.
17998 * config/loongarch/loongarch.opt: Update from loongarch.opt.in.
17999 * config/loongarch/predicates.md: Add support for symbol_type HIGH.
18001 2022-07-26 Lulu Cheng <chenglulu@loongson.cn>
18003 * config/loongarch/constraints.md (a): Delete the constraint.
18004 (b): A constant call not local address.
18005 (h): Delete the constraint.
18006 (t): Delete the constraint.
18007 * config/loongarch/loongarch-opts.cc (loongarch_config_target):
18008 Remove cModel type support other than normal.
18009 * config/loongarch/loongarch-protos.h (enum loongarch_symbol_type):
18010 Add new symbol type 'SYMBOL_PCREL', 'SYMBOL_TLS_IE' and 'SYMBOL_TLS_LE'.
18011 (loongarch_split_symbol): Delete useless function declarations.
18012 (loongarch_split_symbol_type): Delete useless function declarations.
18013 * config/loongarch/loongarch.cc (enum loongarch_address_type):
18014 Delete unnecessary comment information.
18015 (loongarch_symbol_binds_local_p): Modified the judgment order of label
18017 (loongarch_classify_symbol): Return symbol type. If symbol is a label,
18018 or symbol is a local symbol return SYMBOL_PCREL. If is a tls symbol,
18019 return SYMBOL_TLS. If is a not local symbol return SYMBOL_GOT_DISP.
18020 (loongarch_symbolic_constant_p): Add handling of 'SYMBOL_TLS_IE'
18021 'SYMBOL_TLS_LE' and 'SYMBOL_PCREL'.
18022 (loongarch_symbol_insns): Add handling of 'SYMBOL_TLS_IE' 'SYMBOL_TLS_LE'
18023 and 'SYMBOL_PCREL'.
18024 (loongarch_address_insns): Sort code.
18025 (loongarch_12bit_offset_address_p): Sort code.
18026 (loongarch_14bit_shifted_offset_address_p): Sort code.
18027 (loongarch_call_tls_get_addr): Sort code.
18028 (loongarch_legitimize_tls_address): Sort code.
18029 (loongarch_output_move): Remove schema support for cmodel other than normal.
18030 (loongarch_memmodel_needs_release_fence): Sort code.
18031 (loongarch_print_operand): Sort code.
18032 * config/loongarch/loongarch.h (LARCH_U12BIT_OFFSET_P):
18033 Rename to LARCH_12BIT_OFFSET_P.
18034 (LARCH_12BIT_OFFSET_P): New macro.
18035 * config/loongarch/loongarch.md: Reimplement the function call. Remove schema
18036 support for cmodel other than normal.
18037 * config/loongarch/predicates.md (is_const_call_weak_symbol): Delete this predicate.
18038 (is_const_call_plt_symbol): Delete this predicate.
18039 (is_const_call_global_noplt_symbol): Delete this predicate.
18040 (is_const_call_no_local_symbol): New predicate, determines whether it is a local
18043 2022-07-26 Kewen Lin <linkw@linux.ibm.com>
18046 * config/rs6000/rs6000-p8swap.cc (replace_swapped_aligned_store): Copy
18047 REG_EH_REGION when replacing one store insn having it.
18048 (replace_swapped_aligned_load): Likewise.
18050 2022-07-25 Aldy Hernandez <aldyh@redhat.com>
18052 * Makefile.in (OBJS): Add range-op-float.o.
18053 * range-op.cc (get_float_handler): New.
18054 (range_op_handler::range_op_handler): Save code and type for
18056 (range_op_handler::oeprator bool): Move from header file, and
18057 add support for floats.
18058 (range_op_handler::fold_range): Add support for floats.
18059 (range_op_handler::op1_range): Same.
18060 (range_op_handler::op2_range): Same.
18061 (range_op_handler::lhs_op1_relation): Same.
18062 (range_op_handler::lhs_op2_relation): Same.
18063 (range_op_handler::op1_op2_relation): Same.
18064 * range-op.h (class range_operator_float): New.
18065 (class floating_op_table): New.
18066 * value-query.cc (range_query::get_tree_range): Add case for
18068 * range-op-float.cc: New file.
18070 2022-07-25 Roger Sayle <roger@nextmovesoftware.com>
18071 Uroš Bizjak <ubizjak@gmail.com>
18074 * config/i386/i386-expand.cc (split_double_concat): A new helper
18075 function for setting a double word value from two word values.
18076 * config/i386/i386-protos.h (split_double_concat): Prototype here.
18077 * config/i386/i386.md (zero_extendditi2): New define_insn_and_split.
18078 (*add<dwi>3_doubleword_zext): New define_insn_and_split.
18079 (*sub<dwi>3_doubleword_zext): New define_insn_and_split.
18080 (*concat<mode><dwi>3_1): New define_insn_and_split replacing
18081 previous define_split for implementing DST = (HI<<32)|LO as
18082 pair of move instructions, setting lopart and hipart.
18083 (*concat<mode><dwi>3_2): Likewise.
18084 (*concat<mode><dwi>3_3): Likewise, where HI is zero_extended.
18085 (*concat<mode><dwi>3_4): Likewise, where HI is zero_extended.
18087 2022-07-25 Aldy Hernandez <aldyh@redhat.com>
18089 PR middle-end/106432
18090 * gimple-range.cc (gimple_ranger::range_on_edge): Return false
18091 when the result range type is unsupported.
18093 2022-07-25 Sebastian Huber <sebastian.huber@embedded-brains.de>
18095 * config/rs6000/rtems.h (CPLUSPLUS_CPP_SPEC): Undef.
18097 2022-07-25 Richard Biener <rguenther@suse.de>
18099 PR middle-end/106414
18100 * match.pd (~(x ^ y) -> x == y): Restrict to single bit
18103 2022-07-25 Andre Vieira <andre.simoesdiasvieira@arm.com>
18105 * config/aarch64/aarch64.md (rbit<mode>2): Rename this ...
18106 (@aarch64_rbit<mode>): ... to this and change it in...
18107 (ffs<mode>2,ctz<mode>2): ... here.
18108 (@aarch64_rev16<mode>): New.
18109 * config/aarch64/aarch64-builtins.cc: (aarch64_builtins):
18110 Define the following enum AARCH64_REV16, AARCH64_REV16L,
18111 AARCH64_REV16LL, AARCH64_RBIT, AARCH64_RBITL, AARCH64_RBITLL.
18112 (aarch64_init_data_intrinsics): New.
18113 (aarch64_general_init_builtins): Add call to
18114 aarch64_init_data_intrinsics.
18115 (aarch64_expand_builtin_data_intrinsic): New.
18116 (aarch64_general_expand_builtin): Add call to
18117 aarch64_expand_builtin_data_intrinsic.
18118 * config/aarch64/arm_acle.h (__clz, __clzl, __clzll, __cls, __clsl,
18119 __clsll, __rbit, __rbitl, __rbitll, __rev, __revl, __revll, __rev16,
18120 __rev16l, __rev16ll, __ror, __rorl, __rorll, __revsh): New.
18122 2022-07-25 Martin Liska <mliska@suse.cz>
18124 * doc/extend.texi: Remove trailing whitespaces.
18125 * doc/invoke.texi: Likewise.
18127 2022-07-25 Aldy Hernandez <aldyh@redhat.com>
18129 * value-range-pretty-print.cc (vrange_printer::visit): New.
18130 (vrange_printer::print_frange_prop): New.
18131 * value-range-pretty-print.h (class vrange_printer): Add visit and
18133 * value-range-storage.h (vrange_allocator::alloc_vrange): Handle frange.
18134 (vrange_allocator::alloc_frange): New.
18135 * value-range.cc (vrange::operator=): Handle frange.
18136 (vrange::operator==): Same.
18137 (frange::accept): New.
18138 (frange::set): New.
18139 (frange::normalize_kind): New.
18140 (frange::union_): New.
18141 (frange::intersect): New.
18142 (frange::operator=): New.
18143 (frange::operator==): New.
18144 (frange::supports_type_p): New.
18145 (frange::verify_range): New.
18146 * value-range.h (enum value_range_discriminator): Handle frange.
18147 (class fp_prop): New.
18148 (FP_PROP_ACCESSOR): New.
18149 (class frange_props): New.
18150 (FRANGE_PROP_ACCESSOR): New.
18151 (class frange): New.
18152 (Value_Range::init): Handle frange.
18153 (Value_Range::operator=): Same.
18154 (Value_Range::supports_type_p): Same.
18155 (frange_props::operator==): New.
18156 (frange_props::union_): New.
18157 (frange_props::intersect): New
18158 (frange::frange): New.
18159 (frange::type): New.
18160 (frange::set_varying): New.
18161 (frange::set_undefined): New.
18163 2022-07-25 Peter Bergner <bergner@linux.ibm.com>
18164 Kewen Lin <linkw@linux.ibm.com>
18166 PR testsuite/106345
18167 * config/rs6000/rs6000.h (DRIVER_SELF_SPECS): Adjust -mdejagnu-cpu
18168 to filter out all -mtune options.
18170 2022-07-24 Aldy Hernandez <aldyh@redhat.com>
18172 * value-query.cc (range_query::get_value_range): Add assert.
18173 * vr-values.cc (vr_values::range_of_expr): Make sure we don't ICE
18174 on unsupported types in vr_values.
18175 (simplify_using_ranges::simplify): Same.
18177 2022-07-24 Aldy Hernandez <aldyh@redhat.com>
18179 * tree-ssanames.cc (get_nonzero_bits): Return -1 for unsupported
18181 * value-query.cc (get_ssa_name_range_info): Remove precision check.
18183 2022-07-24 Aldy Hernandez <aldyh@redhat.com>
18185 * value-relation.cc (value_relation::set_relation): Remove assert.
18186 (path_oracle::register_relation): Exit when trying to register
18187 same SSA name relations.
18189 2022-07-24 Aldy Hernandez <aldyh@redhat.com>
18191 * gimple-range-cache.cc (ranger_cache::edge_range): Convert to vrange.
18192 (ranger_cache::range_from_dom): Same.
18194 (dom_opt_dom_walker::set_global_ranges_from_unreachable_edges): Same.
18196 2022-07-24 Roger Sayle <roger@nextmovesoftware.com>
18197 H.J. Lu <hjl.tools@gmail.com>
18201 * config/i386/i386-features.cc (make_vector_copies): Move from
18202 general_scalar_chain to scalar_chain.
18203 (convert_reg): Likewise.
18204 (convert_insn_common): New scalar_chain method split out from
18205 general_scalar_chain convert_insn.
18206 (convert_registers): Move from general_scalar_chain to
18208 (scalar_chain::convert): Call convert_insn_common before calling
18210 (timode_remove_non_convertible_regs): Iterate until there are
18211 no further changes to the candidates.
18212 * config/i386/i386-features.h (scalar_chain::hash_map): Move
18213 from general_scalar_chain.
18214 (scalar_chain::convert_reg): Likewise.
18215 (scalar_chain::convert_insn_common): New shared method.
18216 (scalar_chain::make_vector_copies): Move from general_scalar_chain.
18217 (scalar_chain::convert_registers): Likewise. No longer virtual.
18218 (general_scalar_chain::hash_map): Delete. Moved to scalar_chain.
18219 (general_scalar_chain::convert_reg): Likewise.
18220 (general_scalar_chain::make_vector_copies): Likewise.
18221 (general_scalar_chain::convert_registers): Delete virtual method.
18222 (timode_scalar_chain::convert_registers): Likewise.
18224 2022-07-23 Immad Mir <mirimmad@outlook.com>
18226 * doc/extend.texi: Add fd_arg, fd_arg_read and fd_arg_write under
18227 "Common Function Attributes" section.
18228 * doc/invoke.texi: Add docs to -Wanalyzer-fd-access-mode-mismatch,
18229 -Wanalyzer-use-after-close, -Wanalyzer-fd-use-without-check that these
18230 warnings may be emitted through usage of three function attributes used
18231 for static analysis of file descriptors namely fd_arg, fd_arg_read and
18234 2022-07-22 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
18236 * config/xtensa/xtensa.md (*masktrue_const_bitcmpl):
18237 Add a new insn_and_split pattern, and a few split patterns for
18240 2022-07-22 Richard Sandiford <richard.sandiford@arm.com>
18242 * graphds.cc (graphds_scc): Fix algorithm attribution.
18244 2022-07-22 Richard Biener <rguenther@suse.de>
18246 PR tree-optimization/106403
18247 * tree-ssa-sccvn.cc (vn_reference_lookup_3): Move stored
18248 value valueization after check for IFN_MASKED_STORE or
18251 2022-07-22 Richard Biener <rguenther@suse.de>
18253 PR tree-optimization/106397
18254 * tree-ssa-loop-prefetch.cc (emit_mfence_after_loop): Do
18255 not update SSA form here.
18256 (mark_nontemporal_stores): Return whether we marked any
18257 non-temporal stores and inserted mfence.
18258 (loop_prefetch_arrays): Note when we need to update SSA.
18259 (tree_ssa_prefetch_arrays): Perform required (LC) SSA update
18260 at the end of the pass.
18262 2022-07-22 Richard Biener <rguenther@suse.de>
18264 PR tree-optimization/106387
18265 * tree-vect-stmts.cc (vectorizable_load): Use make_ssa_name
18266 if ptr is not an SSA name.
18268 2022-07-22 liuhongt <hongtao.liu@intel.com>
18271 * config/i386/mmx.md (<code><mode>3): New define_expand, it's
18272 original "<code><mode>3".
18273 (*<code><mode>3): New define_insn, it's original
18274 "<code><mode>3" be extended to handle memory and immediate
18275 operand with ix86_binary_operator_ok. Also adjust define_split
18277 (mmxinsnmode): New mode attribute.
18278 (*mov<mode>_imm): Refactor with mmxinsnmode.
18279 * config/i386/predicates.md
18280 (register_or_x86_64_const_vector_operand): New predicate.
18282 2022-07-22 Will Schmidt <will_schmidt@vnet.ibm.com>
18284 * config/rs6000/altivec.md:
18285 (vstrir_code_<mode>): Rename to...
18286 (vstrir_direct_<mode>): ... this.
18287 (vstrir_p_code_<mode>): Rename to...
18288 (vstrir_p_direct_<mode>): ... this.
18289 (vstril_code_<mode>): Rename to...
18290 (vstril_direct_<mode>): ... this.
18291 (vstril_p_code_<mode>): Rename to...
18292 (vstril_p_direct_<mode>): ... this.
18294 2022-07-22 Will Schmidt <will_schmidt@vnet.ibm.com>
18296 * config/rs6000/rs6000-c.cc: Update comments.
18297 (rs6000_target_modify_macros): Remove bu_mask references.
18298 (rs6000_define_or_undefine_macro): Replace bu_mask reference
18299 with a rs6000_cpu value check.
18300 (rs6000_cpu_cpp_builtins): Remove rs6000_builtin_mask_calculate()
18301 parameter from call to rs6000_target_modify_macros.
18302 * config/rs6000/rs6000-protos.h (rs6000_target_modify_macros,
18303 rs6000_target_modify_macros_ptr): Remove parameter from extern
18305 * config/rs6000/rs6000.cc (rs6000_target_modify_macros_ptr): Remove
18306 parameter from prototype, update calls to this function.
18307 (rs6000_print_builtin_options): Remove prototype, call and function.
18308 (rs6000_builtin_mask_calculate): Remove function.
18309 (rs6000_debug_reg_global): Remove call to rs6000_print_builtin_options.
18310 (rs6000_option_override_internal): Remove rs6000_builtin_mask var
18311 and builtin_mask debug output.
18312 (rs6000_builtin_mask_names): Remove.
18313 (rs6000_pragma_target_parse): Remove prev_bumask, cur_bumask,
18314 diff_bumask references; Update calls to rs6000_target_modify_ptr.
18315 * config/rs6000/rs6000.opt (rs6000_builtin_mask): Remove.
18317 2022-07-21 Sam Feifer <sfeifer@redhat.com>
18319 PR tree-optimization/94920
18320 * match.pd (x >= 0 ? x : 0) + (x <= 0 ? -x : 0): New simplification.
18321 (x <= 0 ? -x : 0): New simplification.
18323 2022-07-21 Prathamesh Kulkarni <prathamesh.kulkarni@linaro.org>
18326 2022-07-19 Prathamesh Kulkarni <prathamesh.kulkarni@linaro.org>
18328 * tree-ssa-forwprop.cc (simplify_permutation): Use lhs type
18329 instead of TREE_TYPE (arg0) as result type in folding VEC_PERM_EXPR.
18331 2022-07-21 Richard Biener <rguenther@suse.de>
18333 PR tree-optimization/106379
18334 * match.pd (~(a ^ b) -> a == b): New pattern.
18336 2022-07-21 Richard Biener <rguenther@suse.de>
18338 PR tree-optimization/106378
18339 * tree-ssa-dse.cc (initialize_ao_ref_for_dse): Handle
18340 LEN_STORE, add mode to initialize a may-def and handle
18341 MASK_STORE that way.
18342 (dse_optimize_stmt): Query may-defs. Handle internal
18343 functions LEN_STORE and MASK_STORE similar to how
18344 we handle memory builtins but without byte tracking.
18346 2022-07-21 Richard Biener <rguenther@suse.de>
18348 PR tree-optimization/106365
18349 * tree-ssa-sccvn.cc (pd_data::rhs_off): New field determining
18350 the offset to start encoding of RHS from.
18351 (vn_walk_cb_data::vn_walk_cb_data): Initialize it.
18352 (vn_walk_cb_data::push_partial_def): Allow the first partial
18353 definition to be fully providing the def. Offset RHS
18354 before encoding if requested.
18355 (vn_reference_lookup_3): Initialize def_rhs everywhere.
18356 Add support for .MASK_STORE and .LEN_STORE (partial) definitions.
18358 2022-07-21 Richard Biener <rguenther@suse.de>
18360 * tree-ssa-alias.cc (ref_maybe_used_by_call_p_1): Special-case
18361 store internal functions and IFN_MASK_LOAD, IFN_LEN_LOAD
18362 and IFN_MASK_LOAD_LANES.
18363 (call_may_clobber_ref_p_1): Special-case IFN_MASK_STORE,
18364 IFN_LEN_STORE and IFN_MASK_STORE_LANES.
18366 2022-07-20 Alexander Monakov <amonakov@ispras.ru>
18368 PR rtl-optimization/101347
18369 * builtins.cc (expand_builtin) [BUILT_IN_SETJMP_SETUP]: Move
18370 population of nonlocal_goto_handler_labels from here ...
18371 (expand_builtin) [BUILT_IN_SETJMP_RECEIVER]: ... to here.
18372 * rtlanal.cc (remove_node_from_insn_list): Verify that a
18373 duplicate is not present in the remainder of the list.
18375 2022-07-20 Alexander Monakov <amonakov@ispras.ru>
18377 * rtl.h (remove_node_from_expr_list): Remove declaration.
18378 * rtlanal.cc (remove_node_from_expr_list): Remove (no uses).
18380 2022-07-20 Richard Biener <rguenther@suse.de>
18382 * tree-vect-data-refs.cc (bump_vector_ptr): Return an
18383 invariant updated address when the input was invariant.
18385 2022-07-20 liuhongt <hongtao.liu@intel.com>
18387 * passes.def: (Split pass_cse_sincos to pass_expand_powcabs
18388 and pass_cse_sincos, and move pass_cse_sincos after vectorizer).
18389 * timevar.def (TV_TREE_POWCABS): New timevar.
18390 * tree-pass.h (make_pass_expand_powcabs): Split from pass_cse_sincos.
18391 * tree-ssa-math-opts.cc (gimple_expand_builtin_cabs): Ditto.
18392 (class pass_expand_powcabs): Ditto.
18393 (pass_expand_powcabs::execute): Ditto.
18394 (make_pass_expand_powcabs): Ditto.
18395 (pass_cse_sincos::execute): Remove pow/cabs expand part.
18396 (make_pass_cse_sincos): Ditto.
18398 2022-07-20 Richard Biener <richard.guenther@gmail.com>
18399 Hongtao Liu <hongtao.liu@intel.com>
18401 PR tree-optimization/106010
18402 * tree-complex.cc (init_dont_simulate_again): Lower complex
18404 (expand_complex_move): Also expand COMPLEX_CST for rhs.
18406 2022-07-19 Andrew MacLeod <amacleod@redhat.com>
18408 * gimple-range-cache.cc (ranger_cache::range_from_dom): Check
18409 for incoming ranges on join nodes and add to worklist.
18411 2022-07-19 Andrew MacLeod <amacleod@redhat.com>
18413 * gimple-range-cache.cc (ranger_cache::resolve_dom): New.
18414 (ranger_cache::range_from_dom): Put all nodes to be calculated
18415 in the worklist and resolve after the dom walk.
18416 * gimple-range-cache.h (resolve_dom): New prototype.
18418 2022-07-19 Alexander Monakov <amonakov@ispras.ru>
18420 * tree-cfg.cc (gimple_verify_flow_info): Check placement of
18421 returns_twice calls.
18423 2022-07-19 Alexander Monakov <amonakov@ispras.ru>
18425 * cfghooks.cc (duplicate_block): Expand comment.
18426 * tree-cfg.cc (gimple_can_duplicate_bb_p): Reject blocks with
18427 calls that may return twice.
18429 2022-07-19 Alexander Monakov <amonakov@ispras.ru>
18431 * tree-ssa-sink.cc (select_best_block): Punt if selected block
18432 has incoming abnormal edges.
18434 2022-07-19 Martin Liska <mliska@suse.cz>
18436 * doc/extend.texi: Remove trailing :.
18438 2022-07-19 Prathamesh Kulkarni <prathamesh.kulkarni@linaro.org>
18440 * tree-ssa-forwprop.cc (simplify_permutation): Use lhs type
18441 instead of TREE_TYPE (arg0) as result type in folding VEC_PERM_EXPR.
18443 2022-07-19 Richard Biener <rguenther@suse.de>
18445 PR middle-end/106331
18446 * builtins.cc (get_memory_rtx): Compute alignment from
18447 the original address and set MEM_OFFSET to unknown when
18448 we create a MEM_EXPR from the base object of the address.
18450 2022-07-19 Richard Biener <rguenther@suse.de>
18453 * dwarf2out.cc (dwarf2out_register_external_die): Allow
18454 map entry re-use during WPA.
18456 2022-07-19 Roger Sayle <roger@nextmovesoftware.com>
18457 Richard Biener <rguenther@suse.de>
18460 * builtins.cc (fold_builtin_frexp): Call suppress_warning on
18461 COMPOUND_EXPR to silence spurious warning if result isn't used.
18462 (fold_builtin_modf): Likewise.
18463 (do_mpfr_remquo): Likewise.
18465 2022-07-19 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
18467 * config/xtensa/xtensa.cc (xtensa_rtx_costs):
18468 Change the relative cost of '(set (reg) (const_int N))' where
18469 N fits into signed 12-bit from 4 to 0 if optimizing for size.
18470 And use the appropriate macro instead of the bare number 4.
18472 2022-07-18 Andrew MacLeod <amacleod@redhat.com>
18474 PR tree-optimization/106280
18475 * value-relation.cc (dom_oracle::register_relation): Register
18476 transitives only when it is possible for there to be one.
18477 (dom_oracle::set_one_relation): Return NULL if this is an
18480 2022-07-18 Maciej W. Rozycki <macro@embecosm.com>
18482 * doc/invoke.texi (RISC-V Options): Add index references for
18483 `mrelax' and `mriscv-attribute'.
18485 2022-07-18 Maciej W. Rozycki <macro@embecosm.com>
18487 * doc/invoke.texi (Option Summary): Add missing second space
18488 around `-mstack-protector-guard-reg='.
18490 2022-07-18 Maciej W. Rozycki <macro@embecosm.com>
18492 * doc/invoke.texi (Option Summary): Fix `-mno-riscv-attribute'.
18493 (RISC-V Options): Likewise, and `-mriscv-attribute'.
18495 2022-07-18 Claudiu Zissulescu <claziss@gmail.com>
18497 * config/arc/arc-arch.h (arc_tune_attr): Add
18498 ARC_TUNE_ARCHS4X_REL31A variant.
18499 * config/arc/arc.cc (arc_override_options): Tune options for
18501 (arc_sched_issue_rate): Use correct enum.
18502 (arc600_corereg_hazard): Textual change.
18503 (arc_hazard): Add release 310a tunning.
18504 * config/arc/arc.md (tune): Update and take into consideration new
18506 (tune_dspmpy): Likewise.
18507 (tune_store): New attribute.
18508 * config/arc/arc.opt (mtune): New tune option.
18509 * config/arc/arcHS4x.md (hs4x_brcc0, hs4x_brcc1): New cpu units.
18510 (hs4x_brcc_op): New instruction rezervation.
18511 (hs4x_data_store_1_op): Likewise.
18512 * config/arc/arc-cpus.def (hs4x_rel31): New cpu variant.
18513 * config/arc/arc-tables.opt: Regenerate.
18514 * config/arc/t-multilib: Likewise.
18515 * doc/invoke.texi (ARC): Update mcpu and tune sections.
18517 2022-07-18 Richard Biener <rguenther@suse.de>
18519 * tree-loop-distribution.cc (loop_distribution::distribute_loop):
18520 When computing cost-based merging do not disregard builtin
18521 classified partitions in some cases.
18523 2022-07-18 Richard Sandiford <richard.sandiford@arm.com>
18526 * config/arm/arm-builtins.cc (arm_builtin_vectorized_function):
18528 * config/arm/arm-protos.h (arm_builtin_vectorized_function): Delete.
18529 * config/arm/arm.cc (TARGET_VECTORIZE_BUILTIN_VECTORIZED_FUNCTION):
18531 * config/arm/arm_neon_builtins.def (copysignf): Delete.
18532 * config/arm/iterators.md (nvrint_pattern): New attribute.
18533 * config/arm/neon.md (<NEON_VRINT:nvrint_pattern><VCVTF:mode>2):
18535 (l<NEON_VCVT:nvrint_pattern><su_optab><VCVTF:mode><v_cmp_result>2):
18537 (neon_copysignf<mode>): Rename to...
18538 (copysign<mode>3): ...this.
18540 2022-07-18 Claudiu Zissulescu <claziss@gmail.com>
18542 * config/arc/arc.cc (arc_expand_epilogue): Adjust the frame
18543 pointer first when in interrupts.
18545 2022-07-18 Richard Biener <rguenther@suse.de>
18547 * tree-loop-distribution.cc (copy_loop_before): Add
18548 the ability to replace the original LC PHI defs.
18549 (generate_loops_for_partition): Pass through a flag
18550 whether to redirect original LC PHI defs.
18551 (generate_code_for_partition): Likewise.
18552 (loop_distribution::distribute_loop): Compute the partition
18553 that should provide the LC PHI defs for common reductions
18554 and pass that down.
18556 2022-07-18 Richard Ball <richard.ball@arm.com>
18558 * config/aarch64/aarch64.cc (aarch64_evpc_trn): Use std:swap.
18559 (aarch64_evpc_uzp): Likewise.
18560 (aarch64_evpc_zip): Likewise.
18562 2022-07-18 Roger Sayle <roger@nextmovesoftware.com>
18565 * config/i386/i386.md (*ctzsidi2_<s>ext): New insn_and_split
18566 to recognize any_extend:DI of ctz:SI which is implicitly extended.
18567 (*ctzsidi2_<s>ext_falsedep): New define_insn to model a DImode
18568 extended ctz:SI that has preceding xor to break false dependency.
18570 2022-07-18 Roger Sayle <roger@nextmovesoftware.com>
18572 * config/i386/predicates.md (x86_64_const_vector_operand):
18573 Check the operand's mode matches the specified mode argument.
18575 2022-07-18 Roger Sayle <roger@nextmovesoftware.com>
18577 * config/i386/sse.md (kunpckhi): Add UNSPEC_MASKOP unspec.
18578 (kunpcksi): Likewise, add UNSPEC_MASKOP unspec.
18579 (kunpckdi): Likewise, add UNSPEC_MASKOP unspec.
18580 (vec_pack_trunc_qi): Update to specify the now required
18581 UNSPEC_MASKOP unspec.
18582 (vec_pack_trunc_<mode>): Likewise.
18584 2022-07-16 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
18586 * config/xtensa/xtensa.md
18587 (*masktrue_const_pow2_minus_one, *masktrue_const_negative_pow2,
18588 *masktrue_const_shifted_mask): If the immediate for bitwise AND is
18589 represented as '-(1 << N)', decrease the lower bound of N from 12
18590 to 1. And the other immediate for conditional branch is now no
18591 longer limited to zero, but also one of some positive integers.
18592 Finally, remove the checks of some conditions, because the comparison
18593 expressions that don't satisfy such checks are determined as
18594 compile-time constants and thus will be optimized away before
18597 2022-07-16 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
18599 * config/xtensa/xtensa.cc (xtensa_emit_constantsynth): Remove.
18600 (xtensa_constantsynth_2insn): Change to try all three synthetic
18601 methods and to use the one that fits the immediate value of
18602 the seed into a Narrow Move Immediate instruction "MOVI.N"
18603 when the Code Density Option is configured.
18605 2022-07-15 H.J. Lu <hjl.tools@gmail.com>
18608 * config/i386/i386.cc (ix86_function_ok_for_sibcall): Return
18609 false if callee has indirect_return attribute and caller
18612 2022-07-15 Roger Sayle <roger@nextmovesoftware.com>
18615 * config/i386/i386.md (*andn<dwi>3_doubleword_bmi): Update the
18616 constraints to reflect the output is earlyclobber, unless it is
18617 the same register (pair) as one of the operands.
18619 2022-07-15 David Malcolm <dmalcolm@redhat.com>
18621 * doc/invoke.texi (Static Analyzer Options): Add the new fd
18622 warnings to the initial gccoptlist, and to the list of those
18623 disabled by -fanalyzer-checker=taint.
18625 2022-07-15 Andrew Carlotti <andrew.carlotti@arm.com>
18627 * config/aarch64/aarch64-builtins.cc
18628 (enum aarch64_type_qualifiers): Remove qualifier_internal.
18629 (aarch64_init_simd_builtin_functions): Remove qualifier_internal check.
18631 2022-07-15 Andrew Carlotti <andrew.carlotti@arm.com>
18633 * config/aarch64/aarch64-builtins.cc
18634 (v1di_UP): Add V1DI mode to _UP macros.
18635 * config/aarch64/aarch64-modes.def (VECTOR_MODE): Add V1DI mode.
18636 * config/aarch64/aarch64-simd-builtin-types.def: Use V1DI mode.
18637 * config/aarch64/aarch64-simd.md
18638 (vec_extractv2dfv1df): Replace with...
18639 (vec_extract<mode><V1half>): ...this.
18640 * config/aarch64/aarch64.cc
18641 (aarch64_classify_vector_mode): Add V1DI mode.
18642 * config/aarch64/iterators.md
18643 (VQ_2E, V1HALF, V1half): New.
18644 (nunits): Add V1DI mode.
18646 2022-07-15 Roger Sayle <roger@nextmovesoftware.com>
18649 * config/i386/i386-features.cc (general_scalar_chain::convert_insn):
18650 Fix indentation whitespace.
18651 (timode_scalar_chain::fix_debug_reg_uses): Likewise.
18652 (timode_scalar_chain::convert_insn): Delete dead code.
18653 Update TImode REG_EQUAL_NOTE even if the SET_DEST is already V1TI.
18654 Fix indentation whitespace.
18655 (convertible_comparison_p): Likewise.
18656 (timode_scalar_to_vector_candidate_p): Likewise.
18658 2022-07-15 Aldy Hernandez <aldyh@redhat.com>
18660 * gimple-pretty-print.cc (dump_ssaname_info): Use pp_vrange.
18662 2022-07-15 Aldy Hernandez <aldyh@redhat.com>
18664 * Makefile.in (OBJS): Add value-range-pretty-print.o.
18665 * pretty-print.h (pp_vrange): New.
18666 * value-range.cc (vrange::dump): Call pp version.
18667 (unsupported_range::dump): Move to its own file.
18668 (dump_bound_with_infinite_markers): Same.
18669 (irange::dump): Same.
18670 (irange::dump_bitmasks): Same.
18671 (vrange::debug): Remove.
18672 * value-range.h: Remove virtual designation for dump methods.
18673 Remove dump_bitmasks method.
18674 * value-range-pretty-print.cc: New file.
18675 * value-range-pretty-print.h: New file.
18677 2022-07-15 Aldy Hernandez <aldyh@redhat.com>
18679 * value-range.cc (irange::accept): New.
18680 (unsupported_range::accept): New.
18681 * value-range.h (class vrange_visitor): New.
18682 (class vrange): Add accept method.
18683 (class unsupported_range): Same.
18684 (class Value_Range): Same.
18686 2022-07-15 Jonathan Wakely <jwakely@redhat.com>
18688 * diagnostic-format-json.cc (json_from_location_range): Adjust
18689 to new label_text API.
18690 * diagnostic-format-sarif.cc (sarif_builder::make_location_object):
18692 * diagnostic-show-locus.cc (struct pod_label_text): Likewise.
18693 (layout::print_any_labels): Likewise.
18694 * tree-diagnostic-path.cc (class path_label): Likewise.
18695 (struct event_range): Likewise.
18696 (default_tree_diagnostic_path_printer): Likewise.
18697 (default_tree_make_json_for_path): Likewise.
18699 2022-07-15 konglin1 <lingling.kong@intel.com>
18702 * config/i386/i386-builtin.def (BDESC): Fix [u]comi{ss,sd}
18703 comparison due to intrinsics changed over time.
18704 * config/i386/i386-expand.cc (ix86_ssecom_setcc):
18705 Add unordered check and mode for sse comi codegen.
18706 (ix86_expand_sse_comi): Add unordered check and check a different
18708 (ix86_expand_sse_comi_round):Extract unordered check and mode part
18709 in ix86_ssecom_setcc.
18711 2022-07-15 Prathamesh Kulkarni <prathamesh.kulkarni@linaro.org>
18713 * config/aarch64/aarch64.cc (aarch64_vectorize_vec_perm_const): Use
18714 op_mode instead of vmode in calls to force_reg for op0 and op1.
18716 2022-07-14 H.J. Lu <hjl.tools@gmail.com>
18718 PR tree-optimization/103798
18719 * tree-ssa-forwprop.cc: Include "tree-ssa-strlen.h".
18720 (simplify_builtin_call): Inline memchr with constant strings of
18721 no more than the bytes of a word.
18722 * tree-ssa-strlen.cc (use_in_zero_equality): Make it global.
18723 * tree-ssa-strlen.h (use_in_zero_equality): New.
18725 2022-07-14 Eric Botcazou <ebotcazou@adacore.com>
18727 * gimplify.cc (lookup_tmp_var): Add NOT_GIMPLE_REG boolean parameter
18728 and set DECL_NOT_GIMPLE_REG_P on the variable according to it.
18729 (internal_get_tmp_var): Add NOT_GIMPLE_REG boolean parameter and
18730 pass it in the call to lookup_tmp_var.
18731 (get_formal_tmp_var): Pass false in the call to lookup_tmp_var.
18732 (get_initialized_tmp_var): Likewise.
18733 (prepare_gimple_addressable): Call internal_get_tmp_var instead of
18734 get_initialized_tmp_var with NOT_GIMPLE_REG set to true.
18736 2022-07-14 Martin Liska <mliska@suse.cz>
18738 * doc/gimple.texi: Close properly a deftypefn.
18740 2022-07-14 Martin Liska <mliska@suse.cz>
18742 * doc/gimple.texi: Close properly a deftypefn.
18744 2022-07-14 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
18746 * config/xtensa/xtensa.md:
18747 In FP constant synthesis split pattern, subcontract to
18748 avoid_constant_pool_reference() as in the case of integer,
18749 because it can handle well too. And cast to int32_t before
18750 calling xtensa_constantsynth() in order to ignore upper 32-bit.
18752 2022-07-13 Aldy Hernandez <aldyh@redhat.com>
18754 * range-op.cc (operator_lt::fold_range): Use nonzero bits.
18756 2022-07-13 Aldy Hernandez <aldyh@redhat.com>
18758 * value-range.cc (irange::copy_to_legacy): Set nonzero mask.
18759 (irange::legacy_intersect): Clear nonzero mask.
18760 (irange::legacy_union): Same.
18761 (irange::invert): Same.
18763 2022-07-13 Richard Biener <rguenther@suse.de>
18765 * tree-ssa-dom.h (record_temporary_equivalences): Remove.
18766 * tree-ssa-dom.cc (dom_jt_state::m_blocks_on_stack): New.
18767 (dom_jt_state::get_blocks_on_stack): Likewise.
18768 (dom_opt_dom_walker::dom_opt_dom_walker): Take dom_jt_state.
18769 (back_propagate_equivalences): Remove dominator bitmap
18770 compute and instead use passed in m_blocks_on_stack.
18771 (record_temporary_equivalences): Likewise.
18772 (record_equivalences_from_incoming_edge): Likewise.
18773 (dom_opt_dom_walker::before_dom_children): Maintain and
18774 pass down blocks on stack.
18775 (dom_opt_dom_walker::after_dom_children): Likewise.
18777 2022-07-13 Andrew Carlotti <andrew.carlotti@arm.com>
18779 * config/aarch64/aarch64-builtins.cc
18780 (aarch64_general_gimple_fold_builtin): Add fixup for invalid GIMPLE.
18782 2022-07-13 Richard Biener <rguenther@suse.de>
18784 PR tree-optimization/106249
18785 * tree-ssa-loop-manip.cc (tree_transform_and_unroll_loop):
18786 Only verify LC SSA of the new_loop if we created it.
18787 Use TODO_update_ssa_nophi for the SSA update after versioning
18790 2022-07-12 Aldy Hernandez <aldyh@redhat.com>
18792 * range-op.cc (unsigned_singleton_p): Remove.
18793 (operator_bitwise_and::remove_impossible_ranges): Remove.
18794 (operator_bitwise_and::fold_range): Set nonzero bits. *
18795 * value-range.cc (irange::get_nonzero_bits): Remove
18796 legacy_mode_p assert.
18797 (irange::dump_bitmasks): Remove legacy_mode_p check.
18799 2022-07-12 Richard Sandiford <richard.sandiford@arm.com>
18802 * predict.h (insn_optimization_type): Declare.
18803 * predict.cc (insn_optimization_type): New function.
18804 * internal-fn.def (IFN_ICEIL, IFN_IFLOOR, IFN_IRINT, IFN_IROUND)
18805 (IFN_LCEIL, IFN_LFLOOR, IFN_LRINT, IFN_LROUND, IFN_LLCEIL)
18806 (IFN_LLFLOOR, IFN_LLRINT, IFN_LLROUND): New internal functions.
18807 * internal-fn.cc (unary_convert_direct): New macro.
18808 (expand_convert_optab_fn): New function.
18809 (expand_unary_convert_optab_fn): New macro.
18810 (direct_unary_convert_optab_supported_p): Likewise.
18811 * optabs.cc (expand_sfix_optab): Pass insn_optimization_type to
18812 convert_optab_handler.
18813 * config/aarch64/aarch64-protos.h
18814 (aarch64_builtin_vectorized_function): Delete.
18815 * config/aarch64/aarch64-builtins.cc
18816 (aarch64_builtin_vectorized_function): Delete.
18817 * config/aarch64/aarch64.cc
18818 (TARGET_VECTORIZE_BUILTIN_VECTORIZED_FUNCTION): Delete.
18819 * config/i386/i386.cc (ix86_optab_supported_p): Handle lround_optab.
18820 * config/i386/i386.md (lround<X87MODEF:mode><SWI248x:mode>2): Remove
18821 optimize_insn_for_size_p test.
18823 2022-07-12 Richard Biener <rguenther@suse.de>
18825 * tree-vect-loop-manip.cc (create_lcssa_for_virtual_phi):
18827 (vect_do_peeling): Do not call it, adjust comment.
18829 2022-07-12 Richard Sandiford <richard.sandiford@arm.com>
18831 * config/aarch64/aarch64-builtins.cc
18832 (aarch64_builtin_vectorized_function): Remove handling of
18833 floor, ceil, trunc, round, nearbyint, sqrt, clz and ctz.
18835 2022-07-11 Andrew MacLeod <amacleod@redhat.com>
18837 PR tree-optimization/106234
18838 * gimple-range-cache.cc (ranger_cache::range_from_dom): Check dominator
18839 cache value before recursively resolving it.
18841 2022-07-11 Roger Sayle <roger@nextmovesoftware.com>
18843 * config/i386/i386-features.h (scalar_chain): Add fields
18844 insns_conv, n_sse_to_integer and n_integer_to_sse to this
18845 parent class, moved from general_scalar_chain.
18846 (scalar_chain::convert_compare): Protected method moved
18847 from general_scalar_chain.
18848 (mark_dual_mode_def): Make protected, not private virtual.
18849 (scalar_chain:convert_op): New private virtual method.
18850 (general_scalar_chain::general_scalar_chain): Simplify constructor.
18851 (general_scalar_chain::~general_scalar_chain): Delete destructor.
18852 (general_scalar_chain): Move insns_conv, n_sse_to_integer and
18853 n_integer_to_sse fields to parent class, scalar_chain.
18854 (general_scalar_chain::mark_dual_mode_def): Delete prototype.
18855 (general_scalar_chain::convert_compare): Delete prototype.
18856 (timode_scalar_chain::compute_convert_gain): Remove simplistic
18857 implementation, convert to a method prototype.
18858 (timode_scalar_chain::mark_dual_mode_def): Delete prototype.
18859 (timode_scalar_chain::convert_op): Prototype new virtual method.
18860 * config/i386/i386-features.cc (scalar_chain::scalar_chain):
18861 Allocate insns_conv and initialize n_sse_to_integer and
18862 n_integer_to_sse fields in constructor.
18863 (scalar_chain::scalar_chain): Free insns_conv in destructor.
18864 (general_scalar_chain::general_scalar_chain): Delete
18865 constructor, now defined in the class declaration.
18866 (general_scalar_chain::~general_scalar_chain): Delete destructor.
18867 (scalar_chain::mark_dual_mode_def): Renamed from
18868 general_scalar_chain::mark_dual_mode_def.
18869 (timode_scalar_chain::mark_dual_mode_def): Delete.
18870 (scalar_chain::convert_compare): Renamed from
18871 general_scalar_chain::convert_compare.
18872 (timode_scalar_chain::compute_convert_gain): New method to
18873 determine the gain from converting a TImode chain to V1TImode.
18874 (timode_scalar_chain::convert_op): New method to convert an
18875 operand from TImode to V1TImode.
18876 (timode_scalar_chain::convert_insn) <case REG>: Only PUT_MODE
18877 on REG_EQUAL notes that were originally TImode (not CONST_INT).
18878 Handle AND, ANDN, XOR, IOR, NOT and COMPARE.
18879 (timode_mem_p): Helper predicate to check where operand is
18880 memory reference with sufficient alignment for TImode STV.
18881 (timode_scalar_to_vector_candidate_p): Use convertible_comparison_p
18882 to check whether COMPARE is convertible. Handle SET_DESTs that
18883 that are REG_P or MEM_P and SET_SRCs that are REG, CONST_INT,
18884 CONST_WIDE_INT, MEM, AND, ANDN, IOR, XOR or NOT.
18886 2022-07-11 Richard Sandiford <richard.sandiford@arm.com>
18888 PR tree-optimization/106250
18889 * tree-vect-loop.cc (vectorizable_reduction): Reinstate final
18890 argument to directly_supported_p.
18892 2022-07-11 Richard Biener <rguenther@suse.de>
18894 * tree-into-ssa.cc (update_ssa): Do not forcefully
18895 re-compute dominance fast queries for TODO_update_ssa_no_phi.
18897 2022-07-11 Richard Biener <rguenther@suse.de>
18899 PR tree-optimization/106228
18900 * tree-vect-data-refs.cc (vect_setup_realignment): Adjust
18901 VUSE compute for the non-loop case.
18903 2022-07-11 Richard Biener <rguenther@suse.de>
18905 * tree-into-ssa.cc (rewrite_mode::REWRITE_UPDATE_REGION): New.
18906 (rewrite_update_dom_walker::rewrite_update_dom_walker): Update.
18907 (rewrite_update_dom_walker::m_in_region_flag): New.
18908 (rewrite_update_dom_walker::before_dom_children): If the region
18909 to update is marked, STOP at exits.
18910 (rewrite_blocks): For REWRITE_UPDATE_REGION mark the region
18912 (dump_update_ssa): Use bitmap_empty_p.
18913 (update_ssa): Likewise. Use REWRITE_UPDATE_REGION when
18914 TODO_update_ssa_no_phi.
18915 * tree-cfgcleanup.cc (cleanup_tree_cfg_noloop): Account
18916 pending update_ssa to the caller.
18918 2022-07-11 Richard Biener <rguenthert@suse.de>
18921 * config/i386/i386-options.cc (ix86_set_current_function):
18922 Rebuild the target optimization node whenever necessary,
18923 not only when the optimization node didn't change.
18925 2022-07-11 Richard Biener <rguenther@suse.de>
18927 PR tree-optimization/106228
18928 * tree-vect-data-refs.cc (vect_setup_realignment): Properly
18929 set a VUSE operand on the emitted load.
18931 2022-07-11 Aldy Hernandez <aldyh@redhat.com>
18933 * gimple-range.cc (gimple_ranger::export_global_ranges): Remove
18934 verification against legacy value_range.
18935 (gimple_ranger::register_inferred_ranges): Same.
18936 (gimple_ranger::export_global_ranges): Rename update_global_range
18938 * tree-core.h (struct range_info_def): Remove.
18939 (struct irange_storage_slot): New.
18940 (struct tree_base): Remove SSA_NAME_ANTI_RANGE_P documentation.
18941 (struct tree_ssa_name): Add vrange_storage support.
18942 * tree-ssanames.cc (range_info_p): New.
18943 (range_info_fits_p): New.
18944 (range_info_alloc): New.
18945 (range_info_free): New.
18946 (range_info_get_range): New.
18947 (range_info_set_range): New.
18948 (set_range_info_raw): Remove.
18949 (set_range_info): Adjust to use vrange_storage.
18950 (set_nonzero_bits): Same.
18951 (get_nonzero_bits): Same.
18952 (duplicate_ssa_name_range_info): Remove overload taking
18954 Rewrite tree overload to use vrange_storage.
18955 (duplicate_ssa_name_fn): Adjust to use vrange_storage.
18956 * tree-ssanames.h (struct range_info_def): Remove.
18957 (set_range_info): Adjust prototype to take vrange.
18958 * tree-vrp.cc (vrp_asserts::remove_range_assertions): Call
18959 duplicate_ssa_name_range_info.
18960 * tree.h (SSA_NAME_ANTI_RANGE_P): Remove.
18961 (SSA_NAME_RANGE_TYPE): Remove.
18962 * value-query.cc (get_ssa_name_range_info): Adjust to use
18964 (update_global_range): Remove.
18965 (get_range_global): Remove as_a<irange>.
18966 * value-query.h (update_global_range): Remove.
18967 * tree-ssa-dom.cc (set_global_ranges_from_unreachable_edges):
18968 Rename update_global_range to set_range_info.
18969 * value-range-storage.cc (vrange_storage::alloc_slot): Remove
18972 2022-07-10 Aldy Hernandez <aldyh@redhat.com>
18974 * value-range.cc (irange::operator=): Call verify_range.
18975 (irange::irange_set): Normalize kind after everything else has
18977 (irange::irange_set_anti_range): Same.
18978 (irange::set): Same.
18979 (irange::verify_range): Disallow nonzero masks for VARYING.
18980 (irange::irange_union): Call verify_range.
18981 Handle nonzero masks better.
18982 (irange::irange_intersect): Same.
18983 (irange::set_nonzero_bits): Calculate mask if either range has an
18985 (irange::intersect_nonzero_bits): Same.
18986 (irange::union_nonzero_bits): Same.
18987 (range_tests_nonzero_bits): New.
18988 (range_tests): Call range_tests_nonzero_bits.
18989 * value-range.h (class irange): Remove set_nonzero_bits method
18991 (irange::varying_compatible_p): Set nonzero mask.
18993 2022-07-10 Xi Ruoyao <xry111@xry111.site>
18995 * config/loongarch/loongarch.md (<any_div>di3_fake): Describe
18996 the sign-extend of result in the RTL template.
18997 (<any_div><mode>3): Adjust for <any_div>di3_fake change.
18999 2022-07-10 Xi Ruoyao <xry111@xry111.site>
19001 * config/loongarch/loongarch.cc (loongarch_check_zero_div_p):
19002 Remove static, for use in the machine description file.
19003 * config/loongarch/loongarch-protos.h:
19004 (loongarch_check_zero_div_p): Add prototype.
19005 * config/loongarch/loongarch.md (enabled): New attr.
19006 (*<optab><mode>3): Add (=r,r,r) and (=&r,0,r) alternatives for
19007 idiv. Conditionally enable the alternatives using
19008 loongarch_check_zero_div_p.
19009 (<optab>di3_fake): Likewise.
19011 2022-07-10 Xi Ruoyao <xry111@xry111.site>
19013 * config/loongarch/loongarch.md (mulsidi3_64bit): Use mulw.d.w
19016 2022-07-09 Aldy Hernandez <aldyh@redhat.com>
19018 * value-range.cc (irange::irange_single_pair_union): Set
19019 VR_VARYING when appropriate.
19021 2022-07-09 Vit Kabele <vit.kabele@sysgo.com>
19023 * stor-layout.cc (finalize_record_size): Extend warning message.
19025 2022-07-09 Sam Feifer <sfeifer@redhat.com>
19027 PR tree-optimization/98304
19028 * match.pd (n - (((n > C1) ? n : C1) & -C2)): New simplification.
19030 2022-07-09 Jeff Law <jeffreyalaw@gmail.com>
19032 * expr.cc (store_expr): Identify trailing NULs in a STRING_CST
19033 initializer and use clear_storage rather than copying the
19034 NULs to the destination array.
19036 2022-07-09 Roger Sayle <roger@nextmovesoftware.com>
19038 * config/i386/i386.md (define_peephole2): Use match_operand of
19039 flags_reg_operand to capture and preserve the mode of FLAGS_REG.
19040 (define_peephole2): Likewise.
19041 (define_peephole2): Likewise...
19043 2022-07-09 Roger Sayle <roger@nextmovesoftware.com>
19045 * config/i386/i386-features.cc (convert_compare): Add support
19046 for *testdi_not_doubleword pattern, "(compare (and (not ...)))"
19047 by generating a pandn followed by ptest.
19048 (convertible_comparison_p): Recognize both *cmpdi_doubleword and
19049 recent *testdi_not_doubleword comparison patterns.
19051 2022-07-09 Tamar Christina <tamar.christina@arm.com>
19053 * config/s390/s390.cc (s390_expand_atomic): Pass false to store_bit_field to
19054 indicate that the value is not undefined.
19056 2022-07-09 Andrew Pinski <apinski@marvell.com>
19058 PR tree-optimization/106087
19059 * tree-ssa-dce.cc (simple_dce_from_worklist): Check
19060 to make sure the statement is only defining one operand.
19062 2022-07-08 Lewis Hyatt <lhyatt@gmail.com>
19064 PR preprocessor/91733
19065 * input.cc (find_end_of_line): New helper function.
19066 (file_cache_slot::get_next_line): Recognize \r as a line ending.
19067 * diagnostic-show-locus.cc (test_escaping_bytes_1): Adapt selftest
19068 since \r will now be interpreted as a line-ending.
19070 2022-07-08 Martin Liska <mliska@suse.cz>
19072 PR sanitizer/106132
19073 * opts.cc (finish_options): Use 2 calls to
19074 report_conflicting_sanitizer_options.
19076 2022-07-08 Richard Biener <rguenther@suse.de>
19078 PR tree-optimization/106226
19079 * tree-vect-loop-manip.cc (vect_do_peeling): Assert that
19080 no SSA update is needed. Move virtual SSA update ...
19081 * tree-vectorizer.cc (pass_vectorize::execute): ... here,
19082 via forced virtual renaming when TODO_update_ssa_only_virtuals
19084 (vect_transform_loops): Return TODO_update_ssa_only_virtuals
19085 when virtual SSA update is required.
19086 (try_vectorize_loop_1): Adjust.
19087 * tree-vect-stmts.cc (vectorizable_simd_clone_call): Allow
19088 virtual renaming if the ABI forces an aggregate return
19089 but the original call did not have a virtual definition.
19091 2022-07-08 Martin Liska <mliska@suse.cz>
19093 * toplev.cc (init_asm_output): Do not init asm_out_file.
19095 2022-07-08 Tamar Christina <tamar.christina@arm.com>
19097 PR tree-optimization/106063
19098 * match.pd: Do not apply pattern after veclower is not supported.
19100 2022-07-08 Thomas Schwinge <thomas@codesourcery.com>
19102 * lto-cgraph.cc (input_offload_tables) <LTO_symtab_edge>: Correct
19105 2022-07-08 Tamar Christina <tamar.christina@arm.com>
19107 * expmed.cc (store_bit_field_1): Add parameter that indicates if value is
19108 still undefined and if so emit a subreg move instead.
19109 (store_integral_bit_field): Likewise.
19110 (store_bit_field): Likewise.
19111 * expr.h (write_complex_part): Likewise.
19112 * expmed.h (store_bit_field): Add new parameter.
19113 * builtins.cc (expand_ifn_atomic_compare_exchange_into_call): Use new
19115 (expand_ifn_atomic_compare_exchange): Likewise.
19116 * calls.cc (store_unaligned_arguments_into_pseudos): Likewise.
19117 * emit-rtl.cc (validate_subreg): Likewise.
19118 * expr.cc (emit_group_store): Likewise.
19119 (copy_blkmode_from_reg): Likewise.
19120 (copy_blkmode_to_reg): Likewise.
19121 (clear_storage_hints): Likewise.
19122 (write_complex_part): Likewise.
19123 (emit_move_complex_parts): Likewise.
19124 (expand_assignment): Likewise.
19125 (store_expr): Likewise.
19126 (store_field): Likewise.
19127 (expand_expr_real_2): Likewise.
19128 * ifcvt.cc (noce_emit_move_insn): Likewise.
19129 * internal-fn.cc (expand_arith_set_overflow): Likewise.
19130 (expand_arith_overflow_result_store): Likewise.
19131 (expand_addsub_overflow): Likewise.
19132 (expand_neg_overflow): Likewise.
19133 (expand_mul_overflow): Likewise.
19134 (expand_arith_overflow): Likewise.
19136 2022-07-08 Haochen Jiang <haochen.jiang@intel.com>
19139 * config/i386/sse.md (sse2_cvtps2pd<mask_name>_1):
19140 Rename from *sse2_cvtps2pd<mask_name>_1.
19141 (vec_unpacks_lo_v4sf): Add handler for memory operand.
19143 2022-07-08 Lulu Cheng <chenglulu@loongson.cn>
19145 * config/loongarch/loongarch.cc (loongarch_compute_frame_info):
19146 Modify fp_sp_offset and gp_sp_offset's calculation method,
19147 when frame->mask or frame->fmask is zero, don't minus UNITS_PER_WORD
19148 or UNITS_PER_FP_REG.
19150 2022-07-07 David Malcolm <dmalcolm@redhat.com>
19152 * diagnostic-format-json.cc (json_from_location_range): Update for
19153 removal of label_text::maybe_free in favor of automatic memory
19155 * diagnostic-format-sarif.cc
19156 (sarif_builder::make_location_object): Likewise.
19157 * diagnostic-show-locus.cc (struct pod_label_text): New.
19158 (class line_label): Convert m_text from label_text to pod_label_text.
19159 (layout::print_any_labels): Move "text" to the line_label.
19160 * tree-diagnostic-path.cc (path_label::get_text): Update for
19161 removal of label_text::maybe_free in favor of automatic memory
19163 (event_range::print): Likewise.
19164 (default_tree_diagnostic_path_printer): Likewise.
19165 (default_tree_make_json_for_path): Likewise.
19167 2022-07-07 Prathamesh Kulkarni <prathamesh.kulkarni@linaro.org>
19169 * statistics.cc (get_function_name): Add check to see if fn is not NULL.
19171 2022-07-07 Richard Biener <rguenther@suse.de>
19173 * tree-into-ssa.cc (iterating_old_ssa_names): New.
19174 (add_new_name_mapping): Grow {new,old}_ssa_names separately
19175 and only when actually needed. Assert we are not growing
19176 the old_ssa_names set when iterating over it.
19177 (update_ssa): Remove old_ssa_names copying and empty_p
19178 query, note we are iterating over it and expect no set changes.
19180 2022-07-07 Thomas Schwinge <thomas@codesourcery.com>
19182 * config/i386/intelmic-mkoffload.cc (generate_host_descr_file)
19183 (prepare_target_image, main): Handle OpenMP 'requires'.
19184 (generate_host_descr_file): Switch to 'GOMP_offload_register_ver',
19185 'GOMP_offload_unregister_ver'.
19187 2022-07-07 Richard Biener <rguenther@suse.de>
19190 * config/i386/i386-builtins.cc (ix86_add_new_builtins): Properly
19193 2022-07-07 Richard Biener <rguenther@suse.de>
19195 * tree-ssa-loop-manip.cc (add_exit_phis_var): Return the
19196 number of LC PHIs inserted.
19197 (add_exit_phis): Return whether any variable required
19198 multiple LC PHI nodes.
19199 (rewrite_into_loop_closed_ssa_1): Use TODO_update_ssa_no_phi
19202 2022-07-07 Richard Biener <rguenther@suse.de>
19204 * tree-ssa-loop-manip.cc (compute_live_loop_exits): Take
19205 the def loop exit block bitmap as argument instead of
19206 re-computing it here.
19207 (add_exit_phis_var): Adjust.
19208 (loop_name_cmp): New function.
19209 (add_exit_phis): Sort variables to insert LC PHI nodes
19210 after definition loop, for each definition loop compute
19211 the exit block bitmap once.
19212 (get_loops_exit): Remove.
19213 (rewrite_into_loop_closed_ssa_1): Do not pre-record
19214 all loop exit blocks into bitmaps. Record loop exits
19217 2022-07-07 Dimitrije Milosevic <Dimitrije.Milosevic@Syrmia.com>
19219 * config/mips/mips.cc (mips_asan_shadow_offset): Reformat
19220 to handle the N32 ABI.
19221 * config/mips/mips.h (SUBTARGET_SHADOW_OFFSET): Remove
19222 the macro, as it is not needed anymore.
19224 2022-07-06 Thomas Schwinge <thomas@codesourcery.com>
19226 * config/gcn/mkoffload.cc (process_obj): Clarify 'target_data' ->
19228 * config/nvptx/mkoffload.cc (process): Likewise.
19230 2022-07-06 Thomas Schwinge <thomas@codesourcery.com>
19232 * omp-general.h (enum omp_requires): Use 'GOMP_REQUIRES_[...]'.
19234 2022-07-05 Andrew MacLeod <amacleod@redhat.com>
19236 * value-relation.cc (relation_to_code): New vector.
19237 (relation_oracle::validate_relation): New.
19238 (set_relation): Allow ssa1 == ssa2 to be registered.
19239 * value-relation.h (validate_relation): New prototype.
19240 (query_relation): Make internal variant protected.
19242 2022-07-05 Roger Sayle <roger@nextmovesoftware.com>
19244 * config/i386/i386.cc (ix86_rtx_costs) <COMPARE>: Provide costs
19245 for double word comparisons and tests (comparisons against zero).
19246 * config/i386/i386.md (*test<mode>_not_doubleword): Split DWI
19247 and;cmp into andn;cmp $0 as a pre-reload splitter.
19248 (*andn<dwi>3_doubleword_bmi): Use <dwi> instead of <mode> in name.
19249 (*<any_or><dwi>3_doubleword): Likewise.
19251 2022-07-05 Roger Sayle <roger@nextmovesoftware.com>
19252 Hongtao Liu <hongtao.liu@intel.com>
19254 * config/i386/i386-builtin.def (__builtin_ia32_palignr128): Change
19255 CODE_FOR_ssse3_palignrti to CODE_FOR_ssse3_palignrv1ti.
19256 * config/i386/i386-expand.cc (expand_vec_perm_palignr): Use V1TImode
19257 and gen_ssse3_palignv1ti instead of TImode.
19258 * config/i386/sse.md (SSESCALARMODE): Delete.
19259 (define_mode_attr ssse3_avx2): Handle V1TImode instead of TImode.
19260 (<ssse3_avx2>_palignr<mode>): Use VIMAX_AVX2_AVX512BW as a mode
19261 iterator instead of SSESCALARMODE.
19262 (ssse3_palignrdi): Optimize cases where operands[3] is 0 or 64,
19263 using a single move instruction (if required).
19265 2022-07-05 Roger Sayle <roger@nextmovesoftware.com>
19266 Uroš Bizjak <ubizjak@gmail.com>
19268 PR rtl-optimization/96692
19269 * config/i386/i386.md (define_split): Split ((A | B) ^ C) ^ D
19270 as (X & ~Y) ^ Z on target BMI when either C or D is A or B.
19272 2022-07-05 Richard Biener <rguenther@suse.de>
19274 PR tree-optimization/106198
19275 * tree-cfgcleanup.cc (repair_loop_structures): Always do a
19276 full LC SSA rewrite but only if any blocks changed loop
19279 2022-07-05 Richard Biener <rguenther@suse.de>
19281 * tree-ssa-loop-manip.cc (find_uses_to_rename_def): Remove.
19282 (find_uses_to_rename_in_loop): Likewise.
19283 (rewrite_into_loop_closed_ssa_1): Remove loop parameter and
19285 (rewrite_into_loop_closed_ssa): Adjust.
19287 2022-07-05 Richard Biener <rguenther@suse.de>
19289 PR tree-optimization/106186
19290 * tree-ssa-propagate.cc (clean_up_loop_closed_phi):
19291 Properly handle virtual PHI nodes.
19293 2022-07-05 Richard Biener <rguenther@suse.de>
19295 PR tree-optimization/106196
19296 * tree-vect-stmts.cc (vect_finish_stmt_generation): Properly
19297 handle aggregate returns of calls for VDEF updates.
19299 2022-07-05 Richard Biener <rguenther@suse.de>
19301 * tree-vect-loop-manip.cc (vect_set_loop_condition_normal):
19304 2022-07-05 Richard Sandiford <richard.sandiford@arm.com>
19306 * tree-vect-patterns.cc (vect_convert_input): Expect the input
19307 type to be signed for optab_vector_mixed_sign. Update the vectype
19308 at the same time as type.
19309 (vect_recog_dot_prod_pattern): Update accordingly. If usdot isn't
19310 available, try sdot instead.
19311 * tree-vect-loop.cc (vect_is_emulated_mixed_dot_prod): New function.
19312 (vect_model_reduction_cost): Model the cost of implementing usdot
19314 (vectorizable_reduction): Likewise. Skip target support test
19315 for lane reductions.
19316 (vect_emulate_mixed_dot_prod): New function.
19317 (vect_transform_reduction): Use it to emulate usdot via sdot.
19319 2022-07-05 Richard Biener <rguenther@suse.de>
19321 PR tree-optimization/106182
19322 * loop-init.cc (fix_loop_structure): Return the number
19323 of newly discovered plus the number of deleted loops.
19324 * tree-cfgcleanup.cc (repair_loop_structures): Adjust
19327 2022-07-04 Aldy Hernandez <aldyh@redhat.com>
19329 * gimple-range-fold.cc
19330 (fold_using_range::range_of_ssa_name_with_loop_info): Restrict the
19331 call to SCEV for irange supported types.
19332 (fold_using_range::range_of_builtin_int_call): Convert to vrange.
19333 * gimple-range.cc (gimple_ranger::prefill_stmt_dependencies): Same.
19334 * tree-ssa-dom.cc (cprop_operand): Same.
19336 2022-07-04 Roger Sayle <roger@nextmovesoftware.com>
19339 * calls.cc (precompute_register_parameters): Allow promotion
19340 of floating point values to be passed in wider integer modes
19341 by calling new convert_float_to_wider_int.
19342 (expand_call): Allow floating point results to be returned in
19343 wider integer modes by calling new convert wider_int_to_float.
19344 * cfgexpand.cc (expand_value_return): Allow backends to promote
19345 a scalar floating point return value to a wider integer mode
19346 by calling new convert_float_to_wider_int.
19347 * expr.cc (convert_float_to_wider_int): New function.
19348 (convert_wider_int_to_float): Likewise.
19349 (expand_expr_real_1) <expand_decl_rtl>: Allow backends to promote
19350 scalar FP PARM_DECLs to wider integer modes, by calling new
19351 convert_wider_int_to_float.
19352 * expr.h (convert_modes): Name arguments for improved documentation.
19353 (convert_float_to_wider_int): Prototype new function here.
19354 (convert_wider_int_to_float): Likewise.
19355 * function.cc (assign_parm_setup_stack): Allow floating point
19356 values to be passed on the stack as wider integer modes by
19357 calling new convert_wider_int_to_float.
19359 2022-07-04 Martin Jambor <mjambor@suse.cz>
19361 PR tree-optimization/105860
19362 * tree-sra.cc (build_reconstructed_reference): Start expr
19363 traversal only just below the outermost union.
19365 2022-07-04 Richard Biener <rguenther@suse.de>
19367 * tree-vect-loop-manip.cc (vect_do_peeling): Revert assert
19368 and update virtual SSA form again. Assert we do so for
19369 a known set of transforms only.
19370 * tree-vectorizer.h (vec_info::any_known_not_updated_vssa): New.
19371 * tree-vect-stmts.cc (vectorizable_load): When vectorizing
19372 using load-lanes allow virtual SSA update.
19374 2022-07-04 Tobias Burnus <tobias@codesourcery.com>
19375 Chung-Lin Tang <cltang@codesourcery.com>
19376 Thomas Schwinge <thomas@codesourcery.com>
19378 * config/gcn/mkoffload.cc (process_asm): Write '#include <stdint.h>'.
19379 (process_obj): Pass omp_requires_mask to GOMP_offload_register_ver.
19380 (main): Ask lto1 to obtain omp_requires_mask and pass it on.
19381 * config/nvptx/mkoffload.cc (process, main): Likewise.
19382 * lto-cgraph.cc (omp_requires_to_name): New.
19383 (input_offload_tables): Save omp_requires_mask.
19384 (output_offload_tables): Read it, check for consistency,
19385 save value for mkoffload.
19386 * omp-low.cc (lower_omp_target): Force output_offloadtables
19387 call for OMP_REQUIRES_TARGET_USED.
19389 2022-07-04 Richard Biener <rguenther@suse.de>
19391 * tree-vect-loop-manip.cc (vect_do_peeling): Assert that
19392 no SSA update is needed instead of updating virtual SSA
19394 * tree-vect-stmts.cc (vectorizable_load): For hoisted
19395 invariant load use the loop entry virtual use.
19396 For emulated gather loads use the virtual use of the
19397 original stmt like vect_finish_stmt_generation would do.
19399 2022-07-04 Eric Botcazou <ebotcazou@adacore.com>
19401 * tree-pretty-print.cc (dump_generic_node) <ARRAY_TYPE>: Add guard
19402 for direct circularity.
19404 2022-07-04 Eric Botcazou <ebotcazou@adacore.com>
19406 * dwarf2out.cc (gen_array_type_die): Use the default lower bound of
19407 the language for vector types.
19409 2022-07-04 Aldy Hernandez <aldyh@redhat.com>
19411 * value-range-storage.cc (irange_storage_slot::set_irange): Set
19412 nonzero bits in irange.
19413 (irange_storage_slot::get_irange): Get nonzero bits from irange.
19414 * value-range.cc (irange::operator=): Set nonzero bits.
19415 (irange::irange_set): Same.
19416 (irange::irange_set_anti_range): Same.
19417 (irange::set): Same.
19418 (irange::verify_range): Same.
19419 (irange::legacy_equal_p): Check nonzero bits.
19420 (irange::equal_p): Same.
19421 (irange::contains_p): Handle nonzero bits.
19422 (irange::irange_union): Same.
19423 (irange::irange_intersect): Same.
19424 (irange::dump): Same.
19425 (irange::set_nonzero_bits): New.
19426 (irange::get_nonzero_bits): New.
19427 (irange::intersect_nonzero_bits): New.
19428 (irange::union_nonzero_bits): New.
19429 (irange::dump_bitmasks): New.
19430 * value-range.h (class irange): Add m_nonzero_mask.
19431 (gt_ggc_mx): Handle nonzero bits.
19433 (irange::set_undefined): Set nonzero bits.
19434 (irange::set_varying): Same.
19435 (irange::normalize_kind): Call set_undefined.
19437 2022-07-04 Richard Biener <rguenther@suse.de>
19439 * tree-ssa-loop-manip.h
19440 (rewrite_virtuals_into_loop_closed_ssa): Remove.
19441 (rewrite_into_loop_closed_ssa_1): Likewise.
19442 * tree-ssa-loop-manip.cc (rewrite_into_loop_closed_ssa_1):
19444 (rewrite_into_loop_closed_ssa): Remove loop overload,
19445 always rewrite virtuals into LC SSA.
19446 (check_loop_closed_ssa_bb): Also check virtuals.
19447 * tree-ssa-dce.cc (remove_dead_phis): Preserve virtual
19448 LC PHIs when in LC SSA.
19449 * gimple-loop-jam.cc (fuse_loops): Do not rewrite into
19450 loop-closed SSA here, but ...
19451 (tree_loop_unroll_and_jam): ... here once.
19452 * tree-if-conv.cc (version_loop_for_if_conversion): Use
19453 the cheaper TODO_update_ssa_no_phi.
19454 * tree-loop-distribution.cc (version_loop_by_alias_check):
19456 * tree-ssa-loop-unswitch.cc (tree_unswitch_single_loop):
19458 * tree-vect-loop-manip.cc (vect_loop_versioning): Likewise.
19459 (tree_unswitch_outer_loop): Do not rewrite virtuals into
19461 * tree-parloops.cc (transform_to_exit_first_loop_alt):
19463 (pass_parallelize_loops::execute): After finishing rewrite
19464 into LC SSA again because we do not maintain it properly.
19466 2022-07-04 Richard Biener <rguenther@suse.de>
19468 PR tree-optimization/106055
19469 * graphite.cc (canonicalize_loop_closed_ssa): Check whether
19472 2022-07-04 Haochen Jiang <haochen.jiang@intel.com>
19475 * config/i386/sse.md (extendv2sfv2df2): New define_expand.
19476 (sse2_cvtps2pd_<mask_name>): Change constraint of operands[1].
19477 (*sse2_cvtps2pd_<mask_name>_1): Rename from extendvsdfv2df2.
19479 2022-07-04 Aldy Hernandez <aldyh@redhat.com>
19481 * ipa-cp.cc (ipcp_vr_lattice::meet_with_1): Use operator!=.
19482 * ipa-prop.cc (struct ipa_vr_ggc_hash_traits): Same.
19483 * tree-ssa-loop-unswitch.cc (struct unswitch_predicate): Use set
19484 with two arguments.
19485 (find_unswitching_predicates_for_bb): Same.
19486 * tree-vrp.cc (range_fold_unary_symbolics_p): Same.
19487 * value-range-equiv.cc (value_range_equiv::equal_p): Use operator==.
19488 * value-range.cc (irange::equal_p): Rename to...
19489 (irange::operator==): ...this.
19490 * value-range.h (irange::set): Remove.
19491 (irange::operator==): Remove.
19492 (irange::set_zero): Use set with two arguments.
19493 * vr-values.cc (vr_values::extract_range_from_binary_expr): Same.
19494 (vr_values::extract_range_from_unary_expr): Same.
19495 (check_for_binary_op_overflow): Same.
19496 (bounds_of_var_in_loop): Same.
19498 2022-07-03 H.J. Lu <hjl.tools@gmail.com>
19501 * config/i386/i386-protos.h (ix86_convert_const_vector_to_integer):
19503 * config/i386/i386.cc (ix86_convert_const_vector_to_integer):
19505 * config/i386/mmx.md (V_16_32_64): New.
19506 (*mov<mode>_imm): New patterns for stores with 16-bit, 32-bit
19507 and 64-bit constant vector.
19508 * config/i386/predicates.md (x86_64_const_vector_operand): New.
19510 2022-07-03 Aldy Hernandez <aldyh@redhat.com>
19512 * gimple-range-cache.cc: Include value-range-storage.h.
19513 * gimple-range-cache.h (class block_range_cache): Add "class" to
19515 * gimple-range-edge.cc
19516 (gimple_outgoing_range::gimple_outgoing_range): Allocate allocator.
19517 (gimple_outgoing_range::~gimple_outgoing_range): Free allocator.
19518 (gimple_outgoing_range::calc_switch_ranges): Dereference allocator.
19519 * gimple-range-edge.h: Add "class" to m_range_allocator.
19520 * gimple-range-infer.cc
19521 (infer_range_manager::infer_range_manager): Allocate allocator.
19522 (infer_range_manager::~infer_range_manager): Free allocator.
19523 (infer_range_manager::get_nonzero): Dereference allocator.
19524 (infer_range_manager::add_range): Same.
19525 * gimple-range-infer.h (class vrange_allocator): Add "class" to
19527 * value-range-storage.h (class vrange_allocator): Move from
19529 (class obstack_vrange_allocator): Same.
19530 (class ggc_vrange_allocator): Same.
19531 (vrange_allocator::alloc_vrange): Same.
19532 (vrange_allocator::alloc_irange): Same.
19533 * value-range.h (class vrange_allocator): Move to value-range-storage.h.
19534 (class obstack_vrange_allocator): Same.
19535 (class ggc_vrange_allocator): Same.
19537 2022-07-03 Aldy Hernandez <aldyh@redhat.com>
19539 * Makefile.in (OBJS): Add value-range-storage.o.
19540 (GTFILES): Add value-range-storage.h.
19541 * gengtype.cc (open_base_files): Add value-range-storage.h.
19542 * value-range-storage.cc: New file.
19543 * value-range-storage.h: New file.
19545 2022-07-03 Xi Ruoyao <xry111@xry111.site>
19546 Lulu Cheng <chenglulu@loongson.cn>
19548 * config/loongarch/loongarch.cc (loongarch_check_zero_div_p):
19549 New static function.
19550 (loongarch_idiv_insns): Use loongarch_check_zero_div_p instead
19551 of TARGET_CHECK_ZERO_DIV.
19552 (loongarch_output_division): Likewise.
19553 * common/config/loongarch/loongarch-common.cc
19554 (TARGET_DEFAULT_TARGET_FLAGS): Remove unneeded hook.
19555 * doc/invoke.texi: Update to match the new behavior.
19557 2022-07-03 Ian Lance Taylor <iant@golang.org>
19559 * tree-ssa-dse.cc (dse_optimize_stmt): Only delete a trapping
19560 statement if -fdelete-dead-exceptions.
19562 2022-07-02 Tim Lange <mail@tim-lange.me>
19565 * doc/invoke.texi: Added Wanalyzer-allocation-size.
19567 2022-07-02 Immad Mir <mirimmad17@gmail.com>
19570 * Makefile.in (ANALYZER_OBJS): Add sm-fd.o.
19571 * doc/invoke.texi: Add -Wanalyzer-fd-double-close, -Wanalyzer-fd-leak,
19572 -Wanalyzer-fd-access-mode-mismatch, -Wanalyzer-fd-use-without-check,
19573 -Wanalyzer-fd-use-after-close.
19575 2022-07-01 Aldy Hernandez <aldyh@redhat.com>
19577 * wide-int.h (struct trailing_wide_ints): Add m_num_elements.
19578 (trailing_wide_ints::set_precision): Add num_elements argument.
19579 (trailing_wide_ints::extra_size): Same.
19581 2022-07-01 Uroš Bizjak <ubizjak@gmail.com>
19583 * config/i386/i386.md (*andn<mode>3_doubleword_bmi):
19584 Use "r" constraint for operand 1.
19586 2022-07-01 Richard Biener <rguenther@suse.de>
19588 * tree-into-ssa.cc (rewrite_update_dom_walker::before_dom_children):
19589 Do not look at interesting_blocks which is a copy of
19591 (update_ssa): Do not initialize it.
19592 (pass_build_ssa::execute): Set interesting_blocks to NULL
19593 after releasing it.
19595 2022-07-01 Richard Biener <rguenther@suse.de>
19597 * tree-ssa-sccvn.cc (vn_reference_lookup_3): Revert
19598 back to using maybe_ne (off, -1).
19600 2022-07-01 Richard Biener <rguenther@suse.de>
19602 * tree-ssa-sccvn.cc (vn_nary_op_insert_into): Make
19603 checking dominance check conditional on flag_checking.
19605 2022-07-01 Tobias Burnus <tobias@codesourcery.com>
19607 * config/gcn/gcn-protos.h (print_operand_address): Remove register
19608 keyword on 'rtx addr' argument.
19610 2022-07-01 Roger Sayle <roger@nextmovesoftware.com>
19611 Uroš Bizjak <ubizjak@gmail.com>
19613 * config/i386/i386.md (*andn<mode>3_doubleword_bmi): Add constraints
19614 to post-reload define_insn_and_split.
19616 2022-07-01 Jakub Jelinek <jakub@redhat.com>
19618 PR middle-end/106144
19619 * wide-int.cc (wi::shifted_mask): If end >= prec, return right after
19620 emitting element for shift or if shift is 0 first element after start.
19621 (wide_int_cc_tests): Add tests for equivalency of wi::mask and
19622 wi::shifted_mask with 0 start.
19624 2022-07-01 Roger Sayle <roger@nextmovesoftware.com>
19627 * config/i386/i386.md (peephole2): Avoid generating pop %esp
19628 when optimizing for size.
19630 2022-07-01 Roger Sayle <roger@nextmovesoftware.com>
19631 Uroš Bizjak <ubizjak@gmail.com>
19633 * config/i386/i386.md (general_szext_operand): Add TImode
19634 support using x86_64_hilo_general_operand predicate.
19635 (*cmp<dwi>_doubleword): Use x86_64_hilo_general_operand predicate.
19636 (*add<dwi>3_doubleword): Improved optimization of zero addition.
19637 (and<mode>3): Use SDWIM mode iterator to add support for double
19638 word bit-wise AND in TImode. Use force_reg when double word
19639 immediate operand isn't x86_64_hilo_general_operand.
19640 (and<dwi>3_doubleword): Generalized from anddi3_doubleword and
19641 converted into a post-reload splitter.
19642 (*andndi3_doubleword): Previous define_insn deleted.
19643 (*andn<mode>3_doubleword_bmi): New define_insn_and_split for
19644 TARGET_BMI that splits post-reload.
19645 (*andn<mode>3_doubleword): New define_insn_and_split for
19646 !TARGET_BMI, that lowers/splits before reload.
19647 (<any_or><mode>3): Use SDWIM mode iterator to add suppport for
19648 double word bit-wise XOR and bit-wise IOR in TImode. Use
19649 force_reg when double word immediate operand isn't
19650 x86_64_hilo_general_operand.
19651 (*<any_or>di3_doubleword): Generalized from <any_or>di3_doubleword.
19652 (one_cmpl<mode>2): Use SDWIM mode iterator to add support for
19653 double word bit-wise NOT in TImode.
19654 (one_cmpl<dwi>2_doubleword): Generalize from one_cmpldi2_doubleword
19655 and converted into a post-reload splitter.
19657 2022-07-01 Eric Botcazou <ebotcazou@adacore.com>
19659 PR middle-end/105874
19660 * expr.cc (expand_expr_real_1) <normal_inner_ref>: Force
19661 EXPAND_MEMORY for the expansion of the inner reference only
19662 in the usual cases where a memory reference is required.
19664 2022-07-01 Richard Biener <rguenther@suse.de>
19666 PR tree-optimization/106131
19667 * tree-ssa-sccvn.cc (vn_reference_lookup_3): Force alias-set
19668 zero when offsetting the read looking through an aggregate
19671 2022-07-01 Martin Liska <mliska@suse.cz>
19673 PR tree-optimization/106126
19674 * gimple-if-to-switch.cc (struct condition_info): Save
19676 (find_conditions): Parse all BBs.
19677 (pass_if_to_switch::execute): Allow only side effects for first
19680 2022-07-01 Haochen Jiang <haochen.jiang@intel.com>
19682 * common/config/i386/i386-common.cc (OPTION_MASK_ISA2_AVX512F_UNSET):
19683 Add OPTION_MASK_ISA2_AVX512BW_UNSET, remove
19684 OPTION_MASK_ISA2_AVX512BF16_UNSET and
19685 OPTION_MASK_ISA2_AVX512FP16_UNSET.
19687 2022-06-30 Joseph Myers <joseph@codesourcery.com>
19690 * lto-wrapper.cc (find_option): Add argument start.
19691 (merge_and_complain): Loop over existing_opt_index and
19692 existing_opt2_index for Xassembler check. Update calls to
19694 (find_and_merge_options): Add argument first to determine whether
19695 to merge options with those passed in *opts.
19696 (run_gcc): Update calls to find_and_merge_options.
19698 2022-06-30 Aldy Hernandez <aldyh@redhat.com>
19700 * gimple-range-cache.cc (block_range_cache::block_range_cache):
19701 Rename vrange_allocator to obstack_vrange_allocator.
19702 (ssa_global_cache::ssa_global_cache): Same.
19703 * gimple-range-edge.h (class gimple_outgoing_range): Same.
19704 * gimple-range-infer.h (class infer_range_manager): Same.
19705 * value-range.h (class vrange_allocator): Make abstract.
19706 (class obstack_vrange_allocator): Inherit from vrange_allocator.
19707 (class ggc_vrange_allocator): New.
19709 2022-06-30 Roger Sayle <roger@nextmovesoftware.com>
19710 Uroš Bizjak <ubizjak@gmail.com>
19712 * config/i386/i386.md (swap_mode): Rename from *swap<mode> to
19713 provide gen_swapsi.
19714 (<any_rotate>di3): Handle !TARGET_64BIT rotations by 32 bits
19715 via new gen_<insn>32di2_doubleword below.
19716 (<anyrotate>32di2_doubleword): New define_insn_and_split
19717 that splits after reload as either a pair of move instructions
19718 or an xchgl (using gen_swapsi).
19720 2022-06-30 Richard Biener <rguenther@suse.de>
19722 * domwalk.h (dom_walker::dom_walker): Update comment to
19723 reflect reality and new special argument value for
19725 * domwalk.cc (dom_walker::dom_walker): Recognize -1
19728 (rewrite_update_dom_walker::rewrite_update_dom_walker): Tell
19729 dom_walker to not use RPO.
19731 2022-06-30 Martin Liska <mliska@suse.cz>
19733 * tree-ssa-dom.cc (pass_dominator::execute): Remove m_ranger as
19736 2022-06-30 Andrew MacLeod <amacleod@redhat.com>
19738 PR tree-optimization/106114
19739 * gimple-range-fold.cc (fold_using_range::relation_fold_and_or): Check
19740 statement operands instead of GORI cache.
19742 2022-06-30 Antoni Boucher <bouanto@zoho.com>
19745 * config/i386/sse.md: Fix asm generation.
19747 2022-06-29 Sergei Trofimovich <siarheit@google.com>
19750 * system.h: Introduce INCLUDE_PTHREAD_H macros to include <pthread.h>.
19752 2022-06-29 Joseph Myers <joseph@codesourcery.com>
19754 * config/nios2/nios2.cc (nios2_load_pic_address): Use gen_rtx_MEM
19755 not gen_const_mem for UNSPEC_PIC_CALL_SYM.
19757 2022-06-29 Richard Biener <rguenther@suse.de>
19759 PR rtl-optimization/106082
19760 * combine.cc (distribute_notes): Preserve notes when
19761 they indicate a call doesn't perform a non-local goto.
19763 2022-06-29 Richard Biener <rguenther@suse.de>
19765 PR tree-optimization/106112
19766 * tree-ssa-sccvn.cc (valueized_wider_op): Properly extend
19767 a constant operand according to its type.
19769 2022-06-29 Martin Liska <mliska@suse.cz>
19771 * doc/invoke.texi: Remove removed evrp-mode.
19773 2022-06-29 Lulu Cheng <chenglulu@loongson.cn>
19776 * config/loongarch/loongarch.cc (loongarch_build_integer):
19777 Remove undefined behavior from code.
19779 2022-06-28 Dimitar Dimitrov <dimitar@dinux.eu>
19781 * doc/sourcebuild.texi: Document new no_alignment_constraints
19782 effective target check.
19784 2022-06-28 Aldy Hernandez <aldyh@redhat.com>
19786 * Makefile.in: Remove gimple-ssa-evrp.o and gimple-ssa-evrp-analyze.o.
19787 * flag-types.h (enum evrp_mode): Remove.
19788 * params.opt: Remove --param=evrp-mode.
19789 * tree-vrp.cc (make_pass_early_vrp): New.
19790 (pass_vrp::execute): Call early VRP instance.
19791 * gimple-ssa-evrp-analyze.cc: Removed.
19792 * gimple-ssa-evrp-analyze.h: Removed.
19793 * gimple-ssa-evrp.cc: Removed.
19795 2022-06-28 Alexandre Oliva <oliva@adacore.com>
19797 * Makefile.in (TFLAGS): New.
19798 (GCC_FOR_TARGET): Add TFLAGS.
19799 (FLAGS_TO_PASS): Pass TFLAGS down.
19801 2022-06-28 Richard Biener <rguenther@suse.de>
19803 * tree-ssa-loop-split.cc (fix_loop_bb_probability): Do not
19806 2022-06-28 Richard Biener <rguenther@suse.de>
19808 * tree-into-ssa.cc (insert_updated_phi_nodes_for): Use
19809 mark_block_for_update.
19810 (update_ssa): Adjust.
19812 2022-06-28 Xi Ruoyao <xry111@xry111.site>
19815 * config/loongarch/loongarch.h (REG_CLASS_CONTENTS): Exclude
19816 $r13 from SIBCALL_REGS.
19817 * config/loongarch/loongarch.cc (loongarch_regno_to_class):
19818 Change $r13 to JIRL_REGS.
19820 2022-06-28 Aldy Hernandez <aldyh@redhat.com>
19822 * tree-ssa-loop-ch.cc (entry_loop_condition_is_static): Remove
19823 irange::supports_p.
19825 2022-06-28 Richard Biener <rguenther@suse.de>
19827 PR middle-end/106053
19828 * match.pd ((T)a == (T)b): Avoid folding away sign
19829 changes in a comparison if we'd truncate to a boolean.
19831 2022-06-28 Kewen Lin <linkw@linux.ibm.com>
19833 * config/rs6000/rs6000.md (*rotl<mode>3_insert_4): Replace mode
19834 iterator GPR with SImode, adjust the condition and output template,
19836 (*rotlsi3_insert_4): ... this.
19838 2022-06-27 David Malcolm <dmalcolm@redhat.com>
19840 * genmatch.cc: Add "final" and "override" to various vfunc
19841 implementations, removing redundant "virtual" as appropriate.
19842 * gensupport.cc: Likewise.
19843 * gimple-range-cache.h: Likewise.
19844 * ipa-icf-gimple.h: Likewise.
19845 * ipa-icf.h: Likewise.
19846 * read-md.h: Likewise.
19847 * read-rtl-function.cc: Likewise.
19848 * tree-ssa-loop-ch.cc: Likewise.
19849 * tree-ssa-sccvn.cc: Likewise.
19851 2022-06-27 David Malcolm <dmalcolm@redhat.com>
19853 * config/i386/i386-features.h: Add "final" and "override" to
19854 scalar_chain vfunc implementations as appropriate.
19856 2022-06-27 David Malcolm <dmalcolm@redhat.com>
19858 * tree-switch-conversion.h: Add "final" and "override" to cluster
19859 vfunc implementations as appropriate.
19861 2022-06-27 David Malcolm <dmalcolm@redhat.com>
19863 * expr.cc: Add "final" and "override" to op_by_pieces_d vfunc
19864 implementations as appropriate.
19866 2022-06-27 David Malcolm <dmalcolm@redhat.com>
19868 * compare-elim.cc: Add "final" and "override" to dom_walker vfunc
19869 implementations, removing redundant "virtual" as appropriate.
19870 * gimple-ssa-strength-reduction.cc: Likewise.
19871 * ipa-prop.cc: Likewise.
19872 * rtl-ssa/blocks.cc: Likewise.
19873 * tree-into-ssa.cc: Likewise.
19874 * tree-ssa-dom.cc: Likewise.
19875 * tree-ssa-math-opts.cc: Likewise.
19876 * tree-ssa-phiopt.cc: Likewise.
19877 * tree-ssa-propagate.cc: Likewise.
19878 * tree-ssa-sccvn.cc: Likewise.
19879 * tree-ssa-strlen.cc: Likewise.
19880 * tree-ssa-uncprop.cc: Likewise.
19882 2022-06-27 David Malcolm <dmalcolm@redhat.com>
19884 * adjust-alignment.cc: Add "final" and "override" to opt_pass
19885 vfunc implementations, removing redundant "virtual" as
19887 * asan.cc: Likewise.
19888 * auto-inc-dec.cc: Likewise.
19889 * auto-profile.cc: Likewise.
19890 * bb-reorder.cc: Likewise.
19891 * cfgcleanup.cc: Likewise.
19892 * cfgexpand.cc: Likewise.
19893 * cfgrtl.cc: Likewise.
19894 * cgraphbuild.cc: Likewise.
19895 * combine-stack-adj.cc: Likewise.
19896 * combine.cc: Likewise.
19897 * compare-elim.cc: Likewise.
19898 * config/i386/i386-features.cc: Likewise.
19899 * coroutine-passes.cc: Likewise.
19900 * cprop.cc: Likewise.
19901 * cse.cc: Likewise.
19902 * dce.cc: Likewise.
19903 * df-core.cc: Likewise.
19904 * dse.cc: Likewise.
19905 * dwarf2cfi.cc: Likewise.
19906 * early-remat.cc: Likewise.
19907 * except.cc: Likewise.
19908 * final.cc: Likewise.
19909 * function.cc: Likewise.
19910 * fwprop.cc: Likewise.
19911 * gcse.cc: Likewise.
19912 * gimple-harden-conditionals.cc: Likewise.
19913 * gimple-if-to-switch.cc: Likewise.
19914 * gimple-isel.cc: Likewise.
19915 * gimple-laddress.cc: Likewise.
19916 * gimple-loop-interchange.cc: Likewise.
19917 * gimple-loop-jam.cc: Likewise.
19918 * gimple-loop-versioning.cc: Likewise.
19919 * gimple-low.cc: Likewise.
19920 * gimple-ssa-backprop.cc: Likewise.
19921 * gimple-ssa-evrp.cc: Likewise.
19922 * gimple-ssa-isolate-paths.cc: Likewise.
19923 * gimple-ssa-nonnull-compare.cc: Likewise.
19924 * gimple-ssa-split-paths.cc: Likewise.
19925 * gimple-ssa-store-merging.cc: Likewise.
19926 * gimple-ssa-strength-reduction.cc: Likewise.
19927 * gimple-ssa-warn-access.cc: Likewise.
19928 * gimple-ssa-warn-alloca.cc: Likewise.
19929 * gimple-ssa-warn-restrict.cc: Likewise.
19930 * gimple-warn-recursion.cc: Likewise.
19931 * graphite.cc: Likewise.
19932 * ifcvt.cc: Likewise.
19933 * init-regs.cc: Likewise.
19934 * ipa-comdats.cc: Likewise.
19935 * ipa-cp.cc: Likewise.
19936 * ipa-devirt.cc: Likewise.
19937 * ipa-fnsummary.cc: Likewise.
19938 * ipa-free-lang-data.cc: Likewise.
19939 * ipa-icf.cc: Likewise.
19940 * ipa-inline.cc: Likewise.
19941 * ipa-modref.cc: Likewise.
19942 * ipa-profile.cc: Likewise.
19943 * ipa-pure-const.cc: Likewise.
19944 * ipa-reference.cc: Likewise.
19945 * ipa-split.cc: Likewise.
19946 * ipa-sra.cc: Likewise.
19947 * ipa-visibility.cc: Likewise.
19948 * ipa.cc: Likewise.
19949 * ira.cc: Likewise.
19950 * jump.cc: Likewise.
19951 * loop-init.cc: Likewise.
19952 * lower-subreg.cc: Likewise.
19953 * mode-switching.cc: Likewise.
19954 * modulo-sched.cc: Likewise.
19955 * multiple_target.cc: Likewise.
19956 * omp-expand.cc: Likewise.
19957 * omp-low.cc: Likewise.
19958 * omp-oacc-kernels-decompose.cc: Likewise.
19959 * omp-oacc-neuter-broadcast.cc: Likewise.
19960 * omp-offload.cc: Likewise.
19961 * omp-simd-clone.cc: Likewise.
19962 * passes.cc: Likewise.
19963 * postreload-gcse.cc: Likewise.
19964 * postreload.cc: Likewise.
19965 * predict.cc: Likewise.
19966 * recog.cc: Likewise.
19967 * ree.cc: Likewise.
19968 * reg-stack.cc: Likewise.
19969 * regcprop.cc: Likewise.
19970 * reginfo.cc: Likewise.
19971 * regrename.cc: Likewise.
19972 * reorg.cc: Likewise.
19973 * sancov.cc: Likewise.
19974 * sanopt.cc: Likewise.
19975 * sched-rgn.cc: Likewise.
19976 * stack-ptr-mod.cc: Likewise.
19977 * store-motion.cc: Likewise.
19978 * tracer.cc: Likewise.
19979 * trans-mem.cc: Likewise.
19980 * tree-call-cdce.cc: Likewise.
19981 * tree-cfg.cc: Likewise.
19982 * tree-cfgcleanup.cc: Likewise.
19983 * tree-complex.cc: Likewise.
19984 * tree-eh.cc: Likewise.
19985 * tree-emutls.cc: Likewise.
19986 * tree-if-conv.cc: Likewise.
19987 * tree-into-ssa.cc: Likewise.
19988 * tree-loop-distribution.cc: Likewise.
19989 * tree-nrv.cc: Likewise.
19990 * tree-object-size.cc: Likewise.
19991 * tree-parloops.cc: Likewise.
19992 * tree-predcom.cc: Likewise.
19993 * tree-profile.cc: Likewise.
19994 * tree-sra.cc: Likewise.
19995 * tree-ssa-ccp.cc: Likewise.
19996 * tree-ssa-copy.cc: Likewise.
19997 * tree-ssa-dce.cc: Likewise.
19998 * tree-ssa-dom.cc: Likewise.
19999 * tree-ssa-dse.cc: Likewise.
20000 * tree-ssa-forwprop.cc: Likewise.
20001 * tree-ssa-ifcombine.cc: Likewise.
20002 * tree-ssa-loop-ch.cc: Likewise.
20003 * tree-ssa-loop-im.cc: Likewise.
20004 * tree-ssa-loop-ivcanon.cc: Likewise.
20005 * tree-ssa-loop-prefetch.cc: Likewise.
20006 * tree-ssa-loop-split.cc: Likewise.
20007 * tree-ssa-loop-unswitch.cc: Likewise.
20008 * tree-ssa-loop.cc: Likewise.
20009 * tree-ssa-math-opts.cc: Likewise.
20010 * tree-ssa-phiopt.cc: Likewise.
20011 * tree-ssa-phiprop.cc: Likewise.
20012 * tree-ssa-pre.cc: Likewise.
20013 * tree-ssa-reassoc.cc: Likewise.
20014 * tree-ssa-sccvn.cc: Likewise.
20015 * tree-ssa-sink.cc: Likewise.
20016 * tree-ssa-strlen.cc: Likewise.
20017 * tree-ssa-structalias.cc: Likewise.
20018 * tree-ssa-uncprop.cc: Likewise.
20019 * tree-ssa-uninit.cc: Likewise.
20020 * tree-ssanames.cc: Likewise.
20021 * tree-stdarg.cc: Likewise.
20022 * tree-switch-conversion.cc: Likewise.
20023 * tree-tailcall.cc: Likewise.
20024 * tree-vect-generic.cc: Likewise.
20025 * tree-vectorizer.cc: Likewise.
20026 * tree-vrp.cc: Likewise.
20027 * tsan.cc: Likewise.
20028 * ubsan.cc: Likewise.
20029 * var-tracking.cc: Likewise.
20030 * vtable-verify.cc: Likewise.
20031 * web.cc: Likewise.
20033 2022-06-27 Andrew Stubbs <ams@codesourcery.com>
20035 * config/gcn/gcn.md (*movbi): Remove assembler bug workarounds.
20037 (movdi_symbol_save_scc): Likewise.
20039 2022-06-27 Roger Sayle <roger@nextmovesoftware.com>
20041 PR rtl-optimization/7061
20042 * config/i386/i386.md (*highpartdisi2): New define_insn_and_split.
20044 2022-06-27 Roger Sayle <roger@nextmovesoftware.com>
20046 PR tree-optimization/94026
20047 * match.pd (((X << C1) & C2) eq/ne C3): New simplification.
20048 (((X >> C1) & C2) eq/ne C3): Likewise.
20050 2022-06-27 Prathamesh Kulkarni <prathamesh.kulkarni@linaro.org>
20052 * statistics.cc: Include tree.h.
20053 (get_function_name): New function.
20054 (statistics_fini_pass_2): Call get_function_name instead of
20055 current_function_name.
20056 (statistics_counter_event): Call get_function_name instead of
20058 (statistics_histogram_event): Likewise.
20060 2022-06-27 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
20062 * config/xtensa/xtensa.md:
20063 Suppress unnecessary emitting nop insn in the split patterns for
20064 integer/FP constant synthesis, and add new peephole2 pattern that
20065 folds such synthesized additions.
20067 2022-06-25 Aldy Hernandez <aldyh@redhat.com>
20069 * tree-ssa-dom.cc (dom_jt_state): Pass ranger to constructor
20071 (dom_jt_state::push): Remove m_evrp.
20072 (dom_jt_state::pop): Same.
20073 (dom_jt_state::record_ranges_from_stmt): Remove.
20074 (dom_jt_state::register_equiv): Remove updating of evrp ranges.
20075 (class dom_jt_simplifier): Pass ranger to constructor.
20076 Inherit from hybrid_jt_simplifier.
20077 (dom_jt_simplifier::simplify): Convert to ranger.
20078 (pass_dominator::execute): Same.
20079 (all_uses_feed_or_dominated_by_stmt): New.
20080 (dom_opt_dom_walker::set_global_ranges_from_unreachable_edges): New.
20081 (dom_opt_dom_walker::before_dom_children): Call
20082 set_global_ranges_from_unreachable_edges.
20083 Do not call record_ranges_from_stmt.
20084 (dom_opt_dom_walker::after_dom_children): Remove evrp use.
20085 (cprop_operand): Use int_range<> instead of value_range.
20086 (dom_opt_dom_walker::fold_cond): New.
20087 (dom_opt_dom_walker::optimize_stmt): Pass ranger to
20089 Use fold_cond() instead of vrp_visit_cond_stmt().
20090 * tree-ssa-threadedge.cc (jt_state::register_equivs_stmt): Do not
20091 pass state to simplifier.
20092 * vr-values.h (class vr_values): Make fold_cond public.
20094 2022-06-25 Jeff Law <jeffreyalaw@gmail.com>
20096 * common/config/tilegx/tilegx-common.cc: Removed.
20097 * common/config/tilepro/tilepro-common.cc: Removed.
20098 * config.gcc: Remove tilegx and tilepro entries.
20099 * config/tilegx/constraints.md: Removed.
20100 * config/tilegx/feedback.h: Removed.
20101 * config/tilegx/linux.h: Removed.
20102 * config/tilegx/mul-tables.cc: Removed.
20103 * config/tilegx/predicates.md: Removed.
20104 * config/tilegx/sync.md: Removed.
20105 * config/tilegx/t-tilegx: Removed.
20106 * config/tilegx/tilegx-builtins.h: Removed.
20107 * config/tilegx/tilegx-c.cc: Removed.
20108 * config/tilegx/tilegx-generic.md: Removed.
20109 * config/tilegx/tilegx-modes.def: Removed.
20110 * config/tilegx/tilegx-multiply.h: Removed.
20111 * config/tilegx/tilegx-opts.h: Removed.
20112 * config/tilegx/tilegx-protos.h: Removed.
20113 * config/tilegx/tilegx.cc: Removed.
20114 * config/tilegx/tilegx.h: Removed.
20115 * config/tilegx/tilegx.md: Removed.
20116 * config/tilegx/tilegx.opt: Removed.
20117 * config/tilepro/constraints.md: Removed.
20118 * config/tilepro/feedback.h: Removed.
20119 * config/tilepro/gen-mul-tables.cc: Removed.
20120 * config/tilepro/linux.h: Removed.
20121 * config/tilepro/mul-tables.cc: Removed.
20122 * config/tilepro/predicates.md: Removed.
20123 * config/tilepro/t-tilepro: Removed.
20124 * config/tilepro/tilepro-builtins.h: Removed.
20125 * config/tilepro/tilepro-c.cc: Removed.
20126 * config/tilepro/tilepro-generic.md: Removed.
20127 * config/tilepro/tilepro-modes.def: Removed.
20128 * config/tilepro/tilepro-multiply.h: Removed.
20129 * config/tilepro/tilepro-protos.h: Removed.
20130 * config/tilepro/tilepro.cc: Removed.
20131 * config/tilepro/tilepro.h: Removed.
20132 * config/tilepro/tilepro.md: Removed.
20133 * config/tilepro/tilepro.opt: Removed.
20134 * configure.ac: Remove tilegx and tilepro entries.
20135 * configure: Rebuilt.
20136 * doc/extend.texi: Remove tilegx and tilepro entries.
20137 * doc/install.texi: Remove tilegx and tilepro entries.
20138 * doc/invoke.texi: Remove tilegx and tilepro entries.
20139 * doc/md.texi: Remove tilegx and tilepro entries.
20141 2022-06-25 Roger Sayle <roger@nextmovesoftware.com>
20142 Richard Biener <rguenther@suse.de>
20144 * regcprop.cc (pass_cprop_hardreg::execute): Perform a third
20145 iteration over each basic block that was updated by the second
20148 2022-06-24 Jason Merrill <jason@redhat.com>
20152 * doc/invoke.texi: Document changes.
20154 2022-06-24 Iain Buclaw <ibuclaw@gdcproject.org>
20156 * config/tilepro/gen-mul-tables.cc (tilegx_emit): Adjust loop
20157 condition to avoid overflow.
20159 2022-06-24 Will Schmidt <will_schmidt@vnet.ibm.com>
20161 * config/rs6000/aix71.h (MASK_PPC_GPOPT, MASK_PPC_GFXOPT): Replace with
20162 OPTION_MASK_PPC_GPOPT, OPTION_MASK_PPC_GFXOPT.
20163 * config/rs6000/darwin.h (MASK_PPC_GFXOPT): Replace with
20164 OPTION_MASK_PPC_GFXOPT.
20165 * config/rs6000/darwin64-biarch.h (MASK_PPC_GFXOPT): Same.
20166 * config/rs6000/default64.h (MASK_PPC_GPOPT, MASK_PPC_GFXOPT): Replace with
20167 OPTION_MASK_PPC_GPOPT, OPTION_MASK_PPC_GFXOPT.
20168 * config/rs6000/rs6000-c.cc: Update comment.
20169 * config/rs6000/rs6000-cpus.def: Update RS6000_CPU macro calls.
20170 * config/rs6000/rs6000.cc (rs6000_darwin_file_start): Replace
20171 MASK_PPC_GPOPT with OPTION_MASK_PPC_GPOPT.
20172 (rs6000_builtin_mask_names): Replace MASK_PPC_GFXOPT, MASK_POPCNTB
20173 with OPTION_MASK_PPC_GFXOPT, OPTION_MASK_POPCNTB.
20174 * config/rs6000/rs6000.h: (MASK_P8_VECTOR, MASK_P9_VECTOR,
20175 MASK_P9_MISC, MASK_POPCNTB, MASK_POPCNTD, MASK_PPC_GFXOPT,
20176 MASK_PPC_GPOPT, MASK_RECIP_PRECISION, MASK_SOFT_FLOAT,
20177 MASK_VSX, MASK_POWER10, MASK_P10_FUSION): Delete.
20179 2022-06-24 Will Schmidt <will_schmidt@vnet.ibm.com>
20181 * config/rs6000/aix71.h (TARGET_DEFAULT): Replace MASK_MFCRF with
20183 * config/rs6000/darwin.h (TARGET_DEFAULT): Replace MASK_MULTIPLE with
20184 OPTION_MASK_MULTIPLE.
20185 * config/rs6000/darwin64-biarch.h (TARGET_DEFAULT): Same.
20186 * config/rs6000/default64.h (TARGET_DEFAULT): Replace MASK_MFCRF with
20188 * config/rs6000/eabi.h (TARGET_DEFAULT): Replace MASK_EABI with
20190 * config/rs6000/eabialtivec.h (TARGET_DEFAULT): Same.
20191 * config/rs6000/linuxaltivec.h (TARGET_DEFAULT): Replace
20192 MASK_ALTIVEC with OPTION_MASK_ALTIVEC.
20193 * config/rs6000/rs6000-cpus.def (MASK_ALTIVEC, MASK_CMPB,
20194 MASK_CRYPTO, MASK_DFP, MASK_DIRECT_MOVE, MASK_DLMZB, MASK_EABI,
20195 MASK_FLOAT128_KEYWORD, MASK_FLOAT128_HW, MASK_FPRND,
20196 MASK_P8_FUSION, MASK_HARD_FLOAT, MASK_HTM, MASK_ISEL, MASK_MFCRF,
20197 MASK_MMA, MASK_MULHW, MASK_MULTIPLE, MASK_NO_UPDATE):
20199 OPTION_MASK_ALTIVEC, OPTION_MASK_CMPB, OPTION_MASK_CRYPTO,
20200 OPTION_MASK_DFP, OPTION_MASK_DIRECT_MOVE, OPTION_MASK_DLMZB,
20201 OPTION_MASK_EABI, OPTION_MASK_FLOAT128_KEYWORD,
20202 OPTION_MASK_FLOAT128_HW, OPTION_MASK_FPRND, OPTION_MASK_P8_FUSION,
20203 OPTION_MASK_HARD_FLOAT, OPTION_MASK_HTM, OPTION_MASK_ISEL,
20204 OPTION_MASK_MFCRF, OPTION_MASK_MMA, OPTION_MASK_MULHW,
20205 OPTION_MASK_MULTIPLE, OPTION_MASK_NO_UPDATE.
20206 * config/rs6000/rs6000.cc (rs6000_darwin_file_start): Replace
20207 MASK_MFCRF, MASK_ALTIVEC with OPTION_MASK_MFCRF, OPTION_MASK_ALTIVEC.
20208 * config/rs6000/rs6000.h (TARGET_DEFAULT): Replace MASK_MULTIPLE
20209 with OPTION_MASK_MULTIPLE.
20210 (MASK_ALTIVEC, MASK_CMPB, MASK_CRYPTO, MASK_DFP,
20211 MASK_DIRECT_MOVE, MASK_DLMZB, MASK_EABI, MASK_FLOAT128_KEYWORD,
20212 MASK_FLOAT128_HW, MASK_FPRND, MASK_P8_FUSION, MASK_HARD_FLOAT,
20213 MASK_HTM, MASK_ISEL, MASK_MFCRF, MASK_MMA, MASK_MULHW,
20214 MASK_MULTIPLE, MASK_NO_UPDATE): Delete.
20215 * config/rs6000/vxworks.h (TARGET_DEFAULT): Replace MASK_EABI
20216 with OPTION_MASK_EABI.
20218 2022-06-24 Will Schmidt <will_schmidt@vnet.ibm.com>
20220 * config/rs6000/rs6000.cc (RS6000_BTM_ALTIVEC, RS6000_BTM_CMPB,
20221 RS6000_BTM_VSX, RS6000_BTM_FRE, RS6000_BTM_P8_VECTOR,
20222 RS6000_BTM_P9_VECTOR, RS6000_BTM_P9_MISC, RS6000_BTM_MODULO,
20223 RS6000_BTM_CRYPTO, RS6000_BTM_HTM): Replace with OPTION_MASK_ALTIVEC,
20224 OPTION_MASK_CMPB, OPTION_MASK_VSX, OPTION_MASK_POPCNTB,
20225 OPTION_MASK_P8_VECTOR, OPTION_MASK_P9_VECTOR, OPTION_MASK_P9_MISC,
20226 OPTION_MASK_MODULO, OPTION_MASK_CRYPTO, OPTION_MASK_HTM.
20227 * config/rs6000/rs6000.h (RS6000_BTM_MODULO, RS6000_BTM_ALTIVEC,
20228 RS6000_BTM_CMPB, RS6000_BTM_VSX, RS6000_BTM_P8_VECTOR,
20229 RS6000_BTM_P9_VECTOR, RS6000_BTM_P9_MISC, RS6000_BTM_CRYPTO,
20230 RS6000_BTM_HTM, RS6000_BTM_FRE): Remove.
20232 2022-06-24 Will Schmidt <will_schmidt@vnet.ibm.com>
20234 * config/rs6000/rs6000-c.cc: Update comments.
20235 * config/rs6000/rs6000.cc (RS6000_BTM_FRES, RS6000_BTM_FRSQRTE,
20236 RS6000_BTM_FRSQRTES, RS6000_BTM_POPCNTD, RS6000_BTM_CELL,
20237 RS6000_BTM_64BIT, RS6000_BTM_POWERPC64, RS6000_BTM_DFP,
20238 RS6000_BTM_HARD_FLOAT,RS6000_BTM_LDBL128, RS6000_BTM_FLOAT128,
20239 RS6000_BTM_FLOAT128_HW, RS6000_BTM_MMA, RS6000_BTM_P10): Replace
20240 with OPTION_MASK_PPC_GFXOPT, OPTION_MASK_PPC_GFXOPT,
20241 OPTION_MASK_POPCNTB, OPTION_MASK_POPCNTD,
20242 OPTION_MASK_FPRND, MASK_64BIT, MASK_POWERPC64,
20243 OPTION_MASK_DFP, OPTION_MASK_SOFT_FLOAT, OPTION_MASK_MULTIPLE,
20244 OPTION_MASK_FLOAT128_KEYWORD, OPTION_MASK_FLOAT128_HW,
20245 OPTION_MASK_MMA, OPTION_MASK_POWER10.
20246 * config/rs6000/rs6000.h (RS6000_BTM_FRES, RS6000_BTM_FRSQRTE,
20247 RS6000_BTM_FRSQRTES, RS6000_BTM_POPCNTD, RS6000_BTM_CELL,
20248 RS6000_BTM_DFP, RS6000_BTM_HARD_FLOAT, RS6000_BTM_LDBL128,
20249 RS6000_BTM_64BIT, RS6000_BTM_POWERPC64, RS6000_BTM_FLOAT128,
20250 RS6000_BTM_FLOAT128_HW, RS6000_BTM_MMA, RS6000_BTM_P10): Delete.
20252 2022-06-24 Will Schmidt <will_schmidt@vnet.ibm.com>
20254 * config/rs6000/rs6000.h (RS6000_BTM_COMMON, RS6000_BTM_ALWAYS,
20255 MASK_REGNAMES, OPTION_MASK_REGNAMES, MASK_PROTOTYPE,
20256 OPTION_MASK_PROTOTYPE, MASK_UPDATE, OPTION_MASK_UPDATE): Remove.
20258 2022-06-24 Richard Biener <rguenther@suse.de>
20260 PR middle-end/106070
20261 * match.pd (a != b ? a : b): Fix translation of
20262 operand_equal_for_comparison_p.
20264 2022-06-24 Jan Hubicka <jh@suse.cz>
20267 * tree-ssa-alias.cc (stmt_kills_ref_p): Check for external throw.
20269 2022-06-24 Martin Liska <mliska@suse.cz>
20271 PR middle-end/106059
20272 * profile-count.h: *= and /= operators need to modify this
20275 2022-06-24 Roger Sayle <roger@nextmovesoftware.com>
20276 Uroš Bizjak <ubizjak@gmail.com>
20279 * config/i386/i386.md (*<any_or>di3_doubleword): Split after
20280 reload. Use rtx_equal_p to avoid creating memory-to-memory moves,
20281 and emit NOTE_INSN_DELETED if operand[2] is zero (i.e. with -O0).
20283 2022-06-24 Alexandre Oliva <oliva@adacore.com>
20285 * common.opt (nostdlib++): New.
20286 * doc/invoke.texi (-nostdlib++): Document it.
20288 2022-06-24 Alexandre Oliva <oliva@adacore.com>
20290 * doc/sourcebuild.texi (Environment attributes): Document
20293 2022-06-23 David Malcolm <dmalcolm@redhat.com>
20295 * common.opt (fdiagnostics-show-rules): New option.
20296 * diagnostic-format-json.cc (diagnostic_output_format_init_json):
20297 Fix up context->show_rules.
20298 * diagnostic-format-sarif.cc
20299 (diagnostic_output_format_init_sarif): Likewise.
20300 * diagnostic-metadata.h (diagnostic_metadata::rule): New class.
20301 (diagnostic_metadata::precanned_rule): New class.
20302 (diagnostic_metadata::add_rule): New.
20303 (diagnostic_metadata::get_num_rules): New.
20304 (diagnostic_metadata::get_rule): New.
20305 (diagnostic_metadata::m_rules): New field.
20306 * diagnostic.cc (diagnostic_initialize): Initialize show_rules.
20307 (print_any_rules): New.
20308 (diagnostic_report_diagnostic): Call it.
20309 * diagnostic.h (diagnostic_context::show_rules): New field.
20310 * doc/invoke.texi (-fno-diagnostics-show-rules): New option.
20311 * opts.cc (common_handle_option): Handle
20312 OPT_fdiagnostics_show_rules.
20313 * toplev.cc (general_init): Set up global_dc->show_rules.
20315 2022-06-23 Martin Liska <mliska@suse.cz>
20318 * ubsan.cc (sanitize_unreachable_fn): Change order of calls
20319 in order to initialize UBSAN built-ins.
20321 2022-06-23 Martin Liska <mliska@suse.cz>
20324 * ipa-icf.cc (sem_item_optimizer::filter_removed_items):
20325 Skip variables with body_removed.
20327 2022-06-23 liuhongt <hongtao.liu@intel.com>
20329 * config/i386/sse.md:(sse4_2_pcmpestr): Replace REGNO with
20331 (sse4_2_pcmpistr): Ditto.
20333 2022-06-23 Xionghu Luo <xionghuluo@tencent.com>
20335 * cgraph.cc (cgraph_edge::redirect_call_stmt_to_callee): Fix
20337 * tree-ssa-loop-ivopts.cc (struct iv_cand): Likewise.
20338 * tree-switch-conversion.h: Likewise.
20340 2022-06-22 Jason Merrill <jason@redhat.com>
20343 * common.opt: Add -funreachable-traps.
20344 * doc/invoke.texi (-funreachable-traps): Document it.
20345 * opts.cc (finish_options): Enable at -O0 or -Og.
20346 * tree.cc (build_common_builtin_nodes): Add __builtin_trap.
20347 (builtin_decl_unreachable, build_builtin_unreachable): New.
20348 * tree.h: Declare them.
20349 * ubsan.cc (sanitize_unreachable_fn): Factor out.
20350 (ubsan_instrument_unreachable): Use
20351 gimple_build_builtin_unreachable.
20352 * ubsan.h (sanitize_unreachable_fn): Declare.
20353 * gimple.cc (gimple_build_builtin_unreachable): New.
20354 * gimple.h: Declare it.
20355 * builtins.cc (expand_builtin_unreachable): Add assert.
20356 (fold_builtin_0): Call build_builtin_unreachable.
20357 * sanopt.cc: Don't run for just SANITIZE_RETURN
20358 or SANITIZE_UNREACHABLE when trapping.
20359 * cgraphunit.cc (walk_polymorphic_call_targets): Use new
20360 unreachable functions.
20361 * gimple-fold.cc (gimple_fold_call)
20362 (gimple_get_virt_method_for_vtable)
20363 * ipa-fnsummary.cc (redirect_to_unreachable)
20364 * ipa-prop.cc (ipa_make_edge_direct_to_target)
20365 (ipa_impossible_devirt_target)
20366 * ipa.cc (walk_polymorphic_call_targets)
20367 * tree-cfg.cc (pass_warn_function_return::execute)
20368 (execute_fixup_cfg)
20369 * tree-ssa-loop-ivcanon.cc (remove_exits_and_undefined_stmts)
20371 * tree-ssa-sccvn.cc (eliminate_dom_walker::eliminate_stmt):
20374 2022-06-22 Richard Sandiford <richard.sandiford@arm.com>
20376 PR tree-optimization/106019
20377 * tree-data-ref.cc (dr_may_alias_p): Try using the
20378 innermost_loop_behavior to disambiguate non-loop queries.
20380 2022-06-22 Palmer Dabbelt <palmer@rivosinc.com>
20382 * doc/invoke.texi (RISC-V): Document -mtune=thead-c906.
20384 2022-06-22 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
20386 * config/xtensa/xtensa.md (bswapsi2_internal):
20387 Enlarge the buffer that is obviously smaller than the template
20388 string given to sprintf().
20390 2022-06-21 Roger Sayle <roger@nextmovesoftware.com>
20391 Marek Polacek <polacek@redhat.com>
20392 Segher Boessenkool <segher@kernel.crashing.org>
20393 Kewen Lin <linkw@linux.ibm.com>
20396 * config/rs6000/rs6000.md (rotl<mode>3_insert_3): Check that
20397 exact_log2 doesn't return -1 (or zero).
20398 (plus_xor): New code iterator.
20399 (*rotl<mode>3_insert_3_<code>): New define_insn_and_split.
20401 2022-06-21 Nathan Sidwell <nathan@acm.org>
20403 * doc/invoke.texi (C++ Modules): Remove language-linkage
20404 as missing feature.
20406 2022-06-21 Arjun Shankar <arjun@redhat.com>
20408 PR tree-optimization/94899
20409 * match.pd (X + C < Y + C -> (signed) X < (signed) Y, if C is
20410 0x80000000): New simplification.
20412 2022-06-21 Jakub Jelinek <jakub@redhat.com>
20414 PR rtl-optimization/106032
20415 * ifcvt.cc (noce_try_sign_mask): Punt if !t_unconditional, and
20416 t may_trap_or_fault_p, even if it is cheap.
20418 2022-06-21 Jakub Jelinek <jakub@redhat.com>
20420 PR middle-end/106030
20421 * expr.cc (expand_cond_expr_using_cmove): Pass NULL_RTX instead of
20422 temp to expand_operands if mode has been promoted.
20424 2022-06-21 Xionghu Luo <xionghuluo@tencent.com>
20427 * gimple-if-to-switch.cc (find_conditions): Don't skip the first
20430 2022-06-21 Siddhesh Poyarekar <siddhesh@gotplt.org>
20432 PR tree-optimization/105736
20433 * tree-object-size.cc (addr_object_size): Return size_unknown
20434 when object offset computation returns an error.
20436 2022-06-20 H.J. Lu <hjl.tools@gmail.com>
20439 * config/i386/i386.cc (ix86_function_ok_for_sibcall): Return
20440 false if PIC register is used when calling ifunc functions.
20442 2022-06-20 Richard Biener <rguenther@suse.de>
20444 PR middle-end/106027
20445 * fold-const.cc (fold_to_nonsharp_ineq_using_bound): Use the
20446 type of the prevailing comparison for the new comparison type.
20447 (fold_binary_loc): Use proper types for the A < X && A + 1 > Y
20448 to A < X && A >= Y folding.
20450 2022-06-20 Kewen Lin <linkw@linux.ibm.com>
20452 PR tree-optimization/105940
20453 * tree-vect-loop.cc (vect_analyze_loop_2): Add new parameter
20454 slp_done_for_suggested_uf and adjust with it accordingly.
20455 (vect_analyze_loop_1): Add new variable slp_done_for_suggested_uf,
20456 pass it down to vect_analyze_loop_2 for the initial analysis and
20457 applying suggested unroll factor.
20458 (vect_is_simple_reduction): Add parameter slp and adjust with it.
20459 (vect_analyze_scalar_cycles_1): Add parameter slp and pass down.
20460 (vect_analyze_scalar_cycles): Likewise.
20462 2022-06-20 Martin Liska <mliska@suse.cz>
20464 * bb-reorder.cc (find_traces_1_round): Add operators / and * and
20466 (better_edge_p): Likewise.
20467 * cfgloop.cc (find_subloop_latch_edge_by_profile): Likewise.
20468 * cfgloopmanip.cc (scale_loop_profile): Likewise.
20469 * cfgrtl.cc (force_nonfallthru_and_redirect): Likewise.
20470 * cgraph.cc (cgraph_edge::maybe_hot_p): Likewise.
20471 * config/sh/sh.cc (expand_cbranchdi4): Likewise.
20472 * dojump.cc (do_compare_rtx_and_jump): Likewise.
20473 * final.cc (compute_alignments): Likewise.
20474 * ipa-cp.cc (update_counts_for_self_gen_clones): Likewise.
20475 (decide_about_value): Likewise.
20476 * ipa-inline-analysis.cc (do_estimate_edge_time): Likewise.
20477 * loop-unroll.cc (unroll_loop_runtime_iterations): Likewise.
20478 * modulo-sched.cc (sms_schedule): Likewise.
20479 * omp-expand.cc (extract_omp_for_update_vars): Likewise.
20480 (expand_omp_ordered_sink): Likewise.
20481 (expand_omp_for_ordered_loops): Likewise.
20482 (expand_omp_for_static_nochunk): Likewise.
20483 * predict.cc (maybe_hot_count_p): Likewise.
20484 (probably_never_executed): Likewise.
20485 (set_even_probabilities): Likewise.
20486 (handle_missing_profiles): Likewise.
20487 (expensive_function_p): Likewise.
20488 * profile-count.h: Likewise.
20489 * profile.cc (compute_branch_probabilities): Likewise.
20490 * stmt.cc (emit_case_dispatch_table): Likewise.
20491 * symtab-thunks.cc (expand_thunk): Likewise.
20492 * tree-ssa-loop-manip.cc (tree_transform_and_unroll_loop): Likewise.
20493 * tree-ssa-sink.cc (select_best_block): Likewise.
20494 * tree-switch-conversion.cc (switch_decision_tree::analyze_switch_statement): Likewise.
20495 (switch_decision_tree::balance_case_nodes): Likewise.
20496 (switch_decision_tree::emit_case_nodes): Likewise.
20497 * tree-vect-loop.cc (scale_profile_for_vect_loop): Likewise.
20499 2022-06-20 yulong <shiyulong@iscas.ac.cn>
20501 * config/riscv/riscv-builtins.cc (RISCV_ATYPE_VOID_PTR): New.
20502 * config/riscv/riscv-cmo.def (RISCV_BUILTIN): Changed the FUNCTION_TYPE
20504 * config/riscv/riscv-ftypes.def (0): Remove unused.
20507 2022-06-19 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
20509 * config/xtensa/xtensa.cc (xtensa_is_insn_L32R_p):
20510 Consider relaxed MOVI instructions as L32R.
20512 2022-06-19 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
20514 * config/xtensa/xtensa.cc (xtensa_emit_move_sequence):
20515 Use can_create_pseudo_p(), instead of using individual
20516 reload_in_progress and reload_completed.
20517 (xtensa_expand_block_set_small_loop): Use xtensa_simm8x256(),
20518 the existing predicate function.
20519 (xtensa_is_insn_L32R_p, gen_int_relational, xtensa_emit_sibcall):
20520 Use the standard RTX code predicate macros such as MEM_P,
20521 SYMBOL_REF_P and/or CONST_INT_P.
20522 * config/xtensa/xtensa.md: Avoid using numeric literals to determine
20523 if callee-saved register, at the split patterns for indirect sibcall
20526 2022-06-18 Jakub Jelinek <jakub@redhat.com>
20528 * common.opt (flag_sanitize_trap): New variable.
20529 (fsanitize-trap=, fsanitize-trap): New options.
20530 (fsanitize-undefined-trap-on-error): Change into deprecated alias
20531 for -fsanitize-trap=all.
20532 * opts.h (struct sanitizer_opts_s): Add can_trap member.
20533 * opts.cc (finish_options): Complain about unsupported
20534 -fsanitize-trap= options.
20535 (sanitizer_opts): Add can_trap values to all entries.
20536 (get_closest_sanitizer_option): Ignore -fsanitize-trap=
20537 options which have can_trap false.
20538 (parse_sanitizer_options): Add support for -fsanitize-trap=.
20539 For -fsanitize-trap=all, enable
20540 SANITIZE_UNDEFINED | SANITIZE_UNDEFINED_NONDEFAULT. Disallow
20541 -fsanitize-trap=vptr here.
20542 (common_handle_option): Handle OPT_fsanitize_trap_ and
20543 OPT_fsanitize_trap.
20544 * sanopt.cc (maybe_optimize_ubsan_null_ifn): Check
20545 flag_sanitize_trap & SANITIZE_{NULL,ALIGNMENT} instead of
20546 flag_sanitize_undefined_trap_on_error.
20547 * gcc.cc (sanitize_spec_function): Use
20548 flag_sanitize & ~flag_sanitize_trap instead of flag_sanitize
20549 and drop use of flag_sanitize_undefined_trap_on_error in
20550 "undefined" handling.
20551 * ubsan.cc (ubsan_instrument_unreachable): Use
20552 flag_sanitize_trap & SANITIZE_??? instead of
20553 flag_sanitize_undefined_trap_on_error.
20554 (ubsan_expand_bounds_ifn, ubsan_expand_null_ifn,
20555 ubsan_expand_objsize_ifn, ubsan_expand_ptr_ifn,
20556 ubsan_build_overflow_builtin, instrument_bool_enum_load,
20557 ubsan_instrument_float_cast, instrument_nonnull_arg,
20558 instrument_nonnull_return, instrument_builtin): Likewise.
20559 * doc/invoke.texi (-fsanitize-trap=, -fsanitize-trap): Document.
20560 (-fsanitize-undefined-trap-on-error): Document as deprecated
20561 alias of -fsanitize-trap.
20563 2022-06-18 Jakub Jelinek <jakub@redhat.com>
20565 PR middle-end/105998
20566 * varasm.cc (narrowing_initializer_constant_valid_p): Check
20567 SCALAR_INT_MODE_P instead of INTEGRAL_MODE_P, also break on
20568 ! INTEGRAL_TYPE_P and do the same check also on op{0,1}'s type.
20570 2022-06-18 Roger Sayle <roger@nextmovesoftware.com>
20572 PR tree-optimization/105835
20573 * match.pd (convert (mult zero_one_valued_p@1 INTEGER_CST@2)):
20574 Narrow integer multiplication by a zero_one_valued_p operand.
20575 (convert (cond @1 INTEGER_CST@2 INTEGER_CST@3)): Push integer
20576 conversions inside COND_EXPR where both data operands are
20579 2022-06-18 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
20581 * config/xtensa/constraints.md (Y):
20582 Change to include integer constants until reload begins.
20583 * config/xtensa/predicates.md (move_operand): Ditto.
20584 * config/xtensa/xtensa.cc (xtensa_emit_move_sequence):
20585 Change to allow storing integer constants into litpool only after
20588 2022-06-17 Uroš Bizjak <ubizjak@gmail.com>
20591 * config/alpha/alpha-protos.h (alpha_store_data_bypass_p): New.
20592 * config/alpha/alpha.cc (alpha_store_data_bypass_p): New function.
20593 (alpha_store_data_bypass_p_1): Ditto.
20594 * config/alpha/ev4.md: Use alpha_store_data_bypass_p instead
20595 of generic store_data_bypass_p.
20596 (ev4_ist_c): Remove insn reservation.
20598 2022-06-17 Uroš Bizjak <ubizjak@gmail.com>
20601 * config/i386/i386.cc (ix86_function_arg): Assert that
20602 the mode of pointer argumet is equal to ptr_mode, not Pmode.
20604 2022-06-17 Uroš Bizjak <ubizjak@gmail.com>
20607 * config/i386/sse.md (vpmov splitter): Use (match_dup ...)
20608 instead of REGNO comparisons in combine splitter.
20610 2022-06-17 Segher Boessenkool <segher@kernel.crashing.org>
20612 * config/rs6000/rs6000.cc (rs6000_invalid_conversion): Correct some
20615 2022-06-17 Kito Cheng <kito.cheng@sifive.com>
20617 * config/riscv/bitmanip.md: Supress warning.
20619 2022-06-17 Richard Earnshaw <rearnsha@arm.com>
20622 * config/arm/arm.cc (arm_print_operand, case 'V'): Use UINTVAL.
20623 Clear bits in the mask above bit 31.
20625 2022-06-17 Richard Earnshaw <rearnsha@arm.com>
20627 * config/arm/mve.md (*mve_mov<mode>): Re-order constraints
20628 to avoid spilling trivial literals to the constant pool.
20630 2022-06-16 David Malcolm <dmalcolm@redhat.com>
20632 * gimple-ssa-warn-access.cc (warn_string_no_nul): Add
20633 auto_diagnostic_group to group any warning with its note.
20634 (maybe_warn_for_bound): Likewise.
20635 (check_access): Likewise.
20636 (warn_dealloc_offset): Likewise.
20637 (pass_waccess::maybe_warn_memmodel): Likewise.
20638 (pass_waccess::maybe_check_dealloc_call): Likewise.
20639 (pass_waccess::warn_invalid_pointer): Likewise.
20640 (pass_waccess::check_dangling_stores): Likewise.
20642 2022-06-16 Jason Merrill <jason@redhat.com>
20644 * opts.cc (common_handle_option) [OPT_fsanitize_]: Set
20645 opts_set->x_flag_sanitize.
20647 2022-06-16 Jason Merrill <jason@redhat.com>
20649 * flags.h (issue_strict_overflow_warning): Comment #endif.
20651 2022-06-16 Andrew MacLeod <amacleod@redhat.com>
20653 * gimple-range-cache.cc (ranger_cache::apply_inferred_ranges): If name
20654 was invaraint before, clear the invariant bit.
20655 * gimple-range-gori.cc (gori_map::set_range_invariant): Add a flag.
20656 * gimple-range-gori.h (gori_map::set_range_invariant): Adjust prototype.
20658 2022-06-16 Andrew MacLeod <amacleod@redhat.com>
20660 * tree-ssa-propagate.cc (before_dom_children): Call value_of_stmt.
20662 2022-06-16 Jakub Jelinek <jakub@redhat.com>
20664 PR tree-optimization/105983
20665 * match.pd (y == XXX_MIN || x < y -> x <= y - 1,
20666 y != XXX_MIN && x >= y -> x > y - 1): Use :cs instead of :s
20667 on non-equality comparisons.
20669 2022-06-16 Jakub Jelinek <jakub@redhat.com>
20671 PR tree-optimization/105984
20672 * match.pd (__builtin_mul_overflow_p (x, cst, (stype) 0) ->
20673 x > stype_max / cst || x < stype_min / cst): fold_convert @1
20674 to TREE_TYPE (@0) just once and test for negative divisor
20675 also on that folded constant instead of on @1.
20677 2022-06-16 Jakub Jelinek <jakub@redhat.com>
20679 PR middle-end/105951
20680 * tree-ssa-ccp.cc (optimize_atomic_bit_test_and,
20681 optimize_atomic_op_fetch_cmp_0): Remember gimple_call_fn (call)
20682 as last argument to the internal functions.
20683 * builtins.cc (expand_ifn_atomic_bit_test_and): Adjust for the
20684 extra call argument to ifns. If expand_atomic_fetch_op fails for the
20685 lhs == NULL_TREE case, fall through into the optab code with
20686 gen_reg_rtx (mode) as target. If second expand_atomic_fetch_op
20687 fails, construct a CALL_EXPR and expand that.
20688 (expand_ifn_atomic_op_fetch_cmp_0): Adjust for the extra call argument
20689 to ifns. If expand_atomic_fetch_op fails, construct a CALL_EXPR and
20692 2022-06-16 Haochen Gui <guihaoc@gcc.gnu.org>
20695 * config/rs6000/rs6000-builtin.cc (rs6000_gimple_fold_builtin): Enable
20696 gimple folding for RS6000_BIF_VCMPEQUT, RS6000_BIF_VCMPNET,
20697 RS6000_BIF_CMPGE_1TI, RS6000_BIF_CMPGE_U1TI, RS6000_BIF_VCMPGTUT,
20698 RS6000_BIF_VCMPGTST, RS6000_BIF_CMPLE_1TI, RS6000_BIF_CMPLE_U1TI.
20699 * config/rs6000/vector.md (VEC_IC): New mode iterator. Add support
20700 for new Power10 V1TI instructions.
20701 (vec_cmp<mode><mode>): Set mode iterator to VEC_IC.
20702 (vec_cmpu<mode><mode>): Likewise.
20703 (vector_nlt<mode>): Set mode iterator to VEC_IC.
20704 (vector_nltv1ti): Remove.
20705 (vector_gtu<mode>): Set mode iterator to VEC_IC.
20706 (vector_gtuv1ti): Remove.
20707 (vector_nltu<mode>): Set mode iterator to VEC_IC.
20708 (vector_nltuv1ti): Remove.
20709 (vector_geu<mode>): Set mode iterator to VEC_IC.
20710 (vector_ngt<mode>): Likewise.
20711 (vector_ngtv1ti): Remove.
20712 (vector_ngtu<mode>): Set mode iterator to VEC_IC.
20713 (vector_ngtuv1ti): Remove.
20714 (vector_gtu_<mode>_p): Set mode iterator to VEC_IC.
20715 (vector_gtu_v1ti_p): Remove.
20716 (vrotl<mode>3): Set mode iterator to VEC_IC. Emit insns for V1TI.
20717 (vrotlv1ti3): Remove.
20718 (vashr<mode>3): Set mode iterator to VEC_IC. Emit insns for V1TI.
20719 (vashrv1ti3): Remove.
20721 2022-06-16 Martin Liska <mliska@suse.cz>
20723 * gengtype-state.cc (read_a_state_token): Do not skip extra
20724 character after escaped sequence.
20726 2022-06-16 Martin Liska <mliska@suse.cz>
20729 * spellcheck.cc (test_find_closest_string): Add new test.
20730 * spellcheck.h (class best_match): Prefer a difference in
20731 trailing sign symbol.
20733 2022-06-16 liuhongt <hongtao.liu@intel.com>
20735 PR tree-optimization/53533
20736 * match.pd: Simplify (B * v + C) * D -> BD * v + CD and
20737 (v + B) * C + D -> C * v + BCD when B,C,D are all INTEGER_CST,
20738 and there's no overflow or !TYPE_OVERFLOW_UNDEFINED.
20740 2022-06-15 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
20742 * config/xtensa/xtensa.md (DSC): New split pattern and mode iterator.
20744 2022-06-15 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
20746 * config/xtensa/predicates.md (reload_operand):
20748 * config/xtensa/xtensa.md: New peephole2 pattern.
20750 2022-06-15 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
20752 * config/xtensa/xtensa.md (*round_up_to_even):
20753 New insn-and-split pattern.
20754 (*signed_ge_zero): Ditto.
20756 2022-06-15 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
20758 * config/xtensa/xtensa-protos.h (xtensa_prepare_expand_call,
20759 xtensa_emit_sibcall): New prototypes.
20760 (xtensa_expand_epilogue): Add new argument that specifies whether
20761 or not sibling call.
20762 * config/xtensa/xtensa.cc (TARGET_FUNCTION_OK_FOR_SIBCALL):
20763 New macro definition.
20764 (xtensa_prepare_expand_call): New function in order to share
20766 (xtensa_emit_sibcall, xtensa_function_ok_for_sibcall):
20768 (xtensa_expand_epilogue): Add new argument sibcall_p and use it
20769 for sibling call handling.
20770 * config/xtensa/xtensa.md (call, call_value):
20771 Use xtensa_prepare_expand_call.
20772 (call_internal, call_value_internal):
20773 Add the condition in order to be disabled if sibling call.
20774 (sibcall, sibcall_value, sibcall_epilogue): New expansions.
20775 (sibcall_internal, sibcall_value_internal): New insn patterns,
20776 and split ones in order to take care of the indirect sibcalls.
20778 2022-06-15 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
20780 * doc/invoke.texi: Document -mextra-l32r-costs= option.
20782 2022-06-15 David Malcolm <dmalcolm@redhat.com>
20785 * doc/invoke.texi: Add -fno-analyzer-undo-inlining.
20786 * tree-diagnostic-path.cc (default_tree_diagnostic_path_printer):
20787 Extend -fdiagnostics-path-format=separate-events so that with
20788 -fdiagnostics-show-path-depths it prints fndecls as well as stack
20791 2022-06-15 David Malcolm <dmalcolm@redhat.com>
20793 * value-relation.h: Add "final" and "override" to relation_oracle
20794 vfunc implementations as appropriate.
20796 2022-06-15 Richard Sandiford <richard.sandiford@arm.com>
20798 PR middle-end/105975
20799 Revert everything apart from the expand_fn_using_insn and
20800 expand_direct_optab_fn changes from:
20801 * internal-fn.def (DEF_INTERNAL_INSN_FN): New macro.
20802 (GOMP_SIMT_ENTER_ALLOC, GOMP_SIMT_EXIT, GOMP_SIMT_LANE)
20803 (GOMP_SIMT_LAST_LANE, GOMP_SIMT_ORDERED_PRED, GOMP_SIMT_VOTE_ANY)
20804 (GOMP_SIMT_XCHG_BFLY, GOMP_SIMT_XCHG_IDX): Use it.
20805 * internal-fn.h (direct_internal_fn_info::directly_mapped): New
20807 (direct_internal_fn_info::vectorizable): Reduce to 1 bit.
20808 (direct_internal_fn_p): Also return true for internal functions
20809 that map directly to instructions defined target-insns.def.
20810 (direct_internal_fn): Adjust comment accordingly.
20811 * internal-fn.cc (direct_insn, optab1, optab2, vectorizable_optab1)
20812 (vectorizable_optab2): New local macros.
20813 (not_direct): Initialize directly_mapped.
20814 (mask_load_direct, load_lanes_direct, mask_load_lanes_direct)
20815 (gather_load_direct, len_load_direct, mask_store_direct)
20816 (store_lanes_direct, mask_store_lanes_direct, vec_cond_mask_direct)
20817 (vec_cond_direct, scatter_store_direct, len_store_direct)
20818 (vec_set_direct, unary_direct, binary_direct, ternary_direct)
20819 (cond_unary_direct, cond_binary_direct, cond_ternary_direct)
20820 (while_direct, fold_extract_direct, fold_left_direct)
20821 (mask_fold_left_direct, check_ptrs_direct): Use the macros above.
20822 (expand_GOMP_SIMT_ENTER_ALLOC, expand_GOMP_SIMT_EXIT): Delete
20823 (expand_GOMP_SIMT_LANE, expand_GOMP_SIMT_LAST_LANE): Likewise;
20824 (expand_GOMP_SIMT_ORDERED_PRED, expand_GOMP_SIMT_VOTE_ANY): Likewise.
20825 (expand_GOMP_SIMT_XCHG_BFLY, expand_GOMP_SIMT_XCHG_IDX): Likewise.
20826 (direct_internal_fn_types): Handle functions that map to instructions
20827 defined in target-insns.def.
20828 (direct_internal_fn_types): Likewise.
20829 (direct_internal_fn_supported_p): Likewise.
20830 (internal_fn_expanders): Likewise.
20831 (expand_fn_using_insn): New function,
20832 split out and adapted from...
20833 (expand_direct_optab_fn): ...here.
20834 (expand_GOMP_SIMT_ENTER_ALLOC): Use it.
20835 (expand_GOMP_SIMT_EXIT): Likewise.
20836 (expand_GOMP_SIMT_LANE): Likewise.
20837 (expand_GOMP_SIMT_LAST_LANE): Likewise.
20838 (expand_GOMP_SIMT_ORDERED_PRED): Likewise.
20839 (expand_GOMP_SIMT_VOTE_ANY): Likewise.
20840 (expand_GOMP_SIMT_XCHG_BFLY): Likewise.
20841 (expand_GOMP_SIMT_XCHG_IDX): Likewise.
20843 2022-06-15 Richard Earnshaw <rearnsha@arm.com>
20846 * config/arm/arm.cc (gen_cpymem_ldrd_strd): Rename low_reg and hi_reg
20847 to first_reg and second_reg respectively. Initialize them correctly
20848 when generating big-endian code.
20850 2022-06-15 Richard Earnshaw <rearnsha@arm.com>
20853 * config/arm/arm.cc (arm_bfi_1_p): Use UINTVAL instead of XUINT.
20855 2022-06-15 Richard Biener <rguenther@suse.de>
20857 PR tree-optimization/105971
20858 * tree-ssa-alias.cc (refs_may_alias_p_2): Put bail-out for
20859 FUNCTION_DECL and LABEL_DECL refs after decl-decl disambiguation
20860 to leak less surprising alias results.
20862 2022-06-15 Richard Biener <rguenther@suse.de>
20864 PR tree-optimization/105969
20865 * gimple-ssa-sprintf.cc (get_origin_and_offset_r): Avoid division
20866 by zero in overflow check.
20868 2022-06-15 Richard Sandiford <richard.sandiford@arm.com>
20870 PR tree-optimization/105254
20871 PR tree-optimization/105940
20873 * config/aarch64/aarch64.cc
20874 (aarch64_vector_costs::determine_suggested_unroll_factor): Take a
20875 loop_vec_info as argument. Restrict the unroll factor to values
20876 that divide the VF.
20877 (aarch64_vector_costs::finish_cost): Update call accordingly.
20879 2022-06-15 Richard Sandiford <richard.sandiford@arm.com>
20881 * read-rtl.cc (find_int): Substitute symbolic constants
20882 before converting the string to an integer.
20884 2022-06-15 Roger Sayle <roger@nextmovesoftware.com>
20885 Richard Biener <rguenther@suse.de>
20887 * match.pd (convert (lshift @1 INTEGER_CST@2)): Narrow integer
20888 left shifts by a constant when the result is truncated, and the
20889 shift constant is well-defined.
20890 * tree-vect-patterns.cc (vect_recog_rotate_pattern): Add
20891 support for rotations of signed integer types, by lowering
20892 using unsigned vector shifts.
20894 2022-06-15 liuhongt <hongtao.liu@intel.com>
20897 * config/i386/sse.md (*avx_cmp<mode>3_ltint_not): Force_reg
20900 2022-06-14 Surya Kumari Jangala <jskumari@linux.ibm.com>
20902 PR rtl-optimization/105041
20903 * regrename.cc (check_new_reg_p): Use nregs value from du chain.
20905 2022-06-14 Segher Boessenkool <segher@kernel.crashing.org>
20907 * config/rs6000/vsx.md (VS_scalar): Delete.
20908 (rest of file): Adjust.
20910 2022-06-14 Jan Hubicka <hubicka@ucw.cz>
20913 * ipa-prop.cc (ipa_load_from_parm_agg): Punt on volatile loads.
20915 2022-06-14 Philipp Tomsich <philipp.tomsich@vrull.eu>
20917 * config/riscv/bitmanip.md: Add split to handle opportunities
20918 for slli + sh[123]add.uw
20920 2022-06-14 Philipp Tomsich <philipp.tomsich@vrull.eu>
20922 * config/riscv/predicates.md (consecutive_bits_operand):
20923 Implement new predicate.
20925 2022-06-14 Richard Biener <rguenther@suse.de>
20927 PR tree-optimization/105946
20928 * tree-ssa-uninit.cc (maybe_warn_pass_by_reference):
20929 Do not look at arguments not specified in the function call.
20931 2022-06-14 Richard Biener <rguenther@suse.de>
20933 PR middle-end/105965
20934 * match.pd (view_convert CONSTRUCTOR): Handle single-element
20937 2022-06-14 Eric Botcazou <ebotcazou@adacore.com>
20939 * warning-control.cc (copy_warning) [generic version]: Do not erase
20940 the warning data of the destination location when the no-warning
20941 bit is not set on the source.
20942 (copy_warning) [tree version]: Return early if TO is equal to FROM.
20943 (copy_warning) [gimple version]: Likewise.
20945 2022-06-14 Kewen Lin <linkw@linux.ibm.com>
20947 PR tree-optimization/105940
20948 * tree-vect-loop.cc (vect_analyze_loop_2): Move the place of
20949 applying suggested_unroll_factor after start_over.
20951 2022-06-14 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
20953 * config/xtensa/predicates.md (shifted_mask_operand):
20955 * config/xtensa/xtensa.md (*andsi3_const_pow2_minus_one):
20956 New insn-and-split pattern.
20957 (*andsi3_const_negative_pow2, *andsi3_const_shifted_mask,
20958 *masktrue_const_pow2_minus_one, *masktrue_const_negative_pow2,
20959 *masktrue_const_shifted_mask): Ditto.
20961 2022-06-14 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
20963 * config/xtensa/xtensa.md (*masktrue_bitcmpl): New insn pattern.
20965 2022-06-14 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
20967 * config/xtensa/xtensa-protos.h (xtensa_emit_branch):
20968 Remove the first argument.
20969 (xtensa_emit_bit_branch): Remove it because now called only from the
20970 output statement of *bittrue insn pattern.
20971 * config/xtensa/xtensa.cc (gen_int_relational): Remove the last
20972 argument 'p_invert', and make so that the condition is reversed by
20974 (xtensa_expand_conditional_branch): Share the common path, and remove
20975 condition inversion code.
20976 (xtensa_emit_branch, xtensa_emit_movcc): Simplify by removing the
20977 "false side" pattern.
20978 (xtensa_emit_bit_branch): Remove it because of the abovementioned
20979 reason, and move the function body to *bittrue insn pattern.
20980 * config/xtensa/xtensa.md (*bittrue): Transplant the output
20981 statement from removed xtensa_emit_bit_branch().
20982 (*bfalse, *ubfalse, *bitfalse, *maskfalse): Remove the "false side"
20985 2022-06-14 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
20987 * config/xtensa/predicates.md (logical_shift_operator,
20988 xtensa_shift_per_byte_operator): New predicates.
20989 * config/xtensa/xtensa-protos.h (xtensa_shlrd_which_direction):
20991 * config/xtensa/xtensa.cc (xtensa_shlrd_which_direction):
20992 New helper function for funnel shift patterns.
20993 * config/xtensa/xtensa.md (ior_op): New code iterator.
20994 (*ashlsi3_1): Replace with new split pattern.
20995 (*shift_per_byte): Unify *ashlsi3_3x, *ashrsi3_3x and *lshrsi3_3x.
20996 (*shift_per_byte_omit_AND_0, *shift_per_byte_omit_AND_1):
20997 New insn-and-split patterns that redirect to *xtensa_shift_per_byte,
20998 in order to omit unnecessary bitwise AND operation.
20999 (*shlrd_reg_<code>, *shlrd_const_<code>, *shlrd_per_byte_<code>,
21000 *shlrd_per_byte_<code>_omit_AND):
21001 New insn patterns for funnel shifts.
21003 2022-06-13 Jason Merrill <jason@redhat.com>
21005 * tree-cfg.cc (pass_warn_function_return::execute): Also check
21008 2022-06-13 Maciej W. Rozycki <macro@embecosm.com>
21010 * config/riscv/riscv.md (length): Remove the explicit setting
21013 2022-06-13 H.J. Lu <hjl.tools@gmail.com>
21015 * common/config/i386/cpuinfo.h (get_available_features): Require
21016 AVX for F16C and VAES.
21018 2022-06-13 Uroš Bizjak <ubizjak@gmail.com>
21021 * config/i386/predicates.md (register_no_elim_operand):
21022 Return true for subreg of a memory operand.
21024 2022-06-13 Richard Sandiford <richard.sandiford@arm.com>
21026 * internal-fn.def (DEF_INTERNAL_INSN_FN): New macro.
21027 (GOMP_SIMT_ENTER_ALLOC, GOMP_SIMT_EXIT, GOMP_SIMT_LANE)
21028 (GOMP_SIMT_LAST_LANE, GOMP_SIMT_ORDERED_PRED, GOMP_SIMT_VOTE_ANY)
21029 (GOMP_SIMT_XCHG_BFLY, GOMP_SIMT_XCHG_IDX): Use it.
21030 * internal-fn.h (direct_internal_fn_info::directly_mapped): New
21032 (direct_internal_fn_info::vectorizable): Reduce to 1 bit.
21033 (direct_internal_fn_p): Also return true for internal functions
21034 that map directly to instructions defined target-insns.def.
21035 (direct_internal_fn): Adjust comment accordingly.
21036 * internal-fn.cc (direct_insn, optab1, optab2, vectorizable_optab1)
21037 (vectorizable_optab2): New local macros.
21038 (not_direct): Initialize directly_mapped.
21039 (mask_load_direct, load_lanes_direct, mask_load_lanes_direct)
21040 (gather_load_direct, len_load_direct, mask_store_direct)
21041 (store_lanes_direct, mask_store_lanes_direct, vec_cond_mask_direct)
21042 (vec_cond_direct, scatter_store_direct, len_store_direct)
21043 (vec_set_direct, unary_direct, binary_direct, ternary_direct)
21044 (cond_unary_direct, cond_binary_direct, cond_ternary_direct)
21045 (while_direct, fold_extract_direct, fold_left_direct)
21046 (mask_fold_left_direct, check_ptrs_direct): Use the macros above.
21047 (expand_GOMP_SIMT_ENTER_ALLOC, expand_GOMP_SIMT_EXIT): Delete
21048 (expand_GOMP_SIMT_LANE, expand_GOMP_SIMT_LAST_LANE): Likewise;
21049 (expand_GOMP_SIMT_ORDERED_PRED, expand_GOMP_SIMT_VOTE_ANY): Likewise.
21050 (expand_GOMP_SIMT_XCHG_BFLY, expand_GOMP_SIMT_XCHG_IDX): Likewise.
21051 (direct_internal_fn_types): Handle functions that map to instructions
21052 defined in target-insns.def.
21053 (direct_internal_fn_types): Likewise.
21054 (direct_internal_fn_supported_p): Likewise.
21055 (internal_fn_expanders): Likewise.
21057 2022-06-13 Richard Sandiford <richard.sandiford@arm.com>
21059 * internal-fn.cc (expand_fn_using_insn): New function,
21060 split out and adapted from...
21061 (expand_direct_optab_fn): ...here.
21062 (expand_GOMP_SIMT_ENTER_ALLOC): Use it.
21063 (expand_GOMP_SIMT_EXIT): Likewise.
21064 (expand_GOMP_SIMT_LANE): Likewise.
21065 (expand_GOMP_SIMT_LAST_LANE): Likewise.
21066 (expand_GOMP_SIMT_ORDERED_PRED): Likewise.
21067 (expand_GOMP_SIMT_VOTE_ANY): Likewise.
21068 (expand_GOMP_SIMT_XCHG_BFLY): Likewise.
21069 (expand_GOMP_SIMT_XCHG_IDX): Likewise.
21071 2022-06-13 Jakub Jelinek <jakub@redhat.com>
21073 * omp-expand.cc (expand_omp_target): Remap user provided
21074 device clause arguments, -1 to -2 and -2 to -3, either
21075 at compile time if constant, or at runtime.
21077 2022-06-13 Eric Botcazou <ebotcazou@adacore.com>
21079 * common.opt (finstrument-functions): Set explicit value.
21080 (-finstrument-functions-once): New option.
21081 * doc/invoke.texi (Program Instrumentation Options): Document it.
21082 * gimplify.cc (build_instrumentation_call): New static function.
21083 (gimplify_function_tree): Call it to emit the instrumentation calls
21084 if -finstrument-functions[-once] is specified.
21086 2022-06-13 Eric Botcazou <ebotcazou@adacore.com>
21088 * dwarf2out.cc (output_one_line_info_table): Initialize prev_addr.
21089 * gimple.h (gimple_set_location): Do not copy warning data from
21090 the previous location when it is UNKNOWN_LOCATION.
21091 * optabs.cc (expand_widen_pattern_expr): Always set oprnd{1,2}.
21093 2022-06-13 Jakub Jelinek <jakub@redhat.com>
21096 * config/i386/i386.md (*ashl<dwi>3_doubleword_mask,
21097 *<insn><dwi>3_doubleword_mask): Use operands[3] masked with
21098 (<MODE_SIZE> * BITS_PER_UNIT) - 1 as AND operand instead of
21099 operands[3] unmodified.
21101 2022-06-12 Simon Wright <simon@pushface.org>
21104 * config/darwin-driver.cc (darwin_find_version_from_kernel): If the OS
21105 version is darwin20 (macOS 11) or greater, truncate the version to the
21108 2022-06-12 Mark Mentovai <mark@mentovai.com>
21110 * config/darwin-c.cc: Make -mmacosx-version-min more future-proof.
21112 2022-06-12 Prathamesh Kulkarni <prathamesh.kulkarni@linaro.org>
21115 * config/aarch64/aarch64-sve-builtins-base.cc: Include ssa.h.
21116 (svld1rq_impl::fold): Define.
21117 * config/aarch64/aarch64.cc (expand_vec_perm_d): Define new members
21118 op_mode and op_vec_flags.
21119 (aarch64_evpc_reencode): Initialize newd.op_mode and
21121 (aarch64_evpc_sve_dup): New function.
21122 (aarch64_expand_vec_perm_const_1): Gate existing calls to
21123 aarch64_evpc_* functions under d->vmode == d->op_mode,
21124 and call aarch64_evpc_sve_dup.
21125 (aarch64_vectorize_vec_perm_const): Remove assert
21126 d->vmode != d->op_mode, and initialize d.op_mode and d.op_vec_flags.
21127 * tree-cfg.cc (verify_gimple_assign_ternary): Allow different
21128 vector types for lhs and rhs in VEC_PERM_EXPR if rhs3 is
21131 2022-06-11 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
21133 * config/xtensa/xtensa-protos.h (xtensa_constantsynth):
21135 * config/xtensa/xtensa.cc (xtensa_emit_constantsynth,
21136 xtensa_constantsynth_2insn, xtensa_constantsynth_rtx_SLLI,
21137 xtensa_constantsynth_rtx_ADDSUBX, xtensa_constantsynth):
21138 New backend functions that process the abovementioned logic.
21139 (xtensa_emit_move_sequence): Revert the previous changes.
21140 * config/xtensa/xtensa.md: New split patterns for integer
21141 and floating-point, as the frontend part.
21143 2022-06-11 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
21145 * config/xtensa/xtensa.cc (xtensa_rtx_costs): Correct wrong case
21146 for ABS and NEG, add missing case for BSWAP and CLRSB, and
21147 double the costs for integer divisions using libfuncs if
21148 optimizing for speed, in order to take advantage of fast constant
21149 division by multiplication.
21150 (TARGET_INSN_COST): New macro definition.
21151 (xtensa_is_insn_L32R_p, xtensa_insn_cost): New functions for
21152 calculating relative costs of a RTL insns, for both of speed and
21154 * config/xtensa/xtensa.md (return, nop, trap): Correct values of
21155 the attribute "length" that depends on TARGET_DENSITY.
21156 (define_asm_attributes, blockage, frame_blockage): Add missing
21158 * config/xtensa/xtensa.opt (-mextra-l32r-costs=): New machine-
21159 dependent option, however, preparatory work for now.
21161 2022-06-11 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
21163 * config/xtensa/xtensa.cc (xtensa_expand_block_set_small_loop):
21164 Pass through the block length / loop count conditions if
21165 zero-overhead looping is configured and active,
21167 2022-06-11 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
21169 * config/xtensa/xtensa.md (mulsidi3, umulsidi3):
21170 Split into individual signedness, in order to use libcall
21171 "__umulsidi3" but not the other.
21172 (<u>mulhisi3): Merge into one by using code iterator.
21173 (<u>mulsidi3, mulhisi3, umulhisi3): Remove.
21175 2022-06-11 Michael Meissner <meissner@linux.ibm.com>
21177 * config/rs6000/rs6000.cc (rs6000_option_override_internal): Do
21178 not generate block copies with vector pair instructions if we are
21179 tuning for power10.
21181 2022-06-10 Roger Sayle <roger@nextmovesoftware.com>
21183 PR rtl-optimization/7061
21184 * expr.cc (emit_group_store): For groups that consist of a single
21185 scalar integer register that hold a complex mode value, use
21186 gen_lowpart to generate a SUBREG to "view_convert" to the complex
21187 mode. For modes of different sizes, first convert to an integer
21188 mode of the appropriate size.
21190 2022-06-09 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
21192 * config/xtensa/xtensa.md (clrsbsi2): New insn pattern.
21194 2022-06-09 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
21196 * config/xtensa/xtensa.md (*andsi3_bitcmpl):
21197 New insn_and_split pattern.
21199 2022-06-09 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
21201 * config/xtensa/xtensa.md (one_cmplsi2):
21202 Rearrange as an insn_and_split pattern.
21204 2022-06-09 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
21206 * config/xtensa/xtensa.md (bswaphi2): New insn pattern.
21208 2022-06-09 Segher Boessenkool <segher@kernel.crashing.org>
21210 * config/rs6000/rs6000.md (FP_ISA3): Delete.
21211 (float<QHI:mode><FP_ISA3:mode>2): Rename to...
21212 (float<QHI:mode><SFDF:mode>2): ... this. Adjust.
21213 (*float<QHI:mode><FP_ISA3:mode>2_internal): Rename to...
21214 (*float<QHI:mode><SFDF:mode>2_internal): ... this. Adjust.
21215 (floatuns<QHI:mode><FP_ISA3:mode>2): Rename to...
21216 (floatuns<QHI:mode><SFDF:mode>2): ... this. Adjust.
21217 (*floatuns<QHI:mode><FP_ISA3:mode>2_internal): Rename to...
21218 (*floatuns<QHI:mode><SFDF:mode>2_internal): ... this. Adjust.
21220 2022-06-09 Maciej W. Rozycki <macro@embecosm.com>
21222 * config/riscv/riscv.md
21223 (*f<quiet_pattern>_quiet<ANYF:mode><X:mode>4_default)
21224 (*f<quiet_pattern>_quiet<ANYF:mode><X:mode>4_snan): Emit a tab
21225 rather than space with FSFLAGS.
21227 2022-06-09 Tobias Burnus <tobias@codesourcery.com>
21229 * omp-offload.cc (omp_discover_declare_target_tgt_fn_r,
21230 omp_discover_declare_target_fn_r): Don't walk reverse-offload
21233 2022-06-09 Jakub Jelinek <jakub@redhat.com>
21235 * doc/invoke.texi (-Waddress): Fix a typo in small example.
21236 Fix typos inptr_t -> intptr_t and uinptr_t -> uintptr_t.
21238 2022-06-09 Cui,Lili <lili.cui@intel.com>
21241 * config/i386/x86-tune-costs.h (skylake_cost): Raise the gpr load cost
21242 from 4 to 6 and gpr store cost from 6 to 8. Change SSE loads and
21243 unaligned loads cost from {6, 6, 6, 10, 20} to {8, 8, 8, 8, 16}.
21244 (icelake_cost): Ditto.
21245 (alderlake_cost): Raise the gpr store cost from 6 to 8 and SSE loads,
21246 stores and unaligned stores cost from {6, 6, 6, 10, 15} to
21249 2022-06-09 Haochen Gui <guihaoc@gcc.gnu.org>
21251 * config/rs6000/rs6000.md (define_split for bswapdi load): Merge shift
21252 and ior insns to one rotate and mask insn.
21253 (define_split for bswapdi register): Likewise.
21255 2022-06-08 Roger Sayle <roger@nextmovesoftware.com>
21257 PR middle-end/105874
21258 * expr.cc (expand_expr_real_1) <normal_inner_ref>: New local
21259 variable tem_modifier for calculating the expand_modifier enum to
21260 use for expanding tem. If tem is a VAR_DECL, use EXPAND_MEMORY.
21262 2022-06-08 Max Filippov <jcmvbkbc@gmail.com>
21265 * config/xtensa/xtensa.md (movdi): Rename 'first' and 'second'
21266 to 'lowpart' and 'highpart' so that they match 'gen_lowpart' and
21267 'gen_highpart' bitwise semantics and fix order of highpart and
21268 lowpart depending on target endianness.
21270 2022-06-08 Chung-Ju Wu <jasonwucj@gmail.com>
21272 * config/arm/arm-cpus.in (star-mc1): New cpu.
21273 * config/arm/arm-tables.opt: Regenerate.
21274 * config/arm/arm-tune.md: Regenerate.
21275 * doc/invoke.texi: Update docs.
21277 2022-06-08 liuhongt <hongtao.liu@intel.com>
21281 * config/i386/i386.md (*movsi_internal): Change alternative
21283 (*movdi_internal): Ditto.
21284 * config/i386/sse.md (vec_set<mode>_0): Change alternative *r
21286 (*vec_extractv4sf_mem): Ditto.
21287 (*vec_extracthf): Ditto.
21289 2022-06-07 Richard Earnshaw <rearnsha@arm.com>
21292 * config/arm/arm.cc (arm_bfi_1_p): New function.
21293 (arm_bfi_p): New function.
21294 (arm_rtx_costs_internal): Add costs for BFI idioms.
21295 (arm_print_operand [case 'V']): Format output for BFI/BFC masks.
21296 * config/arm/constraints.md (Dj): New constraint.
21297 * config/arm/arm.md (arm_andsi3_insn): Add alternative to use BFC.
21298 (insv_zero): Convert to an insn with a split.
21299 (*bfi, *bfi_alt1, *bfi_alt2, *bfi_alt3): New patterns.
21301 2022-06-07 liuhongt <hongtao.liu@intel.com>
21304 * config/i386/sse.md (ssse3_palignrdi): Change alternative 2
21307 2022-06-07 Roger Sayle <roger@nextmovesoftware.com>
21309 PR middle-end/105853
21311 * calls.cc (load_register_parameters): Call store_constructor
21312 and int_expr_size directly instead of expanding via expand_expr.
21313 * expr.cc (static void store_constructor): Don't prototype here.
21314 (static HOST_WIDE_INT int_expr_size): Likewise.
21315 (store_constructor): No longer static.
21316 (int_expr_size): Likewise, no longer static.
21317 * expr.h (store_constructor): Prototype here.
21318 (int_expr_size): Prototype here.
21320 2022-06-07 Jan Beulich <jbeulich@suse.com>
21323 2022-06-03 Jan Beulich <jbeulich@suse.com>
21325 * Makefile.in (ORIGINAL_OBJCOPY_FOR_TARGET): New.
21326 * configure.ac: Check for objcopy, producing
21327 ORIGINAL_OBJCOPY_FOR_TARGET.
21328 * configure: Update accordingly.
21329 * exec-tool.in (ORIGINAL_OBJCOPY_FOR_TARGET): New.
21332 2022-06-07 Jakub Jelinek <jakub@redhat.com>
21334 * tree.h (OMP_CLAUSE_LINEAR_OLD_LINEAR_MODIFIER): Define.
21335 * tree-pretty-print.cc (dump_omp_clause) <case OMP_CLAUSE_LINEAR>:
21336 Adjust clause printing style depending on
21337 OMP_CLAUSE_LINEAR_OLD_LINEAR_MODIFIER.
21339 2022-06-07 Jan Beulich <jbeulich@suse.com>
21341 * config/i386/i386-builtin.def (__builtin_ia32_psadbw256):
21343 * config/i386/i386-builtin-types.def: New function type
21344 (V4DI, V32QI, V32QI).
21345 * config/i386/i386-expand.cc (ix86_expand_args_builtin): Handle
21346 V4DI_FTYPE_V32QI_V32QI.
21348 2022-06-07 Jan Beulich <jbeulich@suse.com>
21350 * config/i386/i386.cc (ix86_attr_length_vex_default): Take REX.B
21351 into account for reg-only insns.
21353 2022-06-07 Roger Sayle <roger@nextmovesoftware.com>
21355 * config/i386/i386.cc (ix86_rtx_costs): Add a new case for
21356 IF_THEN_ELSE, and provide costs for TARGET_XOP's vpcmov and
21357 TARGET_CMOVE's (scalar integer) conditional moves.
21358 * config/i386/sse.md (define_split): Recognize XOP's vpcmov
21359 from its equivalent (canonical) pxor;pand;pxor sequence.
21361 2022-06-07 Kewen Lin <linkw@linux.ibm.com>
21363 * machmode.def (VECTOR_MODES_WITH_PREFIX): Update document for
21366 2022-06-06 Andrew Stubbs <ams@codesourcery.com>
21368 * config.in: Regenerate.
21369 * configure: Regenerate.
21370 * configure.ac: Reinstate HAVE_GAS_ARM_EXTENDED_ARCH test.
21372 2022-06-04 Roger Sayle <roger@nextmovesoftware.com>
21374 PR middle-end/95126
21375 * calls.cc (load_register_parameters): When loading a suitable
21376 immediate_const_ctor_p VAR_DECL into a single word_mode register,
21377 construct it directly in a pseudo rather than read it (by parts)
21379 * expr.cc (int_expr_size): Make tree argument a const_tree.
21380 (immediate_const_ctor_p): Helper predicate. Return true for
21381 simple constructors that may be materialized in a register.
21382 (expand_expr_real_1) [VAR_DECL]: When expanding a constant
21383 VAR_DECL with a suitable immediate_const_ctor_p constructor
21384 use store_constructor to materialize it directly in a pseudo.
21385 * expr.h (immediate_const_ctor_p): Prototype here.
21386 * varasm.cc (initializer_constant_valid_for_bitfield_p): Change
21387 VALUE argument from tree to const_tree.
21388 * varasm.h (initializer_constant_valid_for_bitfield_p): Update
21391 2022-06-04 Jakub Jelinek <jakub@redhat.com>
21394 * config/i386/i386.md (*ashl<dwi>3_doubleword_mask,
21395 *<insn><dwi>3_doubleword_mask): If top bit of mask is clear, but lower
21396 bits of mask aren't all set, use operands[2] mode for the AND
21397 operation instead of always SImode.
21399 2022-06-03 Jakub Jelinek <jakub@redhat.com>
21401 PR middle-end/30314
21402 PR middle-end/105777
21403 * match.pd (__builtin_mul_overflow_p (x, cst, (stype) 0) ->
21404 x > stype_max / cst || x < stype_min / cst): New simplification.
21406 2022-06-03 Aldy Hernandez <aldyh@redhat.com>
21408 * gimple-range-cache.cc (ranger_cache::range_from_dom): Use
21410 * gimple-range.cc (gimple_ranger::register_inferred_ranges): Same.
21411 * value-range.h (Value_Range::Value_Range): Implement copy
21412 constructor for Value_Range.
21414 2022-06-03 Aldy Hernandez <aldyh@redhat.com>
21416 * value-range.h (struct vrange_traits): Remove.
21417 (is_a): Rewrite without vrange_traits.
21420 2022-06-03 Aldy Hernandez <aldyh@redhat.com>
21422 * value-range.cc (vrange::contains_p): Implement.
21423 (vrange::type): Return void.
21424 (vrange::supports_type_p): Implement.
21425 (irange::fits_p): Same.
21426 (vrange::set_undefined): Same.
21427 (irange::set_nonnegative): Same.
21428 (vrange::set_varying): Same.
21429 (vrange::union_): Same.
21430 (unsupported_range::set): Move to vrange.
21431 (unsupported_range::type): Move to vrange.
21432 (vrange::intersect): Implement for varying and undefined.
21433 (vrange::zero_p): Implement.
21434 (unsupported_range::supports_type_p): Move to vrange.
21435 (vrange::nonzero_p): Implement.
21436 (unsupported_range::set_undefined): Move to vrange.
21437 (unsupported_range::set_varying): Same.
21438 (unsupported_range::dump): Same.
21439 (unsupported_range::union_): Same. Implement for varying and
21441 (unsupported_range::intersect): Move to vrange.
21442 (unsupported_range::zero_p): Same.
21443 (unsupported_range::nonzero_p): Same.
21444 (unsupported_range::set_nonzero): Same.
21445 (unsupported_range::set_zero): Same.
21446 (unsupported_range::set_nonnegative): Same.
21447 (unsupported_range::fits_p): Same.
21448 * value-range.h (class vrange): Remove abstract markers for most
21450 (class unsupported_range): Remove most methods as they will now be
21451 inherited from vrange.
21453 2022-06-03 Aldy Hernandez <aldyh@redhat.com>
21455 * gimple-range-edge.cc (gimple_outgoing_range_stmt_p): Adjust for
21456 an object level supports_type_p for irange and a static
21457 Value_Range::supports_type_p.
21458 * gimple-range-fold.cc (fold_using_range::range_of_range_op): Same.
21459 (fold_using_range::range_of_address): Same.
21460 (fold_using_range::range_of_builtin_call): Same.
21461 * gimple-range-fold.h (gimple_range_type): Same.
21462 (gimple_range_ssa_p): Same.
21463 * gimple-range-path.cc (path_range_query::internal_range_of_expr):
21465 (path_range_query::range_of_stmt): Same.
21466 (path_range_query::add_to_imports): Same.
21467 * gimple-range.cc (gimple_ranger::range_on_edge): Same.
21468 (gimple_ranger::export_global_ranges): Same.
21469 * gimple-ssa-evrp-analyze.cc
21470 (evrp_range_analyzer::record_ranges_from_phis): Same.
21471 * range-op.cc (range_operator::wi_fold): Same.
21472 (range_operator::fold_range): Same.
21473 * tree-ssa-loop-ch.cc (entry_loop_condition_is_static): Same.
21474 * tree-ssa-loop-unswitch.cc (struct unswitch_predicate): Same.
21475 (evaluate_control_stmt_using_entry_checks): Same.
21476 * tree-ssa-threadedge.cc
21477 (hybrid_jt_simplifier::compute_ranges_from_state): Same.
21478 * tree-vrp.cc (supported_types_p): Same.
21479 * value-query.cc (range_query::value_of_expr): Same.
21480 (range_query::value_on_edge): Same.
21481 (range_query::value_of_stmt): Same.
21482 (range_query::get_tree_range): Same.
21483 (get_range_global): Same.
21484 (global_range_query::range_of_expr): Same.
21485 * value-range-equiv.h (class value_range_equiv): Same.
21486 * value-range.cc (irange::supports_type_p): Same.
21487 (unsupported_range::supports_type_p): Same.
21488 * value-range.h (enum value_range_discriminator): Same.
21489 (Value_Range::init): Same.
21490 (Value_Range::supports_type_p): Same.
21491 (irange::supports_type_p): Same.
21492 (irange::supports_p): Same.
21493 (vrange::supports_type_p): Same.
21494 (vrange_allocator::alloc_vrange): Same.
21496 2022-06-03 Jan Beulich <jbeulich@suse.com>
21498 * Makefile.in (ORIGINAL_OBJCOPY_FOR_TARGET): New.
21499 * configure.ac: Check for objcopy, producing
21500 ORIGINAL_OBJCOPY_FOR_TARGET.
21501 * configure: Update accordingly.
21502 * exec-tool.in (ORIGINAL_OBJCOPY_FOR_TARGET): New.
21505 2022-06-03 Jan Beulich <jbeulich@suse.com>
21507 * config/i386/mmx.md (mmx_psadbw): Convert to expander.
21508 (*mmx_psadbw): New. Mark as commutative.
21509 * config/i386/sse.md (<sse2_avx2>_psadbw): Convert to expander.
21510 (*<sse2_avx2>_psadbw): New. Mark as commutative.
21512 2022-06-03 Alexandre Oliva <oliva@adacore.com>
21514 PR tree-optimization/105665
21515 PR tree-optimization/100810
21516 * tree-ssa-loop-ivopts.cc
21517 (ssa_name_maybe_undef_p, ssa_name_set_maybe_undef): New.
21518 (ssa_name_any_use_dominates_bb_p, mark_ssa_maybe_undefs): New.
21519 (find_ssa_undef): Check precomputed flag and intervening uses.
21520 (tree_ssa_iv_optimize): Call mark_ssa_maybe_undefs.
21522 2022-06-02 David Malcolm <dmalcolm@redhat.com>
21524 * Makefile.in (OBJS): Add tree-diagnostic-client-data-hooks.o and
21525 tree-logical-location.o.
21526 (OBJS-libcommon): Add diagnostic-format-sarif.o; reorder.
21527 (CFLAGS-tree-diagnostic-client-data-hooks.o): Add TARGET_NAME.
21528 * common.opt (fdiagnostics-format=): Add sarif-stderr and sarif-file.
21529 (sarif-stderr, sarif-file): New enum values.
21530 * diagnostic-client-data-hooks.h: New file.
21531 * diagnostic-format-sarif.cc: New file.
21532 * diagnostic-path.h (enum diagnostic_event::verb): New enum.
21533 (enum diagnostic_event::noun): New enum.
21534 (enum diagnostic_event::property): New enum.
21535 (struct diagnostic_event::meaning): New struct.
21536 (diagnostic_event::get_logical_location): New vfunc.
21537 (diagnostic_event::get_meaning): New vfunc.
21538 (simple_diagnostic_event::get_logical_location): New vfunc impl.
21539 (simple_diagnostic_event::get_meaning): New vfunc impl.
21540 * diagnostic.cc: Include "diagnostic-client-data-hooks.h".
21541 (diagnostic_initialize): Initialize m_client_data_hooks.
21542 (diagnostic_finish): Clean up m_client_data_hooks.
21543 (diagnostic_event::meaning::dump_to_pp): New.
21544 (diagnostic_event::meaning::maybe_get_verb_str): New.
21545 (diagnostic_event::meaning::maybe_get_noun_str): New.
21546 (diagnostic_event::meaning::maybe_get_property_str): New.
21547 (get_cwe_url): Make non-static.
21548 (diagnostic_output_format_init): Handle
21549 DIAGNOSTICS_OUTPUT_FORMAT_SARIF_STDERR and
21550 DIAGNOSTICS_OUTPUT_FORMAT_SARIF_FILE.
21551 * diagnostic.h (enum diagnostics_output_format): Add
21552 DIAGNOSTICS_OUTPUT_FORMAT_SARIF_STDERR and
21553 DIAGNOSTICS_OUTPUT_FORMAT_SARIF_FILE.
21554 (class diagnostic_client_data_hooks): New forward decl.
21555 (class logical_location): New forward decl.
21556 (diagnostic_context::m_client_data_hooks): New field.
21557 (diagnostic_output_format_init_sarif_stderr): New decl.
21558 (diagnostic_output_format_init_sarif_file): New decl.
21559 (get_cwe_url): New decl.
21560 * doc/invoke.texi (-fdiagnostics-format=): Add sarif-stderr and
21562 * doc/sourcebuild.texi (Scan a particular file): Add
21563 scan-sarif-file and scan-sarif-file-not.
21564 * langhooks-def.h (lhd_get_sarif_source_language): New decl.
21565 (LANG_HOOKS_GET_SARIF_SOURCE_LANGUAGE): New macro.
21566 (LANG_HOOKS_INITIALIZER): Add
21567 LANG_HOOKS_GET_SARIF_SOURCE_LANGUAGE.
21568 * langhooks.cc (lhd_get_sarif_source_language): New.
21569 * langhooks.h (lang_hooks::get_sarif_source_language): New field.
21570 * logical-location.h: New file.
21571 * plugin.cc (struct for_each_plugin_closure): New.
21572 (for_each_plugin_cb): New.
21573 (for_each_plugin): New.
21574 * plugin.h (for_each_plugin): New decl.
21575 * tree-diagnostic-client-data-hooks.cc: New file.
21576 * tree-diagnostic.cc: Include "diagnostic-client-data-hooks.h".
21577 (tree_diagnostics_defaults): Populate m_client_data_hooks.
21578 * tree-logical-location.cc: New file.
21579 * tree-logical-location.h: New file.
21581 2022-06-02 David Malcolm <dmalcolm@redhat.com>
21583 * common.opt (fdiagnostics-format=): Add json-stderr and json-file
21585 (DIAGNOSTICS_OUTPUT_FORMAT_JSON): Rename to...
21586 (DIAGNOSTICS_OUTPUT_FORMAT_JSON_STDERR): ...this.
21587 (diagnostics_output_format): Add json-stderr and json-file.
21588 * diagnostic-format-json.cc (json_flush_to_file): New.
21589 (json_final_cb): Convert to...
21590 (json_flush_to_file): ...this, ...
21591 (json_stderr_final_cb): ...this, and...
21592 (json_file_final_cb): ...this.
21593 (diagnostic_output_format_init): Move to diagnostic.cc.
21594 (json_output_base_file_name): New.
21595 (diagnostic_output_format_init_json): New.
21596 (diagnostic_output_format_init_json_stderr): New.
21597 (diagnostic_output_format_init_json_file): New.
21598 * diagnostic.cc (diagnostic_output_format_init): Move here from
21599 diagnostic-format-json.cc; update for changes to enum.
21600 * diagnostic.h (enum diagnostics_output_format): Rename
21601 DIAGNOSTICS_OUTPUT_FORMAT_JSON to
21602 DIAGNOSTICS_OUTPUT_FORMAT_JSON_STDERR, and add
21603 DIAGNOSTICS_OUTPUT_FORMAT_JSON_FILE.
21604 (diagnostic_output_format_init): Add base_file_name param.
21605 (diagnostic_output_format_init_json_stderr): New decl.
21606 (diagnostic_output_format_init_json_file): New dec.
21607 * doc/invoke.texi (-fdiagnostics-format=): Add "json-stderr" and
21608 "json-file". Rewrite so that the existing "json" is a synonym of
21610 * gcc.cc (driver_handle_option): Pass dump_base_name to
21611 diagnostic_output_format_init.
21612 * opts.cc (common_handle_option): Likewise.
21614 2022-06-02 David Malcolm <dmalcolm@redhat.com>
21616 * json.cc (string::print): Fix escaping of '\'.
21618 2022-06-02 Philipp Tomsich <philipp.tomsich@vrull.eu>
21620 * config/riscv/riscv.cc (riscv_build_integer_1): Rewrite value as
21621 (-1 << 31) for the single-bit case, when operating on (1 << 31)
21623 * config/riscv/riscv.h (SINGLE_BIT_MASK_OPERAND): Allow for
21624 any single-bit value, moving the special case for (1 << 31) to
21625 riscv_build_integer_1 (in riscv.c).
21627 2022-06-02 Roger Sayle <roger@nextmovesoftware.com>
21630 * config/i386/sse.md (V_128_256):Add V1TI and V2TI.
21631 (define_mode_attr avxsizesuffix): Add support for V1TI and V2TI.
21633 2022-06-02 Jakub Jelinek <jakub@redhat.com>
21636 * config/i386/i386.md (*ashl<dwi>3_doubleword_mask): Remove :SI
21637 from AND and its operands and just verify operands[2] has HImode,
21638 SImode or for TARGET_64BIT DImode. Allow operands[3] to be a mask
21639 with all low 6 (64-bit) or 5 (32-bit) bits set and in that case
21640 just throw away the masking. Use force_reg before calling
21642 (*ashl<dwi>3_doubleword_mask_1): Allow operands[3] to be a mask
21643 with all low 6 (64-bit) or 5 (32-bit) bits set and in that case
21644 just throw away the masking.
21645 (*ashl<mode>3_doubleword): Rename to ...
21646 (ashl<mode>3_doubleword): ... this.
21647 (*ashl<mode>3_mask): Remove :SI from AND and its operands and just
21648 verify operands[2] has HImode, SImode or for TARGET_64BIT DImode.
21649 Use force_reg before calling gen_lowpart.
21650 (*<insn><mode>3_mask): Likewise.
21651 (*<insn><dwi>3_doubleword_mask): Likewise. Allow operands[3] to be
21652 a mask with all low 6 (64-bit) or 5 (32-bit) bits set and in that
21653 case just throw away the masking. Use force_reg before calling
21655 (*<insn><dwi>3_doubleword_mask_1): Allow operands[3] to be a mask
21656 with all low 6 (64-bit) or 5 (32-bit) bits set and in that case just
21657 throw away the masking.
21658 (*<insn><mode>3_doubleword): Rename to ...
21659 (<insn><mode>3_doubleword): ... this.
21660 (*<insn><mode>3_mask): Remove :SI from AND and its operands and just
21661 verify operands[2] has HImode, SImode or for TARGET_64BIT DImode.
21662 Use force_reg before calling gen_lowpart.
21663 (splitter after it): Remove :SI from AND and its operands and just
21664 verify operands[2] has HImode, SImode or for TARGET_64BIT DImode.
21665 (*<btsc><mode>_mask, *<btsc><mode>_mask): Remove :SI from AND and its
21666 operands and just verify operands[1] has HImode, SImode or for
21667 TARGET_64BIT DImode. Use force_reg before calling gen_lowpart.
21668 (*jcc_bt<mode>_mask_1): New define_insn_and_split pattern.
21669 * config/i386/i386.cc (ix86_rtx_costs): For ZERO_EXTRACT with
21670 ZERO_EXTEND QI->SI in last operand ignore the cost of the ZERO_EXTEND.
21672 2022-06-02 Richard Biener <rguenther@suse.de>
21674 PR tree-optimization/101668
21675 * tree-vect-slp.cc (vect_build_slp_tree_1): Allow BIT_FIELD_REFs
21676 for vector types with compatible lane types.
21677 (vect_build_slp_tree_2): Deal with this.
21678 (vect_add_slp_permutation): Adjust. Emit lowpart/concat
21679 special cases without VEC_PERM.
21680 (vectorizable_slp_permutation): Select the operand vector
21681 type and relax requirements. Handle identity permutes
21682 with mismatching operand types.
21683 * optabs-query.cc (can_vec_perm_const_p): Only allow variable
21684 permutes for op_mode == mode.
21686 2022-06-02 Richard Biener <rguenther@suse.de>
21688 PR tree-optimization/105802
21689 * tree-ssa-loop-unswitch.cc (find_unswitching_predicates_for_bb):
21690 Make sure to also compute the range in the type of the switch index.
21692 2022-06-01 David Seifert <soap@gentoo.org>
21695 * configure: Regenerate.
21697 2022-06-01 H.J. Lu <hjl.tools@gmail.com>
21699 PR rtl-optimization/105638
21700 * df-core.cc (df_find_single_def_src): Moved and renamed from
21701 find_single_def_src in loop-iv.cc. Change the argument to rtx
21702 and use rtx_equal_p. Return null for partial or conditional
21704 * df.h (df_find_single_def_src): New prototype.
21705 * dse.cc (record_store): Use the constant source if the source
21706 register is set only once.
21707 * loop-iv.cc (find_single_def_src): Moved to df-core.cc.
21708 (replace_single_def_regs): Replace find_single_def_src with
21709 df_find_single_def_src.
21711 2022-06-01 Wilco Dijkstra <wilco.dijkstra@arm.com>
21713 * config/aarch64/aarch64.opt (explicit_tune_core): Rename to
21715 (explicit_arch): Rename to selected_arch.
21716 (x_aarch64_override_tune_string): Remove.
21717 (aarch64_ra_sign_key): Add as TargetVariable so it gets saved/restored.
21718 (aarch64_override_tune_string): Add Save so it gets saved/restored.
21719 * config/aarch64/aarch64.h (aarch64_architecture_version): Remove.
21720 * config/aarch64/aarch64.cc (aarch64_architecture_version): Remove.
21721 (processor): Remove archtecture_version field.
21722 (selected_arch): Remove global.
21723 (selected_cpu): Remove global.
21724 (selected_tune): Remove global.
21725 (aarch64_ra_sign_key): Move global to aarch64.opt so it is saved.
21726 (aarch64_override_options_internal): Use aarch64_get_tune_cpu.
21727 (aarch64_override_options): Further simplify code to only set
21728 selected_arch and selected_tune globals.
21729 (aarch64_option_save): Remove now that target options are saved.
21730 (aarch64_option_restore): Remove redundant target option restores.
21731 * config/aarch64/aarch64-c.cc (aarch64_update_cpp_builtins): Use
21733 * config/aarch64/aarch64-opts.h (aarch64_key_type): Add, moved from...
21734 * config/aarch64/aarch64-protos.h (aarch64_key_type): Remove.
21735 (aarch64_ra_sign_key): Remove.
21737 2022-06-01 Jakub Jelinek <jakub@redhat.com>
21739 PR middle-end/30314
21740 * match.pd (__builtin_mul_overflow_p (x, cst, (utype) 0) ->
21741 x > ~(utype)0 / cst): New simplification.
21743 2022-06-01 Richard Biener <rguenther@suse.de>
21745 PR tree-optimization/105786
21746 * tree-loop-distribution.cc
21747 (loop_distribution::transform_reduction_loop): Only do strlen
21748 replacement for integer type reductions.
21750 2022-06-01 Jakub Jelinek <jakub@redhat.com>
21752 PR tree-optimization/105770
21753 * tree-ssa-loop-unswitch.cc (find_unswitching_predicates_for_bb): Cast
21754 CASE_LOW and CASE_HIGH to TREE_TYPE (idx) before comparisons with idx.
21756 2022-06-01 Aldy Hernandez <aldyh@redhat.com>
21758 * gimple-range-cache.cc (ssa_block_ranges::dump): Convert to vrange.
21759 (sbr_vector::sbr_vector): Same.
21760 (sbr_vector::grow): Same.
21761 (sbr_vector::set_bb_range): Same.
21762 (sbr_vector::get_bb_range): Same.
21763 (sbr_sparse_bitmap::sbr_sparse_bitmap): Same.
21764 (sbr_sparse_bitmap::set_bb_range): Same.
21765 (sbr_sparse_bitmap::get_bb_range): Same.
21766 (block_range_cache::set_bb_range): Same.
21767 (block_range_cache::get_bb_range): Same.
21768 (block_range_cache::dump): Same.
21769 (ssa_global_cache::get_global_range): Same.
21770 (ssa_global_cache::set_global_range): Same.
21771 (ssa_global_cache::clear): Same.
21772 (ssa_global_cache::dump): Same.
21773 (ranger_cache::get_global_range): Same.
21774 (ranger_cache::set_global_range): Same.
21775 (ranger_cache::range_of_def): Same.
21776 (ranger_cache::entry_range): Same.
21777 (ranger_cache::exit_range): Same.
21778 (ranger_cache::edge_range): Same.
21779 (ranger_cache::range_of_expr): Same.
21780 (ranger_cache::range_on_edge): Same.
21781 (ranger_cache::block_range): Same.
21782 (ranger_cache::propagate_cache): Same.
21783 (ranger_cache::fill_block_cache): Same.
21784 (ranger_cache::range_from_dom): Same.
21785 * gimple-range-cache.h: Same.
21786 * gimple-range-edge.cc (gimple_outgoing_range::get_edge_range):
21788 (gimple_outgoing_range::switch_edge_range): Same.
21789 (gimple_outgoing_range::edge_range_p): Same.
21790 * gimple-range-edge.h: Same.
21791 * gimple-range-fold.cc (fur_source::get_operand): Same.
21792 (fur_source::get_phi_operand): Same.
21793 (fur_edge::get_operand): Same.
21794 (fur_edge::get_phi_operand): Same.
21795 (fur_stmt::get_operand): Same.
21796 (fur_stmt::get_phi_operand): Same.
21797 (fur_list::fur_list): Same.
21798 (fur_list::get_operand): Same.
21799 (fur_list::get_phi_operand): Same.
21800 (fold_range): Same.
21801 (adjust_imagpart_expr): Same.
21802 (adjust_realpart_expr): Same.
21803 (gimple_range_adjustment): Same.
21804 (fold_using_range::fold_stmt): Same.
21805 (fold_using_range::range_of_range_op): Same.
21806 (fold_using_range::range_of_address): Same.
21807 (fold_using_range::range_of_phi): Same.
21808 (fold_using_range::range_of_call): Same.
21809 (fold_using_range::range_of_builtin_call): Same.
21810 (fold_using_range::range_of_builtin_int_call): Same.
21811 (fold_using_range::range_of_cond_expr): Same.
21812 (fur_source::register_outgoing_edges): Same.
21813 * gimple-range-fold.h (fold_range): Same.
21814 (gimple_range_type): Same.
21815 (gimple_range_ssa_p): Same.
21816 * gimple-range-gori.cc (gimple_range_calc_op1): Same.
21817 (gimple_range_calc_op2): Same.
21818 (gori_compute::compute_operand_range_switch): Same.
21819 (gori_compute::compute_operand_range): Same.
21820 (gori_compute::logical_combine): Same.
21821 (gori_compute::compute_logical_operands): Same.
21822 (gori_compute::compute_operand1_range): Same.
21823 (gori_compute::compute_operand2_range): Same.
21824 (gori_compute::compute_operand1_and_operand2_range): Same.
21825 (gori_compute::outgoing_edge_range_p): Same.
21826 (gori_compute::condexpr_adjust): Same.
21827 * gimple-range-gori.h (gimple_range_calc_op1): Same.
21828 (gimple_range_calc_op2): Same.
21829 * gimple-range-path.cc (path_range_query::get_cache): Same.
21830 (path_range_query::set_cache): Same.
21831 (path_range_query::range_on_path_entry): Same.
21832 (path_range_query::internal_range_of_expr): Same.
21833 (path_range_query::range_of_expr): Same.
21834 (path_range_query::ssa_range_in_phi): Same.
21835 (path_range_query::range_defined_in_block): Same.
21836 (path_range_query::compute_ranges_in_phis): Same.
21837 (path_range_query::compute_ranges_in_block): Same.
21838 (path_range_query::add_to_imports): Same.
21839 (path_range_query::range_of_stmt): Same.
21840 * gimple-range-path.h: Same.
21841 * gimple-range-infer.cc (gimple_infer_range::add_range): Same.
21842 (gimple_infer_range::~side_effect_manager): Same.
21843 (gimple_infer_range::get_nonzero): Same.
21844 (gimple_infer_range::maybe_adjust_range): Same.
21845 (gimple_infer_range::add_range): Same.
21846 * gimple-range-infer.h: Same.
21847 * gimple-range-tests.cc: Same.
21848 * gimple-range-trace.cc (range_tracer::trailer): Same.
21849 (debug_seed_ranger): Same.
21850 * gimple-range-trace.h: Same.
21851 * gimple-range.cc (gimple_ranger::range_of_expr): Same.
21852 (gimple_ranger::range_on_entry): Same.
21853 (gimple_ranger::range_on_exit): Same.
21854 (gimple_ranger::range_on_edge): Same.
21855 (gimple_ranger::fold_range_internal): Same.
21856 (gimple_ranger::range_of_stmt): Same.
21857 (gimple_ranger::prefill_name): Same.
21858 (gimple_ranger::prefill_stmt_dependencies): Same.
21859 (gimple_ranger::export_global_ranges): Same.
21860 (gimple_ranger::dump_bb): Same.
21861 * gimple-range.h: Same.
21862 * gimple-ssa-warn-access.cc (check_nul_terminated_array): Same.
21863 (memmodel_to_uhwi): Same.
21864 * tree-ssa-loop-niter.cc (refine_value_range_using_guard): Same.
21865 (determine_value_range): Same.
21866 (record_nonwrapping_iv): Same.
21867 (infer_loop_bounds_from_signedness): Same.
21868 (scev_var_range_cant_overflow): Same.
21869 * tree-ssa-threadedge.cc (hybrid_jt_simplifier::simplify): Same.
21870 * value-query.cc (range_query::range_on_edge): Same.
21871 (range_query::range_of_stmt): Same.
21872 (range_query::value_of_expr): Same.
21873 (range_query::value_on_edge): Same.
21874 (range_query::value_of_stmt): Same.
21875 (range_query::get_tree_range): Same.
21876 (update_global_range): Same.
21877 (get_range_global): Same.
21878 (gimple_range_global): Same.
21879 (global_range_query::range_of_expr): Same.
21880 (range_query::query_relation): Same.
21881 * value-query.h (gimple_range_global): Same.
21882 (update_global_range): Same.
21883 * vr-values.cc (vr_values::range_of_expr): Same.
21884 (bounds_of_var_in_loop): Same.
21885 (simplify_using_ranges::vrp_visit_cond_stmt): Same.
21886 * vr-values.h (class vr_values): Same.
21887 * tree-ssa-loop-unswitch.cc (unswitch_predicate): Same.
21889 2022-06-01 Aldy Hernandez <aldyh@redhat.com>
21891 * gimple-range-cache.cc (sbr_vector::sbr_vector): Adjust for
21893 (sbr_vector::grow): Same.
21894 (sbr_vector::set_bb_range): Same.
21895 (sbr_sparse_bitmap::sbr_sparse_bitmap): Same.
21896 (sbr_sparse_bitmap::set_bb_range): Same.
21897 (block_range_cache::~block_range_cache): Same.
21898 (block_range_cache::set_bb_range): Same.
21899 (ssa_global_cache::ssa_global_cache): Same.
21900 (ssa_global_cache::~ssa_global_cache): Same.
21901 (ssa_global_cache::set_global_range): Same.
21902 * gimple-range-cache.h (block_range_cache): Same.
21903 (ssa_global_cache): Same.
21904 * gimple-range-edge.cc
21905 (gimple_outgoing_range::calc_switch_ranges): Same.
21906 * gimple-range-edge.h (gimple_outgoing_range): Same.
21907 * gimple-range-infer.cc (infer_range_manager::get_nonzero):
21909 (infer_range_manager::add_range): Same.
21910 * gimple-range-infer.h (class infer_range_manager): Same.
21911 * value-range.h (class irange_allocator): Rename to...
21912 (class vrange_allocator): ...this.
21913 (irange_allocator::irange_allocator): New.
21914 (vrange_allocator::vrange_allocator): New.
21915 (irange_allocator::~irange_allocator): New.
21916 (vrange_allocator::~vrange_allocator): New.
21917 (irange_allocator::get_memory): Rename to...
21918 (vrange_allocator::alloc): ...this.
21919 (vrange_allocator::alloc_vrange): Rename from...
21920 (irange_allocator::allocate): ...this.
21921 (vrange_allocator::alloc_irange): New.
21923 2022-06-01 Aldy Hernandez <aldyh@redhat.com>
21925 * gimple-range-edge.cc (gimple_outgoing_range_stmt_p): Adjust for
21926 vrange and convert range_op_handler function calls to use the
21927 identically named object.
21928 * gimple-range-fold.cc (gimple_range_operand1): Same.
21929 (gimple_range_operand2): Same.
21930 (fold_using_range::fold_stmt): Same.
21931 (fold_using_range::range_of_range_op): Same.
21932 (fold_using_range::range_of_builtin_ubsan_call): Same.
21933 (fold_using_range::relation_fold_and_or): Same.
21934 (fur_source::register_outgoing_edges): Same.
21935 * gimple-range-fold.h (gimple_range_handler): Remove.
21936 * gimple-range-gori.cc (gimple_range_calc_op1): Adjust for vrange.
21937 (gimple_range_calc_op2): Same.
21938 (range_def_chain::get_def_chain): Same.
21939 (gori_compute::compute_operand_range): Same.
21940 (gori_compute::condexpr_adjust): Same.
21941 * gimple-range.cc (gimple_ranger::prefill_name): Same.
21942 (gimple_ranger::prefill_stmt_dependencies): Same.
21943 * range-op.cc (get_bool_state): Same.
21944 (class operator_equal): Add using clause.
21945 (class operator_not_equal): Same.
21946 (class operator_lt): Same.
21947 (class operator_le): Same.
21948 (class operator_gt): Same.
21949 (class operator_ge): Same.
21950 (class operator_plus): Same.
21951 (class operator_minus): Same.
21952 (class operator_mult): Same.
21953 (class operator_exact_divide): Same.
21954 (class operator_lshift): Same.
21955 (class operator_rshift): Same.
21956 (class operator_cast): Same.
21957 (class operator_logical_and): Same.
21958 (class operator_bitwise_and): Same.
21959 (class operator_logical_or): Same.
21960 (class operator_bitwise_or): Same.
21961 (class operator_bitwise_xor): Same.
21962 (class operator_trunc_mod): Same.
21963 (class operator_logical_not): Same.
21964 (class operator_bitwise_not): Same.
21965 (class operator_cst): Same.
21966 (class operator_identity): Same.
21967 (class operator_unknown): Same.
21968 (class operator_abs): Same.
21969 (class operator_negate): Same.
21970 (class operator_addr_expr): Same.
21971 (class pointer_or_operator): Same.
21972 (operator_plus::op1_range): Adjust for vrange.
21973 (operator_minus::op1_range): Same.
21974 (operator_mult::op1_range): Same.
21975 (operator_cast::op1_range): Same.
21976 (operator_bitwise_not::fold_range): Same.
21977 (operator_negate::fold_range): Same.
21978 (range_op_handler): Rename to...
21979 (get_handler): ...this.
21980 (range_op_handler::range_op_handler): New.
21981 (range_op_handler::fold_range): New.
21982 (range_op_handler::op1_range): New.
21983 (range_op_handler::op2_range): New.
21984 (range_op_handler::lhs_op1_relation): New.
21985 (range_op_handler::lhs_op2_relation): New.
21986 (range_op_handler::op1_op2_relation): New.
21987 (range_cast): Adjust for vrange.
21988 * range-op.h (range_op_handler): Remove function.
21989 (range_cast): Adjust for vrange.
21990 (class range_op_handler): New.
21991 (get_bool_state): Adjust for vrange.
21992 (empty_range_varying): Same.
21993 (relop_early_resolve): Same.
21994 * tree-data-ref.cc (compute_distributive_range): Same.
21995 * tree-vrp.cc (get_range_op_handler): Remove.
21996 (range_fold_binary_symbolics_p): Use range_op_handler class
21997 instead of get_range_op_handler.
21998 (range_fold_unary_symbolics_p): Same.
21999 (range_fold_binary_expr): Same.
22000 (range_fold_unary_expr): Same.
22001 * value-query.cc (range_query::get_tree_range): Adjust for vrange.
22003 2022-06-01 Aldy Hernandez <aldyh@redhat.com>
22005 * gimple-range-fold.h (gimple_range_type): Check type before
22006 calling supports_type_p.
22007 * gimple-range-path.cc (path_range_query::range_of_stmt): Same.
22008 * value-query.cc (range_query::get_tree_range): Same.
22009 * value-range.cc (Value_Range::lower_bound): New.
22010 (Value_Range::upper_bound): New.
22011 (Value_Range::dump): New.
22012 * value-range.h (class Value_Range): New.
22013 (irange::supports_type_p): Do not check if type is non-zero.
22015 2022-06-01 Aldy Hernandez <aldyh@redhat.com>
22017 * value-range-equiv.cc (value_range_equiv::set): New.
22018 * value-range-equiv.h (class value_range_equiv): Make set method
22020 Remove default bitmap argument from set method.
22021 * value-range.cc (vrange::contains_p): New.
22022 (vrange::singleton_p): New.
22023 (vrange::operator=): New.
22024 (vrange::operator==): New.
22025 (irange::fits_p): Move to .cc file.
22026 (irange::set_nonnegative): New.
22027 (unsupported_range::unsupported_range): New.
22028 (unsupported_range::set): New.
22029 (unsupported_range::type): New.
22030 (unsupported_range::set_undefined): New.
22031 (unsupported_range::set_varying): New.
22032 (unsupported_range::dump): New.
22033 (unsupported_range::union_): New.
22034 (unsupported_range::intersect): New.
22035 (unsupported_range::zero_p): New.
22036 (unsupported_range::nonzero_p): New.
22037 (unsupported_range::set_nonzero): New.
22038 (unsupported_range::set_zero): New.
22039 (unsupported_range::set_nonnegative): New.
22040 (unsupported_range::fits_p): New.
22041 (irange::set): Call irange::set_undefined.
22042 (irange::verify_range): Check discriminator field.
22043 (irange::dump): Dump [irange] marker.
22044 (irange::debug): Move to...
22045 (vrange::debug): ...here.
22046 (dump_value_range): Accept vrange.
22048 * value-range.h (enum value_range_discriminator): New.
22049 (class vrange): New.
22050 (class unsupported_range): New.
22051 (struct vrange_traits): New.
22054 (class irange): Inherit from vrange.
22055 (dump_value_range): Adjust for vrange.
22056 (irange::kind): Rename to...
22057 (vrange::kind): ...this.
22058 (irange::varying_p): Rename to...
22059 (vrange::varying_p): ...this.
22060 (irange::undefined_p): Rename to...
22061 (vrange::undefined_p): ...this.
22062 (irange::irange): Set discriminator.
22063 (irange::union_): Convert to irange before passing to irange
22065 (irange::intersect): Same.
22066 (vrange::supports_type_p): New.
22067 * vr-values.cc (vr_values::extract_range_from_binary_expr): Pass
22068 NULL bitmap argument to value_range_equiv::set.
22069 (vr_values::extract_range_basic): Same.
22071 2022-06-01 Richard Biener <rguenther@suse.de>
22073 PR tree-optimization/105763
22074 * tree-ssa-loop-unswitch.cc (find_unswitching_predicates_for_bb):
22075 Check gimple_range_ssa_p.
22077 2022-05-31 Jason Merrill <jason@redhat.com>
22079 * Makefile.in (TAGS): Look at libcpp/*.cc.
22081 2022-05-31 Christophe Lyon <christophe.lyon@arm.com>
22083 * config/aarch64/aarch64.cc (aarch64_gimplify_va_arg_expr):
22084 Prefix mode names with E_.
22086 2022-05-31 Alan Modra <amodra@gmail.com>
22088 * dwarf2out.cc (gen_namelist_decl): Adjust to suit correct
22089 spelling of DW_AT_namelist_item.
22091 2022-05-31 Jakub Jelinek <jakub@redhat.com>
22093 * omp-low.cc (build_outer_var_ref): For code == OMP_CLAUSE_ALLOCATE
22094 allow var to be private in the outer context.
22095 (lower_private_allocate): Pass OMP_CLAUSE_ALLOCATE as last argument
22096 to build_outer_var_ref.
22098 2022-05-30 Roger Sayle <roger@nextmovesoftware.com>
22100 * config/i386/i386.cc (ix86_modes_tieable_p): Allow SCmode to be
22101 tieable with DImode on TARGET_64BIT, and SCmode tieable with
22102 V2SFmode, and DCmode with V2DFmode.
22104 2022-05-30 Roger Sayle <roger@nextmovesoftware.com>
22106 PR rtl-optimization/101617
22107 * config/i386/i386-expand.cc (ix86_expand_int_movcc): Add a
22108 special case (indicated by negate_cc_compare_p) to generate a
22109 -1/0 mask using neg;sbb.
22110 * config/i386/i386.md (x86_neg<mode>_ccc): New define_expand
22111 to generate an *x86_neg<mode>_ccc instruction.
22112 (x86_mov<mode>cc_0_m1_neg): Likewise, a new define_expand to
22113 generate a *x86_mov<mode>cc_0_m1_neg instruction.
22115 2022-05-30 Roger Sayle <roger@nextmovesoftware.com>
22117 * rtlanal.cc (rtx_cost) <MULT>: Treat FMA, SS_MULT, US_MULT,
22118 SMUL_HIGHPART and UMUL_HIGHPART as having the same cost as MULT.
22119 <DIV>: Likewise, SS_DIV and US_DIV have the same default as DIV.
22121 2022-05-30 Roger Sayle <roger@nextmovesoftware.com>
22124 * config/i386/i386-expand.cc (ix86_expand_branch): Don't decompose
22125 DI mode equality/inequality using XOR here. Instead generate a
22126 COMPARE for doubleword modes (DImode on !TARGET_64BIT or TImode).
22127 * config/i386/i386-features.cc (gen_gpr_to_xmm_move_src): Use
22128 gen_rtx_SUBREG when NUNITS is 1, i.e. for TImode to V1TImode.
22129 (general_scalar_chain::convert_compare): New function to convert
22130 scalar equality/inequality comparison into vector operations.
22131 (general_scalar_chain::convert_insn) [COMPARE]: Refactor. Call
22132 new convert_compare helper method.
22133 (convertible_comparion_p): Update to match doubleword COMPARE
22134 of two register, memory or integer constant operands.
22135 * config/i386/i386-features.h (general_scalar_chain::convert_compare):
22136 Prototype/declare member function here.
22137 * config/i386/i386.md (cstore<mode>4): Change mode to SDWIM, but
22138 only allow new doubleword modes for EQ and NE operators.
22139 (*cmp<dwi>_doubleword): New define_insn_and_split, to split a
22140 doubleword comparison into a pair of XORs followed by an IOR to
22141 set the (zero) flags register, optimizing the XORs if possible.
22142 * config/i386/sse.md (V_AVX): Include V1TI and V2TI in mode
22143 iterator; V_AVX is (currently) only used by ptest.
22144 (sse4_1 mode attribute): Update to support V1TI and V2TI.
22146 2022-05-30 Uroš Bizjak <ubizjak@gmail.com>
22148 * config/i386/i386.md: Remove constraints when used with
22149 const_int_operand, const0_operand, const_1_operand, constm1_operand,
22150 const8_operand, const128_operand, const248_operand, const123_operand,
22151 const2367_operand, const1248_operand, const359_operand,
22152 const_4_or_8_to_11_operand, const48_operand, const_0_to_1_operand,
22153 const_0_to_3_operand, const_0_to_4_operand, const_0_to_5_operand,
22154 const_0_to_7_operand, const_0_to_15_operand, const_0_to_31_operand,
22155 const_0_to_63_operand, const_0_to_127_operand, const_0_to_255_operand,
22156 const_0_to_255_mul_8_operand, const_1_to_31_operand,
22157 const_1_to_63_operand, const_2_to_3_operand, const_4_to_5_operand,
22158 const_4_to_7_operand, const_6_to_7_operand, const_8_to_9_operand,
22159 const_8_to_11_operand, const_8_to_15_operand, const_10_to_11_operand,
22160 const_12_to_13_operand, const_12_to_15_operand, const_14_to_15_operand,
22161 const_16_to_19_operand, const_16_to_31_operand, const_20_to_23_operand,
22162 const_24_to_27_operand and const_28_to_31_operand.
22163 * config/i386/mmx.md: Ditto.
22164 * config/i386/sse.md: Ditto.
22165 * config/i386/subst.md: Ditto.
22166 * config/i386/sync.md: Ditto.
22168 2022-05-30 Jan Beulich <jbeulich@suse.com>
22170 * config/i386/i386.md (bmi2_umul<mode><dwi>3_1): Correct MEM_P()
22173 2022-05-30 Prathamesh Kulkarni <prathamesh.kulkarni@linaro.org>
22175 * config/arm/arm.cc (arm_vectorize_vec_perm_const): Adjust prototype.
22177 2022-05-29 Iain Sandoe <iain@sandoe.co.uk>
22180 * config/darwin.h: Move versions-specific handling of multiply_defined
22181 from SUBTARGET_DRIVER_SELF_SPECS to LINK_SPEC.
22183 2022-05-29 Eric Gallager <egallager@gcc.gnu.org>
22186 * doc/sourcebuild.texi: Add entries for the c++tools,
22187 gotools, libbacktrace, libcc1, libcody, liboffloadmic,
22188 and libsanitizer directories. Remove entry for boehm-gc.
22189 Fix alphabetization for libquadmath.
22191 2022-05-28 Joel Holdsworth <jholdsworth@nvidia.com>
22193 * config/avr/avr-mcus.def: Add device definitions.
22194 * doc/avr-mmcu.texi: Corresponding changes.
22195 * config/avr/gen-avr-mmcu-texi.cc: Added support for avr
22197 * config/avr/gen-avr-mmcu-specs.cc: Prevent -mmcu=avr* flags
22198 from leaking into cc1.
22200 2022-05-28 Vladimir Makarov <vmakarov@gcc.gnu.org>
22203 * config/sh/sh.cc (sh_register_move_cost): Avoid cost "2" (which
22204 is special) for various scenarios.
22206 2022-05-28 Iain Sandoe <iain@sandoe.co.uk>
22208 * config/darwin.h (REAL_LIBGCC_SPEC): Update the comment block
22209 describing this macro.
22211 2022-05-27 Richard Biener <rguenther@suse.de>
22213 * tree-dfa.cc (get_ref_base_and_extent): Avoid shift.
22215 2022-05-27 Martin Jambor <mjambor@suse.cz>
22218 * ipa-prop.cc (propagate_controlled_uses): Check type of the
22219 constant before adding a LOAD reference.
22221 2022-05-27 Jakub Jelinek <jakub@redhat.com>
22223 * tree-core.h (enum omp_clause_code): Rename OMP_CLAUSE_TO_DECLARE
22224 to OMP_CLAUSE_ENTER.
22225 * tree.h (OMP_CLAUSE_ENTER_TO): Define.
22226 * tree.cc (omp_clause_num_ops, omp_clause_code_name): Rename
22227 OMP_CLAUSE_TO_DECLARE to OMP_CLAUSE_ENTER.
22228 * tree-pretty-print.cc (dump_omp_clause): Handle OMP_CLAUSE_ENTER
22229 instead of OMP_CLAUSE_TO_DECLARE, if OMP_CLAUSE_ENTER_TO, print
22230 "to" instead of "enter".
22231 * tree-nested.cc (convert_nonlocal_omp_clauses,
22232 convert_local_omp_clauses): Handle OMP_CLAUSE_ENTER instead of
22233 OMP_CLAUSE_TO_DECLARE.
22235 2022-05-27 Richard Biener <rguenther@suse.de>
22237 PR tree-optimization/105726
22238 * gimple-ssa-warn-restrict.cc (builtin_memref::set_base_and_offset):
22239 Constrain array-of-flexarray case more.
22241 2022-05-27 Jakub Jelinek <jakub@redhat.com>
22243 PR sanitizer/105729
22244 * fold-const.cc (fold_unary_loc): Don't optimize (X &) ((Y *) z + w)
22245 to (X &) z + w if -fsanitize=null during GENERIC folding.
22247 2022-05-27 Roger Sayle <roger@nextmovesoftware.com>
22249 * match.pd (match_zero_one_valued_p): New predicate.
22250 (mult @0 @1): Use zero_one_valued_p for optimization to the
22251 expression "bit_and @0 @1".
22252 (bit_and (negate zero_one_valued_p@0) @1): Optimize to MULT_EXPR.
22253 (plus @0 (mult (minus @1 @0) zero_one_valued_p@2)): New transform.
22254 (minus @0 (mult (minus @0 @1) zero_one_valued_p@2)): Likewise.
22255 (bit_xor @0 (mult (bit_xor @0 @1) zero_one_valued_p@2)): Likewise.
22256 Remove three redundant transforms obsoleted by the three above.
22258 2022-05-27 Roger Sayle <roger@nextmovesoftware.com>
22260 * config/i386/i386.md (*test<mode>_not): New define_insn_and_split
22261 to split a combined "and;cmp" sequence into "not;test".
22263 2022-05-27 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
22265 * config/xtensa/xtensa.md (bswapsi2): New expansion pattern.
22266 (bswapsi2_internal): Revise the template and condition, and add
22267 detection code for preceding the same insn in order to omit a
22268 "SSAI 8" instruction of the latter.
22269 (bswapdi2): Suppress built-in insn expansion with the corresponding
22270 library call when optimizing for size.
22272 2022-05-27 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
22274 * config/xtensa/xtensa-protos.h
22275 (xtensa_expand_block_set_unrolled_loop,
22276 xtensa_expand_block_set_small_loop): New prototypes.
22277 * config/xtensa/xtensa.cc (xtensa_sizeof_MOVI,
22278 xtensa_expand_block_set_unrolled_loop,
22279 xtensa_expand_block_set_small_loop): New functions.
22280 * config/xtensa/xtensa.md (setmemsi): New expansion pattern.
22281 * config/xtensa/xtensa.opt (mlongcalls): Add target mask.
22283 2022-05-26 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
22285 * config/xtensa/xtensa.cc (xtensa_expand_block_move):
22286 Make instruction counting more accurate, and simplify emitting insns.
22288 2022-05-26 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
22290 * config/xtensa/constraints.md (M, O): Use the macro.
22291 * config/xtensa/predicates.md (addsubx_operand, extui_fldsz_operand,
22292 sext_fldsz_operand): Ditto.
22293 * config/xtensa/xtensa.cc (xtensa_simm8, xtensa_simm8x256,
22294 xtensa_simm12b, xtensa_uimm8, xtensa_uimm8x2, xtensa_uimm8x4,
22295 xtensa_mask_immediate, smalloffset_mem_p, printx, xtensa_call_save_reg,
22296 xtensa_expand_prologue): Ditto.
22297 * config/xtensa/xtensa.h (FUNCTION_ARG_REGNO_P): Ditto.
22299 2022-05-26 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
22301 * config/xtensa/predicates.md (extui_fldsz_operand): Simplify.
22302 * config/xtensa/xtensa.cc (xtensa_mask_immediate, print_operand):
22305 2022-05-25 Andrew MacLeod <amacleod@redhat.com>
22307 * gimple-range-cache.cc: Adjust comments.
22308 * gimple-range-infer.cc: Adjust comments.
22309 * gimple-range-infer.h: Adjust comments.
22310 * gimple-range.cc: Adjust comments.
22312 2022-05-25 Andrew MacLeod <amacleod@redhat.com>
22314 * Makefile.in (OBJS): Use gimple-range-infer.o.
22315 * gimple-range-cache.cc (ranger_cache::fill_block_cache): Change msg.
22316 (ranger_cache::range_from_dom): Rename var side_effect to infer.
22317 (ranger_cache::apply_inferred_ranges): Rename from apply_side_effects.
22318 * gimple-range-cache.h: Include gimple-range-infer.h.
22319 (class ranger_cache): Adjust prototypes, use infer_range_manager.
22320 * gimple-range-infer.cc: Rename from gimple-range-side-effects.cc.
22321 (gimple_infer_range::*): Rename from stmt_side_effects.
22322 (infer_range_manager::*): Rename from side_effect_manager.
22323 * gimple-range-side-effect.cc: Rename.
22324 * gimple-range-side-effect.h: Rename.
22325 * gimple-range-infer.h: Rename from gimple-range-side-effects.h.
22326 (class gimple_infer_range): Rename from stmt_side_effects.
22327 (class infer_range_manager): Rename from side_effect_manager.
22328 * gimple-range.cc (gimple_ranger::register_inferred_ranges): Rename
22329 from register_side_effects.
22330 * gimple-range.h (register_inferred_ranges): Adjust prototype.
22331 * range-op.h: Adjust comment.
22332 * tree-vrp.cc (rvrp_folder::pre_fold_bb): Use register_inferred_ranges.
22333 (rvrp_folder::post_fold_bb): Use register_inferred_ranges.
22335 2022-05-25 Simon Cook <simon.cook@embecosm.com>
22337 * config/riscv/arch-canonicalize: Only add mafd extension if
22338 base was rv32/rv64g.
22340 2022-05-25 Tobias Burnus <tobias@codesourcery.com>
22342 * doc/invoke.texi (AMD GCN Options): Add gfx908/gfx90a.
22344 2022-05-25 Jakub Jelinek <jakub@redhat.com>
22346 PR sanitizer/105714
22347 * asan.cc (has_stmt_been_instrumented_p): For assignments which
22348 are both stores and loads, return true only if both destination
22349 and source have been instrumented.
22351 2022-05-25 Martin Liska <mliska@suse.cz>
22352 Richard Biener <rguenther@suse.de>
22354 * dbgcnt.def (DEBUG_COUNTER): Add loop_unswitch counter.
22355 * params.opt (max-unswitch-level): Remove.
22356 * doc/invoke.texi (max-unswitch-level): Likewise.
22357 * tree-cfg.cc (gimple_lv_add_condition_to_bb): Support not
22358 gimplified expressions.
22359 * tree-ssa-loop-unswitch.cc (struct unswitch_predicate): New.
22360 (tree_may_unswitch_on): Rename to ...
22361 (find_unswitching_predicates_for_bb): ... this and handle
22363 (get_predicates_for_bb): Likewise.
22364 (set_predicates_for_bb): Likewise.
22365 (init_loop_unswitch_info): Likewise.
22366 (tree_ssa_unswitch_loops): Prepare stuff before calling
22367 tree_unswitch_single_loop.
22368 (tree_unswitch_single_loop): Rework the function using
22369 pre-computed predicates and with a per original loop cost model.
22371 (add_predicate_to_path): Likewise.
22372 (find_range_for_lhs): Likewise.
22373 (simplify_using_entry_checks): Rename to ...
22374 (evaluate_control_stmt_using_entry_checks): ... this, handle
22375 switch statements and improve simplifications using ranger.
22376 (simplify_loop_version): Rework using
22377 evaluate_control_stmt_using_entry_checks.
22378 (evaluate_bbs): New.
22379 (evaluate_loop_insns_for_predicate): Likewise.
22380 (tree_unswitch_loop): Adjust to allow switch statements and
22381 pass in the edge to unswitch.
22382 (clean_up_after_unswitching): New.
22383 (pass_tree_unswitch::execute): Pass down fun.
22385 2022-05-24 Eugene Rozenfeld <erozen@microsoft.com>
22387 * tree-vect-loop-manip.cc (vect_do_peeling): Save/restore profile
22388 counts for the epilog loop.
22390 2022-05-24 Martin Sebor <msebor@redhat.com>
22391 Richard Biener <rguenther@suse.de>
22393 PR middle-end/105604
22394 * gimple-ssa-sprintf.cc (set_aggregate_size_and_offset): Add comments.
22395 (get_origin_and_offset_r): Remove null handling. Handle variable array
22397 (get_origin_and_offset): Handle null argument here. Simplify.
22398 (alias_offset): Update comment.
22399 * pointer-query.cc (field_at_offset): Update comment. Handle members
22400 of variable-length types.
22402 2022-05-24 Prathamesh Kulkarni <prathamesh.kulkarni@linaro.org>
22404 * target.def (vec_perm_const): Define new parameter op_mode and
22406 * doc/tm.texi: Regenerate.
22407 * config/aarch64/aarch64.cc (aarch64_vectorize_vec_perm_const): Adjust
22408 vec_perm_const hook to add new parameter op_mode and return false
22409 if result and operand modes do not match.
22410 * config/arm/arm.cc (arm_vectorize_vec_perm_const): Likewise.
22411 * config/gcn/gcn.cc (gcn_vectorize_vec_perm_const): Likewise.
22412 * config/ia64/ia64.cc (ia64_vectorize_vec_perm_const): Likewise.
22413 * config/mips/mips.cc (mips_vectorize_vec_perm_const): Likewise.
22414 * config/rs6000/rs6000.cc (rs6000_vectorize_vec_perm_const): Likewise
22415 * config/s390/s390.cc (s390_vectorize_vec_perm_const): Likewise.
22416 * config/sparc/sparc.cc (sparc_vectorize_vec_perm_const): Likewise.
22417 * config/i386/i386-expand.cc (ix86_vectorize_vec_perm_const): Likewise.
22418 * config/i386/i386-expand.h (ix86_vectorize_vec_perm_const): Adjust
22420 * config/i386/sse.md (ashrv4di3): Adjust call to vec_perm_const hook.
22421 (ashrv2di3): Likewise.
22422 * optabs.cc (expand_vec_perm_const): Likewise.
22423 * optabs-query.h (can_vec_perm_const_p): Adjust prototype.
22424 * optabs-query.cc (can_vec_perm_const_p): Define new parameter
22425 op_mode and pass it to vec_perm_const hook.
22426 (can_mult_highpart_p): Adjust call to can_vec_perm_const_p.
22427 * match.pd (vec_perm X Y CST): Likewise.
22428 * tree-ssa-forwprop.cc (simplify_vector_constructor): Likewise.
22429 * tree-vect-data-refs.cc (vect_grouped_store_supported): Likewise.
22430 (vect_grouped_load_supported): Likewise.
22431 (vect_shift_permute_load_chain): Likewise.
22432 * tree-vect-generic.cc (lower_vec_perm): Likewise.
22433 * tree-vect-loop-manip.cc (interleave_supported_p): Likewise.
22434 * tree-vect-loop.cc (have_whole_vector_shift): Likewise.
22435 * tree-vect-patterns.cc (vect_recog_rotate_pattern): Likewise.
22436 * tree-vect-slp.cc (can_duplicate_and_interleave_p): Likewise.
22437 (vect_transform_slp_perm_load): Likewise.
22438 (vectorizable_slp_permutation): Likewise.
22439 * tree-vect-stmts.cc (perm_mask_for_reverse): Likewise.
22440 (vectorizable_bswap): Likewise.
22441 (scan_store_can_perm_p): Likewise.
22442 (vect_gen_perm_mask_checked): Likewise.
22444 2022-05-24 H.J. Lu <hjl.tools@gmail.com>
22447 * config/i386/i386.opt: Remove Undocumented.
22448 * doc/invoke.texi: Document -mcet-switch.
22450 2022-05-24 Andrew Stubbs <ams@codesourcery.com>
22452 * config.gcc (amdgcn): Accept --with-arch=gfx908 and gfx90a.
22453 * config/gcn/gcn-opts.h (enum gcn_isa): New.
22454 (TARGET_GCN3): Use enum gcn_isa.
22455 (TARGET_GCN3_PLUS): Likewise.
22456 (TARGET_GCN5): Likewise.
22457 (TARGET_GCN5_PLUS): Likewise.
22458 (TARGET_CDNA1): New.
22459 (TARGET_CDNA1_PLUS): New.
22460 (TARGET_CDNA2): New.
22461 (TARGET_CDNA2_PLUS): New.
22462 (TARGET_M0_LDS_LIMIT): New.
22463 (TARGET_PACKED_WORK_ITEMS): New.
22464 * config/gcn/gcn.cc (gcn_isa): Change to enum gcn_isa.
22465 (gcn_option_override): Recognise CDNA ISA variants.
22466 (gcn_omp_device_kind_arch_isa): Support gfx90a.
22467 (gcn_expand_prologue): Make m0 init optional.
22468 Add support for packed work items.
22469 (output_file_start): Support gfx90a.
22470 (gcn_hsa_declare_function_name): Support gfx90a metadata.
22471 * config/gcn/gcn.h (TARGET_CPU_CPP_BUILTINS):Add __CDNA1__ and
22473 * config/gcn/gcn.md (<su>mulsi3_highpart): Use TARGET_GCN5_PLUS.
22474 (<su>mulsi3_highpart_imm): Likewise.
22475 (<su>mulsidi3): Likewise.
22476 (<su>mulsidi3_imm): Likewise.
22477 * config/gcn/gcn.opt (gpu_type): Add gfx90a.
22478 * config/gcn/mkoffload.cc (EF_AMDGPU_MACH_AMDGCN_GFX90a): New.
22479 (main): Support gfx90a.
22480 * config/gcn/t-gcn-hsa: Add gfx90a multilib.
22481 * config/gcn/t-omp-device: Add gfx90a isa.
22483 2022-05-24 Andrew Stubbs <ams@codesourcery.com>
22485 * config.in: Regenerate.
22486 * config/gcn/gcn-hsa.h (X_FIJI): Delete.
22494 (NO_XNACK): New macro.
22495 (NO_SRAM_ECC): New macro.
22496 (SRAMOPT): Keep only v4 variant.
22497 (HSACO3_SELECT_OPT): Delete.
22498 (DRIVER_SELF_SPECS): Delete.
22499 (ASM_SPEC): Remove LLVM 9 support.
22500 * config/gcn/gcn-valu.md
22501 (gather<mode>_insn_2offsets<exec>): Remove assembler bug workaround.
22502 (scatter<mode>_insn_2offsets<exec_scatter>): Likewise.
22503 * config/gcn/gcn.cc (output_file_start): Remove LLVM 9 support.
22504 (print_operand_address): Remove assembler bug workaround.
22505 * config/gcn/mkoffload.cc (EF_AMDGPU_XNACK_V3): Delete.
22506 (EF_AMDGPU_SRAM_ECC_V3): Delete.
22507 (SET_XNACK_ON): Delete v3 variants.
22508 (SET_XNACK_OFF): Delete v3 variants.
22509 (TEST_XNACK): Delete v3 variants.
22510 (SET_SRAM_ECC_ON): Delete v3 variants.
22511 (SET_SRAM_ECC_ANY): Delete v3 variants.
22512 (SET_SRAM_ECC_OFF): Delete v3 variants.
22513 (SET_SRAM_ECC_UNSUPPORTED): Delete v3 variants.
22514 (TEST_SRAM_ECC_ANY): Delete v3 variants.
22515 (TEST_SRAM_ECC_ON): Delete v3 variants.
22516 (copy_early_debug_info): Remove v3 support.
22517 (main): Remove v3 support.
22518 * configure: Regenerate.
22519 * configure.ac: Replace all GCN feature checks with a version check.
22521 2022-05-24 Roger Sayle <roger@nextmovesoftware.com>
22523 * config/i386/i386.md (peephole2): Convert xor;neg;adc;neg,
22524 i.e. a double word negation of a zero extended operand, to
22527 2022-05-24 Roger Sayle <roger@nextmovesoftware.com>
22529 PR tree-optimization/105668
22530 * config/i386/i386-expand.cc (ix86_expand_sse_movcc): Support
22531 V1TImode, just like V2DImode.
22532 * config/i386/sse.md (vcond_mask_<mode><sseintvecmodelower>):
22533 Use VI_128 mode iterator instead of VI124_128 to include V2DI.
22534 (vcond_mask_v2div2di): Delete.
22535 (vcond_mask_v1tiv1ti): New define_expand.
22537 2022-05-24 Roger Sayle <roger@nextmovesoftware.com>
22539 * genpreds.cc (write_lookup_constraint_1): Avoid generating a call
22540 to strncmp for strings of length one.
22542 2022-05-24 ShiYulong <shiyulong@iscas.ac.cn>
22544 * config/riscv/predicates.md (imm5_operand): Add a new operand type for
22545 prefetch instructions.
22546 * config/riscv/riscv-builtins.cc (AVAIL): Add new AVAILs for CMO ISA
22548 (RISCV_ATYPE_SI): New.
22549 (RISCV_ATYPE_DI): New.
22550 * config/riscv/riscv-ftypes.def (0): New.
22552 * config/riscv/riscv.md (riscv_clean_<mode>): New.
22553 (riscv_flush_<mode>): New.
22554 (riscv_inval_<mode>): New.
22555 (riscv_zero_<mode>): New.
22557 (riscv_prefetchi_<mode>): New.
22558 * config/riscv/riscv-cmo.def: New file.
22560 2022-05-24 ShiYulong <shiyulong@iscas.ac.cn>
22562 * common/config/riscv/riscv-common.cc: Add zicbom, zicboz, zicbop extensions.
22563 * config/riscv/riscv-opts.h (MASK_ZICBOZ): New.
22564 (MASK_ZICBOM): New.
22565 (MASK_ZICBOP): New.
22566 (TARGET_ZICBOZ): New.
22567 (TARGET_ZICBOM): New.
22568 (TARGET_ZICBOP): New.
22569 * config/riscv/riscv.opt (riscv_zicmo_subext): New.
22571 2022-05-24 David Malcolm <dmalcolm@redhat.com>
22573 * tree-vect-slp-patterns.cc: Add "final" and "override" to
22574 vect_pattern::build impls as appropriate.
22576 2022-05-24 David Malcolm <dmalcolm@redhat.com>
22578 * ipa-cp.cc: Add "final" and "override" to call_summary_base vfunc
22579 implementations, removing redundant "virtual" as appropriate.
22580 * ipa-fnsummary.h: Likewise.
22581 * ipa-modref.cc: Likewise.
22582 * ipa-param-manipulation.cc: Likewise.
22583 * ipa-profile.cc: Likewise.
22584 * ipa-prop.h: Likewise.
22585 * ipa-pure-const.cc: Likewise.
22586 * ipa-reference.cc: Likewise.
22587 * ipa-sra.cc: Likewise.
22588 * symbol-summary.h: Likewise.
22589 * symtab-thunks.cc: Likewise.
22591 2022-05-24 Martin Liska <mliska@suse.cz>
22594 2022-05-24 Martin Liska <mliska@suse.cz>
22596 * expmed.cc (emit_store_flag_1): Mitigate -Wmaybe-uninitialized
22599 2022-05-24 Martin Liska <mliska@suse.cz>
22601 * expmed.cc (emit_store_flag_1): Mitigate -Wmaybe-uninitialized
22604 2022-05-24 Bruno Haible <bruno@clisp.org>
22607 * doc/install.texi (Configuration): Add more details about --with-zstd.
22608 Document --with-zstd-include and --with-zstd-lib
22610 2022-05-24 Richard Biener <rguenther@suse.de>
22612 PR middle-end/105711
22613 * expmed.cc (extract_bit_field_as_subreg): Add op0_mode parameter
22615 (extract_bit_field_1): Pass down the mode of op0 to
22616 extract_bit_field_as_subreg.
22618 2022-05-24 Vineet Gupta <vineetg@rivosinc.com>
22620 * config/riscv/riscv.cc: (struct riscv_tune_param): Add
22622 (rocket_tune_info): Add default fmv_cost 8.
22623 (sifive_7_tune_info): Ditto.
22624 (thead_c906_tune_info): Ditto.
22625 (optimize_size_tune_info): Ditto.
22626 (riscv_register_move_cost): Use fmv_cost for int<->fp moves.
22628 2022-05-24 Jakub Jelinek <jakub@redhat.com>
22631 * omp-builtins.def (BUILT_IN_GOMP_TASKWAIT_DEPEND_NOWAIT): New
22633 * gimplify.cc (gimplify_omp_task): Diagnose taskwait with nowait
22634 clause but no depend clauses.
22635 * omp-expand.cc (expand_taskwait_call): Use
22636 BUILT_IN_GOMP_TASKWAIT_DEPEND_NOWAIT rather than
22637 BUILT_IN_GOMP_TASKWAIT_DEPEND if nowait clause is present.
22639 2022-05-24 Richard Biener <rguenther@suse.de>
22641 PR tree-optimization/100221
22642 * tree-ssa-dse.cc (contains_phi_arg): New function.
22643 (dse_classify_store): Postpone PHI defs that feed another PHI in defs.
22645 2022-05-24 Richard Biener <rguenther@suse.de>
22647 PR tree-optimization/105629
22648 * tree-ssa-phiopt.cc (spaceship_replacement): Allow
22649 a sign-extending conversion.
22651 2022-05-24 Kewen Lin <linkw@linux.ibm.com>
22654 * config/rs6000/rs6000-p8swap.cc (union_defs): Assert def_insn can't
22656 (union_uses): Skip debug use_insn.
22658 2022-05-23 Vineet Gupta <vineetg@rivosinc.com>
22660 * config/riscv/predicates.md (const_0_operand): Remove
22662 * config/riscv/riscv.cc (riscv_rtx_costs): Add check for
22664 * config/riscv/riscv.h (TARGET_SUPPORTS_WIDE_INT): New define.
22666 2022-05-23 Mayshao <mayshao-oc@zhaoxin.com>
22668 * common/config/i386/cpuinfo.h (get_zhaoxin_cpu): Detect
22669 the specific type of Zhaoxin CPU, and return Zhaoxin CPU name.
22670 (cpu_indicator_init): Handle Zhaoxin processors.
22671 * common/config/i386/i386-common.cc: Add lujiazui.
22672 * common/config/i386/i386-cpuinfo.h (enum processor_vendor): Add
22674 (enum processor_types): Add ZHAOXIN_FAM7H.
22675 (enum processor_subtypes): Add ZHAOXIN_FAM7H_LUJIAZUI.
22676 * config.gcc: Add lujiazui.
22677 * config/i386/cpuid.h (signature_SHANGHAI_ebx): Add
22678 Signatures for zhaoxin
22679 (signature_SHANGHAI_ecx): Ditto.
22680 (signature_SHANGHAI_edx): Ditto.
22681 * config/i386/driver-i386.cc (host_detect_local_cpu): Let
22682 -march=native recognize lujiazui processors.
22683 * config/i386/i386-c.cc (ix86_target_macros_internal): Add lujiazui.
22684 * config/i386/i386-options.cc (m_LUJIAZUI): New_definition.
22685 * config/i386/i386.h (enum processor_type): Ditto.
22686 * config/i386/i386.md: Add lujiazui.
22687 * config/i386/x86-tune-costs.h (struct processor_costs): Add
22689 * config/i386/x86-tune-sched.cc (ix86_issue_rate): Add lujiazui.
22690 (ix86_adjust_cost): Ditto.
22691 * config/i386/x86-tune.def (X86_TUNE_SCHEDULE): Add lujiazui Tunnings.
22692 (X86_TUNE_PARTIAL_REG_DEPENDENCY): Ditto.
22693 (X86_TUNE_SSE_PARTIAL_REG_DEPENDENCY): Ditto.
22694 (X86_TUNE_SSE_PARTIAL_REG_FP_CONVERTS_DEPENDENCY): Ditto.
22695 (X86_TUNE_SSE_PARTIAL_REG_CONVERTS_DEPENDENCY): Ditto.
22696 (X86_TUNE_MOVX): Ditto.
22697 (X86_TUNE_MEMORY_MISMATCH_STALL): Ditto.
22698 (X86_TUNE_FUSE_CMP_AND_BRANCH_32): Ditto.
22699 (X86_TUNE_FUSE_CMP_AND_BRANCH_64): Ditto.
22700 (X86_TUNE_FUSE_CMP_AND_BRANCH_SOFLAGS): Ditto.
22701 (X86_TUNE_FUSE_ALU_AND_BRANCH): Ditto.
22702 (X86_TUNE_ACCUMULATE_OUTGOING_ARGS): Ditto.
22703 (X86_TUNE_USE_LEAVE): Ditto.
22704 (X86_TUNE_PUSH_MEMORY): Ditto.
22705 (X86_TUNE_LCP_STALL): Ditto.
22706 (X86_TUNE_USE_INCDEC): Ditto.
22707 (X86_TUNE_INTEGER_DFMODE_MOVES): Ditto.
22708 (X86_TUNE_OPT_AGU): Ditto.
22709 (X86_TUNE_PREFER_KNOWN_REP_MOVSB_STOSB): Ditto.
22710 (X86_TUNE_MISALIGNED_MOVE_STRING_PRO_EPILOGUES): Ditto.
22711 (X86_TUNE_USE_SAHF): Ditto.
22712 (X86_TUNE_USE_BT): Ditto.
22713 (X86_TUNE_AVOID_FALSE_DEP_FOR_BMI): Ditto.
22714 (X86_TUNE_ONE_IF_CONV_INSN): Ditto.
22715 (X86_TUNE_AVOID_MFENCE): Ditto.
22716 (X86_TUNE_EXPAND_ABS): Ditto.
22717 (X86_TUNE_USE_SIMODE_FIOP): Ditto.
22718 (X86_TUNE_USE_FFREEP): Ditto.
22719 (X86_TUNE_EXT_80387_CONSTANTS): Ditto.
22720 (X86_TUNE_SSE_UNALIGNED_LOAD_OPTIMAL): Ditto.
22721 (X86_TUNE_SSE_UNALIGNED_STORE_OPTIMAL): Ditto.
22722 (X86_TUNE_SSE_TYPELESS_STORES): Ditto.
22723 (X86_TUNE_SSE_LOAD0_BY_PXOR): Ditto.
22724 * doc/extend.texi: Add details about lujiazui.
22725 * doc/invoke.texi: Add details about lujiazui.
22726 * config/i386/lujiazui.md: Introduce lujiazui cpu and include new md file.
22728 2022-05-23 Martin Liska <mliska@suse.cz>
22730 * config/tilepro/gen-mul-tables.cc (ARRAY_SIZE): Add new macro.
22732 2022-05-23 Richard Biener <rguenther@suse.de>
22734 * tree-ssa-forwprop.cc (forward_propagate_into_cond): Remove.
22735 (pass_forwprop::execute): Do not propagate into COND_EXPR conditions.
22737 2022-05-23 Richard Biener <rguenther@suse.de>
22739 * gimple-expr.cc (is_gimple_condexpr): Remove.
22740 * gimple-expr.h (is_gimple_condexpr): Likewise.
22741 * gimplify.cc (gimplify_expr): Remove is_gimple_condexpr usage.
22742 * tree-if-conv.cc (set_bb_predicate): Likewie.
22743 (add_to_predicate_list): Likewise.
22744 (gen_phi_arg_condition): Likewise.
22745 (predicate_scalar_phi): Likewise.
22746 (predicate_statements): Likewise.
22748 2022-05-23 Richard Biener <rguenther@suse.de>
22750 * gimple-expr.cc (is_gimple_condexpr): Equate to is_gimple_val.
22751 * gimplify.cc (gimplify_pure_cond_expr): Gimplify the condition
22753 * gimple-fold.cc (valid_gimple_rhs_p): Simplify.
22754 * tree-cfg.cc (verify_gimple_assign_ternary): Likewise.
22755 * gimple-loop-interchange.cc (loop_cand::undo_simple_reduction):
22756 Build the condition of the COND_EXPR separately.
22757 * tree-ssa-loop-im.cc (move_computations_worker): Likewise.
22758 * tree-vect-generic.cc (expand_vector_condition): Likewise.
22759 * tree-vect-loop.cc (vect_create_epilog_for_reduction):
22761 * vr-values.cc (simplify_using_ranges::simplify): Likewise.
22762 * tree-vect-patterns.cc: Add comment indicating we are
22763 building invalid COND_EXPRs and why.
22764 * omp-expand.cc (expand_omp_simd): Gimplify the condition
22765 to the COND_EXPR separately.
22766 (expand_omp_atomic_cas): Note part that should be unreachable
22768 * tree-ssa-forwprop.cc (forward_propagate_into_cond): Adjust
22769 condition for valid replacements.
22770 * tree-if-conv.cc (predicate_bbs): Simulate previous
22771 re-folding of the condition in folded COND_EXPRs which
22772 is necessary because of unfolded GIMPLE_CONDs in the IL
22773 as in for example gcc.dg/fold-bopcond-1.c.
22774 * gimple-range-gori.cc (gori_compute::condexpr_adjust):
22775 Handle that the comparison is now in the def stmt of
22776 the select operand. Required by gcc.dg/pr104526.c.
22778 2022-05-23 Tobias Burnus <tobias@codesourcery.com>
22781 * langhooks-def.h (lhd_omp_array_size): New.
22782 (LANG_HOOKS_OMP_ARRAY_SIZE): Define.
22783 (LANG_HOOKS_DECLS): Add it.
22784 * langhooks.cc (lhd_omp_array_size): New.
22785 * langhooks.h (struct lang_hooks_for_decls): Add hook.
22786 * omp-low.cc (scan_sharing_clauses, lower_omp_target):
22787 Handle GOMP_MAP_FIRSTPRIVATE for array descriptors.
22789 2022-05-23 Roger Sayle <roger@nextmovesoftware.com>
22791 * config/i386/i386.cc (ix86_rtx_costs) <case AND>: Split from
22792 XOR/IOR case. Account for two instructions for double-word
22793 operations. In case of vector pandn, account for single
22794 instruction. Likewise for integer andn with TARGET_BMI.
22795 <case NOT>: Vector NOT requires more than 1 instruction (pxor).
22796 <case NEG>: Double-word negation requires 3 instructions.
22798 2022-05-23 Tsukasa OI <research_trasio@irq.a4lg.com>
22800 * common/config/riscv/riscv-common.cc (riscv_supported_std_ext):
22801 Fix "K" extension prefix to be placed before "J".
22802 * config/riscv/arch-canonicalize: Likewise.
22804 2022-05-23 liuhongt <hongtao.liu@intel.com>
22806 * config/i386/x86-tune-costs.h (skylake_cost): Increase gpr
22807 <-> mask cost from 5 to 6.
22808 (icelake_cost): Ditto.
22810 2022-05-20 Wilco Dijkstra <wilco.dijkstra@arm.com>
22812 * config/aarch64/aarch64.md
22813 (and_<SHIFT:optab><mode>3_compare0): Support rotate left.
22814 (and_<SHIFT:optab>si3_compare0_uxtw): Likewise.
22815 (<LOGICAL:optab>_<SHIFT:optab><mode>3): Likewise.
22816 (<LOGICAL:optab>_<SHIFT:optab>si3_uxtw): Likewise.
22817 (one_cmpl_<optab><mode>2): Likewise.
22818 (<LOGICAL:optab>_one_cmpl_<SHIFT:optab><mode>3): Likewise.
22819 (<LOGICAL:optab>_one_cmpl_<SHIFT:optab>sidi_uxtw): New pattern.
22820 (eor_one_cmpl_<SHIFT:optab><mode>3_alt): Support rotate left.
22821 (eor_one_cmpl_<SHIFT:optab>sidi3_alt_ze): Likewise.
22822 (and_one_cmpl_<SHIFT:optab><mode>3_compare0): Likewise.
22823 (and_one_cmpl_<SHIFT:optab>si3_compare0_uxtw): Likewise.
22824 (and_one_cmpl_<SHIFT:optab><mode>3_compare0_no_reuse): Likewise.
22825 (and_<SHIFT:optab><mode>3nr_compare0): Likewise.
22826 (*<optab>si3_insn_uxtw): Use SHIFT_no_rotate.
22827 (rolsi3_insn_uxtw): New pattern.
22828 * config/aarch64/iterators.md (SHIFT): Add rotate left.
22829 (SHIFT_no_rotate): Add new iterator.
22830 (SHIFT:shift): Print rotate left as ror.
22831 (is_rotl): Add test for left rotate.
22833 2022-05-20 Wilco Dijkstra <wilco.dijkstra@arm.com>
22835 * config.gcc (aarch64*-*-*): Simplify --with-cpu and --with-arch
22836 processing. Add support for architectural extensions.
22837 * config/aarch64/aarch64.h (TARGET_CPU_DEFAULT): Remove
22838 AARCH64_CPU_DEFAULT_FLAGS.
22839 (TARGET_CPU_NBITS): Remove.
22840 (TARGET_CPU_MASK): Remove.
22841 * config/aarch64/aarch64.cc (AARCH64_CPU_DEFAULT_FLAGS): Remove define.
22842 (get_tune_cpu): Assert CPU is always valid.
22843 (get_arch): Assert architecture is always valid.
22844 (aarch64_override_options): Cleanup CPU selection code and simplify logic.
22845 (aarch64_option_restore): Remove unnecessary checks on tune.
22847 2022-05-20 David Malcolm <dmalcolm@redhat.com>
22849 * config/aarch64/aarch64-sve-builtins-base.cc: Replace uses of
22850 "FINAL" and "OVERRIDE" with "final" and "override".
22851 * config/aarch64/aarch64-sve-builtins-functions.h: Likewise.
22852 * config/aarch64/aarch64-sve-builtins-shapes.cc: Likewise.
22853 * config/aarch64/aarch64-sve-builtins-sve2.cc: Likewise.
22854 * diagnostic-path.h: Likewise.
22855 * digraph.cc: Likewise.
22856 * gcc-rich-location.h: Likewise.
22857 * gimple-array-bounds.cc: Likewise.
22858 * gimple-loop-versioning.cc: Likewise.
22859 * gimple-range-cache.cc: Likewise.
22860 * gimple-range-cache.h: Likewise.
22861 * gimple-range-fold.cc: Likewise.
22862 * gimple-range-fold.h: Likewise.
22863 * gimple-range-tests.cc: Likewise.
22864 * gimple-range.h: Likewise.
22865 * gimple-ssa-evrp.cc: Likewise.
22866 * input.cc: Likewise.
22867 * json.h: Likewise.
22868 * read-rtl-function.cc: Likewise.
22869 * tree-complex.cc: Likewise.
22870 * tree-diagnostic-path.cc: Likewise.
22871 * tree-ssa-ccp.cc: Likewise.
22872 * tree-ssa-copy.cc: Likewise.
22873 * tree-vrp.cc: Likewise.
22874 * value-query.h: Likewise.
22875 * vr-values.h: Likewise.
22877 2022-05-20 Marcel Vollweiler <marcel@codesourcery.com>
22879 * omp-low.cc (omp_runtime_api_call): Added target_memcpy_async and
22880 target_memcpy_rect_async to omp_runtime_apis array.
22882 2022-05-20 Christophe Lyon <christophe.lyon@arm.com>
22884 * doc/sourcebuild.texi (Decimal floating point attributes): Document
22885 dfp_bid effective-target.
22887 2022-05-20 Christophe Lyon <christophe.lyon@arm.com>
22889 * config/aarch64/aarch64.cc
22890 (aarch64_split_128bit_move): Handle DFP modes.
22891 (aarch64_mode_valid_for_sched_fusion_p): Likewise.
22892 (aarch64_classify_address): Likewise.
22893 (aarch64_legitimize_address_displacement): Likewise.
22894 (aarch64_reinterpret_float_as_int): Likewise.
22895 (aarch64_float_const_zero_rtx_p): Likewise.
22896 (aarch64_can_const_movi_rtx_p): Likewise.
22897 (aarch64_anchor_offset): Likewise.
22898 (aarch64_secondary_reload): Likewise.
22899 (aarch64_rtx_costs): Likewise.
22900 (aarch64_legitimate_constant_p): Likewise.
22901 (aarch64_gimplify_va_arg_expr): Likewise.
22902 (aapcs_vfp_sub_candidate): Likewise.
22903 (aarch64_vfp_is_call_or_return_candidate): Likewise.
22904 (aarch64_output_scalar_simd_mov_immediate): Likewise.
22905 (aarch64_gen_adjusted_ldpstp): Likewise.
22906 (aarch64_scalar_mode_supported_p): Accept DFP modes if enabled.
22907 * config/aarch64/aarch64.md
22908 (movsf_aarch64): Use SFD iterator and rename into
22910 (movdf_aarch64): Use DFD iterator and rename into
22912 (movtf_aarch64): Use TFD iterator and rename into
22914 (split pattern for move TF mode): Use TFD iterator.
22915 * config/aarch64/iterators.md
22916 (GPF_TF_F16_MOV): Add DFP modes.
22917 (SFD, DFD, TFD): New iterators.
22918 (GPF_TF): Add DFP modes.
22919 (TX, DX, DX2): Likewise.
22921 2022-05-20 Christophe Lyon <christophe.lyon@arm.com>
22923 * configure: Regenerate.
22925 2022-05-19 Roger Sayle <roger@nextmovesoftware.com>
22927 PR middle-end/98865
22928 * expr.cc (expand_expr_real_2) [MULT_EXPR]: Expand X*Y as X&Y
22929 when both X and Y are [0, 1], X*Y as X&-Y when Y is [0,1] and
22930 likewise X*Y as -X&Y when X is [0,1] using tree_nonzero_bits.
22932 2022-05-19 Will Schmidt <will_schmidt@vnet.ibm.com>
22934 * config/rs6000/rs6000-builtins.def: Rephrase
22935 to remove RS6000_BTC_SPECIAL from comment.
22936 * config/rs6000/rs6000.h (RS6000_BTC_UNARY, RS6000_BTC_BINARY,
22937 RS6000_BTC_TERNARY, RS6000_BTC_QUATERNARY,
22938 RS6000_BTC_QUINARY, RS6000_BTC_SENARY, RS6000_BTC_OPND_MASK,
22939 RS6000_BTC_SPECIAL, RS6000_BTC_PREDICATE, RS6000_BTC_ABS,
22940 RS6000_BTC_DST, RS6000_BTC_TYPE_MASK, RS6000_BTC_MISC,
22941 RS6000_BTC_CONST, RS6000_BTC_PURE, RS6000_BTC_FP,
22942 RS6000_BTC_QUAD, RS6000_BTC_PAIR, RS6000_BTC_QUADPAIR,
22943 RS6000_BTC_ATTR_MASK, RS6000_BTC_SPR, RS6000_BTC_VOID,
22944 RS6000_BTC_CR, RS6000_BTC_OVERLOADED, RS6000_BTC_GIMPLE,
22945 RS6000_BTC_MISC_MASK, RS6000_BTC_MEM, RS6000_BTC_SAT,
22946 RS6000_BTM_ALWAYS): Delete.
22948 2022-05-19 Richard Biener <rguenther@suse.de>
22950 * omp-expand.cc (expand_omp_atomic_cas): Do not short-cut
22951 computation of the new value.
22953 2022-05-19 Richard Biener <rguenther@suse.de>
22955 * tree-ssa-pre.cc (get_or_alloc_expression_id): Remove.
22956 (add_to_value): Use get_expression_id.
22957 (bitmap_insert_into_set): Likewise.
22958 (bitmap_value_insert_into_set): Likewise.
22960 2022-05-19 David Malcolm <dmalcolm@redhat.com>
22962 * doc/invoke.texi (-fanalyzer-checker=): Add
22963 -Wanalyzer-va-list-leak and -Wanalyzer-va-list-use-after-va-end to
22964 the list of analyzer warnings disabled by
22965 -fanalyzer-checker=taint.
22967 2022-05-19 Jakub Jelinek <jakub@redhat.com>
22970 * cfgexpand.cc (expand_debug_expr): For VAR_DECL, punt for
22971 global vars without symtab node even when they have DECL_RTL
22974 2022-05-19 Jakub Jelinek <jakub@redhat.com>
22977 * pointer-query.cc (gimple_parm_array_size): Return NULL if var
22978 doesn't have pointer or reference type.
22980 2022-05-18 Marek Polacek <polacek@redhat.com>
22983 * doc/invoke.texi: Document -Wenum-int-mismatch.
22985 2022-05-18 Uros Bizjak <ubizjak@gmail.com>
22987 * config/i386/gnu-user-common.h (defined): Only define
22988 TARGET_CAN_SPLIT_STACK for glibc targets.
22989 * config/i386/gnu.h (defined): Ditto.
22991 2022-05-18 Roger Sayle <roger@nextmovesoftware.com>
22993 * config/i386/i386.cc (ix86_rtx_costs) [MULT]: When mode size
22994 is wider than word_mode, a multiplication costs three word_mode
22995 multiplications and two word_mode additions.
22997 2022-05-18 Roger Sayle <roger@nextmovesoftware.com>
22999 * config/i386/i386.md (define_split): Split *andsi_1
23000 and *andn_si_ccno after reload with -Oz.
23002 2022-05-18 Frederik Harwath <frederik@codesourcery.com>
23004 * graphite-scop-detection.cc (scop_detection::can_represent_loop):
23005 Output reason for failure to dump file.
23006 (scop_detection::harmful_loop_in_region): Likewise.
23007 (scop_detection::graphite_can_represent_expr): Likewise.
23008 (scop_detection::stmt_has_simple_data_refs_p): Likewise.
23009 (scop_detection::stmt_simple_for_scop_p): Likewise.
23010 (print_sese_loop_numbers): New function.
23011 (scop_detection::add_scop): Use from here.
23013 2022-05-18 liuhongt <hongtao.liu@intel.com>
23015 PR middle-end/103462
23016 * match.pd (bitwise_induction_p): New match.
23017 * tree-scalar-evolution.cc (gimple_bitwise_induction_p):
23019 (analyze_and_compute_bitwise_induction_effect): New function.
23020 (enum bit_op_kind): New enum.
23021 (final_value_replacement_loop): Enhanced to handle bitwise
23024 2022-05-18 Haochen Gui <guihaoc@gcc.gnu.org>
23027 * config/rs6000/rs6000.md (*subfsi3_carry_in_xx_64): New.
23029 2022-05-18 liuhongt <hongtao.liu@intel.com>
23032 * config/i386/i386.md (*bmi2_bzhi_zero_extendsidi_4): New
23035 2022-05-18 liuhongt <hongtao.liu@intel.com>
23038 * config/i386/i386-expand.cc (ix86_expand_branch): Use ptest
23039 for QImode when code is EQ or NE.
23040 * config/i386/i386.md (cbranchoi4): New expander.
23042 2022-05-18 Peter Bergner <bergner@linux.ibm.com>
23043 Segher Boessenkool <segher@kernel.crashing.org>
23046 * config/rs6000/mma.md (mma_<vv>, mma_<avv>, mma_<pv>, mma_<apv>,
23047 mma_<vvi4i4i8>, mma_<avvi4i4i8>, mma_<vvi4i4i2>, mma_<avvi4i4i2>,
23048 mma_<vvi4i4>, mma_<avvi4i4>, mma_<pvi4i2>, mma_<apvi4i2>,
23049 mma_<vvi4i4i4>, mma_<avvi4i4i4>): Replace "wa" constraints with "v,?wa".
23050 Update other operands accordingly.
23052 2022-05-17 Marek Polacek <polacek@redhat.com>
23054 * godump.cc (go_output_typedef): Use the DECL_INITIAL of the TREE_VALUE.
23056 2022-05-17 Pat Haugen <pthaugen@linux.ibm.com>
23059 * config/rs6000/rs6000-call.cc (rs6000_function_arg_advance_1): Bump
23060 register count when not splitting IEEE 128-bit Complex.
23062 2022-05-17 Tobias Burnus <tobias@codesourcery.com>
23064 * omp-low.cc (check_omp_nesting_restrictions): Skip warning for
23065 target inside target if inner is reverse offload.
23067 2022-05-17 Tobias Burnus <tobias@codesourcery.com>
23069 * config/gcn/mkoffload.cc (process_obj): Revert: Use ARRAY_SIZE.
23070 * config/nvptx/mkoffload.cc (process): Likewise.
23072 2022-05-17 Andrew MacLeod <amacleod@redhat.com>
23074 * Makefile.in (OBJS): Add gimple-range-side-effect.o.
23075 * gimple-range-cache.cc (non_null_ref::non_null_ref): Delete.
23076 (non_null_ref::~non_null_ref): Delete.
23077 (non_null_ref::set_nonnull): Delete.
23078 (non_null_ref::non_null_deref_p): Delete.
23079 (non_null_ref::process_name): Delete.
23080 (ranger_cache::ranger_cache): Initialize m_exit object.
23081 (ranger_cache::fill_block_cache): Use m_exit object intead of nonnull.
23082 (ranger_cache::range_from_dom): Use side_effect class and m_exit object.
23083 (ranger_cache::update_to_nonnull): Delete.
23084 (non_null_loadstore): Delete.
23085 (ranger_cache::block_apply_nonnull): Delete.
23086 (ranger_cache::apply_side_effects): New.
23087 * gimple-range-cache.h (class non_null_ref): Delete.
23088 (non_null_ref::adjust_range): Delete.
23089 (class ranger_cache): Adjust prototypes, add side effect manager.
23090 * gimple-range-path.cc (path_range_query::range_defined_in_block): Use
23091 side effect manager for queries.
23092 (path_range_query::adjust_for_non_null_uses): Ditto.
23093 * gimple-range-path.h (class path_range_query): Delete non_null_ref.
23094 * gimple-range-side-effect.cc: New.
23095 * gimple-range-side-effect.h: New.
23096 * gimple-range.cc (gimple_ranger::gimple_ranger): Update contructor.
23097 (gimple_ranger::range_of_expr): Check def block for override value.
23098 (gimple_ranger::range_on_entry): Don't scan dominators for non-null.
23099 (gimple_ranger::range_on_edge): Check for outgoing side-effects.
23100 (gimple_ranger::register_side_effects): Call apply_side_effects.
23101 (enable_ranger): Update contructor.
23102 * gimple-range.h (class gimple_ranger): Update prototype.
23103 (enable_ranger): Update prototype.
23104 * tree-vrp.cc (execute_ranger_vrp): Invoke without immediate-use flag.
23106 2022-05-17 Giuliano Belinassi <gbelinassi@suse.de>
23109 * targhooks.cc (default_print_patchable_function_entry_1): Handle COMDAT case.
23110 * varasm.cc (switch_to_comdat_section): New
23111 (handle_vtv_comdat_section): Call switch_to_comdat_section.
23112 * varasm.h: Declare switch_to_comdat_section.
23114 2022-05-17 Richard Biener <rguenther@suse.de>
23116 * cfgloopmanip.cc (duplicate_loop_body_to_header_edge): Do
23117 not clear bb->aux of the copied blocks.
23119 2022-05-17 Andrew MacLeod <amacleod@redhat.com>
23121 PR tree-optimization/105458
23122 * value-relation.cc (path_oracle::register_relation): Merge, then check
23125 2022-05-17 Uroš Bizjak <ubizjak@gmail.com>
23129 * config/i386/i386.md: Remove constraints when used with
23130 const_int_operand, const0_operand, const_1_operand, constm1_operand,
23131 const8_operand, const128_operand, const248_operand, const123_operand,
23132 const2367_operand, const1248_operand, const359_operand,
23133 const_4_or_8_to_11_operand, const48_operand, const_0_to_1_operand,
23134 const_0_to_3_operand, const_0_to_4_operand, const_0_to_5_operand,
23135 const_0_to_7_operand, const_0_to_15_operand, const_0_to_31_operand,
23136 const_0_to_63_operand, const_0_to_127_operand, const_0_to_255_operand,
23137 const_0_to_255_mul_8_operand, const_1_to_31_operand,
23138 const_1_to_63_operand, const_2_to_3_operand, const_4_to_5_operand,
23139 const_4_to_7_operand, const_6_to_7_operand, const_8_to_9_operand,
23140 const_8_to_11_operand, const_8_to_15_operand, const_10_to_11_operand,
23141 const_12_to_13_operand, const_12_to_15_operand, const_14_to_15_operand,
23142 const_16_to_19_operand, const_16_to_31_operand, const_20_to_23_operand,
23143 const_24_to_27_operand and const_28_to_31_operand.
23144 * config/i386/mmx.md: Ditto.
23145 * config/i386/sse.md: Ditto.
23146 * config/i386/subst.md: Ditto.
23147 * config/i386/sync.md: Ditto.
23149 2022-05-17 Thomas Schwinge <thomas@codesourcery.com>
23151 * diagnostic.cc: Don't advise to call 'abort' instead of
23153 * system.h: Advise to call 'internal_error' instead of 'abort' or
23156 2022-05-17 Frederik Harwath <frederik@codesourcery.com>
23158 * graphite-sese-to-poly.cc (build_poly_sr_1): Fix a typo and
23159 a reference to a variable which does not exist.
23160 * graphite-isl-ast-to-gimple.cc (gsi_insert_earliest): Fix typo
23163 2022-05-17 Frederik Harwath <frederik@codesourcery.com>
23165 * graphite-sese-to-poly.cc (isl_id_for_ssa_name): Rename to ...
23166 (isl_id_for_parameter): ... this new function name.
23167 (build_scop_context): Adjust function use.
23169 2022-05-17 Tobias Burnus <tobias@codesourcery.com>
23172 * config/gcn/t-omp-device (arch): Add 'amdgcn' besides existing 'gcn'.
23173 * config/gcn/gcn.cc (gcn_omp_device_kind_arch_isa): Likewise.
23175 2022-05-17 Jakub Jelinek <jakub@redhat.com>
23177 * tree-core.h (enum omp_clause_depend_kind): Add
23178 OMP_CLAUSE_DEPEND_INOUTSET.
23179 * tree-pretty-print.cc (dump_omp_clause): Handle
23180 OMP_CLAUSE_DEPEND_INOUTSET.
23181 * gimplify.cc (gimplify_omp_depend): Likewise.
23182 * omp-low.cc (lower_depend_clauses): Likewise.
23184 2022-05-17 Jakub Jelinek <jakub@redhat.com>
23187 * config/i386/sse.md (vec_cmpeqv2div2di, vec_cmpeqv1tiv1ti): Use
23188 andv4si3 only for EQ, for NE use iorv4si3 instead.
23190 2022-05-17 Richard Biener <rguenther@suse.de>
23192 PR tree-optimization/105618
23193 * tree-ssa-sink.cc (statement_sink_location): For virtual
23194 PHI uses ignore those defining the used virtual operand.
23196 2022-05-17 Jakub Jelinek <jakub@redhat.com>
23198 * tree-ssa-forwprop.cc (simplify_bitfield_ref): Spelling fix:
23199 hanlde -> handle. Fix up comment formatting.
23201 2022-05-17 liuhongt <hongtao.liu@intel.com>
23204 * config/i386/sse.md (*vec_concatv4si): Extend to ..
23205 (*vec_concat<mode>): .. V16QI and V8HImode.
23206 (*vec_concatv16qi_permt2): New pre_reload define_insn_and_split.
23207 (*vec_concatv8hi_permt2): Ditto.
23209 2022-05-17 liuhongt <hongtao.liu@intel.com>
23211 PR tree-optimization/105591
23212 * tree-ssa-forwprop.cc (simplify_bitfield_ref): Clamp
23213 vec_perm_expr index.
23215 2022-05-16 Jason Merrill <jason@redhat.com>
23218 * attribs.cc (decl_attributes): Fix broken typedefs here.
23220 2022-05-16 David Malcolm <dmalcolm@redhat.com>
23223 * Makefile.in (ANALYZER_OBJS): Add analyzer/varargs.o.
23224 * doc/invoke.texi: Add -Wanalyzer-va-arg-type-mismatch,
23225 -Wanalyzer-va-list-exhausted, -Wanalyzer-va-list-leak, and
23226 -Wanalyzer-va-list-use-after-va-end.
23228 2022-05-16 Richard Biener <rguenther@suse.de>
23230 * gimple-match.h (gimple_build): Move code_helper overloads ...
23231 * gimple-fold.h (gimple_build): ... here.
23232 (gimple_build): Transition to new worker API. Provide
23233 overloads from sequence-based API.
23234 (gimple_convert): Likewise.
23235 (gimple_convert_to_ptrofftype): Likewise.
23236 (gimple_build_vector_from_val): Likewise.
23237 (gimple_build_vector): Likewise.
23238 (gimple_build_round_up): Likewise.
23239 * gimple-fold.cc (gimple_build_insert_seq): New helper.
23240 (gimple_build): Use it. Transition combined_fn and code_helper
23242 (gimple_convert): Transition to new worker API.
23243 (gimple_convert_to_ptrofftype): Likewise.
23244 (gimple_build_vector_from_val): Likewise.
23245 (gimple_build_vector): Likewise.
23246 (gimple_build_round_up): Likewise.
23248 2022-05-16 Richard Biener <rguenther@suse.de>
23250 * gimple-match.h (code_helper): Move class ...
23251 * tree.h (code_helper): ... here.
23253 2022-05-16 Martin Liska <mliska@suse.cz>
23255 * opts-global.cc (write_langs): Add comment.
23257 2022-05-16 Eric Botcazou <ebotcazou@adacore.com>
23259 * dwarf2out.cc (loc_list_from_tree_1) <TRUTH_NOT_EXPR>: Do a logical
23260 instead of a bitwise negation.
23261 <COND_EXPR>: Swap the operands if the condition is TRUTH_NOT_EXPR.
23263 2022-05-16 Martin Liska <mliska@suse.cz>
23265 * attribs.cc (diag_attr_exclusions): Use ARRAY_SIZE.
23266 (decls_mismatched_attributes): Likewise.
23267 * builtins.cc (c_strlen): Likewise.
23268 * cfg.cc (DEF_BASIC_BLOCK_FLAG): Likewise.
23269 * common/config/aarch64/aarch64-common.cc (aarch64_option_init_struct): Likewise.
23270 * config/aarch64/aarch64-builtins.cc (aarch64_lookup_simd_builtin_type): Likewise.
23271 (aarch64_init_simd_builtin_types): Likewise.
23272 (aarch64_init_builtin_rsqrt): Likewise.
23273 * config/aarch64/aarch64.cc (is_madd_op): Likewise.
23274 * config/arm/arm-builtins.cc (arm_lookup_simd_builtin_type): Likewise.
23275 (arm_init_simd_builtin_types): Likewise.
23276 * config/avr/gen-avr-mmcu-texi.cc (mcus[ARRAY_SIZE): Likewise.
23277 (c_prefix): Likewise.
23279 * config/c6x/c6x.cc (N_SAVE_ORDER): Likewise.
23280 * config/darwin-c.cc (darwin_register_frameworks): Likewise.
23281 * config/gcn/mkoffload.cc (process_obj): Likewise.
23282 * config/i386/i386-builtins.cc (get_builtin_code_for_version): Likewise.
23283 (fold_builtin_cpu): Likewise.
23284 * config/m32c/m32c.cc (PUSHM_N): Likewise.
23285 * config/nvptx/mkoffload.cc (process): Likewise.
23286 * config/rs6000/driver-rs6000.cc (host_detect_local_cpu): Likewise.
23287 * config/s390/s390.cc (NR_C_MODES): Likewise.
23288 * config/tilepro/gen-mul-tables.cc (find_sequences): Likewise.
23289 (create_insn_code_compression_table): Likewise.
23290 * config/vms/vms.cc (NBR_CRTL_NAMES): Likewise.
23291 * diagnostic-format-json.cc (json_from_expanded_location): Likewise.
23292 * dwarf2out.cc (ARRAY_SIZE): Likewise.
23293 * genhooks.cc (emit_documentation): Likewise.
23294 (emit_init_macros): Likewise.
23295 * gimple-ssa-sprintf.cc (format_floating): Likewise.
23296 * gimple-ssa-warn-access.cc (memmodel_name): Likewise.
23297 * godump.cc (keyword_hash_init): Likewise.
23298 * hash-table.cc (hash_table_higher_prime_index): Likewise.
23299 * input.cc (for_each_line_table_case): Likewise.
23300 * ipa-free-lang-data.cc (free_lang_data): Likewise.
23301 * ipa-inline.cc (sanitize_attrs_match_for_inline_p): Likewise.
23302 * optc-save-gen.awk: Likewise.
23303 * spellcheck.cc (test_metric_conditions): Likewise.
23304 * tree-vect-slp-patterns.cc (sizeof): Likewise.
23305 (ARRAY_SIZE): Likewise.
23306 * tree.cc (build_common_tree_nodes): Likewise.
23308 2022-05-16 Martin Liska <mliska@suse.cz>
23310 * opts-global.cc (write_langs): Allocate at least one byte.
23312 2022-05-16 Richard Biener <rguenther@suse.de>
23314 * match.pd (A cmp B ? A : B -> min/max): New patterns
23315 carried over from fold_cond_expr_with_comparison.
23317 2022-05-16 liuhongt <hongtao.liu@intel.com>
23320 * config/i386/i386-expand.cc
23321 (expand_vec_perm_pslldq_psrldq_por): Fail when (d->perm[i] ==
23322 d->perm[i-1] + 1) && d->perm[i] == nelt && start != -1.
23324 2022-05-15 Uroš Bizjak <ubizjak@gmail.com>
23326 * config/i386/i386.md: Remove constraints when used with
23327 const_int_operand, const0_operand, const_1_operand, constm1_operand,
23328 const8_operand, const128_operand, const248_operand, const123_operand,
23329 const2367_operand, const1248_operand, const359_operand,
23330 const_4_or_8_to_11_operand, const48_operand, const_0_to_1_operand,
23331 const_0_to_3_operand, const_0_to_4_operand, const_0_to_5_operand,
23332 const_0_to_7_operand, const_0_to_15_operand, const_0_to_31_operand,
23333 const_0_to_63_operand, const_0_to_127_operand, const_0_to_255_operand,
23334 const_0_to_255_mul_8_operand, const_1_to_31_operand,
23335 const_1_to_63_operand, const_2_to_3_operand, const_4_to_5_operand,
23336 const_4_to_7_operand, const_6_to_7_operand, const_8_to_9_operand,
23337 const_8_to_11_operand, const_8_to_15_operand, const_10_to_11_operand,
23338 const_12_to_13_operand, const_12_to_15_operand, const_14_to_15_operand,
23339 const_16_to_19_operand, const_16_to_31_operand, const_20_to_23_operand,
23340 const_24_to_27_operand and const_28_to_31_operand.
23341 * config/i386/mmx.md: Ditto.
23342 * config/i386/sse.md: Ditto.
23343 * config/i386/subst.md: Ditto.
23344 * config/i386/sync.md: Ditto.
23346 2022-05-13 Roger Sayle <roger@nextmovesoftware.com>
23347 Uroš Bizjak <ubizjak@gmail.com>
23349 * config/i386/sse.md (vec_cmpeqv2div2di): Enable for TARGET_SSE2.
23350 For !TARGET_SSE4_1, expand as a V4SI vector comparison, followed
23351 by a pshufd and pand.
23352 (vec_cmpeqv1tiv1ti): New define_expand implementing V1TImode
23353 vector equality as a V2DImode vector comparison (see above),
23354 followed by a pshufd and pand.
23356 2022-05-13 Roger Sayle <roger@nextmovesoftware.com>
23358 PR tree-optimization/83907
23359 * tree-ssa-strlen.cc (handle_builtin_memset): Record a strinfo
23360 for memset with an constant char value.
23361 (handle_store): Improved handling of stores with a first byte
23362 of zero, but not storing_all_zeros_p.
23364 2022-05-13 Philipp Tomsich <philipp.tomsich@vrull.eu>
23365 Manolis Tsamis <manolis.tsamis@vrull.eu>
23367 * config/riscv/riscv.h (CLZ_DEFINED_VALUE_AT_ZERO): Implement.
23368 (CTZ_DEFINED_VALUE_AT_ZERO): Same.
23369 * doc/sourcebuild.texi: add documentation for RISC-V specific
23370 test target keywords
23372 2022-05-13 Andrew MacLeod <amacleod@redhat.com>
23374 PR tree-optimization/105597
23375 * range-op.cc (operator_minus::lhs_op1_relation): Use op1 instead
23376 of the lhs and make sure it is not undefined.
23378 2022-05-13 Sebastian Pop <spop@amazon.com>
23381 * config/aarch64/aarch64-protos.h (atomic_ool_names): Increase dimension
23383 * config/aarch64/aarch64.cc (aarch64_atomic_ool_func): Call
23384 memmodel_from_int and handle MEMMODEL_SYNC_*.
23385 (DEF0): Add __aarch64_*_sync functions.
23387 2022-05-13 Andrew MacLeod <amacleod@redhat.com>
23389 * gimple-range-fold.cc (fold_using_range::range_of_phi): Use new VREL_*
23391 * gimple-range-path.cc (maybe_register_phi_relation): Ditto.
23392 * range-op.cc (*::lhs_op1_relation): Return relation_kind, and use
23393 new VREL enumerated values.
23394 (*::lhs_op2_relation): Ditto.
23395 (*::op1_op2_relation): Ditto.
23396 (*::fold_range): Use new VREL enumerated values.
23397 (minus_op1_op2_relation_effect): Ditto.
23398 (range_relational_tests): Ditto.
23399 * range-op.h (fold_range, op1_range, op2_range): Use VREL_VARYING.
23400 (lhs_op1_relation, lhs_op2_relation, op1_op2_relation): Return
23402 (*_op1_op2_relation): Return relation_kind.
23403 (relop_early_resolve): Use VREL_UNDEFINED.
23404 * value-query.cc (range_query::query_relation): Use VREL_VARYING.
23405 * value-relation.cc (VREL_LAST): Change enumerated value.
23406 (vrel_range_assert): Delete.
23407 (print_relation): Remove range assert.
23408 (rr_negate_table): Adjust table to use new enumerated values..
23409 (relation_negate): Remove range assert.
23410 (rr_swap_table): Adjust.
23411 (relation_swap): Remove range assert.
23412 (rr_intersect_table): Adjust.
23413 (relation_intersect): Remove range assert.
23414 (rr_union_table): Adjust.
23415 (relation_union): Remove range assert.
23416 (rr_transitive_table): Adjust.
23417 (relation_transitive): Remove range assert.
23418 (equiv_oracle::query_relation): Use new VREL enumerated values.
23419 (equiv_oracle::register_relation): Ditto.
23420 (relation_oracle::register_stmt): Ditto.
23421 (dom_oracle::set_one_relation): Ditto.
23422 (dom_oracle::register_transitives): Ditto.
23423 (dom_oracle::query_relation): Ditto.
23424 (path_oracle::register_relation): Ditto.
23425 (path_oracle::query_relation): Ditto.
23426 * value-relation.h (enum relation_kind_t): New relation_kind.
23427 (*_op1_op2_relation): Adjust prototypes.
23429 2022-05-13 Andrew MacLeod <amacleod@redhat.com>
23431 * gimple-range-edge.cc (calc_switch_ranges): Check union return value.
23432 * value-range.cc (irange::legacy_verbose_union_): Add return value.
23433 (irange::irange_single_pair_union): New.
23434 (irange::irange_union): Add return value.
23435 * value-range.h (class irange): Adjust prototypes.
23437 2022-05-13 Andrew MacLeod <amacleod@redhat.com>
23439 * value-range.cc (irange::legacy_verbose_intersect): Add return value.
23440 (irange::irange_contains_p): New.
23441 (irange::irange_intersect): Add return value.
23442 * value-range.h (class irange): Adjust prototypes.
23444 2022-05-13 Andrew MacLeod <amacleod@redhat.com>
23446 * gimple-range-cache.cc (ranger_cache::get_global_range): Return the
23447 had_global value instead.
23449 2022-05-13 Andrew MacLeod <amacleod@redhat.com>
23451 PR tree-optimization/104547
23452 * gimple-range-fold.cc (fold_using_range::range_of_range_op): Add
23453 the op1/op2 relation to the relation call.
23454 * range-op.cc (*::lhs_op1_relation): Add param.
23455 (*::lhs_op2_relation): Ditto.
23456 (operator_minus::lhs_op1_relation): New.
23457 (range_relational_tests): Add relation param.
23458 * range-op.h (lhs_op1_relation, lhs_op2_relation): Adjust prototype.
23460 2022-05-13 Andrew MacLeod <amacleod@redhat.com>
23462 * gimple-range.cc (gimple_ranger::register_side_effects): First check
23463 if the DEF should be exported as a global.
23464 * tree-vrp.cc (rvrp_folder::pre_fold_bb): Process PHI side effects,
23465 which will export globals.
23466 (execute_ranger_vrp): Remove call to export_global_ranges.
23468 2022-05-13 Andrew MacLeod <amacleod@redhat.com>
23470 * value-relation.cc (path_oracle::reset_path): Clear killing_defs.
23472 2022-05-13 Andrew MacLeod <amacleod@redhat.com>
23474 * gimple-range-cache.cc (ranger_cache::ranger_cache): Start with
23476 (ranger_cache::entry_range): Add rfd_mode parameter.
23477 (ranger_cache::exit_range): Ditto.
23478 (ranger_cache::edge_range): New. Incorporate from range_on_edge.
23479 (ranger_cache::range_of_expr): Adjust call to entry_range.
23480 (ranger_cache::range_on_edge): Split to edge_range and call.
23481 (ranger_cache::fill_block_cache): Always invoke range_from_dom.
23482 (ranger_cache::range_from_dom): Make reentrant, add search mode, handle
23483 mutiple predecessors.
23484 (ranger_cache::update_to_nonnull): Adjust call to exit_range.
23485 * gimple-range-cache.h (ranger_cache): Add enum rfd_mode. Adjust
23488 2022-05-13 Alexandre Oliva <oliva@adacore.com>
23490 * gimple-harden-conditionals.cc: Include sbitmap.h.
23491 (pass_harden_conditional_branches::execute): Skip new blocks.
23492 (pass_harden_compares::execute): Likewise.
23494 2022-05-13 Richard Earnshaw <rearnsha@arm.com>
23497 * config/arm/mve.md (*movmisalign<mode>_mve_store): Use
23498 mve_memory_operand.
23499 (*movmisalign<mode>_mve_load): Likewise.
23500 * config/arm/vec-common.md (movmisalign<mode>): Convert to generator
23502 (@movmisalign<mode>): ... thus. Use generic predicates and then
23503 rework operands if they are not valid. For MVE rework to a
23504 narrower element size if the alignment is not high enough.
23506 2022-05-13 Richard Earnshaw <rearnsha@arm.com>
23508 * config/arm/arm.cc (mve_vector_mem_operand): Allow SP_REGNUM
23509 when there is no write-back. Fix use when strict is true.
23511 2022-05-13 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
23513 * config/xtensa/xtensa.h (TARGET_HAS_NO_HW_DIVIDE): New macro
23516 2022-05-13 Takayuki 'January June' Suwa <jjsuwa_sys3175@yahoo.co.jp>
23518 * config/xtensa/xtensa.md (extvsi, extvsi_internal, extzvsi,
23519 extzvsi_internal): Rename from extv, extv_internal, extzv and
23520 extzv_internal, respectively.
23522 2022-05-13 Eric Botcazou <ebotcazou@adacore.com>
23524 * tree-sra.cc (sra_modify_assign): Check that scalar storage order
23525 is the same on the LHS and RHS before rewriting one with the model
23528 2022-05-13 Richard Biener <rguenther@suse.de>
23530 * gimple-fold.cc (gimple_build): Adjust for new
23532 * gimple-fold.h (gimple_build): New main APIs with
23533 iterator, insert direction and iterator update.
23534 (gimple_build): New forwarder template.
23535 (clear_padding_type_may_have_padding_p): Remove.
23536 (clear_type_padding_in_mask): Likewise.
23537 (arith_overflowed_p): Likewise.
23538 * fold-const.h (clear_padding_type_may_have_padding_p): Declare.
23539 (clear_type_padding_in_mask): Likewise.
23540 (arith_overflowed_p): Likewise.
23541 * tree-vect-generic.cc (gimplify_build3): Use main gimple_build API.
23542 (gimplify_build2): Likewise.
23543 (gimplify_build1): Likewise.
23544 * ubsan.cc (ubsan_expand_ptr_ifn): Likewise, avoid extra
23546 * gengtype.cc (open_base_files): Re-order includes.
23547 * builtins.cc: Re-order gimple-fold.h include.
23548 * calls.cc: Likewise.
23549 * cgraphbuild.cc: Likewise.
23550 * cgraphunit.cc: Likewise.
23551 * config/rs6000/rs6000-builtin.cc: Likewise.
23552 * config/rs6000/rs6000-call.cc: Likewise.
23553 * config/rs6000/rs6000.cc: Likewise.
23554 * config/s390/s390.cc: Likewise.
23555 * expr.cc: Likewise.
23556 * fold-const.cc: Likewise.
23557 * function-tests.cc: Likewise.
23558 * gimple-match-head.cc: Likewise.
23559 * gimple-range-fold.cc: Likewise.
23560 * gimple-ssa-evrp-analyze.cc: Likewise.
23561 * gimple-ssa-evrp.cc: Likewise.
23562 * gimple-ssa-sprintf.cc: Likewise.
23563 * gimple-ssa-warn-access.cc: Likewise.
23564 * gimplify.cc: Likewise.
23565 * graphite-isl-ast-to-gimple.cc: Likewise.
23566 * ipa-cp.cc: Likewise.
23567 * ipa-devirt.cc: Likewise.
23568 * ipa-prop.cc: Likewise.
23569 * omp-low.cc: Likewise.
23570 * pointer-query.cc: Likewise.
23571 * range-op.cc: Likewise.
23572 * tree-cfg.cc: Likewise.
23573 * tree-if-conv.cc: Likewise.
23574 * tree-inline.cc: Likewise.
23575 * tree-object-size.cc: Likewise.
23576 * tree-ssa-ccp.cc: Likewise.
23577 * tree-ssa-dom.cc: Likewise.
23578 * tree-ssa-forwprop.cc: Likewise.
23579 * tree-ssa-ifcombine.cc: Likewise.
23580 * tree-ssa-loop-ivcanon.cc: Likewise.
23581 * tree-ssa-math-opts.cc: Likewise.
23582 * tree-ssa-pre.cc: Likewise.
23583 * tree-ssa-propagate.cc: Likewise.
23584 * tree-ssa-reassoc.cc: Likewise.
23585 * tree-ssa-sccvn.cc: Likewise.
23586 * tree-ssa-strlen.cc: Likewise.
23587 * tree-ssa.cc: Likewise.
23588 * value-pointer-equiv.cc: Likewise.
23589 * vr-values.cc: Likewise.
23591 2022-05-13 Alexandre Oliva <oliva@adacore.com>
23593 PR rtl-optimization/105455
23594 * gimple-harden-conditionals.cc (insert_check_and_trap): Set
23595 probabilities for newly-conditional edges.
23597 2022-05-13 liuhongt <hongtao.liu@intel.com>
23599 PR tree-optimization/102583
23600 * tree-ssa-forwprop.cc (simplify_bitfield_ref): Extended to a
23601 contiguous stride in the VEC_PERM_EXPR.
23603 2022-05-12 Richard Biener <rguenther@suse.de>
23605 PR rtl-optimization/105577
23606 * dse.cc (rest_of_handle_dse): Make sure to purge dead EH
23607 edges before running fast DCE via df_analyze.
23609 2022-05-12 Richard Biener <rguenther@suse.de>
23611 PR tree-optimization/105562
23612 * tree-ssa-sccvn.cc (vn_reference_lookup_3): Disambiguate
23613 against all CLOBBER defs if there's not an obvious must-alias
23614 and we are not doing redundant store elimination.
23615 (vn_walk_cb_data::redundant_store_removal_p): New field.
23616 (vn_reference_lookup_pieces): Initialize it.
23617 (vn_reference_lookup): Add argument to specify if we are
23618 doing redundant store removal.
23619 (eliminate_dom_walker::eliminate_stmt): Specify we do.
23620 * tree-ssa-sccvn.h (vn_reference_lookup): Adjust.
23622 2022-05-12 Haochen Jiang <haochen.jiang@intel.com>
23625 * config/i386/sse.md (vi1avx2const): New define_mode_attr.
23626 (pxor/pcmpeqb/pmovmskb/cmp 0xffff to ptest splitter):
23627 New define_split pattern.
23629 2022-05-12 Jakub Jelinek <jakub@redhat.com>
23631 * gimplify.cc (gimplify_omp_depend): Don't build_fold_addr_expr
23632 if null_pointer_node.
23633 (gimplify_scan_omp_clauses): Likewise.
23634 * tree-pretty-print.cc (dump_omp_clause): Print null_pointer_node
23637 2022-05-11 Patrick Palka <ppalka@redhat.com>
23639 * tree.h (TREE_VEC_BEGIN): Define.
23640 (TREE_VEC_END): Correct 'length' member access.
23641 (class tree_vec_range): Define.
23643 2022-05-11 Segher Boessenkool <segher@kernel.crashing.org>
23645 * config/rs6000/rs6000.md: Use wa instead of <Fv>.
23647 2022-05-11 Segher Boessenkool <segher@kernel.crashing.org>
23649 * config/rs6000/rs6000.md: Use d instead of <Ff>.
23651 2022-05-11 Segher Boessenkool <segher@kernel.crashing.org>
23653 * config/rs6000/constraints.md (register_constraint "f"): Use
23654 RS6000_CONSTRAINT_d.
23655 * config/rs6000/rs6000.cc (rs6000_debug_reg_global): Do not handle
23656 RS6000_CONSTRAINT_f.
23657 (rs6000_init_hard_regno_mode_ok): Ditto. Reorder and simplify a bit.
23658 * config/rs6000/rs6000.h (r6000_reg_class_enum): Delete
23659 RS6000_CONSTRAINT_d.
23661 2022-05-11 Richard Biener <rguenther@suse.de>
23663 * gimple-fold.h (gimple_build): Use variadic template
23664 functions for the gimple_build API forwarders without
23665 location_t argument.
23667 2022-05-11 Siddhesh Poyarekar <siddhesh@gotplt.org>
23669 PR middle-end/70090
23670 * asan.cc (initialize_sanitizer_builtins): Register
23671 __builtin_dynamic_object_size if necessary.
23673 2022-05-11 Richard Biener <rguenther@suse.de>
23675 PR rtl-optimization/105559
23676 * cfgrtl.cc (delete_insn_and_edges): Only perform search to BB_END
23677 for non-debug insns.
23679 2022-05-11 Richard Biener <rguenther@suse.de>
23681 * generic-match-head.cc: Include tree-eh.h.
23682 * match.pd ((cond ...) cmp X): New simplification inspired
23683 by fold_binary_op_with_conditional_arg.
23684 (eq/ne (cmp ...) true/false): Likewise.
23686 2022-05-11 Thomas Schwinge <thomas@codesourcery.com>
23688 * doc/install.texi: Don't document '--with-hsa-runtime',
23689 '--with-hsa-runtime-include', '--with-hsa-runtime-lib'.
23691 2022-05-11 Martin Liska <mliska@suse.cz>
23694 * doc/install.texi: Document the configure option --with-zstd.
23696 2022-05-11 Martin Liska <mliska@suse.cz>
23698 * common/config/i386/cpuinfo.h (has_cpu_feature): Directly
23699 compute index in cpu_features2.
23700 (set_cpu_feature): Likewise.
23701 * config/i386/i386-builtins.cc (fold_builtin_cpu): Also remove
23702 loop for cpu_features2 and use NOP_EXPRs.
23704 2022-05-11 Richard Biener <rguenther@suse.de>
23706 PR bootstrap/105551
23707 * opts.cc (finish_options): Also disable var-tracking if
23708 !DWARF2_DEBUGGING_INFO.
23710 2022-05-11 liuhongt <hongtao.liu@intel.com>
23713 * config/i386/sse.md (*vec_set<mode>_0_zero_extendhi): New
23714 pre_reload define_insn_and_split.
23715 (*vec_setv2di_0_zero_extendhi_1): Ditto.
23716 (*vec_set<mode>_0_zero_extendsi): Ditto.
23717 (*vec_setv2di_0_zero_extendsi_1): Ditto.
23718 (ssewvecmode): New mode attr.
23719 (ssewvecmodelower): Ditto.
23720 (ssepackmodelower): Ditto.
23722 2022-05-11 Kewen Lin <linkw@linux.ibm.com>
23724 * config/rs6000/constraints.md (register constraint v): Use
23725 rs6000_constraints[RS6000_CONSTRAINT_v] instead of ALTIVEC_REGS.
23727 2022-05-11 Martin Liska <mliska@suse.cz>
23730 * config/riscv/riscv.opt: Remove Separate from
23731 -msmall-data-limit=.
23732 * optc-gen.awk: Report error for the described situation.
23733 * gcc.cc: Use Separate syntax.
23734 * opts.cc (gen_command_line_string): Change option name.
23736 2022-05-11 Haochen Gui <guihaoc@gcc.gnu.org>
23739 * match.pd (minmax): Skip constant folding for fmin/fmax when both
23740 arguments are sNaN or one is sNaN and another is NaN.
23742 2022-05-10 Jakub Jelinek <jakub@redhat.com>
23744 PR tree-optimization/105528
23745 * gimple-isel.cc (gimple_expand_vec_set_expr): After gsi_remove
23746 set *gsi to gsi_for_stmt (ass_stmt). Fix up function comment.
23748 2022-05-10 Segher Boessenkool <segher@kernel.crashing.org>
23750 * doc/md.texi (Defining Mode Iterators): Correct example replacement
23753 2022-05-10 Maciej W. Rozycki <macro@embecosm.com>
23755 * config/riscv/riscv.md (UNSPEC_FMIN, UNSPEC_FMAX): New
23757 (fmin<mode>3, fmax<mode>3): New insns.
23759 2022-05-10 Richard Biener <rguenther@suse.de>
23761 * tree-if-conv.cc (fold_build_cond_expr): Use
23762 match-and-simplify to simplify the condition.
23763 (ifcvt_follow_ssa_use_edges): Remove.
23764 (predicate_scalar_phi): Use follow_all_ssa_edges.
23766 2022-05-10 Thomas Schwinge <thomas@codesourcery.com>
23768 PR middle-end/100400
23769 * omp-oacc-kernels-decompose.cc
23770 (visit_loops_in_gang_single_region) <GIMPLE_OMP_FOR>: Explicitly
23771 call 'internal_error'.
23773 2022-05-10 Richard Biener <rguenther@suse.de>
23775 PR middle-end/105537
23776 * toplev.cc (process_options): Move flag_var_tracking
23778 * opts.cc (finish_options): ... here.
23780 2022-05-10 Martin Liska <mliska@suse.cz>
23782 * basic-block.h (struct basic_block_d): Use void *
23784 * cfgloop.h: Likewise.
23785 * cgraph.h: Likewise.
23786 * gengtype-state.cc (state_ident_by_name): Likewise.
23787 (record_type): Likewise.
23788 (read_state_already_seen_type): Likewise.
23789 * gengtype.cc (dump_type): Likewise.
23790 (input_file_by_name): Likewise.
23792 * ggc-common.cc (ggc_cleared_alloc_ptr_array_two_args): Likewise.
23793 * ipa-utils.h (struct ipa_dfs_info): Likewise.
23794 * plugin.cc (htab_hash_plugin): Likewise.
23796 2022-05-10 Richard Biener <rguenther@suse.de>
23798 * flags.h (dwarf_debuginfo_p): Add opts argument, guard
23799 API with !GENERATOR_FILE.
23800 * opts.cc (global_options): Poison.
23801 (global_options_set): Likewise.
23802 (finish_options): Refer to options via opts.
23804 2022-05-10 Uroš Bizjak <ubizjak@gmail.com>
23805 Roger Sayle <roger@nextmovesoftware.com>
23807 * config/i386/i386.md (*testqi_1_maybe_si): Prefer shorter SImode
23808 alternative when optimizing for size and the immediate operand is
23809 const_0_to_127_operand.
23810 (*andqi_2_maybe_si): Likewise.
23811 * config/i386/predicates.md (const_0_to_127_operand): New predicate.
23813 2022-05-10 Roger Sayle <roger@nextmovesoftware.com>
23815 * tree-chrec.cc (chrec_apply): Attempt to fold the linear chrec
23816 "{a, +, a} (x-1)" as "a*x", as the number of loop iterations, x-1,
23819 2022-05-10 Eric Botcazou <ebotcazou@adacore.com>
23822 * config/sparc/sparc.cc (sparc_vectorize_vec_perm_const): Return
23823 true only for 8-byte vector modes.
23825 2022-05-10 Siddhesh Poyarekar <siddhesh@gotplt.org>
23827 PR middle-end/70090
23828 * ubsan.cc (ubsan_expand_objsize_ifn): Allow non-constant SIZE.
23829 (instrument_object_size): Get dynamic object size expression.
23831 2022-05-10 Jiufu Guo <guojiufu@linux.ibm.com>
23833 PR preprocessor/101168
23834 * config/rs6000/rs6000-c.cc (rs6000_macro_to_expand):
23835 Avoid empty identifier.
23837 2022-05-09 Qing Zhao <qing.zhao@oracle.com>
23840 * config/i386/i386.cc (zero_call_used_regno_mode): use V2SImode
23841 as a generic MMX mode instead of V4HImode.
23842 (zero_all_mm_registers): Use SET to zero instead of MOV for
23843 zeroing scratch registers.
23844 (ix86_zero_call_used_regs): Likewise.
23846 2022-05-09 liuhongt <hongtao.liu@intel.com>
23849 * config/i386/i386-expand.cc
23850 (expand_vec_perm_pslldq_psrldq_por): New function.
23851 (ix86_expand_vec_perm_const_1): Try
23852 expand_vec_perm_pslldq_psrldq_por for both 3-instruction and
23853 4/5-instruction sequence.
23855 2022-05-09 Martin Liška <mliska@suse.cz>
23857 * basic-block.h (STATIC_ASSERT): Use normal STATIC_ASSERT.
23858 * system.h (STATIC_ASSERT): Define as static_assert for C++
23859 and fallback to array index in C.
23861 2022-05-09 Richard Biener <rguenther@suse.de>
23863 PR tree-optimization/105517
23864 * tree-ssa-sccvn.cc (vn_reference_lookup): Make sure the accumulated
23865 offset can be represented in the POINTER_PLUS_EXPR IL.
23866 (vn_reference_insert): Likewise.
23867 * poly-int.h (sext_hwi): Add poly version of sext_hwi.
23869 2022-05-09 Richard Biener <rguenther@suse.de>
23871 * match.pd: Remove #if GIMPLE guards around ! using patterns.
23873 2022-05-09 liuhongt <hongtao.liu@intel.com>
23876 * config/i386/sse.md (*sse4_1_<code>v2qiv2di2<mask_name>_1):
23878 (*sse4_1_zero_extendv2qiv2di2_2): New pre_reload
23879 define_insn_and_split.
23881 2022-05-09 Alex Coplan <alex.coplan@arm.com>
23883 * cgraph.cc (cgraph_node::get_create): Don't dump if the current
23884 symtab state is PARSING.
23886 2022-05-09 Martin Liska <mliska@suse.cz>
23888 * system.h (LIKELY): Define.
23889 (UNLIKELY): Likewise.
23890 * domwalk.cc (sort_bbs_postorder): Use {,UN}LIKELY
23892 * dse.cc (set_position_unneeded): Likewise.
23893 (set_all_positions_unneeded): Likewise.
23894 (any_positions_needed_p): Likewise.
23895 (all_positions_needed_p): Likewise.
23896 * expmed.cc (flip_storage_order): Likewise.
23897 * genmatch.cc (dt_simplify::gen_1): Likewise.
23898 * ggc-common.cc (gt_pch_save): Likewise.
23899 * print-rtl.cc: Likewise.
23900 * rtl-iter.h (T>::array_type::~array_type): Likewise.
23901 (T>::next): Likewise.
23902 * rtl-ssa/internals.inl: Likewise.
23903 * rtl-ssa/member-fns.inl: Likewise.
23904 * rtlanal.cc (T>::add_subrtxes_to_queue): Likewise.
23905 (rtx_properties::try_to_add_dest): Likewise.
23906 * rtlanal.h (growing_rtx_properties::repeat): Likewise.
23907 (vec_rtx_properties_base::~vec_rtx_properties_base): Likewise.
23908 * simplify-rtx.cc (simplify_replace_fn_rtx): Likewise.
23909 * sort.cc (likely): Likewise.
23910 (mergesort): Likewise.
23911 * wide-int.h (wi::eq_p): Likewise.
23912 (wi::ltu_p): Likewise.
23913 (wi::cmpu): Likewise.
23914 (wi::bit_and): Likewise.
23915 (wi::bit_and_not): Likewise.
23916 (wi::bit_or): Likewise.
23917 (wi::bit_or_not): Likewise.
23918 (wi::bit_xor): Likewise.
23919 (wi::add): Likewise.
23920 (wi::sub): Likewise.
23922 2022-05-09 Kito Cheng <kito.cheng@sifive.com>
23924 * config/riscv/arch-canonicalize: Handle g correctly.
23926 2022-05-07 Marek Polacek <polacek@redhat.com>
23930 * tree-core.h (struct attribute_spec): Update comment for HANDLER.
23932 2022-05-06 Jason Merrill <jason@redhat.com>
23934 * diagnostic-color.cc: Add fnname and targs color entries.
23935 * doc/invoke.texi: Document them.
23937 2022-05-06 Jason Merrill <jason@redhat.com>
23939 * vec.h (vec::iterate): Fix comment.
23941 2022-05-06 Michael Meissner <meissner@linux.ibm.com>
23944 * config/rs6000/rs6000.cc (rs6000_can_inline_p): Ignore -mpower8-fusion
23945 and -mpower10-fusion options for inlining purposes.
23947 2022-05-06 Christophe Lyon <christophe.lyon@arm.com>
23949 * config/aarch64/iterators.md (GPF_TF_F16): Delete.
23951 2022-05-06 Marcel Vollweiler <marcel@codesourcery.com>
23953 * omp-low.cc (omp_runtime_api_call): Added target_is_accessible to
23954 omp_runtime_apis array.
23956 2022-05-06 Hafiz Abid Qadeer <abidh@codesourcery.com>
23958 * omp-low.cc (omp_maybe_offloaded_ctx): New prototype.
23959 (scan_sharing_clauses): Check a restriction on allocate clause.
23961 2022-05-05 Sandra Loosemore <sandra@codesourcery.com>
23963 * gimplify.cc (gimplify_omp_for): Update messages for SCHEDULED
23964 and ORDERED clause conflict errors. Add check for GRAINSIZE and
23965 NUM_TASKS on TASKLOOP.
23967 2022-05-05 Martin Liska <mliska@suse.cz>
23969 * genautomata.cc (create_composed_state): Remove dead code.
23970 * graphite-poly.cc (print_pdrs): Likewise.
23971 * lto-wrapper.cc (run_gcc): Likewise.
23972 * tree-switch-conversion.cc (switch_decision_tree::balance_case_nodes):
23975 2022-05-05 Martin Liska <mliska@suse.cz>
23977 * tree-profile.cc (gimple_gen_ic_profiler): Prefix names with
23979 (gimple_gen_time_profiler): Likewise.
23981 2022-05-05 Martin Liska <mliska@suse.cz>
23983 * value-prof.cc (stream_out_histogram_value): Remove sanity
23986 2022-05-05 Richard Biener <rguenther@suse.de>
23988 PR tree-optimization/104162
23989 * tree-ssa-sccvn.cc (vn_reference_lookup): Handle
23990 &MEM[_1 + 5].a[i] like a POINTER_PLUS_EXPR if the offset
23992 (vn_reference_insert): Likewise.
23994 2022-05-05 Richard Biener <rguenther@suse.de>
23996 * tree-core.h (tree_real_cst::real_cst_ptr): Remove pointer
23997 to real_value field.
23998 (tree_real_cst::value): Add real_value field.
23999 * tree.h (TREE_REAL_CST_PTR): Adjust.
24000 * tree.cc (build_real): Remove separate allocation.
24001 * tree-streamer-in.cc (unpack_ts_real_cst_value_fields):
24004 2022-05-05 Richard Biener <rguenther@suse.de>
24006 * tree-ssa-ifcombine.cc (bb_no_side_effects_p): Avoid executing
24007 divisions with undefined overflow unconditionally.
24008 (pass_tree_ifcombine::execute): Rewrite stmts with undefined
24009 overflow to defined.
24011 2022-05-05 Richard Biener <rguenther@suse.de>
24013 PR tree-optimization/105484
24014 * gimple-isel.cc (gimple_expand_vec_set_expr): Clean EH, return
24015 whether the CFG changed.
24016 (gimple_expand_vec_exprs): When the CFG changed, clean it up.
24018 2022-05-05 Richard Biener <rguenther@suse.de>
24020 PR tree-optimization/104595
24021 * tree-vect-patterns.cc (vect_recog_bool_pattern): For
24022 COND_EXPR do not fail if check_bool_pattern returns false.
24024 2022-05-04 Tobias Burnus <tobias@codesourcery.com>
24026 * omp-low.cc (lower_omp_target): Fix use_device_{addr,ptr} with list
24027 item that is in an outer data-sharing clause.
24029 2022-05-04 Richard Biener <rguenther@suse.de>
24031 PR tree-optimization/104658
24032 * tree-vect-slp.cc (vect_slp_convert_to_external): Do not
24033 create VECTOR_BOOLEAN_TYPE_P extern defs. Reset the vector
24034 type on nodes we promote.
24035 (vectorizable_bb_reduc_epilogue): Deal with externalized
24037 * tree-vect-stmts.cc (vect_maybe_update_slp_op_vectype): Do
24038 not allow VECTOR_BOOLEAN_TYPE_P extern defs.
24040 2022-05-04 Richard Biener <rguenther@suse.de>
24042 PR tree-optimization/103116
24043 * tree-vect-stmts.cc (get_group_load_store_type): Handle the
24044 case we need peeling for gaps even though GROUP_GAP is zero.
24046 2022-05-04 Martin Liska <mliska@suse.cz>
24048 * gengtype-state.cc (read_a_state_token): Remove dead code.
24049 * ipa-profile.cc (ipa_profile_read_summary_section): Likewise.
24051 2022-05-04 Richard Biener <rguenther@suse.de>
24053 * gimplify.cc (gimplify_init_constructor): First gimplify,
24054 then simplify the result to a VECTOR_CST.
24056 2022-05-04 Jakub Jelinek <jakub@redhat.com>
24058 * genconditions.cc (write_conditions): Append a { nullptr, -1 }
24059 element at the end of insn_conditions.
24060 (write_writer): Use ARRAY_SIZE (insn_conditions) - 1 instead of
24061 ARRAY_SIZE (insn_conditions).
24063 2022-05-04 Christophe Lyon <christophe.lyon@arm.com>
24066 * config/arm/arm.cc (simd_valid_immediate): Exit when input is a
24067 vector of booleans and MVE is not enabled.
24069 2022-05-04 Richard Biener <rguenther@suse.de>
24072 * tree-cfgcleanup.cc (move_debug_stmts_from_forwarder):
24073 Move debug stmts to the predecessor if moving to the
24074 destination is not possible.
24075 (remove_forwarder_block): Adjust.
24076 (remove_forwarder_block_with_phi): Likewise.
24078 2022-05-03 Roger Sayle <roger@nextmovesoftware.com>
24080 PR tree-optimization/102950
24081 * range-op.cc (wi_optimize_signed_bitwise_op): New function to
24082 determine bounds of bitwise operations on signed types.
24083 (operator_bitwise_and::wi_fold): Call the above function.
24084 (operator_bitwise_or::wi_fold): Likewise.
24085 (operator_bitwise_xor::wi_fold): Likewise. Additionally, the
24086 result can't be zero if the operands can't be equal.
24088 2022-05-03 Iain Sandoe <iain@sandoe.co.uk>
24090 * config/darwin.cc (darwin_label_is_anonymous_local_objc_name): Make
24091 protocol class methods linker-visible.
24093 2022-05-03 Uroš Bizjak <ubizjak@gmail.com>
24096 * config/i386/sse.md (*vec_extract<mode>_0_mem): New pre-reload
24097 define_insn_and_split pattern.
24099 2022-05-03 Richard Biener <rguenther@suse.de>
24101 PR middle-end/105083
24102 * tree-scalar-evolution.cc (scev_initialize): Verify we
24103 have appropriate loop state.
24104 * tree-ssa-dce.cc (perform_tree_ssa_dce): Re-order SCEV and
24105 loop init and finalization.
24107 2022-05-03 Richard Biener <rguenther@suse.de>
24109 PR middle-end/105461
24110 * opts.cc (finish_options): Match the condition to
24111 disable flag_var_tracking to that of process_options.
24113 2022-05-03 Richard Biener <rguenther@suse.de>
24115 * opts.cc: #undef OPTIONS_SET_P.
24116 (finish_options): Use opts_set instead of OPTIONS_SET_P.
24118 2022-05-03 Richard Biener <rguenther@suse.de>
24120 PR tree-optimization/105394
24121 * tree-vect-generic.cc (expand_vector_condition): Adjust
24122 comp_width for non-integer mode masks as well.
24124 2022-05-03 Marcel Vollweiler <marcel@codesourcery.com>
24126 * omp-low.cc (omp_runtime_api_call): Added get_mapped_ptr to
24127 omp_runtime_apis array.
24129 2022-05-02 Richard Biener <rguenther@suse.de>
24131 * tree-scalar-evolution.cc (expression_expensive_p):
24132 Never consider mismatched calls as cheap.
24134 2022-05-02 Richard Biener <rguenther@suse.de>
24136 PR tree-optimization/104240
24137 * tree-vect-slp.cc (op1_op0_map): New.
24138 (vect_get_operand_map): Handle compares.
24139 (vect_build_slp_tree_1): Support swapped operands for
24142 2022-05-02 Jakub Jelinek <jakub@redhat.com>
24145 * cfgexpand.cc (expand_debug_expr): Don't make_decl_rtl_for_debug
24146 if there is no symtab node for the VAR_DECL.
24148 2022-05-02 Sebastian Huber <sebastian.huber@embedded-brains.de>
24150 * gcov-io.cc (gcov_rewrite): Clear the file error status.
24152 2022-05-02 Richard Biener <rguenther@suse.de>
24154 PR tree-optimization/105437
24155 * tree-vect-slp.cc (vect_schedule_slp_node): Handle the
24156 case where last_stmt alters control flow.
24158 2022-05-02 Richard Biener <rguenther@suse.de>
24160 * dojump.cc (do_jump): Use CASE_CONVERT.
24161 * tree-ssa-dom.cc (edge_info::derive_equivalences): Likewise.
24163 2022-05-02 Jakub Jelinek <jakub@redhat.com>
24165 * system.h: Include initializer_list.
24167 2022-05-01 Segher Boessenkool <segher@kernel.crashing.org>
24169 * config/rs6000/constraints.md (Y constraint): Fix comment.
24171 2022-05-01 Aldy Hernandez <aldyh@redhat.com>
24173 * tree-ssanames.cc (set_range_info): Denormalize VR_VARYING to
24174 VR_RANGE before passing a piecewise range to set_range_info_raw.
24176 2022-04-30 Patrick Palka <ppalka@redhat.com>
24178 * gengtype.cc (adjust_field_tree_exp): Remove.
24179 (adjust_field_type): Don't handle the "tree_exp" special attribute.
24180 * tree-core.h (struct tree_exp): Remove "special" and "desc"
24181 attributes. Add "length" attribute.
24183 2022-04-29 Martin Jambor <mjambor@suse.cz>
24186 * cgraph.cc (cgraph_node::remove): Release body of the node this
24187 is clone_of if appropriate.
24189 2022-04-29 Uroš Bizjak <ubizjak@gmail.com>
24192 * config/i386/i386.md (adcl/neg -> sbb peephole): New peephole2.
24194 2022-04-29 Richard Biener <rguenther@suse.de>
24196 * gimple-expr.cc (is_gimple_condexpr): Adjust comment.
24197 (canonicalize_cond_expr_cond): Move here from gimple.cc,
24198 allow both COND_EXPR and GIMPLE_COND forms.
24199 * gimple-expr.h (canonicalize_cond_expr_cond): Declare.
24200 * gimple.cc (canonicalize_cond_expr_cond): Remove here.
24201 * gimple.h (canonicalize_cond_expr_cond): Likewise.
24202 * gimple-loop-versioning.cc (loop_versioning::version_loop):
24203 Use is_gimple_condexpr_for_cond.
24204 * tree-parloops.cc (gen_parallel_loop): Likewise.
24205 * tree-ssa-ifcombine.cc (ifcombine_ifandif): Check for
24206 a proper cond expr after canonicalize_cond_expr_cond.
24207 Use is_gimple_condexpr_for_cond where appropriate.
24208 * tree-ssa-loop-manip.cc (determine_exit_conditions): Likewise.
24209 * tree-vect-loop-manip.cc (slpeel_add_loop_guard): Likewise.
24211 2022-04-29 Richard Biener <rguenther@suse.de>
24213 * gimple-iterator.h (gsi_after_labels): Add overload for
24215 (gsi_start_1): Rename to gsi_start and take a reference.
24216 (gsi_last_1): Likewise.
24217 * gimple-iterator.cc (gsi_for_stmt): Use gsi_start.
24218 * omp-low.cc (lower_rec_input_clauses): Likewise.
24219 (lower_omp_scan): Likewise.
24221 2022-04-29 Richard Biener <rguenther@suse.de>
24223 PR tree-optimization/105431
24224 * tree-ssa-math-opts.cc (powi_as_mults_1): Make n unsigned.
24225 (powi_as_mults): Use absu_hwi.
24226 (gimple_expand_builtin_powi): Remove now pointless n != -n
24229 2022-04-29 Aldy Hernandez <aldyh@redhat.com>
24231 * range-op.cc (empty_range_varying): Move to range-op.h.
24232 (range_true): Move to range.h.
24233 (range_false): Same.
24234 (range_true_and_false): Same.
24235 (enum bool_range_state): Move to range-op.h.
24236 (relop_early_resolve): Same.
24237 (operator_equal::op1_op2_relation): Abstract code to...
24238 (equal_op1_op2_relation): ...here.
24239 (operator_not_equal::op1_op2_relation): Abstract code to...
24240 (not_equal_op1_op2_relation): ...here.
24241 (operator_lt::op1_op2_relation): Abstract code to...
24242 (lt_op1_op2_relation): ...here.
24243 (operator_le::op1_op2_relation): Abstract code to...
24244 (le_op1_op2_relation): ...here.
24245 (operator_gt::op1_op2_relation): Abstract code to...
24246 (gt_op1_op2_relation): ...here.
24247 (operator_ge::op1_op2_relation): Abstract code to...
24248 (ge_op1_op2_relation): ...here.
24249 (class range_op_table): Move to range-op.h.
24250 * range-op.h (equal_op1_op2_relation): Moved from range-op.cc.
24251 (not_equal_op1_op2_relation): Same.
24252 (lt_op1_op2_relation): Same.
24253 (le_op1_op2_relation): Same.
24254 (gt_op1_op2_relation): Same.
24255 (ge_op1_op2_relation): Same.
24256 (enum bool_range_state): Same.
24257 (get_bool_state): Same.
24258 (empty_range_varying): Same.
24259 (relop_early_resolve): Same.
24260 (class range_op_table): Same.
24261 * range.h (range_true): Same.
24262 (range_false): Same.
24263 (range_true_and_false): Same.
24265 2022-04-29 Aldy Hernandez <aldyh@redhat.com>
24267 * gimple-fold.cc (size_must_be_zero_p): Use reference
24269 * gimple-ssa-evrp-analyze.cc
24270 (evrp_range_analyzer::record_ranges_from_incoming_edge): Rename
24271 intersect to legacy_verbose_intersect.
24272 * ipa-cp.cc (ipcp_vr_lattice::meet_with_1): Use reference instead
24274 * tree-ssa-dom.cc (dom_jt_simplifier::simplify): Use value_range
24275 instead of value_range_equiv.
24276 * tree-vrp.cc (extract_range_from_plus_minus_expr): Use reference
24277 instead of pointer.
24278 (find_case_label_range): Same.
24279 * value-range-equiv.cc (value_range_equiv::intersect): Rename to...
24280 (value_range_equiv::legacy_verbose_intersect): ...this.
24281 (value_range_equiv::union_): Rename to...
24282 (value_range_equiv::legacy_verbose_union_): ...this.
24283 * value-range-equiv.h (class value_range_equiv): Rename union and
24284 intersect to legacy_verbose_{intersect,union}.
24285 * value-range.cc (irange::union_): Rename to...
24286 (irange::legacy_verbose_union_): ...this.
24287 (irange::intersect): Rename to...
24288 (irange::legacy_verbose_intersect): ...this.
24289 * value-range.h (irange::union_): Rename union_ to
24290 legacy_verbose_union.
24291 (irange::intersect): Rename intersect to legacy_verbose_intersect.
24292 * vr-values.cc (vr_values::update_value_range): Same.
24293 (vr_values::extract_range_for_var_from_comparison_expr): Same.
24294 (vr_values::extract_range_from_cond_expr): Rename union_ to
24295 legacy_verbose_union.
24296 (vr_values::extract_range_from_phi_node): Same.
24298 2022-04-29 Aldy Hernandez <aldyh@redhat.com>
24300 * gimple-ssa-evrp-analyze.cc
24301 (evrp_range_analyzer::set_ssa_range_info): Use *range_info methods
24303 * gimple-ssa-sprintf.cc (try_substitute_return_value): Same.
24304 * ipa-prop.cc (ipcp_update_vr): Same.
24305 * tree-inline.cc (remap_ssa_name): Same.
24306 * tree-ssa-copy.cc (fini_copy_prop): Same.
24307 * tree-ssa-math-opts.cc (optimize_spaceship): Same.
24308 * tree-ssa-phiopt.cc (replace_phi_edge_with_variable): Same.
24309 * tree-ssa-pre.cc (insert_into_preds_of_block): Same.
24310 * tree-ssa-sccvn.cc (eliminate_dom_walker::eliminate_stmt): Same.
24311 * tree-ssa-strlen.cc (set_strlen_range): Same.
24312 (strlen_pass::handle_builtin_string_cmp): Same.
24313 * tree-ssanames.cc (set_range_info): Make static.
24314 (duplicate_ssa_name_range_info): Make static and add a new variant
24315 calling the static.
24316 * tree-ssanames.h (set_range_info): Remove version taking wide ints.
24317 (duplicate_ssa_name_range_info): Remove version taking a
24318 range_info_def and replace with a version taking SSA names.
24319 * tree-vect-loop-manip.cc (vect_gen_vector_loop_niters): Use *range_info methods
24321 (vect_do_peeling): Same.
24322 * tree-vrp.cc (vrp_asserts::remove_range_assertions): Same.
24323 * vr-values.cc (simplify_truth_ops_using_ranges): Same.
24325 2022-04-29 Aldy Hernandez <aldyh@redhat.com>
24327 * value-range.h (irange::irange): Use set_undefined.
24329 2022-04-29 Aldy Hernandez <aldyh@redhat.com>
24331 * gimple-range-cache.h (non_null_ref::adjust_range): Do not use
24332 irange::intersect (wide_int, wide_int).
24333 * gimple-range-fold.cc (adjust_pointer_diff_expr): Same.
24334 (adjust_imagpart_expr): Same.
24335 * value-range.h (irange::intersect (wide_int, wide_int)): Make
24338 2022-04-29 Richard Biener <rguenther@suse.de>
24340 PR tree-optimization/104322
24341 * tree-vect-loop.cc (vectorizable_reduction): Remove dead code.
24343 2022-04-29 Richard Biener <rguenther@suse.de>
24345 PR middle-end/105376
24346 * tree.cc (build_real): Special case dconst* arguments
24347 for decimal floating point types.
24349 2022-04-28 Sebastian Huber <sebastian.huber@embedded-brains.de>
24351 * doc/gcov.texi (Profiling and Test Coverage in Freestanding
24352 Environments): New section.
24354 2022-04-28 Sebastian Huber <sebastian.huber@embedded-brains.de>
24356 * doc/gcov-tool.texi: Document merge-stream subcommand.
24357 * doc/invoke.texi (fprofile-info-section): Mention merge-stream
24358 subcommand of gcov-tool.
24359 * gcov-tool.cc (gcov_profile_merge_stream): Declare.
24360 (print_merge_stream_usage_message): New.
24361 (merge_stream_usage): Likewise.
24362 (do_merge_stream): Likewise.
24363 (print_usage): Call print_merge_stream_usage_message().
24364 (main): Call do_merge_stream() to execute merge-stream subcommand.
24366 2022-04-28 Sebastian Huber <sebastian.huber@embedded-brains.de>
24368 * gcov-io.cc (gcov_file_error): New enum.
24369 (gcov_var): Use gcov_file_error enum for the error member.
24370 (gcov_open): Use GCOV_FILE_NO_ERROR.
24371 (gcov_close): Use GCOV_FILE_WRITE_ERROR.
24372 (gcov_write): Likewise.
24373 (gcov_write_unsigned): Likewise.
24374 (gcov_write_string): Likewise.
24375 (gcov_read_bytes): Set error code if EOF is reached.
24376 (gcov_read_counter): Use GCOV_FILE_COUNTER_OVERFLOW.
24378 2022-04-28 Sebastian Huber <sebastian.huber@embedded-brains.de>
24380 * gcov-io.cc (GCOV_MODE_STDIN): Define.
24381 (gcov_position): For gcov-tool, return calculated position if file is
24383 (gcov_open): For gcov-tool, use stdin if filename is NULL.
24384 (gcov_close): For gcov-tool, do not close stdin.
24385 (gcov_read_bytes): For gcov-tool, update position if file is stdin.
24386 (gcov_sync): For gcov-tool, discard input if file is stdin.
24388 2022-04-28 Sebastian Huber <sebastian.huber@embedded-brains.de>
24390 * doc/invoke.texi (fprofile-info-section): Mention
24391 __gcov_filename_to_gcfn(). Use "freestanding" to match with C11
24392 standard language. Fix minor example code issues.
24393 * gcov-io.h (GCOV_FILENAME_MAGIC): Define and document.
24395 2022-04-28 Sebastian Huber <sebastian.huber@embedded-brains.de>
24397 * gcov-io.cc (gcov_seek): Make it static.
24398 * gcov-io.h (struct gcov_summary): Do not mention gcov_seek().
24400 2022-04-28 Sebastian Huber <sebastian.huber@embedded-brains.de>
24402 * gcov-tool.cc (gcov_do_dump): Add mode parameter.
24403 (gcov_output_files): Open files for reading and writing.
24405 2022-04-28 Sebastian Huber <sebastian.huber@embedded-brains.de>
24407 * gcov-io.cc (gcov_open): Always use the mode parameter.
24408 * gcov-io.h (gcov_open): Declare it unconditionally.
24410 2022-04-28 Sebastian Huber <sebastian.huber@embedded-brains.de>
24412 * gcov-tool.cc (gcov_profile_merge): Adjust return type.
24413 (profile_merge): Allow merging of directories which contain no profile
24416 2022-04-28 David Malcolm <dmalcolm@redhat.com>
24418 * doc/invoke.texi (-fdump-analyzer-feasibility): Mention the
24421 2022-04-28 Jakub Jelinek <jakub@redhat.com>
24423 * config/i386/i386-expand.cc (ix86_expand_int_movcc): Create a
24424 temporary only if out overlaps compare_op, not when it overlaps
24427 2022-04-28 Jakub Jelinek <jakub@redhat.com>
24430 * cgraph.cc (cgraph_node::verify_node): Don't verify
24431 semantic_interposition flag against
24432 opt_for_fn (decl, flag_semantic_interposition) for aliases in lto1.
24434 2022-04-28 Jakub Jelinek <jakub@redhat.com>
24437 * config/i386/i386.cc (ix86_gimplify_va_arg): Mark va_arg_tmp
24438 temporary TREE_ADDRESSABLE before trying to gimplify ADDR_EXPR
24441 2022-04-28 Jonathan Wakely <jwakely@redhat.com>
24443 * doc/install.texi (Configuration): Remove misleading text
24444 around LE PowerPC Linux multilibs.
24446 2022-04-28 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
24449 * doc/install.texi (Tools/packages necessary for building GCC)
24450 (GDC): Document libphobos requirement.
24451 (Host/target specific installation notes for GCC, *-*-solaris2*):
24452 Document libphobos and GDC specifics.
24454 2022-04-28 Richard Biener <rguenther@suse.de>
24456 PR tree-optimization/105219
24457 * tree-vect-loop.cc (vect_transform_loop): Disable
24458 special code narrowing the vectorized epilogue max
24459 iterations when peeling for alignment or gaps was in effect.
24461 2022-04-28 Xi Ruoyao <xry111@mengyan1223.wang>
24463 * config/loongarch/loongarch.cc
24464 (loongarch_flatten_aggregate_field): Ignore empty fields for
24467 2022-04-27 Lulu Cheng <chenglulu@loongson.cn>
24469 * config/loongarch/loongarch.md: Add fdiv define_expand template,
24470 then generate floating-point division and floating-point reciprocal
24473 2022-04-27 Lulu Cheng <chenglulu@loongson.cn>
24475 * config/loongarch/loongarch.md: Add '(clobber (mem:BLK (scratch)))'
24476 to PLV instruction templates.
24478 2022-04-27 Richard Biener <rguenther@suse.de>
24480 PR middle-end/104492
24481 * gimple-ssa-warn-access.cc
24482 (pass_waccess::warn_invalid_pointer): Exclude equality compare
24483 diagnostics for all kind of invalidations.
24484 (pass_waccess::check_dangling_uses): Fix post-dominator query.
24485 (pass_waccess::check_pointer_uses): Likewise.
24487 2022-04-27 Andreas Krebbel <krebbel@linux.ibm.com>
24490 * config/s390/s390-protos.h (s390_function_arg_vector): Remove
24492 * config/s390/s390.cc (s390_single_field_struct_p): New function.
24493 (s390_function_arg_vector): Invoke s390_single_field_struct_p.
24494 (s390_function_arg_float): Likewise.
24496 2022-04-27 Jakub Jelinek <jakub@redhat.com>
24498 PR sanitizer/105396
24499 * asan.cc (asan_redzone_buffer::emit_redzone_byte): Handle the case
24500 where offset is bigger than off but smaller than m_prev_offset + 32
24501 bits by pushing one or more 0 bytes. Sink the
24502 m_shadow_bytes.safe_push (value); flush_if_full (); statements from
24503 all cases to the end of the function.
24505 2022-04-27 Kewen Lin <linkw@linux.ibm.com>
24508 * config/rs6000/rs6000-builtins.def (NEG_V2DI): Move to [power8-vector]
24511 2022-04-26 Thomas Schwinge <thomas@codesourcery.com>
24513 * config/gcn/gcn.cc (gcn_print_lds_decl): Make "gang-private
24514 data-share memory exhausted" error more verbose.
24516 2022-04-26 Martin Liska <mliska@suse.cz>
24519 * lto-wrapper.cc (print_lto_docs_link): Use global_dc.
24520 (run_gcc): Parse OPT_fdiagnostics_urls_.
24521 (main): Initialize global_dc.
24523 2022-04-26 Jakub Jelinek <jakub@redhat.com>
24525 PR rtl-optimization/105314
24526 * ifcvt.cc (noce_try_store_flag_mask): Don't require that the non-zero
24527 operand is equal to if_info->x, instead use the non-zero operand
24528 as one of the operands of AND with if_info->x as target.
24530 2022-04-26 Jakub Jelinek <jakub@redhat.com>
24532 PR tree-optimization/105374
24533 * tree-ssa-reassoc.cc (eliminate_redundant_comparison): Punt if
24534 !fold_convertible_p rather than assuming fold_convert must succeed.
24536 2022-04-26 Jakub Jelinek <jakub@redhat.com>
24539 * config/i386/i386.cc (ix86_veclibabi_svml, ix86_veclibabi_acml): Pass
24540 el_mode == DFmode ? double_type_node : float_type_node instead of
24541 TREE_TYPE (type_in) as first arguments to mathfn_built_in.
24543 2022-04-25 David Malcolm <dmalcolm@redhat.com>
24546 * gimple-fold.cc (gimple_fold_builtin_memory_op): Explicitly set
24547 the location of new_stmt in all places that don't already set it,
24548 whether explicitly, or via a call to gsi_replace.
24550 2022-04-25 Paul A. Clarke <pc@us.ibm.com>
24552 * doc/extend.texi (Other Builtins): Correct reference to 'modff'.
24554 2022-04-25 Andrew MacLeod <amacleod@redhat.com>
24556 PR tree-optimization/105276
24557 * gimple-range.cc (gimple_ranger::prefill_stmt_dependencies): Include
24558 existing global range with calculated value.
24560 2022-04-25 Richard Biener <rguenther@suse.de>
24562 PR tree-optimization/105368
24563 * tree-ssa-math-opts.cc (powi_cost): Use absu_hwi.
24565 2022-04-25 Richard Biener <rguenther@suse.de>
24567 PR tree-optimization/100810
24568 * tree-ssa-loop-ivopts.cc (struct iv_cand): Add involves_undefs flag.
24569 (find_ssa_undef): New function.
24570 (add_candidate_1): Avoid adding derived candidates with
24571 undefined SSA names and mark the original ones.
24572 (determine_group_iv_cost_generic): Reject rewriting
24573 uses with a different IV when that involves undefined SSA names.
24575 2022-04-25 Steven G. Kargl <kargl@gcc.gnu.org>
24578 * config/freebsd.h: Define TARGET_LIBC_HAS_FUNCTION to be
24579 bsd_libc_has_function.
24580 * targhooks.cc (bsd_libc_has_function): New function.
24581 Expand the supported math functions to inclue C99 libm.
24582 * targhooks.h (bsd_libc_has_function): New Prototype.
24584 2022-04-25 Richard Biener <rguenther@suse.de>
24586 PR rtl-optimization/105231
24587 * combine.cc (distribute_notes): Assert that a REG_EH_REGION
24588 with landing pad > 0 is from i3. Put any REG_EH_REGION note
24589 on i3 or drop it if the insn can not trap.
24590 (try_combine): Ensure that we can merge REG_EH_REGION notes
24591 with non-call exceptions. Ensure we are not splitting a
24592 trapping part of an insn with non-call exceptions when there
24593 is any REG_EH_REGION note to preserve.
24595 2022-04-25 Hongyu Wang <hongyu.wang@intel.com>
24598 * config/i386/avx512fintrin.h (_mm512_scalef_round_pd):
24599 Add parentheses for parameters and djust format.
24600 (_mm512_mask_scalef_round_pd): Ditto.
24601 (_mm512_maskz_scalef_round_pd): Ditto.
24602 (_mm512_scalef_round_ps): Ditto.
24603 (_mm512_mask_scalef_round_ps): Ditto.
24604 (_mm512_maskz_scalef_round_ps): Ditto.
24605 (_mm_scalef_round_sd): Use _mm_undefined_pd.
24606 (_mm_scalef_round_ss): Use _mm_undefined_ps.
24607 (_mm_mask_scalef_round_sd): New macro.
24608 (_mm_mask_scalef_round_ss): Ditto.
24609 (_mm_maskz_scalef_round_sd): Ditto.
24610 (_mm_maskz_scalef_round_ss): Ditto.
24612 2022-04-23 Jakub Jelinek <jakub@redhat.com>
24615 * config/i386/i386-expand.cc (ix86_expand_int_movcc): Handle
24616 op0 == cst1 ? op0 : op3 like op0 == cst1 ? cst1 : op3 for the non-cmov
24619 2022-04-22 Segher Boessenkool <segher@kernel.crashing.org>
24622 * config/rs6000/rs6000.md (pack<mode> for FMOVE128): New expander.
24623 (pack<mode> for FMOVE128): Rename and split the insn_and_split to...
24624 (pack<mode>_hard for FMOVE128): ... this...
24625 (pack<mode>_soft for FMOVE128): ... and this.
24627 2022-04-22 Paul A. Clarke <pc@us.ibm.com>
24629 * doc/extend.texi: Correct "This" to "These".
24631 2022-04-22 Jakub Jelinek <jakub@redhat.com>
24633 PR rtl-optimization/105333
24634 * rtlanal.cc (replace_rtx): Use simplify_subreg or
24635 simplify_unary_operation if CONST_SCALAR_INT_P rather than just
24638 2022-04-21 Segher Boessenkool <segher@kernel.crashing.org>
24642 * config/rs6000/rs6000.md (zero_extendqi<mode>2 for EXTQI): Disparage
24643 the "Z" alternatives in {l,st}{f,xs}iwzx.
24644 (zero_extendhi<mode>2 for EXTHI): Ditto.
24645 (zero_extendsi<mode>2 for EXTSI): Ditto.
24646 (*movsi_internal1): Ditto.
24647 (*mov<mode>_internal1 for QHI): Ditto.
24648 (movsd_hardfloat): Ditto.
24650 2022-04-21 Martin Liska <mliska@suse.cz>
24652 * configure.ac: Enable compressed debug sections for mold
24654 * configure: Regenerate.
24656 2022-04-21 Jakub Jelinek <jakub@redhat.com>
24659 * emit-rtl.cc (emit_copy_of_insn_after): Don't call mark_jump_label
24662 2022-04-20 Richard Biener <rguenther@suse.de>
24664 PR tree-optimization/104912
24665 * tree-vect-loop-manip.cc (vect_loop_versioning): Split
24666 the cost model check to a separate BB to make sure it is
24667 checked first and not combined with other version checks.
24669 2022-04-20 Richard Biener <rguenther@suse.de>
24671 PR tree-optimization/105312
24672 * gimple-isel.cc (gimple_expand_vec_cond_expr): Query both
24673 VCOND and VCONDU for EQ and NE.
24675 2022-04-20 Jan Hubicka <hubicka@ucw.cz>
24678 * ipa-modref-tree.cc (modref_access_node::closer_pair_p): Use
24679 poly_offset_int to avoid overflow.
24680 (modref_access_node::update2): likewise.
24682 2022-04-20 Jakub Jelinek <jakub@redhat.com>
24685 * cgraph.cc (cgraph_node::create): Set node->semantic_interposition
24686 to opt_for_fn (decl, flag_semantic_interposition).
24687 * cgraphclones.cc (cgraph_node::create_clone): Copy over
24688 semantic_interposition flag.
24690 2022-04-19 Sergei Trofimovich <siarheit@google.com>
24692 PR gcov-profile/105282
24693 * value-prof.cc (stream_out_histogram_value): Allow negative counts
24694 on HIST_TYPE_INDIR_CALL.
24696 2022-04-19 Jakub Jelinek <jakub@redhat.com>
24699 * config/sparc/sparc.cc (epilogue_renumber): If ORIGINAL_REGNO,
24700 use gen_raw_REG instead of gen_rtx_REG and copy over also
24701 ORIGINAL_REGNO. Use return 0; instead of /* fallthrough */.
24703 2022-04-19 Richard Biener <rguenther@suse.de>
24705 PR tree-optimization/104010
24706 PR tree-optimization/103941
24707 * tree-vect-slp.cc (vect_bb_slp_scalar_cost): When
24708 we run into stmts in patterns continue walking those
24709 for uses outside of the vectorized region instead of
24710 marking the lane live.
24712 2022-04-18 Hans-Peter Nilsson <hp@axis.com>
24714 * doc/install.texi <CRIS>: Remove references to removed websites and
24715 adjust for cris-*-elf being the only remaining toolchain.
24717 2022-04-18 Hans-Peter Nilsson <hp@axis.com>
24719 * doc/invoke.texi <CRIS>: Remove references to options for removed
24720 subtarget cris-axis-linux-gnu and tweak wording accordingly.
24722 2022-04-16 Gerald Pfeifer <gerald@pfeifer.com>
24724 * doc/install.texi (Specific): Adjust mingw-w64 download link.
24726 2022-04-15 Hongyu Wang <hongyu.wang@intel.com>
24728 * config/i386/smmintrin.h: Correct target pragma from sse4.1
24729 and sse4.2 to crc32 for crc32 intrinsics.
24731 2022-04-14 Indu Bhagat <indu.bhagat@oracle.com>
24734 * ctfc.cc (ctf_dvd_ignore_insert): New function.
24735 (ctf_dvd_ignore_lookup): Likewise.
24736 (ctf_add_variable): Keep track of non-defining decl DIEs.
24737 (new_ctf_container): Initialize the new hash-table.
24738 (ctfc_delete_container): Empty hash-table.
24739 * ctfc.h (struct ctf_container): Add new hash-table.
24740 (ctf_dvd_ignore_lookup): New declaration.
24741 (ctf_add_variable): Add additional argument.
24742 * ctfout.cc (ctf_dvd_preprocess_cb): Skip adding CTF variable
24743 record for non-defining decl for which a defining decl exists
24745 (ctf_preprocess): Defer updating the number of global objts
24747 (output_ctf_header): Use ctfc_vars_list_count as some CTF
24748 variables may not make it to the final output.
24749 (output_ctf_vars): Likewise.
24750 * dwarf2ctf.cc (gen_ctf_variable): Skip generating CTF variable
24751 if this is known to be a non-defining decl DIE.
24753 2022-04-14 Indu Bhagat <indu.bhagat@oracle.com>
24755 * ctfc.h (struct ctf_container): Introduce a new member.
24756 * ctfout.cc (ctf_list_add_ctf_vars): Use it instead of static
24759 2022-04-14 Jakub Jelinek <jakub@redhat.com>
24762 * simplify-rtx.cc (simplify_const_binary_operation): For shifts
24763 or rotates by VOIDmode constant integer shift count use word_mode
24764 for the operand if int_mode is narrower than word.
24766 2022-04-14 Robin Dapp <rdapp@linux.ibm.com>
24768 * config/s390/s390.cc (s390_get_sched_attrmask): Add z16.
24769 (s390_get_unit_mask): Likewise.
24770 (s390_is_fpd): Likewise.
24771 (s390_is_fxd): Likewise.
24772 * config/s390/s390.h (s390_tune_attr): Set max tune level to z16.
24773 * config/s390/s390.md (z900,z990,z9_109,z9_ec,z10,z196,zEC12,z13,z14,z15):
24775 (z900,z990,z9_109,z9_ec,z10,z196,zEC12,z13,z14,z15,z16):
24777 * config/s390/3931.md: New file.
24779 2022-04-13 Richard Sandiford <richard.sandiford@arm.com>
24781 PR tree-optimization/105254
24782 * config/aarch64/aarch64.cc
24783 (aarch64_vector_costs::determine_suggested_unroll_factor): Take a
24784 loop_vec_info as argument. Restrict the unroll factor to values
24785 that divide the VF.
24786 (aarch64_vector_costs::finish_cost): Update call accordingly.
24788 2022-04-13 Richard Biener <rguenther@suse.de>
24790 PR tree-optimization/105263
24791 * tree-ssa-reassoc.cc (try_special_add_to_ops): Do not consume
24792 negates in multiplication chains with DFP.
24794 2022-04-13 Jakub Jelinek <jakub@redhat.com>
24796 PR middle-end/105253
24797 * tree.cc (tree_builtin_call_types_compatible_p): If PROP_gimple,
24798 use useless_type_conversion_p checks instead of TYPE_MAIN_VARIANT
24799 comparisons or tree_nop_conversion_p checks.
24801 2022-04-13 Hongyu Wang <hongyu.wang@intel.com>
24804 * config/i386/i386-expand.cc (ix86_expand_cmpxchg_loop):
24805 Add missing set to target_val at pause label.
24807 2022-04-13 Jakub Jelinek <jakub@redhat.com>
24810 * attribs.cc (decl_attributes): Don't set
24811 DECL_FUNCTION_SPECIFIC_TARGET if target_option_default_node is
24814 2022-04-13 Richard Biener <rguenther@suse.de>
24816 PR tree-optimization/105250
24817 * fold-const.cc (fold_convertible_p): Revert
24818 r12-7979-geaaf77dd85c333, instead check for size equality
24819 of the vector types involved.
24821 2022-04-13 Richard Biener <rguenther@suse.de>
24824 2022-04-13 Richard Biener <rguenther@suse.de>
24826 PR tree-optimization/104912
24827 * tree-vect-loop-manip.cc (vect_loop_versioning): Split
24828 the cost model check to a separate BB to make sure it is
24829 checked first and not combined with other version checks.
24831 2022-04-13 Richard Biener <rguenther@suse.de>
24833 PR tree-optimization/104912
24834 * tree-vect-loop-manip.cc (vect_loop_versioning): Split
24835 the cost model check to a separate BB to make sure it is
24836 checked first and not combined with other version checks.
24838 2022-04-13 Jakub Jelinek <jakub@redhat.com>
24840 * tree-scalar-evolution.cc (expression_expensive_p): Fix a comment typo.
24842 2022-04-12 Antoni Boucher <bouanto@zoho.com>
24845 * reginfo.cc: New functions (clear_global_regs_cache,
24846 reginfo_cc_finalize) to avoid an issue where compiling the same
24847 code multiple times gives an error about assigning the same
24848 register to 2 global variables.
24849 * rtl.h: New function (reginfo_cc_finalize).
24850 * toplev.cc: Call it.
24852 2022-04-12 Antoni Boucher <bouanto@zoho.com>
24855 * toplev.cc: Call the new function tree_cc_finalize in
24857 * tree.cc: New functions (clear_nonstandard_integer_type_cache
24858 and tree_cc_finalize) to clear the cache of non-standard integer
24859 types to avoid having issues with some optimizations of
24860 bitcast where the SSA_NAME will have a size of a cached
24861 integer type that should have been invalidated, causing a
24862 comparison of integer constant to fail.
24863 * tree.h: New function (tree_cc_finalize).
24865 2022-04-12 Thomas Schwinge <thomas@codesourcery.com>
24868 * config/nvptx/nvptx.h (ASM_SPEC): Don't set.
24869 * config/nvptx/nvptx.opt (misa): Adjust comment.
24871 2022-04-12 Thomas Schwinge <thomas@codesourcery.com>
24874 2022-03-03 Tom de Vries <tdevries@suse.de>
24876 * config/nvptx/nvptx.h (ASM_SPEC): Add %{misa=sm_30:--no-verify}.
24878 2022-04-12 Thomas Schwinge <thomas@codesourcery.com>
24881 2022-03-31 Tom de Vries <tdevries@suse.de>
24883 * config/nvptx/nvptx.h (ASM_SPEC): Use "-m sm_35" for -misa=sm_30.
24885 2022-04-12 Richard Biener <rguenther@suse.de>
24888 * tree-ssa-dce.cc (mark_stmt_if_obviously_necessary): Do not
24889 include local escaped memory as obviously necessary stores.
24891 2022-04-12 Richard Biener <rguenther@suse.de>
24893 PR tree-optimization/105235
24894 * tree-ssa-math-opts.cc (execute_cse_conv_1): Clean EH and
24895 return whether the CFG changed.
24896 (execute_cse_sincos_1): Adjust.
24898 2022-04-12 Przemyslaw Wirkus <Przemyslaw.Wirkus@arm.com>
24901 * config/arm/t-aprofile (MULTI_ARCH_OPTS_A): Remove Armv9-a options.
24902 (MULTI_ARCH_DIRS_A): Remove Armv9-a diretories.
24903 (MULTILIB_REQUIRED): Don't require Armv9-a libraries.
24904 (MULTILIB_MATCHES): Treat Armv9-a as equivalent to Armv8-a.
24905 (MULTILIB_REUSE): Remove remap rules for Armv9-a.
24906 * config/arm/t-multilib (v9_a_nosimd_variants): Delete.
24907 (MULTILIB_MATCHES): Remove mappings for v9_a_nosimd_variants.
24909 2022-04-12 Richard Biener <rguenther@suse.de>
24911 PR tree-optimization/105232
24912 * tree.cc (component_ref_size): Bail out for too large
24913 or non-constant sizes.
24915 2022-04-12 Richard Biener <rguenther@suse.de>
24917 PR tree-optimization/105226
24918 * tree-vect-loop-manip.cc (vect_loop_versioning): Verify
24919 we can split the exit of an outer loop we choose to version.
24921 2022-04-12 Jakub Jelinek <jakub@redhat.com>
24923 * config/i386/i386-expand.cc (ix86_emit_i387_sinh, ix86_emit_i387_cosh,
24924 ix86_emit_i387_tanh, ix86_emit_i387_asinh, ix86_emit_i387_acosh,
24925 ix86_emit_i387_atanh, ix86_emit_i387_log1p, ix86_emit_i387_round,
24926 ix86_emit_swdivsf, ix86_emit_swsqrtsf,
24927 ix86_expand_atomic_fetch_op_loop, ix86_expand_cmpxchg_loop):
24929 * config/i386/i386.cc (warn_once_call_ms2sysv_xlogues): Likewise.
24931 2022-04-12 Jakub Jelinek <jakub@redhat.com>
24934 * config/i386/i386-expand.cc (ix86_emit_i387_log1p): Call
24935 do_pending_stack_adjust.
24937 2022-04-12 Jakub Jelinek <jakub@redhat.com>
24939 PR rtl-optimization/105211
24940 * builtins.cc (expand_builtin_int_roundingfn_2): If mathfn_built_in_1
24941 fails for TREE_TYPE (arg), retry it with
24942 TREE_VALUE (TYPE_ARG_TYPES (TREE_TYPE (fndecl))) and if even that
24943 fails, emit call normally.
24945 2022-04-12 Andreas Krebbel <krebbel@linux.ibm.com>
24947 * common/config/s390/s390-common.cc: Rename PF_ARCH14 to PF_Z16.
24948 * config.gcc: Add z16 as march/mtune switch.
24949 * config/s390/driver-native.cc (s390_host_detect_local_cpu):
24950 Recognize z16 with -march=native.
24951 * config/s390/s390-opts.h (enum processor_type): Rename
24952 PROCESSOR_ARCH14 to PROCESSOR_3931_Z16.
24953 * config/s390/s390.cc (PROCESSOR_ARCH14): Rename to ...
24954 (PROCESSOR_3931_Z16): ... throughout the file.
24955 (s390_processor processor_table): Add z16 as cpu string.
24956 * config/s390/s390.h (enum processor_flags): Rename PF_ARCH14 to
24958 (TARGET_CPU_ARCH14): Rename to ...
24959 (TARGET_CPU_Z16): ... this.
24960 (TARGET_CPU_ARCH14_P): Rename to ...
24961 (TARGET_CPU_Z16_P): ... this.
24962 (TARGET_ARCH14): Rename to ...
24963 (TARGET_Z16): ... this.
24964 (TARGET_ARCH14_P): Rename to ...
24965 (TARGET_Z16_P): ... this.
24966 * config/s390/s390.md (cpu_facility): Rename arch14 to z16 and
24967 check TARGET_Z16 instead of TARGET_ARCH14.
24968 * config/s390/s390.opt: Add z16 to processor_type.
24969 * doc/invoke.texi: Document z16 and arch14.
24971 2022-04-12 chenglulu <chenglulu@loongson.cn>
24973 * config/loongarch/loongarch.cc: Fix bug for
24974 tmpdir-g++.dg-struct-layout-1/t033.
24976 2022-04-11 Peter Bergner <bergner@linux.ibm.com>
24979 * config/rs6000/rs6000.cc (rs6000_sibcall_aix): Handle pcrel sibcalls
24980 to longcall functions.
24982 2022-04-11 Jason Merrill <jason@redhat.com>
24984 * ipa-free-lang-data.cc (free_lang_data_in_decl): Fix typos.
24986 2022-04-11 Segher Boessenkool <segher@kernel.crashing.org>
24990 * config/rs6000/rs6000.md (unpack<mode>_nodm): Add m,r,i alternative.
24992 2022-04-11 Jakub Jelinek <jakub@redhat.com>
24994 PR tree-optimization/105218
24995 * tree-ssa-phiopt.cc (value_replacement): If middle_bb has
24996 more than one predecessor or phi's bb more than 2 predecessors,
24997 reset phi result uses instead of adding a debug temp.
24999 2022-04-11 Kito Cheng <kito.cheng@sifive.com>
25002 * config.gcc: Pass -misa-spec to arch-canonicalize and
25003 multilib-generator.
25004 * config/riscv/arch-canonicalize: Adding -misa-spec option.
25005 (SUPPORTED_ISA_SPEC): New.
25006 (arch_canonicalize): New argument `isa_spec`.
25007 Handle multiple ISA spec versions.
25008 * config/riscv/multilib-generator: Adding -misa-spec option.
25010 2022-04-11 Kito Cheng <kito.cheng@sifive.com>
25012 * config/riscv/arch-canonicalize: Add TODO item.
25013 (IMPLIED_EXT): Sync.
25014 (arch_canonicalize): Checking until no change.
25016 2022-04-11 Tamar Christina <tamar.christina@arm.com>
25019 * tree-vect-stmts.cc (vectorizable_condition): Prevent cond swap when
25022 2022-04-11 Jason Merrill <jason@redhat.com>
25025 * pointer-query.cc (compute_objsize_r) [POINTER_PLUS_EXPR]: Require
25028 2022-04-11 Jakub Jelinek <jakub@redhat.com>
25030 PR tree-optimization/104639
25031 * tree-ssa-phiopt.cc: Include tree-ssa-propagate.h.
25032 (value_replacement): Optimize (x != cst1 ? x : cst2) != cst3
25035 2022-04-11 Jeff Law <jeffreyalaw@gmail.com>
25037 * config/bfin/bfin.md (rol_one): Fix pattern to indicate the
25038 sign bit of the source ends up in CC.
25040 2022-04-09 Jan Hubicka <hubicka@ucw.cz>
25043 * cgraphunit.cc (cgraph_node::analyze): update semantic_interposition
25046 2022-04-09 Jan Hubicka <hubicka@ucw.cz>
25048 * ipa-modref.cc (ipa_merge_modref_summary_after_inlining): Propagate
25049 nondeterministic and side_effects flags.
25051 2022-04-08 Andre Vieira <andre.simoesdiasvieira@arm.com>
25054 * config.gcc: Shift ext_mask by TARGET_CPU_NBITS.
25055 * config/aarch64/aarch64.h (TARGET_CPU_NBITS): New macro.
25056 (TARGET_CPU_MASK): Likewise.
25057 (TARGET_CPU_DEFAULT): Use TARGET_CPU_NBITS.
25058 * config/aarch64/aarch64.cc (aarch64_get_tune_cpu): Use TARGET_CPU_MASK.
25059 (aarch64_get_arch): Likewise.
25060 (aarch64_override_options): Use TARGET_CPU_NBITS.
25062 2022-04-08 Richard Biener <rguenther@suse.de>
25064 PR tree-optimization/105198
25065 * tree-predcom.cc (find_looparound_phi): Check whether
25066 the found memory location of the entry value is clobbered
25067 inbetween the value we want to use and loop entry.
25069 2022-04-08 Jakub Jelinek <jakub@redhat.com>
25071 PR tree-optimization/105189
25072 * fold-const.cc (make_range_step): Fix up handling of
25073 (unsigned) x +[low, -] ranges for signed x if low fits into
25076 2022-04-08 Richard Biener <rguenther@suse.de>
25078 PR tree-optimization/105175
25079 * tree-vect-stmts.cc (vectorizable_operation): Suppress
25080 -Wvector-operation-performance if using emulated vectors.
25081 * tree-vect-generic.cc (expand_vector_piecewise): Do not diagnose
25082 -Wvector-operation-performance when suppressed.
25083 (expand_vector_parallel): Likewise.
25084 (expand_vector_comparison): Likewise.
25085 (expand_vector_condition): Likewise.
25086 (lower_vec_perm): Likewise.
25087 (expand_vector_conversion): Likewise.
25089 2022-04-07 Tamar Christina <tamar.christina@arm.com>
25092 * config/aarch64/aarch64-builtins.cc (handle_arm_acle_h): New.
25093 (aarch64_general_init_builtins): Move LS64 code.
25094 * config/aarch64/aarch64-c.cc (aarch64_pragma_aarch64): Support
25096 * config/aarch64/aarch64-protos.h (handle_arm_acle_h): New.
25097 * config/aarch64/arm_acle.h: Add pragma GCC aarch64 "arm_acle.h".
25099 2022-04-07 Richard Biener <rguenther@suse.de>
25100 Jan Hubicka <hubicka@ucw.cz>
25103 * tree-ssa-alias.h (ptr_deref_may_alias_global_p,
25104 ref_may_alias_global_p, ref_may_alias_global_p,
25105 stmt_may_clobber_global_p, pt_solution_includes_global): Add
25106 bool parameters indicating whether escaped locals should be
25108 * tree-ssa-structalias.cc (pt_solution_includes_global):
25109 When the new escaped_nonlocal_p flag is true also consider
25110 pt->vars_contains_escaped.
25111 * tree-ssa-alias.cc (ptr_deref_may_alias_global_p):
25112 Pass down new escaped_nonlocal_p flag.
25113 (ref_may_alias_global_p): Likewise.
25114 (stmt_may_clobber_global_p): Likewise.
25115 (ref_may_alias_global_p_1): Likewise. For decls also
25116 query the escaped solution if true.
25117 (ref_may_access_global_memory_p): Remove.
25118 (modref_may_conflict): Use ref_may_alias_global_p with
25119 escaped locals considered global.
25120 (ref_maybe_used_by_stmt_p): Adjust.
25121 * ipa-fnsummary.cc (points_to_local_or_readonly_memory_p):
25123 * tree-ssa-dse.cc (dse_classify_store): Likewise.
25124 * trans-mem.cc (thread_private_new_memory): Likewise, but
25125 consider escaped locals global.
25126 * tree-ssa-dce.cc (mark_stmt_if_obviously_necessary): Likewise.
25128 2022-04-07 Richard Biener <rguenther@suse.de>
25130 PR tree-optimization/105185
25131 * tree-ssa-sccvn.cc (visit_reference_op_call): Simplify
25132 modref query again.
25134 2022-04-07 Tamar Christina <tamar.christina@arm.com>
25137 * config/aarch64/aarch64-simd.md
25138 (aarch64_reduc_plus_internal<mode>): Fix RTL and rename to...
25139 (reduc_plus_scal_<mode>): ... This.
25140 (reduc_plus_scal_v4sf): Moved.
25141 (aarch64_reduc_plus_internalv2si): Fix RTL and rename to...
25142 (reduc_plus_scal_v2si): ... This.
25144 2022-04-07 Jakub Jelinek <jakub@redhat.com>
25146 PR tree-optimization/102586
25147 * langhooks.h (struct lang_hooks_for_types): Add classtype_as_base
25149 * langhooks-def.h (LANG_HOOKS_CLASSTYPE_AS_BASE): Define.
25150 (LANG_HOOKS_FOR_TYPES_INITIALIZER): Add it.
25151 * gimple-fold.cc (clear_padding_type): Use ftype instead of
25152 TREE_TYPE (field) some more. For artificial FIELD_DECLs without
25153 name try the lang_hooks.types.classtype_as_base langhook and
25154 if it returns non-NULL, use that instead of ftype for recursive call.
25156 2022-04-07 Jakub Jelinek <jakub@redhat.com>
25158 PR tree-optimization/105150
25159 * tree.cc (tree_builtin_call_types_compatible_p): New function.
25160 (get_call_combined_fn): Use it.
25162 2022-04-07 Richard Biener <rguenther@suse.de>
25164 PR middle-end/105165
25165 * tree-complex.cc (expand_complex_asm): Sorry for asm goto
25168 2022-04-07 liuhongt <hongtao.liu@intel.com>
25170 * config/i386/sse.md (<sse2_avx2>_andnot<mode>3_mask):
25172 (<sse>_andnot<mode>3<mask_name>): Disable V*HFmode patterns
25174 (<code><mode>3<mask_name>): Ditto.
25175 (*<code><mode>3<mask_name>): Ditto.
25176 (VFB_128_256): Adjust condition of V8HF/V16HFmode according to
25181 2022-04-06 Jakub Jelinek <jakub@redhat.com>
25183 PR rtl-optimization/104985
25184 * combine.cc (struct undo): Add where.regno member.
25185 (do_SUBST_MODE): Rename to ...
25186 (subst_mode): ... this. Change first argument from rtx * into int,
25187 operate on regno_reg_rtx[regno] and save regno into where.regno.
25188 (SUBST_MODE): Remove.
25189 (try_combine): Use subst_mode instead of SUBST_MODE, change first
25190 argument from regno_reg_rtx[whatever] to whatever. For UNDO_MODE, use
25191 regno_reg_rtx[undo->where.regno] instead of *undo->where.r.
25192 (undo_to_marker): For UNDO_MODE, use regno_reg_rtx[undo->where.regno]
25193 instead of *undo->where.r.
25194 (simplify_set): Use subst_mode instead of SUBST_MODE, change first
25195 argument from regno_reg_rtx[whatever] to whatever.
25197 2022-04-06 Jakub Jelinek <jakub@redhat.com>
25200 * config/sh/sh.opt (mdiv=): Add Save.
25202 2022-04-06 Martin Liska <mliska@suse.cz>
25205 * common.opt: Document properly based on what it does.
25206 * gcc.cc (display_help): Unify with what we have in common.opt.
25207 * opts.cc (common_handle_option): Do not print undocumented
25210 2022-04-06 Xi Ruoyao <xry111@mengyan1223.wang>
25212 * config/mips/mips.cc (mips_fpr_return_fields): Ignore
25213 cxx17_empty_base_field_p fields and set an indicator.
25214 (mips_return_in_msb): Adjust for mips_fpr_return_fields change.
25215 (mips_function_value_1): Inform psABI change about C++17 empty
25218 2022-04-06 Jakub Jelinek <jakub@redhat.com>
25220 PR tree-optimization/105150
25221 * gimple.cc (gimple_builtin_call_types_compatible_p): Use
25222 builtin_decl_explicit here...
25223 (gimple_call_builtin_p, gimple_call_combined_fn): ... rather than
25226 2022-04-06 Richard Biener <rguenther@suse.de>
25228 PR tree-optimization/105173
25229 * tree-ssa-reassoc.cc (find_insert_point): Get extra
25230 insert_before output argument and compute it.
25231 (insert_stmt_before_use): Adjust.
25232 (rewrite_expr_tree): Likewise.
25234 2022-04-06 Richard Biener <rguenther@suse.de>
25237 * ipa-modref-tree.cc (modref_access_node::get_ao_ref ): Bail
25238 out for non-pointer arguments.
25240 2022-04-06 Richard Biener <rguenther@suse.de>
25242 PR tree-optimization/105163
25243 * tree-ssa-reassoc.cc (repropagate_negates): Avoid propagating
25246 2022-04-06 Jakub Jelinek <jakub@redhat.com>
25248 PR tree-optimization/105150
25249 * gimple.cc (gimple_call_builtin_p, gimple_call_combined_fn):
25250 For BUILT_IN_NORMAL calls, call gimple_builtin_call_types_compatible_p
25251 preferrably on builtin_decl_explicit decl rather than fndecl.
25252 * tree-ssa-strlen.cc (valid_builtin_call): Don't call
25253 gimple_builtin_call_types_compatible_p here.
25255 2022-04-06 Richard Sandiford <richard.sandiford@arm.com>
25257 PR tree-optimization/103761
25258 * tree-vect-stmts.cc (check_load_store_for_partial_vectors): Replace
25259 the ncopies parameter with an slp_node parameter. Calculate the
25260 number of vectors based on it and vectype. Rename lambda to
25261 group_memory_nvectors.
25262 (vectorizable_store, vectorizable_load): Update calls accordingly.
25264 2022-04-06 Martin Liska <mliska@suse.cz>
25266 * doc/invoke.texi: Document it.
25268 2022-04-06 Richard Biener <rguenther@suse.de>
25270 PR tree-optimization/105148
25271 * tree-ssa-loop-ivopts.cc (idx_record_use): Walk raw operands
25272 2 and 3 of ARRAY_REFs.
25274 2022-04-06 Roger Sayle <roger@nextmovesoftware.com>
25276 * config/i386/sse.md (ANDNOT_MODE): New mode iterator for TF and V1TI.
25277 (*andnottf3): Replace with...
25278 (*andnot<mode>3): New define_insn using ANDNOT_MODE.
25280 2022-04-06 Richard Biener <rguenther@suse.de>
25282 PR tree-optimization/105142
25283 * gimple-fold.h (maybe_fold_and_comparisons): Add defaulted
25284 basic-block parameter.
25285 (maybe_fold_or_comparisons): Likewise.
25286 * gimple-fold.cc (follow_outer_ssa_edges): New.
25287 (maybe_fold_comparisons_from_match_pd): Use follow_outer_ssa_edges
25288 when an outer condition basic-block is specified.
25289 (and_comparisons_1, and_var_with_comparison,
25290 and_var_with_comparison_1, or_comparisons_1,
25291 or_var_with_comparison, or_var_with_comparison_1): Receive and pass
25292 down the outer condition basic-block.
25293 * tree-ssa-ifcombine.cc (ifcombine_ifandif): Pass down the
25294 basic-block of the outer condition.
25296 2022-04-06 Kewen Lin <linkw@linux.ibm.com>
25299 * config/rs6000/rs6000.cc (rs6000_maybe_emit_maxc_minc): Support more
25300 comparison codes UNLT/UNLE/UNGT/UNGE.
25302 2022-04-05 David Malcolm <dmalcolm@redhat.com>
25304 * doc/extend.texi (Common Function Attributes): Document that
25305 'access' does not imply 'nonnull'.
25307 2022-04-05 Uroš Bizjak <ubizjak@gmail.com>
25310 * config/i386/mmx.md (*movv2qi_internal):
25311 Change insn mode of alternative 5 to HF for TARGET_AVX512FP16.
25313 2022-04-05 Richard Sandiford <richard.sandiford@arm.com>
25315 * config/aarch64/aarch64.md (aarch64_cpymemdi): Turn into a
25316 define_expand and turn operands 0 and 1 from REGs to MEMs.
25317 (*aarch64_cpymemdi): New pattern.
25318 (aarch64_setmemdi): Turn into a define_expand and turn operand 0
25319 from a REG to a MEM.
25320 (*aarch64_setmemdi): New pattern.
25321 * config/aarch64/aarch64.cc (aarch64_expand_cpymem_mops): Use
25322 copy_to_mode_reg on all three registers. Replace the original
25323 MEM addresses rather than creating wild reads and writes.
25324 (aarch64_expand_setmem_mops): Likewise for the size and for the
25325 destination memory and address.
25327 2022-04-05 Richard Sandiford <richard.sandiford@arm.com>
25330 * config/aarch64/aarch64-protos.h (aarch64_simd_switcher): New class.
25331 * config/aarch64/aarch64-sve-builtins.h (sve_switcher): Inherit
25332 from aarch64_simd_switcher.
25333 * config/aarch64/aarch64-builtins.cc (aarch64_simd_tuple_modes):
25335 (aarch64_lookup_simd_builtin_type): Use it instead of TYPE_MODE.
25336 (register_tuple_type): Add more asserts. Expect the alignment
25337 of the structure to be subject to flag_pack_struct and
25338 maximum_field_alignment. Set aarch64_simd_tuple_modes.
25339 (aarch64_simd_switcher::aarch64_simd_switcher): New function.
25340 (aarch64_simd_switcher::~aarch64_simd_switcher): Likewise.
25341 (handle_arm_neon_h): Hold an aarch64_simd_switcher throughout.
25342 (aarch64_general_init_builtins): Hold an aarch64_simd_switcher
25343 while calling aarch64_init_simd_builtins.
25344 * config/aarch64/aarch64-sve-builtins.cc (sve_switcher::sve_switcher)
25345 (sve_switcher::~sve_switcher): Remove code now performed by
25346 aarch64_simd_switcher.
25348 2022-04-05 Richard Sandiford <richard.sandiford@arm.com>
25351 * config/aarch64/aarch64-sve-builtins.cc
25352 (function_resolver::infer_vector_or_tuple_type): Use error_n
25353 for "%d vectors" messages.
25355 2022-04-05 Chung-Lin Tang <cltang@codesourcery.com>
25357 * omp-low.cc (lower_omp_target): Use outer context looked-up 'var' as
25358 argument to lang_hooks.decls.omp_array_data, instead of 'ovar' from
25361 2022-04-05 Richard Biener <rguenther@suse.de>
25364 * passes.def (pass_walloca): Move early instance into
25365 pass_build_ssa_passes to make SSA form available.
25367 2022-04-05 liuhongt <hongtao.liu@intel.com>
25370 * config/i386/i386.cc (ix86_split_stlf_stall_load): New
25372 (ix86_reorg): Call ix86_split_stlf_stall_load.
25373 * config/i386/i386.opt (-param=x86-stlf-window-ninsns=): New
25376 2022-04-05 Alexandre Oliva <oliva@adacore.com>
25378 * targhooks.cc (default_zero_call_used_regs): Attempt to group
25379 regs that the target refuses to use in their natural modes.
25380 (zcur_select_mode_rtx): New.
25381 * regs.h (struct target_regs): Add x_hard_regno_max_nregs.
25382 (hard_regno_max_nregs): Define.
25383 * reginfo.cc (init_reg_modes_target): Set hard_regno_max_nregs.
25385 2022-04-04 Alex Coplan <alex.coplan@arm.com>
25387 * doc/match-and-simplify.texi: Fix typos.
25389 2022-04-04 Jakub Jelinek <jakub@redhat.com>
25392 * config/aarch64/t-aarch64 (s-aarch64-tune-md): Do move-if-change
25393 only if configured with --enable-maintainer-mode, otherwise compare
25394 tmp-aarch64-tune.md with $(srcdir)/config/aarch64/aarch64-tune.md and
25395 if they differ, emit a message and fail.
25397 2022-04-04 Jakub Jelinek <jakub@redhat.com>
25400 * config/aarch64/t-aarch64 (s-mddeps): Depend on s-aarch64-tune-md.
25401 * config/aarch64/aarch64-tune.md: Regenerated.
25403 2022-04-04 Richard Biener <rguenther@suse.de>
25405 PR tree-optimization/105132
25406 * tree-vect-stmts.cc (vectorizable_operation): Check that
25407 the input vectors have the same number of elements.
25409 2022-04-04 Richard Biener <rguenther@suse.de>
25411 PR middle-end/105140
25412 * fold-const.cc (fold_convertible_p): Allow a TYPE_P arg.
25414 2022-04-03 Jeff Law <jeffreyalaw@gmail.com>
25417 * config/iq2000/iq2000.md (bbi): New attribute, default to no.
25418 (delay slot descripts): Use different delay slot description when
25419 the insn as the "bbi" attribute.
25420 (bbi, bbin patterns): Set the bbi attribute to yes.
25422 2022-04-03 Jakub Jelinek <jakub@redhat.com>
25425 * config/i386/i386-expand.cc (ix86_expand_vector_init_general): Avoid
25426 using word as target for expand_simple_binop when doing ASHIFT and
25429 2022-04-02 Xi Ruoyao <xry111@mengyan1223.wang>
25431 * config/mips/mips.cc (mips_function_arg): Check if DECL_SIZE is
25432 NULL before dereferencing it.
25434 2022-04-01 Qing Zhao <qing.zhao@oracle.com>
25436 * config/i386/i386.cc (zero_all_st_registers): Return the value of
25438 (ix86_zero_call_used_regs): Update zeroed_hardregs set according to
25439 the return value of zero_all_st_registers.
25440 * doc/tm.texi: Update the documentation of TARGET_ZERO_CALL_USED_REGS.
25441 * function.cc (gen_call_used_regs_seq): Add an assertion.
25442 * target.def: Update the documentation of TARGET_ZERO_CALL_USED_REGS.
25444 2022-04-01 Xi Ruoyao <xry111@mengyan1223.wang>
25447 * config/mips/mips.cc (mips_function_arg): Ignore zero-width
25448 fields, and inform if it causes a psABI change.
25450 2022-04-01 Xi Ruoyao <xry111@mengyan1223.wang>
25453 * config/mips/mips.cc (mips_fpr_return_fields): Detect C++
25454 zero-width bit-fields and set up an indicator.
25455 (mips_return_in_msb): Adapt for mips_fpr_return_fields change.
25456 (mips_function_value_1): Diagnose when the presense of a C++
25457 zero-width bit-field changes function returning in GCC 12.
25459 2022-04-01 Jakub Jelinek <jakub@redhat.com>
25461 PR tree-optimization/104645
25462 * tree-ssa-phiopt.cc (value_replacement): If assign has
25463 CONVERT_EXPR_CODE_P rhs_code, treat it like a preparation
25464 statement with constant evaluation.
25466 2022-04-01 YunQiang Su <yunqiang.su@cipunited.com>
25468 * config/mips/mips.cc (mips_expand_prologue):
25469 IPL is 8bit for MCU ASE.
25471 2022-03-31 Bill Schmidt <wschmidt@linux.ibm.com>
25474 * config/rs6000/rs6000-builtins.def (MFFSL): Mark nosoft.
25475 (MTFSB0): Likewise.
25476 (MTFSB1): Likewise.
25477 (SET_FPSCR_RN): Likewise.
25478 (SET_FPSCR_DRN): Mark nosoft and no32bit.
25480 2022-03-31 Thomas Schwinge <thomas@codesourcery.com>
25482 * doc/options.texi (Option file format): Clarifications around
25483 option definition records' help texts.
25485 2022-03-31 Thomas Schwinge <thomas@codesourcery.com>
25487 * optc-gen.awk <END>: Fix "Multiple different help strings" error
25490 2022-03-31 Andre Vieira <andre.simoesdiasvieira@arm.com>
25492 * config/aarch64/aarch64.cc (aarch64_vector_costs): Define
25493 determine_suggested_unroll_factor and m_has_avg.
25494 (determine_suggested_unroll_factor): New function.
25495 (aarch64_vector_costs::add_stmt_cost): Check for a qualifying pattern
25496 to set m_nosve_pattern.
25497 (aarch64_vector_costs::finish_costs): Use
25498 determine_suggested_unroll_factor.
25499 * config/aarch64/aarch64.opt (aarch64-vect-unroll-limit): New.
25500 * doc/invoke.texi: (aarch64-vect-unroll-limit): Document new option.
25502 2022-03-31 Martin Jambor <mjambor@suse.cz>
25505 * ipa-prop.h (ipa_ancestor_jf_data): New flag keep_null;
25506 (ipa_get_jf_ancestor_keep_null): New function.
25507 * ipa-prop.cc (ipa_set_ancestor_jf): Initialize keep_null field of the
25509 (compute_complex_assign_jump_func): Pass false to keep_null
25510 parameter of ipa_set_ancestor_jf.
25511 (compute_complex_ancestor_jump_func): Pass true to keep_null
25512 parameter of ipa_set_ancestor_jf.
25513 (update_jump_functions_after_inlining): Carry over keep_null from the
25514 original ancestor jump-function or merge them.
25515 (ipa_write_jump_function): Stream keep_null flag.
25516 (ipa_read_jump_function): Likewise.
25517 (ipa_print_node_jump_functions_for_edge): Print the new flag.
25518 * ipa-cp.cc (class ipcp_bits_lattice): Make various getters const. New
25519 member function known_nonzero_p.
25520 (ipcp_bits_lattice::known_nonzero_p): New.
25521 (ipcp_bits_lattice::meet_with_1): New parameter drop_all_ones,
25523 (ipcp_bits_lattice::meet_with): Likewise.
25524 (propagate_bits_across_jump_function): Simplify. Pass true in
25525 drop_all_ones when it is necessary.
25526 (propagate_aggs_across_jump_function): Take care of keep_null
25528 (ipa_get_jf_ancestor_result): Propagate NULL accross keep_null
25531 2022-03-31 Martin Jambor <mjambor@suse.cz>
25534 * ipa-cp.cc (decide_whether_version_node): Skip scalar values
25535 which do not fit the known value_range.
25537 2022-03-31 Martin Jambor <mjambor@suse.cz>
25540 * ipa-prop.cc (propagate_controlled_uses): Add a LOAD reference
25541 always when an ADDR_EXPR constant is known to reach a load because
25542 of inlining, not just when removing an ADDR reference.
25544 2022-03-31 Richard Biener <rguenther@suse.de>
25546 PR tree-optimization/105109
25547 * tree-ssa.cc (execute_update_addresses_taken): Suppress
25548 diagnostics on the load of the other complex component.
25550 2022-03-31 Tom de Vries <tdevries@suse.de>
25552 * config/nvptx/nvptx.h (ASM_SPEC): Use "-m sm_35" for -misa=sm_30.
25554 2022-03-31 Richard Biener <rguenther@suse.de>
25556 PR rtl-optimization/105091
25557 * gimple-expr.cc (mark_addressable): Handle TARGET_MEM_REF
25560 2022-03-31 Richard Biener <rguenther@suse.de>
25563 2021-09-13 Richard Earnshaw <rearnsha@arm.com>
25566 * gimple-fold.c (gimple_fold_builtin_memory_op): Allow folding
25567 memcpy if the size is not more than MOVE_MAX * MOVE_RATIO.
25569 2022-03-31 Sebastian Huber <sebastian.huber@embedded-brains.de>
25571 * gcov-io.cc (gcov_read_string): Reword documentation comment.
25573 2022-03-30 Bill Schmidt <wschmidt@linux.ibm.com>
25575 * config/rs6000/rs6000-builtins.def (NEG_V16QI): Move to [altivec]
25577 (NEG_V4SF): Likewise.
25578 (NEG_V4SI): Likewise.
25579 (NEG_V8HI): Likewise.
25580 (NEG_V2DF): Move to [vsx] stanza.
25581 (NEG_V2DI): Likewise.
25583 2022-03-30 Vladimir N. Makarov <vmakarov@redhat.com>
25585 PR middle-end/105032
25586 * lra-assigns.cc (find_reload_regno_insns): Modify loop condition.
25588 2022-03-30 Tom de Vries <tdevries@suse.de>
25589 Tobias Burnus <tobias@codesourcery.com>
25591 * doc/invoke.texi (march): Document __PTX_SM__.
25592 (mptx): Document __PTX_ISA_VERSION_MAJOR__ and
25593 __PTX_ISA_VERSION_MINOR__.
25595 2022-03-30 Jakub Jelinek <jakub@redhat.com>
25597 PR sanitizer/105093
25598 * ubsan.cc (instrument_object_size): If t is equal to inner and
25599 is a decl other than global var, punt. When emitting call to
25600 UBSAN_OBJECT_SIZE ifn, make sure base is addressable.
25602 2022-03-30 Jakub Jelinek <jakub@redhat.com>
25604 PR tree-optimization/105094
25605 * gimple-ssa-store-merging.cc (mem_valid_for_store_merging): Punt if
25606 bitsize <= 0 rather than just == 0.
25608 2022-03-30 Tom de Vries <tdevries@suse.de>
25610 * doc/invoke.texi (misa, mptx): Update.
25611 (march, march-map): Add.
25613 2022-03-30 Thomas Schwinge <thomas@codesourcery.com>
25615 * opt-functions.awk (n_args): New function.
25616 (lang_enabled_by): Merge function into...
25617 * optc-gen.awk <END>: ... sole user here.
25618 Improve diagnostics.
25620 2022-03-29 Marek Polacek <polacek@redhat.com>
25621 Jakub Jelinek <jakub@redhat.com>
25623 PR middle-end/103597
25624 * gimplify.cc (collect_fallthrough_labels): Don't push UNUSED_LABEL_Ps
25625 into labels. Maybe set prev to the statement preceding UNUSED_LABEL_P.
25626 (gimplify_cond_expr): Set UNUSED_LABEL_P.
25627 * tree.h (UNUSED_LABEL_P): New.
25629 2022-03-29 Michael Meissner <meissner@linux.ibm.com>
25631 * config/rs6000/vsx.md (vsx_extract_<mode>): Allow destination to
25632 be any VSX register.
25634 2022-03-29 Richard Earnshaw <rearnsha@arm.com>
25637 * config/aarch64/aarch64.cc (aapcs_vfp_sub_candidate): Handle
25638 zero-sized bit-fields. Detect cases where a warning may be needed.
25639 (aarch64_vfp_is_call_or_return_candidate): Emit a note if a
25640 zero-sized bit-field has caused parameter passing to change.
25642 2022-03-29 Richard Earnshaw <rearnsha@arm.com>
25645 * config/arm/arm.cc (aapcs_vfp_sub_candidate): Handle zero-sized
25646 bit-fields. Detect cases where a warning may be needed.
25647 (aapcs_vfp_is_call_or_return_candidate): Emit a note if
25648 a zero-sized bit-field has caused parameter passing to change.
25650 2022-03-29 Richard Earnshaw <rearnsha@arm.com>
25653 * config/arm/arm.cc (arm_get_pcs_model): Disable selection of
25654 ARM_PCS_AAPCS_LOCAL.
25656 2022-03-29 Tom de Vries <tdevries@suse.de>
25659 * config/nvptx/nvptx-c.cc (nvptx_cpu_cpp_builtins): Emit
25660 __PTX_ISA_VERSION_MAJOR__ and __PTX_ISA_VERSION_MINOR__.
25661 * config/nvptx/nvptx.cc (ptx_version_to_number): New function.
25662 * config/nvptx/nvptx-protos.h (ptx_version_to_number): Declare.
25664 2022-03-29 Tom de Vries <tdevries@suse.de>
25666 * config/nvptx/nvptx.opt (m64): Update help text to reflect that it
25669 2022-03-29 Tom de Vries <tdevries@suse.de>
25672 * config/nvptx/nvptx.opt (march-map=*): Add aliases.
25674 2022-03-29 Jan Hubicka <hubicka@ucw.cz>
25676 * config/i386/i386-builtins.cc (ix86_vectorize_builtin_gather): Test
25677 TARGET_USE_GATHER_2PARTS and TARGET_USE_GATHER_4PARTS.
25678 * config/i386/i386.h (TARGET_USE_GATHER_2PARTS): New macro.
25679 (TARGET_USE_GATHER_4PARTS): New macro.
25680 * config/i386/x86-tune.def (X86_TUNE_USE_GATHER_2PARTS): New tune
25681 (X86_TUNE_USE_GATHER_4PARTS): New tune
25683 2022-03-29 Tom de Vries <tdevries@suse.de>
25685 * config/nvptx/nvptx.opt (march): Add alias of misa.
25687 2022-03-29 Chenghua Xu <xuchenghua@loongson.cn>
25688 Lulu Cheng <chenglulu@loongson.cn>
25690 * doc/install.texi: Add LoongArch options section.
25691 * doc/invoke.texi: Add LoongArch options section.
25692 * doc/md.texi: Add LoongArch options section.
25694 2022-03-29 Chenghua Xu <xuchenghua@loongson.cn>
25695 Lulu Cheng <chenglulu@loongson.cn>
25697 * config/loongarch/loongarch-c.cc
25699 2022-03-29 Chenghua Xu <xuchenghua@loongson.cn>
25700 Lulu Cheng <chenglulu@loongson.cn>
25702 * config/loongarch/larchintrin.h: New file.
25703 * config/loongarch/loongarch-builtins.cc: New file.
25705 2022-03-29 Chenghua Xu <xuchenghua@loongson.cn>
25706 Lulu Cheng <chenglulu@loongson.cn>
25708 * config/host-linux.cc: Add LoongArch support.
25709 * config/loongarch/loongarch-protos.h: New file.
25710 * config/loongarch/loongarch-tune.h: Likewise.
25711 * config/loongarch/loongarch.cc: Likewise.
25712 * config/loongarch/loongarch.h: Likewise.
25714 2022-03-29 Chenghua Xu <xuchenghua@loongson.cn>
25715 Lulu Cheng <chenglulu@loongson.cn>
25717 * config/loongarch/constraints.md: New file.
25718 * config/loongarch/generic.md: New file.
25719 * config/loongarch/la464.md: New file.
25720 * config/loongarch/loongarch-ftypes.def: New file.
25721 * config/loongarch/loongarch-modes.def: New file.
25722 * config/loongarch/loongarch.md: New file.
25723 * config/loongarch/predicates.md: New file.
25724 * config/loongarch/sync.md: New file.
25726 2022-03-29 Chenghua Xu <xuchenghua@loongson.cn>
25727 Lulu Cheng <chenglulu@loongson.cn>
25729 * configure: Regenerate file.
25731 2022-03-29 Chenghua Xu <xuchenghua@loongson.cn>
25732 Lulu Cheng <chenglulu@loongson.cn>
25734 * common/config/loongarch/loongarch-common.cc: New file.
25735 * config/loongarch/genopts/genstr.sh: New file.
25736 * config/loongarch/genopts/loongarch-strings: New file.
25737 * config/loongarch/genopts/loongarch.opt.in: New file.
25738 * config/loongarch/loongarch-str.h: New file.
25739 * config/loongarch/gnu-user.h: New file.
25740 * config/loongarch/linux.h: New file.
25741 * config/loongarch/loongarch-cpu.cc: New file.
25742 * config/loongarch/loongarch-cpu.h: New file.
25743 * config/loongarch/loongarch-def.c: New file.
25744 * config/loongarch/loongarch-def.h: New file.
25745 * config/loongarch/loongarch-driver.cc: New file.
25746 * config/loongarch/loongarch-driver.h: New file.
25747 * config/loongarch/loongarch-opts.cc: New file.
25748 * config/loongarch/loongarch-opts.h: New file.
25749 * config/loongarch/loongarch.opt: New file.
25750 * config/loongarch/t-linux: New file.
25751 * config/loongarch/t-loongarch: New file.
25752 * config.gcc: Add LoongArch support.
25753 * configure.ac: Add LoongArch support.
25755 2022-03-29 Thomas Schwinge <thomas@codesourcery.com>
25757 * opt-functions.awk (lang_enabled_by): Fix 'enabledby_negargs'
25760 2022-03-29 Richard Biener <rguenther@suse.de>
25762 PR tree-optimization/105080
25763 * tree-ssa-strlen.cc (printf_strlen_execute): Always init
25766 2022-03-28 Indu Bhagat <indu.bhagat@oracle.com>
25768 * ctfout.cc (ctf_preprocess): Use ctfc_get_num_ctf_vars instead.
25769 (output_ctf_vars): Likewise.
25771 2022-03-28 Jason Merrill <jason@redhat.com>
25774 * doc/extend.texi: Refer to __is_trivial instead of __is_pod.
25776 2022-03-28 H.J. Lu <hjl.tools@gmail.com>
25779 * config/i386/sse.md (*ssse3_pshufbv8qi3): Also replace "Yv" with
25782 2022-03-28 Tom de Vries <tdevries@suse.de>
25785 * config/nvptx/gen-opt.sh (ptx_isa): Improve help text.
25786 * config/nvptx/nvptx-gen.opt: Regenerate.
25787 * config/nvptx/nvptx.opt (misa, mptx, ptx_version): Improve help text.
25788 * config/nvptx/t-nvptx (s-nvptx-gen-opt): Add missing dependency on
25791 2022-03-28 David Malcolm <dmalcolm@redhat.com>
25794 * gimple-fold.cc (gimple_fold_builtin_memory_op): When optimizing
25795 to loads then stores, set the location of the new load stmt.
25797 2022-03-28 Richard Biener <rguenther@suse.de>
25799 PR tree-optimization/105070
25800 * tree-switch-conversion.h
25801 (bit_test_cluster::hoist_edge_and_branch_if_true): Add location
25803 * tree-switch-conversion.cc
25804 (bit_test_cluster::hoist_edge_and_branch_if_true): Annotate
25805 cond with location.
25806 (bit_test_cluster::emit): Annotate all generated expressions
25809 2022-03-28 Andre Vieira <andre.simoesdiasvieira@arm.com>
25811 * config/aarch64/aarch64-cores.def: Update Neoverse N2 core entry.
25813 2022-03-28 liuhongt <hongtao.liu@intel.com>
25816 * config/i386/sse.md (vec_set<mode>_0): Change attr "isa" of
25817 alternative 4 from sse4_noavx to noavx.
25819 2022-03-28 Jakub Jelinek <jakub@redhat.com>
25821 PR tree-optimization/105056
25822 * tree-predcom.cc (component::component): Initialize also comp_step.
25824 2022-03-27 H.J. Lu <hjl.tools@gmail.com>
25827 * config/i386/sse.md (*ssse3_pshufbv8qi3): Replace "Yv" with
25830 2022-03-26 Roger Sayle <roger@nextmovesoftware.com>
25832 PR middle-end/104885
25833 * calls.cc (mark_stack_region_used): Check that the region
25834 is within the allocated size of stack_usage_map.
25836 2022-03-26 Jakub Jelinek <jakub@redhat.com>
25838 PR rtl-optimization/103775
25839 * recog.cc (check_invalid_inc_dec): New function.
25840 (insn_invalid_p): Return 1 if REG_INC operand overlaps
25843 2022-03-26 H.J. Lu <hjl.tools@gmail.com>
25846 * config/i386/sse.md (loadiwkey): Replace "v" with "x".
25847 (aes<aesklvariant>u8): Likewise.
25849 2022-03-26 H.J. Lu <hjl.tools@gmail.com>
25852 * config/i386/sse.md (ssse3_ph<plusminus_mnemonic>wv4hi3):
25853 Replace "Yv" with "x".
25854 (ssse3_ph<plusminus_mnemonic>dv2si3): Likewise.
25855 (ssse3_psign<mode>3): Likewise.
25857 2022-03-26 Hans-Peter Nilsson <hp@axis.com>
25859 * reload.cc (find_reloads): Align comment with code where
25860 considering the intersection of register classes then tweaking the
25861 regclass for the current alternative or rejecting it.
25863 2022-03-25 Christophe Lyon <christohe.lyon@arm.com>
25867 2021-06-11 Christophe Lyon <christophe.lyon@linaro.org>
25869 * config/arm/mve.md (mve_vec_unpack<US>_lo_<mode>): Delete.
25870 (mve_vec_unpack<US>_hi_<mode>): Delete.
25871 (@mve_vec_pack_trunc_lo_<mode>): Delete.
25872 (mve_vmovntq_<supf><mode>): Remove '@' prefix.
25873 * config/arm/neon.md (vec_unpack<US>_hi_<mode>): Move back
25874 from vec-common.md.
25875 (vec_unpack<US>_lo_<mode>): Likewise.
25876 (vec_pack_trunc_<mode>): Rename from
25877 neon_quad_vec_pack_trunc_<mode>.
25878 * config/arm/vec-common.md (vec_unpack<US>_hi_<mode>): Delete.
25879 (vec_unpack<US>_lo_<mode>): Delete.
25880 (vec_pack_trunc_<mode>): Delete.
25882 2022-03-25 Vladimir N. Makarov <vmakarov@redhat.com>
25884 PR middle-end/104971
25885 * lra-lives.cc (process_bb_lives): Check hard_regs_live for hard
25886 regs to clear remove_p flag.
25888 2022-03-25 Richard Biener <rguenther@suse.de>
25890 PR tree-optimization/105053
25891 * tree-vect-loop.cc (vect_create_epilog_for_reduction): Pick
25892 the correct live-out stmt for a reduction chain.
25894 2022-03-25 Richard Biener <rguenther@suse.de>
25896 PR middle-end/105049
25897 * tree.cc (uniform_vector_p): Recurse for VECTOR_CST or
25898 CONSTRUCTOR first elements.
25900 2022-03-25 Tobias Burnus <tobias@codesourcery.com>
25903 * doc/invoke.texi (Static Analyzer Options): Move
25904 @ignore block after @gccoptlist's '}' for 'make pdf'.
25906 2022-03-25 David Malcolm <dmalcolm@redhat.com>
25909 * doc/invoke.texi (Static Analyzer Options): Add
25910 -fdump-analyzer-untracked.
25912 2022-03-25 Avinash Sonawane <rootkea@gmail.com>
25915 * doc/invoke.texi: Document that enabling taint analyzer
25916 checker disables some warnings from `-fanalyzer`.
25918 2022-03-24 Alexandre Oliva <oliva@adacore.com>
25921 * gimple-harden-conditionals.cc (detach_value): Keep temps
25924 2022-03-24 Alexandre Oliva <oliva@adacore.com>
25926 PR middle-end/104975
25927 * gimple-harden-conditionals.cc
25928 (pass_harden_compares::execute): Force split in case of
25931 2022-03-24 Jakub Jelinek <jakub@redhat.com>
25934 * fold-const.cc (operand_equal_p) <case COMPONENT_REF>: If either
25935 field0 or field1 is not a FIELD_DECL, return false.
25937 2022-03-24 Richard Biener <rguenther@suse.de>
25939 * tree-predcom.cc (chain::chain): Add CTOR.
25940 (component::component): Likewise.
25941 (pcom_worker::release_chain): Use delete.
25942 (release_components): Likewise.
25943 (pcom_worker::filter_suitable_components): Likewise.
25944 (pcom_worker::split_data_refs_to_components): Use new.
25945 (make_invariant_chain): Likewise.
25946 (make_rooted_chain): Likewise.
25947 (pcom_worker::combine_chains): Likewise.
25948 * tree-vect-loop.cc (vect_create_epilog_for_reduction):
25949 Make sure to release previously constructed scalar_results.
25950 * tree-vect-stmts.cc (vectorizable_load): Use auto_vec
25952 * vr-values.cc (simplify_using_ranges::~simplify_using_ranges):
25953 Release m_flag_set_edges.
25955 2022-03-24 Siddhesh Poyarekar <siddhesh@gotplt.org>
25957 PR tree-optimization/104970
25958 * tree-object-size.cc (parm_object_size): Restrict size
25959 computation scenarios to explicit access attributes.
25961 2022-03-24 Kewen Lin <linkw@linux.ibm.com>
25964 * config/rs6000/rs6000-c.cc (find_instance): Skip instances with null
25967 2022-03-23 Richard Biener <rguenther@suse.de>
25970 * gimple-fold.cc (gimple_fold_builtin_memory_op): Allow the
25971 use of movmisalign when either the source or destination
25972 decl is properly aligned.
25974 2022-03-23 Richard Biener <rguenther@suse.de>
25976 PR rtl-optimization/105028
25977 * ira-color.cc (form_threads_from_copies): Remove unnecessary
25978 copying of the sorted_copies tail.
25980 2022-03-23 Martin Liska <mliska@suse.cz>
25982 * config/rs6000/rs6000-c.cc (altivec_resolve_overloaded_builtin):
25984 * config/rs6000/rs6000.cc (rs6000_option_override_internal):
25985 Reword the error message.
25987 2022-03-23 liuhongt <hongtao.liu@intel.com>
25990 * config/i386/sse.md (ssePSmodelower): New.
25991 (*avx_cmp<mode>3_ltint_not): Force_reg operand before
25992 lowpart_subreg to avoid NULL_RTX.
25993 (<avx512>_fmaddc_<mode>_mask1<round_expand_name>,
25994 <avx512>_fcmaddc_<mode>_mask1<round_expand_name>,
25995 fma_<mode>_fmaddc_bcst, fma_<mode>_fcmaddc_bcst,
25996 <avx512>_<complexopname>_<mode>_mask<round_name>,
25997 avx512fp16_fcmaddcsh_v8hf_mask1<round_expand_name>,
25998 avx512fp16_fcmaddcsh_v8hf_mask3<round_expand_name>,
25999 avx512fp16_fmaddcsh_v8hf_mask3<round_expand_name>,
26000 avx512fp16_fmaddcsh_v8hf_mask3<round_expand_name>,
26001 float<floatunssuffix><mode>v4hf2,
26002 float<floatunssuffix>v2div2hf2,
26003 fix<fixunssuffix>_truncv4hf<mode>2,
26004 fix<fixunssuffix>_truncv2hfv2di2, extendv4hf<mode>2,
26006 trunc<mode>v4hf2,truncv2dfv2hf2,
26007 *avx512bw_permvar_truncv16siv16hi_1,
26008 *avx512bw_permvar_truncv16siv16hi_1_hf,
26009 *avx512f_permvar_truncv8siv8hi_1,
26010 *avx512f_permvar_truncv8siv8hi_1_hf,
26011 *avx512f_vpermvar_truncv8div8si_1,
26012 *avx512f_permvar_truncv32hiv32qi_1,
26013 *avx512f_permvar_truncv16hiv16qi_1,
26014 *avx512f_permvar_truncv4div4si_1,
26015 *avx512f_pshufb_truncv8hiv8qi_1,
26016 *avx512f_pshufb_truncv4siv4hi_1,
26017 *avx512f_pshufd_truncv2div2si_1,
26018 sdot_prod<mode>, avx2_pblend<ssemodesuffix>_1,
26019 ashrv2di3,ashrv2di3,usdot_prod<mode>): Ditto.
26021 2022-03-22 Tom de Vries <tdevries@suse.de>
26024 * config/nvptx/nvptx.md (define_insn "nvptx_uniform_warp_check"):
26025 Use % as register prefix.
26027 2022-03-22 Tom de Vries <tdevries@suse.de>
26029 * config/nvptx/nvptx.cc (nvptx_scalar_mode_supported_p)
26030 (nvptx_libgcc_floating_mode_supported_p): Only enable HFmode for
26033 2022-03-22 Tom de Vries <tdevries@suse.de>
26035 * config/nvptx/nvptx.opt (mexperimental): New option.
26037 2022-03-22 Tom de Vries <tdevries@suse.de>
26040 * config/nvptx/nvptx-protos.h (nvptx_asm_output_def_from_decls): Declare.
26041 * config/nvptx/nvptx.cc (write_fn_proto_1): Don't add function marker
26043 (SET_ASM_OP, NVPTX_ASM_OUTPUT_DEF): New macro def.
26044 (nvptx_asm_output_def_from_decls): New function.
26045 * config/nvptx/nvptx.h (ASM_OUTPUT_DEF): New macro def, define to
26046 gcc_unreachable ().
26047 (ASM_OUTPUT_DEF_FROM_DECLS): New macro def, define to
26048 nvptx_asm_output_def_from_decls.
26049 * config/nvptx/nvptx.opt (malias): New opt.
26051 2022-03-22 Tom de Vries <tdevries@suse.de>
26055 * config/nvptx/nvptx.md (define_expand "omp_simt_exit"): Emit warp
26056 sync (or uniform warp check for mptx < 6.0).
26058 2022-03-22 Richard Biener <rguenther@suse.de>
26060 PR tree-optimization/105012
26061 * tree-if-conv.cc (ifcvt_local_dce): Only call
26062 dse_classify_store when we have a VDEF.
26064 2022-03-22 Martin Liska <mliska@suse.cz>
26067 * config/nvptx/nvptx.cc (handle_ptx_version_option):
26068 Fix option wrapping in an error message.
26070 2022-03-22 Martin Liska <mliska@suse.cz>
26073 * config/rs6000/rs6000-c.cc (altivec_resolve_overloaded_builtin):
26074 Wrap const keyword.
26076 2022-03-22 Martin Liska <mliska@suse.cz>
26078 * config/v850/v850-c.cc (pop_data_area): Fix typo in pragma
26081 2022-03-22 Martin Liska <mliska@suse.cz>
26084 * config/rs6000/rs6000.cc (rs6000_option_override_internal):
26085 Use %qs instead of (%qs).
26087 2022-03-22 Martin Liska <mliska@suse.cz>
26090 * config/i386/i386-options.cc (ix86_option_override_internal):
26091 Use '%qs' instead of '(%qs)'.
26093 2022-03-22 Martin Liska <mliska@suse.cz>
26096 * config/aarch64/aarch64.cc (aarch64_handle_attr_arch):
26097 Use 'qs' and remove usage '(%qs)'.
26098 (aarch64_handle_attr_cpu): Likewise.
26099 (aarch64_handle_attr_tune): Likewise.
26100 (aarch64_handle_attr_isa_flags): Likewise.
26102 2022-03-22 Tamar Christina <tamar.christina@arm.com>
26103 Andre Vieira <andre.simoesdiasvieira@arm.com>
26105 * config/aarch64/aarch64.cc (neoversev1_regmove_cost): New tuning
26107 (neoversev1_tunings): Use neoversev1_regmove_cost and update store_int
26109 (neoverse512tvb_tunings): Likewise.
26111 2022-03-22 Tamar Christina <tamar.christina@arm.com>
26112 Andre Vieira <andre.simoesdiasvieira@arm.com>
26114 * config/aarch64/aarch64.cc (demeter_addrcost_table,
26115 demeter_regmove_cost, demeter_advsimd_vector_cost,
26116 demeter_sve_vector_cost, demeter_scalar_issue_info,
26117 demeter_advsimd_issue_info, demeter_sve_issue_info,
26118 demeter_vec_issue_info, demeter_vector_cost,
26119 demeter_tunings): New tuning structs.
26120 (aarch64_ve_op_count::rename_cycles_per_iter): Enable for demeter
26122 * config/aarch64/aarch64-cores.def: Add entry for demeter.
26123 * config/aarch64/aarch64-tune.md (tune): Add demeter to list.
26125 2022-03-22 Tamar Christina <tamar.christina@arm.com>
26126 Andre Vieira <andre.simoesdiasvieira@arm.com>
26128 * config/aarch64/aarch64-protos.h (struct cpu_memmov_cost): New struct.
26129 (struct tune_params): Change type of memmov_cost to use cpu_memmov_cost.
26130 * config/aarch64/aarch64.cc (aarch64_memory_move_cost): Update all
26131 tunings to use cpu_memmov_cost struct.
26133 2022-03-22 Tamar Christina <tamar.christina@arm.com>
26134 Andre Vieira <andre.simoesdiasvieira@arm.com>
26136 * config/aarch64/aarch64.cc (neoversen2_addrcost_table,
26137 neoversen2_regmove_cost, neoversen2_advsimd_vector_cost,
26138 neoversen2_sve_vector_cost, neoversen2_scalar_issue_info,
26139 neoversen2_advsimd_issue_info, neoversen2_sve_issue_info,
26140 neoversen2_vec_issue_info, neoversen2_tunings): New structs.
26141 (neoversen2_tunings): Use new structs and update tuning flags.
26142 (aarch64_vec_op_count::rename_cycles_per_iter): Enable for neoversen2
26145 2022-03-22 Andre Vieira <andre.simoesdiasvieira@arm.com>
26147 * config/aarch64/aarch64.h (AARCH64_FL_FOR_ARCH9): Add FP16 feature
26150 2022-03-22 liuhongt <hongtao.liu@intel.com>
26153 * config/i386/i386.md (*jcc_bt<mode>_mask): Extend the
26154 following splitter to reversed condition.
26156 2022-03-22 Jakub Jelinek <jakub@redhat.com>
26158 PR rtl-optimization/104989
26159 * calls.cc (expand_call): Don't set ECF_NORETURN in flags after
26160 sorry for passing too large argument, instead set sibcall_failure
26161 for pass == 0, or a new normal_failure flag otherwise. If
26162 normal_failure is set, don't assert all stack has been deallocated
26163 at the end and throw away the whole insn sequence.
26165 2022-03-22 Qian Jianhua <qianjh@cn.fujitsu.com>
26167 * print-tree.cc: Change array length
26169 2022-03-22 Hongyu Wang <hongyu.wang@intel.com>
26172 * config/i386/sse.md
26173 (avx512fp16_fmaddcsh_v8hf_mask1<round_expand_name):
26174 Use avx512f_movsf_mask instead of vmovaps or vblend, and
26175 force_reg before lowpart_subreg.
26176 (avx512fp16_fcmaddcsh_v8hf_mask1<round_expand_name): Likewise.
26178 2022-03-21 H.J. Lu <hjl.tools@gmail.com>
26181 * common/config/i386/i386-common.cc
26182 (OPTION_MASK_ISA2_GENERAL_REGS_ONLY_UNSET): Replace
26183 OPTION_MASK_ISA2_AVX512F_UNSET with OPTION_MASK_ISA2_SSE_UNSET.
26185 2022-03-21 H.J. Lu <hjl.tools@gmail.com>
26188 * common/config/i386/cpuinfo.h (get_available_features): Pass
26189 0x19 to __cpuid for bit_AESKLE. Enable FEATURE_AESKLE only if
26192 2022-03-21 Richard Sandiford <richard.sandiford@arm.com>
26194 PR middle-end/104869
26195 * rtl-ssa/accesses.h (clobber_group::prev_clobber): Declare.
26196 (clobber_group::next_clobber): Likewise.
26197 (def_lookup::prev_def): Rename to...
26198 (def_lookup::last_def_of_prev_group): ...this.
26199 (def_lookup::next_def): Rename to...
26200 (def_lookup::first_def_of_next_group): ...this.
26201 (def_lookup::matching_or_prev_def): Rename to...
26202 (def_lookup::matching_set_or_last_def_of_prev_group): ...this.
26203 (def_lookup::matching_or_next_def): Rename to...
26204 (def_lookup::matching_set_or_first_def_of_next_group): ...this.
26205 (def_lookup::prev_def): New function, taking the lookup insn as
26207 (def_lookup::next_def): Likewise.
26208 * rtl-ssa/member-fns.inl (def_lookup::prev_def): Rename to...
26209 (def_lookup::last_def_of_prev_group): ...this.
26210 (def_lookup::next_def): Rename to...
26211 (def_lookup::first_def_of_next_group): ...this.
26212 (def_lookup::matching_or_prev_def): Rename to...
26213 (def_lookup::matching_set_or_last_def_of_prev_group): ...this.
26214 (def_lookup::matching_or_next_def): Rename to...
26215 (def_lookup::matching_set_or_first_def_of_next_group): ...this.
26216 * rtl-ssa/movement.h (restrict_movement_for_dead_range): Update after
26218 * rtl-ssa/accesses.cc (clobber_group::prev_clobber): New function.
26219 (clobber_group::next_clobber): Likewise.
26220 (def_lookup::prev_def): Likewise.
26221 (def_lookup::next_def): Likewise.
26222 (function_info::make_use_available): Pass the lookup insn to
26223 def_lookup::prev_def and def_lookup::next_def.
26225 2022-03-21 Martin Liska <mliska@suse.cz>
26227 * doc/invoke.texi: Document min-pagesize parameter.
26229 2022-03-21 Richard Biener <rguenther@suse.de>
26231 * tree-ssa-loop-niter.cc (estimate_numbers_of_iterations): Dump
26232 we are estimating niter of loop.
26234 2022-03-21 Kito Cheng <kito.cheng@sifive.com>
26236 * common/config/riscv/riscv-common.cc (riscv_ext_flag_table):
26237 Update flag name and mask name.
26238 * config/riscv/riscv-c.cc (riscv_cpu_cpp_builtins): Define
26239 misc macro for vector extensions.
26240 * config/riscv/riscv-opts.h (MASK_VECTOR_EEW_32): Rename to ...
26241 (MASK_VECTOR_ELEN_32): ... this.
26242 (MASK_VECTOR_EEW_64): Rename to ...
26243 (MASK_VECTOR_ELEN_64): ... this.
26244 (MASK_VECTOR_EEW_FP_32): Rename to ...
26245 (MASK_VECTOR_ELEN_FP_32): ... this.
26246 (MASK_VECTOR_EEW_FP_64): Rename to ...
26247 (MASK_VECTOR_ELEN_FP_64): ... this.
26248 (TARGET_VECTOR_ELEN_32): New.
26249 (TARGET_VECTOR_ELEN_64): Ditto.
26250 (TARGET_VECTOR_ELEN_FP_32): Ditto.
26251 (TARGET_VECTOR_ELEN_FP_64): Ditto.
26252 (TARGET_MIN_VLEN): Ditto.
26253 * config/riscv/riscv.opt (riscv_vector_eew_flags): Rename to ...
26254 (riscv_vector_elen_flags): ... this.
26256 2022-03-21 Hongyu Wang <hongyu.wang@intel.com>
26259 * config/i386/sse.md
26260 (avx512fp16_fma<complexopname>sh_v8hf<mask_scalarcz_name><round_scalarcz_name>):
26261 Correct round operand for intel dialect.
26263 2022-03-19 Arthur Cohen <arthur.cohen@embecosm.com>
26265 * diagnostic.cc (diagnostic_cc_tests): Rename to...
26266 (c_diagnostic_cc_tests): ...this.
26267 * opt-problem.cc (opt_problem_cc_tests): Rename to...
26268 (c_opt_problem_cc_tests): ...this.
26269 * selftest-run-tests.cc (selftest::run_tests): No longer run
26270 opt_problem_cc_tests or diagnostic_cc_tests.
26271 * selftest.h (diagnostic_cc_tests): Remove declaration.
26272 (opt_problem_cc_tests): Likewise.
26274 2022-03-19 Marc Nieper-Wißkirchen <marc@nieper-wisskirchen.de>
26277 * hash-traits.h (struct typed_const_free_remove): New.
26278 (struct free_string_hash): New.
26279 * pass_manager.h: Use free_string_hash.
26280 * passes.cc (pass_manager::register_pass_name): Use free_string_hash.
26281 (pass_manager::~pass_manager): Delete allocated m_name_to_pass_map.
26283 2022-03-19 Jakub Jelinek <jakub@redhat.com>
26285 PR middle-end/104971
26286 * config/i386/i386-expand.cc
26287 (ix86_expand_builtin) <case IX86_BUILTIN_READ_FLAGS>: If ignore,
26288 don't push/pop anything and just return const0_rtx.
26290 2022-03-18 Vladimir N. Makarov <vmakarov@redhat.com>
26292 PR rtl-optimization/104961
26293 * lra-assigns.cc (find_reload_regno_insns): Process reload pseudo clobber.
26295 2022-03-18 Jason Merrill <jason@redhat.com>
26297 * tree.h (IDENTIFIER_LENGTH): Add comment.
26299 2022-03-18 Jakub Jelinek <jakub@redhat.com>
26301 PR middle-end/99578
26302 PR middle-end/100680
26303 PR tree-optimization/100834
26304 * params.opt (--param=min-pagesize=): New parameter.
26306 (compute_objsize_r) <case ARRAY_REF>: Formatting fix.
26307 (compute_objsize_r) <case INTEGER_CST>: Use maximum object size instead
26308 of zero for pointer constants equal or larger than min-pagesize.
26310 2022-03-18 Tom de Vries <tdevries@suse.de>
26312 * gimplify.cc (gimplify_omp_for): Set location using 'input_location'.
26313 Set gfor location only when dealing with a OMP_TASKLOOP.
26315 2022-03-18 Tom de Vries <tdevries@suse.de>
26317 * gimplify.cc (gimplify_omp_for): Set taskloop location.
26319 2022-03-18 Tom de Vries <tdevries@suse.de>
26322 * omp-low.cc (lower_rec_input_clauses): Make sure GOMP_SIMT_XCHG_BFLY
26323 is executed unconditionally.
26325 2022-03-18 liuhongt <hongtao.liu@intel.com>
26328 * config/i386/i386.md (*movhi_internal): Set attr type from HI
26329 to HF for alternative 12 under TARGET_AVX512FP16.
26331 2022-03-18 Cui,Lili <lili.cui@intel.com>
26334 * config/i386/i386.h (PTA_SAPPHIRERAPIDS): change it to base on ICX.
26335 * doc/invoke.texi: Update documents for Intel sapphirerapids.
26337 2022-03-17 Roger Sayle <roger@nextmovesoftware.com>
26340 PR tree-optimization/90356
26341 * config/i386/i386.md (*movtf_internal): Don't guard
26342 standard_sse_constant_p clause by optimize_function_for_size_p.
26343 (*movdf_internal): Likewise.
26344 (*movsf_internal): Likewise.
26346 2022-03-17 Andrew MacLeod <amacleod@redhat.com>
26348 PR tree-optimization/102943
26349 * gimple-range-cache.cc (ranger_cache::range_from_dom): Find range via
26350 dominators and apply intermediary outgoing edge ranges.
26352 2022-03-17 Richard Biener <rguenther@suse.de>
26354 PR tree-optimization/104960
26355 * passes.def: Add pass parameter to pass_sink_code, mark
26356 last one to unsplit edges.
26357 * tree-ssa-sink.cc (pass_sink_code::set_pass_param): New.
26358 (pass_sink_code::execute): Always execute TODO_cleanup_cfg
26359 when we need to unsplit edges.
26361 2022-03-17 Jakub Jelinek <jakub@redhat.com>
26363 PR middle-end/103984
26364 * gimplify.cc (gimplify_target_expr): Gimplify type sizes and
26365 TARGET_EXPR_INITIAL into a temporary sequence, then push clobbers
26366 and asan unpoisioning, then append the temporary sequence and
26367 finally the TARGET_EXPR_CLEANUP clobbers.
26369 2022-03-16 Roger Sayle <roger@nextmovesoftware.com>
26371 * config/i386/sse.md: Delete corrupt character/typo.
26373 2022-03-16 Roger Sayle <roger@nextmovesoftware.com>
26376 * config/i386/sse.md (sse2_movq128): New define_expand to
26377 preserve previous named instruction.
26378 (*sse2_movq128_<mode>): Renamed from sse2_movq128, and
26379 generalized to VI8F_128 (both V2DI and V2DF).
26381 2022-03-16 Siddhesh Poyarekar <siddhesh@gotplt.org>
26383 PR tree-optimization/104941
26384 * tree-object-size.cc (size_for_offset): Make useless conversion
26385 check lighter and assign result of fold_convert to OFFSET.
26387 2022-03-16 H.J. Lu <hjl.tools@gmail.com>
26390 * config/i386/x86gprintrin.h: Also check _SOFT_FLOAT before
26391 pushing target("general-regs-only").
26393 2022-03-16 Kito Cheng <kito.cheng@sifive.com>
26395 * common/config/riscv/riscv-common.cc (riscv_ext_version_table):
26396 Add version info for zk, zks and zkn.
26398 2022-03-16 LiaoShihua <shihua@iscas.ac.cn>
26400 * common/config/riscv/riscv-common.cc
26401 (riscv_combine_info): New.
26402 (riscv_subset_list::handle_combine_ext): Combine back into zk to
26403 maintain the canonical order in isa strings.
26404 (riscv_subset_list::parse): Ditto.
26405 * config/riscv/riscv-subset.h (handle_combine_ext): New.
26407 2022-03-16 Richard Biener <rguenther@suse.de>
26409 PR tree-optimization/102008
26410 * passes.def: Move the added code sinking pass before the
26411 preceeding phiopt pass.
26413 2022-03-16 Patrick Palka <ppalka@redhat.com>
26416 * doc/invoke.texi (C++ Dialect Options): Document
26417 -ffold-simple-inlines.
26419 2022-03-16 Siddhesh Poyarekar <siddhesh@gotplt.org>
26421 PR tree-optimization/104942
26422 * tree-object-size.cc (alloc_object_size): Remove STRIP_NOPS.
26424 2022-03-16 Jakub Jelinek <jakub@redhat.com>
26427 * config/aarch64/aarch64.cc (aarch64_load_symref_appropriately): Copy
26430 2022-03-16 Roger Sayle <roger@nextmovesoftware.com>
26431 Richard Biener <rguenther@suse.de>
26433 * gimple-match-head.cc (single_use): Implement inline using a
26436 2022-03-16 Roger Sayle <roger@nextmovesoftware.com>
26438 * match.pd (X CMP X -> true): Test tree_expr_maybe_nan_p
26439 instead of HONOR_NANS.
26440 (X LTGT X -> false): Enable if X is not tree_expr_maybe_nan_p, as
26441 this can't trap/signal.
26443 2022-03-16 liuhongt <hongtao.liu@intel.com>
26446 * config/i386/i386-builtin.def (BDESC): Add
26447 CODE_FOR_sse4_1_blendvpd for IX86_BUILTIN_BLENDVPD.
26448 * config/i386/i386.cc (ix86_gimple_fold_builtin): Don't fold
26449 __builtin_ia32_blendvpd w/o sse4.2
26451 2022-03-15 Peter Bergner <bergner@linux.ibm.com>
26454 * config/rs6000/predicates.md (mma_disassemble_output_operand): Restrict
26455 acceptable MEM addresses.
26457 2022-03-15 Jakub Jelinek <jakub@redhat.com>
26460 * config/riscv/riscv.cc (riscv_pass_aggregate_in_fpr_pair_p,
26461 riscv_pass_aggregate_in_fpr_and_gpr_p): Pass OPT_Wpsabi instead of 0
26464 2022-03-15 Jakub Jelinek <jakub@redhat.com>
26467 * config/i386/i386.h (LIBGCC2_UNWIND_ATTRIBUTE): Use no-mmx,no-sse
26468 instead of general-regs-only.
26470 2022-03-15 Roger Sayle <roger@nextmovesoftware.com>
26471 Marc Glisse <marc.glisse@inria.fr>
26472 Richard Biener <rguenther@suse.de>
26474 PR tree-optimization/101895
26475 * match.pd (vec_same_elem_p): Handle CONSTRUCTOR_EXPR def.
26476 (plus (vec_perm (mult ...) ...) ...): New reordering simplification.
26478 2022-03-15 Jakub Jelinek <jakub@redhat.com>
26480 PR rtl-optimization/104814
26481 * ifcvt.cc (find_if_case_1, find_if_case_2): Punt if test_bb doesn't
26482 end with onlyjump_p. Assume BB_END (test_bb) is always non-NULL.
26484 2022-03-15 Martin Sebor <msebor@redhat.com>
26486 PR middle-end/104436
26487 * gimple-ssa-warn-access.cc (pass_waccess::check_dangling_stores):
26488 Check for warning suppression. Avoid by-value arguments transformed
26489 into by-transparent-reference.
26491 2022-03-14 Roger Sayle <roger@nextmovesoftware.com>
26492 Uroš Bizjak <ubizjak@gmail.com>
26494 * config/i386/i386.md (peephole2 xorl;movb -> movzbl): Disable
26495 transformation when *zero_extend<mode>si2 is not available.
26497 2022-03-14 Xi Ruoyao <xry111@mengyan1223.wang>
26499 * config/mips/mips.h (SUBTARGET_SHADOW_OFFSET): Define.
26500 * config/mips/mips.cc (mips_option_override): Make
26501 -fsanitize=address imply -fasynchronous-unwind-tables. This is
26502 needed by libasan for stack backtrace on MIPS.
26503 (mips_asan_shadow_offset): Return SUBTARGET_SHADOW_OFFSET.
26505 2022-03-14 Jakub Jelinek <jakub@redhat.com>
26508 * lra.cc (lra_substitute_pseudo): For debug_p mode, simplify
26509 SUBREG, ZERO_EXTEND, SIGN_EXTEND, FLOAT or UNSIGNED_FLOAT if recursive
26510 call simplified the first operand into VOIDmode constant.
26512 2022-03-14 Jakub Jelinek <jakub@redhat.com>
26514 PR tree-optimization/102586
26515 * doc/extend.texi (__builtin_clear_padding): Clearify that for C++
26516 argument type should be pointer to trivially-copyable type unless it
26517 is address of a variable or parameter.
26519 2022-03-14 Jakub Jelinek <jakub@redhat.com>
26522 * config/i386/emmintrin.h (_mm_loadu_si32): Put loaded value into
26523 first rather than last element of the vector, use __m32_u to do
26524 a really unaligned load, use just 0 instead of (int)0.
26525 (_mm_loadu_si16): Put loaded value into first rather than last
26526 element of the vector, use __m16_u to do a really unaligned load,
26527 use just 0 instead of (short)0.
26529 2022-03-14 Jakub Jelinek <jakub@redhat.com>
26532 * config/bfin/bfin.cc (bfin_handle_longcall_attribute): Fix a typo
26533 in diagnostic message - cannott -> cannot. Use %< and %> around
26534 names of attribute. Avoid too long line.
26535 * range-op.cc (operator_logical_and::op1_range): Fix up a typo
26536 in comment - cannott -> cannot. Use 2 spaces after . instead of one.
26538 2022-03-14 liuhongt <hongtao.liu@intel.com>
26541 * config/i386/i386-expand.cc
26542 (ix86_check_builtin_isa_match): New func.
26543 (ix86_expand_builtin): Move code to
26544 ix86_check_builtin_isa_match and call it.
26545 * config/i386/i386-protos.h
26546 (ix86_check_builtin_isa_match): Declare.
26547 * config/i386/i386.cc (ix86_gimple_fold_builtin): Don't fold
26548 builtin into gimple when isa mismatches.
26550 2022-03-13 Tobias Burnus <tobias@codesourcery.com>
26552 * doc/invoke.texi: Fix typos.
26553 * doc/tm.texi.in: Remove duplicated word.
26554 * doc/tm.texi: Regenerate.
26556 2022-03-12 Segher Boessenkool <segher@kernel.crashing.org>
26559 * config/rs6000/rs6000.cc (rs6000_machine_from_flags): Don't output
26560 "ppc" and "ppc64" based on rs6000_cpu.
26562 2022-03-12 Thomas Schwinge <thomas@codesourcery.com>
26564 PR middle-end/100280
26565 PR middle-end/104892
26566 * omp-oacc-kernels-decompose.cc (omp_oacc_kernels_decompose_1):
26567 Remove special handling of 'GOMP_MAP_FORCE_TOFROM'.
26569 2022-03-12 Thomas Schwinge <thomas@codesourcery.com>
26571 PR middle-end/100280
26572 PR middle-end/104086
26573 * omp-oacc-kernels-decompose.cc (omp_oacc_kernels_decompose_1):
26574 Mark variables used in 'present' clauses as addressable.
26575 * omp-low.cc (scan_sharing_clauses) <OMP_CLAUSE_MAP>: Gracefully
26576 handle duplicate 'OMP_CLAUSE_MAP_DECL_MAKE_ADDRESSABLE'.
26578 2022-03-12 Thomas Schwinge <thomas@codesourcery.com>
26581 * tree-core.h (user_omp_claus_code_name): Declare function.
26582 * tree.cc (user_omp_clause_code_name): New function.
26584 2022-03-12 Roger Sayle <roger@nextmovesoftware.com>
26586 PR middle-end/98420
26587 * match.pd (minus @0 @0): Additional checks for -fno-rounding-math
26588 (the defaut) or -fno-signed-zeros.
26590 2022-03-12 Michael Meissner <meissner@linux.ibm.com>
26593 * config/rs6000/vsx.md (extendditi2): Use a 'b' constraint when
26594 moving from a GPR register to an Altivec register.
26596 2022-03-11 Roger Sayle <roger@nextmovesoftware.com>
26598 PR tree-optimization/98335
26599 * config/i386/i386.md (peephole2): Eliminate redundant insv.
26600 Combine movl followed by movb. Transform xorl followed by
26601 a suitable movb or movw into the equivalent movz[bw]l.
26603 2022-03-11 Roger Sayle <roger@nextmovesoftware.com>
26604 Richard Biener <rguenther@suse.de>
26606 PR tree-optimization/98335
26607 * builtins.cc (get_object_alignment_2): Export.
26608 * builtins.h (get_object_alignment_2): Likewise.
26609 * tree-ssa-alias.cc (ao_ref_alignment): New.
26610 * tree-ssa-alias.h (ao_ref_alignment): Declare.
26611 * tree-ssa-dse.cc (compute_trims): Improve logic deciding whether
26612 to align head/tail, writing more bytes but using fewer store insns.
26614 2022-03-11 Richard Biener <rguenther@suse.de>
26616 PR tree-optimization/104880
26617 * tree-ssa.cc (execute_update_address_taken): Remember if we
26618 optimistically made something not addressable and
26619 prepare to undo it.
26621 2022-03-11 Richard Biener <rguenther@suse.de>
26624 * config/i386/i386.cc (ix86_builtin_vectorization_cost): Do not
26625 cost the first lane of SSE pieces as inserts for vec_construct.
26627 2022-03-10 Roger Sayle <roger@nextmovesoftware.com>
26630 * calls.cc (expand_call): Ignore stack adjustments after sorry.
26632 2022-03-10 Vladimir N. Makarov <vmakarov@redhat.com>
26635 * lra-constraints.cc (split_reg): Set up
26636 check_and_force_assignment_correctness_p when splitting hard
26637 register live range.
26639 2022-03-10 Martin Jambor <mjambor@suse.cz>
26642 * ipa-cp.cc (create_specialized_node): Move removal of
26643 self-recursive calls from callers vector before refrence
26646 2022-03-10 Richard Biener <rguenther@suse.de>
26648 PR tree-optimization/102943
26649 * gimple-range-cache.cc (sbr_sparse_bitmap::bitvec):
26650 Make a bitmap_head.
26651 (sbr_sparse_bitmap::sbr_sparse_bitmap): Adjust and switch
26653 (sbr_sparse_bitmap::set_bb_range): Adjust.
26654 (sbr_sparse_bitmap::get_bb_range): Likewise.
26656 2022-03-10 Richard Biener <rguenther@suse.de>
26658 PR tree-optimization/102943
26659 * tree-ssa-dom.cc (back_propagate_equivalences): Only
26660 populate the dominance bitmap if fast queries are not
26661 available. Use a tree view bitmap.
26662 (record_temporary_equivalences): Cache the dominance bitmap
26663 across all equivalences on the edge.
26665 2022-03-10 Tom de Vries <tdevries@suse.de>
26668 * config/nvptx/nvptx.md (define_attr "predicable"): Use no,yes instead
26671 2022-03-10 Tom de Vries <tdevries@suse.de>
26674 * config/nvptx/nvptx.cc (nvptx_init_unisimt_predicate)
26675 (nvptx_output_unisimt_switch): Handle unisimt_outside_simt_predicate.
26676 (nvptx_get_unisimt_outside_simt_predicate): New function.
26677 (predicate_insn): New function, factored out of ...
26678 (nvptx_reorg_uniform_simt): ... here. Predicate all emitted insns.
26679 * config/nvptx/nvptx.h (struct machine_function): Add
26680 unisimt_outside_simt_predicate field.
26681 * config/nvptx/nvptx.md (define_insn "nvptx_warpsync")
26682 (define_insn "nvptx_uniform_warp_check"): Make predicable.
26684 2022-03-10 Tom de Vries <tdevries@suse.de>
26686 * config/nvptx/nvptx.cc (nvptx_unisimt_handle_set): Handle unused
26689 2022-03-10 Tom de Vries <tdevries@suse.de>
26692 * config/nvptx/nvptx.cc (nvptx_print_operand): Handle 'x' operand
26694 * config/nvptx/nvptx.md: Use %x0 destination operand in atom insns.
26696 2022-03-10 Tom de Vries <tdevries@suse.de>
26698 * config/nvptx/nvptx.md (define_insn "atomic_fetch_<logic><mode>"):
26699 Emit atom.and.b64 instead of atom.b64.and.
26701 2022-03-10 Tom de Vries <tdevries@suse.de>
26703 * config/nvptx/t-nvptx (MULTILIB_EXTRA_OPTS): Move mptx=3.1 ...
26704 (MULTILIB_OPTIONS): ... here.
26706 2022-03-10 Tom de Vries <tdevries@suse.de>
26709 * config/nvptx/nvptx.opt (misa): Set default to sm_30.
26710 * config/nvptx/t-nvptx (MULTILIB_EXTRA_OPTS): Remove misa=sm_30.
26712 2022-03-10 Thomas Schwinge <thomas@codesourcery.com>
26714 PR middle-end/90115
26715 PR middle-end/102330
26716 PR middle-end/104774
26717 * omp-low.cc (oacc_privatization_candidate_p)
26718 (oacc_privatization_scan_clause_chain)
26719 (oacc_privatization_scan_decl_chain, lower_oacc_private_marker):
26720 Analyze 'lookup_decl'-translated DECL.
26722 2022-03-10 Jakub Jelinek <jakub@redhat.com>
26725 * config/rs6000/rs6000.h (enum rs6000_builtin_type_index): Remove
26726 RS6000_BTI_ptr_ieee128_float and RS6000_BTI_ptr_ibm128_float.
26727 (ptr_ieee128_float_type_node, ptr_ibm128_float_type_node): Remove.
26728 * config/rs6000/rs6000-builtin.cc (rs6000_type_string): Return
26729 "**NULL**" if type_node is NULL first. Handle
26730 ieee128_float_type_node.
26731 (rs6000_init_builtins): Don't initialize ptr_ieee128_float_type_node
26732 and ptr_ibm128_float_type_node. Set ibm128_float_type_node and
26733 ieee128_float_type_node to NULL rather than long_double_type_node if
26734 they aren't supported. Do support __ibm128 even if
26735 !TARGET_FLOAT128_TYPE when long double is double double.
26736 (rs6000_expand_builtin): Error if bif_is_ibm128 and
26737 !ibm128_float_type_node. Remap RS6000_BIF_{,UN}PACK_IF to
26738 RS6000_BIF_{,UN}PACK_TF much earlier and only use bif_is_ibm128 check
26740 * config/rs6000/rs6000-c.cc (rs6000_target_modify_macros): Define
26741 __SIZEOF_FLOAT128__ here and only iff __float128 macro is defined.
26742 (rs6000_cpu_cpp_builtins): Don't define __SIZEOF_FLOAT128__ here.
26743 Define __SIZEOF_IBM128__=16 if ieee128_float_type_node is non-NULL.
26745 * config/rs6000/rs6000-gen-builtins.cc: Document ibm128 attribute.
26746 (struct attrinfo): Add isibm128 member.
26747 (TYPE_MAP_SIZE): Remove.
26748 (type_map): Use [] instead of [TYPE_MAP_SIZE]. For "if" use
26749 ibm128_float_type_node only if it is non-NULL, otherwise fall back
26750 to long_double_type_node. Remove "pif" entry.
26751 (parse_bif_attrs): Handle ibm128 attribute and print it for debugging.
26752 (write_decls): Output bif_ibm128_bit and bif_is_ibm128.
26753 (write_type_node): Use sizeof type_map / sizeof type_map[0]
26754 instead of TYPE_MAP_SIZE.
26755 (write_bif_static_init): Handle isibm128.
26756 * config/rs6000/rs6000-builtins.def: Document ibm128 attribute.
26757 (__builtin_pack_ibm128, __builtin_unpack_ibm128): Add ibm128
26760 2022-03-09 Richard Biener <rguenther@suse.de>
26762 * cfgexpand.cc (expand_gimple_asm): Special-case MEM_REF
26763 with non-decl operand, avoiding a copy.
26765 2022-03-09 Jakub Jelinek <jakub@redhat.com>
26768 * config/i386/i386.h (LIBGCC2_UNWIND_ATTRIBUTE): Define for ia32.
26770 2022-03-09 Richard Biener <rguenther@suse.de>
26772 PR middle-end/104786
26773 * cfgexpand.cc (expand_asm_stmt): Do not generate a copy
26774 for VLAs without an upper size bound.
26776 2022-03-09 Xi Ruoyao <xry111@mengyan1223.wang>
26778 PR tree-optimization/104851
26779 * optabs-query.cc (supports_vec_convert_optab_p): Fix off-by-one
26782 2022-03-09 Jakub Jelinek <jakub@redhat.com>
26785 * doc/invoke.texi (-Wextra): Document that -Wshift-negative-value
26786 is enabled by it only for C++11 to C++17 rather than for C++03 or
26788 (-Wshift-negative-value): Similarly (except here we stated
26789 that it is enabled for C++11 or later).
26791 2022-03-09 Jakub Jelinek <jakub@redhat.com>
26793 PR rtl-optimization/104839
26794 * simplify-rtx.cc (simplify_unary_operation_1) <case SIGN_EXTEND>:
26795 Use SRP_SIGNED instead of incorrect 1 in SUBREG_PROMOTED_SET.
26796 (simplify_unary_operation_1) <case ZERO_EXTEND>: Use SRP_UNSIGNED
26797 instead of incorrect 0 in SUBREG_PROMOTED_SET.
26799 2022-03-09 Xi Ruoyao <xry111@mengyan1223.wang>
26802 * config/mips/mips.h (LUI_OPERAND): Cast the input to an unsigned
26803 value before adding an offset.
26805 2022-03-08 Christophe Lyon <christophe.lyon@arm.com>
26807 * config/arm/arm-builtins.cc
26808 (arm_binop_none_none_unone_qualifiers): Delete.
26809 (BINOP_NONE_NONE_UNONE_QUALIFIERS): Delete.
26811 2022-03-08 Iain Sandoe <iain@sandoe.co.uk>
26813 PR translation/104552
26814 * config/host-darwin.cc (darwin_gt_pch_get_address): Amend
26815 the PCH out of memory error message punctuation and wording.
26817 2022-03-08 Marek Polacek <polacek@redhat.com>
26819 PR rtl-optimization/104777
26820 * rtl.cc (classify_insn): For ASM_OPERANDS, return JUMP_INSN only if
26821 ASM_OPERANDS_LABEL_VEC has at least one element.
26823 2022-03-08 H.J. Lu <hjl.tools@gmail.com>
26826 * config/i386/i386.cc (ix86_expand_epilogue): Sorry if there is
26827 stack realignment or regparm nested function with EH return.
26829 2022-03-08 Andre Vieira <andre.simoesdiasvieira@arm.com>
26832 * config/arm/arm.h (MVE_STN_LDW_MODE): New MACRO.
26833 * config/arm/arm.cc (mve_vector_mem_operand): Relax constraint on base
26834 register for non widening loads or narrowing stores.
26836 2022-03-08 Eric Gallager <egallager@gcc.gnu.org>
26838 PR translation/104552
26839 * params.opt: Fix typo.
26841 2022-03-08 Richard Biener <rguenther@suse.de>
26843 PR tree-optimization/84201
26844 * params.opt (-param=vect-induction-float): Add.
26845 * doc/invoke.texi (vect-induction-float): Document.
26846 * tree-vect-loop.cc (vectorizable_induction): Honor
26847 param_vect_induction_float.
26849 2022-03-08 Martin Jambor <mjambor@suse.cz>
26851 PR translation/104552
26852 * params.opt (ipa-cp-recursive-freq-factor): Remove repeated word
26853 "that" in the description.
26855 2022-03-08 Richard Biener <rguenther@suse.de>
26857 PR tree-optimization/104825
26858 * tree-ssa-sccvn.cc (visit_reference_op_call): Properly
26859 guard modref get_ao_ref on a pointer typed argument.
26861 2022-03-08 liuhongt <hongtao.liu@intel.com>
26863 * config/i386/sse.md (*vec_dupv4si): Disable memory operand
26864 for !TARGET_INTER_UNIT_MOVES_TO_VEC when prefer_for_speed.
26866 2022-03-07 Jonathan Wakely <jwakely@redhat.com>
26868 * doc/invoke.texi (C++ Modules): Remove anachronism.
26870 2022-03-07 Martin Liska <mliska@suse.cz>
26872 PR middle-end/104381
26873 * opts.cc (finish_options): If debug info is disabled
26874 (debug_info_level) and -fvar-tracking is unset, disable it.
26876 2022-03-07 Jakub Jelinek <jakub@redhat.com>
26878 * tree-ssa-propagate.cc: Fix up duplicated word issue in a comment.
26879 * config/riscv/riscv.cc: Likewise.
26880 * config/darwin.h: Likewise.
26881 * config/i386/i386.cc: Likewise.
26882 * config/aarch64/thunderx3t110.md: Likewise.
26883 * config/aarch64/fractional-cost.h: Likewise.
26884 * config/vax/vax.cc: Likewise.
26885 * config/rs6000/pcrel-opt.md: Likewise.
26886 * config/rs6000/predicates.md: Likewise.
26887 * ctfc.h: Likewise.
26888 * tree-ssa-uninit.cc: Likewise.
26889 * value-relation.h: Likewise.
26890 * gimple-range-gori.cc: Likewise.
26891 * ipa-polymorphic-call.cc: Likewise.
26892 * pointer-query.cc: Likewise.
26893 * ipa-sra.cc: Likewise.
26894 * internal-fn.cc: Likewise.
26895 * varasm.cc: Likewise.
26896 * gimple-ssa-warn-access.cc: Likewise.
26898 2022-03-07 Martin Liska <mliska@suse.cz>
26901 * config/arm/arm.cc (arm_option_override_internal): Add missing
26904 2022-03-07 Richard Biener <rguenther@suse.de>
26906 PR tree-optimization/104782
26907 * tree-vect-slp.cc (vectorize_slp_instance_root_stmt):
26908 Re-instantiate r10-5979 fix, add comment.
26910 2022-03-07 Martin Liska <mliska@suse.cz>
26913 * config/msp430/msp430.cc (msp430_expand_delay_cycles): Remove
26914 parenthesis from built-in name.
26916 2022-03-07 Martin Liska <mliska@suse.cz>
26919 * config/arm/arm.cc (arm_option_override_internal): Fix quoting
26920 of options in error messages.
26921 (arm_option_reconfigure_globals): Likewise.
26923 2022-03-07 Martin Liska <mliska@suse.cz>
26926 * config/arm/arm-builtins.cc (arm_expand_builtin): Reuse error
26927 message. Fix ARM_BUILTIN_WRORHI and ARM_BUILTIN_WRORH that can
26928 have only range [0,32].
26930 2022-03-07 Jakub Jelinek <jakub@redhat.com>
26933 * config/s390/s390.md (*cmp_and_trap_unsigned_int<mode>): Use
26934 S constraint instead of T in the last alternative.
26936 2022-03-07 Martin Liska <mliska@suse.cz>
26938 * plugin.cc (default_plugin_dir_name): Remove <dir> from error
26941 2022-03-07 Martin Liska <mliska@suse.cz>
26943 PR translation/90148
26944 * config/rs6000/rs6000.cc (rs6000_linux64_override_options): Put
26945 quote to a proper place.
26946 * plugin.cc (default_plugin_dir_name): Likewise.
26948 2022-03-07 Martin Liska <mliska@suse.cz>
26951 * config/rx/rx.cc (rx_expand_builtin_mvtc): Fix translation
26954 2022-03-07 Jakub Jelinek <jakub@redhat.com>
26957 * config/i386/sse.md (avx512dq_mul<mode>3<mask_name>): New
26958 define_expand pattern. Rename define_insn to ...
26959 (*avx512dq_mul<mode>3<mask_name>): ... this.
26960 (<code><mode>3_mask): New any_logic define_expand pattern.
26961 (<mask_codefor><code><mode>3<mask_name>): Rename to ...
26962 (*<code><mode>3<mask_name>): ... this.
26964 2022-03-05 Jakub Jelinek <jakub@redhat.com>
26966 * gimple-ssa-warn-access.cc (pass_waccess::use_after_inval_p): Remove
26967 visited bitmap and its use. Also punt on EDGE_ABNORMAL edges.
26969 2022-03-05 Roger Sayle <roger@nextmovesoftware.com>
26970 Uroš Bizjak <ubizjak@gmail.com>
26972 PR testsuite/104732
26973 * config/i386/i386.md (SWIM1248x): Renamed from SWIM1248s.
26974 Include DI mode unconditionally.
26975 (*anddi3_doubleword): Remove && TARGET_STV && TARGET_SSE2 condition,
26976 i.e. always split on !TARGET_64BIT.
26977 (*<any_or>di3_doubleword): Likewise.
26978 (*one_cmpldi2_doubleword): Likewise.
26979 (and<mode>3 expander): Update to use SWIM1248x from SWIM1248s.
26980 (<any_or><mode>3 expander): Likewise.
26981 (one_cmpl<mode>2 expander): Likewise.
26983 2022-03-05 Michael Meissner <meissner@linux.ibm.com>
26986 * config/rs6000/vsx.md (UNSPEC_MTVSRD_DITI_W1): Delete.
26987 (mtvsrdd_diti_w1): Delete.
26988 (extendditi2): Convert from define_expand to
26989 define_insn_and_split. Replace with code to deal with both GPR
26990 registers and with altivec registers.
26992 2022-03-04 Segher Boessenkool <segher@kernel.crashing.org>
26994 * config/rs6000/rs6000.cc (rs6000_machine_from_flags): Restructure a
26995 bit. Handle most older CPUs.
26997 2022-03-04 Iain Sandoe <iain@sandoe.co.uk>
26999 * config/darwin.cc (darwin_fold_builtin): Make fcode an int to
27000 avoid a mismatch with DECL_MD_FUNCTION_CODE().
27002 2022-03-04 Iain Sandoe <iain@sandoe.co.uk>
27005 * config/rs6000/darwin.md (@machopic_high_<mode>): New.
27006 (@machopic_low_<mode>): New.
27007 * config/rs6000/predicates.md (macho_pic_address): New.
27008 * config/rs6000/rs6000.cc (rs6000_legitimize_address): Do not
27009 apply the TLS processing to Darwin.
27010 * lra-constraints.cc (process_address_1): Revert the changes
27013 2022-03-04 Peter Bergner <bergner@linux.ibm.com>
27017 * config/rs6000/rs6000.cc (rs6000_option_override_internal): Make the
27018 ISA 2.06 requirement for -mabi=ieeelongdouble conditional on
27020 Move the -mabi=ieeelongdouble and -mabi=ibmlongdouble error checking
27022 * common/config/rs6000/rs6000-common.cc (rs6000_handle_option):
27025 2022-03-04 Jakub Jelinek <jakub@redhat.com>
27027 PR middle-end/104529
27028 * gimplify.cc (gimplify_init_constructor): Clear TREE_READONLY
27029 on automatic objects which will be runtime initialized.
27031 2022-03-04 Thomas Schwinge <thomas@codesourcery.com>
27033 PR middle-end/100280
27034 PR middle-end/104132
27035 PR middle-end/104133
27036 * omp-low.cc (task_shared_vars): Rename to
27037 'make_addressable_vars'. Adjust all users.
27038 (scan_sharing_clauses) <OMP_CLAUSE_MAP> Use it for
27039 'OMP_CLAUSE_MAP_DECL_MAKE_ADDRESSABLE' DECLs, too.
27041 2022-03-04 Thomas Schwinge <thomas@codesourcery.com>
27043 PR middle-end/100280
27044 * tree.h (OMP_CLAUSE_MAP_DECL_MAKE_ADDRESSABLE): New.
27045 * tree-core.h: Document it.
27046 * omp-low.cc (scan_sharing_clauses) <OMP_CLAUSE_MAP>: Handle
27047 'OMP_CLAUSE_MAP_DECL_MAKE_ADDRESSABLE'.
27048 * omp-oacc-kernels-decompose.cc (maybe_build_inner_data_region):
27049 Set 'OMP_CLAUSE_MAP_DECL_MAKE_ADDRESSABLE' instead of
27050 'TREE_ADDRESSABLE'.
27052 2022-03-04 Thomas Schwinge <thomas@codesourcery.com>
27054 PR middle-end/100280
27055 * omp-oacc-kernels-decompose.cc (maybe_build_inner_data_region):
27056 Add diagnostic: "note: OpenACC 'kernels' decomposition: variable
27057 '[...]' declared in block made addressable".
27059 2022-03-04 Thomas Schwinge <thomas@codesourcery.com>
27061 PR middle-end/100400
27062 PR middle-end/103836
27063 PR middle-end/104061
27064 * omp-oacc-kernels-decompose.cc (decompose_kernels_region_body):
27065 Catch 'GIMPLE_DEBUG'.
27067 2022-03-04 Jakub Jelinek <jakub@redhat.com>
27070 * tree.cc (warn_deprecated_use): For types prefer to use node
27071 and only use TYPE_MAIN_VARIANT (node) if TYPE_STUB_DECL (node) is
27074 2022-03-04 H.J. Lu <hjl.tools@gmail.com>
27077 * config/i386/i386.cc (ix86_gen_scratch_sse_rtx): Always return
27080 2022-03-03 Martin Sebor <msebor@redhat.com>
27082 PR middle-end/104761
27083 * gimple-ssa-warn-access.cc (pass_waccess::execute): Call
27084 mark_dfs_back_edges.
27086 2022-03-03 Martin Liska <mliska@suse.cz>
27088 * configure.ac: Use linker plug-in by default.
27089 * configure: Regenerate.
27091 2022-03-03 Martin Liska <mliska@suse.cz>
27093 * configure.ac: Now ld.mold support LTO plugin API, use it.
27094 * configure: Regenerate.
27096 2022-03-03 Tom de Vries <tdevries@suse.de>
27098 * config/nvptx/t-nvptx (MULTILIB_EXTRA_OPTS): Add mptx=3.1.
27100 2022-03-03 Tom de Vries <tdevries@suse.de>
27103 * config/nvptx/t-nvptx (MULTILIB_EXTRA_OPTS): Add misa=sm_30.
27105 2022-03-03 Tom de Vries <tdevries@suse.de>
27107 * config/nvptx/nvptx.h (ASM_SPEC): Add %{misa=sm_30:--no-verify}.
27109 2022-03-03 Jakub Jelinek <jakub@redhat.com>
27111 PR middle-end/104757
27112 * gimplify.cc (gimplify_omp_loop): Call gimplify_expr rather than
27114 (gimplify_expr) <case OMP_SIMD>: Temporarily disable
27115 gimplify_ctxp->into_ssa around call to gimplify_omp_for.
27117 2022-03-03 Jakub Jelinek <jakub@redhat.com>
27119 PR middle-end/104558
27120 * calls.cc (store_one_arg): When not calling emit_push_insn
27121 because size_rtx is const0_rtx, call at least anti_adjust_stack
27122 on arg->locate.alignment_pad if !argblock and the alignment might
27125 2022-03-02 Alexandre Oliva <oliva@adacore.com>
27127 * lra-constraints.cc (undo_optional_reloads): Recognize and
27128 drop insns of multi-word move sequences, tolerate removal
27129 iteration on an already-removed clobber, and refuse to
27130 substitute original pseudos into clobbers.
27132 2022-03-02 Qing Zhao <qing.zhao@oracle.com>
27134 PR middle-end/102276
27135 * common.opt (-Wtrivial-auto-var-init): New option.
27136 * doc/invoke.texi (-Wtrivial-auto-var-init): Document new option.
27137 (-ftrivial-auto-var-init): Update option;
27138 * gimplify.cc (emit_warn_switch_unreachable): New function.
27139 (warn_switch_unreachable_r): Rename to ...
27140 (warn_switch_unreachable_and_auto_init_r): This.
27141 (maybe_warn_switch_unreachable): Rename to ...
27142 (maybe_warn_switch_unreachable_and_auto_init): This.
27143 (gimplify_switch_expr): Update calls to renamed function.
27145 2022-03-02 Richard Biener <rguenther@suse.de>
27147 PR rtl-optimization/104686
27148 * ira-color.cc (object_conflicts_with_allocno_p): New function
27149 using a bitvector test instead of iterating when possible.
27150 (allocnos_conflict_p): Choose the best allocno to iterate over
27152 (update_conflict_hard_regno_costs): Do allocnos_conflict_p test
27155 2022-03-02 Jakub Jelinek <jakub@redhat.com>
27157 * cfg.cc (dump_edge_info): Dump goto_locus if present.
27159 2022-03-02 Jakub Jelinek <jakub@redhat.com>
27161 PR rtl-optimization/104589
27162 * cfgrtl.cc (fixup_reorder_chain): Use loc_equal instead of direct
27163 INSN_LOCATION comparison with goto_locus.
27165 2022-03-02 Jakub Jelinek <jakub@redhat.com>
27167 * tree-ssa-strlen.cc (strlen_pass::handle_assign,
27168 strlen_pass::before_dom_children): Comment spelling fixes.
27170 2022-03-02 Jakub Jelinek <jakub@redhat.com>
27172 * ipa-modref-tree.cc (modref_access_node::contains,
27173 modref_access_node::closer_pair_p, modref_access_node::insert,
27174 modref_access_node::insert_kill): Comment spelling fixes.
27175 * ipa-modref.cc: Likewise.
27176 (modref_summary::finalize, ignore_nondeterminism_p,
27177 class modref_access_analysis,
27178 modref_access_analysis::set_side_effects,
27179 modref_access_analysis::set_nondeterministic,
27180 modref_access_analysis::record_global_memory_load,
27181 modref_access_analysis::propagate, modref_access_analysis::analyze,
27182 struct escape_point, class modref_lattice, modref_lattice::merge,
27183 modref_lattice::merge_deref, class modref_eaf_analysis,
27184 modref_eaf_analysis::merge_call_lhs_flags,
27185 modref_eaf_analysis::analyze_ssa_name, modref_eaf_analysis::propagate,
27186 modref_eaf_analysis::record_escape_points, remap_kills,
27187 update_escape_summary, remove_useless_summaries,
27188 ipa_merge_modref_summary_after_inlining, pass_ipa_modref::execute):
27190 * ipa-modref.h (struct modref_summary, interposable_eaf_flags):
27192 * ipa-modref-tree.h (enum modref_special_parms,
27193 struct modref_access_node): Likewise.
27195 2022-03-01 Jakub Jelinek <jakub@redhat.com>
27197 PR tree-optimization/104715
27198 * gimple-ssa-warn-access.cc (pass_waccess::check_pointer_uses): Don't
27199 unnecessarily test if ptr is a SSA_NAME, it has to be. Only push lhs
27200 of a call if gimple_call_return_arg is equal to ptr, not just when it
27203 2022-03-01 Jakub Jelinek <jakub@redhat.com>
27205 * gimple-ssa-warn-access.cc (warn_string_no_nul,
27206 maybe_warn_nonstring_arg, fndecl_alloc_p, new_delete_mismatch_p,
27207 matching_alloc_calls_p, maybe_warn_alloc_args_overflow,
27208 pass_waccess::check_alloca, pass_waccess::check_strcat,
27209 memmodel_to_uhwi, fntype_argno_type,
27210 pass_waccess::maybe_check_access_sizes,
27211 pass_waccess::check_call_access,
27212 pass_waccess::maybe_check_dealloc_call, pass_waccess::check_stmt):
27213 Comment spelling fixes.
27215 2022-03-01 Richard Biener <rguenther@suse.de>
27217 PR tree-optimization/104716
27218 * tree-loop-distribution.cc (find_seed_stmts_for_distribution):
27219 Check if we can copy the loop.
27221 2022-03-01 H.J. Lu <hjl.tools@gmail.com>
27223 PR middle-end/104721
27224 * cfgexpand.cc (expand_gimple_basic_block): Clear
27225 currently_expanding_gimple_stmt when returning inside the loop.
27227 2022-03-01 Martin Liska <mliska@suse.cz>
27230 * multiple_target.cc (get_attr_len): Move to tree.c.
27231 (expand_target_clones): Remove single value checking.
27232 * tree.cc (get_target_clone_attr_len): New fn.
27233 * tree.h (get_target_clone_attr_len): Likewise.
27235 2022-03-01 Martin Liska <mliska@suse.cz>
27237 PR gcov-profile/104677
27238 * doc/invoke.texi: Document more .gcda file name generation.
27240 2022-03-01 Tom de Vries <tdevries@suse.de>
27243 * config/nvptx/nvptx.cc (nvptx_gen_shuffle): Handle DCmode and CDImode.
27244 * config/nvptx/nvptx.md
27245 (define_predicate "nvptx_register_or_complex_di_df_register_operand"):
27247 (define_expand "omp_simt_xchg_bfly", define_expand "omp_simt_xchg_idx"):
27248 Use nvptx_register_or_complex_di_df_register_operand.
27250 2022-03-01 Tom de Vries <tdevries@suse.de>
27252 * config.gcc (nvptx*-*-*): Add nvptx/nvptx-gen.opt to extra_options.
27253 * config/nvptx/gen-copyright.sh: New file.
27254 * config/nvptx/gen-h.sh: New file.
27255 * config/nvptx/gen-opt.sh: New file.
27256 * config/nvptx/nvptx.h (TARGET_SM35, TARGET_SM53, TARGET_SM70)
27257 (TARGET_SM75, TARGET_SM80): Move ...
27258 * config/nvptx/nvptx-gen.h: ... here. New file, generate.
27259 * config/nvptx/nvptx.opt (Enum ptx_isa): Move ...
27260 * config/nvptx/nvptx-gen.opt: ... here. New file, generate.
27261 * config/nvptx/t-nvptx ($(srcdir)/config/nvptx/nvptx-gen.h)
27262 ($(srcdir)/config/nvptx/nvptx-gen.opt): New make target.
27264 2022-03-01 Tom de Vries <tdevries@suse.de>
27266 * config/nvptx/gen-omp-device-properties.sh: New file.
27267 * config/nvptx/t-omp-device: Use gen-omp-device-properties.sh.
27269 2022-03-01 Tom de Vries <tdevries@suse.de>
27271 * config/nvptx/nvptx-sm.def: New file.
27272 * config/nvptx/nvptx-c.cc (nvptx_cpu_cpp_builtins): Use nvptx-sm.def.
27273 * config/nvptx/nvptx-opts.h (enum ptx_isa): Same.
27274 * config/nvptx/nvptx.cc (sm_version_to_string)
27275 (nvptx_omp_device_kind_arch_isa): Same.
27277 2022-03-01 Robin Dapp <rdapp@linux.ibm.com>
27279 PR rtl-optimization/104154
27280 * config/arc/arc.cc (gen_compare_reg): Return the CC-mode
27281 comparison ifcvt passed us.
27283 2022-03-01 Hongyu Wang <hongyu.wang@intel.com>
27286 * config/i386/i386-expand.cc (ix86_expand_vector_init_duplicate):
27287 Use vec_setv8hf_0 for HF to V8HFmode move instead of subreg.
27289 2022-02-28 Roger Sayle <roger@nextmovesoftware.com>
27291 PR tree-optimization/91384
27292 * config/i386/i386.md (peephole2): Eliminate final testl insn
27293 from the sequence *movsi_internal, *negsi_1, *cmpsi_ccno_1 by
27294 transforming using *negsi_2 for the negation.
27296 2022-02-28 Roger Sayle <roger@nextmovesoftware.com>
27297 Eric Botcazou <ebotcazou@adacore.com>
27299 PR middle-end/80270
27300 * expmed.cc (extract_integral_bit_field): If OP0 is a hard
27301 register, copy it to a pseudo before calling simplify_gen_subreg.
27303 2022-02-28 Vladimir N. Makarov <vmakarov@redhat.com>
27305 PR rtl-optimization/104637
27306 * lra-assigns.cc (lra_split_hard_reg_for): Split hard regs as many
27307 as possible on one subpass.
27309 2022-02-28 Qing Zhao <qing.zhao@oracle.com>
27311 PR middle-end/104550
27312 * gimple-fold.cc (clear_padding_flush): Suppress warnings for new
27315 2022-02-28 Martin Liska <mliska@suse.cz>
27318 * main.cc (main): Use flag_checking instead of CHECKING_P
27319 and run toplev::finalize only if there is not error seen.
27321 2022-02-28 Richard Biener <rguenther@suse.de>
27323 * tree-ssa-pre.cc (compute_avail): Revert part of last change.
27325 2022-02-28 Richard Biener <rguenther@suse.de>
27327 PR tree-optimization/104700
27328 * tree-ssa-pre.cc (get_or_alloc_expr_for): Remove and inline
27330 (find_or_generate_expression): ... here, simplifying code.
27332 2022-02-28 Tom de Vries <tdevries@suse.de>
27334 * config/nvptx/nvptx-opts.h (enum ptx_version): Add
27335 PTX_VERSION_default.
27336 * config/nvptx/nvptx.cc (handle_ptx_version_option): Handle
27337 PTX_VERSION_default.
27338 * config/nvptx/nvptx.opt: Add EnumValue "_" / PTX_VERSION_default.
27340 2022-02-28 Richard Biener <rguenther@suse.de>
27342 PR rtl-optimization/104686
27343 * ira-int.h (minmax_set_iter_cond): Use ctz_hwi to elide loop
27344 skipping bits that are zero.
27345 (ira_object_conflict_iter_cond): Likewise.
27347 2022-02-28 Hongyu Wang <hongyu.wang@intel.com>
27349 * config/i386/avx512fintrin.h (_MM_TERNLOG_ENUM): New enum.
27350 (_mm512_ternarylogic_epi64): Truncate imm to unsigned
27351 char to avoid error when using ~enum as parameter.
27352 (_mm512_mask_ternarylogic_epi64): Likewise.
27353 (_mm512_maskz_ternarylogic_epi64): Likewise.
27354 (_mm512_ternarylogic_epi32): Likewise.
27355 (_mm512_mask_ternarylogic_epi32): Likewise.
27356 (_mm512_maskz_ternarylogic_epi32): Likewise.
27357 * config/i386/avx512vlintrin.h (_mm256_ternarylogic_epi64):
27358 Adjust imm param type to unsigned char.
27359 (_mm256_mask_ternarylogic_epi64): Likewise.
27360 (_mm256_maskz_ternarylogic_epi64): Likewise.
27361 (_mm256_ternarylogic_epi32): Likewise.
27362 (_mm256_mask_ternarylogic_epi32): Likewise.
27363 (_mm256_maskz_ternarylogic_epi32): Likewise.
27364 (_mm_ternarylogic_epi64): Likewise.
27365 (_mm_mask_ternarylogic_epi64): Likewise.
27366 (_mm_maskz_ternarylogic_epi64): Likewise.
27367 (_mm_ternarylogic_epi32): Likewise.
27368 (_mm_mask_ternarylogic_epi32): Likewise.
27369 (_mm_maskz_ternarylogic_epi32): Likewise.
27371 2022-02-25 Jakub Jelinek <jakub@redhat.com>
27372 Marc Glisse <marc.glisse@inria.fr>
27374 PR tree-optimization/104675
27375 * match.pd (t * 2U / 2 -> t & (~0 / 2), t / 2U * 2 -> t & ~1):
27376 Restrict simplifications to INTEGRAL_TYPE_P.
27378 2022-02-25 Jakub Jelinek <jakub@redhat.com>
27381 * config/rs6000/vector.md (movmisalign<mode>): Use rs6000_emit_move.
27383 2022-02-25 Claudiu Zissulescu <claziss@synopsys.com>
27385 * config/arc/arc.cc (gen_compare_reg): Return NULL_RTX if the
27386 comparison is not valid.
27387 * config/arc/arc.md (movsicc): Fail if comparison is not valid.
27388 (movdicc): Likewise.
27389 (movsfcc): Likewise.
27390 (movdfcc): Likewise.
27392 2022-02-25 Richard Biener <rguenther@suse.de>
27394 PR tree-optimization/103037
27395 * tree-ssa-sccvn.h (alloc_vn_nary_op_noinit): Declare.
27396 (vn_nary_length_from_stmt): Likewise.
27397 (init_vn_nary_op_from_stmt): Likewise.
27398 (vn_nary_op_compute_hash): Likewise.
27399 * tree-ssa-sccvn.cc (alloc_vn_nary_op_noinit): Export.
27400 (vn_nary_length_from_stmt): Likewise.
27401 (init_vn_nary_op_from_stmt): Likewise.
27402 (vn_nary_op_compute_hash): Likewise.
27403 * tree-ssa-pre.cc (pre_expr_obstack): New obstack.
27404 (get_or_alloc_expr_for_nary): Pass in the value-id to use,
27405 (re-)compute the hash value and if the expression is not
27406 found allocate it from pre_expr_obstack.
27407 (phi_translate_1): Do not insert the NARY found in the
27408 VN tables but build a PRE expression from the valueized
27409 NARY with the value-id we eventually found.
27410 (find_or_generate_expression): Assert we have an entry
27411 for constant values.
27412 (compute_avail): Insert not valueized expressions into
27413 EXP_GEN using the value-id from the VN tables.
27414 (init_pre): Allocate pre_expr_obstack.
27415 (fini_pre): Free pre_expr_obstack.
27417 2022-02-25 Jakub Jelinek <jakub@redhat.com>
27420 * config/i386/i386.h (enum ix86_stack_slot): Add SLOT_FLOATxFDI_387.
27421 * config/i386/i386.md (splitter to floatdi<mode>2_i387_with_xmm): Use
27422 SLOT_FLOATxFDI_387 rather than SLOT_TEMP.
27424 2022-02-25 Jakub Jelinek <jakub@redhat.com>
27426 * warning-control.cc (get_nowarn_spec): Comment spelling fix.
27428 2022-02-25 Jakub Jelinek <jakub@redhat.com>
27430 PR middle-end/104679
27431 * internal-fn.cc (expand_SPACESHIP): Call do_pending_stack_adjust.
27433 2022-02-25 Jakub Jelinek <jakub@redhat.com>
27435 PR tree-optimization/104675
27436 * match.pd (-A - 1 -> ~A, -1 - A -> ~A): Don't simplify for
27439 2022-02-25 Alexandre Oliva <oliva@adacore.com>
27443 * expr.cc (emit_move_multi_word): Restore clobbers during LRA.
27445 2022-02-25 Alexandre Oliva <oliva@adacore.com>
27447 PR middle-end/104540
27448 * dwarf2cfi.cc (cfi_oprnd_equal_p): Cope with NULL
27451 2022-02-25 Alexandre Oliva <oliva@adacore.com>
27453 PR tree-optimization/103856
27454 * gimple-harden-conditionals.cc (non_eh_succ_edge): Enable the
27455 eh edge to be requested through an extra parameter.
27456 (pass_harden_compares::execute): Copy PHI args in the EH dest
27457 block for the new EH edge added for the inverted compare.
27459 2022-02-24 Palmer Dabbelt <palmer@rivosinc.com>
27461 * doc/invoke.texi (RISC-V -mcmodel=medany): Document the degree
27462 of position independence that -mcmodel=medany affords.
27464 2022-02-24 Jose E. Marchesi <jose.marchesi@oracle.com>
27467 * configure.ac: --disable-gcov if targetting bpf-*.
27468 * configure: Regenerate.
27470 2022-02-24 Richard Biener <rguenther@suse.de>
27472 PR tree-optimization/104676
27473 * tree-loop-distribution.cc (loop_distribution::execute):
27474 Do a full scev_reset.
27476 2022-02-24 Jakub Jelinek <jakub@redhat.com>
27478 PR tree-optimization/104601
27479 * tree-ssa-sccvn.cc (visit_reference_op_call): For calls with
27480 non-SSA_NAME lhs value number vdef to itself instead of e.g. the
27483 2022-02-24 Tom de Vries <tdevries@suse.de>
27484 Tobias Burnus <tobias@codesourcery.com>
27486 * config/nvptx/nvptx.cc (nvptx_omp_device_kind_arch_isa): Handle
27487 sm_70, sm_75 and sm_80.
27488 * config/nvptx/t-omp-device: Add sm_53, sm_70, sm_75 and sm_80.
27490 2022-02-24 Tom de Vries <tdevries@suse.de>
27492 * config/nvptx/nvptx.md (define_insn "rotlsi3", define_insn
27493 "rotrsi3"): New define_insn.
27495 2022-02-24 Tom de Vries <tdevries@suse.de>
27497 * config/nvptx/nvptx.cc (gen_comment): Use
27498 DECL_SOURCE_LOCATION (cfun->decl) instead of cfun->function_start_locus.
27500 2022-02-24 liuhongt <hongtao.liu@intel.com>
27502 * config/i386/sse.md (<code>v1ti3): Add suffix and replace
27503 isa attr of alternative 2 from avx to avx512vl.
27505 2022-02-23 Richard Biener <rguenther@suse.de>
27506 Jakub Jelinek <jakub@redhat.com>
27508 PR tree-optimization/104644
27509 * doc/match-and-simplify.texi: Amend ! documentation.
27510 * genmatch.cc (expr::gen_transform): Code-generate ! support
27512 (parser::parse_expr): Allow ! for GENERIC.
27513 * match.pd (cmp (bswap @0) INTEGER_CST@1): Use ! modifier on
27516 2022-02-23 Richard Biener <rguenther@suse.de>
27518 PR tree-optimization/101636
27519 * tree-vect-slp.cc (vect_print_slp_tree): Dump the
27520 vector type of the node.
27521 (vect_slp_analyze_operations): Make sure the CTOR
27522 is vectorized with an expected type.
27523 (vectorize_slp_instance_root_stmt): Revert r10-5979 fix.
27525 2022-02-23 Jakub Jelinek <jakub@redhat.com>
27528 * gimple-warn-recursion.cc (pass_warn_recursion::find_function_exit):
27529 Don't warn about calls to corresponding builtin from extern inline
27530 gnu_inline wrappers.
27532 2022-02-23 Roger Sayle <roger@nextmovesoftware.com>
27535 * config/nvptx/nvptx.md (*movhf_insn): Add subregs_ok attribute.
27537 2022-02-23 Christophe Lyon <christophe.lyon@arm.com>
27541 * config/arm/vfp.md (thumb2_movhi_vfp, thumb2_movhi_fp16): Fix
27544 2022-02-23 Cui,Lili <lili.cui@intel.com>
27546 * doc/invoke.texi: Update documents for Intel architectures.
27548 2022-02-22 Roger Sayle <roger@nextmovesoftware.com>
27550 * config/i386/i386-expand.cc (ix86_expand_cmpxchg_loop): Restore
27553 2022-02-22 Thomas Schwinge <thomas@codesourcery.com>
27555 * omp-low.cc (omp_build_component_ref): Move function...
27556 * omp-general.cc (omp_build_component_ref): ... here. Remove
27558 * omp-general.h (omp_build_component_ref): Declare function.
27559 * omp-oacc-neuter-broadcast.cc (oacc_build_component_ref): Remove
27561 (build_receiver_ref, build_sender_ref): Call
27562 'omp_build_component_ref' instead.
27564 2022-02-22 Thomas Schwinge <thomas@codesourcery.com>
27566 * omp-oacc-neuter-broadcast.cc (record_field_map_t): Further
27567 simplify. Adjust all users.
27569 2022-02-22 Segher Boessenkool <segher@kernel.crashing.org>
27572 * config/rs6000/rs6000.cc (atomic_hold_decl, atomic_clear_decl,
27573 atomic_update_decl): Add GTY markup.
27575 2022-02-22 Christophe Lyon <christophe.lyon@arm.com>
27577 * config/arm/arm.h (REG_CLASS_CONTENTS): Add VPR_REG to ALL_REGS.
27579 2022-02-22 Christophe Lyon <christophe.lyon@arm.com>
27583 * config/arm/arm-builtins.cc (CX_UNARY_UNONE_QUALIFIERS): Use
27585 (CX_BINARY_UNONE_QUALIFIERS): Likewise.
27586 (CX_TERNARY_UNONE_QUALIFIERS): Likewise.
27587 (TERNOP_NONE_NONE_NONE_UNONE_QUALIFIERS): Delete.
27588 (QUADOP_NONE_NONE_NONE_NONE_UNONE_QUALIFIERS): Delete.
27589 (QUADOP_UNONE_UNONE_UNONE_UNONE_UNONE_QUALIFIERS): Delete.
27590 * config/arm/arm_mve_builtins.def: Use predicated qualifiers.
27591 * config/arm/mve.md: Use VxBI instead of HI.
27593 2022-02-22 Christophe Lyon <christophe.lyon@arm.com>
27597 * config/arm/arm-builtins.cc (STRSBS_P_QUALIFIERS): Use predicate
27599 (STRSBU_P_QUALIFIERS): Likewise.
27600 (LDRGBS_Z_QUALIFIERS): Likewise.
27601 (LDRGBU_Z_QUALIFIERS): Likewise.
27602 (LDRGBWBXU_Z_QUALIFIERS): Likewise.
27603 (LDRGBWBS_Z_QUALIFIERS): Likewise.
27604 (LDRGBWBU_Z_QUALIFIERS): Likewise.
27605 (STRSBWBS_P_QUALIFIERS): Likewise.
27606 (STRSBWBU_P_QUALIFIERS): Likewise.
27607 * config/arm/mve.md: Use VxBI instead of HI.
27609 2022-02-22 Christophe Lyon <christophe.lyon@arm.com>
27613 * config/arm/arm-builtins.cc (TERNOP_UNONE_UNONE_NONE_UNONE_QUALIFIERS): Change to ...
27614 (TERNOP_UNONE_UNONE_NONE_PRED_QUALIFIERS): ... this.
27615 (TERNOP_UNONE_UNONE_IMM_UNONE_QUALIFIERS): Change to ...
27616 (TERNOP_UNONE_UNONE_IMM_PRED_QUALIFIERS): ... this.
27617 (TERNOP_NONE_NONE_IMM_UNONE_QUALIFIERS): Change to ...
27618 (TERNOP_NONE_NONE_IMM_PRED_QUALIFIERS): ... this.
27619 (TERNOP_NONE_NONE_UNONE_UNONE_QUALIFIERS): Change to ...
27620 (TERNOP_NONE_NONE_UNONE_PRED_QUALIFIERS): ... this.
27621 (QUADOP_UNONE_UNONE_NONE_NONE_UNONE_QUALIFIERS): Change to ...
27622 (QUADOP_UNONE_UNONE_NONE_NONE_PRED_QUALIFIERS): ... this.
27623 (QUADOP_NONE_NONE_NONE_NONE_PRED_QUALIFIERS): New.
27624 (QUADOP_NONE_NONE_NONE_IMM_UNONE_QUALIFIERS): Change to ...
27625 (QUADOP_NONE_NONE_NONE_IMM_PRED_QUALIFIERS): ... this.
27626 (QUADOP_UNONE_UNONE_UNONE_UNONE_PRED_QUALIFIERS): New.
27627 (QUADOP_UNONE_UNONE_NONE_IMM_UNONE_QUALIFIERS): Change to ...
27628 (QUADOP_UNONE_UNONE_NONE_IMM_PRED_QUALIFIERS): ... this.
27629 (QUADOP_NONE_NONE_UNONE_IMM_UNONE_QUALIFIERS): Change to ...
27630 (QUADOP_NONE_NONE_UNONE_IMM_PRED_QUALIFIERS): ... this.
27631 (QUADOP_UNONE_UNONE_UNONE_IMM_UNONE_QUALIFIERS): Change to ...
27632 (QUADOP_UNONE_UNONE_UNONE_IMM_PRED_QUALIFIERS): ... this.
27633 (QUADOP_UNONE_UNONE_UNONE_NONE_UNONE_QUALIFIERS): Change to ...
27634 (QUADOP_UNONE_UNONE_UNONE_NONE_PRED_QUALIFIERS): ... this.
27635 (STRS_P_QUALIFIERS): Use predicate qualifier.
27636 (STRU_P_QUALIFIERS): Likewise.
27637 (STRSU_P_QUALIFIERS): Likewise.
27638 (STRSS_P_QUALIFIERS): Likewise.
27639 (LDRGS_Z_QUALIFIERS): Likewise.
27640 (LDRGU_Z_QUALIFIERS): Likewise.
27641 (LDRS_Z_QUALIFIERS): Likewise.
27642 (LDRU_Z_QUALIFIERS): Likewise.
27643 (QUINOP_UNONE_UNONE_UNONE_UNONE_IMM_UNONE_QUALIFIERS): Change to ...
27644 (QUINOP_UNONE_UNONE_UNONE_UNONE_IMM_PRED_QUALIFIERS): ... this.
27645 (BINOP_NONE_NONE_PRED_QUALIFIERS): New.
27646 (BINOP_UNONE_UNONE_PRED_QUALIFIERS): New.
27647 * config/arm/arm_mve_builtins.def: Use new predicated qualifiers.
27648 * config/arm/mve.md: Use MVE_VPRED instead of HI.
27650 2022-02-22 Christophe Lyon <christophe.lyon@arm.com>
27654 * config/arm/arm-builtins.cc (BINOP_UNONE_NONE_NONE_QUALIFIERS):
27656 (TERNOP_UNONE_NONE_NONE_UNONE_QUALIFIERS): Change to ...
27657 (TERNOP_PRED_NONE_NONE_PRED_QUALIFIERS): ... this.
27658 (TERNOP_PRED_UNONE_UNONE_PRED_QUALIFIERS): New.
27659 * config/arm/arm_mve_builtins.def (vcmp*q_n_, vcmp*q_m_f): Use new
27660 predicated qualifiers.
27661 * config/arm/mve.md (mve_vcmp<mve_cmp_op>q_n_<mode>)
27662 (mve_vcmp*q_m_f<mode>): Use MVE_VPRED instead of HI.
27664 2022-02-22 Christophe Lyon <christophe.lyon@arm.com>
27667 * config/arm/arm-protos.h (arm_get_mask_mode): New prototype.
27668 (arm_expand_vector_compare): Update prototype.
27669 * config/arm/arm.cc (TARGET_VECTORIZE_GET_MASK_MODE): New.
27670 (arm_vector_mode_supported_p): Add support for VxBI modes.
27671 (arm_expand_vector_compare): Remove useless generation of vpsel.
27672 (arm_expand_vcond): Fix select operands.
27673 (arm_get_mask_mode): New.
27674 * config/arm/mve.md (vec_cmp<mode><MVE_vpred>): New.
27675 (vec_cmpu<mode><MVE_vpred>): New.
27676 (vcond_mask_<mode><MVE_vpred>): New.
27677 * config/arm/vec-common.md (vec_cmp<mode><v_cmp_result>)
27678 (vec_cmpu<mode><mode, vcond_mask_<mode><v_cmp_result>): Move to ...
27679 * config/arm/neon.md (vec_cmp<mode><v_cmp_result>)
27680 (vec_cmpu<mode><mode, vcond_mask_<mode><v_cmp_result>): ... here
27681 and disable for MVE.
27682 * doc/sourcebuild.texi (arm_mve): Document new effective-target.
27684 2022-02-22 Christophe Lyon <christophe.lyon@arm.com>
27688 * config/arm/arm-builtins.cc (BINOP_PRED_UNONE_UNONE_QUALIFIERS)
27689 (BINOP_PRED_NONE_NONE_QUALIFIERS)
27690 (TERNOP_NONE_NONE_NONE_PRED_QUALIFIERS)
27691 (TERNOP_UNONE_UNONE_UNONE_PRED_QUALIFIERS): New.
27692 * config/arm/arm-protos.h (mve_bool_vec_to_const): New.
27693 * config/arm/arm.cc (arm_hard_regno_mode_ok): Handle new VxBI
27695 (arm_mode_to_pred_mode): New.
27696 (arm_expand_vector_compare): Use the right VxBI mode instead of
27698 (arm_expand_vcond): Likewise.
27699 (simd_valid_immediate): Handle MODE_VECTOR_BOOL.
27700 (mve_bool_vec_to_const): New.
27701 (neon_make_constant): Call mve_bool_vec_to_const when needed.
27702 * config/arm/arm_mve_builtins.def (vcmpneq_, vcmphiq_, vcmpcsq_)
27703 (vcmpltq_, vcmpleq_, vcmpgtq_, vcmpgeq_, vcmpeqq_, vcmpneq_f)
27704 (vcmpltq_f, vcmpleq_f, vcmpgtq_f, vcmpgeq_f, vcmpeqq_f, vpselq_u)
27705 (vpselq_s, vpselq_f): Use new predicated qualifiers.
27706 * config/arm/constraints.md (DB): New.
27707 * config/arm/iterators.md (MVE_7, MVE_7_HI): New mode iterators.
27708 (MVE_VPRED, MVE_vpred): New attribute iterators.
27709 * config/arm/mve.md (@mve_vcmp<mve_cmp_op>q_<mode>)
27710 (@mve_vcmp<mve_cmp_op>q_f<mode>, @mve_vpselq_<supf><mode>)
27711 (@mve_vpselq_f<mode>): Use MVE_VPRED instead of HI.
27712 (@mve_vpselq_<supf>v2di): Define separately.
27713 (mov<mode>): New expander for VxBI modes.
27714 * config/arm/vfp.md (thumb2_movhi_vfp, thumb2_movhi_fp16): Use
27715 MVE_7_HI iterator and add support for DB constraint.
27717 2022-02-22 Christophe Lyon <christophe.lyon@arm.com>
27718 Richard Sandiford <richard.sandiford@arm.com>
27722 * config/aarch64/aarch64-modes.def (VNx16BI, VNx8BI, VNx4BI,
27723 VNx2BI): Update definition.
27724 * config/arm/arm-builtins.cc (arm_init_simd_builtin_types): Add new
27726 (arm_init_builtin): Map predicate vectors arguments to HImode.
27727 (arm_expand_builtin_args): Move HImode predicate arguments to VxBI
27728 rtx. Move return value to HImode rtx.
27729 * config/arm/arm-builtins.h (arm_type_qualifiers): Add qualifier_predicate.
27730 * config/arm/arm-modes.def (B2I, B4I, V16BI, V8BI, V4BI): New modes.
27731 * config/arm/arm-simd-builtin-types.def (Pred1x16_t,
27732 Pred2x8_t,Pred4x4_t): New.
27733 * emit-rtl.cc (init_emit_once): Handle all boolean modes.
27734 * genmodes.cc (mode_data): Add boolean field.
27735 (blank_mode): Initialize it.
27736 (make_complex_modes): Fix handling of boolean modes.
27737 (make_vector_modes): Likewise.
27738 (VECTOR_BOOL_MODE): Use new COMPONENT parameter.
27739 (make_vector_bool_mode): Likewise.
27741 (make_bool_mode): New.
27742 (emit_insn_modes_h): Fix generation of boolean modes.
27743 (emit_class_narrowest_mode): Likewise.
27744 * machmode.def: (VECTOR_BOOL_MODE): Document new COMPONENT
27745 parameter. Use new BOOL_MODE instead of FRACTIONAL_INT_MODE to
27747 * rtx-vector-builder.cc (rtx_vector_builder::find_cached_value):
27748 Fix handling of constm1_rtx for VECTOR_BOOL.
27749 * simplify-rtx.cc (native_encode_rtx): Fix support for VECTOR_BOOL.
27750 (native_decode_vector_rtx): Likewise.
27751 (test_vector_ops_duplicate): Skip vec_merge test
27752 with vectors of booleans.
27753 * varasm.cc (output_constant_pool_2): Likewise.
27755 2022-02-22 Christophe Lyon <christophe.lyon@arm.com>
27757 * config/arm/mve.md (mve_vmvnq_n_<supf><mode>): Use V_elem mode
27760 2022-02-22 Christophe Lyon <christophe.lyon@arm.com>
27762 * config/arm/arm.cc (arm_class_likely_spilled_p): Handle VPR_REG.
27764 2022-02-22 Christophe Lyon <christophe.lyon@arm.com>
27766 * config/arm/arm.h (reg_class): Add GENERAL_AND_VPR_REGS.
27767 (REG_CLASS_NAMES): Likewise.
27768 (REG_CLASS_CONTENTS): Likewise.
27769 (CLASS_MAX_NREGS): Handle VPR.
27770 * config/arm/arm.cc (arm_hard_regno_nregs): Handle VPR.
27772 2022-02-22 Tobias Burnus <tobias@codesourcery.com>
27773 Tom de Vries <tdevries@suse.de>
27775 * config/nvptx/nvptx-c.cc (nvptx_cpu_cpp_builtins): Handle SM70.
27776 * config/nvptx/nvptx.cc (first_ptx_version_supporting_sm):
27778 * config/nvptx/nvptx.opt (misa): Add sm_70 alias PTX_ISA_SM70.
27780 2022-02-22 Tobias Burnus <tobias@codesourcery.com>
27781 Tom de Vries <tdevries@suse.de>
27783 * config/nvptx/nvptx.opt (mptx): Add 6.0 alias PTX_VERSION_6_0.
27784 * doc/invoke.texi (-mptx): Update for new values and defaults.
27786 2022-02-22 Tom de Vries <tdevries@suse.de>
27788 * config/nvptx/nvptx.cc (gen_comment): New function.
27789 (workaround_uninit_method_1, workaround_uninit_method_2)
27790 (workaround_uninit_method_3): : Use gen_comment.
27791 * config/nvptx/nvptx.opt (mptx-comment): New option.
27793 2022-02-22 Richard Biener <rguenther@suse.de>
27795 * tree-vect-slp.cc (vect_build_slp_tree_2): Dump the def used
27798 2022-02-22 Roger Sayle <roger@nextmovesoftware.com>
27799 Richard Biener <rguenther@suse.de>
27801 * fold-const.cc (ctor_single_nonzero_element): New function to
27802 return the single non-zero element of a (vector) constructor.
27803 * fold-const.h (ctor_single_nonzero_element): Prototype here.
27804 * match.pd (reduc (constructor@0)): Simplify reductions of a
27805 constructor containing a single non-zero element.
27806 (reduc (@0 op VECTOR_CST) -> (reduc @0) op CONST): Simplify
27807 reductions of vector operations of the same operator with
27808 constant vector operands.
27810 2022-02-22 Jakub Jelinek <jakub@redhat.com>
27812 PR tree-optimization/104604
27813 * gimple-range-fold.cc (adjust_imagpart_expr, adjust_realpart_expr):
27814 Only check if gimple_assign_rhs1 is COMPLEX_CST if
27815 gimple_assign_rhs_code is COMPLEX_CST.
27817 2022-02-22 Jakub Jelinek <jakub@redhat.com>
27820 * config/i386/i386-expand.cc (ix86_expand_copysign): Call force_reg
27821 on input operands before calling lowpart_subreg on it. For output
27822 operand, use a vmode pseudo as destination and then move its lowpart
27823 subreg into operands[0] if lowpart_subreg fails on dest.
27824 (ix86_expand_xorsign): Likewise.
27826 2022-02-22 Richard Biener <rguenther@suse.de>
27828 PR tree-optimization/104582
27830 * config/i386/i386.cc (ix86_vector_costs::add_stmt_cost):
27831 Cost GPR to vector register moves for integer vector construction.
27833 2022-02-22 Richard Biener <rguenther@suse.de>
27835 PR tree-optimization/104582
27836 * tree-vectorizer.h (stmt_info_for_cost::node): New field.
27837 (vector_costs::add_stmt_cost): Add SLP node parameter.
27838 (dump_stmt_cost): Likewise.
27839 (add_stmt_cost): Likewise, new overload and adjust.
27840 (add_stmt_costs): Adjust.
27841 (record_stmt_cost): New overload.
27842 * tree-vectorizer.cc (dump_stmt_cost): Dump the SLP node.
27843 (vector_costs::add_stmt_cost): Adjust.
27844 * tree-vect-loop.cc (vect_estimate_min_profitable_iters):
27846 * tree-vect-slp.cc (vect_prologue_cost_for_slp): Record
27847 the SLP node for costing.
27848 (vectorizable_slp_permutation): Likewise.
27849 * tree-vect-stmts.cc (record_stmt_cost): Adjust and add
27851 * config/i386/i386.cc (ix86_vector_costs::add_stmt_cost):
27853 * config/aarch64/aarch64.cc (aarch64_vector_costs::add_stmt_cost):
27855 * config/rs6000/rs6000.cc (rs6000_vector_costs::add_stmt_cost):
27857 (rs6000_cost_data::adjust_vect_cost_per_loop): Likewise.
27859 2022-02-22 Richard Biener <rguenther@suse.de>
27861 PR tree-optimization/104582
27862 * tree-vectorizer.h (add_stmt_cost): New overload.
27863 (record_stmt_cost): Likewise.
27864 * tree-vect-loop.cc (vect_compute_single_scalar_iteration_cost):
27865 Use add_stmt_costs.
27866 (vect_get_known_peeling_cost): Use new overloads.
27867 (vect_estimate_min_profitable_iters): Likewise. Consistently
27868 use scalar_stmt for costing versioning checks.
27869 * tree-vect-stmts.cc (record_stmt_cost): New overload.
27871 2022-02-22 Hongyu Wang <hongyu.wang@intel.com>
27874 * config/i386/i386-expand.cc (ix86_expand_atomic_fetch_op_loop):
27875 Split atomic fetch and loop part.
27876 (ix86_expand_cmpxchg_loop): New expander for cmpxchg loop.
27877 * config/i386/i386-protos.h (ix86_expand_cmpxchg_loop): New
27879 * config/i386/sync.md (atomic_compare_and_swap<mode>): Call new
27880 expander under TARGET_RELAX_CMPXCHG_LOOP.
27881 (atomic_compare_and_swap<mode>): Likewise for doubleword modes.
27883 2022-02-21 Dan Li <ashimida@linux.alibaba.com>
27885 * config/aarch64/aarch64.cc (SLOT_REQUIRED):
27886 Change wb_candidate[12] to wb_push_candidate[12].
27887 (aarch64_layout_frame): Likewise, and
27888 change callee_adjust when scs is enabled.
27889 (aarch64_save_callee_saves):
27890 Change wb_candidate[12] to wb_push_candidate[12].
27891 (aarch64_restore_callee_saves):
27892 Change wb_candidate[12] to wb_pop_candidate[12].
27893 (aarch64_get_separate_components):
27894 Change wb_candidate[12] to wb_push_candidate[12].
27895 (aarch64_expand_prologue): Push x30 onto SCS before it's
27897 (aarch64_expand_epilogue): Pop x30 frome SCS, while
27898 preventing it from being popped from the regular stack again.
27899 (aarch64_override_options_internal): Add SCS compile option check.
27900 (TARGET_HAVE_SHADOW_CALL_STACK): New hook.
27901 * config/aarch64/aarch64.h (struct GTY): Add is_scs_enabled,
27902 wb_pop_candidate[12], and rename wb_candidate[12] to
27903 wb_push_candidate[12].
27904 * config/aarch64/aarch64.md (scs_push): New template.
27905 (scs_pop): Likewise.
27906 * doc/invoke.texi: Document -fsanitize=shadow-call-stack.
27907 * doc/tm.texi: Regenerate.
27908 * doc/tm.texi.in: Add hook have_shadow_call_stack.
27909 * flag-types.h (enum sanitize_code):
27910 Add SANITIZE_SHADOW_CALL_STACK.
27911 * opts.cc (parse_sanitizer_options): Add shadow-call-stack
27912 and exclude SANITIZE_SHADOW_CALL_STACK.
27913 * target.def: New hook.
27914 * toplev.cc (process_options): Add SCS compile option check.
27915 * ubsan.cc (ubsan_expand_null_ifn): Enum type conversion.
27917 2022-02-21 Tom de Vries <tdevries@suse.de>
27920 * config/nvptx/nvptx.cc (workaround_uninit_method_1)
27921 (workaround_uninit_method_2, workaround_uninit_method_3)
27922 (workaround_uninit): New function.
27923 (nvptx_reorg): Use workaround_uninit.
27924 * config/nvptx/nvptx.opt (minit-regs): New option.
27926 2022-02-21 Andre Vieira <andre.simoesdiasvieira@arm.com>
27928 PR rtl-optimization/104498
27929 * alias.cc (compare_base_symbol_refs): Correct distance computation
27930 when swapping x and y.
27932 2022-02-21 Andrew Pinski <apinski@marvell.com>
27935 * tree-ssa.cc (tree_ssa_useless_type_conversion):
27936 Check the inner type before calling useless_type_conversion_p.
27938 2022-02-19 Tom de Vries <tdevries@suse.de>
27940 * config/nvptx/nvptx.cc (nvptx_reorg_uniform_simt): Handle SET insn.
27941 * config/nvptx/nvptx.md
27942 (define_insn "nvptx_atomic_store<mode>"): Rename to ...
27943 (define_insn "nvptx_atomic_store_sm70<mode>"): This.
27944 (define_insn "nvptx_atomic_store<mode>"): New define_insn.
27945 (define_expand "atomic_store<mode>"): Handle rename. Use
27946 nvptx_atomic_store instead of atomic_exchange.
27948 2022-02-19 Tom de Vries <tdevries@suse.de>
27950 * config/nvptx/nvptx.cc (nvptx_reorg_uniform_simt): Handle all
27951 insns with atomic attribute. Assert that all handled insns are
27953 * config/nvptx/nvptx.md (define_insn "nvptx_atomic_store<mode>"):
27954 Set atomic attribute to false.
27956 2022-02-19 Tom de Vries <tdevries@suse.de>
27958 * config/nvptx/nvptx.cc (nvptx_unisimt_handle_set): Change return
27960 (nvptx_reorg_uniform_simt): Insert nvptx_uniform_warp_check or
27961 nvptx_warpsync, if necessary.
27963 2022-02-19 Jakub Jelinek <jakub@redhat.com>
27965 PR sanitizer/102656
27966 * asan.cc (instrument_derefs): If inner is a RESULT_DECL and access is
27967 known to be within bounds, treat it like automatic variables.
27968 If instrumenting access and inner is {VAR,PARM,RESULT}_DECL from
27969 current function and !TREE_STATIC which is not TREE_ADDRESSABLE, mark
27972 2022-02-18 Pat Haugen <pthaugen@linux.ibm.com>
27974 * config/rs6000/rs6000.opt (mpower10-fusion): Mark Undocumented.
27975 (mpower10-fusion-ld-cmpi, mpower10-fusion-2logical,
27976 mpower10-fusion-logical-add, mpower10-fusion-add-logical,
27977 mpower10-fusion-2add, mpower10-fusion-2store): Remove.
27978 * config/rs6000/rs6000-cpus.def (ISA_3_1_MASKS_SERVER,
27979 OTHER_P9_VECTOR_MASKS): Remove Power10 fusion sub-options.
27980 * config/rs6000/rs6000.cc (rs6000_option_override_internal,
27981 power10_sched_reorder): Likewise.
27982 * config/rs6000/genfusion.pl (gen_ld_cmpi_p10, gen_logical_addsubf,
27983 gen_addadd): Likewise
27984 * config/rs6000/fusion.md: Regenerate.
27986 2022-02-18 Jakub Jelinek <jakub@redhat.com>
27990 * config/rs6000/mm_malloc.h (_mm_malloc): Call posix_memalign
27991 rather than __posix_memalign.
27993 2022-02-18 Richard Biener <rguenther@suse.de>
27996 * config/i386/i386.cc (ix86_avx_u128_mode_source): Remove.
27997 (ix86_avx_u128_mode_needed): Return AVX_U128_DIRTY instead
27998 of calling ix86_avx_u128_mode_source which would eventually
27999 have returned AVX_U128_ANY in some very special case.
28001 2022-02-18 Richard Biener <rguenther@suse.de>
28003 PR tree-optimization/96881
28004 * tree-ssa-dce.cc (mark_stmt_if_obviously_necessary): Comment
28006 (control_parents_preserved_p): New function.
28007 (eliminate_unnecessary_stmts): Check that we preserved control
28008 parents before retaining a CLOBBER.
28009 (perform_tree_ssa_dce): Pass down aggressive flag
28010 to eliminate_unnecessary_stmts.
28012 2022-02-17 Jason Merrill <jason@redhat.com>
28014 * tree.cc (warn_deprecated_use): Look for TYPE_STUB_DECL
28015 on TYPE_MAIN_VARIANT.
28017 2022-02-17 Paul A. Clarke <pc@us.ibm.com>
28020 * config/rs6000/bmi2intrin.h: Uglify local variables.
28021 * config/rs6000/emmintrin.h: Likewise.
28022 * config/rs6000/mm_malloc.h: Likewise.
28023 * config/rs6000/mmintrin.h: Likewise.
28024 * config/rs6000/pmmintrin.h: Likewise.
28025 * config/rs6000/smmintrin.h: Likewise.
28026 * config/rs6000/tmmintrin.h: Likewise.
28027 * config/rs6000/xmmintrin.h: Likewise.
28029 2022-02-17 Robin Dapp <rdapp@linux.ibm.com>
28032 * config/rs6000/rs6000.cc (rs6000_emit_int_cmove): Return false
28033 if the expected comparison's first operand is of mode MODE_CC.
28035 2022-02-17 Vladimir N. Makarov <vmakarov@redhat.com>
28037 PR rtl-optimization/104447
28038 * lra-constraints.cc (spill_hard_reg_in_range): Initiate ignore
28039 hard reg set by lra_no_alloc_regs.
28041 2022-02-17 liuhongt <hongtao.liu@intel.com>
28043 PR tree-optimization/104551
28044 PR tree-optimization/103771
28045 * match.pd (cond_expr_convert_p): Add types_match check when
28046 convert is extension.
28047 * tree-vect-patterns.cc
28048 (gimple_cond_expr_convert_p): Adjust comments.
28049 (vect_recog_cond_expr_convert_pattern): Ditto.
28051 2022-02-17 Jakub Jelinek <jakub@redhat.com>
28054 * valtrack.cc (debug_lowpart_subreg): Don't call gen_rtx_raw_SUBREG
28055 if expr has VOIDmode.
28057 2022-02-17 liuhongt <hongtao.liu@intel.com>
28059 * config/i386/cpuid.h (bit_MPX): Removed.
28060 (bit_BNDREGS): Ditto.
28061 (bit_BNDCSR): Ditto.
28063 2022-02-17 Michael Meissner <meissner@the-meissners.org>
28066 * config/rs6000/rs6000-c.cc (rs6000_cpu_cpp_builtins): Define
28067 __SIZEOF_IBM128__ if the IBM 128-bit long double type is created.
28068 Define __SIZEOF_FLOAT128__ if the IEEE 128-bit floating point type
28071 2022-02-16 Andrew MacLeod <amacleod@redhat.com>
28073 * gimple-range-gori.cc (gori_compute::condexpr_adjust): Use
28074 range_compatible_p instead of direct type comparison.
28076 2022-02-16 Jakub Jelinek <jakub@redhat.com>
28078 PR rtl-optimization/104544
28079 * combine.cc (try_combine): When looking for insn whose links
28080 should be updated from i3 to i2, don't stop on debug insns, instead
28083 2022-02-16 Richard Sandiford <richard.sandiford@arm.com>
28086 * config/aarch64/iterators.md (LOGICAL_OR_PLUS): New iterator.
28087 * config/aarch64/aarch64.md: Extend the PR100056 patterns
28088 to handle plus in the same way as ior, if the operands have
28089 no set bits in common.
28091 2022-02-15 Andrew MacLeod <amacleod@redhat.com>
28093 PR tree-optimization/104526
28094 * gimple-range-fold.cc (fold_using_range::range_of_cond_expr): Call
28096 * gimple-range-gori.cc (range_def_chain::get_def_chain): Force a build
28097 of dependency chain if there isn't one.
28098 (gori_compute::condexpr_adjust): New.
28099 * gimple-range-gori.h (class gori_compute): New prototype.
28101 2022-02-15 Richard Sandiford <richard.sandiford@arm.com>
28104 * config/aarch64/aarch64-protos.h (aarch64_maxmin_plus_const):
28106 * config/aarch64/aarch64.cc (aarch64_maxmin_plus_const): New function.
28107 * config/aarch64/aarch64.md (*aarch64_minmax_plus): New pattern.
28109 2022-02-15 Richard Sandiford <richard.sandiford@arm.com>
28111 * tree-vectorizer.h (vect_scalar_ops_slice): New struct.
28112 (vect_scalar_ops_slice_hash): Likewise.
28113 (vect_scalar_ops_slice::op): New function.
28114 * tree-vect-slp.cc (vect_scalar_ops_slice::all_same_p): New function.
28115 (vect_scalar_ops_slice_hash::hash): Likewise.
28116 (vect_scalar_ops_slice_hash::equal): Likewise.
28117 (vect_prologue_cost_for_slp): Check for duplicate vectors.
28118 * config/aarch64/aarch64.cc
28119 (aarch64_vector_costs::m_stp_sequence_cost): New member variable.
28120 (aarch64_aligned_constant_offset_p): New function.
28121 (aarch64_stp_sequence_cost): Likewise.
28122 (aarch64_vector_costs::add_stmt_cost): Handle new STP heuristic.
28123 (aarch64_vector_costs::finish_cost): Likewise.
28125 2022-02-15 Richard Sandiford <richard.sandiford@arm.com>
28127 * tree-vect-slp.cc (vect_bb_vectorization_profitable_p): Fix
28130 2022-02-15 Richard Biener <rguenther@suse.de>
28132 PR tree-optimization/104543
28133 * gimple-loop-jam.cc (unroll_jam_possible_p): Check outer loop exits
28134 come after the inner loop.
28136 2022-02-15 Jakub Jelinek <jakub@redhat.com>
28139 * config/i386/host-cygwin.cc (cygwin_gt_pch_get_address): Use
28140 cannot instead of can%'t in diagnostics. Formatting fixes.
28142 2022-02-15 Jakub Jelinek <jakub@redhat.com>
28144 PR middle-end/104522
28145 * fold-const.h (native_interpret_real): Declare.
28146 * fold-const.cc (native_interpret_real): No longer static. Don't
28147 perform MODE_COMPOSITE_P verification here.
28148 (native_interpret_expr) <case REAL_TYPE>: But perform it here instead
28150 * gimple-fold.cc (clear_padding_type): Call native_interpret_real
28151 instead of native_interpret_expr.
28152 * simplify-rtx.cc (simplify_immed_subreg): Perform the native_encode_rtx
28153 and comparison verification for all FLOAT_MODE_P modes, not just
28156 2022-02-15 Richard Biener <rguenther@suse.de>
28158 PR tree-optimization/104519
28159 * fold-const.cc (multiple_of_p): Remove never true condition.
28160 * tree-ssa-loop-niter.cc (number_of_iterations_ne): Use
28161 the appropriate types for determining whether the difference
28162 of final and base is a multiple of the step.
28164 2022-02-15 Jakub Jelinek <jakub@redhat.com>
28167 * omp-low.cc (task_cpyfns): New variable.
28168 (delete_omp_context): Don't call finalize_task_copyfn from here.
28169 (create_task_copyfn): Push task_stmt into task_cpyfns.
28170 (execute_lower_omp): Call finalize_task_copyfn here on entries from
28171 task_cpyfns vector and release the vector.
28173 2022-02-14 Martin Sebor <msebor@redhat.com>
28175 PR middle-end/104355
28176 * doc/invoke.texi (-Warray-bounds): Update documentation.
28178 2022-02-14 Michael Meissner <meissner@the-meissners.org>
28181 * config/rs6000/rs6000.cc (init_float128_ibm): Update the
28182 conversion functions used to convert IFmode types.
28184 2022-02-14 Andrew Stubbs <ams@codesourcery.com>
28186 * config/gcn/gcn.cc (gcn_expand_reduc_scalar): Use force_reg.
28188 2022-02-14 Richard Biener <rguenther@suse.de>
28190 PR tree-optimization/104528
28191 * tree-ssa.h (find_released_ssa_name): Declare.
28192 * tree-ssa.cc (find_released_ssa_name): Export.
28193 * cfgloop.cc (verify_loop_structure): Look for released
28194 SSA names in loops nb_iterations.
28195 * tree-ssa-dse.cc (pass_dse::execute): Release number of iteration
28198 2022-02-14 Richard Biener <rguenther@suse.de>
28200 PR tree-optimization/104511
28201 * tree-ssa-forwprop.cc (simplify_vector_constructor): Avoid
28202 touching DFP <-> FP conversions.
28204 2022-02-14 Richard Biener <rguenther@suse.de>
28206 PR middle-end/104497
28207 * gimplify.cc (gimplify_compound_lval): Make sure the
28208 base is a non-register if needed and possible.
28210 2022-02-13 liuhongt <hongtao.liu@intel.com>
28213 * match.pd (cond_expr_convert_p): New match.
28214 * tree-vect-patterns.cc (gimple_cond_expr_convert_p): Declare.
28215 (vect_recog_cond_expr_convert_pattern): New.
28217 2022-02-12 Jakub Jelinek <jakub@redhat.com>
28219 PR sanitizer/104449
28220 * asan.cc: Include tree-eh.h.
28221 (handle_builtin_alloca): Handle the case when __builtin_alloca or
28222 __builtin_alloca_with_align can throw.
28224 2022-02-12 Jakub Jelinek <jakub@redhat.com>
28227 * config/i386/i386.md (cvtsd2ss splitter): If operands[1] is xmm16+
28228 and AVX512VL isn't available, move operands[1] to operands[0] first.
28230 2022-02-12 Uroš Bizjak <ubizjak@gmail.com>
28233 * config/i386/i386.cc (type_natural_mode):
28234 Skip decimal float vector modes.
28236 2022-02-11 Iain Sandoe <iain@sandoe.co.uk>
28237 Vladimir Makarov <vmakarov@redhat.com>
28240 * config/rs6000/rs6000.cc (darwin_rs6000_legitimate_lo_sum_const_p):
28241 Check for UNSPEC_MACHOPIC_OFFSET wrappers on symbolic addresses when
28243 (legitimate_lo_sum_address_p): Likewise.
28244 * lra-constraints.cc (process_address_1): Do not attempt to emit a reg
28245 load from an invalid lo_sum address.
28247 2022-02-11 Jakub Jelinek <jakub@redhat.com>
28249 PR tree-optimization/104499
28250 * match.pd ((X & Y) CMP 0 -> X CMP2 ~Y): Use view_convert instead
28253 2022-02-11 Jakub Jelinek <jakub@redhat.com>
28255 * tree.cc (build_common_builtin_nodes): Fix up formatting in
28256 __builtin_clear_padding decl creation.
28257 * gimplify.cc (gimple_add_padding_init_for_auto_var): Encode
28258 for_auto_init in the value of 2nd BUILT_IN_CLEAR_PADDING
28259 argument rather than in 3rd argument.
28260 (gimplify_call_expr): Likewise. Fix up comment formatting.
28261 * gimple-fold.cc (gimple_fold_builtin_clear_padding): Expect
28262 2 arguments instead of 3, take for_auto_init from the value
28265 2022-02-11 Vladimir N. Makarov <vmakarov@redhat.com>
28267 PR rtl-optimization/104400
28268 * lra-constraints.cc (process_alt_operands): Don't make union of
28269 this_alternative_exclude_start_hard_regs when reg class in insn
28270 alternative covers other reg classes in the same alternative.
28272 2022-02-11 Jakub Jelinek <jakub@redhat.com>
28274 PR middle-end/104446
28275 * combine.cc (subst): Don't substitute CONST_INTs into RTX_AUTOINC
28278 2022-02-11 Richard Biener <rguenther@suse.de>
28280 PR middle-end/104496
28281 * internal-fn.cc (vectorized_internal_fn_supported_p):
28282 Bail out for integer mode vector types.
28284 2022-02-11 Jakub Jelinek <jakub@redhat.com>
28286 PR rtl-optimization/104459
28287 * df-scan.cc (df_insn_change_bb): Don't call df_set_bb_dirty when
28288 moving DEBUG_INSNs between bbs.
28290 2022-02-11 liuhongt <hongtao.liu@intel.com>
28292 PR tree-optimization/104479
28293 * match.pd (uncond_op + vec_cond -> cond_op): Add single_use
28294 for the dest of uncond_op.
28296 2022-02-11 Tom de Vries <tdevries@suse.de>
28299 * config/nvptx/nvptx.cc (prevent_branch_around_nothing): Handle asm
28302 2022-02-10 Qing Zhao <qing.zhao@oracle.com>
28304 PR middle-end/100775
28305 * function.cc (gen_call_used_regs_seq): Call
28306 df_update_exit_block_uses when updating df.
28308 2022-02-10 Uroš Bizjak <ubizjak@gmail.com>
28311 * config/i386/sse.md (vec_unpacks_float_lo_v4si):
28312 Change operand 1 constraint to register_operand.
28314 2022-02-10 Richard Biener <rguenther@suse.de>
28316 PR tree-optimization/104373
28317 * tree-ssa-sccvn.h (do_rpo_vn): New export exposing the
28319 * tree-ssa-sccvn.cc (do_rpo_vn): Export, get the default
28320 walk kind as argument.
28321 (run_rpo_vn): Adjust.
28322 (pass_fre::execute): Likewise.
28323 * tree-ssa-uninit.cc (warn_uninitialized_vars): Skip
28324 blocks not reachable.
28325 (execute_late_warn_uninitialized): Mark all edges as
28327 (execute_early_warn_uninitialized): Use VN to compute
28329 (pass_data_early_warn_uninitialized): Enable a dump file,
28330 change dump name to warn_uninit.
28332 2022-02-10 Richard Biener <rguenther@suse.de>
28334 PR middle-end/104467
28335 * match.pd (vector extract simplification): Multiply the
28336 number of CTOR elements with the number of element elements.
28338 2022-02-10 Richard Biener <rguenther@suse.de>
28340 PR tree-optimization/104466
28341 * tree-ssa-alias.cc (refs_may_alias_p_2): Use rbase1/rbase2
28342 for the MR_DEPENDENCE checks as intended.
28344 2022-02-10 Tom de Vries <tdevries@suse.de>
28346 * config/nvptx/nvptx.md (define_insn "nvptx_atomic_store<mode>"): New
28348 (define_expand "atomic_store<mode>"): Use nvptx_atomic_store<mode> for
28350 (define_c_enum "unspecv"): Add UNSPECV_ST.
28352 2022-02-10 Tom de Vries <tdevries@suse.de>
28354 * config/nvptx/nvptx-protos.h (nvptx_mem_maybe_shared_p): Declare.
28355 * config/nvptx/nvptx.cc (nvptx_mem_data_area): New static function.
28356 (nvptx_mem_maybe_shared_p): New function.
28357 * config/nvptx/nvptx.md (define_expand "atomic_store<mode>"): New
28360 2022-02-10 Tom de Vries <tdevries@suse.de>
28363 * config/nvptx/nvptx.md (define_insn "sub<mode>3"): Workaround
28364 driver JIT bug by using sub.s16 instead of sub.u16.
28366 2022-02-10 Roger Sayle <roger@nextmovesoftware.com>
28368 * config/nvptx/nvptx.md (copysign<mode>3): Allow immediate
28369 floating point constants as operands 1 and/or 2.
28371 2022-02-10 Roger Sayle <roger@nextmovesoftware.com>
28374 * config/nvptx/nvptx.md (sel_true<mode>): Fix indentation.
28375 (sel_false<mode>): Likewise.
28376 (define_code_iterator eqne): New code iterator for EQ and NE.
28377 (*selp<mode>_neg_<code>): New define_insn_and_split to optimize
28378 the negation of a selp instruction.
28379 (*selp<mode>_not_<code>): New define_insn_and_split to optimize
28380 the bitwise not of a selp instruction.
28381 (*setcc_int<mode>): Use set instruction for neg:SI of a selp.
28383 2022-02-10 Roger Sayle <roger@nextmovesoftware.com>
28385 * config/nvptx/nvptx.md (any_logic): Move code iterator earlier
28386 in machine description.
28387 (logic): Move code attribute earlier in machine description.
28388 (ilogic): New code attribute, like logic but "ior" for IOR.
28389 (and<mode>3, ior<mode>3, xor<mode>3): Delete. Replace with...
28390 (<ilogic><mode>3): New define_insn for HSDIM logic operations.
28391 (<ilogic>bi3): New define_insn for BI mode logic operations.
28392 (define_split): Lower logic operations from integer modes to
28393 BI mode predicate operations.
28395 2022-02-10 Roger Sayle <roger@nextmovesoftware.com>
28397 * config/nvptx/nvptx.md (UNSPEC_ISINF): New UNSPEC.
28398 (one_cmplbi2): New define_insn for not.pred.
28399 (mulditi3): New define_expand for signed widening multiply.
28400 (umulditi3): New define_expand for unsigned widening multiply.
28401 (smul<mode>3_highpart): New define_insn for signed highpart mult.
28402 (umul<mode>3_highpart): New define_insn for unsigned highpart mult.
28403 (*smulhi3_highpart_2): Renamed from smulhi3_highpart.
28404 (*smulsi3_highpart_2): Renamed from smulsi3_highpart.
28405 (*umulhi3_highpart_2): Renamed from umulhi3_highpart.
28406 (*umulsi3_highpart_2): Renamed from umulsi3_highpart.
28407 (*setcc<mode>_from_not_bi): New define_insn.
28408 (*setcc_isinf<mode>): New define_insn for testp.infinite.
28409 (isinf<mode>2): New define_expand.
28411 2022-02-10 Roger Sayle <roger@nextmovesoftware.com>
28413 * config/nvptx/nvptx.md (cmp<mode>): Renamed from *cmp<mode>.
28414 (setcc<mode>_from_bi): Additionally support QImode.
28415 (extendbi<mode>2): Additionally support QImode.
28416 (zero_extendbi<mode>2): Additionally support QImode.
28417 (any_sbinary, any_ubinary, any_sunary, any_uunary): New code
28418 iterators for signed and unsigned, binary and unary operations.
28419 (<sbinary>qi3, <ubinary>qi3, <sunary>qi2, <uunary>qi2): New
28420 expanders to perform QImode operations using SImode instructions.
28421 (cstoreqi4): New define_expand.
28422 (*ext_truncsi2_qi): New define_insn.
28423 (*zext_truncsi2_qi): New define_insn.
28425 2022-02-10 Roger Sayle <roger@nextmovesoftware.com>
28427 * config/nvptx/nvptx.md (*cmpf): New define_insn.
28428 (cstorehf4): New define_expand.
28429 (fmahf4): New define_insn.
28430 (neghf2): New define_insn.
28431 (abshf2): New define_insn.
28433 2022-02-10 Gerald Pfeifer <gerald@pfeifer.com>
28435 * doc/install.texi (Specific): Change the www.bitwizard.nl
28436 reference to use https.
28438 2022-02-10 Marcel Vollweiler <marcel@codesourcery.com>
28440 * gimplify.cc (gimplify_scan_omp_clauses): Added cases for
28441 OMP_CLAUSE_HAS_DEVICE_ADDR
28442 and handle array sections.
28443 (gimplify_adjust_omp_clauses): Added OMP_CLAUSE_HAS_DEVICE_ADDR case.
28444 * omp-low.cc (scan_sharing_clauses): Handle OMP_CLAUSE_HAS_DEVICE_ADDR.
28445 (lower_omp_target): Same.
28446 * tree-core.h (enum omp_clause_code): Same.
28447 * tree-nested.cc (convert_nonlocal_omp_clauses): Same.
28448 (convert_local_omp_clauses): Same.
28449 * tree-pretty-print.cc (dump_omp_clause): Same.
28452 2022-02-10 Eugene Rozenfeld <erozen@microsoft.com>
28454 * auto-profile.cc (afdo_indirect_call): Don't attempt to promote indirect calls
28455 that will result in direct recursive calls.
28457 2022-02-10 Andrew Pinski <apinski@marvell.com>
28460 * config/aarch64/aarch64.cc
28461 (aarch64_sve_expand_vector_init_handle_trailing_constants):
28462 Use CONST0_RTX instead of const0_rtx for the non-constant elements.
28464 2022-02-09 Uroš Bizjak <ubizjak@gmail.com>
28467 * common/config/i386/i386-common.cc (OPTION_MASK_ISA2_XSAVE_UNSET):
28468 Also include OPTION_MASK_ISA2_AVX2_UNSET.
28470 2022-02-09 Uroš Bizjak <ubizjak@gmail.com>
28473 * config/i386/i386-expand.cc (ix86_split_idivmod):
28474 Force operands[2] and operands[3] into a register..
28476 2022-02-09 Jeff Law <jeffreyalaw@gmail.com>
28479 * config/v850/v850.md (*v850_fnmasf4): Renamed from fnmasf4.
28480 (*v850_fnmssf4): Renamed from fnmssf4
28482 2022-02-09 Ian Lance Taylor <iant@golang.org>
28484 * godump.cc (go_force_record_alignment): Really name the alignment
28485 field "_" (complete 2021-12-29 change).
28487 2022-02-09 Bill Schmidt <wschmidt@linux.ibm.com>
28489 * config/rs6000/rs6000-builtins.def (VREPLACE_UN_UV2DI): Change
28490 function prototype.
28491 (VREPLACE_UN_UV4SI): Likewise.
28492 (VREPLACE_UN_V2DF): Likewise.
28493 (VREPLACE_UN_V2DI): Likewise.
28494 (VREPLACE_UN_V4SF): Likewise.
28495 (VREPLACE_UN_V4SI): Likewise.
28496 * config/rs6000/rs6000-overload.def (VEC_REPLACE_UN): Change all
28497 function prototypes.
28498 * config/rs6000/vsx.md (vreplace_un_<mode>): Remove define_expand.
28499 (vreplace_un_<mode>): New define_insn.
28501 2022-02-09 Richard Sandiford <richard.sandiford@arm.com>
28503 * config/aarch64/iterators.md (VDCSIF): New mode iterator.
28505 (single_wx, single_type, single_dtype, dblq): New mode attributes.
28506 * config/aarch64/aarch64-simd.md (load_pair_lanes<mode>): Extend
28507 from VDC to VDCSIF.
28508 (store_pair_lanes<mode>): Likewise.
28509 (*aarch64_combine_internal<mode>): Likewise.
28510 (*aarch64_combine_internal_be<mode>): Likewise.
28511 (*aarch64_combinez<mode>): Likewise.
28512 (*aarch64_combinez_be<mode>): Likewise.
28513 * config/aarch64/aarch64.cc (aarch64_classify_address): Handle
28514 8-byte modes for ADDR_QUERY_LDP_STP_N.
28515 (aarch64_print_operand): Likewise for %y.
28517 2022-02-09 Richard Sandiford <richard.sandiford@arm.com>
28519 * config/aarch64/aarch64-simd.md (@aarch64_split_simd_mov<mode>):
28520 Use aarch64_combine instead of move_lo/hi_quad. Tabify.
28521 (move_lo_quad_<mode>, aarch64_simd_move_hi_quad_<mode>): Delete.
28522 (aarch64_simd_move_hi_quad_be_<mode>, move_hi_quad_<mode>): Delete.
28523 (vec_pack_trunc_<mode>): Take general_operand elements and use
28524 aarch64_combine rather than move_lo/hi_quad to combine them.
28525 (vec_pack_trunc_df): Likewise.
28527 2022-02-09 Richard Sandiford <richard.sandiford@arm.com>
28529 * config/aarch64/aarch64-protos.h (aarch64_split_simd_combine):
28531 * config/aarch64/aarch64-simd.md (@aarch64_combinez<mode>): Rename
28533 (*aarch64_combinez<mode>): ...this.
28534 (@aarch64_combinez_be<mode>): Rename to...
28535 (*aarch64_combinez_be<mode>): ...this.
28536 (@aarch64_vec_concat<mode>): New expander.
28537 (aarch64_combine<mode>): Use it.
28538 (@aarch64_simd_combine<mode>): Delete.
28539 * config/aarch64/aarch64.cc (aarch64_split_simd_combine): Delete.
28540 (aarch64_expand_vector_init): Use aarch64_vec_concat.
28542 2022-02-09 Richard Sandiford <richard.sandiford@arm.com>
28544 * config/aarch64/predicates.md (aarch64_reg_or_mem_pair_operand):
28546 * config/aarch64/aarch64-simd.md (*aarch64_combine_internal<mode>)
28547 (*aarch64_combine_internal_be<mode>): New patterns.
28549 2022-02-09 Richard Sandiford <richard.sandiford@arm.com>
28551 * config/aarch64/aarch64-simd.md (move_lo_quad_internal_<mode>)
28552 (move_lo_quad_internal_be_<mode>): Delete.
28553 (move_lo_quad_<mode>): Use aarch64_combine<Vhalf> instead of the above.
28555 2022-02-09 Richard Sandiford <richard.sandiford@arm.com>
28557 * config/aarch64/aarch64-protos.h (aarch64_mergeable_load_pair_p):
28559 * config/aarch64/aarch64-simd.md (load_pair_lanes<mode>): Use
28560 aarch64_mergeable_load_pair_p instead of inline check.
28561 * config/aarch64/aarch64.cc (aarch64_expand_vector_init): Likewise.
28562 (aarch64_check_consecutive_mems): Allow the reversed parameter
28564 (aarch64_mergeable_load_pair_p): New function.
28566 2022-02-09 Richard Sandiford <richard.sandiford@arm.com>
28568 * config/aarch64/aarch64-simd.md (vec_set<mode>): Allow the
28569 element to be an aarch64_simd_nonimmediate_operand.
28571 2022-02-09 Richard Sandiford <richard.sandiford@arm.com>
28573 * config/aarch64/aarch64-simd.md (aarch64_simd_vec_set<mode>): Use
28574 aarch64_simd_nonimmediate_operand instead of
28575 aarch64_simd_general_operand.
28576 (@aarch64_combinez<mode>): Use nonimmediate_operand instead of
28578 (@aarch64_combinez_be<mode>): Likewise.
28580 2022-02-09 Richard Biener <rguenther@suse.de>
28582 PR middle-end/104464
28583 * gimple-isel.cc (gimple_expand_vec_cond_expr): Postpone
28584 throwing check to after unproblematic replacement.
28586 2022-02-09 Roger Sayle <roger@nextmovesoftware.com>
28588 PR tree-optimization/104420
28589 * match.pd (mult @0 real_zerop): Tweak conditions for constant
28590 folding X*0.0 (or X*-0.0) to HONOR_SIGNED_ZEROS when appropriate.
28592 2022-02-09 Jakub Jelinek <jakub@redhat.com>
28595 * dwarf2out.cc (mangle_referenced_decls): New function.
28596 (tree_add_const_value_attribute): Don't call rtl_for_decl_init if
28597 early_dwarf. Instead walk the initializer and try to mangle vars or
28598 functions referenced from it.
28600 2022-02-09 Andrew MacLeod <amacleod@redhat.com>
28602 PR tree-optimization/104288
28603 * gimple-range-cache.cc (non_null_ref::set_nonnull): New.
28604 (non_null_ref::adjust_range): Move to header.
28605 (ranger_cache::range_of_def): Don't check non-null.
28606 (ranger_cache::entry_range): Don't check non-null.
28607 (ranger_cache::range_on_edge): Check for nonnull on normal edges.
28608 (ranger_cache::update_to_nonnull): New.
28609 (non_null_loadstore): New.
28610 (ranger_cache::block_apply_nonnull): New.
28611 * gimple-range-cache.h (class non_null_ref): Update prototypes.
28612 (non_null_ref::adjust_range): Move to here and inline.
28613 (class ranger_cache): Update prototypes.
28614 * gimple-range-path.cc (path_range_query::range_defined_in_block): Do
28615 not search dominators.
28616 (path_range_query::adjust_for_non_null_uses): Ditto.
28617 * gimple-range.cc (gimple_ranger::range_of_expr): Check on-entry for
28618 def overrides. Do not check nonnull.
28619 (gimple_ranger::range_on_entry): Check dominators for nonnull.
28620 (gimple_ranger::range_on_edge): Check for nonnull on normal edges..
28621 (gimple_ranger::register_side_effects): New.
28622 * gimple-range.h (gimple_ranger::register_side_effects): New.
28623 * tree-vrp.cc (rvrp_folder::fold_stmt): Call register_side_effects.
28625 2022-02-09 Richard Biener <rguenther@suse.de>
28627 PR tree-optimization/104445
28628 PR tree-optimization/102832
28629 * optabs-query.h (can_vec_extract): New.
28630 * optabs-query.cc (can_vec_extract): Likewise.
28631 * tree-vect-loop.cc (vect_find_reusable_accumulator): Check
28632 we can extract a hi/lo part from the larger vector, rework
28633 check iteration from larger to smaller sizes.
28635 2022-02-09 H.J. Lu <hjl.tools@gmail.com>
28639 * config/i386/gnu-property.cc: Include "i386-protos.h".
28640 (file_end_indicate_exec_stack_and_gnu_property): Generate
28641 a GNU_PROPERTY_1_NEEDED note for -mno-direct-extern-access or
28642 nodirect_extern_access attribute.
28643 * config/i386/i386-options.cc
28644 (handle_nodirect_extern_access_attribute): New function.
28645 (ix86_attribute_table): Add nodirect_extern_access attribute.
28646 * config/i386/i386-protos.h (ix86_force_load_from_GOT_p): Add a
28648 (ix86_has_no_direct_extern_access): New.
28649 * config/i386/i386.cc (ix86_has_no_direct_extern_access): New.
28650 (ix86_force_load_from_GOT_p): Add a bool argument to indicate
28651 call operand. Force non-call load from GOT for
28652 -mno-direct-extern-access or nodirect_extern_access attribute.
28653 (legitimate_pic_address_disp_p): Avoid copy relocation in PIE
28654 for -mno-direct-extern-access or nodirect_extern_access attribute.
28655 (ix86_print_operand): Pass true to ix86_force_load_from_GOT_p
28657 (asm_preferred_eh_data_format): Use PC-relative format for
28658 -mno-direct-extern-access to avoid copy relocation. Check
28659 ptr_mode instead of TARGET_64BIT when selecting DW_EH_PE_sdata4.
28660 (ix86_binds_local_p): Set ix86_has_no_direct_extern_access to
28661 true for -mno-direct-extern-access or nodirect_extern_access
28662 attribute. Don't treat protected data as extern and avoid copy
28663 relocation on common symbol with -mno-direct-extern-access or
28664 nodirect_extern_access attribute.
28665 (ix86_reloc_rw_mask): New to avoid copy relocation for
28666 -mno-direct-extern-access.
28667 (TARGET_ASM_RELOC_RW_MASK): New.
28668 * config/i386/i386.opt: Add -mdirect-extern-access.
28669 * doc/extend.texi: Document nodirect_extern_access attribute.
28670 * doc/invoke.texi: Document -m[no-]direct-extern-access.
28672 2022-02-09 H.J. Lu <hjl.tools@gmail.com>
28675 * config/i386/i386.cc (ix86_avx_u128_mode_source): New function.
28676 (ix86_avx_u128_mode_needed): Return AVX_U128_ANY for debug INSN.
28677 Call ix86_avx_u128_mode_source to check mode for each component
28680 2022-02-09 liuhongt <hongtao.liu@intel.com>
28683 * config/i386/sse.md (<insn><mode>3): lowpart_subreg
28684 operands[2] from SImode to QImode.
28686 2022-02-09 Richard Biener <rguenther@suse.de>
28688 PR middle-end/104450
28689 * gimple-isel.cc: Pass cfun around.
28690 (+gimple_expand_vec_cond_expr): Do not combine a throwing
28691 comparison with the select.
28693 2022-02-09 Richard Biener <rguenther@suse.de>
28696 * config/i386/i386.cc (ix86_gimple_fold_builtin): Guard shift
28697 folding for NULL LHS.
28699 2022-02-08 Robin Dapp <rdapp@linux.ibm.com>
28701 PR rtl-optimization/104198
28702 PR rtl-optimization/104153
28703 * ifcvt.cc (noce_convert_multiple_sets_1): Copy rtx instead of
28704 using it directly. Rework comparison handling and always
28705 perform a second pass.
28707 2022-02-08 Jakub Jelinek <jakub@redhat.com>
28710 * config/rs6000/rs6000.cc (vspltis_shifted): Return false also if
28711 split1 pass has finished already.
28713 2022-02-08 Bill Schmidt <wschmidt@linux.ibm.com>
28715 * config/rs6000/rs6000-builtins.def (VMSUMCUD): New.
28716 * config/rs6000/rs6000-overload.def (VEC_MSUMC): New.
28717 * config/rs6000/vsx.md (UNSPEC_VMSUMCUD): New constant.
28718 (vmsumcud): New define_insn.
28720 2022-02-08 Tom de Vries <tdevries@suse.de>
28722 * config/nvptx/nvptx-opts.h (enum ptx_isa): Add PTX_ISA_SM70.
28723 * config/nvptx/nvptx.h (TARGET_SM70): Define.
28725 2022-02-08 Robin Dapp <rdapp@linux.ibm.com>
28727 * config/s390/s390.cc (s390_rtx_costs): Increase costs for load
28729 * config/s390/s390.md: Use paradoxical subreg.
28731 2022-02-08 Robin Dapp <rdapp@linux.ibm.com>
28733 * combine.cc (reg_subword_p): Check for paradoxical subreg.
28735 2022-02-08 Tom de Vries <tdevries@suse.de>
28738 * config/nvptx/nvptx-opts.h (enum ptx_version): Add PTX_VERSION_3_0
28739 and PTX_VERSION_4_2.
28740 * config/nvptx/nvptx.cc (first_ptx_version_supporting_sm)
28741 (default_ptx_version_option, ptx_version_to_string)
28742 (sm_version_to_string, handle_ptx_version_option): New function.
28743 (nvptx_option_override): Call handle_ptx_version_option.
28744 (nvptx_file_start): Use ptx_version_to_string and sm_version_to_string.
28745 * config/nvptx/nvptx.md (define_insn "nvptx_shuffle<mode>")
28746 (define_insn "nvptx_vote_ballot"): Use TARGET_PTX_6_0.
28747 * config/nvptx/nvptx.opt (mptx): Remove 'Init'.
28749 2022-02-08 Maciej W. Rozycki <macro@embecosm.com>
28751 * doc/install.texi (Configuration): Document `--with-isa-spec='
28753 * doc/invoke.texi (Option Summary): List `-misa-spec=' RISC-V
28755 (RISC-V Options): Document it.
28757 2022-02-08 Maciej W. Rozycki <macro@embecosm.com>
28759 * config/riscv/t-riscv (riscv-sr.o): Add $(TM_H) dependency.
28761 2022-02-08 Tom de Vries <tdevries@suse.de>
28763 * config/nvptx/nvptx.cc (write_fn_proto_1): Handle 'main (int)'.
28765 2022-02-08 Tom de Vries <tdevries@suse.de>
28768 * config/nvptx/nvptx-protos.h (nvptx_mem_local_p): Declare.
28769 * config/nvptx/nvptx.cc (nvptx_reorg_uniform_simt): Assert that
28770 change is validated.
28771 (nvptx_mem_local_p): New function.
28772 * config/nvptx/nvptx.md: Use nvptx_mem_local_p.
28773 (define_c_enum "unspecv"): Add UNSPECV_CAS_LOCAL.
28774 (define_insn "atomic_compare_and_swap<mode>_1_local"): New
28775 non-atomic, non-predicable define_insn, factored out of ...
28776 (define_insn "atomic_compare_and_swap<mode>_1"): ... here.
28777 Make predicable again.
28778 (define_expand "atomic_compare_and_swap<mode>"): Use
28779 atomic_compare_and_swap<mode>_1_local.
28781 2022-02-08 liuhongt <hongtao.liu@intel.com>
28783 PR rtl-optimization/104059
28784 * regcprop.cc (copyprop_hardreg_forward_1): Don't propagate
28785 for a more expensive reg-reg move.
28787 2022-02-07 Tamar Christina <tamar.christina@arm.com>
28789 * config/arm/arm_neon.h (vusdotq_s32, vusdot_laneq_s32,
28790 vusdotq_laneq_s32, vsudot_laneq_s32, vsudotq_laneq_s32): New
28791 * config/arm/arm_neon_builtins.def (usdot): Add V16QI.
28792 (usdot_laneq, sudot_laneq): New.
28793 * config/arm/neon.md (neon_<sup>dot_laneq<vsi2qi>): New.
28794 (neon_<sup>dot_lane<vsi2qi>): Remote unneeded code.
28796 2022-02-07 Tamar Christina <tamar.christina@arm.com>
28798 * config/arm/arm_neon.h (vdot_laneq_u32, vdotq_laneq_u32,
28799 vdot_laneq_s32, vdotq_laneq_s32): New.
28800 * config/arm/arm_neon_builtins.def (sdot_laneq, udot_laneq): New.
28801 * config/arm/neon.md (neon_<sup>dot<vsi2qi>): New.
28802 (<sup>dot_prod<vsi2qi>): Re-order rtl.
28803 (neon_<sup>dot_lane<vsi2qi>): Fix rtl order and endiannes.
28804 (neon_<sup>dot_laneq<vsi2qi>): New.
28806 2022-02-07 Andreas Krebbel <krebbel@linux.ibm.com>
28809 * config/s390/s390.cc (s390_can_inline_p): Accept a few more flags
28810 if always_inline is set. Don't inline when tune differs without
28813 2022-02-07 Richard Biener <rguenther@suse.de>
28815 PR middle-end/104402
28816 * gimple-expr.cc (is_gimple_condexpr): _Complex typed
28817 compares are not valid.
28818 * tree-cfg.cc (verify_gimple_assign_ternary): For COND_EXPR
28819 check is_gimple_condexpr.
28821 2022-02-07 Kewen Lin <linkw@linux.ibm.com>
28824 * config/rs6000/rs6000.cc (rs6000_option_override_internal): Move the
28825 hunk affecting VSX and ALTIVEC to appropriate place.
28827 2022-02-07 Kewen Lin <linkw@linux.ibm.com>
28830 * config/rs6000/rs6000.cc (rs6000_option_override_internal): Disable
28831 MMA if !TARGET_VSX.
28833 2022-02-06 Jakub Jelinek <jakub@redhat.com>
28837 * fold-const.h (folding_initializer): Adjust comment.
28838 (folding_cxx_constexpr): Declare.
28839 * fold-const.cc (folding_initializer): Adjust comment.
28840 (folding_cxx_constexpr): New variable.
28841 (address_compare): Restrict the decl vs. STRING_CST
28842 or vice versa or STRING_CST vs. STRING_CST or
28843 is_global_var != is_global_var optimizations to !folding_cxx_constexpr.
28844 Punt for FUNCTION_DECLs with non-zero offsets. If folding_initializer,
28845 assume non-aliased functions have non-zero size and have different
28846 addresses. For folding_cxx_constexpr, punt on comparisons of start
28847 of some object and end of another one, regardless whether it is a decl
28848 or string literal. Also punt for folding_cxx_constexpr on
28849 STRING_CST vs. STRING_CST comparisons if the two literals could be
28852 2022-02-05 Jakub Jelinek <jakub@redhat.com>
28854 PR tree-optimization/104389
28855 * match.pd (x * 0 -> 0): Punt if x maybe infinite and NaNs are
28858 2022-02-05 Kito Cheng <kito.cheng@sifive.com>
28860 * configure.ac: Fix detection for zifencei support.
28861 * configure: Regenerate.
28863 2022-02-05 Kito Cheng <kito.cheng@sifive.com>
28866 * config.gcc (riscv*-*-*): Normalize the with_isa_spec value.
28867 (all_defaults): Add isa_spec.
28868 * config/riscv/riscv.h (OPTION_DEFAULT_SPECS): Add isa_spec.
28870 2022-02-04 Bill Schmidt <wschmidt@linux.ibm.com>
28872 * config/rs6000/rs6000-c.cc (resolve_vec_mul): Accept args and types
28873 parameters instead of arglist and nargs. Simplify accordingly. Remove
28874 unnecessary test for argument count mismatch.
28875 (resolve_vec_cmpne): Likewise.
28876 (resolve_vec_adde_sube): Likewise.
28877 (resolve_vec_addec_subec): Likewise.
28878 (altivec_resolve_overloaded_builtin): Move overload special handling
28879 after the gathering of arguments into args[] and types[] and the test
28880 for correct number of arguments. Don't perform the test for correct
28881 number of arguments for certain special cases. Call the other special
28882 cases with args and types instead of arglist and nargs.
28884 2022-02-04 Bill Schmidt <wschmidt@linux.ibm.com>
28887 * doc/extend.texi (Basic PowerPC Built-in Functions Available on ISA
28888 3.1): Provide consistent type names. Remove unnecessary semicolons.
28889 Fix bad line breaks.
28891 2022-02-04 Jakub Jelinek <jakub@redhat.com>
28894 * config/rs6000/rs6000.cc (rs6000_mangle_decl_assembler_name): Also
28895 adjust mangling of __builtin*printf_chk.
28897 2022-02-04 Jonathan Wakely <jwakely@redhat.com>
28899 * doc/cpp.texi (Variadic Macros): Replace C++2a with C++20.
28901 2022-02-04 Richard Biener <rguenther@suse.de>
28902 Bin Cheng <bin.cheng@linux.alibaba.com>
28904 PR tree-optimization/100499
28905 * fold-const.h (multiple_of_p): Add nowrap parameter, defaulted
28907 * fold-const.cc (multiple_of_p): Likewise. Honor it for
28908 MULT_EXPR, PLUS_EXPR and MINUS_EXPR and pass it along,
28909 switching to false for conversions.
28910 * tree-ssa-loop-niter.cc (number_of_iterations_ne): Do not
28911 claim the outermost expression does not wrap when calling
28912 multiple_of_p. Refactor the check done to check the
28913 original IV, avoiding a bias that might wrap.
28915 2022-02-04 Richard Biener <rguenther@suse.de>
28917 * fold-const.cc (multiple_of_p): Re-write and move LSHIFT_EXPR
28920 2022-02-04 Eric Botcazou <ebotcazou@adacore.com>
28923 * dwarf2out.cc (dwarf2out_finish): Empty base_types.
28924 (dwarf2out_early_finish): Likewise.
28926 2022-02-04 Eric Botcazou <ebotcazou@adacore.com>
28928 PR tree-optimization/104356
28929 * match.pd (X / bool_range_Y is X): Add guard.
28930 (X / X is one): Likewise.
28931 (X / abs (X) is X < 0 ? -1 : 1): Likewise.
28932 (X / -X is -1): Likewise.
28933 (1 / X -> X == 1): Likewise.
28935 2022-02-04 Richard Biener <rguenther@suse.de>
28937 PR tree-optimization/103641
28938 * tree-vect-patterns.cc (vect_synth_mult_by_constant):
28939 Pass the vector mode to choose_mult_variant.
28941 2022-02-04 Roger Sayle <roger@nextmovesoftware.com>
28943 PR rtl-optimization/101885
28944 * combine.cc (try_combine): When splitting a parallel into two
28945 sequential sets, check not only that the first doesn't clobber
28946 the second but also that the second doesn't clobber the first.
28948 2022-02-04 Richard Biener <rguenther@suse.de>
28950 PR middle-end/90348
28951 PR middle-end/104092
28952 * tree-core.h (clobber_kind): New enum.
28953 (tree_base::u::bits::address_space): Document use in CONSTRUCTORs.
28954 * tree.h (CLOBBER_KIND): Add.
28955 (build_clobber): Add clobber kind argument, defaulted to
28957 * tree.cc (build_clobber): Likewise.
28958 * gimple.h (gimple_clobber_p): New overload with specified kind.
28959 * tree-streamer-in.cc (streamer_read_tree_bitfields): Stream
28961 * tree-streamer-out.cc (streamer_write_tree_bitfields):
28963 * tree-pretty-print.cc (dump_generic_node): Mark EOL CLOBBERs.
28964 * gimplify.cc (gimplify_bind_expr): Build storage end-of-life clobbers
28966 (gimplify_target_expr): Likewise.
28967 * tree-inline.cc (expand_call_inline): Likewise.
28968 * tree-ssa-ccp.cc (insert_clobber_before_stack_restore): Likewise.
28969 * gimple-ssa-warn-access.cc (pass_waccess::check_stmt): Only treat
28970 CLOBBER_EOL clobbers as ending lifetime of storage.
28972 2022-02-04 Martin Sebor <msebor@redhat.com>
28974 * pointer-query.h (pointer_query::cache_type): Use auto_vec for auto
28977 2022-02-03 Martin Sebor <msebor@redhat.com>
28979 PR middle-end/104260
28980 * passes.def (pass_warn_access): Adjust pass placement.
28982 2022-02-03 Uroš Bizjak <ubizjak@gmail.com>
28985 * config/i386/i386.cc (find_drap_reg): For 32bit targets
28986 return DI_REG if function uses __builtin_eh_return.
28988 2022-02-03 Martin Sebor <msebor@redhat.com>
28990 * gimple-ssa-warn-restrict.cc (class pass_wrestrict): Outline ctor.
28991 (pass_wrestrict::m_ptr_qry): New member.
28992 (wrestrict_walk): Rename...
28993 (pass_wrestrict::check_block): ...to this.
28994 (pass_wrestrict::execute): Set up and tear down pointer_query and
28996 (builtin_memref::builtin_memref): Change ctor argument. Simplify.
28997 (builtin_access::builtin_access): Same.
28998 (builtin_access::m_ptr_qry): New member.
28999 (check_call): Rename...
29000 (pass_wrestrict::check_call): ...to this.
29001 (check_bounds_or_overlap): Change argument.
29002 * gimple-ssa-warn-restrict.h (check_bounds_or_overlap): Same.
29004 2022-02-03 Martin Sebor <msebor@redhat.com>
29006 * gimple-array-bounds.cc (array_bounds_checker::array_bounds_checker):
29008 (array_bounds_checker::get_value_range): Use new member.
29009 (array_bounds_checker::check_mem_ref): Same.
29010 * gimple-array-bounds.h (array_bounds_checker::array_bounds_checker):
29012 (array_bounds_checker::m_ptr_query): New member.
29014 2022-02-03 Martin Sebor <msebor@redhat.com>
29016 * gimple-ssa-warn-access.cc (pass_waccess::pass_waccess): Remove
29017 pointer_query cache.
29018 * pointer-query.cc (pointer_query::pointer_query): Remove cache
29019 argument. Zero-initialize new cache member.
29020 (pointer_query::get_ref): Replace cache pointer with direct access.
29021 (pointer_query::put_ref): Same.
29022 (pointer_query::flush_cache): Same.
29023 (pointer_query::dump): Same.
29024 * pointer-query.h (class pointer_query): Remove cache argument from
29025 ctor. Change cache pointer to cache subobject member.
29026 * tree-ssa-strlen.cc: Remove pointer_query cache.
29028 2022-02-03 Martin Sebor <msebor@redhat.com>
29030 PR tree-optimization/104119
29031 * gimple-ssa-sprintf.cc (struct directive): Change argument type.
29032 (format_none): Same.
29033 (format_percent): Same.
29034 (format_integer): Same.
29035 (format_floating): Same.
29036 (get_string_length): Same.
29037 (format_character): Same.
29038 (format_string): Same.
29039 (format_plain): Same.
29040 (format_directive): Same.
29041 (compute_format_length): Same.
29042 (handle_printf_call): Same.
29043 * tree-ssa-strlen.cc (get_range_strlen_dynamic): Same. Call
29045 (get_range_strlen_phi): Same.
29046 (get_maxbound): New function.
29047 (strlen_pass::get_len_or_size): Adjust to parameter change.
29048 * tree-ssa-strlen.h (get_range_strlen_dynamic): Change argument type.
29050 2022-02-03 Bill Schmidt <wschmidt@linux.ibm.com>
29053 * config/rs6000/rs6000-builtin.cc (rs6000_gimple_fold_builtin): Remove
29054 test for !rs6000_fold_gimple.
29055 * config/rs6000/rs6000.cc (rs6000_option_override_internal): Likewise.
29056 * config/rs6000/rs6000.opt (mfold-gimple): Remove.
29058 2022-02-03 Bill Schmidt <wschmidt@linux.ibm.com>
29061 * config/rs6000/rs6000-builtin.cc (rs6000_expand_builtin): Handle
29062 endianness for vclzlsbb and vctzlsbb.
29063 * config/rs6000/rs6000-builtins.def (VCLZLSBB_V16QI): Change
29064 default pattern and indicate a different pattern will be used for
29066 (VCLZLSBB_V4SI): Likewise.
29067 (VCLZLSBB_V8HI): Likewise.
29068 (VCTZLSBB_V16QI): Likewise.
29069 (VCTZLSBB_V4SI): Likewise.
29070 (VCTZLSBB_V8HI): Likewise.
29072 2022-02-03 Bill Schmidt <wschmidt@linux.ibm.com>
29074 * config.gcc (powerpc*-*-*): Add rs6000-builtin.o to extra_objs.
29075 * config/rs6000/rs6000-builtin.cc: New file, containing code moved
29077 * config/rs6000/rs6000-call.cc (cpu_is_info): Move to
29079 (cpu_supports_info): Likewise.
29080 (rs6000_type_string): Likewise.
29081 (altivec_expand_predicate_builtin): Likewise.
29082 (rs6000_htm_spr_icode): Likewise.
29083 (altivec_expand_vec_init_builtin): Likewise.
29084 (get_element_number): Likewise.
29085 (altivec_expand_vec_set_builtin): Likewise.
29086 (altivec_expand_vec_ext_builtin): Likewise.
29087 (rs6000_invalid_builtin): Likewise.
29088 (rs6000_fold_builtin): Likewise.
29089 (fold_build_vec_cmp): Likewise.
29090 (fold_compare_helper): Likewise.
29091 (map_to_integral_tree_type): Likewise.
29092 (fold_mergehl_helper): Likewise.
29093 (fold_mergeeo_helper): Likewise.
29094 (rs6000_builtin_valid_without_lhs): Likewise.
29095 (rs6000_builtin_is_supported): Likewise.
29096 (rs6000_gimple_fold_mma_builtin): Likewise.
29097 (rs6000_gimple_fold_builtin): Likewise.
29098 (rs6000_expand_ldst_mask): Likewise.
29099 (cpu_expand_builtin): Likewise.
29100 (elemrev_icode): Likewise.
29101 (ldv_expand_builtin): Likewise.
29102 (lxvrse_expand_builtin): Likewise.
29103 (lxvrze_expand_builtin): Likewise.
29104 (stv_expand_builtin): Likewise.
29105 (mma_expand_builtin): Likewise.
29106 (htm_spr_num): Likewise.
29107 (htm_expand_builtin): Likewise.
29108 (rs6000_expand_builtin): Likewise.
29109 (rs6000_vector_type): Likewise.
29110 (rs6000_init_builtins): Likewise. Remove initialization of
29111 builtin_mode_to_type entries.
29112 (rs6000_builtin_decl): Move to rs6000-builtin.cc.
29113 * config/rs6000/rs6000.cc (rs6000_builtin_mask_for_load): New
29114 external declaration.
29115 (rs6000_builtin_md_vectorized_function): Likewise.
29116 (rs6000_builtin_reciprocal): Likewise.
29117 (altivec_builtin_mask_for_load): Move to rs6000-builtin.cc.
29118 (rs6000_builtin_types): Likewise.
29119 (builtin_mode_to_type): Remove.
29120 (rs6000_builtin_mask_for_load): Move to rs6000-builtin.cc. Remove
29122 (rs6000_builtin_md_vectorized_function): Likewise.
29123 (rs6000_builtin_reciprocal): Likewise.
29124 * config/rs6000/rs6000.h (builtin_mode_to_type): Remove.
29125 * config/rs6000/t-rs6000 (rs6000-builtin.o): New target.
29127 2022-02-03 Richard Biener <rguenther@suse.de>
29130 * tree-nrv.cc (pass_nrv::execute): Remove tieing result and found
29131 together via DECL_ABSTRACT_ORIGIN.
29133 2022-02-03 Bill Schmidt <wschmidt@linux.ibm.com>
29135 * config/rs6000/rs6000-call.cc (rs6000_expand_builtin): Revise error
29136 message for RES_BITS case.
29138 2022-02-03 Aldy Hernandez <aldyh@redhat.com>
29140 * gimple-range-fold.cc (fur_list::fur_list): Set m_local[1] correctly.
29142 2022-02-03 Jakub Jelinek <jakub@redhat.com>
29144 * config/arm/arm.opt (mfix-cortex-a57-aes-1742098,
29145 mfix-cortex-a72-aes-1655431): Ensure description ends with full stop.
29147 2022-02-03 Aldy Hernandez <aldyh@redhat.com>
29149 * cfganal.cc (verify_marked_backedges): New.
29150 * cfganal.h (verify_marked_backedges): New.
29151 * gimple-range-path.cc (path_range_query::path_range_query):
29152 Verify freshness of back edges.
29153 * tree-ssa-loop-ch.cc (ch_base::copy_headers): Call
29154 mark_dfs_back_edges.
29155 * tree-ssa-threadbackward.cc (back_threader::back_threader): Move
29156 path_range_query construction after backedges have been
29159 2022-02-03 Richard Sandiford <richard.sandiford@arm.com>
29161 * config/aarch64/aarch64-simd.md (movmisalign<mode>): Extend from
29164 2022-02-03 Richard Sandiford <richard.sandiford@arm.com>
29166 * config/aarch64/iterators.md (VALL_F16MOV): Delete.
29167 * config/aarch64/aarch64-simd.md (mov<mode>): Use VALL_F16 instead
29170 2022-02-03 Martin Liska <mliska@suse.cz>
29172 * config/i386/i386-options.cc (ix86_valid_target_attribute_inner_p):
29173 Change subject and object in the error message.
29174 * config/s390/s390.cc (s390_valid_target_attribute_inner_p):
29177 2022-02-03 Martin Liska <mliska@suse.cz>
29179 * config/s390/s390.cc (s390_valid_target_attribute_inner_p):
29180 Use the error message for i386 target.
29182 2022-02-03 Jakub Jelinek <jakub@redhat.com>
29184 PR tree-optimization/104334
29185 * range-op.cc (range_operator::wi_fold_in_parts): Change lh_range
29186 and rh_range type to widest_int and subtract in widest_int. Remove
29187 ov_rh, ov_lh and sign vars, always perform comparisons as signed
29188 and use >, < and == operators for it.
29190 2022-02-03 Martin Sebor <msebor@redhat.com>
29192 * common.opt (-Wuse-after-free): Correct typos.
29194 2022-02-02 David Malcolm <dmalcolm@redhat.com>
29197 * doc/invoke.texi (-ftrivial-auto-var-init=): Add reference to
29198 -Wanalyzer-use-of-uninitialized-value to paragraph documenting that
29199 -ftrivial-auto-var-init= doesn't suppress warnings.
29201 2022-02-02 Martin Liska <mliska@suse.cz>
29203 * dwarf2out.cc (TEXT_SECTION_NAME): Remove unused macro.
29205 2022-02-02 Bernd Kuhls <bernd.kuhls@t-online.de>
29208 * config/or1k/linux.h (CPP_SPEC): Define.
29210 2022-02-02 Tamar Christina <tamar.christina@arm.com>
29212 PR tree-optimization/102819
29213 PR tree-optimization/103169
29214 * config/arm/vec-common.md (cml<fcmac1><conj_op><mode>4): Use
29217 2022-02-02 Tamar Christina <tamar.christina@arm.com>
29219 PR tree-optimization/102819
29220 PR tree-optimization/103169
29221 * config/aarch64/aarch64-simd.md (cml<fcmac1><conj_op><mode>4): Use
29223 * config/aarch64/aarch64-sve.md (cml<fcmac1><conj_op><mode>4): Likewise.
29225 2022-02-02 Tamar Christina <tamar.christina@arm.com>
29227 PR tree-optimization/102819
29228 PR tree-optimization/103169
29229 * doc/md.texi: Update docs for cfms, cfma.
29230 * tree-data-ref.h (same_data_refs): Accept optional offset.
29231 * tree-vect-slp-patterns.cc (is_linear_load_p): Fix issue with repeating
29233 (vect_normalize_conj_loc): Remove.
29234 (is_eq_or_top): Change to take two nodes.
29235 (enum _conj_status, compatible_complex_nodes_p,
29236 vect_validate_multiplication): New.
29237 (class complex_add_pattern, complex_add_pattern::matches,
29238 complex_add_pattern::recognize, class complex_mul_pattern,
29239 complex_mul_pattern::recognize, class complex_fms_pattern,
29240 complex_fms_pattern::recognize, class complex_operations_pattern,
29241 complex_operations_pattern::recognize, addsub_pattern::recognize): Pass
29243 (complex_fms_pattern::matches, complex_mul_pattern::matches): Pass new
29244 cache and use new validation code.
29245 * tree-vect-slp.cc (vect_match_slp_patterns_2, vect_match_slp_patterns,
29246 vect_analyze_slp): Pass along cache.
29247 (compatible_calls_p): Expose.
29248 * tree-vectorizer.h (compatible_calls_p, slp_node_hash,
29249 slp_compat_nodes_map_t): New.
29250 (class vect_pattern): Update signatures include new cache.
29252 2022-02-02 Hans-Peter Nilsson <hp@axis.com>
29254 * config/cris/cris.cc (cris_preferred_reload_class): Reject
29255 "eliminated" registers and small-enough constants unless
29256 reloaded into a class that is a subset of GENERAL_REGS.
29257 * config/cris/cris.md (attribute "cpu_variant"): New.
29258 (attribute "enabled"): Conditionalize on a matching attribute
29259 cpu_variant, if specified.
29260 ("*movsi_internal<setcc><setnz><setnzvc>"): For moves to and from
29261 memory, add cpu-variant-enabled variants for "r" alternatives on
29262 the far side of the "x" alternatives, preferring the "x" ones
29263 only for variants where MOF is present (in addition to SRP).
29265 2022-02-02 Hans-Peter Nilsson <hp@axis.com>
29267 * config/cris/cris.cc (cris_register_move_cost): Remove special pre-ira
29268 extra cost for ALL_REGS.
29270 2022-02-02 Hans-Peter Nilsson <hp@axis.com>
29272 * config/cris/constraints.md (define_register_constraint "b"): Now
29274 * config/cris/cris.md (CRIS_ACR_REGNUM): Remove.
29275 * config/cris/cris.h: (reg_class, REG_CLASS_NAMES)
29276 (REG_CLASS_CONTENTS): Remove ACR_REGS, SPEC_ACR_REGS, GENNONACR_REGS,
29277 and SPEC_GENNONACR_REGS.
29278 * config/cris/cris.cc (cris_preferred_reload_class): Don't mention
29279 ACR_REGS and return GENERAL_REGS instead of GENNONACR_REGS.
29281 2022-02-02 Hans-Peter Nilsson <hp@axis.com>
29283 * config/cris/cris.md ("*movsi_internal<setcc><setnz><setnzvc>"):
29284 Conditionalize on (sub-)register operands or operand 1 being 0.
29286 2022-02-02 Hans-Peter Nilsson <hp@axis.com>
29288 * config/cris/cris.h (TARGET_DEFAULT): Don't include MASK_MUL_BUG.
29289 (MUL_BUG_ASM_DEFAULT): New macro.
29290 (MAYBE_AS_NO_MUL_BUG_ABORT): Define in terms of MUL_BUG_ASM_DEFAULT.
29291 * doc/invoke.texi (CRIS Options, -mmul-bug-workaround): Adjust
29294 2022-02-01 Eugene Rozenfeld <erozen@microsoft.com>
29296 * opts.cc (common_handle_option): Don't set param_early_inliner_max_iterations
29299 2022-02-01 Eugene Rozenfeld <erozen@microsoft.com>
29301 * auto-profile.cc (auto_profile): Hard-code the number of iterations (10).
29303 2022-02-01 Andrew Pinski <apinski@marvell.com>
29305 * doc/install.texi:
29307 2022-02-01 Ilya Leoshkevich <iii@linux.ibm.com>
29309 * config/s390/s390.cc (s390_code_end): Do not switch back to
29312 2022-02-01 Jakub Jelinek <jakub@redhat.com>
29315 * config/rs6000/t-rs6000 (EXTRA_GTYPE_DEPS): Append rs6000-builtins.h
29316 rather than $(srcdir)/config/rs6000/rs6000-builtins.def.
29317 * config/rs6000/rs6000-gen-builtins.cc (write_decls): Don't use
29318 GTY((user)) for struct bifdata and struct ovlddata. Instead add
29319 GTY((skip(""))) to members with pointer and enum types that don't need
29320 to be tracked. Add GTY(()) to rs6000_builtin_info and rs6000_instance_info
29321 declarations. Don't emit gt_ggc_mx and gt_pch_nx declarations.
29322 (write_extern_fntype, write_fntype): Remove.
29323 (write_fntype_init): Emit the fntype vars as automatic vars instead
29324 of file scope ones.
29325 (write_header_file): Don't iterate with write_extern_fntype.
29326 (write_init_file): Don't iterate with write_fntype. Don't emit
29327 gt_ggc_mx and gt_pch_nx definitions.
29329 2022-02-01 Jason Merrill <jason@redhat.com>
29331 * tree.h (struct tree_vec_map_cache_hasher): Move from...
29332 * tree.cc (struct tree_vec_map_cache_hasher): ...here.
29334 2022-02-01 Tom de Vries <tdevries@suse.de>
29336 * config/nvptx/nvptx.cc (nvptx_single): Use nvptx_uniform_warp_check.
29337 * config/nvptx/nvptx.md (define_c_enum "unspecv"): Add
29338 UNSPECV_UNIFORM_WARP_CHECK.
29339 (define_insn "nvptx_uniform_warp_check"): New define_insn.
29341 2022-02-01 Tom de Vries <tdevries@suse.de>
29343 * config/nvptx/nvptx.cc (nvptx_single): Use nvptx_warpsync.
29344 * config/nvptx/nvptx.md (define_c_enum "unspecv"): Add
29346 (define_insn "nvptx_warpsync"): New define_insn.
29348 2022-02-01 Tom de Vries <tdevries@suse.de>
29350 * config/nvptx/nvptx.opt (mptx): Set to PTX_VERSION_6_3 by default.
29352 2022-02-01 Tom de Vries <tdevries@suse.de>
29354 * config/nvptx/nvptx-opts.h (enum ptx_version): Add PTX_VERSION_6_0.
29355 * config/nvptx/nvptx.h (TARGET_PTX_6_0): New macro.
29356 * config/nvptx/nvptx.md (define_insn "nvptx_barsync"): Use barrier
29357 insn for TARGET_PTX_6_0.
29359 2022-02-01 Tom de Vries <tdevries@suse.de>
29362 * config/nvptx/nvptx.cc (prevent_branch_around_nothing): Handle nop
29365 2022-02-01 Tom de Vries <tdevries@suse.de>
29367 * config/nvptx/nvptx.md (define_insn "atomic_compare_and_swap<mode>_1")
29368 (define_insn "atomic_exchange<mode>")
29369 (define_insn "atomic_fetch_add<mode>")
29370 (define_insn "atomic_fetch_addsf")
29371 (define_insn "atomic_fetch_<logic><mode>"): Output non-atomic version
29372 if memory operands is frame-relative.
29374 2022-02-01 Tom de Vries <tdevries@suse.de>
29376 * config/nvptx/nvptx.cc (enum nvptx_builtins): Add
29377 NVPTX_BUILTIN_MEMBAR_GL and NVPTX_BUILTIN_MEMBAR_CTA.
29379 (nvptx_init_builtins): Add MEMBAR_GL and MEMBAR_CTA.
29380 (nvptx_expand_builtin): Handle NVPTX_BUILTIN_MEMBAR_GL and
29381 NVPTX_BUILTIN_MEMBAR_CTA.
29382 (nvptx_lockfull_update): Add level parameter. Emit barriers.
29383 (nvptx_reduction_update, nvptx_goacc_reduction_fini): Update call to
29384 nvptx_lockfull_update.
29385 * config/nvptx/nvptx.md (define_c_enum "unspecv"): Add
29387 (define_expand "nvptx_membar_gl"): New expand.
29388 (define_insn "*nvptx_membar_gl"): New insn.
29390 2022-02-01 Martin Liska <mliska@suse.cz>
29392 * doc/install.texi: Remove option for GCC < 4.8.
29394 2022-02-01 Jakub Jelinek <jakub@redhat.com>
29396 PR middle-end/104307
29397 * tree-vect-generic.cc (expand_vector_comparison): Don't push debug
29398 stmts to uses vector, just set vec_cond_expr_only to false for
29399 non-VEC_COND_EXPRs instead of pushing them into uses. Treat
29400 VEC_COND_EXPRs that use lhs not just in rhs1, but rhs2 or rhs3 too
29401 like non-VEC_COND_EXPRs.
29403 2022-02-01 Bill Schmidt <wschmidt@linux.ibm.com>
29405 * config/rs6000/rs6000-overload.def (VEC_ABSD): Remove #ifdef token.
29406 (VEC_BLENDV): Likewise.
29407 (VEC_BPERM): Likewise.
29408 (VEC_CFUGE): Likewise.
29409 (VEC_CIPHER_BE): Likewise.
29410 (VEC_CIPHERLAST_BE): Likewise.
29411 (VEC_CLRL): Likewise.
29412 (VEC_CLRR): Likewise.
29413 (VEC_CMPNEZ): Likewise.
29414 (VEC_CNTLZ): Likewise.
29415 (VEC_CNTLZM): Likewise.
29416 (VEC_CNTTZM): Likewise.
29417 (VEC_CNTLZ_LSBB): Likewise.
29418 (VEC_CNTM): Likewise.
29419 (VEC_CNTTZ): Likewise.
29420 (VEC_CNTTZ_LSBB): Likewise.
29421 (VEC_CONVERT_4F32_8F16): Likewise.
29422 (VEC_DIV): Likewise.
29423 (VEC_DIVE): Likewise.
29424 (VEC_EQV): Likewise.
29425 (VEC_EXPANDM): Likewise.
29426 (VEC_EXTRACT_FP_FROM_SHORTH): Likewise.
29427 (VEC_EXTRACT_FP_FROM_SHORTL): Likewise.
29428 (VEC_EXTRACTH): Likewise.
29429 (VEC_EXTRACTL): Likewise.
29430 (VEC_EXTRACTM): Likewise.
29431 (VEC_EXTRACT4B): Likewise.
29432 (VEC_EXTULX): Likewise.
29433 (VEC_EXTURX): Likewise.
29434 (VEC_FIRSTMATCHINDEX): Likewise.
29435 (VEC_FIRSTMACHOREOSINDEX): Likewise.
29436 (VEC_FIRSTMISMATCHINDEX): Likewise.
29437 (VEC_FIRSTMISMATCHOREOSINDEX): Likewise.
29438 (VEC_GB): Likewise.
29439 (VEC_GENBM): Likewise.
29440 (VEC_GENHM): Likewise.
29441 (VEC_GENWM): Likewise.
29442 (VEC_GENDM): Likewise.
29443 (VEC_GENQM): Likewise.
29444 (VEC_GENPCVM): Likewise.
29445 (VEC_GNB): Likewise.
29446 (VEC_INSERTH): Likewise.
29447 (VEC_INSERTL): Likewise.
29448 (VEC_INSERT4B): Likewise.
29449 (VEC_LXVL): Likewise.
29450 (VEC_MERGEE): Likewise.
29451 (VEC_MERGEO): Likewise.
29452 (VEC_MOD): Likewise.
29453 (VEC_MSUB): Likewise.
29454 (VEC_MULH): Likewise.
29455 (VEC_NAND): Likewise.
29456 (VEC_NCIPHER_BE): Likewise.
29457 (VEC_NCIPHERLAST_BE): Likewise.
29458 (VEC_NEARBYINT): Likewise.
29459 (VEC_NMADD): Likewise.
29460 (VEC_ORC): Likewise.
29461 (VEC_PDEP): Likewise.
29462 (VEC_PERMX): Likewise.
29463 (VEC_PEXT): Likewise.
29464 (VEC_POPCNT): Likewise.
29465 (VEC_PARITY_LSBB): Likewise.
29466 (VEC_REPLACE_ELT): Likewise.
29467 (VEC_REPLACE_UN): Likewise.
29468 (VEC_REVB): Likewise.
29469 (VEC_RINT): Likewise.
29470 (VEC_RLMI): Likewise.
29471 (VEC_RLNM): Likewise.
29472 (VEC_SBOX_BE): Likewise.
29473 (VEC_SIGNEXTI): Likewise.
29474 (VEC_SIGNEXTLL): Likewise.
29475 (VEC_SIGNEXTQ): Likewise.
29476 (VEC_SLDB): Likewise.
29477 (VEC_SLV): Likewise.
29478 (VEC_SPLATI): Likewise.
29479 (VEC_SPLATID): Likewise.
29480 (VEC_SPLATI_INS): Likewise.
29481 (VEC_SQRT): Likewise.
29482 (VEC_SRDB): Likewise.
29483 (VEC_SRV): Likewise.
29484 (VEC_STRIL): Likewise.
29485 (VEC_STRIL_P): Likewise.
29486 (VEC_STRIR): Likewise.
29487 (VEC_STRIR_P): Likewise.
29488 (VEC_STXVL): Likewise.
29489 (VEC_TERNARYLOGIC): Likewise.
29490 (VEC_TEST_LSBB_ALL_ONES): Likewise.
29491 (VEC_TEST_LSBB_ALL_ZEROS): Likewise.
29492 (VEC_VEE): Likewise.
29493 (VEC_VES): Likewise.
29494 (VEC_VIE): Likewise.
29495 (VEC_VPRTYB): Likewise.
29496 (VEC_VSCEEQ): Likewise.
29497 (VEC_VSCEGT): Likewise.
29498 (VEC_VSCELT): Likewise.
29499 (VEC_VSCEUO): Likewise.
29500 (VEC_VSEE): Likewise.
29501 (VEC_VSES): Likewise.
29502 (VEC_VSIE): Likewise.
29503 (VEC_VSTDC): Likewise.
29504 (VEC_VSTDCN): Likewise.
29505 (VEC_VTDC): Likewise.
29506 (VEC_XL): Likewise.
29507 (VEC_XL_BE): Likewise.
29508 (VEC_XL_LEN_R): Likewise.
29509 (VEC_XL_SEXT): Likewise.
29510 (VEC_XL_ZEXT): Likewise.
29511 (VEC_XST): Likewise.
29512 (VEC_XST_BE): Likewise.
29513 (VEC_XST_LEN_R): Likewise.
29514 (VEC_XST_TRUNC): Likewise.
29515 (VEC_XXPERMDI): Likewise.
29516 (VEC_XXSLDWI): Likewise.
29517 (VEC_TSTSFI_EQ_DD): Likewise.
29518 (VEC_TSTSFI_EQ_TD): Likewise.
29519 (VEC_TSTSFI_GT_DD): Likewise.
29520 (VEC_TSTSFI_GT_TD): Likewise.
29521 (VEC_TSTSFI_LT_DD): Likewise.
29522 (VEC_TSTSFI_LT_TD): Likewise.
29523 (VEC_TSTSFI_OV_DD): Likewise.
29524 (VEC_TSTSFI_OV_TD): Likewise.
29525 (VEC_VADDCUQ): Likewise.
29526 (VEC_VADDECUQ): Likewise.
29527 (VEC_VADDEUQM): Likewise.
29528 (VEC_VADDUDM): Likewise.
29529 (VEC_VADDUQM): Likewise.
29530 (VEC_VBPERMQ): Likewise.
29531 (VEC_VCLZB): Likewise.
29532 (VEC_VCLZD): Likewise.
29533 (VEC_VCLZH): Likewise.
29534 (VEC_VCLZW): Likewise.
29535 (VEC_VCTZB): Likewise.
29536 (VEC_VCTZD): Likewise.
29537 (VEC_VCTZH): Likewise.
29538 (VEC_VCTZW): Likewise.
29539 (VEC_VEEDP): Likewise.
29540 (VEC_VEESP): Likewise.
29541 (VEC_VESDP): Likewise.
29542 (VEC_VESSP): Likewise.
29543 (VEC_VIEDP): Likewise.
29544 (VEC_VIESP): Likewise.
29545 (VEC_VPKSDSS): Likewise.
29546 (VEC_VPKSDUS): Likewise.
29547 (VEC_VPKUDUM): Likewise.
29548 (VEC_VPKUDUS): Likewise.
29549 (VEC_VPOPCNT): Likewise.
29550 (VEC_VPOPCNTB): Likewise.
29551 (VEC_VPOPCNTD): Likewise.
29552 (VEC_VPOPCNTH): Likewise.
29553 (VEC_VPOPCNTW): Likewise.
29554 (VEC_VPRTYBD): Likewise.
29555 (VEC_VPRTYBQ): Likewise.
29556 (VEC_VPRTYBW): Likewise.
29557 (VEC_VRLD): Likewise.
29558 (VEC_VSLD): Likewise.
29559 (VEC_VSRAD): Likewise.
29560 (VEC_VSRD): Likewise.
29561 (VEC_VSTDCDP): Likewise.
29562 (VEC_VSTDCNDP): Likewise.
29563 (VEC_VSTDCNQP): Likewise.
29564 (VEC_VSTDCNSP): Likewise.
29565 (VEC_VSTDCQP): Likewise.
29566 (VEC_VSTDCSP): Likewise.
29567 (VEC_VSUBECUQ): Likewise.
29568 (VEC_VSUBEUQM): Likewise.
29569 (VEC_VSUBUDM): Likewise.
29570 (VEC_VSUBUQM): Likewise.
29571 (VEC_VTDCDP): Likewise.
29572 (VEC_VTDCSP): Likewise.
29573 (VEC_VUPKHSW): Likewise.
29574 (VEC_VUPKLSW): Likewise.
29576 2022-02-01 Andreas Krebbel <krebbel@linux.ibm.com>
29578 PR rtl-optimization/101260
29579 * regcprop.cc (maybe_mode_change): Invoke mode_change_ok also for
29582 2022-02-01 Xi Ruoyao <xry111@mengyan1223.wang>
29584 PR middle-end/95115
29585 * fold-const.cc (const_binop): Do not fold NaN result from
29588 2022-02-01 Tom de Vries <tdevries@suse.de>
29590 * tree-loop-distribution.cc (generate_reduction_builtin_1): Check for
29591 -ftree-loop-distribute-patterns.
29592 (loop_distribution::execute): Don't call transform_reduction_loop for
29593 -fno-tree-loop-distribute-patterns.
29595 2022-01-31 Andrew Pinski <apinski@marvell.com>
29597 * fold-const.h (operand_compare::operand_equal_p):
29598 Fix comment about OEP_* flags.
29600 2022-01-31 Jakub Jelinek <jakub@redhat.com>
29603 * config/rs6000/aix.h (OPTION_GLIBC): Remove.
29604 * config/rs6000/darwin.h (OPTION_GLIBC): Likewise.
29605 * config/rs6000/option-defaults.h (OPTION_GLIBC): Define to 0
29606 if not already defined.
29608 2022-01-31 Martin Sebor <msebor@redhat.com>
29610 PR middle-end/104232
29611 * gimple-ssa-warn-access.cc (pointers_related_p): Add argument.
29612 Handle PHIs. Add a synonymous overload.
29613 (pass_waccess::check_pointer_uses): Call pointers_related_p.
29615 2022-01-31 Richard Biener <rguenther@suse.de>
29617 PR tree-optimization/100499
29618 * fold-const.cc (multiple_of_p): Pass the correct type of
29619 the expression to the recursive invocation of multiple_of_p
29620 for conversions and use CASE_CONVERT.
29622 2022-01-31 Eric Botcazou <ebotcazou@adacore.com>
29625 * config/sparc/linux64.h (TARGET_DEFAULT): Add MASK_V8PLUS.
29627 2022-01-31 Richard Biener <rguenther@suse.de>
29629 PR tree-optimization/100499
29630 * tree-cfg.cc (verify_gimple_assign_ternary): Use multiple_p
29631 on poly-ints instead of multiple_of_p.
29632 * tree-ssa.cc (maybe_rewrite_mem_ref_base): Likewise.
29633 (non_rewritable_mem_ref_base): Likewise.
29634 (non_rewritable_lvalue_p): Likewise.
29635 (execute_update_addresses_taken): Likewise.
29637 2022-01-29 Jakub Jelinek <jakub@redhat.com>
29638 Andrew Pinski <apinski@marvell.com>
29640 PR tree-optimization/104279
29641 PR tree-optimization/104280
29642 PR tree-optimization/104281
29643 * match.pd (1 / X -> X == 1 for unsigned X): Build eq with
29644 boolean_type_node and convert to type. Formatting fixes.
29646 2022-01-28 Yoshinori Sato <yo-satoh@sios.com>
29648 * config/sh/t-linux (MULTILIB_EXCEPTIONS): Add m1, mb/m1 and m2a.
29650 2022-01-28 Navid Rahimi <navidrahimi@microsoft.com>
29652 PR tree-optimization/103514
29653 * match.pd (a & b) ^ (a == b) -> !(a | b): New optimization.
29654 (a & b) == (a ^ b) -> !(a | b): New optimization.
29656 2022-01-28 Marek Polacek <polacek@redhat.com>
29658 * doc/invoke.texi: Update -Wbidi-chars documentation.
29660 2022-01-28 Iain Sandoe <iain@sandoe.co.uk>
29662 * config/rs6000/darwin.h (OPTION_GLIBC): Define to 0.
29664 2022-01-28 Zhao Wei Liew <zhaoweiliew@gmail.com>
29666 PR tree-optimization/95424
29667 * match.pd: Simplify 1 / X where X is an integer.
29669 2022-01-28 Jakub Jelinek <jakub@redhat.com>
29671 PR tree-optimization/104263
29672 * gimple-ssa-store-merging.cc (get_status_for_store_merging): For
29673 cfun->can_throw_non_call_exceptions && cfun->eh test whether
29674 last non-debug stmt in the bb is store_valid_for_store_merging_p
29675 rather than last stmt.
29677 2022-01-28 Martin Liska <mliska@suse.cz>
29679 * diagnostic.cc (diagnostic_action_after_output): Remove extra
29682 2022-01-28 Martin Liska <mliska@suse.cz>
29684 * config/rs6000/host-darwin.cc (segv_crash_handler):
29685 Do not use leading capital letter.
29686 (segv_handler): Likewise.
29687 * ipa-sra.cc (verify_splitting_accesses): Likewise.
29688 * varasm.cc (get_section): Likewise.
29690 2022-01-28 Richard Biener <rguenther@suse.de>
29692 PR tree-optimization/104267
29693 * tree-vect-stmts.cc (vectorizable_call): Properly use the
29694 per-argument determined vector type for externals and
29697 2022-01-28 Richard Biener <rguenther@suse.de>
29699 PR tree-optimization/104263
29700 * tree-cfg.cc (gimple_purge_dead_abnormal_call_edges):
29701 Purge edges also when !cfun->has_nonlocal_label
29702 and !cfun->calls_setjmp.
29704 2022-01-28 Maciej W. Rozycki <macro@embecosm.com>
29706 * config/riscv/riscv.md: Document `auipc' and `bitmanip' `type'
29709 2022-01-28 Jakub Jelinek <jakub@redhat.com>
29712 * cfgrtl.cc (loc_equal): New function.
29713 (unique_locus_on_edge_between_p): Use it.
29715 2022-01-28 Richard Biener <rguenther@suse.de>
29717 * cfganal.h (mark_dfs_back_edges): Provide API with struct
29719 * cfganal.cc (mark_dfs_back_edges): Take a struct function
29720 to work on, add a wrapper passing cfun.
29721 * graph.cc (draw_cfg_nodes_no_loops): Replace stray cfun
29722 uses with fun which is already passed.
29723 (draw_cfg_edges): Likewise.
29724 (draw_cfg_nodes_for_loop): Do not use draw_cfg_nodes_for_loop
29727 2022-01-27 Patrick Palka <ppalka@redhat.com>
29730 * tree.cc (build_call_vec): Add const to second parameter.
29731 * tree.h (build_call_vec): Likewise.
29733 2022-01-27 Martin Liska <mliska@suse.cz>
29736 * diagnostic.cc (diagnostic_initialize):
29737 Initialize report_bug flag.
29738 (diagnostic_action_after_output):
29739 Explain that -freport-bug option can be used for pre-processed
29740 file creation. Make the message shorter.
29741 (error_recursion): Rename Internal to internal.
29742 * diagnostic.h (struct diagnostic_context): New field.
29743 * opts.cc (common_handle_option): Init the field here.
29745 2022-01-27 Kewen Lin <linkw@linux.ibm.com>
29748 * config/rs6000/rs6000.cc
29749 (rs6000_cost_data::update_target_cost_per_stmt): Fix one wrong
29750 assertion with early return.
29752 2022-01-27 Chung-Lin Tang <cltang@codesourcery.com>
29754 PR middle-end/103642
29755 * gimplify.cc (gimplify_scan_omp_clauses): Do not do indir_p handling
29756 for non-pointer or non-reference-to-pointer cases.
29758 2022-01-27 Jakub Jelinek <jakub@redhat.com>
29760 PR tree-optimization/104196
29761 * gimple-fold.h (rewrite_to_defined_overflow): Add IN_PLACE argument.
29762 * gimple-fold.cc (rewrite_to_defined_overflow): Likewise. If true,
29763 return NULL and emit needed stmts before and after stmt.
29764 * tree-ssa-reassoc.cc (update_range_test): For inter-bb range opt
29765 pick as operand_entry that will hold the merged test the one feeding
29766 earliest condition, ensure that by swapping range->idx with some
29767 other range's idx if needed. If seq is non-NULL, don't actually swap
29768 it but instead rewrite stmts with undefined overflow in between
29770 (maybe_optimize_range_tests): Set ops[]->id to bb->index with the
29771 corresponding condition even if they have non-NULL ops[]->op.
29774 2022-01-26 Jakub Jelinek <jakub@redhat.com>
29777 * config/rs6000/emmintrin.h (_mm_sad_epu8): Use __asm__ instead of
29779 * config/rs6000/smmintrin.h (_mm_minpos_epu16): Declare iterator
29780 before for loop instead of for init clause.
29781 * config/rs6000/bmi2intrin.h (_pext_u64): Likewise.
29783 2022-01-26 Jakub Jelinek <jakub@redhat.com>
29786 * config/rs6000/bmiintrin.h: Test _X86GPRINTRIN_H_INCLUDED instead of
29787 _X86INTRIN_H_INCLUDED and adjust #error wording.
29788 * config/rs6000/bmi2intrin.h: Likewise.
29790 2022-01-26 Jakub Jelinek <jakub@redhat.com>
29793 * dwarf2out.cc (long_double_as_float128): New function.
29794 (modified_type_die): For powerpc64le IEEE 754 quad long double
29795 and complex long double emit those as DW_TAG_typedef to
29796 _Float128 or complex _Float128 base type.
29798 2022-01-26 Marek Polacek <polacek@redhat.com>
29801 * gimple-ssa-warn-access.cc (pass_waccess::warn_invalid_pointer): Don't
29802 warn when the SSA_NAME_VAR of REF has supressed -Wuse-after-free.
29804 2022-01-26 Martin Liska <mliska@suse.cz>
29806 * ipa-modref-tree.cc (modref_access_node::update):
29807 Remove "--param param=foo" with "--param foo".
29808 (modref_access_node::insert): Likewise.
29809 (modref_access_node::insert_kill): Likewise.
29810 * ipa-modref-tree.h (struct modref_ref_node): Likewise.
29811 (struct modref_base_node): Likewise.
29812 (struct modref_tree): Likewise.
29814 2022-01-26 Raoni Fassina Firmino <raoni@linux.ibm.com>
29817 * builtins.cc (expand_builtin_feclear_feraise_except): Add op0
29820 2022-01-25 Martin Sebor <msebor@redhat.com>
29822 PR tree-optimization/104203
29823 * gimple-ssa-warn-access.cc (pass_data pass_data_waccess): Use
29825 * pointer-query.cc (access_ref::merge_ref): Change return type.
29826 Convert failure to a conservative success.
29827 (access_ref::get_ref): Adjust to the change above. Short-circuit
29828 PHI evaluation after first failure turned into conservative success.
29829 * pointer-query.h (access_ref::merge_ref): Change return type.
29830 * timevar.def (TV_WARN_ACCESS): New timer variable.
29832 2022-01-25 David Edelsohn <dje.gcc@gmail.com>
29834 * config/rs6000/aix.h (OPTION_GLIBC): Define as 0.
29836 2022-01-25 Richard Biener <rguenther@suse.de>
29838 PR tree-optimization/104214
29839 * tree-ssa-loop-niter.cc (number_of_iterations_cond): Use
29840 stronger guarantees for relational pointer compares when
29841 rewriting BASE0 + STEP0 cmp BASE1 + STEP1 as
29842 BASE0 + STEP0 - STEP1 cmp BASE1.
29844 2022-01-25 Jakub Jelinek <jakub@redhat.com>
29847 * config/rs6000/rs6000-internal.h (rs6000_passes_ieee128): Don't
29849 * config/rs6000/rs6000.cc (rs6000_passes_ieee128,
29850 ieee128_mangling_gcc_8_1): Remove.
29851 (TARGET_ASM_GLOBALIZE_DECL_NAME): Don't redefine.
29852 (rs6000_mangle_type): Return "u9__ieee128" instead of
29853 ieee128_mangling_gcc_8_1 ? "U10__float128" : "u9__ieee128".
29854 (rs6000_globalize_decl_name): Remove.
29855 * config/rs6000/rs6000-call.cc (init_cumulative_args,
29856 rs6000_function_arg_advance_1): Don't set rs6000_passes_ieee128.
29858 2022-01-24 Martin Sebor <msebor@redhat.com>
29860 * pointer-query.cc (pointer_query::dump): Remove duplicate
29863 2022-01-24 Marek Polacek <polacek@redhat.com>
29865 PR preprocessor/104030
29866 * doc/invoke.texi: Update documentation for -Wbidi-chars.
29868 2022-01-24 Raoni Fassina Firmino <raoni@linux.ibm.com>
29871 * builtins.cc (expand_builtin_fegetround): New function.
29872 (expand_builtin_feclear_feraise_except): New function.
29873 (expand_builtin): Add cases for BUILT_IN_FEGETROUND,
29874 BUILT_IN_FECLEAREXCEPT and BUILT_IN_FERAISEEXCEPT.
29875 * config/rs6000/rs6000.md (fegetroundsi): New pattern.
29876 (feclearexceptsi): New Pattern.
29877 (feraiseexceptsi): New Pattern.
29878 * doc/extend.texi: Add a new introductory paragraph about the
29880 * doc/md.texi: (fegetround@var{m}): Document new optab.
29881 (feclearexcept@var{m}): Document new optab.
29882 (feraiseexcept@var{m}): Document new optab.
29883 * optabs.def (fegetround_optab): New optab.
29884 (feclearexcept_optab): New optab.
29885 (feraiseexcept_optab): New optab.
29887 2022-01-24 Richard Biener <rguenther@suse.de>
29888 Jiufu Guo <guojiufu@linux.ibm.com>
29890 PR tree-optimization/100740
29891 PR tree-optimization/101508
29892 PR tree-optimization/101972
29893 PR tree-optimization/102131
29894 * tree-ssa-loop-niter.cc (number_of_iterations_cond): Properly
29895 constrain BASE0 + STEP0 cmp BASE1 + STEP1 to
29896 BASE0 + STEP0 - STEP1 cmp BASE1 transform.
29898 2022-01-24 Jakub Jelinek <jakub@redhat.com>
29900 PR sanitizer/104158
29901 * opt-functions.awk (var_set): Handle EnumBitSet property.
29902 * optc-gen.awk: Don't disallow RejectNegative if EnumBitSet is
29904 * opts.h (enum cl_enum_var_value): New type.
29905 * opts-common.cc (decode_cmdline_option): Use CLEV_* values.
29906 Handle CLEV_BITSET.
29907 (cmdline_handle_error): Handle CLEV_BITSET.
29908 * opts.cc (test_enum_sets): Also test EnumBitSet requirements.
29909 * doc/options.texi (EnumBitSet): Document.
29910 * common.opt (fsanitize-coverage=): Use EnumBitSet instead of
29912 (trace-pc, trace-cmp): Drop Set properties.
29914 2022-01-24 Jakub Jelinek <jakub@redhat.com>
29916 PR sanitizer/104158
29917 * common.opt (flag_sanitize_coverage): Remove Variable entry.
29918 (fsanitize-coverage=): Remove RejectNegative property, add
29919 Var(flag_sanitize_coverage) and EnumSet properties.
29920 (trace-pc): Add Set(1) property.
29921 (trace-cmp): Add Set(2) property.
29922 * opts.cc (common_handle_option): Don't handle
29923 OPT_fsanitize_coverage_.
29925 2022-01-24 Jakub Jelinek <jakub@redhat.com>
29927 PR sanitizer/104158
29928 * opt-functions.awk (var_set): Handle EnumSet property.
29929 * optc-gen.awk: Don't disallow RejectNegative if EnumSet is
29931 * opt-read.awk: Handle Set property.
29932 * opts.h (CL_ENUM_SET_SHIFT, CL_ERR_ENUM_SET_ARG): Define.
29933 (struct cl_decoded_option): Mention enum in value description.
29935 (set_option): Add mask argument defaulted to 0.
29936 * opts.cc (test_enum_sets): New function.
29937 (opts_cc_tests): Call it.
29938 * opts-common.cc (enum_arg_to_value): Change return argument
29939 from bool to int, on success return index into the cl_enum_arg
29940 array, on failure -1. Add len argument, if non-0, use strncmp
29942 (opt_enum_arg_to_value): Adjust caller.
29943 (decode_cmdline_option): Handle EnumSet represented as
29944 CLVC_ENUM with non-zero var_value. Initialize decoded->mask.
29945 (decode_cmdline_options_to_array): CLear opt_array[0].mask.
29946 (handle_option): Pass decoded->mask to set_options last argument.
29947 (generate_option): Clear decoded->mask.
29948 (generate_option_input_file): Likewise.
29949 (cmdline_handle_error): Handle CL_ERR_ENUM_SET_ARG.
29950 (set_option): Add mask argument, use it for CLVC_ENUM.
29951 (control_warning_option): Adjust enum_arg_to_value caller.
29952 * doc/options.texi: Document Set and EnumSet properties.
29954 2022-01-24 Jakub Jelinek <jakub@redhat.com>
29956 PR bootstrap/104170
29957 * config/linux.h (OPTION_GLIBC_P, OPTION_UCLIBC_P,
29958 OPTION_BIONIC_P, OPTION_MUSL_P): Define.
29959 (OPTION_GLIBC, OPTION_UCLIBC, OPTION_BIONIC, OPTION_MUSL): Redefine
29960 using OPTION_*_P macros.
29961 * config/alpha/linux.h (OPTION_GLIBC_P, OPTION_UCLIBC_P,
29962 OPTION_BIONIC_P, OPTION_MUSL_P): Define.
29963 (OPTION_GLIBC, OPTION_UCLIBC, OPTION_BIONIC, OPTION_MUSL): Redefine
29964 using OPTION_*_P macros.
29965 * config/rs6000/linux.h (OPTION_GLIBC_P, OPTION_UCLIBC_P,
29966 OPTION_BIONIC_P, OPTION_MUSL_P): Define.
29967 (OPTION_GLIBC, OPTION_UCLIBC, OPTION_BIONIC, OPTION_MUSL): Redefine
29968 using OPTION_*_P macros.
29969 * config/rs6000/linux64.h (OPTION_GLIBC_P, OPTION_UCLIBC_P,
29970 OPTION_BIONIC_P, OPTION_MUSL_P): Define.
29971 (OPTION_GLIBC, OPTION_UCLIBC, OPTION_BIONIC, OPTION_MUSL): Redefine
29972 using OPTION_*_P macros.
29973 * config/fuchsia.h (OPTION_MUSL_P): Redefine.
29974 * config/glibc-stdint.h (OPTION_MUSL_P): Define if not defined.
29975 * common/config/s390/s390-common.cc (s390_supports_split_stack): Re-add
29976 ATTRIBUTE_UNUSED to opts parameter. If OPTION_GLIBC_P is defined, use
29977 OPTION_GLIBC_P (opts) as condition, otherwise assume if (false).
29978 * common/config/i386/i386-common.cc (ix86_supports_split_stack): If
29979 OPTION_GLIBC_P is defined use !OPTION_GLIBC_P (opts) as condition,
29980 otherwise assume if (true).
29982 2022-01-24 Kito Cheng <kito.cheng@sifive.com>
29984 * common/config/riscv/riscv-common.cc (riscv_subset_list::to_string):
29985 Skip zicsr and zifencei if I-ext is 2.0.
29987 2022-01-24 Jia-Wei Chen <jiawei@iscas.ac.cn>
29989 * config.gcc: Modify default isa_spec version.
29991 2022-01-24 Jiufu Guo <guojiufu@linux.ibm.com>
29993 PR tree-optimization/102087
29994 * tree-ssa-loop-niter.cc (number_of_iterations_until_wrap):
29995 Correct PLUS result type.
29997 2022-01-24 H.J. Lu <hjl.tools@gmail.com>
30000 * config/i386/predicates.md (bcst_mem_operand): Also check mode
30001 of memory broadcast.
30003 2022-01-23 Andrew Pinski <apinski@marvell.com>
30006 * config/aarch64/aarch64-builtins.cc
30007 (aarch64_general_gimple_fold_builtin): Handle
30008 __builtin_aarch64_sqrt* and simplify into SQRT internal
30011 2022-01-22 Jakub Jelinek <jakub@redhat.com>
30014 * opts-global.cc (handle_common_deferred_options): Quote
30015 --enable-plugin in diagnostics to avoid -Werror=format-diag.
30017 2022-01-21 Michael Meissner <meissner@the-meissners.org>
30020 * config/rs6000/rs6000-protos.h (prefixed_xxsplti_p): Delete.
30021 * config/rs6000/rs6000.cc (prefixed_xxsplti_p): Delete.
30022 * config/rs6000/rs6000.md (prefixed attribute): Delete section
30023 that sets the prefixed attribute for xxspltiw, xxspltidp, and
30024 xxsplti32dx instructions.
30025 (movsf_hardfloat): Explicitly set the prefixed attribute
30026 when xxspltiw and xxspltidp instructions are generated.
30027 (mov<mode>_hardfloat32): Likewise.
30028 (mov<mode>_hardfloat64): Likewise.
30029 * config/rs6000/vsx.md (vsx_mov<mode>_64bit): Explicitly set the
30030 prefixed attribute for xxspltiw and xxspltidp instructions.
30031 (vsx_mov<mode>_32bit): Likewise.
30033 2022-01-21 H.J. Lu <hjl.tools@gmail.com>
30035 PR bootstrap/104170
30036 * common/config/i386/i386-common.cc (ix86_supports_split_stack):
30037 Return true only on glibc.
30038 * config/i386/gnu-user-common.h (STACK_CHECK_STATIC_BUILTIN):
30039 Revert commit c163647ffbc.
30040 * config/i386/gnu.h (TARGET_LIBC_PROVIDES_SSP): Likewise.
30042 2022-01-21 Sören Tempel <soeren@soeren-tempel.net>
30044 * common/config/s390/s390-common.cc (s390_supports_split_stack):
30045 Only support split-stack on glibc targets.
30046 * config/i386/gnu-user-common.h (STACK_CHECK_STATIC_BUILTIN): Ditto.
30047 * config/i386/gnu.h (defined): Ditto.
30049 2022-01-21 Bill Schmidt <wschmidt@linux.ibm.com>
30051 * config/rs6000/rs6000-overload.def (VEC_SLDW): Add instances for
30052 vector float and vector double.
30054 2022-01-21 Bill Seurer <seurer@gcc.gnu.org>
30056 * config/rs6000/rs6000.cc (rs6000_get_function_versions_dispatcher):
30057 Fix mention of ifunc in string.
30059 2022-01-21 Roger Sayle <roger@nextmovesoftware.com>
30061 PR middle-end/104140
30062 * tree-ssa-math-opts.cc (convert_mult_to_highpart): Check that the
30063 operands of the widening multiplication are either both signed or
30064 both unsigned, and abort the conversion if mismatched.
30065 * doc/generic.texi (WIDEN_MULT_EXPR): Describe expression node.
30066 (MULT_HIGHPART_EXPR): Clarify that operands must have the same
30068 * tree.def (MULT_HIGHPART_EXPR): Document both operands must have
30069 integer types with the same precision and signedness.
30070 (WIDEN_MULT_EXPR): Document that operands must have integer types
30071 with the same precision, but possibly differing signedness.
30072 * config/riscv/riscv-c.cc (riscv_cpu_cpp_builtins): Defend against
30073 riscv_current_subset_list returning a NULL pointer (empty list).
30075 2022-01-21 Vladimir N. Makarov <vmakarov@redhat.com>
30078 * ira.h (struct target_ira): Add member
30079 x_ira_exclude_class_mode_regs.
30080 (ira_exclude_class_mode_regs): New macro.
30081 * lra.h (lra_create_new_reg): Add arg exclude_start_hard_regs and
30083 * lra-int.h: ... to here.
30084 (lra_create_new_reg_with_unique_value): Add arg
30085 exclude_start_hard_regs.
30086 (class lra_reg): Add member exclude_start_hard_regs.
30087 * lra-assigns.cc (find_hard_regno_for_1): Setup
30088 impossible_start_hard_regs from exclude_start_hard_regs.
30089 * lra-constraints.cc (get_reload_reg): Add arg exclude_start_hard_regs and pass
30090 it lra_create_new_reg[_with_unique_value].
30091 (match_reload): Ditto.
30092 (check_and_process_move): Pass NULL
30093 exclude_start_hard_regs to lra_create_new_reg_with_unique_value.
30094 (goal_alt_exclude_start_hard_regs): New static variable.
30095 (process_addr_reg, simplify_operand_subreg): Pass NULL
30096 exclude_start_hard_regs to lra_create_new_reg_with_unique_value
30097 and get_reload_reg.
30098 (process_alt_operands): Setup goal_alt_exclude_start_hard_regs.
30099 Use this_alternative_exclude_start_hard_regs additionally to find
30100 winning operand alternative.
30101 (base_to_reg, base_plus_disp_to_reg, index_part_to_reg): Pass NULL
30102 exclude_start_hard_regs to lra_create_new_reg.
30103 (process_address_1, emit_inc): Ditto.
30104 (curr_insn_transform): Pass exclude_start_hard_regs value to
30105 lra_create_new_reg, get_reload_reg, match_reload.
30106 (inherit_reload_reg, split_reg): Pass NULL exclude_start_hard_regs
30107 to lra_create_new_reg.
30108 (process_invariant_for_inheritance): Ditto.
30109 * lra-remat.cc (update_scratch_ops): Ditto.
30110 * lra.cc (lra_create_new_reg_with_unique_value): Add arg
30111 exclude_start_hard_regs. Setup the corresponding member of
30113 (lra_create_new_reg): Add arg exclude_start_hard_regs and pass it
30114 to lra_create_new_reg_with_unique_value.
30115 (initialize_lra_reg_info_element): Initialize member
30116 exclude_start_hard_regs.
30117 (get_scratch_reg): Pass NULL to lra_create_new_reg.
30118 * ira.cc (setup_prohibited_class_mode_regs): Rename to
30119 setup_prohibited_and_exclude_class_mode_regs and calculate
30120 ira_exclude_class_mode_regs.
30122 2022-01-21 Martin Liska <mliska@suse.cz>
30124 * configure.ac: Detect ld_is_mold and use it for
30125 comdat_group=yes and gcc_cv_ld_hidden=yes.
30126 * configure: Regenerate.
30128 2022-01-21 Richard Biener <rguenther@suse.de>
30130 PR tree-optimization/100089
30131 * tree-vect-slp.cc (vect_slp_region): Reject BB vectorization
30132 of if-converted loops with unvectorized COND_EXPRs for
30133 all but the unlimited cost models.
30135 2022-01-21 Ard Biesheuvel <ardb@kernel.org>
30137 * config/arm/arm-opts.h (enum stack_protector_guard): New.
30138 * config/arm/arm-protos.h (arm_stack_protect_tls_canary_mem):
30140 * config/arm/arm.cc (TARGET_STACK_PROTECT_GUARD): Define.
30141 (arm_option_override_internal): Handle and put in error checks.
30142 for stack protector guard options.
30143 (arm_option_reconfigure_globals): Likewise.
30144 (arm_stack_protect_tls_canary_mem): New.
30145 (arm_stack_protect_guard): New.
30146 * config/arm/arm.md (stack_protect_set): New.
30147 (stack_protect_set_tls): Likewise.
30148 (stack_protect_test): Likewise.
30149 (stack_protect_test_tls): Likewise.
30150 (reload_tp_hard): Likewise.
30151 * config/arm/arm.opt (-mstack-protector-guard): New
30152 (-mstack-protector-guard-offset): New.
30153 * doc/invoke.texi: Document new options.
30155 2022-01-21 Richard Biener <rguenther@suse.de>
30157 PR tree-optimization/104156
30158 * tree-ssa-loop-unswitch.cc (tree_unswitch_outer_loop):
30159 Collect and reset debug stmts with out-of-loop uses when
30161 (find_loop_guard): Adjust.
30162 (empty_bb_without_guard_p): Likewise. Ignore debug stmts.
30163 (used_outside_loop_p): Push debug uses to a vector of
30164 debug stmts to reset.
30165 (hoist_guard): Adjust -fopt-info category.
30167 2022-01-21 Richard Biener <rguenther@suse.de>
30169 PR tree-optimization/104152
30170 * tree-vect-slp.cc (vect_build_slp_tree_2): Add missing
30171 can_duplicate_and_interleave_p check.
30173 2022-01-21 Jakub Jelinek <jakub@redhat.com>
30175 * gimple-ssa-warn-access.cc (pass_waccess::warn_invalid_pointer):
30176 Avoid passing var to warning_at when the format string doesn't
30179 2022-01-21 Aldy Hernandez <aldyh@redhat.com>
30181 PR tree-optimization/103721
30182 * gimple-range-path.cc
30183 (path_range_query::relations_may_be_invalidated): New.
30184 (path_range_query::compute_ranges_in_block): Reset relations if
30185 they may be invalidated.
30186 (path_range_query::maybe_register_phi_relation): Exit if relations
30187 may be invalidated on incoming edge.
30188 (path_range_query::compute_phi_relations): Pass incoming PHI edge
30189 to maybe_register_phi_relation.
30190 * gimple-range-path.h (relations_may_be_invalidated): New.
30191 (maybe_register_phi_relation): Pass edge instead of tree.
30192 * tree-ssa-threadbackward.cc (back_threader::back_threader):
30194 * value-relation.cc (path_oracle::path_oracle): Call
30195 mark_dfs_back_edges.
30196 (path_oracle::register_relation): Add SSA names to m_registered
30198 (path_oracle::reset_path): Clear m_registered bitmap.
30199 * value-relation.h (path_oracle::set_root_oracle): New.
30201 2022-01-21 Jakub Jelinek <jakub@redhat.com>
30203 PR rtl-optimization/102478
30204 * optabs.cc (prepare_cmp_insn): If !can_create_pseudo_p (), don't
30205 force_reg constants and for -fnon-call-exceptions fail if copy_to_reg
30208 2022-01-20 Richard Biener <rguenther@suse.de>
30210 PR middle-end/100786
30211 * gimple-fold.cc (get_symbol_constant_value): Only return
30212 values of compatible type to the symbol.
30214 2022-01-20 Andrew MacLeod <amacleod@redhat.com>
30216 * value-relation.cc (relation_oracle::valid_equivs): Query and add
30217 if valid members of a set.
30218 (equiv_oracle::register_equiv): Call valid_equivs rather than
30219 bitmap direct operations.
30220 (path_oracle::register_equiv): Ditto.
30221 * value-relation.h (relation_oracle::valid_equivs): New prototype.
30223 2022-01-20 Richard Biener <rguenther@suse.de>
30226 * config/i386/i386.cc (ix86_gimple_fold_builtin): Check for
30227 LHS before folding __builtin_ia32_shufpd and friends.
30229 2022-01-20 Richard Earnshaw <rearnsha@arm.com>
30231 * config/arm/crypto.md (aes_op_protect): Allow moves from core
30232 registers and from memory.
30233 (aes_op_protect_misalign_load): New pattern.
30234 (aes_op_protect_neon_vld1v16qi): New pattern.
30236 2022-01-20 Richard Earnshaw <rearnsha@arm.com>
30238 * config/arm/crypto.md (crypto_<CRYPTO_AESMC:crypto_pattern>_protected):
30240 (aarch32_crypto_aese_fused_protected): Likewise.
30241 (aarch32_crypto_aesd_fused_protected): Likewise.
30243 2022-01-20 Richard Earnshaw <rearnsha@arm.com>
30245 * config/arm/crypto.md (crypto_<CRYPTO_AES:crypto_pattern>): Convert
30246 to define_expand. Add mitigation for the Cortex-A AES erratum
30248 (*crypto_<CRYPTO_AES:crypto_pattern>_insn): New pattern, based
30249 on original crypto_<CRYPTO_AES:crypto_pattern> insn.
30250 (aes_op_protect): New pattern.
30251 * config/arm/unspecs.md (unspec): Add UNSPEC_AES_PROTECT.
30253 2022-01-20 Richard Earnshaw <rearnsha@arm.com>
30255 * config/arm/arm-cpus.in (quirk_aes_1742098): New quirk feature
30256 (ALL_QUIRKS): Add it.
30257 (cortex-a57, cortex-a72): Enable it.
30258 (cortex-a57.cortex-a53, cortex-a72.cortex-a53): Likewise.
30259 * config/arm/arm.opt (mfix-cortex-a57-aes-1742098): New command-line
30261 (mfix-cortex-a72-aes-1655431): New option alias.
30262 * config/arm/arm.cc (arm_option_override): Handle default settings
30263 for AES erratum switch.
30264 * doc/invoke.texi (Arm Options): Document new options.
30266 2022-01-20 Richard Earnshaw <rearnsha@arm.com>
30268 * config/arm/crypto.md (crypto_<CYRPTO_AES:crypto_pattern>): Use
30269 <crypto_mode> rather than hard-coding the mode.
30270 (crypto_<CRYPTO_AESMC:crypto_pattern>): Fix white space.
30271 (crypto_<CRYPTO_AES:crypto_pattern>): Likewise.
30272 (*aarch32_crypto_aese_fused): Likewise.
30273 (*aarch32_crypto_aesd_fused): Likewise.
30274 (crypto_<CRYPTO_BINARY:crypto_pattern>): Likewise.
30275 (crypto_<CRYPTO_TERNARY:crypto_pattern>): Likewise.
30276 (crypto_sha1h_lb): Likewise.
30277 (crypto_vmullp64): Likewise.
30278 (crypto_<CRYPTO_SELECTING:crypto_pattern>): Likewise.
30279 (crypto_<CRYPTO_SELECTING:crypto_pattern>_lb): Likewise.
30281 2022-01-20 Richard Earnshaw <rearnsha@arm.com>
30283 * config/arm/crypto.md (crypto_<CRYPTO_AESMC:crypto_pattern>): Add
30284 iterator to pattern name to disambiguate.
30285 (crypto_<CRYPTO_AES:crypto_pattern>): Likewise.
30286 (crypto_<CRYPTO_BINARY:crypto_pattern>): Likewise.
30287 (crypto_<CRYPTO_TERNARY:crypto_pattern>): Likewise.
30288 (crypto_<CRYPTO_SELECTING:crypto_pattern>): Likewise.
30289 (crypto_<CRYPTO_SELECTING:crypto_pattern>_lb): Likewise.
30291 2022-01-20 Martin Liska <mliska@suse.cz>
30293 PR bootstrap/104135
30294 * emit-rtl.cc (make_insn_raw): Fix -Wformat-diag warnings.
30295 * rtl.cc: Partially disable -Wformat-diag for RTL checking
30298 2022-01-20 Jakub Jelinek <jakub@redhat.com>
30301 * dwarf2out.cc (index_rnglists): For !HAVE_AS_LEB128 and
30302 block_num > 0, index entry even if !have_multiple_function_sections.
30304 2022-01-20 liuhongt <hongtao.liu@intel.com>
30307 * tree-vect-stmts.cc (supportable_narrowing_operation): Enhance
30308 integral mode mask pack by multi steps which takes
30309 vec_pack_sbool_trunc_optab as start when elements number is
30310 less than BITS_PER_UNITS.
30312 2022-01-20 Richard Biener <rguenther@suse.de>
30314 PR tree-optimization/104114
30315 * tree-vect-generic.cc (expand_vector_piecewise): Do not diagnose
30316 single element vector decomposition.
30318 2022-01-19 Robin Dapp <rdapp@linux.ibm.com>
30320 * ifcvt.cc (noce_convert_multiple_sets_1): New function.
30321 (noce_convert_multiple_sets): Call function a second time if we can
30322 improve the first try.
30324 2022-01-19 Robin Dapp <rdapp@linux.ibm.com>
30326 * ifcvt.cc (cond_exec_get_condition): New parameter to allow getting the
30327 reversed comparison.
30328 (try_emit_cmove_seq): New function to facilitate creating a cmov
30330 (noce_convert_multiple_sets): Create two sequences and use the less
30333 2022-01-19 Robin Dapp <rdapp@linux.ibm.com>
30335 * rtl.h (struct rtx_comparison): New struct that holds an rtx
30337 * config/rs6000/rs6000.cc (rs6000_emit_minmax): Use struct instead of
30339 (rs6000_emit_swsqrt): Likewise.
30340 * expmed.cc (expand_sdiv_pow2): Likewise.
30341 (emit_store_flag): Likewise.
30342 * expr.cc (expand_cond_expr_using_cmove): Likewise.
30343 (expand_expr_real_2): Likewise.
30344 * ifcvt.cc (noce_emit_cmove): Add compare and reversed compare
30346 * optabs.cc (emit_conditional_move_1): New function.
30347 (expand_doubleword_shift_condmove): Use struct.
30348 (emit_conditional_move): Use struct and allow to call directly
30349 without going through preparation steps.
30350 * optabs.h (emit_conditional_move): Use struct.
30352 2022-01-19 Robin Dapp <rdapp@linux.ibm.com>
30354 * ifcvt.cc (bb_ok_for_noce_convert_multiple_sets): Estimate insns costs.
30355 (noce_process_if_block): Use potential costs.
30357 2022-01-19 Robin Dapp <rdapp@linux.ibm.com>
30359 * ifcvt.cc (noce_convert_multiple_sets): Allow constants.
30360 (bb_ok_for_noce_convert_multiple_sets): Likewise.
30362 2022-01-19 Robin Dapp <rdapp@linux.ibm.com>
30364 * ifcvt.cc (need_cmov_or_rewire): New function.
30365 (noce_convert_multiple_sets): Call it.
30367 2022-01-19 David Malcolm <dmalcolm@redhat.com>
30369 * attribs.cc (attribute_c_tests): Rename to...
30370 (attribs_cc_tests): ...this.
30371 * bitmap.cc (bitmap_c_tests): Rename to...
30372 (bitmap_cc_tests): ...this.
30373 * cgraph.cc (cgraph_c_finalize): Rename to...
30374 (cgraph_cc_finalize): ...this.
30375 (cgraph_c_tests): Rename to...
30376 (cgraph_cc_tests): ...this.
30377 * cgraph.h (cgraph_c_finalize): Rename to...
30378 (cgraph_cc_finalize): ...this.
30379 (cgraphunit_c_finalize): Rename to...
30380 (cgraphunit_cc_finalize): ...this.
30381 * cgraphunit.cc (cgraphunit_c_finalize): Rename to...
30382 (cgraphunit_cc_finalize): ...this.
30383 * convert.cc (convert_c_tests): Rename to...
30384 (convert_cc_tests): ...this.
30385 * dbgcnt.cc (dbgcnt_c_tests): Rename to...
30386 (dbgcnt_cc_tests): ...this.
30387 * diagnostic-show-locus.cc (diagnostic_show_locus_c_tests): Rename to...
30388 (diagnostic_show_locus_cc_tests): ...this.
30389 * diagnostic.cc (diagnostic_c_tests): Rename to...
30390 (diagnostic_cc_tests): ...this.
30391 * dumpfile.cc (dumpfile_c_tests): Rename to...
30392 (dumpfile_cc_tests): ...this.
30393 * dwarf2out.cc (dwarf2out_c_finalize): Rename to...
30394 (dwarf2out_cc_finalize): ...this.
30395 * dwarf2out.h (dwarf2out_c_finalize): Rename to...
30396 (dwarf2out_cc_finalize): ...this.
30397 * edit-context.cc (edit_context_c_tests): Rename to...
30398 (edit_context_cc_tests): ...this.
30399 * et-forest.cc (et_forest_c_tests): Rename to...
30400 (et_forest_cc_tests): ...this.
30401 * fibonacci_heap.cc (fibonacci_heap_c_tests): Rename to...
30402 (fibonacci_heap_cc_tests): ...this.
30403 * fold-const.cc (fold_const_c_tests): Rename to...
30404 (fold_const_cc_tests): ...this.
30405 * function-tests.cc (function_tests_c_tests): Rename to...
30406 (function_tests_cc_tests): ...this.
30407 * gcse.cc (gcse_c_finalize): Rename to...
30408 (gcse_cc_finalize): ...this.
30409 * gcse.h (gcse_c_finalize): Rename to...
30410 (gcse_cc_finalize): ...this.
30411 * ggc-tests.cc (ggc_tests_c_tests): Rename to...
30412 (ggc_tests_cc_tests): ...this.
30413 * gimple-ssa-store-merging.cc (store_merging_c_tests): Rename to...
30414 (store_merging_cc_tests): ...this.
30415 * gimple.cc (gimple_c_tests): Rename to...
30416 (gimple_cc_tests): ...this.
30417 * hash-map-tests.cc (hash_map_tests_c_tests): Rename to...
30418 (hash_map_tests_cc_tests): ...this.
30419 * hash-set-tests.cc (hash_set_tests_c_tests): Rename to...
30420 (hash_set_tests_cc_tests): ...this.
30421 * input.cc (input_c_tests): Rename to...
30422 (input_cc_tests): ...this.
30423 * ipa-cp.cc (ipa_cp_c_finalize): Rename to...
30424 (ipa_cp_cc_finalize): ...this.
30425 * ipa-fnsummary.cc (ipa_fnsummary_c_finalize): Rename to...
30426 (ipa_fnsummary_cc_finalize): ...this.
30427 * ipa-fnsummary.h (ipa_fnsummary_c_finalize): Rename to...
30428 (ipa_fnsummary_cc_finalize): ...this.
30429 * ipa-modref-tree.cc (ipa_modref_tree_c_tests): Rename to...
30430 (ipa_modref_tree_cc_tests): ...this.
30431 * ipa-modref-tree.h (modref_c_tests): Delete bogus decl.
30432 * ipa-modref.cc (ipa_modref_c_finalize): Rename to...
30433 (ipa_modref_cc_finalize): ...this.
30434 * ipa-modref.h (ipa_modref_c_finalize): Rename to...
30435 (ipa_modref_cc_finalize): ...this.
30436 * ipa-prop.h (ipa_cp_c_finalize): Rename to...
30437 (ipa_cp_cc_finalize): ...this.
30438 * ipa-reference.cc (ipa_reference_c_finalize): Rename to...
30439 (ipa_reference_cc_finalize): ...this.
30440 * ipa-reference.h (ipa_reference_c_finalize): Rename to...
30441 (ipa_reference_cc_finalize): ...this.
30442 * ira-costs.cc (ira_costs_c_finalize): Rename to...
30443 (ira_costs_cc_finalize): ...this.
30444 * ira.h (ira_costs_c_finalize): Rename to...
30445 (ira_costs_cc_finalize): ...this.
30446 * opt-suggestions.cc (opt_proposer_c_tests): Rename to...
30447 (opt_suggestions_cc_tests): ...this.
30448 * opts.cc (opts_c_tests): Rename to...
30449 (opts_cc_tests): ...this.
30450 * predict.cc (predict_c_tests): Rename to...
30451 (predict_cc_tests): ...this.
30452 * pretty-print.cc (pretty_print_c_tests): Rename to...
30453 (pretty_print_cc_tests): ...this.
30454 * read-rtl-function.cc (read_rtl_function_c_tests): Rename to...
30455 (read_rtl_function_cc_tests): ...this.
30456 * rtl-tests.cc (rtl_tests_c_tests): Rename to...
30457 (rtl_tests_cc_tests): ...this.
30458 * sbitmap.cc (sbitmap_c_tests): Rename to...
30459 (sbitmap_cc_tests): ...this.
30460 * selftest-run-tests.cc (selftest::run_tests): Update calls for
30461 _c_ to _cc_ function renamings; fix names of attribs and
30462 opt-suggestions tests.
30463 * selftest.cc (selftest_c_tests): Rename to...
30464 (selftest_cc_tests): ...this.
30465 * selftest.h (attribute_c_tests): Rename to...
30466 (attribs_cc_tests): ...this.
30467 (bitmap_c_tests): Rename to...
30468 (bitmap_cc_tests): ...this.
30469 (cgraph_c_tests): Rename to...
30470 (cgraph_cc_tests): ...this.
30471 (convert_c_tests): Rename to...
30472 (convert_cc_tests): ...this.
30473 (diagnostic_c_tests): Rename to...
30474 (diagnostic_cc_tests): ...this.
30475 (diagnostic_show_locus_c_tests): Rename to...
30476 (diagnostic_show_locus_cc_tests): ...this.
30477 (dumpfile_c_tests): Rename to...
30478 (dumpfile_cc_tests): ...this.
30479 (edit_context_c_tests): Rename to...
30480 (edit_context_cc_tests): ...this.
30481 (et_forest_c_tests): Rename to...
30482 (et_forest_cc_tests): ...this.
30483 (fibonacci_heap_c_tests): Rename to...
30484 (fibonacci_heap_cc_tests): ...this.
30485 (fold_const_c_tests): Rename to...
30486 (fold_const_cc_tests): ...this.
30487 (function_tests_c_tests): Rename to...
30488 (function_tests_cc_tests): ...this.
30489 (ggc_tests_c_tests): Rename to...
30490 (ggc_tests_cc_tests): ...this.
30491 (gimple_c_tests): Rename to...
30492 (gimple_cc_tests): ...this.
30493 (hash_map_tests_c_tests): Rename to...
30494 (hash_map_tests_cc_tests): ...this.
30495 (hash_set_tests_c_tests): Rename to...
30496 (hash_set_tests_cc_tests): ...this.
30497 (input_c_tests): Rename to...
30498 (input_cc_tests): ...this.
30499 (opts_c_tests): Rename to...
30500 (opts_cc_tests): ...this.
30501 (predict_c_tests): Rename to...
30502 (predict_cc_tests): ...this.
30503 (pretty_print_c_tests): Rename to...
30504 (pretty_print_cc_tests): ...this.
30505 (read_rtl_function_c_tests): Rename to...
30506 (read_rtl_function_cc_tests): ...this.
30507 (rtl_tests_c_tests): Rename to...
30508 (rtl_tests_cc_tests): ...this.
30509 (sbitmap_c_tests): Rename to...
30510 (sbitmap_cc_tests): ...this.
30511 (selftest_c_tests): Rename to...
30512 (selftest_cc_tests): ...this.
30513 (simplify_rtx_c_tests): Rename to...
30514 (simplify_rtx_cc_tests): ...this.
30515 (spellcheck_c_tests): Rename to...
30516 (spellcheck_cc_tests): ...this.
30517 (spellcheck_tree_c_tests): Rename to...
30518 (spellcheck_tree_cc_tests): ...this.
30519 (sreal_c_tests): Rename to...
30520 (sreal_cc_tests): ...this.
30521 (store_merging_c_tests): Rename to...
30522 (store_merging_cc_tests): ...this.
30523 (tree_c_tests): Rename to...
30524 (tree_cc_tests): ...this.
30525 (tree_cfg_c_tests): Rename to...
30526 (tree_cfg_cc_tests): ...this.
30527 (typed_splay_tree_c_tests): Rename to...
30528 (typed_splay_tree_cc_tests): ...this.
30529 (vec_c_tests): Rename to...
30530 (vec_cc_tests): ...this.
30531 (vec_perm_indices_c_tests): Rename to...
30532 (vec_perm_indices_cc_tests): ..this.
30533 (opt_proposer_c_tests): Rename to...
30534 (opt_suggestions_cc_tests): ...this.
30535 (dbgcnt_c_tests): Rename to...
30536 (dbgcnt_cc_tests): ...this.
30537 (ipa_modref_tree_c_tests): Rename to...
30538 (ipa_modref_tree_cc_tests): ...this.
30539 * simplify-rtx.cc (simplify_rtx_c_tests): Rename to...
30540 (simplify_rtx_cc_tests): ...this.
30541 * spellcheck-tree.cc (spellcheck_tree_c_tests): Rename to...
30542 (spellcheck_tree_cc_tests): ...this.
30543 * spellcheck.cc (spellcheck_c_tests): Rename to...
30544 (spellcheck_cc_tests): ...this.
30545 * sreal.cc (sreal_c_tests): Rename to...
30546 (sreal_cc_tests): ...this.
30547 * toplev.cc (toplev::finalize): Update calls for _c_ to _cc_
30548 function renamings.
30549 * tree-cfg.cc (tree_cfg_c_tests): Rename to...
30550 (tree_cfg_cc_tests): ...this.
30551 * tree.cc (tree_c_tests): Rename to...
30552 (tree_cc_tests): ...this.
30553 * typed-splay-tree.cc (typed_splay_tree_c_tests): Rename to...
30554 (typed_splay_tree_cc_tests): ...this.
30555 * vec-perm-indices.cc (vec_perm_indices_c_tests): Rename to...
30556 (vec_perm_indices_cc_tests): ...this.
30557 * vec.cc (vec_c_tests): Rename to...
30558 (vec_cc_tests): ...this.
30560 2022-01-19 Andre Vieira <andre.simoesdiasvieira@arm.com>
30562 PR tree-optimization/103997
30563 * tree-vect-loop.cc (vect_analyze_loop): Fix mode skipping for epilogue
30566 2022-01-19 Jakub Jelinek <jakub@redhat.com>
30568 PR middle-end/102860
30569 * match.pd (x %[fl] y -> x % y): New simplification for
30570 unsigned integral types.
30571 * optabs-tree.cc (optab_for_tree_code): Return unknown_optab
30572 for {CEIL,FLOOR,ROUND}_{DIV,MOD}_EXPR with VECTOR_TYPE.
30574 2022-01-19 Richard Biener <rguenther@suse.de>
30576 PR tree-optimization/104112
30577 * tree-vect-loop.cc (vect_find_reusable_accumulator): Check
30578 for required intermediate vector types.
30580 2022-01-19 Jakub Jelinek <jakub@redhat.com>
30582 * config/rs6000/rs6000.cc (rs6000_machine_from_flags): Add default:.
30584 2022-01-19 Martin Liska <mliska@suse.cz>
30586 * configure.ac: Remove -Wno-error=format-diag.
30587 * configure: Regenerate.
30589 2022-01-19 Martin Liska <mliska@suse.cz>
30591 * config/riscv/riscv.cc (riscv_handle_type_attribute):
30592 Update one -Wformat-diag string in warning message.
30594 2022-01-19 Jakub Jelinek <jakub@redhat.com>
30596 PR middle-end/104103
30597 * gimple-ssa-warn-access.cc (pass_waccess::check_call): Don't check
30600 2022-01-19 Jakub Jelinek <jakub@redhat.com>
30603 * fold-const.cc (address_compare): Consider different STRING_CSTs
30604 with the same lengths that memcmp the same as equal, not different.
30606 2022-01-19 Jakub Jelinek <jakub@redhat.com>
30608 * config/i386/sse.md (*aes<aeswideklvariant>u*): Use %0 instead of
30611 2022-01-19 Martin Liska <mliska@suse.cz>
30612 Thomas Schwinge <thomas@codesourcery.com>
30614 * config/nvptx/nvptx.cc (nvptx_goacc_validate_dims_1): Update
30617 2022-01-19 Sebastian Huber <sebastian.huber@embedded-brains.de>
30620 * config/rs6000/rs6000.cc (rs6000_machine_from_flags): Use also
30623 2022-01-19 Jakub Jelinek <jakub@redhat.com>
30626 * config/i386/sse.md
30627 (<avx512>_<complexopname>_<mode><maskc_name><round_name>,
30628 avx512fp16_<complexopname>sh_v8hf<mask_scalarc_name><round_scalarcz_name>,
30629 avx512dq_mul<mode>3<mask_name>, <avx2_avx512>_permvar<mode><mask_name>,
30630 avx2_perm<mode>_1<mask_name>, avx512f_perm<mode>_1<mask_name>,
30631 avx512dq_rangep<mode><mask_name><round_saeonly_name>,
30632 avx512dq_ranges<mode><mask_scalar_name><round_saeonly_scalar_name>,
30633 <avx512>_getmant<mode><mask_name><round_saeonly_name>,
30634 avx512f_vgetmant<mode><mask_scalar_name><round_saeonly_scalar_name>):
30635 Use vxorps\t%x0, %x0, %x0 instead of vxorps\t{%x0, %x0, %x0}.
30637 2022-01-19 Martin Sebor <msebor@redhat.com>
30639 PR middle-end/104069
30640 * gimple-ssa-warn-access.cc (pointers_related_p): Return false for
30641 an unknown result as documented.
30643 2022-01-18 Andrew Pinski <apinski@marvell.com>
30645 * ipa-split.cc (visit_bb): Fix comment before the
30646 warning/error attribute checking code.
30648 2022-01-18 David Faust <david.faust@oracle.com>
30650 * config/bpf/coreout.cc (bpf_core_reloc_add): Do not account
30651 for base strtab offset yet as it may change.
30652 (output_asm_btfext_core_reloc): Do so here instead.
30653 (output_btfext_core_sections): Likewise.
30655 2022-01-18 David Faust <david.faust@oracle.com>
30657 * config/bpf/coreout.cc (output_btfext_header): Account for
30658 4-byte record size in core_relo_len.
30659 (output_btfext_core_sections): Only write record size once.
30660 * config/bpf/coreout.h (btf_ext_section_header): Delete unused
30663 2022-01-18 Maciej W. Rozycki <macro@embecosm.com>
30665 * common/config/riscv/riscv-common.cc
30666 (riscv_subset_list::parse_multiletter_ext): Move pointer
30667 arithmetic ahead of `free'.
30669 2022-01-18 Jason Merrill <jason@redhat.com>
30672 * gimplify.cc (gimple_push_cleanup): Handle eh_only in conditional
30675 2022-01-18 Sandra Loosemore <sandra@codesourcery.com>
30677 PR middle-end/103163
30678 * emit-rtl.cc (init_emit_regs): Initialize stack_limit_rtx here...
30679 (init_emit_once): ...not here.
30681 2022-01-18 Martin Liska <mliska@suse.cz>
30683 * collect2.cc (scan_libraries): Fix -Wformat-diag issues.
30684 * config/aarch64/aarch64-builtins.cc (aarch64_simd_expand_builtin): Likewise.
30685 * config/arc/arc.md: Likewise.
30686 * config/avr/avr.cc (avr_section_type_flags): Likewise.
30687 * config/bfin/bfin.cc (bfin_option_override): Likewise.
30688 (bfin_handle_longcall_attribute): Likewise.
30689 * config/cris/cris.h (FUNCTION_PROFILER): Likewise.
30690 * config/frv/frv.cc (frv_expand_builtin): Likewise.
30691 * config/ia64/ia64-c.cc (ia64_hpux_handle_builtin_pragma): Likewise.
30692 * config/iq2000/iq2000.cc (save_restore_insns): Likewise.
30693 (iq2000_print_operand_address): Likewise.
30694 (iq2000_print_operand): Likewise.
30695 * config/m32c/m32c-pragma.cc (m32c_pragma_memregs): Likewise.
30696 (m32c_pragma_address): Likewise.
30697 * config/m68k/m68k.cc (m68k_handle_fndecl_attribute): Likewise.
30698 * config/mips/mips.cc (mips_handle_interrupt_attr): Likewise.
30699 (mips_set_compression_mode): Likewise.
30700 * config/mmix/mmix.cc (mmix_function_profiler): Likewise.
30701 (mmix_print_operand): Likewise.
30702 (mmix_output_shiftvalue_op_from_str): Likewise.
30703 (mmix_output_shifted_value): Likewise.
30704 * config/msp430/driver-msp430.cc (msp430_select_hwmult_lib): Likewise.
30705 * config/msp430/msp430.cc (msp430_option_override): Likewise.
30706 (msp430_attr): Likewise.
30707 (msp430_expand_delay_cycles): Likewise.
30708 (msp430_expand_builtin): Likewise.
30709 * config/rs6000/aix73.h: Likewise.
30710 * config/rs6000/rtems.h (INVALID_64BIT): Likewise.
30711 * config/rx/rx.cc (rx_expand_builtin_mvtc): Likewise.
30712 (valid_psw_flag): Likewise.
30713 * config/sh/sh.cc (parse_validate_atomic_model_option): Likewise.
30714 * config/stormy16/stormy16.cc (xstormy16_function_profiler): Likewise.
30715 (xstormy16_expand_builtin_va_start): Likewise.
30716 (xstormy16_handle_below100_attribute): Likewise.
30718 2022-01-18 Martin Liska <mliska@suse.cz>
30720 * config/vms/vms-c.cc (vms_pragma_nostandard): Fix -Wformat-diag
30722 (vms_pragma_standard): Likewise.
30723 (vms_pragma_extern_prefix): Likewise.
30725 2022-01-18 Martin Liska <mliska@suse.cz>
30727 * config/xtensa/xtensa.cc (print_operand): Fix warnings.
30728 (print_operand_address): Likewise.
30729 (xtensa_multibss_section_type_flags): Likewise.
30731 2022-01-18 Martin Liska <mliska@suse.cz>
30733 * config/rs6000/rs6000-call.cc (rs6000_invalid_builtin): Change
30734 wording of an error message.
30736 2022-01-18 Martin Liska <mliska@suse.cz>
30738 * config/v850/v850-c.cc (pop_data_area): Fix -Wformat-diag
30740 (ghs_pragma_section): Likewise.
30741 (ghs_pragma_interrupt): Likewise.
30742 (ghs_pragma_starttda): Likewise.
30743 (ghs_pragma_startsda): Likewise.
30744 (ghs_pragma_startzda): Likewise.
30745 (ghs_pragma_endtda): Likewise.
30746 (ghs_pragma_endsda): Likewise.
30747 (ghs_pragma_endzda): Likewise.
30749 2022-01-18 Martin Liska <mliska@suse.cz>
30751 * config/nds32/nds32-intrinsic.cc (nds32_expand_builtin_impl):
30753 * config/nds32/nds32-intrinsic.md: Likewise.
30754 * config/nds32/nds32-isr.cc (nds32_check_isr_attrs_conflict): Likewise.
30755 * config/nds32/nds32.cc (nds32_print_operand): Likewise.
30756 (nds32_insert_attributes): Likewise.
30758 2022-01-18 Martin Liska <mliska@suse.cz>
30760 * config/nvptx/nvptx.cc (nvptx_goacc_validate_dims_1): Wrap
30762 * config/nvptx/nvptx.md: Remove trailing dot.
30764 2022-01-18 Martin Liska <mliska@suse.cz>
30766 * common/config/riscv/riscv-common.cc (riscv_subset_list::add):
30767 Wrap keywords with quotes and remove trailing dots.
30768 (riscv_subset_list::parsing_subset_version): Likewise.
30769 (riscv_subset_list::parse_std_ext): Likewise.
30770 (riscv_subset_list::parse_multiletter_ext): Likewise.
30771 * config/riscv/riscv.cc (riscv_handle_type_attribute): Likewise.
30773 2022-01-18 Andre Vieira <andre.simoesdiasvieira@arm.com>
30775 * tree-vect-loop.cc (vect_estimate_min_profitable_iters): Pass new
30776 argument suggested_unroll_factor.
30777 (vect_analyze_loop_costing): Likewise.
30778 (_loop_vec_info::_loop_vec_info): Initialize new member
30779 suggested_unroll_factor.
30780 (vect_determine_partial_vectors_and_peeling): Make epilogue of unrolled
30781 main loop use partial vectors.
30782 (vect_analyze_loop_2): Pass and use new argument
30783 suggested_unroll_factor.
30784 (vect_analyze_loop_1): Change to intialize local
30785 suggested_unroll_factor and use it.
30786 (vectorizable_reduction): Don't use single_defuse_cycle when unrolling.
30787 * tree-vectorizer.h (_loop_vec_info::_loop_vec_info): Add new member
30788 suggested_unroll_factor.
30789 (vector_costs::vector_costs): Add new member m_suggested_unroll_factor.
30790 (vector_costs::suggested_unroll_factor): New getter function.
30791 (finish_cost): Set return argument suggested_unroll_factor.
30793 2022-01-18 Andrew MacLeod <amacleod@redhat.com>
30795 PR tree-optimization/104038
30796 * doc/invoke.texi (relation-block-limit): New.
30797 * params.opt (relation-block-limit): New.
30798 * value-relation.cc (dom_oracle::register_relation): Check for NULL
30799 record before invoking transitive registery.
30800 (dom_oracle::set_one_relation): Check limit before creating record.
30801 (dom_oracle::register_transitives): Stop when no record created.
30802 * value-relation.h (relation_chain_head::m_num_relations): New.
30804 2022-01-18 Richard Biener <rguenther@suse.de>
30807 * ipa-inline.cc (inline_small_functions): Do not enqueue call
30808 edges originating in functions compiled with -Og.
30810 2022-01-18 Richard Biener <rguenther@suse.de>
30813 * passes.def (pass_all_optimizations_g): Remove pass_modref
30814 and pass_local_pure_const.
30816 2022-01-18 Martin Liska <mliska@suse.cz>
30818 * config/s390/s390.cc: Fix -Wformat-diag warnings.
30820 2022-01-18 Martin Liska <mliska@suse.cz>
30822 * config/s390/s390-c.cc (s390_expand_overloaded_builtin): Wrap
30824 (s390_resolve_overloaded_builtin): Remove trailing dot.
30825 * config/s390/s390.cc (s390_const_operand_ok): Use - for range.
30826 (s390_expand_builtin): Remove trailing dot.
30827 (s390_emit_prologue): Likewise, use semicolon.
30828 (s390_option_override_internal): Update keyword.
30829 * varasm.cc (do_assemble_alias): Wrap keyword in quotes.
30831 2022-01-18 Martin Liska <mliska@suse.cz>
30833 * config/rs6000/rs6000-call.cc (rs6000_expand_builtin): Wrap
30834 keywords and use %qs instead of %<%s%>.
30836 2022-01-18 Richard Biener <rguenther@suse.de>
30838 PR tree-optimization/103987
30839 * tree-ssa-dse.cc (dse_optimize_call): Properly guard modref
30840 query with a pointer check.
30842 2022-01-18 Richard Sandiford <richard.sandiford@arm.com>
30845 * config/aarch64/aarch64.cc (aarch64_check_consecutive_mems):
30846 When using MEM_EXPR, require the base to be a decl.
30848 2022-01-18 Richard Biener <rguenther@suse.de>
30850 * cgraph.h (struct cgraph_simd_clone_arg): Re-arrange fields to
30852 * function.h (struct function): Likewise.
30854 2022-01-18 Arnaud Charlet <charlet@adacore.com>
30856 * doc/install.texi: Update prerequisites for GNAT
30858 2022-01-18 Andrew Pinski <apinski@marvell.com>
30860 PR tree-optimization/101941
30861 * ipa-split.cc (visit_bb): Disallow function calls where
30862 the function has either error or warning attribute.
30864 2022-01-18 Richard Biener <rguenther@suse.de>
30866 PR tree-optimization/104064
30867 * tree-vect-data-refs.cc (vect_analyze_data_ref_accesses): Check
30868 DR_INIT fits in a signed HWI, represent the difference from the
30869 first DR in unsigned.
30871 2022-01-17 Martin Liska <mliska@suse.cz>
30873 * Makefile.in: Rename .c names to .cc.
30874 * config.gcc: Likewise.
30875 * configure: Regenerate. Likewise.
30876 * configure.ac: Likewise.
30877 * gengtype.cc (set_gc_used): Likewise.
30878 (source_dot_c_frul): Likewise.
30879 (source_dot_cc_frul): Likewise.
30880 (struct file_rule_st): Likewise.
30881 (close_output_files): Likewise.
30882 * config/avr/t-avr: Use CXXFLAGS_* and CXX_FOR_BUILD.
30884 2022-01-17 Martin Liska <mliska@suse.cz>
30886 * Makefile.in: Rename .c names to .cc.
30887 * alias.h: Likewise.
30888 * asan.cc: Likewise.
30889 * auto-profile.h: Likewise.
30890 * basic-block.h (struct basic_block_d): Likewise.
30891 * btfout.cc: Likewise.
30892 * builtins.cc (expand_builtin_longjmp): Likewise.
30893 (validate_arg): Likewise.
30894 (access_ref::offset_bounded): Likewise.
30895 * caller-save.cc (reg_restore_code): Likewise.
30896 (setup_save_areas): Likewise.
30897 * calls.cc (initialize_argument_information): Likewise.
30898 (expand_call): Likewise.
30899 (emit_library_call_value_1): Likewise.
30900 * cfg-flags.def (RTL): Likewise.
30901 (SIBCALL): Likewise.
30902 (CAN_FALLTHRU): Likewise.
30903 * cfganal.cc (post_order_compute): Likewise.
30904 * cfgcleanup.cc (try_simplify_condjump): Likewise.
30905 (merge_blocks_move_predecessor_nojumps): Likewise.
30906 (merge_blocks_move_successor_nojumps): Likewise.
30907 (merge_blocks_move): Likewise.
30908 (old_insns_match_p): Likewise.
30909 (try_crossjump_bb): Likewise.
30910 * cfgexpand.cc (expand_gimple_stmt): Likewise.
30911 * cfghooks.cc (split_block_before_cond_jump): Likewise.
30912 (profile_record_check_consistency): Likewise.
30913 * cfghooks.h: Likewise.
30914 * cfgrtl.cc (pass_free_cfg::execute): Likewise.
30915 (rtl_can_merge_blocks): Likewise.
30916 (try_redirect_by_replacing_jump): Likewise.
30917 (make_pass_outof_cfg_layout_mode): Likewise.
30918 (cfg_layout_can_merge_blocks_p): Likewise.
30919 * cgraph.cc (release_function_body): Likewise.
30920 (cgraph_node::get_fun): Likewise.
30921 * cgraph.h (struct cgraph_node): Likewise.
30922 (asmname_hasher::equal): Likewise.
30923 (cgraph_inline_failed_type): Likewise.
30924 (thunk_adjust): Likewise.
30925 (dump_callgraph_transformation): Likewise.
30926 (record_references_in_initializer): Likewise.
30927 (ipa_discover_variable_flags): Likewise.
30928 * cgraphclones.cc (GTY): Likewise.
30929 * cgraphunit.cc (symbol_table::finalize_compilation_unit): Likewise.
30930 * collect-utils.h (GCC_COLLECT_UTILS_H): Likewise.
30931 * collect2-aix.h (GCC_COLLECT2_AIX_H): Likewise.
30932 * collect2.cc (maybe_run_lto_and_relink): Likewise.
30933 * combine-stack-adj.cc: Likewise.
30934 * combine.cc (setup_incoming_promotions): Likewise.
30935 (combine_simplify_rtx): Likewise.
30936 (count_rtxs): Likewise.
30937 * common.opt: Likewise.
30938 * common/config/aarch64/aarch64-common.cc: Likewise.
30939 * common/config/arm/arm-common.cc (arm_asm_auto_mfpu): Likewise.
30940 * common/config/avr/avr-common.cc: Likewise.
30941 * common/config/i386/i386-isas.h (struct _isa_names_table): Likewise.
30942 * conditions.h: Likewise.
30943 * config.gcc: Likewise.
30944 * config/aarch64/aarch64-builtins.cc (aarch64_resolve_overloaded_memtag): Likewise.
30945 * config/aarch64/aarch64-protos.h (aarch64_classify_address): Likewise.
30946 (aarch64_get_extension_string_for_isa_flags): Likewise.
30947 * config/aarch64/aarch64-sve-builtins.cc (function_builder::add_function): Likewise.
30948 * config/aarch64/aarch64.cc (aarch64_regmode_natural_size): Likewise.
30949 (aarch64_sched_first_cycle_multipass_dfa_lookahead): Likewise.
30950 (aarch64_option_valid_attribute_p): Likewise.
30951 (aarch64_short_vector_p): Likewise.
30952 (aarch64_float_const_representable_p): Likewise.
30953 * config/aarch64/aarch64.h (DBX_REGISTER_NUMBER): Likewise.
30954 (ASM_OUTPUT_POOL_EPILOGUE): Likewise.
30956 * config/aarch64/cortex-a57-fma-steering.cc: Likewise.
30957 * config/aarch64/driver-aarch64.cc (contains_core_p): Likewise.
30958 * config/aarch64/t-aarch64: Likewise.
30959 * config/aarch64/x-aarch64: Likewise.
30960 * config/aarch64/x-darwin: Likewise.
30961 * config/alpha/alpha-protos.h: Likewise.
30962 * config/alpha/alpha.cc (alpha_scalar_mode_supported_p): Likewise.
30963 * config/alpha/alpha.h (LONG_DOUBLE_TYPE_SIZE): Likewise.
30964 (enum reg_class): Likewise.
30965 * config/alpha/alpha.md: Likewise.
30966 * config/alpha/driver-alpha.cc (AMASK_LOCKPFTCHOK): Likewise.
30967 * config/alpha/x-alpha: Likewise.
30968 * config/arc/arc-protos.h (arc_eh_uses): Likewise.
30969 * config/arc/arc.cc (ARC_OPT): Likewise.
30970 (arc_ccfsm_advance): Likewise.
30971 (arc_arg_partial_bytes): Likewise.
30972 (conditionalize_nonjump): Likewise.
30973 * config/arc/arc.md: Likewise.
30974 * config/arc/builtins.def: Likewise.
30975 * config/arc/t-arc: Likewise.
30976 * config/arm/arm-c.cc (arm_resolve_overloaded_builtin): Likewise.
30977 (arm_pragma_target_parse): Likewise.
30978 * config/arm/arm-protos.h (save_restore_target_globals): Likewise.
30979 (arm_cpu_cpp_builtins): Likewise.
30980 * config/arm/arm.cc (vfp3_const_double_index): Likewise.
30981 (shift_op): Likewise.
30982 (thumb2_final_prescan_insn): Likewise.
30983 (arm_final_prescan_insn): Likewise.
30984 (arm_asm_output_labelref): Likewise.
30985 (arm_small_register_classes_for_mode_p): Likewise.
30986 * config/arm/arm.h: Likewise.
30987 * config/arm/arm.md: Likewise.
30988 * config/arm/driver-arm.cc: Likewise.
30989 * config/arm/symbian.h: Likewise.
30990 * config/arm/t-arm: Likewise.
30991 * config/arm/thumb1.md: Likewise.
30992 * config/arm/x-arm: Likewise.
30993 * config/avr/avr-c.cc (avr_register_target_pragmas): Likewise.
30994 * config/avr/avr-fixed.md: Likewise.
30995 * config/avr/avr-log.cc (avr_log_vadump): Likewise.
30996 * config/avr/avr-mcus.def: Likewise.
30997 * config/avr/avr-modes.def (FRACTIONAL_INT_MODE): Likewise.
30998 * config/avr/avr-passes.def (INSERT_PASS_BEFORE): Likewise.
30999 * config/avr/avr-protos.h (make_avr_pass_casesi): Likewise.
31000 * config/avr/avr.cc (avr_option_override): Likewise.
31001 (avr_build_builtin_va_list): Likewise.
31002 (avr_mode_dependent_address_p): Likewise.
31003 (avr_function_arg_advance): Likewise.
31004 (avr_asm_output_aligned_decl_common): Likewise.
31005 * config/avr/avr.h (RETURN_ADDR_RTX): Likewise.
31006 (SUPPORTS_INIT_PRIORITY): Likewise.
31007 * config/avr/avr.md: Likewise.
31008 * config/avr/builtins.def: Likewise.
31009 * config/avr/gen-avr-mmcu-specs.cc (IN_GEN_AVR_MMCU_TEXI): Likewise.
31010 * config/avr/gen-avr-mmcu-texi.cc (IN_GEN_AVR_MMCU_TEXI): Likewise.
31012 * config/avr/t-avr: Likewise.
31013 * config/bfin/bfin.cc (frame_related_constant_load): Likewise.
31014 * config/bpf/bpf-protos.h (GCC_BPF_PROTOS_H): Likewise.
31015 * config/bpf/bpf.h (enum reg_class): Likewise.
31016 * config/bpf/t-bpf: Likewise.
31017 * config/c6x/c6x-protos.h (GCC_C6X_PROTOS_H): Likewise.
31018 * config/cr16/cr16-protos.h: Likewise.
31019 * config/cris/cris.cc (cris_address_cost): Likewise.
31020 (cris_side_effect_mode_ok): Likewise.
31021 (cris_init_machine_status): Likewise.
31022 (cris_emit_movem_store): Likewise.
31023 * config/cris/cris.h (INDEX_REG_CLASS): Likewise.
31024 (enum reg_class): Likewise.
31025 (struct cum_args): Likewise.
31026 * config/cris/cris.opt: Likewise.
31027 * config/cris/sync.md: Likewise.
31028 * config/csky/csky.cc (csky_expand_prologue): Likewise.
31029 * config/darwin-c.cc: Likewise.
31030 * config/darwin-f.cc: Likewise.
31031 * config/darwin-sections.def (zobj_const_section): Likewise.
31032 * config/darwin.cc (output_objc_section_asm_op): Likewise.
31033 (fprintf): Likewise.
31034 * config/darwin.h (GTY): Likewise.
31035 * config/elfos.h: Likewise.
31036 * config/epiphany/epiphany-sched.md: Likewise.
31037 * config/epiphany/epiphany.cc (epiphany_function_value): Likewise.
31038 * config/epiphany/epiphany.h (GTY): Likewise.
31039 (NO_FUNCTION_CSE): Likewise.
31040 * config/epiphany/mode-switch-use.cc: Likewise.
31041 * config/epiphany/predicates.md: Likewise.
31042 * config/epiphany/t-epiphany: Likewise.
31043 * config/fr30/fr30-protos.h: Likewise.
31044 * config/frv/frv-protos.h: Likewise.
31045 * config/frv/frv.cc (TLS_BIAS): Likewise.
31046 * config/frv/frv.h (ASM_OUTPUT_ALIGNED_LOCAL): Likewise.
31047 * config/ft32/ft32-protos.h: Likewise.
31048 * config/gcn/gcn-hsa.h (ASM_APP_OFF): Likewise.
31049 * config/gcn/gcn.cc (gcn_init_libfuncs): Likewise.
31050 * config/gcn/mkoffload.cc (copy_early_debug_info): Likewise.
31051 * config/gcn/t-gcn-hsa: Likewise.
31052 * config/gcn/t-omp-device: Likewise.
31053 * config/h8300/h8300-protos.h (GCC_H8300_PROTOS_H): Likewise.
31054 (same_cmp_following_p): Likewise.
31055 * config/h8300/h8300.cc (F): Likewise.
31056 * config/h8300/h8300.h (struct cum_arg): Likewise.
31057 (BRANCH_COST): Likewise.
31058 * config/i386/cygming.h (DEFAULT_PCC_STRUCT_RETURN): Likewise.
31059 * config/i386/djgpp.h (TARGET_ASM_LTO_END): Likewise.
31060 * config/i386/dragonfly.h (NO_PROFILE_COUNTERS): Likewise.
31061 * config/i386/driver-i386.cc (detect_caches_intel): Likewise.
31062 * config/i386/freebsd.h (NO_PROFILE_COUNTERS): Likewise.
31063 * config/i386/i386-c.cc (ix86_target_macros): Likewise.
31064 * config/i386/i386-expand.cc (get_mode_wider_vector): Likewise.
31065 * config/i386/i386-options.cc (ix86_set_func_type): Likewise.
31066 * config/i386/i386-protos.h (ix86_extract_perm_from_pool_constant): Likewise.
31067 (ix86_register_pragmas): Likewise.
31068 (ix86_d_has_stdcall_convention): Likewise.
31069 (i386_pe_seh_init_sections): Likewise.
31070 * config/i386/i386.cc (ix86_function_arg_regno_p): Likewise.
31071 (ix86_function_value_regno_p): Likewise.
31072 (ix86_compute_frame_layout): Likewise.
31073 (legitimize_pe_coff_symbol): Likewise.
31074 (output_pic_addr_const): Likewise.
31075 * config/i386/i386.h (defined): Likewise.
31076 (host_detect_local_cpu): Likewise.
31077 (CONSTANT_ADDRESS_P): Likewise.
31078 (DEFAULT_LARGE_SECTION_THRESHOLD): Likewise.
31079 (struct machine_frame_state): Likewise.
31080 * config/i386/i386.md: Likewise.
31081 * config/i386/lynx.h (ASM_OUTPUT_ALIGN): Likewise.
31082 * config/i386/mmx.md: Likewise.
31083 * config/i386/sse.md: Likewise.
31084 * config/i386/t-cygming: Likewise.
31085 * config/i386/t-djgpp: Likewise.
31086 * config/i386/t-gnu-property: Likewise.
31087 * config/i386/t-i386: Likewise.
31088 * config/i386/t-intelmic: Likewise.
31089 * config/i386/t-omp-device: Likewise.
31090 * config/i386/winnt-cxx.cc (i386_pe_type_dllimport_p): Likewise.
31091 (i386_pe_adjust_class_at_definition): Likewise.
31092 * config/i386/winnt.cc (gen_stdcall_or_fastcall_suffix): Likewise.
31093 (i386_pe_mangle_decl_assembler_name): Likewise.
31094 (i386_pe_encode_section_info): Likewise.
31095 * config/i386/x-cygwin: Likewise.
31096 * config/i386/x-darwin: Likewise.
31097 * config/i386/x-i386: Likewise.
31098 * config/i386/x-mingw32: Likewise.
31099 * config/i386/x86-tune-sched-core.cc: Likewise.
31100 * config/i386/x86-tune.def: Likewise.
31101 * config/i386/xm-djgpp.h (STANDARD_STARTFILE_PREFIX_1): Likewise.
31102 * config/ia64/freebsd.h: Likewise.
31103 * config/ia64/hpux.h (REGISTER_TARGET_PRAGMAS): Likewise.
31104 * config/ia64/ia64-protos.h (ia64_except_unwind_info): Likewise.
31105 * config/ia64/ia64.cc (ia64_function_value_regno_p): Likewise.
31106 (ia64_secondary_reload_class): Likewise.
31107 (bundling): Likewise.
31108 * config/ia64/ia64.h: Likewise.
31109 * config/ia64/ia64.md: Likewise.
31110 * config/ia64/predicates.md: Likewise.
31111 * config/ia64/sysv4.h: Likewise.
31112 * config/ia64/t-ia64: Likewise.
31113 * config/iq2000/iq2000.h (FUNCTION_MODE): Likewise.
31114 * config/iq2000/iq2000.md: Likewise.
31115 * config/linux.h (TARGET_HAS_BIONIC): Likewise.
31117 * config/m32c/m32c.cc (m32c_function_needs_enter): Likewise.
31118 * config/m32c/m32c.h (MAX_REGS_PER_ADDRESS): Likewise.
31119 * config/m32c/t-m32c: Likewise.
31120 * config/m32r/m32r-protos.h: Likewise.
31121 * config/m32r/m32r.cc (m32r_print_operand): Likewise.
31122 * config/m32r/m32r.h: Likewise.
31123 * config/m32r/m32r.md: Likewise.
31124 * config/m68k/m68k-isas.def: Likewise.
31125 * config/m68k/m68k-microarchs.def: Likewise.
31126 * config/m68k/m68k-protos.h (strict_low_part_peephole_ok): Likewise.
31127 (m68k_epilogue_uses): Likewise.
31128 * config/m68k/m68k.cc (m68k_call_tls_get_addr): Likewise.
31129 (m68k_sched_adjust_cost): Likewise.
31130 (m68k_sched_md_init): Likewise.
31131 * config/m68k/m68k.h (__transfer_from_trampoline): Likewise.
31132 (enum m68k_function_kind): Likewise.
31133 * config/m68k/m68k.md: Likewise.
31134 * config/m68k/m68kemb.h: Likewise.
31135 * config/m68k/uclinux.h (ENDFILE_SPEC): Likewise.
31136 * config/mcore/mcore-protos.h: Likewise.
31137 * config/mcore/mcore.cc (mcore_expand_insv): Likewise.
31138 (mcore_expand_prolog): Likewise.
31139 * config/mcore/mcore.h (TARGET_MCORE): Likewise.
31140 * config/mcore/mcore.md: Likewise.
31141 * config/microblaze/microblaze-protos.h: Likewise.
31142 * config/microblaze/microblaze.cc (microblaze_legitimate_pic_operand): Likewise.
31143 (microblaze_function_prologue): Likewise.
31144 (microblaze_function_epilogue): Likewise.
31145 (microblaze_select_section): Likewise.
31146 (microblaze_asm_output_mi_thunk): Likewise.
31147 (microblaze_eh_return): Likewise.
31148 * config/microblaze/microblaze.h: Likewise.
31149 * config/microblaze/microblaze.md: Likewise.
31150 * config/microblaze/t-microblaze: Likewise.
31151 * config/mips/driver-native.cc: Likewise.
31152 * config/mips/loongson2ef.md: Likewise.
31153 * config/mips/mips-protos.h (mips_expand_vec_cmp_expr): Likewise.
31154 * config/mips/mips.cc (mips_rtx_costs): Likewise.
31155 (mips_output_filename): Likewise.
31156 (mips_output_function_prologue): Likewise.
31157 (mips_output_function_epilogue): Likewise.
31158 (mips_output_mi_thunk): Likewise.
31159 * config/mips/mips.h: Likewise.
31160 * config/mips/mips.md: Likewise.
31161 * config/mips/t-mips: Likewise.
31162 * config/mips/x-native: Likewise.
31163 * config/mmix/mmix-protos.h: Likewise.
31164 * config/mmix/mmix.cc (mmix_option_override): Likewise.
31165 (mmix_dbx_register_number): Likewise.
31166 (mmix_expand_prologue): Likewise.
31167 * config/mmix/mmix.h: Likewise.
31168 * config/mmix/mmix.md: Likewise.
31169 * config/mmix/predicates.md: Likewise.
31170 * config/mn10300/mn10300.cc (mn10300_symbolic_operand): Likewise.
31171 (mn10300_legitimate_pic_operand_p): Likewise.
31172 * config/mn10300/mn10300.h (enum reg_class): Likewise.
31173 (NO_FUNCTION_CSE): Likewise.
31174 * config/moxie/moxie-protos.h: Likewise.
31175 * config/moxie/uclinux.h (TARGET_LIBC_HAS_FUNCTION): Likewise.
31176 * config/msp430/msp430-devices.cc (extract_devices_dir_from_exec_prefix): Likewise.
31177 * config/msp430/msp430.cc (msp430_gimplify_va_arg_expr): Likewise.
31178 (msp430_incoming_return_addr_rtx): Likewise.
31179 * config/msp430/msp430.h (msp430_get_linker_devices_include_path): Likewise.
31180 * config/msp430/t-msp430: Likewise.
31181 * config/nds32/nds32-cost.cc (nds32_rtx_costs_speed_prefer): Likewise.
31182 (nds32_rtx_costs_size_prefer): Likewise.
31183 (nds32_init_rtx_costs): Likewise.
31184 * config/nds32/nds32-doubleword.md: Likewise.
31185 * config/nds32/nds32.cc (nds32_memory_move_cost): Likewise.
31186 (nds32_builtin_decl): Likewise.
31187 * config/nds32/nds32.h (enum nds32_16bit_address_type): Likewise.
31188 (enum nds32_isr_nested_type): Likewise.
31189 (enum reg_class): Likewise.
31190 * config/nds32/predicates.md: Likewise.
31191 * config/nds32/t-nds32: Likewise.
31192 * config/nios2/nios2.cc (nios2_pragma_target_parse): Likewise.
31193 * config/nvptx/nvptx-protos.h: Likewise.
31194 * config/nvptx/nvptx.cc (nvptx_goacc_expand_var_decl): Likewise.
31195 * config/nvptx/nvptx.h (TARGET_CPU_CPP_BUILTINS): Likewise.
31196 * config/nvptx/t-nvptx: Likewise.
31197 * config/nvptx/t-omp-device: Likewise.
31198 * config/pa/elf.h: Likewise.
31199 * config/pa/pa-linux.h (GLOBAL_ASM_OP): Likewise.
31200 * config/pa/pa-netbsd.h (GLOBAL_ASM_OP): Likewise.
31201 * config/pa/pa-openbsd.h (TARGET_ASM_GLOBALIZE_LABEL): Likewise.
31202 * config/pa/pa-protos.h (pa_eh_return_handler_rtx): Likewise.
31203 (pa_legitimize_reload_address): Likewise.
31204 (pa_can_use_return_insn): Likewise.
31205 * config/pa/pa.cc (mem_shadd_or_shadd_rtx_p): Likewise.
31206 (som_output_text_section_asm_op): Likewise.
31207 * config/pa/pa.h (PROFILE_BEFORE_PROLOGUE): Likewise.
31208 * config/pa/pa.md: Likewise.
31209 * config/pa/som.h: Likewise.
31210 * config/pa/t-pa: Likewise.
31211 * config/pdp11/pdp11.cc (decode_pdp11_d): Likewise.
31212 * config/pdp11/pdp11.h: Likewise.
31213 * config/pdp11/pdp11.md: Likewise.
31214 * config/pdp11/t-pdp11: Likewise.
31215 * config/pru/pru.md: Likewise.
31216 * config/pru/t-pru: Likewise.
31217 * config/riscv/riscv-protos.h (NUM_SYMBOL_TYPES): Likewise.
31218 (riscv_gpr_save_operation_p): Likewise.
31219 (riscv_d_register_target_info): Likewise.
31220 (riscv_init_builtins): Likewise.
31221 * config/riscv/riscv.cc (riscv_output_mi_thunk): Likewise.
31222 * config/riscv/riscv.h (CSW_MAX_OFFSET): Likewise.
31223 * config/riscv/t-riscv: Likewise.
31224 * config/rl78/rl78.cc (rl78_asm_ctor_dtor): Likewise.
31225 * config/rl78/t-rl78: Likewise.
31226 * config/rs6000/aix.h: Likewise.
31227 * config/rs6000/aix71.h (ASM_SPEC_COMMON): Likewise.
31228 * config/rs6000/aix72.h (ASM_SPEC_COMMON): Likewise.
31229 * config/rs6000/aix73.h (ASM_SPEC_COMMON): Likewise.
31230 * config/rs6000/darwin.h (TARGET_ASM_GLOBALIZE_LABEL): Likewise.
31231 * config/rs6000/driver-rs6000.cc: Likewise.
31232 * config/rs6000/freebsd.h: Likewise.
31233 * config/rs6000/freebsd64.h: Likewise.
31234 * config/rs6000/lynx.h (ASM_OUTPUT_ALIGN): Likewise.
31235 * config/rs6000/rbtree.cc: Likewise.
31236 * config/rs6000/rbtree.h: Likewise.
31237 * config/rs6000/rs6000-c.cc (rs6000_target_modify_macros): Likewise.
31238 * config/rs6000/rs6000-call.cc (rs6000_invalid_builtin): Likewise.
31239 (rs6000_expand_builtin): Likewise.
31240 (rs6000_init_builtins): Likewise.
31241 * config/rs6000/rs6000-cpus.def: Likewise.
31242 * config/rs6000/rs6000-gen-builtins.cc (write_init_ovld_table): Likewise.
31243 * config/rs6000/rs6000-internal.h (ALTIVEC_REG_BIT): Likewise.
31244 (quad_address_offset_p): Likewise.
31245 * config/rs6000/rs6000-logue.cc (interesting_frame_related_regno): Likewise.
31246 (rs6000_emit_epilogue): Likewise.
31247 * config/rs6000/rs6000-overload.def: Likewise.
31248 * config/rs6000/rs6000-p8swap.cc: Likewise.
31249 * config/rs6000/rs6000-protos.h (GCC_RS6000_PROTOS_H): Likewise.
31250 (rs6000_const_f32_to_i32): Likewise.
31251 * config/rs6000/rs6000.cc (legitimate_lo_sum_address_p): Likewise.
31252 (rs6000_debug_legitimize_address): Likewise.
31253 (rs6000_mode_dependent_address): Likewise.
31254 (rs6000_adjust_priority): Likewise.
31255 (rs6000_c_mode_for_suffix): Likewise.
31256 * config/rs6000/rs6000.h (defined): Likewise.
31257 (LONG_DOUBLE_TYPE_SIZE): Likewise.
31258 * config/rs6000/rs6000.md: Likewise.
31259 * config/rs6000/sysv4.h: Likewise.
31260 * config/rs6000/t-linux: Likewise.
31261 * config/rs6000/t-linux64: Likewise.
31262 * config/rs6000/t-rs6000: Likewise.
31263 * config/rs6000/x-darwin: Likewise.
31264 * config/rs6000/x-darwin64: Likewise.
31265 * config/rs6000/x-rs6000: Likewise.
31266 * config/rs6000/xcoff.h (ASM_OUTPUT_LABELREF): Likewise.
31267 * config/rx/rx.cc (rx_expand_builtin): Likewise.
31268 * config/s390/constraints.md: Likewise.
31269 * config/s390/driver-native.cc: Likewise.
31270 * config/s390/htmxlintrin.h: Likewise.
31271 * config/s390/s390-builtins.def (B_DEF): Likewise.
31272 (OB_DEF_VAR): Likewise.
31273 * config/s390/s390-builtins.h: Likewise.
31274 * config/s390/s390-c.cc: Likewise.
31275 * config/s390/s390-opts.h: Likewise.
31276 * config/s390/s390-protos.h (s390_check_symref_alignment): Likewise.
31277 (s390_register_target_pragmas): Likewise.
31278 * config/s390/s390.cc (s390_init_builtins): Likewise.
31279 (s390_expand_plus_operand): Likewise.
31280 (s390_expand_atomic): Likewise.
31281 (s390_valid_target_attribute_inner_p): Likewise.
31282 * config/s390/s390.h (LONG_DOUBLE_TYPE_SIZE): Likewise.
31283 * config/s390/s390.md: Likewise.
31284 * config/s390/t-s390: Likewise.
31285 * config/s390/vx-builtins.md: Likewise.
31286 * config/s390/x-native: Likewise.
31287 * config/sh/divtab-sh4-300.cc (main): Likewise.
31288 * config/sh/divtab-sh4.cc (main): Likewise.
31289 * config/sh/divtab.cc (main): Likewise.
31290 * config/sh/elf.h: Likewise.
31291 * config/sh/sh-protos.h (sh_fsca_int2sf): Likewise.
31292 * config/sh/sh.cc (SYMBOL_FLAG_FUNCVEC_FUNCTION): Likewise.
31293 (sh_struct_value_rtx): Likewise.
31294 (sh_remove_reg_dead_or_unused_notes): Likewise.
31295 * config/sh/sh.h (MIN_UNITS_PER_WORD): Likewise.
31296 * config/sh/t-sh: Likewise.
31297 * config/sol2-protos.h (solaris_override_options): Likewise.
31298 * config/sol2.h: Likewise.
31299 * config/sparc/driver-sparc.cc: Likewise.
31300 * config/sparc/freebsd.h: Likewise.
31301 * config/sparc/sparc-protos.h (make_pass_work_around_errata): Likewise.
31302 * config/sparc/sparc.cc (sparc_output_mi_thunk): Likewise.
31303 (sparc_asan_shadow_offset): Likewise.
31304 * config/sparc/sparc.h: Likewise.
31305 * config/sparc/sparc.md: Likewise.
31306 * config/sparc/t-sparc: Likewise.
31307 * config/sparc/x-sparc: Likewise.
31308 * config/stormy16/stormy16.cc (xstormy16_mode_dependent_address_p): Likewise.
31309 * config/t-darwin: Likewise.
31310 * config/t-dragonfly: Likewise.
31311 * config/t-freebsd: Likewise.
31312 * config/t-glibc: Likewise.
31313 * config/t-linux: Likewise.
31314 * config/t-netbsd: Likewise.
31315 * config/t-openbsd: Likewise.
31316 * config/t-pnt16-warn: Likewise.
31317 * config/t-sol2: Likewise.
31318 * config/t-vxworks: Likewise.
31319 * config/t-winnt: Likewise.
31320 * config/tilegx/t-tilegx: Likewise.
31321 * config/tilegx/tilegx-c.cc: Likewise.
31322 * config/tilegx/tilegx-protos.h (tilegx_function_profiler): Likewise.
31323 * config/tilegx/tilegx.md: Likewise.
31324 * config/tilepro/t-tilepro: Likewise.
31325 * config/tilepro/tilepro-c.cc: Likewise.
31326 * config/v850/t-v850: Likewise.
31327 * config/v850/v850-protos.h: Likewise.
31328 * config/v850/v850.cc (F): Likewise.
31329 * config/v850/v850.h (enum reg_class): Likewise.
31330 (SLOW_BYTE_ACCESS): Likewise.
31331 * config/vax/vax.cc (vax_mode_dependent_address_p): Likewise.
31332 * config/vax/vax.h (enum reg_class): Likewise.
31333 * config/vax/vax.md: Likewise.
31334 * config/visium/visium.cc (visium_legitimate_address_p): Likewise.
31335 * config/visium/visium.h: Likewise.
31336 * config/vms/t-vms: Likewise.
31337 * config/vms/vms-crtlmap.map: Likewise.
31338 * config/vms/vms-protos.h (vms_c_get_vms_ver): Likewise.
31339 * config/vx-common.h: Likewise.
31340 * config/x-darwin: Likewise.
31341 * config/x-hpux: Likewise.
31342 * config/x-linux: Likewise.
31343 * config/x-netbsd: Likewise.
31344 * config/x-openbsd: Likewise.
31345 * config/x-solaris: Likewise.
31346 * config/xtensa/xtensa-protos.h (xtensa_mem_offset): Likewise.
31347 * config/xtensa/xtensa.cc (xtensa_option_override): Likewise.
31348 * config/xtensa/xtensa.h: Likewise.
31349 * configure.ac: Likewise.
31350 * context.cc: Likewise.
31351 * convert.h: Likewise.
31352 * coretypes.h: Likewise.
31353 * coverage.cc: Likewise.
31354 * coverage.h: Likewise.
31355 * cppdefault.h (struct default_include): Likewise.
31356 * cprop.cc (local_cprop_pass): Likewise.
31357 (one_cprop_pass): Likewise.
31358 * cse.cc (hash_rtx_cb): Likewise.
31359 (fold_rtx): Likewise.
31360 * ctfc.h (ctfc_get_num_vlen_bytes): Likewise.
31361 * data-streamer.h (bp_unpack_var_len_int): Likewise.
31362 (streamer_write_widest_int): Likewise.
31363 * dbgcnt.def: Likewise.
31364 * dbxout.cc (dbxout_early_global_decl): Likewise.
31365 (dbxout_common_check): Likewise.
31366 * dbxout.h: Likewise.
31367 * debug.h (struct gcc_debug_hooks): Likewise.
31368 (dump_go_spec_init): Likewise.
31369 * df-core.cc: Likewise.
31370 * df-scan.cc (df_insn_info_delete): Likewise.
31371 (df_insn_delete): Likewise.
31372 * df.h (debug_df_chain): Likewise.
31373 (can_move_insns_across): Likewise.
31374 * dfp.cc (decimal_from_binary): Likewise.
31375 * diagnostic-color.cc: Likewise.
31376 * diagnostic-event-id.h: Likewise.
31377 * diagnostic-show-locus.cc (test_one_liner_labels): Likewise.
31378 * diagnostic.cc (bt_callback): Likewise.
31379 (num_digits): Likewise.
31380 * doc/avr-mmcu.texi: Likewise.
31381 * doc/cfg.texi: Likewise.
31382 * doc/contrib.texi: Likewise.
31383 * doc/cppinternals.texi: Likewise.
31384 * doc/extend.texi: Likewise.
31385 * doc/generic.texi: Likewise.
31386 * doc/gimple.texi: Likewise.
31387 * doc/gty.texi: Likewise.
31388 * doc/invoke.texi: Likewise.
31389 * doc/loop.texi: Likewise.
31390 * doc/lto.texi: Likewise.
31391 * doc/match-and-simplify.texi: Likewise.
31392 * doc/md.texi: Likewise.
31393 * doc/optinfo.texi: Likewise.
31394 * doc/options.texi: Likewise.
31395 * doc/passes.texi: Likewise.
31396 * doc/plugins.texi: Likewise.
31397 * doc/rtl.texi: Likewise.
31398 * doc/sourcebuild.texi: Likewise.
31399 * doc/tm.texi: Likewise.
31400 * doc/tm.texi.in: Likewise.
31401 * doc/tree-ssa.texi: Likewise.
31402 * dojump.cc (do_jump): Likewise.
31403 * dojump.h: Likewise.
31404 * dumpfile.cc (test_impl_location): Likewise.
31405 (test_capture_of_dump_calls): Likewise.
31406 * dumpfile.h (enum dump_kind): Likewise.
31407 (class dump_location_t): Likewise.
31408 (dump_enabled_p): Likewise.
31409 (enable_rtl_dump_file): Likewise.
31410 (dump_combine_total_stats): Likewise.
31411 * dwarf2asm.cc (dw2_asm_output_delta_uleb128): Likewise.
31412 * dwarf2ctf.h (ctf_debug_finish): Likewise.
31413 * dwarf2out.cc (dwarf2out_begin_prologue): Likewise.
31414 (struct loc_descr_context): Likewise.
31415 (rtl_for_decl_location): Likewise.
31416 (gen_subprogram_die): Likewise.
31417 (gen_label_die): Likewise.
31418 (is_trivial_indirect_ref): Likewise.
31419 (dwarf2out_late_global_decl): Likewise.
31420 (dwarf_file_hasher::hash): Likewise.
31421 (dwarf2out_end_source_file): Likewise.
31422 (dwarf2out_define): Likewise.
31423 (dwarf2out_early_finish): Likewise.
31424 * dwarf2out.h (struct dw_fde_node): Likewise.
31425 (struct dw_discr_list_node): Likewise.
31426 (output_loc_sequence_raw): Likewise.
31427 * emit-rtl.cc (gen_raw_REG): Likewise.
31428 (maybe_set_max_label_num): Likewise.
31429 * emit-rtl.h (struct rtl_data): Likewise.
31430 * errors.cc (internal_error): Likewise.
31431 (trim_filename): Likewise.
31432 * et-forest.cc: Likewise.
31433 * except.cc (init_eh_for_function): Likewise.
31434 * explow.cc (promote_ssa_mode): Likewise.
31435 (get_dynamic_stack_size): Likewise.
31436 * explow.h: Likewise.
31437 * expmed.h: Likewise.
31438 * expr.cc (safe_from_p): Likewise.
31439 (expand_expr_real_2): Likewise.
31440 (expand_expr_real_1): Likewise.
31441 * file-prefix-map.cc (remap_filename): Likewise.
31442 * final.cc (app_enable): Likewise.
31443 (make_pass_compute_alignments): Likewise.
31444 (final_scan_insn_1): Likewise.
31445 (final_scan_insn): Likewise.
31446 * fixed-value.h (fixed_from_string): Likewise.
31447 * flag-types.h (NO_DEBUG): Likewise.
31448 (DWARF2_DEBUG): Likewise.
31449 (VMS_DEBUG): Likewise.
31450 (BTF_DEBUG): Likewise.
31451 (enum ctf_debug_info_levels): Likewise.
31452 * fold-const.cc (const_binop): Likewise.
31453 (fold_binary_loc): Likewise.
31454 (fold_checksum_tree): Likewise.
31455 * fp-test.cc: Likewise.
31456 * function.cc (expand_function_end): Likewise.
31457 * function.h (struct function): Likewise.
31458 * fwprop.cc (should_replace_address): Likewise.
31459 * gcc-main.cc: Likewise.
31460 * gcc-rich-location.h (class gcc_rich_location): Likewise.
31461 * gcc-symtab.h: Likewise.
31462 * gcc.cc (MIN_FATAL_STATUS): Likewise.
31463 (driver_handle_option): Likewise.
31464 (quote_spec_arg): Likewise.
31465 (driver::finalize): Likewise.
31466 * gcc.h (set_input): Likewise.
31467 * gcov-dump.cc: Likewise.
31468 * gcov.cc (solve_flow_graph): Likewise.
31469 * gcse-common.cc: Likewise.
31470 * gcse.cc (make_pass_rtl_hoist): Likewise.
31471 * genattr-common.cc: Likewise.
31472 * genattrtab.cc (min_fn): Likewise.
31473 (write_const_num_delay_slots): Likewise.
31474 * genautomata.cc: Likewise.
31475 * genconditions.cc (write_one_condition): Likewise.
31476 * genconstants.cc: Likewise.
31477 * genemit.cc (gen_exp): Likewise.
31478 * generic-match-head.cc: Likewise.
31479 * genextract.cc: Likewise.
31480 * gengenrtl.cc (always_void_p): Likewise.
31481 * gengtype-parse.cc (gtymarker_opt): Likewise.
31482 * gengtype-state.cc (state_writer::state_writer): Likewise.
31483 (write_state_trailer): Likewise.
31484 (equals_type_number): Likewise.
31485 (read_state): Likewise.
31486 * gengtype.cc (open_base_files): Likewise.
31487 (struct file_rule_st): Likewise.
31488 (header_dot_h_frul): Likewise.
31489 * gengtype.h: Likewise.
31490 * genmatch.cc (main): Likewise.
31491 * genmddeps.cc: Likewise.
31492 * genmodes.cc (emit_mode_inner): Likewise.
31493 (emit_mode_unit_size): Likewise.
31494 * genpeep.cc (gen_peephole): Likewise.
31495 * genpreds.cc (write_tm_preds_h): Likewise.
31496 * genrecog.cc (validate_pattern): Likewise.
31497 (write_header): Likewise.
31499 * gensupport.cc (change_subst_attribute): Likewise.
31500 (traverse_c_tests): Likewise.
31501 (add_predicate): Likewise.
31502 (init_predicate_table): Likewise.
31503 * gensupport.h (struct optab_pattern): Likewise.
31504 (get_num_insn_codes): Likewise.
31505 (maybe_eval_c_test): Likewise.
31506 (struct pred_data): Likewise.
31507 * ggc-internal.h: Likewise.
31508 * gimple-fold.cc (maybe_fold_reference): Likewise.
31509 (get_range_strlen_tree): Likewise.
31510 * gimple-fold.h (gimple_stmt_integer_valued_real_p): Likewise.
31511 * gimple-low.cc: Likewise.
31512 * gimple-match-head.cc (directly_supported_p): Likewise.
31513 * gimple-pretty-print.h: Likewise.
31514 * gimple-ssa-sprintf.cc (format_percent): Likewise.
31515 (adjust_range_for_overflow): Likewise.
31516 * gimple-streamer.h: Likewise.
31517 * gimple.h (struct GTY): Likewise.
31518 (is_gimple_resx): Likewise.
31519 * gimplify.cc (gimplify_expr): Likewise.
31520 (gimplify_init_constructor): Likewise.
31521 (omp_construct_selector_matches): Likewise.
31522 (gimplify_omp_target_update): Likewise.
31523 (gimplify_omp_ordered): Likewise.
31524 (gimplify_va_arg_expr): Likewise.
31525 * graphite-isl-ast-to-gimple.cc (should_copy_to_new_region): Likewise.
31526 * haifa-sched.cc (increase_insn_priority): Likewise.
31527 (try_ready): Likewise.
31528 (sched_create_recovery_edges): Likewise.
31529 * ifcvt.cc (find_if_case_1): Likewise.
31530 (find_if_case_2): Likewise.
31531 * inchash.h: Likewise.
31532 * incpath.cc (add_env_var_paths): Likewise.
31533 * input.cc (dump_location_info): Likewise.
31534 (assert_loceq): Likewise.
31535 (test_lexer_string_locations_concatenation_1): Likewise.
31536 (test_lexer_string_locations_concatenation_2): Likewise.
31537 (test_lexer_string_locations_concatenation_3): Likewise.
31538 * input.h (BUILTINS_LOCATION): Likewise.
31539 (class string_concat_db): Likewise.
31540 * internal-fn.cc (expand_MUL_OVERFLOW): Likewise.
31541 (expand_LOOP_VECTORIZED): Likewise.
31542 * ipa-cp.cc (make_pass_ipa_cp): Likewise.
31543 * ipa-fnsummary.cc (remap_freqcounting_preds_after_dup): Likewise.
31544 (ipa_fn_summary_t::duplicate): Likewise.
31545 (make_pass_ipa_fn_summary): Likewise.
31546 * ipa-fnsummary.h (enum ipa_hints_vals): Likewise.
31547 * ipa-free-lang-data.cc (fld_simplified_type): Likewise.
31548 (free_lang_data_in_decl): Likewise.
31549 * ipa-inline.cc (compute_inlined_call_time): Likewise.
31550 (inline_always_inline_functions): Likewise.
31551 * ipa-inline.h (free_growth_caches): Likewise.
31552 (inline_account_function_p): Likewise.
31553 * ipa-modref.cc (modref_access_analysis::analyze_stmt): Likewise.
31554 (modref_eaf_analysis::analyze_ssa_name): Likewise.
31555 * ipa-param-manipulation.cc (ipa_param_body_adjustments::mark_dead_statements): Likewise.
31556 (ipa_param_body_adjustments::remap_with_debug_expressions): Likewise.
31557 * ipa-prop.cc (ipa_set_node_agg_value_chain): Likewise.
31558 * ipa-prop.h (IPA_UNDESCRIBED_USE): Likewise.
31559 (unadjusted_ptr_and_unit_offset): Likewise.
31560 * ipa-reference.cc (make_pass_ipa_reference): Likewise.
31561 * ipa-reference.h (GCC_IPA_REFERENCE_H): Likewise.
31562 * ipa-split.cc (consider_split): Likewise.
31563 * ipa-sra.cc (isra_read_node_info): Likewise.
31564 * ipa-utils.h (struct ipa_dfs_info): Likewise.
31565 (recursive_call_p): Likewise.
31566 (ipa_make_function_pure): Likewise.
31567 * ira-build.cc (ira_create_allocno): Likewise.
31568 (ira_flattening): Likewise.
31569 * ira-color.cc (do_coloring): Likewise.
31570 (update_curr_costs): Likewise.
31571 * ira-conflicts.cc (process_regs_for_copy): Likewise.
31572 * ira-int.h (struct ira_emit_data): Likewise.
31573 (ira_prohibited_mode_move_regs): Likewise.
31574 (ira_get_dup_out_num): Likewise.
31575 (ira_destroy): Likewise.
31576 (ira_tune_allocno_costs): Likewise.
31577 (ira_implicitly_set_insn_hard_regs): Likewise.
31578 (ira_build_conflicts): Likewise.
31579 (ira_color): Likewise.
31580 * ira-lives.cc (process_bb_node_lives): Likewise.
31581 * ira.cc (class ira_spilled_reg_stack_slot): Likewise.
31582 (setup_uniform_class_p): Likewise.
31583 (def_dominates_uses): Likewise.
31584 * ira.h (ira_nullify_asm_goto): Likewise.
31585 * langhooks.cc (lhd_post_options): Likewise.
31586 * langhooks.h (class substring_loc): Likewise.
31587 (struct lang_hooks_for_tree_inlining): Likewise.
31588 (struct lang_hooks_for_types): Likewise.
31589 (struct lang_hooks): Likewise.
31590 * libfuncs.h (synchronize_libfunc): Likewise.
31591 * loop-doloop.cc (doloop_condition_get): Likewise.
31592 * loop-init.cc (fix_loop_structure): Likewise.
31593 * loop-invariant.cc: Likewise.
31594 * lower-subreg.h: Likewise.
31595 * lra-constraints.cc (curr_insn_transform): Likewise.
31596 * lra-int.h (struct lra_insn_reg): Likewise.
31597 (lra_undo_inheritance): Likewise.
31598 (lra_setup_reload_pseudo_preferenced_hard_reg): Likewise.
31599 (lra_split_hard_reg_for): Likewise.
31600 (lra_coalesce): Likewise.
31601 (lra_final_code_change): Likewise.
31602 * lra-spills.cc (lra_final_code_change): Likewise.
31603 * lra.cc (lra_process_new_insns): Likewise.
31604 * lto-compress.h (struct lto_compression_stream): Likewise.
31605 * lto-streamer-out.cc (DFS::DFS_write_tree_body): Likewise.
31606 (write_symbol): Likewise.
31607 * lto-streamer.h (enum LTO_tags): Likewise.
31608 (lto_value_range_error): Likewise.
31609 (lto_append_block): Likewise.
31610 (lto_streamer_hooks_init): Likewise.
31611 (stream_read_tree_ref): Likewise.
31612 (lto_prepare_function_for_streaming): Likewise.
31613 (select_what_to_stream): Likewise.
31614 (omp_lto_input_declare_variant_alt): Likewise.
31615 (cl_optimization_stream_in): Likewise.
31616 * lto-wrapper.cc (append_compiler_options): Likewise.
31617 * machmode.def: Likewise.
31618 * machmode.h (struct int_n_data_t): Likewise.
31619 * main.cc (main): Likewise.
31620 * match.pd: Likewise.
31621 * omp-builtins.def (BUILT_IN_GOMP_CRITICAL_NAME_END): Likewise.
31622 (BUILT_IN_GOMP_LOOP_ULL_ORDERED_RUNTIME_NEXT): Likewise.
31623 * omp-expand.cc (expand_omp_atomic_fetch_op): Likewise.
31624 (make_pass_expand_omp_ssa): Likewise.
31625 * omp-low.cc (struct omp_context): Likewise.
31626 (struct omp_taskcopy_context): Likewise.
31627 (lower_omp): Likewise.
31628 * omp-oacc-neuter-broadcast.cc (omp_sese_active_worker_call): Likewise.
31629 (mask_name): Likewise.
31630 (omp_sese_dump_pars): Likewise.
31631 (worker_single_simple): Likewise.
31632 * omp-offload.cc (omp_finish_file): Likewise.
31633 (execute_oacc_loop_designation): Likewise.
31634 * optabs-query.cc (lshift_cheap_p): Likewise.
31635 * optc-gen.awk: Likewise.
31636 * optc-save-gen.awk: Likewise.
31637 * optinfo-emit-json.cc (optrecord_json_writer::optrecord_json_writer): Likewise.
31638 * opts-common.cc: Likewise.
31639 * output.h (app_enable): Likewise.
31640 (output_operand_lossage): Likewise.
31641 (insn_current_reference_address): Likewise.
31642 (get_insn_template): Likewise.
31643 (output_quoted_string): Likewise.
31644 * pass_manager.h (struct register_pass_info): Likewise.
31645 * plugin.cc: Likewise.
31646 * plugin.def (PLUGIN_ANALYZER_INIT): Likewise.
31647 * plugin.h (invoke_plugin_callbacks): Likewise.
31648 * pointer-query.cc (handle_mem_ref): Likewise.
31649 * postreload-gcse.cc (alloc_mem): Likewise.
31650 * predict.h (enum prediction): Likewise.
31651 (add_reg_br_prob_note): Likewise.
31652 * prefix.h: Likewise.
31653 * profile.h (get_working_sets): Likewise.
31654 * read-md.cc: Likewise.
31655 * read-md.h (struct mapping): Likewise.
31656 (class md_reader): Likewise.
31657 (class noop_reader): Likewise.
31658 * read-rtl-function.cc (function_reader::create_function): Likewise.
31659 (function_reader::extra_parsing_for_operand_code_0): Likewise.
31660 * read-rtl.cc (initialize_iterators): Likewise.
31661 * real.cc: Likewise.
31662 * real.h (struct real_value): Likewise.
31663 (format_helper::format_helper): Likewise.
31664 (real_hash): Likewise.
31665 (real_can_shorten_arithmetic): Likewise.
31666 * recog.cc (struct target_recog): Likewise.
31667 (offsettable_nonstrict_memref_p): Likewise.
31668 (constrain_operands): Likewise.
31669 * recog.h (MAX_RECOG_ALTERNATIVES): Likewise.
31670 (which_op_alt): Likewise.
31671 (struct insn_gen_fn): Likewise.
31672 * reg-notes.def (REG_NOTE): Likewise.
31673 * reg-stack.cc: Likewise.
31674 * regs.h (reg_is_parm_p): Likewise.
31675 * regset.h: Likewise.
31676 * reload.cc (push_reload): Likewise.
31677 (find_reloads): Likewise.
31678 (find_reloads_address_1): Likewise.
31679 (find_replacement): Likewise.
31680 (refers_to_regno_for_reload_p): Likewise.
31681 (refers_to_mem_for_reload_p): Likewise.
31682 * reload.h (push_reload): Likewise.
31683 (deallocate_reload_reg): Likewise.
31684 * reload1.cc (emit_input_reload_insns): Likewise.
31685 * reorg.cc (relax_delay_slots): Likewise.
31686 * rtl.def (UNKNOWN): Likewise.
31687 (SEQUENCE): Likewise.
31688 (BARRIER): Likewise.
31689 (ASM_OPERANDS): Likewise.
31690 (EQ_ATTR_ALT): Likewise.
31691 * rtl.h (struct GTY): Likewise.
31692 (LABEL_NAME): Likewise.
31693 (LABEL_ALT_ENTRY_P): Likewise.
31694 (SUBREG_BYTE): Likewise.
31695 (get_stack_check_protect): Likewise.
31696 (dump_rtx_statistics): Likewise.
31697 (unwrap_const_vec_duplicate): Likewise.
31698 (subreg_promoted_mode): Likewise.
31699 (gen_lowpart_common): Likewise.
31700 (operand_subword): Likewise.
31701 (immed_wide_int_const): Likewise.
31702 (decide_function_section): Likewise.
31703 (active_insn_p): Likewise.
31704 (delete_related_insns): Likewise.
31705 (try_split): Likewise.
31706 (val_signbit_known_clear_p): Likewise.
31707 (simplifiable_subregs): Likewise.
31708 (set_insn_deleted): Likewise.
31709 (subreg_get_info): Likewise.
31710 (remove_free_EXPR_LIST_node): Likewise.
31711 (finish_subregs_of_mode): Likewise.
31712 (get_mem_attrs): Likewise.
31713 (lookup_constant_def): Likewise.
31714 (rtx_to_tree_code): Likewise.
31715 (hash_rtx): Likewise.
31716 (condjump_in_parallel_p): Likewise.
31717 (validate_subreg): Likewise.
31718 (make_compound_operation): Likewise.
31719 (schedule_ebbs): Likewise.
31720 (print_inline_rtx): Likewise.
31721 (fixup_args_size_notes): Likewise.
31722 (expand_dec): Likewise.
31723 (prepare_copy_insn): Likewise.
31724 (mark_elimination): Likewise.
31725 (valid_mode_changes_for_regno): Likewise.
31726 (make_debug_expr_from_rtl): Likewise.
31727 (delete_vta_debug_insns): Likewise.
31728 (simplify_using_condition): Likewise.
31729 (set_insn_locations): Likewise.
31730 (fatal_insn_not_found): Likewise.
31731 (word_register_operation_p): Likewise.
31732 * rtlanal.cc (get_call_fndecl): Likewise.
31733 (side_effects_p): Likewise.
31734 (subreg_nregs): Likewise.
31735 (rtx_cost): Likewise.
31736 (canonicalize_condition): Likewise.
31737 * rtlanal.h (rtx_properties::try_to_add_note): Likewise.
31738 * run-rtl-passes.cc (run_rtl_passes): Likewise.
31739 * sanitizer.def (BUILT_IN_ASAN_VERSION_MISMATCH_CHECK): Likewise.
31740 * sched-deps.cc (add_dependence_1): Likewise.
31741 * sched-ebb.cc (begin_move_insn): Likewise.
31742 (add_deps_for_risky_insns): Likewise.
31743 (advance_target_bb): Likewise.
31744 * sched-int.h (reemit_notes): Likewise.
31745 (struct _haifa_insn_data): Likewise.
31747 (DEP_CANCELLED): Likewise.
31748 (debug_ds): Likewise.
31749 (number_in_ready): Likewise.
31750 (schedule_ebbs_finish): Likewise.
31751 (find_modifiable_mems): Likewise.
31752 * sched-rgn.cc (debug_rgn_dependencies): Likewise.
31753 * sel-sched-dump.cc (dump_lv_set): Likewise.
31754 * sel-sched-dump.h: Likewise.
31755 * sel-sched-ir.cc (sel_insn_rtx_cost): Likewise.
31756 (setup_id_reg_sets): Likewise.
31757 (has_dependence_p): Likewise.
31758 (sel_num_cfg_preds_gt_1): Likewise.
31759 (bb_ends_ebb_p): Likewise.
31760 * sel-sched-ir.h (struct _list_node): Likewise.
31761 (struct idata_def): Likewise.
31762 (bb_next_bb): Likewise.
31763 * sel-sched.cc (vinsn_writes_one_of_regs_p): Likewise.
31764 (choose_best_pseudo_reg): Likewise.
31765 (verify_target_availability): Likewise.
31766 (can_speculate_dep_p): Likewise.
31767 (sel_rank_for_schedule): Likewise.
31768 * selftest-run-tests.cc (selftest::run_tests): Likewise.
31769 * selftest.h (class auto_fix_quotes): Likewise.
31770 * shrink-wrap.cc (handle_simple_exit): Likewise.
31771 * shrink-wrap.h: Likewise.
31772 * simplify-rtx.cc (simplify_context::simplify_associative_operation): Likewise.
31773 (simplify_context::simplify_gen_vec_select): Likewise.
31774 * spellcheck-tree.h: Likewise.
31775 * spellcheck.h: Likewise.
31776 * statistics.h (struct function): Likewise.
31777 * stmt.cc (conditional_probability): Likewise.
31778 * stmt.h: Likewise.
31779 * stor-layout.h: Likewise.
31780 * streamer-hooks.h: Likewise.
31781 * stringpool.h: Likewise.
31782 * symtab.cc (symbol_table::change_decl_assembler_name): Likewise.
31783 * target.def (HOOK_VECTOR_END): Likewise.
31785 * target.h (union cumulative_args_t): Likewise.
31786 (by_pieces_ninsns): Likewise.
31787 (class predefined_function_abi): Likewise.
31788 * targhooks.cc (default_translate_mode_attribute): Likewise.
31789 * timevar.def: Likewise.
31790 * timevar.h (class timer): Likewise.
31791 * toplev.h (enable_rtl_dump_file): Likewise.
31792 * trans-mem.cc (collect_bb2reg): Likewise.
31793 * tree-call-cdce.cc (gen_conditions_for_pow): Likewise.
31794 * tree-cfg.cc (remove_bb): Likewise.
31795 (verify_gimple_debug): Likewise.
31796 (remove_edge_and_dominated_blocks): Likewise.
31797 (push_fndecl): Likewise.
31798 * tree-cfgcleanup.h (GCC_TREE_CFGCLEANUP_H): Likewise.
31799 * tree-complex.cc (expand_complex_multiplication): Likewise.
31800 (expand_complex_div_straight): Likewise.
31801 * tree-core.h (enum tree_index): Likewise.
31802 (enum operand_equal_flag): Likewise.
31803 * tree-eh.cc (honor_protect_cleanup_actions): Likewise.
31804 * tree-if-conv.cc (if_convertible_gimple_assign_stmt_p): Likewise.
31805 * tree-inline.cc (initialize_inlined_parameters): Likewise.
31806 * tree-inline.h (force_value_to_type): Likewise.
31807 * tree-nested.cc (get_chain_decl): Likewise.
31808 (walk_all_functions): Likewise.
31809 * tree-object-size.h: Likewise.
31810 * tree-outof-ssa.cc: Likewise.
31811 * tree-parloops.cc (create_parallel_loop): Likewise.
31812 * tree-pretty-print.cc (print_generic_expr_to_str): Likewise.
31813 (dump_generic_node): Likewise.
31814 * tree-profile.cc (tree_profiling): Likewise.
31815 * tree-sra.cc (maybe_add_sra_candidate): Likewise.
31816 * tree-ssa-address.cc: Likewise.
31817 * tree-ssa-alias.cc: Likewise.
31818 * tree-ssa-alias.h (ao_ref::max_size_known_p): Likewise.
31819 (dump_alias_stats): Likewise.
31820 * tree-ssa-ccp.cc: Likewise.
31821 * tree-ssa-coalesce.h: Likewise.
31822 * tree-ssa-live.cc (remove_unused_scope_block_p): Likewise.
31823 * tree-ssa-loop-manip.cc (copy_phi_node_args): Likewise.
31824 * tree-ssa-loop-unswitch.cc: Likewise.
31825 * tree-ssa-math-opts.cc: Likewise.
31826 * tree-ssa-operands.cc (class operands_scanner): Likewise.
31827 * tree-ssa-pre.cc: Likewise.
31828 * tree-ssa-reassoc.cc (optimize_ops_list): Likewise.
31829 (debug_range_entry): Likewise.
31830 * tree-ssa-sccvn.cc (eliminate_dom_walker::eliminate_stmt): Likewise.
31831 * tree-ssa-sccvn.h (TREE_SSA_SCCVN_H): Likewise.
31832 * tree-ssa-scopedtables.cc (add_expr_commutative): Likewise.
31833 (equal_mem_array_ref_p): Likewise.
31834 * tree-ssa-strlen.cc (is_strlen_related_p): Likewise.
31835 * tree-ssa-strlen.h (get_range_strlen_dynamic): Likewise.
31836 * tree-ssa-tail-merge.cc (stmt_local_def): Likewise.
31837 * tree-ssa-ter.h: Likewise.
31838 * tree-ssa-threadupdate.h (enum bb_dom_status): Likewise.
31839 * tree-streamer-in.cc (lto_input_ts_block_tree_pointers): Likewise.
31840 * tree-streamer-out.cc (pack_ts_block_value_fields): Likewise.
31841 (write_ts_block_tree_pointers): Likewise.
31842 * tree-streamer.h (struct streamer_tree_cache_d): Likewise.
31843 (streamer_read_tree_bitfields): Likewise.
31844 (streamer_write_integer_cst): Likewise.
31845 * tree-vect-patterns.cc (apply_binop_and_append_stmt): Likewise.
31846 (vect_synth_mult_by_constant): Likewise.
31847 * tree-vect-stmts.cc (vectorizable_operation): Likewise.
31848 * tree-vectorizer.cc: Likewise.
31849 * tree-vectorizer.h (class auto_purge_vect_location): Likewise.
31850 (vect_update_inits_of_drs): Likewise.
31851 (vect_get_mask_type_for_stmt): Likewise.
31852 (vect_rgroup_iv_might_wrap_p): Likewise.
31853 (cse_and_gimplify_to_preheader): Likewise.
31854 (vect_free_slp_tree): Likewise.
31855 (vect_pattern_recog): Likewise.
31856 (vect_stmt_dominates_stmt_p): Likewise.
31857 * tree.cc (initialize_tree_contains_struct): Likewise.
31858 (need_assembler_name_p): Likewise.
31859 (type_with_interoperable_signedness): Likewise.
31860 * tree.def (SWITCH_EXPR): Likewise.
31861 * tree.h (TYPE_SYMTAB_ADDRESS): Likewise.
31862 (poly_int_tree_p): Likewise.
31863 (inlined_function_outer_scope_p): Likewise.
31864 (tree_code_for_canonical_type_merging): Likewise.
31865 * value-prof.cc: Likewise.
31866 * value-prof.h (get_nth_most_common_value): Likewise.
31867 (find_func_by_profile_id): Likewise.
31868 * value-range.cc (vrp_operand_equal_p): Likewise.
31869 * value-range.h: Likewise.
31870 * var-tracking.cc: Likewise.
31871 * varasm.cc (default_function_section): Likewise.
31872 (function_section_1): Likewise.
31873 (assemble_variable): Likewise.
31874 (handle_vtv_comdat_section): Likewise.
31875 * vec.h (struct vec_prefix): Likewise.
31876 * vmsdbgout.cc (full_name): Likewise.
31877 * vtable-verify.cc: Likewise.
31878 * vtable-verify.h (struct vtv_graph_node): Likewise.
31879 * xcoffout.cc: Likewise.
31880 * xcoffout.h (DEBUG_SYMS_TEXT): Likewise.
31882 2022-01-17 Martin Liska <mliska@suse.cz>
31884 * adjust-alignment.c: Moved to...
31885 * adjust-alignment.cc: ...here.
31886 * alias.c: Moved to...
31887 * alias.cc: ...here.
31888 * alloc-pool.c: Moved to...
31889 * alloc-pool.cc: ...here.
31890 * asan.c: Moved to...
31891 * asan.cc: ...here.
31892 * attribs.c: Moved to...
31893 * attribs.cc: ...here.
31894 * auto-inc-dec.c: Moved to...
31895 * auto-inc-dec.cc: ...here.
31896 * auto-profile.c: Moved to...
31897 * auto-profile.cc: ...here.
31898 * bb-reorder.c: Moved to...
31899 * bb-reorder.cc: ...here.
31900 * bitmap.c: Moved to...
31901 * bitmap.cc: ...here.
31902 * btfout.c: Moved to...
31903 * btfout.cc: ...here.
31904 * builtins.c: Moved to...
31905 * builtins.cc: ...here.
31906 * caller-save.c: Moved to...
31907 * caller-save.cc: ...here.
31908 * calls.c: Moved to...
31909 * calls.cc: ...here.
31910 * ccmp.c: Moved to...
31911 * ccmp.cc: ...here.
31912 * cfg.c: Moved to...
31914 * cfganal.c: Moved to...
31915 * cfganal.cc: ...here.
31916 * cfgbuild.c: Moved to...
31917 * cfgbuild.cc: ...here.
31918 * cfgcleanup.c: Moved to...
31919 * cfgcleanup.cc: ...here.
31920 * cfgexpand.c: Moved to...
31921 * cfgexpand.cc: ...here.
31922 * cfghooks.c: Moved to...
31923 * cfghooks.cc: ...here.
31924 * cfgloop.c: Moved to...
31925 * cfgloop.cc: ...here.
31926 * cfgloopanal.c: Moved to...
31927 * cfgloopanal.cc: ...here.
31928 * cfgloopmanip.c: Moved to...
31929 * cfgloopmanip.cc: ...here.
31930 * cfgrtl.c: Moved to...
31931 * cfgrtl.cc: ...here.
31932 * cgraph.c: Moved to...
31933 * cgraph.cc: ...here.
31934 * cgraphbuild.c: Moved to...
31935 * cgraphbuild.cc: ...here.
31936 * cgraphclones.c: Moved to...
31937 * cgraphclones.cc: ...here.
31938 * cgraphunit.c: Moved to...
31939 * cgraphunit.cc: ...here.
31940 * collect-utils.c: Moved to...
31941 * collect-utils.cc: ...here.
31942 * collect2-aix.c: Moved to...
31943 * collect2-aix.cc: ...here.
31944 * collect2.c: Moved to...
31945 * collect2.cc: ...here.
31946 * combine-stack-adj.c: Moved to...
31947 * combine-stack-adj.cc: ...here.
31948 * combine.c: Moved to...
31949 * combine.cc: ...here.
31950 * common/common-targhooks.c: Moved to...
31951 * common/common-targhooks.cc: ...here.
31952 * common/config/aarch64/aarch64-common.c: Moved to...
31953 * common/config/aarch64/aarch64-common.cc: ...here.
31954 * common/config/alpha/alpha-common.c: Moved to...
31955 * common/config/alpha/alpha-common.cc: ...here.
31956 * common/config/arc/arc-common.c: Moved to...
31957 * common/config/arc/arc-common.cc: ...here.
31958 * common/config/arm/arm-common.c: Moved to...
31959 * common/config/arm/arm-common.cc: ...here.
31960 * common/config/avr/avr-common.c: Moved to...
31961 * common/config/avr/avr-common.cc: ...here.
31962 * common/config/bfin/bfin-common.c: Moved to...
31963 * common/config/bfin/bfin-common.cc: ...here.
31964 * common/config/bpf/bpf-common.c: Moved to...
31965 * common/config/bpf/bpf-common.cc: ...here.
31966 * common/config/c6x/c6x-common.c: Moved to...
31967 * common/config/c6x/c6x-common.cc: ...here.
31968 * common/config/cr16/cr16-common.c: Moved to...
31969 * common/config/cr16/cr16-common.cc: ...here.
31970 * common/config/cris/cris-common.c: Moved to...
31971 * common/config/cris/cris-common.cc: ...here.
31972 * common/config/csky/csky-common.c: Moved to...
31973 * common/config/csky/csky-common.cc: ...here.
31974 * common/config/default-common.c: Moved to...
31975 * common/config/default-common.cc: ...here.
31976 * common/config/epiphany/epiphany-common.c: Moved to...
31977 * common/config/epiphany/epiphany-common.cc: ...here.
31978 * common/config/fr30/fr30-common.c: Moved to...
31979 * common/config/fr30/fr30-common.cc: ...here.
31980 * common/config/frv/frv-common.c: Moved to...
31981 * common/config/frv/frv-common.cc: ...here.
31982 * common/config/gcn/gcn-common.c: Moved to...
31983 * common/config/gcn/gcn-common.cc: ...here.
31984 * common/config/h8300/h8300-common.c: Moved to...
31985 * common/config/h8300/h8300-common.cc: ...here.
31986 * common/config/i386/i386-common.c: Moved to...
31987 * common/config/i386/i386-common.cc: ...here.
31988 * common/config/ia64/ia64-common.c: Moved to...
31989 * common/config/ia64/ia64-common.cc: ...here.
31990 * common/config/iq2000/iq2000-common.c: Moved to...
31991 * common/config/iq2000/iq2000-common.cc: ...here.
31992 * common/config/lm32/lm32-common.c: Moved to...
31993 * common/config/lm32/lm32-common.cc: ...here.
31994 * common/config/m32r/m32r-common.c: Moved to...
31995 * common/config/m32r/m32r-common.cc: ...here.
31996 * common/config/m68k/m68k-common.c: Moved to...
31997 * common/config/m68k/m68k-common.cc: ...here.
31998 * common/config/mcore/mcore-common.c: Moved to...
31999 * common/config/mcore/mcore-common.cc: ...here.
32000 * common/config/microblaze/microblaze-common.c: Moved to...
32001 * common/config/microblaze/microblaze-common.cc: ...here.
32002 * common/config/mips/mips-common.c: Moved to...
32003 * common/config/mips/mips-common.cc: ...here.
32004 * common/config/mmix/mmix-common.c: Moved to...
32005 * common/config/mmix/mmix-common.cc: ...here.
32006 * common/config/mn10300/mn10300-common.c: Moved to...
32007 * common/config/mn10300/mn10300-common.cc: ...here.
32008 * common/config/msp430/msp430-common.c: Moved to...
32009 * common/config/msp430/msp430-common.cc: ...here.
32010 * common/config/nds32/nds32-common.c: Moved to...
32011 * common/config/nds32/nds32-common.cc: ...here.
32012 * common/config/nios2/nios2-common.c: Moved to...
32013 * common/config/nios2/nios2-common.cc: ...here.
32014 * common/config/nvptx/nvptx-common.c: Moved to...
32015 * common/config/nvptx/nvptx-common.cc: ...here.
32016 * common/config/or1k/or1k-common.c: Moved to...
32017 * common/config/or1k/or1k-common.cc: ...here.
32018 * common/config/pa/pa-common.c: Moved to...
32019 * common/config/pa/pa-common.cc: ...here.
32020 * common/config/pdp11/pdp11-common.c: Moved to...
32021 * common/config/pdp11/pdp11-common.cc: ...here.
32022 * common/config/pru/pru-common.c: Moved to...
32023 * common/config/pru/pru-common.cc: ...here.
32024 * common/config/riscv/riscv-common.c: Moved to...
32025 * common/config/riscv/riscv-common.cc: ...here.
32026 * common/config/rs6000/rs6000-common.c: Moved to...
32027 * common/config/rs6000/rs6000-common.cc: ...here.
32028 * common/config/rx/rx-common.c: Moved to...
32029 * common/config/rx/rx-common.cc: ...here.
32030 * common/config/s390/s390-common.c: Moved to...
32031 * common/config/s390/s390-common.cc: ...here.
32032 * common/config/sh/sh-common.c: Moved to...
32033 * common/config/sh/sh-common.cc: ...here.
32034 * common/config/sparc/sparc-common.c: Moved to...
32035 * common/config/sparc/sparc-common.cc: ...here.
32036 * common/config/tilegx/tilegx-common.c: Moved to...
32037 * common/config/tilegx/tilegx-common.cc: ...here.
32038 * common/config/tilepro/tilepro-common.c: Moved to...
32039 * common/config/tilepro/tilepro-common.cc: ...here.
32040 * common/config/v850/v850-common.c: Moved to...
32041 * common/config/v850/v850-common.cc: ...here.
32042 * common/config/vax/vax-common.c: Moved to...
32043 * common/config/vax/vax-common.cc: ...here.
32044 * common/config/visium/visium-common.c: Moved to...
32045 * common/config/visium/visium-common.cc: ...here.
32046 * common/config/xstormy16/xstormy16-common.c: Moved to...
32047 * common/config/xstormy16/xstormy16-common.cc: ...here.
32048 * common/config/xtensa/xtensa-common.c: Moved to...
32049 * common/config/xtensa/xtensa-common.cc: ...here.
32050 * compare-elim.c: Moved to...
32051 * compare-elim.cc: ...here.
32052 * config/aarch64/aarch64-bti-insert.c: Moved to...
32053 * config/aarch64/aarch64-bti-insert.cc: ...here.
32054 * config/aarch64/aarch64-builtins.c: Moved to...
32055 * config/aarch64/aarch64-builtins.cc: ...here.
32056 * config/aarch64/aarch64-c.c: Moved to...
32057 * config/aarch64/aarch64-c.cc: ...here.
32058 * config/aarch64/aarch64-d.c: Moved to...
32059 * config/aarch64/aarch64-d.cc: ...here.
32060 * config/aarch64/aarch64.c: Moved to...
32061 * config/aarch64/aarch64.cc: ...here.
32062 * config/aarch64/cortex-a57-fma-steering.c: Moved to...
32063 * config/aarch64/cortex-a57-fma-steering.cc: ...here.
32064 * config/aarch64/driver-aarch64.c: Moved to...
32065 * config/aarch64/driver-aarch64.cc: ...here.
32066 * config/aarch64/falkor-tag-collision-avoidance.c: Moved to...
32067 * config/aarch64/falkor-tag-collision-avoidance.cc: ...here.
32068 * config/aarch64/host-aarch64-darwin.c: Moved to...
32069 * config/aarch64/host-aarch64-darwin.cc: ...here.
32070 * config/alpha/alpha.c: Moved to...
32071 * config/alpha/alpha.cc: ...here.
32072 * config/alpha/driver-alpha.c: Moved to...
32073 * config/alpha/driver-alpha.cc: ...here.
32074 * config/arc/arc-c.c: Moved to...
32075 * config/arc/arc-c.cc: ...here.
32076 * config/arc/arc.c: Moved to...
32077 * config/arc/arc.cc: ...here.
32078 * config/arc/driver-arc.c: Moved to...
32079 * config/arc/driver-arc.cc: ...here.
32080 * config/arm/aarch-common.c: Moved to...
32081 * config/arm/aarch-common.cc: ...here.
32082 * config/arm/arm-builtins.c: Moved to...
32083 * config/arm/arm-builtins.cc: ...here.
32084 * config/arm/arm-c.c: Moved to...
32085 * config/arm/arm-c.cc: ...here.
32086 * config/arm/arm-d.c: Moved to...
32087 * config/arm/arm-d.cc: ...here.
32088 * config/arm/arm.c: Moved to...
32089 * config/arm/arm.cc: ...here.
32090 * config/arm/driver-arm.c: Moved to...
32091 * config/arm/driver-arm.cc: ...here.
32092 * config/avr/avr-c.c: Moved to...
32093 * config/avr/avr-c.cc: ...here.
32094 * config/avr/avr-devices.c: Moved to...
32095 * config/avr/avr-devices.cc: ...here.
32096 * config/avr/avr-log.c: Moved to...
32097 * config/avr/avr-log.cc: ...here.
32098 * config/avr/avr.c: Moved to...
32099 * config/avr/avr.cc: ...here.
32100 * config/avr/driver-avr.c: Moved to...
32101 * config/avr/driver-avr.cc: ...here.
32102 * config/avr/gen-avr-mmcu-specs.c: Moved to...
32103 * config/avr/gen-avr-mmcu-specs.cc: ...here.
32104 * config/avr/gen-avr-mmcu-texi.c: Moved to...
32105 * config/avr/gen-avr-mmcu-texi.cc: ...here.
32106 * config/bfin/bfin.c: Moved to...
32107 * config/bfin/bfin.cc: ...here.
32108 * config/bpf/bpf.c: Moved to...
32109 * config/bpf/bpf.cc: ...here.
32110 * config/bpf/coreout.c: Moved to...
32111 * config/bpf/coreout.cc: ...here.
32112 * config/c6x/c6x.c: Moved to...
32113 * config/c6x/c6x.cc: ...here.
32114 * config/cr16/cr16.c: Moved to...
32115 * config/cr16/cr16.cc: ...here.
32116 * config/cris/cris.c: Moved to...
32117 * config/cris/cris.cc: ...here.
32118 * config/csky/csky.c: Moved to...
32119 * config/csky/csky.cc: ...here.
32120 * config/darwin-c.c: Moved to...
32121 * config/darwin-c.cc: ...here.
32122 * config/darwin-d.c: Moved to...
32123 * config/darwin-d.cc: ...here.
32124 * config/darwin-driver.c: Moved to...
32125 * config/darwin-driver.cc: ...here.
32126 * config/darwin-f.c: Moved to...
32127 * config/darwin-f.cc: ...here.
32128 * config/darwin.c: Moved to...
32129 * config/darwin.cc: ...here.
32130 * config/default-c.c: Moved to...
32131 * config/default-c.cc: ...here.
32132 * config/default-d.c: Moved to...
32133 * config/default-d.cc: ...here.
32134 * config/dragonfly-d.c: Moved to...
32135 * config/dragonfly-d.cc: ...here.
32136 * config/epiphany/epiphany.c: Moved to...
32137 * config/epiphany/epiphany.cc: ...here.
32138 * config/epiphany/mode-switch-use.c: Moved to...
32139 * config/epiphany/mode-switch-use.cc: ...here.
32140 * config/epiphany/resolve-sw-modes.c: Moved to...
32141 * config/epiphany/resolve-sw-modes.cc: ...here.
32142 * config/fr30/fr30.c: Moved to...
32143 * config/fr30/fr30.cc: ...here.
32144 * config/freebsd-d.c: Moved to...
32145 * config/freebsd-d.cc: ...here.
32146 * config/frv/frv.c: Moved to...
32147 * config/frv/frv.cc: ...here.
32148 * config/ft32/ft32.c: Moved to...
32149 * config/ft32/ft32.cc: ...here.
32150 * config/gcn/driver-gcn.c: Moved to...
32151 * config/gcn/driver-gcn.cc: ...here.
32152 * config/gcn/gcn-run.c: Moved to...
32153 * config/gcn/gcn-run.cc: ...here.
32154 * config/gcn/gcn-tree.c: Moved to...
32155 * config/gcn/gcn-tree.cc: ...here.
32156 * config/gcn/gcn.c: Moved to...
32157 * config/gcn/gcn.cc: ...here.
32158 * config/gcn/mkoffload.c: Moved to...
32159 * config/gcn/mkoffload.cc: ...here.
32160 * config/glibc-c.c: Moved to...
32161 * config/glibc-c.cc: ...here.
32162 * config/glibc-d.c: Moved to...
32163 * config/glibc-d.cc: ...here.
32164 * config/h8300/h8300.c: Moved to...
32165 * config/h8300/h8300.cc: ...here.
32166 * config/host-darwin.c: Moved to...
32167 * config/host-darwin.cc: ...here.
32168 * config/host-hpux.c: Moved to...
32169 * config/host-hpux.cc: ...here.
32170 * config/host-linux.c: Moved to...
32171 * config/host-linux.cc: ...here.
32172 * config/host-netbsd.c: Moved to...
32173 * config/host-netbsd.cc: ...here.
32174 * config/host-openbsd.c: Moved to...
32175 * config/host-openbsd.cc: ...here.
32176 * config/host-solaris.c: Moved to...
32177 * config/host-solaris.cc: ...here.
32178 * config/i386/djgpp.c: Moved to...
32179 * config/i386/djgpp.cc: ...here.
32180 * config/i386/driver-i386.c: Moved to...
32181 * config/i386/driver-i386.cc: ...here.
32182 * config/i386/driver-mingw32.c: Moved to...
32183 * config/i386/driver-mingw32.cc: ...here.
32184 * config/i386/gnu-property.c: Moved to...
32185 * config/i386/gnu-property.cc: ...here.
32186 * config/i386/host-cygwin.c: Moved to...
32187 * config/i386/host-cygwin.cc: ...here.
32188 * config/i386/host-i386-darwin.c: Moved to...
32189 * config/i386/host-i386-darwin.cc: ...here.
32190 * config/i386/host-mingw32.c: Moved to...
32191 * config/i386/host-mingw32.cc: ...here.
32192 * config/i386/i386-builtins.c: Moved to...
32193 * config/i386/i386-builtins.cc: ...here.
32194 * config/i386/i386-c.c: Moved to...
32195 * config/i386/i386-c.cc: ...here.
32196 * config/i386/i386-d.c: Moved to...
32197 * config/i386/i386-d.cc: ...here.
32198 * config/i386/i386-expand.c: Moved to...
32199 * config/i386/i386-expand.cc: ...here.
32200 * config/i386/i386-features.c: Moved to...
32201 * config/i386/i386-features.cc: ...here.
32202 * config/i386/i386-options.c: Moved to...
32203 * config/i386/i386-options.cc: ...here.
32204 * config/i386/i386.c: Moved to...
32205 * config/i386/i386.cc: ...here.
32206 * config/i386/intelmic-mkoffload.c: Moved to...
32207 * config/i386/intelmic-mkoffload.cc: ...here.
32208 * config/i386/msformat-c.c: Moved to...
32209 * config/i386/msformat-c.cc: ...here.
32210 * config/i386/winnt-cxx.c: Moved to...
32211 * config/i386/winnt-cxx.cc: ...here.
32212 * config/i386/winnt-d.c: Moved to...
32213 * config/i386/winnt-d.cc: ...here.
32214 * config/i386/winnt-stubs.c: Moved to...
32215 * config/i386/winnt-stubs.cc: ...here.
32216 * config/i386/winnt.c: Moved to...
32217 * config/i386/winnt.cc: ...here.
32218 * config/i386/x86-tune-sched-atom.c: Moved to...
32219 * config/i386/x86-tune-sched-atom.cc: ...here.
32220 * config/i386/x86-tune-sched-bd.c: Moved to...
32221 * config/i386/x86-tune-sched-bd.cc: ...here.
32222 * config/i386/x86-tune-sched-core.c: Moved to...
32223 * config/i386/x86-tune-sched-core.cc: ...here.
32224 * config/i386/x86-tune-sched.c: Moved to...
32225 * config/i386/x86-tune-sched.cc: ...here.
32226 * config/ia64/ia64-c.c: Moved to...
32227 * config/ia64/ia64-c.cc: ...here.
32228 * config/ia64/ia64.c: Moved to...
32229 * config/ia64/ia64.cc: ...here.
32230 * config/iq2000/iq2000.c: Moved to...
32231 * config/iq2000/iq2000.cc: ...here.
32232 * config/linux.c: Moved to...
32233 * config/linux.cc: ...here.
32234 * config/lm32/lm32.c: Moved to...
32235 * config/lm32/lm32.cc: ...here.
32236 * config/m32c/m32c-pragma.c: Moved to...
32237 * config/m32c/m32c-pragma.cc: ...here.
32238 * config/m32c/m32c.c: Moved to...
32239 * config/m32c/m32c.cc: ...here.
32240 * config/m32r/m32r.c: Moved to...
32241 * config/m32r/m32r.cc: ...here.
32242 * config/m68k/m68k.c: Moved to...
32243 * config/m68k/m68k.cc: ...here.
32244 * config/mcore/mcore.c: Moved to...
32245 * config/mcore/mcore.cc: ...here.
32246 * config/microblaze/microblaze-c.c: Moved to...
32247 * config/microblaze/microblaze-c.cc: ...here.
32248 * config/microblaze/microblaze.c: Moved to...
32249 * config/microblaze/microblaze.cc: ...here.
32250 * config/mips/driver-native.c: Moved to...
32251 * config/mips/driver-native.cc: ...here.
32252 * config/mips/frame-header-opt.c: Moved to...
32253 * config/mips/frame-header-opt.cc: ...here.
32254 * config/mips/mips-d.c: Moved to...
32255 * config/mips/mips-d.cc: ...here.
32256 * config/mips/mips.c: Moved to...
32257 * config/mips/mips.cc: ...here.
32258 * config/mmix/mmix.c: Moved to...
32259 * config/mmix/mmix.cc: ...here.
32260 * config/mn10300/mn10300.c: Moved to...
32261 * config/mn10300/mn10300.cc: ...here.
32262 * config/moxie/moxie.c: Moved to...
32263 * config/moxie/moxie.cc: ...here.
32264 * config/msp430/driver-msp430.c: Moved to...
32265 * config/msp430/driver-msp430.cc: ...here.
32266 * config/msp430/msp430-c.c: Moved to...
32267 * config/msp430/msp430-c.cc: ...here.
32268 * config/msp430/msp430-devices.c: Moved to...
32269 * config/msp430/msp430-devices.cc: ...here.
32270 * config/msp430/msp430.c: Moved to...
32271 * config/msp430/msp430.cc: ...here.
32272 * config/nds32/nds32-cost.c: Moved to...
32273 * config/nds32/nds32-cost.cc: ...here.
32274 * config/nds32/nds32-fp-as-gp.c: Moved to...
32275 * config/nds32/nds32-fp-as-gp.cc: ...here.
32276 * config/nds32/nds32-intrinsic.c: Moved to...
32277 * config/nds32/nds32-intrinsic.cc: ...here.
32278 * config/nds32/nds32-isr.c: Moved to...
32279 * config/nds32/nds32-isr.cc: ...here.
32280 * config/nds32/nds32-md-auxiliary.c: Moved to...
32281 * config/nds32/nds32-md-auxiliary.cc: ...here.
32282 * config/nds32/nds32-memory-manipulation.c: Moved to...
32283 * config/nds32/nds32-memory-manipulation.cc: ...here.
32284 * config/nds32/nds32-pipelines-auxiliary.c: Moved to...
32285 * config/nds32/nds32-pipelines-auxiliary.cc: ...here.
32286 * config/nds32/nds32-predicates.c: Moved to...
32287 * config/nds32/nds32-predicates.cc: ...here.
32288 * config/nds32/nds32-relax-opt.c: Moved to...
32289 * config/nds32/nds32-relax-opt.cc: ...here.
32290 * config/nds32/nds32-utils.c: Moved to...
32291 * config/nds32/nds32-utils.cc: ...here.
32292 * config/nds32/nds32.c: Moved to...
32293 * config/nds32/nds32.cc: ...here.
32294 * config/netbsd-d.c: Moved to...
32295 * config/netbsd-d.cc: ...here.
32296 * config/netbsd.c: Moved to...
32297 * config/netbsd.cc: ...here.
32298 * config/nios2/nios2.c: Moved to...
32299 * config/nios2/nios2.cc: ...here.
32300 * config/nvptx/mkoffload.c: Moved to...
32301 * config/nvptx/mkoffload.cc: ...here.
32302 * config/nvptx/nvptx-c.c: Moved to...
32303 * config/nvptx/nvptx-c.cc: ...here.
32304 * config/nvptx/nvptx.c: Moved to...
32305 * config/nvptx/nvptx.cc: ...here.
32306 * config/openbsd-d.c: Moved to...
32307 * config/openbsd-d.cc: ...here.
32308 * config/or1k/or1k.c: Moved to...
32309 * config/or1k/or1k.cc: ...here.
32310 * config/pa/pa-d.c: Moved to...
32311 * config/pa/pa-d.cc: ...here.
32312 * config/pa/pa.c: Moved to...
32313 * config/pa/pa.cc: ...here.
32314 * config/pdp11/pdp11.c: Moved to...
32315 * config/pdp11/pdp11.cc: ...here.
32316 * config/pru/pru-passes.c: Moved to...
32317 * config/pru/pru-passes.cc: ...here.
32318 * config/pru/pru-pragma.c: Moved to...
32319 * config/pru/pru-pragma.cc: ...here.
32320 * config/pru/pru.c: Moved to...
32321 * config/pru/pru.cc: ...here.
32322 * config/riscv/riscv-builtins.c: Moved to...
32323 * config/riscv/riscv-builtins.cc: ...here.
32324 * config/riscv/riscv-c.c: Moved to...
32325 * config/riscv/riscv-c.cc: ...here.
32326 * config/riscv/riscv-d.c: Moved to...
32327 * config/riscv/riscv-d.cc: ...here.
32328 * config/riscv/riscv-shorten-memrefs.c: Moved to...
32329 * config/riscv/riscv-shorten-memrefs.cc: ...here.
32330 * config/riscv/riscv-sr.c: Moved to...
32331 * config/riscv/riscv-sr.cc: ...here.
32332 * config/riscv/riscv.c: Moved to...
32333 * config/riscv/riscv.cc: ...here.
32334 * config/rl78/rl78-c.c: Moved to...
32335 * config/rl78/rl78-c.cc: ...here.
32336 * config/rl78/rl78.c: Moved to...
32337 * config/rl78/rl78.cc: ...here.
32338 * config/rs6000/driver-rs6000.c: Moved to...
32339 * config/rs6000/driver-rs6000.cc: ...here.
32340 * config/rs6000/host-darwin.c: Moved to...
32341 * config/rs6000/host-darwin.cc: ...here.
32342 * config/rs6000/host-ppc64-darwin.c: Moved to...
32343 * config/rs6000/host-ppc64-darwin.cc: ...here.
32344 * config/rs6000/rbtree.c: Moved to...
32345 * config/rs6000/rbtree.cc: ...here.
32346 * config/rs6000/rs6000-c.c: Moved to...
32347 * config/rs6000/rs6000-c.cc: ...here.
32348 * config/rs6000/rs6000-call.c: Moved to...
32349 * config/rs6000/rs6000-call.cc: ...here.
32350 * config/rs6000/rs6000-d.c: Moved to...
32351 * config/rs6000/rs6000-d.cc: ...here.
32352 * config/rs6000/rs6000-gen-builtins.c: Moved to...
32353 * config/rs6000/rs6000-gen-builtins.cc: ...here.
32354 * config/rs6000/rs6000-linux.c: Moved to...
32355 * config/rs6000/rs6000-linux.cc: ...here.
32356 * config/rs6000/rs6000-logue.c: Moved to...
32357 * config/rs6000/rs6000-logue.cc: ...here.
32358 * config/rs6000/rs6000-p8swap.c: Moved to...
32359 * config/rs6000/rs6000-p8swap.cc: ...here.
32360 * config/rs6000/rs6000-pcrel-opt.c: Moved to...
32361 * config/rs6000/rs6000-pcrel-opt.cc: ...here.
32362 * config/rs6000/rs6000-string.c: Moved to...
32363 * config/rs6000/rs6000-string.cc: ...here.
32364 * config/rs6000/rs6000.c: Moved to...
32365 * config/rs6000/rs6000.cc: ...here.
32366 * config/rx/rx.c: Moved to...
32367 * config/rx/rx.cc: ...here.
32368 * config/s390/driver-native.c: Moved to...
32369 * config/s390/driver-native.cc: ...here.
32370 * config/s390/s390-c.c: Moved to...
32371 * config/s390/s390-c.cc: ...here.
32372 * config/s390/s390-d.c: Moved to...
32373 * config/s390/s390-d.cc: ...here.
32374 * config/s390/s390.c: Moved to...
32375 * config/s390/s390.cc: ...here.
32376 * config/sh/divtab-sh4-300.c: Moved to...
32377 * config/sh/divtab-sh4-300.cc: ...here.
32378 * config/sh/divtab-sh4.c: Moved to...
32379 * config/sh/divtab-sh4.cc: ...here.
32380 * config/sh/divtab.c: Moved to...
32381 * config/sh/divtab.cc: ...here.
32382 * config/sh/sh-c.c: Moved to...
32383 * config/sh/sh-c.cc: ...here.
32384 * config/sh/sh.c: Moved to...
32385 * config/sh/sh.cc: ...here.
32386 * config/sol2-c.c: Moved to...
32387 * config/sol2-c.cc: ...here.
32388 * config/sol2-cxx.c: Moved to...
32389 * config/sol2-cxx.cc: ...here.
32390 * config/sol2-d.c: Moved to...
32391 * config/sol2-d.cc: ...here.
32392 * config/sol2-stubs.c: Moved to...
32393 * config/sol2-stubs.cc: ...here.
32394 * config/sol2.c: Moved to...
32395 * config/sol2.cc: ...here.
32396 * config/sparc/driver-sparc.c: Moved to...
32397 * config/sparc/driver-sparc.cc: ...here.
32398 * config/sparc/sparc-c.c: Moved to...
32399 * config/sparc/sparc-c.cc: ...here.
32400 * config/sparc/sparc-d.c: Moved to...
32401 * config/sparc/sparc-d.cc: ...here.
32402 * config/sparc/sparc.c: Moved to...
32403 * config/sparc/sparc.cc: ...here.
32404 * config/stormy16/stormy16.c: Moved to...
32405 * config/stormy16/stormy16.cc: ...here.
32406 * config/tilegx/mul-tables.c: Moved to...
32407 * config/tilegx/mul-tables.cc: ...here.
32408 * config/tilegx/tilegx-c.c: Moved to...
32409 * config/tilegx/tilegx-c.cc: ...here.
32410 * config/tilegx/tilegx.c: Moved to...
32411 * config/tilegx/tilegx.cc: ...here.
32412 * config/tilepro/mul-tables.c: Moved to...
32413 * config/tilepro/mul-tables.cc: ...here.
32414 * config/tilepro/tilepro-c.c: Moved to...
32415 * config/tilepro/tilepro-c.cc: ...here.
32416 * config/tilepro/tilepro.c: Moved to...
32417 * config/tilepro/tilepro.cc: ...here.
32418 * config/v850/v850-c.c: Moved to...
32419 * config/v850/v850-c.cc: ...here.
32420 * config/v850/v850.c: Moved to...
32421 * config/v850/v850.cc: ...here.
32422 * config/vax/vax.c: Moved to...
32423 * config/vax/vax.cc: ...here.
32424 * config/visium/visium.c: Moved to...
32425 * config/visium/visium.cc: ...here.
32426 * config/vms/vms-c.c: Moved to...
32427 * config/vms/vms-c.cc: ...here.
32428 * config/vms/vms-f.c: Moved to...
32429 * config/vms/vms-f.cc: ...here.
32430 * config/vms/vms.c: Moved to...
32431 * config/vms/vms.cc: ...here.
32432 * config/vxworks-c.c: Moved to...
32433 * config/vxworks-c.cc: ...here.
32434 * config/vxworks.c: Moved to...
32435 * config/vxworks.cc: ...here.
32436 * config/winnt-c.c: Moved to...
32437 * config/winnt-c.cc: ...here.
32438 * config/xtensa/xtensa.c: Moved to...
32439 * config/xtensa/xtensa.cc: ...here.
32440 * context.c: Moved to...
32441 * context.cc: ...here.
32442 * convert.c: Moved to...
32443 * convert.cc: ...here.
32444 * coverage.c: Moved to...
32445 * coverage.cc: ...here.
32446 * cppbuiltin.c: Moved to...
32447 * cppbuiltin.cc: ...here.
32448 * cppdefault.c: Moved to...
32449 * cppdefault.cc: ...here.
32450 * cprop.c: Moved to...
32451 * cprop.cc: ...here.
32452 * cse.c: Moved to...
32454 * cselib.c: Moved to...
32455 * cselib.cc: ...here.
32456 * ctfc.c: Moved to...
32457 * ctfc.cc: ...here.
32458 * ctfout.c: Moved to...
32459 * ctfout.cc: ...here.
32460 * data-streamer-in.c: Moved to...
32461 * data-streamer-in.cc: ...here.
32462 * data-streamer-out.c: Moved to...
32463 * data-streamer-out.cc: ...here.
32464 * data-streamer.c: Moved to...
32465 * data-streamer.cc: ...here.
32466 * dbgcnt.c: Moved to...
32467 * dbgcnt.cc: ...here.
32468 * dbxout.c: Moved to...
32469 * dbxout.cc: ...here.
32470 * dce.c: Moved to...
32472 * ddg.c: Moved to...
32474 * debug.c: Moved to...
32475 * debug.cc: ...here.
32476 * df-core.c: Moved to...
32477 * df-core.cc: ...here.
32478 * df-problems.c: Moved to...
32479 * df-problems.cc: ...here.
32480 * df-scan.c: Moved to...
32481 * df-scan.cc: ...here.
32482 * dfp.c: Moved to...
32484 * diagnostic-color.c: Moved to...
32485 * diagnostic-color.cc: ...here.
32486 * diagnostic-show-locus.c: Moved to...
32487 * diagnostic-show-locus.cc: ...here.
32488 * diagnostic-spec.c: Moved to...
32489 * diagnostic-spec.cc: ...here.
32490 * diagnostic.c: Moved to...
32491 * diagnostic.cc: ...here.
32492 * dojump.c: Moved to...
32493 * dojump.cc: ...here.
32494 * dominance.c: Moved to...
32495 * dominance.cc: ...here.
32496 * domwalk.c: Moved to...
32497 * domwalk.cc: ...here.
32498 * double-int.c: Moved to...
32499 * double-int.cc: ...here.
32500 * dse.c: Moved to...
32502 * dumpfile.c: Moved to...
32503 * dumpfile.cc: ...here.
32504 * dwarf2asm.c: Moved to...
32505 * dwarf2asm.cc: ...here.
32506 * dwarf2cfi.c: Moved to...
32507 * dwarf2cfi.cc: ...here.
32508 * dwarf2ctf.c: Moved to...
32509 * dwarf2ctf.cc: ...here.
32510 * dwarf2out.c: Moved to...
32511 * dwarf2out.cc: ...here.
32512 * early-remat.c: Moved to...
32513 * early-remat.cc: ...here.
32514 * edit-context.c: Moved to...
32515 * edit-context.cc: ...here.
32516 * emit-rtl.c: Moved to...
32517 * emit-rtl.cc: ...here.
32518 * errors.c: Moved to...
32519 * errors.cc: ...here.
32520 * et-forest.c: Moved to...
32521 * et-forest.cc: ...here.
32522 * except.c: Moved to...
32523 * except.cc: ...here.
32524 * explow.c: Moved to...
32525 * explow.cc: ...here.
32526 * expmed.c: Moved to...
32527 * expmed.cc: ...here.
32528 * expr.c: Moved to...
32529 * expr.cc: ...here.
32530 * fibonacci_heap.c: Moved to...
32531 * fibonacci_heap.cc: ...here.
32532 * file-find.c: Moved to...
32533 * file-find.cc: ...here.
32534 * file-prefix-map.c: Moved to...
32535 * file-prefix-map.cc: ...here.
32536 * final.c: Moved to...
32537 * final.cc: ...here.
32538 * fixed-value.c: Moved to...
32539 * fixed-value.cc: ...here.
32540 * fold-const-call.c: Moved to...
32541 * fold-const-call.cc: ...here.
32542 * fold-const.c: Moved to...
32543 * fold-const.cc: ...here.
32544 * fp-test.c: Moved to...
32545 * fp-test.cc: ...here.
32546 * function-tests.c: Moved to...
32547 * function-tests.cc: ...here.
32548 * function.c: Moved to...
32549 * function.cc: ...here.
32550 * fwprop.c: Moved to...
32551 * fwprop.cc: ...here.
32552 * gcc-ar.c: Moved to...
32553 * gcc-ar.cc: ...here.
32554 * gcc-main.c: Moved to...
32555 * gcc-main.cc: ...here.
32556 * gcc-rich-location.c: Moved to...
32557 * gcc-rich-location.cc: ...here.
32558 * gcc.c: Moved to...
32560 * gcov-dump.c: Moved to...
32561 * gcov-dump.cc: ...here.
32562 * gcov-io.c: Moved to...
32563 * gcov-io.cc: ...here.
32564 * gcov-tool.c: Moved to...
32565 * gcov-tool.cc: ...here.
32566 * gcov.c: Moved to...
32567 * gcov.cc: ...here.
32568 * gcse-common.c: Moved to...
32569 * gcse-common.cc: ...here.
32570 * gcse.c: Moved to...
32571 * gcse.cc: ...here.
32572 * genattr-common.c: Moved to...
32573 * genattr-common.cc: ...here.
32574 * genattr.c: Moved to...
32575 * genattr.cc: ...here.
32576 * genattrtab.c: Moved to...
32577 * genattrtab.cc: ...here.
32578 * genautomata.c: Moved to...
32579 * genautomata.cc: ...here.
32580 * gencfn-macros.c: Moved to...
32581 * gencfn-macros.cc: ...here.
32582 * gencheck.c: Moved to...
32583 * gencheck.cc: ...here.
32584 * genchecksum.c: Moved to...
32585 * genchecksum.cc: ...here.
32586 * gencodes.c: Moved to...
32587 * gencodes.cc: ...here.
32588 * genconditions.c: Moved to...
32589 * genconditions.cc: ...here.
32590 * genconfig.c: Moved to...
32591 * genconfig.cc: ...here.
32592 * genconstants.c: Moved to...
32593 * genconstants.cc: ...here.
32594 * genemit.c: Moved to...
32595 * genemit.cc: ...here.
32596 * genenums.c: Moved to...
32597 * genenums.cc: ...here.
32598 * generic-match-head.c: Moved to...
32599 * generic-match-head.cc: ...here.
32600 * genextract.c: Moved to...
32601 * genextract.cc: ...here.
32602 * genflags.c: Moved to...
32603 * genflags.cc: ...here.
32604 * gengenrtl.c: Moved to...
32605 * gengenrtl.cc: ...here.
32606 * gengtype-parse.c: Moved to...
32607 * gengtype-parse.cc: ...here.
32608 * gengtype-state.c: Moved to...
32609 * gengtype-state.cc: ...here.
32610 * gengtype.c: Moved to...
32611 * gengtype.cc: ...here.
32612 * genhooks.c: Moved to...
32613 * genhooks.cc: ...here.
32614 * genmatch.c: Moved to...
32615 * genmatch.cc: ...here.
32616 * genmddeps.c: Moved to...
32617 * genmddeps.cc: ...here.
32618 * genmddump.c: Moved to...
32619 * genmddump.cc: ...here.
32620 * genmodes.c: Moved to...
32621 * genmodes.cc: ...here.
32622 * genopinit.c: Moved to...
32623 * genopinit.cc: ...here.
32624 * genoutput.c: Moved to...
32625 * genoutput.cc: ...here.
32626 * genpeep.c: Moved to...
32627 * genpeep.cc: ...here.
32628 * genpreds.c: Moved to...
32629 * genpreds.cc: ...here.
32630 * genrecog.c: Moved to...
32631 * genrecog.cc: ...here.
32632 * gensupport.c: Moved to...
32633 * gensupport.cc: ...here.
32634 * gentarget-def.c: Moved to...
32635 * gentarget-def.cc: ...here.
32636 * genversion.c: Moved to...
32637 * genversion.cc: ...here.
32638 * ggc-common.c: Moved to...
32639 * ggc-common.cc: ...here.
32640 * ggc-none.c: Moved to...
32641 * ggc-none.cc: ...here.
32642 * ggc-page.c: Moved to...
32643 * ggc-page.cc: ...here.
32644 * ggc-tests.c: Moved to...
32645 * ggc-tests.cc: ...here.
32646 * gimple-builder.c: Moved to...
32647 * gimple-builder.cc: ...here.
32648 * gimple-expr.c: Moved to...
32649 * gimple-expr.cc: ...here.
32650 * gimple-fold.c: Moved to...
32651 * gimple-fold.cc: ...here.
32652 * gimple-iterator.c: Moved to...
32653 * gimple-iterator.cc: ...here.
32654 * gimple-laddress.c: Moved to...
32655 * gimple-laddress.cc: ...here.
32656 * gimple-loop-jam.c: Moved to...
32657 * gimple-loop-jam.cc: ...here.
32658 * gimple-low.c: Moved to...
32659 * gimple-low.cc: ...here.
32660 * gimple-match-head.c: Moved to...
32661 * gimple-match-head.cc: ...here.
32662 * gimple-pretty-print.c: Moved to...
32663 * gimple-pretty-print.cc: ...here.
32664 * gimple-ssa-backprop.c: Moved to...
32665 * gimple-ssa-backprop.cc: ...here.
32666 * gimple-ssa-evrp-analyze.c: Moved to...
32667 * gimple-ssa-evrp-analyze.cc: ...here.
32668 * gimple-ssa-evrp.c: Moved to...
32669 * gimple-ssa-evrp.cc: ...here.
32670 * gimple-ssa-isolate-paths.c: Moved to...
32671 * gimple-ssa-isolate-paths.cc: ...here.
32672 * gimple-ssa-nonnull-compare.c: Moved to...
32673 * gimple-ssa-nonnull-compare.cc: ...here.
32674 * gimple-ssa-split-paths.c: Moved to...
32675 * gimple-ssa-split-paths.cc: ...here.
32676 * gimple-ssa-sprintf.c: Moved to...
32677 * gimple-ssa-sprintf.cc: ...here.
32678 * gimple-ssa-store-merging.c: Moved to...
32679 * gimple-ssa-store-merging.cc: ...here.
32680 * gimple-ssa-strength-reduction.c: Moved to...
32681 * gimple-ssa-strength-reduction.cc: ...here.
32682 * gimple-ssa-warn-alloca.c: Moved to...
32683 * gimple-ssa-warn-alloca.cc: ...here.
32684 * gimple-ssa-warn-restrict.c: Moved to...
32685 * gimple-ssa-warn-restrict.cc: ...here.
32686 * gimple-streamer-in.c: Moved to...
32687 * gimple-streamer-in.cc: ...here.
32688 * gimple-streamer-out.c: Moved to...
32689 * gimple-streamer-out.cc: ...here.
32690 * gimple-walk.c: Moved to...
32691 * gimple-walk.cc: ...here.
32692 * gimple-warn-recursion.c: Moved to...
32693 * gimple-warn-recursion.cc: ...here.
32694 * gimple.c: Moved to...
32695 * gimple.cc: ...here.
32696 * gimplify-me.c: Moved to...
32697 * gimplify-me.cc: ...here.
32698 * gimplify.c: Moved to...
32699 * gimplify.cc: ...here.
32700 * godump.c: Moved to...
32701 * godump.cc: ...here.
32702 * graph.c: Moved to...
32703 * graph.cc: ...here.
32704 * graphds.c: Moved to...
32705 * graphds.cc: ...here.
32706 * graphite-dependences.c: Moved to...
32707 * graphite-dependences.cc: ...here.
32708 * graphite-isl-ast-to-gimple.c: Moved to...
32709 * graphite-isl-ast-to-gimple.cc: ...here.
32710 * graphite-optimize-isl.c: Moved to...
32711 * graphite-optimize-isl.cc: ...here.
32712 * graphite-poly.c: Moved to...
32713 * graphite-poly.cc: ...here.
32714 * graphite-scop-detection.c: Moved to...
32715 * graphite-scop-detection.cc: ...here.
32716 * graphite-sese-to-poly.c: Moved to...
32717 * graphite-sese-to-poly.cc: ...here.
32718 * graphite.c: Moved to...
32719 * graphite.cc: ...here.
32720 * haifa-sched.c: Moved to...
32721 * haifa-sched.cc: ...here.
32722 * hash-map-tests.c: Moved to...
32723 * hash-map-tests.cc: ...here.
32724 * hash-set-tests.c: Moved to...
32725 * hash-set-tests.cc: ...here.
32726 * hash-table.c: Moved to...
32727 * hash-table.cc: ...here.
32728 * hooks.c: Moved to...
32729 * hooks.cc: ...here.
32730 * host-default.c: Moved to...
32731 * host-default.cc: ...here.
32732 * hw-doloop.c: Moved to...
32733 * hw-doloop.cc: ...here.
32734 * hwint.c: Moved to...
32735 * hwint.cc: ...here.
32736 * ifcvt.c: Moved to...
32737 * ifcvt.cc: ...here.
32738 * inchash.c: Moved to...
32739 * inchash.cc: ...here.
32740 * incpath.c: Moved to...
32741 * incpath.cc: ...here.
32742 * init-regs.c: Moved to...
32743 * init-regs.cc: ...here.
32744 * input.c: Moved to...
32745 * input.cc: ...here.
32746 * internal-fn.c: Moved to...
32747 * internal-fn.cc: ...here.
32748 * intl.c: Moved to...
32749 * intl.cc: ...here.
32750 * ipa-comdats.c: Moved to...
32751 * ipa-comdats.cc: ...here.
32752 * ipa-cp.c: Moved to...
32753 * ipa-cp.cc: ...here.
32754 * ipa-devirt.c: Moved to...
32755 * ipa-devirt.cc: ...here.
32756 * ipa-fnsummary.c: Moved to...
32757 * ipa-fnsummary.cc: ...here.
32758 * ipa-icf-gimple.c: Moved to...
32759 * ipa-icf-gimple.cc: ...here.
32760 * ipa-icf.c: Moved to...
32761 * ipa-icf.cc: ...here.
32762 * ipa-inline-analysis.c: Moved to...
32763 * ipa-inline-analysis.cc: ...here.
32764 * ipa-inline-transform.c: Moved to...
32765 * ipa-inline-transform.cc: ...here.
32766 * ipa-inline.c: Moved to...
32767 * ipa-inline.cc: ...here.
32768 * ipa-modref-tree.c: Moved to...
32769 * ipa-modref-tree.cc: ...here.
32770 * ipa-modref.c: Moved to...
32771 * ipa-modref.cc: ...here.
32772 * ipa-param-manipulation.c: Moved to...
32773 * ipa-param-manipulation.cc: ...here.
32774 * ipa-polymorphic-call.c: Moved to...
32775 * ipa-polymorphic-call.cc: ...here.
32776 * ipa-predicate.c: Moved to...
32777 * ipa-predicate.cc: ...here.
32778 * ipa-profile.c: Moved to...
32779 * ipa-profile.cc: ...here.
32780 * ipa-prop.c: Moved to...
32781 * ipa-prop.cc: ...here.
32782 * ipa-pure-const.c: Moved to...
32783 * ipa-pure-const.cc: ...here.
32784 * ipa-ref.c: Moved to...
32785 * ipa-ref.cc: ...here.
32786 * ipa-reference.c: Moved to...
32787 * ipa-reference.cc: ...here.
32788 * ipa-split.c: Moved to...
32789 * ipa-split.cc: ...here.
32790 * ipa-sra.c: Moved to...
32791 * ipa-sra.cc: ...here.
32792 * ipa-utils.c: Moved to...
32793 * ipa-utils.cc: ...here.
32794 * ipa-visibility.c: Moved to...
32795 * ipa-visibility.cc: ...here.
32796 * ipa.c: Moved to...
32798 * ira-build.c: Moved to...
32799 * ira-build.cc: ...here.
32800 * ira-color.c: Moved to...
32801 * ira-color.cc: ...here.
32802 * ira-conflicts.c: Moved to...
32803 * ira-conflicts.cc: ...here.
32804 * ira-costs.c: Moved to...
32805 * ira-costs.cc: ...here.
32806 * ira-emit.c: Moved to...
32807 * ira-emit.cc: ...here.
32808 * ira-lives.c: Moved to...
32809 * ira-lives.cc: ...here.
32810 * ira.c: Moved to...
32812 * jump.c: Moved to...
32813 * jump.cc: ...here.
32814 * langhooks.c: Moved to...
32815 * langhooks.cc: ...here.
32816 * lcm.c: Moved to...
32818 * lists.c: Moved to...
32819 * lists.cc: ...here.
32820 * loop-doloop.c: Moved to...
32821 * loop-doloop.cc: ...here.
32822 * loop-init.c: Moved to...
32823 * loop-init.cc: ...here.
32824 * loop-invariant.c: Moved to...
32825 * loop-invariant.cc: ...here.
32826 * loop-iv.c: Moved to...
32827 * loop-iv.cc: ...here.
32828 * loop-unroll.c: Moved to...
32829 * loop-unroll.cc: ...here.
32830 * lower-subreg.c: Moved to...
32831 * lower-subreg.cc: ...here.
32832 * lra-assigns.c: Moved to...
32833 * lra-assigns.cc: ...here.
32834 * lra-coalesce.c: Moved to...
32835 * lra-coalesce.cc: ...here.
32836 * lra-constraints.c: Moved to...
32837 * lra-constraints.cc: ...here.
32838 * lra-eliminations.c: Moved to...
32839 * lra-eliminations.cc: ...here.
32840 * lra-lives.c: Moved to...
32841 * lra-lives.cc: ...here.
32842 * lra-remat.c: Moved to...
32843 * lra-remat.cc: ...here.
32844 * lra-spills.c: Moved to...
32845 * lra-spills.cc: ...here.
32846 * lra.c: Moved to...
32848 * lto-cgraph.c: Moved to...
32849 * lto-cgraph.cc: ...here.
32850 * lto-compress.c: Moved to...
32851 * lto-compress.cc: ...here.
32852 * lto-opts.c: Moved to...
32853 * lto-opts.cc: ...here.
32854 * lto-section-in.c: Moved to...
32855 * lto-section-in.cc: ...here.
32856 * lto-section-out.c: Moved to...
32857 * lto-section-out.cc: ...here.
32858 * lto-streamer-in.c: Moved to...
32859 * lto-streamer-in.cc: ...here.
32860 * lto-streamer-out.c: Moved to...
32861 * lto-streamer-out.cc: ...here.
32862 * lto-streamer.c: Moved to...
32863 * lto-streamer.cc: ...here.
32864 * lto-wrapper.c: Moved to...
32865 * lto-wrapper.cc: ...here.
32866 * main.c: Moved to...
32867 * main.cc: ...here.
32868 * mcf.c: Moved to...
32870 * mode-switching.c: Moved to...
32871 * mode-switching.cc: ...here.
32872 * modulo-sched.c: Moved to...
32873 * modulo-sched.cc: ...here.
32874 * multiple_target.c: Moved to...
32875 * multiple_target.cc: ...here.
32876 * omp-expand.c: Moved to...
32877 * omp-expand.cc: ...here.
32878 * omp-general.c: Moved to...
32879 * omp-general.cc: ...here.
32880 * omp-low.c: Moved to...
32881 * omp-low.cc: ...here.
32882 * omp-offload.c: Moved to...
32883 * omp-offload.cc: ...here.
32884 * omp-simd-clone.c: Moved to...
32885 * omp-simd-clone.cc: ...here.
32886 * opt-suggestions.c: Moved to...
32887 * opt-suggestions.cc: ...here.
32888 * optabs-libfuncs.c: Moved to...
32889 * optabs-libfuncs.cc: ...here.
32890 * optabs-query.c: Moved to...
32891 * optabs-query.cc: ...here.
32892 * optabs-tree.c: Moved to...
32893 * optabs-tree.cc: ...here.
32894 * optabs.c: Moved to...
32895 * optabs.cc: ...here.
32896 * opts-common.c: Moved to...
32897 * opts-common.cc: ...here.
32898 * opts-global.c: Moved to...
32899 * opts-global.cc: ...here.
32900 * opts.c: Moved to...
32901 * opts.cc: ...here.
32902 * passes.c: Moved to...
32903 * passes.cc: ...here.
32904 * plugin.c: Moved to...
32905 * plugin.cc: ...here.
32906 * postreload-gcse.c: Moved to...
32907 * postreload-gcse.cc: ...here.
32908 * postreload.c: Moved to...
32909 * postreload.cc: ...here.
32910 * predict.c: Moved to...
32911 * predict.cc: ...here.
32912 * prefix.c: Moved to...
32913 * prefix.cc: ...here.
32914 * pretty-print.c: Moved to...
32915 * pretty-print.cc: ...here.
32916 * print-rtl-function.c: Moved to...
32917 * print-rtl-function.cc: ...here.
32918 * print-rtl.c: Moved to...
32919 * print-rtl.cc: ...here.
32920 * print-tree.c: Moved to...
32921 * print-tree.cc: ...here.
32922 * profile-count.c: Moved to...
32923 * profile-count.cc: ...here.
32924 * profile.c: Moved to...
32925 * profile.cc: ...here.
32926 * read-md.c: Moved to...
32927 * read-md.cc: ...here.
32928 * read-rtl-function.c: Moved to...
32929 * read-rtl-function.cc: ...here.
32930 * read-rtl.c: Moved to...
32931 * read-rtl.cc: ...here.
32932 * real.c: Moved to...
32933 * real.cc: ...here.
32934 * realmpfr.c: Moved to...
32935 * realmpfr.cc: ...here.
32936 * recog.c: Moved to...
32937 * recog.cc: ...here.
32938 * ree.c: Moved to...
32940 * reg-stack.c: Moved to...
32941 * reg-stack.cc: ...here.
32942 * regcprop.c: Moved to...
32943 * regcprop.cc: ...here.
32944 * reginfo.c: Moved to...
32945 * reginfo.cc: ...here.
32946 * regrename.c: Moved to...
32947 * regrename.cc: ...here.
32948 * regstat.c: Moved to...
32949 * regstat.cc: ...here.
32950 * reload.c: Moved to...
32951 * reload.cc: ...here.
32952 * reload1.c: Moved to...
32953 * reload1.cc: ...here.
32954 * reorg.c: Moved to...
32955 * reorg.cc: ...here.
32956 * resource.c: Moved to...
32957 * resource.cc: ...here.
32958 * rtl-error.c: Moved to...
32959 * rtl-error.cc: ...here.
32960 * rtl-tests.c: Moved to...
32961 * rtl-tests.cc: ...here.
32962 * rtl.c: Moved to...
32964 * rtlanal.c: Moved to...
32965 * rtlanal.cc: ...here.
32966 * rtlhash.c: Moved to...
32967 * rtlhash.cc: ...here.
32968 * rtlhooks.c: Moved to...
32969 * rtlhooks.cc: ...here.
32970 * rtx-vector-builder.c: Moved to...
32971 * rtx-vector-builder.cc: ...here.
32972 * run-rtl-passes.c: Moved to...
32973 * run-rtl-passes.cc: ...here.
32974 * sancov.c: Moved to...
32975 * sancov.cc: ...here.
32976 * sanopt.c: Moved to...
32977 * sanopt.cc: ...here.
32978 * sbitmap.c: Moved to...
32979 * sbitmap.cc: ...here.
32980 * sched-deps.c: Moved to...
32981 * sched-deps.cc: ...here.
32982 * sched-ebb.c: Moved to...
32983 * sched-ebb.cc: ...here.
32984 * sched-rgn.c: Moved to...
32985 * sched-rgn.cc: ...here.
32986 * sel-sched-dump.c: Moved to...
32987 * sel-sched-dump.cc: ...here.
32988 * sel-sched-ir.c: Moved to...
32989 * sel-sched-ir.cc: ...here.
32990 * sel-sched.c: Moved to...
32991 * sel-sched.cc: ...here.
32992 * selftest-diagnostic.c: Moved to...
32993 * selftest-diagnostic.cc: ...here.
32994 * selftest-rtl.c: Moved to...
32995 * selftest-rtl.cc: ...here.
32996 * selftest-run-tests.c: Moved to...
32997 * selftest-run-tests.cc: ...here.
32998 * selftest.c: Moved to...
32999 * selftest.cc: ...here.
33000 * sese.c: Moved to...
33001 * sese.cc: ...here.
33002 * shrink-wrap.c: Moved to...
33003 * shrink-wrap.cc: ...here.
33004 * simplify-rtx.c: Moved to...
33005 * simplify-rtx.cc: ...here.
33006 * sparseset.c: Moved to...
33007 * sparseset.cc: ...here.
33008 * spellcheck-tree.c: Moved to...
33009 * spellcheck-tree.cc: ...here.
33010 * spellcheck.c: Moved to...
33011 * spellcheck.cc: ...here.
33012 * sreal.c: Moved to...
33013 * sreal.cc: ...here.
33014 * stack-ptr-mod.c: Moved to...
33015 * stack-ptr-mod.cc: ...here.
33016 * statistics.c: Moved to...
33017 * statistics.cc: ...here.
33018 * stmt.c: Moved to...
33019 * stmt.cc: ...here.
33020 * stor-layout.c: Moved to...
33021 * stor-layout.cc: ...here.
33022 * store-motion.c: Moved to...
33023 * store-motion.cc: ...here.
33024 * streamer-hooks.c: Moved to...
33025 * streamer-hooks.cc: ...here.
33026 * stringpool.c: Moved to...
33027 * stringpool.cc: ...here.
33028 * substring-locations.c: Moved to...
33029 * substring-locations.cc: ...here.
33030 * symtab.c: Moved to...
33031 * symtab.cc: ...here.
33032 * target-globals.c: Moved to...
33033 * target-globals.cc: ...here.
33034 * targhooks.c: Moved to...
33035 * targhooks.cc: ...here.
33036 * timevar.c: Moved to...
33037 * timevar.cc: ...here.
33038 * toplev.c: Moved to...
33039 * toplev.cc: ...here.
33040 * tracer.c: Moved to...
33041 * tracer.cc: ...here.
33042 * trans-mem.c: Moved to...
33043 * trans-mem.cc: ...here.
33044 * tree-affine.c: Moved to...
33045 * tree-affine.cc: ...here.
33046 * tree-call-cdce.c: Moved to...
33047 * tree-call-cdce.cc: ...here.
33048 * tree-cfg.c: Moved to...
33049 * tree-cfg.cc: ...here.
33050 * tree-cfgcleanup.c: Moved to...
33051 * tree-cfgcleanup.cc: ...here.
33052 * tree-chrec.c: Moved to...
33053 * tree-chrec.cc: ...here.
33054 * tree-complex.c: Moved to...
33055 * tree-complex.cc: ...here.
33056 * tree-data-ref.c: Moved to...
33057 * tree-data-ref.cc: ...here.
33058 * tree-dfa.c: Moved to...
33059 * tree-dfa.cc: ...here.
33060 * tree-diagnostic.c: Moved to...
33061 * tree-diagnostic.cc: ...here.
33062 * tree-dump.c: Moved to...
33063 * tree-dump.cc: ...here.
33064 * tree-eh.c: Moved to...
33065 * tree-eh.cc: ...here.
33066 * tree-emutls.c: Moved to...
33067 * tree-emutls.cc: ...here.
33068 * tree-if-conv.c: Moved to...
33069 * tree-if-conv.cc: ...here.
33070 * tree-inline.c: Moved to...
33071 * tree-inline.cc: ...here.
33072 * tree-into-ssa.c: Moved to...
33073 * tree-into-ssa.cc: ...here.
33074 * tree-iterator.c: Moved to...
33075 * tree-iterator.cc: ...here.
33076 * tree-loop-distribution.c: Moved to...
33077 * tree-loop-distribution.cc: ...here.
33078 * tree-nested.c: Moved to...
33079 * tree-nested.cc: ...here.
33080 * tree-nrv.c: Moved to...
33081 * tree-nrv.cc: ...here.
33082 * tree-object-size.c: Moved to...
33083 * tree-object-size.cc: ...here.
33084 * tree-outof-ssa.c: Moved to...
33085 * tree-outof-ssa.cc: ...here.
33086 * tree-parloops.c: Moved to...
33087 * tree-parloops.cc: ...here.
33088 * tree-phinodes.c: Moved to...
33089 * tree-phinodes.cc: ...here.
33090 * tree-predcom.c: Moved to...
33091 * tree-predcom.cc: ...here.
33092 * tree-pretty-print.c: Moved to...
33093 * tree-pretty-print.cc: ...here.
33094 * tree-profile.c: Moved to...
33095 * tree-profile.cc: ...here.
33096 * tree-scalar-evolution.c: Moved to...
33097 * tree-scalar-evolution.cc: ...here.
33098 * tree-sra.c: Moved to...
33099 * tree-sra.cc: ...here.
33100 * tree-ssa-address.c: Moved to...
33101 * tree-ssa-address.cc: ...here.
33102 * tree-ssa-alias.c: Moved to...
33103 * tree-ssa-alias.cc: ...here.
33104 * tree-ssa-ccp.c: Moved to...
33105 * tree-ssa-ccp.cc: ...here.
33106 * tree-ssa-coalesce.c: Moved to...
33107 * tree-ssa-coalesce.cc: ...here.
33108 * tree-ssa-copy.c: Moved to...
33109 * tree-ssa-copy.cc: ...here.
33110 * tree-ssa-dce.c: Moved to...
33111 * tree-ssa-dce.cc: ...here.
33112 * tree-ssa-dom.c: Moved to...
33113 * tree-ssa-dom.cc: ...here.
33114 * tree-ssa-dse.c: Moved to...
33115 * tree-ssa-dse.cc: ...here.
33116 * tree-ssa-forwprop.c: Moved to...
33117 * tree-ssa-forwprop.cc: ...here.
33118 * tree-ssa-ifcombine.c: Moved to...
33119 * tree-ssa-ifcombine.cc: ...here.
33120 * tree-ssa-live.c: Moved to...
33121 * tree-ssa-live.cc: ...here.
33122 * tree-ssa-loop-ch.c: Moved to...
33123 * tree-ssa-loop-ch.cc: ...here.
33124 * tree-ssa-loop-im.c: Moved to...
33125 * tree-ssa-loop-im.cc: ...here.
33126 * tree-ssa-loop-ivcanon.c: Moved to...
33127 * tree-ssa-loop-ivcanon.cc: ...here.
33128 * tree-ssa-loop-ivopts.c: Moved to...
33129 * tree-ssa-loop-ivopts.cc: ...here.
33130 * tree-ssa-loop-manip.c: Moved to...
33131 * tree-ssa-loop-manip.cc: ...here.
33132 * tree-ssa-loop-niter.c: Moved to...
33133 * tree-ssa-loop-niter.cc: ...here.
33134 * tree-ssa-loop-prefetch.c: Moved to...
33135 * tree-ssa-loop-prefetch.cc: ...here.
33136 * tree-ssa-loop-split.c: Moved to...
33137 * tree-ssa-loop-split.cc: ...here.
33138 * tree-ssa-loop-unswitch.c: Moved to...
33139 * tree-ssa-loop-unswitch.cc: ...here.
33140 * tree-ssa-loop.c: Moved to...
33141 * tree-ssa-loop.cc: ...here.
33142 * tree-ssa-math-opts.c: Moved to...
33143 * tree-ssa-math-opts.cc: ...here.
33144 * tree-ssa-operands.c: Moved to...
33145 * tree-ssa-operands.cc: ...here.
33146 * tree-ssa-phiopt.c: Moved to...
33147 * tree-ssa-phiopt.cc: ...here.
33148 * tree-ssa-phiprop.c: Moved to...
33149 * tree-ssa-phiprop.cc: ...here.
33150 * tree-ssa-pre.c: Moved to...
33151 * tree-ssa-pre.cc: ...here.
33152 * tree-ssa-propagate.c: Moved to...
33153 * tree-ssa-propagate.cc: ...here.
33154 * tree-ssa-reassoc.c: Moved to...
33155 * tree-ssa-reassoc.cc: ...here.
33156 * tree-ssa-sccvn.c: Moved to...
33157 * tree-ssa-sccvn.cc: ...here.
33158 * tree-ssa-scopedtables.c: Moved to...
33159 * tree-ssa-scopedtables.cc: ...here.
33160 * tree-ssa-sink.c: Moved to...
33161 * tree-ssa-sink.cc: ...here.
33162 * tree-ssa-strlen.c: Moved to...
33163 * tree-ssa-strlen.cc: ...here.
33164 * tree-ssa-structalias.c: Moved to...
33165 * tree-ssa-structalias.cc: ...here.
33166 * tree-ssa-tail-merge.c: Moved to...
33167 * tree-ssa-tail-merge.cc: ...here.
33168 * tree-ssa-ter.c: Moved to...
33169 * tree-ssa-ter.cc: ...here.
33170 * tree-ssa-threadbackward.c: Moved to...
33171 * tree-ssa-threadbackward.cc: ...here.
33172 * tree-ssa-threadedge.c: Moved to...
33173 * tree-ssa-threadedge.cc: ...here.
33174 * tree-ssa-threadupdate.c: Moved to...
33175 * tree-ssa-threadupdate.cc: ...here.
33176 * tree-ssa-uncprop.c: Moved to...
33177 * tree-ssa-uncprop.cc: ...here.
33178 * tree-ssa-uninit.c: Moved to...
33179 * tree-ssa-uninit.cc: ...here.
33180 * tree-ssa.c: Moved to...
33181 * tree-ssa.cc: ...here.
33182 * tree-ssanames.c: Moved to...
33183 * tree-ssanames.cc: ...here.
33184 * tree-stdarg.c: Moved to...
33185 * tree-stdarg.cc: ...here.
33186 * tree-streamer-in.c: Moved to...
33187 * tree-streamer-in.cc: ...here.
33188 * tree-streamer-out.c: Moved to...
33189 * tree-streamer-out.cc: ...here.
33190 * tree-streamer.c: Moved to...
33191 * tree-streamer.cc: ...here.
33192 * tree-switch-conversion.c: Moved to...
33193 * tree-switch-conversion.cc: ...here.
33194 * tree-tailcall.c: Moved to...
33195 * tree-tailcall.cc: ...here.
33196 * tree-vect-data-refs.c: Moved to...
33197 * tree-vect-data-refs.cc: ...here.
33198 * tree-vect-generic.c: Moved to...
33199 * tree-vect-generic.cc: ...here.
33200 * tree-vect-loop-manip.c: Moved to...
33201 * tree-vect-loop-manip.cc: ...here.
33202 * tree-vect-loop.c: Moved to...
33203 * tree-vect-loop.cc: ...here.
33204 * tree-vect-patterns.c: Moved to...
33205 * tree-vect-patterns.cc: ...here.
33206 * tree-vect-slp-patterns.c: Moved to...
33207 * tree-vect-slp-patterns.cc: ...here.
33208 * tree-vect-slp.c: Moved to...
33209 * tree-vect-slp.cc: ...here.
33210 * tree-vect-stmts.c: Moved to...
33211 * tree-vect-stmts.cc: ...here.
33212 * tree-vector-builder.c: Moved to...
33213 * tree-vector-builder.cc: ...here.
33214 * tree-vectorizer.c: Moved to...
33215 * tree-vectorizer.cc: ...here.
33216 * tree-vrp.c: Moved to...
33217 * tree-vrp.cc: ...here.
33218 * tree.c: Moved to...
33219 * tree.cc: ...here.
33220 * tsan.c: Moved to...
33221 * tsan.cc: ...here.
33222 * typed-splay-tree.c: Moved to...
33223 * typed-splay-tree.cc: ...here.
33224 * ubsan.c: Moved to...
33225 * ubsan.cc: ...here.
33226 * valtrack.c: Moved to...
33227 * valtrack.cc: ...here.
33228 * value-prof.c: Moved to...
33229 * value-prof.cc: ...here.
33230 * var-tracking.c: Moved to...
33231 * var-tracking.cc: ...here.
33232 * varasm.c: Moved to...
33233 * varasm.cc: ...here.
33234 * varpool.c: Moved to...
33235 * varpool.cc: ...here.
33236 * vec-perm-indices.c: Moved to...
33237 * vec-perm-indices.cc: ...here.
33238 * vec.c: Moved to...
33240 * vmsdbgout.c: Moved to...
33241 * vmsdbgout.cc: ...here.
33242 * vr-values.c: Moved to...
33243 * vr-values.cc: ...here.
33244 * vtable-verify.c: Moved to...
33245 * vtable-verify.cc: ...here.
33246 * web.c: Moved to...
33248 * xcoffout.c: Moved to...
33249 * xcoffout.cc: ...here.
33251 2022-01-17 qing zhao <qing.zhao@oracle.com>
33253 * tree-ssa-uninit.c (warn_uninit): Delete the 4th parameter. Handle
33254 .DEFERRED_INIT call with an anonymous SSA_NAME specially.
33255 (check_defs): Handle .DEFERRED_INIT call with an anonymous SSA_NAME
33257 (warn_uninit_phi_uses): Delete the 4th actual when call warn_uninit.
33258 (warn_uninitialized_vars): Likewise.
33259 (warn_uninitialized_phi): Likewise.
33261 2022-01-17 Jason Merrill <jason@redhat.com>
33263 * diagnostic.h (struct diagnostic_context): Add includes_seen.
33264 * diagnostic.c (diagnostic_initialize): Initialize it.
33265 (diagnostic_finish): Clean it up.
33266 (includes_seen): New function.
33267 (diagnostic_report_current_module): Use it.
33269 2022-01-17 Richard Biener <rguenther@suse.de>
33271 PR middle-end/101292
33272 * diagnostic-spec.c (copy_warning): Make sure to not
33273 reference old hashtable content on possible resize.
33274 * warning-control.cc (copy_warning): Likewise.
33276 2022-01-17 Jakub Jelinek <jakub@redhat.com>
33279 * tree-cfg.h (cond_only_block_p): Declare.
33280 * tree-ssa-phiopt.c (cond_only_block_p): Move function to ...
33281 * tree-cfg.c (cond_only_block_p): ... here. No longer static.
33282 * optabs.def (spaceship_optab): New optab.
33283 * internal-fn.def (SPACESHIP): New internal function.
33284 * internal-fn.h (expand_SPACESHIP): Declare.
33285 * internal-fn.c (expand_PHI): Formatting fix.
33286 (expand_SPACESHIP): New function.
33287 * tree-ssa-math-opts.c (optimize_spaceship): New function.
33288 (math_opts_dom_walker::after_dom_children): Use it.
33289 * config/i386/i386.md (spaceship<mode>3): New define_expand.
33290 * config/i386/i386-protos.h (ix86_expand_fp_spaceship): Declare.
33291 * config/i386/i386-expand.c (ix86_expand_fp_spaceship): New function.
33292 * doc/md.texi (spaceship@var{m}3): Document.
33294 2022-01-17 Kewen Lin <linkw@linux.ibm.com>
33296 * config/rs6000/altivec.md (altivec_vreveti2): Remove.
33297 * config/rs6000/vsx.md (*vsx_extract_si, *vsx_extract_si_<uns>float_df,
33298 *vsx_extract_si_<uns>float_<mode>, *vsx_insert_extract_v4sf_p9): Use
33299 known constant values to simplify code.
33301 2022-01-17 Haochen Gui <guihaoc@gcc.gnu.org>
33304 * config/rs6000/vsx.md (split pattern for TI to V1TI move): Defined.
33306 2022-01-16 wwwhhhyyy <hongyu.wang@intel.com>
33308 * config/i386/i386.h (TARGET_DEST_FALSE_DEP_FOR_GLC): New macro.
33309 * config/i386/sse.md (<avx512>_<complexopname>_<mode><maskc_name><round_name>):
33310 Insert zero-idiom in output template when attr enabled, set new attribute to
33311 true for non-mask/maskz insn.
33312 (avx512fp16_<complexopname>sh_v8hf<mask_scalarc_name><round_scalarcz_name>):
33314 (avx512dq_mul<mode>3<mask_name>): Likewise.
33315 (<avx2_avx512>_permvar<mode><mask_name>): Likewise.
33316 (avx2_perm<mode>_1<mask_name>): Likewise.
33317 (avx512f_perm<mode>_1<mask_name>): Likewise.
33318 (avx512dq_rangep<mode><mask_name><round_saeonly_name>): Likewise.
33319 (avx512dq_ranges<mode><mask_scalar_name><round_saeonly_scalar_name>):
33321 (<avx512>_getmant<mode><mask_name><round_saeonly_name>): Likewise.
33322 (avx512f_vgetmant<mode><mask_scalar_name><round_saeonly_scalar_name>):
33324 * config/i386/subst.md (mask3_dest_false_dep_for_glc_cond): New
33326 (mask4_dest_false_dep_for_glc_cond): Likewise.
33327 (mask6_dest_false_dep_for_glc_cond): Likewise.
33328 (mask10_dest_false_dep_for_glc_cond): Likewise.
33329 (maskc_dest_false_dep_for_glc_cond): Likewise.
33330 (mask_scalar4_dest_false_dep_for_glc_cond): Likewise.
33331 (mask_scalarc_dest_false_dep_for_glc_cond): Likewise.
33332 * config/i386/x86-tune.def (X86_TUNE_DEST_FALSE_DEP_FOR_GLC): New
33333 DEF_TUNE enabled for m_SAPPHIRERAPIDS and m_ALDERLAKE
33335 2022-01-15 Martin Sebor <msebor@redhat.com>
33338 * diagnostic-spec.c (nowarn_spec_t::nowarn_spec_t): Handle
33339 -Wdangling-pointer.
33340 * doc/invoke.texi (-Wdangling-pointer): Document new option.
33341 * gimple-ssa-warn-access.cc (pass_waccess::clone): Set new member.
33342 (pass_waccess::check_pointer_uses): New function.
33343 (pass_waccess::gimple_call_return_arg): New function.
33344 (pass_waccess::gimple_call_return_arg_ref): New function.
33345 (pass_waccess::check_call_dangling): New function.
33346 (pass_waccess::check_dangling_uses): New function overloads.
33347 (pass_waccess::check_dangling_stores): New function.
33348 (pass_waccess::check_dangling_stores): New function.
33349 (pass_waccess::m_clobbers): New data member.
33350 (pass_waccess::m_func): New data member.
33351 (pass_waccess::m_run_number): New data member.
33352 (pass_waccess::m_check_dangling_p): New data member.
33353 (pass_waccess::check_alloca): Check m_early_checks_p.
33354 (pass_waccess::check_alloc_size_call): Same.
33355 (pass_waccess::check_strcat): Same.
33356 (pass_waccess::check_strncat): Same.
33357 (pass_waccess::check_stxcpy): Same.
33358 (pass_waccess::check_stxncpy): Same.
33359 (pass_waccess::check_strncmp): Same.
33360 (pass_waccess::check_memop_access): Same.
33361 (pass_waccess::check_read_access): Same.
33362 (pass_waccess::check_builtin): Call check_pointer_uses.
33363 (pass_waccess::warn_invalid_pointer): Add arguments.
33364 (is_auto_decl): New function.
33365 (pass_waccess::check_stmt): New function.
33366 (pass_waccess::check_block): Call check_stmt.
33367 (pass_waccess::execute): Call check_dangling_uses,
33368 check_dangling_stores. Empty m_clobbers.
33369 * passes.def (pass_warn_access): Invoke pass two more times.
33371 2022-01-15 Martin Sebor <msebor@redhat.com>
33373 PR tree-optimization/80532
33374 * common.opt (-Wuse-after-free): New options.
33375 * diagnostic-spec.c (nowarn_spec_t::nowarn_spec_t): Handle
33376 OPT_Wreturn_local_addr and OPT_Wuse_after_free_.
33377 * diagnostic-spec.h (NW_DANGLING): New enumerator.
33378 * doc/invoke.texi (-Wuse-after-free): Document new option.
33379 * gimple-ssa-warn-access.cc (pass_waccess::check_call): Rename...
33380 (pass_waccess::check_call_access): ...to this.
33381 (pass_waccess::check): Rename...
33382 (pass_waccess::check_block): ...to this.
33383 (pass_waccess::check_pointer_uses): New function.
33384 (pass_waccess::gimple_call_return_arg): New function.
33385 (pass_waccess::warn_invalid_pointer): New function.
33386 (pass_waccess::check_builtin): Handle free and realloc.
33387 (gimple_use_after_inval_p): New function.
33388 (get_realloc_lhs): New function.
33389 (maybe_warn_mismatched_realloc): New function.
33390 (pointers_related_p): New function.
33391 (pass_waccess::check_call): Call check_pointer_uses.
33392 (pass_waccess::execute): Compute and free dominance info.
33394 2022-01-15 Uroš Bizjak <ubizjak@gmail.com>
33396 * config/i386/i386-expand.c (ix86_expand_sse_movcc): Use
33397 expand_simple_unop and expand_simple_binop instead of manually
33398 constructing NOT, AND and IOR RTXes. Use vector_all_ones_operand
33399 consistently. Eliminate common subexpressions and simplify code.
33400 * config/i386/sse.md (<any_logic:code><MODEF:mode>3): New expander.
33401 (<any_logic:code><MODEF:mode>3): Make public.
33403 2022-01-14 Eric Botcazou <ebotcazou@adacore.com>
33405 * ipa-param-manipulation.c (ipa_dump_adjusted_parameters): Dump
33406 reverse flag as "reverse" for the sake of consistency.
33407 * ipa-sra.c: Fix copyright year.
33408 (ipa_sra_function_summaries::duplicate): Copy the reverse flag.
33409 (dump_isra_access): Tweak dump line.
33410 (isra_write_node_summary): Write the reverse flag.
33411 (isra_read_node_info): Read it.
33412 (pull_accesses_from_callee): Test its consistency and copy it.
33414 2022-01-14 Richard Sandiford <richard.sandiford@arm.com>
33416 PR middle-end/104026
33417 * tree-vect-loop.c (_loop_vec_info::_loop_vec_info): Initialize
33418 partial_load_store_bias.
33420 2022-01-14 Martin Sebor <msebor@redhat.com>
33422 PR middle-end/101475
33423 * pointer-query.cc (handle_component_ref): Use the size of
33424 the enclosing object if it's smaller than the member.
33426 2022-01-14 Martin Liska <mliska@suse.cz>
33428 * configure: Regenerate.
33430 2022-01-14 Uroš Bizjak <ubizjak@gmail.com>
33432 * config/i386/i386.md (*add<mode>_1_slp"):
33433 Mark alternative 1 output operand earlyclobbered.
33434 (*sub<mode>_1_slp): Ditto.
33435 (*and<mode>_1_slp): Ditto.
33436 (*<code><mode>_1_slp): Ditto.
33437 (*neg<mode>_1_slp): Ditto.
33438 (*one_cmpl<mode>_1_slp): Ditto.
33439 (*ashl<mode>3_1_slp): Ditto.
33440 (*<insn><mode>3_1_slp): Ditto.
33441 (*<insn><mode>3_1_slp): Ditto.
33443 2022-01-14 Kewen Lin <linkw@linux.ibm.com>
33445 PR tree-optimization/104015
33446 * tree-vect-loop.c (vect_analyze_loop): Check
33447 param_vect_partial_vector_usage for supports_partial_vectors.
33449 2022-01-14 Jakub Jelinek <jakub@redhat.com>
33452 * fold-const.c (address_compare): Punt on comparison of address of
33453 one object with address of end of another object if
33454 folding_initializer.
33456 2022-01-14 Jakub Jelinek <jakub@redhat.com>
33459 * tree-ssa-forwprop.c (simplify_builtin_call): Canonicalize
33460 __atomic_fetch_op (p, x, y) op x into __atomic_op_fetch (p, x, y)
33461 and __atomic_op_fetch (p, x, y) iop x into
33462 __atomic_fetch_op (p, x, y).
33464 2022-01-14 Claudiu Zissulescu <claziss@synopsys.com>
33466 * config/arc/arc.h (DWARF_FRAME_REGNUM): Update definition.
33467 (DWARF_FRAME_RETURN_COLUMN): Use RETURN_ADDR_REGNUM macro.
33468 (INCOMING_RETURN_ADDR_RTX): Likewise.
33469 (DWARF_ALT_FRAME_RETURN_COLUMN): Define.
33471 2022-01-14 Claudiu Zissulescu <claziss@synopsys.com>
33473 * config/arc/arc.c (arc_compute_frame_size): Remove condition when
33474 computin checking accumulator regs.
33475 (arc_expand_prologue): Update comments.
33476 (arc_expand_epilogue): Likewise.
33478 2022-01-14 Roger Sayle <roger@nextmovesoftware.com>
33479 Uroš Bizjak <ubizjak@gmail.com>
33481 * config/i386/i386-expand.c (ix86_expand_v1ti_to_ti): Use force_reg.
33482 (ix86_expand_ti_to_v1ti): Use force_reg.
33483 (ix86_expand_v1ti_shift): Use force_reg.
33484 (ix86_expand_v1ti_rotate): Use force_reg.
33485 (ix86_expand_v1ti_ashiftrt): Provide new three operation
33486 implementations for shifts by 111..126 bits. Use force_reg.
33488 2022-01-14 Martin Liska <mliska@suse.cz>
33490 * common/config/arm/arm-common.c (arm_target_mode): Fix
33491 warning: unterminated quoting directive [-Wformat=].
33493 2022-01-14 Siddhesh Poyarekar <siddhesh@gotplt.org>
33495 PR tree-optimization/104009
33496 * tree-object-size.c (compute_builtin_object_size): Bail out on
33498 (plus_stmt_object_size): Return maximum of wholesize and minimum
33499 of 0 for negative offset.
33501 2022-01-14 liuhongt <hongtao.liu@intel.com>
33506 * config/i386/i386.md (*xor2andn): Refine predicate of
33507 operands[0] from nonimmediate_operand to
33508 register_operand, remove TARGET_AVX512BW from condition.
33510 2022-01-14 David Malcolm <dmalcolm@redhat.com>
33512 * doc/extend.texi (Function Attributes): Note that "tainted_args" can
33513 be used on field decls.
33514 (Common Function Attributes): Add entry on "tainted_args" attribute.
33516 2022-01-13 Anthony Sharp <anthonysharp15@gmail.com>
33517 Jason Merrill <jason@redhat.com>
33520 * doc/invoke.texi: Documentation for Wmissing-template-keyword.
33522 2022-01-13 Uroš Bizjak <ubizjak@gmail.com>
33525 * config/i386/i386.md (*ashlqi_ext<mode>_2): New insn pattern.
33526 (*<any_shiftrt:insn>qi_ext<mode>_2): Ditto.
33527 * config/i386/mmx.md (<any_shift:insn>v2qi):
33528 New insn_and_split pattern.
33530 2022-01-13 Robin Dapp <rdapp@linux.ibm.com>
33532 * internal-fn.c (expand_partial_load_optab_fn): Add bias.
33533 (expand_partial_store_optab_fn): Likewise.
33534 (internal_len_load_store_bias): New function.
33535 * internal-fn.h (VECT_PARTIAL_BIAS_UNSUPPORTED): New define.
33536 (internal_len_load_store_bias): New function.
33537 * tree-vect-loop-manip.c (vect_set_loop_controls_directly): Set bias.
33538 (vect_set_loop_condition_partial_vectors): Add header_seq parameter.
33539 * tree-vect-loop.c (vect_verify_loop_lens): Verify bias.
33540 (vect_estimate_min_profitable_iters): Account for bias.
33541 (vect_get_loop_len): Add bias-adjusted length.
33542 * tree-vect-stmts.c (vectorizable_store): Use.
33543 (vectorizable_load): Use.
33544 * tree-vectorizer.h (struct rgroup_controls): Add bias-adjusted length.
33545 (LOOP_VINFO_PARTIAL_LOAD_STORE_BIAS): New macro.
33546 * config/rs6000/vsx.md: Use const0 bias predicate.
33547 * doc/md.texi: Document bias value.
33549 2022-01-13 Andrew MacLeod <amacleod@redhat.com>
33551 PR tree-optimization/83072
33552 PR tree-optimization/83073
33553 PR tree-optimization/97909
33554 * fold-const.c (expr_not_equal_to): Use a multi-range class.
33556 2022-01-13 Andrew MacLeod <amacleod@redhat.com>
33558 PR tree-optimization/96707
33559 * range-op.cc (operator_rshift::lhs_op1_relation): New.
33561 2022-01-13 Uroš Bizjak <ubizjak@gmail.com>
33563 * config/i386/mmx.md (negv2qi): Disparage GPR alternative a bit.
33564 Disable for TARGET_PARTIAL_REG_STALL unless optimizing for size.
33565 (negv2qi splitters): Use lowpart_subreg instead of
33566 gen_lowpart to create subreg.
33567 (<plusminus:insn>v2qi3): Disparage GPR alternative a bit.
33568 Disable for TARGET_PARTIAL_REG_STALL unless optimizing for size.
33569 (<plusminus:insn>v2qi3 splitters): Use lowpart_subreg instead of
33570 gen_lowpart to create subreg.
33571 * config/i386/i386.md (*subqi_ext<mode>_2): Move.
33573 2022-01-13 Uroš Bizjak <ubizjak@gmail.com>
33576 * config/i386/mmx.md (*xop_pcmov_<mode>): Use VI_16_32 mode iterator.
33578 2022-01-13 Martin Liska <mliska@suse.cz>
33580 * common/config/arm/arm-common.c (arm_target_mode): Wrap
33581 keywords with %<, %> and remove trailing punctuation char.
33582 (arm_canon_arch_option_1): Likewise.
33583 (arm_asm_auto_mfpu): Likewise.
33584 * config/arm/arm-builtins.c (arm_expand_builtin): Likewise.
33585 * config/arm/arm.c (arm_options_perform_arch_sanity_checks): Likewise.
33586 (use_vfp_abi): Likewise.
33587 (aapcs_vfp_is_call_or_return_candidate): Likewise.
33588 (arm_handle_cmse_nonsecure_entry): Likewise.
33589 (arm_handle_cmse_nonsecure_call): Likewise.
33590 (thumb1_md_asm_adjust): Likewise.
33592 2022-01-13 Paul A. Clarke <pc@us.ibm.com>
33594 * config/rs6000/smmintrin.h (_mm_round_pd, _mm_round_ps,
33595 _mm_round_sd, _mm_round_ss, _MM_FROUND_TO_NEAREST_INT,
33596 _MM_FROUND_TO_ZERO, _MM_FROUND_TO_POS_INF, _MM_FROUND_TO_NEG_INF,
33597 _MM_FROUND_CUR_DIRECTION, _MM_FROUND_RAISE_EXC, _MM_FROUND_NO_EXC,
33598 _MM_FROUND_NINT, _MM_FROUND_FLOOR, _MM_FROUND_CEIL, _MM_FROUND_TRUNC,
33599 _MM_FROUND_RINT, _MM_FROUND_NEARBYINT): New.
33600 (_mm_ceil_pd, _mm_ceil_ps, _mm_ceil_sd, _mm_ceil_ss, _mm_floor_pd,
33601 _mm_floor_ps, _mm_floor_sd, _mm_floor_ss): Convert from function to
33604 2022-01-13 Jakub Jelinek <jakub@redhat.com>
33606 PR tree-optimization/103989
33607 * tree-inline.c (setup_one_parameter): Don't copy parms with
33610 2022-01-13 Thomas Schwinge <thomas@codesourcery.com>
33612 * tree-streamer-out.c (pack_ts_base_value_fields): Don't pack
33613 'TYPE_ADDR_SPACE' for offloading.
33614 * tree-streamer-in.c (unpack_ts_base_value_fields): Don't unpack
33615 'TYPE_ADDR_SPACE' for offloading.
33617 2022-01-13 Julian Brown <julian@codesourcery.com>
33618 Thomas Schwinge <thomas@codesourcery.com>
33620 * omp-oacc-kernels-decompose.cc (add_wait): New function, split out
33622 (add_async_clauses_and_wait): ...here. Call new outlined function.
33623 (decompose_kernels_region_body): Add wait at the end of
33624 explicitly-asynchronous kernels regions.
33626 2022-01-13 Thomas Schwinge <thomas@codesourcery.com>
33628 PR middle-end/100280
33629 * omp-oacc-kernels-decompose.cc (maybe_build_inner_data_region):
33630 Mark variables used in synthesized data clauses as addressable.
33632 2022-01-13 Martin Liska <mliska@suse.cz>
33634 * config/epiphany/epiphany.c (epiphany_mode_priority):
33635 Use gcc_unreachable for not handled cases.
33637 2022-01-13 Martin Liska <mliska@suse.cz>
33639 * config/epiphany/epiphany.c (epiphany_handle_interrupt_attribute):
33640 Use %qs format specifier.
33641 (epiphany_override_options): Wrap keyword in %<, %>.
33643 2022-01-13 Haochen Jiang <haochen.jiang@intel.com>
33646 * config/i386/i386.md (*xor2andn): New define_insn_and_split.
33648 2022-01-13 Xionghu Luo <luoxhu@linux.ibm.com>
33650 * config/rs6000/altivec.md (sldoi_to_mov<mode>): New.
33652 2022-01-12 Uroš Bizjak <ubizjak@gmail.com>
33656 * config/i386/i386-expand.c (ix86_emit_vec_binop): New static function.
33657 (ix86_expand_sse_movcc): Use ix86_emit_vec_binop instead of gen_rtx_X
33658 when constructing vector logic RTXes.
33659 (expand_vec_perm_pshufb2): Ditto.
33660 * config/i386/mmx.md (negv2qi): Disparage GPR alternative a bit.
33661 (<plusminus:insn>v2qi3): Ditto.
33662 (vcond<mode><mode>): Re-enable for TARGET_SSE2.
33663 (vcondu<mode><mode>): Ditto.
33664 (vcond_mask_<mode><mode>): Ditto.
33665 (one_cmpl<VI_32:mode>2): Remove expander.
33666 (one_cmpl<VI_16_32:mode>2): Rename from one_cmplv2qi.
33667 Use VI_16_32 mode iterator.
33668 (one_cmpl<VI_16_32:mode>2 splitters): Use VI_16_32 mode iterator.
33669 Use lowpart_subreg instead of gen_lowpart to create subreg.
33670 (*andnot<VI_16_32:mode>3): Merge from "*andnot<VI_32:mode>" and
33671 "*andnotv2qi3" insn patterns using VI_16_32 mode iterator.
33672 Disparage GPR alternative a bit. Add CC clobber.
33673 (*andnot<VI_16_32:mode>3 splitters): Use VI_16_32 mode iterator.
33674 Use lowpart_subreg instead of gen_lowpart to create subreg.
33675 (*<any_logic:code><VI_16_32:mode>3): Merge from
33676 "*<any_logic:code><VI_32:mode>" and "*<any_logic:code>v2qi3" insn patterns
33677 using VI_16_32 mode iterator. Disparage GPR alternative a bit.
33679 (*<any_logic:code><VI_16_32:mode>3 splitters):Use VI_16_32 mode
33680 iterator. Use lowpart_subreg instead of gen_lowpart to create subreg.
33682 2022-01-12 Clément Chigot <clement.chigot@atos.net>
33684 * configure.ac: Check sizeof ino_t and dev_t.
33685 (HOST_STAT_FOR_64BIT_INODES): New AC_DEFINE to provide stat
33686 syscall being able to handle 64bit inodes.
33687 * config.in: Regenerate.
33688 * configure: Regenerate.
33689 * incpath.c (HOST_STAT_FOR_64BIT_INODES): New define.
33690 (remove_duplicates): Use it.
33692 2022-01-12 Andrew MacLeod <amacleod@redhat.com>
33694 PR tree-optimization/103551
33695 * tree-vrp.c (execute_ranger_vrp): Always set EDGE_EXECUTABLE.
33697 2022-01-12 Richard Biener <rguenther@suse.de>
33699 PR tree-optimization/103990
33700 * tree-pass.h (tail_merge_optimize): Drop unused argument.
33701 * tree-ssa-tail-merge.c (tail_merge_optimize): Likewise.
33702 * tree-ssa-pre.c (pass_pre::execute): Retain TODO_cleanup_cfg
33703 and adjust call to tail_merge_optimize.
33705 2022-01-12 Andre Vieira <andre.simoesdiasvieira@arm.com>
33707 * tree-vect-loop.c (vect-analyze-loop): Handle scenario where target
33708 does not add autovectorize_vector_modes.
33710 2022-01-12 Martin Liska <mliska@suse.cz>
33712 * config/aarch64/aarch64.c (aarch64_parse_boolean_options): Use
33713 %qs where possible.
33714 (aarch64_parse_sve_width_string): Likewise.
33715 (aarch64_override_options_internal): Likewise.
33716 (aarch64_print_hint_for_extensions): Likewise.
33717 (aarch64_validate_sls_mitigation): Likewise.
33718 (aarch64_handle_attr_arch): Likewise.
33719 (aarch64_handle_attr_cpu): Likewise.
33720 (aarch64_handle_attr_tune): Likewise.
33721 (aarch64_handle_attr_isa_flags): Likewise.
33723 2022-01-12 Martin Liska <mliska@suse.cz>
33725 * config.gcc: Include elfos.h before ${tm_file}.
33727 2022-01-12 Hans-Peter Nilsson <hp@axis.com>
33729 * config/cris/cris.c: Quote identifiers in parameters to error
33730 and internal_error, and remove extraneous spaces with punctuation.
33731 * config/cris/cris.h (CRIS_ASSERT): When passing on stringified
33732 expression to internal_error, pass it as a parameter instead of
33733 appending it to the format part.
33735 2022-01-12 Hans-Peter Nilsson <hp@axis.com>
33737 * config/cris/cris.c (cris_postdbr_cmpelim): Parenthesize
33740 2022-01-11 qing zhao <qing.zhao@oracle.com>
33742 * gimplify.c (gimple_add_init_for_auto_var): Delete the 3rd argument.
33743 Change the 3rd argument of function .DEFERRED_INIT to the name of the
33745 (gimplify_decl_expr): Delete the 3rd argument when call
33746 gimple_add_init_for_auto_var.
33747 * internal-fn.c (expand_DEFERRED_INIT): Update comments to reflect
33748 the 3rd argument change of function .DEFERRED_INIT.
33749 * tree-cfg.c (verify_gimple_call): Update comments and verification
33750 to reflect the 3rd argument change of function .DEFERRED_INIT.
33751 * tree-sra.c (generate_subtree_deferred_init): Delete the 3rd argument.
33752 (sra_modify_deferred_init): Change the 3rd argument of function
33753 .DEFERRED_INIT to the name of the decl.
33755 2022-01-11 Thomas Koenig <tkoenig@gcc.gnu.org>
33757 * flag-types.h (enum gfc_convert): Add flags for
33760 2022-01-11 Michael Meissner <meissner@the-meissners.org>
33762 * config/rs6000/rs6000.c (rs6000_option_override_internal): Remove
33763 checks for only C/C++ front ends before allowing the long double
33764 format to change without a warning.
33766 2022-01-11 Richard Sandiford <richard.sandiford@arm.com>
33768 PR rtl-optimization/103974
33769 * ira-int.h (ira_subloop_allocnos_can_differ_p): Take an
33770 extra argument, default true, that says whether old-reload
33771 targets should be excluded.
33772 * ira-color.c (color_pass): Pass false.
33774 2022-01-11 Uroš Bizjak <ubizjak@gmail.com>
33777 * config/i386/mmx.md (vcond<mode><mode>):
33778 Use VI_16_32 mode iterator. Enable for TARGET_SSE4_1.
33779 (vcondu<mode><mode>): Ditto.
33780 (vcond_mask_<mode><mode>): Ditto.
33781 (mmx_pblendvb_v8qi): Rename from mmx_pblendvb64.
33782 (mmx_pblendvb_<mode>): Rename from mmx_pblendvb32.
33783 Use VI_16_32 mode iterator.
33784 * config/i386/i386-expand.c (ix86_expand_sse_movcc):
33785 Update for rename. Handle V2QImode.
33786 (expand_vec_perm_blend): Update for rename.
33788 2022-01-11 Jakub Jelinek <jakub@redhat.com>
33791 * tree.def (OBJ_TYPE_REF): Document type of OBJ_TYPE_REF_TOKEN.
33793 2022-01-11 Siddhesh Poyarekar <siddhesh@gotplt.org>
33795 PR middle-end/70090
33796 * tree-object-size.c (size_valid_p): New function.
33797 (size_for_offset): Remove OFFSET constness assertion.
33798 (addr_object_size): Build dynamic expressions for object
33799 sizes and use size_valid_p to decide if it is valid for the
33800 given OBJECT_SIZE_TYPE.
33801 (compute_builtin_object_size): Allow dynamic offsets when
33802 computing size at O0.
33803 (call_object_size): Call size_valid_p.
33804 (plus_stmt_object_size): Allow non-constant offset and use
33805 size_valid_p to decide if it is valid for the given
33808 2022-01-11 Siddhesh Poyarekar <siddhesh@gotplt.org>
33810 PR middle-end/70090
33811 * tree-object-size.c (alloc_object_size): Make and return
33812 non-constant size expression.
33813 (call_object_size): Return expression or unknown based on
33814 whether dynamic object size is requested.
33816 2022-01-11 Siddhesh Poyarekar <siddhesh@gotplt.org>
33818 PR middle-end/70090
33819 * tree-object-size.c: Include tree-dfa.h.
33820 (parm_object_size): New function.
33821 (collect_object_sizes_for): Call it.
33823 2022-01-11 Siddhesh Poyarekar <siddhesh@gotplt.org>
33825 PR middle-end/70090
33826 * builtins.c (fold_builtin_object_size): Adjust for dynamic size
33828 * tree-object-size.c: Include gimplify-me.h.
33829 (struct object_size_info): New member UNKNOWNS.
33830 (size_initval_p, size_usable_p, object_sizes_get_raw): New
33832 (object_sizes_get): Return suitable gimple variable for
33834 (bundle_sizes): New function.
33835 (object_sizes_set): Use it and handle dynamic object size
33837 (object_sizes_set_temp): New function.
33838 (size_for_offset): Adjust for dynamic size expressions.
33839 (emit_phi_nodes, propagate_unknowns, gimplify_size_expressions):
33841 (compute_builtin_object_size): Call gimplify_size_expressions
33843 (dynamic_object_size): New function.
33844 (cond_expr_object_size): Use it.
33845 (phi_dynamic_object_size): New function.
33846 (collect_object_sizes_for): Call it for OST_DYNAMIC. Adjust to
33847 accommodate dynamic object sizes.
33849 2022-01-11 Siddhesh Poyarekar <siddhesh@gotplt.org>
33850 Jakub Jelinek <jakub@redhat.com>
33852 PR tree-optimization/103961
33853 * tree-object-size.c (plus_stmt_object_size): Always avoid
33854 computing offset for -1 size.
33856 2022-01-11 Andrew MacLeod <amacleod@redhat.com>
33858 PR tree-optimization/103821
33859 * range-op.cc (range_operator::fold_range): Only do precise ranges
33860 when there are not too many subranges.
33862 2022-01-11 Richard Sandiford <richard.sandiford@arm.com>
33864 * ira-costs.c (ira_tune_allocno_costs): Fix missing rclass
33865 definition in IRA_HARD_REGNO_ADD_COST_MULTIPLIER code.
33867 2022-01-11 Roger Sayle <roger@nextmovesoftware.com>
33868 Richard Biener <rguenther@suse.de>
33870 * tree-ssa-math-opts.c (struct widen_mul_stats): Add a
33871 highpart_mults_inserted field.
33872 (convert_mult_to_highpart): New function to convert right shift
33873 of a widening multiply into a MULT_HIGHPART_EXPR.
33874 (math_opts_dom_walker::after_dom_children) [RSHIFT_EXPR]:
33875 Call new convert_mult_to_highpart function.
33876 (pass_optimize_widening_mul::execute): Add a statistics counter
33877 for tracking "highpart multiplications inserted" events.
33879 2022-01-11 Xionghu Luo <luoxhu@linux.ibm.com>
33882 * config/rs6000/rs6000-protos.h (rs6000_is_valid_rotate_dot_mask): New
33884 * config/rs6000/rs6000.c (rs6000_is_valid_rotate_dot_mask): New
33886 * config/rs6000/rs6000.md (*branch_anddi3_dot): New.
33888 2022-01-11 Olivier Hainque <hainque@adacore.com>
33890 * gcc.c (driver_handle_option): State --sysroot as
33893 2022-01-11 Kewen Lin <linkw@linux.ibm.com>
33895 * config/rs6000/rs6000.c (rs6000_disable_incompatible_switches): Remove
33896 useless related to option -mno-power10.
33898 2022-01-11 Haochen Jiang <haochen.jiang@intel.com>
33901 * config/i386/sse.md (*andnot<mode>3): Extend predicate of
33902 operands[1] from register_operand to vector_operand.
33904 2022-01-10 Uroš Bizjak <ubizjak@gmail.com>
33907 * config/i386/i386-expand.c (ix86_expand_int_sse_cmp):
33909 * config/i386/mmx.md (<sat_plusminus:insn><mode>3):
33910 Use VI1_16_32 mode iterator.
33911 (*eq<mode>3): Ditto.
33912 (*gt<mode>3): Ditto.
33913 (*xop_maskcmp<mode>3): Ditto.
33914 (*xop_maskcmp_uns<mode>3): Ditto.
33915 (vec_cmp<mode><mode>): Ditto.
33916 (vec_cmpu<mode><mode>): Ditto.
33918 2022-01-10 Richard Biener <rguenther@suse.de>
33920 PR tree-optimization/103948
33921 * tree-vect-generic.c (expand_vector_condition): Return true if
33922 all ones vector is returned for true, all zeros vector for false
33923 and the target defines corresponding vec_cmp{,u}MN named RTX pattern.
33925 2022-01-10 Paul A. Clarke <pc@us.ibm.com>
33927 * config/rs6000/smmintrin.h (_mm_blendv_epi8): Use vec_blendv
33928 when _ARCH_PWR10. Use signed types.
33929 (_mm_blendv_ps): Use vec_blendv when _ARCH_PWR10.
33930 (_mm_blendv_pd): Likewise.
33932 2022-01-10 Andre Vieira <andre.simoesdiasvieira@arm.com>
33934 * tree-vectorizer.c (better_epilogue_loop_than_p): Round factors up for
33936 * tree-vect-loop.c (vect_analyze_loop): Re-analyze all modes for
33937 epilogues, unless we are guaranteed that we can't have partial vectors.
33938 * genopinit.c: (partial_vectors_supported): Generate new function.
33940 2022-01-10 Jakub Jelinek <jakub@redhat.com>
33943 * config/i386/i386.c (classify_argument): Add zero_width_bitfields
33944 argument, when seeing DECL_FIELD_CXX_ZERO_WIDTH_BIT_FIELD bitfields,
33945 always ignore them, when seeing other zero sized bitfields, either
33946 set zero_width_bitfields to 1 and ignore it or if equal to 2 process
33947 it. Pass it to recursive calls. Add wrapper
33948 with old arguments and diagnose ABI differences for C structures
33949 with zero width bitfields. Formatting fixes.
33951 2022-01-10 Richard Sandiford <richard.sandiford@arm.com>
33953 PR rtl-optimization/98782
33954 * ira-int.h (ira_soft_conflict): Declare.
33955 * ira-color.c (max_soft_conflict_loop_depth): New constant.
33956 (ira_soft_conflict): New function.
33957 (spill_soft_conflicts): Likewise.
33958 (assign_hard_reg): Use them to handle the case described by
33959 the comment above ira_soft_conflict.
33960 (improve_allocation): Likewise.
33961 * ira.c (check_allocation): Allow allocnos with "soft" conflicts
33962 to share the same register.
33964 2022-01-10 Richard Sandiford <richard.sandiford@arm.com>
33966 PR rtl-optimization/98782
33967 * ira-int.h (ira_caller_save_cost): New function.
33968 (ira_caller_save_loop_spill_p): Likewise.
33969 * ira-build.c (ira_propagate_hard_reg_costs): Test whether it is
33970 cheaper to spill a call-clobbered register throughout a loop rather
33971 than spill it around each individual call. If so, treat all
33972 call-clobbered registers as conflicts and...
33973 (propagate_allocno_info): ...do not propagate call information
33974 from the child to the parent.
33975 * ira-color.c (move_spill_restore): Update accordingly.
33976 * ira-costs.c (ira_tune_allocno_costs): Use ira_caller_save_cost.
33978 2022-01-10 Richard Sandiford <richard.sandiford@arm.com>
33980 PR rtl-optimization/98782
33981 * ira-int.h (ira_allocno::might_conflict_with_parent_p): New field.
33982 (ALLOCNO_MIGHT_CONFLICT_WITH_PARENT_P): New macro.
33983 (ira_single_region_allocno_p): New function.
33984 (ira_total_conflict_hard_regs): Likewise.
33985 * ira-build.c (ira_create_allocno): Initialize
33986 ALLOCNO_MIGHT_CONFLICT_WITH_PARENT_P.
33987 (ira_propagate_hard_reg_costs): New function.
33988 (propagate_allocno_info): Use it. Try to avoid propagating
33989 hard register conflicts to parent allocnos if we can handle
33990 the conflicts by spilling instead. Limit the propagated
33991 register costs to the cost of spilling throughout the child loop.
33992 * ira-color.c (color_pass): Use ira_single_region_allocno_p to
33993 test whether a child and parent allocno can share the same
33995 (move_spill_restore): Adjust for the new behavior of
33996 propagate_allocno_info.
33998 2022-01-10 Richard Sandiford <richard.sandiford@arm.com>
34000 PR rtl-optimization/98782
34001 * ira-int.h (ira_subloop_allocnos_can_differ_p): New function,
34003 * ira-color.c (color_pass): ...here.
34005 2022-01-10 Richard Sandiford <richard.sandiford@arm.com>
34007 PR rtl-optimization/98782
34008 * ira-color.c (color_pass): Add comments to describe the spill costs.
34009 (move_spill_restore): Likewise. Fix reversed calculation.
34011 2022-01-10 Richard Sandiford <richard.sandiford@arm.com>
34013 PR rtl-optimization/98782
34014 * ira-int.h (ira_loop_border_costs): New class.
34015 * ira-color.c (ira_loop_border_costs::ira_loop_border_costs):
34017 (calculate_allocno_spill_cost): Use ira_loop_border_costs.
34018 (color_pass): Likewise.
34019 (move_spill_restore): Likewise.
34021 2022-01-10 Eric Botcazou <ebotcazou@adacore.com>
34024 * coretypes.h (unwind_info_type): Swap UI_SEH and UI_TARGET.
34026 2022-01-10 Richard Biener <rguenther@suse.de>
34028 PR tree-optimization/100359
34029 * tree-ssa-loop-ivcanon.c (try_unroll_loop_completely):
34030 Allow non-growing peeling with !allow_peel and UL_ALL.
34032 2022-01-08 Roger Sayle <roger@nextmovesoftware.com>
34034 * config/i386/i386-expand.c (ix86_expand_vector_move): Add
34035 special case for TImode to V1TImode moves, going via V2DImode.
34037 2022-01-08 Jakub Jelinek <jakub@redhat.com>
34040 * match.pd ((ptr) (x p+ y) p+ z -> (ptr) (x p+ (y + z))): New GENERIC
34043 2022-01-08 David Malcolm <dmalcolm@redhat.com>
34045 * doc/analyzer.texi
34046 (Special Functions for Debugging the Analyzer): Document
34047 __analyzer_dump_escaped.
34049 2022-01-08 David Malcolm <dmalcolm@redhat.com>
34051 * doc/analyzer.texi (Other Debugging Techniques): Document
34052 region::is_named_decl_p.
34054 2022-01-07 Andrew Pinski <apinski@marvell.com>
34057 * config/arm/aarch-common.c (arm_md_asm_adjust):
34058 Use a temp if !REG_P.
34060 2022-01-07 Uroš Bizjak <ubizjak@gmail.com>
34062 * config/i386/mmx.md (*move<V_32:mode>_internal): Add isa attribute.
34063 (*movv2qi_internal): Remve sse2 requirement for alternatives 4,5.
34065 2022-01-07 liuhongt <hongtao.liu@intel.com>
34067 PR rtl-optimization/103750
34068 * fwprop.c (forward_propagate_into): Allow propagations from
34069 inner loop to outer loop.
34071 2022-01-07 Roger Sayle <roger@nextmovesoftware.com>
34073 * config/nvptx/nvptx.md (*cnot<mode>2): New define_insn.
34075 2022-01-07 Haochen Gui <guihaoc@gcc.gnu.org>
34077 * config/rs6000/rs6000.md (rs6000_mffscrni): Define.
34078 (rs6000_set_fpscr_rn): Change the type of operand[0] from DI to SI.
34079 Call gen_rs6000_mffscrni when operand[0] is a const_0_to_3_operand.
34081 2022-01-07 liuhongt <hongtao.liu@intel.com>
34083 * config/i386/sse.md
34084 (*<avx512>_eq<mode>3<mask_scalar_merge_name>_1): Extend to
34085 UNSPEC_PCMP_UNSIGNED.
34087 2022-01-07 liuhongt <hongtao.liu@intel.com>
34090 * config/i386/i386-expand.c (ix86_expand_vector_set): Not use
34091 gen_avx2_pblendph_1 when elt == 0.
34092 * config/i386/sse.md (avx2_pblendph): Rename to ..
34093 (avx2_pblend<ssemodesuffix>_1).. this, and extend to V16HI.
34094 (*avx2_pblendw): Rename to ..
34095 (*avx2_pblend<ssemodesuffix>): .. this, and extend to V16HF.
34096 (avx2_pblendw): Rename to ..
34097 (*avx2_pblend<ssemodesuffix>): .. this, and extend to V16HF.
34098 (blendsuf): Removed.
34099 (sse4_1_pblend<blendsuf>): Renamed to ..
34100 (sse4_1_pblend<ssemodesuffix>): .. this.
34102 2022-01-06 H.J. Lu <hjl.tools@gmail.com>
34105 * config/i386/i386.c (ix86_output_indirect_function_return):
34106 Generate INT3 after indirect jmp for -mharden-sls=indirect-jmp.
34108 2022-01-06 H.J. Lu <hjl.tools@gmail.com>
34111 * config/i386/i386-opts.h (harden_sls): Replace
34112 harden_sls_indirect_branch with harden_sls_indirect_jmp.
34113 * config/i386/i386.c (ix86_output_jmp_thunk_or_indirect):
34115 (ix86_output_indirect_jmp): Likewise.
34116 (ix86_output_call_insn): Likewise.
34117 * config/i386/i386.opt: Replace indirect-branch with
34118 indirect-jmp. Replace harden_sls_indirect_branch with
34119 harden_sls_indirect_jmp.
34120 * doc/invoke.texi (-harden-sls=): Replace indirect-branch with
34123 2022-01-06 Uroš Bizjak <ubizjak@gmail.com>
34125 * config/i386/i386.c (ix86_output_ssemov) <MODE_DI>:
34126 Add %q modifier for operands in general registers.
34127 <MODE_SI>: Add %q modifier for operands in general registers.
34128 * config/i386/i386.md (*movhi_internal): Change type attribute of
34129 xmm-gpr interunit alternatives 9,10 to ssemov and mode attribute
34130 to SImode for non-avx512fp16 targets.
34131 (*movhf_internal): Ditto for xmm-gpr interunit alternatives 6,8.
34132 * config/i386/mmx.md (*movv2qi_internal):
34133 Ditto for xmm-gpr interunit alternatives 8,9.
34135 2022-01-06 Kito Cheng <kito.cheng@sifive.com>
34137 * common/config/riscv/riscv-common.c (riscv_implied_info): Add
34139 (riscv_ext_version_table): Add version info for vector extensions.
34140 (riscv_ext_flag_table): Add option mask for vector extensions.
34141 * config/riscv/riscv-opts.h (MASK_VECTOR_EEW_32): New.
34142 (MASK_VECTOR_EEW_64): New.
34143 (MASK_VECTOR_EEW_FP_32): New.
34144 (MASK_VECTOR_EEW_FP_64): New.
34145 (MASK_ZVL32B): New.
34146 (MASK_ZVL64B): New.
34147 (MASK_ZVL128B): New.
34148 (MASK_ZVL256B): New.
34149 (MASK_ZVL512B): New.
34150 (MASK_ZVL1024B): New.
34151 (MASK_ZVL2048B): New.
34152 (MASK_ZVL4096B): New.
34153 (MASK_ZVL8192B): New.
34154 (MASK_ZVL16384B): New.
34155 (MASK_ZVL32768B): New.
34156 (MASK_ZVL65536B): New.
34157 (TARGET_ZVL32B): New.
34158 (TARGET_ZVL64B): New.
34159 (TARGET_ZVL128B): New.
34160 (TARGET_ZVL256B): New.
34161 (TARGET_ZVL512B): New.
34162 (TARGET_ZVL1024B): New.
34163 (TARGET_ZVL2048B): New.
34164 (TARGET_ZVL4096B): New.
34165 (TARGET_ZVL8192B): New.
34166 (TARGET_ZVL16384B): New.
34167 (TARGET_ZVL32768B): New.
34168 (TARGET_ZVL65536B): New.
34169 * config/riscv/riscv.opt (Mask(VECTOR)): New.
34170 (riscv_vector_eew_flags): New.
34171 (riscv_zvl_flags): New.
34173 2022-01-06 Kito Cheng <kito.cheng@sifive.com>
34175 * common/config/riscv/riscv-common.c
34176 (riscv_subset_list::parse_multiletter_ext): Allow ext. name has
34179 2022-01-06 Jakub Jelinek <jakub@redhat.com>
34181 PR tree-optimization/103899
34182 * expr.c (expand_expr_real_1): Add a workaround for bogus uninit
34183 warning by moving context variable to the only spot where it is used
34184 and moving gcc_assert into if body.
34186 2022-01-06 Jakub Jelinek <jakub@redhat.com>
34188 PR rtl-optimization/103908
34189 * ifcvt.c (bb_valid_for_noce_process_p): Punt on bbs ending with
34192 2022-01-05 Bill Schmidt <wschmidt@linux.ibm.com>
34195 * config/rs6000/rs6000-c.c (altivec_resolve_overloaded_builtin):
34196 Skip over instances with undefined function types.
34198 2022-01-05 Andrew Pinski <apinski@marvell.com>
34201 * config/i386/i386.h (x86_mfence): Mark with GTY.
34203 2022-01-05 Uroš Bizjak <ubizjak@gmail.com>
34206 * config/i386/mmx.md (VI_16_32): New mode iterator.
34207 (VI1_16_32): Ditto.
34208 (mmxvecsize): Handle V2QI mode.
34209 (<smaxmin:code><mode>3): Rename from <smaxmin:code>v4qi3.
34210 Use VI1_16_32 mode iterator.
34211 (<umaxmin:code><mode>3): Rename from <umaxmin:code>v4qi3.
34212 Use VI1_16_32 mode iterator.
34213 (abs<mode>2): Use VI_16_32 mode iterator.
34214 (uavgv2qi3_ceil): New insn pattern.
34216 2022-01-05 Martin Sebor <msebor@redhat.com>
34218 * gimple-ssa-warn-access.cc (pass_waccess::maybe_warn_memmodel): Use
34219 %qs to avoid -Wformat-diag.
34221 2022-01-05 Uroš Bizjak <ubizjak@gmail.com>
34224 * config/i386/mmx.md (one_cmplv2qi2): Change
34225 alternatives 1,2 type from sselog to sselog1.
34227 2022-01-05 Uroš Bizjak <ubizjak@gmail.com>
34230 * config/i386/i386-expand.c (expand_vec_perm_pshufb): Fix number of
34231 narrow mode remapped elements for !one_operand_p case.
34233 2022-01-05 Richard Biener <rguenther@suse.de>
34235 PR tree-optimization/103816
34236 * tree-vect-data-refs.c (vect_analyze_group_access_1): Also
34237 check DR_GROUP_GAP compute for overflow and representability.
34239 2022-01-05 Jakub Jelinek <jakub@redhat.com>
34242 * gimple-fold.c (fold_stmt_1): Don't call maybe_fold_reference
34243 for DEBUG stmts with ADDR_EXPR gimple_debug_bind_get_value,
34244 it can do unwanted rhs folding like &a[0] into &2.0 etc.
34246 2022-01-05 Kewen Lin <linkw@linux.ibm.com>
34249 * config/rs6000/rs6000.c (TARGET_NEED_IPA_FN_TARGET_INFO): New macro.
34250 (TARGET_UPDATE_IPA_FN_TARGET_INFO): Likewise.
34251 (rs6000_need_ipa_fn_target_info): New function.
34252 (rs6000_update_ipa_fn_target_info): Likewise.
34253 (rs6000_can_inline_p): Adjust for ipa function summary target info.
34254 * config/rs6000/rs6000.h (RS6000_FN_TARGET_INFO_HTM): New macro.
34255 * ipa-fnsummary.c (ipa_dump_fn_summary): Adjust for ipa function
34256 summary target info.
34257 (analyze_function_body): Adjust for ipa function summary target info
34258 and call hook rs6000_need_ipa_fn_target_info and
34259 rs6000_update_ipa_fn_target_info.
34260 (ipa_merge_fn_summary_after_inlining): Adjust for ipa function summary
34262 (inline_read_section): Likewise.
34263 (ipa_fn_summary_write): Likewise.
34264 * ipa-fnsummary.h (ipa_fn_summary::target_info): New member.
34265 * doc/tm.texi: Regenerate.
34266 * doc/tm.texi.in (TARGET_UPDATE_IPA_FN_TARGET_INFO): Document new hook.
34267 (TARGET_NEED_IPA_FN_TARGET_INFO): Likewise.
34268 * target.def (update_ipa_fn_target_info): New hook.
34269 (need_ipa_fn_target_info): Likewise.
34270 * targhooks.c (default_need_ipa_fn_target_info): New function.
34271 (default_update_ipa_fn_target_info): Likewise.
34272 * targhooks.h (default_update_ipa_fn_target_info): New declare.
34273 (default_need_ipa_fn_target_info): Likewise.
34275 2022-01-04 Martin Sebor <msebor@redhat.com>
34277 PR middle-end/99612
34278 * builtins.c (get_memmodel): Move warning code to
34279 gimple-ssa-warn-access.cc.
34280 (expand_builtin_atomic_compare_exchange): Same.
34281 (expand_ifn_atomic_compare_exchange): Same.
34282 (expand_builtin_atomic_load): Same.
34283 (expand_builtin_atomic_store): Same.
34284 (expand_builtin_atomic_clear): Same.
34285 * doc/extend.texi (__atomic_exchange_n): Update valid memory
34287 * gimple-ssa-warn-access.cc (memmodel_to_uhwi): New function.
34288 (struct memmodel_pair): New struct.
34289 (memmodel_name): New function.
34290 (pass_waccess::maybe_warn_memmodel): New function.
34291 (pass_waccess::check_atomic_memmodel): New function.
34292 (pass_waccess::check_atomic_builtin): Handle memory model.
34293 * input.c (expansion_point_location_if_in_system_header): Return
34294 original location if expansion location is in a system header.
34296 2022-01-04 Uroš Bizjak <ubizjak@gmail.com>
34299 * config/i386/mmx.md (one_cmplv2qi3): New insn pattern.
34300 (one_cmplv2qi3 splitters): New post-reload splitters.
34301 (*andnotv2qi3): New insn pattern.
34302 (andnotv2qi3 splitters): New post-reload splitters.
34303 (<any_logic:code>v2qi3): New insn pattern.
34304 (<any_logic:insn>v2qi3 splitters): New post-reload splitters.
34306 2022-01-04 Richard Biener <rguenther@suse.de>
34308 PR tree-optimization/103800
34309 * tree-vect-loop.c (vectorizable_phi): Remove assert and
34312 2022-01-04 Richard Biener <rguenther@suse.de>
34314 PR tree-optimization/103690
34315 * tree-pass.h (tail_merge_optimize): Adjust.
34316 * tree-ssa-tail-merge.c (tail_merge_optimize): Pass in whether
34317 to re-split critical edges, move CFG cleanup ...
34318 * tree-ssa-pre.c (pass_pre::execute): ... here, before
34319 simple_dce_from_worklist and delay freeing inserted_exprs from
34321 (fini_pre): .. here.
34323 2022-01-04 Roger Sayle <roger@nextmovesoftware.com>
34325 * config/nvptx/nvptx.h (STORE_FLAG_VALUE): Change to 1.
34326 * config/nvptx/nvptx.md (movbi): Use P1 constraint for true.
34327 (setcc_from_bi): Remove SImode specific pattern.
34328 (setcc<mode>_from_bi): Provide more general HSDIM pattern.
34329 (extendbi<mode>2, zeroextendbi<mode>2): Provide instructions
34330 for sign- and zero-extending BImode predicates to integers.
34331 (setcc_int<mode>): Remove previous (-1-based) instructions.
34332 (cstorebi4): Remove BImode to SImode specific expander.
34333 (cstore<mode>4): Fix indentation. Expand using setccsi_from_bi.
34334 (cstore<mode>4): For both integer and floating point modes.
34336 2022-01-04 Olivier Hainque <hainque@adacore.com>
34338 * gcc.c (driver_handle_option): do_save --sysroot.
34340 2022-01-04 Richard Biener <rguenther@suse.de>
34342 PR tree-optimization/103864
34343 PR tree-optimization/103544
34344 * tree-vect-slp.c (vect_analyze_slp_instance): Exclude
34345 reductions wrapped in conversions from SLP handling.
34346 (vect_analyze_slp): Revert PR103544 change.
34348 2022-01-04 Jakub Jelinek <jakub@redhat.com>
34350 PR rtl-optimization/103860
34351 * shrink-wrap.c (try_shrink_wrapping): Don't call can_get_prologue
34352 uselessly for blocks for which it has been called already.
34354 2022-01-04 Cui,Lili <lili.cui@intel.com>
34356 * common/config/i386/cpuinfo.h (get_intel_cpu): Add new model values
34357 to Alderlake and Rocketlake.
34359 2022-01-04 Chung-Lin Tang <cltang@codesourcery.com>
34361 PR middle-end/103643
34362 * gimplify.c (gimplify_omp_affinity): Adjust gimplify_expr of entire
34363 OMP_CLAUSE_DECL to use 'is_gimple_lvalue, fb_lvalue'
34365 2022-01-04 liuhongt <hongtao.liu@intel.com>
34368 * config/i386/sse.md (*bit_and_float_vector_all_ones):
34369 Force_reg operand 1 to avoid ICE.
34371 2022-01-04 Jason Merrill <jason@redhat.com>
34373 * tree-pretty-print.c (do_niy): Add spc parameter.
34375 (print_call_name): Add spc local variable.
34377 2022-01-03 Uroš Bizjak <ubizjak@gmail.com>
34380 * config/i386/mmx.md (mov<V_32:mode>): Remove TARGET_SSE2 constraint.
34381 (mov<V_32:mode>_internal): Ditto.
34382 (*push<V_32:mode>_rex64): Ditto.
34383 (movmisalign<V_32:mode>): Ditto.
34384 (*push<V_32:mode>_rex64 splitter): Enable for
34385 TARGET_64BIT && TARGET_SSE.
34386 (*push<V_32:mode>2): Remove insn pattern.
34388 2022-01-03 Andrew Pinski <apinski@marvell.com>
34391 * doc/extend.texi: Extend the documentation about Complex
34392 types for casting and also rewrite the __real__/__imag__
34393 expression portion to use tables.
34394 Move __builtin_complex to the Complex type section.
34396 2022-01-03 Jakub Jelinek <jakub@redhat.com>
34399 * internal-fn.def (ATOMIC_ADD_FETCH_CMP_0, ATOMIC_SUB_FETCH_CMP_0,
34400 ATOMIC_AND_FETCH_CMP_0, ATOMIC_OR_FETCH_CMP_0, ATOMIC_XOR_FETCH_CMP_0):
34402 * internal-fn.h (ATOMIC_OP_FETCH_CMP_0_EQ, ATOMIC_OP_FETCH_CMP_0_NE,
34403 ATOMIC_OP_FETCH_CMP_0_LT, ATOMIC_OP_FETCH_CMP_0_LE,
34404 ATOMIC_OP_FETCH_CMP_0_GT, ATOMIC_OP_FETCH_CMP_0_GE): New enumerators.
34405 * internal-fn.c (expand_ATOMIC_ADD_FETCH_CMP_0,
34406 expand_ATOMIC_SUB_FETCH_CMP_0, expand_ATOMIC_AND_FETCH_CMP_0,
34407 expand_ATOMIC_OR_FETCH_CMP_0, expand_ATOMIC_XOR_FETCH_CMP_0): New
34409 * optabs.def (atomic_add_fetch_cmp_0_optab,
34410 atomic_sub_fetch_cmp_0_optab, atomic_and_fetch_cmp_0_optab,
34411 atomic_or_fetch_cmp_0_optab, atomic_xor_fetch_cmp_0_optab): New
34413 * builtins.h (expand_ifn_atomic_op_fetch_cmp_0): Declare.
34414 * builtins.c (expand_ifn_atomic_op_fetch_cmp_0): New function.
34415 * tree-ssa-ccp.c: Include internal-fn.h.
34416 (optimize_atomic_bit_test_and): Add . before internal fn call
34417 in function comment. Change return type from void to bool and
34418 return true only if successfully replaced.
34419 (optimize_atomic_op_fetch_cmp_0): New function.
34420 (pass_fold_builtins::execute): Use optimize_atomic_op_fetch_cmp_0
34421 for BUILT_IN_ATOMIC_{ADD,SUB,AND,OR,XOR}_FETCH_{1,2,4,8,16} and
34422 BUILT_IN_SYNC_{ADD,SUB,AND,OR,XOR}_AND_FETCH_{1,2,4,8,16},
34423 for *XOR* ones only if optimize_atomic_bit_test_and failed.
34424 * config/i386/sync.md (atomic_<plusminus_mnemonic>_fetch_cmp_0<mode>,
34425 atomic_<logic>_fetch_cmp_0<mode>): New define_expand patterns.
34426 (atomic_add_fetch_cmp_0<mode>_1, atomic_sub_fetch_cmp_0<mode>_1,
34427 atomic_<logic>_fetch_cmp_0<mode>_1): New define_insn patterns.
34428 * doc/md.texi (atomic_add_fetch_cmp_0<mode>,
34429 atomic_sub_fetch_cmp_0<mode>, atomic_and_fetch_cmp_0<mode>,
34430 atomic_or_fetch_cmp_0<mode>, atomic_xor_fetch_cmp_0<mode>): Document
34431 new named patterns.
34433 2022-01-03 Richard Biener <rguenther@suse.de>
34435 PR middle-end/103851
34436 * tree-cfg.c (move_sese_region_to_fn): Always release SSA names.
34438 2022-01-03 Jakub Jelinek <jakub@redhat.com>
34441 * symtab.c: Include fold-const.h.
34442 (symtab_node::equal_address_to): If folding_initializer is true,
34443 handle it like memory_accessed. Simplify.
34445 2022-01-03 Martin Liska <mliska@suse.cz>
34447 * doc/extend.texi: Use ; for function declarations.
34449 2022-01-03 Jakub Jelinek <jakub@redhat.com>
34452 * symtab.c (symtab_node::equal_address_to): Return 0 if one of
34453 VAR_DECLs has "non overlapping" attribute and rs1 != rs2.
34455 2022-01-03 Jakub Jelinek <jakub@redhat.com>
34457 * gcc.c (process_command): Update copyright notice dates.
34458 * gcov-dump.c (print_version): Ditto.
34459 * gcov.c (print_version): Ditto.
34460 * gcov-tool.c (print_version): Ditto.
34461 * gengtype.c (create_file): Ditto.
34462 * doc/cpp.texi: Bump @copying's copyright year.
34463 * doc/cppinternals.texi: Ditto.
34464 * doc/gcc.texi: Ditto.
34465 * doc/gccint.texi: Ditto.
34466 * doc/gcov.texi: Ditto.
34467 * doc/install.texi: Ditto.
34468 * doc/invoke.texi: Ditto.
34470 2022-01-02 Uroš Bizjak <ubizjak@gmail.com>
34473 * config/i386/i386.h (VALID_SSE2_REG_MODE): Add V2QImode.
34474 (VALID_INT_MODE_P): Ditto.
34475 * config/i386/i386.c (ix86_secondary_reload): Handle
34476 V2QImode reloads from SSE register to memory.
34477 (vector_mode_supported_p): Always return true for V2QImode.
34478 * config/i386/i386.md (*subqi_ext<mode>_2): New insn pattern.
34479 (*negqi_ext<mode>_2): Ditto.
34480 * config/i386/mmx.md (movv2qi): New expander.
34481 (movmisalignv2qi): Ditto.
34482 (*movv2qi_internal): New insn pattern.
34483 (*pushv2qi2): Ditto.
34484 (negv2qi2 and splitters): Ditto.
34485 (<plusminus:insn>v2qi3 and splitters): Ditto.
34487 2022-01-02 John David Anglin <danglin@gcc.gnu.org>
34489 * config/pa/pa.md (atomic_storeq): Use optab_libfunc to access
34490 sync_lock_test_and_set libfunc. Call convert_memory_address to
34491 convert memory address to Pmode.
34492 (atomic_storehi, atomic_storesi, atomic_storedi): Likewise.
34495 Copyright (C) 2022 Free Software Foundation, Inc.
34497 Copying and distribution of this file, with or without modification,
34498 are permitted in any medium without royalty provided the copyright
34499 notice and this notice are preserved.