1 2004-10-25 Daniel Jacobowitz <dan@debian.org>
3 * Makefile.am: Run dep-am.
4 * aclocal.m4: Regenerate with automake 1.9.2.
5 * Makefile.in: Regenerate with automake 1.9.2.
6 * doc/Makefile.in: Likewise.
8 * config/tc-arm.c: Include "dw2gencfi.h".
9 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
11 * config/tc-arm.h (TARGET_USE_CFIPOP, DWARF2_DEFAULT_RETURN_COLUMN)
12 (DWARF2_CIE_DATA_ALIGNMENT, tc_regname_to_dw2regnum)
13 (tc_cfi_frame_initial_instructions): Define.
14 (tc_arm_regname_to_dw2regnum, tc_arm_frame_initial_instructions):
17 2004-10-21 Tomer Levi <Tomer.Levi@nsc.com>
19 * config/tc-crx.c (assemble_insn): Check unsigned immediate
21 Update coprocessor id to be unsigned immediate.
23 2004-10-18 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
25 * config/tc-i386.c (O_secrel): Delete.
26 (tc_pe_dwarf2_emit_offset): New function.
27 * config/tc-i386.h (O_secrel): Define as O_md1.
28 (TC_DWARF2_EMIT_OFFSET): Define.
30 2004-10-18 Nick Clifton <nickc@redhat.com>
32 * config/tc-xstormy16.c (xstormy16_cons_fix_new): Accept and
33 ignore @fptr() directives for 4-byte fixups.
35 2004-10-15 Alan Modra <amodra@bigpond.net.au>
37 * Makefile.am: Run "make dep-am"
38 * Makefile.in: Regenerate.
40 2004-10-14 Bob Wilson <bob.wilson@acm.org>
42 * doc/c-xtensa.texi (Xtensa Options, Absolute Literals Directive):
43 Remove comments about placement of literal pools.
44 (Literal Directive): Update description of literal placement.
45 (Literal Prefix Directive): Remove statement that this does not apply
46 to absolute-mode literals. Describe new section naming scheme.
48 2004-10-12 Bob Wilson <bob.wilson@acm.org>
50 * config/tc-xtensa.c (xtensa_frequency_pseudo): Use set_subseg_freq.
51 (is_entry_opcode, is_movi_opcode, is_the_loop_opcode, is_jx_opcode,
52 is_windowed_return_opcode): Delete.
53 (xtensa_frob_label): Use get_subseg_target_freq.
54 (md_assemble): Inline call to is_entry_opcode.
55 (xtensa_handle_align): Inline call to get_frag_is_literal.
56 (relaxation_requirements): Inline call to is_jx_opcode.
57 (emit_single_op): Inline call to is_movi_opcode.
58 (xg_assemble_vliw_tokens): Inline calls to get_frag_is_insn,
59 get_frag_is_no_transform, is_entry_opcode, and
60 set_frag_is_specific_opcode. Use get_subseg_total_freq.
61 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags,
62 xtensa_fix_close_loop_end_frags, relax_frag_immed, convert_frag_immed):
63 Inline calls to get_frag_is_no_transform.
64 (next_instrs_are_b_retw): Inline call to is_windowed_return_opcode.
65 (xtensa_fix_short_loop_frags): Inline calls to is_the_loop_opcode and
66 get_frag_is_no_transform.
67 (convert_frag_immed_finish_loop): Inline calls to get_expression_value
68 and set_frag_is_no_transform.
69 (get_expression_value): Delete.
70 (subseg_map struct): Rename cur_total_freq to total_freq. Rename
71 cur_target_freq to target_freq.
72 (get_subseg_info): Split out code to create a new map entry into ...
73 (add_subseg_info): ... this new function.
74 (get_last_insn_flags): Check if get_subseg_info succeeded.
75 (set_last_insn_flags): Call add_subseg_info if needed.
76 (get_subseg_total_freq, get_subseg_target_freq, set_subseg_freq): New.
77 (xtensa_reorder_segments): Compute last_sec while counting sections.
78 Remove call to get_last_sec.
79 (get_last_sec): Delete.
80 (cache_literal_section): Inline call to retrieve_literal_seg and its
81 callees, seg_present and add_seg_list.
82 (retrieve_literal_seg, seg_present, add_seg_list): Delete.
83 (get_frag_is_insn, get_frag_is_no_transform,
84 set_frag_is_specific_opcode, set_frag_is_no_transform): Delete.
85 * config/tc-xtensa.h (MAX_SLOTS): Reduce from 31 to 15.
87 2004-10-12 Bob Wilson <bob.wilson@acm.org>
89 * config/tc-xtensa.c: Use ISO C90 formatting.
90 * config/tc-xtensa.h: Likewise.
91 * config/xtensa-istack.h: Likewise.
92 * config/xtensa-relax.c: Likewise.
93 * config/xtensa-relax.h: Likewise.
95 2004-10-12 Paul Brook <paul@codesourcery.com>
97 * config/tc-arm.c (md_begin): Change EF_ARM_EABI_VER3 to
100 * doc/c-arm.texi: Document that we actually support -meabi=4, not
103 2004-10-08 Bob Wilson <bob.wilson@acm.org>
105 * doc/as.texinfo (VTableEntry, VTableInherit): Add "directive" to index
107 (Acknowledgements): Use "GAS" instead of AS variable.
109 2004-10-08 Daniel Jacobowitz <dan@debian.org>
111 * config/tc-i386.c: Include "elf/x86-64.h".
112 (i386_elf_section_type): New function.
113 * config/tc-i386.h (md_elf_section_type): Define.
114 (i386_elf_section_type): New prototype.
116 2004-10-08 Linus Nielsen Feltzing <linus@haxx.se>
118 * config/m68k-parse.h (enum m68k_register): New control register,
120 * config/tc-m68k.c: Correct control register set for MCF5249.
122 2004-10-07 Bob Wilson <bob.wilson@acm.org>
123 Sterling Augustine <sterling@tensilica.com>
125 * config/tc-xtensa.c (absolute_literals_supported): New global flag.
126 (UNREACHABLE_MAX_WIDTH): Define.
127 (XTENSA_FETCH_WIDTH): Delete.
128 (cur_vinsn, xtensa_fetch_width, xt_saved_debug_type, past_xtensa_end,
129 prefer_const16, prefer_l32r): New global variables.
130 (LIT4_SECTION_NAME): Define.
131 (lit4_state struct): Add lit4_seg_name and lit4_seg fields.
132 (XTENSA_PROP_*, GET_XTENSA_PROP_*, SET_XTENSA_PROP_*): Define.
133 (frag_flags struct): New.
134 (xtensa_block_info struct): Move from tc-xtensa.h. Add flags field.
135 (subseg_map struct): Add cur_total_freq and cur_target_freq fields.
136 (bitfield, bit_is_set, set_bit, clear_bit): Define.
137 (MAX_FORMATS): Define.
138 (op_placement_info struct, op_placement_table): New.
139 (O_pltrel, O_hi16, O_lo16): Define.
140 (directiveE enum): Rename directive_generics to directive_transform.
141 Delete directive_relax. Add directive_schedule,
142 directive_absolute_literals, and directive_last_directive.
143 (directive_info): Rename "generics" to "transform". Delete "relax".
144 Add "schedule" and "absolute-literals".
145 (directive_state): Adjust entries to match changes in directive_info.
146 (xtensa_relax_statesE, RELAX_IMMED_MAXSTEPS): Move to tc-xtensa.h.
147 (xtensa_const16_opcode, xtensa_movi_opcode, xtensa_movi_n_opcode,
148 xtensa_l32r_opcode, xtensa_nop_opcode, xtensa_rsr_lcount_opcode): New.
149 (xtensa_j_opcode, xtensa_rsr_opcode): Delete.
150 (align_only_targets, software_a0_b_retw_interlock,
151 software_avoid_b_j_loop_end, maybe_has_b_j_loop_end,
152 software_avoid_short_loop, software_avoid_close_loop_end,
153 software_avoid_all_short_loops, specific_opcode): Delete.
154 (warn_unaligned_branch_targets): New.
155 (workaround_a0_b_retw, workaround_b_j_loop_end, workaround_short_loop,
156 workaround_close_loop_end, workaround_all_short_loops): Default FALSE.
157 (option_[no_]link_relax, option_[no_]transform,
158 option_[no_]absolute_literals, option_warn_unaligned_targets,
159 option_prefer_l32r, option_prefer_const16, option_target_hardware):
161 (option_[no_]align_only_targets, option_literal_section_name,
162 option_text_section_name, option_data_section_name,
163 option_bss_section_name, option_eb, option_el): Delete.
164 (md_longopts): Add entries for: [no-]transform, [no-]absolute-literals,
165 warn-unaligned-targets, prefer-l32r, prefer-const16, [no-]link-relax,
166 and target-hardware. Delete entries for [no-]target-align-only,
167 literal-section-name, text-section-name, data-section-name, and
169 (md_parse_option): Handle new options and remove old ones. Accept but
170 ignore [no-]density options. Warn for [no-]generics and [no-]relax
171 and treat them as [no-]transform.
172 (md_show_usage): Add new options and remove old ones.
173 (xtensa_setup_hw_workarounds): New.
174 (md_pseudo_table): Change "word" entry to use xtensa_elf_cons. Add
175 "long", "short", "loc" and "frequency" entries.
176 (use_generics): Rename to ...
177 (use_transform): ... this function. Add past_xtensa_end check.
178 (use_longcalls): Add past_xtensa_end check.
179 (code_density_available, can_relax): Delete.
180 (do_align_targets): New.
181 (get_directive): Accept dashes in directive names. Warn about
182 [no-]generics and [no-]relax directives and treat them as
184 (xtensa_begin_directive): Call md_flush_pending_output only for some
185 directives. Check for directives inside instruction bundles. Warn
186 about deprecated ".begin literal" usage. Warn and ignore [no-]density
187 directives. Handle new directives. Check generating_literals flag
189 (xtensa_end_directive): Check for directives inside instruction
190 bundles. Warn and ignore [no-]density directives. Handle new
191 directives. Call xtensa_set_frag_assembly_state.
192 (xtensa_loc_directive_seen, xtensa_dwarf2_directive_loc,
193 xtensa_dwarf2_emit_insn): New.
194 (xtensa_literal_position): Call md_flush_pending_output. Do not check
195 use_literal_section flag.
196 (xtensa_literal_pseudo): Call md_flush_pending_output. Handle absolute
197 literals. Use xtensa_elf_cons to parse the expression.
198 (xtensa_literal_prefix): Do not check use_literal_section. Support
199 ".lit4" sections for absolute literals. Change prefix convention to
200 replace ".text" (or ".t" in a linkonce section). No need to call
202 (xtensa_frequency_pseudo, xtensa_elf_cons, xtensa_elf_suffix): New.
203 (expression_end): Handle closing braces and colons.
204 (PLT_SUFFIX, plt_suffix): Delete.
205 (expression_maybe_register): Use new xtensa-isa.h functions. Use
206 xtensa_elf_suffix instead of checking for plt suffix, and handle O_lo16
207 and O_hi16 expressions as well.
208 (tokenize_arguments): Handle closing braces and colons.
209 (parse_arguments): Use new xtensa-isa.h functions. Handle "invisible"
210 operands and paired register syntax.
211 (get_invisible_operands): New.
212 (xg_translate_sysreg_op): Handle new Xtensa LX RSR/WSR/XSR syntax. Use
213 new xtensa-isa.h functions.
214 (xtensa_translate_old_userreg_ops, xtensa_translate_zero_immed): New.
215 (xg_translate_idioms): Check if inside bundle. Use use_transform.
216 Handle new Xtensa LX RSR/WSR/XSR syntax. Remove code to widen density
217 instructions. Use xtensa_translate_zero_immed.
218 (operand_is_immed, operand_is_pcrel_label): Delete.
219 (get_relaxable_immed): Use new xtensa-isa.h functions.
220 (get_opcode_from_buf): Add slot parameter. Use new xtensa-isa.h
222 (xtensa_print_insn_table, print_vliw_insn): New.
223 (is_direct_call_opcode): Use new xtensa-isa.h functions.
224 (is_call_opcode, is_loop_opcode, is_conditional_branch_opcode,
225 is_branch_or_jump_opcode): Delete.
226 (is_movi_opcode, decode_reloc, encode_reloc, encode_alt_reloc): New.
227 (opnum_to_reloc, reloc_to_opnum): Delete.
228 (xtensa_insnbuf_set_operand, xtensa_insnbuf_get_operand): Use new
229 xtensa-isa.h functions. Operate on one slot of an instruction.
230 (xtensa_insnbuf_set_immediate_field, is_negatable_branch,
231 xg_get_insn_size): Delete.
232 (xg_get_build_instr_size): Use xg_get_single_size.
233 (xg_is_narrow_insn, xg_is_single_relaxable_insn): Update calls to
234 xg_build_widen_table. Use xg_get_single_size.
235 (xg_get_max_narrow_insn_size): Delete.
236 (xg_get_max_insn_widen_size, xg_get_max_insn_widen_literal_size,
237 xg_is_relaxable_insn): Update calls to xg_build_widen_table. Use
239 (xg_build_to_insn): Record the loc field. Handle OP_OPERAND_HI16U and
240 OP_OPERAND_LOW16U. Check xg_valid_literal_expression.
241 (xg_expand_to_stack, xg_expand_narrow): Update calls to
242 xg_build_widen_table. Use xg_get_single_size.
243 (xg_immeds_fit): Use new xtensa-isa.h functions. Update call to
245 (xg_symbolic_immeds_fit): Likewise. Also handle O_lo16 and O_hi16, and
246 treat weak symbols conservatively.
247 (xg_check_operand): Use new xtensa-isa.h functions.
248 (is_dnrange): Delete.
249 (xg_assembly_relax): Inline previous calls to tinsn_copy.
250 (xg_finish_frag): Specify separate relax states for the frag and slot0.
251 (is_branch_jmp_to_next, xg_add_branch_and_loop_targets): Use new
252 xtensa-isa.h functions.
253 (xg_instruction_matches_option_term, xg_instruction_matches_or_options,
254 xg_instruction_matches_options): New.
255 (xg_instruction_matches_rule): Handle O_register expressions. Call
256 xg_instruction_matches_options.
257 (transition_rule_cmp): New.
258 (xg_instruction_match): Update call to xg_build_simplify_table.
259 (xg_build_token_insn): Record loc fields.
260 (xg_simplify_insn): Check is_specific_opcode field and
261 density_supported flag.
262 (xg_expand_assembly_insn): Skip checking code_density_available. Use
263 new xtensa-isa.h functions. Call use_transform instead of can_relax.
264 (xg_assemble_literal): Add error handling for O_big. Call
265 record_alignment. Handle O_pltrel.
266 (xg_valid_literal_expression): New.
267 (xg_assemble_literal_space): Add slot parameter. Remove call to
268 set_expr_symbol_offset. Add call to record_alignment. Update call to
270 (xg_emit_insn): Delete.
271 (xg_emit_insn_to_buf): Add format parameter. Update calls to
272 xg_add_opcode_fix and xtensa_insnbuf_to_chars.
273 (xg_add_opcode_fix): Change opcode parameter to tinsn and add format
274 and slot parameters. Handle new "alternate" relocations for absolute
275 literals and CONST16 instructions. Check for bad uses of O_lo16 and
276 O_hi16. Use new xtensa-isa.h functions.
277 (xg_assemble_tokens): Delete.
278 (is_register_writer): Use new xtensa-isa.h functions.
279 (is_bad_loopend_opcode): Check for xtensa_rsr_lcount_opcode instead of
280 old-style RSR from LCOUNT.
281 (next_frag_opcode): Delete.
282 (next_frag_opcode_is_loop, next_frag_format_size, frag_format_size,
283 update_next_frag_state): New.
284 (update_next_frag_nop_state): Delete.
285 (next_frag_pre_opcode_bytes): Use next_frag_opcode_is_loop.
286 (xtensa_mark_literal_pool_location): Check use_literal_section flag and
287 the state of the absolute-literals directive. Add calls to
288 record_alignment and xtensa_set_frag_assembly_state. Call
289 xtensa_switch_to_non_abs_literal_fragment instead of
290 xtensa_switch_to_literal_fragment.
292 (assemble_nop): Use build_nop. Update call to xtensa_insnbuf_to_chars.
293 (get_expanded_loop_offset): Change check for undefined opcode to an
295 (xtensa_set_frag_assembly_state, relaxable_section,
296 xtensa_find_unmarked_state_frags, xtensa_find_unaligned_branch_targets,
297 xtensa_find_unaligned_loops, xg_apply_tentative_value): New.
298 (md_begin): Update call to xtensa_isa_init. Initialize linkrelax to 1.
299 Set lit4_seg_name. Call xg_init_vinsn. Initialize new global opcodes.
300 Call init_op_placement_info_table and xtensa_set_frag_assembly_state.
301 (xtensa_init_fix_data): New.
302 (xtensa_frob_label): Reset label symbol to the current frag. Check
303 do_align_targets and generating_literals flag. Propagate frequency
304 info to new alignment frag. Call xtensa_set_frag_assembly_state.
305 (xtensa_unrecognized_line): New.
306 (xtensa_flush_pending_output): Check if inside a bundle. Add a call
307 to xtensa_set_frag_assembly_state.
308 (error_reset_cur_vinsn): New.
309 (md_assemble): Remove check for literal frag. Remove call to
310 istack_init. Call use_transform instead of use_generics. Parse
311 explicit instruction format specifiers. Move code for
312 a0_b_retw_interlock workaround to xg_assemble_vliw_tokens. Call
313 error_reset_cur_vinsn on errors. Add call to get_invisible_operands.
314 Add dwarf2_where call. Remote automatic alignment for ENTRY
315 instructions. Move call to xtensa_clear_insn_labels to the end.
316 Rearrange to handle bundles.
317 (xtensa_cons_fix_new): Delete.
318 (xtensa_handle_align): New.
319 (xtensa_frag_init): Call xtensa_set_frag_assembly_state. Remove
320 assignment to is_no_density field.
321 (md_pcrel_from): Use new xtensa-isa.h functions. Use decode_reloc
322 instead of reloc_to_opnum. Handle "alternate" relocations.
323 (xtensa_force_relocation, xtensa_check_inside_bundle,
324 xtensa_elf_section_change_hook): New.
325 (xtensa_symbol_new_hook): Delete.
326 (xtensa_fix_adjustable): Check for difference of symbols with an
327 offset. Check for external and weak symbols.
328 (md_apply_fix3): Remove cases for XTENSA_OP{0,1,2} relocs.
329 (md_estimate_size_before_relax): Return expansion for the first slot.
330 (tc_gen_reloc): Handle difference of symbols by producing
331 XTENSA_DIFF{8,16,32} relocs and by writing the value of the difference
332 into the output. Handle new XTENSA_SLOT*_OP relocs by storing the
333 tentative values into the output when linkrelax is set.
334 (XTENSA_PROP_SEC_NAME): Define.
335 (xtensa_post_relax_hook): Call xtensa_find_unmarked_state_frags.
336 Create literal tables only if using literal sections. Create new
337 property tables instead of old instruction tables. Check for unaligned
338 branch targets and loops.
339 (finish_vinsn, find_vinsn_conflicts, check_t1_t2_reads_and_writes,
340 new_resource_table, clear_resource_table, resize_resource_table,
341 resources_available, reserve_resources, release_resources,
342 opcode_funcUnit_use_unit, opcode_funcUnit_use_stage,
343 resources_conflict, xg_find_narrowest_format, relaxation_requirements,
344 bundle_single_op, emit_single_op, xg_assemble_vliw_tokens): New.
345 (xtensa_end): Call xtensa_flush_pending_output. Set past_xtensa_end
346 flag. Update checks for workaround options. Call
347 xtensa_mark_narrow_branches and xtensa_mark_zcl_first_insns.
348 (xtensa_cleanup_align_frags): Add special case for branch targets.
349 Check for and mark unreachable frags.
350 (xtensa_fix_target_frags): Remove use of align_only_targets flag.
351 Use RELAX_LOOP_END_BYTES in special case for negatable branch at the
352 end of a zero-overhead loop body.
353 (frag_can_negate_branch): Handle instructions with multiple slots.
354 Use new xtensa-isa.h functions
355 (xtensa_mark_narrow_branches, is_narrow_branch_guaranteed_in_range,
356 xtensa_mark_zcl_first_insns): New.
357 (xtensa_fix_a0_b_retw_frags, xtensa_fix_b_j_loop_end_frags): Error if
358 transformations are disabled.
359 (next_instrs_are_b_retw): Use new xtensa-isa.h functions. Handle
360 multislot instructions.
361 (xtensa_fix_close_loop_end_frags, xtensa_fix_short_loop_frags):
362 Likewise. Also error if transformations are disabled.
363 (unrelaxed_frag_max_size): New.
364 (unrelaxed_frag_min_insn_count, unrelax_frag_has_b_j): Use new
365 xtensa-isa.h functions.
366 (xtensa_sanity_check, is_empty_loop, is_local_forward_loop): Use
367 xtensa_opcode_is_loop instead of is_loop_opcode.
368 (get_text_align_power): Replace as_fatal with assertion.
369 (get_text_align_fill_size): Iterate instead of using modulus when
371 (get_noop_aligned_address): Assert that this is for a machine-dependent
372 RELAX_ALIGN_NEXT_OPCODE frag. Use next_frag_opcode_is_loop,
373 xg_get_single_size, and frag_format_size.
374 (get_widen_aligned_address): Rename to ...
375 (get_aligned_diff): ... this function. Add max_diff parameter.
376 Remove handling of rs_align/rs_align_code frags. Use
377 next_frag_format_size, get_text_align_power, get_text_align_fill_size,
378 next_frag_opcode_is_loop, and xg_get_single_size. Compute max_diff
379 and pass it back to caller.
380 (xtensa_relax_frag): Use relax_frag_loop_align. Add code for new
381 RELAX_SLOTS, RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN,
382 RELAX_FILL_NOP, and RELAX_UNREACHABLE frag types. Check relax_seen.
383 (relax_frag_text_align): Rename to ...
384 (relax_frag_loop_align): ... this function. Assume loops can only be
385 in the first slot of an instruction.
386 (relax_frag_add_nop): Use assemble_nop instead of constructing an OR
387 instruction. Remove call to frag_wane.
388 (relax_frag_narrow): Rename to ...
389 (relax_frag_for_align): ... this function. Extend to handle
390 RELAX_FILL_NOP and RELAX_UNREACHABLE, as well as RELAX_SLOTS with
391 RELAX_NARROW for the first slot.
392 (find_address_of_next_align_frag, bytes_to_stretch): New.
393 (future_alignment_required): Use find_address_of_next_align_frag and
394 bytes_to_stretch. Look ahead to subsequent frags to make smarter
396 (relax_frag_immed): Add format, slot, and estimate_only parameters.
397 Check if transformations are enabled for b_j_loop_end workaround.
398 Use new xtensa-isa.h functions and handle multislot instructions.
399 Update call to xg_assembly_relax.
400 (md_convert_frag): Handle new RELAX_SLOTS, RELAX_UNREACHABLE,
401 RELAX_MAYBE_UNREACHABLE, RELAX_MAYBE_DESIRE_ALIGN, and RELAX_FILL_NOP
403 (convert_frag_narrow): Add segP, format and slot parameters. Call
404 convert_frag_immed for branch instructions. Adjust calls to
405 tinsn_from_chars, tinsn_immed_from_frag, and xg_emit_insn_to_buf. Use
406 xg_get_single_size and xg_get_single_format.
407 (convert_frag_fill_nop): New.
408 (convert_frag_immed): Add format and slot parameters. Handle multislot
409 instructions and use new xtensa-isa.h functions. Update calls to
410 tinsn_immed_from_frag and xg_assembly_relax. Check if transformations
411 enabled for b_j_loop_end workaround. Use build_nop instead of
412 assemble_nop. Check is_specific_opcode flag. Check for unreachable
413 frags. Use xg_get_single_size. Handle O_pltrel.
414 (fix_new_exp_in_seg): Remove check for old plt flag.
415 (convert_frag_immed_finish_loop): Update calls to tinsn_from_chars and
416 xtensa_insnbuf_to_chars. Call tinsn_immed_from_frag. Change check
417 for loop opcode to an assertion. Mark all frags up to the end of the
418 loop as not transformable.
419 (get_last_insn_flags, set_last_insn_flags): Use get_subseg_info.
420 (get_subseg_info): New.
421 (xtensa_move_literals): Call xtensa_set_frag_assembly_state. Add null
423 (xtensa_switch_to_literal_fragment): Rewrite to handle absolute
424 literals and use xtensa_switch_to_non_abs_literal_fragment otherwise.
425 (xtensa_switch_to_non_abs_literal_fragment): New.
426 (cache_literal_section): Add is_code parameter and pass it through to
427 retrieve_literal_seg.
428 (retrieve_literal_seg): Add is_code parameter and use it to set the
429 flags on the literal section. Handle case where head parameter is 0.
430 (get_frag_is_no_transform, set_frag_is_specific_opcode,
431 set_frag_is_no_transform): New.
432 (xtensa_create_property_segments): Add end_property_function parameter
433 and pass it through to add_xt_block_frags. Call bfd_get_section_flags
434 and skip SEC_DEBUGGING and !SEC_ALLOC sections.
435 (xtensa_create_xproperty_segments, section_has_xproperty): New.
436 (add_xt_block_frags): Add end_property_function parameter and call it
437 if it is non-zero. Call xtensa_frag_flags_init.
438 (xtensa_frag_flags_is_empty, xtensa_frag_flags_init,
439 get_frag_property_flags, frag_flags_to_number,
440 xtensa_frag_flags_combinable, xt_block_aligned_size,
441 xtensa_xt_block_combine, add_xt_prop_frags,
442 init_op_placement_info_table, opcode_fits_format_slot,
443 xg_get_single_size, xg_get_single_format): New.
444 (istack_push): Inline call to tinsn_copy.
445 (tinsn_copy): Delete.
446 (tinsn_has_invalid_symbolic_operands): Handle O_hi16 and O_lo16 and
447 CONST16 opcodes. Handle O_big, O_illegal, and O_absent.
448 (tinsn_has_complex_operands): Handle O_hi16 and O_lo16.
449 (tinsn_to_insnbuf): Use xg_get_single_format and new xtensa-isa.h
450 functions. Handle invisible operands.
451 (tinsn_to_slotbuf): New.
452 (tinsn_check_arguments): Use new xtensa-isa.h functions.
453 (tinsn_from_chars): Add slot parameter. Rewrite using xg_init_vinsn,
454 vinsn_from_chars, and xg_free_vinsn.
455 (tinsn_from_insnbuf): New.
456 (tinsn_immed_from_frag): Add slot parameter and handle multislot
457 instructions. Handle symbol differences.
458 (get_num_stack_text_bytes): Use xg_get_single_size.
459 (xg_init_vinsn, xg_clear_vinsn, vinsn_has_specific_opcodes,
460 xg_free_vinsn, vinsn_to_insnbuf, vinsn_from_chars, expr_is_register,
461 get_expr_register, set_expr_symbol_offset_diff): New.
462 * config/tc-xtensa.h (MAX_SLOTS): Define.
463 (xtensa_relax_statesE): Move from tc-xtensa.c. Add
464 RELAX_CHECK_ALIGN_NEXT_OPCODE, RELAX_MAYBE_DESIRE_ALIGN, RELAX_SLOTS,
465 RELAX_FILL_NOP, RELAX_UNREACHABLE, RELAX_MAYBE_UNREACHABLE, and
467 (RELAX_IMMED_MAXSTEPS): Move from tc-xtensa.c.
468 (xtensa_frag_type struct): Add is_assembly_state_set,
469 use_absolute_literals, relax_seen, is_unreachable, is_specific_opcode,
470 is_align, is_text_align, alignment, and is_first_loop_insn fields.
471 Replace is_generics and is_relax fields by is_no_transform field.
472 Delete is_text and is_longcalls fields. Change text_expansion and
473 literal_expansion to arrays of MAX_SLOTS entries. Add arrays of
474 per-slot information: literal_frags, slot_subtypes, slot_symbols,
475 slot_sub_symbols, and slot_offsets. Add fr_prev field.
476 (xtensa_fix_data struct): New.
477 (xtensa_symfield_type struct): Delete plt field.
478 (xtensa_block_info struct): Move definition to tc-xtensa.h. Add
479 forward declaration here.
480 (xt_section_type enum): Delete xt_insn_sec. Add xt_prop_sec.
481 (XTENSA_SECTION_RENAME): Undefine.
482 (TC_FIX_TYPE, TC_INIT_FIX_DATA, TC_FORCE_RELOCATION, NO_PSEUDO_DOT,
483 tc_unrecognized_line, md_do_align, md_elf_section_change_hook,
484 HANDLE_ALIGN, TC_LINKRELAX_FIXUP, SUB_SEGMENT_ALIGN): Define.
485 (TC_CONS_FIX_NEW, tc_symbol_new_hook): Delete.
486 (unit_num_copies_func, opcode_num_units_func,
487 opcode_funcUnit_use_unit_func, opcode_funcUnit_use_stage_func): New.
488 (resource_table struct): New.
489 * config/xtensa-istack.h (MAX_INSN_ARGS): Increase from 6 to 10.
490 (TInsn struct): Add keep_wide, loc, fixup, record_fix, subtype,
491 literal_space, symbol, sub_symbol, offset, and literal_frag fields.
492 (tinsn_copy): Delete prototype.
493 (vliw_insn struct): New.
494 * config/xtensa-relax.c (insn_pattern_struct): Add options field.
495 (widen_spec_list): Add option conditions for density and boolean
496 instructions. Add expansions using CONST16 and conditions for using
497 CONST16 vs. L32R. Use new Xtensa LX RSR/WSR syntax. Add entries for
499 (simplify_spec_list): Add option conditions for density instructions.
500 Add entry for NOP instruction.
501 (append_transition): Add cmp function pointer parameter and use it to
502 insert the new entry in order.
503 (operand_function_LOW16U, operand_function_HI16U): New.
504 (xg_has_userdef_op_fn, xg_apply_userdef_op_fn): Handle
505 OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
506 (enter_opname, split_string): Use xstrdup instead of strdup.
507 (init_insn_pattern): Initialize new options field.
508 (clear_req_or_option_list, clear_req_option_list,
509 clone_req_or_option_list, clone_req_option_list, parse_option_cond):
511 (parse_insn_pattern): Parse option conditions.
512 (transition_applies): New.
513 (build_transition): Use new xtensa-isa.h functions. Fix incorrectly
514 swapped last arguments in calls to append_constant_value_condition.
515 Call clone_req_option_list. Add warning about invalid opcode.
516 Handle LOW16U and HI16U function names.
517 (build_transition_table): Add cmp parameter and use it in calls to
518 append_transition. Use new xtensa-isa.h functions. Check
519 transition_applies before adding entries.
520 (xg_build_widen_table, xg_build_simplify_table): Add cmp parameter and
521 pass it through to build_transition_table.
522 * config/xtensa-relax.h (ReqOrOptionList, ReqOrOption, ReqOptionList,
523 ReqOption, transition_cmp_fn): New types.
524 (OpType enum): Add OP_OPERAND_LOW16U and OP_OPERAND_HI16U.
525 (transition_rule struct): Add options field.
526 * doc/as.texinfo (Overview): Update Xtensa options.
527 * doc/c-xtensa.texi (Xtensa Options): Delete --[no-]density,
528 --[no-]relax, and --[no-]generics options. Update descriptions of
529 --text-section-literals and --[no-]longcalls. Add
530 --[no-]absolute-literals and --[no-]transform.
531 (Xtensa Syntax): Add description of syntax for FLIX instructions.
532 Remove use of "generic" and "specific" terminology for opcodes.
533 (Xtensa Registers): Generalize the syntax description to include
534 user-defined register files.
535 (Xtensa Automatic Alignment): Update.
536 (Xtensa Branch Relaxation): Mention limitation of unconditional jumps.
537 (Xtensa Call Relaxation): Linker can now remove most of the overhead.
538 (Xtensa Directives): Remove confusing rules about precedence.
539 (Density Directive, Relax Directive): Delete.
540 (Schedule Directive): New.
541 (Generics Directive): Rename to ...
542 (Transform Directive): ... this node.
543 (Literal Directive): Update for absolute literals. Missing
544 literal_position directive is now an error.
545 (Literal Position Directive): Update for absolute literals.
546 (Freeregs Directive): Delete.
547 (Absolute Literals Directive): New.
548 (Frame Directive): Minor editing.
549 * Makefile.am (DEPTC_xtensa_elf, DEPOBJ_xtensa_elf, DEP_xtensa_elf):
551 * Makefile.in: Regenerate.
553 2004-10-07 Richard Sandiford <rsandifo@redhat.com>
555 * config/tc-mips.c (append_insn): Use fix_new rather than fix_new_exp
556 to build the second and third fixups for a composite relocation.
557 (macro_read_relocs): New function.
558 (macro_build): Use it.
559 (s_cpsetup): Pass all three composite relocation codes to macro_build.
560 Simplify fragging code accordingly.
561 (s_gpdword): Use fix_new rather than fix_new_exp for the second part
562 of the composite relocation. Set fx_tcbit in both fixups.
564 2004-10-07 Richard Sandiford <rsandifo@redhat.com>
566 * config/tc-mips.c (append_insn): Set fx_tcbit for composite relocs.
567 (md_apply_fix3): Don't treat composite relocs as done.
569 2004-10-07 Jan Beulich <jbeulich@novell.com>
571 * macro.c (macro_expand_body): When ELF, use .LL rather than LL as
572 prefix for symbol names generated from the LOCAL macro directive.
574 * dw2gencfi.c (select_cie_for_fde): When separating CIE out from
575 FDE, treat a DW_CFA_remember_state as we do a DW_CFA_advance_loc.
577 2004-10-07 Tomer Levi <Tomer.Levi@nsc.com>
579 * config/tc-crx.c (preprocess_reglist): Handle Co-processor
581 (md_assemble): Add error checking for Co-Processor instructions.
582 (get_cinv_parameters): Add 'b' option to invalidate the
585 2004-10-05 Paul Brook <paul@codesourcery.com>
587 * config/tc-arm.c (unwind): New variable.
588 (vfp_sp_encode_reg): New function.
589 (vfp_sp_reg_required_here): Use it.
590 (vfp_sp_reg_list, vfp_dp_reg_list): Remove.
591 (vfp_parse_reg_list): New function.
592 (s_arm_unwind_fnstart, s_arm_unwind_fnend, s_arm_unwind_cantunwind,
593 s_arm_unwind_personality, s_arm_unwind_personalityindex,
594 s_arm_unwind_handlerdata, s_arm_unwind_save, s_arm_unwind_movsp,
595 s_arm_unwind_pad, s_arm_unwind_setfp, s_arm_unwind_raw): New
597 (md_pseudo_table): Add them.
598 (do_vfp_reg2_from_sp2): Use vfp_parse_reg_list and vfp_sp_encode_reg.
599 (do_vfp_sp2_from_reg2, vfp_sp_ldstm, vfp_dp_ldstm): Ditto.
600 (set_section, add_unwind_adjustsp, flush_pending_unwind,
601 finish_unwind_opcodes, start_unwind_section, create_unwind_entry,
602 require_hashconst, add_unwind_opcode): New functions.
603 * doc/tc-arm.text: Document unwinding opcodes.
604 * NEWS: Mention the new feature.
606 2004-10-04 Eric Christopher <echristo@redhat.com>
608 * config/tc-mips.c (md_apply_fix3): Remove erroneous assert.
610 2004-10-01 H.J. Lu <hongjiu.lu@intel.com>
612 * config/tc-ppc.c (md_apply_fix3): Call S_SET_THREAD_LOCAL for
614 * config/tc-s390.c (md_apply_fix3): Likewise.
615 * config/tc-sparc.c (md_apply_fix3): Likewise.
617 2004-10-01 Paul Brook <paul@codesourcery.com>
619 * config/tc-arm.c (arm_elf_section_type): New function.
620 (arm_elf_change_section): Set section link for exidx sections.
621 * config/tc-arm.h (arm_elf_section_type): Add prototype.
622 (md_elf_section_type): Define.
624 2004-10-01 Bill Farmer <Bill@the-farmers.freeserve.co.uk>
626 * config/tc-pdp11.c (md_apply_fix3): Change to sign of the SOB
627 instruction's offset.
629 2004-10-01 Adam Nemet <anemet@lnxw.com>
631 * (TARGET_FORMAT): Remove LynxOS COFF definition.
633 2004-10-01 Ravi Ramaseshan <ravi.ramaseshan@codito.com>
635 * config/tc-arc.c (tc_gen_reloc): Don't assume fixP->fx_addsy is an
636 asymbol *, instead use symbol_get_bfdsym.
638 2004-09-30 Linus Nielsen Feltzing <linus@haxx.se>
640 * config/tc-m68k.c (select_control_regs): Add mcf5249.
642 2004-09-30 Paul Brook <paul@codesourcery.com>
644 * config/tc-arm.c (do_smi, do_nop): New functions.
645 (insns): Add ARMv6ZK instructions.
646 (md_apply_fix3): Handle BFD_RELOC_ARM_SMI.
647 (tc_gen_reloc): Ditto.
648 (arm_cpus): Add mpcore and arm1176.
649 (arm_archs): Add armv6{k,z,zk}.
650 * doc/c-arm.texi: Document new cores and architectures.
652 2004-09-30 Nick Clifton <nickc@redhat.com>
654 * config/tc-arm.c: Use ISO C90 formatting.
656 2004-09-30 Vladimir Ivanov <vladitx@nucleusys.com>
658 * config/tc-arm.c (mav_reg_required_here): Allow REG_TYPE_CN
659 as alternative when REG_TYPE_MVF, REG_TYPE_MVD, REG_TYPE_MVFX or
660 REG_TYPE_MVDX is expected.
662 2004-09-29 Marc Bevand <m.bevand@gmail.com>
664 * doc/c-i386.texi (i386-Mnemonics): Fix typo.
666 2004-09-21 James E Wilson <wilson@specifixinc.com>
668 * config/tc-ia64.c (ENCODED_PSP_OFFSET): New.
669 (output_rp_psprel, output_pfs_psprel, output_preds_psprel,
670 output_spill_base, output_unat_psprel, output_lc_psprel,
671 output_fpsr_psprel, output_priunat_psprel, output_bsp_psprel,
672 output_bsprestore_psprel, output_rnat_psprel, output_spill_psprel,
673 output_spill_psprel_p): Use it.
675 2004-09-20 Tomer Levi <Tomer.Levi@nsc.com>
677 * config/tc-crx.c (handle_LoadStor): New function.
678 Handle load/stor unique instructions before parsing.
680 2004-09-17 Paul Brook <paul@codesourcery.com>
682 * config/tc-arm.c (s_arm_rel31): New funciton.
683 (md_pseudo_table): Add .rel31.
684 (md_apply_fix3): Handle BFD_RELOC_ARM_TARGET2,
685 BFD_RELOC_32_PCREL and BFD_RELOC_ARM_PREL31.
686 (tc_gen_reloc): Handle BFD_RELOC_ARM_PREL31 and BFD_RELOC_ARM_TARGET2.
687 (arm_fix_adjustable): Return 0 for BFD_RELOC_ARM_TARGET2.
688 (arm_parse_reloc): Add (target2).
690 2004-09-17 Alan Modra <amodra@bigpond.net.au>
692 * Makefile.am: Run "make dep-am".
693 * Makefile.in: Regenerate.
694 * aclocal.m4: Regenerate.
695 * configure: Regenerate.
696 * doc/Makefile.in: Regenerate.
697 * po/POTFILES.in: Regenerate.
698 * po/gas.pot: Regenerate.
700 2004-09-14 Hideki IWAMOTO <h-iwamoto@kit.hi-ho.ne.jp>
702 * config/tc-mmix.c [!LLONG_MIN]: Correct #elsif to #elif.
705 2004-09-13 Paul Brook <paul@codesourcery.com>
707 * config/tc-arm.c: Rename RELABS to TARGET1.
709 2004-09-13 Alan Modra <amodra@bigpond.net.au>
711 * messages.c (as_internal_value_out_of_range): Cast values passed
712 to as_bad_where or as_warn_where to proper type.
714 2004-09-11 Theodore A. Roth <troth@openavr.org>
716 * config/tc-avr.c: Add support for
717 atmega48, atmega88, atmega168, attiny13, attiny2313, at90can128.
719 2004-09-09 Alan Modra <amodra@bigpond.net.au>
721 * dw2gencfi.c (select_cie_for_fde): When separating CIE out
722 from FDE, treat a CFI_escape as we do a DW_CFA_advance_loc.
724 2004-09-08 Paul Brook <paul@codesourcery.com>
726 * config/obj-elf.c (obj_elf_section_type): Handle init_array,
727 fini_array and preinit_array section types.
728 * config/tc-ia64.c (ia64_elf_section_type): Remove init_array
730 * doc/as.texinfo: Document extra section types.
732 2004-09-02 Mark Mitchell <mark@codesourcery.com>
734 * Makefile.am (TARG_ENV_HFILES): Add te-symbian.h.
735 * Makefile.in: Regenerated.
736 * configure.in: Set em for arm*-*-symbianelf*.
737 * configure: Regenerated.
738 * config/tc-arm.c (elf32_arm_target_format): Use Symbian target
739 vectors when appropriate.
740 * config/te-symbian.h: New file.
742 2004-09-03 Tomer Levi <Tomer.Levi@nsc.com>
744 * config/tc-crx.c (gettrap): Exception vector can be case
746 (process_label_constant): Fix a 32-bit displacement bug in branch
748 (get_operandtype) : Bug fix, wrong operand was used.
749 (process_label_constant): Initialize relocation type to
752 2004-09-01 Richard Earnshaw < reanrsha@arm.com>
754 * tc-arm.c (arm_cpus, arm_fpus): Allow <cpu>-s as well as <cpu>s
755 for synthesizable cores.
757 * doc/c-arm.texi (ARM Options): Document canonical names of CPUs.
759 2004-08-25 Dmitry Diky <diwil@spec.ru>
761 * config/tc-msp430.c: Clean-up the code.
762 (md_relax_table): New relax table.
763 (mcu_types): Sort MCU types.
764 (md_pseudo_table): Add .profiler pseudo handler.
765 (pow2value): New function.
766 (msp430_profiler): New function.
767 (msp430_operands): Add new insns handlers.
768 (msp430_srcoperand): Add register operand handler, allow complex
770 (md_estimate_size_before_relax): Rewritten.
771 (md_convert_frag): Rewritten.
772 (msp430_relax_frag): New function.
773 * config/tc-msp430.h (md_relax_frag): define macro
774 * doc/c-msp430.texi: Update information.
776 2004-08-24 Nick Clifton <nickc@redhat.com>
778 * as.c (std_shortopts): Allow -g to take an optional argument.
779 (parse_args): Pass any switch starting with -g on to the backend
782 2004-08-18 Mark Mitchell <mark@codesourcery.com>
784 * configure.in (arm*-*-symbianelf*): New target.
785 (arm*-*-eabi*): Likewise.
786 * configure: Regenerated.
788 2004-08-18 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
789 * config/tc-mips.c (append_insn): Handle delay slots in branch likely
792 2004-08-18 Jakub Jelinek <jakub@redhat.com>
794 * config/tc-ia64.c (start_unwind_section): Add linkonce_empty
795 argument, don't do anything if current section is not
796 .gnu.linkonce.t.* and linkonce_empty is set.
797 (generate_unwind_image, dot_endp): Adjust callers, call
798 start_unwind_section (*, 1) if nothing will be put into the
801 2004-08-17 Nick Clifton <nickc@redhat.com>
803 * as.c (MD_DEBUG_FORMAT_SELECTOR): Provide default definition.
804 (show_usage): Add -g.
805 (std_longopts): Add --gen-debug. Alpha sort the table.
806 (parse_args): Print an error message if a switch is not handled.
807 Handle the -g switch, calling md_debug_format_selector() if
809 * NEWS: Mention new feature.
810 * doc/as.texinfo: Document new switch.
811 * doc/internals.texi: Document behaviour of md_parse_option.
813 * config/tc-arm.c (md_parse_option): Do not issue an error message
814 if the switch is not recognised.
815 * config/tc-m68k.c (md_parse_option): Likewise.
816 * config/tc-pdp11.c (md_parse_option): Likewise.
817 * config/tc-v850.c (md_parse_option): Likewise.
819 * as.h: Fix up formatting.
822 2004-08-16 Nick Clifton <nickc@redhat.com>
824 * macro.c (macro_set_alternate): Use ISO C90 formatting.
826 * configure.in: Sort architecture based tables alphabetically.
827 * configure: Regenerate.
829 2004-08-16 Alan Modra <amodra@bigpond.net.au>
831 * config/tc-ppc.c (tc_ppc_regname_to_dw2regnum <regnames>): Replace
832 { "cc", 68 }, with { "cr", 70 }.
834 2004-08-13 Jan Beulich <jbeulich@novell.com>
835 Nick Clifton <nickc@redhat.com>
837 * as.c: Add and handle new --alternate command line option.
838 * macro.c (macro_set_alternate): New.
839 * macro.h (macro_set_alternate): Declare.
840 * read.c: Add and handle new .altmacro and .noaltmacro directives.
841 * doc/as.texinfo: Document new command line option and pseudo-ops
842 as well as insert documentation originating from gasp about
843 alternate macro syntax.
844 * NEWS: Mention new command line option and pseudo-ops.
846 2004-08-10 Mark Mitchell <mark@codesourcery.com>
848 * expr.c (operand): Handle the "~", "-", and "!" operators applied
851 2004-08-06 Paul Brook <paul@codesourcery.com>
853 * config/tc-arm.c (md_apply_fix3, tc_gen_reloc, arm_parse_reloc):
854 Handle new relocations.
855 * include/elf/arm.h (elf_arm_reloc_type): Add new EABI relocations.
857 2004-08-05 Bob Wilson <bob.wilson@acm.org>
859 * write.c (relax_segment): Use was_address instead of address when
860 setting fr_fix field for align frag due to backwards .org.
862 2004-07-29 Alexandre Oliva <aoliva@redhat.com>
864 Introduce SH2a support.
865 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
866 * config/tc-sh.c (get_specific): Change arch_sh2a_up to
868 2004-02-24 Corinna Vinschen <vinschen@redhat.com>
869 * config/tc-sh.c (md_parse_option): Add sh2a-nofpu ISA handling.
870 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
871 * config/tc-sh.c (sh_elf_final_processing): Move sh2a recognition
872 to end of conditional expression.
873 2004-02-20 Corinna Vinschen <vinschen@redhat.com>
874 * config/tc-sh.c: Add sh2a-nofpu support.
875 2003-12-29 DJ Delorie <dj@redhat.com>
876 * tc-sh.c: Add sh2a support.
877 (parse_reg): Add tbr.
878 (parse_at): Support @@(disp,tbr).
879 (get_specific): Support sh2a opcodes.
880 (insert4): New, for 4 byte relocs.
881 (build_Mytes): Support sh2a opcodes.
882 (md_apply_fix3_Mytes): Support sh2a opcodes.
883 2003-12-02 Michael Snyder <msnyder@redhat.com>
884 * config/tc-sh.c (md_parse_option): Handle sh2a.
885 (sh_elf_final_processing): Ditto.
887 2004-07-27 Jason Thorpe <thorpej@wasabisystems.com>
889 * config/tc-hppa.h (TARGET_FORMAT): Set to "elf32-hppa-netbsd"
892 2004-07-27 Alan Modra <amodra@bigpond.net.au>
894 * config/tc-ppc.c (ppc_frob_file_before_adjust): Warn if .toc too big.
895 (ppc_arch): Expand comment.
897 2004-07-27 Tomer Levi <Tomer.Levi@nsc.com>
899 * config/tc-crx.c: Support evaluating the difference between two
901 * config/tc-crx.h: Likewise.
903 2004-07-26 H.J. Lu <hongjiu.lu@intel.com>
905 * config/tc-ia64.c (start_unwind_section): Set the linked-to
907 (ia64_elf_section_change_hook): Set the linked-to section for
910 2004-07-26 Dmitry Diky <diwil@spec.ru>
912 * config/tc-msp430.c: Add new subtargets: msp430x1610,
913 msp430x1611, msp430x1612, msp430x415, msp430x417, msp430xG437,
914 msp430xG438, msp430xG439.
916 2004-07-25 Daniel Jacobowitz <dan@debian.org>
918 * doc/as.texinfo (Section, PushSection): Correct documentation
921 2004-07-21 Jan Beulich <jbeulich@novell.com>
923 * config/tc-i386.c (optimize_imm): Adjust immediates to only those
924 permissible for the selected instruction suffix.
925 (match_template): Don't permit 64-bit general purpose operands in
927 (finalize_imm): Permit 64-bit immediates.
928 (build_modrm_byte): Don't treat 32-bit addressing in 64-bit mode
929 specially except for the width of the used base and/or index
930 registers. For 32-bit displacements, use sign-extended
931 relocations only when using 64-bit addressing.
932 Force zero displacement on rip-relative addressing when there is
933 no other displacement.
934 (i386_index_check): Don't treat 32-bit addressing in 64-bit mode
935 specially except for the width of the used base and/or index
937 (parse_register): Disallow Reg64 registers in 32-bit mode.
939 * config/tc-i386.c: For DefaultSize instructions, don't guess a 'q'
940 suffix if the instruction doesn't support it.
942 2004-07-20 Maciej W. Rozycki <macro@linux-mips.org>
944 * config/tc-mips.c (append_insn): Handle constant expressions with
945 no associated relocation.
946 (mips_ip): Cancel the expression after use for the Q format
948 (parse_relocation): Return no relocation for unsupported
950 (my_getSmallExpression): Return no relocation if no relocation
953 2004-07-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
955 * config/obj-som.c (adjust_stab_sections): Add prototype.
956 (obj_som_compiler, obj_som_version, obj_som_copyright,
957 adjust_stab_sections): Add ATTRIBUTE_UNUSED to unused arguments.
958 * config/tc-hppa.c (update_subspace): Likewise.
959 (is_defined_subspace): Amplify comment.
960 * config/obj-som.h (som_frob_file): Add prototype.
962 2004-07-19 Christopher Faylor <cgf@timesys.com>
963 H.J. Lu <hongjiu.lu@intel.com>
965 * subsegs.c (section_symbol): Don't create a new segment when
966 existing segment is undefined.
968 2004-07-16 Richard Earnshaw <rearnsha@arm.com>
970 * config/tc-arm.c: Include include/opcode/arm.h.
971 (ARM_EXT_*, ARM_ARCH_*, ARM_ANY, ARM_ALL, COPROC_ANY): Delete.
972 (FPU_FPA_EXT_* FPU_VFP_EXT_*, FPU_ANY, FPU_NONE, FPU_MAVERICK): Delete.
973 (FPU_ARCH_*): Delete.
974 * Makefile.am: Update dependencies.
975 * Makefile.in: Regenerate.
977 2004-07-15 Nick Clifton <nickc@redhat.com>
979 * configure.in: Accept armbe as a big-endian arm configuration.
980 * configure: Regenerate.
982 2004-07-13 Thomas Nystrom <thn@saeab.se>
984 * config/tc-i386.c (T_SHIFTOP): New constant.
985 (intel_e05_1): Handle '&', '|' and T_SHIFTOP.
986 (intel_el1): Handle '~'.
987 (intel_get_token): Handle '<>', '&', '|' and '~'.
989 2004-07-13 Nick Clifton <nickc@redhat.com>
991 (md_assemble): Remove spurious newline from end of as_bad error
993 (intel_e05_1): Likewise.
994 (intel_e11): Likewise.
995 (intel_match_token): Likewise.
997 2004-07-11 Andreas Schwab <schwab@suse.de>
999 * config/tc-m68k.c: Convert to C90. Remove redundant
1000 declarations. Indentation fixup.
1001 [M68KCOFF]: Include "obj-coff.h" instead of declaring
1002 obj_coff_section ourselves.
1004 2004-07-09 James E Wilson <wilson@specifixinc.com>
1006 * config/tc-ia64.c (default_big_endian): New.
1007 (dot_byteorder, md_begin): Use it.
1008 (md_parse_option): Set it.
1010 2004-07-09 Nick Clifton <nickc@redhat.com>
1012 * configure.in: Change sh-sybmian-elf to sh-*-symbianelf.
1013 * configure: Regenerate.
1014 * NEWS: Change sh-sybmian-elf to sh-*-symbianelf.
1015 * config/tc-sh.c (sh_elf_final_processing): Use renamed version of
1016 sh_find_elf_flags if necessary.
1018 2004-07-08 Richard Sandiford <rsandifo@redhat.com>
1020 * config/tc-mips.c (mips_fix_adjustable): If the full addend is
1021 going to be split into more than one in-place addend, return 0
1022 for relocations against mergeable sections. Associate comments
1025 2004-07-07 Tomer Levi <Tomer.Levi@nsc.com>
1027 * Makefile.am (CPU_TYPES): Add crx.
1028 (TARGET_CPU_CFILES): Add config/tc-crx.c.
1029 (TARGET_CPU_HFILES): Add config/tc-crx.h.
1030 (DEPTC_crx_elf): New target.
1031 (DEPOBJ_crx_elf): Likewise.
1032 (DEP_crx_elf): Likewise.
1033 * Makefile.in: Regenerate.
1034 * configure.in: Add crx* target.
1035 * configure: Regenerate.
1036 * config/tc-crx.c: New file.
1037 * config/tc-crx.h: New file.
1038 * NEWS: Mention new target.
1040 2004-07-06 Nick Clifton <nickc@redhat.com>
1042 * config.in: Undefine TARGET_SYMBIAN by default.
1044 * configure: Regenerate. Add sh-symbian-elf target. If
1045 selected define TARGET_SYMBIAN.
1046 * config/tc-sh.h (TARGET_FORMAT): Select a Symbian target
1047 format if TARGET_SYMBIAN has been defined.
1049 * output-file.c (output_file_create): Report the target format
1050 chosen when bfd_openw reports that it is invalid.
1052 * config/obj-coff.c (coff_pseudo_table): Only define the weak
1053 pseudo for BFD based assemblers.
1055 2004-07-05 Andrew Stubbs <andrew.stubbs@superh.com>
1058 * config/tc-sh.c (md_assemble): Change isspace to ISSPACE.
1059 (md_parse_option): Remove redundant -isa testing.
1060 Make bfd_arch variable const.
1061 (md_show_usage): Make bfd_arch variable const.
1063 2004-07-03 James E Wilson <wilson@specifixinc.com>
1065 * config/tc-ia64.c (emit_one_bundle): Check and set insn_addr.
1066 * config/tc-ia64.h (md_frag_check): Define.
1068 2004-07-03 Aaron W. LaFramboise <aaron98wiridge9@aaronwl.com>
1070 * config/obj-coff.c (obj_coff_weak): New .weak syntax for PE weak
1072 * doc/as.texinfo (Weak): Document PE weak symbols.
1074 2004-07-03 Richard Sandiford <rsandifo@redhat.com>
1076 * config/tc-mips.c (HAVE_IN_PLACE_ADDENDS): New macro.
1077 (reloc_needs_lo_p): Only return true if HAVE_IN_PLACE_ADDENDS.
1078 (mips_frob_file): Rework so that only a single pass through the
1079 relocs is needed. Allow %lo()s to have higher offsets than their
1080 corresponding %hi()s or %got()s.
1082 2004-07-02 Nick Clifton <nickc@redhat.com>
1084 * config/tc-arm.c (md_apply_fix3:BFD_RELOC_ARM_IMMEDIATE): Do not
1085 allow values which have come from undefined symbols.
1086 Always consider this fixup to have been processed as a reloc
1087 cannot be generated for it.
1089 2004-07-02 Alan Modra <amodra@bigpond.net.au>
1091 * frags.h (struct frag): Add has_code and insn_addr fields.
1092 * write.c (cvt_frag_to_fill): Invoke md_frag_check.
1093 * config/tc-ppc.c (md_assemble): Check and set insn_addr.
1094 * config/tc-ppc.h (md_frag_check): Define.
1096 2004-06-28 Maciej W. Rozycki <macro@linux-mips.org>
1098 * doc/Makefile.am (info): Rename goal to...
1099 (info-local): ... this, to preserve implicit dependencies.
1100 * doc/Makefile.in: Regenerate with automake 1.8.5.
1102 2004-06-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1104 * config/tc-m32r.c (md_convert_frag): Changed for @PLT.
1105 (m32r_cgen_record_fixup_exp): Changed for @GOTOFF, @GOT.
1106 (m32r_fix_adjustable): Changed for @GOTOFF, @GOT, @PLT.
1107 (tc_gen_reloc): Likewise.
1108 (m32r_end_of_match): Add for @GOTOFF, @GOT, @PLT.
1109 (m32r_parse_name): Likewise.
1110 (m32r_cgen_parse_fix_exp): Likewise.
1111 * config/tc-m32r.h (md_parse_name): Define for @GOTOFF, @GOT, @PLT.
1112 (O_PIC_reloc): Likewise.
1113 (TC_CGEN_PARSE_FIX_EXP): Likewise..
1114 * cgen.c (gas_cgen_parse_operand): Add TC_CGEN_PARSE_FIX_EXP
1115 for @GOTOFF, @GOT, @PLT.
1117 2004-06-21 Jan Beulich <jbeulich@novell.com>
1119 * gas/symbols.c: While discarding ordinary local absolute symbols
1120 when --strip-local-absolute is in effect, retain file symbols.
1122 2004-06-20 Andreas Schwab <schwab@suse.de>
1124 * config/tc-m68k.c (mri_chip): Replace current_chip, not augment.
1125 (md_parse_option): Likewise.
1127 2004-06-17 Jan Beulich <jbeulich@novell.com>
1129 * config/tc-i386.c: Deal with LEX_QM the same way as with LEX_AT.
1130 * config/te-netware.h: New file.
1131 * config/te-ppcnw.h: Delete: Obsolete.
1132 * configure.in: Eliminate ill NetWare targets. Make generic
1133 NetWare target use proper emulation.
1134 * Makefile.am: Eliminate reference to obsolete te-ppcnw.h, add
1135 reference to new te-netware.h.
1136 * configure: Regenerate.
1137 * Makefile.in: Regenerate.
1139 2004-06-15 Martin Schwidefsky <schwidefsky@de.ibm.com>
1141 * config/tc-s390.c (s390_insn): Avoid incorrect signed/unsigned
1142 comparison in .insn pseudo operation.
1144 2004-06-15 Alan Modra <amodra@bigpond.net.au>
1146 * config/obj-coff.c (coff_adjust_section_syms): Use
1147 bfd_get_section_size instead of bfd_get_section_size_before_reloc.
1148 (coff_frob_section): Likewise.
1149 * config/tc-mips.c (md_apply_fix3): Likewise.
1150 * config/obj-elf.c (elf_frob_file): Use bfd_set_section_size.
1151 (elf_frob_file_after_relocs): Likewise.
1153 2004-06-10 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1155 * config/tc-hppa.c (log2): Rename to exact_log2.
1156 (pa_next_subseg): Delete unused function.
1157 (create_new_space): Mark unused arguments with ATTRIBUTE_UNUSED.
1158 (create_new_subspace): Likewise.
1161 * config/tc-hppa.c (hppa_fix_adjustable): Allow reduction of fake
1162 labels. Fix warning.
1164 2004-05-28 DJ Delorie <dj@redhat.com>
1166 * config/tc-mn10300.h (tc_fix_adjustable): Define.
1167 * config/tc-mn10300.c (mn10300_fix_adjustable): Don't adjust debug
1168 or non-merged symbols.
1170 2004-05-28 H.J. Lu <hongjiu.lu@intel.com>
1172 * config/tc-ia64.c (remove_marked_resource): Save, clear and
1173 restore the old slot when inserting srlz.i/srlz.d.
1175 2004-05-28 Andrew Stubbs <andrew.stubbs@superh.com>
1177 * Makefile.am: Regenerate dependecies.
1178 * Makefile.in: Regenerate.
1179 * config/tc-sh.c (valid_arch): Make unsigned.
1180 (preset_target_arch): Likewise.
1181 (md_begin): Use new architecture flags system.
1182 (get_specific): Likewise.
1183 (assemble_ppi): Likewise.
1184 (md_assemble): Likewise. Also fix error check for bad opcodes.
1185 (md_parse_option): Likewise. Also generate -isa values according
1186 to the table in bfd/cpu-sh.c instead of just constants. Also
1187 allow <arch>-up ISA variants.
1188 (sh_elf_final_processing): Replace if-else chain with a call to
1189 sh_find_elf_flags().
1190 * testsuite/gas/sh/arch: New directory.
1191 * testsuite/gas/sh/arch/arch.exp: New test script.
1192 * testsuite/gas/sh/arch/arch_expected.txt: New file.
1193 * testsuite/gas/sh/arch/sh.s: New file.
1194 * testsuite/gas/sh/arch/sh2.s: New file.
1195 * testsuite/gas/sh/arch/sh-dsp.s: New file.
1196 * testsuite/gas/sh/arch/sh2e.s: New file.
1197 * testsuite/gas/sh/arch/sh3-nommu.s: New file.
1198 * testsuite/gas/sh/arch/sh3.s: New file.
1199 * testsuite/gas/sh/arch/sh3-dsp.s: New file.
1200 * testsuite/gas/sh/arch/sh3e.s: New file.
1201 * testsuite/gas/sh/arch/sh4-nommu-nofpu.s: New file.
1202 * testsuite/gas/sh/arch/sh4-nofpu.s: New file.
1203 * testsuite/gas/sh/arch/sh4.s: New file.
1204 * testsuite/gas/sh/arch/sh4a-nofpu.s: New file.
1205 * testsuite/gas/sh/arch/sh4al-dsp.s: New file.
1206 * testsuite/gas/sh/arch/sh4a.s: New file.
1208 2004-05-28 Peter Barada <peter@the-baradas.com>
1210 * config/gc-m68k.c(m68k_ip): Convert mode 5 addressing
1211 with zero offset into mode 2 addressing to save a word.
1213 2004-05-27 H.J. Lu <hongjiu.lu@intel.com>
1215 * config/tc-ia64.c (ar_is_in_integer_unit): Removed.
1216 (ar_is_only_in_integer_unit): New.
1217 (ar_is_only_in_memory_unit): New.
1218 (generate_unwind_image): Silence gcc on 32bit host.
1219 (md_assemble): Use ar_is_only_in_integer_unit instead of
1220 ar_is_in_integer_unit. Check AR access.
1222 2004-05-27 Peter Barada <peter@the-baradas.com>
1224 * config/tc-m68k.c (md_begin): Sort the opcode table into
1226 (m68k_compare_opcode): New function to do the sorting.
1228 2004-05-24 Peter Barada <peter@the-baradas.com>
1230 * config/m68k-parse.y(operand): Allow for MAC/EMAC mask
1231 addressing on MIT style operands.
1232 * config/m68k-parse.y(yylex): Allow '-&' for predecrement
1234 * config/tc-m68k.c(install_operand): Comment 'G' and 'H' type
1237 2004-05-23 Alan Modra <amodra@bigpond.net.au>
1239 * expr.c (operand, operator): Don't reject '++' and '--'.
1241 2004-05-20 Richard Sandiford <rsandifo@redhat.com>
1243 * config/tc-mips.c (append_insn): Use ISA-encoded addresses in MIPS16
1246 2004-05-17 Adam Nemet <anemet@lnxw.com>
1248 * configure.in: Add ppc-*-lynxos*. Update i386-*-lynxos* to ELF.
1249 * configure: Regenerate.
1251 2004-05-13 Paul Brook <paul@codesourcery.com>
1253 * dw2gencfi.c (output_cie): Handle dwarf3 format CIE entries.
1255 2004-05-13 Joel Sherrill <joel@oarcorp.com>
1257 * configure.in (or32-*-rtems*): Switch to elf.
1258 * configure: Regenerate.
1260 2004-05-13 Nick Clifton <nickc@redhat.com>
1262 * po/fr.po: Updated French translation.
1264 2004-05-11 Nick Clifton <nickc@redhat.com>
1266 * doc/as.texinfo (Section): Document G and T flags to .section
1267 directive. Document the extra arguments that the G flag
1268 requires. Document the #tls flag.
1270 2004-05-11 H.J. Lu <hongjiu.lu@intel.com>
1272 * subsegs.c (section_symbol): Create a new section symbol if
1273 the existing one doesn't match.
1274 * symbols.c (symbol_set_bfdsym): Don't reset BFD section symbol.
1276 2004-05-07 Richard Sandiford <rsandifo@redhat.com>
1278 * config/tc-mips.c (append_insn, mips_emit_delays): Extend -mfix-vr4120
1279 to cope with VR4181A errata MD(1) and MD(4).
1281 2004-05-07 Brian Ford <ford@vss.fsi.com>
1283 * NEWS: Mention .secrel32 for pe[i]-i386.
1285 2004-05-07 Alexandre Oliva <aoliva@redhat.com>
1287 * config/tc-frv.h (MAX_MEM_FOR_RS_ALIGN_CODE): New.
1288 (HANDLE_ALIGN): New.
1290 2004-05-06 Daniel Jacobowitz <dan@debian.org>
1292 * Makefile.am (DIST_SUBDIRS): Define.
1293 * aclocal.m4: Regenerate with automake 1.8.4.
1294 * Makefile.in: Likewise.
1295 * doc/Makefile.in: Likewise.
1297 2004-05-06 David Mosberger-Tang <davidm@hpl.hp.com>
1299 * config/tc-ia64.c (dot_serialize): Declare.
1300 (dot_serialize): New function.
1301 (md_pseudo_table): Add ".serialize.data" and
1302 ".serialize.instruction" directives.
1304 2004-05-06 Nick Clifton <nickc@redhat.com>
1306 * messages (as_internal_value_out_of_range): Print a message about
1307 a value being out of range. Be consistent about whether the
1308 values are printed in decimal or hexadecimal.
1309 (as_warn_value_out_of_range): Generate a warning message about an
1311 (as_bad_value_out_of_range): Generate an error message about an
1313 * as.h: Prototype the new functions.
1314 * config/tc-alpha.c (insert_operand): Use new function.
1315 * config/tc-arc.c (arc_insert_operand): Likewise.
1316 * config/tc-mn10200.c (mn10200_insert_operand): Likewise.
1317 * config/tc-mn10300.c (mn10300_insert_operand): Likewise.
1318 * config/tc-ppc.c (ppc_insert_operand): Likewise.
1319 * config/tc-s390.c (s390_insert_operand): Likewise.
1320 * config/tc-v850.c (v850_insert_operand): Likewise.
1322 2004-05-05 Alexandre Oliva <aoliva@redhat.com>
1324 * configure.in: Set em=linux for frv-*-*linux*.
1325 * configure: Rebuilt.
1326 * config/tc-frv.h (TARGET_FORMAT): Use elf32-frvfdpic if...
1327 (frv_md_fdpic_enabled): New.
1328 * config/tc-frv.c (frv_md_fdpic_enabled): New.
1329 (DEFAULT_FDPIC): New.
1330 (frv_flags): Use DEFAULT_FDPIC.
1331 (frv_pic_flag): Likewise.
1332 (OPTION_NOPIC): New.
1333 (md_longopts): Add -mnopic.
1334 (md_parse_option): Handle it.
1335 (md_show_usage): Add -mfdpic and -mnopic.
1337 2004-05-05 Peter Barada <peter@the-baradas.com>
1339 * config/tc-m68k.c: Add find_cf_chip to print list of valid
1340 chips for invalid coldfire instructions, rename selectors
1341 for ColdFire sub-variants, add 521x,5249,547x,548x and aliases,
1342 add current_chip to track which chip is referred to(including save/restore),
1343 use current_chip to select control registers, not current_arch.
1344 (md_show_usage): Add new chips.
1345 * doc/c-m68k.texi: Document new command line switches.
1347 2004-05-05 Jakub Jelinek <jakub@redhat.com>
1349 * tc-s390.h (md_do_align, HANDLE_ALIGN): Remove.
1350 (NOP_OPCODE): Define.
1351 (s390_align_code): Remove prototype.
1352 * tc-s390.c (s390_align_code): Remove.
1354 2004-05-04 H.J. Lu <hongjiu.lu@intel.com>
1356 * config/tc-ia64.c (make_unw_section_name): Removed.
1357 (start_unwind_section): New function.
1358 (generate_unwind_image): Take const segT instead of const
1360 (dot_handlerdata): Adjusted.
1361 (dot_endp): Likewise.
1363 2004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1365 * config/obj-elf.c (obj_elf_change_section): Allow the
1366 ".note.GNU-stack" section has SHF_EXECINSTR.
1368 2004-05-02 H.J. Lu <hongjiu.lu@intel.com>
1370 * config/obj-elf.c (get_section): Return bfd_boolean.
1371 (obj_elf_change_section): Call bfd_get_section_by_name_if
1372 instead of bfd_map_over_sections.
1374 2004-04-30 H.J. Lu <hongjiu.lu@intel.com>
1376 * config/obj-elf.c (get_section): New function.
1377 (obj_elf_change_section): Support multiple sections with same
1380 2004-04-30 Nick Clifton <nickc@redhat.com>
1382 * config/tc-arm.c (create_register_alias): Fix typo checking for
1383 case sensitive register aliases.
1384 (co_proc_number): Use error message string in all_reg_maps[]
1386 (cp_reg_required_here): Likewise.
1387 (fp_reg_required_here): Likewise.
1389 2004-04-29 Brian Ford <ford@vss.fsi.com>
1391 * dwarf2dbg.c (dwarf2_finish): Add SEC_DEBUGGING to section flags.
1393 2004-04-28 Chris Demetriou <cgd@broadcom.com>
1395 * config/tc-mips.c (HAVE_32BIT_ADDRESSES, append_insn, macro_build)
1396 (load_address, macro, mips_ip, md_parse_option)
1397 (mips_force_relocation, mips_validate_fix, md_apply_fix3)
1398 (s_change_sec, pic_need_relax, tc_gen_reloc): Remove all
1399 embedded-PIC handling, and update comments.
1400 (SWITCH_TABLE): Remove.
1401 * config/tc-mips.h (DIFF_EXPR_OK): Delete.
1402 (enum mips_pic_level): Remove EMBEDDED_PIC.
1403 (EXTERN_FORCE_RELOC): Remove embedded-PIC handling.
1404 (TC_FORCE_RELOCATION): Update comment.
1405 * ecoff.c (ecoff_build_lineno): Add comment about some code that
1406 might be safe to remove now that MIPS embedded-PIC is gone.
1408 2004-04-28 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1410 * config/obj-som.c (obj_som_init_stab_section): Add new arguments in
1411 call to obj_set_subsection_attributes.
1412 (obj_som_init_stab_section): Likewise.
1413 * config/tc-hppa.c (default_subspace_dict): Add comdat field.
1414 (pa_def_subspaces): Provide comdat default.
1415 (pa_subspace): Handle new "comdat" parameter. Set SEC_LINK_ONCE and
1416 not SEC_IS_COMMON if section is comdat, common or dup_common. Update
1417 calls to create_new_subspace and update_subspace to pass comdat flag.
1418 (create_new_subspace, update_subspace): Add new comdat argument. Use
1419 it in calls to obj_set_subsection_attributes.
1420 * doc/c-hppa.texi (.subspa, .nsubspa): Document new comdat parameter
1421 and use of comdat, common and dup_comm parameters.
1423 2004-04-26 H.J. Lu <hongjiu.lu@intel.com>
1425 * config/obj-elf.c (obj_elf_change_section): Check if the old
1426 group name is NULL before comparison.
1428 2004-04-23 Chris Demetriou <cgd@broadcom.com>
1430 * config/tc-mips.h (mips_dwarf2_addr_size): Prototype.
1432 2004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1434 * config/tc-mips.c (s_mipsset): Set default CPU type for .set mipsN.
1436 2004-04-23 Chris Demetriou <cgd@broadcom.com>
1438 * config/tc-mips.c (md_longopts): Remove -membedded-pic option.
1439 (OPTION_MEMBEDDED_PIC): Remove.
1440 (OPTION_TRAP, OPTION_BREAK, OPTION_EB, OPTION_EL)
1441 (OPTION_FP32, OPTION_GP32, OPTION_CONSTRUCT_FLOATS)
1442 (OPTION_NO_CONSTRUCT_FLOATS, OPTIONS_FP64, OPTION_GP64)
1443 (OPTION_RELAX_BRANCH, OPTION_NO_RELAX_BRANCH)
1444 (OPTION_ELF_BASE): Renumber.
1445 (md_parse_option): Remove OPTION_MEMBEDDED_PIC handling.
1446 (md_show_usage): Remove mention of -membedded-pic.
1447 * doc/as.texinfo: Remove mention of -membedded-pic.
1449 2004-04-23 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1451 * config/tc-mips.h (USE_GLOBAL_POINTER_OPT): Remove.
1452 * config/tc-mips.c (RDATA_SECTION_NAME, mips_target_format): Remove
1454 (md_begin, mips_ip, md_parse_option, s_change_sec, s_option,
1455 s_abicalls, nopic_need_relax, tc_gen_reloc): Remove uses of
1456 USE_GLOBAL_POINTER_OPT.
1458 2004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1460 * config/tc-mips.c (macro): One more use of load_delay_nop.
1462 2004-04-22 Atsushi Nemoto <anemo@mba.ocn.ne.jp>
1464 * config/tc-mips.c (load_delay_nop): New function.
1465 (load_address, macro): Use load_delay_nop() to build a nop
1466 which can be omitted with gpr_interlocks.
1468 2004-04-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1470 * config/tc-mips.c (hilo_interlocks, gpr_interlocks,
1471 cop_interlocks): Remove superfluous CPU entries.
1473 2004-04-22 Paul Brook <paul@codesourcery.com>
1475 * config/tc-arm.c (mav_parse_offset): Value must be multiple of 4.
1477 2004-04-22 Peter Barada <peter@the-baradas.com>
1479 * NEWS: Added support for EMAC instructions and MAC/EMAC
1481 * config/m68k-parse.h: Add ACC[123], ACCEXT{01,23}, MAC/EMAC
1482 scale factor tokens, trailing_ampersand to mark mask addressing
1483 for MAC/EMAC instructions.
1484 * config/m68k-parse.y: Add options_ampersand clause, '<<',
1486 (yylex): Handle '>', '<', and '&' following '+'.
1487 * config/tc-m68k.c: Set mcfmac/mcfemac on appropriate ColdFire
1488 architectures in archs[].
1489 (m68k-ip): Add '4', 'e', 'g', 'i', cases to handle mask addressing
1490 for MAC/EMAC instructions, ACC[0123], ACCEXT{01,23}, and '<<'/'>>'
1492 (m68k_ip): Handle trailing '&' on MAC/EMAC insns.
1493 (install_operand): Fix 'n' case, Add 'F', 'f', 'G', 'H', 'I', ']'
1495 Add EMAC operands to init_table[].
1497 2004-04-22 Bruno De Bus <bdebus@elis.ugent.be>
1499 * config/tc-arm.h (enum mstate): Move here, add MAP_UNDEFINED
1501 (TC_SEGMENT_INFO_TYPE): Define to enum mstate.
1502 * config/tc-arm.c (enum mstate): Delete from here.
1503 (mapping_state): Remove the static mapstate variable and instead
1504 store the state in the segment. This allows a per-section mapping
1505 state. Handle and ignore MAP_UNDEFINED states.
1506 (arm_elf_change_section): Get the current mapping state from the
1508 (s_ltorg): Set the mapping state to MAP_DATA.
1509 (arm_cleanup): Use arm_elf_change_section to get the mapping state
1510 for each pool as it is emitted.
1512 2004-04-22 Nick Clifton <nickc@redhat.com>
1514 * config/tc-arm.h: Formatting tidy ups.
1516 2004-04-20 Chris Demetriou <cgd@broadcom.com>
1518 * NEWS: Note that MIPS -membedded-pic option is deprecated.
1520 2004-04-20 DJ Delorie <dj@redhat.com>
1522 * config/tc-i386.h [TE_PE] (TC_CONS_FIX_NEW): Define.
1523 * config/tc-i386.c (md_pseudo_table) [TE_PE]: Add "secrel32".
1524 [TE_PE] (O_secrel): Define.
1525 [TE_PE] (x86_pe_cons_fix_new): New.
1526 [TE_PE] (pe_directive_secrel): Likewise.
1527 (tc_gen_reloc) [TE_PE]: Support BFD_RELOC_32_SECREL.
1529 2004-04-19 Eric Christopher <echristo@redhat.com>
1531 * config/tc-mips.c (mips_dwarf2_addr_size): Revert part
1532 of previous patch for fix in gcc.
1534 2004-04-19 Jakub Jelinek <jakub@redhat.com>
1536 * config/tc-xtensa.c (xg_assembler_literal): Fix a typo.
1538 2004-04-19 Nathan Sidwell <nathan@codesourcery.com>
1540 * read.c (do_align): Call md_flush_pending_output, if defined.
1542 2004-04-16 Alan Modra <amodra@bigpond.net.au>
1544 * expr.c (operand): Correct checks for ++ and --.
1546 2004-04-14 H.J. Lu <hongjiu.lu@intel.com>
1548 * config/tc-generic.c: Add some comments.
1550 2004-04-14 Richard Sandiford <rsandifo@redhat.com>
1552 * doc/c-mips.texi (-m{no-,}fix-vr4120): Renamed from
1553 -{no-}mfix-vr4122-bugs.
1554 * config/tc-mips.c (mips_fix_vr4120): Renamed from mips_fix_4122_bugs.
1555 (append_insn, mips_emit_delays): Update accordingly.
1556 (OPTION_FIX_VR4120, OPTION_NO_FIX_VR4120): Renamed from *VR4122.
1557 (md_longopts): Change -{no-,}mfix-vr4122-bugs to -m{no-,}fix-vr4120.
1558 (md_parse_option): Update after above changes.
1559 (md_show_usage): Add -mfix-vr4120.
1561 2004-04-13 Bob Wilson <bob.wilson@acm.org>
1563 * doc/as.texinfo (Sub-Sections): Conditionalize COFF-specific use
1564 of .section directive; add a reference to the ELF .subsection
1567 2004-04-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1569 * config/tc-m32r.c (md_assemble): Fixed infinite loop bug
1572 2004-04-11 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
1574 * Makefile.am: Remove mips from aout targets.
1575 * Makefile.in: Regenerate.
1576 * configure.in: Remove mips-dec-bsd* target.
1577 * configure: Regenerate.
1579 2004-04-07 Alan Modra <amodra@bigpond.net.au>
1582 * config/tc-ppc.c (ppc_elf_suffix): Add valid32 and valid64 fields
1583 to struct map_bfd. Adjust MAP macro, and define MAP32, MAP64.
1584 Update "mapping". Restrict some @ modifiers to 32 bit.
1586 2004-04-01 Asgari Jinia <asgarij@kpitcummins.com>
1587 Dhananjay Deshpande <dhananjayd@kpitcummins.com>
1589 * config/tc-sh.c (dont_adjust_reloc_32): New variable.
1590 (sh_fix_adjustable): Avoid adjusting BFD_RELOC_32 when
1591 dont_adjust_reloc_32 is set.
1592 (md_longopts): Add option -renesas.
1593 (md_parse_option, md_show_usage): Likewise.
1594 * doc/c-sh.texi: Likewise.
1596 2004-04-01 Dave Korn <dk@artimi.com>
1598 * config/tc-dlx.c (md_assemble): set fx_no_overflow flag for
1599 hi16 and lo16 fixS structs.
1600 (md_assemble): generate bit_fixS for RELOC_DLX_LO16 in
1601 exactly the same way as for RELOC_DLX_REL16.
1602 (machine_ip): properly respect LO flag in the_insn and
1603 output RELOC_DLX_LO16 rather than RELOC_DLX_16.
1604 (md_apply_fix3): apply RELOC_DLX_LO16.
1606 2004-03-30 Stan Shebs <shebs@apple.com>
1608 Remove long-obsolete MPW support.
1609 * mpw-config.in, mpw-make.sed, mac-as.r: Remove files.
1610 * configure.in: Remove mention of ppc-*-mpw* config.
1611 * configure.in: Likewise.
1613 2004-03-30 Nick Clifton <nickc@redhat.com>
1615 * config/tc-arm.c (meabi_flags): Make its use conditional upon
1616 OBJ_ELF being defined.
1618 2004-03-27 Alan Modra <amodra@bigpond.net.au>
1620 * config/obj-aout.c (obj_aout_type): Remove #ifdef BFD_ASSEMBLER code.
1622 2004-03-23 Paul Brook <paul@codesourcery.com>
1624 * config/tc-arm.c (meabi_flags): New variable.
1625 (arm_parse_eabi): New function.
1626 (md_begin): Set flags for EABI v3.
1628 (arm_long_opts): Add meabi.
1629 * doc/as.texinf <ARM>: Document -meabi.
1630 * doc/c-arm.texi: Ditto.
1632 2004-03-22 Bob Wilson <bob.wilson@acm.org>
1634 * config/tc-xtensa.c (xtensa_post_relax_hook): Create literal
1635 tables even when use_literal_section flag is not set.
1637 2004-03-22 Alan Modra <amodra@bigpond.net.au>
1639 * config/tc-sh.c: Remove trailing whitespace.
1641 2004-03-22 Hans-Peter Nilsson <hp@axis.com>
1643 * doc/c-cris.texi (CRIS-Opts): Document --no-mul-bug-abort,
1644 --mul-bug-abort and the default behavior.
1645 * config/tc-cris.c (cris_insn_kind): New member CRIS_INSN_MUL.
1646 (err_for_dangerous_mul_placement): New variable.
1647 (STATE_MUL, OPTION_MULBUG_ABORT_ON, OPTION_MULBUG_ABORT_OFF): New
1649 (md_cris_relax_table): Have placeholder for STATE_MUL.
1650 (md_longopts): New options --mul-bug-abort and --no-mul-bug-abort.
1651 (cris_relax_frag) <case ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: New
1653 (md_estimate_size_before_relax) <case ENCODE_RELAX (STATE_MUL,
1654 STATE_BYTE)>: Ditto.
1655 (md_convert_frag) <ENCODE_RELAX (STATE_MUL, STATE_BYTE)>: Check
1656 alignment and position of this frag, emit error message if
1658 (md_assemble): For a multiply insn and when checking it,
1659 transform the current frag into a special frag for that purpose.
1660 (md_parse_option) <case OPTION_MULBUG_ABORT_OFF, case
1661 OPTION_MULBUG_ABORT_ON>: Handle new options.
1663 2004-03-19 Bob Wilson <bob.wilson@acm.org>
1665 * config/tc-xtensa.c (mark_literal_frags): New function.
1666 (xtensa_move_literals): Call mark_literal_frags for all literal
1667 segments, including init and fini literal segments.
1668 (xtensa_post_relax_hook): Swap use of xt_insn_sec and xt_literal_sec.
1670 2004-03-19 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1672 * tc-hppa.c (cons_fix_new_hppa): Check for PC relative base type.
1673 (pa_comm): Set BSF_OBJECT in symbol flags.
1675 2004-03-19 Alan Modra <amodra@bigpond.net.au>
1677 * Makefile.am: Run "make dep-am".
1678 * Makefile.in: Regenerate.
1679 * doc/Makefile.in: Regenerate.
1680 * config.in: Regenerate.
1681 * po/gas.pot: Regenerate.
1683 2004-03-18 Nathan Sidwell <nathan@codesourcery.com>
1685 * read.c (read_a_source_file): Use demand_empty_rest_of_line.
1686 (demand_empty_rest_of_line): Issue an error here.
1687 (ignore_rest_of_line): Silently skip to end.
1688 (demand_copy_string): Issue an error, not warning.
1690 * config/obj-elf.c (obj_elf_section_name): Likewise.
1691 (obj_elf_section): Likewise.
1692 * config/tc-arc.c (arc_extoper): Remove bogus NULL checks.
1693 (arc_extinst): Likewise.
1694 * config/tc-ia64.c (dot_saveb): Use demand_empty_rest_of_line.
1695 (dot_spill): Likewise.
1696 (dot_unwabi): Likewise.
1697 (dot_prologue): Likewise.
1699 * expr.c (operand): Reject ++ and --.
1700 (operator): Likewise.
1702 2004-03-17 Kaz Kojima <kkojima@rr.iij4u.or.jp>
1704 * config/tc-sh.c: Include dw2gencfi.h.
1705 (sh_cfi_frame_initial_instructions): New function.
1706 (sh_regname_to_dw2regnum): Likewise.
1707 * config/tc-sh.h (DWARF2_LINE_MIN_INSN_LENGTH): Move to the end of
1709 (TARGET_USE_CFIPOP): Define.
1710 (tc_cfi_frame_initial_instructions): Likewise.
1711 (tc_regname_to_dw2regnum): Likewise.
1712 (DWARF2_DEFAULT_RETURN_COLUMN, DWARF2_CIE_DATA_ALIGNMENT): Likewise.
1713 * Makefile.am: Update dependencies.
1714 * Makefile.in: Regenerate.
1716 2004-03-17 Ralf Corsepius <corsepiu@faw.uni-ulm.de>
1718 * configure.in: Switch sh-*-rtems* to ELF. Add sh-*-rtemscoff*.
1719 * configure: Regenerate.
1721 2004-03-12 Bob Wilson <bob.wilson@acm.org>
1723 * read.c (s_leb128): Call md_flush_pending_output.
1725 2004-03-12 Michal Ludvig <mludvig@suse.cz>
1727 * config/tc-i386.c (output_insn): Handle PadLock instructions.
1728 * config/tc-i386.h (CpuPadLock): New define.
1729 (CpuUnknownFlags): Added CpuPadLock.
1731 2004-03-07 Andreas Schwab <schwab@suse.de>
1733 * doc/c-hppa.texi (HPPA Directives): Fix typo.
1735 2004-03-07 Richard Henderson <rth@redhat.com>
1737 * dw2gencfi.c (output_cie): Align length to 4 byte boundary.
1738 (cfi_finish): Likewise for fde.
1740 2004-03-05 H.J. Lu <hongjiu.lu@intel.com>
1742 * config/tc-ia64.c (md_assemble): Properly handle NULL
1744 (ia64_handle_align): Don't abort if failed to add a stop bit.
1746 2004-03-04 H.J. Lu <hongjiu.lu@intel.com>
1748 * Makefile.in: Regenerated.
1749 * aclocal.m4: Likewise.
1750 * configure: Likewise.
1751 * doc/Makefile.in: Likewise.
1753 2004-03-03 H.J. Lu <hongjiu.lu@intel.com>
1755 * config/tc-ia64.c (dot_align): New.
1756 (ia64_do_align): Make it static.
1757 (md_pseudo_table): Use "dot_align" for "align".
1758 (ia64_md_do_align): Don't set align_frag here.
1759 (ia64_handle_align): Add a stop bit to the previous bundle if
1762 * config/tc-ia64.h (ia64_do_align): Removed.
1764 2003-03-03 Andrew Stubbs <andrew.stubbs@superh.com>
1766 * config/tc-sh.c (md_parse_option): Add -isa=sh4-nofpu and
1767 -isa=sh4-nommu-nofpu options. Adjust help messages accordingly.
1768 (sh_elf_final_processing): Output BFD type sh4_nofpu if that is
1769 the most general type or the user specifically requested it.
1770 (md_assemble): Add a new error message for when an instruction
1771 is understood, but is not allowed due to an -isa option.
1773 2004-03-02 H.J. Lu <hongjiu.lu@intel.com>
1775 * config/tc-ia64.c (align_frag): New.
1776 (md_assemble): Set the tc_frag_data field in align_frag for
1777 IA64_OPCODE_FIRST instructions.
1778 (ia64_md_do_align): Set align_frag.
1779 (ia64_handle_align): Add a stop bit if needed.
1781 * config/tc-ia64.h (TC_FRAG_TYPE): New.
1782 (TC_FRAG_INIT): New.
1784 2004-03-01 Richard Sandiford <rsandifo@redhat.com>
1786 * config/tc-frv.c (fr400_audio): New variable.
1787 (md_parse_option, md_show_usage): Add -mcpu=fr405 and -mcpu=fr450.
1788 (md_parse_option): Set fr400_audio for -mcpu=fr400 and -mcpu=fr405.
1789 (target_implements_insn_p): New function.
1790 (md_assemble): Report an error if the processor doesn't implement
1793 2004-02-27 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1795 * config/tc-m32r.c (md_longopts): Added -no-bitinst option.
1796 (md_parse_option): Ditto.
1797 (OPTION_NO_SPECIAL_M32R): Added.
1798 (md_show_usage): Document it.
1799 (enable_speial_m32r): Changed a default value from 0 to 1.
1800 * doc/c-m32r.texi: Document the -no-bitinst option.
1802 2004-02-27 Nick Clifton <nickc@redhat.com>
1804 * config/tc-sh.c (get_operand): Revert previous delta.
1805 (tc_gen_reloc): Check for an unknown reloc type before processing
1808 2004-02-27 Hannes Reinecke <hare@suse.de>
1810 * config/tc-s390.c (s390_insn): Correct range check for opcode in
1811 .insn pseudo operation.
1813 2004-02-27 Anil Paranjpe <anilp1@kpitcummins.com>
1815 * config/tc-sh.c (get_operand): In case of #Imm, check has been
1816 added for wrong syntax.
1818 2004-02-26 Eric Christopher <echristo@redhat.com>
1820 * config/tc-mips.c (mips_dwarf2_addr_size): New.
1821 * config/tc-mips.h (DWARF2_ADDR_SIZE): Use.
1823 2004-02-26 Andrew Stubbs <andrew.stubbs@superh.com>
1825 * config/tc-sh.c (build_Mytes): Add REG_N_D and REG_N_B01
1826 nibble types to assembler.
1828 2004-02-25 Fred Fish <fnf@redhat.com>
1830 * config/tc-iq2000.c: Add missing \n\ in multiline string literal.
1832 2004-02-20 James E Wilson <wilson@specifixinc.com>
1834 * config/tc-ia64.c (slot_index): New arg before_relax. Use instead of
1836 (fixup_unw_records): New arg before_relax. Pass to slot_index.
1837 (ia64_estimate_size_before_relax): New.
1838 (ia64_convert_frag): Pass 0 to fixup_unw_records. Add comment.
1839 (generate_unwind_image): Pass 1 to fixup_unw_records.
1840 * config/tc-ia64.h (ia64_estimate_size_before_relax): Declare.
1841 (md_estimate_size_before_relax): Call ia64_estimate_size_before_relax.
1843 2004-02-19 Jakub Jelinek <jakub@redhat.com>
1845 * stabs.c (generate_asm_file): Avoid warning about use of
1846 uninitialized variable.
1848 2004-02-18 David Mosberger <davidm@hpl.hp.com>
1850 * config/tc-ia64.c (ia64_flush_insns): In addition to prologue,
1851 body, and endp, allow unwind records which do not have a "t"
1852 (time/instruction) field.
1854 2004-02-17 Petko Manolov <petkan@nucleusys.com>
1856 * config/tc-arm.c (do_mav_dspsc_1): Correct offset of CRn.
1857 (do_mav_dspsc_2): Likewise.
1858 Fix accumulator registers move opcodes.
1860 2004-02-13 Hannes Reinecke <hare@suse.de>
1861 Jakub Jelinek <jakub@redhat.com>
1863 * dwarf2dbg.c (get_filenum): Do not read beyond allocated memory.
1865 2004-02-10 Steve Ellcey <sje@cup.hp.com>
1867 * config/tc-ia64.h (ia64_frob_symbol): New declaration.
1868 (tc_frob_symbol): New macro definition.
1869 * config/tc-ia64.c (ia64_frob_symbol): New routine.
1871 2004-02-09 Daniel Jacobowitz <drow@mvista.com>
1873 * config/tc-arm.c (md_begin): Mark .note.gnu.arm.ident as
1876 2004-02-09 Nathan Sidwell <nathan@codesourcery.com>
1878 * read.h (IGNORE_OPCODE_CASE): Do not define. Replace with ...
1879 (TC_CASE_SENSITIVE): ... this.
1880 * read.c: Replace IGNORE_OPCODE_CASE with TC_CASE_SENSITIVE.
1881 * doc/internals.texi (TC_CASE_SENSITIVE): Document.
1883 2004-02-06 James E Wilson <wilson@specifixinc.com>
1885 * config/tc-ia64.c (dot_endp): Delete call to output_endp.
1886 (generate_unwind_image): Re-add it here.
1888 2004-02-06 Nathan Sidwell <nathan@codesourcery.com>
1890 * dwarf2dbg.c (DWARF2_ADDR_SIZE): Remove trailing ';'
1891 * read.h (SKIP_WHITESPACE): Turn into an expression.
1892 * read.c (read_a_source_file): A pseudo is removed by having a
1895 2004-02-05 James E Wilson <wilson@specifixinc.com>
1897 * config/tc-ia64.c (output_endp): New.
1898 (count_bits): Delete.
1899 (ia64_flush_insns, process_one_record, optimize_unw_records): Handle
1900 endp unwind records.
1901 (fixup_unw_records): Handle endp unwind records. Delete code for
1902 shortening prologue regions not followed by a body record.
1903 (dot_endp): Call add_unwind_entry to emit endp unwind record.
1904 * config/tc-ia64.h (unw_record_type): Add endp.
1906 2004-02-03 James E Wilson <wilson@specifixinc.com>
1908 * config/tc-ia64.c (ia64_convert_frag): Call md_number_to_chars to
1909 fill padding bytes with zeroes.
1910 (emit_one_bundle): New locals last_ptr, end_ptr. Rewrite code that
1911 sets unwind_record slot_number and slot_frag fields.
1913 2004-02-02 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
1915 * config/tc-mips.c (add_got_offset_hilo): New function.
1916 (macro): Use load_register() and add_got_offset_hilo() to load
1917 constants instead of hardcoding code sequences throughout.
1919 2004-01-28 H.J. Lu <hongjiu.lu@intel.com>
1921 * config/tc-ia64.c (emit_one_bundle): Add proper indentation.
1923 2004-01-26 Bernardo Innocenti <bernie@develer.com>
1925 * config/tc-m68k.h (EXTERN_FORCE_RELOC): Handle m68k-uclinux specially,
1927 * config/tc-m68k.c (RELAXABLE_SYMBOL): Use EXTERN_FORCE_RELOC instead
1928 of hard-coded test for TARGET_OS=elf.
1930 2004-01-24 Chris Demetriou <cgd@broadcom.com>
1932 * config/tc-mips.c (hilo_interlocks): Change definition
1933 so that MIPS32, MIPS64 and later ISAs are included, along with
1934 the already-included machines. Update comments.
1936 2004-01-23 Daniel Jacobowitz <drow@mvista.com>
1938 * config/tc-arm.c (tc_gen_reloc): Improve error message for
1939 undefined local labels.
1941 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
1943 * config/tc-mips.c (load_address, macro): Update comments about
1944 NewABI GP relaxation.
1946 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
1948 * config/tc-mips.c (macro_build): Remove place and counter arguments.
1949 (mips_build_lui, macro_build_ldst_constoffset): Likewise.
1950 (mips16_macro_build, macro_build_jalr): Remove counter argument.
1951 (set_at, load_register, load_address, move_register): Likewise.
1952 (load_got_offset, add_got_offset): Likewise.
1953 Update all calls and tidy accordingly.
1955 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
1957 * config/tc-mips.c (RELAX_ENCODE): Remove WARN argument.
1958 (RELAX_FIRST, RELAX_SECOND): Turn into 8-bit quantities.
1959 (RELAX_USE_SECOND): Bump to 0x10000.
1960 (RELAX_SECOND_LONGER, RELAX_NOMACRO, RELAX_DELAY_SLOT): New flags.
1961 (mips_macro_warning): New variable.
1962 (md_assemble): Wrap macro expansion in macro_start() and macro_end().
1963 (s_cpload, s_cpsetup, s_cprestore, s_cpreturn): Likewise.
1964 (relax_close_frag): Set mips_macro_warning.first_frag. Adjust use
1966 (append_insn): Update mips_macro_warning.sizes.
1967 (macro_start, macro_warning, macro_end): New functions.
1968 (macro_build): Don't emit warnings here.
1969 (macro_build_lui, md_estimate_size_before_relax): ...or here.
1970 (md_convert_frag): Check for cases where one macro alternative
1971 needs a warning and the other doesn't. Emit a warning if the
1972 longer sequence was chosen.
1974 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
1976 * config/tc-mips.h (tc_frag_data_type, TC_FRAG_TYPE): Remove.
1977 * config/tc-mips.c (RELAX_ENCODE): Take three arguments: the size of
1978 the first sequence, the size of the second sequence, and a flag
1979 that says whether we should warn.
1980 (RELAX_OLD, RELAX_NEW, RELAX_RELOC[123]): Delete.
1981 (RELAX_FIRST, RELAX_SECOND): New.
1982 (mips_relax): New variable.
1983 (relax_close_frag, relax_start, relax_switch, relax_end): New fns.
1984 (append_insn): Remove "place" argument. Use mips_relax.sequence
1985 rather than "place" to check whether we're expanding the second
1986 alternative of a relaxable macro. Remove redundant check for
1987 branch relaxation. If generating a normal insn, and there
1988 is not enough room in the current frag, call relax_close_frag()
1989 to close it. Update mips_relax.sizes[]. Emit fixups for the
1990 second version of a relaxable macro. Record the first relaxable
1991 fixup in mips_relax. Remove tc_gen_reloc workaround.
1992 (macro_build): Remove all uses of "place". Use mips_relax.sequence
1993 in the same way as in append_insn.
1994 (mips16_macro_build): Remove "place" argument.
1995 (macro_build_lui): As for macro_build. Don't drop the add_symbol
1996 when generating the second version of a relaxable macro.
1997 (load_got_offset, add_got_offset): New functions.
1998 (load_address, macro): Use new relaxation machinery. Remove
1999 tc_gen_reloc workarounds.
2000 (md_estimate_size_before_relax): Set RELAX_USE_SECOND if the second
2001 version of a relaxable macro is needed. Return -RELAX_SECOND if the
2002 first version is needed.
2003 (tc_gen_reloc): Remove relaxation handling.
2004 (md_convert_frag): Go through the fixups for a relaxable macro and
2005 mark those that belong to the unneeded alternative as done. If the
2006 second alternative is needed, adjust the fixup addresses to account
2007 for the deleted first alternative.
2009 2004-01-23 Richard Sandiford <rsandifo@redhat.com>
2011 * frags.h (frag_room): Declare.
2012 * frags.c (frag_room): New function.
2013 * doc/internals.texi: Document it.
2015 2004-01-22 Thiemo Seufer <seufer@csv.ica.uni-stuttgart.de>
2017 * config/tc-mips.c (append_insn): Don't do r3900 interlock
2018 optimization for -mtune=r3900, as this will break on other CPUs.
2020 2004-01-11 Tom Rix <tcrix@worldnet.att.net>
2022 * config/tc-m68hc11.c (build_indexed_byte): movb and movw cannot
2023 be relaxed, use fixup.
2024 (md_apply_fix3): Use 5 bit reloc from movb and movw fixup.
2026 2004-01-19 Jakub Jelinek <jakub@redhat.com>
2028 * config/tc-sparc.c (sparc_ip): Disallow %f32-%f63 for single
2031 2004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2033 * config/tc-mips.c (append_insn): Properly detect variant frags
2034 that preclude swapping of relaxed branches. Correctly swap
2035 instructions between frags when dealing with relaxed branches.
2037 2004-01-14 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2039 * acinclude.m4: Quote names of macros to be defined by AC_DEFUN
2041 * aclocal.m4: Regenerate.
2042 * configure: Regenerate.
2044 2004-01-12 Anil Paranjpe <anilp1@KPITCummins.com>
2046 * config/tc-h8300.c (build_bytes): Apply relaxation to bit
2049 2004-01-12 Richard Sandiford <rsandifo@redhat.com>
2051 * config/tc-mips.c (macro_build_jalr): When adding an R_MIPS_JALR
2052 reloc, reserve space for the delay slot as well as the jalr itself.
2054 2004-01-09 Paul Brook <paul@codesourcery.com>
2056 * config/tc-arm.c (do_vfp_reg2_from_sp2): Rename from do_vfp_sp_reg2.
2057 (do_vfp_sp2_from_reg2): New function.
2059 (do_vfp_dp_from_reg2): Check return values properly.
2061 2004-01-08 Ian Lance Taylor <ian@wasabisystems.com>
2063 * config/tc-mips.c (warn_nops): Remove static variable.
2064 (macro): Remove test of warn_nops.
2065 (md_shortops): Remove 'n'.
2066 (md_parse_option): Remove 'n' case.
2067 (md_show_usage): Remove -n.
2068 * doc/as.texinfo (Overview): Remove MIPS -n option.
2069 * doc/c-mips.texi (MIPS Opts): Remove mention -n.
2070 * NEWS: Mention removal of MIPS -n option.
2072 * config/tc-mips.c (ISA_HAS_COPROC_DELAYS): Remove.
2073 (cop_interlocks): Check ISA level.
2074 (cop_mem_interlocks): Define.
2075 (reg_needs_delay): Check cop_interlocks rather than
2076 ISA_HAS_COPROC_DELAYS.
2077 (append_insn): Likewise. Use cop_mem_interlocks rather than
2078 directly checking mips_opts.isa.
2079 (mips_emit_delays): Likewise.
2081 2004-01-07 H.J. Lu <hongjiu.lu@intel.com>
2083 * config/tc-ia64.c (unwind): Move next_slot_number and
2084 next_slot_frag to ...
2085 (unw_rec_list): Here.
2086 (free_list_records): Removed.
2087 (output_unw_records): Likewise.
2088 (generate_unwind_image): Make it void.
2089 (alloc_record): Initialize next_slot_number and next_slot_frag.
2090 (slot_index): Take .org, .space and .align into account.
2091 (fixup_unw_records): Don't set slot_number to 0. Use
2092 list->next_slot_number and list->next_slot_frag instead of
2093 unwind.next_slot_number and unwind.next_slot_frag.
2094 (ia64_convert_frag): New.
2095 (generate_unwind_image): Generate a rs_machine_dependent frag
2097 (emit_one_bundle): Use list->next_slot_number and
2098 list->next_slot_frag instead of unwind.next_slot_number and
2099 unwind.next_slot_frag.
2101 * config/tc-ia64.h (md_convert_frag): Defined as
2103 (md_estimate_size_before_relax): Defined as (f)->fr_var.
2105 2004-01-06 Alexandre Oliva <aoliva@redhat.com>
2107 2003-12-19 Alexandre Oliva <aoliva@redhat.com>
2108 * config/tc-frv.h (md_apply_fix3): Don't define.
2109 * config/tc-frv.c (md_apply_fix3): New. Shift/truncate %hi/%lo
2111 * config/tc-frv.h (TC_FORCE_RELOCATION_SUB_LOCAL): Define.
2112 2003-10-07 Alexandre Oliva <aoliva@redhat.com>
2113 * config/tc-frv.c (line_separator_chars): Add `!'.
2114 2003-09-19 Alexandre Oliva <aoliva@redhat.com>
2115 * config/tc-frv.c (md_assemble): Clear insn upfront.
2116 2003-09-18 Alexandre Oliva <aoliva@redhat.com>
2117 * config/tc-frv.c (OPTION_FDPIC): New macro.
2118 (md_longopts): Add mfdpic.
2119 (md_parse_option): Handle it.
2120 2003-08-04 Alexandre Oliva <aoliva@redhat.com>
2121 * config/tc-frv.c (md_cgen_lookup_reloc) <FRV_OPERAND_D12,
2122 FRV_OPERAND_S12>: Use reloc type encoded in fix-up.
2123 (frv_pic_ptr): Parse funcdesc.
2125 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2127 * doc/as.texinfo: Let texi2pod parse asconfig.texi and
2128 gasver.texi. Remove duplicate symbol definitions for texi2pod.
2130 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2132 * Makefile.am (Makefile): Move the dependency on
2133 $(BFDDIR)/configure.in to...
2134 (CONFIG_STATUS_DEPENDENCIES): ... here.
2135 (AUTOMAKE_OPTIONS): Require automake 1.8.
2136 * Makefile.in: Regenerate.
2137 * doc/Makefile.am (BASEDIR, BFDDIR): Define.
2138 (CONFIG_STATUS_DEPENDENCIES): Add a dependency on
2139 $(BFDDIR)/configure.in here as well.
2140 * doc/Makefile.in: Regenerate.
2142 2004-01-05 Maciej W. Rozycki <macro@ds2.pg.gda.pl>
2144 * Makefile.am (install, install-info, RECURSIVE_TARGETS): Remove.
2145 * Makefile.in: Regenerate.
2146 * aclocal.m4: Regenerate.
2147 * doc/Makefile.am (install, install-info): Remove.
2148 (install-data-local): A new hook for install-info.
2149 (AUTOMAKE_OPTIONS): Require automake 1.8.
2150 * doc/Makefile.in: Regenerate.
2152 2004-01-02 Nutan Singh <nutan@kpitcummins.com>
2154 * doc/c-sh.texi: Update description about floating point behavior
2157 2004-01-02 Bernardo Innocenti <bernie@develer.com>
2159 * configure.in: Add m68k-uClinux target.
2160 * configure: Regenerate.
2162 For older changes see ChangeLog-0203
2168 version-control: never