1 2009-06-04 Andrew Haley <aph@redhat.com>
7 src/powerpc/ffitarget.h,
20 src/alpha/ffitarget.h,
36 src/ia64/ia64_flags.h,
40 src/sparc/ffitarget.h,
50 src/sh/sysv.S: Update license text.
52 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
54 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
55 (.eh_frame): Add FDE for it.
57 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
59 * configure.ac: Also check if assembler supports pc-relative
60 relocs on X86_WIN32 targets.
61 * configure: Regenerate.
62 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
63 (_ffi_call_SYSV): Add missing function type symbol .def and
65 (_ffi_call_STDCALL): Likewise.
66 (_ffi_closure_SYSV): Likewise.
67 (_ffi_closure_raw_SYSV): Likewise.
68 (.eh_frame): Add hand-crafted EH data.
70 2009-04-09 Jakub Jelinek <jakub@redhat.com>
72 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
73 version 3 of the GNU General Public License and to point readers
74 at the COPYING3 file and the FSF's license web page.
75 * testsuite/libffi.call/call.exp: Likewise.
76 * testsuite/libffi.special/special.exp: Likewise.
78 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
80 * configure: Regenerate.
82 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
85 * configure.ac (HAVE_AS_X86_PCREL): New test.
86 * configure: Regenerate.
87 * fficonfig.h.in: Regenerate.
88 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
89 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
90 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
91 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
92 * src/x86/unix64.S (.Lstore_table): Move to .text section.
93 (.Lload_table): Likewise.
94 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
96 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
98 * configure: Regenerate.
100 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
102 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
103 signed/unsigned int8/16 return values.
104 * src/sparc/v8.S (ffi_call_v8): Likewise.
105 (ffi_closure_v8): Likewise.
107 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
108 Steve Ellcey <sje@cup.hp.com>
110 * configure: Regenerate for new libtool.
111 * Makefile.in: Ditto.
112 * include/Makefile.in: Ditto.
115 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
117 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
118 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
119 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
120 Adjust copyright notice.
121 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
122 register or two register to use for FFI_SYSV structs.
123 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
124 (ffi_closure_helper_SYSV): Fix the return type for
125 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
126 Adjust copyright notice.
128 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
130 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
133 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
135 * configure: Regenerate.
136 * include/Makefile.in: Regenerate.
137 * testsuite/Makefile.in: Regenerate.
139 2008-06-07 Joseph Myers <joseph@codesourcery.com>
141 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
142 powerpc-*-beos*): Remove.
143 * configure: Regenerate.
145 2008-05-09 Julian Brown <julian@codesourcery.com>
147 * Makefile.am (LTLDFLAGS): New.
148 (libffi_la_LDFLAGS): Use above.
149 * Makefile.in: Regenerate.
151 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
154 * aclocal.m4: Regenerate.
155 * configure: Regenerate.
157 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
159 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
160 * src/sh64/sysv.S: Likewise.
162 2008-03-26 Daniel Jacobowitz <dan@debian.org>
164 * src/arm/sysv.S: Fix ARM comment marker.
166 2008-03-26 Jakub Jelinek <jakub@redhat.com>
168 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
169 * src/s390/sysv.S: Likewise.
170 * src/powerpc/ppc_closure.S: Likewise.
171 * src/powerpc/sysv.S: Likewise.
172 * src/x86/unix64.S: Likewise.
173 * src/x86/sysv.S: Likewise.
174 * src/sparc/v8.S: Likewise.
175 * src/sparc/v9.S: Likewise.
176 * src/m68k/sysv.S: Likewise.
177 * src/arm/sysv.S: Likewise.
179 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
181 * aclocal.m4: Regenerate.
182 * configure: Likewise.
183 * Makefile.in: Likewise.
184 * include/Makefile.in: Likewise.
185 * testsuite/Makefile.in: Likewise.
187 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
188 Andreas Tobler <a.tobler@schweiz.org>
190 * configure.ac: Add amd64-*-freebsd* target.
191 * configure: Regenerate.
193 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
196 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
199 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
202 2008-01-24 David Edelsohn <edelsohn@gnu.org>
204 * configure: Regenerate.
206 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
208 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
210 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
213 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
214 signed/unsigned int8/16 for X86_DARWIN.
215 Updated copyright info.
216 Handle one and two byte structs with special cif->flags.
217 * src/x86/ffitarget.h: Add special types for one and two byte structs.
218 Updated copyright info.
219 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
221 Remove code to pop args from the stack after call.
222 Special-case signed/unsigned for int8/16, one and two byte structs.
223 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
224 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
226 Updated copyright info.
228 2007-12-08 David Daney <ddaney@avtrex.com>
230 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
231 SUBU, add with ADDU and use smaller code sequences.
233 2007-12-07 David Daney <ddaney@avtrex.com>
235 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
238 2007-12-06 David Daney <ddaney@avtrex.com>
240 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
242 (ffi_java_raw): New typedef.
243 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
244 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
246 (ffi_java_raw_closure) : Same.
247 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
249 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
251 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
252 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
253 sizeof(ffi_java_raw) for alignment calculations.
254 (ffi_java_ptrarray_to_raw): Same.
255 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
256 if FFI_SIZEOF_JAVA_RAW == 4.
257 (ffi_java_raw_to_rvalue): Same.
258 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
259 (ffi_java_translate_args): Same.
260 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
262 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
264 2007-12-06 David Daney <ddaney@avtrex.com>
266 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
269 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
272 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
273 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
274 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
275 set the NUM_FPR_ARG_REGISTERS according to.
276 Add support for potential soft-float support under hard-float
278 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
279 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
280 to the FFI_LINUX_SOFT_FLOAT ABI.
281 (ffi_prep_cif_machdep): Likewise.
282 (ffi_closure_helper_SYSV): Likewise.
283 * src/powerpc/ppc_closure.S: Make sure not to store float/double
284 on archs where __NO_FPRS__ is true.
285 Add FFI_TYPE_UINT128 support.
286 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
287 Adjust copyright notice.
289 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
291 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
292 * include/ffi_common.h: ... here.
295 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
297 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
298 * src/powerpc/linux64.S: Likewise.
299 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
300 * testsuite/libffi.call/return_ldl.c: New test.
302 2007-09-04 <aph@redhat.com>
304 * src/arm/sysv.S (UNWIND): New.
305 (Whole file): Conditionally compile unwinder directives.
306 * src/arm/sysv.S: Add unwinder directives.
308 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
309 Only treat r0 as a struct address if we're actually returning a
311 Only copy the bytes that are actually within a struct.
312 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
313 is returned in r0, not passed by address.
314 (ffi_call): Allocate a word-sized temporary for the case where
315 a composite is returned in r0.
316 (ffi_prep_incoming_args_SYSV): Align as necessary.
318 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
320 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
321 directly using the sys_cacheflush syscall.
323 2007-07-27 Andrew Haley <aph@redhat.com>
325 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
327 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
329 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
330 * configure.ac: Likewise.
331 * Makefile.in: Regenerate.
332 * include/Makefile.in: Likewise.
333 * testsuite/Makefile.in: Likewise.
334 * configure: Likewise.
336 2007-08-24 David Daney <ddaney@avtrex.com>
338 * testsuite/libffi.call/return_sl.c: New test.
340 2007-08-10 David Daney <ddaney@avtrex.com>
342 * testsuite/libffi.call/cls_multi_ushort.c,
343 testsuite/libffi.call/cls_align_uint16.c,
344 testsuite/libffi.call/nested_struct1.c,
345 testsuite/libffi.call/nested_struct3.c,
346 testsuite/libffi.call/cls_7_1_byte.c,
347 testsuite/libffi.call/nested_struct5.c,
348 testsuite/libffi.call/cls_double.c,
349 testsuite/libffi.call/nested_struct7.c,
350 testsuite/libffi.call/cls_sint.c,
351 testsuite/libffi.call/nested_struct9.c,
352 testsuite/libffi.call/cls_20byte1.c,
353 testsuite/libffi.call/cls_multi_sshortchar.c,
354 testsuite/libffi.call/cls_align_sint64.c,
355 testsuite/libffi.call/cls_3byte2.c,
356 testsuite/libffi.call/cls_multi_schar.c,
357 testsuite/libffi.call/cls_multi_uchar.c,
358 testsuite/libffi.call/cls_19byte.c,
359 testsuite/libffi.call/cls_9byte1.c,
360 testsuite/libffi.call/cls_align_float.c,
361 testsuite/libffi.call/closure_fn1.c,
362 testsuite/libffi.call/problem1.c,
363 testsuite/libffi.call/closure_fn3.c,
364 testsuite/libffi.call/cls_sshort.c,
365 testsuite/libffi.call/closure_fn5.c,
366 testsuite/libffi.call/cls_align_double.c,
367 testsuite/libffi.call/nested_struct.c,
368 testsuite/libffi.call/cls_2byte.c,
369 testsuite/libffi.call/nested_struct10.c,
370 testsuite/libffi.call/cls_4byte.c,
371 testsuite/libffi.call/cls_6byte.c,
372 testsuite/libffi.call/cls_8byte.c,
373 testsuite/libffi.call/cls_multi_sshort.c,
374 testsuite/libffi.call/cls_align_sint16.c,
375 testsuite/libffi.call/cls_align_uint32.c,
376 testsuite/libffi.call/cls_20byte.c,
377 testsuite/libffi.call/cls_float.c,
378 testsuite/libffi.call/nested_struct2.c,
379 testsuite/libffi.call/cls_5_1_byte.c,
380 testsuite/libffi.call/nested_struct4.c,
381 testsuite/libffi.call/cls_24byte.c,
382 testsuite/libffi.call/nested_struct6.c,
383 testsuite/libffi.call/cls_64byte.c,
384 testsuite/libffi.call/nested_struct8.c,
385 testsuite/libffi.call/cls_uint.c,
386 testsuite/libffi.call/cls_multi_ushortchar.c,
387 testsuite/libffi.call/cls_schar.c,
388 testsuite/libffi.call/cls_uchar.c,
389 testsuite/libffi.call/cls_align_uint64.c,
390 testsuite/libffi.call/cls_ulonglong.c,
391 testsuite/libffi.call/cls_align_longdouble.c,
392 testsuite/libffi.call/cls_1_1byte.c,
393 testsuite/libffi.call/cls_12byte.c,
394 testsuite/libffi.call/cls_3_1byte.c,
395 testsuite/libffi.call/cls_3byte1.c,
396 testsuite/libffi.call/cls_4_1byte.c,
397 testsuite/libffi.call/cls_6_1_byte.c,
398 testsuite/libffi.call/cls_16byte.c,
399 testsuite/libffi.call/cls_18byte.c,
400 testsuite/libffi.call/closure_fn0.c,
401 testsuite/libffi.call/cls_9byte2.c,
402 testsuite/libffi.call/closure_fn2.c,
403 testsuite/libffi.call/closure_fn4.c,
404 testsuite/libffi.call/cls_ushort.c,
405 testsuite/libffi.call/closure_fn6.c,
406 testsuite/libffi.call/cls_5byte.c,
407 testsuite/libffi.call/cls_align_pointer.c,
408 testsuite/libffi.call/cls_7byte.c,
409 testsuite/libffi.call/cls_align_sint32.c,
410 testsuite/libffi.special/unwindtest_ffi_call.cc,
411 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
413 2007-08-10 David Daney <ddaney@avtrex.com>
416 * configure.ac: Don't treat mips64 as a special case.
417 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
418 * configure: Regenerate
419 * Makefile.in: Ditto.
420 * fficonfig.h.in: Ditto.
421 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
422 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
423 (FFI_DEFAULT_ABI): Set for n64 case.
424 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
425 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
426 (ffi_closure_N32): New function.
427 (.eh_frame): New section
428 * src/mips/o32.S: Clean up comments.
429 (ffi_closure_O32): Pass ffi_closure parameter in $12.
430 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
431 _MIPS_SIM == _ABIN32 throughout.
432 (FFI_MIPS_STOP_HERE): New, use in place of
434 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
435 to support n32/n64 ABIs.
436 (calc_n32_struct_flags): Rewrite.
437 (calc_n32_return_struct_flags): Remove unused variable. Reverse
438 position of flag bits.
439 (ffi_prep_cif_machdep): Rewrite n32 portion.
440 (ffi_call): Enable for n64. Add special handling for small structure
442 (ffi_prep_closure_loc): Add n32 and n64 support.
443 (ffi_closure_mips_inner_O32): Add cast to silence warning.
444 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
446 2007-08-08 David Daney <ddaney@avtrex.com>
448 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
449 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
451 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
452 * testsuite/libffi.call/cls_sint.c (main): Ditto.
453 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
454 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
455 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
456 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
457 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
458 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
459 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
460 * testsuite/libffi.call/return_fl3.c (main): Ditto.
461 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
462 * testsuite/libffi.call/nested_struct.c (main): Ditto.
463 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
464 * testsuite/libffi.call/return_ll1.c (main): Ditto.
465 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
466 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
467 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
468 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
469 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
470 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
471 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
472 * testsuite/libffi.call/cls_uint.c (main): Ditto.
473 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
474 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
475 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
476 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
477 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
478 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
479 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
480 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
481 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
483 2007-08-07 Andrew Haley <aph@redhat.com>
485 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
488 2007-08-06 Andrew Haley <aph@redhat.com>
491 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
492 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
495 2007-08-02 David Daney <ddaney@avtrex.com>
497 * testsuite/libffi.call/return_ul.c (main): Define return type as
498 ffi_arg. Use proper printf conversion specifier.
500 2007-07-30 Andrew Haley <aph@redhat.com>
503 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
504 signed/unsigned int8/16.
505 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
507 Remove code to pop args from the stack after call.
508 Special-case signed/unsigned int8/16.
509 * testsuite/libffi.call/return_sc.c (main): Revert.
511 2007-07-26 Richard Guenther <rguenther@suse.de>
514 * testsuite/libffi.call/return_sc.c (main): Verify call
515 result as signed char, not ffi_arg.
517 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
519 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
520 * configure: Regenerate.
522 2007-07-11 David Daney <ddaney@avtrex.com>
524 * src/mips/ffi.c: Don't include sys/cachectl.h.
525 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
528 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
530 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
531 from (ffi_prep_closure): ... this.
532 (FFI_INIT_TRAMPOLINE): Adjust.
534 2005-12-31 Phil Blundell <pb@reciva.com>
536 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
537 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
538 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
539 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
540 (FFI_CLOSURES): Enable closure support.
542 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
544 * testsuite/libffi.call/cls_multi_ushort.c,
545 testsuite/libffi.call/cls_align_uint16.c,
546 testsuite/libffi.call/nested_struct1.c,
547 testsuite/libffi.call/nested_struct3.c,
548 testsuite/libffi.call/cls_7_1_byte.c,
549 testsuite/libffi.call/cls_double.c,
550 testsuite/libffi.call/nested_struct5.c,
551 testsuite/libffi.call/nested_struct7.c,
552 testsuite/libffi.call/cls_sint.c,
553 testsuite/libffi.call/nested_struct9.c,
554 testsuite/libffi.call/cls_20byte1.c,
555 testsuite/libffi.call/cls_multi_sshortchar.c,
556 testsuite/libffi.call/cls_align_sint64.c,
557 testsuite/libffi.call/cls_3byte2.c,
558 testsuite/libffi.call/cls_multi_schar.c,
559 testsuite/libffi.call/cls_multi_uchar.c,
560 testsuite/libffi.call/cls_19byte.c,
561 testsuite/libffi.call/cls_9byte1.c,
562 testsuite/libffi.call/cls_align_float.c,
563 testsuite/libffi.call/closure_fn1.c,
564 testsuite/libffi.call/problem1.c,
565 testsuite/libffi.call/closure_fn3.c,
566 testsuite/libffi.call/cls_sshort.c,
567 testsuite/libffi.call/closure_fn5.c,
568 testsuite/libffi.call/cls_align_double.c,
569 testsuite/libffi.call/cls_2byte.c,
570 testsuite/libffi.call/nested_struct.c,
571 testsuite/libffi.call/nested_struct10.c,
572 testsuite/libffi.call/cls_4byte.c,
573 testsuite/libffi.call/cls_6byte.c,
574 testsuite/libffi.call/cls_8byte.c,
575 testsuite/libffi.call/cls_multi_sshort.c,
576 testsuite/libffi.call/cls_align_uint32.c,
577 testsuite/libffi.call/cls_align_sint16.c,
578 testsuite/libffi.call/cls_float.c,
579 testsuite/libffi.call/cls_20byte.c,
580 testsuite/libffi.call/cls_5_1_byte.c,
581 testsuite/libffi.call/nested_struct2.c,
582 testsuite/libffi.call/cls_24byte.c,
583 testsuite/libffi.call/nested_struct4.c,
584 testsuite/libffi.call/nested_struct6.c,
585 testsuite/libffi.call/cls_64byte.c,
586 testsuite/libffi.call/nested_struct8.c,
587 testsuite/libffi.call/cls_uint.c,
588 testsuite/libffi.call/cls_multi_ushortchar.c,
589 testsuite/libffi.call/cls_schar.c,
590 testsuite/libffi.call/cls_uchar.c,
591 testsuite/libffi.call/cls_align_uint64.c,
592 testsuite/libffi.call/cls_ulonglong.c,
593 testsuite/libffi.call/cls_align_longdouble.c,
594 testsuite/libffi.call/cls_1_1byte.c,
595 testsuite/libffi.call/cls_12byte.c,
596 testsuite/libffi.call/cls_3_1byte.c,
597 testsuite/libffi.call/cls_3byte1.c,
598 testsuite/libffi.call/cls_4_1byte.c,
599 testsuite/libffi.call/cls_6_1_byte.c,
600 testsuite/libffi.call/cls_16byte.c,
601 testsuite/libffi.call/cls_18byte.c,
602 testsuite/libffi.call/closure_fn0.c,
603 testsuite/libffi.call/cls_9byte2.c,
604 testsuite/libffi.call/closure_fn2.c,
605 testsuite/libffi.call/closure_fn4.c,
606 testsuite/libffi.call/cls_ushort.c,
607 testsuite/libffi.call/closure_fn6.c,
608 testsuite/libffi.call/cls_5byte.c,
609 testsuite/libffi.call/cls_align_pointer.c,
610 testsuite/libffi.call/cls_7byte.c,
611 testsuite/libffi.call/cls_align_sint32.c,
612 testsuite/libffi.special/unwindtest_ffi_call.cc,
613 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
615 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
617 * aclocal.m4: Regenerated.
619 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
621 * configure: Regenerate.
623 2007-05-23 Steve Ellcey <sje@cup.hp.com>
625 * Makefile.in: Regenerate.
626 * configure: Regenerate.
627 * aclocal.m4: Regenerate.
628 * include/Makefile.in: Regenerate.
629 * testsuite/Makefile.in: Regenerate.
631 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
633 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
634 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
635 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
636 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
637 (FFI_CLOSURES): Enable closure support.
639 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
641 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
642 * configure: Regenerate.
643 * fficonfig.h.in: Regenerate.
644 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
645 CFI_OFFSET,CFI_DEF_CFA): New macros.
646 (ffi_call_SYSV): Add callframe annotation.
648 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
650 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
651 numerous test suite failures.
652 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
654 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
656 * Makefile.am (EXTRA_DIST): Bring up to date.
657 * Makefile.in: Regenerate.
658 * src/frv/eabi.S: Remove RCS keyword.
660 2007-04-06 Richard Henderson <rth@redhat.com>
662 * configure.ac: Tidy target case.
663 (HAVE_LONG_DOUBLE): Allow the target to override.
664 * configure: Regenerate.
665 * include/ffi.h.in: Don't define ffi_type_foo if
666 LIBFFI_HIDE_BASIC_TYPES is defined.
667 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
669 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
670 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
671 (ffi_type_longdouble): Special case for Alpha. Don't define
672 if long double == double.
674 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
675 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
676 (ffi_call, ffi_closure_osf_inner): Likewise.
677 (ffi_closure_osf_inner): Likewise. Mark hidden.
678 (ffi_call_osf, ffi_closure_osf): Mark hidden.
679 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
680 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
681 (load_table): Handle 128-bit long double.
683 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
685 2007-04-06 Tom Tromey <tromey@redhat.com>
688 * README: Fixed bug in example.
690 2007-04-03 Jakub Jelinek <jakub@redhat.com>
692 * src/closures.c: Include sys/statfs.h.
693 (_GNU_SOURCE): Define on Linux.
694 (FFI_MMAP_EXEC_SELINUX): Define.
695 (selinux_enabled): New variable.
696 (selinux_enabled_check): New function.
697 (is_selinux_enabled): Define.
700 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
702 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
703 Use 'volatile float sum' to create sum of floats to avoid false
704 negative due to excess precision on ix86 targets.
707 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
709 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
711 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
713 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
715 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
716 (ffi_prep_closure_loc): New.
717 (ffi_prep_raw_closure_loc): New.
718 (ffi_prep_java_raw_closure_loc): New.
719 * src/closures.c: New file.
720 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
721 Replace sflags with exec_offset.
722 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
723 sub_segment_exec_offset): New macros.
724 (get_segment_flags, set_segment_flags, check_segment_merge): New
726 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
727 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
728 destroy_mspace): Use new macros.
729 (sys_alloc): Silence warning.
730 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
731 * Makefile.in: Rebuilt.
732 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
733 terms of ffi_prep_closure_loc.
734 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
736 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
738 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
740 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
742 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
743 (ffi_prep_closure): ... this.
744 * src/pa/ffi.c: Likewise.
745 * src/cris/ffi.c: Likewise. Adjust.
746 * src/frv/ffi.c: Likewise.
747 * src/ia64/ffi.c: Likewise.
748 * src/mips/ffi.c: Likewise.
749 * src/powerpc/ffi_darwin.c: Likewise.
750 * src/s390/ffi.c: Likewise.
751 * src/sh/ffi.c: Likewise.
752 * src/sh64/ffi.c: Likewise.
753 * src/sparc/ffi.c: Likewise.
754 * src/x86/ffi64.c: Likewise.
755 * src/x86/ffi.c: Likewise.
756 (FFI_INIT_TRAMPOLINE): Adjust.
757 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
758 (ffi_prep_raw_closure): ... this.
759 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
760 (ffi_prep_closure): ... this.
761 (flush_icache): Adjust.
763 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
765 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
768 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
770 * Makefile.am: Add dummy install-pdf target.
771 * Makefile.in: Regenerate
773 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
775 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
776 ffi_closure_helper_SYSV): Add long double handling.
778 2007-02-02 Jakub Jelinek <jakub@redhat.com>
780 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
781 immediately after bctrl instruction.
783 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
785 * Makefile.am (all-recursive, install-recursive,
786 mostlyclean-recursive, clean-recursive, distclean-recursive,
787 maintainer-clean-recursive): Add missing targets.
788 * Makefile.in: Rebuilt.
790 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
792 * configure.ac: Add TARGET for x86_64-*-darwin*.
793 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
795 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
796 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
797 * configure: Regenerate.
798 * Makefile.in: Regenerate.
799 * include/Makefile.in: Regenerate.
800 * testsuite/Makefile.in: Regenerate.
801 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
804 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
806 * aclocal.m4: Regenerate with aclocal -I .. as written in the
809 2006-10-31 Geoffrey Keating <geoffk@apple.com>
811 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
812 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
814 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
815 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
817 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
818 Sandro Tolaini <tolaini@libero.it>
820 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
822 * configure: Regenerated.
823 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
824 (EXTRA_DIST): Add src/x86/darwin.S.
825 * Makefile.in: Regenerated.
826 * include/Makefile.in: Regenerated.
827 * testsuite/Makefile.in: Regenerated.
829 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
830 X86_WIN32, and additionally align stack to 16 bytes.
831 * src/x86/darwin.S: New, based on sysv.S.
832 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
834 2006-09-12 David Daney <ddaney@avtrex.com>
837 * include/Makefile.am: Install both ffi.h and ffitarget.h in
838 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
839 * aclocal.m4: Regenerated for automake 1.9.6.
840 * Makefile.in: Regenerated.
841 * include/Makefile.in: Regenerated.
842 * testsuite/Makefile.in: Regenerated.
844 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
846 * include/ffi_common.h (struct): Revert accidental commit.
848 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
850 * include/ffi_common.h: Remove lint directives.
851 * include/ffi.h.in: Likewise.
853 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
855 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
856 for 32-bit architectures.
857 * testsuite/libffi.call/return_ul.c: New test case.
859 2006-07-19 David Daney <ddaney@avtrex.com>
861 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
862 xfail remains for mips64.
864 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
866 * Makefile.am: Add install-html target. Add install-html to .PHONY
867 * Makefile.in: Regenerate.
868 * aclocal.m4: Regenerate.
869 * include/Makefile.in: Regenerate.
870 * testsuite/Makefile.in: Regenerate.
872 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
874 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
877 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
879 * README: Remove notice about 'Crazy Comments'.
880 * src/debug.c: Remove lint directives. Cleanup white spaces.
881 * src/java_raw_api.c: Likewise.
882 * src/prep_cif.c: Likewise.
883 * src/raw_api.c: Likewise.
884 * src/ffitest.c: Delete. No longer needed, all test cases migrated
886 * src/arm/ffi.c: Remove lint directives.
887 * src/m32r/ffi.c: Likewise.
888 * src/pa/ffi.c: Likewise.
889 * src/powerpc/ffi.c: Likewise.
890 * src/powerpc/ffi_darwin.c: Likewise.
891 * src/sh/ffi.c: Likewise.
892 * src/sh64/ffi.c: Likewise.
893 * src/x86/ffi.c: Likewise.
894 * testsuite/libffi.call/float2.c: Likewise.
895 * testsuite/libffi.call/promotion.c: Likewise.
896 * testsuite/libffi.call/struct1.c: Likewise.
898 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
900 * src/pa/hpux32.S: Correct unwind offset calculation for
902 * src/pa/linux.S: Likewise.
904 2006-04-12 James E Wilson <wilson@specifix.com>
907 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
908 (hfa_type_load): Call stf_spill.
909 (hfa_type_store): Call ldf_fill.
910 (ffi_call): Adjust calls to above routines. Add local temps for
913 2006-04-10 Matthias Klose <doko@debian.org>
915 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
916 directory names containing underscores.
918 2006-04-07 James E Wilson <wilson@specifix.com>
920 * testsuite/libffi.call/float4.c: New testcase.
922 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
923 Andreas Tobler <a.tobler@schweiz.ch>
925 * Makefile.am: Add PA_HPUX port.
926 * Makefile.in: Regenerate.
927 * include/Makefile.in: Likewise.
928 * testsuite/Makefile.in: Likewise.
929 * configure.ac: Add PA_HPUX rules.
930 * configure: Regenerate.
931 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
932 Add PA_HPUX and PA64_HPUX.
933 Rename FFI_LINUX ABI to FFI_PA32 ABI.
934 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
935 (FFI_TYPE_SMALL_STRUCT2): Define.
936 (FFI_TYPE_SMALL_STRUCT4): Likewise.
937 (FFI_TYPE_SMALL_STRUCT8): Likewise.
938 (FFI_TYPE_SMALL_STRUCT3): Redefine.
939 (FFI_TYPE_SMALL_STRUCT5): Likewise.
940 (FFI_TYPE_SMALL_STRUCT6): Likewise.
941 (FFI_TYPE_SMALL_STRUCT7): Likewise.
942 * src/pa/ffi.c (ROUND_DOWN): Delete.
943 (fldw, fstw, fldd, fstd): Use '__asm__'.
944 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
945 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
946 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
947 Simplify incrementing of stack slot variable. Change type of local
949 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
951 (ffi_prep_cif_machdep): Likewise.
952 (ffi_call): Likewise.
953 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
954 return type to ffi_status. Simplify incrementing of stack slot
955 variable. Only copy floating point argument registers when PA_LINUX
956 is true. Reformat debug statement.
957 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
958 FFI_TYPE_SMALL_STRUCT8.
959 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
961 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
962 Add nops to cache flush. Add trampoline for PA_HPUX.
963 * src/pa/hpux32.S: New file.
964 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
965 ffi_prep_args_LINUX to ffi_prep_args_pa32.
966 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
967 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
968 argument type checks so that common argument types appear first.
969 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
970 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
972 2006-03-24 Alan Modra <amodra@bigpond.net.au>
974 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
975 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
976 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
978 (ffi_prep_args64): Assert using IBM extended double.
979 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
980 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
981 (ffi_call): Handle FFI_LINUX.
982 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
983 gpr3 return pointer as for struct return. Handle FFI_LINUX
984 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
986 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
987 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
988 Don't use r6 as pointer to results, instead use sp offset. Don't
989 make a special call to load lr with case table address, instead
990 use offset from previous call.
991 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
992 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
995 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
997 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
998 passed with FP registers correctly.
999 (ffi_closure_helper_SYSV): Likewise.
1000 * src/sh64/sysv.S: Likewise.
1002 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1004 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1005 args and userdata unused.
1006 (closure_test_fn1): Mark cif and userdata unused.
1007 (main): Remove unused res.
1009 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1011 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1012 -O2, -O3, -Os and the warning flags -W -Wall.
1013 * testsuite/libffi.special/special.exp: Likewise.
1014 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1015 unused parameter unused for gcc or else do nothing.
1016 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1017 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1018 and userdata unused.
1019 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1020 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1021 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1022 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1023 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1024 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1025 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1026 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1027 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1028 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1029 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1030 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1031 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1032 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1033 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1034 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1035 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1036 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1037 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1038 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1039 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1040 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1041 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1042 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1044 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1046 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1048 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1049 void* to avoid compiler warning.
1051 (cls_struct_align_gn): Mark cif and userdata unused.
1052 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1054 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1056 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1058 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1060 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1062 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1063 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1064 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1066 (main): Cast res_call to silence gcc.
1067 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1069 (main): Cast res_call to silence gcc.
1070 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1072 (main): Cast res_call to silence gcc.
1073 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1075 (main): Cast res_call to silence gcc.
1076 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1078 (main): Cast res_call to silence gcc.
1079 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1081 (main): Cast res_call to silence gcc.
1082 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1084 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1085 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1087 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1088 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1090 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1091 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1093 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1094 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1096 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1097 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1098 and userdata unused.
1099 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1101 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1102 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1103 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1104 Cleanup white spaces.
1105 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1106 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1107 cif and userdata unused.
1108 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1110 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1111 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1112 formatters to silence gcc.
1113 (B_gn): Mark cif and userdata unused.
1114 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1116 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1117 (B_gn): Mark cif and userdata unused.
1118 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1120 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1121 (B_gn): Mark cif and userdata unused.
1122 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1124 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1125 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1126 * testsuite/libffi.call/problem1.c (stub): Likewise.
1127 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1129 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1130 in the last commit for this test case in the test case itself.
1131 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1133 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1134 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1135 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1136 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1137 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1138 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1140 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1142 * src/sh/sysv.S: Fix register numbers in the FDE for
1145 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1147 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1148 declaration to avoid a false negative on ix86. See PR323.
1150 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1152 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1153 and cast integer to void * if needed. Update the pointer to
1154 the FP register saved area correctly.
1156 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1158 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1160 * testsuite/libffi.call/nested_struct4.c: Likewise.
1162 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1164 * testsuite/libffi.call/return_dbl.c: New test case.
1165 * testsuite/libffi.call/return_dbl1.c: Likewise.
1166 * testsuite/libffi.call/return_dbl2.c: Likewise.
1167 * testsuite/libffi.call/return_fl.c: Likewise.
1168 * testsuite/libffi.call/return_fl1.c: Likewise.
1169 * testsuite/libffi.call/return_fl2.c: Likewise.
1170 * testsuite/libffi.call/return_fl3.c: Likewise.
1171 * testsuite/libffi.call/closure_fn6.c: Likewise.
1173 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1175 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1176 here to be used by other test cases too.
1178 * testsuite/libffi.call/nested_struct10.c: New test case.
1179 * testsuite/libffi.call/nested_struct9.c: Likewise.
1180 * testsuite/libffi.call/nested_struct8.c: Likewise.
1181 * testsuite/libffi.call/nested_struct7.c: Likewise.
1182 * testsuite/libffi.call/nested_struct6.c: Likewise.
1183 * testsuite/libffi.call/nested_struct5.c: Likewise.
1184 * testsuite/libffi.call/nested_struct4.c: Likewise.
1186 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1188 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1189 * configure: Rebuilt.
1191 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1193 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1194 instead do the shifting inline.
1195 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1196 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1197 and 8 byte structs, for the remaining struct sizes don't call
1198 __lshrdi3, instead do the shifting inline.
1200 2005-12-07 Thiemo Seufer <ths@networkno.de>
1202 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1203 missing parentheses.
1204 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1205 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1206 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1207 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1208 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1210 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1212 (ffi_prep_closure): Improve trampoline instruction scheduling.
1213 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1215 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1217 * src/powerpc/ffi.c: Formatting.
1218 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1219 (ffi_prep_args64): Likewise.
1221 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1223 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1224 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1226 2005-09-26 Tom Tromey <tromey@redhat.com>
1228 * testsuite/libffi.call/float1.c (value_type): New typedef.
1229 (CANARY): New define.
1230 (main): Check for result buffer overflow.
1231 * src/powerpc/linux64.S: Handle linux64 long double returns.
1232 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1233 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1235 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1238 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1240 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1242 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1244 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1245 (AH_BOTTOM): Add FFI_HIDDEN definition.
1246 * configure: Rebuilt.
1247 * fficonfig.h.in: Rebuilt.
1248 * src/powerpc/ffi.c (hidden): Remove.
1249 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1250 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1251 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1252 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1253 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1254 add FFI_HIDDEN to its prototype.
1255 (ffi_closure_SYSV_inner): New.
1256 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1257 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1259 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1262 * configure: Rebuilt.
1263 * configure.ac: Handle i*86-*-gnu*.
1265 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1267 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1268 DW_CFA_offset_extended_sf rather than
1269 DW_CFA_GNU_negative_offset_extended.
1270 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1272 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1274 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1276 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1277 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1278 partially on register.
1279 (ffi_closure_helper_SYSV): Likewise.
1280 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1282 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1284 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1286 * src/sh64/ffi.c (simple_type): Remove.
1287 (return_type): Handle small structures correctly.
1288 (ffi_prep_args): Likewise.
1289 (ffi_call): Likewise.
1290 (ffi_closure_helper_SYSV): Likewise.
1291 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1292 Emit position independent code if PIC and remove wrong datalabel
1293 prefixes from EH data.
1295 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1297 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1298 * Makefile.in: Regenerate.
1299 * include/Makefile.in: Likewise.
1300 * testsuite/Makefile.in: Likewise.
1301 * configure.ac: Add POWERPC_FREEBSD rules.
1302 * configure: Regenerate.
1303 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1304 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1305 * src/powerpc/ffi.c: Add flags to handle small structure returns
1307 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1309 (ffi_closure_helper_SYSV): Likewise.
1310 * src/powerpc/ppc_closure.S: Add return types for small structures.
1311 * src/powerpc/sysv.S: Add bits to handle small structures for
1314 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1316 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1317 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1318 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1320 2005-07-05 Randolph Chung <tausq@debian.org>
1322 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1323 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1324 structures. Kill compilation warnings.
1325 (ffi_closure_inner_LINUX): Print return values as hex in debug
1326 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1327 Properly handle 5-7 byte structure returns.
1328 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1329 (FFI_TYPE_SMALL_STRUCT2): Remove.
1330 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1331 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1332 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1333 (checksmst1, checksmst2): Remove.
1334 (checksmst3): Optimize handling of 3-byte struct returns.
1335 (checksmst567): Properly handle 5-7 byte struct returns.
1337 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1340 * src/mips/n32.S: Enforce PIC code.
1341 * src/mips/o32.S: Likewise.
1343 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1345 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1346 * configure: Regenerate.
1348 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1350 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1351 to call ffi_closure_helper_SYSV. Append @local instead.
1352 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1354 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1356 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1357 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1358 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1359 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1360 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1362 2005-05-09 Mike Stump <mrs@apple.com>
1364 * configure: Regenerate.
1366 2005-05-08 Richard Henderson <rth@redhat.com>
1369 * src/alpha/osf.S: Update unwind into to match code.
1371 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1372 Richard Henderson <rth@redhat.com>
1374 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1376 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1377 (ffi_prep_closure): Set carry bit if sse-used flag set.
1378 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1379 Only load sse registers if ssecount non-zero.
1380 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1382 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1384 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1385 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1386 * configure: Regenerate.
1388 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1390 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1391 have Tcl8.3-compatible intermediate variable.
1393 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1394 Hans-Peter Nilsson <hp@axis.com>
1396 * Makefile.am: Add CRIS support.
1397 * configure.ac: Likewise.
1398 * Makefile.in, configure, testsuite/Makefile.in,
1399 include/Makefile.in: Regenerate.
1400 * src/cris: New directory.
1401 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1402 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1404 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1405 \r?\n in output tests.
1407 2005-04-12 Mike Stump <mrs@apple.com>
1409 * configure: Regenerate.
1411 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1413 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1415 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1417 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1419 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1420 to 64 bits in ILP32 mode.
1421 Load 64 bits even for short data.
1423 2005-03-23 Mike Stump <mrs@apple.com>
1425 * src/powerpc/darwin.S: Update for -m64 multilib.
1426 * src/powerpc/darwin_closure.S: Likewise.
1428 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1430 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1431 Do not set tool_include_dir.
1432 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1434 * include/Makefile.am: Set gcc_version and toollibffidir.
1435 * include/Makefile.in: Regenerate.
1437 2005-02-22 Andrew Haley <aph@redhat.com>
1439 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1440 odd-numbered register pairs for 64-bit integer types.
1442 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1445 * testsuite/libffi.call/return_ll1.c: New test case.
1447 2005-02-11 Janis Johnson <janis187@us.ibm.com>
1449 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1450 * testsuite/libffi.call/float.c: Ditto.
1451 * testsuite/libffi.call/float2.c: Ditto.
1452 * testsuite/libffi.call/float3.c: Ditto.
1454 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1456 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1458 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1460 * testsuite/libffi.special/special.exp (cxx_options): Add
1463 2004-12-31 Richard Henderson <rth@redhat.com>
1465 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1466 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1467 offset parameters with a type parameter; deduce size and structure
1468 alignment. Update all users.
1470 2004-12-31 Richard Henderson <rth@redhat.com>
1472 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1473 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
1474 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
1475 into ffi_prep_closure.
1476 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1479 2004-12-27 Richard Henderson <rth@redhat.com>
1481 * src/x86/unix64.S: Fix typo in unwind info.
1483 2004-12-25 Richard Henderson <rth@redhat.com>
1485 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1486 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1487 (merge_classes): Check for it.
1489 (classify_argument): Pass byte_offset by value; perform all updates
1491 (examine_argument): Add classes argument; handle
1492 X86_64_COMPLEX_X87_CLASS.
1493 (ffi_prep_args): Merge into ...
1494 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1495 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1496 (ffi_fill_return_value): Remove.
1497 (ffi_prep_closure): Remove dead assert.
1498 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1499 Rewrite to use struct register_args instead of va_list. Create
1500 flags for handling structure returns.
1501 * src/x86/unix64.S: Remove dead strings.
1502 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1503 stack frame with ffi_call. Handle structure returns properly.
1504 (float2sse, floatfloat2sse, double2sse): Remove.
1505 (sse2float, sse2double, sse2floatfloat): Remove.
1506 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1507 to handle structure returns properly.
1509 2004-12-08 David Edelsohn <edelsohn@gnu.org>
1511 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1513 * Makefile.in: Regenerated.
1515 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
1517 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1518 * configure, aclocal.m4, Makefile.in: Regenerate.
1519 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1521 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1523 * configure: Regenerate for libtool change.
1525 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1527 * configure: Regenerate for libtool reversion.
1529 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1531 * configure: Regenerate for libtool change.
1533 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1535 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1537 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
1539 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1540 of jal. Use an absolute encoding for the frame information.
1542 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1544 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1545 * acinclude.m4: Delete logic for sincludes.
1546 * aclocal.m4, Makefile.in, configure: Regenerate.
1547 * include/Makefile: Likewise.
1548 * testsuite/Makefile: Likewise.
1550 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1552 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1553 on a 8-byte boundary.
1554 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1556 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1558 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1559 long long values. Round stack allocation to a multiple of 8 bytes
1560 for ATPCS compatibility.
1561 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1562 names. Handle returning long long types. Add Thumb and interworking
1563 support. Improve soft-float code.
1565 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1567 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1568 (libffi_exit): New function.
1569 (libffi_init): Build the testglue wrapper if needed.
1571 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1574 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1576 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1578 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1580 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1582 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1583 * testsuite/libffi.call/float3.c: New test case.
1585 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1587 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1588 the function returning a structure pointed with R2.
1589 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1590 the structure return value if T bit set. Emit position
1591 independent code and EH data if PIC.
1593 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1595 * Makefile.am: Add m32r support.
1596 * configure.ac: Likewise.
1597 * Makefile.in: Regenerate.
1598 * confiugre: Regenerate.
1599 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1600 (uint64, sint64, double, longdouble)
1601 * src/m32r: New directory.
1602 * src/m32r/ffi.c: New file.
1603 * src/m32r/sysv.S: Likewise.
1604 * src/m32r/ffitarget.h: Likewise.
1606 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1608 * testsuite/libffi.call/negint.c: New test case.
1610 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1613 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1614 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1615 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1618 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1620 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1621 * testsuite/libffi.call/promotion.c: Likewise.
1622 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1624 * testsuite/libffi.call/return_sc.c: Likewise.
1625 * testsuite/libffi.call/return_uc.c: Likewise.
1627 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1629 * src/powerpc/darwin.S: Fix comments and identation.
1630 * src/powerpc/darwin_closure.S: Likewise.
1632 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1634 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1635 (ffi_prep_args): Handle longdouble arguments.
1636 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1638 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1639 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1641 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1642 * src/types.c: Defined longdouble size and alignment for darwin.
1644 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1646 * src/powerpc/aix.S: Remove whitespaces.
1647 * src/powerpc/aix_closure.S: Likewise.
1648 * src/powerpc/asm.h: Likewise.
1649 * src/powerpc/ffi.c: Likewise.
1650 * src/powerpc/ffitarget.h: Likewise.
1651 * src/powerpc/linux64.S: Likewise.
1652 * src/powerpc/linux64_closure.S: Likewise.
1653 * src/powerpc/ppc_closure.S: Likewise.
1654 * src/powerpc/sysv.S: Likewise.
1656 2004-08-30 Anthony Green <green@redhat.com>
1658 * Makefile.am: Add frv support.
1659 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1660 * configure.ac: Read configure.host.
1661 * configure.in: Read configure.host.
1662 * configure.host: New file. frv-elf needs libgloss.
1663 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
1664 alignment. This is needed to frv and shouldn't harm the others.
1665 * include/ffi_common.h (ALIGN_DOWN): New macro.
1666 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1668 2004-08-24 David Daney <daney@avtrex.com>
1670 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1671 * testsuite/libffi.call/closure_fn1.c: Likewise.
1672 * testsuite/libffi.call/closure_fn2.c Likewise.
1673 * testsuite/libffi.call/closure_fn3.c: Likewise.
1674 * testsuite/libffi.call/closure_fn4.c: Likewise.
1675 * testsuite/libffi.call/closure_fn5.c: Likewise.
1676 * testsuite/libffi.call/cls_18byte.c: Likewise.
1677 * testsuite/libffi.call/cls_19byte.c: Likewise.
1678 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1679 * testsuite/libffi.call/cls_20byte.c: Likewise.
1680 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1681 * testsuite/libffi.call/cls_24byte.c: Likewise.
1682 * testsuite/libffi.call/cls_2byte.c: Likewise.
1683 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1684 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1685 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1686 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1687 * testsuite/libffi.call/cls_4byte.c: Likewise.
1688 * testsuite/libffi.call/cls_64byte.c: Likewise.
1689 * testsuite/libffi.call/cls_6byte.c: Likewise.
1690 * testsuite/libffi.call/cls_7byte.c: Likewise.
1691 * testsuite/libffi.call/cls_8byte.c: Likewise.
1692 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1693 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1694 * testsuite/libffi.call/cls_align_double.c: Likewise.
1695 * testsuite/libffi.call/cls_align_float.c: Likewise.
1696 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1697 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1698 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1699 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1700 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1701 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1702 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1703 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1704 * testsuite/libffi.call/cls_double.c: Likewise.
1705 * testsuite/libffi.call/cls_float.c: Likewise.
1706 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1707 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1708 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1709 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1710 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1711 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1712 * testsuite/libffi.call/cls_schar.c: Likewise.
1713 * testsuite/libffi.call/cls_sint.c: Likewise.
1714 * testsuite/libffi.call/cls_sshort.c: Likewise.
1715 * testsuite/libffi.call/cls_uchar.c: Likewise.
1716 * testsuite/libffi.call/cls_uint.c: Likewise.
1717 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1718 * testsuite/libffi.call/cls_ushort.c: Likewise.
1719 * testsuite/libffi.call/nested_struct.c: Likewise.
1720 * testsuite/libffi.call/nested_struct1.c: Likewise.
1721 * testsuite/libffi.call/nested_struct2.c: Likewise.
1722 * testsuite/libffi.call/nested_struct3.c: Likewise.
1723 * testsuite/libffi.call/problem1.c: Likewise.
1724 * testsuite/libffi.special/unwindtest.cc: Likewise.
1725 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1727 * testsuite/libffi.call/cls_16byte.c: Likewise.
1728 * testsuite/libffi.call/cls_5byte.c: Likewise.
1730 2004-08-23 David Daney <daney@avtrex.com>
1733 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1734 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1735 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1736 parameters and return types.
1737 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1738 (ffi_prep_closure): Ditto.
1739 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1740 alignment calculations.
1741 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1742 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1744 2004-08-14 Casey Marshall <csm@gnu.org>
1746 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1747 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1748 integer (O32 ABI only).
1749 (ffi_prep_closure): new function.
1750 (ffi_closure_mips_inner_O32): new function.
1751 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
1752 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1753 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1754 64 bit integers correctly.
1755 (ffi_closure_O32): new function.
1756 Added DWARF-2 unwind info for both functions.
1758 2004-08-10 Andrew Haley <aph@redhat.com>
1760 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1762 2004-08-01 Robert Millan <robertmh@gnu.org>
1764 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1765 * configure: Regenerate.
1767 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1769 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1770 and mmap() explicitly instead of relying on preset autoconf cache
1772 * aclocal.m4: Regenerate.
1773 * configure: Regenerate.
1775 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1777 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1778 (ffi_check_float_struct): Remove unused prototype.
1780 2004-06-30 Geoffrey Keating <geoffk@apple.com>
1782 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1783 character on Darwin, use '\n\t' instead.
1785 2004-06-26 Matthias Klose <doko@debian.org>
1787 * libtool-version: Fix typo in revision/age.
1789 2004-06-17 Matthias Klose <doko@debian.org>
1791 * libtool-version: New.
1792 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1793 * Makefile.in: Regenerate.
1795 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1797 * Makefile.am: Remove useless multilib rules.
1798 * Makefile.in: Regenerate.
1799 * aclocal.m4: Regenerate with automake 1.8.5.
1800 * configure.ac: Remove useless multilib configury.
1801 * configure: Regenerate.
1803 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1805 * .cvsignore: New file.
1807 2004-06-10 Jakub Jelinek <jakub@redhat.com>
1809 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1811 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1814 2004-06-06 Sean McNeil <sean@mcneil.com>
1816 * configure.ac: Add x86_64-*-freebsd* support.
1817 * configure: Regenerate.
1819 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
1822 * configure.ac: Test for existence of mmap and sys/mman.h before
1823 checking blacklist. Fix suggested by Jim Wilson.
1824 * configure: Regenerate.
1826 2004-04-26 Matt Austern <austern@apple.com>
1828 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1830 * src/powerpc/darwin_closure.S: Likewise.
1832 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1834 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1835 error. Reported by Thomas Heller <theller@python.net>.
1836 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1837 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1839 2004-03-20 Matthias Klose <doko@debian.org>
1841 * src/pa/linux.S: Fix typo.
1843 2004-03-19 Matthias Klose <doko@debian.org>
1845 * Makefile.am: Update.
1846 * Makefile.in: Regenerate.
1847 * src/pa/ffi.h.in: Remove.
1848 * src/pa/ffitarget.h: New file.
1850 2004-02-10 Randolph Chung <tausq@debian.org>
1852 * Makefile.am: Add PA support.
1853 * Makefile.in: Regenerate.
1854 * include/Makefile.in: Regenerate.
1855 * configure.ac: Add PA target.
1856 * configure: Regenerate.
1857 * src/pa/ffi.c: New file.
1858 * src/pa/ffi.h.in: Add PA support.
1859 * src/pa/linux.S: New file.
1860 * prep_cif.c: Add PA support.
1862 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
1864 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
1866 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
1867 with ecif->cif->flags.
1868 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
1870 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
1871 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
1872 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
1873 sc_retstruct2b): Add for 1 or 2-bytes struct case.
1875 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
1877 * configure.in: Rename file to ...
1878 * configure.ac: ... this.
1879 * fficonfig.h.in: Regenerate.
1880 * Makefile.in: Regenerate.
1881 * include/Makefile.in: Regenerate.
1882 * testsuite/Makefile.in: Regenerate.
1884 2004-03-12 Matt Austern <austern@apple.com>
1886 * src/powerpc/darwin.S: Fix EH information so it corresponds to
1887 changes in EH format resulting from addition of linkonce support.
1888 * src/powerpc/darwin_closure.S: Likewise.
1890 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
1891 Paolo Bonzini <bonzini@gnu.org>
1893 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
1894 Remove VPATH. Remove rules for object files. Remove multilib support.
1895 (AM_CCASFLAGS): Add.
1896 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
1897 (AC_PREREQ): Bump version to 2.59.
1898 (AC_INIT): Fill with version info and bug address.
1899 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
1900 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
1901 De-precious CC so that the right flags are passed down to multilibs.
1902 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
1903 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
1904 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
1905 * configure: Rebuilt.
1906 * aclocal.m4: Likewise.
1907 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
1908 * fficonfig.h.in: Likewise.
1910 2004-03-11 Andreas Schwab <schwab@suse.de>
1912 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
1913 arguments from fp registers only for the first 8 parameter slots.
1914 Don't convert a float parameter when passed in memory.
1916 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
1918 * configure: Regenerate for config/accross.m4 correction.
1920 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
1922 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
1923 ecif->cif->bytes to bytes.
1924 (ffi_prep_cif_machdep): Add braces around nested if statement.
1926 2004-02-09 Alan Modra <amodra@bigpond.net.au>
1928 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
1930 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
1931 (ffi_closure_helper_LINUX64): Fix typo.
1932 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
1934 * testsuite/libffi.call/float.c: Likewise.
1935 * testsuite/libffi.call/float2.c: Likewise.
1937 2004-02-08 Alan Modra <amodra@bigpond.net.au>
1939 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
1940 long double function return and long double arg handling.
1941 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
1942 Use "end_pfr" instead of "nf". Correct long double handling.
1944 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
1946 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
1947 space for long double return value. Adjust stack frame and offsets.
1948 Load f2 long double return.
1950 2004-02-07 Alan Modra <amodra@bigpond.net.au>
1952 * src/types.c: Use 16 byte long double for POWERPC64.
1954 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1956 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
1957 when the structure return address is passed in %o0.
1958 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
1959 (ffi_v9_layout_struct): Align the field following a nested structure
1960 on a word boundary. Use memmove instead of memcpy.
1961 (ffi_call): Update call to ffi_V9_return_struct.
1962 (ffi_prep_closure): Define 'ctx' only for V8.
1963 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
1964 and ffi_closure_sparc_inner_v9.
1965 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
1966 Always skip the structure return address. For structures and long
1967 doubles, copy the argument directly.
1968 (ffi_closure_sparc_inner_v9): Skip the structure return address only
1969 if required. Shift the maximum floating-point slot accordingly. For
1970 big structures, copy the argument directly; otherwise, left-justify the
1971 argument and call ffi_v9_layout_struct to lay out the structure on
1973 * src/sparc/v8.S: Undef STACKFRAME before defining it.
1974 (ffi_closure_v8): Pass the structure return address. Update call to
1975 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
1976 Skip the 'unimp' insn when returning long doubles and structures.
1977 * src/sparc/v9.S: Undef STACKFRAME before defining it.
1978 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
1979 FFI_TYPE_INT handling. Load structures both in integers and
1980 floating-point registers on return.
1981 * README: Update status of the SPARC port.
1983 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
1985 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
1987 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
1989 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
1991 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
1992 value as of type ffi_arg, not unsigned int.
1994 2004-01-21 Michael Ritzert <ritzert@t-online.de>
1996 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
1999 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2001 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2004 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2006 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2009 2003-12-10 Richard Henderson <rth@redhat.com>
2011 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2012 size_t instead of int.
2014 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2016 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2017 * testsuite/libffi.call/many_win32.c (main): Replace variable
2018 int i with unsigned long ul.
2020 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2021 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2022 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2023 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2024 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2025 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2026 * testsuite/libffi.call/cls_align_float.c: Likewise.
2027 * testsuite/libffi.call/cls_align_double.c: Likewise.
2028 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2029 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2031 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2034 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2035 Align arguments to 32 bits.
2037 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2040 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2041 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2042 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2043 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2044 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2045 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2047 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2049 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2051 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2052 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2054 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2056 * Makefile.in: Rebuilt.
2057 * configure: Likewise.
2058 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2061 2003-11-21 Andreas Jaeger <aj@suse.de>
2062 Andreas Tobler <a.tobler@schweiz.ch>
2064 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2065 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2066 * Makefile.in: Rebuilt.
2067 * aclocal.m4: Likewise.
2068 * configure: Likewise.
2069 * fficonfig.h.in: Likewise.
2070 * testsuite/lib/libffi-dg.exp: Add include dir.
2071 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2072 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2073 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2074 for ffi_closure if available.
2075 * testsuite/libffi.call/closure_fn1.c: Likewise.
2076 * testsuite/libffi.call/closure_fn2.c: Likewise.
2077 * testsuite/libffi.call/closure_fn3.c: Likewise.
2078 * testsuite/libffi.call/closure_fn4.c: Likewise.
2079 * testsuite/libffi.call/closure_fn5.c: Likewise.
2080 * testsuite/libffi.call/cls_12byte.c: Likewise.
2081 * testsuite/libffi.call/cls_16byte.c: Likewise.
2082 * testsuite/libffi.call/cls_18byte.c: Likewise.
2083 * testsuite/libffi.call/cls_19byte.c: Likewise.
2084 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2085 * testsuite/libffi.call/cls_20byte.c: Likewise.
2086 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2087 * testsuite/libffi.call/cls_24byte.c: Likewise.
2088 * testsuite/libffi.call/cls_2byte.c: Likewise.
2089 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2090 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2091 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2092 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2093 * testsuite/libffi.call/cls_4byte.c: Likewise.
2094 * testsuite/libffi.call/cls_5byte.c: Likewise.
2095 * testsuite/libffi.call/cls_64byte.c: Likewise.
2096 * testsuite/libffi.call/cls_6byte.c: Likewise.
2097 * testsuite/libffi.call/cls_7byte.c: Likewise.
2098 * testsuite/libffi.call/cls_8byte.c: Likewise.
2099 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2100 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2101 * testsuite/libffi.call/cls_double.c: Likewise.
2102 * testsuite/libffi.call/cls_float.c: Likewise.
2103 * testsuite/libffi.call/cls_schar.c: Likewise.
2104 * testsuite/libffi.call/cls_sint.c: Likewise.
2105 * testsuite/libffi.call/cls_sshort.c: Likewise.
2106 * testsuite/libffi.call/cls_uchar.c: Likewise.
2107 * testsuite/libffi.call/cls_uint.c: Likewise.
2108 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2109 * testsuite/libffi.call/cls_ushort.c: Likewise.
2110 * testsuite/libffi.call/nested_struct.c: Likewise.
2111 * testsuite/libffi.call/nested_struct1.c: Likewise.
2112 * testsuite/libffi.call/nested_struct2.c: Likewise.
2113 * testsuite/libffi.call/nested_struct3.c: Likewise.
2114 * testsuite/libffi.call/problem1.c: Likewise.
2115 * testsuite/libffi.special/unwindtest.cc: Likewise.
2117 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2119 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2121 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2123 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2124 Add -lgcc_s to additional flags.
2126 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2128 * configure.in, include/Makefile.am: PR libgcj/11147, install
2129 the ffitarget.h header file in a gcc versioned and target
2131 * configure: Regenerated.
2132 * Makefile.in, include/Makefile.in: Likewise.
2133 * testsuite/Makefile.in: Likewise.
2135 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2137 * testsuite/libffi.call/closure_fn0.c: Print result and check
2138 with dg-output to make debugging easier.
2139 * testsuite/libffi.call/closure_fn1.c: Likewise.
2140 * testsuite/libffi.call/closure_fn2.c: Likewise.
2141 * testsuite/libffi.call/closure_fn3.c: Likewise.
2142 * testsuite/libffi.call/closure_fn4.c: Likewise.
2143 * testsuite/libffi.call/closure_fn5.c: Likewise.
2144 * testsuite/libffi.call/cls_12byte.c: Likewise.
2145 * testsuite/libffi.call/cls_16byte.c: Likewise.
2146 * testsuite/libffi.call/cls_18byte.c: Likewise.
2147 * testsuite/libffi.call/cls_19byte.c: Likewise.
2148 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2149 * testsuite/libffi.call/cls_20byte.c: Likewise.
2150 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2151 * testsuite/libffi.call/cls_24byte.c: Likewise.
2152 * testsuite/libffi.call/cls_2byte.c: Likewise.
2153 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2154 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2155 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2156 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2157 * testsuite/libffi.call/cls_4byte.c: Likewise.
2158 * testsuite/libffi.call/cls_5byte.c: Likewise.
2159 * testsuite/libffi.call/cls_64byte.c: Likewise.
2160 * testsuite/libffi.call/cls_6byte.c: Likewise.
2161 * testsuite/libffi.call/cls_7byte.c: Likewise.
2162 * testsuite/libffi.call/cls_8byte.c: Likewise.
2163 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2164 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2165 * testsuite/libffi.call/cls_double.c: Likewise.
2166 * testsuite/libffi.call/cls_float.c: Likewise.
2167 * testsuite/libffi.call/cls_schar.c: Likewise.
2168 * testsuite/libffi.call/cls_sint.c: Likewise.
2169 * testsuite/libffi.call/cls_sshort.c: Likewise.
2170 * testsuite/libffi.call/cls_uchar.c: Likewise.
2171 * testsuite/libffi.call/cls_uint.c: Likewise.
2172 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2173 * testsuite/libffi.call/cls_ushort.c: Likewise.
2174 * testsuite/libffi.call/problem1.c: Likewise.
2176 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2179 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2181 * testsuite/libffi.call/cls_9byte2.c: New test case.
2182 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2183 * testsuite/libffi.call/cls_64byte.c: Likewise.
2184 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2185 * testsuite/libffi.call/cls_19byte.c: Likewise.
2186 * testsuite/libffi.call/cls_18byte.c: Likewise.
2187 * testsuite/libffi.call/closure_fn4.c: Likewise.
2188 * testsuite/libffi.call/closure_fn5.c: Likewise.
2189 * testsuite/libffi.call/cls_schar.c: Likewise.
2190 * testsuite/libffi.call/cls_sint.c: Likewise.
2191 * testsuite/libffi.call/cls_sshort.c: Likewise.
2192 * testsuite/libffi.call/nested_struct2.c: Likewise.
2193 * testsuite/libffi.call/nested_struct3.c: Likewise.
2195 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2197 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2198 * testsuite/libffi.call/cls_uchar.c: Likewise.
2199 * testsuite/libffi.call/cls_uint.c: Likewise.
2200 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2201 * testsuite/libffi.call/cls_ushort.c: Likewise.
2202 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2204 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2206 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2209 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2211 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2212 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2214 2003-10-22 David Daney <ddaney@avtrex.com>
2216 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2217 __attribute__((__mode__(__SI__))) and friends.
2219 2003-10-22 Andreas Schwab <schwab@suse.de>
2221 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2223 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2225 * configure.in: AC_LINK_FILES(ffitarget.h).
2226 * configure: Regenerate.
2227 * Makefile.in: Likewise.
2228 * include/Makefile.in: Likewise.
2229 * testsuite/Makefile.in: Likewise.
2230 * fficonfig.h.in: Likewise.
2232 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2233 Richard Henderson <rth@redhat.com>
2235 Avoid that ffi.h includes fficonfig.h.
2237 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2238 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2239 (TARGET_SRC_MIPS_SGI): Removed.
2240 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2241 (MIPS_SGI): Removed.
2242 (CLEANFILES): Removed.
2243 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2245 * acconfig.h: Removed.
2246 * configure.in: Compute sizeofs only for double and long double.
2247 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2248 into AC_DEFINE instead of using acconfig.h. Create
2249 include/ffitarget.h instead of include/fficonfig.h. Rename
2250 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2251 AC_DEFINE EH_FRAME_FLAGS.
2252 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2253 (hack_DATA): Add ffitarget.h.
2254 * include/ffi.h.in: Remove all system specific definitions.
2255 Declare raw API even if it is not installed, why bother?
2256 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2257 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2258 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2259 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2260 * include/ffi_common.h (bool): Do not define.
2261 (ffi_assert): Accept failed assertion.
2262 (ffi_type_test): Return void and accept file/line.
2263 (FFI_ASSERT): Pass stringized failed assertion.
2264 (FFI_ASSERT_AT): New macro.
2265 (FFI_ASSERT_VALID_TYPE): New macro.
2266 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2267 UINT64, SINT64): Define here with gcc's __attribute__ macro
2269 (FLOAT32, ALIGN): Define here instead of in ffi.h
2270 * include/ffi-mips.h: Removed. Its content moved to
2271 src/mips/ffitarget.h after separating assembly and C sections.
2272 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2273 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2274 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2275 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2276 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2277 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2278 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2279 (ffi_type_test): Rewritten.
2280 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2281 FFI_ASSERT_VALID_TYPE.
2282 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2283 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2284 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2285 src/s390/ffitarget.h, src/sh/ffitarget.h,
2286 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2287 src/x86/ffitarget.h: New files.
2288 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2289 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2290 src/powerpc/aix.S, src/powerpc/darwin.S,
2291 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2292 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2293 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2294 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2295 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2298 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2300 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2301 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2303 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2305 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2306 Used when FFI_DEBUG = 1.
2308 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2310 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2313 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2315 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2316 FFI_MIPS_O32 for O32 ABI.
2318 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2320 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2321 SPARC64. Cleanup whitespaces.
2323 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2325 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2326 strongarm, xscale. Cleanup whitespaces.
2327 * testsuite/libffi.call/closure_fn1.c: Likewise.
2328 * testsuite/libffi.call/closure_fn2.c: Likewise.
2329 * testsuite/libffi.call/closure_fn3.c: Likewise.
2330 * testsuite/libffi.call/cls_12byte.c: Likewise.
2331 * testsuite/libffi.call/cls_16byte.c: Likewise.
2332 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2333 * testsuite/libffi.call/cls_20byte.c: Likewise.
2334 * testsuite/libffi.call/cls_24byte.c: Likewise.
2335 * testsuite/libffi.call/cls_2byte.c: Likewise.
2336 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2337 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2338 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2339 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2340 * testsuite/libffi.call/cls_4byte.c: Likewise.
2341 * testsuite/libffi.call/cls_5byte.c: Likewise.
2342 * testsuite/libffi.call/cls_6byte.c: Likewise.
2343 * testsuite/libffi.call/cls_7byte.c: Likewise.
2344 * testsuite/libffi.call/cls_8byte.c: Likewise.
2345 * testsuite/libffi.call/cls_double.c: Likewise.
2346 * testsuite/libffi.call/cls_float.c: Likewise.
2347 * testsuite/libffi.call/cls_uchar.c: Likewise.
2348 * testsuite/libffi.call/cls_uint.c: Likewise.
2349 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2350 * testsuite/libffi.call/cls_ushort.c: Likewise.
2351 * testsuite/libffi.call/nested_struct.c: Likewise.
2352 * testsuite/libffi.call/nested_struct1.c: Likewise.
2353 * testsuite/libffi.call/problem1.c: Likewise.
2354 * testsuite/libffi.special/unwindtest.cc: Likewise.
2355 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2357 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2359 * src/powerpc/aix.S: Cleanup whitespaces.
2360 * src/powerpc/aix_closure.S: Likewise.
2362 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2364 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2365 * src/powerpc/darwin_closure.S: Likewise.
2366 * src/powerpc/ffi_darwin.c: Likewise.
2368 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2369 David Edelsohn <edelsohn@gnu.org>
2371 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2372 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2374 * src/powerpc/darwin_closure.S: Likewise.
2375 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2376 according to the Darwin/AIX ABI.
2377 (ffi_prep_cif_machdep): Likewise.
2378 (ffi_closure_helper_DARWIN): Likewise.
2379 Remove the outgoing parameter stack logic. Simplify the evaluation
2380 of the different CASE types.
2381 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2382 statement in the trampoline code.
2384 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2386 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2387 for the register size.
2388 (ffi_closure_helper_SYSV): Handle the structure return value
2390 (ffi_closure_helper_SYSV): Return the appropriate type when
2391 the registers are used for the structure return value.
2392 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2393 the 64-bit return value. Update copyright years.
2395 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2397 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2398 srcdir for ffi_mips.h.
2400 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2402 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2404 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2405 placement of float result.
2406 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2407 cast of "resp" for big-endian 64 bit machines.
2409 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2411 * src/types.c (double, longdouble): Merge identical SH and ARM
2412 typedefs, and add POWERPC64.
2413 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2414 struct split over gpr and rest.
2415 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2416 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2418 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2420 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2423 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2425 * configure: Regenerate.
2427 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2429 * Makefile.am: Remove build rules for ffitest.
2430 * Makefile.in: Rebuilt.
2432 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2434 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2435 about implicit declaration of abort().
2437 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2439 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2440 * Makefile.in: Rebuilt.
2441 * configure.in: Add dejagnu test framework.
2442 * configure: Rebuilt.
2444 * testsuite/Makefile.am: New file.
2445 * testsuite/Makefile.in: Built
2446 * testsuite/lib/libffi-dg.exp: New file.
2447 * testsuite/config/default.exp: Likewise.
2448 * testsuite/libffi.call/call.exp: Likewise.
2449 * testsuite/libffi.call/ffitest.h: Likewise.
2450 * testsuite/libffi.call/closure_fn0.c: Likewise.
2451 * testsuite/libffi.call/closure_fn1.c: Likewise.
2452 * testsuite/libffi.call/closure_fn2.c: Likewise.
2453 * testsuite/libffi.call/closure_fn3.c: Likewise.
2454 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2455 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2456 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2457 * testsuite/libffi.call/cls_2byte.c: Likewise.
2458 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2459 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2460 * testsuite/libffi.call/cls_4byte.c: Likewise.
2461 * testsuite/libffi.call/cls_5byte.c: Likewise.
2462 * testsuite/libffi.call/cls_6byte.c: Likewise.
2463 * testsuite/libffi.call/cls_7byte.c: Likewise.
2464 * testsuite/libffi.call/cls_8byte.c: Likewise.
2465 * testsuite/libffi.call/cls_12byte.c: Likewise.
2466 * testsuite/libffi.call/cls_16byte.c: Likewise.
2467 * testsuite/libffi.call/cls_20byte.c: Likewise.
2468 * testsuite/libffi.call/cls_24byte.c: Likewise.
2469 * testsuite/libffi.call/cls_double.c: Likewise.
2470 * testsuite/libffi.call/cls_float.c: Likewise.
2471 * testsuite/libffi.call/cls_uchar.c: Likewise.
2472 * testsuite/libffi.call/cls_uint.c: Likewise.
2473 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2474 * testsuite/libffi.call/cls_ushort.c: Likewise.
2475 * testsuite/libffi.call/float.c: Likewise.
2476 * testsuite/libffi.call/float1.c: Likewise.
2477 * testsuite/libffi.call/float2.c: Likewise.
2478 * testsuite/libffi.call/many.c: Likewise.
2479 * testsuite/libffi.call/many_win32.c: Likewise.
2480 * testsuite/libffi.call/nested_struct.c: Likewise.
2481 * testsuite/libffi.call/nested_struct1.c: Likewise.
2482 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2483 * testsuite/libffi.call/problem1.c: Likewise.
2484 * testsuite/libffi.call/promotion.c: Likewise.
2485 * testsuite/libffi.call/return_ll.c: Likewise.
2486 * testsuite/libffi.call/return_sc.c: Likewise.
2487 * testsuite/libffi.call/return_uc.c: Likewise.
2488 * testsuite/libffi.call/strlen.c: Likewise.
2489 * testsuite/libffi.call/strlen_win32.c: Likewise.
2490 * testsuite/libffi.call/struct1.c: Likewise.
2491 * testsuite/libffi.call/struct2.c: Likewise.
2492 * testsuite/libffi.call/struct3.c: Likewise.
2493 * testsuite/libffi.call/struct4.c: Likewise.
2494 * testsuite/libffi.call/struct5.c: Likewise.
2495 * testsuite/libffi.call/struct6.c: Likewise.
2496 * testsuite/libffi.call/struct7.c: Likewise.
2497 * testsuite/libffi.call/struct8.c: Likewise.
2498 * testsuite/libffi.call/struct9.c: Likewise.
2499 * testsuite/libffi.special/special.exp: New file.
2500 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2501 * testsuite/libffi.special/unwindtest.cc: Likewise.
2504 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2506 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2509 2003-08-02 Alan Modra <amodra@bigpond.net.au>
2511 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2513 (ffi_closure_helper_LINUX64): Likewise.
2514 * src/powerpc/linux64.S: Remove code writing to parm save area.
2515 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2516 address in lr from ffi_closure_helper_LINUX64 call to calculate
2517 table address. Optimize function tail.
2519 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2521 * src/sparc/ffi.c: Handle all floating point registers.
2522 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2524 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2526 * README: Note that libffi is not part of GCC. Update the project
2529 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2531 * src/powerpc/ppc_closure.S: Include ffi.h.
2533 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2535 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2536 Use C style comments.
2538 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2540 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2541 * Makefile.in: Regenerate.
2542 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2543 * configure: Regenerate.
2544 * include/ffi.h.in: Add SHmedia support.
2545 * src/sh64/ffi.c: New file.
2546 * src/sh64/sysv.S: New file.
2548 2003-05-16 Jakub Jelinek <jakub@redhat.com>
2550 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2551 should be read-only.
2552 * configure: Rebuilt.
2553 * fficonfig.h.in: Rebuilt.
2554 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2555 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2556 * src/powerpc/linux64.S: Likewise.
2557 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2558 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2559 if -fpic/-fPIC/-mrelocatable.
2560 * src/powerpc/powerpc_closure.S: Likewise.
2561 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2562 #write in .eh_frame flags.
2563 * src/sparc/v9.S: Likewise.
2564 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2565 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2566 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2568 2003-05-07 Jeff Sturm <jsturm@one-point.com>
2570 Fixes PR bootstrap/10656
2571 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2572 support for .register pseudo-op.
2573 * src/sparc/v8.S: Use it.
2574 * fficonfig.h.in: Rebuilt.
2575 * configure: Rebuilt.
2577 2003-04-18 Jakub Jelinek <jakub@redhat.com>
2579 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2580 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2581 Make it the default on POWERPC64.
2582 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2583 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2584 * configure: Rebuilt.
2585 * src/powerpc/ffi.c (hidden): Define.
2586 (ffi_prep_args_SYSV): Renamed from
2587 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2588 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2589 ASM_NEEDS_REGISTERS64): New.
2590 (ffi_prep_args64): New function.
2591 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2592 (ffi_call): Likewise.
2593 (ffi_prep_closure): Likewise.
2594 (flush_icache): Surround by #ifndef POWERPC64.
2595 (ffi_dblfl): New union type.
2596 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2597 (ffi_closure_helper_LINUX64): New function.
2598 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2600 * src/powerpc/sysv.S: Likewise.
2601 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2602 * src/powerpc/linux64.S: New file.
2603 * src/powerpc/linux64_closure.S: New file.
2604 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2605 src/powerpc/linux64_closure.S.
2606 (TARGET_SRC_POWERPC): Likewise.
2608 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2609 closure_test_fn3): Fix result printing on big-endian 64-bit
2611 (main): Print tst2_arg instead of uninitialized tst2_result.
2613 * src/ffitest.c (main): Hide what closure pointer really points to
2616 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
2618 * configure.in (arm-*-netbsdelf*): Add configuration.
2619 (configure): Regenerated.
2621 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
2623 * include/Makefile.in: Regenerate.
2625 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2627 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2629 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2630 Receive closure pointer through parameter, read args using
2631 __builtin_dwarf_cfa.
2632 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2634 2003-03-12 Andreas Schwab <schwab@suse.de>
2636 * configure.in: Avoid trailing /. in toolexeclibdir.
2637 * configure: Rebuilt.
2639 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2641 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
2643 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2645 * libffi/src/powerpc/darwin_closure.S:
2646 Fix alignement bug, allocate 8 bytes for the result.
2647 * libffi/src/powerpc/aix_closure.S:
2649 * libffi/src/powerpc/ffi_darwin.c:
2650 Update stackframe description for aix/darwin_closure.S.
2652 2003-02-06 Jakub Jelinek <jakub@redhat.com>
2654 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2657 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2658 Andreas Schwab <schwab@suse.de>
2660 * configure.in: Adjust command to source config-ml.in to account
2661 for changes to the libffi_basedir definition.
2662 (libffi_basedir): Remove ${srcdir} from value and include trailing
2665 * configure: Regenerate.
2667 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2669 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2671 2003-01-28 Andrew Haley <aph@redhat.com>
2673 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2674 * src/x86/ffi64.c (ffi_prep_closure): New.
2675 (ffi_closure_UNIX64_inner): New.
2676 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2678 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
2680 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2681 Remove USE_LIBDIR conditional.
2682 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2683 * Makefile.in, configure: Rebuilt.
2685 2003-01027 David Edelsohn <edelsohn@gnu.org>
2687 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2688 * Makefile.in: Regenerate.
2690 2003-01-22 Andrew Haley <aph@redhat.com>
2692 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2695 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2697 * src/powerpc/darwin.S: Add unwind info.
2698 * src/powerpc/darwin_closure.S: Likewise.
2700 2003-01-14 Andrew Haley <aph@redhat.com>
2702 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2703 (ffi_prep_cif_machdep): Likewise.
2704 * src/x86/unix64.S: Add unwind info.
2706 2003-01-14 Andreas Jaeger <aj@suse.de>
2708 * src/ffitest.c (main): Only use ffi_closures if those are
2711 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2713 * libffi/src/ffitest.c
2714 add closure testcases
2716 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2718 * libffi/src/powerpc/ffi.c
2719 fix alignment bug for float (4 byte aligned iso 8 byte)
2721 2003-01-09 Geoffrey Keating <geoffk@apple.com>
2723 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2724 * src/powerpc/darwin.S: Remove RCS version string.
2726 2003-01-03 Jeff Sturm <jsturm@one-point.com>
2728 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2729 * src/ffitest.c (main): Use static storage for closure.
2730 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2731 * src/sparc/v8.S (ffi_closure_v8): New.
2732 * src/sparc/v9.S (ffi_closure_v9): New.
2734 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2736 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2737 enumeration for X86_WIN32.
2738 * src/x86/win32.S: Added ffi_call_STDCALL function
2740 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2741 switch cases for recognising FFI_STDCALL and
2742 calling ffi_call_STDCALL if target is X86_WIN32.
2743 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2744 stdcall versions of the "my_strlen" and "many"
2745 test functions (for X86_WIN32).
2746 Added test cases to test stdcall invocation using
2749 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2751 * src/sh/sysv.S: Add DWARF2 unwind info.
2753 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2755 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2757 2002-11-26 Jim Wilson <wilson@redhat.com>
2759 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2761 2002-11-23 H.J. Lu <hjl@gnu.org>
2763 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2764 Include ../config/accross.m4.
2765 * aclocal.m4; Rebuild.
2766 * configure: Likewise.
2768 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2770 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2772 2002-11-11 DJ Delorie <dj@redhat.com>
2774 * configure.in: Look for common files in the right place.
2776 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2778 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2779 raw data as _Jv_word values, not ffi_raw.
2780 (ffi_java_ptrarray_to_raw): Likewise.
2781 (ffi_java_rvalue_to_raw): New function.
2782 (ffi_java_raw_call): Call it.
2783 (ffi_java_raw_to_rvalue): New function.
2784 (ffi_java_translate_args): Call it.
2785 * src/ffitest.c (closure_test_fn): Interpret return value
2786 as ffi_arg, not int.
2787 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2788 FFI_TYPE_POINTER case.
2789 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2790 values extended to word size.
2792 2002-10-02 Andreas Jaeger <aj@suse.de>
2794 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2796 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
2798 * include/ffi.h.in: Fix i386 win32 compilation.
2800 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2802 * configure.in: Add s390x-*-linux-* target.
2803 * configure: Regenerate.
2804 * include/ffi.h.in: Define S390X for s390x targets.
2805 (FFI_CLOSURES): Define for s390/s390x.
2806 (FFI_TRAMPOLINE_SIZE): Likewise.
2807 (FFI_NATIVE_RAW_API): Likewise.
2808 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2809 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
2810 * src/s390/ffi.c: Major rework of existing code. Add support for
2811 s390x targets. Add closure support.
2812 * src/s390/sysv.S: Likewise.
2814 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
2816 * src/arm/sysv.S: Fix typo.
2818 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
2820 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2821 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2822 (ffi_call_SYSV): Handle soft-float.
2824 2002-09-27 Bo Thorsen <bo@suse.de>
2826 * include/ffi.h.in: Fix multilib x86-64 support.
2828 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2830 * Makefile.am (all-multi): Fix multilib parallel build.
2832 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2834 * configure.in (sh[34]*-*-linux*): Add brackets.
2835 * configure: Regenerate.
2837 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2839 * Makefile.am: Add SH support.
2840 * Makefile.in: Regenerate.
2841 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2842 * configure: Regenerate.
2843 * include/ffi.h.in: Add SH support.
2844 * src/sh/ffi.c: New file.
2845 * src/sh/sysv.S: New file.
2846 * src/types.c: Add SH support.
2848 2002-07-16 Bo Thorsen <bo@suse.de>
2850 * src/x86/ffi64.c: New file that adds x86-64 support.
2851 * src/x86/unix64.S: New file that handles argument setup for
2853 * src/x86/sysv.S: Don't use this on x86-64.
2854 * src/x86/ffi.c: Don't use this on x86-64.
2856 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2858 * src/ffitest.c (struct6): New test that tests a special case in
2860 (struct7): Likewise.
2861 (struct8): Likewise.
2862 (struct9): Likewise.
2863 (closure_test_fn): Silence warning about this when it's not used.
2864 (main): Add the new tests.
2865 (main): Fix a couple of wrong casts and silence some compiler warnings.
2866 * include/ffi.h.in: Add x86-64 ABI definition.
2867 * fficonfig.h.in: Regenerate.
2868 * Makefile.am: Add x86-64 support.
2869 * configure.in: Likewise.
2870 * Makefile.in: Regenerate.
2871 * configure: Likewise.
2873 2002-06-24 Bo Thorsen <bo@suse.de>
2875 * src/types.c: Merge settings for similar architectures.
2876 Add x86-64 sizes and alignments.
2878 2002-06-23 Bo Thorsen <bo@suse.de>
2880 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
2881 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
2882 * src/mips/ffi.c (ffi_prep_args): Likewise.
2883 * src/m68k/ffi.c (ffi_prep_args): Likewise.
2885 2002-07-18 H.J. Lu (hjl@gnu.org)
2887 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
2888 (libffi_la_SOURCES): Support MIPS_LINUX.
2889 (libffi_convenience_la_SOURCES): Likewise.
2890 * Makefile.in: Regenerated.
2892 * configure.in (mips64*-*): Skip.
2893 (mips*-*-linux*): New.
2894 * configure: Regenerated.
2896 * src/mips/ffi.c: Include <sgidefs.h>.
2898 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
2900 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
2902 2002-05-27 Roger Sayle <roger@eyesopen.com>
2904 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
2906 2002-05-27 Bo Thorsen <bo@suse.de>
2908 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
2911 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
2913 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
2914 beginning of function (for older apple cc).
2916 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
2918 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
2919 script entry, and set LD to it when configuring multilibs.
2920 * configure: Rebuilt.
2922 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
2924 * configure.in (sparc64-*-netbsd*): Add target.
2925 (sparc-*-netbsdelf*): Likewise.
2926 * configure: Regenerate.
2928 2002-04-28 David S. Miller <davem@redhat.com>
2930 * configure.in, configure: Fix SPARC test in previous change.
2932 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
2934 * Makefile.am: Add Linux for S/390 support.
2935 * Makefile.in: Regenerate.
2936 * configure.in: Add Linux for S/390 support.
2937 * configure: Regenerate.
2938 * include/ffi.h.in: Add Linux for S/390 support.
2939 * src/s390/ffi.c: New file from libffi CVS tree.
2940 * src/s390/sysv.S: New file from libffi CVS tree.
2942 2002-04-28 Jakub Jelinek <jakub@redhat.com>
2944 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
2946 * src/sparc/v8.S: Use it.
2947 * src/sparc/v9.S: Likewise.
2948 * fficonfig.h.in: Rebuilt.
2949 * configure: Rebuilt.
2951 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
2953 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
2955 * src/ia64/unix.S: Add unwind information. Fix comments.
2956 Save sp in a way that's compatible with unwind info.
2957 (ffi_call_unix): Correctly restore sp in all cases.
2958 * src/ia64/ffi.c: Add, fix comments.
2960 2002-04-08 Jakub Jelinek <jakub@redhat.com>
2962 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
2964 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
2966 * configure.in (alpha*-*-netbsd*): Add target.
2967 * configure: Regenerate.
2969 2002-04-04 Jeff Sturm <jsturm@one-point.com>
2971 * src/sparc/v8.S: Add unwind info.
2972 * src/sparc/v9.S: Likewise.
2974 2002-03-30 Krister Walfridsson <cato@df.lth.se>
2976 * configure.in: Enable i*86-*-netbsdelf*.
2977 * configure: Rebuilt.
2979 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
2982 * src/mips/n32.s: Delete
2983 * src/mips/o32.s: Delete
2985 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
2987 * configure.in: Enable alpha*-*-freebsd*.
2988 * configure: Rebuilt.
2990 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2992 * Makefile.am: libfficonvenience -> libffi_convenience.
2993 * Makefile.in: Rebuilt.
2995 * Makefile.am: Define ffitest_OBJECTS.
2996 * Makefile.in: Rebuilt.
2998 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
2999 David Edelsohn <edelsohn@gnu.org>
3001 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3002 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3003 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3004 * Makefile.in: Regenerate.
3005 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3006 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3007 (flush_icache, flush_range): New functions.
3008 (ffi_closure_helper_DARWIN): New function.
3009 * src/powerpc/aix_closure.S: New file.
3010 * src/powerpc/darwin_closure.S: New file.
3012 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3014 * include/ffi.h.in: Add typedef for ffi_arg.
3015 * src/ffitest.c (main): Declare rint with ffi_arg.
3017 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3019 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3020 number of GPRs for floating-point arguments.
3022 2002-01-31 Anthony Green <green@redhat.com>
3024 * configure: Rebuilt.
3025 * configure.in: Replace CHECK_SIZEOF and endian tests with
3026 cross-compiler friendly macros.
3027 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3030 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3032 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3033 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3035 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3037 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3038 (TARGET_SRC_POWERPC_AIX): New.
3039 (POWERPC_AIX): New stanza.
3040 * Makefile.in: Regenerate.
3041 * configure.in: Add AIX case.
3042 * configure: Regenerate.
3043 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3044 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3045 size. Fix "long double" support.
3046 (ffi_call): Add FFI_AIX case.
3047 * src/powerpc/aix.S: New.
3049 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3051 Implement Darwin PowerPC ABI.
3052 * configure.in: Handle powerpc-*-darwin*.
3053 * Makefile.am: Set source files for POWERPC_DARWIN.
3054 * configure: Rebuilt.
3055 * Makefile.in: Rebuilt.
3056 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3058 * src/powerpc/darwin.S: New file.
3059 * src/powerpc/ffi_darwin.c: New file.
3061 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3063 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3065 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3067 * src/x86/sysv.S: Avoid gas-only .balign directive.
3068 Use C style comments.
3070 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3072 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3073 Fixes PR bootstrap/3563.
3075 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3077 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3079 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3081 * configure.in: Recognize sparc*-sun-* host.
3082 * configure: Regenerate.
3084 2001-06-06 Andrew Haley <aph@redhat.com>
3086 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
3088 2001-06-03 Andrew Haley <aph@redhat.com>
3090 * src/alpha/osf.S: Add unwind info.
3091 * src/powerpc/sysv.S: Add unwind info.
3092 * src/powerpc/ppc_closure.S: Likewise.
3094 2000-05-31 Jeff Sturm <jsturm@one-point.com>
3096 * configure.in: Fix AC_ARG_ENABLE usage.
3097 * configure: Rebuilt.
3099 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3101 * configure.in: Remove warning about beta code.
3102 * configure: Rebuilt.
3104 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3106 * src/ia64/unix.S: Restore stack pointer when returning from
3108 * src/ia64/ffi.c: Fix typo in comment.
3110 2001-04-18 Jim Wilson <wilson@redhat.com>
3112 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3113 to eliminate RAW DV.
3115 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3117 * Makefile.am: Make a libtool convenience library.
3118 * Makefile.in: Rebuilt.
3120 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3122 * configure.in: Use different syntax for subdirectory creation.
3123 * configure: Rebuilt.
3125 2001-03-27 Jon Beniston <jon@beniston.com>
3127 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3128 * configure: Rebuilt.
3129 * Makefile.am: Added X86_WIN32 target support.
3130 * Makefile.in: Rebuilt.
3132 * include/ffi.h.in: Added X86_WIN32 target support.
3134 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3135 * src/types.c: Added X86_WIN32 target support.
3137 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3138 stuff removed and made to work with CygWin's gas.
3140 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3142 * configure.in: Make target subdirectory in build dir.
3143 * Makefile.am: Override suffix based rules to specify correct output
3145 * Makefile.in: Rebuilt.
3146 * configure: Rebuilt.
3148 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3150 * src/powerpc/ppc_closure.S: New file.
3151 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3152 involving long long and register pairs.
3153 (ffi_prep_closure): New function.
3154 (flush_icache): Likewise.
3155 (ffi_closure_helper_SYSV): Likewise.
3156 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3157 (FFI_TRAMPOLINE_SIZE): Likewise.
3158 (FFI_NATIVE_RAW_API): Likewise.
3159 * Makefile.in: Rebuilt.
3160 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3161 (TARGET_SRC_POWERPC): Likewise.
3163 2001-03-19 Tom Tromey <tromey@redhat.com>
3165 * Makefile.in: Rebuilt.
3166 * Makefile.am (ffitest_LDFLAGS): New macro.
3168 2001-03-02 Nick Clifton <nickc@redhat.com>
3170 * include/ffi.h.in: Remove RCS ident string.
3171 * include/ffi_mips.h: Remove RCS ident string.
3172 * src/debug.c: Remove RCS ident string.
3173 * src/ffitest.c: Remove RCS ident string.
3174 * src/prep_cif.c: Remove RCS ident string.
3175 * src/types.c: Remove RCS ident string.
3176 * src/alpha/ffi.c: Remove RCS ident string.
3177 * src/alpha/osf.S: Remove RCS ident string.
3178 * src/arm/ffi.c: Remove RCS ident string.
3179 * src/arm/sysv.S: Remove RCS ident string.
3180 * src/mips/ffi.c: Remove RCS ident string.
3181 * src/mips/n32.S: Remove RCS ident string.
3182 * src/mips/o32.S: Remove RCS ident string.
3183 * src/sparc/ffi.c: Remove RCS ident string.
3184 * src/sparc/v8.S: Remove RCS ident string.
3185 * src/sparc/v9.S: Remove RCS ident string.
3186 * src/x86/ffi.c: Remove RCS ident string.
3187 * src/x86/sysv.S: Remove RCS ident string.
3189 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3191 * include/ffi.h.in: Change sourceware.cygnus.com references to
3194 2000-12-09 Richard Henderson <rth@redhat.com>
3196 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3197 (ffi_closure_osf_inner): Index rather than increment avalue
3198 and arg_types. Give ffi_closure_osf the raw return value type.
3199 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3202 2000-12-07 Richard Henderson <rth@redhat.com>
3204 * src/raw_api.c (ffi_translate_args): Fix typo.
3205 (ffi_prep_closure): Likewise.
3207 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3208 FFI_TRAMPOLINE_SIZE.
3209 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3210 cif->bytes for new ffi_call_osf implementation.
3211 (ffi_prep_args): Absorb into ...
3212 (ffi_call): ... here. Do all stack allocation here and
3213 avoid a callback function.
3214 (ffi_prep_closure, ffi_closure_osf_inner): New.
3215 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3216 (ffi_closure_osf): New.
3218 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3220 * config.guess, config.sub, install-sh: Removed.
3221 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3222 * Makefile.in: Rebuilt.
3224 * acinclude.m4: Include libtool macros from the top level.
3225 * aclocal.m4, configure: Rebuilt.
3227 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3229 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3230 * configure: Rebuilt.
3232 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3234 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3235 memory correctly. Use conditional instructions, not branches where
3238 2000-05-04 Tom Tromey <tromey@cygnus.com>
3240 * configure: Rebuilt.
3241 * configure.in: Match `arm*-*-linux-*'.
3242 From Chris Dornan <cdornan@arm.com>.
3244 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3246 * Makefile.am (SUBDIRS): Define.
3247 (AM_MAKEFLAGS): Likewise.
3248 (Multilib support.): Add section.
3249 * Makefile.in: Rebuilt.
3250 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3251 New variables. Set for gcc using -print-multi-lib. Export them
3253 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3254 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3256 (extra_compiler_flags_value, extra_compiler_flags): Check these
3257 for extra compiler options which need to be passed down in
3260 2000-04-16 Anthony Green <green@redhat.com>
3262 * configure: Rebuilt.
3263 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3265 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3267 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3268 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3269 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3270 Replace all void * sizeofs with sizeof(int).
3271 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3272 different than DOUBLE.
3273 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3274 (ffi_prep_args_v9): New function.
3275 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3276 (ffi_V9_return_struct): New function.
3277 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3278 32bit code (not yet cross-arch calls).
3279 * src/sparc/v8.S: Add struct return delay nop.
3281 * src/sparc/v9.S: New file.
3282 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3283 is used on sparc64 only for structures larger than 32 bytes.
3284 Pass by reference for structures is done for structure arguments
3285 larger than 16 bytes.
3286 * src/ffitest.c (main): Use 64bit rint on sparc64.
3287 Run long long tests on sparc.
3288 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3290 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3292 * configure.in (sparc-*-linux*): New supported target.
3293 (sparc64-*-linux*): Likewise.
3294 * configure: Rebuilt.
3295 * Makefile.am: Add v9.S to SPARC files.
3296 * Makefile.in: Likewise.
3297 (LINK): Surround $(CCLD) into double quotes, so that multilib
3298 compiles work correctly.
3300 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3302 * configure: Rebuilt.
3303 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3304 proposed by Bryce McKinlay.
3306 2000-03-20 Tom Tromey <tromey@cygnus.com>
3308 * Makefile.in: Hand edit for java_raw_api.lo.
3310 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3312 * config.guess, config.sub: Update from the gcc tree.
3313 Fix for PR libgcj/168.
3315 2000-03-03 Tom Tromey <tromey@cygnus.com>
3317 * Makefile.in: Fixed ia64 by hand.
3319 * configure: Rebuilt.
3320 * configure.in (--enable-multilib): New option.
3321 (libffi_basedir): New subst.
3322 (AC_OUTPUT): Added multilib code.
3324 2000-03-02 Tom Tromey <tromey@cygnus.com>
3326 * Makefile.in: Rebuilt.
3327 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3330 2000-02-25 Hans Boehm <boehm@acm.org>
3332 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3334 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3336 (ffi_prep_raw_closure): Use ffi_translate_args, not
3337 ffi_closure_translate.
3338 * src/java_raw_api.c: New file.
3339 * src/ffitest.c (closure_test_fn): New function.
3340 (main): Define `rint' as long long on IA64. Added new test when
3341 FFI_CLOSURES is defined.
3342 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3343 (ffi_abi): Recognize IA64.
3344 (ffi_raw): Added `flt' field.
3345 Added "Java raw API" code.
3346 * configure.in: Recognize ia64.
3347 * Makefile.am (TARGET_SRC_IA64): New macro.
3348 (libffi_la_common_SOURCES): Added java_raw_api.c.
3349 (libffi_la_SOURCES): Define in IA64 case.
3351 2000-01-04 Tom Tromey <tromey@cygnus.com>
3353 * Makefile.in: Rebuilt with newer automake.
3355 1999-12-31 Tom Tromey <tromey@cygnus.com>
3357 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3359 1999-09-01 Tom Tromey <tromey@cygnus.com>
3361 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3363 * fficonfig.h.in: Rebuilt.
3364 * configure: Rebuilt.
3365 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3366 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3367 * acconfig.h: Don't #undef PACKAGE or VERSION.
3369 1999-08-09 Anthony Green <green@cygnus.com>
3371 * include/ffi.h.in: Try to work around messy header problem
3372 with PACKAGE and VERSION.
3374 * configure: Rebuilt.
3375 * configure.in: Change version to 2.00-beta.
3377 * fficonfig.h.in: Rebuilt.
3378 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3380 * src/x86/ffi.c (ffi_raw_call): Rename.
3382 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3384 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3385 (ffi_prep_incoming_args_SYSV): Ditto.
3386 (ffi_prep_closure): Ditto.
3387 (ffi_closure_raw_SYSV): Ditto.
3388 (ffi_prep_raw_closure): More ditto.
3389 (ffi_call_raw): Final ditto.
3391 * include/ffi.h.in: Add definitions for closure and raw API.
3393 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3396 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3398 * src/raw_api.c: New file.
3400 * include/ffi.h.in (ffi_raw): New type.
3401 (UINT_ARG, SINT_ARG): New defines.
3402 (ffi_closure, ffi_raw_closure): New types.
3403 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3405 * configure.in: Add check for endianness and sizeof void*.
3407 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3408 instead of directly.
3410 * configure: Rebuilt.
3412 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3414 * configure.in: Add x86 and powerpc BeOS configurations.
3415 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3417 1999-05-09 Anthony Green <green@cygnus.com>
3419 * configure.in: Add warning about this being beta code.
3420 Remove src/Makefile.am from the picture.
3421 * configure: Rebuilt.
3423 * Makefile.am: Move logic from src/Makefile.am. Add changes
3424 to support libffi as a target library.
3425 * Makefile.in: Rebuilt.
3427 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3428 Upgraded to new autoconf, automake, libtool.
3432 * LICENSE: Update copyright date.
3434 * src/Makefile.am, src/Makefile.in: Removed.
3436 1998-11-29 Anthony Green <green@cygnus.com>
3438 * include/ChangeLog: Removed.
3439 * src/ChangeLog: Removed.
3440 * src/mips/ChangeLog: Removed.
3441 * src/sparc/ChangeLog: Remboved.
3442 * src/x86/ChangeLog: Removed.
3444 * ChangeLog.v1: Created.