1 2012-01-23 Andreas Schwab <schwab@linux-m68k.org>
3 * src/m68k/sysv.S (ffi_call_SYSV): Properly test for plain
4 mc68000. Test for __HAVE_68881__ in addition to __MC68881__.
6 2012-01-19 Jakub Jelinek <jakub@redhat.com>
8 PR rtl-optimization/48496
9 * src/ia64/ffi.c (ffi_call): Fix up aliasing violations.
11 2012-01-09 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
13 * configure.ac (i?86-*-*): Set TARGET to X86_64.
14 * configure: Regenerate.
16 2011-12-07 Andrew Pinski <apinski@cavium.com>
19 * src/mips/n32.S: Add ".set mips4".
21 2011-11-21 Andreas Tobler <andreast@fgznet.ch>
23 * configure: Regenerate.
25 2011-11-10 Richard Henderson <rth@redhat.com>
27 * configure.ac (GCC_AS_CFI_PSEUDO_OP): Use it instead of inline check.
28 * configure, aclocal.m4: Rebuild.
30 2011-09-04 Iain Sandoe <iains@gcc.gnu.org>
33 * src/powerpc/darwin_closure.S (stubs): Make the stub binding
34 helper reference track the architecture pointer size.
36 2011-08-25 Andrew Haley <aph@redhat.com>
38 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Remove hard-coded assembly
40 * src/arm/sysv.S (ffi_arm_trampoline): Put them here instead.
42 2011-07-11 Andrew Haley <aph@redhat.com>
44 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Clear icache.
46 2011-06-29 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
48 * testsuite/libffi.call/cls_double_va.c: Move PR number to comment.
49 * testsuite/libffi.call/cls_longdouble_va.c: Likewise.
51 2011-06-29 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
54 * testsuite/libffi.call/cls_double_va.c: xfail dg-output on
56 * testsuite/libffi.call/cls_longdouble_va.c: Likewise.
58 2011-06-14 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
60 * testsuite/libffi.call/huge_struct.c (test_large_fn): Use PRIu8,
61 PRId8 instead of %hhu, %hhd.
62 * testsuite/libffi.call/ffitest.h [__alpha__ && __osf__] (PRId8,
64 [__sgi__] (PRId8, PRIu8): Define.
66 2011-04-29 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
68 * src/alpha/osf.S (UA_SI, FDE_ENCODING, FDE_ENCODE, FDE_ARANGE):
70 Use them to handle ELF vs. ECOFF differences.
71 [__osf__] (_GLOBAL__F_ffi_call_osf): Define.
73 2011-02-13 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
75 * configure: Regenerate.
77 2011-02-09 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
80 * testsuite/libffi.call/cls_pointer.c (main): Cast void * to
82 * testsuite/libffi.call/cls_pointer_stack.c (main): Likewise.
84 2011-02-07 Joel Sherrill <joel.sherrill@oarcorp.com>
86 * libffi/src/m68k/ffi.c: Add RTEMS support for cache flushing.
87 Handle case when CPU variant does not have long double support.
88 * libffi/src/m68k/sysv.S: Add support for mc68000, Coldfire,
89 and cores with soft floating point.
91 2011-02-07 Joel Sherrill <joel.sherrill@oarcorp.com>
93 * configure.ac: Add mips*-*-rtems* support.
94 * configure: Regenerate.
95 * src/mips/ffitarget.h: Ensure needed constants are available
96 for targets which do not have sgidefs.h.
98 2011-01-26 Dave Korn <dave.korn.cygwin@gmail.com>
101 * configure.ac (AM_LTLDFLAGS): Add -bindir option for windows DLLs.
102 * configure: Regenerate.
104 2010-12-18 Iain Sandoe <iains@gcc.gnu.org>
108 * src/powerpc/darwin_closure.S: Provide Darwin64 implementation,
110 * src/powerpc/ffitarget.h (POWERPC_DARWIN64): New,
111 (FFI_TRAMPOLINE_SIZE): Update for Darwin64.
112 * src/powerpc/darwin.S: Provide Darwin64 implementation,
114 * src/powerpc/ffi_darwin.c: Likewise.
116 2010-12-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
118 * configure.ac (libffi_cv_as_ascii_pseudo_op): Use double
120 (libffi_cv_as_string_pseudo_op): Likewise.
121 * configure: Regenerate.
123 2010-12-03 Chung-Lin Tang <cltang@codesourcery.com>
125 * src/arm/sysv.S (ffi_closure_SYSV): Add UNWIND to .pad directive.
126 (ffi_closure_VFP): Same.
127 (ffi_call_VFP): Move down to before ffi_closure_VFP. Add '.fpu vfp'
130 2010-12-01 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
132 * testsuite/libffi.call/ffitest.h [__sgi] (PRId64, PRIu64): Define.
135 2010-11-29 Richard Henderson <rth@redhat.com>
136 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
138 * src/x86/sysv.S (FDE_ENCODING, FDE_ENCODE): Define.
139 (.eh_frame): Use FDE_ENCODING.
140 (.LASFDE1, .LASFDE2, LASFDE3): Simplify with FDE_ENCODE.
142 2010-11-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
144 * testsuite/lib/libffi-dg.exp: Rename ...
145 * testsuite/lib/libffi.exp: ... to this.
146 * libffi/testsuite/libffi.call/call.exp: Don't load libffi-dg.exp.
147 * libffi/testsuite/libffi.special/special.exp: Likewise.
149 2010-10-28 Chung-Lin Tang <cltang@codesourcery.com>
151 * src/arm/ffi.c (ffi_prep_args): Add VFP register argument handling
152 code, new parameter, and return value. Update comments.
153 (ffi_prep_cif_machdep): Add case for VFP struct return values. Add
154 call to layout_vfp_args().
155 (ffi_call_SYSV): Update declaration.
156 (ffi_call_VFP): New declaration.
157 (ffi_call): Add VFP struct return conditions. Call ffi_call_VFP()
159 (ffi_closure_VFP): New declaration.
160 (ffi_closure_SYSV_inner): Add new vfp_args parameter, update call to
161 ffi_prep_incoming_args_SYSV().
162 (ffi_prep_incoming_args_SYSV): Update parameters. Add VFP argument
164 (ffi_prep_closure_loc): Pass ffi_closure_VFP to trampoline
165 construction under VFP hard-float.
166 (rec_vfp_type_p): New function.
168 (place_vfp_arg): Same.
169 (layout_vfp_args): Same.
170 * src/arm/ffitarget.h (ffi_abi): Add FFI_VFP. Define FFI_DEFAULT_ABI
171 based on __ARM_PCS_VFP.
172 (FFI_EXTRA_CIF_FIELDS): Define for adding VFP hard-float specific
174 (FFI_TYPE_STRUCT_VFP_FLOAT): Define internally used type code.
175 (FFI_TYPE_STRUCT_VFP_DOUBLE): Same.
176 * src/arm/sysv.S (ffi_call_SYSV): Change call of ffi_prep_args() to
177 direct call. Move function pointer load upwards.
178 (ffi_call_VFP): New function.
179 (ffi_closure_VFP): Same.
181 * testsuite/lib/libffi-dg.exp (check-flags): New function.
182 (dg-skip-if): New function.
183 * testsuite/libffi.call/cls_double_va.c: Skip if target is arm*-*-*
184 and compiler options include -mfloat-abi=hard.
185 * testsuite/libffi.call/cls_longdouble_va.c: Same.
187 2010-10-01 Jakub Jelinek <jakub@redhat.com>
190 * src/x86/ffi64.c (ffi_prep_cif_machdep): Ensure cif->bytes is
192 * testsuite/libffi.call/many2.c: New test.
194 2010-08-20 Mark Wielaard <mjw@redhat.com>
196 * src/closures.c (open_temp_exec_file_mnt): Check if getmntent_r
199 2010-08-09 Andreas Tobler <andreast@fgznet.ch>
201 * configure.ac: Add target powerpc64-*-freebsd*.
202 * configure: Regenerate.
203 * testsuite/libffi.call/cls_align_longdouble_split.c: Pass
204 -mlong-double-128 only to linux targets.
205 * testsuite/libffi.call/cls_align_longdouble_split2.c: Likewise.
206 * testsuite/libffi.call/cls_longdouble.c: Likewise.
207 * testsuite/libffi.call/huge_struct.c: Likewise.
209 2010-07-10 Evan Phoenix <evan@fallingsnow.net>
211 * src/closures.c (selinux_enabled_check): Fix strncmp usage bug.
213 2010-07-07 Dan HorĂ¡k <dan@danny.cz>
215 * include/ffi.h.in: Protect #define with #ifndef.
216 * src/powerpc/ffitarget.h: Ditto.
217 * src/s390/ffitarget.h: Ditto.
218 * src/sparc/ffitarget.h: Ditto.
220 2010-07-07 Neil Roberts <neil@linux.intel.com>
222 * src/x86/sysv.S (ffi_call_SYSV): Align the stack pointer to
225 2010-07-02 Jakub Jelinek <jakub@redhat.com>
227 * Makefile.am (AM_MAKEFLAGS): Pass also mandir to submakes.
228 * Makefile.in: Regenerated.
230 2010-05-19 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
232 * configure.ac (libffi_cv_as_x86_pcrel): Check for illegal in as
234 (libffi_cv_as_ascii_pseudo_op): Check for .ascii.
235 (libffi_cv_as_string_pseudo_op): Check for .string.
236 * configure: Regenerate.
237 * fficonfig.h.in: Regenerate.
238 * src/x86/sysv.S (.eh_frame): Use .ascii, .string or error.
240 2010-05-05 Michael Kohler <michaelkohler@live.com>
242 * src/dlmalloc.c (dlfree): Fix spelling.
243 * src/ia64/ffi.c (ffi_prep_cif_machdep): Ditto.
244 * configure.ac: Ditto.
245 * configure: Rebuilt.
247 2010-04-13 Dan Witte <dwitte@mozilla.com>
249 * msvcc.sh: Build with -W3 instead of -Wall.
250 * src/powerpc/ffi_darwin.c: Remove build warnings.
251 * src/x86/ffi.c: Ditto.
252 * src/x86/ffitarget.h: Ditto.
254 2010-04-12 Dan Witte <dwitte@mozilla.com>
255 Walter Meinl <wuno@lsvw.de>
257 * configure.ac: Add OS/2 support.
258 * configure: Rebuilt.
259 * src/closures.c: Ditto.
260 * src/dlmalloc.c: Ditto.
261 * src/x86/win32.S: Ditto.
263 2010-04-07 Jakub Jelinek <jakub@redhat.com>
265 * testsuite/libffi.call/err_bad_abi.c: Remove unused args variable.
267 2010-04-02 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
269 * Makefile.in: Regenerate.
270 * aclocal.m4: Regenerate.
271 * include/Makefile.in: Regenerate.
272 * man/Makefile.in: Regenerate.
273 * testsuite/Makefile.in: Regenerate.
275 2010-03-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
277 * configure.ac (libffi_cv_as_x86_64_unwind_section_type): New test.
278 * configure: Regenerate.
279 * fficonfig.h.in: Regenerate.
280 * libffi/src/x86/unix64.S (.eh_frame)
281 [HAVE_AS_X86_64_UNWIND_SECTION_TYPE]: Use @unwind section type.
283 2010-03-14 Matthias Klose <doko@ubuntu.com>
285 * src/x86/ffi64.c: Fix typo in comment.
286 * src/x86/ffi.c: Use /* ... */ comment style.
288 2010-02-24 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
290 * doc/libffi.texi (The Closure API): Fix typo.
291 * doc/libffi.info: Remove.
293 2010-02-15 Matthias Klose <doko@ubuntu.com>
295 * src/arm/sysv.S (__ARM_ARCH__): Define for processor
298 2010-01-15 Anthony Green <green@redhat.com>
300 * README: Add notes on building with Microsoft Visual C++.
302 2010-01-15 Daniel Witte <dwitte@mozilla.com>
304 * msvcc.sh: New file.
306 * src/x86/win32.S: Port assembly routines to MSVC and #ifdef.
307 * src/x86/ffi.c: Tweak function declaration and remove excess
309 * include/ffi.h.in: Add __declspec(align(8)) to typedef struct
312 * src/x86/ffi.c: Merge ffi_call_SYSV and ffi_call_STDCALL into new
313 function ffi_call_win32 on X86_WIN32.
314 * src/x86/win32.S (ffi_call_SYSV): Rename to ffi_call_win32.
315 (ffi_call_STDCALL): Remove.
317 * src/prep_cif.c (ffi_prep_cif): Move stack space allocation code
318 to ffi_prep_cif_machdep for x86.
319 * src/x86/ffi.c (ffi_prep_cif_machdep): To here.
321 2010-01-15 Oliver Kiddle <okiddle@yahoo.co.uk>
323 * src/x86/ffitarget.h (ffi_abi): Check for __i386 and __amd64 for
324 Sun Studio compiler compatibility.
326 2010-01-12 Conrad Irwin <conrad.irwin@gmail.com>
328 * doc/libffi.texi: Add closure example.
330 2010-01-07 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
333 * testsuite/libffi.call/ffitest.h [__alpha__ && __osf__] (PRIdLL,
334 PRIuLL, PRId64, PRIu64, PRIuPTR): Define.
335 * testsuite/libffi.call/cls_align_sint64.c: Add -Wno-format on
337 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
338 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
339 * testsuite/libffi.call/return_ll1.c: Likewise.
340 * testsuite/libffi.call/stret_medium2.c: Likewise.
341 * testsuite/libffi.special/ffitestcxx.h (allocate_mmap): Cast
342 MAP_FAILED to char *.
344 2010-01-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
346 * src/mips/n32.S: Use .abicalls and .eh_frame with __GNUC__.
348 2009-12-31 Anthony Green <green@redhat.com>
350 * README: Update for libffi 3.0.9.
352 2009-12-27 Matthias Klose <doko@ubuntu.com>
354 * configure.ac (HAVE_LONG_DOUBLE): Define for mips when
356 * configure: Rebuilt.
358 2009-12-26 Anthony Green <green@redhat.com>
360 * testsuite/libffi.call/cls_longdouble_va.c: Mark as xfail for
362 * testsuite/libffi.call/cls_double_va.c: Ditto.
364 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
366 * testsuite/libffi.call/ffitest.h: Conditionally include stdint.h
368 * testsuite/libffi.special/unwindtest.cc: Ditto.
370 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
372 * configure.ac: Add amd64-*-openbsd*.
373 * configure: Rebuilt.
374 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Link
375 openbsd programs with -lpthread.
377 2009-12-26 Anthony Green <green@redhat.com>
379 * testsuite/libffi.call/cls_double_va.c,
380 testsuite/libffi.call/cls_longdouble.c,
381 testsuite/libffi.call/cls_longdouble_va.c,
382 testsuite/libffi.call/cls_pointer.c,
383 testsuite/libffi.call/cls_pointer_stack.c: Remove xfail for
384 mips*-*-* and arm*-*-*.
385 * testsuite/libffi.call/cls_align_longdouble_split.c,
386 testsuite/libffi.call/cls_align_longdouble_split2.c,
387 testsuite/libffi.call/stret_medium2.c,
388 testsuite/libffi.call/stret_medium.c,
389 testsuite/libffi.call/stret_large.c,
390 testsuite/libffi.call/stret_large2.c: Remove xfail for arm*-*-*.
392 2009-12-31 Kay Tietz <ktietz70@googlemail.com>
394 * testsuite/libffi.call/ffitest.h,
395 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRuLL): Fix
398 2009-12-31 Carlo Bramini <carlo.bramix@libero.it>
400 * configure.ac (AM_LTLDFLAGS): Define for windows hosts.
401 * Makefile.am (libffi_la_LDFLAGS): Add AM_LTLDFLAGS.
402 * configure: Rebuilt.
403 * Makefile.in: Rebuilt.
405 2009-12-31 Anthony Green <green@redhat.com>
408 * testsuite/libffi.call/huge_struct.c: New test case from Blake
411 2009-12-28 David Edelsohn <edelsohn@gnu.org>
413 * src/powerpc/ffi_darwin.c (ffi_prep_args): Copy abi and nargs to
415 (aix_adjust_aggregate_sizes): New function.
416 (ffi_prep_cif_machdep): Call it.
418 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
420 * configure.ac: Define FFI_MMAP_EXEC_WRIT for the given targets.
421 * configure: Regenerate.
422 * fficonfig.h.in: Likewise.
423 * src/closures.c: Remove the FFI_MMAP_EXEC_WRIT definition for
426 2009-12-26 Andreas Schwab <schwab@linux-m68k.org>
428 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Advance intarg_count
429 when a float arguments is passed in memory.
430 (ffi_closure_helper_SYSV): Mark general registers as used up when
431 a 64bit or soft-float long double argument is passed in memory.
433 2009-12-25 Matthias Klose <doko@ubuntu.com>
435 * man/ffi_call.3: Fix #include in examples.
436 * doc/libffi.texi: Add dircategory.
438 2009-12-25 Frank Everdij <f.p.x.everdij@tudelft.nl>
440 * include/ffi.h.in: Placed '__GNUC__' ifdef around
441 '__attribute__((aligned(8)))' in ffi_closure, fixes compile for
443 * include/ffi_common.h: Added '__sgi' define to non
444 '__attribute__((__mode__()))' integer typedefs.
445 * src/mips/ffi.c (ffi_call, ffi_closure_mips_inner_O32,
446 ffi_closure_mips_inner_N32): Added 'defined(_MIPSEB)' to BE check.
447 (ffi_closure_mips_inner_O32, ffi_closure_mips_inner_N32): Added
448 FFI_LONGDOUBLE support and alignment(N32 only).
449 * src/mips/ffitarget.h: Corrected '#include <sgidefs.h>' for IRIX and
450 fixed non '__attribute__((__mode__()))' integer typedefs.
451 * src/mips/n32.S: Put '#ifdef linux' around '.abicalls' and '.eh_frame'
452 since they are Linux/GNU Assembler specific.
454 2009-12-25 Bradley Smith <brad@brad-smith.co.uk>
456 * configure.ac, Makefile.am, src/avr32/ffi.c,
457 src/avr32/ffitarget.h,
458 src/avr32/sysv.S: Add AVR32 port.
459 * configure, Makefile.in: Rebuilt.
461 2009-12-21 Andreas Tobler <a.tobler@schweiz.org>
463 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
464 * configure: Regenerate.
466 2009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
468 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
470 2009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
472 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
475 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
477 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
480 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
482 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
484 2009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
487 * src/closures.c [X86_64 && __sun__ && __svr4__]
488 (FFI_MMAP_EXEC_WRIT): Define.
490 2009-12-08 David Daney <ddaney@caviumnetworks.com>
492 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
493 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
494 * testsuite/libffi.call/stret_large.c: Same.
495 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
496 * testsuite/libffi.call/stret_large2.c: Same.
497 * testsuite/libffi.call/stret_medium2.c: Same.
499 2009-12-07 David Edelsohn <edelsohn@gnu.org>
501 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
504 2009-12-05 David Edelsohn <edelsohn@gnu.org>
506 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
508 * src/powerpc/aix_closure.S: Same.
510 2009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
512 * Makefile.in: Regenerate.
513 * configure: Regenerate.
514 * include/Makefile.in: Regenerate.
515 * man/Makefile.in: Regenerate.
516 * testsuite/Makefile.in: Regenerate.
518 2009-12-04 David Edelsohn <edelsohn@gnu.org>
520 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
523 2009-12-04 Uros Bizjak <ubizjak@gmail.com>
526 * src/x86/ffi64.c (classify_argument): Update from
527 gcc/config/i386/i386.c.
528 (ffi_closure_unix64_inner): Do not use the address of two consecutive
529 SSE registers directly.
530 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
531 for x86_64 linux targets.
533 2009-12-04 David Edelsohn <edelsohn@gnu.org>
535 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
536 pfr for long double split between fpr13 and stack.
538 2009-12-03 David Edelsohn <edelsohn@gnu.org>
540 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
541 fparg_count twice for long double.
543 2009-12-03 David Edelsohn <edelsohn@gnu.org>
546 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
548 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
550 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
551 Remove xfails for x86 linux targets.
553 2009-12-02 David Edelsohn <edelsohn@gnu.org>
555 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
558 2009-12-01 David Edelsohn <edelsohn@gnu.org>
560 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
561 register usage. Call ffi_prep_args directly. Add long double
562 return value support.
563 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
564 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
565 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
566 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
568 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
570 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
571 Allocate result area between params and FPRs.
573 2009-11-30 David Edelsohn <edelsohn@gnu.org>
576 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
578 * src/powerpc/aix.S: Implement AIX64 version.
579 * src/powerpc/aix_closure.S: Implement AIX64 version.
580 (ffi_closure_ASM): Use extsb, lha and displament addresses.
581 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
583 (ffi_prep_cif_machdep): Same.
585 (ffi_closure_helper_DARWIN): Same.
587 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
590 * testsuite/libffi.call/testclosure.c: New test.
592 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
594 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
595 assembly version use of ___chkstk.
597 2009-09-23 Matthias Klose <doko@ubuntu.com>
599 PR libffi/40242, PR libffi/41443
600 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
601 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
602 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
603 Change the conditionals to __SOFTFP__ || __ARM_EABI__
604 for -mfloat-abi=softfp to work.
606 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
608 PR testsuite/32843 (strikes again)
609 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
610 enable proper extension on char and short.
612 2009-09-15 David Daney <ddaney@caviumnetworks.com>
614 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
615 handling for FFI_TYPE_POINTER.
616 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
617 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
618 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
619 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
620 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
621 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
622 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
623 structure and pointer returns.
624 (ffi_closure_N32): Add handling for pointer returns.
625 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
626 calc_n32_return_struct_flags): Handle soft-float.
627 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
628 (ffi_call_N32): Declare proper argument types.
629 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
632 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
634 * configure.ac (AC_PREREQ): Bump to 2.64.
636 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
638 * Makefile.am (install-html, install-pdf): Remove.
639 * Makefile.in: Regenerate.
641 * Makefile.in: Regenerate.
642 * aclocal.m4: Regenerate.
643 * configure: Regenerate.
644 * fficonfig.h.in: Regenerate.
645 * include/Makefile.in: Regenerate.
646 * man/Makefile.in: Regenerate.
647 * testsuite/Makefile.in: Regenerate.
649 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
651 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
653 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
656 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
657 return types for X86_WIN32.
658 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
659 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
660 _ffi_closure_STDCALL): Likewise.
662 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
663 (dlmmap, dlmunmap): Also use these functions on Cygwin.
665 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
670 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
673 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
675 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
676 to "" before adding paths. (This reinstates an assignment that
677 was removed by my 2009-06-30 commit, but changes the initial
678 value from "." to "".)
680 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
683 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
684 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
687 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
689 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
690 to ld_library_path. Use add_path. Add just find_libgcc_s
691 to ld_library_path, not every libgcc multilib directory.
693 2009-06-16 Wim Lewis <wiml@hhhh.org>
695 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
696 supposed to be callee-saved.
697 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
698 return buffer for odd-size structs.
700 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
703 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
704 allow_stack_execute for Darwin.
706 2009-06-16 Andrew Haley <aph@redhat.com>
708 * configure.ac (TARGETDIR): Add missing blank lines.
709 * configure: Regenerate.
711 2009-06-16 Andrew Haley <aph@redhat.com>
713 * testsuite/libffi.call/cls_align_sint64.c,
714 testsuite/libffi.call/cls_align_uint64.c,
715 testsuite/libffi.call/cls_longdouble_va.c,
716 testsuite/libffi.call/cls_ulonglong.c,
717 testsuite/libffi.call/return_ll1.c,
718 testsuite/libffi.call/stret_medium2.c: Fix printf format
720 * testsuite/libffi.call/ffitest.h,
721 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
723 2009-06-15 Andrew Haley <aph@redhat.com>
725 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
726 * testsuite/libffi.call/err_bad_abi.c: Likewise.
728 2009-06-12 Andrew Haley <aph@redhat.com>
730 * Makefile.am: Remove info_TEXINFOS.
732 2009-06-12 Andrew Haley <aph@redhat.com>
734 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
735 testsuite/libffi.call/cls_align_uint64.c,
736 testsuite/libffi.call/cls_ulonglong.c,
737 testsuite/libffi.call/return_ll1.c,
738 testsuite/libffi.call/stret_medium2.c: Fix printf format
740 testsuite/libffi.special/unwindtest.cc: include stdint.h.
742 2009-06-11 Timothy Wall <twall@users.sf.net>
747 include/ffi_common.h,
752 src/x86/win64.S (new),
753 README: Added win64 support (mingw or MSVC)
757 testsuite/Makefile.in,
759 aclocal.m4: Regenerated
760 * ltcf-c.sh: properly escape cygwin/w32 path
761 * man/ffi_call.3: Clarify size requirements for return value.
762 * src/x86/ffi64.c: Fix filename in comment.
763 * src/x86/win32.S: Remove unused extern.
765 * testsuite/libffi.call/closure_fn0.c,
766 testsuite/libffi.call/closure_fn1.c,
767 testsuite/libffi.call/closure_fn2.c,
768 testsuite/libffi.call/closure_fn3.c,
769 testsuite/libffi.call/closure_fn4.c,
770 testsuite/libffi.call/closure_fn5.c,
771 testsuite/libffi.call/closure_fn6.c,
772 testsuite/libffi.call/closure_stdcall.c,
773 testsuite/libffi.call/cls_12byte.c,
774 testsuite/libffi.call/cls_16byte.c,
775 testsuite/libffi.call/cls_18byte.c,
776 testsuite/libffi.call/cls_19byte.c,
777 testsuite/libffi.call/cls_1_1byte.c,
778 testsuite/libffi.call/cls_20byte.c,
779 testsuite/libffi.call/cls_20byte1.c,
780 testsuite/libffi.call/cls_24byte.c,
781 testsuite/libffi.call/cls_2byte.c,
782 testsuite/libffi.call/cls_3_1byte.c,
783 testsuite/libffi.call/cls_3byte1.c,
784 testsuite/libffi.call/cls_3byte2.c,
785 testsuite/libffi.call/cls_4_1byte.c,
786 testsuite/libffi.call/cls_4byte.c,
787 testsuite/libffi.call/cls_5_1_byte.c,
788 testsuite/libffi.call/cls_5byte.c,
789 testsuite/libffi.call/cls_64byte.c,
790 testsuite/libffi.call/cls_6_1_byte.c,
791 testsuite/libffi.call/cls_6byte.c,
792 testsuite/libffi.call/cls_7_1_byte.c,
793 testsuite/libffi.call/cls_7byte.c,
794 testsuite/libffi.call/cls_8byte.c,
795 testsuite/libffi.call/cls_9byte1.c,
796 testsuite/libffi.call/cls_9byte2.c,
797 testsuite/libffi.call/cls_align_double.c,
798 testsuite/libffi.call/cls_align_float.c,
799 testsuite/libffi.call/cls_align_longdouble.c,
800 testsuite/libffi.call/cls_align_longdouble_split.c,
801 testsuite/libffi.call/cls_align_longdouble_split2.c,
802 testsuite/libffi.call/cls_align_pointer.c,
803 testsuite/libffi.call/cls_align_sint16.c,
804 testsuite/libffi.call/cls_align_sint32.c,
805 testsuite/libffi.call/cls_align_sint64.c,
806 testsuite/libffi.call/cls_align_uint16.c,
807 testsuite/libffi.call/cls_align_uint32.c,
808 testsuite/libffi.call/cls_align_uint64.c,
809 testsuite/libffi.call/cls_dbls_struct.c,
810 testsuite/libffi.call/cls_double.c,
811 testsuite/libffi.call/cls_double_va.c,
812 testsuite/libffi.call/cls_float.c,
813 testsuite/libffi.call/cls_longdouble.c,
814 testsuite/libffi.call/cls_longdouble_va.c,
815 testsuite/libffi.call/cls_multi_schar.c,
816 testsuite/libffi.call/cls_multi_sshort.c,
817 testsuite/libffi.call/cls_multi_sshortchar.c,
818 testsuite/libffi.call/cls_multi_uchar.c,
819 testsuite/libffi.call/cls_multi_ushort.c,
820 testsuite/libffi.call/cls_multi_ushortchar.c,
821 testsuite/libffi.call/cls_pointer.c,
822 testsuite/libffi.call/cls_pointer_stack.c,
823 testsuite/libffi.call/cls_schar.c,
824 testsuite/libffi.call/cls_sint.c,
825 testsuite/libffi.call/cls_sshort.c,
826 testsuite/libffi.call/cls_uchar.c,
827 testsuite/libffi.call/cls_uint.c,
828 testsuite/libffi.call/cls_ulonglong.c,
829 testsuite/libffi.call/cls_ushort.c,
830 testsuite/libffi.call/err_bad_abi.c,
831 testsuite/libffi.call/err_bad_typedef.c,
832 testsuite/libffi.call/float2.c,
833 testsuite/libffi.call/huge_struct.c,
834 testsuite/libffi.call/nested_struct.c,
835 testsuite/libffi.call/nested_struct1.c,
836 testsuite/libffi.call/nested_struct10.c,
837 testsuite/libffi.call/nested_struct2.c,
838 testsuite/libffi.call/nested_struct3.c,
839 testsuite/libffi.call/nested_struct4.c,
840 testsuite/libffi.call/nested_struct5.c,
841 testsuite/libffi.call/nested_struct6.c,
842 testsuite/libffi.call/nested_struct7.c,
843 testsuite/libffi.call/nested_struct8.c,
844 testsuite/libffi.call/nested_struct9.c,
845 testsuite/libffi.call/problem1.c,
846 testsuite/libffi.call/return_ldl.c,
847 testsuite/libffi.call/return_ll1.c,
848 testsuite/libffi.call/stret_large.c,
849 testsuite/libffi.call/stret_large2.c,
850 testsuite/libffi.call/stret_medium.c,
851 testsuite/libffi.call/stret_medium2.c,
852 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
853 of checking for MMAP. Use intptr_t instead of long casts.
855 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
857 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
858 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
859 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
861 2009-06-09 Andrew Haley <aph@redhat.com>
863 * src/x86/freebsd.S: Add missing file.
865 2009-06-08 Andrew Haley <aph@redhat.com>
867 Import from libffi 3.0.8:
869 * doc/libffi.texi: New file.
870 * doc/libffi.info: Likewise.
871 * doc/stamp-vti: Likewise.
872 * man/Makefile.am: New file.
873 * man/ffi_call.3: New file.
875 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
877 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
879 * configure.ac: Bump version to 3.0.8.
880 parisc*-*-linux*: Add.
881 i386-*-freebsd* | i386-*-openbsd*: Add.
882 powerpc-*-beos*: Add.
883 AM_CONDITIONAL X86_FREEBSD: Add.
884 AC_CONFIG_FILES: Add man/Makefile.
886 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
888 2009-06-08 Andrew Haley <aph@redhat.com>
890 * README: Import from libffi 3.0.8.
892 2009-06-08 Andrew Haley <aph@redhat.com>
894 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
895 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
896 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
897 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
899 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
900 * testsuite/libffi.call/stret_medium.c: Likewise.
901 * testsuite/libffi.call/stret_large2.c: Likewise.
902 * testsuite/libffi.call/stret_large.c: Likewise.
904 2008-12-26 Timothy Wall <twall@users.sf.net>
906 * testsuite/libffi.call/cls_longdouble.c,
907 testsuite/libffi.call/cls_longdouble_va.c,
908 testsuite/libffi.call/cls_align_longdouble.c,
909 testsuite/libffi.call/cls_align_longdouble_split.c,
910 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
911 failures on x86_64 cygwin/mingw.
913 2008-12-22 Timothy Wall <twall@users.sf.net>
915 * testsuite/libffi.call/closure_fn0.c,
916 testsuite/libffi.call/closure_fn1.c,
917 testsuite/libffi.call/closure_fn2.c,
918 testsuite/libffi.call/closure_fn3.c,
919 testsuite/libffi.call/closure_fn4.c,
920 testsuite/libffi.call/closure_fn5.c,
921 testsuite/libffi.call/closure_fn6.c,
922 testsuite/libffi.call/closure_loc_fn0.c,
923 testsuite/libffi.call/closure_stdcall.c,
924 testsuite/libffi.call/cls_align_pointer.c,
925 testsuite/libffi.call/cls_pointer.c,
926 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
927 pointer to integer (intptr_t).
928 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
930 2008-07-24 Anthony Green <green@redhat.com>
932 * testsuite/libffi.call/cls_dbls_struct.c,
933 testsuite/libffi.call/cls_double_va.c,
934 testsuite/libffi.call/cls_longdouble.c,
935 testsuite/libffi.call/cls_longdouble_va.c,
936 testsuite/libffi.call/cls_pointer.c,
937 testsuite/libffi.call/cls_pointer_stack.c,
938 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
941 2008-03-04 Anthony Green <green@redhat.com>
945 * testsuite/libffi.call/cls_align_longdouble_split2.c
946 testsuite/libffi.call/cls_align_longdouble_split.c
947 testsuite/libffi.call/cls_dbls_struct.c
948 testsuite/libffi.call/cls_double_va.c
949 testsuite/libffi.call/cls_longdouble.c
950 testsuite/libffi.call/cls_longdouble_va.c
951 testsuite/libffi.call/cls_pointer.c
952 testsuite/libffi.call/cls_pointer_stack.c
953 testsuite/libffi.call/err_bad_abi.c
954 testsuite/libffi.call/err_bad_typedef.c
955 testsuite/libffi.call/stret_large2.c
956 testsuite/libffi.call/stret_large.c
957 testsuite/libffi.call/stret_medium2.c
958 testsuite/libffi.call/stret_medium.c: New tests from Apple.
960 2009-06-05 Andrew Haley <aph@redhat.com>
962 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
965 2009-06-04 Andrew Haley <aph@redhat.com>
967 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
970 2008-02-26 Anthony Green <green@redhat.com>
971 Thomas Heller <theller@ctypes.org>
973 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
976 2008-02-03 Timothy Wall <twall@users.sf.net>
978 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
979 offset based on code pointer, not data pointer.
981 2008-01-31 Timothy Wall <twall@users.sf.net>
983 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
985 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
987 * src/x86/win32.S: Add assembly for stdcall closure.
988 * src/x86/ffi.c: Initialize stdcall closure trampoline.
990 2009-06-04 Andrew Haley <aph@redhat.com>
992 * include/ffi.h.in: Change void (*)() to void (*)(void).
993 * src/x86/ffi.c: Likewise.
995 2009-06-04 Andrew Haley <aph@redhat.com>
997 * src/powerpc/ppc_closure.S: Insert licence header.
998 * src/powerpc/linux64_closure.S: Likewise.
999 * src/m68k/sysv.S: Likewise.
1001 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
1002 * src/powerpc/ffi.c: Likewise.
1003 * src/powerpc/ffi_darwin.c: Likewise.
1004 * src/m32r/ffi.c: Likewise.
1005 * src/sh64/ffi.c: Likewise.
1006 * src/x86/ffi64.c: Likewise.
1007 * src/alpha/ffi.c: Likewise.
1008 * src/alpha/osf.S: Likewise.
1009 * src/frv/ffi.c: Likewise.
1010 * src/s390/ffi.c: Likewise.
1011 * src/pa/ffi.c: Likewise.
1012 * src/pa/hpux32.S: Likewise.
1013 * src/ia64/unix.S: Likewise.
1014 * src/ia64/ffi.c: Likewise.
1015 * src/sparc/ffi.c: Likewise.
1016 * src/mips/ffi.c: Likewise.
1017 * src/sh/ffi.c: Likewise.
1019 2008-02-15 David Daney <ddaney@avtrex.com>
1021 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
1022 Define (conditionally), and use it to include cachectl.h.
1023 (ffi_prep_closure_loc): Fix cache flushing.
1024 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
1026 2009-06-04 Andrew Haley <aph@redhat.com>
1029 src/arm/ffitarget.h,
1032 src/powerpc/ffitarget.h,
1034 src/sh64/ffitarget.h,
1039 src/x86/ffitarget.h,
1045 src/alpha/ffitarget.h,
1048 src/m68k/ffitarget.h,
1049 src/frv/ffitarget.h,
1051 src/s390/ffitarget.h,
1053 src/cris/ffitarget.h,
1058 src/ia64/ffitarget.h,
1061 src/ia64/ia64_flags.h,
1065 src/sparc/ffitarget.h,
1068 src/mips/ffitarget.h,
1075 src/sh/sysv.S: Update license text.
1077 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
1079 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
1080 (.eh_frame): Add FDE for it.
1082 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
1084 * configure.ac: Also check if assembler supports pc-relative
1085 relocs on X86_WIN32 targets.
1086 * configure: Regenerate.
1087 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
1088 (_ffi_call_SYSV): Add missing function type symbol .def and
1089 add EH markup labels.
1090 (_ffi_call_STDCALL): Likewise.
1091 (_ffi_closure_SYSV): Likewise.
1092 (_ffi_closure_raw_SYSV): Likewise.
1093 (.eh_frame): Add hand-crafted EH data.
1095 2009-04-09 Jakub Jelinek <jakub@redhat.com>
1097 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
1098 version 3 of the GNU General Public License and to point readers
1099 at the COPYING3 file and the FSF's license web page.
1100 * testsuite/libffi.call/call.exp: Likewise.
1101 * testsuite/libffi.special/special.exp: Likewise.
1103 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1105 * configure: Regenerate.
1107 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1110 * configure.ac (HAVE_AS_X86_PCREL): New test.
1111 * configure: Regenerate.
1112 * fficonfig.h.in: Regenerate.
1113 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
1114 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
1115 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
1116 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
1117 * src/x86/unix64.S (.Lstore_table): Move to .text section.
1118 (.Lload_table): Likewise.
1119 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
1121 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1123 * configure: Regenerate.
1125 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
1127 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
1128 signed/unsigned int8/16 return values.
1129 * src/sparc/v8.S (ffi_call_v8): Likewise.
1130 (ffi_closure_v8): Likewise.
1132 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
1133 Steve Ellcey <sje@cup.hp.com>
1135 * configure: Regenerate for new libtool.
1136 * Makefile.in: Ditto.
1137 * include/Makefile.in: Ditto.
1138 * aclocal.m4: Ditto.
1140 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
1142 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
1143 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
1144 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
1145 Adjust copyright notice.
1146 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
1147 register or two register to use for FFI_SYSV structs.
1148 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
1149 (ffi_closure_helper_SYSV): Fix the return type for
1150 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
1151 Adjust copyright notice.
1153 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
1155 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
1158 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1160 * configure: Regenerate.
1161 * include/Makefile.in: Regenerate.
1162 * testsuite/Makefile.in: Regenerate.
1164 2008-06-07 Joseph Myers <joseph@codesourcery.com>
1166 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
1167 powerpc-*-beos*): Remove.
1168 * configure: Regenerate.
1170 2008-05-09 Julian Brown <julian@codesourcery.com>
1172 * Makefile.am (LTLDFLAGS): New.
1173 (libffi_la_LDFLAGS): Use above.
1174 * Makefile.in: Regenerate.
1176 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
1179 * aclocal.m4: Regenerate.
1180 * configure: Regenerate.
1182 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
1184 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
1185 * src/sh64/sysv.S: Likewise.
1187 2008-03-26 Daniel Jacobowitz <dan@debian.org>
1189 * src/arm/sysv.S: Fix ARM comment marker.
1191 2008-03-26 Jakub Jelinek <jakub@redhat.com>
1193 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
1194 * src/s390/sysv.S: Likewise.
1195 * src/powerpc/ppc_closure.S: Likewise.
1196 * src/powerpc/sysv.S: Likewise.
1197 * src/x86/unix64.S: Likewise.
1198 * src/x86/sysv.S: Likewise.
1199 * src/sparc/v8.S: Likewise.
1200 * src/sparc/v9.S: Likewise.
1201 * src/m68k/sysv.S: Likewise.
1202 * src/arm/sysv.S: Likewise.
1204 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1206 * aclocal.m4: Regenerate.
1207 * configure: Likewise.
1208 * Makefile.in: Likewise.
1209 * include/Makefile.in: Likewise.
1210 * testsuite/Makefile.in: Likewise.
1212 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
1213 Andreas Tobler <a.tobler@schweiz.org>
1215 * configure.ac: Add amd64-*-freebsd* target.
1216 * configure: Regenerate.
1218 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
1221 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
1224 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
1227 2008-01-24 David Edelsohn <edelsohn@gnu.org>
1229 * configure: Regenerate.
1231 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
1233 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
1235 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
1238 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
1239 signed/unsigned int8/16 for X86_DARWIN.
1240 Updated copyright info.
1241 Handle one and two byte structs with special cif->flags.
1242 * src/x86/ffitarget.h: Add special types for one and two byte structs.
1243 Updated copyright info.
1244 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
1246 Remove code to pop args from the stack after call.
1247 Special-case signed/unsigned for int8/16, one and two byte structs.
1248 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1249 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1251 Updated copyright info.
1253 2007-12-08 David Daney <ddaney@avtrex.com>
1255 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
1256 SUBU, add with ADDU and use smaller code sequences.
1258 2007-12-07 David Daney <ddaney@avtrex.com>
1260 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
1263 2007-12-06 David Daney <ddaney@avtrex.com>
1265 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
1267 (ffi_java_raw): New typedef.
1268 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
1269 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
1271 (ffi_java_raw_closure) : Same.
1272 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
1274 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
1275 FFI_SIZEOF_JAVA_RAW.
1276 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
1277 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
1278 sizeof(ffi_java_raw) for alignment calculations.
1279 (ffi_java_ptrarray_to_raw): Same.
1280 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
1281 if FFI_SIZEOF_JAVA_RAW == 4.
1282 (ffi_java_raw_to_rvalue): Same.
1283 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
1284 (ffi_java_translate_args): Same.
1285 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
1287 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
1289 2007-12-06 David Daney <ddaney@avtrex.com>
1291 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
1294 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
1297 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
1298 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
1299 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
1300 set the NUM_FPR_ARG_REGISTERS according to.
1301 Add support for potential soft-float support under hard-float
1303 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
1304 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
1305 to the FFI_LINUX_SOFT_FLOAT ABI.
1306 (ffi_prep_cif_machdep): Likewise.
1307 (ffi_closure_helper_SYSV): Likewise.
1308 * src/powerpc/ppc_closure.S: Make sure not to store float/double
1309 on archs where __NO_FPRS__ is true.
1310 Add FFI_TYPE_UINT128 support.
1311 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
1312 Adjust copyright notice.
1314 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
1316 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
1317 * include/ffi_common.h: ... here.
1320 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
1322 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
1323 * src/powerpc/linux64.S: Likewise.
1324 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
1325 * testsuite/libffi.call/return_ldl.c: New test.
1327 2007-09-04 <aph@redhat.com>
1329 * src/arm/sysv.S (UNWIND): New.
1330 (Whole file): Conditionally compile unwinder directives.
1331 * src/arm/sysv.S: Add unwinder directives.
1333 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
1334 Only treat r0 as a struct address if we're actually returning a
1336 Only copy the bytes that are actually within a struct.
1337 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
1338 is returned in r0, not passed by address.
1339 (ffi_call): Allocate a word-sized temporary for the case where
1340 a composite is returned in r0.
1341 (ffi_prep_incoming_args_SYSV): Align as necessary.
1343 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
1345 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
1346 directly using the sys_cacheflush syscall.
1348 2007-07-27 Andrew Haley <aph@redhat.com>
1350 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
1352 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
1354 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
1355 * configure.ac: Likewise.
1356 * Makefile.in: Regenerate.
1357 * include/Makefile.in: Likewise.
1358 * testsuite/Makefile.in: Likewise.
1359 * configure: Likewise.
1361 2007-08-24 David Daney <ddaney@avtrex.com>
1363 * testsuite/libffi.call/return_sl.c: New test.
1365 2007-08-10 David Daney <ddaney@avtrex.com>
1367 * testsuite/libffi.call/cls_multi_ushort.c,
1368 testsuite/libffi.call/cls_align_uint16.c,
1369 testsuite/libffi.call/nested_struct1.c,
1370 testsuite/libffi.call/nested_struct3.c,
1371 testsuite/libffi.call/cls_7_1_byte.c,
1372 testsuite/libffi.call/nested_struct5.c,
1373 testsuite/libffi.call/cls_double.c,
1374 testsuite/libffi.call/nested_struct7.c,
1375 testsuite/libffi.call/cls_sint.c,
1376 testsuite/libffi.call/nested_struct9.c,
1377 testsuite/libffi.call/cls_20byte1.c,
1378 testsuite/libffi.call/cls_multi_sshortchar.c,
1379 testsuite/libffi.call/cls_align_sint64.c,
1380 testsuite/libffi.call/cls_3byte2.c,
1381 testsuite/libffi.call/cls_multi_schar.c,
1382 testsuite/libffi.call/cls_multi_uchar.c,
1383 testsuite/libffi.call/cls_19byte.c,
1384 testsuite/libffi.call/cls_9byte1.c,
1385 testsuite/libffi.call/cls_align_float.c,
1386 testsuite/libffi.call/closure_fn1.c,
1387 testsuite/libffi.call/problem1.c,
1388 testsuite/libffi.call/closure_fn3.c,
1389 testsuite/libffi.call/cls_sshort.c,
1390 testsuite/libffi.call/closure_fn5.c,
1391 testsuite/libffi.call/cls_align_double.c,
1392 testsuite/libffi.call/nested_struct.c,
1393 testsuite/libffi.call/cls_2byte.c,
1394 testsuite/libffi.call/nested_struct10.c,
1395 testsuite/libffi.call/cls_4byte.c,
1396 testsuite/libffi.call/cls_6byte.c,
1397 testsuite/libffi.call/cls_8byte.c,
1398 testsuite/libffi.call/cls_multi_sshort.c,
1399 testsuite/libffi.call/cls_align_sint16.c,
1400 testsuite/libffi.call/cls_align_uint32.c,
1401 testsuite/libffi.call/cls_20byte.c,
1402 testsuite/libffi.call/cls_float.c,
1403 testsuite/libffi.call/nested_struct2.c,
1404 testsuite/libffi.call/cls_5_1_byte.c,
1405 testsuite/libffi.call/nested_struct4.c,
1406 testsuite/libffi.call/cls_24byte.c,
1407 testsuite/libffi.call/nested_struct6.c,
1408 testsuite/libffi.call/cls_64byte.c,
1409 testsuite/libffi.call/nested_struct8.c,
1410 testsuite/libffi.call/cls_uint.c,
1411 testsuite/libffi.call/cls_multi_ushortchar.c,
1412 testsuite/libffi.call/cls_schar.c,
1413 testsuite/libffi.call/cls_uchar.c,
1414 testsuite/libffi.call/cls_align_uint64.c,
1415 testsuite/libffi.call/cls_ulonglong.c,
1416 testsuite/libffi.call/cls_align_longdouble.c,
1417 testsuite/libffi.call/cls_1_1byte.c,
1418 testsuite/libffi.call/cls_12byte.c,
1419 testsuite/libffi.call/cls_3_1byte.c,
1420 testsuite/libffi.call/cls_3byte1.c,
1421 testsuite/libffi.call/cls_4_1byte.c,
1422 testsuite/libffi.call/cls_6_1_byte.c,
1423 testsuite/libffi.call/cls_16byte.c,
1424 testsuite/libffi.call/cls_18byte.c,
1425 testsuite/libffi.call/closure_fn0.c,
1426 testsuite/libffi.call/cls_9byte2.c,
1427 testsuite/libffi.call/closure_fn2.c,
1428 testsuite/libffi.call/closure_fn4.c,
1429 testsuite/libffi.call/cls_ushort.c,
1430 testsuite/libffi.call/closure_fn6.c,
1431 testsuite/libffi.call/cls_5byte.c,
1432 testsuite/libffi.call/cls_align_pointer.c,
1433 testsuite/libffi.call/cls_7byte.c,
1434 testsuite/libffi.call/cls_align_sint32.c,
1435 testsuite/libffi.special/unwindtest_ffi_call.cc,
1436 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
1438 2007-08-10 David Daney <ddaney@avtrex.com>
1441 * configure.ac: Don't treat mips64 as a special case.
1442 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
1443 * configure: Regenerate
1444 * Makefile.in: Ditto.
1445 * fficonfig.h.in: Ditto.
1446 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
1447 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
1448 (FFI_DEFAULT_ABI): Set for n64 case.
1449 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
1450 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
1451 (ffi_closure_N32): New function.
1452 (.eh_frame): New section
1453 * src/mips/o32.S: Clean up comments.
1454 (ffi_closure_O32): Pass ffi_closure parameter in $12.
1455 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
1456 _MIPS_SIM == _ABIN32 throughout.
1457 (FFI_MIPS_STOP_HERE): New, use in place of
1459 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1460 to support n32/n64 ABIs.
1461 (calc_n32_struct_flags): Rewrite.
1462 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1463 position of flag bits.
1464 (ffi_prep_cif_machdep): Rewrite n32 portion.
1465 (ffi_call): Enable for n64. Add special handling for small structure
1467 (ffi_prep_closure_loc): Add n32 and n64 support.
1468 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1469 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1471 2007-08-08 David Daney <ddaney@avtrex.com>
1473 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1474 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1476 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1477 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1478 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1479 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1480 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1481 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1482 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1483 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1484 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1485 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1486 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1487 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1488 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1489 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1490 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1491 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1492 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1493 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1494 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1495 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1496 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1497 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1498 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1499 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1500 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1501 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1502 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1503 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1504 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1505 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1506 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1508 2007-08-07 Andrew Haley <aph@redhat.com>
1510 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1513 2007-08-06 Andrew Haley <aph@redhat.com>
1516 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1517 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1520 2007-08-02 David Daney <ddaney@avtrex.com>
1522 * testsuite/libffi.call/return_ul.c (main): Define return type as
1523 ffi_arg. Use proper printf conversion specifier.
1525 2007-07-30 Andrew Haley <aph@redhat.com>
1528 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1529 signed/unsigned int8/16.
1530 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1532 Remove code to pop args from the stack after call.
1533 Special-case signed/unsigned int8/16.
1534 * testsuite/libffi.call/return_sc.c (main): Revert.
1536 2007-07-26 Richard Guenther <rguenther@suse.de>
1539 * testsuite/libffi.call/return_sc.c (main): Verify call
1540 result as signed char, not ffi_arg.
1542 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1544 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1545 * configure: Regenerate.
1547 2007-07-11 David Daney <ddaney@avtrex.com>
1549 * src/mips/ffi.c: Don't include sys/cachectl.h.
1550 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1553 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1555 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1556 from (ffi_prep_closure): ... this.
1557 (FFI_INIT_TRAMPOLINE): Adjust.
1559 2005-12-31 Phil Blundell <pb@reciva.com>
1561 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1562 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1563 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1564 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1565 (FFI_CLOSURES): Enable closure support.
1567 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1569 * testsuite/libffi.call/cls_multi_ushort.c,
1570 testsuite/libffi.call/cls_align_uint16.c,
1571 testsuite/libffi.call/nested_struct1.c,
1572 testsuite/libffi.call/nested_struct3.c,
1573 testsuite/libffi.call/cls_7_1_byte.c,
1574 testsuite/libffi.call/cls_double.c,
1575 testsuite/libffi.call/nested_struct5.c,
1576 testsuite/libffi.call/nested_struct7.c,
1577 testsuite/libffi.call/cls_sint.c,
1578 testsuite/libffi.call/nested_struct9.c,
1579 testsuite/libffi.call/cls_20byte1.c,
1580 testsuite/libffi.call/cls_multi_sshortchar.c,
1581 testsuite/libffi.call/cls_align_sint64.c,
1582 testsuite/libffi.call/cls_3byte2.c,
1583 testsuite/libffi.call/cls_multi_schar.c,
1584 testsuite/libffi.call/cls_multi_uchar.c,
1585 testsuite/libffi.call/cls_19byte.c,
1586 testsuite/libffi.call/cls_9byte1.c,
1587 testsuite/libffi.call/cls_align_float.c,
1588 testsuite/libffi.call/closure_fn1.c,
1589 testsuite/libffi.call/problem1.c,
1590 testsuite/libffi.call/closure_fn3.c,
1591 testsuite/libffi.call/cls_sshort.c,
1592 testsuite/libffi.call/closure_fn5.c,
1593 testsuite/libffi.call/cls_align_double.c,
1594 testsuite/libffi.call/cls_2byte.c,
1595 testsuite/libffi.call/nested_struct.c,
1596 testsuite/libffi.call/nested_struct10.c,
1597 testsuite/libffi.call/cls_4byte.c,
1598 testsuite/libffi.call/cls_6byte.c,
1599 testsuite/libffi.call/cls_8byte.c,
1600 testsuite/libffi.call/cls_multi_sshort.c,
1601 testsuite/libffi.call/cls_align_uint32.c,
1602 testsuite/libffi.call/cls_align_sint16.c,
1603 testsuite/libffi.call/cls_float.c,
1604 testsuite/libffi.call/cls_20byte.c,
1605 testsuite/libffi.call/cls_5_1_byte.c,
1606 testsuite/libffi.call/nested_struct2.c,
1607 testsuite/libffi.call/cls_24byte.c,
1608 testsuite/libffi.call/nested_struct4.c,
1609 testsuite/libffi.call/nested_struct6.c,
1610 testsuite/libffi.call/cls_64byte.c,
1611 testsuite/libffi.call/nested_struct8.c,
1612 testsuite/libffi.call/cls_uint.c,
1613 testsuite/libffi.call/cls_multi_ushortchar.c,
1614 testsuite/libffi.call/cls_schar.c,
1615 testsuite/libffi.call/cls_uchar.c,
1616 testsuite/libffi.call/cls_align_uint64.c,
1617 testsuite/libffi.call/cls_ulonglong.c,
1618 testsuite/libffi.call/cls_align_longdouble.c,
1619 testsuite/libffi.call/cls_1_1byte.c,
1620 testsuite/libffi.call/cls_12byte.c,
1621 testsuite/libffi.call/cls_3_1byte.c,
1622 testsuite/libffi.call/cls_3byte1.c,
1623 testsuite/libffi.call/cls_4_1byte.c,
1624 testsuite/libffi.call/cls_6_1_byte.c,
1625 testsuite/libffi.call/cls_16byte.c,
1626 testsuite/libffi.call/cls_18byte.c,
1627 testsuite/libffi.call/closure_fn0.c,
1628 testsuite/libffi.call/cls_9byte2.c,
1629 testsuite/libffi.call/closure_fn2.c,
1630 testsuite/libffi.call/closure_fn4.c,
1631 testsuite/libffi.call/cls_ushort.c,
1632 testsuite/libffi.call/closure_fn6.c,
1633 testsuite/libffi.call/cls_5byte.c,
1634 testsuite/libffi.call/cls_align_pointer.c,
1635 testsuite/libffi.call/cls_7byte.c,
1636 testsuite/libffi.call/cls_align_sint32.c,
1637 testsuite/libffi.special/unwindtest_ffi_call.cc,
1638 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1640 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1642 * aclocal.m4: Regenerated.
1644 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1646 * configure: Regenerate.
1648 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1650 * Makefile.in: Regenerate.
1651 * configure: Regenerate.
1652 * aclocal.m4: Regenerate.
1653 * include/Makefile.in: Regenerate.
1654 * testsuite/Makefile.in: Regenerate.
1656 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1658 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1659 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1660 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1661 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1662 (FFI_CLOSURES): Enable closure support.
1664 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1666 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1667 * configure: Regenerate.
1668 * fficonfig.h.in: Regenerate.
1669 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1670 CFI_OFFSET,CFI_DEF_CFA): New macros.
1671 (ffi_call_SYSV): Add callframe annotation.
1673 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1675 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1676 numerous test suite failures.
1677 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1679 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1681 * Makefile.am (EXTRA_DIST): Bring up to date.
1682 * Makefile.in: Regenerate.
1683 * src/frv/eabi.S: Remove RCS keyword.
1685 2007-04-06 Richard Henderson <rth@redhat.com>
1687 * configure.ac: Tidy target case.
1688 (HAVE_LONG_DOUBLE): Allow the target to override.
1689 * configure: Regenerate.
1690 * include/ffi.h.in: Don't define ffi_type_foo if
1691 LIBFFI_HIDE_BASIC_TYPES is defined.
1692 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1694 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1695 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1696 (ffi_type_longdouble): Special case for Alpha. Don't define
1697 if long double == double.
1699 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1700 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1701 (ffi_call, ffi_closure_osf_inner): Likewise.
1702 (ffi_closure_osf_inner): Likewise. Mark hidden.
1703 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1704 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1705 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1706 (load_table): Handle 128-bit long double.
1708 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1710 2007-04-06 Tom Tromey <tromey@redhat.com>
1713 * README: Fixed bug in example.
1715 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1717 * src/closures.c: Include sys/statfs.h.
1718 (_GNU_SOURCE): Define on Linux.
1719 (FFI_MMAP_EXEC_SELINUX): Define.
1720 (selinux_enabled): New variable.
1721 (selinux_enabled_check): New function.
1722 (is_selinux_enabled): Define.
1725 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1727 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1728 Use 'volatile float sum' to create sum of floats to avoid false
1729 negative due to excess precision on ix86 targets.
1732 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1734 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1736 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1738 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1740 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1741 (ffi_prep_closure_loc): New.
1742 (ffi_prep_raw_closure_loc): New.
1743 (ffi_prep_java_raw_closure_loc): New.
1744 * src/closures.c: New file.
1745 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1746 Replace sflags with exec_offset.
1747 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1748 sub_segment_exec_offset): New macros.
1749 (get_segment_flags, set_segment_flags, check_segment_merge): New
1751 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1752 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1753 destroy_mspace): Use new macros.
1754 (sys_alloc): Silence warning.
1755 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1756 * Makefile.in: Rebuilt.
1757 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1758 terms of ffi_prep_closure_loc.
1759 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1761 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1763 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1765 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1766 the renamed version.
1767 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1768 (ffi_prep_closure): ... this.
1769 * src/pa/ffi.c: Likewise.
1770 * src/cris/ffi.c: Likewise. Adjust.
1771 * src/frv/ffi.c: Likewise.
1772 * src/ia64/ffi.c: Likewise.
1773 * src/mips/ffi.c: Likewise.
1774 * src/powerpc/ffi_darwin.c: Likewise.
1775 * src/s390/ffi.c: Likewise.
1776 * src/sh/ffi.c: Likewise.
1777 * src/sh64/ffi.c: Likewise.
1778 * src/sparc/ffi.c: Likewise.
1779 * src/x86/ffi64.c: Likewise.
1780 * src/x86/ffi.c: Likewise.
1781 (FFI_INIT_TRAMPOLINE): Adjust.
1782 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1783 (ffi_prep_raw_closure): ... this.
1784 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1785 (ffi_prep_closure): ... this.
1786 (flush_icache): Adjust.
1788 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1790 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1793 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1795 * Makefile.am: Add dummy install-pdf target.
1796 * Makefile.in: Regenerate
1798 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1800 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1801 ffi_closure_helper_SYSV): Add long double handling.
1803 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1805 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1806 immediately after bctrl instruction.
1808 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1810 * Makefile.am (all-recursive, install-recursive,
1811 mostlyclean-recursive, clean-recursive, distclean-recursive,
1812 maintainer-clean-recursive): Add missing targets.
1813 * Makefile.in: Rebuilt.
1815 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1817 * configure.ac: Add TARGET for x86_64-*-darwin*.
1818 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1820 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1821 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1822 * configure: Regenerate.
1823 * Makefile.in: Regenerate.
1824 * include/Makefile.in: Regenerate.
1825 * testsuite/Makefile.in: Regenerate.
1826 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1829 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1831 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1834 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1836 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1837 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1839 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1840 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1842 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1843 Sandro Tolaini <tolaini@libero.it>
1845 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1847 * configure: Regenerated.
1848 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1849 (EXTRA_DIST): Add src/x86/darwin.S.
1850 * Makefile.in: Regenerated.
1851 * include/Makefile.in: Regenerated.
1852 * testsuite/Makefile.in: Regenerated.
1854 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1855 X86_WIN32, and additionally align stack to 16 bytes.
1856 * src/x86/darwin.S: New, based on sysv.S.
1857 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1859 2006-09-12 David Daney <ddaney@avtrex.com>
1862 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1863 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1864 * aclocal.m4: Regenerated for automake 1.9.6.
1865 * Makefile.in: Regenerated.
1866 * include/Makefile.in: Regenerated.
1867 * testsuite/Makefile.in: Regenerated.
1869 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1871 * include/ffi_common.h (struct): Revert accidental commit.
1873 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1875 * include/ffi_common.h: Remove lint directives.
1876 * include/ffi.h.in: Likewise.
1878 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1880 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1881 for 32-bit architectures.
1882 * testsuite/libffi.call/return_ul.c: New test case.
1884 2006-07-19 David Daney <ddaney@avtrex.com>
1886 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1887 xfail remains for mips64.
1889 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1891 * Makefile.am: Add install-html target. Add install-html to .PHONY
1892 * Makefile.in: Regenerate.
1893 * aclocal.m4: Regenerate.
1894 * include/Makefile.in: Regenerate.
1895 * testsuite/Makefile.in: Regenerate.
1897 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1899 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1902 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1904 * README: Remove notice about 'Crazy Comments'.
1905 * src/debug.c: Remove lint directives. Cleanup white spaces.
1906 * src/java_raw_api.c: Likewise.
1907 * src/prep_cif.c: Likewise.
1908 * src/raw_api.c: Likewise.
1909 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1911 * src/arm/ffi.c: Remove lint directives.
1912 * src/m32r/ffi.c: Likewise.
1913 * src/pa/ffi.c: Likewise.
1914 * src/powerpc/ffi.c: Likewise.
1915 * src/powerpc/ffi_darwin.c: Likewise.
1916 * src/sh/ffi.c: Likewise.
1917 * src/sh64/ffi.c: Likewise.
1918 * src/x86/ffi.c: Likewise.
1919 * testsuite/libffi.call/float2.c: Likewise.
1920 * testsuite/libffi.call/promotion.c: Likewise.
1921 * testsuite/libffi.call/struct1.c: Likewise.
1923 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1925 * src/pa/hpux32.S: Correct unwind offset calculation for
1927 * src/pa/linux.S: Likewise.
1929 2006-04-12 James E Wilson <wilson@specifix.com>
1932 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1933 (hfa_type_load): Call stf_spill.
1934 (hfa_type_store): Call ldf_fill.
1935 (ffi_call): Adjust calls to above routines. Add local temps for
1938 2006-04-10 Matthias Klose <doko@debian.org>
1940 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1941 directory names containing underscores.
1943 2006-04-07 James E Wilson <wilson@specifix.com>
1945 * testsuite/libffi.call/float4.c: New testcase.
1947 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1948 Andreas Tobler <a.tobler@schweiz.ch>
1950 * Makefile.am: Add PA_HPUX port.
1951 * Makefile.in: Regenerate.
1952 * include/Makefile.in: Likewise.
1953 * testsuite/Makefile.in: Likewise.
1954 * configure.ac: Add PA_HPUX rules.
1955 * configure: Regenerate.
1956 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1957 Add PA_HPUX and PA64_HPUX.
1958 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1959 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1960 (FFI_TYPE_SMALL_STRUCT2): Define.
1961 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1962 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1963 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1964 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1965 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1966 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1967 * src/pa/ffi.c (ROUND_DOWN): Delete.
1968 (fldw, fstw, fldd, fstd): Use '__asm__'.
1969 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1970 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1971 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1972 Simplify incrementing of stack slot variable. Change type of local
1973 'n' to unsigned int.
1974 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1976 (ffi_prep_cif_machdep): Likewise.
1977 (ffi_call): Likewise.
1978 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1979 return type to ffi_status. Simplify incrementing of stack slot
1980 variable. Only copy floating point argument registers when PA_LINUX
1981 is true. Reformat debug statement.
1982 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1983 FFI_TYPE_SMALL_STRUCT8.
1984 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1986 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1987 Add nops to cache flush. Add trampoline for PA_HPUX.
1988 * src/pa/hpux32.S: New file.
1989 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1990 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1991 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1992 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1993 argument type checks so that common argument types appear first.
1994 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1995 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1997 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1999 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
2000 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
2001 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
2002 FFI_TYPE_LONGDOUBLE.
2003 (ffi_prep_args64): Assert using IBM extended double.
2004 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
2005 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
2006 (ffi_call): Handle FFI_LINUX.
2007 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
2008 gpr3 return pointer as for struct return. Handle FFI_LINUX
2009 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
2011 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
2012 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
2013 Don't use r6 as pointer to results, instead use sp offset. Don't
2014 make a special call to load lr with case table address, instead
2015 use offset from previous call.
2016 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
2017 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
2020 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
2022 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
2023 passed with FP registers correctly.
2024 (ffi_closure_helper_SYSV): Likewise.
2025 * src/sh64/sysv.S: Likewise.
2027 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
2029 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
2030 args and userdata unused.
2031 (closure_test_fn1): Mark cif and userdata unused.
2032 (main): Remove unused res.
2034 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
2036 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
2037 -O2, -O3, -Os and the warning flags -W -Wall.
2038 * testsuite/libffi.special/special.exp: Likewise.
2039 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
2040 unused parameter unused for gcc or else do nothing.
2041 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2042 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
2043 and userdata unused.
2044 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
2045 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
2046 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
2047 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
2048 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
2049 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
2050 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
2051 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
2052 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
2053 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
2054 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
2055 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
2056 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
2057 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
2058 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
2059 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
2060 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
2061 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
2062 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
2063 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
2064 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
2065 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
2066 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
2067 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
2069 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
2071 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
2073 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
2074 void* to avoid compiler warning.
2076 (cls_struct_align_gn): Mark cif and userdata unused.
2077 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
2079 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
2081 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
2083 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
2085 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
2087 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
2088 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
2089 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
2091 (main): Cast res_call to silence gcc.
2092 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
2094 (main): Cast res_call to silence gcc.
2095 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
2097 (main): Cast res_call to silence gcc.
2098 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
2100 (main): Cast res_call to silence gcc.
2101 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
2103 (main): Cast res_call to silence gcc.
2104 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
2106 (main): Cast res_call to silence gcc.
2107 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
2109 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
2110 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
2112 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
2113 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
2115 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
2116 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
2118 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
2119 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
2121 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
2122 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
2123 and userdata unused.
2124 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
2126 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
2127 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
2128 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
2129 Cleanup white spaces.
2130 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
2131 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
2132 cif and userdata unused.
2133 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
2135 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
2136 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
2137 formatters to silence gcc.
2138 (B_gn): Mark cif and userdata unused.
2139 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
2141 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
2142 (B_gn): Mark cif and userdata unused.
2143 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
2145 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
2146 (B_gn): Mark cif and userdata unused.
2147 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
2149 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
2150 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
2151 * testsuite/libffi.call/problem1.c (stub): Likewise.
2152 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
2154 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
2155 in the last commit for this test case in the test case itself.
2156 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
2158 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
2159 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
2160 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
2161 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
2162 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
2163 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
2165 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
2167 * src/sh/sysv.S: Fix register numbers in the FDE for
2170 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
2172 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
2173 declaration to avoid a false negative on ix86. See PR323.
2175 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
2177 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
2178 and cast integer to void * if needed. Update the pointer to
2179 the FP register saved area correctly.
2181 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
2183 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
2185 * testsuite/libffi.call/nested_struct4.c: Likewise.
2187 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
2189 * testsuite/libffi.call/return_dbl.c: New test case.
2190 * testsuite/libffi.call/return_dbl1.c: Likewise.
2191 * testsuite/libffi.call/return_dbl2.c: Likewise.
2192 * testsuite/libffi.call/return_fl.c: Likewise.
2193 * testsuite/libffi.call/return_fl1.c: Likewise.
2194 * testsuite/libffi.call/return_fl2.c: Likewise.
2195 * testsuite/libffi.call/return_fl3.c: Likewise.
2196 * testsuite/libffi.call/closure_fn6.c: Likewise.
2198 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
2200 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
2201 here to be used by other test cases too.
2203 * testsuite/libffi.call/nested_struct10.c: New test case.
2204 * testsuite/libffi.call/nested_struct9.c: Likewise.
2205 * testsuite/libffi.call/nested_struct8.c: Likewise.
2206 * testsuite/libffi.call/nested_struct7.c: Likewise.
2207 * testsuite/libffi.call/nested_struct6.c: Likewise.
2208 * testsuite/libffi.call/nested_struct5.c: Likewise.
2209 * testsuite/libffi.call/nested_struct4.c: Likewise.
2211 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2213 * configure.ac: Enable libffi for sparc64-*-freebsd*.
2214 * configure: Rebuilt.
2216 2006-01-18 Jakub Jelinek <jakub@redhat.com>
2218 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
2219 instead do the shifting inline.
2220 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
2221 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
2222 and 8 byte structs, for the remaining struct sizes don't call
2223 __lshrdi3, instead do the shifting inline.
2225 2005-12-07 Thiemo Seufer <ths@networkno.de>
2227 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
2228 missing parentheses.
2229 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
2230 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
2231 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
2232 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
2233 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
2235 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
2237 (ffi_prep_closure): Improve trampoline instruction scheduling.
2238 (ffi_closure_mips_inner_O32): Fix endianness bugs.
2240 2005-12-03 Alan Modra <amodra@bigpond.net.au>
2242 * src/powerpc/ffi.c: Formatting.
2243 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
2244 (ffi_prep_args64): Likewise.
2246 2005-09-30 Geoffrey Keating <geoffk@apple.com>
2248 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
2249 darwin, use -shared-libgcc not -lgcc_s, and explain why.
2251 2005-09-26 Tom Tromey <tromey@redhat.com>
2253 * testsuite/libffi.call/float1.c (value_type): New typedef.
2254 (CANARY): New define.
2255 (main): Check for result buffer overflow.
2256 * src/powerpc/linux64.S: Handle linux64 long double returns.
2257 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
2258 (ffi_prep_cif_machdep): Handle linux64 long double returns.
2260 2005-08-25 Alan Modra <amodra@bigpond.net.au>
2263 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
2265 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
2267 2005-08-11 Jakub Jelinek <jakub@redhat.com>
2269 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
2270 (AH_BOTTOM): Add FFI_HIDDEN definition.
2271 * configure: Rebuilt.
2272 * fficonfig.h.in: Rebuilt.
2273 * src/powerpc/ffi.c (hidden): Remove.
2274 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
2275 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
2276 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
2277 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
2278 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
2279 add FFI_HIDDEN to its prototype.
2280 (ffi_closure_SYSV_inner): New.
2281 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2282 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2284 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
2287 * configure: Rebuilt.
2288 * configure.ac: Handle i*86-*-gnu*.
2290 2005-08-09 Jakub Jelinek <jakub@redhat.com>
2292 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
2293 DW_CFA_offset_extended_sf rather than
2294 DW_CFA_GNU_negative_offset_extended.
2295 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
2297 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
2299 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
2301 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
2302 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
2303 partially on register.
2304 (ffi_closure_helper_SYSV): Likewise.
2305 (ffi_prep_cif_machdep): Don't set too many cif->flags.
2307 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
2309 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
2311 * src/sh64/ffi.c (simple_type): Remove.
2312 (return_type): Handle small structures correctly.
2313 (ffi_prep_args): Likewise.
2314 (ffi_call): Likewise.
2315 (ffi_closure_helper_SYSV): Likewise.
2316 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
2317 Emit position independent code if PIC and remove wrong datalabel
2318 prefixes from EH data.
2320 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
2322 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
2323 * Makefile.in: Regenerate.
2324 * include/Makefile.in: Likewise.
2325 * testsuite/Makefile.in: Likewise.
2326 * configure.ac: Add POWERPC_FREEBSD rules.
2327 * configure: Regenerate.
2328 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
2329 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
2330 * src/powerpc/ffi.c: Add flags to handle small structure returns
2332 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
2334 (ffi_closure_helper_SYSV): Likewise.
2335 * src/powerpc/ppc_closure.S: Add return types for small structures.
2336 * src/powerpc/sysv.S: Add bits to handle small structures for
2339 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
2341 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
2342 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
2343 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
2345 2005-07-05 Randolph Chung <tausq@debian.org>
2347 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
2348 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
2349 structures. Kill compilation warnings.
2350 (ffi_closure_inner_LINUX): Print return values as hex in debug
2351 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
2352 Properly handle 5-7 byte structure returns.
2353 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
2354 (FFI_TYPE_SMALL_STRUCT2): Remove.
2355 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
2356 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
2357 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
2358 (checksmst1, checksmst2): Remove.
2359 (checksmst3): Optimize handling of 3-byte struct returns.
2360 (checksmst567): Properly handle 5-7 byte struct returns.
2362 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2365 * src/mips/n32.S: Enforce PIC code.
2366 * src/mips/o32.S: Likewise.
2368 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2370 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
2371 * configure: Regenerate.
2373 2005-06-01 Alan Modra <amodra@bigpond.net.au>
2375 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
2376 to call ffi_closure_helper_SYSV. Append @local instead.
2377 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
2379 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
2381 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
2382 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
2383 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
2384 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
2385 include/Makefile.in, testsuite/Makefile.in: Regenerate.
2387 2005-05-09 Mike Stump <mrs@apple.com>
2389 * configure: Regenerate.
2391 2005-05-08 Richard Henderson <rth@redhat.com>
2394 * src/alpha/osf.S: Update unwind into to match code.
2396 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
2397 Richard Henderson <rth@redhat.com>
2399 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
2401 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
2402 (ffi_prep_closure): Set carry bit if sse-used flag set.
2403 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
2404 Only load sse registers if ssecount non-zero.
2405 (ffi_closure_unix64): Only save sse registers if carry set on entry.
2407 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
2409 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
2410 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
2411 * configure: Regenerate.
2413 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
2415 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
2416 have Tcl8.3-compatible intermediate variable.
2418 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
2419 Hans-Peter Nilsson <hp@axis.com>
2421 * Makefile.am: Add CRIS support.
2422 * configure.ac: Likewise.
2423 * Makefile.in, configure, testsuite/Makefile.in,
2424 include/Makefile.in: Regenerate.
2425 * src/cris: New directory.
2426 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
2427 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
2429 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
2430 \r?\n in output tests.
2432 2005-04-12 Mike Stump <mrs@apple.com>
2434 * configure: Regenerate.
2436 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
2438 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
2440 2005-03-30 Steve Ellcey <sje@cup.hp.com>
2442 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
2444 * src/ia64/unix.S (ffi_closure_unix): Extend gp
2445 to 64 bits in ILP32 mode.
2446 Load 64 bits even for short data.
2448 2005-03-23 Mike Stump <mrs@apple.com>
2450 * src/powerpc/darwin.S: Update for -m64 multilib.
2451 * src/powerpc/darwin_closure.S: Likewise.
2453 2005-03-21 Zack Weinberg <zack@codesourcery.com>
2455 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
2456 Do not set tool_include_dir.
2457 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
2459 * include/Makefile.am: Set gcc_version and toollibffidir.
2460 * include/Makefile.in: Regenerate.
2462 2005-02-22 Andrew Haley <aph@redhat.com>
2464 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2465 odd-numbered register pairs for 64-bit integer types.
2467 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2470 * testsuite/libffi.call/return_ll1.c: New test case.
2472 2005-02-11 Janis Johnson <janis187@us.ibm.com>
2474 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2475 * testsuite/libffi.call/float.c: Ditto.
2476 * testsuite/libffi.call/float2.c: Ditto.
2477 * testsuite/libffi.call/float3.c: Ditto.
2479 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2481 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2483 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2485 * testsuite/libffi.special/special.exp (cxx_options): Add
2488 2004-12-31 Richard Henderson <rth@redhat.com>
2490 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2491 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2492 offset parameters with a type parameter; deduce size and structure
2493 alignment. Update all users.
2495 2004-12-31 Richard Henderson <rth@redhat.com>
2497 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2498 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
2499 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
2500 into ffi_prep_closure.
2501 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2504 2004-12-27 Richard Henderson <rth@redhat.com>
2506 * src/x86/unix64.S: Fix typo in unwind info.
2508 2004-12-25 Richard Henderson <rth@redhat.com>
2510 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2511 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2512 (merge_classes): Check for it.
2514 (classify_argument): Pass byte_offset by value; perform all updates
2516 (examine_argument): Add classes argument; handle
2517 X86_64_COMPLEX_X87_CLASS.
2518 (ffi_prep_args): Merge into ...
2519 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2520 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2521 (ffi_fill_return_value): Remove.
2522 (ffi_prep_closure): Remove dead assert.
2523 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2524 Rewrite to use struct register_args instead of va_list. Create
2525 flags for handling structure returns.
2526 * src/x86/unix64.S: Remove dead strings.
2527 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2528 stack frame with ffi_call. Handle structure returns properly.
2529 (float2sse, floatfloat2sse, double2sse): Remove.
2530 (sse2float, sse2double, sse2floatfloat): Remove.
2531 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2532 to handle structure returns properly.
2534 2004-12-08 David Edelsohn <edelsohn@gnu.org>
2536 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2538 * Makefile.in: Regenerated.
2540 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2542 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2543 * configure, aclocal.m4, Makefile.in: Regenerate.
2544 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2546 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2548 * configure: Regenerate for libtool change.
2550 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2552 * configure: Regenerate for libtool reversion.
2554 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2556 * configure: Regenerate for libtool change.
2558 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2560 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2562 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2564 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2565 of jal. Use an absolute encoding for the frame information.
2567 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2569 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2570 * acinclude.m4: Delete logic for sincludes.
2571 * aclocal.m4, Makefile.in, configure: Regenerate.
2572 * include/Makefile: Likewise.
2573 * testsuite/Makefile: Likewise.
2575 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2577 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2578 on a 8-byte boundary.
2579 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2581 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2583 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2584 long long values. Round stack allocation to a multiple of 8 bytes
2585 for ATPCS compatibility.
2586 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2587 names. Handle returning long long types. Add Thumb and interworking
2588 support. Improve soft-float code.
2590 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2592 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2593 (libffi_exit): New function.
2594 (libffi_init): Build the testglue wrapper if needed.
2596 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2599 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2601 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2603 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2605 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2607 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2608 * testsuite/libffi.call/float3.c: New test case.
2610 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2612 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2613 the function returning a structure pointed with R2.
2614 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2615 the structure return value if T bit set. Emit position
2616 independent code and EH data if PIC.
2618 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2620 * Makefile.am: Add m32r support.
2621 * configure.ac: Likewise.
2622 * Makefile.in: Regenerate.
2623 * confiugre: Regenerate.
2624 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2625 (uint64, sint64, double, longdouble)
2626 * src/m32r: New directory.
2627 * src/m32r/ffi.c: New file.
2628 * src/m32r/sysv.S: Likewise.
2629 * src/m32r/ffitarget.h: Likewise.
2631 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2633 * testsuite/libffi.call/negint.c: New test case.
2635 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2638 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2639 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2640 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2643 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2645 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2646 * testsuite/libffi.call/promotion.c: Likewise.
2647 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2649 * testsuite/libffi.call/return_sc.c: Likewise.
2650 * testsuite/libffi.call/return_uc.c: Likewise.
2652 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2654 * src/powerpc/darwin.S: Fix comments and identation.
2655 * src/powerpc/darwin_closure.S: Likewise.
2657 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2659 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2660 (ffi_prep_args): Handle longdouble arguments.
2661 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2663 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2664 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2666 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2667 * src/types.c: Defined longdouble size and alignment for darwin.
2669 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2671 * src/powerpc/aix.S: Remove whitespaces.
2672 * src/powerpc/aix_closure.S: Likewise.
2673 * src/powerpc/asm.h: Likewise.
2674 * src/powerpc/ffi.c: Likewise.
2675 * src/powerpc/ffitarget.h: Likewise.
2676 * src/powerpc/linux64.S: Likewise.
2677 * src/powerpc/linux64_closure.S: Likewise.
2678 * src/powerpc/ppc_closure.S: Likewise.
2679 * src/powerpc/sysv.S: Likewise.
2681 2004-08-30 Anthony Green <green@redhat.com>
2683 * Makefile.am: Add frv support.
2684 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2685 * configure.ac: Read configure.host.
2686 * configure.in: Read configure.host.
2687 * configure.host: New file. frv-elf needs libgloss.
2688 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2689 alignment. This is needed to frv and shouldn't harm the others.
2690 * include/ffi_common.h (ALIGN_DOWN): New macro.
2691 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2693 2004-08-24 David Daney <daney@avtrex.com>
2695 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2696 * testsuite/libffi.call/closure_fn1.c: Likewise.
2697 * testsuite/libffi.call/closure_fn2.c Likewise.
2698 * testsuite/libffi.call/closure_fn3.c: Likewise.
2699 * testsuite/libffi.call/closure_fn4.c: Likewise.
2700 * testsuite/libffi.call/closure_fn5.c: Likewise.
2701 * testsuite/libffi.call/cls_18byte.c: Likewise.
2702 * testsuite/libffi.call/cls_19byte.c: Likewise.
2703 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2704 * testsuite/libffi.call/cls_20byte.c: Likewise.
2705 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2706 * testsuite/libffi.call/cls_24byte.c: Likewise.
2707 * testsuite/libffi.call/cls_2byte.c: Likewise.
2708 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2709 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2710 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2711 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2712 * testsuite/libffi.call/cls_4byte.c: Likewise.
2713 * testsuite/libffi.call/cls_64byte.c: Likewise.
2714 * testsuite/libffi.call/cls_6byte.c: Likewise.
2715 * testsuite/libffi.call/cls_7byte.c: Likewise.
2716 * testsuite/libffi.call/cls_8byte.c: Likewise.
2717 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2718 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2719 * testsuite/libffi.call/cls_align_double.c: Likewise.
2720 * testsuite/libffi.call/cls_align_float.c: Likewise.
2721 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2722 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2723 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2724 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2725 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2726 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2727 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2728 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2729 * testsuite/libffi.call/cls_double.c: Likewise.
2730 * testsuite/libffi.call/cls_float.c: Likewise.
2731 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2732 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2733 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2734 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2735 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2736 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2737 * testsuite/libffi.call/cls_schar.c: Likewise.
2738 * testsuite/libffi.call/cls_sint.c: Likewise.
2739 * testsuite/libffi.call/cls_sshort.c: Likewise.
2740 * testsuite/libffi.call/cls_uchar.c: Likewise.
2741 * testsuite/libffi.call/cls_uint.c: Likewise.
2742 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2743 * testsuite/libffi.call/cls_ushort.c: Likewise.
2744 * testsuite/libffi.call/nested_struct.c: Likewise.
2745 * testsuite/libffi.call/nested_struct1.c: Likewise.
2746 * testsuite/libffi.call/nested_struct2.c: Likewise.
2747 * testsuite/libffi.call/nested_struct3.c: Likewise.
2748 * testsuite/libffi.call/problem1.c: Likewise.
2749 * testsuite/libffi.special/unwindtest.cc: Likewise.
2750 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2752 * testsuite/libffi.call/cls_16byte.c: Likewise.
2753 * testsuite/libffi.call/cls_5byte.c: Likewise.
2755 2004-08-23 David Daney <daney@avtrex.com>
2758 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2759 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2760 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2761 parameters and return types.
2762 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2763 (ffi_prep_closure): Ditto.
2764 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2765 alignment calculations.
2766 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2767 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2769 2004-08-14 Casey Marshall <csm@gnu.org>
2771 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2772 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2773 integer (O32 ABI only).
2774 (ffi_prep_closure): new function.
2775 (ffi_closure_mips_inner_O32): new function.
2776 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2777 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2778 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2779 64 bit integers correctly.
2780 (ffi_closure_O32): new function.
2781 Added DWARF-2 unwind info for both functions.
2783 2004-08-10 Andrew Haley <aph@redhat.com>
2785 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2787 2004-08-01 Robert Millan <robertmh@gnu.org>
2789 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2790 * configure: Regenerate.
2792 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2794 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2795 and mmap() explicitly instead of relying on preset autoconf cache
2797 * aclocal.m4: Regenerate.
2798 * configure: Regenerate.
2800 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2802 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2803 (ffi_check_float_struct): Remove unused prototype.
2805 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2807 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2808 character on Darwin, use '\n\t' instead.
2810 2004-06-26 Matthias Klose <doko@debian.org>
2812 * libtool-version: Fix typo in revision/age.
2814 2004-06-17 Matthias Klose <doko@debian.org>
2816 * libtool-version: New.
2817 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2818 * Makefile.in: Regenerate.
2820 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2822 * Makefile.am: Remove useless multilib rules.
2823 * Makefile.in: Regenerate.
2824 * aclocal.m4: Regenerate with automake 1.8.5.
2825 * configure.ac: Remove useless multilib configury.
2826 * configure: Regenerate.
2828 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2830 * .cvsignore: New file.
2832 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2834 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2836 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2839 2004-06-06 Sean McNeil <sean@mcneil.com>
2841 * configure.ac: Add x86_64-*-freebsd* support.
2842 * configure: Regenerate.
2844 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2847 * configure.ac: Test for existence of mmap and sys/mman.h before
2848 checking blacklist. Fix suggested by Jim Wilson.
2849 * configure: Regenerate.
2851 2004-04-26 Matt Austern <austern@apple.com>
2853 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2855 * src/powerpc/darwin_closure.S: Likewise.
2857 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2859 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2860 error. Reported by Thomas Heller <theller@python.net>.
2861 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2862 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2864 2004-03-20 Matthias Klose <doko@debian.org>
2866 * src/pa/linux.S: Fix typo.
2868 2004-03-19 Matthias Klose <doko@debian.org>
2870 * Makefile.am: Update.
2871 * Makefile.in: Regenerate.
2872 * src/pa/ffi.h.in: Remove.
2873 * src/pa/ffitarget.h: New file.
2875 2004-02-10 Randolph Chung <tausq@debian.org>
2877 * Makefile.am: Add PA support.
2878 * Makefile.in: Regenerate.
2879 * include/Makefile.in: Regenerate.
2880 * configure.ac: Add PA target.
2881 * configure: Regenerate.
2882 * src/pa/ffi.c: New file.
2883 * src/pa/ffi.h.in: Add PA support.
2884 * src/pa/linux.S: New file.
2885 * prep_cif.c: Add PA support.
2887 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2889 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2891 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2892 with ecif->cif->flags.
2893 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2895 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2896 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2897 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2898 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2900 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2902 * configure.in: Rename file to ...
2903 * configure.ac: ... this.
2904 * fficonfig.h.in: Regenerate.
2905 * Makefile.in: Regenerate.
2906 * include/Makefile.in: Regenerate.
2907 * testsuite/Makefile.in: Regenerate.
2909 2004-03-12 Matt Austern <austern@apple.com>
2911 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2912 changes in EH format resulting from addition of linkonce support.
2913 * src/powerpc/darwin_closure.S: Likewise.
2915 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2916 Paolo Bonzini <bonzini@gnu.org>
2918 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2919 Remove VPATH. Remove rules for object files. Remove multilib support.
2920 (AM_CCASFLAGS): Add.
2921 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2922 (AC_PREREQ): Bump version to 2.59.
2923 (AC_INIT): Fill with version info and bug address.
2924 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2925 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2926 De-precious CC so that the right flags are passed down to multilibs.
2927 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2928 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2929 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2930 * configure: Rebuilt.
2931 * aclocal.m4: Likewise.
2932 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2933 * fficonfig.h.in: Likewise.
2935 2004-03-11 Andreas Schwab <schwab@suse.de>
2937 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2938 arguments from fp registers only for the first 8 parameter slots.
2939 Don't convert a float parameter when passed in memory.
2941 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2943 * configure: Regenerate for config/accross.m4 correction.
2945 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2947 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2948 ecif->cif->bytes to bytes.
2949 (ffi_prep_cif_machdep): Add braces around nested if statement.
2951 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2953 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2955 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2956 (ffi_closure_helper_LINUX64): Fix typo.
2957 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2959 * testsuite/libffi.call/float.c: Likewise.
2960 * testsuite/libffi.call/float2.c: Likewise.
2962 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2964 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2965 long double function return and long double arg handling.
2966 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2967 Use "end_pfr" instead of "nf". Correct long double handling.
2969 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2971 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2972 space for long double return value. Adjust stack frame and offsets.
2973 Load f2 long double return.
2975 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2977 * src/types.c: Use 16 byte long double for POWERPC64.
2979 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2981 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2982 when the structure return address is passed in %o0.
2983 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2984 (ffi_v9_layout_struct): Align the field following a nested structure
2985 on a word boundary. Use memmove instead of memcpy.
2986 (ffi_call): Update call to ffi_V9_return_struct.
2987 (ffi_prep_closure): Define 'ctx' only for V8.
2988 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2989 and ffi_closure_sparc_inner_v9.
2990 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2991 Always skip the structure return address. For structures and long
2992 doubles, copy the argument directly.
2993 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2994 if required. Shift the maximum floating-point slot accordingly. For
2995 big structures, copy the argument directly; otherwise, left-justify the
2996 argument and call ffi_v9_layout_struct to lay out the structure on
2998 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2999 (ffi_closure_v8): Pass the structure return address. Update call to
3000 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
3001 Skip the 'unimp' insn when returning long doubles and structures.
3002 * src/sparc/v9.S: Undef STACKFRAME before defining it.
3003 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
3004 FFI_TYPE_INT handling. Load structures both in integers and
3005 floating-point registers on return.
3006 * README: Update status of the SPARC port.
3008 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
3010 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
3012 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
3014 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
3016 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
3017 value as of type ffi_arg, not unsigned int.
3019 2004-01-21 Michael Ritzert <ritzert@t-online.de>
3021 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
3024 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
3026 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
3029 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3031 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
3034 2003-12-10 Richard Henderson <rth@redhat.com>
3036 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
3037 size_t instead of int.
3039 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
3041 * testsuite/libffi.call/many_win32.c: Include <float.h>.
3042 * testsuite/libffi.call/many_win32.c (main): Replace variable
3043 int i with unsigned long ul.
3045 * testsuite/libffi.call/cls_align_uint64.c: New test case.
3046 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
3047 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
3048 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
3049 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
3050 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
3051 * testsuite/libffi.call/cls_align_float.c: Likewise.
3052 * testsuite/libffi.call/cls_align_double.c: Likewise.
3053 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
3054 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
3056 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
3059 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
3060 Align arguments to 32 bits.
3062 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
3065 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
3066 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
3067 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
3068 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
3069 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
3070 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
3072 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
3074 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3076 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
3077 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3079 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
3081 * Makefile.in: Rebuilt.
3082 * configure: Likewise.
3083 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
3086 2003-11-21 Andreas Jaeger <aj@suse.de>
3087 Andreas Tobler <a.tobler@schweiz.ch>
3089 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
3090 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
3091 * Makefile.in: Rebuilt.
3092 * aclocal.m4: Likewise.
3093 * configure: Likewise.
3094 * fficonfig.h.in: Likewise.
3095 * testsuite/lib/libffi-dg.exp: Add include dir.
3096 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
3097 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3098 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
3099 for ffi_closure if available.
3100 * testsuite/libffi.call/closure_fn1.c: Likewise.
3101 * testsuite/libffi.call/closure_fn2.c: Likewise.
3102 * testsuite/libffi.call/closure_fn3.c: Likewise.
3103 * testsuite/libffi.call/closure_fn4.c: Likewise.
3104 * testsuite/libffi.call/closure_fn5.c: Likewise.
3105 * testsuite/libffi.call/cls_12byte.c: Likewise.
3106 * testsuite/libffi.call/cls_16byte.c: Likewise.
3107 * testsuite/libffi.call/cls_18byte.c: Likewise.
3108 * testsuite/libffi.call/cls_19byte.c: Likewise.
3109 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3110 * testsuite/libffi.call/cls_20byte.c: Likewise.
3111 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3112 * testsuite/libffi.call/cls_24byte.c: Likewise.
3113 * testsuite/libffi.call/cls_2byte.c: Likewise.
3114 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3115 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3116 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3117 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3118 * testsuite/libffi.call/cls_4byte.c: Likewise.
3119 * testsuite/libffi.call/cls_5byte.c: Likewise.
3120 * testsuite/libffi.call/cls_64byte.c: Likewise.
3121 * testsuite/libffi.call/cls_6byte.c: Likewise.
3122 * testsuite/libffi.call/cls_7byte.c: Likewise.
3123 * testsuite/libffi.call/cls_8byte.c: Likewise.
3124 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3125 * testsuite/libffi.call/cls_9byte2.c: Likewise.
3126 * testsuite/libffi.call/cls_double.c: Likewise.
3127 * testsuite/libffi.call/cls_float.c: Likewise.
3128 * testsuite/libffi.call/cls_schar.c: Likewise.
3129 * testsuite/libffi.call/cls_sint.c: Likewise.
3130 * testsuite/libffi.call/cls_sshort.c: Likewise.
3131 * testsuite/libffi.call/cls_uchar.c: Likewise.
3132 * testsuite/libffi.call/cls_uint.c: Likewise.
3133 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3134 * testsuite/libffi.call/cls_ushort.c: Likewise.
3135 * testsuite/libffi.call/nested_struct.c: Likewise.
3136 * testsuite/libffi.call/nested_struct1.c: Likewise.
3137 * testsuite/libffi.call/nested_struct2.c: Likewise.
3138 * testsuite/libffi.call/nested_struct3.c: Likewise.
3139 * testsuite/libffi.call/problem1.c: Likewise.
3140 * testsuite/libffi.special/unwindtest.cc: Likewise.
3142 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
3144 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
3146 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
3148 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
3149 Add -lgcc_s to additional flags.
3151 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
3153 * configure.in, include/Makefile.am: PR libgcj/11147, install
3154 the ffitarget.h header file in a gcc versioned and target
3156 * configure: Regenerated.
3157 * Makefile.in, include/Makefile.in: Likewise.
3158 * testsuite/Makefile.in: Likewise.
3160 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
3162 * testsuite/libffi.call/closure_fn0.c: Print result and check
3163 with dg-output to make debugging easier.
3164 * testsuite/libffi.call/closure_fn1.c: Likewise.
3165 * testsuite/libffi.call/closure_fn2.c: Likewise.
3166 * testsuite/libffi.call/closure_fn3.c: Likewise.
3167 * testsuite/libffi.call/closure_fn4.c: Likewise.
3168 * testsuite/libffi.call/closure_fn5.c: Likewise.
3169 * testsuite/libffi.call/cls_12byte.c: Likewise.
3170 * testsuite/libffi.call/cls_16byte.c: Likewise.
3171 * testsuite/libffi.call/cls_18byte.c: Likewise.
3172 * testsuite/libffi.call/cls_19byte.c: Likewise.
3173 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3174 * testsuite/libffi.call/cls_20byte.c: Likewise.
3175 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3176 * testsuite/libffi.call/cls_24byte.c: Likewise.
3177 * testsuite/libffi.call/cls_2byte.c: Likewise.
3178 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3179 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3180 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3181 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3182 * testsuite/libffi.call/cls_4byte.c: Likewise.
3183 * testsuite/libffi.call/cls_5byte.c: Likewise.
3184 * testsuite/libffi.call/cls_64byte.c: Likewise.
3185 * testsuite/libffi.call/cls_6byte.c: Likewise.
3186 * testsuite/libffi.call/cls_7byte.c: Likewise.
3187 * testsuite/libffi.call/cls_8byte.c: Likewise.
3188 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3189 * testsuite/libffi.call/cls_9byte2.c: Likewise.
3190 * testsuite/libffi.call/cls_double.c: Likewise.
3191 * testsuite/libffi.call/cls_float.c: Likewise.
3192 * testsuite/libffi.call/cls_schar.c: Likewise.
3193 * testsuite/libffi.call/cls_sint.c: Likewise.
3194 * testsuite/libffi.call/cls_sshort.c: Likewise.
3195 * testsuite/libffi.call/cls_uchar.c: Likewise.
3196 * testsuite/libffi.call/cls_uint.c: Likewise.
3197 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3198 * testsuite/libffi.call/cls_ushort.c: Likewise.
3199 * testsuite/libffi.call/problem1.c: Likewise.
3201 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
3204 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
3206 * testsuite/libffi.call/cls_9byte2.c: New test case.
3207 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3208 * testsuite/libffi.call/cls_64byte.c: Likewise.
3209 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3210 * testsuite/libffi.call/cls_19byte.c: Likewise.
3211 * testsuite/libffi.call/cls_18byte.c: Likewise.
3212 * testsuite/libffi.call/closure_fn4.c: Likewise.
3213 * testsuite/libffi.call/closure_fn5.c: Likewise.
3214 * testsuite/libffi.call/cls_schar.c: Likewise.
3215 * testsuite/libffi.call/cls_sint.c: Likewise.
3216 * testsuite/libffi.call/cls_sshort.c: Likewise.
3217 * testsuite/libffi.call/nested_struct2.c: Likewise.
3218 * testsuite/libffi.call/nested_struct3.c: Likewise.
3220 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
3222 * testsuite/libffi.call/cls_double.c: Do a check on the result.
3223 * testsuite/libffi.call/cls_uchar.c: Likewise.
3224 * testsuite/libffi.call/cls_uint.c: Likewise.
3225 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3226 * testsuite/libffi.call/cls_ushort.c: Likewise.
3227 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
3229 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
3231 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
3234 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
3236 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
3237 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
3239 2003-10-22 David Daney <ddaney@avtrex.com>
3241 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
3242 __attribute__((__mode__(__SI__))) and friends.
3244 2003-10-22 Andreas Schwab <schwab@suse.de>
3246 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
3248 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
3250 * configure.in: AC_LINK_FILES(ffitarget.h).
3251 * configure: Regenerate.
3252 * Makefile.in: Likewise.
3253 * include/Makefile.in: Likewise.
3254 * testsuite/Makefile.in: Likewise.
3255 * fficonfig.h.in: Likewise.
3257 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
3258 Richard Henderson <rth@redhat.com>
3260 Avoid that ffi.h includes fficonfig.h.
3262 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
3263 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
3264 (TARGET_SRC_MIPS_SGI): Removed.
3265 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
3266 (MIPS_SGI): Removed.
3267 (CLEANFILES): Removed.
3268 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
3270 * acconfig.h: Removed.
3271 * configure.in: Compute sizeofs only for double and long double.
3272 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
3273 into AC_DEFINE instead of using acconfig.h. Create
3274 include/ffitarget.h instead of include/fficonfig.h. Rename
3275 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
3276 AC_DEFINE EH_FRAME_FLAGS.
3277 * include/Makefile.am (DISTCLEANFILES): New automake macro.
3278 (hack_DATA): Add ffitarget.h.
3279 * include/ffi.h.in: Remove all system specific definitions.
3280 Declare raw API even if it is not installed, why bother?
3281 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
3282 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
3283 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
3284 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
3285 * include/ffi_common.h (bool): Do not define.
3286 (ffi_assert): Accept failed assertion.
3287 (ffi_type_test): Return void and accept file/line.
3288 (FFI_ASSERT): Pass stringized failed assertion.
3289 (FFI_ASSERT_AT): New macro.
3290 (FFI_ASSERT_VALID_TYPE): New macro.
3291 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
3292 UINT64, SINT64): Define here with gcc's __attribute__ macro
3294 (FLOAT32, ALIGN): Define here instead of in ffi.h
3295 * include/ffi-mips.h: Removed. Its content moved to
3296 src/mips/ffitarget.h after separating assembly and C sections.
3297 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
3298 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
3299 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
3300 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
3301 SIZEOF_ARG -> FFI_SIZEOF_ARG.
3302 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
3303 * src/debug.c (ffi_assert): Accept stringized failed assertion.
3304 (ffi_type_test): Rewritten.
3305 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
3306 FFI_ASSERT_VALID_TYPE.
3307 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
3308 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
3309 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
3310 src/s390/ffitarget.h, src/sh/ffitarget.h,
3311 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
3312 src/x86/ffitarget.h: New files.
3313 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
3314 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
3315 src/powerpc/aix.S, src/powerpc/darwin.S,
3316 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
3317 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
3318 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
3319 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
3320 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
3323 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3325 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
3326 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
3328 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
3330 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
3331 Used when FFI_DEBUG = 1.
3333 2003-10-14 Alan Modra <amodra@bigpond.net.au>
3335 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
3338 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3340 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
3341 FFI_MIPS_O32 for O32 ABI.
3343 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
3345 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
3346 SPARC64. Cleanup whitespaces.
3348 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
3350 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
3351 strongarm, xscale. Cleanup whitespaces.
3352 * testsuite/libffi.call/closure_fn1.c: Likewise.
3353 * testsuite/libffi.call/closure_fn2.c: Likewise.
3354 * testsuite/libffi.call/closure_fn3.c: Likewise.
3355 * testsuite/libffi.call/cls_12byte.c: Likewise.
3356 * testsuite/libffi.call/cls_16byte.c: Likewise.
3357 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3358 * testsuite/libffi.call/cls_20byte.c: Likewise.
3359 * testsuite/libffi.call/cls_24byte.c: Likewise.
3360 * testsuite/libffi.call/cls_2byte.c: Likewise.
3361 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3362 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3363 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3364 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3365 * testsuite/libffi.call/cls_4byte.c: Likewise.
3366 * testsuite/libffi.call/cls_5byte.c: Likewise.
3367 * testsuite/libffi.call/cls_6byte.c: Likewise.
3368 * testsuite/libffi.call/cls_7byte.c: Likewise.
3369 * testsuite/libffi.call/cls_8byte.c: Likewise.
3370 * testsuite/libffi.call/cls_double.c: Likewise.
3371 * testsuite/libffi.call/cls_float.c: Likewise.
3372 * testsuite/libffi.call/cls_uchar.c: Likewise.
3373 * testsuite/libffi.call/cls_uint.c: Likewise.
3374 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3375 * testsuite/libffi.call/cls_ushort.c: Likewise.
3376 * testsuite/libffi.call/nested_struct.c: Likewise.
3377 * testsuite/libffi.call/nested_struct1.c: Likewise.
3378 * testsuite/libffi.call/problem1.c: Likewise.
3379 * testsuite/libffi.special/unwindtest.cc: Likewise.
3380 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
3382 2003-09-18 David Edelsohn <edelsohn@gnu.org>
3384 * src/powerpc/aix.S: Cleanup whitespaces.
3385 * src/powerpc/aix_closure.S: Likewise.
3387 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3389 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
3390 * src/powerpc/darwin_closure.S: Likewise.
3391 * src/powerpc/ffi_darwin.c: Likewise.
3393 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3394 David Edelsohn <edelsohn@gnu.org>
3396 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
3397 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
3399 * src/powerpc/darwin_closure.S: Likewise.
3400 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
3401 according to the Darwin/AIX ABI.
3402 (ffi_prep_cif_machdep): Likewise.
3403 (ffi_closure_helper_DARWIN): Likewise.
3404 Remove the outgoing parameter stack logic. Simplify the evaluation
3405 of the different CASE types.
3406 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
3407 statement in the trampoline code.
3409 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
3411 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
3412 for the register size.
3413 (ffi_closure_helper_SYSV): Handle the structure return value
3415 (ffi_closure_helper_SYSV): Return the appropriate type when
3416 the registers are used for the structure return value.
3417 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
3418 the 64-bit return value. Update copyright years.
3420 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3422 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
3423 srcdir for ffi_mips.h.
3425 2003-09-12 Alan Modra <amodra@bigpond.net.au>
3427 * src/prep_cif.c (initialize_aggregate): Include tail padding in
3429 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
3430 placement of float result.
3431 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
3432 cast of "resp" for big-endian 64 bit machines.
3434 2003-09-11 Alan Modra <amodra@bigpond.net.au>
3436 * src/types.c (double, longdouble): Merge identical SH and ARM
3437 typedefs, and add POWERPC64.
3438 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
3439 struct split over gpr and rest.
3440 (ffi_prep_cif_machdep): Correct intarg_count for structures.
3441 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
3443 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
3445 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
3448 2003-09-09 Alan Modra <amodra@bigpond.net.au>
3450 * configure: Regenerate.
3452 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3454 * Makefile.am: Remove build rules for ffitest.
3455 * Makefile.in: Rebuilt.
3457 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3459 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3460 about implicit declaration of abort().
3462 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3464 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3465 * Makefile.in: Rebuilt.
3466 * configure.in: Add dejagnu test framework.
3467 * configure: Rebuilt.
3469 * testsuite/Makefile.am: New file.
3470 * testsuite/Makefile.in: Built
3471 * testsuite/lib/libffi-dg.exp: New file.
3472 * testsuite/config/default.exp: Likewise.
3473 * testsuite/libffi.call/call.exp: Likewise.
3474 * testsuite/libffi.call/ffitest.h: Likewise.
3475 * testsuite/libffi.call/closure_fn0.c: Likewise.
3476 * testsuite/libffi.call/closure_fn1.c: Likewise.
3477 * testsuite/libffi.call/closure_fn2.c: Likewise.
3478 * testsuite/libffi.call/closure_fn3.c: Likewise.
3479 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3480 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3481 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3482 * testsuite/libffi.call/cls_2byte.c: Likewise.
3483 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3484 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3485 * testsuite/libffi.call/cls_4byte.c: Likewise.
3486 * testsuite/libffi.call/cls_5byte.c: Likewise.
3487 * testsuite/libffi.call/cls_6byte.c: Likewise.
3488 * testsuite/libffi.call/cls_7byte.c: Likewise.
3489 * testsuite/libffi.call/cls_8byte.c: Likewise.
3490 * testsuite/libffi.call/cls_12byte.c: Likewise.
3491 * testsuite/libffi.call/cls_16byte.c: Likewise.
3492 * testsuite/libffi.call/cls_20byte.c: Likewise.
3493 * testsuite/libffi.call/cls_24byte.c: Likewise.
3494 * testsuite/libffi.call/cls_double.c: Likewise.
3495 * testsuite/libffi.call/cls_float.c: Likewise.
3496 * testsuite/libffi.call/cls_uchar.c: Likewise.
3497 * testsuite/libffi.call/cls_uint.c: Likewise.
3498 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3499 * testsuite/libffi.call/cls_ushort.c: Likewise.
3500 * testsuite/libffi.call/float.c: Likewise.
3501 * testsuite/libffi.call/float1.c: Likewise.
3502 * testsuite/libffi.call/float2.c: Likewise.
3503 * testsuite/libffi.call/many.c: Likewise.
3504 * testsuite/libffi.call/many_win32.c: Likewise.
3505 * testsuite/libffi.call/nested_struct.c: Likewise.
3506 * testsuite/libffi.call/nested_struct1.c: Likewise.
3507 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3508 * testsuite/libffi.call/problem1.c: Likewise.
3509 * testsuite/libffi.call/promotion.c: Likewise.
3510 * testsuite/libffi.call/return_ll.c: Likewise.
3511 * testsuite/libffi.call/return_sc.c: Likewise.
3512 * testsuite/libffi.call/return_uc.c: Likewise.
3513 * testsuite/libffi.call/strlen.c: Likewise.
3514 * testsuite/libffi.call/strlen_win32.c: Likewise.
3515 * testsuite/libffi.call/struct1.c: Likewise.
3516 * testsuite/libffi.call/struct2.c: Likewise.
3517 * testsuite/libffi.call/struct3.c: Likewise.
3518 * testsuite/libffi.call/struct4.c: Likewise.
3519 * testsuite/libffi.call/struct5.c: Likewise.
3520 * testsuite/libffi.call/struct6.c: Likewise.
3521 * testsuite/libffi.call/struct7.c: Likewise.
3522 * testsuite/libffi.call/struct8.c: Likewise.
3523 * testsuite/libffi.call/struct9.c: Likewise.
3524 * testsuite/libffi.special/special.exp: New file.
3525 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3526 * testsuite/libffi.special/unwindtest.cc: Likewise.
3529 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3531 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3534 2003-08-02 Alan Modra <amodra@bigpond.net.au>
3536 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3538 (ffi_closure_helper_LINUX64): Likewise.
3539 * src/powerpc/linux64.S: Remove code writing to parm save area.
3540 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3541 address in lr from ffi_closure_helper_LINUX64 call to calculate
3542 table address. Optimize function tail.
3544 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3546 * src/sparc/ffi.c: Handle all floating point registers.
3547 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3549 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3551 * README: Note that libffi is not part of GCC. Update the project
3554 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3556 * src/powerpc/ppc_closure.S: Include ffi.h.
3558 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3560 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3561 Use C style comments.
3563 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3565 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3566 * Makefile.in: Regenerate.
3567 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3568 * configure: Regenerate.
3569 * include/ffi.h.in: Add SHmedia support.
3570 * src/sh64/ffi.c: New file.
3571 * src/sh64/sysv.S: New file.
3573 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3575 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3576 should be read-only.
3577 * configure: Rebuilt.
3578 * fficonfig.h.in: Rebuilt.
3579 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3580 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3581 * src/powerpc/linux64.S: Likewise.
3582 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3583 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3584 if -fpic/-fPIC/-mrelocatable.
3585 * src/powerpc/powerpc_closure.S: Likewise.
3586 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3587 #write in .eh_frame flags.
3588 * src/sparc/v9.S: Likewise.
3589 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3590 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3591 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3593 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3595 Fixes PR bootstrap/10656
3596 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3597 support for .register pseudo-op.
3598 * src/sparc/v8.S: Use it.
3599 * fficonfig.h.in: Rebuilt.
3600 * configure: Rebuilt.
3602 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3604 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3605 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3606 Make it the default on POWERPC64.
3607 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3608 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3609 * configure: Rebuilt.
3610 * src/powerpc/ffi.c (hidden): Define.
3611 (ffi_prep_args_SYSV): Renamed from
3612 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3613 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3614 ASM_NEEDS_REGISTERS64): New.
3615 (ffi_prep_args64): New function.
3616 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3617 (ffi_call): Likewise.
3618 (ffi_prep_closure): Likewise.
3619 (flush_icache): Surround by #ifndef POWERPC64.
3620 (ffi_dblfl): New union type.
3621 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3622 (ffi_closure_helper_LINUX64): New function.
3623 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3625 * src/powerpc/sysv.S: Likewise.
3626 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3627 * src/powerpc/linux64.S: New file.
3628 * src/powerpc/linux64_closure.S: New file.
3629 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3630 src/powerpc/linux64_closure.S.
3631 (TARGET_SRC_POWERPC): Likewise.
3633 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3634 closure_test_fn3): Fix result printing on big-endian 64-bit
3636 (main): Print tst2_arg instead of uninitialized tst2_result.
3638 * src/ffitest.c (main): Hide what closure pointer really points to
3641 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3643 * configure.in (arm-*-netbsdelf*): Add configuration.
3644 (configure): Regenerated.
3646 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3648 * include/Makefile.in: Regenerate.
3650 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3652 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3654 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3655 Receive closure pointer through parameter, read args using
3656 __builtin_dwarf_cfa.
3657 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3659 2003-03-12 Andreas Schwab <schwab@suse.de>
3661 * configure.in: Avoid trailing /. in toolexeclibdir.
3662 * configure: Rebuilt.
3664 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3666 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3668 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3670 * libffi/src/powerpc/darwin_closure.S:
3671 Fix alignement bug, allocate 8 bytes for the result.
3672 * libffi/src/powerpc/aix_closure.S:
3674 * libffi/src/powerpc/ffi_darwin.c:
3675 Update stackframe description for aix/darwin_closure.S.
3677 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3679 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3682 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3683 Andreas Schwab <schwab@suse.de>
3685 * configure.in: Adjust command to source config-ml.in to account
3686 for changes to the libffi_basedir definition.
3687 (libffi_basedir): Remove ${srcdir} from value and include trailing
3690 * configure: Regenerate.
3692 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3694 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3696 2003-01-28 Andrew Haley <aph@redhat.com>
3698 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3699 * src/x86/ffi64.c (ffi_prep_closure): New.
3700 (ffi_closure_UNIX64_inner): New.
3701 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3703 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
3705 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3706 Remove USE_LIBDIR conditional.
3707 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3708 * Makefile.in, configure: Rebuilt.
3710 2003-01027 David Edelsohn <edelsohn@gnu.org>
3712 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3713 * Makefile.in: Regenerate.
3715 2003-01-22 Andrew Haley <aph@redhat.com>
3717 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3720 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3722 * src/powerpc/darwin.S: Add unwind info.
3723 * src/powerpc/darwin_closure.S: Likewise.
3725 2003-01-14 Andrew Haley <aph@redhat.com>
3727 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3728 (ffi_prep_cif_machdep): Likewise.
3729 * src/x86/unix64.S: Add unwind info.
3731 2003-01-14 Andreas Jaeger <aj@suse.de>
3733 * src/ffitest.c (main): Only use ffi_closures if those are
3736 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3738 * libffi/src/ffitest.c
3739 add closure testcases
3741 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3743 * libffi/src/powerpc/ffi.c
3744 fix alignment bug for float (4 byte aligned iso 8 byte)
3746 2003-01-09 Geoffrey Keating <geoffk@apple.com>
3748 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3749 * src/powerpc/darwin.S: Remove RCS version string.
3751 2003-01-03 Jeff Sturm <jsturm@one-point.com>
3753 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3754 * src/ffitest.c (main): Use static storage for closure.
3755 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3756 * src/sparc/v8.S (ffi_closure_v8): New.
3757 * src/sparc/v9.S (ffi_closure_v9): New.
3759 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3761 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3762 enumeration for X86_WIN32.
3763 * src/x86/win32.S: Added ffi_call_STDCALL function
3765 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3766 switch cases for recognising FFI_STDCALL and
3767 calling ffi_call_STDCALL if target is X86_WIN32.
3768 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3769 stdcall versions of the "my_strlen" and "many"
3770 test functions (for X86_WIN32).
3771 Added test cases to test stdcall invocation using
3774 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3776 * src/sh/sysv.S: Add DWARF2 unwind info.
3778 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3780 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3782 2002-11-26 Jim Wilson <wilson@redhat.com>
3784 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3786 2002-11-23 H.J. Lu <hjl@gnu.org>
3788 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3789 Include ../config/accross.m4.
3790 * aclocal.m4; Rebuild.
3791 * configure: Likewise.
3793 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3795 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3797 2002-11-11 DJ Delorie <dj@redhat.com>
3799 * configure.in: Look for common files in the right place.
3801 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3803 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3804 raw data as _Jv_word values, not ffi_raw.
3805 (ffi_java_ptrarray_to_raw): Likewise.
3806 (ffi_java_rvalue_to_raw): New function.
3807 (ffi_java_raw_call): Call it.
3808 (ffi_java_raw_to_rvalue): New function.
3809 (ffi_java_translate_args): Call it.
3810 * src/ffitest.c (closure_test_fn): Interpret return value
3811 as ffi_arg, not int.
3812 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3813 FFI_TYPE_POINTER case.
3814 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3815 values extended to word size.
3817 2002-10-02 Andreas Jaeger <aj@suse.de>
3819 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3821 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
3823 * include/ffi.h.in: Fix i386 win32 compilation.
3825 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3827 * configure.in: Add s390x-*-linux-* target.
3828 * configure: Regenerate.
3829 * include/ffi.h.in: Define S390X for s390x targets.
3830 (FFI_CLOSURES): Define for s390/s390x.
3831 (FFI_TRAMPOLINE_SIZE): Likewise.
3832 (FFI_NATIVE_RAW_API): Likewise.
3833 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3834 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
3835 * src/s390/ffi.c: Major rework of existing code. Add support for
3836 s390x targets. Add closure support.
3837 * src/s390/sysv.S: Likewise.
3839 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
3841 * src/arm/sysv.S: Fix typo.
3843 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
3845 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3846 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3847 (ffi_call_SYSV): Handle soft-float.
3849 2002-09-27 Bo Thorsen <bo@suse.de>
3851 * include/ffi.h.in: Fix multilib x86-64 support.
3853 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3855 * Makefile.am (all-multi): Fix multilib parallel build.
3857 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3859 * configure.in (sh[34]*-*-linux*): Add brackets.
3860 * configure: Regenerate.
3862 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3864 * Makefile.am: Add SH support.
3865 * Makefile.in: Regenerate.
3866 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3867 * configure: Regenerate.
3868 * include/ffi.h.in: Add SH support.
3869 * src/sh/ffi.c: New file.
3870 * src/sh/sysv.S: New file.
3871 * src/types.c: Add SH support.
3873 2002-07-16 Bo Thorsen <bo@suse.de>
3875 * src/x86/ffi64.c: New file that adds x86-64 support.
3876 * src/x86/unix64.S: New file that handles argument setup for
3878 * src/x86/sysv.S: Don't use this on x86-64.
3879 * src/x86/ffi.c: Don't use this on x86-64.
3881 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3883 * src/ffitest.c (struct6): New test that tests a special case in
3885 (struct7): Likewise.
3886 (struct8): Likewise.
3887 (struct9): Likewise.
3888 (closure_test_fn): Silence warning about this when it's not used.
3889 (main): Add the new tests.
3890 (main): Fix a couple of wrong casts and silence some compiler warnings.
3891 * include/ffi.h.in: Add x86-64 ABI definition.
3892 * fficonfig.h.in: Regenerate.
3893 * Makefile.am: Add x86-64 support.
3894 * configure.in: Likewise.
3895 * Makefile.in: Regenerate.
3896 * configure: Likewise.
3898 2002-06-24 Bo Thorsen <bo@suse.de>
3900 * src/types.c: Merge settings for similar architectures.
3901 Add x86-64 sizes and alignments.
3903 2002-06-23 Bo Thorsen <bo@suse.de>
3905 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3906 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3907 * src/mips/ffi.c (ffi_prep_args): Likewise.
3908 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3910 2002-07-18 H.J. Lu (hjl@gnu.org)
3912 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3913 (libffi_la_SOURCES): Support MIPS_LINUX.
3914 (libffi_convenience_la_SOURCES): Likewise.
3915 * Makefile.in: Regenerated.
3917 * configure.in (mips64*-*): Skip.
3918 (mips*-*-linux*): New.
3919 * configure: Regenerated.
3921 * src/mips/ffi.c: Include <sgidefs.h>.
3923 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3925 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3927 2002-05-27 Roger Sayle <roger@eyesopen.com>
3929 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3931 2002-05-27 Bo Thorsen <bo@suse.de>
3933 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3936 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3938 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3939 beginning of function (for older apple cc).
3941 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3943 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3944 script entry, and set LD to it when configuring multilibs.
3945 * configure: Rebuilt.
3947 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3949 * configure.in (sparc64-*-netbsd*): Add target.
3950 (sparc-*-netbsdelf*): Likewise.
3951 * configure: Regenerate.
3953 2002-04-28 David S. Miller <davem@redhat.com>
3955 * configure.in, configure: Fix SPARC test in previous change.
3957 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3959 * Makefile.am: Add Linux for S/390 support.
3960 * Makefile.in: Regenerate.
3961 * configure.in: Add Linux for S/390 support.
3962 * configure: Regenerate.
3963 * include/ffi.h.in: Add Linux for S/390 support.
3964 * src/s390/ffi.c: New file from libffi CVS tree.
3965 * src/s390/sysv.S: New file from libffi CVS tree.
3967 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3969 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3971 * src/sparc/v8.S: Use it.
3972 * src/sparc/v9.S: Likewise.
3973 * fficonfig.h.in: Rebuilt.
3974 * configure: Rebuilt.
3976 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3978 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3980 * src/ia64/unix.S: Add unwind information. Fix comments.
3981 Save sp in a way that's compatible with unwind info.
3982 (ffi_call_unix): Correctly restore sp in all cases.
3983 * src/ia64/ffi.c: Add, fix comments.
3985 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3987 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3989 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3991 * configure.in (alpha*-*-netbsd*): Add target.
3992 * configure: Regenerate.
3994 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3996 * src/sparc/v8.S: Add unwind info.
3997 * src/sparc/v9.S: Likewise.
3999 2002-03-30 Krister Walfridsson <cato@df.lth.se>
4001 * configure.in: Enable i*86-*-netbsdelf*.
4002 * configure: Rebuilt.
4004 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
4007 * src/mips/n32.s: Delete
4008 * src/mips/o32.s: Delete
4010 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
4012 * configure.in: Enable alpha*-*-freebsd*.
4013 * configure: Rebuilt.
4015 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
4017 * Makefile.am: libfficonvenience -> libffi_convenience.
4018 * Makefile.in: Rebuilt.
4020 * Makefile.am: Define ffitest_OBJECTS.
4021 * Makefile.in: Rebuilt.
4023 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
4024 David Edelsohn <edelsohn@gnu.org>
4026 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
4027 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
4028 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
4029 * Makefile.in: Regenerate.
4030 * include/ffi.h.in: Add AIX and Darwin closure definitions.
4031 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
4032 (flush_icache, flush_range): New functions.
4033 (ffi_closure_helper_DARWIN): New function.
4034 * src/powerpc/aix_closure.S: New file.
4035 * src/powerpc/darwin_closure.S: New file.
4037 2002-02-24 Jeff Sturm <jsturm@one-point.com>
4039 * include/ffi.h.in: Add typedef for ffi_arg.
4040 * src/ffitest.c (main): Declare rint with ffi_arg.
4042 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
4044 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
4045 number of GPRs for floating-point arguments.
4047 2002-01-31 Anthony Green <green@redhat.com>
4049 * configure: Rebuilt.
4050 * configure.in: Replace CHECK_SIZEOF and endian tests with
4051 cross-compiler friendly macros.
4052 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
4055 2002-01-18 David Edelsohn <edelsohn@gnu.org>
4057 * src/powerpc/darwin.S (_ffi_call_AIX): New.
4058 * src/powerpc/aix.S (ffi_call_DARWIN): New.
4060 2002-01-17 David Edelsohn <edelsohn@gnu.org>
4062 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
4063 (TARGET_SRC_POWERPC_AIX): New.
4064 (POWERPC_AIX): New stanza.
4065 * Makefile.in: Regenerate.
4066 * configure.in: Add AIX case.
4067 * configure: Regenerate.
4068 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
4069 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
4070 size. Fix "long double" support.
4071 (ffi_call): Add FFI_AIX case.
4072 * src/powerpc/aix.S: New.
4074 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
4076 Implement Darwin PowerPC ABI.
4077 * configure.in: Handle powerpc-*-darwin*.
4078 * Makefile.am: Set source files for POWERPC_DARWIN.
4079 * configure: Rebuilt.
4080 * Makefile.in: Rebuilt.
4081 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
4083 * src/powerpc/darwin.S: New file.
4084 * src/powerpc/ffi_darwin.c: New file.
4086 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
4088 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
4090 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4092 * src/x86/sysv.S: Avoid gas-only .balign directive.
4093 Use C style comments.
4095 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4097 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
4098 Fixes PR bootstrap/3563.
4100 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4102 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
4104 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4106 * configure.in: Recognize sparc*-sun-* host.
4107 * configure: Regenerate.
4109 2001-06-06 Andrew Haley <aph@redhat.com>
4111 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
4113 2001-06-03 Andrew Haley <aph@redhat.com>
4115 * src/alpha/osf.S: Add unwind info.
4116 * src/powerpc/sysv.S: Add unwind info.
4117 * src/powerpc/ppc_closure.S: Likewise.
4119 2000-05-31 Jeff Sturm <jsturm@one-point.com>
4121 * configure.in: Fix AC_ARG_ENABLE usage.
4122 * configure: Rebuilt.
4124 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
4126 * configure.in: Remove warning about beta code.
4127 * configure: Rebuilt.
4129 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
4131 * src/ia64/unix.S: Restore stack pointer when returning from
4133 * src/ia64/ffi.c: Fix typo in comment.
4135 2001-04-18 Jim Wilson <wilson@redhat.com>
4137 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
4138 to eliminate RAW DV.
4140 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
4142 * Makefile.am: Make a libtool convenience library.
4143 * Makefile.in: Rebuilt.
4145 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
4147 * configure.in: Use different syntax for subdirectory creation.
4148 * configure: Rebuilt.
4150 2001-03-27 Jon Beniston <jon@beniston.com>
4152 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
4153 * configure: Rebuilt.
4154 * Makefile.am: Added X86_WIN32 target support.
4155 * Makefile.in: Rebuilt.
4157 * include/ffi.h.in: Added X86_WIN32 target support.
4159 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
4160 * src/types.c: Added X86_WIN32 target support.
4162 * src/x86/win32.S: New file. Based on sysv.S, but with EH
4163 stuff removed and made to work with CygWin's gas.
4165 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
4167 * configure.in: Make target subdirectory in build dir.
4168 * Makefile.am: Override suffix based rules to specify correct output
4170 * Makefile.in: Rebuilt.
4171 * configure: Rebuilt.
4173 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
4175 * src/powerpc/ppc_closure.S: New file.
4176 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
4177 involving long long and register pairs.
4178 (ffi_prep_closure): New function.
4179 (flush_icache): Likewise.
4180 (ffi_closure_helper_SYSV): Likewise.
4181 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
4182 (FFI_TRAMPOLINE_SIZE): Likewise.
4183 (FFI_NATIVE_RAW_API): Likewise.
4184 * Makefile.in: Rebuilt.
4185 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
4186 (TARGET_SRC_POWERPC): Likewise.
4188 2001-03-19 Tom Tromey <tromey@redhat.com>
4190 * Makefile.in: Rebuilt.
4191 * Makefile.am (ffitest_LDFLAGS): New macro.
4193 2001-03-02 Nick Clifton <nickc@redhat.com>
4195 * include/ffi.h.in: Remove RCS ident string.
4196 * include/ffi_mips.h: Remove RCS ident string.
4197 * src/debug.c: Remove RCS ident string.
4198 * src/ffitest.c: Remove RCS ident string.
4199 * src/prep_cif.c: Remove RCS ident string.
4200 * src/types.c: Remove RCS ident string.
4201 * src/alpha/ffi.c: Remove RCS ident string.
4202 * src/alpha/osf.S: Remove RCS ident string.
4203 * src/arm/ffi.c: Remove RCS ident string.
4204 * src/arm/sysv.S: Remove RCS ident string.
4205 * src/mips/ffi.c: Remove RCS ident string.
4206 * src/mips/n32.S: Remove RCS ident string.
4207 * src/mips/o32.S: Remove RCS ident string.
4208 * src/sparc/ffi.c: Remove RCS ident string.
4209 * src/sparc/v8.S: Remove RCS ident string.
4210 * src/sparc/v9.S: Remove RCS ident string.
4211 * src/x86/ffi.c: Remove RCS ident string.
4212 * src/x86/sysv.S: Remove RCS ident string.
4214 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
4216 * include/ffi.h.in: Change sourceware.cygnus.com references to
4219 2000-12-09 Richard Henderson <rth@redhat.com>
4221 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
4222 (ffi_closure_osf_inner): Index rather than increment avalue
4223 and arg_types. Give ffi_closure_osf the raw return value type.
4224 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
4227 2000-12-07 Richard Henderson <rth@redhat.com>
4229 * src/raw_api.c (ffi_translate_args): Fix typo.
4230 (ffi_prep_closure): Likewise.
4232 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
4233 FFI_TRAMPOLINE_SIZE.
4234 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
4235 cif->bytes for new ffi_call_osf implementation.
4236 (ffi_prep_args): Absorb into ...
4237 (ffi_call): ... here. Do all stack allocation here and
4238 avoid a callback function.
4239 (ffi_prep_closure, ffi_closure_osf_inner): New.
4240 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
4241 (ffi_closure_osf): New.
4243 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
4245 * config.guess, config.sub, install-sh: Removed.
4246 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
4247 * Makefile.in: Rebuilt.
4249 * acinclude.m4: Include libtool macros from the top level.
4250 * aclocal.m4, configure: Rebuilt.
4252 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
4254 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
4255 * configure: Rebuilt.
4257 2000-05-11 Scott Bambrough <scottb@netwinder.org>
4259 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
4260 memory correctly. Use conditional instructions, not branches where
4263 2000-05-04 Tom Tromey <tromey@cygnus.com>
4265 * configure: Rebuilt.
4266 * configure.in: Match `arm*-*-linux-*'.
4267 From Chris Dornan <cdornan@arm.com>.
4269 2000-04-28 Jakub Jelinek <jakub@redhat.com>
4271 * Makefile.am (SUBDIRS): Define.
4272 (AM_MAKEFLAGS): Likewise.
4273 (Multilib support.): Add section.
4274 * Makefile.in: Rebuilt.
4275 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
4276 New variables. Set for gcc using -print-multi-lib. Export them
4278 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
4279 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
4281 (extra_compiler_flags_value, extra_compiler_flags): Check these
4282 for extra compiler options which need to be passed down in
4285 2000-04-16 Anthony Green <green@redhat.com>
4287 * configure: Rebuilt.
4288 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
4290 2000-04-14 Jakub Jelinek <jakub@redhat.com>
4292 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
4293 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
4294 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
4295 Replace all void * sizeofs with sizeof(int).
4296 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
4297 different than DOUBLE.
4298 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
4299 (ffi_prep_args_v9): New function.
4300 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
4301 (ffi_V9_return_struct): New function.
4302 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
4303 32bit code (not yet cross-arch calls).
4304 * src/sparc/v8.S: Add struct return delay nop.
4306 * src/sparc/v9.S: New file.
4307 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
4308 is used on sparc64 only for structures larger than 32 bytes.
4309 Pass by reference for structures is done for structure arguments
4310 larger than 16 bytes.
4311 * src/ffitest.c (main): Use 64bit rint on sparc64.
4312 Run long long tests on sparc.
4313 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
4315 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
4317 * configure.in (sparc-*-linux*): New supported target.
4318 (sparc64-*-linux*): Likewise.
4319 * configure: Rebuilt.
4320 * Makefile.am: Add v9.S to SPARC files.
4321 * Makefile.in: Likewise.
4322 (LINK): Surround $(CCLD) into double quotes, so that multilib
4323 compiles work correctly.
4325 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
4327 * configure: Rebuilt.
4328 * configure.in: (i*86-*-solaris*): New libffi target. Patch
4329 proposed by Bryce McKinlay.
4331 2000-03-20 Tom Tromey <tromey@cygnus.com>
4333 * Makefile.in: Hand edit for java_raw_api.lo.
4335 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
4337 * config.guess, config.sub: Update from the gcc tree.
4338 Fix for PR libgcj/168.
4340 2000-03-03 Tom Tromey <tromey@cygnus.com>
4342 * Makefile.in: Fixed ia64 by hand.
4344 * configure: Rebuilt.
4345 * configure.in (--enable-multilib): New option.
4346 (libffi_basedir): New subst.
4347 (AC_OUTPUT): Added multilib code.
4349 2000-03-02 Tom Tromey <tromey@cygnus.com>
4351 * Makefile.in: Rebuilt.
4352 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
4355 2000-02-25 Hans Boehm <boehm@acm.org>
4357 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
4359 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
4361 (ffi_prep_raw_closure): Use ffi_translate_args, not
4362 ffi_closure_translate.
4363 * src/java_raw_api.c: New file.
4364 * src/ffitest.c (closure_test_fn): New function.
4365 (main): Define `rint' as long long on IA64. Added new test when
4366 FFI_CLOSURES is defined.
4367 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
4368 (ffi_abi): Recognize IA64.
4369 (ffi_raw): Added `flt' field.
4370 Added "Java raw API" code.
4371 * configure.in: Recognize ia64.
4372 * Makefile.am (TARGET_SRC_IA64): New macro.
4373 (libffi_la_common_SOURCES): Added java_raw_api.c.
4374 (libffi_la_SOURCES): Define in IA64 case.
4376 2000-01-04 Tom Tromey <tromey@cygnus.com>
4378 * Makefile.in: Rebuilt with newer automake.
4380 1999-12-31 Tom Tromey <tromey@cygnus.com>
4382 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
4384 1999-09-01 Tom Tromey <tromey@cygnus.com>
4386 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
4388 * fficonfig.h.in: Rebuilt.
4389 * configure: Rebuilt.
4390 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
4391 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
4392 * acconfig.h: Don't #undef PACKAGE or VERSION.
4394 1999-08-09 Anthony Green <green@cygnus.com>
4396 * include/ffi.h.in: Try to work around messy header problem
4397 with PACKAGE and VERSION.
4399 * configure: Rebuilt.
4400 * configure.in: Change version to 2.00-beta.
4402 * fficonfig.h.in: Rebuilt.
4403 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
4405 * src/x86/ffi.c (ffi_raw_call): Rename.
4407 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
4409 * src/x86/ffi.c (ffi_closure_SYSV): New function.
4410 (ffi_prep_incoming_args_SYSV): Ditto.
4411 (ffi_prep_closure): Ditto.
4412 (ffi_closure_raw_SYSV): Ditto.
4413 (ffi_prep_raw_closure): More ditto.
4414 (ffi_call_raw): Final ditto.
4416 * include/ffi.h.in: Add definitions for closure and raw API.
4418 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4421 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
4423 * src/raw_api.c: New file.
4425 * include/ffi.h.in (ffi_raw): New type.
4426 (UINT_ARG, SINT_ARG): New defines.
4427 (ffi_closure, ffi_raw_closure): New types.
4428 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4430 * configure.in: Add check for endianness and sizeof void*.
4432 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
4433 instead of directly.
4435 * configure: Rebuilt.
4437 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
4439 * configure.in: Add x86 and powerpc BeOS configurations.
4440 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
4442 1999-05-09 Anthony Green <green@cygnus.com>
4444 * configure.in: Add warning about this being beta code.
4445 Remove src/Makefile.am from the picture.
4446 * configure: Rebuilt.
4448 * Makefile.am: Move logic from src/Makefile.am. Add changes
4449 to support libffi as a target library.
4450 * Makefile.in: Rebuilt.
4452 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
4453 Upgraded to new autoconf, automake, libtool.
4457 * LICENSE: Update copyright date.
4459 * src/Makefile.am, src/Makefile.in: Removed.
4461 1998-11-29 Anthony Green <green@cygnus.com>
4463 * include/ChangeLog: Removed.
4464 * src/ChangeLog: Removed.
4465 * src/mips/ChangeLog: Removed.
4466 * src/sparc/ChangeLog: Remboved.
4467 * src/x86/ChangeLog: Removed.
4469 * ChangeLog.v1: Created.