1 2010-07-07 Neil Roberts <neil@linux.intel.com>
3 * src/x86/sysv.S (ffi_call_SYSV): Align the stack pointer to
6 2010-07-02 Jakub Jelinek <jakub@redhat.com>
8 * Makefile.am (AM_MAKEFLAGS): Pass also mandir to submakes.
9 * Makefile.in: Regenerated.
11 2010-05-19 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
13 * configure.ac (libffi_cv_as_x86_pcrel): Check for illegal in as
15 (libffi_cv_as_ascii_pseudo_op): Check for .ascii.
16 (libffi_cv_as_string_pseudo_op): Check for .string.
17 * configure: Regenerate.
18 * fficonfig.h.in: Regenerate.
19 * src/x86/sysv.S (.eh_frame): Use .ascii, .string or error.
21 2010-04-07 Jakub Jelinek <jakub@redhat.com>
23 * regex.c (byte_re_match_2_internal): Avoid set but not used
26 2010-04-02 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
28 * Makefile.in: Regenerate.
29 * aclocal.m4: Regenerate.
30 * include/Makefile.in: Regenerate.
31 * man/Makefile.in: Regenerate.
32 * testsuite/Makefile.in: Regenerate.
34 2010-03-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
36 * configure.ac (libffi_cv_as_x86_64_unwind_section_type): New test.
37 * configure: Regenerate.
38 * fficonfig.h.in: Regenerate.
39 * libffi/src/x86/unix64.S (.eh_frame)
40 [HAVE_AS_X86_64_UNWIND_SECTION_TYPE]: Use @unwind section type.
42 2010-03-14 Matthias Klose <doko@ubuntu.com>
44 * src/x86/ffi64.c: Fix typo in comment.
45 * src/x86/ffi.c: Use /* ... */ comment style.
47 2010-02-24 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
49 * doc/libffi.texi (The Closure API): Fix typo.
50 * doc/libffi.info: Remove.
52 2010-02-15 Matthias Klose <doko@ubuntu.com>
54 * src/arm/sysv.S (__ARM_ARCH__): Define for processor
57 2010-01-15 Anthony Green <green@redhat.com>
59 * README: Add notes on building with Microsoft Visual C++.
61 2010-01-15 Daniel Witte <dwitte@mozilla.com>
65 * src/x86/win32.S: Port assembly routines to MSVC and #ifdef.
66 * src/x86/ffi.c: Tweak function declaration and remove excess
68 * include/ffi.h.in: Add __declspec(align(8)) to typedef struct
71 * src/x86/ffi.c: Merge ffi_call_SYSV and ffi_call_STDCALL into new
72 function ffi_call_win32 on X86_WIN32.
73 * src/x86/win32.S (ffi_call_SYSV): Rename to ffi_call_win32.
74 (ffi_call_STDCALL): Remove.
76 * src/prep_cif.c (ffi_prep_cif): Move stack space allocation code
77 to ffi_prep_cif_machdep for x86.
78 * src/x86/ffi.c (ffi_prep_cif_machdep): To here.
80 2010-01-15 Oliver Kiddle <okiddle@yahoo.co.uk>
82 * src/x86/ffitarget.h (ffi_abi): Check for __i386 and __amd64 for
83 Sun Studio compiler compatibility.
85 2010-01-12 Conrad Irwin <conrad.irwin@gmail.com>
87 * doc/libffi.texi: Add closure example.
89 2010-01-07 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
92 * testsuite/libffi.call/ffitest.h [__alpha__ && __osf__] (PRIdLL,
93 PRIuLL, PRId64, PRIu64, PRIuPTR): Define.
94 * testsuite/libffi.call/cls_align_sint64.c: Add -Wno-format on
96 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
97 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
98 * testsuite/libffi.call/return_ll1.c: Likewise.
99 * testsuite/libffi.call/stret_medium2.c: Likewise.
100 * testsuite/libffi.special/ffitestcxx.h (allocate_mmap): Cast
101 MAP_FAILED to char *.
103 2010-01-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
105 * src/mips/n32.S: Use .abicalls and .eh_frame with __GNUC__.
107 2009-12-31 Anthony Green <green@redhat.com>
109 * README: Update for libffi 3.0.9.
111 2009-12-27 Matthias Klose <doko@ubuntu.com>
113 * configure.ac (HAVE_LONG_DOUBLE): Define for mips when
115 * configure: Rebuilt.
117 2009-12-26 Anthony Green <green@redhat.com>
119 * testsuite/libffi.call/cls_longdouble_va.c: Mark as xfail for
121 * testsuite/libffi.call/cls_double_va.c: Ditto.
123 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
125 * testsuite/libffi.call/ffitest.h: Conditionally include stdint.h
127 * testsuite/libffi.special/unwindtest.cc: Ditto.
129 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
131 * configure.ac: Add amd64-*-openbsd*.
132 * configure: Rebuilt.
133 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Link
134 openbsd programs with -lpthread.
136 2009-12-26 Anthony Green <green@redhat.com>
138 * testsuite/libffi.call/cls_double_va.c,
139 testsuite/libffi.call/cls_longdouble.c,
140 testsuite/libffi.call/cls_longdouble_va.c,
141 testsuite/libffi.call/cls_pointer.c,
142 testsuite/libffi.call/cls_pointer_stack.c: Remove xfail for
143 mips*-*-* and arm*-*-*.
144 * testsuite/libffi.call/cls_align_longdouble_split.c,
145 testsuite/libffi.call/cls_align_longdouble_split2.c,
146 testsuite/libffi.call/stret_medium2.c,
147 testsuite/libffi.call/stret_medium.c,
148 testsuite/libffi.call/stret_large.c,
149 testsuite/libffi.call/stret_large2.c: Remove xfail for arm*-*-*.
151 2009-12-31 Kay Tietz <ktietz70@googlemail.com>
153 * testsuite/libffi.call/ffitest.h,
154 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRuLL): Fix
157 2009-12-31 Carlo Bramini <carlo.bramix@libero.it>
159 * configure.ac (AM_LTLDFLAGS): Define for windows hosts.
160 * Makefile.am (libffi_la_LDFLAGS): Add AM_LTLDFLAGS.
161 * configure: Rebuilt.
162 * Makefile.in: Rebuilt.
164 2009-12-31 Anthony Green <green@redhat.com>
167 * testsuite/libffi.call/huge_struct.c: New test case from Blake
170 2009-12-28 David Edelsohn <edelsohn@gnu.org>
172 * src/powerpc/ffi_darwin.c (ffi_prep_args): Copy abi and nargs to
174 (aix_adjust_aggregate_sizes): New function.
175 (ffi_prep_cif_machdep): Call it.
177 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
179 * configure.ac: Define FFI_MMAP_EXEC_WRIT for the given targets.
180 * configure: Regenerate.
181 * fficonfig.h.in: Likewise.
182 * src/closures.c: Remove the FFI_MMAP_EXEC_WRIT definition for
185 2009-12-26 Andreas Schwab <schwab@linux-m68k.org>
187 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Advance intarg_count
188 when a float arguments is passed in memory.
189 (ffi_closure_helper_SYSV): Mark general registers as used up when
190 a 64bit or soft-float long double argument is passed in memory.
192 2009-12-25 Matthias Klose <doko@ubuntu.com>
194 * man/ffi_call.3: Fix #include in examples.
195 * doc/libffi.texi: Add dircategory.
197 2009-12-25 Frank Everdij <f.p.x.everdij@tudelft.nl>
199 * include/ffi.h.in: Placed '__GNUC__' ifdef around
200 '__attribute__((aligned(8)))' in ffi_closure, fixes compile for
202 * include/ffi_common.h: Added '__sgi' define to non
203 '__attribute__((__mode__()))' integer typedefs.
204 * src/mips/ffi.c (ffi_call, ffi_closure_mips_inner_O32,
205 ffi_closure_mips_inner_N32): Added 'defined(_MIPSEB)' to BE check.
206 (ffi_closure_mips_inner_O32, ffi_closure_mips_inner_N32): Added
207 FFI_LONGDOUBLE support and alignment(N32 only).
208 * src/mips/ffitarget.h: Corrected '#include <sgidefs.h>' for IRIX and
209 fixed non '__attribute__((__mode__()))' integer typedefs.
210 * src/mips/n32.S: Put '#ifdef linux' around '.abicalls' and '.eh_frame'
211 since they are Linux/GNU Assembler specific.
213 2009-12-25 Bradley Smith <brad@brad-smith.co.uk>
215 * configure.ac, Makefile.am, src/avr32/ffi.c,
216 src/avr32/ffitarget.h,
217 src/avr32/sysv.S: Add AVR32 port.
218 * configure, Makefile.in: Rebuilt.
220 2009-12-21 Andreas Tobler <a.tobler@schweiz.org>
222 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
223 * configure: Regenerate.
225 2009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
227 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
229 2009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
231 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
234 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
236 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
239 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
241 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
243 2009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
246 * src/closures.c [X86_64 && __sun__ && __svr4__]
247 (FFI_MMAP_EXEC_WRIT): Define.
249 2009-12-08 David Daney <ddaney@caviumnetworks.com>
251 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
252 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
253 * testsuite/libffi.call/stret_large.c: Same.
254 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
255 * testsuite/libffi.call/stret_large2.c: Same.
256 * testsuite/libffi.call/stret_medium2.c: Same.
258 2009-12-07 David Edelsohn <edelsohn@gnu.org>
260 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
263 2009-12-05 David Edelsohn <edelsohn@gnu.org>
265 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
267 * src/powerpc/aix_closure.S: Same.
269 2009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
271 * Makefile.in: Regenerate.
272 * configure: Regenerate.
273 * include/Makefile.in: Regenerate.
274 * man/Makefile.in: Regenerate.
275 * testsuite/Makefile.in: Regenerate.
277 2009-12-04 David Edelsohn <edelsohn@gnu.org>
279 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
282 2009-12-04 Uros Bizjak <ubizjak@gmail.com>
285 * src/x86/ffi64.c (classify_argument): Update from
286 gcc/config/i386/i386.c.
287 (ffi_closure_unix64_inner): Do not use the address of two consecutive
288 SSE registers directly.
289 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
290 for x86_64 linux targets.
292 2009-12-04 David Edelsohn <edelsohn@gnu.org>
294 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
295 pfr for long double split between fpr13 and stack.
297 2009-12-03 David Edelsohn <edelsohn@gnu.org>
299 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
300 fparg_count twice for long double.
302 2009-12-03 David Edelsohn <edelsohn@gnu.org>
305 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
307 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
309 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
310 Remove xfails for x86 linux targets.
312 2009-12-02 David Edelsohn <edelsohn@gnu.org>
314 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
317 2009-12-01 David Edelsohn <edelsohn@gnu.org>
319 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
320 register usage. Call ffi_prep_args directly. Add long double
321 return value support.
322 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
323 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
324 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
325 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
327 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
329 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
330 Allocate result area between params and FPRs.
332 2009-11-30 David Edelsohn <edelsohn@gnu.org>
335 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
337 * src/powerpc/aix.S: Implement AIX64 version.
338 * src/powerpc/aix_closure.S: Implement AIX64 version.
339 (ffi_closure_ASM): Use extsb, lha and displament addresses.
340 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
342 (ffi_prep_cif_machdep): Same.
344 (ffi_closure_helper_DARWIN): Same.
346 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
349 * testsuite/libffi.call/testclosure.c: New test.
351 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
353 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
354 assembly version use of ___chkstk.
356 2009-09-23 Matthias Klose <doko@ubuntu.com>
358 PR libffi/40242, PR libffi/41443
359 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
360 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
361 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
362 Change the conditionals to __SOFTFP__ || __ARM_EABI__
363 for -mfloat-abi=softfp to work.
365 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
367 PR testsuite/32843 (strikes again)
368 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
369 enable proper extension on char and short.
371 2009-09-15 David Daney <ddaney@caviumnetworks.com>
373 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
374 handling for FFI_TYPE_POINTER.
375 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
376 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
377 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
378 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
379 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
380 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
381 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
382 structure and pointer returns.
383 (ffi_closure_N32): Add handling for pointer returns.
384 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
385 calc_n32_return_struct_flags): Handle soft-float.
386 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
387 (ffi_call_N32): Declare proper argument types.
388 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
391 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
393 * configure.ac (AC_PREREQ): Bump to 2.64.
395 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
397 * Makefile.am (install-html, install-pdf): Remove.
398 * Makefile.in: Regenerate.
400 * Makefile.in: Regenerate.
401 * aclocal.m4: Regenerate.
402 * configure: Regenerate.
403 * fficonfig.h.in: Regenerate.
404 * include/Makefile.in: Regenerate.
405 * man/Makefile.in: Regenerate.
406 * testsuite/Makefile.in: Regenerate.
408 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
410 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
412 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
415 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
416 return types for X86_WIN32.
417 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
418 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
419 _ffi_closure_STDCALL): Likewise.
421 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
422 (dlmmap, dlmunmap): Also use these functions on Cygwin.
424 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
429 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
432 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
434 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
435 to "" before adding paths. (This reinstates an assignment that
436 was removed by my 2009-06-30 commit, but changes the initial
437 value from "." to "".)
439 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
442 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
443 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
446 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
448 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
449 to ld_library_path. Use add_path. Add just find_libgcc_s
450 to ld_library_path, not every libgcc multilib directory.
452 2009-06-16 Wim Lewis <wiml@hhhh.org>
454 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
455 supposed to be callee-saved.
456 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
457 return buffer for odd-size structs.
459 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
462 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
463 allow_stack_execute for Darwin.
465 2009-06-16 Andrew Haley <aph@redhat.com>
467 * configure.ac (TARGETDIR): Add missing blank lines.
468 * configure: Regenerate.
470 2009-06-16 Andrew Haley <aph@redhat.com>
472 * testsuite/libffi.call/cls_align_sint64.c,
473 testsuite/libffi.call/cls_align_uint64.c,
474 testsuite/libffi.call/cls_longdouble_va.c,
475 testsuite/libffi.call/cls_ulonglong.c,
476 testsuite/libffi.call/return_ll1.c,
477 testsuite/libffi.call/stret_medium2.c: Fix printf format
479 * testsuite/libffi.call/ffitest.h,
480 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
482 2009-06-15 Andrew Haley <aph@redhat.com>
484 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
485 * testsuite/libffi.call/err_bad_abi.c: Likewise.
487 2009-06-12 Andrew Haley <aph@redhat.com>
489 * Makefile.am: Remove info_TEXINFOS.
491 2009-06-12 Andrew Haley <aph@redhat.com>
493 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
494 testsuite/libffi.call/cls_align_uint64.c,
495 testsuite/libffi.call/cls_ulonglong.c,
496 testsuite/libffi.call/return_ll1.c,
497 testsuite/libffi.call/stret_medium2.c: Fix printf format
499 testsuite/libffi.special/unwindtest.cc: include stdint.h.
501 2009-06-11 Timothy Wall <twall@users.sf.net>
506 include/ffi_common.h,
511 src/x86/win64.S (new),
512 README: Added win64 support (mingw or MSVC)
516 testsuite/Makefile.in,
518 aclocal.m4: Regenerated
519 * ltcf-c.sh: properly escape cygwin/w32 path
520 * man/ffi_call.3: Clarify size requirements for return value.
521 * src/x86/ffi64.c: Fix filename in comment.
522 * src/x86/win32.S: Remove unused extern.
524 * testsuite/libffi.call/closure_fn0.c,
525 testsuite/libffi.call/closure_fn1.c,
526 testsuite/libffi.call/closure_fn2.c,
527 testsuite/libffi.call/closure_fn3.c,
528 testsuite/libffi.call/closure_fn4.c,
529 testsuite/libffi.call/closure_fn5.c,
530 testsuite/libffi.call/closure_fn6.c,
531 testsuite/libffi.call/closure_stdcall.c,
532 testsuite/libffi.call/cls_12byte.c,
533 testsuite/libffi.call/cls_16byte.c,
534 testsuite/libffi.call/cls_18byte.c,
535 testsuite/libffi.call/cls_19byte.c,
536 testsuite/libffi.call/cls_1_1byte.c,
537 testsuite/libffi.call/cls_20byte.c,
538 testsuite/libffi.call/cls_20byte1.c,
539 testsuite/libffi.call/cls_24byte.c,
540 testsuite/libffi.call/cls_2byte.c,
541 testsuite/libffi.call/cls_3_1byte.c,
542 testsuite/libffi.call/cls_3byte1.c,
543 testsuite/libffi.call/cls_3byte2.c,
544 testsuite/libffi.call/cls_4_1byte.c,
545 testsuite/libffi.call/cls_4byte.c,
546 testsuite/libffi.call/cls_5_1_byte.c,
547 testsuite/libffi.call/cls_5byte.c,
548 testsuite/libffi.call/cls_64byte.c,
549 testsuite/libffi.call/cls_6_1_byte.c,
550 testsuite/libffi.call/cls_6byte.c,
551 testsuite/libffi.call/cls_7_1_byte.c,
552 testsuite/libffi.call/cls_7byte.c,
553 testsuite/libffi.call/cls_8byte.c,
554 testsuite/libffi.call/cls_9byte1.c,
555 testsuite/libffi.call/cls_9byte2.c,
556 testsuite/libffi.call/cls_align_double.c,
557 testsuite/libffi.call/cls_align_float.c,
558 testsuite/libffi.call/cls_align_longdouble.c,
559 testsuite/libffi.call/cls_align_longdouble_split.c,
560 testsuite/libffi.call/cls_align_longdouble_split2.c,
561 testsuite/libffi.call/cls_align_pointer.c,
562 testsuite/libffi.call/cls_align_sint16.c,
563 testsuite/libffi.call/cls_align_sint32.c,
564 testsuite/libffi.call/cls_align_sint64.c,
565 testsuite/libffi.call/cls_align_uint16.c,
566 testsuite/libffi.call/cls_align_uint32.c,
567 testsuite/libffi.call/cls_align_uint64.c,
568 testsuite/libffi.call/cls_dbls_struct.c,
569 testsuite/libffi.call/cls_double.c,
570 testsuite/libffi.call/cls_double_va.c,
571 testsuite/libffi.call/cls_float.c,
572 testsuite/libffi.call/cls_longdouble.c,
573 testsuite/libffi.call/cls_longdouble_va.c,
574 testsuite/libffi.call/cls_multi_schar.c,
575 testsuite/libffi.call/cls_multi_sshort.c,
576 testsuite/libffi.call/cls_multi_sshortchar.c,
577 testsuite/libffi.call/cls_multi_uchar.c,
578 testsuite/libffi.call/cls_multi_ushort.c,
579 testsuite/libffi.call/cls_multi_ushortchar.c,
580 testsuite/libffi.call/cls_pointer.c,
581 testsuite/libffi.call/cls_pointer_stack.c,
582 testsuite/libffi.call/cls_schar.c,
583 testsuite/libffi.call/cls_sint.c,
584 testsuite/libffi.call/cls_sshort.c,
585 testsuite/libffi.call/cls_uchar.c,
586 testsuite/libffi.call/cls_uint.c,
587 testsuite/libffi.call/cls_ulonglong.c,
588 testsuite/libffi.call/cls_ushort.c,
589 testsuite/libffi.call/err_bad_abi.c,
590 testsuite/libffi.call/err_bad_typedef.c,
591 testsuite/libffi.call/float2.c,
592 testsuite/libffi.call/huge_struct.c,
593 testsuite/libffi.call/nested_struct.c,
594 testsuite/libffi.call/nested_struct1.c,
595 testsuite/libffi.call/nested_struct10.c,
596 testsuite/libffi.call/nested_struct2.c,
597 testsuite/libffi.call/nested_struct3.c,
598 testsuite/libffi.call/nested_struct4.c,
599 testsuite/libffi.call/nested_struct5.c,
600 testsuite/libffi.call/nested_struct6.c,
601 testsuite/libffi.call/nested_struct7.c,
602 testsuite/libffi.call/nested_struct8.c,
603 testsuite/libffi.call/nested_struct9.c,
604 testsuite/libffi.call/problem1.c,
605 testsuite/libffi.call/return_ldl.c,
606 testsuite/libffi.call/return_ll1.c,
607 testsuite/libffi.call/stret_large.c,
608 testsuite/libffi.call/stret_large2.c,
609 testsuite/libffi.call/stret_medium.c,
610 testsuite/libffi.call/stret_medium2.c,
611 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
612 of checking for MMAP. Use intptr_t instead of long casts.
614 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
616 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
617 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
618 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
620 2009-06-09 Andrew Haley <aph@redhat.com>
622 * src/x86/freebsd.S: Add missing file.
624 2009-06-08 Andrew Haley <aph@redhat.com>
626 Import from libffi 3.0.8:
628 * doc/libffi.texi: New file.
629 * doc/libffi.info: Likewise.
630 * doc/stamp-vti: Likewise.
631 * man/Makefile.am: New file.
632 * man/ffi_call.3: New file.
634 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
636 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
638 * configure.ac: Bump version to 3.0.8.
639 parisc*-*-linux*: Add.
640 i386-*-freebsd* | i386-*-openbsd*: Add.
641 powerpc-*-beos*: Add.
642 AM_CONDITIONAL X86_FREEBSD: Add.
643 AC_CONFIG_FILES: Add man/Makefile.
645 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
647 2009-06-08 Andrew Haley <aph@redhat.com>
649 * README: Import from libffi 3.0.8.
651 2009-06-08 Andrew Haley <aph@redhat.com>
653 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
654 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
655 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
656 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
658 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
659 * testsuite/libffi.call/stret_medium.c: Likewise.
660 * testsuite/libffi.call/stret_large2.c: Likewise.
661 * testsuite/libffi.call/stret_large.c: Likewise.
663 2008-12-26 Timothy Wall <twall@users.sf.net>
665 * testsuite/libffi.call/cls_longdouble.c,
666 testsuite/libffi.call/cls_longdouble_va.c,
667 testsuite/libffi.call/cls_align_longdouble.c,
668 testsuite/libffi.call/cls_align_longdouble_split.c,
669 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
670 failures on x86_64 cygwin/mingw.
672 2008-12-22 Timothy Wall <twall@users.sf.net>
674 * testsuite/libffi.call/closure_fn0.c,
675 testsuite/libffi.call/closure_fn1.c,
676 testsuite/libffi.call/closure_fn2.c,
677 testsuite/libffi.call/closure_fn3.c,
678 testsuite/libffi.call/closure_fn4.c,
679 testsuite/libffi.call/closure_fn5.c,
680 testsuite/libffi.call/closure_fn6.c,
681 testsuite/libffi.call/closure_loc_fn0.c,
682 testsuite/libffi.call/closure_stdcall.c,
683 testsuite/libffi.call/cls_align_pointer.c,
684 testsuite/libffi.call/cls_pointer.c,
685 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
686 pointer to integer (intptr_t).
687 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
689 2008-07-24 Anthony Green <green@redhat.com>
691 * testsuite/libffi.call/cls_dbls_struct.c,
692 testsuite/libffi.call/cls_double_va.c,
693 testsuite/libffi.call/cls_longdouble.c,
694 testsuite/libffi.call/cls_longdouble_va.c,
695 testsuite/libffi.call/cls_pointer.c,
696 testsuite/libffi.call/cls_pointer_stack.c,
697 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
700 2008-03-04 Anthony Green <green@redhat.com>
704 * testsuite/libffi.call/cls_align_longdouble_split2.c
705 testsuite/libffi.call/cls_align_longdouble_split.c
706 testsuite/libffi.call/cls_dbls_struct.c
707 testsuite/libffi.call/cls_double_va.c
708 testsuite/libffi.call/cls_longdouble.c
709 testsuite/libffi.call/cls_longdouble_va.c
710 testsuite/libffi.call/cls_pointer.c
711 testsuite/libffi.call/cls_pointer_stack.c
712 testsuite/libffi.call/err_bad_abi.c
713 testsuite/libffi.call/err_bad_typedef.c
714 testsuite/libffi.call/stret_large2.c
715 testsuite/libffi.call/stret_large.c
716 testsuite/libffi.call/stret_medium2.c
717 testsuite/libffi.call/stret_medium.c: New tests from Apple.
719 2009-06-05 Andrew Haley <aph@redhat.com>
721 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
724 2009-06-04 Andrew Haley <aph@redhat.com>
726 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
729 2008-02-26 Anthony Green <green@redhat.com>
730 Thomas Heller <theller@ctypes.org>
732 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
735 2008-02-03 Timothy Wall <twall@users.sf.net>
737 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
738 offset based on code pointer, not data pointer.
740 2008-01-31 Timothy Wall <twall@users.sf.net>
742 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
744 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
746 * src/x86/win32.S: Add assembly for stdcall closure.
747 * src/x86/ffi.c: Initialize stdcall closure trampoline.
749 2009-06-04 Andrew Haley <aph@redhat.com>
751 * include/ffi.h.in: Change void (*)() to void (*)(void).
752 * src/x86/ffi.c: Likewise.
754 2009-06-04 Andrew Haley <aph@redhat.com>
756 * src/powerpc/ppc_closure.S: Insert licence header.
757 * src/powerpc/linux64_closure.S: Likewise.
758 * src/m68k/sysv.S: Likewise.
760 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
761 * src/powerpc/ffi.c: Likewise.
762 * src/powerpc/ffi_darwin.c: Likewise.
763 * src/m32r/ffi.c: Likewise.
764 * src/sh64/ffi.c: Likewise.
765 * src/x86/ffi64.c: Likewise.
766 * src/alpha/ffi.c: Likewise.
767 * src/alpha/osf.S: Likewise.
768 * src/frv/ffi.c: Likewise.
769 * src/s390/ffi.c: Likewise.
770 * src/pa/ffi.c: Likewise.
771 * src/pa/hpux32.S: Likewise.
772 * src/ia64/unix.S: Likewise.
773 * src/ia64/ffi.c: Likewise.
774 * src/sparc/ffi.c: Likewise.
775 * src/mips/ffi.c: Likewise.
776 * src/sh/ffi.c: Likewise.
778 2008-02-15 David Daney <ddaney@avtrex.com>
780 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
781 Define (conditionally), and use it to include cachectl.h.
782 (ffi_prep_closure_loc): Fix cache flushing.
783 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
785 2009-06-04 Andrew Haley <aph@redhat.com>
791 src/powerpc/ffitarget.h,
793 src/sh64/ffitarget.h,
804 src/alpha/ffitarget.h,
807 src/m68k/ffitarget.h,
810 src/s390/ffitarget.h,
812 src/cris/ffitarget.h,
817 src/ia64/ffitarget.h,
820 src/ia64/ia64_flags.h,
824 src/sparc/ffitarget.h,
827 src/mips/ffitarget.h,
834 src/sh/sysv.S: Update license text.
836 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
838 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
839 (.eh_frame): Add FDE for it.
841 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
843 * configure.ac: Also check if assembler supports pc-relative
844 relocs on X86_WIN32 targets.
845 * configure: Regenerate.
846 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
847 (_ffi_call_SYSV): Add missing function type symbol .def and
848 add EH markup labels.
849 (_ffi_call_STDCALL): Likewise.
850 (_ffi_closure_SYSV): Likewise.
851 (_ffi_closure_raw_SYSV): Likewise.
852 (.eh_frame): Add hand-crafted EH data.
854 2009-04-09 Jakub Jelinek <jakub@redhat.com>
856 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
857 version 3 of the GNU General Public License and to point readers
858 at the COPYING3 file and the FSF's license web page.
859 * testsuite/libffi.call/call.exp: Likewise.
860 * testsuite/libffi.special/special.exp: Likewise.
862 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
864 * configure: Regenerate.
866 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
869 * configure.ac (HAVE_AS_X86_PCREL): New test.
870 * configure: Regenerate.
871 * fficonfig.h.in: Regenerate.
872 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
873 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
874 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
875 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
876 * src/x86/unix64.S (.Lstore_table): Move to .text section.
877 (.Lload_table): Likewise.
878 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
880 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
882 * configure: Regenerate.
884 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
886 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
887 signed/unsigned int8/16 return values.
888 * src/sparc/v8.S (ffi_call_v8): Likewise.
889 (ffi_closure_v8): Likewise.
891 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
892 Steve Ellcey <sje@cup.hp.com>
894 * configure: Regenerate for new libtool.
895 * Makefile.in: Ditto.
896 * include/Makefile.in: Ditto.
899 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
901 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
902 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
903 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
904 Adjust copyright notice.
905 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
906 register or two register to use for FFI_SYSV structs.
907 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
908 (ffi_closure_helper_SYSV): Fix the return type for
909 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
910 Adjust copyright notice.
912 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
914 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
917 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
919 * configure: Regenerate.
920 * include/Makefile.in: Regenerate.
921 * testsuite/Makefile.in: Regenerate.
923 2008-06-07 Joseph Myers <joseph@codesourcery.com>
925 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
926 powerpc-*-beos*): Remove.
927 * configure: Regenerate.
929 2008-05-09 Julian Brown <julian@codesourcery.com>
931 * Makefile.am (LTLDFLAGS): New.
932 (libffi_la_LDFLAGS): Use above.
933 * Makefile.in: Regenerate.
935 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
938 * aclocal.m4: Regenerate.
939 * configure: Regenerate.
941 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
943 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
944 * src/sh64/sysv.S: Likewise.
946 2008-03-26 Daniel Jacobowitz <dan@debian.org>
948 * src/arm/sysv.S: Fix ARM comment marker.
950 2008-03-26 Jakub Jelinek <jakub@redhat.com>
952 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
953 * src/s390/sysv.S: Likewise.
954 * src/powerpc/ppc_closure.S: Likewise.
955 * src/powerpc/sysv.S: Likewise.
956 * src/x86/unix64.S: Likewise.
957 * src/x86/sysv.S: Likewise.
958 * src/sparc/v8.S: Likewise.
959 * src/sparc/v9.S: Likewise.
960 * src/m68k/sysv.S: Likewise.
961 * src/arm/sysv.S: Likewise.
963 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
965 * aclocal.m4: Regenerate.
966 * configure: Likewise.
967 * Makefile.in: Likewise.
968 * include/Makefile.in: Likewise.
969 * testsuite/Makefile.in: Likewise.
971 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
972 Andreas Tobler <a.tobler@schweiz.org>
974 * configure.ac: Add amd64-*-freebsd* target.
975 * configure: Regenerate.
977 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
980 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
983 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
986 2008-01-24 David Edelsohn <edelsohn@gnu.org>
988 * configure: Regenerate.
990 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
992 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
994 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
997 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
998 signed/unsigned int8/16 for X86_DARWIN.
999 Updated copyright info.
1000 Handle one and two byte structs with special cif->flags.
1001 * src/x86/ffitarget.h: Add special types for one and two byte structs.
1002 Updated copyright info.
1003 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
1005 Remove code to pop args from the stack after call.
1006 Special-case signed/unsigned for int8/16, one and two byte structs.
1007 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1008 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1010 Updated copyright info.
1012 2007-12-08 David Daney <ddaney@avtrex.com>
1014 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
1015 SUBU, add with ADDU and use smaller code sequences.
1017 2007-12-07 David Daney <ddaney@avtrex.com>
1019 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
1022 2007-12-06 David Daney <ddaney@avtrex.com>
1024 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
1026 (ffi_java_raw): New typedef.
1027 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
1028 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
1030 (ffi_java_raw_closure) : Same.
1031 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
1033 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
1034 FFI_SIZEOF_JAVA_RAW.
1035 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
1036 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
1037 sizeof(ffi_java_raw) for alignment calculations.
1038 (ffi_java_ptrarray_to_raw): Same.
1039 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
1040 if FFI_SIZEOF_JAVA_RAW == 4.
1041 (ffi_java_raw_to_rvalue): Same.
1042 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
1043 (ffi_java_translate_args): Same.
1044 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
1046 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
1048 2007-12-06 David Daney <ddaney@avtrex.com>
1050 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
1053 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
1056 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
1057 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
1058 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
1059 set the NUM_FPR_ARG_REGISTERS according to.
1060 Add support for potential soft-float support under hard-float
1062 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
1063 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
1064 to the FFI_LINUX_SOFT_FLOAT ABI.
1065 (ffi_prep_cif_machdep): Likewise.
1066 (ffi_closure_helper_SYSV): Likewise.
1067 * src/powerpc/ppc_closure.S: Make sure not to store float/double
1068 on archs where __NO_FPRS__ is true.
1069 Add FFI_TYPE_UINT128 support.
1070 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
1071 Adjust copyright notice.
1073 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
1075 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
1076 * include/ffi_common.h: ... here.
1079 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
1081 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
1082 * src/powerpc/linux64.S: Likewise.
1083 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
1084 * testsuite/libffi.call/return_ldl.c: New test.
1086 2007-09-04 <aph@redhat.com>
1088 * src/arm/sysv.S (UNWIND): New.
1089 (Whole file): Conditionally compile unwinder directives.
1090 * src/arm/sysv.S: Add unwinder directives.
1092 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
1093 Only treat r0 as a struct address if we're actually returning a
1095 Only copy the bytes that are actually within a struct.
1096 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
1097 is returned in r0, not passed by address.
1098 (ffi_call): Allocate a word-sized temporary for the case where
1099 a composite is returned in r0.
1100 (ffi_prep_incoming_args_SYSV): Align as necessary.
1102 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
1104 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
1105 directly using the sys_cacheflush syscall.
1107 2007-07-27 Andrew Haley <aph@redhat.com>
1109 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
1111 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
1113 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
1114 * configure.ac: Likewise.
1115 * Makefile.in: Regenerate.
1116 * include/Makefile.in: Likewise.
1117 * testsuite/Makefile.in: Likewise.
1118 * configure: Likewise.
1120 2007-08-24 David Daney <ddaney@avtrex.com>
1122 * testsuite/libffi.call/return_sl.c: New test.
1124 2007-08-10 David Daney <ddaney@avtrex.com>
1126 * testsuite/libffi.call/cls_multi_ushort.c,
1127 testsuite/libffi.call/cls_align_uint16.c,
1128 testsuite/libffi.call/nested_struct1.c,
1129 testsuite/libffi.call/nested_struct3.c,
1130 testsuite/libffi.call/cls_7_1_byte.c,
1131 testsuite/libffi.call/nested_struct5.c,
1132 testsuite/libffi.call/cls_double.c,
1133 testsuite/libffi.call/nested_struct7.c,
1134 testsuite/libffi.call/cls_sint.c,
1135 testsuite/libffi.call/nested_struct9.c,
1136 testsuite/libffi.call/cls_20byte1.c,
1137 testsuite/libffi.call/cls_multi_sshortchar.c,
1138 testsuite/libffi.call/cls_align_sint64.c,
1139 testsuite/libffi.call/cls_3byte2.c,
1140 testsuite/libffi.call/cls_multi_schar.c,
1141 testsuite/libffi.call/cls_multi_uchar.c,
1142 testsuite/libffi.call/cls_19byte.c,
1143 testsuite/libffi.call/cls_9byte1.c,
1144 testsuite/libffi.call/cls_align_float.c,
1145 testsuite/libffi.call/closure_fn1.c,
1146 testsuite/libffi.call/problem1.c,
1147 testsuite/libffi.call/closure_fn3.c,
1148 testsuite/libffi.call/cls_sshort.c,
1149 testsuite/libffi.call/closure_fn5.c,
1150 testsuite/libffi.call/cls_align_double.c,
1151 testsuite/libffi.call/nested_struct.c,
1152 testsuite/libffi.call/cls_2byte.c,
1153 testsuite/libffi.call/nested_struct10.c,
1154 testsuite/libffi.call/cls_4byte.c,
1155 testsuite/libffi.call/cls_6byte.c,
1156 testsuite/libffi.call/cls_8byte.c,
1157 testsuite/libffi.call/cls_multi_sshort.c,
1158 testsuite/libffi.call/cls_align_sint16.c,
1159 testsuite/libffi.call/cls_align_uint32.c,
1160 testsuite/libffi.call/cls_20byte.c,
1161 testsuite/libffi.call/cls_float.c,
1162 testsuite/libffi.call/nested_struct2.c,
1163 testsuite/libffi.call/cls_5_1_byte.c,
1164 testsuite/libffi.call/nested_struct4.c,
1165 testsuite/libffi.call/cls_24byte.c,
1166 testsuite/libffi.call/nested_struct6.c,
1167 testsuite/libffi.call/cls_64byte.c,
1168 testsuite/libffi.call/nested_struct8.c,
1169 testsuite/libffi.call/cls_uint.c,
1170 testsuite/libffi.call/cls_multi_ushortchar.c,
1171 testsuite/libffi.call/cls_schar.c,
1172 testsuite/libffi.call/cls_uchar.c,
1173 testsuite/libffi.call/cls_align_uint64.c,
1174 testsuite/libffi.call/cls_ulonglong.c,
1175 testsuite/libffi.call/cls_align_longdouble.c,
1176 testsuite/libffi.call/cls_1_1byte.c,
1177 testsuite/libffi.call/cls_12byte.c,
1178 testsuite/libffi.call/cls_3_1byte.c,
1179 testsuite/libffi.call/cls_3byte1.c,
1180 testsuite/libffi.call/cls_4_1byte.c,
1181 testsuite/libffi.call/cls_6_1_byte.c,
1182 testsuite/libffi.call/cls_16byte.c,
1183 testsuite/libffi.call/cls_18byte.c,
1184 testsuite/libffi.call/closure_fn0.c,
1185 testsuite/libffi.call/cls_9byte2.c,
1186 testsuite/libffi.call/closure_fn2.c,
1187 testsuite/libffi.call/closure_fn4.c,
1188 testsuite/libffi.call/cls_ushort.c,
1189 testsuite/libffi.call/closure_fn6.c,
1190 testsuite/libffi.call/cls_5byte.c,
1191 testsuite/libffi.call/cls_align_pointer.c,
1192 testsuite/libffi.call/cls_7byte.c,
1193 testsuite/libffi.call/cls_align_sint32.c,
1194 testsuite/libffi.special/unwindtest_ffi_call.cc,
1195 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
1197 2007-08-10 David Daney <ddaney@avtrex.com>
1200 * configure.ac: Don't treat mips64 as a special case.
1201 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
1202 * configure: Regenerate
1203 * Makefile.in: Ditto.
1204 * fficonfig.h.in: Ditto.
1205 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
1206 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
1207 (FFI_DEFAULT_ABI): Set for n64 case.
1208 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
1209 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
1210 (ffi_closure_N32): New function.
1211 (.eh_frame): New section
1212 * src/mips/o32.S: Clean up comments.
1213 (ffi_closure_O32): Pass ffi_closure parameter in $12.
1214 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
1215 _MIPS_SIM == _ABIN32 throughout.
1216 (FFI_MIPS_STOP_HERE): New, use in place of
1218 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1219 to support n32/n64 ABIs.
1220 (calc_n32_struct_flags): Rewrite.
1221 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1222 position of flag bits.
1223 (ffi_prep_cif_machdep): Rewrite n32 portion.
1224 (ffi_call): Enable for n64. Add special handling for small structure
1226 (ffi_prep_closure_loc): Add n32 and n64 support.
1227 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1228 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1230 2007-08-08 David Daney <ddaney@avtrex.com>
1232 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1233 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1235 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1236 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1237 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1238 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1239 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1240 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1241 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1242 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1243 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1244 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1245 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1246 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1247 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1248 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1249 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1250 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1251 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1252 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1253 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1254 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1255 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1256 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1257 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1258 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1259 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1260 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1261 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1262 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1263 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1264 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1265 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1267 2007-08-07 Andrew Haley <aph@redhat.com>
1269 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1272 2007-08-06 Andrew Haley <aph@redhat.com>
1275 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1276 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1279 2007-08-02 David Daney <ddaney@avtrex.com>
1281 * testsuite/libffi.call/return_ul.c (main): Define return type as
1282 ffi_arg. Use proper printf conversion specifier.
1284 2007-07-30 Andrew Haley <aph@redhat.com>
1287 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1288 signed/unsigned int8/16.
1289 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1291 Remove code to pop args from the stack after call.
1292 Special-case signed/unsigned int8/16.
1293 * testsuite/libffi.call/return_sc.c (main): Revert.
1295 2007-07-26 Richard Guenther <rguenther@suse.de>
1298 * testsuite/libffi.call/return_sc.c (main): Verify call
1299 result as signed char, not ffi_arg.
1301 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1303 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1304 * configure: Regenerate.
1306 2007-07-11 David Daney <ddaney@avtrex.com>
1308 * src/mips/ffi.c: Don't include sys/cachectl.h.
1309 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1312 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1314 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1315 from (ffi_prep_closure): ... this.
1316 (FFI_INIT_TRAMPOLINE): Adjust.
1318 2005-12-31 Phil Blundell <pb@reciva.com>
1320 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1321 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1322 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1323 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1324 (FFI_CLOSURES): Enable closure support.
1326 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1328 * testsuite/libffi.call/cls_multi_ushort.c,
1329 testsuite/libffi.call/cls_align_uint16.c,
1330 testsuite/libffi.call/nested_struct1.c,
1331 testsuite/libffi.call/nested_struct3.c,
1332 testsuite/libffi.call/cls_7_1_byte.c,
1333 testsuite/libffi.call/cls_double.c,
1334 testsuite/libffi.call/nested_struct5.c,
1335 testsuite/libffi.call/nested_struct7.c,
1336 testsuite/libffi.call/cls_sint.c,
1337 testsuite/libffi.call/nested_struct9.c,
1338 testsuite/libffi.call/cls_20byte1.c,
1339 testsuite/libffi.call/cls_multi_sshortchar.c,
1340 testsuite/libffi.call/cls_align_sint64.c,
1341 testsuite/libffi.call/cls_3byte2.c,
1342 testsuite/libffi.call/cls_multi_schar.c,
1343 testsuite/libffi.call/cls_multi_uchar.c,
1344 testsuite/libffi.call/cls_19byte.c,
1345 testsuite/libffi.call/cls_9byte1.c,
1346 testsuite/libffi.call/cls_align_float.c,
1347 testsuite/libffi.call/closure_fn1.c,
1348 testsuite/libffi.call/problem1.c,
1349 testsuite/libffi.call/closure_fn3.c,
1350 testsuite/libffi.call/cls_sshort.c,
1351 testsuite/libffi.call/closure_fn5.c,
1352 testsuite/libffi.call/cls_align_double.c,
1353 testsuite/libffi.call/cls_2byte.c,
1354 testsuite/libffi.call/nested_struct.c,
1355 testsuite/libffi.call/nested_struct10.c,
1356 testsuite/libffi.call/cls_4byte.c,
1357 testsuite/libffi.call/cls_6byte.c,
1358 testsuite/libffi.call/cls_8byte.c,
1359 testsuite/libffi.call/cls_multi_sshort.c,
1360 testsuite/libffi.call/cls_align_uint32.c,
1361 testsuite/libffi.call/cls_align_sint16.c,
1362 testsuite/libffi.call/cls_float.c,
1363 testsuite/libffi.call/cls_20byte.c,
1364 testsuite/libffi.call/cls_5_1_byte.c,
1365 testsuite/libffi.call/nested_struct2.c,
1366 testsuite/libffi.call/cls_24byte.c,
1367 testsuite/libffi.call/nested_struct4.c,
1368 testsuite/libffi.call/nested_struct6.c,
1369 testsuite/libffi.call/cls_64byte.c,
1370 testsuite/libffi.call/nested_struct8.c,
1371 testsuite/libffi.call/cls_uint.c,
1372 testsuite/libffi.call/cls_multi_ushortchar.c,
1373 testsuite/libffi.call/cls_schar.c,
1374 testsuite/libffi.call/cls_uchar.c,
1375 testsuite/libffi.call/cls_align_uint64.c,
1376 testsuite/libffi.call/cls_ulonglong.c,
1377 testsuite/libffi.call/cls_align_longdouble.c,
1378 testsuite/libffi.call/cls_1_1byte.c,
1379 testsuite/libffi.call/cls_12byte.c,
1380 testsuite/libffi.call/cls_3_1byte.c,
1381 testsuite/libffi.call/cls_3byte1.c,
1382 testsuite/libffi.call/cls_4_1byte.c,
1383 testsuite/libffi.call/cls_6_1_byte.c,
1384 testsuite/libffi.call/cls_16byte.c,
1385 testsuite/libffi.call/cls_18byte.c,
1386 testsuite/libffi.call/closure_fn0.c,
1387 testsuite/libffi.call/cls_9byte2.c,
1388 testsuite/libffi.call/closure_fn2.c,
1389 testsuite/libffi.call/closure_fn4.c,
1390 testsuite/libffi.call/cls_ushort.c,
1391 testsuite/libffi.call/closure_fn6.c,
1392 testsuite/libffi.call/cls_5byte.c,
1393 testsuite/libffi.call/cls_align_pointer.c,
1394 testsuite/libffi.call/cls_7byte.c,
1395 testsuite/libffi.call/cls_align_sint32.c,
1396 testsuite/libffi.special/unwindtest_ffi_call.cc,
1397 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1399 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1401 * aclocal.m4: Regenerated.
1403 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1405 * configure: Regenerate.
1407 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1409 * Makefile.in: Regenerate.
1410 * configure: Regenerate.
1411 * aclocal.m4: Regenerate.
1412 * include/Makefile.in: Regenerate.
1413 * testsuite/Makefile.in: Regenerate.
1415 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1417 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1418 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1419 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1420 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1421 (FFI_CLOSURES): Enable closure support.
1423 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1425 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1426 * configure: Regenerate.
1427 * fficonfig.h.in: Regenerate.
1428 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1429 CFI_OFFSET,CFI_DEF_CFA): New macros.
1430 (ffi_call_SYSV): Add callframe annotation.
1432 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1434 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1435 numerous test suite failures.
1436 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1438 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1440 * Makefile.am (EXTRA_DIST): Bring up to date.
1441 * Makefile.in: Regenerate.
1442 * src/frv/eabi.S: Remove RCS keyword.
1444 2007-04-06 Richard Henderson <rth@redhat.com>
1446 * configure.ac: Tidy target case.
1447 (HAVE_LONG_DOUBLE): Allow the target to override.
1448 * configure: Regenerate.
1449 * include/ffi.h.in: Don't define ffi_type_foo if
1450 LIBFFI_HIDE_BASIC_TYPES is defined.
1451 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1453 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1454 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1455 (ffi_type_longdouble): Special case for Alpha. Don't define
1456 if long double == double.
1458 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1459 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1460 (ffi_call, ffi_closure_osf_inner): Likewise.
1461 (ffi_closure_osf_inner): Likewise. Mark hidden.
1462 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1463 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1464 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1465 (load_table): Handle 128-bit long double.
1467 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1469 2007-04-06 Tom Tromey <tromey@redhat.com>
1472 * README: Fixed bug in example.
1474 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1476 * src/closures.c: Include sys/statfs.h.
1477 (_GNU_SOURCE): Define on Linux.
1478 (FFI_MMAP_EXEC_SELINUX): Define.
1479 (selinux_enabled): New variable.
1480 (selinux_enabled_check): New function.
1481 (is_selinux_enabled): Define.
1484 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1486 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1487 Use 'volatile float sum' to create sum of floats to avoid false
1488 negative due to excess precision on ix86 targets.
1491 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1493 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1495 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1497 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1499 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1500 (ffi_prep_closure_loc): New.
1501 (ffi_prep_raw_closure_loc): New.
1502 (ffi_prep_java_raw_closure_loc): New.
1503 * src/closures.c: New file.
1504 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1505 Replace sflags with exec_offset.
1506 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1507 sub_segment_exec_offset): New macros.
1508 (get_segment_flags, set_segment_flags, check_segment_merge): New
1510 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1511 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1512 destroy_mspace): Use new macros.
1513 (sys_alloc): Silence warning.
1514 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1515 * Makefile.in: Rebuilt.
1516 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1517 terms of ffi_prep_closure_loc.
1518 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1520 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1522 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1524 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1525 the renamed version.
1526 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1527 (ffi_prep_closure): ... this.
1528 * src/pa/ffi.c: Likewise.
1529 * src/cris/ffi.c: Likewise. Adjust.
1530 * src/frv/ffi.c: Likewise.
1531 * src/ia64/ffi.c: Likewise.
1532 * src/mips/ffi.c: Likewise.
1533 * src/powerpc/ffi_darwin.c: Likewise.
1534 * src/s390/ffi.c: Likewise.
1535 * src/sh/ffi.c: Likewise.
1536 * src/sh64/ffi.c: Likewise.
1537 * src/sparc/ffi.c: Likewise.
1538 * src/x86/ffi64.c: Likewise.
1539 * src/x86/ffi.c: Likewise.
1540 (FFI_INIT_TRAMPOLINE): Adjust.
1541 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1542 (ffi_prep_raw_closure): ... this.
1543 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1544 (ffi_prep_closure): ... this.
1545 (flush_icache): Adjust.
1547 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1549 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1552 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1554 * Makefile.am: Add dummy install-pdf target.
1555 * Makefile.in: Regenerate
1557 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1559 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1560 ffi_closure_helper_SYSV): Add long double handling.
1562 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1564 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1565 immediately after bctrl instruction.
1567 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1569 * Makefile.am (all-recursive, install-recursive,
1570 mostlyclean-recursive, clean-recursive, distclean-recursive,
1571 maintainer-clean-recursive): Add missing targets.
1572 * Makefile.in: Rebuilt.
1574 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1576 * configure.ac: Add TARGET for x86_64-*-darwin*.
1577 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1579 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1580 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1581 * configure: Regenerate.
1582 * Makefile.in: Regenerate.
1583 * include/Makefile.in: Regenerate.
1584 * testsuite/Makefile.in: Regenerate.
1585 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1588 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1590 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1593 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1595 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1596 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1598 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1599 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1601 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1602 Sandro Tolaini <tolaini@libero.it>
1604 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1606 * configure: Regenerated.
1607 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1608 (EXTRA_DIST): Add src/x86/darwin.S.
1609 * Makefile.in: Regenerated.
1610 * include/Makefile.in: Regenerated.
1611 * testsuite/Makefile.in: Regenerated.
1613 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1614 X86_WIN32, and additionally align stack to 16 bytes.
1615 * src/x86/darwin.S: New, based on sysv.S.
1616 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1618 2006-09-12 David Daney <ddaney@avtrex.com>
1621 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1622 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1623 * aclocal.m4: Regenerated for automake 1.9.6.
1624 * Makefile.in: Regenerated.
1625 * include/Makefile.in: Regenerated.
1626 * testsuite/Makefile.in: Regenerated.
1628 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1630 * include/ffi_common.h (struct): Revert accidental commit.
1632 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1634 * include/ffi_common.h: Remove lint directives.
1635 * include/ffi.h.in: Likewise.
1637 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1639 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1640 for 32-bit architectures.
1641 * testsuite/libffi.call/return_ul.c: New test case.
1643 2006-07-19 David Daney <ddaney@avtrex.com>
1645 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1646 xfail remains for mips64.
1648 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1650 * Makefile.am: Add install-html target. Add install-html to .PHONY
1651 * Makefile.in: Regenerate.
1652 * aclocal.m4: Regenerate.
1653 * include/Makefile.in: Regenerate.
1654 * testsuite/Makefile.in: Regenerate.
1656 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1658 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1661 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1663 * README: Remove notice about 'Crazy Comments'.
1664 * src/debug.c: Remove lint directives. Cleanup white spaces.
1665 * src/java_raw_api.c: Likewise.
1666 * src/prep_cif.c: Likewise.
1667 * src/raw_api.c: Likewise.
1668 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1670 * src/arm/ffi.c: Remove lint directives.
1671 * src/m32r/ffi.c: Likewise.
1672 * src/pa/ffi.c: Likewise.
1673 * src/powerpc/ffi.c: Likewise.
1674 * src/powerpc/ffi_darwin.c: Likewise.
1675 * src/sh/ffi.c: Likewise.
1676 * src/sh64/ffi.c: Likewise.
1677 * src/x86/ffi.c: Likewise.
1678 * testsuite/libffi.call/float2.c: Likewise.
1679 * testsuite/libffi.call/promotion.c: Likewise.
1680 * testsuite/libffi.call/struct1.c: Likewise.
1682 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1684 * src/pa/hpux32.S: Correct unwind offset calculation for
1686 * src/pa/linux.S: Likewise.
1688 2006-04-12 James E Wilson <wilson@specifix.com>
1691 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1692 (hfa_type_load): Call stf_spill.
1693 (hfa_type_store): Call ldf_fill.
1694 (ffi_call): Adjust calls to above routines. Add local temps for
1697 2006-04-10 Matthias Klose <doko@debian.org>
1699 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1700 directory names containing underscores.
1702 2006-04-07 James E Wilson <wilson@specifix.com>
1704 * testsuite/libffi.call/float4.c: New testcase.
1706 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1707 Andreas Tobler <a.tobler@schweiz.ch>
1709 * Makefile.am: Add PA_HPUX port.
1710 * Makefile.in: Regenerate.
1711 * include/Makefile.in: Likewise.
1712 * testsuite/Makefile.in: Likewise.
1713 * configure.ac: Add PA_HPUX rules.
1714 * configure: Regenerate.
1715 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1716 Add PA_HPUX and PA64_HPUX.
1717 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1718 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1719 (FFI_TYPE_SMALL_STRUCT2): Define.
1720 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1721 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1722 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1723 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1724 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1725 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1726 * src/pa/ffi.c (ROUND_DOWN): Delete.
1727 (fldw, fstw, fldd, fstd): Use '__asm__'.
1728 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1729 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1730 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1731 Simplify incrementing of stack slot variable. Change type of local
1732 'n' to unsigned int.
1733 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1735 (ffi_prep_cif_machdep): Likewise.
1736 (ffi_call): Likewise.
1737 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1738 return type to ffi_status. Simplify incrementing of stack slot
1739 variable. Only copy floating point argument registers when PA_LINUX
1740 is true. Reformat debug statement.
1741 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1742 FFI_TYPE_SMALL_STRUCT8.
1743 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1745 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1746 Add nops to cache flush. Add trampoline for PA_HPUX.
1747 * src/pa/hpux32.S: New file.
1748 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1749 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1750 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1751 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1752 argument type checks so that common argument types appear first.
1753 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1754 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1756 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1758 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1759 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1760 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1761 FFI_TYPE_LONGDOUBLE.
1762 (ffi_prep_args64): Assert using IBM extended double.
1763 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1764 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1765 (ffi_call): Handle FFI_LINUX.
1766 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1767 gpr3 return pointer as for struct return. Handle FFI_LINUX
1768 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1770 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1771 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1772 Don't use r6 as pointer to results, instead use sp offset. Don't
1773 make a special call to load lr with case table address, instead
1774 use offset from previous call.
1775 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1776 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1779 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1781 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1782 passed with FP registers correctly.
1783 (ffi_closure_helper_SYSV): Likewise.
1784 * src/sh64/sysv.S: Likewise.
1786 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1788 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1789 args and userdata unused.
1790 (closure_test_fn1): Mark cif and userdata unused.
1791 (main): Remove unused res.
1793 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1795 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1796 -O2, -O3, -Os and the warning flags -W -Wall.
1797 * testsuite/libffi.special/special.exp: Likewise.
1798 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1799 unused parameter unused for gcc or else do nothing.
1800 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1801 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1802 and userdata unused.
1803 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1804 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1805 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1806 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1807 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1808 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1809 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1810 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1811 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1812 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1813 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1814 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1815 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1816 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1817 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1818 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1819 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1820 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1821 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1822 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1823 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1824 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1825 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1826 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1828 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1830 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1832 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1833 void* to avoid compiler warning.
1835 (cls_struct_align_gn): Mark cif and userdata unused.
1836 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1838 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1840 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1842 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1844 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1846 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1847 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1848 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1850 (main): Cast res_call to silence gcc.
1851 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1853 (main): Cast res_call to silence gcc.
1854 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1856 (main): Cast res_call to silence gcc.
1857 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1859 (main): Cast res_call to silence gcc.
1860 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1862 (main): Cast res_call to silence gcc.
1863 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1865 (main): Cast res_call to silence gcc.
1866 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1868 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1869 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1871 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1872 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1874 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1875 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1877 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1878 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1880 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1881 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1882 and userdata unused.
1883 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1885 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1886 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1887 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1888 Cleanup white spaces.
1889 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1890 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1891 cif and userdata unused.
1892 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1894 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1895 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1896 formatters to silence gcc.
1897 (B_gn): Mark cif and userdata unused.
1898 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1900 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1901 (B_gn): Mark cif and userdata unused.
1902 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1904 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1905 (B_gn): Mark cif and userdata unused.
1906 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1908 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1909 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1910 * testsuite/libffi.call/problem1.c (stub): Likewise.
1911 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1913 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1914 in the last commit for this test case in the test case itself.
1915 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1917 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1918 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1919 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1920 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1921 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1922 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1924 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1926 * src/sh/sysv.S: Fix register numbers in the FDE for
1929 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1931 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1932 declaration to avoid a false negative on ix86. See PR323.
1934 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1936 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1937 and cast integer to void * if needed. Update the pointer to
1938 the FP register saved area correctly.
1940 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1942 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1944 * testsuite/libffi.call/nested_struct4.c: Likewise.
1946 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1948 * testsuite/libffi.call/return_dbl.c: New test case.
1949 * testsuite/libffi.call/return_dbl1.c: Likewise.
1950 * testsuite/libffi.call/return_dbl2.c: Likewise.
1951 * testsuite/libffi.call/return_fl.c: Likewise.
1952 * testsuite/libffi.call/return_fl1.c: Likewise.
1953 * testsuite/libffi.call/return_fl2.c: Likewise.
1954 * testsuite/libffi.call/return_fl3.c: Likewise.
1955 * testsuite/libffi.call/closure_fn6.c: Likewise.
1957 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1959 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1960 here to be used by other test cases too.
1962 * testsuite/libffi.call/nested_struct10.c: New test case.
1963 * testsuite/libffi.call/nested_struct9.c: Likewise.
1964 * testsuite/libffi.call/nested_struct8.c: Likewise.
1965 * testsuite/libffi.call/nested_struct7.c: Likewise.
1966 * testsuite/libffi.call/nested_struct6.c: Likewise.
1967 * testsuite/libffi.call/nested_struct5.c: Likewise.
1968 * testsuite/libffi.call/nested_struct4.c: Likewise.
1970 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1972 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1973 * configure: Rebuilt.
1975 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1977 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1978 instead do the shifting inline.
1979 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1980 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1981 and 8 byte structs, for the remaining struct sizes don't call
1982 __lshrdi3, instead do the shifting inline.
1984 2005-12-07 Thiemo Seufer <ths@networkno.de>
1986 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1987 missing parentheses.
1988 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1989 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1990 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1991 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1992 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1994 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1996 (ffi_prep_closure): Improve trampoline instruction scheduling.
1997 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1999 2005-12-03 Alan Modra <amodra@bigpond.net.au>
2001 * src/powerpc/ffi.c: Formatting.
2002 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
2003 (ffi_prep_args64): Likewise.
2005 2005-09-30 Geoffrey Keating <geoffk@apple.com>
2007 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
2008 darwin, use -shared-libgcc not -lgcc_s, and explain why.
2010 2005-09-26 Tom Tromey <tromey@redhat.com>
2012 * testsuite/libffi.call/float1.c (value_type): New typedef.
2013 (CANARY): New define.
2014 (main): Check for result buffer overflow.
2015 * src/powerpc/linux64.S: Handle linux64 long double returns.
2016 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
2017 (ffi_prep_cif_machdep): Handle linux64 long double returns.
2019 2005-08-25 Alan Modra <amodra@bigpond.net.au>
2022 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
2024 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
2026 2005-08-11 Jakub Jelinek <jakub@redhat.com>
2028 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
2029 (AH_BOTTOM): Add FFI_HIDDEN definition.
2030 * configure: Rebuilt.
2031 * fficonfig.h.in: Rebuilt.
2032 * src/powerpc/ffi.c (hidden): Remove.
2033 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
2034 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
2035 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
2036 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
2037 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
2038 add FFI_HIDDEN to its prototype.
2039 (ffi_closure_SYSV_inner): New.
2040 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2041 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2043 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
2046 * configure: Rebuilt.
2047 * configure.ac: Handle i*86-*-gnu*.
2049 2005-08-09 Jakub Jelinek <jakub@redhat.com>
2051 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
2052 DW_CFA_offset_extended_sf rather than
2053 DW_CFA_GNU_negative_offset_extended.
2054 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
2056 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
2058 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
2060 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
2061 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
2062 partially on register.
2063 (ffi_closure_helper_SYSV): Likewise.
2064 (ffi_prep_cif_machdep): Don't set too many cif->flags.
2066 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
2068 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
2070 * src/sh64/ffi.c (simple_type): Remove.
2071 (return_type): Handle small structures correctly.
2072 (ffi_prep_args): Likewise.
2073 (ffi_call): Likewise.
2074 (ffi_closure_helper_SYSV): Likewise.
2075 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
2076 Emit position independent code if PIC and remove wrong datalabel
2077 prefixes from EH data.
2079 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
2081 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
2082 * Makefile.in: Regenerate.
2083 * include/Makefile.in: Likewise.
2084 * testsuite/Makefile.in: Likewise.
2085 * configure.ac: Add POWERPC_FREEBSD rules.
2086 * configure: Regenerate.
2087 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
2088 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
2089 * src/powerpc/ffi.c: Add flags to handle small structure returns
2091 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
2093 (ffi_closure_helper_SYSV): Likewise.
2094 * src/powerpc/ppc_closure.S: Add return types for small structures.
2095 * src/powerpc/sysv.S: Add bits to handle small structures for
2098 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
2100 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
2101 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
2102 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
2104 2005-07-05 Randolph Chung <tausq@debian.org>
2106 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
2107 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
2108 structures. Kill compilation warnings.
2109 (ffi_closure_inner_LINUX): Print return values as hex in debug
2110 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
2111 Properly handle 5-7 byte structure returns.
2112 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
2113 (FFI_TYPE_SMALL_STRUCT2): Remove.
2114 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
2115 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
2116 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
2117 (checksmst1, checksmst2): Remove.
2118 (checksmst3): Optimize handling of 3-byte struct returns.
2119 (checksmst567): Properly handle 5-7 byte struct returns.
2121 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2124 * src/mips/n32.S: Enforce PIC code.
2125 * src/mips/o32.S: Likewise.
2127 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2129 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
2130 * configure: Regenerate.
2132 2005-06-01 Alan Modra <amodra@bigpond.net.au>
2134 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
2135 to call ffi_closure_helper_SYSV. Append @local instead.
2136 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
2138 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
2140 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
2141 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
2142 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
2143 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
2144 include/Makefile.in, testsuite/Makefile.in: Regenerate.
2146 2005-05-09 Mike Stump <mrs@apple.com>
2148 * configure: Regenerate.
2150 2005-05-08 Richard Henderson <rth@redhat.com>
2153 * src/alpha/osf.S: Update unwind into to match code.
2155 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
2156 Richard Henderson <rth@redhat.com>
2158 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
2160 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
2161 (ffi_prep_closure): Set carry bit if sse-used flag set.
2162 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
2163 Only load sse registers if ssecount non-zero.
2164 (ffi_closure_unix64): Only save sse registers if carry set on entry.
2166 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
2168 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
2169 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
2170 * configure: Regenerate.
2172 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
2174 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
2175 have Tcl8.3-compatible intermediate variable.
2177 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
2178 Hans-Peter Nilsson <hp@axis.com>
2180 * Makefile.am: Add CRIS support.
2181 * configure.ac: Likewise.
2182 * Makefile.in, configure, testsuite/Makefile.in,
2183 include/Makefile.in: Regenerate.
2184 * src/cris: New directory.
2185 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
2186 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
2188 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
2189 \r?\n in output tests.
2191 2005-04-12 Mike Stump <mrs@apple.com>
2193 * configure: Regenerate.
2195 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
2197 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
2199 2005-03-30 Steve Ellcey <sje@cup.hp.com>
2201 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
2203 * src/ia64/unix.S (ffi_closure_unix): Extend gp
2204 to 64 bits in ILP32 mode.
2205 Load 64 bits even for short data.
2207 2005-03-23 Mike Stump <mrs@apple.com>
2209 * src/powerpc/darwin.S: Update for -m64 multilib.
2210 * src/powerpc/darwin_closure.S: Likewise.
2212 2005-03-21 Zack Weinberg <zack@codesourcery.com>
2214 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
2215 Do not set tool_include_dir.
2216 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
2218 * include/Makefile.am: Set gcc_version and toollibffidir.
2219 * include/Makefile.in: Regenerate.
2221 2005-02-22 Andrew Haley <aph@redhat.com>
2223 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2224 odd-numbered register pairs for 64-bit integer types.
2226 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2229 * testsuite/libffi.call/return_ll1.c: New test case.
2231 2005-02-11 Janis Johnson <janis187@us.ibm.com>
2233 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2234 * testsuite/libffi.call/float.c: Ditto.
2235 * testsuite/libffi.call/float2.c: Ditto.
2236 * testsuite/libffi.call/float3.c: Ditto.
2238 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2240 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2242 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2244 * testsuite/libffi.special/special.exp (cxx_options): Add
2247 2004-12-31 Richard Henderson <rth@redhat.com>
2249 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2250 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2251 offset parameters with a type parameter; deduce size and structure
2252 alignment. Update all users.
2254 2004-12-31 Richard Henderson <rth@redhat.com>
2256 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2257 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
2258 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
2259 into ffi_prep_closure.
2260 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2263 2004-12-27 Richard Henderson <rth@redhat.com>
2265 * src/x86/unix64.S: Fix typo in unwind info.
2267 2004-12-25 Richard Henderson <rth@redhat.com>
2269 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2270 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2271 (merge_classes): Check for it.
2273 (classify_argument): Pass byte_offset by value; perform all updates
2275 (examine_argument): Add classes argument; handle
2276 X86_64_COMPLEX_X87_CLASS.
2277 (ffi_prep_args): Merge into ...
2278 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2279 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2280 (ffi_fill_return_value): Remove.
2281 (ffi_prep_closure): Remove dead assert.
2282 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2283 Rewrite to use struct register_args instead of va_list. Create
2284 flags for handling structure returns.
2285 * src/x86/unix64.S: Remove dead strings.
2286 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2287 stack frame with ffi_call. Handle structure returns properly.
2288 (float2sse, floatfloat2sse, double2sse): Remove.
2289 (sse2float, sse2double, sse2floatfloat): Remove.
2290 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2291 to handle structure returns properly.
2293 2004-12-08 David Edelsohn <edelsohn@gnu.org>
2295 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2297 * Makefile.in: Regenerated.
2299 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2301 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2302 * configure, aclocal.m4, Makefile.in: Regenerate.
2303 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2305 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2307 * configure: Regenerate for libtool change.
2309 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2311 * configure: Regenerate for libtool reversion.
2313 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2315 * configure: Regenerate for libtool change.
2317 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2319 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2321 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2323 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2324 of jal. Use an absolute encoding for the frame information.
2326 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2328 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2329 * acinclude.m4: Delete logic for sincludes.
2330 * aclocal.m4, Makefile.in, configure: Regenerate.
2331 * include/Makefile: Likewise.
2332 * testsuite/Makefile: Likewise.
2334 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2336 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2337 on a 8-byte boundary.
2338 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2340 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2342 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2343 long long values. Round stack allocation to a multiple of 8 bytes
2344 for ATPCS compatibility.
2345 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2346 names. Handle returning long long types. Add Thumb and interworking
2347 support. Improve soft-float code.
2349 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2351 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2352 (libffi_exit): New function.
2353 (libffi_init): Build the testglue wrapper if needed.
2355 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2358 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2360 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2362 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2364 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2366 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2367 * testsuite/libffi.call/float3.c: New test case.
2369 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2371 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2372 the function returning a structure pointed with R2.
2373 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2374 the structure return value if T bit set. Emit position
2375 independent code and EH data if PIC.
2377 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2379 * Makefile.am: Add m32r support.
2380 * configure.ac: Likewise.
2381 * Makefile.in: Regenerate.
2382 * confiugre: Regenerate.
2383 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2384 (uint64, sint64, double, longdouble)
2385 * src/m32r: New directory.
2386 * src/m32r/ffi.c: New file.
2387 * src/m32r/sysv.S: Likewise.
2388 * src/m32r/ffitarget.h: Likewise.
2390 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2392 * testsuite/libffi.call/negint.c: New test case.
2394 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2397 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2398 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2399 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2402 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2404 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2405 * testsuite/libffi.call/promotion.c: Likewise.
2406 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2408 * testsuite/libffi.call/return_sc.c: Likewise.
2409 * testsuite/libffi.call/return_uc.c: Likewise.
2411 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2413 * src/powerpc/darwin.S: Fix comments and identation.
2414 * src/powerpc/darwin_closure.S: Likewise.
2416 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2418 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2419 (ffi_prep_args): Handle longdouble arguments.
2420 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2422 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2423 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2425 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2426 * src/types.c: Defined longdouble size and alignment for darwin.
2428 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2430 * src/powerpc/aix.S: Remove whitespaces.
2431 * src/powerpc/aix_closure.S: Likewise.
2432 * src/powerpc/asm.h: Likewise.
2433 * src/powerpc/ffi.c: Likewise.
2434 * src/powerpc/ffitarget.h: Likewise.
2435 * src/powerpc/linux64.S: Likewise.
2436 * src/powerpc/linux64_closure.S: Likewise.
2437 * src/powerpc/ppc_closure.S: Likewise.
2438 * src/powerpc/sysv.S: Likewise.
2440 2004-08-30 Anthony Green <green@redhat.com>
2442 * Makefile.am: Add frv support.
2443 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2444 * configure.ac: Read configure.host.
2445 * configure.in: Read configure.host.
2446 * configure.host: New file. frv-elf needs libgloss.
2447 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2448 alignment. This is needed to frv and shouldn't harm the others.
2449 * include/ffi_common.h (ALIGN_DOWN): New macro.
2450 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2452 2004-08-24 David Daney <daney@avtrex.com>
2454 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2455 * testsuite/libffi.call/closure_fn1.c: Likewise.
2456 * testsuite/libffi.call/closure_fn2.c Likewise.
2457 * testsuite/libffi.call/closure_fn3.c: Likewise.
2458 * testsuite/libffi.call/closure_fn4.c: Likewise.
2459 * testsuite/libffi.call/closure_fn5.c: Likewise.
2460 * testsuite/libffi.call/cls_18byte.c: Likewise.
2461 * testsuite/libffi.call/cls_19byte.c: Likewise.
2462 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2463 * testsuite/libffi.call/cls_20byte.c: Likewise.
2464 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2465 * testsuite/libffi.call/cls_24byte.c: Likewise.
2466 * testsuite/libffi.call/cls_2byte.c: Likewise.
2467 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2468 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2469 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2470 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2471 * testsuite/libffi.call/cls_4byte.c: Likewise.
2472 * testsuite/libffi.call/cls_64byte.c: Likewise.
2473 * testsuite/libffi.call/cls_6byte.c: Likewise.
2474 * testsuite/libffi.call/cls_7byte.c: Likewise.
2475 * testsuite/libffi.call/cls_8byte.c: Likewise.
2476 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2477 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2478 * testsuite/libffi.call/cls_align_double.c: Likewise.
2479 * testsuite/libffi.call/cls_align_float.c: Likewise.
2480 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2481 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2482 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2483 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2484 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2485 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2486 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2487 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2488 * testsuite/libffi.call/cls_double.c: Likewise.
2489 * testsuite/libffi.call/cls_float.c: Likewise.
2490 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2491 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2492 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2493 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2494 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2495 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2496 * testsuite/libffi.call/cls_schar.c: Likewise.
2497 * testsuite/libffi.call/cls_sint.c: Likewise.
2498 * testsuite/libffi.call/cls_sshort.c: Likewise.
2499 * testsuite/libffi.call/cls_uchar.c: Likewise.
2500 * testsuite/libffi.call/cls_uint.c: Likewise.
2501 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2502 * testsuite/libffi.call/cls_ushort.c: Likewise.
2503 * testsuite/libffi.call/nested_struct.c: Likewise.
2504 * testsuite/libffi.call/nested_struct1.c: Likewise.
2505 * testsuite/libffi.call/nested_struct2.c: Likewise.
2506 * testsuite/libffi.call/nested_struct3.c: Likewise.
2507 * testsuite/libffi.call/problem1.c: Likewise.
2508 * testsuite/libffi.special/unwindtest.cc: Likewise.
2509 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2511 * testsuite/libffi.call/cls_16byte.c: Likewise.
2512 * testsuite/libffi.call/cls_5byte.c: Likewise.
2514 2004-08-23 David Daney <daney@avtrex.com>
2517 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2518 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2519 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2520 parameters and return types.
2521 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2522 (ffi_prep_closure): Ditto.
2523 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2524 alignment calculations.
2525 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2526 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2528 2004-08-14 Casey Marshall <csm@gnu.org>
2530 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2531 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2532 integer (O32 ABI only).
2533 (ffi_prep_closure): new function.
2534 (ffi_closure_mips_inner_O32): new function.
2535 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2536 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2537 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2538 64 bit integers correctly.
2539 (ffi_closure_O32): new function.
2540 Added DWARF-2 unwind info for both functions.
2542 2004-08-10 Andrew Haley <aph@redhat.com>
2544 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2546 2004-08-01 Robert Millan <robertmh@gnu.org>
2548 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2549 * configure: Regenerate.
2551 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2553 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2554 and mmap() explicitly instead of relying on preset autoconf cache
2556 * aclocal.m4: Regenerate.
2557 * configure: Regenerate.
2559 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2561 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2562 (ffi_check_float_struct): Remove unused prototype.
2564 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2566 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2567 character on Darwin, use '\n\t' instead.
2569 2004-06-26 Matthias Klose <doko@debian.org>
2571 * libtool-version: Fix typo in revision/age.
2573 2004-06-17 Matthias Klose <doko@debian.org>
2575 * libtool-version: New.
2576 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2577 * Makefile.in: Regenerate.
2579 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2581 * Makefile.am: Remove useless multilib rules.
2582 * Makefile.in: Regenerate.
2583 * aclocal.m4: Regenerate with automake 1.8.5.
2584 * configure.ac: Remove useless multilib configury.
2585 * configure: Regenerate.
2587 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2589 * .cvsignore: New file.
2591 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2593 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2595 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2598 2004-06-06 Sean McNeil <sean@mcneil.com>
2600 * configure.ac: Add x86_64-*-freebsd* support.
2601 * configure: Regenerate.
2603 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2606 * configure.ac: Test for existence of mmap and sys/mman.h before
2607 checking blacklist. Fix suggested by Jim Wilson.
2608 * configure: Regenerate.
2610 2004-04-26 Matt Austern <austern@apple.com>
2612 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2614 * src/powerpc/darwin_closure.S: Likewise.
2616 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2618 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2619 error. Reported by Thomas Heller <theller@python.net>.
2620 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2621 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2623 2004-03-20 Matthias Klose <doko@debian.org>
2625 * src/pa/linux.S: Fix typo.
2627 2004-03-19 Matthias Klose <doko@debian.org>
2629 * Makefile.am: Update.
2630 * Makefile.in: Regenerate.
2631 * src/pa/ffi.h.in: Remove.
2632 * src/pa/ffitarget.h: New file.
2634 2004-02-10 Randolph Chung <tausq@debian.org>
2636 * Makefile.am: Add PA support.
2637 * Makefile.in: Regenerate.
2638 * include/Makefile.in: Regenerate.
2639 * configure.ac: Add PA target.
2640 * configure: Regenerate.
2641 * src/pa/ffi.c: New file.
2642 * src/pa/ffi.h.in: Add PA support.
2643 * src/pa/linux.S: New file.
2644 * prep_cif.c: Add PA support.
2646 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2648 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2650 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2651 with ecif->cif->flags.
2652 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2654 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2655 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2656 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2657 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2659 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2661 * configure.in: Rename file to ...
2662 * configure.ac: ... this.
2663 * fficonfig.h.in: Regenerate.
2664 * Makefile.in: Regenerate.
2665 * include/Makefile.in: Regenerate.
2666 * testsuite/Makefile.in: Regenerate.
2668 2004-03-12 Matt Austern <austern@apple.com>
2670 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2671 changes in EH format resulting from addition of linkonce support.
2672 * src/powerpc/darwin_closure.S: Likewise.
2674 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2675 Paolo Bonzini <bonzini@gnu.org>
2677 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2678 Remove VPATH. Remove rules for object files. Remove multilib support.
2679 (AM_CCASFLAGS): Add.
2680 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2681 (AC_PREREQ): Bump version to 2.59.
2682 (AC_INIT): Fill with version info and bug address.
2683 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2684 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2685 De-precious CC so that the right flags are passed down to multilibs.
2686 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2687 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2688 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2689 * configure: Rebuilt.
2690 * aclocal.m4: Likewise.
2691 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2692 * fficonfig.h.in: Likewise.
2694 2004-03-11 Andreas Schwab <schwab@suse.de>
2696 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2697 arguments from fp registers only for the first 8 parameter slots.
2698 Don't convert a float parameter when passed in memory.
2700 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2702 * configure: Regenerate for config/accross.m4 correction.
2704 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2706 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2707 ecif->cif->bytes to bytes.
2708 (ffi_prep_cif_machdep): Add braces around nested if statement.
2710 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2712 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2714 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2715 (ffi_closure_helper_LINUX64): Fix typo.
2716 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2718 * testsuite/libffi.call/float.c: Likewise.
2719 * testsuite/libffi.call/float2.c: Likewise.
2721 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2723 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2724 long double function return and long double arg handling.
2725 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2726 Use "end_pfr" instead of "nf". Correct long double handling.
2728 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2730 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2731 space for long double return value. Adjust stack frame and offsets.
2732 Load f2 long double return.
2734 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2736 * src/types.c: Use 16 byte long double for POWERPC64.
2738 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2740 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2741 when the structure return address is passed in %o0.
2742 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2743 (ffi_v9_layout_struct): Align the field following a nested structure
2744 on a word boundary. Use memmove instead of memcpy.
2745 (ffi_call): Update call to ffi_V9_return_struct.
2746 (ffi_prep_closure): Define 'ctx' only for V8.
2747 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2748 and ffi_closure_sparc_inner_v9.
2749 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2750 Always skip the structure return address. For structures and long
2751 doubles, copy the argument directly.
2752 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2753 if required. Shift the maximum floating-point slot accordingly. For
2754 big structures, copy the argument directly; otherwise, left-justify the
2755 argument and call ffi_v9_layout_struct to lay out the structure on
2757 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2758 (ffi_closure_v8): Pass the structure return address. Update call to
2759 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2760 Skip the 'unimp' insn when returning long doubles and structures.
2761 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2762 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2763 FFI_TYPE_INT handling. Load structures both in integers and
2764 floating-point registers on return.
2765 * README: Update status of the SPARC port.
2767 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2769 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2771 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2773 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2775 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2776 value as of type ffi_arg, not unsigned int.
2778 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2780 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2783 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2785 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2788 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2790 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2793 2003-12-10 Richard Henderson <rth@redhat.com>
2795 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2796 size_t instead of int.
2798 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2800 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2801 * testsuite/libffi.call/many_win32.c (main): Replace variable
2802 int i with unsigned long ul.
2804 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2805 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2806 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2807 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2808 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2809 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2810 * testsuite/libffi.call/cls_align_float.c: Likewise.
2811 * testsuite/libffi.call/cls_align_double.c: Likewise.
2812 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2813 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2815 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2818 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2819 Align arguments to 32 bits.
2821 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2824 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2825 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2826 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2827 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2828 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2829 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2831 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2833 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2835 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2836 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2838 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2840 * Makefile.in: Rebuilt.
2841 * configure: Likewise.
2842 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2845 2003-11-21 Andreas Jaeger <aj@suse.de>
2846 Andreas Tobler <a.tobler@schweiz.ch>
2848 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2849 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2850 * Makefile.in: Rebuilt.
2851 * aclocal.m4: Likewise.
2852 * configure: Likewise.
2853 * fficonfig.h.in: Likewise.
2854 * testsuite/lib/libffi-dg.exp: Add include dir.
2855 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2856 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2857 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2858 for ffi_closure if available.
2859 * testsuite/libffi.call/closure_fn1.c: Likewise.
2860 * testsuite/libffi.call/closure_fn2.c: Likewise.
2861 * testsuite/libffi.call/closure_fn3.c: Likewise.
2862 * testsuite/libffi.call/closure_fn4.c: Likewise.
2863 * testsuite/libffi.call/closure_fn5.c: Likewise.
2864 * testsuite/libffi.call/cls_12byte.c: Likewise.
2865 * testsuite/libffi.call/cls_16byte.c: Likewise.
2866 * testsuite/libffi.call/cls_18byte.c: Likewise.
2867 * testsuite/libffi.call/cls_19byte.c: Likewise.
2868 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2869 * testsuite/libffi.call/cls_20byte.c: Likewise.
2870 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2871 * testsuite/libffi.call/cls_24byte.c: Likewise.
2872 * testsuite/libffi.call/cls_2byte.c: Likewise.
2873 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2874 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2875 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2876 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2877 * testsuite/libffi.call/cls_4byte.c: Likewise.
2878 * testsuite/libffi.call/cls_5byte.c: Likewise.
2879 * testsuite/libffi.call/cls_64byte.c: Likewise.
2880 * testsuite/libffi.call/cls_6byte.c: Likewise.
2881 * testsuite/libffi.call/cls_7byte.c: Likewise.
2882 * testsuite/libffi.call/cls_8byte.c: Likewise.
2883 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2884 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2885 * testsuite/libffi.call/cls_double.c: Likewise.
2886 * testsuite/libffi.call/cls_float.c: Likewise.
2887 * testsuite/libffi.call/cls_schar.c: Likewise.
2888 * testsuite/libffi.call/cls_sint.c: Likewise.
2889 * testsuite/libffi.call/cls_sshort.c: Likewise.
2890 * testsuite/libffi.call/cls_uchar.c: Likewise.
2891 * testsuite/libffi.call/cls_uint.c: Likewise.
2892 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2893 * testsuite/libffi.call/cls_ushort.c: Likewise.
2894 * testsuite/libffi.call/nested_struct.c: Likewise.
2895 * testsuite/libffi.call/nested_struct1.c: Likewise.
2896 * testsuite/libffi.call/nested_struct2.c: Likewise.
2897 * testsuite/libffi.call/nested_struct3.c: Likewise.
2898 * testsuite/libffi.call/problem1.c: Likewise.
2899 * testsuite/libffi.special/unwindtest.cc: Likewise.
2901 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2903 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2905 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2907 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2908 Add -lgcc_s to additional flags.
2910 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2912 * configure.in, include/Makefile.am: PR libgcj/11147, install
2913 the ffitarget.h header file in a gcc versioned and target
2915 * configure: Regenerated.
2916 * Makefile.in, include/Makefile.in: Likewise.
2917 * testsuite/Makefile.in: Likewise.
2919 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2921 * testsuite/libffi.call/closure_fn0.c: Print result and check
2922 with dg-output to make debugging easier.
2923 * testsuite/libffi.call/closure_fn1.c: Likewise.
2924 * testsuite/libffi.call/closure_fn2.c: Likewise.
2925 * testsuite/libffi.call/closure_fn3.c: Likewise.
2926 * testsuite/libffi.call/closure_fn4.c: Likewise.
2927 * testsuite/libffi.call/closure_fn5.c: Likewise.
2928 * testsuite/libffi.call/cls_12byte.c: Likewise.
2929 * testsuite/libffi.call/cls_16byte.c: Likewise.
2930 * testsuite/libffi.call/cls_18byte.c: Likewise.
2931 * testsuite/libffi.call/cls_19byte.c: Likewise.
2932 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2933 * testsuite/libffi.call/cls_20byte.c: Likewise.
2934 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2935 * testsuite/libffi.call/cls_24byte.c: Likewise.
2936 * testsuite/libffi.call/cls_2byte.c: Likewise.
2937 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2938 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2939 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2940 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2941 * testsuite/libffi.call/cls_4byte.c: Likewise.
2942 * testsuite/libffi.call/cls_5byte.c: Likewise.
2943 * testsuite/libffi.call/cls_64byte.c: Likewise.
2944 * testsuite/libffi.call/cls_6byte.c: Likewise.
2945 * testsuite/libffi.call/cls_7byte.c: Likewise.
2946 * testsuite/libffi.call/cls_8byte.c: Likewise.
2947 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2948 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2949 * testsuite/libffi.call/cls_double.c: Likewise.
2950 * testsuite/libffi.call/cls_float.c: Likewise.
2951 * testsuite/libffi.call/cls_schar.c: Likewise.
2952 * testsuite/libffi.call/cls_sint.c: Likewise.
2953 * testsuite/libffi.call/cls_sshort.c: Likewise.
2954 * testsuite/libffi.call/cls_uchar.c: Likewise.
2955 * testsuite/libffi.call/cls_uint.c: Likewise.
2956 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2957 * testsuite/libffi.call/cls_ushort.c: Likewise.
2958 * testsuite/libffi.call/problem1.c: Likewise.
2960 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2963 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2965 * testsuite/libffi.call/cls_9byte2.c: New test case.
2966 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2967 * testsuite/libffi.call/cls_64byte.c: Likewise.
2968 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2969 * testsuite/libffi.call/cls_19byte.c: Likewise.
2970 * testsuite/libffi.call/cls_18byte.c: Likewise.
2971 * testsuite/libffi.call/closure_fn4.c: Likewise.
2972 * testsuite/libffi.call/closure_fn5.c: Likewise.
2973 * testsuite/libffi.call/cls_schar.c: Likewise.
2974 * testsuite/libffi.call/cls_sint.c: Likewise.
2975 * testsuite/libffi.call/cls_sshort.c: Likewise.
2976 * testsuite/libffi.call/nested_struct2.c: Likewise.
2977 * testsuite/libffi.call/nested_struct3.c: Likewise.
2979 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2981 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2982 * testsuite/libffi.call/cls_uchar.c: Likewise.
2983 * testsuite/libffi.call/cls_uint.c: Likewise.
2984 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2985 * testsuite/libffi.call/cls_ushort.c: Likewise.
2986 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2988 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2990 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2993 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2995 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2996 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2998 2003-10-22 David Daney <ddaney@avtrex.com>
3000 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
3001 __attribute__((__mode__(__SI__))) and friends.
3003 2003-10-22 Andreas Schwab <schwab@suse.de>
3005 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
3007 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
3009 * configure.in: AC_LINK_FILES(ffitarget.h).
3010 * configure: Regenerate.
3011 * Makefile.in: Likewise.
3012 * include/Makefile.in: Likewise.
3013 * testsuite/Makefile.in: Likewise.
3014 * fficonfig.h.in: Likewise.
3016 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
3017 Richard Henderson <rth@redhat.com>
3019 Avoid that ffi.h includes fficonfig.h.
3021 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
3022 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
3023 (TARGET_SRC_MIPS_SGI): Removed.
3024 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
3025 (MIPS_SGI): Removed.
3026 (CLEANFILES): Removed.
3027 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
3029 * acconfig.h: Removed.
3030 * configure.in: Compute sizeofs only for double and long double.
3031 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
3032 into AC_DEFINE instead of using acconfig.h. Create
3033 include/ffitarget.h instead of include/fficonfig.h. Rename
3034 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
3035 AC_DEFINE EH_FRAME_FLAGS.
3036 * include/Makefile.am (DISTCLEANFILES): New automake macro.
3037 (hack_DATA): Add ffitarget.h.
3038 * include/ffi.h.in: Remove all system specific definitions.
3039 Declare raw API even if it is not installed, why bother?
3040 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
3041 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
3042 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
3043 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
3044 * include/ffi_common.h (bool): Do not define.
3045 (ffi_assert): Accept failed assertion.
3046 (ffi_type_test): Return void and accept file/line.
3047 (FFI_ASSERT): Pass stringized failed assertion.
3048 (FFI_ASSERT_AT): New macro.
3049 (FFI_ASSERT_VALID_TYPE): New macro.
3050 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
3051 UINT64, SINT64): Define here with gcc's __attribute__ macro
3053 (FLOAT32, ALIGN): Define here instead of in ffi.h
3054 * include/ffi-mips.h: Removed. Its content moved to
3055 src/mips/ffitarget.h after separating assembly and C sections.
3056 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
3057 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
3058 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
3059 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
3060 SIZEOF_ARG -> FFI_SIZEOF_ARG.
3061 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
3062 * src/debug.c (ffi_assert): Accept stringized failed assertion.
3063 (ffi_type_test): Rewritten.
3064 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
3065 FFI_ASSERT_VALID_TYPE.
3066 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
3067 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
3068 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
3069 src/s390/ffitarget.h, src/sh/ffitarget.h,
3070 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
3071 src/x86/ffitarget.h: New files.
3072 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
3073 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
3074 src/powerpc/aix.S, src/powerpc/darwin.S,
3075 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
3076 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
3077 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
3078 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
3079 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
3082 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3084 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
3085 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
3087 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
3089 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
3090 Used when FFI_DEBUG = 1.
3092 2003-10-14 Alan Modra <amodra@bigpond.net.au>
3094 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
3097 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3099 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
3100 FFI_MIPS_O32 for O32 ABI.
3102 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
3104 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
3105 SPARC64. Cleanup whitespaces.
3107 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
3109 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
3110 strongarm, xscale. Cleanup whitespaces.
3111 * testsuite/libffi.call/closure_fn1.c: Likewise.
3112 * testsuite/libffi.call/closure_fn2.c: Likewise.
3113 * testsuite/libffi.call/closure_fn3.c: Likewise.
3114 * testsuite/libffi.call/cls_12byte.c: Likewise.
3115 * testsuite/libffi.call/cls_16byte.c: Likewise.
3116 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3117 * testsuite/libffi.call/cls_20byte.c: Likewise.
3118 * testsuite/libffi.call/cls_24byte.c: Likewise.
3119 * testsuite/libffi.call/cls_2byte.c: Likewise.
3120 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3121 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3122 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3123 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3124 * testsuite/libffi.call/cls_4byte.c: Likewise.
3125 * testsuite/libffi.call/cls_5byte.c: Likewise.
3126 * testsuite/libffi.call/cls_6byte.c: Likewise.
3127 * testsuite/libffi.call/cls_7byte.c: Likewise.
3128 * testsuite/libffi.call/cls_8byte.c: Likewise.
3129 * testsuite/libffi.call/cls_double.c: Likewise.
3130 * testsuite/libffi.call/cls_float.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/problem1.c: Likewise.
3138 * testsuite/libffi.special/unwindtest.cc: Likewise.
3139 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
3141 2003-09-18 David Edelsohn <edelsohn@gnu.org>
3143 * src/powerpc/aix.S: Cleanup whitespaces.
3144 * src/powerpc/aix_closure.S: Likewise.
3146 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3148 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
3149 * src/powerpc/darwin_closure.S: Likewise.
3150 * src/powerpc/ffi_darwin.c: Likewise.
3152 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3153 David Edelsohn <edelsohn@gnu.org>
3155 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
3156 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
3158 * src/powerpc/darwin_closure.S: Likewise.
3159 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
3160 according to the Darwin/AIX ABI.
3161 (ffi_prep_cif_machdep): Likewise.
3162 (ffi_closure_helper_DARWIN): Likewise.
3163 Remove the outgoing parameter stack logic. Simplify the evaluation
3164 of the different CASE types.
3165 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
3166 statement in the trampoline code.
3168 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
3170 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
3171 for the register size.
3172 (ffi_closure_helper_SYSV): Handle the structure return value
3174 (ffi_closure_helper_SYSV): Return the appropriate type when
3175 the registers are used for the structure return value.
3176 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
3177 the 64-bit return value. Update copyright years.
3179 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3181 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
3182 srcdir for ffi_mips.h.
3184 2003-09-12 Alan Modra <amodra@bigpond.net.au>
3186 * src/prep_cif.c (initialize_aggregate): Include tail padding in
3188 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
3189 placement of float result.
3190 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
3191 cast of "resp" for big-endian 64 bit machines.
3193 2003-09-11 Alan Modra <amodra@bigpond.net.au>
3195 * src/types.c (double, longdouble): Merge identical SH and ARM
3196 typedefs, and add POWERPC64.
3197 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
3198 struct split over gpr and rest.
3199 (ffi_prep_cif_machdep): Correct intarg_count for structures.
3200 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
3202 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
3204 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
3207 2003-09-09 Alan Modra <amodra@bigpond.net.au>
3209 * configure: Regenerate.
3211 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3213 * Makefile.am: Remove build rules for ffitest.
3214 * Makefile.in: Rebuilt.
3216 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3218 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3219 about implicit declaration of abort().
3221 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3223 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3224 * Makefile.in: Rebuilt.
3225 * configure.in: Add dejagnu test framework.
3226 * configure: Rebuilt.
3228 * testsuite/Makefile.am: New file.
3229 * testsuite/Makefile.in: Built
3230 * testsuite/lib/libffi-dg.exp: New file.
3231 * testsuite/config/default.exp: Likewise.
3232 * testsuite/libffi.call/call.exp: Likewise.
3233 * testsuite/libffi.call/ffitest.h: Likewise.
3234 * testsuite/libffi.call/closure_fn0.c: Likewise.
3235 * testsuite/libffi.call/closure_fn1.c: Likewise.
3236 * testsuite/libffi.call/closure_fn2.c: Likewise.
3237 * testsuite/libffi.call/closure_fn3.c: Likewise.
3238 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3239 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3240 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3241 * testsuite/libffi.call/cls_2byte.c: Likewise.
3242 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3243 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3244 * testsuite/libffi.call/cls_4byte.c: Likewise.
3245 * testsuite/libffi.call/cls_5byte.c: Likewise.
3246 * testsuite/libffi.call/cls_6byte.c: Likewise.
3247 * testsuite/libffi.call/cls_7byte.c: Likewise.
3248 * testsuite/libffi.call/cls_8byte.c: Likewise.
3249 * testsuite/libffi.call/cls_12byte.c: Likewise.
3250 * testsuite/libffi.call/cls_16byte.c: Likewise.
3251 * testsuite/libffi.call/cls_20byte.c: Likewise.
3252 * testsuite/libffi.call/cls_24byte.c: Likewise.
3253 * testsuite/libffi.call/cls_double.c: Likewise.
3254 * testsuite/libffi.call/cls_float.c: Likewise.
3255 * testsuite/libffi.call/cls_uchar.c: Likewise.
3256 * testsuite/libffi.call/cls_uint.c: Likewise.
3257 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3258 * testsuite/libffi.call/cls_ushort.c: Likewise.
3259 * testsuite/libffi.call/float.c: Likewise.
3260 * testsuite/libffi.call/float1.c: Likewise.
3261 * testsuite/libffi.call/float2.c: Likewise.
3262 * testsuite/libffi.call/many.c: Likewise.
3263 * testsuite/libffi.call/many_win32.c: Likewise.
3264 * testsuite/libffi.call/nested_struct.c: Likewise.
3265 * testsuite/libffi.call/nested_struct1.c: Likewise.
3266 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3267 * testsuite/libffi.call/problem1.c: Likewise.
3268 * testsuite/libffi.call/promotion.c: Likewise.
3269 * testsuite/libffi.call/return_ll.c: Likewise.
3270 * testsuite/libffi.call/return_sc.c: Likewise.
3271 * testsuite/libffi.call/return_uc.c: Likewise.
3272 * testsuite/libffi.call/strlen.c: Likewise.
3273 * testsuite/libffi.call/strlen_win32.c: Likewise.
3274 * testsuite/libffi.call/struct1.c: Likewise.
3275 * testsuite/libffi.call/struct2.c: Likewise.
3276 * testsuite/libffi.call/struct3.c: Likewise.
3277 * testsuite/libffi.call/struct4.c: Likewise.
3278 * testsuite/libffi.call/struct5.c: Likewise.
3279 * testsuite/libffi.call/struct6.c: Likewise.
3280 * testsuite/libffi.call/struct7.c: Likewise.
3281 * testsuite/libffi.call/struct8.c: Likewise.
3282 * testsuite/libffi.call/struct9.c: Likewise.
3283 * testsuite/libffi.special/special.exp: New file.
3284 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3285 * testsuite/libffi.special/unwindtest.cc: Likewise.
3288 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3290 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3293 2003-08-02 Alan Modra <amodra@bigpond.net.au>
3295 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3297 (ffi_closure_helper_LINUX64): Likewise.
3298 * src/powerpc/linux64.S: Remove code writing to parm save area.
3299 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3300 address in lr from ffi_closure_helper_LINUX64 call to calculate
3301 table address. Optimize function tail.
3303 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3305 * src/sparc/ffi.c: Handle all floating point registers.
3306 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3308 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3310 * README: Note that libffi is not part of GCC. Update the project
3313 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3315 * src/powerpc/ppc_closure.S: Include ffi.h.
3317 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3319 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3320 Use C style comments.
3322 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3324 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3325 * Makefile.in: Regenerate.
3326 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3327 * configure: Regenerate.
3328 * include/ffi.h.in: Add SHmedia support.
3329 * src/sh64/ffi.c: New file.
3330 * src/sh64/sysv.S: New file.
3332 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3334 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3335 should be read-only.
3336 * configure: Rebuilt.
3337 * fficonfig.h.in: Rebuilt.
3338 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3339 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3340 * src/powerpc/linux64.S: Likewise.
3341 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3342 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3343 if -fpic/-fPIC/-mrelocatable.
3344 * src/powerpc/powerpc_closure.S: Likewise.
3345 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3346 #write in .eh_frame flags.
3347 * src/sparc/v9.S: Likewise.
3348 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3349 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3350 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3352 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3354 Fixes PR bootstrap/10656
3355 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3356 support for .register pseudo-op.
3357 * src/sparc/v8.S: Use it.
3358 * fficonfig.h.in: Rebuilt.
3359 * configure: Rebuilt.
3361 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3363 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3364 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3365 Make it the default on POWERPC64.
3366 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3367 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3368 * configure: Rebuilt.
3369 * src/powerpc/ffi.c (hidden): Define.
3370 (ffi_prep_args_SYSV): Renamed from
3371 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3372 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3373 ASM_NEEDS_REGISTERS64): New.
3374 (ffi_prep_args64): New function.
3375 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3376 (ffi_call): Likewise.
3377 (ffi_prep_closure): Likewise.
3378 (flush_icache): Surround by #ifndef POWERPC64.
3379 (ffi_dblfl): New union type.
3380 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3381 (ffi_closure_helper_LINUX64): New function.
3382 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3384 * src/powerpc/sysv.S: Likewise.
3385 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3386 * src/powerpc/linux64.S: New file.
3387 * src/powerpc/linux64_closure.S: New file.
3388 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3389 src/powerpc/linux64_closure.S.
3390 (TARGET_SRC_POWERPC): Likewise.
3392 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3393 closure_test_fn3): Fix result printing on big-endian 64-bit
3395 (main): Print tst2_arg instead of uninitialized tst2_result.
3397 * src/ffitest.c (main): Hide what closure pointer really points to
3400 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3402 * configure.in (arm-*-netbsdelf*): Add configuration.
3403 (configure): Regenerated.
3405 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3407 * include/Makefile.in: Regenerate.
3409 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3411 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3413 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3414 Receive closure pointer through parameter, read args using
3415 __builtin_dwarf_cfa.
3416 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3418 2003-03-12 Andreas Schwab <schwab@suse.de>
3420 * configure.in: Avoid trailing /. in toolexeclibdir.
3421 * configure: Rebuilt.
3423 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3425 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3427 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3429 * libffi/src/powerpc/darwin_closure.S:
3430 Fix alignement bug, allocate 8 bytes for the result.
3431 * libffi/src/powerpc/aix_closure.S:
3433 * libffi/src/powerpc/ffi_darwin.c:
3434 Update stackframe description for aix/darwin_closure.S.
3436 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3438 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add&n
OSDN Git Service
