1 2010-12-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
3 * configure.ac (libffi_cv_as_ascii_pseudo_op): Use double
5 (libffi_cv_as_string_pseudo_op): Likewise.
6 * configure: Regenerate.
8 2010-12-03 Chung-Lin Tang <cltang@codesourcery.com>
10 * src/arm/sysv.S (ffi_closure_SYSV): Add UNWIND to .pad directive.
11 (ffi_closure_VFP): Same.
12 (ffi_call_VFP): Move down to before ffi_closure_VFP. Add '.fpu vfp'
15 2010-12-01 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
17 * testsuite/libffi.call/ffitest.h [__sgi] (PRId64, PRIu64): Define.
20 2010-11-29 Richard Henderson <rth@redhat.com>
21 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
23 * src/x86/sysv.S (FDE_ENCODING, FDE_ENCODE): Define.
24 (.eh_frame): Use FDE_ENCODING.
25 (.LASFDE1, .LASFDE2, LASFDE3): Simplify with FDE_ENCODE.
27 2010-11-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
29 * testsuite/lib/libffi-dg.exp: Rename ...
30 * testsuite/lib/libffi.exp: ... to this.
31 * libffi/testsuite/libffi.call/call.exp: Don't load libffi-dg.exp.
32 * libffi/testsuite/libffi.special/special.exp: Likewise.
34 2010-10-28 Chung-Lin Tang <cltang@codesourcery.com>
36 * src/arm/ffi.c (ffi_prep_args): Add VFP register argument handling
37 code, new parameter, and return value. Update comments.
38 (ffi_prep_cif_machdep): Add case for VFP struct return values. Add
39 call to layout_vfp_args().
40 (ffi_call_SYSV): Update declaration.
41 (ffi_call_VFP): New declaration.
42 (ffi_call): Add VFP struct return conditions. Call ffi_call_VFP()
44 (ffi_closure_VFP): New declaration.
45 (ffi_closure_SYSV_inner): Add new vfp_args parameter, update call to
46 ffi_prep_incoming_args_SYSV().
47 (ffi_prep_incoming_args_SYSV): Update parameters. Add VFP argument
49 (ffi_prep_closure_loc): Pass ffi_closure_VFP to trampoline
50 construction under VFP hard-float.
51 (rec_vfp_type_p): New function.
53 (place_vfp_arg): Same.
54 (layout_vfp_args): Same.
55 * src/arm/ffitarget.h (ffi_abi): Add FFI_VFP. Define FFI_DEFAULT_ABI
56 based on __ARM_PCS_VFP.
57 (FFI_EXTRA_CIF_FIELDS): Define for adding VFP hard-float specific
59 (FFI_TYPE_STRUCT_VFP_FLOAT): Define internally used type code.
60 (FFI_TYPE_STRUCT_VFP_DOUBLE): Same.
61 * src/arm/sysv.S (ffi_call_SYSV): Change call of ffi_prep_args() to
62 direct call. Move function pointer load upwards.
63 (ffi_call_VFP): New function.
64 (ffi_closure_VFP): Same.
66 * testsuite/lib/libffi-dg.exp (check-flags): New function.
67 (dg-skip-if): New function.
68 * testsuite/libffi.call/cls_double_va.c: Skip if target is arm*-*-*
69 and compiler options include -mfloat-abi=hard.
70 * testsuite/libffi.call/cls_longdouble_va.c: Same.
72 2010-10-01 Jakub Jelinek <jakub@redhat.com>
75 * src/x86/ffi64.c (ffi_prep_cif_machdep): Ensure cif->bytes is
77 * testsuite/libffi.call/many2.c: New test.
79 2010-08-20 Mark Wielaard <mjw@redhat.com>
81 * src/closures.c (open_temp_exec_file_mnt): Check if getmntent_r
84 2010-08-09 Andreas Tobler <andreast@fgznet.ch>
86 * configure.ac: Add target powerpc64-*-freebsd*.
87 * configure: Regenerate.
88 * testsuite/libffi.call/cls_align_longdouble_split.c: Pass
89 -mlong-double-128 only to linux targets.
90 * testsuite/libffi.call/cls_align_longdouble_split2.c: Likewise.
91 * testsuite/libffi.call/cls_longdouble.c: Likewise.
92 * testsuite/libffi.call/huge_struct.c: Likewise.
94 2010-07-10 Evan Phoenix <evan@fallingsnow.net>
96 * src/closures.c (selinux_enabled_check): Fix strncmp usage bug.
98 2010-07-07 Dan HorĂ¡k <dan@danny.cz>
100 * include/ffi.h.in: Protect #define with #ifndef.
101 * src/powerpc/ffitarget.h: Ditto.
102 * src/s390/ffitarget.h: Ditto.
103 * src/sparc/ffitarget.h: Ditto.
105 2010-07-07 Neil Roberts <neil@linux.intel.com>
107 * src/x86/sysv.S (ffi_call_SYSV): Align the stack pointer to
110 2010-07-02 Jakub Jelinek <jakub@redhat.com>
112 * Makefile.am (AM_MAKEFLAGS): Pass also mandir to submakes.
113 * Makefile.in: Regenerated.
115 2010-05-19 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
117 * configure.ac (libffi_cv_as_x86_pcrel): Check for illegal in as
119 (libffi_cv_as_ascii_pseudo_op): Check for .ascii.
120 (libffi_cv_as_string_pseudo_op): Check for .string.
121 * configure: Regenerate.
122 * fficonfig.h.in: Regenerate.
123 * src/x86/sysv.S (.eh_frame): Use .ascii, .string or error.
125 2010-05-05 Michael Kohler <michaelkohler@live.com>
127 * src/dlmalloc.c (dlfree): Fix spelling.
128 * src/ia64/ffi.c (ffi_prep_cif_machdep): Ditto.
129 * configure.ac: Ditto.
130 * configure: Rebuilt.
132 2010-04-13 Dan Witte <dwitte@mozilla.com>
134 * msvcc.sh: Build with -W3 instead of -Wall.
135 * src/powerpc/ffi_darwin.c: Remove build warnings.
136 * src/x86/ffi.c: Ditto.
137 * src/x86/ffitarget.h: Ditto.
139 2010-04-12 Dan Witte <dwitte@mozilla.com>
140 Walter Meinl <wuno@lsvw.de>
142 * configure.ac: Add OS/2 support.
143 * configure: Rebuilt.
144 * src/closures.c: Ditto.
145 * src/dlmalloc.c: Ditto.
146 * src/x86/win32.S: Ditto.
148 2010-04-07 Jakub Jelinek <jakub@redhat.com>
150 * testsuite/libffi.call/err_bad_abi.c: Remove unused args variable.
152 2010-04-02 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
154 * Makefile.in: Regenerate.
155 * aclocal.m4: Regenerate.
156 * include/Makefile.in: Regenerate.
157 * man/Makefile.in: Regenerate.
158 * testsuite/Makefile.in: Regenerate.
160 2010-03-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
162 * configure.ac (libffi_cv_as_x86_64_unwind_section_type): New test.
163 * configure: Regenerate.
164 * fficonfig.h.in: Regenerate.
165 * libffi/src/x86/unix64.S (.eh_frame)
166 [HAVE_AS_X86_64_UNWIND_SECTION_TYPE]: Use @unwind section type.
168 2010-03-14 Matthias Klose <doko@ubuntu.com>
170 * src/x86/ffi64.c: Fix typo in comment.
171 * src/x86/ffi.c: Use /* ... */ comment style.
173 2010-02-24 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
175 * doc/libffi.texi (The Closure API): Fix typo.
176 * doc/libffi.info: Remove.
178 2010-02-15 Matthias Klose <doko@ubuntu.com>
180 * src/arm/sysv.S (__ARM_ARCH__): Define for processor
183 2010-01-15 Anthony Green <green@redhat.com>
185 * README: Add notes on building with Microsoft Visual C++.
187 2010-01-15 Daniel Witte <dwitte@mozilla.com>
189 * msvcc.sh: New file.
191 * src/x86/win32.S: Port assembly routines to MSVC and #ifdef.
192 * src/x86/ffi.c: Tweak function declaration and remove excess
194 * include/ffi.h.in: Add __declspec(align(8)) to typedef struct
197 * src/x86/ffi.c: Merge ffi_call_SYSV and ffi_call_STDCALL into new
198 function ffi_call_win32 on X86_WIN32.
199 * src/x86/win32.S (ffi_call_SYSV): Rename to ffi_call_win32.
200 (ffi_call_STDCALL): Remove.
202 * src/prep_cif.c (ffi_prep_cif): Move stack space allocation code
203 to ffi_prep_cif_machdep for x86.
204 * src/x86/ffi.c (ffi_prep_cif_machdep): To here.
206 2010-01-15 Oliver Kiddle <okiddle@yahoo.co.uk>
208 * src/x86/ffitarget.h (ffi_abi): Check for __i386 and __amd64 for
209 Sun Studio compiler compatibility.
211 2010-01-12 Conrad Irwin <conrad.irwin@gmail.com>
213 * doc/libffi.texi: Add closure example.
215 2010-01-07 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
218 * testsuite/libffi.call/ffitest.h [__alpha__ && __osf__] (PRIdLL,
219 PRIuLL, PRId64, PRIu64, PRIuPTR): Define.
220 * testsuite/libffi.call/cls_align_sint64.c: Add -Wno-format on
222 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
223 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
224 * testsuite/libffi.call/return_ll1.c: Likewise.
225 * testsuite/libffi.call/stret_medium2.c: Likewise.
226 * testsuite/libffi.special/ffitestcxx.h (allocate_mmap): Cast
227 MAP_FAILED to char *.
229 2010-01-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
231 * src/mips/n32.S: Use .abicalls and .eh_frame with __GNUC__.
233 2009-12-31 Anthony Green <green@redhat.com>
235 * README: Update for libffi 3.0.9.
237 2009-12-27 Matthias Klose <doko@ubuntu.com>
239 * configure.ac (HAVE_LONG_DOUBLE): Define for mips when
241 * configure: Rebuilt.
243 2009-12-26 Anthony Green <green@redhat.com>
245 * testsuite/libffi.call/cls_longdouble_va.c: Mark as xfail for
247 * testsuite/libffi.call/cls_double_va.c: Ditto.
249 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
251 * testsuite/libffi.call/ffitest.h: Conditionally include stdint.h
253 * testsuite/libffi.special/unwindtest.cc: Ditto.
255 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
257 * configure.ac: Add amd64-*-openbsd*.
258 * configure: Rebuilt.
259 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Link
260 openbsd programs with -lpthread.
262 2009-12-26 Anthony Green <green@redhat.com>
264 * testsuite/libffi.call/cls_double_va.c,
265 testsuite/libffi.call/cls_longdouble.c,
266 testsuite/libffi.call/cls_longdouble_va.c,
267 testsuite/libffi.call/cls_pointer.c,
268 testsuite/libffi.call/cls_pointer_stack.c: Remove xfail for
269 mips*-*-* and arm*-*-*.
270 * testsuite/libffi.call/cls_align_longdouble_split.c,
271 testsuite/libffi.call/cls_align_longdouble_split2.c,
272 testsuite/libffi.call/stret_medium2.c,
273 testsuite/libffi.call/stret_medium.c,
274 testsuite/libffi.call/stret_large.c,
275 testsuite/libffi.call/stret_large2.c: Remove xfail for arm*-*-*.
277 2009-12-31 Kay Tietz <ktietz70@googlemail.com>
279 * testsuite/libffi.call/ffitest.h,
280 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRuLL): Fix
283 2009-12-31 Carlo Bramini <carlo.bramix@libero.it>
285 * configure.ac (AM_LTLDFLAGS): Define for windows hosts.
286 * Makefile.am (libffi_la_LDFLAGS): Add AM_LTLDFLAGS.
287 * configure: Rebuilt.
288 * Makefile.in: Rebuilt.
290 2009-12-31 Anthony Green <green@redhat.com>
293 * testsuite/libffi.call/huge_struct.c: New test case from Blake
296 2009-12-28 David Edelsohn <edelsohn@gnu.org>
298 * src/powerpc/ffi_darwin.c (ffi_prep_args): Copy abi and nargs to
300 (aix_adjust_aggregate_sizes): New function.
301 (ffi_prep_cif_machdep): Call it.
303 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
305 * configure.ac: Define FFI_MMAP_EXEC_WRIT for the given targets.
306 * configure: Regenerate.
307 * fficonfig.h.in: Likewise.
308 * src/closures.c: Remove the FFI_MMAP_EXEC_WRIT definition for
311 2009-12-26 Andreas Schwab <schwab@linux-m68k.org>
313 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Advance intarg_count
314 when a float arguments is passed in memory.
315 (ffi_closure_helper_SYSV): Mark general registers as used up when
316 a 64bit or soft-float long double argument is passed in memory.
318 2009-12-25 Matthias Klose <doko@ubuntu.com>
320 * man/ffi_call.3: Fix #include in examples.
321 * doc/libffi.texi: Add dircategory.
323 2009-12-25 Frank Everdij <f.p.x.everdij@tudelft.nl>
325 * include/ffi.h.in: Placed '__GNUC__' ifdef around
326 '__attribute__((aligned(8)))' in ffi_closure, fixes compile for
328 * include/ffi_common.h: Added '__sgi' define to non
329 '__attribute__((__mode__()))' integer typedefs.
330 * src/mips/ffi.c (ffi_call, ffi_closure_mips_inner_O32,
331 ffi_closure_mips_inner_N32): Added 'defined(_MIPSEB)' to BE check.
332 (ffi_closure_mips_inner_O32, ffi_closure_mips_inner_N32): Added
333 FFI_LONGDOUBLE support and alignment(N32 only).
334 * src/mips/ffitarget.h: Corrected '#include <sgidefs.h>' for IRIX and
335 fixed non '__attribute__((__mode__()))' integer typedefs.
336 * src/mips/n32.S: Put '#ifdef linux' around '.abicalls' and '.eh_frame'
337 since they are Linux/GNU Assembler specific.
339 2009-12-25 Bradley Smith <brad@brad-smith.co.uk>
341 * configure.ac, Makefile.am, src/avr32/ffi.c,
342 src/avr32/ffitarget.h,
343 src/avr32/sysv.S: Add AVR32 port.
344 * configure, Makefile.in: Rebuilt.
346 2009-12-21 Andreas Tobler <a.tobler@schweiz.org>
348 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
349 * configure: Regenerate.
351 2009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
353 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
355 2009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
357 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
360 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
362 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
365 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
367 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
369 2009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
372 * src/closures.c [X86_64 && __sun__ && __svr4__]
373 (FFI_MMAP_EXEC_WRIT): Define.
375 2009-12-08 David Daney <ddaney@caviumnetworks.com>
377 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
378 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
379 * testsuite/libffi.call/stret_large.c: Same.
380 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
381 * testsuite/libffi.call/stret_large2.c: Same.
382 * testsuite/libffi.call/stret_medium2.c: Same.
384 2009-12-07 David Edelsohn <edelsohn@gnu.org>
386 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
389 2009-12-05 David Edelsohn <edelsohn@gnu.org>
391 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
393 * src/powerpc/aix_closure.S: Same.
395 2009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
397 * Makefile.in: Regenerate.
398 * configure: Regenerate.
399 * include/Makefile.in: Regenerate.
400 * man/Makefile.in: Regenerate.
401 * testsuite/Makefile.in: Regenerate.
403 2009-12-04 David Edelsohn <edelsohn@gnu.org>
405 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
408 2009-12-04 Uros Bizjak <ubizjak@gmail.com>
411 * src/x86/ffi64.c (classify_argument): Update from
412 gcc/config/i386/i386.c.
413 (ffi_closure_unix64_inner): Do not use the address of two consecutive
414 SSE registers directly.
415 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
416 for x86_64 linux targets.
418 2009-12-04 David Edelsohn <edelsohn@gnu.org>
420 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
421 pfr for long double split between fpr13 and stack.
423 2009-12-03 David Edelsohn <edelsohn@gnu.org>
425 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
426 fparg_count twice for long double.
428 2009-12-03 David Edelsohn <edelsohn@gnu.org>
431 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
433 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
435 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
436 Remove xfails for x86 linux targets.
438 2009-12-02 David Edelsohn <edelsohn@gnu.org>
440 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
443 2009-12-01 David Edelsohn <edelsohn@gnu.org>
445 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
446 register usage. Call ffi_prep_args directly. Add long double
447 return value support.
448 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
449 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
450 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
451 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
453 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
455 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
456 Allocate result area between params and FPRs.
458 2009-11-30 David Edelsohn <edelsohn@gnu.org>
461 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
463 * src/powerpc/aix.S: Implement AIX64 version.
464 * src/powerpc/aix_closure.S: Implement AIX64 version.
465 (ffi_closure_ASM): Use extsb, lha and displament addresses.
466 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
468 (ffi_prep_cif_machdep): Same.
470 (ffi_closure_helper_DARWIN): Same.
472 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
475 * testsuite/libffi.call/testclosure.c: New test.
477 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
479 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
480 assembly version use of ___chkstk.
482 2009-09-23 Matthias Klose <doko@ubuntu.com>
484 PR libffi/40242, PR libffi/41443
485 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
486 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
487 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
488 Change the conditionals to __SOFTFP__ || __ARM_EABI__
489 for -mfloat-abi=softfp to work.
491 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
493 PR testsuite/32843 (strikes again)
494 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
495 enable proper extension on char and short.
497 2009-09-15 David Daney <ddaney@caviumnetworks.com>
499 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
500 handling for FFI_TYPE_POINTER.
501 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
502 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
503 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
504 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
505 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
506 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
507 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
508 structure and pointer returns.
509 (ffi_closure_N32): Add handling for pointer returns.
510 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
511 calc_n32_return_struct_flags): Handle soft-float.
512 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
513 (ffi_call_N32): Declare proper argument types.
514 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
517 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
519 * configure.ac (AC_PREREQ): Bump to 2.64.
521 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
523 * Makefile.am (install-html, install-pdf): Remove.
524 * Makefile.in: Regenerate.
526 * Makefile.in: Regenerate.
527 * aclocal.m4: Regenerate.
528 * configure: Regenerate.
529 * fficonfig.h.in: Regenerate.
530 * include/Makefile.in: Regenerate.
531 * man/Makefile.in: Regenerate.
532 * testsuite/Makefile.in: Regenerate.
534 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
536 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
538 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
541 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
542 return types for X86_WIN32.
543 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
544 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
545 _ffi_closure_STDCALL): Likewise.
547 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
548 (dlmmap, dlmunmap): Also use these functions on Cygwin.
550 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
555 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
558 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
560 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
561 to "" before adding paths. (This reinstates an assignment that
562 was removed by my 2009-06-30 commit, but changes the initial
563 value from "." to "".)
565 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
568 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
569 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
572 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
574 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
575 to ld_library_path. Use add_path. Add just find_libgcc_s
576 to ld_library_path, not every libgcc multilib directory.
578 2009-06-16 Wim Lewis <wiml@hhhh.org>
580 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
581 supposed to be callee-saved.
582 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
583 return buffer for odd-size structs.
585 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
588 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
589 allow_stack_execute for Darwin.
591 2009-06-16 Andrew Haley <aph@redhat.com>
593 * configure.ac (TARGETDIR): Add missing blank lines.
594 * configure: Regenerate.
596 2009-06-16 Andrew Haley <aph@redhat.com>
598 * testsuite/libffi.call/cls_align_sint64.c,
599 testsuite/libffi.call/cls_align_uint64.c,
600 testsuite/libffi.call/cls_longdouble_va.c,
601 testsuite/libffi.call/cls_ulonglong.c,
602 testsuite/libffi.call/return_ll1.c,
603 testsuite/libffi.call/stret_medium2.c: Fix printf format
605 * testsuite/libffi.call/ffitest.h,
606 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
608 2009-06-15 Andrew Haley <aph@redhat.com>
610 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
611 * testsuite/libffi.call/err_bad_abi.c: Likewise.
613 2009-06-12 Andrew Haley <aph@redhat.com>
615 * Makefile.am: Remove info_TEXINFOS.
617 2009-06-12 Andrew Haley <aph@redhat.com>
619 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
620 testsuite/libffi.call/cls_align_uint64.c,
621 testsuite/libffi.call/cls_ulonglong.c,
622 testsuite/libffi.call/return_ll1.c,
623 testsuite/libffi.call/stret_medium2.c: Fix printf format
625 testsuite/libffi.special/unwindtest.cc: include stdint.h.
627 2009-06-11 Timothy Wall <twall@users.sf.net>
632 include/ffi_common.h,
637 src/x86/win64.S (new),
638 README: Added win64 support (mingw or MSVC)
642 testsuite/Makefile.in,
644 aclocal.m4: Regenerated
645 * ltcf-c.sh: properly escape cygwin/w32 path
646 * man/ffi_call.3: Clarify size requirements for return value.
647 * src/x86/ffi64.c: Fix filename in comment.
648 * src/x86/win32.S: Remove unused extern.
650 * testsuite/libffi.call/closure_fn0.c,
651 testsuite/libffi.call/closure_fn1.c,
652 testsuite/libffi.call/closure_fn2.c,
653 testsuite/libffi.call/closure_fn3.c,
654 testsuite/libffi.call/closure_fn4.c,
655 testsuite/libffi.call/closure_fn5.c,
656 testsuite/libffi.call/closure_fn6.c,
657 testsuite/libffi.call/closure_stdcall.c,
658 testsuite/libffi.call/cls_12byte.c,
659 testsuite/libffi.call/cls_16byte.c,
660 testsuite/libffi.call/cls_18byte.c,
661 testsuite/libffi.call/cls_19byte.c,
662 testsuite/libffi.call/cls_1_1byte.c,
663 testsuite/libffi.call/cls_20byte.c,
664 testsuite/libffi.call/cls_20byte1.c,
665 testsuite/libffi.call/cls_24byte.c,
666 testsuite/libffi.call/cls_2byte.c,
667 testsuite/libffi.call/cls_3_1byte.c,
668 testsuite/libffi.call/cls_3byte1.c,
669 testsuite/libffi.call/cls_3byte2.c,
670 testsuite/libffi.call/cls_4_1byte.c,
671 testsuite/libffi.call/cls_4byte.c,
672 testsuite/libffi.call/cls_5_1_byte.c,
673 testsuite/libffi.call/cls_5byte.c,
674 testsuite/libffi.call/cls_64byte.c,
675 testsuite/libffi.call/cls_6_1_byte.c,
676 testsuite/libffi.call/cls_6byte.c,
677 testsuite/libffi.call/cls_7_1_byte.c,
678 testsuite/libffi.call/cls_7byte.c,
679 testsuite/libffi.call/cls_8byte.c,
680 testsuite/libffi.call/cls_9byte1.c,
681 testsuite/libffi.call/cls_9byte2.c,
682 testsuite/libffi.call/cls_align_double.c,
683 testsuite/libffi.call/cls_align_float.c,
684 testsuite/libffi.call/cls_align_longdouble.c,
685 testsuite/libffi.call/cls_align_longdouble_split.c,
686 testsuite/libffi.call/cls_align_longdouble_split2.c,
687 testsuite/libffi.call/cls_align_pointer.c,
688 testsuite/libffi.call/cls_align_sint16.c,
689 testsuite/libffi.call/cls_align_sint32.c,
690 testsuite/libffi.call/cls_align_sint64.c,
691 testsuite/libffi.call/cls_align_uint16.c,
692 testsuite/libffi.call/cls_align_uint32.c,
693 testsuite/libffi.call/cls_align_uint64.c,
694 testsuite/libffi.call/cls_dbls_struct.c,
695 testsuite/libffi.call/cls_double.c,
696 testsuite/libffi.call/cls_double_va.c,
697 testsuite/libffi.call/cls_float.c,
698 testsuite/libffi.call/cls_longdouble.c,
699 testsuite/libffi.call/cls_longdouble_va.c,
700 testsuite/libffi.call/cls_multi_schar.c,
701 testsuite/libffi.call/cls_multi_sshort.c,
702 testsuite/libffi.call/cls_multi_sshortchar.c,
703 testsuite/libffi.call/cls_multi_uchar.c,
704 testsuite/libffi.call/cls_multi_ushort.c,
705 testsuite/libffi.call/cls_multi_ushortchar.c,
706 testsuite/libffi.call/cls_pointer.c,
707 testsuite/libffi.call/cls_pointer_stack.c,
708 testsuite/libffi.call/cls_schar.c,
709 testsuite/libffi.call/cls_sint.c,
710 testsuite/libffi.call/cls_sshort.c,
711 testsuite/libffi.call/cls_uchar.c,
712 testsuite/libffi.call/cls_uint.c,
713 testsuite/libffi.call/cls_ulonglong.c,
714 testsuite/libffi.call/cls_ushort.c,
715 testsuite/libffi.call/err_bad_abi.c,
716 testsuite/libffi.call/err_bad_typedef.c,
717 testsuite/libffi.call/float2.c,
718 testsuite/libffi.call/huge_struct.c,
719 testsuite/libffi.call/nested_struct.c,
720 testsuite/libffi.call/nested_struct1.c,
721 testsuite/libffi.call/nested_struct10.c,
722 testsuite/libffi.call/nested_struct2.c,
723 testsuite/libffi.call/nested_struct3.c,
724 testsuite/libffi.call/nested_struct4.c,
725 testsuite/libffi.call/nested_struct5.c,
726 testsuite/libffi.call/nested_struct6.c,
727 testsuite/libffi.call/nested_struct7.c,
728 testsuite/libffi.call/nested_struct8.c,
729 testsuite/libffi.call/nested_struct9.c,
730 testsuite/libffi.call/problem1.c,
731 testsuite/libffi.call/return_ldl.c,
732 testsuite/libffi.call/return_ll1.c,
733 testsuite/libffi.call/stret_large.c,
734 testsuite/libffi.call/stret_large2.c,
735 testsuite/libffi.call/stret_medium.c,
736 testsuite/libffi.call/stret_medium2.c,
737 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
738 of checking for MMAP. Use intptr_t instead of long casts.
740 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
742 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
743 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
744 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
746 2009-06-09 Andrew Haley <aph@redhat.com>
748 * src/x86/freebsd.S: Add missing file.
750 2009-06-08 Andrew Haley <aph@redhat.com>
752 Import from libffi 3.0.8:
754 * doc/libffi.texi: New file.
755 * doc/libffi.info: Likewise.
756 * doc/stamp-vti: Likewise.
757 * man/Makefile.am: New file.
758 * man/ffi_call.3: New file.
760 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
762 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
764 * configure.ac: Bump version to 3.0.8.
765 parisc*-*-linux*: Add.
766 i386-*-freebsd* | i386-*-openbsd*: Add.
767 powerpc-*-beos*: Add.
768 AM_CONDITIONAL X86_FREEBSD: Add.
769 AC_CONFIG_FILES: Add man/Makefile.
771 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
773 2009-06-08 Andrew Haley <aph@redhat.com>
775 * README: Import from libffi 3.0.8.
777 2009-06-08 Andrew Haley <aph@redhat.com>
779 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
780 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
781 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
782 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
784 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
785 * testsuite/libffi.call/stret_medium.c: Likewise.
786 * testsuite/libffi.call/stret_large2.c: Likewise.
787 * testsuite/libffi.call/stret_large.c: Likewise.
789 2008-12-26 Timothy Wall <twall@users.sf.net>
791 * testsuite/libffi.call/cls_longdouble.c,
792 testsuite/libffi.call/cls_longdouble_va.c,
793 testsuite/libffi.call/cls_align_longdouble.c,
794 testsuite/libffi.call/cls_align_longdouble_split.c,
795 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
796 failures on x86_64 cygwin/mingw.
798 2008-12-22 Timothy Wall <twall@users.sf.net>
800 * testsuite/libffi.call/closure_fn0.c,
801 testsuite/libffi.call/closure_fn1.c,
802 testsuite/libffi.call/closure_fn2.c,
803 testsuite/libffi.call/closure_fn3.c,
804 testsuite/libffi.call/closure_fn4.c,
805 testsuite/libffi.call/closure_fn5.c,
806 testsuite/libffi.call/closure_fn6.c,
807 testsuite/libffi.call/closure_loc_fn0.c,
808 testsuite/libffi.call/closure_stdcall.c,
809 testsuite/libffi.call/cls_align_pointer.c,
810 testsuite/libffi.call/cls_pointer.c,
811 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
812 pointer to integer (intptr_t).
813 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
815 2008-07-24 Anthony Green <green@redhat.com>
817 * testsuite/libffi.call/cls_dbls_struct.c,
818 testsuite/libffi.call/cls_double_va.c,
819 testsuite/libffi.call/cls_longdouble.c,
820 testsuite/libffi.call/cls_longdouble_va.c,
821 testsuite/libffi.call/cls_pointer.c,
822 testsuite/libffi.call/cls_pointer_stack.c,
823 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
826 2008-03-04 Anthony Green <green@redhat.com>
830 * testsuite/libffi.call/cls_align_longdouble_split2.c
831 testsuite/libffi.call/cls_align_longdouble_split.c
832 testsuite/libffi.call/cls_dbls_struct.c
833 testsuite/libffi.call/cls_double_va.c
834 testsuite/libffi.call/cls_longdouble.c
835 testsuite/libffi.call/cls_longdouble_va.c
836 testsuite/libffi.call/cls_pointer.c
837 testsuite/libffi.call/cls_pointer_stack.c
838 testsuite/libffi.call/err_bad_abi.c
839 testsuite/libffi.call/err_bad_typedef.c
840 testsuite/libffi.call/stret_large2.c
841 testsuite/libffi.call/stret_large.c
842 testsuite/libffi.call/stret_medium2.c
843 testsuite/libffi.call/stret_medium.c: New tests from Apple.
845 2009-06-05 Andrew Haley <aph@redhat.com>
847 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
850 2009-06-04 Andrew Haley <aph@redhat.com>
852 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
855 2008-02-26 Anthony Green <green@redhat.com>
856 Thomas Heller <theller@ctypes.org>
858 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
861 2008-02-03 Timothy Wall <twall@users.sf.net>
863 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
864 offset based on code pointer, not data pointer.
866 2008-01-31 Timothy Wall <twall@users.sf.net>
868 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
870 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
872 * src/x86/win32.S: Add assembly for stdcall closure.
873 * src/x86/ffi.c: Initialize stdcall closure trampoline.
875 2009-06-04 Andrew Haley <aph@redhat.com>
877 * include/ffi.h.in: Change void (*)() to void (*)(void).
878 * src/x86/ffi.c: Likewise.
880 2009-06-04 Andrew Haley <aph@redhat.com>
882 * src/powerpc/ppc_closure.S: Insert licence header.
883 * src/powerpc/linux64_closure.S: Likewise.
884 * src/m68k/sysv.S: Likewise.
886 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
887 * src/powerpc/ffi.c: Likewise.
888 * src/powerpc/ffi_darwin.c: Likewise.
889 * src/m32r/ffi.c: Likewise.
890 * src/sh64/ffi.c: Likewise.
891 * src/x86/ffi64.c: Likewise.
892 * src/alpha/ffi.c: Likewise.
893 * src/alpha/osf.S: Likewise.
894 * src/frv/ffi.c: Likewise.
895 * src/s390/ffi.c: Likewise.
896 * src/pa/ffi.c: Likewise.
897 * src/pa/hpux32.S: Likewise.
898 * src/ia64/unix.S: Likewise.
899 * src/ia64/ffi.c: Likewise.
900 * src/sparc/ffi.c: Likewise.
901 * src/mips/ffi.c: Likewise.
902 * src/sh/ffi.c: Likewise.
904 2008-02-15 David Daney <ddaney@avtrex.com>
906 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
907 Define (conditionally), and use it to include cachectl.h.
908 (ffi_prep_closure_loc): Fix cache flushing.
909 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
911 2009-06-04 Andrew Haley <aph@redhat.com>
917 src/powerpc/ffitarget.h,
919 src/sh64/ffitarget.h,
930 src/alpha/ffitarget.h,
933 src/m68k/ffitarget.h,
936 src/s390/ffitarget.h,
938 src/cris/ffitarget.h,
943 src/ia64/ffitarget.h,
946 src/ia64/ia64_flags.h,
950 src/sparc/ffitarget.h,
953 src/mips/ffitarget.h,
960 src/sh/sysv.S: Update license text.
962 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
964 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
965 (.eh_frame): Add FDE for it.
967 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
969 * configure.ac: Also check if assembler supports pc-relative
970 relocs on X86_WIN32 targets.
971 * configure: Regenerate.
972 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
973 (_ffi_call_SYSV): Add missing function type symbol .def and
974 add EH markup labels.
975 (_ffi_call_STDCALL): Likewise.
976 (_ffi_closure_SYSV): Likewise.
977 (_ffi_closure_raw_SYSV): Likewise.
978 (.eh_frame): Add hand-crafted EH data.
980 2009-04-09 Jakub Jelinek <jakub@redhat.com>
982 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
983 version 3 of the GNU General Public License and to point readers
984 at the COPYING3 file and the FSF's license web page.
985 * testsuite/libffi.call/call.exp: Likewise.
986 * testsuite/libffi.special/special.exp: Likewise.
988 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
990 * configure: Regenerate.
992 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
995 * configure.ac (HAVE_AS_X86_PCREL): New test.
996 * configure: Regenerate.
997 * fficonfig.h.in: Regenerate.
998 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
999 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
1000 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
1001 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
1002 * src/x86/unix64.S (.Lstore_table): Move to .text section.
1003 (.Lload_table): Likewise.
1004 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
1006 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1008 * configure: Regenerate.
1010 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
1012 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
1013 signed/unsigned int8/16 return values.
1014 * src/sparc/v8.S (ffi_call_v8): Likewise.
1015 (ffi_closure_v8): Likewise.
1017 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
1018 Steve Ellcey <sje@cup.hp.com>
1020 * configure: Regenerate for new libtool.
1021 * Makefile.in: Ditto.
1022 * include/Makefile.in: Ditto.
1023 * aclocal.m4: Ditto.
1025 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
1027 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
1028 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
1029 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
1030 Adjust copyright notice.
1031 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
1032 register or two register to use for FFI_SYSV structs.
1033 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
1034 (ffi_closure_helper_SYSV): Fix the return type for
1035 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
1036 Adjust copyright notice.
1038 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
1040 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
1043 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1045 * configure: Regenerate.
1046 * include/Makefile.in: Regenerate.
1047 * testsuite/Makefile.in: Regenerate.
1049 2008-06-07 Joseph Myers <joseph@codesourcery.com>
1051 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
1052 powerpc-*-beos*): Remove.
1053 * configure: Regenerate.
1055 2008-05-09 Julian Brown <julian@codesourcery.com>
1057 * Makefile.am (LTLDFLAGS): New.
1058 (libffi_la_LDFLAGS): Use above.
1059 * Makefile.in: Regenerate.
1061 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
1064 * aclocal.m4: Regenerate.
1065 * configure: Regenerate.
1067 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
1069 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
1070 * src/sh64/sysv.S: Likewise.
1072 2008-03-26 Daniel Jacobowitz <dan@debian.org>
1074 * src/arm/sysv.S: Fix ARM comment marker.
1076 2008-03-26 Jakub Jelinek <jakub@redhat.com>
1078 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
1079 * src/s390/sysv.S: Likewise.
1080 * src/powerpc/ppc_closure.S: Likewise.
1081 * src/powerpc/sysv.S: Likewise.
1082 * src/x86/unix64.S: Likewise.
1083 * src/x86/sysv.S: Likewise.
1084 * src/sparc/v8.S: Likewise.
1085 * src/sparc/v9.S: Likewise.
1086 * src/m68k/sysv.S: Likewise.
1087 * src/arm/sysv.S: Likewise.
1089 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1091 * aclocal.m4: Regenerate.
1092 * configure: Likewise.
1093 * Makefile.in: Likewise.
1094 * include/Makefile.in: Likewise.
1095 * testsuite/Makefile.in: Likewise.
1097 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
1098 Andreas Tobler <a.tobler@schweiz.org>
1100 * configure.ac: Add amd64-*-freebsd* target.
1101 * configure: Regenerate.
1103 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
1106 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
1109 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
1112 2008-01-24 David Edelsohn <edelsohn@gnu.org>
1114 * configure: Regenerate.
1116 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
1118 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
1120 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
1123 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
1124 signed/unsigned int8/16 for X86_DARWIN.
1125 Updated copyright info.
1126 Handle one and two byte structs with special cif->flags.
1127 * src/x86/ffitarget.h: Add special types for one and two byte structs.
1128 Updated copyright info.
1129 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
1131 Remove code to pop args from the stack after call.
1132 Special-case signed/unsigned for int8/16, one and two byte structs.
1133 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1134 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1136 Updated copyright info.
1138 2007-12-08 David Daney <ddaney@avtrex.com>
1140 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
1141 SUBU, add with ADDU and use smaller code sequences.
1143 2007-12-07 David Daney <ddaney@avtrex.com>
1145 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
1148 2007-12-06 David Daney <ddaney@avtrex.com>
1150 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
1152 (ffi_java_raw): New typedef.
1153 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
1154 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
1156 (ffi_java_raw_closure) : Same.
1157 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
1159 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
1160 FFI_SIZEOF_JAVA_RAW.
1161 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
1162 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
1163 sizeof(ffi_java_raw) for alignment calculations.
1164 (ffi_java_ptrarray_to_raw): Same.
1165 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
1166 if FFI_SIZEOF_JAVA_RAW == 4.
1167 (ffi_java_raw_to_rvalue): Same.
1168 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
1169 (ffi_java_translate_args): Same.
1170 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
1172 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
1174 2007-12-06 David Daney <ddaney@avtrex.com>
1176 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
1179 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
1182 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
1183 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
1184 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
1185 set the NUM_FPR_ARG_REGISTERS according to.
1186 Add support for potential soft-float support under hard-float
1188 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
1189 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
1190 to the FFI_LINUX_SOFT_FLOAT ABI.
1191 (ffi_prep_cif_machdep): Likewise.
1192 (ffi_closure_helper_SYSV): Likewise.
1193 * src/powerpc/ppc_closure.S: Make sure not to store float/double
1194 on archs where __NO_FPRS__ is true.
1195 Add FFI_TYPE_UINT128 support.
1196 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
1197 Adjust copyright notice.
1199 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
1201 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
1202 * include/ffi_common.h: ... here.
1205 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
1207 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
1208 * src/powerpc/linux64.S: Likewise.
1209 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
1210 * testsuite/libffi.call/return_ldl.c: New test.
1212 2007-09-04 <aph@redhat.com>
1214 * src/arm/sysv.S (UNWIND): New.
1215 (Whole file): Conditionally compile unwinder directives.
1216 * src/arm/sysv.S: Add unwinder directives.
1218 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
1219 Only treat r0 as a struct address if we're actually returning a
1221 Only copy the bytes that are actually within a struct.
1222 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
1223 is returned in r0, not passed by address.
1224 (ffi_call): Allocate a word-sized temporary for the case where
1225 a composite is returned in r0.
1226 (ffi_prep_incoming_args_SYSV): Align as necessary.
1228 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
1230 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
1231 directly using the sys_cacheflush syscall.
1233 2007-07-27 Andrew Haley <aph@redhat.com>
1235 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
1237 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
1239 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
1240 * configure.ac: Likewise.
1241 * Makefile.in: Regenerate.
1242 * include/Makefile.in: Likewise.
1243 * testsuite/Makefile.in: Likewise.
1244 * configure: Likewise.
1246 2007-08-24 David Daney <ddaney@avtrex.com>
1248 * testsuite/libffi.call/return_sl.c: New test.
1250 2007-08-10 David Daney <ddaney@avtrex.com>
1252 * testsuite/libffi.call/cls_multi_ushort.c,
1253 testsuite/libffi.call/cls_align_uint16.c,
1254 testsuite/libffi.call/nested_struct1.c,
1255 testsuite/libffi.call/nested_struct3.c,
1256 testsuite/libffi.call/cls_7_1_byte.c,
1257 testsuite/libffi.call/nested_struct5.c,
1258 testsuite/libffi.call/cls_double.c,
1259 testsuite/libffi.call/nested_struct7.c,
1260 testsuite/libffi.call/cls_sint.c,
1261 testsuite/libffi.call/nested_struct9.c,
1262 testsuite/libffi.call/cls_20byte1.c,
1263 testsuite/libffi.call/cls_multi_sshortchar.c,
1264 testsuite/libffi.call/cls_align_sint64.c,
1265 testsuite/libffi.call/cls_3byte2.c,
1266 testsuite/libffi.call/cls_multi_schar.c,
1267 testsuite/libffi.call/cls_multi_uchar.c,
1268 testsuite/libffi.call/cls_19byte.c,
1269 testsuite/libffi.call/cls_9byte1.c,
1270 testsuite/libffi.call/cls_align_float.c,
1271 testsuite/libffi.call/closure_fn1.c,
1272 testsuite/libffi.call/problem1.c,
1273 testsuite/libffi.call/closure_fn3.c,
1274 testsuite/libffi.call/cls_sshort.c,
1275 testsuite/libffi.call/closure_fn5.c,
1276 testsuite/libffi.call/cls_align_double.c,
1277 testsuite/libffi.call/nested_struct.c,
1278 testsuite/libffi.call/cls_2byte.c,
1279 testsuite/libffi.call/nested_struct10.c,
1280 testsuite/libffi.call/cls_4byte.c,
1281 testsuite/libffi.call/cls_6byte.c,
1282 testsuite/libffi.call/cls_8byte.c,
1283 testsuite/libffi.call/cls_multi_sshort.c,
1284 testsuite/libffi.call/cls_align_sint16.c,
1285 testsuite/libffi.call/cls_align_uint32.c,
1286 testsuite/libffi.call/cls_20byte.c,
1287 testsuite/libffi.call/cls_float.c,
1288 testsuite/libffi.call/nested_struct2.c,
1289 testsuite/libffi.call/cls_5_1_byte.c,
1290 testsuite/libffi.call/nested_struct4.c,
1291 testsuite/libffi.call/cls_24byte.c,
1292 testsuite/libffi.call/nested_struct6.c,
1293 testsuite/libffi.call/cls_64byte.c,
1294 testsuite/libffi.call/nested_struct8.c,
1295 testsuite/libffi.call/cls_uint.c,
1296 testsuite/libffi.call/cls_multi_ushortchar.c,
1297 testsuite/libffi.call/cls_schar.c,
1298 testsuite/libffi.call/cls_uchar.c,
1299 testsuite/libffi.call/cls_align_uint64.c,
1300 testsuite/libffi.call/cls_ulonglong.c,
1301 testsuite/libffi.call/cls_align_longdouble.c,
1302 testsuite/libffi.call/cls_1_1byte.c,
1303 testsuite/libffi.call/cls_12byte.c,
1304 testsuite/libffi.call/cls_3_1byte.c,
1305 testsuite/libffi.call/cls_3byte1.c,
1306 testsuite/libffi.call/cls_4_1byte.c,
1307 testsuite/libffi.call/cls_6_1_byte.c,
1308 testsuite/libffi.call/cls_16byte.c,
1309 testsuite/libffi.call/cls_18byte.c,
1310 testsuite/libffi.call/closure_fn0.c,
1311 testsuite/libffi.call/cls_9byte2.c,
1312 testsuite/libffi.call/closure_fn2.c,
1313 testsuite/libffi.call/closure_fn4.c,
1314 testsuite/libffi.call/cls_ushort.c,
1315 testsuite/libffi.call/closure_fn6.c,
1316 testsuite/libffi.call/cls_5byte.c,
1317 testsuite/libffi.call/cls_align_pointer.c,
1318 testsuite/libffi.call/cls_7byte.c,
1319 testsuite/libffi.call/cls_align_sint32.c,
1320 testsuite/libffi.special/unwindtest_ffi_call.cc,
1321 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
1323 2007-08-10 David Daney <ddaney@avtrex.com>
1326 * configure.ac: Don't treat mips64 as a special case.
1327 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
1328 * configure: Regenerate
1329 * Makefile.in: Ditto.
1330 * fficonfig.h.in: Ditto.
1331 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
1332 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
1333 (FFI_DEFAULT_ABI): Set for n64 case.
1334 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
1335 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
1336 (ffi_closure_N32): New function.
1337 (.eh_frame): New section
1338 * src/mips/o32.S: Clean up comments.
1339 (ffi_closure_O32): Pass ffi_closure parameter in $12.
1340 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
1341 _MIPS_SIM == _ABIN32 throughout.
1342 (FFI_MIPS_STOP_HERE): New, use in place of
1344 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1345 to support n32/n64 ABIs.
1346 (calc_n32_struct_flags): Rewrite.
1347 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1348 position of flag bits.
1349 (ffi_prep_cif_machdep): Rewrite n32 portion.
1350 (ffi_call): Enable for n64. Add special handling for small structure
1352 (ffi_prep_closure_loc): Add n32 and n64 support.
1353 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1354 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1356 2007-08-08 David Daney <ddaney@avtrex.com>
1358 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1359 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1361 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1362 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1363 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1364 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1365 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1366 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1367 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1368 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1369 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1370 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1371 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1372 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1373 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1374 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1375 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1376 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1377 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1378 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1379 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1380 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1381 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1382 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1383 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1384 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1385 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1386 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1387 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1388 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1389 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1390 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1391 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1393 2007-08-07 Andrew Haley <aph@redhat.com>
1395 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1398 2007-08-06 Andrew Haley <aph@redhat.com>
1401 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1402 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1405 2007-08-02 David Daney <ddaney@avtrex.com>
1407 * testsuite/libffi.call/return_ul.c (main): Define return type as
1408 ffi_arg. Use proper printf conversion specifier.
1410 2007-07-30 Andrew Haley <aph@redhat.com>
1413 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1414 signed/unsigned int8/16.
1415 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1417 Remove code to pop args from the stack after call.
1418 Special-case signed/unsigned int8/16.
1419 * testsuite/libffi.call/return_sc.c (main): Revert.
1421 2007-07-26 Richard Guenther <rguenther@suse.de>
1424 * testsuite/libffi.call/return_sc.c (main): Verify call
1425 result as signed char, not ffi_arg.
1427 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1429 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1430 * configure: Regenerate.
1432 2007-07-11 David Daney <ddaney@avtrex.com>
1434 * src/mips/ffi.c: Don't include sys/cachectl.h.
1435 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1438 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1440 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1441 from (ffi_prep_closure): ... this.
1442 (FFI_INIT_TRAMPOLINE): Adjust.
1444 2005-12-31 Phil Blundell <pb@reciva.com>
1446 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1447 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1448 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1449 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1450 (FFI_CLOSURES): Enable closure support.
1452 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1454 * testsuite/libffi.call/cls_multi_ushort.c,
1455 testsuite/libffi.call/cls_align_uint16.c,
1456 testsuite/libffi.call/nested_struct1.c,
1457 testsuite/libffi.call/nested_struct3.c,
1458 testsuite/libffi.call/cls_7_1_byte.c,
1459 testsuite/libffi.call/cls_double.c,
1460 testsuite/libffi.call/nested_struct5.c,
1461 testsuite/libffi.call/nested_struct7.c,
1462 testsuite/libffi.call/cls_sint.c,
1463 testsuite/libffi.call/nested_struct9.c,
1464 testsuite/libffi.call/cls_20byte1.c,
1465 testsuite/libffi.call/cls_multi_sshortchar.c,
1466 testsuite/libffi.call/cls_align_sint64.c,
1467 testsuite/libffi.call/cls_3byte2.c,
1468 testsuite/libffi.call/cls_multi_schar.c,
1469 testsuite/libffi.call/cls_multi_uchar.c,
1470 testsuite/libffi.call/cls_19byte.c,
1471 testsuite/libffi.call/cls_9byte1.c,
1472 testsuite/libffi.call/cls_align_float.c,
1473 testsuite/libffi.call/closure_fn1.c,
1474 testsuite/libffi.call/problem1.c,
1475 testsuite/libffi.call/closure_fn3.c,
1476 testsuite/libffi.call/cls_sshort.c,
1477 testsuite/libffi.call/closure_fn5.c,
1478 testsuite/libffi.call/cls_align_double.c,
1479 testsuite/libffi.call/cls_2byte.c,
1480 testsuite/libffi.call/nested_struct.c,
1481 testsuite/libffi.call/nested_struct10.c,
1482 testsuite/libffi.call/cls_4byte.c,
1483 testsuite/libffi.call/cls_6byte.c,
1484 testsuite/libffi.call/cls_8byte.c,
1485 testsuite/libffi.call/cls_multi_sshort.c,
1486 testsuite/libffi.call/cls_align_uint32.c,
1487 testsuite/libffi.call/cls_align_sint16.c,
1488 testsuite/libffi.call/cls_float.c,
1489 testsuite/libffi.call/cls_20byte.c,
1490 testsuite/libffi.call/cls_5_1_byte.c,
1491 testsuite/libffi.call/nested_struct2.c,
1492 testsuite/libffi.call/cls_24byte.c,
1493 testsuite/libffi.call/nested_struct4.c,
1494 testsuite/libffi.call/nested_struct6.c,
1495 testsuite/libffi.call/cls_64byte.c,
1496 testsuite/libffi.call/nested_struct8.c,
1497 testsuite/libffi.call/cls_uint.c,
1498 testsuite/libffi.call/cls_multi_ushortchar.c,
1499 testsuite/libffi.call/cls_schar.c,
1500 testsuite/libffi.call/cls_uchar.c,
1501 testsuite/libffi.call/cls_align_uint64.c,
1502 testsuite/libffi.call/cls_ulonglong.c,
1503 testsuite/libffi.call/cls_align_longdouble.c,
1504 testsuite/libffi.call/cls_1_1byte.c,
1505 testsuite/libffi.call/cls_12byte.c,
1506 testsuite/libffi.call/cls_3_1byte.c,
1507 testsuite/libffi.call/cls_3byte1.c,
1508 testsuite/libffi.call/cls_4_1byte.c,
1509 testsuite/libffi.call/cls_6_1_byte.c,
1510 testsuite/libffi.call/cls_16byte.c,
1511 testsuite/libffi.call/cls_18byte.c,
1512 testsuite/libffi.call/closure_fn0.c,
1513 testsuite/libffi.call/cls_9byte2.c,
1514 testsuite/libffi.call/closure_fn2.c,
1515 testsuite/libffi.call/closure_fn4.c,
1516 testsuite/libffi.call/cls_ushort.c,
1517 testsuite/libffi.call/closure_fn6.c,
1518 testsuite/libffi.call/cls_5byte.c,
1519 testsuite/libffi.call/cls_align_pointer.c,
1520 testsuite/libffi.call/cls_7byte.c,
1521 testsuite/libffi.call/cls_align_sint32.c,
1522 testsuite/libffi.special/unwindtest_ffi_call.cc,
1523 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1525 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1527 * aclocal.m4: Regenerated.
1529 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1531 * configure: Regenerate.
1533 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1535 * Makefile.in: Regenerate.
1536 * configure: Regenerate.
1537 * aclocal.m4: Regenerate.
1538 * include/Makefile.in: Regenerate.
1539 * testsuite/Makefile.in: Regenerate.
1541 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1543 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1544 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1545 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1546 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1547 (FFI_CLOSURES): Enable closure support.
1549 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1551 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1552 * configure: Regenerate.
1553 * fficonfig.h.in: Regenerate.
1554 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1555 CFI_OFFSET,CFI_DEF_CFA): New macros.
1556 (ffi_call_SYSV): Add callframe annotation.
1558 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1560 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1561 numerous test suite failures.
1562 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1564 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1566 * Makefile.am (EXTRA_DIST): Bring up to date.
1567 * Makefile.in: Regenerate.
1568 * src/frv/eabi.S: Remove RCS keyword.
1570 2007-04-06 Richard Henderson <rth@redhat.com>
1572 * configure.ac: Tidy target case.
1573 (HAVE_LONG_DOUBLE): Allow the target to override.
1574 * configure: Regenerate.
1575 * include/ffi.h.in: Don't define ffi_type_foo if
1576 LIBFFI_HIDE_BASIC_TYPES is defined.
1577 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1579 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1580 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1581 (ffi_type_longdouble): Special case for Alpha. Don't define
1582 if long double == double.
1584 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1585 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1586 (ffi_call, ffi_closure_osf_inner): Likewise.
1587 (ffi_closure_osf_inner): Likewise. Mark hidden.
1588 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1589 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1590 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1591 (load_table): Handle 128-bit long double.
1593 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1595 2007-04-06 Tom Tromey <tromey@redhat.com>
1598 * README: Fixed bug in example.
1600 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1602 * src/closures.c: Include sys/statfs.h.
1603 (_GNU_SOURCE): Define on Linux.
1604 (FFI_MMAP_EXEC_SELINUX): Define.
1605 (selinux_enabled): New variable.
1606 (selinux_enabled_check): New function.
1607 (is_selinux_enabled): Define.
1610 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1612 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1613 Use 'volatile float sum' to create sum of floats to avoid false
1614 negative due to excess precision on ix86 targets.
1617 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1619 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1621 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1623 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1625 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1626 (ffi_prep_closure_loc): New.
1627 (ffi_prep_raw_closure_loc): New.
1628 (ffi_prep_java_raw_closure_loc): New.
1629 * src/closures.c: New file.
1630 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1631 Replace sflags with exec_offset.
1632 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1633 sub_segment_exec_offset): New macros.
1634 (get_segment_flags, set_segment_flags, check_segment_merge): New
1636 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1637 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1638 destroy_mspace): Use new macros.
1639 (sys_alloc): Silence warning.
1640 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1641 * Makefile.in: Rebuilt.
1642 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1643 terms of ffi_prep_closure_loc.
1644 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1646 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1648 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1650 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1651 the renamed version.
1652 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1653 (ffi_prep_closure): ... this.
1654 * src/pa/ffi.c: Likewise.
1655 * src/cris/ffi.c: Likewise. Adjust.
1656 * src/frv/ffi.c: Likewise.
1657 * src/ia64/ffi.c: Likewise.
1658 * src/mips/ffi.c: Likewise.
1659 * src/powerpc/ffi_darwin.c: Likewise.
1660 * src/s390/ffi.c: Likewise.
1661 * src/sh/ffi.c: Likewise.
1662 * src/sh64/ffi.c: Likewise.
1663 * src/sparc/ffi.c: Likewise.
1664 * src/x86/ffi64.c: Likewise.
1665 * src/x86/ffi.c: Likewise.
1666 (FFI_INIT_TRAMPOLINE): Adjust.
1667 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1668 (ffi_prep_raw_closure): ... this.
1669 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1670 (ffi_prep_closure): ... this.
1671 (flush_icache): Adjust.
1673 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1675 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1678 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1680 * Makefile.am: Add dummy install-pdf target.
1681 * Makefile.in: Regenerate
1683 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1685 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1686 ffi_closure_helper_SYSV): Add long double handling.
1688 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1690 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1691 immediately after bctrl instruction.
1693 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1695 * Makefile.am (all-recursive, install-recursive,
1696 mostlyclean-recursive, clean-recursive, distclean-recursive,
1697 maintainer-clean-recursive): Add missing targets.
1698 * Makefile.in: Rebuilt.
1700 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1702 * configure.ac: Add TARGET for x86_64-*-darwin*.
1703 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1705 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1706 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1707 * configure: Regenerate.
1708 * Makefile.in: Regenerate.
1709 * include/Makefile.in: Regenerate.
1710 * testsuite/Makefile.in: Regenerate.
1711 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1714 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1716 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1719 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1721 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1722 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1724 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1725 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1727 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1728 Sandro Tolaini <tolaini@libero.it>
1730 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1732 * configure: Regenerated.
1733 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1734 (EXTRA_DIST): Add src/x86/darwin.S.
1735 * Makefile.in: Regenerated.
1736 * include/Makefile.in: Regenerated.
1737 * testsuite/Makefile.in: Regenerated.
1739 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1740 X86_WIN32, and additionally align stack to 16 bytes.
1741 * src/x86/darwin.S: New, based on sysv.S.
1742 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1744 2006-09-12 David Daney <ddaney@avtrex.com>
1747 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1748 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1749 * aclocal.m4: Regenerated for automake 1.9.6.
1750 * Makefile.in: Regenerated.
1751 * include/Makefile.in: Regenerated.
1752 * testsuite/Makefile.in: Regenerated.
1754 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1756 * include/ffi_common.h (struct): Revert accidental commit.
1758 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1760 * include/ffi_common.h: Remove lint directives.
1761 * include/ffi.h.in: Likewise.
1763 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1765 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1766 for 32-bit architectures.
1767 * testsuite/libffi.call/return_ul.c: New test case.
1769 2006-07-19 David Daney <ddaney@avtrex.com>
1771 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1772 xfail remains for mips64.
1774 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1776 * Makefile.am: Add install-html target. Add install-html to .PHONY
1777 * Makefile.in: Regenerate.
1778 * aclocal.m4: Regenerate.
1779 * include/Makefile.in: Regenerate.
1780 * testsuite/Makefile.in: Regenerate.
1782 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1784 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1787 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1789 * README: Remove notice about 'Crazy Comments'.
1790 * src/debug.c: Remove lint directives. Cleanup white spaces.
1791 * src/java_raw_api.c: Likewise.
1792 * src/prep_cif.c: Likewise.
1793 * src/raw_api.c: Likewise.
1794 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1796 * src/arm/ffi.c: Remove lint directives.
1797 * src/m32r/ffi.c: Likewise.
1798 * src/pa/ffi.c: Likewise.
1799 * src/powerpc/ffi.c: Likewise.
1800 * src/powerpc/ffi_darwin.c: Likewise.
1801 * src/sh/ffi.c: Likewise.
1802 * src/sh64/ffi.c: Likewise.
1803 * src/x86/ffi.c: Likewise.
1804 * testsuite/libffi.call/float2.c: Likewise.
1805 * testsuite/libffi.call/promotion.c: Likewise.
1806 * testsuite/libffi.call/struct1.c: Likewise.
1808 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1810 * src/pa/hpux32.S: Correct unwind offset calculation for
1812 * src/pa/linux.S: Likewise.
1814 2006-04-12 James E Wilson <wilson@specifix.com>
1817 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1818 (hfa_type_load): Call stf_spill.
1819 (hfa_type_store): Call ldf_fill.
1820 (ffi_call): Adjust calls to above routines. Add local temps for
1823 2006-04-10 Matthias Klose <doko@debian.org>
1825 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1826 directory names containing underscores.
1828 2006-04-07 James E Wilson <wilson@specifix.com>
1830 * testsuite/libffi.call/float4.c: New testcase.
1832 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1833 Andreas Tobler <a.tobler@schweiz.ch>
1835 * Makefile.am: Add PA_HPUX port.
1836 * Makefile.in: Regenerate.
1837 * include/Makefile.in: Likewise.
1838 * testsuite/Makefile.in: Likewise.
1839 * configure.ac: Add PA_HPUX rules.
1840 * configure: Regenerate.
1841 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1842 Add PA_HPUX and PA64_HPUX.
1843 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1844 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1845 (FFI_TYPE_SMALL_STRUCT2): Define.
1846 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1847 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1848 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1849 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1850 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1851 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1852 * src/pa/ffi.c (ROUND_DOWN): Delete.
1853 (fldw, fstw, fldd, fstd): Use '__asm__'.
1854 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1855 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1856 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1857 Simplify incrementing of stack slot variable. Change type of local
1858 'n' to unsigned int.
1859 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1861 (ffi_prep_cif_machdep): Likewise.
1862 (ffi_call): Likewise.
1863 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1864 return type to ffi_status. Simplify incrementing of stack slot
1865 variable. Only copy floating point argument registers when PA_LINUX
1866 is true. Reformat debug statement.
1867 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1868 FFI_TYPE_SMALL_STRUCT8.
1869 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1871 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1872 Add nops to cache flush. Add trampoline for PA_HPUX.
1873 * src/pa/hpux32.S: New file.
1874 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1875 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1876 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1877 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1878 argument type checks so that common argument types appear first.
1879 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1880 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1882 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1884 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1885 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1886 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1887 FFI_TYPE_LONGDOUBLE.
1888 (ffi_prep_args64): Assert using IBM extended double.
1889 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1890 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1891 (ffi_call): Handle FFI_LINUX.
1892 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1893 gpr3 return pointer as for struct return. Handle FFI_LINUX
1894 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1896 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1897 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1898 Don't use r6 as pointer to results, instead use sp offset. Don't
1899 make a special call to load lr with case table address, instead
1900 use offset from previous call.
1901 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1902 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1905 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1907 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1908 passed with FP registers correctly.
1909 (ffi_closure_helper_SYSV): Likewise.
1910 * src/sh64/sysv.S: Likewise.
1912 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1914 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1915 args and userdata unused.
1916 (closure_test_fn1): Mark cif and userdata unused.
1917 (main): Remove unused res.
1919 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1921 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1922 -O2, -O3, -Os and the warning flags -W -Wall.
1923 * testsuite/libffi.special/special.exp: Likewise.
1924 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1925 unused parameter unused for gcc or else do nothing.
1926 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1927 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1928 and userdata unused.
1929 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1930 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1931 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1932 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1933 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1934 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1935 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1936 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1937 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1938 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1939 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1940 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1941 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1942 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1943 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1944 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1945 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1946 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1947 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1948 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1949 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1950 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1951 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1952 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1954 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1956 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1958 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1959 void* to avoid compiler warning.
1961 (cls_struct_align_gn): Mark cif and userdata unused.
1962 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1964 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1966 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1968 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1970 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1972 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1973 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1974 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1976 (main): Cast res_call to silence gcc.
1977 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1979 (main): Cast res_call to silence gcc.
1980 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1982 (main): Cast res_call to silence gcc.
1983 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1985 (main): Cast res_call to silence gcc.
1986 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1988 (main): Cast res_call to silence gcc.
1989 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1991 (main): Cast res_call to silence gcc.
1992 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1994 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1995 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1997 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1998 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
2000 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
2001 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
2003 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
2004 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
2006 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
2007 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
2008 and userdata unused.
2009 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
2011 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
2012 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
2013 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
2014 Cleanup white spaces.
2015 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
2016 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
2017 cif and userdata unused.
2018 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
2020 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
2021 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
2022 formatters to silence gcc.
2023 (B_gn): Mark cif and userdata unused.
2024 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
2026 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
2027 (B_gn): Mark cif and userdata unused.
2028 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
2030 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
2031 (B_gn): Mark cif and userdata unused.
2032 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
2034 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
2035 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
2036 * testsuite/libffi.call/problem1.c (stub): Likewise.
2037 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
2039 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
2040 in the last commit for this test case in the test case itself.
2041 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
2043 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
2044 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
2045 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
2046 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
2047 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
2048 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
2050 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
2052 * src/sh/sysv.S: Fix register numbers in the FDE for
2055 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
2057 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
2058 declaration to avoid a false negative on ix86. See PR323.
2060 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
2062 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
2063 and cast integer to void * if needed. Update the pointer to
2064 the FP register saved area correctly.
2066 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
2068 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
2070 * testsuite/libffi.call/nested_struct4.c: Likewise.
2072 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
2074 * testsuite/libffi.call/return_dbl.c: New test case.
2075 * testsuite/libffi.call/return_dbl1.c: Likewise.
2076 * testsuite/libffi.call/return_dbl2.c: Likewise.
2077 * testsuite/libffi.call/return_fl.c: Likewise.
2078 * testsuite/libffi.call/return_fl1.c: Likewise.
2079 * testsuite/libffi.call/return_fl2.c: Likewise.
2080 * testsuite/libffi.call/return_fl3.c: Likewise.
2081 * testsuite/libffi.call/closure_fn6.c: Likewise.
2083 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
2085 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
2086 here to be used by other test cases too.
2088 * testsuite/libffi.call/nested_struct10.c: New test case.
2089 * testsuite/libffi.call/nested_struct9.c: Likewise.
2090 * testsuite/libffi.call/nested_struct8.c: Likewise.
2091 * testsuite/libffi.call/nested_struct7.c: Likewise.
2092 * testsuite/libffi.call/nested_struct6.c: Likewise.
2093 * testsuite/libffi.call/nested_struct5.c: Likewise.
2094 * testsuite/libffi.call/nested_struct4.c: Likewise.
2096 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2098 * configure.ac: Enable libffi for sparc64-*-freebsd*.
2099 * configure: Rebuilt.
2101 2006-01-18 Jakub Jelinek <jakub@redhat.com>
2103 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
2104 instead do the shifting inline.
2105 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
2106 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
2107 and 8 byte structs, for the remaining struct sizes don't call
2108 __lshrdi3, instead do the shifting inline.
2110 2005-12-07 Thiemo Seufer <ths@networkno.de>
2112 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
2113 missing parentheses.
2114 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
2115 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
2116 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
2117 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
2118 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
2120 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
2122 (ffi_prep_closure): Improve trampoline instruction scheduling.
2123 (ffi_closure_mips_inner_O32): Fix endianness bugs.
2125 2005-12-03 Alan Modra <amodra@bigpond.net.au>
2127 * src/powerpc/ffi.c: Formatting.
2128 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
2129 (ffi_prep_args64): Likewise.
2131 2005-09-30 Geoffrey Keating <geoffk@apple.com>
2133 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
2134 darwin, use -shared-libgcc not -lgcc_s, and explain why.
2136 2005-09-26 Tom Tromey <tromey@redhat.com>
2138 * testsuite/libffi.call/float1.c (value_type): New typedef.
2139 (CANARY): New define.
2140 (main): Check for result buffer overflow.
2141 * src/powerpc/linux64.S: Handle linux64 long double returns.
2142 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
2143 (ffi_prep_cif_machdep): Handle linux64 long double returns.
2145 2005-08-25 Alan Modra <amodra@bigpond.net.au>
2148 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
2150 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
2152 2005-08-11 Jakub Jelinek <jakub@redhat.com>
2154 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
2155 (AH_BOTTOM): Add FFI_HIDDEN definition.
2156 * configure: Rebuilt.
2157 * fficonfig.h.in: Rebuilt.
2158 * src/powerpc/ffi.c (hidden): Remove.
2159 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
2160 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
2161 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
2162 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
2163 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
2164 add FFI_HIDDEN to its prototype.
2165 (ffi_closure_SYSV_inner): New.
2166 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2167 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2169 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
2172 * configure: Rebuilt.
2173 * configure.ac: Handle i*86-*-gnu*.
2175 2005-08-09 Jakub Jelinek <jakub@redhat.com>
2177 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
2178 DW_CFA_offset_extended_sf rather than
2179 DW_CFA_GNU_negative_offset_extended.
2180 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
2182 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
2184 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
2186 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
2187 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
2188 partially on register.
2189 (ffi_closure_helper_SYSV): Likewise.
2190 (ffi_prep_cif_machdep): Don't set too many cif->flags.
2192 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
2194 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
2196 * src/sh64/ffi.c (simple_type): Remove.
2197 (return_type): Handle small structures correctly.
2198 (ffi_prep_args): Likewise.
2199 (ffi_call): Likewise.
2200 (ffi_closure_helper_SYSV): Likewise.
2201 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
2202 Emit position independent code if PIC and remove wrong datalabel
2203 prefixes from EH data.
2205 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
2207 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
2208 * Makefile.in: Regenerate.
2209 * include/Makefile.in: Likewise.
2210 * testsuite/Makefile.in: Likewise.
2211 * configure.ac: Add POWERPC_FREEBSD rules.
2212 * configure: Regenerate.
2213 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
2214 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
2215 * src/powerpc/ffi.c: Add flags to handle small structure returns
2217 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
2219 (ffi_closure_helper_SYSV): Likewise.
2220 * src/powerpc/ppc_closure.S: Add return types for small structures.
2221 * src/powerpc/sysv.S: Add bits to handle small structures for
2224 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
2226 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
2227 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
2228 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
2230 2005-07-05 Randolph Chung <tausq@debian.org>
2232 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
2233 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
2234 structures. Kill compilation warnings.
2235 (ffi_closure_inner_LINUX): Print return values as hex in debug
2236 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
2237 Properly handle 5-7 byte structure returns.
2238 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
2239 (FFI_TYPE_SMALL_STRUCT2): Remove.
2240 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
2241 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
2242 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
2243 (checksmst1, checksmst2): Remove.
2244 (checksmst3): Optimize handling of 3-byte struct returns.
2245 (checksmst567): Properly handle 5-7 byte struct returns.
2247 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2250 * src/mips/n32.S: Enforce PIC code.
2251 * src/mips/o32.S: Likewise.
2253 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2255 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
2256 * configure: Regenerate.
2258 2005-06-01 Alan Modra <amodra@bigpond.net.au>
2260 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
2261 to call ffi_closure_helper_SYSV. Append @local instead.
2262 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
2264 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
2266 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
2267 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
2268 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
2269 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
2270 include/Makefile.in, testsuite/Makefile.in: Regenerate.
2272 2005-05-09 Mike Stump <mrs@apple.com>
2274 * configure: Regenerate.
2276 2005-05-08 Richard Henderson <rth@redhat.com>
2279 * src/alpha/osf.S: Update unwind into to match code.
2281 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
2282 Richard Henderson <rth@redhat.com>
2284 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
2286 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
2287 (ffi_prep_closure): Set carry bit if sse-used flag set.
2288 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
2289 Only load sse registers if ssecount non-zero.
2290 (ffi_closure_unix64): Only save sse registers if carry set on entry.
2292 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
2294 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
2295 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
2296 * configure: Regenerate.
2298 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
2300 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
2301 have Tcl8.3-compatible intermediate variable.
2303 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
2304 Hans-Peter Nilsson <hp@axis.com>
2306 * Makefile.am: Add CRIS support.
2307 * configure.ac: Likewise.
2308 * Makefile.in, configure, testsuite/Makefile.in,
2309 include/Makefile.in: Regenerate.
2310 * src/cris: New directory.
2311 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
2312 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
2314 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
2315 \r?\n in output tests.
2317 2005-04-12 Mike Stump <mrs@apple.com>
2319 * configure: Regenerate.
2321 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
2323 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
2325 2005-03-30 Steve Ellcey <sje@cup.hp.com>
2327 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
2329 * src/ia64/unix.S (ffi_closure_unix): Extend gp
2330 to 64 bits in ILP32 mode.
2331 Load 64 bits even for short data.
2333 2005-03-23 Mike Stump <mrs@apple.com>
2335 * src/powerpc/darwin.S: Update for -m64 multilib.
2336 * src/powerpc/darwin_closure.S: Likewise.
2338 2005-03-21 Zack Weinberg <zack@codesourcery.com>
2340 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
2341 Do not set tool_include_dir.
2342 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
2344 * include/Makefile.am: Set gcc_version and toollibffidir.
2345 * include/Makefile.in: Regenerate.
2347 2005-02-22 Andrew Haley <aph@redhat.com>
2349 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2350 odd-numbered register pairs for 64-bit integer types.
2352 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2355 * testsuite/libffi.call/return_ll1.c: New test case.
2357 2005-02-11 Janis Johnson <janis187@us.ibm.com>
2359 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2360 * testsuite/libffi.call/float.c: Ditto.
2361 * testsuite/libffi.call/float2.c: Ditto.
2362 * testsuite/libffi.call/float3.c: Ditto.
2364 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2366 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2368 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2370 * testsuite/libffi.special/special.exp (cxx_options): Add
2373 2004-12-31 Richard Henderson <rth@redhat.com>
2375 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2376 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2377 offset parameters with a type parameter; deduce size and structure
2378 alignment. Update all users.
2380 2004-12-31 Richard Henderson <rth@redhat.com>
2382 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2383 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
2384 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
2385 into ffi_prep_closure.
2386 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2389 2004-12-27 Richard Henderson <rth@redhat.com>
2391 * src/x86/unix64.S: Fix typo in unwind info.
2393 2004-12-25 Richard Henderson <rth@redhat.com>
2395 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2396 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2397 (merge_classes): Check for it.
2399 (classify_argument): Pass byte_offset by value; perform all updates
2401 (examine_argument): Add classes argument; handle
2402 X86_64_COMPLEX_X87_CLASS.
2403 (ffi_prep_args): Merge into ...
2404 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2405 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2406 (ffi_fill_return_value): Remove.
2407 (ffi_prep_closure): Remove dead assert.
2408 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2409 Rewrite to use struct register_args instead of va_list. Create
2410 flags for handling structure returns.
2411 * src/x86/unix64.S: Remove dead strings.
2412 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2413 stack frame with ffi_call. Handle structure returns properly.
2414 (float2sse, floatfloat2sse, double2sse): Remove.
2415 (sse2float, sse2double, sse2floatfloat): Remove.
2416 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2417 to handle structure returns properly.
2419 2004-12-08 David Edelsohn <edelsohn@gnu.org>
2421 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2423 * Makefile.in: Regenerated.
2425 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2427 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2428 * configure, aclocal.m4, Makefile.in: Regenerate.
2429 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2431 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2433 * configure: Regenerate for libtool change.
2435 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2437 * configure: Regenerate for libtool reversion.
2439 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2441 * configure: Regenerate for libtool change.
2443 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2445 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2447 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2449 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2450 of jal. Use an absolute encoding for the frame information.
2452 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2454 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2455 * acinclude.m4: Delete logic for sincludes.
2456 * aclocal.m4, Makefile.in, configure: Regenerate.
2457 * include/Makefile: Likewise.
2458 * testsuite/Makefile: Likewise.
2460 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2462 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2463 on a 8-byte boundary.
2464 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2466 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2468 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2469 long long values. Round stack allocation to a multiple of 8 bytes
2470 for ATPCS compatibility.
2471 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2472 names. Handle returning long long types. Add Thumb and interworking
2473 support. Improve soft-float code.
2475 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2477 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2478 (libffi_exit): New function.
2479 (libffi_init): Build the testglue wrapper if needed.
2481 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2484 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2486 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2488 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2490 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2492 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2493 * testsuite/libffi.call/float3.c: New test case.
2495 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2497 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2498 the function returning a structure pointed with R2.
2499 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2500 the structure return value if T bit set. Emit position
2501 independent code and EH data if PIC.
2503 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2505 * Makefile.am: Add m32r support.
2506 * configure.ac: Likewise.
2507 * Makefile.in: Regenerate.
2508 * confiugre: Regenerate.
2509 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2510 (uint64, sint64, double, longdouble)
2511 * src/m32r: New directory.
2512 * src/m32r/ffi.c: New file.
2513 * src/m32r/sysv.S: Likewise.
2514 * src/m32r/ffitarget.h: Likewise.
2516 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2518 * testsuite/libffi.call/negint.c: New test case.
2520 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2523 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2524 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2525 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2528 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2530 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2531 * testsuite/libffi.call/promotion.c: Likewise.
2532 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2534 * testsuite/libffi.call/return_sc.c: Likewise.
2535 * testsuite/libffi.call/return_uc.c: Likewise.
2537 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2539 * src/powerpc/darwin.S: Fix comments and identation.
2540 * src/powerpc/darwin_closure.S: Likewise.
2542 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2544 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2545 (ffi_prep_args): Handle longdouble arguments.
2546 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2548 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2549 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2551 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2552 * src/types.c: Defined longdouble size and alignment for darwin.
2554 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2556 * src/powerpc/aix.S: Remove whitespaces.
2557 * src/powerpc/aix_closure.S: Likewise.
2558 * src/powerpc/asm.h: Likewise.
2559 * src/powerpc/ffi.c: Likewise.
2560 * src/powerpc/ffitarget.h: Likewise.
2561 * src/powerpc/linux64.S: Likewise.
2562 * src/powerpc/linux64_closure.S: Likewise.
2563 * src/powerpc/ppc_closure.S: Likewise.
2564 * src/powerpc/sysv.S: Likewise.
2566 2004-08-30 Anthony Green <green@redhat.com>
2568 * Makefile.am: Add frv support.
2569 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2570 * configure.ac: Read configure.host.
2571 * configure.in: Read configure.host.
2572 * configure.host: New file. frv-elf needs libgloss.
2573 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2574 alignment. This is needed to frv and shouldn't harm the others.
2575 * include/ffi_common.h (ALIGN_DOWN): New macro.
2576 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2578 2004-08-24 David Daney <daney@avtrex.com>
2580 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2581 * testsuite/libffi.call/closure_fn1.c: Likewise.
2582 * testsuite/libffi.call/closure_fn2.c Likewise.
2583 * testsuite/libffi.call/closure_fn3.c: Likewise.
2584 * testsuite/libffi.call/closure_fn4.c: Likewise.
2585 * testsuite/libffi.call/closure_fn5.c: Likewise.
2586 * testsuite/libffi.call/cls_18byte.c: Likewise.
2587 * testsuite/libffi.call/cls_19byte.c: Likewise.
2588 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2589 * testsuite/libffi.call/cls_20byte.c: Likewise.
2590 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2591 * testsuite/libffi.call/cls_24byte.c: Likewise.
2592 * testsuite/libffi.call/cls_2byte.c: Likewise.
2593 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2594 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2595 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2596 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2597 * testsuite/libffi.call/cls_4byte.c: Likewise.
2598 * testsuite/libffi.call/cls_64byte.c: Likewise.
2599 * testsuite/libffi.call/cls_6byte.c: Likewise.
2600 * testsuite/libffi.call/cls_7byte.c: Likewise.
2601 * testsuite/libffi.call/cls_8byte.c: Likewise.
2602 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2603 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2604 * testsuite/libffi.call/cls_align_double.c: Likewise.
2605 * testsuite/libffi.call/cls_align_float.c: Likewise.
2606 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2607 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2608 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2609 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2610 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2611 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2612 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2613 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2614 * testsuite/libffi.call/cls_double.c: Likewise.
2615 * testsuite/libffi.call/cls_float.c: Likewise.
2616 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2617 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2618 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2619 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2620 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2621 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2622 * testsuite/libffi.call/cls_schar.c: Likewise.
2623 * testsuite/libffi.call/cls_sint.c: Likewise.
2624 * testsuite/libffi.call/cls_sshort.c: Likewise.
2625 * testsuite/libffi.call/cls_uchar.c: Likewise.
2626 * testsuite/libffi.call/cls_uint.c: Likewise.
2627 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2628 * testsuite/libffi.call/cls_ushort.c: Likewise.
2629 * testsuite/libffi.call/nested_struct.c: Likewise.
2630 * testsuite/libffi.call/nested_struct1.c: Likewise.
2631 * testsuite/libffi.call/nested_struct2.c: Likewise.
2632 * testsuite/libffi.call/nested_struct3.c: Likewise.
2633 * testsuite/libffi.call/problem1.c: Likewise.
2634 * testsuite/libffi.special/unwindtest.cc: Likewise.
2635 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2637 * testsuite/libffi.call/cls_16byte.c: Likewise.
2638 * testsuite/libffi.call/cls_5byte.c: Likewise.
2640 2004-08-23 David Daney <daney@avtrex.com>
2643 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2644 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2645 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2646 parameters and return types.
2647 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2648 (ffi_prep_closure): Ditto.
2649 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2650 alignment calculations.
2651 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2652 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2654 2004-08-14 Casey Marshall <csm@gnu.org>
2656 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2657 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2658 integer (O32 ABI only).
2659 (ffi_prep_closure): new function.
2660 (ffi_closure_mips_inner_O32): new function.
2661 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2662 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2663 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2664 64 bit integers correctly.
2665 (ffi_closure_O32): new function.
2666 Added DWARF-2 unwind info for both functions.
2668 2004-08-10 Andrew Haley <aph@redhat.com>
2670 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2672 2004-08-01 Robert Millan <robertmh@gnu.org>
2674 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2675 * configure: Regenerate.
2677 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2679 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2680 and mmap() explicitly instead of relying on preset autoconf cache
2682 * aclocal.m4: Regenerate.
2683 * configure: Regenerate.
2685 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2687 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2688 (ffi_check_float_struct): Remove unused prototype.
2690 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2692 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2693 character on Darwin, use '\n\t' instead.
2695 2004-06-26 Matthias Klose <doko@debian.org>
2697 * libtool-version: Fix typo in revision/age.
2699 2004-06-17 Matthias Klose <doko@debian.org>
2701 * libtool-version: New.
2702 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2703 * Makefile.in: Regenerate.
2705 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2707 * Makefile.am: Remove useless multilib rules.
2708 * Makefile.in: Regenerate.
2709 * aclocal.m4: Regenerate with automake 1.8.5.
2710 * configure.ac: Remove useless multilib configury.
2711 * configure: Regenerate.
2713 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2715 * .cvsignore: New file.
2717 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2719 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2721 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2724 2004-06-06 Sean McNeil <sean@mcneil.com>
2726 * configure.ac: Add x86_64-*-freebsd* support.
2727 * configure: Regenerate.
2729 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2732 * configure.ac: Test for existence of mmap and sys/mman.h before
2733 checking blacklist. Fix suggested by Jim Wilson.
2734 * configure: Regenerate.
2736 2004-04-26 Matt Austern <austern@apple.com>
2738 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2740 * src/powerpc/darwin_closure.S: Likewise.
2742 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2744 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2745 error. Reported by Thomas Heller <theller@python.net>.
2746 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2747 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2749 2004-03-20 Matthias Klose <doko@debian.org>
2751 * src/pa/linux.S: Fix typo.
2753 2004-03-19 Matthias Klose <doko@debian.org>
2755 * Makefile.am: Update.
2756 * Makefile.in: Regenerate.
2757 * src/pa/ffi.h.in: Remove.
2758 * src/pa/ffitarget.h: New file.
2760 2004-02-10 Randolph Chung <tausq@debian.org>
2762 * Makefile.am: Add PA support.
2763 * Makefile.in: Regenerate.
2764 * include/Makefile.in: Regenerate.
2765 * configure.ac: Add PA target.
2766 * configure: Regenerate.
2767 * src/pa/ffi.c: New file.
2768 * src/pa/ffi.h.in: Add PA support.
2769 * src/pa/linux.S: New file.
2770 * prep_cif.c: Add PA support.
2772 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2774 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2776 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2777 with ecif->cif->flags.
2778 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2780 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2781 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2782 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2783 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2785 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2787 * configure.in: Rename file to ...
2788 * configure.ac: ... this.
2789 * fficonfig.h.in: Regenerate.
2790 * Makefile.in: Regenerate.
2791 * include/Makefile.in: Regenerate.
2792 * testsuite/Makefile.in: Regenerate.
2794 2004-03-12 Matt Austern <austern@apple.com>
2796 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2797 changes in EH format resulting from addition of linkonce support.
2798 * src/powerpc/darwin_closure.S: Likewise.
2800 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2801 Paolo Bonzini <bonzini@gnu.org>
2803 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2804 Remove VPATH. Remove rules for object files. Remove multilib support.
2805 (AM_CCASFLAGS): Add.
2806 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2807 (AC_PREREQ): Bump version to 2.59.
2808 (AC_INIT): Fill with version info and bug address.
2809 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2810 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2811 De-precious CC so that the right flags are passed down to multilibs.
2812 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2813 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2814 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2815 * configure: Rebuilt.
2816 * aclocal.m4: Likewise.
2817 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2818 * fficonfig.h.in: Likewise.
2820 2004-03-11 Andreas Schwab <schwab@suse.de>
2822 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2823 arguments from fp registers only for the first 8 parameter slots.
2824 Don't convert a float parameter when passed in memory.
2826 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2828 * configure: Regenerate for config/accross.m4 correction.
2830 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2832 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2833 ecif->cif->bytes to bytes.
2834 (ffi_prep_cif_machdep): Add braces around nested if statement.
2836 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2838 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2840 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2841 (ffi_closure_helper_LINUX64): Fix typo.
2842 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2844 * testsuite/libffi.call/float.c: Likewise.
2845 * testsuite/libffi.call/float2.c: Likewise.
2847 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2849 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2850 long double function return and long double arg handling.
2851 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2852 Use "end_pfr" instead of "nf". Correct long double handling.
2854 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2856 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2857 space for long double return value. Adjust stack frame and offsets.
2858 Load f2 long double return.
2860 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2862 * src/types.c: Use 16 byte long double for POWERPC64.
2864 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2866 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2867 when the structure return address is passed in %o0.
2868 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2869 (ffi_v9_layout_struct): Align the field following a nested structure
2870 on a word boundary. Use memmove instead of memcpy.
2871 (ffi_call): Update call to ffi_V9_return_struct.
2872 (ffi_prep_closure): Define 'ctx' only for V8.
2873 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2874 and ffi_closure_sparc_inner_v9.
2875 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2876 Always skip the structure return address. For structures and long
2877 doubles, copy the argument directly.
2878 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2879 if required. Shift the maximum floating-point slot accordingly. For
2880 big structures, copy the argument directly; otherwise, left-justify the
2881 argument and call ffi_v9_layout_struct to lay out the structure on
2883 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2884 (ffi_closure_v8): Pass the structure return address. Update call to
2885 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2886 Skip the 'unimp' insn when returning long doubles and structures.
2887 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2888 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2889 FFI_TYPE_INT handling. Load structures both in integers and
2890 floating-point registers on return.
2891 * README: Update status of the SPARC port.
2893 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2895 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2897 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2899 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2901 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2902 value as of type ffi_arg, not unsigned int.
2904 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2906 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2909 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2911 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2914 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2916 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2919 2003-12-10 Richard Henderson <rth@redhat.com>
2921 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2922 size_t instead of int.
2924 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2926 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2927 * testsuite/libffi.call/many_win32.c (main): Replace variable
2928 int i with unsigned long ul.
2930 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2931 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2932 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2933 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2934 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2935 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2936 * testsuite/libffi.call/cls_align_float.c: Likewise.
2937 * testsuite/libffi.call/cls_align_double.c: Likewise.
2938 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2939 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2941 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2944 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2945 Align arguments to 32 bits.
2947 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2950 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2951 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2952 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2953 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2954 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2955 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2957 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2959 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2961 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2962 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2964 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2966 * Makefile.in: Rebuilt.
2967 * configure: Likewise.
2968 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2971 2003-11-21 Andreas Jaeger <aj@suse.de>
2972 Andreas Tobler <a.tobler@schweiz.ch>
2974 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2975 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2976 * Makefile.in: Rebuilt.
2977 * aclocal.m4: Likewise.
2978 * configure: Likewise.
2979 * fficonfig.h.in: Likewise.
2980 * testsuite/lib/libffi-dg.exp: Add include dir.
2981 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2982 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2983 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2984 for ffi_closure if available.
2985 * testsuite/libffi.call/closure_fn1.c: Likewise.
2986 * testsuite/libffi.call/closure_fn2.c: Likewise.
2987 * testsuite/libffi.call/closure_fn3.c: Likewise.
2988 * testsuite/libffi.call/closure_fn4.c: Likewise.
2989 * testsuite/libffi.call/closure_fn5.c: Likewise.
2990 * testsuite/libffi.call/cls_12byte.c: Likewise.
2991 * testsuite/libffi.call/cls_16byte.c: Likewise.
2992 * testsuite/libffi.call/cls_18byte.c: Likewise.
2993 * testsuite/libffi.call/cls_19byte.c: Likewise.
2994 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2995 * testsuite/libffi.call/cls_20byte.c: Likewise.
2996 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2997 * testsuite/libffi.call/cls_24byte.c: Likewise.
2998 * testsuite/libffi.call/cls_2byte.c: Likewise.
2999 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3000 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3001 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3002 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3003 * testsuite/libffi.call/cls_4byte.c: Likewise.
3004 * testsuite/libffi.call/cls_5byte.c: Likewise.
3005 * testsuite/libffi.call/cls_64byte.c: Likewise.
3006 * testsuite/libffi.call/cls_6byte.c: Likewise.
3007 * testsuite/libffi.call/cls_7byte.c: Likewise.
3008 * testsuite/libffi.call/cls_8byte.c: Likewise.
3009 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3010 * testsuite/libffi.call/cls_9byte2.c: Likewise.
3011 * testsuite/libffi.call/cls_double.c: Likewise.
3012 * testsuite/libffi.call/cls_float.c: Likewise.
3013 * testsuite/libffi.call/cls_schar.c: Likewise.
3014 * testsuite/libffi.call/cls_sint.c: Likewise.
3015 * testsuite/libffi.call/cls_sshort.c: Likewise.
3016 * testsuite/libffi.call/cls_uchar.c: Likewise.
3017 * testsuite/libffi.call/cls_uint.c: Likewise.
3018 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3019 * testsuite/libffi.call/cls_ushort.c: Likewise.
3020 * testsuite/libffi.call/nested_struct.c: Likewise.
3021 * testsuite/libffi.call/nested_struct1.c: Likewise.
3022 * testsuite/libffi.call/nested_struct2.c: Likewise.
3023 * testsuite/libffi.call/nested_struct3.c: Likewise.
3024 * testsuite/libffi.call/problem1.c: Likewise.
3025 * testsuite/libffi.special/unwindtest.cc: Likewise.
3027 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
3029 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
3031 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
3033 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
3034 Add -lgcc_s to additional flags.
3036 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
3038 * configure.in, include/Makefile.am: PR libgcj/11147, install
3039 the ffitarget.h header file in a gcc versioned and target
3041 * configure: Regenerated.
3042 * Makefile.in, include/Makefile.in: Likewise.
3043 * testsuite/Makefile.in: Likewise.
3045 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
3047 * testsuite/libffi.call/closure_fn0.c: Print result and check
3048 with dg-output to make debugging easier.
3049 * testsuite/libffi.call/closure_fn1.c: Likewise.
3050 * testsuite/libffi.call/closure_fn2.c: Likewise.
3051 * testsuite/libffi.call/closure_fn3.c: Likewise.
3052 * testsuite/libffi.call/closure_fn4.c: Likewise.
3053 * testsuite/libffi.call/closure_fn5.c: Likewise.
3054 * testsuite/libffi.call/cls_12byte.c: Likewise.
3055 * testsuite/libffi.call/cls_16byte.c: Likewise.
3056 * testsuite/libffi.call/cls_18byte.c: Likewise.
3057 * testsuite/libffi.call/cls_19byte.c: Likewise.
3058 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3059 * testsuite/libffi.call/cls_20byte.c: Likewise.
3060 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3061 * testsuite/libffi.call/cls_24byte.c: Likewise.
3062 * testsuite/libffi.call/cls_2byte.c: Likewise.
3063 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3064 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3065 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3066 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3067 * testsuite/libffi.call/cls_4byte.c: Likewise.
3068 * testsuite/libffi.call/cls_5byte.c: Likewise.
3069 * testsuite/libffi.call/cls_64byte.c: Likewise.
3070 * testsuite/libffi.call/cls_6byte.c: Likewise.
3071 * testsuite/libffi.call/cls_7byte.c: Likewise.
3072 * testsuite/libffi.call/cls_8byte.c: Likewise.
3073 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3074 * testsuite/libffi.call/cls_9byte2.c: Likewise.
3075 * testsuite/libffi.call/cls_double.c: Likewise.
3076 * testsuite/libffi.call/cls_float.c: Likewise.
3077 * testsuite/libffi.call/cls_schar.c: Likewise.
3078 * testsuite/libffi.call/cls_sint.c: Likewise.
3079 * testsuite/libffi.call/cls_sshort.c: Likewise.
3080 * testsuite/libffi.call/cls_uchar.c: Likewise.
3081 * testsuite/libffi.call/cls_uint.c: Likewise.
3082 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3083 * testsuite/libffi.call/cls_ushort.c: Likewise.
3084 * testsuite/libffi.call/problem1.c: Likewise.
3086 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
3089 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
3091 * testsuite/libffi.call/cls_9byte2.c: New test case.
3092 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3093 * testsuite/libffi.call/cls_64byte.c: Likewise.
3094 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3095 * testsuite/libffi.call/cls_19byte.c: Likewise.
3096 * testsuite/libffi.call/cls_18byte.c: Likewise.
3097 * testsuite/libffi.call/closure_fn4.c: Likewise.
3098 * testsuite/libffi.call/closure_fn5.c: Likewise.
3099 * testsuite/libffi.call/cls_schar.c: Likewise.
3100 * testsuite/libffi.call/cls_sint.c: Likewise.
3101 * testsuite/libffi.call/cls_sshort.c: Likewise.
3102 * testsuite/libffi.call/nested_struct2.c: Likewise.
3103 * testsuite/libffi.call/nested_struct3.c: Likewise.
3105 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
3107 * testsuite/libffi.call/cls_double.c: Do a check on the result.
3108 * testsuite/libffi.call/cls_uchar.c: Likewise.
3109 * testsuite/libffi.call/cls_uint.c: Likewise.
3110 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3111 * testsuite/libffi.call/cls_ushort.c: Likewise.
3112 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
3114 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
3116 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
3119 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
3121 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
3122 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
3124 2003-10-22 David Daney <ddaney@avtrex.com>
3126 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
3127 __attribute__((__mode__(__SI__))) and friends.
3129 2003-10-22 Andreas Schwab <schwab@suse.de>
3131 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
3133 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
3135 * configure.in: AC_LINK_FILES(ffitarget.h).
3136 * configure: Regenerate.
3137 * Makefile.in: Likewise.
3138 * include/Makefile.in: Likewise.
3139 * testsuite/Makefile.in: Likewise.
3140 * fficonfig.h.in: Likewise.
3142 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
3143 Richard Henderson <rth@redhat.com>
3145 Avoid that ffi.h includes fficonfig.h.
3147 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
3148 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
3149 (TARGET_SRC_MIPS_SGI): Removed.
3150 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
3151 (MIPS_SGI): Removed.
3152 (CLEANFILES): Removed.
3153 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
3155 * acconfig.h: Removed.
3156 * configure.in: Compute sizeofs only for double and long double.
3157 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
3158 into AC_DEFINE instead of using acconfig.h. Create
3159 include/ffitarget.h instead of include/fficonfig.h. Rename
3160 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
3161 AC_DEFINE EH_FRAME_FLAGS.
3162 * include/Makefile.am (DISTCLEANFILES): New automake macro.
3163 (hack_DATA): Add ffitarget.h.
3164 * include/ffi.h.in: Remove all system specific definitions.
3165 Declare raw API even if it is not installed, why bother?
3166 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
3167 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
3168 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
3169 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
3170 * include/ffi_common.h (bool): Do not define.
3171 (ffi_assert): Accept failed assertion.
3172 (ffi_type_test): Return void and accept file/line.
3173 (FFI_ASSERT): Pass stringized failed assertion.
3174 (FFI_ASSERT_AT): New macro.
3175 (FFI_ASSERT_VALID_TYPE): New macro.
3176 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
3177 UINT64, SINT64): Define here with gcc's __attribute__ macro
3179 (FLOAT32, ALIGN): Define here instead of in ffi.h
3180 * include/ffi-mips.h: Removed. Its content moved to
3181 src/mips/ffitarget.h after separating assembly and C sections.
3182 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
3183 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
3184 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
3185 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
3186 SIZEOF_ARG -> FFI_SIZEOF_ARG.
3187 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
3188 * src/debug.c (ffi_assert): Accept stringized failed assertion.
3189 (ffi_type_test): Rewritten.
3190 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
3191 FFI_ASSERT_VALID_TYPE.
3192 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
3193 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
3194 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
3195 src/s390/ffitarget.h, src/sh/ffitarget.h,
3196 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
3197 src/x86/ffitarget.h: New files.
3198 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
3199 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
3200 src/powerpc/aix.S, src/powerpc/darwin.S,
3201 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
3202 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
3203 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
3204 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
3205 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
3208 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3210 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
3211 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
3213 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
3215 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
3216 Used when FFI_DEBUG = 1.
3218 2003-10-14 Alan Modra <amodra@bigpond.net.au>
3220 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
3223 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3225 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
3226 FFI_MIPS_O32 for O32 ABI.
3228 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
3230 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
3231 SPARC64. Cleanup whitespaces.
3233 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
3235 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
3236 strongarm, xscale. Cleanup whitespaces.
3237 * testsuite/libffi.call/closure_fn1.c: Likewise.
3238 * testsuite/libffi.call/closure_fn2.c: Likewise.
3239 * testsuite/libffi.call/closure_fn3.c: Likewise.
3240 * testsuite/libffi.call/cls_12byte.c: Likewise.
3241 * testsuite/libffi.call/cls_16byte.c: Likewise.
3242 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3243 * testsuite/libffi.call/cls_20byte.c: Likewise.
3244 * testsuite/libffi.call/cls_24byte.c: Likewise.
3245 * testsuite/libffi.call/cls_2byte.c: Likewise.
3246 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3247 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3248 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3249 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3250 * testsuite/libffi.call/cls_4byte.c: Likewise.
3251 * testsuite/libffi.call/cls_5byte.c: Likewise.
3252 * testsuite/libffi.call/cls_6byte.c: Likewise.
3253 * testsuite/libffi.call/cls_7byte.c: Likewise.
3254 * testsuite/libffi.call/cls_8byte.c: Likewise.
3255 * testsuite/libffi.call/cls_double.c: Likewise.
3256 * testsuite/libffi.call/cls_float.c: Likewise.
3257 * testsuite/libffi.call/cls_uchar.c: Likewise.
3258 * testsuite/libffi.call/cls_uint.c: Likewise.
3259 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3260 * testsuite/libffi.call/cls_ushort.c: Likewise.
3261 * testsuite/libffi.call/nested_struct.c: Likewise.
3262 * testsuite/libffi.call/nested_struct1.c: Likewise.
3263 * testsuite/libffi.call/problem1.c: Likewise.
3264 * testsuite/libffi.special/unwindtest.cc: Likewise.
3265 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
3267 2003-09-18 David Edelsohn <edelsohn@gnu.org>
3269 * src/powerpc/aix.S: Cleanup whitespaces.
3270 * src/powerpc/aix_closure.S: Likewise.
3272 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3274 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
3275 * src/powerpc/darwin_closure.S: Likewise.
3276 * src/powerpc/ffi_darwin.c: Likewise.
3278 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3279 David Edelsohn <edelsohn@gnu.org>
3281 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
3282 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
3284 * src/powerpc/darwin_closure.S: Likewise.
3285 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
3286 according to the Darwin/AIX ABI.
3287 (ffi_prep_cif_machdep): Likewise.
3288 (ffi_closure_helper_DARWIN): Likewise.
3289 Remove the outgoing parameter stack logic. Simplify the evaluation
3290 of the different CASE types.
3291 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
3292 statement in the trampoline code.
3294 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
3296 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
3297 for the register size.
3298 (ffi_closure_helper_SYSV): Handle the structure return value
3300 (ffi_closure_helper_SYSV): Return the appropriate type when
3301 the registers are used for the structure return value.
3302 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
3303 the 64-bit return value. Update copyright years.
3305 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3307 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
3308 srcdir for ffi_mips.h.
3310 2003-09-12 Alan Modra <amodra@bigpond.net.au>
3312 * src/prep_cif.c (initialize_aggregate): Include tail padding in
3314 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
3315 placement of float result.
3316 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
3317 cast of "resp" for big-endian 64 bit machines.
3319 2003-09-11 Alan Modra <amodra@bigpond.net.au>
3321 * src/types.c (double, longdouble): Merge identical SH and ARM
3322 typedefs, and add POWERPC64.
3323 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
3324 struct split over gpr and rest.
3325 (ffi_prep_cif_machdep): Correct intarg_count for structures.
3326 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
3328 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
3330 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
3333 2003-09-09 Alan Modra <amodra@bigpond.net.au>
3335 * configure: Regenerate.
3337 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3339 * Makefile.am: Remove build rules for ffitest.
3340 * Makefile.in: Rebuilt.
3342 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3344 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3345 about implicit declaration of abort().
3347 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3349 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3350 * Makefile.in: Rebuilt.
3351 * configure.in: Add dejagnu test framework.
3352 * configure: Rebuilt.
3354 * testsuite/Makefile.am: New file.
3355 * testsuite/Makefile.in: Built
3356 * testsuite/lib/libffi-dg.exp: New file.
3357 * testsuite/config/default.exp: Likewise.
3358 * testsuite/libffi.call/call.exp: Likewise.
3359 * testsuite/libffi.call/ffitest.h: Likewise.
3360 * testsuite/libffi.call/closure_fn0.c: Likewise.
3361 * testsuite/libffi.call/closure_fn1.c: Likewise.
3362 * testsuite/libffi.call/closure_fn2.c: Likewise.
3363 * testsuite/libffi.call/closure_fn3.c: Likewise.
3364 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3365 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3366 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3367 * testsuite/libffi.call/cls_2byte.c: Likewise.
3368 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3369 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3370 * testsuite/libffi.call/cls_4byte.c: Likewise.
3371 * testsuite/libffi.call/cls_5byte.c: Likewise.
3372 * testsuite/libffi.call/cls_6byte.c: Likewise.
3373 * testsuite/libffi.call/cls_7byte.c: Likewise.
3374 * testsuite/libffi.call/cls_8byte.c: Likewise.
3375 * testsuite/libffi.call/cls_12byte.c: Likewise.
3376 * testsuite/libffi.call/cls_16byte.c: Likewise.
3377 * testsuite/libffi.call/cls_20byte.c: Likewise.
3378 * testsuite/libffi.call/cls_24byte.c: Likewise.
3379 * testsuite/libffi.call/cls_double.c: Likewise.
3380 * testsuite/libffi.call/cls_float.c: Likewise.
3381 * testsuite/libffi.call/cls_uchar.c: Likewise.
3382 * testsuite/libffi.call/cls_uint.c: Likewise.
3383 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3384 * testsuite/libffi.call/cls_ushort.c: Likewise.
3385 * testsuite/libffi.call/float.c: Likewise.
3386 * testsuite/libffi.call/float1.c: Likewise.
3387 * testsuite/libffi.call/float2.c: Likewise.
3388 * testsuite/libffi.call/many.c: Likewise.
3389 * testsuite/libffi.call/many_win32.c: Likewise.
3390 * testsuite/libffi.call/nested_struct.c: Likewise.
3391 * testsuite/libffi.call/nested_struct1.c: Likewise.
3392 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3393 * testsuite/libffi.call/problem1.c: Likewise.
3394 * testsuite/libffi.call/promotion.c: Likewise.
3395 * testsuite/libffi.call/return_ll.c: Likewise.
3396 * testsuite/libffi.call/return_sc.c: Likewise.
3397 * testsuite/libffi.call/return_uc.c: Likewise.
3398 * testsuite/libffi.call/strlen.c: Likewise.
3399 * testsuite/libffi.call/strlen_win32.c: Likewise.
3400 * testsuite/libffi.call/struct1.c: Likewise.
3401 * testsuite/libffi.call/struct2.c: Likewise.
3402 * testsuite/libffi.call/struct3.c: Likewise.
3403 * testsuite/libffi.call/struct4.c: Likewise.
3404 * testsuite/libffi.call/struct5.c: Likewise.
3405 * testsuite/libffi.call/struct6.c: Likewise.
3406 * testsuite/libffi.call/struct7.c: Likewise.
3407 * testsuite/libffi.call/struct8.c: Likewise.
3408 * testsuite/libffi.call/struct9.c: Likewise.
3409 * testsuite/libffi.special/special.exp: New file.
3410 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3411 * testsuite/libffi.special/unwindtest.cc: Likewise.
3414 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3416 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3419 2003-08-02 Alan Modra <amodra@bigpond.net.au>
3421 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3423 (ffi_closure_helper_LINUX64): Likewise.
3424 * src/powerpc/linux64.S: Remove code writing to parm save area.
3425 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3426 address in lr from ffi_closure_helper_LINUX64 call to calculate
3427 table address. Optimize function tail.
3429 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3431 * src/sparc/ffi.c: Handle all floating point registers.
3432 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3434 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3436 * README: Note that libffi is not part of GCC. Update the project
3439 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3441 * src/powerpc/ppc_closure.S: Include ffi.h.
3443 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3445 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3446 Use C style comments.
3448 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3450 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3451 * Makefile.in: Regenerate.
3452 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3453 * configure: Regenerate.
3454 * include/ffi.h.in: Add SHmedia support.
3455 * src/sh64/ffi.c: New file.
3456 * src/sh64/sysv.S: New file.
3458 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3460 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3461 should be read-only.
3462 * configure: Rebuilt.
3463 * fficonfig.h.in: Rebuilt.
3464 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3465 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3466 * src/powerpc/linux64.S: Likewise.
3467 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3468 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3469 if -fpic/-fPIC/-mrelocatable.
3470 * src/powerpc/powerpc_closure.S: Likewise.
3471 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3472 #write in .eh_frame flags.
3473 * src/sparc/v9.S: Likewise.
3474 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3475 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3476 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3478 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3480 Fixes PR bootstrap/10656
3481 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3482 support for .register pseudo-op.
3483 * src/sparc/v8.S: Use it.
3484 * fficonfig.h.in: Rebuilt.
3485 * configure: Rebuilt.
3487 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3489 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3490 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3491 Make it the default on POWERPC64.
3492 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3493 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3494 * configure: Rebuilt.
3495 * src/powerpc/ffi.c (hidden): Define.
3496 (ffi_prep_args_SYSV): Renamed from
3497 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3498 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3499 ASM_NEEDS_REGISTERS64): New.
3500 (ffi_prep_args64): New function.
3501 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3502 (ffi_call): Likewise.
3503 (ffi_prep_closure): Likewise.
3504 (flush_icache): Surround by #ifndef POWERPC64.
3505 (ffi_dblfl): New union type.
3506 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3507 (ffi_closure_helper_LINUX64): New function.
3508 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3510 * src/powerpc/sysv.S: Likewise.
3511 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3512 * src/powerpc/linux64.S: New file.
3513 * src/powerpc/linux64_closure.S: New file.
3514 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3515 src/powerpc/linux64_closure.S.
3516 (TARGET_SRC_POWERPC): Likewise.
3518 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3519 closure_test_fn3): Fix result printing on big-endian 64-bit
3521 (main): Print tst2_arg instead of uninitialized tst2_result.
3523 * src/ffitest.c (main): Hide what closure pointer really points to
3526 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3528 * configure.in (arm-*-netbsdelf*): Add configuration.
3529 (configure): Regenerated.
3531 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3533 * include/Makefile.in: Regenerate.
3535 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3537 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3539 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3540 Receive closure pointer through parameter, read args using
3541 __builtin_dwarf_cfa.
3542 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3544 2003-03-12 Andreas Schwab <schwab@suse.de>
3546 * configure.in: Avoid trailing /. in toolexeclibdir.
3547 * configure: Rebuilt.
3549 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3551 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3553 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3555 * libffi/src/powerpc/darwin_closure.S:
3556 Fix alignement bug, allocate 8 bytes for the result.
3557 * libffi/src/powerpc/aix_closure.S:
3559 * libffi/src/powerpc/ffi_darwin.c:
3560 Update stackframe description for aix/darwin_closure.S.
3562 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3564 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3567 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3568 Andreas Schwab <schwab@suse.de>
3570 * configure.in: Adjust command to source config-ml.in to account
3571 for changes to the libffi_basedir definition.
3572 (libffi_basedir): Remove ${srcdir} from value and include trailing
3575 * configure: Regenerate.
3577 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3579 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3581 2003-01-28 Andrew Haley <aph@redhat.com>
3583 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3584 * src/x86/ffi64.c (ffi_prep_closure): New.
3585 (ffi_closure_UNIX64_inner): New.
3586 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3588 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
3590 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3591 Remove USE_LIBDIR conditional.
3592 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3593 * Makefile.in, configure: Rebuilt.
3595 2003-01027 David Edelsohn <edelsohn@gnu.org>
3597 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3598 * Makefile.in: Regenerate.
3600 2003-01-22 Andrew Haley <aph@redhat.com>
3602 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3605 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3607 * src/powerpc/darwin.S: Add unwind info.
3608 * src/powerpc/darwin_closure.S: Likewise.
3610 2003-01-14 Andrew Haley <aph@redhat.com>
3612 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3613 (ffi_prep_cif_machdep): Likewise.
3614 * src/x86/unix64.S: Add unwind info.
3616 2003-01-14 Andreas Jaeger <aj@suse.de>
3618 * src/ffitest.c (main): Only use ffi_closures if those are
3621 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3623 * libffi/src/ffitest.c
3624 add closure testcases
3626 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3628 * libffi/src/powerpc/ffi.c
3629 fix alignment bug for float (4 byte aligned iso 8 byte)
3631 2003-01-09 Geoffrey Keating <geoffk@apple.com>
3633 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3634 * src/powerpc/darwin.S: Remove RCS version string.
3636 2003-01-03 Jeff Sturm <jsturm@one-point.com>
3638 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3639 * src/ffitest.c (main): Use static storage for closure.
3640 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3641 * src/sparc/v8.S (ffi_closure_v8): New.
3642 * src/sparc/v9.S (ffi_closure_v9): New.
3644 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3646 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3647 enumeration for X86_WIN32.
3648 * src/x86/win32.S: Added ffi_call_STDCALL function
3650 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3651 switch cases for recognising FFI_STDCALL and
3652 calling ffi_call_STDCALL if target is X86_WIN32.
3653 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3654 stdcall versions of the "my_strlen" and "many"
3655 test functions (for X86_WIN32).
3656 Added test cases to test stdcall invocation using
3659 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3661 * src/sh/sysv.S: Add DWARF2 unwind info.
3663 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3665 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3667 2002-11-26 Jim Wilson <wilson@redhat.com>
3669 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3671 2002-11-23 H.J. Lu <hjl@gnu.org>
3673 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3674 Include ../config/accross.m4.
3675 * aclocal.m4; Rebuild.
3676 * configure: Likewise.
3678 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3680 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3682 2002-11-11 DJ Delorie <dj@redhat.com>
3684 * configure.in: Look for common files in the right place.
3686 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3688 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3689 raw data as _Jv_word values, not ffi_raw.
3690 (ffi_java_ptrarray_to_raw): Likewise.
3691 (ffi_java_rvalue_to_raw): New function.
3692 (ffi_java_raw_call): Call it.
3693 (ffi_java_raw_to_rvalue): New function.
3694 (ffi_java_translate_args): Call it.
3695 * src/ffitest.c (closure_test_fn): Interpret return value
3696 as ffi_arg, not int.
3697 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3698 FFI_TYPE_POINTER case.
3699 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3700 values extended to word size.
3702 2002-10-02 Andreas Jaeger <aj@suse.de>
3704 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3706 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
3708 * include/ffi.h.in: Fix i386 win32 compilation.
3710 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3712 * configure.in: Add s390x-*-linux-* target.
3713 * configure: Regenerate.
3714 * include/ffi.h.in: Define S390X for s390x targets.
3715 (FFI_CLOSURES): Define for s390/s390x.
3716 (FFI_TRAMPOLINE_SIZE): Likewise.
3717 (FFI_NATIVE_RAW_API): Likewise.
3718 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3719 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
3720 * src/s390/ffi.c: Major rework of existing code. Add support for
3721 s390x targets. Add closure support.
3722 * src/s390/sysv.S: Likewise.
3724 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
3726 * src/arm/sysv.S: Fix typo.
3728 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
3730 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3731 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3732 (ffi_call_SYSV): Handle soft-float.
3734 2002-09-27 Bo Thorsen <bo@suse.de>
3736 * include/ffi.h.in: Fix multilib x86-64 support.
3738 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3740 * Makefile.am (all-multi): Fix multilib parallel build.
3742 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3744 * configure.in (sh[34]*-*-linux*): Add brackets.
3745 * configure: Regenerate.
3747 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3749 * Makefile.am: Add SH support.
3750 * Makefile.in: Regenerate.
3751 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3752 * configure: Regenerate.
3753 * include/ffi.h.in: Add SH support.
3754 * src/sh/ffi.c: New file.
3755 * src/sh/sysv.S: New file.
3756 * src/types.c: Add SH support.
3758 2002-07-16 Bo Thorsen <bo@suse.de>
3760 * src/x86/ffi64.c: New file that adds x86-64 support.
3761 * src/x86/unix64.S: New file that handles argument setup for
3763 * src/x86/sysv.S: Don't use this on x86-64.
3764 * src/x86/ffi.c: Don't use this on x86-64.
3766 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3768 * src/ffitest.c (struct6): New test that tests a special case in
3770 (struct7): Likewise.
3771 (struct8): Likewise.
3772 (struct9): Likewise.
3773 (closure_test_fn): Silence warning about this when it's not used.
3774 (main): Add the new tests.
3775 (main): Fix a couple of wrong casts and silence some compiler warnings.
3776 * include/ffi.h.in: Add x86-64 ABI definition.
3777 * fficonfig.h.in: Regenerate.
3778 * Makefile.am: Add x86-64 support.
3779 * configure.in: Likewise.
3780 * Makefile.in: Regenerate.
3781 * configure: Likewise.
3783 2002-06-24 Bo Thorsen <bo@suse.de>
3785 * src/types.c: Merge settings for similar architectures.
3786 Add x86-64 sizes and alignments.
3788 2002-06-23 Bo Thorsen <bo@suse.de>
3790 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3791 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3792 * src/mips/ffi.c (ffi_prep_args): Likewise.
3793 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3795 2002-07-18 H.J. Lu (hjl@gnu.org)
3797 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3798 (libffi_la_SOURCES): Support MIPS_LINUX.
3799 (libffi_convenience_la_SOURCES): Likewise.
3800 * Makefile.in: Regenerated.
3802 * configure.in (mips64*-*): Skip.
3803 (mips*-*-linux*): New.
3804 * configure: Regenerated.
3806 * src/mips/ffi.c: Include <sgidefs.h>.
3808 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3810 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3812 2002-05-27 Roger Sayle <roger@eyesopen.com>
3814 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3816 2002-05-27 Bo Thorsen <bo@suse.de>
3818 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3821 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3823 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3824 beginning of function (for older apple cc).
3826 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3828 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3829 script entry, and set LD to it when configuring multilibs.
3830 * configure: Rebuilt.
3832 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3834 * configure.in (sparc64-*-netbsd*): Add target.
3835 (sparc-*-netbsdelf*): Likewise.
3836 * configure: Regenerate.
3838 2002-04-28 David S. Miller <davem@redhat.com>
3840 * configure.in, configure: Fix SPARC test in previous change.
3842 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3844 * Makefile.am: Add Linux for S/390 support.
3845 * Makefile.in: Regenerate.
3846 * configure.in: Add Linux for S/390 support.
3847 * configure: Regenerate.
3848 * include/ffi.h.in: Add Linux for S/390 support.
3849 * src/s390/ffi.c: New file from libffi CVS tree.
3850 * src/s390/sysv.S: New file from libffi CVS tree.
3852 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3854 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3856 * src/sparc/v8.S: Use it.
3857 * src/sparc/v9.S: Likewise.
3858 * fficonfig.h.in: Rebuilt.
3859 * configure: Rebuilt.
3861 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3863 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3865 * src/ia64/unix.S: Add unwind information. Fix comments.
3866 Save sp in a way that's compatible with unwind info.
3867 (ffi_call_unix): Correctly restore sp in all cases.
3868 * src/ia64/ffi.c: Add, fix comments.
3870 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3872 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3874 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3876 * configure.in (alpha*-*-netbsd*): Add target.
3877 * configure: Regenerate.
3879 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3881 * src/sparc/v8.S: Add unwind info.
3882 * src/sparc/v9.S: Likewise.
3884 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3886 * configure.in: Enable i*86-*-netbsdelf*.
3887 * configure: Rebuilt.
3889 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3892 * src/mips/n32.s: Delete
3893 * src/mips/o32.s: Delete
3895 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3897 * configure.in: Enable alpha*-*-freebsd*.
3898 * configure: Rebuilt.
3900 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3902 * Makefile.am: libfficonvenience -> libffi_convenience.
3903 * Makefile.in: Rebuilt.
3905 * Makefile.am: Define ffitest_OBJECTS.
3906 * Makefile.in: Rebuilt.
3908 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3909 David Edelsohn <edelsohn@gnu.org>
3911 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3912 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3913 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3914 * Makefile.in: Regenerate.
3915 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3916 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3917 (flush_icache, flush_range): New functions.
3918 (ffi_closure_helper_DARWIN): New function.
3919 * src/powerpc/aix_closure.S: New file.
3920 * src/powerpc/darwin_closure.S: New file.
3922 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3924 * include/ffi.h.in: Add typedef for ffi_arg.
3925 * src/ffitest.c (main): Declare rint with ffi_arg.
3927 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3929 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3930 number of GPRs for floating-point arguments.
3932 2002-01-31 Anthony Green <green@redhat.com>
3934 * configure: Rebuilt.
3935 * configure.in: Replace CHECK_SIZEOF and endian tests with
3936 cross-compiler friendly macros.
3937 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3940 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3942 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3943 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3945 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3947 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3948 (TARGET_SRC_POWERPC_AIX): New.
3949 (POWERPC_AIX): New stanza.
3950 * Makefile.in: Regenerate.
3951 * configure.in: Add AIX case.
3952 * configure: Regenerate.
3953 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3954 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3955 size. Fix "long double" support.
3956 (ffi_call): Add FFI_AIX case.
3957 * src/powerpc/aix.S: New.
3959 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3961 Implement Darwin PowerPC ABI.
3962 * configure.in: Handle powerpc-*-darwin*.
3963 * Makefile.am: Set source files for POWERPC_DARWIN.
3964 * configure: Rebuilt.
3965 * Makefile.in: Rebuilt.
3966 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3968 * src/powerpc/darwin.S: New file.
3969 * src/powerpc/ffi_darwin.c: New file.
3971 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3973 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3975 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3977 * src/x86/sysv.S: Avoid gas-only .balign directive.
3978 Use C style comments.
3980 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3982 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3983 Fixes PR bootstrap/3563.
3985 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3987 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3989 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3991 * configure.in: Recognize sparc*-sun-* host.
3992 * configure: Regenerate.
3994 2001-06-06 Andrew Haley <aph@redhat.com>
3996 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
3998 2001-06-03 Andrew Haley <aph@redhat.com>
4000 * src/alpha/osf.S: Add unwind info.
4001 * src/powerpc/sysv.S: Add unwind info.
4002 * src/powerpc/ppc_closure.S: Likewise.
4004 2000-05-31 Jeff Sturm <jsturm@one-point.com>
4006 * configure.in: Fix AC_ARG_ENABLE usage.
4007 * configure: Rebuilt.
4009 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
4011 * configure.in: Remove warning about beta code.
4012 * configure: Rebuilt.
4014 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
4016 * src/ia64/unix.S: Restore stack pointer when returning from
4018 * src/ia64/ffi.c: Fix typo in comment.
4020 2001-04-18 Jim Wilson <wilson@redhat.com>
4022 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
4023 to eliminate RAW DV.
4025 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
4027 * Makefile.am: Make a libtool convenience library.
4028 * Makefile.in: Rebuilt.
4030 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
4032 * configure.in: Use different syntax for subdirectory creation.
4033 * configure: Rebuilt.
4035 2001-03-27 Jon Beniston <jon@beniston.com>
4037 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
4038 * configure: Rebuilt.
4039 * Makefile.am: Added X86_WIN32 target support.
4040 * Makefile.in: Rebuilt.
4042 * include/ffi.h.in: Added X86_WIN32 target support.
4044 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
4045 * src/types.c: Added X86_WIN32 target support.
4047 * src/x86/win32.S: New file. Based on sysv.S, but with EH
4048 stuff removed and made to work with CygWin's gas.
4050 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
4052 * configure.in: Make target subdirectory in build dir.
4053 * Makefile.am: Override suffix based rules to specify correct output
4055 * Makefile.in: Rebuilt.
4056 * configure: Rebuilt.
4058 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
4060 * src/powerpc/ppc_closure.S: New file.
4061 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
4062 involving long long and register pairs.
4063 (ffi_prep_closure): New function.
4064 (flush_icache): Likewise.
4065 (ffi_closure_helper_SYSV): Likewise.
4066 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
4067 (FFI_TRAMPOLINE_SIZE): Likewise.
4068 (FFI_NATIVE_RAW_API): Likewise.
4069 * Makefile.in: Rebuilt.
4070 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
4071 (TARGET_SRC_POWERPC): Likewise.
4073 2001-03-19 Tom Tromey <tromey@redhat.com>
4075 * Makefile.in: Rebuilt.
4076 * Makefile.am (ffitest_LDFLAGS): New macro.
4078 2001-03-02 Nick Clifton <nickc@redhat.com>
4080 * include/ffi.h.in: Remove RCS ident string.
4081 * include/ffi_mips.h: Remove RCS ident string.
4082 * src/debug.c: Remove RCS ident string.
4083 * src/ffitest.c: Remove RCS ident string.
4084 * src/prep_cif.c: Remove RCS ident string.
4085 * src/types.c: Remove RCS ident string.
4086 * src/alpha/ffi.c: Remove RCS ident string.
4087 * src/alpha/osf.S: Remove RCS ident string.
4088 * src/arm/ffi.c: Remove RCS ident string.
4089 * src/arm/sysv.S: Remove RCS ident string.
4090 * src/mips/ffi.c: Remove RCS ident string.
4091 * src/mips/n32.S: Remove RCS ident string.
4092 * src/mips/o32.S: Remove RCS ident string.
4093 * src/sparc/ffi.c: Remove RCS ident string.
4094 * src/sparc/v8.S: Remove RCS ident string.
4095 * src/sparc/v9.S: Remove RCS ident string.
4096 * src/x86/ffi.c: Remove RCS ident string.
4097 * src/x86/sysv.S: Remove RCS ident string.
4099 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
4101 * include/ffi.h.in: Change sourceware.cygnus.com references to
4104 2000-12-09 Richard Henderson <rth@redhat.com>
4106 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
4107 (ffi_closure_osf_inner): Index rather than increment avalue
4108 and arg_types. Give ffi_closure_osf the raw return value type.
4109 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
4112 2000-12-07 Richard Henderson <rth@redhat.com>
4114 * src/raw_api.c (ffi_translate_args): Fix typo.
4115 (ffi_prep_closure): Likewise.
4117 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
4118 FFI_TRAMPOLINE_SIZE.
4119 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
4120 cif->bytes for new ffi_call_osf implementation.
4121 (ffi_prep_args): Absorb into ...
4122 (ffi_call): ... here. Do all stack allocation here and
4123 avoid a callback function.
4124 (ffi_prep_closure, ffi_closure_osf_inner): New.
4125 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
4126 (ffi_closure_osf): New.
4128 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
4130 * config.guess, config.sub, install-sh: Removed.
4131 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
4132 * Makefile.in: Rebuilt.
4134 * acinclude.m4: Include libtool macros from the top level.
4135 * aclocal.m4, configure: Rebuilt.
4137 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
4139 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
4140 * configure: Rebuilt.
4142 2000-05-11 Scott Bambrough <scottb@netwinder.org>
4144 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
4145 memory correctly. Use conditional instructions, not branches where
4148 2000-05-04 Tom Tromey <tromey@cygnus.com>
4150 * configure: Rebuilt.
4151 * configure.in: Match `arm*-*-linux-*'.
4152 From Chris Dornan <cdornan@arm.com>.
4154 2000-04-28 Jakub Jelinek <jakub@redhat.com>
4156 * Makefile.am (SUBDIRS): Define.
4157 (AM_MAKEFLAGS): Likewise.
4158 (Multilib support.): Add section.
4159 * Makefile.in: Rebuilt.
4160 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
4161 New variables. Set for gcc using -print-multi-lib. Export them
4163 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
4164 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
4166 (extra_compiler_flags_value, extra_compiler_flags): Check these
4167 for extra compiler options which need to be passed down in
4170 2000-04-16 Anthony Green <green@redhat.com>
4172 * configure: Rebuilt.
4173 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
4175 2000-04-14 Jakub Jelinek <jakub@redhat.com>
4177 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
4178 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
4179 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
4180 Replace all void * sizeofs with sizeof(int).
4181 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
4182 different than DOUBLE.
4183 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
4184 (ffi_prep_args_v9): New function.
4185 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
4186 (ffi_V9_return_struct): New function.
4187 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
4188 32bit code (not yet cross-arch calls).
4189 * src/sparc/v8.S: Add struct return delay nop.
4191 * src/sparc/v9.S: New file.
4192 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
4193 is used on sparc64 only for structures larger than 32 bytes.
4194 Pass by reference for structures is done for structure arguments
4195 larger than 16 bytes.
4196 * src/ffitest.c (main): Use 64bit rint on sparc64.
4197 Run long long tests on sparc.
4198 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
4200 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
4202 * configure.in (sparc-*-linux*): New supported target.
4203 (sparc64-*-linux*): Likewise.
4204 * configure: Rebuilt.
4205 * Makefile.am: Add v9.S to SPARC files.
4206 * Makefile.in: Likewise.
4207 (LINK): Surround $(CCLD) into double quotes, so that multilib
4208 compiles work correctly.
4210 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
4212 * configure: Rebuilt.
4213 * configure.in: (i*86-*-solaris*): New libffi target. Patch
4214 proposed by Bryce McKinlay.
4216 2000-03-20 Tom Tromey <tromey@cygnus.com>
4218 * Makefile.in: Hand edit for java_raw_api.lo.
4220 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
4222 * config.guess, config.sub: Update from the gcc tree.
4223 Fix for PR libgcj/168.
4225 2000-03-03 Tom Tromey <tromey@cygnus.com>
4227 * Makefile.in: Fixed ia64 by hand.
4229 * configure: Rebuilt.
4230 * configure.in (--enable-multilib): New option.
4231 (libffi_basedir): New subst.
4232 (AC_OUTPUT): Added multilib code.
4234 2000-03-02 Tom Tromey <tromey@cygnus.com>
4236 * Makefile.in: Rebuilt.
4237 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
4240 2000-02-25 Hans Boehm <boehm@acm.org>
4242 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
4244 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
4246 (ffi_prep_raw_closure): Use ffi_translate_args, not
4247 ffi_closure_translate.
4248 * src/java_raw_api.c: New file.
4249 * src/ffitest.c (closure_test_fn): New function.
4250 (main): Define `rint' as long long on IA64. Added new test when
4251 FFI_CLOSURES is defined.
4252 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
4253 (ffi_abi): Recognize IA64.
4254 (ffi_raw): Added `flt' field.
4255 Added "Java raw API" code.
4256 * configure.in: Recognize ia64.
4257 * Makefile.am (TARGET_SRC_IA64): New macro.
4258 (libffi_la_common_SOURCES): Added java_raw_api.c.
4259 (libffi_la_SOURCES): Define in IA64 case.
4261 2000-01-04 Tom Tromey <tromey@cygnus.com>
4263 * Makefile.in: Rebuilt with newer automake.
4265 1999-12-31 Tom Tromey <tromey@cygnus.com>
4267 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
4269 1999-09-01 Tom Tromey <tromey@cygnus.com>
4271 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
4273 * fficonfig.h.in: Rebuilt.
4274 * configure: Rebuilt.
4275 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
4276 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
4277 * acconfig.h: Don't #undef PACKAGE or VERSION.
4279 1999-08-09 Anthony Green <green@cygnus.com>
4281 * include/ffi.h.in: Try to work around messy header problem
4282 with PACKAGE and VERSION.
4284 * configure: Rebuilt.
4285 * configure.in: Change version to 2.00-beta.
4287 * fficonfig.h.in: Rebuilt.
4288 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
4290 * src/x86/ffi.c (ffi_raw_call): Rename.
4292 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
4294 * src/x86/ffi.c (ffi_closure_SYSV): New function.
4295 (ffi_prep_incoming_args_SYSV): Ditto.
4296 (ffi_prep_closure): Ditto.
4297 (ffi_closure_raw_SYSV): Ditto.
4298 (ffi_prep_raw_closure): More ditto.
4299 (ffi_call_raw): Final ditto.
4301 * include/ffi.h.in: Add definitions for closure and raw API.
4303 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4306 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
4308 * src/raw_api.c: New file.
4310 * include/ffi.h.in (ffi_raw): New type.
4311 (UINT_ARG, SINT_ARG): New defines.
4312 (ffi_closure, ffi_raw_closure): New types.
4313 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4315 * configure.in: Add check for endianness and sizeof void*.
4317 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
4318 instead of directly.
4320 * configure: Rebuilt.
4322 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
4324 * configure.in: Add x86 and powerpc BeOS configurations.
4325 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
4327 1999-05-09 Anthony Green <green@cygnus.com>
4329 * configure.in: Add warning about this being beta code.
4330 Remove src/Makefile.am from the picture.
4331 * configure: Rebuilt.
4333 * Makefile.am: Move logic from src/Makefile.am. Add changes
4334 to support libffi as a target library.
4335 * Makefile.in: Rebuilt.
4337 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
4338 Upgraded to new autoconf, automake, libtool.
4342 * LICENSE: Update copyright date.
4344 * src/Makefile.am, src/Makefile.in: Removed.
4346 1998-11-29 Anthony Green <green@cygnus.com>
4348 * include/ChangeLog: Removed.
4349 * src/ChangeLog: Removed.
4350 * src/mips/ChangeLog: Removed.
4351 * src/sparc/ChangeLog: Remboved.
4352 * src/x86/ChangeLog: Removed.
4354 * ChangeLog.v1: Created.