1 2011-06-29 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
4 * testsuite/libffi.call/cls_double_va.c: xfail dg-output on
6 * testsuite/libffi.call/cls_longdouble_va.c: Likewise.
8 2011-06-14 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
10 * testsuite/libffi.call/huge_struct.c (test_large_fn): Use PRIu8,
11 PRId8 instead of %hhu, %hhd.
12 * testsuite/libffi.call/ffitest.h [__alpha__ && __osf__] (PRId8,
14 [__sgi__] (PRId8, PRIu8): Define.
16 2011-04-29 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
18 * src/alpha/osf.S (UA_SI, FDE_ENCODING, FDE_ENCODE, FDE_ARANGE):
20 Use them to handle ELF vs. ECOFF differences.
21 [__osf__] (_GLOBAL__F_ffi_call_osf): Define.
23 2011-02-13 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
25 * configure: Regenerate.
27 2011-02-09 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
30 * testsuite/libffi.call/cls_pointer.c (main): Cast void * to
32 * testsuite/libffi.call/cls_pointer_stack.c (main): Likewise.
34 2011-02-07 Joel Sherrill <joel.sherrill@oarcorp.com>
36 * libffi/src/m68k/ffi.c: Add RTEMS support for cache flushing.
37 Handle case when CPU variant does not have long double support.
38 * libffi/src/m68k/sysv.S: Add support for mc68000, Coldfire,
39 and cores with soft floating point.
41 2011-02-07 Joel Sherrill <joel.sherrill@oarcorp.com>
43 * configure.ac: Add mips*-*-rtems* support.
44 * configure: Regenerate.
45 * src/mips/ffitarget.h: Ensure needed constants are available
46 for targets which do not have sgidefs.h.
48 2011-01-26 Dave Korn <dave.korn.cygwin@gmail.com>
51 * configure.ac (AM_LTLDFLAGS): Add -bindir option for windows DLLs.
52 * configure: Regenerate.
54 2010-12-18 Iain Sandoe <iains@gcc.gnu.org>
58 * src/powerpc/darwin_closure.S: Provide Darwin64 implementation,
60 * src/powerpc/ffitarget.h (POWERPC_DARWIN64): New,
61 (FFI_TRAMPOLINE_SIZE): Update for Darwin64.
62 * src/powerpc/darwin.S: Provide Darwin64 implementation,
64 * src/powerpc/ffi_darwin.c: Likewise.
66 2010-12-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
68 * configure.ac (libffi_cv_as_ascii_pseudo_op): Use double
70 (libffi_cv_as_string_pseudo_op): Likewise.
71 * configure: Regenerate.
73 2010-12-03 Chung-Lin Tang <cltang@codesourcery.com>
75 * src/arm/sysv.S (ffi_closure_SYSV): Add UNWIND to .pad directive.
76 (ffi_closure_VFP): Same.
77 (ffi_call_VFP): Move down to before ffi_closure_VFP. Add '.fpu vfp'
80 2010-12-01 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
82 * testsuite/libffi.call/ffitest.h [__sgi] (PRId64, PRIu64): Define.
85 2010-11-29 Richard Henderson <rth@redhat.com>
86 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
88 * src/x86/sysv.S (FDE_ENCODING, FDE_ENCODE): Define.
89 (.eh_frame): Use FDE_ENCODING.
90 (.LASFDE1, .LASFDE2, LASFDE3): Simplify with FDE_ENCODE.
92 2010-11-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
94 * testsuite/lib/libffi-dg.exp: Rename ...
95 * testsuite/lib/libffi.exp: ... to this.
96 * libffi/testsuite/libffi.call/call.exp: Don't load libffi-dg.exp.
97 * libffi/testsuite/libffi.special/special.exp: Likewise.
99 2010-10-28 Chung-Lin Tang <cltang@codesourcery.com>
101 * src/arm/ffi.c (ffi_prep_args): Add VFP register argument handling
102 code, new parameter, and return value. Update comments.
103 (ffi_prep_cif_machdep): Add case for VFP struct return values. Add
104 call to layout_vfp_args().
105 (ffi_call_SYSV): Update declaration.
106 (ffi_call_VFP): New declaration.
107 (ffi_call): Add VFP struct return conditions. Call ffi_call_VFP()
109 (ffi_closure_VFP): New declaration.
110 (ffi_closure_SYSV_inner): Add new vfp_args parameter, update call to
111 ffi_prep_incoming_args_SYSV().
112 (ffi_prep_incoming_args_SYSV): Update parameters. Add VFP argument
114 (ffi_prep_closure_loc): Pass ffi_closure_VFP to trampoline
115 construction under VFP hard-float.
116 (rec_vfp_type_p): New function.
118 (place_vfp_arg): Same.
119 (layout_vfp_args): Same.
120 * src/arm/ffitarget.h (ffi_abi): Add FFI_VFP. Define FFI_DEFAULT_ABI
121 based on __ARM_PCS_VFP.
122 (FFI_EXTRA_CIF_FIELDS): Define for adding VFP hard-float specific
124 (FFI_TYPE_STRUCT_VFP_FLOAT): Define internally used type code.
125 (FFI_TYPE_STRUCT_VFP_DOUBLE): Same.
126 * src/arm/sysv.S (ffi_call_SYSV): Change call of ffi_prep_args() to
127 direct call. Move function pointer load upwards.
128 (ffi_call_VFP): New function.
129 (ffi_closure_VFP): Same.
131 * testsuite/lib/libffi-dg.exp (check-flags): New function.
132 (dg-skip-if): New function.
133 * testsuite/libffi.call/cls_double_va.c: Skip if target is arm*-*-*
134 and compiler options include -mfloat-abi=hard.
135 * testsuite/libffi.call/cls_longdouble_va.c: Same.
137 2010-10-01 Jakub Jelinek <jakub@redhat.com>
140 * src/x86/ffi64.c (ffi_prep_cif_machdep): Ensure cif->bytes is
142 * testsuite/libffi.call/many2.c: New test.
144 2010-08-20 Mark Wielaard <mjw@redhat.com>
146 * src/closures.c (open_temp_exec_file_mnt): Check if getmntent_r
149 2010-08-09 Andreas Tobler <andreast@fgznet.ch>
151 * configure.ac: Add target powerpc64-*-freebsd*.
152 * configure: Regenerate.
153 * testsuite/libffi.call/cls_align_longdouble_split.c: Pass
154 -mlong-double-128 only to linux targets.
155 * testsuite/libffi.call/cls_align_longdouble_split2.c: Likewise.
156 * testsuite/libffi.call/cls_longdouble.c: Likewise.
157 * testsuite/libffi.call/huge_struct.c: Likewise.
159 2010-07-10 Evan Phoenix <evan@fallingsnow.net>
161 * src/closures.c (selinux_enabled_check): Fix strncmp usage bug.
163 2010-07-07 Dan HorĂ¡k <dan@danny.cz>
165 * include/ffi.h.in: Protect #define with #ifndef.
166 * src/powerpc/ffitarget.h: Ditto.
167 * src/s390/ffitarget.h: Ditto.
168 * src/sparc/ffitarget.h: Ditto.
170 2010-07-07 Neil Roberts <neil@linux.intel.com>
172 * src/x86/sysv.S (ffi_call_SYSV): Align the stack pointer to
175 2010-07-02 Jakub Jelinek <jakub@redhat.com>
177 * Makefile.am (AM_MAKEFLAGS): Pass also mandir to submakes.
178 * Makefile.in: Regenerated.
180 2010-05-19 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
182 * configure.ac (libffi_cv_as_x86_pcrel): Check for illegal in as
184 (libffi_cv_as_ascii_pseudo_op): Check for .ascii.
185 (libffi_cv_as_string_pseudo_op): Check for .string.
186 * configure: Regenerate.
187 * fficonfig.h.in: Regenerate.
188 * src/x86/sysv.S (.eh_frame): Use .ascii, .string or error.
190 2010-05-05 Michael Kohler <michaelkohler@live.com>
192 * src/dlmalloc.c (dlfree): Fix spelling.
193 * src/ia64/ffi.c (ffi_prep_cif_machdep): Ditto.
194 * configure.ac: Ditto.
195 * configure: Rebuilt.
197 2010-04-13 Dan Witte <dwitte@mozilla.com>
199 * msvcc.sh: Build with -W3 instead of -Wall.
200 * src/powerpc/ffi_darwin.c: Remove build warnings.
201 * src/x86/ffi.c: Ditto.
202 * src/x86/ffitarget.h: Ditto.
204 2010-04-12 Dan Witte <dwitte@mozilla.com>
205 Walter Meinl <wuno@lsvw.de>
207 * configure.ac: Add OS/2 support.
208 * configure: Rebuilt.
209 * src/closures.c: Ditto.
210 * src/dlmalloc.c: Ditto.
211 * src/x86/win32.S: Ditto.
213 2010-04-07 Jakub Jelinek <jakub@redhat.com>
215 * testsuite/libffi.call/err_bad_abi.c: Remove unused args variable.
217 2010-04-02 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
219 * Makefile.in: Regenerate.
220 * aclocal.m4: Regenerate.
221 * include/Makefile.in: Regenerate.
222 * man/Makefile.in: Regenerate.
223 * testsuite/Makefile.in: Regenerate.
225 2010-03-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
227 * configure.ac (libffi_cv_as_x86_64_unwind_section_type): New test.
228 * configure: Regenerate.
229 * fficonfig.h.in: Regenerate.
230 * libffi/src/x86/unix64.S (.eh_frame)
231 [HAVE_AS_X86_64_UNWIND_SECTION_TYPE]: Use @unwind section type.
233 2010-03-14 Matthias Klose <doko@ubuntu.com>
235 * src/x86/ffi64.c: Fix typo in comment.
236 * src/x86/ffi.c: Use /* ... */ comment style.
238 2010-02-24 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
240 * doc/libffi.texi (The Closure API): Fix typo.
241 * doc/libffi.info: Remove.
243 2010-02-15 Matthias Klose <doko@ubuntu.com>
245 * src/arm/sysv.S (__ARM_ARCH__): Define for processor
248 2010-01-15 Anthony Green <green@redhat.com>
250 * README: Add notes on building with Microsoft Visual C++.
252 2010-01-15 Daniel Witte <dwitte@mozilla.com>
254 * msvcc.sh: New file.
256 * src/x86/win32.S: Port assembly routines to MSVC and #ifdef.
257 * src/x86/ffi.c: Tweak function declaration and remove excess
259 * include/ffi.h.in: Add __declspec(align(8)) to typedef struct
262 * src/x86/ffi.c: Merge ffi_call_SYSV and ffi_call_STDCALL into new
263 function ffi_call_win32 on X86_WIN32.
264 * src/x86/win32.S (ffi_call_SYSV): Rename to ffi_call_win32.
265 (ffi_call_STDCALL): Remove.
267 * src/prep_cif.c (ffi_prep_cif): Move stack space allocation code
268 to ffi_prep_cif_machdep for x86.
269 * src/x86/ffi.c (ffi_prep_cif_machdep): To here.
271 2010-01-15 Oliver Kiddle <okiddle@yahoo.co.uk>
273 * src/x86/ffitarget.h (ffi_abi): Check for __i386 and __amd64 for
274 Sun Studio compiler compatibility.
276 2010-01-12 Conrad Irwin <conrad.irwin@gmail.com>
278 * doc/libffi.texi: Add closure example.
280 2010-01-07 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
283 * testsuite/libffi.call/ffitest.h [__alpha__ && __osf__] (PRIdLL,
284 PRIuLL, PRId64, PRIu64, PRIuPTR): Define.
285 * testsuite/libffi.call/cls_align_sint64.c: Add -Wno-format on
287 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
288 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
289 * testsuite/libffi.call/return_ll1.c: Likewise.
290 * testsuite/libffi.call/stret_medium2.c: Likewise.
291 * testsuite/libffi.special/ffitestcxx.h (allocate_mmap): Cast
292 MAP_FAILED to char *.
294 2010-01-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
296 * src/mips/n32.S: Use .abicalls and .eh_frame with __GNUC__.
298 2009-12-31 Anthony Green <green@redhat.com>
300 * README: Update for libffi 3.0.9.
302 2009-12-27 Matthias Klose <doko@ubuntu.com>
304 * configure.ac (HAVE_LONG_DOUBLE): Define for mips when
306 * configure: Rebuilt.
308 2009-12-26 Anthony Green <green@redhat.com>
310 * testsuite/libffi.call/cls_longdouble_va.c: Mark as xfail for
312 * testsuite/libffi.call/cls_double_va.c: Ditto.
314 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
316 * testsuite/libffi.call/ffitest.h: Conditionally include stdint.h
318 * testsuite/libffi.special/unwindtest.cc: Ditto.
320 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
322 * configure.ac: Add amd64-*-openbsd*.
323 * configure: Rebuilt.
324 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Link
325 openbsd programs with -lpthread.
327 2009-12-26 Anthony Green <green@redhat.com>
329 * testsuite/libffi.call/cls_double_va.c,
330 testsuite/libffi.call/cls_longdouble.c,
331 testsuite/libffi.call/cls_longdouble_va.c,
332 testsuite/libffi.call/cls_pointer.c,
333 testsuite/libffi.call/cls_pointer_stack.c: Remove xfail for
334 mips*-*-* and arm*-*-*.
335 * testsuite/libffi.call/cls_align_longdouble_split.c,
336 testsuite/libffi.call/cls_align_longdouble_split2.c,
337 testsuite/libffi.call/stret_medium2.c,
338 testsuite/libffi.call/stret_medium.c,
339 testsuite/libffi.call/stret_large.c,
340 testsuite/libffi.call/stret_large2.c: Remove xfail for arm*-*-*.
342 2009-12-31 Kay Tietz <ktietz70@googlemail.com>
344 * testsuite/libffi.call/ffitest.h,
345 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRuLL): Fix
348 2009-12-31 Carlo Bramini <carlo.bramix@libero.it>
350 * configure.ac (AM_LTLDFLAGS): Define for windows hosts.
351 * Makefile.am (libffi_la_LDFLAGS): Add AM_LTLDFLAGS.
352 * configure: Rebuilt.
353 * Makefile.in: Rebuilt.
355 2009-12-31 Anthony Green <green@redhat.com>
358 * testsuite/libffi.call/huge_struct.c: New test case from Blake
361 2009-12-28 David Edelsohn <edelsohn@gnu.org>
363 * src/powerpc/ffi_darwin.c (ffi_prep_args): Copy abi and nargs to
365 (aix_adjust_aggregate_sizes): New function.
366 (ffi_prep_cif_machdep): Call it.
368 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
370 * configure.ac: Define FFI_MMAP_EXEC_WRIT for the given targets.
371 * configure: Regenerate.
372 * fficonfig.h.in: Likewise.
373 * src/closures.c: Remove the FFI_MMAP_EXEC_WRIT definition for
376 2009-12-26 Andreas Schwab <schwab@linux-m68k.org>
378 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Advance intarg_count
379 when a float arguments is passed in memory.
380 (ffi_closure_helper_SYSV): Mark general registers as used up when
381 a 64bit or soft-float long double argument is passed in memory.
383 2009-12-25 Matthias Klose <doko@ubuntu.com>
385 * man/ffi_call.3: Fix #include in examples.
386 * doc/libffi.texi: Add dircategory.
388 2009-12-25 Frank Everdij <f.p.x.everdij@tudelft.nl>
390 * include/ffi.h.in: Placed '__GNUC__' ifdef around
391 '__attribute__((aligned(8)))' in ffi_closure, fixes compile for
393 * include/ffi_common.h: Added '__sgi' define to non
394 '__attribute__((__mode__()))' integer typedefs.
395 * src/mips/ffi.c (ffi_call, ffi_closure_mips_inner_O32,
396 ffi_closure_mips_inner_N32): Added 'defined(_MIPSEB)' to BE check.
397 (ffi_closure_mips_inner_O32, ffi_closure_mips_inner_N32): Added
398 FFI_LONGDOUBLE support and alignment(N32 only).
399 * src/mips/ffitarget.h: Corrected '#include <sgidefs.h>' for IRIX and
400 fixed non '__attribute__((__mode__()))' integer typedefs.
401 * src/mips/n32.S: Put '#ifdef linux' around '.abicalls' and '.eh_frame'
402 since they are Linux/GNU Assembler specific.
404 2009-12-25 Bradley Smith <brad@brad-smith.co.uk>
406 * configure.ac, Makefile.am, src/avr32/ffi.c,
407 src/avr32/ffitarget.h,
408 src/avr32/sysv.S: Add AVR32 port.
409 * configure, Makefile.in: Rebuilt.
411 2009-12-21 Andreas Tobler <a.tobler@schweiz.org>
413 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
414 * configure: Regenerate.
416 2009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
418 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
420 2009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
422 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
425 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
427 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
430 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
432 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
434 2009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
437 * src/closures.c [X86_64 && __sun__ && __svr4__]
438 (FFI_MMAP_EXEC_WRIT): Define.
440 2009-12-08 David Daney <ddaney@caviumnetworks.com>
442 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
443 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
444 * testsuite/libffi.call/stret_large.c: Same.
445 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
446 * testsuite/libffi.call/stret_large2.c: Same.
447 * testsuite/libffi.call/stret_medium2.c: Same.
449 2009-12-07 David Edelsohn <edelsohn@gnu.org>
451 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
454 2009-12-05 David Edelsohn <edelsohn@gnu.org>
456 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
458 * src/powerpc/aix_closure.S: Same.
460 2009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
462 * Makefile.in: Regenerate.
463 * configure: Regenerate.
464 * include/Makefile.in: Regenerate.
465 * man/Makefile.in: Regenerate.
466 * testsuite/Makefile.in: Regenerate.
468 2009-12-04 David Edelsohn <edelsohn@gnu.org>
470 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
473 2009-12-04 Uros Bizjak <ubizjak@gmail.com>
476 * src/x86/ffi64.c (classify_argument): Update from
477 gcc/config/i386/i386.c.
478 (ffi_closure_unix64_inner): Do not use the address of two consecutive
479 SSE registers directly.
480 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
481 for x86_64 linux targets.
483 2009-12-04 David Edelsohn <edelsohn@gnu.org>
485 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
486 pfr for long double split between fpr13 and stack.
488 2009-12-03 David Edelsohn <edelsohn@gnu.org>
490 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
491 fparg_count twice for long double.
493 2009-12-03 David Edelsohn <edelsohn@gnu.org>
496 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
498 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
500 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
501 Remove xfails for x86 linux targets.
503 2009-12-02 David Edelsohn <edelsohn@gnu.org>
505 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
508 2009-12-01 David Edelsohn <edelsohn@gnu.org>
510 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
511 register usage. Call ffi_prep_args directly. Add long double
512 return value support.
513 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
514 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
515 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
516 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
518 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
520 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
521 Allocate result area between params and FPRs.
523 2009-11-30 David Edelsohn <edelsohn@gnu.org>
526 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
528 * src/powerpc/aix.S: Implement AIX64 version.
529 * src/powerpc/aix_closure.S: Implement AIX64 version.
530 (ffi_closure_ASM): Use extsb, lha and displament addresses.
531 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
533 (ffi_prep_cif_machdep): Same.
535 (ffi_closure_helper_DARWIN): Same.
537 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
540 * testsuite/libffi.call/testclosure.c: New test.
542 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
544 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
545 assembly version use of ___chkstk.
547 2009-09-23 Matthias Klose <doko@ubuntu.com>
549 PR libffi/40242, PR libffi/41443
550 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
551 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
552 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
553 Change the conditionals to __SOFTFP__ || __ARM_EABI__
554 for -mfloat-abi=softfp to work.
556 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
558 PR testsuite/32843 (strikes again)
559 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
560 enable proper extension on char and short.
562 2009-09-15 David Daney <ddaney@caviumnetworks.com>
564 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
565 handling for FFI_TYPE_POINTER.
566 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
567 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
568 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
569 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
570 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
571 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
572 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
573 structure and pointer returns.
574 (ffi_closure_N32): Add handling for pointer returns.
575 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
576 calc_n32_return_struct_flags): Handle soft-float.
577 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
578 (ffi_call_N32): Declare proper argument types.
579 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
582 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
584 * configure.ac (AC_PREREQ): Bump to 2.64.
586 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
588 * Makefile.am (install-html, install-pdf): Remove.
589 * Makefile.in: Regenerate.
591 * Makefile.in: Regenerate.
592 * aclocal.m4: Regenerate.
593 * configure: Regenerate.
594 * fficonfig.h.in: Regenerate.
595 * include/Makefile.in: Regenerate.
596 * man/Makefile.in: Regenerate.
597 * testsuite/Makefile.in: Regenerate.
599 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
601 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
603 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
606 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
607 return types for X86_WIN32.
608 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
609 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
610 _ffi_closure_STDCALL): Likewise.
612 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
613 (dlmmap, dlmunmap): Also use these functions on Cygwin.
615 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
620 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
623 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
625 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
626 to "" before adding paths. (This reinstates an assignment that
627 was removed by my 2009-06-30 commit, but changes the initial
628 value from "." to "".)
630 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
633 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
634 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
637 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
639 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
640 to ld_library_path. Use add_path. Add just find_libgcc_s
641 to ld_library_path, not every libgcc multilib directory.
643 2009-06-16 Wim Lewis <wiml@hhhh.org>
645 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
646 supposed to be callee-saved.
647 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
648 return buffer for odd-size structs.
650 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
653 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
654 allow_stack_execute for Darwin.
656 2009-06-16 Andrew Haley <aph@redhat.com>
658 * configure.ac (TARGETDIR): Add missing blank lines.
659 * configure: Regenerate.
661 2009-06-16 Andrew Haley <aph@redhat.com>
663 * testsuite/libffi.call/cls_align_sint64.c,
664 testsuite/libffi.call/cls_align_uint64.c,
665 testsuite/libffi.call/cls_longdouble_va.c,
666 testsuite/libffi.call/cls_ulonglong.c,
667 testsuite/libffi.call/return_ll1.c,
668 testsuite/libffi.call/stret_medium2.c: Fix printf format
670 * testsuite/libffi.call/ffitest.h,
671 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
673 2009-06-15 Andrew Haley <aph@redhat.com>
675 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
676 * testsuite/libffi.call/err_bad_abi.c: Likewise.
678 2009-06-12 Andrew Haley <aph@redhat.com>
680 * Makefile.am: Remove info_TEXINFOS.
682 2009-06-12 Andrew Haley <aph@redhat.com>
684 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
685 testsuite/libffi.call/cls_align_uint64.c,
686 testsuite/libffi.call/cls_ulonglong.c,
687 testsuite/libffi.call/return_ll1.c,
688 testsuite/libffi.call/stret_medium2.c: Fix printf format
690 testsuite/libffi.special/unwindtest.cc: include stdint.h.
692 2009-06-11 Timothy Wall <twall@users.sf.net>
697 include/ffi_common.h,
702 src/x86/win64.S (new),
703 README: Added win64 support (mingw or MSVC)
707 testsuite/Makefile.in,
709 aclocal.m4: Regenerated
710 * ltcf-c.sh: properly escape cygwin/w32 path
711 * man/ffi_call.3: Clarify size requirements for return value.
712 * src/x86/ffi64.c: Fix filename in comment.
713 * src/x86/win32.S: Remove unused extern.
715 * testsuite/libffi.call/closure_fn0.c,
716 testsuite/libffi.call/closure_fn1.c,
717 testsuite/libffi.call/closure_fn2.c,
718 testsuite/libffi.call/closure_fn3.c,
719 testsuite/libffi.call/closure_fn4.c,
720 testsuite/libffi.call/closure_fn5.c,
721 testsuite/libffi.call/closure_fn6.c,
722 testsuite/libffi.call/closure_stdcall.c,
723 testsuite/libffi.call/cls_12byte.c,
724 testsuite/libffi.call/cls_16byte.c,
725 testsuite/libffi.call/cls_18byte.c,
726 testsuite/libffi.call/cls_19byte.c,
727 testsuite/libffi.call/cls_1_1byte.c,
728 testsuite/libffi.call/cls_20byte.c,
729 testsuite/libffi.call/cls_20byte1.c,
730 testsuite/libffi.call/cls_24byte.c,
731 testsuite/libffi.call/cls_2byte.c,
732 testsuite/libffi.call/cls_3_1byte.c,
733 testsuite/libffi.call/cls_3byte1.c,
734 testsuite/libffi.call/cls_3byte2.c,
735 testsuite/libffi.call/cls_4_1byte.c,
736 testsuite/libffi.call/cls_4byte.c,
737 testsuite/libffi.call/cls_5_1_byte.c,
738 testsuite/libffi.call/cls_5byte.c,
739 testsuite/libffi.call/cls_64byte.c,
740 testsuite/libffi.call/cls_6_1_byte.c,
741 testsuite/libffi.call/cls_6byte.c,
742 testsuite/libffi.call/cls_7_1_byte.c,
743 testsuite/libffi.call/cls_7byte.c,
744 testsuite/libffi.call/cls_8byte.c,
745 testsuite/libffi.call/cls_9byte1.c,
746 testsuite/libffi.call/cls_9byte2.c,
747 testsuite/libffi.call/cls_align_double.c,
748 testsuite/libffi.call/cls_align_float.c,
749 testsuite/libffi.call/cls_align_longdouble.c,
750 testsuite/libffi.call/cls_align_longdouble_split.c,
751 testsuite/libffi.call/cls_align_longdouble_split2.c,
752 testsuite/libffi.call/cls_align_pointer.c,
753 testsuite/libffi.call/cls_align_sint16.c,
754 testsuite/libffi.call/cls_align_sint32.c,
755 testsuite/libffi.call/cls_align_sint64.c,
756 testsuite/libffi.call/cls_align_uint16.c,
757 testsuite/libffi.call/cls_align_uint32.c,
758 testsuite/libffi.call/cls_align_uint64.c,
759 testsuite/libffi.call/cls_dbls_struct.c,
760 testsuite/libffi.call/cls_double.c,
761 testsuite/libffi.call/cls_double_va.c,
762 testsuite/libffi.call/cls_float.c,
763 testsuite/libffi.call/cls_longdouble.c,
764 testsuite/libffi.call/cls_longdouble_va.c,
765 testsuite/libffi.call/cls_multi_schar.c,
766 testsuite/libffi.call/cls_multi_sshort.c,
767 testsuite/libffi.call/cls_multi_sshortchar.c,
768 testsuite/libffi.call/cls_multi_uchar.c,
769 testsuite/libffi.call/cls_multi_ushort.c,
770 testsuite/libffi.call/cls_multi_ushortchar.c,
771 testsuite/libffi.call/cls_pointer.c,
772 testsuite/libffi.call/cls_pointer_stack.c,
773 testsuite/libffi.call/cls_schar.c,
774 testsuite/libffi.call/cls_sint.c,
775 testsuite/libffi.call/cls_sshort.c,
776 testsuite/libffi.call/cls_uchar.c,
777 testsuite/libffi.call/cls_uint.c,
778 testsuite/libffi.call/cls_ulonglong.c,
779 testsuite/libffi.call/cls_ushort.c,
780 testsuite/libffi.call/err_bad_abi.c,
781 testsuite/libffi.call/err_bad_typedef.c,
782 testsuite/libffi.call/float2.c,
783 testsuite/libffi.call/huge_struct.c,
784 testsuite/libffi.call/nested_struct.c,
785 testsuite/libffi.call/nested_struct1.c,
786 testsuite/libffi.call/nested_struct10.c,
787 testsuite/libffi.call/nested_struct2.c,
788 testsuite/libffi.call/nested_struct3.c,
789 testsuite/libffi.call/nested_struct4.c,
790 testsuite/libffi.call/nested_struct5.c,
791 testsuite/libffi.call/nested_struct6.c,
792 testsuite/libffi.call/nested_struct7.c,
793 testsuite/libffi.call/nested_struct8.c,
794 testsuite/libffi.call/nested_struct9.c,
795 testsuite/libffi.call/problem1.c,
796 testsuite/libffi.call/return_ldl.c,
797 testsuite/libffi.call/return_ll1.c,
798 testsuite/libffi.call/stret_large.c,
799 testsuite/libffi.call/stret_large2.c,
800 testsuite/libffi.call/stret_medium.c,
801 testsuite/libffi.call/stret_medium2.c,
802 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
803 of checking for MMAP. Use intptr_t instead of long casts.
805 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
807 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
808 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
809 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
811 2009-06-09 Andrew Haley <aph@redhat.com>
813 * src/x86/freebsd.S: Add missing file.
815 2009-06-08 Andrew Haley <aph@redhat.com>
817 Import from libffi 3.0.8:
819 * doc/libffi.texi: New file.
820 * doc/libffi.info: Likewise.
821 * doc/stamp-vti: Likewise.
822 * man/Makefile.am: New file.
823 * man/ffi_call.3: New file.
825 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
827 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
829 * configure.ac: Bump version to 3.0.8.
830 parisc*-*-linux*: Add.
831 i386-*-freebsd* | i386-*-openbsd*: Add.
832 powerpc-*-beos*: Add.
833 AM_CONDITIONAL X86_FREEBSD: Add.
834 AC_CONFIG_FILES: Add man/Makefile.
836 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
838 2009-06-08 Andrew Haley <aph@redhat.com>
840 * README: Import from libffi 3.0.8.
842 2009-06-08 Andrew Haley <aph@redhat.com>
844 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
845 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
846 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
847 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
849 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
850 * testsuite/libffi.call/stret_medium.c: Likewise.
851 * testsuite/libffi.call/stret_large2.c: Likewise.
852 * testsuite/libffi.call/stret_large.c: Likewise.
854 2008-12-26 Timothy Wall <twall@users.sf.net>
856 * testsuite/libffi.call/cls_longdouble.c,
857 testsuite/libffi.call/cls_longdouble_va.c,
858 testsuite/libffi.call/cls_align_longdouble.c,
859 testsuite/libffi.call/cls_align_longdouble_split.c,
860 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
861 failures on x86_64 cygwin/mingw.
863 2008-12-22 Timothy Wall <twall@users.sf.net>
865 * testsuite/libffi.call/closure_fn0.c,
866 testsuite/libffi.call/closure_fn1.c,
867 testsuite/libffi.call/closure_fn2.c,
868 testsuite/libffi.call/closure_fn3.c,
869 testsuite/libffi.call/closure_fn4.c,
870 testsuite/libffi.call/closure_fn5.c,
871 testsuite/libffi.call/closure_fn6.c,
872 testsuite/libffi.call/closure_loc_fn0.c,
873 testsuite/libffi.call/closure_stdcall.c,
874 testsuite/libffi.call/cls_align_pointer.c,
875 testsuite/libffi.call/cls_pointer.c,
876 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
877 pointer to integer (intptr_t).
878 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
880 2008-07-24 Anthony Green <green@redhat.com>
882 * testsuite/libffi.call/cls_dbls_struct.c,
883 testsuite/libffi.call/cls_double_va.c,
884 testsuite/libffi.call/cls_longdouble.c,
885 testsuite/libffi.call/cls_longdouble_va.c,
886 testsuite/libffi.call/cls_pointer.c,
887 testsuite/libffi.call/cls_pointer_stack.c,
888 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
891 2008-03-04 Anthony Green <green@redhat.com>
895 * testsuite/libffi.call/cls_align_longdouble_split2.c
896 testsuite/libffi.call/cls_align_longdouble_split.c
897 testsuite/libffi.call/cls_dbls_struct.c
898 testsuite/libffi.call/cls_double_va.c
899 testsuite/libffi.call/cls_longdouble.c
900 testsuite/libffi.call/cls_longdouble_va.c
901 testsuite/libffi.call/cls_pointer.c
902 testsuite/libffi.call/cls_pointer_stack.c
903 testsuite/libffi.call/err_bad_abi.c
904 testsuite/libffi.call/err_bad_typedef.c
905 testsuite/libffi.call/stret_large2.c
906 testsuite/libffi.call/stret_large.c
907 testsuite/libffi.call/stret_medium2.c
908 testsuite/libffi.call/stret_medium.c: New tests from Apple.
910 2009-06-05 Andrew Haley <aph@redhat.com>
912 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
915 2009-06-04 Andrew Haley <aph@redhat.com>
917 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
920 2008-02-26 Anthony Green <green@redhat.com>
921 Thomas Heller <theller@ctypes.org>
923 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
926 2008-02-03 Timothy Wall <twall@users.sf.net>
928 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
929 offset based on code pointer, not data pointer.
931 2008-01-31 Timothy Wall <twall@users.sf.net>
933 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
935 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
937 * src/x86/win32.S: Add assembly for stdcall closure.
938 * src/x86/ffi.c: Initialize stdcall closure trampoline.
940 2009-06-04 Andrew Haley <aph@redhat.com>
942 * include/ffi.h.in: Change void (*)() to void (*)(void).
943 * src/x86/ffi.c: Likewise.
945 2009-06-04 Andrew Haley <aph@redhat.com>
947 * src/powerpc/ppc_closure.S: Insert licence header.
948 * src/powerpc/linux64_closure.S: Likewise.
949 * src/m68k/sysv.S: Likewise.
951 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
952 * src/powerpc/ffi.c: Likewise.
953 * src/powerpc/ffi_darwin.c: Likewise.
954 * src/m32r/ffi.c: Likewise.
955 * src/sh64/ffi.c: Likewise.
956 * src/x86/ffi64.c: Likewise.
957 * src/alpha/ffi.c: Likewise.
958 * src/alpha/osf.S: Likewise.
959 * src/frv/ffi.c: Likewise.
960 * src/s390/ffi.c: Likewise.
961 * src/pa/ffi.c: Likewise.
962 * src/pa/hpux32.S: Likewise.
963 * src/ia64/unix.S: Likewise.
964 * src/ia64/ffi.c: Likewise.
965 * src/sparc/ffi.c: Likewise.
966 * src/mips/ffi.c: Likewise.
967 * src/sh/ffi.c: Likewise.
969 2008-02-15 David Daney <ddaney@avtrex.com>
971 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
972 Define (conditionally), and use it to include cachectl.h.
973 (ffi_prep_closure_loc): Fix cache flushing.
974 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
976 2009-06-04 Andrew Haley <aph@redhat.com>
982 src/powerpc/ffitarget.h,
984 src/sh64/ffitarget.h,
995 src/alpha/ffitarget.h,
998 src/m68k/ffitarget.h,
1001 src/s390/ffitarget.h,
1003 src/cris/ffitarget.h,
1008 src/ia64/ffitarget.h,
1011 src/ia64/ia64_flags.h,
1015 src/sparc/ffitarget.h,
1018 src/mips/ffitarget.h,
1025 src/sh/sysv.S: Update license text.
1027 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
1029 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
1030 (.eh_frame): Add FDE for it.
1032 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
1034 * configure.ac: Also check if assembler supports pc-relative
1035 relocs on X86_WIN32 targets.
1036 * configure: Regenerate.
1037 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
1038 (_ffi_call_SYSV): Add missing function type symbol .def and
1039 add EH markup labels.
1040 (_ffi_call_STDCALL): Likewise.
1041 (_ffi_closure_SYSV): Likewise.
1042 (_ffi_closure_raw_SYSV): Likewise.
1043 (.eh_frame): Add hand-crafted EH data.
1045 2009-04-09 Jakub Jelinek <jakub@redhat.com>
1047 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
1048 version 3 of the GNU General Public License and to point readers
1049 at the COPYING3 file and the FSF's license web page.
1050 * testsuite/libffi.call/call.exp: Likewise.
1051 * testsuite/libffi.special/special.exp: Likewise.
1053 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1055 * configure: Regenerate.
1057 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1060 * configure.ac (HAVE_AS_X86_PCREL): New test.
1061 * configure: Regenerate.
1062 * fficonfig.h.in: Regenerate.
1063 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
1064 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
1065 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
1066 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
1067 * src/x86/unix64.S (.Lstore_table): Move to .text section.
1068 (.Lload_table): Likewise.
1069 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
1071 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1073 * configure: Regenerate.
1075 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
1077 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
1078 signed/unsigned int8/16 return values.
1079 * src/sparc/v8.S (ffi_call_v8): Likewise.
1080 (ffi_closure_v8): Likewise.
1082 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
1083 Steve Ellcey <sje@cup.hp.com>
1085 * configure: Regenerate for new libtool.
1086 * Makefile.in: Ditto.
1087 * include/Makefile.in: Ditto.
1088 * aclocal.m4: Ditto.
1090 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
1092 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
1093 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
1094 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
1095 Adjust copyright notice.
1096 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
1097 register or two register to use for FFI_SYSV structs.
1098 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
1099 (ffi_closure_helper_SYSV): Fix the return type for
1100 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
1101 Adjust copyright notice.
1103 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
1105 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
1108 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1110 * configure: Regenerate.
1111 * include/Makefile.in: Regenerate.
1112 * testsuite/Makefile.in: Regenerate.
1114 2008-06-07 Joseph Myers <joseph@codesourcery.com>
1116 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
1117 powerpc-*-beos*): Remove.
1118 * configure: Regenerate.
1120 2008-05-09 Julian Brown <julian@codesourcery.com>
1122 * Makefile.am (LTLDFLAGS): New.
1123 (libffi_la_LDFLAGS): Use above.
1124 * Makefile.in: Regenerate.
1126 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
1129 * aclocal.m4: Regenerate.
1130 * configure: Regenerate.
1132 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
1134 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
1135 * src/sh64/sysv.S: Likewise.
1137 2008-03-26 Daniel Jacobowitz <dan@debian.org>
1139 * src/arm/sysv.S: Fix ARM comment marker.
1141 2008-03-26 Jakub Jelinek <jakub@redhat.com>
1143 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
1144 * src/s390/sysv.S: Likewise.
1145 * src/powerpc/ppc_closure.S: Likewise.
1146 * src/powerpc/sysv.S: Likewise.
1147 * src/x86/unix64.S: Likewise.
1148 * src/x86/sysv.S: Likewise.
1149 * src/sparc/v8.S: Likewise.
1150 * src/sparc/v9.S: Likewise.
1151 * src/m68k/sysv.S: Likewise.
1152 * src/arm/sysv.S: Likewise.
1154 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1156 * aclocal.m4: Regenerate.
1157 * configure: Likewise.
1158 * Makefile.in: Likewise.
1159 * include/Makefile.in: Likewise.
1160 * testsuite/Makefile.in: Likewise.
1162 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
1163 Andreas Tobler <a.tobler@schweiz.org>
1165 * configure.ac: Add amd64-*-freebsd* target.
1166 * configure: Regenerate.
1168 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
1171 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
1174 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
1177 2008-01-24 David Edelsohn <edelsohn@gnu.org>
1179 * configure: Regenerate.
1181 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
1183 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
1185 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
1188 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
1189 signed/unsigned int8/16 for X86_DARWIN.
1190 Updated copyright info.
1191 Handle one and two byte structs with special cif->flags.
1192 * src/x86/ffitarget.h: Add special types for one and two byte structs.
1193 Updated copyright info.
1194 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
1196 Remove code to pop args from the stack after call.
1197 Special-case signed/unsigned for int8/16, one and two byte structs.
1198 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1199 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1201 Updated copyright info.
1203 2007-12-08 David Daney <ddaney@avtrex.com>
1205 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
1206 SUBU, add with ADDU and use smaller code sequences.
1208 2007-12-07 David Daney <ddaney@avtrex.com>
1210 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
1213 2007-12-06 David Daney <ddaney@avtrex.com>
1215 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
1217 (ffi_java_raw): New typedef.
1218 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
1219 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
1221 (ffi_java_raw_closure) : Same.
1222 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
1224 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
1225 FFI_SIZEOF_JAVA_RAW.
1226 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
1227 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
1228 sizeof(ffi_java_raw) for alignment calculations.
1229 (ffi_java_ptrarray_to_raw): Same.
1230 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
1231 if FFI_SIZEOF_JAVA_RAW == 4.
1232 (ffi_java_raw_to_rvalue): Same.
1233 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
1234 (ffi_java_translate_args): Same.
1235 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
1237 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
1239 2007-12-06 David Daney <ddaney@avtrex.com>
1241 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
1244 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
1247 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
1248 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
1249 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
1250 set the NUM_FPR_ARG_REGISTERS according to.
1251 Add support for potential soft-float support under hard-float
1253 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
1254 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
1255 to the FFI_LINUX_SOFT_FLOAT ABI.
1256 (ffi_prep_cif_machdep): Likewise.
1257 (ffi_closure_helper_SYSV): Likewise.
1258 * src/powerpc/ppc_closure.S: Make sure not to store float/double
1259 on archs where __NO_FPRS__ is true.
1260 Add FFI_TYPE_UINT128 support.
1261 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
1262 Adjust copyright notice.
1264 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
1266 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
1267 * include/ffi_common.h: ... here.
1270 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
1272 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
1273 * src/powerpc/linux64.S: Likewise.
1274 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
1275 * testsuite/libffi.call/return_ldl.c: New test.
1277 2007-09-04 <aph@redhat.com>
1279 * src/arm/sysv.S (UNWIND): New.
1280 (Whole file): Conditionally compile unwinder directives.
1281 * src/arm/sysv.S: Add unwinder directives.
1283 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
1284 Only treat r0 as a struct address if we're actually returning a
1286 Only copy the bytes that are actually within a struct.
1287 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
1288 is returned in r0, not passed by address.
1289 (ffi_call): Allocate a word-sized temporary for the case where
1290 a composite is returned in r0.
1291 (ffi_prep_incoming_args_SYSV): Align as necessary.
1293 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
1295 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
1296 directly using the sys_cacheflush syscall.
1298 2007-07-27 Andrew Haley <aph@redhat.com>
1300 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
1302 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
1304 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
1305 * configure.ac: Likewise.
1306 * Makefile.in: Regenerate.
1307 * include/Makefile.in: Likewise.
1308 * testsuite/Makefile.in: Likewise.
1309 * configure: Likewise.
1311 2007-08-24 David Daney <ddaney@avtrex.com>
1313 * testsuite/libffi.call/return_sl.c: New test.
1315 2007-08-10 David Daney <ddaney@avtrex.com>
1317 * testsuite/libffi.call/cls_multi_ushort.c,
1318 testsuite/libffi.call/cls_align_uint16.c,
1319 testsuite/libffi.call/nested_struct1.c,
1320 testsuite/libffi.call/nested_struct3.c,
1321 testsuite/libffi.call/cls_7_1_byte.c,
1322 testsuite/libffi.call/nested_struct5.c,
1323 testsuite/libffi.call/cls_double.c,
1324 testsuite/libffi.call/nested_struct7.c,
1325 testsuite/libffi.call/cls_sint.c,
1326 testsuite/libffi.call/nested_struct9.c,
1327 testsuite/libffi.call/cls_20byte1.c,
1328 testsuite/libffi.call/cls_multi_sshortchar.c,
1329 testsuite/libffi.call/cls_align_sint64.c,
1330 testsuite/libffi.call/cls_3byte2.c,
1331 testsuite/libffi.call/cls_multi_schar.c,
1332 testsuite/libffi.call/cls_multi_uchar.c,
1333 testsuite/libffi.call/cls_19byte.c,
1334 testsuite/libffi.call/cls_9byte1.c,
1335 testsuite/libffi.call/cls_align_float.c,
1336 testsuite/libffi.call/closure_fn1.c,
1337 testsuite/libffi.call/problem1.c,
1338 testsuite/libffi.call/closure_fn3.c,
1339 testsuite/libffi.call/cls_sshort.c,
1340 testsuite/libffi.call/closure_fn5.c,
1341 testsuite/libffi.call/cls_align_double.c,
1342 testsuite/libffi.call/nested_struct.c,
1343 testsuite/libffi.call/cls_2byte.c,
1344 testsuite/libffi.call/nested_struct10.c,
1345 testsuite/libffi.call/cls_4byte.c,
1346 testsuite/libffi.call/cls_6byte.c,
1347 testsuite/libffi.call/cls_8byte.c,
1348 testsuite/libffi.call/cls_multi_sshort.c,
1349 testsuite/libffi.call/cls_align_sint16.c,
1350 testsuite/libffi.call/cls_align_uint32.c,
1351 testsuite/libffi.call/cls_20byte.c,
1352 testsuite/libffi.call/cls_float.c,
1353 testsuite/libffi.call/nested_struct2.c,
1354 testsuite/libffi.call/cls_5_1_byte.c,
1355 testsuite/libffi.call/nested_struct4.c,
1356 testsuite/libffi.call/cls_24byte.c,
1357 testsuite/libffi.call/nested_struct6.c,
1358 testsuite/libffi.call/cls_64byte.c,
1359 testsuite/libffi.call/nested_struct8.c,
1360 testsuite/libffi.call/cls_uint.c,
1361 testsuite/libffi.call/cls_multi_ushortchar.c,
1362 testsuite/libffi.call/cls_schar.c,
1363 testsuite/libffi.call/cls_uchar.c,
1364 testsuite/libffi.call/cls_align_uint64.c,
1365 testsuite/libffi.call/cls_ulonglong.c,
1366 testsuite/libffi.call/cls_align_longdouble.c,
1367 testsuite/libffi.call/cls_1_1byte.c,
1368 testsuite/libffi.call/cls_12byte.c,
1369 testsuite/libffi.call/cls_3_1byte.c,
1370 testsuite/libffi.call/cls_3byte1.c,
1371 testsuite/libffi.call/cls_4_1byte.c,
1372 testsuite/libffi.call/cls_6_1_byte.c,
1373 testsuite/libffi.call/cls_16byte.c,
1374 testsuite/libffi.call/cls_18byte.c,
1375 testsuite/libffi.call/closure_fn0.c,
1376 testsuite/libffi.call/cls_9byte2.c,
1377 testsuite/libffi.call/closure_fn2.c,
1378 testsuite/libffi.call/closure_fn4.c,
1379 testsuite/libffi.call/cls_ushort.c,
1380 testsuite/libffi.call/closure_fn6.c,
1381 testsuite/libffi.call/cls_5byte.c,
1382 testsuite/libffi.call/cls_align_pointer.c,
1383 testsuite/libffi.call/cls_7byte.c,
1384 testsuite/libffi.call/cls_align_sint32.c,
1385 testsuite/libffi.special/unwindtest_ffi_call.cc,
1386 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
1388 2007-08-10 David Daney <ddaney@avtrex.com>
1391 * configure.ac: Don't treat mips64 as a special case.
1392 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
1393 * configure: Regenerate
1394 * Makefile.in: Ditto.
1395 * fficonfig.h.in: Ditto.
1396 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
1397 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
1398 (FFI_DEFAULT_ABI): Set for n64 case.
1399 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
1400 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
1401 (ffi_closure_N32): New function.
1402 (.eh_frame): New section
1403 * src/mips/o32.S: Clean up comments.
1404 (ffi_closure_O32): Pass ffi_closure parameter in $12.
1405 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
1406 _MIPS_SIM == _ABIN32 throughout.
1407 (FFI_MIPS_STOP_HERE): New, use in place of
1409 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1410 to support n32/n64 ABIs.
1411 (calc_n32_struct_flags): Rewrite.
1412 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1413 position of flag bits.
1414 (ffi_prep_cif_machdep): Rewrite n32 portion.
1415 (ffi_call): Enable for n64. Add special handling for small structure
1417 (ffi_prep_closure_loc): Add n32 and n64 support.
1418 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1419 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1421 2007-08-08 David Daney <ddaney@avtrex.com>
1423 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1424 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1426 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1427 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1428 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1429 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1430 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1431 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1432 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1433 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1434 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1435 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1436 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1437 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1438 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1439 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1440 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1441 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1442 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1443 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1444 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1445 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1446 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1447 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1448 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1449 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1450 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1451 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1452 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1453 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1454 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1455 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1456 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1458 2007-08-07 Andrew Haley <aph@redhat.com>
1460 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1463 2007-08-06 Andrew Haley <aph@redhat.com>
1466 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1467 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1470 2007-08-02 David Daney <ddaney@avtrex.com>
1472 * testsuite/libffi.call/return_ul.c (main): Define return type as
1473 ffi_arg. Use proper printf conversion specifier.
1475 2007-07-30 Andrew Haley <aph@redhat.com>
1478 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1479 signed/unsigned int8/16.
1480 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1482 Remove code to pop args from the stack after call.
1483 Special-case signed/unsigned int8/16.
1484 * testsuite/libffi.call/return_sc.c (main): Revert.
1486 2007-07-26 Richard Guenther <rguenther@suse.de>
1489 * testsuite/libffi.call/return_sc.c (main): Verify call
1490 result as signed char, not ffi_arg.
1492 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1494 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1495 * configure: Regenerate.
1497 2007-07-11 David Daney <ddaney@avtrex.com>
1499 * src/mips/ffi.c: Don't include sys/cachectl.h.
1500 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1503 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1505 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1506 from (ffi_prep_closure): ... this.
1507 (FFI_INIT_TRAMPOLINE): Adjust.
1509 2005-12-31 Phil Blundell <pb@reciva.com>
1511 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1512 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1513 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1514 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1515 (FFI_CLOSURES): Enable closure support.
1517 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1519 * testsuite/libffi.call/cls_multi_ushort.c,
1520 testsuite/libffi.call/cls_align_uint16.c,
1521 testsuite/libffi.call/nested_struct1.c,
1522 testsuite/libffi.call/nested_struct3.c,
1523 testsuite/libffi.call/cls_7_1_byte.c,
1524 testsuite/libffi.call/cls_double.c,
1525 testsuite/libffi.call/nested_struct5.c,
1526 testsuite/libffi.call/nested_struct7.c,
1527 testsuite/libffi.call/cls_sint.c,
1528 testsuite/libffi.call/nested_struct9.c,
1529 testsuite/libffi.call/cls_20byte1.c,
1530 testsuite/libffi.call/cls_multi_sshortchar.c,
1531 testsuite/libffi.call/cls_align_sint64.c,
1532 testsuite/libffi.call/cls_3byte2.c,
1533 testsuite/libffi.call/cls_multi_schar.c,
1534 testsuite/libffi.call/cls_multi_uchar.c,
1535 testsuite/libffi.call/cls_19byte.c,
1536 testsuite/libffi.call/cls_9byte1.c,
1537 testsuite/libffi.call/cls_align_float.c,
1538 testsuite/libffi.call/closure_fn1.c,
1539 testsuite/libffi.call/problem1.c,
1540 testsuite/libffi.call/closure_fn3.c,
1541 testsuite/libffi.call/cls_sshort.c,
1542 testsuite/libffi.call/closure_fn5.c,
1543 testsuite/libffi.call/cls_align_double.c,
1544 testsuite/libffi.call/cls_2byte.c,
1545 testsuite/libffi.call/nested_struct.c,
1546 testsuite/libffi.call/nested_struct10.c,
1547 testsuite/libffi.call/cls_4byte.c,
1548 testsuite/libffi.call/cls_6byte.c,
1549 testsuite/libffi.call/cls_8byte.c,
1550 testsuite/libffi.call/cls_multi_sshort.c,
1551 testsuite/libffi.call/cls_align_uint32.c,
1552 testsuite/libffi.call/cls_align_sint16.c,
1553 testsuite/libffi.call/cls_float.c,
1554 testsuite/libffi.call/cls_20byte.c,
1555 testsuite/libffi.call/cls_5_1_byte.c,
1556 testsuite/libffi.call/nested_struct2.c,
1557 testsuite/libffi.call/cls_24byte.c,
1558 testsuite/libffi.call/nested_struct4.c,
1559 testsuite/libffi.call/nested_struct6.c,
1560 testsuite/libffi.call/cls_64byte.c,
1561 testsuite/libffi.call/nested_struct8.c,
1562 testsuite/libffi.call/cls_uint.c,
1563 testsuite/libffi.call/cls_multi_ushortchar.c,
1564 testsuite/libffi.call/cls_schar.c,
1565 testsuite/libffi.call/cls_uchar.c,
1566 testsuite/libffi.call/cls_align_uint64.c,
1567 testsuite/libffi.call/cls_ulonglong.c,
1568 testsuite/libffi.call/cls_align_longdouble.c,
1569 testsuite/libffi.call/cls_1_1byte.c,
1570 testsuite/libffi.call/cls_12byte.c,
1571 testsuite/libffi.call/cls_3_1byte.c,
1572 testsuite/libffi.call/cls_3byte1.c,
1573 testsuite/libffi.call/cls_4_1byte.c,
1574 testsuite/libffi.call/cls_6_1_byte.c,
1575 testsuite/libffi.call/cls_16byte.c,
1576 testsuite/libffi.call/cls_18byte.c,
1577 testsuite/libffi.call/closure_fn0.c,
1578 testsuite/libffi.call/cls_9byte2.c,
1579 testsuite/libffi.call/closure_fn2.c,
1580 testsuite/libffi.call/closure_fn4.c,
1581 testsuite/libffi.call/cls_ushort.c,
1582 testsuite/libffi.call/closure_fn6.c,
1583 testsuite/libffi.call/cls_5byte.c,
1584 testsuite/libffi.call/cls_align_pointer.c,
1585 testsuite/libffi.call/cls_7byte.c,
1586 testsuite/libffi.call/cls_align_sint32.c,
1587 testsuite/libffi.special/unwindtest_ffi_call.cc,
1588 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1590 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1592 * aclocal.m4: Regenerated.
1594 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1596 * configure: Regenerate.
1598 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1600 * Makefile.in: Regenerate.
1601 * configure: Regenerate.
1602 * aclocal.m4: Regenerate.
1603 * include/Makefile.in: Regenerate.
1604 * testsuite/Makefile.in: Regenerate.
1606 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1608 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1609 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1610 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1611 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1612 (FFI_CLOSURES): Enable closure support.
1614 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1616 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1617 * configure: Regenerate.
1618 * fficonfig.h.in: Regenerate.
1619 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1620 CFI_OFFSET,CFI_DEF_CFA): New macros.
1621 (ffi_call_SYSV): Add callframe annotation.
1623 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1625 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1626 numerous test suite failures.
1627 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1629 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1631 * Makefile.am (EXTRA_DIST): Bring up to date.
1632 * Makefile.in: Regenerate.
1633 * src/frv/eabi.S: Remove RCS keyword.
1635 2007-04-06 Richard Henderson <rth@redhat.com>
1637 * configure.ac: Tidy target case.
1638 (HAVE_LONG_DOUBLE): Allow the target to override.
1639 * configure: Regenerate.
1640 * include/ffi.h.in: Don't define ffi_type_foo if
1641 LIBFFI_HIDE_BASIC_TYPES is defined.
1642 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1644 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1645 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1646 (ffi_type_longdouble): Special case for Alpha. Don't define
1647 if long double == double.
1649 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1650 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1651 (ffi_call, ffi_closure_osf_inner): Likewise.
1652 (ffi_closure_osf_inner): Likewise. Mark hidden.
1653 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1654 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1655 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1656 (load_table): Handle 128-bit long double.
1658 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1660 2007-04-06 Tom Tromey <tromey@redhat.com>
1663 * README: Fixed bug in example.
1665 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1667 * src/closures.c: Include sys/statfs.h.
1668 (_GNU_SOURCE): Define on Linux.
1669 (FFI_MMAP_EXEC_SELINUX): Define.
1670 (selinux_enabled): New variable.
1671 (selinux_enabled_check): New function.
1672 (is_selinux_enabled): Define.
1675 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1677 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1678 Use 'volatile float sum' to create sum of floats to avoid false
1679 negative due to excess precision on ix86 targets.
1682 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1684 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1686 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1688 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1690 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1691 (ffi_prep_closure_loc): New.
1692 (ffi_prep_raw_closure_loc): New.
1693 (ffi_prep_java_raw_closure_loc): New.
1694 * src/closures.c: New file.
1695 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1696 Replace sflags with exec_offset.
1697 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1698 sub_segment_exec_offset): New macros.
1699 (get_segment_flags, set_segment_flags, check_segment_merge): New
1701 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1702 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1703 destroy_mspace): Use new macros.
1704 (sys_alloc): Silence warning.
1705 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1706 * Makefile.in: Rebuilt.
1707 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1708 terms of ffi_prep_closure_loc.
1709 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1711 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1713 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1715 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1716 the renamed version.
1717 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1718 (ffi_prep_closure): ... this.
1719 * src/pa/ffi.c: Likewise.
1720 * src/cris/ffi.c: Likewise. Adjust.
1721 * src/frv/ffi.c: Likewise.
1722 * src/ia64/ffi.c: Likewise.
1723 * src/mips/ffi.c: Likewise.
1724 * src/powerpc/ffi_darwin.c: Likewise.
1725 * src/s390/ffi.c: Likewise.
1726 * src/sh/ffi.c: Likewise.
1727 * src/sh64/ffi.c: Likewise.
1728 * src/sparc/ffi.c: Likewise.
1729 * src/x86/ffi64.c: Likewise.
1730 * src/x86/ffi.c: Likewise.
1731 (FFI_INIT_TRAMPOLINE): Adjust.
1732 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1733 (ffi_prep_raw_closure): ... this.
1734 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1735 (ffi_prep_closure): ... this.
1736 (flush_icache): Adjust.
1738 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1740 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1743 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1745 * Makefile.am: Add dummy install-pdf target.
1746 * Makefile.in: Regenerate
1748 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1750 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1751 ffi_closure_helper_SYSV): Add long double handling.
1753 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1755 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1756 immediately after bctrl instruction.
1758 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1760 * Makefile.am (all-recursive, install-recursive,
1761 mostlyclean-recursive, clean-recursive, distclean-recursive,
1762 maintainer-clean-recursive): Add missing targets.
1763 * Makefile.in: Rebuilt.
1765 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1767 * configure.ac: Add TARGET for x86_64-*-darwin*.
1768 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1770 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1771 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1772 * configure: Regenerate.
1773 * Makefile.in: Regenerate.
1774 * include/Makefile.in: Regenerate.
1775 * testsuite/Makefile.in: Regenerate.
1776 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1779 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1781 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1784 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1786 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1787 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1789 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1790 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1792 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1793 Sandro Tolaini <tolaini@libero.it>
1795 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1797 * configure: Regenerated.
1798 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1799 (EXTRA_DIST): Add src/x86/darwin.S.
1800 * Makefile.in: Regenerated.
1801 * include/Makefile.in: Regenerated.
1802 * testsuite/Makefile.in: Regenerated.
1804 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1805 X86_WIN32, and additionally align stack to 16 bytes.
1806 * src/x86/darwin.S: New, based on sysv.S.
1807 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1809 2006-09-12 David Daney <ddaney@avtrex.com>
1812 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1813 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1814 * aclocal.m4: Regenerated for automake 1.9.6.
1815 * Makefile.in: Regenerated.
1816 * include/Makefile.in: Regenerated.
1817 * testsuite/Makefile.in: Regenerated.
1819 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1821 * include/ffi_common.h (struct): Revert accidental commit.
1823 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1825 * include/ffi_common.h: Remove lint directives.
1826 * include/ffi.h.in: Likewise.
1828 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1830 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1831 for 32-bit architectures.
1832 * testsuite/libffi.call/return_ul.c: New test case.
1834 2006-07-19 David Daney <ddaney@avtrex.com>
1836 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1837 xfail remains for mips64.
1839 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1841 * Makefile.am: Add install-html target. Add install-html to .PHONY
1842 * Makefile.in: Regenerate.
1843 * aclocal.m4: Regenerate.
1844 * include/Makefile.in: Regenerate.
1845 * testsuite/Makefile.in: Regenerate.
1847 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1849 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1852 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1854 * README: Remove notice about 'Crazy Comments'.
1855 * src/debug.c: Remove lint directives. Cleanup white spaces.
1856 * src/java_raw_api.c: Likewise.
1857 * src/prep_cif.c: Likewise.
1858 * src/raw_api.c: Likewise.
1859 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1861 * src/arm/ffi.c: Remove lint directives.
1862 * src/m32r/ffi.c: Likewise.
1863 * src/pa/ffi.c: Likewise.
1864 * src/powerpc/ffi.c: Likewise.
1865 * src/powerpc/ffi_darwin.c: Likewise.
1866 * src/sh/ffi.c: Likewise.
1867 * src/sh64/ffi.c: Likewise.
1868 * src/x86/ffi.c: Likewise.
1869 * testsuite/libffi.call/float2.c: Likewise.
1870 * testsuite/libffi.call/promotion.c: Likewise.
1871 * testsuite/libffi.call/struct1.c: Likewise.
1873 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1875 * src/pa/hpux32.S: Correct unwind offset calculation for
1877 * src/pa/linux.S: Likewise.
1879 2006-04-12 James E Wilson <wilson@specifix.com>
1882 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1883 (hfa_type_load): Call stf_spill.
1884 (hfa_type_store): Call ldf_fill.
1885 (ffi_call): Adjust calls to above routines. Add local temps for
1888 2006-04-10 Matthias Klose <doko@debian.org>
1890 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1891 directory names containing underscores.
1893 2006-04-07 James E Wilson <wilson@specifix.com>
1895 * testsuite/libffi.call/float4.c: New testcase.
1897 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1898 Andreas Tobler <a.tobler@schweiz.ch>
1900 * Makefile.am: Add PA_HPUX port.
1901 * Makefile.in: Regenerate.
1902 * include/Makefile.in: Likewise.
1903 * testsuite/Makefile.in: Likewise.
1904 * configure.ac: Add PA_HPUX rules.
1905 * configure: Regenerate.
1906 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1907 Add PA_HPUX and PA64_HPUX.
1908 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1909 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1910 (FFI_TYPE_SMALL_STRUCT2): Define.
1911 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1912 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1913 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1914 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1915 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1916 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1917 * src/pa/ffi.c (ROUND_DOWN): Delete.
1918 (fldw, fstw, fldd, fstd): Use '__asm__'.
1919 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1920 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1921 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1922 Simplify incrementing of stack slot variable. Change type of local
1923 'n' to unsigned int.
1924 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1926 (ffi_prep_cif_machdep): Likewise.
1927 (ffi_call): Likewise.
1928 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1929 return type to ffi_status. Simplify incrementing of stack slot
1930 variable. Only copy floating point argument registers when PA_LINUX
1931 is true. Reformat debug statement.
1932 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1933 FFI_TYPE_SMALL_STRUCT8.
1934 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1936 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1937 Add nops to cache flush. Add trampoline for PA_HPUX.
1938 * src/pa/hpux32.S: New file.
1939 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1940 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1941 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1942 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1943 argument type checks so that common argument types appear first.
1944 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1945 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1947 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1949 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1950 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1951 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1952 FFI_TYPE_LONGDOUBLE.
1953 (ffi_prep_args64): Assert using IBM extended double.
1954 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1955 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1956 (ffi_call): Handle FFI_LINUX.
1957 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1958 gpr3 return pointer as for struct return. Handle FFI_LINUX
1959 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1961 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1962 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1963 Don't use r6 as pointer to results, instead use sp offset. Don't
1964 make a special call to load lr with case table address, instead
1965 use offset from previous call.
1966 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1967 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1970 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1972 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1973 passed with FP registers correctly.
1974 (ffi_closure_helper_SYSV): Likewise.
1975 * src/sh64/sysv.S: Likewise.
1977 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1979 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1980 args and userdata unused.
1981 (closure_test_fn1): Mark cif and userdata unused.
1982 (main): Remove unused res.
1984 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1986 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1987 -O2, -O3, -Os and the warning flags -W -Wall.
1988 * testsuite/libffi.special/special.exp: Likewise.
1989 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1990 unused parameter unused for gcc or else do nothing.
1991 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1992 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1993 and userdata unused.
1994 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1995 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1996 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1997 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1998 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1999 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
2000 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
2001 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
2002 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
2003 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
2004 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
2005 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
2006 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
2007 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
2008 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
2009 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
2010 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
2011 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
2012 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
2013 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
2014 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
2015 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
2016 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
2017 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
2019 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
2021 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
2023 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
2024 void* to avoid compiler warning.
2026 (cls_struct_align_gn): Mark cif and userdata unused.
2027 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
2029 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
2031 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
2033 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
2035 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
2037 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
2038 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
2039 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
2041 (main): Cast res_call to silence gcc.
2042 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
2044 (main): Cast res_call to silence gcc.
2045 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
2047 (main): Cast res_call to silence gcc.
2048 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
2050 (main): Cast res_call to silence gcc.
2051 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
2053 (main): Cast res_call to silence gcc.
2054 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
2056 (main): Cast res_call to silence gcc.
2057 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
2059 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
2060 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
2062 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
2063 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
2065 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
2066 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
2068 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
2069 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
2071 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
2072 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
2073 and userdata unused.
2074 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
2076 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
2077 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
2078 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
2079 Cleanup white spaces.
2080 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
2081 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
2082 cif and userdata unused.
2083 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
2085 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
2086 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
2087 formatters to silence gcc.
2088 (B_gn): Mark cif and userdata unused.
2089 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
2091 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
2092 (B_gn): Mark cif and userdata unused.
2093 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
2095 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
2096 (B_gn): Mark cif and userdata unused.
2097 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
2099 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
2100 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
2101 * testsuite/libffi.call/problem1.c (stub): Likewise.
2102 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
2104 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
2105 in the last commit for this test case in the test case itself.
2106 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
2108 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
2109 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
2110 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
2111 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
2112 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
2113 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
2115 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
2117 * src/sh/sysv.S: Fix register numbers in the FDE for
2120 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
2122 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
2123 declaration to avoid a false negative on ix86. See PR323.
2125 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
2127 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
2128 and cast integer to void * if needed. Update the pointer to
2129 the FP register saved area correctly.
2131 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
2133 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
2135 * testsuite/libffi.call/nested_struct4.c: Likewise.
2137 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
2139 * testsuite/libffi.call/return_dbl.c: New test case.
2140 * testsuite/libffi.call/return_dbl1.c: Likewise.
2141 * testsuite/libffi.call/return_dbl2.c: Likewise.
2142 * testsuite/libffi.call/return_fl.c: Likewise.
2143 * testsuite/libffi.call/return_fl1.c: Likewise.
2144 * testsuite/libffi.call/return_fl2.c: Likewise.
2145 * testsuite/libffi.call/return_fl3.c: Likewise.
2146 * testsuite/libffi.call/closure_fn6.c: Likewise.
2148 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
2150 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
2151 here to be used by other test cases too.
2153 * testsuite/libffi.call/nested_struct10.c: New test case.
2154 * testsuite/libffi.call/nested_struct9.c: Likewise.
2155 * testsuite/libffi.call/nested_struct8.c: Likewise.
2156 * testsuite/libffi.call/nested_struct7.c: Likewise.
2157 * testsuite/libffi.call/nested_struct6.c: Likewise.
2158 * testsuite/libffi.call/nested_struct5.c: Likewise.
2159 * testsuite/libffi.call/nested_struct4.c: Likewise.
2161 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2163 * configure.ac: Enable libffi for sparc64-*-freebsd*.
2164 * configure: Rebuilt.
2166 2006-01-18 Jakub Jelinek <jakub@redhat.com>
2168 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
2169 instead do the shifting inline.
2170 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
2171 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
2172 and 8 byte structs, for the remaining struct sizes don't call
2173 __lshrdi3, instead do the shifting inline.
2175 2005-12-07 Thiemo Seufer <ths@networkno.de>
2177 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
2178 missing parentheses.
2179 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
2180 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
2181 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
2182 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
2183 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
2185 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
2187 (ffi_prep_closure): Improve trampoline instruction scheduling.
2188 (ffi_closure_mips_inner_O32): Fix endianness bugs.
2190 2005-12-03 Alan Modra <amodra@bigpond.net.au>
2192 * src/powerpc/ffi.c: Formatting.
2193 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
2194 (ffi_prep_args64): Likewise.
2196 2005-09-30 Geoffrey Keating <geoffk@apple.com>
2198 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
2199 darwin, use -shared-libgcc not -lgcc_s, and explain why.
2201 2005-09-26 Tom Tromey <tromey@redhat.com>
2203 * testsuite/libffi.call/float1.c (value_type): New typedef.
2204 (CANARY): New define.
2205 (main): Check for result buffer overflow.
2206 * src/powerpc/linux64.S: Handle linux64 long double returns.
2207 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
2208 (ffi_prep_cif_machdep): Handle linux64 long double returns.
2210 2005-08-25 Alan Modra <amodra@bigpond.net.au>
2213 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
2215 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
2217 2005-08-11 Jakub Jelinek <jakub@redhat.com>
2219 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
2220 (AH_BOTTOM): Add FFI_HIDDEN definition.
2221 * configure: Rebuilt.
2222 * fficonfig.h.in: Rebuilt.
2223 * src/powerpc/ffi.c (hidden): Remove.
2224 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
2225 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
2226 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
2227 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
2228 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
2229 add FFI_HIDDEN to its prototype.
2230 (ffi_closure_SYSV_inner): New.
2231 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2232 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2234 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
2237 * configure: Rebuilt.
2238 * configure.ac: Handle i*86-*-gnu*.
2240 2005-08-09 Jakub Jelinek <jakub@redhat.com>
2242 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
2243 DW_CFA_offset_extended_sf rather than
2244 DW_CFA_GNU_negative_offset_extended.
2245 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
2247 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
2249 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
2251 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
2252 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
2253 partially on register.
2254 (ffi_closure_helper_SYSV): Likewise.
2255 (ffi_prep_cif_machdep): Don't set too many cif->flags.
2257 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
2259 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
2261 * src/sh64/ffi.c (simple_type): Remove.
2262 (return_type): Handle small structures correctly.
2263 (ffi_prep_args): Likewise.
2264 (ffi_call): Likewise.
2265 (ffi_closure_helper_SYSV): Likewise.
2266 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
2267 Emit position independent code if PIC and remove wrong datalabel
2268 prefixes from EH data.
2270 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
2272 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
2273 * Makefile.in: Regenerate.
2274 * include/Makefile.in: Likewise.
2275 * testsuite/Makefile.in: Likewise.
2276 * configure.ac: Add POWERPC_FREEBSD rules.
2277 * configure: Regenerate.
2278 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
2279 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
2280 * src/powerpc/ffi.c: Add flags to handle small structure returns
2282 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
2284 (ffi_closure_helper_SYSV): Likewise.
2285 * src/powerpc/ppc_closure.S: Add return types for small structures.
2286 * src/powerpc/sysv.S: Add bits to handle small structures for
2289 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
2291 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
2292 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
2293 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
2295 2005-07-05 Randolph Chung <tausq@debian.org>
2297 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
2298 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
2299 structures. Kill compilation warnings.
2300 (ffi_closure_inner_LINUX): Print return values as hex in debug
2301 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
2302 Properly handle 5-7 byte structure returns.
2303 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
2304 (FFI_TYPE_SMALL_STRUCT2): Remove.
2305 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
2306 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
2307 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
2308 (checksmst1, checksmst2): Remove.
2309 (checksmst3): Optimize handling of 3-byte struct returns.
2310 (checksmst567): Properly handle 5-7 byte struct returns.
2312 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2315 * src/mips/n32.S: Enforce PIC code.
2316 * src/mips/o32.S: Likewise.
2318 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2320 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
2321 * configure: Regenerate.
2323 2005-06-01 Alan Modra <amodra@bigpond.net.au>
2325 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
2326 to call ffi_closure_helper_SYSV. Append @local instead.
2327 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
2329 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
2331 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
2332 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
2333 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
2334 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
2335 include/Makefile.in, testsuite/Makefile.in: Regenerate.
2337 2005-05-09 Mike Stump <mrs@apple.com>
2339 * configure: Regenerate.
2341 2005-05-08 Richard Henderson <rth@redhat.com>
2344 * src/alpha/osf.S: Update unwind into to match code.
2346 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
2347 Richard Henderson <rth@redhat.com>
2349 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
2351 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
2352 (ffi_prep_closure): Set carry bit if sse-used flag set.
2353 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
2354 Only load sse registers if ssecount non-zero.
2355 (ffi_closure_unix64): Only save sse registers if carry set on entry.
2357 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
2359 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
2360 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
2361 * configure: Regenerate.
2363 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
2365 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
2366 have Tcl8.3-compatible intermediate variable.
2368 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
2369 Hans-Peter Nilsson <hp@axis.com>
2371 * Makefile.am: Add CRIS support.
2372 * configure.ac: Likewise.
2373 * Makefile.in, configure, testsuite/Makefile.in,
2374 include/Makefile.in: Regenerate.
2375 * src/cris: New directory.
2376 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
2377 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
2379 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
2380 \r?\n in output tests.
2382 2005-04-12 Mike Stump <mrs@apple.com>
2384 * configure: Regenerate.
2386 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
2388 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
2390 2005-03-30 Steve Ellcey <sje@cup.hp.com>
2392 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
2394 * src/ia64/unix.S (ffi_closure_unix): Extend gp
2395 to 64 bits in ILP32 mode.
2396 Load 64 bits even for short data.
2398 2005-03-23 Mike Stump <mrs@apple.com>
2400 * src/powerpc/darwin.S: Update for -m64 multilib.
2401 * src/powerpc/darwin_closure.S: Likewise.
2403 2005-03-21 Zack Weinberg <zack@codesourcery.com>
2405 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
2406 Do not set tool_include_dir.
2407 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
2409 * include/Makefile.am: Set gcc_version and toollibffidir.
2410 * include/Makefile.in: Regenerate.
2412 2005-02-22 Andrew Haley <aph@redhat.com>
2414 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2415 odd-numbered register pairs for 64-bit integer types.
2417 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2420 * testsuite/libffi.call/return_ll1.c: New test case.
2422 2005-02-11 Janis Johnson <janis187@us.ibm.com>
2424 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2425 * testsuite/libffi.call/float.c: Ditto.
2426 * testsuite/libffi.call/float2.c: Ditto.
2427 * testsuite/libffi.call/float3.c: Ditto.
2429 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2431 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2433 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2435 * testsuite/libffi.special/special.exp (cxx_options): Add
2438 2004-12-31 Richard Henderson <rth@redhat.com>
2440 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2441 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2442 offset parameters with a type parameter; deduce size and structure
2443 alignment. Update all users.
2445 2004-12-31 Richard Henderson <rth@redhat.com>
2447 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2448 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
2449 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
2450 into ffi_prep_closure.
2451 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2454 2004-12-27 Richard Henderson <rth@redhat.com>
2456 * src/x86/unix64.S: Fix typo in unwind info.
2458 2004-12-25 Richard Henderson <rth@redhat.com>
2460 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2461 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2462 (merge_classes): Check for it.
2464 (classify_argument): Pass byte_offset by value; perform all updates
2466 (examine_argument): Add classes argument; handle
2467 X86_64_COMPLEX_X87_CLASS.
2468 (ffi_prep_args): Merge into ...
2469 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2470 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2471 (ffi_fill_return_value): Remove.
2472 (ffi_prep_closure): Remove dead assert.
2473 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2474 Rewrite to use struct register_args instead of va_list. Create
2475 flags for handling structure returns.
2476 * src/x86/unix64.S: Remove dead strings.
2477 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2478 stack frame with ffi_call. Handle structure returns properly.
2479 (float2sse, floatfloat2sse, double2sse): Remove.
2480 (sse2float, sse2double, sse2floatfloat): Remove.
2481 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2482 to handle structure returns properly.
2484 2004-12-08 David Edelsohn <edelsohn@gnu.org>
2486 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2488 * Makefile.in: Regenerated.
2490 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2492 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2493 * configure, aclocal.m4, Makefile.in: Regenerate.
2494 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2496 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2498 * configure: Regenerate for libtool change.
2500 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2502 * configure: Regenerate for libtool reversion.
2504 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2506 * configure: Regenerate for libtool change.
2508 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2510 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2512 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2514 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2515 of jal. Use an absolute encoding for the frame information.
2517 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2519 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2520 * acinclude.m4: Delete logic for sincludes.
2521 * aclocal.m4, Makefile.in, configure: Regenerate.
2522 * include/Makefile: Likewise.
2523 * testsuite/Makefile: Likewise.
2525 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2527 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2528 on a 8-byte boundary.
2529 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2531 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2533 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2534 long long values. Round stack allocation to a multiple of 8 bytes
2535 for ATPCS compatibility.
2536 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2537 names. Handle returning long long types. Add Thumb and interworking
2538 support. Improve soft-float code.
2540 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2542 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2543 (libffi_exit): New function.
2544 (libffi_init): Build the testglue wrapper if needed.
2546 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2549 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2551 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2553 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2555 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2557 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2558 * testsuite/libffi.call/float3.c: New test case.
2560 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2562 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2563 the function returning a structure pointed with R2.
2564 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2565 the structure return value if T bit set. Emit position
2566 independent code and EH data if PIC.
2568 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2570 * Makefile.am: Add m32r support.
2571 * configure.ac: Likewise.
2572 * Makefile.in: Regenerate.
2573 * confiugre: Regenerate.
2574 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2575 (uint64, sint64, double, longdouble)
2576 * src/m32r: New directory.
2577 * src/m32r/ffi.c: New file.
2578 * src/m32r/sysv.S: Likewise.
2579 * src/m32r/ffitarget.h: Likewise.
2581 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2583 * testsuite/libffi.call/negint.c: New test case.
2585 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2588 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2589 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2590 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2593 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2595 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2596 * testsuite/libffi.call/promotion.c: Likewise.
2597 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2599 * testsuite/libffi.call/return_sc.c: Likewise.
2600 * testsuite/libffi.call/return_uc.c: Likewise.
2602 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2604 * src/powerpc/darwin.S: Fix comments and identation.
2605 * src/powerpc/darwin_closure.S: Likewise.
2607 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2609 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2610 (ffi_prep_args): Handle longdouble arguments.
2611 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2613 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2614 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2616 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2617 * src/types.c: Defined longdouble size and alignment for darwin.
2619 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2621 * src/powerpc/aix.S: Remove whitespaces.
2622 * src/powerpc/aix_closure.S: Likewise.
2623 * src/powerpc/asm.h: Likewise.
2624 * src/powerpc/ffi.c: Likewise.
2625 * src/powerpc/ffitarget.h: Likewise.
2626 * src/powerpc/linux64.S: Likewise.
2627 * src/powerpc/linux64_closure.S: Likewise.
2628 * src/powerpc/ppc_closure.S: Likewise.
2629 * src/powerpc/sysv.S: Likewise.
2631 2004-08-30 Anthony Green <green@redhat.com>
2633 * Makefile.am: Add frv support.
2634 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2635 * configure.ac: Read configure.host.
2636 * configure.in: Read configure.host.
2637 * configure.host: New file. frv-elf needs libgloss.
2638 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2639 alignment. This is needed to frv and shouldn't harm the others.
2640 * include/ffi_common.h (ALIGN_DOWN): New macro.
2641 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2643 2004-08-24 David Daney <daney@avtrex.com>
2645 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2646 * testsuite/libffi.call/closure_fn1.c: Likewise.
2647 * testsuite/libffi.call/closure_fn2.c Likewise.
2648 * testsuite/libffi.call/closure_fn3.c: Likewise.
2649 * testsuite/libffi.call/closure_fn4.c: Likewise.
2650 * testsuite/libffi.call/closure_fn5.c: Likewise.
2651 * testsuite/libffi.call/cls_18byte.c: Likewise.
2652 * testsuite/libffi.call/cls_19byte.c: Likewise.
2653 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2654 * testsuite/libffi.call/cls_20byte.c: Likewise.
2655 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2656 * testsuite/libffi.call/cls_24byte.c: Likewise.
2657 * testsuite/libffi.call/cls_2byte.c: Likewise.
2658 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2659 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2660 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2661 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2662 * testsuite/libffi.call/cls_4byte.c: Likewise.
2663 * testsuite/libffi.call/cls_64byte.c: Likewise.
2664 * testsuite/libffi.call/cls_6byte.c: Likewise.
2665 * testsuite/libffi.call/cls_7byte.c: Likewise.
2666 * testsuite/libffi.call/cls_8byte.c: Likewise.
2667 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2668 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2669 * testsuite/libffi.call/cls_align_double.c: Likewise.
2670 * testsuite/libffi.call/cls_align_float.c: Likewise.
2671 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2672 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2673 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2674 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2675 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2676 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2677 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2678 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2679 * testsuite/libffi.call/cls_double.c: Likewise.
2680 * testsuite/libffi.call/cls_float.c: Likewise.
2681 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2682 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2683 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2684 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2685 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2686 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2687 * testsuite/libffi.call/cls_schar.c: Likewise.
2688 * testsuite/libffi.call/cls_sint.c: Likewise.
2689 * testsuite/libffi.call/cls_sshort.c: Likewise.
2690 * testsuite/libffi.call/cls_uchar.c: Likewise.
2691 * testsuite/libffi.call/cls_uint.c: Likewise.
2692 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2693 * testsuite/libffi.call/cls_ushort.c: Likewise.
2694 * testsuite/libffi.call/nested_struct.c: Likewise.
2695 * testsuite/libffi.call/nested_struct1.c: Likewise.
2696 * testsuite/libffi.call/nested_struct2.c: Likewise.
2697 * testsuite/libffi.call/nested_struct3.c: Likewise.
2698 * testsuite/libffi.call/problem1.c: Likewise.
2699 * testsuite/libffi.special/unwindtest.cc: Likewise.
2700 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2702 * testsuite/libffi.call/cls_16byte.c: Likewise.
2703 * testsuite/libffi.call/cls_5byte.c: Likewise.
2705 2004-08-23 David Daney <daney@avtrex.com>
2708 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2709 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2710 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2711 parameters and return types.
2712 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2713 (ffi_prep_closure): Ditto.
2714 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2715 alignment calculations.
2716 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2717 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2719 2004-08-14 Casey Marshall <csm@gnu.org>
2721 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2722 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2723 integer (O32 ABI only).
2724 (ffi_prep_closure): new function.
2725 (ffi_closure_mips_inner_O32): new function.
2726 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2727 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2728 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2729 64 bit integers correctly.
2730 (ffi_closure_O32): new function.
2731 Added DWARF-2 unwind info for both functions.
2733 2004-08-10 Andrew Haley <aph@redhat.com>
2735 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2737 2004-08-01 Robert Millan <robertmh@gnu.org>
2739 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2740 * configure: Regenerate.
2742 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2744 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2745 and mmap() explicitly instead of relying on preset autoconf cache
2747 * aclocal.m4: Regenerate.
2748 * configure: Regenerate.
2750 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2752 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2753 (ffi_check_float_struct): Remove unused prototype.
2755 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2757 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2758 character on Darwin, use '\n\t' instead.
2760 2004-06-26 Matthias Klose <doko@debian.org>
2762 * libtool-version: Fix typo in revision/age.
2764 2004-06-17 Matthias Klose <doko@debian.org>
2766 * libtool-version: New.
2767 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2768 * Makefile.in: Regenerate.
2770 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2772 * Makefile.am: Remove useless multilib rules.
2773 * Makefile.in: Regenerate.
2774 * aclocal.m4: Regenerate with automake 1.8.5.
2775 * configure.ac: Remove useless multilib configury.
2776 * configure: Regenerate.
2778 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2780 * .cvsignore: New file.
2782 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2784 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2786 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2789 2004-06-06 Sean McNeil <sean@mcneil.com>
2791 * configure.ac: Add x86_64-*-freebsd* support.
2792 * configure: Regenerate.
2794 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2797 * configure.ac: Test for existence of mmap and sys/mman.h before
2798 checking blacklist. Fix suggested by Jim Wilson.
2799 * configure: Regenerate.
2801 2004-04-26 Matt Austern <austern@apple.com>
2803 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2805 * src/powerpc/darwin_closure.S: Likewise.
2807 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2809 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2810 error. Reported by Thomas Heller <theller@python.net>.
2811 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2812 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2814 2004-03-20 Matthias Klose <doko@debian.org>
2816 * src/pa/linux.S: Fix typo.
2818 2004-03-19 Matthias Klose <doko@debian.org>
2820 * Makefile.am: Update.
2821 * Makefile.in: Regenerate.
2822 * src/pa/ffi.h.in: Remove.
2823 * src/pa/ffitarget.h: New file.
2825 2004-02-10 Randolph Chung <tausq@debian.org>
2827 * Makefile.am: Add PA support.
2828 * Makefile.in: Regenerate.
2829 * include/Makefile.in: Regenerate.
2830 * configure.ac: Add PA target.
2831 * configure: Regenerate.
2832 * src/pa/ffi.c: New file.
2833 * src/pa/ffi.h.in: Add PA support.
2834 * src/pa/linux.S: New file.
2835 * prep_cif.c: Add PA support.
2837 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2839 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2841 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2842 with ecif->cif->flags.
2843 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2845 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2846 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2847 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2848 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2850 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2852 * configure.in: Rename file to ...
2853 * configure.ac: ... this.
2854 * fficonfig.h.in: Regenerate.
2855 * Makefile.in: Regenerate.
2856 * include/Makefile.in: Regenerate.
2857 * testsuite/Makefile.in: Regenerate.
2859 2004-03-12 Matt Austern <austern@apple.com>
2861 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2862 changes in EH format resulting from addition of linkonce support.
2863 * src/powerpc/darwin_closure.S: Likewise.
2865 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2866 Paolo Bonzini <bonzini@gnu.org>
2868 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2869 Remove VPATH. Remove rules for object files. Remove multilib support.
2870 (AM_CCASFLAGS): Add.
2871 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2872 (AC_PREREQ): Bump version to 2.59.
2873 (AC_INIT): Fill with version info and bug address.
2874 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2875 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2876 De-precious CC so that the right flags are passed down to multilibs.
2877 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2878 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2879 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2880 * configure: Rebuilt.
2881 * aclocal.m4: Likewise.
2882 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2883 * fficonfig.h.in: Likewise.
2885 2004-03-11 Andreas Schwab <schwab@suse.de>
2887 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2888 arguments from fp registers only for the first 8 parameter slots.
2889 Don't convert a float parameter when passed in memory.
2891 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2893 * configure: Regenerate for config/accross.m4 correction.
2895 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2897 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2898 ecif->cif->bytes to bytes.
2899 (ffi_prep_cif_machdep): Add braces around nested if statement.
2901 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2903 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2905 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2906 (ffi_closure_helper_LINUX64): Fix typo.
2907 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2909 * testsuite/libffi.call/float.c: Likewise.
2910 * testsuite/libffi.call/float2.c: Likewise.
2912 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2914 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2915 long double function return and long double arg handling.
2916 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2917 Use "end_pfr" instead of "nf". Correct long double handling.
2919 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2921 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2922 space for long double return value. Adjust stack frame and offsets.
2923 Load f2 long double return.
2925 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2927 * src/types.c: Use 16 byte long double for POWERPC64.
2929 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2931 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2932 when the structure return address is passed in %o0.
2933 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2934 (ffi_v9_layout_struct): Align the field following a nested structure
2935 on a word boundary. Use memmove instead of memcpy.
2936 (ffi_call): Update call to ffi_V9_return_struct.
2937 (ffi_prep_closure): Define 'ctx' only for V8.
2938 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2939 and ffi_closure_sparc_inner_v9.
2940 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2941 Always skip the structure return address. For structures and long
2942 doubles, copy the argument directly.
2943 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2944 if required. Shift the maximum floating-point slot accordingly. For
2945 big structures, copy the argument directly; otherwise, left-justify the
2946 argument and call ffi_v9_layout_struct to lay out the structure on
2948 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2949 (ffi_closure_v8): Pass the structure return address. Update call to
2950 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2951 Skip the 'unimp' insn when returning long doubles and structures.
2952 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2953 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2954 FFI_TYPE_INT handling. Load structures both in integers and
2955 floating-point registers on return.
2956 * README: Update status of the SPARC port.
2958 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2960 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2962 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2964 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2966 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2967 value as of type ffi_arg, not unsigned int.
2969 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2971 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2974 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2976 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2979 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2981 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2984 2003-12-10 Richard Henderson <rth@redhat.com>
2986 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2987 size_t instead of int.
2989 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2991 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2992 * testsuite/libffi.call/many_win32.c (main): Replace variable
2993 int i with unsigned long ul.
2995 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2996 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2997 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2998 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2999 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
3000 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
3001 * testsuite/libffi.call/cls_align_float.c: Likewise.
3002 * testsuite/libffi.call/cls_align_double.c: Likewise.
3003 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
3004 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
3006 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
3009 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
3010 Align arguments to 32 bits.
3012 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
3015 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
3016 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
3017 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
3018 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
3019 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
3020 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
3022 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
3024 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3026 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
3027 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3029 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
3031 * Makefile.in: Rebuilt.
3032 * configure: Likewise.
3033 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
3036 2003-11-21 Andreas Jaeger <aj@suse.de>
3037 Andreas Tobler <a.tobler@schweiz.ch>
3039 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
3040 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
3041 * Makefile.in: Rebuilt.
3042 * aclocal.m4: Likewise.
3043 * configure: Likewise.
3044 * fficonfig.h.in: Likewise.
3045 * testsuite/lib/libffi-dg.exp: Add include dir.
3046 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
3047 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3048 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
3049 for ffi_closure if available.
3050 * testsuite/libffi.call/closure_fn1.c: Likewise.
3051 * testsuite/libffi.call/closure_fn2.c: Likewise.
3052 * testsuite/libffi.call/closure_fn3.c: Likewise.
3053 * testsuite/libffi.call/closure_fn4.c: Likewise.
3054 * testsuite/libffi.call/closure_fn5.c: Likewise.
3055 * testsuite/libffi.call/cls_12byte.c: Likewise.
3056 * testsuite/libffi.call/cls_16byte.c: Likewise.
3057 * testsuite/libffi.call/cls_18byte.c: Likewise.
3058 * testsuite/libffi.call/cls_19byte.c: Likewise.
3059 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3060 * testsuite/libffi.call/cls_20byte.c: Likewise.
3061 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3062 * testsuite/libffi.call/cls_24byte.c: Likewise.
3063 * testsuite/libffi.call/cls_2byte.c: Likewise.
3064 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3065 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3066 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3067 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3068 * testsuite/libffi.call/cls_4byte.c: Likewise.
3069 * testsuite/libffi.call/cls_5byte.c: Likewise.
3070 * testsuite/libffi.call/cls_64byte.c: Likewise.
3071 * testsuite/libffi.call/cls_6byte.c: Likewise.
3072 * testsuite/libffi.call/cls_7byte.c: Likewise.
3073 * testsuite/libffi.call/cls_8byte.c: Likewise.
3074 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3075 * testsuite/libffi.call/cls_9byte2.c: Likewise.
3076 * testsuite/libffi.call/cls_double.c: Likewise.
3077 * testsuite/libffi.call/cls_float.c: Likewise.
3078 * testsuite/libffi.call/cls_schar.c: Likewise.
3079 * testsuite/libffi.call/cls_sint.c: Likewise.
3080 * testsuite/libffi.call/cls_sshort.c: Likewise.
3081 * testsuite/libffi.call/cls_uchar.c: Likewise.
3082 * testsuite/libffi.call/cls_uint.c: Likewise.
3083 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3084 * testsuite/libffi.call/cls_ushort.c: Likewise.
3085 * testsuite/libffi.call/nested_struct.c: Likewise.
3086 * testsuite/libffi.call/nested_struct1.c: Likewise.
3087 * testsuite/libffi.call/nested_struct2.c: Likewise.
3088 * testsuite/libffi.call/nested_struct3.c: Likewise.
3089 * testsuite/libffi.call/problem1.c: Likewise.
3090 * testsuite/libffi.special/unwindtest.cc: Likewise.
3092 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
3094 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
3096 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
3098 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
3099 Add -lgcc_s to additional flags.
3101 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
3103 * configure.in, include/Makefile.am: PR libgcj/11147, install
3104 the ffitarget.h header file in a gcc versioned and target
3106 * configure: Regenerated.
3107 * Makefile.in, include/Makefile.in: Likewise.
3108 * testsuite/Makefile.in: Likewise.
3110 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
3112 * testsuite/libffi.call/closure_fn0.c: Print result and check
3113 with dg-output to make debugging easier.
3114 * testsuite/libffi.call/closure_fn1.c: Likewise.
3115 * testsuite/libffi.call/closure_fn2.c: Likewise.
3116 * testsuite/libffi.call/closure_fn3.c: Likewise.
3117 * testsuite/libffi.call/closure_fn4.c: Likewise.
3118 * testsuite/libffi.call/closure_fn5.c: Likewise.
3119 * testsuite/libffi.call/cls_12byte.c: Likewise.
3120 * testsuite/libffi.call/cls_16byte.c: Likewise.
3121 * testsuite/libffi.call/cls_18byte.c: Likewise.
3122 * testsuite/libffi.call/cls_19byte.c: Likewise.
3123 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3124 * testsuite/libffi.call/cls_20byte.c: Likewise.
3125 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3126 * testsuite/libffi.call/cls_24byte.c: Likewise.
3127 * testsuite/libffi.call/cls_2byte.c: Likewise.
3128 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3129 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3130 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3131 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3132 * testsuite/libffi.call/cls_4byte.c: Likewise.
3133 * testsuite/libffi.call/cls_5byte.c: Likewise.
3134 * testsuite/libffi.call/cls_64byte.c: Likewise.
3135 * testsuite/libffi.call/cls_6byte.c: Likewise.
3136 * testsuite/libffi.call/cls_7byte.c: Likewise.
3137 * testsuite/libffi.call/cls_8byte.c: Likewise.
3138 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3139 * testsuite/libffi.call/cls_9byte2.c: Likewise.
3140 * testsuite/libffi.call/cls_double.c: Likewise.
3141 * testsuite/libffi.call/cls_float.c: Likewise.
3142 * testsuite/libffi.call/cls_schar.c: Likewise.
3143 * testsuite/libffi.call/cls_sint.c: Likewise.
3144 * testsuite/libffi.call/cls_sshort.c: Likewise.
3145 * testsuite/libffi.call/cls_uchar.c: Likewise.
3146 * testsuite/libffi.call/cls_uint.c: Likewise.
3147 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3148 * testsuite/libffi.call/cls_ushort.c: Likewise.
3149 * testsuite/libffi.call/problem1.c: Likewise.
3151 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
3154 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
3156 * testsuite/libffi.call/cls_9byte2.c: New test case.
3157 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3158 * testsuite/libffi.call/cls_64byte.c: Likewise.
3159 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3160 * testsuite/libffi.call/cls_19byte.c: Likewise.
3161 * testsuite/libffi.call/cls_18byte.c: Likewise.
3162 * testsuite/libffi.call/closure_fn4.c: Likewise.
3163 * testsuite/libffi.call/closure_fn5.c: Likewise.
3164 * testsuite/libffi.call/cls_schar.c: Likewise.
3165 * testsuite/libffi.call/cls_sint.c: Likewise.
3166 * testsuite/libffi.call/cls_sshort.c: Likewise.
3167 * testsuite/libffi.call/nested_struct2.c: Likewise.
3168 * testsuite/libffi.call/nested_struct3.c: Likewise.
3170 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
3172 * testsuite/libffi.call/cls_double.c: Do a check on the result.
3173 * testsuite/libffi.call/cls_uchar.c: Likewise.
3174 * testsuite/libffi.call/cls_uint.c: Likewise.
3175 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3176 * testsuite/libffi.call/cls_ushort.c: Likewise.
3177 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
3179 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
3181 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
3184 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
3186 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
3187 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
3189 2003-10-22 David Daney <ddaney@avtrex.com>
3191 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
3192 __attribute__((__mode__(__SI__))) and friends.
3194 2003-10-22 Andreas Schwab <schwab@suse.de>
3196 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
3198 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
3200 * configure.in: AC_LINK_FILES(ffitarget.h).
3201 * configure: Regenerate.
3202 * Makefile.in: Likewise.
3203 * include/Makefile.in: Likewise.
3204 * testsuite/Makefile.in: Likewise.
3205 * fficonfig.h.in: Likewise.
3207 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
3208 Richard Henderson <rth@redhat.com>
3210 Avoid that ffi.h includes fficonfig.h.
3212 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
3213 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
3214 (TARGET_SRC_MIPS_SGI): Removed.
3215 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
3216 (MIPS_SGI): Removed.
3217 (CLEANFILES): Removed.
3218 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
3220 * acconfig.h: Removed.
3221 * configure.in: Compute sizeofs only for double and long double.
3222 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
3223 into AC_DEFINE instead of using acconfig.h. Create
3224 include/ffitarget.h instead of include/fficonfig.h. Rename
3225 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
3226 AC_DEFINE EH_FRAME_FLAGS.
3227 * include/Makefile.am (DISTCLEANFILES): New automake macro.
3228 (hack_DATA): Add ffitarget.h.
3229 * include/ffi.h.in: Remove all system specific definitions.
3230 Declare raw API even if it is not installed, why bother?
3231 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
3232 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
3233 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
3234 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
3235 * include/ffi_common.h (bool): Do not define.
3236 (ffi_assert): Accept failed assertion.
3237 (ffi_type_test): Return void and accept file/line.
3238 (FFI_ASSERT): Pass stringized failed assertion.
3239 (FFI_ASSERT_AT): New macro.
3240 (FFI_ASSERT_VALID_TYPE): New macro.
3241 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
3242 UINT64, SINT64): Define here with gcc's __attribute__ macro
3244 (FLOAT32, ALIGN): Define here instead of in ffi.h
3245 * include/ffi-mips.h: Removed. Its content moved to
3246 src/mips/ffitarget.h after separating assembly and C sections.
3247 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
3248 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
3249 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
3250 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
3251 SIZEOF_ARG -> FFI_SIZEOF_ARG.
3252 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
3253 * src/debug.c (ffi_assert): Accept stringized failed assertion.
3254 (ffi_type_test): Rewritten.
3255 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
3256 FFI_ASSERT_VALID_TYPE.
3257 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
3258 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
3259 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
3260 src/s390/ffitarget.h, src/sh/ffitarget.h,
3261 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
3262 src/x86/ffitarget.h: New files.
3263 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
3264 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
3265 src/powerpc/aix.S, src/powerpc/darwin.S,
3266 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
3267 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
3268 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
3269 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
3270 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
3273 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3275 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
3276 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
3278 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
3280 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
3281 Used when FFI_DEBUG = 1.
3283 2003-10-14 Alan Modra <amodra@bigpond.net.au>
3285 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
3288 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3290 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
3291 FFI_MIPS_O32 for O32 ABI.
3293 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
3295 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
3296 SPARC64. Cleanup whitespaces.
3298 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
3300 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
3301 strongarm, xscale. Cleanup whitespaces.
3302 * testsuite/libffi.call/closure_fn1.c: Likewise.
3303 * testsuite/libffi.call/closure_fn2.c: Likewise.
3304 * testsuite/libffi.call/closure_fn3.c: Likewise.
3305 * testsuite/libffi.call/cls_12byte.c: Likewise.
3306 * testsuite/libffi.call/cls_16byte.c: Likewise.
3307 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3308 * testsuite/libffi.call/cls_20byte.c: Likewise.
3309 * testsuite/libffi.call/cls_24byte.c: Likewise.
3310 * testsuite/libffi.call/cls_2byte.c: Likewise.
3311 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3312 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3313 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3314 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3315 * testsuite/libffi.call/cls_4byte.c: Likewise.
3316 * testsuite/libffi.call/cls_5byte.c: Likewise.
3317 * testsuite/libffi.call/cls_6byte.c: Likewise.
3318 * testsuite/libffi.call/cls_7byte.c: Likewise.
3319 * testsuite/libffi.call/cls_8byte.c: Likewise.
3320 * testsuite/libffi.call/cls_double.c: Likewise.
3321 * testsuite/libffi.call/cls_float.c: Likewise.
3322 * testsuite/libffi.call/cls_uchar.c: Likewise.
3323 * testsuite/libffi.call/cls_uint.c: Likewise.
3324 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3325 * testsuite/libffi.call/cls_ushort.c: Likewise.
3326 * testsuite/libffi.call/nested_struct.c: Likewise.
3327 * testsuite/libffi.call/nested_struct1.c: Likewise.
3328 * testsuite/libffi.call/problem1.c: Likewise.
3329 * testsuite/libffi.special/unwindtest.cc: Likewise.
3330 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
3332 2003-09-18 David Edelsohn <edelsohn@gnu.org>
3334 * src/powerpc/aix.S: Cleanup whitespaces.
3335 * src/powerpc/aix_closure.S: Likewise.
3337 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3339 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
3340 * src/powerpc/darwin_closure.S: Likewise.
3341 * src/powerpc/ffi_darwin.c: Likewise.
3343 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3344 David Edelsohn <edelsohn@gnu.org>
3346 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
3347 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
3349 * src/powerpc/darwin_closure.S: Likewise.
3350 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
3351 according to the Darwin/AIX ABI.
3352 (ffi_prep_cif_machdep): Likewise.
3353 (ffi_closure_helper_DARWIN): Likewise.
3354 Remove the outgoing parameter stack logic. Simplify the evaluation
3355 of the different CASE types.
3356 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
3357 statement in the trampoline code.
3359 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
3361 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
3362 for the register size.
3363 (ffi_closure_helper_SYSV): Handle the structure return value
3365 (ffi_closure_helper_SYSV): Return the appropriate type when
3366 the registers are used for the structure return value.
3367 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
3368 the 64-bit return value. Update copyright years.
3370 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3372 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
3373 srcdir for ffi_mips.h.
3375 2003-09-12 Alan Modra <amodra@bigpond.net.au>
3377 * src/prep_cif.c (initialize_aggregate): Include tail padding in
3379 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
3380 placement of float result.
3381 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
3382 cast of "resp" for big-endian 64 bit machines.
3384 2003-09-11 Alan Modra <amodra@bigpond.net.au>
3386 * src/types.c (double, longdouble): Merge identical SH and ARM
3387 typedefs, and add POWERPC64.
3388 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
3389 struct split over gpr and rest.
3390 (ffi_prep_cif_machdep): Correct intarg_count for structures.
3391 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
3393 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
3395 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
3398 2003-09-09 Alan Modra <amodra@bigpond.net.au>
3400 * configure: Regenerate.
3402 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3404 * Makefile.am: Remove build rules for ffitest.
3405 * Makefile.in: Rebuilt.
3407 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3409 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3410 about implicit declaration of abort().
3412 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3414 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3415 * Makefile.in: Rebuilt.
3416 * configure.in: Add dejagnu test framework.
3417 * configure: Rebuilt.
3419 * testsuite/Makefile.am: New file.
3420 * testsuite/Makefile.in: Built
3421 * testsuite/lib/libffi-dg.exp: New file.
3422 * testsuite/config/default.exp: Likewise.
3423 * testsuite/libffi.call/call.exp: Likewise.
3424 * testsuite/libffi.call/ffitest.h: Likewise.
3425 * testsuite/libffi.call/closure_fn0.c: Likewise.
3426 * testsuite/libffi.call/closure_fn1.c: Likewise.
3427 * testsuite/libffi.call/closure_fn2.c: Likewise.
3428 * testsuite/libffi.call/closure_fn3.c: Likewise.
3429 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3430 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3431 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3432 * testsuite/libffi.call/cls_2byte.c: Likewise.
3433 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3434 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3435 * testsuite/libffi.call/cls_4byte.c: Likewise.
3436 * testsuite/libffi.call/cls_5byte.c: Likewise.
3437 * testsuite/libffi.call/cls_6byte.c: Likewise.
3438 * testsuite/libffi.call/cls_7byte.c: Likewise.
3439 * testsuite/libffi.call/cls_8byte.c: Likewise.
3440 * testsuite/libffi.call/cls_12byte.c: Likewise.
3441 * testsuite/libffi.call/cls_16byte.c: Likewise.
3442 * testsuite/libffi.call/cls_20byte.c: Likewise.
3443 * testsuite/libffi.call/cls_24byte.c: Likewise.
3444 * testsuite/libffi.call/cls_double.c: Likewise.
3445 * testsuite/libffi.call/cls_float.c: Likewise.
3446 * testsuite/libffi.call/cls_uchar.c: Likewise.
3447 * testsuite/libffi.call/cls_uint.c: Likewise.
3448 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3449 * testsuite/libffi.call/cls_ushort.c: Likewise.
3450 * testsuite/libffi.call/float.c: Likewise.
3451 * testsuite/libffi.call/float1.c: Likewise.
3452 * testsuite/libffi.call/float2.c: Likewise.
3453 * testsuite/libffi.call/many.c: Likewise.
3454 * testsuite/libffi.call/many_win32.c: Likewise.
3455 * testsuite/libffi.call/nested_struct.c: Likewise.
3456 * testsuite/libffi.call/nested_struct1.c: Likewise.
3457 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3458 * testsuite/libffi.call/problem1.c: Likewise.
3459 * testsuite/libffi.call/promotion.c: Likewise.
3460 * testsuite/libffi.call/return_ll.c: Likewise.
3461 * testsuite/libffi.call/return_sc.c: Likewise.
3462 * testsuite/libffi.call/return_uc.c: Likewise.
3463 * testsuite/libffi.call/strlen.c: Likewise.
3464 * testsuite/libffi.call/strlen_win32.c: Likewise.
3465 * testsuite/libffi.call/struct1.c: Likewise.
3466 * testsuite/libffi.call/struct2.c: Likewise.
3467 * testsuite/libffi.call/struct3.c: Likewise.
3468 * testsuite/libffi.call/struct4.c: Likewise.
3469 * testsuite/libffi.call/struct5.c: Likewise.
3470 * testsuite/libffi.call/struct6.c: Likewise.
3471 * testsuite/libffi.call/struct7.c: Likewise.
3472 * testsuite/libffi.call/struct8.c: Likewise.
3473 * testsuite/libffi.call/struct9.c: Likewise.
3474 * testsuite/libffi.special/special.exp: New file.
3475 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3476 * testsuite/libffi.special/unwindtest.cc: Likewise.
3479 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3481 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3484 2003-08-02 Alan Modra <amodra@bigpond.net.au>
3486 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3488 (ffi_closure_helper_LINUX64): Likewise.
3489 * src/powerpc/linux64.S: Remove code writing to parm save area.
3490 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3491 address in lr from ffi_closure_helper_LINUX64 call to calculate
3492 table address. Optimize function tail.
3494 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3496 * src/sparc/ffi.c: Handle all floating point registers.
3497 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3499 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3501 * README: Note that libffi is not part of GCC. Update the project
3504 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3506 * src/powerpc/ppc_closure.S: Include ffi.h.
3508 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3510 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3511 Use C style comments.
3513 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3515 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3516 * Makefile.in: Regenerate.
3517 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3518 * configure: Regenerate.
3519 * include/ffi.h.in: Add SHmedia support.
3520 * src/sh64/ffi.c: New file.
3521 * src/sh64/sysv.S: New file.
3523 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3525 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3526 should be read-only.
3527 * configure: Rebuilt.
3528 * fficonfig.h.in: Rebuilt.
3529 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3530 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3531 * src/powerpc/linux64.S: Likewise.
3532 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3533 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3534 if -fpic/-fPIC/-mrelocatable.
3535 * src/powerpc/powerpc_closure.S: Likewise.
3536 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3537 #write in .eh_frame flags.
3538 * src/sparc/v9.S: Likewise.
3539 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3540 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3541 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3543 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3545 Fixes PR bootstrap/10656
3546 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3547 support for .register pseudo-op.
3548 * src/sparc/v8.S: Use it.
3549 * fficonfig.h.in: Rebuilt.
3550 * configure: Rebuilt.
3552 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3554 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3555 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3556 Make it the default on POWERPC64.
3557 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3558 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3559 * configure: Rebuilt.
3560 * src/powerpc/ffi.c (hidden): Define.
3561 (ffi_prep_args_SYSV): Renamed from
3562 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3563 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3564 ASM_NEEDS_REGISTERS64): New.
3565 (ffi_prep_args64): New function.
3566 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3567 (ffi_call): Likewise.
3568 (ffi_prep_closure): Likewise.
3569 (flush_icache): Surround by #ifndef POWERPC64.
3570 (ffi_dblfl): New union type.
3571 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3572 (ffi_closure_helper_LINUX64): New function.
3573 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3575 * src/powerpc/sysv.S: Likewise.
3576 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3577 * src/powerpc/linux64.S: New file.
3578 * src/powerpc/linux64_closure.S: New file.
3579 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3580 src/powerpc/linux64_closure.S.
3581 (TARGET_SRC_POWERPC): Likewise.
3583 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3584 closure_test_fn3): Fix result printing on big-endian 64-bit
3586 (main): Print tst2_arg instead of uninitialized tst2_result.
3588 * src/ffitest.c (main): Hide what closure pointer really points to
3591 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3593 * configure.in (arm-*-netbsdelf*): Add configuration.
3594 (configure): Regenerated.
3596 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3598 * include/Makefile.in: Regenerate.
3600 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3602 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3604 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3605 Receive closure pointer through parameter, read args using
3606 __builtin_dwarf_cfa.
3607 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3609 2003-03-12 Andreas Schwab <schwab@suse.de>
3611 * configure.in: Avoid trailing /. in toolexeclibdir.
3612 * configure: Rebuilt.
3614 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3616 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3618 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3620 * libffi/src/powerpc/darwin_closure.S:
3621 Fix alignement bug, allocate 8 bytes for the result.
3622 * libffi/src/powerpc/aix_closure.S:
3624 * libffi/src/powerpc/ffi_darwin.c:
3625 Update stackframe description for aix/darwin_closure.S.
3627 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3629 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3632 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3633 Andreas Schwab <schwab@suse.de>
3635 * configure.in: Adjust command to source config-ml.in to account
3636 for changes to the libffi_basedir definition.
3637 (libffi_basedir): Remove ${srcdir} from value and include trailing
3640 * configure: Regenerate.
3642 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3644 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3646 2003-01-28 Andrew Haley <aph@redhat.com>
3648 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3649 * src/x86/ffi64.c (ffi_prep_closure): New.
3650 (ffi_closure_UNIX64_inner): New.
3651 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3653 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
3655 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3656 Remove USE_LIBDIR conditional.
3657 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3658 * Makefile.in, configure: Rebuilt.
3660 2003-01027 David Edelsohn <edelsohn@gnu.org>
3662 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3663 * Makefile.in: Regenerate.
3665 2003-01-22 Andrew Haley <aph@redhat.com>
3667 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3670 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3672 * src/powerpc/darwin.S: Add unwind info.
3673 * src/powerpc/darwin_closure.S: Likewise.
3675 2003-01-14 Andrew Haley <aph@redhat.com>
3677 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3678 (ffi_prep_cif_machdep): Likewise.
3679 * src/x86/unix64.S: Add unwind info.
3681 2003-01-14 Andreas Jaeger <aj@suse.de>
3683 * src/ffitest.c (main): Only use ffi_closures if those are
3686 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3688 * libffi/src/ffitest.c
3689 add closure testcases
3691 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3693 * libffi/src/powerpc/ffi.c
3694 fix alignment bug for float (4 byte aligned iso 8 byte)
3696 2003-01-09 Geoffrey Keating <geoffk@apple.com>
3698 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3699 * src/powerpc/darwin.S: Remove RCS version string.
3701 2003-01-03 Jeff Sturm <jsturm@one-point.com>
3703 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3704 * src/ffitest.c (main): Use static storage for closure.
3705 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3706 * src/sparc/v8.S (ffi_closure_v8): New.
3707 * src/sparc/v9.S (ffi_closure_v9): New.
3709 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3711 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3712 enumeration for X86_WIN32.
3713 * src/x86/win32.S: Added ffi_call_STDCALL function
3715 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3716 switch cases for recognising FFI_STDCALL and
3717 calling ffi_call_STDCALL if target is X86_WIN32.
3718 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3719 stdcall versions of the "my_strlen" and "many"
3720 test functions (for X86_WIN32).
3721 Added test cases to test stdcall invocation using
3724 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3726 * src/sh/sysv.S: Add DWARF2 unwind info.
3728 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3730 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3732 2002-11-26 Jim Wilson <wilson@redhat.com>
3734 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3736 2002-11-23 H.J. Lu <hjl@gnu.org>
3738 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3739 Include ../config/accross.m4.
3740 * aclocal.m4; Rebuild.
3741 * configure: Likewise.
3743 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3745 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3747 2002-11-11 DJ Delorie <dj@redhat.com>
3749 * configure.in: Look for common files in the right place.
3751 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3753 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3754 raw data as _Jv_word values, not ffi_raw.
3755 (ffi_java_ptrarray_to_raw): Likewise.
3756 (ffi_java_rvalue_to_raw): New function.
3757 (ffi_java_raw_call): Call it.
3758 (ffi_java_raw_to_rvalue): New function.
3759 (ffi_java_translate_args): Call it.
3760 * src/ffitest.c (closure_test_fn): Interpret return value
3761 as ffi_arg, not int.
3762 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3763 FFI_TYPE_POINTER case.
3764 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3765 values extended to word size.
3767 2002-10-02 Andreas Jaeger <aj@suse.de>
3769 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3771 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
3773 * include/ffi.h.in: Fix i386 win32 compilation.
3775 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3777 * configure.in: Add s390x-*-linux-* target.
3778 * configure: Regenerate.
3779 * include/ffi.h.in: Define S390X for s390x targets.
3780 (FFI_CLOSURES): Define for s390/s390x.
3781 (FFI_TRAMPOLINE_SIZE): Likewise.
3782 (FFI_NATIVE_RAW_API): Likewise.
3783 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3784 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
3785 * src/s390/ffi.c: Major rework of existing code. Add support for
3786 s390x targets. Add closure support.
3787 * src/s390/sysv.S: Likewise.
3789 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
3791 * src/arm/sysv.S: Fix typo.
3793 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
3795 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3796 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3797 (ffi_call_SYSV): Handle soft-float.
3799 2002-09-27 Bo Thorsen <bo@suse.de>
3801 * include/ffi.h.in: Fix multilib x86-64 support.
3803 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3805 * Makefile.am (all-multi): Fix multilib parallel build.
3807 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3809 * configure.in (sh[34]*-*-linux*): Add brackets.
3810 * configure: Regenerate.
3812 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3814 * Makefile.am: Add SH support.
3815 * Makefile.in: Regenerate.
3816 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3817 * configure: Regenerate.
3818 * include/ffi.h.in: Add SH support.
3819 * src/sh/ffi.c: New file.
3820 * src/sh/sysv.S: New file.
3821 * src/types.c: Add SH support.
3823 2002-07-16 Bo Thorsen <bo@suse.de>
3825 * src/x86/ffi64.c: New file that adds x86-64 support.
3826 * src/x86/unix64.S: New file that handles argument setup for
3828 * src/x86/sysv.S: Don't use this on x86-64.
3829 * src/x86/ffi.c: Don't use this on x86-64.
3831 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3833 * src/ffitest.c (struct6): New test that tests a special case in
3835 (struct7): Likewise.
3836 (struct8): Likewise.
3837 (struct9): Likewise.
3838 (closure_test_fn): Silence warning about this when it's not used.
3839 (main): Add the new tests.
3840 (main): Fix a couple of wrong casts and silence some compiler warnings.
3841 * include/ffi.h.in: Add x86-64 ABI definition.
3842 * fficonfig.h.in: Regenerate.
3843 * Makefile.am: Add x86-64 support.
3844 * configure.in: Likewise.
3845 * Makefile.in: Regenerate.
3846 * configure: Likewise.
3848 2002-06-24 Bo Thorsen <bo@suse.de>
3850 * src/types.c: Merge settings for similar architectures.
3851 Add x86-64 sizes and alignments.
3853 2002-06-23 Bo Thorsen <bo@suse.de>
3855 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3856 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3857 * src/mips/ffi.c (ffi_prep_args): Likewise.
3858 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3860 2002-07-18 H.J. Lu (hjl@gnu.org)
3862 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3863 (libffi_la_SOURCES): Support MIPS_LINUX.
3864 (libffi_convenience_la_SOURCES): Likewise.
3865 * Makefile.in: Regenerated.
3867 * configure.in (mips64*-*): Skip.
3868 (mips*-*-linux*): New.
3869 * configure: Regenerated.
3871 * src/mips/ffi.c: Include <sgidefs.h>.
3873 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3875 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3877 2002-05-27 Roger Sayle <roger@eyesopen.com>
3879 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3881 2002-05-27 Bo Thorsen <bo@suse.de>
3883 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3886 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3888 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3889 beginning of function (for older apple cc).
3891 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3893 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3894 script entry, and set LD to it when configuring multilibs.
3895 * configure: Rebuilt.
3897 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3899 * configure.in (sparc64-*-netbsd*): Add target.
3900 (sparc-*-netbsdelf*): Likewise.
3901 * configure: Regenerate.
3903 2002-04-28 David S. Miller <davem@redhat.com>
3905 * configure.in, configure: Fix SPARC test in previous change.
3907 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3909 * Makefile.am: Add Linux for S/390 support.
3910 * Makefile.in: Regenerate.
3911 * configure.in: Add Linux for S/390 support.
3912 * configure: Regenerate.
3913 * include/ffi.h.in: Add Linux for S/390 support.
3914 * src/s390/ffi.c: New file from libffi CVS tree.
3915 * src/s390/sysv.S: New file from libffi CVS tree.
3917 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3919 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3921 * src/sparc/v8.S: Use it.
3922 * src/sparc/v9.S: Likewise.
3923 * fficonfig.h.in: Rebuilt.
3924 * configure: Rebuilt.
3926 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3928 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3930 * src/ia64/unix.S: Add unwind information. Fix comments.
3931 Save sp in a way that's compatible with unwind info.
3932 (ffi_call_unix): Correctly restore sp in all cases.
3933 * src/ia64/ffi.c: Add, fix comments.
3935 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3937 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3939 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3941 * configure.in (alpha*-*-netbsd*): Add target.
3942 * configure: Regenerate.
3944 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3946 * src/sparc/v8.S: Add unwind info.
3947 * src/sparc/v9.S: Likewise.
3949 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3951 * configure.in: Enable i*86-*-netbsdelf*.
3952 * configure: Rebuilt.
3954 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3957 * src/mips/n32.s: Delete
3958 * src/mips/o32.s: Delete
3960 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3962 * configure.in: Enable alpha*-*-freebsd*.
3963 * configure: Rebuilt.
3965 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3967 * Makefile.am: libfficonvenience -> libffi_convenience.
3968 * Makefile.in: Rebuilt.
3970 * Makefile.am: Define ffitest_OBJECTS.
3971 * Makefile.in: Rebuilt.
3973 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3974 David Edelsohn <edelsohn@gnu.org>
3976 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3977 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3978 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3979 * Makefile.in: Regenerate.
3980 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3981 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3982 (flush_icache, flush_range): New functions.
3983 (ffi_closure_helper_DARWIN): New function.
3984 * src/powerpc/aix_closure.S: New file.
3985 * src/powerpc/darwin_closure.S: New file.
3987 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3989 * include/ffi.h.in: Add typedef for ffi_arg.
3990 * src/ffitest.c (main): Declare rint with ffi_arg.
3992 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3994 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3995 number of GPRs for floating-point arguments.
3997 2002-01-31 Anthony Green <green@redhat.com>
3999 * configure: Rebuilt.
4000 * configure.in: Replace CHECK_SIZEOF and endian tests with
4001 cross-compiler friendly macros.
4002 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
4005 2002-01-18 David Edelsohn <edelsohn@gnu.org>
4007 * src/powerpc/darwin.S (_ffi_call_AIX): New.
4008 * src/powerpc/aix.S (ffi_call_DARWIN): New.
4010 2002-01-17 David Edelsohn <edelsohn@gnu.org>
4012 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
4013 (TARGET_SRC_POWERPC_AIX): New.
4014 (POWERPC_AIX): New stanza.
4015 * Makefile.in: Regenerate.
4016 * configure.in: Add AIX case.
4017 * configure: Regenerate.
4018 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
4019 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
4020 size. Fix "long double" support.
4021 (ffi_call): Add FFI_AIX case.
4022 * src/powerpc/aix.S: New.
4024 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
4026 Implement Darwin PowerPC ABI.
4027 * configure.in: Handle powerpc-*-darwin*.
4028 * Makefile.am: Set source files for POWERPC_DARWIN.
4029 * configure: Rebuilt.
4030 * Makefile.in: Rebuilt.
4031 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
4033 * src/powerpc/darwin.S: New file.
4034 * src/powerpc/ffi_darwin.c: New file.
4036 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
4038 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
4040 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4042 * src/x86/sysv.S: Avoid gas-only .balign directive.
4043 Use C style comments.
4045 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4047 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
4048 Fixes PR bootstrap/3563.
4050 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4052 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
4054 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4056 * configure.in: Recognize sparc*-sun-* host.
4057 * configure: Regenerate.
4059 2001-06-06 Andrew Haley <aph@redhat.com>
4061 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
4063 2001-06-03 Andrew Haley <aph@redhat.com>
4065 * src/alpha/osf.S: Add unwind info.
4066 * src/powerpc/sysv.S: Add unwind info.
4067 * src/powerpc/ppc_closure.S: Likewise.
4069 2000-05-31 Jeff Sturm <jsturm@one-point.com>
4071 * configure.in: Fix AC_ARG_ENABLE usage.
4072 * configure: Rebuilt.
4074 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
4076 * configure.in: Remove warning about beta code.
4077 * configure: Rebuilt.
4079 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
4081 * src/ia64/unix.S: Restore stack pointer when returning from
4083 * src/ia64/ffi.c: Fix typo in comment.
4085 2001-04-18 Jim Wilson <wilson@redhat.com>
4087 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
4088 to eliminate RAW DV.
4090 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
4092 * Makefile.am: Make a libtool convenience library.
4093 * Makefile.in: Rebuilt.
4095 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
4097 * configure.in: Use different syntax for subdirectory creation.
4098 * configure: Rebuilt.
4100 2001-03-27 Jon Beniston <jon@beniston.com>
4102 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
4103 * configure: Rebuilt.
4104 * Makefile.am: Added X86_WIN32 target support.
4105 * Makefile.in: Rebuilt.
4107 * include/ffi.h.in: Added X86_WIN32 target support.
4109 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
4110 * src/types.c: Added X86_WIN32 target support.
4112 * src/x86/win32.S: New file. Based on sysv.S, but with EH
4113 stuff removed and made to work with CygWin's gas.
4115 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
4117 * configure.in: Make target subdirectory in build dir.
4118 * Makefile.am: Override suffix based rules to specify correct output
4120 * Makefile.in: Rebuilt.
4121 * configure: Rebuilt.
4123 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
4125 * src/powerpc/ppc_closure.S: New file.
4126 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
4127 involving long long and register pairs.
4128 (ffi_prep_closure): New function.
4129 (flush_icache): Likewise.
4130 (ffi_closure_helper_SYSV): Likewise.
4131 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
4132 (FFI_TRAMPOLINE_SIZE): Likewise.
4133 (FFI_NATIVE_RAW_API): Likewise.
4134 * Makefile.in: Rebuilt.
4135 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
4136 (TARGET_SRC_POWERPC): Likewise.
4138 2001-03-19 Tom Tromey <tromey@redhat.com>
4140 * Makefile.in: Rebuilt.
4141 * Makefile.am (ffitest_LDFLAGS): New macro.
4143 2001-03-02 Nick Clifton <nickc@redhat.com>
4145 * include/ffi.h.in: Remove RCS ident string.
4146 * include/ffi_mips.h: Remove RCS ident string.
4147 * src/debug.c: Remove RCS ident string.
4148 * src/ffitest.c: Remove RCS ident string.
4149 * src/prep_cif.c: Remove RCS ident string.
4150 * src/types.c: Remove RCS ident string.
4151 * src/alpha/ffi.c: Remove RCS ident string.
4152 * src/alpha/osf.S: Remove RCS ident string.
4153 * src/arm/ffi.c: Remove RCS ident string.
4154 * src/arm/sysv.S: Remove RCS ident string.
4155 * src/mips/ffi.c: Remove RCS ident string.
4156 * src/mips/n32.S: Remove RCS ident string.
4157 * src/mips/o32.S: Remove RCS ident string.
4158 * src/sparc/ffi.c: Remove RCS ident string.
4159 * src/sparc/v8.S: Remove RCS ident string.
4160 * src/sparc/v9.S: Remove RCS ident string.
4161 * src/x86/ffi.c: Remove RCS ident string.
4162 * src/x86/sysv.S: Remove RCS ident string.
4164 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
4166 * include/ffi.h.in: Change sourceware.cygnus.com references to
4169 2000-12-09 Richard Henderson <rth@redhat.com>
4171 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
4172 (ffi_closure_osf_inner): Index rather than increment avalue
4173 and arg_types. Give ffi_closure_osf the raw return value type.
4174 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
4177 2000-12-07 Richard Henderson <rth@redhat.com>
4179 * src/raw_api.c (ffi_translate_args): Fix typo.
4180 (ffi_prep_closure): Likewise.
4182 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
4183 FFI_TRAMPOLINE_SIZE.
4184 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
4185 cif->bytes for new ffi_call_osf implementation.
4186 (ffi_prep_args): Absorb into ...
4187 (ffi_call): ... here. Do all stack allocation here and
4188 avoid a callback function.
4189 (ffi_prep_closure, ffi_closure_osf_inner): New.
4190 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
4191 (ffi_closure_osf): New.
4193 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
4195 * config.guess, config.sub, install-sh: Removed.
4196 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
4197 * Makefile.in: Rebuilt.
4199 * acinclude.m4: Include libtool macros from the top level.
4200 * aclocal.m4, configure: Rebuilt.
4202 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
4204 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
4205 * configure: Rebuilt.
4207 2000-05-11 Scott Bambrough <scottb@netwinder.org>
4209 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
4210 memory correctly. Use conditional instructions, not branches where
4213 2000-05-04 Tom Tromey <tromey@cygnus.com>
4215 * configure: Rebuilt.
4216 * configure.in: Match `arm*-*-linux-*'.
4217 From Chris Dornan <cdornan@arm.com>.
4219 2000-04-28 Jakub Jelinek <jakub@redhat.com>
4221 * Makefile.am (SUBDIRS): Define.
4222 (AM_MAKEFLAGS): Likewise.
4223 (Multilib support.): Add section.
4224 * Makefile.in: Rebuilt.
4225 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
4226 New variables. Set for gcc using -print-multi-lib. Export them
4228 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
4229 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
4231 (extra_compiler_flags_value, extra_compiler_flags): Check these
4232 for extra compiler options which need to be passed down in
4235 2000-04-16 Anthony Green <green@redhat.com>
4237 * configure: Rebuilt.
4238 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
4240 2000-04-14 Jakub Jelinek <jakub@redhat.com>
4242 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
4243 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
4244 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
4245 Replace all void * sizeofs with sizeof(int).
4246 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
4247 different than DOUBLE.
4248 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
4249 (ffi_prep_args_v9): New function.
4250 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
4251 (ffi_V9_return_struct): New function.
4252 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
4253 32bit code (not yet cross-arch calls).
4254 * src/sparc/v8.S: Add struct return delay nop.
4256 * src/sparc/v9.S: New file.
4257 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
4258 is used on sparc64 only for structures larger than 32 bytes.
4259 Pass by reference for structures is done for structure arguments
4260 larger than 16 bytes.
4261 * src/ffitest.c (main): Use 64bit rint on sparc64.
4262 Run long long tests on sparc.
4263 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
4265 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
4267 * configure.in (sparc-*-linux*): New supported target.
4268 (sparc64-*-linux*): Likewise.
4269 * configure: Rebuilt.
4270 * Makefile.am: Add v9.S to SPARC files.
4271 * Makefile.in: Likewise.
4272 (LINK): Surround $(CCLD) into double quotes, so that multilib
4273 compiles work correctly.
4275 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
4277 * configure: Rebuilt.
4278 * configure.in: (i*86-*-solaris*): New libffi target. Patch
4279 proposed by Bryce McKinlay.
4281 2000-03-20 Tom Tromey <tromey@cygnus.com>
4283 * Makefile.in: Hand edit for java_raw_api.lo.
4285 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
4287 * config.guess, config.sub: Update from the gcc tree.
4288 Fix for PR libgcj/168.
4290 2000-03-03 Tom Tromey <tromey@cygnus.com>
4292 * Makefile.in: Fixed ia64 by hand.
4294 * configure: Rebuilt.
4295 * configure.in (--enable-multilib): New option.
4296 (libffi_basedir): New subst.
4297 (AC_OUTPUT): Added multilib code.
4299 2000-03-02 Tom Tromey <tromey@cygnus.com>
4301 * Makefile.in: Rebuilt.
4302 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
4305 2000-02-25 Hans Boehm <boehm@acm.org>
4307 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
4309 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
4311 (ffi_prep_raw_closure): Use ffi_translate_args, not
4312 ffi_closure_translate.
4313 * src/java_raw_api.c: New file.
4314 * src/ffitest.c (closure_test_fn): New function.
4315 (main): Define `rint' as long long on IA64. Added new test when
4316 FFI_CLOSURES is defined.
4317 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
4318 (ffi_abi): Recognize IA64.
4319 (ffi_raw): Added `flt' field.
4320 Added "Java raw API" code.
4321 * configure.in: Recognize ia64.
4322 * Makefile.am (TARGET_SRC_IA64): New macro.
4323 (libffi_la_common_SOURCES): Added java_raw_api.c.
4324 (libffi_la_SOURCES): Define in IA64 case.
4326 2000-01-04 Tom Tromey <tromey@cygnus.com>
4328 * Makefile.in: Rebuilt with newer automake.
4330 1999-12-31 Tom Tromey <tromey@cygnus.com>
4332 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
4334 1999-09-01 Tom Tromey <tromey@cygnus.com>
4336 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
4338 * fficonfig.h.in: Rebuilt.
4339 * configure: Rebuilt.
4340 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
4341 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
4342 * acconfig.h: Don't #undef PACKAGE or VERSION.
4344 1999-08-09 Anthony Green <green@cygnus.com>
4346 * include/ffi.h.in: Try to work around messy header problem
4347 with PACKAGE and VERSION.
4349 * configure: Rebuilt.
4350 * configure.in: Change version to 2.00-beta.
4352 * fficonfig.h.in: Rebuilt.
4353 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
4355 * src/x86/ffi.c (ffi_raw_call): Rename.
4357 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
4359 * src/x86/ffi.c (ffi_closure_SYSV): New function.
4360 (ffi_prep_incoming_args_SYSV): Ditto.
4361 (ffi_prep_closure): Ditto.
4362 (ffi_closure_raw_SYSV): Ditto.
4363 (ffi_prep_raw_closure): More ditto.
4364 (ffi_call_raw): Final ditto.
4366 * include/ffi.h.in: Add definitions for closure and raw API.
4368 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4371 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
4373 * src/raw_api.c: New file.
4375 * include/ffi.h.in (ffi_raw): New type.
4376 (UINT_ARG, SINT_ARG): New defines.
4377 (ffi_closure, ffi_raw_closure): New types.
4378 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4380 * configure.in: Add check for endianness and sizeof void*.
4382 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
4383 instead of directly.
4385 * configure: Rebuilt.
4387 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
4389 * configure.in: Add x86 and powerpc BeOS configurations.
4390 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
4392 1999-05-09 Anthony Green <green@cygnus.com>
4394 * configure.in: Add warning about this being beta code.
4395 Remove src/Makefile.am from the picture.
4396 * configure: Rebuilt.
4398 * Makefile.am: Move logic from src/Makefile.am. Add changes
4399 to support libffi as a target library.
4400 * Makefile.in: Rebuilt.
4402 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
4403 Upgraded to new autoconf, automake, libtool.
4407 * LICENSE: Update copyright date.
4409 * src/Makefile.am, src/Makefile.in: Removed.
4411 1998-11-29 Anthony Green <green@cygnus.com>
4413 * include/ChangeLog: Removed.
4414 * src/ChangeLog: Removed.
4415 * src/mips/ChangeLog: Removed.
4416 * src/sparc/ChangeLog: Remboved.
4417 * src/x86/ChangeLog: Removed.
4419 * ChangeLog.v1: Created.