1 2011-04-29 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
3 * src/alpha/osf.S (UA_SI, FDE_ENCODING, FDE_ENCODE, FDE_ARANGE):
5 Use them to handle ELF vs. ECOFF differences.
6 [__osf__] (_GLOBAL__F_ffi_call_osf): Define.
8 2011-02-13 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
10 * configure: Regenerate.
12 2011-02-09 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
15 * testsuite/libffi.call/cls_pointer.c (main): Cast void * to
17 * testsuite/libffi.call/cls_pointer_stack.c (main): Likewise.
19 2011-02-07 Joel Sherrill <joel.sherrill@oarcorp.com>
21 * libffi/src/m68k/ffi.c: Add RTEMS support for cache flushing.
22 Handle case when CPU variant does not have long double support.
23 * libffi/src/m68k/sysv.S: Add support for mc68000, Coldfire,
24 and cores with soft floating point.
26 2011-02-07 Joel Sherrill <joel.sherrill@oarcorp.com>
28 * configure.ac: Add mips*-*-rtems* support.
29 * configure: Regenerate.
30 * src/mips/ffitarget.h: Ensure needed constants are available
31 for targets which do not have sgidefs.h.
33 2011-01-26 Dave Korn <dave.korn.cygwin@gmail.com>
36 * configure.ac (AM_LTLDFLAGS): Add -bindir option for windows DLLs.
37 * configure: Regenerate.
39 2010-12-18 Iain Sandoe <iains@gcc.gnu.org>
43 * src/powerpc/darwin_closure.S: Provide Darwin64 implementation,
45 * src/powerpc/ffitarget.h (POWERPC_DARWIN64): New,
46 (FFI_TRAMPOLINE_SIZE): Update for Darwin64.
47 * src/powerpc/darwin.S: Provide Darwin64 implementation,
49 * src/powerpc/ffi_darwin.c: Likewise.
51 2010-12-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
53 * configure.ac (libffi_cv_as_ascii_pseudo_op): Use double
55 (libffi_cv_as_string_pseudo_op): Likewise.
56 * configure: Regenerate.
58 2010-12-03 Chung-Lin Tang <cltang@codesourcery.com>
60 * src/arm/sysv.S (ffi_closure_SYSV): Add UNWIND to .pad directive.
61 (ffi_closure_VFP): Same.
62 (ffi_call_VFP): Move down to before ffi_closure_VFP. Add '.fpu vfp'
65 2010-12-01 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
67 * testsuite/libffi.call/ffitest.h [__sgi] (PRId64, PRIu64): Define.
70 2010-11-29 Richard Henderson <rth@redhat.com>
71 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
73 * src/x86/sysv.S (FDE_ENCODING, FDE_ENCODE): Define.
74 (.eh_frame): Use FDE_ENCODING.
75 (.LASFDE1, .LASFDE2, LASFDE3): Simplify with FDE_ENCODE.
77 2010-11-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
79 * testsuite/lib/libffi-dg.exp: Rename ...
80 * testsuite/lib/libffi.exp: ... to this.
81 * libffi/testsuite/libffi.call/call.exp: Don't load libffi-dg.exp.
82 * libffi/testsuite/libffi.special/special.exp: Likewise.
84 2010-10-28 Chung-Lin Tang <cltang@codesourcery.com>
86 * src/arm/ffi.c (ffi_prep_args): Add VFP register argument handling
87 code, new parameter, and return value. Update comments.
88 (ffi_prep_cif_machdep): Add case for VFP struct return values. Add
89 call to layout_vfp_args().
90 (ffi_call_SYSV): Update declaration.
91 (ffi_call_VFP): New declaration.
92 (ffi_call): Add VFP struct return conditions. Call ffi_call_VFP()
94 (ffi_closure_VFP): New declaration.
95 (ffi_closure_SYSV_inner): Add new vfp_args parameter, update call to
96 ffi_prep_incoming_args_SYSV().
97 (ffi_prep_incoming_args_SYSV): Update parameters. Add VFP argument
99 (ffi_prep_closure_loc): Pass ffi_closure_VFP to trampoline
100 construction under VFP hard-float.
101 (rec_vfp_type_p): New function.
103 (place_vfp_arg): Same.
104 (layout_vfp_args): Same.
105 * src/arm/ffitarget.h (ffi_abi): Add FFI_VFP. Define FFI_DEFAULT_ABI
106 based on __ARM_PCS_VFP.
107 (FFI_EXTRA_CIF_FIELDS): Define for adding VFP hard-float specific
109 (FFI_TYPE_STRUCT_VFP_FLOAT): Define internally used type code.
110 (FFI_TYPE_STRUCT_VFP_DOUBLE): Same.
111 * src/arm/sysv.S (ffi_call_SYSV): Change call of ffi_prep_args() to
112 direct call. Move function pointer load upwards.
113 (ffi_call_VFP): New function.
114 (ffi_closure_VFP): Same.
116 * testsuite/lib/libffi-dg.exp (check-flags): New function.
117 (dg-skip-if): New function.
118 * testsuite/libffi.call/cls_double_va.c: Skip if target is arm*-*-*
119 and compiler options include -mfloat-abi=hard.
120 * testsuite/libffi.call/cls_longdouble_va.c: Same.
122 2010-10-01 Jakub Jelinek <jakub@redhat.com>
125 * src/x86/ffi64.c (ffi_prep_cif_machdep): Ensure cif->bytes is
127 * testsuite/libffi.call/many2.c: New test.
129 2010-08-20 Mark Wielaard <mjw@redhat.com>
131 * src/closures.c (open_temp_exec_file_mnt): Check if getmntent_r
134 2010-08-09 Andreas Tobler <andreast@fgznet.ch>
136 * configure.ac: Add target powerpc64-*-freebsd*.
137 * configure: Regenerate.
138 * testsuite/libffi.call/cls_align_longdouble_split.c: Pass
139 -mlong-double-128 only to linux targets.
140 * testsuite/libffi.call/cls_align_longdouble_split2.c: Likewise.
141 * testsuite/libffi.call/cls_longdouble.c: Likewise.
142 * testsuite/libffi.call/huge_struct.c: Likewise.
144 2010-07-10 Evan Phoenix <evan@fallingsnow.net>
146 * src/closures.c (selinux_enabled_check): Fix strncmp usage bug.
148 2010-07-07 Dan HorĂ¡k <dan@danny.cz>
150 * include/ffi.h.in: Protect #define with #ifndef.
151 * src/powerpc/ffitarget.h: Ditto.
152 * src/s390/ffitarget.h: Ditto.
153 * src/sparc/ffitarget.h: Ditto.
155 2010-07-07 Neil Roberts <neil@linux.intel.com>
157 * src/x86/sysv.S (ffi_call_SYSV): Align the stack pointer to
160 2010-07-02 Jakub Jelinek <jakub@redhat.com>
162 * Makefile.am (AM_MAKEFLAGS): Pass also mandir to submakes.
163 * Makefile.in: Regenerated.
165 2010-05-19 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
167 * configure.ac (libffi_cv_as_x86_pcrel): Check for illegal in as
169 (libffi_cv_as_ascii_pseudo_op): Check for .ascii.
170 (libffi_cv_as_string_pseudo_op): Check for .string.
171 * configure: Regenerate.
172 * fficonfig.h.in: Regenerate.
173 * src/x86/sysv.S (.eh_frame): Use .ascii, .string or error.
175 2010-05-05 Michael Kohler <michaelkohler@live.com>
177 * src/dlmalloc.c (dlfree): Fix spelling.
178 * src/ia64/ffi.c (ffi_prep_cif_machdep): Ditto.
179 * configure.ac: Ditto.
180 * configure: Rebuilt.
182 2010-04-13 Dan Witte <dwitte@mozilla.com>
184 * msvcc.sh: Build with -W3 instead of -Wall.
185 * src/powerpc/ffi_darwin.c: Remove build warnings.
186 * src/x86/ffi.c: Ditto.
187 * src/x86/ffitarget.h: Ditto.
189 2010-04-12 Dan Witte <dwitte@mozilla.com>
190 Walter Meinl <wuno@lsvw.de>
192 * configure.ac: Add OS/2 support.
193 * configure: Rebuilt.
194 * src/closures.c: Ditto.
195 * src/dlmalloc.c: Ditto.
196 * src/x86/win32.S: Ditto.
198 2010-04-07 Jakub Jelinek <jakub@redhat.com>
200 * testsuite/libffi.call/err_bad_abi.c: Remove unused args variable.
202 2010-04-02 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
204 * Makefile.in: Regenerate.
205 * aclocal.m4: Regenerate.
206 * include/Makefile.in: Regenerate.
207 * man/Makefile.in: Regenerate.
208 * testsuite/Makefile.in: Regenerate.
210 2010-03-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
212 * configure.ac (libffi_cv_as_x86_64_unwind_section_type): New test.
213 * configure: Regenerate.
214 * fficonfig.h.in: Regenerate.
215 * libffi/src/x86/unix64.S (.eh_frame)
216 [HAVE_AS_X86_64_UNWIND_SECTION_TYPE]: Use @unwind section type.
218 2010-03-14 Matthias Klose <doko@ubuntu.com>
220 * src/x86/ffi64.c: Fix typo in comment.
221 * src/x86/ffi.c: Use /* ... */ comment style.
223 2010-02-24 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
225 * doc/libffi.texi (The Closure API): Fix typo.
226 * doc/libffi.info: Remove.
228 2010-02-15 Matthias Klose <doko@ubuntu.com>
230 * src/arm/sysv.S (__ARM_ARCH__): Define for processor
233 2010-01-15 Anthony Green <green@redhat.com>
235 * README: Add notes on building with Microsoft Visual C++.
237 2010-01-15 Daniel Witte <dwitte@mozilla.com>
239 * msvcc.sh: New file.
241 * src/x86/win32.S: Port assembly routines to MSVC and #ifdef.
242 * src/x86/ffi.c: Tweak function declaration and remove excess
244 * include/ffi.h.in: Add __declspec(align(8)) to typedef struct
247 * src/x86/ffi.c: Merge ffi_call_SYSV and ffi_call_STDCALL into new
248 function ffi_call_win32 on X86_WIN32.
249 * src/x86/win32.S (ffi_call_SYSV): Rename to ffi_call_win32.
250 (ffi_call_STDCALL): Remove.
252 * src/prep_cif.c (ffi_prep_cif): Move stack space allocation code
253 to ffi_prep_cif_machdep for x86.
254 * src/x86/ffi.c (ffi_prep_cif_machdep): To here.
256 2010-01-15 Oliver Kiddle <okiddle@yahoo.co.uk>
258 * src/x86/ffitarget.h (ffi_abi): Check for __i386 and __amd64 for
259 Sun Studio compiler compatibility.
261 2010-01-12 Conrad Irwin <conrad.irwin@gmail.com>
263 * doc/libffi.texi: Add closure example.
265 2010-01-07 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
268 * testsuite/libffi.call/ffitest.h [__alpha__ && __osf__] (PRIdLL,
269 PRIuLL, PRId64, PRIu64, PRIuPTR): Define.
270 * testsuite/libffi.call/cls_align_sint64.c: Add -Wno-format on
272 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
273 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
274 * testsuite/libffi.call/return_ll1.c: Likewise.
275 * testsuite/libffi.call/stret_medium2.c: Likewise.
276 * testsuite/libffi.special/ffitestcxx.h (allocate_mmap): Cast
277 MAP_FAILED to char *.
279 2010-01-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
281 * src/mips/n32.S: Use .abicalls and .eh_frame with __GNUC__.
283 2009-12-31 Anthony Green <green@redhat.com>
285 * README: Update for libffi 3.0.9.
287 2009-12-27 Matthias Klose <doko@ubuntu.com>
289 * configure.ac (HAVE_LONG_DOUBLE): Define for mips when
291 * configure: Rebuilt.
293 2009-12-26 Anthony Green <green@redhat.com>
295 * testsuite/libffi.call/cls_longdouble_va.c: Mark as xfail for
297 * testsuite/libffi.call/cls_double_va.c: Ditto.
299 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
301 * testsuite/libffi.call/ffitest.h: Conditionally include stdint.h
303 * testsuite/libffi.special/unwindtest.cc: Ditto.
305 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
307 * configure.ac: Add amd64-*-openbsd*.
308 * configure: Rebuilt.
309 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Link
310 openbsd programs with -lpthread.
312 2009-12-26 Anthony Green <green@redhat.com>
314 * testsuite/libffi.call/cls_double_va.c,
315 testsuite/libffi.call/cls_longdouble.c,
316 testsuite/libffi.call/cls_longdouble_va.c,
317 testsuite/libffi.call/cls_pointer.c,
318 testsuite/libffi.call/cls_pointer_stack.c: Remove xfail for
319 mips*-*-* and arm*-*-*.
320 * testsuite/libffi.call/cls_align_longdouble_split.c,
321 testsuite/libffi.call/cls_align_longdouble_split2.c,
322 testsuite/libffi.call/stret_medium2.c,
323 testsuite/libffi.call/stret_medium.c,
324 testsuite/libffi.call/stret_large.c,
325 testsuite/libffi.call/stret_large2.c: Remove xfail for arm*-*-*.
327 2009-12-31 Kay Tietz <ktietz70@googlemail.com>
329 * testsuite/libffi.call/ffitest.h,
330 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRuLL): Fix
333 2009-12-31 Carlo Bramini <carlo.bramix@libero.it>
335 * configure.ac (AM_LTLDFLAGS): Define for windows hosts.
336 * Makefile.am (libffi_la_LDFLAGS): Add AM_LTLDFLAGS.
337 * configure: Rebuilt.
338 * Makefile.in: Rebuilt.
340 2009-12-31 Anthony Green <green@redhat.com>
343 * testsuite/libffi.call/huge_struct.c: New test case from Blake
346 2009-12-28 David Edelsohn <edelsohn@gnu.org>
348 * src/powerpc/ffi_darwin.c (ffi_prep_args): Copy abi and nargs to
350 (aix_adjust_aggregate_sizes): New function.
351 (ffi_prep_cif_machdep): Call it.
353 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
355 * configure.ac: Define FFI_MMAP_EXEC_WRIT for the given targets.
356 * configure: Regenerate.
357 * fficonfig.h.in: Likewise.
358 * src/closures.c: Remove the FFI_MMAP_EXEC_WRIT definition for
361 2009-12-26 Andreas Schwab <schwab@linux-m68k.org>
363 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Advance intarg_count
364 when a float arguments is passed in memory.
365 (ffi_closure_helper_SYSV): Mark general registers as used up when
366 a 64bit or soft-float long double argument is passed in memory.
368 2009-12-25 Matthias Klose <doko@ubuntu.com>
370 * man/ffi_call.3: Fix #include in examples.
371 * doc/libffi.texi: Add dircategory.
373 2009-12-25 Frank Everdij <f.p.x.everdij@tudelft.nl>
375 * include/ffi.h.in: Placed '__GNUC__' ifdef around
376 '__attribute__((aligned(8)))' in ffi_closure, fixes compile for
378 * include/ffi_common.h: Added '__sgi' define to non
379 '__attribute__((__mode__()))' integer typedefs.
380 * src/mips/ffi.c (ffi_call, ffi_closure_mips_inner_O32,
381 ffi_closure_mips_inner_N32): Added 'defined(_MIPSEB)' to BE check.
382 (ffi_closure_mips_inner_O32, ffi_closure_mips_inner_N32): Added
383 FFI_LONGDOUBLE support and alignment(N32 only).
384 * src/mips/ffitarget.h: Corrected '#include <sgidefs.h>' for IRIX and
385 fixed non '__attribute__((__mode__()))' integer typedefs.
386 * src/mips/n32.S: Put '#ifdef linux' around '.abicalls' and '.eh_frame'
387 since they are Linux/GNU Assembler specific.
389 2009-12-25 Bradley Smith <brad@brad-smith.co.uk>
391 * configure.ac, Makefile.am, src/avr32/ffi.c,
392 src/avr32/ffitarget.h,
393 src/avr32/sysv.S: Add AVR32 port.
394 * configure, Makefile.in: Rebuilt.
396 2009-12-21 Andreas Tobler <a.tobler@schweiz.org>
398 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
399 * configure: Regenerate.
401 2009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
403 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
405 2009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
407 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
410 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
412 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
415 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
417 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
419 2009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
422 * src/closures.c [X86_64 && __sun__ && __svr4__]
423 (FFI_MMAP_EXEC_WRIT): Define.
425 2009-12-08 David Daney <ddaney@caviumnetworks.com>
427 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
428 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
429 * testsuite/libffi.call/stret_large.c: Same.
430 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
431 * testsuite/libffi.call/stret_large2.c: Same.
432 * testsuite/libffi.call/stret_medium2.c: Same.
434 2009-12-07 David Edelsohn <edelsohn@gnu.org>
436 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
439 2009-12-05 David Edelsohn <edelsohn@gnu.org>
441 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
443 * src/powerpc/aix_closure.S: Same.
445 2009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
447 * Makefile.in: Regenerate.
448 * configure: Regenerate.
449 * include/Makefile.in: Regenerate.
450 * man/Makefile.in: Regenerate.
451 * testsuite/Makefile.in: Regenerate.
453 2009-12-04 David Edelsohn <edelsohn@gnu.org>
455 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
458 2009-12-04 Uros Bizjak <ubizjak@gmail.com>
461 * src/x86/ffi64.c (classify_argument): Update from
462 gcc/config/i386/i386.c.
463 (ffi_closure_unix64_inner): Do not use the address of two consecutive
464 SSE registers directly.
465 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
466 for x86_64 linux targets.
468 2009-12-04 David Edelsohn <edelsohn@gnu.org>
470 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
471 pfr for long double split between fpr13 and stack.
473 2009-12-03 David Edelsohn <edelsohn@gnu.org>
475 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
476 fparg_count twice for long double.
478 2009-12-03 David Edelsohn <edelsohn@gnu.org>
481 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
483 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
485 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
486 Remove xfails for x86 linux targets.
488 2009-12-02 David Edelsohn <edelsohn@gnu.org>
490 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
493 2009-12-01 David Edelsohn <edelsohn@gnu.org>
495 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
496 register usage. Call ffi_prep_args directly. Add long double
497 return value support.
498 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
499 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
500 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
501 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
503 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
505 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
506 Allocate result area between params and FPRs.
508 2009-11-30 David Edelsohn <edelsohn@gnu.org>
511 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
513 * src/powerpc/aix.S: Implement AIX64 version.
514 * src/powerpc/aix_closure.S: Implement AIX64 version.
515 (ffi_closure_ASM): Use extsb, lha and displament addresses.
516 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
518 (ffi_prep_cif_machdep): Same.
520 (ffi_closure_helper_DARWIN): Same.
522 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
525 * testsuite/libffi.call/testclosure.c: New test.
527 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
529 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
530 assembly version use of ___chkstk.
532 2009-09-23 Matthias Klose <doko@ubuntu.com>
534 PR libffi/40242, PR libffi/41443
535 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
536 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
537 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
538 Change the conditionals to __SOFTFP__ || __ARM_EABI__
539 for -mfloat-abi=softfp to work.
541 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
543 PR testsuite/32843 (strikes again)
544 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
545 enable proper extension on char and short.
547 2009-09-15 David Daney <ddaney@caviumnetworks.com>
549 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
550 handling for FFI_TYPE_POINTER.
551 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
552 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
553 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
554 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
555 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
556 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
557 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
558 structure and pointer returns.
559 (ffi_closure_N32): Add handling for pointer returns.
560 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
561 calc_n32_return_struct_flags): Handle soft-float.
562 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
563 (ffi_call_N32): Declare proper argument types.
564 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
567 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
569 * configure.ac (AC_PREREQ): Bump to 2.64.
571 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
573 * Makefile.am (install-html, install-pdf): Remove.
574 * Makefile.in: Regenerate.
576 * Makefile.in: Regenerate.
577 * aclocal.m4: Regenerate.
578 * configure: Regenerate.
579 * fficonfig.h.in: Regenerate.
580 * include/Makefile.in: Regenerate.
581 * man/Makefile.in: Regenerate.
582 * testsuite/Makefile.in: Regenerate.
584 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
586 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
588 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
591 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
592 return types for X86_WIN32.
593 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
594 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
595 _ffi_closure_STDCALL): Likewise.
597 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
598 (dlmmap, dlmunmap): Also use these functions on Cygwin.
600 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
605 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
608 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
610 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
611 to "" before adding paths. (This reinstates an assignment that
612 was removed by my 2009-06-30 commit, but changes the initial
613 value from "." to "".)
615 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
618 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
619 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
622 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
624 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
625 to ld_library_path. Use add_path. Add just find_libgcc_s
626 to ld_library_path, not every libgcc multilib directory.
628 2009-06-16 Wim Lewis <wiml@hhhh.org>
630 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
631 supposed to be callee-saved.
632 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
633 return buffer for odd-size structs.
635 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
638 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
639 allow_stack_execute for Darwin.
641 2009-06-16 Andrew Haley <aph@redhat.com>
643 * configure.ac (TARGETDIR): Add missing blank lines.
644 * configure: Regenerate.
646 2009-06-16 Andrew Haley <aph@redhat.com>
648 * testsuite/libffi.call/cls_align_sint64.c,
649 testsuite/libffi.call/cls_align_uint64.c,
650 testsuite/libffi.call/cls_longdouble_va.c,
651 testsuite/libffi.call/cls_ulonglong.c,
652 testsuite/libffi.call/return_ll1.c,
653 testsuite/libffi.call/stret_medium2.c: Fix printf format
655 * testsuite/libffi.call/ffitest.h,
656 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
658 2009-06-15 Andrew Haley <aph@redhat.com>
660 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
661 * testsuite/libffi.call/err_bad_abi.c: Likewise.
663 2009-06-12 Andrew Haley <aph@redhat.com>
665 * Makefile.am: Remove info_TEXINFOS.
667 2009-06-12 Andrew Haley <aph@redhat.com>
669 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
670 testsuite/libffi.call/cls_align_uint64.c,
671 testsuite/libffi.call/cls_ulonglong.c,
672 testsuite/libffi.call/return_ll1.c,
673 testsuite/libffi.call/stret_medium2.c: Fix printf format
675 testsuite/libffi.special/unwindtest.cc: include stdint.h.
677 2009-06-11 Timothy Wall <twall@users.sf.net>
682 include/ffi_common.h,
687 src/x86/win64.S (new),
688 README: Added win64 support (mingw or MSVC)
692 testsuite/Makefile.in,
694 aclocal.m4: Regenerated
695 * ltcf-c.sh: properly escape cygwin/w32 path
696 * man/ffi_call.3: Clarify size requirements for return value.
697 * src/x86/ffi64.c: Fix filename in comment.
698 * src/x86/win32.S: Remove unused extern.
700 * testsuite/libffi.call/closure_fn0.c,
701 testsuite/libffi.call/closure_fn1.c,
702 testsuite/libffi.call/closure_fn2.c,
703 testsuite/libffi.call/closure_fn3.c,
704 testsuite/libffi.call/closure_fn4.c,
705 testsuite/libffi.call/closure_fn5.c,
706 testsuite/libffi.call/closure_fn6.c,
707 testsuite/libffi.call/closure_stdcall.c,
708 testsuite/libffi.call/cls_12byte.c,
709 testsuite/libffi.call/cls_16byte.c,
710 testsuite/libffi.call/cls_18byte.c,
711 testsuite/libffi.call/cls_19byte.c,
712 testsuite/libffi.call/cls_1_1byte.c,
713 testsuite/libffi.call/cls_20byte.c,
714 testsuite/libffi.call/cls_20byte1.c,
715 testsuite/libffi.call/cls_24byte.c,
716 testsuite/libffi.call/cls_2byte.c,
717 testsuite/libffi.call/cls_3_1byte.c,
718 testsuite/libffi.call/cls_3byte1.c,
719 testsuite/libffi.call/cls_3byte2.c,
720 testsuite/libffi.call/cls_4_1byte.c,
721 testsuite/libffi.call/cls_4byte.c,
722 testsuite/libffi.call/cls_5_1_byte.c,
723 testsuite/libffi.call/cls_5byte.c,
724 testsuite/libffi.call/cls_64byte.c,
725 testsuite/libffi.call/cls_6_1_byte.c,
726 testsuite/libffi.call/cls_6byte.c,
727 testsuite/libffi.call/cls_7_1_byte.c,
728 testsuite/libffi.call/cls_7byte.c,
729 testsuite/libffi.call/cls_8byte.c,
730 testsuite/libffi.call/cls_9byte1.c,
731 testsuite/libffi.call/cls_9byte2.c,
732 testsuite/libffi.call/cls_align_double.c,
733 testsuite/libffi.call/cls_align_float.c,
734 testsuite/libffi.call/cls_align_longdouble.c,
735 testsuite/libffi.call/cls_align_longdouble_split.c,
736 testsuite/libffi.call/cls_align_longdouble_split2.c,
737 testsuite/libffi.call/cls_align_pointer.c,
738 testsuite/libffi.call/cls_align_sint16.c,
739 testsuite/libffi.call/cls_align_sint32.c,
740 testsuite/libffi.call/cls_align_sint64.c,
741 testsuite/libffi.call/cls_align_uint16.c,
742 testsuite/libffi.call/cls_align_uint32.c,
743 testsuite/libffi.call/cls_align_uint64.c,
744 testsuite/libffi.call/cls_dbls_struct.c,
745 testsuite/libffi.call/cls_double.c,
746 testsuite/libffi.call/cls_double_va.c,
747 testsuite/libffi.call/cls_float.c,
748 testsuite/libffi.call/cls_longdouble.c,
749 testsuite/libffi.call/cls_longdouble_va.c,
750 testsuite/libffi.call/cls_multi_schar.c,
751 testsuite/libffi.call/cls_multi_sshort.c,
752 testsuite/libffi.call/cls_multi_sshortchar.c,
753 testsuite/libffi.call/cls_multi_uchar.c,
754 testsuite/libffi.call/cls_multi_ushort.c,
755 testsuite/libffi.call/cls_multi_ushortchar.c,
756 testsuite/libffi.call/cls_pointer.c,
757 testsuite/libffi.call/cls_pointer_stack.c,
758 testsuite/libffi.call/cls_schar.c,
759 testsuite/libffi.call/cls_sint.c,
760 testsuite/libffi.call/cls_sshort.c,
761 testsuite/libffi.call/cls_uchar.c,
762 testsuite/libffi.call/cls_uint.c,
763 testsuite/libffi.call/cls_ulonglong.c,
764 testsuite/libffi.call/cls_ushort.c,
765 testsuite/libffi.call/err_bad_abi.c,
766 testsuite/libffi.call/err_bad_typedef.c,
767 testsuite/libffi.call/float2.c,
768 testsuite/libffi.call/huge_struct.c,
769 testsuite/libffi.call/nested_struct.c,
770 testsuite/libffi.call/nested_struct1.c,
771 testsuite/libffi.call/nested_struct10.c,
772 testsuite/libffi.call/nested_struct2.c,
773 testsuite/libffi.call/nested_struct3.c,
774 testsuite/libffi.call/nested_struct4.c,
775 testsuite/libffi.call/nested_struct5.c,
776 testsuite/libffi.call/nested_struct6.c,
777 testsuite/libffi.call/nested_struct7.c,
778 testsuite/libffi.call/nested_struct8.c,
779 testsuite/libffi.call/nested_struct9.c,
780 testsuite/libffi.call/problem1.c,
781 testsuite/libffi.call/return_ldl.c,
782 testsuite/libffi.call/return_ll1.c,
783 testsuite/libffi.call/stret_large.c,
784 testsuite/libffi.call/stret_large2.c,
785 testsuite/libffi.call/stret_medium.c,
786 testsuite/libffi.call/stret_medium2.c,
787 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
788 of checking for MMAP. Use intptr_t instead of long casts.
790 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
792 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
793 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
794 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
796 2009-06-09 Andrew Haley <aph@redhat.com>
798 * src/x86/freebsd.S: Add missing file.
800 2009-06-08 Andrew Haley <aph@redhat.com>
802 Import from libffi 3.0.8:
804 * doc/libffi.texi: New file.
805 * doc/libffi.info: Likewise.
806 * doc/stamp-vti: Likewise.
807 * man/Makefile.am: New file.
808 * man/ffi_call.3: New file.
810 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
812 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
814 * configure.ac: Bump version to 3.0.8.
815 parisc*-*-linux*: Add.
816 i386-*-freebsd* | i386-*-openbsd*: Add.
817 powerpc-*-beos*: Add.
818 AM_CONDITIONAL X86_FREEBSD: Add.
819 AC_CONFIG_FILES: Add man/Makefile.
821 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
823 2009-06-08 Andrew Haley <aph@redhat.com>
825 * README: Import from libffi 3.0.8.
827 2009-06-08 Andrew Haley <aph@redhat.com>
829 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
830 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
831 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
832 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
834 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
835 * testsuite/libffi.call/stret_medium.c: Likewise.
836 * testsuite/libffi.call/stret_large2.c: Likewise.
837 * testsuite/libffi.call/stret_large.c: Likewise.
839 2008-12-26 Timothy Wall <twall@users.sf.net>
841 * testsuite/libffi.call/cls_longdouble.c,
842 testsuite/libffi.call/cls_longdouble_va.c,
843 testsuite/libffi.call/cls_align_longdouble.c,
844 testsuite/libffi.call/cls_align_longdouble_split.c,
845 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
846 failures on x86_64 cygwin/mingw.
848 2008-12-22 Timothy Wall <twall@users.sf.net>
850 * testsuite/libffi.call/closure_fn0.c,
851 testsuite/libffi.call/closure_fn1.c,
852 testsuite/libffi.call/closure_fn2.c,
853 testsuite/libffi.call/closure_fn3.c,
854 testsuite/libffi.call/closure_fn4.c,
855 testsuite/libffi.call/closure_fn5.c,
856 testsuite/libffi.call/closure_fn6.c,
857 testsuite/libffi.call/closure_loc_fn0.c,
858 testsuite/libffi.call/closure_stdcall.c,
859 testsuite/libffi.call/cls_align_pointer.c,
860 testsuite/libffi.call/cls_pointer.c,
861 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
862 pointer to integer (intptr_t).
863 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
865 2008-07-24 Anthony Green <green@redhat.com>
867 * testsuite/libffi.call/cls_dbls_struct.c,
868 testsuite/libffi.call/cls_double_va.c,
869 testsuite/libffi.call/cls_longdouble.c,
870 testsuite/libffi.call/cls_longdouble_va.c,
871 testsuite/libffi.call/cls_pointer.c,
872 testsuite/libffi.call/cls_pointer_stack.c,
873 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
876 2008-03-04 Anthony Green <green@redhat.com>
880 * testsuite/libffi.call/cls_align_longdouble_split2.c
881 testsuite/libffi.call/cls_align_longdouble_split.c
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
889 testsuite/libffi.call/err_bad_typedef.c
890 testsuite/libffi.call/stret_large2.c
891 testsuite/libffi.call/stret_large.c
892 testsuite/libffi.call/stret_medium2.c
893 testsuite/libffi.call/stret_medium.c: New tests from Apple.
895 2009-06-05 Andrew Haley <aph@redhat.com>
897 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
900 2009-06-04 Andrew Haley <aph@redhat.com>
902 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
905 2008-02-26 Anthony Green <green@redhat.com>
906 Thomas Heller <theller@ctypes.org>
908 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
911 2008-02-03 Timothy Wall <twall@users.sf.net>
913 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
914 offset based on code pointer, not data pointer.
916 2008-01-31 Timothy Wall <twall@users.sf.net>
918 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
920 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
922 * src/x86/win32.S: Add assembly for stdcall closure.
923 * src/x86/ffi.c: Initialize stdcall closure trampoline.
925 2009-06-04 Andrew Haley <aph@redhat.com>
927 * include/ffi.h.in: Change void (*)() to void (*)(void).
928 * src/x86/ffi.c: Likewise.
930 2009-06-04 Andrew Haley <aph@redhat.com>
932 * src/powerpc/ppc_closure.S: Insert licence header.
933 * src/powerpc/linux64_closure.S: Likewise.
934 * src/m68k/sysv.S: Likewise.
936 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
937 * src/powerpc/ffi.c: Likewise.
938 * src/powerpc/ffi_darwin.c: Likewise.
939 * src/m32r/ffi.c: Likewise.
940 * src/sh64/ffi.c: Likewise.
941 * src/x86/ffi64.c: Likewise.
942 * src/alpha/ffi.c: Likewise.
943 * src/alpha/osf.S: Likewise.
944 * src/frv/ffi.c: Likewise.
945 * src/s390/ffi.c: Likewise.
946 * src/pa/ffi.c: Likewise.
947 * src/pa/hpux32.S: Likewise.
948 * src/ia64/unix.S: Likewise.
949 * src/ia64/ffi.c: Likewise.
950 * src/sparc/ffi.c: Likewise.
951 * src/mips/ffi.c: Likewise.
952 * src/sh/ffi.c: Likewise.
954 2008-02-15 David Daney <ddaney@avtrex.com>
956 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
957 Define (conditionally), and use it to include cachectl.h.
958 (ffi_prep_closure_loc): Fix cache flushing.
959 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
961 2009-06-04 Andrew Haley <aph@redhat.com>
967 src/powerpc/ffitarget.h,
969 src/sh64/ffitarget.h,
980 src/alpha/ffitarget.h,
983 src/m68k/ffitarget.h,
986 src/s390/ffitarget.h,
988 src/cris/ffitarget.h,
993 src/ia64/ffitarget.h,
996 src/ia64/ia64_flags.h,
1000 src/sparc/ffitarget.h,
1003 src/mips/ffitarget.h,
1010 src/sh/sysv.S: Update license text.
1012 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
1014 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
1015 (.eh_frame): Add FDE for it.
1017 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
1019 * configure.ac: Also check if assembler supports pc-relative
1020 relocs on X86_WIN32 targets.
1021 * configure: Regenerate.
1022 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
1023 (_ffi_call_SYSV): Add missing function type symbol .def and
1024 add EH markup labels.
1025 (_ffi_call_STDCALL): Likewise.
1026 (_ffi_closure_SYSV): Likewise.
1027 (_ffi_closure_raw_SYSV): Likewise.
1028 (.eh_frame): Add hand-crafted EH data.
1030 2009-04-09 Jakub Jelinek <jakub@redhat.com>
1032 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
1033 version 3 of the GNU General Public License and to point readers
1034 at the COPYING3 file and the FSF's license web page.
1035 * testsuite/libffi.call/call.exp: Likewise.
1036 * testsuite/libffi.special/special.exp: Likewise.
1038 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1040 * configure: Regenerate.
1042 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1045 * configure.ac (HAVE_AS_X86_PCREL): New test.
1046 * configure: Regenerate.
1047 * fficonfig.h.in: Regenerate.
1048 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
1049 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
1050 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
1051 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
1052 * src/x86/unix64.S (.Lstore_table): Move to .text section.
1053 (.Lload_table): Likewise.
1054 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
1056 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1058 * configure: Regenerate.
1060 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
1062 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
1063 signed/unsigned int8/16 return values.
1064 * src/sparc/v8.S (ffi_call_v8): Likewise.
1065 (ffi_closure_v8): Likewise.
1067 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
1068 Steve Ellcey <sje@cup.hp.com>
1070 * configure: Regenerate for new libtool.
1071 * Makefile.in: Ditto.
1072 * include/Makefile.in: Ditto.
1073 * aclocal.m4: Ditto.
1075 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
1077 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
1078 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
1079 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
1080 Adjust copyright notice.
1081 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
1082 register or two register to use for FFI_SYSV structs.
1083 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
1084 (ffi_closure_helper_SYSV): Fix the return type for
1085 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
1086 Adjust copyright notice.
1088 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
1090 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
1093 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1095 * configure: Regenerate.
1096 * include/Makefile.in: Regenerate.
1097 * testsuite/Makefile.in: Regenerate.
1099 2008-06-07 Joseph Myers <joseph@codesourcery.com>
1101 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
1102 powerpc-*-beos*): Remove.
1103 * configure: Regenerate.
1105 2008-05-09 Julian Brown <julian@codesourcery.com>
1107 * Makefile.am (LTLDFLAGS): New.
1108 (libffi_la_LDFLAGS): Use above.
1109 * Makefile.in: Regenerate.
1111 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
1114 * aclocal.m4: Regenerate.
1115 * configure: Regenerate.
1117 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
1119 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
1120 * src/sh64/sysv.S: Likewise.
1122 2008-03-26 Daniel Jacobowitz <dan@debian.org>
1124 * src/arm/sysv.S: Fix ARM comment marker.
1126 2008-03-26 Jakub Jelinek <jakub@redhat.com>
1128 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
1129 * src/s390/sysv.S: Likewise.
1130 * src/powerpc/ppc_closure.S: Likewise.
1131 * src/powerpc/sysv.S: Likewise.
1132 * src/x86/unix64.S: Likewise.
1133 * src/x86/sysv.S: Likewise.
1134 * src/sparc/v8.S: Likewise.
1135 * src/sparc/v9.S: Likewise.
1136 * src/m68k/sysv.S: Likewise.
1137 * src/arm/sysv.S: Likewise.
1139 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1141 * aclocal.m4: Regenerate.
1142 * configure: Likewise.
1143 * Makefile.in: Likewise.
1144 * include/Makefile.in: Likewise.
1145 * testsuite/Makefile.in: Likewise.
1147 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
1148 Andreas Tobler <a.tobler@schweiz.org>
1150 * configure.ac: Add amd64-*-freebsd* target.
1151 * configure: Regenerate.
1153 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
1156 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
1159 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
1162 2008-01-24 David Edelsohn <edelsohn@gnu.org>
1164 * configure: Regenerate.
1166 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
1168 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
1170 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
1173 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
1174 signed/unsigned int8/16 for X86_DARWIN.
1175 Updated copyright info.
1176 Handle one and two byte structs with special cif->flags.
1177 * src/x86/ffitarget.h: Add special types for one and two byte structs.
1178 Updated copyright info.
1179 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
1181 Remove code to pop args from the stack after call.
1182 Special-case signed/unsigned for int8/16, one and two byte structs.
1183 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1184 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1186 Updated copyright info.
1188 2007-12-08 David Daney <ddaney@avtrex.com>
1190 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
1191 SUBU, add with ADDU and use smaller code sequences.
1193 2007-12-07 David Daney <ddaney@avtrex.com>
1195 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
1198 2007-12-06 David Daney <ddaney@avtrex.com>
1200 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
1202 (ffi_java_raw): New typedef.
1203 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
1204 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
1206 (ffi_java_raw_closure) : Same.
1207 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
1209 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
1210 FFI_SIZEOF_JAVA_RAW.
1211 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
1212 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
1213 sizeof(ffi_java_raw) for alignment calculations.
1214 (ffi_java_ptrarray_to_raw): Same.
1215 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
1216 if FFI_SIZEOF_JAVA_RAW == 4.
1217 (ffi_java_raw_to_rvalue): Same.
1218 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
1219 (ffi_java_translate_args): Same.
1220 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
1222 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
1224 2007-12-06 David Daney <ddaney@avtrex.com>
1226 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
1229 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
1232 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
1233 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
1234 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
1235 set the NUM_FPR_ARG_REGISTERS according to.
1236 Add support for potential soft-float support under hard-float
1238 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
1239 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
1240 to the FFI_LINUX_SOFT_FLOAT ABI.
1241 (ffi_prep_cif_machdep): Likewise.
1242 (ffi_closure_helper_SYSV): Likewise.
1243 * src/powerpc/ppc_closure.S: Make sure not to store float/double
1244 on archs where __NO_FPRS__ is true.
1245 Add FFI_TYPE_UINT128 support.
1246 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
1247 Adjust copyright notice.
1249 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
1251 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
1252 * include/ffi_common.h: ... here.
1255 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
1257 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
1258 * src/powerpc/linux64.S: Likewise.
1259 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
1260 * testsuite/libffi.call/return_ldl.c: New test.
1262 2007-09-04 <aph@redhat.com>
1264 * src/arm/sysv.S (UNWIND): New.
1265 (Whole file): Conditionally compile unwinder directives.
1266 * src/arm/sysv.S: Add unwinder directives.
1268 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
1269 Only treat r0 as a struct address if we're actually returning a
1271 Only copy the bytes that are actually within a struct.
1272 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
1273 is returned in r0, not passed by address.
1274 (ffi_call): Allocate a word-sized temporary for the case where
1275 a composite is returned in r0.
1276 (ffi_prep_incoming_args_SYSV): Align as necessary.
1278 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
1280 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
1281 directly using the sys_cacheflush syscall.
1283 2007-07-27 Andrew Haley <aph@redhat.com>
1285 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
1287 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
1289 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
1290 * configure.ac: Likewise.
1291 * Makefile.in: Regenerate.
1292 * include/Makefile.in: Likewise.
1293 * testsuite/Makefile.in: Likewise.
1294 * configure: Likewise.
1296 2007-08-24 David Daney <ddaney@avtrex.com>
1298 * testsuite/libffi.call/return_sl.c: New test.
1300 2007-08-10 David Daney <ddaney@avtrex.com>
1302 * testsuite/libffi.call/cls_multi_ushort.c,
1303 testsuite/libffi.call/cls_align_uint16.c,
1304 testsuite/libffi.call/nested_struct1.c,
1305 testsuite/libffi.call/nested_struct3.c,
1306 testsuite/libffi.call/cls_7_1_byte.c,
1307 testsuite/libffi.call/nested_struct5.c,
1308 testsuite/libffi.call/cls_double.c,
1309 testsuite/libffi.call/nested_struct7.c,
1310 testsuite/libffi.call/cls_sint.c,
1311 testsuite/libffi.call/nested_struct9.c,
1312 testsuite/libffi.call/cls_20byte1.c,
1313 testsuite/libffi.call/cls_multi_sshortchar.c,
1314 testsuite/libffi.call/cls_align_sint64.c,
1315 testsuite/libffi.call/cls_3byte2.c,
1316 testsuite/libffi.call/cls_multi_schar.c,
1317 testsuite/libffi.call/cls_multi_uchar.c,
1318 testsuite/libffi.call/cls_19byte.c,
1319 testsuite/libffi.call/cls_9byte1.c,
1320 testsuite/libffi.call/cls_align_float.c,
1321 testsuite/libffi.call/closure_fn1.c,
1322 testsuite/libffi.call/problem1.c,
1323 testsuite/libffi.call/closure_fn3.c,
1324 testsuite/libffi.call/cls_sshort.c,
1325 testsuite/libffi.call/closure_fn5.c,
1326 testsuite/libffi.call/cls_align_double.c,
1327 testsuite/libffi.call/nested_struct.c,
1328 testsuite/libffi.call/cls_2byte.c,
1329 testsuite/libffi.call/nested_struct10.c,
1330 testsuite/libffi.call/cls_4byte.c,
1331 testsuite/libffi.call/cls_6byte.c,
1332 testsuite/libffi.call/cls_8byte.c,
1333 testsuite/libffi.call/cls_multi_sshort.c,
1334 testsuite/libffi.call/cls_align_sint16.c,
1335 testsuite/libffi.call/cls_align_uint32.c,
1336 testsuite/libffi.call/cls_20byte.c,
1337 testsuite/libffi.call/cls_float.c,
1338 testsuite/libffi.call/nested_struct2.c,
1339 testsuite/libffi.call/cls_5_1_byte.c,
1340 testsuite/libffi.call/nested_struct4.c,
1341 testsuite/libffi.call/cls_24byte.c,
1342 testsuite/libffi.call/nested_struct6.c,
1343 testsuite/libffi.call/cls_64byte.c,
1344 testsuite/libffi.call/nested_struct8.c,
1345 testsuite/libffi.call/cls_uint.c,
1346 testsuite/libffi.call/cls_multi_ushortchar.c,
1347 testsuite/libffi.call/cls_schar.c,
1348 testsuite/libffi.call/cls_uchar.c,
1349 testsuite/libffi.call/cls_align_uint64.c,
1350 testsuite/libffi.call/cls_ulonglong.c,
1351 testsuite/libffi.call/cls_align_longdouble.c,
1352 testsuite/libffi.call/cls_1_1byte.c,
1353 testsuite/libffi.call/cls_12byte.c,
1354 testsuite/libffi.call/cls_3_1byte.c,
1355 testsuite/libffi.call/cls_3byte1.c,
1356 testsuite/libffi.call/cls_4_1byte.c,
1357 testsuite/libffi.call/cls_6_1_byte.c,
1358 testsuite/libffi.call/cls_16byte.c,
1359 testsuite/libffi.call/cls_18byte.c,
1360 testsuite/libffi.call/closure_fn0.c,
1361 testsuite/libffi.call/cls_9byte2.c,
1362 testsuite/libffi.call/closure_fn2.c,
1363 testsuite/libffi.call/closure_fn4.c,
1364 testsuite/libffi.call/cls_ushort.c,
1365 testsuite/libffi.call/closure_fn6.c,
1366 testsuite/libffi.call/cls_5byte.c,
1367 testsuite/libffi.call/cls_align_pointer.c,
1368 testsuite/libffi.call/cls_7byte.c,
1369 testsuite/libffi.call/cls_align_sint32.c,
1370 testsuite/libffi.special/unwindtest_ffi_call.cc,
1371 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
1373 2007-08-10 David Daney <ddaney@avtrex.com>
1376 * configure.ac: Don't treat mips64 as a special case.
1377 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
1378 * configure: Regenerate
1379 * Makefile.in: Ditto.
1380 * fficonfig.h.in: Ditto.
1381 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
1382 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
1383 (FFI_DEFAULT_ABI): Set for n64 case.
1384 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
1385 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
1386 (ffi_closure_N32): New function.
1387 (.eh_frame): New section
1388 * src/mips/o32.S: Clean up comments.
1389 (ffi_closure_O32): Pass ffi_closure parameter in $12.
1390 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
1391 _MIPS_SIM == _ABIN32 throughout.
1392 (FFI_MIPS_STOP_HERE): New, use in place of
1394 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1395 to support n32/n64 ABIs.
1396 (calc_n32_struct_flags): Rewrite.
1397 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1398 position of flag bits.
1399 (ffi_prep_cif_machdep): Rewrite n32 portion.
1400 (ffi_call): Enable for n64. Add special handling for small structure
1402 (ffi_prep_closure_loc): Add n32 and n64 support.
1403 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1404 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1406 2007-08-08 David Daney <ddaney@avtrex.com>
1408 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1409 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1411 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1412 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1413 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1414 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1415 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1416 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1417 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1418 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1419 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1420 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1421 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1422 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1423 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1424 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1425 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1426 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1427 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1428 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1429 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1430 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1431 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1432 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1433 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1434 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1435 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1436 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1437 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1438 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1439 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1440 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1441 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1443 2007-08-07 Andrew Haley <aph@redhat.com>
1445 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1448 2007-08-06 Andrew Haley <aph@redhat.com>
1451 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1452 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1455 2007-08-02 David Daney <ddaney@avtrex.com>
1457 * testsuite/libffi.call/return_ul.c (main): Define return type as
1458 ffi_arg. Use proper printf conversion specifier.
1460 2007-07-30 Andrew Haley <aph@redhat.com>
1463 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1464 signed/unsigned int8/16.
1465 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1467 Remove code to pop args from the stack after call.
1468 Special-case signed/unsigned int8/16.
1469 * testsuite/libffi.call/return_sc.c (main): Revert.
1471 2007-07-26 Richard Guenther <rguenther@suse.de>
1474 * testsuite/libffi.call/return_sc.c (main): Verify call
1475 result as signed char, not ffi_arg.
1477 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1479 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1480 * configure: Regenerate.
1482 2007-07-11 David Daney <ddaney@avtrex.com>
1484 * src/mips/ffi.c: Don't include sys/cachectl.h.
1485 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1488 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1490 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1491 from (ffi_prep_closure): ... this.
1492 (FFI_INIT_TRAMPOLINE): Adjust.
1494 2005-12-31 Phil Blundell <pb@reciva.com>
1496 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1497 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1498 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1499 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1500 (FFI_CLOSURES): Enable closure support.
1502 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1504 * testsuite/libffi.call/cls_multi_ushort.c,
1505 testsuite/libffi.call/cls_align_uint16.c,
1506 testsuite/libffi.call/nested_struct1.c,
1507 testsuite/libffi.call/nested_struct3.c,
1508 testsuite/libffi.call/cls_7_1_byte.c,
1509 testsuite/libffi.call/cls_double.c,
1510 testsuite/libffi.call/nested_struct5.c,
1511 testsuite/libffi.call/nested_struct7.c,
1512 testsuite/libffi.call/cls_sint.c,
1513 testsuite/libffi.call/nested_struct9.c,
1514 testsuite/libffi.call/cls_20byte1.c,
1515 testsuite/libffi.call/cls_multi_sshortchar.c,
1516 testsuite/libffi.call/cls_align_sint64.c,
1517 testsuite/libffi.call/cls_3byte2.c,
1518 testsuite/libffi.call/cls_multi_schar.c,
1519 testsuite/libffi.call/cls_multi_uchar.c,
1520 testsuite/libffi.call/cls_19byte.c,
1521 testsuite/libffi.call/cls_9byte1.c,
1522 testsuite/libffi.call/cls_align_float.c,
1523 testsuite/libffi.call/closure_fn1.c,
1524 testsuite/libffi.call/problem1.c,
1525 testsuite/libffi.call/closure_fn3.c,
1526 testsuite/libffi.call/cls_sshort.c,
1527 testsuite/libffi.call/closure_fn5.c,
1528 testsuite/libffi.call/cls_align_double.c,
1529 testsuite/libffi.call/cls_2byte.c,
1530 testsuite/libffi.call/nested_struct.c,
1531 testsuite/libffi.call/nested_struct10.c,
1532 testsuite/libffi.call/cls_4byte.c,
1533 testsuite/libffi.call/cls_6byte.c,
1534 testsuite/libffi.call/cls_8byte.c,
1535 testsuite/libffi.call/cls_multi_sshort.c,
1536 testsuite/libffi.call/cls_align_uint32.c,
1537 testsuite/libffi.call/cls_align_sint16.c,
1538 testsuite/libffi.call/cls_float.c,
1539 testsuite/libffi.call/cls_20byte.c,
1540 testsuite/libffi.call/cls_5_1_byte.c,
1541 testsuite/libffi.call/nested_struct2.c,
1542 testsuite/libffi.call/cls_24byte.c,
1543 testsuite/libffi.call/nested_struct4.c,
1544 testsuite/libffi.call/nested_struct6.c,
1545 testsuite/libffi.call/cls_64byte.c,
1546 testsuite/libffi.call/nested_struct8.c,
1547 testsuite/libffi.call/cls_uint.c,
1548 testsuite/libffi.call/cls_multi_ushortchar.c,
1549 testsuite/libffi.call/cls_schar.c,
1550 testsuite/libffi.call/cls_uchar.c,
1551 testsuite/libffi.call/cls_align_uint64.c,
1552 testsuite/libffi.call/cls_ulonglong.c,
1553 testsuite/libffi.call/cls_align_longdouble.c,
1554 testsuite/libffi.call/cls_1_1byte.c,
1555 testsuite/libffi.call/cls_12byte.c,
1556 testsuite/libffi.call/cls_3_1byte.c,
1557 testsuite/libffi.call/cls_3byte1.c,
1558 testsuite/libffi.call/cls_4_1byte.c,
1559 testsuite/libffi.call/cls_6_1_byte.c,
1560 testsuite/libffi.call/cls_16byte.c,
1561 testsuite/libffi.call/cls_18byte.c,
1562 testsuite/libffi.call/closure_fn0.c,
1563 testsuite/libffi.call/cls_9byte2.c,
1564 testsuite/libffi.call/closure_fn2.c,
1565 testsuite/libffi.call/closure_fn4.c,
1566 testsuite/libffi.call/cls_ushort.c,
1567 testsuite/libffi.call/closure_fn6.c,
1568 testsuite/libffi.call/cls_5byte.c,
1569 testsuite/libffi.call/cls_align_pointer.c,
1570 testsuite/libffi.call/cls_7byte.c,
1571 testsuite/libffi.call/cls_align_sint32.c,
1572 testsuite/libffi.special/unwindtest_ffi_call.cc,
1573 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1575 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1577 * aclocal.m4: Regenerated.
1579 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1581 * configure: Regenerate.
1583 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1585 * Makefile.in: Regenerate.
1586 * configure: Regenerate.
1587 * aclocal.m4: Regenerate.
1588 * include/Makefile.in: Regenerate.
1589 * testsuite/Makefile.in: Regenerate.
1591 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1593 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1594 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1595 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1596 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1597 (FFI_CLOSURES): Enable closure support.
1599 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1601 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1602 * configure: Regenerate.
1603 * fficonfig.h.in: Regenerate.
1604 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1605 CFI_OFFSET,CFI_DEF_CFA): New macros.
1606 (ffi_call_SYSV): Add callframe annotation.
1608 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1610 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1611 numerous test suite failures.
1612 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1614 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1616 * Makefile.am (EXTRA_DIST): Bring up to date.
1617 * Makefile.in: Regenerate.
1618 * src/frv/eabi.S: Remove RCS keyword.
1620 2007-04-06 Richard Henderson <rth@redhat.com>
1622 * configure.ac: Tidy target case.
1623 (HAVE_LONG_DOUBLE): Allow the target to override.
1624 * configure: Regenerate.
1625 * include/ffi.h.in: Don't define ffi_type_foo if
1626 LIBFFI_HIDE_BASIC_TYPES is defined.
1627 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1629 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1630 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1631 (ffi_type_longdouble): Special case for Alpha. Don't define
1632 if long double == double.
1634 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1635 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1636 (ffi_call, ffi_closure_osf_inner): Likewise.
1637 (ffi_closure_osf_inner): Likewise. Mark hidden.
1638 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1639 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1640 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1641 (load_table): Handle 128-bit long double.
1643 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1645 2007-04-06 Tom Tromey <tromey@redhat.com>
1648 * README: Fixed bug in example.
1650 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1652 * src/closures.c: Include sys/statfs.h.
1653 (_GNU_SOURCE): Define on Linux.
1654 (FFI_MMAP_EXEC_SELINUX): Define.
1655 (selinux_enabled): New variable.
1656 (selinux_enabled_check): New function.
1657 (is_selinux_enabled): Define.
1660 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1662 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1663 Use 'volatile float sum' to create sum of floats to avoid false
1664 negative due to excess precision on ix86 targets.
1667 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1669 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1671 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1673 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1675 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1676 (ffi_prep_closure_loc): New.
1677 (ffi_prep_raw_closure_loc): New.
1678 (ffi_prep_java_raw_closure_loc): New.
1679 * src/closures.c: New file.
1680 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1681 Replace sflags with exec_offset.
1682 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1683 sub_segment_exec_offset): New macros.
1684 (get_segment_flags, set_segment_flags, check_segment_merge): New
1686 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1687 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1688 destroy_mspace): Use new macros.
1689 (sys_alloc): Silence warning.
1690 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1691 * Makefile.in: Rebuilt.
1692 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1693 terms of ffi_prep_closure_loc.
1694 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1696 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1698 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1700 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1701 the renamed version.
1702 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1703 (ffi_prep_closure): ... this.
1704 * src/pa/ffi.c: Likewise.
1705 * src/cris/ffi.c: Likewise. Adjust.
1706 * src/frv/ffi.c: Likewise.
1707 * src/ia64/ffi.c: Likewise.
1708 * src/mips/ffi.c: Likewise.
1709 * src/powerpc/ffi_darwin.c: Likewise.
1710 * src/s390/ffi.c: Likewise.
1711 * src/sh/ffi.c: Likewise.
1712 * src/sh64/ffi.c: Likewise.
1713 * src/sparc/ffi.c: Likewise.
1714 * src/x86/ffi64.c: Likewise.
1715 * src/x86/ffi.c: Likewise.
1716 (FFI_INIT_TRAMPOLINE): Adjust.
1717 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1718 (ffi_prep_raw_closure): ... this.
1719 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1720 (ffi_prep_closure): ... this.
1721 (flush_icache): Adjust.
1723 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1725 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1728 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1730 * Makefile.am: Add dummy install-pdf target.
1731 * Makefile.in: Regenerate
1733 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1735 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1736 ffi_closure_helper_SYSV): Add long double handling.
1738 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1740 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1741 immediately after bctrl instruction.
1743 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1745 * Makefile.am (all-recursive, install-recursive,
1746 mostlyclean-recursive, clean-recursive, distclean-recursive,
1747 maintainer-clean-recursive): Add missing targets.
1748 * Makefile.in: Rebuilt.
1750 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1752 * configure.ac: Add TARGET for x86_64-*-darwin*.
1753 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1755 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1756 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1757 * configure: Regenerate.
1758 * Makefile.in: Regenerate.
1759 * include/Makefile.in: Regenerate.
1760 * testsuite/Makefile.in: Regenerate.
1761 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1764 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1766 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1769 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1771 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1772 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1774 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1775 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1777 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1778 Sandro Tolaini <tolaini@libero.it>
1780 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1782 * configure: Regenerated.
1783 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1784 (EXTRA_DIST): Add src/x86/darwin.S.
1785 * Makefile.in: Regenerated.
1786 * include/Makefile.in: Regenerated.
1787 * testsuite/Makefile.in: Regenerated.
1789 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1790 X86_WIN32, and additionally align stack to 16 bytes.
1791 * src/x86/darwin.S: New, based on sysv.S.
1792 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1794 2006-09-12 David Daney <ddaney@avtrex.com>
1797 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1798 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1799 * aclocal.m4: Regenerated for automake 1.9.6.
1800 * Makefile.in: Regenerated.
1801 * include/Makefile.in: Regenerated.
1802 * testsuite/Makefile.in: Regenerated.
1804 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1806 * include/ffi_common.h (struct): Revert accidental commit.
1808 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1810 * include/ffi_common.h: Remove lint directives.
1811 * include/ffi.h.in: Likewise.
1813 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1815 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1816 for 32-bit architectures.
1817 * testsuite/libffi.call/return_ul.c: New test case.
1819 2006-07-19 David Daney <ddaney@avtrex.com>
1821 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1822 xfail remains for mips64.
1824 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1826 * Makefile.am: Add install-html target. Add install-html to .PHONY
1827 * Makefile.in: Regenerate.
1828 * aclocal.m4: Regenerate.
1829 * include/Makefile.in: Regenerate.
1830 * testsuite/Makefile.in: Regenerate.
1832 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1834 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1837 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1839 * README: Remove notice about 'Crazy Comments'.
1840 * src/debug.c: Remove lint directives. Cleanup white spaces.
1841 * src/java_raw_api.c: Likewise.
1842 * src/prep_cif.c: Likewise.
1843 * src/raw_api.c: Likewise.
1844 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1846 * src/arm/ffi.c: Remove lint directives.
1847 * src/m32r/ffi.c: Likewise.
1848 * src/pa/ffi.c: Likewise.
1849 * src/powerpc/ffi.c: Likewise.
1850 * src/powerpc/ffi_darwin.c: Likewise.
1851 * src/sh/ffi.c: Likewise.
1852 * src/sh64/ffi.c: Likewise.
1853 * src/x86/ffi.c: Likewise.
1854 * testsuite/libffi.call/float2.c: Likewise.
1855 * testsuite/libffi.call/promotion.c: Likewise.
1856 * testsuite/libffi.call/struct1.c: Likewise.
1858 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1860 * src/pa/hpux32.S: Correct unwind offset calculation for
1862 * src/pa/linux.S: Likewise.
1864 2006-04-12 James E Wilson <wilson@specifix.com>
1867 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1868 (hfa_type_load): Call stf_spill.
1869 (hfa_type_store): Call ldf_fill.
1870 (ffi_call): Adjust calls to above routines. Add local temps for
1873 2006-04-10 Matthias Klose <doko@debian.org>
1875 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1876 directory names containing underscores.
1878 2006-04-07 James E Wilson <wilson@specifix.com>
1880 * testsuite/libffi.call/float4.c: New testcase.
1882 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1883 Andreas Tobler <a.tobler@schweiz.ch>
1885 * Makefile.am: Add PA_HPUX port.
1886 * Makefile.in: Regenerate.
1887 * include/Makefile.in: Likewise.
1888 * testsuite/Makefile.in: Likewise.
1889 * configure.ac: Add PA_HPUX rules.
1890 * configure: Regenerate.
1891 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1892 Add PA_HPUX and PA64_HPUX.
1893 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1894 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1895 (FFI_TYPE_SMALL_STRUCT2): Define.
1896 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1897 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1898 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1899 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1900 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1901 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1902 * src/pa/ffi.c (ROUND_DOWN): Delete.
1903 (fldw, fstw, fldd, fstd): Use '__asm__'.
1904 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1905 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1906 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1907 Simplify incrementing of stack slot variable. Change type of local
1908 'n' to unsigned int.
1909 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1911 (ffi_prep_cif_machdep): Likewise.
1912 (ffi_call): Likewise.
1913 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1914 return type to ffi_status. Simplify incrementing of stack slot
1915 variable. Only copy floating point argument registers when PA_LINUX
1916 is true. Reformat debug statement.
1917 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1918 FFI_TYPE_SMALL_STRUCT8.
1919 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1921 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1922 Add nops to cache flush. Add trampoline for PA_HPUX.
1923 * src/pa/hpux32.S: New file.
1924 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1925 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1926 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1927 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1928 argument type checks so that common argument types appear first.
1929 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1930 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1932 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1934 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1935 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1936 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1937 FFI_TYPE_LONGDOUBLE.
1938 (ffi_prep_args64): Assert using IBM extended double.
1939 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1940 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1941 (ffi_call): Handle FFI_LINUX.
1942 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1943 gpr3 return pointer as for struct return. Handle FFI_LINUX
1944 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1946 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1947 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1948 Don't use r6 as pointer to results, instead use sp offset. Don't
1949 make a special call to load lr with case table address, instead
1950 use offset from previous call.
1951 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1952 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1955 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1957 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1958 passed with FP registers correctly.
1959 (ffi_closure_helper_SYSV): Likewise.
1960 * src/sh64/sysv.S: Likewise.
1962 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1964 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1965 args and userdata unused.
1966 (closure_test_fn1): Mark cif and userdata unused.
1967 (main): Remove unused res.
1969 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1971 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1972 -O2, -O3, -Os and the warning flags -W -Wall.
1973 * testsuite/libffi.special/special.exp: Likewise.
1974 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1975 unused parameter unused for gcc or else do nothing.
1976 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1977 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1978 and userdata unused.
1979 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1980 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1981 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1982 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1983 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1984 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1985 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1986 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1987 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1988 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1989 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1990 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1991 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1992 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1993 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1994 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1995 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1996 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1997 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1998 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1999 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
2000 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
2001 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
2002 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
2004 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
2006 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
2008 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
2009 void* to avoid compiler warning.
2011 (cls_struct_align_gn): Mark cif and userdata unused.
2012 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
2014 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
2016 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
2018 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
2020 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
2022 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
2023 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
2024 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
2026 (main): Cast res_call to silence gcc.
2027 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
2029 (main): Cast res_call to silence gcc.
2030 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
2032 (main): Cast res_call to silence gcc.
2033 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
2035 (main): Cast res_call to silence gcc.
2036 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
2038 (main): Cast res_call to silence gcc.
2039 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
2041 (main): Cast res_call to silence gcc.
2042 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
2044 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
2045 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
2047 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
2048 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
2050 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
2051 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
2053 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
2054 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
2056 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
2057 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
2058 and userdata unused.
2059 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
2061 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
2062 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
2063 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
2064 Cleanup white spaces.
2065 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
2066 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
2067 cif and userdata unused.
2068 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
2070 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
2071 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
2072 formatters to silence gcc.
2073 (B_gn): Mark cif and userdata unused.
2074 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
2076 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
2077 (B_gn): Mark cif and userdata unused.
2078 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
2080 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
2081 (B_gn): Mark cif and userdata unused.
2082 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
2084 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
2085 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
2086 * testsuite/libffi.call/problem1.c (stub): Likewise.
2087 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
2089 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
2090 in the last commit for this test case in the test case itself.
2091 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
2093 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
2094 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
2095 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
2096 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
2097 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
2098 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
2100 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
2102 * src/sh/sysv.S: Fix register numbers in the FDE for
2105 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
2107 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
2108 declaration to avoid a false negative on ix86. See PR323.
2110 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
2112 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
2113 and cast integer to void * if needed. Update the pointer to
2114 the FP register saved area correctly.
2116 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
2118 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
2120 * testsuite/libffi.call/nested_struct4.c: Likewise.
2122 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
2124 * testsuite/libffi.call/return_dbl.c: New test case.
2125 * testsuite/libffi.call/return_dbl1.c: Likewise.
2126 * testsuite/libffi.call/return_dbl2.c: Likewise.
2127 * testsuite/libffi.call/return_fl.c: Likewise.
2128 * testsuite/libffi.call/return_fl1.c: Likewise.
2129 * testsuite/libffi.call/return_fl2.c: Likewise.
2130 * testsuite/libffi.call/return_fl3.c: Likewise.
2131 * testsuite/libffi.call/closure_fn6.c: Likewise.
2133 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
2135 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
2136 here to be used by other test cases too.
2138 * testsuite/libffi.call/nested_struct10.c: New test case.
2139 * testsuite/libffi.call/nested_struct9.c: Likewise.
2140 * testsuite/libffi.call/nested_struct8.c: Likewise.
2141 * testsuite/libffi.call/nested_struct7.c: Likewise.
2142 * testsuite/libffi.call/nested_struct6.c: Likewise.
2143 * testsuite/libffi.call/nested_struct5.c: Likewise.
2144 * testsuite/libffi.call/nested_struct4.c: Likewise.
2146 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2148 * configure.ac: Enable libffi for sparc64-*-freebsd*.
2149 * configure: Rebuilt.
2151 2006-01-18 Jakub Jelinek <jakub@redhat.com>
2153 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
2154 instead do the shifting inline.
2155 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
2156 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
2157 and 8 byte structs, for the remaining struct sizes don't call
2158 __lshrdi3, instead do the shifting inline.
2160 2005-12-07 Thiemo Seufer <ths@networkno.de>
2162 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
2163 missing parentheses.
2164 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
2165 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
2166 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
2167 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
2168 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
2170 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
2172 (ffi_prep_closure): Improve trampoline instruction scheduling.
2173 (ffi_closure_mips_inner_O32): Fix endianness bugs.
2175 2005-12-03 Alan Modra <amodra@bigpond.net.au>
2177 * src/powerpc/ffi.c: Formatting.
2178 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
2179 (ffi_prep_args64): Likewise.
2181 2005-09-30 Geoffrey Keating <geoffk@apple.com>
2183 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
2184 darwin, use -shared-libgcc not -lgcc_s, and explain why.
2186 2005-09-26 Tom Tromey <tromey@redhat.com>
2188 * testsuite/libffi.call/float1.c (value_type): New typedef.
2189 (CANARY): New define.
2190 (main): Check for result buffer overflow.
2191 * src/powerpc/linux64.S: Handle linux64 long double returns.
2192 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
2193 (ffi_prep_cif_machdep): Handle linux64 long double returns.
2195 2005-08-25 Alan Modra <amodra@bigpond.net.au>
2198 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
2200 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
2202 2005-08-11 Jakub Jelinek <jakub@redhat.com>
2204 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
2205 (AH_BOTTOM): Add FFI_HIDDEN definition.
2206 * configure: Rebuilt.
2207 * fficonfig.h.in: Rebuilt.
2208 * src/powerpc/ffi.c (hidden): Remove.
2209 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
2210 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
2211 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
2212 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
2213 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
2214 add FFI_HIDDEN to its prototype.
2215 (ffi_closure_SYSV_inner): New.
2216 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2217 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2219 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
2222 * configure: Rebuilt.
2223 * configure.ac: Handle i*86-*-gnu*.
2225 2005-08-09 Jakub Jelinek <jakub@redhat.com>
2227 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
2228 DW_CFA_offset_extended_sf rather than
2229 DW_CFA_GNU_negative_offset_extended.
2230 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
2232 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
2234 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
2236 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
2237 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
2238 partially on register.
2239 (ffi_closure_helper_SYSV): Likewise.
2240 (ffi_prep_cif_machdep): Don't set too many cif->flags.
2242 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
2244 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
2246 * src/sh64/ffi.c (simple_type): Remove.
2247 (return_type): Handle small structures correctly.
2248 (ffi_prep_args): Likewise.
2249 (ffi_call): Likewise.
2250 (ffi_closure_helper_SYSV): Likewise.
2251 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
2252 Emit position independent code if PIC and remove wrong datalabel
2253 prefixes from EH data.
2255 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
2257 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
2258 * Makefile.in: Regenerate.
2259 * include/Makefile.in: Likewise.
2260 * testsuite/Makefile.in: Likewise.
2261 * configure.ac: Add POWERPC_FREEBSD rules.
2262 * configure: Regenerate.
2263 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
2264 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
2265 * src/powerpc/ffi.c: Add flags to handle small structure returns
2267 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
2269 (ffi_closure_helper_SYSV): Likewise.
2270 * src/powerpc/ppc_closure.S: Add return types for small structures.
2271 * src/powerpc/sysv.S: Add bits to handle small structures for
2274 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
2276 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
2277 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
2278 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
2280 2005-07-05 Randolph Chung <tausq@debian.org>
2282 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
2283 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
2284 structures. Kill compilation warnings.
2285 (ffi_closure_inner_LINUX): Print return values as hex in debug
2286 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
2287 Properly handle 5-7 byte structure returns.
2288 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
2289 (FFI_TYPE_SMALL_STRUCT2): Remove.
2290 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
2291 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
2292 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
2293 (checksmst1, checksmst2): Remove.
2294 (checksmst3): Optimize handling of 3-byte struct returns.
2295 (checksmst567): Properly handle 5-7 byte struct returns.
2297 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2300 * src/mips/n32.S: Enforce PIC code.
2301 * src/mips/o32.S: Likewise.
2303 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2305 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
2306 * configure: Regenerate.
2308 2005-06-01 Alan Modra <amodra@bigpond.net.au>
2310 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
2311 to call ffi_closure_helper_SYSV. Append @local instead.
2312 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
2314 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
2316 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
2317 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
2318 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
2319 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
2320 include/Makefile.in, testsuite/Makefile.in: Regenerate.
2322 2005-05-09 Mike Stump <mrs@apple.com>
2324 * configure: Regenerate.
2326 2005-05-08 Richard Henderson <rth@redhat.com>
2329 * src/alpha/osf.S: Update unwind into to match code.
2331 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
2332 Richard Henderson <rth@redhat.com>
2334 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
2336 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
2337 (ffi_prep_closure): Set carry bit if sse-used flag set.
2338 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
2339 Only load sse registers if ssecount non-zero.
2340 (ffi_closure_unix64): Only save sse registers if carry set on entry.
2342 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
2344 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
2345 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
2346 * configure: Regenerate.
2348 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
2350 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
2351 have Tcl8.3-compatible intermediate variable.
2353 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
2354 Hans-Peter Nilsson <hp@axis.com>
2356 * Makefile.am: Add CRIS support.
2357 * configure.ac: Likewise.
2358 * Makefile.in, configure, testsuite/Makefile.in,
2359 include/Makefile.in: Regenerate.
2360 * src/cris: New directory.
2361 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
2362 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
2364 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
2365 \r?\n in output tests.
2367 2005-04-12 Mike Stump <mrs@apple.com>
2369 * configure: Regenerate.
2371 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
2373 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
2375 2005-03-30 Steve Ellcey <sje@cup.hp.com>
2377 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
2379 * src/ia64/unix.S (ffi_closure_unix): Extend gp
2380 to 64 bits in ILP32 mode.
2381 Load 64 bits even for short data.
2383 2005-03-23 Mike Stump <mrs@apple.com>
2385 * src/powerpc/darwin.S: Update for -m64 multilib.
2386 * src/powerpc/darwin_closure.S: Likewise.
2388 2005-03-21 Zack Weinberg <zack@codesourcery.com>
2390 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
2391 Do not set tool_include_dir.
2392 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
2394 * include/Makefile.am: Set gcc_version and toollibffidir.
2395 * include/Makefile.in: Regenerate.
2397 2005-02-22 Andrew Haley <aph@redhat.com>
2399 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2400 odd-numbered register pairs for 64-bit integer types.
2402 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2405 * testsuite/libffi.call/return_ll1.c: New test case.
2407 2005-02-11 Janis Johnson <janis187@us.ibm.com>
2409 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2410 * testsuite/libffi.call/float.c: Ditto.
2411 * testsuite/libffi.call/float2.c: Ditto.
2412 * testsuite/libffi.call/float3.c: Ditto.
2414 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2416 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2418 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2420 * testsuite/libffi.special/special.exp (cxx_options): Add
2423 2004-12-31 Richard Henderson <rth@redhat.com>
2425 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2426 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2427 offset parameters with a type parameter; deduce size and structure
2428 alignment. Update all users.
2430 2004-12-31 Richard Henderson <rth@redhat.com>
2432 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2433 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
2434 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
2435 into ffi_prep_closure.
2436 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2439 2004-12-27 Richard Henderson <rth@redhat.com>
2441 * src/x86/unix64.S: Fix typo in unwind info.
2443 2004-12-25 Richard Henderson <rth@redhat.com>
2445 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2446 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2447 (merge_classes): Check for it.
2449 (classify_argument): Pass byte_offset by value; perform all updates
2451 (examine_argument): Add classes argument; handle
2452 X86_64_COMPLEX_X87_CLASS.
2453 (ffi_prep_args): Merge into ...
2454 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2455 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2456 (ffi_fill_return_value): Remove.
2457 (ffi_prep_closure): Remove dead assert.
2458 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2459 Rewrite to use struct register_args instead of va_list. Create
2460 flags for handling structure returns.
2461 * src/x86/unix64.S: Remove dead strings.
2462 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2463 stack frame with ffi_call. Handle structure returns properly.
2464 (float2sse, floatfloat2sse, double2sse): Remove.
2465 (sse2float, sse2double, sse2floatfloat): Remove.
2466 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2467 to handle structure returns properly.
2469 2004-12-08 David Edelsohn <edelsohn@gnu.org>
2471 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2473 * Makefile.in: Regenerated.
2475 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2477 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2478 * configure, aclocal.m4, Makefile.in: Regenerate.
2479 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2481 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2483 * configure: Regenerate for libtool change.
2485 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2487 * configure: Regenerate for libtool reversion.
2489 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2491 * configure: Regenerate for libtool change.
2493 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2495 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2497 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2499 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2500 of jal. Use an absolute encoding for the frame information.
2502 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2504 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2505 * acinclude.m4: Delete logic for sincludes.
2506 * aclocal.m4, Makefile.in, configure: Regenerate.
2507 * include/Makefile: Likewise.
2508 * testsuite/Makefile: Likewise.
2510 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2512 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2513 on a 8-byte boundary.
2514 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2516 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2518 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2519 long long values. Round stack allocation to a multiple of 8 bytes
2520 for ATPCS compatibility.
2521 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2522 names. Handle returning long long types. Add Thumb and interworking
2523 support. Improve soft-float code.
2525 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2527 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2528 (libffi_exit): New function.
2529 (libffi_init): Build the testglue wrapper if needed.
2531 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2534 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2536 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2538 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2540 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2542 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2543 * testsuite/libffi.call/float3.c: New test case.
2545 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2547 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2548 the function returning a structure pointed with R2.
2549 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2550 the structure return value if T bit set. Emit position
2551 independent code and EH data if PIC.
2553 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2555 * Makefile.am: Add m32r support.
2556 * configure.ac: Likewise.
2557 * Makefile.in: Regenerate.
2558 * confiugre: Regenerate.
2559 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2560 (uint64, sint64, double, longdouble)
2561 * src/m32r: New directory.
2562 * src/m32r/ffi.c: New file.
2563 * src/m32r/sysv.S: Likewise.
2564 * src/m32r/ffitarget.h: Likewise.
2566 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2568 * testsuite/libffi.call/negint.c: New test case.
2570 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2573 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2574 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2575 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2578 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2580 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2581 * testsuite/libffi.call/promotion.c: Likewise.
2582 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2584 * testsuite/libffi.call/return_sc.c: Likewise.
2585 * testsuite/libffi.call/return_uc.c: Likewise.
2587 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2589 * src/powerpc/darwin.S: Fix comments and identation.
2590 * src/powerpc/darwin_closure.S: Likewise.
2592 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2594 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2595 (ffi_prep_args): Handle longdouble arguments.
2596 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2598 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2599 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2601 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2602 * src/types.c: Defined longdouble size and alignment for darwin.
2604 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2606 * src/powerpc/aix.S: Remove whitespaces.
2607 * src/powerpc/aix_closure.S: Likewise.
2608 * src/powerpc/asm.h: Likewise.
2609 * src/powerpc/ffi.c: Likewise.
2610 * src/powerpc/ffitarget.h: Likewise.
2611 * src/powerpc/linux64.S: Likewise.
2612 * src/powerpc/linux64_closure.S: Likewise.
2613 * src/powerpc/ppc_closure.S: Likewise.
2614 * src/powerpc/sysv.S: Likewise.
2616 2004-08-30 Anthony Green <green@redhat.com>
2618 * Makefile.am: Add frv support.
2619 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2620 * configure.ac: Read configure.host.
2621 * configure.in: Read configure.host.
2622 * configure.host: New file. frv-elf needs libgloss.
2623 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2624 alignment. This is needed to frv and shouldn't harm the others.
2625 * include/ffi_common.h (ALIGN_DOWN): New macro.
2626 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2628 2004-08-24 David Daney <daney@avtrex.com>
2630 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2631 * testsuite/libffi.call/closure_fn1.c: Likewise.
2632 * testsuite/libffi.call/closure_fn2.c Likewise.
2633 * testsuite/libffi.call/closure_fn3.c: Likewise.
2634 * testsuite/libffi.call/closure_fn4.c: Likewise.
2635 * testsuite/libffi.call/closure_fn5.c: Likewise.
2636 * testsuite/libffi.call/cls_18byte.c: Likewise.
2637 * testsuite/libffi.call/cls_19byte.c: Likewise.
2638 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2639 * testsuite/libffi.call/cls_20byte.c: Likewise.
2640 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2641 * testsuite/libffi.call/cls_24byte.c: Likewise.
2642 * testsuite/libffi.call/cls_2byte.c: Likewise.
2643 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2644 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2645 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2646 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2647 * testsuite/libffi.call/cls_4byte.c: Likewise.
2648 * testsuite/libffi.call/cls_64byte.c: Likewise.
2649 * testsuite/libffi.call/cls_6byte.c: Likewise.
2650 * testsuite/libffi.call/cls_7byte.c: Likewise.
2651 * testsuite/libffi.call/cls_8byte.c: Likewise.
2652 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2653 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2654 * testsuite/libffi.call/cls_align_double.c: Likewise.
2655 * testsuite/libffi.call/cls_align_float.c: Likewise.
2656 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2657 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2658 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2659 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2660 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2661 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2662 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2663 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2664 * testsuite/libffi.call/cls_double.c: Likewise.
2665 * testsuite/libffi.call/cls_float.c: Likewise.
2666 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2667 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2668 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2669 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2670 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2671 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2672 * testsuite/libffi.call/cls_schar.c: Likewise.
2673 * testsuite/libffi.call/cls_sint.c: Likewise.
2674 * testsuite/libffi.call/cls_sshort.c: Likewise.
2675 * testsuite/libffi.call/cls_uchar.c: Likewise.
2676 * testsuite/libffi.call/cls_uint.c: Likewise.
2677 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2678 * testsuite/libffi.call/cls_ushort.c: Likewise.
2679 * testsuite/libffi.call/nested_struct.c: Likewise.
2680 * testsuite/libffi.call/nested_struct1.c: Likewise.
2681 * testsuite/libffi.call/nested_struct2.c: Likewise.
2682 * testsuite/libffi.call/nested_struct3.c: Likewise.
2683 * testsuite/libffi.call/problem1.c: Likewise.
2684 * testsuite/libffi.special/unwindtest.cc: Likewise.
2685 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2687 * testsuite/libffi.call/cls_16byte.c: Likewise.
2688 * testsuite/libffi.call/cls_5byte.c: Likewise.
2690 2004-08-23 David Daney <daney@avtrex.com>
2693 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2694 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2695 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2696 parameters and return types.
2697 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2698 (ffi_prep_closure): Ditto.
2699 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2700 alignment calculations.
2701 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2702 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2704 2004-08-14 Casey Marshall <csm@gnu.org>
2706 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2707 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2708 integer (O32 ABI only).
2709 (ffi_prep_closure): new function.
2710 (ffi_closure_mips_inner_O32): new function.
2711 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2712 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2713 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2714 64 bit integers correctly.
2715 (ffi_closure_O32): new function.
2716 Added DWARF-2 unwind info for both functions.
2718 2004-08-10 Andrew Haley <aph@redhat.com>
2720 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2722 2004-08-01 Robert Millan <robertmh@gnu.org>
2724 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2725 * configure: Regenerate.
2727 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2729 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2730 and mmap() explicitly instead of relying on preset autoconf cache
2732 * aclocal.m4: Regenerate.
2733 * configure: Regenerate.
2735 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2737 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2738 (ffi_check_float_struct): Remove unused prototype.
2740 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2742 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2743 character on Darwin, use '\n\t' instead.
2745 2004-06-26 Matthias Klose <doko@debian.org>
2747 * libtool-version: Fix typo in revision/age.
2749 2004-06-17 Matthias Klose <doko@debian.org>
2751 * libtool-version: New.
2752 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2753 * Makefile.in: Regenerate.
2755 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2757 * Makefile.am: Remove useless multilib rules.
2758 * Makefile.in: Regenerate.
2759 * aclocal.m4: Regenerate with automake 1.8.5.
2760 * configure.ac: Remove useless multilib configury.
2761 * configure: Regenerate.
2763 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2765 * .cvsignore: New file.
2767 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2769 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2771 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2774 2004-06-06 Sean McNeil <sean@mcneil.com>
2776 * configure.ac: Add x86_64-*-freebsd* support.
2777 * configure: Regenerate.
2779 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2782 * configure.ac: Test for existence of mmap and sys/mman.h before
2783 checking blacklist. Fix suggested by Jim Wilson.
2784 * configure: Regenerate.
2786 2004-04-26 Matt Austern <austern@apple.com>
2788 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2790 * src/powerpc/darwin_closure.S: Likewise.
2792 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2794 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2795 error. Reported by Thomas Heller <theller@python.net>.
2796 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2797 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2799 2004-03-20 Matthias Klose <doko@debian.org>
2801 * src/pa/linux.S: Fix typo.
2803 2004-03-19 Matthias Klose <doko@debian.org>
2805 * Makefile.am: Update.
2806 * Makefile.in: Regenerate.
2807 * src/pa/ffi.h.in: Remove.
2808 * src/pa/ffitarget.h: New file.
2810 2004-02-10 Randolph Chung <tausq@debian.org>
2812 * Makefile.am: Add PA support.
2813 * Makefile.in: Regenerate.
2814 * include/Makefile.in: Regenerate.
2815 * configure.ac: Add PA target.
2816 * configure: Regenerate.
2817 * src/pa/ffi.c: New file.
2818 * src/pa/ffi.h.in: Add PA support.
2819 * src/pa/linux.S: New file.
2820 * prep_cif.c: Add PA support.
2822 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2824 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2826 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2827 with ecif->cif->flags.
2828 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2830 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2831 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2832 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2833 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2835 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2837 * configure.in: Rename file to ...
2838 * configure.ac: ... this.
2839 * fficonfig.h.in: Regenerate.
2840 * Makefile.in: Regenerate.
2841 * include/Makefile.in: Regenerate.
2842 * testsuite/Makefile.in: Regenerate.
2844 2004-03-12 Matt Austern <austern@apple.com>
2846 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2847 changes in EH format resulting from addition of linkonce support.
2848 * src/powerpc/darwin_closure.S: Likewise.
2850 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2851 Paolo Bonzini <bonzini@gnu.org>
2853 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2854 Remove VPATH. Remove rules for object files. Remove multilib support.
2855 (AM_CCASFLAGS): Add.
2856 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2857 (AC_PREREQ): Bump version to 2.59.
2858 (AC_INIT): Fill with version info and bug address.
2859 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2860 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2861 De-precious CC so that the right flags are passed down to multilibs.
2862 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2863 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2864 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2865 * configure: Rebuilt.
2866 * aclocal.m4: Likewise.
2867 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2868 * fficonfig.h.in: Likewise.
2870 2004-03-11 Andreas Schwab <schwab@suse.de>
2872 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2873 arguments from fp registers only for the first 8 parameter slots.
2874 Don't convert a float parameter when passed in memory.
2876 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2878 * configure: Regenerate for config/accross.m4 correction.
2880 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2882 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2883 ecif->cif->bytes to bytes.
2884 (ffi_prep_cif_machdep): Add braces around nested if statement.
2886 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2888 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2890 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2891 (ffi_closure_helper_LINUX64): Fix typo.
2892 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2894 * testsuite/libffi.call/float.c: Likewise.
2895 * testsuite/libffi.call/float2.c: Likewise.
2897 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2899 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2900 long double function return and long double arg handling.
2901 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2902 Use "end_pfr" instead of "nf". Correct long double handling.
2904 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2906 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2907 space for long double return value. Adjust stack frame and offsets.
2908 Load f2 long double return.
2910 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2912 * src/types.c: Use 16 byte long double for POWERPC64.
2914 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2916 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2917 when the structure return address is passed in %o0.
2918 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2919 (ffi_v9_layout_struct): Align the field following a nested structure
2920 on a word boundary. Use memmove instead of memcpy.
2921 (ffi_call): Update call to ffi_V9_return_struct.
2922 (ffi_prep_closure): Define 'ctx' only for V8.
2923 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2924 and ffi_closure_sparc_inner_v9.
2925 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2926 Always skip the structure return address. For structures and long
2927 doubles, copy the argument directly.
2928 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2929 if required. Shift the maximum floating-point slot accordingly. For
2930 big structures, copy the argument directly; otherwise, left-justify the
2931 argument and call ffi_v9_layout_struct to lay out the structure on
2933 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2934 (ffi_closure_v8): Pass the structure return address. Update call to
2935 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2936 Skip the 'unimp' insn when returning long doubles and structures.
2937 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2938 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2939 FFI_TYPE_INT handling. Load structures both in integers and
2940 floating-point registers on return.
2941 * README: Update status of the SPARC port.
2943 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2945 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2947 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2949 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2951 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2952 value as of type ffi_arg, not unsigned int.
2954 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2956 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2959 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2961 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2964 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2966 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2969 2003-12-10 Richard Henderson <rth@redhat.com>
2971 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2972 size_t instead of int.
2974 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2976 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2977 * testsuite/libffi.call/many_win32.c (main): Replace variable
2978 int i with unsigned long ul.
2980 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2981 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2982 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2983 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2984 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2985 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2986 * testsuite/libffi.call/cls_align_float.c: Likewise.
2987 * testsuite/libffi.call/cls_align_double.c: Likewise.
2988 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2989 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2991 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2994 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2995 Align arguments to 32 bits.
2997 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
3000 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
3001 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
3002 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
3003 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
3004 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
3005 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
3007 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
3009 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3011 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
3012 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3014 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
3016 * Makefile.in: Rebuilt.
3017 * configure: Likewise.
3018 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
3021 2003-11-21 Andreas Jaeger <aj@suse.de>
3022 Andreas Tobler <a.tobler@schweiz.ch>
3024 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
3025 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
3026 * Makefile.in: Rebuilt.
3027 * aclocal.m4: Likewise.
3028 * configure: Likewise.
3029 * fficonfig.h.in: Likewise.
3030 * testsuite/lib/libffi-dg.exp: Add include dir.
3031 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
3032 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3033 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
3034 for ffi_closure if available.
3035 * testsuite/libffi.call/closure_fn1.c: Likewise.
3036 * testsuite/libffi.call/closure_fn2.c: Likewise.
3037 * testsuite/libffi.call/closure_fn3.c: Likewise.
3038 * testsuite/libffi.call/closure_fn4.c: Likewise.
3039 * testsuite/libffi.call/closure_fn5.c: Likewise.
3040 * testsuite/libffi.call/cls_12byte.c: Likewise.
3041 * testsuite/libffi.call/cls_16byte.c: Likewise.
3042 * testsuite/libffi.call/cls_18byte.c: Likewise.
3043 * testsuite/libffi.call/cls_19byte.c: Likewise.
3044 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3045 * testsuite/libffi.call/cls_20byte.c: Likewise.
3046 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3047 * testsuite/libffi.call/cls_24byte.c: Likewise.
3048 * testsuite/libffi.call/cls_2byte.c: Likewise.
3049 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3050 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3051 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3052 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3053 * testsuite/libffi.call/cls_4byte.c: Likewise.
3054 * testsuite/libffi.call/cls_5byte.c: Likewise.
3055 * testsuite/libffi.call/cls_64byte.c: Likewise.
3056 * testsuite/libffi.call/cls_6byte.c: Likewise.
3057 * testsuite/libffi.call/cls_7byte.c: Likewise.
3058 * testsuite/libffi.call/cls_8byte.c: Likewise.
3059 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3060 * testsuite/libffi.call/cls_9byte2.c: Likewise.
3061 * testsuite/libffi.call/cls_double.c: Likewise.
3062 * testsuite/libffi.call/cls_float.c: Likewise.
3063 * testsuite/libffi.call/cls_schar.c: Likewise.
3064 * testsuite/libffi.call/cls_sint.c: Likewise.
3065 * testsuite/libffi.call/cls_sshort.c: Likewise.
3066 * testsuite/libffi.call/cls_uchar.c: Likewise.
3067 * testsuite/libffi.call/cls_uint.c: Likewise.
3068 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3069 * testsuite/libffi.call/cls_ushort.c: Likewise.
3070 * testsuite/libffi.call/nested_struct.c: Likewise.
3071 * testsuite/libffi.call/nested_struct1.c: Likewise.
3072 * testsuite/libffi.call/nested_struct2.c: Likewise.
3073 * testsuite/libffi.call/nested_struct3.c: Likewise.
3074 * testsuite/libffi.call/problem1.c: Likewise.
3075 * testsuite/libffi.special/unwindtest.cc: Likewise.
3077 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
3079 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
3081 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
3083 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
3084 Add -lgcc_s to additional flags.
3086 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
3088 * configure.in, include/Makefile.am: PR libgcj/11147, install
3089 the ffitarget.h header file in a gcc versioned and target
3091 * configure: Regenerated.
3092 * Makefile.in, include/Makefile.in: Likewise.
3093 * testsuite/Makefile.in: Likewise.
3095 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
3097 * testsuite/libffi.call/closure_fn0.c: Print result and check
3098 with dg-output to make debugging easier.
3099 * testsuite/libffi.call/closure_fn1.c: Likewise.
3100 * testsuite/libffi.call/closure_fn2.c: Likewise.
3101 * testsuite/libffi.call/closure_fn3.c: Likewise.
3102 * testsuite/libffi.call/closure_fn4.c: Likewise.
3103 * testsuite/libffi.call/closure_fn5.c: Likewise.
3104 * testsuite/libffi.call/cls_12byte.c: Likewise.
3105 * testsuite/libffi.call/cls_16byte.c: Likewise.
3106 * testsuite/libffi.call/cls_18byte.c: Likewise.
3107 * testsuite/libffi.call/cls_19byte.c: Likewise.
3108 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3109 * testsuite/libffi.call/cls_20byte.c: Likewise.
3110 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3111 * testsuite/libffi.call/cls_24byte.c: Likewise.
3112 * testsuite/libffi.call/cls_2byte.c: Likewise.
3113 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3114 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3115 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3116 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3117 * testsuite/libffi.call/cls_4byte.c: Likewise.
3118 * testsuite/libffi.call/cls_5byte.c: Likewise.
3119 * testsuite/libffi.call/cls_64byte.c: Likewise.
3120 * testsuite/libffi.call/cls_6byte.c: Likewise.
3121 * testsuite/libffi.call/cls_7byte.c: Likewise.
3122 * testsuite/libffi.call/cls_8byte.c: Likewise.
3123 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3124 * testsuite/libffi.call/cls_9byte2.c: Likewise.
3125 * testsuite/libffi.call/cls_double.c: Likewise.
3126 * testsuite/libffi.call/cls_float.c: Likewise.
3127 * testsuite/libffi.call/cls_schar.c: Likewise.
3128 * testsuite/libffi.call/cls_sint.c: Likewise.
3129 * testsuite/libffi.call/cls_sshort.c: Likewise.
3130 * testsuite/libffi.call/cls_uchar.c: Likewise.
3131 * testsuite/libffi.call/cls_uint.c: Likewise.
3132 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3133 * testsuite/libffi.call/cls_ushort.c: Likewise.
3134 * testsuite/libffi.call/problem1.c: Likewise.
3136 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
3139 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
3141 * testsuite/libffi.call/cls_9byte2.c: New test case.
3142 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3143 * testsuite/libffi.call/cls_64byte.c: Likewise.
3144 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3145 * testsuite/libffi.call/cls_19byte.c: Likewise.
3146 * testsuite/libffi.call/cls_18byte.c: Likewise.
3147 * testsuite/libffi.call/closure_fn4.c: Likewise.
3148 * testsuite/libffi.call/closure_fn5.c: Likewise.
3149 * testsuite/libffi.call/cls_schar.c: Likewise.
3150 * testsuite/libffi.call/cls_sint.c: Likewise.
3151 * testsuite/libffi.call/cls_sshort.c: Likewise.
3152 * testsuite/libffi.call/nested_struct2.c: Likewise.
3153 * testsuite/libffi.call/nested_struct3.c: Likewise.
3155 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
3157 * testsuite/libffi.call/cls_double.c: Do a check on the result.
3158 * testsuite/libffi.call/cls_uchar.c: Likewise.
3159 * testsuite/libffi.call/cls_uint.c: Likewise.
3160 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3161 * testsuite/libffi.call/cls_ushort.c: Likewise.
3162 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
3164 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
3166 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
3169 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
3171 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
3172 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
3174 2003-10-22 David Daney <ddaney@avtrex.com>
3176 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
3177 __attribute__((__mode__(__SI__))) and friends.
3179 2003-10-22 Andreas Schwab <schwab@suse.de>
3181 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
3183 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
3185 * configure.in: AC_LINK_FILES(ffitarget.h).
3186 * configure: Regenerate.
3187 * Makefile.in: Likewise.
3188 * include/Makefile.in: Likewise.
3189 * testsuite/Makefile.in: Likewise.
3190 * fficonfig.h.in: Likewise.
3192 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
3193 Richard Henderson <rth@redhat.com>
3195 Avoid that ffi.h includes fficonfig.h.
3197 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
3198 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
3199 (TARGET_SRC_MIPS_SGI): Removed.
3200 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
3201 (MIPS_SGI): Removed.
3202 (CLEANFILES): Removed.
3203 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
3205 * acconfig.h: Removed.
3206 * configure.in: Compute sizeofs only for double and long double.
3207 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
3208 into AC_DEFINE instead of using acconfig.h. Create
3209 include/ffitarget.h instead of include/fficonfig.h. Rename
3210 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
3211 AC_DEFINE EH_FRAME_FLAGS.
3212 * include/Makefile.am (DISTCLEANFILES): New automake macro.
3213 (hack_DATA): Add ffitarget.h.
3214 * include/ffi.h.in: Remove all system specific definitions.
3215 Declare raw API even if it is not installed, why bother?
3216 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
3217 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
3218 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
3219 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
3220 * include/ffi_common.h (bool): Do not define.
3221 (ffi_assert): Accept failed assertion.
3222 (ffi_type_test): Return void and accept file/line.
3223 (FFI_ASSERT): Pass stringized failed assertion.
3224 (FFI_ASSERT_AT): New macro.
3225 (FFI_ASSERT_VALID_TYPE): New macro.
3226 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
3227 UINT64, SINT64): Define here with gcc's __attribute__ macro
3229 (FLOAT32, ALIGN): Define here instead of in ffi.h
3230 * include/ffi-mips.h: Removed. Its content moved to
3231 src/mips/ffitarget.h after separating assembly and C sections.
3232 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
3233 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
3234 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
3235 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
3236 SIZEOF_ARG -> FFI_SIZEOF_ARG.
3237 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
3238 * src/debug.c (ffi_assert): Accept stringized failed assertion.
3239 (ffi_type_test): Rewritten.
3240 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
3241 FFI_ASSERT_VALID_TYPE.
3242 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
3243 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
3244 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
3245 src/s390/ffitarget.h, src/sh/ffitarget.h,
3246 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
3247 src/x86/ffitarget.h: New files.
3248 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
3249 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
3250 src/powerpc/aix.S, src/powerpc/darwin.S,
3251 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
3252 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
3253 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
3254 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
3255 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
3258 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3260 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
3261 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
3263 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
3265 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
3266 Used when FFI_DEBUG = 1.
3268 2003-10-14 Alan Modra <amodra@bigpond.net.au>
3270 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
3273 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3275 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
3276 FFI_MIPS_O32 for O32 ABI.
3278 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
3280 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
3281 SPARC64. Cleanup whitespaces.
3283 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
3285 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
3286 strongarm, xscale. Cleanup whitespaces.
3287 * testsuite/libffi.call/closure_fn1.c: Likewise.
3288 * testsuite/libffi.call/closure_fn2.c: Likewise.
3289 * testsuite/libffi.call/closure_fn3.c: Likewise.
3290 * testsuite/libffi.call/cls_12byte.c: Likewise.
3291 * testsuite/libffi.call/cls_16byte.c: Likewise.
3292 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3293 * testsuite/libffi.call/cls_20byte.c: Likewise.
3294 * testsuite/libffi.call/cls_24byte.c: Likewise.
3295 * testsuite/libffi.call/cls_2byte.c: Likewise.
3296 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3297 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3298 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3299 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3300 * testsuite/libffi.call/cls_4byte.c: Likewise.
3301 * testsuite/libffi.call/cls_5byte.c: Likewise.
3302 * testsuite/libffi.call/cls_6byte.c: Likewise.
3303 * testsuite/libffi.call/cls_7byte.c: Likewise.
3304 * testsuite/libffi.call/cls_8byte.c: Likewise.
3305 * testsuite/libffi.call/cls_double.c: Likewise.
3306 * testsuite/libffi.call/cls_float.c: Likewise.
3307 * testsuite/libffi.call/cls_uchar.c: Likewise.
3308 * testsuite/libffi.call/cls_uint.c: Likewise.
3309 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3310 * testsuite/libffi.call/cls_ushort.c: Likewise.
3311 * testsuite/libffi.call/nested_struct.c: Likewise.
3312 * testsuite/libffi.call/nested_struct1.c: Likewise.
3313 * testsuite/libffi.call/problem1.c: Likewise.
3314 * testsuite/libffi.special/unwindtest.cc: Likewise.
3315 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
3317 2003-09-18 David Edelsohn <edelsohn@gnu.org>
3319 * src/powerpc/aix.S: Cleanup whitespaces.
3320 * src/powerpc/aix_closure.S: Likewise.
3322 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3324 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
3325 * src/powerpc/darwin_closure.S: Likewise.
3326 * src/powerpc/ffi_darwin.c: Likewise.
3328 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3329 David Edelsohn <edelsohn@gnu.org>
3331 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
3332 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
3334 * src/powerpc/darwin_closure.S: Likewise.
3335 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
3336 according to the Darwin/AIX ABI.
3337 (ffi_prep_cif_machdep): Likewise.
3338 (ffi_closure_helper_DARWIN): Likewise.
3339 Remove the outgoing parameter stack logic. Simplify the evaluation
3340 of the different CASE types.
3341 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
3342 statement in the trampoline code.
3344 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
3346 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
3347 for the register size.
3348 (ffi_closure_helper_SYSV): Handle the structure return value
3350 (ffi_closure_helper_SYSV): Return the appropriate type when
3351 the registers are used for the structure return value.
3352 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
3353 the 64-bit return value. Update copyright years.
3355 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3357 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
3358 srcdir for ffi_mips.h.
3360 2003-09-12 Alan Modra <amodra@bigpond.net.au>
3362 * src/prep_cif.c (initialize_aggregate): Include tail padding in
3364 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
3365 placement of float result.
3366 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
3367 cast of "resp" for big-endian 64 bit machines.
3369 2003-09-11 Alan Modra <amodra@bigpond.net.au>
3371 * src/types.c (double, longdouble): Merge identical SH and ARM
3372 typedefs, and add POWERPC64.
3373 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
3374 struct split over gpr and rest.
3375 (ffi_prep_cif_machdep): Correct intarg_count for structures.
3376 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
3378 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
3380 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
3383 2003-09-09 Alan Modra <amodra@bigpond.net.au>
3385 * configure: Regenerate.
3387 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3389 * Makefile.am: Remove build rules for ffitest.
3390 * Makefile.in: Rebuilt.
3392 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3394 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3395 about implicit declaration of abort().
3397 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3399 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3400 * Makefile.in: Rebuilt.
3401 * configure.in: Add dejagnu test framework.
3402 * configure: Rebuilt.
3404 * testsuite/Makefile.am: New file.
3405 * testsuite/Makefile.in: Built
3406 * testsuite/lib/libffi-dg.exp: New file.
3407 * testsuite/config/default.exp: Likewise.
3408 * testsuite/libffi.call/call.exp: Likewise.
3409 * testsuite/libffi.call/ffitest.h: Likewise.
3410 * testsuite/libffi.call/closure_fn0.c: Likewise.
3411 * testsuite/libffi.call/closure_fn1.c: Likewise.
3412 * testsuite/libffi.call/closure_fn2.c: Likewise.
3413 * testsuite/libffi.call/closure_fn3.c: Likewise.
3414 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3415 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3416 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3417 * testsuite/libffi.call/cls_2byte.c: Likewise.
3418 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3419 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3420 * testsuite/libffi.call/cls_4byte.c: Likewise.
3421 * testsuite/libffi.call/cls_5byte.c: Likewise.
3422 * testsuite/libffi.call/cls_6byte.c: Likewise.
3423 * testsuite/libffi.call/cls_7byte.c: Likewise.
3424 * testsuite/libffi.call/cls_8byte.c: Likewise.
3425 * testsuite/libffi.call/cls_12byte.c: Likewise.
3426 * testsuite/libffi.call/cls_16byte.c: Likewise.
3427 * testsuite/libffi.call/cls_20byte.c: Likewise.
3428 * testsuite/libffi.call/cls_24byte.c: Likewise.
3429 * testsuite/libffi.call/cls_double.c: Likewise.
3430 * testsuite/libffi.call/cls_float.c: Likewise.
3431 * testsuite/libffi.call/cls_uchar.c: Likewise.
3432 * testsuite/libffi.call/cls_uint.c: Likewise.
3433 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3434 * testsuite/libffi.call/cls_ushort.c: Likewise.
3435 * testsuite/libffi.call/float.c: Likewise.
3436 * testsuite/libffi.call/float1.c: Likewise.
3437 * testsuite/libffi.call/float2.c: Likewise.
3438 * testsuite/libffi.call/many.c: Likewise.
3439 * testsuite/libffi.call/many_win32.c: Likewise.
3440 * testsuite/libffi.call/nested_struct.c: Likewise.
3441 * testsuite/libffi.call/nested_struct1.c: Likewise.
3442 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3443 * testsuite/libffi.call/problem1.c: Likewise.
3444 * testsuite/libffi.call/promotion.c: Likewise.
3445 * testsuite/libffi.call/return_ll.c: Likewise.
3446 * testsuite/libffi.call/return_sc.c: Likewise.
3447 * testsuite/libffi.call/return_uc.c: Likewise.
3448 * testsuite/libffi.call/strlen.c: Likewise.
3449 * testsuite/libffi.call/strlen_win32.c: Likewise.
3450 * testsuite/libffi.call/struct1.c: Likewise.
3451 * testsuite/libffi.call/struct2.c: Likewise.
3452 * testsuite/libffi.call/struct3.c: Likewise.
3453 * testsuite/libffi.call/struct4.c: Likewise.
3454 * testsuite/libffi.call/struct5.c: Likewise.
3455 * testsuite/libffi.call/struct6.c: Likewise.
3456 * testsuite/libffi.call/struct7.c: Likewise.
3457 * testsuite/libffi.call/struct8.c: Likewise.
3458 * testsuite/libffi.call/struct9.c: Likewise.
3459 * testsuite/libffi.special/special.exp: New file.
3460 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3461 * testsuite/libffi.special/unwindtest.cc: Likewise.
3464 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3466 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3469 2003-08-02 Alan Modra <amodra@bigpond.net.au>
3471 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3473 (ffi_closure_helper_LINUX64): Likewise.
3474 * src/powerpc/linux64.S: Remove code writing to parm save area.
3475 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3476 address in lr from ffi_closure_helper_LINUX64 call to calculate
3477 table address. Optimize function tail.
3479 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3481 * src/sparc/ffi.c: Handle all floating point registers.
3482 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3484 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3486 * README: Note that libffi is not part of GCC. Update the project
3489 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3491 * src/powerpc/ppc_closure.S: Include ffi.h.
3493 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3495 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3496 Use C style comments.
3498 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3500 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3501 * Makefile.in: Regenerate.
3502 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3503 * configure: Regenerate.
3504 * include/ffi.h.in: Add SHmedia support.
3505 * src/sh64/ffi.c: New file.
3506 * src/sh64/sysv.S: New file.
3508 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3510 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3511 should be read-only.
3512 * configure: Rebuilt.
3513 * fficonfig.h.in: Rebuilt.
3514 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3515 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3516 * src/powerpc/linux64.S: Likewise.
3517 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3518 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3519 if -fpic/-fPIC/-mrelocatable.
3520 * src/powerpc/powerpc_closure.S: Likewise.
3521 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3522 #write in .eh_frame flags.
3523 * src/sparc/v9.S: Likewise.
3524 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3525 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3526 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3528 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3530 Fixes PR bootstrap/10656
3531 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3532 support for .register pseudo-op.
3533 * src/sparc/v8.S: Use it.
3534 * fficonfig.h.in: Rebuilt.
3535 * configure: Rebuilt.
3537 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3539 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3540 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3541 Make it the default on POWERPC64.
3542 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3543 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3544 * configure: Rebuilt.
3545 * src/powerpc/ffi.c (hidden): Define.
3546 (ffi_prep_args_SYSV): Renamed from
3547 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3548 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3549 ASM_NEEDS_REGISTERS64): New.
3550 (ffi_prep_args64): New function.
3551 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3552 (ffi_call): Likewise.
3553 (ffi_prep_closure): Likewise.
3554 (flush_icache): Surround by #ifndef POWERPC64.
3555 (ffi_dblfl): New union type.
3556 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3557 (ffi_closure_helper_LINUX64): New function.
3558 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3560 * src/powerpc/sysv.S: Likewise.
3561 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3562 * src/powerpc/linux64.S: New file.
3563 * src/powerpc/linux64_closure.S: New file.
3564 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3565 src/powerpc/linux64_closure.S.
3566 (TARGET_SRC_POWERPC): Likewise.
3568 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3569 closure_test_fn3): Fix result printing on big-endian 64-bit
3571 (main): Print tst2_arg instead of uninitialized tst2_result.
3573 * src/ffitest.c (main): Hide what closure pointer really points to
3576 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3578 * configure.in (arm-*-netbsdelf*): Add configuration.
3579 (configure): Regenerated.
3581 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3583 * include/Makefile.in: Regenerate.
3585 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3587 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3589 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3590 Receive closure pointer through parameter, read args using
3591 __builtin_dwarf_cfa.
3592 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3594 2003-03-12 Andreas Schwab <schwab@suse.de>
3596 * configure.in: Avoid trailing /. in toolexeclibdir.
3597 * configure: Rebuilt.
3599 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3601 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3603 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3605 * libffi/src/powerpc/darwin_closure.S:
3606 Fix alignement bug, allocate 8 bytes for the result.
3607 * libffi/src/powerpc/aix_closure.S:
3609 * libffi/src/powerpc/ffi_darwin.c:
3610 Update stackframe description for aix/darwin_closure.S.
3612 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3614 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3617 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3618 Andreas Schwab <schwab@suse.de>
3620 * configure.in: Adjust command to source config-ml.in to account
3621 for changes to the libffi_basedir definition.
3622 (libffi_basedir): Remove ${srcdir} from value and include trailing
3625 * configure: Regenerate.
3627 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3629 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3631 2003-01-28 Andrew Haley <aph@redhat.com>
3633 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3634 * src/x86/ffi64.c (ffi_prep_closure): New.
3635 (ffi_closure_UNIX64_inner): New.
3636 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3638 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
3640 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3641 Remove USE_LIBDIR conditional.
3642 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3643 * Makefile.in, configure: Rebuilt.
3645 2003-01027 David Edelsohn <edelsohn@gnu.org>
3647 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3648 * Makefile.in: Regenerate.
3650 2003-01-22 Andrew Haley <aph@redhat.com>
3652 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3655 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3657 * src/powerpc/darwin.S: Add unwind info.
3658 * src/powerpc/darwin_closure.S: Likewise.
3660 2003-01-14 Andrew Haley <aph@redhat.com>
3662 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3663 (ffi_prep_cif_machdep): Likewise.
3664 * src/x86/unix64.S: Add unwind info.
3666 2003-01-14 Andreas Jaeger <aj@suse.de>
3668 * src/ffitest.c (main): Only use ffi_closures if those are
3671 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3673 * libffi/src/ffitest.c
3674 add closure testcases
3676 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3678 * libffi/src/powerpc/ffi.c
3679 fix alignment bug for float (4 byte aligned iso 8 byte)
3681 2003-01-09 Geoffrey Keating <geoffk@apple.com>
3683 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3684 * src/powerpc/darwin.S: Remove RCS version string.
3686 2003-01-03 Jeff Sturm <jsturm@one-point.com>
3688 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3689 * src/ffitest.c (main): Use static storage for closure.
3690 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3691 * src/sparc/v8.S (ffi_closure_v8): New.
3692 * src/sparc/v9.S (ffi_closure_v9): New.
3694 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3696 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3697 enumeration for X86_WIN32.
3698 * src/x86/win32.S: Added ffi_call_STDCALL function
3700 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3701 switch cases for recognising FFI_STDCALL and
3702 calling ffi_call_STDCALL if target is X86_WIN32.
3703 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3704 stdcall versions of the "my_strlen" and "many"
3705 test functions (for X86_WIN32).
3706 Added test cases to test stdcall invocation using
3709 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3711 * src/sh/sysv.S: Add DWARF2 unwind info.
3713 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3715 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3717 2002-11-26 Jim Wilson <wilson@redhat.com>
3719 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3721 2002-11-23 H.J. Lu <hjl@gnu.org>
3723 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3724 Include ../config/accross.m4.
3725 * aclocal.m4; Rebuild.
3726 * configure: Likewise.
3728 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3730 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3732 2002-11-11 DJ Delorie <dj@redhat.com>
3734 * configure.in: Look for common files in the right place.
3736 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3738 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3739 raw data as _Jv_word values, not ffi_raw.
3740 (ffi_java_ptrarray_to_raw): Likewise.
3741 (ffi_java_rvalue_to_raw): New function.
3742 (ffi_java_raw_call): Call it.
3743 (ffi_java_raw_to_rvalue): New function.
3744 (ffi_java_translate_args): Call it.
3745 * src/ffitest.c (closure_test_fn): Interpret return value
3746 as ffi_arg, not int.
3747 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3748 FFI_TYPE_POINTER case.
3749 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3750 values extended to word size.
3752 2002-10-02 Andreas Jaeger <aj@suse.de>
3754 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3756 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
3758 * include/ffi.h.in: Fix i386 win32 compilation.
3760 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3762 * configure.in: Add s390x-*-linux-* target.
3763 * configure: Regenerate.
3764 * include/ffi.h.in: Define S390X for s390x targets.
3765 (FFI_CLOSURES): Define for s390/s390x.
3766 (FFI_TRAMPOLINE_SIZE): Likewise.
3767 (FFI_NATIVE_RAW_API): Likewise.
3768 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3769 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
3770 * src/s390/ffi.c: Major rework of existing code. Add support for
3771 s390x targets. Add closure support.
3772 * src/s390/sysv.S: Likewise.
3774 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
3776 * src/arm/sysv.S: Fix typo.
3778 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
3780 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3781 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3782 (ffi_call_SYSV): Handle soft-float.
3784 2002-09-27 Bo Thorsen <bo@suse.de>
3786 * include/ffi.h.in: Fix multilib x86-64 support.
3788 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3790 * Makefile.am (all-multi): Fix multilib parallel build.
3792 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3794 * configure.in (sh[34]*-*-linux*): Add brackets.
3795 * configure: Regenerate.
3797 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3799 * Makefile.am: Add SH support.
3800 * Makefile.in: Regenerate.
3801 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3802 * configure: Regenerate.
3803 * include/ffi.h.in: Add SH support.
3804 * src/sh/ffi.c: New file.
3805 * src/sh/sysv.S: New file.
3806 * src/types.c: Add SH support.
3808 2002-07-16 Bo Thorsen <bo@suse.de>
3810 * src/x86/ffi64.c: New file that adds x86-64 support.
3811 * src/x86/unix64.S: New file that handles argument setup for
3813 * src/x86/sysv.S: Don't use this on x86-64.
3814 * src/x86/ffi.c: Don't use this on x86-64.
3816 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3818 * src/ffitest.c (struct6): New test that tests a special case in
3820 (struct7): Likewise.
3821 (struct8): Likewise.
3822 (struct9): Likewise.
3823 (closure_test_fn): Silence warning about this when it's not used.
3824 (main): Add the new tests.
3825 (main): Fix a couple of wrong casts and silence some compiler warnings.
3826 * include/ffi.h.in: Add x86-64 ABI definition.
3827 * fficonfig.h.in: Regenerate.
3828 * Makefile.am: Add x86-64 support.
3829 * configure.in: Likewise.
3830 * Makefile.in: Regenerate.
3831 * configure: Likewise.
3833 2002-06-24 Bo Thorsen <bo@suse.de>
3835 * src/types.c: Merge settings for similar architectures.
3836 Add x86-64 sizes and alignments.
3838 2002-06-23 Bo Thorsen <bo@suse.de>
3840 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3841 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3842 * src/mips/ffi.c (ffi_prep_args): Likewise.
3843 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3845 2002-07-18 H.J. Lu (hjl@gnu.org)
3847 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3848 (libffi_la_SOURCES): Support MIPS_LINUX.
3849 (libffi_convenience_la_SOURCES): Likewise.
3850 * Makefile.in: Regenerated.
3852 * configure.in (mips64*-*): Skip.
3853 (mips*-*-linux*): New.
3854 * configure: Regenerated.
3856 * src/mips/ffi.c: Include <sgidefs.h>.
3858 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3860 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3862 2002-05-27 Roger Sayle <roger@eyesopen.com>
3864 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3866 2002-05-27 Bo Thorsen <bo@suse.de>
3868 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3871 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3873 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3874 beginning of function (for older apple cc).
3876 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3878 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3879 script entry, and set LD to it when configuring multilibs.
3880 * configure: Rebuilt.
3882 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3884 * configure.in (sparc64-*-netbsd*): Add target.
3885 (sparc-*-netbsdelf*): Likewise.
3886 * configure: Regenerate.
3888 2002-04-28 David S. Miller <davem@redhat.com>
3890 * configure.in, configure: Fix SPARC test in previous change.
3892 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3894 * Makefile.am: Add Linux for S/390 support.
3895 * Makefile.in: Regenerate.
3896 * configure.in: Add Linux for S/390 support.
3897 * configure: Regenerate.
3898 * include/ffi.h.in: Add Linux for S/390 support.
3899 * src/s390/ffi.c: New file from libffi CVS tree.
3900 * src/s390/sysv.S: New file from libffi CVS tree.
3902 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3904 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3906 * src/sparc/v8.S: Use it.
3907 * src/sparc/v9.S: Likewise.
3908 * fficonfig.h.in: Rebuilt.
3909 * configure: Rebuilt.
3911 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3913 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3915 * src/ia64/unix.S: Add unwind information. Fix comments.
3916 Save sp in a way that's compatible with unwind info.
3917 (ffi_call_unix): Correctly restore sp in all cases.
3918 * src/ia64/ffi.c: Add, fix comments.
3920 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3922 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3924 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3926 * configure.in (alpha*-*-netbsd*): Add target.
3927 * configure: Regenerate.
3929 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3931 * src/sparc/v8.S: Add unwind info.
3932 * src/sparc/v9.S: Likewise.
3934 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3936 * configure.in: Enable i*86-*-netbsdelf*.
3937 * configure: Rebuilt.
3939 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3942 * src/mips/n32.s: Delete
3943 * src/mips/o32.s: Delete
3945 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3947 * configure.in: Enable alpha*-*-freebsd*.
3948 * configure: Rebuilt.
3950 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3952 * Makefile.am: libfficonvenience -> libffi_convenience.
3953 * Makefile.in: Rebuilt.
3955 * Makefile.am: Define ffitest_OBJECTS.
3956 * Makefile.in: Rebuilt.
3958 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3959 David Edelsohn <edelsohn@gnu.org>
3961 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3962 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3963 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3964 * Makefile.in: Regenerate.
3965 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3966 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3967 (flush_icache, flush_range): New functions.
3968 (ffi_closure_helper_DARWIN): New function.
3969 * src/powerpc/aix_closure.S: New file.
3970 * src/powerpc/darwin_closure.S: New file.
3972 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3974 * include/ffi.h.in: Add typedef for ffi_arg.
3975 * src/ffitest.c (main): Declare rint with ffi_arg.
3977 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3979 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3980 number of GPRs for floating-point arguments.
3982 2002-01-31 Anthony Green <green@redhat.com>
3984 * configure: Rebuilt.
3985 * configure.in: Replace CHECK_SIZEOF and endian tests with
3986 cross-compiler friendly macros.
3987 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3990 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3992 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3993 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3995 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3997 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3998 (TARGET_SRC_POWERPC_AIX): New.
3999 (POWERPC_AIX): New stanza.
4000 * Makefile.in: Regenerate.
4001 * configure.in: Add AIX case.
4002 * configure: Regenerate.
4003 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
4004 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
4005 size. Fix "long double" support.
4006 (ffi_call): Add FFI_AIX case.
4007 * src/powerpc/aix.S: New.
4009 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
4011 Implement Darwin PowerPC ABI.
4012 * configure.in: Handle powerpc-*-darwin*.
4013 * Makefile.am: Set source files for POWERPC_DARWIN.
4014 * configure: Rebuilt.
4015 * Makefile.in: Rebuilt.
4016 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
4018 * src/powerpc/darwin.S: New file.
4019 * src/powerpc/ffi_darwin.c: New file.
4021 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
4023 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
4025 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4027 * src/x86/sysv.S: Avoid gas-only .balign directive.
4028 Use C style comments.
4030 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4032 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
4033 Fixes PR bootstrap/3563.
4035 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4037 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
4039 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4041 * configure.in: Recognize sparc*-sun-* host.
4042 * configure: Regenerate.
4044 2001-06-06 Andrew Haley <aph@redhat.com>
4046 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
4048 2001-06-03 Andrew Haley <aph@redhat.com>
4050 * src/alpha/osf.S: Add unwind info.
4051 * src/powerpc/sysv.S: Add unwind info.
4052 * src/powerpc/ppc_closure.S: Likewise.
4054 2000-05-31 Jeff Sturm <jsturm@one-point.com>
4056 * configure.in: Fix AC_ARG_ENABLE usage.
4057 * configure: Rebuilt.
4059 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
4061 * configure.in: Remove warning about beta code.
4062 * configure: Rebuilt.
4064 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
4066 * src/ia64/unix.S: Restore stack pointer when returning from
4068 * src/ia64/ffi.c: Fix typo in comment.
4070 2001-04-18 Jim Wilson <wilson@redhat.com>
4072 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
4073 to eliminate RAW DV.
4075 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
4077 * Makefile.am: Make a libtool convenience library.
4078 * Makefile.in: Rebuilt.
4080 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
4082 * configure.in: Use different syntax for subdirectory creation.
4083 * configure: Rebuilt.
4085 2001-03-27 Jon Beniston <jon@beniston.com>
4087 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
4088 * configure: Rebuilt.
4089 * Makefile.am: Added X86_WIN32 target support.
4090 * Makefile.in: Rebuilt.
4092 * include/ffi.h.in: Added X86_WIN32 target support.
4094 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
4095 * src/types.c: Added X86_WIN32 target support.
4097 * src/x86/win32.S: New file. Based on sysv.S, but with EH
4098 stuff removed and made to work with CygWin's gas.
4100 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
4102 * configure.in: Make target subdirectory in build dir.
4103 * Makefile.am: Override suffix based rules to specify correct output
4105 * Makefile.in: Rebuilt.
4106 * configure: Rebuilt.
4108 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
4110 * src/powerpc/ppc_closure.S: New file.
4111 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
4112 involving long long and register pairs.
4113 (ffi_prep_closure): New function.
4114 (flush_icache): Likewise.
4115 (ffi_closure_helper_SYSV): Likewise.
4116 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
4117 (FFI_TRAMPOLINE_SIZE): Likewise.
4118 (FFI_NATIVE_RAW_API): Likewise.
4119 * Makefile.in: Rebuilt.
4120 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
4121 (TARGET_SRC_POWERPC): Likewise.
4123 2001-03-19 Tom Tromey <tromey@redhat.com>
4125 * Makefile.in: Rebuilt.
4126 * Makefile.am (ffitest_LDFLAGS): New macro.
4128 2001-03-02 Nick Clifton <nickc@redhat.com>
4130 * include/ffi.h.in: Remove RCS ident string.
4131 * include/ffi_mips.h: Remove RCS ident string.
4132 * src/debug.c: Remove RCS ident string.
4133 * src/ffitest.c: Remove RCS ident string.
4134 * src/prep_cif.c: Remove RCS ident string.
4135 * src/types.c: Remove RCS ident string.
4136 * src/alpha/ffi.c: Remove RCS ident string.
4137 * src/alpha/osf.S: Remove RCS ident string.
4138 * src/arm/ffi.c: Remove RCS ident string.
4139 * src/arm/sysv.S: Remove RCS ident string.
4140 * src/mips/ffi.c: Remove RCS ident string.
4141 * src/mips/n32.S: Remove RCS ident string.
4142 * src/mips/o32.S: Remove RCS ident string.
4143 * src/sparc/ffi.c: Remove RCS ident string.
4144 * src/sparc/v8.S: Remove RCS ident string.
4145 * src/sparc/v9.S: Remove RCS ident string.
4146 * src/x86/ffi.c: Remove RCS ident string.
4147 * src/x86/sysv.S: Remove RCS ident string.
4149 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
4151 * include/ffi.h.in: Change sourceware.cygnus.com references to
4154 2000-12-09 Richard Henderson <rth@redhat.com>
4156 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
4157 (ffi_closure_osf_inner): Index rather than increment avalue
4158 and arg_types. Give ffi_closure_osf the raw return value type.
4159 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
4162 2000-12-07 Richard Henderson <rth@redhat.com>
4164 * src/raw_api.c (ffi_translate_args): Fix typo.
4165 (ffi_prep_closure): Likewise.
4167 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
4168 FFI_TRAMPOLINE_SIZE.
4169 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
4170 cif->bytes for new ffi_call_osf implementation.
4171 (ffi_prep_args): Absorb into ...
4172 (ffi_call): ... here. Do all stack allocation here and
4173 avoid a callback function.
4174 (ffi_prep_closure, ffi_closure_osf_inner): New.
4175 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
4176 (ffi_closure_osf): New.
4178 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
4180 * config.guess, config.sub, install-sh: Removed.
4181 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
4182 * Makefile.in: Rebuilt.
4184 * acinclude.m4: Include libtool macros from the top level.
4185 * aclocal.m4, configure: Rebuilt.
4187 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
4189 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
4190 * configure: Rebuilt.
4192 2000-05-11 Scott Bambrough <scottb@netwinder.org>
4194 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
4195 memory correctly. Use conditional instructions, not branches where
4198 2000-05-04 Tom Tromey <tromey@cygnus.com>
4200 * configure: Rebuilt.
4201 * configure.in: Match `arm*-*-linux-*'.
4202 From Chris Dornan <cdornan@arm.com>.
4204 2000-04-28 Jakub Jelinek <jakub@redhat.com>
4206 * Makefile.am (SUBDIRS): Define.
4207 (AM_MAKEFLAGS): Likewise.
4208 (Multilib support.): Add section.
4209 * Makefile.in: Rebuilt.
4210 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
4211 New variables. Set for gcc using -print-multi-lib. Export them
4213 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
4214 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
4216 (extra_compiler_flags_value, extra_compiler_flags): Check these
4217 for extra compiler options which need to be passed down in
4220 2000-04-16 Anthony Green <green@redhat.com>
4222 * configure: Rebuilt.
4223 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
4225 2000-04-14 Jakub Jelinek <jakub@redhat.com>
4227 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
4228 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
4229 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
4230 Replace all void * sizeofs with sizeof(int).
4231 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
4232 different than DOUBLE.
4233 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
4234 (ffi_prep_args_v9): New function.
4235 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
4236 (ffi_V9_return_struct): New function.
4237 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
4238 32bit code (not yet cross-arch calls).
4239 * src/sparc/v8.S: Add struct return delay nop.
4241 * src/sparc/v9.S: New file.
4242 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
4243 is used on sparc64 only for structures larger than 32 bytes.
4244 Pass by reference for structures is done for structure arguments
4245 larger than 16 bytes.
4246 * src/ffitest.c (main): Use 64bit rint on sparc64.
4247 Run long long tests on sparc.
4248 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
4250 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
4252 * configure.in (sparc-*-linux*): New supported target.
4253 (sparc64-*-linux*): Likewise.
4254 * configure: Rebuilt.
4255 * Makefile.am: Add v9.S to SPARC files.
4256 * Makefile.in: Likewise.
4257 (LINK): Surround $(CCLD) into double quotes, so that multilib
4258 compiles work correctly.
4260 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
4262 * configure: Rebuilt.
4263 * configure.in: (i*86-*-solaris*): New libffi target. Patch
4264 proposed by Bryce McKinlay.
4266 2000-03-20 Tom Tromey <tromey@cygnus.com>
4268 * Makefile.in: Hand edit for java_raw_api.lo.
4270 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
4272 * config.guess, config.sub: Update from the gcc tree.
4273 Fix for PR libgcj/168.
4275 2000-03-03 Tom Tromey <tromey@cygnus.com>
4277 * Makefile.in: Fixed ia64 by hand.
4279 * configure: Rebuilt.
4280 * configure.in (--enable-multilib): New option.
4281 (libffi_basedir): New subst.
4282 (AC_OUTPUT): Added multilib code.
4284 2000-03-02 Tom Tromey <tromey@cygnus.com>
4286 * Makefile.in: Rebuilt.
4287 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
4290 2000-02-25 Hans Boehm <boehm@acm.org>
4292 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
4294 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
4296 (ffi_prep_raw_closure): Use ffi_translate_args, not
4297 ffi_closure_translate.
4298 * src/java_raw_api.c: New file.
4299 * src/ffitest.c (closure_test_fn): New function.
4300 (main): Define `rint' as long long on IA64. Added new test when
4301 FFI_CLOSURES is defined.
4302 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
4303 (ffi_abi): Recognize IA64.
4304 (ffi_raw): Added `flt' field.
4305 Added "Java raw API" code.
4306 * configure.in: Recognize ia64.
4307 * Makefile.am (TARGET_SRC_IA64): New macro.
4308 (libffi_la_common_SOURCES): Added java_raw_api.c.
4309 (libffi_la_SOURCES): Define in IA64 case.
4311 2000-01-04 Tom Tromey <tromey@cygnus.com>
4313 * Makefile.in: Rebuilt with newer automake.
4315 1999-12-31 Tom Tromey <tromey@cygnus.com>
4317 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
4319 1999-09-01 Tom Tromey <tromey@cygnus.com>
4321 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
4323 * fficonfig.h.in: Rebuilt.
4324 * configure: Rebuilt.
4325 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
4326 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
4327 * acconfig.h: Don't #undef PACKAGE or VERSION.
4329 1999-08-09 Anthony Green <green@cygnus.com>
4331 * include/ffi.h.in: Try to work around messy header problem
4332 with PACKAGE and VERSION.
4334 * configure: Rebuilt.
4335 * configure.in: Change version to 2.00-beta.
4337 * fficonfig.h.in: Rebuilt.
4338 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
4340 * src/x86/ffi.c (ffi_raw_call): Rename.
4342 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
4344 * src/x86/ffi.c (ffi_closure_SYSV): New function.
4345 (ffi_prep_incoming_args_SYSV): Ditto.
4346 (ffi_prep_closure): Ditto.
4347 (ffi_closure_raw_SYSV): Ditto.
4348 (ffi_prep_raw_closure): More ditto.
4349 (ffi_call_raw): Final ditto.
4351 * include/ffi.h.in: Add definitions for closure and raw API.
4353 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4356 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
4358 * src/raw_api.c: New file.
4360 * include/ffi.h.in (ffi_raw): New type.
4361 (UINT_ARG, SINT_ARG): New defines.
4362 (ffi_closure, ffi_raw_closure): New types.
4363 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4365 * configure.in: Add check for endianness and sizeof void*.
4367 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
4368 instead of directly.
4370 * configure: Rebuilt.
4372 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
4374 * configure.in: Add x86 and powerpc BeOS configurations.
4375 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
4377 1999-05-09 Anthony Green <green@cygnus.com>
4379 * configure.in: Add warning about this being beta code.
4380 Remove src/Makefile.am from the picture.
4381 * configure: Rebuilt.
4383 * Makefile.am: Move logic from src/Makefile.am. Add changes
4384 to support libffi as a target library.
4385 * Makefile.in: Rebuilt.
4387 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
4388 Upgraded to new autoconf, automake, libtool.
4392 * LICENSE: Update copyright date.
4394 * src/Makefile.am, src/Makefile.in: Removed.
4396 1998-11-29 Anthony Green <green@cygnus.com>
4398 * include/ChangeLog: Removed.
4399 * src/ChangeLog: Removed.
4400 * src/mips/ChangeLog: Removed.
4401 * src/sparc/ChangeLog: Remboved.
4402 * src/x86/ChangeLog: Removed.
4404 * ChangeLog.v1: Created.