1 2011-02-09 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
4 * testsuite/libffi.call/cls_pointer.c (main): Cast void * to
6 * testsuite/libffi.call/cls_pointer_stack.c (main): Likewise.
8 2011-02-07 Joel Sherrill <joel.sherrill@oarcorp.com>
10 * libffi/src/m68k/ffi.c: Add RTEMS support for cache flushing.
11 Handle case when CPU variant does not have long double support.
12 * libffi/src/m68k/sysv.S: Add support for mc68000, Coldfire,
13 and cores with soft floating point.
15 2011-02-07 Joel Sherrill <joel.sherrill@oarcorp.com>
17 * configure.ac: Add mips*-*-rtems* support.
18 * configure: Regenerate.
19 * src/mips/ffitarget.h: Ensure needed constants are available
20 for targets which do not have sgidefs.h.
22 2011-01-26 Dave Korn <dave.korn.cygwin@gmail.com>
25 * configure.ac (AM_LTLDFLAGS): Add -bindir option for windows DLLs.
26 * configure: Regenerate.
28 2010-12-18 Iain Sandoe <iains@gcc.gnu.org>
32 * src/powerpc/darwin_closure.S: Provide Darwin64 implementation,
34 * src/powerpc/ffitarget.h (POWERPC_DARWIN64): New,
35 (FFI_TRAMPOLINE_SIZE): Update for Darwin64.
36 * src/powerpc/darwin.S: Provide Darwin64 implementation,
38 * src/powerpc/ffi_darwin.c: Likewise.
40 2010-12-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
42 * configure.ac (libffi_cv_as_ascii_pseudo_op): Use double
44 (libffi_cv_as_string_pseudo_op): Likewise.
45 * configure: Regenerate.
47 2010-12-03 Chung-Lin Tang <cltang@codesourcery.com>
49 * src/arm/sysv.S (ffi_closure_SYSV): Add UNWIND to .pad directive.
50 (ffi_closure_VFP): Same.
51 (ffi_call_VFP): Move down to before ffi_closure_VFP. Add '.fpu vfp'
54 2010-12-01 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
56 * testsuite/libffi.call/ffitest.h [__sgi] (PRId64, PRIu64): Define.
59 2010-11-29 Richard Henderson <rth@redhat.com>
60 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
62 * src/x86/sysv.S (FDE_ENCODING, FDE_ENCODE): Define.
63 (.eh_frame): Use FDE_ENCODING.
64 (.LASFDE1, .LASFDE2, LASFDE3): Simplify with FDE_ENCODE.
66 2010-11-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
68 * testsuite/lib/libffi-dg.exp: Rename ...
69 * testsuite/lib/libffi.exp: ... to this.
70 * libffi/testsuite/libffi.call/call.exp: Don't load libffi-dg.exp.
71 * libffi/testsuite/libffi.special/special.exp: Likewise.
73 2010-10-28 Chung-Lin Tang <cltang@codesourcery.com>
75 * src/arm/ffi.c (ffi_prep_args): Add VFP register argument handling
76 code, new parameter, and return value. Update comments.
77 (ffi_prep_cif_machdep): Add case for VFP struct return values. Add
78 call to layout_vfp_args().
79 (ffi_call_SYSV): Update declaration.
80 (ffi_call_VFP): New declaration.
81 (ffi_call): Add VFP struct return conditions. Call ffi_call_VFP()
83 (ffi_closure_VFP): New declaration.
84 (ffi_closure_SYSV_inner): Add new vfp_args parameter, update call to
85 ffi_prep_incoming_args_SYSV().
86 (ffi_prep_incoming_args_SYSV): Update parameters. Add VFP argument
88 (ffi_prep_closure_loc): Pass ffi_closure_VFP to trampoline
89 construction under VFP hard-float.
90 (rec_vfp_type_p): New function.
92 (place_vfp_arg): Same.
93 (layout_vfp_args): Same.
94 * src/arm/ffitarget.h (ffi_abi): Add FFI_VFP. Define FFI_DEFAULT_ABI
95 based on __ARM_PCS_VFP.
96 (FFI_EXTRA_CIF_FIELDS): Define for adding VFP hard-float specific
98 (FFI_TYPE_STRUCT_VFP_FLOAT): Define internally used type code.
99 (FFI_TYPE_STRUCT_VFP_DOUBLE): Same.
100 * src/arm/sysv.S (ffi_call_SYSV): Change call of ffi_prep_args() to
101 direct call. Move function pointer load upwards.
102 (ffi_call_VFP): New function.
103 (ffi_closure_VFP): Same.
105 * testsuite/lib/libffi-dg.exp (check-flags): New function.
106 (dg-skip-if): New function.
107 * testsuite/libffi.call/cls_double_va.c: Skip if target is arm*-*-*
108 and compiler options include -mfloat-abi=hard.
109 * testsuite/libffi.call/cls_longdouble_va.c: Same.
111 2010-10-01 Jakub Jelinek <jakub@redhat.com>
114 * src/x86/ffi64.c (ffi_prep_cif_machdep): Ensure cif->bytes is
116 * testsuite/libffi.call/many2.c: New test.
118 2010-08-20 Mark Wielaard <mjw@redhat.com>
120 * src/closures.c (open_temp_exec_file_mnt): Check if getmntent_r
123 2010-08-09 Andreas Tobler <andreast@fgznet.ch>
125 * configure.ac: Add target powerpc64-*-freebsd*.
126 * configure: Regenerate.
127 * testsuite/libffi.call/cls_align_longdouble_split.c: Pass
128 -mlong-double-128 only to linux targets.
129 * testsuite/libffi.call/cls_align_longdouble_split2.c: Likewise.
130 * testsuite/libffi.call/cls_longdouble.c: Likewise.
131 * testsuite/libffi.call/huge_struct.c: Likewise.
133 2010-07-10 Evan Phoenix <evan@fallingsnow.net>
135 * src/closures.c (selinux_enabled_check): Fix strncmp usage bug.
137 2010-07-07 Dan HorĂ¡k <dan@danny.cz>
139 * include/ffi.h.in: Protect #define with #ifndef.
140 * src/powerpc/ffitarget.h: Ditto.
141 * src/s390/ffitarget.h: Ditto.
142 * src/sparc/ffitarget.h: Ditto.
144 2010-07-07 Neil Roberts <neil@linux.intel.com>
146 * src/x86/sysv.S (ffi_call_SYSV): Align the stack pointer to
149 2010-07-02 Jakub Jelinek <jakub@redhat.com>
151 * Makefile.am (AM_MAKEFLAGS): Pass also mandir to submakes.
152 * Makefile.in: Regenerated.
154 2010-05-19 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
156 * configure.ac (libffi_cv_as_x86_pcrel): Check for illegal in as
158 (libffi_cv_as_ascii_pseudo_op): Check for .ascii.
159 (libffi_cv_as_string_pseudo_op): Check for .string.
160 * configure: Regenerate.
161 * fficonfig.h.in: Regenerate.
162 * src/x86/sysv.S (.eh_frame): Use .ascii, .string or error.
164 2010-05-05 Michael Kohler <michaelkohler@live.com>
166 * src/dlmalloc.c (dlfree): Fix spelling.
167 * src/ia64/ffi.c (ffi_prep_cif_machdep): Ditto.
168 * configure.ac: Ditto.
169 * configure: Rebuilt.
171 2010-04-13 Dan Witte <dwitte@mozilla.com>
173 * msvcc.sh: Build with -W3 instead of -Wall.
174 * src/powerpc/ffi_darwin.c: Remove build warnings.
175 * src/x86/ffi.c: Ditto.
176 * src/x86/ffitarget.h: Ditto.
178 2010-04-12 Dan Witte <dwitte@mozilla.com>
179 Walter Meinl <wuno@lsvw.de>
181 * configure.ac: Add OS/2 support.
182 * configure: Rebuilt.
183 * src/closures.c: Ditto.
184 * src/dlmalloc.c: Ditto.
185 * src/x86/win32.S: Ditto.
187 2010-04-07 Jakub Jelinek <jakub@redhat.com>
189 * testsuite/libffi.call/err_bad_abi.c: Remove unused args variable.
191 2010-04-02 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
193 * Makefile.in: Regenerate.
194 * aclocal.m4: Regenerate.
195 * include/Makefile.in: Regenerate.
196 * man/Makefile.in: Regenerate.
197 * testsuite/Makefile.in: Regenerate.
199 2010-03-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
201 * configure.ac (libffi_cv_as_x86_64_unwind_section_type): New test.
202 * configure: Regenerate.
203 * fficonfig.h.in: Regenerate.
204 * libffi/src/x86/unix64.S (.eh_frame)
205 [HAVE_AS_X86_64_UNWIND_SECTION_TYPE]: Use @unwind section type.
207 2010-03-14 Matthias Klose <doko@ubuntu.com>
209 * src/x86/ffi64.c: Fix typo in comment.
210 * src/x86/ffi.c: Use /* ... */ comment style.
212 2010-02-24 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
214 * doc/libffi.texi (The Closure API): Fix typo.
215 * doc/libffi.info: Remove.
217 2010-02-15 Matthias Klose <doko@ubuntu.com>
219 * src/arm/sysv.S (__ARM_ARCH__): Define for processor
222 2010-01-15 Anthony Green <green@redhat.com>
224 * README: Add notes on building with Microsoft Visual C++.
226 2010-01-15 Daniel Witte <dwitte@mozilla.com>
228 * msvcc.sh: New file.
230 * src/x86/win32.S: Port assembly routines to MSVC and #ifdef.
231 * src/x86/ffi.c: Tweak function declaration and remove excess
233 * include/ffi.h.in: Add __declspec(align(8)) to typedef struct
236 * src/x86/ffi.c: Merge ffi_call_SYSV and ffi_call_STDCALL into new
237 function ffi_call_win32 on X86_WIN32.
238 * src/x86/win32.S (ffi_call_SYSV): Rename to ffi_call_win32.
239 (ffi_call_STDCALL): Remove.
241 * src/prep_cif.c (ffi_prep_cif): Move stack space allocation code
242 to ffi_prep_cif_machdep for x86.
243 * src/x86/ffi.c (ffi_prep_cif_machdep): To here.
245 2010-01-15 Oliver Kiddle <okiddle@yahoo.co.uk>
247 * src/x86/ffitarget.h (ffi_abi): Check for __i386 and __amd64 for
248 Sun Studio compiler compatibility.
250 2010-01-12 Conrad Irwin <conrad.irwin@gmail.com>
252 * doc/libffi.texi: Add closure example.
254 2010-01-07 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
257 * testsuite/libffi.call/ffitest.h [__alpha__ && __osf__] (PRIdLL,
258 PRIuLL, PRId64, PRIu64, PRIuPTR): Define.
259 * testsuite/libffi.call/cls_align_sint64.c: Add -Wno-format on
261 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
262 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
263 * testsuite/libffi.call/return_ll1.c: Likewise.
264 * testsuite/libffi.call/stret_medium2.c: Likewise.
265 * testsuite/libffi.special/ffitestcxx.h (allocate_mmap): Cast
266 MAP_FAILED to char *.
268 2010-01-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
270 * src/mips/n32.S: Use .abicalls and .eh_frame with __GNUC__.
272 2009-12-31 Anthony Green <green@redhat.com>
274 * README: Update for libffi 3.0.9.
276 2009-12-27 Matthias Klose <doko@ubuntu.com>
278 * configure.ac (HAVE_LONG_DOUBLE): Define for mips when
280 * configure: Rebuilt.
282 2009-12-26 Anthony Green <green@redhat.com>
284 * testsuite/libffi.call/cls_longdouble_va.c: Mark as xfail for
286 * testsuite/libffi.call/cls_double_va.c: Ditto.
288 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
290 * testsuite/libffi.call/ffitest.h: Conditionally include stdint.h
292 * testsuite/libffi.special/unwindtest.cc: Ditto.
294 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
296 * configure.ac: Add amd64-*-openbsd*.
297 * configure: Rebuilt.
298 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Link
299 openbsd programs with -lpthread.
301 2009-12-26 Anthony Green <green@redhat.com>
303 * testsuite/libffi.call/cls_double_va.c,
304 testsuite/libffi.call/cls_longdouble.c,
305 testsuite/libffi.call/cls_longdouble_va.c,
306 testsuite/libffi.call/cls_pointer.c,
307 testsuite/libffi.call/cls_pointer_stack.c: Remove xfail for
308 mips*-*-* and arm*-*-*.
309 * testsuite/libffi.call/cls_align_longdouble_split.c,
310 testsuite/libffi.call/cls_align_longdouble_split2.c,
311 testsuite/libffi.call/stret_medium2.c,
312 testsuite/libffi.call/stret_medium.c,
313 testsuite/libffi.call/stret_large.c,
314 testsuite/libffi.call/stret_large2.c: Remove xfail for arm*-*-*.
316 2009-12-31 Kay Tietz <ktietz70@googlemail.com>
318 * testsuite/libffi.call/ffitest.h,
319 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRuLL): Fix
322 2009-12-31 Carlo Bramini <carlo.bramix@libero.it>
324 * configure.ac (AM_LTLDFLAGS): Define for windows hosts.
325 * Makefile.am (libffi_la_LDFLAGS): Add AM_LTLDFLAGS.
326 * configure: Rebuilt.
327 * Makefile.in: Rebuilt.
329 2009-12-31 Anthony Green <green@redhat.com>
332 * testsuite/libffi.call/huge_struct.c: New test case from Blake
335 2009-12-28 David Edelsohn <edelsohn@gnu.org>
337 * src/powerpc/ffi_darwin.c (ffi_prep_args): Copy abi and nargs to
339 (aix_adjust_aggregate_sizes): New function.
340 (ffi_prep_cif_machdep): Call it.
342 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
344 * configure.ac: Define FFI_MMAP_EXEC_WRIT for the given targets.
345 * configure: Regenerate.
346 * fficonfig.h.in: Likewise.
347 * src/closures.c: Remove the FFI_MMAP_EXEC_WRIT definition for
350 2009-12-26 Andreas Schwab <schwab@linux-m68k.org>
352 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Advance intarg_count
353 when a float arguments is passed in memory.
354 (ffi_closure_helper_SYSV): Mark general registers as used up when
355 a 64bit or soft-float long double argument is passed in memory.
357 2009-12-25 Matthias Klose <doko@ubuntu.com>
359 * man/ffi_call.3: Fix #include in examples.
360 * doc/libffi.texi: Add dircategory.
362 2009-12-25 Frank Everdij <f.p.x.everdij@tudelft.nl>
364 * include/ffi.h.in: Placed '__GNUC__' ifdef around
365 '__attribute__((aligned(8)))' in ffi_closure, fixes compile for
367 * include/ffi_common.h: Added '__sgi' define to non
368 '__attribute__((__mode__()))' integer typedefs.
369 * src/mips/ffi.c (ffi_call, ffi_closure_mips_inner_O32,
370 ffi_closure_mips_inner_N32): Added 'defined(_MIPSEB)' to BE check.
371 (ffi_closure_mips_inner_O32, ffi_closure_mips_inner_N32): Added
372 FFI_LONGDOUBLE support and alignment(N32 only).
373 * src/mips/ffitarget.h: Corrected '#include <sgidefs.h>' for IRIX and
374 fixed non '__attribute__((__mode__()))' integer typedefs.
375 * src/mips/n32.S: Put '#ifdef linux' around '.abicalls' and '.eh_frame'
376 since they are Linux/GNU Assembler specific.
378 2009-12-25 Bradley Smith <brad@brad-smith.co.uk>
380 * configure.ac, Makefile.am, src/avr32/ffi.c,
381 src/avr32/ffitarget.h,
382 src/avr32/sysv.S: Add AVR32 port.
383 * configure, Makefile.in: Rebuilt.
385 2009-12-21 Andreas Tobler <a.tobler@schweiz.org>
387 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
388 * configure: Regenerate.
390 2009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
392 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
394 2009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
396 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
399 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
401 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
404 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
406 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
408 2009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
411 * src/closures.c [X86_64 && __sun__ && __svr4__]
412 (FFI_MMAP_EXEC_WRIT): Define.
414 2009-12-08 David Daney <ddaney@caviumnetworks.com>
416 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
417 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
418 * testsuite/libffi.call/stret_large.c: Same.
419 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
420 * testsuite/libffi.call/stret_large2.c: Same.
421 * testsuite/libffi.call/stret_medium2.c: Same.
423 2009-12-07 David Edelsohn <edelsohn@gnu.org>
425 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
428 2009-12-05 David Edelsohn <edelsohn@gnu.org>
430 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
432 * src/powerpc/aix_closure.S: Same.
434 2009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
436 * Makefile.in: Regenerate.
437 * configure: Regenerate.
438 * include/Makefile.in: Regenerate.
439 * man/Makefile.in: Regenerate.
440 * testsuite/Makefile.in: Regenerate.
442 2009-12-04 David Edelsohn <edelsohn@gnu.org>
444 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
447 2009-12-04 Uros Bizjak <ubizjak@gmail.com>
450 * src/x86/ffi64.c (classify_argument): Update from
451 gcc/config/i386/i386.c.
452 (ffi_closure_unix64_inner): Do not use the address of two consecutive
453 SSE registers directly.
454 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
455 for x86_64 linux targets.
457 2009-12-04 David Edelsohn <edelsohn@gnu.org>
459 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
460 pfr for long double split between fpr13 and stack.
462 2009-12-03 David Edelsohn <edelsohn@gnu.org>
464 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
465 fparg_count twice for long double.
467 2009-12-03 David Edelsohn <edelsohn@gnu.org>
470 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
472 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
474 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
475 Remove xfails for x86 linux targets.
477 2009-12-02 David Edelsohn <edelsohn@gnu.org>
479 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
482 2009-12-01 David Edelsohn <edelsohn@gnu.org>
484 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
485 register usage. Call ffi_prep_args directly. Add long double
486 return value support.
487 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
488 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
489 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
490 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
492 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
494 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
495 Allocate result area between params and FPRs.
497 2009-11-30 David Edelsohn <edelsohn@gnu.org>
500 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
502 * src/powerpc/aix.S: Implement AIX64 version.
503 * src/powerpc/aix_closure.S: Implement AIX64 version.
504 (ffi_closure_ASM): Use extsb, lha and displament addresses.
505 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
507 (ffi_prep_cif_machdep): Same.
509 (ffi_closure_helper_DARWIN): Same.
511 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
514 * testsuite/libffi.call/testclosure.c: New test.
516 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
518 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
519 assembly version use of ___chkstk.
521 2009-09-23 Matthias Klose <doko@ubuntu.com>
523 PR libffi/40242, PR libffi/41443
524 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
525 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
526 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
527 Change the conditionals to __SOFTFP__ || __ARM_EABI__
528 for -mfloat-abi=softfp to work.
530 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
532 PR testsuite/32843 (strikes again)
533 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
534 enable proper extension on char and short.
536 2009-09-15 David Daney <ddaney@caviumnetworks.com>
538 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
539 handling for FFI_TYPE_POINTER.
540 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
541 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
542 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
543 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
544 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
545 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
546 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
547 structure and pointer returns.
548 (ffi_closure_N32): Add handling for pointer returns.
549 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
550 calc_n32_return_struct_flags): Handle soft-float.
551 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
552 (ffi_call_N32): Declare proper argument types.
553 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
556 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
558 * configure.ac (AC_PREREQ): Bump to 2.64.
560 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
562 * Makefile.am (install-html, install-pdf): Remove.
563 * Makefile.in: Regenerate.
565 * Makefile.in: Regenerate.
566 * aclocal.m4: Regenerate.
567 * configure: Regenerate.
568 * fficonfig.h.in: Regenerate.
569 * include/Makefile.in: Regenerate.
570 * man/Makefile.in: Regenerate.
571 * testsuite/Makefile.in: Regenerate.
573 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
575 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
577 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
580 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
581 return types for X86_WIN32.
582 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
583 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
584 _ffi_closure_STDCALL): Likewise.
586 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
587 (dlmmap, dlmunmap): Also use these functions on Cygwin.
589 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
594 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
597 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
599 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
600 to "" before adding paths. (This reinstates an assignment that
601 was removed by my 2009-06-30 commit, but changes the initial
602 value from "." to "".)
604 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
607 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
608 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
611 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
613 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
614 to ld_library_path. Use add_path. Add just find_libgcc_s
615 to ld_library_path, not every libgcc multilib directory.
617 2009-06-16 Wim Lewis <wiml@hhhh.org>
619 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
620 supposed to be callee-saved.
621 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
622 return buffer for odd-size structs.
624 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
627 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
628 allow_stack_execute for Darwin.
630 2009-06-16 Andrew Haley <aph@redhat.com>
632 * configure.ac (TARGETDIR): Add missing blank lines.
633 * configure: Regenerate.
635 2009-06-16 Andrew Haley <aph@redhat.com>
637 * testsuite/libffi.call/cls_align_sint64.c,
638 testsuite/libffi.call/cls_align_uint64.c,
639 testsuite/libffi.call/cls_longdouble_va.c,
640 testsuite/libffi.call/cls_ulonglong.c,
641 testsuite/libffi.call/return_ll1.c,
642 testsuite/libffi.call/stret_medium2.c: Fix printf format
644 * testsuite/libffi.call/ffitest.h,
645 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
647 2009-06-15 Andrew Haley <aph@redhat.com>
649 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
650 * testsuite/libffi.call/err_bad_abi.c: Likewise.
652 2009-06-12 Andrew Haley <aph@redhat.com>
654 * Makefile.am: Remove info_TEXINFOS.
656 2009-06-12 Andrew Haley <aph@redhat.com>
658 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
659 testsuite/libffi.call/cls_align_uint64.c,
660 testsuite/libffi.call/cls_ulonglong.c,
661 testsuite/libffi.call/return_ll1.c,
662 testsuite/libffi.call/stret_medium2.c: Fix printf format
664 testsuite/libffi.special/unwindtest.cc: include stdint.h.
666 2009-06-11 Timothy Wall <twall@users.sf.net>
671 include/ffi_common.h,
676 src/x86/win64.S (new),
677 README: Added win64 support (mingw or MSVC)
681 testsuite/Makefile.in,
683 aclocal.m4: Regenerated
684 * ltcf-c.sh: properly escape cygwin/w32 path
685 * man/ffi_call.3: Clarify size requirements for return value.
686 * src/x86/ffi64.c: Fix filename in comment.
687 * src/x86/win32.S: Remove unused extern.
689 * testsuite/libffi.call/closure_fn0.c,
690 testsuite/libffi.call/closure_fn1.c,
691 testsuite/libffi.call/closure_fn2.c,
692 testsuite/libffi.call/closure_fn3.c,
693 testsuite/libffi.call/closure_fn4.c,
694 testsuite/libffi.call/closure_fn5.c,
695 testsuite/libffi.call/closure_fn6.c,
696 testsuite/libffi.call/closure_stdcall.c,
697 testsuite/libffi.call/cls_12byte.c,
698 testsuite/libffi.call/cls_16byte.c,
699 testsuite/libffi.call/cls_18byte.c,
700 testsuite/libffi.call/cls_19byte.c,
701 testsuite/libffi.call/cls_1_1byte.c,
702 testsuite/libffi.call/cls_20byte.c,
703 testsuite/libffi.call/cls_20byte1.c,
704 testsuite/libffi.call/cls_24byte.c,
705 testsuite/libffi.call/cls_2byte.c,
706 testsuite/libffi.call/cls_3_1byte.c,
707 testsuite/libffi.call/cls_3byte1.c,
708 testsuite/libffi.call/cls_3byte2.c,
709 testsuite/libffi.call/cls_4_1byte.c,
710 testsuite/libffi.call/cls_4byte.c,
711 testsuite/libffi.call/cls_5_1_byte.c,
712 testsuite/libffi.call/cls_5byte.c,
713 testsuite/libffi.call/cls_64byte.c,
714 testsuite/libffi.call/cls_6_1_byte.c,
715 testsuite/libffi.call/cls_6byte.c,
716 testsuite/libffi.call/cls_7_1_byte.c,
717 testsuite/libffi.call/cls_7byte.c,
718 testsuite/libffi.call/cls_8byte.c,
719 testsuite/libffi.call/cls_9byte1.c,
720 testsuite/libffi.call/cls_9byte2.c,
721 testsuite/libffi.call/cls_align_double.c,
722 testsuite/libffi.call/cls_align_float.c,
723 testsuite/libffi.call/cls_align_longdouble.c,
724 testsuite/libffi.call/cls_align_longdouble_split.c,
725 testsuite/libffi.call/cls_align_longdouble_split2.c,
726 testsuite/libffi.call/cls_align_pointer.c,
727 testsuite/libffi.call/cls_align_sint16.c,
728 testsuite/libffi.call/cls_align_sint32.c,
729 testsuite/libffi.call/cls_align_sint64.c,
730 testsuite/libffi.call/cls_align_uint16.c,
731 testsuite/libffi.call/cls_align_uint32.c,
732 testsuite/libffi.call/cls_align_uint64.c,
733 testsuite/libffi.call/cls_dbls_struct.c,
734 testsuite/libffi.call/cls_double.c,
735 testsuite/libffi.call/cls_double_va.c,
736 testsuite/libffi.call/cls_float.c,
737 testsuite/libffi.call/cls_longdouble.c,
738 testsuite/libffi.call/cls_longdouble_va.c,
739 testsuite/libffi.call/cls_multi_schar.c,
740 testsuite/libffi.call/cls_multi_sshort.c,
741 testsuite/libffi.call/cls_multi_sshortchar.c,
742 testsuite/libffi.call/cls_multi_uchar.c,
743 testsuite/libffi.call/cls_multi_ushort.c,
744 testsuite/libffi.call/cls_multi_ushortchar.c,
745 testsuite/libffi.call/cls_pointer.c,
746 testsuite/libffi.call/cls_pointer_stack.c,
747 testsuite/libffi.call/cls_schar.c,
748 testsuite/libffi.call/cls_sint.c,
749 testsuite/libffi.call/cls_sshort.c,
750 testsuite/libffi.call/cls_uchar.c,
751 testsuite/libffi.call/cls_uint.c,
752 testsuite/libffi.call/cls_ulonglong.c,
753 testsuite/libffi.call/cls_ushort.c,
754 testsuite/libffi.call/err_bad_abi.c,
755 testsuite/libffi.call/err_bad_typedef.c,
756 testsuite/libffi.call/float2.c,
757 testsuite/libffi.call/huge_struct.c,
758 testsuite/libffi.call/nested_struct.c,
759 testsuite/libffi.call/nested_struct1.c,
760 testsuite/libffi.call/nested_struct10.c,
761 testsuite/libffi.call/nested_struct2.c,
762 testsuite/libffi.call/nested_struct3.c,
763 testsuite/libffi.call/nested_struct4.c,
764 testsuite/libffi.call/nested_struct5.c,
765 testsuite/libffi.call/nested_struct6.c,
766 testsuite/libffi.call/nested_struct7.c,
767 testsuite/libffi.call/nested_struct8.c,
768 testsuite/libffi.call/nested_struct9.c,
769 testsuite/libffi.call/problem1.c,
770 testsuite/libffi.call/return_ldl.c,
771 testsuite/libffi.call/return_ll1.c,
772 testsuite/libffi.call/stret_large.c,
773 testsuite/libffi.call/stret_large2.c,
774 testsuite/libffi.call/stret_medium.c,
775 testsuite/libffi.call/stret_medium2.c,
776 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
777 of checking for MMAP. Use intptr_t instead of long casts.
779 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
781 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
782 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
783 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
785 2009-06-09 Andrew Haley <aph@redhat.com>
787 * src/x86/freebsd.S: Add missing file.
789 2009-06-08 Andrew Haley <aph@redhat.com>
791 Import from libffi 3.0.8:
793 * doc/libffi.texi: New file.
794 * doc/libffi.info: Likewise.
795 * doc/stamp-vti: Likewise.
796 * man/Makefile.am: New file.
797 * man/ffi_call.3: New file.
799 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
801 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
803 * configure.ac: Bump version to 3.0.8.
804 parisc*-*-linux*: Add.
805 i386-*-freebsd* | i386-*-openbsd*: Add.
806 powerpc-*-beos*: Add.
807 AM_CONDITIONAL X86_FREEBSD: Add.
808 AC_CONFIG_FILES: Add man/Makefile.
810 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
812 2009-06-08 Andrew Haley <aph@redhat.com>
814 * README: Import from libffi 3.0.8.
816 2009-06-08 Andrew Haley <aph@redhat.com>
818 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
819 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
820 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
821 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
823 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
824 * testsuite/libffi.call/stret_medium.c: Likewise.
825 * testsuite/libffi.call/stret_large2.c: Likewise.
826 * testsuite/libffi.call/stret_large.c: Likewise.
828 2008-12-26 Timothy Wall <twall@users.sf.net>
830 * testsuite/libffi.call/cls_longdouble.c,
831 testsuite/libffi.call/cls_longdouble_va.c,
832 testsuite/libffi.call/cls_align_longdouble.c,
833 testsuite/libffi.call/cls_align_longdouble_split.c,
834 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
835 failures on x86_64 cygwin/mingw.
837 2008-12-22 Timothy Wall <twall@users.sf.net>
839 * testsuite/libffi.call/closure_fn0.c,
840 testsuite/libffi.call/closure_fn1.c,
841 testsuite/libffi.call/closure_fn2.c,
842 testsuite/libffi.call/closure_fn3.c,
843 testsuite/libffi.call/closure_fn4.c,
844 testsuite/libffi.call/closure_fn5.c,
845 testsuite/libffi.call/closure_fn6.c,
846 testsuite/libffi.call/closure_loc_fn0.c,
847 testsuite/libffi.call/closure_stdcall.c,
848 testsuite/libffi.call/cls_align_pointer.c,
849 testsuite/libffi.call/cls_pointer.c,
850 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
851 pointer to integer (intptr_t).
852 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
854 2008-07-24 Anthony Green <green@redhat.com>
856 * testsuite/libffi.call/cls_dbls_struct.c,
857 testsuite/libffi.call/cls_double_va.c,
858 testsuite/libffi.call/cls_longdouble.c,
859 testsuite/libffi.call/cls_longdouble_va.c,
860 testsuite/libffi.call/cls_pointer.c,
861 testsuite/libffi.call/cls_pointer_stack.c,
862 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
865 2008-03-04 Anthony Green <green@redhat.com>
869 * testsuite/libffi.call/cls_align_longdouble_split2.c
870 testsuite/libffi.call/cls_align_longdouble_split.c
871 testsuite/libffi.call/cls_dbls_struct.c
872 testsuite/libffi.call/cls_double_va.c
873 testsuite/libffi.call/cls_longdouble.c
874 testsuite/libffi.call/cls_longdouble_va.c
875 testsuite/libffi.call/cls_pointer.c
876 testsuite/libffi.call/cls_pointer_stack.c
877 testsuite/libffi.call/err_bad_abi.c
878 testsuite/libffi.call/err_bad_typedef.c
879 testsuite/libffi.call/stret_large2.c
880 testsuite/libffi.call/stret_large.c
881 testsuite/libffi.call/stret_medium2.c
882 testsuite/libffi.call/stret_medium.c: New tests from Apple.
884 2009-06-05 Andrew Haley <aph@redhat.com>
886 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
889 2009-06-04 Andrew Haley <aph@redhat.com>
891 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
894 2008-02-26 Anthony Green <green@redhat.com>
895 Thomas Heller <theller@ctypes.org>
897 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
900 2008-02-03 Timothy Wall <twall@users.sf.net>
902 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
903 offset based on code pointer, not data pointer.
905 2008-01-31 Timothy Wall <twall@users.sf.net>
907 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
909 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
911 * src/x86/win32.S: Add assembly for stdcall closure.
912 * src/x86/ffi.c: Initialize stdcall closure trampoline.
914 2009-06-04 Andrew Haley <aph@redhat.com>
916 * include/ffi.h.in: Change void (*)() to void (*)(void).
917 * src/x86/ffi.c: Likewise.
919 2009-06-04 Andrew Haley <aph@redhat.com>
921 * src/powerpc/ppc_closure.S: Insert licence header.
922 * src/powerpc/linux64_closure.S: Likewise.
923 * src/m68k/sysv.S: Likewise.
925 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
926 * src/powerpc/ffi.c: Likewise.
927 * src/powerpc/ffi_darwin.c: Likewise.
928 * src/m32r/ffi.c: Likewise.
929 * src/sh64/ffi.c: Likewise.
930 * src/x86/ffi64.c: Likewise.
931 * src/alpha/ffi.c: Likewise.
932 * src/alpha/osf.S: Likewise.
933 * src/frv/ffi.c: Likewise.
934 * src/s390/ffi.c: Likewise.
935 * src/pa/ffi.c: Likewise.
936 * src/pa/hpux32.S: Likewise.
937 * src/ia64/unix.S: Likewise.
938 * src/ia64/ffi.c: Likewise.
939 * src/sparc/ffi.c: Likewise.
940 * src/mips/ffi.c: Likewise.
941 * src/sh/ffi.c: Likewise.
943 2008-02-15 David Daney <ddaney@avtrex.com>
945 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
946 Define (conditionally), and use it to include cachectl.h.
947 (ffi_prep_closure_loc): Fix cache flushing.
948 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
950 2009-06-04 Andrew Haley <aph@redhat.com>
956 src/powerpc/ffitarget.h,
958 src/sh64/ffitarget.h,
969 src/alpha/ffitarget.h,
972 src/m68k/ffitarget.h,
975 src/s390/ffitarget.h,
977 src/cris/ffitarget.h,
982 src/ia64/ffitarget.h,
985 src/ia64/ia64_flags.h,
989 src/sparc/ffitarget.h,
992 src/mips/ffitarget.h,
999 src/sh/sysv.S: Update license text.
1001 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
1003 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
1004 (.eh_frame): Add FDE for it.
1006 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
1008 * configure.ac: Also check if assembler supports pc-relative
1009 relocs on X86_WIN32 targets.
1010 * configure: Regenerate.
1011 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
1012 (_ffi_call_SYSV): Add missing function type symbol .def and
1013 add EH markup labels.
1014 (_ffi_call_STDCALL): Likewise.
1015 (_ffi_closure_SYSV): Likewise.
1016 (_ffi_closure_raw_SYSV): Likewise.
1017 (.eh_frame): Add hand-crafted EH data.
1019 2009-04-09 Jakub Jelinek <jakub@redhat.com>
1021 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
1022 version 3 of the GNU General Public License and to point readers
1023 at the COPYING3 file and the FSF's license web page.
1024 * testsuite/libffi.call/call.exp: Likewise.
1025 * testsuite/libffi.special/special.exp: Likewise.
1027 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1029 * configure: Regenerate.
1031 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1034 * configure.ac (HAVE_AS_X86_PCREL): New test.
1035 * configure: Regenerate.
1036 * fficonfig.h.in: Regenerate.
1037 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
1038 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
1039 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
1040 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
1041 * src/x86/unix64.S (.Lstore_table): Move to .text section.
1042 (.Lload_table): Likewise.
1043 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
1045 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1047 * configure: Regenerate.
1049 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
1051 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
1052 signed/unsigned int8/16 return values.
1053 * src/sparc/v8.S (ffi_call_v8): Likewise.
1054 (ffi_closure_v8): Likewise.
1056 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
1057 Steve Ellcey <sje@cup.hp.com>
1059 * configure: Regenerate for new libtool.
1060 * Makefile.in: Ditto.
1061 * include/Makefile.in: Ditto.
1062 * aclocal.m4: Ditto.
1064 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
1066 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
1067 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
1068 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
1069 Adjust copyright notice.
1070 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
1071 register or two register to use for FFI_SYSV structs.
1072 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
1073 (ffi_closure_helper_SYSV): Fix the return type for
1074 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
1075 Adjust copyright notice.
1077 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
1079 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
1082 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1084 * configure: Regenerate.
1085 * include/Makefile.in: Regenerate.
1086 * testsuite/Makefile.in: Regenerate.
1088 2008-06-07 Joseph Myers <joseph@codesourcery.com>
1090 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
1091 powerpc-*-beos*): Remove.
1092 * configure: Regenerate.
1094 2008-05-09 Julian Brown <julian@codesourcery.com>
1096 * Makefile.am (LTLDFLAGS): New.
1097 (libffi_la_LDFLAGS): Use above.
1098 * Makefile.in: Regenerate.
1100 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
1103 * aclocal.m4: Regenerate.
1104 * configure: Regenerate.
1106 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
1108 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
1109 * src/sh64/sysv.S: Likewise.
1111 2008-03-26 Daniel Jacobowitz <dan@debian.org>
1113 * src/arm/sysv.S: Fix ARM comment marker.
1115 2008-03-26 Jakub Jelinek <jakub@redhat.com>
1117 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
1118 * src/s390/sysv.S: Likewise.
1119 * src/powerpc/ppc_closure.S: Likewise.
1120 * src/powerpc/sysv.S: Likewise.
1121 * src/x86/unix64.S: Likewise.
1122 * src/x86/sysv.S: Likewise.
1123 * src/sparc/v8.S: Likewise.
1124 * src/sparc/v9.S: Likewise.
1125 * src/m68k/sysv.S: Likewise.
1126 * src/arm/sysv.S: Likewise.
1128 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1130 * aclocal.m4: Regenerate.
1131 * configure: Likewise.
1132 * Makefile.in: Likewise.
1133 * include/Makefile.in: Likewise.
1134 * testsuite/Makefile.in: Likewise.
1136 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
1137 Andreas Tobler <a.tobler@schweiz.org>
1139 * configure.ac: Add amd64-*-freebsd* target.
1140 * configure: Regenerate.
1142 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
1145 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
1148 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
1151 2008-01-24 David Edelsohn <edelsohn@gnu.org>
1153 * configure: Regenerate.
1155 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
1157 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
1159 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
1162 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
1163 signed/unsigned int8/16 for X86_DARWIN.
1164 Updated copyright info.
1165 Handle one and two byte structs with special cif->flags.
1166 * src/x86/ffitarget.h: Add special types for one and two byte structs.
1167 Updated copyright info.
1168 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
1170 Remove code to pop args from the stack after call.
1171 Special-case signed/unsigned for int8/16, one and two byte structs.
1172 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1173 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1175 Updated copyright info.
1177 2007-12-08 David Daney <ddaney@avtrex.com>
1179 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
1180 SUBU, add with ADDU and use smaller code sequences.
1182 2007-12-07 David Daney <ddaney@avtrex.com>
1184 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
1187 2007-12-06 David Daney <ddaney@avtrex.com>
1189 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
1191 (ffi_java_raw): New typedef.
1192 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
1193 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
1195 (ffi_java_raw_closure) : Same.
1196 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
1198 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
1199 FFI_SIZEOF_JAVA_RAW.
1200 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
1201 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
1202 sizeof(ffi_java_raw) for alignment calculations.
1203 (ffi_java_ptrarray_to_raw): Same.
1204 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
1205 if FFI_SIZEOF_JAVA_RAW == 4.
1206 (ffi_java_raw_to_rvalue): Same.
1207 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
1208 (ffi_java_translate_args): Same.
1209 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
1211 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
1213 2007-12-06 David Daney <ddaney@avtrex.com>
1215 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
1218 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
1221 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
1222 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
1223 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
1224 set the NUM_FPR_ARG_REGISTERS according to.
1225 Add support for potential soft-float support under hard-float
1227 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
1228 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
1229 to the FFI_LINUX_SOFT_FLOAT ABI.
1230 (ffi_prep_cif_machdep): Likewise.
1231 (ffi_closure_helper_SYSV): Likewise.
1232 * src/powerpc/ppc_closure.S: Make sure not to store float/double
1233 on archs where __NO_FPRS__ is true.
1234 Add FFI_TYPE_UINT128 support.
1235 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
1236 Adjust copyright notice.
1238 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
1240 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
1241 * include/ffi_common.h: ... here.
1244 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
1246 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
1247 * src/powerpc/linux64.S: Likewise.
1248 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
1249 * testsuite/libffi.call/return_ldl.c: New test.
1251 2007-09-04 <aph@redhat.com>
1253 * src/arm/sysv.S (UNWIND): New.
1254 (Whole file): Conditionally compile unwinder directives.
1255 * src/arm/sysv.S: Add unwinder directives.
1257 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
1258 Only treat r0 as a struct address if we're actually returning a
1260 Only copy the bytes that are actually within a struct.
1261 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
1262 is returned in r0, not passed by address.
1263 (ffi_call): Allocate a word-sized temporary for the case where
1264 a composite is returned in r0.
1265 (ffi_prep_incoming_args_SYSV): Align as necessary.
1267 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
1269 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
1270 directly using the sys_cacheflush syscall.
1272 2007-07-27 Andrew Haley <aph@redhat.com>
1274 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
1276 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
1278 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
1279 * configure.ac: Likewise.
1280 * Makefile.in: Regenerate.
1281 * include/Makefile.in: Likewise.
1282 * testsuite/Makefile.in: Likewise.
1283 * configure: Likewise.
1285 2007-08-24 David Daney <ddaney@avtrex.com>
1287 * testsuite/libffi.call/return_sl.c: New test.
1289 2007-08-10 David Daney <ddaney@avtrex.com>
1291 * testsuite/libffi.call/cls_multi_ushort.c,
1292 testsuite/libffi.call/cls_align_uint16.c,
1293 testsuite/libffi.call/nested_struct1.c,
1294 testsuite/libffi.call/nested_struct3.c,
1295 testsuite/libffi.call/cls_7_1_byte.c,
1296 testsuite/libffi.call/nested_struct5.c,
1297 testsuite/libffi.call/cls_double.c,
1298 testsuite/libffi.call/nested_struct7.c,
1299 testsuite/libffi.call/cls_sint.c,
1300 testsuite/libffi.call/nested_struct9.c,
1301 testsuite/libffi.call/cls_20byte1.c,
1302 testsuite/libffi.call/cls_multi_sshortchar.c,
1303 testsuite/libffi.call/cls_align_sint64.c,
1304 testsuite/libffi.call/cls_3byte2.c,
1305 testsuite/libffi.call/cls_multi_schar.c,
1306 testsuite/libffi.call/cls_multi_uchar.c,
1307 testsuite/libffi.call/cls_19byte.c,
1308 testsuite/libffi.call/cls_9byte1.c,
1309 testsuite/libffi.call/cls_align_float.c,
1310 testsuite/libffi.call/closure_fn1.c,
1311 testsuite/libffi.call/problem1.c,
1312 testsuite/libffi.call/closure_fn3.c,
1313 testsuite/libffi.call/cls_sshort.c,
1314 testsuite/libffi.call/closure_fn5.c,
1315 testsuite/libffi.call/cls_align_double.c,
1316 testsuite/libffi.call/nested_struct.c,
1317 testsuite/libffi.call/cls_2byte.c,
1318 testsuite/libffi.call/nested_struct10.c,
1319 testsuite/libffi.call/cls_4byte.c,
1320 testsuite/libffi.call/cls_6byte.c,
1321 testsuite/libffi.call/cls_8byte.c,
1322 testsuite/libffi.call/cls_multi_sshort.c,
1323 testsuite/libffi.call/cls_align_sint16.c,
1324 testsuite/libffi.call/cls_align_uint32.c,
1325 testsuite/libffi.call/cls_20byte.c,
1326 testsuite/libffi.call/cls_float.c,
1327 testsuite/libffi.call/nested_struct2.c,
1328 testsuite/libffi.call/cls_5_1_byte.c,
1329 testsuite/libffi.call/nested_struct4.c,
1330 testsuite/libffi.call/cls_24byte.c,
1331 testsuite/libffi.call/nested_struct6.c,
1332 testsuite/libffi.call/cls_64byte.c,
1333 testsuite/libffi.call/nested_struct8.c,
1334 testsuite/libffi.call/cls_uint.c,
1335 testsuite/libffi.call/cls_multi_ushortchar.c,
1336 testsuite/libffi.call/cls_schar.c,
1337 testsuite/libffi.call/cls_uchar.c,
1338 testsuite/libffi.call/cls_align_uint64.c,
1339 testsuite/libffi.call/cls_ulonglong.c,
1340 testsuite/libffi.call/cls_align_longdouble.c,
1341 testsuite/libffi.call/cls_1_1byte.c,
1342 testsuite/libffi.call/cls_12byte.c,
1343 testsuite/libffi.call/cls_3_1byte.c,
1344 testsuite/libffi.call/cls_3byte1.c,
1345 testsuite/libffi.call/cls_4_1byte.c,
1346 testsuite/libffi.call/cls_6_1_byte.c,
1347 testsuite/libffi.call/cls_16byte.c,
1348 testsuite/libffi.call/cls_18byte.c,
1349 testsuite/libffi.call/closure_fn0.c,
1350 testsuite/libffi.call/cls_9byte2.c,
1351 testsuite/libffi.call/closure_fn2.c,
1352 testsuite/libffi.call/closure_fn4.c,
1353 testsuite/libffi.call/cls_ushort.c,
1354 testsuite/libffi.call/closure_fn6.c,
1355 testsuite/libffi.call/cls_5byte.c,
1356 testsuite/libffi.call/cls_align_pointer.c,
1357 testsuite/libffi.call/cls_7byte.c,
1358 testsuite/libffi.call/cls_align_sint32.c,
1359 testsuite/libffi.special/unwindtest_ffi_call.cc,
1360 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
1362 2007-08-10 David Daney <ddaney@avtrex.com>
1365 * configure.ac: Don't treat mips64 as a special case.
1366 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
1367 * configure: Regenerate
1368 * Makefile.in: Ditto.
1369 * fficonfig.h.in: Ditto.
1370 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
1371 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
1372 (FFI_DEFAULT_ABI): Set for n64 case.
1373 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
1374 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
1375 (ffi_closure_N32): New function.
1376 (.eh_frame): New section
1377 * src/mips/o32.S: Clean up comments.
1378 (ffi_closure_O32): Pass ffi_closure parameter in $12.
1379 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
1380 _MIPS_SIM == _ABIN32 throughout.
1381 (FFI_MIPS_STOP_HERE): New, use in place of
1383 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1384 to support n32/n64 ABIs.
1385 (calc_n32_struct_flags): Rewrite.
1386 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1387 position of flag bits.
1388 (ffi_prep_cif_machdep): Rewrite n32 portion.
1389 (ffi_call): Enable for n64. Add special handling for small structure
1391 (ffi_prep_closure_loc): Add n32 and n64 support.
1392 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1393 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1395 2007-08-08 David Daney <ddaney@avtrex.com>
1397 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1398 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1400 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1401 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1402 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1403 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1404 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1405 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1406 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1407 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1408 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1409 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1410 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1411 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1412 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1413 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1414 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1415 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1416 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1417 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1418 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1419 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1420 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1421 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1422 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1423 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1424 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1425 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1426 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1427 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1428 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1429 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1430 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1432 2007-08-07 Andrew Haley <aph@redhat.com>
1434 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1437 2007-08-06 Andrew Haley <aph@redhat.com>
1440 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1441 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1444 2007-08-02 David Daney <ddaney@avtrex.com>
1446 * testsuite/libffi.call/return_ul.c (main): Define return type as
1447 ffi_arg. Use proper printf conversion specifier.
1449 2007-07-30 Andrew Haley <aph@redhat.com>
1452 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1453 signed/unsigned int8/16.
1454 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1456 Remove code to pop args from the stack after call.
1457 Special-case signed/unsigned int8/16.
1458 * testsuite/libffi.call/return_sc.c (main): Revert.
1460 2007-07-26 Richard Guenther <rguenther@suse.de>
1463 * testsuite/libffi.call/return_sc.c (main): Verify call
1464 result as signed char, not ffi_arg.
1466 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1468 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1469 * configure: Regenerate.
1471 2007-07-11 David Daney <ddaney@avtrex.com>
1473 * src/mips/ffi.c: Don't include sys/cachectl.h.
1474 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1477 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1479 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1480 from (ffi_prep_closure): ... this.
1481 (FFI_INIT_TRAMPOLINE): Adjust.
1483 2005-12-31 Phil Blundell <pb@reciva.com>
1485 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1486 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1487 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1488 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1489 (FFI_CLOSURES): Enable closure support.
1491 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1493 * testsuite/libffi.call/cls_multi_ushort.c,
1494 testsuite/libffi.call/cls_align_uint16.c,
1495 testsuite/libffi.call/nested_struct1.c,
1496 testsuite/libffi.call/nested_struct3.c,
1497 testsuite/libffi.call/cls_7_1_byte.c,
1498 testsuite/libffi.call/cls_double.c,
1499 testsuite/libffi.call/nested_struct5.c,
1500 testsuite/libffi.call/nested_struct7.c,
1501 testsuite/libffi.call/cls_sint.c,
1502 testsuite/libffi.call/nested_struct9.c,
1503 testsuite/libffi.call/cls_20byte1.c,
1504 testsuite/libffi.call/cls_multi_sshortchar.c,
1505 testsuite/libffi.call/cls_align_sint64.c,
1506 testsuite/libffi.call/cls_3byte2.c,
1507 testsuite/libffi.call/cls_multi_schar.c,
1508 testsuite/libffi.call/cls_multi_uchar.c,
1509 testsuite/libffi.call/cls_19byte.c,
1510 testsuite/libffi.call/cls_9byte1.c,
1511 testsuite/libffi.call/cls_align_float.c,
1512 testsuite/libffi.call/closure_fn1.c,
1513 testsuite/libffi.call/problem1.c,
1514 testsuite/libffi.call/closure_fn3.c,
1515 testsuite/libffi.call/cls_sshort.c,
1516 testsuite/libffi.call/closure_fn5.c,
1517 testsuite/libffi.call/cls_align_double.c,
1518 testsuite/libffi.call/cls_2byte.c,
1519 testsuite/libffi.call/nested_struct.c,
1520 testsuite/libffi.call/nested_struct10.c,
1521 testsuite/libffi.call/cls_4byte.c,
1522 testsuite/libffi.call/cls_6byte.c,
1523 testsuite/libffi.call/cls_8byte.c,
1524 testsuite/libffi.call/cls_multi_sshort.c,
1525 testsuite/libffi.call/cls_align_uint32.c,
1526 testsuite/libffi.call/cls_align_sint16.c,
1527 testsuite/libffi.call/cls_float.c,
1528 testsuite/libffi.call/cls_20byte.c,
1529 testsuite/libffi.call/cls_5_1_byte.c,
1530 testsuite/libffi.call/nested_struct2.c,
1531 testsuite/libffi.call/cls_24byte.c,
1532 testsuite/libffi.call/nested_struct4.c,
1533 testsuite/libffi.call/nested_struct6.c,
1534 testsuite/libffi.call/cls_64byte.c,
1535 testsuite/libffi.call/nested_struct8.c,
1536 testsuite/libffi.call/cls_uint.c,
1537 testsuite/libffi.call/cls_multi_ushortchar.c,
1538 testsuite/libffi.call/cls_schar.c,
1539 testsuite/libffi.call/cls_uchar.c,
1540 testsuite/libffi.call/cls_align_uint64.c,
1541 testsuite/libffi.call/cls_ulonglong.c,
1542 testsuite/libffi.call/cls_align_longdouble.c,
1543 testsuite/libffi.call/cls_1_1byte.c,
1544 testsuite/libffi.call/cls_12byte.c,
1545 testsuite/libffi.call/cls_3_1byte.c,
1546 testsuite/libffi.call/cls_3byte1.c,
1547 testsuite/libffi.call/cls_4_1byte.c,
1548 testsuite/libffi.call/cls_6_1_byte.c,
1549 testsuite/libffi.call/cls_16byte.c,
1550 testsuite/libffi.call/cls_18byte.c,
1551 testsuite/libffi.call/closure_fn0.c,
1552 testsuite/libffi.call/cls_9byte2.c,
1553 testsuite/libffi.call/closure_fn2.c,
1554 testsuite/libffi.call/closure_fn4.c,
1555 testsuite/libffi.call/cls_ushort.c,
1556 testsuite/libffi.call/closure_fn6.c,
1557 testsuite/libffi.call/cls_5byte.c,
1558 testsuite/libffi.call/cls_align_pointer.c,
1559 testsuite/libffi.call/cls_7byte.c,
1560 testsuite/libffi.call/cls_align_sint32.c,
1561 testsuite/libffi.special/unwindtest_ffi_call.cc,
1562 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1564 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1566 * aclocal.m4: Regenerated.
1568 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1570 * configure: Regenerate.
1572 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1574 * Makefile.in: Regenerate.
1575 * configure: Regenerate.
1576 * aclocal.m4: Regenerate.
1577 * include/Makefile.in: Regenerate.
1578 * testsuite/Makefile.in: Regenerate.
1580 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1582 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1583 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1584 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1585 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1586 (FFI_CLOSURES): Enable closure support.
1588 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1590 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1591 * configure: Regenerate.
1592 * fficonfig.h.in: Regenerate.
1593 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1594 CFI_OFFSET,CFI_DEF_CFA): New macros.
1595 (ffi_call_SYSV): Add callframe annotation.
1597 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1599 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1600 numerous test suite failures.
1601 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1603 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1605 * Makefile.am (EXTRA_DIST): Bring up to date.
1606 * Makefile.in: Regenerate.
1607 * src/frv/eabi.S: Remove RCS keyword.
1609 2007-04-06 Richard Henderson <rth@redhat.com>
1611 * configure.ac: Tidy target case.
1612 (HAVE_LONG_DOUBLE): Allow the target to override.
1613 * configure: Regenerate.
1614 * include/ffi.h.in: Don't define ffi_type_foo if
1615 LIBFFI_HIDE_BASIC_TYPES is defined.
1616 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1618 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1619 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1620 (ffi_type_longdouble): Special case for Alpha. Don't define
1621 if long double == double.
1623 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1624 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1625 (ffi_call, ffi_closure_osf_inner): Likewise.
1626 (ffi_closure_osf_inner): Likewise. Mark hidden.
1627 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1628 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1629 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1630 (load_table): Handle 128-bit long double.
1632 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1634 2007-04-06 Tom Tromey <tromey@redhat.com>
1637 * README: Fixed bug in example.
1639 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1641 * src/closures.c: Include sys/statfs.h.
1642 (_GNU_SOURCE): Define on Linux.
1643 (FFI_MMAP_EXEC_SELINUX): Define.
1644 (selinux_enabled): New variable.
1645 (selinux_enabled_check): New function.
1646 (is_selinux_enabled): Define.
1649 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1651 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1652 Use 'volatile float sum' to create sum of floats to avoid false
1653 negative due to excess precision on ix86 targets.
1656 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1658 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1660 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1662 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1664 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1665 (ffi_prep_closure_loc): New.
1666 (ffi_prep_raw_closure_loc): New.
1667 (ffi_prep_java_raw_closure_loc): New.
1668 * src/closures.c: New file.
1669 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1670 Replace sflags with exec_offset.
1671 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1672 sub_segment_exec_offset): New macros.
1673 (get_segment_flags, set_segment_flags, check_segment_merge): New
1675 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1676 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1677 destroy_mspace): Use new macros.
1678 (sys_alloc): Silence warning.
1679 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1680 * Makefile.in: Rebuilt.
1681 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1682 terms of ffi_prep_closure_loc.
1683 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1685 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1687 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1689 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1690 the renamed version.
1691 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1692 (ffi_prep_closure): ... this.
1693 * src/pa/ffi.c: Likewise.
1694 * src/cris/ffi.c: Likewise. Adjust.
1695 * src/frv/ffi.c: Likewise.
1696 * src/ia64/ffi.c: Likewise.
1697 * src/mips/ffi.c: Likewise.
1698 * src/powerpc/ffi_darwin.c: Likewise.
1699 * src/s390/ffi.c: Likewise.
1700 * src/sh/ffi.c: Likewise.
1701 * src/sh64/ffi.c: Likewise.
1702 * src/sparc/ffi.c: Likewise.
1703 * src/x86/ffi64.c: Likewise.
1704 * src/x86/ffi.c: Likewise.
1705 (FFI_INIT_TRAMPOLINE): Adjust.
1706 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1707 (ffi_prep_raw_closure): ... this.
1708 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1709 (ffi_prep_closure): ... this.
1710 (flush_icache): Adjust.
1712 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1714 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1717 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1719 * Makefile.am: Add dummy install-pdf target.
1720 * Makefile.in: Regenerate
1722 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1724 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1725 ffi_closure_helper_SYSV): Add long double handling.
1727 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1729 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1730 immediately after bctrl instruction.
1732 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1734 * Makefile.am (all-recursive, install-recursive,
1735 mostlyclean-recursive, clean-recursive, distclean-recursive,
1736 maintainer-clean-recursive): Add missing targets.
1737 * Makefile.in: Rebuilt.
1739 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1741 * configure.ac: Add TARGET for x86_64-*-darwin*.
1742 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1744 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1745 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1746 * configure: Regenerate.
1747 * Makefile.in: Regenerate.
1748 * include/Makefile.in: Regenerate.
1749 * testsuite/Makefile.in: Regenerate.
1750 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1753 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1755 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1758 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1760 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1761 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1763 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1764 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1766 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1767 Sandro Tolaini <tolaini@libero.it>
1769 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1771 * configure: Regenerated.
1772 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1773 (EXTRA_DIST): Add src/x86/darwin.S.
1774 * Makefile.in: Regenerated.
1775 * include/Makefile.in: Regenerated.
1776 * testsuite/Makefile.in: Regenerated.
1778 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1779 X86_WIN32, and additionally align stack to 16 bytes.
1780 * src/x86/darwin.S: New, based on sysv.S.
1781 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1783 2006-09-12 David Daney <ddaney@avtrex.com>
1786 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1787 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1788 * aclocal.m4: Regenerated for automake 1.9.6.
1789 * Makefile.in: Regenerated.
1790 * include/Makefile.in: Regenerated.
1791 * testsuite/Makefile.in: Regenerated.
1793 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1795 * include/ffi_common.h (struct): Revert accidental commit.
1797 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1799 * include/ffi_common.h: Remove lint directives.
1800 * include/ffi.h.in: Likewise.
1802 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1804 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1805 for 32-bit architectures.
1806 * testsuite/libffi.call/return_ul.c: New test case.
1808 2006-07-19 David Daney <ddaney@avtrex.com>
1810 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1811 xfail remains for mips64.
1813 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1815 * Makefile.am: Add install-html target. Add install-html to .PHONY
1816 * Makefile.in: Regenerate.
1817 * aclocal.m4: Regenerate.
1818 * include/Makefile.in: Regenerate.
1819 * testsuite/Makefile.in: Regenerate.
1821 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1823 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1826 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1828 * README: Remove notice about 'Crazy Comments'.
1829 * src/debug.c: Remove lint directives. Cleanup white spaces.
1830 * src/java_raw_api.c: Likewise.
1831 * src/prep_cif.c: Likewise.
1832 * src/raw_api.c: Likewise.
1833 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1835 * src/arm/ffi.c: Remove lint directives.
1836 * src/m32r/ffi.c: Likewise.
1837 * src/pa/ffi.c: Likewise.
1838 * src/powerpc/ffi.c: Likewise.
1839 * src/powerpc/ffi_darwin.c: Likewise.
1840 * src/sh/ffi.c: Likewise.
1841 * src/sh64/ffi.c: Likewise.
1842 * src/x86/ffi.c: Likewise.
1843 * testsuite/libffi.call/float2.c: Likewise.
1844 * testsuite/libffi.call/promotion.c: Likewise.
1845 * testsuite/libffi.call/struct1.c: Likewise.
1847 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1849 * src/pa/hpux32.S: Correct unwind offset calculation for
1851 * src/pa/linux.S: Likewise.
1853 2006-04-12 James E Wilson <wilson@specifix.com>
1856 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1857 (hfa_type_load): Call stf_spill.
1858 (hfa_type_store): Call ldf_fill.
1859 (ffi_call): Adjust calls to above routines. Add local temps for
1862 2006-04-10 Matthias Klose <doko@debian.org>
1864 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1865 directory names containing underscores.
1867 2006-04-07 James E Wilson <wilson@specifix.com>
1869 * testsuite/libffi.call/float4.c: New testcase.
1871 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1872 Andreas Tobler <a.tobler@schweiz.ch>
1874 * Makefile.am: Add PA_HPUX port.
1875 * Makefile.in: Regenerate.
1876 * include/Makefile.in: Likewise.
1877 * testsuite/Makefile.in: Likewise.
1878 * configure.ac: Add PA_HPUX rules.
1879 * configure: Regenerate.
1880 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1881 Add PA_HPUX and PA64_HPUX.
1882 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1883 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1884 (FFI_TYPE_SMALL_STRUCT2): Define.
1885 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1886 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1887 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1888 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1889 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1890 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1891 * src/pa/ffi.c (ROUND_DOWN): Delete.
1892 (fldw, fstw, fldd, fstd): Use '__asm__'.
1893 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1894 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1895 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1896 Simplify incrementing of stack slot variable. Change type of local
1897 'n' to unsigned int.
1898 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1900 (ffi_prep_cif_machdep): Likewise.
1901 (ffi_call): Likewise.
1902 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1903 return type to ffi_status. Simplify incrementing of stack slot
1904 variable. Only copy floating point argument registers when PA_LINUX
1905 is true. Reformat debug statement.
1906 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1907 FFI_TYPE_SMALL_STRUCT8.
1908 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1910 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1911 Add nops to cache flush. Add trampoline for PA_HPUX.
1912 * src/pa/hpux32.S: New file.
1913 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1914 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1915 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1916 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1917 argument type checks so that common argument types appear first.
1918 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1919 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1921 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1923 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1924 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1925 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1926 FFI_TYPE_LONGDOUBLE.
1927 (ffi_prep_args64): Assert using IBM extended double.
1928 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1929 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1930 (ffi_call): Handle FFI_LINUX.
1931 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1932 gpr3 return pointer as for struct return. Handle FFI_LINUX
1933 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1935 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1936 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1937 Don't use r6 as pointer to results, instead use sp offset. Don't
1938 make a special call to load lr with case table address, instead
1939 use offset from previous call.
1940 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1941 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1944 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1946 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1947 passed with FP registers correctly.
1948 (ffi_closure_helper_SYSV): Likewise.
1949 * src/sh64/sysv.S: Likewise.
1951 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1953 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1954 args and userdata unused.
1955 (closure_test_fn1): Mark cif and userdata unused.
1956 (main): Remove unused res.
1958 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1960 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1961 -O2, -O3, -Os and the warning flags -W -Wall.
1962 * testsuite/libffi.special/special.exp: Likewise.
1963 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1964 unused parameter unused for gcc or else do nothing.
1965 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1966 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1967 and userdata unused.
1968 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1969 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1970 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1971 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1972 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1973 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1974 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1975 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1976 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1977 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1978 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1979 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1980 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1981 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1982 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1983 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1984 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1985 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1986 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1987 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1988 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1989 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1990 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1991 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1993 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1995 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1997 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1998 void* to avoid compiler warning.
2000 (cls_struct_align_gn): Mark cif and userdata unused.
2001 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
2003 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
2005 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
2007 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
2009 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
2011 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
2012 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
2013 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
2015 (main): Cast res_call to silence gcc.
2016 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
2018 (main): Cast res_call to silence gcc.
2019 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
2021 (main): Cast res_call to silence gcc.
2022 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
2024 (main): Cast res_call to silence gcc.
2025 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
2027 (main): Cast res_call to silence gcc.
2028 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
2030 (main): Cast res_call to silence gcc.
2031 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
2033 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
2034 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
2036 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
2037 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
2039 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
2040 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
2042 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
2043 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
2045 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
2046 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
2047 and userdata unused.
2048 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
2050 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
2051 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
2052 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
2053 Cleanup white spaces.
2054 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
2055 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
2056 cif and userdata unused.
2057 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
2059 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
2060 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
2061 formatters to silence gcc.
2062 (B_gn): Mark cif and userdata unused.
2063 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
2065 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
2066 (B_gn): Mark cif and userdata unused.
2067 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
2069 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
2070 (B_gn): Mark cif and userdata unused.
2071 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
2073 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
2074 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
2075 * testsuite/libffi.call/problem1.c (stub): Likewise.
2076 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
2078 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
2079 in the last commit for this test case in the test case itself.
2080 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
2082 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
2083 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
2084 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
2085 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
2086 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
2087 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
2089 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
2091 * src/sh/sysv.S: Fix register numbers in the FDE for
2094 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
2096 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
2097 declaration to avoid a false negative on ix86. See PR323.
2099 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
2101 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
2102 and cast integer to void * if needed. Update the pointer to
2103 the FP register saved area correctly.
2105 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
2107 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
2109 * testsuite/libffi.call/nested_struct4.c: Likewise.
2111 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
2113 * testsuite/libffi.call/return_dbl.c: New test case.
2114 * testsuite/libffi.call/return_dbl1.c: Likewise.
2115 * testsuite/libffi.call/return_dbl2.c: Likewise.
2116 * testsuite/libffi.call/return_fl.c: Likewise.
2117 * testsuite/libffi.call/return_fl1.c: Likewise.
2118 * testsuite/libffi.call/return_fl2.c: Likewise.
2119 * testsuite/libffi.call/return_fl3.c: Likewise.
2120 * testsuite/libffi.call/closure_fn6.c: Likewise.
2122 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
2124 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
2125 here to be used by other test cases too.
2127 * testsuite/libffi.call/nested_struct10.c: New test case.
2128 * testsuite/libffi.call/nested_struct9.c: Likewise.
2129 * testsuite/libffi.call/nested_struct8.c: Likewise.
2130 * testsuite/libffi.call/nested_struct7.c: Likewise.
2131 * testsuite/libffi.call/nested_struct6.c: Likewise.
2132 * testsuite/libffi.call/nested_struct5.c: Likewise.
2133 * testsuite/libffi.call/nested_struct4.c: Likewise.
2135 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2137 * configure.ac: Enable libffi for sparc64-*-freebsd*.
2138 * configure: Rebuilt.
2140 2006-01-18 Jakub Jelinek <jakub@redhat.com>
2142 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
2143 instead do the shifting inline.
2144 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
2145 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
2146 and 8 byte structs, for the remaining struct sizes don't call
2147 __lshrdi3, instead do the shifting inline.
2149 2005-12-07 Thiemo Seufer <ths@networkno.de>
2151 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
2152 missing parentheses.
2153 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
2154 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
2155 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
2156 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
2157 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
2159 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
2161 (ffi_prep_closure): Improve trampoline instruction scheduling.
2162 (ffi_closure_mips_inner_O32): Fix endianness bugs.
2164 2005-12-03 Alan Modra <amodra@bigpond.net.au>
2166 * src/powerpc/ffi.c: Formatting.
2167 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
2168 (ffi_prep_args64): Likewise.
2170 2005-09-30 Geoffrey Keating <geoffk@apple.com>
2172 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
2173 darwin, use -shared-libgcc not -lgcc_s, and explain why.
2175 2005-09-26 Tom Tromey <tromey@redhat.com>
2177 * testsuite/libffi.call/float1.c (value_type): New typedef.
2178 (CANARY): New define.
2179 (main): Check for result buffer overflow.
2180 * src/powerpc/linux64.S: Handle linux64 long double returns.
2181 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
2182 (ffi_prep_cif_machdep): Handle linux64 long double returns.
2184 2005-08-25 Alan Modra <amodra@bigpond.net.au>
2187 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
2189 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
2191 2005-08-11 Jakub Jelinek <jakub@redhat.com>
2193 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
2194 (AH_BOTTOM): Add FFI_HIDDEN definition.
2195 * configure: Rebuilt.
2196 * fficonfig.h.in: Rebuilt.
2197 * src/powerpc/ffi.c (hidden): Remove.
2198 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
2199 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
2200 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
2201 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
2202 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
2203 add FFI_HIDDEN to its prototype.
2204 (ffi_closure_SYSV_inner): New.
2205 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2206 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2208 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
2211 * configure: Rebuilt.
2212 * configure.ac: Handle i*86-*-gnu*.
2214 2005-08-09 Jakub Jelinek <jakub@redhat.com>
2216 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
2217 DW_CFA_offset_extended_sf rather than
2218 DW_CFA_GNU_negative_offset_extended.
2219 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
2221 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
2223 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
2225 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
2226 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
2227 partially on register.
2228 (ffi_closure_helper_SYSV): Likewise.
2229 (ffi_prep_cif_machdep): Don't set too many cif->flags.
2231 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
2233 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
2235 * src/sh64/ffi.c (simple_type): Remove.
2236 (return_type): Handle small structures correctly.
2237 (ffi_prep_args): Likewise.
2238 (ffi_call): Likewise.
2239 (ffi_closure_helper_SYSV): Likewise.
2240 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
2241 Emit position independent code if PIC and remove wrong datalabel
2242 prefixes from EH data.
2244 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
2246 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
2247 * Makefile.in: Regenerate.
2248 * include/Makefile.in: Likewise.
2249 * testsuite/Makefile.in: Likewise.
2250 * configure.ac: Add POWERPC_FREEBSD rules.
2251 * configure: Regenerate.
2252 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
2253 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
2254 * src/powerpc/ffi.c: Add flags to handle small structure returns
2256 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
2258 (ffi_closure_helper_SYSV): Likewise.
2259 * src/powerpc/ppc_closure.S: Add return types for small structures.
2260 * src/powerpc/sysv.S: Add bits to handle small structures for
2263 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
2265 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
2266 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
2267 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
2269 2005-07-05 Randolph Chung <tausq@debian.org>
2271 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
2272 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
2273 structures. Kill compilation warnings.
2274 (ffi_closure_inner_LINUX): Print return values as hex in debug
2275 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
2276 Properly handle 5-7 byte structure returns.
2277 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
2278 (FFI_TYPE_SMALL_STRUCT2): Remove.
2279 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
2280 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
2281 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
2282 (checksmst1, checksmst2): Remove.
2283 (checksmst3): Optimize handling of 3-byte struct returns.
2284 (checksmst567): Properly handle 5-7 byte struct returns.
2286 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2289 * src/mips/n32.S: Enforce PIC code.
2290 * src/mips/o32.S: Likewise.
2292 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2294 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
2295 * configure: Regenerate.
2297 2005-06-01 Alan Modra <amodra@bigpond.net.au>
2299 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
2300 to call ffi_closure_helper_SYSV. Append @local instead.
2301 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
2303 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
2305 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
2306 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
2307 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
2308 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
2309 include/Makefile.in, testsuite/Makefile.in: Regenerate.
2311 2005-05-09 Mike Stump <mrs@apple.com>
2313 * configure: Regenerate.
2315 2005-05-08 Richard Henderson <rth@redhat.com>
2318 * src/alpha/osf.S: Update unwind into to match code.
2320 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
2321 Richard Henderson <rth@redhat.com>
2323 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
2325 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
2326 (ffi_prep_closure): Set carry bit if sse-used flag set.
2327 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
2328 Only load sse registers if ssecount non-zero.
2329 (ffi_closure_unix64): Only save sse registers if carry set on entry.
2331 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
2333 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
2334 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
2335 * configure: Regenerate.
2337 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
2339 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
2340 have Tcl8.3-compatible intermediate variable.
2342 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
2343 Hans-Peter Nilsson <hp@axis.com>
2345 * Makefile.am: Add CRIS support.
2346 * configure.ac: Likewise.
2347 * Makefile.in, configure, testsuite/Makefile.in,
2348 include/Makefile.in: Regenerate.
2349 * src/cris: New directory.
2350 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
2351 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
2353 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
2354 \r?\n in output tests.
2356 2005-04-12 Mike Stump <mrs@apple.com>
2358 * configure: Regenerate.
2360 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
2362 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
2364 2005-03-30 Steve Ellcey <sje@cup.hp.com>
2366 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
2368 * src/ia64/unix.S (ffi_closure_unix): Extend gp
2369 to 64 bits in ILP32 mode.
2370 Load 64 bits even for short data.
2372 2005-03-23 Mike Stump <mrs@apple.com>
2374 * src/powerpc/darwin.S: Update for -m64 multilib.
2375 * src/powerpc/darwin_closure.S: Likewise.
2377 2005-03-21 Zack Weinberg <zack@codesourcery.com>
2379 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
2380 Do not set tool_include_dir.
2381 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
2383 * include/Makefile.am: Set gcc_version and toollibffidir.
2384 * include/Makefile.in: Regenerate.
2386 2005-02-22 Andrew Haley <aph@redhat.com>
2388 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2389 odd-numbered register pairs for 64-bit integer types.
2391 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2394 * testsuite/libffi.call/return_ll1.c: New test case.
2396 2005-02-11 Janis Johnson <janis187@us.ibm.com>
2398 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2399 * testsuite/libffi.call/float.c: Ditto.
2400 * testsuite/libffi.call/float2.c: Ditto.
2401 * testsuite/libffi.call/float3.c: Ditto.
2403 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2405 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2407 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2409 * testsuite/libffi.special/special.exp (cxx_options): Add
2412 2004-12-31 Richard Henderson <rth@redhat.com>
2414 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2415 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2416 offset parameters with a type parameter; deduce size and structure
2417 alignment. Update all users.
2419 2004-12-31 Richard Henderson <rth@redhat.com>
2421 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2422 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
2423 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
2424 into ffi_prep_closure.
2425 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2428 2004-12-27 Richard Henderson <rth@redhat.com>
2430 * src/x86/unix64.S: Fix typo in unwind info.
2432 2004-12-25 Richard Henderson <rth@redhat.com>
2434 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2435 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2436 (merge_classes): Check for it.
2438 (classify_argument): Pass byte_offset by value; perform all updates
2440 (examine_argument): Add classes argument; handle
2441 X86_64_COMPLEX_X87_CLASS.
2442 (ffi_prep_args): Merge into ...
2443 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2444 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2445 (ffi_fill_return_value): Remove.
2446 (ffi_prep_closure): Remove dead assert.
2447 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2448 Rewrite to use struct register_args instead of va_list. Create
2449 flags for handling structure returns.
2450 * src/x86/unix64.S: Remove dead strings.
2451 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2452 stack frame with ffi_call. Handle structure returns properly.
2453 (float2sse, floatfloat2sse, double2sse): Remove.
2454 (sse2float, sse2double, sse2floatfloat): Remove.
2455 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2456 to handle structure returns properly.
2458 2004-12-08 David Edelsohn <edelsohn@gnu.org>
2460 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2462 * Makefile.in: Regenerated.
2464 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2466 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2467 * configure, aclocal.m4, Makefile.in: Regenerate.
2468 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2470 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2472 * configure: Regenerate for libtool change.
2474 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2476 * configure: Regenerate for libtool reversion.
2478 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2480 * configure: Regenerate for libtool change.
2482 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2484 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2486 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2488 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2489 of jal. Use an absolute encoding for the frame information.
2491 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2493 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2494 * acinclude.m4: Delete logic for sincludes.
2495 * aclocal.m4, Makefile.in, configure: Regenerate.
2496 * include/Makefile: Likewise.
2497 * testsuite/Makefile: Likewise.
2499 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2501 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2502 on a 8-byte boundary.
2503 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2505 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2507 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2508 long long values. Round stack allocation to a multiple of 8 bytes
2509 for ATPCS compatibility.
2510 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2511 names. Handle returning long long types. Add Thumb and interworking
2512 support. Improve soft-float code.
2514 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2516 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2517 (libffi_exit): New function.
2518 (libffi_init): Build the testglue wrapper if needed.
2520 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2523 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2525 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2527 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2529 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2531 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2532 * testsuite/libffi.call/float3.c: New test case.
2534 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2536 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2537 the function returning a structure pointed with R2.
2538 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2539 the structure return value if T bit set. Emit position
2540 independent code and EH data if PIC.
2542 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2544 * Makefile.am: Add m32r support.
2545 * configure.ac: Likewise.
2546 * Makefile.in: Regenerate.
2547 * confiugre: Regenerate.
2548 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2549 (uint64, sint64, double, longdouble)
2550 * src/m32r: New directory.
2551 * src/m32r/ffi.c: New file.
2552 * src/m32r/sysv.S: Likewise.
2553 * src/m32r/ffitarget.h: Likewise.
2555 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2557 * testsuite/libffi.call/negint.c: New test case.
2559 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2562 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2563 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2564 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2567 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2569 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2570 * testsuite/libffi.call/promotion.c: Likewise.
2571 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2573 * testsuite/libffi.call/return_sc.c: Likewise.
2574 * testsuite/libffi.call/return_uc.c: Likewise.
2576 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2578 * src/powerpc/darwin.S: Fix comments and identation.
2579 * src/powerpc/darwin_closure.S: Likewise.
2581 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2583 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2584 (ffi_prep_args): Handle longdouble arguments.
2585 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2587 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2588 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2590 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2591 * src/types.c: Defined longdouble size and alignment for darwin.
2593 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2595 * src/powerpc/aix.S: Remove whitespaces.
2596 * src/powerpc/aix_closure.S: Likewise.
2597 * src/powerpc/asm.h: Likewise.
2598 * src/powerpc/ffi.c: Likewise.
2599 * src/powerpc/ffitarget.h: Likewise.
2600 * src/powerpc/linux64.S: Likewise.
2601 * src/powerpc/linux64_closure.S: Likewise.
2602 * src/powerpc/ppc_closure.S: Likewise.
2603 * src/powerpc/sysv.S: Likewise.
2605 2004-08-30 Anthony Green <green@redhat.com>
2607 * Makefile.am: Add frv support.
2608 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2609 * configure.ac: Read configure.host.
2610 * configure.in: Read configure.host.
2611 * configure.host: New file. frv-elf needs libgloss.
2612 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2613 alignment. This is needed to frv and shouldn't harm the others.
2614 * include/ffi_common.h (ALIGN_DOWN): New macro.
2615 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2617 2004-08-24 David Daney <daney@avtrex.com>
2619 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2620 * testsuite/libffi.call/closure_fn1.c: Likewise.
2621 * testsuite/libffi.call/closure_fn2.c Likewise.
2622 * testsuite/libffi.call/closure_fn3.c: Likewise.
2623 * testsuite/libffi.call/closure_fn4.c: Likewise.
2624 * testsuite/libffi.call/closure_fn5.c: Likewise.
2625 * testsuite/libffi.call/cls_18byte.c: Likewise.
2626 * testsuite/libffi.call/cls_19byte.c: Likewise.
2627 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2628 * testsuite/libffi.call/cls_20byte.c: Likewise.
2629 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2630 * testsuite/libffi.call/cls_24byte.c: Likewise.
2631 * testsuite/libffi.call/cls_2byte.c: Likewise.
2632 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2633 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2634 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2635 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2636 * testsuite/libffi.call/cls_4byte.c: Likewise.
2637 * testsuite/libffi.call/cls_64byte.c: Likewise.
2638 * testsuite/libffi.call/cls_6byte.c: Likewise.
2639 * testsuite/libffi.call/cls_7byte.c: Likewise.
2640 * testsuite/libffi.call/cls_8byte.c: Likewise.
2641 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2642 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2643 * testsuite/libffi.call/cls_align_double.c: Likewise.
2644 * testsuite/libffi.call/cls_align_float.c: Likewise.
2645 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2646 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2647 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2648 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2649 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2650 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2651 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2652 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2653 * testsuite/libffi.call/cls_double.c: Likewise.
2654 * testsuite/libffi.call/cls_float.c: Likewise.
2655 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2656 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2657 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2658 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2659 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2660 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2661 * testsuite/libffi.call/cls_schar.c: Likewise.
2662 * testsuite/libffi.call/cls_sint.c: Likewise.
2663 * testsuite/libffi.call/cls_sshort.c: Likewise.
2664 * testsuite/libffi.call/cls_uchar.c: Likewise.
2665 * testsuite/libffi.call/cls_uint.c: Likewise.
2666 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2667 * testsuite/libffi.call/cls_ushort.c: Likewise.
2668 * testsuite/libffi.call/nested_struct.c: Likewise.
2669 * testsuite/libffi.call/nested_struct1.c: Likewise.
2670 * testsuite/libffi.call/nested_struct2.c: Likewise.
2671 * testsuite/libffi.call/nested_struct3.c: Likewise.
2672 * testsuite/libffi.call/problem1.c: Likewise.
2673 * testsuite/libffi.special/unwindtest.cc: Likewise.
2674 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2676 * testsuite/libffi.call/cls_16byte.c: Likewise.
2677 * testsuite/libffi.call/cls_5byte.c: Likewise.
2679 2004-08-23 David Daney <daney@avtrex.com>
2682 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2683 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2684 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2685 parameters and return types.
2686 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2687 (ffi_prep_closure): Ditto.
2688 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2689 alignment calculations.
2690 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2691 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2693 2004-08-14 Casey Marshall <csm@gnu.org>
2695 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2696 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2697 integer (O32 ABI only).
2698 (ffi_prep_closure): new function.
2699 (ffi_closure_mips_inner_O32): new function.
2700 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2701 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2702 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2703 64 bit integers correctly.
2704 (ffi_closure_O32): new function.
2705 Added DWARF-2 unwind info for both functions.
2707 2004-08-10 Andrew Haley <aph@redhat.com>
2709 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2711 2004-08-01 Robert Millan <robertmh@gnu.org>
2713 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2714 * configure: Regenerate.
2716 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2718 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2719 and mmap() explicitly instead of relying on preset autoconf cache
2721 * aclocal.m4: Regenerate.
2722 * configure: Regenerate.
2724 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2726 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2727 (ffi_check_float_struct): Remove unused prototype.
2729 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2731 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2732 character on Darwin, use '\n\t' instead.
2734 2004-06-26 Matthias Klose <doko@debian.org>
2736 * libtool-version: Fix typo in revision/age.
2738 2004-06-17 Matthias Klose <doko@debian.org>
2740 * libtool-version: New.
2741 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2742 * Makefile.in: Regenerate.
2744 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2746 * Makefile.am: Remove useless multilib rules.
2747 * Makefile.in: Regenerate.
2748 * aclocal.m4: Regenerate with automake 1.8.5.
2749 * configure.ac: Remove useless multilib configury.
2750 * configure: Regenerate.
2752 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2754 * .cvsignore: New file.
2756 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2758 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2760 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2763 2004-06-06 Sean McNeil <sean@mcneil.com>
2765 * configure.ac: Add x86_64-*-freebsd* support.
2766 * configure: Regenerate.
2768 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2771 * configure.ac: Test for existence of mmap and sys/mman.h before
2772 checking blacklist. Fix suggested by Jim Wilson.
2773 * configure: Regenerate.
2775 2004-04-26 Matt Austern <austern@apple.com>
2777 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2779 * src/powerpc/darwin_closure.S: Likewise.
2781 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2783 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2784 error. Reported by Thomas Heller <theller@python.net>.
2785 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2786 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2788 2004-03-20 Matthias Klose <doko@debian.org>
2790 * src/pa/linux.S: Fix typo.
2792 2004-03-19 Matthias Klose <doko@debian.org>
2794 * Makefile.am: Update.
2795 * Makefile.in: Regenerate.
2796 * src/pa/ffi.h.in: Remove.
2797 * src/pa/ffitarget.h: New file.
2799 2004-02-10 Randolph Chung <tausq@debian.org>
2801 * Makefile.am: Add PA support.
2802 * Makefile.in: Regenerate.
2803 * include/Makefile.in: Regenerate.
2804 * configure.ac: Add PA target.
2805 * configure: Regenerate.
2806 * src/pa/ffi.c: New file.
2807 * src/pa/ffi.h.in: Add PA support.
2808 * src/pa/linux.S: New file.
2809 * prep_cif.c: Add PA support.
2811 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2813 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2815 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2816 with ecif->cif->flags.
2817 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2819 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2820 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2821 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2822 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2824 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2826 * configure.in: Rename file to ...
2827 * configure.ac: ... this.
2828 * fficonfig.h.in: Regenerate.
2829 * Makefile.in: Regenerate.
2830 * include/Makefile.in: Regenerate.
2831 * testsuite/Makefile.in: Regenerate.
2833 2004-03-12 Matt Austern <austern@apple.com>
2835 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2836 changes in EH format resulting from addition of linkonce support.
2837 * src/powerpc/darwin_closure.S: Likewise.
2839 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2840 Paolo Bonzini <bonzini@gnu.org>
2842 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2843 Remove VPATH. Remove rules for object files. Remove multilib support.
2844 (AM_CCASFLAGS): Add.
2845 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2846 (AC_PREREQ): Bump version to 2.59.
2847 (AC_INIT): Fill with version info and bug address.
2848 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2849 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2850 De-precious CC so that the right flags are passed down to multilibs.
2851 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2852 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2853 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2854 * configure: Rebuilt.
2855 * aclocal.m4: Likewise.
2856 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2857 * fficonfig.h.in: Likewise.
2859 2004-03-11 Andreas Schwab <schwab@suse.de>
2861 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2862 arguments from fp registers only for the first 8 parameter slots.
2863 Don't convert a float parameter when passed in memory.
2865 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2867 * configure: Regenerate for config/accross.m4 correction.
2869 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2871 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2872 ecif->cif->bytes to bytes.
2873 (ffi_prep_cif_machdep): Add braces around nested if statement.
2875 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2877 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2879 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2880 (ffi_closure_helper_LINUX64): Fix typo.
2881 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2883 * testsuite/libffi.call/float.c: Likewise.
2884 * testsuite/libffi.call/float2.c: Likewise.
2886 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2888 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2889 long double function return and long double arg handling.
2890 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2891 Use "end_pfr" instead of "nf". Correct long double handling.
2893 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2895 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2896 space for long double return value. Adjust stack frame and offsets.
2897 Load f2 long double return.
2899 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2901 * src/types.c: Use 16 byte long double for POWERPC64.
2903 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2905 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2906 when the structure return address is passed in %o0.
2907 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2908 (ffi_v9_layout_struct): Align the field following a nested structure
2909 on a word boundary. Use memmove instead of memcpy.
2910 (ffi_call): Update call to ffi_V9_return_struct.
2911 (ffi_prep_closure): Define 'ctx' only for V8.
2912 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2913 and ffi_closure_sparc_inner_v9.
2914 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2915 Always skip the structure return address. For structures and long
2916 doubles, copy the argument directly.
2917 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2918 if required. Shift the maximum floating-point slot accordingly. For
2919 big structures, copy the argument directly; otherwise, left-justify the
2920 argument and call ffi_v9_layout_struct to lay out the structure on
2922 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2923 (ffi_closure_v8): Pass the structure return address. Update call to
2924 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2925 Skip the 'unimp' insn when returning long doubles and structures.
2926 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2927 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2928 FFI_TYPE_INT handling. Load structures both in integers and
2929 floating-point registers on return.
2930 * README: Update status of the SPARC port.
2932 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2934 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2936 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2938 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2940 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2941 value as of type ffi_arg, not unsigned int.
2943 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2945 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2948 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2950 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2953 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2955 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2958 2003-12-10 Richard Henderson <rth@redhat.com>
2960 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2961 size_t instead of int.
2963 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2965 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2966 * testsuite/libffi.call/many_win32.c (main): Replace variable
2967 int i with unsigned long ul.
2969 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2970 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2971 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2972 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2973 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2974 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2975 * testsuite/libffi.call/cls_align_float.c: Likewise.
2976 * testsuite/libffi.call/cls_align_double.c: Likewise.
2977 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2978 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2980 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2983 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2984 Align arguments to 32 bits.
2986 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2989 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2990 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2991 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2992 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2993 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2994 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2996 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2998 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3000 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
3001 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3003 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
3005 * Makefile.in: Rebuilt.
3006 * configure: Likewise.
3007 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
3010 2003-11-21 Andreas Jaeger <aj@suse.de>
3011 Andreas Tobler <a.tobler@schweiz.ch>
3013 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
3014 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
3015 * Makefile.in: Rebuilt.
3016 * aclocal.m4: Likewise.
3017 * configure: Likewise.
3018 * fficonfig.h.in: Likewise.
3019 * testsuite/lib/libffi-dg.exp: Add include dir.
3020 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
3021 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3022 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
3023 for ffi_closure if available.
3024 * testsuite/libffi.call/closure_fn1.c: Likewise.
3025 * testsuite/libffi.call/closure_fn2.c: Likewise.
3026 * testsuite/libffi.call/closure_fn3.c: Likewise.
3027 * testsuite/libffi.call/closure_fn4.c: Likewise.
3028 * testsuite/libffi.call/closure_fn5.c: Likewise.
3029 * testsuite/libffi.call/cls_12byte.c: Likewise.
3030 * testsuite/libffi.call/cls_16byte.c: Likewise.
3031 * testsuite/libffi.call/cls_18byte.c: Likewise.
3032 * testsuite/libffi.call/cls_19byte.c: Likewise.
3033 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3034 * testsuite/libffi.call/cls_20byte.c: Likewise.
3035 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3036 * testsuite/libffi.call/cls_24byte.c: Likewise.
3037 * testsuite/libffi.call/cls_2byte.c: Likewise.
3038 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3039 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3040 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3041 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3042 * testsuite/libffi.call/cls_4byte.c: Likewise.
3043 * testsuite/libffi.call/cls_5byte.c: Likewise.
3044 * testsuite/libffi.call/cls_64byte.c: Likewise.
3045 * testsuite/libffi.call/cls_6byte.c: Likewise.
3046 * testsuite/libffi.call/cls_7byte.c: Likewise.
3047 * testsuite/libffi.call/cls_8byte.c: Likewise.
3048 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3049 * testsuite/libffi.call/cls_9byte2.c: Likewise.
3050 * testsuite/libffi.call/cls_double.c: Likewise.
3051 * testsuite/libffi.call/cls_float.c: Likewise.
3052 * testsuite/libffi.call/cls_schar.c: Likewise.
3053 * testsuite/libffi.call/cls_sint.c: Likewise.
3054 * testsuite/libffi.call/cls_sshort.c: Likewise.
3055 * testsuite/libffi.call/cls_uchar.c: Likewise.
3056 * testsuite/libffi.call/cls_uint.c: Likewise.
3057 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3058 * testsuite/libffi.call/cls_ushort.c: Likewise.
3059 * testsuite/libffi.call/nested_struct.c: Likewise.
3060 * testsuite/libffi.call/nested_struct1.c: Likewise.
3061 * testsuite/libffi.call/nested_struct2.c: Likewise.
3062 * testsuite/libffi.call/nested_struct3.c: Likewise.
3063 * testsuite/libffi.call/problem1.c: Likewise.
3064 * testsuite/libffi.special/unwindtest.cc: Likewise.
3066 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
3068 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
3070 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
3072 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
3073 Add -lgcc_s to additional flags.
3075 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
3077 * configure.in, include/Makefile.am: PR libgcj/11147, install
3078 the ffitarget.h header file in a gcc versioned and target
3080 * configure: Regenerated.
3081 * Makefile.in, include/Makefile.in: Likewise.
3082 * testsuite/Makefile.in: Likewise.
3084 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
3086 * testsuite/libffi.call/closure_fn0.c: Print result and check
3087 with dg-output to make debugging easier.
3088 * testsuite/libffi.call/closure_fn1.c: Likewise.
3089 * testsuite/libffi.call/closure_fn2.c: Likewise.
3090 * testsuite/libffi.call/closure_fn3.c: Likewise.
3091 * testsuite/libffi.call/closure_fn4.c: Likewise.
3092 * testsuite/libffi.call/closure_fn5.c: Likewise.
3093 * testsuite/libffi.call/cls_12byte.c: Likewise.
3094 * testsuite/libffi.call/cls_16byte.c: Likewise.
3095 * testsuite/libffi.call/cls_18byte.c: Likewise.
3096 * testsuite/libffi.call/cls_19byte.c: Likewise.
3097 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3098 * testsuite/libffi.call/cls_20byte.c: Likewise.
3099 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3100 * testsuite/libffi.call/cls_24byte.c: Likewise.
3101 * testsuite/libffi.call/cls_2byte.c: Likewise.
3102 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3103 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3104 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3105 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3106 * testsuite/libffi.call/cls_4byte.c: Likewise.
3107 * testsuite/libffi.call/cls_5byte.c: Likewise.
3108 * testsuite/libffi.call/cls_64byte.c: Likewise.
3109 * testsuite/libffi.call/cls_6byte.c: Likewise.
3110 * testsuite/libffi.call/cls_7byte.c: Likewise.
3111 * testsuite/libffi.call/cls_8byte.c: Likewise.
3112 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3113 * testsuite/libffi.call/cls_9byte2.c: Likewise.
3114 * testsuite/libffi.call/cls_double.c: Likewise.
3115 * testsuite/libffi.call/cls_float.c: Likewise.
3116 * testsuite/libffi.call/cls_schar.c: Likewise.
3117 * testsuite/libffi.call/cls_sint.c: Likewise.
3118 * testsuite/libffi.call/cls_sshort.c: Likewise.
3119 * testsuite/libffi.call/cls_uchar.c: Likewise.
3120 * testsuite/libffi.call/cls_uint.c: Likewise.
3121 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3122 * testsuite/libffi.call/cls_ushort.c: Likewise.
3123 * testsuite/libffi.call/problem1.c: Likewise.
3125 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
3128 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
3130 * testsuite/libffi.call/cls_9byte2.c: New test case.
3131 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3132 * testsuite/libffi.call/cls_64byte.c: Likewise.
3133 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3134 * testsuite/libffi.call/cls_19byte.c: Likewise.
3135 * testsuite/libffi.call/cls_18byte.c: Likewise.
3136 * testsuite/libffi.call/closure_fn4.c: Likewise.
3137 * testsuite/libffi.call/closure_fn5.c: Likewise.
3138 * testsuite/libffi.call/cls_schar.c: Likewise.
3139 * testsuite/libffi.call/cls_sint.c: Likewise.
3140 * testsuite/libffi.call/cls_sshort.c: Likewise.
3141 * testsuite/libffi.call/nested_struct2.c: Likewise.
3142 * testsuite/libffi.call/nested_struct3.c: Likewise.
3144 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
3146 * testsuite/libffi.call/cls_double.c: Do a check on the result.
3147 * testsuite/libffi.call/cls_uchar.c: Likewise.
3148 * testsuite/libffi.call/cls_uint.c: Likewise.
3149 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3150 * testsuite/libffi.call/cls_ushort.c: Likewise.
3151 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
3153 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
3155 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
3158 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
3160 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
3161 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
3163 2003-10-22 David Daney <ddaney@avtrex.com>
3165 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
3166 __attribute__((__mode__(__SI__))) and friends.
3168 2003-10-22 Andreas Schwab <schwab@suse.de>
3170 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
3172 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
3174 * configure.in: AC_LINK_FILES(ffitarget.h).
3175 * configure: Regenerate.
3176 * Makefile.in: Likewise.
3177 * include/Makefile.in: Likewise.
3178 * testsuite/Makefile.in: Likewise.
3179 * fficonfig.h.in: Likewise.
3181 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
3182 Richard Henderson <rth@redhat.com>
3184 Avoid that ffi.h includes fficonfig.h.
3186 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
3187 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
3188 (TARGET_SRC_MIPS_SGI): Removed.
3189 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
3190 (MIPS_SGI): Removed.
3191 (CLEANFILES): Removed.
3192 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
3194 * acconfig.h: Removed.
3195 * configure.in: Compute sizeofs only for double and long double.
3196 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
3197 into AC_DEFINE instead of using acconfig.h. Create
3198 include/ffitarget.h instead of include/fficonfig.h. Rename
3199 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
3200 AC_DEFINE EH_FRAME_FLAGS.
3201 * include/Makefile.am (DISTCLEANFILES): New automake macro.
3202 (hack_DATA): Add ffitarget.h.
3203 * include/ffi.h.in: Remove all system specific definitions.
3204 Declare raw API even if it is not installed, why bother?
3205 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
3206 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
3207 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
3208 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
3209 * include/ffi_common.h (bool): Do not define.
3210 (ffi_assert): Accept failed assertion.
3211 (ffi_type_test): Return void and accept file/line.
3212 (FFI_ASSERT): Pass stringized failed assertion.
3213 (FFI_ASSERT_AT): New macro.
3214 (FFI_ASSERT_VALID_TYPE): New macro.
3215 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
3216 UINT64, SINT64): Define here with gcc's __attribute__ macro
3218 (FLOAT32, ALIGN): Define here instead of in ffi.h
3219 * include/ffi-mips.h: Removed. Its content moved to
3220 src/mips/ffitarget.h after separating assembly and C sections.
3221 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
3222 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
3223 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
3224 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
3225 SIZEOF_ARG -> FFI_SIZEOF_ARG.
3226 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
3227 * src/debug.c (ffi_assert): Accept stringized failed assertion.
3228 (ffi_type_test): Rewritten.
3229 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
3230 FFI_ASSERT_VALID_TYPE.
3231 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
3232 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
3233 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
3234 src/s390/ffitarget.h, src/sh/ffitarget.h,
3235 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
3236 src/x86/ffitarget.h: New files.
3237 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
3238 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
3239 src/powerpc/aix.S, src/powerpc/darwin.S,
3240 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
3241 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
3242 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
3243 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
3244 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
3247 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3249 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
3250 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
3252 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
3254 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
3255 Used when FFI_DEBUG = 1.
3257 2003-10-14 Alan Modra <amodra@bigpond.net.au>
3259 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
3262 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3264 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
3265 FFI_MIPS_O32 for O32 ABI.
3267 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
3269 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
3270 SPARC64. Cleanup whitespaces.
3272 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
3274 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
3275 strongarm, xscale. Cleanup whitespaces.
3276 * testsuite/libffi.call/closure_fn1.c: Likewise.
3277 * testsuite/libffi.call/closure_fn2.c: Likewise.
3278 * testsuite/libffi.call/closure_fn3.c: Likewise.
3279 * testsuite/libffi.call/cls_12byte.c: Likewise.
3280 * testsuite/libffi.call/cls_16byte.c: Likewise.
3281 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3282 * testsuite/libffi.call/cls_20byte.c: Likewise.
3283 * testsuite/libffi.call/cls_24byte.c: Likewise.
3284 * testsuite/libffi.call/cls_2byte.c: Likewise.
3285 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3286 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3287 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3288 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3289 * testsuite/libffi.call/cls_4byte.c: Likewise.
3290 * testsuite/libffi.call/cls_5byte.c: Likewise.
3291 * testsuite/libffi.call/cls_6byte.c: Likewise.
3292 * testsuite/libffi.call/cls_7byte.c: Likewise.
3293 * testsuite/libffi.call/cls_8byte.c: Likewise.
3294 * testsuite/libffi.call/cls_double.c: Likewise.
3295 * testsuite/libffi.call/cls_float.c: Likewise.
3296 * testsuite/libffi.call/cls_uchar.c: Likewise.
3297 * testsuite/libffi.call/cls_uint.c: Likewise.
3298 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3299 * testsuite/libffi.call/cls_ushort.c: Likewise.
3300 * testsuite/libffi.call/nested_struct.c: Likewise.
3301 * testsuite/libffi.call/nested_struct1.c: Likewise.
3302 * testsuite/libffi.call/problem1.c: Likewise.
3303 * testsuite/libffi.special/unwindtest.cc: Likewise.
3304 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
3306 2003-09-18 David Edelsohn <edelsohn@gnu.org>
3308 * src/powerpc/aix.S: Cleanup whitespaces.
3309 * src/powerpc/aix_closure.S: Likewise.
3311 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3313 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
3314 * src/powerpc/darwin_closure.S: Likewise.
3315 * src/powerpc/ffi_darwin.c: Likewise.
3317 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3318 David Edelsohn <edelsohn@gnu.org>
3320 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
3321 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
3323 * src/powerpc/darwin_closure.S: Likewise.
3324 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
3325 according to the Darwin/AIX ABI.
3326 (ffi_prep_cif_machdep): Likewise.
3327 (ffi_closure_helper_DARWIN): Likewise.
3328 Remove the outgoing parameter stack logic. Simplify the evaluation
3329 of the different CASE types.
3330 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
3331 statement in the trampoline code.
3333 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
3335 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
3336 for the register size.
3337 (ffi_closure_helper_SYSV): Handle the structure return value
3339 (ffi_closure_helper_SYSV): Return the appropriate type when
3340 the registers are used for the structure return value.
3341 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
3342 the 64-bit return value. Update copyright years.
3344 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3346 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
3347 srcdir for ffi_mips.h.
3349 2003-09-12 Alan Modra <amodra@bigpond.net.au>
3351 * src/prep_cif.c (initialize_aggregate): Include tail padding in
3353 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
3354 placement of float result.
3355 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
3356 cast of "resp" for big-endian 64 bit machines.
3358 2003-09-11 Alan Modra <amodra@bigpond.net.au>
3360 * src/types.c (double, longdouble): Merge identical SH and ARM
3361 typedefs, and add POWERPC64.
3362 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
3363 struct split over gpr and rest.
3364 (ffi_prep_cif_machdep): Correct intarg_count for structures.
3365 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
3367 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
3369 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
3372 2003-09-09 Alan Modra <amodra@bigpond.net.au>
3374 * configure: Regenerate.
3376 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3378 * Makefile.am: Remove build rules for ffitest.
3379 * Makefile.in: Rebuilt.
3381 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3383 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3384 about implicit declaration of abort().
3386 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3388 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3389 * Makefile.in: Rebuilt.
3390 * configure.in: Add dejagnu test framework.
3391 * configure: Rebuilt.
3393 * testsuite/Makefile.am: New file.
3394 * testsuite/Makefile.in: Built
3395 * testsuite/lib/libffi-dg.exp: New file.
3396 * testsuite/config/default.exp: Likewise.
3397 * testsuite/libffi.call/call.exp: Likewise.
3398 * testsuite/libffi.call/ffitest.h: Likewise.
3399 * testsuite/libffi.call/closure_fn0.c: Likewise.
3400 * testsuite/libffi.call/closure_fn1.c: Likewise.
3401 * testsuite/libffi.call/closure_fn2.c: Likewise.
3402 * testsuite/libffi.call/closure_fn3.c: Likewise.
3403 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3404 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3405 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3406 * testsuite/libffi.call/cls_2byte.c: Likewise.
3407 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3408 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3409 * testsuite/libffi.call/cls_4byte.c: Likewise.
3410 * testsuite/libffi.call/cls_5byte.c: Likewise.
3411 * testsuite/libffi.call/cls_6byte.c: Likewise.
3412 * testsuite/libffi.call/cls_7byte.c: Likewise.
3413 * testsuite/libffi.call/cls_8byte.c: Likewise.
3414 * testsuite/libffi.call/cls_12byte.c: Likewise.
3415 * testsuite/libffi.call/cls_16byte.c: Likewise.
3416 * testsuite/libffi.call/cls_20byte.c: Likewise.
3417 * testsuite/libffi.call/cls_24byte.c: Likewise.
3418 * testsuite/libffi.call/cls_double.c: Likewise.
3419 * testsuite/libffi.call/cls_float.c: Likewise.
3420 * testsuite/libffi.call/cls_uchar.c: Likewise.
3421 * testsuite/libffi.call/cls_uint.c: Likewise.
3422 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3423 * testsuite/libffi.call/cls_ushort.c: Likewise.
3424 * testsuite/libffi.call/float.c: Likewise.
3425 * testsuite/libffi.call/float1.c: Likewise.
3426 * testsuite/libffi.call/float2.c: Likewise.
3427 * testsuite/libffi.call/many.c: Likewise.
3428 * testsuite/libffi.call/many_win32.c: Likewise.
3429 * testsuite/libffi.call/nested_struct.c: Likewise.
3430 * testsuite/libffi.call/nested_struct1.c: Likewise.
3431 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3432 * testsuite/libffi.call/problem1.c: Likewise.
3433 * testsuite/libffi.call/promotion.c: Likewise.
3434 * testsuite/libffi.call/return_ll.c: Likewise.
3435 * testsuite/libffi.call/return_sc.c: Likewise.
3436 * testsuite/libffi.call/return_uc.c: Likewise.
3437 * testsuite/libffi.call/strlen.c: Likewise.
3438 * testsuite/libffi.call/strlen_win32.c: Likewise.
3439 * testsuite/libffi.call/struct1.c: Likewise.
3440 * testsuite/libffi.call/struct2.c: Likewise.
3441 * testsuite/libffi.call/struct3.c: Likewise.
3442 * testsuite/libffi.call/struct4.c: Likewise.
3443 * testsuite/libffi.call/struct5.c: Likewise.
3444 * testsuite/libffi.call/struct6.c: Likewise.
3445 * testsuite/libffi.call/struct7.c: Likewise.
3446 * testsuite/libffi.call/struct8.c: Likewise.
3447 * testsuite/libffi.call/struct9.c: Likewise.
3448 * testsuite/libffi.special/special.exp: New file.
3449 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3450 * testsuite/libffi.special/unwindtest.cc: Likewise.
3453 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3455 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3458 2003-08-02 Alan Modra <amodra@bigpond.net.au>
3460 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3462 (ffi_closure_helper_LINUX64): Likewise.
3463 * src/powerpc/linux64.S: Remove code writing to parm save area.
3464 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3465 address in lr from ffi_closure_helper_LINUX64 call to calculate
3466 table address. Optimize function tail.
3468 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3470 * src/sparc/ffi.c: Handle all floating point registers.
3471 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3473 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3475 * README: Note that libffi is not part of GCC. Update the project
3478 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3480 * src/powerpc/ppc_closure.S: Include ffi.h.
3482 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3484 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3485 Use C style comments.
3487 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3489 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3490 * Makefile.in: Regenerate.
3491 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3492 * configure: Regenerate.
3493 * include/ffi.h.in: Add SHmedia support.
3494 * src/sh64/ffi.c: New file.
3495 * src/sh64/sysv.S: New file.
3497 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3499 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3500 should be read-only.
3501 * configure: Rebuilt.
3502 * fficonfig.h.in: Rebuilt.
3503 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3504 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3505 * src/powerpc/linux64.S: Likewise.
3506 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3507 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3508 if -fpic/-fPIC/-mrelocatable.
3509 * src/powerpc/powerpc_closure.S: Likewise.
3510 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3511 #write in .eh_frame flags.
3512 * src/sparc/v9.S: Likewise.
3513 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3514 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3515 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3517 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3519 Fixes PR bootstrap/10656
3520 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3521 support for .register pseudo-op.
3522 * src/sparc/v8.S: Use it.
3523 * fficonfig.h.in: Rebuilt.
3524 * configure: Rebuilt.
3526 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3528 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3529 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3530 Make it the default on POWERPC64.
3531 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3532 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3533 * configure: Rebuilt.
3534 * src/powerpc/ffi.c (hidden): Define.
3535 (ffi_prep_args_SYSV): Renamed from
3536 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3537 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3538 ASM_NEEDS_REGISTERS64): New.
3539 (ffi_prep_args64): New function.
3540 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3541 (ffi_call): Likewise.
3542 (ffi_prep_closure): Likewise.
3543 (flush_icache): Surround by #ifndef POWERPC64.
3544 (ffi_dblfl): New union type.
3545 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3546 (ffi_closure_helper_LINUX64): New function.
3547 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3549 * src/powerpc/sysv.S: Likewise.
3550 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3551 * src/powerpc/linux64.S: New file.
3552 * src/powerpc/linux64_closure.S: New file.
3553 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3554 src/powerpc/linux64_closure.S.
3555 (TARGET_SRC_POWERPC): Likewise.
3557 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3558 closure_test_fn3): Fix result printing on big-endian 64-bit
3560 (main): Print tst2_arg instead of uninitialized tst2_result.
3562 * src/ffitest.c (main): Hide what closure pointer really points to
3565 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3567 * configure.in (arm-*-netbsdelf*): Add configuration.
3568 (configure): Regenerated.
3570 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3572 * include/Makefile.in: Regenerate.
3574 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3576 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3578 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3579 Receive closure pointer through parameter, read args using
3580 __builtin_dwarf_cfa.
3581 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3583 2003-03-12 Andreas Schwab <schwab@suse.de>
3585 * configure.in: Avoid trailing /. in toolexeclibdir.
3586 * configure: Rebuilt.
3588 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3590 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3592 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3594 * libffi/src/powerpc/darwin_closure.S:
3595 Fix alignement bug, allocate 8 bytes for the result.
3596 * libffi/src/powerpc/aix_closure.S:
3598 * libffi/src/powerpc/ffi_darwin.c:
3599 Update stackframe description for aix/darwin_closure.S.
3601 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3603 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3606 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3607 Andreas Schwab <schwab@suse.de>
3609 * configure.in: Adjust command to source config-ml.in to account
3610 for changes to the libffi_basedir definition.
3611 (libffi_basedir): Remove ${srcdir} from value and include trailing
3614 * configure: Regenerate.
3616 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3618 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3620 2003-01-28 Andrew Haley <aph@redhat.com>
3622 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3623 * src/x86/ffi64.c (ffi_prep_closure): New.
3624 (ffi_closure_UNIX64_inner): New.
3625 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3627 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
3629 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3630 Remove USE_LIBDIR conditional.
3631 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3632 * Makefile.in, configure: Rebuilt.
3634 2003-01027 David Edelsohn <edelsohn@gnu.org>
3636 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3637 * Makefile.in: Regenerate.
3639 2003-01-22 Andrew Haley <aph@redhat.com>
3641 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3644 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3646 * src/powerpc/darwin.S: Add unwind info.
3647 * src/powerpc/darwin_closure.S: Likewise.
3649 2003-01-14 Andrew Haley <aph@redhat.com>
3651 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3652 (ffi_prep_cif_machdep): Likewise.
3653 * src/x86/unix64.S: Add unwind info.
3655 2003-01-14 Andreas Jaeger <aj@suse.de>
3657 * src/ffitest.c (main): Only use ffi_closures if those are
3660 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3662 * libffi/src/ffitest.c
3663 add closure testcases
3665 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3667 * libffi/src/powerpc/ffi.c
3668 fix alignment bug for float (4 byte aligned iso 8 byte)
3670 2003-01-09 Geoffrey Keating <geoffk@apple.com>
3672 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3673 * src/powerpc/darwin.S: Remove RCS version string.
3675 2003-01-03 Jeff Sturm <jsturm@one-point.com>
3677 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3678 * src/ffitest.c (main): Use static storage for closure.
3679 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3680 * src/sparc/v8.S (ffi_closure_v8): New.
3681 * src/sparc/v9.S (ffi_closure_v9): New.
3683 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3685 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3686 enumeration for X86_WIN32.
3687 * src/x86/win32.S: Added ffi_call_STDCALL function
3689 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3690 switch cases for recognising FFI_STDCALL and
3691 calling ffi_call_STDCALL if target is X86_WIN32.
3692 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3693 stdcall versions of the "my_strlen" and "many"
3694 test functions (for X86_WIN32).
3695 Added test cases to test stdcall invocation using
3698 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3700 * src/sh/sysv.S: Add DWARF2 unwind info.
3702 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3704 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3706 2002-11-26 Jim Wilson <wilson@redhat.com>
3708 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3710 2002-11-23 H.J. Lu <hjl@gnu.org>
3712 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3713 Include ../config/accross.m4.
3714 * aclocal.m4; Rebuild.
3715 * configure: Likewise.
3717 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3719 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3721 2002-11-11 DJ Delorie <dj@redhat.com>
3723 * configure.in: Look for common files in the right place.
3725 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3727 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3728 raw data as _Jv_word values, not ffi_raw.
3729 (ffi_java_ptrarray_to_raw): Likewise.
3730 (ffi_java_rvalue_to_raw): New function.
3731 (ffi_java_raw_call): Call it.
3732 (ffi_java_raw_to_rvalue): New function.
3733 (ffi_java_translate_args): Call it.
3734 * src/ffitest.c (closure_test_fn): Interpret return value
3735 as ffi_arg, not int.
3736 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3737 FFI_TYPE_POINTER case.
3738 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3739 values extended to word size.
3741 2002-10-02 Andreas Jaeger <aj@suse.de>
3743 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3745 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
3747 * include/ffi.h.in: Fix i386 win32 compilation.
3749 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3751 * configure.in: Add s390x-*-linux-* target.
3752 * configure: Regenerate.
3753 * include/ffi.h.in: Define S390X for s390x targets.
3754 (FFI_CLOSURES): Define for s390/s390x.
3755 (FFI_TRAMPOLINE_SIZE): Likewise.
3756 (FFI_NATIVE_RAW_API): Likewise.
3757 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3758 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
3759 * src/s390/ffi.c: Major rework of existing code. Add support for
3760 s390x targets. Add closure support.
3761 * src/s390/sysv.S: Likewise.
3763 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
3765 * src/arm/sysv.S: Fix typo.
3767 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
3769 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3770 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3771 (ffi_call_SYSV): Handle soft-float.
3773 2002-09-27 Bo Thorsen <bo@suse.de>
3775 * include/ffi.h.in: Fix multilib x86-64 support.
3777 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3779 * Makefile.am (all-multi): Fix multilib parallel build.
3781 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3783 * configure.in (sh[34]*-*-linux*): Add brackets.
3784 * configure: Regenerate.
3786 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3788 * Makefile.am: Add SH support.
3789 * Makefile.in: Regenerate.
3790 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3791 * configure: Regenerate.
3792 * include/ffi.h.in: Add SH support.
3793 * src/sh/ffi.c: New file.
3794 * src/sh/sysv.S: New file.
3795 * src/types.c: Add SH support.
3797 2002-07-16 Bo Thorsen <bo@suse.de>
3799 * src/x86/ffi64.c: New file that adds x86-64 support.
3800 * src/x86/unix64.S: New file that handles argument setup for
3802 * src/x86/sysv.S: Don't use this on x86-64.
3803 * src/x86/ffi.c: Don't use this on x86-64.
3805 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3807 * src/ffitest.c (struct6): New test that tests a special case in
3809 (struct7): Likewise.
3810 (struct8): Likewise.
3811 (struct9): Likewise.
3812 (closure_test_fn): Silence warning about this when it's not used.
3813 (main): Add the new tests.
3814 (main): Fix a couple of wrong casts and silence some compiler warnings.
3815 * include/ffi.h.in: Add x86-64 ABI definition.
3816 * fficonfig.h.in: Regenerate.
3817 * Makefile.am: Add x86-64 support.
3818 * configure.in: Likewise.
3819 * Makefile.in: Regenerate.
3820 * configure: Likewise.
3822 2002-06-24 Bo Thorsen <bo@suse.de>
3824 * src/types.c: Merge settings for similar architectures.
3825 Add x86-64 sizes and alignments.
3827 2002-06-23 Bo Thorsen <bo@suse.de>
3829 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3830 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3831 * src/mips/ffi.c (ffi_prep_args): Likewise.
3832 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3834 2002-07-18 H.J. Lu (hjl@gnu.org)
3836 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3837 (libffi_la_SOURCES): Support MIPS_LINUX.
3838 (libffi_convenience_la_SOURCES): Likewise.
3839 * Makefile.in: Regenerated.
3841 * configure.in (mips64*-*): Skip.
3842 (mips*-*-linux*): New.
3843 * configure: Regenerated.
3845 * src/mips/ffi.c: Include <sgidefs.h>.
3847 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3849 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3851 2002-05-27 Roger Sayle <roger@eyesopen.com>
3853 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3855 2002-05-27 Bo Thorsen <bo@suse.de>
3857 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3860 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3862 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3863 beginning of function (for older apple cc).
3865 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3867 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3868 script entry, and set LD to it when configuring multilibs.
3869 * configure: Rebuilt.
3871 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3873 * configure.in (sparc64-*-netbsd*): Add target.
3874 (sparc-*-netbsdelf*): Likewise.
3875 * configure: Regenerate.
3877 2002-04-28 David S. Miller <davem@redhat.com>
3879 * configure.in, configure: Fix SPARC test in previous change.
3881 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3883 * Makefile.am: Add Linux for S/390 support.
3884 * Makefile.in: Regenerate.
3885 * configure.in: Add Linux for S/390 support.
3886 * configure: Regenerate.
3887 * include/ffi.h.in: Add Linux for S/390 support.
3888 * src/s390/ffi.c: New file from libffi CVS tree.
3889 * src/s390/sysv.S: New file from libffi CVS tree.
3891 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3893 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3895 * src/sparc/v8.S: Use it.
3896 * src/sparc/v9.S: Likewise.
3897 * fficonfig.h.in: Rebuilt.
3898 * configure: Rebuilt.
3900 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3902 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3904 * src/ia64/unix.S: Add unwind information. Fix comments.
3905 Save sp in a way that's compatible with unwind info.
3906 (ffi_call_unix): Correctly restore sp in all cases.
3907 * src/ia64/ffi.c: Add, fix comments.
3909 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3911 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3913 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3915 * configure.in (alpha*-*-netbsd*): Add target.
3916 * configure: Regenerate.
3918 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3920 * src/sparc/v8.S: Add unwind info.
3921 * src/sparc/v9.S: Likewise.
3923 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3925 * configure.in: Enable i*86-*-netbsdelf*.
3926 * configure: Rebuilt.
3928 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3931 * src/mips/n32.s: Delete
3932 * src/mips/o32.s: Delete
3934 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3936 * configure.in: Enable alpha*-*-freebsd*.
3937 * configure: Rebuilt.
3939 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3941 * Makefile.am: libfficonvenience -> libffi_convenience.
3942 * Makefile.in: Rebuilt.
3944 * Makefile.am: Define ffitest_OBJECTS.
3945 * Makefile.in: Rebuilt.
3947 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3948 David Edelsohn <edelsohn@gnu.org>
3950 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3951 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3952 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3953 * Makefile.in: Regenerate.
3954 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3955 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3956 (flush_icache, flush_range): New functions.
3957 (ffi_closure_helper_DARWIN): New function.
3958 * src/powerpc/aix_closure.S: New file.
3959 * src/powerpc/darwin_closure.S: New file.
3961 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3963 * include/ffi.h.in: Add typedef for ffi_arg.
3964 * src/ffitest.c (main): Declare rint with ffi_arg.
3966 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3968 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3969 number of GPRs for floating-point arguments.
3971 2002-01-31 Anthony Green <green@redhat.com>
3973 * configure: Rebuilt.
3974 * configure.in: Replace CHECK_SIZEOF and endian tests with
3975 cross-compiler friendly macros.
3976 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3979 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3981 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3982 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3984 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3986 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3987 (TARGET_SRC_POWERPC_AIX): New.
3988 (POWERPC_AIX): New stanza.
3989 * Makefile.in: Regenerate.
3990 * configure.in: Add AIX case.
3991 * configure: Regenerate.
3992 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3993 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3994 size. Fix "long double" support.
3995 (ffi_call): Add FFI_AIX case.
3996 * src/powerpc/aix.S: New.
3998 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
4000 Implement Darwin PowerPC ABI.
4001 * configure.in: Handle powerpc-*-darwin*.
4002 * Makefile.am: Set source files for POWERPC_DARWIN.
4003 * configure: Rebuilt.
4004 * Makefile.in: Rebuilt.
4005 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
4007 * src/powerpc/darwin.S: New file.
4008 * src/powerpc/ffi_darwin.c: New file.
4010 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
4012 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
4014 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4016 * src/x86/sysv.S: Avoid gas-only .balign directive.
4017 Use C style comments.
4019 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4021 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
4022 Fixes PR bootstrap/3563.
4024 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4026 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
4028 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4030 * configure.in: Recognize sparc*-sun-* host.
4031 * configure: Regenerate.
4033 2001-06-06 Andrew Haley <aph@redhat.com>
4035 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
4037 2001-06-03 Andrew Haley <aph@redhat.com>
4039 * src/alpha/osf.S: Add unwind info.
4040 * src/powerpc/sysv.S: Add unwind info.
4041 * src/powerpc/ppc_closure.S: Likewise.
4043 2000-05-31 Jeff Sturm <jsturm@one-point.com>
4045 * configure.in: Fix AC_ARG_ENABLE usage.
4046 * configure: Rebuilt.
4048 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
4050 * configure.in: Remove warning about beta code.
4051 * configure: Rebuilt.
4053 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
4055 * src/ia64/unix.S: Restore stack pointer when returning from
4057 * src/ia64/ffi.c: Fix typo in comment.
4059 2001-04-18 Jim Wilson <wilson@redhat.com>
4061 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
4062 to eliminate RAW DV.
4064 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
4066 * Makefile.am: Make a libtool convenience library.
4067 * Makefile.in: Rebuilt.
4069 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
4071 * configure.in: Use different syntax for subdirectory creation.
4072 * configure: Rebuilt.
4074 2001-03-27 Jon Beniston <jon@beniston.com>
4076 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
4077 * configure: Rebuilt.
4078 * Makefile.am: Added X86_WIN32 target support.
4079 * Makefile.in: Rebuilt.
4081 * include/ffi.h.in: Added X86_WIN32 target support.
4083 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
4084 * src/types.c: Added X86_WIN32 target support.
4086 * src/x86/win32.S: New file. Based on sysv.S, but with EH
4087 stuff removed and made to work with CygWin's gas.
4089 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
4091 * configure.in: Make target subdirectory in build dir.
4092 * Makefile.am: Override suffix based rules to specify correct output
4094 * Makefile.in: Rebuilt.
4095 * configure: Rebuilt.
4097 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
4099 * src/powerpc/ppc_closure.S: New file.
4100 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
4101 involving long long and register pairs.
4102 (ffi_prep_closure): New function.
4103 (flush_icache): Likewise.
4104 (ffi_closure_helper_SYSV): Likewise.
4105 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
4106 (FFI_TRAMPOLINE_SIZE): Likewise.
4107 (FFI_NATIVE_RAW_API): Likewise.
4108 * Makefile.in: Rebuilt.
4109 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
4110 (TARGET_SRC_POWERPC): Likewise.
4112 2001-03-19 Tom Tromey <tromey@redhat.com>
4114 * Makefile.in: Rebuilt.
4115 * Makefile.am (ffitest_LDFLAGS): New macro.
4117 2001-03-02 Nick Clifton <nickc@redhat.com>
4119 * include/ffi.h.in: Remove RCS ident string.
4120 * include/ffi_mips.h: Remove RCS ident string.
4121 * src/debug.c: Remove RCS ident string.
4122 * src/ffitest.c: Remove RCS ident string.
4123 * src/prep_cif.c: Remove RCS ident string.
4124 * src/types.c: Remove RCS ident string.
4125 * src/alpha/ffi.c: Remove RCS ident string.
4126 * src/alpha/osf.S: Remove RCS ident string.
4127 * src/arm/ffi.c: Remove RCS ident string.
4128 * src/arm/sysv.S: Remove RCS ident string.
4129 * src/mips/ffi.c: Remove RCS ident string.
4130 * src/mips/n32.S: Remove RCS ident string.
4131 * src/mips/o32.S: Remove RCS ident string.
4132 * src/sparc/ffi.c: Remove RCS ident string.
4133 * src/sparc/v8.S: Remove RCS ident string.
4134 * src/sparc/v9.S: Remove RCS ident string.
4135 * src/x86/ffi.c: Remove RCS ident string.
4136 * src/x86/sysv.S: Remove RCS ident string.
4138 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
4140 * include/ffi.h.in: Change sourceware.cygnus.com references to
4143 2000-12-09 Richard Henderson <rth@redhat.com>
4145 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
4146 (ffi_closure_osf_inner): Index rather than increment avalue
4147 and arg_types. Give ffi_closure_osf the raw return value type.
4148 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
4151 2000-12-07 Richard Henderson <rth@redhat.com>
4153 * src/raw_api.c (ffi_translate_args): Fix typo.
4154 (ffi_prep_closure): Likewise.
4156 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
4157 FFI_TRAMPOLINE_SIZE.
4158 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
4159 cif->bytes for new ffi_call_osf implementation.
4160 (ffi_prep_args): Absorb into ...
4161 (ffi_call): ... here. Do all stack allocation here and
4162 avoid a callback function.
4163 (ffi_prep_closure, ffi_closure_osf_inner): New.
4164 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
4165 (ffi_closure_osf): New.
4167 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
4169 * config.guess, config.sub, install-sh: Removed.
4170 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
4171 * Makefile.in: Rebuilt.
4173 * acinclude.m4: Include libtool macros from the top level.
4174 * aclocal.m4, configure: Rebuilt.
4176 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
4178 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
4179 * configure: Rebuilt.
4181 2000-05-11 Scott Bambrough <scottb@netwinder.org>
4183 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
4184 memory correctly. Use conditional instructions, not branches where
4187 2000-05-04 Tom Tromey <tromey@cygnus.com>
4189 * configure: Rebuilt.
4190 * configure.in: Match `arm*-*-linux-*'.
4191 From Chris Dornan <cdornan@arm.com>.
4193 2000-04-28 Jakub Jelinek <jakub@redhat.com>
4195 * Makefile.am (SUBDIRS): Define.
4196 (AM_MAKEFLAGS): Likewise.
4197 (Multilib support.): Add section.
4198 * Makefile.in: Rebuilt.
4199 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
4200 New variables. Set for gcc using -print-multi-lib. Export them
4202 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
4203 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
4205 (extra_compiler_flags_value, extra_compiler_flags): Check these
4206 for extra compiler options which need to be passed down in
4209 2000-04-16 Anthony Green <green@redhat.com>
4211 * configure: Rebuilt.
4212 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
4214 2000-04-14 Jakub Jelinek <jakub@redhat.com>
4216 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
4217 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
4218 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
4219 Replace all void * sizeofs with sizeof(int).
4220 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
4221 different than DOUBLE.
4222 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
4223 (ffi_prep_args_v9): New function.
4224 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
4225 (ffi_V9_return_struct): New function.
4226 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
4227 32bit code (not yet cross-arch calls).
4228 * src/sparc/v8.S: Add struct return delay nop.
4230 * src/sparc/v9.S: New file.
4231 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
4232 is used on sparc64 only for structures larger than 32 bytes.
4233 Pass by reference for structures is done for structure arguments
4234 larger than 16 bytes.
4235 * src/ffitest.c (main): Use 64bit rint on sparc64.
4236 Run long long tests on sparc.
4237 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
4239 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
4241 * configure.in (sparc-*-linux*): New supported target.
4242 (sparc64-*-linux*): Likewise.
4243 * configure: Rebuilt.
4244 * Makefile.am: Add v9.S to SPARC files.
4245 * Makefile.in: Likewise.
4246 (LINK): Surround $(CCLD) into double quotes, so that multilib
4247 compiles work correctly.
4249 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
4251 * configure: Rebuilt.
4252 * configure.in: (i*86-*-solaris*): New libffi target. Patch
4253 proposed by Bryce McKinlay.
4255 2000-03-20 Tom Tromey <tromey@cygnus.com>
4257 * Makefile.in: Hand edit for java_raw_api.lo.
4259 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
4261 * config.guess, config.sub: Update from the gcc tree.
4262 Fix for PR libgcj/168.
4264 2000-03-03 Tom Tromey <tromey@cygnus.com>
4266 * Makefile.in: Fixed ia64 by hand.
4268 * configure: Rebuilt.
4269 * configure.in (--enable-multilib): New option.
4270 (libffi_basedir): New subst.
4271 (AC_OUTPUT): Added multilib code.
4273 2000-03-02 Tom Tromey <tromey@cygnus.com>
4275 * Makefile.in: Rebuilt.
4276 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
4279 2000-02-25 Hans Boehm <boehm@acm.org>
4281 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
4283 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
4285 (ffi_prep_raw_closure): Use ffi_translate_args, not
4286 ffi_closure_translate.
4287 * src/java_raw_api.c: New file.
4288 * src/ffitest.c (closure_test_fn): New function.
4289 (main): Define `rint' as long long on IA64. Added new test when
4290 FFI_CLOSURES is defined.
4291 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
4292 (ffi_abi): Recognize IA64.
4293 (ffi_raw): Added `flt' field.
4294 Added "Java raw API" code.
4295 * configure.in: Recognize ia64.
4296 * Makefile.am (TARGET_SRC_IA64): New macro.
4297 (libffi_la_common_SOURCES): Added java_raw_api.c.
4298 (libffi_la_SOURCES): Define in IA64 case.
4300 2000-01-04 Tom Tromey <tromey@cygnus.com>
4302 * Makefile.in: Rebuilt with newer automake.
4304 1999-12-31 Tom Tromey <tromey@cygnus.com>
4306 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
4308 1999-09-01 Tom Tromey <tromey@cygnus.com>
4310 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
4312 * fficonfig.h.in: Rebuilt.
4313 * configure: Rebuilt.
4314 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
4315 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
4316 * acconfig.h: Don't #undef PACKAGE or VERSION.
4318 1999-08-09 Anthony Green <green@cygnus.com>
4320 * include/ffi.h.in: Try to work around messy header problem
4321 with PACKAGE and VERSION.
4323 * configure: Rebuilt.
4324 * configure.in: Change version to 2.00-beta.
4326 * fficonfig.h.in: Rebuilt.
4327 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
4329 * src/x86/ffi.c (ffi_raw_call): Rename.
4331 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
4333 * src/x86/ffi.c (ffi_closure_SYSV): New function.
4334 (ffi_prep_incoming_args_SYSV): Ditto.
4335 (ffi_prep_closure): Ditto.
4336 (ffi_closure_raw_SYSV): Ditto.
4337 (ffi_prep_raw_closure): More ditto.
4338 (ffi_call_raw): Final ditto.
4340 * include/ffi.h.in: Add definitions for closure and raw API.
4342 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4345 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
4347 * src/raw_api.c: New file.
4349 * include/ffi.h.in (ffi_raw): New type.
4350 (UINT_ARG, SINT_ARG): New defines.
4351 (ffi_closure, ffi_raw_closure): New types.
4352 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4354 * configure.in: Add check for endianness and sizeof void*.
4356 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
4357 instead of directly.
4359 * configure: Rebuilt.
4361 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
4363 * configure.in: Add x86 and powerpc BeOS configurations.
4364 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
4366 1999-05-09 Anthony Green <green@cygnus.com>
4368 * configure.in: Add warning about this being beta code.
4369 Remove src/Makefile.am from the picture.
4370 * configure: Rebuilt.
4372 * Makefile.am: Move logic from src/Makefile.am. Add changes
4373 to support libffi as a target library.
4374 * Makefile.in: Rebuilt.
4376 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
4377 Upgraded to new autoconf, automake, libtool.
4381 * LICENSE: Update copyright date.
4383 * src/Makefile.am, src/Makefile.in: Removed.
4385 1998-11-29 Anthony Green <green@cygnus.com>
4387 * include/ChangeLog: Removed.
4388 * src/ChangeLog: Removed.
4389 * src/mips/ChangeLog: Removed.
4390 * src/sparc/ChangeLog: Remboved.
4391 * src/x86/ChangeLog: Removed.
4393 * ChangeLog.v1: Created.