1 2011-01-26 Dave Korn <dave.korn.cygwin@gmail.com>
4 * configure.ac (AM_LTLDFLAGS): Add -bindir option for windows DLLs.
5 * configure: Regenerate.
7 2010-12-18 Iain Sandoe <iains@gcc.gnu.org>
11 * src/powerpc/darwin_closure.S: Provide Darwin64 implementation,
13 * src/powerpc/ffitarget.h (POWERPC_DARWIN64): New,
14 (FFI_TRAMPOLINE_SIZE): Update for Darwin64.
15 * src/powerpc/darwin.S: Provide Darwin64 implementation,
17 * src/powerpc/ffi_darwin.c: Likewise.
19 2010-12-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
21 * configure.ac (libffi_cv_as_ascii_pseudo_op): Use double
23 (libffi_cv_as_string_pseudo_op): Likewise.
24 * configure: Regenerate.
26 2010-12-03 Chung-Lin Tang <cltang@codesourcery.com>
28 * src/arm/sysv.S (ffi_closure_SYSV): Add UNWIND to .pad directive.
29 (ffi_closure_VFP): Same.
30 (ffi_call_VFP): Move down to before ffi_closure_VFP. Add '.fpu vfp'
33 2010-12-01 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
35 * testsuite/libffi.call/ffitest.h [__sgi] (PRId64, PRIu64): Define.
38 2010-11-29 Richard Henderson <rth@redhat.com>
39 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
41 * src/x86/sysv.S (FDE_ENCODING, FDE_ENCODE): Define.
42 (.eh_frame): Use FDE_ENCODING.
43 (.LASFDE1, .LASFDE2, LASFDE3): Simplify with FDE_ENCODE.
45 2010-11-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
47 * testsuite/lib/libffi-dg.exp: Rename ...
48 * testsuite/lib/libffi.exp: ... to this.
49 * libffi/testsuite/libffi.call/call.exp: Don't load libffi-dg.exp.
50 * libffi/testsuite/libffi.special/special.exp: Likewise.
52 2010-10-28 Chung-Lin Tang <cltang@codesourcery.com>
54 * src/arm/ffi.c (ffi_prep_args): Add VFP register argument handling
55 code, new parameter, and return value. Update comments.
56 (ffi_prep_cif_machdep): Add case for VFP struct return values. Add
57 call to layout_vfp_args().
58 (ffi_call_SYSV): Update declaration.
59 (ffi_call_VFP): New declaration.
60 (ffi_call): Add VFP struct return conditions. Call ffi_call_VFP()
62 (ffi_closure_VFP): New declaration.
63 (ffi_closure_SYSV_inner): Add new vfp_args parameter, update call to
64 ffi_prep_incoming_args_SYSV().
65 (ffi_prep_incoming_args_SYSV): Update parameters. Add VFP argument
67 (ffi_prep_closure_loc): Pass ffi_closure_VFP to trampoline
68 construction under VFP hard-float.
69 (rec_vfp_type_p): New function.
71 (place_vfp_arg): Same.
72 (layout_vfp_args): Same.
73 * src/arm/ffitarget.h (ffi_abi): Add FFI_VFP. Define FFI_DEFAULT_ABI
74 based on __ARM_PCS_VFP.
75 (FFI_EXTRA_CIF_FIELDS): Define for adding VFP hard-float specific
77 (FFI_TYPE_STRUCT_VFP_FLOAT): Define internally used type code.
78 (FFI_TYPE_STRUCT_VFP_DOUBLE): Same.
79 * src/arm/sysv.S (ffi_call_SYSV): Change call of ffi_prep_args() to
80 direct call. Move function pointer load upwards.
81 (ffi_call_VFP): New function.
82 (ffi_closure_VFP): Same.
84 * testsuite/lib/libffi-dg.exp (check-flags): New function.
85 (dg-skip-if): New function.
86 * testsuite/libffi.call/cls_double_va.c: Skip if target is arm*-*-*
87 and compiler options include -mfloat-abi=hard.
88 * testsuite/libffi.call/cls_longdouble_va.c: Same.
90 2010-10-01 Jakub Jelinek <jakub@redhat.com>
93 * src/x86/ffi64.c (ffi_prep_cif_machdep): Ensure cif->bytes is
95 * testsuite/libffi.call/many2.c: New test.
97 2010-08-20 Mark Wielaard <mjw@redhat.com>
99 * src/closures.c (open_temp_exec_file_mnt): Check if getmntent_r
102 2010-08-09 Andreas Tobler <andreast@fgznet.ch>
104 * configure.ac: Add target powerpc64-*-freebsd*.
105 * configure: Regenerate.
106 * testsuite/libffi.call/cls_align_longdouble_split.c: Pass
107 -mlong-double-128 only to linux targets.
108 * testsuite/libffi.call/cls_align_longdouble_split2.c: Likewise.
109 * testsuite/libffi.call/cls_longdouble.c: Likewise.
110 * testsuite/libffi.call/huge_struct.c: Likewise.
112 2010-07-10 Evan Phoenix <evan@fallingsnow.net>
114 * src/closures.c (selinux_enabled_check): Fix strncmp usage bug.
116 2010-07-07 Dan HorĂ¡k <dan@danny.cz>
118 * include/ffi.h.in: Protect #define with #ifndef.
119 * src/powerpc/ffitarget.h: Ditto.
120 * src/s390/ffitarget.h: Ditto.
121 * src/sparc/ffitarget.h: Ditto.
123 2010-07-07 Neil Roberts <neil@linux.intel.com>
125 * src/x86/sysv.S (ffi_call_SYSV): Align the stack pointer to
128 2010-07-02 Jakub Jelinek <jakub@redhat.com>
130 * Makefile.am (AM_MAKEFLAGS): Pass also mandir to submakes.
131 * Makefile.in: Regenerated.
133 2010-05-19 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
135 * configure.ac (libffi_cv_as_x86_pcrel): Check for illegal in as
137 (libffi_cv_as_ascii_pseudo_op): Check for .ascii.
138 (libffi_cv_as_string_pseudo_op): Check for .string.
139 * configure: Regenerate.
140 * fficonfig.h.in: Regenerate.
141 * src/x86/sysv.S (.eh_frame): Use .ascii, .string or error.
143 2010-05-05 Michael Kohler <michaelkohler@live.com>
145 * src/dlmalloc.c (dlfree): Fix spelling.
146 * src/ia64/ffi.c (ffi_prep_cif_machdep): Ditto.
147 * configure.ac: Ditto.
148 * configure: Rebuilt.
150 2010-04-13 Dan Witte <dwitte@mozilla.com>
152 * msvcc.sh: Build with -W3 instead of -Wall.
153 * src/powerpc/ffi_darwin.c: Remove build warnings.
154 * src/x86/ffi.c: Ditto.
155 * src/x86/ffitarget.h: Ditto.
157 2010-04-12 Dan Witte <dwitte@mozilla.com>
158 Walter Meinl <wuno@lsvw.de>
160 * configure.ac: Add OS/2 support.
161 * configure: Rebuilt.
162 * src/closures.c: Ditto.
163 * src/dlmalloc.c: Ditto.
164 * src/x86/win32.S: Ditto.
166 2010-04-07 Jakub Jelinek <jakub@redhat.com>
168 * testsuite/libffi.call/err_bad_abi.c: Remove unused args variable.
170 2010-04-02 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
172 * Makefile.in: Regenerate.
173 * aclocal.m4: Regenerate.
174 * include/Makefile.in: Regenerate.
175 * man/Makefile.in: Regenerate.
176 * testsuite/Makefile.in: Regenerate.
178 2010-03-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
180 * configure.ac (libffi_cv_as_x86_64_unwind_section_type): New test.
181 * configure: Regenerate.
182 * fficonfig.h.in: Regenerate.
183 * libffi/src/x86/unix64.S (.eh_frame)
184 [HAVE_AS_X86_64_UNWIND_SECTION_TYPE]: Use @unwind section type.
186 2010-03-14 Matthias Klose <doko@ubuntu.com>
188 * src/x86/ffi64.c: Fix typo in comment.
189 * src/x86/ffi.c: Use /* ... */ comment style.
191 2010-02-24 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
193 * doc/libffi.texi (The Closure API): Fix typo.
194 * doc/libffi.info: Remove.
196 2010-02-15 Matthias Klose <doko@ubuntu.com>
198 * src/arm/sysv.S (__ARM_ARCH__): Define for processor
201 2010-01-15 Anthony Green <green@redhat.com>
203 * README: Add notes on building with Microsoft Visual C++.
205 2010-01-15 Daniel Witte <dwitte@mozilla.com>
207 * msvcc.sh: New file.
209 * src/x86/win32.S: Port assembly routines to MSVC and #ifdef.
210 * src/x86/ffi.c: Tweak function declaration and remove excess
212 * include/ffi.h.in: Add __declspec(align(8)) to typedef struct
215 * src/x86/ffi.c: Merge ffi_call_SYSV and ffi_call_STDCALL into new
216 function ffi_call_win32 on X86_WIN32.
217 * src/x86/win32.S (ffi_call_SYSV): Rename to ffi_call_win32.
218 (ffi_call_STDCALL): Remove.
220 * src/prep_cif.c (ffi_prep_cif): Move stack space allocation code
221 to ffi_prep_cif_machdep for x86.
222 * src/x86/ffi.c (ffi_prep_cif_machdep): To here.
224 2010-01-15 Oliver Kiddle <okiddle@yahoo.co.uk>
226 * src/x86/ffitarget.h (ffi_abi): Check for __i386 and __amd64 for
227 Sun Studio compiler compatibility.
229 2010-01-12 Conrad Irwin <conrad.irwin@gmail.com>
231 * doc/libffi.texi: Add closure example.
233 2010-01-07 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
236 * testsuite/libffi.call/ffitest.h [__alpha__ && __osf__] (PRIdLL,
237 PRIuLL, PRId64, PRIu64, PRIuPTR): Define.
238 * testsuite/libffi.call/cls_align_sint64.c: Add -Wno-format on
240 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
241 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
242 * testsuite/libffi.call/return_ll1.c: Likewise.
243 * testsuite/libffi.call/stret_medium2.c: Likewise.
244 * testsuite/libffi.special/ffitestcxx.h (allocate_mmap): Cast
245 MAP_FAILED to char *.
247 2010-01-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
249 * src/mips/n32.S: Use .abicalls and .eh_frame with __GNUC__.
251 2009-12-31 Anthony Green <green@redhat.com>
253 * README: Update for libffi 3.0.9.
255 2009-12-27 Matthias Klose <doko@ubuntu.com>
257 * configure.ac (HAVE_LONG_DOUBLE): Define for mips when
259 * configure: Rebuilt.
261 2009-12-26 Anthony Green <green@redhat.com>
263 * testsuite/libffi.call/cls_longdouble_va.c: Mark as xfail for
265 * testsuite/libffi.call/cls_double_va.c: Ditto.
267 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
269 * testsuite/libffi.call/ffitest.h: Conditionally include stdint.h
271 * testsuite/libffi.special/unwindtest.cc: Ditto.
273 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
275 * configure.ac: Add amd64-*-openbsd*.
276 * configure: Rebuilt.
277 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Link
278 openbsd programs with -lpthread.
280 2009-12-26 Anthony Green <green@redhat.com>
282 * testsuite/libffi.call/cls_double_va.c,
283 testsuite/libffi.call/cls_longdouble.c,
284 testsuite/libffi.call/cls_longdouble_va.c,
285 testsuite/libffi.call/cls_pointer.c,
286 testsuite/libffi.call/cls_pointer_stack.c: Remove xfail for
287 mips*-*-* and arm*-*-*.
288 * testsuite/libffi.call/cls_align_longdouble_split.c,
289 testsuite/libffi.call/cls_align_longdouble_split2.c,
290 testsuite/libffi.call/stret_medium2.c,
291 testsuite/libffi.call/stret_medium.c,
292 testsuite/libffi.call/stret_large.c,
293 testsuite/libffi.call/stret_large2.c: Remove xfail for arm*-*-*.
295 2009-12-31 Kay Tietz <ktietz70@googlemail.com>
297 * testsuite/libffi.call/ffitest.h,
298 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRuLL): Fix
301 2009-12-31 Carlo Bramini <carlo.bramix@libero.it>
303 * configure.ac (AM_LTLDFLAGS): Define for windows hosts.
304 * Makefile.am (libffi_la_LDFLAGS): Add AM_LTLDFLAGS.
305 * configure: Rebuilt.
306 * Makefile.in: Rebuilt.
308 2009-12-31 Anthony Green <green@redhat.com>
311 * testsuite/libffi.call/huge_struct.c: New test case from Blake
314 2009-12-28 David Edelsohn <edelsohn@gnu.org>
316 * src/powerpc/ffi_darwin.c (ffi_prep_args): Copy abi and nargs to
318 (aix_adjust_aggregate_sizes): New function.
319 (ffi_prep_cif_machdep): Call it.
321 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
323 * configure.ac: Define FFI_MMAP_EXEC_WRIT for the given targets.
324 * configure: Regenerate.
325 * fficonfig.h.in: Likewise.
326 * src/closures.c: Remove the FFI_MMAP_EXEC_WRIT definition for
329 2009-12-26 Andreas Schwab <schwab@linux-m68k.org>
331 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Advance intarg_count
332 when a float arguments is passed in memory.
333 (ffi_closure_helper_SYSV): Mark general registers as used up when
334 a 64bit or soft-float long double argument is passed in memory.
336 2009-12-25 Matthias Klose <doko@ubuntu.com>
338 * man/ffi_call.3: Fix #include in examples.
339 * doc/libffi.texi: Add dircategory.
341 2009-12-25 Frank Everdij <f.p.x.everdij@tudelft.nl>
343 * include/ffi.h.in: Placed '__GNUC__' ifdef around
344 '__attribute__((aligned(8)))' in ffi_closure, fixes compile for
346 * include/ffi_common.h: Added '__sgi' define to non
347 '__attribute__((__mode__()))' integer typedefs.
348 * src/mips/ffi.c (ffi_call, ffi_closure_mips_inner_O32,
349 ffi_closure_mips_inner_N32): Added 'defined(_MIPSEB)' to BE check.
350 (ffi_closure_mips_inner_O32, ffi_closure_mips_inner_N32): Added
351 FFI_LONGDOUBLE support and alignment(N32 only).
352 * src/mips/ffitarget.h: Corrected '#include <sgidefs.h>' for IRIX and
353 fixed non '__attribute__((__mode__()))' integer typedefs.
354 * src/mips/n32.S: Put '#ifdef linux' around '.abicalls' and '.eh_frame'
355 since they are Linux/GNU Assembler specific.
357 2009-12-25 Bradley Smith <brad@brad-smith.co.uk>
359 * configure.ac, Makefile.am, src/avr32/ffi.c,
360 src/avr32/ffitarget.h,
361 src/avr32/sysv.S: Add AVR32 port.
362 * configure, Makefile.in: Rebuilt.
364 2009-12-21 Andreas Tobler <a.tobler@schweiz.org>
366 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
367 * configure: Regenerate.
369 2009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
371 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
373 2009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
375 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
378 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
380 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
383 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
385 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
387 2009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
390 * src/closures.c [X86_64 && __sun__ && __svr4__]
391 (FFI_MMAP_EXEC_WRIT): Define.
393 2009-12-08 David Daney <ddaney@caviumnetworks.com>
395 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
396 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
397 * testsuite/libffi.call/stret_large.c: Same.
398 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
399 * testsuite/libffi.call/stret_large2.c: Same.
400 * testsuite/libffi.call/stret_medium2.c: Same.
402 2009-12-07 David Edelsohn <edelsohn@gnu.org>
404 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
407 2009-12-05 David Edelsohn <edelsohn@gnu.org>
409 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
411 * src/powerpc/aix_closure.S: Same.
413 2009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
415 * Makefile.in: Regenerate.
416 * configure: Regenerate.
417 * include/Makefile.in: Regenerate.
418 * man/Makefile.in: Regenerate.
419 * testsuite/Makefile.in: Regenerate.
421 2009-12-04 David Edelsohn <edelsohn@gnu.org>
423 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
426 2009-12-04 Uros Bizjak <ubizjak@gmail.com>
429 * src/x86/ffi64.c (classify_argument): Update from
430 gcc/config/i386/i386.c.
431 (ffi_closure_unix64_inner): Do not use the address of two consecutive
432 SSE registers directly.
433 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
434 for x86_64 linux targets.
436 2009-12-04 David Edelsohn <edelsohn@gnu.org>
438 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
439 pfr for long double split between fpr13 and stack.
441 2009-12-03 David Edelsohn <edelsohn@gnu.org>
443 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
444 fparg_count twice for long double.
446 2009-12-03 David Edelsohn <edelsohn@gnu.org>
449 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
451 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
453 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
454 Remove xfails for x86 linux targets.
456 2009-12-02 David Edelsohn <edelsohn@gnu.org>
458 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
461 2009-12-01 David Edelsohn <edelsohn@gnu.org>
463 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
464 register usage. Call ffi_prep_args directly. Add long double
465 return value support.
466 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
467 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
468 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
469 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
471 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
473 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
474 Allocate result area between params and FPRs.
476 2009-11-30 David Edelsohn <edelsohn@gnu.org>
479 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
481 * src/powerpc/aix.S: Implement AIX64 version.
482 * src/powerpc/aix_closure.S: Implement AIX64 version.
483 (ffi_closure_ASM): Use extsb, lha and displament addresses.
484 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
486 (ffi_prep_cif_machdep): Same.
488 (ffi_closure_helper_DARWIN): Same.
490 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
493 * testsuite/libffi.call/testclosure.c: New test.
495 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
497 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
498 assembly version use of ___chkstk.
500 2009-09-23 Matthias Klose <doko@ubuntu.com>
502 PR libffi/40242, PR libffi/41443
503 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
504 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
505 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
506 Change the conditionals to __SOFTFP__ || __ARM_EABI__
507 for -mfloat-abi=softfp to work.
509 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
511 PR testsuite/32843 (strikes again)
512 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
513 enable proper extension on char and short.
515 2009-09-15 David Daney <ddaney@caviumnetworks.com>
517 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
518 handling for FFI_TYPE_POINTER.
519 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
520 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
521 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
522 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
523 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
524 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
525 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
526 structure and pointer returns.
527 (ffi_closure_N32): Add handling for pointer returns.
528 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
529 calc_n32_return_struct_flags): Handle soft-float.
530 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
531 (ffi_call_N32): Declare proper argument types.
532 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
535 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
537 * configure.ac (AC_PREREQ): Bump to 2.64.
539 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
541 * Makefile.am (install-html, install-pdf): Remove.
542 * Makefile.in: Regenerate.
544 * Makefile.in: Regenerate.
545 * aclocal.m4: Regenerate.
546 * configure: Regenerate.
547 * fficonfig.h.in: Regenerate.
548 * include/Makefile.in: Regenerate.
549 * man/Makefile.in: Regenerate.
550 * testsuite/Makefile.in: Regenerate.
552 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
554 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
556 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
559 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
560 return types for X86_WIN32.
561 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
562 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
563 _ffi_closure_STDCALL): Likewise.
565 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
566 (dlmmap, dlmunmap): Also use these functions on Cygwin.
568 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
573 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
576 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
578 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
579 to "" before adding paths. (This reinstates an assignment that
580 was removed by my 2009-06-30 commit, but changes the initial
581 value from "." to "".)
583 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
586 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
587 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
590 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
592 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
593 to ld_library_path. Use add_path. Add just find_libgcc_s
594 to ld_library_path, not every libgcc multilib directory.
596 2009-06-16 Wim Lewis <wiml@hhhh.org>
598 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
599 supposed to be callee-saved.
600 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
601 return buffer for odd-size structs.
603 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
606 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
607 allow_stack_execute for Darwin.
609 2009-06-16 Andrew Haley <aph@redhat.com>
611 * configure.ac (TARGETDIR): Add missing blank lines.
612 * configure: Regenerate.
614 2009-06-16 Andrew Haley <aph@redhat.com>
616 * testsuite/libffi.call/cls_align_sint64.c,
617 testsuite/libffi.call/cls_align_uint64.c,
618 testsuite/libffi.call/cls_longdouble_va.c,
619 testsuite/libffi.call/cls_ulonglong.c,
620 testsuite/libffi.call/return_ll1.c,
621 testsuite/libffi.call/stret_medium2.c: Fix printf format
623 * testsuite/libffi.call/ffitest.h,
624 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
626 2009-06-15 Andrew Haley <aph@redhat.com>
628 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
629 * testsuite/libffi.call/err_bad_abi.c: Likewise.
631 2009-06-12 Andrew Haley <aph@redhat.com>
633 * Makefile.am: Remove info_TEXINFOS.
635 2009-06-12 Andrew Haley <aph@redhat.com>
637 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
638 testsuite/libffi.call/cls_align_uint64.c,
639 testsuite/libffi.call/cls_ulonglong.c,
640 testsuite/libffi.call/return_ll1.c,
641 testsuite/libffi.call/stret_medium2.c: Fix printf format
643 testsuite/libffi.special/unwindtest.cc: include stdint.h.
645 2009-06-11 Timothy Wall <twall@users.sf.net>
650 include/ffi_common.h,
655 src/x86/win64.S (new),
656 README: Added win64 support (mingw or MSVC)
660 testsuite/Makefile.in,
662 aclocal.m4: Regenerated
663 * ltcf-c.sh: properly escape cygwin/w32 path
664 * man/ffi_call.3: Clarify size requirements for return value.
665 * src/x86/ffi64.c: Fix filename in comment.
666 * src/x86/win32.S: Remove unused extern.
668 * testsuite/libffi.call/closure_fn0.c,
669 testsuite/libffi.call/closure_fn1.c,
670 testsuite/libffi.call/closure_fn2.c,
671 testsuite/libffi.call/closure_fn3.c,
672 testsuite/libffi.call/closure_fn4.c,
673 testsuite/libffi.call/closure_fn5.c,
674 testsuite/libffi.call/closure_fn6.c,
675 testsuite/libffi.call/closure_stdcall.c,
676 testsuite/libffi.call/cls_12byte.c,
677 testsuite/libffi.call/cls_16byte.c,
678 testsuite/libffi.call/cls_18byte.c,
679 testsuite/libffi.call/cls_19byte.c,
680 testsuite/libffi.call/cls_1_1byte.c,
681 testsuite/libffi.call/cls_20byte.c,
682 testsuite/libffi.call/cls_20byte1.c,
683 testsuite/libffi.call/cls_24byte.c,
684 testsuite/libffi.call/cls_2byte.c,
685 testsuite/libffi.call/cls_3_1byte.c,
686 testsuite/libffi.call/cls_3byte1.c,
687 testsuite/libffi.call/cls_3byte2.c,
688 testsuite/libffi.call/cls_4_1byte.c,
689 testsuite/libffi.call/cls_4byte.c,
690 testsuite/libffi.call/cls_5_1_byte.c,
691 testsuite/libffi.call/cls_5byte.c,
692 testsuite/libffi.call/cls_64byte.c,
693 testsuite/libffi.call/cls_6_1_byte.c,
694 testsuite/libffi.call/cls_6byte.c,
695 testsuite/libffi.call/cls_7_1_byte.c,
696 testsuite/libffi.call/cls_7byte.c,
697 testsuite/libffi.call/cls_8byte.c,
698 testsuite/libffi.call/cls_9byte1.c,
699 testsuite/libffi.call/cls_9byte2.c,
700 testsuite/libffi.call/cls_align_double.c,
701 testsuite/libffi.call/cls_align_float.c,
702 testsuite/libffi.call/cls_align_longdouble.c,
703 testsuite/libffi.call/cls_align_longdouble_split.c,
704 testsuite/libffi.call/cls_align_longdouble_split2.c,
705 testsuite/libffi.call/cls_align_pointer.c,
706 testsuite/libffi.call/cls_align_sint16.c,
707 testsuite/libffi.call/cls_align_sint32.c,
708 testsuite/libffi.call/cls_align_sint64.c,
709 testsuite/libffi.call/cls_align_uint16.c,
710 testsuite/libffi.call/cls_align_uint32.c,
711 testsuite/libffi.call/cls_align_uint64.c,
712 testsuite/libffi.call/cls_dbls_struct.c,
713 testsuite/libffi.call/cls_double.c,
714 testsuite/libffi.call/cls_double_va.c,
715 testsuite/libffi.call/cls_float.c,
716 testsuite/libffi.call/cls_longdouble.c,
717 testsuite/libffi.call/cls_longdouble_va.c,
718 testsuite/libffi.call/cls_multi_schar.c,
719 testsuite/libffi.call/cls_multi_sshort.c,
720 testsuite/libffi.call/cls_multi_sshortchar.c,
721 testsuite/libffi.call/cls_multi_uchar.c,
722 testsuite/libffi.call/cls_multi_ushort.c,
723 testsuite/libffi.call/cls_multi_ushortchar.c,
724 testsuite/libffi.call/cls_pointer.c,
725 testsuite/libffi.call/cls_pointer_stack.c,
726 testsuite/libffi.call/cls_schar.c,
727 testsuite/libffi.call/cls_sint.c,
728 testsuite/libffi.call/cls_sshort.c,
729 testsuite/libffi.call/cls_uchar.c,
730 testsuite/libffi.call/cls_uint.c,
731 testsuite/libffi.call/cls_ulonglong.c,
732 testsuite/libffi.call/cls_ushort.c,
733 testsuite/libffi.call/err_bad_abi.c,
734 testsuite/libffi.call/err_bad_typedef.c,
735 testsuite/libffi.call/float2.c,
736 testsuite/libffi.call/huge_struct.c,
737 testsuite/libffi.call/nested_struct.c,
738 testsuite/libffi.call/nested_struct1.c,
739 testsuite/libffi.call/nested_struct10.c,
740 testsuite/libffi.call/nested_struct2.c,
741 testsuite/libffi.call/nested_struct3.c,
742 testsuite/libffi.call/nested_struct4.c,
743 testsuite/libffi.call/nested_struct5.c,
744 testsuite/libffi.call/nested_struct6.c,
745 testsuite/libffi.call/nested_struct7.c,
746 testsuite/libffi.call/nested_struct8.c,
747 testsuite/libffi.call/nested_struct9.c,
748 testsuite/libffi.call/problem1.c,
749 testsuite/libffi.call/return_ldl.c,
750 testsuite/libffi.call/return_ll1.c,
751 testsuite/libffi.call/stret_large.c,
752 testsuite/libffi.call/stret_large2.c,
753 testsuite/libffi.call/stret_medium.c,
754 testsuite/libffi.call/stret_medium2.c,
755 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
756 of checking for MMAP. Use intptr_t instead of long casts.
758 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
760 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
761 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
762 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
764 2009-06-09 Andrew Haley <aph@redhat.com>
766 * src/x86/freebsd.S: Add missing file.
768 2009-06-08 Andrew Haley <aph@redhat.com>
770 Import from libffi 3.0.8:
772 * doc/libffi.texi: New file.
773 * doc/libffi.info: Likewise.
774 * doc/stamp-vti: Likewise.
775 * man/Makefile.am: New file.
776 * man/ffi_call.3: New file.
778 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
780 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
782 * configure.ac: Bump version to 3.0.8.
783 parisc*-*-linux*: Add.
784 i386-*-freebsd* | i386-*-openbsd*: Add.
785 powerpc-*-beos*: Add.
786 AM_CONDITIONAL X86_FREEBSD: Add.
787 AC_CONFIG_FILES: Add man/Makefile.
789 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
791 2009-06-08 Andrew Haley <aph@redhat.com>
793 * README: Import from libffi 3.0.8.
795 2009-06-08 Andrew Haley <aph@redhat.com>
797 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
798 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
799 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
800 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
802 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
803 * testsuite/libffi.call/stret_medium.c: Likewise.
804 * testsuite/libffi.call/stret_large2.c: Likewise.
805 * testsuite/libffi.call/stret_large.c: Likewise.
807 2008-12-26 Timothy Wall <twall@users.sf.net>
809 * testsuite/libffi.call/cls_longdouble.c,
810 testsuite/libffi.call/cls_longdouble_va.c,
811 testsuite/libffi.call/cls_align_longdouble.c,
812 testsuite/libffi.call/cls_align_longdouble_split.c,
813 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
814 failures on x86_64 cygwin/mingw.
816 2008-12-22 Timothy Wall <twall@users.sf.net>
818 * testsuite/libffi.call/closure_fn0.c,
819 testsuite/libffi.call/closure_fn1.c,
820 testsuite/libffi.call/closure_fn2.c,
821 testsuite/libffi.call/closure_fn3.c,
822 testsuite/libffi.call/closure_fn4.c,
823 testsuite/libffi.call/closure_fn5.c,
824 testsuite/libffi.call/closure_fn6.c,
825 testsuite/libffi.call/closure_loc_fn0.c,
826 testsuite/libffi.call/closure_stdcall.c,
827 testsuite/libffi.call/cls_align_pointer.c,
828 testsuite/libffi.call/cls_pointer.c,
829 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
830 pointer to integer (intptr_t).
831 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
833 2008-07-24 Anthony Green <green@redhat.com>
835 * testsuite/libffi.call/cls_dbls_struct.c,
836 testsuite/libffi.call/cls_double_va.c,
837 testsuite/libffi.call/cls_longdouble.c,
838 testsuite/libffi.call/cls_longdouble_va.c,
839 testsuite/libffi.call/cls_pointer.c,
840 testsuite/libffi.call/cls_pointer_stack.c,
841 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
844 2008-03-04 Anthony Green <green@redhat.com>
848 * testsuite/libffi.call/cls_align_longdouble_split2.c
849 testsuite/libffi.call/cls_align_longdouble_split.c
850 testsuite/libffi.call/cls_dbls_struct.c
851 testsuite/libffi.call/cls_double_va.c
852 testsuite/libffi.call/cls_longdouble.c
853 testsuite/libffi.call/cls_longdouble_va.c
854 testsuite/libffi.call/cls_pointer.c
855 testsuite/libffi.call/cls_pointer_stack.c
856 testsuite/libffi.call/err_bad_abi.c
857 testsuite/libffi.call/err_bad_typedef.c
858 testsuite/libffi.call/stret_large2.c
859 testsuite/libffi.call/stret_large.c
860 testsuite/libffi.call/stret_medium2.c
861 testsuite/libffi.call/stret_medium.c: New tests from Apple.
863 2009-06-05 Andrew Haley <aph@redhat.com>
865 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
868 2009-06-04 Andrew Haley <aph@redhat.com>
870 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
873 2008-02-26 Anthony Green <green@redhat.com>
874 Thomas Heller <theller@ctypes.org>
876 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
879 2008-02-03 Timothy Wall <twall@users.sf.net>
881 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
882 offset based on code pointer, not data pointer.
884 2008-01-31 Timothy Wall <twall@users.sf.net>
886 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
888 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
890 * src/x86/win32.S: Add assembly for stdcall closure.
891 * src/x86/ffi.c: Initialize stdcall closure trampoline.
893 2009-06-04 Andrew Haley <aph@redhat.com>
895 * include/ffi.h.in: Change void (*)() to void (*)(void).
896 * src/x86/ffi.c: Likewise.
898 2009-06-04 Andrew Haley <aph@redhat.com>
900 * src/powerpc/ppc_closure.S: Insert licence header.
901 * src/powerpc/linux64_closure.S: Likewise.
902 * src/m68k/sysv.S: Likewise.
904 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
905 * src/powerpc/ffi.c: Likewise.
906 * src/powerpc/ffi_darwin.c: Likewise.
907 * src/m32r/ffi.c: Likewise.
908 * src/sh64/ffi.c: Likewise.
909 * src/x86/ffi64.c: Likewise.
910 * src/alpha/ffi.c: Likewise.
911 * src/alpha/osf.S: Likewise.
912 * src/frv/ffi.c: Likewise.
913 * src/s390/ffi.c: Likewise.
914 * src/pa/ffi.c: Likewise.
915 * src/pa/hpux32.S: Likewise.
916 * src/ia64/unix.S: Likewise.
917 * src/ia64/ffi.c: Likewise.
918 * src/sparc/ffi.c: Likewise.
919 * src/mips/ffi.c: Likewise.
920 * src/sh/ffi.c: Likewise.
922 2008-02-15 David Daney <ddaney@avtrex.com>
924 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
925 Define (conditionally), and use it to include cachectl.h.
926 (ffi_prep_closure_loc): Fix cache flushing.
927 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
929 2009-06-04 Andrew Haley <aph@redhat.com>
935 src/powerpc/ffitarget.h,
937 src/sh64/ffitarget.h,
948 src/alpha/ffitarget.h,
951 src/m68k/ffitarget.h,
954 src/s390/ffitarget.h,
956 src/cris/ffitarget.h,
961 src/ia64/ffitarget.h,
964 src/ia64/ia64_flags.h,
968 src/sparc/ffitarget.h,
971 src/mips/ffitarget.h,
978 src/sh/sysv.S: Update license text.
980 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
982 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
983 (.eh_frame): Add FDE for it.
985 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
987 * configure.ac: Also check if assembler supports pc-relative
988 relocs on X86_WIN32 targets.
989 * configure: Regenerate.
990 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
991 (_ffi_call_SYSV): Add missing function type symbol .def and
992 add EH markup labels.
993 (_ffi_call_STDCALL): Likewise.
994 (_ffi_closure_SYSV): Likewise.
995 (_ffi_closure_raw_SYSV): Likewise.
996 (.eh_frame): Add hand-crafted EH data.
998 2009-04-09 Jakub Jelinek <jakub@redhat.com>
1000 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
1001 version 3 of the GNU General Public License and to point readers
1002 at the COPYING3 file and the FSF's license web page.
1003 * testsuite/libffi.call/call.exp: Likewise.
1004 * testsuite/libffi.special/special.exp: Likewise.
1006 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1008 * configure: Regenerate.
1010 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1013 * configure.ac (HAVE_AS_X86_PCREL): New test.
1014 * configure: Regenerate.
1015 * fficonfig.h.in: Regenerate.
1016 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
1017 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
1018 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
1019 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
1020 * src/x86/unix64.S (.Lstore_table): Move to .text section.
1021 (.Lload_table): Likewise.
1022 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
1024 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1026 * configure: Regenerate.
1028 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
1030 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
1031 signed/unsigned int8/16 return values.
1032 * src/sparc/v8.S (ffi_call_v8): Likewise.
1033 (ffi_closure_v8): Likewise.
1035 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
1036 Steve Ellcey <sje@cup.hp.com>
1038 * configure: Regenerate for new libtool.
1039 * Makefile.in: Ditto.
1040 * include/Makefile.in: Ditto.
1041 * aclocal.m4: Ditto.
1043 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
1045 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
1046 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
1047 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
1048 Adjust copyright notice.
1049 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
1050 register or two register to use for FFI_SYSV structs.
1051 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
1052 (ffi_closure_helper_SYSV): Fix the return type for
1053 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
1054 Adjust copyright notice.
1056 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
1058 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
1061 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1063 * configure: Regenerate.
1064 * include/Makefile.in: Regenerate.
1065 * testsuite/Makefile.in: Regenerate.
1067 2008-06-07 Joseph Myers <joseph@codesourcery.com>
1069 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
1070 powerpc-*-beos*): Remove.
1071 * configure: Regenerate.
1073 2008-05-09 Julian Brown <julian@codesourcery.com>
1075 * Makefile.am (LTLDFLAGS): New.
1076 (libffi_la_LDFLAGS): Use above.
1077 * Makefile.in: Regenerate.
1079 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
1082 * aclocal.m4: Regenerate.
1083 * configure: Regenerate.
1085 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
1087 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
1088 * src/sh64/sysv.S: Likewise.
1090 2008-03-26 Daniel Jacobowitz <dan@debian.org>
1092 * src/arm/sysv.S: Fix ARM comment marker.
1094 2008-03-26 Jakub Jelinek <jakub@redhat.com>
1096 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
1097 * src/s390/sysv.S: Likewise.
1098 * src/powerpc/ppc_closure.S: Likewise.
1099 * src/powerpc/sysv.S: Likewise.
1100 * src/x86/unix64.S: Likewise.
1101 * src/x86/sysv.S: Likewise.
1102 * src/sparc/v8.S: Likewise.
1103 * src/sparc/v9.S: Likewise.
1104 * src/m68k/sysv.S: Likewise.
1105 * src/arm/sysv.S: Likewise.
1107 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1109 * aclocal.m4: Regenerate.
1110 * configure: Likewise.
1111 * Makefile.in: Likewise.
1112 * include/Makefile.in: Likewise.
1113 * testsuite/Makefile.in: Likewise.
1115 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
1116 Andreas Tobler <a.tobler@schweiz.org>
1118 * configure.ac: Add amd64-*-freebsd* target.
1119 * configure: Regenerate.
1121 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
1124 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
1127 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
1130 2008-01-24 David Edelsohn <edelsohn@gnu.org>
1132 * configure: Regenerate.
1134 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
1136 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
1138 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
1141 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
1142 signed/unsigned int8/16 for X86_DARWIN.
1143 Updated copyright info.
1144 Handle one and two byte structs with special cif->flags.
1145 * src/x86/ffitarget.h: Add special types for one and two byte structs.
1146 Updated copyright info.
1147 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
1149 Remove code to pop args from the stack after call.
1150 Special-case signed/unsigned for int8/16, one and two byte structs.
1151 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1152 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1154 Updated copyright info.
1156 2007-12-08 David Daney <ddaney@avtrex.com>
1158 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
1159 SUBU, add with ADDU and use smaller code sequences.
1161 2007-12-07 David Daney <ddaney@avtrex.com>
1163 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
1166 2007-12-06 David Daney <ddaney@avtrex.com>
1168 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
1170 (ffi_java_raw): New typedef.
1171 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
1172 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
1174 (ffi_java_raw_closure) : Same.
1175 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
1177 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
1178 FFI_SIZEOF_JAVA_RAW.
1179 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
1180 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
1181 sizeof(ffi_java_raw) for alignment calculations.
1182 (ffi_java_ptrarray_to_raw): Same.
1183 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
1184 if FFI_SIZEOF_JAVA_RAW == 4.
1185 (ffi_java_raw_to_rvalue): Same.
1186 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
1187 (ffi_java_translate_args): Same.
1188 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
1190 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
1192 2007-12-06 David Daney <ddaney@avtrex.com>
1194 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
1197 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
1200 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
1201 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
1202 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
1203 set the NUM_FPR_ARG_REGISTERS according to.
1204 Add support for potential soft-float support under hard-float
1206 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
1207 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
1208 to the FFI_LINUX_SOFT_FLOAT ABI.
1209 (ffi_prep_cif_machdep): Likewise.
1210 (ffi_closure_helper_SYSV): Likewise.
1211 * src/powerpc/ppc_closure.S: Make sure not to store float/double
1212 on archs where __NO_FPRS__ is true.
1213 Add FFI_TYPE_UINT128 support.
1214 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
1215 Adjust copyright notice.
1217 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
1219 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
1220 * include/ffi_common.h: ... here.
1223 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
1225 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
1226 * src/powerpc/linux64.S: Likewise.
1227 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
1228 * testsuite/libffi.call/return_ldl.c: New test.
1230 2007-09-04 <aph@redhat.com>
1232 * src/arm/sysv.S (UNWIND): New.
1233 (Whole file): Conditionally compile unwinder directives.
1234 * src/arm/sysv.S: Add unwinder directives.
1236 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
1237 Only treat r0 as a struct address if we're actually returning a
1239 Only copy the bytes that are actually within a struct.
1240 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
1241 is returned in r0, not passed by address.
1242 (ffi_call): Allocate a word-sized temporary for the case where
1243 a composite is returned in r0.
1244 (ffi_prep_incoming_args_SYSV): Align as necessary.
1246 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
1248 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
1249 directly using the sys_cacheflush syscall.
1251 2007-07-27 Andrew Haley <aph@redhat.com>
1253 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
1255 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
1257 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
1258 * configure.ac: Likewise.
1259 * Makefile.in: Regenerate.
1260 * include/Makefile.in: Likewise.
1261 * testsuite/Makefile.in: Likewise.
1262 * configure: Likewise.
1264 2007-08-24 David Daney <ddaney@avtrex.com>
1266 * testsuite/libffi.call/return_sl.c: New test.
1268 2007-08-10 David Daney <ddaney@avtrex.com>
1270 * testsuite/libffi.call/cls_multi_ushort.c,
1271 testsuite/libffi.call/cls_align_uint16.c,
1272 testsuite/libffi.call/nested_struct1.c,
1273 testsuite/libffi.call/nested_struct3.c,
1274 testsuite/libffi.call/cls_7_1_byte.c,
1275 testsuite/libffi.call/nested_struct5.c,
1276 testsuite/libffi.call/cls_double.c,
1277 testsuite/libffi.call/nested_struct7.c,
1278 testsuite/libffi.call/cls_sint.c,
1279 testsuite/libffi.call/nested_struct9.c,
1280 testsuite/libffi.call/cls_20byte1.c,
1281 testsuite/libffi.call/cls_multi_sshortchar.c,
1282 testsuite/libffi.call/cls_align_sint64.c,
1283 testsuite/libffi.call/cls_3byte2.c,
1284 testsuite/libffi.call/cls_multi_schar.c,
1285 testsuite/libffi.call/cls_multi_uchar.c,
1286 testsuite/libffi.call/cls_19byte.c,
1287 testsuite/libffi.call/cls_9byte1.c,
1288 testsuite/libffi.call/cls_align_float.c,
1289 testsuite/libffi.call/closure_fn1.c,
1290 testsuite/libffi.call/problem1.c,
1291 testsuite/libffi.call/closure_fn3.c,
1292 testsuite/libffi.call/cls_sshort.c,
1293 testsuite/libffi.call/closure_fn5.c,
1294 testsuite/libffi.call/cls_align_double.c,
1295 testsuite/libffi.call/nested_struct.c,
1296 testsuite/libffi.call/cls_2byte.c,
1297 testsuite/libffi.call/nested_struct10.c,
1298 testsuite/libffi.call/cls_4byte.c,
1299 testsuite/libffi.call/cls_6byte.c,
1300 testsuite/libffi.call/cls_8byte.c,
1301 testsuite/libffi.call/cls_multi_sshort.c,
1302 testsuite/libffi.call/cls_align_sint16.c,
1303 testsuite/libffi.call/cls_align_uint32.c,
1304 testsuite/libffi.call/cls_20byte.c,
1305 testsuite/libffi.call/cls_float.c,
1306 testsuite/libffi.call/nested_struct2.c,
1307 testsuite/libffi.call/cls_5_1_byte.c,
1308 testsuite/libffi.call/nested_struct4.c,
1309 testsuite/libffi.call/cls_24byte.c,
1310 testsuite/libffi.call/nested_struct6.c,
1311 testsuite/libffi.call/cls_64byte.c,
1312 testsuite/libffi.call/nested_struct8.c,
1313 testsuite/libffi.call/cls_uint.c,
1314 testsuite/libffi.call/cls_multi_ushortchar.c,
1315 testsuite/libffi.call/cls_schar.c,
1316 testsuite/libffi.call/cls_uchar.c,
1317 testsuite/libffi.call/cls_align_uint64.c,
1318 testsuite/libffi.call/cls_ulonglong.c,
1319 testsuite/libffi.call/cls_align_longdouble.c,
1320 testsuite/libffi.call/cls_1_1byte.c,
1321 testsuite/libffi.call/cls_12byte.c,
1322 testsuite/libffi.call/cls_3_1byte.c,
1323 testsuite/libffi.call/cls_3byte1.c,
1324 testsuite/libffi.call/cls_4_1byte.c,
1325 testsuite/libffi.call/cls_6_1_byte.c,
1326 testsuite/libffi.call/cls_16byte.c,
1327 testsuite/libffi.call/cls_18byte.c,
1328 testsuite/libffi.call/closure_fn0.c,
1329 testsuite/libffi.call/cls_9byte2.c,
1330 testsuite/libffi.call/closure_fn2.c,
1331 testsuite/libffi.call/closure_fn4.c,
1332 testsuite/libffi.call/cls_ushort.c,
1333 testsuite/libffi.call/closure_fn6.c,
1334 testsuite/libffi.call/cls_5byte.c,
1335 testsuite/libffi.call/cls_align_pointer.c,
1336 testsuite/libffi.call/cls_7byte.c,
1337 testsuite/libffi.call/cls_align_sint32.c,
1338 testsuite/libffi.special/unwindtest_ffi_call.cc,
1339 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
1341 2007-08-10 David Daney <ddaney@avtrex.com>
1344 * configure.ac: Don't treat mips64 as a special case.
1345 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
1346 * configure: Regenerate
1347 * Makefile.in: Ditto.
1348 * fficonfig.h.in: Ditto.
1349 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
1350 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
1351 (FFI_DEFAULT_ABI): Set for n64 case.
1352 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
1353 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
1354 (ffi_closure_N32): New function.
1355 (.eh_frame): New section
1356 * src/mips/o32.S: Clean up comments.
1357 (ffi_closure_O32): Pass ffi_closure parameter in $12.
1358 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
1359 _MIPS_SIM == _ABIN32 throughout.
1360 (FFI_MIPS_STOP_HERE): New, use in place of
1362 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1363 to support n32/n64 ABIs.
1364 (calc_n32_struct_flags): Rewrite.
1365 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1366 position of flag bits.
1367 (ffi_prep_cif_machdep): Rewrite n32 portion.
1368 (ffi_call): Enable for n64. Add special handling for small structure
1370 (ffi_prep_closure_loc): Add n32 and n64 support.
1371 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1372 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1374 2007-08-08 David Daney <ddaney@avtrex.com>
1376 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1377 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1379 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1380 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1381 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1382 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1383 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1384 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1385 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1386 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1387 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1388 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1389 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1390 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1391 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1392 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1393 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1394 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1395 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1396 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1397 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1398 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1399 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1400 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1401 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1402 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1403 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1404 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1405 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1406 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1407 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1408 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1409 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1411 2007-08-07 Andrew Haley <aph@redhat.com>
1413 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1416 2007-08-06 Andrew Haley <aph@redhat.com>
1419 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1420 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1423 2007-08-02 David Daney <ddaney@avtrex.com>
1425 * testsuite/libffi.call/return_ul.c (main): Define return type as
1426 ffi_arg. Use proper printf conversion specifier.
1428 2007-07-30 Andrew Haley <aph@redhat.com>
1431 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1432 signed/unsigned int8/16.
1433 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1435 Remove code to pop args from the stack after call.
1436 Special-case signed/unsigned int8/16.
1437 * testsuite/libffi.call/return_sc.c (main): Revert.
1439 2007-07-26 Richard Guenther <rguenther@suse.de>
1442 * testsuite/libffi.call/return_sc.c (main): Verify call
1443 result as signed char, not ffi_arg.
1445 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1447 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1448 * configure: Regenerate.
1450 2007-07-11 David Daney <ddaney@avtrex.com>
1452 * src/mips/ffi.c: Don't include sys/cachectl.h.
1453 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1456 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1458 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1459 from (ffi_prep_closure): ... this.
1460 (FFI_INIT_TRAMPOLINE): Adjust.
1462 2005-12-31 Phil Blundell <pb@reciva.com>
1464 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1465 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1466 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1467 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1468 (FFI_CLOSURES): Enable closure support.
1470 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1472 * testsuite/libffi.call/cls_multi_ushort.c,
1473 testsuite/libffi.call/cls_align_uint16.c,
1474 testsuite/libffi.call/nested_struct1.c,
1475 testsuite/libffi.call/nested_struct3.c,
1476 testsuite/libffi.call/cls_7_1_byte.c,
1477 testsuite/libffi.call/cls_double.c,
1478 testsuite/libffi.call/nested_struct5.c,
1479 testsuite/libffi.call/nested_struct7.c,
1480 testsuite/libffi.call/cls_sint.c,
1481 testsuite/libffi.call/nested_struct9.c,
1482 testsuite/libffi.call/cls_20byte1.c,
1483 testsuite/libffi.call/cls_multi_sshortchar.c,
1484 testsuite/libffi.call/cls_align_sint64.c,
1485 testsuite/libffi.call/cls_3byte2.c,
1486 testsuite/libffi.call/cls_multi_schar.c,
1487 testsuite/libffi.call/cls_multi_uchar.c,
1488 testsuite/libffi.call/cls_19byte.c,
1489 testsuite/libffi.call/cls_9byte1.c,
1490 testsuite/libffi.call/cls_align_float.c,
1491 testsuite/libffi.call/closure_fn1.c,
1492 testsuite/libffi.call/problem1.c,
1493 testsuite/libffi.call/closure_fn3.c,
1494 testsuite/libffi.call/cls_sshort.c,
1495 testsuite/libffi.call/closure_fn5.c,
1496 testsuite/libffi.call/cls_align_double.c,
1497 testsuite/libffi.call/cls_2byte.c,
1498 testsuite/libffi.call/nested_struct.c,
1499 testsuite/libffi.call/nested_struct10.c,
1500 testsuite/libffi.call/cls_4byte.c,
1501 testsuite/libffi.call/cls_6byte.c,
1502 testsuite/libffi.call/cls_8byte.c,
1503 testsuite/libffi.call/cls_multi_sshort.c,
1504 testsuite/libffi.call/cls_align_uint32.c,
1505 testsuite/libffi.call/cls_align_sint16.c,
1506 testsuite/libffi.call/cls_float.c,
1507 testsuite/libffi.call/cls_20byte.c,
1508 testsuite/libffi.call/cls_5_1_byte.c,
1509 testsuite/libffi.call/nested_struct2.c,
1510 testsuite/libffi.call/cls_24byte.c,
1511 testsuite/libffi.call/nested_struct4.c,
1512 testsuite/libffi.call/nested_struct6.c,
1513 testsuite/libffi.call/cls_64byte.c,
1514 testsuite/libffi.call/nested_struct8.c,
1515 testsuite/libffi.call/cls_uint.c,
1516 testsuite/libffi.call/cls_multi_ushortchar.c,
1517 testsuite/libffi.call/cls_schar.c,
1518 testsuite/libffi.call/cls_uchar.c,
1519 testsuite/libffi.call/cls_align_uint64.c,
1520 testsuite/libffi.call/cls_ulonglong.c,
1521 testsuite/libffi.call/cls_align_longdouble.c,
1522 testsuite/libffi.call/cls_1_1byte.c,
1523 testsuite/libffi.call/cls_12byte.c,
1524 testsuite/libffi.call/cls_3_1byte.c,
1525 testsuite/libffi.call/cls_3byte1.c,
1526 testsuite/libffi.call/cls_4_1byte.c,
1527 testsuite/libffi.call/cls_6_1_byte.c,
1528 testsuite/libffi.call/cls_16byte.c,
1529 testsuite/libffi.call/cls_18byte.c,
1530 testsuite/libffi.call/closure_fn0.c,
1531 testsuite/libffi.call/cls_9byte2.c,
1532 testsuite/libffi.call/closure_fn2.c,
1533 testsuite/libffi.call/closure_fn4.c,
1534 testsuite/libffi.call/cls_ushort.c,
1535 testsuite/libffi.call/closure_fn6.c,
1536 testsuite/libffi.call/cls_5byte.c,
1537 testsuite/libffi.call/cls_align_pointer.c,
1538 testsuite/libffi.call/cls_7byte.c,
1539 testsuite/libffi.call/cls_align_sint32.c,
1540 testsuite/libffi.special/unwindtest_ffi_call.cc,
1541 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1543 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1545 * aclocal.m4: Regenerated.
1547 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1549 * configure: Regenerate.
1551 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1553 * Makefile.in: Regenerate.
1554 * configure: Regenerate.
1555 * aclocal.m4: Regenerate.
1556 * include/Makefile.in: Regenerate.
1557 * testsuite/Makefile.in: Regenerate.
1559 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1561 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1562 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1563 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1564 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1565 (FFI_CLOSURES): Enable closure support.
1567 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1569 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1570 * configure: Regenerate.
1571 * fficonfig.h.in: Regenerate.
1572 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1573 CFI_OFFSET,CFI_DEF_CFA): New macros.
1574 (ffi_call_SYSV): Add callframe annotation.
1576 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1578 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1579 numerous test suite failures.
1580 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1582 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1584 * Makefile.am (EXTRA_DIST): Bring up to date.
1585 * Makefile.in: Regenerate.
1586 * src/frv/eabi.S: Remove RCS keyword.
1588 2007-04-06 Richard Henderson <rth@redhat.com>
1590 * configure.ac: Tidy target case.
1591 (HAVE_LONG_DOUBLE): Allow the target to override.
1592 * configure: Regenerate.
1593 * include/ffi.h.in: Don't define ffi_type_foo if
1594 LIBFFI_HIDE_BASIC_TYPES is defined.
1595 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1597 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1598 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1599 (ffi_type_longdouble): Special case for Alpha. Don't define
1600 if long double == double.
1602 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1603 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1604 (ffi_call, ffi_closure_osf_inner): Likewise.
1605 (ffi_closure_osf_inner): Likewise. Mark hidden.
1606 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1607 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1608 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1609 (load_table): Handle 128-bit long double.
1611 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1613 2007-04-06 Tom Tromey <tromey@redhat.com>
1616 * README: Fixed bug in example.
1618 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1620 * src/closures.c: Include sys/statfs.h.
1621 (_GNU_SOURCE): Define on Linux.
1622 (FFI_MMAP_EXEC_SELINUX): Define.
1623 (selinux_enabled): New variable.
1624 (selinux_enabled_check): New function.
1625 (is_selinux_enabled): Define.
1628 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1630 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1631 Use 'volatile float sum' to create sum of floats to avoid false
1632 negative due to excess precision on ix86 targets.
1635 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1637 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1639 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1641 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1643 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1644 (ffi_prep_closure_loc): New.
1645 (ffi_prep_raw_closure_loc): New.
1646 (ffi_prep_java_raw_closure_loc): New.
1647 * src/closures.c: New file.
1648 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1649 Replace sflags with exec_offset.
1650 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1651 sub_segment_exec_offset): New macros.
1652 (get_segment_flags, set_segment_flags, check_segment_merge): New
1654 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1655 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1656 destroy_mspace): Use new macros.
1657 (sys_alloc): Silence warning.
1658 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1659 * Makefile.in: Rebuilt.
1660 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1661 terms of ffi_prep_closure_loc.
1662 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1664 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1666 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1668 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1669 the renamed version.
1670 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1671 (ffi_prep_closure): ... this.
1672 * src/pa/ffi.c: Likewise.
1673 * src/cris/ffi.c: Likewise. Adjust.
1674 * src/frv/ffi.c: Likewise.
1675 * src/ia64/ffi.c: Likewise.
1676 * src/mips/ffi.c: Likewise.
1677 * src/powerpc/ffi_darwin.c: Likewise.
1678 * src/s390/ffi.c: Likewise.
1679 * src/sh/ffi.c: Likewise.
1680 * src/sh64/ffi.c: Likewise.
1681 * src/sparc/ffi.c: Likewise.
1682 * src/x86/ffi64.c: Likewise.
1683 * src/x86/ffi.c: Likewise.
1684 (FFI_INIT_TRAMPOLINE): Adjust.
1685 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1686 (ffi_prep_raw_closure): ... this.
1687 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1688 (ffi_prep_closure): ... this.
1689 (flush_icache): Adjust.
1691 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1693 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1696 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1698 * Makefile.am: Add dummy install-pdf target.
1699 * Makefile.in: Regenerate
1701 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1703 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1704 ffi_closure_helper_SYSV): Add long double handling.
1706 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1708 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1709 immediately after bctrl instruction.
1711 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1713 * Makefile.am (all-recursive, install-recursive,
1714 mostlyclean-recursive, clean-recursive, distclean-recursive,
1715 maintainer-clean-recursive): Add missing targets.
1716 * Makefile.in: Rebuilt.
1718 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1720 * configure.ac: Add TARGET for x86_64-*-darwin*.
1721 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1723 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1724 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1725 * configure: Regenerate.
1726 * Makefile.in: Regenerate.
1727 * include/Makefile.in: Regenerate.
1728 * testsuite/Makefile.in: Regenerate.
1729 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1732 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1734 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1737 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1739 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1740 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1742 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1743 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1745 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1746 Sandro Tolaini <tolaini@libero.it>
1748 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1750 * configure: Regenerated.
1751 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1752 (EXTRA_DIST): Add src/x86/darwin.S.
1753 * Makefile.in: Regenerated.
1754 * include/Makefile.in: Regenerated.
1755 * testsuite/Makefile.in: Regenerated.
1757 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1758 X86_WIN32, and additionally align stack to 16 bytes.
1759 * src/x86/darwin.S: New, based on sysv.S.
1760 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1762 2006-09-12 David Daney <ddaney@avtrex.com>
1765 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1766 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1767 * aclocal.m4: Regenerated for automake 1.9.6.
1768 * Makefile.in: Regenerated.
1769 * include/Makefile.in: Regenerated.
1770 * testsuite/Makefile.in: Regenerated.
1772 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1774 * include/ffi_common.h (struct): Revert accidental commit.
1776 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1778 * include/ffi_common.h: Remove lint directives.
1779 * include/ffi.h.in: Likewise.
1781 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1783 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1784 for 32-bit architectures.
1785 * testsuite/libffi.call/return_ul.c: New test case.
1787 2006-07-19 David Daney <ddaney@avtrex.com>
1789 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1790 xfail remains for mips64.
1792 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1794 * Makefile.am: Add install-html target. Add install-html to .PHONY
1795 * Makefile.in: Regenerate.
1796 * aclocal.m4: Regenerate.
1797 * include/Makefile.in: Regenerate.
1798 * testsuite/Makefile.in: Regenerate.
1800 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1802 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1805 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1807 * README: Remove notice about 'Crazy Comments'.
1808 * src/debug.c: Remove lint directives. Cleanup white spaces.
1809 * src/java_raw_api.c: Likewise.
1810 * src/prep_cif.c: Likewise.
1811 * src/raw_api.c: Likewise.
1812 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1814 * src/arm/ffi.c: Remove lint directives.
1815 * src/m32r/ffi.c: Likewise.
1816 * src/pa/ffi.c: Likewise.
1817 * src/powerpc/ffi.c: Likewise.
1818 * src/powerpc/ffi_darwin.c: Likewise.
1819 * src/sh/ffi.c: Likewise.
1820 * src/sh64/ffi.c: Likewise.
1821 * src/x86/ffi.c: Likewise.
1822 * testsuite/libffi.call/float2.c: Likewise.
1823 * testsuite/libffi.call/promotion.c: Likewise.
1824 * testsuite/libffi.call/struct1.c: Likewise.
1826 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1828 * src/pa/hpux32.S: Correct unwind offset calculation for
1830 * src/pa/linux.S: Likewise.
1832 2006-04-12 James E Wilson <wilson@specifix.com>
1835 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1836 (hfa_type_load): Call stf_spill.
1837 (hfa_type_store): Call ldf_fill.
1838 (ffi_call): Adjust calls to above routines. Add local temps for
1841 2006-04-10 Matthias Klose <doko@debian.org>
1843 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1844 directory names containing underscores.
1846 2006-04-07 James E Wilson <wilson@specifix.com>
1848 * testsuite/libffi.call/float4.c: New testcase.
1850 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1851 Andreas Tobler <a.tobler@schweiz.ch>
1853 * Makefile.am: Add PA_HPUX port.
1854 * Makefile.in: Regenerate.
1855 * include/Makefile.in: Likewise.
1856 * testsuite/Makefile.in: Likewise.
1857 * configure.ac: Add PA_HPUX rules.
1858 * configure: Regenerate.
1859 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1860 Add PA_HPUX and PA64_HPUX.
1861 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1862 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1863 (FFI_TYPE_SMALL_STRUCT2): Define.
1864 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1865 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1866 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1867 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1868 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1869 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1870 * src/pa/ffi.c (ROUND_DOWN): Delete.
1871 (fldw, fstw, fldd, fstd): Use '__asm__'.
1872 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1873 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1874 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1875 Simplify incrementing of stack slot variable. Change type of local
1876 'n' to unsigned int.
1877 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1879 (ffi_prep_cif_machdep): Likewise.
1880 (ffi_call): Likewise.
1881 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1882 return type to ffi_status. Simplify incrementing of stack slot
1883 variable. Only copy floating point argument registers when PA_LINUX
1884 is true. Reformat debug statement.
1885 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1886 FFI_TYPE_SMALL_STRUCT8.
1887 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1889 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1890 Add nops to cache flush. Add trampoline for PA_HPUX.
1891 * src/pa/hpux32.S: New file.
1892 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1893 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1894 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1895 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1896 argument type checks so that common argument types appear first.
1897 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1898 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1900 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1902 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1903 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1904 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1905 FFI_TYPE_LONGDOUBLE.
1906 (ffi_prep_args64): Assert using IBM extended double.
1907 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1908 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1909 (ffi_call): Handle FFI_LINUX.
1910 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1911 gpr3 return pointer as for struct return. Handle FFI_LINUX
1912 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1914 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1915 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1916 Don't use r6 as pointer to results, instead use sp offset. Don't
1917 make a special call to load lr with case table address, instead
1918 use offset from previous call.
1919 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1920 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1923 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1925 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1926 passed with FP registers correctly.
1927 (ffi_closure_helper_SYSV): Likewise.
1928 * src/sh64/sysv.S: Likewise.
1930 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1932 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1933 args and userdata unused.
1934 (closure_test_fn1): Mark cif and userdata unused.
1935 (main): Remove unused res.
1937 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1939 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1940 -O2, -O3, -Os and the warning flags -W -Wall.
1941 * testsuite/libffi.special/special.exp: Likewise.
1942 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1943 unused parameter unused for gcc or else do nothing.
1944 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1945 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1946 and userdata unused.
1947 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1948 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1949 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1950 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1951 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1952 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1953 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1954 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1955 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1956 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1957 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1958 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1959 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1960 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1961 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1962 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1963 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1964 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1965 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1966 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1967 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1968 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1969 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1970 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1972 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1974 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1976 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1977 void* to avoid compiler warning.
1979 (cls_struct_align_gn): Mark cif and userdata unused.
1980 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1982 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1984 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1986 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1988 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1990 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1991 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1992 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1994 (main): Cast res_call to silence gcc.
1995 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1997 (main): Cast res_call to silence gcc.
1998 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
2000 (main): Cast res_call to silence gcc.
2001 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
2003 (main): Cast res_call to silence gcc.
2004 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
2006 (main): Cast res_call to silence gcc.
2007 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
2009 (main): Cast res_call to silence gcc.
2010 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
2012 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
2013 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
2015 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
2016 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
2018 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
2019 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
2021 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
2022 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
2024 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
2025 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
2026 and userdata unused.
2027 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
2029 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
2030 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
2031 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
2032 Cleanup white spaces.
2033 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
2034 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
2035 cif and userdata unused.
2036 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
2038 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
2039 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
2040 formatters to silence gcc.
2041 (B_gn): Mark cif and userdata unused.
2042 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
2044 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
2045 (B_gn): Mark cif and userdata unused.
2046 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
2048 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
2049 (B_gn): Mark cif and userdata unused.
2050 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
2052 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
2053 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
2054 * testsuite/libffi.call/problem1.c (stub): Likewise.
2055 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
2057 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
2058 in the last commit for this test case in the test case itself.
2059 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
2061 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
2062 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
2063 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
2064 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
2065 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
2066 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
2068 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
2070 * src/sh/sysv.S: Fix register numbers in the FDE for
2073 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
2075 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
2076 declaration to avoid a false negative on ix86. See PR323.
2078 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
2080 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
2081 and cast integer to void * if needed. Update the pointer to
2082 the FP register saved area correctly.
2084 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
2086 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
2088 * testsuite/libffi.call/nested_struct4.c: Likewise.
2090 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
2092 * testsuite/libffi.call/return_dbl.c: New test case.
2093 * testsuite/libffi.call/return_dbl1.c: Likewise.
2094 * testsuite/libffi.call/return_dbl2.c: Likewise.
2095 * testsuite/libffi.call/return_fl.c: Likewise.
2096 * testsuite/libffi.call/return_fl1.c: Likewise.
2097 * testsuite/libffi.call/return_fl2.c: Likewise.
2098 * testsuite/libffi.call/return_fl3.c: Likewise.
2099 * testsuite/libffi.call/closure_fn6.c: Likewise.
2101 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
2103 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
2104 here to be used by other test cases too.
2106 * testsuite/libffi.call/nested_struct10.c: New test case.
2107 * testsuite/libffi.call/nested_struct9.c: Likewise.
2108 * testsuite/libffi.call/nested_struct8.c: Likewise.
2109 * testsuite/libffi.call/nested_struct7.c: Likewise.
2110 * testsuite/libffi.call/nested_struct6.c: Likewise.
2111 * testsuite/libffi.call/nested_struct5.c: Likewise.
2112 * testsuite/libffi.call/nested_struct4.c: Likewise.
2114 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2116 * configure.ac: Enable libffi for sparc64-*-freebsd*.
2117 * configure: Rebuilt.
2119 2006-01-18 Jakub Jelinek <jakub@redhat.com>
2121 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
2122 instead do the shifting inline.
2123 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
2124 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
2125 and 8 byte structs, for the remaining struct sizes don't call
2126 __lshrdi3, instead do the shifting inline.
2128 2005-12-07 Thiemo Seufer <ths@networkno.de>
2130 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
2131 missing parentheses.
2132 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
2133 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
2134 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
2135 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
2136 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
2138 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
2140 (ffi_prep_closure): Improve trampoline instruction scheduling.
2141 (ffi_closure_mips_inner_O32): Fix endianness bugs.
2143 2005-12-03 Alan Modra <amodra@bigpond.net.au>
2145 * src/powerpc/ffi.c: Formatting.
2146 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
2147 (ffi_prep_args64): Likewise.
2149 2005-09-30 Geoffrey Keating <geoffk@apple.com>
2151 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
2152 darwin, use -shared-libgcc not -lgcc_s, and explain why.
2154 2005-09-26 Tom Tromey <tromey@redhat.com>
2156 * testsuite/libffi.call/float1.c (value_type): New typedef.
2157 (CANARY): New define.
2158 (main): Check for result buffer overflow.
2159 * src/powerpc/linux64.S: Handle linux64 long double returns.
2160 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
2161 (ffi_prep_cif_machdep): Handle linux64 long double returns.
2163 2005-08-25 Alan Modra <amodra@bigpond.net.au>
2166 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
2168 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
2170 2005-08-11 Jakub Jelinek <jakub@redhat.com>
2172 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
2173 (AH_BOTTOM): Add FFI_HIDDEN definition.
2174 * configure: Rebuilt.
2175 * fficonfig.h.in: Rebuilt.
2176 * src/powerpc/ffi.c (hidden): Remove.
2177 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
2178 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
2179 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
2180 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
2181 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
2182 add FFI_HIDDEN to its prototype.
2183 (ffi_closure_SYSV_inner): New.
2184 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2185 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2187 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
2190 * configure: Rebuilt.
2191 * configure.ac: Handle i*86-*-gnu*.
2193 2005-08-09 Jakub Jelinek <jakub@redhat.com>
2195 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
2196 DW_CFA_offset_extended_sf rather than
2197 DW_CFA_GNU_negative_offset_extended.
2198 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
2200 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
2202 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
2204 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
2205 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
2206 partially on register.
2207 (ffi_closure_helper_SYSV): Likewise.
2208 (ffi_prep_cif_machdep): Don't set too many cif->flags.
2210 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
2212 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
2214 * src/sh64/ffi.c (simple_type): Remove.
2215 (return_type): Handle small structures correctly.
2216 (ffi_prep_args): Likewise.
2217 (ffi_call): Likewise.
2218 (ffi_closure_helper_SYSV): Likewise.
2219 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
2220 Emit position independent code if PIC and remove wrong datalabel
2221 prefixes from EH data.
2223 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
2225 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
2226 * Makefile.in: Regenerate.
2227 * include/Makefile.in: Likewise.
2228 * testsuite/Makefile.in: Likewise.
2229 * configure.ac: Add POWERPC_FREEBSD rules.
2230 * configure: Regenerate.
2231 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
2232 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
2233 * src/powerpc/ffi.c: Add flags to handle small structure returns
2235 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
2237 (ffi_closure_helper_SYSV): Likewise.
2238 * src/powerpc/ppc_closure.S: Add return types for small structures.
2239 * src/powerpc/sysv.S: Add bits to handle small structures for
2242 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
2244 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
2245 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
2246 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
2248 2005-07-05 Randolph Chung <tausq@debian.org>
2250 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
2251 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
2252 structures. Kill compilation warnings.
2253 (ffi_closure_inner_LINUX): Print return values as hex in debug
2254 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
2255 Properly handle 5-7 byte structure returns.
2256 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
2257 (FFI_TYPE_SMALL_STRUCT2): Remove.
2258 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
2259 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
2260 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
2261 (checksmst1, checksmst2): Remove.
2262 (checksmst3): Optimize handling of 3-byte struct returns.
2263 (checksmst567): Properly handle 5-7 byte struct returns.
2265 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2268 * src/mips/n32.S: Enforce PIC code.
2269 * src/mips/o32.S: Likewise.
2271 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2273 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
2274 * configure: Regenerate.
2276 2005-06-01 Alan Modra <amodra@bigpond.net.au>
2278 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
2279 to call ffi_closure_helper_SYSV. Append @local instead.
2280 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
2282 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
2284 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
2285 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
2286 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
2287 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
2288 include/Makefile.in, testsuite/Makefile.in: Regenerate.
2290 2005-05-09 Mike Stump <mrs@apple.com>
2292 * configure: Regenerate.
2294 2005-05-08 Richard Henderson <rth@redhat.com>
2297 * src/alpha/osf.S: Update unwind into to match code.
2299 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
2300 Richard Henderson <rth@redhat.com>
2302 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
2304 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
2305 (ffi_prep_closure): Set carry bit if sse-used flag set.
2306 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
2307 Only load sse registers if ssecount non-zero.
2308 (ffi_closure_unix64): Only save sse registers if carry set on entry.
2310 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
2312 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
2313 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
2314 * configure: Regenerate.
2316 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
2318 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
2319 have Tcl8.3-compatible intermediate variable.
2321 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
2322 Hans-Peter Nilsson <hp@axis.com>
2324 * Makefile.am: Add CRIS support.
2325 * configure.ac: Likewise.
2326 * Makefile.in, configure, testsuite/Makefile.in,
2327 include/Makefile.in: Regenerate.
2328 * src/cris: New directory.
2329 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
2330 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
2332 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
2333 \r?\n in output tests.
2335 2005-04-12 Mike Stump <mrs@apple.com>
2337 * configure: Regenerate.
2339 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
2341 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
2343 2005-03-30 Steve Ellcey <sje@cup.hp.com>
2345 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
2347 * src/ia64/unix.S (ffi_closure_unix): Extend gp
2348 to 64 bits in ILP32 mode.
2349 Load 64 bits even for short data.
2351 2005-03-23 Mike Stump <mrs@apple.com>
2353 * src/powerpc/darwin.S: Update for -m64 multilib.
2354 * src/powerpc/darwin_closure.S: Likewise.
2356 2005-03-21 Zack Weinberg <zack@codesourcery.com>
2358 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
2359 Do not set tool_include_dir.
2360 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
2362 * include/Makefile.am: Set gcc_version and toollibffidir.
2363 * include/Makefile.in: Regenerate.
2365 2005-02-22 Andrew Haley <aph@redhat.com>
2367 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2368 odd-numbered register pairs for 64-bit integer types.
2370 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2373 * testsuite/libffi.call/return_ll1.c: New test case.
2375 2005-02-11 Janis Johnson <janis187@us.ibm.com>
2377 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2378 * testsuite/libffi.call/float.c: Ditto.
2379 * testsuite/libffi.call/float2.c: Ditto.
2380 * testsuite/libffi.call/float3.c: Ditto.
2382 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2384 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2386 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2388 * testsuite/libffi.special/special.exp (cxx_options): Add
2391 2004-12-31 Richard Henderson <rth@redhat.com>
2393 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2394 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2395 offset parameters with a type parameter; deduce size and structure
2396 alignment. Update all users.
2398 2004-12-31 Richard Henderson <rth@redhat.com>
2400 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2401 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
2402 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
2403 into ffi_prep_closure.
2404 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2407 2004-12-27 Richard Henderson <rth@redhat.com>
2409 * src/x86/unix64.S: Fix typo in unwind info.
2411 2004-12-25 Richard Henderson <rth@redhat.com>
2413 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2414 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2415 (merge_classes): Check for it.
2417 (classify_argument): Pass byte_offset by value; perform all updates
2419 (examine_argument): Add classes argument; handle
2420 X86_64_COMPLEX_X87_CLASS.
2421 (ffi_prep_args): Merge into ...
2422 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2423 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2424 (ffi_fill_return_value): Remove.
2425 (ffi_prep_closure): Remove dead assert.
2426 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2427 Rewrite to use struct register_args instead of va_list. Create
2428 flags for handling structure returns.
2429 * src/x86/unix64.S: Remove dead strings.
2430 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2431 stack frame with ffi_call. Handle structure returns properly.
2432 (float2sse, floatfloat2sse, double2sse): Remove.
2433 (sse2float, sse2double, sse2floatfloat): Remove.
2434 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2435 to handle structure returns properly.
2437 2004-12-08 David Edelsohn <edelsohn@gnu.org>
2439 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2441 * Makefile.in: Regenerated.
2443 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2445 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2446 * configure, aclocal.m4, Makefile.in: Regenerate.
2447 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2449 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2451 * configure: Regenerate for libtool change.
2453 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2455 * configure: Regenerate for libtool reversion.
2457 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2459 * configure: Regenerate for libtool change.
2461 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2463 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2465 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2467 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2468 of jal. Use an absolute encoding for the frame information.
2470 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2472 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2473 * acinclude.m4: Delete logic for sincludes.
2474 * aclocal.m4, Makefile.in, configure: Regenerate.
2475 * include/Makefile: Likewise.
2476 * testsuite/Makefile: Likewise.
2478 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2480 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2481 on a 8-byte boundary.
2482 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2484 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2486 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2487 long long values. Round stack allocation to a multiple of 8 bytes
2488 for ATPCS compatibility.
2489 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2490 names. Handle returning long long types. Add Thumb and interworking
2491 support. Improve soft-float code.
2493 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2495 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2496 (libffi_exit): New function.
2497 (libffi_init): Build the testglue wrapper if needed.
2499 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2502 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2504 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2506 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2508 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2510 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2511 * testsuite/libffi.call/float3.c: New test case.
2513 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2515 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2516 the function returning a structure pointed with R2.
2517 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2518 the structure return value if T bit set. Emit position
2519 independent code and EH data if PIC.
2521 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2523 * Makefile.am: Add m32r support.
2524 * configure.ac: Likewise.
2525 * Makefile.in: Regenerate.
2526 * confiugre: Regenerate.
2527 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2528 (uint64, sint64, double, longdouble)
2529 * src/m32r: New directory.
2530 * src/m32r/ffi.c: New file.
2531 * src/m32r/sysv.S: Likewise.
2532 * src/m32r/ffitarget.h: Likewise.
2534 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2536 * testsuite/libffi.call/negint.c: New test case.
2538 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2541 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2542 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2543 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2546 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2548 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2549 * testsuite/libffi.call/promotion.c: Likewise.
2550 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2552 * testsuite/libffi.call/return_sc.c: Likewise.
2553 * testsuite/libffi.call/return_uc.c: Likewise.
2555 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2557 * src/powerpc/darwin.S: Fix comments and identation.
2558 * src/powerpc/darwin_closure.S: Likewise.
2560 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2562 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2563 (ffi_prep_args): Handle longdouble arguments.
2564 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2566 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2567 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2569 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2570 * src/types.c: Defined longdouble size and alignment for darwin.
2572 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2574 * src/powerpc/aix.S: Remove whitespaces.
2575 * src/powerpc/aix_closure.S: Likewise.
2576 * src/powerpc/asm.h: Likewise.
2577 * src/powerpc/ffi.c: Likewise.
2578 * src/powerpc/ffitarget.h: Likewise.
2579 * src/powerpc/linux64.S: Likewise.
2580 * src/powerpc/linux64_closure.S: Likewise.
2581 * src/powerpc/ppc_closure.S: Likewise.
2582 * src/powerpc/sysv.S: Likewise.
2584 2004-08-30 Anthony Green <green@redhat.com>
2586 * Makefile.am: Add frv support.
2587 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2588 * configure.ac: Read configure.host.
2589 * configure.in: Read configure.host.
2590 * configure.host: New file. frv-elf needs libgloss.
2591 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2592 alignment. This is needed to frv and shouldn't harm the others.
2593 * include/ffi_common.h (ALIGN_DOWN): New macro.
2594 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2596 2004-08-24 David Daney <daney@avtrex.com>
2598 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2599 * testsuite/libffi.call/closure_fn1.c: Likewise.
2600 * testsuite/libffi.call/closure_fn2.c Likewise.
2601 * testsuite/libffi.call/closure_fn3.c: Likewise.
2602 * testsuite/libffi.call/closure_fn4.c: Likewise.
2603 * testsuite/libffi.call/closure_fn5.c: Likewise.
2604 * testsuite/libffi.call/cls_18byte.c: Likewise.
2605 * testsuite/libffi.call/cls_19byte.c: Likewise.
2606 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2607 * testsuite/libffi.call/cls_20byte.c: Likewise.
2608 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2609 * testsuite/libffi.call/cls_24byte.c: Likewise.
2610 * testsuite/libffi.call/cls_2byte.c: Likewise.
2611 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2612 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2613 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2614 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2615 * testsuite/libffi.call/cls_4byte.c: Likewise.
2616 * testsuite/libffi.call/cls_64byte.c: Likewise.
2617 * testsuite/libffi.call/cls_6byte.c: Likewise.
2618 * testsuite/libffi.call/cls_7byte.c: Likewise.
2619 * testsuite/libffi.call/cls_8byte.c: Likewise.
2620 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2621 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2622 * testsuite/libffi.call/cls_align_double.c: Likewise.
2623 * testsuite/libffi.call/cls_align_float.c: Likewise.
2624 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2625 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2626 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2627 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2628 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2629 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2630 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2631 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2632 * testsuite/libffi.call/cls_double.c: Likewise.
2633 * testsuite/libffi.call/cls_float.c: Likewise.
2634 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2635 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2636 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2637 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2638 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2639 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2640 * testsuite/libffi.call/cls_schar.c: Likewise.
2641 * testsuite/libffi.call/cls_sint.c: Likewise.
2642 * testsuite/libffi.call/cls_sshort.c: Likewise.
2643 * testsuite/libffi.call/cls_uchar.c: Likewise.
2644 * testsuite/libffi.call/cls_uint.c: Likewise.
2645 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2646 * testsuite/libffi.call/cls_ushort.c: Likewise.
2647 * testsuite/libffi.call/nested_struct.c: Likewise.
2648 * testsuite/libffi.call/nested_struct1.c: Likewise.
2649 * testsuite/libffi.call/nested_struct2.c: Likewise.
2650 * testsuite/libffi.call/nested_struct3.c: Likewise.
2651 * testsuite/libffi.call/problem1.c: Likewise.
2652 * testsuite/libffi.special/unwindtest.cc: Likewise.
2653 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2655 * testsuite/libffi.call/cls_16byte.c: Likewise.
2656 * testsuite/libffi.call/cls_5byte.c: Likewise.
2658 2004-08-23 David Daney <daney@avtrex.com>
2661 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2662 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2663 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2664 parameters and return types.
2665 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2666 (ffi_prep_closure): Ditto.
2667 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2668 alignment calculations.
2669 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2670 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2672 2004-08-14 Casey Marshall <csm@gnu.org>
2674 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2675 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2676 integer (O32 ABI only).
2677 (ffi_prep_closure): new function.
2678 (ffi_closure_mips_inner_O32): new function.
2679 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2680 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2681 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2682 64 bit integers correctly.
2683 (ffi_closure_O32): new function.
2684 Added DWARF-2 unwind info for both functions.
2686 2004-08-10 Andrew Haley <aph@redhat.com>
2688 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2690 2004-08-01 Robert Millan <robertmh@gnu.org>
2692 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2693 * configure: Regenerate.
2695 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2697 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2698 and mmap() explicitly instead of relying on preset autoconf cache
2700 * aclocal.m4: Regenerate.
2701 * configure: Regenerate.
2703 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2705 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2706 (ffi_check_float_struct): Remove unused prototype.
2708 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2710 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2711 character on Darwin, use '\n\t' instead.
2713 2004-06-26 Matthias Klose <doko@debian.org>
2715 * libtool-version: Fix typo in revision/age.
2717 2004-06-17 Matthias Klose <doko@debian.org>
2719 * libtool-version: New.
2720 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2721 * Makefile.in: Regenerate.
2723 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2725 * Makefile.am: Remove useless multilib rules.
2726 * Makefile.in: Regenerate.
2727 * aclocal.m4: Regenerate with automake 1.8.5.
2728 * configure.ac: Remove useless multilib configury.
2729 * configure: Regenerate.
2731 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2733 * .cvsignore: New file.
2735 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2737 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2739 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2742 2004-06-06 Sean McNeil <sean@mcneil.com>
2744 * configure.ac: Add x86_64-*-freebsd* support.
2745 * configure: Regenerate.
2747 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2750 * configure.ac: Test for existence of mmap and sys/mman.h before
2751 checking blacklist. Fix suggested by Jim Wilson.
2752 * configure: Regenerate.
2754 2004-04-26 Matt Austern <austern@apple.com>
2756 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2758 * src/powerpc/darwin_closure.S: Likewise.
2760 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2762 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2763 error. Reported by Thomas Heller <theller@python.net>.
2764 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2765 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2767 2004-03-20 Matthias Klose <doko@debian.org>
2769 * src/pa/linux.S: Fix typo.
2771 2004-03-19 Matthias Klose <doko@debian.org>
2773 * Makefile.am: Update.
2774 * Makefile.in: Regenerate.
2775 * src/pa/ffi.h.in: Remove.
2776 * src/pa/ffitarget.h: New file.
2778 2004-02-10 Randolph Chung <tausq@debian.org>
2780 * Makefile.am: Add PA support.
2781 * Makefile.in: Regenerate.
2782 * include/Makefile.in: Regenerate.
2783 * configure.ac: Add PA target.
2784 * configure: Regenerate.
2785 * src/pa/ffi.c: New file.
2786 * src/pa/ffi.h.in: Add PA support.
2787 * src/pa/linux.S: New file.
2788 * prep_cif.c: Add PA support.
2790 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2792 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2794 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2795 with ecif->cif->flags.
2796 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2798 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2799 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2800 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2801 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2803 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2805 * configure.in: Rename file to ...
2806 * configure.ac: ... this.
2807 * fficonfig.h.in: Regenerate.
2808 * Makefile.in: Regenerate.
2809 * include/Makefile.in: Regenerate.
2810 * testsuite/Makefile.in: Regenerate.
2812 2004-03-12 Matt Austern <austern@apple.com>
2814 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2815 changes in EH format resulting from addition of linkonce support.
2816 * src/powerpc/darwin_closure.S: Likewise.
2818 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2819 Paolo Bonzini <bonzini@gnu.org>
2821 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2822 Remove VPATH. Remove rules for object files. Remove multilib support.
2823 (AM_CCASFLAGS): Add.
2824 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2825 (AC_PREREQ): Bump version to 2.59.
2826 (AC_INIT): Fill with version info and bug address.
2827 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2828 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2829 De-precious CC so that the right flags are passed down to multilibs.
2830 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2831 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2832 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2833 * configure: Rebuilt.
2834 * aclocal.m4: Likewise.
2835 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2836 * fficonfig.h.in: Likewise.
2838 2004-03-11 Andreas Schwab <schwab@suse.de>
2840 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2841 arguments from fp registers only for the first 8 parameter slots.
2842 Don't convert a float parameter when passed in memory.
2844 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2846 * configure: Regenerate for config/accross.m4 correction.
2848 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2850 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2851 ecif->cif->bytes to bytes.
2852 (ffi_prep_cif_machdep): Add braces around nested if statement.
2854 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2856 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2858 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2859 (ffi_closure_helper_LINUX64): Fix typo.
2860 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2862 * testsuite/libffi.call/float.c: Likewise.
2863 * testsuite/libffi.call/float2.c: Likewise.
2865 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2867 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2868 long double function return and long double arg handling.
2869 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2870 Use "end_pfr" instead of "nf". Correct long double handling.
2872 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2874 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2875 space for long double return value. Adjust stack frame and offsets.
2876 Load f2 long double return.
2878 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2880 * src/types.c: Use 16 byte long double for POWERPC64.
2882 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2884 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2885 when the structure return address is passed in %o0.
2886 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2887 (ffi_v9_layout_struct): Align the field following a nested structure
2888 on a word boundary. Use memmove instead of memcpy.
2889 (ffi_call): Update call to ffi_V9_return_struct.
2890 (ffi_prep_closure): Define 'ctx' only for V8.
2891 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2892 and ffi_closure_sparc_inner_v9.
2893 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2894 Always skip the structure return address. For structures and long
2895 doubles, copy the argument directly.
2896 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2897 if required. Shift the maximum floating-point slot accordingly. For
2898 big structures, copy the argument directly; otherwise, left-justify the
2899 argument and call ffi_v9_layout_struct to lay out the structure on
2901 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2902 (ffi_closure_v8): Pass the structure return address. Update call to
2903 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2904 Skip the 'unimp' insn when returning long doubles and structures.
2905 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2906 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2907 FFI_TYPE_INT handling. Load structures both in integers and
2908 floating-point registers on return.
2909 * README: Update status of the SPARC port.
2911 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2913 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2915 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2917 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2919 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2920 value as of type ffi_arg, not unsigned int.
2922 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2924 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2927 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2929 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2932 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2934 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2937 2003-12-10 Richard Henderson <rth@redhat.com>
2939 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2940 size_t instead of int.
2942 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2944 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2945 * testsuite/libffi.call/many_win32.c (main): Replace variable
2946 int i with unsigned long ul.
2948 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2949 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2950 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2951 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2952 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2953 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2954 * testsuite/libffi.call/cls_align_float.c: Likewise.
2955 * testsuite/libffi.call/cls_align_double.c: Likewise.
2956 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2957 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2959 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2962 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2963 Align arguments to 32 bits.
2965 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2968 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2969 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2970 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2971 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2972 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2973 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2975 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2977 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2979 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2980 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2982 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2984 * Makefile.in: Rebuilt.
2985 * configure: Likewise.
2986 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2989 2003-11-21 Andreas Jaeger <aj@suse.de>
2990 Andreas Tobler <a.tobler@schweiz.ch>
2992 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2993 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2994 * Makefile.in: Rebuilt.
2995 * aclocal.m4: Likewise.
2996 * configure: Likewise.
2997 * fficonfig.h.in: Likewise.
2998 * testsuite/lib/libffi-dg.exp: Add include dir.
2999 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
3000 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3001 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
3002 for ffi_closure if available.
3003 * testsuite/libffi.call/closure_fn1.c: Likewise.
3004 * testsuite/libffi.call/closure_fn2.c: Likewise.
3005 * testsuite/libffi.call/closure_fn3.c: Likewise.
3006 * testsuite/libffi.call/closure_fn4.c: Likewise.
3007 * testsuite/libffi.call/closure_fn5.c: Likewise.
3008 * testsuite/libffi.call/cls_12byte.c: Likewise.
3009 * testsuite/libffi.call/cls_16byte.c: Likewise.
3010 * testsuite/libffi.call/cls_18byte.c: Likewise.
3011 * testsuite/libffi.call/cls_19byte.c: Likewise.
3012 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3013 * testsuite/libffi.call/cls_20byte.c: Likewise.
3014 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3015 * testsuite/libffi.call/cls_24byte.c: Likewise.
3016 * testsuite/libffi.call/cls_2byte.c: Likewise.
3017 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3018 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3019 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3020 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3021 * testsuite/libffi.call/cls_4byte.c: Likewise.
3022 * testsuite/libffi.call/cls_5byte.c: Likewise.
3023 * testsuite/libffi.call/cls_64byte.c: Likewise.
3024 * testsuite/libffi.call/cls_6byte.c: Likewise.
3025 * testsuite/libffi.call/cls_7byte.c: Likewise.
3026 * testsuite/libffi.call/cls_8byte.c: Likewise.
3027 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3028 * testsuite/libffi.call/cls_9byte2.c: Likewise.
3029 * testsuite/libffi.call/cls_double.c: Likewise.
3030 * testsuite/libffi.call/cls_float.c: Likewise.
3031 * testsuite/libffi.call/cls_schar.c: Likewise.
3032 * testsuite/libffi.call/cls_sint.c: Likewise.
3033 * testsuite/libffi.call/cls_sshort.c: Likewise.
3034 * testsuite/libffi.call/cls_uchar.c: Likewise.
3035 * testsuite/libffi.call/cls_uint.c: Likewise.
3036 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3037 * testsuite/libffi.call/cls_ushort.c: Likewise.
3038 * testsuite/libffi.call/nested_struct.c: Likewise.
3039 * testsuite/libffi.call/nested_struct1.c: Likewise.
3040 * testsuite/libffi.call/nested_struct2.c: Likewise.
3041 * testsuite/libffi.call/nested_struct3.c: Likewise.
3042 * testsuite/libffi.call/problem1.c: Likewise.
3043 * testsuite/libffi.special/unwindtest.cc: Likewise.
3045 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
3047 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
3049 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
3051 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
3052 Add -lgcc_s to additional flags.
3054 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
3056 * configure.in, include/Makefile.am: PR libgcj/11147, install
3057 the ffitarget.h header file in a gcc versioned and target
3059 * configure: Regenerated.
3060 * Makefile.in, include/Makefile.in: Likewise.
3061 * testsuite/Makefile.in: Likewise.
3063 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
3065 * testsuite/libffi.call/closure_fn0.c: Print result and check
3066 with dg-output to make debugging easier.
3067 * testsuite/libffi.call/closure_fn1.c: Likewise.
3068 * testsuite/libffi.call/closure_fn2.c: Likewise.
3069 * testsuite/libffi.call/closure_fn3.c: Likewise.
3070 * testsuite/libffi.call/closure_fn4.c: Likewise.
3071 * testsuite/libffi.call/closure_fn5.c: Likewise.
3072 * testsuite/libffi.call/cls_12byte.c: Likewise.
3073 * testsuite/libffi.call/cls_16byte.c: Likewise.
3074 * testsuite/libffi.call/cls_18byte.c: Likewise.
3075 * testsuite/libffi.call/cls_19byte.c: Likewise.
3076 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3077 * testsuite/libffi.call/cls_20byte.c: Likewise.
3078 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3079 * testsuite/libffi.call/cls_24byte.c: Likewise.
3080 * testsuite/libffi.call/cls_2byte.c: Likewise.
3081 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3082 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3083 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3084 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3085 * testsuite/libffi.call/cls_4byte.c: Likewise.
3086 * testsuite/libffi.call/cls_5byte.c: Likewise.
3087 * testsuite/libffi.call/cls_64byte.c: Likewise.
3088 * testsuite/libffi.call/cls_6byte.c: Likewise.
3089 * testsuite/libffi.call/cls_7byte.c: Likewise.
3090 * testsuite/libffi.call/cls_8byte.c: Likewise.
3091 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3092 * testsuite/libffi.call/cls_9byte2.c: Likewise.
3093 * testsuite/libffi.call/cls_double.c: Likewise.
3094 * testsuite/libffi.call/cls_float.c: Likewise.
3095 * testsuite/libffi.call/cls_schar.c: Likewise.
3096 * testsuite/libffi.call/cls_sint.c: Likewise.
3097 * testsuite/libffi.call/cls_sshort.c: Likewise.
3098 * testsuite/libffi.call/cls_uchar.c: Likewise.
3099 * testsuite/libffi.call/cls_uint.c: Likewise.
3100 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3101 * testsuite/libffi.call/cls_ushort.c: Likewise.
3102 * testsuite/libffi.call/problem1.c: Likewise.
3104 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
3107 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
3109 * testsuite/libffi.call/cls_9byte2.c: New test case.
3110 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3111 * testsuite/libffi.call/cls_64byte.c: Likewise.
3112 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3113 * testsuite/libffi.call/cls_19byte.c: Likewise.
3114 * testsuite/libffi.call/cls_18byte.c: Likewise.
3115 * testsuite/libffi.call/closure_fn4.c: Likewise.
3116 * testsuite/libffi.call/closure_fn5.c: Likewise.
3117 * testsuite/libffi.call/cls_schar.c: Likewise.
3118 * testsuite/libffi.call/cls_sint.c: Likewise.
3119 * testsuite/libffi.call/cls_sshort.c: Likewise.
3120 * testsuite/libffi.call/nested_struct2.c: Likewise.
3121 * testsuite/libffi.call/nested_struct3.c: Likewise.
3123 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
3125 * testsuite/libffi.call/cls_double.c: Do a check on the result.
3126 * testsuite/libffi.call/cls_uchar.c: Likewise.
3127 * testsuite/libffi.call/cls_uint.c: Likewise.
3128 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3129 * testsuite/libffi.call/cls_ushort.c: Likewise.
3130 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
3132 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
3134 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
3137 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
3139 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
3140 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
3142 2003-10-22 David Daney <ddaney@avtrex.com>
3144 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
3145 __attribute__((__mode__(__SI__))) and friends.
3147 2003-10-22 Andreas Schwab <schwab@suse.de>
3149 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
3151 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
3153 * configure.in: AC_LINK_FILES(ffitarget.h).
3154 * configure: Regenerate.
3155 * Makefile.in: Likewise.
3156 * include/Makefile.in: Likewise.
3157 * testsuite/Makefile.in: Likewise.
3158 * fficonfig.h.in: Likewise.
3160 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
3161 Richard Henderson <rth@redhat.com>
3163 Avoid that ffi.h includes fficonfig.h.
3165 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
3166 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
3167 (TARGET_SRC_MIPS_SGI): Removed.
3168 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
3169 (MIPS_SGI): Removed.
3170 (CLEANFILES): Removed.
3171 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
3173 * acconfig.h: Removed.
3174 * configure.in: Compute sizeofs only for double and long double.
3175 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
3176 into AC_DEFINE instead of using acconfig.h. Create
3177 include/ffitarget.h instead of include/fficonfig.h. Rename
3178 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
3179 AC_DEFINE EH_FRAME_FLAGS.
3180 * include/Makefile.am (DISTCLEANFILES): New automake macro.
3181 (hack_DATA): Add ffitarget.h.
3182 * include/ffi.h.in: Remove all system specific definitions.
3183 Declare raw API even if it is not installed, why bother?
3184 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
3185 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
3186 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
3187 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
3188 * include/ffi_common.h (bool): Do not define.
3189 (ffi_assert): Accept failed assertion.
3190 (ffi_type_test): Return void and accept file/line.
3191 (FFI_ASSERT): Pass stringized failed assertion.
3192 (FFI_ASSERT_AT): New macro.
3193 (FFI_ASSERT_VALID_TYPE): New macro.
3194 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
3195 UINT64, SINT64): Define here with gcc's __attribute__ macro
3197 (FLOAT32, ALIGN): Define here instead of in ffi.h
3198 * include/ffi-mips.h: Removed. Its content moved to
3199 src/mips/ffitarget.h after separating assembly and C sections.
3200 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
3201 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
3202 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
3203 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
3204 SIZEOF_ARG -> FFI_SIZEOF_ARG.
3205 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
3206 * src/debug.c (ffi_assert): Accept stringized failed assertion.
3207 (ffi_type_test): Rewritten.
3208 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
3209 FFI_ASSERT_VALID_TYPE.
3210 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
3211 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
3212 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
3213 src/s390/ffitarget.h, src/sh/ffitarget.h,
3214 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
3215 src/x86/ffitarget.h: New files.
3216 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
3217 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
3218 src/powerpc/aix.S, src/powerpc/darwin.S,
3219 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
3220 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
3221 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
3222 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
3223 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
3226 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3228 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
3229 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
3231 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
3233 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
3234 Used when FFI_DEBUG = 1.
3236 2003-10-14 Alan Modra <amodra@bigpond.net.au>
3238 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
3241 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3243 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
3244 FFI_MIPS_O32 for O32 ABI.
3246 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
3248 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
3249 SPARC64. Cleanup whitespaces.
3251 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
3253 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
3254 strongarm, xscale. Cleanup whitespaces.
3255 * testsuite/libffi.call/closure_fn1.c: Likewise.
3256 * testsuite/libffi.call/closure_fn2.c: Likewise.
3257 * testsuite/libffi.call/closure_fn3.c: Likewise.
3258 * testsuite/libffi.call/cls_12byte.c: Likewise.
3259 * testsuite/libffi.call/cls_16byte.c: Likewise.
3260 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3261 * testsuite/libffi.call/cls_20byte.c: Likewise.
3262 * testsuite/libffi.call/cls_24byte.c: Likewise.
3263 * testsuite/libffi.call/cls_2byte.c: Likewise.
3264 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3265 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3266 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3267 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3268 * testsuite/libffi.call/cls_4byte.c: Likewise.
3269 * testsuite/libffi.call/cls_5byte.c: Likewise.
3270 * testsuite/libffi.call/cls_6byte.c: Likewise.
3271 * testsuite/libffi.call/cls_7byte.c: Likewise.
3272 * testsuite/libffi.call/cls_8byte.c: Likewise.
3273 * testsuite/libffi.call/cls_double.c: Likewise.
3274 * testsuite/libffi.call/cls_float.c: Likewise.
3275 * testsuite/libffi.call/cls_uchar.c: Likewise.
3276 * testsuite/libffi.call/cls_uint.c: Likewise.
3277 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3278 * testsuite/libffi.call/cls_ushort.c: Likewise.
3279 * testsuite/libffi.call/nested_struct.c: Likewise.
3280 * testsuite/libffi.call/nested_struct1.c: Likewise.
3281 * testsuite/libffi.call/problem1.c: Likewise.
3282 * testsuite/libffi.special/unwindtest.cc: Likewise.
3283 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
3285 2003-09-18 David Edelsohn <edelsohn@gnu.org>
3287 * src/powerpc/aix.S: Cleanup whitespaces.
3288 * src/powerpc/aix_closure.S: Likewise.
3290 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3292 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
3293 * src/powerpc/darwin_closure.S: Likewise.
3294 * src/powerpc/ffi_darwin.c: Likewise.
3296 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3297 David Edelsohn <edelsohn@gnu.org>
3299 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
3300 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
3302 * src/powerpc/darwin_closure.S: Likewise.
3303 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
3304 according to the Darwin/AIX ABI.
3305 (ffi_prep_cif_machdep): Likewise.
3306 (ffi_closure_helper_DARWIN): Likewise.
3307 Remove the outgoing parameter stack logic. Simplify the evaluation
3308 of the different CASE types.
3309 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
3310 statement in the trampoline code.
3312 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
3314 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
3315 for the register size.
3316 (ffi_closure_helper_SYSV): Handle the structure return value
3318 (ffi_closure_helper_SYSV): Return the appropriate type when
3319 the registers are used for the structure return value.
3320 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
3321 the 64-bit return value. Update copyright years.
3323 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3325 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
3326 srcdir for ffi_mips.h.
3328 2003-09-12 Alan Modra <amodra@bigpond.net.au>
3330 * src/prep_cif.c (initialize_aggregate): Include tail padding in
3332 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
3333 placement of float result.
3334 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
3335 cast of "resp" for big-endian 64 bit machines.
3337 2003-09-11 Alan Modra <amodra@bigpond.net.au>
3339 * src/types.c (double, longdouble): Merge identical SH and ARM
3340 typedefs, and add POWERPC64.
3341 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
3342 struct split over gpr and rest.
3343 (ffi_prep_cif_machdep): Correct intarg_count for structures.
3344 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
3346 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
3348 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
3351 2003-09-09 Alan Modra <amodra@bigpond.net.au>
3353 * configure: Regenerate.
3355 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3357 * Makefile.am: Remove build rules for ffitest.
3358 * Makefile.in: Rebuilt.
3360 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3362 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3363 about implicit declaration of abort().
3365 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3367 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3368 * Makefile.in: Rebuilt.
3369 * configure.in: Add dejagnu test framework.
3370 * configure: Rebuilt.
3372 * testsuite/Makefile.am: New file.
3373 * testsuite/Makefile.in: Built
3374 * testsuite/lib/libffi-dg.exp: New file.
3375 * testsuite/config/default.exp: Likewise.
3376 * testsuite/libffi.call/call.exp: Likewise.
3377 * testsuite/libffi.call/ffitest.h: Likewise.
3378 * testsuite/libffi.call/closure_fn0.c: Likewise.
3379 * testsuite/libffi.call/closure_fn1.c: Likewise.
3380 * testsuite/libffi.call/closure_fn2.c: Likewise.
3381 * testsuite/libffi.call/closure_fn3.c: Likewise.
3382 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3383 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3384 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3385 * testsuite/libffi.call/cls_2byte.c: Likewise.
3386 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3387 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3388 * testsuite/libffi.call/cls_4byte.c: Likewise.
3389 * testsuite/libffi.call/cls_5byte.c: Likewise.
3390 * testsuite/libffi.call/cls_6byte.c: Likewise.
3391 * testsuite/libffi.call/cls_7byte.c: Likewise.
3392 * testsuite/libffi.call/cls_8byte.c: Likewise.
3393 * testsuite/libffi.call/cls_12byte.c: Likewise.
3394 * testsuite/libffi.call/cls_16byte.c: Likewise.
3395 * testsuite/libffi.call/cls_20byte.c: Likewise.
3396 * testsuite/libffi.call/cls_24byte.c: Likewise.
3397 * testsuite/libffi.call/cls_double.c: Likewise.
3398 * testsuite/libffi.call/cls_float.c: Likewise.
3399 * testsuite/libffi.call/cls_uchar.c: Likewise.
3400 * testsuite/libffi.call/cls_uint.c: Likewise.
3401 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3402 * testsuite/libffi.call/cls_ushort.c: Likewise.
3403 * testsuite/libffi.call/float.c: Likewise.
3404 * testsuite/libffi.call/float1.c: Likewise.
3405 * testsuite/libffi.call/float2.c: Likewise.
3406 * testsuite/libffi.call/many.c: Likewise.
3407 * testsuite/libffi.call/many_win32.c: Likewise.
3408 * testsuite/libffi.call/nested_struct.c: Likewise.
3409 * testsuite/libffi.call/nested_struct1.c: Likewise.
3410 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3411 * testsuite/libffi.call/problem1.c: Likewise.
3412 * testsuite/libffi.call/promotion.c: Likewise.
3413 * testsuite/libffi.call/return_ll.c: Likewise.
3414 * testsuite/libffi.call/return_sc.c: Likewise.
3415 * testsuite/libffi.call/return_uc.c: Likewise.
3416 * testsuite/libffi.call/strlen.c: Likewise.
3417 * testsuite/libffi.call/strlen_win32.c: Likewise.
3418 * testsuite/libffi.call/struct1.c: Likewise.
3419 * testsuite/libffi.call/struct2.c: Likewise.
3420 * testsuite/libffi.call/struct3.c: Likewise.
3421 * testsuite/libffi.call/struct4.c: Likewise.
3422 * testsuite/libffi.call/struct5.c: Likewise.
3423 * testsuite/libffi.call/struct6.c: Likewise.
3424 * testsuite/libffi.call/struct7.c: Likewise.
3425 * testsuite/libffi.call/struct8.c: Likewise.
3426 * testsuite/libffi.call/struct9.c: Likewise.
3427 * testsuite/libffi.special/special.exp: New file.
3428 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3429 * testsuite/libffi.special/unwindtest.cc: Likewise.
3432 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3434 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3437 2003-08-02 Alan Modra <amodra@bigpond.net.au>
3439 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3441 (ffi_closure_helper_LINUX64): Likewise.
3442 * src/powerpc/linux64.S: Remove code writing to parm save area.
3443 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3444 address in lr from ffi_closure_helper_LINUX64 call to calculate
3445 table address. Optimize function tail.
3447 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3449 * src/sparc/ffi.c: Handle all floating point registers.
3450 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3452 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3454 * README: Note that libffi is not part of GCC. Update the project
3457 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3459 * src/powerpc/ppc_closure.S: Include ffi.h.
3461 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3463 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3464 Use C style comments.
3466 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3468 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3469 * Makefile.in: Regenerate.
3470 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3471 * configure: Regenerate.
3472 * include/ffi.h.in: Add SHmedia support.
3473 * src/sh64/ffi.c: New file.
3474 * src/sh64/sysv.S: New file.
3476 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3478 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3479 should be read-only.
3480 * configure: Rebuilt.
3481 * fficonfig.h.in: Rebuilt.
3482 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3483 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3484 * src/powerpc/linux64.S: Likewise.
3485 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3486 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3487 if -fpic/-fPIC/-mrelocatable.
3488 * src/powerpc/powerpc_closure.S: Likewise.
3489 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3490 #write in .eh_frame flags.
3491 * src/sparc/v9.S: Likewise.
3492 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3493 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3494 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3496 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3498 Fixes PR bootstrap/10656
3499 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3500 support for .register pseudo-op.
3501 * src/sparc/v8.S: Use it.
3502 * fficonfig.h.in: Rebuilt.
3503 * configure: Rebuilt.
3505 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3507 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3508 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3509 Make it the default on POWERPC64.
3510 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3511 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3512 * configure: Rebuilt.
3513 * src/powerpc/ffi.c (hidden): Define.
3514 (ffi_prep_args_SYSV): Renamed from
3515 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3516 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3517 ASM_NEEDS_REGISTERS64): New.
3518 (ffi_prep_args64): New function.
3519 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3520 (ffi_call): Likewise.
3521 (ffi_prep_closure): Likewise.
3522 (flush_icache): Surround by #ifndef POWERPC64.
3523 (ffi_dblfl): New union type.
3524 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3525 (ffi_closure_helper_LINUX64): New function.
3526 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3528 * src/powerpc/sysv.S: Likewise.
3529 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3530 * src/powerpc/linux64.S: New file.
3531 * src/powerpc/linux64_closure.S: New file.
3532 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3533 src/powerpc/linux64_closure.S.
3534 (TARGET_SRC_POWERPC): Likewise.
3536 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3537 closure_test_fn3): Fix result printing on big-endian 64-bit
3539 (main): Print tst2_arg instead of uninitialized tst2_result.
3541 * src/ffitest.c (main): Hide what closure pointer really points to
3544 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3546 * configure.in (arm-*-netbsdelf*): Add configuration.
3547 (configure): Regenerated.
3549 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3551 * include/Makefile.in: Regenerate.
3553 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3555 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3557 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3558 Receive closure pointer through parameter, read args using
3559 __builtin_dwarf_cfa.
3560 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3562 2003-03-12 Andreas Schwab <schwab@suse.de>
3564 * configure.in: Avoid trailing /. in toolexeclibdir.
3565 * configure: Rebuilt.
3567 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3569 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3571 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3573 * libffi/src/powerpc/darwin_closure.S:
3574 Fix alignement bug, allocate 8 bytes for the result.
3575 * libffi/src/powerpc/aix_closure.S:
3577 * libffi/src/powerpc/ffi_darwin.c:
3578 Update stackframe description for aix/darwin_closure.S.
3580 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3582 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3585 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3586 Andreas Schwab <schwab@suse.de>
3588 * configure.in: Adjust command to source config-ml.in to account
3589 for changes to the libffi_basedir definition.
3590 (libffi_basedir): Remove ${srcdir} from value and include trailing
3593 * configure: Regenerate.
3595 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3597 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3599 2003-01-28 Andrew Haley <aph@redhat.com>
3601 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3602 * src/x86/ffi64.c (ffi_prep_closure): New.
3603 (ffi_closure_UNIX64_inner): New.
3604 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3606 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
3608 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3609 Remove USE_LIBDIR conditional.
3610 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3611 * Makefile.in, configure: Rebuilt.
3613 2003-01027 David Edelsohn <edelsohn@gnu.org>
3615 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3616 * Makefile.in: Regenerate.
3618 2003-01-22 Andrew Haley <aph@redhat.com>
3620 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3623 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3625 * src/powerpc/darwin.S: Add unwind info.
3626 * src/powerpc/darwin_closure.S: Likewise.
3628 2003-01-14 Andrew Haley <aph@redhat.com>
3630 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3631 (ffi_prep_cif_machdep): Likewise.
3632 * src/x86/unix64.S: Add unwind info.
3634 2003-01-14 Andreas Jaeger <aj@suse.de>
3636 * src/ffitest.c (main): Only use ffi_closures if those are
3639 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3641 * libffi/src/ffitest.c
3642 add closure testcases
3644 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3646 * libffi/src/powerpc/ffi.c
3647 fix alignment bug for float (4 byte aligned iso 8 byte)
3649 2003-01-09 Geoffrey Keating <geoffk@apple.com>
3651 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3652 * src/powerpc/darwin.S: Remove RCS version string.
3654 2003-01-03 Jeff Sturm <jsturm@one-point.com>
3656 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3657 * src/ffitest.c (main): Use static storage for closure.
3658 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3659 * src/sparc/v8.S (ffi_closure_v8): New.
3660 * src/sparc/v9.S (ffi_closure_v9): New.
3662 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3664 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3665 enumeration for X86_WIN32.
3666 * src/x86/win32.S: Added ffi_call_STDCALL function
3668 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3669 switch cases for recognising FFI_STDCALL and
3670 calling ffi_call_STDCALL if target is X86_WIN32.
3671 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3672 stdcall versions of the "my_strlen" and "many"
3673 test functions (for X86_WIN32).
3674 Added test cases to test stdcall invocation using
3677 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3679 * src/sh/sysv.S: Add DWARF2 unwind info.
3681 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3683 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3685 2002-11-26 Jim Wilson <wilson@redhat.com>
3687 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3689 2002-11-23 H.J. Lu <hjl@gnu.org>
3691 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3692 Include ../config/accross.m4.
3693 * aclocal.m4; Rebuild.
3694 * configure: Likewise.
3696 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3698 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3700 2002-11-11 DJ Delorie <dj@redhat.com>
3702 * configure.in: Look for common files in the right place.
3704 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3706 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3707 raw data as _Jv_word values, not ffi_raw.
3708 (ffi_java_ptrarray_to_raw): Likewise.
3709 (ffi_java_rvalue_to_raw): New function.
3710 (ffi_java_raw_call): Call it.
3711 (ffi_java_raw_to_rvalue): New function.
3712 (ffi_java_translate_args): Call it.
3713 * src/ffitest.c (closure_test_fn): Interpret return value
3714 as ffi_arg, not int.
3715 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3716 FFI_TYPE_POINTER case.
3717 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3718 values extended to word size.
3720 2002-10-02 Andreas Jaeger <aj@suse.de>
3722 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3724 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
3726 * include/ffi.h.in: Fix i386 win32 compilation.
3728 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3730 * configure.in: Add s390x-*-linux-* target.
3731 * configure: Regenerate.
3732 * include/ffi.h.in: Define S390X for s390x targets.
3733 (FFI_CLOSURES): Define for s390/s390x.
3734 (FFI_TRAMPOLINE_SIZE): Likewise.
3735 (FFI_NATIVE_RAW_API): Likewise.
3736 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3737 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
3738 * src/s390/ffi.c: Major rework of existing code. Add support for
3739 s390x targets. Add closure support.
3740 * src/s390/sysv.S: Likewise.
3742 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
3744 * src/arm/sysv.S: Fix typo.
3746 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
3748 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3749 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3750 (ffi_call_SYSV): Handle soft-float.
3752 2002-09-27 Bo Thorsen <bo@suse.de>
3754 * include/ffi.h.in: Fix multilib x86-64 support.
3756 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3758 * Makefile.am (all-multi): Fix multilib parallel build.
3760 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3762 * configure.in (sh[34]*-*-linux*): Add brackets.
3763 * configure: Regenerate.
3765 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3767 * Makefile.am: Add SH support.
3768 * Makefile.in: Regenerate.
3769 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3770 * configure: Regenerate.
3771 * include/ffi.h.in: Add SH support.
3772 * src/sh/ffi.c: New file.
3773 * src/sh/sysv.S: New file.
3774 * src/types.c: Add SH support.
3776 2002-07-16 Bo Thorsen <bo@suse.de>
3778 * src/x86/ffi64.c: New file that adds x86-64 support.
3779 * src/x86/unix64.S: New file that handles argument setup for
3781 * src/x86/sysv.S: Don't use this on x86-64.
3782 * src/x86/ffi.c: Don't use this on x86-64.
3784 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3786 * src/ffitest.c (struct6): New test that tests a special case in
3788 (struct7): Likewise.
3789 (struct8): Likewise.
3790 (struct9): Likewise.
3791 (closure_test_fn): Silence warning about this when it's not used.
3792 (main): Add the new tests.
3793 (main): Fix a couple of wrong casts and silence some compiler warnings.
3794 * include/ffi.h.in: Add x86-64 ABI definition.
3795 * fficonfig.h.in: Regenerate.
3796 * Makefile.am: Add x86-64 support.
3797 * configure.in: Likewise.
3798 * Makefile.in: Regenerate.
3799 * configure: Likewise.
3801 2002-06-24 Bo Thorsen <bo@suse.de>
3803 * src/types.c: Merge settings for similar architectures.
3804 Add x86-64 sizes and alignments.
3806 2002-06-23 Bo Thorsen <bo@suse.de>
3808 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3809 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3810 * src/mips/ffi.c (ffi_prep_args): Likewise.
3811 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3813 2002-07-18 H.J. Lu (hjl@gnu.org)
3815 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3816 (libffi_la_SOURCES): Support MIPS_LINUX.
3817 (libffi_convenience_la_SOURCES): Likewise.
3818 * Makefile.in: Regenerated.
3820 * configure.in (mips64*-*): Skip.
3821 (mips*-*-linux*): New.
3822 * configure: Regenerated.
3824 * src/mips/ffi.c: Include <sgidefs.h>.
3826 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3828 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3830 2002-05-27 Roger Sayle <roger@eyesopen.com>
3832 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3834 2002-05-27 Bo Thorsen <bo@suse.de>
3836 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3839 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3841 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3842 beginning of function (for older apple cc).
3844 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3846 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3847 script entry, and set LD to it when configuring multilibs.
3848 * configure: Rebuilt.
3850 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3852 * configure.in (sparc64-*-netbsd*): Add target.
3853 (sparc-*-netbsdelf*): Likewise.
3854 * configure: Regenerate.
3856 2002-04-28 David S. Miller <davem@redhat.com>
3858 * configure.in, configure: Fix SPARC test in previous change.
3860 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3862 * Makefile.am: Add Linux for S/390 support.
3863 * Makefile.in: Regenerate.
3864 * configure.in: Add Linux for S/390 support.
3865 * configure: Regenerate.
3866 * include/ffi.h.in: Add Linux for S/390 support.
3867 * src/s390/ffi.c: New file from libffi CVS tree.
3868 * src/s390/sysv.S: New file from libffi CVS tree.
3870 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3872 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3874 * src/sparc/v8.S: Use it.
3875 * src/sparc/v9.S: Likewise.
3876 * fficonfig.h.in: Rebuilt.
3877 * configure: Rebuilt.
3879 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3881 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3883 * src/ia64/unix.S: Add unwind information. Fix comments.
3884 Save sp in a way that's compatible with unwind info.
3885 (ffi_call_unix): Correctly restore sp in all cases.
3886 * src/ia64/ffi.c: Add, fix comments.
3888 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3890 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3892 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3894 * configure.in (alpha*-*-netbsd*): Add target.
3895 * configure: Regenerate.
3897 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3899 * src/sparc/v8.S: Add unwind info.
3900 * src/sparc/v9.S: Likewise.
3902 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3904 * configure.in: Enable i*86-*-netbsdelf*.
3905 * configure: Rebuilt.
3907 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3910 * src/mips/n32.s: Delete
3911 * src/mips/o32.s: Delete
3913 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3915 * configure.in: Enable alpha*-*-freebsd*.
3916 * configure: Rebuilt.
3918 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3920 * Makefile.am: libfficonvenience -> libffi_convenience.
3921 * Makefile.in: Rebuilt.
3923 * Makefile.am: Define ffitest_OBJECTS.
3924 * Makefile.in: Rebuilt.
3926 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3927 David Edelsohn <edelsohn@gnu.org>
3929 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3930 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3931 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3932 * Makefile.in: Regenerate.
3933 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3934 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3935 (flush_icache, flush_range): New functions.
3936 (ffi_closure_helper_DARWIN): New function.
3937 * src/powerpc/aix_closure.S: New file.
3938 * src/powerpc/darwin_closure.S: New file.
3940 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3942 * include/ffi.h.in: Add typedef for ffi_arg.
3943 * src/ffitest.c (main): Declare rint with ffi_arg.
3945 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3947 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3948 number of GPRs for floating-point arguments.
3950 2002-01-31 Anthony Green <green@redhat.com>
3952 * configure: Rebuilt.
3953 * configure.in: Replace CHECK_SIZEOF and endian tests with
3954 cross-compiler friendly macros.
3955 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3958 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3960 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3961 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3963 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3965 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3966 (TARGET_SRC_POWERPC_AIX): New.
3967 (POWERPC_AIX): New stanza.
3968 * Makefile.in: Regenerate.
3969 * configure.in: Add AIX case.
3970 * configure: Regenerate.
3971 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3972 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3973 size. Fix "long double" support.
3974 (ffi_call): Add FFI_AIX case.
3975 * src/powerpc/aix.S: New.
3977 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3979 Implement Darwin PowerPC ABI.
3980 * configure.in: Handle powerpc-*-darwin*.
3981 * Makefile.am: Set source files for POWERPC_DARWIN.
3982 * configure: Rebuilt.
3983 * Makefile.in: Rebuilt.
3984 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3986 * src/powerpc/darwin.S: New file.
3987 * src/powerpc/ffi_darwin.c: New file.
3989 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3991 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3993 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3995 * src/x86/sysv.S: Avoid gas-only .balign directive.
3996 Use C style comments.
3998 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4000 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
4001 Fixes PR bootstrap/3563.
4003 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4005 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
4007 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4009 * configure.in: Recognize sparc*-sun-* host.
4010 * configure: Regenerate.
4012 2001-06-06 Andrew Haley <aph@redhat.com>
4014 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
4016 2001-06-03 Andrew Haley <aph@redhat.com>
4018 * src/alpha/osf.S: Add unwind info.
4019 * src/powerpc/sysv.S: Add unwind info.
4020 * src/powerpc/ppc_closure.S: Likewise.
4022 2000-05-31 Jeff Sturm <jsturm@one-point.com>
4024 * configure.in: Fix AC_ARG_ENABLE usage.
4025 * configure: Rebuilt.
4027 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
4029 * configure.in: Remove warning about beta code.
4030 * configure: Rebuilt.
4032 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
4034 * src/ia64/unix.S: Restore stack pointer when returning from
4036 * src/ia64/ffi.c: Fix typo in comment.
4038 2001-04-18 Jim Wilson <wilson@redhat.com>
4040 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
4041 to eliminate RAW DV.
4043 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
4045 * Makefile.am: Make a libtool convenience library.
4046 * Makefile.in: Rebuilt.
4048 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
4050 * configure.in: Use different syntax for subdirectory creation.
4051 * configure: Rebuilt.
4053 2001-03-27 Jon Beniston <jon@beniston.com>
4055 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
4056 * configure: Rebuilt.
4057 * Makefile.am: Added X86_WIN32 target support.
4058 * Makefile.in: Rebuilt.
4060 * include/ffi.h.in: Added X86_WIN32 target support.
4062 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
4063 * src/types.c: Added X86_WIN32 target support.
4065 * src/x86/win32.S: New file. Based on sysv.S, but with EH
4066 stuff removed and made to work with CygWin's gas.
4068 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
4070 * configure.in: Make target subdirectory in build dir.
4071 * Makefile.am: Override suffix based rules to specify correct output
4073 * Makefile.in: Rebuilt.
4074 * configure: Rebuilt.
4076 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
4078 * src/powerpc/ppc_closure.S: New file.
4079 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
4080 involving long long and register pairs.
4081 (ffi_prep_closure): New function.
4082 (flush_icache): Likewise.
4083 (ffi_closure_helper_SYSV): Likewise.
4084 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
4085 (FFI_TRAMPOLINE_SIZE): Likewise.
4086 (FFI_NATIVE_RAW_API): Likewise.
4087 * Makefile.in: Rebuilt.
4088 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
4089 (TARGET_SRC_POWERPC): Likewise.
4091 2001-03-19 Tom Tromey <tromey@redhat.com>
4093 * Makefile.in: Rebuilt.
4094 * Makefile.am (ffitest_LDFLAGS): New macro.
4096 2001-03-02 Nick Clifton <nickc@redhat.com>
4098 * include/ffi.h.in: Remove RCS ident string.
4099 * include/ffi_mips.h: Remove RCS ident string.
4100 * src/debug.c: Remove RCS ident string.
4101 * src/ffitest.c: Remove RCS ident string.
4102 * src/prep_cif.c: Remove RCS ident string.
4103 * src/types.c: Remove RCS ident string.
4104 * src/alpha/ffi.c: Remove RCS ident string.
4105 * src/alpha/osf.S: Remove RCS ident string.
4106 * src/arm/ffi.c: Remove RCS ident string.
4107 * src/arm/sysv.S: Remove RCS ident string.
4108 * src/mips/ffi.c: Remove RCS ident string.
4109 * src/mips/n32.S: Remove RCS ident string.
4110 * src/mips/o32.S: Remove RCS ident string.
4111 * src/sparc/ffi.c: Remove RCS ident string.
4112 * src/sparc/v8.S: Remove RCS ident string.
4113 * src/sparc/v9.S: Remove RCS ident string.
4114 * src/x86/ffi.c: Remove RCS ident string.
4115 * src/x86/sysv.S: Remove RCS ident string.
4117 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
4119 * include/ffi.h.in: Change sourceware.cygnus.com references to
4122 2000-12-09 Richard Henderson <rth@redhat.com>
4124 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
4125 (ffi_closure_osf_inner): Index rather than increment avalue
4126 and arg_types. Give ffi_closure_osf the raw return value type.
4127 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
4130 2000-12-07 Richard Henderson <rth@redhat.com>
4132 * src/raw_api.c (ffi_translate_args): Fix typo.
4133 (ffi_prep_closure): Likewise.
4135 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
4136 FFI_TRAMPOLINE_SIZE.
4137 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
4138 cif->bytes for new ffi_call_osf implementation.
4139 (ffi_prep_args): Absorb into ...
4140 (ffi_call): ... here. Do all stack allocation here and
4141 avoid a callback function.
4142 (ffi_prep_closure, ffi_closure_osf_inner): New.
4143 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
4144 (ffi_closure_osf): New.
4146 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
4148 * config.guess, config.sub, install-sh: Removed.
4149 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
4150 * Makefile.in: Rebuilt.
4152 * acinclude.m4: Include libtool macros from the top level.
4153 * aclocal.m4, configure: Rebuilt.
4155 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
4157 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
4158 * configure: Rebuilt.
4160 2000-05-11 Scott Bambrough <scottb@netwinder.org>
4162 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
4163 memory correctly. Use conditional instructions, not branches where
4166 2000-05-04 Tom Tromey <tromey@cygnus.com>
4168 * configure: Rebuilt.
4169 * configure.in: Match `arm*-*-linux-*'.
4170 From Chris Dornan <cdornan@arm.com>.
4172 2000-04-28 Jakub Jelinek <jakub@redhat.com>
4174 * Makefile.am (SUBDIRS): Define.
4175 (AM_MAKEFLAGS): Likewise.
4176 (Multilib support.): Add section.
4177 * Makefile.in: Rebuilt.
4178 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
4179 New variables. Set for gcc using -print-multi-lib. Export them
4181 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
4182 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
4184 (extra_compiler_flags_value, extra_compiler_flags): Check these
4185 for extra compiler options which need to be passed down in
4188 2000-04-16 Anthony Green <green@redhat.com>
4190 * configure: Rebuilt.
4191 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
4193 2000-04-14 Jakub Jelinek <jakub@redhat.com>
4195 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
4196 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
4197 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
4198 Replace all void * sizeofs with sizeof(int).
4199 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
4200 different than DOUBLE.
4201 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
4202 (ffi_prep_args_v9): New function.
4203 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
4204 (ffi_V9_return_struct): New function.
4205 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
4206 32bit code (not yet cross-arch calls).
4207 * src/sparc/v8.S: Add struct return delay nop.
4209 * src/sparc/v9.S: New file.
4210 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
4211 is used on sparc64 only for structures larger than 32 bytes.
4212 Pass by reference for structures is done for structure arguments
4213 larger than 16 bytes.
4214 * src/ffitest.c (main): Use 64bit rint on sparc64.
4215 Run long long tests on sparc.
4216 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
4218 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
4220 * configure.in (sparc-*-linux*): New supported target.
4221 (sparc64-*-linux*): Likewise.
4222 * configure: Rebuilt.
4223 * Makefile.am: Add v9.S to SPARC files.
4224 * Makefile.in: Likewise.
4225 (LINK): Surround $(CCLD) into double quotes, so that multilib
4226 compiles work correctly.
4228 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
4230 * configure: Rebuilt.
4231 * configure.in: (i*86-*-solaris*): New libffi target. Patch
4232 proposed by Bryce McKinlay.
4234 2000-03-20 Tom Tromey <tromey@cygnus.com>
4236 * Makefile.in: Hand edit for java_raw_api.lo.
4238 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
4240 * config.guess, config.sub: Update from the gcc tree.
4241 Fix for PR libgcj/168.
4243 2000-03-03 Tom Tromey <tromey@cygnus.com>
4245 * Makefile.in: Fixed ia64 by hand.
4247 * configure: Rebuilt.
4248 * configure.in (--enable-multilib): New option.
4249 (libffi_basedir): New subst.
4250 (AC_OUTPUT): Added multilib code.
4252 2000-03-02 Tom Tromey <tromey@cygnus.com>
4254 * Makefile.in: Rebuilt.
4255 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
4258 2000-02-25 Hans Boehm <boehm@acm.org>
4260 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
4262 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
4264 (ffi_prep_raw_closure): Use ffi_translate_args, not
4265 ffi_closure_translate.
4266 * src/java_raw_api.c: New file.
4267 * src/ffitest.c (closure_test_fn): New function.
4268 (main): Define `rint' as long long on IA64. Added new test when
4269 FFI_CLOSURES is defined.
4270 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
4271 (ffi_abi): Recognize IA64.
4272 (ffi_raw): Added `flt' field.
4273 Added "Java raw API" code.
4274 * configure.in: Recognize ia64.
4275 * Makefile.am (TARGET_SRC_IA64): New macro.
4276 (libffi_la_common_SOURCES): Added java_raw_api.c.
4277 (libffi_la_SOURCES): Define in IA64 case.
4279 2000-01-04 Tom Tromey <tromey@cygnus.com>
4281 * Makefile.in: Rebuilt with newer automake.
4283 1999-12-31 Tom Tromey <tromey@cygnus.com>
4285 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
4287 1999-09-01 Tom Tromey <tromey@cygnus.com>
4289 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
4291 * fficonfig.h.in: Rebuilt.
4292 * configure: Rebuilt.
4293 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
4294 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
4295 * acconfig.h: Don't #undef PACKAGE or VERSION.
4297 1999-08-09 Anthony Green <green@cygnus.com>
4299 * include/ffi.h.in: Try to work around messy header problem
4300 with PACKAGE and VERSION.
4302 * configure: Rebuilt.
4303 * configure.in: Change version to 2.00-beta.
4305 * fficonfig.h.in: Rebuilt.
4306 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
4308 * src/x86/ffi.c (ffi_raw_call): Rename.
4310 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
4312 * src/x86/ffi.c (ffi_closure_SYSV): New function.
4313 (ffi_prep_incoming_args_SYSV): Ditto.
4314 (ffi_prep_closure): Ditto.
4315 (ffi_closure_raw_SYSV): Ditto.
4316 (ffi_prep_raw_closure): More ditto.
4317 (ffi_call_raw): Final ditto.
4319 * include/ffi.h.in: Add definitions for closure and raw API.
4321 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4324 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
4326 * src/raw_api.c: New file.
4328 * include/ffi.h.in (ffi_raw): New type.
4329 (UINT_ARG, SINT_ARG): New defines.
4330 (ffi_closure, ffi_raw_closure): New types.
4331 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4333 * configure.in: Add check for endianness and sizeof void*.
4335 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
4336 instead of directly.
4338 * configure: Rebuilt.
4340 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
4342 * configure.in: Add x86 and powerpc BeOS configurations.
4343 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
4345 1999-05-09 Anthony Green <green@cygnus.com>
4347 * configure.in: Add warning about this being beta code.
4348 Remove src/Makefile.am from the picture.
4349 * configure: Rebuilt.
4351 * Makefile.am: Move logic from src/Makefile.am. Add changes
4352 to support libffi as a target library.
4353 * Makefile.in: Rebuilt.
4355 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
4356 Upgraded to new autoconf, automake, libtool.
4360 * LICENSE: Update copyright date.
4362 * src/Makefile.am, src/Makefile.in: Removed.
4364 1998-11-29 Anthony Green <green@cygnus.com>
4366 * include/ChangeLog: Removed.
4367 * src/ChangeLog: Removed.
4368 * src/mips/ChangeLog: Removed.
4369 * src/sparc/ChangeLog: Remboved.
4370 * src/x86/ChangeLog: Removed.
4372 * ChangeLog.v1: Created.