1 2012-01-09 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
3 * configure.ac (i?86-*-*): Set TARGET to X86_64.
4 * configure: Regenerate.
6 2011-12-07 Andrew Pinski <apinski@cavium.com>
9 * src/mips/n32.S: Add ".set mips4".
11 2011-11-21 Andreas Tobler <andreast@fgznet.ch>
13 * configure: Regenerate.
15 2011-11-10 Richard Henderson <rth@redhat.com>
17 * configure.ac (GCC_AS_CFI_PSEUDO_OP): Use it instead of inline check.
18 * configure, aclocal.m4: Rebuild.
20 2011-09-04 Iain Sandoe <iains@gcc.gnu.org>
23 * src/powerpc/darwin_closure.S (stubs): Make the stub binding
24 helper reference track the architecture pointer size.
26 2011-08-25 Andrew Haley <aph@redhat.com>
28 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Remove hard-coded assembly
30 * src/arm/sysv.S (ffi_arm_trampoline): Put them here instead.
32 2011-07-11 Andrew Haley <aph@redhat.com>
34 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Clear icache.
36 2011-06-29 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
38 * testsuite/libffi.call/cls_double_va.c: Move PR number to comment.
39 * testsuite/libffi.call/cls_longdouble_va.c: Likewise.
41 2011-06-29 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
44 * testsuite/libffi.call/cls_double_va.c: xfail dg-output on
46 * testsuite/libffi.call/cls_longdouble_va.c: Likewise.
48 2011-06-14 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
50 * testsuite/libffi.call/huge_struct.c (test_large_fn): Use PRIu8,
51 PRId8 instead of %hhu, %hhd.
52 * testsuite/libffi.call/ffitest.h [__alpha__ && __osf__] (PRId8,
54 [__sgi__] (PRId8, PRIu8): Define.
56 2011-04-29 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
58 * src/alpha/osf.S (UA_SI, FDE_ENCODING, FDE_ENCODE, FDE_ARANGE):
60 Use them to handle ELF vs. ECOFF differences.
61 [__osf__] (_GLOBAL__F_ffi_call_osf): Define.
63 2011-02-13 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
65 * configure: Regenerate.
67 2011-02-09 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
70 * testsuite/libffi.call/cls_pointer.c (main): Cast void * to
72 * testsuite/libffi.call/cls_pointer_stack.c (main): Likewise.
74 2011-02-07 Joel Sherrill <joel.sherrill@oarcorp.com>
76 * libffi/src/m68k/ffi.c: Add RTEMS support for cache flushing.
77 Handle case when CPU variant does not have long double support.
78 * libffi/src/m68k/sysv.S: Add support for mc68000, Coldfire,
79 and cores with soft floating point.
81 2011-02-07 Joel Sherrill <joel.sherrill@oarcorp.com>
83 * configure.ac: Add mips*-*-rtems* support.
84 * configure: Regenerate.
85 * src/mips/ffitarget.h: Ensure needed constants are available
86 for targets which do not have sgidefs.h.
88 2011-01-26 Dave Korn <dave.korn.cygwin@gmail.com>
91 * configure.ac (AM_LTLDFLAGS): Add -bindir option for windows DLLs.
92 * configure: Regenerate.
94 2010-12-18 Iain Sandoe <iains@gcc.gnu.org>
98 * src/powerpc/darwin_closure.S: Provide Darwin64 implementation,
100 * src/powerpc/ffitarget.h (POWERPC_DARWIN64): New,
101 (FFI_TRAMPOLINE_SIZE): Update for Darwin64.
102 * src/powerpc/darwin.S: Provide Darwin64 implementation,
104 * src/powerpc/ffi_darwin.c: Likewise.
106 2010-12-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
108 * configure.ac (libffi_cv_as_ascii_pseudo_op): Use double
110 (libffi_cv_as_string_pseudo_op): Likewise.
111 * configure: Regenerate.
113 2010-12-03 Chung-Lin Tang <cltang@codesourcery.com>
115 * src/arm/sysv.S (ffi_closure_SYSV): Add UNWIND to .pad directive.
116 (ffi_closure_VFP): Same.
117 (ffi_call_VFP): Move down to before ffi_closure_VFP. Add '.fpu vfp'
120 2010-12-01 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
122 * testsuite/libffi.call/ffitest.h [__sgi] (PRId64, PRIu64): Define.
125 2010-11-29 Richard Henderson <rth@redhat.com>
126 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
128 * src/x86/sysv.S (FDE_ENCODING, FDE_ENCODE): Define.
129 (.eh_frame): Use FDE_ENCODING.
130 (.LASFDE1, .LASFDE2, LASFDE3): Simplify with FDE_ENCODE.
132 2010-11-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
134 * testsuite/lib/libffi-dg.exp: Rename ...
135 * testsuite/lib/libffi.exp: ... to this.
136 * libffi/testsuite/libffi.call/call.exp: Don't load libffi-dg.exp.
137 * libffi/testsuite/libffi.special/special.exp: Likewise.
139 2010-10-28 Chung-Lin Tang <cltang@codesourcery.com>
141 * src/arm/ffi.c (ffi_prep_args): Add VFP register argument handling
142 code, new parameter, and return value. Update comments.
143 (ffi_prep_cif_machdep): Add case for VFP struct return values. Add
144 call to layout_vfp_args().
145 (ffi_call_SYSV): Update declaration.
146 (ffi_call_VFP): New declaration.
147 (ffi_call): Add VFP struct return conditions. Call ffi_call_VFP()
149 (ffi_closure_VFP): New declaration.
150 (ffi_closure_SYSV_inner): Add new vfp_args parameter, update call to
151 ffi_prep_incoming_args_SYSV().
152 (ffi_prep_incoming_args_SYSV): Update parameters. Add VFP argument
154 (ffi_prep_closure_loc): Pass ffi_closure_VFP to trampoline
155 construction under VFP hard-float.
156 (rec_vfp_type_p): New function.
158 (place_vfp_arg): Same.
159 (layout_vfp_args): Same.
160 * src/arm/ffitarget.h (ffi_abi): Add FFI_VFP. Define FFI_DEFAULT_ABI
161 based on __ARM_PCS_VFP.
162 (FFI_EXTRA_CIF_FIELDS): Define for adding VFP hard-float specific
164 (FFI_TYPE_STRUCT_VFP_FLOAT): Define internally used type code.
165 (FFI_TYPE_STRUCT_VFP_DOUBLE): Same.
166 * src/arm/sysv.S (ffi_call_SYSV): Change call of ffi_prep_args() to
167 direct call. Move function pointer load upwards.
168 (ffi_call_VFP): New function.
169 (ffi_closure_VFP): Same.
171 * testsuite/lib/libffi-dg.exp (check-flags): New function.
172 (dg-skip-if): New function.
173 * testsuite/libffi.call/cls_double_va.c: Skip if target is arm*-*-*
174 and compiler options include -mfloat-abi=hard.
175 * testsuite/libffi.call/cls_longdouble_va.c: Same.
177 2010-10-01 Jakub Jelinek <jakub@redhat.com>
180 * src/x86/ffi64.c (ffi_prep_cif_machdep): Ensure cif->bytes is
182 * testsuite/libffi.call/many2.c: New test.
184 2010-08-20 Mark Wielaard <mjw@redhat.com>
186 * src/closures.c (open_temp_exec_file_mnt): Check if getmntent_r
189 2010-08-09 Andreas Tobler <andreast@fgznet.ch>
191 * configure.ac: Add target powerpc64-*-freebsd*.
192 * configure: Regenerate.
193 * testsuite/libffi.call/cls_align_longdouble_split.c: Pass
194 -mlong-double-128 only to linux targets.
195 * testsuite/libffi.call/cls_align_longdouble_split2.c: Likewise.
196 * testsuite/libffi.call/cls_longdouble.c: Likewise.
197 * testsuite/libffi.call/huge_struct.c: Likewise.
199 2010-07-10 Evan Phoenix <evan@fallingsnow.net>
201 * src/closures.c (selinux_enabled_check): Fix strncmp usage bug.
203 2010-07-07 Dan HorĂ¡k <dan@danny.cz>
205 * include/ffi.h.in: Protect #define with #ifndef.
206 * src/powerpc/ffitarget.h: Ditto.
207 * src/s390/ffitarget.h: Ditto.
208 * src/sparc/ffitarget.h: Ditto.
210 2010-07-07 Neil Roberts <neil@linux.intel.com>
212 * src/x86/sysv.S (ffi_call_SYSV): Align the stack pointer to
215 2010-07-02 Jakub Jelinek <jakub@redhat.com>
217 * Makefile.am (AM_MAKEFLAGS): Pass also mandir to submakes.
218 * Makefile.in: Regenerated.
220 2010-05-19 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
222 * configure.ac (libffi_cv_as_x86_pcrel): Check for illegal in as
224 (libffi_cv_as_ascii_pseudo_op): Check for .ascii.
225 (libffi_cv_as_string_pseudo_op): Check for .string.
226 * configure: Regenerate.
227 * fficonfig.h.in: Regenerate.
228 * src/x86/sysv.S (.eh_frame): Use .ascii, .string or error.
230 2010-05-05 Michael Kohler <michaelkohler@live.com>
232 * src/dlmalloc.c (dlfree): Fix spelling.
233 * src/ia64/ffi.c (ffi_prep_cif_machdep): Ditto.
234 * configure.ac: Ditto.
235 * configure: Rebuilt.
237 2010-04-13 Dan Witte <dwitte@mozilla.com>
239 * msvcc.sh: Build with -W3 instead of -Wall.
240 * src/powerpc/ffi_darwin.c: Remove build warnings.
241 * src/x86/ffi.c: Ditto.
242 * src/x86/ffitarget.h: Ditto.
244 2010-04-12 Dan Witte <dwitte@mozilla.com>
245 Walter Meinl <wuno@lsvw.de>
247 * configure.ac: Add OS/2 support.
248 * configure: Rebuilt.
249 * src/closures.c: Ditto.
250 * src/dlmalloc.c: Ditto.
251 * src/x86/win32.S: Ditto.
253 2010-04-07 Jakub Jelinek <jakub@redhat.com>
255 * testsuite/libffi.call/err_bad_abi.c: Remove unused args variable.
257 2010-04-02 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
259 * Makefile.in: Regenerate.
260 * aclocal.m4: Regenerate.
261 * include/Makefile.in: Regenerate.
262 * man/Makefile.in: Regenerate.
263 * testsuite/Makefile.in: Regenerate.
265 2010-03-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
267 * configure.ac (libffi_cv_as_x86_64_unwind_section_type): New test.
268 * configure: Regenerate.
269 * fficonfig.h.in: Regenerate.
270 * libffi/src/x86/unix64.S (.eh_frame)
271 [HAVE_AS_X86_64_UNWIND_SECTION_TYPE]: Use @unwind section type.
273 2010-03-14 Matthias Klose <doko@ubuntu.com>
275 * src/x86/ffi64.c: Fix typo in comment.
276 * src/x86/ffi.c: Use /* ... */ comment style.
278 2010-02-24 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
280 * doc/libffi.texi (The Closure API): Fix typo.
281 * doc/libffi.info: Remove.
283 2010-02-15 Matthias Klose <doko@ubuntu.com>
285 * src/arm/sysv.S (__ARM_ARCH__): Define for processor
288 2010-01-15 Anthony Green <green@redhat.com>
290 * README: Add notes on building with Microsoft Visual C++.
292 2010-01-15 Daniel Witte <dwitte@mozilla.com>
294 * msvcc.sh: New file.
296 * src/x86/win32.S: Port assembly routines to MSVC and #ifdef.
297 * src/x86/ffi.c: Tweak function declaration and remove excess
299 * include/ffi.h.in: Add __declspec(align(8)) to typedef struct
302 * src/x86/ffi.c: Merge ffi_call_SYSV and ffi_call_STDCALL into new
303 function ffi_call_win32 on X86_WIN32.
304 * src/x86/win32.S (ffi_call_SYSV): Rename to ffi_call_win32.
305 (ffi_call_STDCALL): Remove.
307 * src/prep_cif.c (ffi_prep_cif): Move stack space allocation code
308 to ffi_prep_cif_machdep for x86.
309 * src/x86/ffi.c (ffi_prep_cif_machdep): To here.
311 2010-01-15 Oliver Kiddle <okiddle@yahoo.co.uk>
313 * src/x86/ffitarget.h (ffi_abi): Check for __i386 and __amd64 for
314 Sun Studio compiler compatibility.
316 2010-01-12 Conrad Irwin <conrad.irwin@gmail.com>
318 * doc/libffi.texi: Add closure example.
320 2010-01-07 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
323 * testsuite/libffi.call/ffitest.h [__alpha__ && __osf__] (PRIdLL,
324 PRIuLL, PRId64, PRIu64, PRIuPTR): Define.
325 * testsuite/libffi.call/cls_align_sint64.c: Add -Wno-format on
327 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
328 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
329 * testsuite/libffi.call/return_ll1.c: Likewise.
330 * testsuite/libffi.call/stret_medium2.c: Likewise.
331 * testsuite/libffi.special/ffitestcxx.h (allocate_mmap): Cast
332 MAP_FAILED to char *.
334 2010-01-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
336 * src/mips/n32.S: Use .abicalls and .eh_frame with __GNUC__.
338 2009-12-31 Anthony Green <green@redhat.com>
340 * README: Update for libffi 3.0.9.
342 2009-12-27 Matthias Klose <doko@ubuntu.com>
344 * configure.ac (HAVE_LONG_DOUBLE): Define for mips when
346 * configure: Rebuilt.
348 2009-12-26 Anthony Green <green@redhat.com>
350 * testsuite/libffi.call/cls_longdouble_va.c: Mark as xfail for
352 * testsuite/libffi.call/cls_double_va.c: Ditto.
354 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
356 * testsuite/libffi.call/ffitest.h: Conditionally include stdint.h
358 * testsuite/libffi.special/unwindtest.cc: Ditto.
360 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
362 * configure.ac: Add amd64-*-openbsd*.
363 * configure: Rebuilt.
364 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Link
365 openbsd programs with -lpthread.
367 2009-12-26 Anthony Green <green@redhat.com>
369 * testsuite/libffi.call/cls_double_va.c,
370 testsuite/libffi.call/cls_longdouble.c,
371 testsuite/libffi.call/cls_longdouble_va.c,
372 testsuite/libffi.call/cls_pointer.c,
373 testsuite/libffi.call/cls_pointer_stack.c: Remove xfail for
374 mips*-*-* and arm*-*-*.
375 * testsuite/libffi.call/cls_align_longdouble_split.c,
376 testsuite/libffi.call/cls_align_longdouble_split2.c,
377 testsuite/libffi.call/stret_medium2.c,
378 testsuite/libffi.call/stret_medium.c,
379 testsuite/libffi.call/stret_large.c,
380 testsuite/libffi.call/stret_large2.c: Remove xfail for arm*-*-*.
382 2009-12-31 Kay Tietz <ktietz70@googlemail.com>
384 * testsuite/libffi.call/ffitest.h,
385 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRuLL): Fix
388 2009-12-31 Carlo Bramini <carlo.bramix@libero.it>
390 * configure.ac (AM_LTLDFLAGS): Define for windows hosts.
391 * Makefile.am (libffi_la_LDFLAGS): Add AM_LTLDFLAGS.
392 * configure: Rebuilt.
393 * Makefile.in: Rebuilt.
395 2009-12-31 Anthony Green <green@redhat.com>
398 * testsuite/libffi.call/huge_struct.c: New test case from Blake
401 2009-12-28 David Edelsohn <edelsohn@gnu.org>
403 * src/powerpc/ffi_darwin.c (ffi_prep_args): Copy abi and nargs to
405 (aix_adjust_aggregate_sizes): New function.
406 (ffi_prep_cif_machdep): Call it.
408 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
410 * configure.ac: Define FFI_MMAP_EXEC_WRIT for the given targets.
411 * configure: Regenerate.
412 * fficonfig.h.in: Likewise.
413 * src/closures.c: Remove the FFI_MMAP_EXEC_WRIT definition for
416 2009-12-26 Andreas Schwab <schwab@linux-m68k.org>
418 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Advance intarg_count
419 when a float arguments is passed in memory.
420 (ffi_closure_helper_SYSV): Mark general registers as used up when
421 a 64bit or soft-float long double argument is passed in memory.
423 2009-12-25 Matthias Klose <doko@ubuntu.com>
425 * man/ffi_call.3: Fix #include in examples.
426 * doc/libffi.texi: Add dircategory.
428 2009-12-25 Frank Everdij <f.p.x.everdij@tudelft.nl>
430 * include/ffi.h.in: Placed '__GNUC__' ifdef around
431 '__attribute__((aligned(8)))' in ffi_closure, fixes compile for
433 * include/ffi_common.h: Added '__sgi' define to non
434 '__attribute__((__mode__()))' integer typedefs.
435 * src/mips/ffi.c (ffi_call, ffi_closure_mips_inner_O32,
436 ffi_closure_mips_inner_N32): Added 'defined(_MIPSEB)' to BE check.
437 (ffi_closure_mips_inner_O32, ffi_closure_mips_inner_N32): Added
438 FFI_LONGDOUBLE support and alignment(N32 only).
439 * src/mips/ffitarget.h: Corrected '#include <sgidefs.h>' for IRIX and
440 fixed non '__attribute__((__mode__()))' integer typedefs.
441 * src/mips/n32.S: Put '#ifdef linux' around '.abicalls' and '.eh_frame'
442 since they are Linux/GNU Assembler specific.
444 2009-12-25 Bradley Smith <brad@brad-smith.co.uk>
446 * configure.ac, Makefile.am, src/avr32/ffi.c,
447 src/avr32/ffitarget.h,
448 src/avr32/sysv.S: Add AVR32 port.
449 * configure, Makefile.in: Rebuilt.
451 2009-12-21 Andreas Tobler <a.tobler@schweiz.org>
453 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
454 * configure: Regenerate.
456 2009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
458 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
460 2009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
462 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
465 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
467 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
470 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
472 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
474 2009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
477 * src/closures.c [X86_64 && __sun__ && __svr4__]
478 (FFI_MMAP_EXEC_WRIT): Define.
480 2009-12-08 David Daney <ddaney@caviumnetworks.com>
482 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
483 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
484 * testsuite/libffi.call/stret_large.c: Same.
485 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
486 * testsuite/libffi.call/stret_large2.c: Same.
487 * testsuite/libffi.call/stret_medium2.c: Same.
489 2009-12-07 David Edelsohn <edelsohn@gnu.org>
491 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
494 2009-12-05 David Edelsohn <edelsohn@gnu.org>
496 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
498 * src/powerpc/aix_closure.S: Same.
500 2009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
502 * Makefile.in: Regenerate.
503 * configure: Regenerate.
504 * include/Makefile.in: Regenerate.
505 * man/Makefile.in: Regenerate.
506 * testsuite/Makefile.in: Regenerate.
508 2009-12-04 David Edelsohn <edelsohn@gnu.org>
510 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
513 2009-12-04 Uros Bizjak <ubizjak@gmail.com>
516 * src/x86/ffi64.c (classify_argument): Update from
517 gcc/config/i386/i386.c.
518 (ffi_closure_unix64_inner): Do not use the address of two consecutive
519 SSE registers directly.
520 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
521 for x86_64 linux targets.
523 2009-12-04 David Edelsohn <edelsohn@gnu.org>
525 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
526 pfr for long double split between fpr13 and stack.
528 2009-12-03 David Edelsohn <edelsohn@gnu.org>
530 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
531 fparg_count twice for long double.
533 2009-12-03 David Edelsohn <edelsohn@gnu.org>
536 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
538 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
540 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
541 Remove xfails for x86 linux targets.
543 2009-12-02 David Edelsohn <edelsohn@gnu.org>
545 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
548 2009-12-01 David Edelsohn <edelsohn@gnu.org>
550 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
551 register usage. Call ffi_prep_args directly. Add long double
552 return value support.
553 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
554 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
555 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
556 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
558 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
560 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
561 Allocate result area between params and FPRs.
563 2009-11-30 David Edelsohn <edelsohn@gnu.org>
566 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
568 * src/powerpc/aix.S: Implement AIX64 version.
569 * src/powerpc/aix_closure.S: Implement AIX64 version.
570 (ffi_closure_ASM): Use extsb, lha and displament addresses.
571 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
573 (ffi_prep_cif_machdep): Same.
575 (ffi_closure_helper_DARWIN): Same.
577 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
580 * testsuite/libffi.call/testclosure.c: New test.
582 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
584 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
585 assembly version use of ___chkstk.
587 2009-09-23 Matthias Klose <doko@ubuntu.com>
589 PR libffi/40242, PR libffi/41443
590 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
591 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
592 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
593 Change the conditionals to __SOFTFP__ || __ARM_EABI__
594 for -mfloat-abi=softfp to work.
596 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
598 PR testsuite/32843 (strikes again)
599 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
600 enable proper extension on char and short.
602 2009-09-15 David Daney <ddaney@caviumnetworks.com>
604 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
605 handling for FFI_TYPE_POINTER.
606 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
607 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
608 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
609 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
610 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
611 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
612 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
613 structure and pointer returns.
614 (ffi_closure_N32): Add handling for pointer returns.
615 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
616 calc_n32_return_struct_flags): Handle soft-float.
617 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
618 (ffi_call_N32): Declare proper argument types.
619 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
622 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
624 * configure.ac (AC_PREREQ): Bump to 2.64.
626 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
628 * Makefile.am (install-html, install-pdf): Remove.
629 * Makefile.in: Regenerate.
631 * Makefile.in: Regenerate.
632 * aclocal.m4: Regenerate.
633 * configure: Regenerate.
634 * fficonfig.h.in: Regenerate.
635 * include/Makefile.in: Regenerate.
636 * man/Makefile.in: Regenerate.
637 * testsuite/Makefile.in: Regenerate.
639 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
641 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
643 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
646 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
647 return types for X86_WIN32.
648 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
649 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
650 _ffi_closure_STDCALL): Likewise.
652 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
653 (dlmmap, dlmunmap): Also use these functions on Cygwin.
655 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
660 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
663 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
665 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
666 to "" before adding paths. (This reinstates an assignment that
667 was removed by my 2009-06-30 commit, but changes the initial
668 value from "." to "".)
670 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
673 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
674 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
677 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
679 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
680 to ld_library_path. Use add_path. Add just find_libgcc_s
681 to ld_library_path, not every libgcc multilib directory.
683 2009-06-16 Wim Lewis <wiml@hhhh.org>
685 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
686 supposed to be callee-saved.
687 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
688 return buffer for odd-size structs.
690 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
693 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
694 allow_stack_execute for Darwin.
696 2009-06-16 Andrew Haley <aph@redhat.com>
698 * configure.ac (TARGETDIR): Add missing blank lines.
699 * configure: Regenerate.
701 2009-06-16 Andrew Haley <aph@redhat.com>
703 * testsuite/libffi.call/cls_align_sint64.c,
704 testsuite/libffi.call/cls_align_uint64.c,
705 testsuite/libffi.call/cls_longdouble_va.c,
706 testsuite/libffi.call/cls_ulonglong.c,
707 testsuite/libffi.call/return_ll1.c,
708 testsuite/libffi.call/stret_medium2.c: Fix printf format
710 * testsuite/libffi.call/ffitest.h,
711 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
713 2009-06-15 Andrew Haley <aph@redhat.com>
715 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
716 * testsuite/libffi.call/err_bad_abi.c: Likewise.
718 2009-06-12 Andrew Haley <aph@redhat.com>
720 * Makefile.am: Remove info_TEXINFOS.
722 2009-06-12 Andrew Haley <aph@redhat.com>
724 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
725 testsuite/libffi.call/cls_align_uint64.c,
726 testsuite/libffi.call/cls_ulonglong.c,
727 testsuite/libffi.call/return_ll1.c,
728 testsuite/libffi.call/stret_medium2.c: Fix printf format
730 testsuite/libffi.special/unwindtest.cc: include stdint.h.
732 2009-06-11 Timothy Wall <twall@users.sf.net>
737 include/ffi_common.h,
742 src/x86/win64.S (new),
743 README: Added win64 support (mingw or MSVC)
747 testsuite/Makefile.in,
749 aclocal.m4: Regenerated
750 * ltcf-c.sh: properly escape cygwin/w32 path
751 * man/ffi_call.3: Clarify size requirements for return value.
752 * src/x86/ffi64.c: Fix filename in comment.
753 * src/x86/win32.S: Remove unused extern.
755 * testsuite/libffi.call/closure_fn0.c,
756 testsuite/libffi.call/closure_fn1.c,
757 testsuite/libffi.call/closure_fn2.c,
758 testsuite/libffi.call/closure_fn3.c,
759 testsuite/libffi.call/closure_fn4.c,
760 testsuite/libffi.call/closure_fn5.c,
761 testsuite/libffi.call/closure_fn6.c,
762 testsuite/libffi.call/closure_stdcall.c,
763 testsuite/libffi.call/cls_12byte.c,
764 testsuite/libffi.call/cls_16byte.c,
765 testsuite/libffi.call/cls_18byte.c,
766 testsuite/libffi.call/cls_19byte.c,
767 testsuite/libffi.call/cls_1_1byte.c,
768 testsuite/libffi.call/cls_20byte.c,
769 testsuite/libffi.call/cls_20byte1.c,
770 testsuite/libffi.call/cls_24byte.c,
771 testsuite/libffi.call/cls_2byte.c,
772 testsuite/libffi.call/cls_3_1byte.c,
773 testsuite/libffi.call/cls_3byte1.c,
774 testsuite/libffi.call/cls_3byte2.c,
775 testsuite/libffi.call/cls_4_1byte.c,
776 testsuite/libffi.call/cls_4byte.c,
777 testsuite/libffi.call/cls_5_1_byte.c,
778 testsuite/libffi.call/cls_5byte.c,
779 testsuite/libffi.call/cls_64byte.c,
780 testsuite/libffi.call/cls_6_1_byte.c,
781 testsuite/libffi.call/cls_6byte.c,
782 testsuite/libffi.call/cls_7_1_byte.c,
783 testsuite/libffi.call/cls_7byte.c,
784 testsuite/libffi.call/cls_8byte.c,
785 testsuite/libffi.call/cls_9byte1.c,
786 testsuite/libffi.call/cls_9byte2.c,
787 testsuite/libffi.call/cls_align_double.c,
788 testsuite/libffi.call/cls_align_float.c,
789 testsuite/libffi.call/cls_align_longdouble.c,
790 testsuite/libffi.call/cls_align_longdouble_split.c,
791 testsuite/libffi.call/cls_align_longdouble_split2.c,
792 testsuite/libffi.call/cls_align_pointer.c,
793 testsuite/libffi.call/cls_align_sint16.c,
794 testsuite/libffi.call/cls_align_sint32.c,
795 testsuite/libffi.call/cls_align_sint64.c,
796 testsuite/libffi.call/cls_align_uint16.c,
797 testsuite/libffi.call/cls_align_uint32.c,
798 testsuite/libffi.call/cls_align_uint64.c,
799 testsuite/libffi.call/cls_dbls_struct.c,
800 testsuite/libffi.call/cls_double.c,
801 testsuite/libffi.call/cls_double_va.c,
802 testsuite/libffi.call/cls_float.c,
803 testsuite/libffi.call/cls_longdouble.c,
804 testsuite/libffi.call/cls_longdouble_va.c,
805 testsuite/libffi.call/cls_multi_schar.c,
806 testsuite/libffi.call/cls_multi_sshort.c,
807 testsuite/libffi.call/cls_multi_sshortchar.c,
808 testsuite/libffi.call/cls_multi_uchar.c,
809 testsuite/libffi.call/cls_multi_ushort.c,
810 testsuite/libffi.call/cls_multi_ushortchar.c,
811 testsuite/libffi.call/cls_pointer.c,
812 testsuite/libffi.call/cls_pointer_stack.c,
813 testsuite/libffi.call/cls_schar.c,
814 testsuite/libffi.call/cls_sint.c,
815 testsuite/libffi.call/cls_sshort.c,
816 testsuite/libffi.call/cls_uchar.c,
817 testsuite/libffi.call/cls_uint.c,
818 testsuite/libffi.call/cls_ulonglong.c,
819 testsuite/libffi.call/cls_ushort.c,
820 testsuite/libffi.call/err_bad_abi.c,
821 testsuite/libffi.call/err_bad_typedef.c,
822 testsuite/libffi.call/float2.c,
823 testsuite/libffi.call/huge_struct.c,
824 testsuite/libffi.call/nested_struct.c,
825 testsuite/libffi.call/nested_struct1.c,
826 testsuite/libffi.call/nested_struct10.c,
827 testsuite/libffi.call/nested_struct2.c,
828 testsuite/libffi.call/nested_struct3.c,
829 testsuite/libffi.call/nested_struct4.c,
830 testsuite/libffi.call/nested_struct5.c,
831 testsuite/libffi.call/nested_struct6.c,
832 testsuite/libffi.call/nested_struct7.c,
833 testsuite/libffi.call/nested_struct8.c,
834 testsuite/libffi.call/nested_struct9.c,
835 testsuite/libffi.call/problem1.c,
836 testsuite/libffi.call/return_ldl.c,
837 testsuite/libffi.call/return_ll1.c,
838 testsuite/libffi.call/stret_large.c,
839 testsuite/libffi.call/stret_large2.c,
840 testsuite/libffi.call/stret_medium.c,
841 testsuite/libffi.call/stret_medium2.c,
842 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
843 of checking for MMAP. Use intptr_t instead of long casts.
845 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
847 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
848 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
849 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
851 2009-06-09 Andrew Haley <aph@redhat.com>
853 * src/x86/freebsd.S: Add missing file.
855 2009-06-08 Andrew Haley <aph@redhat.com>
857 Import from libffi 3.0.8:
859 * doc/libffi.texi: New file.
860 * doc/libffi.info: Likewise.
861 * doc/stamp-vti: Likewise.
862 * man/Makefile.am: New file.
863 * man/ffi_call.3: New file.
865 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
867 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
869 * configure.ac: Bump version to 3.0.8.
870 parisc*-*-linux*: Add.
871 i386-*-freebsd* | i386-*-openbsd*: Add.
872 powerpc-*-beos*: Add.
873 AM_CONDITIONAL X86_FREEBSD: Add.
874 AC_CONFIG_FILES: Add man/Makefile.
876 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
878 2009-06-08 Andrew Haley <aph@redhat.com>
880 * README: Import from libffi 3.0.8.
882 2009-06-08 Andrew Haley <aph@redhat.com>
884 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
885 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
886 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
887 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
889 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
890 * testsuite/libffi.call/stret_medium.c: Likewise.
891 * testsuite/libffi.call/stret_large2.c: Likewise.
892 * testsuite/libffi.call/stret_large.c: Likewise.
894 2008-12-26 Timothy Wall <twall@users.sf.net>
896 * testsuite/libffi.call/cls_longdouble.c,
897 testsuite/libffi.call/cls_longdouble_va.c,
898 testsuite/libffi.call/cls_align_longdouble.c,
899 testsuite/libffi.call/cls_align_longdouble_split.c,
900 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
901 failures on x86_64 cygwin/mingw.
903 2008-12-22 Timothy Wall <twall@users.sf.net>
905 * testsuite/libffi.call/closure_fn0.c,
906 testsuite/libffi.call/closure_fn1.c,
907 testsuite/libffi.call/closure_fn2.c,
908 testsuite/libffi.call/closure_fn3.c,
909 testsuite/libffi.call/closure_fn4.c,
910 testsuite/libffi.call/closure_fn5.c,
911 testsuite/libffi.call/closure_fn6.c,
912 testsuite/libffi.call/closure_loc_fn0.c,
913 testsuite/libffi.call/closure_stdcall.c,
914 testsuite/libffi.call/cls_align_pointer.c,
915 testsuite/libffi.call/cls_pointer.c,
916 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
917 pointer to integer (intptr_t).
918 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
920 2008-07-24 Anthony Green <green@redhat.com>
922 * testsuite/libffi.call/cls_dbls_struct.c,
923 testsuite/libffi.call/cls_double_va.c,
924 testsuite/libffi.call/cls_longdouble.c,
925 testsuite/libffi.call/cls_longdouble_va.c,
926 testsuite/libffi.call/cls_pointer.c,
927 testsuite/libffi.call/cls_pointer_stack.c,
928 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
931 2008-03-04 Anthony Green <green@redhat.com>
935 * testsuite/libffi.call/cls_align_longdouble_split2.c
936 testsuite/libffi.call/cls_align_longdouble_split.c
937 testsuite/libffi.call/cls_dbls_struct.c
938 testsuite/libffi.call/cls_double_va.c
939 testsuite/libffi.call/cls_longdouble.c
940 testsuite/libffi.call/cls_longdouble_va.c
941 testsuite/libffi.call/cls_pointer.c
942 testsuite/libffi.call/cls_pointer_stack.c
943 testsuite/libffi.call/err_bad_abi.c
944 testsuite/libffi.call/err_bad_typedef.c
945 testsuite/libffi.call/stret_large2.c
946 testsuite/libffi.call/stret_large.c
947 testsuite/libffi.call/stret_medium2.c
948 testsuite/libffi.call/stret_medium.c: New tests from Apple.
950 2009-06-05 Andrew Haley <aph@redhat.com>
952 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
955 2009-06-04 Andrew Haley <aph@redhat.com>
957 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
960 2008-02-26 Anthony Green <green@redhat.com>
961 Thomas Heller <theller@ctypes.org>
963 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
966 2008-02-03 Timothy Wall <twall@users.sf.net>
968 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
969 offset based on code pointer, not data pointer.
971 2008-01-31 Timothy Wall <twall@users.sf.net>
973 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
975 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
977 * src/x86/win32.S: Add assembly for stdcall closure.
978 * src/x86/ffi.c: Initialize stdcall closure trampoline.
980 2009-06-04 Andrew Haley <aph@redhat.com>
982 * include/ffi.h.in: Change void (*)() to void (*)(void).
983 * src/x86/ffi.c: Likewise.
985 2009-06-04 Andrew Haley <aph@redhat.com>
987 * src/powerpc/ppc_closure.S: Insert licence header.
988 * src/powerpc/linux64_closure.S: Likewise.
989 * src/m68k/sysv.S: Likewise.
991 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
992 * src/powerpc/ffi.c: Likewise.
993 * src/powerpc/ffi_darwin.c: Likewise.
994 * src/m32r/ffi.c: Likewise.
995 * src/sh64/ffi.c: Likewise.
996 * src/x86/ffi64.c: Likewise.
997 * src/alpha/ffi.c: Likewise.
998 * src/alpha/osf.S: Likewise.
999 * src/frv/ffi.c: Likewise.
1000 * src/s390/ffi.c: Likewise.
1001 * src/pa/ffi.c: Likewise.
1002 * src/pa/hpux32.S: Likewise.
1003 * src/ia64/unix.S: Likewise.
1004 * src/ia64/ffi.c: Likewise.
1005 * src/sparc/ffi.c: Likewise.
1006 * src/mips/ffi.c: Likewise.
1007 * src/sh/ffi.c: Likewise.
1009 2008-02-15 David Daney <ddaney@avtrex.com>
1011 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
1012 Define (conditionally), and use it to include cachectl.h.
1013 (ffi_prep_closure_loc): Fix cache flushing.
1014 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
1016 2009-06-04 Andrew Haley <aph@redhat.com>
1019 src/arm/ffitarget.h,
1022 src/powerpc/ffitarget.h,
1024 src/sh64/ffitarget.h,
1029 src/x86/ffitarget.h,
1035 src/alpha/ffitarget.h,
1038 src/m68k/ffitarget.h,
1039 src/frv/ffitarget.h,
1041 src/s390/ffitarget.h,
1043 src/cris/ffitarget.h,
1048 src/ia64/ffitarget.h,
1051 src/ia64/ia64_flags.h,
1055 src/sparc/ffitarget.h,
1058 src/mips/ffitarget.h,
1065 src/sh/sysv.S: Update license text.
1067 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
1069 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
1070 (.eh_frame): Add FDE for it.
1072 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
1074 * configure.ac: Also check if assembler supports pc-relative
1075 relocs on X86_WIN32 targets.
1076 * configure: Regenerate.
1077 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
1078 (_ffi_call_SYSV): Add missing function type symbol .def and
1079 add EH markup labels.
1080 (_ffi_call_STDCALL): Likewise.
1081 (_ffi_closure_SYSV): Likewise.
1082 (_ffi_closure_raw_SYSV): Likewise.
1083 (.eh_frame): Add hand-crafted EH data.
1085 2009-04-09 Jakub Jelinek <jakub@redhat.com>
1087 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
1088 version 3 of the GNU General Public License and to point readers
1089 at the COPYING3 file and the FSF's license web page.
1090 * testsuite/libffi.call/call.exp: Likewise.
1091 * testsuite/libffi.special/special.exp: Likewise.
1093 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1095 * configure: Regenerate.
1097 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1100 * configure.ac (HAVE_AS_X86_PCREL): New test.
1101 * configure: Regenerate.
1102 * fficonfig.h.in: Regenerate.
1103 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
1104 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
1105 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
1106 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
1107 * src/x86/unix64.S (.Lstore_table): Move to .text section.
1108 (.Lload_table): Likewise.
1109 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
1111 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1113 * configure: Regenerate.
1115 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
1117 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
1118 signed/unsigned int8/16 return values.
1119 * src/sparc/v8.S (ffi_call_v8): Likewise.
1120 (ffi_closure_v8): Likewise.
1122 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
1123 Steve Ellcey <sje@cup.hp.com>
1125 * configure: Regenerate for new libtool.
1126 * Makefile.in: Ditto.
1127 * include/Makefile.in: Ditto.
1128 * aclocal.m4: Ditto.
1130 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
1132 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
1133 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
1134 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
1135 Adjust copyright notice.
1136 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
1137 register or two register to use for FFI_SYSV structs.
1138 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
1139 (ffi_closure_helper_SYSV): Fix the return type for
1140 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
1141 Adjust copyright notice.
1143 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
1145 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
1148 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1150 * configure: Regenerate.
1151 * include/Makefile.in: Regenerate.
1152 * testsuite/Makefile.in: Regenerate.
1154 2008-06-07 Joseph Myers <joseph@codesourcery.com>
1156 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
1157 powerpc-*-beos*): Remove.
1158 * configure: Regenerate.
1160 2008-05-09 Julian Brown <julian@codesourcery.com>
1162 * Makefile.am (LTLDFLAGS): New.
1163 (libffi_la_LDFLAGS): Use above.
1164 * Makefile.in: Regenerate.
1166 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
1169 * aclocal.m4: Regenerate.
1170 * configure: Regenerate.
1172 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
1174 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
1175 * src/sh64/sysv.S: Likewise.
1177 2008-03-26 Daniel Jacobowitz <dan@debian.org>
1179 * src/arm/sysv.S: Fix ARM comment marker.
1181 2008-03-26 Jakub Jelinek <jakub@redhat.com>
1183 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
1184 * src/s390/sysv.S: Likewise.
1185 * src/powerpc/ppc_closure.S: Likewise.
1186 * src/powerpc/sysv.S: Likewise.
1187 * src/x86/unix64.S: Likewise.
1188 * src/x86/sysv.S: Likewise.
1189 * src/sparc/v8.S: Likewise.
1190 * src/sparc/v9.S: Likewise.
1191 * src/m68k/sysv.S: Likewise.
1192 * src/arm/sysv.S: Likewise.
1194 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1196 * aclocal.m4: Regenerate.
1197 * configure: Likewise.
1198 * Makefile.in: Likewise.
1199 * include/Makefile.in: Likewise.
1200 * testsuite/Makefile.in: Likewise.
1202 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
1203 Andreas Tobler <a.tobler@schweiz.org>
1205 * configure.ac: Add amd64-*-freebsd* target.
1206 * configure: Regenerate.
1208 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
1211 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
1214 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
1217 2008-01-24 David Edelsohn <edelsohn@gnu.org>
1219 * configure: Regenerate.
1221 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
1223 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
1225 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
1228 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
1229 signed/unsigned int8/16 for X86_DARWIN.
1230 Updated copyright info.
1231 Handle one and two byte structs with special cif->flags.
1232 * src/x86/ffitarget.h: Add special types for one and two byte structs.
1233 Updated copyright info.
1234 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
1236 Remove code to pop args from the stack after call.
1237 Special-case signed/unsigned for int8/16, one and two byte structs.
1238 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1239 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1241 Updated copyright info.
1243 2007-12-08 David Daney <ddaney@avtrex.com>
1245 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
1246 SUBU, add with ADDU and use smaller code sequences.
1248 2007-12-07 David Daney <ddaney@avtrex.com>
1250 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
1253 2007-12-06 David Daney <ddaney@avtrex.com>
1255 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
1257 (ffi_java_raw): New typedef.
1258 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
1259 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
1261 (ffi_java_raw_closure) : Same.
1262 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
1264 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
1265 FFI_SIZEOF_JAVA_RAW.
1266 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
1267 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
1268 sizeof(ffi_java_raw) for alignment calculations.
1269 (ffi_java_ptrarray_to_raw): Same.
1270 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
1271 if FFI_SIZEOF_JAVA_RAW == 4.
1272 (ffi_java_raw_to_rvalue): Same.
1273 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
1274 (ffi_java_translate_args): Same.
1275 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
1277 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
1279 2007-12-06 David Daney <ddaney@avtrex.com>
1281 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
1284 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
1287 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
1288 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
1289 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
1290 set the NUM_FPR_ARG_REGISTERS according to.
1291 Add support for potential soft-float support under hard-float
1293 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
1294 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
1295 to the FFI_LINUX_SOFT_FLOAT ABI.
1296 (ffi_prep_cif_machdep): Likewise.
1297 (ffi_closure_helper_SYSV): Likewise.
1298 * src/powerpc/ppc_closure.S: Make sure not to store float/double
1299 on archs where __NO_FPRS__ is true.
1300 Add FFI_TYPE_UINT128 support.
1301 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
1302 Adjust copyright notice.
1304 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
1306 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
1307 * include/ffi_common.h: ... here.
1310 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
1312 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
1313 * src/powerpc/linux64.S: Likewise.
1314 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
1315 * testsuite/libffi.call/return_ldl.c: New test.
1317 2007-09-04 <aph@redhat.com>
1319 * src/arm/sysv.S (UNWIND): New.
1320 (Whole file): Conditionally compile unwinder directives.
1321 * src/arm/sysv.S: Add unwinder directives.
1323 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
1324 Only treat r0 as a struct address if we're actually returning a
1326 Only copy the bytes that are actually within a struct.
1327 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
1328 is returned in r0, not passed by address.
1329 (ffi_call): Allocate a word-sized temporary for the case where
1330 a composite is returned in r0.
1331 (ffi_prep_incoming_args_SYSV): Align as necessary.
1333 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
1335 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
1336 directly using the sys_cacheflush syscall.
1338 2007-07-27 Andrew Haley <aph@redhat.com>
1340 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
1342 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
1344 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
1345 * configure.ac: Likewise.
1346 * Makefile.in: Regenerate.
1347 * include/Makefile.in: Likewise.
1348 * testsuite/Makefile.in: Likewise.
1349 * configure: Likewise.
1351 2007-08-24 David Daney <ddaney@avtrex.com>
1353 * testsuite/libffi.call/return_sl.c: New test.
1355 2007-08-10 David Daney <ddaney@avtrex.com>
1357 * testsuite/libffi.call/cls_multi_ushort.c,
1358 testsuite/libffi.call/cls_align_uint16.c,
1359 testsuite/libffi.call/nested_struct1.c,
1360 testsuite/libffi.call/nested_struct3.c,
1361 testsuite/libffi.call/cls_7_1_byte.c,
1362 testsuite/libffi.call/nested_struct5.c,
1363 testsuite/libffi.call/cls_double.c,
1364 testsuite/libffi.call/nested_struct7.c,
1365 testsuite/libffi.call/cls_sint.c,
1366 testsuite/libffi.call/nested_struct9.c,
1367 testsuite/libffi.call/cls_20byte1.c,
1368 testsuite/libffi.call/cls_multi_sshortchar.c,
1369 testsuite/libffi.call/cls_align_sint64.c,
1370 testsuite/libffi.call/cls_3byte2.c,
1371 testsuite/libffi.call/cls_multi_schar.c,
1372 testsuite/libffi.call/cls_multi_uchar.c,
1373 testsuite/libffi.call/cls_19byte.c,
1374 testsuite/libffi.call/cls_9byte1.c,
1375 testsuite/libffi.call/cls_align_float.c,
1376 testsuite/libffi.call/closure_fn1.c,
1377 testsuite/libffi.call/problem1.c,
1378 testsuite/libffi.call/closure_fn3.c,
1379 testsuite/libffi.call/cls_sshort.c,
1380 testsuite/libffi.call/closure_fn5.c,
1381 testsuite/libffi.call/cls_align_double.c,
1382 testsuite/libffi.call/nested_struct.c,
1383 testsuite/libffi.call/cls_2byte.c,
1384 testsuite/libffi.call/nested_struct10.c,
1385 testsuite/libffi.call/cls_4byte.c,
1386 testsuite/libffi.call/cls_6byte.c,
1387 testsuite/libffi.call/cls_8byte.c,
1388 testsuite/libffi.call/cls_multi_sshort.c,
1389 testsuite/libffi.call/cls_align_sint16.c,
1390 testsuite/libffi.call/cls_align_uint32.c,
1391 testsuite/libffi.call/cls_20byte.c,
1392 testsuite/libffi.call/cls_float.c,
1393 testsuite/libffi.call/nested_struct2.c,
1394 testsuite/libffi.call/cls_5_1_byte.c,
1395 testsuite/libffi.call/nested_struct4.c,
1396 testsuite/libffi.call/cls_24byte.c,
1397 testsuite/libffi.call/nested_struct6.c,
1398 testsuite/libffi.call/cls_64byte.c,
1399 testsuite/libffi.call/nested_struct8.c,
1400 testsuite/libffi.call/cls_uint.c,
1401 testsuite/libffi.call/cls_multi_ushortchar.c,
1402 testsuite/libffi.call/cls_schar.c,
1403 testsuite/libffi.call/cls_uchar.c,
1404 testsuite/libffi.call/cls_align_uint64.c,
1405 testsuite/libffi.call/cls_ulonglong.c,
1406 testsuite/libffi.call/cls_align_longdouble.c,
1407 testsuite/libffi.call/cls_1_1byte.c,
1408 testsuite/libffi.call/cls_12byte.c,
1409 testsuite/libffi.call/cls_3_1byte.c,
1410 testsuite/libffi.call/cls_3byte1.c,
1411 testsuite/libffi.call/cls_4_1byte.c,
1412 testsuite/libffi.call/cls_6_1_byte.c,
1413 testsuite/libffi.call/cls_16byte.c,
1414 testsuite/libffi.call/cls_18byte.c,
1415 testsuite/libffi.call/closure_fn0.c,
1416 testsuite/libffi.call/cls_9byte2.c,
1417 testsuite/libffi.call/closure_fn2.c,
1418 testsuite/libffi.call/closure_fn4.c,
1419 testsuite/libffi.call/cls_ushort.c,
1420 testsuite/libffi.call/closure_fn6.c,
1421 testsuite/libffi.call/cls_5byte.c,
1422 testsuite/libffi.call/cls_align_pointer.c,
1423 testsuite/libffi.call/cls_7byte.c,
1424 testsuite/libffi.call/cls_align_sint32.c,
1425 testsuite/libffi.special/unwindtest_ffi_call.cc,
1426 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
1428 2007-08-10 David Daney <ddaney@avtrex.com>
1431 * configure.ac: Don't treat mips64 as a special case.
1432 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
1433 * configure: Regenerate
1434 * Makefile.in: Ditto.
1435 * fficonfig.h.in: Ditto.
1436 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
1437 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
1438 (FFI_DEFAULT_ABI): Set for n64 case.
1439 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
1440 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
1441 (ffi_closure_N32): New function.
1442 (.eh_frame): New section
1443 * src/mips/o32.S: Clean up comments.
1444 (ffi_closure_O32): Pass ffi_closure parameter in $12.
1445 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
1446 _MIPS_SIM == _ABIN32 throughout.
1447 (FFI_MIPS_STOP_HERE): New, use in place of
1449 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1450 to support n32/n64 ABIs.
1451 (calc_n32_struct_flags): Rewrite.
1452 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1453 position of flag bits.
1454 (ffi_prep_cif_machdep): Rewrite n32 portion.
1455 (ffi_call): Enable for n64. Add special handling for small structure
1457 (ffi_prep_closure_loc): Add n32 and n64 support.
1458 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1459 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1461 2007-08-08 David Daney <ddaney@avtrex.com>
1463 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1464 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1466 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1467 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1468 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1469 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1470 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1471 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1472 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1473 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1474 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1475 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1476 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1477 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1478 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1479 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1480 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1481 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1482 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1483 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1484 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1485 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1486 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1487 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1488 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1489 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1490 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1491 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1492 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1493 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1494 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1495 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1496 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1498 2007-08-07 Andrew Haley <aph@redhat.com>
1500 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1503 2007-08-06 Andrew Haley <aph@redhat.com>
1506 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1507 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1510 2007-08-02 David Daney <ddaney@avtrex.com>
1512 * testsuite/libffi.call/return_ul.c (main): Define return type as
1513 ffi_arg. Use proper printf conversion specifier.
1515 2007-07-30 Andrew Haley <aph@redhat.com>
1518 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1519 signed/unsigned int8/16.
1520 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1522 Remove code to pop args from the stack after call.
1523 Special-case signed/unsigned int8/16.
1524 * testsuite/libffi.call/return_sc.c (main): Revert.
1526 2007-07-26 Richard Guenther <rguenther@suse.de>
1529 * testsuite/libffi.call/return_sc.c (main): Verify call
1530 result as signed char, not ffi_arg.
1532 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1534 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1535 * configure: Regenerate.
1537 2007-07-11 David Daney <ddaney@avtrex.com>
1539 * src/mips/ffi.c: Don't include sys/cachectl.h.
1540 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1543 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1545 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1546 from (ffi_prep_closure): ... this.
1547 (FFI_INIT_TRAMPOLINE): Adjust.
1549 2005-12-31 Phil Blundell <pb@reciva.com>
1551 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1552 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1553 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1554 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1555 (FFI_CLOSURES): Enable closure support.
1557 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1559 * testsuite/libffi.call/cls_multi_ushort.c,
1560 testsuite/libffi.call/cls_align_uint16.c,
1561 testsuite/libffi.call/nested_struct1.c,
1562 testsuite/libffi.call/nested_struct3.c,
1563 testsuite/libffi.call/cls_7_1_byte.c,
1564 testsuite/libffi.call/cls_double.c,
1565 testsuite/libffi.call/nested_struct5.c,
1566 testsuite/libffi.call/nested_struct7.c,
1567 testsuite/libffi.call/cls_sint.c,
1568 testsuite/libffi.call/nested_struct9.c,
1569 testsuite/libffi.call/cls_20byte1.c,
1570 testsuite/libffi.call/cls_multi_sshortchar.c,
1571 testsuite/libffi.call/cls_align_sint64.c,
1572 testsuite/libffi.call/cls_3byte2.c,
1573 testsuite/libffi.call/cls_multi_schar.c,
1574 testsuite/libffi.call/cls_multi_uchar.c,
1575 testsuite/libffi.call/cls_19byte.c,
1576 testsuite/libffi.call/cls_9byte1.c,
1577 testsuite/libffi.call/cls_align_float.c,
1578 testsuite/libffi.call/closure_fn1.c,
1579 testsuite/libffi.call/problem1.c,
1580 testsuite/libffi.call/closure_fn3.c,
1581 testsuite/libffi.call/cls_sshort.c,
1582 testsuite/libffi.call/closure_fn5.c,
1583 testsuite/libffi.call/cls_align_double.c,
1584 testsuite/libffi.call/cls_2byte.c,
1585 testsuite/libffi.call/nested_struct.c,
1586 testsuite/libffi.call/nested_struct10.c,
1587 testsuite/libffi.call/cls_4byte.c,
1588 testsuite/libffi.call/cls_6byte.c,
1589 testsuite/libffi.call/cls_8byte.c,
1590 testsuite/libffi.call/cls_multi_sshort.c,
1591 testsuite/libffi.call/cls_align_uint32.c,
1592 testsuite/libffi.call/cls_align_sint16.c,
1593 testsuite/libffi.call/cls_float.c,
1594 testsuite/libffi.call/cls_20byte.c,
1595 testsuite/libffi.call/cls_5_1_byte.c,
1596 testsuite/libffi.call/nested_struct2.c,
1597 testsuite/libffi.call/cls_24byte.c,
1598 testsuite/libffi.call/nested_struct4.c,
1599 testsuite/libffi.call/nested_struct6.c,
1600 testsuite/libffi.call/cls_64byte.c,
1601 testsuite/libffi.call/nested_struct8.c,
1602 testsuite/libffi.call/cls_uint.c,
1603 testsuite/libffi.call/cls_multi_ushortchar.c,
1604 testsuite/libffi.call/cls_schar.c,
1605 testsuite/libffi.call/cls_uchar.c,
1606 testsuite/libffi.call/cls_align_uint64.c,
1607 testsuite/libffi.call/cls_ulonglong.c,
1608 testsuite/libffi.call/cls_align_longdouble.c,
1609 testsuite/libffi.call/cls_1_1byte.c,
1610 testsuite/libffi.call/cls_12byte.c,
1611 testsuite/libffi.call/cls_3_1byte.c,
1612 testsuite/libffi.call/cls_3byte1.c,
1613 testsuite/libffi.call/cls_4_1byte.c,
1614 testsuite/libffi.call/cls_6_1_byte.c,
1615 testsuite/libffi.call/cls_16byte.c,
1616 testsuite/libffi.call/cls_18byte.c,
1617 testsuite/libffi.call/closure_fn0.c,
1618 testsuite/libffi.call/cls_9byte2.c,
1619 testsuite/libffi.call/closure_fn2.c,
1620 testsuite/libffi.call/closure_fn4.c,
1621 testsuite/libffi.call/cls_ushort.c,
1622 testsuite/libffi.call/closure_fn6.c,
1623 testsuite/libffi.call/cls_5byte.c,
1624 testsuite/libffi.call/cls_align_pointer.c,
1625 testsuite/libffi.call/cls_7byte.c,
1626 testsuite/libffi.call/cls_align_sint32.c,
1627 testsuite/libffi.special/unwindtest_ffi_call.cc,
1628 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1630 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1632 * aclocal.m4: Regenerated.
1634 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1636 * configure: Regenerate.
1638 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1640 * Makefile.in: Regenerate.
1641 * configure: Regenerate.
1642 * aclocal.m4: Regenerate.
1643 * include/Makefile.in: Regenerate.
1644 * testsuite/Makefile.in: Regenerate.
1646 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1648 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1649 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1650 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1651 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1652 (FFI_CLOSURES): Enable closure support.
1654 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1656 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1657 * configure: Regenerate.
1658 * fficonfig.h.in: Regenerate.
1659 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1660 CFI_OFFSET,CFI_DEF_CFA): New macros.
1661 (ffi_call_SYSV): Add callframe annotation.
1663 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1665 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1666 numerous test suite failures.
1667 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1669 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1671 * Makefile.am (EXTRA_DIST): Bring up to date.
1672 * Makefile.in: Regenerate.
1673 * src/frv/eabi.S: Remove RCS keyword.
1675 2007-04-06 Richard Henderson <rth@redhat.com>
1677 * configure.ac: Tidy target case.
1678 (HAVE_LONG_DOUBLE): Allow the target to override.
1679 * configure: Regenerate.
1680 * include/ffi.h.in: Don't define ffi_type_foo if
1681 LIBFFI_HIDE_BASIC_TYPES is defined.
1682 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1684 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1685 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1686 (ffi_type_longdouble): Special case for Alpha. Don't define
1687 if long double == double.
1689 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1690 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1691 (ffi_call, ffi_closure_osf_inner): Likewise.
1692 (ffi_closure_osf_inner): Likewise. Mark hidden.
1693 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1694 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1695 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1696 (load_table): Handle 128-bit long double.
1698 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1700 2007-04-06 Tom Tromey <tromey@redhat.com>
1703 * README: Fixed bug in example.
1705 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1707 * src/closures.c: Include sys/statfs.h.
1708 (_GNU_SOURCE): Define on Linux.
1709 (FFI_MMAP_EXEC_SELINUX): Define.
1710 (selinux_enabled): New variable.
1711 (selinux_enabled_check): New function.
1712 (is_selinux_enabled): Define.
1715 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1717 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1718 Use 'volatile float sum' to create sum of floats to avoid false
1719 negative due to excess precision on ix86 targets.
1722 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1724 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1726 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1728 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1730 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1731 (ffi_prep_closure_loc): New.
1732 (ffi_prep_raw_closure_loc): New.
1733 (ffi_prep_java_raw_closure_loc): New.
1734 * src/closures.c: New file.
1735 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1736 Replace sflags with exec_offset.
1737 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1738 sub_segment_exec_offset): New macros.
1739 (get_segment_flags, set_segment_flags, check_segment_merge): New
1741 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1742 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1743 destroy_mspace): Use new macros.
1744 (sys_alloc): Silence warning.
1745 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1746 * Makefile.in: Rebuilt.
1747 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1748 terms of ffi_prep_closure_loc.
1749 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1751 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1753 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1755 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1756 the renamed version.
1757 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1758 (ffi_prep_closure): ... this.
1759 * src/pa/ffi.c: Likewise.
1760 * src/cris/ffi.c: Likewise. Adjust.
1761 * src/frv/ffi.c: Likewise.
1762 * src/ia64/ffi.c: Likewise.
1763 * src/mips/ffi.c: Likewise.
1764 * src/powerpc/ffi_darwin.c: Likewise.
1765 * src/s390/ffi.c: Likewise.
1766 * src/sh/ffi.c: Likewise.
1767 * src/sh64/ffi.c: Likewise.
1768 * src/sparc/ffi.c: Likewise.
1769 * src/x86/ffi64.c: Likewise.
1770 * src/x86/ffi.c: Likewise.
1771 (FFI_INIT_TRAMPOLINE): Adjust.
1772 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1773 (ffi_prep_raw_closure): ... this.
1774 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1775 (ffi_prep_closure): ... this.
1776 (flush_icache): Adjust.
1778 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1780 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1783 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1785 * Makefile.am: Add dummy install-pdf target.
1786 * Makefile.in: Regenerate
1788 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1790 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1791 ffi_closure_helper_SYSV): Add long double handling.
1793 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1795 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1796 immediately after bctrl instruction.
1798 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1800 * Makefile.am (all-recursive, install-recursive,
1801 mostlyclean-recursive, clean-recursive, distclean-recursive,
1802 maintainer-clean-recursive): Add missing targets.
1803 * Makefile.in: Rebuilt.
1805 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1807 * configure.ac: Add TARGET for x86_64-*-darwin*.
1808 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1810 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1811 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1812 * configure: Regenerate.
1813 * Makefile.in: Regenerate.
1814 * include/Makefile.in: Regenerate.
1815 * testsuite/Makefile.in: Regenerate.
1816 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1819 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1821 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1824 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1826 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1827 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1829 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1830 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1832 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1833 Sandro Tolaini <tolaini@libero.it>
1835 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1837 * configure: Regenerated.
1838 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1839 (EXTRA_DIST): Add src/x86/darwin.S.
1840 * Makefile.in: Regenerated.
1841 * include/Makefile.in: Regenerated.
1842 * testsuite/Makefile.in: Regenerated.
1844 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1845 X86_WIN32, and additionally align stack to 16 bytes.
1846 * src/x86/darwin.S: New, based on sysv.S.
1847 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1849 2006-09-12 David Daney <ddaney@avtrex.com>
1852 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1853 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1854 * aclocal.m4: Regenerated for automake 1.9.6.
1855 * Makefile.in: Regenerated.
1856 * include/Makefile.in: Regenerated.
1857 * testsuite/Makefile.in: Regenerated.
1859 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1861 * include/ffi_common.h (struct): Revert accidental commit.
1863 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1865 * include/ffi_common.h: Remove lint directives.
1866 * include/ffi.h.in: Likewise.
1868 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1870 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1871 for 32-bit architectures.
1872 * testsuite/libffi.call/return_ul.c: New test case.
1874 2006-07-19 David Daney <ddaney@avtrex.com>
1876 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1877 xfail remains for mips64.
1879 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1881 * Makefile.am: Add install-html target. Add install-html to .PHONY
1882 * Makefile.in: Regenerate.
1883 * aclocal.m4: Regenerate.
1884 * include/Makefile.in: Regenerate.
1885 * testsuite/Makefile.in: Regenerate.
1887 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1889 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1892 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1894 * README: Remove notice about 'Crazy Comments'.
1895 * src/debug.c: Remove lint directives. Cleanup white spaces.
1896 * src/java_raw_api.c: Likewise.
1897 * src/prep_cif.c: Likewise.
1898 * src/raw_api.c: Likewise.
1899 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1901 * src/arm/ffi.c: Remove lint directives.
1902 * src/m32r/ffi.c: Likewise.
1903 * src/pa/ffi.c: Likewise.
1904 * src/powerpc/ffi.c: Likewise.
1905 * src/powerpc/ffi_darwin.c: Likewise.
1906 * src/sh/ffi.c: Likewise.
1907 * src/sh64/ffi.c: Likewise.
1908 * src/x86/ffi.c: Likewise.
1909 * testsuite/libffi.call/float2.c: Likewise.
1910 * testsuite/libffi.call/promotion.c: Likewise.
1911 * testsuite/libffi.call/struct1.c: Likewise.
1913 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1915 * src/pa/hpux32.S: Correct unwind offset calculation for
1917 * src/pa/linux.S: Likewise.
1919 2006-04-12 James E Wilson <wilson@specifix.com>
1922 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1923 (hfa_type_load): Call stf_spill.
1924 (hfa_type_store): Call ldf_fill.
1925 (ffi_call): Adjust calls to above routines. Add local temps for
1928 2006-04-10 Matthias Klose <doko@debian.org>
1930 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1931 directory names containing underscores.
1933 2006-04-07 James E Wilson <wilson@specifix.com>
1935 * testsuite/libffi.call/float4.c: New testcase.
1937 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1938 Andreas Tobler <a.tobler@schweiz.ch>
1940 * Makefile.am: Add PA_HPUX port.
1941 * Makefile.in: Regenerate.
1942 * include/Makefile.in: Likewise.
1943 * testsuite/Makefile.in: Likewise.
1944 * configure.ac: Add PA_HPUX rules.
1945 * configure: Regenerate.
1946 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1947 Add PA_HPUX and PA64_HPUX.
1948 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1949 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1950 (FFI_TYPE_SMALL_STRUCT2): Define.
1951 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1952 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1953 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1954 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1955 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1956 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1957 * src/pa/ffi.c (ROUND_DOWN): Delete.
1958 (fldw, fstw, fldd, fstd): Use '__asm__'.
1959 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1960 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1961 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1962 Simplify incrementing of stack slot variable. Change type of local
1963 'n' to unsigned int.
1964 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1966 (ffi_prep_cif_machdep): Likewise.
1967 (ffi_call): Likewise.
1968 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1969 return type to ffi_status. Simplify incrementing of stack slot
1970 variable. Only copy floating point argument registers when PA_LINUX
1971 is true. Reformat debug statement.
1972 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1973 FFI_TYPE_SMALL_STRUCT8.
1974 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1976 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1977 Add nops to cache flush. Add trampoline for PA_HPUX.
1978 * src/pa/hpux32.S: New file.
1979 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1980 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1981 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1982 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1983 argument type checks so that common argument types appear first.
1984 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1985 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1987 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1989 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1990 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1991 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1992 FFI_TYPE_LONGDOUBLE.
1993 (ffi_prep_args64): Assert using IBM extended double.
1994 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1995 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1996 (ffi_call): Handle FFI_LINUX.
1997 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1998 gpr3 return pointer as for struct return. Handle FFI_LINUX
1999 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
2001 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
2002 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
2003 Don't use r6 as pointer to results, instead use sp offset. Don't
2004 make a special call to load lr with case table address, instead
2005 use offset from previous call.
2006 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
2007 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
2010 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
2012 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
2013 passed with FP registers correctly.
2014 (ffi_closure_helper_SYSV): Likewise.
2015 * src/sh64/sysv.S: Likewise.
2017 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
2019 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
2020 args and userdata unused.
2021 (closure_test_fn1): Mark cif and userdata unused.
2022 (main): Remove unused res.
2024 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
2026 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
2027 -O2, -O3, -Os and the warning flags -W -Wall.
2028 * testsuite/libffi.special/special.exp: Likewise.
2029 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
2030 unused parameter unused for gcc or else do nothing.
2031 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2032 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
2033 and userdata unused.
2034 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
2035 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
2036 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
2037 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
2038 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
2039 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
2040 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
2041 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
2042 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
2043 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
2044 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
2045 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
2046 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
2047 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
2048 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
2049 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
2050 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
2051 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
2052 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
2053 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
2054 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
2055 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
2056 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
2057 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
2059 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
2061 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
2063 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
2064 void* to avoid compiler warning.
2066 (cls_struct_align_gn): Mark cif and userdata unused.
2067 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
2069 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
2071 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
2073 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
2075 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
2077 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
2078 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
2079 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
2081 (main): Cast res_call to silence gcc.
2082 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
2084 (main): Cast res_call to silence gcc.
2085 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
2087 (main): Cast res_call to silence gcc.
2088 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
2090 (main): Cast res_call to silence gcc.
2091 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
2093 (main): Cast res_call to silence gcc.
2094 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
2096 (main): Cast res_call to silence gcc.
2097 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
2099 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
2100 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
2102 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
2103 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
2105 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
2106 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
2108 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
2109 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
2111 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
2112 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
2113 and userdata unused.
2114 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
2116 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
2117 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
2118 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
2119 Cleanup white spaces.
2120 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
2121 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
2122 cif and userdata unused.
2123 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
2125 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
2126 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
2127 formatters to silence gcc.
2128 (B_gn): Mark cif and userdata unused.
2129 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
2131 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
2132 (B_gn): Mark cif and userdata unused.
2133 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
2135 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
2136 (B_gn): Mark cif and userdata unused.
2137 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
2139 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
2140 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
2141 * testsuite/libffi.call/problem1.c (stub): Likewise.
2142 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
2144 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
2145 in the last commit for this test case in the test case itself.
2146 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
2148 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
2149 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
2150 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
2151 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
2152 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
2153 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
2155 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
2157 * src/sh/sysv.S: Fix register numbers in the FDE for
2160 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
2162 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
2163 declaration to avoid a false negative on ix86. See PR323.
2165 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
2167 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
2168 and cast integer to void * if needed. Update the pointer to
2169 the FP register saved area correctly.
2171 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
2173 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
2175 * testsuite/libffi.call/nested_struct4.c: Likewise.
2177 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
2179 * testsuite/libffi.call/return_dbl.c: New test case.
2180 * testsuite/libffi.call/return_dbl1.c: Likewise.
2181 * testsuite/libffi.call/return_dbl2.c: Likewise.
2182 * testsuite/libffi.call/return_fl.c: Likewise.
2183 * testsuite/libffi.call/return_fl1.c: Likewise.
2184 * testsuite/libffi.call/return_fl2.c: Likewise.
2185 * testsuite/libffi.call/return_fl3.c: Likewise.
2186 * testsuite/libffi.call/closure_fn6.c: Likewise.
2188 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
2190 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
2191 here to be used by other test cases too.
2193 * testsuite/libffi.call/nested_struct10.c: New test case.
2194 * testsuite/libffi.call/nested_struct9.c: Likewise.
2195 * testsuite/libffi.call/nested_struct8.c: Likewise.
2196 * testsuite/libffi.call/nested_struct7.c: Likewise.
2197 * testsuite/libffi.call/nested_struct6.c: Likewise.
2198 * testsuite/libffi.call/nested_struct5.c: Likewise.
2199 * testsuite/libffi.call/nested_struct4.c: Likewise.
2201 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2203 * configure.ac: Enable libffi for sparc64-*-freebsd*.
2204 * configure: Rebuilt.
2206 2006-01-18 Jakub Jelinek <jakub@redhat.com>
2208 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
2209 instead do the shifting inline.
2210 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
2211 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
2212 and 8 byte structs, for the remaining struct sizes don't call
2213 __lshrdi3, instead do the shifting inline.
2215 2005-12-07 Thiemo Seufer <ths@networkno.de>
2217 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
2218 missing parentheses.
2219 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
2220 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
2221 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
2222 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
2223 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
2225 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
2227 (ffi_prep_closure): Improve trampoline instruction scheduling.
2228 (ffi_closure_mips_inner_O32): Fix endianness bugs.
2230 2005-12-03 Alan Modra <amodra@bigpond.net.au>
2232 * src/powerpc/ffi.c: Formatting.
2233 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
2234 (ffi_prep_args64): Likewise.
2236 2005-09-30 Geoffrey Keating <geoffk@apple.com>
2238 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
2239 darwin, use -shared-libgcc not -lgcc_s, and explain why.
2241 2005-09-26 Tom Tromey <tromey@redhat.com>
2243 * testsuite/libffi.call/float1.c (value_type): New typedef.
2244 (CANARY): New define.
2245 (main): Check for result buffer overflow.
2246 * src/powerpc/linux64.S: Handle linux64 long double returns.
2247 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
2248 (ffi_prep_cif_machdep): Handle linux64 long double returns.
2250 2005-08-25 Alan Modra <amodra@bigpond.net.au>
2253 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
2255 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
2257 2005-08-11 Jakub Jelinek <jakub@redhat.com>
2259 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
2260 (AH_BOTTOM): Add FFI_HIDDEN definition.
2261 * configure: Rebuilt.
2262 * fficonfig.h.in: Rebuilt.
2263 * src/powerpc/ffi.c (hidden): Remove.
2264 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
2265 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
2266 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
2267 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
2268 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
2269 add FFI_HIDDEN to its prototype.
2270 (ffi_closure_SYSV_inner): New.
2271 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2272 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2274 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
2277 * configure: Rebuilt.
2278 * configure.ac: Handle i*86-*-gnu*.
2280 2005-08-09 Jakub Jelinek <jakub@redhat.com>
2282 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
2283 DW_CFA_offset_extended_sf rather than
2284 DW_CFA_GNU_negative_offset_extended.
2285 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
2287 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
2289 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
2291 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
2292 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
2293 partially on register.
2294 (ffi_closure_helper_SYSV): Likewise.
2295 (ffi_prep_cif_machdep): Don't set too many cif->flags.
2297 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
2299 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
2301 * src/sh64/ffi.c (simple_type): Remove.
2302 (return_type): Handle small structures correctly.
2303 (ffi_prep_args): Likewise.
2304 (ffi_call): Likewise.
2305 (ffi_closure_helper_SYSV): Likewise.
2306 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
2307 Emit position independent code if PIC and remove wrong datalabel
2308 prefixes from EH data.
2310 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
2312 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
2313 * Makefile.in: Regenerate.
2314 * include/Makefile.in: Likewise.
2315 * testsuite/Makefile.in: Likewise.
2316 * configure.ac: Add POWERPC_FREEBSD rules.
2317 * configure: Regenerate.
2318 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
2319 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
2320 * src/powerpc/ffi.c: Add flags to handle small structure returns
2322 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
2324 (ffi_closure_helper_SYSV): Likewise.
2325 * src/powerpc/ppc_closure.S: Add return types for small structures.
2326 * src/powerpc/sysv.S: Add bits to handle small structures for
2329 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
2331 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
2332 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
2333 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
2335 2005-07-05 Randolph Chung <tausq@debian.org>
2337 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
2338 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
2339 structures. Kill compilation warnings.
2340 (ffi_closure_inner_LINUX): Print return values as hex in debug
2341 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
2342 Properly handle 5-7 byte structure returns.
2343 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
2344 (FFI_TYPE_SMALL_STRUCT2): Remove.
2345 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
2346 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
2347 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
2348 (checksmst1, checksmst2): Remove.
2349 (checksmst3): Optimize handling of 3-byte struct returns.
2350 (checksmst567): Properly handle 5-7 byte struct returns.
2352 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2355 * src/mips/n32.S: Enforce PIC code.
2356 * src/mips/o32.S: Likewise.
2358 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2360 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
2361 * configure: Regenerate.
2363 2005-06-01 Alan Modra <amodra@bigpond.net.au>
2365 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
2366 to call ffi_closure_helper_SYSV. Append @local instead.
2367 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
2369 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
2371 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
2372 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
2373 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
2374 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
2375 include/Makefile.in, testsuite/Makefile.in: Regenerate.
2377 2005-05-09 Mike Stump <mrs@apple.com>
2379 * configure: Regenerate.
2381 2005-05-08 Richard Henderson <rth@redhat.com>
2384 * src/alpha/osf.S: Update unwind into to match code.
2386 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
2387 Richard Henderson <rth@redhat.com>
2389 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
2391 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
2392 (ffi_prep_closure): Set carry bit if sse-used flag set.
2393 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
2394 Only load sse registers if ssecount non-zero.
2395 (ffi_closure_unix64): Only save sse registers if carry set on entry.
2397 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
2399 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
2400 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
2401 * configure: Regenerate.
2403 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
2405 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
2406 have Tcl8.3-compatible intermediate variable.
2408 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
2409 Hans-Peter Nilsson <hp@axis.com>
2411 * Makefile.am: Add CRIS support.
2412 * configure.ac: Likewise.
2413 * Makefile.in, configure, testsuite/Makefile.in,
2414 include/Makefile.in: Regenerate.
2415 * src/cris: New directory.
2416 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
2417 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
2419 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
2420 \r?\n in output tests.
2422 2005-04-12 Mike Stump <mrs@apple.com>
2424 * configure: Regenerate.
2426 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
2428 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
2430 2005-03-30 Steve Ellcey <sje@cup.hp.com>
2432 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
2434 * src/ia64/unix.S (ffi_closure_unix): Extend gp
2435 to 64 bits in ILP32 mode.
2436 Load 64 bits even for short data.
2438 2005-03-23 Mike Stump <mrs@apple.com>
2440 * src/powerpc/darwin.S: Update for -m64 multilib.
2441 * src/powerpc/darwin_closure.S: Likewise.
2443 2005-03-21 Zack Weinberg <zack@codesourcery.com>
2445 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
2446 Do not set tool_include_dir.
2447 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
2449 * include/Makefile.am: Set gcc_version and toollibffidir.
2450 * include/Makefile.in: Regenerate.
2452 2005-02-22 Andrew Haley <aph@redhat.com>
2454 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2455 odd-numbered register pairs for 64-bit integer types.
2457 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2460 * testsuite/libffi.call/return_ll1.c: New test case.
2462 2005-02-11 Janis Johnson <janis187@us.ibm.com>
2464 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2465 * testsuite/libffi.call/float.c: Ditto.
2466 * testsuite/libffi.call/float2.c: Ditto.
2467 * testsuite/libffi.call/float3.c: Ditto.
2469 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2471 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2473 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2475 * testsuite/libffi.special/special.exp (cxx_options): Add
2478 2004-12-31 Richard Henderson <rth@redhat.com>
2480 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2481 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2482 offset parameters with a type parameter; deduce size and structure
2483 alignment. Update all users.
2485 2004-12-31 Richard Henderson <rth@redhat.com>
2487 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2488 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
2489 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
2490 into ffi_prep_closure.
2491 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2494 2004-12-27 Richard Henderson <rth@redhat.com>
2496 * src/x86/unix64.S: Fix typo in unwind info.
2498 2004-12-25 Richard Henderson <rth@redhat.com>
2500 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2501 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2502 (merge_classes): Check for it.
2504 (classify_argument): Pass byte_offset by value; perform all updates
2506 (examine_argument): Add classes argument; handle
2507 X86_64_COMPLEX_X87_CLASS.
2508 (ffi_prep_args): Merge into ...
2509 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2510 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2511 (ffi_fill_return_value): Remove.
2512 (ffi_prep_closure): Remove dead assert.
2513 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2514 Rewrite to use struct register_args instead of va_list. Create
2515 flags for handling structure returns.
2516 * src/x86/unix64.S: Remove dead strings.
2517 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2518 stack frame with ffi_call. Handle structure returns properly.
2519 (float2sse, floatfloat2sse, double2sse): Remove.
2520 (sse2float, sse2double, sse2floatfloat): Remove.
2521 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2522 to handle structure returns properly.
2524 2004-12-08 David Edelsohn <edelsohn@gnu.org>
2526 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2528 * Makefile.in: Regenerated.
2530 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2532 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2533 * configure, aclocal.m4, Makefile.in: Regenerate.
2534 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2536 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2538 * configure: Regenerate for libtool change.
2540 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2542 * configure: Regenerate for libtool reversion.
2544 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2546 * configure: Regenerate for libtool change.
2548 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2550 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2552 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2554 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2555 of jal. Use an absolute encoding for the frame information.
2557 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2559 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2560 * acinclude.m4: Delete logic for sincludes.
2561 * aclocal.m4, Makefile.in, configure: Regenerate.
2562 * include/Makefile: Likewise.
2563 * testsuite/Makefile: Likewise.
2565 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2567 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2568 on a 8-byte boundary.
2569 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2571 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2573 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2574 long long values. Round stack allocation to a multiple of 8 bytes
2575 for ATPCS compatibility.
2576 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2577 names. Handle returning long long types. Add Thumb and interworking
2578 support. Improve soft-float code.
2580 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2582 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2583 (libffi_exit): New function.
2584 (libffi_init): Build the testglue wrapper if needed.
2586 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2589 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2591 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2593 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2595 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2597 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2598 * testsuite/libffi.call/float3.c: New test case.
2600 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2602 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2603 the function returning a structure pointed with R2.
2604 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2605 the structure return value if T bit set. Emit position
2606 independent code and EH data if PIC.
2608 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2610 * Makefile.am: Add m32r support.
2611 * configure.ac: Likewise.
2612 * Makefile.in: Regenerate.
2613 * confiugre: Regenerate.
2614 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2615 (uint64, sint64, double, longdouble)
2616 * src/m32r: New directory.
2617 * src/m32r/ffi.c: New file.
2618 * src/m32r/sysv.S: Likewise.
2619 * src/m32r/ffitarget.h: Likewise.
2621 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2623 * testsuite/libffi.call/negint.c: New test case.
2625 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2628 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2629 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2630 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2633 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2635 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2636 * testsuite/libffi.call/promotion.c: Likewise.
2637 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2639 * testsuite/libffi.call/return_sc.c: Likewise.
2640 * testsuite/libffi.call/return_uc.c: Likewise.
2642 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2644 * src/powerpc/darwin.S: Fix comments and identation.
2645 * src/powerpc/darwin_closure.S: Likewise.
2647 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2649 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2650 (ffi_prep_args): Handle longdouble arguments.
2651 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2653 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2654 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2656 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2657 * src/types.c: Defined longdouble size and alignment for darwin.
2659 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2661 * src/powerpc/aix.S: Remove whitespaces.
2662 * src/powerpc/aix_closure.S: Likewise.
2663 * src/powerpc/asm.h: Likewise.
2664 * src/powerpc/ffi.c: Likewise.
2665 * src/powerpc/ffitarget.h: Likewise.
2666 * src/powerpc/linux64.S: Likewise.
2667 * src/powerpc/linux64_closure.S: Likewise.
2668 * src/powerpc/ppc_closure.S: Likewise.
2669 * src/powerpc/sysv.S: Likewise.
2671 2004-08-30 Anthony Green <green@redhat.com>
2673 * Makefile.am: Add frv support.
2674 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2675 * configure.ac: Read configure.host.
2676 * configure.in: Read configure.host.
2677 * configure.host: New file. frv-elf needs libgloss.
2678 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2679 alignment. This is needed to frv and shouldn't harm the others.
2680 * include/ffi_common.h (ALIGN_DOWN): New macro.
2681 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2683 2004-08-24 David Daney <daney@avtrex.com>
2685 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2686 * testsuite/libffi.call/closure_fn1.c: Likewise.
2687 * testsuite/libffi.call/closure_fn2.c Likewise.
2688 * testsuite/libffi.call/closure_fn3.c: Likewise.
2689 * testsuite/libffi.call/closure_fn4.c: Likewise.
2690 * testsuite/libffi.call/closure_fn5.c: Likewise.
2691 * testsuite/libffi.call/cls_18byte.c: Likewise.
2692 * testsuite/libffi.call/cls_19byte.c: Likewise.
2693 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2694 * testsuite/libffi.call/cls_20byte.c: Likewise.
2695 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2696 * testsuite/libffi.call/cls_24byte.c: Likewise.
2697 * testsuite/libffi.call/cls_2byte.c: Likewise.
2698 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2699 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2700 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2701 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2702 * testsuite/libffi.call/cls_4byte.c: Likewise.
2703 * testsuite/libffi.call/cls_64byte.c: Likewise.
2704 * testsuite/libffi.call/cls_6byte.c: Likewise.
2705 * testsuite/libffi.call/cls_7byte.c: Likewise.
2706 * testsuite/libffi.call/cls_8byte.c: Likewise.
2707 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2708 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2709 * testsuite/libffi.call/cls_align_double.c: Likewise.
2710 * testsuite/libffi.call/cls_align_float.c: Likewise.
2711 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2712 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2713 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2714 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2715 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2716 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2717 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2718 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2719 * testsuite/libffi.call/cls_double.c: Likewise.
2720 * testsuite/libffi.call/cls_float.c: Likewise.
2721 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2722 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2723 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2724 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2725 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2726 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2727 * testsuite/libffi.call/cls_schar.c: Likewise.
2728 * testsuite/libffi.call/cls_sint.c: Likewise.
2729 * testsuite/libffi.call/cls_sshort.c: Likewise.
2730 * testsuite/libffi.call/cls_uchar.c: Likewise.
2731 * testsuite/libffi.call/cls_uint.c: Likewise.
2732 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2733 * testsuite/libffi.call/cls_ushort.c: Likewise.
2734 * testsuite/libffi.call/nested_struct.c: Likewise.
2735 * testsuite/libffi.call/nested_struct1.c: Likewise.
2736 * testsuite/libffi.call/nested_struct2.c: Likewise.
2737 * testsuite/libffi.call/nested_struct3.c: Likewise.
2738 * testsuite/libffi.call/problem1.c: Likewise.
2739 * testsuite/libffi.special/unwindtest.cc: Likewise.
2740 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2742 * testsuite/libffi.call/cls_16byte.c: Likewise.
2743 * testsuite/libffi.call/cls_5byte.c: Likewise.
2745 2004-08-23 David Daney <daney@avtrex.com>
2748 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2749 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2750 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2751 parameters and return types.
2752 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2753 (ffi_prep_closure): Ditto.
2754 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2755 alignment calculations.
2756 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2757 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2759 2004-08-14 Casey Marshall <csm@gnu.org>
2761 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2762 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2763 integer (O32 ABI only).
2764 (ffi_prep_closure): new function.
2765 (ffi_closure_mips_inner_O32): new function.
2766 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2767 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2768 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2769 64 bit integers correctly.
2770 (ffi_closure_O32): new function.
2771 Added DWARF-2 unwind info for both functions.
2773 2004-08-10 Andrew Haley <aph@redhat.com>
2775 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2777 2004-08-01 Robert Millan <robertmh@gnu.org>
2779 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2780 * configure: Regenerate.
2782 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2784 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2785 and mmap() explicitly instead of relying on preset autoconf cache
2787 * aclocal.m4: Regenerate.
2788 * configure: Regenerate.
2790 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2792 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2793 (ffi_check_float_struct): Remove unused prototype.
2795 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2797 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2798 character on Darwin, use '\n\t' instead.
2800 2004-06-26 Matthias Klose <doko@debian.org>
2802 * libtool-version: Fix typo in revision/age.
2804 2004-06-17 Matthias Klose <doko@debian.org>
2806 * libtool-version: New.
2807 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2808 * Makefile.in: Regenerate.
2810 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2812 * Makefile.am: Remove useless multilib rules.
2813 * Makefile.in: Regenerate.
2814 * aclocal.m4: Regenerate with automake 1.8.5.
2815 * configure.ac: Remove useless multilib configury.
2816 * configure: Regenerate.
2818 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2820 * .cvsignore: New file.
2822 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2824 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2826 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2829 2004-06-06 Sean McNeil <sean@mcneil.com>
2831 * configure.ac: Add x86_64-*-freebsd* support.
2832 * configure: Regenerate.
2834 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2837 * configure.ac: Test for existence of mmap and sys/mman.h before
2838 checking blacklist. Fix suggested by Jim Wilson.
2839 * configure: Regenerate.
2841 2004-04-26 Matt Austern <austern@apple.com>
2843 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2845 * src/powerpc/darwin_closure.S: Likewise.
2847 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2849 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2850 error. Reported by Thomas Heller <theller@python.net>.
2851 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2852 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2854 2004-03-20 Matthias Klose <doko@debian.org>
2856 * src/pa/linux.S: Fix typo.
2858 2004-03-19 Matthias Klose <doko@debian.org>
2860 * Makefile.am: Update.
2861 * Makefile.in: Regenerate.
2862 * src/pa/ffi.h.in: Remove.
2863 * src/pa/ffitarget.h: New file.
2865 2004-02-10 Randolph Chung <tausq@debian.org>
2867 * Makefile.am: Add PA support.
2868 * Makefile.in: Regenerate.
2869 * include/Makefile.in: Regenerate.
2870 * configure.ac: Add PA target.
2871 * configure: Regenerate.
2872 * src/pa/ffi.c: New file.
2873 * src/pa/ffi.h.in: Add PA support.
2874 * src/pa/linux.S: New file.
2875 * prep_cif.c: Add PA support.
2877 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2879 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2881 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2882 with ecif->cif->flags.
2883 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2885 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2886 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2887 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2888 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2890 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2892 * configure.in: Rename file to ...
2893 * configure.ac: ... this.
2894 * fficonfig.h.in: Regenerate.
2895 * Makefile.in: Regenerate.
2896 * include/Makefile.in: Regenerate.
2897 * testsuite/Makefile.in: Regenerate.
2899 2004-03-12 Matt Austern <austern@apple.com>
2901 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2902 changes in EH format resulting from addition of linkonce support.
2903 * src/powerpc/darwin_closure.S: Likewise.
2905 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2906 Paolo Bonzini <bonzini@gnu.org>
2908 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2909 Remove VPATH. Remove rules for object files. Remove multilib support.
2910 (AM_CCASFLAGS): Add.
2911 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2912 (AC_PREREQ): Bump version to 2.59.
2913 (AC_INIT): Fill with version info and bug address.
2914 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2915 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2916 De-precious CC so that the right flags are passed down to multilibs.
2917 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2918 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2919 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2920 * configure: Rebuilt.
2921 * aclocal.m4: Likewise.
2922 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2923 * fficonfig.h.in: Likewise.
2925 2004-03-11 Andreas Schwab <schwab@suse.de>
2927 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2928 arguments from fp registers only for the first 8 parameter slots.
2929 Don't convert a float parameter when passed in memory.
2931 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2933 * configure: Regenerate for config/accross.m4 correction.
2935 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2937 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2938 ecif->cif->bytes to bytes.
2939 (ffi_prep_cif_machdep): Add braces around nested if statement.
2941 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2943 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2945 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2946 (ffi_closure_helper_LINUX64): Fix typo.
2947 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2949 * testsuite/libffi.call/float.c: Likewise.
2950 * testsuite/libffi.call/float2.c: Likewise.
2952 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2954 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2955 long double function return and long double arg handling.
2956 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2957 Use "end_pfr" instead of "nf". Correct long double handling.
2959 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2961 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2962 space for long double return value. Adjust stack frame and offsets.
2963 Load f2 long double return.
2965 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2967 * src/types.c: Use 16 byte long double for POWERPC64.
2969 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2971 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2972 when the structure return address is passed in %o0.
2973 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2974 (ffi_v9_layout_struct): Align the field following a nested structure
2975 on a word boundary. Use memmove instead of memcpy.
2976 (ffi_call): Update call to ffi_V9_return_struct.
2977 (ffi_prep_closure): Define 'ctx' only for V8.
2978 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2979 and ffi_closure_sparc_inner_v9.
2980 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2981 Always skip the structure return address. For structures and long
2982 doubles, copy the argument directly.
2983 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2984 if required. Shift the maximum floating-point slot accordingly. For
2985 big structures, copy the argument directly; otherwise, left-justify the
2986 argument and call ffi_v9_layout_struct to lay out the structure on
2988 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2989 (ffi_closure_v8): Pass the structure return address. Update call to
2990 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2991 Skip the 'unimp' insn when returning long doubles and structures.
2992 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2993 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2994 FFI_TYPE_INT handling. Load structures both in integers and
2995 floating-point registers on return.
2996 * README: Update status of the SPARC port.
2998 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
3000 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
3002 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
3004 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
3006 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
3007 value as of type ffi_arg, not unsigned int.
3009 2004-01-21 Michael Ritzert <ritzert@t-online.de>
3011 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
3014 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
3016 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
3019 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3021 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
3024 2003-12-10 Richard Henderson <rth@redhat.com>
3026 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
3027 size_t instead of int.
3029 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
3031 * testsuite/libffi.call/many_win32.c: Include <float.h>.
3032 * testsuite/libffi.call/many_win32.c (main): Replace variable
3033 int i with unsigned long ul.
3035 * testsuite/libffi.call/cls_align_uint64.c: New test case.
3036 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
3037 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
3038 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
3039 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
3040 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
3041 * testsuite/libffi.call/cls_align_float.c: Likewise.
3042 * testsuite/libffi.call/cls_align_double.c: Likewise.
3043 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
3044 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
3046 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
3049 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
3050 Align arguments to 32 bits.
3052 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
3055 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
3056 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
3057 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
3058 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
3059 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
3060 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
3062 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
3064 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3066 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
3067 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3069 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
3071 * Makefile.in: Rebuilt.
3072 * configure: Likewise.
3073 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
3076 2003-11-21 Andreas Jaeger <aj@suse.de>
3077 Andreas Tobler <a.tobler@schweiz.ch>
3079 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
3080 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
3081 * Makefile.in: Rebuilt.
3082 * aclocal.m4: Likewise.
3083 * configure: Likewise.
3084 * fficonfig.h.in: Likewise.
3085 * testsuite/lib/libffi-dg.exp: Add include dir.
3086 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
3087 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3088 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
3089 for ffi_closure if available.
3090 * testsuite/libffi.call/closure_fn1.c: Likewise.
3091 * testsuite/libffi.call/closure_fn2.c: Likewise.
3092 * testsuite/libffi.call/closure_fn3.c: Likewise.
3093 * testsuite/libffi.call/closure_fn4.c: Likewise.
3094 * testsuite/libffi.call/closure_fn5.c: Likewise.
3095 * testsuite/libffi.call/cls_12byte.c: Likewise.
3096 * testsuite/libffi.call/cls_16byte.c: Likewise.
3097 * testsuite/libffi.call/cls_18byte.c: Likewise.
3098 * testsuite/libffi.call/cls_19byte.c: Likewise.
3099 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3100 * testsuite/libffi.call/cls_20byte.c: Likewise.
3101 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3102 * testsuite/libffi.call/cls_24byte.c: Likewise.
3103 * testsuite/libffi.call/cls_2byte.c: Likewise.
3104 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3105 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3106 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3107 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3108 * testsuite/libffi.call/cls_4byte.c: Likewise.
3109 * testsuite/libffi.call/cls_5byte.c: Likewise.
3110 * testsuite/libffi.call/cls_64byte.c: Likewise.
3111 * testsuite/libffi.call/cls_6byte.c: Likewise.
3112 * testsuite/libffi.call/cls_7byte.c: Likewise.
3113 * testsuite/libffi.call/cls_8byte.c: Likewise.
3114 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3115 * testsuite/libffi.call/cls_9byte2.c: Likewise.
3116 * testsuite/libffi.call/cls_double.c: Likewise.
3117 * testsuite/libffi.call/cls_float.c: Likewise.
3118 * testsuite/libffi.call/cls_schar.c: Likewise.
3119 * testsuite/libffi.call/cls_sint.c: Likewise.
3120 * testsuite/libffi.call/cls_sshort.c: Likewise.
3121 * testsuite/libffi.call/cls_uchar.c: Likewise.
3122 * testsuite/libffi.call/cls_uint.c: Likewise.
3123 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3124 * testsuite/libffi.call/cls_ushort.c: Likewise.
3125 * testsuite/libffi.call/nested_struct.c: Likewise.
3126 * testsuite/libffi.call/nested_struct1.c: Likewise.
3127 * testsuite/libffi.call/nested_struct2.c: Likewise.
3128 * testsuite/libffi.call/nested_struct3.c: Likewise.
3129 * testsuite/libffi.call/problem1.c: Likewise.
3130 * testsuite/libffi.special/unwindtest.cc: Likewise.
3132 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
3134 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
3136 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
3138 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
3139 Add -lgcc_s to additional flags.
3141 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
3143 * configure.in, include/Makefile.am: PR libgcj/11147, install
3144 the ffitarget.h header file in a gcc versioned and target
3146 * configure: Regenerated.
3147 * Makefile.in, include/Makefile.in: Likewise.
3148 * testsuite/Makefile.in: Likewise.
3150 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
3152 * testsuite/libffi.call/closure_fn0.c: Print result and check
3153 with dg-output to make debugging easier.
3154 * testsuite/libffi.call/closure_fn1.c: Likewise.
3155 * testsuite/libffi.call/closure_fn2.c: Likewise.
3156 * testsuite/libffi.call/closure_fn3.c: Likewise.
3157 * testsuite/libffi.call/closure_fn4.c: Likewise.
3158 * testsuite/libffi.call/closure_fn5.c: Likewise.
3159 * testsuite/libffi.call/cls_12byte.c: Likewise.
3160 * testsuite/libffi.call/cls_16byte.c: Likewise.
3161 * testsuite/libffi.call/cls_18byte.c: Likewise.
3162 * testsuite/libffi.call/cls_19byte.c: Likewise.
3163 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3164 * testsuite/libffi.call/cls_20byte.c: Likewise.
3165 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3166 * testsuite/libffi.call/cls_24byte.c: Likewise.
3167 * testsuite/libffi.call/cls_2byte.c: Likewise.
3168 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3169 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3170 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3171 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3172 * testsuite/libffi.call/cls_4byte.c: Likewise.
3173 * testsuite/libffi.call/cls_5byte.c: Likewise.
3174 * testsuite/libffi.call/cls_64byte.c: Likewise.
3175 * testsuite/libffi.call/cls_6byte.c: Likewise.
3176 * testsuite/libffi.call/cls_7byte.c: Likewise.
3177 * testsuite/libffi.call/cls_8byte.c: Likewise.
3178 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3179 * testsuite/libffi.call/cls_9byte2.c: Likewise.
3180 * testsuite/libffi.call/cls_double.c: Likewise.
3181 * testsuite/libffi.call/cls_float.c: Likewise.
3182 * testsuite/libffi.call/cls_schar.c: Likewise.
3183 * testsuite/libffi.call/cls_sint.c: Likewise.
3184 * testsuite/libffi.call/cls_sshort.c: Likewise.
3185 * testsuite/libffi.call/cls_uchar.c: Likewise.
3186 * testsuite/libffi.call/cls_uint.c: Likewise.
3187 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3188 * testsuite/libffi.call/cls_ushort.c: Likewise.
3189 * testsuite/libffi.call/problem1.c: Likewise.
3191 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
3194 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
3196 * testsuite/libffi.call/cls_9byte2.c: New test case.
3197 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3198 * testsuite/libffi.call/cls_64byte.c: Likewise.
3199 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3200 * testsuite/libffi.call/cls_19byte.c: Likewise.
3201 * testsuite/libffi.call/cls_18byte.c: Likewise.
3202 * testsuite/libffi.call/closure_fn4.c: Likewise.
3203 * testsuite/libffi.call/closure_fn5.c: Likewise.
3204 * testsuite/libffi.call/cls_schar.c: Likewise.
3205 * testsuite/libffi.call/cls_sint.c: Likewise.
3206 * testsuite/libffi.call/cls_sshort.c: Likewise.
3207 * testsuite/libffi.call/nested_struct2.c: Likewise.
3208 * testsuite/libffi.call/nested_struct3.c: Likewise.
3210 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
3212 * testsuite/libffi.call/cls_double.c: Do a check on the result.
3213 * testsuite/libffi.call/cls_uchar.c: Likewise.
3214 * testsuite/libffi.call/cls_uint.c: Likewise.
3215 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3216 * testsuite/libffi.call/cls_ushort.c: Likewise.
3217 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
3219 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
3221 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
3224 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
3226 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
3227 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
3229 2003-10-22 David Daney <ddaney@avtrex.com>
3231 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
3232 __attribute__((__mode__(__SI__))) and friends.
3234 2003-10-22 Andreas Schwab <schwab@suse.de>
3236 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
3238 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
3240 * configure.in: AC_LINK_FILES(ffitarget.h).
3241 * configure: Regenerate.
3242 * Makefile.in: Likewise.
3243 * include/Makefile.in: Likewise.
3244 * testsuite/Makefile.in: Likewise.
3245 * fficonfig.h.in: Likewise.
3247 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
3248 Richard Henderson <rth@redhat.com>
3250 Avoid that ffi.h includes fficonfig.h.
3252 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
3253 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
3254 (TARGET_SRC_MIPS_SGI): Removed.
3255 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
3256 (MIPS_SGI): Removed.
3257 (CLEANFILES): Removed.
3258 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
3260 * acconfig.h: Removed.
3261 * configure.in: Compute sizeofs only for double and long double.
3262 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
3263 into AC_DEFINE instead of using acconfig.h. Create
3264 include/ffitarget.h instead of include/fficonfig.h. Rename
3265 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
3266 AC_DEFINE EH_FRAME_FLAGS.
3267 * include/Makefile.am (DISTCLEANFILES): New automake macro.
3268 (hack_DATA): Add ffitarget.h.
3269 * include/ffi.h.in: Remove all system specific definitions.
3270 Declare raw API even if it is not installed, why bother?
3271 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
3272 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
3273 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
3274 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
3275 * include/ffi_common.h (bool): Do not define.
3276 (ffi_assert): Accept failed assertion.
3277 (ffi_type_test): Return void and accept file/line.
3278 (FFI_ASSERT): Pass stringized failed assertion.
3279 (FFI_ASSERT_AT): New macro.
3280 (FFI_ASSERT_VALID_TYPE): New macro.
3281 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
3282 UINT64, SINT64): Define here with gcc's __attribute__ macro
3284 (FLOAT32, ALIGN): Define here instead of in ffi.h
3285 * include/ffi-mips.h: Removed. Its content moved to
3286 src/mips/ffitarget.h after separating assembly and C sections.
3287 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
3288 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
3289 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
3290 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
3291 SIZEOF_ARG -> FFI_SIZEOF_ARG.
3292 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
3293 * src/debug.c (ffi_assert): Accept stringized failed assertion.
3294 (ffi_type_test): Rewritten.
3295 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
3296 FFI_ASSERT_VALID_TYPE.
3297 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
3298 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
3299 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
3300 src/s390/ffitarget.h, src/sh/ffitarget.h,
3301 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
3302 src/x86/ffitarget.h: New files.
3303 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
3304 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
3305 src/powerpc/aix.S, src/powerpc/darwin.S,
3306 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
3307 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
3308 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
3309 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
3310 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
3313 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3315 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
3316 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
3318 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
3320 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
3321 Used when FFI_DEBUG = 1.
3323 2003-10-14 Alan Modra <amodra@bigpond.net.au>
3325 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
3328 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3330 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
3331 FFI_MIPS_O32 for O32 ABI.
3333 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
3335 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
3336 SPARC64. Cleanup whitespaces.
3338 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
3340 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
3341 strongarm, xscale. Cleanup whitespaces.
3342 * testsuite/libffi.call/closure_fn1.c: Likewise.
3343 * testsuite/libffi.call/closure_fn2.c: Likewise.
3344 * testsuite/libffi.call/closure_fn3.c: Likewise.
3345 * testsuite/libffi.call/cls_12byte.c: Likewise.
3346 * testsuite/libffi.call/cls_16byte.c: Likewise.
3347 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3348 * testsuite/libffi.call/cls_20byte.c: Likewise.
3349 * testsuite/libffi.call/cls_24byte.c: Likewise.
3350 * testsuite/libffi.call/cls_2byte.c: Likewise.
3351 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3352 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3353 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3354 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3355 * testsuite/libffi.call/cls_4byte.c: Likewise.
3356 * testsuite/libffi.call/cls_5byte.c: Likewise.
3357 * testsuite/libffi.call/cls_6byte.c: Likewise.
3358 * testsuite/libffi.call/cls_7byte.c: Likewise.
3359 * testsuite/libffi.call/cls_8byte.c: Likewise.
3360 * testsuite/libffi.call/cls_double.c: Likewise.
3361 * testsuite/libffi.call/cls_float.c: Likewise.
3362 * testsuite/libffi.call/cls_uchar.c: Likewise.
3363 * testsuite/libffi.call/cls_uint.c: Likewise.
3364 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3365 * testsuite/libffi.call/cls_ushort.c: Likewise.
3366 * testsuite/libffi.call/nested_struct.c: Likewise.
3367 * testsuite/libffi.call/nested_struct1.c: Likewise.
3368 * testsuite/libffi.call/problem1.c: Likewise.
3369 * testsuite/libffi.special/unwindtest.cc: Likewise.
3370 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
3372 2003-09-18 David Edelsohn <edelsohn@gnu.org>
3374 * src/powerpc/aix.S: Cleanup whitespaces.
3375 * src/powerpc/aix_closure.S: Likewise.
3377 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3379 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
3380 * src/powerpc/darwin_closure.S: Likewise.
3381 * src/powerpc/ffi_darwin.c: Likewise.
3383 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3384 David Edelsohn <edelsohn@gnu.org>
3386 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
3387 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
3389 * src/powerpc/darwin_closure.S: Likewise.
3390 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
3391 according to the Darwin/AIX ABI.
3392 (ffi_prep_cif_machdep): Likewise.
3393 (ffi_closure_helper_DARWIN): Likewise.
3394 Remove the outgoing parameter stack logic. Simplify the evaluation
3395 of the different CASE types.
3396 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
3397 statement in the trampoline code.
3399 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
3401 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
3402 for the register size.
3403 (ffi_closure_helper_SYSV): Handle the structure return value
3405 (ffi_closure_helper_SYSV): Return the appropriate type when
3406 the registers are used for the structure return value.
3407 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
3408 the 64-bit return value. Update copyright years.
3410 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3412 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
3413 srcdir for ffi_mips.h.
3415 2003-09-12 Alan Modra <amodra@bigpond.net.au>
3417 * src/prep_cif.c (initialize_aggregate): Include tail padding in
3419 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
3420 placement of float result.
3421 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
3422 cast of "resp" for big-endian 64 bit machines.
3424 2003-09-11 Alan Modra <amodra@bigpond.net.au>
3426 * src/types.c (double, longdouble): Merge identical SH and ARM
3427 typedefs, and add POWERPC64.
3428 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
3429 struct split over gpr and rest.
3430 (ffi_prep_cif_machdep): Correct intarg_count for structures.
3431 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
3433 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
3435 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
3438 2003-09-09 Alan Modra <amodra@bigpond.net.au>
3440 * configure: Regenerate.
3442 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3444 * Makefile.am: Remove build rules for ffitest.
3445 * Makefile.in: Rebuilt.
3447 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3449 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3450 about implicit declaration of abort().
3452 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3454 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3455 * Makefile.in: Rebuilt.
3456 * configure.in: Add dejagnu test framework.
3457 * configure: Rebuilt.
3459 * testsuite/Makefile.am: New file.
3460 * testsuite/Makefile.in: Built
3461 * testsuite/lib/libffi-dg.exp: New file.
3462 * testsuite/config/default.exp: Likewise.
3463 * testsuite/libffi.call/call.exp: Likewise.
3464 * testsuite/libffi.call/ffitest.h: Likewise.
3465 * testsuite/libffi.call/closure_fn0.c: Likewise.
3466 * testsuite/libffi.call/closure_fn1.c: Likewise.
3467 * testsuite/libffi.call/closure_fn2.c: Likewise.
3468 * testsuite/libffi.call/closure_fn3.c: Likewise.
3469 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3470 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3471 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3472 * testsuite/libffi.call/cls_2byte.c: Likewise.
3473 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3474 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3475 * testsuite/libffi.call/cls_4byte.c: Likewise.
3476 * testsuite/libffi.call/cls_5byte.c: Likewise.
3477 * testsuite/libffi.call/cls_6byte.c: Likewise.
3478 * testsuite/libffi.call/cls_7byte.c: Likewise.
3479 * testsuite/libffi.call/cls_8byte.c: Likewise.
3480 * testsuite/libffi.call/cls_12byte.c: Likewise.
3481 * testsuite/libffi.call/cls_16byte.c: Likewise.
3482 * testsuite/libffi.call/cls_20byte.c: Likewise.
3483 * testsuite/libffi.call/cls_24byte.c: Likewise.
3484 * testsuite/libffi.call/cls_double.c: Likewise.
3485 * testsuite/libffi.call/cls_float.c: Likewise.
3486 * testsuite/libffi.call/cls_uchar.c: Likewise.
3487 * testsuite/libffi.call/cls_uint.c: Likewise.
3488 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3489 * testsuite/libffi.call/cls_ushort.c: Likewise.
3490 * testsuite/libffi.call/float.c: Likewise.
3491 * testsuite/libffi.call/float1.c: Likewise.
3492 * testsuite/libffi.call/float2.c: Likewise.
3493 * testsuite/libffi.call/many.c: Likewise.
3494 * testsuite/libffi.call/many_win32.c: Likewise.
3495 * testsuite/libffi.call/nested_struct.c: Likewise.
3496 * testsuite/libffi.call/nested_struct1.c: Likewise.
3497 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3498 * testsuite/libffi.call/problem1.c: Likewise.
3499 * testsuite/libffi.call/promotion.c: Likewise.
3500 * testsuite/libffi.call/return_ll.c: Likewise.
3501 * testsuite/libffi.call/return_sc.c: Likewise.
3502 * testsuite/libffi.call/return_uc.c: Likewise.
3503 * testsuite/libffi.call/strlen.c: Likewise.
3504 * testsuite/libffi.call/strlen_win32.c: Likewise.
3505 * testsuite/libffi.call/struct1.c: Likewise.
3506 * testsuite/libffi.call/struct2.c: Likewise.
3507 * testsuite/libffi.call/struct3.c: Likewise.
3508 * testsuite/libffi.call/struct4.c: Likewise.
3509 * testsuite/libffi.call/struct5.c: Likewise.
3510 * testsuite/libffi.call/struct6.c: Likewise.
3511 * testsuite/libffi.call/struct7.c: Likewise.
3512 * testsuite/libffi.call/struct8.c: Likewise.
3513 * testsuite/libffi.call/struct9.c: Likewise.
3514 * testsuite/libffi.special/special.exp: New file.
3515 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3516 * testsuite/libffi.special/unwindtest.cc: Likewise.
3519 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3521 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3524 2003-08-02 Alan Modra <amodra@bigpond.net.au>
3526 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3528 (ffi_closure_helper_LINUX64): Likewise.
3529 * src/powerpc/linux64.S: Remove code writing to parm save area.
3530 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3531 address in lr from ffi_closure_helper_LINUX64 call to calculate
3532 table address. Optimize function tail.
3534 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3536 * src/sparc/ffi.c: Handle all floating point registers.
3537 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3539 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3541 * README: Note that libffi is not part of GCC. Update the project
3544 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3546 * src/powerpc/ppc_closure.S: Include ffi.h.
3548 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3550 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3551 Use C style comments.
3553 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3555 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3556 * Makefile.in: Regenerate.
3557 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3558 * configure: Regenerate.
3559 * include/ffi.h.in: Add SHmedia support.
3560 * src/sh64/ffi.c: New file.
3561 * src/sh64/sysv.S: New file.
3563 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3565 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3566 should be read-only.
3567 * configure: Rebuilt.
3568 * fficonfig.h.in: Rebuilt.
3569 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3570 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3571 * src/powerpc/linux64.S: Likewise.
3572 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3573 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3574 if -fpic/-fPIC/-mrelocatable.
3575 * src/powerpc/powerpc_closure.S: Likewise.
3576 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3577 #write in .eh_frame flags.
3578 * src/sparc/v9.S: Likewise.
3579 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3580 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3581 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3583 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3585 Fixes PR bootstrap/10656
3586 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3587 support for .register pseudo-op.
3588 * src/sparc/v8.S: Use it.
3589 * fficonfig.h.in: Rebuilt.
3590 * configure: Rebuilt.
3592 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3594 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3595 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3596 Make it the default on POWERPC64.
3597 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3598 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3599 * configure: Rebuilt.
3600 * src/powerpc/ffi.c (hidden): Define.
3601 (ffi_prep_args_SYSV): Renamed from
3602 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3603 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3604 ASM_NEEDS_REGISTERS64): New.
3605 (ffi_prep_args64): New function.
3606 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3607 (ffi_call): Likewise.
3608 (ffi_prep_closure): Likewise.
3609 (flush_icache): Surround by #ifndef POWERPC64.
3610 (ffi_dblfl): New union type.
3611 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3612 (ffi_closure_helper_LINUX64): New function.
3613 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3615 * src/powerpc/sysv.S: Likewise.
3616 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3617 * src/powerpc/linux64.S: New file.
3618 * src/powerpc/linux64_closure.S: New file.
3619 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3620 src/powerpc/linux64_closure.S.
3621 (TARGET_SRC_POWERPC): Likewise.
3623 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3624 closure_test_fn3): Fix result printing on big-endian 64-bit
3626 (main): Print tst2_arg instead of uninitialized tst2_result.
3628 * src/ffitest.c (main): Hide what closure pointer really points to
3631 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3633 * configure.in (arm-*-netbsdelf*): Add configuration.
3634 (configure): Regenerated.
3636 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3638 * include/Makefile.in: Regenerate.
3640 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3642 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3644 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3645 Receive closure pointer through parameter, read args using
3646 __builtin_dwarf_cfa.
3647 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3649 2003-03-12 Andreas Schwab <schwab@suse.de>
3651 * configure.in: Avoid trailing /. in toolexeclibdir.
3652 * configure: Rebuilt.
3654 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3656 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3658 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3660 * libffi/src/powerpc/darwin_closure.S:
3661 Fix alignement bug, allocate 8 bytes for the result.
3662 * libffi/src/powerpc/aix_closure.S:
3664 * libffi/src/powerpc/ffi_darwin.c:
3665 Update stackframe description for aix/darwin_closure.S.
3667 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3669 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3672 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3673 Andreas Schwab <schwab@suse.de>
3675 * configure.in: Adjust command to source config-ml.in to account
3676 for changes to the libffi_basedir definition.
3677 (libffi_basedir): Remove ${srcdir} from value and include trailing
3680 * configure: Regenerate.
3682 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3684 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3686 2003-01-28 Andrew Haley <aph@redhat.com>
3688 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3689 * src/x86/ffi64.c (ffi_prep_closure): New.
3690 (ffi_closure_UNIX64_inner): New.
3691 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3693 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
3695 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3696 Remove USE_LIBDIR conditional.
3697 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3698 * Makefile.in, configure: Rebuilt.
3700 2003-01027 David Edelsohn <edelsohn@gnu.org>
3702 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3703 * Makefile.in: Regenerate.
3705 2003-01-22 Andrew Haley <aph@redhat.com>
3707 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3710 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3712 * src/powerpc/darwin.S: Add unwind info.
3713 * src/powerpc/darwin_closure.S: Likewise.
3715 2003-01-14 Andrew Haley <aph@redhat.com>
3717 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3718 (ffi_prep_cif_machdep): Likewise.
3719 * src/x86/unix64.S: Add unwind info.
3721 2003-01-14 Andreas Jaeger <aj@suse.de>
3723 * src/ffitest.c (main): Only use ffi_closures if those are
3726 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3728 * libffi/src/ffitest.c
3729 add closure testcases
3731 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3733 * libffi/src/powerpc/ffi.c
3734 fix alignment bug for float (4 byte aligned iso 8 byte)
3736 2003-01-09 Geoffrey Keating <geoffk@apple.com>
3738 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3739 * src/powerpc/darwin.S: Remove RCS version string.
3741 2003-01-03 Jeff Sturm <jsturm@one-point.com>
3743 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3744 * src/ffitest.c (main): Use static storage for closure.
3745 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3746 * src/sparc/v8.S (ffi_closure_v8): New.
3747 * src/sparc/v9.S (ffi_closure_v9): New.
3749 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3751 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3752 enumeration for X86_WIN32.
3753 * src/x86/win32.S: Added ffi_call_STDCALL function
3755 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3756 switch cases for recognising FFI_STDCALL and
3757 calling ffi_call_STDCALL if target is X86_WIN32.
3758 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3759 stdcall versions of the "my_strlen" and "many"
3760 test functions (for X86_WIN32).
3761 Added test cases to test stdcall invocation using
3764 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3766 * src/sh/sysv.S: Add DWARF2 unwind info.
3768 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3770 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3772 2002-11-26 Jim Wilson <wilson@redhat.com>
3774 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3776 2002-11-23 H.J. Lu <hjl@gnu.org>
3778 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3779 Include ../config/accross.m4.
3780 * aclocal.m4; Rebuild.
3781 * configure: Likewise.
3783 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3785 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3787 2002-11-11 DJ Delorie <dj@redhat.com>
3789 * configure.in: Look for common files in the right place.
3791 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3793 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3794 raw data as _Jv_word values, not ffi_raw.
3795 (ffi_java_ptrarray_to_raw): Likewise.
3796 (ffi_java_rvalue_to_raw): New function.
3797 (ffi_java_raw_call): Call it.
3798 (ffi_java_raw_to_rvalue): New function.
3799 (ffi_java_translate_args): Call it.
3800 * src/ffitest.c (closure_test_fn): Interpret return value
3801 as ffi_arg, not int.
3802 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3803 FFI_TYPE_POINTER case.
3804 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3805 values extended to word size.
3807 2002-10-02 Andreas Jaeger <aj@suse.de>
3809 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3811 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
3813 * include/ffi.h.in: Fix i386 win32 compilation.
3815 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3817 * configure.in: Add s390x-*-linux-* target.
3818 * configure: Regenerate.
3819 * include/ffi.h.in: Define S390X for s390x targets.
3820 (FFI_CLOSURES): Define for s390/s390x.
3821 (FFI_TRAMPOLINE_SIZE): Likewise.
3822 (FFI_NATIVE_RAW_API): Likewise.
3823 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3824 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
3825 * src/s390/ffi.c: Major rework of existing code. Add support for
3826 s390x targets. Add closure support.
3827 * src/s390/sysv.S: Likewise.
3829 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
3831 * src/arm/sysv.S: Fix typo.
3833 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
3835 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3836 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3837 (ffi_call_SYSV): Handle soft-float.
3839 2002-09-27 Bo Thorsen <bo@suse.de>
3841 * include/ffi.h.in: Fix multilib x86-64 support.
3843 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3845 * Makefile.am (all-multi): Fix multilib parallel build.
3847 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3849 * configure.in (sh[34]*-*-linux*): Add brackets.
3850 * configure: Regenerate.
3852 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3854 * Makefile.am: Add SH support.
3855 * Makefile.in: Regenerate.
3856 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3857 * configure: Regenerate.
3858 * include/ffi.h.in: Add SH support.
3859 * src/sh/ffi.c: New file.
3860 * src/sh/sysv.S: New file.
3861 * src/types.c: Add SH support.
3863 2002-07-16 Bo Thorsen <bo@suse.de>
3865 * src/x86/ffi64.c: New file that adds x86-64 support.
3866 * src/x86/unix64.S: New file that handles argument setup for
3868 * src/x86/sysv.S: Don't use this on x86-64.
3869 * src/x86/ffi.c: Don't use this on x86-64.
3871 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3873 * src/ffitest.c (struct6): New test that tests a special case in
3875 (struct7): Likewise.
3876 (struct8): Likewise.
3877 (struct9): Likewise.
3878 (closure_test_fn): Silence warning about this when it's not used.
3879 (main): Add the new tests.
3880 (main): Fix a couple of wrong casts and silence some compiler warnings.
3881 * include/ffi.h.in: Add x86-64 ABI definition.
3882 * fficonfig.h.in: Regenerate.
3883 * Makefile.am: Add x86-64 support.
3884 * configure.in: Likewise.
3885 * Makefile.in: Regenerate.
3886 * configure: Likewise.
3888 2002-06-24 Bo Thorsen <bo@suse.de>
3890 * src/types.c: Merge settings for similar architectures.
3891 Add x86-64 sizes and alignments.
3893 2002-06-23 Bo Thorsen <bo@suse.de>
3895 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3896 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3897 * src/mips/ffi.c (ffi_prep_args): Likewise.
3898 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3900 2002-07-18 H.J. Lu (hjl@gnu.org)
3902 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3903 (libffi_la_SOURCES): Support MIPS_LINUX.
3904 (libffi_convenience_la_SOURCES): Likewise.
3905 * Makefile.in: Regenerated.
3907 * configure.in (mips64*-*): Skip.
3908 (mips*-*-linux*): New.
3909 * configure: Regenerated.
3911 * src/mips/ffi.c: Include <sgidefs.h>.
3913 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3915 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3917 2002-05-27 Roger Sayle <roger@eyesopen.com>
3919 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3921 2002-05-27 Bo Thorsen <bo@suse.de>
3923 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3926 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3928 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3929 beginning of function (for older apple cc).
3931 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3933 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3934 script entry, and set LD to it when configuring multilibs.
3935 * configure: Rebuilt.
3937 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3939 * configure.in (sparc64-*-netbsd*): Add target.
3940 (sparc-*-netbsdelf*): Likewise.
3941 * configure: Regenerate.
3943 2002-04-28 David S. Miller <davem@redhat.com>
3945 * configure.in, configure: Fix SPARC test in previous change.
3947 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3949 * Makefile.am: Add Linux for S/390 support.
3950 * Makefile.in: Regenerate.
3951 * configure.in: Add Linux for S/390 support.
3952 * configure: Regenerate.
3953 * include/ffi.h.in: Add Linux for S/390 support.
3954 * src/s390/ffi.c: New file from libffi CVS tree.
3955 * src/s390/sysv.S: New file from libffi CVS tree.
3957 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3959 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3961 * src/sparc/v8.S: Use it.
3962 * src/sparc/v9.S: Likewise.
3963 * fficonfig.h.in: Rebuilt.
3964 * configure: Rebuilt.
3966 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3968 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3970 * src/ia64/unix.S: Add unwind information. Fix comments.
3971 Save sp in a way that's compatible with unwind info.
3972 (ffi_call_unix): Correctly restore sp in all cases.
3973 * src/ia64/ffi.c: Add, fix comments.
3975 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3977 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3979 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3981 * configure.in (alpha*-*-netbsd*): Add target.
3982 * configure: Regenerate.
3984 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3986 * src/sparc/v8.S: Add unwind info.
3987 * src/sparc/v9.S: Likewise.
3989 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3991 * configure.in: Enable i*86-*-netbsdelf*.
3992 * configure: Rebuilt.
3994 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3997 * src/mips/n32.s: Delete
3998 * src/mips/o32.s: Delete
4000 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
4002 * configure.in: Enable alpha*-*-freebsd*.
4003 * configure: Rebuilt.
4005 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
4007 * Makefile.am: libfficonvenience -> libffi_convenience.
4008 * Makefile.in: Rebuilt.
4010 * Makefile.am: Define ffitest_OBJECTS.
4011 * Makefile.in: Rebuilt.
4013 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
4014 David Edelsohn <edelsohn@gnu.org>
4016 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
4017 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
4018 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
4019 * Makefile.in: Regenerate.
4020 * include/ffi.h.in: Add AIX and Darwin closure definitions.
4021 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
4022 (flush_icache, flush_range): New functions.
4023 (ffi_closure_helper_DARWIN): New function.
4024 * src/powerpc/aix_closure.S: New file.
4025 * src/powerpc/darwin_closure.S: New file.
4027 2002-02-24 Jeff Sturm <jsturm@one-point.com>
4029 * include/ffi.h.in: Add typedef for ffi_arg.
4030 * src/ffitest.c (main): Declare rint with ffi_arg.
4032 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
4034 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
4035 number of GPRs for floating-point arguments.
4037 2002-01-31 Anthony Green <green@redhat.com>
4039 * configure: Rebuilt.
4040 * configure.in: Replace CHECK_SIZEOF and endian tests with
4041 cross-compiler friendly macros.
4042 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
4045 2002-01-18 David Edelsohn <edelsohn@gnu.org>
4047 * src/powerpc/darwin.S (_ffi_call_AIX): New.
4048 * src/powerpc/aix.S (ffi_call_DARWIN): New.
4050 2002-01-17 David Edelsohn <edelsohn@gnu.org>
4052 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
4053 (TARGET_SRC_POWERPC_AIX): New.
4054 (POWERPC_AIX): New stanza.
4055 * Makefile.in: Regenerate.
4056 * configure.in: Add AIX case.
4057 * configure: Regenerate.
4058 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
4059 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
4060 size. Fix "long double" support.
4061 (ffi_call): Add FFI_AIX case.
4062 * src/powerpc/aix.S: New.
4064 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
4066 Implement Darwin PowerPC ABI.
4067 * configure.in: Handle powerpc-*-darwin*.
4068 * Makefile.am: Set source files for POWERPC_DARWIN.
4069 * configure: Rebuilt.
4070 * Makefile.in: Rebuilt.
4071 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
4073 * src/powerpc/darwin.S: New file.
4074 * src/powerpc/ffi_darwin.c: New file.
4076 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
4078 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
4080 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4082 * src/x86/sysv.S: Avoid gas-only .balign directive.
4083 Use C style comments.
4085 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4087 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
4088 Fixes PR bootstrap/3563.
4090 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4092 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
4094 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4096 * configure.in: Recognize sparc*-sun-* host.
4097 * configure: Regenerate.
4099 2001-06-06 Andrew Haley <aph@redhat.com>
4101 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
4103 2001-06-03 Andrew Haley <aph@redhat.com>
4105 * src/alpha/osf.S: Add unwind info.
4106 * src/powerpc/sysv.S: Add unwind info.
4107 * src/powerpc/ppc_closure.S: Likewise.
4109 2000-05-31 Jeff Sturm <jsturm@one-point.com>
4111 * configure.in: Fix AC_ARG_ENABLE usage.
4112 * configure: Rebuilt.
4114 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
4116 * configure.in: Remove warning about beta code.
4117 * configure: Rebuilt.
4119 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
4121 * src/ia64/unix.S: Restore stack pointer when returning from
4123 * src/ia64/ffi.c: Fix typo in comment.
4125 2001-04-18 Jim Wilson <wilson@redhat.com>
4127 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
4128 to eliminate RAW DV.
4130 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
4132 * Makefile.am: Make a libtool convenience library.
4133 * Makefile.in: Rebuilt.
4135 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
4137 * configure.in: Use different syntax for subdirectory creation.
4138 * configure: Rebuilt.
4140 2001-03-27 Jon Beniston <jon@beniston.com>
4142 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
4143 * configure: Rebuilt.
4144 * Makefile.am: Added X86_WIN32 target support.
4145 * Makefile.in: Rebuilt.
4147 * include/ffi.h.in: Added X86_WIN32 target support.
4149 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
4150 * src/types.c: Added X86_WIN32 target support.
4152 * src/x86/win32.S: New file. Based on sysv.S, but with EH
4153 stuff removed and made to work with CygWin's gas.
4155 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
4157 * configure.in: Make target subdirectory in build dir.
4158 * Makefile.am: Override suffix based rules to specify correct output
4160 * Makefile.in: Rebuilt.
4161 * configure: Rebuilt.
4163 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
4165 * src/powerpc/ppc_closure.S: New file.
4166 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
4167 involving long long and register pairs.
4168 (ffi_prep_closure): New function.
4169 (flush_icache): Likewise.
4170 (ffi_closure_helper_SYSV): Likewise.
4171 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
4172 (FFI_TRAMPOLINE_SIZE): Likewise.
4173 (FFI_NATIVE_RAW_API): Likewise.
4174 * Makefile.in: Rebuilt.
4175 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
4176 (TARGET_SRC_POWERPC): Likewise.
4178 2001-03-19 Tom Tromey <tromey@redhat.com>
4180 * Makefile.in: Rebuilt.
4181 * Makefile.am (ffitest_LDFLAGS): New macro.
4183 2001-03-02 Nick Clifton <nickc@redhat.com>
4185 * include/ffi.h.in: Remove RCS ident string.
4186 * include/ffi_mips.h: Remove RCS ident string.
4187 * src/debug.c: Remove RCS ident string.
4188 * src/ffitest.c: Remove RCS ident string.
4189 * src/prep_cif.c: Remove RCS ident string.
4190 * src/types.c: Remove RCS ident string.
4191 * src/alpha/ffi.c: Remove RCS ident string.
4192 * src/alpha/osf.S: Remove RCS ident string.
4193 * src/arm/ffi.c: Remove RCS ident string.
4194 * src/arm/sysv.S: Remove RCS ident string.
4195 * src/mips/ffi.c: Remove RCS ident string.
4196 * src/mips/n32.S: Remove RCS ident string.
4197 * src/mips/o32.S: Remove RCS ident string.
4198 * src/sparc/ffi.c: Remove RCS ident string.
4199 * src/sparc/v8.S: Remove RCS ident string.
4200 * src/sparc/v9.S: Remove RCS ident string.
4201 * src/x86/ffi.c: Remove RCS ident string.
4202 * src/x86/sysv.S: Remove RCS ident string.
4204 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
4206 * include/ffi.h.in: Change sourceware.cygnus.com references to
4209 2000-12-09 Richard Henderson <rth@redhat.com>
4211 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
4212 (ffi_closure_osf_inner): Index rather than increment avalue
4213 and arg_types. Give ffi_closure_osf the raw return value type.
4214 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
4217 2000-12-07 Richard Henderson <rth@redhat.com>
4219 * src/raw_api.c (ffi_translate_args): Fix typo.
4220 (ffi_prep_closure): Likewise.
4222 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
4223 FFI_TRAMPOLINE_SIZE.
4224 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
4225 cif->bytes for new ffi_call_osf implementation.
4226 (ffi_prep_args): Absorb into ...
4227 (ffi_call): ... here. Do all stack allocation here and
4228 avoid a callback function.
4229 (ffi_prep_closure, ffi_closure_osf_inner): New.
4230 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
4231 (ffi_closure_osf): New.
4233 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
4235 * config.guess, config.sub, install-sh: Removed.
4236 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
4237 * Makefile.in: Rebuilt.
4239 * acinclude.m4: Include libtool macros from the top level.
4240 * aclocal.m4, configure: Rebuilt.
4242 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
4244 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
4245 * configure: Rebuilt.
4247 2000-05-11 Scott Bambrough <scottb@netwinder.org>
4249 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
4250 memory correctly. Use conditional instructions, not branches where
4253 2000-05-04 Tom Tromey <tromey@cygnus.com>
4255 * configure: Rebuilt.
4256 * configure.in: Match `arm*-*-linux-*'.
4257 From Chris Dornan <cdornan@arm.com>.
4259 2000-04-28 Jakub Jelinek <jakub@redhat.com>
4261 * Makefile.am (SUBDIRS): Define.
4262 (AM_MAKEFLAGS): Likewise.
4263 (Multilib support.): Add section.
4264 * Makefile.in: Rebuilt.
4265 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
4266 New variables. Set for gcc using -print-multi-lib. Export them
4268 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
4269 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
4271 (extra_compiler_flags_value, extra_compiler_flags): Check these
4272 for extra compiler options which need to be passed down in
4275 2000-04-16 Anthony Green <green@redhat.com>
4277 * configure: Rebuilt.
4278 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
4280 2000-04-14 Jakub Jelinek <jakub@redhat.com>
4282 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
4283 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
4284 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
4285 Replace all void * sizeofs with sizeof(int).
4286 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
4287 different than DOUBLE.
4288 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
4289 (ffi_prep_args_v9): New function.
4290 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
4291 (ffi_V9_return_struct): New function.
4292 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
4293 32bit code (not yet cross-arch calls).
4294 * src/sparc/v8.S: Add struct return delay nop.
4296 * src/sparc/v9.S: New file.
4297 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
4298 is used on sparc64 only for structures larger than 32 bytes.
4299 Pass by reference for structures is done for structure arguments
4300 larger than 16 bytes.
4301 * src/ffitest.c (main): Use 64bit rint on sparc64.
4302 Run long long tests on sparc.
4303 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
4305 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
4307 * configure.in (sparc-*-linux*): New supported target.
4308 (sparc64-*-linux*): Likewise.
4309 * configure: Rebuilt.
4310 * Makefile.am: Add v9.S to SPARC files.
4311 * Makefile.in: Likewise.
4312 (LINK): Surround $(CCLD) into double quotes, so that multilib
4313 compiles work correctly.
4315 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
4317 * configure: Rebuilt.
4318 * configure.in: (i*86-*-solaris*): New libffi target. Patch
4319 proposed by Bryce McKinlay.
4321 2000-03-20 Tom Tromey <tromey@cygnus.com>
4323 * Makefile.in: Hand edit for java_raw_api.lo.
4325 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
4327 * config.guess, config.sub: Update from the gcc tree.
4328 Fix for PR libgcj/168.
4330 2000-03-03 Tom Tromey <tromey@cygnus.com>
4332 * Makefile.in: Fixed ia64 by hand.
4334 * configure: Rebuilt.
4335 * configure.in (--enable-multilib): New option.
4336 (libffi_basedir): New subst.
4337 (AC_OUTPUT): Added multilib code.
4339 2000-03-02 Tom Tromey <tromey@cygnus.com>
4341 * Makefile.in: Rebuilt.
4342 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
4345 2000-02-25 Hans Boehm <boehm@acm.org>
4347 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
4349 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
4351 (ffi_prep_raw_closure): Use ffi_translate_args, not
4352 ffi_closure_translate.
4353 * src/java_raw_api.c: New file.
4354 * src/ffitest.c (closure_test_fn): New function.
4355 (main): Define `rint' as long long on IA64. Added new test when
4356 FFI_CLOSURES is defined.
4357 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
4358 (ffi_abi): Recognize IA64.
4359 (ffi_raw): Added `flt' field.
4360 Added "Java raw API" code.
4361 * configure.in: Recognize ia64.
4362 * Makefile.am (TARGET_SRC_IA64): New macro.
4363 (libffi_la_common_SOURCES): Added java_raw_api.c.
4364 (libffi_la_SOURCES): Define in IA64 case.
4366 2000-01-04 Tom Tromey <tromey@cygnus.com>
4368 * Makefile.in: Rebuilt with newer automake.
4370 1999-12-31 Tom Tromey <tromey@cygnus.com>
4372 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
4374 1999-09-01 Tom Tromey <tromey@cygnus.com>
4376 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
4378 * fficonfig.h.in: Rebuilt.
4379 * configure: Rebuilt.
4380 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
4381 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
4382 * acconfig.h: Don't #undef PACKAGE or VERSION.
4384 1999-08-09 Anthony Green <green@cygnus.com>
4386 * include/ffi.h.in: Try to work around messy header problem
4387 with PACKAGE and VERSION.
4389 * configure: Rebuilt.
4390 * configure.in: Change version to 2.00-beta.
4392 * fficonfig.h.in: Rebuilt.
4393 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
4395 * src/x86/ffi.c (ffi_raw_call): Rename.
4397 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
4399 * src/x86/ffi.c (ffi_closure_SYSV): New function.
4400 (ffi_prep_incoming_args_SYSV): Ditto.
4401 (ffi_prep_closure): Ditto.
4402 (ffi_closure_raw_SYSV): Ditto.
4403 (ffi_prep_raw_closure): More ditto.
4404 (ffi_call_raw): Final ditto.
4406 * include/ffi.h.in: Add definitions for closure and raw API.
4408 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4411 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
4413 * src/raw_api.c: New file.
4415 * include/ffi.h.in (ffi_raw): New type.
4416 (UINT_ARG, SINT_ARG): New defines.
4417 (ffi_closure, ffi_raw_closure): New types.
4418 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4420 * configure.in: Add check for endianness and sizeof void*.
4422 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
4423 instead of directly.
4425 * configure: Rebuilt.
4427 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
4429 * configure.in: Add x86 and powerpc BeOS configurations.
4430 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
4432 1999-05-09 Anthony Green <green@cygnus.com>
4434 * configure.in: Add warning about this being beta code.
4435 Remove src/Makefile.am from the picture.
4436 * configure: Rebuilt.
4438 * Makefile.am: Move logic from src/Makefile.am. Add changes
4439 to support libffi as a target library.
4440 * Makefile.in: Rebuilt.
4442 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
4443 Upgraded to new autoconf, automake, libtool.
4447 * LICENSE: Update copyright date.
4449 * src/Makefile.am, src/Makefile.in: Removed.
4451 1998-11-29 Anthony Green <green@cygnus.com>
4453 * include/ChangeLog: Removed.
4454 * src/ChangeLog: Removed.
4455 * src/mips/ChangeLog: Removed.
4456 * src/sparc/ChangeLog: Remboved.
4457 * src/x86/ChangeLog: Removed.
4459 * ChangeLog.v1: Created.