1 2012-01-19 Jakub Jelinek <jakub@redhat.com>
3 PR rtl-optimization/48496
4 * src/ia64/ffi.c (ffi_call): Fix up aliasing violations.
6 2012-01-09 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
8 * configure.ac (i?86-*-*): Set TARGET to X86_64.
9 * configure: Regenerate.
11 2011-12-07 Andrew Pinski <apinski@cavium.com>
14 * src/mips/n32.S: Add ".set mips4".
16 2011-11-21 Andreas Tobler <andreast@fgznet.ch>
18 * configure: Regenerate.
20 2011-11-10 Richard Henderson <rth@redhat.com>
22 * configure.ac (GCC_AS_CFI_PSEUDO_OP): Use it instead of inline check.
23 * configure, aclocal.m4: Rebuild.
25 2011-09-04 Iain Sandoe <iains@gcc.gnu.org>
28 * src/powerpc/darwin_closure.S (stubs): Make the stub binding
29 helper reference track the architecture pointer size.
31 2011-08-25 Andrew Haley <aph@redhat.com>
33 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Remove hard-coded assembly
35 * src/arm/sysv.S (ffi_arm_trampoline): Put them here instead.
37 2011-07-11 Andrew Haley <aph@redhat.com>
39 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Clear icache.
41 2011-06-29 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
43 * testsuite/libffi.call/cls_double_va.c: Move PR number to comment.
44 * testsuite/libffi.call/cls_longdouble_va.c: Likewise.
46 2011-06-29 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
49 * testsuite/libffi.call/cls_double_va.c: xfail dg-output on
51 * testsuite/libffi.call/cls_longdouble_va.c: Likewise.
53 2011-06-14 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
55 * testsuite/libffi.call/huge_struct.c (test_large_fn): Use PRIu8,
56 PRId8 instead of %hhu, %hhd.
57 * testsuite/libffi.call/ffitest.h [__alpha__ && __osf__] (PRId8,
59 [__sgi__] (PRId8, PRIu8): Define.
61 2011-04-29 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
63 * src/alpha/osf.S (UA_SI, FDE_ENCODING, FDE_ENCODE, FDE_ARANGE):
65 Use them to handle ELF vs. ECOFF differences.
66 [__osf__] (_GLOBAL__F_ffi_call_osf): Define.
68 2011-02-13 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
70 * configure: Regenerate.
72 2011-02-09 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
75 * testsuite/libffi.call/cls_pointer.c (main): Cast void * to
77 * testsuite/libffi.call/cls_pointer_stack.c (main): Likewise.
79 2011-02-07 Joel Sherrill <joel.sherrill@oarcorp.com>
81 * libffi/src/m68k/ffi.c: Add RTEMS support for cache flushing.
82 Handle case when CPU variant does not have long double support.
83 * libffi/src/m68k/sysv.S: Add support for mc68000, Coldfire,
84 and cores with soft floating point.
86 2011-02-07 Joel Sherrill <joel.sherrill@oarcorp.com>
88 * configure.ac: Add mips*-*-rtems* support.
89 * configure: Regenerate.
90 * src/mips/ffitarget.h: Ensure needed constants are available
91 for targets which do not have sgidefs.h.
93 2011-01-26 Dave Korn <dave.korn.cygwin@gmail.com>
96 * configure.ac (AM_LTLDFLAGS): Add -bindir option for windows DLLs.
97 * configure: Regenerate.
99 2010-12-18 Iain Sandoe <iains@gcc.gnu.org>
103 * src/powerpc/darwin_closure.S: Provide Darwin64 implementation,
105 * src/powerpc/ffitarget.h (POWERPC_DARWIN64): New,
106 (FFI_TRAMPOLINE_SIZE): Update for Darwin64.
107 * src/powerpc/darwin.S: Provide Darwin64 implementation,
109 * src/powerpc/ffi_darwin.c: Likewise.
111 2010-12-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
113 * configure.ac (libffi_cv_as_ascii_pseudo_op): Use double
115 (libffi_cv_as_string_pseudo_op): Likewise.
116 * configure: Regenerate.
118 2010-12-03 Chung-Lin Tang <cltang@codesourcery.com>
120 * src/arm/sysv.S (ffi_closure_SYSV): Add UNWIND to .pad directive.
121 (ffi_closure_VFP): Same.
122 (ffi_call_VFP): Move down to before ffi_closure_VFP. Add '.fpu vfp'
125 2010-12-01 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
127 * testsuite/libffi.call/ffitest.h [__sgi] (PRId64, PRIu64): Define.
130 2010-11-29 Richard Henderson <rth@redhat.com>
131 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
133 * src/x86/sysv.S (FDE_ENCODING, FDE_ENCODE): Define.
134 (.eh_frame): Use FDE_ENCODING.
135 (.LASFDE1, .LASFDE2, LASFDE3): Simplify with FDE_ENCODE.
137 2010-11-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
139 * testsuite/lib/libffi-dg.exp: Rename ...
140 * testsuite/lib/libffi.exp: ... to this.
141 * libffi/testsuite/libffi.call/call.exp: Don't load libffi-dg.exp.
142 * libffi/testsuite/libffi.special/special.exp: Likewise.
144 2010-10-28 Chung-Lin Tang <cltang@codesourcery.com>
146 * src/arm/ffi.c (ffi_prep_args): Add VFP register argument handling
147 code, new parameter, and return value. Update comments.
148 (ffi_prep_cif_machdep): Add case for VFP struct return values. Add
149 call to layout_vfp_args().
150 (ffi_call_SYSV): Update declaration.
151 (ffi_call_VFP): New declaration.
152 (ffi_call): Add VFP struct return conditions. Call ffi_call_VFP()
154 (ffi_closure_VFP): New declaration.
155 (ffi_closure_SYSV_inner): Add new vfp_args parameter, update call to
156 ffi_prep_incoming_args_SYSV().
157 (ffi_prep_incoming_args_SYSV): Update parameters. Add VFP argument
159 (ffi_prep_closure_loc): Pass ffi_closure_VFP to trampoline
160 construction under VFP hard-float.
161 (rec_vfp_type_p): New function.
163 (place_vfp_arg): Same.
164 (layout_vfp_args): Same.
165 * src/arm/ffitarget.h (ffi_abi): Add FFI_VFP. Define FFI_DEFAULT_ABI
166 based on __ARM_PCS_VFP.
167 (FFI_EXTRA_CIF_FIELDS): Define for adding VFP hard-float specific
169 (FFI_TYPE_STRUCT_VFP_FLOAT): Define internally used type code.
170 (FFI_TYPE_STRUCT_VFP_DOUBLE): Same.
171 * src/arm/sysv.S (ffi_call_SYSV): Change call of ffi_prep_args() to
172 direct call. Move function pointer load upwards.
173 (ffi_call_VFP): New function.
174 (ffi_closure_VFP): Same.
176 * testsuite/lib/libffi-dg.exp (check-flags): New function.
177 (dg-skip-if): New function.
178 * testsuite/libffi.call/cls_double_va.c: Skip if target is arm*-*-*
179 and compiler options include -mfloat-abi=hard.
180 * testsuite/libffi.call/cls_longdouble_va.c: Same.
182 2010-10-01 Jakub Jelinek <jakub@redhat.com>
185 * src/x86/ffi64.c (ffi_prep_cif_machdep): Ensure cif->bytes is
187 * testsuite/libffi.call/many2.c: New test.
189 2010-08-20 Mark Wielaard <mjw@redhat.com>
191 * src/closures.c (open_temp_exec_file_mnt): Check if getmntent_r
194 2010-08-09 Andreas Tobler <andreast@fgznet.ch>
196 * configure.ac: Add target powerpc64-*-freebsd*.
197 * configure: Regenerate.
198 * testsuite/libffi.call/cls_align_longdouble_split.c: Pass
199 -mlong-double-128 only to linux targets.
200 * testsuite/libffi.call/cls_align_longdouble_split2.c: Likewise.
201 * testsuite/libffi.call/cls_longdouble.c: Likewise.
202 * testsuite/libffi.call/huge_struct.c: Likewise.
204 2010-07-10 Evan Phoenix <evan@fallingsnow.net>
206 * src/closures.c (selinux_enabled_check): Fix strncmp usage bug.
208 2010-07-07 Dan HorĂ¡k <dan@danny.cz>
210 * include/ffi.h.in: Protect #define with #ifndef.
211 * src/powerpc/ffitarget.h: Ditto.
212 * src/s390/ffitarget.h: Ditto.
213 * src/sparc/ffitarget.h: Ditto.
215 2010-07-07 Neil Roberts <neil@linux.intel.com>
217 * src/x86/sysv.S (ffi_call_SYSV): Align the stack pointer to
220 2010-07-02 Jakub Jelinek <jakub@redhat.com>
222 * Makefile.am (AM_MAKEFLAGS): Pass also mandir to submakes.
223 * Makefile.in: Regenerated.
225 2010-05-19 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
227 * configure.ac (libffi_cv_as_x86_pcrel): Check for illegal in as
229 (libffi_cv_as_ascii_pseudo_op): Check for .ascii.
230 (libffi_cv_as_string_pseudo_op): Check for .string.
231 * configure: Regenerate.
232 * fficonfig.h.in: Regenerate.
233 * src/x86/sysv.S (.eh_frame): Use .ascii, .string or error.
235 2010-05-05 Michael Kohler <michaelkohler@live.com>
237 * src/dlmalloc.c (dlfree): Fix spelling.
238 * src/ia64/ffi.c (ffi_prep_cif_machdep): Ditto.
239 * configure.ac: Ditto.
240 * configure: Rebuilt.
242 2010-04-13 Dan Witte <dwitte@mozilla.com>
244 * msvcc.sh: Build with -W3 instead of -Wall.
245 * src/powerpc/ffi_darwin.c: Remove build warnings.
246 * src/x86/ffi.c: Ditto.
247 * src/x86/ffitarget.h: Ditto.
249 2010-04-12 Dan Witte <dwitte@mozilla.com>
250 Walter Meinl <wuno@lsvw.de>
252 * configure.ac: Add OS/2 support.
253 * configure: Rebuilt.
254 * src/closures.c: Ditto.
255 * src/dlmalloc.c: Ditto.
256 * src/x86/win32.S: Ditto.
258 2010-04-07 Jakub Jelinek <jakub@redhat.com>
260 * testsuite/libffi.call/err_bad_abi.c: Remove unused args variable.
262 2010-04-02 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
264 * Makefile.in: Regenerate.
265 * aclocal.m4: Regenerate.
266 * include/Makefile.in: Regenerate.
267 * man/Makefile.in: Regenerate.
268 * testsuite/Makefile.in: Regenerate.
270 2010-03-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
272 * configure.ac (libffi_cv_as_x86_64_unwind_section_type): New test.
273 * configure: Regenerate.
274 * fficonfig.h.in: Regenerate.
275 * libffi/src/x86/unix64.S (.eh_frame)
276 [HAVE_AS_X86_64_UNWIND_SECTION_TYPE]: Use @unwind section type.
278 2010-03-14 Matthias Klose <doko@ubuntu.com>
280 * src/x86/ffi64.c: Fix typo in comment.
281 * src/x86/ffi.c: Use /* ... */ comment style.
283 2010-02-24 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
285 * doc/libffi.texi (The Closure API): Fix typo.
286 * doc/libffi.info: Remove.
288 2010-02-15 Matthias Klose <doko@ubuntu.com>
290 * src/arm/sysv.S (__ARM_ARCH__): Define for processor
293 2010-01-15 Anthony Green <green@redhat.com>
295 * README: Add notes on building with Microsoft Visual C++.
297 2010-01-15 Daniel Witte <dwitte@mozilla.com>
299 * msvcc.sh: New file.
301 * src/x86/win32.S: Port assembly routines to MSVC and #ifdef.
302 * src/x86/ffi.c: Tweak function declaration and remove excess
304 * include/ffi.h.in: Add __declspec(align(8)) to typedef struct
307 * src/x86/ffi.c: Merge ffi_call_SYSV and ffi_call_STDCALL into new
308 function ffi_call_win32 on X86_WIN32.
309 * src/x86/win32.S (ffi_call_SYSV): Rename to ffi_call_win32.
310 (ffi_call_STDCALL): Remove.
312 * src/prep_cif.c (ffi_prep_cif): Move stack space allocation code
313 to ffi_prep_cif_machdep for x86.
314 * src/x86/ffi.c (ffi_prep_cif_machdep): To here.
316 2010-01-15 Oliver Kiddle <okiddle@yahoo.co.uk>
318 * src/x86/ffitarget.h (ffi_abi): Check for __i386 and __amd64 for
319 Sun Studio compiler compatibility.
321 2010-01-12 Conrad Irwin <conrad.irwin@gmail.com>
323 * doc/libffi.texi: Add closure example.
325 2010-01-07 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
328 * testsuite/libffi.call/ffitest.h [__alpha__ && __osf__] (PRIdLL,
329 PRIuLL, PRId64, PRIu64, PRIuPTR): Define.
330 * testsuite/libffi.call/cls_align_sint64.c: Add -Wno-format on
332 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
333 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
334 * testsuite/libffi.call/return_ll1.c: Likewise.
335 * testsuite/libffi.call/stret_medium2.c: Likewise.
336 * testsuite/libffi.special/ffitestcxx.h (allocate_mmap): Cast
337 MAP_FAILED to char *.
339 2010-01-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
341 * src/mips/n32.S: Use .abicalls and .eh_frame with __GNUC__.
343 2009-12-31 Anthony Green <green@redhat.com>
345 * README: Update for libffi 3.0.9.
347 2009-12-27 Matthias Klose <doko@ubuntu.com>
349 * configure.ac (HAVE_LONG_DOUBLE): Define for mips when
351 * configure: Rebuilt.
353 2009-12-26 Anthony Green <green@redhat.com>
355 * testsuite/libffi.call/cls_longdouble_va.c: Mark as xfail for
357 * testsuite/libffi.call/cls_double_va.c: Ditto.
359 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
361 * testsuite/libffi.call/ffitest.h: Conditionally include stdint.h
363 * testsuite/libffi.special/unwindtest.cc: Ditto.
365 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
367 * configure.ac: Add amd64-*-openbsd*.
368 * configure: Rebuilt.
369 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Link
370 openbsd programs with -lpthread.
372 2009-12-26 Anthony Green <green@redhat.com>
374 * testsuite/libffi.call/cls_double_va.c,
375 testsuite/libffi.call/cls_longdouble.c,
376 testsuite/libffi.call/cls_longdouble_va.c,
377 testsuite/libffi.call/cls_pointer.c,
378 testsuite/libffi.call/cls_pointer_stack.c: Remove xfail for
379 mips*-*-* and arm*-*-*.
380 * testsuite/libffi.call/cls_align_longdouble_split.c,
381 testsuite/libffi.call/cls_align_longdouble_split2.c,
382 testsuite/libffi.call/stret_medium2.c,
383 testsuite/libffi.call/stret_medium.c,
384 testsuite/libffi.call/stret_large.c,
385 testsuite/libffi.call/stret_large2.c: Remove xfail for arm*-*-*.
387 2009-12-31 Kay Tietz <ktietz70@googlemail.com>
389 * testsuite/libffi.call/ffitest.h,
390 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRuLL): Fix
393 2009-12-31 Carlo Bramini <carlo.bramix@libero.it>
395 * configure.ac (AM_LTLDFLAGS): Define for windows hosts.
396 * Makefile.am (libffi_la_LDFLAGS): Add AM_LTLDFLAGS.
397 * configure: Rebuilt.
398 * Makefile.in: Rebuilt.
400 2009-12-31 Anthony Green <green@redhat.com>
403 * testsuite/libffi.call/huge_struct.c: New test case from Blake
406 2009-12-28 David Edelsohn <edelsohn@gnu.org>
408 * src/powerpc/ffi_darwin.c (ffi_prep_args): Copy abi and nargs to
410 (aix_adjust_aggregate_sizes): New function.
411 (ffi_prep_cif_machdep): Call it.
413 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
415 * configure.ac: Define FFI_MMAP_EXEC_WRIT for the given targets.
416 * configure: Regenerate.
417 * fficonfig.h.in: Likewise.
418 * src/closures.c: Remove the FFI_MMAP_EXEC_WRIT definition for
421 2009-12-26 Andreas Schwab <schwab@linux-m68k.org>
423 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Advance intarg_count
424 when a float arguments is passed in memory.
425 (ffi_closure_helper_SYSV): Mark general registers as used up when
426 a 64bit or soft-float long double argument is passed in memory.
428 2009-12-25 Matthias Klose <doko@ubuntu.com>
430 * man/ffi_call.3: Fix #include in examples.
431 * doc/libffi.texi: Add dircategory.
433 2009-12-25 Frank Everdij <f.p.x.everdij@tudelft.nl>
435 * include/ffi.h.in: Placed '__GNUC__' ifdef around
436 '__attribute__((aligned(8)))' in ffi_closure, fixes compile for
438 * include/ffi_common.h: Added '__sgi' define to non
439 '__attribute__((__mode__()))' integer typedefs.
440 * src/mips/ffi.c (ffi_call, ffi_closure_mips_inner_O32,
441 ffi_closure_mips_inner_N32): Added 'defined(_MIPSEB)' to BE check.
442 (ffi_closure_mips_inner_O32, ffi_closure_mips_inner_N32): Added
443 FFI_LONGDOUBLE support and alignment(N32 only).
444 * src/mips/ffitarget.h: Corrected '#include <sgidefs.h>' for IRIX and
445 fixed non '__attribute__((__mode__()))' integer typedefs.
446 * src/mips/n32.S: Put '#ifdef linux' around '.abicalls' and '.eh_frame'
447 since they are Linux/GNU Assembler specific.
449 2009-12-25 Bradley Smith <brad@brad-smith.co.uk>
451 * configure.ac, Makefile.am, src/avr32/ffi.c,
452 src/avr32/ffitarget.h,
453 src/avr32/sysv.S: Add AVR32 port.
454 * configure, Makefile.in: Rebuilt.
456 2009-12-21 Andreas Tobler <a.tobler@schweiz.org>
458 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
459 * configure: Regenerate.
461 2009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
463 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
465 2009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
467 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
470 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
472 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
475 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
477 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
479 2009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
482 * src/closures.c [X86_64 && __sun__ && __svr4__]
483 (FFI_MMAP_EXEC_WRIT): Define.
485 2009-12-08 David Daney <ddaney@caviumnetworks.com>
487 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
488 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
489 * testsuite/libffi.call/stret_large.c: Same.
490 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
491 * testsuite/libffi.call/stret_large2.c: Same.
492 * testsuite/libffi.call/stret_medium2.c: Same.
494 2009-12-07 David Edelsohn <edelsohn@gnu.org>
496 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
499 2009-12-05 David Edelsohn <edelsohn@gnu.org>
501 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
503 * src/powerpc/aix_closure.S: Same.
505 2009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
507 * Makefile.in: Regenerate.
508 * configure: Regenerate.
509 * include/Makefile.in: Regenerate.
510 * man/Makefile.in: Regenerate.
511 * testsuite/Makefile.in: Regenerate.
513 2009-12-04 David Edelsohn <edelsohn@gnu.org>
515 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
518 2009-12-04 Uros Bizjak <ubizjak@gmail.com>
521 * src/x86/ffi64.c (classify_argument): Update from
522 gcc/config/i386/i386.c.
523 (ffi_closure_unix64_inner): Do not use the address of two consecutive
524 SSE registers directly.
525 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
526 for x86_64 linux targets.
528 2009-12-04 David Edelsohn <edelsohn@gnu.org>
530 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
531 pfr for long double split between fpr13 and stack.
533 2009-12-03 David Edelsohn <edelsohn@gnu.org>
535 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
536 fparg_count twice for long double.
538 2009-12-03 David Edelsohn <edelsohn@gnu.org>
541 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
543 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
545 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
546 Remove xfails for x86 linux targets.
548 2009-12-02 David Edelsohn <edelsohn@gnu.org>
550 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
553 2009-12-01 David Edelsohn <edelsohn@gnu.org>
555 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
556 register usage. Call ffi_prep_args directly. Add long double
557 return value support.
558 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
559 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
560 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
561 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
563 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
565 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
566 Allocate result area between params and FPRs.
568 2009-11-30 David Edelsohn <edelsohn@gnu.org>
571 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
573 * src/powerpc/aix.S: Implement AIX64 version.
574 * src/powerpc/aix_closure.S: Implement AIX64 version.
575 (ffi_closure_ASM): Use extsb, lha and displament addresses.
576 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
578 (ffi_prep_cif_machdep): Same.
580 (ffi_closure_helper_DARWIN): Same.
582 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
585 * testsuite/libffi.call/testclosure.c: New test.
587 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
589 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
590 assembly version use of ___chkstk.
592 2009-09-23 Matthias Klose <doko@ubuntu.com>
594 PR libffi/40242, PR libffi/41443
595 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
596 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
597 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
598 Change the conditionals to __SOFTFP__ || __ARM_EABI__
599 for -mfloat-abi=softfp to work.
601 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
603 PR testsuite/32843 (strikes again)
604 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
605 enable proper extension on char and short.
607 2009-09-15 David Daney <ddaney@caviumnetworks.com>
609 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
610 handling for FFI_TYPE_POINTER.
611 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
612 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
613 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
614 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
615 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
616 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
617 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
618 structure and pointer returns.
619 (ffi_closure_N32): Add handling for pointer returns.
620 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
621 calc_n32_return_struct_flags): Handle soft-float.
622 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
623 (ffi_call_N32): Declare proper argument types.
624 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
627 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
629 * configure.ac (AC_PREREQ): Bump to 2.64.
631 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
633 * Makefile.am (install-html, install-pdf): Remove.
634 * Makefile.in: Regenerate.
636 * Makefile.in: Regenerate.
637 * aclocal.m4: Regenerate.
638 * configure: Regenerate.
639 * fficonfig.h.in: Regenerate.
640 * include/Makefile.in: Regenerate.
641 * man/Makefile.in: Regenerate.
642 * testsuite/Makefile.in: Regenerate.
644 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
646 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
648 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
651 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
652 return types for X86_WIN32.
653 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
654 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
655 _ffi_closure_STDCALL): Likewise.
657 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
658 (dlmmap, dlmunmap): Also use these functions on Cygwin.
660 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
665 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
668 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
670 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
671 to "" before adding paths. (This reinstates an assignment that
672 was removed by my 2009-06-30 commit, but changes the initial
673 value from "." to "".)
675 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
678 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
679 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
682 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
684 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
685 to ld_library_path. Use add_path. Add just find_libgcc_s
686 to ld_library_path, not every libgcc multilib directory.
688 2009-06-16 Wim Lewis <wiml@hhhh.org>
690 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
691 supposed to be callee-saved.
692 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
693 return buffer for odd-size structs.
695 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
698 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
699 allow_stack_execute for Darwin.
701 2009-06-16 Andrew Haley <aph@redhat.com>
703 * configure.ac (TARGETDIR): Add missing blank lines.
704 * configure: Regenerate.
706 2009-06-16 Andrew Haley <aph@redhat.com>
708 * testsuite/libffi.call/cls_align_sint64.c,
709 testsuite/libffi.call/cls_align_uint64.c,
710 testsuite/libffi.call/cls_longdouble_va.c,
711 testsuite/libffi.call/cls_ulonglong.c,
712 testsuite/libffi.call/return_ll1.c,
713 testsuite/libffi.call/stret_medium2.c: Fix printf format
715 * testsuite/libffi.call/ffitest.h,
716 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
718 2009-06-15 Andrew Haley <aph@redhat.com>
720 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
721 * testsuite/libffi.call/err_bad_abi.c: Likewise.
723 2009-06-12 Andrew Haley <aph@redhat.com>
725 * Makefile.am: Remove info_TEXINFOS.
727 2009-06-12 Andrew Haley <aph@redhat.com>
729 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
730 testsuite/libffi.call/cls_align_uint64.c,
731 testsuite/libffi.call/cls_ulonglong.c,
732 testsuite/libffi.call/return_ll1.c,
733 testsuite/libffi.call/stret_medium2.c: Fix printf format
735 testsuite/libffi.special/unwindtest.cc: include stdint.h.
737 2009-06-11 Timothy Wall <twall@users.sf.net>
742 include/ffi_common.h,
747 src/x86/win64.S (new),
748 README: Added win64 support (mingw or MSVC)
752 testsuite/Makefile.in,
754 aclocal.m4: Regenerated
755 * ltcf-c.sh: properly escape cygwin/w32 path
756 * man/ffi_call.3: Clarify size requirements for return value.
757 * src/x86/ffi64.c: Fix filename in comment.
758 * src/x86/win32.S: Remove unused extern.
760 * testsuite/libffi.call/closure_fn0.c,
761 testsuite/libffi.call/closure_fn1.c,
762 testsuite/libffi.call/closure_fn2.c,
763 testsuite/libffi.call/closure_fn3.c,
764 testsuite/libffi.call/closure_fn4.c,
765 testsuite/libffi.call/closure_fn5.c,
766 testsuite/libffi.call/closure_fn6.c,
767 testsuite/libffi.call/closure_stdcall.c,
768 testsuite/libffi.call/cls_12byte.c,
769 testsuite/libffi.call/cls_16byte.c,
770 testsuite/libffi.call/cls_18byte.c,
771 testsuite/libffi.call/cls_19byte.c,
772 testsuite/libffi.call/cls_1_1byte.c,
773 testsuite/libffi.call/cls_20byte.c,
774 testsuite/libffi.call/cls_20byte1.c,
775 testsuite/libffi.call/cls_24byte.c,
776 testsuite/libffi.call/cls_2byte.c,
777 testsuite/libffi.call/cls_3_1byte.c,
778 testsuite/libffi.call/cls_3byte1.c,
779 testsuite/libffi.call/cls_3byte2.c,
780 testsuite/libffi.call/cls_4_1byte.c,
781 testsuite/libffi.call/cls_4byte.c,
782 testsuite/libffi.call/cls_5_1_byte.c,
783 testsuite/libffi.call/cls_5byte.c,
784 testsuite/libffi.call/cls_64byte.c,
785 testsuite/libffi.call/cls_6_1_byte.c,
786 testsuite/libffi.call/cls_6byte.c,
787 testsuite/libffi.call/cls_7_1_byte.c,
788 testsuite/libffi.call/cls_7byte.c,
789 testsuite/libffi.call/cls_8byte.c,
790 testsuite/libffi.call/cls_9byte1.c,
791 testsuite/libffi.call/cls_9byte2.c,
792 testsuite/libffi.call/cls_align_double.c,
793 testsuite/libffi.call/cls_align_float.c,
794 testsuite/libffi.call/cls_align_longdouble.c,
795 testsuite/libffi.call/cls_align_longdouble_split.c,
796 testsuite/libffi.call/cls_align_longdouble_split2.c,
797 testsuite/libffi.call/cls_align_pointer.c,
798 testsuite/libffi.call/cls_align_sint16.c,
799 testsuite/libffi.call/cls_align_sint32.c,
800 testsuite/libffi.call/cls_align_sint64.c,
801 testsuite/libffi.call/cls_align_uint16.c,
802 testsuite/libffi.call/cls_align_uint32.c,
803 testsuite/libffi.call/cls_align_uint64.c,
804 testsuite/libffi.call/cls_dbls_struct.c,
805 testsuite/libffi.call/cls_double.c,
806 testsuite/libffi.call/cls_double_va.c,
807 testsuite/libffi.call/cls_float.c,
808 testsuite/libffi.call/cls_longdouble.c,
809 testsuite/libffi.call/cls_longdouble_va.c,
810 testsuite/libffi.call/cls_multi_schar.c,
811 testsuite/libffi.call/cls_multi_sshort.c,
812 testsuite/libffi.call/cls_multi_sshortchar.c,
813 testsuite/libffi.call/cls_multi_uchar.c,
814 testsuite/libffi.call/cls_multi_ushort.c,
815 testsuite/libffi.call/cls_multi_ushortchar.c,
816 testsuite/libffi.call/cls_pointer.c,
817 testsuite/libffi.call/cls_pointer_stack.c,
818 testsuite/libffi.call/cls_schar.c,
819 testsuite/libffi.call/cls_sint.c,
820 testsuite/libffi.call/cls_sshort.c,
821 testsuite/libffi.call/cls_uchar.c,
822 testsuite/libffi.call/cls_uint.c,
823 testsuite/libffi.call/cls_ulonglong.c,
824 testsuite/libffi.call/cls_ushort.c,
825 testsuite/libffi.call/err_bad_abi.c,
826 testsuite/libffi.call/err_bad_typedef.c,
827 testsuite/libffi.call/float2.c,
828 testsuite/libffi.call/huge_struct.c,
829 testsuite/libffi.call/nested_struct.c,
830 testsuite/libffi.call/nested_struct1.c,
831 testsuite/libffi.call/nested_struct10.c,
832 testsuite/libffi.call/nested_struct2.c,
833 testsuite/libffi.call/nested_struct3.c,
834 testsuite/libffi.call/nested_struct4.c,
835 testsuite/libffi.call/nested_struct5.c,
836 testsuite/libffi.call/nested_struct6.c,
837 testsuite/libffi.call/nested_struct7.c,
838 testsuite/libffi.call/nested_struct8.c,
839 testsuite/libffi.call/nested_struct9.c,
840 testsuite/libffi.call/problem1.c,
841 testsuite/libffi.call/return_ldl.c,
842 testsuite/libffi.call/return_ll1.c,
843 testsuite/libffi.call/stret_large.c,
844 testsuite/libffi.call/stret_large2.c,
845 testsuite/libffi.call/stret_medium.c,
846 testsuite/libffi.call/stret_medium2.c,
847 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
848 of checking for MMAP. Use intptr_t instead of long casts.
850 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
852 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
853 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
854 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
856 2009-06-09 Andrew Haley <aph@redhat.com>
858 * src/x86/freebsd.S: Add missing file.
860 2009-06-08 Andrew Haley <aph@redhat.com>
862 Import from libffi 3.0.8:
864 * doc/libffi.texi: New file.
865 * doc/libffi.info: Likewise.
866 * doc/stamp-vti: Likewise.
867 * man/Makefile.am: New file.
868 * man/ffi_call.3: New file.
870 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
872 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
874 * configure.ac: Bump version to 3.0.8.
875 parisc*-*-linux*: Add.
876 i386-*-freebsd* | i386-*-openbsd*: Add.
877 powerpc-*-beos*: Add.
878 AM_CONDITIONAL X86_FREEBSD: Add.
879 AC_CONFIG_FILES: Add man/Makefile.
881 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
883 2009-06-08 Andrew Haley <aph@redhat.com>
885 * README: Import from libffi 3.0.8.
887 2009-06-08 Andrew Haley <aph@redhat.com>
889 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
890 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
891 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
892 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
894 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
895 * testsuite/libffi.call/stret_medium.c: Likewise.
896 * testsuite/libffi.call/stret_large2.c: Likewise.
897 * testsuite/libffi.call/stret_large.c: Likewise.
899 2008-12-26 Timothy Wall <twall@users.sf.net>
901 * testsuite/libffi.call/cls_longdouble.c,
902 testsuite/libffi.call/cls_longdouble_va.c,
903 testsuite/libffi.call/cls_align_longdouble.c,
904 testsuite/libffi.call/cls_align_longdouble_split.c,
905 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
906 failures on x86_64 cygwin/mingw.
908 2008-12-22 Timothy Wall <twall@users.sf.net>
910 * testsuite/libffi.call/closure_fn0.c,
911 testsuite/libffi.call/closure_fn1.c,
912 testsuite/libffi.call/closure_fn2.c,
913 testsuite/libffi.call/closure_fn3.c,
914 testsuite/libffi.call/closure_fn4.c,
915 testsuite/libffi.call/closure_fn5.c,
916 testsuite/libffi.call/closure_fn6.c,
917 testsuite/libffi.call/closure_loc_fn0.c,
918 testsuite/libffi.call/closure_stdcall.c,
919 testsuite/libffi.call/cls_align_pointer.c,
920 testsuite/libffi.call/cls_pointer.c,
921 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
922 pointer to integer (intptr_t).
923 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
925 2008-07-24 Anthony Green <green@redhat.com>
927 * testsuite/libffi.call/cls_dbls_struct.c,
928 testsuite/libffi.call/cls_double_va.c,
929 testsuite/libffi.call/cls_longdouble.c,
930 testsuite/libffi.call/cls_longdouble_va.c,
931 testsuite/libffi.call/cls_pointer.c,
932 testsuite/libffi.call/cls_pointer_stack.c,
933 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
936 2008-03-04 Anthony Green <green@redhat.com>
940 * testsuite/libffi.call/cls_align_longdouble_split2.c
941 testsuite/libffi.call/cls_align_longdouble_split.c
942 testsuite/libffi.call/cls_dbls_struct.c
943 testsuite/libffi.call/cls_double_va.c
944 testsuite/libffi.call/cls_longdouble.c
945 testsuite/libffi.call/cls_longdouble_va.c
946 testsuite/libffi.call/cls_pointer.c
947 testsuite/libffi.call/cls_pointer_stack.c
948 testsuite/libffi.call/err_bad_abi.c
949 testsuite/libffi.call/err_bad_typedef.c
950 testsuite/libffi.call/stret_large2.c
951 testsuite/libffi.call/stret_large.c
952 testsuite/libffi.call/stret_medium2.c
953 testsuite/libffi.call/stret_medium.c: New tests from Apple.
955 2009-06-05 Andrew Haley <aph@redhat.com>
957 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
960 2009-06-04 Andrew Haley <aph@redhat.com>
962 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
965 2008-02-26 Anthony Green <green@redhat.com>
966 Thomas Heller <theller@ctypes.org>
968 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
971 2008-02-03 Timothy Wall <twall@users.sf.net>
973 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
974 offset based on code pointer, not data pointer.
976 2008-01-31 Timothy Wall <twall@users.sf.net>
978 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
980 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
982 * src/x86/win32.S: Add assembly for stdcall closure.
983 * src/x86/ffi.c: Initialize stdcall closure trampoline.
985 2009-06-04 Andrew Haley <aph@redhat.com>
987 * include/ffi.h.in: Change void (*)() to void (*)(void).
988 * src/x86/ffi.c: Likewise.
990 2009-06-04 Andrew Haley <aph@redhat.com>
992 * src/powerpc/ppc_closure.S: Insert licence header.
993 * src/powerpc/linux64_closure.S: Likewise.
994 * src/m68k/sysv.S: Likewise.
996 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
997 * src/powerpc/ffi.c: Likewise.
998 * src/powerpc/ffi_darwin.c: Likewise.
999 * src/m32r/ffi.c: Likewise.
1000 * src/sh64/ffi.c: Likewise.
1001 * src/x86/ffi64.c: Likewise.
1002 * src/alpha/ffi.c: Likewise.
1003 * src/alpha/osf.S: Likewise.
1004 * src/frv/ffi.c: Likewise.
1005 * src/s390/ffi.c: Likewise.
1006 * src/pa/ffi.c: Likewise.
1007 * src/pa/hpux32.S: Likewise.
1008 * src/ia64/unix.S: Likewise.
1009 * src/ia64/ffi.c: Likewise.
1010 * src/sparc/ffi.c: Likewise.
1011 * src/mips/ffi.c: Likewise.
1012 * src/sh/ffi.c: Likewise.
1014 2008-02-15 David Daney <ddaney@avtrex.com>
1016 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
1017 Define (conditionally), and use it to include cachectl.h.
1018 (ffi_prep_closure_loc): Fix cache flushing.
1019 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
1021 2009-06-04 Andrew Haley <aph@redhat.com>
1024 src/arm/ffitarget.h,
1027 src/powerpc/ffitarget.h,
1029 src/sh64/ffitarget.h,
1034 src/x86/ffitarget.h,
1040 src/alpha/ffitarget.h,
1043 src/m68k/ffitarget.h,
1044 src/frv/ffitarget.h,
1046 src/s390/ffitarget.h,
1048 src/cris/ffitarget.h,
1053 src/ia64/ffitarget.h,
1056 src/ia64/ia64_flags.h,
1060 src/sparc/ffitarget.h,
1063 src/mips/ffitarget.h,
1070 src/sh/sysv.S: Update license text.
1072 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
1074 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
1075 (.eh_frame): Add FDE for it.
1077 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
1079 * configure.ac: Also check if assembler supports pc-relative
1080 relocs on X86_WIN32 targets.
1081 * configure: Regenerate.
1082 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
1083 (_ffi_call_SYSV): Add missing function type symbol .def and
1084 add EH markup labels.
1085 (_ffi_call_STDCALL): Likewise.
1086 (_ffi_closure_SYSV): Likewise.
1087 (_ffi_closure_raw_SYSV): Likewise.
1088 (.eh_frame): Add hand-crafted EH data.
1090 2009-04-09 Jakub Jelinek <jakub@redhat.com>
1092 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
1093 version 3 of the GNU General Public License and to point readers
1094 at the COPYING3 file and the FSF's license web page.
1095 * testsuite/libffi.call/call.exp: Likewise.
1096 * testsuite/libffi.special/special.exp: Likewise.
1098 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1100 * configure: Regenerate.
1102 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1105 * configure.ac (HAVE_AS_X86_PCREL): New test.
1106 * configure: Regenerate.
1107 * fficonfig.h.in: Regenerate.
1108 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
1109 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
1110 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
1111 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
1112 * src/x86/unix64.S (.Lstore_table): Move to .text section.
1113 (.Lload_table): Likewise.
1114 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
1116 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1118 * configure: Regenerate.
1120 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
1122 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
1123 signed/unsigned int8/16 return values.
1124 * src/sparc/v8.S (ffi_call_v8): Likewise.
1125 (ffi_closure_v8): Likewise.
1127 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
1128 Steve Ellcey <sje@cup.hp.com>
1130 * configure: Regenerate for new libtool.
1131 * Makefile.in: Ditto.
1132 * include/Makefile.in: Ditto.
1133 * aclocal.m4: Ditto.
1135 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
1137 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
1138 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
1139 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
1140 Adjust copyright notice.
1141 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
1142 register or two register to use for FFI_SYSV structs.
1143 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
1144 (ffi_closure_helper_SYSV): Fix the return type for
1145 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
1146 Adjust copyright notice.
1148 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
1150 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
1153 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1155 * configure: Regenerate.
1156 * include/Makefile.in: Regenerate.
1157 * testsuite/Makefile.in: Regenerate.
1159 2008-06-07 Joseph Myers <joseph@codesourcery.com>
1161 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
1162 powerpc-*-beos*): Remove.
1163 * configure: Regenerate.
1165 2008-05-09 Julian Brown <julian@codesourcery.com>
1167 * Makefile.am (LTLDFLAGS): New.
1168 (libffi_la_LDFLAGS): Use above.
1169 * Makefile.in: Regenerate.
1171 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
1174 * aclocal.m4: Regenerate.
1175 * configure: Regenerate.
1177 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
1179 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
1180 * src/sh64/sysv.S: Likewise.
1182 2008-03-26 Daniel Jacobowitz <dan@debian.org>
1184 * src/arm/sysv.S: Fix ARM comment marker.
1186 2008-03-26 Jakub Jelinek <jakub@redhat.com>
1188 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
1189 * src/s390/sysv.S: Likewise.
1190 * src/powerpc/ppc_closure.S: Likewise.
1191 * src/powerpc/sysv.S: Likewise.
1192 * src/x86/unix64.S: Likewise.
1193 * src/x86/sysv.S: Likewise.
1194 * src/sparc/v8.S: Likewise.
1195 * src/sparc/v9.S: Likewise.
1196 * src/m68k/sysv.S: Likewise.
1197 * src/arm/sysv.S: Likewise.
1199 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
1201 * aclocal.m4: Regenerate.
1202 * configure: Likewise.
1203 * Makefile.in: Likewise.
1204 * include/Makefile.in: Likewise.
1205 * testsuite/Makefile.in: Likewise.
1207 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
1208 Andreas Tobler <a.tobler@schweiz.org>
1210 * configure.ac: Add amd64-*-freebsd* target.
1211 * configure: Regenerate.
1213 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
1216 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
1219 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
1222 2008-01-24 David Edelsohn <edelsohn@gnu.org>
1224 * configure: Regenerate.
1226 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
1228 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
1230 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
1233 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
1234 signed/unsigned int8/16 for X86_DARWIN.
1235 Updated copyright info.
1236 Handle one and two byte structs with special cif->flags.
1237 * src/x86/ffitarget.h: Add special types for one and two byte structs.
1238 Updated copyright info.
1239 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
1241 Remove code to pop args from the stack after call.
1242 Special-case signed/unsigned for int8/16, one and two byte structs.
1243 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1244 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1246 Updated copyright info.
1248 2007-12-08 David Daney <ddaney@avtrex.com>
1250 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
1251 SUBU, add with ADDU and use smaller code sequences.
1253 2007-12-07 David Daney <ddaney@avtrex.com>
1255 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
1258 2007-12-06 David Daney <ddaney@avtrex.com>
1260 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
1262 (ffi_java_raw): New typedef.
1263 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
1264 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
1266 (ffi_java_raw_closure) : Same.
1267 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
1269 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
1270 FFI_SIZEOF_JAVA_RAW.
1271 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
1272 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
1273 sizeof(ffi_java_raw) for alignment calculations.
1274 (ffi_java_ptrarray_to_raw): Same.
1275 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
1276 if FFI_SIZEOF_JAVA_RAW == 4.
1277 (ffi_java_raw_to_rvalue): Same.
1278 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
1279 (ffi_java_translate_args): Same.
1280 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
1282 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
1284 2007-12-06 David Daney <ddaney@avtrex.com>
1286 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
1289 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
1292 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
1293 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
1294 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
1295 set the NUM_FPR_ARG_REGISTERS according to.
1296 Add support for potential soft-float support under hard-float
1298 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
1299 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
1300 to the FFI_LINUX_SOFT_FLOAT ABI.
1301 (ffi_prep_cif_machdep): Likewise.
1302 (ffi_closure_helper_SYSV): Likewise.
1303 * src/powerpc/ppc_closure.S: Make sure not to store float/double
1304 on archs where __NO_FPRS__ is true.
1305 Add FFI_TYPE_UINT128 support.
1306 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
1307 Adjust copyright notice.
1309 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
1311 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
1312 * include/ffi_common.h: ... here.
1315 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
1317 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
1318 * src/powerpc/linux64.S: Likewise.
1319 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
1320 * testsuite/libffi.call/return_ldl.c: New test.
1322 2007-09-04 <aph@redhat.com>
1324 * src/arm/sysv.S (UNWIND): New.
1325 (Whole file): Conditionally compile unwinder directives.
1326 * src/arm/sysv.S: Add unwinder directives.
1328 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
1329 Only treat r0 as a struct address if we're actually returning a
1331 Only copy the bytes that are actually within a struct.
1332 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
1333 is returned in r0, not passed by address.
1334 (ffi_call): Allocate a word-sized temporary for the case where
1335 a composite is returned in r0.
1336 (ffi_prep_incoming_args_SYSV): Align as necessary.
1338 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
1340 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
1341 directly using the sys_cacheflush syscall.
1343 2007-07-27 Andrew Haley <aph@redhat.com>
1345 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
1347 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
1349 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
1350 * configure.ac: Likewise.
1351 * Makefile.in: Regenerate.
1352 * include/Makefile.in: Likewise.
1353 * testsuite/Makefile.in: Likewise.
1354 * configure: Likewise.
1356 2007-08-24 David Daney <ddaney@avtrex.com>
1358 * testsuite/libffi.call/return_sl.c: New test.
1360 2007-08-10 David Daney <ddaney@avtrex.com>
1362 * testsuite/libffi.call/cls_multi_ushort.c,
1363 testsuite/libffi.call/cls_align_uint16.c,
1364 testsuite/libffi.call/nested_struct1.c,
1365 testsuite/libffi.call/nested_struct3.c,
1366 testsuite/libffi.call/cls_7_1_byte.c,
1367 testsuite/libffi.call/nested_struct5.c,
1368 testsuite/libffi.call/cls_double.c,
1369 testsuite/libffi.call/nested_struct7.c,
1370 testsuite/libffi.call/cls_sint.c,
1371 testsuite/libffi.call/nested_struct9.c,
1372 testsuite/libffi.call/cls_20byte1.c,
1373 testsuite/libffi.call/cls_multi_sshortchar.c,
1374 testsuite/libffi.call/cls_align_sint64.c,
1375 testsuite/libffi.call/cls_3byte2.c,
1376 testsuite/libffi.call/cls_multi_schar.c,
1377 testsuite/libffi.call/cls_multi_uchar.c,
1378 testsuite/libffi.call/cls_19byte.c,
1379 testsuite/libffi.call/cls_9byte1.c,
1380 testsuite/libffi.call/cls_align_float.c,
1381 testsuite/libffi.call/closure_fn1.c,
1382 testsuite/libffi.call/problem1.c,
1383 testsuite/libffi.call/closure_fn3.c,
1384 testsuite/libffi.call/cls_sshort.c,
1385 testsuite/libffi.call/closure_fn5.c,
1386 testsuite/libffi.call/cls_align_double.c,
1387 testsuite/libffi.call/nested_struct.c,
1388 testsuite/libffi.call/cls_2byte.c,
1389 testsuite/libffi.call/nested_struct10.c,
1390 testsuite/libffi.call/cls_4byte.c,
1391 testsuite/libffi.call/cls_6byte.c,
1392 testsuite/libffi.call/cls_8byte.c,
1393 testsuite/libffi.call/cls_multi_sshort.c,
1394 testsuite/libffi.call/cls_align_sint16.c,
1395 testsuite/libffi.call/cls_align_uint32.c,
1396 testsuite/libffi.call/cls_20byte.c,
1397 testsuite/libffi.call/cls_float.c,
1398 testsuite/libffi.call/nested_struct2.c,
1399 testsuite/libffi.call/cls_5_1_byte.c,
1400 testsuite/libffi.call/nested_struct4.c,
1401 testsuite/libffi.call/cls_24byte.c,
1402 testsuite/libffi.call/nested_struct6.c,
1403 testsuite/libffi.call/cls_64byte.c,
1404 testsuite/libffi.call/nested_struct8.c,
1405 testsuite/libffi.call/cls_uint.c,
1406 testsuite/libffi.call/cls_multi_ushortchar.c,
1407 testsuite/libffi.call/cls_schar.c,
1408 testsuite/libffi.call/cls_uchar.c,
1409 testsuite/libffi.call/cls_align_uint64.c,
1410 testsuite/libffi.call/cls_ulonglong.c,
1411 testsuite/libffi.call/cls_align_longdouble.c,
1412 testsuite/libffi.call/cls_1_1byte.c,
1413 testsuite/libffi.call/cls_12byte.c,
1414 testsuite/libffi.call/cls_3_1byte.c,
1415 testsuite/libffi.call/cls_3byte1.c,
1416 testsuite/libffi.call/cls_4_1byte.c,
1417 testsuite/libffi.call/cls_6_1_byte.c,
1418 testsuite/libffi.call/cls_16byte.c,
1419 testsuite/libffi.call/cls_18byte.c,
1420 testsuite/libffi.call/closure_fn0.c,
1421 testsuite/libffi.call/cls_9byte2.c,
1422 testsuite/libffi.call/closure_fn2.c,
1423 testsuite/libffi.call/closure_fn4.c,
1424 testsuite/libffi.call/cls_ushort.c,
1425 testsuite/libffi.call/closure_fn6.c,
1426 testsuite/libffi.call/cls_5byte.c,
1427 testsuite/libffi.call/cls_align_pointer.c,
1428 testsuite/libffi.call/cls_7byte.c,
1429 testsuite/libffi.call/cls_align_sint32.c,
1430 testsuite/libffi.special/unwindtest_ffi_call.cc,
1431 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
1433 2007-08-10 David Daney <ddaney@avtrex.com>
1436 * configure.ac: Don't treat mips64 as a special case.
1437 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
1438 * configure: Regenerate
1439 * Makefile.in: Ditto.
1440 * fficonfig.h.in: Ditto.
1441 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
1442 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
1443 (FFI_DEFAULT_ABI): Set for n64 case.
1444 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
1445 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
1446 (ffi_closure_N32): New function.
1447 (.eh_frame): New section
1448 * src/mips/o32.S: Clean up comments.
1449 (ffi_closure_O32): Pass ffi_closure parameter in $12.
1450 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
1451 _MIPS_SIM == _ABIN32 throughout.
1452 (FFI_MIPS_STOP_HERE): New, use in place of
1454 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1455 to support n32/n64 ABIs.
1456 (calc_n32_struct_flags): Rewrite.
1457 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1458 position of flag bits.
1459 (ffi_prep_cif_machdep): Rewrite n32 portion.
1460 (ffi_call): Enable for n64. Add special handling for small structure
1462 (ffi_prep_closure_loc): Add n32 and n64 support.
1463 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1464 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1466 2007-08-08 David Daney <ddaney@avtrex.com>
1468 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1469 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1471 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1472 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1473 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1474 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1475 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1476 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1477 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1478 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1479 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1480 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1481 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1482 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1483 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1484 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1485 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1486 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1487 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1488 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1489 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1490 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1491 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1492 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1493 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1494 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1495 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1496 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1497 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1498 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1499 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1500 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1501 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1503 2007-08-07 Andrew Haley <aph@redhat.com>
1505 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1508 2007-08-06 Andrew Haley <aph@redhat.com>
1511 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1512 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1515 2007-08-02 David Daney <ddaney@avtrex.com>
1517 * testsuite/libffi.call/return_ul.c (main): Define return type as
1518 ffi_arg. Use proper printf conversion specifier.
1520 2007-07-30 Andrew Haley <aph@redhat.com>
1523 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1524 signed/unsigned int8/16.
1525 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1527 Remove code to pop args from the stack after call.
1528 Special-case signed/unsigned int8/16.
1529 * testsuite/libffi.call/return_sc.c (main): Revert.
1531 2007-07-26 Richard Guenther <rguenther@suse.de>
1534 * testsuite/libffi.call/return_sc.c (main): Verify call
1535 result as signed char, not ffi_arg.
1537 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1539 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1540 * configure: Regenerate.
1542 2007-07-11 David Daney <ddaney@avtrex.com>
1544 * src/mips/ffi.c: Don't include sys/cachectl.h.
1545 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1548 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1550 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1551 from (ffi_prep_closure): ... this.
1552 (FFI_INIT_TRAMPOLINE): Adjust.
1554 2005-12-31 Phil Blundell <pb@reciva.com>
1556 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1557 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1558 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1559 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1560 (FFI_CLOSURES): Enable closure support.
1562 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1564 * testsuite/libffi.call/cls_multi_ushort.c,
1565 testsuite/libffi.call/cls_align_uint16.c,
1566 testsuite/libffi.call/nested_struct1.c,
1567 testsuite/libffi.call/nested_struct3.c,
1568 testsuite/libffi.call/cls_7_1_byte.c,
1569 testsuite/libffi.call/cls_double.c,
1570 testsuite/libffi.call/nested_struct5.c,
1571 testsuite/libffi.call/nested_struct7.c,
1572 testsuite/libffi.call/cls_sint.c,
1573 testsuite/libffi.call/nested_struct9.c,
1574 testsuite/libffi.call/cls_20byte1.c,
1575 testsuite/libffi.call/cls_multi_sshortchar.c,
1576 testsuite/libffi.call/cls_align_sint64.c,
1577 testsuite/libffi.call/cls_3byte2.c,
1578 testsuite/libffi.call/cls_multi_schar.c,
1579 testsuite/libffi.call/cls_multi_uchar.c,
1580 testsuite/libffi.call/cls_19byte.c,
1581 testsuite/libffi.call/cls_9byte1.c,
1582 testsuite/libffi.call/cls_align_float.c,
1583 testsuite/libffi.call/closure_fn1.c,
1584 testsuite/libffi.call/problem1.c,
1585 testsuite/libffi.call/closure_fn3.c,
1586 testsuite/libffi.call/cls_sshort.c,
1587 testsuite/libffi.call/closure_fn5.c,
1588 testsuite/libffi.call/cls_align_double.c,
1589 testsuite/libffi.call/cls_2byte.c,
1590 testsuite/libffi.call/nested_struct.c,
1591 testsuite/libffi.call/nested_struct10.c,
1592 testsuite/libffi.call/cls_4byte.c,
1593 testsuite/libffi.call/cls_6byte.c,
1594 testsuite/libffi.call/cls_8byte.c,
1595 testsuite/libffi.call/cls_multi_sshort.c,
1596 testsuite/libffi.call/cls_align_uint32.c,
1597 testsuite/libffi.call/cls_align_sint16.c,
1598 testsuite/libffi.call/cls_float.c,
1599 testsuite/libffi.call/cls_20byte.c,
1600 testsuite/libffi.call/cls_5_1_byte.c,
1601 testsuite/libffi.call/nested_struct2.c,
1602 testsuite/libffi.call/cls_24byte.c,
1603 testsuite/libffi.call/nested_struct4.c,
1604 testsuite/libffi.call/nested_struct6.c,
1605 testsuite/libffi.call/cls_64byte.c,
1606 testsuite/libffi.call/nested_struct8.c,
1607 testsuite/libffi.call/cls_uint.c,
1608 testsuite/libffi.call/cls_multi_ushortchar.c,
1609 testsuite/libffi.call/cls_schar.c,
1610 testsuite/libffi.call/cls_uchar.c,
1611 testsuite/libffi.call/cls_align_uint64.c,
1612 testsuite/libffi.call/cls_ulonglong.c,
1613 testsuite/libffi.call/cls_align_longdouble.c,
1614 testsuite/libffi.call/cls_1_1byte.c,
1615 testsuite/libffi.call/cls_12byte.c,
1616 testsuite/libffi.call/cls_3_1byte.c,
1617 testsuite/libffi.call/cls_3byte1.c,
1618 testsuite/libffi.call/cls_4_1byte.c,
1619 testsuite/libffi.call/cls_6_1_byte.c,
1620 testsuite/libffi.call/cls_16byte.c,
1621 testsuite/libffi.call/cls_18byte.c,
1622 testsuite/libffi.call/closure_fn0.c,
1623 testsuite/libffi.call/cls_9byte2.c,
1624 testsuite/libffi.call/closure_fn2.c,
1625 testsuite/libffi.call/closure_fn4.c,
1626 testsuite/libffi.call/cls_ushort.c,
1627 testsuite/libffi.call/closure_fn6.c,
1628 testsuite/libffi.call/cls_5byte.c,
1629 testsuite/libffi.call/cls_align_pointer.c,
1630 testsuite/libffi.call/cls_7byte.c,
1631 testsuite/libffi.call/cls_align_sint32.c,
1632 testsuite/libffi.special/unwindtest_ffi_call.cc,
1633 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1635 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1637 * aclocal.m4: Regenerated.
1639 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1641 * configure: Regenerate.
1643 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1645 * Makefile.in: Regenerate.
1646 * configure: Regenerate.
1647 * aclocal.m4: Regenerate.
1648 * include/Makefile.in: Regenerate.
1649 * testsuite/Makefile.in: Regenerate.
1651 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1653 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1654 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1655 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1656 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1657 (FFI_CLOSURES): Enable closure support.
1659 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1661 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1662 * configure: Regenerate.
1663 * fficonfig.h.in: Regenerate.
1664 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1665 CFI_OFFSET,CFI_DEF_CFA): New macros.
1666 (ffi_call_SYSV): Add callframe annotation.
1668 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1670 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1671 numerous test suite failures.
1672 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1674 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1676 * Makefile.am (EXTRA_DIST): Bring up to date.
1677 * Makefile.in: Regenerate.
1678 * src/frv/eabi.S: Remove RCS keyword.
1680 2007-04-06 Richard Henderson <rth@redhat.com>
1682 * configure.ac: Tidy target case.
1683 (HAVE_LONG_DOUBLE): Allow the target to override.
1684 * configure: Regenerate.
1685 * include/ffi.h.in: Don't define ffi_type_foo if
1686 LIBFFI_HIDE_BASIC_TYPES is defined.
1687 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1689 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1690 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1691 (ffi_type_longdouble): Special case for Alpha. Don't define
1692 if long double == double.
1694 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1695 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1696 (ffi_call, ffi_closure_osf_inner): Likewise.
1697 (ffi_closure_osf_inner): Likewise. Mark hidden.
1698 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1699 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1700 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1701 (load_table): Handle 128-bit long double.
1703 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1705 2007-04-06 Tom Tromey <tromey@redhat.com>
1708 * README: Fixed bug in example.
1710 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1712 * src/closures.c: Include sys/statfs.h.
1713 (_GNU_SOURCE): Define on Linux.
1714 (FFI_MMAP_EXEC_SELINUX): Define.
1715 (selinux_enabled): New variable.
1716 (selinux_enabled_check): New function.
1717 (is_selinux_enabled): Define.
1720 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1722 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1723 Use 'volatile float sum' to create sum of floats to avoid false
1724 negative due to excess precision on ix86 targets.
1727 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1729 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1731 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1733 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1735 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1736 (ffi_prep_closure_loc): New.
1737 (ffi_prep_raw_closure_loc): New.
1738 (ffi_prep_java_raw_closure_loc): New.
1739 * src/closures.c: New file.
1740 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1741 Replace sflags with exec_offset.
1742 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1743 sub_segment_exec_offset): New macros.
1744 (get_segment_flags, set_segment_flags, check_segment_merge): New
1746 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1747 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1748 destroy_mspace): Use new macros.
1749 (sys_alloc): Silence warning.
1750 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1751 * Makefile.in: Rebuilt.
1752 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1753 terms of ffi_prep_closure_loc.
1754 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1756 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1758 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1760 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1761 the renamed version.
1762 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1763 (ffi_prep_closure): ... this.
1764 * src/pa/ffi.c: Likewise.
1765 * src/cris/ffi.c: Likewise. Adjust.
1766 * src/frv/ffi.c: Likewise.
1767 * src/ia64/ffi.c: Likewise.
1768 * src/mips/ffi.c: Likewise.
1769 * src/powerpc/ffi_darwin.c: Likewise.
1770 * src/s390/ffi.c: Likewise.
1771 * src/sh/ffi.c: Likewise.
1772 * src/sh64/ffi.c: Likewise.
1773 * src/sparc/ffi.c: Likewise.
1774 * src/x86/ffi64.c: Likewise.
1775 * src/x86/ffi.c: Likewise.
1776 (FFI_INIT_TRAMPOLINE): Adjust.
1777 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1778 (ffi_prep_raw_closure): ... this.
1779 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1780 (ffi_prep_closure): ... this.
1781 (flush_icache): Adjust.
1783 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1785 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1788 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1790 * Makefile.am: Add dummy install-pdf target.
1791 * Makefile.in: Regenerate
1793 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1795 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1796 ffi_closure_helper_SYSV): Add long double handling.
1798 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1800 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1801 immediately after bctrl instruction.
1803 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1805 * Makefile.am (all-recursive, install-recursive,
1806 mostlyclean-recursive, clean-recursive, distclean-recursive,
1807 maintainer-clean-recursive): Add missing targets.
1808 * Makefile.in: Rebuilt.
1810 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1812 * configure.ac: Add TARGET for x86_64-*-darwin*.
1813 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1815 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1816 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1817 * configure: Regenerate.
1818 * Makefile.in: Regenerate.
1819 * include/Makefile.in: Regenerate.
1820 * testsuite/Makefile.in: Regenerate.
1821 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1824 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1826 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1829 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1831 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1832 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1834 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1835 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1837 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1838 Sandro Tolaini <tolaini@libero.it>
1840 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1842 * configure: Regenerated.
1843 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1844 (EXTRA_DIST): Add src/x86/darwin.S.
1845 * Makefile.in: Regenerated.
1846 * include/Makefile.in: Regenerated.
1847 * testsuite/Makefile.in: Regenerated.
1849 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1850 X86_WIN32, and additionally align stack to 16 bytes.
1851 * src/x86/darwin.S: New, based on sysv.S.
1852 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1854 2006-09-12 David Daney <ddaney@avtrex.com>
1857 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1858 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1859 * aclocal.m4: Regenerated for automake 1.9.6.
1860 * Makefile.in: Regenerated.
1861 * include/Makefile.in: Regenerated.
1862 * testsuite/Makefile.in: Regenerated.
1864 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1866 * include/ffi_common.h (struct): Revert accidental commit.
1868 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1870 * include/ffi_common.h: Remove lint directives.
1871 * include/ffi.h.in: Likewise.
1873 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1875 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1876 for 32-bit architectures.
1877 * testsuite/libffi.call/return_ul.c: New test case.
1879 2006-07-19 David Daney <ddaney@avtrex.com>
1881 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1882 xfail remains for mips64.
1884 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1886 * Makefile.am: Add install-html target. Add install-html to .PHONY
1887 * Makefile.in: Regenerate.
1888 * aclocal.m4: Regenerate.
1889 * include/Makefile.in: Regenerate.
1890 * testsuite/Makefile.in: Regenerate.
1892 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1894 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1897 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1899 * README: Remove notice about 'Crazy Comments'.
1900 * src/debug.c: Remove lint directives. Cleanup white spaces.
1901 * src/java_raw_api.c: Likewise.
1902 * src/prep_cif.c: Likewise.
1903 * src/raw_api.c: Likewise.
1904 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1906 * src/arm/ffi.c: Remove lint directives.
1907 * src/m32r/ffi.c: Likewise.
1908 * src/pa/ffi.c: Likewise.
1909 * src/powerpc/ffi.c: Likewise.
1910 * src/powerpc/ffi_darwin.c: Likewise.
1911 * src/sh/ffi.c: Likewise.
1912 * src/sh64/ffi.c: Likewise.
1913 * src/x86/ffi.c: Likewise.
1914 * testsuite/libffi.call/float2.c: Likewise.
1915 * testsuite/libffi.call/promotion.c: Likewise.
1916 * testsuite/libffi.call/struct1.c: Likewise.
1918 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1920 * src/pa/hpux32.S: Correct unwind offset calculation for
1922 * src/pa/linux.S: Likewise.
1924 2006-04-12 James E Wilson <wilson@specifix.com>
1927 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1928 (hfa_type_load): Call stf_spill.
1929 (hfa_type_store): Call ldf_fill.
1930 (ffi_call): Adjust calls to above routines. Add local temps for
1933 2006-04-10 Matthias Klose <doko@debian.org>
1935 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1936 directory names containing underscores.
1938 2006-04-07 James E Wilson <wilson@specifix.com>
1940 * testsuite/libffi.call/float4.c: New testcase.
1942 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1943 Andreas Tobler <a.tobler@schweiz.ch>
1945 * Makefile.am: Add PA_HPUX port.
1946 * Makefile.in: Regenerate.
1947 * include/Makefile.in: Likewise.
1948 * testsuite/Makefile.in: Likewise.
1949 * configure.ac: Add PA_HPUX rules.
1950 * configure: Regenerate.
1951 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1952 Add PA_HPUX and PA64_HPUX.
1953 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1954 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1955 (FFI_TYPE_SMALL_STRUCT2): Define.
1956 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1957 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1958 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1959 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1960 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1961 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1962 * src/pa/ffi.c (ROUND_DOWN): Delete.
1963 (fldw, fstw, fldd, fstd): Use '__asm__'.
1964 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1965 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1966 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1967 Simplify incrementing of stack slot variable. Change type of local
1968 'n' to unsigned int.
1969 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1971 (ffi_prep_cif_machdep): Likewise.
1972 (ffi_call): Likewise.
1973 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1974 return type to ffi_status. Simplify incrementing of stack slot
1975 variable. Only copy floating point argument registers when PA_LINUX
1976 is true. Reformat debug statement.
1977 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1978 FFI_TYPE_SMALL_STRUCT8.
1979 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1981 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1982 Add nops to cache flush. Add trampoline for PA_HPUX.
1983 * src/pa/hpux32.S: New file.
1984 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1985 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1986 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1987 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1988 argument type checks so that common argument types appear first.
1989 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1990 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1992 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1994 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1995 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1996 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1997 FFI_TYPE_LONGDOUBLE.
1998 (ffi_prep_args64): Assert using IBM extended double.
1999 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
2000 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
2001 (ffi_call): Handle FFI_LINUX.
2002 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
2003 gpr3 return pointer as for struct return. Handle FFI_LINUX
2004 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
2006 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
2007 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
2008 Don't use r6 as pointer to results, instead use sp offset. Don't
2009 make a special call to load lr with case table address, instead
2010 use offset from previous call.
2011 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
2012 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
2015 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
2017 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
2018 passed with FP registers correctly.
2019 (ffi_closure_helper_SYSV): Likewise.
2020 * src/sh64/sysv.S: Likewise.
2022 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
2024 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
2025 args and userdata unused.
2026 (closure_test_fn1): Mark cif and userdata unused.
2027 (main): Remove unused res.
2029 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
2031 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
2032 -O2, -O3, -Os and the warning flags -W -Wall.
2033 * testsuite/libffi.special/special.exp: Likewise.
2034 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
2035 unused parameter unused for gcc or else do nothing.
2036 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2037 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
2038 and userdata unused.
2039 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
2040 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
2041 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
2042 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
2043 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
2044 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
2045 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
2046 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
2047 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
2048 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
2049 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
2050 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
2051 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
2052 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
2053 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
2054 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
2055 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
2056 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
2057 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
2058 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
2059 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
2060 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
2061 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
2062 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
2064 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
2066 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
2068 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
2069 void* to avoid compiler warning.
2071 (cls_struct_align_gn): Mark cif and userdata unused.
2072 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
2074 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
2076 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
2078 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
2080 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
2082 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
2083 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
2084 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
2086 (main): Cast res_call to silence gcc.
2087 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
2089 (main): Cast res_call to silence gcc.
2090 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
2092 (main): Cast res_call to silence gcc.
2093 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
2095 (main): Cast res_call to silence gcc.
2096 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
2098 (main): Cast res_call to silence gcc.
2099 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
2101 (main): Cast res_call to silence gcc.
2102 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
2104 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
2105 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
2107 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
2108 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
2110 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
2111 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
2113 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
2114 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
2116 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
2117 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
2118 and userdata unused.
2119 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
2121 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
2122 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
2123 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
2124 Cleanup white spaces.
2125 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
2126 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
2127 cif and userdata unused.
2128 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
2130 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
2131 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
2132 formatters to silence gcc.
2133 (B_gn): Mark cif and userdata unused.
2134 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
2136 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
2137 (B_gn): Mark cif and userdata unused.
2138 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
2140 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
2141 (B_gn): Mark cif and userdata unused.
2142 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
2144 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
2145 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
2146 * testsuite/libffi.call/problem1.c (stub): Likewise.
2147 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
2149 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
2150 in the last commit for this test case in the test case itself.
2151 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
2153 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
2154 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
2155 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
2156 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
2157 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
2158 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
2160 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
2162 * src/sh/sysv.S: Fix register numbers in the FDE for
2165 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
2167 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
2168 declaration to avoid a false negative on ix86. See PR323.
2170 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
2172 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
2173 and cast integer to void * if needed. Update the pointer to
2174 the FP register saved area correctly.
2176 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
2178 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
2180 * testsuite/libffi.call/nested_struct4.c: Likewise.
2182 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
2184 * testsuite/libffi.call/return_dbl.c: New test case.
2185 * testsuite/libffi.call/return_dbl1.c: Likewise.
2186 * testsuite/libffi.call/return_dbl2.c: Likewise.
2187 * testsuite/libffi.call/return_fl.c: Likewise.
2188 * testsuite/libffi.call/return_fl1.c: Likewise.
2189 * testsuite/libffi.call/return_fl2.c: Likewise.
2190 * testsuite/libffi.call/return_fl3.c: Likewise.
2191 * testsuite/libffi.call/closure_fn6.c: Likewise.
2193 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
2195 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
2196 here to be used by other test cases too.
2198 * testsuite/libffi.call/nested_struct10.c: New test case.
2199 * testsuite/libffi.call/nested_struct9.c: Likewise.
2200 * testsuite/libffi.call/nested_struct8.c: Likewise.
2201 * testsuite/libffi.call/nested_struct7.c: Likewise.
2202 * testsuite/libffi.call/nested_struct6.c: Likewise.
2203 * testsuite/libffi.call/nested_struct5.c: Likewise.
2204 * testsuite/libffi.call/nested_struct4.c: Likewise.
2206 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2208 * configure.ac: Enable libffi for sparc64-*-freebsd*.
2209 * configure: Rebuilt.
2211 2006-01-18 Jakub Jelinek <jakub@redhat.com>
2213 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
2214 instead do the shifting inline.
2215 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
2216 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
2217 and 8 byte structs, for the remaining struct sizes don't call
2218 __lshrdi3, instead do the shifting inline.
2220 2005-12-07 Thiemo Seufer <ths@networkno.de>
2222 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
2223 missing parentheses.
2224 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
2225 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
2226 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
2227 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
2228 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
2230 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
2232 (ffi_prep_closure): Improve trampoline instruction scheduling.
2233 (ffi_closure_mips_inner_O32): Fix endianness bugs.
2235 2005-12-03 Alan Modra <amodra@bigpond.net.au>
2237 * src/powerpc/ffi.c: Formatting.
2238 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
2239 (ffi_prep_args64): Likewise.
2241 2005-09-30 Geoffrey Keating <geoffk@apple.com>
2243 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
2244 darwin, use -shared-libgcc not -lgcc_s, and explain why.
2246 2005-09-26 Tom Tromey <tromey@redhat.com>
2248 * testsuite/libffi.call/float1.c (value_type): New typedef.
2249 (CANARY): New define.
2250 (main): Check for result buffer overflow.
2251 * src/powerpc/linux64.S: Handle linux64 long double returns.
2252 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
2253 (ffi_prep_cif_machdep): Handle linux64 long double returns.
2255 2005-08-25 Alan Modra <amodra@bigpond.net.au>
2258 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
2260 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
2262 2005-08-11 Jakub Jelinek <jakub@redhat.com>
2264 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
2265 (AH_BOTTOM): Add FFI_HIDDEN definition.
2266 * configure: Rebuilt.
2267 * fficonfig.h.in: Rebuilt.
2268 * src/powerpc/ffi.c (hidden): Remove.
2269 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
2270 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
2271 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
2272 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
2273 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
2274 add FFI_HIDDEN to its prototype.
2275 (ffi_closure_SYSV_inner): New.
2276 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2277 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2279 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
2282 * configure: Rebuilt.
2283 * configure.ac: Handle i*86-*-gnu*.
2285 2005-08-09 Jakub Jelinek <jakub@redhat.com>
2287 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
2288 DW_CFA_offset_extended_sf rather than
2289 DW_CFA_GNU_negative_offset_extended.
2290 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
2292 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
2294 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
2296 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
2297 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
2298 partially on register.
2299 (ffi_closure_helper_SYSV): Likewise.
2300 (ffi_prep_cif_machdep): Don't set too many cif->flags.
2302 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
2304 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
2306 * src/sh64/ffi.c (simple_type): Remove.
2307 (return_type): Handle small structures correctly.
2308 (ffi_prep_args): Likewise.
2309 (ffi_call): Likewise.
2310 (ffi_closure_helper_SYSV): Likewise.
2311 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
2312 Emit position independent code if PIC and remove wrong datalabel
2313 prefixes from EH data.
2315 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
2317 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
2318 * Makefile.in: Regenerate.
2319 * include/Makefile.in: Likewise.
2320 * testsuite/Makefile.in: Likewise.
2321 * configure.ac: Add POWERPC_FREEBSD rules.
2322 * configure: Regenerate.
2323 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
2324 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
2325 * src/powerpc/ffi.c: Add flags to handle small structure returns
2327 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
2329 (ffi_closure_helper_SYSV): Likewise.
2330 * src/powerpc/ppc_closure.S: Add return types for small structures.
2331 * src/powerpc/sysv.S: Add bits to handle small structures for
2334 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
2336 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
2337 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
2338 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
2340 2005-07-05 Randolph Chung <tausq@debian.org>
2342 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
2343 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
2344 structures. Kill compilation warnings.
2345 (ffi_closure_inner_LINUX): Print return values as hex in debug
2346 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
2347 Properly handle 5-7 byte structure returns.
2348 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
2349 (FFI_TYPE_SMALL_STRUCT2): Remove.
2350 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
2351 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
2352 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
2353 (checksmst1, checksmst2): Remove.
2354 (checksmst3): Optimize handling of 3-byte struct returns.
2355 (checksmst567): Properly handle 5-7 byte struct returns.
2357 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2360 * src/mips/n32.S: Enforce PIC code.
2361 * src/mips/o32.S: Likewise.
2363 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2365 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
2366 * configure: Regenerate.
2368 2005-06-01 Alan Modra <amodra@bigpond.net.au>
2370 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
2371 to call ffi_closure_helper_SYSV. Append @local instead.
2372 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
2374 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
2376 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
2377 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
2378 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
2379 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
2380 include/Makefile.in, testsuite/Makefile.in: Regenerate.
2382 2005-05-09 Mike Stump <mrs@apple.com>
2384 * configure: Regenerate.
2386 2005-05-08 Richard Henderson <rth@redhat.com>
2389 * src/alpha/osf.S: Update unwind into to match code.
2391 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
2392 Richard Henderson <rth@redhat.com>
2394 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
2396 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
2397 (ffi_prep_closure): Set carry bit if sse-used flag set.
2398 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
2399 Only load sse registers if ssecount non-zero.
2400 (ffi_closure_unix64): Only save sse registers if carry set on entry.
2402 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
2404 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
2405 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
2406 * configure: Regenerate.
2408 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
2410 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
2411 have Tcl8.3-compatible intermediate variable.
2413 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
2414 Hans-Peter Nilsson <hp@axis.com>
2416 * Makefile.am: Add CRIS support.
2417 * configure.ac: Likewise.
2418 * Makefile.in, configure, testsuite/Makefile.in,
2419 include/Makefile.in: Regenerate.
2420 * src/cris: New directory.
2421 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
2422 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
2424 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
2425 \r?\n in output tests.
2427 2005-04-12 Mike Stump <mrs@apple.com>
2429 * configure: Regenerate.
2431 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
2433 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
2435 2005-03-30 Steve Ellcey <sje@cup.hp.com>
2437 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
2439 * src/ia64/unix.S (ffi_closure_unix): Extend gp
2440 to 64 bits in ILP32 mode.
2441 Load 64 bits even for short data.
2443 2005-03-23 Mike Stump <mrs@apple.com>
2445 * src/powerpc/darwin.S: Update for -m64 multilib.
2446 * src/powerpc/darwin_closure.S: Likewise.
2448 2005-03-21 Zack Weinberg <zack@codesourcery.com>
2450 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
2451 Do not set tool_include_dir.
2452 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
2454 * include/Makefile.am: Set gcc_version and toollibffidir.
2455 * include/Makefile.in: Regenerate.
2457 2005-02-22 Andrew Haley <aph@redhat.com>
2459 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2460 odd-numbered register pairs for 64-bit integer types.
2462 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2465 * testsuite/libffi.call/return_ll1.c: New test case.
2467 2005-02-11 Janis Johnson <janis187@us.ibm.com>
2469 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2470 * testsuite/libffi.call/float.c: Ditto.
2471 * testsuite/libffi.call/float2.c: Ditto.
2472 * testsuite/libffi.call/float3.c: Ditto.
2474 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2476 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2478 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2480 * testsuite/libffi.special/special.exp (cxx_options): Add
2483 2004-12-31 Richard Henderson <rth@redhat.com>
2485 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2486 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2487 offset parameters with a type parameter; deduce size and structure
2488 alignment. Update all users.
2490 2004-12-31 Richard Henderson <rth@redhat.com>
2492 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2493 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
2494 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
2495 into ffi_prep_closure.
2496 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2499 2004-12-27 Richard Henderson <rth@redhat.com>
2501 * src/x86/unix64.S: Fix typo in unwind info.
2503 2004-12-25 Richard Henderson <rth@redhat.com>
2505 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2506 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2507 (merge_classes): Check for it.
2509 (classify_argument): Pass byte_offset by value; perform all updates
2511 (examine_argument): Add classes argument; handle
2512 X86_64_COMPLEX_X87_CLASS.
2513 (ffi_prep_args): Merge into ...
2514 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2515 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2516 (ffi_fill_return_value): Remove.
2517 (ffi_prep_closure): Remove dead assert.
2518 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2519 Rewrite to use struct register_args instead of va_list. Create
2520 flags for handling structure returns.
2521 * src/x86/unix64.S: Remove dead strings.
2522 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2523 stack frame with ffi_call. Handle structure returns properly.
2524 (float2sse, floatfloat2sse, double2sse): Remove.
2525 (sse2float, sse2double, sse2floatfloat): Remove.
2526 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2527 to handle structure returns properly.
2529 2004-12-08 David Edelsohn <edelsohn@gnu.org>
2531 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2533 * Makefile.in: Regenerated.
2535 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2537 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2538 * configure, aclocal.m4, Makefile.in: Regenerate.
2539 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2541 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2543 * configure: Regenerate for libtool change.
2545 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2547 * configure: Regenerate for libtool reversion.
2549 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2551 * configure: Regenerate for libtool change.
2553 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2555 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2557 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2559 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2560 of jal. Use an absolute encoding for the frame information.
2562 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2564 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2565 * acinclude.m4: Delete logic for sincludes.
2566 * aclocal.m4, Makefile.in, configure: Regenerate.
2567 * include/Makefile: Likewise.
2568 * testsuite/Makefile: Likewise.
2570 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2572 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2573 on a 8-byte boundary.
2574 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2576 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2578 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2579 long long values. Round stack allocation to a multiple of 8 bytes
2580 for ATPCS compatibility.
2581 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2582 names. Handle returning long long types. Add Thumb and interworking
2583 support. Improve soft-float code.
2585 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2587 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2588 (libffi_exit): New function.
2589 (libffi_init): Build the testglue wrapper if needed.
2591 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2594 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2596 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2598 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2600 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2602 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2603 * testsuite/libffi.call/float3.c: New test case.
2605 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2607 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2608 the function returning a structure pointed with R2.
2609 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2610 the structure return value if T bit set. Emit position
2611 independent code and EH data if PIC.
2613 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2615 * Makefile.am: Add m32r support.
2616 * configure.ac: Likewise.
2617 * Makefile.in: Regenerate.
2618 * confiugre: Regenerate.
2619 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2620 (uint64, sint64, double, longdouble)
2621 * src/m32r: New directory.
2622 * src/m32r/ffi.c: New file.
2623 * src/m32r/sysv.S: Likewise.
2624 * src/m32r/ffitarget.h: Likewise.
2626 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2628 * testsuite/libffi.call/negint.c: New test case.
2630 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2633 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2634 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2635 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2638 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2640 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2641 * testsuite/libffi.call/promotion.c: Likewise.
2642 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2644 * testsuite/libffi.call/return_sc.c: Likewise.
2645 * testsuite/libffi.call/return_uc.c: Likewise.
2647 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2649 * src/powerpc/darwin.S: Fix comments and identation.
2650 * src/powerpc/darwin_closure.S: Likewise.
2652 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2654 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2655 (ffi_prep_args): Handle longdouble arguments.
2656 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2658 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2659 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2661 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2662 * src/types.c: Defined longdouble size and alignment for darwin.
2664 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2666 * src/powerpc/aix.S: Remove whitespaces.
2667 * src/powerpc/aix_closure.S: Likewise.
2668 * src/powerpc/asm.h: Likewise.
2669 * src/powerpc/ffi.c: Likewise.
2670 * src/powerpc/ffitarget.h: Likewise.
2671 * src/powerpc/linux64.S: Likewise.
2672 * src/powerpc/linux64_closure.S: Likewise.
2673 * src/powerpc/ppc_closure.S: Likewise.
2674 * src/powerpc/sysv.S: Likewise.
2676 2004-08-30 Anthony Green <green@redhat.com>
2678 * Makefile.am: Add frv support.
2679 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2680 * configure.ac: Read configure.host.
2681 * configure.in: Read configure.host.
2682 * configure.host: New file. frv-elf needs libgloss.
2683 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2684 alignment. This is needed to frv and shouldn't harm the others.
2685 * include/ffi_common.h (ALIGN_DOWN): New macro.
2686 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2688 2004-08-24 David Daney <daney@avtrex.com>
2690 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2691 * testsuite/libffi.call/closure_fn1.c: Likewise.
2692 * testsuite/libffi.call/closure_fn2.c Likewise.
2693 * testsuite/libffi.call/closure_fn3.c: Likewise.
2694 * testsuite/libffi.call/closure_fn4.c: Likewise.
2695 * testsuite/libffi.call/closure_fn5.c: Likewise.
2696 * testsuite/libffi.call/cls_18byte.c: Likewise.
2697 * testsuite/libffi.call/cls_19byte.c: Likewise.
2698 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2699 * testsuite/libffi.call/cls_20byte.c: Likewise.
2700 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2701 * testsuite/libffi.call/cls_24byte.c: Likewise.
2702 * testsuite/libffi.call/cls_2byte.c: Likewise.
2703 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2704 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2705 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2706 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2707 * testsuite/libffi.call/cls_4byte.c: Likewise.
2708 * testsuite/libffi.call/cls_64byte.c: Likewise.
2709 * testsuite/libffi.call/cls_6byte.c: Likewise.
2710 * testsuite/libffi.call/cls_7byte.c: Likewise.
2711 * testsuite/libffi.call/cls_8byte.c: Likewise.
2712 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2713 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2714 * testsuite/libffi.call/cls_align_double.c: Likewise.
2715 * testsuite/libffi.call/cls_align_float.c: Likewise.
2716 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2717 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2718 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2719 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2720 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2721 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2722 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2723 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2724 * testsuite/libffi.call/cls_double.c: Likewise.
2725 * testsuite/libffi.call/cls_float.c: Likewise.
2726 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2727 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2728 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2729 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2730 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2731 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2732 * testsuite/libffi.call/cls_schar.c: Likewise.
2733 * testsuite/libffi.call/cls_sint.c: Likewise.
2734 * testsuite/libffi.call/cls_sshort.c: Likewise.
2735 * testsuite/libffi.call/cls_uchar.c: Likewise.
2736 * testsuite/libffi.call/cls_uint.c: Likewise.
2737 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2738 * testsuite/libffi.call/cls_ushort.c: Likewise.
2739 * testsuite/libffi.call/nested_struct.c: Likewise.
2740 * testsuite/libffi.call/nested_struct1.c: Likewise.
2741 * testsuite/libffi.call/nested_struct2.c: Likewise.
2742 * testsuite/libffi.call/nested_struct3.c: Likewise.
2743 * testsuite/libffi.call/problem1.c: Likewise.
2744 * testsuite/libffi.special/unwindtest.cc: Likewise.
2745 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2747 * testsuite/libffi.call/cls_16byte.c: Likewise.
2748 * testsuite/libffi.call/cls_5byte.c: Likewise.
2750 2004-08-23 David Daney <daney@avtrex.com>
2753 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2754 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2755 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2756 parameters and return types.
2757 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2758 (ffi_prep_closure): Ditto.
2759 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2760 alignment calculations.
2761 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2762 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2764 2004-08-14 Casey Marshall <csm@gnu.org>
2766 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2767 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2768 integer (O32 ABI only).
2769 (ffi_prep_closure): new function.
2770 (ffi_closure_mips_inner_O32): new function.
2771 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2772 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2773 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2774 64 bit integers correctly.
2775 (ffi_closure_O32): new function.
2776 Added DWARF-2 unwind info for both functions.
2778 2004-08-10 Andrew Haley <aph@redhat.com>
2780 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2782 2004-08-01 Robert Millan <robertmh@gnu.org>
2784 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2785 * configure: Regenerate.
2787 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2789 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2790 and mmap() explicitly instead of relying on preset autoconf cache
2792 * aclocal.m4: Regenerate.
2793 * configure: Regenerate.
2795 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2797 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2798 (ffi_check_float_struct): Remove unused prototype.
2800 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2802 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2803 character on Darwin, use '\n\t' instead.
2805 2004-06-26 Matthias Klose <doko@debian.org>
2807 * libtool-version: Fix typo in revision/age.
2809 2004-06-17 Matthias Klose <doko@debian.org>
2811 * libtool-version: New.
2812 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2813 * Makefile.in: Regenerate.
2815 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2817 * Makefile.am: Remove useless multilib rules.
2818 * Makefile.in: Regenerate.
2819 * aclocal.m4: Regenerate with automake 1.8.5.
2820 * configure.ac: Remove useless multilib configury.
2821 * configure: Regenerate.
2823 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2825 * .cvsignore: New file.
2827 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2829 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2831 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2834 2004-06-06 Sean McNeil <sean@mcneil.com>
2836 * configure.ac: Add x86_64-*-freebsd* support.
2837 * configure: Regenerate.
2839 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2842 * configure.ac: Test for existence of mmap and sys/mman.h before
2843 checking blacklist. Fix suggested by Jim Wilson.
2844 * configure: Regenerate.
2846 2004-04-26 Matt Austern <austern@apple.com>
2848 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2850 * src/powerpc/darwin_closure.S: Likewise.
2852 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2854 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2855 error. Reported by Thomas Heller <theller@python.net>.
2856 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2857 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2859 2004-03-20 Matthias Klose <doko@debian.org>
2861 * src/pa/linux.S: Fix typo.
2863 2004-03-19 Matthias Klose <doko@debian.org>
2865 * Makefile.am: Update.
2866 * Makefile.in: Regenerate.
2867 * src/pa/ffi.h.in: Remove.
2868 * src/pa/ffitarget.h: New file.
2870 2004-02-10 Randolph Chung <tausq@debian.org>
2872 * Makefile.am: Add PA support.
2873 * Makefile.in: Regenerate.
2874 * include/Makefile.in: Regenerate.
2875 * configure.ac: Add PA target.
2876 * configure: Regenerate.
2877 * src/pa/ffi.c: New file.
2878 * src/pa/ffi.h.in: Add PA support.
2879 * src/pa/linux.S: New file.
2880 * prep_cif.c: Add PA support.
2882 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2884 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2886 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2887 with ecif->cif->flags.
2888 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2890 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2891 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2892 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2893 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2895 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2897 * configure.in: Rename file to ...
2898 * configure.ac: ... this.
2899 * fficonfig.h.in: Regenerate.
2900 * Makefile.in: Regenerate.
2901 * include/Makefile.in: Regenerate.
2902 * testsuite/Makefile.in: Regenerate.
2904 2004-03-12 Matt Austern <austern@apple.com>
2906 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2907 changes in EH format resulting from addition of linkonce support.
2908 * src/powerpc/darwin_closure.S: Likewise.
2910 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2911 Paolo Bonzini <bonzini@gnu.org>
2913 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2914 Remove VPATH. Remove rules for object files. Remove multilib support.
2915 (AM_CCASFLAGS): Add.
2916 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2917 (AC_PREREQ): Bump version to 2.59.
2918 (AC_INIT): Fill with version info and bug address.
2919 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2920 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2921 De-precious CC so that the right flags are passed down to multilibs.
2922 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2923 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2924 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2925 * configure: Rebuilt.
2926 * aclocal.m4: Likewise.
2927 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2928 * fficonfig.h.in: Likewise.
2930 2004-03-11 Andreas Schwab <schwab@suse.de>
2932 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2933 arguments from fp registers only for the first 8 parameter slots.
2934 Don't convert a float parameter when passed in memory.
2936 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2938 * configure: Regenerate for config/accross.m4 correction.
2940 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2942 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2943 ecif->cif->bytes to bytes.
2944 (ffi_prep_cif_machdep): Add braces around nested if statement.
2946 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2948 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2950 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2951 (ffi_closure_helper_LINUX64): Fix typo.
2952 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2954 * testsuite/libffi.call/float.c: Likewise.
2955 * testsuite/libffi.call/float2.c: Likewise.
2957 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2959 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2960 long double function return and long double arg handling.
2961 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2962 Use "end_pfr" instead of "nf". Correct long double handling.
2964 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2966 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2967 space for long double return value. Adjust stack frame and offsets.
2968 Load f2 long double return.
2970 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2972 * src/types.c: Use 16 byte long double for POWERPC64.
2974 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2976 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2977 when the structure return address is passed in %o0.
2978 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2979 (ffi_v9_layout_struct): Align the field following a nested structure
2980 on a word boundary. Use memmove instead of memcpy.
2981 (ffi_call): Update call to ffi_V9_return_struct.
2982 (ffi_prep_closure): Define 'ctx' only for V8.
2983 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2984 and ffi_closure_sparc_inner_v9.
2985 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2986 Always skip the structure return address. For structures and long
2987 doubles, copy the argument directly.
2988 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2989 if required. Shift the maximum floating-point slot accordingly. For
2990 big structures, copy the argument directly; otherwise, left-justify the
2991 argument and call ffi_v9_layout_struct to lay out the structure on
2993 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2994 (ffi_closure_v8): Pass the structure return address. Update call to
2995 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2996 Skip the 'unimp' insn when returning long doubles and structures.
2997 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2998 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2999 FFI_TYPE_INT handling. Load structures both in integers and
3000 floating-point registers on return.
3001 * README: Update status of the SPARC port.
3003 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
3005 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
3007 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
3009 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
3011 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
3012 value as of type ffi_arg, not unsigned int.
3014 2004-01-21 Michael Ritzert <ritzert@t-online.de>
3016 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
3019 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
3021 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
3024 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3026 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
3029 2003-12-10 Richard Henderson <rth@redhat.com>
3031 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
3032 size_t instead of int.
3034 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
3036 * testsuite/libffi.call/many_win32.c: Include <float.h>.
3037 * testsuite/libffi.call/many_win32.c (main): Replace variable
3038 int i with unsigned long ul.
3040 * testsuite/libffi.call/cls_align_uint64.c: New test case.
3041 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
3042 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
3043 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
3044 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
3045 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
3046 * testsuite/libffi.call/cls_align_float.c: Likewise.
3047 * testsuite/libffi.call/cls_align_double.c: Likewise.
3048 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
3049 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
3051 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
3054 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
3055 Align arguments to 32 bits.
3057 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
3060 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
3061 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
3062 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
3063 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
3064 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
3065 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
3067 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
3069 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3071 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
3072 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3074 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
3076 * Makefile.in: Rebuilt.
3077 * configure: Likewise.
3078 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
3081 2003-11-21 Andreas Jaeger <aj@suse.de>
3082 Andreas Tobler <a.tobler@schweiz.ch>
3084 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
3085 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
3086 * Makefile.in: Rebuilt.
3087 * aclocal.m4: Likewise.
3088 * configure: Likewise.
3089 * fficonfig.h.in: Likewise.
3090 * testsuite/lib/libffi-dg.exp: Add include dir.
3091 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
3092 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3093 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
3094 for ffi_closure if available.
3095 * testsuite/libffi.call/closure_fn1.c: Likewise.
3096 * testsuite/libffi.call/closure_fn2.c: Likewise.
3097 * testsuite/libffi.call/closure_fn3.c: Likewise.
3098 * testsuite/libffi.call/closure_fn4.c: Likewise.
3099 * testsuite/libffi.call/closure_fn5.c: Likewise.
3100 * testsuite/libffi.call/cls_12byte.c: Likewise.
3101 * testsuite/libffi.call/cls_16byte.c: Likewise.
3102 * testsuite/libffi.call/cls_18byte.c: Likewise.
3103 * testsuite/libffi.call/cls_19byte.c: Likewise.
3104 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3105 * testsuite/libffi.call/cls_20byte.c: Likewise.
3106 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3107 * testsuite/libffi.call/cls_24byte.c: Likewise.
3108 * testsuite/libffi.call/cls_2byte.c: Likewise.
3109 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3110 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3111 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3112 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3113 * testsuite/libffi.call/cls_4byte.c: Likewise.
3114 * testsuite/libffi.call/cls_5byte.c: Likewise.
3115 * testsuite/libffi.call/cls_64byte.c: Likewise.
3116 * testsuite/libffi.call/cls_6byte.c: Likewise.
3117 * testsuite/libffi.call/cls_7byte.c: Likewise.
3118 * testsuite/libffi.call/cls_8byte.c: Likewise.
3119 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3120 * testsuite/libffi.call/cls_9byte2.c: Likewise.
3121 * testsuite/libffi.call/cls_double.c: Likewise.
3122 * testsuite/libffi.call/cls_float.c: Likewise.
3123 * testsuite/libffi.call/cls_schar.c: Likewise.
3124 * testsuite/libffi.call/cls_sint.c: Likewise.
3125 * testsuite/libffi.call/cls_sshort.c: Likewise.
3126 * testsuite/libffi.call/cls_uchar.c: Likewise.
3127 * testsuite/libffi.call/cls_uint.c: Likewise.
3128 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3129 * testsuite/libffi.call/cls_ushort.c: Likewise.
3130 * testsuite/libffi.call/nested_struct.c: Likewise.
3131 * testsuite/libffi.call/nested_struct1.c: Likewise.
3132 * testsuite/libffi.call/nested_struct2.c: Likewise.
3133 * testsuite/libffi.call/nested_struct3.c: Likewise.
3134 * testsuite/libffi.call/problem1.c: Likewise.
3135 * testsuite/libffi.special/unwindtest.cc: Likewise.
3137 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
3139 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
3141 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
3143 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
3144 Add -lgcc_s to additional flags.
3146 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
3148 * configure.in, include/Makefile.am: PR libgcj/11147, install
3149 the ffitarget.h header file in a gcc versioned and target
3151 * configure: Regenerated.
3152 * Makefile.in, include/Makefile.in: Likewise.
3153 * testsuite/Makefile.in: Likewise.
3155 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
3157 * testsuite/libffi.call/closure_fn0.c: Print result and check
3158 with dg-output to make debugging easier.
3159 * testsuite/libffi.call/closure_fn1.c: Likewise.
3160 * testsuite/libffi.call/closure_fn2.c: Likewise.
3161 * testsuite/libffi.call/closure_fn3.c: Likewise.
3162 * testsuite/libffi.call/closure_fn4.c: Likewise.
3163 * testsuite/libffi.call/closure_fn5.c: Likewise.
3164 * testsuite/libffi.call/cls_12byte.c: Likewise.
3165 * testsuite/libffi.call/cls_16byte.c: Likewise.
3166 * testsuite/libffi.call/cls_18byte.c: Likewise.
3167 * testsuite/libffi.call/cls_19byte.c: Likewise.
3168 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3169 * testsuite/libffi.call/cls_20byte.c: Likewise.
3170 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3171 * testsuite/libffi.call/cls_24byte.c: Likewise.
3172 * testsuite/libffi.call/cls_2byte.c: Likewise.
3173 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3174 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3175 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3176 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3177 * testsuite/libffi.call/cls_4byte.c: Likewise.
3178 * testsuite/libffi.call/cls_5byte.c: Likewise.
3179 * testsuite/libffi.call/cls_64byte.c: Likewise.
3180 * testsuite/libffi.call/cls_6byte.c: Likewise.
3181 * testsuite/libffi.call/cls_7byte.c: Likewise.
3182 * testsuite/libffi.call/cls_8byte.c: Likewise.
3183 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3184 * testsuite/libffi.call/cls_9byte2.c: Likewise.
3185 * testsuite/libffi.call/cls_double.c: Likewise.
3186 * testsuite/libffi.call/cls_float.c: Likewise.
3187 * testsuite/libffi.call/cls_schar.c: Likewise.
3188 * testsuite/libffi.call/cls_sint.c: Likewise.
3189 * testsuite/libffi.call/cls_sshort.c: Likewise.
3190 * testsuite/libffi.call/cls_uchar.c: Likewise.
3191 * testsuite/libffi.call/cls_uint.c: Likewise.
3192 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3193 * testsuite/libffi.call/cls_ushort.c: Likewise.
3194 * testsuite/libffi.call/problem1.c: Likewise.
3196 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
3199 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
3201 * testsuite/libffi.call/cls_9byte2.c: New test case.
3202 * testsuite/libffi.call/cls_9byte1.c: Likewise.
3203 * testsuite/libffi.call/cls_64byte.c: Likewise.
3204 * testsuite/libffi.call/cls_20byte1.c: Likewise.
3205 * testsuite/libffi.call/cls_19byte.c: Likewise.
3206 * testsuite/libffi.call/cls_18byte.c: Likewise.
3207 * testsuite/libffi.call/closure_fn4.c: Likewise.
3208 * testsuite/libffi.call/closure_fn5.c: Likewise.
3209 * testsuite/libffi.call/cls_schar.c: Likewise.
3210 * testsuite/libffi.call/cls_sint.c: Likewise.
3211 * testsuite/libffi.call/cls_sshort.c: Likewise.
3212 * testsuite/libffi.call/nested_struct2.c: Likewise.
3213 * testsuite/libffi.call/nested_struct3.c: Likewise.
3215 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
3217 * testsuite/libffi.call/cls_double.c: Do a check on the result.
3218 * testsuite/libffi.call/cls_uchar.c: Likewise.
3219 * testsuite/libffi.call/cls_uint.c: Likewise.
3220 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3221 * testsuite/libffi.call/cls_ushort.c: Likewise.
3222 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
3224 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
3226 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
3229 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
3231 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
3232 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
3234 2003-10-22 David Daney <ddaney@avtrex.com>
3236 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
3237 __attribute__((__mode__(__SI__))) and friends.
3239 2003-10-22 Andreas Schwab <schwab@suse.de>
3241 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
3243 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
3245 * configure.in: AC_LINK_FILES(ffitarget.h).
3246 * configure: Regenerate.
3247 * Makefile.in: Likewise.
3248 * include/Makefile.in: Likewise.
3249 * testsuite/Makefile.in: Likewise.
3250 * fficonfig.h.in: Likewise.
3252 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
3253 Richard Henderson <rth@redhat.com>
3255 Avoid that ffi.h includes fficonfig.h.
3257 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
3258 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
3259 (TARGET_SRC_MIPS_SGI): Removed.
3260 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
3261 (MIPS_SGI): Removed.
3262 (CLEANFILES): Removed.
3263 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
3265 * acconfig.h: Removed.
3266 * configure.in: Compute sizeofs only for double and long double.
3267 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
3268 into AC_DEFINE instead of using acconfig.h. Create
3269 include/ffitarget.h instead of include/fficonfig.h. Rename
3270 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
3271 AC_DEFINE EH_FRAME_FLAGS.
3272 * include/Makefile.am (DISTCLEANFILES): New automake macro.
3273 (hack_DATA): Add ffitarget.h.
3274 * include/ffi.h.in: Remove all system specific definitions.
3275 Declare raw API even if it is not installed, why bother?
3276 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
3277 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
3278 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
3279 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
3280 * include/ffi_common.h (bool): Do not define.
3281 (ffi_assert): Accept failed assertion.
3282 (ffi_type_test): Return void and accept file/line.
3283 (FFI_ASSERT): Pass stringized failed assertion.
3284 (FFI_ASSERT_AT): New macro.
3285 (FFI_ASSERT_VALID_TYPE): New macro.
3286 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
3287 UINT64, SINT64): Define here with gcc's __attribute__ macro
3289 (FLOAT32, ALIGN): Define here instead of in ffi.h
3290 * include/ffi-mips.h: Removed. Its content moved to
3291 src/mips/ffitarget.h after separating assembly and C sections.
3292 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
3293 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
3294 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
3295 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
3296 SIZEOF_ARG -> FFI_SIZEOF_ARG.
3297 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
3298 * src/debug.c (ffi_assert): Accept stringized failed assertion.
3299 (ffi_type_test): Rewritten.
3300 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
3301 FFI_ASSERT_VALID_TYPE.
3302 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
3303 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
3304 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
3305 src/s390/ffitarget.h, src/sh/ffitarget.h,
3306 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
3307 src/x86/ffitarget.h: New files.
3308 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
3309 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
3310 src/powerpc/aix.S, src/powerpc/darwin.S,
3311 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
3312 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
3313 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
3314 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
3315 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
3318 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3320 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
3321 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
3323 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
3325 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
3326 Used when FFI_DEBUG = 1.
3328 2003-10-14 Alan Modra <amodra@bigpond.net.au>
3330 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
3333 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3335 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
3336 FFI_MIPS_O32 for O32 ABI.
3338 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
3340 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
3341 SPARC64. Cleanup whitespaces.
3343 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
3345 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
3346 strongarm, xscale. Cleanup whitespaces.
3347 * testsuite/libffi.call/closure_fn1.c: Likewise.
3348 * testsuite/libffi.call/closure_fn2.c: Likewise.
3349 * testsuite/libffi.call/closure_fn3.c: Likewise.
3350 * testsuite/libffi.call/cls_12byte.c: Likewise.
3351 * testsuite/libffi.call/cls_16byte.c: Likewise.
3352 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3353 * testsuite/libffi.call/cls_20byte.c: Likewise.
3354 * testsuite/libffi.call/cls_24byte.c: Likewise.
3355 * testsuite/libffi.call/cls_2byte.c: Likewise.
3356 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3357 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3358 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3359 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3360 * testsuite/libffi.call/cls_4byte.c: Likewise.
3361 * testsuite/libffi.call/cls_5byte.c: Likewise.
3362 * testsuite/libffi.call/cls_6byte.c: Likewise.
3363 * testsuite/libffi.call/cls_7byte.c: Likewise.
3364 * testsuite/libffi.call/cls_8byte.c: Likewise.
3365 * testsuite/libffi.call/cls_double.c: Likewise.
3366 * testsuite/libffi.call/cls_float.c: Likewise.
3367 * testsuite/libffi.call/cls_uchar.c: Likewise.
3368 * testsuite/libffi.call/cls_uint.c: Likewise.
3369 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3370 * testsuite/libffi.call/cls_ushort.c: Likewise.
3371 * testsuite/libffi.call/nested_struct.c: Likewise.
3372 * testsuite/libffi.call/nested_struct1.c: Likewise.
3373 * testsuite/libffi.call/problem1.c: Likewise.
3374 * testsuite/libffi.special/unwindtest.cc: Likewise.
3375 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
3377 2003-09-18 David Edelsohn <edelsohn@gnu.org>
3379 * src/powerpc/aix.S: Cleanup whitespaces.
3380 * src/powerpc/aix_closure.S: Likewise.
3382 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3384 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
3385 * src/powerpc/darwin_closure.S: Likewise.
3386 * src/powerpc/ffi_darwin.c: Likewise.
3388 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3389 David Edelsohn <edelsohn@gnu.org>
3391 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
3392 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
3394 * src/powerpc/darwin_closure.S: Likewise.
3395 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
3396 according to the Darwin/AIX ABI.
3397 (ffi_prep_cif_machdep): Likewise.
3398 (ffi_closure_helper_DARWIN): Likewise.
3399 Remove the outgoing parameter stack logic. Simplify the evaluation
3400 of the different CASE types.
3401 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
3402 statement in the trampoline code.
3404 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
3406 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
3407 for the register size.
3408 (ffi_closure_helper_SYSV): Handle the structure return value
3410 (ffi_closure_helper_SYSV): Return the appropriate type when
3411 the registers are used for the structure return value.
3412 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
3413 the 64-bit return value. Update copyright years.
3415 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3417 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
3418 srcdir for ffi_mips.h.
3420 2003-09-12 Alan Modra <amodra@bigpond.net.au>
3422 * src/prep_cif.c (initialize_aggregate): Include tail padding in
3424 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
3425 placement of float result.
3426 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
3427 cast of "resp" for big-endian 64 bit machines.
3429 2003-09-11 Alan Modra <amodra@bigpond.net.au>
3431 * src/types.c (double, longdouble): Merge identical SH and ARM
3432 typedefs, and add POWERPC64.
3433 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
3434 struct split over gpr and rest.
3435 (ffi_prep_cif_machdep): Correct intarg_count for structures.
3436 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
3438 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
3440 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
3443 2003-09-09 Alan Modra <amodra@bigpond.net.au>
3445 * configure: Regenerate.
3447 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3449 * Makefile.am: Remove build rules for ffitest.
3450 * Makefile.in: Rebuilt.
3452 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3454 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3455 about implicit declaration of abort().
3457 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3459 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3460 * Makefile.in: Rebuilt.
3461 * configure.in: Add dejagnu test framework.
3462 * configure: Rebuilt.
3464 * testsuite/Makefile.am: New file.
3465 * testsuite/Makefile.in: Built
3466 * testsuite/lib/libffi-dg.exp: New file.
3467 * testsuite/config/default.exp: Likewise.
3468 * testsuite/libffi.call/call.exp: Likewise.
3469 * testsuite/libffi.call/ffitest.h: Likewise.
3470 * testsuite/libffi.call/closure_fn0.c: Likewise.
3471 * testsuite/libffi.call/closure_fn1.c: Likewise.
3472 * testsuite/libffi.call/closure_fn2.c: Likewise.
3473 * testsuite/libffi.call/closure_fn3.c: Likewise.
3474 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3475 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3476 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3477 * testsuite/libffi.call/cls_2byte.c: Likewise.
3478 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3479 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3480 * testsuite/libffi.call/cls_4byte.c: Likewise.
3481 * testsuite/libffi.call/cls_5byte.c: Likewise.
3482 * testsuite/libffi.call/cls_6byte.c: Likewise.
3483 * testsuite/libffi.call/cls_7byte.c: Likewise.
3484 * testsuite/libffi.call/cls_8byte.c: Likewise.
3485 * testsuite/libffi.call/cls_12byte.c: Likewise.
3486 * testsuite/libffi.call/cls_16byte.c: Likewise.
3487 * testsuite/libffi.call/cls_20byte.c: Likewise.
3488 * testsuite/libffi.call/cls_24byte.c: Likewise.
3489 * testsuite/libffi.call/cls_double.c: Likewise.
3490 * testsuite/libffi.call/cls_float.c: Likewise.
3491 * testsuite/libffi.call/cls_uchar.c: Likewise.
3492 * testsuite/libffi.call/cls_uint.c: Likewise.
3493 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3494 * testsuite/libffi.call/cls_ushort.c: Likewise.
3495 * testsuite/libffi.call/float.c: Likewise.
3496 * testsuite/libffi.call/float1.c: Likewise.
3497 * testsuite/libffi.call/float2.c: Likewise.
3498 * testsuite/libffi.call/many.c: Likewise.
3499 * testsuite/libffi.call/many_win32.c: Likewise.
3500 * testsuite/libffi.call/nested_struct.c: Likewise.
3501 * testsuite/libffi.call/nested_struct1.c: Likewise.
3502 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3503 * testsuite/libffi.call/problem1.c: Likewise.
3504 * testsuite/libffi.call/promotion.c: Likewise.
3505 * testsuite/libffi.call/return_ll.c: Likewise.
3506 * testsuite/libffi.call/return_sc.c: Likewise.
3507 * testsuite/libffi.call/return_uc.c: Likewise.
3508 * testsuite/libffi.call/strlen.c: Likewise.
3509 * testsuite/libffi.call/strlen_win32.c: Likewise.
3510 * testsuite/libffi.call/struct1.c: Likewise.
3511 * testsuite/libffi.call/struct2.c: Likewise.
3512 * testsuite/libffi.call/struct3.c: Likewise.
3513 * testsuite/libffi.call/struct4.c: Likewise.
3514 * testsuite/libffi.call/struct5.c: Likewise.
3515 * testsuite/libffi.call/struct6.c: Likewise.
3516 * testsuite/libffi.call/struct7.c: Likewise.
3517 * testsuite/libffi.call/struct8.c: Likewise.
3518 * testsuite/libffi.call/struct9.c: Likewise.
3519 * testsuite/libffi.special/special.exp: New file.
3520 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3521 * testsuite/libffi.special/unwindtest.cc: Likewise.
3524 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3526 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3529 2003-08-02 Alan Modra <amodra@bigpond.net.au>
3531 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3533 (ffi_closure_helper_LINUX64): Likewise.
3534 * src/powerpc/linux64.S: Remove code writing to parm save area.
3535 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3536 address in lr from ffi_closure_helper_LINUX64 call to calculate
3537 table address. Optimize function tail.
3539 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3541 * src/sparc/ffi.c: Handle all floating point registers.
3542 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3544 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3546 * README: Note that libffi is not part of GCC. Update the project
3549 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3551 * src/powerpc/ppc_closure.S: Include ffi.h.
3553 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3555 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3556 Use C style comments.
3558 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3560 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3561 * Makefile.in: Regenerate.
3562 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3563 * configure: Regenerate.
3564 * include/ffi.h.in: Add SHmedia support.
3565 * src/sh64/ffi.c: New file.
3566 * src/sh64/sysv.S: New file.
3568 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3570 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3571 should be read-only.
3572 * configure: Rebuilt.
3573 * fficonfig.h.in: Rebuilt.
3574 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3575 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3576 * src/powerpc/linux64.S: Likewise.
3577 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3578 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3579 if -fpic/-fPIC/-mrelocatable.
3580 * src/powerpc/powerpc_closure.S: Likewise.
3581 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3582 #write in .eh_frame flags.
3583 * src/sparc/v9.S: Likewise.
3584 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3585 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3586 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3588 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3590 Fixes PR bootstrap/10656
3591 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3592 support for .register pseudo-op.
3593 * src/sparc/v8.S: Use it.
3594 * fficonfig.h.in: Rebuilt.
3595 * configure: Rebuilt.
3597 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3599 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3600 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3601 Make it the default on POWERPC64.
3602 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3603 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3604 * configure: Rebuilt.
3605 * src/powerpc/ffi.c (hidden): Define.
3606 (ffi_prep_args_SYSV): Renamed from
3607 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3608 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3609 ASM_NEEDS_REGISTERS64): New.
3610 (ffi_prep_args64): New function.
3611 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3612 (ffi_call): Likewise.
3613 (ffi_prep_closure): Likewise.
3614 (flush_icache): Surround by #ifndef POWERPC64.
3615 (ffi_dblfl): New union type.
3616 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3617 (ffi_closure_helper_LINUX64): New function.
3618 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3620 * src/powerpc/sysv.S: Likewise.
3621 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3622 * src/powerpc/linux64.S: New file.
3623 * src/powerpc/linux64_closure.S: New file.
3624 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3625 src/powerpc/linux64_closure.S.
3626 (TARGET_SRC_POWERPC): Likewise.
3628 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3629 closure_test_fn3): Fix result printing on big-endian 64-bit
3631 (main): Print tst2_arg instead of uninitialized tst2_result.
3633 * src/ffitest.c (main): Hide what closure pointer really points to
3636 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3638 * configure.in (arm-*-netbsdelf*): Add configuration.
3639 (configure): Regenerated.
3641 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3643 * include/Makefile.in: Regenerate.
3645 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3647 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3649 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3650 Receive closure pointer through parameter, read args using
3651 __builtin_dwarf_cfa.
3652 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3654 2003-03-12 Andreas Schwab <schwab@suse.de>
3656 * configure.in: Avoid trailing /. in toolexeclibdir.
3657 * configure: Rebuilt.
3659 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3661 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3663 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3665 * libffi/src/powerpc/darwin_closure.S:
3666 Fix alignement bug, allocate 8 bytes for the result.
3667 * libffi/src/powerpc/aix_closure.S:
3669 * libffi/src/powerpc/ffi_darwin.c:
3670 Update stackframe description for aix/darwin_closure.S.
3672 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3674 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3677 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3678 Andreas Schwab <schwab@suse.de>
3680 * configure.in: Adjust command to source config-ml.in to account
3681 for changes to the libffi_basedir definition.
3682 (libffi_basedir): Remove ${srcdir} from value and include trailing
3685 * configure: Regenerate.
3687 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3689 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3691 2003-01-28 Andrew Haley <aph@redhat.com>
3693 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3694 * src/x86/ffi64.c (ffi_prep_closure): New.
3695 (ffi_closure_UNIX64_inner): New.
3696 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3698 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
3700 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3701 Remove USE_LIBDIR conditional.
3702 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3703 * Makefile.in, configure: Rebuilt.
3705 2003-01027 David Edelsohn <edelsohn@gnu.org>
3707 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3708 * Makefile.in: Regenerate.
3710 2003-01-22 Andrew Haley <aph@redhat.com>
3712 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3715 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3717 * src/powerpc/darwin.S: Add unwind info.
3718 * src/powerpc/darwin_closure.S: Likewise.
3720 2003-01-14 Andrew Haley <aph@redhat.com>
3722 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3723 (ffi_prep_cif_machdep): Likewise.
3724 * src/x86/unix64.S: Add unwind info.
3726 2003-01-14 Andreas Jaeger <aj@suse.de>
3728 * src/ffitest.c (main): Only use ffi_closures if those are
3731 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3733 * libffi/src/ffitest.c
3734 add closure testcases
3736 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3738 * libffi/src/powerpc/ffi.c
3739 fix alignment bug for float (4 byte aligned iso 8 byte)
3741 2003-01-09 Geoffrey Keating <geoffk@apple.com>
3743 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3744 * src/powerpc/darwin.S: Remove RCS version string.
3746 2003-01-03 Jeff Sturm <jsturm@one-point.com>
3748 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3749 * src/ffitest.c (main): Use static storage for closure.
3750 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3751 * src/sparc/v8.S (ffi_closure_v8): New.
3752 * src/sparc/v9.S (ffi_closure_v9): New.
3754 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3756 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3757 enumeration for X86_WIN32.
3758 * src/x86/win32.S: Added ffi_call_STDCALL function
3760 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3761 switch cases for recognising FFI_STDCALL and
3762 calling ffi_call_STDCALL if target is X86_WIN32.
3763 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3764 stdcall versions of the "my_strlen" and "many"
3765 test functions (for X86_WIN32).
3766 Added test cases to test stdcall invocation using
3769 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3771 * src/sh/sysv.S: Add DWARF2 unwind info.
3773 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3775 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3777 2002-11-26 Jim Wilson <wilson@redhat.com>
3779 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3781 2002-11-23 H.J. Lu <hjl@gnu.org>
3783 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3784 Include ../config/accross.m4.
3785 * aclocal.m4; Rebuild.
3786 * configure: Likewise.
3788 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3790 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3792 2002-11-11 DJ Delorie <dj@redhat.com>
3794 * configure.in: Look for common files in the right place.
3796 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3798 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3799 raw data as _Jv_word values, not ffi_raw.
3800 (ffi_java_ptrarray_to_raw): Likewise.
3801 (ffi_java_rvalue_to_raw): New function.
3802 (ffi_java_raw_call): Call it.
3803 (ffi_java_raw_to_rvalue): New function.
3804 (ffi_java_translate_args): Call it.
3805 * src/ffitest.c (closure_test_fn): Interpret return value
3806 as ffi_arg, not int.
3807 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3808 FFI_TYPE_POINTER case.
3809 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3810 values extended to word size.
3812 2002-10-02 Andreas Jaeger <aj@suse.de>
3814 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3816 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
3818 * include/ffi.h.in: Fix i386 win32 compilation.
3820 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3822 * configure.in: Add s390x-*-linux-* target.
3823 * configure: Regenerate.
3824 * include/ffi.h.in: Define S390X for s390x targets.
3825 (FFI_CLOSURES): Define for s390/s390x.
3826 (FFI_TRAMPOLINE_SIZE): Likewise.
3827 (FFI_NATIVE_RAW_API): Likewise.
3828 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3829 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
3830 * src/s390/ffi.c: Major rework of existing code. Add support for
3831 s390x targets. Add closure support.
3832 * src/s390/sysv.S: Likewise.
3834 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
3836 * src/arm/sysv.S: Fix typo.
3838 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
3840 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3841 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3842 (ffi_call_SYSV): Handle soft-float.
3844 2002-09-27 Bo Thorsen <bo@suse.de>
3846 * include/ffi.h.in: Fix multilib x86-64 support.
3848 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3850 * Makefile.am (all-multi): Fix multilib parallel build.
3852 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3854 * configure.in (sh[34]*-*-linux*): Add brackets.
3855 * configure: Regenerate.
3857 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3859 * Makefile.am: Add SH support.
3860 * Makefile.in: Regenerate.
3861 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3862 * configure: Regenerate.
3863 * include/ffi.h.in: Add SH support.
3864 * src/sh/ffi.c: New file.
3865 * src/sh/sysv.S: New file.
3866 * src/types.c: Add SH support.
3868 2002-07-16 Bo Thorsen <bo@suse.de>
3870 * src/x86/ffi64.c: New file that adds x86-64 support.
3871 * src/x86/unix64.S: New file that handles argument setup for
3873 * src/x86/sysv.S: Don't use this on x86-64.
3874 * src/x86/ffi.c: Don't use this on x86-64.
3876 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3878 * src/ffitest.c (struct6): New test that tests a special case in
3880 (struct7): Likewise.
3881 (struct8): Likewise.
3882 (struct9): Likewise.
3883 (closure_test_fn): Silence warning about this when it's not used.
3884 (main): Add the new tests.
3885 (main): Fix a couple of wrong casts and silence some compiler warnings.
3886 * include/ffi.h.in: Add x86-64 ABI definition.
3887 * fficonfig.h.in: Regenerate.
3888 * Makefile.am: Add x86-64 support.
3889 * configure.in: Likewise.
3890 * Makefile.in: Regenerate.
3891 * configure: Likewise.
3893 2002-06-24 Bo Thorsen <bo@suse.de>
3895 * src/types.c: Merge settings for similar architectures.
3896 Add x86-64 sizes and alignments.
3898 2002-06-23 Bo Thorsen <bo@suse.de>
3900 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3901 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3902 * src/mips/ffi.c (ffi_prep_args): Likewise.
3903 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3905 2002-07-18 H.J. Lu (hjl@gnu.org)
3907 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3908 (libffi_la_SOURCES): Support MIPS_LINUX.
3909 (libffi_convenience_la_SOURCES): Likewise.
3910 * Makefile.in: Regenerated.
3912 * configure.in (mips64*-*): Skip.
3913 (mips*-*-linux*): New.
3914 * configure: Regenerated.
3916 * src/mips/ffi.c: Include <sgidefs.h>.
3918 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3920 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3922 2002-05-27 Roger Sayle <roger@eyesopen.com>
3924 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3926 2002-05-27 Bo Thorsen <bo@suse.de>
3928 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3931 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3933 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3934 beginning of function (for older apple cc).
3936 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3938 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3939 script entry, and set LD to it when configuring multilibs.
3940 * configure: Rebuilt.
3942 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3944 * configure.in (sparc64-*-netbsd*): Add target.
3945 (sparc-*-netbsdelf*): Likewise.
3946 * configure: Regenerate.
3948 2002-04-28 David S. Miller <davem@redhat.com>
3950 * configure.in, configure: Fix SPARC test in previous change.
3952 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3954 * Makefile.am: Add Linux for S/390 support.
3955 * Makefile.in: Regenerate.
3956 * configure.in: Add Linux for S/390 support.
3957 * configure: Regenerate.
3958 * include/ffi.h.in: Add Linux for S/390 support.
3959 * src/s390/ffi.c: New file from libffi CVS tree.
3960 * src/s390/sysv.S: New file from libffi CVS tree.
3962 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3964 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3966 * src/sparc/v8.S: Use it.
3967 * src/sparc/v9.S: Likewise.
3968 * fficonfig.h.in: Rebuilt.
3969 * configure: Rebuilt.
3971 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3973 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3975 * src/ia64/unix.S: Add unwind information. Fix comments.
3976 Save sp in a way that's compatible with unwind info.
3977 (ffi_call_unix): Correctly restore sp in all cases.
3978 * src/ia64/ffi.c: Add, fix comments.
3980 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3982 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3984 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3986 * configure.in (alpha*-*-netbsd*): Add target.
3987 * configure: Regenerate.
3989 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3991 * src/sparc/v8.S: Add unwind info.
3992 * src/sparc/v9.S: Likewise.
3994 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3996 * configure.in: Enable i*86-*-netbsdelf*.
3997 * configure: Rebuilt.
3999 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
4002 * src/mips/n32.s: Delete
4003 * src/mips/o32.s: Delete
4005 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
4007 * configure.in: Enable alpha*-*-freebsd*.
4008 * configure: Rebuilt.
4010 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
4012 * Makefile.am: libfficonvenience -> libffi_convenience.
4013 * Makefile.in: Rebuilt.
4015 * Makefile.am: Define ffitest_OBJECTS.
4016 * Makefile.in: Rebuilt.
4018 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
4019 David Edelsohn <edelsohn@gnu.org>
4021 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
4022 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
4023 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
4024 * Makefile.in: Regenerate.
4025 * include/ffi.h.in: Add AIX and Darwin closure definitions.
4026 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
4027 (flush_icache, flush_range): New functions.
4028 (ffi_closure_helper_DARWIN): New function.
4029 * src/powerpc/aix_closure.S: New file.
4030 * src/powerpc/darwin_closure.S: New file.
4032 2002-02-24 Jeff Sturm <jsturm@one-point.com>
4034 * include/ffi.h.in: Add typedef for ffi_arg.
4035 * src/ffitest.c (main): Declare rint with ffi_arg.
4037 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
4039 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
4040 number of GPRs for floating-point arguments.
4042 2002-01-31 Anthony Green <green@redhat.com>
4044 * configure: Rebuilt.
4045 * configure.in: Replace CHECK_SIZEOF and endian tests with
4046 cross-compiler friendly macros.
4047 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
4050 2002-01-18 David Edelsohn <edelsohn@gnu.org>
4052 * src/powerpc/darwin.S (_ffi_call_AIX): New.
4053 * src/powerpc/aix.S (ffi_call_DARWIN): New.
4055 2002-01-17 David Edelsohn <edelsohn@gnu.org>
4057 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
4058 (TARGET_SRC_POWERPC_AIX): New.
4059 (POWERPC_AIX): New stanza.
4060 * Makefile.in: Regenerate.
4061 * configure.in: Add AIX case.
4062 * configure: Regenerate.
4063 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
4064 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
4065 size. Fix "long double" support.
4066 (ffi_call): Add FFI_AIX case.
4067 * src/powerpc/aix.S: New.
4069 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
4071 Implement Darwin PowerPC ABI.
4072 * configure.in: Handle powerpc-*-darwin*.
4073 * Makefile.am: Set source files for POWERPC_DARWIN.
4074 * configure: Rebuilt.
4075 * Makefile.in: Rebuilt.
4076 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
4078 * src/powerpc/darwin.S: New file.
4079 * src/powerpc/ffi_darwin.c: New file.
4081 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
4083 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
4085 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4087 * src/x86/sysv.S: Avoid gas-only .balign directive.
4088 Use C style comments.
4090 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4092 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
4093 Fixes PR bootstrap/3563.
4095 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4097 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
4099 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
4101 * configure.in: Recognize sparc*-sun-* host.
4102 * configure: Regenerate.
4104 2001-06-06 Andrew Haley <aph@redhat.com>
4106 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
4108 2001-06-03 Andrew Haley <aph@redhat.com>
4110 * src/alpha/osf.S: Add unwind info.
4111 * src/powerpc/sysv.S: Add unwind info.
4112 * src/powerpc/ppc_closure.S: Likewise.
4114 2000-05-31 Jeff Sturm <jsturm@one-point.com>
4116 * configure.in: Fix AC_ARG_ENABLE usage.
4117 * configure: Rebuilt.
4119 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
4121 * configure.in: Remove warning about beta code.
4122 * configure: Rebuilt.
4124 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
4126 * src/ia64/unix.S: Restore stack pointer when returning from
4128 * src/ia64/ffi.c: Fix typo in comment.
4130 2001-04-18 Jim Wilson <wilson@redhat.com>
4132 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
4133 to eliminate RAW DV.
4135 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
4137 * Makefile.am: Make a libtool convenience library.
4138 * Makefile.in: Rebuilt.
4140 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
4142 * configure.in: Use different syntax for subdirectory creation.
4143 * configure: Rebuilt.
4145 2001-03-27 Jon Beniston <jon@beniston.com>
4147 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
4148 * configure: Rebuilt.
4149 * Makefile.am: Added X86_WIN32 target support.
4150 * Makefile.in: Rebuilt.
4152 * include/ffi.h.in: Added X86_WIN32 target support.
4154 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
4155 * src/types.c: Added X86_WIN32 target support.
4157 * src/x86/win32.S: New file. Based on sysv.S, but with EH
4158 stuff removed and made to work with CygWin's gas.
4160 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
4162 * configure.in: Make target subdirectory in build dir.
4163 * Makefile.am: Override suffix based rules to specify correct output
4165 * Makefile.in: Rebuilt.
4166 * configure: Rebuilt.
4168 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
4170 * src/powerpc/ppc_closure.S: New file.
4171 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
4172 involving long long and register pairs.
4173 (ffi_prep_closure): New function.
4174 (flush_icache): Likewise.
4175 (ffi_closure_helper_SYSV): Likewise.
4176 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
4177 (FFI_TRAMPOLINE_SIZE): Likewise.
4178 (FFI_NATIVE_RAW_API): Likewise.
4179 * Makefile.in: Rebuilt.
4180 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
4181 (TARGET_SRC_POWERPC): Likewise.
4183 2001-03-19 Tom Tromey <tromey@redhat.com>
4185 * Makefile.in: Rebuilt.
4186 * Makefile.am (ffitest_LDFLAGS): New macro.
4188 2001-03-02 Nick Clifton <nickc@redhat.com>
4190 * include/ffi.h.in: Remove RCS ident string.
4191 * include/ffi_mips.h: Remove RCS ident string.
4192 * src/debug.c: Remove RCS ident string.
4193 * src/ffitest.c: Remove RCS ident string.
4194 * src/prep_cif.c: Remove RCS ident string.
4195 * src/types.c: Remove RCS ident string.
4196 * src/alpha/ffi.c: Remove RCS ident string.
4197 * src/alpha/osf.S: Remove RCS ident string.
4198 * src/arm/ffi.c: Remove RCS ident string.
4199 * src/arm/sysv.S: Remove RCS ident string.
4200 * src/mips/ffi.c: Remove RCS ident string.
4201 * src/mips/n32.S: Remove RCS ident string.
4202 * src/mips/o32.S: Remove RCS ident string.
4203 * src/sparc/ffi.c: Remove RCS ident string.
4204 * src/sparc/v8.S: Remove RCS ident string.
4205 * src/sparc/v9.S: Remove RCS ident string.
4206 * src/x86/ffi.c: Remove RCS ident string.
4207 * src/x86/sysv.S: Remove RCS ident string.
4209 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
4211 * include/ffi.h.in: Change sourceware.cygnus.com references to
4214 2000-12-09 Richard Henderson <rth@redhat.com>
4216 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
4217 (ffi_closure_osf_inner): Index rather than increment avalue
4218 and arg_types. Give ffi_closure_osf the raw return value type.
4219 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
4222 2000-12-07 Richard Henderson <rth@redhat.com>
4224 * src/raw_api.c (ffi_translate_args): Fix typo.
4225 (ffi_prep_closure): Likewise.
4227 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
4228 FFI_TRAMPOLINE_SIZE.
4229 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
4230 cif->bytes for new ffi_call_osf implementation.
4231 (ffi_prep_args): Absorb into ...
4232 (ffi_call): ... here. Do all stack allocation here and
4233 avoid a callback function.
4234 (ffi_prep_closure, ffi_closure_osf_inner): New.
4235 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
4236 (ffi_closure_osf): New.
4238 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
4240 * config.guess, config.sub, install-sh: Removed.
4241 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
4242 * Makefile.in: Rebuilt.
4244 * acinclude.m4: Include libtool macros from the top level.
4245 * aclocal.m4, configure: Rebuilt.
4247 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
4249 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
4250 * configure: Rebuilt.
4252 2000-05-11 Scott Bambrough <scottb@netwinder.org>
4254 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
4255 memory correctly. Use conditional instructions, not branches where
4258 2000-05-04 Tom Tromey <tromey@cygnus.com>
4260 * configure: Rebuilt.
4261 * configure.in: Match `arm*-*-linux-*'.
4262 From Chris Dornan <cdornan@arm.com>.
4264 2000-04-28 Jakub Jelinek <jakub@redhat.com>
4266 * Makefile.am (SUBDIRS): Define.
4267 (AM_MAKEFLAGS): Likewise.
4268 (Multilib support.): Add section.
4269 * Makefile.in: Rebuilt.
4270 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
4271 New variables. Set for gcc using -print-multi-lib. Export them
4273 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
4274 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
4276 (extra_compiler_flags_value, extra_compiler_flags): Check these
4277 for extra compiler options which need to be passed down in
4280 2000-04-16 Anthony Green <green@redhat.com>
4282 * configure: Rebuilt.
4283 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
4285 2000-04-14 Jakub Jelinek <jakub@redhat.com>
4287 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
4288 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
4289 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
4290 Replace all void * sizeofs with sizeof(int).
4291 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
4292 different than DOUBLE.
4293 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
4294 (ffi_prep_args_v9): New function.
4295 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
4296 (ffi_V9_return_struct): New function.
4297 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
4298 32bit code (not yet cross-arch calls).
4299 * src/sparc/v8.S: Add struct return delay nop.
4301 * src/sparc/v9.S: New file.
4302 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
4303 is used on sparc64 only for structures larger than 32 bytes.
4304 Pass by reference for structures is done for structure arguments
4305 larger than 16 bytes.
4306 * src/ffitest.c (main): Use 64bit rint on sparc64.
4307 Run long long tests on sparc.
4308 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
4310 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
4312 * configure.in (sparc-*-linux*): New supported target.
4313 (sparc64-*-linux*): Likewise.
4314 * configure: Rebuilt.
4315 * Makefile.am: Add v9.S to SPARC files.
4316 * Makefile.in: Likewise.
4317 (LINK): Surround $(CCLD) into double quotes, so that multilib
4318 compiles work correctly.
4320 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
4322 * configure: Rebuilt.
4323 * configure.in: (i*86-*-solaris*): New libffi target. Patch
4324 proposed by Bryce McKinlay.
4326 2000-03-20 Tom Tromey <tromey@cygnus.com>
4328 * Makefile.in: Hand edit for java_raw_api.lo.
4330 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
4332 * config.guess, config.sub: Update from the gcc tree.
4333 Fix for PR libgcj/168.
4335 2000-03-03 Tom Tromey <tromey@cygnus.com>
4337 * Makefile.in: Fixed ia64 by hand.
4339 * configure: Rebuilt.
4340 * configure.in (--enable-multilib): New option.
4341 (libffi_basedir): New subst.
4342 (AC_OUTPUT): Added multilib code.
4344 2000-03-02 Tom Tromey <tromey@cygnus.com>
4346 * Makefile.in: Rebuilt.
4347 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
4350 2000-02-25 Hans Boehm <boehm@acm.org>
4352 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
4354 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
4356 (ffi_prep_raw_closure): Use ffi_translate_args, not
4357 ffi_closure_translate.
4358 * src/java_raw_api.c: New file.
4359 * src/ffitest.c (closure_test_fn): New function.
4360 (main): Define `rint' as long long on IA64. Added new test when
4361 FFI_CLOSURES is defined.
4362 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
4363 (ffi_abi): Recognize IA64.
4364 (ffi_raw): Added `flt' field.
4365 Added "Java raw API" code.
4366 * configure.in: Recognize ia64.
4367 * Makefile.am (TARGET_SRC_IA64): New macro.
4368 (libffi_la_common_SOURCES): Added java_raw_api.c.
4369 (libffi_la_SOURCES): Define in IA64 case.
4371 2000-01-04 Tom Tromey <tromey@cygnus.com>
4373 * Makefile.in: Rebuilt with newer automake.
4375 1999-12-31 Tom Tromey <tromey@cygnus.com>
4377 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
4379 1999-09-01 Tom Tromey <tromey@cygnus.com>
4381 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
4383 * fficonfig.h.in: Rebuilt.
4384 * configure: Rebuilt.
4385 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
4386 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
4387 * acconfig.h: Don't #undef PACKAGE or VERSION.
4389 1999-08-09 Anthony Green <green@cygnus.com>
4391 * include/ffi.h.in: Try to work around messy header problem
4392 with PACKAGE and VERSION.
4394 * configure: Rebuilt.
4395 * configure.in: Change version to 2.00-beta.
4397 * fficonfig.h.in: Rebuilt.
4398 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
4400 * src/x86/ffi.c (ffi_raw_call): Rename.
4402 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
4404 * src/x86/ffi.c (ffi_closure_SYSV): New function.
4405 (ffi_prep_incoming_args_SYSV): Ditto.
4406 (ffi_prep_closure): Ditto.
4407 (ffi_closure_raw_SYSV): Ditto.
4408 (ffi_prep_raw_closure): More ditto.
4409 (ffi_call_raw): Final ditto.
4411 * include/ffi.h.in: Add definitions for closure and raw API.
4413 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4416 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
4418 * src/raw_api.c: New file.
4420 * include/ffi.h.in (ffi_raw): New type.
4421 (UINT_ARG, SINT_ARG): New defines.
4422 (ffi_closure, ffi_raw_closure): New types.
4423 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4425 * configure.in: Add check for endianness and sizeof void*.
4427 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
4428 instead of directly.
4430 * configure: Rebuilt.
4432 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
4434 * configure.in: Add x86 and powerpc BeOS configurations.
4435 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
4437 1999-05-09 Anthony Green <green@cygnus.com>
4439 * configure.in: Add warning about this being beta code.
4440 Remove src/Makefile.am from the picture.
4441 * configure: Rebuilt.
4443 * Makefile.am: Move logic from src/Makefile.am. Add changes
4444 to support libffi as a target library.
4445 * Makefile.in: Rebuilt.
4447 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
4448 Upgraded to new autoconf, automake, libtool.
4452 * LICENSE: Update copyright date.
4454 * src/Makefile.am, src/Makefile.in: Removed.
4456 1998-11-29 Anthony Green <green@cygnus.com>
4458 * include/ChangeLog: Removed.
4459 * src/ChangeLog: Removed.
4460 * src/mips/ChangeLog: Removed.
4461 * src/sparc/ChangeLog: Remboved.
4462 * src/x86/ChangeLog: Removed.
4464 * ChangeLog.v1: Created.