1 2010-03-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
3 * configure.ac (libffi_cv_as_x86_64_unwind_section_type): New test.
4 * configure: Regenerate.
5 * fficonfig.h.in: Regenerate.
6 * libffi/src/x86/unix64.S (.eh_frame)
7 [HAVE_AS_X86_64_UNWIND_SECTION_TYPE]: Use @unwind section type.
9 2010-02-24 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
11 * doc/libffi.texi (The Closure API): Fix typo.
12 * doc/libffi.info: Remove.
14 2010-02-15 Matthias Klose <doko@ubuntu.com>
16 * src/arm/sysv.S (__ARM_ARCH__): Define for processor
19 2010-01-07 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
22 * testsuite/libffi.call/ffitest.h [__alpha__ && __osf__] (PRIdLL,
23 PRIuLL, PRId64, PRIu64, PRIuPTR): Define.
24 * testsuite/libffi.call/cls_align_sint64.c: Add -Wno-format on
26 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
27 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
28 * testsuite/libffi.call/return_ll1.c: Likewise.
29 * testsuite/libffi.call/stret_medium2.c: Likewise.
30 * testsuite/libffi.special/ffitestcxx.h (allocate_mmap): Cast
33 2010-01-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
35 * src/mips/n32.S: Use .abicalls and .eh_frame with __GNUC__.
37 2009-12-31 Anthony Green <green@redhat.com>
39 * README: Update for libffi 3.0.9.
41 2009-12-27 Matthias Klose <doko@ubuntu.com>
43 * configure.ac (HAVE_LONG_DOUBLE): Define for mips when
47 2009-12-26 Anthony Green <green@redhat.com>
49 * testsuite/libffi.call/cls_longdouble_va.c: Mark as xfail for
51 * testsuite/libffi.call/cls_double_va.c: Ditto.
53 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
55 * testsuite/libffi.call/ffitest.h: Conditionally include stdint.h
57 * testsuite/libffi.special/unwindtest.cc: Ditto.
59 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
61 * configure.ac: Add amd64-*-openbsd*.
63 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Link
64 openbsd programs with -lpthread.
66 2009-12-26 Anthony Green <green@redhat.com>
68 * testsuite/libffi.call/cls_double_va.c,
69 testsuite/libffi.call/cls_longdouble.c,
70 testsuite/libffi.call/cls_longdouble_va.c,
71 testsuite/libffi.call/cls_pointer.c,
72 testsuite/libffi.call/cls_pointer_stack.c: Remove xfail for
73 mips*-*-* and arm*-*-*.
74 * testsuite/libffi.call/cls_align_longdouble_split.c,
75 testsuite/libffi.call/cls_align_longdouble_split2.c,
76 testsuite/libffi.call/stret_medium2.c,
77 testsuite/libffi.call/stret_medium.c,
78 testsuite/libffi.call/stret_large.c,
79 testsuite/libffi.call/stret_large2.c: Remove xfail for arm*-*-*.
81 2009-12-31 Kay Tietz <ktietz70@googlemail.com>
83 * testsuite/libffi.call/ffitest.h,
84 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRuLL): Fix
87 2009-12-31 Carlo Bramini <carlo.bramix@libero.it>
89 * configure.ac (AM_LTLDFLAGS): Define for windows hosts.
90 * Makefile.am (libffi_la_LDFLAGS): Add AM_LTLDFLAGS.
92 * Makefile.in: Rebuilt.
94 2009-12-31 Anthony Green <green@redhat.com>
97 * testsuite/libffi.call/huge_struct.c: New test case from Blake
100 2009-12-28 David Edelsohn <edelsohn@gnu.org>
102 * src/powerpc/ffi_darwin.c (ffi_prep_args): Copy abi and nargs to
104 (aix_adjust_aggregate_sizes): New function.
105 (ffi_prep_cif_machdep): Call it.
107 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
109 * configure.ac: Define FFI_MMAP_EXEC_WRIT for the given targets.
110 * configure: Regenerate.
111 * fficonfig.h.in: Likewise.
112 * src/closures.c: Remove the FFI_MMAP_EXEC_WRIT definition for
115 2009-12-26 Andreas Schwab <schwab@linux-m68k.org>
117 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Advance intarg_count
118 when a float arguments is passed in memory.
119 (ffi_closure_helper_SYSV): Mark general registers as used up when
120 a 64bit or soft-float long double argument is passed in memory.
122 2009-12-25 Matthias Klose <doko@ubuntu.com>
124 * man/ffi_call.3: Fix #include in examples.
125 * doc/libffi.texi: Add dircategory.
127 2009-12-25 Frank Everdij <f.p.x.everdij@tudelft.nl>
129 * include/ffi.h.in: Placed '__GNUC__' ifdef around
130 '__attribute__((aligned(8)))' in ffi_closure, fixes compile for
132 * include/ffi_common.h: Added '__sgi' define to non
133 '__attribute__((__mode__()))' integer typedefs.
134 * src/mips/ffi.c (ffi_call, ffi_closure_mips_inner_O32,
135 ffi_closure_mips_inner_N32): Added 'defined(_MIPSEB)' to BE check.
136 (ffi_closure_mips_inner_O32, ffi_closure_mips_inner_N32): Added
137 FFI_LONGDOUBLE support and alignment(N32 only).
138 * src/mips/ffitarget.h: Corrected '#include <sgidefs.h>' for IRIX and
139 fixed non '__attribute__((__mode__()))' integer typedefs.
140 * src/mips/n32.S: Put '#ifdef linux' around '.abicalls' and '.eh_frame'
141 since they are Linux/GNU Assembler specific.
143 2009-12-25 Bradley Smith <brad@brad-smith.co.uk>
145 * configure.ac, Makefile.am, src/avr32/ffi.c,
146 src/avr32/ffitarget.h,
147 src/avr32/sysv.S: Add AVR32 port.
148 * configure, Makefile.in: Rebuilt.
150 2009-12-21 Andreas Tobler <a.tobler@schweiz.org>
152 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
153 * configure: Regenerate.
155 2009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
157 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
159 2009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
161 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
164 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
166 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
169 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
171 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
173 2009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
176 * src/closures.c [X86_64 && __sun__ && __svr4__]
177 (FFI_MMAP_EXEC_WRIT): Define.
179 2009-12-08 David Daney <ddaney@caviumnetworks.com>
181 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
182 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
183 * testsuite/libffi.call/stret_large.c: Same.
184 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
185 * testsuite/libffi.call/stret_large2.c: Same.
186 * testsuite/libffi.call/stret_medium2.c: Same.
188 2009-12-07 David Edelsohn <edelsohn@gnu.org>
190 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
193 2009-12-05 David Edelsohn <edelsohn@gnu.org>
195 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
197 * src/powerpc/aix_closure.S: Same.
199 2009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
201 * Makefile.in: Regenerate.
202 * configure: Regenerate.
203 * include/Makefile.in: Regenerate.
204 * man/Makefile.in: Regenerate.
205 * testsuite/Makefile.in: Regenerate.
207 2009-12-04 David Edelsohn <edelsohn@gnu.org>
209 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
212 2009-12-04 Uros Bizjak <ubizjak@gmail.com>
215 * src/x86/ffi64.c (classify_argument): Update from
216 gcc/config/i386/i386.c.
217 (ffi_closure_unix64_inner): Do not use the address of two consecutive
218 SSE registers directly.
219 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
220 for x86_64 linux targets.
222 2009-12-04 David Edelsohn <edelsohn@gnu.org>
224 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
225 pfr for long double split between fpr13 and stack.
227 2009-12-03 David Edelsohn <edelsohn@gnu.org>
229 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
230 fparg_count twice for long double.
232 2009-12-03 David Edelsohn <edelsohn@gnu.org>
235 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
237 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
239 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
240 Remove xfails for x86 linux targets.
242 2009-12-02 David Edelsohn <edelsohn@gnu.org>
244 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
247 2009-12-01 David Edelsohn <edelsohn@gnu.org>
249 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
250 register usage. Call ffi_prep_args directly. Add long double
251 return value support.
252 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
253 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
254 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
255 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
257 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
259 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
260 Allocate result area between params and FPRs.
262 2009-11-30 David Edelsohn <edelsohn@gnu.org>
265 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
267 * src/powerpc/aix.S: Implement AIX64 version.
268 * src/powerpc/aix_closure.S: Implement AIX64 version.
269 (ffi_closure_ASM): Use extsb, lha and displament addresses.
270 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
272 (ffi_prep_cif_machdep): Same.
274 (ffi_closure_helper_DARWIN): Same.
276 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
279 * testsuite/libffi.call/testclosure.c: New test.
281 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
283 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
284 assembly version use of ___chkstk.
286 2009-09-23 Matthias Klose <doko@ubuntu.com>
288 PR libffi/40242, PR libffi/41443
289 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
290 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
291 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
292 Change the conditionals to __SOFTFP__ || __ARM_EABI__
293 for -mfloat-abi=softfp to work.
295 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
297 PR testsuite/32843 (strikes again)
298 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
299 enable proper extension on char and short.
301 2009-09-15 David Daney <ddaney@caviumnetworks.com>
303 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
304 handling for FFI_TYPE_POINTER.
305 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
306 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
307 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
308 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
309 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
310 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
311 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
312 structure and pointer returns.
313 (ffi_closure_N32): Add handling for pointer returns.
314 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
315 calc_n32_return_struct_flags): Handle soft-float.
316 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
317 (ffi_call_N32): Declare proper argument types.
318 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
321 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
323 * configure.ac (AC_PREREQ): Bump to 2.64.
325 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
327 * Makefile.am (install-html, install-pdf): Remove.
328 * Makefile.in: Regenerate.
330 * Makefile.in: Regenerate.
331 * aclocal.m4: Regenerate.
332 * configure: Regenerate.
333 * fficonfig.h.in: Regenerate.
334 * include/Makefile.in: Regenerate.
335 * man/Makefile.in: Regenerate.
336 * testsuite/Makefile.in: Regenerate.
338 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
340 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
342 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
345 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
346 return types for X86_WIN32.
347 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
348 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
349 _ffi_closure_STDCALL): Likewise.
351 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
352 (dlmmap, dlmunmap): Also use these functions on Cygwin.
354 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
359 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
362 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
364 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
365 to "" before adding paths. (This reinstates an assignment that
366 was removed by my 2009-06-30 commit, but changes the initial
367 value from "." to "".)
369 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
372 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
373 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
376 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
378 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
379 to ld_library_path. Use add_path. Add just find_libgcc_s
380 to ld_library_path, not every libgcc multilib directory.
382 2009-06-16 Wim Lewis <wiml@hhhh.org>
384 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
385 supposed to be callee-saved.
386 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
387 return buffer for odd-size structs.
389 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
392 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
393 allow_stack_execute for Darwin.
395 2009-06-16 Andrew Haley <aph@redhat.com>
397 * configure.ac (TARGETDIR): Add missing blank lines.
398 * configure: Regenerate.
400 2009-06-16 Andrew Haley <aph@redhat.com>
402 * testsuite/libffi.call/cls_align_sint64.c,
403 testsuite/libffi.call/cls_align_uint64.c,
404 testsuite/libffi.call/cls_longdouble_va.c,
405 testsuite/libffi.call/cls_ulonglong.c,
406 testsuite/libffi.call/return_ll1.c,
407 testsuite/libffi.call/stret_medium2.c: Fix printf format
409 * testsuite/libffi.call/ffitest.h,
410 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
412 2009-06-15 Andrew Haley <aph@redhat.com>
414 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
415 * testsuite/libffi.call/err_bad_abi.c: Likewise.
417 2009-06-12 Andrew Haley <aph@redhat.com>
419 * Makefile.am: Remove info_TEXINFOS.
421 2009-06-12 Andrew Haley <aph@redhat.com>
423 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
424 testsuite/libffi.call/cls_align_uint64.c,
425 testsuite/libffi.call/cls_ulonglong.c,
426 testsuite/libffi.call/return_ll1.c,
427 testsuite/libffi.call/stret_medium2.c: Fix printf format
429 testsuite/libffi.special/unwindtest.cc: include stdint.h.
431 2009-06-11 Timothy Wall <twall@users.sf.net>
436 include/ffi_common.h,
441 src/x86/win64.S (new),
442 README: Added win64 support (mingw or MSVC)
446 testsuite/Makefile.in,
448 aclocal.m4: Regenerated
449 * ltcf-c.sh: properly escape cygwin/w32 path
450 * man/ffi_call.3: Clarify size requirements for return value.
451 * src/x86/ffi64.c: Fix filename in comment.
452 * src/x86/win32.S: Remove unused extern.
454 * testsuite/libffi.call/closure_fn0.c,
455 testsuite/libffi.call/closure_fn1.c,
456 testsuite/libffi.call/closure_fn2.c,
457 testsuite/libffi.call/closure_fn3.c,
458 testsuite/libffi.call/closure_fn4.c,
459 testsuite/libffi.call/closure_fn5.c,
460 testsuite/libffi.call/closure_fn6.c,
461 testsuite/libffi.call/closure_stdcall.c,
462 testsuite/libffi.call/cls_12byte.c,
463 testsuite/libffi.call/cls_16byte.c,
464 testsuite/libffi.call/cls_18byte.c,
465 testsuite/libffi.call/cls_19byte.c,
466 testsuite/libffi.call/cls_1_1byte.c,
467 testsuite/libffi.call/cls_20byte.c,
468 testsuite/libffi.call/cls_20byte1.c,
469 testsuite/libffi.call/cls_24byte.c,
470 testsuite/libffi.call/cls_2byte.c,
471 testsuite/libffi.call/cls_3_1byte.c,
472 testsuite/libffi.call/cls_3byte1.c,
473 testsuite/libffi.call/cls_3byte2.c,
474 testsuite/libffi.call/cls_4_1byte.c,
475 testsuite/libffi.call/cls_4byte.c,
476 testsuite/libffi.call/cls_5_1_byte.c,
477 testsuite/libffi.call/cls_5byte.c,
478 testsuite/libffi.call/cls_64byte.c,
479 testsuite/libffi.call/cls_6_1_byte.c,
480 testsuite/libffi.call/cls_6byte.c,
481 testsuite/libffi.call/cls_7_1_byte.c,
482 testsuite/libffi.call/cls_7byte.c,
483 testsuite/libffi.call/cls_8byte.c,
484 testsuite/libffi.call/cls_9byte1.c,
485 testsuite/libffi.call/cls_9byte2.c,
486 testsuite/libffi.call/cls_align_double.c,
487 testsuite/libffi.call/cls_align_float.c,
488 testsuite/libffi.call/cls_align_longdouble.c,
489 testsuite/libffi.call/cls_align_longdouble_split.c,
490 testsuite/libffi.call/cls_align_longdouble_split2.c,
491 testsuite/libffi.call/cls_align_pointer.c,
492 testsuite/libffi.call/cls_align_sint16.c,
493 testsuite/libffi.call/cls_align_sint32.c,
494 testsuite/libffi.call/cls_align_sint64.c,
495 testsuite/libffi.call/cls_align_uint16.c,
496 testsuite/libffi.call/cls_align_uint32.c,
497 testsuite/libffi.call/cls_align_uint64.c,
498 testsuite/libffi.call/cls_dbls_struct.c,
499 testsuite/libffi.call/cls_double.c,
500 testsuite/libffi.call/cls_double_va.c,
501 testsuite/libffi.call/cls_float.c,
502 testsuite/libffi.call/cls_longdouble.c,
503 testsuite/libffi.call/cls_longdouble_va.c,
504 testsuite/libffi.call/cls_multi_schar.c,
505 testsuite/libffi.call/cls_multi_sshort.c,
506 testsuite/libffi.call/cls_multi_sshortchar.c,
507 testsuite/libffi.call/cls_multi_uchar.c,
508 testsuite/libffi.call/cls_multi_ushort.c,
509 testsuite/libffi.call/cls_multi_ushortchar.c,
510 testsuite/libffi.call/cls_pointer.c,
511 testsuite/libffi.call/cls_pointer_stack.c,
512 testsuite/libffi.call/cls_schar.c,
513 testsuite/libffi.call/cls_sint.c,
514 testsuite/libffi.call/cls_sshort.c,
515 testsuite/libffi.call/cls_uchar.c,
516 testsuite/libffi.call/cls_uint.c,
517 testsuite/libffi.call/cls_ulonglong.c,
518 testsuite/libffi.call/cls_ushort.c,
519 testsuite/libffi.call/err_bad_abi.c,
520 testsuite/libffi.call/err_bad_typedef.c,
521 testsuite/libffi.call/float2.c,
522 testsuite/libffi.call/huge_struct.c,
523 testsuite/libffi.call/nested_struct.c,
524 testsuite/libffi.call/nested_struct1.c,
525 testsuite/libffi.call/nested_struct10.c,
526 testsuite/libffi.call/nested_struct2.c,
527 testsuite/libffi.call/nested_struct3.c,
528 testsuite/libffi.call/nested_struct4.c,
529 testsuite/libffi.call/nested_struct5.c,
530 testsuite/libffi.call/nested_struct6.c,
531 testsuite/libffi.call/nested_struct7.c,
532 testsuite/libffi.call/nested_struct8.c,
533 testsuite/libffi.call/nested_struct9.c,
534 testsuite/libffi.call/problem1.c,
535 testsuite/libffi.call/return_ldl.c,
536 testsuite/libffi.call/return_ll1.c,
537 testsuite/libffi.call/stret_large.c,
538 testsuite/libffi.call/stret_large2.c,
539 testsuite/libffi.call/stret_medium.c,
540 testsuite/libffi.call/stret_medium2.c,
541 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
542 of checking for MMAP. Use intptr_t instead of long casts.
544 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
546 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
547 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
548 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
550 2009-06-09 Andrew Haley <aph@redhat.com>
552 * src/x86/freebsd.S: Add missing file.
554 2009-06-08 Andrew Haley <aph@redhat.com>
556 Import from libffi 3.0.8:
558 * doc/libffi.texi: New file.
559 * doc/libffi.info: Likewise.
560 * doc/stamp-vti: Likewise.
561 * man/Makefile.am: New file.
562 * man/ffi_call.3: New file.
564 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
566 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
568 * configure.ac: Bump version to 3.0.8.
569 parisc*-*-linux*: Add.
570 i386-*-freebsd* | i386-*-openbsd*: Add.
571 powerpc-*-beos*: Add.
572 AM_CONDITIONAL X86_FREEBSD: Add.
573 AC_CONFIG_FILES: Add man/Makefile.
575 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
577 2009-06-08 Andrew Haley <aph@redhat.com>
579 * README: Import from libffi 3.0.8.
581 2009-06-08 Andrew Haley <aph@redhat.com>
583 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
584 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
585 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
586 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
588 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
589 * testsuite/libffi.call/stret_medium.c: Likewise.
590 * testsuite/libffi.call/stret_large2.c: Likewise.
591 * testsuite/libffi.call/stret_large.c: Likewise.
593 2008-12-26 Timothy Wall <twall@users.sf.net>
595 * testsuite/libffi.call/cls_longdouble.c,
596 testsuite/libffi.call/cls_longdouble_va.c,
597 testsuite/libffi.call/cls_align_longdouble.c,
598 testsuite/libffi.call/cls_align_longdouble_split.c,
599 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
600 failures on x86_64 cygwin/mingw.
602 2008-12-22 Timothy Wall <twall@users.sf.net>
604 * testsuite/libffi.call/closure_fn0.c,
605 testsuite/libffi.call/closure_fn1.c,
606 testsuite/libffi.call/closure_fn2.c,
607 testsuite/libffi.call/closure_fn3.c,
608 testsuite/libffi.call/closure_fn4.c,
609 testsuite/libffi.call/closure_fn5.c,
610 testsuite/libffi.call/closure_fn6.c,
611 testsuite/libffi.call/closure_loc_fn0.c,
612 testsuite/libffi.call/closure_stdcall.c,
613 testsuite/libffi.call/cls_align_pointer.c,
614 testsuite/libffi.call/cls_pointer.c,
615 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
616 pointer to integer (intptr_t).
617 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
619 2008-07-24 Anthony Green <green@redhat.com>
621 * testsuite/libffi.call/cls_dbls_struct.c,
622 testsuite/libffi.call/cls_double_va.c,
623 testsuite/libffi.call/cls_longdouble.c,
624 testsuite/libffi.call/cls_longdouble_va.c,
625 testsuite/libffi.call/cls_pointer.c,
626 testsuite/libffi.call/cls_pointer_stack.c,
627 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
630 2008-03-04 Anthony Green <green@redhat.com>
634 * testsuite/libffi.call/cls_align_longdouble_split2.c
635 testsuite/libffi.call/cls_align_longdouble_split.c
636 testsuite/libffi.call/cls_dbls_struct.c
637 testsuite/libffi.call/cls_double_va.c
638 testsuite/libffi.call/cls_longdouble.c
639 testsuite/libffi.call/cls_longdouble_va.c
640 testsuite/libffi.call/cls_pointer.c
641 testsuite/libffi.call/cls_pointer_stack.c
642 testsuite/libffi.call/err_bad_abi.c
643 testsuite/libffi.call/err_bad_typedef.c
644 testsuite/libffi.call/stret_large2.c
645 testsuite/libffi.call/stret_large.c
646 testsuite/libffi.call/stret_medium2.c
647 testsuite/libffi.call/stret_medium.c: New tests from Apple.
649 2009-06-05 Andrew Haley <aph@redhat.com>
651 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
654 2009-06-04 Andrew Haley <aph@redhat.com>
656 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
659 2008-02-26 Anthony Green <green@redhat.com>
660 Thomas Heller <theller@ctypes.org>
662 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
665 2008-02-03 Timothy Wall <twall@users.sf.net>
667 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
668 offset based on code pointer, not data pointer.
670 2008-01-31 Timothy Wall <twall@users.sf.net>
672 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
674 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
676 * src/x86/win32.S: Add assembly for stdcall closure.
677 * src/x86/ffi.c: Initialize stdcall closure trampoline.
679 2009-06-04 Andrew Haley <aph@redhat.com>
681 * include/ffi.h.in: Change void (*)() to void (*)(void).
682 * src/x86/ffi.c: Likewise.
684 2009-06-04 Andrew Haley <aph@redhat.com>
686 * src/powerpc/ppc_closure.S: Insert licence header.
687 * src/powerpc/linux64_closure.S: Likewise.
688 * src/m68k/sysv.S: Likewise.
690 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
691 * src/powerpc/ffi.c: Likewise.
692 * src/powerpc/ffi_darwin.c: Likewise.
693 * src/m32r/ffi.c: Likewise.
694 * src/sh64/ffi.c: Likewise.
695 * src/x86/ffi64.c: Likewise.
696 * src/alpha/ffi.c: Likewise.
697 * src/alpha/osf.S: Likewise.
698 * src/frv/ffi.c: Likewise.
699 * src/s390/ffi.c: Likewise.
700 * src/pa/ffi.c: Likewise.
701 * src/pa/hpux32.S: Likewise.
702 * src/ia64/unix.S: Likewise.
703 * src/ia64/ffi.c: Likewise.
704 * src/sparc/ffi.c: Likewise.
705 * src/mips/ffi.c: Likewise.
706 * src/sh/ffi.c: Likewise.
708 2008-02-15 David Daney <ddaney@avtrex.com>
710 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
711 Define (conditionally), and use it to include cachectl.h.
712 (ffi_prep_closure_loc): Fix cache flushing.
713 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
715 2009-06-04 Andrew Haley <aph@redhat.com>
721 src/powerpc/ffitarget.h,
723 src/sh64/ffitarget.h,
734 src/alpha/ffitarget.h,
737 src/m68k/ffitarget.h,
740 src/s390/ffitarget.h,
742 src/cris/ffitarget.h,
747 src/ia64/ffitarget.h,
750 src/ia64/ia64_flags.h,
754 src/sparc/ffitarget.h,
757 src/mips/ffitarget.h,
764 src/sh/sysv.S: Update license text.
766 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
768 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
769 (.eh_frame): Add FDE for it.
771 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
773 * configure.ac: Also check if assembler supports pc-relative
774 relocs on X86_WIN32 targets.
775 * configure: Regenerate.
776 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
777 (_ffi_call_SYSV): Add missing function type symbol .def and
778 add EH markup labels.
779 (_ffi_call_STDCALL): Likewise.
780 (_ffi_closure_SYSV): Likewise.
781 (_ffi_closure_raw_SYSV): Likewise.
782 (.eh_frame): Add hand-crafted EH data.
784 2009-04-09 Jakub Jelinek <jakub@redhat.com>
786 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
787 version 3 of the GNU General Public License and to point readers
788 at the COPYING3 file and the FSF's license web page.
789 * testsuite/libffi.call/call.exp: Likewise.
790 * testsuite/libffi.special/special.exp: Likewise.
792 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
794 * configure: Regenerate.
796 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
799 * configure.ac (HAVE_AS_X86_PCREL): New test.
800 * configure: Regenerate.
801 * fficonfig.h.in: Regenerate.
802 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
803 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
804 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
805 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
806 * src/x86/unix64.S (.Lstore_table): Move to .text section.
807 (.Lload_table): Likewise.
808 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
810 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
812 * configure: Regenerate.
814 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
816 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
817 signed/unsigned int8/16 return values.
818 * src/sparc/v8.S (ffi_call_v8): Likewise.
819 (ffi_closure_v8): Likewise.
821 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
822 Steve Ellcey <sje@cup.hp.com>
824 * configure: Regenerate for new libtool.
825 * Makefile.in: Ditto.
826 * include/Makefile.in: Ditto.
829 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
831 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
832 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
833 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
834 Adjust copyright notice.
835 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
836 register or two register to use for FFI_SYSV structs.
837 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
838 (ffi_closure_helper_SYSV): Fix the return type for
839 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
840 Adjust copyright notice.
842 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
844 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
847 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
849 * configure: Regenerate.
850 * include/Makefile.in: Regenerate.
851 * testsuite/Makefile.in: Regenerate.
853 2008-06-07 Joseph Myers <joseph@codesourcery.com>
855 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
856 powerpc-*-beos*): Remove.
857 * configure: Regenerate.
859 2008-05-09 Julian Brown <julian@codesourcery.com>
861 * Makefile.am (LTLDFLAGS): New.
862 (libffi_la_LDFLAGS): Use above.
863 * Makefile.in: Regenerate.
865 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
868 * aclocal.m4: Regenerate.
869 * configure: Regenerate.
871 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
873 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
874 * src/sh64/sysv.S: Likewise.
876 2008-03-26 Daniel Jacobowitz <dan@debian.org>
878 * src/arm/sysv.S: Fix ARM comment marker.
880 2008-03-26 Jakub Jelinek <jakub@redhat.com>
882 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
883 * src/s390/sysv.S: Likewise.
884 * src/powerpc/ppc_closure.S: Likewise.
885 * src/powerpc/sysv.S: Likewise.
886 * src/x86/unix64.S: Likewise.
887 * src/x86/sysv.S: Likewise.
888 * src/sparc/v8.S: Likewise.
889 * src/sparc/v9.S: Likewise.
890 * src/m68k/sysv.S: Likewise.
891 * src/arm/sysv.S: Likewise.
893 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
895 * aclocal.m4: Regenerate.
896 * configure: Likewise.
897 * Makefile.in: Likewise.
898 * include/Makefile.in: Likewise.
899 * testsuite/Makefile.in: Likewise.
901 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
902 Andreas Tobler <a.tobler@schweiz.org>
904 * configure.ac: Add amd64-*-freebsd* target.
905 * configure: Regenerate.
907 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
910 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
913 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
916 2008-01-24 David Edelsohn <edelsohn@gnu.org>
918 * configure: Regenerate.
920 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
922 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
924 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
927 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
928 signed/unsigned int8/16 for X86_DARWIN.
929 Updated copyright info.
930 Handle one and two byte structs with special cif->flags.
931 * src/x86/ffitarget.h: Add special types for one and two byte structs.
932 Updated copyright info.
933 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
935 Remove code to pop args from the stack after call.
936 Special-case signed/unsigned for int8/16, one and two byte structs.
937 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
938 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
940 Updated copyright info.
942 2007-12-08 David Daney <ddaney@avtrex.com>
944 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
945 SUBU, add with ADDU and use smaller code sequences.
947 2007-12-07 David Daney <ddaney@avtrex.com>
949 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
952 2007-12-06 David Daney <ddaney@avtrex.com>
954 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
956 (ffi_java_raw): New typedef.
957 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
958 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
960 (ffi_java_raw_closure) : Same.
961 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
963 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
965 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
966 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
967 sizeof(ffi_java_raw) for alignment calculations.
968 (ffi_java_ptrarray_to_raw): Same.
969 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
970 if FFI_SIZEOF_JAVA_RAW == 4.
971 (ffi_java_raw_to_rvalue): Same.
972 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
973 (ffi_java_translate_args): Same.
974 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
976 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
978 2007-12-06 David Daney <ddaney@avtrex.com>
980 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
983 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
986 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
987 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
988 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
989 set the NUM_FPR_ARG_REGISTERS according to.
990 Add support for potential soft-float support under hard-float
992 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
993 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
994 to the FFI_LINUX_SOFT_FLOAT ABI.
995 (ffi_prep_cif_machdep): Likewise.
996 (ffi_closure_helper_SYSV): Likewise.
997 * src/powerpc/ppc_closure.S: Make sure not to store float/double
998 on archs where __NO_FPRS__ is true.
999 Add FFI_TYPE_UINT128 support.
1000 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
1001 Adjust copyright notice.
1003 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
1005 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
1006 * include/ffi_common.h: ... here.
1009 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
1011 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
1012 * src/powerpc/linux64.S: Likewise.
1013 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
1014 * testsuite/libffi.call/return_ldl.c: New test.
1016 2007-09-04 <aph@redhat.com>
1018 * src/arm/sysv.S (UNWIND): New.
1019 (Whole file): Conditionally compile unwinder directives.
1020 * src/arm/sysv.S: Add unwinder directives.
1022 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
1023 Only treat r0 as a struct address if we're actually returning a
1025 Only copy the bytes that are actually within a struct.
1026 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
1027 is returned in r0, not passed by address.
1028 (ffi_call): Allocate a word-sized temporary for the case where
1029 a composite is returned in r0.
1030 (ffi_prep_incoming_args_SYSV): Align as necessary.
1032 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
1034 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
1035 directly using the sys_cacheflush syscall.
1037 2007-07-27 Andrew Haley <aph@redhat.com>
1039 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
1041 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
1043 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
1044 * configure.ac: Likewise.
1045 * Makefile.in: Regenerate.
1046 * include/Makefile.in: Likewise.
1047 * testsuite/Makefile.in: Likewise.
1048 * configure: Likewise.
1050 2007-08-24 David Daney <ddaney@avtrex.com>
1052 * testsuite/libffi.call/return_sl.c: New test.
1054 2007-08-10 David Daney <ddaney@avtrex.com>
1056 * testsuite/libffi.call/cls_multi_ushort.c,
1057 testsuite/libffi.call/cls_align_uint16.c,
1058 testsuite/libffi.call/nested_struct1.c,
1059 testsuite/libffi.call/nested_struct3.c,
1060 testsuite/libffi.call/cls_7_1_byte.c,
1061 testsuite/libffi.call/nested_struct5.c,
1062 testsuite/libffi.call/cls_double.c,
1063 testsuite/libffi.call/nested_struct7.c,
1064 testsuite/libffi.call/cls_sint.c,
1065 testsuite/libffi.call/nested_struct9.c,
1066 testsuite/libffi.call/cls_20byte1.c,
1067 testsuite/libffi.call/cls_multi_sshortchar.c,
1068 testsuite/libffi.call/cls_align_sint64.c,
1069 testsuite/libffi.call/cls_3byte2.c,
1070 testsuite/libffi.call/cls_multi_schar.c,
1071 testsuite/libffi.call/cls_multi_uchar.c,
1072 testsuite/libffi.call/cls_19byte.c,
1073 testsuite/libffi.call/cls_9byte1.c,
1074 testsuite/libffi.call/cls_align_float.c,
1075 testsuite/libffi.call/closure_fn1.c,
1076 testsuite/libffi.call/problem1.c,
1077 testsuite/libffi.call/closure_fn3.c,
1078 testsuite/libffi.call/cls_sshort.c,
1079 testsuite/libffi.call/closure_fn5.c,
1080 testsuite/libffi.call/cls_align_double.c,
1081 testsuite/libffi.call/nested_struct.c,
1082 testsuite/libffi.call/cls_2byte.c,
1083 testsuite/libffi.call/nested_struct10.c,
1084 testsuite/libffi.call/cls_4byte.c,
1085 testsuite/libffi.call/cls_6byte.c,
1086 testsuite/libffi.call/cls_8byte.c,
1087 testsuite/libffi.call/cls_multi_sshort.c,
1088 testsuite/libffi.call/cls_align_sint16.c,
1089 testsuite/libffi.call/cls_align_uint32.c,
1090 testsuite/libffi.call/cls_20byte.c,
1091 testsuite/libffi.call/cls_float.c,
1092 testsuite/libffi.call/nested_struct2.c,
1093 testsuite/libffi.call/cls_5_1_byte.c,
1094 testsuite/libffi.call/nested_struct4.c,
1095 testsuite/libffi.call/cls_24byte.c,
1096 testsuite/libffi.call/nested_struct6.c,
1097 testsuite/libffi.call/cls_64byte.c,
1098 testsuite/libffi.call/nested_struct8.c,
1099 testsuite/libffi.call/cls_uint.c,
1100 testsuite/libffi.call/cls_multi_ushortchar.c,
1101 testsuite/libffi.call/cls_schar.c,
1102 testsuite/libffi.call/cls_uchar.c,
1103 testsuite/libffi.call/cls_align_uint64.c,
1104 testsuite/libffi.call/cls_ulonglong.c,
1105 testsuite/libffi.call/cls_align_longdouble.c,
1106 testsuite/libffi.call/cls_1_1byte.c,
1107 testsuite/libffi.call/cls_12byte.c,
1108 testsuite/libffi.call/cls_3_1byte.c,
1109 testsuite/libffi.call/cls_3byte1.c,
1110 testsuite/libffi.call/cls_4_1byte.c,
1111 testsuite/libffi.call/cls_6_1_byte.c,
1112 testsuite/libffi.call/cls_16byte.c,
1113 testsuite/libffi.call/cls_18byte.c,
1114 testsuite/libffi.call/closure_fn0.c,
1115 testsuite/libffi.call/cls_9byte2.c,
1116 testsuite/libffi.call/closure_fn2.c,
1117 testsuite/libffi.call/closure_fn4.c,
1118 testsuite/libffi.call/cls_ushort.c,
1119 testsuite/libffi.call/closure_fn6.c,
1120 testsuite/libffi.call/cls_5byte.c,
1121 testsuite/libffi.call/cls_align_pointer.c,
1122 testsuite/libffi.call/cls_7byte.c,
1123 testsuite/libffi.call/cls_align_sint32.c,
1124 testsuite/libffi.special/unwindtest_ffi_call.cc,
1125 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
1127 2007-08-10 David Daney <ddaney@avtrex.com>
1130 * configure.ac: Don't treat mips64 as a special case.
1131 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
1132 * configure: Regenerate
1133 * Makefile.in: Ditto.
1134 * fficonfig.h.in: Ditto.
1135 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
1136 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
1137 (FFI_DEFAULT_ABI): Set for n64 case.
1138 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
1139 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
1140 (ffi_closure_N32): New function.
1141 (.eh_frame): New section
1142 * src/mips/o32.S: Clean up comments.
1143 (ffi_closure_O32): Pass ffi_closure parameter in $12.
1144 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
1145 _MIPS_SIM == _ABIN32 throughout.
1146 (FFI_MIPS_STOP_HERE): New, use in place of
1148 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1149 to support n32/n64 ABIs.
1150 (calc_n32_struct_flags): Rewrite.
1151 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1152 position of flag bits.
1153 (ffi_prep_cif_machdep): Rewrite n32 portion.
1154 (ffi_call): Enable for n64. Add special handling for small structure
1156 (ffi_prep_closure_loc): Add n32 and n64 support.
1157 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1158 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1160 2007-08-08 David Daney <ddaney@avtrex.com>
1162 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1163 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1165 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1166 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1167 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1168 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1169 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1170 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1171 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1172 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1173 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1174 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1175 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1176 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1177 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1178 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1179 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1180 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1181 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1182 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1183 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1184 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1185 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1186 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1187 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1188 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1189 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1190 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1191 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1192 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1193 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1194 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1195 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1197 2007-08-07 Andrew Haley <aph@redhat.com>
1199 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1202 2007-08-06 Andrew Haley <aph@redhat.com>
1205 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1206 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1209 2007-08-02 David Daney <ddaney@avtrex.com>
1211 * testsuite/libffi.call/return_ul.c (main): Define return type as
1212 ffi_arg. Use proper printf conversion specifier.
1214 2007-07-30 Andrew Haley <aph@redhat.com>
1217 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1218 signed/unsigned int8/16.
1219 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1221 Remove code to pop args from the stack after call.
1222 Special-case signed/unsigned int8/16.
1223 * testsuite/libffi.call/return_sc.c (main): Revert.
1225 2007-07-26 Richard Guenther <rguenther@suse.de>
1228 * testsuite/libffi.call/return_sc.c (main): Verify call
1229 result as signed char, not ffi_arg.
1231 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1233 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1234 * configure: Regenerate.
1236 2007-07-11 David Daney <ddaney@avtrex.com>
1238 * src/mips/ffi.c: Don't include sys/cachectl.h.
1239 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1242 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1244 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1245 from (ffi_prep_closure): ... this.
1246 (FFI_INIT_TRAMPOLINE): Adjust.
1248 2005-12-31 Phil Blundell <pb@reciva.com>
1250 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1251 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1252 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1253 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1254 (FFI_CLOSURES): Enable closure support.
1256 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1258 * testsuite/libffi.call/cls_multi_ushort.c,
1259 testsuite/libffi.call/cls_align_uint16.c,
1260 testsuite/libffi.call/nested_struct1.c,
1261 testsuite/libffi.call/nested_struct3.c,
1262 testsuite/libffi.call/cls_7_1_byte.c,
1263 testsuite/libffi.call/cls_double.c,
1264 testsuite/libffi.call/nested_struct5.c,
1265 testsuite/libffi.call/nested_struct7.c,
1266 testsuite/libffi.call/cls_sint.c,
1267 testsuite/libffi.call/nested_struct9.c,
1268 testsuite/libffi.call/cls_20byte1.c,
1269 testsuite/libffi.call/cls_multi_sshortchar.c,
1270 testsuite/libffi.call/cls_align_sint64.c,
1271 testsuite/libffi.call/cls_3byte2.c,
1272 testsuite/libffi.call/cls_multi_schar.c,
1273 testsuite/libffi.call/cls_multi_uchar.c,
1274 testsuite/libffi.call/cls_19byte.c,
1275 testsuite/libffi.call/cls_9byte1.c,
1276 testsuite/libffi.call/cls_align_float.c,
1277 testsuite/libffi.call/closure_fn1.c,
1278 testsuite/libffi.call/problem1.c,
1279 testsuite/libffi.call/closure_fn3.c,
1280 testsuite/libffi.call/cls_sshort.c,
1281 testsuite/libffi.call/closure_fn5.c,
1282 testsuite/libffi.call/cls_align_double.c,
1283 testsuite/libffi.call/cls_2byte.c,
1284 testsuite/libffi.call/nested_struct.c,
1285 testsuite/libffi.call/nested_struct10.c,
1286 testsuite/libffi.call/cls_4byte.c,
1287 testsuite/libffi.call/cls_6byte.c,
1288 testsuite/libffi.call/cls_8byte.c,
1289 testsuite/libffi.call/cls_multi_sshort.c,
1290 testsuite/libffi.call/cls_align_uint32.c,
1291 testsuite/libffi.call/cls_align_sint16.c,
1292 testsuite/libffi.call/cls_float.c,
1293 testsuite/libffi.call/cls_20byte.c,
1294 testsuite/libffi.call/cls_5_1_byte.c,
1295 testsuite/libffi.call/nested_struct2.c,
1296 testsuite/libffi.call/cls_24byte.c,
1297 testsuite/libffi.call/nested_struct4.c,
1298 testsuite/libffi.call/nested_struct6.c,
1299 testsuite/libffi.call/cls_64byte.c,
1300 testsuite/libffi.call/nested_struct8.c,
1301 testsuite/libffi.call/cls_uint.c,
1302 testsuite/libffi.call/cls_multi_ushortchar.c,
1303 testsuite/libffi.call/cls_schar.c,
1304 testsuite/libffi.call/cls_uchar.c,
1305 testsuite/libffi.call/cls_align_uint64.c,
1306 testsuite/libffi.call/cls_ulonglong.c,
1307 testsuite/libffi.call/cls_align_longdouble.c,
1308 testsuite/libffi.call/cls_1_1byte.c,
1309 testsuite/libffi.call/cls_12byte.c,
1310 testsuite/libffi.call/cls_3_1byte.c,
1311 testsuite/libffi.call/cls_3byte1.c,
1312 testsuite/libffi.call/cls_4_1byte.c,
1313 testsuite/libffi.call/cls_6_1_byte.c,
1314 testsuite/libffi.call/cls_16byte.c,
1315 testsuite/libffi.call/cls_18byte.c,
1316 testsuite/libffi.call/closure_fn0.c,
1317 testsuite/libffi.call/cls_9byte2.c,
1318 testsuite/libffi.call/closure_fn2.c,
1319 testsuite/libffi.call/closure_fn4.c,
1320 testsuite/libffi.call/cls_ushort.c,
1321 testsuite/libffi.call/closure_fn6.c,
1322 testsuite/libffi.call/cls_5byte.c,
1323 testsuite/libffi.call/cls_align_pointer.c,
1324 testsuite/libffi.call/cls_7byte.c,
1325 testsuite/libffi.call/cls_align_sint32.c,
1326 testsuite/libffi.special/unwindtest_ffi_call.cc,
1327 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1329 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1331 * aclocal.m4: Regenerated.
1333 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1335 * configure: Regenerate.
1337 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1339 * Makefile.in: Regenerate.
1340 * configure: Regenerate.
1341 * aclocal.m4: Regenerate.
1342 * include/Makefile.in: Regenerate.
1343 * testsuite/Makefile.in: Regenerate.
1345 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1347 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1348 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1349 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1350 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1351 (FFI_CLOSURES): Enable closure support.
1353 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1355 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1356 * configure: Regenerate.
1357 * fficonfig.h.in: Regenerate.
1358 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1359 CFI_OFFSET,CFI_DEF_CFA): New macros.
1360 (ffi_call_SYSV): Add callframe annotation.
1362 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1364 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1365 numerous test suite failures.
1366 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1368 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1370 * Makefile.am (EXTRA_DIST): Bring up to date.
1371 * Makefile.in: Regenerate.
1372 * src/frv/eabi.S: Remove RCS keyword.
1374 2007-04-06 Richard Henderson <rth@redhat.com>
1376 * configure.ac: Tidy target case.
1377 (HAVE_LONG_DOUBLE): Allow the target to override.
1378 * configure: Regenerate.
1379 * include/ffi.h.in: Don't define ffi_type_foo if
1380 LIBFFI_HIDE_BASIC_TYPES is defined.
1381 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1383 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1384 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1385 (ffi_type_longdouble): Special case for Alpha. Don't define
1386 if long double == double.
1388 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1389 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1390 (ffi_call, ffi_closure_osf_inner): Likewise.
1391 (ffi_closure_osf_inner): Likewise. Mark hidden.
1392 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1393 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1394 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1395 (load_table): Handle 128-bit long double.
1397 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1399 2007-04-06 Tom Tromey <tromey@redhat.com>
1402 * README: Fixed bug in example.
1404 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1406 * src/closures.c: Include sys/statfs.h.
1407 (_GNU_SOURCE): Define on Linux.
1408 (FFI_MMAP_EXEC_SELINUX): Define.
1409 (selinux_enabled): New variable.
1410 (selinux_enabled_check): New function.
1411 (is_selinux_enabled): Define.
1414 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1416 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1417 Use 'volatile float sum' to create sum of floats to avoid false
1418 negative due to excess precision on ix86 targets.
1421 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1423 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1425 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1427 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1429 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1430 (ffi_prep_closure_loc): New.
1431 (ffi_prep_raw_closure_loc): New.
1432 (ffi_prep_java_raw_closure_loc): New.
1433 * src/closures.c: New file.
1434 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1435 Replace sflags with exec_offset.
1436 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1437 sub_segment_exec_offset): New macros.
1438 (get_segment_flags, set_segment_flags, check_segment_merge): New
1440 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1441 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1442 destroy_mspace): Use new macros.
1443 (sys_alloc): Silence warning.
1444 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1445 * Makefile.in: Rebuilt.
1446 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1447 terms of ffi_prep_closure_loc.
1448 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1450 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1452 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1454 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1455 the renamed version.
1456 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1457 (ffi_prep_closure): ... this.
1458 * src/pa/ffi.c: Likewise.
1459 * src/cris/ffi.c: Likewise. Adjust.
1460 * src/frv/ffi.c: Likewise.
1461 * src/ia64/ffi.c: Likewise.
1462 * src/mips/ffi.c: Likewise.
1463 * src/powerpc/ffi_darwin.c: Likewise.
1464 * src/s390/ffi.c: Likewise.
1465 * src/sh/ffi.c: Likewise.
1466 * src/sh64/ffi.c: Likewise.
1467 * src/sparc/ffi.c: Likewise.
1468 * src/x86/ffi64.c: Likewise.
1469 * src/x86/ffi.c: Likewise.
1470 (FFI_INIT_TRAMPOLINE): Adjust.
1471 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1472 (ffi_prep_raw_closure): ... this.
1473 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1474 (ffi_prep_closure): ... this.
1475 (flush_icache): Adjust.
1477 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1479 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1482 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1484 * Makefile.am: Add dummy install-pdf target.
1485 * Makefile.in: Regenerate
1487 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1489 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1490 ffi_closure_helper_SYSV): Add long double handling.
1492 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1494 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1495 immediately after bctrl instruction.
1497 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1499 * Makefile.am (all-recursive, install-recursive,
1500 mostlyclean-recursive, clean-recursive, distclean-recursive,
1501 maintainer-clean-recursive): Add missing targets.
1502 * Makefile.in: Rebuilt.
1504 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1506 * configure.ac: Add TARGET for x86_64-*-darwin*.
1507 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1509 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1510 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1511 * configure: Regenerate.
1512 * Makefile.in: Regenerate.
1513 * include/Makefile.in: Regenerate.
1514 * testsuite/Makefile.in: Regenerate.
1515 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1518 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1520 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1523 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1525 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1526 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1528 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1529 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1531 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1532 Sandro Tolaini <tolaini@libero.it>
1534 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1536 * configure: Regenerated.
1537 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1538 (EXTRA_DIST): Add src/x86/darwin.S.
1539 * Makefile.in: Regenerated.
1540 * include/Makefile.in: Regenerated.
1541 * testsuite/Makefile.in: Regenerated.
1543 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1544 X86_WIN32, and additionally align stack to 16 bytes.
1545 * src/x86/darwin.S: New, based on sysv.S.
1546 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1548 2006-09-12 David Daney <ddaney@avtrex.com>
1551 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1552 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1553 * aclocal.m4: Regenerated for automake 1.9.6.
1554 * Makefile.in: Regenerated.
1555 * include/Makefile.in: Regenerated.
1556 * testsuite/Makefile.in: Regenerated.
1558 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1560 * include/ffi_common.h (struct): Revert accidental commit.
1562 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1564 * include/ffi_common.h: Remove lint directives.
1565 * include/ffi.h.in: Likewise.
1567 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1569 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1570 for 32-bit architectures.
1571 * testsuite/libffi.call/return_ul.c: New test case.
1573 2006-07-19 David Daney <ddaney@avtrex.com>
1575 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1576 xfail remains for mips64.
1578 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1580 * Makefile.am: Add install-html target. Add install-html to .PHONY
1581 * Makefile.in: Regenerate.
1582 * aclocal.m4: Regenerate.
1583 * include/Makefile.in: Regenerate.
1584 * testsuite/Makefile.in: Regenerate.
1586 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1588 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1591 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1593 * README: Remove notice about 'Crazy Comments'.
1594 * src/debug.c: Remove lint directives. Cleanup white spaces.
1595 * src/java_raw_api.c: Likewise.
1596 * src/prep_cif.c: Likewise.
1597 * src/raw_api.c: Likewise.
1598 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1600 * src/arm/ffi.c: Remove lint directives.
1601 * src/m32r/ffi.c: Likewise.
1602 * src/pa/ffi.c: Likewise.
1603 * src/powerpc/ffi.c: Likewise.
1604 * src/powerpc/ffi_darwin.c: Likewise.
1605 * src/sh/ffi.c: Likewise.
1606 * src/sh64/ffi.c: Likewise.
1607 * src/x86/ffi.c: Likewise.
1608 * testsuite/libffi.call/float2.c: Likewise.
1609 * testsuite/libffi.call/promotion.c: Likewise.
1610 * testsuite/libffi.call/struct1.c: Likewise.
1612 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1614 * src/pa/hpux32.S: Correct unwind offset calculation for
1616 * src/pa/linux.S: Likewise.
1618 2006-04-12 James E Wilson <wilson@specifix.com>
1621 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1622 (hfa_type_load): Call stf_spill.
1623 (hfa_type_store): Call ldf_fill.
1624 (ffi_call): Adjust calls to above routines. Add local temps for
1627 2006-04-10 Matthias Klose <doko@debian.org>
1629 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1630 directory names containing underscores.
1632 2006-04-07 James E Wilson <wilson@specifix.com>
1634 * testsuite/libffi.call/float4.c: New testcase.
1636 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1637 Andreas Tobler <a.tobler@schweiz.ch>
1639 * Makefile.am: Add PA_HPUX port.
1640 * Makefile.in: Regenerate.
1641 * include/Makefile.in: Likewise.
1642 * testsuite/Makefile.in: Likewise.
1643 * configure.ac: Add PA_HPUX rules.
1644 * configure: Regenerate.
1645 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1646 Add PA_HPUX and PA64_HPUX.
1647 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1648 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1649 (FFI_TYPE_SMALL_STRUCT2): Define.
1650 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1651 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1652 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1653 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1654 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1655 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1656 * src/pa/ffi.c (ROUND_DOWN): Delete.
1657 (fldw, fstw, fldd, fstd): Use '__asm__'.
1658 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1659 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1660 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1661 Simplify incrementing of stack slot variable. Change type of local
1662 'n' to unsigned int.
1663 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1665 (ffi_prep_cif_machdep): Likewise.
1666 (ffi_call): Likewise.
1667 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1668 return type to ffi_status. Simplify incrementing of stack slot
1669 variable. Only copy floating point argument registers when PA_LINUX
1670 is true. Reformat debug statement.
1671 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1672 FFI_TYPE_SMALL_STRUCT8.
1673 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1675 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1676 Add nops to cache flush. Add trampoline for PA_HPUX.
1677 * src/pa/hpux32.S: New file.
1678 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1679 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1680 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1681 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1682 argument type checks so that common argument types appear first.
1683 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1684 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1686 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1688 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1689 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1690 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1691 FFI_TYPE_LONGDOUBLE.
1692 (ffi_prep_args64): Assert using IBM extended double.
1693 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1694 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1695 (ffi_call): Handle FFI_LINUX.
1696 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1697 gpr3 return pointer as for struct return. Handle FFI_LINUX
1698 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1700 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1701 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1702 Don't use r6 as pointer to results, instead use sp offset. Don't
1703 make a special call to load lr with case table address, instead
1704 use offset from previous call.
1705 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1706 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1709 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1711 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1712 passed with FP registers correctly.
1713 (ffi_closure_helper_SYSV): Likewise.
1714 * src/sh64/sysv.S: Likewise.
1716 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1718 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1719 args and userdata unused.
1720 (closure_test_fn1): Mark cif and userdata unused.
1721 (main): Remove unused res.
1723 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1725 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1726 -O2, -O3, -Os and the warning flags -W -Wall.
1727 * testsuite/libffi.special/special.exp: Likewise.
1728 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1729 unused parameter unused for gcc or else do nothing.
1730 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1731 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1732 and userdata unused.
1733 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1734 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1735 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1736 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1737 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1738 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1739 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1740 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1741 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1742 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1743 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1744 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1745 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1746 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1747 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1748 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1749 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1750 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1751 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1752 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1753 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1754 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1755 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1756 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1758 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1760 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1762 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1763 void* to avoid compiler warning.
1765 (cls_struct_align_gn): Mark cif and userdata unused.
1766 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1768 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1770 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1772 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1774 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1776 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1777 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1778 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1780 (main): Cast res_call to silence gcc.
1781 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1783 (main): Cast res_call to silence gcc.
1784 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1786 (main): Cast res_call to silence gcc.
1787 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1789 (main): Cast res_call to silence gcc.
1790 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1792 (main): Cast res_call to silence gcc.
1793 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1795 (main): Cast res_call to silence gcc.
1796 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1798 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1799 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1801 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1802 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1804 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1805 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1807 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1808 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1810 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1811 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1812 and userdata unused.
1813 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1815 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1816 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1817 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1818 Cleanup white spaces.
1819 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1820 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1821 cif and userdata unused.
1822 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1824 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1825 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1826 formatters to silence gcc.
1827 (B_gn): Mark cif and userdata unused.
1828 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1830 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1831 (B_gn): Mark cif and userdata unused.
1832 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1834 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1835 (B_gn): Mark cif and userdata unused.
1836 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1838 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1839 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1840 * testsuite/libffi.call/problem1.c (stub): Likewise.
1841 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1843 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1844 in the last commit for this test case in the test case itself.
1845 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1847 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1848 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1849 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1850 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1851 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1852 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1854 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1856 * src/sh/sysv.S: Fix register numbers in the FDE for
1859 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1861 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1862 declaration to avoid a false negative on ix86. See PR323.
1864 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1866 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1867 and cast integer to void * if needed. Update the pointer to
1868 the FP register saved area correctly.
1870 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1872 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1874 * testsuite/libffi.call/nested_struct4.c: Likewise.
1876 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1878 * testsuite/libffi.call/return_dbl.c: New test case.
1879 * testsuite/libffi.call/return_dbl1.c: Likewise.
1880 * testsuite/libffi.call/return_dbl2.c: Likewise.
1881 * testsuite/libffi.call/return_fl.c: Likewise.
1882 * testsuite/libffi.call/return_fl1.c: Likewise.
1883 * testsuite/libffi.call/return_fl2.c: Likewise.
1884 * testsuite/libffi.call/return_fl3.c: Likewise.
1885 * testsuite/libffi.call/closure_fn6.c: Likewise.
1887 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1889 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1890 here to be used by other test cases too.
1892 * testsuite/libffi.call/nested_struct10.c: New test case.
1893 * testsuite/libffi.call/nested_struct9.c: Likewise.
1894 * testsuite/libffi.call/nested_struct8.c: Likewise.
1895 * testsuite/libffi.call/nested_struct7.c: Likewise.
1896 * testsuite/libffi.call/nested_struct6.c: Likewise.
1897 * testsuite/libffi.call/nested_struct5.c: Likewise.
1898 * testsuite/libffi.call/nested_struct4.c: Likewise.
1900 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1902 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1903 * configure: Rebuilt.
1905 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1907 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1908 instead do the shifting inline.
1909 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1910 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1911 and 8 byte structs, for the remaining struct sizes don't call
1912 __lshrdi3, instead do the shifting inline.
1914 2005-12-07 Thiemo Seufer <ths@networkno.de>
1916 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1917 missing parentheses.
1918 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1919 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1920 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1921 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1922 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1924 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1926 (ffi_prep_closure): Improve trampoline instruction scheduling.
1927 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1929 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1931 * src/powerpc/ffi.c: Formatting.
1932 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1933 (ffi_prep_args64): Likewise.
1935 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1937 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1938 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1940 2005-09-26 Tom Tromey <tromey@redhat.com>
1942 * testsuite/libffi.call/float1.c (value_type): New typedef.
1943 (CANARY): New define.
1944 (main): Check for result buffer overflow.
1945 * src/powerpc/linux64.S: Handle linux64 long double returns.
1946 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1947 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1949 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1952 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1954 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1956 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1958 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1959 (AH_BOTTOM): Add FFI_HIDDEN definition.
1960 * configure: Rebuilt.
1961 * fficonfig.h.in: Rebuilt.
1962 * src/powerpc/ffi.c (hidden): Remove.
1963 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1964 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1965 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1966 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1967 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1968 add FFI_HIDDEN to its prototype.
1969 (ffi_closure_SYSV_inner): New.
1970 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1971 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1973 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1976 * configure: Rebuilt.
1977 * configure.ac: Handle i*86-*-gnu*.
1979 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1981 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1982 DW_CFA_offset_extended_sf rather than
1983 DW_CFA_GNU_negative_offset_extended.
1984 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1986 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1988 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1990 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1991 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1992 partially on register.
1993 (ffi_closure_helper_SYSV): Likewise.
1994 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1996 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1998 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
2000 * src/sh64/ffi.c (simple_type): Remove.
2001 (return_type): Handle small structures correctly.
2002 (ffi_prep_args): Likewise.
2003 (ffi_call): Likewise.
2004 (ffi_closure_helper_SYSV): Likewise.
2005 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
2006 Emit position independent code if PIC and remove wrong datalabel
2007 prefixes from EH data.
2009 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
2011 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
2012 * Makefile.in: Regenerate.
2013 * include/Makefile.in: Likewise.
2014 * testsuite/Makefile.in: Likewise.
2015 * configure.ac: Add POWERPC_FREEBSD rules.
2016 * configure: Regenerate.
2017 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
2018 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
2019 * src/powerpc/ffi.c: Add flags to handle small structure returns
2021 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
2023 (ffi_closure_helper_SYSV): Likewise.
2024 * src/powerpc/ppc_closure.S: Add return types for small structures.
2025 * src/powerpc/sysv.S: Add bits to handle small structures for
2028 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
2030 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
2031 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
2032 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
2034 2005-07-05 Randolph Chung <tausq@debian.org>
2036 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
2037 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
2038 structures. Kill compilation warnings.
2039 (ffi_closure_inner_LINUX): Print return values as hex in debug
2040 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
2041 Properly handle 5-7 byte structure returns.
2042 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
2043 (FFI_TYPE_SMALL_STRUCT2): Remove.
2044 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
2045 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
2046 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
2047 (checksmst1, checksmst2): Remove.
2048 (checksmst3): Optimize handling of 3-byte struct returns.
2049 (checksmst567): Properly handle 5-7 byte struct returns.
2051 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2054 * src/mips/n32.S: Enforce PIC code.
2055 * src/mips/o32.S: Likewise.
2057 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2059 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
2060 * configure: Regenerate.
2062 2005-06-01 Alan Modra <amodra@bigpond.net.au>
2064 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
2065 to call ffi_closure_helper_SYSV. Append @local instead.
2066 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
2068 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
2070 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
2071 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
2072 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
2073 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
2074 include/Makefile.in, testsuite/Makefile.in: Regenerate.
2076 2005-05-09 Mike Stump <mrs@apple.com>
2078 * configure: Regenerate.
2080 2005-05-08 Richard Henderson <rth@redhat.com>
2083 * src/alpha/osf.S: Update unwind into to match code.
2085 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
2086 Richard Henderson <rth@redhat.com>
2088 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
2090 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
2091 (ffi_prep_closure): Set carry bit if sse-used flag set.
2092 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
2093 Only load sse registers if ssecount non-zero.
2094 (ffi_closure_unix64): Only save sse registers if carry set on entry.
2096 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
2098 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
2099 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
2100 * configure: Regenerate.
2102 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
2104 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
2105 have Tcl8.3-compatible intermediate variable.
2107 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
2108 Hans-Peter Nilsson <hp@axis.com>
2110 * Makefile.am: Add CRIS support.
2111 * configure.ac: Likewise.
2112 * Makefile.in, configure, testsuite/Makefile.in,
2113 include/Makefile.in: Regenerate.
2114 * src/cris: New directory.
2115 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
2116 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
2118 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
2119 \r?\n in output tests.
2121 2005-04-12 Mike Stump <mrs@apple.com>
2123 * configure: Regenerate.
2125 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
2127 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
2129 2005-03-30 Steve Ellcey <sje@cup.hp.com>
2131 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
2133 * src/ia64/unix.S (ffi_closure_unix): Extend gp
2134 to 64 bits in ILP32 mode.
2135 Load 64 bits even for short data.
2137 2005-03-23 Mike Stump <mrs@apple.com>
2139 * src/powerpc/darwin.S: Update for -m64 multilib.
2140 * src/powerpc/darwin_closure.S: Likewise.
2142 2005-03-21 Zack Weinberg <zack@codesourcery.com>
2144 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
2145 Do not set tool_include_dir.
2146 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
2148 * include/Makefile.am: Set gcc_version and toollibffidir.
2149 * include/Makefile.in: Regenerate.
2151 2005-02-22 Andrew Haley <aph@redhat.com>
2153 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2154 odd-numbered register pairs for 64-bit integer types.
2156 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2159 * testsuite/libffi.call/return_ll1.c: New test case.
2161 2005-02-11 Janis Johnson <janis187@us.ibm.com>
2163 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2164 * testsuite/libffi.call/float.c: Ditto.
2165 * testsuite/libffi.call/float2.c: Ditto.
2166 * testsuite/libffi.call/float3.c: Ditto.
2168 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2170 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2172 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2174 * testsuite/libffi.special/special.exp (cxx_options): Add
2177 2004-12-31 Richard Henderson <rth@redhat.com>
2179 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2180 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2181 offset parameters with a type parameter; deduce size and structure
2182 alignment. Update all users.
2184 2004-12-31 Richard Henderson <rth@redhat.com>
2186 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2187 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
2188 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
2189 into ffi_prep_closure.
2190 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2193 2004-12-27 Richard Henderson <rth@redhat.com>
2195 * src/x86/unix64.S: Fix typo in unwind info.
2197 2004-12-25 Richard Henderson <rth@redhat.com>
2199 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2200 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2201 (merge_classes): Check for it.
2203 (classify_argument): Pass byte_offset by value; perform all updates
2205 (examine_argument): Add classes argument; handle
2206 X86_64_COMPLEX_X87_CLASS.
2207 (ffi_prep_args): Merge into ...
2208 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2209 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2210 (ffi_fill_return_value): Remove.
2211 (ffi_prep_closure): Remove dead assert.
2212 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2213 Rewrite to use struct register_args instead of va_list. Create
2214 flags for handling structure returns.
2215 * src/x86/unix64.S: Remove dead strings.
2216 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2217 stack frame with ffi_call. Handle structure returns properly.
2218 (float2sse, floatfloat2sse, double2sse): Remove.
2219 (sse2float, sse2double, sse2floatfloat): Remove.
2220 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2221 to handle structure returns properly.
2223 2004-12-08 David Edelsohn <edelsohn@gnu.org>
2225 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2227 * Makefile.in: Regenerated.
2229 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2231 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2232 * configure, aclocal.m4, Makefile.in: Regenerate.
2233 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2235 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2237 * configure: Regenerate for libtool change.
2239 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2241 * configure: Regenerate for libtool reversion.
2243 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2245 * configure: Regenerate for libtool change.
2247 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2249 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2251 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2253 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2254 of jal. Use an absolute encoding for the frame information.
2256 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2258 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2259 * acinclude.m4: Delete logic for sincludes.
2260 * aclocal.m4, Makefile.in, configure: Regenerate.
2261 * include/Makefile: Likewise.
2262 * testsuite/Makefile: Likewise.
2264 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2266 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2267 on a 8-byte boundary.
2268 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2270 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2272 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2273 long long values. Round stack allocation to a multiple of 8 bytes
2274 for ATPCS compatibility.
2275 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2276 names. Handle returning long long types. Add Thumb and interworking
2277 support. Improve soft-float code.
2279 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2281 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2282 (libffi_exit): New function.
2283 (libffi_init): Build the testglue wrapper if needed.
2285 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2288 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2290 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2292 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2294 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2296 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2297 * testsuite/libffi.call/float3.c: New test case.
2299 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2301 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2302 the function returning a structure pointed with R2.
2303 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2304 the structure return value if T bit set. Emit position
2305 independent code and EH data if PIC.
2307 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2309 * Makefile.am: Add m32r support.
2310 * configure.ac: Likewise.
2311 * Makefile.in: Regenerate.
2312 * confiugre: Regenerate.
2313 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2314 (uint64, sint64, double, longdouble)
2315 * src/m32r: New directory.
2316 * src/m32r/ffi.c: New file.
2317 * src/m32r/sysv.S: Likewise.
2318 * src/m32r/ffitarget.h: Likewise.
2320 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2322 * testsuite/libffi.call/negint.c: New test case.
2324 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2327 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2328 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2329 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2332 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2334 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2335 * testsuite/libffi.call/promotion.c: Likewise.
2336 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2338 * testsuite/libffi.call/return_sc.c: Likewise.
2339 * testsuite/libffi.call/return_uc.c: Likewise.
2341 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2343 * src/powerpc/darwin.S: Fix comments and identation.
2344 * src/powerpc/darwin_closure.S: Likewise.
2346 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2348 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2349 (ffi_prep_args): Handle longdouble arguments.
2350 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2352 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2353 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2355 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2356 * src/types.c: Defined longdouble size and alignment for darwin.
2358 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2360 * src/powerpc/aix.S: Remove whitespaces.
2361 * src/powerpc/aix_closure.S: Likewise.
2362 * src/powerpc/asm.h: Likewise.
2363 * src/powerpc/ffi.c: Likewise.
2364 * src/powerpc/ffitarget.h: Likewise.
2365 * src/powerpc/linux64.S: Likewise.
2366 * src/powerpc/linux64_closure.S: Likewise.
2367 * src/powerpc/ppc_closure.S: Likewise.
2368 * src/powerpc/sysv.S: Likewise.
2370 2004-08-30 Anthony Green <green@redhat.com>
2372 * Makefile.am: Add frv support.
2373 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2374 * configure.ac: Read configure.host.
2375 * configure.in: Read configure.host.
2376 * configure.host: New file. frv-elf needs libgloss.
2377 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2378 alignment. This is needed to frv and shouldn't harm the others.
2379 * include/ffi_common.h (ALIGN_DOWN): New macro.
2380 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2382 2004-08-24 David Daney <daney@avtrex.com>
2384 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2385 * testsuite/libffi.call/closure_fn1.c: Likewise.
2386 * testsuite/libffi.call/closure_fn2.c Likewise.
2387 * testsuite/libffi.call/closure_fn3.c: Likewise.
2388 * testsuite/libffi.call/closure_fn4.c: Likewise.
2389 * testsuite/libffi.call/closure_fn5.c: Likewise.
2390 * testsuite/libffi.call/cls_18byte.c: Likewise.
2391 * testsuite/libffi.call/cls_19byte.c: Likewise.
2392 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2393 * testsuite/libffi.call/cls_20byte.c: Likewise.
2394 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2395 * testsuite/libffi.call/cls_24byte.c: Likewise.
2396 * testsuite/libffi.call/cls_2byte.c: Likewise.
2397 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2398 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2399 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2400 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2401 * testsuite/libffi.call/cls_4byte.c: Likewise.
2402 * testsuite/libffi.call/cls_64byte.c: Likewise.
2403 * testsuite/libffi.call/cls_6byte.c: Likewise.
2404 * testsuite/libffi.call/cls_7byte.c: Likewise.
2405 * testsuite/libffi.call/cls_8byte.c: Likewise.
2406 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2407 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2408 * testsuite/libffi.call/cls_align_double.c: Likewise.
2409 * testsuite/libffi.call/cls_align_float.c: Likewise.
2410 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2411 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2412 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2413 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2414 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2415 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2416 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2417 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2418 * testsuite/libffi.call/cls_double.c: Likewise.
2419 * testsuite/libffi.call/cls_float.c: Likewise.
2420 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2421 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2422 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2423 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2424 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2425 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2426 * testsuite/libffi.call/cls_schar.c: Likewise.
2427 * testsuite/libffi.call/cls_sint.c: Likewise.
2428 * testsuite/libffi.call/cls_sshort.c: Likewise.
2429 * testsuite/libffi.call/cls_uchar.c: Likewise.
2430 * testsuite/libffi.call/cls_uint.c: Likewise.
2431 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2432 * testsuite/libffi.call/cls_ushort.c: Likewise.
2433 * testsuite/libffi.call/nested_struct.c: Likewise.
2434 * testsuite/libffi.call/nested_struct1.c: Likewise.
2435 * testsuite/libffi.call/nested_struct2.c: Likewise.
2436 * testsuite/libffi.call/nested_struct3.c: Likewise.
2437 * testsuite/libffi.call/problem1.c: Likewise.
2438 * testsuite/libffi.special/unwindtest.cc: Likewise.
2439 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2441 * testsuite/libffi.call/cls_16byte.c: Likewise.
2442 * testsuite/libffi.call/cls_5byte.c: Likewise.
2444 2004-08-23 David Daney <daney@avtrex.com>
2447 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2448 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2449 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2450 parameters and return types.
2451 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2452 (ffi_prep_closure): Ditto.
2453 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2454 alignment calculations.
2455 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2456 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2458 2004-08-14 Casey Marshall <csm@gnu.org>
2460 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2461 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2462 integer (O32 ABI only).
2463 (ffi_prep_closure): new function.
2464 (ffi_closure_mips_inner_O32): new function.
2465 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2466 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2467 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2468 64 bit integers correctly.
2469 (ffi_closure_O32): new function.
2470 Added DWARF-2 unwind info for both functions.
2472 2004-08-10 Andrew Haley <aph@redhat.com>
2474 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2476 2004-08-01 Robert Millan <robertmh@gnu.org>
2478 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2479 * configure: Regenerate.
2481 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2483 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2484 and mmap() explicitly instead of relying on preset autoconf cache
2486 * aclocal.m4: Regenerate.
2487 * configure: Regenerate.
2489 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2491 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2492 (ffi_check_float_struct): Remove unused prototype.
2494 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2496 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2497 character on Darwin, use '\n\t' instead.
2499 2004-06-26 Matthias Klose <doko@debian.org>
2501 * libtool-version: Fix typo in revision/age.
2503 2004-06-17 Matthias Klose <doko@debian.org>
2505 * libtool-version: New.
2506 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2507 * Makefile.in: Regenerate.
2509 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2511 * Makefile.am: Remove useless multilib rules.
2512 * Makefile.in: Regenerate.
2513 * aclocal.m4: Regenerate with automake 1.8.5.
2514 * configure.ac: Remove useless multilib configury.
2515 * configure: Regenerate.
2517 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2519 * .cvsignore: New file.
2521 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2523 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2525 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2528 2004-06-06 Sean McNeil <sean@mcneil.com>
2530 * configure.ac: Add x86_64-*-freebsd* support.
2531 * configure: Regenerate.
2533 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2536 * configure.ac: Test for existence of mmap and sys/mman.h before
2537 checking blacklist. Fix suggested by Jim Wilson.
2538 * configure: Regenerate.
2540 2004-04-26 Matt Austern <austern@apple.com>
2542 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2544 * src/powerpc/darwin_closure.S: Likewise.
2546 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2548 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2549 error. Reported by Thomas Heller <theller@python.net>.
2550 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2551 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2553 2004-03-20 Matthias Klose <doko@debian.org>
2555 * src/pa/linux.S: Fix typo.
2557 2004-03-19 Matthias Klose <doko@debian.org>
2559 * Makefile.am: Update.
2560 * Makefile.in: Regenerate.
2561 * src/pa/ffi.h.in: Remove.
2562 * src/pa/ffitarget.h: New file.
2564 2004-02-10 Randolph Chung <tausq@debian.org>
2566 * Makefile.am: Add PA support.
2567 * Makefile.in: Regenerate.
2568 * include/Makefile.in: Regenerate.
2569 * configure.ac: Add PA target.
2570 * configure: Regenerate.
2571 * src/pa/ffi.c: New file.
2572 * src/pa/ffi.h.in: Add PA support.
2573 * src/pa/linux.S: New file.
2574 * prep_cif.c: Add PA support.
2576 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2578 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2580 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2581 with ecif->cif->flags.
2582 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2584 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2585 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2586 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2587 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2589 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2591 * configure.in: Rename file to ...
2592 * configure.ac: ... this.
2593 * fficonfig.h.in: Regenerate.
2594 * Makefile.in: Regenerate.
2595 * include/Makefile.in: Regenerate.
2596 * testsuite/Makefile.in: Regenerate.
2598 2004-03-12 Matt Austern <austern@apple.com>
2600 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2601 changes in EH format resulting from addition of linkonce support.
2602 * src/powerpc/darwin_closure.S: Likewise.
2604 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2605 Paolo Bonzini <bonzini@gnu.org>
2607 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2608 Remove VPATH. Remove rules for object files. Remove multilib support.
2609 (AM_CCASFLAGS): Add.
2610 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2611 (AC_PREREQ): Bump version to 2.59.
2612 (AC_INIT): Fill with version info and bug address.
2613 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2614 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2615 De-precious CC so that the right flags are passed down to multilibs.
2616 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2617 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2618 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2619 * configure: Rebuilt.
2620 * aclocal.m4: Likewise.
2621 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2622 * fficonfig.h.in: Likewise.
2624 2004-03-11 Andreas Schwab <schwab@suse.de>
2626 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2627 arguments from fp registers only for the first 8 parameter slots.
2628 Don't convert a float parameter when passed in memory.
2630 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2632 * configure: Regenerate for config/accross.m4 correction.
2634 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2636 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2637 ecif->cif->bytes to bytes.
2638 (ffi_prep_cif_machdep): Add braces around nested if statement.
2640 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2642 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2644 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2645 (ffi_closure_helper_LINUX64): Fix typo.
2646 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2648 * testsuite/libffi.call/float.c: Likewise.
2649 * testsuite/libffi.call/float2.c: Likewise.
2651 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2653 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2654 long double function return and long double arg handling.
2655 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2656 Use "end_pfr" instead of "nf". Correct long double handling.
2658 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2660 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2661 space for long double return value. Adjust stack frame and offsets.
2662 Load f2 long double return.
2664 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2666 * src/types.c: Use 16 byte long double for POWERPC64.
2668 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2670 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2671 when the structure return address is passed in %o0.
2672 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2673 (ffi_v9_layout_struct): Align the field following a nested structure
2674 on a word boundary. Use memmove instead of memcpy.
2675 (ffi_call): Update call to ffi_V9_return_struct.
2676 (ffi_prep_closure): Define 'ctx' only for V8.
2677 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2678 and ffi_closure_sparc_inner_v9.
2679 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2680 Always skip the structure return address. For structures and long
2681 doubles, copy the argument directly.
2682 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2683 if required. Shift the maximum floating-point slot accordingly. For
2684 big structures, copy the argument directly; otherwise, left-justify the
2685 argument and call ffi_v9_layout_struct to lay out the structure on
2687 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2688 (ffi_closure_v8): Pass the structure return address. Update call to
2689 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2690 Skip the 'unimp' insn when returning long doubles and structures.
2691 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2692 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2693 FFI_TYPE_INT handling. Load structures both in integers and
2694 floating-point registers on return.
2695 * README: Update status of the SPARC port.
2697 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2699 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2701 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2703 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2705 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2706 value as of type ffi_arg, not unsigned int.
2708 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2710 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2713 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2715 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2718 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2720 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2723 2003-12-10 Richard Henderson <rth@redhat.com>
2725 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2726 size_t instead of int.
2728 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2730 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2731 * testsuite/libffi.call/many_win32.c (main): Replace variable
2732 int i with unsigned long ul.
2734 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2735 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2736 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2737 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2738 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2739 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2740 * testsuite/libffi.call/cls_align_float.c: Likewise.
2741 * testsuite/libffi.call/cls_align_double.c: Likewise.
2742 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2743 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2745 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2748 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2749 Align arguments to 32 bits.
2751 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2754 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2755 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2756 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2757 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2758 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2759 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2761 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2763 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2765 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2766 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2768 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2770 * Makefile.in: Rebuilt.
2771 * configure: Likewise.
2772 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2775 2003-11-21 Andreas Jaeger <aj@suse.de>
2776 Andreas Tobler <a.tobler@schweiz.ch>
2778 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2779 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2780 * Makefile.in: Rebuilt.
2781 * aclocal.m4: Likewise.
2782 * configure: Likewise.
2783 * fficonfig.h.in: Likewise.
2784 * testsuite/lib/libffi-dg.exp: Add include dir.
2785 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2786 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2787 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2788 for ffi_closure if available.
2789 * testsuite/libffi.call/closure_fn1.c: Likewise.
2790 * testsuite/libffi.call/closure_fn2.c: Likewise.
2791 * testsuite/libffi.call/closure_fn3.c: Likewise.
2792 * testsuite/libffi.call/closure_fn4.c: Likewise.
2793 * testsuite/libffi.call/closure_fn5.c: Likewise.
2794 * testsuite/libffi.call/cls_12byte.c: Likewise.
2795 * testsuite/libffi.call/cls_16byte.c: Likewise.
2796 * testsuite/libffi.call/cls_18byte.c: Likewise.
2797 * testsuite/libffi.call/cls_19byte.c: Likewise.
2798 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2799 * testsuite/libffi.call/cls_20byte.c: Likewise.
2800 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2801 * testsuite/libffi.call/cls_24byte.c: Likewise.
2802 * testsuite/libffi.call/cls_2byte.c: Likewise.
2803 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2804 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2805 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2806 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2807 * testsuite/libffi.call/cls_4byte.c: Likewise.
2808 * testsuite/libffi.call/cls_5byte.c: Likewise.
2809 * testsuite/libffi.call/cls_64byte.c: Likewise.
2810 * testsuite/libffi.call/cls_6byte.c: Likewise.
2811 * testsuite/libffi.call/cls_7byte.c: Likewise.
2812 * testsuite/libffi.call/cls_8byte.c: Likewise.
2813 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2814 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2815 * testsuite/libffi.call/cls_double.c: Likewise.
2816 * testsuite/libffi.call/cls_float.c: Likewise.
2817 * testsuite/libffi.call/cls_schar.c: Likewise.
2818 * testsuite/libffi.call/cls_sint.c: Likewise.
2819 * testsuite/libffi.call/cls_sshort.c: Likewise.
2820 * testsuite/libffi.call/cls_uchar.c: Likewise.
2821 * testsuite/libffi.call/cls_uint.c: Likewise.
2822 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2823 * testsuite/libffi.call/cls_ushort.c: Likewise.
2824 * testsuite/libffi.call/nested_struct.c: Likewise.
2825 * testsuite/libffi.call/nested_struct1.c: Likewise.
2826 * testsuite/libffi.call/nested_struct2.c: Likewise.
2827 * testsuite/libffi.call/nested_struct3.c: Likewise.
2828 * testsuite/libffi.call/problem1.c: Likewise.
2829 * testsuite/libffi.special/unwindtest.cc: Likewise.
2831 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2833 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2835 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2837 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2838 Add -lgcc_s to additional flags.
2840 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2842 * configure.in, include/Makefile.am: PR libgcj/11147, install
2843 the ffitarget.h header file in a gcc versioned and target
2845 * configure: Regenerated.
2846 * Makefile.in, include/Makefile.in: Likewise.
2847 * testsuite/Makefile.in: Likewise.
2849 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2851 * testsuite/libffi.call/closure_fn0.c: Print result and check
2852 with dg-output to make debugging easier.
2853 * testsuite/libffi.call/closure_fn1.c: Likewise.
2854 * testsuite/libffi.call/closure_fn2.c: Likewise.
2855 * testsuite/libffi.call/closure_fn3.c: Likewise.
2856 * testsuite/libffi.call/closure_fn4.c: Likewise.
2857 * testsuite/libffi.call/closure_fn5.c: Likewise.
2858 * testsuite/libffi.call/cls_12byte.c: Likewise.
2859 * testsuite/libffi.call/cls_16byte.c: Likewise.
2860 * testsuite/libffi.call/cls_18byte.c: Likewise.
2861 * testsuite/libffi.call/cls_19byte.c: Likewise.
2862 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2863 * testsuite/libffi.call/cls_20byte.c: Likewise.
2864 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2865 * testsuite/libffi.call/cls_24byte.c: Likewise.
2866 * testsuite/libffi.call/cls_2byte.c: Likewise.
2867 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2868 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2869 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2870 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2871 * testsuite/libffi.call/cls_4byte.c: Likewise.
2872 * testsuite/libffi.call/cls_5byte.c: Likewise.
2873 * testsuite/libffi.call/cls_64byte.c: Likewise.
2874 * testsuite/libffi.call/cls_6byte.c: Likewise.
2875 * testsuite/libffi.call/cls_7byte.c: Likewise.
2876 * testsuite/libffi.call/cls_8byte.c: Likewise.
2877 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2878 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2879 * testsuite/libffi.call/cls_double.c: Likewise.
2880 * testsuite/libffi.call/cls_float.c: Likewise.
2881 * testsuite/libffi.call/cls_schar.c: Likewise.
2882 * testsuite/libffi.call/cls_sint.c: Likewise.
2883 * testsuite/libffi.call/cls_sshort.c: Likewise.
2884 * testsuite/libffi.call/cls_uchar.c: Likewise.
2885 * testsuite/libffi.call/cls_uint.c: Likewise.
2886 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2887 * testsuite/libffi.call/cls_ushort.c: Likewise.
2888 * testsuite/libffi.call/problem1.c: Likewise.
2890 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2893 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2895 * testsuite/libffi.call/cls_9byte2.c: New test case.
2896 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2897 * testsuite/libffi.call/cls_64byte.c: Likewise.
2898 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2899 * testsuite/libffi.call/cls_19byte.c: Likewise.
2900 * testsuite/libffi.call/cls_18byte.c: Likewise.
2901 * testsuite/libffi.call/closure_fn4.c: Likewise.
2902 * testsuite/libffi.call/closure_fn5.c: Likewise.
2903 * testsuite/libffi.call/cls_schar.c: Likewise.
2904 * testsuite/libffi.call/cls_sint.c: Likewise.
2905 * testsuite/libffi.call/cls_sshort.c: Likewise.
2906 * testsuite/libffi.call/nested_struct2.c: Likewise.
2907 * testsuite/libffi.call/nested_struct3.c: Likewise.
2909 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2911 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2912 * testsuite/libffi.call/cls_uchar.c: Likewise.
2913 * testsuite/libffi.call/cls_uint.c: Likewise.
2914 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2915 * testsuite/libffi.call/cls_ushort.c: Likewise.
2916 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2918 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2920 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2923 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2925 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2926 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2928 2003-10-22 David Daney <ddaney@avtrex.com>
2930 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2931 __attribute__((__mode__(__SI__))) and friends.
2933 2003-10-22 Andreas Schwab <schwab@suse.de>
2935 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2937 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2939 * configure.in: AC_LINK_FILES(ffitarget.h).
2940 * configure: Regenerate.
2941 * Makefile.in: Likewise.
2942 * include/Makefile.in: Likewise.
2943 * testsuite/Makefile.in: Likewise.
2944 * fficonfig.h.in: Likewise.
2946 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2947 Richard Henderson <rth@redhat.com>
2949 Avoid that ffi.h includes fficonfig.h.
2951 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2952 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2953 (TARGET_SRC_MIPS_SGI): Removed.
2954 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2955 (MIPS_SGI): Removed.
2956 (CLEANFILES): Removed.
2957 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2959 * acconfig.h: Removed.
2960 * configure.in: Compute sizeofs only for double and long double.
2961 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2962 into AC_DEFINE instead of using acconfig.h. Create
2963 include/ffitarget.h instead of include/fficonfig.h. Rename
2964 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2965 AC_DEFINE EH_FRAME_FLAGS.
2966 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2967 (hack_DATA): Add ffitarget.h.
2968 * include/ffi.h.in: Remove all system specific definitions.
2969 Declare raw API even if it is not installed, why bother?
2970 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2971 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2972 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2973 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2974 * include/ffi_common.h (bool): Do not define.
2975 (ffi_assert): Accept failed assertion.
2976 (ffi_type_test): Return void and accept file/line.
2977 (FFI_ASSERT): Pass stringized failed assertion.
2978 (FFI_ASSERT_AT): New macro.
2979 (FFI_ASSERT_VALID_TYPE): New macro.
2980 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2981 UINT64, SINT64): Define here with gcc's __attribute__ macro
2983 (FLOAT32, ALIGN): Define here instead of in ffi.h
2984 * include/ffi-mips.h: Removed. Its content moved to
2985 src/mips/ffitarget.h after separating assembly and C sections.
2986 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2987 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2988 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2989 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2990 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2991 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2992 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2993 (ffi_type_test): Rewritten.
2994 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2995 FFI_ASSERT_VALID_TYPE.
2996 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2997 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2998 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2999 src/s390/ffitarget.h, src/sh/ffitarget.h,
3000 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
3001 src/x86/ffitarget.h: New files.
3002 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
3003 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
3004 src/powerpc/aix.S, src/powerpc/darwin.S,
3005 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
3006 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
3007 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
3008 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
3009 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
3012 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3014 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
3015 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
3017 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
3019 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
3020 Used when FFI_DEBUG = 1.
3022 2003-10-14 Alan Modra <amodra@bigpond.net.au>
3024 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
3027 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3029 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
3030 FFI_MIPS_O32 for O32 ABI.
3032 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
3034 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
3035 SPARC64. Cleanup whitespaces.
3037 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
3039 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
3040 strongarm, xscale. Cleanup whitespaces.
3041 * testsuite/libffi.call/closure_fn1.c: Likewise.
3042 * testsuite/libffi.call/closure_fn2.c: Likewise.
3043 * testsuite/libffi.call/closure_fn3.c: Likewise.
3044 * testsuite/libffi.call/cls_12byte.c: Likewise.
3045 * testsuite/libffi.call/cls_16byte.c: Likewise.
3046 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3047 * testsuite/libffi.call/cls_20byte.c: Likewise.
3048 * testsuite/libffi.call/cls_24byte.c: Likewise.
3049 * testsuite/libffi.call/cls_2byte.c: Likewise.
3050 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3051 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3052 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3053 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3054 * testsuite/libffi.call/cls_4byte.c: Likewise.
3055 * testsuite/libffi.call/cls_5byte.c: Likewise.
3056 * testsuite/libffi.call/cls_6byte.c: Likewise.
3057 * testsuite/libffi.call/cls_7byte.c: Likewise.
3058 * testsuite/libffi.call/cls_8byte.c: Likewise.
3059 * testsuite/libffi.call/cls_double.c: Likewise.
3060 * testsuite/libffi.call/cls_float.c: Likewise.
3061 * testsuite/libffi.call/cls_uchar.c: Likewise.
3062 * testsuite/libffi.call/cls_uint.c: Likewise.
3063 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3064 * testsuite/libffi.call/cls_ushort.c: Likewise.
3065 * testsuite/libffi.call/nested_struct.c: Likewise.
3066 * testsuite/libffi.call/nested_struct1.c: Likewise.
3067 * testsuite/libffi.call/problem1.c: Likewise.
3068 * testsuite/libffi.special/unwindtest.cc: Likewise.
3069 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
3071 2003-09-18 David Edelsohn <edelsohn@gnu.org>
3073 * src/powerpc/aix.S: Cleanup whitespaces.
3074 * src/powerpc/aix_closure.S: Likewise.
3076 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3078 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
3079 * src/powerpc/darwin_closure.S: Likewise.
3080 * src/powerpc/ffi_darwin.c: Likewise.
3082 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3083 David Edelsohn <edelsohn@gnu.org>
3085 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
3086 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
3088 * src/powerpc/darwin_closure.S: Likewise.
3089 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
3090 according to the Darwin/AIX ABI.
3091 (ffi_prep_cif_machdep): Likewise.
3092 (ffi_closure_helper_DARWIN): Likewise.
3093 Remove the outgoing parameter stack logic. Simplify the evaluation
3094 of the different CASE types.
3095 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
3096 statement in the trampoline code.
3098 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
3100 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
3101 for the register size.
3102 (ffi_closure_helper_SYSV): Handle the structure return value
3104 (ffi_closure_helper_SYSV): Return the appropriate type when
3105 the registers are used for the structure return value.
3106 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
3107 the 64-bit return value. Update copyright years.
3109 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3111 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
3112 srcdir for ffi_mips.h.
3114 2003-09-12 Alan Modra <amodra@bigpond.net.au>
3116 * src/prep_cif.c (initialize_aggregate): Include tail padding in
3118 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
3119 placement of float result.
3120 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
3121 cast of "resp" for big-endian 64 bit machines.
3123 2003-09-11 Alan Modra <amodra@bigpond.net.au>
3125 * src/types.c (double, longdouble): Merge identical SH and ARM
3126 typedefs, and add POWERPC64.
3127 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
3128 struct split over gpr and rest.
3129 (ffi_prep_cif_machdep): Correct intarg_count for structures.
3130 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
3132 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
3134 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
3137 2003-09-09 Alan Modra <amodra@bigpond.net.au>
3139 * configure: Regenerate.
3141 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3143 * Makefile.am: Remove build rules for ffitest.
3144 * Makefile.in: Rebuilt.
3146 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3148 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3149 about implicit declaration of abort().
3151 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3153 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3154 * Makefile.in: Rebuilt.
3155 * configure.in: Add dejagnu test framework.
3156 * configure: Rebuilt.
3158 * testsuite/Makefile.am: New file.
3159 * testsuite/Makefile.in: Built
3160 * testsuite/lib/libffi-dg.exp: New file.
3161 * testsuite/config/default.exp: Likewise.
3162 * testsuite/libffi.call/call.exp: Likewise.
3163 * testsuite/libffi.call/ffitest.h: Likewise.
3164 * testsuite/libffi.call/closure_fn0.c: Likewise.
3165 * testsuite/libffi.call/closure_fn1.c: Likewise.
3166 * testsuite/libffi.call/closure_fn2.c: Likewise.
3167 * testsuite/libffi.call/closure_fn3.c: Likewise.
3168 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3169 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3170 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3171 * testsuite/libffi.call/cls_2byte.c: Likewise.
3172 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3173 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3174 * testsuite/libffi.call/cls_4byte.c: Likewise.
3175 * testsuite/libffi.call/cls_5byte.c: Likewise.
3176 * testsuite/libffi.call/cls_6byte.c: Likewise.
3177 * testsuite/libffi.call/cls_7byte.c: Likewise.
3178 * testsuite/libffi.call/cls_8byte.c: Likewise.
3179 * testsuite/libffi.call/cls_12byte.c: Likewise.
3180 * testsuite/libffi.call/cls_16byte.c: Likewise.
3181 * testsuite/libffi.call/cls_20byte.c: Likewise.
3182 * testsuite/libffi.call/cls_24byte.c: Likewise.
3183 * testsuite/libffi.call/cls_double.c: Likewise.
3184 * testsuite/libffi.call/cls_float.c: Likewise.
3185 * testsuite/libffi.call/cls_uchar.c: Likewise.
3186 * testsuite/libffi.call/cls_uint.c: Likewise.
3187 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3188 * testsuite/libffi.call/cls_ushort.c: Likewise.
3189 * testsuite/libffi.call/float.c: Likewise.
3190 * testsuite/libffi.call/float1.c: Likewise.
3191 * testsuite/libffi.call/float2.c: Likewise.
3192 * testsuite/libffi.call/many.c: Likewise.
3193 * testsuite/libffi.call/many_win32.c: Likewise.
3194 * testsuite/libffi.call/nested_struct.c: Likewise.
3195 * testsuite/libffi.call/nested_struct1.c: Likewise.
3196 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3197 * testsuite/libffi.call/problem1.c: Likewise.
3198 * testsuite/libffi.call/promotion.c: Likewise.
3199 * testsuite/libffi.call/return_ll.c: Likewise.
3200 * testsuite/libffi.call/return_sc.c: Likewise.
3201 * testsuite/libffi.call/return_uc.c: Likewise.
3202 * testsuite/libffi.call/strlen.c: Likewise.
3203 * testsuite/libffi.call/strlen_win32.c: Likewise.
3204 * testsuite/libffi.call/struct1.c: Likewise.
3205 * testsuite/libffi.call/struct2.c: Likewise.
3206 * testsuite/libffi.call/struct3.c: Likewise.
3207 * testsuite/libffi.call/struct4.c: Likewise.
3208 * testsuite/libffi.call/struct5.c: Likewise.
3209 * testsuite/libffi.call/struct6.c: Likewise.
3210 * testsuite/libffi.call/struct7.c: Likewise.
3211 * testsuite/libffi.call/struct8.c: Likewise.
3212 * testsuite/libffi.call/struct9.c: Likewise.
3213 * testsuite/libffi.special/special.exp: New file.
3214 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3215 * testsuite/libffi.special/unwindtest.cc: Likewise.
3218 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3220 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3223 2003-08-02 Alan Modra <amodra@bigpond.net.au>
3225 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3227 (ffi_closure_helper_LINUX64): Likewise.
3228 * src/powerpc/linux64.S: Remove code writing to parm save area.
3229 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3230 address in lr from ffi_closure_helper_LINUX64 call to calculate
3231 table address. Optimize function tail.
3233 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3235 * src/sparc/ffi.c: Handle all floating point registers.
3236 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3238 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3240 * README: Note that libffi is not part of GCC. Update the project
3243 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3245 * src/powerpc/ppc_closure.S: Include ffi.h.
3247 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3249 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3250 Use C style comments.
3252 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3254 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3255 * Makefile.in: Regenerate.
3256 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3257 * configure: Regenerate.
3258 * include/ffi.h.in: Add SHmedia support.
3259 * src/sh64/ffi.c: New file.
3260 * src/sh64/sysv.S: New file.
3262 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3264 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3265 should be read-only.
3266 * configure: Rebuilt.
3267 * fficonfig.h.in: Rebuilt.
3268 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3269 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3270 * src/powerpc/linux64.S: Likewise.
3271 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3272 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3273 if -fpic/-fPIC/-mrelocatable.
3274 * src/powerpc/powerpc_closure.S: Likewise.
3275 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3276 #write in .eh_frame flags.
3277 * src/sparc/v9.S: Likewise.
3278 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3279 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3280 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3282 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3284 Fixes PR bootstrap/10656
3285 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3286 support for .register pseudo-op.
3287 * src/sparc/v8.S: Use it.
3288 * fficonfig.h.in: Rebuilt.
3289 * configure: Rebuilt.
3291 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3293 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3294 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3295 Make it the default on POWERPC64.
3296 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3297 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3298 * configure: Rebuilt.
3299 * src/powerpc/ffi.c (hidden): Define.
3300 (ffi_prep_args_SYSV): Renamed from
3301 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3302 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3303 ASM_NEEDS_REGISTERS64): New.
3304 (ffi_prep_args64): New function.
3305 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3306 (ffi_call): Likewise.
3307 (ffi_prep_closure): Likewise.
3308 (flush_icache): Surround by #ifndef POWERPC64.
3309 (ffi_dblfl): New union type.
3310 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3311 (ffi_closure_helper_LINUX64): New function.
3312 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3314 * src/powerpc/sysv.S: Likewise.
3315 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3316 * src/powerpc/linux64.S: New file.
3317 * src/powerpc/linux64_closure.S: New file.
3318 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3319 src/powerpc/linux64_closure.S.
3320 (TARGET_SRC_POWERPC): Likewise.
3322 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3323 closure_test_fn3): Fix result printing on big-endian 64-bit
3325 (main): Print tst2_arg instead of uninitialized tst2_result.
3327 * src/ffitest.c (main): Hide what closure pointer really points to
3330 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3332 * configure.in (arm-*-netbsdelf*): Add configuration.
3333 (configure): Regenerated.
3335 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3337 * include/Makefile.in: Regenerate.
3339 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3341 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3343 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3344 Receive closure pointer through parameter, read args using
3345 __builtin_dwarf_cfa.
3346 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3348 2003-03-12 Andreas Schwab <schwab@suse.de>
3350 * configure.in: Avoid trailing /. in toolexeclibdir.
3351 * configure: Rebuilt.
3353 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3355 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3357 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3359 * libffi/src/powerpc/darwin_closure.S:
3360 Fix alignement bug, allocate 8 bytes for the result.
3361 * libffi/src/powerpc/aix_closure.S:
3363 * libffi/src/powerpc/ffi_darwin.c:
3364 Update stackframe description for aix/darwin_closure.S.
3366 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3368 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3371 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3372 Andreas Schwab <schwab@suse.de>
3374 * configure.in: Adjust command to source config-ml.in to account
3375 for changes to the libffi_basedir definition.
3376 (libffi_basedir): Remove ${srcdir} from value and include trailing
3379 * configure: Regenerate.
3381 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3383 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3385 2003-01-28 Andrew Haley <aph@redhat.com>
3387 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3388 * src/x86/ffi64.c (ffi_prep_closure): New.
3389 (ffi_closure_UNIX64_inner): New.
3390 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3392 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
3394 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3395 Remove USE_LIBDIR conditional.
3396 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3397 * Makefile.in, configure: Rebuilt.
3399 2003-01027 David Edelsohn <edelsohn@gnu.org>
3401 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3402 * Makefile.in: Regenerate.
3404 2003-01-22 Andrew Haley <aph@redhat.com>
3406 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3409 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3411 * src/powerpc/darwin.S: Add unwind info.
3412 * src/powerpc/darwin_closure.S: Likewise.
3414 2003-01-14 Andrew Haley <aph@redhat.com>
3416 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3417 (ffi_prep_cif_machdep): Likewise.
3418 * src/x86/unix64.S: Add unwind info.
3420 2003-01-14 Andreas Jaeger <aj@suse.de>
3422 * src/ffitest.c (main): Only use ffi_closures if those are
3425 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3427 * libffi/src/ffitest.c
3428 add closure testcases
3430 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3432 * libffi/src/powerpc/ffi.c
3433 fix alignment bug for float (4 byte aligned iso 8 byte)
3435 2003-01-09 Geoffrey Keating <geoffk@apple.com>
3437 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3438 * src/powerpc/darwin.S: Remove RCS version string.
3440 2003-01-03 Jeff Sturm <jsturm@one-point.com>
3442 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3443 * src/ffitest.c (main): Use static storage for closure.
3444 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3445 * src/sparc/v8.S (ffi_closure_v8): New.
3446 * src/sparc/v9.S (ffi_closure_v9): New.
3448 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3450 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3451 enumeration for X86_WIN32.
3452 * src/x86/win32.S: Added ffi_call_STDCALL function
3454 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3455 switch cases for recognising FFI_STDCALL and
3456 calling ffi_call_STDCALL if target is X86_WIN32.
3457 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3458 stdcall versions of the "my_strlen" and "many"
3459 test functions (for X86_WIN32).
3460 Added test cases to test stdcall invocation using
3463 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3465 * src/sh/sysv.S: Add DWARF2 unwind info.
3467 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3469 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3471 2002-11-26 Jim Wilson <wilson@redhat.com>
3473 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3475 2002-11-23 H.J. Lu <hjl@gnu.org>
3477 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3478 Include ../config/accross.m4.
3479 * aclocal.m4; Rebuild.
3480 * configure: Likewise.
3482 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3484 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3486 2002-11-11 DJ Delorie <dj@redhat.com>
3488 * configure.in: Look for common files in the right place.
3490 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3492 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3493 raw data as _Jv_word values, not ffi_raw.
3494 (ffi_java_ptrarray_to_raw): Likewise.
3495 (ffi_java_rvalue_to_raw): New function.
3496 (ffi_java_raw_call): Call it.
3497 (ffi_java_raw_to_rvalue): New function.
3498 (ffi_java_translate_args): Call it.
3499 * src/ffitest.c (closure_test_fn): Interpret return value
3500 as ffi_arg, not int.
3501 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3502 FFI_TYPE_POINTER case.
3503 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3504 values extended to word size.
3506 2002-10-02 Andreas Jaeger <aj@suse.de>
3508 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3510 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
3512 * include/ffi.h.in: Fix i386 win32 compilation.
3514 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3516 * configure.in: Add s390x-*-linux-* target.
3517 * configure: Regenerate.
3518 * include/ffi.h.in: Define S390X for s390x targets.
3519 (FFI_CLOSURES): Define for s390/s390x.
3520 (FFI_TRAMPOLINE_SIZE): Likewise.
3521 (FFI_NATIVE_RAW_API): Likewise.
3522 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3523 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
3524 * src/s390/ffi.c: Major rework of existing code. Add support for
3525 s390x targets. Add closure support.
3526 * src/s390/sysv.S: Likewise.
3528 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
3530 * src/arm/sysv.S: Fix typo.
3532 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
3534 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3535 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3536 (ffi_call_SYSV): Handle soft-float.
3538 2002-09-27 Bo Thorsen <bo@suse.de>
3540 * include/ffi.h.in: Fix multilib x86-64 support.
3542 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3544 * Makefile.am (all-multi): Fix multilib parallel build.
3546 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3548 * configure.in (sh[34]*-*-linux*): Add brackets.
3549 * configure: Regenerate.
3551 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3553 * Makefile.am: Add SH support.
3554 * Makefile.in: Regenerate.
3555 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3556 * configure: Regenerate.
3557 * include/ffi.h.in: Add SH support.
3558 * src/sh/ffi.c: New file.
3559 * src/sh/sysv.S: New file.
3560 * src/types.c: Add SH support.
3562 2002-07-16 Bo Thorsen <bo@suse.de>
3564 * src/x86/ffi64.c: New file that adds x86-64 support.
3565 * src/x86/unix64.S: New file that handles argument setup for
3567 * src/x86/sysv.S: Don't use this on x86-64.
3568 * src/x86/ffi.c: Don't use this on x86-64.
3570 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3572 * src/ffitest.c (struct6): New test that tests a special case in
3574 (struct7): Likewise.
3575 (struct8): Likewise.
3576 (struct9): Likewise.
3577 (closure_test_fn): Silence warning about this when it's not used.
3578 (main): Add the new tests.
3579 (main): Fix a couple of wrong casts and silence some compiler warnings.
3580 * include/ffi.h.in: Add x86-64 ABI definition.
3581 * fficonfig.h.in: Regenerate.
3582 * Makefile.am: Add x86-64 support.
3583 * configure.in: Likewise.
3584 * Makefile.in: Regenerate.
3585 * configure: Likewise.
3587 2002-06-24 Bo Thorsen <bo@suse.de>
3589 * src/types.c: Merge settings for similar architectures.
3590 Add x86-64 sizes and alignments.
3592 2002-06-23 Bo Thorsen <bo@suse.de>
3594 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3595 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3596 * src/mips/ffi.c (ffi_prep_args): Likewise.
3597 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3599 2002-07-18 H.J. Lu (hjl@gnu.org)
3601 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3602 (libffi_la_SOURCES): Support MIPS_LINUX.
3603 (libffi_convenience_la_SOURCES): Likewise.
3604 * Makefile.in: Regenerated.
3606 * configure.in (mips64*-*): Skip.
3607 (mips*-*-linux*): New.
3608 * configure: Regenerated.
3610 * src/mips/ffi.c: Include <sgidefs.h>.
3612 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3614 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3616 2002-05-27 Roger Sayle <roger@eyesopen.com>
3618 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3620 2002-05-27 Bo Thorsen <bo@suse.de>
3622 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3625 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3627 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3628 beginning of function (for older apple cc).
3630 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3632 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3633 script entry, and set LD to it when configuring multilibs.
3634 * configure: Rebuilt.
3636 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3638 * configure.in (sparc64-*-netbsd*): Add target.
3639 (sparc-*-netbsdelf*): Likewise.
3640 * configure: Regenerate.
3642 2002-04-28 David S. Miller <davem@redhat.com>
3644 * configure.in, configure: Fix SPARC test in previous change.
3646 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3648 * Makefile.am: Add Linux for S/390 support.
3649 * Makefile.in: Regenerate.
3650 * configure.in: Add Linux for S/390 support.
3651 * configure: Regenerate.
3652 * include/ffi.h.in: Add Linux for S/390 support.
3653 * src/s390/ffi.c: New file from libffi CVS tree.
3654 * src/s390/sysv.S: New file from libffi CVS tree.
3656 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3658 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3660 * src/sparc/v8.S: Use it.
3661 * src/sparc/v9.S: Likewise.
3662 * fficonfig.h.in: Rebuilt.
3663 * configure: Rebuilt.
3665 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3667 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3669 * src/ia64/unix.S: Add unwind information. Fix comments.
3670 Save sp in a way that's compatible with unwind info.
3671 (ffi_call_unix): Correctly restore sp in all cases.
3672 * src/ia64/ffi.c: Add, fix comments.
3674 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3676 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3678 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3680 * configure.in (alpha*-*-netbsd*): Add target.
3681 * configure: Regenerate.
3683 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3685 * src/sparc/v8.S: Add unwind info.
3686 * src/sparc/v9.S: Likewise.
3688 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3690 * configure.in: Enable i*86-*-netbsdelf*.
3691 * configure: Rebuilt.
3693 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3696 * src/mips/n32.s: Delete
3697 * src/mips/o32.s: Delete
3699 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3701 * configure.in: Enable alpha*-*-freebsd*.
3702 * configure: Rebuilt.
3704 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3706 * Makefile.am: libfficonvenience -> libffi_convenience.
3707 * Makefile.in: Rebuilt.
3709 * Makefile.am: Define ffitest_OBJECTS.
3710 * Makefile.in: Rebuilt.
3712 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3713 David Edelsohn <edelsohn@gnu.org>
3715 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3716 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3717 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3718 * Makefile.in: Regenerate.
3719 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3720 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3721 (flush_icache, flush_range): New functions.
3722 (ffi_closure_helper_DARWIN): New function.
3723 * src/powerpc/aix_closure.S: New file.
3724 * src/powerpc/darwin_closure.S: New file.
3726 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3728 * include/ffi.h.in: Add typedef for ffi_arg.
3729 * src/ffitest.c (main): Declare rint with ffi_arg.
3731 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3733 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3734 number of GPRs for floating-point arguments.
3736 2002-01-31 Anthony Green <green@redhat.com>
3738 * configure: Rebuilt.
3739 * configure.in: Replace CHECK_SIZEOF and endian tests with
3740 cross-compiler friendly macros.
3741 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3744 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3746 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3747 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3749 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3751 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3752 (TARGET_SRC_POWERPC_AIX): New.
3753 (POWERPC_AIX): New stanza.
3754 * Makefile.in: Regenerate.
3755 * configure.in: Add AIX case.
3756 * configure: Regenerate.
3757 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3758 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3759 size. Fix "long double" support.
3760 (ffi_call): Add FFI_AIX case.
3761 * src/powerpc/aix.S: New.
3763 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3765 Implement Darwin PowerPC ABI.
3766 * configure.in: Handle powerpc-*-darwin*.
3767 * Makefile.am: Set source files for POWERPC_DARWIN.
3768 * configure: Rebuilt.
3769 * Makefile.in: Rebuilt.
3770 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3772 * src/powerpc/darwin.S: New file.
3773 * src/powerpc/ffi_darwin.c: New file.
3775 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3777 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3779 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3781 * src/x86/sysv.S: Avoid gas-only .balign directive.
3782 Use C style comments.
3784 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3786 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3787 Fixes PR bootstrap/3563.
3789 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3791 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3793 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3795 * configure.in: Recognize sparc*-sun-* host.
3796 * configure: Regenerate.
3798 2001-06-06 Andrew Haley <aph@redhat.com>
3800 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
3802 2001-06-03 Andrew Haley <aph@redhat.com>
3804 * src/alpha/osf.S: Add unwind info.
3805 * src/powerpc/sysv.S: Add unwind info.
3806 * src/powerpc/ppc_closure.S: Likewise.
3808 2000-05-31 Jeff Sturm <jsturm@one-point.com>
3810 * configure.in: Fix AC_ARG_ENABLE usage.
3811 * configure: Rebuilt.
3813 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3815 * configure.in: Remove warning about beta code.
3816 * configure: Rebuilt.
3818 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3820 * src/ia64/unix.S: Restore stack pointer when returning from
3822 * src/ia64/ffi.c: Fix typo in comment.
3824 2001-04-18 Jim Wilson <wilson@redhat.com>
3826 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3827 to eliminate RAW DV.
3829 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3831 * Makefile.am: Make a libtool convenience library.
3832 * Makefile.in: Rebuilt.
3834 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3836 * configure.in: Use different syntax for subdirectory creation.
3837 * configure: Rebuilt.
3839 2001-03-27 Jon Beniston <jon@beniston.com>
3841 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3842 * configure: Rebuilt.
3843 * Makefile.am: Added X86_WIN32 target support.
3844 * Makefile.in: Rebuilt.
3846 * include/ffi.h.in: Added X86_WIN32 target support.
3848 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3849 * src/types.c: Added X86_WIN32 target support.
3851 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3852 stuff removed and made to work with CygWin's gas.
3854 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3856 * configure.in: Make target subdirectory in build dir.
3857 * Makefile.am: Override suffix based rules to specify correct output
3859 * Makefile.in: Rebuilt.
3860 * configure: Rebuilt.
3862 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3864 * src/powerpc/ppc_closure.S: New file.
3865 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3866 involving long long and register pairs.
3867 (ffi_prep_closure): New function.
3868 (flush_icache): Likewise.
3869 (ffi_closure_helper_SYSV): Likewise.
3870 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3871 (FFI_TRAMPOLINE_SIZE): Likewise.
3872 (FFI_NATIVE_RAW_API): Likewise.
3873 * Makefile.in: Rebuilt.
3874 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3875 (TARGET_SRC_POWERPC): Likewise.
3877 2001-03-19 Tom Tromey <tromey@redhat.com>
3879 * Makefile.in: Rebuilt.
3880 * Makefile.am (ffitest_LDFLAGS): New macro.
3882 2001-03-02 Nick Clifton <nickc@redhat.com>
3884 * include/ffi.h.in: Remove RCS ident string.
3885 * include/ffi_mips.h: Remove RCS ident string.
3886 * src/debug.c: Remove RCS ident string.
3887 * src/ffitest.c: Remove RCS ident string.
3888 * src/prep_cif.c: Remove RCS ident string.
3889 * src/types.c: Remove RCS ident string.
3890 * src/alpha/ffi.c: Remove RCS ident string.
3891 * src/alpha/osf.S: Remove RCS ident string.
3892 * src/arm/ffi.c: Remove RCS ident string.
3893 * src/arm/sysv.S: Remove RCS ident string.
3894 * src/mips/ffi.c: Remove RCS ident string.
3895 * src/mips/n32.S: Remove RCS ident string.
3896 * src/mips/o32.S: Remove RCS ident string.
3897 * src/sparc/ffi.c: Remove RCS ident string.
3898 * src/sparc/v8.S: Remove RCS ident string.
3899 * src/sparc/v9.S: Remove RCS ident string.
3900 * src/x86/ffi.c: Remove RCS ident string.
3901 * src/x86/sysv.S: Remove RCS ident string.
3903 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3905 * include/ffi.h.in: Change sourceware.cygnus.com references to
3908 2000-12-09 Richard Henderson <rth@redhat.com>
3910 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3911 (ffi_closure_osf_inner): Index rather than increment avalue
3912 and arg_types. Give ffi_closure_osf the raw return value type.
3913 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3916 2000-12-07 Richard Henderson <rth@redhat.com>
3918 * src/raw_api.c (ffi_translate_args): Fix typo.
3919 (ffi_prep_closure): Likewise.
3921 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3922 FFI_TRAMPOLINE_SIZE.
3923 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3924 cif->bytes for new ffi_call_osf implementation.
3925 (ffi_prep_args): Absorb into ...
3926 (ffi_call): ... here. Do all stack allocation here and
3927 avoid a callback function.
3928 (ffi_prep_closure, ffi_closure_osf_inner): New.
3929 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3930 (ffi_closure_osf): New.
3932 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3934 * config.guess, config.sub, install-sh: Removed.
3935 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3936 * Makefile.in: Rebuilt.
3938 * acinclude.m4: Include libtool macros from the top level.
3939 * aclocal.m4, configure: Rebuilt.
3941 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3943 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3944 * configure: Rebuilt.
3946 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3948 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3949 memory correctly. Use conditional instructions, not branches where
3952 2000-05-04 Tom Tromey <tromey@cygnus.com>
3954 * configure: Rebuilt.
3955 * configure.in: Match `arm*-*-linux-*'.
3956 From Chris Dornan <cdornan@arm.com>.
3958 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3960 * Makefile.am (SUBDIRS): Define.
3961 (AM_MAKEFLAGS): Likewise.
3962 (Multilib support.): Add section.
3963 * Makefile.in: Rebuilt.
3964 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3965 New variables. Set for gcc using -print-multi-lib. Export them
3967 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3968 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3970 (extra_compiler_flags_value, extra_compiler_flags): Check these
3971 for extra compiler options which need to be passed down in
3974 2000-04-16 Anthony Green <green@redhat.com>
3976 * configure: Rebuilt.
3977 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3979 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3981 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3982 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3983 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3984 Replace all void * sizeofs with sizeof(int).
3985 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3986 different than DOUBLE.
3987 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3988 (ffi_prep_args_v9): New function.
3989 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3990 (ffi_V9_return_struct): New function.
3991 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3992 32bit code (not yet cross-arch calls).
3993 * src/sparc/v8.S: Add struct return delay nop.
3995 * src/sparc/v9.S: New file.
3996 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3997 is used on sparc64 only for structures larger than 32 bytes.
3998 Pass by reference for structures is done for structure arguments
3999 larger than 16 bytes.
4000 * src/ffitest.c (main): Use 64bit rint on sparc64.
4001 Run long long tests on sparc.
4002 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
4004 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
4006 * configure.in (sparc-*-linux*): New supported target.
4007 (sparc64-*-linux*): Likewise.
4008 * configure: Rebuilt.
4009 * Makefile.am: Add v9.S to SPARC files.
4010 * Makefile.in: Likewise.
4011 (LINK): Surround $(CCLD) into double quotes, so that multilib
4012 compiles work correctly.
4014 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
4016 * configure: Rebuilt.
4017 * configure.in: (i*86-*-solaris*): New libffi target. Patch
4018 proposed by Bryce McKinlay.
4020 2000-03-20 Tom Tromey <tromey@cygnus.com>
4022 * Makefile.in: Hand edit for java_raw_api.lo.
4024 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
4026 * config.guess, config.sub: Update from the gcc tree.
4027 Fix for PR libgcj/168.
4029 2000-03-03 Tom Tromey <tromey@cygnus.com>
4031 * Makefile.in: Fixed ia64 by hand.
4033 * configure: Rebuilt.
4034 * configure.in (--enable-multilib): New option.
4035 (libffi_basedir): New subst.
4036 (AC_OUTPUT): Added multilib code.
4038 2000-03-02 Tom Tromey <tromey@cygnus.com>
4040 * Makefile.in: Rebuilt.
4041 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
4044 2000-02-25 Hans Boehm <boehm@acm.org>
4046 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
4048 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
4050 (ffi_prep_raw_closure): Use ffi_translate_args, not
4051 ffi_closure_translate.
4052 * src/java_raw_api.c: New file.
4053 * src/ffitest.c (closure_test_fn): New function.
4054 (main): Define `rint' as long long on IA64. Added new test when
4055 FFI_CLOSURES is defined.
4056 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
4057 (ffi_abi): Recognize IA64.
4058 (ffi_raw): Added `flt' field.
4059 Added "Java raw API" code.
4060 * configure.in: Recognize ia64.
4061 * Makefile.am (TARGET_SRC_IA64): New macro.
4062 (libffi_la_common_SOURCES): Added java_raw_api.c.
4063 (libffi_la_SOURCES): Define in IA64 case.
4065 2000-01-04 Tom Tromey <tromey@cygnus.com>
4067 * Makefile.in: Rebuilt with newer automake.
4069 1999-12-31 Tom Tromey <tromey@cygnus.com>
4071 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
4073 1999-09-01 Tom Tromey <tromey@cygnus.com>
4075 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
4077 * fficonfig.h.in: Rebuilt.
4078 * configure: Rebuilt.
4079 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
4080 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
4081 * acconfig.h: Don't #undef PACKAGE or VERSION.
4083 1999-08-09 Anthony Green <green@cygnus.com>
4085 * include/ffi.h.in: Try to work around messy header problem
4086 with PACKAGE and VERSION.
4088 * configure: Rebuilt.
4089 * configure.in: Change version to 2.00-beta.
4091 * fficonfig.h.in: Rebuilt.
4092 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
4094 * src/x86/ffi.c (ffi_raw_call): Rename.
4096 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
4098 * src/x86/ffi.c (ffi_closure_SYSV): New function.
4099 (ffi_prep_incoming_args_SYSV): Ditto.
4100 (ffi_prep_closure): Ditto.
4101 (ffi_closure_raw_SYSV): Ditto.
4102 (ffi_prep_raw_closure): More ditto.
4103 (ffi_call_raw): Final ditto.
4105 * include/ffi.h.in: Add definitions for closure and raw API.
4107 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4110 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
4112 * src/raw_api.c: New file.
4114 * include/ffi.h.in (ffi_raw): New type.
4115 (UINT_ARG, SINT_ARG): New defines.
4116 (ffi_closure, ffi_raw_closure): New types.
4117 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4119 * configure.in: Add check for endianness and sizeof void*.
4121 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
4122 instead of directly.
4124 * configure: Rebuilt.
4126 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
4128 * configure.in: Add x86 and powerpc BeOS configurations.
4129 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
4131 1999-05-09 Anthony Green <green@cygnus.com>
4133 * configure.in: Add warning about this being beta code.
4134 Remove src/Makefile.am from the picture.
4135 * configure: Rebuilt.
4137 * Makefile.am: Move logic from src/Makefile.am. Add changes
4138 to support libffi as a target library.
4139 * Makefile.in: Rebuilt.
4141 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
4142 Upgraded to new autoconf, automake, libtool.
4146 * LICENSE: Update copyright date.
4148 * src/Makefile.am, src/Makefile.in: Removed.
4150 1998-11-29 Anthony Green <green@cygnus.com>
4152 * include/ChangeLog: Removed.
4153 * src/ChangeLog: Removed.
4154 * src/mips/ChangeLog: Removed.
4155 * src/sparc/ChangeLog: Remboved.
4156 * src/x86/ChangeLog: Removed.
4158 * ChangeLog.v1: Created.