| blob | bccfef30349384b553b0441198ca093c013bc1bf |
1 2010-07-07 Neil Roberts <neil@linux.intel.com>
3 * src/x86/sysv.S (ffi_call_SYSV): Align the stack pointer to
4 16-bytes.
6 2010-07-02 Jakub Jelinek <jakub@redhat.com>
8 * Makefile.am (AM_MAKEFLAGS): Pass also mandir to submakes.
9 * Makefile.in: Regenerated.
11 2010-05-19 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
13 * configure.ac (libffi_cv_as_x86_pcrel): Check for illegal in as
14 output, too.
15 (libffi_cv_as_ascii_pseudo_op): Check for .ascii.
16 (libffi_cv_as_string_pseudo_op): Check for .string.
17 * configure: Regenerate.
18 * fficonfig.h.in: Regenerate.
19 * src/x86/sysv.S (.eh_frame): Use .ascii, .string or error.
21 2010-04-07 Jakub Jelinek <jakub@redhat.com>
23 * regex.c (byte_re_match_2_internal): Avoid set but not used
24 warning.
26 2010-04-02 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
28 * Makefile.in: Regenerate.
29 * aclocal.m4: Regenerate.
30 * include/Makefile.in: Regenerate.
31 * man/Makefile.in: Regenerate.
32 * testsuite/Makefile.in: Regenerate.
34 2010-03-15 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
36 * configure.ac (libffi_cv_as_x86_64_unwind_section_type): New test.
37 * configure: Regenerate.
38 * fficonfig.h.in: Regenerate.
39 * libffi/src/x86/unix64.S (.eh_frame)
40 [HAVE_AS_X86_64_UNWIND_SECTION_TYPE]: Use @unwind section type.
42 2010-02-24 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
44 * doc/libffi.texi (The Closure API): Fix typo.
45 * doc/libffi.info: Remove.
47 2010-02-15 Matthias Klose <doko@ubuntu.com>
49 * src/arm/sysv.S (__ARM_ARCH__): Define for processor
50 __ARM_ARCH_7EM__.
52 2010-01-07 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
54 PR libffi/40701
55 * testsuite/libffi.call/ffitest.h [__alpha__ && __osf__] (PRIdLL,
56 PRIuLL, PRId64, PRIu64, PRIuPTR): Define.
57 * testsuite/libffi.call/cls_align_sint64.c: Add -Wno-format on
58 alpha*-dec-osf*.
59 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
60 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
61 * testsuite/libffi.call/return_ll1.c: Likewise.
62 * testsuite/libffi.call/stret_medium2.c: Likewise.
63 * testsuite/libffi.special/ffitestcxx.h (allocate_mmap): Cast
64 MAP_FAILED to char *.
66 2010-01-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
68 * src/mips/n32.S: Use .abicalls and .eh_frame with __GNUC__.
70 2009-12-31 Anthony Green <green@redhat.com>
72 * README: Update for libffi 3.0.9.
74 2009-12-27 Matthias Klose <doko@ubuntu.com>
76 * configure.ac (HAVE_LONG_DOUBLE): Define for mips when
77 appropriate.
78 * configure: Rebuilt.
80 2009-12-26 Anthony Green <green@redhat.com>
82 * testsuite/libffi.call/cls_longdouble_va.c: Mark as xfail for
83 avr32*-*-*.
84 * testsuite/libffi.call/cls_double_va.c: Ditto.
86 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
88 * testsuite/libffi.call/ffitest.h: Conditionally include stdint.h
89 and inttypes.h.
90 * testsuite/libffi.special/unwindtest.cc: Ditto.
92 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
94 * configure.ac: Add amd64-*-openbsd*.
95 * configure: Rebuilt.
96 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Link
97 openbsd programs with -lpthread.
99 2009-12-26 Anthony Green <green@redhat.com>
101 * testsuite/libffi.call/cls_double_va.c,
102 testsuite/libffi.call/cls_longdouble.c,
103 testsuite/libffi.call/cls_longdouble_va.c,
104 testsuite/libffi.call/cls_pointer.c,
105 testsuite/libffi.call/cls_pointer_stack.c: Remove xfail for
106 mips*-*-* and arm*-*-*.
107 * testsuite/libffi.call/cls_align_longdouble_split.c,
108 testsuite/libffi.call/cls_align_longdouble_split2.c,
109 testsuite/libffi.call/stret_medium2.c,
110 testsuite/libffi.call/stret_medium.c,
111 testsuite/libffi.call/stret_large.c,
112 testsuite/libffi.call/stret_large2.c: Remove xfail for arm*-*-*.
114 2009-12-31 Kay Tietz <ktietz70@googlemail.com>
116 * testsuite/libffi.call/ffitest.h,
117 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRuLL): Fix
118 definitions.
120 2009-12-31 Carlo Bramini <carlo.bramix@libero.it>
122 * configure.ac (AM_LTLDFLAGS): Define for windows hosts.
123 * Makefile.am (libffi_la_LDFLAGS): Add AM_LTLDFLAGS.
124 * configure: Rebuilt.
125 * Makefile.in: Rebuilt.
127 2009-12-31 Anthony Green <green@redhat.com>
128 Blake Chaffin.
130 * testsuite/libffi.call/huge_struct.c: New test case from Blake
131 Chaffin @ Apple.
133 2009-12-28 David Edelsohn <edelsohn@gnu.org>
135 * src/powerpc/ffi_darwin.c (ffi_prep_args): Copy abi and nargs to
136 local variables.
137 (aix_adjust_aggregate_sizes): New function.
138 (ffi_prep_cif_machdep): Call it.
140 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
142 * configure.ac: Define FFI_MMAP_EXEC_WRIT for the given targets.
143 * configure: Regenerate.
144 * fficonfig.h.in: Likewise.
145 * src/closures.c: Remove the FFI_MMAP_EXEC_WRIT definition for
146 Solaris/x86.
148 2009-12-26 Andreas Schwab <schwab@linux-m68k.org>
150 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Advance intarg_count
151 when a float arguments is passed in memory.
152 (ffi_closure_helper_SYSV): Mark general registers as used up when
153 a 64bit or soft-float long double argument is passed in memory.
155 2009-12-25 Matthias Klose <doko@ubuntu.com>
157 * man/ffi_call.3: Fix #include in examples.
158 * doc/libffi.texi: Add dircategory.
160 2009-12-25 Frank Everdij <f.p.x.everdij@tudelft.nl>
162 * include/ffi.h.in: Placed '__GNUC__' ifdef around
163 '__attribute__((aligned(8)))' in ffi_closure, fixes compile for
164 IRIX MIPSPro c99.
165 * include/ffi_common.h: Added '__sgi' define to non
166 '__attribute__((__mode__()))' integer typedefs.
167 * src/mips/ffi.c (ffi_call, ffi_closure_mips_inner_O32,
168 ffi_closure_mips_inner_N32): Added 'defined(_MIPSEB)' to BE check.
169 (ffi_closure_mips_inner_O32, ffi_closure_mips_inner_N32): Added
170 FFI_LONGDOUBLE support and alignment(N32 only).
171 * src/mips/ffitarget.h: Corrected '#include <sgidefs.h>' for IRIX and
172 fixed non '__attribute__((__mode__()))' integer typedefs.
173 * src/mips/n32.S: Put '#ifdef linux' around '.abicalls' and '.eh_frame'
174 since they are Linux/GNU Assembler specific.
176 2009-12-25 Bradley Smith <brad@brad-smith.co.uk>
178 * configure.ac, Makefile.am, src/avr32/ffi.c,
179 src/avr32/ffitarget.h,
180 src/avr32/sysv.S: Add AVR32 port.
181 * configure, Makefile.in: Rebuilt.
183 2009-12-21 Andreas Tobler <a.tobler@schweiz.org>
185 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
186 * configure: Regenerate.
188 2009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
190 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
192 2009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
194 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
195 type on HP-UX.
197 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
199 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
200 double' arguments.
202 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
204 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
206 2009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
208 PR libffi/40700
209 * src/closures.c [X86_64 && __sun__ && __svr4__]
210 (FFI_MMAP_EXEC_WRIT): Define.
212 2009-12-08 David Daney <ddaney@caviumnetworks.com>
214 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
215 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
216 * testsuite/libffi.call/stret_large.c: Same.
217 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
218 * testsuite/libffi.call/stret_large2.c: Same.
219 * testsuite/libffi.call/stret_medium2.c: Same.
221 2009-12-07 David Edelsohn <edelsohn@gnu.org>
223 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
224 typo.
226 2009-12-05 David Edelsohn <edelsohn@gnu.org>
228 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
229 code.
230 * src/powerpc/aix_closure.S: Same.
232 2009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
234 * Makefile.in: Regenerate.
235 * configure: Regenerate.
236 * include/Makefile.in: Regenerate.
237 * man/Makefile.in: Regenerate.
238 * testsuite/Makefile.in: Regenerate.
240 2009-12-04 David Edelsohn <edelsohn@gnu.org>
242 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
243 linux64_closure.S.
245 2009-12-04 Uros Bizjak <ubizjak@gmail.com>
247 PR libffi/41908
248 * src/x86/ffi64.c (classify_argument): Update from
249 gcc/config/i386/i386.c.
250 (ffi_closure_unix64_inner): Do not use the address of two consecutive
251 SSE registers directly.
252 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
253 for x86_64 linux targets.
255 2009-12-04 David Edelsohn <edelsohn@gnu.org>
257 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
258 pfr for long double split between fpr13 and stack.
260 2009-12-03 David Edelsohn <edelsohn@gnu.org>
262 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
263 fparg_count twice for long double.
265 2009-12-03 David Edelsohn <edelsohn@gnu.org>
267 PR libffi/42243
268 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
270 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
272 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
273 Remove xfails for x86 linux targets.
275 2009-12-02 David Edelsohn <edelsohn@gnu.org>
277 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
278 case.
280 2009-12-01 David Edelsohn <edelsohn@gnu.org>
282 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
283 register usage. Call ffi_prep_args directly. Add long double
284 return value support.
285 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
286 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
287 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
288 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
289 mode.
290 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
291 into case.
292 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
293 Allocate result area between params and FPRs.
295 2009-11-30 David Edelsohn <edelsohn@gnu.org>
297 PR target/35484
298 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
299 AIX64.
300 * src/powerpc/aix.S: Implement AIX64 version.
301 * src/powerpc/aix_closure.S: Implement AIX64 version.
302 (ffi_closure_ASM): Use extsb, lha and displament addresses.
303 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
304 support.
305 (ffi_prep_cif_machdep): Same.
306 (ffi_call): Same.
307 (ffi_closure_helper_DARWIN): Same.
309 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
311 PR libffi/41908
312 * testsuite/libffi.call/testclosure.c: New test.
314 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
316 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
317 assembly version use of ___chkstk.
319 2009-09-23 Matthias Klose <doko@ubuntu.com>
321 PR libffi/40242, PR libffi/41443
322 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
323 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
324 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
325 Change the conditionals to __SOFTFP__ || __ARM_EABI__
326 for -mfloat-abi=softfp to work.
328 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
330 PR testsuite/32843 (strikes again)
331 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
332 enable proper extension on char and short.
334 2009-09-15 David Daney <ddaney@caviumnetworks.com>
336 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
337 handling for FFI_TYPE_POINTER.
338 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
339 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
340 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
341 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
342 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
343 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
344 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
345 structure and pointer returns.
346 (ffi_closure_N32): Add handling for pointer returns.
347 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
348 calc_n32_return_struct_flags): Handle soft-float.
349 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
350 (ffi_call_N32): Declare proper argument types.
351 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
352 soft-float.
354 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
356 * configure.ac (AC_PREREQ): Bump to 2.64.
358 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
360 * Makefile.am (install-html, install-pdf): Remove.
361 * Makefile.in: Regenerate.
363 * Makefile.in: Regenerate.
364 * aclocal.m4: Regenerate.
365 * configure: Regenerate.
366 * fficonfig.h.in: Regenerate.
367 * include/Makefile.in: Regenerate.
368 * man/Makefile.in: Regenerate.
369 * testsuite/Makefile.in: Regenerate.
371 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
373 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
375 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
377 PR libffi/40807
378 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
379 return types for X86_WIN32.
380 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
381 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
382 _ffi_closure_STDCALL): Likewise.
384 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
385 (dlmmap, dlmunmap): Also use these functions on Cygwin.
387 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
389 PR testsuite/40699
390 PR testsuite/40707
391 PR testsuite/40709
392 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
393 2009-06-30 commits.
395 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
397 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
398 to "" before adding paths. (This reinstates an assignment that
399 was removed by my 2009-06-30 commit, but changes the initial
400 value from "." to "".)
402 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
404 PR testsuite/40601
405 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
406 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
407 empty.
409 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
411 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
412 to ld_library_path. Use add_path. Add just find_libgcc_s
413 to ld_library_path, not every libgcc multilib directory.
415 2009-06-16 Wim Lewis <wiml@hhhh.org>
417 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
418 supposed to be callee-saved.
419 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
420 return buffer for odd-size structs.
422 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
424 PR libffi/40444
425 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
426 allow_stack_execute for Darwin.
428 2009-06-16 Andrew Haley <aph@redhat.com>
430 * configure.ac (TARGETDIR): Add missing blank lines.
431 * configure: Regenerate.
433 2009-06-16 Andrew Haley <aph@redhat.com>
435 * testsuite/libffi.call/cls_align_sint64.c,
436 testsuite/libffi.call/cls_align_uint64.c,
437 testsuite/libffi.call/cls_longdouble_va.c,
438 testsuite/libffi.call/cls_ulonglong.c,
439 testsuite/libffi.call/return_ll1.c,
440 testsuite/libffi.call/stret_medium2.c: Fix printf format
441 specifiers.
442 * testsuite/libffi.call/ffitest.h,
443 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
445 2009-06-15 Andrew Haley <aph@redhat.com>
447 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
448 * testsuite/libffi.call/err_bad_abi.c: Likewise.
450 2009-06-12 Andrew Haley <aph@redhat.com>
452 * Makefile.am: Remove info_TEXINFOS.
454 2009-06-12 Andrew Haley <aph@redhat.com>
456 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
457 testsuite/libffi.call/cls_align_uint64.c,
458 testsuite/libffi.call/cls_ulonglong.c,
459 testsuite/libffi.call/return_ll1.c,
460 testsuite/libffi.call/stret_medium2.c: Fix printf format
461 specifiers.
462 testsuite/libffi.special/unwindtest.cc: include stdint.h.
464 2009-06-11 Timothy Wall <twall@users.sf.net>
466 * Makefile.am,
467 configure.ac,
468 include/ffi.h.in,
469 include/ffi_common.h,
470 src/closures.c,
471 src/dlmalloc.c,
472 src/x86/ffi.c,
473 src/x86/ffitarget.h,
474 src/x86/win64.S (new),
475 README: Added win64 support (mingw or MSVC)
476 * Makefile.in,
477 include/Makefile.in,
478 man/Makefile.in,
479 testsuite/Makefile.in,
480 configure,
481 aclocal.m4: Regenerated
482 * ltcf-c.sh: properly escape cygwin/w32 path
483 * man/ffi_call.3: Clarify size requirements for return value.
484 * src/x86/ffi64.c: Fix filename in comment.
485 * src/x86/win32.S: Remove unused extern.
487 * testsuite/libffi.call/closure_fn0.c,
488 testsuite/libffi.call/closure_fn1.c,
489 testsuite/libffi.call/closure_fn2.c,
490 testsuite/libffi.call/closure_fn3.c,
491 testsuite/libffi.call/closure_fn4.c,
492 testsuite/libffi.call/closure_fn5.c,
493 testsuite/libffi.call/closure_fn6.c,
494 testsuite/libffi.call/closure_stdcall.c,
495 testsuite/libffi.call/cls_12byte.c,
496 testsuite/libffi.call/cls_16byte.c,
497 testsuite/libffi.call/cls_18byte.c,
498 testsuite/libffi.call/cls_19byte.c,
499 testsuite/libffi.call/cls_1_1byte.c,
500 testsuite/libffi.call/cls_20byte.c,
501 testsuite/libffi.call/cls_20byte1.c,
502 testsuite/libffi.call/cls_24byte.c,
503 testsuite/libffi.call/cls_2byte.c,
504 testsuite/libffi.call/cls_3_1byte.c,
505 testsuite/libffi.call/cls_3byte1.c,
506 testsuite/libffi.call/cls_3byte2.c,
507 testsuite/libffi.call/cls_4_1byte.c,
508 testsuite/libffi.call/cls_4byte.c,
509 testsuite/libffi.call/cls_5_1_byte.c,
510 testsuite/libffi.call/cls_5byte.c,
511 testsuite/libffi.call/cls_64byte.c,
512 testsuite/libffi.call/cls_6_1_byte.c,
513 testsuite/libffi.call/cls_6byte.c,
514 testsuite/libffi.call/cls_7_1_byte.c,
515 testsuite/libffi.call/cls_7byte.c,
516 testsuite/libffi.call/cls_8byte.c,
517 testsuite/libffi.call/cls_9byte1.c,
518 testsuite/libffi.call/cls_9byte2.c,
519 testsuite/libffi.call/cls_align_double.c,
520 testsuite/libffi.call/cls_align_float.c,
521 testsuite/libffi.call/cls_align_longdouble.c,
522 testsuite/libffi.call/cls_align_longdouble_split.c,
523 testsuite/libffi.call/cls_align_longdouble_split2.c,
524 testsuite/libffi.call/cls_align_pointer.c,
525 testsuite/libffi.call/cls_align_sint16.c,
526 testsuite/libffi.call/cls_align_sint32.c,
527 testsuite/libffi.call/cls_align_sint64.c,
528 testsuite/libffi.call/cls_align_uint16.c,
529 testsuite/libffi.call/cls_align_uint32.c,
530 testsuite/libffi.call/cls_align_uint64.c,
531 testsuite/libffi.call/cls_dbls_struct.c,
532 testsuite/libffi.call/cls_double.c,
533 testsuite/libffi.call/cls_double_va.c,
534 testsuite/libffi.call/cls_float.c,
535 testsuite/libffi.call/cls_longdouble.c,
536 testsuite/libffi.call/cls_longdouble_va.c,
537 testsuite/libffi.call/cls_multi_schar.c,
538 testsuite/libffi.call/cls_multi_sshort.c,
539 testsuite/libffi.call/cls_multi_sshortchar.c,
540 testsuite/libffi.call/cls_multi_uchar.c,
541 testsuite/libffi.call/cls_multi_ushort.c,
542 testsuite/libffi.call/cls_multi_ushortchar.c,
543 testsuite/libffi.call/cls_pointer.c,
544 testsuite/libffi.call/cls_pointer_stack.c,
545 testsuite/libffi.call/cls_schar.c,
546 testsuite/libffi.call/cls_sint.c,
547 testsuite/libffi.call/cls_sshort.c,
548 testsuite/libffi.call/cls_uchar.c,
549 testsuite/libffi.call/cls_uint.c,
550 testsuite/libffi.call/cls_ulonglong.c,
551 testsuite/libffi.call/cls_ushort.c,
552 testsuite/libffi.call/err_bad_abi.c,
553 testsuite/libffi.call/err_bad_typedef.c,
554 testsuite/libffi.call/float2.c,
555 testsuite/libffi.call/huge_struct.c,
556 testsuite/libffi.call/nested_struct.c,
557 testsuite/libffi.call/nested_struct1.c,
558 testsuite/libffi.call/nested_struct10.c,
559 testsuite/libffi.call/nested_struct2.c,
560 testsuite/libffi.call/nested_struct3.c,
561 testsuite/libffi.call/nested_struct4.c,
562 testsuite/libffi.call/nested_struct5.c,
563 testsuite/libffi.call/nested_struct6.c,
564 testsuite/libffi.call/nested_struct7.c,
565 testsuite/libffi.call/nested_struct8.c,
566 testsuite/libffi.call/nested_struct9.c,
567 testsuite/libffi.call/problem1.c,
568 testsuite/libffi.call/return_ldl.c,
569 testsuite/libffi.call/return_ll1.c,
570 testsuite/libffi.call/stret_large.c,
571 testsuite/libffi.call/stret_large2.c,
572 testsuite/libffi.call/stret_medium.c,
573 testsuite/libffi.call/stret_medium2.c,
574 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
575 of checking for MMAP. Use intptr_t instead of long casts.
577 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
579 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
580 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
581 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
583 2009-06-09 Andrew Haley <aph@redhat.com>
585 * src/x86/freebsd.S: Add missing file.
587 2009-06-08 Andrew Haley <aph@redhat.com>
589 Import from libffi 3.0.8:
591 * doc/libffi.texi: New file.
592 * doc/libffi.info: Likewise.
593 * doc/stamp-vti: Likewise.
594 * man/Makefile.am: New file.
595 * man/ffi_call.3: New file.
597 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
598 src/dlmalloc.c.
599 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
601 * configure.ac: Bump version to 3.0.8.
602 parisc*-*-linux*: Add.
603 i386-*-freebsd* | i386-*-openbsd*: Add.
604 powerpc-*-beos*: Add.
605 AM_CONDITIONAL X86_FREEBSD: Add.
606 AC_CONFIG_FILES: Add man/Makefile.
608 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
610 2009-06-08 Andrew Haley <aph@redhat.com>
612 * README: Import from libffi 3.0.8.
614 2009-06-08 Andrew Haley <aph@redhat.com>
616 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
617 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
618 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
619 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
621 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
622 * testsuite/libffi.call/stret_medium.c: Likewise.
623 * testsuite/libffi.call/stret_large2.c: Likewise.
624 * testsuite/libffi.call/stret_large.c: Likewise.
626 2008-12-26 Timothy Wall <twall@users.sf.net>
628 * testsuite/libffi.call/cls_longdouble.c,
629 testsuite/libffi.call/cls_longdouble_va.c,
630 testsuite/libffi.call/cls_align_longdouble.c,
631 testsuite/libffi.call/cls_align_longdouble_split.c,
632 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
633 failures on x86_64 cygwin/mingw.
635 2008-12-22 Timothy Wall <twall@users.sf.net>
637 * testsuite/libffi.call/closure_fn0.c,
638 testsuite/libffi.call/closure_fn1.c,
639 testsuite/libffi.call/closure_fn2.c,
640 testsuite/libffi.call/closure_fn3.c,
641 testsuite/libffi.call/closure_fn4.c,
642 testsuite/libffi.call/closure_fn5.c,
643 testsuite/libffi.call/closure_fn6.c,
644 testsuite/libffi.call/closure_loc_fn0.c,
645 testsuite/libffi.call/closure_stdcall.c,
646 testsuite/libffi.call/cls_align_pointer.c,
647 testsuite/libffi.call/cls_pointer.c,
648 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
649 pointer to integer (intptr_t).
650 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
652 2008-07-24 Anthony Green <green@redhat.com>
654 * testsuite/libffi.call/cls_dbls_struct.c,
655 testsuite/libffi.call/cls_double_va.c,
656 testsuite/libffi.call/cls_longdouble.c,
657 testsuite/libffi.call/cls_longdouble_va.c,
658 testsuite/libffi.call/cls_pointer.c,
659 testsuite/libffi.call/cls_pointer_stack.c,
660 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
661 compiler warnings.
663 2008-03-04 Anthony Green <green@redhat.com>
664 Blake Chaffin
665 hos@tamanegi.org
667 * testsuite/libffi.call/cls_align_longdouble_split2.c
668 testsuite/libffi.call/cls_align_longdouble_split.c
669 testsuite/libffi.call/cls_dbls_struct.c
670 testsuite/libffi.call/cls_double_va.c
671 testsuite/libffi.call/cls_longdouble.c
672 testsuite/libffi.call/cls_longdouble_va.c
673 testsuite/libffi.call/cls_pointer.c
674 testsuite/libffi.call/cls_pointer_stack.c
675 testsuite/libffi.call/err_bad_abi.c
676 testsuite/libffi.call/err_bad_typedef.c
677 testsuite/libffi.call/stret_large2.c
678 testsuite/libffi.call/stret_large.c
679 testsuite/libffi.call/stret_medium2.c
680 testsuite/libffi.call/stret_medium.c: New tests from Apple.
682 2009-06-05 Andrew Haley <aph@redhat.com>
684 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
685 libffi.
687 2009-06-04 Andrew Haley <aph@redhat.com>
689 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
690 stdcall changes.
692 2008-02-26 Anthony Green <green@redhat.com>
693 Thomas Heller <theller@ctypes.org>
695 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
696 comment.
698 2008-02-03 Timothy Wall <twall@users.sf.net>
700 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
701 offset based on code pointer, not data pointer.
703 2008-01-31 Timothy Wall <twall@users.sf.net>
705 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
706 closures.
707 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
708 closures.
709 * src/x86/win32.S: Add assembly for stdcall closure.
710 * src/x86/ffi.c: Initialize stdcall closure trampoline.
712 2009-06-04 Andrew Haley <aph@redhat.com>
714 * include/ffi.h.in: Change void (*)() to void (*)(void).
715 * src/x86/ffi.c: Likewise.
717 2009-06-04 Andrew Haley <aph@redhat.com>
719 * src/powerpc/ppc_closure.S: Insert licence header.
720 * src/powerpc/linux64_closure.S: Likewise.
721 * src/m68k/sysv.S: Likewise.
723 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
724 * src/powerpc/ffi.c: Likewise.
725 * src/powerpc/ffi_darwin.c: Likewise.
726 * src/m32r/ffi.c: Likewise.
727 * src/sh64/ffi.c: Likewise.
728 * src/x86/ffi64.c: Likewise.
729 * src/alpha/ffi.c: Likewise.
730 * src/alpha/osf.S: Likewise.
731 * src/frv/ffi.c: Likewise.
732 * src/s390/ffi.c: Likewise.
733 * src/pa/ffi.c: Likewise.
734 * src/pa/hpux32.S: Likewise.
735 * src/ia64/unix.S: Likewise.
736 * src/ia64/ffi.c: Likewise.
737 * src/sparc/ffi.c: Likewise.
738 * src/mips/ffi.c: Likewise.
739 * src/sh/ffi.c: Likewise.
741 2008-02-15 David Daney <ddaney@avtrex.com>
743 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
744 Define (conditionally), and use it to include cachectl.h.
745 (ffi_prep_closure_loc): Fix cache flushing.
746 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
748 2009-06-04 Andrew Haley <aph@redhat.com>
750 include/ffi.h.in,
751 src/arm/ffitarget.h,
752 src/arm/ffi.c,
753 src/arm/sysv.S,
754 src/powerpc/ffitarget.h,
755 src/closures.c,
756 src/sh64/ffitarget.h,
757 src/sh64/ffi.c,
758 src/sh64/sysv.S,
759 src/types.c,
760 src/x86/ffi64.c,
761 src/x86/ffitarget.h,
762 src/x86/win32.S,
763 src/x86/darwin.S,
764 src/x86/ffi.c,
765 src/x86/sysv.S,
766 src/x86/unix64.S,
767 src/alpha/ffitarget.h,
768 src/alpha/ffi.c,
769 src/alpha/osf.S,
770 src/m68k/ffitarget.h,
771 src/frv/ffitarget.h,
772 src/frv/ffi.c,
773 src/s390/ffitarget.h,
774 src/s390/sysv.S,
775 src/cris/ffitarget.h,
776 src/pa/linux.S,
777 src/pa/ffitarget.h,
778 src/pa/ffi.c,
779 src/raw_api.c,
780 src/ia64/ffitarget.h,
781 src/ia64/unix.S,
782 src/ia64/ffi.c,
783 src/ia64/ia64_flags.h,
784 src/java_raw_api.c,
785 src/debug.c,
786 src/sparc/v9.S,
787 src/sparc/ffitarget.h,
788 src/sparc/ffi.c,
789 src/sparc/v8.S,
790 src/mips/ffitarget.h,
791 src/mips/n32.S,
792 src/mips/o32.S,
793 src/mips/ffi.c,
794 src/prep_cif.c,
795 src/sh/ffitarget.h,
796 src/sh/ffi.c,
797 src/sh/sysv.S: Update license text.
799 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
801 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
802 (.eh_frame): Add FDE for it.
804 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
806 * configure.ac: Also check if assembler supports pc-relative
807 relocs on X86_WIN32 targets.
808 * configure: Regenerate.
809 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
810 (_ffi_call_SYSV): Add missing function type symbol .def and
811 add EH markup labels.
812 (_ffi_call_STDCALL): Likewise.
813 (_ffi_closure_SYSV): Likewise.
814 (_ffi_closure_raw_SYSV): Likewise.
815 (.eh_frame): Add hand-crafted EH data.
817 2009-04-09 Jakub Jelinek <jakub@redhat.com>
819 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
820 version 3 of the GNU General Public License and to point readers
821 at the COPYING3 file and the FSF's license web page.
822 * testsuite/libffi.call/call.exp: Likewise.
823 * testsuite/libffi.special/special.exp: Likewise.
825 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
827 * configure: Regenerate.
829 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
831 PR libffi/26048
832 * configure.ac (HAVE_AS_X86_PCREL): New test.
833 * configure: Regenerate.
834 * fficonfig.h.in: Regenerate.
835 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
836 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
837 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
838 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
839 * src/x86/unix64.S (.Lstore_table): Move to .text section.
840 (.Lload_table): Likewise.
841 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
843 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
845 * configure: Regenerate.
847 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
849 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
850 signed/unsigned int8/16 return values.
851 * src/sparc/v8.S (ffi_call_v8): Likewise.
852 (ffi_closure_v8): Likewise.
854 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
855 Steve Ellcey <sje@cup.hp.com>
857 * configure: Regenerate for new libtool.
858 * Makefile.in: Ditto.
859 * include/Makefile.in: Ditto.
860 * aclocal.m4: Ditto.
862 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
864 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
865 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
866 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
867 Adjust copyright notice.
868 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
869 register or two register to use for FFI_SYSV structs.
870 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
871 (ffi_closure_helper_SYSV): Fix the return type for
872 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
873 Adjust copyright notice.
875 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
877 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
878 int.
880 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
882 * configure: Regenerate.
883 * include/Makefile.in: Regenerate.
884 * testsuite/Makefile.in: Regenerate.
886 2008-06-07 Joseph Myers <joseph@codesourcery.com>
888 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
889 powerpc-*-beos*): Remove.
890 * configure: Regenerate.
892 2008-05-09 Julian Brown <julian@codesourcery.com>
894 * Makefile.am (LTLDFLAGS): New.
895 (libffi_la_LDFLAGS): Use above.
896 * Makefile.in: Regenerate.
898 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
900 PR bootstrap/35457
901 * aclocal.m4: Regenerate.
902 * configure: Regenerate.
904 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
906 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
907 * src/sh64/sysv.S: Likewise.
909 2008-03-26 Daniel Jacobowitz <dan@debian.org>
911 * src/arm/sysv.S: Fix ARM comment marker.
913 2008-03-26 Jakub Jelinek <jakub@redhat.com>
915 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
916 * src/s390/sysv.S: Likewise.
917 * src/powerpc/ppc_closure.S: Likewise.
918 * src/powerpc/sysv.S: Likewise.
919 * src/x86/unix64.S: Likewise.
920 * src/x86/sysv.S: Likewise.
921 * src/sparc/v8.S: Likewise.
922 * src/sparc/v9.S: Likewise.
923 * src/m68k/sysv.S: Likewise.
924 * src/arm/sysv.S: Likewise.
926 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
928 * aclocal.m4: Regenerate.
929 * configure: Likewise.
930 * Makefile.in: Likewise.
931 * include/Makefile.in: Likewise.
932 * testsuite/Makefile.in: Likewise.
934 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
935 Andreas Tobler <a.tobler@schweiz.org>
937 * configure.ac: Add amd64-*-freebsd* target.
938 * configure: Regenerate.
940 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
942 PR libffi/34612
943 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
944 returning struct.
946 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
947 tests.
949 2008-01-24 David Edelsohn <edelsohn@gnu.org>
951 * configure: Regenerate.
953 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
955 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
957 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
959 PR testsuite/32843
960 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
961 signed/unsigned int8/16 for X86_DARWIN.
962 Updated copyright info.
963 Handle one and two byte structs with special cif->flags.
964 * src/x86/ffitarget.h: Add special types for one and two byte structs.
965 Updated copyright info.
966 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
967 sysv.S
968 Remove code to pop args from the stack after call.
969 Special-case signed/unsigned for int8/16, one and two byte structs.
970 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
971 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
972 FFI_TYPE_SINT32.
973 Updated copyright info.
975 2007-12-08 David Daney <ddaney@avtrex.com>
977 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
978 SUBU, add with ADDU and use smaller code sequences.
980 2007-12-07 David Daney <ddaney@avtrex.com>
982 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
983 type.
985 2007-12-06 David Daney <ddaney@avtrex.com>
987 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
988 defined.
989 (ffi_java_raw): New typedef.
990 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
991 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
992 ffi_java_raw.
993 (ffi_java_raw_closure) : Same.
994 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
995 parameter types.
996 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
997 FFI_SIZEOF_JAVA_RAW.
998 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
999 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
1000 sizeof(ffi_java_raw) for alignment calculations.
1001 (ffi_java_ptrarray_to_raw): Same.
1002 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
1003 if FFI_SIZEOF_JAVA_RAW == 4.
1004 (ffi_java_raw_to_rvalue): Same.
1005 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
1006 (ffi_java_translate_args): Same.
1007 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
1008 parameter types.
1009 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
1011 2007-12-06 David Daney <ddaney@avtrex.com>
1013 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
1014 pointer values.
1016 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
1018 PR libffi/31937
1019 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
1020 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
1021 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
1022 set the NUM_FPR_ARG_REGISTERS according to.
1023 Add support for potential soft-float support under hard-float
1024 architecture.
1025 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
1026 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
1027 to the FFI_LINUX_SOFT_FLOAT ABI.
1028 (ffi_prep_cif_machdep): Likewise.
1029 (ffi_closure_helper_SYSV): Likewise.
1030 * src/powerpc/ppc_closure.S: Make sure not to store float/double
1031 on archs where __NO_FPRS__ is true.
1032 Add FFI_TYPE_UINT128 support.
1033 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
1034 Adjust copyright notice.
1036 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
1038 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
1039 * include/ffi_common.h: ... here.
1040 Update copyright.
1042 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
1044 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
1045 * src/powerpc/linux64.S: Likewise.
1046 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
1047 * testsuite/libffi.call/return_ldl.c: New test.
1049 2007-09-04 <aph@redhat.com>
1051 * src/arm/sysv.S (UNWIND): New.
1052 (Whole file): Conditionally compile unwinder directives.
1053 * src/arm/sysv.S: Add unwinder directives.
1055 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
1056 Only treat r0 as a struct address if we're actually returning a
1057 struct by address.
1058 Only copy the bytes that are actually within a struct.
1059 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
1060 is returned in r0, not passed by address.
1061 (ffi_call): Allocate a word-sized temporary for the case where
1062 a composite is returned in r0.
1063 (ffi_prep_incoming_args_SYSV): Align as necessary.
1065 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
1067 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
1068 directly using the sys_cacheflush syscall.
1070 2007-07-27 Andrew Haley <aph@redhat.com>
1072 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
1074 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
1076 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
1077 * configure.ac: Likewise.
1078 * Makefile.in: Regenerate.
1079 * include/Makefile.in: Likewise.
1080 * testsuite/Makefile.in: Likewise.
1081 * configure: Likewise.
1083 2007-08-24 David Daney <ddaney@avtrex.com>
1085 * testsuite/libffi.call/return_sl.c: New test.
1087 2007-08-10 David Daney <ddaney@avtrex.com>
1089 * testsuite/libffi.call/cls_multi_ushort.c,
1090 testsuite/libffi.call/cls_align_uint16.c,
1091 testsuite/libffi.call/nested_struct1.c,
1092 testsuite/libffi.call/nested_struct3.c,
1093 testsuite/libffi.call/cls_7_1_byte.c,
1094 testsuite/libffi.call/nested_struct5.c,
1095 testsuite/libffi.call/cls_double.c,
1096 testsuite/libffi.call/nested_struct7.c,
1097 testsuite/libffi.call/cls_sint.c,
1098 testsuite/libffi.call/nested_struct9.c,
1099 testsuite/libffi.call/cls_20byte1.c,
1100 testsuite/libffi.call/cls_multi_sshortchar.c,
1101 testsuite/libffi.call/cls_align_sint64.c,
1102 testsuite/libffi.call/cls_3byte2.c,
1103 testsuite/libffi.call/cls_multi_schar.c,
1104 testsuite/libffi.call/cls_multi_uchar.c,
1105 testsuite/libffi.call/cls_19byte.c,
1106 testsuite/libffi.call/cls_9byte1.c,
1107 testsuite/libffi.call/cls_align_float.c,
1108 testsuite/libffi.call/closure_fn1.c,
1109 testsuite/libffi.call/problem1.c,
1110 testsuite/libffi.call/closure_fn3.c,
1111 testsuite/libffi.call/cls_sshort.c,
1112 testsuite/libffi.call/closure_fn5.c,
1113 testsuite/libffi.call/cls_align_double.c,
1114 testsuite/libffi.call/nested_struct.c,
1115 testsuite/libffi.call/cls_2byte.c,
1116 testsuite/libffi.call/nested_struct10.c,
1117 testsuite/libffi.call/cls_4byte.c,
1118 testsuite/libffi.call/cls_6byte.c,
1119 testsuite/libffi.call/cls_8byte.c,
1120 testsuite/libffi.call/cls_multi_sshort.c,
1121 testsuite/libffi.call/cls_align_sint16.c,
1122 testsuite/libffi.call/cls_align_uint32.c,
1123 testsuite/libffi.call/cls_20byte.c,
1124 testsuite/libffi.call/cls_float.c,
1125 testsuite/libffi.call/nested_struct2.c,
1126 testsuite/libffi.call/cls_5_1_byte.c,
1127 testsuite/libffi.call/nested_struct4.c,
1128 testsuite/libffi.call/cls_24byte.c,
1129 testsuite/libffi.call/nested_struct6.c,
1130 testsuite/libffi.call/cls_64byte.c,
1131 testsuite/libffi.call/nested_struct8.c,
1132 testsuite/libffi.call/cls_uint.c,
1133 testsuite/libffi.call/cls_multi_ushortchar.c,
1134 testsuite/libffi.call/cls_schar.c,
1135 testsuite/libffi.call/cls_uchar.c,
1136 testsuite/libffi.call/cls_align_uint64.c,
1137 testsuite/libffi.call/cls_ulonglong.c,
1138 testsuite/libffi.call/cls_align_longdouble.c,
1139 testsuite/libffi.call/cls_1_1byte.c,
1140 testsuite/libffi.call/cls_12byte.c,
1141 testsuite/libffi.call/cls_3_1byte.c,
1142 testsuite/libffi.call/cls_3byte1.c,
1143 testsuite/libffi.call/cls_4_1byte.c,
1144 testsuite/libffi.call/cls_6_1_byte.c,
1145 testsuite/libffi.call/cls_16byte.c,
1146 testsuite/libffi.call/cls_18byte.c,
1147 testsuite/libffi.call/closure_fn0.c,
1148 testsuite/libffi.call/cls_9byte2.c,
1149 testsuite/libffi.call/closure_fn2.c,
1150 testsuite/libffi.call/closure_fn4.c,
1151 testsuite/libffi.call/cls_ushort.c,
1152 testsuite/libffi.call/closure_fn6.c,
1153 testsuite/libffi.call/cls_5byte.c,
1154 testsuite/libffi.call/cls_align_pointer.c,
1155 testsuite/libffi.call/cls_7byte.c,
1156 testsuite/libffi.call/cls_align_sint32.c,
1157 testsuite/libffi.special/unwindtest_ffi_call.cc,
1158 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
1160 2007-08-10 David Daney <ddaney@avtrex.com>
1162 PR libffi/28313
1163 * configure.ac: Don't treat mips64 as a special case.
1164 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
1165 * configure: Regenerate
1166 * Makefile.in: Ditto.
1167 * fficonfig.h.in: Ditto.
1168 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
1169 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
1170 (FFI_DEFAULT_ABI): Set for n64 case.
1171 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
1172 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
1173 (ffi_closure_N32): New function.
1174 (.eh_frame): New section
1175 * src/mips/o32.S: Clean up comments.
1176 (ffi_closure_O32): Pass ffi_closure parameter in $12.
1177 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
1178 _MIPS_SIM == _ABIN32 throughout.
1179 (FFI_MIPS_STOP_HERE): New, use in place of
1180 ffi_stop_here.
1181 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1182 to support n32/n64 ABIs.
1183 (calc_n32_struct_flags): Rewrite.
1184 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1185 position of flag bits.
1186 (ffi_prep_cif_machdep): Rewrite n32 portion.
1187 (ffi_call): Enable for n64. Add special handling for small structure
1188 return values.
1189 (ffi_prep_closure_loc): Add n32 and n64 support.
1190 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1191 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1193 2007-08-08 David Daney <ddaney@avtrex.com>
1195 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1196 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1197 specifiers.
1198 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1199 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1200 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1201 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1202 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1203 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1204 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1205 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1206 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1207 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1208 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1209 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1210 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1211 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1212 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1213 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1214 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1215 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1216 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1217 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1218 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1219 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1220 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1221 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1222 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1223 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1224 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1225 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1226 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1227 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1228 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1230 2007-08-07 Andrew Haley <aph@redhat.com>
1232 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1233 checkin.
1235 2007-08-06 Andrew Haley <aph@redhat.com>
1237 PR testsuite/32843
1238 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1239 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1240 FFI_TYPE_SINT32.
1242 2007-08-02 David Daney <ddaney@avtrex.com>
1244 * testsuite/libffi.call/return_ul.c (main): Define return type as
1245 ffi_arg. Use proper printf conversion specifier.
1247 2007-07-30 Andrew Haley <aph@redhat.com>
1249 PR testsuite/32843
1250 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1251 signed/unsigned int8/16.
1252 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1253 Use a jump table.
1254 Remove code to pop args from the stack after call.
1255 Special-case signed/unsigned int8/16.
1256 * testsuite/libffi.call/return_sc.c (main): Revert.
1258 2007-07-26 Richard Guenther <rguenther@suse.de>
1260 PR testsuite/32843
1261 * testsuite/libffi.call/return_sc.c (main): Verify call
1262 result as signed char, not ffi_arg.
1264 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1266 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1267 * configure: Regenerate.
1269 2007-07-11 David Daney <ddaney@avtrex.com>
1271 * src/mips/ffi.c: Don't include sys/cachectl.h.
1272 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1273 cacheflush().
1275 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1277 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1278 from (ffi_prep_closure): ... this.
1279 (FFI_INIT_TRAMPOLINE): Adjust.
1281 2005-12-31 Phil Blundell <pb@reciva.com>
1283 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1284 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1285 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1286 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1287 (FFI_CLOSURES): Enable closure support.
1289 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1291 * testsuite/libffi.call/cls_multi_ushort.c,
1292 testsuite/libffi.call/cls_align_uint16.c,
1293 testsuite/libffi.call/nested_struct1.c,
1294 testsuite/libffi.call/nested_struct3.c,
1295 testsuite/libffi.call/cls_7_1_byte.c,
1296 testsuite/libffi.call/cls_double.c,
1297 testsuite/libffi.call/nested_struct5.c,
1298 testsuite/libffi.call/nested_struct7.c,
1299 testsuite/libffi.call/cls_sint.c,
1300 testsuite/libffi.call/nested_struct9.c,
1301 testsuite/libffi.call/cls_20byte1.c,
1302 testsuite/libffi.call/cls_multi_sshortchar.c,
1303 testsuite/libffi.call/cls_align_sint64.c,
1304 testsuite/libffi.call/cls_3byte2.c,
1305 testsuite/libffi.call/cls_multi_schar.c,
1306 testsuite/libffi.call/cls_multi_uchar.c,
1307 testsuite/libffi.call/cls_19byte.c,
1308 testsuite/libffi.call/cls_9byte1.c,
1309 testsuite/libffi.call/cls_align_float.c,
1310 testsuite/libffi.call/closure_fn1.c,
1311 testsuite/libffi.call/problem1.c,
1312 testsuite/libffi.call/closure_fn3.c,
1313 testsuite/libffi.call/cls_sshort.c,
1314 testsuite/libffi.call/closure_fn5.c,
1315 testsuite/libffi.call/cls_align_double.c,
1316 testsuite/libffi.call/cls_2byte.c,
1317 testsuite/libffi.call/nested_struct.c,
1318 testsuite/libffi.call/nested_struct10.c,
1319 testsuite/libffi.call/cls_4byte.c,
1320 testsuite/libffi.call/cls_6byte.c,
1321 testsuite/libffi.call/cls_8byte.c,
1322 testsuite/libffi.call/cls_multi_sshort.c,
1323 testsuite/libffi.call/cls_align_uint32.c,
1324 testsuite/libffi.call/cls_align_sint16.c,
1325 testsuite/libffi.call/cls_float.c,
1326 testsuite/libffi.call/cls_20byte.c,
1327 testsuite/libffi.call/cls_5_1_byte.c,
1328 testsuite/libffi.call/nested_struct2.c,
1329 testsuite/libffi.call/cls_24byte.c,
1330 testsuite/libffi.call/nested_struct4.c,
1331 testsuite/libffi.call/nested_struct6.c,
1332 testsuite/libffi.call/cls_64byte.c,
1333 testsuite/libffi.call/nested_struct8.c,
1334 testsuite/libffi.call/cls_uint.c,
1335 testsuite/libffi.call/cls_multi_ushortchar.c,
1336 testsuite/libffi.call/cls_schar.c,
1337 testsuite/libffi.call/cls_uchar.c,
1338 testsuite/libffi.call/cls_align_uint64.c,
1339 testsuite/libffi.call/cls_ulonglong.c,
1340 testsuite/libffi.call/cls_align_longdouble.c,
1341 testsuite/libffi.call/cls_1_1byte.c,
1342 testsuite/libffi.call/cls_12byte.c,
1343 testsuite/libffi.call/cls_3_1byte.c,
1344 testsuite/libffi.call/cls_3byte1.c,
1345 testsuite/libffi.call/cls_4_1byte.c,
1346 testsuite/libffi.call/cls_6_1_byte.c,
1347 testsuite/libffi.call/cls_16byte.c,
1348 testsuite/libffi.call/cls_18byte.c,
1349 testsuite/libffi.call/closure_fn0.c,
1350 testsuite/libffi.call/cls_9byte2.c,
1351 testsuite/libffi.call/closure_fn2.c,
1352 testsuite/libffi.call/closure_fn4.c,
1353 testsuite/libffi.call/cls_ushort.c,
1354 testsuite/libffi.call/closure_fn6.c,
1355 testsuite/libffi.call/cls_5byte.c,
1356 testsuite/libffi.call/cls_align_pointer.c,
1357 testsuite/libffi.call/cls_7byte.c,
1358 testsuite/libffi.call/cls_align_sint32.c,
1359 testsuite/libffi.special/unwindtest_ffi_call.cc,
1360 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1362 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1364 * aclocal.m4: Regenerated.
1366 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1368 * configure: Regenerate.
1370 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1372 * Makefile.in: Regenerate.
1373 * configure: Regenerate.
1374 * aclocal.m4: Regenerate.
1375 * include/Makefile.in: Regenerate.
1376 * testsuite/Makefile.in: Regenerate.
1378 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1380 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1381 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1382 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1383 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1384 (FFI_CLOSURES): Enable closure support.
1386 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1388 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1389 * configure: Regenerate.
1390 * fficonfig.h.in: Regenerate.
1391 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1392 CFI_OFFSET,CFI_DEF_CFA): New macros.
1393 (ffi_call_SYSV): Add callframe annotation.
1395 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1397 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1398 numerous test suite failures.
1399 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1401 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1403 * Makefile.am (EXTRA_DIST): Bring up to date.
1404 * Makefile.in: Regenerate.
1405 * src/frv/eabi.S: Remove RCS keyword.
1407 2007-04-06 Richard Henderson <rth@redhat.com>
1409 * configure.ac: Tidy target case.
1410 (HAVE_LONG_DOUBLE): Allow the target to override.
1411 * configure: Regenerate.
1412 * include/ffi.h.in: Don't define ffi_type_foo if
1413 LIBFFI_HIDE_BASIC_TYPES is defined.
1414 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1415 to ffi_type_double.
1416 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1417 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1418 (ffi_type_longdouble): Special case for Alpha. Don't define
1419 if long double == double.
1421 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1422 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1423 (ffi_call, ffi_closure_osf_inner): Likewise.
1424 (ffi_closure_osf_inner): Likewise. Mark hidden.
1425 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1426 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1427 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1428 (load_table): Handle 128-bit long double.
1430 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1432 2007-04-06 Tom Tromey <tromey@redhat.com>
1434 PR libffi/31491:
1435 * README: Fixed bug in example.
1437 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1439 * src/closures.c: Include sys/statfs.h.
1440 (_GNU_SOURCE): Define on Linux.
1441 (FFI_MMAP_EXEC_SELINUX): Define.
1442 (selinux_enabled): New variable.
1443 (selinux_enabled_check): New function.
1444 (is_selinux_enabled): Define.
1445 (dlmmap): Use it.
1447 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1449 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1450 Use 'volatile float sum' to create sum of floats to avoid false
1451 negative due to excess precision on ix86 targets.
1452 (main): Ditto.
1454 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1456 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1457 patch.
1458 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1460 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1462 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1463 (ffi_prep_closure_loc): New.
1464 (ffi_prep_raw_closure_loc): New.
1465 (ffi_prep_java_raw_closure_loc): New.
1466 * src/closures.c: New file.
1467 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1468 Replace sflags with exec_offset.
1469 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1470 sub_segment_exec_offset): New macros.
1471 (get_segment_flags, set_segment_flags, check_segment_merge): New
1472 macros.
1473 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1474 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1475 destroy_mspace): Use new macros.
1476 (sys_alloc): Silence warning.
1477 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1478 * Makefile.in: Rebuilt.
1479 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1480 terms of ffi_prep_closure_loc.
1481 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1482 from...
1483 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1484 renamed version.
1485 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1486 adjusted from...
1487 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1488 the renamed version.
1489 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1490 (ffi_prep_closure): ... this.
1491 * src/pa/ffi.c: Likewise.
1492 * src/cris/ffi.c: Likewise. Adjust.
1493 * src/frv/ffi.c: Likewise.
1494 * src/ia64/ffi.c: Likewise.
1495 * src/mips/ffi.c: Likewise.
1496 * src/powerpc/ffi_darwin.c: Likewise.
1497 * src/s390/ffi.c: Likewise.
1498 * src/sh/ffi.c: Likewise.
1499 * src/sh64/ffi.c: Likewise.
1500 * src/sparc/ffi.c: Likewise.
1501 * src/x86/ffi64.c: Likewise.
1502 * src/x86/ffi.c: Likewise.
1503 (FFI_INIT_TRAMPOLINE): Adjust.
1504 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1505 (ffi_prep_raw_closure): ... this.
1506 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1507 (ffi_prep_closure): ... this.
1508 (flush_icache): Adjust.
1510 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1512 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1513 Lea's malloc.
1515 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1517 * Makefile.am: Add dummy install-pdf target.
1518 * Makefile.in: Regenerate
1520 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1522 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1523 ffi_closure_helper_SYSV): Add long double handling.
1525 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1527 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1528 immediately after bctrl instruction.
1530 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1532 * Makefile.am (all-recursive, install-recursive,
1533 mostlyclean-recursive, clean-recursive, distclean-recursive,
1534 maintainer-clean-recursive): Add missing targets.
1535 * Makefile.in: Rebuilt.
1537 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1539 * configure.ac: Add TARGET for x86_64-*-darwin*.
1540 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1541 for X86_DARWIN.
1542 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1543 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1544 * configure: Regenerate.
1545 * Makefile.in: Regenerate.
1546 * include/Makefile.in: Regenerate.
1547 * testsuite/Makefile.in: Regenerate.
1548 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1549 ffi_call only.
1551 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1553 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1554 Makefile.am.
1556 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1558 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1559 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1560 Darwin.
1561 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1562 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1564 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1565 Sandro Tolaini <tolaini@libero.it>
1567 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1568 conditional.
1569 * configure: Regenerated.
1570 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1571 (EXTRA_DIST): Add src/x86/darwin.S.
1572 * Makefile.in: Regenerated.
1573 * include/Makefile.in: Regenerated.
1574 * testsuite/Makefile.in: Regenerated.
1576 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1577 X86_WIN32, and additionally align stack to 16 bytes.
1578 * src/x86/darwin.S: New, based on sysv.S.
1579 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1581 2006-09-12 David Daney <ddaney@avtrex.com>
1583 PR libffi/23935
1584 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1585 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1586 * aclocal.m4: Regenerated for automake 1.9.6.
1587 * Makefile.in: Regenerated.
1588 * include/Makefile.in: Regenerated.
1589 * testsuite/Makefile.in: Regenerated.
1591 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1593 * include/ffi_common.h (struct): Revert accidental commit.
1595 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1597 * include/ffi_common.h: Remove lint directives.
1598 * include/ffi.h.in: Likewise.
1600 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1602 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1603 for 32-bit architectures.
1604 * testsuite/libffi.call/return_ul.c: New test case.
1606 2006-07-19 David Daney <ddaney@avtrex.com>
1608 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1609 xfail remains for mips64.
1611 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1613 * Makefile.am: Add install-html target. Add install-html to .PHONY
1614 * Makefile.in: Regenerate.
1615 * aclocal.m4: Regenerate.
1616 * include/Makefile.in: Regenerate.
1617 * testsuite/Makefile.in: Regenerate.
1619 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1621 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1622 stack slot.
1624 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1626 * README: Remove notice about 'Crazy Comments'.
1627 * src/debug.c: Remove lint directives. Cleanup white spaces.
1628 * src/java_raw_api.c: Likewise.
1629 * src/prep_cif.c: Likewise.
1630 * src/raw_api.c: Likewise.
1631 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1632 to the testsuite.
1633 * src/arm/ffi.c: Remove lint directives.
1634 * src/m32r/ffi.c: Likewise.
1635 * src/pa/ffi.c: Likewise.
1636 * src/powerpc/ffi.c: Likewise.
1637 * src/powerpc/ffi_darwin.c: Likewise.
1638 * src/sh/ffi.c: Likewise.
1639 * src/sh64/ffi.c: Likewise.
1640 * src/x86/ffi.c: Likewise.
1641 * testsuite/libffi.call/float2.c: Likewise.
1642 * testsuite/libffi.call/promotion.c: Likewise.
1643 * testsuite/libffi.call/struct1.c: Likewise.
1645 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1647 * src/pa/hpux32.S: Correct unwind offset calculation for
1648 ffi_closure_pa32.
1649 * src/pa/linux.S: Likewise.
1651 2006-04-12 James E Wilson <wilson@specifix.com>
1653 PR libgcj/26483
1654 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1655 (hfa_type_load): Call stf_spill.
1656 (hfa_type_store): Call ldf_fill.
1657 (ffi_call): Adjust calls to above routines. Add local temps for
1658 macro result.
1660 2006-04-10 Matthias Klose <doko@debian.org>
1662 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1663 directory names containing underscores.
1665 2006-04-07 James E Wilson <wilson@specifix.com>
1667 * testsuite/libffi.call/float4.c: New testcase.
1669 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1670 Andreas Tobler <a.tobler@schweiz.ch>
1672 * Makefile.am: Add PA_HPUX port.
1673 * Makefile.in: Regenerate.
1674 * include/Makefile.in: Likewise.
1675 * testsuite/Makefile.in: Likewise.
1676 * configure.ac: Add PA_HPUX rules.
1677 * configure: Regenerate.
1678 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1679 Add PA_HPUX and PA64_HPUX.
1680 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1681 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1682 (FFI_TYPE_SMALL_STRUCT2): Define.
1683 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1684 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1685 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1686 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1687 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1688 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1689 * src/pa/ffi.c (ROUND_DOWN): Delete.
1690 (fldw, fstw, fldd, fstd): Use '__asm__'.
1691 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1692 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1693 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1694 Simplify incrementing of stack slot variable. Change type of local
1695 'n' to unsigned int.
1696 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1697 double on PA_HPUX.
1698 (ffi_prep_cif_machdep): Likewise.
1699 (ffi_call): Likewise.
1700 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1701 return type to ffi_status. Simplify incrementing of stack slot
1702 variable. Only copy floating point argument registers when PA_LINUX
1703 is true. Reformat debug statement.
1704 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1705 FFI_TYPE_SMALL_STRUCT8.
1706 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1707 declaration.
1708 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1709 Add nops to cache flush. Add trampoline for PA_HPUX.
1710 * src/pa/hpux32.S: New file.
1711 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1712 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1713 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1714 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1715 argument type checks so that common argument types appear first.
1716 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1717 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1719 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1721 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1722 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1723 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1724 FFI_TYPE_LONGDOUBLE.
1725 (ffi_prep_args64): Assert using IBM extended double.
1726 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1727 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1728 (ffi_call): Handle FFI_LINUX.
1729 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1730 gpr3 return pointer as for struct return. Handle FFI_LINUX
1731 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1732 unnecessarily.
1733 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1734 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1735 Don't use r6 as pointer to results, instead use sp offset. Don't
1736 make a special call to load lr with case table address, instead
1737 use offset from previous call.
1738 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1739 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1740 return.
1742 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1744 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1745 passed with FP registers correctly.
1746 (ffi_closure_helper_SYSV): Likewise.
1747 * src/sh64/sysv.S: Likewise.
1749 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1751 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1752 args and userdata unused.
1753 (closure_test_fn1): Mark cif and userdata unused.
1754 (main): Remove unused res.
1756 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1758 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1759 -O2, -O3, -Os and the warning flags -W -Wall.
1760 * testsuite/libffi.special/special.exp: Likewise.
1761 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1762 unused parameter unused for gcc or else do nothing.
1763 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1764 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1765 and userdata unused.
1766 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1767 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1768 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1769 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1770 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1771 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1772 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1773 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1774 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1775 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1776 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1777 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1778 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1779 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1780 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1781 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1782 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1783 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1784 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1785 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1786 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1787 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1788 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1789 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1790 Likewise.
1791 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1792 Likewise.
1793 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1794 Likewise.
1795 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1796 void* to avoid compiler warning.
1797 (main): Likewise.
1798 (cls_struct_align_gn): Mark cif and userdata unused.
1799 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1800 Likewise.
1801 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1802 Likewise.
1803 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1804 Likewise.
1805 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1806 Likewise.
1807 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1808 Likewise.
1809 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1810 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1811 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1812 data unused.
1813 (main): Cast res_call to silence gcc.
1814 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1815 data unused.
1816 (main): Cast res_call to silence gcc.
1817 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1818 and data unused.
1819 (main): Cast res_call to silence gcc.
1820 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1821 data unused.
1822 (main): Cast res_call to silence gcc.
1823 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1824 data unused.
1825 (main): Cast res_call to silence gcc.
1826 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1827 and data unused.
1828 (main): Cast res_call to silence gcc.
1829 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1830 userdata unused.
1831 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1832 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1833 userdata unused.
1834 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1835 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1836 userdata unused.
1837 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1838 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1839 userdata unused.
1840 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1841 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1842 userdata unused.
1843 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1844 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1845 and userdata unused.
1846 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1847 userdata unused.
1848 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1849 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1850 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1851 Cleanup white spaces.
1852 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1853 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1854 cif and userdata unused.
1855 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1856 Likewise.
1857 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1858 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1859 formatters to silence gcc.
1860 (B_gn): Mark cif and userdata unused.
1861 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1862 unused.
1863 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1864 (B_gn): Mark cif and userdata unused.
1865 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1866 unused.
1867 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1868 (B_gn): Mark cif and userdata unused.
1869 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1870 unused.
1871 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1872 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1873 * testsuite/libffi.call/problem1.c (stub): Likewise.
1874 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1875 gcc.
1876 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1877 in the last commit for this test case in the test case itself.
1878 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1879 unused.
1880 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1881 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1882 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1883 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1884 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1885 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1887 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1889 * src/sh/sysv.S: Fix register numbers in the FDE for
1890 ffi_closure_SYSV.
1892 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1894 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1895 declaration to avoid a false negative on ix86. See PR323.
1897 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1899 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1900 and cast integer to void * if needed. Update the pointer to
1901 the FP register saved area correctly.
1903 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1905 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1906 is fixed.
1907 * testsuite/libffi.call/nested_struct4.c: Likewise.
1909 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1911 * testsuite/libffi.call/return_dbl.c: New test case.
1912 * testsuite/libffi.call/return_dbl1.c: Likewise.
1913 * testsuite/libffi.call/return_dbl2.c: Likewise.
1914 * testsuite/libffi.call/return_fl.c: Likewise.
1915 * testsuite/libffi.call/return_fl1.c: Likewise.
1916 * testsuite/libffi.call/return_fl2.c: Likewise.
1917 * testsuite/libffi.call/return_fl3.c: Likewise.
1918 * testsuite/libffi.call/closure_fn6.c: Likewise.
1920 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1921 definition.
1922 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1923 here to be used by other test cases too.
1925 * testsuite/libffi.call/nested_struct10.c: New test case.
1926 * testsuite/libffi.call/nested_struct9.c: Likewise.
1927 * testsuite/libffi.call/nested_struct8.c: Likewise.
1928 * testsuite/libffi.call/nested_struct7.c: Likewise.
1929 * testsuite/libffi.call/nested_struct6.c: Likewise.
1930 * testsuite/libffi.call/nested_struct5.c: Likewise.
1931 * testsuite/libffi.call/nested_struct4.c: Likewise.
1933 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1935 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1936 * configure: Rebuilt.
1938 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1940 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1941 instead do the shifting inline.
1942 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1943 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1944 and 8 byte structs, for the remaining struct sizes don't call
1945 __lshrdi3, instead do the shifting inline.
1947 2005-12-07 Thiemo Seufer <ths@networkno.de>
1949 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1950 missing parentheses.
1951 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1952 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1953 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1954 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1955 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1956 FA_0_0_OFF2.
1957 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1958 endianness bugs.
1959 (ffi_prep_closure): Improve trampoline instruction scheduling.
1960 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1962 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1964 * src/powerpc/ffi.c: Formatting.
1965 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1966 (ffi_prep_args64): Likewise.
1968 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1970 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1971 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1973 2005-09-26 Tom Tromey <tromey@redhat.com>
1975 * testsuite/libffi.call/float1.c (value_type): New typedef.
1976 (CANARY): New define.
1977 (main): Check for result buffer overflow.
1978 * src/powerpc/linux64.S: Handle linux64 long double returns.
1979 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1980 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1982 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1984 PR target/23404
1985 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1986 homed fp args.
1987 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1989 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1991 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1992 (AH_BOTTOM): Add FFI_HIDDEN definition.
1993 * configure: Rebuilt.
1994 * fficonfig.h.in: Rebuilt.
1995 * src/powerpc/ffi.c (hidden): Remove.
1996 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1997 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1998 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1999 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
2000 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
2001 add FFI_HIDDEN to its prototype.
2002 (ffi_closure_SYSV_inner): New.
2003 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2004 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
2006 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
2008 PR libffi/21819:
2009 * configure: Rebuilt.
2010 * configure.ac: Handle i*86-*-gnu*.
2012 2005-08-09 Jakub Jelinek <jakub@redhat.com>
2014 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
2015 DW_CFA_offset_extended_sf rather than
2016 DW_CFA_GNU_negative_offset_extended.
2017 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
2019 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
2021 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
2022 on sh3.
2023 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
2024 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
2025 partially on register.
2026 (ffi_closure_helper_SYSV): Likewise.
2027 (ffi_prep_cif_machdep): Don't set too many cif->flags.
2029 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
2031 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
2032 Remove empty line.
2033 * src/sh64/ffi.c (simple_type): Remove.
2034 (return_type): Handle small structures correctly.
2035 (ffi_prep_args): Likewise.
2036 (ffi_call): Likewise.
2037 (ffi_closure_helper_SYSV): Likewise.
2038 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
2039 Emit position independent code if PIC and remove wrong datalabel
2040 prefixes from EH data.
2042 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
2044 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
2045 * Makefile.in: Regenerate.
2046 * include/Makefile.in: Likewise.
2047 * testsuite/Makefile.in: Likewise.
2048 * configure.ac: Add POWERPC_FREEBSD rules.
2049 * configure: Regenerate.
2050 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
2051 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
2052 * src/powerpc/ffi.c: Add flags to handle small structure returns
2053 in ffi_call_SYSV.
2054 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
2055 Aka FFI_SYSV.
2056 (ffi_closure_helper_SYSV): Likewise.
2057 * src/powerpc/ppc_closure.S: Add return types for small structures.
2058 * src/powerpc/sysv.S: Add bits to handle small structures for
2059 final SYSV 4 ABI.
2061 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
2063 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
2064 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
2065 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
2067 2005-07-05 Randolph Chung <tausq@debian.org>
2069 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
2070 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
2071 structures. Kill compilation warnings.
2072 (ffi_closure_inner_LINUX): Print return values as hex in debug
2073 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
2074 Properly handle 5-7 byte structure returns.
2075 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
2076 (FFI_TYPE_SMALL_STRUCT2): Remove.
2077 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
2078 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
2079 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
2080 (checksmst1, checksmst2): Remove.
2081 (checksmst3): Optimize handling of 3-byte struct returns.
2082 (checksmst567): Properly handle 5-7 byte struct returns.
2084 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2086 PR libgcj/21943
2087 * src/mips/n32.S: Enforce PIC code.
2088 * src/mips/o32.S: Likewise.
2090 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2092 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
2093 * configure: Regenerate.
2095 2005-06-01 Alan Modra <amodra@bigpond.net.au>
2097 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
2098 to call ffi_closure_helper_SYSV. Append @local instead.
2099 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
2101 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
2103 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
2104 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
2105 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
2106 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
2107 include/Makefile.in, testsuite/Makefile.in: Regenerate.
2109 2005-05-09 Mike Stump <mrs@apple.com>
2111 * configure: Regenerate.
2113 2005-05-08 Richard Henderson <rth@redhat.com>
2115 PR libffi/21285
2116 * src/alpha/osf.S: Update unwind into to match code.
2118 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
2119 Richard Henderson <rth@redhat.com>
2121 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
2122 bit 11 of flags.
2123 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
2124 (ffi_prep_closure): Set carry bit if sse-used flag set.
2125 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
2126 Only load sse registers if ssecount non-zero.
2127 (ffi_closure_unix64): Only save sse registers if carry set on entry.
2129 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
2131 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
2132 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
2133 * configure: Regenerate.
2135 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
2137 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
2138 have Tcl8.3-compatible intermediate variable.
2140 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
2141 Hans-Peter Nilsson <hp@axis.com>
2143 * Makefile.am: Add CRIS support.
2144 * configure.ac: Likewise.
2145 * Makefile.in, configure, testsuite/Makefile.in,
2146 include/Makefile.in: Regenerate.
2147 * src/cris: New directory.
2148 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
2149 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
2151 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
2152 \r?\n in output tests.
2154 2005-04-12 Mike Stump <mrs@apple.com>
2156 * configure: Regenerate.
2158 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
2160 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
2162 2005-03-30 Steve Ellcey <sje@cup.hp.com>
2164 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
2165 (ffi_sarg) Ditto.
2166 * src/ia64/unix.S (ffi_closure_unix): Extend gp
2167 to 64 bits in ILP32 mode.
2168 Load 64 bits even for short data.
2170 2005-03-23 Mike Stump <mrs@apple.com>
2172 * src/powerpc/darwin.S: Update for -m64 multilib.
2173 * src/powerpc/darwin_closure.S: Likewise.
2175 2005-03-21 Zack Weinberg <zack@codesourcery.com>
2177 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
2178 Do not set tool_include_dir.
2179 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
2180 Regenerate.
2181 * include/Makefile.am: Set gcc_version and toollibffidir.
2182 * include/Makefile.in: Regenerate.
2184 2005-02-22 Andrew Haley <aph@redhat.com>
2186 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2187 odd-numbered register pairs for 64-bit integer types.
2189 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2191 PR libffi/20104
2192 * testsuite/libffi.call/return_ll1.c: New test case.
2194 2005-02-11 Janis Johnson <janis187@us.ibm.com>
2196 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2197 * testsuite/libffi.call/float.c: Ditto.
2198 * testsuite/libffi.call/float2.c: Ditto.
2199 * testsuite/libffi.call/float3.c: Ditto.
2201 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2203 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2205 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2207 * testsuite/libffi.special/special.exp (cxx_options): Add
2208 -shared-libgcc.
2210 2004-12-31 Richard Henderson <rth@redhat.com>
2212 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2213 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2214 offset parameters with a type parameter; deduce size and structure
2215 alignment. Update all users.
2217 2004-12-31 Richard Henderson <rth@redhat.com>
2219 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2220 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
2221 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
2222 into ffi_prep_closure.
2223 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2224 from scratch.
2226 2004-12-27 Richard Henderson <rth@redhat.com>
2228 * src/x86/unix64.S: Fix typo in unwind info.
2230 2004-12-25 Richard Henderson <rth@redhat.com>
2232 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2233 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2234 (merge_classes): Check for it.
2235 (SSE_CLASS_P): New.
2236 (classify_argument): Pass byte_offset by value; perform all updates
2237 inside struct case.
2238 (examine_argument): Add classes argument; handle
2239 X86_64_COMPLEX_X87_CLASS.
2240 (ffi_prep_args): Merge into ...
2241 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2242 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2243 (ffi_fill_return_value): Remove.
2244 (ffi_prep_closure): Remove dead assert.
2245 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2246 Rewrite to use struct register_args instead of va_list. Create
2247 flags for handling structure returns.
2248 * src/x86/unix64.S: Remove dead strings.
2249 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2250 stack frame with ffi_call. Handle structure returns properly.
2251 (float2sse, floatfloat2sse, double2sse): Remove.
2252 (sse2float, sse2double, sse2floatfloat): Remove.
2253 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2254 to handle structure returns properly.
2256 2004-12-08 David Edelsohn <edelsohn@gnu.org>
2258 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2259 PICFLAG.
2260 * Makefile.in: Regenerated.
2262 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2264 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2265 * configure, aclocal.m4, Makefile.in: Regenerate.
2266 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2268 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2270 * configure: Regenerate for libtool change.
2272 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2274 * configure: Regenerate for libtool reversion.
2276 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2278 * configure: Regenerate for libtool change.
2280 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2282 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2284 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2286 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2287 of jal. Use an absolute encoding for the frame information.
2289 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2291 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2292 * acinclude.m4: Delete logic for sincludes.
2293 * aclocal.m4, Makefile.in, configure: Regenerate.
2294 * include/Makefile: Likewise.
2295 * testsuite/Makefile: Likewise.
2297 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2299 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2300 on a 8-byte boundary.
2301 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2303 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2305 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2306 long long values. Round stack allocation to a multiple of 8 bytes
2307 for ATPCS compatibility.
2308 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2309 names. Handle returning long long types. Add Thumb and interworking
2310 support. Improve soft-float code.
2312 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2314 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2315 (libffi_exit): New function.
2316 (libffi_init): Build the testglue wrapper if needed.
2318 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2320 PR other/18138
2321 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2323 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2325 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2327 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2329 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2330 * testsuite/libffi.call/float3.c: New test case.
2332 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2334 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2335 the function returning a structure pointed with R2.
2336 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2337 the structure return value if T bit set. Emit position
2338 independent code and EH data if PIC.
2340 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2342 * Makefile.am: Add m32r support.
2343 * configure.ac: Likewise.
2344 * Makefile.in: Regenerate.
2345 * confiugre: Regenerate.
2346 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2347 (uint64, sint64, double, longdouble)
2348 * src/m32r: New directory.
2349 * src/m32r/ffi.c: New file.
2350 * src/m32r/sysv.S: Likewise.
2351 * src/m32r/ffitarget.h: Likewise.
2353 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2355 * testsuite/libffi.call/negint.c: New test case.
2357 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2359 PR libgcj/17465
2360 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2361 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2362 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2363 DYLD_LIBRARY_PATH.
2365 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2367 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2368 * testsuite/libffi.call/promotion.c: Likewise.
2369 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2370 whitespaces.
2371 * testsuite/libffi.call/return_sc.c: Likewise.
2372 * testsuite/libffi.call/return_uc.c: Likewise.
2374 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2376 * src/powerpc/darwin.S: Fix comments and identation.
2377 * src/powerpc/darwin_closure.S: Likewise.
2379 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2381 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2382 (ffi_prep_args): Handle longdouble arguments.
2383 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2384 longdouble.
2385 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2386 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2387 values.
2388 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2389 * src/types.c: Defined longdouble size and alignment for darwin.
2391 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2393 * src/powerpc/aix.S: Remove whitespaces.
2394 * src/powerpc/aix_closure.S: Likewise.
2395 * src/powerpc/asm.h: Likewise.
2396 * src/powerpc/ffi.c: Likewise.
2397 * src/powerpc/ffitarget.h: Likewise.
2398 * src/powerpc/linux64.S: Likewise.
2399 * src/powerpc/linux64_closure.S: Likewise.
2400 * src/powerpc/ppc_closure.S: Likewise.
2401 * src/powerpc/sysv.S: Likewise.
2403 2004-08-30 Anthony Green <green@redhat.com>
2405 * Makefile.am: Add frv support.
2406 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2407 * configure.ac: Read configure.host.
2408 * configure.in: Read configure.host.
2409 * configure.host: New file. frv-elf needs libgloss.
2410 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2411 alignment. This is needed to frv and shouldn't harm the others.
2412 * include/ffi_common.h (ALIGN_DOWN): New macro.
2413 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2415 2004-08-24 David Daney <daney@avtrex.com>
2417 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2418 * testsuite/libffi.call/closure_fn1.c: Likewise.
2419 * testsuite/libffi.call/closure_fn2.c Likewise.
2420 * testsuite/libffi.call/closure_fn3.c: Likewise.
2421 * testsuite/libffi.call/closure_fn4.c: Likewise.
2422 * testsuite/libffi.call/closure_fn5.c: Likewise.
2423 * testsuite/libffi.call/cls_18byte.c: Likewise.
2424 * testsuite/libffi.call/cls_19byte.c: Likewise.
2425 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2426 * testsuite/libffi.call/cls_20byte.c: Likewise.
2427 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2428 * testsuite/libffi.call/cls_24byte.c: Likewise.
2429 * testsuite/libffi.call/cls_2byte.c: Likewise.
2430 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2431 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2432 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2433 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2434 * testsuite/libffi.call/cls_4byte.c: Likewise.
2435 * testsuite/libffi.call/cls_64byte.c: Likewise.
2436 * testsuite/libffi.call/cls_6byte.c: Likewise.
2437 * testsuite/libffi.call/cls_7byte.c: Likewise.
2438 * testsuite/libffi.call/cls_8byte.c: Likewise.
2439 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2440 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2441 * testsuite/libffi.call/cls_align_double.c: Likewise.
2442 * testsuite/libffi.call/cls_align_float.c: Likewise.
2443 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2444 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2445 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2446 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2447 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2448 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2449 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2450 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2451 * testsuite/libffi.call/cls_double.c: Likewise.
2452 * testsuite/libffi.call/cls_float.c: Likewise.
2453 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2454 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2455 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2456 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2457 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2458 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2459 * testsuite/libffi.call/cls_schar.c: Likewise.
2460 * testsuite/libffi.call/cls_sint.c: Likewise.
2461 * testsuite/libffi.call/cls_sshort.c: Likewise.
2462 * testsuite/libffi.call/cls_uchar.c: Likewise.
2463 * testsuite/libffi.call/cls_uint.c: Likewise.
2464 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2465 * testsuite/libffi.call/cls_ushort.c: Likewise.
2466 * testsuite/libffi.call/nested_struct.c: Likewise.
2467 * testsuite/libffi.call/nested_struct1.c: Likewise.
2468 * testsuite/libffi.call/nested_struct2.c: Likewise.
2469 * testsuite/libffi.call/nested_struct3.c: Likewise.
2470 * testsuite/libffi.call/problem1.c: Likewise.
2471 * testsuite/libffi.special/unwindtest.cc: Likewise.
2472 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2473 to zero.
2474 * testsuite/libffi.call/cls_16byte.c: Likewise.
2475 * testsuite/libffi.call/cls_5byte.c: Likewise.
2477 2004-08-23 David Daney <daney@avtrex.com>
2479 PR libgcj/13141
2480 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2481 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2482 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2483 parameters and return types.
2484 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2485 (ffi_prep_closure): Ditto.
2486 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2487 alignment calculations.
2488 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2489 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2491 2004-08-14 Casey Marshall <csm@gnu.org>
2493 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2494 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2495 integer (O32 ABI only).
2496 (ffi_prep_closure): new function.
2497 (ffi_closure_mips_inner_O32): new function.
2498 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2499 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2500 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2501 64 bit integers correctly.
2502 (ffi_closure_O32): new function.
2503 Added DWARF-2 unwind info for both functions.
2505 2004-08-10 Andrew Haley <aph@redhat.com>
2507 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2509 2004-08-01 Robert Millan <robertmh@gnu.org>
2511 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2512 * configure: Regenerate.
2514 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2516 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2517 and mmap() explicitly instead of relying on preset autoconf cache
2518 variables.
2519 * aclocal.m4: Regenerate.
2520 * configure: Regenerate.
2522 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2524 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2525 (ffi_check_float_struct): Remove unused prototype.
2527 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2529 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2530 character on Darwin, use '\n\t' instead.
2532 2004-06-26 Matthias Klose <doko@debian.org>
2534 * libtool-version: Fix typo in revision/age.
2536 2004-06-17 Matthias Klose <doko@debian.org>
2538 * libtool-version: New.
2539 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2540 * Makefile.in: Regenerate.
2542 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2544 * Makefile.am: Remove useless multilib rules.
2545 * Makefile.in: Regenerate.
2546 * aclocal.m4: Regenerate with automake 1.8.5.
2547 * configure.ac: Remove useless multilib configury.
2548 * configure: Regenerate.
2550 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2552 * .cvsignore: New file.
2554 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2556 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2557 fp_done.
2558 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2559 changed from 8.
2561 2004-06-06 Sean McNeil <sean@mcneil.com>
2563 * configure.ac: Add x86_64-*-freebsd* support.
2564 * configure: Regenerate.
2566 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2568 Bug 15093
2569 * configure.ac: Test for existence of mmap and sys/mman.h before
2570 checking blacklist. Fix suggested by Jim Wilson.
2571 * configure: Regenerate.
2573 2004-04-26 Matt Austern <austern@apple.com>
2575 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2576 FDE location.
2577 * src/powerpc/darwin_closure.S: Likewise.
2579 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2581 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2582 error. Reported by Thomas Heller <theller@python.net>.
2583 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2584 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2586 2004-03-20 Matthias Klose <doko@debian.org>
2588 * src/pa/linux.S: Fix typo.
2590 2004-03-19 Matthias Klose <doko@debian.org>
2592 * Makefile.am: Update.
2593 * Makefile.in: Regenerate.
2594 * src/pa/ffi.h.in: Remove.
2595 * src/pa/ffitarget.h: New file.
2597 2004-02-10 Randolph Chung <tausq@debian.org>
2599 * Makefile.am: Add PA support.
2600 * Makefile.in: Regenerate.
2601 * include/Makefile.in: Regenerate.
2602 * configure.ac: Add PA target.
2603 * configure: Regenerate.
2604 * src/pa/ffi.c: New file.
2605 * src/pa/ffi.h.in: Add PA support.
2606 * src/pa/linux.S: New file.
2607 * prep_cif.c: Add PA support.
2609 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2611 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2612 double.
2613 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2614 with ecif->cif->flags.
2615 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2616 with cif->flags.
2617 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2618 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2619 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2620 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2622 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2624 * configure.in: Rename file to ...
2625 * configure.ac: ... this.
2626 * fficonfig.h.in: Regenerate.
2627 * Makefile.in: Regenerate.
2628 * include/Makefile.in: Regenerate.
2629 * testsuite/Makefile.in: Regenerate.
2631 2004-03-12 Matt Austern <austern@apple.com>
2633 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2634 changes in EH format resulting from addition of linkonce support.
2635 * src/powerpc/darwin_closure.S: Likewise.
2637 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2638 Paolo Bonzini <bonzini@gnu.org>
2640 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2641 Remove VPATH. Remove rules for object files. Remove multilib support.
2642 (AM_CCASFLAGS): Add.
2643 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2644 (AC_PREREQ): Bump version to 2.59.
2645 (AC_INIT): Fill with version info and bug address.
2646 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2647 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2648 De-precious CC so that the right flags are passed down to multilibs.
2649 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2650 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2651 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2652 * configure: Rebuilt.
2653 * aclocal.m4: Likewise.
2654 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2655 * fficonfig.h.in: Likewise.
2657 2004-03-11 Andreas Schwab <schwab@suse.de>
2659 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2660 arguments from fp registers only for the first 8 parameter slots.
2661 Don't convert a float parameter when passed in memory.
2663 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2665 * configure: Regenerate for config/accross.m4 correction.
2667 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2669 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2670 ecif->cif->bytes to bytes.
2671 (ffi_prep_cif_machdep): Add braces around nested if statement.
2673 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2675 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2677 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2678 (ffi_closure_helper_LINUX64): Fix typo.
2679 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2680 for powerpc64-*-*.
2681 * testsuite/libffi.call/float.c: Likewise.
2682 * testsuite/libffi.call/float2.c: Likewise.
2684 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2686 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2687 long double function return and long double arg handling.
2688 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2689 Use "end_pfr" instead of "nf". Correct long double handling.
2690 Localise "temp".
2691 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2692 return value.
2693 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2694 space for long double return value. Adjust stack frame and offsets.
2695 Load f2 long double return.
2697 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2699 * src/types.c: Use 16 byte long double for POWERPC64.
2701 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2703 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2704 when the structure return address is passed in %o0.
2705 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2706 (ffi_v9_layout_struct): Align the field following a nested structure
2707 on a word boundary. Use memmove instead of memcpy.
2708 (ffi_call): Update call to ffi_V9_return_struct.
2709 (ffi_prep_closure): Define 'ctx' only for V8.
2710 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2711 and ffi_closure_sparc_inner_v9.
2712 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2713 Always skip the structure return address. For structures and long
2714 doubles, copy the argument directly.
2715 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2716 if required. Shift the maximum floating-point slot accordingly. For
2717 big structures, copy the argument directly; otherwise, left-justify the
2718 argument and call ffi_v9_layout_struct to lay out the structure on
2719 the stack.
2720 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2721 (ffi_closure_v8): Pass the structure return address. Update call to
2722 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2723 Skip the 'unimp' insn when returning long doubles and structures.
2724 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2725 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2726 FFI_TYPE_INT handling. Load structures both in integers and
2727 floating-point registers on return.
2728 * README: Update status of the SPARC port.
2730 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2732 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2733 as of type ffi_arg.
2734 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2736 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2738 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2739 value as of type ffi_arg, not unsigned int.
2741 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2743 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2744 of the LHS.
2746 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2748 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2749 Solaris.
2751 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2753 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2754 to void *.
2756 2003-12-10 Richard Henderson <rth@redhat.com>
2758 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2759 size_t instead of int.
2761 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2763 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2764 * testsuite/libffi.call/many_win32.c (main): Replace variable
2765 int i with unsigned long ul.
2767 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2768 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2769 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2770 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2771 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2772 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2773 * testsuite/libffi.call/cls_align_float.c: Likewise.
2774 * testsuite/libffi.call/cls_align_double.c: Likewise.
2775 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2776 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2778 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2780 PR other/13221
2781 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2782 Align arguments to 32 bits.
2784 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2786 PR other/13221
2787 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2788 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2789 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2790 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2791 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2792 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2794 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2796 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2798 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2799 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2801 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2803 * Makefile.in: Rebuilt.
2804 * configure: Likewise.
2805 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2806 the right type.
2808 2003-11-21 Andreas Jaeger <aj@suse.de>
2809 Andreas Tobler <a.tobler@schweiz.ch>
2811 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2812 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2813 * Makefile.in: Rebuilt.
2814 * aclocal.m4: Likewise.
2815 * configure: Likewise.
2816 * fficonfig.h.in: Likewise.
2817 * testsuite/lib/libffi-dg.exp: Add include dir.
2818 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2819 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2820 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2821 for ffi_closure if available.
2822 * testsuite/libffi.call/closure_fn1.c: Likewise.
2823 * testsuite/libffi.call/closure_fn2.c: Likewise.
2824 * testsuite/libffi.call/closure_fn3.c: Likewise.
2825 * testsuite/libffi.call/closure_fn4.c: Likewise.
2826 * testsuite/libffi.call/closure_fn5.c: Likewise.
2827 * testsuite/libffi.call/cls_12byte.c: Likewise.
2828 * testsuite/libffi.call/cls_16byte.c: Likewise.
2829 * testsuite/libffi.call/cls_18byte.c: Likewise.
2830 * testsuite/libffi.call/cls_19byte.c: Likewise.
2831 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2832 * testsuite/libffi.call/cls_20byte.c: Likewise.
2833 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2834 * testsuite/libffi.call/cls_24byte.c: Likewise.
2835 * testsuite/libffi.call/cls_2byte.c: Likewise.
2836 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2837 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2838 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2839 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2840 * testsuite/libffi.call/cls_4byte.c: Likewise.
2841 * testsuite/libffi.call/cls_5byte.c: Likewise.
2842 * testsuite/libffi.call/cls_64byte.c: Likewise.
2843 * testsuite/libffi.call/cls_6byte.c: Likewise.
2844 * testsuite/libffi.call/cls_7byte.c: Likewise.
2845 * testsuite/libffi.call/cls_8byte.c: Likewise.
2846 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2847 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2848 * testsuite/libffi.call/cls_double.c: Likewise.
2849 * testsuite/libffi.call/cls_float.c: Likewise.
2850 * testsuite/libffi.call/cls_schar.c: Likewise.
2851 * testsuite/libffi.call/cls_sint.c: Likewise.
2852 * testsuite/libffi.call/cls_sshort.c: Likewise.
2853 * testsuite/libffi.call/cls_uchar.c: Likewise.
2854 * testsuite/libffi.call/cls_uint.c: Likewise.
2855 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2856 * testsuite/libffi.call/cls_ushort.c: Likewise.
2857 * testsuite/libffi.call/nested_struct.c: Likewise.
2858 * testsuite/libffi.call/nested_struct1.c: Likewise.
2859 * testsuite/libffi.call/nested_struct2.c: Likewise.
2860 * testsuite/libffi.call/nested_struct3.c: Likewise.
2861 * testsuite/libffi.call/problem1.c: Likewise.
2862 * testsuite/libffi.special/unwindtest.cc: Likewise.
2864 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2866 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2868 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2870 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2871 Add -lgcc_s to additional flags.
2873 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2875 * configure.in, include/Makefile.am: PR libgcj/11147, install
2876 the ffitarget.h header file in a gcc versioned and target
2877 dependent place.
2878 * configure: Regenerated.
2879 * Makefile.in, include/Makefile.in: Likewise.
2880 * testsuite/Makefile.in: Likewise.
2882 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2884 * testsuite/libffi.call/closure_fn0.c: Print result and check
2885 with dg-output to make debugging easier.
2886 * testsuite/libffi.call/closure_fn1.c: Likewise.
2887 * testsuite/libffi.call/closure_fn2.c: Likewise.
2888 * testsuite/libffi.call/closure_fn3.c: Likewise.
2889 * testsuite/libffi.call/closure_fn4.c: Likewise.
2890 * testsuite/libffi.call/closure_fn5.c: Likewise.
2891 * testsuite/libffi.call/cls_12byte.c: Likewise.
2892 * testsuite/libffi.call/cls_16byte.c: Likewise.
2893 * testsuite/libffi.call/cls_18byte.c: Likewise.
2894 * testsuite/libffi.call/cls_19byte.c: Likewise.
2895 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2896 * testsuite/libffi.call/cls_20byte.c: Likewise.
2897 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2898 * testsuite/libffi.call/cls_24byte.c: Likewise.
2899 * testsuite/libffi.call/cls_2byte.c: Likewise.
2900 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2901 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2902 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2903 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2904 * testsuite/libffi.call/cls_4byte.c: Likewise.
2905 * testsuite/libffi.call/cls_5byte.c: Likewise.
2906 * testsuite/libffi.call/cls_64byte.c: Likewise.
2907 * testsuite/libffi.call/cls_6byte.c: Likewise.
2908 * testsuite/libffi.call/cls_7byte.c: Likewise.
2909 * testsuite/libffi.call/cls_8byte.c: Likewise.
2910 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2911 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2912 * testsuite/libffi.call/cls_double.c: Likewise.
2913 * testsuite/libffi.call/cls_float.c: Likewise.
2914 * testsuite/libffi.call/cls_schar.c: Likewise.
2915 * testsuite/libffi.call/cls_sint.c: Likewise.
2916 * testsuite/libffi.call/cls_sshort.c: Likewise.
2917 * testsuite/libffi.call/cls_uchar.c: Likewise.
2918 * testsuite/libffi.call/cls_uint.c: Likewise.
2919 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2920 * testsuite/libffi.call/cls_ushort.c: Likewise.
2921 * testsuite/libffi.call/problem1.c: Likewise.
2923 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2924 static.
2926 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2928 * testsuite/libffi.call/cls_9byte2.c: New test case.
2929 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2930 * testsuite/libffi.call/cls_64byte.c: Likewise.
2931 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2932 * testsuite/libffi.call/cls_19byte.c: Likewise.
2933 * testsuite/libffi.call/cls_18byte.c: Likewise.
2934 * testsuite/libffi.call/closure_fn4.c: Likewise.
2935 * testsuite/libffi.call/closure_fn5.c: Likewise.
2936 * testsuite/libffi.call/cls_schar.c: Likewise.
2937 * testsuite/libffi.call/cls_sint.c: Likewise.
2938 * testsuite/libffi.call/cls_sshort.c: Likewise.
2939 * testsuite/libffi.call/nested_struct2.c: Likewise.
2940 * testsuite/libffi.call/nested_struct3.c: Likewise.
2942 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2944 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2945 * testsuite/libffi.call/cls_uchar.c: Likewise.
2946 * testsuite/libffi.call/cls_uint.c: Likewise.
2947 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2948 * testsuite/libffi.call/cls_ushort.c: Likewise.
2949 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2951 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2953 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2954 the initialization.
2956 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2958 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2959 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2961 2003-10-22 David Daney <ddaney@avtrex.com>
2963 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2964 __attribute__((__mode__(__SI__))) and friends.
2966 2003-10-22 Andreas Schwab <schwab@suse.de>
2968 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2970 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2972 * configure.in: AC_LINK_FILES(ffitarget.h).
2973 * configure: Regenerate.
2974 * Makefile.in: Likewise.
2975 * include/Makefile.in: Likewise.
2976 * testsuite/Makefile.in: Likewise.
2977 * fficonfig.h.in: Likewise.
2979 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2980 Richard Henderson <rth@redhat.com>
2982 Avoid that ffi.h includes fficonfig.h.
2984 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2985 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2986 (TARGET_SRC_MIPS_SGI): Removed.
2987 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2988 (MIPS_SGI): Removed.
2989 (CLEANFILES): Removed.
2990 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2991 targets.
2992 * acconfig.h: Removed.
2993 * configure.in: Compute sizeofs only for double and long double.
2994 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2995 into AC_DEFINE instead of using acconfig.h. Create
2996 include/ffitarget.h instead of include/fficonfig.h. Rename
2997 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2998 AC_DEFINE EH_FRAME_FLAGS.
2999 * include/Makefile.am (DISTCLEANFILES): New automake macro.
3000 (hack_DATA): Add ffitarget.h.
3001 * include/ffi.h.in: Remove all system specific definitions.
3002 Declare raw API even if it is not installed, why bother?
3003 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
3004 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
3005 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
3006 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
3007 * include/ffi_common.h (bool): Do not define.
3008 (ffi_assert): Accept failed assertion.
3009 (ffi_type_test): Return void and accept file/line.
3010 (FFI_ASSERT): Pass stringized failed assertion.
3011 (FFI_ASSERT_AT): New macro.
3012 (FFI_ASSERT_VALID_TYPE): New macro.
3013 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
3014 UINT64, SINT64): Define here with gcc's __attribute__ macro
3015 instead of in ffi.h
3016 (FLOAT32, ALIGN): Define here instead of in ffi.h
3017 * include/ffi-mips.h: Removed. Its content moved to
3018 src/mips/ffitarget.h after separating assembly and C sections.
3019 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
3020 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
3021 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
3022 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
3023 SIZEOF_ARG -> FFI_SIZEOF_ARG.
3024 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
3025 * src/debug.c (ffi_assert): Accept stringized failed assertion.
3026 (ffi_type_test): Rewritten.
3027 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
3028 FFI_ASSERT_VALID_TYPE.
3029 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
3030 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
3031 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
3032 src/s390/ffitarget.h, src/sh/ffitarget.h,
3033 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
3034 src/x86/ffitarget.h: New files.
3035 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
3036 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
3037 src/powerpc/aix.S, src/powerpc/darwin.S,
3038 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
3039 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
3040 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
3041 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
3042 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
3043 include fficonfig.h
3045 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3047 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
3048 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
3050 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
3052 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
3053 Used when FFI_DEBUG = 1.
3055 2003-10-14 Alan Modra <amodra@bigpond.net.au>
3057 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
3058 and align.
3060 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3062 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
3063 FFI_MIPS_O32 for O32 ABI.
3065 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
3067 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
3068 SPARC64. Cleanup whitespaces.
3070 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
3072 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
3073 strongarm, xscale. Cleanup whitespaces.
3074 * testsuite/libffi.call/closure_fn1.c: Likewise.
3075 * testsuite/libffi.call/closure_fn2.c: Likewise.
3076 * testsuite/libffi.call/closure_fn3.c: Likewise.
3077 * testsuite/libffi.call/cls_12byte.c: Likewise.
3078 * testsuite/libffi.call/cls_16byte.c: Likewise.
3079 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3080 * testsuite/libffi.call/cls_20byte.c: Likewise.
3081 * testsuite/libffi.call/cls_24byte.c: Likewise.
3082 * testsuite/libffi.call/cls_2byte.c: Likewise.
3083 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3084 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3085 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3086 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3087 * testsuite/libffi.call/cls_4byte.c: Likewise.
3088 * testsuite/libffi.call/cls_5byte.c: Likewise.
3089 * testsuite/libffi.call/cls_6byte.c: Likewise.
3090 * testsuite/libffi.call/cls_7byte.c: Likewise.
3091 * testsuite/libffi.call/cls_8byte.c: Likewise.
3092 * testsuite/libffi.call/cls_double.c: Likewise.
3093 * testsuite/libffi.call/cls_float.c: Likewise.
3094 * testsuite/libffi.call/cls_uchar.c: Likewise.
3095 * testsuite/libffi.call/cls_uint.c: Likewise.
3096 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3097 * testsuite/libffi.call/cls_ushort.c: Likewise.
3098 * testsuite/libffi.call/nested_struct.c: Likewise.
3099 * testsuite/libffi.call/nested_struct1.c: Likewise.
3100 * testsuite/libffi.call/problem1.c: Likewise.
3101 * testsuite/libffi.special/unwindtest.cc: Likewise.
3102 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
3104 2003-09-18 David Edelsohn <edelsohn@gnu.org>
3106 * src/powerpc/aix.S: Cleanup whitespaces.
3107 * src/powerpc/aix_closure.S: Likewise.
3109 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3111 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
3112 * src/powerpc/darwin_closure.S: Likewise.
3113 * src/powerpc/ffi_darwin.c: Likewise.
3115 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3116 David Edelsohn <edelsohn@gnu.org>
3118 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
3119 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
3120 parameter stack.
3121 * src/powerpc/darwin_closure.S: Likewise.
3122 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
3123 according to the Darwin/AIX ABI.
3124 (ffi_prep_cif_machdep): Likewise.
3125 (ffi_closure_helper_DARWIN): Likewise.
3126 Remove the outgoing parameter stack logic. Simplify the evaluation
3127 of the different CASE types.
3128 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
3129 statement in the trampoline code.
3131 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
3133 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
3134 for the register size.
3135 (ffi_closure_helper_SYSV): Handle the structure return value
3136 address correctly.
3137 (ffi_closure_helper_SYSV): Return the appropriate type when
3138 the registers are used for the structure return value.
3139 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
3140 the 64-bit return value. Update copyright years.
3142 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3144 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
3145 srcdir for ffi_mips.h.
3147 2003-09-12 Alan Modra <amodra@bigpond.net.au>
3149 * src/prep_cif.c (initialize_aggregate): Include tail padding in
3150 structure size.
3151 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
3152 placement of float result.
3153 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
3154 cast of "resp" for big-endian 64 bit machines.
3156 2003-09-11 Alan Modra <amodra@bigpond.net.au>
3158 * src/types.c (double, longdouble): Merge identical SH and ARM
3159 typedefs, and add POWERPC64.
3160 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
3161 struct split over gpr and rest.
3162 (ffi_prep_cif_machdep): Correct intarg_count for structures.
3163 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
3165 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
3167 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
3168 passing correctly.
3170 2003-09-09 Alan Modra <amodra@bigpond.net.au>
3172 * configure: Regenerate.
3174 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3176 * Makefile.am: Remove build rules for ffitest.
3177 * Makefile.in: Rebuilt.
3179 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3181 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3182 about implicit declaration of abort().
3184 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3186 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3187 * Makefile.in: Rebuilt.
3188 * configure.in: Add dejagnu test framework.
3189 * configure: Rebuilt.
3191 * testsuite/Makefile.am: New file.
3192 * testsuite/Makefile.in: Built
3193 * testsuite/lib/libffi-dg.exp: New file.
3194 * testsuite/config/default.exp: Likewise.
3195 * testsuite/libffi.call/call.exp: Likewise.
3196 * testsuite/libffi.call/ffitest.h: Likewise.
3197 * testsuite/libffi.call/closure_fn0.c: Likewise.
3198 * testsuite/libffi.call/closure_fn1.c: Likewise.
3199 * testsuite/libffi.call/closure_fn2.c: Likewise.
3200 * testsuite/libffi.call/closure_fn3.c: Likewise.
3201 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3202 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3203 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3204 * testsuite/libffi.call/cls_2byte.c: Likewise.
3205 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3206 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3207 * testsuite/libffi.call/cls_4byte.c: Likewise.
3208 * testsuite/libffi.call/cls_5byte.c: Likewise.
3209 * testsuite/libffi.call/cls_6byte.c: Likewise.
3210 * testsuite/libffi.call/cls_7byte.c: Likewise.
3211 * testsuite/libffi.call/cls_8byte.c: Likewise.
3212 * testsuite/libffi.call/cls_12byte.c: Likewise.
3213 * testsuite/libffi.call/cls_16byte.c: Likewise.
3214 * testsuite/libffi.call/cls_20byte.c: Likewise.
3215 * testsuite/libffi.call/cls_24byte.c: Likewise.
3216 * testsuite/libffi.call/cls_double.c: Likewise.
3217 * testsuite/libffi.call/cls_float.c: Likewise.
3218 * testsuite/libffi.call/cls_uchar.c: Likewise.
3219 * testsuite/libffi.call/cls_uint.c: Likewise.
3220 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3221 * testsuite/libffi.call/cls_ushort.c: Likewise.
3222 * testsuite/libffi.call/float.c: Likewise.
3223 * testsuite/libffi.call/float1.c: Likewise.
3224 * testsuite/libffi.call/float2.c: Likewise.
3225 * testsuite/libffi.call/many.c: Likewise.
3226 * testsuite/libffi.call/many_win32.c: Likewise.
3227 * testsuite/libffi.call/nested_struct.c: Likewise.
3228 * testsuite/libffi.call/nested_struct1.c: Likewise.
3229 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3230 * testsuite/libffi.call/problem1.c: Likewise.
3231 * testsuite/libffi.call/promotion.c: Likewise.
3232 * testsuite/libffi.call/return_ll.c: Likewise.
3233 * testsuite/libffi.call/return_sc.c: Likewise.
3234 * testsuite/libffi.call/return_uc.c: Likewise.
3235 * testsuite/libffi.call/strlen.c: Likewise.
3236 * testsuite/libffi.call/strlen_win32.c: Likewise.
3237 * testsuite/libffi.call/struct1.c: Likewise.
3238 * testsuite/libffi.call/struct2.c: Likewise.
3239 * testsuite/libffi.call/struct3.c: Likewise.
3240 * testsuite/libffi.call/struct4.c: Likewise.
3241 * testsuite/libffi.call/struct5.c: Likewise.
3242 * testsuite/libffi.call/struct6.c: Likewise.
3243 * testsuite/libffi.call/struct7.c: Likewise.
3244 * testsuite/libffi.call/struct8.c: Likewise.
3245 * testsuite/libffi.call/struct9.c: Likewise.
3246 * testsuite/libffi.special/special.exp: New file.
3247 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3248 * testsuite/libffi.special/unwindtest.cc: Likewise.
3251 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3253 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3254 copyright years.
3256 2003-08-02 Alan Modra <amodra@bigpond.net.au>
3258 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3259 structure passing.
3260 (ffi_closure_helper_LINUX64): Likewise.
3261 * src/powerpc/linux64.S: Remove code writing to parm save area.
3262 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3263 address in lr from ffi_closure_helper_LINUX64 call to calculate
3264 table address. Optimize function tail.
3266 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3268 * src/sparc/ffi.c: Handle all floating point registers.
3269 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3271 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3273 * README: Note that libffi is not part of GCC. Update the project
3274 URL and status.
3276 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3278 * src/powerpc/ppc_closure.S: Include ffi.h.
3280 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3282 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3283 Use C style comments.
3285 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3287 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3288 * Makefile.in: Regenerate.
3289 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3290 * configure: Regenerate.
3291 * include/ffi.h.in: Add SHmedia support.
3292 * src/sh64/ffi.c: New file.
3293 * src/sh64/sysv.S: New file.
3295 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3297 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3298 should be read-only.
3299 * configure: Rebuilt.
3300 * fficonfig.h.in: Rebuilt.
3301 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3302 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3303 * src/powerpc/linux64.S: Likewise.
3304 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3305 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3306 if -fpic/-fPIC/-mrelocatable.
3307 * src/powerpc/powerpc_closure.S: Likewise.
3308 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3309 #write in .eh_frame flags.
3310 * src/sparc/v9.S: Likewise.
3311 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3312 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3313 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3315 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3317 Fixes PR bootstrap/10656
3318 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3319 support for .register pseudo-op.
3320 * src/sparc/v8.S: Use it.
3321 * fficonfig.h.in: Rebuilt.
3322 * configure: Rebuilt.
3324 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3326 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3327 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3328 Make it the default on POWERPC64.
3329 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3330 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3331 * configure: Rebuilt.
3332 * src/powerpc/ffi.c (hidden): Define.
3333 (ffi_prep_args_SYSV): Renamed from
3334 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3335 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3336 ASM_NEEDS_REGISTERS64): New.
3337 (ffi_prep_args64): New function.
3338 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3339 (ffi_call): Likewise.
3340 (ffi_prep_closure): Likewise.
3341 (flush_icache): Surround by #ifndef POWERPC64.
3342 (ffi_dblfl): New union type.
3343 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3344 (ffi_closure_helper_LINUX64): New function.
3345 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3346 __powerpc64__.
3347 * src/powerpc/sysv.S: Likewise.
3348 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3349 * src/powerpc/linux64.S: New file.
3350 * src/powerpc/linux64_closure.S: New file.
3351 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3352 src/powerpc/linux64_closure.S.
3353 (TARGET_SRC_POWERPC): Likewise.
3355 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3356 closure_test_fn3): Fix result printing on big-endian 64-bit
3357 machines.
3358 (main): Print tst2_arg instead of uninitialized tst2_result.
3360 * src/ffitest.c (main): Hide what closure pointer really points to
3361 from the compiler.
3363 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3365 * configure.in (arm-*-netbsdelf*): Add configuration.
3366 (configure): Regenerated.
3368 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3370 * include/Makefile.in: Regenerate.
3372 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3374 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3375 bit mode.
3376 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3377 Receive closure pointer through parameter, read args using
3378 __builtin_dwarf_cfa.
3379 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3381 2003-03-12 Andreas Schwab <schwab@suse.de>
3383 * configure.in: Avoid trailing /. in toolexeclibdir.
3384 * configure: Rebuilt.
3386 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3388 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3390 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3392 * libffi/src/powerpc/darwin_closure.S:
3393 Fix alignement bug, allocate 8 bytes for the result.
3394 * libffi/src/powerpc/aix_closure.S:
3395 Likewise.
3396 * libffi/src/powerpc/ffi_darwin.c:
3397 Update stackframe description for aix/darwin_closure.S.
3399 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3401 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3402 attribute.
3404 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3405 Andreas Schwab <schwab@suse.de>
3407 * configure.in: Adjust command to source config-ml.in to account
3408 for changes to the libffi_basedir definition.
3409 (libffi_basedir): Remove ${srcdir} from value and include trailing
3410 slash if nonempty.
3412 * configure: Regenerate.
3414 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3416 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3418 2003-01-28 Andrew Haley <aph@redhat.com>
3420 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3421 * src/x86/ffi64.c (ffi_prep_closure): New.
3422 (ffi_closure_UNIX64_inner): New.
3423 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3425 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
3427 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3428 Remove USE_LIBDIR conditional.
3429 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3430 * Makefile.in, configure: Rebuilt.
3432 2003-01027 David Edelsohn <edelsohn@gnu.org>
3434 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3435 * Makefile.in: Regenerate.
3437 2003-01-22 Andrew Haley <aph@redhat.com>
3439 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3440 unwind info.
3442 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3444 * src/powerpc/darwin.S: Add unwind info.
3445 * src/powerpc/darwin_closure.S: Likewise.
3447 2003-01-14 Andrew Haley <aph@redhat.com>
3449 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3450 (ffi_prep_cif_machdep): Likewise.
3451 * src/x86/unix64.S: Add unwind info.
3453 2003-01-14 Andreas Jaeger <aj@suse.de>
3455 * src/ffitest.c (main): Only use ffi_closures if those are
3456 supported.
3458 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3460 * libffi/src/ffitest.c
3461 add closure testcases
3463 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3465 * libffi/src/powerpc/ffi.c
3466 fix alignment bug for float (4 byte aligned iso 8 byte)
3468 2003-01-09 Geoffrey Keating <geoffk@apple.com>
3470 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3471 * src/powerpc/darwin.S: Remove RCS version string.
3473 2003-01-03 Jeff Sturm <jsturm@one-point.com>
3475 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3476 * src/ffitest.c (main): Use static storage for closure.
3477 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3478 * src/sparc/v8.S (ffi_closure_v8): New.
3479 * src/sparc/v9.S (ffi_closure_v9): New.
3481 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3483 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3484 enumeration for X86_WIN32.
3485 * src/x86/win32.S: Added ffi_call_STDCALL function
3486 definition.
3487 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3488 switch cases for recognising FFI_STDCALL and
3489 calling ffi_call_STDCALL if target is X86_WIN32.
3490 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3491 stdcall versions of the "my_strlen" and "many"
3492 test functions (for X86_WIN32).
3493 Added test cases to test stdcall invocation using
3494 these functions.
3496 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3498 * src/sh/sysv.S: Add DWARF2 unwind info.
3500 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3502 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3504 2002-11-26 Jim Wilson <wilson@redhat.com>
3506 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3508 2002-11-23 H.J. Lu <hjl@gnu.org>
3510 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3511 Include ../config/accross.m4.
3512 * aclocal.m4; Rebuild.
3513 * configure: Likewise.
3515 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3517 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3519 2002-11-11 DJ Delorie <dj@redhat.com>
3521 * configure.in: Look for common files in the right place.
3523 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3525 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3526 raw data as _Jv_word values, not ffi_raw.
3527 (ffi_java_ptrarray_to_raw): Likewise.
3528 (ffi_java_rvalue_to_raw): New function.
3529 (ffi_java_raw_call): Call it.
3530 (ffi_java_raw_to_rvalue): New function.
3531 (ffi_java_translate_args): Call it.
3532 * src/ffitest.c (closure_test_fn): Interpret return value
3533 as ffi_arg, not int.
3534 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3535 FFI_TYPE_POINTER case.
3536 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3537 values extended to word size.
3539 2002-10-02 Andreas Jaeger <aj@suse.de>
3541 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3543 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
3545 * include/ffi.h.in: Fix i386 win32 compilation.
3547 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3549 * configure.in: Add s390x-*-linux-* target.
3550 * configure: Regenerate.
3551 * include/ffi.h.in: Define S390X for s390x targets.
3552 (FFI_CLOSURES): Define for s390/s390x.
3553 (FFI_TRAMPOLINE_SIZE): Likewise.
3554 (FFI_NATIVE_RAW_API): Likewise.
3555 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3556 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
3557 * src/s390/ffi.c: Major rework of existing code. Add support for
3558 s390x targets. Add closure support.
3559 * src/s390/sysv.S: Likewise.
3561 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
3563 * src/arm/sysv.S: Fix typo.
3565 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
3567 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3568 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3569 (ffi_call_SYSV): Handle soft-float.
3571 2002-09-27 Bo Thorsen <bo@suse.de>
3573 * include/ffi.h.in: Fix multilib x86-64 support.
3575 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3577 * Makefile.am (all-multi): Fix multilib parallel build.
3579 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3581 * configure.in (sh[34]*-*-linux*): Add brackets.
3582 * configure: Regenerate.
3584 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3586 * Makefile.am: Add SH support.
3587 * Makefile.in: Regenerate.
3588 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3589 * configure: Regenerate.
3590 * include/ffi.h.in: Add SH support.
3591 * src/sh/ffi.c: New file.
3592 * src/sh/sysv.S: New file.
3593 * src/types.c: Add SH support.
3595 2002-07-16 Bo Thorsen <bo@suse.de>
3597 * src/x86/ffi64.c: New file that adds x86-64 support.
3598 * src/x86/unix64.S: New file that handles argument setup for
3599 x86-64.
3600 * src/x86/sysv.S: Don't use this on x86-64.
3601 * src/x86/ffi.c: Don't use this on x86-64.
3602 Remove unused vars.
3603 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3604 for x86-64.
3605 * src/ffitest.c (struct6): New test that tests a special case in
3606 the x86-64 ABI.
3607 (struct7): Likewise.
3608 (struct8): Likewise.
3609 (struct9): Likewise.
3610 (closure_test_fn): Silence warning about this when it's not used.
3611 (main): Add the new tests.
3612 (main): Fix a couple of wrong casts and silence some compiler warnings.
3613 * include/ffi.h.in: Add x86-64 ABI definition.
3614 * fficonfig.h.in: Regenerate.
3615 * Makefile.am: Add x86-64 support.
3616 * configure.in: Likewise.
3617 * Makefile.in: Regenerate.
3618 * configure: Likewise.
3620 2002-06-24 Bo Thorsen <bo@suse.de>
3622 * src/types.c: Merge settings for similar architectures.
3623 Add x86-64 sizes and alignments.
3625 2002-06-23 Bo Thorsen <bo@suse.de>
3627 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3628 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3629 * src/mips/ffi.c (ffi_prep_args): Likewise.
3630 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3632 2002-07-18 H.J. Lu (hjl@gnu.org)
3634 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3635 (libffi_la_SOURCES): Support MIPS_LINUX.
3636 (libffi_convenience_la_SOURCES): Likewise.
3637 * Makefile.in: Regenerated.
3639 * configure.in (mips64*-*): Skip.
3640 (mips*-*-linux*): New.
3641 * configure: Regenerated.
3643 * src/mips/ffi.c: Include <sgidefs.h>.
3645 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3647 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3649 2002-05-27 Roger Sayle <roger@eyesopen.com>
3651 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3653 2002-05-27 Bo Thorsen <bo@suse.de>
3655 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3656 fix formatting.
3658 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3660 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3661 beginning of function (for older apple cc).
3663 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3665 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3666 script entry, and set LD to it when configuring multilibs.
3667 * configure: Rebuilt.
3669 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3671 * configure.in (sparc64-*-netbsd*): Add target.
3672 (sparc-*-netbsdelf*): Likewise.
3673 * configure: Regenerate.
3675 2002-04-28 David S. Miller <davem@redhat.com>
3677 * configure.in, configure: Fix SPARC test in previous change.
3679 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3681 * Makefile.am: Add Linux for S/390 support.
3682 * Makefile.in: Regenerate.
3683 * configure.in: Add Linux for S/390 support.
3684 * configure: Regenerate.
3685 * include/ffi.h.in: Add Linux for S/390 support.
3686 * src/s390/ffi.c: New file from libffi CVS tree.
3687 * src/s390/sysv.S: New file from libffi CVS tree.
3689 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3691 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3692 %r_disp32().
3693 * src/sparc/v8.S: Use it.
3694 * src/sparc/v9.S: Likewise.
3695 * fficonfig.h.in: Rebuilt.
3696 * configure: Rebuilt.
3698 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3700 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3701 correctly.
3702 * src/ia64/unix.S: Add unwind information. Fix comments.
3703 Save sp in a way that's compatible with unwind info.
3704 (ffi_call_unix): Correctly restore sp in all cases.
3705 * src/ia64/ffi.c: Add, fix comments.
3707 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3709 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3711 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3713 * configure.in (alpha*-*-netbsd*): Add target.
3714 * configure: Regenerate.
3716 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3718 * src/sparc/v8.S: Add unwind info.
3719 * src/sparc/v9.S: Likewise.
3721 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3723 * configure.in: Enable i*86-*-netbsdelf*.
3724 * configure: Rebuilt.
3726 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3728 PR other/2620
3729 * src/mips/n32.s: Delete
3730 * src/mips/o32.s: Delete
3732 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3734 * configure.in: Enable alpha*-*-freebsd*.
3735 * configure: Rebuilt.
3737 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3739 * Makefile.am: libfficonvenience -> libffi_convenience.
3740 * Makefile.in: Rebuilt.
3742 * Makefile.am: Define ffitest_OBJECTS.
3743 * Makefile.in: Rebuilt.
3745 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3746 David Edelsohn <edelsohn@gnu.org>
3748 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3749 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3750 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3751 * Makefile.in: Regenerate.
3752 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3753 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3754 (flush_icache, flush_range): New functions.
3755 (ffi_closure_helper_DARWIN): New function.
3756 * src/powerpc/aix_closure.S: New file.
3757 * src/powerpc/darwin_closure.S: New file.
3759 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3761 * include/ffi.h.in: Add typedef for ffi_arg.
3762 * src/ffitest.c (main): Declare rint with ffi_arg.
3764 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3766 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3767 number of GPRs for floating-point arguments.
3769 2002-01-31 Anthony Green <green@redhat.com>
3771 * configure: Rebuilt.
3772 * configure.in: Replace CHECK_SIZEOF and endian tests with
3773 cross-compiler friendly macros.
3774 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3775 macros.
3777 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3779 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3780 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3782 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3784 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3785 (TARGET_SRC_POWERPC_AIX): New.
3786 (POWERPC_AIX): New stanza.
3787 * Makefile.in: Regenerate.
3788 * configure.in: Add AIX case.
3789 * configure: Regenerate.
3790 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3791 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3792 size. Fix "long double" support.
3793 (ffi_call): Add FFI_AIX case.
3794 * src/powerpc/aix.S: New.
3796 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3798 Implement Darwin PowerPC ABI.
3799 * configure.in: Handle powerpc-*-darwin*.
3800 * Makefile.am: Set source files for POWERPC_DARWIN.
3801 * configure: Rebuilt.
3802 * Makefile.in: Rebuilt.
3803 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3804 POWERPC_DARWIN.
3805 * src/powerpc/darwin.S: New file.
3806 * src/powerpc/ffi_darwin.c: New file.
3808 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3810 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3812 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3814 * src/x86/sysv.S: Avoid gas-only .balign directive.
3815 Use C style comments.
3817 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3819 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3820 Fixes PR bootstrap/3563.
3822 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3824 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3826 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3828 * configure.in: Recognize sparc*-sun-* host.
3829 * configure: Regenerate.
3831 2001-06-06 Andrew Haley <aph@redhat.com>
3833 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
3835 2001-06-03 Andrew Haley <aph@redhat.com>
3837 * src/alpha/osf.S: Add unwind info.
3838 * src/powerpc/sysv.S: Add unwind info.
3839 * src/powerpc/ppc_closure.S: Likewise.
3841 2000-05-31 Jeff Sturm <jsturm@one-point.com>
3843 * configure.in: Fix AC_ARG_ENABLE usage.
3844 * configure: Rebuilt.
3846 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3848 * configure.in: Remove warning about beta code.
3849 * configure: Rebuilt.
3851 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3853 * src/ia64/unix.S: Restore stack pointer when returning from
3854 ffi_closure_UNIX.
3855 * src/ia64/ffi.c: Fix typo in comment.
3857 2001-04-18 Jim Wilson <wilson@redhat.com>
3859 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3860 to eliminate RAW DV.
3862 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3864 * Makefile.am: Make a libtool convenience library.
3865 * Makefile.in: Rebuilt.
3867 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3869 * configure.in: Use different syntax for subdirectory creation.
3870 * configure: Rebuilt.
3872 2001-03-27 Jon Beniston <jon@beniston.com>
3874 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3875 * configure: Rebuilt.
3876 * Makefile.am: Added X86_WIN32 target support.
3877 * Makefile.in: Rebuilt.
3879 * include/ffi.h.in: Added X86_WIN32 target support.
3881 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3882 * src/types.c: Added X86_WIN32 target support.
3884 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3885 stuff removed and made to work with CygWin's gas.
3887 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3889 * configure.in: Make target subdirectory in build dir.
3890 * Makefile.am: Override suffix based rules to specify correct output
3891 subdirectory.
3892 * Makefile.in: Rebuilt.
3893 * configure: Rebuilt.
3895 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3897 * src/powerpc/ppc_closure.S: New file.
3898 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3899 involving long long and register pairs.
3900 (ffi_prep_closure): New function.
3901 (flush_icache): Likewise.
3902 (ffi_closure_helper_SYSV): Likewise.
3903 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3904 (FFI_TRAMPOLINE_SIZE): Likewise.
3905 (FFI_NATIVE_RAW_API): Likewise.
3906 * Makefile.in: Rebuilt.
3907 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3908 (TARGET_SRC_POWERPC): Likewise.
3910 2001-03-19 Tom Tromey <tromey@redhat.com>
3912 * Makefile.in: Rebuilt.
3913 * Makefile.am (ffitest_LDFLAGS): New macro.
3915 2001-03-02 Nick Clifton <nickc@redhat.com>
3917 * include/ffi.h.in: Remove RCS ident string.
3918 * include/ffi_mips.h: Remove RCS ident string.
3919 * src/debug.c: Remove RCS ident string.
3920 * src/ffitest.c: Remove RCS ident string.
3921 * src/prep_cif.c: Remove RCS ident string.
3922 * src/types.c: Remove RCS ident string.
3923 * src/alpha/ffi.c: Remove RCS ident string.
3924 * src/alpha/osf.S: Remove RCS ident string.
3925 * src/arm/ffi.c: Remove RCS ident string.
3926 * src/arm/sysv.S: Remove RCS ident string.
3927 * src/mips/ffi.c: Remove RCS ident string.
3928 * src/mips/n32.S: Remove RCS ident string.
3929 * src/mips/o32.S: Remove RCS ident string.
3930 * src/sparc/ffi.c: Remove RCS ident string.
3931 * src/sparc/v8.S: Remove RCS ident string.
3932 * src/sparc/v9.S: Remove RCS ident string.
3933 * src/x86/ffi.c: Remove RCS ident string.
3934 * src/x86/sysv.S: Remove RCS ident string.
3936 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3938 * include/ffi.h.in: Change sourceware.cygnus.com references to
3939 gcc.gnu.org.
3941 2000-12-09 Richard Henderson <rth@redhat.com>
3943 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3944 (ffi_closure_osf_inner): Index rather than increment avalue
3945 and arg_types. Give ffi_closure_osf the raw return value type.
3946 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3947 promotion.
3949 2000-12-07 Richard Henderson <rth@redhat.com>
3951 * src/raw_api.c (ffi_translate_args): Fix typo.
3952 (ffi_prep_closure): Likewise.
3954 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3955 FFI_TRAMPOLINE_SIZE.
3956 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3957 cif->bytes for new ffi_call_osf implementation.
3958 (ffi_prep_args): Absorb into ...
3959 (ffi_call): ... here. Do all stack allocation here and
3960 avoid a callback function.
3961 (ffi_prep_closure, ffi_closure_osf_inner): New.
3962 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3963 (ffi_closure_osf): New.
3965 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3967 * config.guess, config.sub, install-sh: Removed.
3968 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3969 * Makefile.in: Rebuilt.
3971 * acinclude.m4: Include libtool macros from the top level.
3972 * aclocal.m4, configure: Rebuilt.
3974 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3976 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3977 * configure: Rebuilt.
3979 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3981 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3982 memory correctly. Use conditional instructions, not branches where
3983 possible.
3985 2000-05-04 Tom Tromey <tromey@cygnus.com>
3987 * configure: Rebuilt.
3988 * configure.in: Match `arm*-*-linux-*'.
3989 From Chris Dornan <cdornan@arm.com>.
3991 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3993 * Makefile.am (SUBDIRS): Define.
3994 (AM_MAKEFLAGS): Likewise.
3995 (Multilib support.): Add section.
3996 * Makefile.in: Rebuilt.
3997 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3998 New variables. Set for gcc using -print-multi-lib. Export them
3999 to libtool.
4000 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
4001 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
4002 for -shared links.
4003 (extra_compiler_flags_value, extra_compiler_flags): Check these
4004 for extra compiler options which need to be passed down in
4005 compiler_flags.
4007 2000-04-16 Anthony Green <green@redhat.com>
4009 * configure: Rebuilt.
4010 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
4012 2000-04-14 Jakub Jelinek <jakub@redhat.com>
4014 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
4015 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
4016 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
4017 Replace all void * sizeofs with sizeof(int).
4018 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
4019 different than DOUBLE.
4020 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
4021 (ffi_prep_args_v9): New function.
4022 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
4023 (ffi_V9_return_struct): New function.
4024 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
4025 32bit code (not yet cross-arch calls).
4026 * src/sparc/v8.S: Add struct return delay nop.
4027 Handle long long.
4028 * src/sparc/v9.S: New file.
4029 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
4030 is used on sparc64 only for structures larger than 32 bytes.
4031 Pass by reference for structures is done for structure arguments
4032 larger than 16 bytes.
4033 * src/ffitest.c (main): Use 64bit rint on sparc64.
4034 Run long long tests on sparc.
4035 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
4036 sparc64.
4037 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
4038 on sparc64.
4039 * configure.in (sparc-*-linux*): New supported target.
4040 (sparc64-*-linux*): Likewise.
4041 * configure: Rebuilt.
4042 * Makefile.am: Add v9.S to SPARC files.
4043 * Makefile.in: Likewise.
4044 (LINK): Surround $(CCLD) into double quotes, so that multilib
4045 compiles work correctly.
4047 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
4049 * configure: Rebuilt.
4050 * configure.in: (i*86-*-solaris*): New libffi target. Patch
4051 proposed by Bryce McKinlay.
4053 2000-03-20 Tom Tromey <tromey@cygnus.com>
4055 * Makefile.in: Hand edit for java_raw_api.lo.
4057 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
4059 * config.guess, config.sub: Update from the gcc tree.
4060 Fix for PR libgcj/168.
4062 2000-03-03 Tom Tromey <tromey@cygnus.com>
4064 * Makefile.in: Fixed ia64 by hand.
4066 * configure: Rebuilt.
4067 * configure.in (--enable-multilib): New option.
4068 (libffi_basedir): New subst.
4069 (AC_OUTPUT): Added multilib code.
4071 2000-03-02 Tom Tromey <tromey@cygnus.com>
4073 * Makefile.in: Rebuilt.
4074 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
4075 directory name.
4077 2000-02-25 Hans Boehm <boehm@acm.org>
4079 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
4080 files.
4081 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
4082 list.
4083 (ffi_prep_raw_closure): Use ffi_translate_args, not
4084 ffi_closure_translate.
4085 * src/java_raw_api.c: New file.
4086 * src/ffitest.c (closure_test_fn): New function.
4087 (main): Define `rint' as long long on IA64. Added new test when
4088 FFI_CLOSURES is defined.
4089 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
4090 (ffi_abi): Recognize IA64.
4091 (ffi_raw): Added `flt' field.
4092 Added "Java raw API" code.
4093 * configure.in: Recognize ia64.
4094 * Makefile.am (TARGET_SRC_IA64): New macro.
4095 (libffi_la_common_SOURCES): Added java_raw_api.c.
4096 (libffi_la_SOURCES): Define in IA64 case.
4098 2000-01-04 Tom Tromey <tromey@cygnus.com>
4100 * Makefile.in: Rebuilt with newer automake.
4102 1999-12-31 Tom Tromey <tromey@cygnus.com>
4104 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
4106 1999-09-01 Tom Tromey <tromey@cygnus.com>
4108 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
4109 undefs.
4110 * fficonfig.h.in: Rebuilt.
4111 * configure: Rebuilt.
4112 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
4113 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
4114 * acconfig.h: Don't #undef PACKAGE or VERSION.
4116 1999-08-09 Anthony Green <green@cygnus.com>
4118 * include/ffi.h.in: Try to work around messy header problem
4119 with PACKAGE and VERSION.
4121 * configure: Rebuilt.
4122 * configure.in: Change version to 2.00-beta.
4124 * fficonfig.h.in: Rebuilt.
4125 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
4127 * src/x86/ffi.c (ffi_raw_call): Rename.
4129 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
4131 * src/x86/ffi.c (ffi_closure_SYSV): New function.
4132 (ffi_prep_incoming_args_SYSV): Ditto.
4133 (ffi_prep_closure): Ditto.
4134 (ffi_closure_raw_SYSV): Ditto.
4135 (ffi_prep_raw_closure): More ditto.
4136 (ffi_call_raw): Final ditto.
4138 * include/ffi.h.in: Add definitions for closure and raw API.
4140 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4141 FFI_TYPE_UINT64.
4143 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
4145 * src/raw_api.c: New file.
4147 * include/ffi.h.in (ffi_raw): New type.
4148 (UINT_ARG, SINT_ARG): New defines.
4149 (ffi_closure, ffi_raw_closure): New types.
4150 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4152 * configure.in: Add check for endianness and sizeof void*.
4154 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
4155 instead of directly.
4157 * configure: Rebuilt.
4159 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
4161 * configure.in: Add x86 and powerpc BeOS configurations.
4162 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
4164 1999-05-09 Anthony Green <green@cygnus.com>
4166 * configure.in: Add warning about this being beta code.
4167 Remove src/Makefile.am from the picture.
4168 * configure: Rebuilt.
4170 * Makefile.am: Move logic from src/Makefile.am. Add changes
4171 to support libffi as a target library.
4172 * Makefile.in: Rebuilt.
4174 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
4175 Upgraded to new autoconf, automake, libtool.
4177 * README: Tweaks.
4179 * LICENSE: Update copyright date.
4181 * src/Makefile.am, src/Makefile.in: Removed.
4183 1998-11-29 Anthony Green <green@cygnus.com>
4185 * include/ChangeLog: Removed.
4186 * src/ChangeLog: Removed.
4187 * src/mips/ChangeLog: Removed.
4188 * src/sparc/ChangeLog: Remboved.
4189 * src/x86/ChangeLog: Removed.
4191 * ChangeLog.v1: Created.