1 2010-02-24 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
3 * doc/libffi.texi (The Closure API): Fix typo.
4 * doc/libffi.info: Remove.
6 2010-02-15 Matthias Klose <doko@ubuntu.com>
8 * src/arm/sysv.S (__ARM_ARCH__): Define for processor
11 2010-01-07 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
14 * testsuite/libffi.call/ffitest.h [__alpha__ && __osf__] (PRIdLL,
15 PRIuLL, PRId64, PRIu64, PRIuPTR): Define.
16 * testsuite/libffi.call/cls_align_sint64.c: Add -Wno-format on
18 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
19 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
20 * testsuite/libffi.call/return_ll1.c: Likewise.
21 * testsuite/libffi.call/stret_medium2.c: Likewise.
22 * testsuite/libffi.special/ffitestcxx.h (allocate_mmap): Cast
25 2010-01-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
27 * src/mips/n32.S: Use .abicalls and .eh_frame with __GNUC__.
29 2009-12-31 Anthony Green <green@redhat.com>
31 * README: Update for libffi 3.0.9.
33 2009-12-27 Matthias Klose <doko@ubuntu.com>
35 * configure.ac (HAVE_LONG_DOUBLE): Define for mips when
39 2009-12-26 Anthony Green <green@redhat.com>
41 * testsuite/libffi.call/cls_longdouble_va.c: Mark as xfail for
43 * testsuite/libffi.call/cls_double_va.c: Ditto.
45 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
47 * testsuite/libffi.call/ffitest.h: Conditionally include stdint.h
49 * testsuite/libffi.special/unwindtest.cc: Ditto.
51 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
53 * configure.ac: Add amd64-*-openbsd*.
55 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Link
56 openbsd programs with -lpthread.
58 2009-12-26 Anthony Green <green@redhat.com>
60 * testsuite/libffi.call/cls_double_va.c,
61 testsuite/libffi.call/cls_longdouble.c,
62 testsuite/libffi.call/cls_longdouble_va.c,
63 testsuite/libffi.call/cls_pointer.c,
64 testsuite/libffi.call/cls_pointer_stack.c: Remove xfail for
65 mips*-*-* and arm*-*-*.
66 * testsuite/libffi.call/cls_align_longdouble_split.c,
67 testsuite/libffi.call/cls_align_longdouble_split2.c,
68 testsuite/libffi.call/stret_medium2.c,
69 testsuite/libffi.call/stret_medium.c,
70 testsuite/libffi.call/stret_large.c,
71 testsuite/libffi.call/stret_large2.c: Remove xfail for arm*-*-*.
73 2009-12-31 Kay Tietz <ktietz70@googlemail.com>
75 * testsuite/libffi.call/ffitest.h,
76 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRuLL): Fix
79 2009-12-31 Carlo Bramini <carlo.bramix@libero.it>
81 * configure.ac (AM_LTLDFLAGS): Define for windows hosts.
82 * Makefile.am (libffi_la_LDFLAGS): Add AM_LTLDFLAGS.
84 * Makefile.in: Rebuilt.
86 2009-12-31 Anthony Green <green@redhat.com>
89 * testsuite/libffi.call/huge_struct.c: New test case from Blake
92 2009-12-28 David Edelsohn <edelsohn@gnu.org>
94 * src/powerpc/ffi_darwin.c (ffi_prep_args): Copy abi and nargs to
96 (aix_adjust_aggregate_sizes): New function.
97 (ffi_prep_cif_machdep): Call it.
99 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
101 * configure.ac: Define FFI_MMAP_EXEC_WRIT for the given targets.
102 * configure: Regenerate.
103 * fficonfig.h.in: Likewise.
104 * src/closures.c: Remove the FFI_MMAP_EXEC_WRIT definition for
107 2009-12-26 Andreas Schwab <schwab@linux-m68k.org>
109 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Advance intarg_count
110 when a float arguments is passed in memory.
111 (ffi_closure_helper_SYSV): Mark general registers as used up when
112 a 64bit or soft-float long double argument is passed in memory.
114 2009-12-25 Matthias Klose <doko@ubuntu.com>
116 * man/ffi_call.3: Fix #include in examples.
117 * doc/libffi.texi: Add dircategory.
119 2009-12-25 Frank Everdij <f.p.x.everdij@tudelft.nl>
121 * include/ffi.h.in: Placed '__GNUC__' ifdef around
122 '__attribute__((aligned(8)))' in ffi_closure, fixes compile for
124 * include/ffi_common.h: Added '__sgi' define to non
125 '__attribute__((__mode__()))' integer typedefs.
126 * src/mips/ffi.c (ffi_call, ffi_closure_mips_inner_O32,
127 ffi_closure_mips_inner_N32): Added 'defined(_MIPSEB)' to BE check.
128 (ffi_closure_mips_inner_O32, ffi_closure_mips_inner_N32): Added
129 FFI_LONGDOUBLE support and alignment(N32 only).
130 * src/mips/ffitarget.h: Corrected '#include <sgidefs.h>' for IRIX and
131 fixed non '__attribute__((__mode__()))' integer typedefs.
132 * src/mips/n32.S: Put '#ifdef linux' around '.abicalls' and '.eh_frame'
133 since they are Linux/GNU Assembler specific.
135 2009-12-25 Bradley Smith <brad@brad-smith.co.uk>
137 * configure.ac, Makefile.am, src/avr32/ffi.c,
138 src/avr32/ffitarget.h,
139 src/avr32/sysv.S: Add AVR32 port.
140 * configure, Makefile.in: Rebuilt.
142 2009-12-21 Andreas Tobler <a.tobler@schweiz.org>
144 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
145 * configure: Regenerate.
147 2009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
149 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
151 2009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
153 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
156 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
158 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
161 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
163 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
165 2009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
168 * src/closures.c [X86_64 && __sun__ && __svr4__]
169 (FFI_MMAP_EXEC_WRIT): Define.
171 2009-12-08 David Daney <ddaney@caviumnetworks.com>
173 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
174 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
175 * testsuite/libffi.call/stret_large.c: Same.
176 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
177 * testsuite/libffi.call/stret_large2.c: Same.
178 * testsuite/libffi.call/stret_medium2.c: Same.
180 2009-12-07 David Edelsohn <edelsohn@gnu.org>
182 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
185 2009-12-05 David Edelsohn <edelsohn@gnu.org>
187 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
189 * src/powerpc/aix_closure.S: Same.
191 2009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
193 * Makefile.in: Regenerate.
194 * configure: Regenerate.
195 * include/Makefile.in: Regenerate.
196 * man/Makefile.in: Regenerate.
197 * testsuite/Makefile.in: Regenerate.
199 2009-12-04 David Edelsohn <edelsohn@gnu.org>
201 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
204 2009-12-04 Uros Bizjak <ubizjak@gmail.com>
207 * src/x86/ffi64.c (classify_argument): Update from
208 gcc/config/i386/i386.c.
209 (ffi_closure_unix64_inner): Do not use the address of two consecutive
210 SSE registers directly.
211 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
212 for x86_64 linux targets.
214 2009-12-04 David Edelsohn <edelsohn@gnu.org>
216 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
217 pfr for long double split between fpr13 and stack.
219 2009-12-03 David Edelsohn <edelsohn@gnu.org>
221 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
222 fparg_count twice for long double.
224 2009-12-03 David Edelsohn <edelsohn@gnu.org>
227 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
229 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
231 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
232 Remove xfails for x86 linux targets.
234 2009-12-02 David Edelsohn <edelsohn@gnu.org>
236 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
239 2009-12-01 David Edelsohn <edelsohn@gnu.org>
241 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
242 register usage. Call ffi_prep_args directly. Add long double
243 return value support.
244 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
245 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
246 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
247 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
249 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
251 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
252 Allocate result area between params and FPRs.
254 2009-11-30 David Edelsohn <edelsohn@gnu.org>
257 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
259 * src/powerpc/aix.S: Implement AIX64 version.
260 * src/powerpc/aix_closure.S: Implement AIX64 version.
261 (ffi_closure_ASM): Use extsb, lha and displament addresses.
262 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
264 (ffi_prep_cif_machdep): Same.
266 (ffi_closure_helper_DARWIN): Same.
268 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
271 * testsuite/libffi.call/testclosure.c: New test.
273 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
275 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
276 assembly version use of ___chkstk.
278 2009-09-23 Matthias Klose <doko@ubuntu.com>
280 PR libffi/40242, PR libffi/41443
281 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
282 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
283 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
284 Change the conditionals to __SOFTFP__ || __ARM_EABI__
285 for -mfloat-abi=softfp to work.
287 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
289 PR testsuite/32843 (strikes again)
290 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
291 enable proper extension on char and short.
293 2009-09-15 David Daney <ddaney@caviumnetworks.com>
295 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
296 handling for FFI_TYPE_POINTER.
297 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
298 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
299 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
300 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
301 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
302 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
303 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
304 structure and pointer returns.
305 (ffi_closure_N32): Add handling for pointer returns.
306 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
307 calc_n32_return_struct_flags): Handle soft-float.
308 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
309 (ffi_call_N32): Declare proper argument types.
310 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
313 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
315 * configure.ac (AC_PREREQ): Bump to 2.64.
317 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
319 * Makefile.am (install-html, install-pdf): Remove.
320 * Makefile.in: Regenerate.
322 * Makefile.in: Regenerate.
323 * aclocal.m4: Regenerate.
324 * configure: Regenerate.
325 * fficonfig.h.in: Regenerate.
326 * include/Makefile.in: Regenerate.
327 * man/Makefile.in: Regenerate.
328 * testsuite/Makefile.in: Regenerate.
330 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
332 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
334 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
337 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
338 return types for X86_WIN32.
339 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
340 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
341 _ffi_closure_STDCALL): Likewise.
343 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
344 (dlmmap, dlmunmap): Also use these functions on Cygwin.
346 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
351 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
354 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
356 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
357 to "" before adding paths. (This reinstates an assignment that
358 was removed by my 2009-06-30 commit, but changes the initial
359 value from "." to "".)
361 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
364 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
365 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
368 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
370 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
371 to ld_library_path. Use add_path. Add just find_libgcc_s
372 to ld_library_path, not every libgcc multilib directory.
374 2009-06-16 Wim Lewis <wiml@hhhh.org>
376 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
377 supposed to be callee-saved.
378 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
379 return buffer for odd-size structs.
381 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
384 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
385 allow_stack_execute for Darwin.
387 2009-06-16 Andrew Haley <aph@redhat.com>
389 * configure.ac (TARGETDIR): Add missing blank lines.
390 * configure: Regenerate.
392 2009-06-16 Andrew Haley <aph@redhat.com>
394 * testsuite/libffi.call/cls_align_sint64.c,
395 testsuite/libffi.call/cls_align_uint64.c,
396 testsuite/libffi.call/cls_longdouble_va.c,
397 testsuite/libffi.call/cls_ulonglong.c,
398 testsuite/libffi.call/return_ll1.c,
399 testsuite/libffi.call/stret_medium2.c: Fix printf format
401 * testsuite/libffi.call/ffitest.h,
402 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
404 2009-06-15 Andrew Haley <aph@redhat.com>
406 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
407 * testsuite/libffi.call/err_bad_abi.c: Likewise.
409 2009-06-12 Andrew Haley <aph@redhat.com>
411 * Makefile.am: Remove info_TEXINFOS.
413 2009-06-12 Andrew Haley <aph@redhat.com>
415 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
416 testsuite/libffi.call/cls_align_uint64.c,
417 testsuite/libffi.call/cls_ulonglong.c,
418 testsuite/libffi.call/return_ll1.c,
419 testsuite/libffi.call/stret_medium2.c: Fix printf format
421 testsuite/libffi.special/unwindtest.cc: include stdint.h.
423 2009-06-11 Timothy Wall <twall@users.sf.net>
428 include/ffi_common.h,
433 src/x86/win64.S (new),
434 README: Added win64 support (mingw or MSVC)
438 testsuite/Makefile.in,
440 aclocal.m4: Regenerated
441 * ltcf-c.sh: properly escape cygwin/w32 path
442 * man/ffi_call.3: Clarify size requirements for return value.
443 * src/x86/ffi64.c: Fix filename in comment.
444 * src/x86/win32.S: Remove unused extern.
446 * testsuite/libffi.call/closure_fn0.c,
447 testsuite/libffi.call/closure_fn1.c,
448 testsuite/libffi.call/closure_fn2.c,
449 testsuite/libffi.call/closure_fn3.c,
450 testsuite/libffi.call/closure_fn4.c,
451 testsuite/libffi.call/closure_fn5.c,
452 testsuite/libffi.call/closure_fn6.c,
453 testsuite/libffi.call/closure_stdcall.c,
454 testsuite/libffi.call/cls_12byte.c,
455 testsuite/libffi.call/cls_16byte.c,
456 testsuite/libffi.call/cls_18byte.c,
457 testsuite/libffi.call/cls_19byte.c,
458 testsuite/libffi.call/cls_1_1byte.c,
459 testsuite/libffi.call/cls_20byte.c,
460 testsuite/libffi.call/cls_20byte1.c,
461 testsuite/libffi.call/cls_24byte.c,
462 testsuite/libffi.call/cls_2byte.c,
463 testsuite/libffi.call/cls_3_1byte.c,
464 testsuite/libffi.call/cls_3byte1.c,
465 testsuite/libffi.call/cls_3byte2.c,
466 testsuite/libffi.call/cls_4_1byte.c,
467 testsuite/libffi.call/cls_4byte.c,
468 testsuite/libffi.call/cls_5_1_byte.c,
469 testsuite/libffi.call/cls_5byte.c,
470 testsuite/libffi.call/cls_64byte.c,
471 testsuite/libffi.call/cls_6_1_byte.c,
472 testsuite/libffi.call/cls_6byte.c,
473 testsuite/libffi.call/cls_7_1_byte.c,
474 testsuite/libffi.call/cls_7byte.c,
475 testsuite/libffi.call/cls_8byte.c,
476 testsuite/libffi.call/cls_9byte1.c,
477 testsuite/libffi.call/cls_9byte2.c,
478 testsuite/libffi.call/cls_align_double.c,
479 testsuite/libffi.call/cls_align_float.c,
480 testsuite/libffi.call/cls_align_longdouble.c,
481 testsuite/libffi.call/cls_align_longdouble_split.c,
482 testsuite/libffi.call/cls_align_longdouble_split2.c,
483 testsuite/libffi.call/cls_align_pointer.c,
484 testsuite/libffi.call/cls_align_sint16.c,
485 testsuite/libffi.call/cls_align_sint32.c,
486 testsuite/libffi.call/cls_align_sint64.c,
487 testsuite/libffi.call/cls_align_uint16.c,
488 testsuite/libffi.call/cls_align_uint32.c,
489 testsuite/libffi.call/cls_align_uint64.c,
490 testsuite/libffi.call/cls_dbls_struct.c,
491 testsuite/libffi.call/cls_double.c,
492 testsuite/libffi.call/cls_double_va.c,
493 testsuite/libffi.call/cls_float.c,
494 testsuite/libffi.call/cls_longdouble.c,
495 testsuite/libffi.call/cls_longdouble_va.c,
496 testsuite/libffi.call/cls_multi_schar.c,
497 testsuite/libffi.call/cls_multi_sshort.c,
498 testsuite/libffi.call/cls_multi_sshortchar.c,
499 testsuite/libffi.call/cls_multi_uchar.c,
500 testsuite/libffi.call/cls_multi_ushort.c,
501 testsuite/libffi.call/cls_multi_ushortchar.c,
502 testsuite/libffi.call/cls_pointer.c,
503 testsuite/libffi.call/cls_pointer_stack.c,
504 testsuite/libffi.call/cls_schar.c,
505 testsuite/libffi.call/cls_sint.c,
506 testsuite/libffi.call/cls_sshort.c,
507 testsuite/libffi.call/cls_uchar.c,
508 testsuite/libffi.call/cls_uint.c,
509 testsuite/libffi.call/cls_ulonglong.c,
510 testsuite/libffi.call/cls_ushort.c,
511 testsuite/libffi.call/err_bad_abi.c,
512 testsuite/libffi.call/err_bad_typedef.c,
513 testsuite/libffi.call/float2.c,
514 testsuite/libffi.call/huge_struct.c,
515 testsuite/libffi.call/nested_struct.c,
516 testsuite/libffi.call/nested_struct1.c,
517 testsuite/libffi.call/nested_struct10.c,
518 testsuite/libffi.call/nested_struct2.c,
519 testsuite/libffi.call/nested_struct3.c,
520 testsuite/libffi.call/nested_struct4.c,
521 testsuite/libffi.call/nested_struct5.c,
522 testsuite/libffi.call/nested_struct6.c,
523 testsuite/libffi.call/nested_struct7.c,
524 testsuite/libffi.call/nested_struct8.c,
525 testsuite/libffi.call/nested_struct9.c,
526 testsuite/libffi.call/problem1.c,
527 testsuite/libffi.call/return_ldl.c,
528 testsuite/libffi.call/return_ll1.c,
529 testsuite/libffi.call/stret_large.c,
530 testsuite/libffi.call/stret_large2.c,
531 testsuite/libffi.call/stret_medium.c,
532 testsuite/libffi.call/stret_medium2.c,
533 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
534 of checking for MMAP. Use intptr_t instead of long casts.
536 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
538 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
539 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
540 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
542 2009-06-09 Andrew Haley <aph@redhat.com>
544 * src/x86/freebsd.S: Add missing file.
546 2009-06-08 Andrew Haley <aph@redhat.com>
548 Import from libffi 3.0.8:
550 * doc/libffi.texi: New file.
551 * doc/libffi.info: Likewise.
552 * doc/stamp-vti: Likewise.
553 * man/Makefile.am: New file.
554 * man/ffi_call.3: New file.
556 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
558 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
560 * configure.ac: Bump version to 3.0.8.
561 parisc*-*-linux*: Add.
562 i386-*-freebsd* | i386-*-openbsd*: Add.
563 powerpc-*-beos*: Add.
564 AM_CONDITIONAL X86_FREEBSD: Add.
565 AC_CONFIG_FILES: Add man/Makefile.
567 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
569 2009-06-08 Andrew Haley <aph@redhat.com>
571 * README: Import from libffi 3.0.8.
573 2009-06-08 Andrew Haley <aph@redhat.com>
575 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
576 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
577 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
578 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
580 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
581 * testsuite/libffi.call/stret_medium.c: Likewise.
582 * testsuite/libffi.call/stret_large2.c: Likewise.
583 * testsuite/libffi.call/stret_large.c: Likewise.
585 2008-12-26 Timothy Wall <twall@users.sf.net>
587 * testsuite/libffi.call/cls_longdouble.c,
588 testsuite/libffi.call/cls_longdouble_va.c,
589 testsuite/libffi.call/cls_align_longdouble.c,
590 testsuite/libffi.call/cls_align_longdouble_split.c,
591 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
592 failures on x86_64 cygwin/mingw.
594 2008-12-22 Timothy Wall <twall@users.sf.net>
596 * testsuite/libffi.call/closure_fn0.c,
597 testsuite/libffi.call/closure_fn1.c,
598 testsuite/libffi.call/closure_fn2.c,
599 testsuite/libffi.call/closure_fn3.c,
600 testsuite/libffi.call/closure_fn4.c,
601 testsuite/libffi.call/closure_fn5.c,
602 testsuite/libffi.call/closure_fn6.c,
603 testsuite/libffi.call/closure_loc_fn0.c,
604 testsuite/libffi.call/closure_stdcall.c,
605 testsuite/libffi.call/cls_align_pointer.c,
606 testsuite/libffi.call/cls_pointer.c,
607 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
608 pointer to integer (intptr_t).
609 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
611 2008-07-24 Anthony Green <green@redhat.com>
613 * testsuite/libffi.call/cls_dbls_struct.c,
614 testsuite/libffi.call/cls_double_va.c,
615 testsuite/libffi.call/cls_longdouble.c,
616 testsuite/libffi.call/cls_longdouble_va.c,
617 testsuite/libffi.call/cls_pointer.c,
618 testsuite/libffi.call/cls_pointer_stack.c,
619 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
622 2008-03-04 Anthony Green <green@redhat.com>
626 * testsuite/libffi.call/cls_align_longdouble_split2.c
627 testsuite/libffi.call/cls_align_longdouble_split.c
628 testsuite/libffi.call/cls_dbls_struct.c
629 testsuite/libffi.call/cls_double_va.c
630 testsuite/libffi.call/cls_longdouble.c
631 testsuite/libffi.call/cls_longdouble_va.c
632 testsuite/libffi.call/cls_pointer.c
633 testsuite/libffi.call/cls_pointer_stack.c
634 testsuite/libffi.call/err_bad_abi.c
635 testsuite/libffi.call/err_bad_typedef.c
636 testsuite/libffi.call/stret_large2.c
637 testsuite/libffi.call/stret_large.c
638 testsuite/libffi.call/stret_medium2.c
639 testsuite/libffi.call/stret_medium.c: New tests from Apple.
641 2009-06-05 Andrew Haley <aph@redhat.com>
643 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
646 2009-06-04 Andrew Haley <aph@redhat.com>
648 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
651 2008-02-26 Anthony Green <green@redhat.com>
652 Thomas Heller <theller@ctypes.org>
654 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
657 2008-02-03 Timothy Wall <twall@users.sf.net>
659 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
660 offset based on code pointer, not data pointer.
662 2008-01-31 Timothy Wall <twall@users.sf.net>
664 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
666 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
668 * src/x86/win32.S: Add assembly for stdcall closure.
669 * src/x86/ffi.c: Initialize stdcall closure trampoline.
671 2009-06-04 Andrew Haley <aph@redhat.com>
673 * include/ffi.h.in: Change void (*)() to void (*)(void).
674 * src/x86/ffi.c: Likewise.
676 2009-06-04 Andrew Haley <aph@redhat.com>
678 * src/powerpc/ppc_closure.S: Insert licence header.
679 * src/powerpc/linux64_closure.S: Likewise.
680 * src/m68k/sysv.S: Likewise.
682 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
683 * src/powerpc/ffi.c: Likewise.
684 * src/powerpc/ffi_darwin.c: Likewise.
685 * src/m32r/ffi.c: Likewise.
686 * src/sh64/ffi.c: Likewise.
687 * src/x86/ffi64.c: Likewise.
688 * src/alpha/ffi.c: Likewise.
689 * src/alpha/osf.S: Likewise.
690 * src/frv/ffi.c: Likewise.
691 * src/s390/ffi.c: Likewise.
692 * src/pa/ffi.c: Likewise.
693 * src/pa/hpux32.S: Likewise.
694 * src/ia64/unix.S: Likewise.
695 * src/ia64/ffi.c: Likewise.
696 * src/sparc/ffi.c: Likewise.
697 * src/mips/ffi.c: Likewise.
698 * src/sh/ffi.c: Likewise.
700 2008-02-15 David Daney <ddaney@avtrex.com>
702 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
703 Define (conditionally), and use it to include cachectl.h.
704 (ffi_prep_closure_loc): Fix cache flushing.
705 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
707 2009-06-04 Andrew Haley <aph@redhat.com>
713 src/powerpc/ffitarget.h,
715 src/sh64/ffitarget.h,
726 src/alpha/ffitarget.h,
729 src/m68k/ffitarget.h,
732 src/s390/ffitarget.h,
734 src/cris/ffitarget.h,
739 src/ia64/ffitarget.h,
742 src/ia64/ia64_flags.h,
746 src/sparc/ffitarget.h,
749 src/mips/ffitarget.h,
756 src/sh/sysv.S: Update license text.
758 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
760 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
761 (.eh_frame): Add FDE for it.
763 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
765 * configure.ac: Also check if assembler supports pc-relative
766 relocs on X86_WIN32 targets.
767 * configure: Regenerate.
768 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
769 (_ffi_call_SYSV): Add missing function type symbol .def and
770 add EH markup labels.
771 (_ffi_call_STDCALL): Likewise.
772 (_ffi_closure_SYSV): Likewise.
773 (_ffi_closure_raw_SYSV): Likewise.
774 (.eh_frame): Add hand-crafted EH data.
776 2009-04-09 Jakub Jelinek <jakub@redhat.com>
778 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
779 version 3 of the GNU General Public License and to point readers
780 at the COPYING3 file and the FSF's license web page.
781 * testsuite/libffi.call/call.exp: Likewise.
782 * testsuite/libffi.special/special.exp: Likewise.
784 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
786 * configure: Regenerate.
788 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
791 * configure.ac (HAVE_AS_X86_PCREL): New test.
792 * configure: Regenerate.
793 * fficonfig.h.in: Regenerate.
794 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
795 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
796 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
797 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
798 * src/x86/unix64.S (.Lstore_table): Move to .text section.
799 (.Lload_table): Likewise.
800 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
802 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
804 * configure: Regenerate.
806 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
808 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
809 signed/unsigned int8/16 return values.
810 * src/sparc/v8.S (ffi_call_v8): Likewise.
811 (ffi_closure_v8): Likewise.
813 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
814 Steve Ellcey <sje@cup.hp.com>
816 * configure: Regenerate for new libtool.
817 * Makefile.in: Ditto.
818 * include/Makefile.in: Ditto.
821 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
823 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
824 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
825 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
826 Adjust copyright notice.
827 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
828 register or two register to use for FFI_SYSV structs.
829 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
830 (ffi_closure_helper_SYSV): Fix the return type for
831 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
832 Adjust copyright notice.
834 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
836 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
839 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
841 * configure: Regenerate.
842 * include/Makefile.in: Regenerate.
843 * testsuite/Makefile.in: Regenerate.
845 2008-06-07 Joseph Myers <joseph@codesourcery.com>
847 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
848 powerpc-*-beos*): Remove.
849 * configure: Regenerate.
851 2008-05-09 Julian Brown <julian@codesourcery.com>
853 * Makefile.am (LTLDFLAGS): New.
854 (libffi_la_LDFLAGS): Use above.
855 * Makefile.in: Regenerate.
857 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
860 * aclocal.m4: Regenerate.
861 * configure: Regenerate.
863 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
865 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
866 * src/sh64/sysv.S: Likewise.
868 2008-03-26 Daniel Jacobowitz <dan@debian.org>
870 * src/arm/sysv.S: Fix ARM comment marker.
872 2008-03-26 Jakub Jelinek <jakub@redhat.com>
874 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
875 * src/s390/sysv.S: Likewise.
876 * src/powerpc/ppc_closure.S: Likewise.
877 * src/powerpc/sysv.S: Likewise.
878 * src/x86/unix64.S: Likewise.
879 * src/x86/sysv.S: Likewise.
880 * src/sparc/v8.S: Likewise.
881 * src/sparc/v9.S: Likewise.
882 * src/m68k/sysv.S: Likewise.
883 * src/arm/sysv.S: Likewise.
885 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
887 * aclocal.m4: Regenerate.
888 * configure: Likewise.
889 * Makefile.in: Likewise.
890 * include/Makefile.in: Likewise.
891 * testsuite/Makefile.in: Likewise.
893 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
894 Andreas Tobler <a.tobler@schweiz.org>
896 * configure.ac: Add amd64-*-freebsd* target.
897 * configure: Regenerate.
899 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
902 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
905 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
908 2008-01-24 David Edelsohn <edelsohn@gnu.org>
910 * configure: Regenerate.
912 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
914 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
916 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
919 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
920 signed/unsigned int8/16 for X86_DARWIN.
921 Updated copyright info.
922 Handle one and two byte structs with special cif->flags.
923 * src/x86/ffitarget.h: Add special types for one and two byte structs.
924 Updated copyright info.
925 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
927 Remove code to pop args from the stack after call.
928 Special-case signed/unsigned for int8/16, one and two byte structs.
929 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
930 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
932 Updated copyright info.
934 2007-12-08 David Daney <ddaney@avtrex.com>
936 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
937 SUBU, add with ADDU and use smaller code sequences.
939 2007-12-07 David Daney <ddaney@avtrex.com>
941 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
944 2007-12-06 David Daney <ddaney@avtrex.com>
946 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
948 (ffi_java_raw): New typedef.
949 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
950 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
952 (ffi_java_raw_closure) : Same.
953 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
955 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
957 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
958 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
959 sizeof(ffi_java_raw) for alignment calculations.
960 (ffi_java_ptrarray_to_raw): Same.
961 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
962 if FFI_SIZEOF_JAVA_RAW == 4.
963 (ffi_java_raw_to_rvalue): Same.
964 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
965 (ffi_java_translate_args): Same.
966 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
968 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
970 2007-12-06 David Daney <ddaney@avtrex.com>
972 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
975 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
978 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
979 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
980 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
981 set the NUM_FPR_ARG_REGISTERS according to.
982 Add support for potential soft-float support under hard-float
984 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
985 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
986 to the FFI_LINUX_SOFT_FLOAT ABI.
987 (ffi_prep_cif_machdep): Likewise.
988 (ffi_closure_helper_SYSV): Likewise.
989 * src/powerpc/ppc_closure.S: Make sure not to store float/double
990 on archs where __NO_FPRS__ is true.
991 Add FFI_TYPE_UINT128 support.
992 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
993 Adjust copyright notice.
995 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
997 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
998 * include/ffi_common.h: ... here.
1001 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
1003 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
1004 * src/powerpc/linux64.S: Likewise.
1005 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
1006 * testsuite/libffi.call/return_ldl.c: New test.
1008 2007-09-04 <aph@redhat.com>
1010 * src/arm/sysv.S (UNWIND): New.
1011 (Whole file): Conditionally compile unwinder directives.
1012 * src/arm/sysv.S: Add unwinder directives.
1014 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
1015 Only treat r0 as a struct address if we're actually returning a
1017 Only copy the bytes that are actually within a struct.
1018 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
1019 is returned in r0, not passed by address.
1020 (ffi_call): Allocate a word-sized temporary for the case where
1021 a composite is returned in r0.
1022 (ffi_prep_incoming_args_SYSV): Align as necessary.
1024 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
1026 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
1027 directly using the sys_cacheflush syscall.
1029 2007-07-27 Andrew Haley <aph@redhat.com>
1031 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
1033 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
1035 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
1036 * configure.ac: Likewise.
1037 * Makefile.in: Regenerate.
1038 * include/Makefile.in: Likewise.
1039 * testsuite/Makefile.in: Likewise.
1040 * configure: Likewise.
1042 2007-08-24 David Daney <ddaney@avtrex.com>
1044 * testsuite/libffi.call/return_sl.c: New test.
1046 2007-08-10 David Daney <ddaney@avtrex.com>
1048 * testsuite/libffi.call/cls_multi_ushort.c,
1049 testsuite/libffi.call/cls_align_uint16.c,
1050 testsuite/libffi.call/nested_struct1.c,
1051 testsuite/libffi.call/nested_struct3.c,
1052 testsuite/libffi.call/cls_7_1_byte.c,
1053 testsuite/libffi.call/nested_struct5.c,
1054 testsuite/libffi.call/cls_double.c,
1055 testsuite/libffi.call/nested_struct7.c,
1056 testsuite/libffi.call/cls_sint.c,
1057 testsuite/libffi.call/nested_struct9.c,
1058 testsuite/libffi.call/cls_20byte1.c,
1059 testsuite/libffi.call/cls_multi_sshortchar.c,
1060 testsuite/libffi.call/cls_align_sint64.c,
1061 testsuite/libffi.call/cls_3byte2.c,
1062 testsuite/libffi.call/cls_multi_schar.c,
1063 testsuite/libffi.call/cls_multi_uchar.c,
1064 testsuite/libffi.call/cls_19byte.c,
1065 testsuite/libffi.call/cls_9byte1.c,
1066 testsuite/libffi.call/cls_align_float.c,
1067 testsuite/libffi.call/closure_fn1.c,
1068 testsuite/libffi.call/problem1.c,
1069 testsuite/libffi.call/closure_fn3.c,
1070 testsuite/libffi.call/cls_sshort.c,
1071 testsuite/libffi.call/closure_fn5.c,
1072 testsuite/libffi.call/cls_align_double.c,
1073 testsuite/libffi.call/nested_struct.c,
1074 testsuite/libffi.call/cls_2byte.c,
1075 testsuite/libffi.call/nested_struct10.c,
1076 testsuite/libffi.call/cls_4byte.c,
1077 testsuite/libffi.call/cls_6byte.c,
1078 testsuite/libffi.call/cls_8byte.c,
1079 testsuite/libffi.call/cls_multi_sshort.c,
1080 testsuite/libffi.call/cls_align_sint16.c,
1081 testsuite/libffi.call/cls_align_uint32.c,
1082 testsuite/libffi.call/cls_20byte.c,
1083 testsuite/libffi.call/cls_float.c,
1084 testsuite/libffi.call/nested_struct2.c,
1085 testsuite/libffi.call/cls_5_1_byte.c,
1086 testsuite/libffi.call/nested_struct4.c,
1087 testsuite/libffi.call/cls_24byte.c,
1088 testsuite/libffi.call/nested_struct6.c,
1089 testsuite/libffi.call/cls_64byte.c,
1090 testsuite/libffi.call/nested_struct8.c,
1091 testsuite/libffi.call/cls_uint.c,
1092 testsuite/libffi.call/cls_multi_ushortchar.c,
1093 testsuite/libffi.call/cls_schar.c,
1094 testsuite/libffi.call/cls_uchar.c,
1095 testsuite/libffi.call/cls_align_uint64.c,
1096 testsuite/libffi.call/cls_ulonglong.c,
1097 testsuite/libffi.call/cls_align_longdouble.c,
1098 testsuite/libffi.call/cls_1_1byte.c,
1099 testsuite/libffi.call/cls_12byte.c,
1100 testsuite/libffi.call/cls_3_1byte.c,
1101 testsuite/libffi.call/cls_3byte1.c,
1102 testsuite/libffi.call/cls_4_1byte.c,
1103 testsuite/libffi.call/cls_6_1_byte.c,
1104 testsuite/libffi.call/cls_16byte.c,
1105 testsuite/libffi.call/cls_18byte.c,
1106 testsuite/libffi.call/closure_fn0.c,
1107 testsuite/libffi.call/cls_9byte2.c,
1108 testsuite/libffi.call/closure_fn2.c,
1109 testsuite/libffi.call/closure_fn4.c,
1110 testsuite/libffi.call/cls_ushort.c,
1111 testsuite/libffi.call/closure_fn6.c,
1112 testsuite/libffi.call/cls_5byte.c,
1113 testsuite/libffi.call/cls_align_pointer.c,
1114 testsuite/libffi.call/cls_7byte.c,
1115 testsuite/libffi.call/cls_align_sint32.c,
1116 testsuite/libffi.special/unwindtest_ffi_call.cc,
1117 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
1119 2007-08-10 David Daney <ddaney@avtrex.com>
1122 * configure.ac: Don't treat mips64 as a special case.
1123 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
1124 * configure: Regenerate
1125 * Makefile.in: Ditto.
1126 * fficonfig.h.in: Ditto.
1127 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
1128 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
1129 (FFI_DEFAULT_ABI): Set for n64 case.
1130 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
1131 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
1132 (ffi_closure_N32): New function.
1133 (.eh_frame): New section
1134 * src/mips/o32.S: Clean up comments.
1135 (ffi_closure_O32): Pass ffi_closure parameter in $12.
1136 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
1137 _MIPS_SIM == _ABIN32 throughout.
1138 (FFI_MIPS_STOP_HERE): New, use in place of
1140 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1141 to support n32/n64 ABIs.
1142 (calc_n32_struct_flags): Rewrite.
1143 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1144 position of flag bits.
1145 (ffi_prep_cif_machdep): Rewrite n32 portion.
1146 (ffi_call): Enable for n64. Add special handling for small structure
1148 (ffi_prep_closure_loc): Add n32 and n64 support.
1149 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1150 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1152 2007-08-08 David Daney <ddaney@avtrex.com>
1154 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1155 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1157 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1158 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1159 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1160 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1161 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1162 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1163 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1164 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1165 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1166 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1167 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1168 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1169 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1170 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1171 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1172 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1173 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1174 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1175 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1176 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1177 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1178 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1179 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1180 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1181 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1182 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1183 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1184 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1185 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1186 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1187 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1189 2007-08-07 Andrew Haley <aph@redhat.com>
1191 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1194 2007-08-06 Andrew Haley <aph@redhat.com>
1197 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1198 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1201 2007-08-02 David Daney <ddaney@avtrex.com>
1203 * testsuite/libffi.call/return_ul.c (main): Define return type as
1204 ffi_arg. Use proper printf conversion specifier.
1206 2007-07-30 Andrew Haley <aph@redhat.com>
1209 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1210 signed/unsigned int8/16.
1211 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1213 Remove code to pop args from the stack after call.
1214 Special-case signed/unsigned int8/16.
1215 * testsuite/libffi.call/return_sc.c (main): Revert.
1217 2007-07-26 Richard Guenther <rguenther@suse.de>
1220 * testsuite/libffi.call/return_sc.c (main): Verify call
1221 result as signed char, not ffi_arg.
1223 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1225 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1226 * configure: Regenerate.
1228 2007-07-11 David Daney <ddaney@avtrex.com>
1230 * src/mips/ffi.c: Don't include sys/cachectl.h.
1231 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1234 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1236 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1237 from (ffi_prep_closure): ... this.
1238 (FFI_INIT_TRAMPOLINE): Adjust.
1240 2005-12-31 Phil Blundell <pb@reciva.com>
1242 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1243 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1244 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1245 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1246 (FFI_CLOSURES): Enable closure support.
1248 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1250 * testsuite/libffi.call/cls_multi_ushort.c,
1251 testsuite/libffi.call/cls_align_uint16.c,
1252 testsuite/libffi.call/nested_struct1.c,
1253 testsuite/libffi.call/nested_struct3.c,
1254 testsuite/libffi.call/cls_7_1_byte.c,
1255 testsuite/libffi.call/cls_double.c,
1256 testsuite/libffi.call/nested_struct5.c,
1257 testsuite/libffi.call/nested_struct7.c,
1258 testsuite/libffi.call/cls_sint.c,
1259 testsuite/libffi.call/nested_struct9.c,
1260 testsuite/libffi.call/cls_20byte1.c,
1261 testsuite/libffi.call/cls_multi_sshortchar.c,
1262 testsuite/libffi.call/cls_align_sint64.c,
1263 testsuite/libffi.call/cls_3byte2.c,
1264 testsuite/libffi.call/cls_multi_schar.c,
1265 testsuite/libffi.call/cls_multi_uchar.c,
1266 testsuite/libffi.call/cls_19byte.c,
1267 testsuite/libffi.call/cls_9byte1.c,
1268 testsuite/libffi.call/cls_align_float.c,
1269 testsuite/libffi.call/closure_fn1.c,
1270 testsuite/libffi.call/problem1.c,
1271 testsuite/libffi.call/closure_fn3.c,
1272 testsuite/libffi.call/cls_sshort.c,
1273 testsuite/libffi.call/closure_fn5.c,
1274 testsuite/libffi.call/cls_align_double.c,
1275 testsuite/libffi.call/cls_2byte.c,
1276 testsuite/libffi.call/nested_struct.c,
1277 testsuite/libffi.call/nested_struct10.c,
1278 testsuite/libffi.call/cls_4byte.c,
1279 testsuite/libffi.call/cls_6byte.c,
1280 testsuite/libffi.call/cls_8byte.c,
1281 testsuite/libffi.call/cls_multi_sshort.c,
1282 testsuite/libffi.call/cls_align_uint32.c,
1283 testsuite/libffi.call/cls_align_sint16.c,
1284 testsuite/libffi.call/cls_float.c,
1285 testsuite/libffi.call/cls_20byte.c,
1286 testsuite/libffi.call/cls_5_1_byte.c,
1287 testsuite/libffi.call/nested_struct2.c,
1288 testsuite/libffi.call/cls_24byte.c,
1289 testsuite/libffi.call/nested_struct4.c,
1290 testsuite/libffi.call/nested_struct6.c,
1291 testsuite/libffi.call/cls_64byte.c,
1292 testsuite/libffi.call/nested_struct8.c,
1293 testsuite/libffi.call/cls_uint.c,
1294 testsuite/libffi.call/cls_multi_ushortchar.c,
1295 testsuite/libffi.call/cls_schar.c,
1296 testsuite/libffi.call/cls_uchar.c,
1297 testsuite/libffi.call/cls_align_uint64.c,
1298 testsuite/libffi.call/cls_ulonglong.c,
1299 testsuite/libffi.call/cls_align_longdouble.c,
1300 testsuite/libffi.call/cls_1_1byte.c,
1301 testsuite/libffi.call/cls_12byte.c,
1302 testsuite/libffi.call/cls_3_1byte.c,
1303 testsuite/libffi.call/cls_3byte1.c,
1304 testsuite/libffi.call/cls_4_1byte.c,
1305 testsuite/libffi.call/cls_6_1_byte.c,
1306 testsuite/libffi.call/cls_16byte.c,
1307 testsuite/libffi.call/cls_18byte.c,
1308 testsuite/libffi.call/closure_fn0.c,
1309 testsuite/libffi.call/cls_9byte2.c,
1310 testsuite/libffi.call/closure_fn2.c,
1311 testsuite/libffi.call/closure_fn4.c,
1312 testsuite/libffi.call/cls_ushort.c,
1313 testsuite/libffi.call/closure_fn6.c,
1314 testsuite/libffi.call/cls_5byte.c,
1315 testsuite/libffi.call/cls_align_pointer.c,
1316 testsuite/libffi.call/cls_7byte.c,
1317 testsuite/libffi.call/cls_align_sint32.c,
1318 testsuite/libffi.special/unwindtest_ffi_call.cc,
1319 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1321 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1323 * aclocal.m4: Regenerated.
1325 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1327 * configure: Regenerate.
1329 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1331 * Makefile.in: Regenerate.
1332 * configure: Regenerate.
1333 * aclocal.m4: Regenerate.
1334 * include/Makefile.in: Regenerate.
1335 * testsuite/Makefile.in: Regenerate.
1337 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1339 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1340 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1341 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1342 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1343 (FFI_CLOSURES): Enable closure support.
1345 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1347 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1348 * configure: Regenerate.
1349 * fficonfig.h.in: Regenerate.
1350 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1351 CFI_OFFSET,CFI_DEF_CFA): New macros.
1352 (ffi_call_SYSV): Add callframe annotation.
1354 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1356 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1357 numerous test suite failures.
1358 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1360 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1362 * Makefile.am (EXTRA_DIST): Bring up to date.
1363 * Makefile.in: Regenerate.
1364 * src/frv/eabi.S: Remove RCS keyword.
1366 2007-04-06 Richard Henderson <rth@redhat.com>
1368 * configure.ac: Tidy target case.
1369 (HAVE_LONG_DOUBLE): Allow the target to override.
1370 * configure: Regenerate.
1371 * include/ffi.h.in: Don't define ffi_type_foo if
1372 LIBFFI_HIDE_BASIC_TYPES is defined.
1373 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1375 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1376 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1377 (ffi_type_longdouble): Special case for Alpha. Don't define
1378 if long double == double.
1380 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1381 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1382 (ffi_call, ffi_closure_osf_inner): Likewise.
1383 (ffi_closure_osf_inner): Likewise. Mark hidden.
1384 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1385 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1386 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1387 (load_table): Handle 128-bit long double.
1389 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1391 2007-04-06 Tom Tromey <tromey@redhat.com>
1394 * README: Fixed bug in example.
1396 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1398 * src/closures.c: Include sys/statfs.h.
1399 (_GNU_SOURCE): Define on Linux.
1400 (FFI_MMAP_EXEC_SELINUX): Define.
1401 (selinux_enabled): New variable.
1402 (selinux_enabled_check): New function.
1403 (is_selinux_enabled): Define.
1406 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1408 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1409 Use 'volatile float sum' to create sum of floats to avoid false
1410 negative due to excess precision on ix86 targets.
1413 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1415 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1417 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1419 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1421 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1422 (ffi_prep_closure_loc): New.
1423 (ffi_prep_raw_closure_loc): New.
1424 (ffi_prep_java_raw_closure_loc): New.
1425 * src/closures.c: New file.
1426 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1427 Replace sflags with exec_offset.
1428 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1429 sub_segment_exec_offset): New macros.
1430 (get_segment_flags, set_segment_flags, check_segment_merge): New
1432 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1433 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1434 destroy_mspace): Use new macros.
1435 (sys_alloc): Silence warning.
1436 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1437 * Makefile.in: Rebuilt.
1438 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1439 terms of ffi_prep_closure_loc.
1440 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1442 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1444 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1446 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1447 the renamed version.
1448 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1449 (ffi_prep_closure): ... this.
1450 * src/pa/ffi.c: Likewise.
1451 * src/cris/ffi.c: Likewise. Adjust.
1452 * src/frv/ffi.c: Likewise.
1453 * src/ia64/ffi.c: Likewise.
1454 * src/mips/ffi.c: Likewise.
1455 * src/powerpc/ffi_darwin.c: Likewise.
1456 * src/s390/ffi.c: Likewise.
1457 * src/sh/ffi.c: Likewise.
1458 * src/sh64/ffi.c: Likewise.
1459 * src/sparc/ffi.c: Likewise.
1460 * src/x86/ffi64.c: Likewise.
1461 * src/x86/ffi.c: Likewise.
1462 (FFI_INIT_TRAMPOLINE): Adjust.
1463 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1464 (ffi_prep_raw_closure): ... this.
1465 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1466 (ffi_prep_closure): ... this.
1467 (flush_icache): Adjust.
1469 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1471 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1474 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1476 * Makefile.am: Add dummy install-pdf target.
1477 * Makefile.in: Regenerate
1479 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1481 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1482 ffi_closure_helper_SYSV): Add long double handling.
1484 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1486 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1487 immediately after bctrl instruction.
1489 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1491 * Makefile.am (all-recursive, install-recursive,
1492 mostlyclean-recursive, clean-recursive, distclean-recursive,
1493 maintainer-clean-recursive): Add missing targets.
1494 * Makefile.in: Rebuilt.
1496 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1498 * configure.ac: Add TARGET for x86_64-*-darwin*.
1499 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1501 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1502 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1503 * configure: Regenerate.
1504 * Makefile.in: Regenerate.
1505 * include/Makefile.in: Regenerate.
1506 * testsuite/Makefile.in: Regenerate.
1507 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1510 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1512 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1515 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1517 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1518 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1520 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1521 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1523 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1524 Sandro Tolaini <tolaini@libero.it>
1526 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1528 * configure: Regenerated.
1529 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1530 (EXTRA_DIST): Add src/x86/darwin.S.
1531 * Makefile.in: Regenerated.
1532 * include/Makefile.in: Regenerated.
1533 * testsuite/Makefile.in: Regenerated.
1535 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1536 X86_WIN32, and additionally align stack to 16 bytes.
1537 * src/x86/darwin.S: New, based on sysv.S.
1538 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1540 2006-09-12 David Daney <ddaney@avtrex.com>
1543 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1544 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1545 * aclocal.m4: Regenerated for automake 1.9.6.
1546 * Makefile.in: Regenerated.
1547 * include/Makefile.in: Regenerated.
1548 * testsuite/Makefile.in: Regenerated.
1550 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1552 * include/ffi_common.h (struct): Revert accidental commit.
1554 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1556 * include/ffi_common.h: Remove lint directives.
1557 * include/ffi.h.in: Likewise.
1559 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1561 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1562 for 32-bit architectures.
1563 * testsuite/libffi.call/return_ul.c: New test case.
1565 2006-07-19 David Daney <ddaney@avtrex.com>
1567 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1568 xfail remains for mips64.
1570 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1572 * Makefile.am: Add install-html target. Add install-html to .PHONY
1573 * Makefile.in: Regenerate.
1574 * aclocal.m4: Regenerate.
1575 * include/Makefile.in: Regenerate.
1576 * testsuite/Makefile.in: Regenerate.
1578 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1580 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1583 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1585 * README: Remove notice about 'Crazy Comments'.
1586 * src/debug.c: Remove lint directives. Cleanup white spaces.
1587 * src/java_raw_api.c: Likewise.
1588 * src/prep_cif.c: Likewise.
1589 * src/raw_api.c: Likewise.
1590 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1592 * src/arm/ffi.c: Remove lint directives.
1593 * src/m32r/ffi.c: Likewise.
1594 * src/pa/ffi.c: Likewise.
1595 * src/powerpc/ffi.c: Likewise.
1596 * src/powerpc/ffi_darwin.c: Likewise.
1597 * src/sh/ffi.c: Likewise.
1598 * src/sh64/ffi.c: Likewise.
1599 * src/x86/ffi.c: Likewise.
1600 * testsuite/libffi.call/float2.c: Likewise.
1601 * testsuite/libffi.call/promotion.c: Likewise.
1602 * testsuite/libffi.call/struct1.c: Likewise.
1604 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1606 * src/pa/hpux32.S: Correct unwind offset calculation for
1608 * src/pa/linux.S: Likewise.
1610 2006-04-12 James E Wilson <wilson@specifix.com>
1613 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1614 (hfa_type_load): Call stf_spill.
1615 (hfa_type_store): Call ldf_fill.
1616 (ffi_call): Adjust calls to above routines. Add local temps for
1619 2006-04-10 Matthias Klose <doko@debian.org>
1621 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1622 directory names containing underscores.
1624 2006-04-07 James E Wilson <wilson@specifix.com>
1626 * testsuite/libffi.call/float4.c: New testcase.
1628 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1629 Andreas Tobler <a.tobler@schweiz.ch>
1631 * Makefile.am: Add PA_HPUX port.
1632 * Makefile.in: Regenerate.
1633 * include/Makefile.in: Likewise.
1634 * testsuite/Makefile.in: Likewise.
1635 * configure.ac: Add PA_HPUX rules.
1636 * configure: Regenerate.
1637 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1638 Add PA_HPUX and PA64_HPUX.
1639 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1640 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1641 (FFI_TYPE_SMALL_STRUCT2): Define.
1642 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1643 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1644 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1645 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1646 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1647 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1648 * src/pa/ffi.c (ROUND_DOWN): Delete.
1649 (fldw, fstw, fldd, fstd): Use '__asm__'.
1650 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1651 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1652 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1653 Simplify incrementing of stack slot variable. Change type of local
1654 'n' to unsigned int.
1655 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1657 (ffi_prep_cif_machdep): Likewise.
1658 (ffi_call): Likewise.
1659 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1660 return type to ffi_status. Simplify incrementing of stack slot
1661 variable. Only copy floating point argument registers when PA_LINUX
1662 is true. Reformat debug statement.
1663 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1664 FFI_TYPE_SMALL_STRUCT8.
1665 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1667 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1668 Add nops to cache flush. Add trampoline for PA_HPUX.
1669 * src/pa/hpux32.S: New file.
1670 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1671 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1672 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1673 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1674 argument type checks so that common argument types appear first.
1675 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1676 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1678 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1680 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1681 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1682 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1683 FFI_TYPE_LONGDOUBLE.
1684 (ffi_prep_args64): Assert using IBM extended double.
1685 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1686 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1687 (ffi_call): Handle FFI_LINUX.
1688 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1689 gpr3 return pointer as for struct return. Handle FFI_LINUX
1690 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1692 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1693 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1694 Don't use r6 as pointer to results, instead use sp offset. Don't
1695 make a special call to load lr with case table address, instead
1696 use offset from previous call.
1697 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1698 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1701 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1703 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1704 passed with FP registers correctly.
1705 (ffi_closure_helper_SYSV): Likewise.
1706 * src/sh64/sysv.S: Likewise.
1708 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1710 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1711 args and userdata unused.
1712 (closure_test_fn1): Mark cif and userdata unused.
1713 (main): Remove unused res.
1715 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1717 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1718 -O2, -O3, -Os and the warning flags -W -Wall.
1719 * testsuite/libffi.special/special.exp: Likewise.
1720 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1721 unused parameter unused for gcc or else do nothing.
1722 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1723 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1724 and userdata unused.
1725 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1726 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1727 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1728 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1729 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1730 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1731 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1732 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1733 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1734 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1735 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1736 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1737 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1738 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1739 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1740 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1741 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1742 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1743 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1744 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1745 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1746 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1747 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1748 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1750 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1752 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1754 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1755 void* to avoid compiler warning.
1757 (cls_struct_align_gn): Mark cif and userdata unused.
1758 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1760 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1762 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1764 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1766 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1768 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1769 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1770 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1772 (main): Cast res_call to silence gcc.
1773 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1775 (main): Cast res_call to silence gcc.
1776 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1778 (main): Cast res_call to silence gcc.
1779 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1781 (main): Cast res_call to silence gcc.
1782 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1784 (main): Cast res_call to silence gcc.
1785 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1787 (main): Cast res_call to silence gcc.
1788 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1790 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1791 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1793 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1794 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1796 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1797 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1799 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1800 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1802 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1803 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1804 and userdata unused.
1805 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1807 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1808 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1809 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1810 Cleanup white spaces.
1811 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1812 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1813 cif and userdata unused.
1814 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1816 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1817 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1818 formatters to silence gcc.
1819 (B_gn): Mark cif and userdata unused.
1820 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1822 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1823 (B_gn): Mark cif and userdata unused.
1824 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1826 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1827 (B_gn): Mark cif and userdata unused.
1828 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1830 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1831 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1832 * testsuite/libffi.call/problem1.c (stub): Likewise.
1833 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1835 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1836 in the last commit for this test case in the test case itself.
1837 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1839 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1840 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1841 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1842 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1843 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1844 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1846 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1848 * src/sh/sysv.S: Fix register numbers in the FDE for
1851 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1853 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1854 declaration to avoid a false negative on ix86. See PR323.
1856 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1858 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1859 and cast integer to void * if needed. Update the pointer to
1860 the FP register saved area correctly.
1862 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1864 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1866 * testsuite/libffi.call/nested_struct4.c: Likewise.
1868 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1870 * testsuite/libffi.call/return_dbl.c: New test case.
1871 * testsuite/libffi.call/return_dbl1.c: Likewise.
1872 * testsuite/libffi.call/return_dbl2.c: Likewise.
1873 * testsuite/libffi.call/return_fl.c: Likewise.
1874 * testsuite/libffi.call/return_fl1.c: Likewise.
1875 * testsuite/libffi.call/return_fl2.c: Likewise.
1876 * testsuite/libffi.call/return_fl3.c: Likewise.
1877 * testsuite/libffi.call/closure_fn6.c: Likewise.
1879 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1881 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1882 here to be used by other test cases too.
1884 * testsuite/libffi.call/nested_struct10.c: New test case.
1885 * testsuite/libffi.call/nested_struct9.c: Likewise.
1886 * testsuite/libffi.call/nested_struct8.c: Likewise.
1887 * testsuite/libffi.call/nested_struct7.c: Likewise.
1888 * testsuite/libffi.call/nested_struct6.c: Likewise.
1889 * testsuite/libffi.call/nested_struct5.c: Likewise.
1890 * testsuite/libffi.call/nested_struct4.c: Likewise.
1892 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1894 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1895 * configure: Rebuilt.
1897 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1899 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1900 instead do the shifting inline.
1901 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1902 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1903 and 8 byte structs, for the remaining struct sizes don't call
1904 __lshrdi3, instead do the shifting inline.
1906 2005-12-07 Thiemo Seufer <ths@networkno.de>
1908 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1909 missing parentheses.
1910 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1911 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1912 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1913 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1914 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1916 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1918 (ffi_prep_closure): Improve trampoline instruction scheduling.
1919 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1921 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1923 * src/powerpc/ffi.c: Formatting.
1924 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1925 (ffi_prep_args64): Likewise.
1927 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1929 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1930 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1932 2005-09-26 Tom Tromey <tromey@redhat.com>
1934 * testsuite/libffi.call/float1.c (value_type): New typedef.
1935 (CANARY): New define.
1936 (main): Check for result buffer overflow.
1937 * src/powerpc/linux64.S: Handle linux64 long double returns.
1938 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1939 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1941 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1944 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1946 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1948 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1950 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1951 (AH_BOTTOM): Add FFI_HIDDEN definition.
1952 * configure: Rebuilt.
1953 * fficonfig.h.in: Rebuilt.
1954 * src/powerpc/ffi.c (hidden): Remove.
1955 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1956 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1957 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1958 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1959 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1960 add FFI_HIDDEN to its prototype.
1961 (ffi_closure_SYSV_inner): New.
1962 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1963 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1965 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1968 * configure: Rebuilt.
1969 * configure.ac: Handle i*86-*-gnu*.
1971 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1973 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1974 DW_CFA_offset_extended_sf rather than
1975 DW_CFA_GNU_negative_offset_extended.
1976 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1978 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1980 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1982 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1983 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1984 partially on register.
1985 (ffi_closure_helper_SYSV): Likewise.
1986 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1988 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1990 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1992 * src/sh64/ffi.c (simple_type): Remove.
1993 (return_type): Handle small structures correctly.
1994 (ffi_prep_args): Likewise.
1995 (ffi_call): Likewise.
1996 (ffi_closure_helper_SYSV): Likewise.
1997 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1998 Emit position independent code if PIC and remove wrong datalabel
1999 prefixes from EH data.
2001 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
2003 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
2004 * Makefile.in: Regenerate.
2005 * include/Makefile.in: Likewise.
2006 * testsuite/Makefile.in: Likewise.
2007 * configure.ac: Add POWERPC_FREEBSD rules.
2008 * configure: Regenerate.
2009 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
2010 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
2011 * src/powerpc/ffi.c: Add flags to handle small structure returns
2013 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
2015 (ffi_closure_helper_SYSV): Likewise.
2016 * src/powerpc/ppc_closure.S: Add return types for small structures.
2017 * src/powerpc/sysv.S: Add bits to handle small structures for
2020 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
2022 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
2023 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
2024 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
2026 2005-07-05 Randolph Chung <tausq@debian.org>
2028 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
2029 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
2030 structures. Kill compilation warnings.
2031 (ffi_closure_inner_LINUX): Print return values as hex in debug
2032 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
2033 Properly handle 5-7 byte structure returns.
2034 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
2035 (FFI_TYPE_SMALL_STRUCT2): Remove.
2036 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
2037 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
2038 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
2039 (checksmst1, checksmst2): Remove.
2040 (checksmst3): Optimize handling of 3-byte struct returns.
2041 (checksmst567): Properly handle 5-7 byte struct returns.
2043 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2046 * src/mips/n32.S: Enforce PIC code.
2047 * src/mips/o32.S: Likewise.
2049 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2051 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
2052 * configure: Regenerate.
2054 2005-06-01 Alan Modra <amodra@bigpond.net.au>
2056 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
2057 to call ffi_closure_helper_SYSV. Append @local instead.
2058 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
2060 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
2062 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
2063 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
2064 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
2065 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
2066 include/Makefile.in, testsuite/Makefile.in: Regenerate.
2068 2005-05-09 Mike Stump <mrs@apple.com>
2070 * configure: Regenerate.
2072 2005-05-08 Richard Henderson <rth@redhat.com>
2075 * src/alpha/osf.S: Update unwind into to match code.
2077 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
2078 Richard Henderson <rth@redhat.com>
2080 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
2082 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
2083 (ffi_prep_closure): Set carry bit if sse-used flag set.
2084 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
2085 Only load sse registers if ssecount non-zero.
2086 (ffi_closure_unix64): Only save sse registers if carry set on entry.
2088 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
2090 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
2091 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
2092 * configure: Regenerate.
2094 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
2096 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
2097 have Tcl8.3-compatible intermediate variable.
2099 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
2100 Hans-Peter Nilsson <hp@axis.com>
2102 * Makefile.am: Add CRIS support.
2103 * configure.ac: Likewise.
2104 * Makefile.in, configure, testsuite/Makefile.in,
2105 include/Makefile.in: Regenerate.
2106 * src/cris: New directory.
2107 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
2108 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
2110 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
2111 \r?\n in output tests.
2113 2005-04-12 Mike Stump <mrs@apple.com>
2115 * configure: Regenerate.
2117 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
2119 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
2121 2005-03-30 Steve Ellcey <sje@cup.hp.com>
2123 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
2125 * src/ia64/unix.S (ffi_closure_unix): Extend gp
2126 to 64 bits in ILP32 mode.
2127 Load 64 bits even for short data.
2129 2005-03-23 Mike Stump <mrs@apple.com>
2131 * src/powerpc/darwin.S: Update for -m64 multilib.
2132 * src/powerpc/darwin_closure.S: Likewise.
2134 2005-03-21 Zack Weinberg <zack@codesourcery.com>
2136 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
2137 Do not set tool_include_dir.
2138 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
2140 * include/Makefile.am: Set gcc_version and toollibffidir.
2141 * include/Makefile.in: Regenerate.
2143 2005-02-22 Andrew Haley <aph@redhat.com>
2145 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2146 odd-numbered register pairs for 64-bit integer types.
2148 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2151 * testsuite/libffi.call/return_ll1.c: New test case.
2153 2005-02-11 Janis Johnson <janis187@us.ibm.com>
2155 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2156 * testsuite/libffi.call/float.c: Ditto.
2157 * testsuite/libffi.call/float2.c: Ditto.
2158 * testsuite/libffi.call/float3.c: Ditto.
2160 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2162 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2164 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2166 * testsuite/libffi.special/special.exp (cxx_options): Add
2169 2004-12-31 Richard Henderson <rth@redhat.com>
2171 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2172 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2173 offset parameters with a type parameter; deduce size and structure
2174 alignment. Update all users.
2176 2004-12-31 Richard Henderson <rth@redhat.com>
2178 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2179 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
2180 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
2181 into ffi_prep_closure.
2182 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2185 2004-12-27 Richard Henderson <rth@redhat.com>
2187 * src/x86/unix64.S: Fix typo in unwind info.
2189 2004-12-25 Richard Henderson <rth@redhat.com>
2191 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2192 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2193 (merge_classes): Check for it.
2195 (classify_argument): Pass byte_offset by value; perform all updates
2197 (examine_argument): Add classes argument; handle
2198 X86_64_COMPLEX_X87_CLASS.
2199 (ffi_prep_args): Merge into ...
2200 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2201 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2202 (ffi_fill_return_value): Remove.
2203 (ffi_prep_closure): Remove dead assert.
2204 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2205 Rewrite to use struct register_args instead of va_list. Create
2206 flags for handling structure returns.
2207 * src/x86/unix64.S: Remove dead strings.
2208 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2209 stack frame with ffi_call. Handle structure returns properly.
2210 (float2sse, floatfloat2sse, double2sse): Remove.
2211 (sse2float, sse2double, sse2floatfloat): Remove.
2212 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2213 to handle structure returns properly.
2215 2004-12-08 David Edelsohn <edelsohn@gnu.org>
2217 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2219 * Makefile.in: Regenerated.
2221 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2223 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2224 * configure, aclocal.m4, Makefile.in: Regenerate.
2225 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2227 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2229 * configure: Regenerate for libtool change.
2231 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2233 * configure: Regenerate for libtool reversion.
2235 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2237 * configure: Regenerate for libtool change.
2239 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2241 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2243 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2245 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2246 of jal. Use an absolute encoding for the frame information.
2248 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2250 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2251 * acinclude.m4: Delete logic for sincludes.
2252 * aclocal.m4, Makefile.in, configure: Regenerate.
2253 * include/Makefile: Likewise.
2254 * testsuite/Makefile: Likewise.
2256 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2258 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2259 on a 8-byte boundary.
2260 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2262 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2264 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2265 long long values. Round stack allocation to a multiple of 8 bytes
2266 for ATPCS compatibility.
2267 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2268 names. Handle returning long long types. Add Thumb and interworking
2269 support. Improve soft-float code.
2271 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2273 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2274 (libffi_exit): New function.
2275 (libffi_init): Build the testglue wrapper if needed.
2277 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2280 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2282 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2284 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2286 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2288 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2289 * testsuite/libffi.call/float3.c: New test case.
2291 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2293 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2294 the function returning a structure pointed with R2.
2295 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2296 the structure return value if T bit set. Emit position
2297 independent code and EH data if PIC.
2299 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2301 * Makefile.am: Add m32r support.
2302 * configure.ac: Likewise.
2303 * Makefile.in: Regenerate.
2304 * confiugre: Regenerate.
2305 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2306 (uint64, sint64, double, longdouble)
2307 * src/m32r: New directory.
2308 * src/m32r/ffi.c: New file.
2309 * src/m32r/sysv.S: Likewise.
2310 * src/m32r/ffitarget.h: Likewise.
2312 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2314 * testsuite/libffi.call/negint.c: New test case.
2316 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2319 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2320 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2321 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2324 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2326 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2327 * testsuite/libffi.call/promotion.c: Likewise.
2328 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2330 * testsuite/libffi.call/return_sc.c: Likewise.
2331 * testsuite/libffi.call/return_uc.c: Likewise.
2333 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2335 * src/powerpc/darwin.S: Fix comments and identation.
2336 * src/powerpc/darwin_closure.S: Likewise.
2338 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2340 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2341 (ffi_prep_args): Handle longdouble arguments.
2342 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2344 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2345 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2347 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2348 * src/types.c: Defined longdouble size and alignment for darwin.
2350 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2352 * src/powerpc/aix.S: Remove whitespaces.
2353 * src/powerpc/aix_closure.S: Likewise.
2354 * src/powerpc/asm.h: Likewise.
2355 * src/powerpc/ffi.c: Likewise.
2356 * src/powerpc/ffitarget.h: Likewise.
2357 * src/powerpc/linux64.S: Likewise.
2358 * src/powerpc/linux64_closure.S: Likewise.
2359 * src/powerpc/ppc_closure.S: Likewise.
2360 * src/powerpc/sysv.S: Likewise.
2362 2004-08-30 Anthony Green <green@redhat.com>
2364 * Makefile.am: Add frv support.
2365 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2366 * configure.ac: Read configure.host.
2367 * configure.in: Read configure.host.
2368 * configure.host: New file. frv-elf needs libgloss.
2369 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2370 alignment. This is needed to frv and shouldn't harm the others.
2371 * include/ffi_common.h (ALIGN_DOWN): New macro.
2372 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2374 2004-08-24 David Daney <daney@avtrex.com>
2376 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2377 * testsuite/libffi.call/closure_fn1.c: Likewise.
2378 * testsuite/libffi.call/closure_fn2.c Likewise.
2379 * testsuite/libffi.call/closure_fn3.c: Likewise.
2380 * testsuite/libffi.call/closure_fn4.c: Likewise.
2381 * testsuite/libffi.call/closure_fn5.c: Likewise.
2382 * testsuite/libffi.call/cls_18byte.c: Likewise.
2383 * testsuite/libffi.call/cls_19byte.c: Likewise.
2384 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2385 * testsuite/libffi.call/cls_20byte.c: Likewise.
2386 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2387 * testsuite/libffi.call/cls_24byte.c: Likewise.
2388 * testsuite/libffi.call/cls_2byte.c: Likewise.
2389 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2390 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2391 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2392 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2393 * testsuite/libffi.call/cls_4byte.c: Likewise.
2394 * testsuite/libffi.call/cls_64byte.c: Likewise.
2395 * testsuite/libffi.call/cls_6byte.c: Likewise.
2396 * testsuite/libffi.call/cls_7byte.c: Likewise.
2397 * testsuite/libffi.call/cls_8byte.c: Likewise.
2398 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2399 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2400 * testsuite/libffi.call/cls_align_double.c: Likewise.
2401 * testsuite/libffi.call/cls_align_float.c: Likewise.
2402 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2403 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2404 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2405 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2406 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2407 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2408 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2409 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2410 * testsuite/libffi.call/cls_double.c: Likewise.
2411 * testsuite/libffi.call/cls_float.c: Likewise.
2412 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2413 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2414 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2415 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2416 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2417 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2418 * testsuite/libffi.call/cls_schar.c: Likewise.
2419 * testsuite/libffi.call/cls_sint.c: Likewise.
2420 * testsuite/libffi.call/cls_sshort.c: Likewise.
2421 * testsuite/libffi.call/cls_uchar.c: Likewise.
2422 * testsuite/libffi.call/cls_uint.c: Likewise.
2423 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2424 * testsuite/libffi.call/cls_ushort.c: Likewise.
2425 * testsuite/libffi.call/nested_struct.c: Likewise.
2426 * testsuite/libffi.call/nested_struct1.c: Likewise.
2427 * testsuite/libffi.call/nested_struct2.c: Likewise.
2428 * testsuite/libffi.call/nested_struct3.c: Likewise.
2429 * testsuite/libffi.call/problem1.c: Likewise.
2430 * testsuite/libffi.special/unwindtest.cc: Likewise.
2431 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2433 * testsuite/libffi.call/cls_16byte.c: Likewise.
2434 * testsuite/libffi.call/cls_5byte.c: Likewise.
2436 2004-08-23 David Daney <daney@avtrex.com>
2439 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2440 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2441 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2442 parameters and return types.
2443 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2444 (ffi_prep_closure): Ditto.
2445 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2446 alignment calculations.
2447 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2448 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2450 2004-08-14 Casey Marshall <csm@gnu.org>
2452 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2453 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2454 integer (O32 ABI only).
2455 (ffi_prep_closure): new function.
2456 (ffi_closure_mips_inner_O32): new function.
2457 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2458 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2459 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2460 64 bit integers correctly.
2461 (ffi_closure_O32): new function.
2462 Added DWARF-2 unwind info for both functions.
2464 2004-08-10 Andrew Haley <aph@redhat.com>
2466 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2468 2004-08-01 Robert Millan <robertmh@gnu.org>
2470 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2471 * configure: Regenerate.
2473 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2475 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2476 and mmap() explicitly instead of relying on preset autoconf cache
2478 * aclocal.m4: Regenerate.
2479 * configure: Regenerate.
2481 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2483 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2484 (ffi_check_float_struct): Remove unused prototype.
2486 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2488 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2489 character on Darwin, use '\n\t' instead.
2491 2004-06-26 Matthias Klose <doko@debian.org>
2493 * libtool-version: Fix typo in revision/age.
2495 2004-06-17 Matthias Klose <doko@debian.org>
2497 * libtool-version: New.
2498 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2499 * Makefile.in: Regenerate.
2501 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2503 * Makefile.am: Remove useless multilib rules.
2504 * Makefile.in: Regenerate.
2505 * aclocal.m4: Regenerate with automake 1.8.5.
2506 * configure.ac: Remove useless multilib configury.
2507 * configure: Regenerate.
2509 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2511 * .cvsignore: New file.
2513 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2515 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2517 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2520 2004-06-06 Sean McNeil <sean@mcneil.com>
2522 * configure.ac: Add x86_64-*-freebsd* support.
2523 * configure: Regenerate.
2525 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2528 * configure.ac: Test for existence of mmap and sys/mman.h before
2529 checking blacklist. Fix suggested by Jim Wilson.
2530 * configure: Regenerate.
2532 2004-04-26 Matt Austern <austern@apple.com>
2534 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2536 * src/powerpc/darwin_closure.S: Likewise.
2538 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2540 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2541 error. Reported by Thomas Heller <theller@python.net>.
2542 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2543 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2545 2004-03-20 Matthias Klose <doko@debian.org>
2547 * src/pa/linux.S: Fix typo.
2549 2004-03-19 Matthias Klose <doko@debian.org>
2551 * Makefile.am: Update.
2552 * Makefile.in: Regenerate.
2553 * src/pa/ffi.h.in: Remove.
2554 * src/pa/ffitarget.h: New file.
2556 2004-02-10 Randolph Chung <tausq@debian.org>
2558 * Makefile.am: Add PA support.
2559 * Makefile.in: Regenerate.
2560 * include/Makefile.in: Regenerate.
2561 * configure.ac: Add PA target.
2562 * configure: Regenerate.
2563 * src/pa/ffi.c: New file.
2564 * src/pa/ffi.h.in: Add PA support.
2565 * src/pa/linux.S: New file.
2566 * prep_cif.c: Add PA support.
2568 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2570 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2572 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2573 with ecif->cif->flags.
2574 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2576 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2577 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2578 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2579 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2581 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2583 * configure.in: Rename file to ...
2584 * configure.ac: ... this.
2585 * fficonfig.h.in: Regenerate.
2586 * Makefile.in: Regenerate.
2587 * include/Makefile.in: Regenerate.
2588 * testsuite/Makefile.in: Regenerate.
2590 2004-03-12 Matt Austern <austern@apple.com>
2592 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2593 changes in EH format resulting from addition of linkonce support.
2594 * src/powerpc/darwin_closure.S: Likewise.
2596 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2597 Paolo Bonzini <bonzini@gnu.org>
2599 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2600 Remove VPATH. Remove rules for object files. Remove multilib support.
2601 (AM_CCASFLAGS): Add.
2602 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2603 (AC_PREREQ): Bump version to 2.59.
2604 (AC_INIT): Fill with version info and bug address.
2605 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2606 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2607 De-precious CC so that the right flags are passed down to multilibs.
2608 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2609 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2610 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2611 * configure: Rebuilt.
2612 * aclocal.m4: Likewise.
2613 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2614 * fficonfig.h.in: Likewise.
2616 2004-03-11 Andreas Schwab <schwab@suse.de>
2618 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2619 arguments from fp registers only for the first 8 parameter slots.
2620 Don't convert a float parameter when passed in memory.
2622 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2624 * configure: Regenerate for config/accross.m4 correction.
2626 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2628 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2629 ecif->cif->bytes to bytes.
2630 (ffi_prep_cif_machdep): Add braces around nested if statement.
2632 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2634 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2636 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2637 (ffi_closure_helper_LINUX64): Fix typo.
2638 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2640 * testsuite/libffi.call/float.c: Likewise.
2641 * testsuite/libffi.call/float2.c: Likewise.
2643 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2645 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2646 long double function return and long double arg handling.
2647 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2648 Use "end_pfr" instead of "nf". Correct long double handling.
2650 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2652 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2653 space for long double return value. Adjust stack frame and offsets.
2654 Load f2 long double return.
2656 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2658 * src/types.c: Use 16 byte long double for POWERPC64.
2660 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2662 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2663 when the structure return address is passed in %o0.
2664 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2665 (ffi_v9_layout_struct): Align the field following a nested structure
2666 on a word boundary. Use memmove instead of memcpy.
2667 (ffi_call): Update call to ffi_V9_return_struct.
2668 (ffi_prep_closure): Define 'ctx' only for V8.
2669 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2670 and ffi_closure_sparc_inner_v9.
2671 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2672 Always skip the structure return address. For structures and long
2673 doubles, copy the argument directly.
2674 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2675 if required. Shift the maximum floating-point slot accordingly. For
2676 big structures, copy the argument directly; otherwise, left-justify the
2677 argument and call ffi_v9_layout_struct to lay out the structure on
2679 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2680 (ffi_closure_v8): Pass the structure return address. Update call to
2681 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2682 Skip the 'unimp' insn when returning long doubles and structures.
2683 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2684 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2685 FFI_TYPE_INT handling. Load structures both in integers and
2686 floating-point registers on return.
2687 * README: Update status of the SPARC port.
2689 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2691 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2693 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2695 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2697 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2698 value as of type ffi_arg, not unsigned int.
2700 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2702 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2705 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2707 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2710 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2712 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2715 2003-12-10 Richard Henderson <rth@redhat.com>
2717 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2718 size_t instead of int.
2720 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2722 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2723 * testsuite/libffi.call/many_win32.c (main): Replace variable
2724 int i with unsigned long ul.
2726 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2727 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2728 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2729 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2730 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2731 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2732 * testsuite/libffi.call/cls_align_float.c: Likewise.
2733 * testsuite/libffi.call/cls_align_double.c: Likewise.
2734 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2735 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2737 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2740 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2741 Align arguments to 32 bits.
2743 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2746 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2747 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2748 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2749 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2750 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2751 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2753 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2755 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2757 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2758 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2760 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2762 * Makefile.in: Rebuilt.
2763 * configure: Likewise.
2764 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2767 2003-11-21 Andreas Jaeger <aj@suse.de>
2768 Andreas Tobler <a.tobler@schweiz.ch>
2770 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2771 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2772 * Makefile.in: Rebuilt.
2773 * aclocal.m4: Likewise.
2774 * configure: Likewise.
2775 * fficonfig.h.in: Likewise.
2776 * testsuite/lib/libffi-dg.exp: Add include dir.
2777 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2778 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2779 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2780 for ffi_closure if available.
2781 * testsuite/libffi.call/closure_fn1.c: Likewise.
2782 * testsuite/libffi.call/closure_fn2.c: Likewise.
2783 * testsuite/libffi.call/closure_fn3.c: Likewise.
2784 * testsuite/libffi.call/closure_fn4.c: Likewise.
2785 * testsuite/libffi.call/closure_fn5.c: Likewise.
2786 * testsuite/libffi.call/cls_12byte.c: Likewise.
2787 * testsuite/libffi.call/cls_16byte.c: Likewise.
2788 * testsuite/libffi.call/cls_18byte.c: Likewise.
2789 * testsuite/libffi.call/cls_19byte.c: Likewise.
2790 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2791 * testsuite/libffi.call/cls_20byte.c: Likewise.
2792 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2793 * testsuite/libffi.call/cls_24byte.c: Likewise.
2794 * testsuite/libffi.call/cls_2byte.c: Likewise.
2795 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2796 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2797 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2798 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2799 * testsuite/libffi.call/cls_4byte.c: Likewise.
2800 * testsuite/libffi.call/cls_5byte.c: Likewise.
2801 * testsuite/libffi.call/cls_64byte.c: Likewise.
2802 * testsuite/libffi.call/cls_6byte.c: Likewise.
2803 * testsuite/libffi.call/cls_7byte.c: Likewise.
2804 * testsuite/libffi.call/cls_8byte.c: Likewise.
2805 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2806 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2807 * testsuite/libffi.call/cls_double.c: Likewise.
2808 * testsuite/libffi.call/cls_float.c: Likewise.
2809 * testsuite/libffi.call/cls_schar.c: Likewise.
2810 * testsuite/libffi.call/cls_sint.c: Likewise.
2811 * testsuite/libffi.call/cls_sshort.c: Likewise.
2812 * testsuite/libffi.call/cls_uchar.c: Likewise.
2813 * testsuite/libffi.call/cls_uint.c: Likewise.
2814 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2815 * testsuite/libffi.call/cls_ushort.c: Likewise.
2816 * testsuite/libffi.call/nested_struct.c: Likewise.
2817 * testsuite/libffi.call/nested_struct1.c: Likewise.
2818 * testsuite/libffi.call/nested_struct2.c: Likewise.
2819 * testsuite/libffi.call/nested_struct3.c: Likewise.
2820 * testsuite/libffi.call/problem1.c: Likewise.
2821 * testsuite/libffi.special/unwindtest.cc: Likewise.
2823 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2825 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2827 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2829 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2830 Add -lgcc_s to additional flags.
2832 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2834 * configure.in, include/Makefile.am: PR libgcj/11147, install
2835 the ffitarget.h header file in a gcc versioned and target
2837 * configure: Regenerated.
2838 * Makefile.in, include/Makefile.in: Likewise.
2839 * testsuite/Makefile.in: Likewise.
2841 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2843 * testsuite/libffi.call/closure_fn0.c: Print result and check
2844 with dg-output to make debugging easier.
2845 * testsuite/libffi.call/closure_fn1.c: Likewise.
2846 * testsuite/libffi.call/closure_fn2.c: Likewise.
2847 * testsuite/libffi.call/closure_fn3.c: Likewise.
2848 * testsuite/libffi.call/closure_fn4.c: Likewise.
2849 * testsuite/libffi.call/closure_fn5.c: Likewise.
2850 * testsuite/libffi.call/cls_12byte.c: Likewise.
2851 * testsuite/libffi.call/cls_16byte.c: Likewise.
2852 * testsuite/libffi.call/cls_18byte.c: Likewise.
2853 * testsuite/libffi.call/cls_19byte.c: Likewise.
2854 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2855 * testsuite/libffi.call/cls_20byte.c: Likewise.
2856 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2857 * testsuite/libffi.call/cls_24byte.c: Likewise.
2858 * testsuite/libffi.call/cls_2byte.c: Likewise.
2859 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2860 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2861 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2862 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2863 * testsuite/libffi.call/cls_4byte.c: Likewise.
2864 * testsuite/libffi.call/cls_5byte.c: Likewise.
2865 * testsuite/libffi.call/cls_64byte.c: Likewise.
2866 * testsuite/libffi.call/cls_6byte.c: Likewise.
2867 * testsuite/libffi.call/cls_7byte.c: Likewise.
2868 * testsuite/libffi.call/cls_8byte.c: Likewise.
2869 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2870 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2871 * testsuite/libffi.call/cls_double.c: Likewise.
2872 * testsuite/libffi.call/cls_float.c: Likewise.
2873 * testsuite/libffi.call/cls_schar.c: Likewise.
2874 * testsuite/libffi.call/cls_sint.c: Likewise.
2875 * testsuite/libffi.call/cls_sshort.c: Likewise.
2876 * testsuite/libffi.call/cls_uchar.c: Likewise.
2877 * testsuite/libffi.call/cls_uint.c: Likewise.
2878 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2879 * testsuite/libffi.call/cls_ushort.c: Likewise.
2880 * testsuite/libffi.call/problem1.c: Likewise.
2882 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2885 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2887 * testsuite/libffi.call/cls_9byte2.c: New test case.
2888 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2889 * testsuite/libffi.call/cls_64byte.c: Likewise.
2890 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2891 * testsuite/libffi.call/cls_19byte.c: Likewise.
2892 * testsuite/libffi.call/cls_18byte.c: Likewise.
2893 * testsuite/libffi.call/closure_fn4.c: Likewise.
2894 * testsuite/libffi.call/closure_fn5.c: Likewise.
2895 * testsuite/libffi.call/cls_schar.c: Likewise.
2896 * testsuite/libffi.call/cls_sint.c: Likewise.
2897 * testsuite/libffi.call/cls_sshort.c: Likewise.
2898 * testsuite/libffi.call/nested_struct2.c: Likewise.
2899 * testsuite/libffi.call/nested_struct3.c: Likewise.
2901 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2903 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2904 * testsuite/libffi.call/cls_uchar.c: Likewise.
2905 * testsuite/libffi.call/cls_uint.c: Likewise.
2906 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2907 * testsuite/libffi.call/cls_ushort.c: Likewise.
2908 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2910 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2912 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2915 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2917 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2918 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2920 2003-10-22 David Daney <ddaney@avtrex.com>
2922 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2923 __attribute__((__mode__(__SI__))) and friends.
2925 2003-10-22 Andreas Schwab <schwab@suse.de>
2927 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2929 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2931 * configure.in: AC_LINK_FILES(ffitarget.h).
2932 * configure: Regenerate.
2933 * Makefile.in: Likewise.
2934 * include/Makefile.in: Likewise.
2935 * testsuite/Makefile.in: Likewise.
2936 * fficonfig.h.in: Likewise.
2938 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2939 Richard Henderson <rth@redhat.com>
2941 Avoid that ffi.h includes fficonfig.h.
2943 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2944 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2945 (TARGET_SRC_MIPS_SGI): Removed.
2946 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2947 (MIPS_SGI): Removed.
2948 (CLEANFILES): Removed.
2949 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2951 * acconfig.h: Removed.
2952 * configure.in: Compute sizeofs only for double and long double.
2953 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2954 into AC_DEFINE instead of using acconfig.h. Create
2955 include/ffitarget.h instead of include/fficonfig.h. Rename
2956 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2957 AC_DEFINE EH_FRAME_FLAGS.
2958 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2959 (hack_DATA): Add ffitarget.h.
2960 * include/ffi.h.in: Remove all system specific definitions.
2961 Declare raw API even if it is not installed, why bother?
2962 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2963 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2964 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2965 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2966 * include/ffi_common.h (bool): Do not define.
2967 (ffi_assert): Accept failed assertion.
2968 (ffi_type_test): Return void and accept file/line.
2969 (FFI_ASSERT): Pass stringized failed assertion.
2970 (FFI_ASSERT_AT): New macro.
2971 (FFI_ASSERT_VALID_TYPE): New macro.
2972 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2973 UINT64, SINT64): Define here with gcc's __attribute__ macro
2975 (FLOAT32, ALIGN): Define here instead of in ffi.h
2976 * include/ffi-mips.h: Removed. Its content moved to
2977 src/mips/ffitarget.h after separating assembly and C sections.
2978 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2979 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2980 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2981 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2982 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2983 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2984 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2985 (ffi_type_test): Rewritten.
2986 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2987 FFI_ASSERT_VALID_TYPE.
2988 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2989 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2990 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2991 src/s390/ffitarget.h, src/sh/ffitarget.h,
2992 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2993 src/x86/ffitarget.h: New files.
2994 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2995 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2996 src/powerpc/aix.S, src/powerpc/darwin.S,
2997 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2998 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2999 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
3000 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
3001 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
3004 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3006 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
3007 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
3009 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
3011 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
3012 Used when FFI_DEBUG = 1.
3014 2003-10-14 Alan Modra <amodra@bigpond.net.au>
3016 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
3019 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3021 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
3022 FFI_MIPS_O32 for O32 ABI.
3024 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
3026 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
3027 SPARC64. Cleanup whitespaces.
3029 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
3031 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
3032 strongarm, xscale. Cleanup whitespaces.
3033 * testsuite/libffi.call/closure_fn1.c: Likewise.
3034 * testsuite/libffi.call/closure_fn2.c: Likewise.
3035 * testsuite/libffi.call/closure_fn3.c: Likewise.
3036 * testsuite/libffi.call/cls_12byte.c: Likewise.
3037 * testsuite/libffi.call/cls_16byte.c: Likewise.
3038 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3039 * testsuite/libffi.call/cls_20byte.c: Likewise.
3040 * testsuite/libffi.call/cls_24byte.c: Likewise.
3041 * testsuite/libffi.call/cls_2byte.c: Likewise.
3042 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3043 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3044 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3045 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3046 * testsuite/libffi.call/cls_4byte.c: Likewise.
3047 * testsuite/libffi.call/cls_5byte.c: Likewise.
3048 * testsuite/libffi.call/cls_6byte.c: Likewise.
3049 * testsuite/libffi.call/cls_7byte.c: Likewise.
3050 * testsuite/libffi.call/cls_8byte.c: Likewise.
3051 * testsuite/libffi.call/cls_double.c: Likewise.
3052 * testsuite/libffi.call/cls_float.c: Likewise.
3053 * testsuite/libffi.call/cls_uchar.c: Likewise.
3054 * testsuite/libffi.call/cls_uint.c: Likewise.
3055 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3056 * testsuite/libffi.call/cls_ushort.c: Likewise.
3057 * testsuite/libffi.call/nested_struct.c: Likewise.
3058 * testsuite/libffi.call/nested_struct1.c: Likewise.
3059 * testsuite/libffi.call/problem1.c: Likewise.
3060 * testsuite/libffi.special/unwindtest.cc: Likewise.
3061 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
3063 2003-09-18 David Edelsohn <edelsohn@gnu.org>
3065 * src/powerpc/aix.S: Cleanup whitespaces.
3066 * src/powerpc/aix_closure.S: Likewise.
3068 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3070 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
3071 * src/powerpc/darwin_closure.S: Likewise.
3072 * src/powerpc/ffi_darwin.c: Likewise.
3074 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3075 David Edelsohn <edelsohn@gnu.org>
3077 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
3078 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
3080 * src/powerpc/darwin_closure.S: Likewise.
3081 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
3082 according to the Darwin/AIX ABI.
3083 (ffi_prep_cif_machdep): Likewise.
3084 (ffi_closure_helper_DARWIN): Likewise.
3085 Remove the outgoing parameter stack logic. Simplify the evaluation
3086 of the different CASE types.
3087 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
3088 statement in the trampoline code.
3090 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
3092 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
3093 for the register size.
3094 (ffi_closure_helper_SYSV): Handle the structure return value
3096 (ffi_closure_helper_SYSV): Return the appropriate type when
3097 the registers are used for the structure return value.
3098 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
3099 the 64-bit return value. Update copyright years.
3101 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3103 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
3104 srcdir for ffi_mips.h.
3106 2003-09-12 Alan Modra <amodra@bigpond.net.au>
3108 * src/prep_cif.c (initialize_aggregate): Include tail padding in
3110 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
3111 placement of float result.
3112 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
3113 cast of "resp" for big-endian 64 bit machines.
3115 2003-09-11 Alan Modra <amodra@bigpond.net.au>
3117 * src/types.c (double, longdouble): Merge identical SH and ARM
3118 typedefs, and add POWERPC64.
3119 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
3120 struct split over gpr and rest.
3121 (ffi_prep_cif_machdep): Correct intarg_count for structures.
3122 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
3124 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
3126 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
3129 2003-09-09 Alan Modra <amodra@bigpond.net.au>
3131 * configure: Regenerate.
3133 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3135 * Makefile.am: Remove build rules for ffitest.
3136 * Makefile.in: Rebuilt.
3138 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3140 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3141 about implicit declaration of abort().
3143 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3145 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3146 * Makefile.in: Rebuilt.
3147 * configure.in: Add dejagnu test framework.
3148 * configure: Rebuilt.
3150 * testsuite/Makefile.am: New file.
3151 * testsuite/Makefile.in: Built
3152 * testsuite/lib/libffi-dg.exp: New file.
3153 * testsuite/config/default.exp: Likewise.
3154 * testsuite/libffi.call/call.exp: Likewise.
3155 * testsuite/libffi.call/ffitest.h: Likewise.
3156 * testsuite/libffi.call/closure_fn0.c: Likewise.
3157 * testsuite/libffi.call/closure_fn1.c: Likewise.
3158 * testsuite/libffi.call/closure_fn2.c: Likewise.
3159 * testsuite/libffi.call/closure_fn3.c: Likewise.
3160 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3161 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3162 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3163 * testsuite/libffi.call/cls_2byte.c: Likewise.
3164 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3165 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3166 * testsuite/libffi.call/cls_4byte.c: Likewise.
3167 * testsuite/libffi.call/cls_5byte.c: Likewise.
3168 * testsuite/libffi.call/cls_6byte.c: Likewise.
3169 * testsuite/libffi.call/cls_7byte.c: Likewise.
3170 * testsuite/libffi.call/cls_8byte.c: Likewise.
3171 * testsuite/libffi.call/cls_12byte.c: Likewise.
3172 * testsuite/libffi.call/cls_16byte.c: Likewise.
3173 * testsuite/libffi.call/cls_20byte.c: Likewise.
3174 * testsuite/libffi.call/cls_24byte.c: Likewise.
3175 * testsuite/libffi.call/cls_double.c: Likewise.
3176 * testsuite/libffi.call/cls_float.c: Likewise.
3177 * testsuite/libffi.call/cls_uchar.c: Likewise.
3178 * testsuite/libffi.call/cls_uint.c: Likewise.
3179 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3180 * testsuite/libffi.call/cls_ushort.c: Likewise.
3181 * testsuite/libffi.call/float.c: Likewise.
3182 * testsuite/libffi.call/float1.c: Likewise.
3183 * testsuite/libffi.call/float2.c: Likewise.
3184 * testsuite/libffi.call/many.c: Likewise.
3185 * testsuite/libffi.call/many_win32.c: Likewise.
3186 * testsuite/libffi.call/nested_struct.c: Likewise.
3187 * testsuite/libffi.call/nested_struct1.c: Likewise.
3188 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3189 * testsuite/libffi.call/problem1.c: Likewise.
3190 * testsuite/libffi.call/promotion.c: Likewise.
3191 * testsuite/libffi.call/return_ll.c: Likewise.
3192 * testsuite/libffi.call/return_sc.c: Likewise.
3193 * testsuite/libffi.call/return_uc.c: Likewise.
3194 * testsuite/libffi.call/strlen.c: Likewise.
3195 * testsuite/libffi.call/strlen_win32.c: Likewise.
3196 * testsuite/libffi.call/struct1.c: Likewise.
3197 * testsuite/libffi.call/struct2.c: Likewise.
3198 * testsuite/libffi.call/struct3.c: Likewise.
3199 * testsuite/libffi.call/struct4.c: Likewise.
3200 * testsuite/libffi.call/struct5.c: Likewise.
3201 * testsuite/libffi.call/struct6.c: Likewise.
3202 * testsuite/libffi.call/struct7.c: Likewise.
3203 * testsuite/libffi.call/struct8.c: Likewise.
3204 * testsuite/libffi.call/struct9.c: Likewise.
3205 * testsuite/libffi.special/special.exp: New file.
3206 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3207 * testsuite/libffi.special/unwindtest.cc: Likewise.
3210 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3212 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3215 2003-08-02 Alan Modra <amodra@bigpond.net.au>
3217 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3219 (ffi_closure_helper_LINUX64): Likewise.
3220 * src/powerpc/linux64.S: Remove code writing to parm save area.
3221 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3222 address in lr from ffi_closure_helper_LINUX64 call to calculate
3223 table address. Optimize function tail.
3225 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3227 * src/sparc/ffi.c: Handle all floating point registers.
3228 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3230 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3232 * README: Note that libffi is not part of GCC. Update the project
3235 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3237 * src/powerpc/ppc_closure.S: Include ffi.h.
3239 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3241 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3242 Use C style comments.
3244 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3246 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3247 * Makefile.in: Regenerate.
3248 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3249 * configure: Regenerate.
3250 * include/ffi.h.in: Add SHmedia support.
3251 * src/sh64/ffi.c: New file.
3252 * src/sh64/sysv.S: New file.
3254 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3256 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3257 should be read-only.
3258 * configure: Rebuilt.
3259 * fficonfig.h.in: Rebuilt.
3260 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3261 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3262 * src/powerpc/linux64.S: Likewise.
3263 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3264 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3265 if -fpic/-fPIC/-mrelocatable.
3266 * src/powerpc/powerpc_closure.S: Likewise.
3267 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3268 #write in .eh_frame flags.
3269 * src/sparc/v9.S: Likewise.
3270 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3271 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3272 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3274 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3276 Fixes PR bootstrap/10656
3277 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3278 support for .register pseudo-op.
3279 * src/sparc/v8.S: Use it.
3280 * fficonfig.h.in: Rebuilt.
3281 * configure: Rebuilt.
3283 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3285 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3286 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3287 Make it the default on POWERPC64.
3288 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3289 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3290 * configure: Rebuilt.
3291 * src/powerpc/ffi.c (hidden): Define.
3292 (ffi_prep_args_SYSV): Renamed from
3293 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3294 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3295 ASM_NEEDS_REGISTERS64): New.
3296 (ffi_prep_args64): New function.
3297 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3298 (ffi_call): Likewise.
3299 (ffi_prep_closure): Likewise.
3300 (flush_icache): Surround by #ifndef POWERPC64.
3301 (ffi_dblfl): New union type.
3302 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3303 (ffi_closure_helper_LINUX64): New function.
3304 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3306 * src/powerpc/sysv.S: Likewise.
3307 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3308 * src/powerpc/linux64.S: New file.
3309 * src/powerpc/linux64_closure.S: New file.
3310 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3311 src/powerpc/linux64_closure.S.
3312 (TARGET_SRC_POWERPC): Likewise.
3314 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3315 closure_test_fn3): Fix result printing on big-endian 64-bit
3317 (main): Print tst2_arg instead of uninitialized tst2_result.
3319 * src/ffitest.c (main): Hide what closure pointer really points to
3322 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3324 * configure.in (arm-*-netbsdelf*): Add configuration.
3325 (configure): Regenerated.
3327 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3329 * include/Makefile.in: Regenerate.
3331 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3333 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3335 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3336 Receive closure pointer through parameter, read args using
3337 __builtin_dwarf_cfa.
3338 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3340 2003-03-12 Andreas Schwab <schwab@suse.de>
3342 * configure.in: Avoid trailing /. in toolexeclibdir.
3343 * configure: Rebuilt.
3345 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3347 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3349 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3351 * libffi/src/powerpc/darwin_closure.S:
3352 Fix alignement bug, allocate 8 bytes for the result.
3353 * libffi/src/powerpc/aix_closure.S:
3355 * libffi/src/powerpc/ffi_darwin.c:
3356 Update stackframe description for aix/darwin_closure.S.
3358 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3360 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3363 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3364 Andreas Schwab <schwab@suse.de>
3366 * configure.in: Adjust command to source config-ml.in to account
3367 for changes to the libffi_basedir definition.
3368 (libffi_basedir): Remove ${srcdir} from value and include trailing
3371 * configure: Regenerate.
3373 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3375 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3377 2003-01-28 Andrew Haley <aph@redhat.com>
3379 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3380 * src/x86/ffi64.c (ffi_prep_closure): New.
3381 (ffi_closure_UNIX64_inner): New.
3382 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3384 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
3386 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3387 Remove USE_LIBDIR conditional.
3388 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3389 * Makefile.in, configure: Rebuilt.
3391 2003-01027 David Edelsohn <edelsohn@gnu.org>
3393 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3394 * Makefile.in: Regenerate.
3396 2003-01-22 Andrew Haley <aph@redhat.com>
3398 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3401 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3403 * src/powerpc/darwin.S: Add unwind info.
3404 * src/powerpc/darwin_closure.S: Likewise.
3406 2003-01-14 Andrew Haley <aph@redhat.com>
3408 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3409 (ffi_prep_cif_machdep): Likewise.
3410 * src/x86/unix64.S: Add unwind info.
3412 2003-01-14 Andreas Jaeger <aj@suse.de>
3414 * src/ffitest.c (main): Only use ffi_closures if those are
3417 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3419 * libffi/src/ffitest.c
3420 add closure testcases
3422 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3424 * libffi/src/powerpc/ffi.c
3425 fix alignment bug for float (4 byte aligned iso 8 byte)
3427 2003-01-09 Geoffrey Keating <geoffk@apple.com>
3429 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3430 * src/powerpc/darwin.S: Remove RCS version string.
3432 2003-01-03 Jeff Sturm <jsturm@one-point.com>
3434 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3435 * src/ffitest.c (main): Use static storage for closure.
3436 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3437 * src/sparc/v8.S (ffi_closure_v8): New.
3438 * src/sparc/v9.S (ffi_closure_v9): New.
3440 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3442 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3443 enumeration for X86_WIN32.
3444 * src/x86/win32.S: Added ffi_call_STDCALL function
3446 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3447 switch cases for recognising FFI_STDCALL and
3448 calling ffi_call_STDCALL if target is X86_WIN32.
3449 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3450 stdcall versions of the "my_strlen" and "many"
3451 test functions (for X86_WIN32).
3452 Added test cases to test stdcall invocation using
3455 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3457 * src/sh/sysv.S: Add DWARF2 unwind info.
3459 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3461 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3463 2002-11-26 Jim Wilson <wilson@redhat.com>
3465 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3467 2002-11-23 H.J. Lu <hjl@gnu.org>
3469 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3470 Include ../config/accross.m4.
3471 * aclocal.m4; Rebuild.
3472 * configure: Likewise.
3474 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3476 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3478 2002-11-11 DJ Delorie <dj@redhat.com>
3480 * configure.in: Look for common files in the right place.
3482 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3484 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3485 raw data as _Jv_word values, not ffi_raw.
3486 (ffi_java_ptrarray_to_raw): Likewise.
3487 (ffi_java_rvalue_to_raw): New function.
3488 (ffi_java_raw_call): Call it.
3489 (ffi_java_raw_to_rvalue): New function.
3490 (ffi_java_translate_args): Call it.
3491 * src/ffitest.c (closure_test_fn): Interpret return value
3492 as ffi_arg, not int.
3493 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3494 FFI_TYPE_POINTER case.
3495 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3496 values extended to word size.
3498 2002-10-02 Andreas Jaeger <aj@suse.de>
3500 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3502 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
3504 * include/ffi.h.in: Fix i386 win32 compilation.
3506 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3508 * configure.in: Add s390x-*-linux-* target.
3509 * configure: Regenerate.
3510 * include/ffi.h.in: Define S390X for s390x targets.
3511 (FFI_CLOSURES): Define for s390/s390x.
3512 (FFI_TRAMPOLINE_SIZE): Likewise.
3513 (FFI_NATIVE_RAW_API): Likewise.
3514 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3515 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
3516 * src/s390/ffi.c: Major rework of existing code. Add support for
3517 s390x targets. Add closure support.
3518 * src/s390/sysv.S: Likewise.
3520 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
3522 * src/arm/sysv.S: Fix typo.
3524 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
3526 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3527 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3528 (ffi_call_SYSV): Handle soft-float.
3530 2002-09-27 Bo Thorsen <bo@suse.de>
3532 * include/ffi.h.in: Fix multilib x86-64 support.
3534 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3536 * Makefile.am (all-multi): Fix multilib parallel build.
3538 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3540 * configure.in (sh[34]*-*-linux*): Add brackets.
3541 * configure: Regenerate.
3543 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3545 * Makefile.am: Add SH support.
3546 * Makefile.in: Regenerate.
3547 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3548 * configure: Regenerate.
3549 * include/ffi.h.in: Add SH support.
3550 * src/sh/ffi.c: New file.
3551 * src/sh/sysv.S: New file.
3552 * src/types.c: Add SH support.
3554 2002-07-16 Bo Thorsen <bo@suse.de>
3556 * src/x86/ffi64.c: New file that adds x86-64 support.
3557 * src/x86/unix64.S: New file that handles argument setup for
3559 * src/x86/sysv.S: Don't use this on x86-64.
3560 * src/x86/ffi.c: Don't use this on x86-64.
3562 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3564 * src/ffitest.c (struct6): New test that tests a special case in
3566 (struct7): Likewise.
3567 (struct8): Likewise.
3568 (struct9): Likewise.
3569 (closure_test_fn): Silence warning about this when it's not used.
3570 (main): Add the new tests.
3571 (main): Fix a couple of wrong casts and silence some compiler warnings.
3572 * include/ffi.h.in: Add x86-64 ABI definition.
3573 * fficonfig.h.in: Regenerate.
3574 * Makefile.am: Add x86-64 support.
3575 * configure.in: Likewise.
3576 * Makefile.in: Regenerate.
3577 * configure: Likewise.
3579 2002-06-24 Bo Thorsen <bo@suse.de>
3581 * src/types.c: Merge settings for similar architectures.
3582 Add x86-64 sizes and alignments.
3584 2002-06-23 Bo Thorsen <bo@suse.de>
3586 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3587 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3588 * src/mips/ffi.c (ffi_prep_args): Likewise.
3589 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3591 2002-07-18 H.J. Lu (hjl@gnu.org)
3593 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3594 (libffi_la_SOURCES): Support MIPS_LINUX.
3595 (libffi_convenience_la_SOURCES): Likewise.
3596 * Makefile.in: Regenerated.
3598 * configure.in (mips64*-*): Skip.
3599 (mips*-*-linux*): New.
3600 * configure: Regenerated.
3602 * src/mips/ffi.c: Include <sgidefs.h>.
3604 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3606 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3608 2002-05-27 Roger Sayle <roger@eyesopen.com>
3610 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3612 2002-05-27 Bo Thorsen <bo@suse.de>
3614 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3617 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3619 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3620 beginning of function (for older apple cc).
3622 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3624 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3625 script entry, and set LD to it when configuring multilibs.
3626 * configure: Rebuilt.
3628 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3630 * configure.in (sparc64-*-netbsd*): Add target.
3631 (sparc-*-netbsdelf*): Likewise.
3632 * configure: Regenerate.
3634 2002-04-28 David S. Miller <davem@redhat.com>
3636 * configure.in, configure: Fix SPARC test in previous change.
3638 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3640 * Makefile.am: Add Linux for S/390 support.
3641 * Makefile.in: Regenerate.
3642 * configure.in: Add Linux for S/390 support.
3643 * configure: Regenerate.
3644 * include/ffi.h.in: Add Linux for S/390 support.
3645 * src/s390/ffi.c: New file from libffi CVS tree.
3646 * src/s390/sysv.S: New file from libffi CVS tree.
3648 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3650 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3652 * src/sparc/v8.S: Use it.
3653 * src/sparc/v9.S: Likewise.
3654 * fficonfig.h.in: Rebuilt.
3655 * configure: Rebuilt.
3657 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3659 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3661 * src/ia64/unix.S: Add unwind information. Fix comments.
3662 Save sp in a way that's compatible with unwind info.
3663 (ffi_call_unix): Correctly restore sp in all cases.
3664 * src/ia64/ffi.c: Add, fix comments.
3666 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3668 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3670 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3672 * configure.in (alpha*-*-netbsd*): Add target.
3673 * configure: Regenerate.
3675 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3677 * src/sparc/v8.S: Add unwind info.
3678 * src/sparc/v9.S: Likewise.
3680 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3682 * configure.in: Enable i*86-*-netbsdelf*.
3683 * configure: Rebuilt.
3685 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3688 * src/mips/n32.s: Delete
3689 * src/mips/o32.s: Delete
3691 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3693 * configure.in: Enable alpha*-*-freebsd*.
3694 * configure: Rebuilt.
3696 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3698 * Makefile.am: libfficonvenience -> libffi_convenience.
3699 * Makefile.in: Rebuilt.
3701 * Makefile.am: Define ffitest_OBJECTS.
3702 * Makefile.in: Rebuilt.
3704 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3705 David Edelsohn <edelsohn@gnu.org>
3707 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3708 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3709 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3710 * Makefile.in: Regenerate.
3711 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3712 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3713 (flush_icache, flush_range): New functions.
3714 (ffi_closure_helper_DARWIN): New function.
3715 * src/powerpc/aix_closure.S: New file.
3716 * src/powerpc/darwin_closure.S: New file.
3718 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3720 * include/ffi.h.in: Add typedef for ffi_arg.
3721 * src/ffitest.c (main): Declare rint with ffi_arg.
3723 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3725 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3726 number of GPRs for floating-point arguments.
3728 2002-01-31 Anthony Green <green@redhat.com>
3730 * configure: Rebuilt.
3731 * configure.in: Replace CHECK_SIZEOF and endian tests with
3732 cross-compiler friendly macros.
3733 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3736 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3738 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3739 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3741 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3743 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3744 (TARGET_SRC_POWERPC_AIX): New.
3745 (POWERPC_AIX): New stanza.
3746 * Makefile.in: Regenerate.
3747 * configure.in: Add AIX case.
3748 * configure: Regenerate.
3749 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3750 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3751 size. Fix "long double" support.
3752 (ffi_call): Add FFI_AIX case.
3753 * src/powerpc/aix.S: New.
3755 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3757 Implement Darwin PowerPC ABI.
3758 * configure.in: Handle powerpc-*-darwin*.
3759 * Makefile.am: Set source files for POWERPC_DARWIN.
3760 * configure: Rebuilt.
3761 * Makefile.in: Rebuilt.
3762 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3764 * src/powerpc/darwin.S: New file.
3765 * src/powerpc/ffi_darwin.c: New file.
3767 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3769 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3771 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3773 * src/x86/sysv.S: Avoid gas-only .balign directive.
3774 Use C style comments.
3776 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3778 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3779 Fixes PR bootstrap/3563.
3781 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3783 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3785 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3787 * configure.in: Recognize sparc*-sun-* host.
3788 * configure: Regenerate.
3790 2001-06-06 Andrew Haley <aph@redhat.com>
3792 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
3794 2001-06-03 Andrew Haley <aph@redhat.com>
3796 * src/alpha/osf.S: Add unwind info.
3797 * src/powerpc/sysv.S: Add unwind info.
3798 * src/powerpc/ppc_closure.S: Likewise.
3800 2000-05-31 Jeff Sturm <jsturm@one-point.com>
3802 * configure.in: Fix AC_ARG_ENABLE usage.
3803 * configure: Rebuilt.
3805 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3807 * configure.in: Remove warning about beta code.
3808 * configure: Rebuilt.
3810 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3812 * src/ia64/unix.S: Restore stack pointer when returning from
3814 * src/ia64/ffi.c: Fix typo in comment.
3816 2001-04-18 Jim Wilson <wilson@redhat.com>
3818 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3819 to eliminate RAW DV.
3821 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3823 * Makefile.am: Make a libtool convenience library.
3824 * Makefile.in: Rebuilt.
3826 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3828 * configure.in: Use different syntax for subdirectory creation.
3829 * configure: Rebuilt.
3831 2001-03-27 Jon Beniston <jon@beniston.com>
3833 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3834 * configure: Rebuilt.
3835 * Makefile.am: Added X86_WIN32 target support.
3836 * Makefile.in: Rebuilt.
3838 * include/ffi.h.in: Added X86_WIN32 target support.
3840 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3841 * src/types.c: Added X86_WIN32 target support.
3843 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3844 stuff removed and made to work with CygWin's gas.
3846 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3848 * configure.in: Make target subdirectory in build dir.
3849 * Makefile.am: Override suffix based rules to specify correct output
3851 * Makefile.in: Rebuilt.
3852 * configure: Rebuilt.
3854 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3856 * src/powerpc/ppc_closure.S: New file.
3857 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3858 involving long long and register pairs.
3859 (ffi_prep_closure): New function.
3860 (flush_icache): Likewise.
3861 (ffi_closure_helper_SYSV): Likewise.
3862 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3863 (FFI_TRAMPOLINE_SIZE): Likewise.
3864 (FFI_NATIVE_RAW_API): Likewise.
3865 * Makefile.in: Rebuilt.
3866 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3867 (TARGET_SRC_POWERPC): Likewise.
3869 2001-03-19 Tom Tromey <tromey@redhat.com>
3871 * Makefile.in: Rebuilt.
3872 * Makefile.am (ffitest_LDFLAGS): New macro.
3874 2001-03-02 Nick Clifton <nickc@redhat.com>
3876 * include/ffi.h.in: Remove RCS ident string.
3877 * include/ffi_mips.h: Remove RCS ident string.
3878 * src/debug.c: Remove RCS ident string.
3879 * src/ffitest.c: Remove RCS ident string.
3880 * src/prep_cif.c: Remove RCS ident string.
3881 * src/types.c: Remove RCS ident string.
3882 * src/alpha/ffi.c: Remove RCS ident string.
3883 * src/alpha/osf.S: Remove RCS ident string.
3884 * src/arm/ffi.c: Remove RCS ident string.
3885 * src/arm/sysv.S: Remove RCS ident string.
3886 * src/mips/ffi.c: Remove RCS ident string.
3887 * src/mips/n32.S: Remove RCS ident string.
3888 * src/mips/o32.S: Remove RCS ident string.
3889 * src/sparc/ffi.c: Remove RCS ident string.
3890 * src/sparc/v8.S: Remove RCS ident string.
3891 * src/sparc/v9.S: Remove RCS ident string.
3892 * src/x86/ffi.c: Remove RCS ident string.
3893 * src/x86/sysv.S: Remove RCS ident string.
3895 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3897 * include/ffi.h.in: Change sourceware.cygnus.com references to
3900 2000-12-09 Richard Henderson <rth@redhat.com>
3902 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3903 (ffi_closure_osf_inner): Index rather than increment avalue
3904 and arg_types. Give ffi_closure_osf the raw return value type.
3905 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3908 2000-12-07 Richard Henderson <rth@redhat.com>
3910 * src/raw_api.c (ffi_translate_args): Fix typo.
3911 (ffi_prep_closure): Likewise.
3913 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3914 FFI_TRAMPOLINE_SIZE.
3915 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3916 cif->bytes for new ffi_call_osf implementation.
3917 (ffi_prep_args): Absorb into ...
3918 (ffi_call): ... here. Do all stack allocation here and
3919 avoid a callback function.
3920 (ffi_prep_closure, ffi_closure_osf_inner): New.
3921 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3922 (ffi_closure_osf): New.
3924 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3926 * config.guess, config.sub, install-sh: Removed.
3927 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3928 * Makefile.in: Rebuilt.
3930 * acinclude.m4: Include libtool macros from the top level.
3931 * aclocal.m4, configure: Rebuilt.
3933 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3935 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3936 * configure: Rebuilt.
3938 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3940 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3941 memory correctly. Use conditional instructions, not branches where
3944 2000-05-04 Tom Tromey <tromey@cygnus.com>
3946 * configure: Rebuilt.
3947 * configure.in: Match `arm*-*-linux-*'.
3948 From Chris Dornan <cdornan@arm.com>.
3950 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3952 * Makefile.am (SUBDIRS): Define.
3953 (AM_MAKEFLAGS): Likewise.
3954 (Multilib support.): Add section.
3955 * Makefile.in: Rebuilt.
3956 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3957 New variables. Set for gcc using -print-multi-lib. Export them
3959 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3960 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3962 (extra_compiler_flags_value, extra_compiler_flags): Check these
3963 for extra compiler options which need to be passed down in
3966 2000-04-16 Anthony Green <green@redhat.com>
3968 * configure: Rebuilt.
3969 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3971 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3973 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3974 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3975 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3976 Replace all void * sizeofs with sizeof(int).
3977 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3978 different than DOUBLE.
3979 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3980 (ffi_prep_args_v9): New function.
3981 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3982 (ffi_V9_return_struct): New function.
3983 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3984 32bit code (not yet cross-arch calls).
3985 * src/sparc/v8.S: Add struct return delay nop.
3987 * src/sparc/v9.S: New file.
3988 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3989 is used on sparc64 only for structures larger than 32 bytes.
3990 Pass by reference for structures is done for structure arguments
3991 larger than 16 bytes.
3992 * src/ffitest.c (main): Use 64bit rint on sparc64.
3993 Run long long tests on sparc.
3994 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3996 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3998 * configure.in (sparc-*-linux*): New supported target.
3999 (sparc64-*-linux*): Likewise.
4000 * configure: Rebuilt.
4001 * Makefile.am: Add v9.S to SPARC files.
4002 * Makefile.in: Likewise.
4003 (LINK): Surround $(CCLD) into double quotes, so that multilib
4004 compiles work correctly.
4006 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
4008 * configure: Rebuilt.
4009 * configure.in: (i*86-*-solaris*): New libffi target. Patch
4010 proposed by Bryce McKinlay.
4012 2000-03-20 Tom Tromey <tromey@cygnus.com>
4014 * Makefile.in: Hand edit for java_raw_api.lo.
4016 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
4018 * config.guess, config.sub: Update from the gcc tree.
4019 Fix for PR libgcj/168.
4021 2000-03-03 Tom Tromey <tromey@cygnus.com>
4023 * Makefile.in: Fixed ia64 by hand.
4025 * configure: Rebuilt.
4026 * configure.in (--enable-multilib): New option.
4027 (libffi_basedir): New subst.
4028 (AC_OUTPUT): Added multilib code.
4030 2000-03-02 Tom Tromey <tromey@cygnus.com>
4032 * Makefile.in: Rebuilt.
4033 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
4036 2000-02-25 Hans Boehm <boehm@acm.org>
4038 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
4040 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
4042 (ffi_prep_raw_closure): Use ffi_translate_args, not
4043 ffi_closure_translate.
4044 * src/java_raw_api.c: New file.
4045 * src/ffitest.c (closure_test_fn): New function.
4046 (main): Define `rint' as long long on IA64. Added new test when
4047 FFI_CLOSURES is defined.
4048 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
4049 (ffi_abi): Recognize IA64.
4050 (ffi_raw): Added `flt' field.
4051 Added "Java raw API" code.
4052 * configure.in: Recognize ia64.
4053 * Makefile.am (TARGET_SRC_IA64): New macro.
4054 (libffi_la_common_SOURCES): Added java_raw_api.c.
4055 (libffi_la_SOURCES): Define in IA64 case.
4057 2000-01-04 Tom Tromey <tromey@cygnus.com>
4059 * Makefile.in: Rebuilt with newer automake.
4061 1999-12-31 Tom Tromey <tromey@cygnus.com>
4063 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
4065 1999-09-01 Tom Tromey <tromey@cygnus.com>
4067 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
4069 * fficonfig.h.in: Rebuilt.
4070 * configure: Rebuilt.
4071 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
4072 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
4073 * acconfig.h: Don't #undef PACKAGE or VERSION.
4075 1999-08-09 Anthony Green <green@cygnus.com>
4077 * include/ffi.h.in: Try to work around messy header problem
4078 with PACKAGE and VERSION.
4080 * configure: Rebuilt.
4081 * configure.in: Change version to 2.00-beta.
4083 * fficonfig.h.in: Rebuilt.
4084 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
4086 * src/x86/ffi.c (ffi_raw_call): Rename.
4088 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
4090 * src/x86/ffi.c (ffi_closure_SYSV): New function.
4091 (ffi_prep_incoming_args_SYSV): Ditto.
4092 (ffi_prep_closure): Ditto.
4093 (ffi_closure_raw_SYSV): Ditto.
4094 (ffi_prep_raw_closure): More ditto.
4095 (ffi_call_raw): Final ditto.
4097 * include/ffi.h.in: Add definitions for closure and raw API.
4099 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4102 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
4104 * src/raw_api.c: New file.
4106 * include/ffi.h.in (ffi_raw): New type.
4107 (UINT_ARG, SINT_ARG): New defines.
4108 (ffi_closure, ffi_raw_closure): New types.
4109 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4111 * configure.in: Add check for endianness and sizeof void*.
4113 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
4114 instead of directly.
4116 * configure: Rebuilt.
4118 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
4120 * configure.in: Add x86 and powerpc BeOS configurations.
4121 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
4123 1999-05-09 Anthony Green <green@cygnus.com>
4125 * configure.in: Add warning about this being beta code.
4126 Remove src/Makefile.am from the picture.
4127 * configure: Rebuilt.
4129 * Makefile.am: Move logic from src/Makefile.am. Add changes
4130 to support libffi as a target library.
4131 * Makefile.in: Rebuilt.
4133 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
4134 Upgraded to new autoconf, automake, libtool.
4138 * LICENSE: Update copyright date.
4140 * src/Makefile.am, src/Makefile.in: Removed.
4142 1998-11-29 Anthony Green <green@cygnus.com>
4144 * include/ChangeLog: Removed.
4145 * src/ChangeLog: Removed.
4146 * src/mips/ChangeLog: Removed.
4147 * src/sparc/ChangeLog: Remboved.
4148 * src/x86/ChangeLog: Removed.
4150 * ChangeLog.v1: Created.