1 2009-12-26 Anthony Green <green@redhat.com>
3 * testsuite/libffi.call/cls_double_va.c,
4 testsuite/libffi.call/cls_longdouble.c,
5 testsuite/libffi.call/cls_longdouble_va.c,
6 testsuite/libffi.call/cls_pointer.c,
7 testsuite/libffi.call/cls_pointer_stack.c: Remove xfail for
8 mips*-*-* and arm*-*-*.
9 * testsuite/libffi.call/cls_align_longdouble_split.c,
10 testsuite/libffi.call/cls_align_longdouble_split2.c,
11 testsuite/libffi.call/stret_medium2.c,
12 testsuite/libffi.call/stret_medium.c,
13 testsuite/libffi.call/stret_large.c,
14 testsuite/libffi.call/stret_large2.c: Remove xfail for arm*-*-*.
16 2009-12-31 Kay Tietz <ktietz70@googlemail.com>
18 * testsuite/libffi.call/ffitest.h,
19 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRuLL): Fix
22 2009-12-31 Carlo Bramini <carlo.bramix@libero.it>
24 * configure.ac (AM_LTLDFLAGS): Define for windows hosts.
25 * Makefile.am (libffi_la_LDFLAGS): Add AM_LTLDFLAGS.
27 * Makefile.in: Rebuilt.
29 2009-12-31 Anthony Green <green@redhat.com>
32 * testsuite/libffi.call/huge_struct.c: New test case from Blake
35 2009-12-28 David Edelsohn <edelsohn@gnu.org>
37 * src/powerpc/ffi_darwin.c (ffi_prep_args): Copy abi and nargs to
39 (aix_adjust_aggregate_sizes): New function.
40 (ffi_prep_cif_machdep): Call it.
42 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
44 * configure.ac: Define FFI_MMAP_EXEC_WRIT for the given targets.
45 * configure: Regenerate.
46 * fficonfig.h.in: Likewise.
47 * src/closures.c: Remove the FFI_MMAP_EXEC_WRIT definition for
50 2009-12-26 Andreas Schwab <schwab@linux-m68k.org>
52 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Advance intarg_count
53 when a float arguments is passed in memory.
54 (ffi_closure_helper_SYSV): Mark general registers as used up when
55 a 64bit or soft-float long double argument is passed in memory.
57 2009-12-25 Matthias Klose <doko@ubuntu.com>
59 * man/ffi_call.3: Fix #include in examples.
60 * doc/libffi.texi: Add dircategory.
62 2009-12-25 Frank Everdij <f.p.x.everdij@tudelft.nl>
64 * include/ffi.h.in: Placed '__GNUC__' ifdef around
65 '__attribute__((aligned(8)))' in ffi_closure, fixes compile for
67 * include/ffi_common.h: Added '__sgi' define to non
68 '__attribute__((__mode__()))' integer typedefs.
69 * src/mips/ffi.c (ffi_call, ffi_closure_mips_inner_O32,
70 ffi_closure_mips_inner_N32): Added 'defined(_MIPSEB)' to BE check.
71 (ffi_closure_mips_inner_O32, ffi_closure_mips_inner_N32): Added
72 FFI_LONGDOUBLE support and alignment(N32 only).
73 * src/mips/ffitarget.h: Corrected '#include <sgidefs.h>' for IRIX and
74 fixed non '__attribute__((__mode__()))' integer typedefs.
75 * src/mips/n32.S: Put '#ifdef linux' around '.abicalls' and '.eh_frame'
76 since they are Linux/GNU Assembler specific.
78 2009-12-25 Bradley Smith <brad@brad-smith.co.uk>
80 * configure.ac, Makefile.am, src/avr32/ffi.c,
81 src/avr32/ffitarget.h,
82 src/avr32/sysv.S: Add AVR32 port.
83 * configure, Makefile.in: Rebuilt.
85 2009-12-21 Andreas Tobler <a.tobler@schweiz.org>
87 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
88 * configure: Regenerate.
90 2009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
92 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
94 2009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
96 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
99 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
101 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
104 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
106 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
108 2009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
111 * src/closures.c [X86_64 && __sun__ && __svr4__]
112 (FFI_MMAP_EXEC_WRIT): Define.
114 2009-12-08 David Daney <ddaney@caviumnetworks.com>
116 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
117 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
118 * testsuite/libffi.call/stret_large.c: Same.
119 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
120 * testsuite/libffi.call/stret_large2.c: Same.
121 * testsuite/libffi.call/stret_medium2.c: Same.
123 2009-12-07 David Edelsohn <edelsohn@gnu.org>
125 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
128 2009-12-05 David Edelsohn <edelsohn@gnu.org>
130 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
132 * src/powerpc/aix_closure.S: Same.
134 2009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
136 * Makefile.in: Regenerate.
137 * configure: Regenerate.
138 * include/Makefile.in: Regenerate.
139 * man/Makefile.in: Regenerate.
140 * testsuite/Makefile.in: Regenerate.
142 2009-12-04 David Edelsohn <edelsohn@gnu.org>
144 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
147 2009-12-04 Uros Bizjak <ubizjak@gmail.com>
150 * src/x86/ffi64.c (classify_argument): Update from
151 gcc/config/i386/i386.c.
152 (ffi_closure_unix64_inner): Do not use the address of two consecutive
153 SSE registers directly.
154 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
155 for x86_64 linux targets.
157 2009-12-04 David Edelsohn <edelsohn@gnu.org>
159 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
160 pfr for long double split between fpr13 and stack.
162 2009-12-03 David Edelsohn <edelsohn@gnu.org>
164 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
165 fparg_count twice for long double.
167 2009-12-03 David Edelsohn <edelsohn@gnu.org>
170 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
172 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
174 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
175 Remove xfails for x86 linux targets.
177 2009-12-02 David Edelsohn <edelsohn@gnu.org>
179 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
182 2009-12-01 David Edelsohn <edelsohn@gnu.org>
184 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
185 register usage. Call ffi_prep_args directly. Add long double
186 return value support.
187 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
188 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
189 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
190 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
192 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
194 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
195 Allocate result area between params and FPRs.
197 2009-11-30 David Edelsohn <edelsohn@gnu.org>
200 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
202 * src/powerpc/aix.S: Implement AIX64 version.
203 * src/powerpc/aix_closure.S: Implement AIX64 version.
204 (ffi_closure_ASM): Use extsb, lha and displament addresses.
205 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
207 (ffi_prep_cif_machdep): Same.
209 (ffi_closure_helper_DARWIN): Same.
211 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
214 * testsuite/libffi.call/testclosure.c: New test.
216 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
218 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
219 assembly version use of ___chkstk.
221 2009-09-23 Matthias Klose <doko@ubuntu.com>
223 PR libffi/40242, PR libffi/41443
224 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
225 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
226 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
227 Change the conditionals to __SOFTFP__ || __ARM_EABI__
228 for -mfloat-abi=softfp to work.
230 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
232 PR testsuite/32843 (strikes again)
233 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
234 enable proper extension on char and short.
236 2009-09-15 David Daney <ddaney@caviumnetworks.com>
238 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
239 handling for FFI_TYPE_POINTER.
240 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
241 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
242 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
243 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
244 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
245 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
246 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
247 structure and pointer returns.
248 (ffi_closure_N32): Add handling for pointer returns.
249 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
250 calc_n32_return_struct_flags): Handle soft-float.
251 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
252 (ffi_call_N32): Declare proper argument types.
253 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
256 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
258 * configure.ac (AC_PREREQ): Bump to 2.64.
260 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
262 * Makefile.am (install-html, install-pdf): Remove.
263 * Makefile.in: Regenerate.
265 * Makefile.in: Regenerate.
266 * aclocal.m4: Regenerate.
267 * configure: Regenerate.
268 * fficonfig.h.in: Regenerate.
269 * include/Makefile.in: Regenerate.
270 * man/Makefile.in: Regenerate.
271 * testsuite/Makefile.in: Regenerate.
273 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
275 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
277 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
280 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
281 return types for X86_WIN32.
282 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
283 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
284 _ffi_closure_STDCALL): Likewise.
286 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
287 (dlmmap, dlmunmap): Also use these functions on Cygwin.
289 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
294 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
297 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
299 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
300 to "" before adding paths. (This reinstates an assignment that
301 was removed by my 2009-06-30 commit, but changes the initial
302 value from "." to "".)
304 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
307 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
308 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
311 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
313 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
314 to ld_library_path. Use add_path. Add just find_libgcc_s
315 to ld_library_path, not every libgcc multilib directory.
317 2009-06-16 Wim Lewis <wiml@hhhh.org>
319 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
320 supposed to be callee-saved.
321 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
322 return buffer for odd-size structs.
324 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
327 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
328 allow_stack_execute for Darwin.
330 2009-06-16 Andrew Haley <aph@redhat.com>
332 * configure.ac (TARGETDIR): Add missing blank lines.
333 * configure: Regenerate.
335 2009-06-16 Andrew Haley <aph@redhat.com>
337 * testsuite/libffi.call/cls_align_sint64.c,
338 testsuite/libffi.call/cls_align_uint64.c,
339 testsuite/libffi.call/cls_longdouble_va.c,
340 testsuite/libffi.call/cls_ulonglong.c,
341 testsuite/libffi.call/return_ll1.c,
342 testsuite/libffi.call/stret_medium2.c: Fix printf format
344 * testsuite/libffi.call/ffitest.h,
345 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
347 2009-06-15 Andrew Haley <aph@redhat.com>
349 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
350 * testsuite/libffi.call/err_bad_abi.c: Likewise.
352 2009-06-12 Andrew Haley <aph@redhat.com>
354 * Makefile.am: Remove info_TEXINFOS.
356 2009-06-12 Andrew Haley <aph@redhat.com>
358 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
359 testsuite/libffi.call/cls_align_uint64.c,
360 testsuite/libffi.call/cls_ulonglong.c,
361 testsuite/libffi.call/return_ll1.c,
362 testsuite/libffi.call/stret_medium2.c: Fix printf format
364 testsuite/libffi.special/unwindtest.cc: include stdint.h.
366 2009-06-11 Timothy Wall <twall@users.sf.net>
371 include/ffi_common.h,
376 src/x86/win64.S (new),
377 README: Added win64 support (mingw or MSVC)
381 testsuite/Makefile.in,
383 aclocal.m4: Regenerated
384 * ltcf-c.sh: properly escape cygwin/w32 path
385 * man/ffi_call.3: Clarify size requirements for return value.
386 * src/x86/ffi64.c: Fix filename in comment.
387 * src/x86/win32.S: Remove unused extern.
389 * testsuite/libffi.call/closure_fn0.c,
390 testsuite/libffi.call/closure_fn1.c,
391 testsuite/libffi.call/closure_fn2.c,
392 testsuite/libffi.call/closure_fn3.c,
393 testsuite/libffi.call/closure_fn4.c,
394 testsuite/libffi.call/closure_fn5.c,
395 testsuite/libffi.call/closure_fn6.c,
396 testsuite/libffi.call/closure_stdcall.c,
397 testsuite/libffi.call/cls_12byte.c,
398 testsuite/libffi.call/cls_16byte.c,
399 testsuite/libffi.call/cls_18byte.c,
400 testsuite/libffi.call/cls_19byte.c,
401 testsuite/libffi.call/cls_1_1byte.c,
402 testsuite/libffi.call/cls_20byte.c,
403 testsuite/libffi.call/cls_20byte1.c,
404 testsuite/libffi.call/cls_24byte.c,
405 testsuite/libffi.call/cls_2byte.c,
406 testsuite/libffi.call/cls_3_1byte.c,
407 testsuite/libffi.call/cls_3byte1.c,
408 testsuite/libffi.call/cls_3byte2.c,
409 testsuite/libffi.call/cls_4_1byte.c,
410 testsuite/libffi.call/cls_4byte.c,
411 testsuite/libffi.call/cls_5_1_byte.c,
412 testsuite/libffi.call/cls_5byte.c,
413 testsuite/libffi.call/cls_64byte.c,
414 testsuite/libffi.call/cls_6_1_byte.c,
415 testsuite/libffi.call/cls_6byte.c,
416 testsuite/libffi.call/cls_7_1_byte.c,
417 testsuite/libffi.call/cls_7byte.c,
418 testsuite/libffi.call/cls_8byte.c,
419 testsuite/libffi.call/cls_9byte1.c,
420 testsuite/libffi.call/cls_9byte2.c,
421 testsuite/libffi.call/cls_align_double.c,
422 testsuite/libffi.call/cls_align_float.c,
423 testsuite/libffi.call/cls_align_longdouble.c,
424 testsuite/libffi.call/cls_align_longdouble_split.c,
425 testsuite/libffi.call/cls_align_longdouble_split2.c,
426 testsuite/libffi.call/cls_align_pointer.c,
427 testsuite/libffi.call/cls_align_sint16.c,
428 testsuite/libffi.call/cls_align_sint32.c,
429 testsuite/libffi.call/cls_align_sint64.c,
430 testsuite/libffi.call/cls_align_uint16.c,
431 testsuite/libffi.call/cls_align_uint32.c,
432 testsuite/libffi.call/cls_align_uint64.c,
433 testsuite/libffi.call/cls_dbls_struct.c,
434 testsuite/libffi.call/cls_double.c,
435 testsuite/libffi.call/cls_double_va.c,
436 testsuite/libffi.call/cls_float.c,
437 testsuite/libffi.call/cls_longdouble.c,
438 testsuite/libffi.call/cls_longdouble_va.c,
439 testsuite/libffi.call/cls_multi_schar.c,
440 testsuite/libffi.call/cls_multi_sshort.c,
441 testsuite/libffi.call/cls_multi_sshortchar.c,
442 testsuite/libffi.call/cls_multi_uchar.c,
443 testsuite/libffi.call/cls_multi_ushort.c,
444 testsuite/libffi.call/cls_multi_ushortchar.c,
445 testsuite/libffi.call/cls_pointer.c,
446 testsuite/libffi.call/cls_pointer_stack.c,
447 testsuite/libffi.call/cls_schar.c,
448 testsuite/libffi.call/cls_sint.c,
449 testsuite/libffi.call/cls_sshort.c,
450 testsuite/libffi.call/cls_uchar.c,
451 testsuite/libffi.call/cls_uint.c,
452 testsuite/libffi.call/cls_ulonglong.c,
453 testsuite/libffi.call/cls_ushort.c,
454 testsuite/libffi.call/err_bad_abi.c,
455 testsuite/libffi.call/err_bad_typedef.c,
456 testsuite/libffi.call/float2.c,
457 testsuite/libffi.call/huge_struct.c,
458 testsuite/libffi.call/nested_struct.c,
459 testsuite/libffi.call/nested_struct1.c,
460 testsuite/libffi.call/nested_struct10.c,
461 testsuite/libffi.call/nested_struct2.c,
462 testsuite/libffi.call/nested_struct3.c,
463 testsuite/libffi.call/nested_struct4.c,
464 testsuite/libffi.call/nested_struct5.c,
465 testsuite/libffi.call/nested_struct6.c,
466 testsuite/libffi.call/nested_struct7.c,
467 testsuite/libffi.call/nested_struct8.c,
468 testsuite/libffi.call/nested_struct9.c,
469 testsuite/libffi.call/problem1.c,
470 testsuite/libffi.call/return_ldl.c,
471 testsuite/libffi.call/return_ll1.c,
472 testsuite/libffi.call/stret_large.c,
473 testsuite/libffi.call/stret_large2.c,
474 testsuite/libffi.call/stret_medium.c,
475 testsuite/libffi.call/stret_medium2.c,
476 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
477 of checking for MMAP. Use intptr_t instead of long casts.
479 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
481 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
482 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
483 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
485 2009-06-09 Andrew Haley <aph@redhat.com>
487 * src/x86/freebsd.S: Add missing file.
489 2009-06-08 Andrew Haley <aph@redhat.com>
491 Import from libffi 3.0.8:
493 * doc/libffi.texi: New file.
494 * doc/libffi.info: Likewise.
495 * doc/stamp-vti: Likewise.
496 * man/Makefile.am: New file.
497 * man/ffi_call.3: New file.
499 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
501 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
503 * configure.ac: Bump version to 3.0.8.
504 parisc*-*-linux*: Add.
505 i386-*-freebsd* | i386-*-openbsd*: Add.
506 powerpc-*-beos*: Add.
507 AM_CONDITIONAL X86_FREEBSD: Add.
508 AC_CONFIG_FILES: Add man/Makefile.
510 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
512 2009-06-08 Andrew Haley <aph@redhat.com>
514 * README: Import from libffi 3.0.8.
516 2009-06-08 Andrew Haley <aph@redhat.com>
518 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
519 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
520 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
521 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
523 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
524 * testsuite/libffi.call/stret_medium.c: Likewise.
525 * testsuite/libffi.call/stret_large2.c: Likewise.
526 * testsuite/libffi.call/stret_large.c: Likewise.
528 2008-12-26 Timothy Wall <twall@users.sf.net>
530 * testsuite/libffi.call/cls_longdouble.c,
531 testsuite/libffi.call/cls_longdouble_va.c,
532 testsuite/libffi.call/cls_align_longdouble.c,
533 testsuite/libffi.call/cls_align_longdouble_split.c,
534 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
535 failures on x86_64 cygwin/mingw.
537 2008-12-22 Timothy Wall <twall@users.sf.net>
539 * testsuite/libffi.call/closure_fn0.c,
540 testsuite/libffi.call/closure_fn1.c,
541 testsuite/libffi.call/closure_fn2.c,
542 testsuite/libffi.call/closure_fn3.c,
543 testsuite/libffi.call/closure_fn4.c,
544 testsuite/libffi.call/closure_fn5.c,
545 testsuite/libffi.call/closure_fn6.c,
546 testsuite/libffi.call/closure_loc_fn0.c,
547 testsuite/libffi.call/closure_stdcall.c,
548 testsuite/libffi.call/cls_align_pointer.c,
549 testsuite/libffi.call/cls_pointer.c,
550 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
551 pointer to integer (intptr_t).
552 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
554 2008-07-24 Anthony Green <green@redhat.com>
556 * testsuite/libffi.call/cls_dbls_struct.c,
557 testsuite/libffi.call/cls_double_va.c,
558 testsuite/libffi.call/cls_longdouble.c,
559 testsuite/libffi.call/cls_longdouble_va.c,
560 testsuite/libffi.call/cls_pointer.c,
561 testsuite/libffi.call/cls_pointer_stack.c,
562 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
565 2008-03-04 Anthony Green <green@redhat.com>
569 * testsuite/libffi.call/cls_align_longdouble_split2.c
570 testsuite/libffi.call/cls_align_longdouble_split.c
571 testsuite/libffi.call/cls_dbls_struct.c
572 testsuite/libffi.call/cls_double_va.c
573 testsuite/libffi.call/cls_longdouble.c
574 testsuite/libffi.call/cls_longdouble_va.c
575 testsuite/libffi.call/cls_pointer.c
576 testsuite/libffi.call/cls_pointer_stack.c
577 testsuite/libffi.call/err_bad_abi.c
578 testsuite/libffi.call/err_bad_typedef.c
579 testsuite/libffi.call/stret_large2.c
580 testsuite/libffi.call/stret_large.c
581 testsuite/libffi.call/stret_medium2.c
582 testsuite/libffi.call/stret_medium.c: New tests from Apple.
584 2009-06-05 Andrew Haley <aph@redhat.com>
586 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
589 2009-06-04 Andrew Haley <aph@redhat.com>
591 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
594 2008-02-26 Anthony Green <green@redhat.com>
595 Thomas Heller <theller@ctypes.org>
597 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
600 2008-02-03 Timothy Wall <twall@users.sf.net>
602 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
603 offset based on code pointer, not data pointer.
605 2008-01-31 Timothy Wall <twall@users.sf.net>
607 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
609 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
611 * src/x86/win32.S: Add assembly for stdcall closure.
612 * src/x86/ffi.c: Initialize stdcall closure trampoline.
614 2009-06-04 Andrew Haley <aph@redhat.com>
616 * include/ffi.h.in: Change void (*)() to void (*)(void).
617 * src/x86/ffi.c: Likewise.
619 2009-06-04 Andrew Haley <aph@redhat.com>
621 * src/powerpc/ppc_closure.S: Insert licence header.
622 * src/powerpc/linux64_closure.S: Likewise.
623 * src/m68k/sysv.S: Likewise.
625 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
626 * src/powerpc/ffi.c: Likewise.
627 * src/powerpc/ffi_darwin.c: Likewise.
628 * src/m32r/ffi.c: Likewise.
629 * src/sh64/ffi.c: Likewise.
630 * src/x86/ffi64.c: Likewise.
631 * src/alpha/ffi.c: Likewise.
632 * src/alpha/osf.S: Likewise.
633 * src/frv/ffi.c: Likewise.
634 * src/s390/ffi.c: Likewise.
635 * src/pa/ffi.c: Likewise.
636 * src/pa/hpux32.S: Likewise.
637 * src/ia64/unix.S: Likewise.
638 * src/ia64/ffi.c: Likewise.
639 * src/sparc/ffi.c: Likewise.
640 * src/mips/ffi.c: Likewise.
641 * src/sh/ffi.c: Likewise.
643 2008-02-15 David Daney <ddaney@avtrex.com>
645 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
646 Define (conditionally), and use it to include cachectl.h.
647 (ffi_prep_closure_loc): Fix cache flushing.
648 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
650 2009-06-04 Andrew Haley <aph@redhat.com>
656 src/powerpc/ffitarget.h,
658 src/sh64/ffitarget.h,
669 src/alpha/ffitarget.h,
672 src/m68k/ffitarget.h,
675 src/s390/ffitarget.h,
677 src/cris/ffitarget.h,
682 src/ia64/ffitarget.h,
685 src/ia64/ia64_flags.h,
689 src/sparc/ffitarget.h,
692 src/mips/ffitarget.h,
699 src/sh/sysv.S: Update license text.
701 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
703 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
704 (.eh_frame): Add FDE for it.
706 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
708 * configure.ac: Also check if assembler supports pc-relative
709 relocs on X86_WIN32 targets.
710 * configure: Regenerate.
711 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
712 (_ffi_call_SYSV): Add missing function type symbol .def and
713 add EH markup labels.
714 (_ffi_call_STDCALL): Likewise.
715 (_ffi_closure_SYSV): Likewise.
716 (_ffi_closure_raw_SYSV): Likewise.
717 (.eh_frame): Add hand-crafted EH data.
719 2009-04-09 Jakub Jelinek <jakub@redhat.com>
721 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
722 version 3 of the GNU General Public License and to point readers
723 at the COPYING3 file and the FSF's license web page.
724 * testsuite/libffi.call/call.exp: Likewise.
725 * testsuite/libffi.special/special.exp: Likewise.
727 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
729 * configure: Regenerate.
731 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
734 * configure.ac (HAVE_AS_X86_PCREL): New test.
735 * configure: Regenerate.
736 * fficonfig.h.in: Regenerate.
737 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
738 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
739 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
740 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
741 * src/x86/unix64.S (.Lstore_table): Move to .text section.
742 (.Lload_table): Likewise.
743 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
745 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
747 * configure: Regenerate.
749 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
751 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
752 signed/unsigned int8/16 return values.
753 * src/sparc/v8.S (ffi_call_v8): Likewise.
754 (ffi_closure_v8): Likewise.
756 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
757 Steve Ellcey <sje@cup.hp.com>
759 * configure: Regenerate for new libtool.
760 * Makefile.in: Ditto.
761 * include/Makefile.in: Ditto.
764 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
766 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
767 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
768 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
769 Adjust copyright notice.
770 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
771 register or two register to use for FFI_SYSV structs.
772 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
773 (ffi_closure_helper_SYSV): Fix the return type for
774 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
775 Adjust copyright notice.
777 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
779 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
782 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
784 * configure: Regenerate.
785 * include/Makefile.in: Regenerate.
786 * testsuite/Makefile.in: Regenerate.
788 2008-06-07 Joseph Myers <joseph@codesourcery.com>
790 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
791 powerpc-*-beos*): Remove.
792 * configure: Regenerate.
794 2008-05-09 Julian Brown <julian@codesourcery.com>
796 * Makefile.am (LTLDFLAGS): New.
797 (libffi_la_LDFLAGS): Use above.
798 * Makefile.in: Regenerate.
800 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
803 * aclocal.m4: Regenerate.
804 * configure: Regenerate.
806 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
808 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
809 * src/sh64/sysv.S: Likewise.
811 2008-03-26 Daniel Jacobowitz <dan@debian.org>
813 * src/arm/sysv.S: Fix ARM comment marker.
815 2008-03-26 Jakub Jelinek <jakub@redhat.com>
817 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
818 * src/s390/sysv.S: Likewise.
819 * src/powerpc/ppc_closure.S: Likewise.
820 * src/powerpc/sysv.S: Likewise.
821 * src/x86/unix64.S: Likewise.
822 * src/x86/sysv.S: Likewise.
823 * src/sparc/v8.S: Likewise.
824 * src/sparc/v9.S: Likewise.
825 * src/m68k/sysv.S: Likewise.
826 * src/arm/sysv.S: Likewise.
828 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
830 * aclocal.m4: Regenerate.
831 * configure: Likewise.
832 * Makefile.in: Likewise.
833 * include/Makefile.in: Likewise.
834 * testsuite/Makefile.in: Likewise.
836 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
837 Andreas Tobler <a.tobler@schweiz.org>
839 * configure.ac: Add amd64-*-freebsd* target.
840 * configure: Regenerate.
842 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
845 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
848 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
851 2008-01-24 David Edelsohn <edelsohn@gnu.org>
853 * configure: Regenerate.
855 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
857 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
859 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
862 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
863 signed/unsigned int8/16 for X86_DARWIN.
864 Updated copyright info.
865 Handle one and two byte structs with special cif->flags.
866 * src/x86/ffitarget.h: Add special types for one and two byte structs.
867 Updated copyright info.
868 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
870 Remove code to pop args from the stack after call.
871 Special-case signed/unsigned for int8/16, one and two byte structs.
872 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
873 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
875 Updated copyright info.
877 2007-12-08 David Daney <ddaney@avtrex.com>
879 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
880 SUBU, add with ADDU and use smaller code sequences.
882 2007-12-07 David Daney <ddaney@avtrex.com>
884 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
887 2007-12-06 David Daney <ddaney@avtrex.com>
889 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
891 (ffi_java_raw): New typedef.
892 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
893 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
895 (ffi_java_raw_closure) : Same.
896 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
898 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
900 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
901 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
902 sizeof(ffi_java_raw) for alignment calculations.
903 (ffi_java_ptrarray_to_raw): Same.
904 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
905 if FFI_SIZEOF_JAVA_RAW == 4.
906 (ffi_java_raw_to_rvalue): Same.
907 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
908 (ffi_java_translate_args): Same.
909 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
911 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
913 2007-12-06 David Daney <ddaney@avtrex.com>
915 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
918 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
921 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
922 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
923 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
924 set the NUM_FPR_ARG_REGISTERS according to.
925 Add support for potential soft-float support under hard-float
927 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
928 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
929 to the FFI_LINUX_SOFT_FLOAT ABI.
930 (ffi_prep_cif_machdep): Likewise.
931 (ffi_closure_helper_SYSV): Likewise.
932 * src/powerpc/ppc_closure.S: Make sure not to store float/double
933 on archs where __NO_FPRS__ is true.
934 Add FFI_TYPE_UINT128 support.
935 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
936 Adjust copyright notice.
938 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
940 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
941 * include/ffi_common.h: ... here.
944 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
946 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
947 * src/powerpc/linux64.S: Likewise.
948 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
949 * testsuite/libffi.call/return_ldl.c: New test.
951 2007-09-04 <aph@redhat.com>
953 * src/arm/sysv.S (UNWIND): New.
954 (Whole file): Conditionally compile unwinder directives.
955 * src/arm/sysv.S: Add unwinder directives.
957 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
958 Only treat r0 as a struct address if we're actually returning a
960 Only copy the bytes that are actually within a struct.
961 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
962 is returned in r0, not passed by address.
963 (ffi_call): Allocate a word-sized temporary for the case where
964 a composite is returned in r0.
965 (ffi_prep_incoming_args_SYSV): Align as necessary.
967 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
969 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
970 directly using the sys_cacheflush syscall.
972 2007-07-27 Andrew Haley <aph@redhat.com>
974 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
976 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
978 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
979 * configure.ac: Likewise.
980 * Makefile.in: Regenerate.
981 * include/Makefile.in: Likewise.
982 * testsuite/Makefile.in: Likewise.
983 * configure: Likewise.
985 2007-08-24 David Daney <ddaney@avtrex.com>
987 * testsuite/libffi.call/return_sl.c: New test.
989 2007-08-10 David Daney <ddaney@avtrex.com>
991 * testsuite/libffi.call/cls_multi_ushort.c,
992 testsuite/libffi.call/cls_align_uint16.c,
993 testsuite/libffi.call/nested_struct1.c,
994 testsuite/libffi.call/nested_struct3.c,
995 testsuite/libffi.call/cls_7_1_byte.c,
996 testsuite/libffi.call/nested_struct5.c,
997 testsuite/libffi.call/cls_double.c,
998 testsuite/libffi.call/nested_struct7.c,
999 testsuite/libffi.call/cls_sint.c,
1000 testsuite/libffi.call/nested_struct9.c,
1001 testsuite/libffi.call/cls_20byte1.c,
1002 testsuite/libffi.call/cls_multi_sshortchar.c,
1003 testsuite/libffi.call/cls_align_sint64.c,
1004 testsuite/libffi.call/cls_3byte2.c,
1005 testsuite/libffi.call/cls_multi_schar.c,
1006 testsuite/libffi.call/cls_multi_uchar.c,
1007 testsuite/libffi.call/cls_19byte.c,
1008 testsuite/libffi.call/cls_9byte1.c,
1009 testsuite/libffi.call/cls_align_float.c,
1010 testsuite/libffi.call/closure_fn1.c,
1011 testsuite/libffi.call/problem1.c,
1012 testsuite/libffi.call/closure_fn3.c,
1013 testsuite/libffi.call/cls_sshort.c,
1014 testsuite/libffi.call/closure_fn5.c,
1015 testsuite/libffi.call/cls_align_double.c,
1016 testsuite/libffi.call/nested_struct.c,
1017 testsuite/libffi.call/cls_2byte.c,
1018 testsuite/libffi.call/nested_struct10.c,
1019 testsuite/libffi.call/cls_4byte.c,
1020 testsuite/libffi.call/cls_6byte.c,
1021 testsuite/libffi.call/cls_8byte.c,
1022 testsuite/libffi.call/cls_multi_sshort.c,
1023 testsuite/libffi.call/cls_align_sint16.c,
1024 testsuite/libffi.call/cls_align_uint32.c,
1025 testsuite/libffi.call/cls_20byte.c,
1026 testsuite/libffi.call/cls_float.c,
1027 testsuite/libffi.call/nested_struct2.c,
1028 testsuite/libffi.call/cls_5_1_byte.c,
1029 testsuite/libffi.call/nested_struct4.c,
1030 testsuite/libffi.call/cls_24byte.c,
1031 testsuite/libffi.call/nested_struct6.c,
1032 testsuite/libffi.call/cls_64byte.c,
1033 testsuite/libffi.call/nested_struct8.c,
1034 testsuite/libffi.call/cls_uint.c,
1035 testsuite/libffi.call/cls_multi_ushortchar.c,
1036 testsuite/libffi.call/cls_schar.c,
1037 testsuite/libffi.call/cls_uchar.c,
1038 testsuite/libffi.call/cls_align_uint64.c,
1039 testsuite/libffi.call/cls_ulonglong.c,
1040 testsuite/libffi.call/cls_align_longdouble.c,
1041 testsuite/libffi.call/cls_1_1byte.c,
1042 testsuite/libffi.call/cls_12byte.c,
1043 testsuite/libffi.call/cls_3_1byte.c,
1044 testsuite/libffi.call/cls_3byte1.c,
1045 testsuite/libffi.call/cls_4_1byte.c,
1046 testsuite/libffi.call/cls_6_1_byte.c,
1047 testsuite/libffi.call/cls_16byte.c,
1048 testsuite/libffi.call/cls_18byte.c,
1049 testsuite/libffi.call/closure_fn0.c,
1050 testsuite/libffi.call/cls_9byte2.c,
1051 testsuite/libffi.call/closure_fn2.c,
1052 testsuite/libffi.call/closure_fn4.c,
1053 testsuite/libffi.call/cls_ushort.c,
1054 testsuite/libffi.call/closure_fn6.c,
1055 testsuite/libffi.call/cls_5byte.c,
1056 testsuite/libffi.call/cls_align_pointer.c,
1057 testsuite/libffi.call/cls_7byte.c,
1058 testsuite/libffi.call/cls_align_sint32.c,
1059 testsuite/libffi.special/unwindtest_ffi_call.cc,
1060 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
1062 2007-08-10 David Daney <ddaney@avtrex.com>
1065 * configure.ac: Don't treat mips64 as a special case.
1066 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
1067 * configure: Regenerate
1068 * Makefile.in: Ditto.
1069 * fficonfig.h.in: Ditto.
1070 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
1071 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
1072 (FFI_DEFAULT_ABI): Set for n64 case.
1073 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
1074 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
1075 (ffi_closure_N32): New function.
1076 (.eh_frame): New section
1077 * src/mips/o32.S: Clean up comments.
1078 (ffi_closure_O32): Pass ffi_closure parameter in $12.
1079 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
1080 _MIPS_SIM == _ABIN32 throughout.
1081 (FFI_MIPS_STOP_HERE): New, use in place of
1083 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1084 to support n32/n64 ABIs.
1085 (calc_n32_struct_flags): Rewrite.
1086 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1087 position of flag bits.
1088 (ffi_prep_cif_machdep): Rewrite n32 portion.
1089 (ffi_call): Enable for n64. Add special handling for small structure
1091 (ffi_prep_closure_loc): Add n32 and n64 support.
1092 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1093 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1095 2007-08-08 David Daney <ddaney@avtrex.com>
1097 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1098 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1100 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1101 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1102 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1103 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1104 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1105 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1106 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1107 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1108 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1109 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1110 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1111 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1112 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1113 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1114 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1115 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1116 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1117 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1118 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1119 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1120 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1121 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1122 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1123 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1124 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1125 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1126 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1127 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1128 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1129 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1130 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1132 2007-08-07 Andrew Haley <aph@redhat.com>
1134 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1137 2007-08-06 Andrew Haley <aph@redhat.com>
1140 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1141 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1144 2007-08-02 David Daney <ddaney@avtrex.com>
1146 * testsuite/libffi.call/return_ul.c (main): Define return type as
1147 ffi_arg. Use proper printf conversion specifier.
1149 2007-07-30 Andrew Haley <aph@redhat.com>
1152 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1153 signed/unsigned int8/16.
1154 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1156 Remove code to pop args from the stack after call.
1157 Special-case signed/unsigned int8/16.
1158 * testsuite/libffi.call/return_sc.c (main): Revert.
1160 2007-07-26 Richard Guenther <rguenther@suse.de>
1163 * testsuite/libffi.call/return_sc.c (main): Verify call
1164 result as signed char, not ffi_arg.
1166 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1168 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1169 * configure: Regenerate.
1171 2007-07-11 David Daney <ddaney@avtrex.com>
1173 * src/mips/ffi.c: Don't include sys/cachectl.h.
1174 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1177 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1179 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1180 from (ffi_prep_closure): ... this.
1181 (FFI_INIT_TRAMPOLINE): Adjust.
1183 2005-12-31 Phil Blundell <pb@reciva.com>
1185 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1186 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1187 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1188 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1189 (FFI_CLOSURES): Enable closure support.
1191 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1193 * testsuite/libffi.call/cls_multi_ushort.c,
1194 testsuite/libffi.call/cls_align_uint16.c,
1195 testsuite/libffi.call/nested_struct1.c,
1196 testsuite/libffi.call/nested_struct3.c,
1197 testsuite/libffi.call/cls_7_1_byte.c,
1198 testsuite/libffi.call/cls_double.c,
1199 testsuite/libffi.call/nested_struct5.c,
1200 testsuite/libffi.call/nested_struct7.c,
1201 testsuite/libffi.call/cls_sint.c,
1202 testsuite/libffi.call/nested_struct9.c,
1203 testsuite/libffi.call/cls_20byte1.c,
1204 testsuite/libffi.call/cls_multi_sshortchar.c,
1205 testsuite/libffi.call/cls_align_sint64.c,
1206 testsuite/libffi.call/cls_3byte2.c,
1207 testsuite/libffi.call/cls_multi_schar.c,
1208 testsuite/libffi.call/cls_multi_uchar.c,
1209 testsuite/libffi.call/cls_19byte.c,
1210 testsuite/libffi.call/cls_9byte1.c,
1211 testsuite/libffi.call/cls_align_float.c,
1212 testsuite/libffi.call/closure_fn1.c,
1213 testsuite/libffi.call/problem1.c,
1214 testsuite/libffi.call/closure_fn3.c,
1215 testsuite/libffi.call/cls_sshort.c,
1216 testsuite/libffi.call/closure_fn5.c,
1217 testsuite/libffi.call/cls_align_double.c,
1218 testsuite/libffi.call/cls_2byte.c,
1219 testsuite/libffi.call/nested_struct.c,
1220 testsuite/libffi.call/nested_struct10.c,
1221 testsuite/libffi.call/cls_4byte.c,
1222 testsuite/libffi.call/cls_6byte.c,
1223 testsuite/libffi.call/cls_8byte.c,
1224 testsuite/libffi.call/cls_multi_sshort.c,
1225 testsuite/libffi.call/cls_align_uint32.c,
1226 testsuite/libffi.call/cls_align_sint16.c,
1227 testsuite/libffi.call/cls_float.c,
1228 testsuite/libffi.call/cls_20byte.c,
1229 testsuite/libffi.call/cls_5_1_byte.c,
1230 testsuite/libffi.call/nested_struct2.c,
1231 testsuite/libffi.call/cls_24byte.c,
1232 testsuite/libffi.call/nested_struct4.c,
1233 testsuite/libffi.call/nested_struct6.c,
1234 testsuite/libffi.call/cls_64byte.c,
1235 testsuite/libffi.call/nested_struct8.c,
1236 testsuite/libffi.call/cls_uint.c,
1237 testsuite/libffi.call/cls_multi_ushortchar.c,
1238 testsuite/libffi.call/cls_schar.c,
1239 testsuite/libffi.call/cls_uchar.c,
1240 testsuite/libffi.call/cls_align_uint64.c,
1241 testsuite/libffi.call/cls_ulonglong.c,
1242 testsuite/libffi.call/cls_align_longdouble.c,
1243 testsuite/libffi.call/cls_1_1byte.c,
1244 testsuite/libffi.call/cls_12byte.c,
1245 testsuite/libffi.call/cls_3_1byte.c,
1246 testsuite/libffi.call/cls_3byte1.c,
1247 testsuite/libffi.call/cls_4_1byte.c,
1248 testsuite/libffi.call/cls_6_1_byte.c,
1249 testsuite/libffi.call/cls_16byte.c,
1250 testsuite/libffi.call/cls_18byte.c,
1251 testsuite/libffi.call/closure_fn0.c,
1252 testsuite/libffi.call/cls_9byte2.c,
1253 testsuite/libffi.call/closure_fn2.c,
1254 testsuite/libffi.call/closure_fn4.c,
1255 testsuite/libffi.call/cls_ushort.c,
1256 testsuite/libffi.call/closure_fn6.c,
1257 testsuite/libffi.call/cls_5byte.c,
1258 testsuite/libffi.call/cls_align_pointer.c,
1259 testsuite/libffi.call/cls_7byte.c,
1260 testsuite/libffi.call/cls_align_sint32.c,
1261 testsuite/libffi.special/unwindtest_ffi_call.cc,
1262 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1264 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1266 * aclocal.m4: Regenerated.
1268 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1270 * configure: Regenerate.
1272 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1274 * Makefile.in: Regenerate.
1275 * configure: Regenerate.
1276 * aclocal.m4: Regenerate.
1277 * include/Makefile.in: Regenerate.
1278 * testsuite/Makefile.in: Regenerate.
1280 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1282 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1283 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1284 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1285 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1286 (FFI_CLOSURES): Enable closure support.
1288 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1290 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1291 * configure: Regenerate.
1292 * fficonfig.h.in: Regenerate.
1293 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1294 CFI_OFFSET,CFI_DEF_CFA): New macros.
1295 (ffi_call_SYSV): Add callframe annotation.
1297 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1299 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1300 numerous test suite failures.
1301 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1303 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1305 * Makefile.am (EXTRA_DIST): Bring up to date.
1306 * Makefile.in: Regenerate.
1307 * src/frv/eabi.S: Remove RCS keyword.
1309 2007-04-06 Richard Henderson <rth@redhat.com>
1311 * configure.ac: Tidy target case.
1312 (HAVE_LONG_DOUBLE): Allow the target to override.
1313 * configure: Regenerate.
1314 * include/ffi.h.in: Don't define ffi_type_foo if
1315 LIBFFI_HIDE_BASIC_TYPES is defined.
1316 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1318 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1319 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1320 (ffi_type_longdouble): Special case for Alpha. Don't define
1321 if long double == double.
1323 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1324 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1325 (ffi_call, ffi_closure_osf_inner): Likewise.
1326 (ffi_closure_osf_inner): Likewise. Mark hidden.
1327 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1328 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1329 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1330 (load_table): Handle 128-bit long double.
1332 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1334 2007-04-06 Tom Tromey <tromey@redhat.com>
1337 * README: Fixed bug in example.
1339 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1341 * src/closures.c: Include sys/statfs.h.
1342 (_GNU_SOURCE): Define on Linux.
1343 (FFI_MMAP_EXEC_SELINUX): Define.
1344 (selinux_enabled): New variable.
1345 (selinux_enabled_check): New function.
1346 (is_selinux_enabled): Define.
1349 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1351 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1352 Use 'volatile float sum' to create sum of floats to avoid false
1353 negative due to excess precision on ix86 targets.
1356 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1358 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1360 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1362 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1364 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1365 (ffi_prep_closure_loc): New.
1366 (ffi_prep_raw_closure_loc): New.
1367 (ffi_prep_java_raw_closure_loc): New.
1368 * src/closures.c: New file.
1369 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1370 Replace sflags with exec_offset.
1371 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1372 sub_segment_exec_offset): New macros.
1373 (get_segment_flags, set_segment_flags, check_segment_merge): New
1375 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1376 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1377 destroy_mspace): Use new macros.
1378 (sys_alloc): Silence warning.
1379 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1380 * Makefile.in: Rebuilt.
1381 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1382 terms of ffi_prep_closure_loc.
1383 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1385 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1387 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1389 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1390 the renamed version.
1391 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1392 (ffi_prep_closure): ... this.
1393 * src/pa/ffi.c: Likewise.
1394 * src/cris/ffi.c: Likewise. Adjust.
1395 * src/frv/ffi.c: Likewise.
1396 * src/ia64/ffi.c: Likewise.
1397 * src/mips/ffi.c: Likewise.
1398 * src/powerpc/ffi_darwin.c: Likewise.
1399 * src/s390/ffi.c: Likewise.
1400 * src/sh/ffi.c: Likewise.
1401 * src/sh64/ffi.c: Likewise.
1402 * src/sparc/ffi.c: Likewise.
1403 * src/x86/ffi64.c: Likewise.
1404 * src/x86/ffi.c: Likewise.
1405 (FFI_INIT_TRAMPOLINE): Adjust.
1406 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1407 (ffi_prep_raw_closure): ... this.
1408 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1409 (ffi_prep_closure): ... this.
1410 (flush_icache): Adjust.
1412 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1414 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1417 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1419 * Makefile.am: Add dummy install-pdf target.
1420 * Makefile.in: Regenerate
1422 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1424 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1425 ffi_closure_helper_SYSV): Add long double handling.
1427 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1429 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1430 immediately after bctrl instruction.
1432 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1434 * Makefile.am (all-recursive, install-recursive,
1435 mostlyclean-recursive, clean-recursive, distclean-recursive,
1436 maintainer-clean-recursive): Add missing targets.
1437 * Makefile.in: Rebuilt.
1439 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1441 * configure.ac: Add TARGET for x86_64-*-darwin*.
1442 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1444 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1445 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1446 * configure: Regenerate.
1447 * Makefile.in: Regenerate.
1448 * include/Makefile.in: Regenerate.
1449 * testsuite/Makefile.in: Regenerate.
1450 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1453 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1455 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1458 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1460 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1461 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1463 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1464 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1466 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1467 Sandro Tolaini <tolaini@libero.it>
1469 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1471 * configure: Regenerated.
1472 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1473 (EXTRA_DIST): Add src/x86/darwin.S.
1474 * Makefile.in: Regenerated.
1475 * include/Makefile.in: Regenerated.
1476 * testsuite/Makefile.in: Regenerated.
1478 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1479 X86_WIN32, and additionally align stack to 16 bytes.
1480 * src/x86/darwin.S: New, based on sysv.S.
1481 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1483 2006-09-12 David Daney <ddaney@avtrex.com>
1486 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1487 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1488 * aclocal.m4: Regenerated for automake 1.9.6.
1489 * Makefile.in: Regenerated.
1490 * include/Makefile.in: Regenerated.
1491 * testsuite/Makefile.in: Regenerated.
1493 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1495 * include/ffi_common.h (struct): Revert accidental commit.
1497 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1499 * include/ffi_common.h: Remove lint directives.
1500 * include/ffi.h.in: Likewise.
1502 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1504 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1505 for 32-bit architectures.
1506 * testsuite/libffi.call/return_ul.c: New test case.
1508 2006-07-19 David Daney <ddaney@avtrex.com>
1510 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1511 xfail remains for mips64.
1513 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1515 * Makefile.am: Add install-html target. Add install-html to .PHONY
1516 * Makefile.in: Regenerate.
1517 * aclocal.m4: Regenerate.
1518 * include/Makefile.in: Regenerate.
1519 * testsuite/Makefile.in: Regenerate.
1521 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1523 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1526 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1528 * README: Remove notice about 'Crazy Comments'.
1529 * src/debug.c: Remove lint directives. Cleanup white spaces.
1530 * src/java_raw_api.c: Likewise.
1531 * src/prep_cif.c: Likewise.
1532 * src/raw_api.c: Likewise.
1533 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1535 * src/arm/ffi.c: Remove lint directives.
1536 * src/m32r/ffi.c: Likewise.
1537 * src/pa/ffi.c: Likewise.
1538 * src/powerpc/ffi.c: Likewise.
1539 * src/powerpc/ffi_darwin.c: Likewise.
1540 * src/sh/ffi.c: Likewise.
1541 * src/sh64/ffi.c: Likewise.
1542 * src/x86/ffi.c: Likewise.
1543 * testsuite/libffi.call/float2.c: Likewise.
1544 * testsuite/libffi.call/promotion.c: Likewise.
1545 * testsuite/libffi.call/struct1.c: Likewise.
1547 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1549 * src/pa/hpux32.S: Correct unwind offset calculation for
1551 * src/pa/linux.S: Likewise.
1553 2006-04-12 James E Wilson <wilson@specifix.com>
1556 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1557 (hfa_type_load): Call stf_spill.
1558 (hfa_type_store): Call ldf_fill.
1559 (ffi_call): Adjust calls to above routines. Add local temps for
1562 2006-04-10 Matthias Klose <doko@debian.org>
1564 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1565 directory names containing underscores.
1567 2006-04-07 James E Wilson <wilson@specifix.com>
1569 * testsuite/libffi.call/float4.c: New testcase.
1571 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1572 Andreas Tobler <a.tobler@schweiz.ch>
1574 * Makefile.am: Add PA_HPUX port.
1575 * Makefile.in: Regenerate.
1576 * include/Makefile.in: Likewise.
1577 * testsuite/Makefile.in: Likewise.
1578 * configure.ac: Add PA_HPUX rules.
1579 * configure: Regenerate.
1580 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1581 Add PA_HPUX and PA64_HPUX.
1582 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1583 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1584 (FFI_TYPE_SMALL_STRUCT2): Define.
1585 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1586 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1587 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1588 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1589 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1590 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1591 * src/pa/ffi.c (ROUND_DOWN): Delete.
1592 (fldw, fstw, fldd, fstd): Use '__asm__'.
1593 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1594 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1595 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1596 Simplify incrementing of stack slot variable. Change type of local
1597 'n' to unsigned int.
1598 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1600 (ffi_prep_cif_machdep): Likewise.
1601 (ffi_call): Likewise.
1602 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1603 return type to ffi_status. Simplify incrementing of stack slot
1604 variable. Only copy floating point argument registers when PA_LINUX
1605 is true. Reformat debug statement.
1606 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1607 FFI_TYPE_SMALL_STRUCT8.
1608 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1610 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1611 Add nops to cache flush. Add trampoline for PA_HPUX.
1612 * src/pa/hpux32.S: New file.
1613 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1614 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1615 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1616 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1617 argument type checks so that common argument types appear first.
1618 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1619 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1621 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1623 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1624 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1625 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1626 FFI_TYPE_LONGDOUBLE.
1627 (ffi_prep_args64): Assert using IBM extended double.
1628 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1629 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1630 (ffi_call): Handle FFI_LINUX.
1631 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1632 gpr3 return pointer as for struct return. Handle FFI_LINUX
1633 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1635 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1636 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1637 Don't use r6 as pointer to results, instead use sp offset. Don't
1638 make a special call to load lr with case table address, instead
1639 use offset from previous call.
1640 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1641 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1644 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1646 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1647 passed with FP registers correctly.
1648 (ffi_closure_helper_SYSV): Likewise.
1649 * src/sh64/sysv.S: Likewise.
1651 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1653 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1654 args and userdata unused.
1655 (closure_test_fn1): Mark cif and userdata unused.
1656 (main): Remove unused res.
1658 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1660 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1661 -O2, -O3, -Os and the warning flags -W -Wall.
1662 * testsuite/libffi.special/special.exp: Likewise.
1663 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1664 unused parameter unused for gcc or else do nothing.
1665 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1666 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1667 and userdata unused.
1668 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1669 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1670 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1671 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1672 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1673 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1674 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1675 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1676 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1677 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1678 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1679 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1680 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1681 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1682 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1683 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1684 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1685 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1686 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1687 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1688 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1689 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1690 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1691 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1693 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1695 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1697 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1698 void* to avoid compiler warning.
1700 (cls_struct_align_gn): Mark cif and userdata unused.
1701 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1703 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1705 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1707 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1709 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1711 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1712 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1713 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1715 (main): Cast res_call to silence gcc.
1716 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1718 (main): Cast res_call to silence gcc.
1719 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1721 (main): Cast res_call to silence gcc.
1722 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1724 (main): Cast res_call to silence gcc.
1725 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1727 (main): Cast res_call to silence gcc.
1728 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1730 (main): Cast res_call to silence gcc.
1731 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1733 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1734 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1736 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1737 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1739 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1740 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1742 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1743 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1745 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1746 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1747 and userdata unused.
1748 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1750 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1751 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1752 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1753 Cleanup white spaces.
1754 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1755 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1756 cif and userdata unused.
1757 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1759 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1760 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1761 formatters to silence gcc.
1762 (B_gn): Mark cif and userdata unused.
1763 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1765 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1766 (B_gn): Mark cif and userdata unused.
1767 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1769 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1770 (B_gn): Mark cif and userdata unused.
1771 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1773 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1774 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1775 * testsuite/libffi.call/problem1.c (stub): Likewise.
1776 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1778 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1779 in the last commit for this test case in the test case itself.
1780 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1782 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1783 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1784 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1785 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1786 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1787 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1789 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1791 * src/sh/sysv.S: Fix register numbers in the FDE for
1794 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1796 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1797 declaration to avoid a false negative on ix86. See PR323.
1799 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1801 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1802 and cast integer to void * if needed. Update the pointer to
1803 the FP register saved area correctly.
1805 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1807 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1809 * testsuite/libffi.call/nested_struct4.c: Likewise.
1811 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1813 * testsuite/libffi.call/return_dbl.c: New test case.
1814 * testsuite/libffi.call/return_dbl1.c: Likewise.
1815 * testsuite/libffi.call/return_dbl2.c: Likewise.
1816 * testsuite/libffi.call/return_fl.c: Likewise.
1817 * testsuite/libffi.call/return_fl1.c: Likewise.
1818 * testsuite/libffi.call/return_fl2.c: Likewise.
1819 * testsuite/libffi.call/return_fl3.c: Likewise.
1820 * testsuite/libffi.call/closure_fn6.c: Likewise.
1822 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1824 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1825 here to be used by other test cases too.
1827 * testsuite/libffi.call/nested_struct10.c: New test case.
1828 * testsuite/libffi.call/nested_struct9.c: Likewise.
1829 * testsuite/libffi.call/nested_struct8.c: Likewise.
1830 * testsuite/libffi.call/nested_struct7.c: Likewise.
1831 * testsuite/libffi.call/nested_struct6.c: Likewise.
1832 * testsuite/libffi.call/nested_struct5.c: Likewise.
1833 * testsuite/libffi.call/nested_struct4.c: Likewise.
1835 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1837 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1838 * configure: Rebuilt.
1840 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1842 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1843 instead do the shifting inline.
1844 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1845 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1846 and 8 byte structs, for the remaining struct sizes don't call
1847 __lshrdi3, instead do the shifting inline.
1849 2005-12-07 Thiemo Seufer <ths@networkno.de>
1851 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1852 missing parentheses.
1853 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1854 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1855 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1856 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1857 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1859 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1861 (ffi_prep_closure): Improve trampoline instruction scheduling.
1862 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1864 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1866 * src/powerpc/ffi.c: Formatting.
1867 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1868 (ffi_prep_args64): Likewise.
1870 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1872 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1873 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1875 2005-09-26 Tom Tromey <tromey@redhat.com>
1877 * testsuite/libffi.call/float1.c (value_type): New typedef.
1878 (CANARY): New define.
1879 (main): Check for result buffer overflow.
1880 * src/powerpc/linux64.S: Handle linux64 long double returns.
1881 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1882 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1884 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1887 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1889 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1891 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1893 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1894 (AH_BOTTOM): Add FFI_HIDDEN definition.
1895 * configure: Rebuilt.
1896 * fficonfig.h.in: Rebuilt.
1897 * src/powerpc/ffi.c (hidden): Remove.
1898 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1899 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1900 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1901 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1902 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1903 add FFI_HIDDEN to its prototype.
1904 (ffi_closure_SYSV_inner): New.
1905 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1906 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1908 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1911 * configure: Rebuilt.
1912 * configure.ac: Handle i*86-*-gnu*.
1914 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1916 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1917 DW_CFA_offset_extended_sf rather than
1918 DW_CFA_GNU_negative_offset_extended.
1919 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1921 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1923 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1925 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1926 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1927 partially on register.
1928 (ffi_closure_helper_SYSV): Likewise.
1929 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1931 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1933 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1935 * src/sh64/ffi.c (simple_type): Remove.
1936 (return_type): Handle small structures correctly.
1937 (ffi_prep_args): Likewise.
1938 (ffi_call): Likewise.
1939 (ffi_closure_helper_SYSV): Likewise.
1940 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1941 Emit position independent code if PIC and remove wrong datalabel
1942 prefixes from EH data.
1944 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1946 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1947 * Makefile.in: Regenerate.
1948 * include/Makefile.in: Likewise.
1949 * testsuite/Makefile.in: Likewise.
1950 * configure.ac: Add POWERPC_FREEBSD rules.
1951 * configure: Regenerate.
1952 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1953 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1954 * src/powerpc/ffi.c: Add flags to handle small structure returns
1956 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1958 (ffi_closure_helper_SYSV): Likewise.
1959 * src/powerpc/ppc_closure.S: Add return types for small structures.
1960 * src/powerpc/sysv.S: Add bits to handle small structures for
1963 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1965 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1966 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1967 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1969 2005-07-05 Randolph Chung <tausq@debian.org>
1971 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1972 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1973 structures. Kill compilation warnings.
1974 (ffi_closure_inner_LINUX): Print return values as hex in debug
1975 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1976 Properly handle 5-7 byte structure returns.
1977 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1978 (FFI_TYPE_SMALL_STRUCT2): Remove.
1979 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1980 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1981 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1982 (checksmst1, checksmst2): Remove.
1983 (checksmst3): Optimize handling of 3-byte struct returns.
1984 (checksmst567): Properly handle 5-7 byte struct returns.
1986 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1989 * src/mips/n32.S: Enforce PIC code.
1990 * src/mips/o32.S: Likewise.
1992 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1994 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1995 * configure: Regenerate.
1997 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1999 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
2000 to call ffi_closure_helper_SYSV. Append @local instead.
2001 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
2003 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
2005 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
2006 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
2007 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
2008 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
2009 include/Makefile.in, testsuite/Makefile.in: Regenerate.
2011 2005-05-09 Mike Stump <mrs@apple.com>
2013 * configure: Regenerate.
2015 2005-05-08 Richard Henderson <rth@redhat.com>
2018 * src/alpha/osf.S: Update unwind into to match code.
2020 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
2021 Richard Henderson <rth@redhat.com>
2023 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
2025 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
2026 (ffi_prep_closure): Set carry bit if sse-used flag set.
2027 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
2028 Only load sse registers if ssecount non-zero.
2029 (ffi_closure_unix64): Only save sse registers if carry set on entry.
2031 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
2033 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
2034 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
2035 * configure: Regenerate.
2037 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
2039 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
2040 have Tcl8.3-compatible intermediate variable.
2042 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
2043 Hans-Peter Nilsson <hp@axis.com>
2045 * Makefile.am: Add CRIS support.
2046 * configure.ac: Likewise.
2047 * Makefile.in, configure, testsuite/Makefile.in,
2048 include/Makefile.in: Regenerate.
2049 * src/cris: New directory.
2050 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
2051 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
2053 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
2054 \r?\n in output tests.
2056 2005-04-12 Mike Stump <mrs@apple.com>
2058 * configure: Regenerate.
2060 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
2062 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
2064 2005-03-30 Steve Ellcey <sje@cup.hp.com>
2066 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
2068 * src/ia64/unix.S (ffi_closure_unix): Extend gp
2069 to 64 bits in ILP32 mode.
2070 Load 64 bits even for short data.
2072 2005-03-23 Mike Stump <mrs@apple.com>
2074 * src/powerpc/darwin.S: Update for -m64 multilib.
2075 * src/powerpc/darwin_closure.S: Likewise.
2077 2005-03-21 Zack Weinberg <zack@codesourcery.com>
2079 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
2080 Do not set tool_include_dir.
2081 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
2083 * include/Makefile.am: Set gcc_version and toollibffidir.
2084 * include/Makefile.in: Regenerate.
2086 2005-02-22 Andrew Haley <aph@redhat.com>
2088 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2089 odd-numbered register pairs for 64-bit integer types.
2091 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2094 * testsuite/libffi.call/return_ll1.c: New test case.
2096 2005-02-11 Janis Johnson <janis187@us.ibm.com>
2098 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2099 * testsuite/libffi.call/float.c: Ditto.
2100 * testsuite/libffi.call/float2.c: Ditto.
2101 * testsuite/libffi.call/float3.c: Ditto.
2103 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2105 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2107 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2109 * testsuite/libffi.special/special.exp (cxx_options): Add
2112 2004-12-31 Richard Henderson <rth@redhat.com>
2114 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2115 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2116 offset parameters with a type parameter; deduce size and structure
2117 alignment. Update all users.
2119 2004-12-31 Richard Henderson <rth@redhat.com>
2121 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2122 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
2123 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
2124 into ffi_prep_closure.
2125 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2128 2004-12-27 Richard Henderson <rth@redhat.com>
2130 * src/x86/unix64.S: Fix typo in unwind info.
2132 2004-12-25 Richard Henderson <rth@redhat.com>
2134 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2135 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2136 (merge_classes): Check for it.
2138 (classify_argument): Pass byte_offset by value; perform all updates
2140 (examine_argument): Add classes argument; handle
2141 X86_64_COMPLEX_X87_CLASS.
2142 (ffi_prep_args): Merge into ...
2143 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2144 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2145 (ffi_fill_return_value): Remove.
2146 (ffi_prep_closure): Remove dead assert.
2147 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2148 Rewrite to use struct register_args instead of va_list. Create
2149 flags for handling structure returns.
2150 * src/x86/unix64.S: Remove dead strings.
2151 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2152 stack frame with ffi_call. Handle structure returns properly.
2153 (float2sse, floatfloat2sse, double2sse): Remove.
2154 (sse2float, sse2double, sse2floatfloat): Remove.
2155 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2156 to handle structure returns properly.
2158 2004-12-08 David Edelsohn <edelsohn@gnu.org>
2160 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2162 * Makefile.in: Regenerated.
2164 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2166 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2167 * configure, aclocal.m4, Makefile.in: Regenerate.
2168 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2170 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2172 * configure: Regenerate for libtool change.
2174 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2176 * configure: Regenerate for libtool reversion.
2178 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2180 * configure: Regenerate for libtool change.
2182 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2184 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2186 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2188 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2189 of jal. Use an absolute encoding for the frame information.
2191 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2193 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2194 * acinclude.m4: Delete logic for sincludes.
2195 * aclocal.m4, Makefile.in, configure: Regenerate.
2196 * include/Makefile: Likewise.
2197 * testsuite/Makefile: Likewise.
2199 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2201 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2202 on a 8-byte boundary.
2203 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2205 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2207 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2208 long long values. Round stack allocation to a multiple of 8 bytes
2209 for ATPCS compatibility.
2210 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2211 names. Handle returning long long types. Add Thumb and interworking
2212 support. Improve soft-float code.
2214 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2216 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2217 (libffi_exit): New function.
2218 (libffi_init): Build the testglue wrapper if needed.
2220 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2223 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2225 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2227 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2229 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2231 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2232 * testsuite/libffi.call/float3.c: New test case.
2234 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2236 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2237 the function returning a structure pointed with R2.
2238 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2239 the structure return value if T bit set. Emit position
2240 independent code and EH data if PIC.
2242 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2244 * Makefile.am: Add m32r support.
2245 * configure.ac: Likewise.
2246 * Makefile.in: Regenerate.
2247 * confiugre: Regenerate.
2248 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2249 (uint64, sint64, double, longdouble)
2250 * src/m32r: New directory.
2251 * src/m32r/ffi.c: New file.
2252 * src/m32r/sysv.S: Likewise.
2253 * src/m32r/ffitarget.h: Likewise.
2255 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2257 * testsuite/libffi.call/negint.c: New test case.
2259 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2262 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2263 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2264 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2267 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2269 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2270 * testsuite/libffi.call/promotion.c: Likewise.
2271 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2273 * testsuite/libffi.call/return_sc.c: Likewise.
2274 * testsuite/libffi.call/return_uc.c: Likewise.
2276 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2278 * src/powerpc/darwin.S: Fix comments and identation.
2279 * src/powerpc/darwin_closure.S: Likewise.
2281 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2283 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2284 (ffi_prep_args): Handle longdouble arguments.
2285 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2287 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2288 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2290 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2291 * src/types.c: Defined longdouble size and alignment for darwin.
2293 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2295 * src/powerpc/aix.S: Remove whitespaces.
2296 * src/powerpc/aix_closure.S: Likewise.
2297 * src/powerpc/asm.h: Likewise.
2298 * src/powerpc/ffi.c: Likewise.
2299 * src/powerpc/ffitarget.h: Likewise.
2300 * src/powerpc/linux64.S: Likewise.
2301 * src/powerpc/linux64_closure.S: Likewise.
2302 * src/powerpc/ppc_closure.S: Likewise.
2303 * src/powerpc/sysv.S: Likewise.
2305 2004-08-30 Anthony Green <green@redhat.com>
2307 * Makefile.am: Add frv support.
2308 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2309 * configure.ac: Read configure.host.
2310 * configure.in: Read configure.host.
2311 * configure.host: New file. frv-elf needs libgloss.
2312 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2313 alignment. This is needed to frv and shouldn't harm the others.
2314 * include/ffi_common.h (ALIGN_DOWN): New macro.
2315 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2317 2004-08-24 David Daney <daney@avtrex.com>
2319 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2320 * testsuite/libffi.call/closure_fn1.c: Likewise.
2321 * testsuite/libffi.call/closure_fn2.c Likewise.
2322 * testsuite/libffi.call/closure_fn3.c: Likewise.
2323 * testsuite/libffi.call/closure_fn4.c: Likewise.
2324 * testsuite/libffi.call/closure_fn5.c: Likewise.
2325 * testsuite/libffi.call/cls_18byte.c: Likewise.
2326 * testsuite/libffi.call/cls_19byte.c: Likewise.
2327 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2328 * testsuite/libffi.call/cls_20byte.c: Likewise.
2329 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2330 * testsuite/libffi.call/cls_24byte.c: Likewise.
2331 * testsuite/libffi.call/cls_2byte.c: Likewise.
2332 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2333 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2334 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2335 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2336 * testsuite/libffi.call/cls_4byte.c: Likewise.
2337 * testsuite/libffi.call/cls_64byte.c: Likewise.
2338 * testsuite/libffi.call/cls_6byte.c: Likewise.
2339 * testsuite/libffi.call/cls_7byte.c: Likewise.
2340 * testsuite/libffi.call/cls_8byte.c: Likewise.
2341 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2342 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2343 * testsuite/libffi.call/cls_align_double.c: Likewise.
2344 * testsuite/libffi.call/cls_align_float.c: Likewise.
2345 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2346 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2347 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2348 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2349 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2350 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2351 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2352 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2353 * testsuite/libffi.call/cls_double.c: Likewise.
2354 * testsuite/libffi.call/cls_float.c: Likewise.
2355 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2356 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2357 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2358 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2359 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2360 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2361 * testsuite/libffi.call/cls_schar.c: Likewise.
2362 * testsuite/libffi.call/cls_sint.c: Likewise.
2363 * testsuite/libffi.call/cls_sshort.c: Likewise.
2364 * testsuite/libffi.call/cls_uchar.c: Likewise.
2365 * testsuite/libffi.call/cls_uint.c: Likewise.
2366 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2367 * testsuite/libffi.call/cls_ushort.c: Likewise.
2368 * testsuite/libffi.call/nested_struct.c: Likewise.
2369 * testsuite/libffi.call/nested_struct1.c: Likewise.
2370 * testsuite/libffi.call/nested_struct2.c: Likewise.
2371 * testsuite/libffi.call/nested_struct3.c: Likewise.
2372 * testsuite/libffi.call/problem1.c: Likewise.
2373 * testsuite/libffi.special/unwindtest.cc: Likewise.
2374 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2376 * testsuite/libffi.call/cls_16byte.c: Likewise.
2377 * testsuite/libffi.call/cls_5byte.c: Likewise.
2379 2004-08-23 David Daney <daney@avtrex.com>
2382 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2383 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2384 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2385 parameters and return types.
2386 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2387 (ffi_prep_closure): Ditto.
2388 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2389 alignment calculations.
2390 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2391 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2393 2004-08-14 Casey Marshall <csm@gnu.org>
2395 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2396 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2397 integer (O32 ABI only).
2398 (ffi_prep_closure): new function.
2399 (ffi_closure_mips_inner_O32): new function.
2400 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2401 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2402 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2403 64 bit integers correctly.
2404 (ffi_closure_O32): new function.
2405 Added DWARF-2 unwind info for both functions.
2407 2004-08-10 Andrew Haley <aph@redhat.com>
2409 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2411 2004-08-01 Robert Millan <robertmh@gnu.org>
2413 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2414 * configure: Regenerate.
2416 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2418 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2419 and mmap() explicitly instead of relying on preset autoconf cache
2421 * aclocal.m4: Regenerate.
2422 * configure: Regenerate.
2424 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2426 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2427 (ffi_check_float_struct): Remove unused prototype.
2429 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2431 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2432 character on Darwin, use '\n\t' instead.
2434 2004-06-26 Matthias Klose <doko@debian.org>
2436 * libtool-version: Fix typo in revision/age.
2438 2004-06-17 Matthias Klose <doko@debian.org>
2440 * libtool-version: New.
2441 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2442 * Makefile.in: Regenerate.
2444 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2446 * Makefile.am: Remove useless multilib rules.
2447 * Makefile.in: Regenerate.
2448 * aclocal.m4: Regenerate with automake 1.8.5.
2449 * configure.ac: Remove useless multilib configury.
2450 * configure: Regenerate.
2452 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2454 * .cvsignore: New file.
2456 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2458 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2460 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2463 2004-06-06 Sean McNeil <sean@mcneil.com>
2465 * configure.ac: Add x86_64-*-freebsd* support.
2466 * configure: Regenerate.
2468 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2471 * configure.ac: Test for existence of mmap and sys/mman.h before
2472 checking blacklist. Fix suggested by Jim Wilson.
2473 * configure: Regenerate.
2475 2004-04-26 Matt Austern <austern@apple.com>
2477 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2479 * src/powerpc/darwin_closure.S: Likewise.
2481 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2483 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2484 error. Reported by Thomas Heller <theller@python.net>.
2485 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2486 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2488 2004-03-20 Matthias Klose <doko@debian.org>
2490 * src/pa/linux.S: Fix typo.
2492 2004-03-19 Matthias Klose <doko@debian.org>
2494 * Makefile.am: Update.
2495 * Makefile.in: Regenerate.
2496 * src/pa/ffi.h.in: Remove.
2497 * src/pa/ffitarget.h: New file.
2499 2004-02-10 Randolph Chung <tausq@debian.org>
2501 * Makefile.am: Add PA support.
2502 * Makefile.in: Regenerate.
2503 * include/Makefile.in: Regenerate.
2504 * configure.ac: Add PA target.
2505 * configure: Regenerate.
2506 * src/pa/ffi.c: New file.
2507 * src/pa/ffi.h.in: Add PA support.
2508 * src/pa/linux.S: New file.
2509 * prep_cif.c: Add PA support.
2511 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2513 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2515 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2516 with ecif->cif->flags.
2517 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2519 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2520 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2521 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2522 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2524 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2526 * configure.in: Rename file to ...
2527 * configure.ac: ... this.
2528 * fficonfig.h.in: Regenerate.
2529 * Makefile.in: Regenerate.
2530 * include/Makefile.in: Regenerate.
2531 * testsuite/Makefile.in: Regenerate.
2533 2004-03-12 Matt Austern <austern@apple.com>
2535 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2536 changes in EH format resulting from addition of linkonce support.
2537 * src/powerpc/darwin_closure.S: Likewise.
2539 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2540 Paolo Bonzini <bonzini@gnu.org>
2542 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2543 Remove VPATH. Remove rules for object files. Remove multilib support.
2544 (AM_CCASFLAGS): Add.
2545 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2546 (AC_PREREQ): Bump version to 2.59.
2547 (AC_INIT): Fill with version info and bug address.
2548 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2549 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2550 De-precious CC so that the right flags are passed down to multilibs.
2551 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2552 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2553 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2554 * configure: Rebuilt.
2555 * aclocal.m4: Likewise.
2556 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2557 * fficonfig.h.in: Likewise.
2559 2004-03-11 Andreas Schwab <schwab@suse.de>
2561 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2562 arguments from fp registers only for the first 8 parameter slots.
2563 Don't convert a float parameter when passed in memory.
2565 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2567 * configure: Regenerate for config/accross.m4 correction.
2569 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2571 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2572 ecif->cif->bytes to bytes.
2573 (ffi_prep_cif_machdep): Add braces around nested if statement.
2575 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2577 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2579 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2580 (ffi_closure_helper_LINUX64): Fix typo.
2581 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2583 * testsuite/libffi.call/float.c: Likewise.
2584 * testsuite/libffi.call/float2.c: Likewise.
2586 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2588 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2589 long double function return and long double arg handling.
2590 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2591 Use "end_pfr" instead of "nf". Correct long double handling.
2593 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2595 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2596 space for long double return value. Adjust stack frame and offsets.
2597 Load f2 long double return.
2599 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2601 * src/types.c: Use 16 byte long double for POWERPC64.
2603 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2605 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2606 when the structure return address is passed in %o0.
2607 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2608 (ffi_v9_layout_struct): Align the field following a nested structure
2609 on a word boundary. Use memmove instead of memcpy.
2610 (ffi_call): Update call to ffi_V9_return_struct.
2611 (ffi_prep_closure): Define 'ctx' only for V8.
2612 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2613 and ffi_closure_sparc_inner_v9.
2614 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2615 Always skip the structure return address. For structures and long
2616 doubles, copy the argument directly.
2617 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2618 if required. Shift the maximum floating-point slot accordingly. For
2619 big structures, copy the argument directly; otherwise, left-justify the
2620 argument and call ffi_v9_layout_struct to lay out the structure on
2622 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2623 (ffi_closure_v8): Pass the structure return address. Update call to
2624 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2625 Skip the 'unimp' insn when returning long doubles and structures.
2626 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2627 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2628 FFI_TYPE_INT handling. Load structures both in integers and
2629 floating-point registers on return.
2630 * README: Update status of the SPARC port.
2632 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2634 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2636 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2638 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2640 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2641 value as of type ffi_arg, not unsigned int.
2643 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2645 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2648 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2650 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2653 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2655 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2658 2003-12-10 Richard Henderson <rth@redhat.com>
2660 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2661 size_t instead of int.
2663 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2665 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2666 * testsuite/libffi.call/many_win32.c (main): Replace variable
2667 int i with unsigned long ul.
2669 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2670 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2671 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2672 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2673 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2674 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2675 * testsuite/libffi.call/cls_align_float.c: Likewise.
2676 * testsuite/libffi.call/cls_align_double.c: Likewise.
2677 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2678 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2680 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2683 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2684 Align arguments to 32 bits.
2686 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2689 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2690 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2691 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2692 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2693 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2694 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2696 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2698 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2700 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2701 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2703 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2705 * Makefile.in: Rebuilt.
2706 * configure: Likewise.
2707 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2710 2003-11-21 Andreas Jaeger <aj@suse.de>
2711 Andreas Tobler <a.tobler@schweiz.ch>
2713 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2714 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2715 * Makefile.in: Rebuilt.
2716 * aclocal.m4: Likewise.
2717 * configure: Likewise.
2718 * fficonfig.h.in: Likewise.
2719 * testsuite/lib/libffi-dg.exp: Add include dir.
2720 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2721 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2722 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2723 for ffi_closure if available.
2724 * testsuite/libffi.call/closure_fn1.c: Likewise.
2725 * testsuite/libffi.call/closure_fn2.c: Likewise.
2726 * testsuite/libffi.call/closure_fn3.c: Likewise.
2727 * testsuite/libffi.call/closure_fn4.c: Likewise.
2728 * testsuite/libffi.call/closure_fn5.c: Likewise.
2729 * testsuite/libffi.call/cls_12byte.c: Likewise.
2730 * testsuite/libffi.call/cls_16byte.c: Likewise.
2731 * testsuite/libffi.call/cls_18byte.c: Likewise.
2732 * testsuite/libffi.call/cls_19byte.c: Likewise.
2733 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2734 * testsuite/libffi.call/cls_20byte.c: Likewise.
2735 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2736 * testsuite/libffi.call/cls_24byte.c: Likewise.
2737 * testsuite/libffi.call/cls_2byte.c: Likewise.
2738 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2739 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2740 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2741 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2742 * testsuite/libffi.call/cls_4byte.c: Likewise.
2743 * testsuite/libffi.call/cls_5byte.c: Likewise.
2744 * testsuite/libffi.call/cls_64byte.c: Likewise.
2745 * testsuite/libffi.call/cls_6byte.c: Likewise.
2746 * testsuite/libffi.call/cls_7byte.c: Likewise.
2747 * testsuite/libffi.call/cls_8byte.c: Likewise.
2748 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2749 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2750 * testsuite/libffi.call/cls_double.c: Likewise.
2751 * testsuite/libffi.call/cls_float.c: Likewise.
2752 * testsuite/libffi.call/cls_schar.c: Likewise.
2753 * testsuite/libffi.call/cls_sint.c: Likewise.
2754 * testsuite/libffi.call/cls_sshort.c: Likewise.
2755 * testsuite/libffi.call/cls_uchar.c: Likewise.
2756 * testsuite/libffi.call/cls_uint.c: Likewise.
2757 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2758 * testsuite/libffi.call/cls_ushort.c: Likewise.
2759 * testsuite/libffi.call/nested_struct.c: Likewise.
2760 * testsuite/libffi.call/nested_struct1.c: Likewise.
2761 * testsuite/libffi.call/nested_struct2.c: Likewise.
2762 * testsuite/libffi.call/nested_struct3.c: Likewise.
2763 * testsuite/libffi.call/problem1.c: Likewise.
2764 * testsuite/libffi.special/unwindtest.cc: Likewise.
2766 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2768 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2770 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2772 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2773 Add -lgcc_s to additional flags.
2775 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2777 * configure.in, include/Makefile.am: PR libgcj/11147, install
2778 the ffitarget.h header file in a gcc versioned and target
2780 * configure: Regenerated.
2781 * Makefile.in, include/Makefile.in: Likewise.
2782 * testsuite/Makefile.in: Likewise.
2784 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2786 * testsuite/libffi.call/closure_fn0.c: Print result and check
2787 with dg-output to make debugging easier.
2788 * testsuite/libffi.call/closure_fn1.c: Likewise.
2789 * testsuite/libffi.call/closure_fn2.c: Likewise.
2790 * testsuite/libffi.call/closure_fn3.c: Likewise.
2791 * testsuite/libffi.call/closure_fn4.c: Likewise.
2792 * testsuite/libffi.call/closure_fn5.c: Likewise.
2793 * testsuite/libffi.call/cls_12byte.c: Likewise.
2794 * testsuite/libffi.call/cls_16byte.c: Likewise.
2795 * testsuite/libffi.call/cls_18byte.c: Likewise.
2796 * testsuite/libffi.call/cls_19byte.c: Likewise.
2797 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2798 * testsuite/libffi.call/cls_20byte.c: Likewise.
2799 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2800 * testsuite/libffi.call/cls_24byte.c: Likewise.
2801 * testsuite/libffi.call/cls_2byte.c: Likewise.
2802 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2803 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2804 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2805 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2806 * testsuite/libffi.call/cls_4byte.c: Likewise.
2807 * testsuite/libffi.call/cls_5byte.c: Likewise.
2808 * testsuite/libffi.call/cls_64byte.c: Likewise.
2809 * testsuite/libffi.call/cls_6byte.c: Likewise.
2810 * testsuite/libffi.call/cls_7byte.c: Likewise.
2811 * testsuite/libffi.call/cls_8byte.c: Likewise.
2812 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2813 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2814 * testsuite/libffi.call/cls_double.c: Likewise.
2815 * testsuite/libffi.call/cls_float.c: Likewise.
2816 * testsuite/libffi.call/cls_schar.c: Likewise.
2817 * testsuite/libffi.call/cls_sint.c: Likewise.
2818 * testsuite/libffi.call/cls_sshort.c: Likewise.
2819 * testsuite/libffi.call/cls_uchar.c: Likewise.
2820 * testsuite/libffi.call/cls_uint.c: Likewise.
2821 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2822 * testsuite/libffi.call/cls_ushort.c: Likewise.
2823 * testsuite/libffi.call/problem1.c: Likewise.
2825 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2828 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2830 * testsuite/libffi.call/cls_9byte2.c: New test case.
2831 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2832 * testsuite/libffi.call/cls_64byte.c: Likewise.
2833 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2834 * testsuite/libffi.call/cls_19byte.c: Likewise.
2835 * testsuite/libffi.call/cls_18byte.c: Likewise.
2836 * testsuite/libffi.call/closure_fn4.c: Likewise.
2837 * testsuite/libffi.call/closure_fn5.c: Likewise.
2838 * testsuite/libffi.call/cls_schar.c: Likewise.
2839 * testsuite/libffi.call/cls_sint.c: Likewise.
2840 * testsuite/libffi.call/cls_sshort.c: Likewise.
2841 * testsuite/libffi.call/nested_struct2.c: Likewise.
2842 * testsuite/libffi.call/nested_struct3.c: Likewise.
2844 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2846 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2847 * testsuite/libffi.call/cls_uchar.c: Likewise.
2848 * testsuite/libffi.call/cls_uint.c: Likewise.
2849 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2850 * testsuite/libffi.call/cls_ushort.c: Likewise.
2851 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2853 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2855 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2858 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2860 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2861 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2863 2003-10-22 David Daney <ddaney@avtrex.com>
2865 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2866 __attribute__((__mode__(__SI__))) and friends.
2868 2003-10-22 Andreas Schwab <schwab@suse.de>
2870 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2872 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2874 * configure.in: AC_LINK_FILES(ffitarget.h).
2875 * configure: Regenerate.
2876 * Makefile.in: Likewise.
2877 * include/Makefile.in: Likewise.
2878 * testsuite/Makefile.in: Likewise.
2879 * fficonfig.h.in: Likewise.
2881 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2882 Richard Henderson <rth@redhat.com>
2884 Avoid that ffi.h includes fficonfig.h.
2886 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2887 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2888 (TARGET_SRC_MIPS_SGI): Removed.
2889 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2890 (MIPS_SGI): Removed.
2891 (CLEANFILES): Removed.
2892 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2894 * acconfig.h: Removed.
2895 * configure.in: Compute sizeofs only for double and long double.
2896 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2897 into AC_DEFINE instead of using acconfig.h. Create
2898 include/ffitarget.h instead of include/fficonfig.h. Rename
2899 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2900 AC_DEFINE EH_FRAME_FLAGS.
2901 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2902 (hack_DATA): Add ffitarget.h.
2903 * include/ffi.h.in: Remove all system specific definitions.
2904 Declare raw API even if it is not installed, why bother?
2905 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2906 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2907 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2908 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2909 * include/ffi_common.h (bool): Do not define.
2910 (ffi_assert): Accept failed assertion.
2911 (ffi_type_test): Return void and accept file/line.
2912 (FFI_ASSERT): Pass stringized failed assertion.
2913 (FFI_ASSERT_AT): New macro.
2914 (FFI_ASSERT_VALID_TYPE): New macro.
2915 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2916 UINT64, SINT64): Define here with gcc's __attribute__ macro
2918 (FLOAT32, ALIGN): Define here instead of in ffi.h
2919 * include/ffi-mips.h: Removed. Its content moved to
2920 src/mips/ffitarget.h after separating assembly and C sections.
2921 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2922 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2923 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2924 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2925 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2926 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2927 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2928 (ffi_type_test): Rewritten.
2929 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2930 FFI_ASSERT_VALID_TYPE.
2931 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2932 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2933 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2934 src/s390/ffitarget.h, src/sh/ffitarget.h,
2935 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2936 src/x86/ffitarget.h: New files.
2937 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2938 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2939 src/powerpc/aix.S, src/powerpc/darwin.S,
2940 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2941 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2942 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2943 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2944 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2947 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2949 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2950 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2952 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2954 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2955 Used when FFI_DEBUG = 1.
2957 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2959 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2962 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2964 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2965 FFI_MIPS_O32 for O32 ABI.
2967 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2969 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2970 SPARC64. Cleanup whitespaces.
2972 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2974 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2975 strongarm, xscale. Cleanup whitespaces.
2976 * testsuite/libffi.call/closure_fn1.c: Likewise.
2977 * testsuite/libffi.call/closure_fn2.c: Likewise.
2978 * testsuite/libffi.call/closure_fn3.c: Likewise.
2979 * testsuite/libffi.call/cls_12byte.c: Likewise.
2980 * testsuite/libffi.call/cls_16byte.c: Likewise.
2981 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2982 * testsuite/libffi.call/cls_20byte.c: Likewise.
2983 * testsuite/libffi.call/cls_24byte.c: Likewise.
2984 * testsuite/libffi.call/cls_2byte.c: Likewise.
2985 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2986 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2987 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2988 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2989 * testsuite/libffi.call/cls_4byte.c: Likewise.
2990 * testsuite/libffi.call/cls_5byte.c: Likewise.
2991 * testsuite/libffi.call/cls_6byte.c: Likewise.
2992 * testsuite/libffi.call/cls_7byte.c: Likewise.
2993 * testsuite/libffi.call/cls_8byte.c: Likewise.
2994 * testsuite/libffi.call/cls_double.c: Likewise.
2995 * testsuite/libffi.call/cls_float.c: Likewise.
2996 * testsuite/libffi.call/cls_uchar.c: Likewise.
2997 * testsuite/libffi.call/cls_uint.c: Likewise.
2998 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2999 * testsuite/libffi.call/cls_ushort.c: Likewise.
3000 * testsuite/libffi.call/nested_struct.c: Likewise.
3001 * testsuite/libffi.call/nested_struct1.c: Likewise.
3002 * testsuite/libffi.call/problem1.c: Likewise.
3003 * testsuite/libffi.special/unwindtest.cc: Likewise.
3004 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
3006 2003-09-18 David Edelsohn <edelsohn@gnu.org>
3008 * src/powerpc/aix.S: Cleanup whitespaces.
3009 * src/powerpc/aix_closure.S: Likewise.
3011 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3013 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
3014 * src/powerpc/darwin_closure.S: Likewise.
3015 * src/powerpc/ffi_darwin.c: Likewise.
3017 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3018 David Edelsohn <edelsohn@gnu.org>
3020 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
3021 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
3023 * src/powerpc/darwin_closure.S: Likewise.
3024 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
3025 according to the Darwin/AIX ABI.
3026 (ffi_prep_cif_machdep): Likewise.
3027 (ffi_closure_helper_DARWIN): Likewise.
3028 Remove the outgoing parameter stack logic. Simplify the evaluation
3029 of the different CASE types.
3030 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
3031 statement in the trampoline code.
3033 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
3035 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
3036 for the register size.
3037 (ffi_closure_helper_SYSV): Handle the structure return value
3039 (ffi_closure_helper_SYSV): Return the appropriate type when
3040 the registers are used for the structure return value.
3041 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
3042 the 64-bit return value. Update copyright years.
3044 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3046 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
3047 srcdir for ffi_mips.h.
3049 2003-09-12 Alan Modra <amodra@bigpond.net.au>
3051 * src/prep_cif.c (initialize_aggregate): Include tail padding in
3053 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
3054 placement of float result.
3055 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
3056 cast of "resp" for big-endian 64 bit machines.
3058 2003-09-11 Alan Modra <amodra@bigpond.net.au>
3060 * src/types.c (double, longdouble): Merge identical SH and ARM
3061 typedefs, and add POWERPC64.
3062 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
3063 struct split over gpr and rest.
3064 (ffi_prep_cif_machdep): Correct intarg_count for structures.
3065 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
3067 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
3069 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
3072 2003-09-09 Alan Modra <amodra@bigpond.net.au>
3074 * configure: Regenerate.
3076 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3078 * Makefile.am: Remove build rules for ffitest.
3079 * Makefile.in: Rebuilt.
3081 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3083 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3084 about implicit declaration of abort().
3086 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3088 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3089 * Makefile.in: Rebuilt.
3090 * configure.in: Add dejagnu test framework.
3091 * configure: Rebuilt.
3093 * testsuite/Makefile.am: New file.
3094 * testsuite/Makefile.in: Built
3095 * testsuite/lib/libffi-dg.exp: New file.
3096 * testsuite/config/default.exp: Likewise.
3097 * testsuite/libffi.call/call.exp: Likewise.
3098 * testsuite/libffi.call/ffitest.h: Likewise.
3099 * testsuite/libffi.call/closure_fn0.c: Likewise.
3100 * testsuite/libffi.call/closure_fn1.c: Likewise.
3101 * testsuite/libffi.call/closure_fn2.c: Likewise.
3102 * testsuite/libffi.call/closure_fn3.c: Likewise.
3103 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3104 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3105 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3106 * testsuite/libffi.call/cls_2byte.c: Likewise.
3107 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3108 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3109 * testsuite/libffi.call/cls_4byte.c: Likewise.
3110 * testsuite/libffi.call/cls_5byte.c: Likewise.
3111 * testsuite/libffi.call/cls_6byte.c: Likewise.
3112 * testsuite/libffi.call/cls_7byte.c: Likewise.
3113 * testsuite/libffi.call/cls_8byte.c: Likewise.
3114 * testsuite/libffi.call/cls_12byte.c: Likewise.
3115 * testsuite/libffi.call/cls_16byte.c: Likewise.
3116 * testsuite/libffi.call/cls_20byte.c: Likewise.
3117 * testsuite/libffi.call/cls_24byte.c: Likewise.
3118 * testsuite/libffi.call/cls_double.c: Likewise.
3119 * testsuite/libffi.call/cls_float.c: Likewise.
3120 * testsuite/libffi.call/cls_uchar.c: Likewise.
3121 * testsuite/libffi.call/cls_uint.c: Likewise.
3122 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3123 * testsuite/libffi.call/cls_ushort.c: Likewise.
3124 * testsuite/libffi.call/float.c: Likewise.
3125 * testsuite/libffi.call/float1.c: Likewise.
3126 * testsuite/libffi.call/float2.c: Likewise.
3127 * testsuite/libffi.call/many.c: Likewise.
3128 * testsuite/libffi.call/many_win32.c: Likewise.
3129 * testsuite/libffi.call/nested_struct.c: Likewise.
3130 * testsuite/libffi.call/nested_struct1.c: Likewise.
3131 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3132 * testsuite/libffi.call/problem1.c: Likewise.
3133 * testsuite/libffi.call/promotion.c: Likewise.
3134 * testsuite/libffi.call/return_ll.c: Likewise.
3135 * testsuite/libffi.call/return_sc.c: Likewise.
3136 * testsuite/libffi.call/return_uc.c: Likewise.
3137 * testsuite/libffi.call/strlen.c: Likewise.
3138 * testsuite/libffi.call/strlen_win32.c: Likewise.
3139 * testsuite/libffi.call/struct1.c: Likewise.
3140 * testsuite/libffi.call/struct2.c: Likewise.
3141 * testsuite/libffi.call/struct3.c: Likewise.
3142 * testsuite/libffi.call/struct4.c: Likewise.
3143 * testsuite/libffi.call/struct5.c: Likewise.
3144 * testsuite/libffi.call/struct6.c: Likewise.
3145 * testsuite/libffi.call/struct7.c: Likewise.
3146 * testsuite/libffi.call/struct8.c: Likewise.
3147 * testsuite/libffi.call/struct9.c: Likewise.
3148 * testsuite/libffi.special/special.exp: New file.
3149 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3150 * testsuite/libffi.special/unwindtest.cc: Likewise.
3153 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3155 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3158 2003-08-02 Alan Modra <amodra@bigpond.net.au>
3160 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3162 (ffi_closure_helper_LINUX64): Likewise.
3163 * src/powerpc/linux64.S: Remove code writing to parm save area.
3164 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3165 address in lr from ffi_closure_helper_LINUX64 call to calculate
3166 table address. Optimize function tail.
3168 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3170 * src/sparc/ffi.c: Handle all floating point registers.
3171 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3173 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3175 * README: Note that libffi is not part of GCC. Update the project
3178 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3180 * src/powerpc/ppc_closure.S: Include ffi.h.
3182 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3184 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3185 Use C style comments.
3187 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3189 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3190 * Makefile.in: Regenerate.
3191 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3192 * configure: Regenerate.
3193 * include/ffi.h.in: Add SHmedia support.
3194 * src/sh64/ffi.c: New file.
3195 * src/sh64/sysv.S: New file.
3197 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3199 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3200 should be read-only.
3201 * configure: Rebuilt.
3202 * fficonfig.h.in: Rebuilt.
3203 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3204 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3205 * src/powerpc/linux64.S: Likewise.
3206 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3207 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3208 if -fpic/-fPIC/-mrelocatable.
3209 * src/powerpc/powerpc_closure.S: Likewise.
3210 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3211 #write in .eh_frame flags.
3212 * src/sparc/v9.S: Likewise.
3213 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3214 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3215 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3217 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3219 Fixes PR bootstrap/10656
3220 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3221 support for .register pseudo-op.
3222 * src/sparc/v8.S: Use it.
3223 * fficonfig.h.in: Rebuilt.
3224 * configure: Rebuilt.
3226 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3228 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3229 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3230 Make it the default on POWERPC64.
3231 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3232 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3233 * configure: Rebuilt.
3234 * src/powerpc/ffi.c (hidden): Define.
3235 (ffi_prep_args_SYSV): Renamed from
3236 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3237 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3238 ASM_NEEDS_REGISTERS64): New.
3239 (ffi_prep_args64): New function.
3240 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3241 (ffi_call): Likewise.
3242 (ffi_prep_closure): Likewise.
3243 (flush_icache): Surround by #ifndef POWERPC64.
3244 (ffi_dblfl): New union type.
3245 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3246 (ffi_closure_helper_LINUX64): New function.
3247 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3249 * src/powerpc/sysv.S: Likewise.
3250 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3251 * src/powerpc/linux64.S: New file.
3252 * src/powerpc/linux64_closure.S: New file.
3253 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3254 src/powerpc/linux64_closure.S.
3255 (TARGET_SRC_POWERPC): Likewise.
3257 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3258 closure_test_fn3): Fix result printing on big-endian 64-bit
3260 (main): Print tst2_arg instead of uninitialized tst2_result.
3262 * src/ffitest.c (main): Hide what closure pointer really points to
3265 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3267 * configure.in (arm-*-netbsdelf*): Add configuration.
3268 (configure): Regenerated.
3270 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3272 * include/Makefile.in: Regenerate.
3274 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3276 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3278 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3279 Receive closure pointer through parameter, read args using
3280 __builtin_dwarf_cfa.
3281 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3283 2003-03-12 Andreas Schwab <schwab@suse.de>
3285 * configure.in: Avoid trailing /. in toolexeclibdir.
3286 * configure: Rebuilt.
3288 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3290 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3292 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3294 * libffi/src/powerpc/darwin_closure.S:
3295 Fix alignement bug, allocate 8 bytes for the result.
3296 * libffi/src/powerpc/aix_closure.S:
3298 * libffi/src/powerpc/ffi_darwin.c:
3299 Update stackframe description for aix/darwin_closure.S.
3301 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3303 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3306 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3307 Andreas Schwab <schwab@suse.de>
3309 * configure.in: Adjust command to source config-ml.in to account
3310 for changes to the libffi_basedir definition.
3311 (libffi_basedir): Remove ${srcdir} from value and include trailing
3314 * configure: Regenerate.
3316 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3318 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3320 2003-01-28 Andrew Haley <aph@redhat.com>
3322 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3323 * src/x86/ffi64.c (ffi_prep_closure): New.
3324 (ffi_closure_UNIX64_inner): New.
3325 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3327 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
3329 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3330 Remove USE_LIBDIR conditional.
3331 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3332 * Makefile.in, configure: Rebuilt.
3334 2003-01027 David Edelsohn <edelsohn@gnu.org>
3336 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3337 * Makefile.in: Regenerate.
3339 2003-01-22 Andrew Haley <aph@redhat.com>
3341 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3344 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3346 * src/powerpc/darwin.S: Add unwind info.
3347 * src/powerpc/darwin_closure.S: Likewise.
3349 2003-01-14 Andrew Haley <aph@redhat.com>
3351 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3352 (ffi_prep_cif_machdep): Likewise.
3353 * src/x86/unix64.S: Add unwind info.
3355 2003-01-14 Andreas Jaeger <aj@suse.de>
3357 * src/ffitest.c (main): Only use ffi_closures if those are
3360 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3362 * libffi/src/ffitest.c
3363 add closure testcases
3365 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3367 * libffi/src/powerpc/ffi.c
3368 fix alignment bug for float (4 byte aligned iso 8 byte)
3370 2003-01-09 Geoffrey Keating <geoffk@apple.com>
3372 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3373 * src/powerpc/darwin.S: Remove RCS version string.
3375 2003-01-03 Jeff Sturm <jsturm@one-point.com>
3377 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3378 * src/ffitest.c (main): Use static storage for closure.
3379 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3380 * src/sparc/v8.S (ffi_closure_v8): New.
3381 * src/sparc/v9.S (ffi_closure_v9): New.
3383 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3385 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3386 enumeration for X86_WIN32.
3387 * src/x86/win32.S: Added ffi_call_STDCALL function
3389 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3390 switch cases for recognising FFI_STDCALL and
3391 calling ffi_call_STDCALL if target is X86_WIN32.
3392 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3393 stdcall versions of the "my_strlen" and "many"
3394 test functions (for X86_WIN32).
3395 Added test cases to test stdcall invocation using
3398 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3400 * src/sh/sysv.S: Add DWARF2 unwind info.
3402 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3404 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3406 2002-11-26 Jim Wilson <wilson@redhat.com>
3408 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3410 2002-11-23 H.J. Lu <hjl@gnu.org>
3412 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3413 Include ../config/accross.m4.
3414 * aclocal.m4; Rebuild.
3415 * configure: Likewise.
3417 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3419 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3421 2002-11-11 DJ Delorie <dj@redhat.com>
3423 * configure.in: Look for common files in the right place.
3425 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3427 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3428 raw data as _Jv_word values, not ffi_raw.
3429 (ffi_java_ptrarray_to_raw): Likewise.
3430 (ffi_java_rvalue_to_raw): New function.
3431 (ffi_java_raw_call): Call it.
3432 (ffi_java_raw_to_rvalue): New function.
3433 (ffi_java_translate_args): Call it.
3434 * src/ffitest.c (closure_test_fn): Interpret return value
3435 as ffi_arg, not int.
3436 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3437 FFI_TYPE_POINTER case.
3438 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3439 values extended to word size.
3441 2002-10-02 Andreas Jaeger <aj@suse.de>
3443 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3445 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
3447 * include/ffi.h.in: Fix i386 win32 compilation.
3449 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3451 * configure.in: Add s390x-*-linux-* target.
3452 * configure: Regenerate.
3453 * include/ffi.h.in: Define S390X for s390x targets.
3454 (FFI_CLOSURES): Define for s390/s390x.
3455 (FFI_TRAMPOLINE_SIZE): Likewise.
3456 (FFI_NATIVE_RAW_API): Likewise.
3457 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3458 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
3459 * src/s390/ffi.c: Major rework of existing code. Add support for
3460 s390x targets. Add closure support.
3461 * src/s390/sysv.S: Likewise.
3463 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
3465 * src/arm/sysv.S: Fix typo.
3467 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
3469 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3470 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3471 (ffi_call_SYSV): Handle soft-float.
3473 2002-09-27 Bo Thorsen <bo@suse.de>
3475 * include/ffi.h.in: Fix multilib x86-64 support.
3477 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3479 * Makefile.am (all-multi): Fix multilib parallel build.
3481 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3483 * configure.in (sh[34]*-*-linux*): Add brackets.
3484 * configure: Regenerate.
3486 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3488 * Makefile.am: Add SH support.
3489 * Makefile.in: Regenerate.
3490 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3491 * configure: Regenerate.
3492 * include/ffi.h.in: Add SH support.
3493 * src/sh/ffi.c: New file.
3494 * src/sh/sysv.S: New file.
3495 * src/types.c: Add SH support.
3497 2002-07-16 Bo Thorsen <bo@suse.de>
3499 * src/x86/ffi64.c: New file that adds x86-64 support.
3500 * src/x86/unix64.S: New file that handles argument setup for
3502 * src/x86/sysv.S: Don't use this on x86-64.
3503 * src/x86/ffi.c: Don't use this on x86-64.
3505 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3507 * src/ffitest.c (struct6): New test that tests a special case in
3509 (struct7): Likewise.
3510 (struct8): Likewise.
3511 (struct9): Likewise.
3512 (closure_test_fn): Silence warning about this when it's not used.
3513 (main): Add the new tests.
3514 (main): Fix a couple of wrong casts and silence some compiler warnings.
3515 * include/ffi.h.in: Add x86-64 ABI definition.
3516 * fficonfig.h.in: Regenerate.
3517 * Makefile.am: Add x86-64 support.
3518 * configure.in: Likewise.
3519 * Makefile.in: Regenerate.
3520 * configure: Likewise.
3522 2002-06-24 Bo Thorsen <bo@suse.de>
3524 * src/types.c: Merge settings for similar architectures.
3525 Add x86-64 sizes and alignments.
3527 2002-06-23 Bo Thorsen <bo@suse.de>
3529 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3530 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3531 * src/mips/ffi.c (ffi_prep_args): Likewise.
3532 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3534 2002-07-18 H.J. Lu (hjl@gnu.org)
3536 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3537 (libffi_la_SOURCES): Support MIPS_LINUX.
3538 (libffi_convenience_la_SOURCES): Likewise.
3539 * Makefile.in: Regenerated.
3541 * configure.in (mips64*-*): Skip.
3542 (mips*-*-linux*): New.
3543 * configure: Regenerated.
3545 * src/mips/ffi.c: Include <sgidefs.h>.
3547 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3549 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3551 2002-05-27 Roger Sayle <roger@eyesopen.com>
3553 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3555 2002-05-27 Bo Thorsen <bo@suse.de>
3557 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3560 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3562 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3563 beginning of function (for older apple cc).
3565 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3567 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3568 script entry, and set LD to it when configuring multilibs.
3569 * configure: Rebuilt.
3571 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3573 * configure.in (sparc64-*-netbsd*): Add target.
3574 (sparc-*-netbsdelf*): Likewise.
3575 * configure: Regenerate.
3577 2002-04-28 David S. Miller <davem@redhat.com>
3579 * configure.in, configure: Fix SPARC test in previous change.
3581 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3583 * Makefile.am: Add Linux for S/390 support.
3584 * Makefile.in: Regenerate.
3585 * configure.in: Add Linux for S/390 support.
3586 * configure: Regenerate.
3587 * include/ffi.h.in: Add Linux for S/390 support.
3588 * src/s390/ffi.c: New file from libffi CVS tree.
3589 * src/s390/sysv.S: New file from libffi CVS tree.
3591 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3593 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3595 * src/sparc/v8.S: Use it.
3596 * src/sparc/v9.S: Likewise.
3597 * fficonfig.h.in: Rebuilt.
3598 * configure: Rebuilt.
3600 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3602 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3604 * src/ia64/unix.S: Add unwind information. Fix comments.
3605 Save sp in a way that's compatible with unwind info.
3606 (ffi_call_unix): Correctly restore sp in all cases.
3607 * src/ia64/ffi.c: Add, fix comments.
3609 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3611 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3613 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3615 * configure.in (alpha*-*-netbsd*): Add target.
3616 * configure: Regenerate.
3618 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3620 * src/sparc/v8.S: Add unwind info.
3621 * src/sparc/v9.S: Likewise.
3623 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3625 * configure.in: Enable i*86-*-netbsdelf*.
3626 * configure: Rebuilt.
3628 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3631 * src/mips/n32.s: Delete
3632 * src/mips/o32.s: Delete
3634 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3636 * configure.in: Enable alpha*-*-freebsd*.
3637 * configure: Rebuilt.
3639 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3641 * Makefile.am: libfficonvenience -> libffi_convenience.
3642 * Makefile.in: Rebuilt.
3644 * Makefile.am: Define ffitest_OBJECTS.
3645 * Makefile.in: Rebuilt.
3647 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3648 David Edelsohn <edelsohn@gnu.org>
3650 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3651 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3652 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3653 * Makefile.in: Regenerate.
3654 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3655 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3656 (flush_icache, flush_range): New functions.
3657 (ffi_closure_helper_DARWIN): New function.
3658 * src/powerpc/aix_closure.S: New file.
3659 * src/powerpc/darwin_closure.S: New file.
3661 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3663 * include/ffi.h.in: Add typedef for ffi_arg.
3664 * src/ffitest.c (main): Declare rint with ffi_arg.
3666 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3668 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3669 number of GPRs for floating-point arguments.
3671 2002-01-31 Anthony Green <green@redhat.com>
3673 * configure: Rebuilt.
3674 * configure.in: Replace CHECK_SIZEOF and endian tests with
3675 cross-compiler friendly macros.
3676 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3679 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3681 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3682 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3684 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3686 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3687 (TARGET_SRC_POWERPC_AIX): New.
3688 (POWERPC_AIX): New stanza.
3689 * Makefile.in: Regenerate.
3690 * configure.in: Add AIX case.
3691 * configure: Regenerate.
3692 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3693 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3694 size. Fix "long double" support.
3695 (ffi_call): Add FFI_AIX case.
3696 * src/powerpc/aix.S: New.
3698 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3700 Implement Darwin PowerPC ABI.
3701 * configure.in: Handle powerpc-*-darwin*.
3702 * Makefile.am: Set source files for POWERPC_DARWIN.
3703 * configure: Rebuilt.
3704 * Makefile.in: Rebuilt.
3705 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3707 * src/powerpc/darwin.S: New file.
3708 * src/powerpc/ffi_darwin.c: New file.
3710 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3712 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3714 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3716 * src/x86/sysv.S: Avoid gas-only .balign directive.
3717 Use C style comments.
3719 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3721 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3722 Fixes PR bootstrap/3563.
3724 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3726 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3728 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3730 * configure.in: Recognize sparc*-sun-* host.
3731 * configure: Regenerate.
3733 2001-06-06 Andrew Haley <aph@redhat.com>
3735 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
3737 2001-06-03 Andrew Haley <aph@redhat.com>
3739 * src/alpha/osf.S: Add unwind info.
3740 * src/powerpc/sysv.S: Add unwind info.
3741 * src/powerpc/ppc_closure.S: Likewise.
3743 2000-05-31 Jeff Sturm <jsturm@one-point.com>
3745 * configure.in: Fix AC_ARG_ENABLE usage.
3746 * configure: Rebuilt.
3748 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3750 * configure.in: Remove warning about beta code.
3751 * configure: Rebuilt.
3753 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3755 * src/ia64/unix.S: Restore stack pointer when returning from
3757 * src/ia64/ffi.c: Fix typo in comment.
3759 2001-04-18 Jim Wilson <wilson@redhat.com>
3761 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3762 to eliminate RAW DV.
3764 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3766 * Makefile.am: Make a libtool convenience library.
3767 * Makefile.in: Rebuilt.
3769 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3771 * configure.in: Use different syntax for subdirectory creation.
3772 * configure: Rebuilt.
3774 2001-03-27 Jon Beniston <jon@beniston.com>
3776 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3777 * configure: Rebuilt.
3778 * Makefile.am: Added X86_WIN32 target support.
3779 * Makefile.in: Rebuilt.
3781 * include/ffi.h.in: Added X86_WIN32 target support.
3783 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3784 * src/types.c: Added X86_WIN32 target support.
3786 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3787 stuff removed and made to work with CygWin's gas.
3789 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3791 * configure.in: Make target subdirectory in build dir.
3792 * Makefile.am: Override suffix based rules to specify correct output
3794 * Makefile.in: Rebuilt.
3795 * configure: Rebuilt.
3797 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3799 * src/powerpc/ppc_closure.S: New file.
3800 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3801 involving long long and register pairs.
3802 (ffi_prep_closure): New function.
3803 (flush_icache): Likewise.
3804 (ffi_closure_helper_SYSV): Likewise.
3805 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3806 (FFI_TRAMPOLINE_SIZE): Likewise.
3807 (FFI_NATIVE_RAW_API): Likewise.
3808 * Makefile.in: Rebuilt.
3809 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3810 (TARGET_SRC_POWERPC): Likewise.
3812 2001-03-19 Tom Tromey <tromey@redhat.com>
3814 * Makefile.in: Rebuilt.
3815 * Makefile.am (ffitest_LDFLAGS): New macro.
3817 2001-03-02 Nick Clifton <nickc@redhat.com>
3819 * include/ffi.h.in: Remove RCS ident string.
3820 * include/ffi_mips.h: Remove RCS ident string.
3821 * src/debug.c: Remove RCS ident string.
3822 * src/ffitest.c: Remove RCS ident string.
3823 * src/prep_cif.c: Remove RCS ident string.
3824 * src/types.c: Remove RCS ident string.
3825 * src/alpha/ffi.c: Remove RCS ident string.
3826 * src/alpha/osf.S: Remove RCS ident string.
3827 * src/arm/ffi.c: Remove RCS ident string.
3828 * src/arm/sysv.S: Remove RCS ident string.
3829 * src/mips/ffi.c: Remove RCS ident string.
3830 * src/mips/n32.S: Remove RCS ident string.
3831 * src/mips/o32.S: Remove RCS ident string.
3832 * src/sparc/ffi.c: Remove RCS ident string.
3833 * src/sparc/v8.S: Remove RCS ident string.
3834 * src/sparc/v9.S: Remove RCS ident string.
3835 * src/x86/ffi.c: Remove RCS ident string.
3836 * src/x86/sysv.S: Remove RCS ident string.
3838 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3840 * include/ffi.h.in: Change sourceware.cygnus.com references to
3843 2000-12-09 Richard Henderson <rth@redhat.com>
3845 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3846 (ffi_closure_osf_inner): Index rather than increment avalue
3847 and arg_types. Give ffi_closure_osf the raw return value type.
3848 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3851 2000-12-07 Richard Henderson <rth@redhat.com>
3853 * src/raw_api.c (ffi_translate_args): Fix typo.
3854 (ffi_prep_closure): Likewise.
3856 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3857 FFI_TRAMPOLINE_SIZE.
3858 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3859 cif->bytes for new ffi_call_osf implementation.
3860 (ffi_prep_args): Absorb into ...
3861 (ffi_call): ... here. Do all stack allocation here and
3862 avoid a callback function.
3863 (ffi_prep_closure, ffi_closure_osf_inner): New.
3864 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3865 (ffi_closure_osf): New.
3867 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3869 * config.guess, config.sub, install-sh: Removed.
3870 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3871 * Makefile.in: Rebuilt.
3873 * acinclude.m4: Include libtool macros from the top level.
3874 * aclocal.m4, configure: Rebuilt.
3876 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3878 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3879 * configure: Rebuilt.
3881 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3883 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3884 memory correctly. Use conditional instructions, not branches where
3887 2000-05-04 Tom Tromey <tromey@cygnus.com>
3889 * configure: Rebuilt.
3890 * configure.in: Match `arm*-*-linux-*'.
3891 From Chris Dornan <cdornan@arm.com>.
3893 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3895 * Makefile.am (SUBDIRS): Define.
3896 (AM_MAKEFLAGS): Likewise.
3897 (Multilib support.): Add section.
3898 * Makefile.in: Rebuilt.
3899 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3900 New variables. Set for gcc using -print-multi-lib. Export them
3902 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3903 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3905 (extra_compiler_flags_value, extra_compiler_flags): Check these
3906 for extra compiler options which need to be passed down in
3909 2000-04-16 Anthony Green <green@redhat.com>
3911 * configure: Rebuilt.
3912 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3914 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3916 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3917 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3918 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3919 Replace all void * sizeofs with sizeof(int).
3920 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3921 different than DOUBLE.
3922 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3923 (ffi_prep_args_v9): New function.
3924 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3925 (ffi_V9_return_struct): New function.
3926 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3927 32bit code (not yet cross-arch calls).
3928 * src/sparc/v8.S: Add struct return delay nop.
3930 * src/sparc/v9.S: New file.
3931 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3932 is used on sparc64 only for structures larger than 32 bytes.
3933 Pass by reference for structures is done for structure arguments
3934 larger than 16 bytes.
3935 * src/ffitest.c (main): Use 64bit rint on sparc64.
3936 Run long long tests on sparc.
3937 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3939 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3941 * configure.in (sparc-*-linux*): New supported target.
3942 (sparc64-*-linux*): Likewise.
3943 * configure: Rebuilt.
3944 * Makefile.am: Add v9.S to SPARC files.
3945 * Makefile.in: Likewise.
3946 (LINK): Surround $(CCLD) into double quotes, so that multilib
3947 compiles work correctly.
3949 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3951 * configure: Rebuilt.
3952 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3953 proposed by Bryce McKinlay.
3955 2000-03-20 Tom Tromey <tromey@cygnus.com>
3957 * Makefile.in: Hand edit for java_raw_api.lo.
3959 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3961 * config.guess, config.sub: Update from the gcc tree.
3962 Fix for PR libgcj/168.
3964 2000-03-03 Tom Tromey <tromey@cygnus.com>
3966 * Makefile.in: Fixed ia64 by hand.
3968 * configure: Rebuilt.
3969 * configure.in (--enable-multilib): New option.
3970 (libffi_basedir): New subst.
3971 (AC_OUTPUT): Added multilib code.
3973 2000-03-02 Tom Tromey <tromey@cygnus.com>
3975 * Makefile.in: Rebuilt.
3976 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3979 2000-02-25 Hans Boehm <boehm@acm.org>
3981 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3983 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3985 (ffi_prep_raw_closure): Use ffi_translate_args, not
3986 ffi_closure_translate.
3987 * src/java_raw_api.c: New file.
3988 * src/ffitest.c (closure_test_fn): New function.
3989 (main): Define `rint' as long long on IA64. Added new test when
3990 FFI_CLOSURES is defined.
3991 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3992 (ffi_abi): Recognize IA64.
3993 (ffi_raw): Added `flt' field.
3994 Added "Java raw API" code.
3995 * configure.in: Recognize ia64.
3996 * Makefile.am (TARGET_SRC_IA64): New macro.
3997 (libffi_la_common_SOURCES): Added java_raw_api.c.
3998 (libffi_la_SOURCES): Define in IA64 case.
4000 2000-01-04 Tom Tromey <tromey@cygnus.com>
4002 * Makefile.in: Rebuilt with newer automake.
4004 1999-12-31 Tom Tromey <tromey@cygnus.com>
4006 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
4008 1999-09-01 Tom Tromey <tromey@cygnus.com>
4010 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
4012 * fficonfig.h.in: Rebuilt.
4013 * configure: Rebuilt.
4014 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
4015 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
4016 * acconfig.h: Don't #undef PACKAGE or VERSION.
4018 1999-08-09 Anthony Green <green@cygnus.com>
4020 * include/ffi.h.in: Try to work around messy header problem
4021 with PACKAGE and VERSION.
4023 * configure: Rebuilt.
4024 * configure.in: Change version to 2.00-beta.
4026 * fficonfig.h.in: Rebuilt.
4027 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
4029 * src/x86/ffi.c (ffi_raw_call): Rename.
4031 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
4033 * src/x86/ffi.c (ffi_closure_SYSV): New function.
4034 (ffi_prep_incoming_args_SYSV): Ditto.
4035 (ffi_prep_closure): Ditto.
4036 (ffi_closure_raw_SYSV): Ditto.
4037 (ffi_prep_raw_closure): More ditto.
4038 (ffi_call_raw): Final ditto.
4040 * include/ffi.h.in: Add definitions for closure and raw API.
4042 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4045 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
4047 * src/raw_api.c: New file.
4049 * include/ffi.h.in (ffi_raw): New type.
4050 (UINT_ARG, SINT_ARG): New defines.
4051 (ffi_closure, ffi_raw_closure): New types.
4052 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4054 * configure.in: Add check for endianness and sizeof void*.
4056 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
4057 instead of directly.
4059 * configure: Rebuilt.
4061 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
4063 * configure.in: Add x86 and powerpc BeOS configurations.
4064 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
4066 1999-05-09 Anthony Green <green@cygnus.com>
4068 * configure.in: Add warning about this being beta code.
4069 Remove src/Makefile.am from the picture.
4070 * configure: Rebuilt.
4072 * Makefile.am: Move logic from src/Makefile.am. Add changes
4073 to support libffi as a target library.
4074 * Makefile.in: Rebuilt.
4076 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
4077 Upgraded to new autoconf, automake, libtool.
4081 * LICENSE: Update copyright date.
4083 * src/Makefile.am, src/Makefile.in: Removed.
4085 1998-11-29 Anthony Green <green@cygnus.com>
4087 * include/ChangeLog: Removed.
4088 * src/ChangeLog: Removed.
4089 * src/mips/ChangeLog: Removed.
4090 * src/sparc/ChangeLog: Remboved.
4091 * src/x86/ChangeLog: Removed.
4093 * ChangeLog.v1: Created.