1 2009-12-26 Anthony Green <green@redhat.com>
3 * testsuite/libffi.call/cls_longdouble_va.c: Mark as xfail for
5 * testsuite/libffi.call/cls_double_va.c: Ditto.
7 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
9 * testsuite/libffi.call/ffitest.h: Conditionally include stdint.h
11 * testsuite/libffi.special/unwindtest.cc: Ditto.
13 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
15 * configure.ac: Add amd64-*-openbsd*.
17 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Link
18 openbsd programs with -lpthread.
20 2009-12-26 Anthony Green <green@redhat.com>
22 * testsuite/libffi.call/cls_double_va.c,
23 testsuite/libffi.call/cls_longdouble.c,
24 testsuite/libffi.call/cls_longdouble_va.c,
25 testsuite/libffi.call/cls_pointer.c,
26 testsuite/libffi.call/cls_pointer_stack.c: Remove xfail for
27 mips*-*-* and arm*-*-*.
28 * testsuite/libffi.call/cls_align_longdouble_split.c,
29 testsuite/libffi.call/cls_align_longdouble_split2.c,
30 testsuite/libffi.call/stret_medium2.c,
31 testsuite/libffi.call/stret_medium.c,
32 testsuite/libffi.call/stret_large.c,
33 testsuite/libffi.call/stret_large2.c: Remove xfail for arm*-*-*.
35 2009-12-31 Kay Tietz <ktietz70@googlemail.com>
37 * testsuite/libffi.call/ffitest.h,
38 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRuLL): Fix
41 2009-12-31 Carlo Bramini <carlo.bramix@libero.it>
43 * configure.ac (AM_LTLDFLAGS): Define for windows hosts.
44 * Makefile.am (libffi_la_LDFLAGS): Add AM_LTLDFLAGS.
46 * Makefile.in: Rebuilt.
48 2009-12-31 Anthony Green <green@redhat.com>
51 * testsuite/libffi.call/huge_struct.c: New test case from Blake
54 2009-12-28 David Edelsohn <edelsohn@gnu.org>
56 * src/powerpc/ffi_darwin.c (ffi_prep_args): Copy abi and nargs to
58 (aix_adjust_aggregate_sizes): New function.
59 (ffi_prep_cif_machdep): Call it.
61 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
63 * configure.ac: Define FFI_MMAP_EXEC_WRIT for the given targets.
64 * configure: Regenerate.
65 * fficonfig.h.in: Likewise.
66 * src/closures.c: Remove the FFI_MMAP_EXEC_WRIT definition for
69 2009-12-26 Andreas Schwab <schwab@linux-m68k.org>
71 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Advance intarg_count
72 when a float arguments is passed in memory.
73 (ffi_closure_helper_SYSV): Mark general registers as used up when
74 a 64bit or soft-float long double argument is passed in memory.
76 2009-12-25 Matthias Klose <doko@ubuntu.com>
78 * man/ffi_call.3: Fix #include in examples.
79 * doc/libffi.texi: Add dircategory.
81 2009-12-25 Frank Everdij <f.p.x.everdij@tudelft.nl>
83 * include/ffi.h.in: Placed '__GNUC__' ifdef around
84 '__attribute__((aligned(8)))' in ffi_closure, fixes compile for
86 * include/ffi_common.h: Added '__sgi' define to non
87 '__attribute__((__mode__()))' integer typedefs.
88 * src/mips/ffi.c (ffi_call, ffi_closure_mips_inner_O32,
89 ffi_closure_mips_inner_N32): Added 'defined(_MIPSEB)' to BE check.
90 (ffi_closure_mips_inner_O32, ffi_closure_mips_inner_N32): Added
91 FFI_LONGDOUBLE support and alignment(N32 only).
92 * src/mips/ffitarget.h: Corrected '#include <sgidefs.h>' for IRIX and
93 fixed non '__attribute__((__mode__()))' integer typedefs.
94 * src/mips/n32.S: Put '#ifdef linux' around '.abicalls' and '.eh_frame'
95 since they are Linux/GNU Assembler specific.
97 2009-12-25 Bradley Smith <brad@brad-smith.co.uk>
99 * configure.ac, Makefile.am, src/avr32/ffi.c,
100 src/avr32/ffitarget.h,
101 src/avr32/sysv.S: Add AVR32 port.
102 * configure, Makefile.in: Rebuilt.
104 2009-12-21 Andreas Tobler <a.tobler@schweiz.org>
106 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
107 * configure: Regenerate.
109 2009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
111 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
113 2009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
115 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
118 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
120 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
123 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
125 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
127 2009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
130 * src/closures.c [X86_64 && __sun__ && __svr4__]
131 (FFI_MMAP_EXEC_WRIT): Define.
133 2009-12-08 David Daney <ddaney@caviumnetworks.com>
135 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
136 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
137 * testsuite/libffi.call/stret_large.c: Same.
138 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
139 * testsuite/libffi.call/stret_large2.c: Same.
140 * testsuite/libffi.call/stret_medium2.c: Same.
142 2009-12-07 David Edelsohn <edelsohn@gnu.org>
144 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
147 2009-12-05 David Edelsohn <edelsohn@gnu.org>
149 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
151 * src/powerpc/aix_closure.S: Same.
153 2009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
155 * Makefile.in: Regenerate.
156 * configure: Regenerate.
157 * include/Makefile.in: Regenerate.
158 * man/Makefile.in: Regenerate.
159 * testsuite/Makefile.in: Regenerate.
161 2009-12-04 David Edelsohn <edelsohn@gnu.org>
163 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
166 2009-12-04 Uros Bizjak <ubizjak@gmail.com>
169 * src/x86/ffi64.c (classify_argument): Update from
170 gcc/config/i386/i386.c.
171 (ffi_closure_unix64_inner): Do not use the address of two consecutive
172 SSE registers directly.
173 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
174 for x86_64 linux targets.
176 2009-12-04 David Edelsohn <edelsohn@gnu.org>
178 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
179 pfr for long double split between fpr13 and stack.
181 2009-12-03 David Edelsohn <edelsohn@gnu.org>
183 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
184 fparg_count twice for long double.
186 2009-12-03 David Edelsohn <edelsohn@gnu.org>
189 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
191 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
193 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
194 Remove xfails for x86 linux targets.
196 2009-12-02 David Edelsohn <edelsohn@gnu.org>
198 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
201 2009-12-01 David Edelsohn <edelsohn@gnu.org>
203 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
204 register usage. Call ffi_prep_args directly. Add long double
205 return value support.
206 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
207 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
208 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
209 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
211 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
213 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
214 Allocate result area between params and FPRs.
216 2009-11-30 David Edelsohn <edelsohn@gnu.org>
219 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
221 * src/powerpc/aix.S: Implement AIX64 version.
222 * src/powerpc/aix_closure.S: Implement AIX64 version.
223 (ffi_closure_ASM): Use extsb, lha and displament addresses.
224 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
226 (ffi_prep_cif_machdep): Same.
228 (ffi_closure_helper_DARWIN): Same.
230 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
233 * testsuite/libffi.call/testclosure.c: New test.
235 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
237 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
238 assembly version use of ___chkstk.
240 2009-09-23 Matthias Klose <doko@ubuntu.com>
242 PR libffi/40242, PR libffi/41443
243 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
244 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
245 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
246 Change the conditionals to __SOFTFP__ || __ARM_EABI__
247 for -mfloat-abi=softfp to work.
249 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
251 PR testsuite/32843 (strikes again)
252 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
253 enable proper extension on char and short.
255 2009-09-15 David Daney <ddaney@caviumnetworks.com>
257 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
258 handling for FFI_TYPE_POINTER.
259 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
260 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
261 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
262 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
263 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
264 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
265 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
266 structure and pointer returns.
267 (ffi_closure_N32): Add handling for pointer returns.
268 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
269 calc_n32_return_struct_flags): Handle soft-float.
270 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
271 (ffi_call_N32): Declare proper argument types.
272 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
275 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
277 * configure.ac (AC_PREREQ): Bump to 2.64.
279 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
281 * Makefile.am (install-html, install-pdf): Remove.
282 * Makefile.in: Regenerate.
284 * Makefile.in: Regenerate.
285 * aclocal.m4: Regenerate.
286 * configure: Regenerate.
287 * fficonfig.h.in: Regenerate.
288 * include/Makefile.in: Regenerate.
289 * man/Makefile.in: Regenerate.
290 * testsuite/Makefile.in: Regenerate.
292 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
294 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
296 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
299 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
300 return types for X86_WIN32.
301 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
302 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
303 _ffi_closure_STDCALL): Likewise.
305 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
306 (dlmmap, dlmunmap): Also use these functions on Cygwin.
308 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
313 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
316 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
318 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
319 to "" before adding paths. (This reinstates an assignment that
320 was removed by my 2009-06-30 commit, but changes the initial
321 value from "." to "".)
323 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
326 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
327 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
330 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
332 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
333 to ld_library_path. Use add_path. Add just find_libgcc_s
334 to ld_library_path, not every libgcc multilib directory.
336 2009-06-16 Wim Lewis <wiml@hhhh.org>
338 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
339 supposed to be callee-saved.
340 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
341 return buffer for odd-size structs.
343 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
346 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
347 allow_stack_execute for Darwin.
349 2009-06-16 Andrew Haley <aph@redhat.com>
351 * configure.ac (TARGETDIR): Add missing blank lines.
352 * configure: Regenerate.
354 2009-06-16 Andrew Haley <aph@redhat.com>
356 * testsuite/libffi.call/cls_align_sint64.c,
357 testsuite/libffi.call/cls_align_uint64.c,
358 testsuite/libffi.call/cls_longdouble_va.c,
359 testsuite/libffi.call/cls_ulonglong.c,
360 testsuite/libffi.call/return_ll1.c,
361 testsuite/libffi.call/stret_medium2.c: Fix printf format
363 * testsuite/libffi.call/ffitest.h,
364 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
366 2009-06-15 Andrew Haley <aph@redhat.com>
368 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
369 * testsuite/libffi.call/err_bad_abi.c: Likewise.
371 2009-06-12 Andrew Haley <aph@redhat.com>
373 * Makefile.am: Remove info_TEXINFOS.
375 2009-06-12 Andrew Haley <aph@redhat.com>
377 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
378 testsuite/libffi.call/cls_align_uint64.c,
379 testsuite/libffi.call/cls_ulonglong.c,
380 testsuite/libffi.call/return_ll1.c,
381 testsuite/libffi.call/stret_medium2.c: Fix printf format
383 testsuite/libffi.special/unwindtest.cc: include stdint.h.
385 2009-06-11 Timothy Wall <twall@users.sf.net>
390 include/ffi_common.h,
395 src/x86/win64.S (new),
396 README: Added win64 support (mingw or MSVC)
400 testsuite/Makefile.in,
402 aclocal.m4: Regenerated
403 * ltcf-c.sh: properly escape cygwin/w32 path
404 * man/ffi_call.3: Clarify size requirements for return value.
405 * src/x86/ffi64.c: Fix filename in comment.
406 * src/x86/win32.S: Remove unused extern.
408 * testsuite/libffi.call/closure_fn0.c,
409 testsuite/libffi.call/closure_fn1.c,
410 testsuite/libffi.call/closure_fn2.c,
411 testsuite/libffi.call/closure_fn3.c,
412 testsuite/libffi.call/closure_fn4.c,
413 testsuite/libffi.call/closure_fn5.c,
414 testsuite/libffi.call/closure_fn6.c,
415 testsuite/libffi.call/closure_stdcall.c,
416 testsuite/libffi.call/cls_12byte.c,
417 testsuite/libffi.call/cls_16byte.c,
418 testsuite/libffi.call/cls_18byte.c,
419 testsuite/libffi.call/cls_19byte.c,
420 testsuite/libffi.call/cls_1_1byte.c,
421 testsuite/libffi.call/cls_20byte.c,
422 testsuite/libffi.call/cls_20byte1.c,
423 testsuite/libffi.call/cls_24byte.c,
424 testsuite/libffi.call/cls_2byte.c,
425 testsuite/libffi.call/cls_3_1byte.c,
426 testsuite/libffi.call/cls_3byte1.c,
427 testsuite/libffi.call/cls_3byte2.c,
428 testsuite/libffi.call/cls_4_1byte.c,
429 testsuite/libffi.call/cls_4byte.c,
430 testsuite/libffi.call/cls_5_1_byte.c,
431 testsuite/libffi.call/cls_5byte.c,
432 testsuite/libffi.call/cls_64byte.c,
433 testsuite/libffi.call/cls_6_1_byte.c,
434 testsuite/libffi.call/cls_6byte.c,
435 testsuite/libffi.call/cls_7_1_byte.c,
436 testsuite/libffi.call/cls_7byte.c,
437 testsuite/libffi.call/cls_8byte.c,
438 testsuite/libffi.call/cls_9byte1.c,
439 testsuite/libffi.call/cls_9byte2.c,
440 testsuite/libffi.call/cls_align_double.c,
441 testsuite/libffi.call/cls_align_float.c,
442 testsuite/libffi.call/cls_align_longdouble.c,
443 testsuite/libffi.call/cls_align_longdouble_split.c,
444 testsuite/libffi.call/cls_align_longdouble_split2.c,
445 testsuite/libffi.call/cls_align_pointer.c,
446 testsuite/libffi.call/cls_align_sint16.c,
447 testsuite/libffi.call/cls_align_sint32.c,
448 testsuite/libffi.call/cls_align_sint64.c,
449 testsuite/libffi.call/cls_align_uint16.c,
450 testsuite/libffi.call/cls_align_uint32.c,
451 testsuite/libffi.call/cls_align_uint64.c,
452 testsuite/libffi.call/cls_dbls_struct.c,
453 testsuite/libffi.call/cls_double.c,
454 testsuite/libffi.call/cls_double_va.c,
455 testsuite/libffi.call/cls_float.c,
456 testsuite/libffi.call/cls_longdouble.c,
457 testsuite/libffi.call/cls_longdouble_va.c,
458 testsuite/libffi.call/cls_multi_schar.c,
459 testsuite/libffi.call/cls_multi_sshort.c,
460 testsuite/libffi.call/cls_multi_sshortchar.c,
461 testsuite/libffi.call/cls_multi_uchar.c,
462 testsuite/libffi.call/cls_multi_ushort.c,
463 testsuite/libffi.call/cls_multi_ushortchar.c,
464 testsuite/libffi.call/cls_pointer.c,
465 testsuite/libffi.call/cls_pointer_stack.c,
466 testsuite/libffi.call/cls_schar.c,
467 testsuite/libffi.call/cls_sint.c,
468 testsuite/libffi.call/cls_sshort.c,
469 testsuite/libffi.call/cls_uchar.c,
470 testsuite/libffi.call/cls_uint.c,
471 testsuite/libffi.call/cls_ulonglong.c,
472 testsuite/libffi.call/cls_ushort.c,
473 testsuite/libffi.call/err_bad_abi.c,
474 testsuite/libffi.call/err_bad_typedef.c,
475 testsuite/libffi.call/float2.c,
476 testsuite/libffi.call/huge_struct.c,
477 testsuite/libffi.call/nested_struct.c,
478 testsuite/libffi.call/nested_struct1.c,
479 testsuite/libffi.call/nested_struct10.c,
480 testsuite/libffi.call/nested_struct2.c,
481 testsuite/libffi.call/nested_struct3.c,
482 testsuite/libffi.call/nested_struct4.c,
483 testsuite/libffi.call/nested_struct5.c,
484 testsuite/libffi.call/nested_struct6.c,
485 testsuite/libffi.call/nested_struct7.c,
486 testsuite/libffi.call/nested_struct8.c,
487 testsuite/libffi.call/nested_struct9.c,
488 testsuite/libffi.call/problem1.c,
489 testsuite/libffi.call/return_ldl.c,
490 testsuite/libffi.call/return_ll1.c,
491 testsuite/libffi.call/stret_large.c,
492 testsuite/libffi.call/stret_large2.c,
493 testsuite/libffi.call/stret_medium.c,
494 testsuite/libffi.call/stret_medium2.c,
495 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
496 of checking for MMAP. Use intptr_t instead of long casts.
498 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
500 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
501 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
502 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
504 2009-06-09 Andrew Haley <aph@redhat.com>
506 * src/x86/freebsd.S: Add missing file.
508 2009-06-08 Andrew Haley <aph@redhat.com>
510 Import from libffi 3.0.8:
512 * doc/libffi.texi: New file.
513 * doc/libffi.info: Likewise.
514 * doc/stamp-vti: Likewise.
515 * man/Makefile.am: New file.
516 * man/ffi_call.3: New file.
518 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
520 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
522 * configure.ac: Bump version to 3.0.8.
523 parisc*-*-linux*: Add.
524 i386-*-freebsd* | i386-*-openbsd*: Add.
525 powerpc-*-beos*: Add.
526 AM_CONDITIONAL X86_FREEBSD: Add.
527 AC_CONFIG_FILES: Add man/Makefile.
529 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
531 2009-06-08 Andrew Haley <aph@redhat.com>
533 * README: Import from libffi 3.0.8.
535 2009-06-08 Andrew Haley <aph@redhat.com>
537 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
538 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
539 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
540 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
542 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
543 * testsuite/libffi.call/stret_medium.c: Likewise.
544 * testsuite/libffi.call/stret_large2.c: Likewise.
545 * testsuite/libffi.call/stret_large.c: Likewise.
547 2008-12-26 Timothy Wall <twall@users.sf.net>
549 * testsuite/libffi.call/cls_longdouble.c,
550 testsuite/libffi.call/cls_longdouble_va.c,
551 testsuite/libffi.call/cls_align_longdouble.c,
552 testsuite/libffi.call/cls_align_longdouble_split.c,
553 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
554 failures on x86_64 cygwin/mingw.
556 2008-12-22 Timothy Wall <twall@users.sf.net>
558 * testsuite/libffi.call/closure_fn0.c,
559 testsuite/libffi.call/closure_fn1.c,
560 testsuite/libffi.call/closure_fn2.c,
561 testsuite/libffi.call/closure_fn3.c,
562 testsuite/libffi.call/closure_fn4.c,
563 testsuite/libffi.call/closure_fn5.c,
564 testsuite/libffi.call/closure_fn6.c,
565 testsuite/libffi.call/closure_loc_fn0.c,
566 testsuite/libffi.call/closure_stdcall.c,
567 testsuite/libffi.call/cls_align_pointer.c,
568 testsuite/libffi.call/cls_pointer.c,
569 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
570 pointer to integer (intptr_t).
571 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
573 2008-07-24 Anthony Green <green@redhat.com>
575 * testsuite/libffi.call/cls_dbls_struct.c,
576 testsuite/libffi.call/cls_double_va.c,
577 testsuite/libffi.call/cls_longdouble.c,
578 testsuite/libffi.call/cls_longdouble_va.c,
579 testsuite/libffi.call/cls_pointer.c,
580 testsuite/libffi.call/cls_pointer_stack.c,
581 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
584 2008-03-04 Anthony Green <green@redhat.com>
588 * testsuite/libffi.call/cls_align_longdouble_split2.c
589 testsuite/libffi.call/cls_align_longdouble_split.c
590 testsuite/libffi.call/cls_dbls_struct.c
591 testsuite/libffi.call/cls_double_va.c
592 testsuite/libffi.call/cls_longdouble.c
593 testsuite/libffi.call/cls_longdouble_va.c
594 testsuite/libffi.call/cls_pointer.c
595 testsuite/libffi.call/cls_pointer_stack.c
596 testsuite/libffi.call/err_bad_abi.c
597 testsuite/libffi.call/err_bad_typedef.c
598 testsuite/libffi.call/stret_large2.c
599 testsuite/libffi.call/stret_large.c
600 testsuite/libffi.call/stret_medium2.c
601 testsuite/libffi.call/stret_medium.c: New tests from Apple.
603 2009-06-05 Andrew Haley <aph@redhat.com>
605 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
608 2009-06-04 Andrew Haley <aph@redhat.com>
610 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
613 2008-02-26 Anthony Green <green@redhat.com>
614 Thomas Heller <theller@ctypes.org>
616 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
619 2008-02-03 Timothy Wall <twall@users.sf.net>
621 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
622 offset based on code pointer, not data pointer.
624 2008-01-31 Timothy Wall <twall@users.sf.net>
626 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
628 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
630 * src/x86/win32.S: Add assembly for stdcall closure.
631 * src/x86/ffi.c: Initialize stdcall closure trampoline.
633 2009-06-04 Andrew Haley <aph@redhat.com>
635 * include/ffi.h.in: Change void (*)() to void (*)(void).
636 * src/x86/ffi.c: Likewise.
638 2009-06-04 Andrew Haley <aph@redhat.com>
640 * src/powerpc/ppc_closure.S: Insert licence header.
641 * src/powerpc/linux64_closure.S: Likewise.
642 * src/m68k/sysv.S: Likewise.
644 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
645 * src/powerpc/ffi.c: Likewise.
646 * src/powerpc/ffi_darwin.c: Likewise.
647 * src/m32r/ffi.c: Likewise.
648 * src/sh64/ffi.c: Likewise.
649 * src/x86/ffi64.c: Likewise.
650 * src/alpha/ffi.c: Likewise.
651 * src/alpha/osf.S: Likewise.
652 * src/frv/ffi.c: Likewise.
653 * src/s390/ffi.c: Likewise.
654 * src/pa/ffi.c: Likewise.
655 * src/pa/hpux32.S: Likewise.
656 * src/ia64/unix.S: Likewise.
657 * src/ia64/ffi.c: Likewise.
658 * src/sparc/ffi.c: Likewise.
659 * src/mips/ffi.c: Likewise.
660 * src/sh/ffi.c: Likewise.
662 2008-02-15 David Daney <ddaney@avtrex.com>
664 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
665 Define (conditionally), and use it to include cachectl.h.
666 (ffi_prep_closure_loc): Fix cache flushing.
667 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
669 2009-06-04 Andrew Haley <aph@redhat.com>
675 src/powerpc/ffitarget.h,
677 src/sh64/ffitarget.h,
688 src/alpha/ffitarget.h,
691 src/m68k/ffitarget.h,
694 src/s390/ffitarget.h,
696 src/cris/ffitarget.h,
701 src/ia64/ffitarget.h,
704 src/ia64/ia64_flags.h,
708 src/sparc/ffitarget.h,
711 src/mips/ffitarget.h,
718 src/sh/sysv.S: Update license text.
720 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
722 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
723 (.eh_frame): Add FDE for it.
725 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
727 * configure.ac: Also check if assembler supports pc-relative
728 relocs on X86_WIN32 targets.
729 * configure: Regenerate.
730 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
731 (_ffi_call_SYSV): Add missing function type symbol .def and
732 add EH markup labels.
733 (_ffi_call_STDCALL): Likewise.
734 (_ffi_closure_SYSV): Likewise.
735 (_ffi_closure_raw_SYSV): Likewise.
736 (.eh_frame): Add hand-crafted EH data.
738 2009-04-09 Jakub Jelinek <jakub@redhat.com>
740 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
741 version 3 of the GNU General Public License and to point readers
742 at the COPYING3 file and the FSF's license web page.
743 * testsuite/libffi.call/call.exp: Likewise.
744 * testsuite/libffi.special/special.exp: Likewise.
746 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
748 * configure: Regenerate.
750 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
753 * configure.ac (HAVE_AS_X86_PCREL): New test.
754 * configure: Regenerate.
755 * fficonfig.h.in: Regenerate.
756 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
757 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
758 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
759 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
760 * src/x86/unix64.S (.Lstore_table): Move to .text section.
761 (.Lload_table): Likewise.
762 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
764 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
766 * configure: Regenerate.
768 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
770 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
771 signed/unsigned int8/16 return values.
772 * src/sparc/v8.S (ffi_call_v8): Likewise.
773 (ffi_closure_v8): Likewise.
775 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
776 Steve Ellcey <sje@cup.hp.com>
778 * configure: Regenerate for new libtool.
779 * Makefile.in: Ditto.
780 * include/Makefile.in: Ditto.
783 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
785 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
786 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
787 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
788 Adjust copyright notice.
789 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
790 register or two register to use for FFI_SYSV structs.
791 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
792 (ffi_closure_helper_SYSV): Fix the return type for
793 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
794 Adjust copyright notice.
796 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
798 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
801 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
803 * configure: Regenerate.
804 * include/Makefile.in: Regenerate.
805 * testsuite/Makefile.in: Regenerate.
807 2008-06-07 Joseph Myers <joseph@codesourcery.com>
809 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
810 powerpc-*-beos*): Remove.
811 * configure: Regenerate.
813 2008-05-09 Julian Brown <julian@codesourcery.com>
815 * Makefile.am (LTLDFLAGS): New.
816 (libffi_la_LDFLAGS): Use above.
817 * Makefile.in: Regenerate.
819 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
822 * aclocal.m4: Regenerate.
823 * configure: Regenerate.
825 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
827 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
828 * src/sh64/sysv.S: Likewise.
830 2008-03-26 Daniel Jacobowitz <dan@debian.org>
832 * src/arm/sysv.S: Fix ARM comment marker.
834 2008-03-26 Jakub Jelinek <jakub@redhat.com>
836 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
837 * src/s390/sysv.S: Likewise.
838 * src/powerpc/ppc_closure.S: Likewise.
839 * src/powerpc/sysv.S: Likewise.
840 * src/x86/unix64.S: Likewise.
841 * src/x86/sysv.S: Likewise.
842 * src/sparc/v8.S: Likewise.
843 * src/sparc/v9.S: Likewise.
844 * src/m68k/sysv.S: Likewise.
845 * src/arm/sysv.S: Likewise.
847 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
849 * aclocal.m4: Regenerate.
850 * configure: Likewise.
851 * Makefile.in: Likewise.
852 * include/Makefile.in: Likewise.
853 * testsuite/Makefile.in: Likewise.
855 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
856 Andreas Tobler <a.tobler@schweiz.org>
858 * configure.ac: Add amd64-*-freebsd* target.
859 * configure: Regenerate.
861 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
864 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
867 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
870 2008-01-24 David Edelsohn <edelsohn@gnu.org>
872 * configure: Regenerate.
874 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
876 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
878 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
881 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
882 signed/unsigned int8/16 for X86_DARWIN.
883 Updated copyright info.
884 Handle one and two byte structs with special cif->flags.
885 * src/x86/ffitarget.h: Add special types for one and two byte structs.
886 Updated copyright info.
887 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
889 Remove code to pop args from the stack after call.
890 Special-case signed/unsigned for int8/16, one and two byte structs.
891 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
892 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
894 Updated copyright info.
896 2007-12-08 David Daney <ddaney@avtrex.com>
898 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
899 SUBU, add with ADDU and use smaller code sequences.
901 2007-12-07 David Daney <ddaney@avtrex.com>
903 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
906 2007-12-06 David Daney <ddaney@avtrex.com>
908 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
910 (ffi_java_raw): New typedef.
911 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
912 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
914 (ffi_java_raw_closure) : Same.
915 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
917 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
919 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
920 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
921 sizeof(ffi_java_raw) for alignment calculations.
922 (ffi_java_ptrarray_to_raw): Same.
923 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
924 if FFI_SIZEOF_JAVA_RAW == 4.
925 (ffi_java_raw_to_rvalue): Same.
926 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
927 (ffi_java_translate_args): Same.
928 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
930 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
932 2007-12-06 David Daney <ddaney@avtrex.com>
934 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
937 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
940 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
941 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
942 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
943 set the NUM_FPR_ARG_REGISTERS according to.
944 Add support for potential soft-float support under hard-float
946 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
947 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
948 to the FFI_LINUX_SOFT_FLOAT ABI.
949 (ffi_prep_cif_machdep): Likewise.
950 (ffi_closure_helper_SYSV): Likewise.
951 * src/powerpc/ppc_closure.S: Make sure not to store float/double
952 on archs where __NO_FPRS__ is true.
953 Add FFI_TYPE_UINT128 support.
954 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
955 Adjust copyright notice.
957 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
959 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
960 * include/ffi_common.h: ... here.
963 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
965 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
966 * src/powerpc/linux64.S: Likewise.
967 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
968 * testsuite/libffi.call/return_ldl.c: New test.
970 2007-09-04 <aph@redhat.com>
972 * src/arm/sysv.S (UNWIND): New.
973 (Whole file): Conditionally compile unwinder directives.
974 * src/arm/sysv.S: Add unwinder directives.
976 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
977 Only treat r0 as a struct address if we're actually returning a
979 Only copy the bytes that are actually within a struct.
980 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
981 is returned in r0, not passed by address.
982 (ffi_call): Allocate a word-sized temporary for the case where
983 a composite is returned in r0.
984 (ffi_prep_incoming_args_SYSV): Align as necessary.
986 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
988 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
989 directly using the sys_cacheflush syscall.
991 2007-07-27 Andrew Haley <aph@redhat.com>
993 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
995 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
997 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
998 * configure.ac: Likewise.
999 * Makefile.in: Regenerate.
1000 * include/Makefile.in: Likewise.
1001 * testsuite/Makefile.in: Likewise.
1002 * configure: Likewise.
1004 2007-08-24 David Daney <ddaney@avtrex.com>
1006 * testsuite/libffi.call/return_sl.c: New test.
1008 2007-08-10 David Daney <ddaney@avtrex.com>
1010 * testsuite/libffi.call/cls_multi_ushort.c,
1011 testsuite/libffi.call/cls_align_uint16.c,
1012 testsuite/libffi.call/nested_struct1.c,
1013 testsuite/libffi.call/nested_struct3.c,
1014 testsuite/libffi.call/cls_7_1_byte.c,
1015 testsuite/libffi.call/nested_struct5.c,
1016 testsuite/libffi.call/cls_double.c,
1017 testsuite/libffi.call/nested_struct7.c,
1018 testsuite/libffi.call/cls_sint.c,
1019 testsuite/libffi.call/nested_struct9.c,
1020 testsuite/libffi.call/cls_20byte1.c,
1021 testsuite/libffi.call/cls_multi_sshortchar.c,
1022 testsuite/libffi.call/cls_align_sint64.c,
1023 testsuite/libffi.call/cls_3byte2.c,
1024 testsuite/libffi.call/cls_multi_schar.c,
1025 testsuite/libffi.call/cls_multi_uchar.c,
1026 testsuite/libffi.call/cls_19byte.c,
1027 testsuite/libffi.call/cls_9byte1.c,
1028 testsuite/libffi.call/cls_align_float.c,
1029 testsuite/libffi.call/closure_fn1.c,
1030 testsuite/libffi.call/problem1.c,
1031 testsuite/libffi.call/closure_fn3.c,
1032 testsuite/libffi.call/cls_sshort.c,
1033 testsuite/libffi.call/closure_fn5.c,
1034 testsuite/libffi.call/cls_align_double.c,
1035 testsuite/libffi.call/nested_struct.c,
1036 testsuite/libffi.call/cls_2byte.c,
1037 testsuite/libffi.call/nested_struct10.c,
1038 testsuite/libffi.call/cls_4byte.c,
1039 testsuite/libffi.call/cls_6byte.c,
1040 testsuite/libffi.call/cls_8byte.c,
1041 testsuite/libffi.call/cls_multi_sshort.c,
1042 testsuite/libffi.call/cls_align_sint16.c,
1043 testsuite/libffi.call/cls_align_uint32.c,
1044 testsuite/libffi.call/cls_20byte.c,
1045 testsuite/libffi.call/cls_float.c,
1046 testsuite/libffi.call/nested_struct2.c,
1047 testsuite/libffi.call/cls_5_1_byte.c,
1048 testsuite/libffi.call/nested_struct4.c,
1049 testsuite/libffi.call/cls_24byte.c,
1050 testsuite/libffi.call/nested_struct6.c,
1051 testsuite/libffi.call/cls_64byte.c,
1052 testsuite/libffi.call/nested_struct8.c,
1053 testsuite/libffi.call/cls_uint.c,
1054 testsuite/libffi.call/cls_multi_ushortchar.c,
1055 testsuite/libffi.call/cls_schar.c,
1056 testsuite/libffi.call/cls_uchar.c,
1057 testsuite/libffi.call/cls_align_uint64.c,
1058 testsuite/libffi.call/cls_ulonglong.c,
1059 testsuite/libffi.call/cls_align_longdouble.c,
1060 testsuite/libffi.call/cls_1_1byte.c,
1061 testsuite/libffi.call/cls_12byte.c,
1062 testsuite/libffi.call/cls_3_1byte.c,
1063 testsuite/libffi.call/cls_3byte1.c,
1064 testsuite/libffi.call/cls_4_1byte.c,
1065 testsuite/libffi.call/cls_6_1_byte.c,
1066 testsuite/libffi.call/cls_16byte.c,
1067 testsuite/libffi.call/cls_18byte.c,
1068 testsuite/libffi.call/closure_fn0.c,
1069 testsuite/libffi.call/cls_9byte2.c,
1070 testsuite/libffi.call/closure_fn2.c,
1071 testsuite/libffi.call/closure_fn4.c,
1072 testsuite/libffi.call/cls_ushort.c,
1073 testsuite/libffi.call/closure_fn6.c,
1074 testsuite/libffi.call/cls_5byte.c,
1075 testsuite/libffi.call/cls_align_pointer.c,
1076 testsuite/libffi.call/cls_7byte.c,
1077 testsuite/libffi.call/cls_align_sint32.c,
1078 testsuite/libffi.special/unwindtest_ffi_call.cc,
1079 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
1081 2007-08-10 David Daney <ddaney@avtrex.com>
1084 * configure.ac: Don't treat mips64 as a special case.
1085 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
1086 * configure: Regenerate
1087 * Makefile.in: Ditto.
1088 * fficonfig.h.in: Ditto.
1089 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
1090 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
1091 (FFI_DEFAULT_ABI): Set for n64 case.
1092 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
1093 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
1094 (ffi_closure_N32): New function.
1095 (.eh_frame): New section
1096 * src/mips/o32.S: Clean up comments.
1097 (ffi_closure_O32): Pass ffi_closure parameter in $12.
1098 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
1099 _MIPS_SIM == _ABIN32 throughout.
1100 (FFI_MIPS_STOP_HERE): New, use in place of
1102 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1103 to support n32/n64 ABIs.
1104 (calc_n32_struct_flags): Rewrite.
1105 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1106 position of flag bits.
1107 (ffi_prep_cif_machdep): Rewrite n32 portion.
1108 (ffi_call): Enable for n64. Add special handling for small structure
1110 (ffi_prep_closure_loc): Add n32 and n64 support.
1111 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1112 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1114 2007-08-08 David Daney <ddaney@avtrex.com>
1116 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1117 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1119 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1120 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1121 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1122 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1123 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1124 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1125 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1126 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1127 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1128 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1129 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1130 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1131 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1132 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1133 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1134 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1135 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1136 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1137 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1138 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1139 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1140 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1141 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1142 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1143 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1144 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1145 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1146 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1147 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1148 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1149 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1151 2007-08-07 Andrew Haley <aph@redhat.com>
1153 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1156 2007-08-06 Andrew Haley <aph@redhat.com>
1159 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1160 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1163 2007-08-02 David Daney <ddaney@avtrex.com>
1165 * testsuite/libffi.call/return_ul.c (main): Define return type as
1166 ffi_arg. Use proper printf conversion specifier.
1168 2007-07-30 Andrew Haley <aph@redhat.com>
1171 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1172 signed/unsigned int8/16.
1173 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1175 Remove code to pop args from the stack after call.
1176 Special-case signed/unsigned int8/16.
1177 * testsuite/libffi.call/return_sc.c (main): Revert.
1179 2007-07-26 Richard Guenther <rguenther@suse.de>
1182 * testsuite/libffi.call/return_sc.c (main): Verify call
1183 result as signed char, not ffi_arg.
1185 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1187 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1188 * configure: Regenerate.
1190 2007-07-11 David Daney <ddaney@avtrex.com>
1192 * src/mips/ffi.c: Don't include sys/cachectl.h.
1193 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1196 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1198 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1199 from (ffi_prep_closure): ... this.
1200 (FFI_INIT_TRAMPOLINE): Adjust.
1202 2005-12-31 Phil Blundell <pb@reciva.com>
1204 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1205 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1206 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1207 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1208 (FFI_CLOSURES): Enable closure support.
1210 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1212 * testsuite/libffi.call/cls_multi_ushort.c,
1213 testsuite/libffi.call/cls_align_uint16.c,
1214 testsuite/libffi.call/nested_struct1.c,
1215 testsuite/libffi.call/nested_struct3.c,
1216 testsuite/libffi.call/cls_7_1_byte.c,
1217 testsuite/libffi.call/cls_double.c,
1218 testsuite/libffi.call/nested_struct5.c,
1219 testsuite/libffi.call/nested_struct7.c,
1220 testsuite/libffi.call/cls_sint.c,
1221 testsuite/libffi.call/nested_struct9.c,
1222 testsuite/libffi.call/cls_20byte1.c,
1223 testsuite/libffi.call/cls_multi_sshortchar.c,
1224 testsuite/libffi.call/cls_align_sint64.c,
1225 testsuite/libffi.call/cls_3byte2.c,
1226 testsuite/libffi.call/cls_multi_schar.c,
1227 testsuite/libffi.call/cls_multi_uchar.c,
1228 testsuite/libffi.call/cls_19byte.c,
1229 testsuite/libffi.call/cls_9byte1.c,
1230 testsuite/libffi.call/cls_align_float.c,
1231 testsuite/libffi.call/closure_fn1.c,
1232 testsuite/libffi.call/problem1.c,
1233 testsuite/libffi.call/closure_fn3.c,
1234 testsuite/libffi.call/cls_sshort.c,
1235 testsuite/libffi.call/closure_fn5.c,
1236 testsuite/libffi.call/cls_align_double.c,
1237 testsuite/libffi.call/cls_2byte.c,
1238 testsuite/libffi.call/nested_struct.c,
1239 testsuite/libffi.call/nested_struct10.c,
1240 testsuite/libffi.call/cls_4byte.c,
1241 testsuite/libffi.call/cls_6byte.c,
1242 testsuite/libffi.call/cls_8byte.c,
1243 testsuite/libffi.call/cls_multi_sshort.c,
1244 testsuite/libffi.call/cls_align_uint32.c,
1245 testsuite/libffi.call/cls_align_sint16.c,
1246 testsuite/libffi.call/cls_float.c,
1247 testsuite/libffi.call/cls_20byte.c,
1248 testsuite/libffi.call/cls_5_1_byte.c,
1249 testsuite/libffi.call/nested_struct2.c,
1250 testsuite/libffi.call/cls_24byte.c,
1251 testsuite/libffi.call/nested_struct4.c,
1252 testsuite/libffi.call/nested_struct6.c,
1253 testsuite/libffi.call/cls_64byte.c,
1254 testsuite/libffi.call/nested_struct8.c,
1255 testsuite/libffi.call/cls_uint.c,
1256 testsuite/libffi.call/cls_multi_ushortchar.c,
1257 testsuite/libffi.call/cls_schar.c,
1258 testsuite/libffi.call/cls_uchar.c,
1259 testsuite/libffi.call/cls_align_uint64.c,
1260 testsuite/libffi.call/cls_ulonglong.c,
1261 testsuite/libffi.call/cls_align_longdouble.c,
1262 testsuite/libffi.call/cls_1_1byte.c,
1263 testsuite/libffi.call/cls_12byte.c,
1264 testsuite/libffi.call/cls_3_1byte.c,
1265 testsuite/libffi.call/cls_3byte1.c,
1266 testsuite/libffi.call/cls_4_1byte.c,
1267 testsuite/libffi.call/cls_6_1_byte.c,
1268 testsuite/libffi.call/cls_16byte.c,
1269 testsuite/libffi.call/cls_18byte.c,
1270 testsuite/libffi.call/closure_fn0.c,
1271 testsuite/libffi.call/cls_9byte2.c,
1272 testsuite/libffi.call/closure_fn2.c,
1273 testsuite/libffi.call/closure_fn4.c,
1274 testsuite/libffi.call/cls_ushort.c,
1275 testsuite/libffi.call/closure_fn6.c,
1276 testsuite/libffi.call/cls_5byte.c,
1277 testsuite/libffi.call/cls_align_pointer.c,
1278 testsuite/libffi.call/cls_7byte.c,
1279 testsuite/libffi.call/cls_align_sint32.c,
1280 testsuite/libffi.special/unwindtest_ffi_call.cc,
1281 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1283 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1285 * aclocal.m4: Regenerated.
1287 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1289 * configure: Regenerate.
1291 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1293 * Makefile.in: Regenerate.
1294 * configure: Regenerate.
1295 * aclocal.m4: Regenerate.
1296 * include/Makefile.in: Regenerate.
1297 * testsuite/Makefile.in: Regenerate.
1299 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1301 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1302 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1303 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1304 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1305 (FFI_CLOSURES): Enable closure support.
1307 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1309 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1310 * configure: Regenerate.
1311 * fficonfig.h.in: Regenerate.
1312 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1313 CFI_OFFSET,CFI_DEF_CFA): New macros.
1314 (ffi_call_SYSV): Add callframe annotation.
1316 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1318 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1319 numerous test suite failures.
1320 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1322 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1324 * Makefile.am (EXTRA_DIST): Bring up to date.
1325 * Makefile.in: Regenerate.
1326 * src/frv/eabi.S: Remove RCS keyword.
1328 2007-04-06 Richard Henderson <rth@redhat.com>
1330 * configure.ac: Tidy target case.
1331 (HAVE_LONG_DOUBLE): Allow the target to override.
1332 * configure: Regenerate.
1333 * include/ffi.h.in: Don't define ffi_type_foo if
1334 LIBFFI_HIDE_BASIC_TYPES is defined.
1335 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1337 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1338 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1339 (ffi_type_longdouble): Special case for Alpha. Don't define
1340 if long double == double.
1342 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1343 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1344 (ffi_call, ffi_closure_osf_inner): Likewise.
1345 (ffi_closure_osf_inner): Likewise. Mark hidden.
1346 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1347 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1348 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1349 (load_table): Handle 128-bit long double.
1351 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1353 2007-04-06 Tom Tromey <tromey@redhat.com>
1356 * README: Fixed bug in example.
1358 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1360 * src/closures.c: Include sys/statfs.h.
1361 (_GNU_SOURCE): Define on Linux.
1362 (FFI_MMAP_EXEC_SELINUX): Define.
1363 (selinux_enabled): New variable.
1364 (selinux_enabled_check): New function.
1365 (is_selinux_enabled): Define.
1368 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1370 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1371 Use 'volatile float sum' to create sum of floats to avoid false
1372 negative due to excess precision on ix86 targets.
1375 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1377 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1379 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1381 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1383 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1384 (ffi_prep_closure_loc): New.
1385 (ffi_prep_raw_closure_loc): New.
1386 (ffi_prep_java_raw_closure_loc): New.
1387 * src/closures.c: New file.
1388 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1389 Replace sflags with exec_offset.
1390 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1391 sub_segment_exec_offset): New macros.
1392 (get_segment_flags, set_segment_flags, check_segment_merge): New
1394 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1395 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1396 destroy_mspace): Use new macros.
1397 (sys_alloc): Silence warning.
1398 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1399 * Makefile.in: Rebuilt.
1400 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1401 terms of ffi_prep_closure_loc.
1402 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1404 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1406 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1408 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1409 the renamed version.
1410 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1411 (ffi_prep_closure): ... this.
1412 * src/pa/ffi.c: Likewise.
1413 * src/cris/ffi.c: Likewise. Adjust.
1414 * src/frv/ffi.c: Likewise.
1415 * src/ia64/ffi.c: Likewise.
1416 * src/mips/ffi.c: Likewise.
1417 * src/powerpc/ffi_darwin.c: Likewise.
1418 * src/s390/ffi.c: Likewise.
1419 * src/sh/ffi.c: Likewise.
1420 * src/sh64/ffi.c: Likewise.
1421 * src/sparc/ffi.c: Likewise.
1422 * src/x86/ffi64.c: Likewise.
1423 * src/x86/ffi.c: Likewise.
1424 (FFI_INIT_TRAMPOLINE): Adjust.
1425 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1426 (ffi_prep_raw_closure): ... this.
1427 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1428 (ffi_prep_closure): ... this.
1429 (flush_icache): Adjust.
1431 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1433 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1436 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1438 * Makefile.am: Add dummy install-pdf target.
1439 * Makefile.in: Regenerate
1441 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1443 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1444 ffi_closure_helper_SYSV): Add long double handling.
1446 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1448 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1449 immediately after bctrl instruction.
1451 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1453 * Makefile.am (all-recursive, install-recursive,
1454 mostlyclean-recursive, clean-recursive, distclean-recursive,
1455 maintainer-clean-recursive): Add missing targets.
1456 * Makefile.in: Rebuilt.
1458 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1460 * configure.ac: Add TARGET for x86_64-*-darwin*.
1461 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1463 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1464 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1465 * configure: Regenerate.
1466 * Makefile.in: Regenerate.
1467 * include/Makefile.in: Regenerate.
1468 * testsuite/Makefile.in: Regenerate.
1469 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1472 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1474 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1477 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1479 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1480 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1482 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1483 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1485 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1486 Sandro Tolaini <tolaini@libero.it>
1488 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1490 * configure: Regenerated.
1491 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1492 (EXTRA_DIST): Add src/x86/darwin.S.
1493 * Makefile.in: Regenerated.
1494 * include/Makefile.in: Regenerated.
1495 * testsuite/Makefile.in: Regenerated.
1497 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1498 X86_WIN32, and additionally align stack to 16 bytes.
1499 * src/x86/darwin.S: New, based on sysv.S.
1500 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1502 2006-09-12 David Daney <ddaney@avtrex.com>
1505 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1506 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1507 * aclocal.m4: Regenerated for automake 1.9.6.
1508 * Makefile.in: Regenerated.
1509 * include/Makefile.in: Regenerated.
1510 * testsuite/Makefile.in: Regenerated.
1512 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1514 * include/ffi_common.h (struct): Revert accidental commit.
1516 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1518 * include/ffi_common.h: Remove lint directives.
1519 * include/ffi.h.in: Likewise.
1521 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1523 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1524 for 32-bit architectures.
1525 * testsuite/libffi.call/return_ul.c: New test case.
1527 2006-07-19 David Daney <ddaney@avtrex.com>
1529 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1530 xfail remains for mips64.
1532 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1534 * Makefile.am: Add install-html target. Add install-html to .PHONY
1535 * Makefile.in: Regenerate.
1536 * aclocal.m4: Regenerate.
1537 * include/Makefile.in: Regenerate.
1538 * testsuite/Makefile.in: Regenerate.
1540 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1542 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1545 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1547 * README: Remove notice about 'Crazy Comments'.
1548 * src/debug.c: Remove lint directives. Cleanup white spaces.
1549 * src/java_raw_api.c: Likewise.
1550 * src/prep_cif.c: Likewise.
1551 * src/raw_api.c: Likewise.
1552 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1554 * src/arm/ffi.c: Remove lint directives.
1555 * src/m32r/ffi.c: Likewise.
1556 * src/pa/ffi.c: Likewise.
1557 * src/powerpc/ffi.c: Likewise.
1558 * src/powerpc/ffi_darwin.c: Likewise.
1559 * src/sh/ffi.c: Likewise.
1560 * src/sh64/ffi.c: Likewise.
1561 * src/x86/ffi.c: Likewise.
1562 * testsuite/libffi.call/float2.c: Likewise.
1563 * testsuite/libffi.call/promotion.c: Likewise.
1564 * testsuite/libffi.call/struct1.c: Likewise.
1566 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1568 * src/pa/hpux32.S: Correct unwind offset calculation for
1570 * src/pa/linux.S: Likewise.
1572 2006-04-12 James E Wilson <wilson@specifix.com>
1575 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1576 (hfa_type_load): Call stf_spill.
1577 (hfa_type_store): Call ldf_fill.
1578 (ffi_call): Adjust calls to above routines. Add local temps for
1581 2006-04-10 Matthias Klose <doko@debian.org>
1583 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1584 directory names containing underscores.
1586 2006-04-07 James E Wilson <wilson@specifix.com>
1588 * testsuite/libffi.call/float4.c: New testcase.
1590 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1591 Andreas Tobler <a.tobler@schweiz.ch>
1593 * Makefile.am: Add PA_HPUX port.
1594 * Makefile.in: Regenerate.
1595 * include/Makefile.in: Likewise.
1596 * testsuite/Makefile.in: Likewise.
1597 * configure.ac: Add PA_HPUX rules.
1598 * configure: Regenerate.
1599 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1600 Add PA_HPUX and PA64_HPUX.
1601 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1602 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1603 (FFI_TYPE_SMALL_STRUCT2): Define.
1604 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1605 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1606 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1607 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1608 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1609 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1610 * src/pa/ffi.c (ROUND_DOWN): Delete.
1611 (fldw, fstw, fldd, fstd): Use '__asm__'.
1612 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1613 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1614 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1615 Simplify incrementing of stack slot variable. Change type of local
1616 'n' to unsigned int.
1617 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1619 (ffi_prep_cif_machdep): Likewise.
1620 (ffi_call): Likewise.
1621 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1622 return type to ffi_status. Simplify incrementing of stack slot
1623 variable. Only copy floating point argument registers when PA_LINUX
1624 is true. Reformat debug statement.
1625 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1626 FFI_TYPE_SMALL_STRUCT8.
1627 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1629 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1630 Add nops to cache flush. Add trampoline for PA_HPUX.
1631 * src/pa/hpux32.S: New file.
1632 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1633 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1634 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1635 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1636 argument type checks so that common argument types appear first.
1637 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1638 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1640 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1642 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1643 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1644 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1645 FFI_TYPE_LONGDOUBLE.
1646 (ffi_prep_args64): Assert using IBM extended double.
1647 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1648 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1649 (ffi_call): Handle FFI_LINUX.
1650 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1651 gpr3 return pointer as for struct return. Handle FFI_LINUX
1652 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1654 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1655 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1656 Don't use r6 as pointer to results, instead use sp offset. Don't
1657 make a special call to load lr with case table address, instead
1658 use offset from previous call.
1659 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1660 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1663 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1665 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1666 passed with FP registers correctly.
1667 (ffi_closure_helper_SYSV): Likewise.
1668 * src/sh64/sysv.S: Likewise.
1670 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1672 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1673 args and userdata unused.
1674 (closure_test_fn1): Mark cif and userdata unused.
1675 (main): Remove unused res.
1677 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1679 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1680 -O2, -O3, -Os and the warning flags -W -Wall.
1681 * testsuite/libffi.special/special.exp: Likewise.
1682 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1683 unused parameter unused for gcc or else do nothing.
1684 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1685 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1686 and userdata unused.
1687 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1688 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1689 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1690 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1691 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1692 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1693 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1694 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1695 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1696 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1697 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1698 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1699 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1700 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1701 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1702 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1703 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1704 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1705 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1706 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1707 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1708 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1709 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1710 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1712 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1714 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1716 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1717 void* to avoid compiler warning.
1719 (cls_struct_align_gn): Mark cif and userdata unused.
1720 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1722 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1724 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1726 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1728 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1730 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1731 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1732 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1734 (main): Cast res_call to silence gcc.
1735 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1737 (main): Cast res_call to silence gcc.
1738 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1740 (main): Cast res_call to silence gcc.
1741 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1743 (main): Cast res_call to silence gcc.
1744 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1746 (main): Cast res_call to silence gcc.
1747 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1749 (main): Cast res_call to silence gcc.
1750 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1752 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1753 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1755 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1756 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1758 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1759 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1761 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1762 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1764 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1765 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1766 and userdata unused.
1767 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1769 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1770 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1771 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1772 Cleanup white spaces.
1773 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1774 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1775 cif and userdata unused.
1776 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1778 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1779 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1780 formatters to silence gcc.
1781 (B_gn): Mark cif and userdata unused.
1782 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1784 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1785 (B_gn): Mark cif and userdata unused.
1786 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1788 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1789 (B_gn): Mark cif and userdata unused.
1790 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1792 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1793 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1794 * testsuite/libffi.call/problem1.c (stub): Likewise.
1795 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1797 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1798 in the last commit for this test case in the test case itself.
1799 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1801 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1802 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1803 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1804 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1805 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1806 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1808 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1810 * src/sh/sysv.S: Fix register numbers in the FDE for
1813 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1815 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1816 declaration to avoid a false negative on ix86. See PR323.
1818 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1820 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1821 and cast integer to void * if needed. Update the pointer to
1822 the FP register saved area correctly.
1824 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1826 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1828 * testsuite/libffi.call/nested_struct4.c: Likewise.
1830 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1832 * testsuite/libffi.call/return_dbl.c: New test case.
1833 * testsuite/libffi.call/return_dbl1.c: Likewise.
1834 * testsuite/libffi.call/return_dbl2.c: Likewise.
1835 * testsuite/libffi.call/return_fl.c: Likewise.
1836 * testsuite/libffi.call/return_fl1.c: Likewise.
1837 * testsuite/libffi.call/return_fl2.c: Likewise.
1838 * testsuite/libffi.call/return_fl3.c: Likewise.
1839 * testsuite/libffi.call/closure_fn6.c: Likewise.
1841 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1843 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1844 here to be used by other test cases too.
1846 * testsuite/libffi.call/nested_struct10.c: New test case.
1847 * testsuite/libffi.call/nested_struct9.c: Likewise.
1848 * testsuite/libffi.call/nested_struct8.c: Likewise.
1849 * testsuite/libffi.call/nested_struct7.c: Likewise.
1850 * testsuite/libffi.call/nested_struct6.c: Likewise.
1851 * testsuite/libffi.call/nested_struct5.c: Likewise.
1852 * testsuite/libffi.call/nested_struct4.c: Likewise.
1854 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1856 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1857 * configure: Rebuilt.
1859 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1861 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1862 instead do the shifting inline.
1863 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1864 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1865 and 8 byte structs, for the remaining struct sizes don't call
1866 __lshrdi3, instead do the shifting inline.
1868 2005-12-07 Thiemo Seufer <ths@networkno.de>
1870 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1871 missing parentheses.
1872 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1873 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1874 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1875 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1876 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1878 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1880 (ffi_prep_closure): Improve trampoline instruction scheduling.
1881 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1883 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1885 * src/powerpc/ffi.c: Formatting.
1886 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1887 (ffi_prep_args64): Likewise.
1889 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1891 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1892 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1894 2005-09-26 Tom Tromey <tromey@redhat.com>
1896 * testsuite/libffi.call/float1.c (value_type): New typedef.
1897 (CANARY): New define.
1898 (main): Check for result buffer overflow.
1899 * src/powerpc/linux64.S: Handle linux64 long double returns.
1900 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1901 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1903 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1906 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1908 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1910 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1912 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1913 (AH_BOTTOM): Add FFI_HIDDEN definition.
1914 * configure: Rebuilt.
1915 * fficonfig.h.in: Rebuilt.
1916 * src/powerpc/ffi.c (hidden): Remove.
1917 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1918 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1919 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1920 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1921 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1922 add FFI_HIDDEN to its prototype.
1923 (ffi_closure_SYSV_inner): New.
1924 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1925 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1927 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1930 * configure: Rebuilt.
1931 * configure.ac: Handle i*86-*-gnu*.
1933 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1935 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1936 DW_CFA_offset_extended_sf rather than
1937 DW_CFA_GNU_negative_offset_extended.
1938 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1940 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1942 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1944 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1945 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1946 partially on register.
1947 (ffi_closure_helper_SYSV): Likewise.
1948 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1950 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1952 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1954 * src/sh64/ffi.c (simple_type): Remove.
1955 (return_type): Handle small structures correctly.
1956 (ffi_prep_args): Likewise.
1957 (ffi_call): Likewise.
1958 (ffi_closure_helper_SYSV): Likewise.
1959 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1960 Emit position independent code if PIC and remove wrong datalabel
1961 prefixes from EH data.
1963 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1965 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1966 * Makefile.in: Regenerate.
1967 * include/Makefile.in: Likewise.
1968 * testsuite/Makefile.in: Likewise.
1969 * configure.ac: Add POWERPC_FREEBSD rules.
1970 * configure: Regenerate.
1971 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1972 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1973 * src/powerpc/ffi.c: Add flags to handle small structure returns
1975 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1977 (ffi_closure_helper_SYSV): Likewise.
1978 * src/powerpc/ppc_closure.S: Add return types for small structures.
1979 * src/powerpc/sysv.S: Add bits to handle small structures for
1982 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1984 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1985 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1986 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1988 2005-07-05 Randolph Chung <tausq@debian.org>
1990 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1991 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1992 structures. Kill compilation warnings.
1993 (ffi_closure_inner_LINUX): Print return values as hex in debug
1994 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1995 Properly handle 5-7 byte structure returns.
1996 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1997 (FFI_TYPE_SMALL_STRUCT2): Remove.
1998 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1999 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
2000 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
2001 (checksmst1, checksmst2): Remove.
2002 (checksmst3): Optimize handling of 3-byte struct returns.
2003 (checksmst567): Properly handle 5-7 byte struct returns.
2005 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2008 * src/mips/n32.S: Enforce PIC code.
2009 * src/mips/o32.S: Likewise.
2011 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2013 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
2014 * configure: Regenerate.
2016 2005-06-01 Alan Modra <amodra@bigpond.net.au>
2018 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
2019 to call ffi_closure_helper_SYSV. Append @local instead.
2020 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
2022 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
2024 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
2025 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
2026 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
2027 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
2028 include/Makefile.in, testsuite/Makefile.in: Regenerate.
2030 2005-05-09 Mike Stump <mrs@apple.com>
2032 * configure: Regenerate.
2034 2005-05-08 Richard Henderson <rth@redhat.com>
2037 * src/alpha/osf.S: Update unwind into to match code.
2039 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
2040 Richard Henderson <rth@redhat.com>
2042 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
2044 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
2045 (ffi_prep_closure): Set carry bit if sse-used flag set.
2046 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
2047 Only load sse registers if ssecount non-zero.
2048 (ffi_closure_unix64): Only save sse registers if carry set on entry.
2050 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
2052 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
2053 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
2054 * configure: Regenerate.
2056 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
2058 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
2059 have Tcl8.3-compatible intermediate variable.
2061 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
2062 Hans-Peter Nilsson <hp@axis.com>
2064 * Makefile.am: Add CRIS support.
2065 * configure.ac: Likewise.
2066 * Makefile.in, configure, testsuite/Makefile.in,
2067 include/Makefile.in: Regenerate.
2068 * src/cris: New directory.
2069 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
2070 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
2072 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
2073 \r?\n in output tests.
2075 2005-04-12 Mike Stump <mrs@apple.com>
2077 * configure: Regenerate.
2079 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
2081 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
2083 2005-03-30 Steve Ellcey <sje@cup.hp.com>
2085 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
2087 * src/ia64/unix.S (ffi_closure_unix): Extend gp
2088 to 64 bits in ILP32 mode.
2089 Load 64 bits even for short data.
2091 2005-03-23 Mike Stump <mrs@apple.com>
2093 * src/powerpc/darwin.S: Update for -m64 multilib.
2094 * src/powerpc/darwin_closure.S: Likewise.
2096 2005-03-21 Zack Weinberg <zack@codesourcery.com>
2098 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
2099 Do not set tool_include_dir.
2100 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
2102 * include/Makefile.am: Set gcc_version and toollibffidir.
2103 * include/Makefile.in: Regenerate.
2105 2005-02-22 Andrew Haley <aph@redhat.com>
2107 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2108 odd-numbered register pairs for 64-bit integer types.
2110 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2113 * testsuite/libffi.call/return_ll1.c: New test case.
2115 2005-02-11 Janis Johnson <janis187@us.ibm.com>
2117 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2118 * testsuite/libffi.call/float.c: Ditto.
2119 * testsuite/libffi.call/float2.c: Ditto.
2120 * testsuite/libffi.call/float3.c: Ditto.
2122 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2124 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2126 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2128 * testsuite/libffi.special/special.exp (cxx_options): Add
2131 2004-12-31 Richard Henderson <rth@redhat.com>
2133 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2134 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2135 offset parameters with a type parameter; deduce size and structure
2136 alignment. Update all users.
2138 2004-12-31 Richard Henderson <rth@redhat.com>
2140 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2141 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
2142 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
2143 into ffi_prep_closure.
2144 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2147 2004-12-27 Richard Henderson <rth@redhat.com>
2149 * src/x86/unix64.S: Fix typo in unwind info.
2151 2004-12-25 Richard Henderson <rth@redhat.com>
2153 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2154 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2155 (merge_classes): Check for it.
2157 (classify_argument): Pass byte_offset by value; perform all updates
2159 (examine_argument): Add classes argument; handle
2160 X86_64_COMPLEX_X87_CLASS.
2161 (ffi_prep_args): Merge into ...
2162 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2163 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2164 (ffi_fill_return_value): Remove.
2165 (ffi_prep_closure): Remove dead assert.
2166 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2167 Rewrite to use struct register_args instead of va_list. Create
2168 flags for handling structure returns.
2169 * src/x86/unix64.S: Remove dead strings.
2170 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2171 stack frame with ffi_call. Handle structure returns properly.
2172 (float2sse, floatfloat2sse, double2sse): Remove.
2173 (sse2float, sse2double, sse2floatfloat): Remove.
2174 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2175 to handle structure returns properly.
2177 2004-12-08 David Edelsohn <edelsohn@gnu.org>
2179 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2181 * Makefile.in: Regenerated.
2183 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2185 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2186 * configure, aclocal.m4, Makefile.in: Regenerate.
2187 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2189 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2191 * configure: Regenerate for libtool change.
2193 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2195 * configure: Regenerate for libtool reversion.
2197 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2199 * configure: Regenerate for libtool change.
2201 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2203 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2205 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2207 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2208 of jal. Use an absolute encoding for the frame information.
2210 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2212 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2213 * acinclude.m4: Delete logic for sincludes.
2214 * aclocal.m4, Makefile.in, configure: Regenerate.
2215 * include/Makefile: Likewise.
2216 * testsuite/Makefile: Likewise.
2218 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2220 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2221 on a 8-byte boundary.
2222 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2224 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2226 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2227 long long values. Round stack allocation to a multiple of 8 bytes
2228 for ATPCS compatibility.
2229 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2230 names. Handle returning long long types. Add Thumb and interworking
2231 support. Improve soft-float code.
2233 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2235 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2236 (libffi_exit): New function.
2237 (libffi_init): Build the testglue wrapper if needed.
2239 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2242 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2244 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2246 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2248 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2250 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2251 * testsuite/libffi.call/float3.c: New test case.
2253 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2255 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2256 the function returning a structure pointed with R2.
2257 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2258 the structure return value if T bit set. Emit position
2259 independent code and EH data if PIC.
2261 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2263 * Makefile.am: Add m32r support.
2264 * configure.ac: Likewise.
2265 * Makefile.in: Regenerate.
2266 * confiugre: Regenerate.
2267 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2268 (uint64, sint64, double, longdouble)
2269 * src/m32r: New directory.
2270 * src/m32r/ffi.c: New file.
2271 * src/m32r/sysv.S: Likewise.
2272 * src/m32r/ffitarget.h: Likewise.
2274 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2276 * testsuite/libffi.call/negint.c: New test case.
2278 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2281 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2282 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2283 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2286 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2288 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2289 * testsuite/libffi.call/promotion.c: Likewise.
2290 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2292 * testsuite/libffi.call/return_sc.c: Likewise.
2293 * testsuite/libffi.call/return_uc.c: Likewise.
2295 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2297 * src/powerpc/darwin.S: Fix comments and identation.
2298 * src/powerpc/darwin_closure.S: Likewise.
2300 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2302 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2303 (ffi_prep_args): Handle longdouble arguments.
2304 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2306 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2307 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2309 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2310 * src/types.c: Defined longdouble size and alignment for darwin.
2312 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2314 * src/powerpc/aix.S: Remove whitespaces.
2315 * src/powerpc/aix_closure.S: Likewise.
2316 * src/powerpc/asm.h: Likewise.
2317 * src/powerpc/ffi.c: Likewise.
2318 * src/powerpc/ffitarget.h: Likewise.
2319 * src/powerpc/linux64.S: Likewise.
2320 * src/powerpc/linux64_closure.S: Likewise.
2321 * src/powerpc/ppc_closure.S: Likewise.
2322 * src/powerpc/sysv.S: Likewise.
2324 2004-08-30 Anthony Green <green@redhat.com>
2326 * Makefile.am: Add frv support.
2327 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2328 * configure.ac: Read configure.host.
2329 * configure.in: Read configure.host.
2330 * configure.host: New file. frv-elf needs libgloss.
2331 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2332 alignment. This is needed to frv and shouldn't harm the others.
2333 * include/ffi_common.h (ALIGN_DOWN): New macro.
2334 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2336 2004-08-24 David Daney <daney@avtrex.com>
2338 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2339 * testsuite/libffi.call/closure_fn1.c: Likewise.
2340 * testsuite/libffi.call/closure_fn2.c Likewise.
2341 * testsuite/libffi.call/closure_fn3.c: Likewise.
2342 * testsuite/libffi.call/closure_fn4.c: Likewise.
2343 * testsuite/libffi.call/closure_fn5.c: Likewise.
2344 * testsuite/libffi.call/cls_18byte.c: Likewise.
2345 * testsuite/libffi.call/cls_19byte.c: Likewise.
2346 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2347 * testsuite/libffi.call/cls_20byte.c: Likewise.
2348 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2349 * testsuite/libffi.call/cls_24byte.c: Likewise.
2350 * testsuite/libffi.call/cls_2byte.c: Likewise.
2351 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2352 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2353 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2354 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2355 * testsuite/libffi.call/cls_4byte.c: Likewise.
2356 * testsuite/libffi.call/cls_64byte.c: Likewise.
2357 * testsuite/libffi.call/cls_6byte.c: Likewise.
2358 * testsuite/libffi.call/cls_7byte.c: Likewise.
2359 * testsuite/libffi.call/cls_8byte.c: Likewise.
2360 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2361 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2362 * testsuite/libffi.call/cls_align_double.c: Likewise.
2363 * testsuite/libffi.call/cls_align_float.c: Likewise.
2364 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2365 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2366 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2367 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2368 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2369 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2370 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2371 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2372 * testsuite/libffi.call/cls_double.c: Likewise.
2373 * testsuite/libffi.call/cls_float.c: Likewise.
2374 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2375 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2376 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2377 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2378 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2379 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2380 * testsuite/libffi.call/cls_schar.c: Likewise.
2381 * testsuite/libffi.call/cls_sint.c: Likewise.
2382 * testsuite/libffi.call/cls_sshort.c: Likewise.
2383 * testsuite/libffi.call/cls_uchar.c: Likewise.
2384 * testsuite/libffi.call/cls_uint.c: Likewise.
2385 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2386 * testsuite/libffi.call/cls_ushort.c: Likewise.
2387 * testsuite/libffi.call/nested_struct.c: Likewise.
2388 * testsuite/libffi.call/nested_struct1.c: Likewise.
2389 * testsuite/libffi.call/nested_struct2.c: Likewise.
2390 * testsuite/libffi.call/nested_struct3.c: Likewise.
2391 * testsuite/libffi.call/problem1.c: Likewise.
2392 * testsuite/libffi.special/unwindtest.cc: Likewise.
2393 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2395 * testsuite/libffi.call/cls_16byte.c: Likewise.
2396 * testsuite/libffi.call/cls_5byte.c: Likewise.
2398 2004-08-23 David Daney <daney@avtrex.com>
2401 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2402 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2403 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2404 parameters and return types.
2405 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2406 (ffi_prep_closure): Ditto.
2407 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2408 alignment calculations.
2409 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2410 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2412 2004-08-14 Casey Marshall <csm@gnu.org>
2414 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2415 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2416 integer (O32 ABI only).
2417 (ffi_prep_closure): new function.
2418 (ffi_closure_mips_inner_O32): new function.
2419 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2420 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2421 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2422 64 bit integers correctly.
2423 (ffi_closure_O32): new function.
2424 Added DWARF-2 unwind info for both functions.
2426 2004-08-10 Andrew Haley <aph@redhat.com>
2428 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2430 2004-08-01 Robert Millan <robertmh@gnu.org>
2432 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2433 * configure: Regenerate.
2435 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2437 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2438 and mmap() explicitly instead of relying on preset autoconf cache
2440 * aclocal.m4: Regenerate.
2441 * configure: Regenerate.
2443 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2445 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2446 (ffi_check_float_struct): Remove unused prototype.
2448 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2450 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2451 character on Darwin, use '\n\t' instead.
2453 2004-06-26 Matthias Klose <doko@debian.org>
2455 * libtool-version: Fix typo in revision/age.
2457 2004-06-17 Matthias Klose <doko@debian.org>
2459 * libtool-version: New.
2460 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2461 * Makefile.in: Regenerate.
2463 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2465 * Makefile.am: Remove useless multilib rules.
2466 * Makefile.in: Regenerate.
2467 * aclocal.m4: Regenerate with automake 1.8.5.
2468 * configure.ac: Remove useless multilib configury.
2469 * configure: Regenerate.
2471 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2473 * .cvsignore: New file.
2475 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2477 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2479 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2482 2004-06-06 Sean McNeil <sean@mcneil.com>
2484 * configure.ac: Add x86_64-*-freebsd* support.
2485 * configure: Regenerate.
2487 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2490 * configure.ac: Test for existence of mmap and sys/mman.h before
2491 checking blacklist. Fix suggested by Jim Wilson.
2492 * configure: Regenerate.
2494 2004-04-26 Matt Austern <austern@apple.com>
2496 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2498 * src/powerpc/darwin_closure.S: Likewise.
2500 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2502 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2503 error. Reported by Thomas Heller <theller@python.net>.
2504 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2505 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2507 2004-03-20 Matthias Klose <doko@debian.org>
2509 * src/pa/linux.S: Fix typo.
2511 2004-03-19 Matthias Klose <doko@debian.org>
2513 * Makefile.am: Update.
2514 * Makefile.in: Regenerate.
2515 * src/pa/ffi.h.in: Remove.
2516 * src/pa/ffitarget.h: New file.
2518 2004-02-10 Randolph Chung <tausq@debian.org>
2520 * Makefile.am: Add PA support.
2521 * Makefile.in: Regenerate.
2522 * include/Makefile.in: Regenerate.
2523 * configure.ac: Add PA target.
2524 * configure: Regenerate.
2525 * src/pa/ffi.c: New file.
2526 * src/pa/ffi.h.in: Add PA support.
2527 * src/pa/linux.S: New file.
2528 * prep_cif.c: Add PA support.
2530 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2532 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2534 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2535 with ecif->cif->flags.
2536 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2538 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2539 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2540 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2541 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2543 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2545 * configure.in: Rename file to ...
2546 * configure.ac: ... this.
2547 * fficonfig.h.in: Regenerate.
2548 * Makefile.in: Regenerate.
2549 * include/Makefile.in: Regenerate.
2550 * testsuite/Makefile.in: Regenerate.
2552 2004-03-12 Matt Austern <austern@apple.com>
2554 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2555 changes in EH format resulting from addition of linkonce support.
2556 * src/powerpc/darwin_closure.S: Likewise.
2558 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2559 Paolo Bonzini <bonzini@gnu.org>
2561 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2562 Remove VPATH. Remove rules for object files. Remove multilib support.
2563 (AM_CCASFLAGS): Add.
2564 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2565 (AC_PREREQ): Bump version to 2.59.
2566 (AC_INIT): Fill with version info and bug address.
2567 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2568 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2569 De-precious CC so that the right flags are passed down to multilibs.
2570 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2571 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2572 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2573 * configure: Rebuilt.
2574 * aclocal.m4: Likewise.
2575 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2576 * fficonfig.h.in: Likewise.
2578 2004-03-11 Andreas Schwab <schwab@suse.de>
2580 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2581 arguments from fp registers only for the first 8 parameter slots.
2582 Don't convert a float parameter when passed in memory.
2584 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2586 * configure: Regenerate for config/accross.m4 correction.
2588 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2590 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2591 ecif->cif->bytes to bytes.
2592 (ffi_prep_cif_machdep): Add braces around nested if statement.
2594 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2596 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2598 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2599 (ffi_closure_helper_LINUX64): Fix typo.
2600 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2602 * testsuite/libffi.call/float.c: Likewise.
2603 * testsuite/libffi.call/float2.c: Likewise.
2605 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2607 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2608 long double function return and long double arg handling.
2609 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2610 Use "end_pfr" instead of "nf". Correct long double handling.
2612 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2614 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2615 space for long double return value. Adjust stack frame and offsets.
2616 Load f2 long double return.
2618 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2620 * src/types.c: Use 16 byte long double for POWERPC64.
2622 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2624 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2625 when the structure return address is passed in %o0.
2626 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2627 (ffi_v9_layout_struct): Align the field following a nested structure
2628 on a word boundary. Use memmove instead of memcpy.
2629 (ffi_call): Update call to ffi_V9_return_struct.
2630 (ffi_prep_closure): Define 'ctx' only for V8.
2631 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2632 and ffi_closure_sparc_inner_v9.
2633 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2634 Always skip the structure return address. For structures and long
2635 doubles, copy the argument directly.
2636 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2637 if required. Shift the maximum floating-point slot accordingly. For
2638 big structures, copy the argument directly; otherwise, left-justify the
2639 argument and call ffi_v9_layout_struct to lay out the structure on
2641 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2642 (ffi_closure_v8): Pass the structure return address. Update call to
2643 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2644 Skip the 'unimp' insn when returning long doubles and structures.
2645 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2646 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2647 FFI_TYPE_INT handling. Load structures both in integers and
2648 floating-point registers on return.
2649 * README: Update status of the SPARC port.
2651 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2653 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2655 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2657 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2659 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2660 value as of type ffi_arg, not unsigned int.
2662 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2664 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2667 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2669 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2672 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2674 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2677 2003-12-10 Richard Henderson <rth@redhat.com>
2679 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2680 size_t instead of int.
2682 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2684 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2685 * testsuite/libffi.call/many_win32.c (main): Replace variable
2686 int i with unsigned long ul.
2688 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2689 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2690 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2691 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2692 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2693 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2694 * testsuite/libffi.call/cls_align_float.c: Likewise.
2695 * testsuite/libffi.call/cls_align_double.c: Likewise.
2696 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2697 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2699 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2702 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2703 Align arguments to 32 bits.
2705 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2708 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2709 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2710 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2711 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2712 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2713 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2715 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2717 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2719 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2720 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2722 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2724 * Makefile.in: Rebuilt.
2725 * configure: Likewise.
2726 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2729 2003-11-21 Andreas Jaeger <aj@suse.de>
2730 Andreas Tobler <a.tobler@schweiz.ch>
2732 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2733 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2734 * Makefile.in: Rebuilt.
2735 * aclocal.m4: Likewise.
2736 * configure: Likewise.
2737 * fficonfig.h.in: Likewise.
2738 * testsuite/lib/libffi-dg.exp: Add include dir.
2739 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2740 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2741 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2742 for ffi_closure if available.
2743 * testsuite/libffi.call/closure_fn1.c: Likewise.
2744 * testsuite/libffi.call/closure_fn2.c: Likewise.
2745 * testsuite/libffi.call/closure_fn3.c: Likewise.
2746 * testsuite/libffi.call/closure_fn4.c: Likewise.
2747 * testsuite/libffi.call/closure_fn5.c: Likewise.
2748 * testsuite/libffi.call/cls_12byte.c: Likewise.
2749 * testsuite/libffi.call/cls_16byte.c: Likewise.
2750 * testsuite/libffi.call/cls_18byte.c: Likewise.
2751 * testsuite/libffi.call/cls_19byte.c: Likewise.
2752 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2753 * testsuite/libffi.call/cls_20byte.c: Likewise.
2754 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2755 * testsuite/libffi.call/cls_24byte.c: Likewise.
2756 * testsuite/libffi.call/cls_2byte.c: Likewise.
2757 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2758 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2759 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2760 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2761 * testsuite/libffi.call/cls_4byte.c: Likewise.
2762 * testsuite/libffi.call/cls_5byte.c: Likewise.
2763 * testsuite/libffi.call/cls_64byte.c: Likewise.
2764 * testsuite/libffi.call/cls_6byte.c: Likewise.
2765 * testsuite/libffi.call/cls_7byte.c: Likewise.
2766 * testsuite/libffi.call/cls_8byte.c: Likewise.
2767 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2768 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2769 * testsuite/libffi.call/cls_double.c: Likewise.
2770 * testsuite/libffi.call/cls_float.c: Likewise.
2771 * testsuite/libffi.call/cls_schar.c: Likewise.
2772 * testsuite/libffi.call/cls_sint.c: Likewise.
2773 * testsuite/libffi.call/cls_sshort.c: Likewise.
2774 * testsuite/libffi.call/cls_uchar.c: Likewise.
2775 * testsuite/libffi.call/cls_uint.c: Likewise.
2776 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2777 * testsuite/libffi.call/cls_ushort.c: Likewise.
2778 * testsuite/libffi.call/nested_struct.c: Likewise.
2779 * testsuite/libffi.call/nested_struct1.c: Likewise.
2780 * testsuite/libffi.call/nested_struct2.c: Likewise.
2781 * testsuite/libffi.call/nested_struct3.c: Likewise.
2782 * testsuite/libffi.call/problem1.c: Likewise.
2783 * testsuite/libffi.special/unwindtest.cc: Likewise.
2785 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2787 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2789 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2791 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2792 Add -lgcc_s to additional flags.
2794 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2796 * configure.in, include/Makefile.am: PR libgcj/11147, install
2797 the ffitarget.h header file in a gcc versioned and target
2799 * configure: Regenerated.
2800 * Makefile.in, include/Makefile.in: Likewise.
2801 * testsuite/Makefile.in: Likewise.
2803 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2805 * testsuite/libffi.call/closure_fn0.c: Print result and check
2806 with dg-output to make debugging easier.
2807 * testsuite/libffi.call/closure_fn1.c: Likewise.
2808 * testsuite/libffi.call/closure_fn2.c: Likewise.
2809 * testsuite/libffi.call/closure_fn3.c: Likewise.
2810 * testsuite/libffi.call/closure_fn4.c: Likewise.
2811 * testsuite/libffi.call/closure_fn5.c: Likewise.
2812 * testsuite/libffi.call/cls_12byte.c: Likewise.
2813 * testsuite/libffi.call/cls_16byte.c: Likewise.
2814 * testsuite/libffi.call/cls_18byte.c: Likewise.
2815 * testsuite/libffi.call/cls_19byte.c: Likewise.
2816 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2817 * testsuite/libffi.call/cls_20byte.c: Likewise.
2818 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2819 * testsuite/libffi.call/cls_24byte.c: Likewise.
2820 * testsuite/libffi.call/cls_2byte.c: Likewise.
2821 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2822 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2823 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2824 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2825 * testsuite/libffi.call/cls_4byte.c: Likewise.
2826 * testsuite/libffi.call/cls_5byte.c: Likewise.
2827 * testsuite/libffi.call/cls_64byte.c: Likewise.
2828 * testsuite/libffi.call/cls_6byte.c: Likewise.
2829 * testsuite/libffi.call/cls_7byte.c: Likewise.
2830 * testsuite/libffi.call/cls_8byte.c: Likewise.
2831 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2832 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2833 * testsuite/libffi.call/cls_double.c: Likewise.
2834 * testsuite/libffi.call/cls_float.c: Likewise.
2835 * testsuite/libffi.call/cls_schar.c: Likewise.
2836 * testsuite/libffi.call/cls_sint.c: Likewise.
2837 * testsuite/libffi.call/cls_sshort.c: Likewise.
2838 * testsuite/libffi.call/cls_uchar.c: Likewise.
2839 * testsuite/libffi.call/cls_uint.c: Likewise.
2840 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2841 * testsuite/libffi.call/cls_ushort.c: Likewise.
2842 * testsuite/libffi.call/problem1.c: Likewise.
2844 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2847 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2849 * testsuite/libffi.call/cls_9byte2.c: New test case.
2850 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2851 * testsuite/libffi.call/cls_64byte.c: Likewise.
2852 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2853 * testsuite/libffi.call/cls_19byte.c: Likewise.
2854 * testsuite/libffi.call/cls_18byte.c: Likewise.
2855 * testsuite/libffi.call/closure_fn4.c: Likewise.
2856 * testsuite/libffi.call/closure_fn5.c: Likewise.
2857 * testsuite/libffi.call/cls_schar.c: Likewise.
2858 * testsuite/libffi.call/cls_sint.c: Likewise.
2859 * testsuite/libffi.call/cls_sshort.c: Likewise.
2860 * testsuite/libffi.call/nested_struct2.c: Likewise.
2861 * testsuite/libffi.call/nested_struct3.c: Likewise.
2863 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2865 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2866 * testsuite/libffi.call/cls_uchar.c: Likewise.
2867 * testsuite/libffi.call/cls_uint.c: Likewise.
2868 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2869 * testsuite/libffi.call/cls_ushort.c: Likewise.
2870 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2872 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2874 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2877 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2879 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2880 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2882 2003-10-22 David Daney <ddaney@avtrex.com>
2884 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2885 __attribute__((__mode__(__SI__))) and friends.
2887 2003-10-22 Andreas Schwab <schwab@suse.de>
2889 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2891 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2893 * configure.in: AC_LINK_FILES(ffitarget.h).
2894 * configure: Regenerate.
2895 * Makefile.in: Likewise.
2896 * include/Makefile.in: Likewise.
2897 * testsuite/Makefile.in: Likewise.
2898 * fficonfig.h.in: Likewise.
2900 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2901 Richard Henderson <rth@redhat.com>
2903 Avoid that ffi.h includes fficonfig.h.
2905 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2906 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2907 (TARGET_SRC_MIPS_SGI): Removed.
2908 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2909 (MIPS_SGI): Removed.
2910 (CLEANFILES): Removed.
2911 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2913 * acconfig.h: Removed.
2914 * configure.in: Compute sizeofs only for double and long double.
2915 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2916 into AC_DEFINE instead of using acconfig.h. Create
2917 include/ffitarget.h instead of include/fficonfig.h. Rename
2918 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2919 AC_DEFINE EH_FRAME_FLAGS.
2920 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2921 (hack_DATA): Add ffitarget.h.
2922 * include/ffi.h.in: Remove all system specific definitions.
2923 Declare raw API even if it is not installed, why bother?
2924 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2925 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2926 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2927 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2928 * include/ffi_common.h (bool): Do not define.
2929 (ffi_assert): Accept failed assertion.
2930 (ffi_type_test): Return void and accept file/line.
2931 (FFI_ASSERT): Pass stringized failed assertion.
2932 (FFI_ASSERT_AT): New macro.
2933 (FFI_ASSERT_VALID_TYPE): New macro.
2934 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2935 UINT64, SINT64): Define here with gcc's __attribute__ macro
2937 (FLOAT32, ALIGN): Define here instead of in ffi.h
2938 * include/ffi-mips.h: Removed. Its content moved to
2939 src/mips/ffitarget.h after separating assembly and C sections.
2940 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2941 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2942 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2943 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2944 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2945 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2946 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2947 (ffi_type_test): Rewritten.
2948 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2949 FFI_ASSERT_VALID_TYPE.
2950 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2951 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2952 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2953 src/s390/ffitarget.h, src/sh/ffitarget.h,
2954 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2955 src/x86/ffitarget.h: New files.
2956 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2957 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2958 src/powerpc/aix.S, src/powerpc/darwin.S,
2959 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2960 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2961 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2962 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2963 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2966 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2968 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2969 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2971 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2973 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2974 Used when FFI_DEBUG = 1.
2976 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2978 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2981 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2983 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2984 FFI_MIPS_O32 for O32 ABI.
2986 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2988 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2989 SPARC64. Cleanup whitespaces.
2991 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2993 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2994 strongarm, xscale. Cleanup whitespaces.
2995 * testsuite/libffi.call/closure_fn1.c: Likewise.
2996 * testsuite/libffi.call/closure_fn2.c: Likewise.
2997 * testsuite/libffi.call/closure_fn3.c: Likewise.
2998 * testsuite/libffi.call/cls_12byte.c: Likewise.
2999 * testsuite/libffi.call/cls_16byte.c: Likewise.
3000 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3001 * testsuite/libffi.call/cls_20byte.c: Likewise.
3002 * testsuite/libffi.call/cls_24byte.c: Likewise.
3003 * testsuite/libffi.call/cls_2byte.c: Likewise.
3004 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3005 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3006 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3007 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3008 * testsuite/libffi.call/cls_4byte.c: Likewise.
3009 * testsuite/libffi.call/cls_5byte.c: Likewise.
3010 * testsuite/libffi.call/cls_6byte.c: Likewise.
3011 * testsuite/libffi.call/cls_7byte.c: Likewise.
3012 * testsuite/libffi.call/cls_8byte.c: Likewise.
3013 * testsuite/libffi.call/cls_double.c: Likewise.
3014 * testsuite/libffi.call/cls_float.c: Likewise.
3015 * testsuite/libffi.call/cls_uchar.c: Likewise.
3016 * testsuite/libffi.call/cls_uint.c: Likewise.
3017 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3018 * testsuite/libffi.call/cls_ushort.c: Likewise.
3019 * testsuite/libffi.call/nested_struct.c: Likewise.
3020 * testsuite/libffi.call/nested_struct1.c: Likewise.
3021 * testsuite/libffi.call/problem1.c: Likewise.
3022 * testsuite/libffi.special/unwindtest.cc: Likewise.
3023 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
3025 2003-09-18 David Edelsohn <edelsohn@gnu.org>
3027 * src/powerpc/aix.S: Cleanup whitespaces.
3028 * src/powerpc/aix_closure.S: Likewise.
3030 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3032 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
3033 * src/powerpc/darwin_closure.S: Likewise.
3034 * src/powerpc/ffi_darwin.c: Likewise.
3036 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3037 David Edelsohn <edelsohn@gnu.org>
3039 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
3040 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
3042 * src/powerpc/darwin_closure.S: Likewise.
3043 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
3044 according to the Darwin/AIX ABI.
3045 (ffi_prep_cif_machdep): Likewise.
3046 (ffi_closure_helper_DARWIN): Likewise.
3047 Remove the outgoing parameter stack logic. Simplify the evaluation
3048 of the different CASE types.
3049 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
3050 statement in the trampoline code.
3052 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
3054 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
3055 for the register size.
3056 (ffi_closure_helper_SYSV): Handle the structure return value
3058 (ffi_closure_helper_SYSV): Return the appropriate type when
3059 the registers are used for the structure return value.
3060 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
3061 the 64-bit return value. Update copyright years.
3063 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3065 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
3066 srcdir for ffi_mips.h.
3068 2003-09-12 Alan Modra <amodra@bigpond.net.au>
3070 * src/prep_cif.c (initialize_aggregate): Include tail padding in
3072 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
3073 placement of float result.
3074 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
3075 cast of "resp" for big-endian 64 bit machines.
3077 2003-09-11 Alan Modra <amodra@bigpond.net.au>
3079 * src/types.c (double, longdouble): Merge identical SH and ARM
3080 typedefs, and add POWERPC64.
3081 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
3082 struct split over gpr and rest.
3083 (ffi_prep_cif_machdep): Correct intarg_count for structures.
3084 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
3086 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
3088 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
3091 2003-09-09 Alan Modra <amodra@bigpond.net.au>
3093 * configure: Regenerate.
3095 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3097 * Makefile.am: Remove build rules for ffitest.
3098 * Makefile.in: Rebuilt.
3100 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3102 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3103 about implicit declaration of abort().
3105 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3107 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3108 * Makefile.in: Rebuilt.
3109 * configure.in: Add dejagnu test framework.
3110 * configure: Rebuilt.
3112 * testsuite/Makefile.am: New file.
3113 * testsuite/Makefile.in: Built
3114 * testsuite/lib/libffi-dg.exp: New file.
3115 * testsuite/config/default.exp: Likewise.
3116 * testsuite/libffi.call/call.exp: Likewise.
3117 * testsuite/libffi.call/ffitest.h: Likewise.
3118 * testsuite/libffi.call/closure_fn0.c: Likewise.
3119 * testsuite/libffi.call/closure_fn1.c: Likewise.
3120 * testsuite/libffi.call/closure_fn2.c: Likewise.
3121 * testsuite/libffi.call/closure_fn3.c: Likewise.
3122 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3123 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3124 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3125 * testsuite/libffi.call/cls_2byte.c: Likewise.
3126 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3127 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3128 * testsuite/libffi.call/cls_4byte.c: Likewise.
3129 * testsuite/libffi.call/cls_5byte.c: Likewise.
3130 * testsuite/libffi.call/cls_6byte.c: Likewise.
3131 * testsuite/libffi.call/cls_7byte.c: Likewise.
3132 * testsuite/libffi.call/cls_8byte.c: Likewise.
3133 * testsuite/libffi.call/cls_12byte.c: Likewise.
3134 * testsuite/libffi.call/cls_16byte.c: Likewise.
3135 * testsuite/libffi.call/cls_20byte.c: Likewise.
3136 * testsuite/libffi.call/cls_24byte.c: Likewise.
3137 * testsuite/libffi.call/cls_double.c: Likewise.
3138 * testsuite/libffi.call/cls_float.c: Likewise.
3139 * testsuite/libffi.call/cls_uchar.c: Likewise.
3140 * testsuite/libffi.call/cls_uint.c: Likewise.
3141 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3142 * testsuite/libffi.call/cls_ushort.c: Likewise.
3143 * testsuite/libffi.call/float.c: Likewise.
3144 * testsuite/libffi.call/float1.c: Likewise.
3145 * testsuite/libffi.call/float2.c: Likewise.
3146 * testsuite/libffi.call/many.c: Likewise.
3147 * testsuite/libffi.call/many_win32.c: Likewise.
3148 * testsuite/libffi.call/nested_struct.c: Likewise.
3149 * testsuite/libffi.call/nested_struct1.c: Likewise.
3150 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3151 * testsuite/libffi.call/problem1.c: Likewise.
3152 * testsuite/libffi.call/promotion.c: Likewise.
3153 * testsuite/libffi.call/return_ll.c: Likewise.
3154 * testsuite/libffi.call/return_sc.c: Likewise.
3155 * testsuite/libffi.call/return_uc.c: Likewise.
3156 * testsuite/libffi.call/strlen.c: Likewise.
3157 * testsuite/libffi.call/strlen_win32.c: Likewise.
3158 * testsuite/libffi.call/struct1.c: Likewise.
3159 * testsuite/libffi.call/struct2.c: Likewise.
3160 * testsuite/libffi.call/struct3.c: Likewise.
3161 * testsuite/libffi.call/struct4.c: Likewise.
3162 * testsuite/libffi.call/struct5.c: Likewise.
3163 * testsuite/libffi.call/struct6.c: Likewise.
3164 * testsuite/libffi.call/struct7.c: Likewise.
3165 * testsuite/libffi.call/struct8.c: Likewise.
3166 * testsuite/libffi.call/struct9.c: Likewise.
3167 * testsuite/libffi.special/special.exp: New file.
3168 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3169 * testsuite/libffi.special/unwindtest.cc: Likewise.
3172 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3174 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3177 2003-08-02 Alan Modra <amodra@bigpond.net.au>
3179 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3181 (ffi_closure_helper_LINUX64): Likewise.
3182 * src/powerpc/linux64.S: Remove code writing to parm save area.
3183 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3184 address in lr from ffi_closure_helper_LINUX64 call to calculate
3185 table address. Optimize function tail.
3187 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3189 * src/sparc/ffi.c: Handle all floating point registers.
3190 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3192 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3194 * README: Note that libffi is not part of GCC. Update the project
3197 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3199 * src/powerpc/ppc_closure.S: Include ffi.h.
3201 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3203 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3204 Use C style comments.
3206 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3208 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3209 * Makefile.in: Regenerate.
3210 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3211 * configure: Regenerate.
3212 * include/ffi.h.in: Add SHmedia support.
3213 * src/sh64/ffi.c: New file.
3214 * src/sh64/sysv.S: New file.
3216 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3218 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3219 should be read-only.
3220 * configure: Rebuilt.
3221 * fficonfig.h.in: Rebuilt.
3222 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3223 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3224 * src/powerpc/linux64.S: Likewise.
3225 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3226 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3227 if -fpic/-fPIC/-mrelocatable.
3228 * src/powerpc/powerpc_closure.S: Likewise.
3229 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3230 #write in .eh_frame flags.
3231 * src/sparc/v9.S: Likewise.
3232 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3233 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3234 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3236 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3238 Fixes PR bootstrap/10656
3239 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3240 support for .register pseudo-op.
3241 * src/sparc/v8.S: Use it.
3242 * fficonfig.h.in: Rebuilt.
3243 * configure: Rebuilt.
3245 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3247 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3248 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3249 Make it the default on POWERPC64.
3250 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3251 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3252 * configure: Rebuilt.
3253 * src/powerpc/ffi.c (hidden): Define.
3254 (ffi_prep_args_SYSV): Renamed from
3255 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3256 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3257 ASM_NEEDS_REGISTERS64): New.
3258 (ffi_prep_args64): New function.
3259 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3260 (ffi_call): Likewise.
3261 (ffi_prep_closure): Likewise.
3262 (flush_icache): Surround by #ifndef POWERPC64.
3263 (ffi_dblfl): New union type.
3264 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3265 (ffi_closure_helper_LINUX64): New function.
3266 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3268 * src/powerpc/sysv.S: Likewise.
3269 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3270 * src/powerpc/linux64.S: New file.
3271 * src/powerpc/linux64_closure.S: New file.
3272 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3273 src/powerpc/linux64_closure.S.
3274 (TARGET_SRC_POWERPC): Likewise.
3276 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3277 closure_test_fn3): Fix result printing on big-endian 64-bit
3279 (main): Print tst2_arg instead of uninitialized tst2_result.
3281 * src/ffitest.c (main): Hide what closure pointer really points to
3284 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3286 * configure.in (arm-*-netbsdelf*): Add configuration.
3287 (configure): Regenerated.
3289 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3291 * include/Makefile.in: Regenerate.
3293 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3295 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3297 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3298 Receive closure pointer through parameter, read args using
3299 __builtin_dwarf_cfa.
3300 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3302 2003-03-12 Andreas Schwab <schwab@suse.de>
3304 * configure.in: Avoid trailing /. in toolexeclibdir.
3305 * configure: Rebuilt.
3307 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3309 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3311 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3313 * libffi/src/powerpc/darwin_closure.S:
3314 Fix alignement bug, allocate 8 bytes for the result.
3315 * libffi/src/powerpc/aix_closure.S:
3317 * libffi/src/powerpc/ffi_darwin.c:
3318 Update stackframe description for aix/darwin_closure.S.
3320 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3322 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3325 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3326 Andreas Schwab <schwab@suse.de>
3328 * configure.in: Adjust command to source config-ml.in to account
3329 for changes to the libffi_basedir definition.
3330 (libffi_basedir): Remove ${srcdir} from value and include trailing
3333 * configure: Regenerate.
3335 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3337 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3339 2003-01-28 Andrew Haley <aph@redhat.com>
3341 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3342 * src/x86/ffi64.c (ffi_prep_closure): New.
3343 (ffi_closure_UNIX64_inner): New.
3344 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3346 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
3348 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3349 Remove USE_LIBDIR conditional.
3350 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3351 * Makefile.in, configure: Rebuilt.
3353 2003-01027 David Edelsohn <edelsohn@gnu.org>
3355 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3356 * Makefile.in: Regenerate.
3358 2003-01-22 Andrew Haley <aph@redhat.com>
3360 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3363 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3365 * src/powerpc/darwin.S: Add unwind info.
3366 * src/powerpc/darwin_closure.S: Likewise.
3368 2003-01-14 Andrew Haley <aph@redhat.com>
3370 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3371 (ffi_prep_cif_machdep): Likewise.
3372 * src/x86/unix64.S: Add unwind info.
3374 2003-01-14 Andreas Jaeger <aj@suse.de>
3376 * src/ffitest.c (main): Only use ffi_closures if those are
3379 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3381 * libffi/src/ffitest.c
3382 add closure testcases
3384 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3386 * libffi/src/powerpc/ffi.c
3387 fix alignment bug for float (4 byte aligned iso 8 byte)
3389 2003-01-09 Geoffrey Keating <geoffk@apple.com>
3391 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3392 * src/powerpc/darwin.S: Remove RCS version string.
3394 2003-01-03 Jeff Sturm <jsturm@one-point.com>
3396 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3397 * src/ffitest.c (main): Use static storage for closure.
3398 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3399 * src/sparc/v8.S (ffi_closure_v8): New.
3400 * src/sparc/v9.S (ffi_closure_v9): New.
3402 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3404 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3405 enumeration for X86_WIN32.
3406 * src/x86/win32.S: Added ffi_call_STDCALL function
3408 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3409 switch cases for recognising FFI_STDCALL and
3410 calling ffi_call_STDCALL if target is X86_WIN32.
3411 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3412 stdcall versions of the "my_strlen" and "many"
3413 test functions (for X86_WIN32).
3414 Added test cases to test stdcall invocation using
3417 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3419 * src/sh/sysv.S: Add DWARF2 unwind info.
3421 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3423 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3425 2002-11-26 Jim Wilson <wilson@redhat.com>
3427 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3429 2002-11-23 H.J. Lu <hjl@gnu.org>
3431 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3432 Include ../config/accross.m4.
3433 * aclocal.m4; Rebuild.
3434 * configure: Likewise.
3436 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3438 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3440 2002-11-11 DJ Delorie <dj@redhat.com>
3442 * configure.in: Look for common files in the right place.
3444 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3446 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3447 raw data as _Jv_word values, not ffi_raw.
3448 (ffi_java_ptrarray_to_raw): Likewise.
3449 (ffi_java_rvalue_to_raw): New function.
3450 (ffi_java_raw_call): Call it.
3451 (ffi_java_raw_to_rvalue): New function.
3452 (ffi_java_translate_args): Call it.
3453 * src/ffitest.c (closure_test_fn): Interpret return value
3454 as ffi_arg, not int.
3455 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3456 FFI_TYPE_POINTER case.
3457 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3458 values extended to word size.
3460 2002-10-02 Andreas Jaeger <aj@suse.de>
3462 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3464 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
3466 * include/ffi.h.in: Fix i386 win32 compilation.
3468 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3470 * configure.in: Add s390x-*-linux-* target.
3471 * configure: Regenerate.
3472 * include/ffi.h.in: Define S390X for s390x targets.
3473 (FFI_CLOSURES): Define for s390/s390x.
3474 (FFI_TRAMPOLINE_SIZE): Likewise.
3475 (FFI_NATIVE_RAW_API): Likewise.
3476 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3477 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
3478 * src/s390/ffi.c: Major rework of existing code. Add support for
3479 s390x targets. Add closure support.
3480 * src/s390/sysv.S: Likewise.
3482 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
3484 * src/arm/sysv.S: Fix typo.
3486 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
3488 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3489 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3490 (ffi_call_SYSV): Handle soft-float.
3492 2002-09-27 Bo Thorsen <bo@suse.de>
3494 * include/ffi.h.in: Fix multilib x86-64 support.
3496 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3498 * Makefile.am (all-multi): Fix multilib parallel build.
3500 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3502 * configure.in (sh[34]*-*-linux*): Add brackets.
3503 * configure: Regenerate.
3505 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3507 * Makefile.am: Add SH support.
3508 * Makefile.in: Regenerate.
3509 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3510 * configure: Regenerate.
3511 * include/ffi.h.in: Add SH support.
3512 * src/sh/ffi.c: New file.
3513 * src/sh/sysv.S: New file.
3514 * src/types.c: Add SH support.
3516 2002-07-16 Bo Thorsen <bo@suse.de>
3518 * src/x86/ffi64.c: New file that adds x86-64 support.
3519 * src/x86/unix64.S: New file that handles argument setup for
3521 * src/x86/sysv.S: Don't use this on x86-64.
3522 * src/x86/ffi.c: Don't use this on x86-64.
3524 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3526 * src/ffitest.c (struct6): New test that tests a special case in
3528 (struct7): Likewise.
3529 (struct8): Likewise.
3530 (struct9): Likewise.
3531 (closure_test_fn): Silence warning about this when it's not used.
3532 (main): Add the new tests.
3533 (main): Fix a couple of wrong casts and silence some compiler warnings.
3534 * include/ffi.h.in: Add x86-64 ABI definition.
3535 * fficonfig.h.in: Regenerate.
3536 * Makefile.am: Add x86-64 support.
3537 * configure.in: Likewise.
3538 * Makefile.in: Regenerate.
3539 * configure: Likewise.
3541 2002-06-24 Bo Thorsen <bo@suse.de>
3543 * src/types.c: Merge settings for similar architectures.
3544 Add x86-64 sizes and alignments.
3546 2002-06-23 Bo Thorsen <bo@suse.de>
3548 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3549 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3550 * src/mips/ffi.c (ffi_prep_args): Likewise.
3551 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3553 2002-07-18 H.J. Lu (hjl@gnu.org)
3555 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3556 (libffi_la_SOURCES): Support MIPS_LINUX.
3557 (libffi_convenience_la_SOURCES): Likewise.
3558 * Makefile.in: Regenerated.
3560 * configure.in (mips64*-*): Skip.
3561 (mips*-*-linux*): New.
3562 * configure: Regenerated.
3564 * src/mips/ffi.c: Include <sgidefs.h>.
3566 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3568 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3570 2002-05-27 Roger Sayle <roger@eyesopen.com>
3572 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3574 2002-05-27 Bo Thorsen <bo@suse.de>
3576 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3579 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3581 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3582 beginning of function (for older apple cc).
3584 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3586 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3587 script entry, and set LD to it when configuring multilibs.
3588 * configure: Rebuilt.
3590 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3592 * configure.in (sparc64-*-netbsd*): Add target.
3593 (sparc-*-netbsdelf*): Likewise.
3594 * configure: Regenerate.
3596 2002-04-28 David S. Miller <davem@redhat.com>
3598 * configure.in, configure: Fix SPARC test in previous change.
3600 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3602 * Makefile.am: Add Linux for S/390 support.
3603 * Makefile.in: Regenerate.
3604 * configure.in: Add Linux for S/390 support.
3605 * configure: Regenerate.
3606 * include/ffi.h.in: Add Linux for S/390 support.
3607 * src/s390/ffi.c: New file from libffi CVS tree.
3608 * src/s390/sysv.S: New file from libffi CVS tree.
3610 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3612 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3614 * src/sparc/v8.S: Use it.
3615 * src/sparc/v9.S: Likewise.
3616 * fficonfig.h.in: Rebuilt.
3617 * configure: Rebuilt.
3619 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3621 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3623 * src/ia64/unix.S: Add unwind information. Fix comments.
3624 Save sp in a way that's compatible with unwind info.
3625 (ffi_call_unix): Correctly restore sp in all cases.
3626 * src/ia64/ffi.c: Add, fix comments.
3628 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3630 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3632 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3634 * configure.in (alpha*-*-netbsd*): Add target.
3635 * configure: Regenerate.
3637 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3639 * src/sparc/v8.S: Add unwind info.
3640 * src/sparc/v9.S: Likewise.
3642 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3644 * configure.in: Enable i*86-*-netbsdelf*.
3645 * configure: Rebuilt.
3647 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3650 * src/mips/n32.s: Delete
3651 * src/mips/o32.s: Delete
3653 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3655 * configure.in: Enable alpha*-*-freebsd*.
3656 * configure: Rebuilt.
3658 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3660 * Makefile.am: libfficonvenience -> libffi_convenience.
3661 * Makefile.in: Rebuilt.
3663 * Makefile.am: Define ffitest_OBJECTS.
3664 * Makefile.in: Rebuilt.
3666 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3667 David Edelsohn <edelsohn@gnu.org>
3669 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3670 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3671 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3672 * Makefile.in: Regenerate.
3673 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3674 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3675 (flush_icache, flush_range): New functions.
3676 (ffi_closure_helper_DARWIN): New function.
3677 * src/powerpc/aix_closure.S: New file.
3678 * src/powerpc/darwin_closure.S: New file.
3680 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3682 * include/ffi.h.in: Add typedef for ffi_arg.
3683 * src/ffitest.c (main): Declare rint with ffi_arg.
3685 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3687 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3688 number of GPRs for floating-point arguments.
3690 2002-01-31 Anthony Green <green@redhat.com>
3692 * configure: Rebuilt.
3693 * configure.in: Replace CHECK_SIZEOF and endian tests with
3694 cross-compiler friendly macros.
3695 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3698 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3700 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3701 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3703 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3705 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3706 (TARGET_SRC_POWERPC_AIX): New.
3707 (POWERPC_AIX): New stanza.
3708 * Makefile.in: Regenerate.
3709 * configure.in: Add AIX case.
3710 * configure: Regenerate.
3711 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3712 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3713 size. Fix "long double" support.
3714 (ffi_call): Add FFI_AIX case.
3715 * src/powerpc/aix.S: New.
3717 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3719 Implement Darwin PowerPC ABI.
3720 * configure.in: Handle powerpc-*-darwin*.
3721 * Makefile.am: Set source files for POWERPC_DARWIN.
3722 * configure: Rebuilt.
3723 * Makefile.in: Rebuilt.
3724 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3726 * src/powerpc/darwin.S: New file.
3727 * src/powerpc/ffi_darwin.c: New file.
3729 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3731 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3733 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3735 * src/x86/sysv.S: Avoid gas-only .balign directive.
3736 Use C style comments.
3738 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3740 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3741 Fixes PR bootstrap/3563.
3743 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3745 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3747 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3749 * configure.in: Recognize sparc*-sun-* host.
3750 * configure: Regenerate.
3752 2001-06-06 Andrew Haley <aph@redhat.com>
3754 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
3756 2001-06-03 Andrew Haley <aph@redhat.com>
3758 * src/alpha/osf.S: Add unwind info.
3759 * src/powerpc/sysv.S: Add unwind info.
3760 * src/powerpc/ppc_closure.S: Likewise.
3762 2000-05-31 Jeff Sturm <jsturm@one-point.com>
3764 * configure.in: Fix AC_ARG_ENABLE usage.
3765 * configure: Rebuilt.
3767 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3769 * configure.in: Remove warning about beta code.
3770 * configure: Rebuilt.
3772 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3774 * src/ia64/unix.S: Restore stack pointer when returning from
3776 * src/ia64/ffi.c: Fix typo in comment.
3778 2001-04-18 Jim Wilson <wilson@redhat.com>
3780 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3781 to eliminate RAW DV.
3783 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3785 * Makefile.am: Make a libtool convenience library.
3786 * Makefile.in: Rebuilt.
3788 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3790 * configure.in: Use different syntax for subdirectory creation.
3791 * configure: Rebuilt.
3793 2001-03-27 Jon Beniston <jon@beniston.com>
3795 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3796 * configure: Rebuilt.
3797 * Makefile.am: Added X86_WIN32 target support.
3798 * Makefile.in: Rebuilt.
3800 * include/ffi.h.in: Added X86_WIN32 target support.
3802 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3803 * src/types.c: Added X86_WIN32 target support.
3805 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3806 stuff removed and made to work with CygWin's gas.
3808 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3810 * configure.in: Make target subdirectory in build dir.
3811 * Makefile.am: Override suffix based rules to specify correct output
3813 * Makefile.in: Rebuilt.
3814 * configure: Rebuilt.
3816 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3818 * src/powerpc/ppc_closure.S: New file.
3819 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3820 involving long long and register pairs.
3821 (ffi_prep_closure): New function.
3822 (flush_icache): Likewise.
3823 (ffi_closure_helper_SYSV): Likewise.
3824 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3825 (FFI_TRAMPOLINE_SIZE): Likewise.
3826 (FFI_NATIVE_RAW_API): Likewise.
3827 * Makefile.in: Rebuilt.
3828 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3829 (TARGET_SRC_POWERPC): Likewise.
3831 2001-03-19 Tom Tromey <tromey@redhat.com>
3833 * Makefile.in: Rebuilt.
3834 * Makefile.am (ffitest_LDFLAGS): New macro.
3836 2001-03-02 Nick Clifton <nickc@redhat.com>
3838 * include/ffi.h.in: Remove RCS ident string.
3839 * include/ffi_mips.h: Remove RCS ident string.
3840 * src/debug.c: Remove RCS ident string.
3841 * src/ffitest.c: Remove RCS ident string.
3842 * src/prep_cif.c: Remove RCS ident string.
3843 * src/types.c: Remove RCS ident string.
3844 * src/alpha/ffi.c: Remove RCS ident string.
3845 * src/alpha/osf.S: Remove RCS ident string.
3846 * src/arm/ffi.c: Remove RCS ident string.
3847 * src/arm/sysv.S: Remove RCS ident string.
3848 * src/mips/ffi.c: Remove RCS ident string.
3849 * src/mips/n32.S: Remove RCS ident string.
3850 * src/mips/o32.S: Remove RCS ident string.
3851 * src/sparc/ffi.c: Remove RCS ident string.
3852 * src/sparc/v8.S: Remove RCS ident string.
3853 * src/sparc/v9.S: Remove RCS ident string.
3854 * src/x86/ffi.c: Remove RCS ident string.
3855 * src/x86/sysv.S: Remove RCS ident string.
3857 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3859 * include/ffi.h.in: Change sourceware.cygnus.com references to
3862 2000-12-09 Richard Henderson <rth@redhat.com>
3864 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3865 (ffi_closure_osf_inner): Index rather than increment avalue
3866 and arg_types. Give ffi_closure_osf the raw return value type.
3867 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3870 2000-12-07 Richard Henderson <rth@redhat.com>
3872 * src/raw_api.c (ffi_translate_args): Fix typo.
3873 (ffi_prep_closure): Likewise.
3875 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3876 FFI_TRAMPOLINE_SIZE.
3877 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3878 cif->bytes for new ffi_call_osf implementation.
3879 (ffi_prep_args): Absorb into ...
3880 (ffi_call): ... here. Do all stack allocation here and
3881 avoid a callback function.
3882 (ffi_prep_closure, ffi_closure_osf_inner): New.
3883 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3884 (ffi_closure_osf): New.
3886 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3888 * config.guess, config.sub, install-sh: Removed.
3889 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3890 * Makefile.in: Rebuilt.
3892 * acinclude.m4: Include libtool macros from the top level.
3893 * aclocal.m4, configure: Rebuilt.
3895 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3897 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3898 * configure: Rebuilt.
3900 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3902 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3903 memory correctly. Use conditional instructions, not branches where
3906 2000-05-04 Tom Tromey <tromey@cygnus.com>
3908 * configure: Rebuilt.
3909 * configure.in: Match `arm*-*-linux-*'.
3910 From Chris Dornan <cdornan@arm.com>.
3912 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3914 * Makefile.am (SUBDIRS): Define.
3915 (AM_MAKEFLAGS): Likewise.
3916 (Multilib support.): Add section.
3917 * Makefile.in: Rebuilt.
3918 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3919 New variables. Set for gcc using -print-multi-lib. Export them
3921 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3922 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3924 (extra_compiler_flags_value, extra_compiler_flags): Check these
3925 for extra compiler options which need to be passed down in
3928 2000-04-16 Anthony Green <green@redhat.com>
3930 * configure: Rebuilt.
3931 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3933 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3935 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3936 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3937 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3938 Replace all void * sizeofs with sizeof(int).
3939 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3940 different than DOUBLE.
3941 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3942 (ffi_prep_args_v9): New function.
3943 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3944 (ffi_V9_return_struct): New function.
3945 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3946 32bit code (not yet cross-arch calls).
3947 * src/sparc/v8.S: Add struct return delay nop.
3949 * src/sparc/v9.S: New file.
3950 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3951 is used on sparc64 only for structures larger than 32 bytes.
3952 Pass by reference for structures is done for structure arguments
3953 larger than 16 bytes.
3954 * src/ffitest.c (main): Use 64bit rint on sparc64.
3955 Run long long tests on sparc.
3956 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3958 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3960 * configure.in (sparc-*-linux*): New supported target.
3961 (sparc64-*-linux*): Likewise.
3962 * configure: Rebuilt.
3963 * Makefile.am: Add v9.S to SPARC files.
3964 * Makefile.in: Likewise.
3965 (LINK): Surround $(CCLD) into double quotes, so that multilib
3966 compiles work correctly.
3968 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3970 * configure: Rebuilt.
3971 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3972 proposed by Bryce McKinlay.
3974 2000-03-20 Tom Tromey <tromey@cygnus.com>
3976 * Makefile.in: Hand edit for java_raw_api.lo.
3978 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3980 * config.guess, config.sub: Update from the gcc tree.
3981 Fix for PR libgcj/168.
3983 2000-03-03 Tom Tromey <tromey@cygnus.com>
3985 * Makefile.in: Fixed ia64 by hand.
3987 * configure: Rebuilt.
3988 * configure.in (--enable-multilib): New option.
3989 (libffi_basedir): New subst.
3990 (AC_OUTPUT): Added multilib code.
3992 2000-03-02 Tom Tromey <tromey@cygnus.com>
3994 * Makefile.in: Rebuilt.
3995 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3998 2000-02-25 Hans Boehm <boehm@acm.org>
4000 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
4002 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
4004 (ffi_prep_raw_closure): Use ffi_translate_args, not
4005 ffi_closure_translate.
4006 * src/java_raw_api.c: New file.
4007 * src/ffitest.c (closure_test_fn): New function.
4008 (main): Define `rint' as long long on IA64. Added new test when
4009 FFI_CLOSURES is defined.
4010 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
4011 (ffi_abi): Recognize IA64.
4012 (ffi_raw): Added `flt' field.
4013 Added "Java raw API" code.
4014 * configure.in: Recognize ia64.
4015 * Makefile.am (TARGET_SRC_IA64): New macro.
4016 (libffi_la_common_SOURCES): Added java_raw_api.c.
4017 (libffi_la_SOURCES): Define in IA64 case.
4019 2000-01-04 Tom Tromey <tromey@cygnus.com>
4021 * Makefile.in: Rebuilt with newer automake.
4023 1999-12-31 Tom Tromey <tromey@cygnus.com>
4025 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
4027 1999-09-01 Tom Tromey <tromey@cygnus.com>
4029 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
4031 * fficonfig.h.in: Rebuilt.
4032 * configure: Rebuilt.
4033 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
4034 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
4035 * acconfig.h: Don't #undef PACKAGE or VERSION.
4037 1999-08-09 Anthony Green <green@cygnus.com>
4039 * include/ffi.h.in: Try to work around messy header problem
4040 with PACKAGE and VERSION.
4042 * configure: Rebuilt.
4043 * configure.in: Change version to 2.00-beta.
4045 * fficonfig.h.in: Rebuilt.
4046 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
4048 * src/x86/ffi.c (ffi_raw_call): Rename.
4050 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
4052 * src/x86/ffi.c (ffi_closure_SYSV): New function.
4053 (ffi_prep_incoming_args_SYSV): Ditto.
4054 (ffi_prep_closure): Ditto.
4055 (ffi_closure_raw_SYSV): Ditto.
4056 (ffi_prep_raw_closure): More ditto.
4057 (ffi_call_raw): Final ditto.
4059 * include/ffi.h.in: Add definitions for closure and raw API.
4061 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4064 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
4066 * src/raw_api.c: New file.
4068 * include/ffi.h.in (ffi_raw): New type.
4069 (UINT_ARG, SINT_ARG): New defines.
4070 (ffi_closure, ffi_raw_closure): New types.
4071 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4073 * configure.in: Add check for endianness and sizeof void*.
4075 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
4076 instead of directly.
4078 * configure: Rebuilt.
4080 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
4082 * configure.in: Add x86 and powerpc BeOS configurations.
4083 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
4085 1999-05-09 Anthony Green <green@cygnus.com>
4087 * configure.in: Add warning about this being beta code.
4088 Remove src/Makefile.am from the picture.
4089 * configure: Rebuilt.
4091 * Makefile.am: Move logic from src/Makefile.am. Add changes
4092 to support libffi as a target library.
4093 * Makefile.in: Rebuilt.
4095 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
4096 Upgraded to new autoconf, automake, libtool.
4100 * LICENSE: Update copyright date.
4102 * src/Makefile.am, src/Makefile.in: Removed.
4104 1998-11-29 Anthony Green <green@cygnus.com>
4106 * include/ChangeLog: Removed.
4107 * src/ChangeLog: Removed.
4108 * src/mips/ChangeLog: Removed.
4109 * src/sparc/ChangeLog: Remboved.
4110 * src/x86/ChangeLog: Removed.
4112 * ChangeLog.v1: Created.