1 2009-06-09 Andrew Haley <aph@redhat.com>
3 * src/x86/freebsd.S: Add missing file.
5 2009-06-08 Andrew Haley <aph@redhat.com>
7 Import from libffi 3.0.8:
9 * doc/libffi.texi: New file.
10 * doc/libffi.info: Likewise.
11 * doc/stamp-vti: Likewise.
12 * man/Makefile.am: New file.
13 * man/ffi_call.3: New file.
15 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
17 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
19 * configure.ac: Bump version to 3.0.8.
20 parisc*-*-linux*: Add.
21 i386-*-freebsd* | i386-*-openbsd*: Add.
23 AM_CONDITIONAL X86_FREEBSD: Add.
24 AC_CONFIG_FILES: Add man/Makefile.
26 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
28 2009-06-08 Andrew Haley <aph@redhat.com>
30 * README: Import from libffi 3.0.8.
32 2009-06-08 Andrew Haley <aph@redhat.com>
34 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
35 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
36 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
37 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
39 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
40 * testsuite/libffi.call/stret_medium.c: Likewise.
41 * testsuite/libffi.call/stret_large2.c: Likewise.
42 * testsuite/libffi.call/stret_large.c: Likewise.
44 2008-12-26 Timothy Wall <twall@users.sf.net>
46 * testsuite/libffi.call/cls_longdouble.c,
47 testsuite/libffi.call/cls_longdouble_va.c,
48 testsuite/libffi.call/cls_align_longdouble.c,
49 testsuite/libffi.call/cls_align_longdouble_split.c,
50 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
51 failures on x86_64 cygwin/mingw.
53 2008-12-22 Timothy Wall <twall@users.sf.net>
55 * testsuite/libffi.call/closure_fn0.c,
56 testsuite/libffi.call/closure_fn1.c,
57 testsuite/libffi.call/closure_fn2.c,
58 testsuite/libffi.call/closure_fn3.c,
59 testsuite/libffi.call/closure_fn4.c,
60 testsuite/libffi.call/closure_fn5.c,
61 testsuite/libffi.call/closure_fn6.c,
62 testsuite/libffi.call/closure_loc_fn0.c,
63 testsuite/libffi.call/closure_stdcall.c,
64 testsuite/libffi.call/cls_align_pointer.c,
65 testsuite/libffi.call/cls_pointer.c,
66 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
67 pointer to integer (intptr_t).
68 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
70 2008-07-24 Anthony Green <green@redhat.com>
72 * testsuite/libffi.call/cls_dbls_struct.c,
73 testsuite/libffi.call/cls_double_va.c,
74 testsuite/libffi.call/cls_longdouble.c,
75 testsuite/libffi.call/cls_longdouble_va.c,
76 testsuite/libffi.call/cls_pointer.c,
77 testsuite/libffi.call/cls_pointer_stack.c,
78 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
81 2008-03-04 Anthony Green <green@redhat.com>
85 * testsuite/libffi.call/cls_align_longdouble_split2.c
86 testsuite/libffi.call/cls_align_longdouble_split.c
87 testsuite/libffi.call/cls_dbls_struct.c
88 testsuite/libffi.call/cls_double_va.c
89 testsuite/libffi.call/cls_longdouble.c
90 testsuite/libffi.call/cls_longdouble_va.c
91 testsuite/libffi.call/cls_pointer.c
92 testsuite/libffi.call/cls_pointer_stack.c
93 testsuite/libffi.call/err_bad_abi.c
94 testsuite/libffi.call/err_bad_typedef.c
95 testsuite/libffi.call/stret_large2.c
96 testsuite/libffi.call/stret_large.c
97 testsuite/libffi.call/stret_medium2.c
98 testsuite/libffi.call/stret_medium.c: New tests from Apple.
100 2009-06-05 Andrew Haley <aph@redhat.com>
102 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
105 2009-06-04 Andrew Haley <aph@redhat.com>
107 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
110 2008-02-26 Anthony Green <green@redhat.com>
111 Thomas Heller <theller@ctypes.org>
113 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
116 2008-02-03 Timothy Wall <twall@users.sf.net>
118 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
119 offset based on code pointer, not data pointer.
121 2008-01-31 Timothy Wall <twall@users.sf.net>
123 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
125 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
127 * src/x86/win32.S: Add assembly for stdcall closure.
128 * src/x86/ffi.c: Initialize stdcall closure trampoline.
130 2009-06-04 Andrew Haley <aph@redhat.com>
132 * include/ffi.h.in: Change void (*)() to void (*)(void).
133 * src/x86/ffi.c: Likewise.
135 2009-06-04 Andrew Haley <aph@redhat.com>
137 * src/powerpc/ppc_closure.S: Insert licence header.
138 * src/powerpc/linux64_closure.S: Likewise.
139 * src/m68k/sysv.S: Likewise.
141 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
142 * src/powerpc/ffi.c: Likewise.
143 * src/powerpc/ffi_darwin.c: Likewise.
144 * src/m32r/ffi.c: Likewise.
145 * src/sh64/ffi.c: Likewise.
146 * src/x86/ffi64.c: Likewise.
147 * src/alpha/ffi.c: Likewise.
148 * src/alpha/osf.S: Likewise.
149 * src/frv/ffi.c: Likewise.
150 * src/s390/ffi.c: Likewise.
151 * src/pa/ffi.c: Likewise.
152 * src/pa/hpux32.S: Likewise.
153 * src/ia64/unix.S: Likewise.
154 * src/ia64/ffi.c: Likewise.
155 * src/sparc/ffi.c: Likewise.
156 * src/mips/ffi.c: Likewise.
157 * src/sh/ffi.c: Likewise.
159 2008-02-15 David Daney <ddaney@avtrex.com>
161 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
162 Define (conditionally), and use it to include cachectl.h.
163 (ffi_prep_closure_loc): Fix cache flushing.
164 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
166 2009-06-04 Andrew Haley <aph@redhat.com>
172 src/powerpc/ffitarget.h,
174 src/sh64/ffitarget.h,
185 src/alpha/ffitarget.h,
188 src/m68k/ffitarget.h,
191 src/s390/ffitarget.h,
193 src/cris/ffitarget.h,
198 src/ia64/ffitarget.h,
201 src/ia64/ia64_flags.h,
205 src/sparc/ffitarget.h,
208 src/mips/ffitarget.h,
215 src/sh/sysv.S: Update license text.
217 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
219 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
220 (.eh_frame): Add FDE for it.
222 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
224 * configure.ac: Also check if assembler supports pc-relative
225 relocs on X86_WIN32 targets.
226 * configure: Regenerate.
227 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
228 (_ffi_call_SYSV): Add missing function type symbol .def and
229 add EH markup labels.
230 (_ffi_call_STDCALL): Likewise.
231 (_ffi_closure_SYSV): Likewise.
232 (_ffi_closure_raw_SYSV): Likewise.
233 (.eh_frame): Add hand-crafted EH data.
235 2009-04-09 Jakub Jelinek <jakub@redhat.com>
237 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
238 version 3 of the GNU General Public License and to point readers
239 at the COPYING3 file and the FSF's license web page.
240 * testsuite/libffi.call/call.exp: Likewise.
241 * testsuite/libffi.special/special.exp: Likewise.
243 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
245 * configure: Regenerate.
247 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
250 * configure.ac (HAVE_AS_X86_PCREL): New test.
251 * configure: Regenerate.
252 * fficonfig.h.in: Regenerate.
253 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
254 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
255 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
256 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
257 * src/x86/unix64.S (.Lstore_table): Move to .text section.
258 (.Lload_table): Likewise.
259 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
261 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
263 * configure: Regenerate.
265 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
267 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
268 signed/unsigned int8/16 return values.
269 * src/sparc/v8.S (ffi_call_v8): Likewise.
270 (ffi_closure_v8): Likewise.
272 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
273 Steve Ellcey <sje@cup.hp.com>
275 * configure: Regenerate for new libtool.
276 * Makefile.in: Ditto.
277 * include/Makefile.in: Ditto.
280 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
282 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
283 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
284 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
285 Adjust copyright notice.
286 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
287 register or two register to use for FFI_SYSV structs.
288 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
289 (ffi_closure_helper_SYSV): Fix the return type for
290 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
291 Adjust copyright notice.
293 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
295 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
298 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
300 * configure: Regenerate.
301 * include/Makefile.in: Regenerate.
302 * testsuite/Makefile.in: Regenerate.
304 2008-06-07 Joseph Myers <joseph@codesourcery.com>
306 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
307 powerpc-*-beos*): Remove.
308 * configure: Regenerate.
310 2008-05-09 Julian Brown <julian@codesourcery.com>
312 * Makefile.am (LTLDFLAGS): New.
313 (libffi_la_LDFLAGS): Use above.
314 * Makefile.in: Regenerate.
316 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
319 * aclocal.m4: Regenerate.
320 * configure: Regenerate.
322 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
324 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
325 * src/sh64/sysv.S: Likewise.
327 2008-03-26 Daniel Jacobowitz <dan@debian.org>
329 * src/arm/sysv.S: Fix ARM comment marker.
331 2008-03-26 Jakub Jelinek <jakub@redhat.com>
333 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
334 * src/s390/sysv.S: Likewise.
335 * src/powerpc/ppc_closure.S: Likewise.
336 * src/powerpc/sysv.S: Likewise.
337 * src/x86/unix64.S: Likewise.
338 * src/x86/sysv.S: Likewise.
339 * src/sparc/v8.S: Likewise.
340 * src/sparc/v9.S: Likewise.
341 * src/m68k/sysv.S: Likewise.
342 * src/arm/sysv.S: Likewise.
344 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
346 * aclocal.m4: Regenerate.
347 * configure: Likewise.
348 * Makefile.in: Likewise.
349 * include/Makefile.in: Likewise.
350 * testsuite/Makefile.in: Likewise.
352 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
353 Andreas Tobler <a.tobler@schweiz.org>
355 * configure.ac: Add amd64-*-freebsd* target.
356 * configure: Regenerate.
358 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
361 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
364 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
367 2008-01-24 David Edelsohn <edelsohn@gnu.org>
369 * configure: Regenerate.
371 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
373 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
375 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
378 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
379 signed/unsigned int8/16 for X86_DARWIN.
380 Updated copyright info.
381 Handle one and two byte structs with special cif->flags.
382 * src/x86/ffitarget.h: Add special types for one and two byte structs.
383 Updated copyright info.
384 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
386 Remove code to pop args from the stack after call.
387 Special-case signed/unsigned for int8/16, one and two byte structs.
388 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
389 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
391 Updated copyright info.
393 2007-12-08 David Daney <ddaney@avtrex.com>
395 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
396 SUBU, add with ADDU and use smaller code sequences.
398 2007-12-07 David Daney <ddaney@avtrex.com>
400 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
403 2007-12-06 David Daney <ddaney@avtrex.com>
405 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
407 (ffi_java_raw): New typedef.
408 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
409 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
411 (ffi_java_raw_closure) : Same.
412 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
414 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
416 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
417 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
418 sizeof(ffi_java_raw) for alignment calculations.
419 (ffi_java_ptrarray_to_raw): Same.
420 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
421 if FFI_SIZEOF_JAVA_RAW == 4.
422 (ffi_java_raw_to_rvalue): Same.
423 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
424 (ffi_java_translate_args): Same.
425 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
427 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
429 2007-12-06 David Daney <ddaney@avtrex.com>
431 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
434 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
437 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
438 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
439 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
440 set the NUM_FPR_ARG_REGISTERS according to.
441 Add support for potential soft-float support under hard-float
443 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
444 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
445 to the FFI_LINUX_SOFT_FLOAT ABI.
446 (ffi_prep_cif_machdep): Likewise.
447 (ffi_closure_helper_SYSV): Likewise.
448 * src/powerpc/ppc_closure.S: Make sure not to store float/double
449 on archs where __NO_FPRS__ is true.
450 Add FFI_TYPE_UINT128 support.
451 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
452 Adjust copyright notice.
454 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
456 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
457 * include/ffi_common.h: ... here.
460 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
462 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
463 * src/powerpc/linux64.S: Likewise.
464 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
465 * testsuite/libffi.call/return_ldl.c: New test.
467 2007-09-04 <aph@redhat.com>
469 * src/arm/sysv.S (UNWIND): New.
470 (Whole file): Conditionally compile unwinder directives.
471 * src/arm/sysv.S: Add unwinder directives.
473 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
474 Only treat r0 as a struct address if we're actually returning a
476 Only copy the bytes that are actually within a struct.
477 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
478 is returned in r0, not passed by address.
479 (ffi_call): Allocate a word-sized temporary for the case where
480 a composite is returned in r0.
481 (ffi_prep_incoming_args_SYSV): Align as necessary.
483 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
485 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
486 directly using the sys_cacheflush syscall.
488 2007-07-27 Andrew Haley <aph@redhat.com>
490 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
492 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
494 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
495 * configure.ac: Likewise.
496 * Makefile.in: Regenerate.
497 * include/Makefile.in: Likewise.
498 * testsuite/Makefile.in: Likewise.
499 * configure: Likewise.
501 2007-08-24 David Daney <ddaney@avtrex.com>
503 * testsuite/libffi.call/return_sl.c: New test.
505 2007-08-10 David Daney <ddaney@avtrex.com>
507 * testsuite/libffi.call/cls_multi_ushort.c,
508 testsuite/libffi.call/cls_align_uint16.c,
509 testsuite/libffi.call/nested_struct1.c,
510 testsuite/libffi.call/nested_struct3.c,
511 testsuite/libffi.call/cls_7_1_byte.c,
512 testsuite/libffi.call/nested_struct5.c,
513 testsuite/libffi.call/cls_double.c,
514 testsuite/libffi.call/nested_struct7.c,
515 testsuite/libffi.call/cls_sint.c,
516 testsuite/libffi.call/nested_struct9.c,
517 testsuite/libffi.call/cls_20byte1.c,
518 testsuite/libffi.call/cls_multi_sshortchar.c,
519 testsuite/libffi.call/cls_align_sint64.c,
520 testsuite/libffi.call/cls_3byte2.c,
521 testsuite/libffi.call/cls_multi_schar.c,
522 testsuite/libffi.call/cls_multi_uchar.c,
523 testsuite/libffi.call/cls_19byte.c,
524 testsuite/libffi.call/cls_9byte1.c,
525 testsuite/libffi.call/cls_align_float.c,
526 testsuite/libffi.call/closure_fn1.c,
527 testsuite/libffi.call/problem1.c,
528 testsuite/libffi.call/closure_fn3.c,
529 testsuite/libffi.call/cls_sshort.c,
530 testsuite/libffi.call/closure_fn5.c,
531 testsuite/libffi.call/cls_align_double.c,
532 testsuite/libffi.call/nested_struct.c,
533 testsuite/libffi.call/cls_2byte.c,
534 testsuite/libffi.call/nested_struct10.c,
535 testsuite/libffi.call/cls_4byte.c,
536 testsuite/libffi.call/cls_6byte.c,
537 testsuite/libffi.call/cls_8byte.c,
538 testsuite/libffi.call/cls_multi_sshort.c,
539 testsuite/libffi.call/cls_align_sint16.c,
540 testsuite/libffi.call/cls_align_uint32.c,
541 testsuite/libffi.call/cls_20byte.c,
542 testsuite/libffi.call/cls_float.c,
543 testsuite/libffi.call/nested_struct2.c,
544 testsuite/libffi.call/cls_5_1_byte.c,
545 testsuite/libffi.call/nested_struct4.c,
546 testsuite/libffi.call/cls_24byte.c,
547 testsuite/libffi.call/nested_struct6.c,
548 testsuite/libffi.call/cls_64byte.c,
549 testsuite/libffi.call/nested_struct8.c,
550 testsuite/libffi.call/cls_uint.c,
551 testsuite/libffi.call/cls_multi_ushortchar.c,
552 testsuite/libffi.call/cls_schar.c,
553 testsuite/libffi.call/cls_uchar.c,
554 testsuite/libffi.call/cls_align_uint64.c,
555 testsuite/libffi.call/cls_ulonglong.c,
556 testsuite/libffi.call/cls_align_longdouble.c,
557 testsuite/libffi.call/cls_1_1byte.c,
558 testsuite/libffi.call/cls_12byte.c,
559 testsuite/libffi.call/cls_3_1byte.c,
560 testsuite/libffi.call/cls_3byte1.c,
561 testsuite/libffi.call/cls_4_1byte.c,
562 testsuite/libffi.call/cls_6_1_byte.c,
563 testsuite/libffi.call/cls_16byte.c,
564 testsuite/libffi.call/cls_18byte.c,
565 testsuite/libffi.call/closure_fn0.c,
566 testsuite/libffi.call/cls_9byte2.c,
567 testsuite/libffi.call/closure_fn2.c,
568 testsuite/libffi.call/closure_fn4.c,
569 testsuite/libffi.call/cls_ushort.c,
570 testsuite/libffi.call/closure_fn6.c,
571 testsuite/libffi.call/cls_5byte.c,
572 testsuite/libffi.call/cls_align_pointer.c,
573 testsuite/libffi.call/cls_7byte.c,
574 testsuite/libffi.call/cls_align_sint32.c,
575 testsuite/libffi.special/unwindtest_ffi_call.cc,
576 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
578 2007-08-10 David Daney <ddaney@avtrex.com>
581 * configure.ac: Don't treat mips64 as a special case.
582 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
583 * configure: Regenerate
584 * Makefile.in: Ditto.
585 * fficonfig.h.in: Ditto.
586 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
587 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
588 (FFI_DEFAULT_ABI): Set for n64 case.
589 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
590 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
591 (ffi_closure_N32): New function.
592 (.eh_frame): New section
593 * src/mips/o32.S: Clean up comments.
594 (ffi_closure_O32): Pass ffi_closure parameter in $12.
595 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
596 _MIPS_SIM == _ABIN32 throughout.
597 (FFI_MIPS_STOP_HERE): New, use in place of
599 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
600 to support n32/n64 ABIs.
601 (calc_n32_struct_flags): Rewrite.
602 (calc_n32_return_struct_flags): Remove unused variable. Reverse
603 position of flag bits.
604 (ffi_prep_cif_machdep): Rewrite n32 portion.
605 (ffi_call): Enable for n64. Add special handling for small structure
607 (ffi_prep_closure_loc): Add n32 and n64 support.
608 (ffi_closure_mips_inner_O32): Add cast to silence warning.
609 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
611 2007-08-08 David Daney <ddaney@avtrex.com>
613 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
614 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
616 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
617 * testsuite/libffi.call/cls_sint.c (main): Ditto.
618 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
619 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
620 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
621 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
622 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
623 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
624 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
625 * testsuite/libffi.call/return_fl3.c (main): Ditto.
626 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
627 * testsuite/libffi.call/nested_struct.c (main): Ditto.
628 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
629 * testsuite/libffi.call/return_ll1.c (main): Ditto.
630 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
631 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
632 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
633 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
634 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
635 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
636 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
637 * testsuite/libffi.call/cls_uint.c (main): Ditto.
638 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
639 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
640 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
641 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
642 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
643 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
644 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
645 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
646 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
648 2007-08-07 Andrew Haley <aph@redhat.com>
650 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
653 2007-08-06 Andrew Haley <aph@redhat.com>
656 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
657 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
660 2007-08-02 David Daney <ddaney@avtrex.com>
662 * testsuite/libffi.call/return_ul.c (main): Define return type as
663 ffi_arg. Use proper printf conversion specifier.
665 2007-07-30 Andrew Haley <aph@redhat.com>
668 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
669 signed/unsigned int8/16.
670 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
672 Remove code to pop args from the stack after call.
673 Special-case signed/unsigned int8/16.
674 * testsuite/libffi.call/return_sc.c (main): Revert.
676 2007-07-26 Richard Guenther <rguenther@suse.de>
679 * testsuite/libffi.call/return_sc.c (main): Verify call
680 result as signed char, not ffi_arg.
682 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
684 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
685 * configure: Regenerate.
687 2007-07-11 David Daney <ddaney@avtrex.com>
689 * src/mips/ffi.c: Don't include sys/cachectl.h.
690 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
693 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
695 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
696 from (ffi_prep_closure): ... this.
697 (FFI_INIT_TRAMPOLINE): Adjust.
699 2005-12-31 Phil Blundell <pb@reciva.com>
701 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
702 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
703 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
704 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
705 (FFI_CLOSURES): Enable closure support.
707 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
709 * testsuite/libffi.call/cls_multi_ushort.c,
710 testsuite/libffi.call/cls_align_uint16.c,
711 testsuite/libffi.call/nested_struct1.c,
712 testsuite/libffi.call/nested_struct3.c,
713 testsuite/libffi.call/cls_7_1_byte.c,
714 testsuite/libffi.call/cls_double.c,
715 testsuite/libffi.call/nested_struct5.c,
716 testsuite/libffi.call/nested_struct7.c,
717 testsuite/libffi.call/cls_sint.c,
718 testsuite/libffi.call/nested_struct9.c,
719 testsuite/libffi.call/cls_20byte1.c,
720 testsuite/libffi.call/cls_multi_sshortchar.c,
721 testsuite/libffi.call/cls_align_sint64.c,
722 testsuite/libffi.call/cls_3byte2.c,
723 testsuite/libffi.call/cls_multi_schar.c,
724 testsuite/libffi.call/cls_multi_uchar.c,
725 testsuite/libffi.call/cls_19byte.c,
726 testsuite/libffi.call/cls_9byte1.c,
727 testsuite/libffi.call/cls_align_float.c,
728 testsuite/libffi.call/closure_fn1.c,
729 testsuite/libffi.call/problem1.c,
730 testsuite/libffi.call/closure_fn3.c,
731 testsuite/libffi.call/cls_sshort.c,
732 testsuite/libffi.call/closure_fn5.c,
733 testsuite/libffi.call/cls_align_double.c,
734 testsuite/libffi.call/cls_2byte.c,
735 testsuite/libffi.call/nested_struct.c,
736 testsuite/libffi.call/nested_struct10.c,
737 testsuite/libffi.call/cls_4byte.c,
738 testsuite/libffi.call/cls_6byte.c,
739 testsuite/libffi.call/cls_8byte.c,
740 testsuite/libffi.call/cls_multi_sshort.c,
741 testsuite/libffi.call/cls_align_uint32.c,
742 testsuite/libffi.call/cls_align_sint16.c,
743 testsuite/libffi.call/cls_float.c,
744 testsuite/libffi.call/cls_20byte.c,
745 testsuite/libffi.call/cls_5_1_byte.c,
746 testsuite/libffi.call/nested_struct2.c,
747 testsuite/libffi.call/cls_24byte.c,
748 testsuite/libffi.call/nested_struct4.c,
749 testsuite/libffi.call/nested_struct6.c,
750 testsuite/libffi.call/cls_64byte.c,
751 testsuite/libffi.call/nested_struct8.c,
752 testsuite/libffi.call/cls_uint.c,
753 testsuite/libffi.call/cls_multi_ushortchar.c,
754 testsuite/libffi.call/cls_schar.c,
755 testsuite/libffi.call/cls_uchar.c,
756 testsuite/libffi.call/cls_align_uint64.c,
757 testsuite/libffi.call/cls_ulonglong.c,
758 testsuite/libffi.call/cls_align_longdouble.c,
759 testsuite/libffi.call/cls_1_1byte.c,
760 testsuite/libffi.call/cls_12byte.c,
761 testsuite/libffi.call/cls_3_1byte.c,
762 testsuite/libffi.call/cls_3byte1.c,
763 testsuite/libffi.call/cls_4_1byte.c,
764 testsuite/libffi.call/cls_6_1_byte.c,
765 testsuite/libffi.call/cls_16byte.c,
766 testsuite/libffi.call/cls_18byte.c,
767 testsuite/libffi.call/closure_fn0.c,
768 testsuite/libffi.call/cls_9byte2.c,
769 testsuite/libffi.call/closure_fn2.c,
770 testsuite/libffi.call/closure_fn4.c,
771 testsuite/libffi.call/cls_ushort.c,
772 testsuite/libffi.call/closure_fn6.c,
773 testsuite/libffi.call/cls_5byte.c,
774 testsuite/libffi.call/cls_align_pointer.c,
775 testsuite/libffi.call/cls_7byte.c,
776 testsuite/libffi.call/cls_align_sint32.c,
777 testsuite/libffi.special/unwindtest_ffi_call.cc,
778 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
780 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
782 * aclocal.m4: Regenerated.
784 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
786 * configure: Regenerate.
788 2007-05-23 Steve Ellcey <sje@cup.hp.com>
790 * Makefile.in: Regenerate.
791 * configure: Regenerate.
792 * aclocal.m4: Regenerate.
793 * include/Makefile.in: Regenerate.
794 * testsuite/Makefile.in: Regenerate.
796 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
798 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
799 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
800 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
801 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
802 (FFI_CLOSURES): Enable closure support.
804 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
806 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
807 * configure: Regenerate.
808 * fficonfig.h.in: Regenerate.
809 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
810 CFI_OFFSET,CFI_DEF_CFA): New macros.
811 (ffi_call_SYSV): Add callframe annotation.
813 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
815 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
816 numerous test suite failures.
817 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
819 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
821 * Makefile.am (EXTRA_DIST): Bring up to date.
822 * Makefile.in: Regenerate.
823 * src/frv/eabi.S: Remove RCS keyword.
825 2007-04-06 Richard Henderson <rth@redhat.com>
827 * configure.ac: Tidy target case.
828 (HAVE_LONG_DOUBLE): Allow the target to override.
829 * configure: Regenerate.
830 * include/ffi.h.in: Don't define ffi_type_foo if
831 LIBFFI_HIDE_BASIC_TYPES is defined.
832 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
834 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
835 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
836 (ffi_type_longdouble): Special case for Alpha. Don't define
837 if long double == double.
839 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
840 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
841 (ffi_call, ffi_closure_osf_inner): Likewise.
842 (ffi_closure_osf_inner): Likewise. Mark hidden.
843 (ffi_call_osf, ffi_closure_osf): Mark hidden.
844 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
845 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
846 (load_table): Handle 128-bit long double.
848 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
850 2007-04-06 Tom Tromey <tromey@redhat.com>
853 * README: Fixed bug in example.
855 2007-04-03 Jakub Jelinek <jakub@redhat.com>
857 * src/closures.c: Include sys/statfs.h.
858 (_GNU_SOURCE): Define on Linux.
859 (FFI_MMAP_EXEC_SELINUX): Define.
860 (selinux_enabled): New variable.
861 (selinux_enabled_check): New function.
862 (is_selinux_enabled): Define.
865 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
867 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
868 Use 'volatile float sum' to create sum of floats to avoid false
869 negative due to excess precision on ix86 targets.
872 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
874 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
876 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
878 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
880 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
881 (ffi_prep_closure_loc): New.
882 (ffi_prep_raw_closure_loc): New.
883 (ffi_prep_java_raw_closure_loc): New.
884 * src/closures.c: New file.
885 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
886 Replace sflags with exec_offset.
887 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
888 sub_segment_exec_offset): New macros.
889 (get_segment_flags, set_segment_flags, check_segment_merge): New
891 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
892 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
893 destroy_mspace): Use new macros.
894 (sys_alloc): Silence warning.
895 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
896 * Makefile.in: Rebuilt.
897 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
898 terms of ffi_prep_closure_loc.
899 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
901 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
903 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
905 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
907 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
908 (ffi_prep_closure): ... this.
909 * src/pa/ffi.c: Likewise.
910 * src/cris/ffi.c: Likewise. Adjust.
911 * src/frv/ffi.c: Likewise.
912 * src/ia64/ffi.c: Likewise.
913 * src/mips/ffi.c: Likewise.
914 * src/powerpc/ffi_darwin.c: Likewise.
915 * src/s390/ffi.c: Likewise.
916 * src/sh/ffi.c: Likewise.
917 * src/sh64/ffi.c: Likewise.
918 * src/sparc/ffi.c: Likewise.
919 * src/x86/ffi64.c: Likewise.
920 * src/x86/ffi.c: Likewise.
921 (FFI_INIT_TRAMPOLINE): Adjust.
922 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
923 (ffi_prep_raw_closure): ... this.
924 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
925 (ffi_prep_closure): ... this.
926 (flush_icache): Adjust.
928 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
930 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
933 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
935 * Makefile.am: Add dummy install-pdf target.
936 * Makefile.in: Regenerate
938 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
940 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
941 ffi_closure_helper_SYSV): Add long double handling.
943 2007-02-02 Jakub Jelinek <jakub@redhat.com>
945 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
946 immediately after bctrl instruction.
948 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
950 * Makefile.am (all-recursive, install-recursive,
951 mostlyclean-recursive, clean-recursive, distclean-recursive,
952 maintainer-clean-recursive): Add missing targets.
953 * Makefile.in: Rebuilt.
955 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
957 * configure.ac: Add TARGET for x86_64-*-darwin*.
958 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
960 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
961 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
962 * configure: Regenerate.
963 * Makefile.in: Regenerate.
964 * include/Makefile.in: Regenerate.
965 * testsuite/Makefile.in: Regenerate.
966 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
969 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
971 * aclocal.m4: Regenerate with aclocal -I .. as written in the
974 2006-10-31 Geoffrey Keating <geoffk@apple.com>
976 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
977 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
979 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
980 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
982 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
983 Sandro Tolaini <tolaini@libero.it>
985 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
987 * configure: Regenerated.
988 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
989 (EXTRA_DIST): Add src/x86/darwin.S.
990 * Makefile.in: Regenerated.
991 * include/Makefile.in: Regenerated.
992 * testsuite/Makefile.in: Regenerated.
994 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
995 X86_WIN32, and additionally align stack to 16 bytes.
996 * src/x86/darwin.S: New, based on sysv.S.
997 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
999 2006-09-12 David Daney <ddaney@avtrex.com>
1002 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1003 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1004 * aclocal.m4: Regenerated for automake 1.9.6.
1005 * Makefile.in: Regenerated.
1006 * include/Makefile.in: Regenerated.
1007 * testsuite/Makefile.in: Regenerated.
1009 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1011 * include/ffi_common.h (struct): Revert accidental commit.
1013 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1015 * include/ffi_common.h: Remove lint directives.
1016 * include/ffi.h.in: Likewise.
1018 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1020 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1021 for 32-bit architectures.
1022 * testsuite/libffi.call/return_ul.c: New test case.
1024 2006-07-19 David Daney <ddaney@avtrex.com>
1026 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1027 xfail remains for mips64.
1029 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1031 * Makefile.am: Add install-html target. Add install-html to .PHONY
1032 * Makefile.in: Regenerate.
1033 * aclocal.m4: Regenerate.
1034 * include/Makefile.in: Regenerate.
1035 * testsuite/Makefile.in: Regenerate.
1037 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1039 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1042 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1044 * README: Remove notice about 'Crazy Comments'.
1045 * src/debug.c: Remove lint directives. Cleanup white spaces.
1046 * src/java_raw_api.c: Likewise.
1047 * src/prep_cif.c: Likewise.
1048 * src/raw_api.c: Likewise.
1049 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1051 * src/arm/ffi.c: Remove lint directives.
1052 * src/m32r/ffi.c: Likewise.
1053 * src/pa/ffi.c: Likewise.
1054 * src/powerpc/ffi.c: Likewise.
1055 * src/powerpc/ffi_darwin.c: Likewise.
1056 * src/sh/ffi.c: Likewise.
1057 * src/sh64/ffi.c: Likewise.
1058 * src/x86/ffi.c: Likewise.
1059 * testsuite/libffi.call/float2.c: Likewise.
1060 * testsuite/libffi.call/promotion.c: Likewise.
1061 * testsuite/libffi.call/struct1.c: Likewise.
1063 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1065 * src/pa/hpux32.S: Correct unwind offset calculation for
1067 * src/pa/linux.S: Likewise.
1069 2006-04-12 James E Wilson <wilson@specifix.com>
1072 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1073 (hfa_type_load): Call stf_spill.
1074 (hfa_type_store): Call ldf_fill.
1075 (ffi_call): Adjust calls to above routines. Add local temps for
1078 2006-04-10 Matthias Klose <doko@debian.org>
1080 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1081 directory names containing underscores.
1083 2006-04-07 James E Wilson <wilson@specifix.com>
1085 * testsuite/libffi.call/float4.c: New testcase.
1087 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1088 Andreas Tobler <a.tobler@schweiz.ch>
1090 * Makefile.am: Add PA_HPUX port.
1091 * Makefile.in: Regenerate.
1092 * include/Makefile.in: Likewise.
1093 * testsuite/Makefile.in: Likewise.
1094 * configure.ac: Add PA_HPUX rules.
1095 * configure: Regenerate.
1096 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1097 Add PA_HPUX and PA64_HPUX.
1098 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1099 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1100 (FFI_TYPE_SMALL_STRUCT2): Define.
1101 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1102 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1103 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1104 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1105 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1106 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1107 * src/pa/ffi.c (ROUND_DOWN): Delete.
1108 (fldw, fstw, fldd, fstd): Use '__asm__'.
1109 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1110 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1111 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1112 Simplify incrementing of stack slot variable. Change type of local
1113 'n' to unsigned int.
1114 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1116 (ffi_prep_cif_machdep): Likewise.
1117 (ffi_call): Likewise.
1118 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1119 return type to ffi_status. Simplify incrementing of stack slot
1120 variable. Only copy floating point argument registers when PA_LINUX
1121 is true. Reformat debug statement.
1122 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1123 FFI_TYPE_SMALL_STRUCT8.
1124 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1126 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1127 Add nops to cache flush. Add trampoline for PA_HPUX.
1128 * src/pa/hpux32.S: New file.
1129 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1130 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1131 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1132 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1133 argument type checks so that common argument types appear first.
1134 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1135 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1137 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1139 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1140 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1141 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1142 FFI_TYPE_LONGDOUBLE.
1143 (ffi_prep_args64): Assert using IBM extended double.
1144 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1145 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1146 (ffi_call): Handle FFI_LINUX.
1147 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1148 gpr3 return pointer as for struct return. Handle FFI_LINUX
1149 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1151 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1152 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1153 Don't use r6 as pointer to results, instead use sp offset. Don't
1154 make a special call to load lr with case table address, instead
1155 use offset from previous call.
1156 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1157 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1160 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1162 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1163 passed with FP registers correctly.
1164 (ffi_closure_helper_SYSV): Likewise.
1165 * src/sh64/sysv.S: Likewise.
1167 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1169 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1170 args and userdata unused.
1171 (closure_test_fn1): Mark cif and userdata unused.
1172 (main): Remove unused res.
1174 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1176 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1177 -O2, -O3, -Os and the warning flags -W -Wall.
1178 * testsuite/libffi.special/special.exp: Likewise.
1179 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1180 unused parameter unused for gcc or else do nothing.
1181 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1182 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1183 and userdata unused.
1184 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1185 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1186 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1187 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1188 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1189 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1190 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1191 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1192 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1193 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1194 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1195 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1196 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1197 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1198 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1199 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1200 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1201 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1202 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1203 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1204 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1205 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1206 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1207 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1209 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1211 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1213 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1214 void* to avoid compiler warning.
1216 (cls_struct_align_gn): Mark cif and userdata unused.
1217 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1219 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1221 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1223 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1225 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1227 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1228 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1229 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1231 (main): Cast res_call to silence gcc.
1232 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1234 (main): Cast res_call to silence gcc.
1235 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1237 (main): Cast res_call to silence gcc.
1238 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1240 (main): Cast res_call to silence gcc.
1241 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1243 (main): Cast res_call to silence gcc.
1244 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1246 (main): Cast res_call to silence gcc.
1247 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1249 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1250 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1252 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1253 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1255 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1256 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1258 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1259 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1261 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1262 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1263 and userdata unused.
1264 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1266 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1267 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1268 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1269 Cleanup white spaces.
1270 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1271 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1272 cif and userdata unused.
1273 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1275 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1276 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1277 formatters to silence gcc.
1278 (B_gn): Mark cif and userdata unused.
1279 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1281 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1282 (B_gn): Mark cif and userdata unused.
1283 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1285 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1286 (B_gn): Mark cif and userdata unused.
1287 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1289 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1290 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1291 * testsuite/libffi.call/problem1.c (stub): Likewise.
1292 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1294 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1295 in the last commit for this test case in the test case itself.
1296 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1298 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1299 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1300 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1301 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1302 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1303 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1305 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1307 * src/sh/sysv.S: Fix register numbers in the FDE for
1310 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1312 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1313 declaration to avoid a false negative on ix86. See PR323.
1315 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1317 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1318 and cast integer to void * if needed. Update the pointer to
1319 the FP register saved area correctly.
1321 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1323 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1325 * testsuite/libffi.call/nested_struct4.c: Likewise.
1327 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1329 * testsuite/libffi.call/return_dbl.c: New test case.
1330 * testsuite/libffi.call/return_dbl1.c: Likewise.
1331 * testsuite/libffi.call/return_dbl2.c: Likewise.
1332 * testsuite/libffi.call/return_fl.c: Likewise.
1333 * testsuite/libffi.call/return_fl1.c: Likewise.
1334 * testsuite/libffi.call/return_fl2.c: Likewise.
1335 * testsuite/libffi.call/return_fl3.c: Likewise.
1336 * testsuite/libffi.call/closure_fn6.c: Likewise.
1338 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1340 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1341 here to be used by other test cases too.
1343 * testsuite/libffi.call/nested_struct10.c: New test case.
1344 * testsuite/libffi.call/nested_struct9.c: Likewise.
1345 * testsuite/libffi.call/nested_struct8.c: Likewise.
1346 * testsuite/libffi.call/nested_struct7.c: Likewise.
1347 * testsuite/libffi.call/nested_struct6.c: Likewise.
1348 * testsuite/libffi.call/nested_struct5.c: Likewise.
1349 * testsuite/libffi.call/nested_struct4.c: Likewise.
1351 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1353 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1354 * configure: Rebuilt.
1356 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1358 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1359 instead do the shifting inline.
1360 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1361 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1362 and 8 byte structs, for the remaining struct sizes don't call
1363 __lshrdi3, instead do the shifting inline.
1365 2005-12-07 Thiemo Seufer <ths@networkno.de>
1367 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1368 missing parentheses.
1369 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1370 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1371 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1372 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1373 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1375 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1377 (ffi_prep_closure): Improve trampoline instruction scheduling.
1378 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1380 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1382 * src/powerpc/ffi.c: Formatting.
1383 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1384 (ffi_prep_args64): Likewise.
1386 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1388 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1389 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1391 2005-09-26 Tom Tromey <tromey@redhat.com>
1393 * testsuite/libffi.call/float1.c (value_type): New typedef.
1394 (CANARY): New define.
1395 (main): Check for result buffer overflow.
1396 * src/powerpc/linux64.S: Handle linux64 long double returns.
1397 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1398 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1400 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1403 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1405 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1407 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1409 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1410 (AH_BOTTOM): Add FFI_HIDDEN definition.
1411 * configure: Rebuilt.
1412 * fficonfig.h.in: Rebuilt.
1413 * src/powerpc/ffi.c (hidden): Remove.
1414 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1415 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1416 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1417 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1418 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1419 add FFI_HIDDEN to its prototype.
1420 (ffi_closure_SYSV_inner): New.
1421 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1422 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1424 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1427 * configure: Rebuilt.
1428 * configure.ac: Handle i*86-*-gnu*.
1430 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1432 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1433 DW_CFA_offset_extended_sf rather than
1434 DW_CFA_GNU_negative_offset_extended.
1435 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1437 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1439 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1441 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1442 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1443 partially on register.
1444 (ffi_closure_helper_SYSV): Likewise.
1445 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1447 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1449 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1451 * src/sh64/ffi.c (simple_type): Remove.
1452 (return_type): Handle small structures correctly.
1453 (ffi_prep_args): Likewise.
1454 (ffi_call): Likewise.
1455 (ffi_closure_helper_SYSV): Likewise.
1456 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1457 Emit position independent code if PIC and remove wrong datalabel
1458 prefixes from EH data.
1460 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1462 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1463 * Makefile.in: Regenerate.
1464 * include/Makefile.in: Likewise.
1465 * testsuite/Makefile.in: Likewise.
1466 * configure.ac: Add POWERPC_FREEBSD rules.
1467 * configure: Regenerate.
1468 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1469 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1470 * src/powerpc/ffi.c: Add flags to handle small structure returns
1472 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1474 (ffi_closure_helper_SYSV): Likewise.
1475 * src/powerpc/ppc_closure.S: Add return types for small structures.
1476 * src/powerpc/sysv.S: Add bits to handle small structures for
1479 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1481 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1482 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1483 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1485 2005-07-05 Randolph Chung <tausq@debian.org>
1487 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1488 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1489 structures. Kill compilation warnings.
1490 (ffi_closure_inner_LINUX): Print return values as hex in debug
1491 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1492 Properly handle 5-7 byte structure returns.
1493 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1494 (FFI_TYPE_SMALL_STRUCT2): Remove.
1495 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1496 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1497 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1498 (checksmst1, checksmst2): Remove.
1499 (checksmst3): Optimize handling of 3-byte struct returns.
1500 (checksmst567): Properly handle 5-7 byte struct returns.
1502 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1505 * src/mips/n32.S: Enforce PIC code.
1506 * src/mips/o32.S: Likewise.
1508 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1510 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1511 * configure: Regenerate.
1513 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1515 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1516 to call ffi_closure_helper_SYSV. Append @local instead.
1517 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1519 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1521 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1522 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1523 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1524 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1525 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1527 2005-05-09 Mike Stump <mrs@apple.com>
1529 * configure: Regenerate.
1531 2005-05-08 Richard Henderson <rth@redhat.com>
1534 * src/alpha/osf.S: Update unwind into to match code.
1536 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1537 Richard Henderson <rth@redhat.com>
1539 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1541 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1542 (ffi_prep_closure): Set carry bit if sse-used flag set.
1543 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1544 Only load sse registers if ssecount non-zero.
1545 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1547 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1549 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1550 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1551 * configure: Regenerate.
1553 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1555 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1556 have Tcl8.3-compatible intermediate variable.
1558 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1559 Hans-Peter Nilsson <hp@axis.com>
1561 * Makefile.am: Add CRIS support.
1562 * configure.ac: Likewise.
1563 * Makefile.in, configure, testsuite/Makefile.in,
1564 include/Makefile.in: Regenerate.
1565 * src/cris: New directory.
1566 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1567 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1569 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1570 \r?\n in output tests.
1572 2005-04-12 Mike Stump <mrs@apple.com>
1574 * configure: Regenerate.
1576 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1578 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1580 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1582 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1584 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1585 to 64 bits in ILP32 mode.
1586 Load 64 bits even for short data.
1588 2005-03-23 Mike Stump <mrs@apple.com>
1590 * src/powerpc/darwin.S: Update for -m64 multilib.
1591 * src/powerpc/darwin_closure.S: Likewise.
1593 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1595 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1596 Do not set tool_include_dir.
1597 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1599 * include/Makefile.am: Set gcc_version and toollibffidir.
1600 * include/Makefile.in: Regenerate.
1602 2005-02-22 Andrew Haley <aph@redhat.com>
1604 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1605 odd-numbered register pairs for 64-bit integer types.
1607 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1610 * testsuite/libffi.call/return_ll1.c: New test case.
1612 2005-02-11 Janis Johnson <janis187@us.ibm.com>
1614 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1615 * testsuite/libffi.call/float.c: Ditto.
1616 * testsuite/libffi.call/float2.c: Ditto.
1617 * testsuite/libffi.call/float3.c: Ditto.
1619 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1621 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1623 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1625 * testsuite/libffi.special/special.exp (cxx_options): Add
1628 2004-12-31 Richard Henderson <rth@redhat.com>
1630 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1631 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1632 offset parameters with a type parameter; deduce size and structure
1633 alignment. Update all users.
1635 2004-12-31 Richard Henderson <rth@redhat.com>
1637 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1638 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
1639 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
1640 into ffi_prep_closure.
1641 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1644 2004-12-27 Richard Henderson <rth@redhat.com>
1646 * src/x86/unix64.S: Fix typo in unwind info.
1648 2004-12-25 Richard Henderson <rth@redhat.com>
1650 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1651 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1652 (merge_classes): Check for it.
1654 (classify_argument): Pass byte_offset by value; perform all updates
1656 (examine_argument): Add classes argument; handle
1657 X86_64_COMPLEX_X87_CLASS.
1658 (ffi_prep_args): Merge into ...
1659 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1660 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1661 (ffi_fill_return_value): Remove.
1662 (ffi_prep_closure): Remove dead assert.
1663 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1664 Rewrite to use struct register_args instead of va_list. Create
1665 flags for handling structure returns.
1666 * src/x86/unix64.S: Remove dead strings.
1667 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1668 stack frame with ffi_call. Handle structure returns properly.
1669 (float2sse, floatfloat2sse, double2sse): Remove.
1670 (sse2float, sse2double, sse2floatfloat): Remove.
1671 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1672 to handle structure returns properly.
1674 2004-12-08 David Edelsohn <edelsohn@gnu.org>
1676 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1678 * Makefile.in: Regenerated.
1680 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
1682 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1683 * configure, aclocal.m4, Makefile.in: Regenerate.
1684 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1686 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1688 * configure: Regenerate for libtool change.
1690 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1692 * configure: Regenerate for libtool reversion.
1694 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1696 * configure: Regenerate for libtool change.
1698 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1700 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1702 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
1704 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1705 of jal. Use an absolute encoding for the frame information.
1707 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1709 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1710 * acinclude.m4: Delete logic for sincludes.
1711 * aclocal.m4, Makefile.in, configure: Regenerate.
1712 * include/Makefile: Likewise.
1713 * testsuite/Makefile: Likewise.
1715 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1717 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1718 on a 8-byte boundary.
1719 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1721 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1723 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1724 long long values. Round stack allocation to a multiple of 8 bytes
1725 for ATPCS compatibility.
1726 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1727 names. Handle returning long long types. Add Thumb and interworking
1728 support. Improve soft-float code.
1730 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1732 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1733 (libffi_exit): New function.
1734 (libffi_init): Build the testglue wrapper if needed.
1736 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1739 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1741 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1743 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1745 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1747 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1748 * testsuite/libffi.call/float3.c: New test case.
1750 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1752 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1753 the function returning a structure pointed with R2.
1754 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1755 the structure return value if T bit set. Emit position
1756 independent code and EH data if PIC.
1758 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1760 * Makefile.am: Add m32r support.
1761 * configure.ac: Likewise.
1762 * Makefile.in: Regenerate.
1763 * confiugre: Regenerate.
1764 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1765 (uint64, sint64, double, longdouble)
1766 * src/m32r: New directory.
1767 * src/m32r/ffi.c: New file.
1768 * src/m32r/sysv.S: Likewise.
1769 * src/m32r/ffitarget.h: Likewise.
1771 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1773 * testsuite/libffi.call/negint.c: New test case.
1775 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1778 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1779 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1780 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1783 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1785 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1786 * testsuite/libffi.call/promotion.c: Likewise.
1787 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1789 * testsuite/libffi.call/return_sc.c: Likewise.
1790 * testsuite/libffi.call/return_uc.c: Likewise.
1792 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1794 * src/powerpc/darwin.S: Fix comments and identation.
1795 * src/powerpc/darwin_closure.S: Likewise.
1797 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1799 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1800 (ffi_prep_args): Handle longdouble arguments.
1801 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1803 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1804 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1806 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1807 * src/types.c: Defined longdouble size and alignment for darwin.
1809 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1811 * src/powerpc/aix.S: Remove whitespaces.
1812 * src/powerpc/aix_closure.S: Likewise.
1813 * src/powerpc/asm.h: Likewise.
1814 * src/powerpc/ffi.c: Likewise.
1815 * src/powerpc/ffitarget.h: Likewise.
1816 * src/powerpc/linux64.S: Likewise.
1817 * src/powerpc/linux64_closure.S: Likewise.
1818 * src/powerpc/ppc_closure.S: Likewise.
1819 * src/powerpc/sysv.S: Likewise.
1821 2004-08-30 Anthony Green <green@redhat.com>
1823 * Makefile.am: Add frv support.
1824 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1825 * configure.ac: Read configure.host.
1826 * configure.in: Read configure.host.
1827 * configure.host: New file. frv-elf needs libgloss.
1828 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
1829 alignment. This is needed to frv and shouldn't harm the others.
1830 * include/ffi_common.h (ALIGN_DOWN): New macro.
1831 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1833 2004-08-24 David Daney <daney@avtrex.com>
1835 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1836 * testsuite/libffi.call/closure_fn1.c: Likewise.
1837 * testsuite/libffi.call/closure_fn2.c Likewise.
1838 * testsuite/libffi.call/closure_fn3.c: Likewise.
1839 * testsuite/libffi.call/closure_fn4.c: Likewise.
1840 * testsuite/libffi.call/closure_fn5.c: Likewise.
1841 * testsuite/libffi.call/cls_18byte.c: Likewise.
1842 * testsuite/libffi.call/cls_19byte.c: Likewise.
1843 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1844 * testsuite/libffi.call/cls_20byte.c: Likewise.
1845 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1846 * testsuite/libffi.call/cls_24byte.c: Likewise.
1847 * testsuite/libffi.call/cls_2byte.c: Likewise.
1848 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1849 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1850 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1851 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1852 * testsuite/libffi.call/cls_4byte.c: Likewise.
1853 * testsuite/libffi.call/cls_64byte.c: Likewise.
1854 * testsuite/libffi.call/cls_6byte.c: Likewise.
1855 * testsuite/libffi.call/cls_7byte.c: Likewise.
1856 * testsuite/libffi.call/cls_8byte.c: Likewise.
1857 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1858 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1859 * testsuite/libffi.call/cls_align_double.c: Likewise.
1860 * testsuite/libffi.call/cls_align_float.c: Likewise.
1861 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1862 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1863 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1864 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1865 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1866 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1867 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1868 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1869 * testsuite/libffi.call/cls_double.c: Likewise.
1870 * testsuite/libffi.call/cls_float.c: Likewise.
1871 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1872 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1873 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1874 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1875 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1876 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1877 * testsuite/libffi.call/cls_schar.c: Likewise.
1878 * testsuite/libffi.call/cls_sint.c: Likewise.
1879 * testsuite/libffi.call/cls_sshort.c: Likewise.
1880 * testsuite/libffi.call/cls_uchar.c: Likewise.
1881 * testsuite/libffi.call/cls_uint.c: Likewise.
1882 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1883 * testsuite/libffi.call/cls_ushort.c: Likewise.
1884 * testsuite/libffi.call/nested_struct.c: Likewise.
1885 * testsuite/libffi.call/nested_struct1.c: Likewise.
1886 * testsuite/libffi.call/nested_struct2.c: Likewise.
1887 * testsuite/libffi.call/nested_struct3.c: Likewise.
1888 * testsuite/libffi.call/problem1.c: Likewise.
1889 * testsuite/libffi.special/unwindtest.cc: Likewise.
1890 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1892 * testsuite/libffi.call/cls_16byte.c: Likewise.
1893 * testsuite/libffi.call/cls_5byte.c: Likewise.
1895 2004-08-23 David Daney <daney@avtrex.com>
1898 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1899 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1900 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1901 parameters and return types.
1902 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1903 (ffi_prep_closure): Ditto.
1904 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1905 alignment calculations.
1906 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1907 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1909 2004-08-14 Casey Marshall <csm@gnu.org>
1911 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1912 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1913 integer (O32 ABI only).
1914 (ffi_prep_closure): new function.
1915 (ffi_closure_mips_inner_O32): new function.
1916 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
1917 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1918 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1919 64 bit integers correctly.
1920 (ffi_closure_O32): new function.
1921 Added DWARF-2 unwind info for both functions.
1923 2004-08-10 Andrew Haley <aph@redhat.com>
1925 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1927 2004-08-01 Robert Millan <robertmh@gnu.org>
1929 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1930 * configure: Regenerate.
1932 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1934 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1935 and mmap() explicitly instead of relying on preset autoconf cache
1937 * aclocal.m4: Regenerate.
1938 * configure: Regenerate.
1940 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1942 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1943 (ffi_check_float_struct): Remove unused prototype.
1945 2004-06-30 Geoffrey Keating <geoffk@apple.com>
1947 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1948 character on Darwin, use '\n\t' instead.
1950 2004-06-26 Matthias Klose <doko@debian.org>
1952 * libtool-version: Fix typo in revision/age.
1954 2004-06-17 Matthias Klose <doko@debian.org>
1956 * libtool-version: New.
1957 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1958 * Makefile.in: Regenerate.
1960 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1962 * Makefile.am: Remove useless multilib rules.
1963 * Makefile.in: Regenerate.
1964 * aclocal.m4: Regenerate with automake 1.8.5.
1965 * configure.ac: Remove useless multilib configury.
1966 * configure: Regenerate.
1968 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1970 * .cvsignore: New file.
1972 2004-06-10 Jakub Jelinek <jakub@redhat.com>
1974 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1976 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1979 2004-06-06 Sean McNeil <sean@mcneil.com>
1981 * configure.ac: Add x86_64-*-freebsd* support.
1982 * configure: Regenerate.
1984 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
1987 * configure.ac: Test for existence of mmap and sys/mman.h before
1988 checking blacklist. Fix suggested by Jim Wilson.
1989 * configure: Regenerate.
1991 2004-04-26 Matt Austern <austern@apple.com>
1993 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1995 * src/powerpc/darwin_closure.S: Likewise.
1997 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1999 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2000 error. Reported by Thomas Heller <theller@python.net>.
2001 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2002 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2004 2004-03-20 Matthias Klose <doko@debian.org>
2006 * src/pa/linux.S: Fix typo.
2008 2004-03-19 Matthias Klose <doko@debian.org>
2010 * Makefile.am: Update.
2011 * Makefile.in: Regenerate.
2012 * src/pa/ffi.h.in: Remove.
2013 * src/pa/ffitarget.h: New file.
2015 2004-02-10 Randolph Chung <tausq@debian.org>
2017 * Makefile.am: Add PA support.
2018 * Makefile.in: Regenerate.
2019 * include/Makefile.in: Regenerate.
2020 * configure.ac: Add PA target.
2021 * configure: Regenerate.
2022 * src/pa/ffi.c: New file.
2023 * src/pa/ffi.h.in: Add PA support.
2024 * src/pa/linux.S: New file.
2025 * prep_cif.c: Add PA support.
2027 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2029 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2031 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2032 with ecif->cif->flags.
2033 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2035 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2036 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2037 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2038 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2040 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2042 * configure.in: Rename file to ...
2043 * configure.ac: ... this.
2044 * fficonfig.h.in: Regenerate.
2045 * Makefile.in: Regenerate.
2046 * include/Makefile.in: Regenerate.
2047 * testsuite/Makefile.in: Regenerate.
2049 2004-03-12 Matt Austern <austern@apple.com>
2051 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2052 changes in EH format resulting from addition of linkonce support.
2053 * src/powerpc/darwin_closure.S: Likewise.
2055 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2056 Paolo Bonzini <bonzini@gnu.org>
2058 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2059 Remove VPATH. Remove rules for object files. Remove multilib support.
2060 (AM_CCASFLAGS): Add.
2061 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2062 (AC_PREREQ): Bump version to 2.59.
2063 (AC_INIT): Fill with version info and bug address.
2064 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2065 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2066 De-precious CC so that the right flags are passed down to multilibs.
2067 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2068 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2069 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2070 * configure: Rebuilt.
2071 * aclocal.m4: Likewise.
2072 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2073 * fficonfig.h.in: Likewise.
2075 2004-03-11 Andreas Schwab <schwab@suse.de>
2077 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2078 arguments from fp registers only for the first 8 parameter slots.
2079 Don't convert a float parameter when passed in memory.
2081 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2083 * configure: Regenerate for config/accross.m4 correction.
2085 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2087 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2088 ecif->cif->bytes to bytes.
2089 (ffi_prep_cif_machdep): Add braces around nested if statement.
2091 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2093 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2095 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2096 (ffi_closure_helper_LINUX64): Fix typo.
2097 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2099 * testsuite/libffi.call/float.c: Likewise.
2100 * testsuite/libffi.call/float2.c: Likewise.
2102 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2104 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2105 long double function return and long double arg handling.
2106 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2107 Use "end_pfr" instead of "nf". Correct long double handling.
2109 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2111 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2112 space for long double return value. Adjust stack frame and offsets.
2113 Load f2 long double return.
2115 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2117 * src/types.c: Use 16 byte long double for POWERPC64.
2119 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2121 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2122 when the structure return address is passed in %o0.
2123 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2124 (ffi_v9_layout_struct): Align the field following a nested structure
2125 on a word boundary. Use memmove instead of memcpy.
2126 (ffi_call): Update call to ffi_V9_return_struct.
2127 (ffi_prep_closure): Define 'ctx' only for V8.
2128 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2129 and ffi_closure_sparc_inner_v9.
2130 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2131 Always skip the structure return address. For structures and long
2132 doubles, copy the argument directly.
2133 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2134 if required. Shift the maximum floating-point slot accordingly. For
2135 big structures, copy the argument directly; otherwise, left-justify the
2136 argument and call ffi_v9_layout_struct to lay out the structure on
2138 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2139 (ffi_closure_v8): Pass the structure return address. Update call to
2140 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2141 Skip the 'unimp' insn when returning long doubles and structures.
2142 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2143 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2144 FFI_TYPE_INT handling. Load structures both in integers and
2145 floating-point registers on return.
2146 * README: Update status of the SPARC port.
2148 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2150 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2152 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2154 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2156 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2157 value as of type ffi_arg, not unsigned int.
2159 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2161 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2164 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2166 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2169 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2171 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2174 2003-12-10 Richard Henderson <rth@redhat.com>
2176 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2177 size_t instead of int.
2179 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2181 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2182 * testsuite/libffi.call/many_win32.c (main): Replace variable
2183 int i with unsigned long ul.
2185 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2186 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2187 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2188 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2189 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2190 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2191 * testsuite/libffi.call/cls_align_float.c: Likewise.
2192 * testsuite/libffi.call/cls_align_double.c: Likewise.
2193 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2194 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2196 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2199 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2200 Align arguments to 32 bits.
2202 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2205 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2206 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2207 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2208 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2209 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2210 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2212 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2214 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2216 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2217 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2219 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2221 * Makefile.in: Rebuilt.
2222 * configure: Likewise.
2223 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2226 2003-11-21 Andreas Jaeger <aj@suse.de>
2227 Andreas Tobler <a.tobler@schweiz.ch>
2229 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2230 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2231 * Makefile.in: Rebuilt.
2232 * aclocal.m4: Likewise.
2233 * configure: Likewise.
2234 * fficonfig.h.in: Likewise.
2235 * testsuite/lib/libffi-dg.exp: Add include dir.
2236 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2237 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2238 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2239 for ffi_closure if available.
2240 * testsuite/libffi.call/closure_fn1.c: Likewise.
2241 * testsuite/libffi.call/closure_fn2.c: Likewise.
2242 * testsuite/libffi.call/closure_fn3.c: Likewise.
2243 * testsuite/libffi.call/closure_fn4.c: Likewise.
2244 * testsuite/libffi.call/closure_fn5.c: Likewise.
2245 * testsuite/libffi.call/cls_12byte.c: Likewise.
2246 * testsuite/libffi.call/cls_16byte.c: Likewise.
2247 * testsuite/libffi.call/cls_18byte.c: Likewise.
2248 * testsuite/libffi.call/cls_19byte.c: Likewise.
2249 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2250 * testsuite/libffi.call/cls_20byte.c: Likewise.
2251 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2252 * testsuite/libffi.call/cls_24byte.c: Likewise.
2253 * testsuite/libffi.call/cls_2byte.c: Likewise.
2254 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2255 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2256 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2257 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2258 * testsuite/libffi.call/cls_4byte.c: Likewise.
2259 * testsuite/libffi.call/cls_5byte.c: Likewise.
2260 * testsuite/libffi.call/cls_64byte.c: Likewise.
2261 * testsuite/libffi.call/cls_6byte.c: Likewise.
2262 * testsuite/libffi.call/cls_7byte.c: Likewise.
2263 * testsuite/libffi.call/cls_8byte.c: Likewise.
2264 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2265 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2266 * testsuite/libffi.call/cls_double.c: Likewise.
2267 * testsuite/libffi.call/cls_float.c: Likewise.
2268 * testsuite/libffi.call/cls_schar.c: Likewise.
2269 * testsuite/libffi.call/cls_sint.c: Likewise.
2270 * testsuite/libffi.call/cls_sshort.c: Likewise.
2271 * testsuite/libffi.call/cls_uchar.c: Likewise.
2272 * testsuite/libffi.call/cls_uint.c: Likewise.
2273 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2274 * testsuite/libffi.call/cls_ushort.c: Likewise.
2275 * testsuite/libffi.call/nested_struct.c: Likewise.
2276 * testsuite/libffi.call/nested_struct1.c: Likewise.
2277 * testsuite/libffi.call/nested_struct2.c: Likewise.
2278 * testsuite/libffi.call/nested_struct3.c: Likewise.
2279 * testsuite/libffi.call/problem1.c: Likewise.
2280 * testsuite/libffi.special/unwindtest.cc: Likewise.
2282 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2284 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2286 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2288 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2289 Add -lgcc_s to additional flags.
2291 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2293 * configure.in, include/Makefile.am: PR libgcj/11147, install
2294 the ffitarget.h header file in a gcc versioned and target
2296 * configure: Regenerated.
2297 * Makefile.in, include/Makefile.in: Likewise.
2298 * testsuite/Makefile.in: Likewise.
2300 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2302 * testsuite/libffi.call/closure_fn0.c: Print result and check
2303 with dg-output to make debugging easier.
2304 * testsuite/libffi.call/closure_fn1.c: Likewise.
2305 * testsuite/libffi.call/closure_fn2.c: Likewise.
2306 * testsuite/libffi.call/closure_fn3.c: Likewise.
2307 * testsuite/libffi.call/closure_fn4.c: Likewise.
2308 * testsuite/libffi.call/closure_fn5.c: Likewise.
2309 * testsuite/libffi.call/cls_12byte.c: Likewise.
2310 * testsuite/libffi.call/cls_16byte.c: Likewise.
2311 * testsuite/libffi.call/cls_18byte.c: Likewise.
2312 * testsuite/libffi.call/cls_19byte.c: Likewise.
2313 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2314 * testsuite/libffi.call/cls_20byte.c: Likewise.
2315 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2316 * testsuite/libffi.call/cls_24byte.c: Likewise.
2317 * testsuite/libffi.call/cls_2byte.c: Likewise.
2318 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2319 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2320 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2321 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2322 * testsuite/libffi.call/cls_4byte.c: Likewise.
2323 * testsuite/libffi.call/cls_5byte.c: Likewise.
2324 * testsuite/libffi.call/cls_64byte.c: Likewise.
2325 * testsuite/libffi.call/cls_6byte.c: Likewise.
2326 * testsuite/libffi.call/cls_7byte.c: Likewise.
2327 * testsuite/libffi.call/cls_8byte.c: Likewise.
2328 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2329 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2330 * testsuite/libffi.call/cls_double.c: Likewise.
2331 * testsuite/libffi.call/cls_float.c: Likewise.
2332 * testsuite/libffi.call/cls_schar.c: Likewise.
2333 * testsuite/libffi.call/cls_sint.c: Likewise.
2334 * testsuite/libffi.call/cls_sshort.c: Likewise.
2335 * testsuite/libffi.call/cls_uchar.c: Likewise.
2336 * testsuite/libffi.call/cls_uint.c: Likewise.
2337 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2338 * testsuite/libffi.call/cls_ushort.c: Likewise.
2339 * testsuite/libffi.call/problem1.c: Likewise.
2341 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2344 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2346 * testsuite/libffi.call/cls_9byte2.c: New test case.
2347 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2348 * testsuite/libffi.call/cls_64byte.c: Likewise.
2349 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2350 * testsuite/libffi.call/cls_19byte.c: Likewise.
2351 * testsuite/libffi.call/cls_18byte.c: Likewise.
2352 * testsuite/libffi.call/closure_fn4.c: Likewise.
2353 * testsuite/libffi.call/closure_fn5.c: Likewise.
2354 * testsuite/libffi.call/cls_schar.c: Likewise.
2355 * testsuite/libffi.call/cls_sint.c: Likewise.
2356 * testsuite/libffi.call/cls_sshort.c: Likewise.
2357 * testsuite/libffi.call/nested_struct2.c: Likewise.
2358 * testsuite/libffi.call/nested_struct3.c: Likewise.
2360 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2362 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2363 * testsuite/libffi.call/cls_uchar.c: Likewise.
2364 * testsuite/libffi.call/cls_uint.c: Likewise.
2365 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2366 * testsuite/libffi.call/cls_ushort.c: Likewise.
2367 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2369 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2371 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2374 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2376 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2377 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2379 2003-10-22 David Daney <ddaney@avtrex.com>
2381 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2382 __attribute__((__mode__(__SI__))) and friends.
2384 2003-10-22 Andreas Schwab <schwab@suse.de>
2386 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2388 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2390 * configure.in: AC_LINK_FILES(ffitarget.h).
2391 * configure: Regenerate.
2392 * Makefile.in: Likewise.
2393 * include/Makefile.in: Likewise.
2394 * testsuite/Makefile.in: Likewise.
2395 * fficonfig.h.in: Likewise.
2397 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2398 Richard Henderson <rth@redhat.com>
2400 Avoid that ffi.h includes fficonfig.h.
2402 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2403 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2404 (TARGET_SRC_MIPS_SGI): Removed.
2405 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2406 (MIPS_SGI): Removed.
2407 (CLEANFILES): Removed.
2408 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2410 * acconfig.h: Removed.
2411 * configure.in: Compute sizeofs only for double and long double.
2412 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2413 into AC_DEFINE instead of using acconfig.h. Create
2414 include/ffitarget.h instead of include/fficonfig.h. Rename
2415 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2416 AC_DEFINE EH_FRAME_FLAGS.
2417 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2418 (hack_DATA): Add ffitarget.h.
2419 * include/ffi.h.in: Remove all system specific definitions.
2420 Declare raw API even if it is not installed, why bother?
2421 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2422 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2423 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2424 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2425 * include/ffi_common.h (bool): Do not define.
2426 (ffi_assert): Accept failed assertion.
2427 (ffi_type_test): Return void and accept file/line.
2428 (FFI_ASSERT): Pass stringized failed assertion.
2429 (FFI_ASSERT_AT): New macro.
2430 (FFI_ASSERT_VALID_TYPE): New macro.
2431 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2432 UINT64, SINT64): Define here with gcc's __attribute__ macro
2434 (FLOAT32, ALIGN): Define here instead of in ffi.h
2435 * include/ffi-mips.h: Removed. Its content moved to
2436 src/mips/ffitarget.h after separating assembly and C sections.
2437 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2438 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2439 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2440 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2441 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2442 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2443 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2444 (ffi_type_test): Rewritten.
2445 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2446 FFI_ASSERT_VALID_TYPE.
2447 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2448 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2449 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2450 src/s390/ffitarget.h, src/sh/ffitarget.h,
2451 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2452 src/x86/ffitarget.h: New files.
2453 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2454 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2455 src/powerpc/aix.S, src/powerpc/darwin.S,
2456 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2457 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2458 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2459 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2460 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2463 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2465 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2466 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2468 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2470 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2471 Used when FFI_DEBUG = 1.
2473 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2475 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2478 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2480 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2481 FFI_MIPS_O32 for O32 ABI.
2483 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2485 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2486 SPARC64. Cleanup whitespaces.
2488 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2490 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2491 strongarm, xscale. Cleanup whitespaces.
2492 * testsuite/libffi.call/closure_fn1.c: Likewise.
2493 * testsuite/libffi.call/closure_fn2.c: Likewise.
2494 * testsuite/libffi.call/closure_fn3.c: Likewise.
2495 * testsuite/libffi.call/cls_12byte.c: Likewise.
2496 * testsuite/libffi.call/cls_16byte.c: Likewise.
2497 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2498 * testsuite/libffi.call/cls_20byte.c: Likewise.
2499 * testsuite/libffi.call/cls_24byte.c: Likewise.
2500 * testsuite/libffi.call/cls_2byte.c: Likewise.
2501 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2502 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2503 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2504 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2505 * testsuite/libffi.call/cls_4byte.c: Likewise.
2506 * testsuite/libffi.call/cls_5byte.c: Likewise.
2507 * testsuite/libffi.call/cls_6byte.c: Likewise.
2508 * testsuite/libffi.call/cls_7byte.c: Likewise.
2509 * testsuite/libffi.call/cls_8byte.c: Likewise.
2510 * testsuite/libffi.call/cls_double.c: Likewise.
2511 * testsuite/libffi.call/cls_float.c: Likewise.
2512 * testsuite/libffi.call/cls_uchar.c: Likewise.
2513 * testsuite/libffi.call/cls_uint.c: Likewise.
2514 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2515 * testsuite/libffi.call/cls_ushort.c: Likewise.
2516 * testsuite/libffi.call/nested_struct.c: Likewise.
2517 * testsuite/libffi.call/nested_struct1.c: Likewise.
2518 * testsuite/libffi.call/problem1.c: Likewise.
2519 * testsuite/libffi.special/unwindtest.cc: Likewise.
2520 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2522 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2524 * src/powerpc/aix.S: Cleanup whitespaces.
2525 * src/powerpc/aix_closure.S: Likewise.
2527 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2529 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2530 * src/powerpc/darwin_closure.S: Likewise.
2531 * src/powerpc/ffi_darwin.c: Likewise.
2533 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2534 David Edelsohn <edelsohn@gnu.org>
2536 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2537 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2539 * src/powerpc/darwin_closure.S: Likewise.
2540 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2541 according to the Darwin/AIX ABI.
2542 (ffi_prep_cif_machdep): Likewise.
2543 (ffi_closure_helper_DARWIN): Likewise.
2544 Remove the outgoing parameter stack logic. Simplify the evaluation
2545 of the different CASE types.
2546 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2547 statement in the trampoline code.
2549 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2551 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2552 for the register size.
2553 (ffi_closure_helper_SYSV): Handle the structure return value
2555 (ffi_closure_helper_SYSV): Return the appropriate type when
2556 the registers are used for the structure return value.
2557 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2558 the 64-bit return value. Update copyright years.
2560 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2562 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2563 srcdir for ffi_mips.h.
2565 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2567 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2569 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2570 placement of float result.
2571 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2572 cast of "resp" for big-endian 64 bit machines.
2574 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2576 * src/types.c (double, longdouble): Merge identical SH and ARM
2577 typedefs, and add POWERPC64.
2578 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2579 struct split over gpr and rest.
2580 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2581 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2583 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2585 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2588 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2590 * configure: Regenerate.
2592 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2594 * Makefile.am: Remove build rules for ffitest.
2595 * Makefile.in: Rebuilt.
2597 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2599 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2600 about implicit declaration of abort().
2602 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2604 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2605 * Makefile.in: Rebuilt.
2606 * configure.in: Add dejagnu test framework.
2607 * configure: Rebuilt.
2609 * testsuite/Makefile.am: New file.
2610 * testsuite/Makefile.in: Built
2611 * testsuite/lib/libffi-dg.exp: New file.
2612 * testsuite/config/default.exp: Likewise.
2613 * testsuite/libffi.call/call.exp: Likewise.
2614 * testsuite/libffi.call/ffitest.h: Likewise.
2615 * testsuite/libffi.call/closure_fn0.c: Likewise.
2616 * testsuite/libffi.call/closure_fn1.c: Likewise.
2617 * testsuite/libffi.call/closure_fn2.c: Likewise.
2618 * testsuite/libffi.call/closure_fn3.c: Likewise.
2619 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2620 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2621 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2622 * testsuite/libffi.call/cls_2byte.c: Likewise.
2623 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2624 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2625 * testsuite/libffi.call/cls_4byte.c: Likewise.
2626 * testsuite/libffi.call/cls_5byte.c: Likewise.
2627 * testsuite/libffi.call/cls_6byte.c: Likewise.
2628 * testsuite/libffi.call/cls_7byte.c: Likewise.
2629 * testsuite/libffi.call/cls_8byte.c: Likewise.
2630 * testsuite/libffi.call/cls_12byte.c: Likewise.
2631 * testsuite/libffi.call/cls_16byte.c: Likewise.
2632 * testsuite/libffi.call/cls_20byte.c: Likewise.
2633 * testsuite/libffi.call/cls_24byte.c: Likewise.
2634 * testsuite/libffi.call/cls_double.c: Likewise.
2635 * testsuite/libffi.call/cls_float.c: Likewise.
2636 * testsuite/libffi.call/cls_uchar.c: Likewise.
2637 * testsuite/libffi.call/cls_uint.c: Likewise.
2638 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2639 * testsuite/libffi.call/cls_ushort.c: Likewise.
2640 * testsuite/libffi.call/float.c: Likewise.
2641 * testsuite/libffi.call/float1.c: Likewise.
2642 * testsuite/libffi.call/float2.c: Likewise.
2643 * testsuite/libffi.call/many.c: Likewise.
2644 * testsuite/libffi.call/many_win32.c: Likewise.
2645 * testsuite/libffi.call/nested_struct.c: Likewise.
2646 * testsuite/libffi.call/nested_struct1.c: Likewise.
2647 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2648 * testsuite/libffi.call/problem1.c: Likewise.
2649 * testsuite/libffi.call/promotion.c: Likewise.
2650 * testsuite/libffi.call/return_ll.c: Likewise.
2651 * testsuite/libffi.call/return_sc.c: Likewise.
2652 * testsuite/libffi.call/return_uc.c: Likewise.
2653 * testsuite/libffi.call/strlen.c: Likewise.
2654 * testsuite/libffi.call/strlen_win32.c: Likewise.
2655 * testsuite/libffi.call/struct1.c: Likewise.
2656 * testsuite/libffi.call/struct2.c: Likewise.
2657 * testsuite/libffi.call/struct3.c: Likewise.
2658 * testsuite/libffi.call/struct4.c: Likewise.
2659 * testsuite/libffi.call/struct5.c: Likewise.
2660 * testsuite/libffi.call/struct6.c: Likewise.
2661 * testsuite/libffi.call/struct7.c: Likewise.
2662 * testsuite/libffi.call/struct8.c: Likewise.
2663 * testsuite/libffi.call/struct9.c: Likewise.
2664 * testsuite/libffi.special/special.exp: New file.
2665 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2666 * testsuite/libffi.special/unwindtest.cc: Likewise.
2669 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2671 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2674 2003-08-02 Alan Modra <amodra@bigpond.net.au>
2676 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2678 (ffi_closure_helper_LINUX64): Likewise.
2679 * src/powerpc/linux64.S: Remove code writing to parm save area.
2680 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2681 address in lr from ffi_closure_helper_LINUX64 call to calculate
2682 table address. Optimize function tail.
2684 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2686 * src/sparc/ffi.c: Handle all floating point registers.
2687 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2689 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2691 * README: Note that libffi is not part of GCC. Update the project
2694 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2696 * src/powerpc/ppc_closure.S: Include ffi.h.
2698 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2700 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2701 Use C style comments.
2703 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2705 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2706 * Makefile.in: Regenerate.
2707 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2708 * configure: Regenerate.
2709 * include/ffi.h.in: Add SHmedia support.
2710 * src/sh64/ffi.c: New file.
2711 * src/sh64/sysv.S: New file.
2713 2003-05-16 Jakub Jelinek <jakub@redhat.com>
2715 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2716 should be read-only.
2717 * configure: Rebuilt.
2718 * fficonfig.h.in: Rebuilt.
2719 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2720 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2721 * src/powerpc/linux64.S: Likewise.
2722 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2723 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2724 if -fpic/-fPIC/-mrelocatable.
2725 * src/powerpc/powerpc_closure.S: Likewise.
2726 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2727 #write in .eh_frame flags.
2728 * src/sparc/v9.S: Likewise.
2729 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2730 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2731 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2733 2003-05-07 Jeff Sturm <jsturm@one-point.com>
2735 Fixes PR bootstrap/10656
2736 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2737 support for .register pseudo-op.
2738 * src/sparc/v8.S: Use it.
2739 * fficonfig.h.in: Rebuilt.
2740 * configure: Rebuilt.
2742 2003-04-18 Jakub Jelinek <jakub@redhat.com>
2744 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2745 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2746 Make it the default on POWERPC64.
2747 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2748 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2749 * configure: Rebuilt.
2750 * src/powerpc/ffi.c (hidden): Define.
2751 (ffi_prep_args_SYSV): Renamed from
2752 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2753 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2754 ASM_NEEDS_REGISTERS64): New.
2755 (ffi_prep_args64): New function.
2756 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2757 (ffi_call): Likewise.
2758 (ffi_prep_closure): Likewise.
2759 (flush_icache): Surround by #ifndef POWERPC64.
2760 (ffi_dblfl): New union type.
2761 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2762 (ffi_closure_helper_LINUX64): New function.
2763 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2765 * src/powerpc/sysv.S: Likewise.
2766 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2767 * src/powerpc/linux64.S: New file.
2768 * src/powerpc/linux64_closure.S: New file.
2769 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2770 src/powerpc/linux64_closure.S.
2771 (TARGET_SRC_POWERPC): Likewise.
2773 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2774 closure_test_fn3): Fix result printing on big-endian 64-bit
2776 (main): Print tst2_arg instead of uninitialized tst2_result.
2778 * src/ffitest.c (main): Hide what closure pointer really points to
2781 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
2783 * configure.in (arm-*-netbsdelf*): Add configuration.
2784 (configure): Regenerated.
2786 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
2788 * include/Makefile.in: Regenerate.
2790 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2792 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2794 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2795 Receive closure pointer through parameter, read args using
2796 __builtin_dwarf_cfa.
2797 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2799 2003-03-12 Andreas Schwab <schwab@suse.de>
2801 * configure.in: Avoid trailing /. in toolexeclibdir.
2802 * configure: Rebuilt.
2804 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2806 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
2808 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2810 * libffi/src/powerpc/darwin_closure.S:
2811 Fix alignement bug, allocate 8 bytes for the result.
2812 * libffi/src/powerpc/aix_closure.S:
2814 * libffi/src/powerpc/ffi_darwin.c:
2815 Update stackframe description for aix/darwin_closure.S.
2817 2003-02-06 Jakub Jelinek <jakub@redhat.com>
2819 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2822 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2823 Andreas Schwab <schwab@suse.de>
2825 * configure.in: Adjust command to source config-ml.in to account
2826 for changes to the libffi_basedir definition.
2827 (libffi_basedir): Remove ${srcdir} from value and include trailing
2830 * configure: Regenerate.
2832 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2834 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2836 2003-01-28 Andrew Haley <aph@redhat.com>
2838 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2839 * src/x86/ffi64.c (ffi_prep_closure): New.
2840 (ffi_closure_UNIX64_inner): New.
2841 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2843 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
2845 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2846 Remove USE_LIBDIR conditional.
2847 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2848 * Makefile.in, configure: Rebuilt.
2850 2003-01027 David Edelsohn <edelsohn@gnu.org>
2852 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2853 * Makefile.in: Regenerate.
2855 2003-01-22 Andrew Haley <aph@redhat.com>
2857 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2860 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2862 * src/powerpc/darwin.S: Add unwind info.
2863 * src/powerpc/darwin_closure.S: Likewise.
2865 2003-01-14 Andrew Haley <aph@redhat.com>
2867 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2868 (ffi_prep_cif_machdep): Likewise.
2869 * src/x86/unix64.S: Add unwind info.
2871 2003-01-14 Andreas Jaeger <aj@suse.de>
2873 * src/ffitest.c (main): Only use ffi_closures if those are
2876 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2878 * libffi/src/ffitest.c
2879 add closure testcases
2881 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2883 * libffi/src/powerpc/ffi.c
2884 fix alignment bug for float (4 byte aligned iso 8 byte)
2886 2003-01-09 Geoffrey Keating <geoffk@apple.com>
2888 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2889 * src/powerpc/darwin.S: Remove RCS version string.
2891 2003-01-03 Jeff Sturm <jsturm@one-point.com>
2893 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2894 * src/ffitest.c (main): Use static storage for closure.
2895 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2896 * src/sparc/v8.S (ffi_closure_v8): New.
2897 * src/sparc/v9.S (ffi_closure_v9): New.
2899 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2901 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2902 enumeration for X86_WIN32.
2903 * src/x86/win32.S: Added ffi_call_STDCALL function
2905 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2906 switch cases for recognising FFI_STDCALL and
2907 calling ffi_call_STDCALL if target is X86_WIN32.
2908 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2909 stdcall versions of the "my_strlen" and "many"
2910 test functions (for X86_WIN32).
2911 Added test cases to test stdcall invocation using
2914 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2916 * src/sh/sysv.S: Add DWARF2 unwind info.
2918 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2920 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2922 2002-11-26 Jim Wilson <wilson@redhat.com>
2924 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2926 2002-11-23 H.J. Lu <hjl@gnu.org>
2928 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2929 Include ../config/accross.m4.
2930 * aclocal.m4; Rebuild.
2931 * configure: Likewise.
2933 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2935 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2937 2002-11-11 DJ Delorie <dj@redhat.com>
2939 * configure.in: Look for common files in the right place.
2941 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2943 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2944 raw data as _Jv_word values, not ffi_raw.
2945 (ffi_java_ptrarray_to_raw): Likewise.
2946 (ffi_java_rvalue_to_raw): New function.
2947 (ffi_java_raw_call): Call it.
2948 (ffi_java_raw_to_rvalue): New function.
2949 (ffi_java_translate_args): Call it.
2950 * src/ffitest.c (closure_test_fn): Interpret return value
2951 as ffi_arg, not int.
2952 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2953 FFI_TYPE_POINTER case.
2954 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2955 values extended to word size.
2957 2002-10-02 Andreas Jaeger <aj@suse.de>
2959 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2961 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
2963 * include/ffi.h.in: Fix i386 win32 compilation.
2965 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2967 * configure.in: Add s390x-*-linux-* target.
2968 * configure: Regenerate.
2969 * include/ffi.h.in: Define S390X for s390x targets.
2970 (FFI_CLOSURES): Define for s390/s390x.
2971 (FFI_TRAMPOLINE_SIZE): Likewise.
2972 (FFI_NATIVE_RAW_API): Likewise.
2973 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2974 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
2975 * src/s390/ffi.c: Major rework of existing code. Add support for
2976 s390x targets. Add closure support.
2977 * src/s390/sysv.S: Likewise.
2979 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
2981 * src/arm/sysv.S: Fix typo.
2983 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
2985 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2986 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2987 (ffi_call_SYSV): Handle soft-float.
2989 2002-09-27 Bo Thorsen <bo@suse.de>
2991 * include/ffi.h.in: Fix multilib x86-64 support.
2993 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2995 * Makefile.am (all-multi): Fix multilib parallel build.
2997 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2999 * configure.in (sh[34]*-*-linux*): Add brackets.
3000 * configure: Regenerate.
3002 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3004 * Makefile.am: Add SH support.
3005 * Makefile.in: Regenerate.
3006 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3007 * configure: Regenerate.
3008 * include/ffi.h.in: Add SH support.
3009 * src/sh/ffi.c: New file.
3010 * src/sh/sysv.S: New file.
3011 * src/types.c: Add SH support.
3013 2002-07-16 Bo Thorsen <bo@suse.de>
3015 * src/x86/ffi64.c: New file that adds x86-64 support.
3016 * src/x86/unix64.S: New file that handles argument setup for
3018 * src/x86/sysv.S: Don't use this on x86-64.
3019 * src/x86/ffi.c: Don't use this on x86-64.
3021 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3023 * src/ffitest.c (struct6): New test that tests a special case in
3025 (struct7): Likewise.
3026 (struct8): Likewise.
3027 (struct9): Likewise.
3028 (closure_test_fn): Silence warning about this when it's not used.
3029 (main): Add the new tests.
3030 (main): Fix a couple of wrong casts and silence some compiler warnings.
3031 * include/ffi.h.in: Add x86-64 ABI definition.
3032 * fficonfig.h.in: Regenerate.
3033 * Makefile.am: Add x86-64 support.
3034 * configure.in: Likewise.
3035 * Makefile.in: Regenerate.
3036 * configure: Likewise.
3038 2002-06-24 Bo Thorsen <bo@suse.de>
3040 * src/types.c: Merge settings for similar architectures.
3041 Add x86-64 sizes and alignments.
3043 2002-06-23 Bo Thorsen <bo@suse.de>
3045 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3046 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3047 * src/mips/ffi.c (ffi_prep_args): Likewise.
3048 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3050 2002-07-18 H.J. Lu (hjl@gnu.org)
3052 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3053 (libffi_la_SOURCES): Support MIPS_LINUX.
3054 (libffi_convenience_la_SOURCES): Likewise.
3055 * Makefile.in: Regenerated.
3057 * configure.in (mips64*-*): Skip.
3058 (mips*-*-linux*): New.
3059 * configure: Regenerated.
3061 * src/mips/ffi.c: Include <sgidefs.h>.
3063 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3065 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3067 2002-05-27 Roger Sayle <roger@eyesopen.com>
3069 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3071 2002-05-27 Bo Thorsen <bo@suse.de>
3073 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3076 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3078 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3079 beginning of function (for older apple cc).
3081 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3083 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3084 script entry, and set LD to it when configuring multilibs.
3085 * configure: Rebuilt.
3087 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3089 * configure.in (sparc64-*-netbsd*): Add target.
3090 (sparc-*-netbsdelf*): Likewise.
3091 * configure: Regenerate.
3093 2002-04-28 David S. Miller <davem@redhat.com>
3095 * configure.in, configure: Fix SPARC test in previous change.
3097 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3099 * Makefile.am: Add Linux for S/390 support.
3100 * Makefile.in: Regenerate.
3101 * configure.in: Add Linux for S/390 support.
3102 * configure: Regenerate.
3103 * include/ffi.h.in: Add Linux for S/390 support.
3104 * src/s390/ffi.c: New file from libffi CVS tree.
3105 * src/s390/sysv.S: New file from libffi CVS tree.
3107 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3109 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3111 * src/sparc/v8.S: Use it.
3112 * src/sparc/v9.S: Likewise.
3113 * fficonfig.h.in: Rebuilt.
3114 * configure: Rebuilt.
3116 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3118 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3120 * src/ia64/unix.S: Add unwind information. Fix comments.
3121 Save sp in a way that's compatible with unwind info.
3122 (ffi_call_unix): Correctly restore sp in all cases.
3123 * src/ia64/ffi.c: Add, fix comments.
3125 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3127 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3129 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3131 * configure.in (alpha*-*-netbsd*): Add target.
3132 * configure: Regenerate.
3134 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3136 * src/sparc/v8.S: Add unwind info.
3137 * src/sparc/v9.S: Likewise.
3139 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3141 * configure.in: Enable i*86-*-netbsdelf*.
3142 * configure: Rebuilt.
3144 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3147 * src/mips/n32.s: Delete
3148 * src/mips/o32.s: Delete
3150 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3152 * configure.in: Enable alpha*-*-freebsd*.
3153 * configure: Rebuilt.
3155 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3157 * Makefile.am: libfficonvenience -> libffi_convenience.
3158 * Makefile.in: Rebuilt.
3160 * Makefile.am: Define ffitest_OBJECTS.
3161 * Makefile.in: Rebuilt.
3163 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3164 David Edelsohn <edelsohn@gnu.org>
3166 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3167 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3168 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3169 * Makefile.in: Regenerate.
3170 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3171 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3172 (flush_icache, flush_range): New functions.
3173 (ffi_closure_helper_DARWIN): New function.
3174 * src/powerpc/aix_closure.S: New file.
3175 * src/powerpc/darwin_closure.S: New file.
3177 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3179 * include/ffi.h.in: Add typedef for ffi_arg.
3180 * src/ffitest.c (main): Declare rint with ffi_arg.
3182 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3184 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3185 number of GPRs for floating-point arguments.
3187 2002-01-31 Anthony Green <green@redhat.com>
3189 * configure: Rebuilt.
3190 * configure.in: Replace CHECK_SIZEOF and endian tests with
3191 cross-compiler friendly macros.
3192 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3195 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3197 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3198 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3200 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3202 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3203 (TARGET_SRC_POWERPC_AIX): New.
3204 (POWERPC_AIX): New stanza.
3205 * Makefile.in: Regenerate.
3206 * configure.in: Add AIX case.
3207 * configure: Regenerate.
3208 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3209 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3210 size. Fix "long double" support.
3211 (ffi_call): Add FFI_AIX case.
3212 * src/powerpc/aix.S: New.
3214 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3216 Implement Darwin PowerPC ABI.
3217 * configure.in: Handle powerpc-*-darwin*.
3218 * Makefile.am: Set source files for POWERPC_DARWIN.
3219 * configure: Rebuilt.
3220 * Makefile.in: Rebuilt.
3221 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3223 * src/powerpc/darwin.S: New file.
3224 * src/powerpc/ffi_darwin.c: New file.
3226 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3228 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3230 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3232 * src/x86/sysv.S: Avoid gas-only .balign directive.
3233 Use C style comments.
3235 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3237 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3238 Fixes PR bootstrap/3563.
3240 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3242 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3244 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3246 * configure.in: Recognize sparc*-sun-* host.
3247 * configure: Regenerate.
3249 2001-06-06 Andrew Haley <aph@redhat.com>
3251 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
3253 2001-06-03 Andrew Haley <aph@redhat.com>
3255 * src/alpha/osf.S: Add unwind info.
3256 * src/powerpc/sysv.S: Add unwind info.
3257 * src/powerpc/ppc_closure.S: Likewise.
3259 2000-05-31 Jeff Sturm <jsturm@one-point.com>
3261 * configure.in: Fix AC_ARG_ENABLE usage.
3262 * configure: Rebuilt.
3264 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3266 * configure.in: Remove warning about beta code.
3267 * configure: Rebuilt.
3269 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3271 * src/ia64/unix.S: Restore stack pointer when returning from
3273 * src/ia64/ffi.c: Fix typo in comment.
3275 2001-04-18 Jim Wilson <wilson@redhat.com>
3277 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3278 to eliminate RAW DV.
3280 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3282 * Makefile.am: Make a libtool convenience library.
3283 * Makefile.in: Rebuilt.
3285 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3287 * configure.in: Use different syntax for subdirectory creation.
3288 * configure: Rebuilt.
3290 2001-03-27 Jon Beniston <jon@beniston.com>
3292 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3293 * configure: Rebuilt.
3294 * Makefile.am: Added X86_WIN32 target support.
3295 * Makefile.in: Rebuilt.
3297 * include/ffi.h.in: Added X86_WIN32 target support.
3299 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3300 * src/types.c: Added X86_WIN32 target support.
3302 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3303 stuff removed and made to work with CygWin's gas.
3305 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3307 * configure.in: Make target subdirectory in build dir.
3308 * Makefile.am: Override suffix based rules to specify correct output
3310 * Makefile.in: Rebuilt.
3311 * configure: Rebuilt.
3313 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3315 * src/powerpc/ppc_closure.S: New file.
3316 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3317 involving long long and register pairs.
3318 (ffi_prep_closure): New function.
3319 (flush_icache): Likewise.
3320 (ffi_closure_helper_SYSV): Likewise.
3321 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3322 (FFI_TRAMPOLINE_SIZE): Likewise.
3323 (FFI_NATIVE_RAW_API): Likewise.
3324 * Makefile.in: Rebuilt.
3325 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3326 (TARGET_SRC_POWERPC): Likewise.
3328 2001-03-19 Tom Tromey <tromey@redhat.com>
3330 * Makefile.in: Rebuilt.
3331 * Makefile.am (ffitest_LDFLAGS): New macro.
3333 2001-03-02 Nick Clifton <nickc@redhat.com>
3335 * include/ffi.h.in: Remove RCS ident string.
3336 * include/ffi_mips.h: Remove RCS ident string.
3337 * src/debug.c: Remove RCS ident string.
3338 * src/ffitest.c: Remove RCS ident string.
3339 * src/prep_cif.c: Remove RCS ident string.
3340 * src/types.c: Remove RCS ident string.
3341 * src/alpha/ffi.c: Remove RCS ident string.
3342 * src/alpha/osf.S: Remove RCS ident string.
3343 * src/arm/ffi.c: Remove RCS ident string.
3344 * src/arm/sysv.S: Remove RCS ident string.
3345 * src/mips/ffi.c: Remove RCS ident string.
3346 * src/mips/n32.S: Remove RCS ident string.
3347 * src/mips/o32.S: Remove RCS ident string.
3348 * src/sparc/ffi.c: Remove RCS ident string.
3349 * src/sparc/v8.S: Remove RCS ident string.
3350 * src/sparc/v9.S: Remove RCS ident string.
3351 * src/x86/ffi.c: Remove RCS ident string.
3352 * src/x86/sysv.S: Remove RCS ident string.
3354 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3356 * include/ffi.h.in: Change sourceware.cygnus.com references to
3359 2000-12-09 Richard Henderson <rth@redhat.com>
3361 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3362 (ffi_closure_osf_inner): Index rather than increment avalue
3363 and arg_types. Give ffi_closure_osf the raw return value type.
3364 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3367 2000-12-07 Richard Henderson <rth@redhat.com>
3369 * src/raw_api.c (ffi_translate_args): Fix typo.
3370 (ffi_prep_closure): Likewise.
3372 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3373 FFI_TRAMPOLINE_SIZE.
3374 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3375 cif->bytes for new ffi_call_osf implementation.
3376 (ffi_prep_args): Absorb into ...
3377 (ffi_call): ... here. Do all stack allocation here and
3378 avoid a callback function.
3379 (ffi_prep_closure, ffi_closure_osf_inner): New.
3380 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3381 (ffi_closure_osf): New.
3383 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3385 * config.guess, config.sub, install-sh: Removed.
3386 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3387 * Makefile.in: Rebuilt.
3389 * acinclude.m4: Include libtool macros from the top level.
3390 * aclocal.m4, configure: Rebuilt.
3392 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3394 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3395 * configure: Rebuilt.
3397 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3399 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3400 memory correctly. Use conditional instructions, not branches where
3403 2000-05-04 Tom Tromey <tromey@cygnus.com>
3405 * configure: Rebuilt.
3406 * configure.in: Match `arm*-*-linux-*'.
3407 From Chris Dornan <cdornan@arm.com>.
3409 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3411 * Makefile.am (SUBDIRS): Define.
3412 (AM_MAKEFLAGS): Likewise.
3413 (Multilib support.): Add section.
3414 * Makefile.in: Rebuilt.
3415 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3416 New variables. Set for gcc using -print-multi-lib. Export them
3418 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3419 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3421 (extra_compiler_flags_value, extra_compiler_flags): Check these
3422 for extra compiler options which need to be passed down in
3425 2000-04-16 Anthony Green <green@redhat.com>
3427 * configure: Rebuilt.
3428 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3430 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3432 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3433 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3434 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3435 Replace all void * sizeofs with sizeof(int).
3436 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3437 different than DOUBLE.
3438 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3439 (ffi_prep_args_v9): New function.
3440 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3441 (ffi_V9_return_struct): New function.
3442 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3443 32bit code (not yet cross-arch calls).
3444 * src/sparc/v8.S: Add struct return delay nop.
3446 * src/sparc/v9.S: New file.
3447 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3448 is used on sparc64 only for structures larger than 32 bytes.
3449 Pass by reference for structures is done for structure arguments
3450 larger than 16 bytes.
3451 * src/ffitest.c (main): Use 64bit rint on sparc64.
3452 Run long long tests on sparc.
3453 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3455 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3457 * configure.in (sparc-*-linux*): New supported target.
3458 (sparc64-*-linux*): Likewise.
3459 * configure: Rebuilt.
3460 * Makefile.am: Add v9.S to SPARC files.
3461 * Makefile.in: Likewise.
3462 (LINK): Surround $(CCLD) into double quotes, so that multilib
3463 compiles work correctly.
3465 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3467 * configure: Rebuilt.
3468 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3469 proposed by Bryce McKinlay.
3471 2000-03-20 Tom Tromey <tromey@cygnus.com>
3473 * Makefile.in: Hand edit for java_raw_api.lo.
3475 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3477 * config.guess, config.sub: Update from the gcc tree.
3478 Fix for PR libgcj/168.
3480 2000-03-03 Tom Tromey <tromey@cygnus.com>
3482 * Makefile.in: Fixed ia64 by hand.
3484 * configure: Rebuilt.
3485 * configure.in (--enable-multilib): New option.
3486 (libffi_basedir): New subst.
3487 (AC_OUTPUT): Added multilib code.
3489 2000-03-02 Tom Tromey <tromey@cygnus.com>
3491 * Makefile.in: Rebuilt.
3492 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3495 2000-02-25 Hans Boehm <boehm@acm.org>
3497 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3499 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3501 (ffi_prep_raw_closure): Use ffi_translate_args, not
3502 ffi_closure_translate.
3503 * src/java_raw_api.c: New file.
3504 * src/ffitest.c (closure_test_fn): New function.
3505 (main): Define `rint' as long long on IA64. Added new test when
3506 FFI_CLOSURES is defined.
3507 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3508 (ffi_abi): Recognize IA64.
3509 (ffi_raw): Added `flt' field.
3510 Added "Java raw API" code.
3511 * configure.in: Recognize ia64.
3512 * Makefile.am (TARGET_SRC_IA64): New macro.
3513 (libffi_la_common_SOURCES): Added java_raw_api.c.
3514 (libffi_la_SOURCES): Define in IA64 case.
3516 2000-01-04 Tom Tromey <tromey@cygnus.com>
3518 * Makefile.in: Rebuilt with newer automake.
3520 1999-12-31 Tom Tromey <tromey@cygnus.com>
3522 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3524 1999-09-01 Tom Tromey <tromey@cygnus.com>
3526 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3528 * fficonfig.h.in: Rebuilt.
3529 * configure: Rebuilt.
3530 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3531 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3532 * acconfig.h: Don't #undef PACKAGE or VERSION.
3534 1999-08-09 Anthony Green <green@cygnus.com>
3536 * include/ffi.h.in: Try to work around messy header problem
3537 with PACKAGE and VERSION.
3539 * configure: Rebuilt.
3540 * configure.in: Change version to 2.00-beta.
3542 * fficonfig.h.in: Rebuilt.
3543 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3545 * src/x86/ffi.c (ffi_raw_call): Rename.
3547 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3549 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3550 (ffi_prep_incoming_args_SYSV): Ditto.
3551 (ffi_prep_closure): Ditto.
3552 (ffi_closure_raw_SYSV): Ditto.
3553 (ffi_prep_raw_closure): More ditto.
3554 (ffi_call_raw): Final ditto.
3556 * include/ffi.h.in: Add definitions for closure and raw API.
3558 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3561 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3563 * src/raw_api.c: New file.
3565 * include/ffi.h.in (ffi_raw): New type.
3566 (UINT_ARG, SINT_ARG): New defines.
3567 (ffi_closure, ffi_raw_closure): New types.
3568 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3570 * configure.in: Add check for endianness and sizeof void*.
3572 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3573 instead of directly.
3575 * configure: Rebuilt.
3577 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3579 * configure.in: Add x86 and powerpc BeOS configurations.
3580 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3582 1999-05-09 Anthony Green <green@cygnus.com>
3584 * configure.in: Add warning about this being beta code.
3585 Remove src/Makefile.am from the picture.
3586 * configure: Rebuilt.
3588 * Makefile.am: Move logic from src/Makefile.am. Add changes
3589 to support libffi as a target library.
3590 * Makefile.in: Rebuilt.
3592 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3593 Upgraded to new autoconf, automake, libtool.
3597 * LICENSE: Update copyright date.
3599 * src/Makefile.am, src/Makefile.in: Removed.
3601 1998-11-29 Anthony Green <green@cygnus.com>
3603 * include/ChangeLog: Removed.
3604 * src/ChangeLog: Removed.
3605 * src/mips/ChangeLog: Removed.
3606 * src/sparc/ChangeLog: Remboved.
3607 * src/x86/ChangeLog: Removed.
3609 * ChangeLog.v1: Created.