1 2009-06-08 Andrew Haley <aph@redhat.com>
3 * README: Import from libffi 3.0.8.
5 2009-06-08 Andrew Haley <aph@redhat.com>
7 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
8 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
9 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
10 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
12 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
13 * testsuite/libffi.call/stret_medium.c: Likewise.
14 * testsuite/libffi.call/stret_large2.c: Likewise.
15 * testsuite/libffi.call/stret_large.c: Likewise.
17 2008-12-26 Timothy Wall <twall@users.sf.net>
19 * testsuite/libffi.call/cls_longdouble.c,
20 testsuite/libffi.call/cls_longdouble_va.c,
21 testsuite/libffi.call/cls_align_longdouble.c,
22 testsuite/libffi.call/cls_align_longdouble_split.c,
23 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
24 failures on x86_64 cygwin/mingw.
26 2008-12-22 Timothy Wall <twall@users.sf.net>
28 * testsuite/libffi.call/closure_fn0.c,
29 testsuite/libffi.call/closure_fn1.c,
30 testsuite/libffi.call/closure_fn2.c,
31 testsuite/libffi.call/closure_fn3.c,
32 testsuite/libffi.call/closure_fn4.c,
33 testsuite/libffi.call/closure_fn5.c,
34 testsuite/libffi.call/closure_fn6.c,
35 testsuite/libffi.call/closure_loc_fn0.c,
36 testsuite/libffi.call/closure_stdcall.c,
37 testsuite/libffi.call/cls_align_pointer.c,
38 testsuite/libffi.call/cls_pointer.c,
39 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
40 pointer to integer (intptr_t).
41 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
43 2008-07-24 Anthony Green <green@redhat.com>
45 * testsuite/libffi.call/cls_dbls_struct.c,
46 testsuite/libffi.call/cls_double_va.c,
47 testsuite/libffi.call/cls_longdouble.c,
48 testsuite/libffi.call/cls_longdouble_va.c,
49 testsuite/libffi.call/cls_pointer.c,
50 testsuite/libffi.call/cls_pointer_stack.c,
51 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
54 2008-03-04 Anthony Green <green@redhat.com>
58 * testsuite/libffi.call/cls_align_longdouble_split2.c
59 testsuite/libffi.call/cls_align_longdouble_split.c
60 testsuite/libffi.call/cls_dbls_struct.c
61 testsuite/libffi.call/cls_double_va.c
62 testsuite/libffi.call/cls_longdouble.c
63 testsuite/libffi.call/cls_longdouble_va.c
64 testsuite/libffi.call/cls_pointer.c
65 testsuite/libffi.call/cls_pointer_stack.c
66 testsuite/libffi.call/err_bad_abi.c
67 testsuite/libffi.call/err_bad_typedef.c
68 testsuite/libffi.call/stret_large2.c
69 testsuite/libffi.call/stret_large.c
70 testsuite/libffi.call/stret_medium2.c
71 testsuite/libffi.call/stret_medium.c: New tests from Apple.
73 2009-06-05 Andrew Haley <aph@redhat.com>
75 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
78 2009-06-04 Andrew Haley <aph@redhat.com>
80 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
83 2008-02-26 Anthony Green <green@redhat.com>
84 Thomas Heller <theller@ctypes.org>
86 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
89 2008-02-03 Timothy Wall <twall@users.sf.net>
91 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
92 offset based on code pointer, not data pointer.
94 2008-01-31 Timothy Wall <twall@users.sf.net>
96 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
98 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
100 * src/x86/win32.S: Add assembly for stdcall closure.
101 * src/x86/ffi.c: Initialize stdcall closure trampoline.
103 2009-06-04 Andrew Haley <aph@redhat.com>
105 * include/ffi.h.in: Change void (*)() to void (*)(void).
106 * src/x86/ffi.c: Likewise.
108 2009-06-04 Andrew Haley <aph@redhat.com>
110 * src/powerpc/ppc_closure.S: Insert licence header.
111 * src/powerpc/linux64_closure.S: Likewise.
112 * src/m68k/sysv.S: Likewise.
114 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
115 * src/powerpc/ffi.c: Likewise.
116 * src/powerpc/ffi_darwin.c: Likewise.
117 * src/m32r/ffi.c: Likewise.
118 * src/sh64/ffi.c: Likewise.
119 * src/x86/ffi64.c: Likewise.
120 * src/alpha/ffi.c: Likewise.
121 * src/alpha/osf.S: Likewise.
122 * src/frv/ffi.c: Likewise.
123 * src/s390/ffi.c: Likewise.
124 * src/pa/ffi.c: Likewise.
125 * src/pa/hpux32.S: Likewise.
126 * src/ia64/unix.S: Likewise.
127 * src/ia64/ffi.c: Likewise.
128 * src/sparc/ffi.c: Likewise.
129 * src/mips/ffi.c: Likewise.
130 * src/sh/ffi.c: Likewise.
132 2008-02-15 David Daney <ddaney@avtrex.com>
134 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
135 Define (conditionally), and use it to include cachectl.h.
136 (ffi_prep_closure_loc): Fix cache flushing.
137 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
139 2009-06-04 Andrew Haley <aph@redhat.com>
145 src/powerpc/ffitarget.h,
147 src/sh64/ffitarget.h,
158 src/alpha/ffitarget.h,
161 src/m68k/ffitarget.h,
164 src/s390/ffitarget.h,
166 src/cris/ffitarget.h,
171 src/ia64/ffitarget.h,
174 src/ia64/ia64_flags.h,
178 src/sparc/ffitarget.h,
181 src/mips/ffitarget.h,
188 src/sh/sysv.S: Update license text.
190 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
192 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
193 (.eh_frame): Add FDE for it.
195 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
197 * configure.ac: Also check if assembler supports pc-relative
198 relocs on X86_WIN32 targets.
199 * configure: Regenerate.
200 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
201 (_ffi_call_SYSV): Add missing function type symbol .def and
202 add EH markup labels.
203 (_ffi_call_STDCALL): Likewise.
204 (_ffi_closure_SYSV): Likewise.
205 (_ffi_closure_raw_SYSV): Likewise.
206 (.eh_frame): Add hand-crafted EH data.
208 2009-04-09 Jakub Jelinek <jakub@redhat.com>
210 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
211 version 3 of the GNU General Public License and to point readers
212 at the COPYING3 file and the FSF's license web page.
213 * testsuite/libffi.call/call.exp: Likewise.
214 * testsuite/libffi.special/special.exp: Likewise.
216 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
218 * configure: Regenerate.
220 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
223 * configure.ac (HAVE_AS_X86_PCREL): New test.
224 * configure: Regenerate.
225 * fficonfig.h.in: Regenerate.
226 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
227 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
228 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
229 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
230 * src/x86/unix64.S (.Lstore_table): Move to .text section.
231 (.Lload_table): Likewise.
232 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
234 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
236 * configure: Regenerate.
238 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
240 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
241 signed/unsigned int8/16 return values.
242 * src/sparc/v8.S (ffi_call_v8): Likewise.
243 (ffi_closure_v8): Likewise.
245 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
246 Steve Ellcey <sje@cup.hp.com>
248 * configure: Regenerate for new libtool.
249 * Makefile.in: Ditto.
250 * include/Makefile.in: Ditto.
253 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
255 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
256 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
257 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
258 Adjust copyright notice.
259 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
260 register or two register to use for FFI_SYSV structs.
261 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
262 (ffi_closure_helper_SYSV): Fix the return type for
263 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
264 Adjust copyright notice.
266 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
268 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
271 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
273 * configure: Regenerate.
274 * include/Makefile.in: Regenerate.
275 * testsuite/Makefile.in: Regenerate.
277 2008-06-07 Joseph Myers <joseph@codesourcery.com>
279 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
280 powerpc-*-beos*): Remove.
281 * configure: Regenerate.
283 2008-05-09 Julian Brown <julian@codesourcery.com>
285 * Makefile.am (LTLDFLAGS): New.
286 (libffi_la_LDFLAGS): Use above.
287 * Makefile.in: Regenerate.
289 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
292 * aclocal.m4: Regenerate.
293 * configure: Regenerate.
295 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
297 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
298 * src/sh64/sysv.S: Likewise.
300 2008-03-26 Daniel Jacobowitz <dan@debian.org>
302 * src/arm/sysv.S: Fix ARM comment marker.
304 2008-03-26 Jakub Jelinek <jakub@redhat.com>
306 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
307 * src/s390/sysv.S: Likewise.
308 * src/powerpc/ppc_closure.S: Likewise.
309 * src/powerpc/sysv.S: Likewise.
310 * src/x86/unix64.S: Likewise.
311 * src/x86/sysv.S: Likewise.
312 * src/sparc/v8.S: Likewise.
313 * src/sparc/v9.S: Likewise.
314 * src/m68k/sysv.S: Likewise.
315 * src/arm/sysv.S: Likewise.
317 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
319 * aclocal.m4: Regenerate.
320 * configure: Likewise.
321 * Makefile.in: Likewise.
322 * include/Makefile.in: Likewise.
323 * testsuite/Makefile.in: Likewise.
325 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
326 Andreas Tobler <a.tobler@schweiz.org>
328 * configure.ac: Add amd64-*-freebsd* target.
329 * configure: Regenerate.
331 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
334 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
337 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
340 2008-01-24 David Edelsohn <edelsohn@gnu.org>
342 * configure: Regenerate.
344 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
346 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
348 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
351 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
352 signed/unsigned int8/16 for X86_DARWIN.
353 Updated copyright info.
354 Handle one and two byte structs with special cif->flags.
355 * src/x86/ffitarget.h: Add special types for one and two byte structs.
356 Updated copyright info.
357 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
359 Remove code to pop args from the stack after call.
360 Special-case signed/unsigned for int8/16, one and two byte structs.
361 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
362 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
364 Updated copyright info.
366 2007-12-08 David Daney <ddaney@avtrex.com>
368 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
369 SUBU, add with ADDU and use smaller code sequences.
371 2007-12-07 David Daney <ddaney@avtrex.com>
373 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
376 2007-12-06 David Daney <ddaney@avtrex.com>
378 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
380 (ffi_java_raw): New typedef.
381 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
382 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
384 (ffi_java_raw_closure) : Same.
385 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
387 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
389 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
390 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
391 sizeof(ffi_java_raw) for alignment calculations.
392 (ffi_java_ptrarray_to_raw): Same.
393 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
394 if FFI_SIZEOF_JAVA_RAW == 4.
395 (ffi_java_raw_to_rvalue): Same.
396 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
397 (ffi_java_translate_args): Same.
398 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
400 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
402 2007-12-06 David Daney <ddaney@avtrex.com>
404 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
407 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
410 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
411 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
412 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
413 set the NUM_FPR_ARG_REGISTERS according to.
414 Add support for potential soft-float support under hard-float
416 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
417 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
418 to the FFI_LINUX_SOFT_FLOAT ABI.
419 (ffi_prep_cif_machdep): Likewise.
420 (ffi_closure_helper_SYSV): Likewise.
421 * src/powerpc/ppc_closure.S: Make sure not to store float/double
422 on archs where __NO_FPRS__ is true.
423 Add FFI_TYPE_UINT128 support.
424 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
425 Adjust copyright notice.
427 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
429 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
430 * include/ffi_common.h: ... here.
433 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
435 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
436 * src/powerpc/linux64.S: Likewise.
437 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
438 * testsuite/libffi.call/return_ldl.c: New test.
440 2007-09-04 <aph@redhat.com>
442 * src/arm/sysv.S (UNWIND): New.
443 (Whole file): Conditionally compile unwinder directives.
444 * src/arm/sysv.S: Add unwinder directives.
446 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
447 Only treat r0 as a struct address if we're actually returning a
449 Only copy the bytes that are actually within a struct.
450 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
451 is returned in r0, not passed by address.
452 (ffi_call): Allocate a word-sized temporary for the case where
453 a composite is returned in r0.
454 (ffi_prep_incoming_args_SYSV): Align as necessary.
456 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
458 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
459 directly using the sys_cacheflush syscall.
461 2007-07-27 Andrew Haley <aph@redhat.com>
463 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
465 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
467 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
468 * configure.ac: Likewise.
469 * Makefile.in: Regenerate.
470 * include/Makefile.in: Likewise.
471 * testsuite/Makefile.in: Likewise.
472 * configure: Likewise.
474 2007-08-24 David Daney <ddaney@avtrex.com>
476 * testsuite/libffi.call/return_sl.c: New test.
478 2007-08-10 David Daney <ddaney@avtrex.com>
480 * testsuite/libffi.call/cls_multi_ushort.c,
481 testsuite/libffi.call/cls_align_uint16.c,
482 testsuite/libffi.call/nested_struct1.c,
483 testsuite/libffi.call/nested_struct3.c,
484 testsuite/libffi.call/cls_7_1_byte.c,
485 testsuite/libffi.call/nested_struct5.c,
486 testsuite/libffi.call/cls_double.c,
487 testsuite/libffi.call/nested_struct7.c,
488 testsuite/libffi.call/cls_sint.c,
489 testsuite/libffi.call/nested_struct9.c,
490 testsuite/libffi.call/cls_20byte1.c,
491 testsuite/libffi.call/cls_multi_sshortchar.c,
492 testsuite/libffi.call/cls_align_sint64.c,
493 testsuite/libffi.call/cls_3byte2.c,
494 testsuite/libffi.call/cls_multi_schar.c,
495 testsuite/libffi.call/cls_multi_uchar.c,
496 testsuite/libffi.call/cls_19byte.c,
497 testsuite/libffi.call/cls_9byte1.c,
498 testsuite/libffi.call/cls_align_float.c,
499 testsuite/libffi.call/closure_fn1.c,
500 testsuite/libffi.call/problem1.c,
501 testsuite/libffi.call/closure_fn3.c,
502 testsuite/libffi.call/cls_sshort.c,
503 testsuite/libffi.call/closure_fn5.c,
504 testsuite/libffi.call/cls_align_double.c,
505 testsuite/libffi.call/nested_struct.c,
506 testsuite/libffi.call/cls_2byte.c,
507 testsuite/libffi.call/nested_struct10.c,
508 testsuite/libffi.call/cls_4byte.c,
509 testsuite/libffi.call/cls_6byte.c,
510 testsuite/libffi.call/cls_8byte.c,
511 testsuite/libffi.call/cls_multi_sshort.c,
512 testsuite/libffi.call/cls_align_sint16.c,
513 testsuite/libffi.call/cls_align_uint32.c,
514 testsuite/libffi.call/cls_20byte.c,
515 testsuite/libffi.call/cls_float.c,
516 testsuite/libffi.call/nested_struct2.c,
517 testsuite/libffi.call/cls_5_1_byte.c,
518 testsuite/libffi.call/nested_struct4.c,
519 testsuite/libffi.call/cls_24byte.c,
520 testsuite/libffi.call/nested_struct6.c,
521 testsuite/libffi.call/cls_64byte.c,
522 testsuite/libffi.call/nested_struct8.c,
523 testsuite/libffi.call/cls_uint.c,
524 testsuite/libffi.call/cls_multi_ushortchar.c,
525 testsuite/libffi.call/cls_schar.c,
526 testsuite/libffi.call/cls_uchar.c,
527 testsuite/libffi.call/cls_align_uint64.c,
528 testsuite/libffi.call/cls_ulonglong.c,
529 testsuite/libffi.call/cls_align_longdouble.c,
530 testsuite/libffi.call/cls_1_1byte.c,
531 testsuite/libffi.call/cls_12byte.c,
532 testsuite/libffi.call/cls_3_1byte.c,
533 testsuite/libffi.call/cls_3byte1.c,
534 testsuite/libffi.call/cls_4_1byte.c,
535 testsuite/libffi.call/cls_6_1_byte.c,
536 testsuite/libffi.call/cls_16byte.c,
537 testsuite/libffi.call/cls_18byte.c,
538 testsuite/libffi.call/closure_fn0.c,
539 testsuite/libffi.call/cls_9byte2.c,
540 testsuite/libffi.call/closure_fn2.c,
541 testsuite/libffi.call/closure_fn4.c,
542 testsuite/libffi.call/cls_ushort.c,
543 testsuite/libffi.call/closure_fn6.c,
544 testsuite/libffi.call/cls_5byte.c,
545 testsuite/libffi.call/cls_align_pointer.c,
546 testsuite/libffi.call/cls_7byte.c,
547 testsuite/libffi.call/cls_align_sint32.c,
548 testsuite/libffi.special/unwindtest_ffi_call.cc,
549 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
551 2007-08-10 David Daney <ddaney@avtrex.com>
554 * configure.ac: Don't treat mips64 as a special case.
555 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
556 * configure: Regenerate
557 * Makefile.in: Ditto.
558 * fficonfig.h.in: Ditto.
559 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
560 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
561 (FFI_DEFAULT_ABI): Set for n64 case.
562 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
563 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
564 (ffi_closure_N32): New function.
565 (.eh_frame): New section
566 * src/mips/o32.S: Clean up comments.
567 (ffi_closure_O32): Pass ffi_closure parameter in $12.
568 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
569 _MIPS_SIM == _ABIN32 throughout.
570 (FFI_MIPS_STOP_HERE): New, use in place of
572 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
573 to support n32/n64 ABIs.
574 (calc_n32_struct_flags): Rewrite.
575 (calc_n32_return_struct_flags): Remove unused variable. Reverse
576 position of flag bits.
577 (ffi_prep_cif_machdep): Rewrite n32 portion.
578 (ffi_call): Enable for n64. Add special handling for small structure
580 (ffi_prep_closure_loc): Add n32 and n64 support.
581 (ffi_closure_mips_inner_O32): Add cast to silence warning.
582 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
584 2007-08-08 David Daney <ddaney@avtrex.com>
586 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
587 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
589 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
590 * testsuite/libffi.call/cls_sint.c (main): Ditto.
591 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
592 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
593 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
594 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
595 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
596 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
597 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
598 * testsuite/libffi.call/return_fl3.c (main): Ditto.
599 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
600 * testsuite/libffi.call/nested_struct.c (main): Ditto.
601 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
602 * testsuite/libffi.call/return_ll1.c (main): Ditto.
603 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
604 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
605 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
606 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
607 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
608 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
609 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
610 * testsuite/libffi.call/cls_uint.c (main): Ditto.
611 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
612 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
613 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
614 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
615 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
616 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
617 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
618 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
619 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
621 2007-08-07 Andrew Haley <aph@redhat.com>
623 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
626 2007-08-06 Andrew Haley <aph@redhat.com>
629 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
630 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
633 2007-08-02 David Daney <ddaney@avtrex.com>
635 * testsuite/libffi.call/return_ul.c (main): Define return type as
636 ffi_arg. Use proper printf conversion specifier.
638 2007-07-30 Andrew Haley <aph@redhat.com>
641 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
642 signed/unsigned int8/16.
643 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
645 Remove code to pop args from the stack after call.
646 Special-case signed/unsigned int8/16.
647 * testsuite/libffi.call/return_sc.c (main): Revert.
649 2007-07-26 Richard Guenther <rguenther@suse.de>
652 * testsuite/libffi.call/return_sc.c (main): Verify call
653 result as signed char, not ffi_arg.
655 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
657 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
658 * configure: Regenerate.
660 2007-07-11 David Daney <ddaney@avtrex.com>
662 * src/mips/ffi.c: Don't include sys/cachectl.h.
663 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
666 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
668 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
669 from (ffi_prep_closure): ... this.
670 (FFI_INIT_TRAMPOLINE): Adjust.
672 2005-12-31 Phil Blundell <pb@reciva.com>
674 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
675 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
676 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
677 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
678 (FFI_CLOSURES): Enable closure support.
680 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
682 * testsuite/libffi.call/cls_multi_ushort.c,
683 testsuite/libffi.call/cls_align_uint16.c,
684 testsuite/libffi.call/nested_struct1.c,
685 testsuite/libffi.call/nested_struct3.c,
686 testsuite/libffi.call/cls_7_1_byte.c,
687 testsuite/libffi.call/cls_double.c,
688 testsuite/libffi.call/nested_struct5.c,
689 testsuite/libffi.call/nested_struct7.c,
690 testsuite/libffi.call/cls_sint.c,
691 testsuite/libffi.call/nested_struct9.c,
692 testsuite/libffi.call/cls_20byte1.c,
693 testsuite/libffi.call/cls_multi_sshortchar.c,
694 testsuite/libffi.call/cls_align_sint64.c,
695 testsuite/libffi.call/cls_3byte2.c,
696 testsuite/libffi.call/cls_multi_schar.c,
697 testsuite/libffi.call/cls_multi_uchar.c,
698 testsuite/libffi.call/cls_19byte.c,
699 testsuite/libffi.call/cls_9byte1.c,
700 testsuite/libffi.call/cls_align_float.c,
701 testsuite/libffi.call/closure_fn1.c,
702 testsuite/libffi.call/problem1.c,
703 testsuite/libffi.call/closure_fn3.c,
704 testsuite/libffi.call/cls_sshort.c,
705 testsuite/libffi.call/closure_fn5.c,
706 testsuite/libffi.call/cls_align_double.c,
707 testsuite/libffi.call/cls_2byte.c,
708 testsuite/libffi.call/nested_struct.c,
709 testsuite/libffi.call/nested_struct10.c,
710 testsuite/libffi.call/cls_4byte.c,
711 testsuite/libffi.call/cls_6byte.c,
712 testsuite/libffi.call/cls_8byte.c,
713 testsuite/libffi.call/cls_multi_sshort.c,
714 testsuite/libffi.call/cls_align_uint32.c,
715 testsuite/libffi.call/cls_align_sint16.c,
716 testsuite/libffi.call/cls_float.c,
717 testsuite/libffi.call/cls_20byte.c,
718 testsuite/libffi.call/cls_5_1_byte.c,
719 testsuite/libffi.call/nested_struct2.c,
720 testsuite/libffi.call/cls_24byte.c,
721 testsuite/libffi.call/nested_struct4.c,
722 testsuite/libffi.call/nested_struct6.c,
723 testsuite/libffi.call/cls_64byte.c,
724 testsuite/libffi.call/nested_struct8.c,
725 testsuite/libffi.call/cls_uint.c,
726 testsuite/libffi.call/cls_multi_ushortchar.c,
727 testsuite/libffi.call/cls_schar.c,
728 testsuite/libffi.call/cls_uchar.c,
729 testsuite/libffi.call/cls_align_uint64.c,
730 testsuite/libffi.call/cls_ulonglong.c,
731 testsuite/libffi.call/cls_align_longdouble.c,
732 testsuite/libffi.call/cls_1_1byte.c,
733 testsuite/libffi.call/cls_12byte.c,
734 testsuite/libffi.call/cls_3_1byte.c,
735 testsuite/libffi.call/cls_3byte1.c,
736 testsuite/libffi.call/cls_4_1byte.c,
737 testsuite/libffi.call/cls_6_1_byte.c,
738 testsuite/libffi.call/cls_16byte.c,
739 testsuite/libffi.call/cls_18byte.c,
740 testsuite/libffi.call/closure_fn0.c,
741 testsuite/libffi.call/cls_9byte2.c,
742 testsuite/libffi.call/closure_fn2.c,
743 testsuite/libffi.call/closure_fn4.c,
744 testsuite/libffi.call/cls_ushort.c,
745 testsuite/libffi.call/closure_fn6.c,
746 testsuite/libffi.call/cls_5byte.c,
747 testsuite/libffi.call/cls_align_pointer.c,
748 testsuite/libffi.call/cls_7byte.c,
749 testsuite/libffi.call/cls_align_sint32.c,
750 testsuite/libffi.special/unwindtest_ffi_call.cc,
751 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
753 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
755 * aclocal.m4: Regenerated.
757 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
759 * configure: Regenerate.
761 2007-05-23 Steve Ellcey <sje@cup.hp.com>
763 * Makefile.in: Regenerate.
764 * configure: Regenerate.
765 * aclocal.m4: Regenerate.
766 * include/Makefile.in: Regenerate.
767 * testsuite/Makefile.in: Regenerate.
769 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
771 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
772 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
773 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
774 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
775 (FFI_CLOSURES): Enable closure support.
777 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
779 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
780 * configure: Regenerate.
781 * fficonfig.h.in: Regenerate.
782 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
783 CFI_OFFSET,CFI_DEF_CFA): New macros.
784 (ffi_call_SYSV): Add callframe annotation.
786 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
788 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
789 numerous test suite failures.
790 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
792 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
794 * Makefile.am (EXTRA_DIST): Bring up to date.
795 * Makefile.in: Regenerate.
796 * src/frv/eabi.S: Remove RCS keyword.
798 2007-04-06 Richard Henderson <rth@redhat.com>
800 * configure.ac: Tidy target case.
801 (HAVE_LONG_DOUBLE): Allow the target to override.
802 * configure: Regenerate.
803 * include/ffi.h.in: Don't define ffi_type_foo if
804 LIBFFI_HIDE_BASIC_TYPES is defined.
805 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
807 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
808 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
809 (ffi_type_longdouble): Special case for Alpha. Don't define
810 if long double == double.
812 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
813 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
814 (ffi_call, ffi_closure_osf_inner): Likewise.
815 (ffi_closure_osf_inner): Likewise. Mark hidden.
816 (ffi_call_osf, ffi_closure_osf): Mark hidden.
817 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
818 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
819 (load_table): Handle 128-bit long double.
821 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
823 2007-04-06 Tom Tromey <tromey@redhat.com>
826 * README: Fixed bug in example.
828 2007-04-03 Jakub Jelinek <jakub@redhat.com>
830 * src/closures.c: Include sys/statfs.h.
831 (_GNU_SOURCE): Define on Linux.
832 (FFI_MMAP_EXEC_SELINUX): Define.
833 (selinux_enabled): New variable.
834 (selinux_enabled_check): New function.
835 (is_selinux_enabled): Define.
838 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
840 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
841 Use 'volatile float sum' to create sum of floats to avoid false
842 negative due to excess precision on ix86 targets.
845 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
847 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
849 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
851 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
853 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
854 (ffi_prep_closure_loc): New.
855 (ffi_prep_raw_closure_loc): New.
856 (ffi_prep_java_raw_closure_loc): New.
857 * src/closures.c: New file.
858 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
859 Replace sflags with exec_offset.
860 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
861 sub_segment_exec_offset): New macros.
862 (get_segment_flags, set_segment_flags, check_segment_merge): New
864 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
865 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
866 destroy_mspace): Use new macros.
867 (sys_alloc): Silence warning.
868 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
869 * Makefile.in: Rebuilt.
870 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
871 terms of ffi_prep_closure_loc.
872 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
874 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
876 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
878 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
880 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
881 (ffi_prep_closure): ... this.
882 * src/pa/ffi.c: Likewise.
883 * src/cris/ffi.c: Likewise. Adjust.
884 * src/frv/ffi.c: Likewise.
885 * src/ia64/ffi.c: Likewise.
886 * src/mips/ffi.c: Likewise.
887 * src/powerpc/ffi_darwin.c: Likewise.
888 * src/s390/ffi.c: Likewise.
889 * src/sh/ffi.c: Likewise.
890 * src/sh64/ffi.c: Likewise.
891 * src/sparc/ffi.c: Likewise.
892 * src/x86/ffi64.c: Likewise.
893 * src/x86/ffi.c: Likewise.
894 (FFI_INIT_TRAMPOLINE): Adjust.
895 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
896 (ffi_prep_raw_closure): ... this.
897 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
898 (ffi_prep_closure): ... this.
899 (flush_icache): Adjust.
901 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
903 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
906 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
908 * Makefile.am: Add dummy install-pdf target.
909 * Makefile.in: Regenerate
911 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
913 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
914 ffi_closure_helper_SYSV): Add long double handling.
916 2007-02-02 Jakub Jelinek <jakub@redhat.com>
918 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
919 immediately after bctrl instruction.
921 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
923 * Makefile.am (all-recursive, install-recursive,
924 mostlyclean-recursive, clean-recursive, distclean-recursive,
925 maintainer-clean-recursive): Add missing targets.
926 * Makefile.in: Rebuilt.
928 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
930 * configure.ac: Add TARGET for x86_64-*-darwin*.
931 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
933 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
934 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
935 * configure: Regenerate.
936 * Makefile.in: Regenerate.
937 * include/Makefile.in: Regenerate.
938 * testsuite/Makefile.in: Regenerate.
939 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
942 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
944 * aclocal.m4: Regenerate with aclocal -I .. as written in the
947 2006-10-31 Geoffrey Keating <geoffk@apple.com>
949 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
950 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
952 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
953 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
955 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
956 Sandro Tolaini <tolaini@libero.it>
958 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
960 * configure: Regenerated.
961 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
962 (EXTRA_DIST): Add src/x86/darwin.S.
963 * Makefile.in: Regenerated.
964 * include/Makefile.in: Regenerated.
965 * testsuite/Makefile.in: Regenerated.
967 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
968 X86_WIN32, and additionally align stack to 16 bytes.
969 * src/x86/darwin.S: New, based on sysv.S.
970 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
972 2006-09-12 David Daney <ddaney@avtrex.com>
975 * include/Makefile.am: Install both ffi.h and ffitarget.h in
976 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
977 * aclocal.m4: Regenerated for automake 1.9.6.
978 * Makefile.in: Regenerated.
979 * include/Makefile.in: Regenerated.
980 * testsuite/Makefile.in: Regenerated.
982 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
984 * include/ffi_common.h (struct): Revert accidental commit.
986 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
988 * include/ffi_common.h: Remove lint directives.
989 * include/ffi.h.in: Likewise.
991 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
993 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
994 for 32-bit architectures.
995 * testsuite/libffi.call/return_ul.c: New test case.
997 2006-07-19 David Daney <ddaney@avtrex.com>
999 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1000 xfail remains for mips64.
1002 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1004 * Makefile.am: Add install-html target. Add install-html to .PHONY
1005 * Makefile.in: Regenerate.
1006 * aclocal.m4: Regenerate.
1007 * include/Makefile.in: Regenerate.
1008 * testsuite/Makefile.in: Regenerate.
1010 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1012 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1015 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1017 * README: Remove notice about 'Crazy Comments'.
1018 * src/debug.c: Remove lint directives. Cleanup white spaces.
1019 * src/java_raw_api.c: Likewise.
1020 * src/prep_cif.c: Likewise.
1021 * src/raw_api.c: Likewise.
1022 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1024 * src/arm/ffi.c: Remove lint directives.
1025 * src/m32r/ffi.c: Likewise.
1026 * src/pa/ffi.c: Likewise.
1027 * src/powerpc/ffi.c: Likewise.
1028 * src/powerpc/ffi_darwin.c: Likewise.
1029 * src/sh/ffi.c: Likewise.
1030 * src/sh64/ffi.c: Likewise.
1031 * src/x86/ffi.c: Likewise.
1032 * testsuite/libffi.call/float2.c: Likewise.
1033 * testsuite/libffi.call/promotion.c: Likewise.
1034 * testsuite/libffi.call/struct1.c: Likewise.
1036 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1038 * src/pa/hpux32.S: Correct unwind offset calculation for
1040 * src/pa/linux.S: Likewise.
1042 2006-04-12 James E Wilson <wilson@specifix.com>
1045 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1046 (hfa_type_load): Call stf_spill.
1047 (hfa_type_store): Call ldf_fill.
1048 (ffi_call): Adjust calls to above routines. Add local temps for
1051 2006-04-10 Matthias Klose <doko@debian.org>
1053 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1054 directory names containing underscores.
1056 2006-04-07 James E Wilson <wilson@specifix.com>
1058 * testsuite/libffi.call/float4.c: New testcase.
1060 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1061 Andreas Tobler <a.tobler@schweiz.ch>
1063 * Makefile.am: Add PA_HPUX port.
1064 * Makefile.in: Regenerate.
1065 * include/Makefile.in: Likewise.
1066 * testsuite/Makefile.in: Likewise.
1067 * configure.ac: Add PA_HPUX rules.
1068 * configure: Regenerate.
1069 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1070 Add PA_HPUX and PA64_HPUX.
1071 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1072 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1073 (FFI_TYPE_SMALL_STRUCT2): Define.
1074 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1075 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1076 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1077 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1078 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1079 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1080 * src/pa/ffi.c (ROUND_DOWN): Delete.
1081 (fldw, fstw, fldd, fstd): Use '__asm__'.
1082 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1083 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1084 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1085 Simplify incrementing of stack slot variable. Change type of local
1086 'n' to unsigned int.
1087 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1089 (ffi_prep_cif_machdep): Likewise.
1090 (ffi_call): Likewise.
1091 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1092 return type to ffi_status. Simplify incrementing of stack slot
1093 variable. Only copy floating point argument registers when PA_LINUX
1094 is true. Reformat debug statement.
1095 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1096 FFI_TYPE_SMALL_STRUCT8.
1097 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1099 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1100 Add nops to cache flush. Add trampoline for PA_HPUX.
1101 * src/pa/hpux32.S: New file.
1102 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1103 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1104 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1105 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1106 argument type checks so that common argument types appear first.
1107 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1108 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1110 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1112 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1113 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1114 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1115 FFI_TYPE_LONGDOUBLE.
1116 (ffi_prep_args64): Assert using IBM extended double.
1117 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1118 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1119 (ffi_call): Handle FFI_LINUX.
1120 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1121 gpr3 return pointer as for struct return. Handle FFI_LINUX
1122 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1124 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1125 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1126 Don't use r6 as pointer to results, instead use sp offset. Don't
1127 make a special call to load lr with case table address, instead
1128 use offset from previous call.
1129 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1130 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1133 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1135 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1136 passed with FP registers correctly.
1137 (ffi_closure_helper_SYSV): Likewise.
1138 * src/sh64/sysv.S: Likewise.
1140 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1142 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1143 args and userdata unused.
1144 (closure_test_fn1): Mark cif and userdata unused.
1145 (main): Remove unused res.
1147 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1149 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1150 -O2, -O3, -Os and the warning flags -W -Wall.
1151 * testsuite/libffi.special/special.exp: Likewise.
1152 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1153 unused parameter unused for gcc or else do nothing.
1154 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1155 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1156 and userdata unused.
1157 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1158 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1159 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1160 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1161 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1162 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1163 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1164 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1165 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1166 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1167 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1168 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1169 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1170 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1171 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1172 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1173 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1174 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1175 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1176 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1177 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1178 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1179 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1180 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1182 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1184 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1186 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1187 void* to avoid compiler warning.
1189 (cls_struct_align_gn): Mark cif and userdata unused.
1190 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1192 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1194 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1196 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1198 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1200 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1201 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1202 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1204 (main): Cast res_call to silence gcc.
1205 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1207 (main): Cast res_call to silence gcc.
1208 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1210 (main): Cast res_call to silence gcc.
1211 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1213 (main): Cast res_call to silence gcc.
1214 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1216 (main): Cast res_call to silence gcc.
1217 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1219 (main): Cast res_call to silence gcc.
1220 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1222 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1223 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1225 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1226 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1228 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1229 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1231 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1232 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1234 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1235 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1236 and userdata unused.
1237 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1239 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1240 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1241 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1242 Cleanup white spaces.
1243 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1244 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1245 cif and userdata unused.
1246 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1248 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1249 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1250 formatters to silence gcc.
1251 (B_gn): Mark cif and userdata unused.
1252 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1254 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1255 (B_gn): Mark cif and userdata unused.
1256 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1258 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1259 (B_gn): Mark cif and userdata unused.
1260 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1262 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1263 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1264 * testsuite/libffi.call/problem1.c (stub): Likewise.
1265 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1267 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1268 in the last commit for this test case in the test case itself.
1269 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1271 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1272 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1273 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1274 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1275 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1276 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1278 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1280 * src/sh/sysv.S: Fix register numbers in the FDE for
1283 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1285 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1286 declaration to avoid a false negative on ix86. See PR323.
1288 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1290 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1291 and cast integer to void * if needed. Update the pointer to
1292 the FP register saved area correctly.
1294 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1296 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1298 * testsuite/libffi.call/nested_struct4.c: Likewise.
1300 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1302 * testsuite/libffi.call/return_dbl.c: New test case.
1303 * testsuite/libffi.call/return_dbl1.c: Likewise.
1304 * testsuite/libffi.call/return_dbl2.c: Likewise.
1305 * testsuite/libffi.call/return_fl.c: Likewise.
1306 * testsuite/libffi.call/return_fl1.c: Likewise.
1307 * testsuite/libffi.call/return_fl2.c: Likewise.
1308 * testsuite/libffi.call/return_fl3.c: Likewise.
1309 * testsuite/libffi.call/closure_fn6.c: Likewise.
1311 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1313 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1314 here to be used by other test cases too.
1316 * testsuite/libffi.call/nested_struct10.c: New test case.
1317 * testsuite/libffi.call/nested_struct9.c: Likewise.
1318 * testsuite/libffi.call/nested_struct8.c: Likewise.
1319 * testsuite/libffi.call/nested_struct7.c: Likewise.
1320 * testsuite/libffi.call/nested_struct6.c: Likewise.
1321 * testsuite/libffi.call/nested_struct5.c: Likewise.
1322 * testsuite/libffi.call/nested_struct4.c: Likewise.
1324 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1326 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1327 * configure: Rebuilt.
1329 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1331 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1332 instead do the shifting inline.
1333 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1334 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1335 and 8 byte structs, for the remaining struct sizes don't call
1336 __lshrdi3, instead do the shifting inline.
1338 2005-12-07 Thiemo Seufer <ths@networkno.de>
1340 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1341 missing parentheses.
1342 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1343 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1344 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1345 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1346 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1348 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1350 (ffi_prep_closure): Improve trampoline instruction scheduling.
1351 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1353 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1355 * src/powerpc/ffi.c: Formatting.
1356 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1357 (ffi_prep_args64): Likewise.
1359 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1361 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1362 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1364 2005-09-26 Tom Tromey <tromey@redhat.com>
1366 * testsuite/libffi.call/float1.c (value_type): New typedef.
1367 (CANARY): New define.
1368 (main): Check for result buffer overflow.
1369 * src/powerpc/linux64.S: Handle linux64 long double returns.
1370 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1371 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1373 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1376 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1378 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1380 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1382 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1383 (AH_BOTTOM): Add FFI_HIDDEN definition.
1384 * configure: Rebuilt.
1385 * fficonfig.h.in: Rebuilt.
1386 * src/powerpc/ffi.c (hidden): Remove.
1387 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1388 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1389 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1390 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1391 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1392 add FFI_HIDDEN to its prototype.
1393 (ffi_closure_SYSV_inner): New.
1394 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1395 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1397 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1400 * configure: Rebuilt.
1401 * configure.ac: Handle i*86-*-gnu*.
1403 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1405 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1406 DW_CFA_offset_extended_sf rather than
1407 DW_CFA_GNU_negative_offset_extended.
1408 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1410 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1412 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1414 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1415 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1416 partially on register.
1417 (ffi_closure_helper_SYSV): Likewise.
1418 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1420 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1422 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1424 * src/sh64/ffi.c (simple_type): Remove.
1425 (return_type): Handle small structures correctly.
1426 (ffi_prep_args): Likewise.
1427 (ffi_call): Likewise.
1428 (ffi_closure_helper_SYSV): Likewise.
1429 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1430 Emit position independent code if PIC and remove wrong datalabel
1431 prefixes from EH data.
1433 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1435 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1436 * Makefile.in: Regenerate.
1437 * include/Makefile.in: Likewise.
1438 * testsuite/Makefile.in: Likewise.
1439 * configure.ac: Add POWERPC_FREEBSD rules.
1440 * configure: Regenerate.
1441 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1442 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1443 * src/powerpc/ffi.c: Add flags to handle small structure returns
1445 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1447 (ffi_closure_helper_SYSV): Likewise.
1448 * src/powerpc/ppc_closure.S: Add return types for small structures.
1449 * src/powerpc/sysv.S: Add bits to handle small structures for
1452 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1454 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1455 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1456 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1458 2005-07-05 Randolph Chung <tausq@debian.org>
1460 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1461 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1462 structures. Kill compilation warnings.
1463 (ffi_closure_inner_LINUX): Print return values as hex in debug
1464 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1465 Properly handle 5-7 byte structure returns.
1466 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1467 (FFI_TYPE_SMALL_STRUCT2): Remove.
1468 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1469 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1470 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1471 (checksmst1, checksmst2): Remove.
1472 (checksmst3): Optimize handling of 3-byte struct returns.
1473 (checksmst567): Properly handle 5-7 byte struct returns.
1475 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1478 * src/mips/n32.S: Enforce PIC code.
1479 * src/mips/o32.S: Likewise.
1481 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1483 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1484 * configure: Regenerate.
1486 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1488 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1489 to call ffi_closure_helper_SYSV. Append @local instead.
1490 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1492 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1494 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1495 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1496 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1497 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1498 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1500 2005-05-09 Mike Stump <mrs@apple.com>
1502 * configure: Regenerate.
1504 2005-05-08 Richard Henderson <rth@redhat.com>
1507 * src/alpha/osf.S: Update unwind into to match code.
1509 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1510 Richard Henderson <rth@redhat.com>
1512 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1514 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1515 (ffi_prep_closure): Set carry bit if sse-used flag set.
1516 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1517 Only load sse registers if ssecount non-zero.
1518 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1520 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1522 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1523 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1524 * configure: Regenerate.
1526 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1528 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1529 have Tcl8.3-compatible intermediate variable.
1531 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1532 Hans-Peter Nilsson <hp@axis.com>
1534 * Makefile.am: Add CRIS support.
1535 * configure.ac: Likewise.
1536 * Makefile.in, configure, testsuite/Makefile.in,
1537 include/Makefile.in: Regenerate.
1538 * src/cris: New directory.
1539 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1540 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1542 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1543 \r?\n in output tests.
1545 2005-04-12 Mike Stump <mrs@apple.com>
1547 * configure: Regenerate.
1549 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1551 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1553 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1555 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1557 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1558 to 64 bits in ILP32 mode.
1559 Load 64 bits even for short data.
1561 2005-03-23 Mike Stump <mrs@apple.com>
1563 * src/powerpc/darwin.S: Update for -m64 multilib.
1564 * src/powerpc/darwin_closure.S: Likewise.
1566 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1568 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1569 Do not set tool_include_dir.
1570 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1572 * include/Makefile.am: Set gcc_version and toollibffidir.
1573 * include/Makefile.in: Regenerate.
1575 2005-02-22 Andrew Haley <aph@redhat.com>
1577 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1578 odd-numbered register pairs for 64-bit integer types.
1580 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1583 * testsuite/libffi.call/return_ll1.c: New test case.
1585 2005-02-11 Janis Johnson <janis187@us.ibm.com>
1587 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1588 * testsuite/libffi.call/float.c: Ditto.
1589 * testsuite/libffi.call/float2.c: Ditto.
1590 * testsuite/libffi.call/float3.c: Ditto.
1592 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1594 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1596 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1598 * testsuite/libffi.special/special.exp (cxx_options): Add
1601 2004-12-31 Richard Henderson <rth@redhat.com>
1603 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1604 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1605 offset parameters with a type parameter; deduce size and structure
1606 alignment. Update all users.
1608 2004-12-31 Richard Henderson <rth@redhat.com>
1610 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1611 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
1612 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
1613 into ffi_prep_closure.
1614 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1617 2004-12-27 Richard Henderson <rth@redhat.com>
1619 * src/x86/unix64.S: Fix typo in unwind info.
1621 2004-12-25 Richard Henderson <rth@redhat.com>
1623 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1624 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1625 (merge_classes): Check for it.
1627 (classify_argument): Pass byte_offset by value; perform all updates
1629 (examine_argument): Add classes argument; handle
1630 X86_64_COMPLEX_X87_CLASS.
1631 (ffi_prep_args): Merge into ...
1632 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1633 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1634 (ffi_fill_return_value): Remove.
1635 (ffi_prep_closure): Remove dead assert.
1636 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1637 Rewrite to use struct register_args instead of va_list. Create
1638 flags for handling structure returns.
1639 * src/x86/unix64.S: Remove dead strings.
1640 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1641 stack frame with ffi_call. Handle structure returns properly.
1642 (float2sse, floatfloat2sse, double2sse): Remove.
1643 (sse2float, sse2double, sse2floatfloat): Remove.
1644 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1645 to handle structure returns properly.
1647 2004-12-08 David Edelsohn <edelsohn@gnu.org>
1649 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1651 * Makefile.in: Regenerated.
1653 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
1655 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1656 * configure, aclocal.m4, Makefile.in: Regenerate.
1657 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1659 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1661 * configure: Regenerate for libtool change.
1663 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1665 * configure: Regenerate for libtool reversion.
1667 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1669 * configure: Regenerate for libtool change.
1671 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1673 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1675 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
1677 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1678 of jal. Use an absolute encoding for the frame information.
1680 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1682 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1683 * acinclude.m4: Delete logic for sincludes.
1684 * aclocal.m4, Makefile.in, configure: Regenerate.
1685 * include/Makefile: Likewise.
1686 * testsuite/Makefile: Likewise.
1688 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1690 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1691 on a 8-byte boundary.
1692 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1694 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1696 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1697 long long values. Round stack allocation to a multiple of 8 bytes
1698 for ATPCS compatibility.
1699 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1700 names. Handle returning long long types. Add Thumb and interworking
1701 support. Improve soft-float code.
1703 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1705 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1706 (libffi_exit): New function.
1707 (libffi_init): Build the testglue wrapper if needed.
1709 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1712 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1714 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1716 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1718 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1720 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1721 * testsuite/libffi.call/float3.c: New test case.
1723 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1725 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1726 the function returning a structure pointed with R2.
1727 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1728 the structure return value if T bit set. Emit position
1729 independent code and EH data if PIC.
1731 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1733 * Makefile.am: Add m32r support.
1734 * configure.ac: Likewise.
1735 * Makefile.in: Regenerate.
1736 * confiugre: Regenerate.
1737 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1738 (uint64, sint64, double, longdouble)
1739 * src/m32r: New directory.
1740 * src/m32r/ffi.c: New file.
1741 * src/m32r/sysv.S: Likewise.
1742 * src/m32r/ffitarget.h: Likewise.
1744 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1746 * testsuite/libffi.call/negint.c: New test case.
1748 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1751 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1752 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1753 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1756 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1758 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1759 * testsuite/libffi.call/promotion.c: Likewise.
1760 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1762 * testsuite/libffi.call/return_sc.c: Likewise.
1763 * testsuite/libffi.call/return_uc.c: Likewise.
1765 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1767 * src/powerpc/darwin.S: Fix comments and identation.
1768 * src/powerpc/darwin_closure.S: Likewise.
1770 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1772 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1773 (ffi_prep_args): Handle longdouble arguments.
1774 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1776 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1777 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1779 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1780 * src/types.c: Defined longdouble size and alignment for darwin.
1782 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1784 * src/powerpc/aix.S: Remove whitespaces.
1785 * src/powerpc/aix_closure.S: Likewise.
1786 * src/powerpc/asm.h: Likewise.
1787 * src/powerpc/ffi.c: Likewise.
1788 * src/powerpc/ffitarget.h: Likewise.
1789 * src/powerpc/linux64.S: Likewise.
1790 * src/powerpc/linux64_closure.S: Likewise.
1791 * src/powerpc/ppc_closure.S: Likewise.
1792 * src/powerpc/sysv.S: Likewise.
1794 2004-08-30 Anthony Green <green@redhat.com>
1796 * Makefile.am: Add frv support.
1797 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1798 * configure.ac: Read configure.host.
1799 * configure.in: Read configure.host.
1800 * configure.host: New file. frv-elf needs libgloss.
1801 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
1802 alignment. This is needed to frv and shouldn't harm the others.
1803 * include/ffi_common.h (ALIGN_DOWN): New macro.
1804 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1806 2004-08-24 David Daney <daney@avtrex.com>
1808 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1809 * testsuite/libffi.call/closure_fn1.c: Likewise.
1810 * testsuite/libffi.call/closure_fn2.c Likewise.
1811 * testsuite/libffi.call/closure_fn3.c: Likewise.
1812 * testsuite/libffi.call/closure_fn4.c: Likewise.
1813 * testsuite/libffi.call/closure_fn5.c: Likewise.
1814 * testsuite/libffi.call/cls_18byte.c: Likewise.
1815 * testsuite/libffi.call/cls_19byte.c: Likewise.
1816 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1817 * testsuite/libffi.call/cls_20byte.c: Likewise.
1818 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1819 * testsuite/libffi.call/cls_24byte.c: Likewise.
1820 * testsuite/libffi.call/cls_2byte.c: Likewise.
1821 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1822 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1823 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1824 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1825 * testsuite/libffi.call/cls_4byte.c: Likewise.
1826 * testsuite/libffi.call/cls_64byte.c: Likewise.
1827 * testsuite/libffi.call/cls_6byte.c: Likewise.
1828 * testsuite/libffi.call/cls_7byte.c: Likewise.
1829 * testsuite/libffi.call/cls_8byte.c: Likewise.
1830 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1831 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1832 * testsuite/libffi.call/cls_align_double.c: Likewise.
1833 * testsuite/libffi.call/cls_align_float.c: Likewise.
1834 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1835 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1836 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1837 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1838 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1839 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1840 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1841 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1842 * testsuite/libffi.call/cls_double.c: Likewise.
1843 * testsuite/libffi.call/cls_float.c: Likewise.
1844 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1845 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1846 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1847 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1848 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1849 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1850 * testsuite/libffi.call/cls_schar.c: Likewise.
1851 * testsuite/libffi.call/cls_sint.c: Likewise.
1852 * testsuite/libffi.call/cls_sshort.c: Likewise.
1853 * testsuite/libffi.call/cls_uchar.c: Likewise.
1854 * testsuite/libffi.call/cls_uint.c: Likewise.
1855 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1856 * testsuite/libffi.call/cls_ushort.c: Likewise.
1857 * testsuite/libffi.call/nested_struct.c: Likewise.
1858 * testsuite/libffi.call/nested_struct1.c: Likewise.
1859 * testsuite/libffi.call/nested_struct2.c: Likewise.
1860 * testsuite/libffi.call/nested_struct3.c: Likewise.
1861 * testsuite/libffi.call/problem1.c: Likewise.
1862 * testsuite/libffi.special/unwindtest.cc: Likewise.
1863 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1865 * testsuite/libffi.call/cls_16byte.c: Likewise.
1866 * testsuite/libffi.call/cls_5byte.c: Likewise.
1868 2004-08-23 David Daney <daney@avtrex.com>
1871 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1872 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1873 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1874 parameters and return types.
1875 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1876 (ffi_prep_closure): Ditto.
1877 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1878 alignment calculations.
1879 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1880 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1882 2004-08-14 Casey Marshall <csm@gnu.org>
1884 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1885 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1886 integer (O32 ABI only).
1887 (ffi_prep_closure): new function.
1888 (ffi_closure_mips_inner_O32): new function.
1889 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
1890 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1891 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1892 64 bit integers correctly.
1893 (ffi_closure_O32): new function.
1894 Added DWARF-2 unwind info for both functions.
1896 2004-08-10 Andrew Haley <aph@redhat.com>
1898 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1900 2004-08-01 Robert Millan <robertmh@gnu.org>
1902 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1903 * configure: Regenerate.
1905 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1907 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1908 and mmap() explicitly instead of relying on preset autoconf cache
1910 * aclocal.m4: Regenerate.
1911 * configure: Regenerate.
1913 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1915 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1916 (ffi_check_float_struct): Remove unused prototype.
1918 2004-06-30 Geoffrey Keating <geoffk@apple.com>
1920 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1921 character on Darwin, use '\n\t' instead.
1923 2004-06-26 Matthias Klose <doko@debian.org>
1925 * libtool-version: Fix typo in revision/age.
1927 2004-06-17 Matthias Klose <doko@debian.org>
1929 * libtool-version: New.
1930 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1931 * Makefile.in: Regenerate.
1933 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1935 * Makefile.am: Remove useless multilib rules.
1936 * Makefile.in: Regenerate.
1937 * aclocal.m4: Regenerate with automake 1.8.5.
1938 * configure.ac: Remove useless multilib configury.
1939 * configure: Regenerate.
1941 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1943 * .cvsignore: New file.
1945 2004-06-10 Jakub Jelinek <jakub@redhat.com>
1947 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1949 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1952 2004-06-06 Sean McNeil <sean@mcneil.com>
1954 * configure.ac: Add x86_64-*-freebsd* support.
1955 * configure: Regenerate.
1957 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
1960 * configure.ac: Test for existence of mmap and sys/mman.h before
1961 checking blacklist. Fix suggested by Jim Wilson.
1962 * configure: Regenerate.
1964 2004-04-26 Matt Austern <austern@apple.com>
1966 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1968 * src/powerpc/darwin_closure.S: Likewise.
1970 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1972 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1973 error. Reported by Thomas Heller <theller@python.net>.
1974 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1975 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1977 2004-03-20 Matthias Klose <doko@debian.org>
1979 * src/pa/linux.S: Fix typo.
1981 2004-03-19 Matthias Klose <doko@debian.org>
1983 * Makefile.am: Update.
1984 * Makefile.in: Regenerate.
1985 * src/pa/ffi.h.in: Remove.
1986 * src/pa/ffitarget.h: New file.
1988 2004-02-10 Randolph Chung <tausq@debian.org>
1990 * Makefile.am: Add PA support.
1991 * Makefile.in: Regenerate.
1992 * include/Makefile.in: Regenerate.
1993 * configure.ac: Add PA target.
1994 * configure: Regenerate.
1995 * src/pa/ffi.c: New file.
1996 * src/pa/ffi.h.in: Add PA support.
1997 * src/pa/linux.S: New file.
1998 * prep_cif.c: Add PA support.
2000 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2002 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2004 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2005 with ecif->cif->flags.
2006 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2008 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2009 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2010 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2011 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2013 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2015 * configure.in: Rename file to ...
2016 * configure.ac: ... this.
2017 * fficonfig.h.in: Regenerate.
2018 * Makefile.in: Regenerate.
2019 * include/Makefile.in: Regenerate.
2020 * testsuite/Makefile.in: Regenerate.
2022 2004-03-12 Matt Austern <austern@apple.com>
2024 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2025 changes in EH format resulting from addition of linkonce support.
2026 * src/powerpc/darwin_closure.S: Likewise.
2028 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2029 Paolo Bonzini <bonzini@gnu.org>
2031 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2032 Remove VPATH. Remove rules for object files. Remove multilib support.
2033 (AM_CCASFLAGS): Add.
2034 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2035 (AC_PREREQ): Bump version to 2.59.
2036 (AC_INIT): Fill with version info and bug address.
2037 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2038 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2039 De-precious CC so that the right flags are passed down to multilibs.
2040 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2041 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2042 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2043 * configure: Rebuilt.
2044 * aclocal.m4: Likewise.
2045 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2046 * fficonfig.h.in: Likewise.
2048 2004-03-11 Andreas Schwab <schwab@suse.de>
2050 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2051 arguments from fp registers only for the first 8 parameter slots.
2052 Don't convert a float parameter when passed in memory.
2054 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2056 * configure: Regenerate for config/accross.m4 correction.
2058 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2060 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2061 ecif->cif->bytes to bytes.
2062 (ffi_prep_cif_machdep): Add braces around nested if statement.
2064 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2066 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2068 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2069 (ffi_closure_helper_LINUX64): Fix typo.
2070 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2072 * testsuite/libffi.call/float.c: Likewise.
2073 * testsuite/libffi.call/float2.c: Likewise.
2075 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2077 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2078 long double function return and long double arg handling.
2079 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2080 Use "end_pfr" instead of "nf". Correct long double handling.
2082 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2084 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2085 space for long double return value. Adjust stack frame and offsets.
2086 Load f2 long double return.
2088 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2090 * src/types.c: Use 16 byte long double for POWERPC64.
2092 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2094 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2095 when the structure return address is passed in %o0.
2096 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2097 (ffi_v9_layout_struct): Align the field following a nested structure
2098 on a word boundary. Use memmove instead of memcpy.
2099 (ffi_call): Update call to ffi_V9_return_struct.
2100 (ffi_prep_closure): Define 'ctx' only for V8.
2101 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2102 and ffi_closure_sparc_inner_v9.
2103 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2104 Always skip the structure return address. For structures and long
2105 doubles, copy the argument directly.
2106 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2107 if required. Shift the maximum floating-point slot accordingly. For
2108 big structures, copy the argument directly; otherwise, left-justify the
2109 argument and call ffi_v9_layout_struct to lay out the structure on
2111 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2112 (ffi_closure_v8): Pass the structure return address. Update call to
2113 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2114 Skip the 'unimp' insn when returning long doubles and structures.
2115 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2116 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2117 FFI_TYPE_INT handling. Load structures both in integers and
2118 floating-point registers on return.
2119 * README: Update status of the SPARC port.
2121 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2123 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2125 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2127 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2129 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2130 value as of type ffi_arg, not unsigned int.
2132 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2134 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2137 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2139 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2142 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2144 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2147 2003-12-10 Richard Henderson <rth@redhat.com>
2149 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2150 size_t instead of int.
2152 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2154 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2155 * testsuite/libffi.call/many_win32.c (main): Replace variable
2156 int i with unsigned long ul.
2158 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2159 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2160 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2161 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2162 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2163 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2164 * testsuite/libffi.call/cls_align_float.c: Likewise.
2165 * testsuite/libffi.call/cls_align_double.c: Likewise.
2166 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2167 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2169 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2172 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2173 Align arguments to 32 bits.
2175 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2178 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2179 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2180 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2181 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2182 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2183 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2185 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2187 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2189 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2190 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2192 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2194 * Makefile.in: Rebuilt.
2195 * configure: Likewise.
2196 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2199 2003-11-21 Andreas Jaeger <aj@suse.de>
2200 Andreas Tobler <a.tobler@schweiz.ch>
2202 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2203 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2204 * Makefile.in: Rebuilt.
2205 * aclocal.m4: Likewise.
2206 * configure: Likewise.
2207 * fficonfig.h.in: Likewise.
2208 * testsuite/lib/libffi-dg.exp: Add include dir.
2209 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2210 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2211 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2212 for ffi_closure if available.
2213 * testsuite/libffi.call/closure_fn1.c: Likewise.
2214 * testsuite/libffi.call/closure_fn2.c: Likewise.
2215 * testsuite/libffi.call/closure_fn3.c: Likewise.
2216 * testsuite/libffi.call/closure_fn4.c: Likewise.
2217 * testsuite/libffi.call/closure_fn5.c: Likewise.
2218 * testsuite/libffi.call/cls_12byte.c: Likewise.
2219 * testsuite/libffi.call/cls_16byte.c: Likewise.
2220 * testsuite/libffi.call/cls_18byte.c: Likewise.
2221 * testsuite/libffi.call/cls_19byte.c: Likewise.
2222 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2223 * testsuite/libffi.call/cls_20byte.c: Likewise.
2224 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2225 * testsuite/libffi.call/cls_24byte.c: Likewise.
2226 * testsuite/libffi.call/cls_2byte.c: Likewise.
2227 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2228 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2229 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2230 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2231 * testsuite/libffi.call/cls_4byte.c: Likewise.
2232 * testsuite/libffi.call/cls_5byte.c: Likewise.
2233 * testsuite/libffi.call/cls_64byte.c: Likewise.
2234 * testsuite/libffi.call/cls_6byte.c: Likewise.
2235 * testsuite/libffi.call/cls_7byte.c: Likewise.
2236 * testsuite/libffi.call/cls_8byte.c: Likewise.
2237 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2238 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2239 * testsuite/libffi.call/cls_double.c: Likewise.
2240 * testsuite/libffi.call/cls_float.c: Likewise.
2241 * testsuite/libffi.call/cls_schar.c: Likewise.
2242 * testsuite/libffi.call/cls_sint.c: Likewise.
2243 * testsuite/libffi.call/cls_sshort.c: Likewise.
2244 * testsuite/libffi.call/cls_uchar.c: Likewise.
2245 * testsuite/libffi.call/cls_uint.c: Likewise.
2246 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2247 * testsuite/libffi.call/cls_ushort.c: Likewise.
2248 * testsuite/libffi.call/nested_struct.c: Likewise.
2249 * testsuite/libffi.call/nested_struct1.c: Likewise.
2250 * testsuite/libffi.call/nested_struct2.c: Likewise.
2251 * testsuite/libffi.call/nested_struct3.c: Likewise.
2252 * testsuite/libffi.call/problem1.c: Likewise.
2253 * testsuite/libffi.special/unwindtest.cc: Likewise.
2255 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2257 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2259 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2261 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2262 Add -lgcc_s to additional flags.
2264 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2266 * configure.in, include/Makefile.am: PR libgcj/11147, install
2267 the ffitarget.h header file in a gcc versioned and target
2269 * configure: Regenerated.
2270 * Makefile.in, include/Makefile.in: Likewise.
2271 * testsuite/Makefile.in: Likewise.
2273 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2275 * testsuite/libffi.call/closure_fn0.c: Print result and check
2276 with dg-output to make debugging easier.
2277 * testsuite/libffi.call/closure_fn1.c: Likewise.
2278 * testsuite/libffi.call/closure_fn2.c: Likewise.
2279 * testsuite/libffi.call/closure_fn3.c: Likewise.
2280 * testsuite/libffi.call/closure_fn4.c: Likewise.
2281 * testsuite/libffi.call/closure_fn5.c: Likewise.
2282 * testsuite/libffi.call/cls_12byte.c: Likewise.
2283 * testsuite/libffi.call/cls_16byte.c: Likewise.
2284 * testsuite/libffi.call/cls_18byte.c: Likewise.
2285 * testsuite/libffi.call/cls_19byte.c: Likewise.
2286 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2287 * testsuite/libffi.call/cls_20byte.c: Likewise.
2288 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2289 * testsuite/libffi.call/cls_24byte.c: Likewise.
2290 * testsuite/libffi.call/cls_2byte.c: Likewise.
2291 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2292 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2293 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2294 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2295 * testsuite/libffi.call/cls_4byte.c: Likewise.
2296 * testsuite/libffi.call/cls_5byte.c: Likewise.
2297 * testsuite/libffi.call/cls_64byte.c: Likewise.
2298 * testsuite/libffi.call/cls_6byte.c: Likewise.
2299 * testsuite/libffi.call/cls_7byte.c: Likewise.
2300 * testsuite/libffi.call/cls_8byte.c: Likewise.
2301 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2302 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2303 * testsuite/libffi.call/cls_double.c: Likewise.
2304 * testsuite/libffi.call/cls_float.c: Likewise.
2305 * testsuite/libffi.call/cls_schar.c: Likewise.
2306 * testsuite/libffi.call/cls_sint.c: Likewise.
2307 * testsuite/libffi.call/cls_sshort.c: Likewise.
2308 * testsuite/libffi.call/cls_uchar.c: Likewise.
2309 * testsuite/libffi.call/cls_uint.c: Likewise.
2310 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2311 * testsuite/libffi.call/cls_ushort.c: Likewise.
2312 * testsuite/libffi.call/problem1.c: Likewise.
2314 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2317 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2319 * testsuite/libffi.call/cls_9byte2.c: New test case.
2320 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2321 * testsuite/libffi.call/cls_64byte.c: Likewise.
2322 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2323 * testsuite/libffi.call/cls_19byte.c: Likewise.
2324 * testsuite/libffi.call/cls_18byte.c: Likewise.
2325 * testsuite/libffi.call/closure_fn4.c: Likewise.
2326 * testsuite/libffi.call/closure_fn5.c: Likewise.
2327 * testsuite/libffi.call/cls_schar.c: Likewise.
2328 * testsuite/libffi.call/cls_sint.c: Likewise.
2329 * testsuite/libffi.call/cls_sshort.c: Likewise.
2330 * testsuite/libffi.call/nested_struct2.c: Likewise.
2331 * testsuite/libffi.call/nested_struct3.c: Likewise.
2333 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2335 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2336 * testsuite/libffi.call/cls_uchar.c: Likewise.
2337 * testsuite/libffi.call/cls_uint.c: Likewise.
2338 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2339 * testsuite/libffi.call/cls_ushort.c: Likewise.
2340 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2342 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2344 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2347 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2349 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2350 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2352 2003-10-22 David Daney <ddaney@avtrex.com>
2354 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2355 __attribute__((__mode__(__SI__))) and friends.
2357 2003-10-22 Andreas Schwab <schwab@suse.de>
2359 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2361 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2363 * configure.in: AC_LINK_FILES(ffitarget.h).
2364 * configure: Regenerate.
2365 * Makefile.in: Likewise.
2366 * include/Makefile.in: Likewise.
2367 * testsuite/Makefile.in: Likewise.
2368 * fficonfig.h.in: Likewise.
2370 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2371 Richard Henderson <rth@redhat.com>
2373 Avoid that ffi.h includes fficonfig.h.
2375 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2376 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2377 (TARGET_SRC_MIPS_SGI): Removed.
2378 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2379 (MIPS_SGI): Removed.
2380 (CLEANFILES): Removed.
2381 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2383 * acconfig.h: Removed.
2384 * configure.in: Compute sizeofs only for double and long double.
2385 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2386 into AC_DEFINE instead of using acconfig.h. Create
2387 include/ffitarget.h instead of include/fficonfig.h. Rename
2388 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2389 AC_DEFINE EH_FRAME_FLAGS.
2390 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2391 (hack_DATA): Add ffitarget.h.
2392 * include/ffi.h.in: Remove all system specific definitions.
2393 Declare raw API even if it is not installed, why bother?
2394 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2395 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2396 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2397 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2398 * include/ffi_common.h (bool): Do not define.
2399 (ffi_assert): Accept failed assertion.
2400 (ffi_type_test): Return void and accept file/line.
2401 (FFI_ASSERT): Pass stringized failed assertion.
2402 (FFI_ASSERT_AT): New macro.
2403 (FFI_ASSERT_VALID_TYPE): New macro.
2404 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2405 UINT64, SINT64): Define here with gcc's __attribute__ macro
2407 (FLOAT32, ALIGN): Define here instead of in ffi.h
2408 * include/ffi-mips.h: Removed. Its content moved to
2409 src/mips/ffitarget.h after separating assembly and C sections.
2410 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2411 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2412 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2413 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2414 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2415 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2416 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2417 (ffi_type_test): Rewritten.
2418 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2419 FFI_ASSERT_VALID_TYPE.
2420 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2421 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2422 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2423 src/s390/ffitarget.h, src/sh/ffitarget.h,
2424 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2425 src/x86/ffitarget.h: New files.
2426 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2427 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2428 src/powerpc/aix.S, src/powerpc/darwin.S,
2429 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2430 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2431 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2432 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2433 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2436 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2438 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2439 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2441 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2443 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2444 Used when FFI_DEBUG = 1.
2446 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2448 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2451 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2453 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2454 FFI_MIPS_O32 for O32 ABI.
2456 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2458 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2459 SPARC64. Cleanup whitespaces.
2461 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2463 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2464 strongarm, xscale. Cleanup whitespaces.
2465 * testsuite/libffi.call/closure_fn1.c: Likewise.
2466 * testsuite/libffi.call/closure_fn2.c: Likewise.
2467 * testsuite/libffi.call/closure_fn3.c: Likewise.
2468 * testsuite/libffi.call/cls_12byte.c: Likewise.
2469 * testsuite/libffi.call/cls_16byte.c: Likewise.
2470 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2471 * testsuite/libffi.call/cls_20byte.c: Likewise.
2472 * testsuite/libffi.call/cls_24byte.c: Likewise.
2473 * testsuite/libffi.call/cls_2byte.c: Likewise.
2474 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2475 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2476 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2477 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2478 * testsuite/libffi.call/cls_4byte.c: Likewise.
2479 * testsuite/libffi.call/cls_5byte.c: Likewise.
2480 * testsuite/libffi.call/cls_6byte.c: Likewise.
2481 * testsuite/libffi.call/cls_7byte.c: Likewise.
2482 * testsuite/libffi.call/cls_8byte.c: Likewise.
2483 * testsuite/libffi.call/cls_double.c: Likewise.
2484 * testsuite/libffi.call/cls_float.c: Likewise.
2485 * testsuite/libffi.call/cls_uchar.c: Likewise.
2486 * testsuite/libffi.call/cls_uint.c: Likewise.
2487 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2488 * testsuite/libffi.call/cls_ushort.c: Likewise.
2489 * testsuite/libffi.call/nested_struct.c: Likewise.
2490 * testsuite/libffi.call/nested_struct1.c: Likewise.
2491 * testsuite/libffi.call/problem1.c: Likewise.
2492 * testsuite/libffi.special/unwindtest.cc: Likewise.
2493 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2495 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2497 * src/powerpc/aix.S: Cleanup whitespaces.
2498 * src/powerpc/aix_closure.S: Likewise.
2500 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2502 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2503 * src/powerpc/darwin_closure.S: Likewise.
2504 * src/powerpc/ffi_darwin.c: Likewise.
2506 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2507 David Edelsohn <edelsohn@gnu.org>
2509 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2510 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2512 * src/powerpc/darwin_closure.S: Likewise.
2513 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2514 according to the Darwin/AIX ABI.
2515 (ffi_prep_cif_machdep): Likewise.
2516 (ffi_closure_helper_DARWIN): Likewise.
2517 Remove the outgoing parameter stack logic. Simplify the evaluation
2518 of the different CASE types.
2519 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2520 statement in the trampoline code.
2522 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2524 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2525 for the register size.
2526 (ffi_closure_helper_SYSV): Handle the structure return value
2528 (ffi_closure_helper_SYSV): Return the appropriate type when
2529 the registers are used for the structure return value.
2530 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2531 the 64-bit return value. Update copyright years.
2533 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2535 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2536 srcdir for ffi_mips.h.
2538 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2540 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2542 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2543 placement of float result.
2544 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2545 cast of "resp" for big-endian 64 bit machines.
2547 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2549 * src/types.c (double, longdouble): Merge identical SH and ARM
2550 typedefs, and add POWERPC64.
2551 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2552 struct split over gpr and rest.
2553 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2554 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2556 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2558 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2561 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2563 * configure: Regenerate.
2565 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2567 * Makefile.am: Remove build rules for ffitest.
2568 * Makefile.in: Rebuilt.
2570 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2572 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2573 about implicit declaration of abort().
2575 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2577 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2578 * Makefile.in: Rebuilt.
2579 * configure.in: Add dejagnu test framework.
2580 * configure: Rebuilt.
2582 * testsuite/Makefile.am: New file.
2583 * testsuite/Makefile.in: Built
2584 * testsuite/lib/libffi-dg.exp: New file.
2585 * testsuite/config/default.exp: Likewise.
2586 * testsuite/libffi.call/call.exp: Likewise.
2587 * testsuite/libffi.call/ffitest.h: Likewise.
2588 * testsuite/libffi.call/closure_fn0.c: Likewise.
2589 * testsuite/libffi.call/closure_fn1.c: Likewise.
2590 * testsuite/libffi.call/closure_fn2.c: Likewise.
2591 * testsuite/libffi.call/closure_fn3.c: Likewise.
2592 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2593 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2594 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2595 * testsuite/libffi.call/cls_2byte.c: Likewise.
2596 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2597 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2598 * testsuite/libffi.call/cls_4byte.c: Likewise.
2599 * testsuite/libffi.call/cls_5byte.c: Likewise.
2600 * testsuite/libffi.call/cls_6byte.c: Likewise.
2601 * testsuite/libffi.call/cls_7byte.c: Likewise.
2602 * testsuite/libffi.call/cls_8byte.c: Likewise.
2603 * testsuite/libffi.call/cls_12byte.c: Likewise.
2604 * testsuite/libffi.call/cls_16byte.c: Likewise.
2605 * testsuite/libffi.call/cls_20byte.c: Likewise.
2606 * testsuite/libffi.call/cls_24byte.c: Likewise.
2607 * testsuite/libffi.call/cls_double.c: Likewise.
2608 * testsuite/libffi.call/cls_float.c: Likewise.
2609 * testsuite/libffi.call/cls_uchar.c: Likewise.
2610 * testsuite/libffi.call/cls_uint.c: Likewise.
2611 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2612 * testsuite/libffi.call/cls_ushort.c: Likewise.
2613 * testsuite/libffi.call/float.c: Likewise.
2614 * testsuite/libffi.call/float1.c: Likewise.
2615 * testsuite/libffi.call/float2.c: Likewise.
2616 * testsuite/libffi.call/many.c: Likewise.
2617 * testsuite/libffi.call/many_win32.c: Likewise.
2618 * testsuite/libffi.call/nested_struct.c: Likewise.
2619 * testsuite/libffi.call/nested_struct1.c: Likewise.
2620 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2621 * testsuite/libffi.call/problem1.c: Likewise.
2622 * testsuite/libffi.call/promotion.c: Likewise.
2623 * testsuite/libffi.call/return_ll.c: Likewise.
2624 * testsuite/libffi.call/return_sc.c: Likewise.
2625 * testsuite/libffi.call/return_uc.c: Likewise.
2626 * testsuite/libffi.call/strlen.c: Likewise.
2627 * testsuite/libffi.call/strlen_win32.c: Likewise.
2628 * testsuite/libffi.call/struct1.c: Likewise.
2629 * testsuite/libffi.call/struct2.c: Likewise.
2630 * testsuite/libffi.call/struct3.c: Likewise.
2631 * testsuite/libffi.call/struct4.c: Likewise.
2632 * testsuite/libffi.call/struct5.c: Likewise.
2633 * testsuite/libffi.call/struct6.c: Likewise.
2634 * testsuite/libffi.call/struct7.c: Likewise.
2635 * testsuite/libffi.call/struct8.c: Likewise.
2636 * testsuite/libffi.call/struct9.c: Likewise.
2637 * testsuite/libffi.special/special.exp: New file.
2638 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2639 * testsuite/libffi.special/unwindtest.cc: Likewise.
2642 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2644 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2647 2003-08-02 Alan Modra <amodra@bigpond.net.au>
2649 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2651 (ffi_closure_helper_LINUX64): Likewise.
2652 * src/powerpc/linux64.S: Remove code writing to parm save area.
2653 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2654 address in lr from ffi_closure_helper_LINUX64 call to calculate
2655 table address. Optimize function tail.
2657 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2659 * src/sparc/ffi.c: Handle all floating point registers.
2660 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2662 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2664 * README: Note that libffi is not part of GCC. Update the project
2667 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2669 * src/powerpc/ppc_closure.S: Include ffi.h.
2671 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2673 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2674 Use C style comments.
2676 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2678 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2679 * Makefile.in: Regenerate.
2680 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2681 * configure: Regenerate.
2682 * include/ffi.h.in: Add SHmedia support.
2683 * src/sh64/ffi.c: New file.
2684 * src/sh64/sysv.S: New file.
2686 2003-05-16 Jakub Jelinek <jakub@redhat.com>
2688 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2689 should be read-only.
2690 * configure: Rebuilt.
2691 * fficonfig.h.in: Rebuilt.
2692 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2693 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2694 * src/powerpc/linux64.S: Likewise.
2695 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2696 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2697 if -fpic/-fPIC/-mrelocatable.
2698 * src/powerpc/powerpc_closure.S: Likewise.
2699 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2700 #write in .eh_frame flags.
2701 * src/sparc/v9.S: Likewise.
2702 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2703 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2704 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2706 2003-05-07 Jeff Sturm <jsturm@one-point.com>
2708 Fixes PR bootstrap/10656
2709 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2710 support for .register pseudo-op.
2711 * src/sparc/v8.S: Use it.
2712 * fficonfig.h.in: Rebuilt.
2713 * configure: Rebuilt.
2715 2003-04-18 Jakub Jelinek <jakub@redhat.com>
2717 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2718 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2719 Make it the default on POWERPC64.
2720 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2721 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2722 * configure: Rebuilt.
2723 * src/powerpc/ffi.c (hidden): Define.
2724 (ffi_prep_args_SYSV): Renamed from
2725 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2726 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2727 ASM_NEEDS_REGISTERS64): New.
2728 (ffi_prep_args64): New function.
2729 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2730 (ffi_call): Likewise.
2731 (ffi_prep_closure): Likewise.
2732 (flush_icache): Surround by #ifndef POWERPC64.
2733 (ffi_dblfl): New union type.
2734 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2735 (ffi_closure_helper_LINUX64): New function.
2736 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2738 * src/powerpc/sysv.S: Likewise.
2739 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2740 * src/powerpc/linux64.S: New file.
2741 * src/powerpc/linux64_closure.S: New file.
2742 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2743 src/powerpc/linux64_closure.S.
2744 (TARGET_SRC_POWERPC): Likewise.
2746 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2747 closure_test_fn3): Fix result printing on big-endian 64-bit
2749 (main): Print tst2_arg instead of uninitialized tst2_result.
2751 * src/ffitest.c (main): Hide what closure pointer really points to
2754 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
2756 * configure.in (arm-*-netbsdelf*): Add configuration.
2757 (configure): Regenerated.
2759 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
2761 * include/Makefile.in: Regenerate.
2763 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2765 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2767 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2768 Receive closure pointer through parameter, read args using
2769 __builtin_dwarf_cfa.
2770 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2772 2003-03-12 Andreas Schwab <schwab@suse.de>
2774 * configure.in: Avoid trailing /. in toolexeclibdir.
2775 * configure: Rebuilt.
2777 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2779 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
2781 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2783 * libffi/src/powerpc/darwin_closure.S:
2784 Fix alignement bug, allocate 8 bytes for the result.
2785 * libffi/src/powerpc/aix_closure.S:
2787 * libffi/src/powerpc/ffi_darwin.c:
2788 Update stackframe description for aix/darwin_closure.S.
2790 2003-02-06 Jakub Jelinek <jakub@redhat.com>
2792 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2795 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2796 Andreas Schwab <schwab@suse.de>
2798 * configure.in: Adjust command to source config-ml.in to account
2799 for changes to the libffi_basedir definition.
2800 (libffi_basedir): Remove ${srcdir} from value and include trailing
2803 * configure: Regenerate.
2805 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2807 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2809 2003-01-28 Andrew Haley <aph@redhat.com>
2811 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2812 * src/x86/ffi64.c (ffi_prep_closure): New.
2813 (ffi_closure_UNIX64_inner): New.
2814 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2816 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
2818 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2819 Remove USE_LIBDIR conditional.
2820 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2821 * Makefile.in, configure: Rebuilt.
2823 2003-01027 David Edelsohn <edelsohn@gnu.org>
2825 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2826 * Makefile.in: Regenerate.
2828 2003-01-22 Andrew Haley <aph@redhat.com>
2830 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2833 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2835 * src/powerpc/darwin.S: Add unwind info.
2836 * src/powerpc/darwin_closure.S: Likewise.
2838 2003-01-14 Andrew Haley <aph@redhat.com>
2840 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2841 (ffi_prep_cif_machdep): Likewise.
2842 * src/x86/unix64.S: Add unwind info.
2844 2003-01-14 Andreas Jaeger <aj@suse.de>
2846 * src/ffitest.c (main): Only use ffi_closures if those are
2849 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2851 * libffi/src/ffitest.c
2852 add closure testcases
2854 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2856 * libffi/src/powerpc/ffi.c
2857 fix alignment bug for float (4 byte aligned iso 8 byte)
2859 2003-01-09 Geoffrey Keating <geoffk@apple.com>
2861 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2862 * src/powerpc/darwin.S: Remove RCS version string.
2864 2003-01-03 Jeff Sturm <jsturm@one-point.com>
2866 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2867 * src/ffitest.c (main): Use static storage for closure.
2868 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2869 * src/sparc/v8.S (ffi_closure_v8): New.
2870 * src/sparc/v9.S (ffi_closure_v9): New.
2872 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2874 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2875 enumeration for X86_WIN32.
2876 * src/x86/win32.S: Added ffi_call_STDCALL function
2878 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2879 switch cases for recognising FFI_STDCALL and
2880 calling ffi_call_STDCALL if target is X86_WIN32.
2881 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2882 stdcall versions of the "my_strlen" and "many"
2883 test functions (for X86_WIN32).
2884 Added test cases to test stdcall invocation using
2887 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2889 * src/sh/sysv.S: Add DWARF2 unwind info.
2891 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2893 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2895 2002-11-26 Jim Wilson <wilson@redhat.com>
2897 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2899 2002-11-23 H.J. Lu <hjl@gnu.org>
2901 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2902 Include ../config/accross.m4.
2903 * aclocal.m4; Rebuild.
2904 * configure: Likewise.
2906 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2908 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2910 2002-11-11 DJ Delorie <dj@redhat.com>
2912 * configure.in: Look for common files in the right place.
2914 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2916 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2917 raw data as _Jv_word values, not ffi_raw.
2918 (ffi_java_ptrarray_to_raw): Likewise.
2919 (ffi_java_rvalue_to_raw): New function.
2920 (ffi_java_raw_call): Call it.
2921 (ffi_java_raw_to_rvalue): New function.
2922 (ffi_java_translate_args): Call it.
2923 * src/ffitest.c (closure_test_fn): Interpret return value
2924 as ffi_arg, not int.
2925 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2926 FFI_TYPE_POINTER case.
2927 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2928 values extended to word size.
2930 2002-10-02 Andreas Jaeger <aj@suse.de>
2932 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2934 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
2936 * include/ffi.h.in: Fix i386 win32 compilation.
2938 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2940 * configure.in: Add s390x-*-linux-* target.
2941 * configure: Regenerate.
2942 * include/ffi.h.in: Define S390X for s390x targets.
2943 (FFI_CLOSURES): Define for s390/s390x.
2944 (FFI_TRAMPOLINE_SIZE): Likewise.
2945 (FFI_NATIVE_RAW_API): Likewise.
2946 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2947 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
2948 * src/s390/ffi.c: Major rework of existing code. Add support for
2949 s390x targets. Add closure support.
2950 * src/s390/sysv.S: Likewise.
2952 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
2954 * src/arm/sysv.S: Fix typo.
2956 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
2958 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2959 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2960 (ffi_call_SYSV): Handle soft-float.
2962 2002-09-27 Bo Thorsen <bo@suse.de>
2964 * include/ffi.h.in: Fix multilib x86-64 support.
2966 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2968 * Makefile.am (all-multi): Fix multilib parallel build.
2970 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2972 * configure.in (sh[34]*-*-linux*): Add brackets.
2973 * configure: Regenerate.
2975 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2977 * Makefile.am: Add SH support.
2978 * Makefile.in: Regenerate.
2979 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2980 * configure: Regenerate.
2981 * include/ffi.h.in: Add SH support.
2982 * src/sh/ffi.c: New file.
2983 * src/sh/sysv.S: New file.
2984 * src/types.c: Add SH support.
2986 2002-07-16 Bo Thorsen <bo@suse.de>
2988 * src/x86/ffi64.c: New file that adds x86-64 support.
2989 * src/x86/unix64.S: New file that handles argument setup for
2991 * src/x86/sysv.S: Don't use this on x86-64.
2992 * src/x86/ffi.c: Don't use this on x86-64.
2994 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2996 * src/ffitest.c (struct6): New test that tests a special case in
2998 (struct7): Likewise.
2999 (struct8): Likewise.
3000 (struct9): Likewise.
3001 (closure_test_fn): Silence warning about this when it's not used.
3002 (main): Add the new tests.
3003 (main): Fix a couple of wrong casts and silence some compiler warnings.
3004 * include/ffi.h.in: Add x86-64 ABI definition.
3005 * fficonfig.h.in: Regenerate.
3006 * Makefile.am: Add x86-64 support.
3007 * configure.in: Likewise.
3008 * Makefile.in: Regenerate.
3009 * configure: Likewise.
3011 2002-06-24 Bo Thorsen <bo@suse.de>
3013 * src/types.c: Merge settings for similar architectures.
3014 Add x86-64 sizes and alignments.
3016 2002-06-23 Bo Thorsen <bo@suse.de>
3018 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3019 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3020 * src/mips/ffi.c (ffi_prep_args): Likewise.
3021 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3023 2002-07-18 H.J. Lu (hjl@gnu.org)
3025 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3026 (libffi_la_SOURCES): Support MIPS_LINUX.
3027 (libffi_convenience_la_SOURCES): Likewise.
3028 * Makefile.in: Regenerated.
3030 * configure.in (mips64*-*): Skip.
3031 (mips*-*-linux*): New.
3032 * configure: Regenerated.
3034 * src/mips/ffi.c: Include <sgidefs.h>.
3036 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3038 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3040 2002-05-27 Roger Sayle <roger@eyesopen.com>
3042 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3044 2002-05-27 Bo Thorsen <bo@suse.de>
3046 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3049 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3051 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3052 beginning of function (for older apple cc).
3054 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3056 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3057 script entry, and set LD to it when configuring multilibs.
3058 * configure: Rebuilt.
3060 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3062 * configure.in (sparc64-*-netbsd*): Add target.
3063 (sparc-*-netbsdelf*): Likewise.
3064 * configure: Regenerate.
3066 2002-04-28 David S. Miller <davem@redhat.com>
3068 * configure.in, configure: Fix SPARC test in previous change.
3070 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3072 * Makefile.am: Add Linux for S/390 support.
3073 * Makefile.in: Regenerate.
3074 * configure.in: Add Linux for S/390 support.
3075 * configure: Regenerate.
3076 * include/ffi.h.in: Add Linux for S/390 support.
3077 * src/s390/ffi.c: New file from libffi CVS tree.
3078 * src/s390/sysv.S: New file from libffi CVS tree.
3080 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3082 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3084 * src/sparc/v8.S: Use it.
3085 * src/sparc/v9.S: Likewise.
3086 * fficonfig.h.in: Rebuilt.
3087 * configure: Rebuilt.
3089 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3091 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3093 * src/ia64/unix.S: Add unwind information. Fix comments.
3094 Save sp in a way that's compatible with unwind info.
3095 (ffi_call_unix): Correctly restore sp in all cases.
3096 * src/ia64/ffi.c: Add, fix comments.
3098 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3100 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3102 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3104 * configure.in (alpha*-*-netbsd*): Add target.
3105 * configure: Regenerate.
3107 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3109 * src/sparc/v8.S: Add unwind info.
3110 * src/sparc/v9.S: Likewise.
3112 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3114 * configure.in: Enable i*86-*-netbsdelf*.
3115 * configure: Rebuilt.
3117 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3120 * src/mips/n32.s: Delete
3121 * src/mips/o32.s: Delete
3123 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3125 * configure.in: Enable alpha*-*-freebsd*.
3126 * configure: Rebuilt.
3128 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3130 * Makefile.am: libfficonvenience -> libffi_convenience.
3131 * Makefile.in: Rebuilt.
3133 * Makefile.am: Define ffitest_OBJECTS.
3134 * Makefile.in: Rebuilt.
3136 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3137 David Edelsohn <edelsohn@gnu.org>
3139 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3140 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3141 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3142 * Makefile.in: Regenerate.
3143 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3144 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3145 (flush_icache, flush_range): New functions.
3146 (ffi_closure_helper_DARWIN): New function.
3147 * src/powerpc/aix_closure.S: New file.
3148 * src/powerpc/darwin_closure.S: New file.
3150 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3152 * include/ffi.h.in: Add typedef for ffi_arg.
3153 * src/ffitest.c (main): Declare rint with ffi_arg.
3155 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3157 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3158 number of GPRs for floating-point arguments.
3160 2002-01-31 Anthony Green <green@redhat.com>
3162 * configure: Rebuilt.
3163 * configure.in: Replace CHECK_SIZEOF and endian tests with
3164 cross-compiler friendly macros.
3165 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3168 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3170 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3171 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3173 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3175 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3176 (TARGET_SRC_POWERPC_AIX): New.
3177 (POWERPC_AIX): New stanza.
3178 * Makefile.in: Regenerate.
3179 * configure.in: Add AIX case.
3180 * configure: Regenerate.
3181 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3182 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3183 size. Fix "long double" support.
3184 (ffi_call): Add FFI_AIX case.
3185 * src/powerpc/aix.S: New.
3187 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3189 Implement Darwin PowerPC ABI.
3190 * configure.in: Handle powerpc-*-darwin*.
3191 * Makefile.am: Set source files for POWERPC_DARWIN.
3192 * configure: Rebuilt.
3193 * Makefile.in: Rebuilt.
3194 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3196 * src/powerpc/darwin.S: New file.
3197 * src/powerpc/ffi_darwin.c: New file.
3199 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3201 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3203 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3205 * src/x86/sysv.S: Avoid gas-only .balign directive.
3206 Use C style comments.
3208 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3210 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3211 Fixes PR bootstrap/3563.
3213 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3215 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3217 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3219 * configure.in: Recognize sparc*-sun-* host.
3220 * configure: Regenerate.
3222 2001-06-06 Andrew Haley <aph@redhat.com>
3224 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
3226 2001-06-03 Andrew Haley <aph@redhat.com>
3228 * src/alpha/osf.S: Add unwind info.
3229 * src/powerpc/sysv.S: Add unwind info.
3230 * src/powerpc/ppc_closure.S: Likewise.
3232 2000-05-31 Jeff Sturm <jsturm@one-point.com>
3234 * configure.in: Fix AC_ARG_ENABLE usage.
3235 * configure: Rebuilt.
3237 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3239 * configure.in: Remove warning about beta code.
3240 * configure: Rebuilt.
3242 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3244 * src/ia64/unix.S: Restore stack pointer when returning from
3246 * src/ia64/ffi.c: Fix typo in comment.
3248 2001-04-18 Jim Wilson <wilson@redhat.com>
3250 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3251 to eliminate RAW DV.
3253 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3255 * Makefile.am: Make a libtool convenience library.
3256 * Makefile.in: Rebuilt.
3258 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3260 * configure.in: Use different syntax for subdirectory creation.
3261 * configure: Rebuilt.
3263 2001-03-27 Jon Beniston <jon@beniston.com>
3265 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3266 * configure: Rebuilt.
3267 * Makefile.am: Added X86_WIN32 target support.
3268 * Makefile.in: Rebuilt.
3270 * include/ffi.h.in: Added X86_WIN32 target support.
3272 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3273 * src/types.c: Added X86_WIN32 target support.
3275 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3276 stuff removed and made to work with CygWin's gas.
3278 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3280 * configure.in: Make target subdirectory in build dir.
3281 * Makefile.am: Override suffix based rules to specify correct output
3283 * Makefile.in: Rebuilt.
3284 * configure: Rebuilt.
3286 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3288 * src/powerpc/ppc_closure.S: New file.
3289 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3290 involving long long and register pairs.
3291 (ffi_prep_closure): New function.
3292 (flush_icache): Likewise.
3293 (ffi_closure_helper_SYSV): Likewise.
3294 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3295 (FFI_TRAMPOLINE_SIZE): Likewise.
3296 (FFI_NATIVE_RAW_API): Likewise.
3297 * Makefile.in: Rebuilt.
3298 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3299 (TARGET_SRC_POWERPC): Likewise.
3301 2001-03-19 Tom Tromey <tromey@redhat.com>
3303 * Makefile.in: Rebuilt.
3304 * Makefile.am (ffitest_LDFLAGS): New macro.
3306 2001-03-02 Nick Clifton <nickc@redhat.com>
3308 * include/ffi.h.in: Remove RCS ident string.
3309 * include/ffi_mips.h: Remove RCS ident string.
3310 * src/debug.c: Remove RCS ident string.
3311 * src/ffitest.c: Remove RCS ident string.
3312 * src/prep_cif.c: Remove RCS ident string.
3313 * src/types.c: Remove RCS ident string.
3314 * src/alpha/ffi.c: Remove RCS ident string.
3315 * src/alpha/osf.S: Remove RCS ident string.
3316 * src/arm/ffi.c: Remove RCS ident string.
3317 * src/arm/sysv.S: Remove RCS ident string.
3318 * src/mips/ffi.c: Remove RCS ident string.
3319 * src/mips/n32.S: Remove RCS ident string.
3320 * src/mips/o32.S: Remove RCS ident string.
3321 * src/sparc/ffi.c: Remove RCS ident string.
3322 * src/sparc/v8.S: Remove RCS ident string.
3323 * src/sparc/v9.S: Remove RCS ident string.
3324 * src/x86/ffi.c: Remove RCS ident string.
3325 * src/x86/sysv.S: Remove RCS ident string.
3327 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3329 * include/ffi.h.in: Change sourceware.cygnus.com references to
3332 2000-12-09 Richard Henderson <rth@redhat.com>
3334 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3335 (ffi_closure_osf_inner): Index rather than increment avalue
3336 and arg_types. Give ffi_closure_osf the raw return value type.
3337 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3340 2000-12-07 Richard Henderson <rth@redhat.com>
3342 * src/raw_api.c (ffi_translate_args): Fix typo.
3343 (ffi_prep_closure): Likewise.
3345 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3346 FFI_TRAMPOLINE_SIZE.
3347 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3348 cif->bytes for new ffi_call_osf implementation.
3349 (ffi_prep_args): Absorb into ...
3350 (ffi_call): ... here. Do all stack allocation here and
3351 avoid a callback function.
3352 (ffi_prep_closure, ffi_closure_osf_inner): New.
3353 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3354 (ffi_closure_osf): New.
3356 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3358 * config.guess, config.sub, install-sh: Removed.
3359 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3360 * Makefile.in: Rebuilt.
3362 * acinclude.m4: Include libtool macros from the top level.
3363 * aclocal.m4, configure: Rebuilt.
3365 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3367 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3368 * configure: Rebuilt.
3370 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3372 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3373 memory correctly. Use conditional instructions, not branches where
3376 2000-05-04 Tom Tromey <tromey@cygnus.com>
3378 * configure: Rebuilt.
3379 * configure.in: Match `arm*-*-linux-*'.
3380 From Chris Dornan <cdornan@arm.com>.
3382 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3384 * Makefile.am (SUBDIRS): Define.
3385 (AM_MAKEFLAGS): Likewise.
3386 (Multilib support.): Add section.
3387 * Makefile.in: Rebuilt.
3388 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3389 New variables. Set for gcc using -print-multi-lib. Export them
3391 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3392 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3394 (extra_compiler_flags_value, extra_compiler_flags): Check these
3395 for extra compiler options which need to be passed down in
3398 2000-04-16 Anthony Green <green@redhat.com>
3400 * configure: Rebuilt.
3401 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3403 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3405 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3406 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3407 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3408 Replace all void * sizeofs with sizeof(int).
3409 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3410 different than DOUBLE.
3411 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3412 (ffi_prep_args_v9): New function.
3413 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3414 (ffi_V9_return_struct): New function.
3415 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3416 32bit code (not yet cross-arch calls).
3417 * src/sparc/v8.S: Add struct return delay nop.
3419 * src/sparc/v9.S: New file.
3420 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3421 is used on sparc64 only for structures larger than 32 bytes.
3422 Pass by reference for structures is done for structure arguments
3423 larger than 16 bytes.
3424 * src/ffitest.c (main): Use 64bit rint on sparc64.
3425 Run long long tests on sparc.
3426 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3428 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3430 * configure.in (sparc-*-linux*): New supported target.
3431 (sparc64-*-linux*): Likewise.
3432 * configure: Rebuilt.
3433 * Makefile.am: Add v9.S to SPARC files.
3434 * Makefile.in: Likewise.
3435 (LINK): Surround $(CCLD) into double quotes, so that multilib
3436 compiles work correctly.
3438 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3440 * configure: Rebuilt.
3441 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3442 proposed by Bryce McKinlay.
3444 2000-03-20 Tom Tromey <tromey@cygnus.com>
3446 * Makefile.in: Hand edit for java_raw_api.lo.
3448 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3450 * config.guess, config.sub: Update from the gcc tree.
3451 Fix for PR libgcj/168.
3453 2000-03-03 Tom Tromey <tromey@cygnus.com>
3455 * Makefile.in: Fixed ia64 by hand.
3457 * configure: Rebuilt.
3458 * configure.in (--enable-multilib): New option.
3459 (libffi_basedir): New subst.
3460 (AC_OUTPUT): Added multilib code.
3462 2000-03-02 Tom Tromey <tromey@cygnus.com>
3464 * Makefile.in: Rebuilt.
3465 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3468 2000-02-25 Hans Boehm <boehm@acm.org>
3470 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3472 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3474 (ffi_prep_raw_closure): Use ffi_translate_args, not
3475 ffi_closure_translate.
3476 * src/java_raw_api.c: New file.
3477 * src/ffitest.c (closure_test_fn): New function.
3478 (main): Define `rint' as long long on IA64. Added new test when
3479 FFI_CLOSURES is defined.
3480 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3481 (ffi_abi): Recognize IA64.
3482 (ffi_raw): Added `flt' field.
3483 Added "Java raw API" code.
3484 * configure.in: Recognize ia64.
3485 * Makefile.am (TARGET_SRC_IA64): New macro.
3486 (libffi_la_common_SOURCES): Added java_raw_api.c.
3487 (libffi_la_SOURCES): Define in IA64 case.
3489 2000-01-04 Tom Tromey <tromey@cygnus.com>
3491 * Makefile.in: Rebuilt with newer automake.
3493 1999-12-31 Tom Tromey <tromey@cygnus.com>
3495 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3497 1999-09-01 Tom Tromey <tromey@cygnus.com>
3499 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3501 * fficonfig.h.in: Rebuilt.
3502 * configure: Rebuilt.
3503 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3504 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3505 * acconfig.h: Don't #undef PACKAGE or VERSION.
3507 1999-08-09 Anthony Green <green@cygnus.com>
3509 * include/ffi.h.in: Try to work around messy header problem
3510 with PACKAGE and VERSION.
3512 * configure: Rebuilt.
3513 * configure.in: Change version to 2.00-beta.
3515 * fficonfig.h.in: Rebuilt.
3516 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3518 * src/x86/ffi.c (ffi_raw_call): Rename.
3520 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3522 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3523 (ffi_prep_incoming_args_SYSV): Ditto.
3524 (ffi_prep_closure): Ditto.
3525 (ffi_closure_raw_SYSV): Ditto.
3526 (ffi_prep_raw_closure): More ditto.
3527 (ffi_call_raw): Final ditto.
3529 * include/ffi.h.in: Add definitions for closure and raw API.
3531 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3534 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3536 * src/raw_api.c: New file.
3538 * include/ffi.h.in (ffi_raw): New type.
3539 (UINT_ARG, SINT_ARG): New defines.
3540 (ffi_closure, ffi_raw_closure): New types.
3541 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3543 * configure.in: Add check for endianness and sizeof void*.
3545 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3546 instead of directly.
3548 * configure: Rebuilt.
3550 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3552 * configure.in: Add x86 and powerpc BeOS configurations.
3553 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3555 1999-05-09 Anthony Green <green@cygnus.com>
3557 * configure.in: Add warning about this being beta code.
3558 Remove src/Makefile.am from the picture.
3559 * configure: Rebuilt.
3561 * Makefile.am: Move logic from src/Makefile.am. Add changes
3562 to support libffi as a target library.
3563 * Makefile.in: Rebuilt.
3565 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3566 Upgraded to new autoconf, automake, libtool.
3570 * LICENSE: Update copyright date.
3572 * src/Makefile.am, src/Makefile.in: Removed.
3574 1998-11-29 Anthony Green <green@cygnus.com>
3576 * include/ChangeLog: Removed.
3577 * src/ChangeLog: Removed.
3578 * src/mips/ChangeLog: Removed.
3579 * src/sparc/ChangeLog: Remboved.
3580 * src/x86/ChangeLog: Removed.
3582 * ChangeLog.v1: Created.