]> git.ipfire.org Git - thirdparty/gcc.git/blame - libffi/ChangeLog
projects / thirdparty / gcc.git / blame
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | blame (incremental) | history | HEAD
2009-06-04 Andrew Haley <aph@redhat.com>
[thirdparty/gcc.git] / libffi / ChangeLog
CommitLineData
6e59f66f 12009-06-04 Andrew Haley <aph@redhat.com>
2
3 * src/powerpc/ppc_closure.S: Insert licence header.
4 * src/powerpc/linux64_closure.S: Likewise.
5 * src/m68k/sysv.S: Likewise.
6
7 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
8 * src/powerpc/ffi.c: Likewise.
9 * src/powerpc/ffi_darwin.c: Likewise.
10 * src/m32r/ffi.c: Likewise.
11 * src/sh64/ffi.c: Likewise.
12 * src/x86/ffi64.c: Likewise.
13 * src/alpha/ffi.c: Likewise.
14 * src/alpha/osf.S: Likewise.
15 * src/frv/ffi.c: Likewise.
16 * src/s390/ffi.c: Likewise.
17 * src/pa/ffi.c: Likewise.
18 * src/pa/hpux32.S: Likewise.
19 * src/ia64/unix.S: Likewise.
20 * src/ia64/ffi.c: Likewise.
21 * src/sparc/ffi.c: Likewise.
22 * src/mips/ffi.c: Likewise.
23 * src/sh/ffi.c: Likewise.
24
252008-02-15 David Daney <ddaney@avtrex.com>
26
27 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
28 Define (conditionally), and use it to include cachectl.h.
29 (ffi_prep_closure_loc): Fix cache flushing.
30 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
31
9d98ec1b 322009-06-04 Andrew Haley <aph@redhat.com>
33
34 include/ffi.h.in,
35 src/arm/ffitarget.h,
36 src/arm/ffi.c,
37 src/arm/sysv.S,
38 src/powerpc/ffitarget.h,
39 src/closures.c,
40 src/sh64/ffitarget.h,
41 src/sh64/ffi.c,
42 src/sh64/sysv.S,
43 src/types.c,
44 src/x86/ffi64.c,
45 src/x86/ffitarget.h,
46 src/x86/win32.S,
47 src/x86/darwin.S,
48 src/x86/ffi.c,
49 src/x86/sysv.S,
50 src/x86/unix64.S,
51 src/alpha/ffitarget.h,
52 src/alpha/ffi.c,
53 src/alpha/osf.S,
54 src/m68k/ffitarget.h,
55 src/frv/ffitarget.h,
56 src/frv/ffi.c,
57 src/s390/ffitarget.h,
58 src/s390/sysv.S,
59 src/cris/ffitarget.h,
60 src/pa/linux.S,
61 src/pa/ffitarget.h,
62 src/pa/ffi.c,
63 src/raw_api.c,
64 src/ia64/ffitarget.h,
65 src/ia64/unix.S,
66 src/ia64/ffi.c,
67 src/ia64/ia64_flags.h,
68 src/java_raw_api.c,
69 src/debug.c,
70 src/sparc/v9.S,
71 src/sparc/ffitarget.h,
72 src/sparc/ffi.c,
73 src/sparc/v8.S,
74 src/mips/ffitarget.h,
75 src/mips/n32.S,
76 src/mips/o32.S,
77 src/mips/ffi.c,
78 src/prep_cif.c,
79 src/sh/ffitarget.h,
80 src/sh/ffi.c,
81 src/sh/sysv.S: Update license text.
82
ad81e87b 832009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
84
85 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
86 (.eh_frame): Add FDE for it.
87
1cde303d 882009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
89
90 * configure.ac: Also check if assembler supports pc-relative
91 relocs on X86_WIN32 targets.
92 * configure: Regenerate.
93 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
94 (_ffi_call_SYSV): Add missing function type symbol .def and
95 add EH markup labels.
96 (_ffi_call_STDCALL): Likewise.
97 (_ffi_closure_SYSV): Likewise.
98 (_ffi_closure_raw_SYSV): Likewise.
99 (.eh_frame): Add hand-crafted EH data.
100
6bc9506f 1012009-04-09 Jakub Jelinek <jakub@redhat.com>
102
103 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
104 version 3 of the GNU General Public License and to point readers
105 at the COPYING3 file and the FSF's license web page.
106 * testsuite/libffi.call/call.exp: Likewise.
107 * testsuite/libffi.special/special.exp: Likewise.
108
9af964d4 1092009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
110
111 * configure: Regenerate.
112
95ff4d7d 1132008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
114
115 PR libffi/26048
116 * configure.ac (HAVE_AS_X86_PCREL): New test.
117 * configure: Regenerate.
118 * fficonfig.h.in: Regenerate.
119 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
120 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
121 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
122 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
123 * src/x86/unix64.S (.Lstore_table): Move to .text section.
124 (.Lload_table): Likewise.
125 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
126
9e6fd619 1272008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
128
129 * configure: Regenerate.
130
95c3349b 1312008-11-21 Eric Botcazou <ebotcazou@adacore.com>
132
133 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
134 signed/unsigned int8/16 return values.
135 * src/sparc/v8.S (ffi_call_v8): Likewise.
136 (ffi_closure_v8): Likewise.
137
4bc2b84f 1382008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
139 Steve Ellcey <sje@cup.hp.com>
140
141 * configure: Regenerate for new libtool.
142 * Makefile.in: Ditto.
143 * include/Makefile.in: Ditto.
144 * aclocal.m4: Ditto.
145
a2c1e278 1462008-08-25 Andreas Tobler <a.tobler@schweiz.org>
147
148 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
149 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
150 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
151 Adjust copyright notice.
152 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
153 register or two register to use for FFI_SYSV structs.
154 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
155 (ffi_closure_helper_SYSV): Fix the return type for
156 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
157 Adjust copyright notice.
158
0733c867 1592008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
160
161 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
162 int.
163
5b2c7553 1642008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
165
166 * configure: Regenerate.
167 * include/Makefile.in: Regenerate.
168 * testsuite/Makefile.in: Regenerate.
169
6af9f7ea 1702008-06-07 Joseph Myers <joseph@codesourcery.com>
171
172 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
173 powerpc-*-beos*): Remove.
174 * configure: Regenerate.
175
ac051731 1762008-05-09 Julian Brown <julian@codesourcery.com>
177
178 * Makefile.am (LTLDFLAGS): New.
179 (libffi_la_LDFLAGS): Use above.
180 * Makefile.in: Regenerate.
181
6a88da51 1822008-04-18 Paolo Bonzini <bonzini@gnu.org>
183
184 PR bootstrap/35457
185 * aclocal.m4: Regenerate.
186 * configure: Regenerate.
187
644af649 1882008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
189
190 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
191 * src/sh64/sysv.S: Likewise.
192
595c2b68 1932008-03-26 Daniel Jacobowitz <dan@debian.org>
194
195 * src/arm/sysv.S: Fix ARM comment marker.
196
1972008-03-26 Jakub Jelinek <jakub@redhat.com>
198
199 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
200 * src/s390/sysv.S: Likewise.
201 * src/powerpc/ppc_closure.S: Likewise.
202 * src/powerpc/sysv.S: Likewise.
203 * src/x86/unix64.S: Likewise.
204 * src/x86/sysv.S: Likewise.
205 * src/sparc/v8.S: Likewise.
206 * src/sparc/v9.S: Likewise.
207 * src/m68k/sysv.S: Likewise.
208 * src/arm/sysv.S: Likewise.
209
82eaf560 2102008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
211
212 * aclocal.m4: Regenerate.
213 * configure: Likewise.
214 * Makefile.in: Likewise.
215 * include/Makefile.in: Likewise.
216 * testsuite/Makefile.in: Likewise.
217
55e813ce 2182008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
219 Andreas Tobler <a.tobler@schweiz.org>
220
221 * configure.ac: Add amd64-*-freebsd* target.
222 * configure: Regenerate.
223
00eae123 2242008-01-30 H.J. Lu <hongjiu.lu@intel.com>
225
226 PR libffi/34612
227 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
228 returning struct.
229
230 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
231 tests.
232
ff3fb914 2332008-01-24 David Edelsohn <edelsohn@gnu.org>
234
235 * configure: Regenerate.
236
8f1f10d3 2372008-01-06 Andreas Tobler <a.tobler@schweiz.org>
238
239 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
240
ad73551f 2412008-01-05 Andreas Tobler <a.tobler@schweiz.org>
242
243 PR testsuite/32843
244 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
245 signed/unsigned int8/16 for X86_DARWIN.
246 Updated copyright info.
247 Handle one and two byte structs with special cif->flags.
248 * src/x86/ffitarget.h: Add special types for one and two byte structs.
249 Updated copyright info.
250 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
251 sysv.S
252 Remove code to pop args from the stack after call.
253 Special-case signed/unsigned for int8/16, one and two byte structs.
254 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
255 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
256 FFI_TYPE_SINT32.
257 Updated copyright info.
258
aaf16623 2592007-12-08 David Daney <ddaney@avtrex.com>
260
261 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
262 SUBU, add with ADDU and use smaller code sequences.
263
2c3f7870 2642007-12-07 David Daney <ddaney@avtrex.com>
2af77f57 265
2c3f7870 266 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
267 type.
268
5be4032c 2692007-12-06 David Daney <ddaney@avtrex.com>
270
271 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
272 defined.
273 (ffi_java_raw): New typedef.
274 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
275 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
276 ffi_java_raw.
277 (ffi_java_raw_closure) : Same.
278 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
279 parameter types.
280 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
281 FFI_SIZEOF_JAVA_RAW.
282 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
283 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
284 sizeof(ffi_java_raw) for alignment calculations.
285 (ffi_java_ptrarray_to_raw): Same.
286 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
287 if FFI_SIZEOF_JAVA_RAW == 4.
288 (ffi_java_raw_to_rvalue): Same.
289 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
290 (ffi_java_translate_args): Same.
291 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
292 parameter types.
293 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
294
79369c38 2952007-12-06 David Daney <ddaney@avtrex.com>
296
297 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
298 pointer values.
299
7cf0754c 3002007-12-01 Andreas Tobler <a.tobler@schweiz.org>
301
302 PR libffi/31937
303 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
304 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
305 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
306 set the NUM_FPR_ARG_REGISTERS according to.
307 Add support for potential soft-float support under hard-float
308 architecture.
309 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
310 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
311 to the FFI_LINUX_SOFT_FLOAT ABI.
312 (ffi_prep_cif_machdep): Likewise.
313 (ffi_closure_helper_SYSV): Likewise.
314 * src/powerpc/ppc_closure.S: Make sure not to store float/double
315 on archs where __NO_FPRS__ is true.
316 Add FFI_TYPE_UINT128 support.
317 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
318 Adjust copyright notice.
319
5036bf25 3202007-11-25 Andreas Tobler <a.tobler@schweiz.org>
321
322 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
323 * include/ffi_common.h: ... here.
324 Update copyright.
325
c3f75f11 3262007-11-17 Andreas Tobler <a.tobler@schweiz.org>
327
328 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
329 * src/powerpc/linux64.S: Likewise.
330 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
331 * testsuite/libffi.call/return_ldl.c: New test.
332
244c3f18 3332007-09-04 <aph@redhat.com>
334
335 * src/arm/sysv.S (UNWIND): New.
336 (Whole file): Conditionally compile unwinder directives.
337 * src/arm/sysv.S: Add unwinder directives.
338
339 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
340 Only treat r0 as a struct address if we're actually returning a
341 struct by address.
342 Only copy the bytes that are actually within a struct.
343 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
344 is returned in r0, not passed by address.
345 (ffi_call): Allocate a word-sized temporary for the case where
346 a composite is returned in r0.
347 (ffi_prep_incoming_args_SYSV): Align as necessary.
348
3492007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
350
351 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
352 directly using the sys_cacheflush syscall.
353
3542007-07-27 Andrew Haley <aph@redhat.com>
355
356 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
357
7a659e46 3582007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
359
360 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
361 * configure.ac: Likewise.
362 * Makefile.in: Regenerate.
363 * include/Makefile.in: Likewise.
364 * testsuite/Makefile.in: Likewise.
365 * configure: Likewise.
366
99b65499 3672007-08-24 David Daney <ddaney@avtrex.com>
368
369 * testsuite/libffi.call/return_sl.c: New test.
370
b5c438f4 3712007-08-10 David Daney <ddaney@avtrex.com>
372
373 * testsuite/libffi.call/cls_multi_ushort.c,
374 testsuite/libffi.call/cls_align_uint16.c,
375 testsuite/libffi.call/nested_struct1.c,
376 testsuite/libffi.call/nested_struct3.c,
377 testsuite/libffi.call/cls_7_1_byte.c,
378 testsuite/libffi.call/nested_struct5.c,
379 testsuite/libffi.call/cls_double.c,
380 testsuite/libffi.call/nested_struct7.c,
381 testsuite/libffi.call/cls_sint.c,
382 testsuite/libffi.call/nested_struct9.c,
383 testsuite/libffi.call/cls_20byte1.c,
384 testsuite/libffi.call/cls_multi_sshortchar.c,
385 testsuite/libffi.call/cls_align_sint64.c,
386 testsuite/libffi.call/cls_3byte2.c,
387 testsuite/libffi.call/cls_multi_schar.c,
388 testsuite/libffi.call/cls_multi_uchar.c,
389 testsuite/libffi.call/cls_19byte.c,
390 testsuite/libffi.call/cls_9byte1.c,
391 testsuite/libffi.call/cls_align_float.c,
392 testsuite/libffi.call/closure_fn1.c,
393 testsuite/libffi.call/problem1.c,
394 testsuite/libffi.call/closure_fn3.c,
395 testsuite/libffi.call/cls_sshort.c,
396 testsuite/libffi.call/closure_fn5.c,
397 testsuite/libffi.call/cls_align_double.c,
398 testsuite/libffi.call/nested_struct.c,
399 testsuite/libffi.call/cls_2byte.c,
400 testsuite/libffi.call/nested_struct10.c,
401 testsuite/libffi.call/cls_4byte.c,
402 testsuite/libffi.call/cls_6byte.c,
403 testsuite/libffi.call/cls_8byte.c,
404 testsuite/libffi.call/cls_multi_sshort.c,
405 testsuite/libffi.call/cls_align_sint16.c,
406 testsuite/libffi.call/cls_align_uint32.c,
407 testsuite/libffi.call/cls_20byte.c,
408 testsuite/libffi.call/cls_float.c,
409 testsuite/libffi.call/nested_struct2.c,
410 testsuite/libffi.call/cls_5_1_byte.c,
411 testsuite/libffi.call/nested_struct4.c,
412 testsuite/libffi.call/cls_24byte.c,
413 testsuite/libffi.call/nested_struct6.c,
414 testsuite/libffi.call/cls_64byte.c,
415 testsuite/libffi.call/nested_struct8.c,
416 testsuite/libffi.call/cls_uint.c,
417 testsuite/libffi.call/cls_multi_ushortchar.c,
418 testsuite/libffi.call/cls_schar.c,
419 testsuite/libffi.call/cls_uchar.c,
420 testsuite/libffi.call/cls_align_uint64.c,
421 testsuite/libffi.call/cls_ulonglong.c,
422 testsuite/libffi.call/cls_align_longdouble.c,
423 testsuite/libffi.call/cls_1_1byte.c,
424 testsuite/libffi.call/cls_12byte.c,
425 testsuite/libffi.call/cls_3_1byte.c,
426 testsuite/libffi.call/cls_3byte1.c,
427 testsuite/libffi.call/cls_4_1byte.c,
428 testsuite/libffi.call/cls_6_1_byte.c,
429 testsuite/libffi.call/cls_16byte.c,
430 testsuite/libffi.call/cls_18byte.c,
431 testsuite/libffi.call/closure_fn0.c,
432 testsuite/libffi.call/cls_9byte2.c,
433 testsuite/libffi.call/closure_fn2.c,
434 testsuite/libffi.call/closure_fn4.c,
435 testsuite/libffi.call/cls_ushort.c,
436 testsuite/libffi.call/closure_fn6.c,
437 testsuite/libffi.call/cls_5byte.c,
438 testsuite/libffi.call/cls_align_pointer.c,
439 testsuite/libffi.call/cls_7byte.c,
440 testsuite/libffi.call/cls_align_sint32.c,
441 testsuite/libffi.special/unwindtest_ffi_call.cc,
442 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
443
edd57a0a 4442007-08-10 David Daney <ddaney@avtrex.com>
445
446 PR libffi/28313
447 * configure.ac: Don't treat mips64 as a special case.
448 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
449 * configure: Regenerate
450 * Makefile.in: Ditto.
451 * fficonfig.h.in: Ditto.
452 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
453 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
454 (FFI_DEFAULT_ABI): Set for n64 case.
455 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
456 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
457 (ffi_closure_N32): New function.
458 (.eh_frame): New section
459 * src/mips/o32.S: Clean up comments.
460 (ffi_closure_O32): Pass ffi_closure parameter in $12.
461 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
462 _MIPS_SIM == _ABIN32 throughout.
463 (FFI_MIPS_STOP_HERE): New, use in place of
464 ffi_stop_here.
465 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
466 to support n32/n64 ABIs.
467 (calc_n32_struct_flags): Rewrite.
468 (calc_n32_return_struct_flags): Remove unused variable. Reverse
469 position of flag bits.
470 (ffi_prep_cif_machdep): Rewrite n32 portion.
471 (ffi_call): Enable for n64. Add special handling for small structure
472 return values.
473 (ffi_prep_closure_loc): Add n32 and n64 support.
474 (ffi_closure_mips_inner_O32): Add cast to silence warning.
475 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
476
3471c5eb 4772007-08-08 David Daney <ddaney@avtrex.com>
478
479 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
480 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
481 specifiers.
482 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
483 * testsuite/libffi.call/cls_sint.c (main): Ditto.
484 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
485 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
486 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
487 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
488 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
489 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
490 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
491 * testsuite/libffi.call/return_fl3.c (main): Ditto.
492 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
493 * testsuite/libffi.call/nested_struct.c (main): Ditto.
494 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
495 * testsuite/libffi.call/return_ll1.c (main): Ditto.
496 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
497 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
498 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
499 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
500 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
501 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
502 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
503 * testsuite/libffi.call/cls_uint.c (main): Ditto.
504 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
505 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
506 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
507 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
508 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
509 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
510 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
511 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
512 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
513
ad7bf8e7 5142007-08-07 Andrew Haley <aph@redhat.com>
515
516 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
517 checkin.
518
6cc1286e 5192007-08-06 Andrew Haley <aph@redhat.com>
520
521 PR testsuite/32843
522 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
523 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
524 FFI_TYPE_SINT32.
525
c6f5483d 5262007-08-02 David Daney <ddaney@avtrex.com>
527
528 * testsuite/libffi.call/return_ul.c (main): Define return type as
529 ffi_arg. Use proper printf conversion specifier.
530
f6179f0e 5312007-07-30 Andrew Haley <aph@redhat.com>
532
533 PR testsuite/32843
534 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
535 signed/unsigned int8/16.
536 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
537 Use a jump table.
538 Remove code to pop args from the stack after call.
539 Special-case signed/unsigned int8/16.
540 * testsuite/libffi.call/return_sc.c (main): Revert.
541
7babed79 5422007-07-26 Richard Guenther <rguenther@suse.de>
543
544 PR testsuite/32843
545 * testsuite/libffi.call/return_sc.c (main): Verify call
546 result as signed char, not ffi_arg.
547
f54a4242 5482007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
549
550 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
551 * configure: Regenerate.
552
e96a69da 5532007-07-11 David Daney <ddaney@avtrex.com>
554
555 * src/mips/ffi.c: Don't include sys/cachectl.h.
556 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
557 cacheflush().
558
187f6543 5592007-05-18 Aurelien Jarno <aurelien@aurel32.net>
560
561 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
562 from (ffi_prep_closure): ... this.
563 (FFI_INIT_TRAMPOLINE): Adjust.
564
5652005-12-31 Phil Blundell <pb@reciva.com>
566
567 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
568 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
569 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
570 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
571 (FFI_CLOSURES): Enable closure support.
572
5732007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
574
575 * testsuite/libffi.call/cls_multi_ushort.c,
576 testsuite/libffi.call/cls_align_uint16.c,
577 testsuite/libffi.call/nested_struct1.c,
578 testsuite/libffi.call/nested_struct3.c,
579 testsuite/libffi.call/cls_7_1_byte.c,
580 testsuite/libffi.call/cls_double.c,
581 testsuite/libffi.call/nested_struct5.c,
582 testsuite/libffi.call/nested_struct7.c,
583 testsuite/libffi.call/cls_sint.c,
584 testsuite/libffi.call/nested_struct9.c,
585 testsuite/libffi.call/cls_20byte1.c,
586 testsuite/libffi.call/cls_multi_sshortchar.c,
587 testsuite/libffi.call/cls_align_sint64.c,
588 testsuite/libffi.call/cls_3byte2.c,
589 testsuite/libffi.call/cls_multi_schar.c,
590 testsuite/libffi.call/cls_multi_uchar.c,
591 testsuite/libffi.call/cls_19byte.c,
592 testsuite/libffi.call/cls_9byte1.c,
593 testsuite/libffi.call/cls_align_float.c,
594 testsuite/libffi.call/closure_fn1.c,
595 testsuite/libffi.call/problem1.c,
596 testsuite/libffi.call/closure_fn3.c,
597 testsuite/libffi.call/cls_sshort.c,
598 testsuite/libffi.call/closure_fn5.c,
599 testsuite/libffi.call/cls_align_double.c,
600 testsuite/libffi.call/cls_2byte.c,
601 testsuite/libffi.call/nested_struct.c,
602 testsuite/libffi.call/nested_struct10.c,
603 testsuite/libffi.call/cls_4byte.c,
604 testsuite/libffi.call/cls_6byte.c,
605 testsuite/libffi.call/cls_8byte.c,
606 testsuite/libffi.call/cls_multi_sshort.c,
607 testsuite/libffi.call/cls_align_uint32.c,
608 testsuite/libffi.call/cls_align_sint16.c,
609 testsuite/libffi.call/cls_float.c,
610 testsuite/libffi.call/cls_20byte.c,
611 testsuite/libffi.call/cls_5_1_byte.c,
612 testsuite/libffi.call/nested_struct2.c,
613 testsuite/libffi.call/cls_24byte.c,
614 testsuite/libffi.call/nested_struct4.c,
615 testsuite/libffi.call/nested_struct6.c,
616 testsuite/libffi.call/cls_64byte.c,
617 testsuite/libffi.call/nested_struct8.c,
618 testsuite/libffi.call/cls_uint.c,
619 testsuite/libffi.call/cls_multi_ushortchar.c,
620 testsuite/libffi.call/cls_schar.c,
621 testsuite/libffi.call/cls_uchar.c,
622 testsuite/libffi.call/cls_align_uint64.c,
623 testsuite/libffi.call/cls_ulonglong.c,
624 testsuite/libffi.call/cls_align_longdouble.c,
625 testsuite/libffi.call/cls_1_1byte.c,
626 testsuite/libffi.call/cls_12byte.c,
627 testsuite/libffi.call/cls_3_1byte.c,
628 testsuite/libffi.call/cls_3byte1.c,
629 testsuite/libffi.call/cls_4_1byte.c,
630 testsuite/libffi.call/cls_6_1_byte.c,
631 testsuite/libffi.call/cls_16byte.c,
632 testsuite/libffi.call/cls_18byte.c,
633 testsuite/libffi.call/closure_fn0.c,
634 testsuite/libffi.call/cls_9byte2.c,
635 testsuite/libffi.call/closure_fn2.c,
636 testsuite/libffi.call/closure_fn4.c,
637 testsuite/libffi.call/cls_ushort.c,
638 testsuite/libffi.call/closure_fn6.c,
639 testsuite/libffi.call/cls_5byte.c,
640 testsuite/libffi.call/cls_align_pointer.c,
641 testsuite/libffi.call/cls_7byte.c,
642 testsuite/libffi.call/cls_align_sint32.c,
643 testsuite/libffi.special/unwindtest_ffi_call.cc,
644 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
645
cecc75a9 6462007-07-05 H.J. Lu <hongjiu.lu@intel.com>
647
648 * aclocal.m4: Regenerated.
649
07c46636 6502007-06-02 Paolo Bonzini <bonzini@gnu.org>
651
652 * configure: Regenerate.
653
542f9784 6542007-05-23 Steve Ellcey <sje@cup.hp.com>
655
656 * Makefile.in: Regenerate.
657 * configure: Regenerate.
658 * aclocal.m4: Regenerate.
659 * include/Makefile.in: Regenerate.
660 * testsuite/Makefile.in: Regenerate.
661
0e3a3bdb 6622007-05-10 Roman Zippel <zippel@linux-m68k.org>
663
664 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
665 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
666 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
667 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
668 (FFI_CLOSURES): Enable closure support.
669
a53add80 6702007-05-10 Roman Zippel <zippel@linux-m68k.org>
671
672 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
673 * configure: Regenerate.
674 * fficonfig.h.in: Regenerate.
675 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
676 CFI_OFFSET,CFI_DEF_CFA): New macros.
677 (ffi_call_SYSV): Add callframe annotation.
678
bb3218e9 6792007-05-10 Roman Zippel <zippel@linux-m68k.org>
680
681 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
682 numerous test suite failures.
683 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
684
5d0df87f 6852007-04-11 Paolo Bonzini <bonzini@gnu.org>
686
687 * Makefile.am (EXTRA_DIST): Bring up to date.
688 * Makefile.in: Regenerate.
689 * src/frv/eabi.S: Remove RCS keyword.
690
ea4d299a 6912007-04-06 Richard Henderson <rth@redhat.com>
692
693 * configure.ac: Tidy target case.
694 (HAVE_LONG_DOUBLE): Allow the target to override.
695 * configure: Regenerate.
696 * include/ffi.h.in: Don't define ffi_type_foo if
697 LIBFFI_HIDE_BASIC_TYPES is defined.
698 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
699 to ffi_type_double.
700 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
701 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
702 (ffi_type_longdouble): Special case for Alpha. Don't define
703 if long double == double.
704
705 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
706 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
707 (ffi_call, ffi_closure_osf_inner): Likewise.
708 (ffi_closure_osf_inner): Likewise. Mark hidden.
709 (ffi_call_osf, ffi_closure_osf): Mark hidden.
710 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
711 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
712 (load_table): Handle 128-bit long double.
713
714 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
715
770e494d 7162007-04-06 Tom Tromey <tromey@redhat.com>
717
718 PR libffi/31491:
719 * README: Fixed bug in example.
720
58606959 7212007-04-03 Jakub Jelinek <jakub@redhat.com>
722
723 * src/closures.c: Include sys/statfs.h.
724 (_GNU_SOURCE): Define on Linux.
725 (FFI_MMAP_EXEC_SELINUX): Define.
726 (selinux_enabled): New variable.
727 (selinux_enabled_check): New function.
728 (is_selinux_enabled): Define.
729 (dlmmap): Use it.
730
7fed151e 7312007-03-24 Uros Bizjak <ubizjak@gmail.com>
732
733 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
734 Use 'volatile float sum' to create sum of floats to avoid false
735 negative due to excess precision on ix86 targets.
736 (main): Ditto.
737
8fb4628d 7382007-03-08 Alexandre Oliva <aoliva@redhat.com>
739
740 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
741 patch.
742 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
743
f20fe604 7442007-03-07 Alexandre Oliva <aoliva@redhat.com>
745
746 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
747 (ffi_prep_closure_loc): New.
748 (ffi_prep_raw_closure_loc): New.
749 (ffi_prep_java_raw_closure_loc): New.
750 * src/closures.c: New file.
751 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
752 Replace sflags with exec_offset.
753 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
754 sub_segment_exec_offset): New macros.
755 (get_segment_flags, set_segment_flags, check_segment_merge): New
756 macros.
757 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
758 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
759 destroy_mspace): Use new macros.
760 (sys_alloc): Silence warning.
761 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
762 * Makefile.in: Rebuilt.
763 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
764 terms of ffi_prep_closure_loc.
765 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
766 from...
767 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
768 renamed version.
769 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
770 adjusted from...
771 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
772 the renamed version.
773 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
774 (ffi_prep_closure): ... this.
775 * src/pa/ffi.c: Likewise.
776 * src/cris/ffi.c: Likewise. Adjust.
777 * src/frv/ffi.c: Likewise.
778 * src/ia64/ffi.c: Likewise.
779 * src/mips/ffi.c: Likewise.
780 * src/powerpc/ffi_darwin.c: Likewise.
781 * src/s390/ffi.c: Likewise.
782 * src/sh/ffi.c: Likewise.
783 * src/sh64/ffi.c: Likewise.
784 * src/sparc/ffi.c: Likewise.
785 * src/x86/ffi64.c: Likewise.
786 * src/x86/ffi.c: Likewise.
787 (FFI_INIT_TRAMPOLINE): Adjust.
788 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
789 (ffi_prep_raw_closure): ... this.
790 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
791 (ffi_prep_closure): ... this.
792 (flush_icache): Adjust.
793
c3265eac 7942007-03-07 Alexandre Oliva <aoliva@redhat.com>
795
796 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
797 Lea's malloc.
798
0ce4a5ce 7992007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
800
801 * Makefile.am: Add dummy install-pdf target.
802 * Makefile.in: Regenerate
803
45444175 8042007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
805
806 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
807 ffi_closure_helper_SYSV): Add long double handling.
808
1df875cc 8092007-02-02 Jakub Jelinek <jakub@redhat.com>
810
811 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
812 immediately after bctrl instruction.
813
27c4194f 8142007-01-18 Alexandre Oliva <aoliva@redhat.com>
815
816 * Makefile.am (all-recursive, install-recursive,
817 mostlyclean-recursive, clean-recursive, distclean-recursive,
818 maintainer-clean-recursive): Add missing targets.
819 * Makefile.in: Rebuilt.
820
eb6316a2 8212006-12-14 Andreas Tobler <a.tobler@schweiz.org>
822
823 * configure.ac: Add TARGET for x86_64-*-darwin*.
824 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
825 for X86_DARWIN.
826 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
827 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
828 * configure: Regenerate.
829 * Makefile.in: Regenerate.
830 * include/Makefile.in: Regenerate.
831 * testsuite/Makefile.in: Regenerate.
832 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
833 ffi_call only.
834
75a99c5e 8352006-12-13 Andreas Tobler <a.tobler@schweiz.org>
836
837 * aclocal.m4: Regenerate with aclocal -I .. as written in the
838 Makefile.am.
839
a39e926c 8402006-10-31 Geoffrey Keating <geoffk@apple.com>
c8dfb8ef 841
842 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
843 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
844 Darwin.
845 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
846 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
847
eb819ecd 8482006-10-10 Paolo Bonzini <bonzini@gnu.org>
849 Sandro Tolaini <tolaini@libero.it>
850
851 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
852 conditional.
853 * configure: Regenerated.
854 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
855 (EXTRA_DIST): Add src/x86/darwin.S.
856 * Makefile.in: Regenerated.
857 * include/Makefile.in: Regenerated.
858 * testsuite/Makefile.in: Regenerated.
859
860 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
861 X86_WIN32, and additionally align stack to 16 bytes.
862 * src/x86/darwin.S: New, based on sysv.S.
863 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
864
23d9d417 8652006-09-12 David Daney <ddaney@avtrex.com>
866
867 PR libffi/23935
868 * include/Makefile.am: Install both ffi.h and ffitarget.h in
869 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
870 * aclocal.m4: Regenerated for automake 1.9.6.
871 * Makefile.in: Regenerated.
872 * include/Makefile.in: Regenerated.
873 * testsuite/Makefile.in: Regenerated.
874
ec045626 8752006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
876
877 * include/ffi_common.h (struct): Revert accidental commit.
878
b9463909 8792006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
880
881 * include/ffi_common.h: Remove lint directives.
882 * include/ffi.h.in: Likewise.
883
c695c024 8842006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
885
886 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
887 for 32-bit architectures.
888 * testsuite/libffi.call/return_ul.c: New test case.
889
3a9d2a82 8902006-07-19 David Daney <ddaney@avtrex.com>
891
892 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
893 xfail remains for mips64.
894
8f16e3ad 8952006-05-23 Carlos O'Donell <carlos@codesourcery.com>
896
897 * Makefile.am: Add install-html target. Add install-html to .PHONY
898 * Makefile.in: Regenerate.
899 * aclocal.m4: Regenerate.
900 * include/Makefile.in: Regenerate.
901 * testsuite/Makefile.in: Regenerate.
902
db951627 9032006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
904
905 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
906 stack slot.
907
9a333c86 9082006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
909
910 * README: Remove notice about 'Crazy Comments'.
911 * src/debug.c: Remove lint directives. Cleanup white spaces.
912 * src/java_raw_api.c: Likewise.
913 * src/prep_cif.c: Likewise.
914 * src/raw_api.c: Likewise.
915 * src/ffitest.c: Delete. No longer needed, all test cases migrated
916 to the testsuite.
917 * src/arm/ffi.c: Remove lint directives.
918 * src/m32r/ffi.c: Likewise.
919 * src/pa/ffi.c: Likewise.
920 * src/powerpc/ffi.c: Likewise.
921 * src/powerpc/ffi_darwin.c: Likewise.
922 * src/sh/ffi.c: Likewise.
923 * src/sh64/ffi.c: Likewise.
924 * src/x86/ffi.c: Likewise.
925 * testsuite/libffi.call/float2.c: Likewise.
926 * testsuite/libffi.call/promotion.c: Likewise.
927 * testsuite/libffi.call/struct1.c: Likewise.
928
7610fe2b 9292006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
930
931 * src/pa/hpux32.S: Correct unwind offset calculation for
932 ffi_closure_pa32.
933 * src/pa/linux.S: Likewise.
934
5adc6320 9352006-04-12 James E Wilson <wilson@specifix.com>
936
937 PR libgcj/26483
938 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
939 (hfa_type_load): Call stf_spill.
940 (hfa_type_store): Call ldf_fill.
941 (ffi_call): Adjust calls to above routines. Add local temps for
942 macro result.
7610fe2b 943
d35f63a9 9442006-04-10 Matthias Klose <doko@debian.org>
945
946 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
947 directory names containing underscores.
948
2d3dcfaa 9492006-04-07 James E Wilson <wilson@specifix.com>
950
951 * testsuite/libffi.call/float4.c: New testcase.
952
1a82d6f6 9532006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
954 Andreas Tobler <a.tobler@schweiz.ch>
955
956 * Makefile.am: Add PA_HPUX port.
957 * Makefile.in: Regenerate.
958 * include/Makefile.in: Likewise.
959 * testsuite/Makefile.in: Likewise.
960 * configure.ac: Add PA_HPUX rules.
961 * configure: Regenerate.
962 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
963 Add PA_HPUX and PA64_HPUX.
964 Rename FFI_LINUX ABI to FFI_PA32 ABI.
965 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
966 (FFI_TYPE_SMALL_STRUCT2): Define.
967 (FFI_TYPE_SMALL_STRUCT4): Likewise.
968 (FFI_TYPE_SMALL_STRUCT8): Likewise.
969 (FFI_TYPE_SMALL_STRUCT3): Redefine.
970 (FFI_TYPE_SMALL_STRUCT5): Likewise.
971 (FFI_TYPE_SMALL_STRUCT6): Likewise.
972 (FFI_TYPE_SMALL_STRUCT7): Likewise.
973 * src/pa/ffi.c (ROUND_DOWN): Delete.
974 (fldw, fstw, fldd, fstd): Use '__asm__'.
975 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
976 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
977 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
978 Simplify incrementing of stack slot variable. Change type of local
979 'n' to unsigned int.
980 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
981 double on PA_HPUX.
982 (ffi_prep_cif_machdep): Likewise.
983 (ffi_call): Likewise.
984 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
985 return type to ffi_status. Simplify incrementing of stack slot
986 variable. Only copy floating point argument registers when PA_LINUX
987 is true. Reformat debug statement.
988 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
989 FFI_TYPE_SMALL_STRUCT8.
990 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
991 declaration.
992 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
993 Add nops to cache flush. Add trampoline for PA_HPUX.
994 * src/pa/hpux32.S: New file.
995 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
996 ffi_prep_args_LINUX to ffi_prep_args_pa32.
997 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
998 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
999 argument type checks so that common argument types appear first.
1000 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1001 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1002
7f5ddc97 10032006-03-24 Alan Modra <amodra@bigpond.net.au>
1004
1005 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1006 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1007 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1008 FFI_TYPE_LONGDOUBLE.
1009 (ffi_prep_args64): Assert using IBM extended double.
1010 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1011 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1012 (ffi_call): Handle FFI_LINUX.
1013 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1014 gpr3 return pointer as for struct return. Handle FFI_LINUX
1015 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1016 unnecessarily.
1017 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1018 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1019 Don't use r6 as pointer to results, instead use sp offset. Don't
1020 make a special call to load lr with case table address, instead
1021 use offset from previous call.
1022 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1023 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1024 return.
1025
cd10a5a7 10262006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1027
1028 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1029 passed with FP registers correctly.
1030 (ffi_closure_helper_SYSV): Likewise.
1031 * src/sh64/sysv.S: Likewise.
1032
a07533b6 10332006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1034
1035 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1036 args and userdata unused.
1037 (closure_test_fn1): Mark cif and userdata unused.
1038 (main): Remove unused res.
1039
bdc6f488 10402006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1041
1042 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1043 -O2, -O3, -Os and the warning flags -W -Wall.
1044 * testsuite/libffi.special/special.exp: Likewise.
1045 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1046 unused parameter unused for gcc or else do nothing.
1047 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1048 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1049 and userdata unused.
1050 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1051 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1052 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1053 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1054 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1055 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1056 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1057 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1058 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1059 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1060 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1061 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1062 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1063 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1064 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1065 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1066 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1067 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1068 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1069 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1070 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1071 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1072 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1073 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1074 Likewise.
1075 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1076 Likewise.
1077 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1078 Likewise.
1079 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1080 void* to avoid compiler warning.
1081 (main): Likewise.
1082 (cls_struct_align_gn): Mark cif and userdata unused.
1083 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1084 Likewise.
1085 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1086 Likewise.
1087 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1088 Likewise.
1089 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1090 Likewise.
1091 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1092 Likewise.
1093 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1094 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1095 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1096 data unused.
1097 (main): Cast res_call to silence gcc.
1098 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1099 data unused.
1100 (main): Cast res_call to silence gcc.
1101 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1102 and data unused.
1103 (main): Cast res_call to silence gcc.
1104 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1105 data unused.
1106 (main): Cast res_call to silence gcc.
1107 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1108 data unused.
1109 (main): Cast res_call to silence gcc.
1110 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1111 and data unused.
1112 (main): Cast res_call to silence gcc.
1113 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1114 userdata unused.
1115 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1116 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1117 userdata unused.
1118 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1119 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1120 userdata unused.
1121 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1122 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1123 userdata unused.
1124 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1125 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1126 userdata unused.
1127 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1128 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1129 and userdata unused.
1130 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1131 userdata unused.
1132 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1133 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1134 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1135 Cleanup white spaces.
1136 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1137 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1138 cif and userdata unused.
1139 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1140 Likewise.
1141 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1142 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1143 formatters to silence gcc.
1144 (B_gn): Mark cif and userdata unused.
1145 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1146 unused.
1147 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1148 (B_gn): Mark cif and userdata unused.
1149 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1150 unused.
1151 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1152 (B_gn): Mark cif and userdata unused.
1153 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1154 unused.
1155 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1156 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1157 * testsuite/libffi.call/problem1.c (stub): Likewise.
1158 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1159 gcc.
1160 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1161 in the last commit for this test case in the test case itself.
1162 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1163 unused.
1164 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1165 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1166 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1167 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1168 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1169 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1170
5ab2e6e4 11712006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1172
1173 * src/sh/sysv.S: Fix register numbers in the FDE for
1174 ffi_closure_SYSV.
1175
849b9f91 11762006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1177
1178 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1179 declaration to avoid a false negative on ix86. See PR323.
1180
003f258f 11812006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1182
1183 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1184 and cast integer to void * if needed. Update the pointer to
1185 the FP register saved area correctly.
1186
9121151b 11872006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1188
1189 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1190 is fixed.
1191 * testsuite/libffi.call/nested_struct4.c: Likewise.
1192
019a3b64 11932006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1194
6cbfba8c 1195 * testsuite/libffi.call/return_dbl.c: New test case.
1196 * testsuite/libffi.call/return_dbl1.c: Likewise.
1197 * testsuite/libffi.call/return_dbl2.c: Likewise.
1198 * testsuite/libffi.call/return_fl.c: Likewise.
1199 * testsuite/libffi.call/return_fl1.c: Likewise.
1200 * testsuite/libffi.call/return_fl2.c: Likewise.
1201 * testsuite/libffi.call/return_fl3.c: Likewise.
1202 * testsuite/libffi.call/closure_fn6.c: Likewise.
1203
019a3b64 1204 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1205 definition.
1206 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1207 here to be used by other test cases too.
1208
1209 * testsuite/libffi.call/nested_struct10.c: New test case.
1210 * testsuite/libffi.call/nested_struct9.c: Likewise.
1211 * testsuite/libffi.call/nested_struct8.c: Likewise.
1212 * testsuite/libffi.call/nested_struct7.c: Likewise.
1213 * testsuite/libffi.call/nested_struct6.c: Likewise.
1214 * testsuite/libffi.call/nested_struct5.c: Likewise.
1215 * testsuite/libffi.call/nested_struct4.c: Likewise.
1216
946a8d62 12172006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1218
1219 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1220 * configure: Rebuilt.
1221
cb99d24e 12222006-01-18 Jakub Jelinek <jakub@redhat.com>
1223
1224 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1225 instead do the shifting inline.
1226 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1227 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1228 and 8 byte structs, for the remaining struct sizes don't call
1229 __lshrdi3, instead do the shifting inline.
1230
06bf2366 12312005-12-07 Thiemo Seufer <ths@networkno.de>
1232
1233 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1234 missing parentheses.
1235 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1236 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1237 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1238 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1239 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1240 FA_0_0_OFF2.
1241 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1242 endianness bugs.
1243 (ffi_prep_closure): Improve trampoline instruction scheduling.
1244 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1245
9a385484 12462005-12-03 Alan Modra <amodra@bigpond.net.au>
1247
1248 * src/powerpc/ffi.c: Formatting.
1249 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1250 (ffi_prep_args64): Likewise.
1251
193c2942 12522005-09-30 Geoffrey Keating <geoffk@apple.com>
1253
1254 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1255 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1256
d3eb23fb 12572005-09-26 Tom Tromey <tromey@redhat.com>
1258
1259 * testsuite/libffi.call/float1.c (value_type): New typedef.
1260 (CANARY): New define.
1261 (main): Check for result buffer overflow.
1262 * src/powerpc/linux64.S: Handle linux64 long double returns.
1263 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1264 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1265
5497bbdf 12662005-08-25 Alan Modra <amodra@bigpond.net.au>
1267
1268 PR target/23404
1269 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1270 homed fp args.
1271 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1272
33b4ca90 12732005-08-11 Jakub Jelinek <jakub@redhat.com>
1274
1275 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1276 (AH_BOTTOM): Add FFI_HIDDEN definition.
1277 * configure: Rebuilt.
1278 * fficonfig.h.in: Rebuilt.
1279 * src/powerpc/ffi.c (hidden): Remove.
1280 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1281 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1282 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1283 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1284 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1285 add FFI_HIDDEN to its prototype.
1286 (ffi_closure_SYSV_inner): New.
1287 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1288 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1289
967ec51b 12902005-08-10 Alfred M. Szmidt <ams@gnu.org>
1291
1292 PR libffi/21819:
1293 * configure: Rebuilt.
1294 * configure.ac: Handle i*86-*-gnu*.
1295
80207f81 12962005-08-09 Jakub Jelinek <jakub@redhat.com>
1297
1298 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1299 DW_CFA_offset_extended_sf rather than
1300 DW_CFA_GNU_negative_offset_extended.
1301 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1302
6a0e74dc 13032005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1304
1305 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1306 on sh3.
1307 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1308 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1309 partially on register.
1310 (ffi_closure_helper_SYSV): Likewise.
1311 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1312
cef3f52b 13132005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1314
1315 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1316 Remove empty line.
1317 * src/sh64/ffi.c (simple_type): Remove.
1318 (return_type): Handle small structures correctly.
1319 (ffi_prep_args): Likewise.
1320 (ffi_call): Likewise.
1321 (ffi_closure_helper_SYSV): Likewise.
1322 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1323 Emit position independent code if PIC and remove wrong datalabel
1324 prefixes from EH data.
1325
66c44239 13262005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1327
1328 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1329 * Makefile.in: Regenerate.
1330 * include/Makefile.in: Likewise.
1331 * testsuite/Makefile.in: Likewise.
1332 * configure.ac: Add POWERPC_FREEBSD rules.
1333 * configure: Regenerate.
1334 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1335 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1336 * src/powerpc/ffi.c: Add flags to handle small structure returns
1337 in ffi_call_SYSV.
1338 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1339 Aka FFI_SYSV.
1340 (ffi_closure_helper_SYSV): Likewise.
1341 * src/powerpc/ppc_closure.S: Add return types for small structures.
1342 * src/powerpc/sysv.S: Add bits to handle small structures for
1343 final SYSV 4 ABI.
1344
a21bd97f 13452005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1346
1347 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1348 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1349 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1350
c350327d 13512005-07-05 Randolph Chung <tausq@debian.org>
1352
1353 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1354 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1355 structures. Kill compilation warnings.
1356 (ffi_closure_inner_LINUX): Print return values as hex in debug
1357 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1358 Properly handle 5-7 byte structure returns.
1359 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1360 (FFI_TYPE_SMALL_STRUCT2): Remove.
1361 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1362 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1363 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1364 (checksmst1, checksmst2): Remove.
1365 (checksmst3): Optimize handling of 3-byte struct returns.
1366 (checksmst567): Properly handle 5-7 byte struct returns.
1367
20750e69 13682005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1369
1370 PR libgcj/21943
1371 * src/mips/n32.S: Enforce PIC code.
1372 * src/mips/o32.S: Likewise.
1373
58841df8 13742005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1375
1376 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1377 * configure: Regenerate.
1378
cc46fb30 13792005-06-01 Alan Modra <amodra@bigpond.net.au>
1380
1381 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1382 to call ffi_closure_helper_SYSV. Append @local instead.
1383 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1384
6e912c1e 13852005-05-17 Kelley Cook <kcook@gcc.gnu.org>
69738b31 1386
1387 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1388 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1389 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1390 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1391 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1392
5f0a087e 13932005-05-09 Mike Stump <mrs@apple.com>
1394
1395 * configure: Regenerate.
1396
81734468 13972005-05-08 Richard Henderson <rth@redhat.com>
1398
1399 PR libffi/21285
1400 * src/alpha/osf.S: Update unwind into to match code.
1401
4e9df534 14022005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1403 Richard Henderson <rth@redhat.com>
1404
1405 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1406 bit 11 of flags.
1407 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1408 (ffi_prep_closure): Set carry bit if sse-used flag set.
1409 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1410 Only load sse registers if ssecount non-zero.
1411 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1412
14132005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
6155bf60 1414
1415 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1416 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1417 * configure: Regenerate.
1418
643e4a19 14192005-04-20 Hans-Peter Nilsson <hp@axis.com>
1420
1421 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1422 have Tcl8.3-compatible intermediate variable.
1423
7e5823cf 14242005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1425 Hans-Peter Nilsson <hp@axis.com>
1426
1427 * Makefile.am: Add CRIS support.
1428 * configure.ac: Likewise.
942132a7 1429 * Makefile.in, configure, testsuite/Makefile.in,
1430 include/Makefile.in: Regenerate.
7e5823cf 1431 * src/cris: New directory.
1432 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1433 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
f0a2eecf 1434
1435 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1436 \r?\n in output tests.
1437
3bcf2cd5 14382005-04-12 Mike Stump <mrs@apple.com>
1439
1440 * configure: Regenerate.
1441
b7292ee2 14422005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1443
1444 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1445
3bdc9b0e 14462005-03-30 Steve Ellcey <sje@cup.hp.com>
1447
1448 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1449 (ffi_sarg) Ditto.
1450 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1451 to 64 bits in ILP32 mode.
1452 Load 64 bits even for short data.
1453
7abd8367 14542005-03-23 Mike Stump <mrs@apple.com>
1455
1456 * src/powerpc/darwin.S: Update for -m64 multilib.
1457 * src/powerpc/darwin_closure.S: Likewise.
1458
bbb07d35 14592005-03-21 Zack Weinberg <zack@codesourcery.com>
1460
1461 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1462 Do not set tool_include_dir.
1463 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1464 Regenerate.
1465 * include/Makefile.am: Set gcc_version and toollibffidir.
1466 * include/Makefile.in: Regenerate.
1467
e6fd470b 14682005-02-22 Andrew Haley <aph@redhat.com>
1469
1470 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1471 odd-numbered register pairs for 64-bit integer types.
1472
4f373876 14732005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1474
1475 PR libffi/20104
1476 * testsuite/libffi.call/return_ll1.c: New test case.
1477
c2cf854f 14782005-02-11 Janis Johnson <janis187@us.ibm.com>
1479
1480 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1481 * testsuite/libffi.call/float.c: Ditto.
1482 * testsuite/libffi.call/float2.c: Ditto.
1483 * testsuite/libffi.call/float3.c: Ditto.
1484
9e71f2c8 14852005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1486
1487 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1488
45e075b4 14892005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1490
1491 * testsuite/libffi.special/special.exp (cxx_options): Add
1492 -shared-libgcc.
1493
f587fca2 14942004-12-31 Richard Henderson <rth@redhat.com>
1495
1496 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1497 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1498 offset parameters with a type parameter; deduce size and structure
1499 alignment. Update all users.
1500
717c502e 15012004-12-31 Richard Henderson <rth@redhat.com>
1502
1503 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1504 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
bbb07d35 1505 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
717c502e 1506 into ffi_prep_closure.
1507 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1508 from scratch.
1509
4a60643d 15102004-12-27 Richard Henderson <rth@redhat.com>
1511
1512 * src/x86/unix64.S: Fix typo in unwind info.
1513
66d912b3 15142004-12-25 Richard Henderson <rth@redhat.com>
1515
1516 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1517 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1518 (merge_classes): Check for it.
1519 (SSE_CLASS_P): New.
1520 (classify_argument): Pass byte_offset by value; perform all updates
1521 inside struct case.
1522 (examine_argument): Add classes argument; handle
1523 X86_64_COMPLEX_X87_CLASS.
1524 (ffi_prep_args): Merge into ...
1525 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1526 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1527 (ffi_fill_return_value): Remove.
1528 (ffi_prep_closure): Remove dead assert.
1529 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1530 Rewrite to use struct register_args instead of va_list. Create
1531 flags for handling structure returns.
1532 * src/x86/unix64.S: Remove dead strings.
1533 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1534 stack frame with ffi_call. Handle structure returns properly.
1535 (float2sse, floatfloat2sse, double2sse): Remove.
1536 (sse2float, sse2double, sse2floatfloat): Remove.
1537 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1538 to handle structure returns properly.
1539
a8594486 15402004-12-08 David Edelsohn <edelsohn@gnu.org>
1541
1542 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1543 PICFLAG.
1544 * Makefile.in: Regenerated.
1545
14ac9144 15462004-12-02 Richard Sandiford <rsandifo@redhat.com>
1547
1548 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1549 * configure, aclocal.m4, Makefile.in: Regenerate.
1550 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1551
d081dd29 15522004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1553
1554 * configure: Regenerate for libtool change.
1555
a234d196 15562004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1557
1558 * configure: Regenerate for libtool reversion.
1559
7f93fe7a 15602004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1561
1562 * configure: Regenerate for libtool change.
1563
c0e31427 15642004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1565
1566 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1567
15682004-11-23 Richard Sandiford <rsandifo@redhat.com>
1569
1570 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1571 of jal. Use an absolute encoding for the frame information.
1572
15732004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1574
1575 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1576 * acinclude.m4: Delete logic for sincludes.
1577 * aclocal.m4, Makefile.in, configure: Regenerate.
1578 * include/Makefile: Likewise.
1579 * testsuite/Makefile: Likewise.
1580
68fa8e89 15812004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1582
1583 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1584 on a 8-byte boundary.
1585 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1586
c11820d9 15872004-10-27 Richard Earnshaw <rearnsha@arm.com>
1588
1589 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1590 long long values. Round stack allocation to a multiple of 8 bytes
1591 for ATPCS compatibility.
1592 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1593 names. Handle returning long long types. Add Thumb and interworking
1594 support. Improve soft-float code.
1595
03b2b968 15962004-10-27 Richard Earnshaw <rearnsha@arm.com>
1597
1598 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1599 (libffi_exit): New function.
1600 (libffi_init): Build the testglue wrapper if needed.
1601
9f234538 16022004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1603
1604 PR other/18138
1605 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1606
61dc6540 16072004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1608
1609 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1610
3b77fb00 16112004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1612
1613 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1614 * testsuite/libffi.call/float3.c: New test case.
1615
7c2fd652 16162004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1617
1618 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1619 the function returning a structure pointed with R2.
1620 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1621 the structure return value if T bit set. Emit position
1622 independent code and EH data if PIC.
1623
7e74b6d4 16242004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1625
1626 * Makefile.am: Add m32r support.
1627 * configure.ac: Likewise.
1628 * Makefile.in: Regenerate.
1629 * confiugre: Regenerate.
1630 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1631 (uint64, sint64, double, longdouble)
1632 * src/m32r: New directory.
1633 * src/m32r/ffi.c: New file.
1634 * src/m32r/sysv.S: Likewise.
1635 * src/m32r/ffitarget.h: Likewise.
1636
d601026d 16372004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1638
1639 * testsuite/libffi.call/negint.c: New test case.
1640
ec1bc4ab 16412004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1642
1643 PR libgcj/17465
1644 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1645 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1646 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1647 DYLD_LIBRARY_PATH.
1648
41067a17 16492004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1650
1651 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1652 * testsuite/libffi.call/promotion.c: Likewise.
1653 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1654 whitespaces.
1655 * testsuite/libffi.call/return_sc.c: Likewise.
1656 * testsuite/libffi.call/return_uc.c: Likewise.
1657
acbdb39a 16582004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1659
1660 * src/powerpc/darwin.S: Fix comments and identation.
1661 * src/powerpc/darwin_closure.S: Likewise.
1662
de7fad36 16632004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1664
1665 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1666 (ffi_prep_args): Handle longdouble arguments.
1667 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1668 longdouble.
1669 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1670 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1671 values.
1672 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1673 * src/types.c: Defined longdouble size and alignment for darwin.
1674
8bf7bfd1 16752004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1676
1677 * src/powerpc/aix.S: Remove whitespaces.
1678 * src/powerpc/aix_closure.S: Likewise.
1679 * src/powerpc/asm.h: Likewise.
1680 * src/powerpc/ffi.c: Likewise.
1681 * src/powerpc/ffitarget.h: Likewise.
1682 * src/powerpc/linux64.S: Likewise.
1683 * src/powerpc/linux64_closure.S: Likewise.
1684 * src/powerpc/ppc_closure.S: Likewise.
1685 * src/powerpc/sysv.S: Likewise.
1686
2e818f50 16872004-08-30 Anthony Green <green@redhat.com>
1688
1689 * Makefile.am: Add frv support.
1690 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1691 * configure.ac: Read configure.host.
1692 * configure.in: Read configure.host.
1693 * configure.host: New file. frv-elf needs libgloss.
1694 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
752da755 1695 alignment. This is needed to frv and shouldn't harm the others.
2e818f50 1696 * include/ffi_common.h (ALIGN_DOWN): New macro.
1697 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1698
d36998ee 16992004-08-24 David Daney <daney@avtrex.com>
1700
1701 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1702 * testsuite/libffi.call/closure_fn1.c: Likewise.
1703 * testsuite/libffi.call/closure_fn2.c Likewise.
1704 * testsuite/libffi.call/closure_fn3.c: Likewise.
1705 * testsuite/libffi.call/closure_fn4.c: Likewise.
1706 * testsuite/libffi.call/closure_fn5.c: Likewise.
1707 * testsuite/libffi.call/cls_18byte.c: Likewise.
1708 * testsuite/libffi.call/cls_19byte.c: Likewise.
1709 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1710 * testsuite/libffi.call/cls_20byte.c: Likewise.
1711 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1712 * testsuite/libffi.call/cls_24byte.c: Likewise.
1713 * testsuite/libffi.call/cls_2byte.c: Likewise.
1714 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1715 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1716 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1717 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1718 * testsuite/libffi.call/cls_4byte.c: Likewise.
1719 * testsuite/libffi.call/cls_64byte.c: Likewise.
1720 * testsuite/libffi.call/cls_6byte.c: Likewise.
1721 * testsuite/libffi.call/cls_7byte.c: Likewise.
1722 * testsuite/libffi.call/cls_8byte.c: Likewise.
1723 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1724 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1725 * testsuite/libffi.call/cls_align_double.c: Likewise.
1726 * testsuite/libffi.call/cls_align_float.c: Likewise.
1727 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1728 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1729 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1730 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1731 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1732 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1733 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1734 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1735 * testsuite/libffi.call/cls_double.c: Likewise.
1736 * testsuite/libffi.call/cls_float.c: Likewise.
1737 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1738 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1739 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1740 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1741 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1742 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1743 * testsuite/libffi.call/cls_schar.c: Likewise.
1744 * testsuite/libffi.call/cls_sint.c: Likewise.
1745 * testsuite/libffi.call/cls_sshort.c: Likewise.
1746 * testsuite/libffi.call/cls_uchar.c: Likewise.
1747 * testsuite/libffi.call/cls_uint.c: Likewise.
1748 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1749 * testsuite/libffi.call/cls_ushort.c: Likewise.
1750 * testsuite/libffi.call/nested_struct.c: Likewise.
1751 * testsuite/libffi.call/nested_struct1.c: Likewise.
1752 * testsuite/libffi.call/nested_struct2.c: Likewise.
1753 * testsuite/libffi.call/nested_struct3.c: Likewise.
1754 * testsuite/libffi.call/problem1.c: Likewise.
1755 * testsuite/libffi.special/unwindtest.cc: Likewise.
1756 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1757 to zero.
1758 * testsuite/libffi.call/cls_16byte.c: Likewise.
1759 * testsuite/libffi.call/cls_5byte.c: Likewise.
1760
29ac6c83 17612004-08-23 David Daney <daney@avtrex.com>
1762
1763 PR libgcj/13141
1764 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1765 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1766 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1767 parameters and return types.
1768 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1769 (ffi_prep_closure): Ditto.
1770 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1771 alignment calculations.
1772 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1773 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1774
4487a7e8 17752004-08-14 Casey Marshall <csm@gnu.org>
1776
1777 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1778 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1779 integer (O32 ABI only).
1780 (ffi_prep_closure): new function.
1781 (ffi_closure_mips_inner_O32): new function.
bbb07d35 1782 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
4487a7e8 1783 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1784 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1785 64 bit integers correctly.
1786 (ffi_closure_O32): new function.
1787 Added DWARF-2 unwind info for both functions.
1788
bb2d2363 17892004-08-10 Andrew Haley <aph@redhat.com>
1790
1791 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1792
cb8a7007 17932004-08-01 Robert Millan <robertmh@gnu.org>
1794
1795 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1796 * configure: Regenerate.
1797
e2f66690 17982004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1799
1800 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1801 and mmap() explicitly instead of relying on preset autoconf cache
1802 variables.
1803 * aclocal.m4: Regenerate.
1804 * configure: Regenerate.
1805
e395522d 18062004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1807
1808 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1809 (ffi_check_float_struct): Remove unused prototype.
1810
b589ba75 18112004-06-30 Geoffrey Keating <geoffk@apple.com>
1812
1813 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1814 character on Darwin, use '\n\t' instead.
1815
7b709c61 18162004-06-26 Matthias Klose <doko@debian.org>
1817
1818 * libtool-version: Fix typo in revision/age.
1819
06e2077b 18202004-06-17 Matthias Klose <doko@debian.org>
1821
1822 * libtool-version: New.
1823 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1824 * Makefile.in: Regenerate.
1825
4889e098 18262004-06-15 Paolo Bonzini <bonzini@gnu.org>
d947df39 1827
1828 * Makefile.am: Remove useless multilib rules.
1829 * Makefile.in: Regenerate.
1830 * aclocal.m4: Regenerate with automake 1.8.5.
1831 * configure.ac: Remove useless multilib configury.
1832 * configure: Regenerate.
1833
4889e098 18342004-06-15 Paolo Bonzini <bonzini@gnu.org>
b2191698 1835
1836 * .cvsignore: New file.
1837
6bc491fa 18382004-06-10 Jakub Jelinek <jakub@redhat.com>
1839
1840 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1841 fp_done.
1842 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1843 changed from 8.
1844
e1c1d65f 18452004-06-06 Sean McNeil <sean@mcneil.com>
1846
1847 * configure.ac: Add x86_64-*-freebsd* support.
1848 * configure: Regenerate.
1849
2be99030 18502004-04-26 Joe Buck <jbuck@welsh-buck.org>
1851
1852 Bug 15093
1853 * configure.ac: Test for existence of mmap and sys/mman.h before
1854 checking blacklist. Fix suggested by Jim Wilson.
1855 * configure: Regenerate.
1856
efefde65 18572004-04-26 Matt Austern <austern@apple.com>
1858
1859 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1860 FDE location.
1861 * src/powerpc/darwin_closure.S: Likewise.
bbb07d35 1862
5766842b 18632004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1864
1865 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1866 error. Reported by Thomas Heller <theller@python.net>.
1867 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1868 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1869
43935228 18702004-03-20 Matthias Klose <doko@debian.org>
1871
1872 * src/pa/linux.S: Fix typo.
1873
f35c338f 18742004-03-19 Matthias Klose <doko@debian.org>
1875
5766842b 1876 * Makefile.am: Update.
f35c338f 1877 * Makefile.in: Regenerate.
1878 * src/pa/ffi.h.in: Remove.
1879 * src/pa/ffitarget.h: New file.
1880
18812004-02-10 Randolph Chung <tausq@debian.org>
1882
1883 * Makefile.am: Add PA support.
1884 * Makefile.in: Regenerate.
1885 * include/Makefile.in: Regenerate.
1886 * configure.ac: Add PA target.
1887 * configure: Regenerate.
1888 * src/pa/ffi.c: New file.
1889 * src/pa/ffi.h.in: Add PA support.
1890 * src/pa/linux.S: New file.
1891 * prep_cif.c: Add PA support.
1892
021bf237 18932004-03-16 Hosaka Yuji <hos@tamanegi.org>
1894
1895 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
1896 double.
1897 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
1898 with ecif->cif->flags.
1899 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
1900 with cif->flags.
1901 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
1902 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
1903 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
1904 sc_retstruct2b): Add for 1 or 2-bytes struct case.
1905
4a900809 19062004-03-15 Kelley Cook <kcook@gcc.gnu.org>
1907
1908 * configure.in: Rename file to ...
1909 * configure.ac: ... this.
1910 * fficonfig.h.in: Regenerate.
1911 * Makefile.in: Regenerate.
1912 * include/Makefile.in: Regenerate.
1913 * testsuite/Makefile.in: Regenerate.
1914
d128584a 19152004-03-12 Matt Austern <austern@apple.com>
1916
1917 * src/powerpc/darwin.S: Fix EH information so it corresponds to
1918 changes in EH format resulting from addition of linkonce support.
1919 * src/powerpc/darwin_closure.S: Likewise.
021bf237 1920
b375837d 19212004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
1922 Paolo Bonzini <bonzini@gnu.org>
1923
1924 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
1925 Remove VPATH. Remove rules for object files. Remove multilib support.
1926 (AM_CCASFLAGS): Add.
1927 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
1928 (AC_PREREQ): Bump version to 2.59.
1929 (AC_INIT): Fill with version info and bug address.
1930 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
1931 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
1932 De-precious CC so that the right flags are passed down to multilibs.
1933 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
1934 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
1935 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
1936 * configure: Rebuilt.
1937 * aclocal.m4: Likewise.
1938 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
1939 * fficonfig.h.in: Likewise.
1940
52cee448 19412004-03-11 Andreas Schwab <schwab@suse.de>
1942
1943 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
1944 arguments from fp registers only for the first 8 parameter slots.
1945 Don't convert a float parameter when passed in memory.
1946
ac09b42b 19472004-03-09 Hans-Peter Nilsson <hp@axis.com>
1948
1949 * configure: Regenerate for config/accross.m4 correction.
1950
497c7020 19512004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
1952
1953 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
1954 ecif->cif->bytes to bytes.
1955 (ffi_prep_cif_machdep): Add braces around nested if statement.
1956
0ad895fb 19572004-02-09 Alan Modra <amodra@bigpond.net.au>
1958
ad98ab60 1959 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
1960
0ad895fb 1961 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
1962 (ffi_closure_helper_LINUX64): Fix typo.
1963 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
1964 for powerpc64-*-*.
1965 * testsuite/libffi.call/float.c: Likewise.
1966 * testsuite/libffi.call/float2.c: Likewise.
1967
de9c7178 19682004-02-08 Alan Modra <amodra@bigpond.net.au>
1969
1970 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
1971 long double function return and long double arg handling.
1972 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
1973 Use "end_pfr" instead of "nf". Correct long double handling.
1974 Localise "temp".
1975 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
1976 return value.
1977 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
1978 space for long double return value. Adjust stack frame and offsets.
1979 Load f2 long double return.
1980
3474a085 19812004-02-07 Alan Modra <amodra@bigpond.net.au>
1982
1983 * src/types.c: Use 16 byte long double for POWERPC64.
1984
071ff5d6 19852004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1986
1987 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
1988 when the structure return address is passed in %o0.
1989 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
1990 (ffi_v9_layout_struct): Align the field following a nested structure
1991 on a word boundary. Use memmove instead of memcpy.
1992 (ffi_call): Update call to ffi_V9_return_struct.
1993 (ffi_prep_closure): Define 'ctx' only for V8.
1994 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
1995 and ffi_closure_sparc_inner_v9.
1996 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
1997 Always skip the structure return address. For structures and long
1998 doubles, copy the argument directly.
1999 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2000 if required. Shift the maximum floating-point slot accordingly. For
2001 big structures, copy the argument directly; otherwise, left-justify the
2002 argument and call ffi_v9_layout_struct to lay out the structure on
2003 the stack.
2004 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2005 (ffi_closure_v8): Pass the structure return address. Update call to
2006 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2007 Skip the 'unimp' insn when returning long doubles and structures.
2008 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2009 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2010 FFI_TYPE_INT handling. Load structures both in integers and
2011 floating-point registers on return.
2012 * README: Update status of the SPARC port.
2013
24ee5f14 20142004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2015
2016 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2017 as of type ffi_arg.
2018 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2019
1e0ee4d8 20202004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2021
2022 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2023 value as of type ffi_arg, not unsigned int.
2024
c886ae3f 20252004-01-21 Michael Ritzert <ritzert@t-online.de>
2026
2027 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2028 of the LHS.
2029
fcd85db3 20302004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2031
2032 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2033 Solaris.
2034
dcf910e2 20352004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2036
2037 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2038 to void *.
2039
3d3042b8 20402003-12-10 Richard Henderson <rth@redhat.com>
2041
2042 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2043 size_t instead of int.
2044
dc36d4d1 20452003-12-04 Hosaka Yuji <hos@tamanegi.org>
2046
2047 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2048 * testsuite/libffi.call/many_win32.c (main): Replace variable
2049 int i with unsigned long ul.
2050
2051 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2052 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2053 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2054 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2055 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2056 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2057 * testsuite/libffi.call/cls_align_float.c: Likewise.
2058 * testsuite/libffi.call/cls_align_double.c: Likewise.
2059 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2060 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2061
f90351a5 20622003-12-02 Hosaka Yuji <hos@tamanegi.org>
2063
2064 PR other/13221
2065 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2066 Align arguments to 32 bits.
2067
1c4a2a48 20682003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2069
2070 PR other/13221
2071 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2072 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2073 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2074 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2075 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2076 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2077
2078 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2079
8c899267 20802003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2081
2082 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2083 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2084
857794c9 20852003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2086
2087 * Makefile.in: Rebuilt.
2088 * configure: Likewise.
2089 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2090 the right type.
2091
e7f352cc 20922003-11-21 Andreas Jaeger <aj@suse.de>
2093 Andreas Tobler <a.tobler@schweiz.ch>
2094
2095 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2096 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2097 * Makefile.in: Rebuilt.
2098 * aclocal.m4: Likewise.
2099 * configure: Likewise.
2100 * fficonfig.h.in: Likewise.
2101 * testsuite/lib/libffi-dg.exp: Add include dir.
2102 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2103 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2104 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2105 for ffi_closure if available.
2106 * testsuite/libffi.call/closure_fn1.c: Likewise.
2107 * testsuite/libffi.call/closure_fn2.c: Likewise.
2108 * testsuite/libffi.call/closure_fn3.c: Likewise.
2109 * testsuite/libffi.call/closure_fn4.c: Likewise.
2110 * testsuite/libffi.call/closure_fn5.c: Likewise.
2111 * testsuite/libffi.call/cls_12byte.c: Likewise.
2112 * testsuite/libffi.call/cls_16byte.c: Likewise.
2113 * testsuite/libffi.call/cls_18byte.c: Likewise.
2114 * testsuite/libffi.call/cls_19byte.c: Likewise.
2115 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2116 * testsuite/libffi.call/cls_20byte.c: Likewise.
2117 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2118 * testsuite/libffi.call/cls_24byte.c: Likewise.
2119 * testsuite/libffi.call/cls_2byte.c: Likewise.
2120 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2121 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2122 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2123 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2124 * testsuite/libffi.call/cls_4byte.c: Likewise.
2125 * testsuite/libffi.call/cls_5byte.c: Likewise.
2126 * testsuite/libffi.call/cls_64byte.c: Likewise.
2127 * testsuite/libffi.call/cls_6byte.c: Likewise.
2128 * testsuite/libffi.call/cls_7byte.c: Likewise.
2129 * testsuite/libffi.call/cls_8byte.c: Likewise.
2130 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2131 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2132 * testsuite/libffi.call/cls_double.c: Likewise.
2133 * testsuite/libffi.call/cls_float.c: Likewise.
2134 * testsuite/libffi.call/cls_schar.c: Likewise.
2135 * testsuite/libffi.call/cls_sint.c: Likewise.
2136 * testsuite/libffi.call/cls_sshort.c: Likewise.
2137 * testsuite/libffi.call/cls_uchar.c: Likewise.
2138 * testsuite/libffi.call/cls_uint.c: Likewise.
2139 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2140 * testsuite/libffi.call/cls_ushort.c: Likewise.
2141 * testsuite/libffi.call/nested_struct.c: Likewise.
2142 * testsuite/libffi.call/nested_struct1.c: Likewise.
2143 * testsuite/libffi.call/nested_struct2.c: Likewise.
2144 * testsuite/libffi.call/nested_struct3.c: Likewise.
2145 * testsuite/libffi.call/problem1.c: Likewise.
2146 * testsuite/libffi.special/unwindtest.cc: Likewise.
2147
71e96ac5 21482003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2149
2150 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2151
fd0a337b 21522003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2153
2154 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2155 Add -lgcc_s to additional flags.
2156
c2dfd8f6 21572003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2158
2159 * configure.in, include/Makefile.am: PR libgcj/11147, install
2160 the ffitarget.h header file in a gcc versioned and target
2161 dependent place.
2162 * configure: Regenerated.
2163 * Makefile.in, include/Makefile.in: Likewise.
2164 * testsuite/Makefile.in: Likewise.
2165
17c9ab63 21662003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2167
2168 * testsuite/libffi.call/closure_fn0.c: Print result and check
2169 with dg-output to make debugging easier.
2170 * testsuite/libffi.call/closure_fn1.c: Likewise.
2171 * testsuite/libffi.call/closure_fn2.c: Likewise.
2172 * testsuite/libffi.call/closure_fn3.c: Likewise.
2173 * testsuite/libffi.call/closure_fn4.c: Likewise.
2174 * testsuite/libffi.call/closure_fn5.c: Likewise.
2175 * testsuite/libffi.call/cls_12byte.c: Likewise.
2176 * testsuite/libffi.call/cls_16byte.c: Likewise.
2177 * testsuite/libffi.call/cls_18byte.c: Likewise.
2178 * testsuite/libffi.call/cls_19byte.c: Likewise.
2179 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2180 * testsuite/libffi.call/cls_20byte.c: Likewise.
2181 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2182 * testsuite/libffi.call/cls_24byte.c: Likewise.
2183 * testsuite/libffi.call/cls_2byte.c: Likewise.
2184 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2185 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2186 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2187 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2188 * testsuite/libffi.call/cls_4byte.c: Likewise.
2189 * testsuite/libffi.call/cls_5byte.c: Likewise.
2190 * testsuite/libffi.call/cls_64byte.c: Likewise.
2191 * testsuite/libffi.call/cls_6byte.c: Likewise.
2192 * testsuite/libffi.call/cls_7byte.c: Likewise.
2193 * testsuite/libffi.call/cls_8byte.c: Likewise.
2194 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2195 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2196 * testsuite/libffi.call/cls_double.c: Likewise.
2197 * testsuite/libffi.call/cls_float.c: Likewise.
2198 * testsuite/libffi.call/cls_schar.c: Likewise.
2199 * testsuite/libffi.call/cls_sint.c: Likewise.
2200 * testsuite/libffi.call/cls_sshort.c: Likewise.
2201 * testsuite/libffi.call/cls_uchar.c: Likewise.
2202 * testsuite/libffi.call/cls_uint.c: Likewise.
2203 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2204 * testsuite/libffi.call/cls_ushort.c: Likewise.
2205 * testsuite/libffi.call/problem1.c: Likewise.
2206
2207 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2208 static.
2209
a0b53ce0 22102003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2211
2212 * testsuite/libffi.call/cls_9byte2.c: New test case.
2213 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2214 * testsuite/libffi.call/cls_64byte.c: Likewise.
2215 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2216 * testsuite/libffi.call/cls_19byte.c: Likewise.
2217 * testsuite/libffi.call/cls_18byte.c: Likewise.
2218 * testsuite/libffi.call/closure_fn4.c: Likewise.
2219 * testsuite/libffi.call/closure_fn5.c: Likewise.
2220 * testsuite/libffi.call/cls_schar.c: Likewise.
2221 * testsuite/libffi.call/cls_sint.c: Likewise.
2222 * testsuite/libffi.call/cls_sshort.c: Likewise.
2223 * testsuite/libffi.call/nested_struct2.c: Likewise.
2224 * testsuite/libffi.call/nested_struct3.c: Likewise.
2225
ed0ed2e9 22262003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2227
2228 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2229 * testsuite/libffi.call/cls_uchar.c: Likewise.
2230 * testsuite/libffi.call/cls_uint.c: Likewise.
2231 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2232 * testsuite/libffi.call/cls_ushort.c: Likewise.
2233 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2234
61641cc0 22352003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2236
2237 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2238 the initialization.
2239
6a759884 22402003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2241
61641cc0 2242 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
6a759884 2243 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2244
4f388831 22452003-10-22 David Daney <ddaney@avtrex.com>
2246
2247 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2248 __attribute__((__mode__(__SI__))) and friends.
2249
b4891aa6 22502003-10-22 Andreas Schwab <schwab@suse.de>
2251
2252 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2253
3d37e05c 22542003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2255
2256 * configure.in: AC_LINK_FILES(ffitarget.h).
2257 * configure: Regenerate.
2258 * Makefile.in: Likewise.
2259 * include/Makefile.in: Likewise.
2260 * testsuite/Makefile.in: Likewise.
2261 * fficonfig.h.in: Likewise.
2262
22632003-10-21 Paolo Bonzini <bonzini@gnu.org>
2264 Richard Henderson <rth@redhat.com>
2265
2266 Avoid that ffi.h includes fficonfig.h.
2267
2268 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2269 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2270 (TARGET_SRC_MIPS_SGI): Removed.
2271 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2272 (MIPS_SGI): Removed.
2273 (CLEANFILES): Removed.
2274 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2275 targets.
2276 * acconfig.h: Removed.
2277 * configure.in: Compute sizeofs only for double and long double.
2278 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2279 into AC_DEFINE instead of using acconfig.h. Create
2280 include/ffitarget.h instead of include/fficonfig.h. Rename
2281 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2282 AC_DEFINE EH_FRAME_FLAGS.
2283 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2284 (hack_DATA): Add ffitarget.h.
2285 * include/ffi.h.in: Remove all system specific definitions.
2286 Declare raw API even if it is not installed, why bother?
2287 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2288 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2289 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2290 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2291 * include/ffi_common.h (bool): Do not define.
2292 (ffi_assert): Accept failed assertion.
2293 (ffi_type_test): Return void and accept file/line.
2294 (FFI_ASSERT): Pass stringized failed assertion.
2295 (FFI_ASSERT_AT): New macro.
2296 (FFI_ASSERT_VALID_TYPE): New macro.
2297 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2298 UINT64, SINT64): Define here with gcc's __attribute__ macro
2299 instead of in ffi.h
2300 (FLOAT32, ALIGN): Define here instead of in ffi.h
2301 * include/ffi-mips.h: Removed. Its content moved to
2302 src/mips/ffitarget.h after separating assembly and C sections.
2303 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2304 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2305 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2306 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2307 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2308 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2309 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2310 (ffi_type_test): Rewritten.
2311 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2312 FFI_ASSERT_VALID_TYPE.
2313 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2314 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2315 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2316 src/s390/ffitarget.h, src/sh/ffitarget.h,
2317 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2318 src/x86/ffitarget.h: New files.
2319 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2320 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2321 src/powerpc/aix.S, src/powerpc/darwin.S,
2322 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2323 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2324 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2325 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2326 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2327 include fficonfig.h
2328
2a5f9ffc 23292003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2330
2331 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2332 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2333
51c5f96c 23342003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2335
2336 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2337 Used when FFI_DEBUG = 1.
2338
ee46e78f 23392003-10-14 Alan Modra <amodra@bigpond.net.au>
2340
2341 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2342 and align.
2343
a3abe7fe 23442003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2345
2346 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2347 FFI_MIPS_O32 for O32 ABI.
2348
1ee4561e 23492003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2350
2351 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2352 SPARC64. Cleanup whitespaces.
2353
93421fe6 23542003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2355
2356 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2357 strongarm, xscale. Cleanup whitespaces.
2358 * testsuite/libffi.call/closure_fn1.c: Likewise.
2359 * testsuite/libffi.call/closure_fn2.c: Likewise.
2360 * testsuite/libffi.call/closure_fn3.c: Likewise.
2361 * testsuite/libffi.call/cls_12byte.c: Likewise.
2362 * testsuite/libffi.call/cls_16byte.c: Likewise.
2363 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2364 * testsuite/libffi.call/cls_20byte.c: Likewise.
2365 * testsuite/libffi.call/cls_24byte.c: Likewise.
2366 * testsuite/libffi.call/cls_2byte.c: Likewise.
2367 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2368 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2369 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2370 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2371 * testsuite/libffi.call/cls_4byte.c: Likewise.
2372 * testsuite/libffi.call/cls_5byte.c: Likewise.
2373 * testsuite/libffi.call/cls_6byte.c: Likewise.
2374 * testsuite/libffi.call/cls_7byte.c: Likewise.
2375 * testsuite/libffi.call/cls_8byte.c: Likewise.
2376 * testsuite/libffi.call/cls_double.c: Likewise.
2377 * testsuite/libffi.call/cls_float.c: Likewise.
2378 * testsuite/libffi.call/cls_uchar.c: Likewise.
2379 * testsuite/libffi.call/cls_uint.c: Likewise.
2380 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2381 * testsuite/libffi.call/cls_ushort.c: Likewise.
2382 * testsuite/libffi.call/nested_struct.c: Likewise.
2383 * testsuite/libffi.call/nested_struct1.c: Likewise.
2384 * testsuite/libffi.call/problem1.c: Likewise.
2385 * testsuite/libffi.special/unwindtest.cc: Likewise.
2386 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2387
8a94dd75 23882003-09-18 David Edelsohn <edelsohn@gnu.org>
2389
2390 * src/powerpc/aix.S: Cleanup whitespaces.
2391 * src/powerpc/aix_closure.S: Likewise.
2392
bd1f3533 23932003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2394
2395 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2396 * src/powerpc/darwin_closure.S: Likewise.
2397 * src/powerpc/ffi_darwin.c: Likewise.
2398
7a142533 23992003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2400 David Edelsohn <edelsohn@gnu.org>
2401
2402 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2403 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2404 parameter stack.
2405 * src/powerpc/darwin_closure.S: Likewise.
2406 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2407 according to the Darwin/AIX ABI.
2408 (ffi_prep_cif_machdep): Likewise.
2409 (ffi_closure_helper_DARWIN): Likewise.
2410 Remove the outgoing parameter stack logic. Simplify the evaluation
2411 of the different CASE types.
2412 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2413 statement in the trampoline code.
2414
d80727ae 24152003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2416
2417 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2418 for the register size.
2419 (ffi_closure_helper_SYSV): Handle the structure return value
2420 address correctly.
2421 (ffi_closure_helper_SYSV): Return the appropriate type when
2422 the registers are used for the structure return value.
2423 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2424 the 64-bit return value. Update copyright years.
2425
47fab6c3 24262003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2427
2428 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2429 srcdir for ffi_mips.h.
2430
f397026a 24312003-09-12 Alan Modra <amodra@bigpond.net.au>
2432
2433 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2434 structure size.
2435 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2436 placement of float result.
2437 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2438 cast of "resp" for big-endian 64 bit machines.
2439
341b7dcf 24402003-09-11 Alan Modra <amodra@bigpond.net.au>
2441
2442 * src/types.c (double, longdouble): Merge identical SH and ARM
2443 typedefs, and add POWERPC64.
2444 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2445 struct split over gpr and rest.
2446 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2447 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2448
06855f3a 24492003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2450
2451 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2452 passing correctly.
c886ae3f 2453
9165b0af 24542003-09-09 Alan Modra <amodra@bigpond.net.au>
2455
2456 * configure: Regenerate.
2457
98d58335 24582003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2459
2460 * Makefile.am: Remove build rules for ffitest.
2461 * Makefile.in: Rebuilt.
c886ae3f 2462
98d58335 24632003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
0262ee1f 2464
2465 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2466 about implicit declaration of abort().
2467
3743c332 24682003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2469
2470 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2471 * Makefile.in: Rebuilt.
2472 * configure.in: Add dejagnu test framework.
2473 * configure: Rebuilt.
2474
2475 * testsuite/Makefile.am: New file.
2476 * testsuite/Makefile.in: Built
c886ae3f 2477 * testsuite/lib/libffi-dg.exp: New file.
3743c332 2478 * testsuite/config/default.exp: Likewise.
2479 * testsuite/libffi.call/call.exp: Likewise.
2480 * testsuite/libffi.call/ffitest.h: Likewise.
2481 * testsuite/libffi.call/closure_fn0.c: Likewise.
2482 * testsuite/libffi.call/closure_fn1.c: Likewise.
2483 * testsuite/libffi.call/closure_fn2.c: Likewise.
2484 * testsuite/libffi.call/closure_fn3.c: Likewise.
2485 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2486 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2487 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2488 * testsuite/libffi.call/cls_2byte.c: Likewise.
2489 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2490 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2491 * testsuite/libffi.call/cls_4byte.c: Likewise.
2492 * testsuite/libffi.call/cls_5byte.c: Likewise.
2493 * testsuite/libffi.call/cls_6byte.c: Likewise.
2494 * testsuite/libffi.call/cls_7byte.c: Likewise.
2495 * testsuite/libffi.call/cls_8byte.c: Likewise.
2496 * testsuite/libffi.call/cls_12byte.c: Likewise.
2497 * testsuite/libffi.call/cls_16byte.c: Likewise.
2498 * testsuite/libffi.call/cls_20byte.c: Likewise.
2499 * testsuite/libffi.call/cls_24byte.c: Likewise.
2500 * testsuite/libffi.call/cls_double.c: Likewise.
2501 * testsuite/libffi.call/cls_float.c: Likewise.
2502 * testsuite/libffi.call/cls_uchar.c: Likewise.
2503 * testsuite/libffi.call/cls_uint.c: Likewise.
2504 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2505 * testsuite/libffi.call/cls_ushort.c: Likewise.
2506 * testsuite/libffi.call/float.c: Likewise.
2507 * testsuite/libffi.call/float1.c: Likewise.
2508 * testsuite/libffi.call/float2.c: Likewise.
2509 * testsuite/libffi.call/many.c: Likewise.
2510 * testsuite/libffi.call/many_win32.c: Likewise.
2511 * testsuite/libffi.call/nested_struct.c: Likewise.
2512 * testsuite/libffi.call/nested_struct1.c: Likewise.
2513 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2514 * testsuite/libffi.call/problem1.c: Likewise.
2515 * testsuite/libffi.call/promotion.c: Likewise.
2516 * testsuite/libffi.call/return_ll.c: Likewise.
2517 * testsuite/libffi.call/return_sc.c: Likewise.
2518 * testsuite/libffi.call/return_uc.c: Likewise.
2519 * testsuite/libffi.call/strlen.c: Likewise.
2520 * testsuite/libffi.call/strlen_win32.c: Likewise.
2521 * testsuite/libffi.call/struct1.c: Likewise.
2522 * testsuite/libffi.call/struct2.c: Likewise.
2523 * testsuite/libffi.call/struct3.c: Likewise.
2524 * testsuite/libffi.call/struct4.c: Likewise.
2525 * testsuite/libffi.call/struct5.c: Likewise.
2526 * testsuite/libffi.call/struct6.c: Likewise.
2527 * testsuite/libffi.call/struct7.c: Likewise.
2528 * testsuite/libffi.call/struct8.c: Likewise.
c886ae3f 2529 * testsuite/libffi.call/struct9.c: Likewise.
3743c332 2530 * testsuite/libffi.special/special.exp: New file.
2531 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2532 * testsuite/libffi.special/unwindtest.cc: Likewise.
2533
c886ae3f 2534
3a845f75 25352003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2536
2537 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2538 copyright years.
2539
4637dfa6 25402003-08-02 Alan Modra <amodra@bigpond.net.au>
2541
2542 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2543 structure passing.
2544 (ffi_closure_helper_LINUX64): Likewise.
2545 * src/powerpc/linux64.S: Remove code writing to parm save area.
2546 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2547 address in lr from ffi_closure_helper_LINUX64 call to calculate
2548 table address. Optimize function tail.
2549
8bcdc7f0 25502003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2551
2552 * src/sparc/ffi.c: Handle all floating point registers.
2553 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2554
3d6cc549 25552003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2556
2557 * README: Note that libffi is not part of GCC. Update the project
2558 URL and status.
2559
9b770b5a 25602003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2561
2562 * src/powerpc/ppc_closure.S: Include ffi.h.
2563
d46ab9f3 25642003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2565
2566 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2567 Use C style comments.
2568
81f536a3 25692003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2570
2571 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2572 * Makefile.in: Regenerate.
2573 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2574 * configure: Regenerate.
2575 * include/ffi.h.in: Add SHmedia support.
2576 * src/sh64/ffi.c: New file.
2577 * src/sh64/sysv.S: New file.
2578
1da888ec 25792003-05-16 Jakub Jelinek <jakub@redhat.com>
2580
2581 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2582 should be read-only.
2583 * configure: Rebuilt.
2584 * fficonfig.h.in: Rebuilt.
2585 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2586 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2587 * src/powerpc/linux64.S: Likewise.
2588 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2589 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2590 if -fpic/-fPIC/-mrelocatable.
2591 * src/powerpc/powerpc_closure.S: Likewise.
2592 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2593 #write in .eh_frame flags.
2594 * src/sparc/v9.S: Likewise.
2595 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2596 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2597 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2598
f3bd7a2c 25992003-05-07 Jeff Sturm <jsturm@one-point.com>
2600
2601 Fixes PR bootstrap/10656
2602 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2603 support for .register pseudo-op.
2604 * src/sparc/v8.S: Use it.
2605 * fficonfig.h.in: Rebuilt.
2606 * configure: Rebuilt.
2607
3647950d 26082003-04-18 Jakub Jelinek <jakub@redhat.com>
2609
2610 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2611 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2612 Make it the default on POWERPC64.
2613 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2614 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2615 * configure: Rebuilt.
2616 * src/powerpc/ffi.c (hidden): Define.
2617 (ffi_prep_args_SYSV): Renamed from
2618 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2619 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2620 ASM_NEEDS_REGISTERS64): New.
2621 (ffi_prep_args64): New function.
2622 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2623 (ffi_call): Likewise.
2624 (ffi_prep_closure): Likewise.
2625 (flush_icache): Surround by #ifndef POWERPC64.
2626 (ffi_dblfl): New union type.
2627 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2628 (ffi_closure_helper_LINUX64): New function.
2629 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2630 __powerpc64__.
2631 * src/powerpc/sysv.S: Likewise.
2632 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2633 * src/powerpc/linux64.S: New file.
2634 * src/powerpc/linux64_closure.S: New file.
2635 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2636 src/powerpc/linux64_closure.S.
2637 (TARGET_SRC_POWERPC): Likewise.
2638
2639 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2640 closure_test_fn3): Fix result printing on big-endian 64-bit
2641 machines.
2642 (main): Print tst2_arg instead of uninitialized tst2_result.
2643
2644 * src/ffitest.c (main): Hide what closure pointer really points to
2645 from the compiler.
2646
1636e920 26472003-04-16 Richard Earnshaw <rearnsha@arm.com>
2648
2649 * configure.in (arm-*-netbsdelf*): Add configuration.
2650 (configure): Regenerated.
2651
e663564a 26522003-04-04 Loren J. Rittle <ljrittle@acm.org>
2653
2654 * include/Makefile.in: Regenerate.
2655
02f5d4c9 26562003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2657
2658 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2659 bit mode.
2660 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2661 Receive closure pointer through parameter, read args using
2662 __builtin_dwarf_cfa.
2663 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2664
b8e27232 26652003-03-12 Andreas Schwab <schwab@suse.de>
2666
2667 * configure.in: Avoid trailing /. in toolexeclibdir.
2668 * configure: Rebuilt.
2669
5fee68c6 26702003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2671
bbb07d35 2672 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
5fee68c6 2673
ecadab3f 26742003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2675
2676 * libffi/src/powerpc/darwin_closure.S:
2677 Fix alignement bug, allocate 8 bytes for the result.
c886ae3f 2678 * libffi/src/powerpc/aix_closure.S:
ecadab3f 2679 Likewise.
2680 * libffi/src/powerpc/ffi_darwin.c:
2681 Update stackframe description for aix/darwin_closure.S.
c886ae3f 2682
a0294641 26832003-02-06 Jakub Jelinek <jakub@redhat.com>
2684
2685 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2686 attribute.
2687
624476ed 26882003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2689 Andreas Schwab <schwab@suse.de>
2690
bbb07d35 2691 * configure.in: Adjust command to source config-ml.in to account
624476ed 2692 for changes to the libffi_basedir definition.
2693 (libffi_basedir): Remove ${srcdir} from value and include trailing
2694 slash if nonempty.
c886ae3f 2695
bbb07d35 2696 * configure: Regenerate.
624476ed 2697
1143cc9b 26982003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2699
2700 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2701
967ae8fb 27022003-01-28 Andrew Haley <aph@redhat.com>
2703
2704 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2705 * src/x86/ffi64.c (ffi_prep_closure): New.
2706 (ffi_closure_UNIX64_inner): New.
2707 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2708
c8d4fa8c 27092003-01-27 Alexandre Oliva <aoliva@redhat.com>
2710
2711 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2712 Remove USE_LIBDIR conditional.
2713 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2714 * Makefile.in, configure: Rebuilt.
2715
2dd7852c 27162003-01027 David Edelsohn <edelsohn@gnu.org>
2717
2718 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2719 * Makefile.in: Regenerate.
2720
008eadcd 27212003-01-22 Andrew Haley <aph@redhat.com>
2722
2723 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2724 unwind info.
2725
cb440058 27262003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2727
2728 * src/powerpc/darwin.S: Add unwind info.
2729 * src/powerpc/darwin_closure.S: Likewise.
2730
149417bc 27312003-01-14 Andrew Haley <aph@redhat.com>
2732
2733 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2734 (ffi_prep_cif_machdep): Likewise.
2735 * src/x86/unix64.S: Add unwind info.
2736
d4dcd973 27372003-01-14 Andreas Jaeger <aj@suse.de>
2738
2739 * src/ffitest.c (main): Only use ffi_closures if those are
2740 supported.
2741
a7898d0c 27422003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2743
2744 * libffi/src/ffitest.c
bbb07d35 2745 add closure testcases
a7898d0c 2746
27472003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2748
2749 * libffi/src/powerpc/ffi.c
bbb07d35 2750 fix alignment bug for float (4 byte aligned iso 8 byte)
a7898d0c 2751
4e844289 27522003-01-09 Geoffrey Keating <geoffk@apple.com>
2753
2754 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2755 * src/powerpc/darwin.S: Remove RCS version string.
2756
0ef3e9c5 27572003-01-03 Jeff Sturm <jsturm@one-point.com>
2758
2759 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2760 * src/ffitest.c (main): Use static storage for closure.
2761 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2762 * src/sparc/v8.S (ffi_closure_v8): New.
2763 * src/sparc/v9.S (ffi_closure_v9): New.
2764
acdc3fc3 27652002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2766
2767 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2768 enumeration for X86_WIN32.
2769 * src/x86/win32.S: Added ffi_call_STDCALL function
2770 definition.
2771 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2772 switch cases for recognising FFI_STDCALL and
2773 calling ffi_call_STDCALL if target is X86_WIN32.
2774 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2775 stdcall versions of the "my_strlen" and "many"
2776 test functions (for X86_WIN32).
2777 Added test cases to test stdcall invocation using
2778 these functions.
2779
805e22b2 27802002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2781
2782 * src/sh/sysv.S: Add DWARF2 unwind info.
2783
6d219176 27842002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2785
2786 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2787
4827afc9 27882002-11-26 Jim Wilson <wilson@redhat.com>
2789
2790 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2791
805e22b2 27922002-11-23 H.J. Lu <hjl@gnu.org>
2793
2794 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2795 Include ../config/accross.m4.
2796 * aclocal.m4; Rebuild.
2797 * configure: Likewise.
2798
e3a04290 27992002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2800
2801 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2802
be36f77d 28032002-11-11 DJ Delorie <dj@redhat.com>
2804
2805 * configure.in: Look for common files in the right place.
2806
dd985e54 28072002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2808
2809 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2810 raw data as _Jv_word values, not ffi_raw.
2811 (ffi_java_ptrarray_to_raw): Likewise.
2812 (ffi_java_rvalue_to_raw): New function.
2813 (ffi_java_raw_call): Call it.
2814 (ffi_java_raw_to_rvalue): New function.
2815 (ffi_java_translate_args): Call it.
2816 * src/ffitest.c (closure_test_fn): Interpret return value
2817 as ffi_arg, not int.
2818 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2819 FFI_TYPE_POINTER case.
2820 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2821 values extended to word size.
2822
4a18e5fa 28232002-10-02 Andreas Jaeger <aj@suse.de>
2824
2825 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2826
3d8ac088 28272002-10-01 Bo Thorsen <bo@smetana.suse.de>
2828
2829 * include/ffi.h.in: Fix i386 win32 compilation.
2830
e0aaecf3 28312002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2832
2833 * configure.in: Add s390x-*-linux-* target.
2834 * configure: Regenerate.
2835 * include/ffi.h.in: Define S390X for s390x targets.
2836 (FFI_CLOSURES): Define for s390/s390x.
2837 (FFI_TRAMPOLINE_SIZE): Likewise.
2838 (FFI_NATIVE_RAW_API): Likewise.
2839 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2840 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
805e22b2 2841 * src/s390/ffi.c: Major rework of existing code. Add support for
e0aaecf3 2842 s390x targets. Add closure support.
2843 * src/s390/sysv.S: Likewise.
2844
c3e9c301 28452002-09-29 Richard Earnshaw <rearnsha@arm.com>
2846
2847 * src/arm/sysv.S: Fix typo.
2848
d13886be 28492002-09-28 Richard Earnshaw <rearnsha@arm.com>
2850
2851 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2852 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2853 (ffi_call_SYSV): Handle soft-float.
2854
aaff811e 28552002-09-27 Bo Thorsen <bo@suse.de>
2856
2857 * include/ffi.h.in: Fix multilib x86-64 support.
2858
5f7912d8 28592002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2860
2861 * Makefile.am (all-multi): Fix multilib parallel build.
2862
ccda5e55 28632002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2864
2865 * configure.in (sh[34]*-*-linux*): Add brackets.
2866 * configure: Regenerate.
2867
b1e5d71d 28682002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2869
2870 * Makefile.am: Add SH support.
2871 * Makefile.in: Regenerate.
2872 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2873 * configure: Regenerate.
2874 * include/ffi.h.in: Add SH support.
2875 * src/sh/ffi.c: New file.
2876 * src/sh/sysv.S: New file.
2877 * src/types.c: Add SH support.
2878
42c71854 28792002-07-16 Bo Thorsen <bo@suse.de>
2880
2881 * src/x86/ffi64.c: New file that adds x86-64 support.
2882 * src/x86/unix64.S: New file that handles argument setup for
2883 x86-64.
2884 * src/x86/sysv.S: Don't use this on x86-64.
2885 * src/x86/ffi.c: Don't use this on x86-64.
2886 Remove unused vars.
2887 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2888 for x86-64.
2889 * src/ffitest.c (struct6): New test that tests a special case in
2890 the x86-64 ABI.
2891 (struct7): Likewise.
2892 (struct8): Likewise.
2893 (struct9): Likewise.
2894 (closure_test_fn): Silence warning about this when it's not used.
2895 (main): Add the new tests.
2896 (main): Fix a couple of wrong casts and silence some compiler warnings.
2897 * include/ffi.h.in: Add x86-64 ABI definition.
2898 * fficonfig.h.in: Regenerate.
2899 * Makefile.am: Add x86-64 support.
2900 * configure.in: Likewise.
2901 * Makefile.in: Regenerate.
2902 * configure: Likewise.
2903
29042002-06-24 Bo Thorsen <bo@suse.de>
2905
2906 * src/types.c: Merge settings for similar architectures.
2907 Add x86-64 sizes and alignments.
2908
29092002-06-23 Bo Thorsen <bo@suse.de>
2910
2911 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
2912 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
2913 * src/mips/ffi.c (ffi_prep_args): Likewise.
2914 * src/m68k/ffi.c (ffi_prep_args): Likewise.
2915
72726889 29162002-07-18 H.J. Lu (hjl@gnu.org)
2917
2918 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
2919 (libffi_la_SOURCES): Support MIPS_LINUX.
2920 (libffi_convenience_la_SOURCES): Likewise.
2921 * Makefile.in: Regenerated.
2922
2923 * configure.in (mips64*-*): Skip.
2924 (mips*-*-linux*): New.
2925 * configure: Regenerated.
2926
2927 * src/mips/ffi.c: Include <sgidefs.h>.
2928
e17e640b 29292002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
2930
2931 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
2932
b5d69d52 29332002-05-27 Roger Sayle <roger@eyesopen.com>
2934
2935 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
2936
eadca1d5 29372002-05-27 Bo Thorsen <bo@suse.de>
2938
2939 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
2940 fix formatting.
2941
2abdadc9 29422002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
2943
2944 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
2945 beginning of function (for older apple cc).
2946
0b045cf7 29472002-05-08 Alexandre Oliva <aoliva@redhat.com>
2948
2949 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
2950 script entry, and set LD to it when configuring multilibs.
2951 * configure: Rebuilt.
2952
7f0e056d 29532002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
2954
2955 * configure.in (sparc64-*-netbsd*): Add target.
2956 (sparc-*-netbsdelf*): Likewise.
2957 * configure: Regenerate.
2958
b19313f4 29592002-04-28 David S. Miller <davem@redhat.com>
2960
2961 * configure.in, configure: Fix SPARC test in previous change.
2962
ea1f83c5 29632002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
2964
2965 * Makefile.am: Add Linux for S/390 support.
2966 * Makefile.in: Regenerate.
2967 * configure.in: Add Linux for S/390 support.
2968 * configure: Regenerate.
2969 * include/ffi.h.in: Add Linux for S/390 support.
2970 * src/s390/ffi.c: New file from libffi CVS tree.
2971 * src/s390/sysv.S: New file from libffi CVS tree.
2972
b40c762b 29732002-04-28 Jakub Jelinek <jakub@redhat.com>
2974
2975 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
2976 %r_disp32().
2977 * src/sparc/v8.S: Use it.
2978 * src/sparc/v9.S: Likewise.
2979 * fficonfig.h.in: Rebuilt.
2980 * configure: Rebuilt.
2981
00f87a50 29822002-04-08 Hans Boehm <Hans_Boehm@hp.com>
2983
2984 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
2985 correctly.
2986 * src/ia64/unix.S: Add unwind information. Fix comments.
2987 Save sp in a way that's compatible with unwind info.
2988 (ffi_call_unix): Correctly restore sp in all cases.
2989 * src/ia64/ffi.c: Add, fix comments.
2990
a622ae9d 29912002-04-08 Jakub Jelinek <jakub@redhat.com>
2992
2993 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
2994
c1d86bcd 29952002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
2996
2997 * configure.in (alpha*-*-netbsd*): Add target.
2998 * configure: Regenerate.
2999
3faa4340 30002002-04-04 Jeff Sturm <jsturm@one-point.com>
3001
3002 * src/sparc/v8.S: Add unwind info.
3003 * src/sparc/v9.S: Likewise.
3004
ec66e65b 30052002-03-30 Krister Walfridsson <cato@df.lth.se>
3006
3007 * configure.in: Enable i*86-*-netbsdelf*.
3008 * configure: Rebuilt.
805e22b2 3009
9ee9948b 30102002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3011
3012 PR other/2620
3013 * src/mips/n32.s: Delete
3014 * src/mips/o32.s: Delete
3015
6c74b4f2 30162002-03-21 Loren J. Rittle <ljrittle@acm.org>
3017
3018 * configure.in: Enable alpha*-*-freebsd*.
3019 * configure: Rebuilt.
3020
9d0a2dd5 30212002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3022
3023 * Makefile.am: libfficonvenience -> libffi_convenience.
3024 * Makefile.in: Rebuilt.
805e22b2 3025
ee7750c2 3026 * Makefile.am: Define ffitest_OBJECTS.
3027 * Makefile.in: Rebuilt.
9d0a2dd5 3028
f6a16bd8 30292002-03-07 Andreas Tobler <toa@pop.agri.ch>
3030 David Edelsohn <edelsohn@gnu.org>
3031
3032 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3033 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3034 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3035 * Makefile.in: Regenerate.
3036 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3037 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3038 (flush_icache, flush_range): New functions.
3039 (ffi_closure_helper_DARWIN): New function.
3040 * src/powerpc/aix_closure.S: New file.
3041 * src/powerpc/darwin_closure.S: New file.
805e22b2 3042
fed8d18d 30432002-02-24 Jeff Sturm <jsturm@one-point.com>
3044
3045 * include/ffi.h.in: Add typedef for ffi_arg.
3046 * src/ffitest.c (main): Declare rint with ffi_arg.
805e22b2 3047
ddac7e16 30482002-02-21 Andreas Tobler <toa@pop.agri.ch>
3049
3050 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3051 number of GPRs for floating-point arguments.
3052
ed35b400 30532002-01-31 Anthony Green <green@redhat.com>
3054
3055 * configure: Rebuilt.
3056 * configure.in: Replace CHECK_SIZEOF and endian tests with
3057 cross-compiler friendly macros.
3058 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3059 macros.
3060
bf270768 30612002-01-18 David Edelsohn <edelsohn@gnu.org>
3062
3063 * src/powerpc/darwin.S (_ffi_call_AIX): New.
dc155f85 3064 * src/powerpc/aix.S (ffi_call_DARWIN): New.
bf270768 3065
faae36ce 30662002-01-17 David Edelsohn <edelsohn@gnu.org>
3067
3068 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3069 (TARGET_SRC_POWERPC_AIX): New.
3070 (POWERPC_AIX): New stanza.
3071 * Makefile.in: Regenerate.
3072 * configure.in: Add AIX case.
3073 * configure: Regenerate.
3074 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3075 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3076 size. Fix "long double" support.
3077 (ffi_call): Add FFI_AIX case.
3078 * src/powerpc/aix.S: New.
3079
b81d6fe1 30802001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3081
3082 Implement Darwin PowerPC ABI.
3083 * configure.in: Handle powerpc-*-darwin*.
3084 * Makefile.am: Set source files for POWERPC_DARWIN.
3085 * configure: Rebuilt.
3086 * Makefile.in: Rebuilt.
3087 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3088 POWERPC_DARWIN.
3089 * src/powerpc/darwin.S: New file.
3090 * src/powerpc/ffi_darwin.c: New file.
3091
9bccb7a3 30922001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3093
3094 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3095
ac1f371e 30962001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3097
3098 * src/x86/sysv.S: Avoid gas-only .balign directive.
3099 Use C style comments.
3100
e1e36390 31012001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3102
3103 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3104 Fixes PR bootstrap/3563.
3105
deb29e7a 31062001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3107
3108 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3109
8008de24 31102001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3111
3112 * configure.in: Recognize sparc*-sun-* host.
3113 * configure: Regenerate.
805e22b2 3114
5fc46419 31152001-06-06 Andrew Haley <aph@redhat.com>
3116
968b0d1e 3117 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
5fc46419 3118
49ff2ded 31192001-06-03 Andrew Haley <aph@redhat.com>
3120
968b0d1e 3121 * src/alpha/osf.S: Add unwind info.
3122 * src/powerpc/sysv.S: Add unwind info.
3123 * src/powerpc/ppc_closure.S: Likewise.
49ff2ded 3124
070883fe 31252000-05-31 Jeff Sturm <jsturm@one-point.com>
3126
3127 * configure.in: Fix AC_ARG_ENABLE usage.
3128 * configure: Rebuilt.
3129
09da9506 31302001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3131
3132 * configure.in: Remove warning about beta code.
3133 * configure: Rebuilt.
3134
31352001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3136
6c1b002f 3137 * src/ia64/unix.S: Restore stack pointer when returning from
3138 ffi_closure_UNIX.
3139 * src/ia64/ffi.c: Fix typo in comment.
3140
c4312026 31412001-04-18 Jim Wilson <wilson@redhat.com>
3142
3143 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3144 to eliminate RAW DV.
3145
db6ca0af 31462001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3147
3148 * Makefile.am: Make a libtool convenience library.
3149 * Makefile.in: Rebuilt.
3150
9461b350 31512001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3152
3153 * configure.in: Use different syntax for subdirectory creation.
3154 * configure: Rebuilt.
3155
a4cc90ca 31562001-03-27 Jon Beniston <jon@beniston.com>
3157
3158 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3159 * configure: Rebuilt.
3160 * Makefile.am: Added X86_WIN32 target support.
3161 * Makefile.in: Rebuilt.
3162
3163 * include/ffi.h.in: Added X86_WIN32 target support.
3164
3165 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3166 * src/types.c: Added X86_WIN32 target support.
3167
3168 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3169 stuff removed and made to work with CygWin's gas.
3170
21f4579d 31712001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3172
3173 * configure.in: Make target subdirectory in build dir.
3174 * Makefile.am: Override suffix based rules to specify correct output
3175 subdirectory.
3176 * Makefile.in: Rebuilt.
3177 * configure: Rebuilt.
3178
29183c57 31792001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
384eea1f 3180
3181 * src/powerpc/ppc_closure.S: New file.
3182 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3183 involving long long and register pairs.
3184 (ffi_prep_closure): New function.
3185 (flush_icache): Likewise.
3186 (ffi_closure_helper_SYSV): Likewise.
3187 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3188 (FFI_TRAMPOLINE_SIZE): Likewise.
3189 (FFI_NATIVE_RAW_API): Likewise.
3190 * Makefile.in: Rebuilt.
3191 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3192 (TARGET_SRC_POWERPC): Likewise.
3193
49604b5b 31942001-03-19 Tom Tromey <tromey@redhat.com>
3195
3196 * Makefile.in: Rebuilt.
3197 * Makefile.am (ffitest_LDFLAGS): New macro.
3198
c8fdcc07 31992001-03-02 Nick Clifton <nickc@redhat.com>
3200
3201 * include/ffi.h.in: Remove RCS ident string.
3202 * include/ffi_mips.h: Remove RCS ident string.
3203 * src/debug.c: Remove RCS ident string.
3204 * src/ffitest.c: Remove RCS ident string.
3205 * src/prep_cif.c: Remove RCS ident string.
3206 * src/types.c: Remove RCS ident string.
3207 * src/alpha/ffi.c: Remove RCS ident string.
3208 * src/alpha/osf.S: Remove RCS ident string.
3209 * src/arm/ffi.c: Remove RCS ident string.
3210 * src/arm/sysv.S: Remove RCS ident string.
3211 * src/mips/ffi.c: Remove RCS ident string.
3212 * src/mips/n32.S: Remove RCS ident string.
3213 * src/mips/o32.S: Remove RCS ident string.
3214 * src/sparc/ffi.c: Remove RCS ident string.
3215 * src/sparc/v8.S: Remove RCS ident string.
3216 * src/sparc/v9.S: Remove RCS ident string.
3217 * src/x86/ffi.c: Remove RCS ident string.
3218 * src/x86/sysv.S: Remove RCS ident string.
3219
8cb0f8ea 32202001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3221
3222 * include/ffi.h.in: Change sourceware.cygnus.com references to
3223 gcc.gnu.org.
3224
18f704a9 32252000-12-09 Richard Henderson <rth@redhat.com>
3226
3227 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3228 (ffi_closure_osf_inner): Index rather than increment avalue
3229 and arg_types. Give ffi_closure_osf the raw return value type.
3230 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3231 promotion.
3232
32332000-12-07 Richard Henderson <rth@redhat.com>
fdf04128 3234
3235 * src/raw_api.c (ffi_translate_args): Fix typo.
3236 (ffi_prep_closure): Likewise.
3237
3238 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3239 FFI_TRAMPOLINE_SIZE.
3240 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3241 cif->bytes for new ffi_call_osf implementation.
3242 (ffi_prep_args): Absorb into ...
3243 (ffi_call): ... here. Do all stack allocation here and
3244 avoid a callback function.
3245 (ffi_prep_closure, ffi_closure_osf_inner): New.
3246 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3247 (ffi_closure_osf): New.
3248
652f285b 32492000-09-10 Alexandre Oliva <aoliva@redhat.com>
3250
4bfa769b 3251 * config.guess, config.sub, install-sh: Removed.
3252 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3253 * Makefile.in: Rebuilt.
3254
652f285b 3255 * acinclude.m4: Include libtool macros from the top level.
3256 * aclocal.m4, configure: Rebuilt.
3257
6f52caf7 32582000-08-22 Alexandre Oliva <aoliva@redhat.com>
3259
3260 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3261 * configure: Rebuilt.
3262
d03e800c 32632000-05-11 Scott Bambrough <scottb@netwinder.org>
3264
805e22b2 3265 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3266 memory correctly. Use conditional instructions, not branches where
d03e800c 3267 possible.
3268
f3e32e0b 32692000-05-04 Tom Tromey <tromey@cygnus.com>
3270
3271 * configure: Rebuilt.
3272 * configure.in: Match `arm*-*-linux-*'.
3273 From Chris Dornan <cdornan@arm.com>.
3274
b23e0d76 32752000-04-28 Jakub Jelinek <jakub@redhat.com>
3276
3277 * Makefile.am (SUBDIRS): Define.
3278 (AM_MAKEFLAGS): Likewise.
3279 (Multilib support.): Add section.
3280 * Makefile.in: Rebuilt.
3281 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
968b0d1e 3282 New variables. Set for gcc using -print-multi-lib. Export them
3283 to libtool.
3284 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3285 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3286 for -shared links.
3287 (extra_compiler_flags_value, extra_compiler_flags): Check these
3288 for extra compiler options which need to be passed down in
3289 compiler_flags.
b23e0d76 3290
2c9b297c 32912000-04-16 Anthony Green <green@redhat.com>
3292
3293 * configure: Rebuilt.
3294 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3295
d7a35c6c 32962000-04-14 Jakub Jelinek <jakub@redhat.com>
3297
968b0d1e 3298 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
d7a35c6c 3299 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3300 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3301 Replace all void * sizeofs with sizeof(int).
3302 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3303 different than DOUBLE.
3304 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3305 (ffi_prep_args_v9): New function.
3306 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3307 (ffi_V9_return_struct): New function.
3308 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3309 32bit code (not yet cross-arch calls).
3310 * src/sparc/v8.S: Add struct return delay nop.
3311 Handle long long.
3312 * src/sparc/v9.S: New file.
3313 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3314 is used on sparc64 only for structures larger than 32 bytes.
3315 Pass by reference for structures is done for structure arguments
3316 larger than 16 bytes.
3317 * src/ffitest.c (main): Use 64bit rint on sparc64.
3318 Run long long tests on sparc.
3319 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3320 sparc64.
3321 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3322 on sparc64.
3323 * configure.in (sparc-*-linux*): New supported target.
3324 (sparc64-*-linux*): Likewise.
3325 * configure: Rebuilt.
3326 * Makefile.am: Add v9.S to SPARC files.
3327 * Makefile.in: Likewise.
3328 (LINK): Surround $(CCLD) into double quotes, so that multilib
3329 compiles work correctly.
3330
800d12c7 33312000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3332
3333 * configure: Rebuilt.
3334 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3335 proposed by Bryce McKinlay.
3336
c44e8808 33372000-03-20 Tom Tromey <tromey@cygnus.com>
3338
3339 * Makefile.in: Hand edit for java_raw_api.lo.
3340
f764cb60 33412000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3342
3343 * config.guess, config.sub: Update from the gcc tree.
3344 Fix for PR libgcj/168.
3345
3b43a26e 33462000-03-03 Tom Tromey <tromey@cygnus.com>
3347
96215ab3 3348 * Makefile.in: Fixed ia64 by hand.
3349
3b43a26e 3350 * configure: Rebuilt.
3351 * configure.in (--enable-multilib): New option.
3352 (libffi_basedir): New subst.
3353 (AC_OUTPUT): Added multilib code.
3354
03ae52bb 33552000-03-02 Tom Tromey <tromey@cygnus.com>
3356
3357 * Makefile.in: Rebuilt.
3358 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3359 directory name.
3360
28c07da5 33612000-02-25 Hans Boehm <boehm@acm.org>
3362
3363 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3364 files.
3365 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3366 list.
3367 (ffi_prep_raw_closure): Use ffi_translate_args, not
3368 ffi_closure_translate.
3369 * src/java_raw_api.c: New file.
3370 * src/ffitest.c (closure_test_fn): New function.
3371 (main): Define `rint' as long long on IA64. Added new test when
3372 FFI_CLOSURES is defined.
3373 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3374 (ffi_abi): Recognize IA64.
3375 (ffi_raw): Added `flt' field.
3376 Added "Java raw API" code.
3377 * configure.in: Recognize ia64.
3378 * Makefile.am (TARGET_SRC_IA64): New macro.
3379 (libffi_la_common_SOURCES): Added java_raw_api.c.
3380 (libffi_la_SOURCES): Define in IA64 case.
3381
0bcdf5c5 33822000-01-04 Tom Tromey <tromey@cygnus.com>
3383
3384 * Makefile.in: Rebuilt with newer automake.
3385
e9de8c8b 33861999-12-31 Tom Tromey <tromey@cygnus.com>
3387
3388 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3389
20e58c02 33901999-09-01 Tom Tromey <tromey@cygnus.com>
3391
3392 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3393 undefs.
3394 * fficonfig.h.in: Rebuilt.
3395 * configure: Rebuilt.
3396 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3397 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3398 * acconfig.h: Don't #undef PACKAGE or VERSION.
3399
47c5fab6 34001999-08-09 Anthony Green <green@cygnus.com>
3401
3402 * include/ffi.h.in: Try to work around messy header problem
3403 with PACKAGE and VERSION.
3404
3405 * configure: Rebuilt.
3406 * configure.in: Change version to 2.00-beta.
3407
968b0d1e 3408 * fficonfig.h.in: Rebuilt.
47c5fab6 3409 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3410
3411 * src/x86/ffi.c (ffi_raw_call): Rename.
3412
34131999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3414
3415 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3416 (ffi_prep_incoming_args_SYSV): Ditto.
3417 (ffi_prep_closure): Ditto.
3418 (ffi_closure_raw_SYSV): Ditto.
3419 (ffi_prep_raw_closure): More ditto.
3420 (ffi_call_raw): Final ditto.
3421
3422 * include/ffi.h.in: Add definitions for closure and raw API.
3423
3424 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
805e22b2 3425 FFI_TYPE_UINT64.
47c5fab6 3426
3427 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3428
3429 * src/raw_api.c: New file.
3430
3431 * include/ffi.h.in (ffi_raw): New type.
3432 (UINT_ARG, SINT_ARG): New defines.
3433 (ffi_closure, ffi_raw_closure): New types.
3434 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
805e22b2 3435
47c5fab6 3436 * configure.in: Add check for endianness and sizeof void*.
3437
3438 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3439 instead of directly.
3440
3441 * configure: Rebuilt.
3442
3443Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3444
3445 * configure.in: Add x86 and powerpc BeOS configurations.
3446 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3447
34481999-05-09 Anthony Green <green@cygnus.com>
3449
3450 * configure.in: Add warning about this being beta code.
3451 Remove src/Makefile.am from the picture.
3452 * configure: Rebuilt.
3453
3454 * Makefile.am: Move logic from src/Makefile.am. Add changes
3455 to support libffi as a target library.
3456 * Makefile.in: Rebuilt.
3457
3458 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3459 Upgraded to new autoconf, automake, libtool.
3460
3461 * README: Tweaks.
3462
3463 * LICENSE: Update copyright date.
3464
3465 * src/Makefile.am, src/Makefile.in: Removed.
3466
34671998-11-29 Anthony Green <green@cygnus.com>
3468
3469 * include/ChangeLog: Removed.
3470 * src/ChangeLog: Removed.
3471 * src/mips/ChangeLog: Removed.
3472 * src/sparc/ChangeLog: Remboved.
3473 * src/x86/ChangeLog: Removed.
3474
3475 * ChangeLog.v1: Created.
Mirror of https://gcc.gnu.org/git/gcc.git