]> 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
re PR libffi/26048 (libffi doesn't build on Solaris 10/x86 with native assembler)
[thirdparty/gcc.git] / libffi / ChangeLog
CommitLineData
4d3d4078
RO		 12008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2
3 PR libffi/26048
4 * configure.ac (HAVE_AS_X86_PCREL): New test.
5 * configure: Regenerate.
6 * fficonfig.h.in: Regenerate.
7 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
8 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
9 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
10 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
11 * src/x86/unix64.S (.Lstore_table): Move to .text section.
12 (.Lload_table): Likewise.
13 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
14
baafc534
RW		 152008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
16
17 * configure: Regenerate.
18
da17a98b
EB		 192008-11-21 Eric Botcazou <ebotcazou@adacore.com>
20
21 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
22 signed/unsigned int8/16 return values.
23 * src/sparc/v8.S (ffi_call_v8): Likewise.
24 (ffi_closure_v8): Likewise.
25
80c6c2cb
PG		 262008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
27 Steve Ellcey <sje@cup.hp.com>
28
29 * configure: Regenerate for new libtool.
30 * Makefile.in: Ditto.
31 * include/Makefile.in: Ditto.
32 * aclocal.m4: Ditto.
33
d7f68a67
AT		 342008-08-25 Andreas Tobler <a.tobler@schweiz.org>
35
36 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
37 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
38 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
39 Adjust copyright notice.
40 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
41 register or two register to use for FFI_SYSV structs.
42 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
43 (ffi_closure_helper_SYSV): Fix the return type for
44 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
45 Adjust copyright notice.
46
130f4c50
KK		 472008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
48
49 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
50 int.
51
18c04407
RW		 522008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
53
54 * configure: Regenerate.
55 * include/Makefile.in: Regenerate.
56 * testsuite/Makefile.in: Regenerate.
57
cd985f66
JM		 582008-06-07 Joseph Myers <joseph@codesourcery.com>
59
60 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
61 powerpc-*-beos*): Remove.
62 * configure: Regenerate.
63
7afa92c5
JB		 642008-05-09 Julian Brown <julian@codesourcery.com>
65
66 * Makefile.am (LTLDFLAGS): New.
67 (libffi_la_LDFLAGS): Use above.
68 * Makefile.in: Regenerate.
69
deb984e6
PB		 702008-04-18 Paolo Bonzini <bonzini@gnu.org>
71
72 PR bootstrap/35457
73 * aclocal.m4: Regenerate.
74 * configure: Regenerate.
75
3e9608a5
KK		 762008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
77
78 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
79 * src/sh64/sysv.S: Likewise.
80
feb51530
MK		 812008-03-26 Daniel Jacobowitz <dan@debian.org>
82
83 * src/arm/sysv.S: Fix ARM comment marker.
84
852008-03-26 Jakub Jelinek <jakub@redhat.com>
86
87 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
88 * src/s390/sysv.S: Likewise.
89 * src/powerpc/ppc_closure.S: Likewise.
90 * src/powerpc/sysv.S: Likewise.
91 * src/x86/unix64.S: Likewise.
92 * src/x86/sysv.S: Likewise.
93 * src/sparc/v8.S: Likewise.
94 * src/sparc/v9.S: Likewise.
95 * src/m68k/sysv.S: Likewise.
96 * src/arm/sysv.S: Likewise.
97
38d24731
RW		 982008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
99
100 * aclocal.m4: Regenerate.
101 * configure: Likewise.
102 * Makefile.in: Likewise.
103 * include/Makefile.in: Likewise.
104 * testsuite/Makefile.in: Likewise.
105
2e250fa2
BK		 1062008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
107 Andreas Tobler <a.tobler@schweiz.org>
108
109 * configure.ac: Add amd64-*-freebsd* target.
110 * configure: Regenerate.
111
7f92e242
L		 1122008-01-30 H.J. Lu <hongjiu.lu@intel.com>
113
114 PR libffi/34612
115 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
116 returning struct.
117
118 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
119 tests.
120
95d4f544
DE		 1212008-01-24 David Edelsohn <edelsohn@gnu.org>
122
123 * configure: Regenerate.
124
79ca8fc0
AT		 1252008-01-06 Andreas Tobler <a.tobler@schweiz.org>
126
127 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
128
d7b07cc7
AT		 1292008-01-05 Andreas Tobler <a.tobler@schweiz.org>
130
131 PR testsuite/32843
132 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
133 signed/unsigned int8/16 for X86_DARWIN.
134 Updated copyright info.
135 Handle one and two byte structs with special cif->flags.
136 * src/x86/ffitarget.h: Add special types for one and two byte structs.
137 Updated copyright info.
138 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
139 sysv.S
140 Remove code to pop args from the stack after call.
141 Special-case signed/unsigned for int8/16, one and two byte structs.
142 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
143 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
144 FFI_TYPE_SINT32.
145 Updated copyright info.
146
c94ae2e4
DD		 1472007-12-08 David Daney <ddaney@avtrex.com>
148
149 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
150 SUBU, add with ADDU and use smaller code sequences.
151
3875b6d7 1522007-12-07 David Daney <ddaney@avtrex.com>
44be66a1 153
3875b6d7
DD		 154 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
155 type.
156
4c42b3d8
DD		 1572007-12-06 David Daney <ddaney@avtrex.com>
158
159 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
160 defined.
161 (ffi_java_raw): New typedef.
162 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
163 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
164 ffi_java_raw.
165 (ffi_java_raw_closure) : Same.
166 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
167 parameter types.
168 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
169 FFI_SIZEOF_JAVA_RAW.
170 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
171 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
172 sizeof(ffi_java_raw) for alignment calculations.
173 (ffi_java_ptrarray_to_raw): Same.
174 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
175 if FFI_SIZEOF_JAVA_RAW == 4.
176 (ffi_java_raw_to_rvalue): Same.
177 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
178 (ffi_java_translate_args): Same.
179 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
180 parameter types.
181 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
182
3ef50b62
DD		 1832007-12-06 David Daney <ddaney@avtrex.com>
184
185 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
186 pointer values.
187
16287148
AT		 1882007-12-01 Andreas Tobler <a.tobler@schweiz.org>
189
190 PR libffi/31937
191 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
192 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
193 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
194 set the NUM_FPR_ARG_REGISTERS according to.
195 Add support for potential soft-float support under hard-float
196 architecture.
197 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
198 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
199 to the FFI_LINUX_SOFT_FLOAT ABI.
200 (ffi_prep_cif_machdep): Likewise.
201 (ffi_closure_helper_SYSV): Likewise.
202 * src/powerpc/ppc_closure.S: Make sure not to store float/double
203 on archs where __NO_FPRS__ is true.
204 Add FFI_TYPE_UINT128 support.
205 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
206 Adjust copyright notice.
207
a4b7c9cd
AT		 2082007-11-25 Andreas Tobler <a.tobler@schweiz.org>
209
210 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
211 * include/ffi_common.h: ... here.
212 Update copyright.
213
e84296c6
AT		 2142007-11-17 Andreas Tobler <a.tobler@schweiz.org>
215
216 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
217 * src/powerpc/linux64.S: Likewise.
218 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
219 * testsuite/libffi.call/return_ldl.c: New test.
220
cb642590
AH		 2212007-09-04 <aph@redhat.com>
222
223 * src/arm/sysv.S (UNWIND): New.
224 (Whole file): Conditionally compile unwinder directives.
225 * src/arm/sysv.S: Add unwinder directives.
226
227 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
228 Only treat r0 as a struct address if we're actually returning a
229 struct by address.
230 Only copy the bytes that are actually within a struct.
231 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
232 is returned in r0, not passed by address.
233 (ffi_call): Allocate a word-sized temporary for the case where
234 a composite is returned in r0.
235 (ffi_prep_incoming_args_SYSV): Align as necessary.
236
2372007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
238
239 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
240 directly using the sys_cacheflush syscall.
241
2422007-07-27 Andrew Haley <aph@redhat.com>
243
244 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
245
8d5c5e35
MR		 2462007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
247
248 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
249 * configure.ac: Likewise.
250 * Makefile.in: Regenerate.
251 * include/Makefile.in: Likewise.
252 * testsuite/Makefile.in: Likewise.
253 * configure: Likewise.
254
787b0d19
DD		 2552007-08-24 David Daney <ddaney@avtrex.com>
256
257 * testsuite/libffi.call/return_sl.c: New test.
258
dfbb3e36
DD		 2592007-08-10 David Daney <ddaney@avtrex.com>
260
261 * testsuite/libffi.call/cls_multi_ushort.c,
262 testsuite/libffi.call/cls_align_uint16.c,
263 testsuite/libffi.call/nested_struct1.c,
264 testsuite/libffi.call/nested_struct3.c,
265 testsuite/libffi.call/cls_7_1_byte.c,
266 testsuite/libffi.call/nested_struct5.c,
267 testsuite/libffi.call/cls_double.c,
268 testsuite/libffi.call/nested_struct7.c,
269 testsuite/libffi.call/cls_sint.c,
270 testsuite/libffi.call/nested_struct9.c,
271 testsuite/libffi.call/cls_20byte1.c,
272 testsuite/libffi.call/cls_multi_sshortchar.c,
273 testsuite/libffi.call/cls_align_sint64.c,
274 testsuite/libffi.call/cls_3byte2.c,
275 testsuite/libffi.call/cls_multi_schar.c,
276 testsuite/libffi.call/cls_multi_uchar.c,
277 testsuite/libffi.call/cls_19byte.c,
278 testsuite/libffi.call/cls_9byte1.c,
279 testsuite/libffi.call/cls_align_float.c,
280 testsuite/libffi.call/closure_fn1.c,
281 testsuite/libffi.call/problem1.c,
282 testsuite/libffi.call/closure_fn3.c,
283 testsuite/libffi.call/cls_sshort.c,
284 testsuite/libffi.call/closure_fn5.c,
285 testsuite/libffi.call/cls_align_double.c,
286 testsuite/libffi.call/nested_struct.c,
287 testsuite/libffi.call/cls_2byte.c,
288 testsuite/libffi.call/nested_struct10.c,
289 testsuite/libffi.call/cls_4byte.c,
290 testsuite/libffi.call/cls_6byte.c,
291 testsuite/libffi.call/cls_8byte.c,
292 testsuite/libffi.call/cls_multi_sshort.c,
293 testsuite/libffi.call/cls_align_sint16.c,
294 testsuite/libffi.call/cls_align_uint32.c,
295 testsuite/libffi.call/cls_20byte.c,
296 testsuite/libffi.call/cls_float.c,
297 testsuite/libffi.call/nested_struct2.c,
298 testsuite/libffi.call/cls_5_1_byte.c,
299 testsuite/libffi.call/nested_struct4.c,
300 testsuite/libffi.call/cls_24byte.c,
301 testsuite/libffi.call/nested_struct6.c,
302 testsuite/libffi.call/cls_64byte.c,
303 testsuite/libffi.call/nested_struct8.c,
304 testsuite/libffi.call/cls_uint.c,
305 testsuite/libffi.call/cls_multi_ushortchar.c,
306 testsuite/libffi.call/cls_schar.c,
307 testsuite/libffi.call/cls_uchar.c,
308 testsuite/libffi.call/cls_align_uint64.c,
309 testsuite/libffi.call/cls_ulonglong.c,
310 testsuite/libffi.call/cls_align_longdouble.c,
311 testsuite/libffi.call/cls_1_1byte.c,
312 testsuite/libffi.call/cls_12byte.c,
313 testsuite/libffi.call/cls_3_1byte.c,
314 testsuite/libffi.call/cls_3byte1.c,
315 testsuite/libffi.call/cls_4_1byte.c,
316 testsuite/libffi.call/cls_6_1_byte.c,
317 testsuite/libffi.call/cls_16byte.c,
318 testsuite/libffi.call/cls_18byte.c,
319 testsuite/libffi.call/closure_fn0.c,
320 testsuite/libffi.call/cls_9byte2.c,
321 testsuite/libffi.call/closure_fn2.c,
322 testsuite/libffi.call/closure_fn4.c,
323 testsuite/libffi.call/cls_ushort.c,
324 testsuite/libffi.call/closure_fn6.c,
325 testsuite/libffi.call/cls_5byte.c,
326 testsuite/libffi.call/cls_align_pointer.c,
327 testsuite/libffi.call/cls_7byte.c,
328 testsuite/libffi.call/cls_align_sint32.c,
329 testsuite/libffi.special/unwindtest_ffi_call.cc,
330 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
331
89d9d98a
DD		 3322007-08-10 David Daney <ddaney@avtrex.com>
333
334 PR libffi/28313
335 * configure.ac: Don't treat mips64 as a special case.
336 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
337 * configure: Regenerate
338 * Makefile.in: Ditto.
339 * fficonfig.h.in: Ditto.
340 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
341 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
342 (FFI_DEFAULT_ABI): Set for n64 case.
343 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
344 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
345 (ffi_closure_N32): New function.
346 (.eh_frame): New section
347 * src/mips/o32.S: Clean up comments.
348 (ffi_closure_O32): Pass ffi_closure parameter in $12.
349 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
350 _MIPS_SIM == _ABIN32 throughout.
351 (FFI_MIPS_STOP_HERE): New, use in place of
352 ffi_stop_here.
353 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
354 to support n32/n64 ABIs.
355 (calc_n32_struct_flags): Rewrite.
356 (calc_n32_return_struct_flags): Remove unused variable. Reverse
357 position of flag bits.
358 (ffi_prep_cif_machdep): Rewrite n32 portion.
359 (ffi_call): Enable for n64. Add special handling for small structure
360 return values.
361 (ffi_prep_closure_loc): Add n32 and n64 support.
362 (ffi_closure_mips_inner_O32): Add cast to silence warning.
363 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
364
d23869d4
DD		 3652007-08-08 David Daney <ddaney@avtrex.com>
366
367 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
368 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
369 specifiers.
370 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
371 * testsuite/libffi.call/cls_sint.c (main): Ditto.
372 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
373 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
374 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
375 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
376 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
377 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
378 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
379 * testsuite/libffi.call/return_fl3.c (main): Ditto.
380 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
381 * testsuite/libffi.call/nested_struct.c (main): Ditto.
382 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
383 * testsuite/libffi.call/return_ll1.c (main): Ditto.
384 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
385 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
386 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
387 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
388 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
389 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
390 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
391 * testsuite/libffi.call/cls_uint.c (main): Ditto.
392 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
393 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
394 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
395 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
396 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
397 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
398 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
399 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
400 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
401
536e0a8f
AH		 4022007-08-07 Andrew Haley <aph@redhat.com>
403
404 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
405 checkin.
406
7665c9be
AH		 4072007-08-06 Andrew Haley <aph@redhat.com>
408
409 PR testsuite/32843
410 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
411 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
412 FFI_TYPE_SINT32.
413
58e11497
DD		 4142007-08-02 David Daney <ddaney@avtrex.com>
415
416 * testsuite/libffi.call/return_ul.c (main): Define return type as
417 ffi_arg. Use proper printf conversion specifier.
418
387cfbba
AH		 4192007-07-30 Andrew Haley <aph@redhat.com>
420
421 PR testsuite/32843
422 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
423 signed/unsigned int8/16.
424 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
425 Use a jump table.
426 Remove code to pop args from the stack after call.
427 Special-case signed/unsigned int8/16.
428 * testsuite/libffi.call/return_sc.c (main): Revert.
429
0ee0208e
RG		 4302007-07-26 Richard Guenther <rguenther@suse.de>
431
432 PR testsuite/32843
433 * testsuite/libffi.call/return_sc.c (main): Verify call
434 result as signed char, not ffi_arg.
435
04a49983
RO		 4362007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
437
438 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
439 * configure: Regenerate.
440
433e6a8c
DD		 4412007-07-11 David Daney <ddaney@avtrex.com>
442
443 * src/mips/ffi.c: Don't include sys/cachectl.h.
444 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
445 cacheflush().
446
0959e2b8
AH		 4472007-05-18 Aurelien Jarno <aurelien@aurel32.net>
448
449 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
450 from (ffi_prep_closure): ... this.
451 (FFI_INIT_TRAMPOLINE): Adjust.
452
4532005-12-31 Phil Blundell <pb@reciva.com>
454
455 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
456 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
457 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
458 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
459 (FFI_CLOSURES): Enable closure support.
460
4612007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
462
463 * testsuite/libffi.call/cls_multi_ushort.c,
464 testsuite/libffi.call/cls_align_uint16.c,
465 testsuite/libffi.call/nested_struct1.c,
466 testsuite/libffi.call/nested_struct3.c,
467 testsuite/libffi.call/cls_7_1_byte.c,
468 testsuite/libffi.call/cls_double.c,
469 testsuite/libffi.call/nested_struct5.c,
470 testsuite/libffi.call/nested_struct7.c,
471 testsuite/libffi.call/cls_sint.c,
472 testsuite/libffi.call/nested_struct9.c,
473 testsuite/libffi.call/cls_20byte1.c,
474 testsuite/libffi.call/cls_multi_sshortchar.c,
475 testsuite/libffi.call/cls_align_sint64.c,
476 testsuite/libffi.call/cls_3byte2.c,
477 testsuite/libffi.call/cls_multi_schar.c,
478 testsuite/libffi.call/cls_multi_uchar.c,
479 testsuite/libffi.call/cls_19byte.c,
480 testsuite/libffi.call/cls_9byte1.c,
481 testsuite/libffi.call/cls_align_float.c,
482 testsuite/libffi.call/closure_fn1.c,
483 testsuite/libffi.call/problem1.c,
484 testsuite/libffi.call/closure_fn3.c,
485 testsuite/libffi.call/cls_sshort.c,
486 testsuite/libffi.call/closure_fn5.c,
487 testsuite/libffi.call/cls_align_double.c,
488 testsuite/libffi.call/cls_2byte.c,
489 testsuite/libffi.call/nested_struct.c,
490 testsuite/libffi.call/nested_struct10.c,
491 testsuite/libffi.call/cls_4byte.c,
492 testsuite/libffi.call/cls_6byte.c,
493 testsuite/libffi.call/cls_8byte.c,
494 testsuite/libffi.call/cls_multi_sshort.c,
495 testsuite/libffi.call/cls_align_uint32.c,
496 testsuite/libffi.call/cls_align_sint16.c,
497 testsuite/libffi.call/cls_float.c,
498 testsuite/libffi.call/cls_20byte.c,
499 testsuite/libffi.call/cls_5_1_byte.c,
500 testsuite/libffi.call/nested_struct2.c,
501 testsuite/libffi.call/cls_24byte.c,
502 testsuite/libffi.call/nested_struct4.c,
503 testsuite/libffi.call/nested_struct6.c,
504 testsuite/libffi.call/cls_64byte.c,
505 testsuite/libffi.call/nested_struct8.c,
506 testsuite/libffi.call/cls_uint.c,
507 testsuite/libffi.call/cls_multi_ushortchar.c,
508 testsuite/libffi.call/cls_schar.c,
509 testsuite/libffi.call/cls_uchar.c,
510 testsuite/libffi.call/cls_align_uint64.c,
511 testsuite/libffi.call/cls_ulonglong.c,
512 testsuite/libffi.call/cls_align_longdouble.c,
513 testsuite/libffi.call/cls_1_1byte.c,
514 testsuite/libffi.call/cls_12byte.c,
515 testsuite/libffi.call/cls_3_1byte.c,
516 testsuite/libffi.call/cls_3byte1.c,
517 testsuite/libffi.call/cls_4_1byte.c,
518 testsuite/libffi.call/cls_6_1_byte.c,
519 testsuite/libffi.call/cls_16byte.c,
520 testsuite/libffi.call/cls_18byte.c,
521 testsuite/libffi.call/closure_fn0.c,
522 testsuite/libffi.call/cls_9byte2.c,
523 testsuite/libffi.call/closure_fn2.c,
524 testsuite/libffi.call/closure_fn4.c,
525 testsuite/libffi.call/cls_ushort.c,
526 testsuite/libffi.call/closure_fn6.c,
527 testsuite/libffi.call/cls_5byte.c,
528 testsuite/libffi.call/cls_align_pointer.c,
529 testsuite/libffi.call/cls_7byte.c,
530 testsuite/libffi.call/cls_align_sint32.c,
531 testsuite/libffi.special/unwindtest_ffi_call.cc,
532 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
533
12a27363
L		 5342007-07-05 H.J. Lu <hongjiu.lu@intel.com>
535
536 * aclocal.m4: Regenerated.
537
82a6cadf
PB		 5382007-06-02 Paolo Bonzini <bonzini@gnu.org>
539
540 * configure: Regenerate.
541
1cf3d07d
SE		 5422007-05-23 Steve Ellcey <sje@cup.hp.com>
543
544 * Makefile.in: Regenerate.
545 * configure: Regenerate.
546 * aclocal.m4: Regenerate.
547 * include/Makefile.in: Regenerate.
548 * testsuite/Makefile.in: Regenerate.
549
5635c26e
RZ		 5502007-05-10 Roman Zippel <zippel@linux-m68k.org>
551
552 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
553 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
554 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
555 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
556 (FFI_CLOSURES): Enable closure support.
557
e6fbf387
RZ		 5582007-05-10 Roman Zippel <zippel@linux-m68k.org>
559
560 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
561 * configure: Regenerate.
562 * fficonfig.h.in: Regenerate.
563 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
564 CFI_OFFSET,CFI_DEF_CFA): New macros.
565 (ffi_call_SYSV): Add callframe annotation.
566
b5296e52
RZ		 5672007-05-10 Roman Zippel <zippel@linux-m68k.org>
568
569 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
570 numerous test suite failures.
571 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
572
bd61d708
PB		 5732007-04-11 Paolo Bonzini <bonzini@gnu.org>
574
575 * Makefile.am (EXTRA_DIST): Bring up to date.
576 * Makefile.in: Regenerate.
577 * src/frv/eabi.S: Remove RCS keyword.
578
bf8da5fc
RH		 5792007-04-06 Richard Henderson <rth@redhat.com>
580
581 * configure.ac: Tidy target case.
582 (HAVE_LONG_DOUBLE): Allow the target to override.
583 * configure: Regenerate.
584 * include/ffi.h.in: Don't define ffi_type_foo if
585 LIBFFI_HIDE_BASIC_TYPES is defined.
586 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
587 to ffi_type_double.
588 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
589 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
590 (ffi_type_longdouble): Special case for Alpha. Don't define
591 if long double == double.
592
593 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
594 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
595 (ffi_call, ffi_closure_osf_inner): Likewise.
596 (ffi_closure_osf_inner): Likewise. Mark hidden.
597 (ffi_call_osf, ffi_closure_osf): Mark hidden.
598 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
599 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
600 (load_table): Handle 128-bit long double.
601
602 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
603
8669fed1
TT		 6042007-04-06 Tom Tromey <tromey@redhat.com>
605
606 PR libffi/31491:
607 * README: Fixed bug in example.
608
f0920e6c
JJ		 6092007-04-03 Jakub Jelinek <jakub@redhat.com>
610
611 * src/closures.c: Include sys/statfs.h.
612 (_GNU_SOURCE): Define on Linux.
613 (FFI_MMAP_EXEC_SELINUX): Define.
614 (selinux_enabled): New variable.
615 (selinux_enabled_check): New function.
616 (is_selinux_enabled): Define.
617 (dlmmap): Use it.
618
34a6ccda
UB		 6192007-03-24 Uros Bizjak <ubizjak@gmail.com>
620
621 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
622 Use 'volatile float sum' to create sum of floats to avoid false
623 negative due to excess precision on ix86 targets.
624 (main): Ditto.
625
8b9cf39e
AO		 6262007-03-08 Alexandre Oliva <aoliva@redhat.com>
627
628 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
629 patch.
630 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
631
18fa3240
AO		 6322007-03-07 Alexandre Oliva <aoliva@redhat.com>
633
634 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
635 (ffi_prep_closure_loc): New.
636 (ffi_prep_raw_closure_loc): New.
637 (ffi_prep_java_raw_closure_loc): New.
638 * src/closures.c: New file.
639 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
640 Replace sflags with exec_offset.
641 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
642 sub_segment_exec_offset): New macros.
643 (get_segment_flags, set_segment_flags, check_segment_merge): New
644 macros.
645 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
646 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
647 destroy_mspace): Use new macros.
648 (sys_alloc): Silence warning.
649 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
650 * Makefile.in: Rebuilt.
651 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
652 terms of ffi_prep_closure_loc.
653 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
654 from...
655 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
656 renamed version.
657 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
658 adjusted from...
659 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
660 the renamed version.
661 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
662 (ffi_prep_closure): ... this.
663 * src/pa/ffi.c: Likewise.
664 * src/cris/ffi.c: Likewise. Adjust.
665 * src/frv/ffi.c: Likewise.
666 * src/ia64/ffi.c: Likewise.
667 * src/mips/ffi.c: Likewise.
668 * src/powerpc/ffi_darwin.c: Likewise.
669 * src/s390/ffi.c: Likewise.
670 * src/sh/ffi.c: Likewise.
671 * src/sh64/ffi.c: Likewise.
672 * src/sparc/ffi.c: Likewise.
673 * src/x86/ffi64.c: Likewise.
674 * src/x86/ffi.c: Likewise.
675 (FFI_INIT_TRAMPOLINE): Adjust.
676 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
677 (ffi_prep_raw_closure): ... this.
678 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
679 (ffi_prep_closure): ... this.
680 (flush_icache): Adjust.
681
dd778331
AO		 6822007-03-07 Alexandre Oliva <aoliva@redhat.com>
683
684 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
685 Lea's malloc.
686
b5f8d49b
BM		 6872007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
688
689 * Makefile.am: Add dummy install-pdf target.
690 * Makefile.in: Regenerate
691
7a23933b
AK		 6922007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
693
694 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
695 ffi_closure_helper_SYSV): Add long double handling.
696
8aaf4681
JJ		 6972007-02-02 Jakub Jelinek <jakub@redhat.com>
698
699 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
700 immediately after bctrl instruction.
701
0eb5e2b9
AO		 7022007-01-18 Alexandre Oliva <aoliva@redhat.com>
703
704 * Makefile.am (all-recursive, install-recursive,
705 mostlyclean-recursive, clean-recursive, distclean-recursive,
706 maintainer-clean-recursive): Add missing targets.
707 * Makefile.in: Rebuilt.
708
f4f7486b
AT		 7092006-12-14 Andreas Tobler <a.tobler@schweiz.org>
710
711 * configure.ac: Add TARGET for x86_64-*-darwin*.
712 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
713 for X86_DARWIN.
714 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
715 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
716 * configure: Regenerate.
717 * Makefile.in: Regenerate.
718 * include/Makefile.in: Regenerate.
719 * testsuite/Makefile.in: Regenerate.
720 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
721 ffi_call only.
722
3d5eafe2
AT		 7232006-12-13 Andreas Tobler <a.tobler@schweiz.org>
724
725 * aclocal.m4: Regenerate with aclocal -I .. as written in the
726 Makefile.am.
727
d5523ce1 7282006-10-31 Geoffrey Keating <geoffk@apple.com>
58182de3
GK		 729
730 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
731 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
732 Darwin.
733 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
734 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
735
a3bd50cc
PB		 7362006-10-10 Paolo Bonzini <bonzini@gnu.org>
737 Sandro Tolaini <tolaini@libero.it>
738
739 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
740 conditional.
741 * configure: Regenerated.
742 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
743 (EXTRA_DIST): Add src/x86/darwin.S.
744 * Makefile.in: Regenerated.
745 * include/Makefile.in: Regenerated.
746 * testsuite/Makefile.in: Regenerated.
747
748 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
749 X86_WIN32, and additionally align stack to 16 bytes.
750 * src/x86/darwin.S: New, based on sysv.S.
751 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
752
b095810f
DD		 7532006-09-12 David Daney <ddaney@avtrex.com>
754
755 PR libffi/23935
756 * include/Makefile.am: Install both ffi.h and ffitarget.h in
757 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
758 * aclocal.m4: Regenerated for automake 1.9.6.
759 * Makefile.in: Regenerated.
760 * include/Makefile.in: Regenerated.
761 * testsuite/Makefile.in: Regenerated.
762
110b2db4
AT		 7632006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
764
765 * include/ffi_common.h (struct): Revert accidental commit.
766
8e5fde87
AT		 7672006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
768
769 * include/ffi_common.h: Remove lint directives.
770 * include/ffi.h.in: Likewise.
771
cb4132fe
TS		 7722006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
773
774 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
775 for 32-bit architectures.
776 * testsuite/libffi.call/return_ul.c: New test case.
777
0d24c3fa
DD		 7782006-07-19 David Daney <ddaney@avtrex.com>
779
780 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
781 xfail remains for mips64.
782
8665824c
CD		 7832006-05-23 Carlos O'Donell <carlos@codesourcery.com>
784
785 * Makefile.am: Add install-html target. Add install-html to .PHONY
786 * Makefile.in: Regenerate.
787 * aclocal.m4: Regenerate.
788 * include/Makefile.in: Regenerate.
789 * testsuite/Makefile.in: Regenerate.
790
84e2e370
JDA		 7912006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
792
793 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
794 stack slot.
795
ac6ed182
AT		 7962006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
797
798 * README: Remove notice about 'Crazy Comments'.
799 * src/debug.c: Remove lint directives. Cleanup white spaces.
800 * src/java_raw_api.c: Likewise.
801 * src/prep_cif.c: Likewise.
802 * src/raw_api.c: Likewise.
803 * src/ffitest.c: Delete. No longer needed, all test cases migrated
804 to the testsuite.
805 * src/arm/ffi.c: Remove lint directives.
806 * src/m32r/ffi.c: Likewise.
807 * src/pa/ffi.c: Likewise.
808 * src/powerpc/ffi.c: Likewise.
809 * src/powerpc/ffi_darwin.c: Likewise.
810 * src/sh/ffi.c: Likewise.
811 * src/sh64/ffi.c: Likewise.
812 * src/x86/ffi.c: Likewise.
813 * testsuite/libffi.call/float2.c: Likewise.
814 * testsuite/libffi.call/promotion.c: Likewise.
815 * testsuite/libffi.call/struct1.c: Likewise.
816
b7211585
AT		 8172006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
818
819 * src/pa/hpux32.S: Correct unwind offset calculation for
820 ffi_closure_pa32.
821 * src/pa/linux.S: Likewise.
822
86066f9b
JW		 8232006-04-12 James E Wilson <wilson@specifix.com>
824
825 PR libgcj/26483
826 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
827 (hfa_type_load): Call stf_spill.
828 (hfa_type_store): Call ldf_fill.
829 (ffi_call): Adjust calls to above routines. Add local temps for
830 macro result.
b7211585 831
615baed7
MK		 8322006-04-10 Matthias Klose <doko@debian.org>
833
834 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
835 directory names containing underscores.
836
8e7a2b58
JW		 8372006-04-07 James E Wilson <wilson@specifix.com>
838
839 * testsuite/libffi.call/float4.c: New testcase.
840
e6c300ef
JDA		 8412006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
842 Andreas Tobler <a.tobler@schweiz.ch>
843
844 * Makefile.am: Add PA_HPUX port.
845 * Makefile.in: Regenerate.
846 * include/Makefile.in: Likewise.
847 * testsuite/Makefile.in: Likewise.
848 * configure.ac: Add PA_HPUX rules.
849 * configure: Regenerate.
850 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
851 Add PA_HPUX and PA64_HPUX.
852 Rename FFI_LINUX ABI to FFI_PA32 ABI.
853 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
854 (FFI_TYPE_SMALL_STRUCT2): Define.
855 (FFI_TYPE_SMALL_STRUCT4): Likewise.
856 (FFI_TYPE_SMALL_STRUCT8): Likewise.
857 (FFI_TYPE_SMALL_STRUCT3): Redefine.
858 (FFI_TYPE_SMALL_STRUCT5): Likewise.
859 (FFI_TYPE_SMALL_STRUCT6): Likewise.
860 (FFI_TYPE_SMALL_STRUCT7): Likewise.
861 * src/pa/ffi.c (ROUND_DOWN): Delete.
862 (fldw, fstw, fldd, fstd): Use '__asm__'.
863 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
864 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
865 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
866 Simplify incrementing of stack slot variable. Change type of local
867 'n' to unsigned int.
868 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
869 double on PA_HPUX.
870 (ffi_prep_cif_machdep): Likewise.
871 (ffi_call): Likewise.
872 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
873 return type to ffi_status. Simplify incrementing of stack slot
874 variable. Only copy floating point argument registers when PA_LINUX
875 is true. Reformat debug statement.
876 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
877 FFI_TYPE_SMALL_STRUCT8.
878 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
879 declaration.
880 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
881 Add nops to cache flush. Add trampoline for PA_HPUX.
882 * src/pa/hpux32.S: New file.
883 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
884 ffi_prep_args_LINUX to ffi_prep_args_pa32.
885 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
886 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
887 argument type checks so that common argument types appear first.
888 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
889 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
890
75b8b1be
AM		 8912006-03-24 Alan Modra <amodra@bigpond.net.au>
892
893 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
894 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
895 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
896 FFI_TYPE_LONGDOUBLE.
897 (ffi_prep_args64): Assert using IBM extended double.
898 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
899 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
900 (ffi_call): Handle FFI_LINUX.
901 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
902 gpr3 return pointer as for struct return. Handle FFI_LINUX
903 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
904 unnecessarily.
905 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
906 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
907 Don't use r6 as pointer to results, instead use sp offset. Don't
908 make a special call to load lr with case table address, instead
909 use offset from previous call.
910 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
911 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
912 return.
913
697701ad
KK		 9142006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
915
916 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
917 passed with FP registers correctly.
918 (ffi_closure_helper_SYSV): Likewise.
919 * src/sh64/sysv.S: Likewise.
920
b295c471
AT		 9212006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
922
923 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
924 args and userdata unused.
925 (closure_test_fn1): Mark cif and userdata unused.
926 (main): Remove unused res.
927
9fdeb13b
AT		 9282006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
929
930 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
931 -O2, -O3, -Os and the warning flags -W -Wall.
932 * testsuite/libffi.special/special.exp: Likewise.
933 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
934 unused parameter unused for gcc or else do nothing.
935 * testsuite/libffi.special/ffitestcxx.h: Likewise.
936 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
937 and userdata unused.
938 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
939 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
940 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
941 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
942 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
943 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
944 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
945 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
946 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
947 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
948 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
949 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
950 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
951 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
952 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
953 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
954 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
955 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
956 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
957 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
958 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
959 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
960 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
961 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
962 Likewise.
963 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
964 Likewise.
965 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
966 Likewise.
967 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
968 void* to avoid compiler warning.
969 (main): Likewise.
970 (cls_struct_align_gn): Mark cif and userdata unused.
971 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
972 Likewise.
973 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
974 Likewise.
975 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
976 Likewise.
977 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
978 Likewise.
979 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
980 Likewise.
981 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
982 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
983 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
984 data unused.
985 (main): Cast res_call to silence gcc.
986 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
987 data unused.
988 (main): Cast res_call to silence gcc.
989 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
990 and data unused.
991 (main): Cast res_call to silence gcc.
992 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
993 data unused.
994 (main): Cast res_call to silence gcc.
995 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
996 data unused.
997 (main): Cast res_call to silence gcc.
998 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
999 and data unused.
1000 (main): Cast res_call to silence gcc.
1001 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1002 userdata unused.
1003 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1004 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1005 userdata unused.
1006 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1007 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1008 userdata unused.
1009 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1010 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1011 userdata unused.
1012 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1013 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1014 userdata unused.
1015 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1016 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1017 and userdata unused.
1018 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1019 userdata unused.
1020 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1021 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1022 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1023 Cleanup white spaces.
1024 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1025 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1026 cif and userdata unused.
1027 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1028 Likewise.
1029 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1030 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1031 formatters to silence gcc.
1032 (B_gn): Mark cif and userdata unused.
1033 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1034 unused.
1035 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1036 (B_gn): Mark cif and userdata unused.
1037 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1038 unused.
1039 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1040 (B_gn): Mark cif and userdata unused.
1041 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1042 unused.
1043 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1044 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1045 * testsuite/libffi.call/problem1.c (stub): Likewise.
1046 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1047 gcc.
1048 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1049 in the last commit for this test case in the test case itself.
1050 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1051 unused.
1052 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1053 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1054 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1055 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1056 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1057 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1058
0a4e2bb5
KK		 10592006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1060
1061 * src/sh/sysv.S: Fix register numbers in the FDE for
1062 ffi_closure_SYSV.
1063
723e5d5a
AT		 10642006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1065
1066 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1067 declaration to avoid a false negative on ix86. See PR323.
1068
e097f887
KK		 10692006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1070
1071 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1072 and cast integer to void * if needed. Update the pointer to
1073 the FP register saved area correctly.
1074
1a4878f7
AT		 10752006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1076
1077 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1078 is fixed.
1079 * testsuite/libffi.call/nested_struct4.c: Likewise.
1080
b47b5e08
AT		 10812006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1082
6429d58f
AT		 1083 * testsuite/libffi.call/return_dbl.c: New test case.
1084 * testsuite/libffi.call/return_dbl1.c: Likewise.
1085 * testsuite/libffi.call/return_dbl2.c: Likewise.
1086 * testsuite/libffi.call/return_fl.c: Likewise.
1087 * testsuite/libffi.call/return_fl1.c: Likewise.
1088 * testsuite/libffi.call/return_fl2.c: Likewise.
1089 * testsuite/libffi.call/return_fl3.c: Likewise.
1090 * testsuite/libffi.call/closure_fn6.c: Likewise.
1091
b47b5e08
AT		 1092 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1093 definition.
1094 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1095 here to be used by other test cases too.
1096
1097 * testsuite/libffi.call/nested_struct10.c: New test case.
1098 * testsuite/libffi.call/nested_struct9.c: Likewise.
1099 * testsuite/libffi.call/nested_struct8.c: Likewise.
1100 * testsuite/libffi.call/nested_struct7.c: Likewise.
1101 * testsuite/libffi.call/nested_struct6.c: Likewise.
1102 * testsuite/libffi.call/nested_struct5.c: Likewise.
1103 * testsuite/libffi.call/nested_struct4.c: Likewise.
1104
9abe31d4
AT		 11052006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1106
1107 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1108 * configure: Rebuilt.
1109
71aa1a70
JJ		 11102006-01-18 Jakub Jelinek <jakub@redhat.com>
1111
1112 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1113 instead do the shifting inline.
1114 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1115 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1116 and 8 byte structs, for the remaining struct sizes don't call
1117 __lshrdi3, instead do the shifting inline.
1118
8a39029d
TS		 11192005-12-07 Thiemo Seufer <ths@networkno.de>
1120
1121 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1122 missing parentheses.
1123 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1124 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1125 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1126 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1127 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1128 FA_0_0_OFF2.
1129 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1130 endianness bugs.
1131 (ffi_prep_closure): Improve trampoline instruction scheduling.
1132 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1133
cd4241aa
AM		 11342005-12-03 Alan Modra <amodra@bigpond.net.au>
1135
1136 * src/powerpc/ffi.c: Formatting.
1137 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1138 (ffi_prep_args64): Likewise.
1139
0363db46
GK		 11402005-09-30 Geoffrey Keating <geoffk@apple.com>
1141
1142 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1143 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1144
d674eb2f
TT		 11452005-09-26 Tom Tromey <tromey@redhat.com>
1146
1147 * testsuite/libffi.call/float1.c (value_type): New typedef.
1148 (CANARY): New define.
1149 (main): Check for result buffer overflow.
1150 * src/powerpc/linux64.S: Handle linux64 long double returns.
1151 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1152 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1153
2d9c5743
AM		 11542005-08-25 Alan Modra <amodra@bigpond.net.au>
1155
1156 PR target/23404
1157 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1158 homed fp args.
1159 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1160
8a42356f
JJ		 11612005-08-11 Jakub Jelinek <jakub@redhat.com>
1162
1163 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1164 (AH_BOTTOM): Add FFI_HIDDEN definition.
1165 * configure: Rebuilt.
1166 * fficonfig.h.in: Rebuilt.
1167 * src/powerpc/ffi.c (hidden): Remove.
1168 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1169 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1170 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1171 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1172 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1173 add FFI_HIDDEN to its prototype.
1174 (ffi_closure_SYSV_inner): New.
1175 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1176 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1177
863c5508
AS		 11782005-08-10 Alfred M. Szmidt <ams@gnu.org>
1179
1180 PR libffi/21819:
1181 * configure: Rebuilt.
1182 * configure.ac: Handle i*86-*-gnu*.
1183
6a26ea58
JJ		 11842005-08-09 Jakub Jelinek <jakub@redhat.com>
1185
1186 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1187 DW_CFA_offset_extended_sf rather than
1188 DW_CFA_GNU_negative_offset_extended.
1189 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1190
583a92e8
ST		 11912005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1192
1193 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1194 on sh3.
1195 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1196 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1197 partially on register.
1198 (ffi_closure_helper_SYSV): Likewise.
1199 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1200
5d0317fc
KK		 12012005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1202
1203 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1204 Remove empty line.
1205 * src/sh64/ffi.c (simple_type): Remove.
1206 (return_type): Handle small structures correctly.
1207 (ffi_prep_args): Likewise.
1208 (ffi_call): Likewise.
1209 (ffi_closure_helper_SYSV): Likewise.
1210 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1211 Emit position independent code if PIC and remove wrong datalabel
1212 prefixes from EH data.
1213
8d9fd5f5
AT		 12142005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1215
1216 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1217 * Makefile.in: Regenerate.
1218 * include/Makefile.in: Likewise.
1219 * testsuite/Makefile.in: Likewise.
1220 * configure.ac: Add POWERPC_FREEBSD rules.
1221 * configure: Regenerate.
1222 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1223 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1224 * src/powerpc/ffi.c: Add flags to handle small structure returns
1225 in ffi_call_SYSV.
1226 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1227 Aka FFI_SYSV.
1228 (ffi_closure_helper_SYSV): Likewise.
1229 * src/powerpc/ppc_closure.S: Add return types for small structures.
1230 * src/powerpc/sysv.S: Add bits to handle small structures for
1231 final SYSV 4 ABI.
1232
93c76f81
AT		 12332005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1234
1235 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1236 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1237 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1238
b65da304
RC		 12392005-07-05 Randolph Chung <tausq@debian.org>
1240
1241 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1242 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1243 structures. Kill compilation warnings.
1244 (ffi_closure_inner_LINUX): Print return values as hex in debug
1245 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1246 Properly handle 5-7 byte structure returns.
1247 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1248 (FFI_TYPE_SMALL_STRUCT2): Remove.
1249 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1250 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1251 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1252 (checksmst1, checksmst2): Remove.
1253 (checksmst3): Optimize handling of 3-byte struct returns.
1254 (checksmst567): Properly handle 5-7 byte struct returns.
1255
2572f1c6
RO		 12562005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1257
1258 PR libgcj/21943
1259 * src/mips/n32.S: Enforce PIC code.
1260 * src/mips/o32.S: Likewise.
1261
cc786582
RO		 12622005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1263
1264 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1265 * configure: Regenerate.
1266
6d0d376c
AM		 12672005-06-01 Alan Modra <amodra@bigpond.net.au>
1268
1269 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1270 to call ffi_closure_helper_SYSV. Append @local instead.
1271 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1272
a21946f5 12732005-05-17 Kelley Cook <kcook@gcc.gnu.org>
a097ef55
KC		 1274
1275 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1276 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1277 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1278 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1279 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1280
f4c767fd
MS		 12812005-05-09 Mike Stump <mrs@apple.com>
1282
1283 * configure: Regenerate.
1284
7e15b5fe
RH		 12852005-05-08 Richard Henderson <rth@redhat.com>
1286
1287 PR libffi/21285
1288 * src/alpha/osf.S: Update unwind into to match code.
1289
d56ea8d9
RH		 12902005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1291 Richard Henderson <rth@redhat.com>
1292
1293 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1294 bit 11 of flags.
1295 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1296 (ffi_prep_closure): Set carry bit if sse-used flag set.
1297 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1298 Only load sse registers if ssecount non-zero.
1299 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1300
13012005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
d948c201
RC		 1302
1303 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1304 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1305 * configure: Regenerate.
1306
fc0ad8d9
HPN		 13072005-04-20 Hans-Peter Nilsson <hp@axis.com>
1308
1309 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1310 have Tcl8.3-compatible intermediate variable.
1311
614348bb
SP		 13122005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1313 Hans-Peter Nilsson <hp@axis.com>
1314
1315 * Makefile.am: Add CRIS support.
1316 * configure.ac: Likewise.
29ede6db
HPN		 1317 * Makefile.in, configure, testsuite/Makefile.in,
1318 include/Makefile.in: Regenerate.
614348bb
SP		 1319 * src/cris: New directory.
1320 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1321 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
727e729b
HPN		 1322
1323 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1324 \r?\n in output tests.
1325
6dab073b
MS		 13262005-04-12 Mike Stump <mrs@apple.com>
1327
1328 * configure: Regenerate.
1329
51f55d7c
HB		 13302005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1331
1332 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1333
7efa52a3
SE		 13342005-03-30 Steve Ellcey <sje@cup.hp.com>
1335
1336 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1337 (ffi_sarg) Ditto.
1338 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1339 to 64 bits in ILP32 mode.
1340 Load 64 bits even for short data.
1341
1488177a
MS		 13422005-03-23 Mike Stump <mrs@apple.com>
1343
1344 * src/powerpc/darwin.S: Update for -m64 multilib.
1345 * src/powerpc/darwin_closure.S: Likewise.
1346
c6900272
ZW		 13472005-03-21 Zack Weinberg <zack@codesourcery.com>
1348
1349 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1350 Do not set tool_include_dir.
1351 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1352 Regenerate.
1353 * include/Makefile.am: Set gcc_version and toollibffidir.
1354 * include/Makefile.in: Regenerate.
1355
962eea0a
AH		 13562005-02-22 Andrew Haley <aph@redhat.com>
1357
1358 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1359 odd-numbered register pairs for 64-bit integer types.
1360
1b05e8bf
AT		 13612005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1362
1363 PR libffi/20104
1364 * testsuite/libffi.call/return_ll1.c: New test case.
1365
6027770b
JJ		 13662005-02-11 Janis Johnson <janis187@us.ibm.com>
1367
1368 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1369 * testsuite/libffi.call/float.c: Ditto.
1370 * testsuite/libffi.call/float2.c: Ditto.
1371 * testsuite/libffi.call/float3.c: Ditto.
1372
ee31ea68
AT		 13732005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1374
1375 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1376
c16be7d5
EB		 13772005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1378
1379 * testsuite/libffi.special/special.exp (cxx_options): Add
1380 -shared-libgcc.
1381
5df07255
RH		 13822004-12-31 Richard Henderson <rth@redhat.com>
1383
1384 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1385 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1386 offset parameters with a type parameter; deduce size and structure
1387 alignment. Update all users.
1388
81a69b13
RH		 13892004-12-31 Richard Henderson <rth@redhat.com>
1390
1391 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1392 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
c6900272 1393 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
81a69b13
RH		 1394 into ffi_prep_closure.
1395 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1396 from scratch.
1397
d833cfb0
RH		 13982004-12-27 Richard Henderson <rth@redhat.com>
1399
1400 * src/x86/unix64.S: Fix typo in unwind info.
1401
1a0f488c
RH		 14022004-12-25 Richard Henderson <rth@redhat.com>
1403
1404 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1405 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1406 (merge_classes): Check for it.
1407 (SSE_CLASS_P): New.
1408 (classify_argument): Pass byte_offset by value; perform all updates
1409 inside struct case.
1410 (examine_argument): Add classes argument; handle
1411 X86_64_COMPLEX_X87_CLASS.
1412 (ffi_prep_args): Merge into ...
1413 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1414 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1415 (ffi_fill_return_value): Remove.
1416 (ffi_prep_closure): Remove dead assert.
1417 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1418 Rewrite to use struct register_args instead of va_list. Create
1419 flags for handling structure returns.
1420 * src/x86/unix64.S: Remove dead strings.
1421 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1422 stack frame with ffi_call. Handle structure returns properly.
1423 (float2sse, floatfloat2sse, double2sse): Remove.
1424 (sse2float, sse2double, sse2floatfloat): Remove.
1425 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1426 to handle structure returns properly.
1427
7994d50c
DE		 14282004-12-08 David Edelsohn <edelsohn@gnu.org>
1429
1430 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1431 PICFLAG.
1432 * Makefile.in: Regenerated.
1433
afabd8e6
RS		 14342004-12-02 Richard Sandiford <rsandifo@redhat.com>
1435
1436 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1437 * configure, aclocal.m4, Makefile.in: Regenerate.
1438 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1439
2539bffe
KC		 14402004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1441
1442 * configure: Regenerate for libtool change.
1443
767f726e
KC		 14442004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1445
1446 * configure: Regenerate for libtool reversion.
1447
2c92e90f
KC		 14482004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1449
1450 * configure: Regenerate for libtool change.
1451
1df8e834
JDA		 14522004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1453
1454 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1455
14562004-11-23 Richard Sandiford <rsandifo@redhat.com>
1457
1458 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1459 of jal. Use an absolute encoding for the frame information.
1460
14612004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1462
1463 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1464 * acinclude.m4: Delete logic for sincludes.
1465 * aclocal.m4, Makefile.in, configure: Regenerate.
1466 * include/Makefile: Likewise.
1467 * testsuite/Makefile: Likewise.
1468
bf02e7fa
EB		 14692004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1470
1471 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1472 on a 8-byte boundary.
1473 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1474
f20459f1
RE		 14752004-10-27 Richard Earnshaw <rearnsha@arm.com>
1476
1477 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1478 long long values. Round stack allocation to a multiple of 8 bytes
1479 for ATPCS compatibility.
1480 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1481 names. Handle returning long long types. Add Thumb and interworking
1482 support. Improve soft-float code.
1483
fdc69597
RE		 14842004-10-27 Richard Earnshaw <rearnsha@arm.com>
1485
1486 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1487 (libffi_exit): New function.
1488 (libffi_init): Build the testglue wrapper if needed.
1489
1e02510f
EB		 14902004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1491
1492 PR other/18138
1493 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1494
d9a193b2
KI		 14952004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1496
1497 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1498
1fddb058
KK		 14992004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1500
1501 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1502 * testsuite/libffi.call/float3.c: New test case.
1503
e0d3b452
KK		 15042004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1505
1506 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1507 the function returning a structure pointed with R2.
1508 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1509 the structure return value if T bit set. Emit position
1510 independent code and EH data if PIC.
1511
a472e9d4
KI		 15122004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1513
1514 * Makefile.am: Add m32r support.
1515 * configure.ac: Likewise.
1516 * Makefile.in: Regenerate.
1517 * confiugre: Regenerate.
1518 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1519 (uint64, sint64, double, longdouble)
1520 * src/m32r: New directory.
1521 * src/m32r/ffi.c: New file.
1522 * src/m32r/sysv.S: Likewise.
1523 * src/m32r/ffitarget.h: Likewise.
1524
3d47e9ca
KK		 15252004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1526
1527 * testsuite/libffi.call/negint.c: New test case.
1528
fc9051dd
L		 15292004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1530
1531 PR libgcj/17465
1532 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1533 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1534 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1535 DYLD_LIBRARY_PATH.
1536
2010caea
AT		 15372004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1538
1539 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1540 * testsuite/libffi.call/promotion.c: Likewise.
1541 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1542 whitespaces.
1543 * testsuite/libffi.call/return_sc.c: Likewise.
1544 * testsuite/libffi.call/return_uc.c: Likewise.
1545
894be816
AT		 15462004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1547
1548 * src/powerpc/darwin.S: Fix comments and identation.
1549 * src/powerpc/darwin_closure.S: Likewise.
1550
2fe7404a
AT		 15512004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1552
1553 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1554 (ffi_prep_args): Handle longdouble arguments.
1555 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1556 longdouble.
1557 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1558 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1559 values.
1560 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1561 * src/types.c: Defined longdouble size and alignment for darwin.
1562
16070e45
AT		 15632004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1564
1565 * src/powerpc/aix.S: Remove whitespaces.
1566 * src/powerpc/aix_closure.S: Likewise.
1567 * src/powerpc/asm.h: Likewise.
1568 * src/powerpc/ffi.c: Likewise.
1569 * src/powerpc/ffitarget.h: Likewise.
1570 * src/powerpc/linux64.S: Likewise.
1571 * src/powerpc/linux64_closure.S: Likewise.
1572 * src/powerpc/ppc_closure.S: Likewise.
1573 * src/powerpc/sysv.S: Likewise.
1574
88d8c771
AG		 15752004-08-30 Anthony Green <green@redhat.com>
1576
1577 * Makefile.am: Add frv support.
1578 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1579 * configure.ac: Read configure.host.
1580 * configure.in: Read configure.host.
1581 * configure.host: New file. frv-elf needs libgloss.
1582 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
a23adb4d 1583 alignment. This is needed to frv and shouldn't harm the others.
88d8c771
AG		 1584 * include/ffi_common.h (ALIGN_DOWN): New macro.
1585 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1586
4c5f37d5
DD		 15872004-08-24 David Daney <daney@avtrex.com>
1588
1589 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1590 * testsuite/libffi.call/closure_fn1.c: Likewise.
1591 * testsuite/libffi.call/closure_fn2.c Likewise.
1592 * testsuite/libffi.call/closure_fn3.c: Likewise.
1593 * testsuite/libffi.call/closure_fn4.c: Likewise.
1594 * testsuite/libffi.call/closure_fn5.c: Likewise.
1595 * testsuite/libffi.call/cls_18byte.c: Likewise.
1596 * testsuite/libffi.call/cls_19byte.c: Likewise.
1597 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1598 * testsuite/libffi.call/cls_20byte.c: Likewise.
1599 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1600 * testsuite/libffi.call/cls_24byte.c: Likewise.
1601 * testsuite/libffi.call/cls_2byte.c: Likewise.
1602 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1603 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1604 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1605 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1606 * testsuite/libffi.call/cls_4byte.c: Likewise.
1607 * testsuite/libffi.call/cls_64byte.c: Likewise.
1608 * testsuite/libffi.call/cls_6byte.c: Likewise.
1609 * testsuite/libffi.call/cls_7byte.c: Likewise.
1610 * testsuite/libffi.call/cls_8byte.c: Likewise.
1611 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1612 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1613 * testsuite/libffi.call/cls_align_double.c: Likewise.
1614 * testsuite/libffi.call/cls_align_float.c: Likewise.
1615 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1616 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1617 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1618 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1619 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1620 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1621 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1622 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1623 * testsuite/libffi.call/cls_double.c: Likewise.
1624 * testsuite/libffi.call/cls_float.c: Likewise.
1625 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1626 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1627 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1628 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1629 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1630 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1631 * testsuite/libffi.call/cls_schar.c: Likewise.
1632 * testsuite/libffi.call/cls_sint.c: Likewise.
1633 * testsuite/libffi.call/cls_sshort.c: Likewise.
1634 * testsuite/libffi.call/cls_uchar.c: Likewise.
1635 * testsuite/libffi.call/cls_uint.c: Likewise.
1636 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1637 * testsuite/libffi.call/cls_ushort.c: Likewise.
1638 * testsuite/libffi.call/nested_struct.c: Likewise.
1639 * testsuite/libffi.call/nested_struct1.c: Likewise.
1640 * testsuite/libffi.call/nested_struct2.c: Likewise.
1641 * testsuite/libffi.call/nested_struct3.c: Likewise.
1642 * testsuite/libffi.call/problem1.c: Likewise.
1643 * testsuite/libffi.special/unwindtest.cc: Likewise.
1644 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1645 to zero.
1646 * testsuite/libffi.call/cls_16byte.c: Likewise.
1647 * testsuite/libffi.call/cls_5byte.c: Likewise.
1648
39dca114
DD		 16492004-08-23 David Daney <daney@avtrex.com>
1650
1651 PR libgcj/13141
1652 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1653 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1654 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1655 parameters and return types.
1656 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1657 (ffi_prep_closure): Ditto.
1658 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1659 alignment calculations.
1660 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1661 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1662
b790003a
CM		 16632004-08-14 Casey Marshall <csm@gnu.org>
1664
1665 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1666 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1667 integer (O32 ABI only).
1668 (ffi_prep_closure): new function.
1669 (ffi_closure_mips_inner_O32): new function.
c6900272 1670 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
b790003a
CM		 1671 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1672 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1673 64 bit integers correctly.
1674 (ffi_closure_O32): new function.
1675 Added DWARF-2 unwind info for both functions.
1676
547d71a1
AH		 16772004-08-10 Andrew Haley <aph@redhat.com>
1678
1679 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1680
98a5bb3b
RM		 16812004-08-01 Robert Millan <robertmh@gnu.org>
1682
1683 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1684 * configure: Regenerate.
1685
4d447364
MR		 16862004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1687
1688 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1689 and mmap() explicitly instead of relying on preset autoconf cache
1690 variables.
1691 * aclocal.m4: Regenerate.
1692 * configure: Regenerate.
1693
e5dce82f
UW		 16942004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1695
1696 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1697 (ffi_check_float_struct): Remove unused prototype.
1698
e6a0022b
GK		 16992004-06-30 Geoffrey Keating <geoffk@apple.com>
1700
1701 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1702 character on Darwin, use '\n\t' instead.
1703
5a2d52ca
MK		 17042004-06-26 Matthias Klose <doko@debian.org>
1705
1706 * libtool-version: Fix typo in revision/age.
1707
401844ef
MK		 17082004-06-17 Matthias Klose <doko@debian.org>
1709
1710 * libtool-version: New.
1711 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1712 * Makefile.in: Regenerate.
1713
94632047 17142004-06-15 Paolo Bonzini <bonzini@gnu.org>
076e5443
PB		 1715
1716 * Makefile.am: Remove useless multilib rules.
1717 * Makefile.in: Regenerate.
1718 * aclocal.m4: Regenerate with automake 1.8.5.
1719 * configure.ac: Remove useless multilib configury.
1720 * configure: Regenerate.
1721
94632047 17222004-06-15 Paolo Bonzini <bonzini@gnu.org>
9ac926b1
PB		 1723
1724 * .cvsignore: New file.
1725
37c849e4
JJ		 17262004-06-10 Jakub Jelinek <jakub@redhat.com>
1727
1728 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1729 fp_done.
1730 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1731 changed from 8.
1732
cb623172
SM		 17332004-06-06 Sean McNeil <sean@mcneil.com>
1734
1735 * configure.ac: Add x86_64-*-freebsd* support.
1736 * configure: Regenerate.
1737
c9b93377
JB		 17382004-04-26 Joe Buck <jbuck@welsh-buck.org>
1739
1740 Bug 15093
1741 * configure.ac: Test for existence of mmap and sys/mman.h before
1742 checking blacklist. Fix suggested by Jim Wilson.
1743 * configure: Regenerate.
1744
cf90fd1b
MA		 17452004-04-26 Matt Austern <austern@apple.com>
1746
1747 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1748 FDE location.
1749 * src/powerpc/darwin_closure.S: Likewise.
c6900272 1750
67781b44
AT		 17512004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1752
1753 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1754 error. Reported by Thomas Heller <theller@python.net>.
1755 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1756 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1757
14b078a8
MK		 17582004-03-20 Matthias Klose <doko@debian.org>
1759
1760 * src/pa/linux.S: Fix typo.
1761
d65a810f
MK		 17622004-03-19 Matthias Klose <doko@debian.org>
1763
67781b44 1764 * Makefile.am: Update.
d65a810f
MK		 1765 * Makefile.in: Regenerate.
1766 * src/pa/ffi.h.in: Remove.
1767 * src/pa/ffitarget.h: New file.
1768
17692004-02-10 Randolph Chung <tausq@debian.org>
1770
1771 * Makefile.am: Add PA support.
1772 * Makefile.in: Regenerate.
1773 * include/Makefile.in: Regenerate.
1774 * configure.ac: Add PA target.
1775 * configure: Regenerate.
1776 * src/pa/ffi.c: New file.
1777 * src/pa/ffi.h.in: Add PA support.
1778 * src/pa/linux.S: New file.
1779 * prep_cif.c: Add PA support.
1780
09cababc
HY		 17812004-03-16 Hosaka Yuji <hos@tamanegi.org>
1782
1783 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
1784 double.
1785 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
1786 with ecif->cif->flags.
1787 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
1788 with cif->flags.
1789 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
1790 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
1791 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
1792 sc_retstruct2b): Add for 1 or 2-bytes struct case.
1793
0e8d0c5f
KC		 17942004-03-15 Kelley Cook <kcook@gcc.gnu.org>
1795
1796 * configure.in: Rename file to ...
1797 * configure.ac: ... this.
1798 * fficonfig.h.in: Regenerate.
1799 * Makefile.in: Regenerate.
1800 * include/Makefile.in: Regenerate.
1801 * testsuite/Makefile.in: Regenerate.
1802
b8d50739
MA		 18032004-03-12 Matt Austern <austern@apple.com>
1804
1805 * src/powerpc/darwin.S: Fix EH information so it corresponds to
1806 changes in EH format resulting from addition of linkonce support.
1807 * src/powerpc/darwin_closure.S: Likewise.
09cababc 1808
3ed8e5b6
AT		 18092004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
1810 Paolo Bonzini <bonzini@gnu.org>
1811
1812 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
1813 Remove VPATH. Remove rules for object files. Remove multilib support.
1814 (AM_CCASFLAGS): Add.
1815 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
1816 (AC_PREREQ): Bump version to 2.59.
1817 (AC_INIT): Fill with version info and bug address.
1818 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
1819 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
1820 De-precious CC so that the right flags are passed down to multilibs.
1821 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
1822 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
1823 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
1824 * configure: Rebuilt.
1825 * aclocal.m4: Likewise.
1826 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
1827 * fficonfig.h.in: Likewise.
1828
254ea477
AS		 18292004-03-11 Andreas Schwab <schwab@suse.de>
1830
1831 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
1832 arguments from fp registers only for the first 8 parameter slots.
1833 Don't convert a float parameter when passed in memory.
1834
3beb1315
HPN		 18352004-03-09 Hans-Peter Nilsson <hp@axis.com>
1836
1837 * configure: Regenerate for config/accross.m4 correction.
1838
fecf735c
MK		 18392004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
1840
1841 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
1842 ecif->cif->bytes to bytes.
1843 (ffi_prep_cif_machdep): Add braces around nested if statement.
1844
4243752c
AM		 18452004-02-09 Alan Modra <amodra@bigpond.net.au>
1846
12001e0f
AM		 1847 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
1848
4243752c
AM		 1849 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
1850 (ffi_closure_helper_LINUX64): Fix typo.
1851 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
1852 for powerpc64-*-*.
1853 * testsuite/libffi.call/float.c: Likewise.
1854 * testsuite/libffi.call/float2.c: Likewise.
1855
bf310028
AM		 18562004-02-08 Alan Modra <amodra@bigpond.net.au>
1857
1858 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
1859 long double function return and long double arg handling.
1860 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
1861 Use "end_pfr" instead of "nf". Correct long double handling.
1862 Localise "temp".
1863 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
1864 return value.
1865 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
1866 space for long double return value. Adjust stack frame and offsets.
1867 Load f2 long double return.
1868
dd5f9ea2
AM		 18692004-02-07 Alan Modra <amodra@bigpond.net.au>
1870
1871 * src/types.c: Use 16 byte long double for POWERPC64.
1872
0ce78f01
EB		 18732004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1874
1875 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
1876 when the structure return address is passed in %o0.
1877 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
1878 (ffi_v9_layout_struct): Align the field following a nested structure
1879 on a word boundary. Use memmove instead of memcpy.
1880 (ffi_call): Update call to ffi_V9_return_struct.
1881 (ffi_prep_closure): Define 'ctx' only for V8.
1882 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
1883 and ffi_closure_sparc_inner_v9.
1884 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
1885 Always skip the structure return address. For structures and long
1886 doubles, copy the argument directly.
1887 (ffi_closure_sparc_inner_v9): Skip the structure return address only
1888 if required. Shift the maximum floating-point slot accordingly. For
1889 big structures, copy the argument directly; otherwise, left-justify the
1890 argument and call ffi_v9_layout_struct to lay out the structure on
1891 the stack.
1892 * src/sparc/v8.S: Undef STACKFRAME before defining it.
1893 (ffi_closure_v8): Pass the structure return address. Update call to
1894 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
1895 Skip the 'unimp' insn when returning long doubles and structures.
1896 * src/sparc/v9.S: Undef STACKFRAME before defining it.
1897 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
1898 FFI_TYPE_INT handling. Load structures both in integers and
1899 floating-point registers on return.
1900 * README: Update status of the SPARC port.
1901
7fcc1d9b
AT		 19022004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
1903
1904 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
1905 as of type ffi_arg.
1906 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
1907
6ed0ecaf
UW		 19082004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
1909
1910 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
1911 value as of type ffi_arg, not unsigned int.
1912
c930d8a5
MR		 19132004-01-21 Michael Ritzert <ritzert@t-online.de>
1914
1915 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
1916 of the LHS.
1917
7b2e3e60
AT		 19182004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
1919
1920 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
1921 Solaris.
1922
2e89756b
RO		 19232004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1924
1925 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
1926 to void *.
1927
628b26f8
RH		 19282003-12-10 Richard Henderson <rth@redhat.com>
1929
1930 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
1931 size_t instead of int.
1932
0ee646d6
HY		 19332003-12-04 Hosaka Yuji <hos@tamanegi.org>
1934
1935 * testsuite/libffi.call/many_win32.c: Include <float.h>.
1936 * testsuite/libffi.call/many_win32.c (main): Replace variable
1937 int i with unsigned long ul.
1938
1939 * testsuite/libffi.call/cls_align_uint64.c: New test case.
1940 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1941 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1942 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1943 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1944 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1945 * testsuite/libffi.call/cls_align_float.c: Likewise.
1946 * testsuite/libffi.call/cls_align_double.c: Likewise.
1947 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1948 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1949
2c4ed456
HY		 19502003-12-02 Hosaka Yuji <hos@tamanegi.org>
1951
1952 PR other/13221
1953 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
1954 Align arguments to 32 bits.
1955
d18c3386
AT		 19562003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
1957
1958 PR other/13221
1959 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
1960 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1961 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1962 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1963 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1964 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1965
1966 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
1967
88096b2a
KG		 19682003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
1969
1970 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
1971 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1972
3eee102e
AT		 19732003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
1974
1975 * Makefile.in: Rebuilt.
1976 * configure: Likewise.
1977 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
1978 the right type.
1979
5196736e
AJ		 19802003-11-21 Andreas Jaeger <aj@suse.de>
1981 Andreas Tobler <a.tobler@schweiz.ch>
1982
1983 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
1984 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
1985 * Makefile.in: Rebuilt.
1986 * aclocal.m4: Likewise.
1987 * configure: Likewise.
1988 * fficonfig.h.in: Likewise.
1989 * testsuite/lib/libffi-dg.exp: Add include dir.
1990 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
1991 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1992 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
1993 for ffi_closure if available.
1994 * testsuite/libffi.call/closure_fn1.c: Likewise.
1995 * testsuite/libffi.call/closure_fn2.c: Likewise.
1996 * testsuite/libffi.call/closure_fn3.c: Likewise.
1997 * testsuite/libffi.call/closure_fn4.c: Likewise.
1998 * testsuite/libffi.call/closure_fn5.c: Likewise.
1999 * testsuite/libffi.call/cls_12byte.c: Likewise.
2000 * testsuite/libffi.call/cls_16byte.c: Likewise.
2001 * testsuite/libffi.call/cls_18byte.c: Likewise.
2002 * testsuite/libffi.call/cls_19byte.c: Likewise.
2003 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2004 * testsuite/libffi.call/cls_20byte.c: Likewise.
2005 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2006 * testsuite/libffi.call/cls_24byte.c: Likewise.
2007 * testsuite/libffi.call/cls_2byte.c: Likewise.
2008 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2009 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2010 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2011 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2012 * testsuite/libffi.call/cls_4byte.c: Likewise.
2013 * testsuite/libffi.call/cls_5byte.c: Likewise.
2014 * testsuite/libffi.call/cls_64byte.c: Likewise.
2015 * testsuite/libffi.call/cls_6byte.c: Likewise.
2016 * testsuite/libffi.call/cls_7byte.c: Likewise.
2017 * testsuite/libffi.call/cls_8byte.c: Likewise.
2018 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2019 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2020 * testsuite/libffi.call/cls_double.c: Likewise.
2021 * testsuite/libffi.call/cls_float.c: Likewise.
2022 * testsuite/libffi.call/cls_schar.c: Likewise.
2023 * testsuite/libffi.call/cls_sint.c: Likewise.
2024 * testsuite/libffi.call/cls_sshort.c: Likewise.
2025 * testsuite/libffi.call/cls_uchar.c: Likewise.
2026 * testsuite/libffi.call/cls_uint.c: Likewise.
2027 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2028 * testsuite/libffi.call/cls_ushort.c: Likewise.
2029 * testsuite/libffi.call/nested_struct.c: Likewise.
2030 * testsuite/libffi.call/nested_struct1.c: Likewise.
2031 * testsuite/libffi.call/nested_struct2.c: Likewise.
2032 * testsuite/libffi.call/nested_struct3.c: Likewise.
2033 * testsuite/libffi.call/problem1.c: Likewise.
2034 * testsuite/libffi.special/unwindtest.cc: Likewise.
2035
9e6c3ecb
AT		 20362003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2037
2038 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2039
70d8f2a1
AT		 20402003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2041
2042 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2043 Add -lgcc_s to additional flags.
2044
ea81d2a3
AT		 20452003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2046
2047 * configure.in, include/Makefile.am: PR libgcj/11147, install
2048 the ffitarget.h header file in a gcc versioned and target
2049 dependent place.
2050 * configure: Regenerated.
2051 * Makefile.in, include/Makefile.in: Likewise.
2052 * testsuite/Makefile.in: Likewise.
2053
647af4a9
AT		 20542003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2055
2056 * testsuite/libffi.call/closure_fn0.c: Print result and check
2057 with dg-output to make debugging easier.
2058 * testsuite/libffi.call/closure_fn1.c: Likewise.
2059 * testsuite/libffi.call/closure_fn2.c: Likewise.
2060 * testsuite/libffi.call/closure_fn3.c: Likewise.
2061 * testsuite/libffi.call/closure_fn4.c: Likewise.
2062 * testsuite/libffi.call/closure_fn5.c: Likewise.
2063 * testsuite/libffi.call/cls_12byte.c: Likewise.
2064 * testsuite/libffi.call/cls_16byte.c: Likewise.
2065 * testsuite/libffi.call/cls_18byte.c: Likewise.
2066 * testsuite/libffi.call/cls_19byte.c: Likewise.
2067 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2068 * testsuite/libffi.call/cls_20byte.c: Likewise.
2069 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2070 * testsuite/libffi.call/cls_24byte.c: Likewise.
2071 * testsuite/libffi.call/cls_2byte.c: Likewise.
2072 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2073 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2074 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2075 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2076 * testsuite/libffi.call/cls_4byte.c: Likewise.
2077 * testsuite/libffi.call/cls_5byte.c: Likewise.
2078 * testsuite/libffi.call/cls_64byte.c: Likewise.
2079 * testsuite/libffi.call/cls_6byte.c: Likewise.
2080 * testsuite/libffi.call/cls_7byte.c: Likewise.
2081 * testsuite/libffi.call/cls_8byte.c: Likewise.
2082 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2083 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2084 * testsuite/libffi.call/cls_double.c: Likewise.
2085 * testsuite/libffi.call/cls_float.c: Likewise.
2086 * testsuite/libffi.call/cls_schar.c: Likewise.
2087 * testsuite/libffi.call/cls_sint.c: Likewise.
2088 * testsuite/libffi.call/cls_sshort.c: Likewise.
2089 * testsuite/libffi.call/cls_uchar.c: Likewise.
2090 * testsuite/libffi.call/cls_uint.c: Likewise.
2091 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2092 * testsuite/libffi.call/cls_ushort.c: Likewise.
2093 * testsuite/libffi.call/problem1.c: Likewise.
2094
2095 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2096 static.
2097
ac47cc13
AT		 20982003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2099
2100 * testsuite/libffi.call/cls_9byte2.c: New test case.
2101 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2102 * testsuite/libffi.call/cls_64byte.c: Likewise.
2103 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2104 * testsuite/libffi.call/cls_19byte.c: Likewise.
2105 * testsuite/libffi.call/cls_18byte.c: Likewise.
2106 * testsuite/libffi.call/closure_fn4.c: Likewise.
2107 * testsuite/libffi.call/closure_fn5.c: Likewise.
2108 * testsuite/libffi.call/cls_schar.c: Likewise.
2109 * testsuite/libffi.call/cls_sint.c: Likewise.
2110 * testsuite/libffi.call/cls_sshort.c: Likewise.
2111 * testsuite/libffi.call/nested_struct2.c: Likewise.
2112 * testsuite/libffi.call/nested_struct3.c: Likewise.
2113
f145ed6b
AT		 21142003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2115
2116 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2117 * testsuite/libffi.call/cls_uchar.c: Likewise.
2118 * testsuite/libffi.call/cls_uint.c: Likewise.
2119 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2120 * testsuite/libffi.call/cls_ushort.c: Likewise.
2121 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2122
2b04d015
AT		 21232003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2124
2125 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2126 the initialization.
2127
57f542b3
AT		 21282003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2129
2b04d015 2130 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
57f542b3
AT		 2131 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2132
e61acf72
DD		 21332003-10-22 David Daney <ddaney@avtrex.com>
2134
2135 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2136 __attribute__((__mode__(__SI__))) and friends.
2137
2461c08b
AS		 21382003-10-22 Andreas Schwab <schwab@suse.de>
2139
2140 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2141
1450eb7a
AT		 21422003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2143
2144 * configure.in: AC_LINK_FILES(ffitarget.h).
2145 * configure: Regenerate.
2146 * Makefile.in: Likewise.
2147 * include/Makefile.in: Likewise.
2148 * testsuite/Makefile.in: Likewise.
2149 * fficonfig.h.in: Likewise.
2150
21512003-10-21 Paolo Bonzini <bonzini@gnu.org>
2152 Richard Henderson <rth@redhat.com>
2153
2154 Avoid that ffi.h includes fficonfig.h.
2155
2156 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2157 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2158 (TARGET_SRC_MIPS_SGI): Removed.
2159 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2160 (MIPS_SGI): Removed.
2161 (CLEANFILES): Removed.
2162 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2163 targets.
2164 * acconfig.h: Removed.
2165 * configure.in: Compute sizeofs only for double and long double.
2166 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2167 into AC_DEFINE instead of using acconfig.h. Create
2168 include/ffitarget.h instead of include/fficonfig.h. Rename
2169 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2170 AC_DEFINE EH_FRAME_FLAGS.
2171 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2172 (hack_DATA): Add ffitarget.h.
2173 * include/ffi.h.in: Remove all system specific definitions.
2174 Declare raw API even if it is not installed, why bother?
2175 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2176 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2177 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2178 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2179 * include/ffi_common.h (bool): Do not define.
2180 (ffi_assert): Accept failed assertion.
2181 (ffi_type_test): Return void and accept file/line.
2182 (FFI_ASSERT): Pass stringized failed assertion.
2183 (FFI_ASSERT_AT): New macro.
2184 (FFI_ASSERT_VALID_TYPE): New macro.
2185 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2186 UINT64, SINT64): Define here with gcc's __attribute__ macro
2187 instead of in ffi.h
2188 (FLOAT32, ALIGN): Define here instead of in ffi.h
2189 * include/ffi-mips.h: Removed. Its content moved to
2190 src/mips/ffitarget.h after separating assembly and C sections.
2191 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2192 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2193 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2194 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2195 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2196 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2197 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2198 (ffi_type_test): Rewritten.
2199 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2200 FFI_ASSERT_VALID_TYPE.
2201 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2202 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2203 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2204 src/s390/ffitarget.h, src/sh/ffitarget.h,
2205 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2206 src/x86/ffitarget.h: New files.
2207 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2208 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2209 src/powerpc/aix.S, src/powerpc/darwin.S,
2210 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2211 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2212 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2213 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2214 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2215 include fficonfig.h
2216
27d54b2a
RO		 22172003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2218
2219 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2220 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2221
0df6970d
AT		 22222003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2223
2224 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2225 Used when FFI_DEBUG = 1.
2226
2612ffce
AM		 22272003-10-14 Alan Modra <amodra@bigpond.net.au>
2228
2229 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2230 and align.
2231
d54f7354
RO		 22322003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2233
2234 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2235 FFI_MIPS_O32 for O32 ABI.
2236
7a54c850
AT		 22372003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2238
2239 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2240 SPARC64. Cleanup whitespaces.
2241
9e3bfa9b
AT		 22422003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2243
2244 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2245 strongarm, xscale. Cleanup whitespaces.
2246 * testsuite/libffi.call/closure_fn1.c: Likewise.
2247 * testsuite/libffi.call/closure_fn2.c: Likewise.
2248 * testsuite/libffi.call/closure_fn3.c: Likewise.
2249 * testsuite/libffi.call/cls_12byte.c: Likewise.
2250 * testsuite/libffi.call/cls_16byte.c: Likewise.
2251 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2252 * testsuite/libffi.call/cls_20byte.c: Likewise.
2253 * testsuite/libffi.call/cls_24byte.c: Likewise.
2254 * testsuite/libffi.call/cls_2byte.c: Likewise.
2255 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2256 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2257 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2258 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2259 * testsuite/libffi.call/cls_4byte.c: Likewise.
2260 * testsuite/libffi.call/cls_5byte.c: Likewise.
2261 * testsuite/libffi.call/cls_6byte.c: Likewise.
2262 * testsuite/libffi.call/cls_7byte.c: Likewise.
2263 * testsuite/libffi.call/cls_8byte.c: Likewise.
2264 * testsuite/libffi.call/cls_double.c: Likewise.
2265 * testsuite/libffi.call/cls_float.c: Likewise.
2266 * testsuite/libffi.call/cls_uchar.c: Likewise.
2267 * testsuite/libffi.call/cls_uint.c: Likewise.
2268 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2269 * testsuite/libffi.call/cls_ushort.c: Likewise.
2270 * testsuite/libffi.call/nested_struct.c: Likewise.
2271 * testsuite/libffi.call/nested_struct1.c: Likewise.
2272 * testsuite/libffi.call/problem1.c: Likewise.
2273 * testsuite/libffi.special/unwindtest.cc: Likewise.
2274 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2275
3e86b595
DE		 22762003-09-18 David Edelsohn <edelsohn@gnu.org>
2277
2278 * src/powerpc/aix.S: Cleanup whitespaces.
2279 * src/powerpc/aix_closure.S: Likewise.
2280
8d8369f8
AT		 22812003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2282
2283 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2284 * src/powerpc/darwin_closure.S: Likewise.
2285 * src/powerpc/ffi_darwin.c: Likewise.
2286
69b182fd
AT		 22872003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2288 David Edelsohn <edelsohn@gnu.org>
2289
2290 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2291 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2292 parameter stack.
2293 * src/powerpc/darwin_closure.S: Likewise.
2294 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2295 according to the Darwin/AIX ABI.
2296 (ffi_prep_cif_machdep): Likewise.
2297 (ffi_closure_helper_DARWIN): Likewise.
2298 Remove the outgoing parameter stack logic. Simplify the evaluation
2299 of the different CASE types.
2300 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2301 statement in the trampoline code.
2302
df805cdc
KK		 23032003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2304
2305 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2306 for the register size.
2307 (ffi_closure_helper_SYSV): Handle the structure return value
2308 address correctly.
2309 (ffi_closure_helper_SYSV): Return the appropriate type when
2310 the registers are used for the structure return value.
2311 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2312 the 64-bit return value. Update copyright years.
2313
eb26c76c
RO		 23142003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2315
2316 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2317 srcdir for ffi_mips.h.
2318
6350c02c
AM		 23192003-09-12 Alan Modra <amodra@bigpond.net.au>
2320
2321 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2322 structure size.
2323 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2324 placement of float result.
2325 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2326 cast of "resp" for big-endian 64 bit machines.
2327
5af1c806
AM		 23282003-09-11 Alan Modra <amodra@bigpond.net.au>
2329
2330 * src/types.c (double, longdouble): Merge identical SH and ARM
2331 typedefs, and add POWERPC64.
2332 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2333 struct split over gpr and rest.
2334 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2335 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2336
086a4bd7
AT		 23372003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2338
2339 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2340 passing correctly.
c930d8a5 2341
e517f636
AM		 23422003-09-09 Alan Modra <amodra@bigpond.net.au>
2343
2344 * configure: Regenerate.
2345
6f20d627
AT		 23462003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2347
2348 * Makefile.am: Remove build rules for ffitest.
2349 * Makefile.in: Rebuilt.
c930d8a5 2350
6f20d627 23512003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
226660af
AT		 2352
2353 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2354 about implicit declaration of abort().
2355
bbf3057b
AT		 23562003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2357
2358 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2359 * Makefile.in: Rebuilt.
2360 * configure.in: Add dejagnu test framework.
2361 * configure: Rebuilt.
2362
2363 * testsuite/Makefile.am: New file.
2364 * testsuite/Makefile.in: Built
c930d8a5 2365 * testsuite/lib/libffi-dg.exp: New file.
bbf3057b
AT		 2366 * testsuite/config/default.exp: Likewise.
2367 * testsuite/libffi.call/call.exp: Likewise.
2368 * testsuite/libffi.call/ffitest.h: Likewise.
2369 * testsuite/libffi.call/closure_fn0.c: Likewise.
2370 * testsuite/libffi.call/closure_fn1.c: Likewise.
2371 * testsuite/libffi.call/closure_fn2.c: Likewise.
2372 * testsuite/libffi.call/closure_fn3.c: Likewise.
2373 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2374 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2375 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2376 * testsuite/libffi.call/cls_2byte.c: Likewise.
2377 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2378 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2379 * testsuite/libffi.call/cls_4byte.c: Likewise.
2380 * testsuite/libffi.call/cls_5byte.c: Likewise.
2381 * testsuite/libffi.call/cls_6byte.c: Likewise.
2382 * testsuite/libffi.call/cls_7byte.c: Likewise.
2383 * testsuite/libffi.call/cls_8byte.c: Likewise.
2384 * testsuite/libffi.call/cls_12byte.c: Likewise.
2385 * testsuite/libffi.call/cls_16byte.c: Likewise.
2386 * testsuite/libffi.call/cls_20byte.c: Likewise.
2387 * testsuite/libffi.call/cls_24byte.c: Likewise.
2388 * testsuite/libffi.call/cls_double.c: Likewise.
2389 * testsuite/libffi.call/cls_float.c: Likewise.
2390 * testsuite/libffi.call/cls_uchar.c: Likewise.
2391 * testsuite/libffi.call/cls_uint.c: Likewise.
2392 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2393 * testsuite/libffi.call/cls_ushort.c: Likewise.
2394 * testsuite/libffi.call/float.c: Likewise.
2395 * testsuite/libffi.call/float1.c: Likewise.
2396 * testsuite/libffi.call/float2.c: Likewise.
2397 * testsuite/libffi.call/many.c: Likewise.
2398 * testsuite/libffi.call/many_win32.c: Likewise.
2399 * testsuite/libffi.call/nested_struct.c: Likewise.
2400 * testsuite/libffi.call/nested_struct1.c: Likewise.
2401 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2402 * testsuite/libffi.call/problem1.c: Likewise.
2403 * testsuite/libffi.call/promotion.c: Likewise.
2404 * testsuite/libffi.call/return_ll.c: Likewise.
2405 * testsuite/libffi.call/return_sc.c: Likewise.
2406 * testsuite/libffi.call/return_uc.c: Likewise.
2407 * testsuite/libffi.call/strlen.c: Likewise.
2408 * testsuite/libffi.call/strlen_win32.c: Likewise.
2409 * testsuite/libffi.call/struct1.c: Likewise.
2410 * testsuite/libffi.call/struct2.c: Likewise.
2411 * testsuite/libffi.call/struct3.c: Likewise.
2412 * testsuite/libffi.call/struct4.c: Likewise.
2413 * testsuite/libffi.call/struct5.c: Likewise.
2414 * testsuite/libffi.call/struct6.c: Likewise.
2415 * testsuite/libffi.call/struct7.c: Likewise.
2416 * testsuite/libffi.call/struct8.c: Likewise.
c930d8a5 2417 * testsuite/libffi.call/struct9.c: Likewise.
bbf3057b
AT		 2418 * testsuite/libffi.special/special.exp: New file.
2419 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2420 * testsuite/libffi.special/unwindtest.cc: Likewise.
2421
c930d8a5 2422
996b4f8f
KK		 24232003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2424
2425 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2426 copyright years.
2427
b00badcd
AM		 24282003-08-02 Alan Modra <amodra@bigpond.net.au>
2429
2430 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2431 structure passing.
2432 (ffi_closure_helper_LINUX64): Likewise.
2433 * src/powerpc/linux64.S: Remove code writing to parm save area.
2434 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2435 address in lr from ffi_closure_helper_LINUX64 call to calculate
2436 table address. Optimize function tail.
2437
c1516eae
AT		 24382003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2439
2440 * src/sparc/ffi.c: Handle all floating point registers.
2441 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2442
fc0e6222
GP		 24432003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2444
2445 * README: Note that libffi is not part of GCC. Update the project
2446 URL and status.
2447
a1569df8
FS		 24482003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2449
2450 * src/powerpc/ppc_closure.S: Include ffi.h.
2451
12e816bf
RO		 24522003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2453
2454 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2455 Use C style comments.
2456
e860ed6d
KK		 24572003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2458
2459 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2460 * Makefile.in: Regenerate.
2461 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2462 * configure: Regenerate.
2463 * include/ffi.h.in: Add SHmedia support.
2464 * src/sh64/ffi.c: New file.
2465 * src/sh64/sysv.S: New file.
2466
a733b15e
JJ		 24672003-05-16 Jakub Jelinek <jakub@redhat.com>
2468
2469 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2470 should be read-only.
2471 * configure: Rebuilt.
2472 * fficonfig.h.in: Rebuilt.
2473 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2474 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2475 * src/powerpc/linux64.S: Likewise.
2476 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2477 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2478 if -fpic/-fPIC/-mrelocatable.
2479 * src/powerpc/powerpc_closure.S: Likewise.
2480 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2481 #write in .eh_frame flags.
2482 * src/sparc/v9.S: Likewise.
2483 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2484 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2485 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2486
bd19c9f4
JS		 24872003-05-07 Jeff Sturm <jsturm@one-point.com>
2488
2489 Fixes PR bootstrap/10656
2490 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2491 support for .register pseudo-op.
2492 * src/sparc/v8.S: Use it.
2493 * fficonfig.h.in: Rebuilt.
2494 * configure: Rebuilt.
2495
e9b84181
JJ		 24962003-04-18 Jakub Jelinek <jakub@redhat.com>
2497
2498 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2499 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2500 Make it the default on POWERPC64.
2501 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2502 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2503 * configure: Rebuilt.
2504 * src/powerpc/ffi.c (hidden): Define.
2505 (ffi_prep_args_SYSV): Renamed from
2506 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2507 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2508 ASM_NEEDS_REGISTERS64): New.
2509 (ffi_prep_args64): New function.
2510 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2511 (ffi_call): Likewise.
2512 (ffi_prep_closure): Likewise.
2513 (flush_icache): Surround by #ifndef POWERPC64.
2514 (ffi_dblfl): New union type.
2515 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2516 (ffi_closure_helper_LINUX64): New function.
2517 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2518 __powerpc64__.
2519 * src/powerpc/sysv.S: Likewise.
2520 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2521 * src/powerpc/linux64.S: New file.
2522 * src/powerpc/linux64_closure.S: New file.
2523 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2524 src/powerpc/linux64_closure.S.
2525 (TARGET_SRC_POWERPC): Likewise.
2526
2527 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2528 closure_test_fn3): Fix result printing on big-endian 64-bit
2529 machines.
2530 (main): Print tst2_arg instead of uninitialized tst2_result.
2531
2532 * src/ffitest.c (main): Hide what closure pointer really points to
2533 from the compiler.
2534
908a31e1
RE		 25352003-04-16 Richard Earnshaw <rearnsha@arm.com>
2536
2537 * configure.in (arm-*-netbsdelf*): Add configuration.
2538 (configure): Regenerated.
2539
41a73c4c
LR		 25402003-04-04 Loren J. Rittle <ljrittle@acm.org>
2541
2542 * include/Makefile.in: Regenerate.
2543
cd41c847
ZD		 25442003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2545
2546 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2547 bit mode.
2548 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2549 Receive closure pointer through parameter, read args using
2550 __builtin_dwarf_cfa.
2551 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2552
6db450f9
AS		 25532003-03-12 Andreas Schwab <schwab@suse.de>
2554
2555 * configure.in: Avoid trailing /. in toolexeclibdir.
2556 * configure: Rebuilt.
2557
70144e9c
AT		 25582003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2559
c6900272 2560 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
70144e9c 2561
79e43f89
AT		 25622003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2563
2564 * libffi/src/powerpc/darwin_closure.S:
2565 Fix alignement bug, allocate 8 bytes for the result.
c930d8a5 2566 * libffi/src/powerpc/aix_closure.S:
79e43f89
AT		 2567 Likewise.
2568 * libffi/src/powerpc/ffi_darwin.c:
2569 Update stackframe description for aix/darwin_closure.S.
c930d8a5 2570
0fcc9e55
JJ		 25712003-02-06 Jakub Jelinek <jakub@redhat.com>
2572
2573 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2574 attribute.
2575
63b87c2e
CC		 25762003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2577 Andreas Schwab <schwab@suse.de>
2578
c6900272 2579 * configure.in: Adjust command to source config-ml.in to account
63b87c2e
CC		 2580 for changes to the libffi_basedir definition.
2581 (libffi_basedir): Remove ${srcdir} from value and include trailing
2582 slash if nonempty.
c930d8a5 2583
c6900272 2584 * configure: Regenerate.
63b87c2e 2585
c33402e2
FS		 25862003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2587
2588 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2589
fd2e49b2
AH		 25902003-01-28 Andrew Haley <aph@redhat.com>
2591
2592 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2593 * src/x86/ffi64.c (ffi_prep_closure): New.
2594 (ffi_closure_UNIX64_inner): New.
2595 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2596
0e292c01
AO		 25972003-01-27 Alexandre Oliva <aoliva@redhat.com>
2598
2599 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2600 Remove USE_LIBDIR conditional.
2601 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2602 * Makefile.in, configure: Rebuilt.
2603
acabc220
DE		 26042003-01027 David Edelsohn <edelsohn@gnu.org>
2605
2606 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2607 * Makefile.in: Regenerate.
2608
be0acd42
AH		 26092003-01-22 Andrew Haley <aph@redhat.com>
2610
2611 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2612 unwind info.
2613
ed695c8e
AT		 26142003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2615
2616 * src/powerpc/darwin.S: Add unwind info.
2617 * src/powerpc/darwin_closure.S: Likewise.
2618
45ba15f4
AH		 26192003-01-14 Andrew Haley <aph@redhat.com>
2620
2621 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2622 (ffi_prep_cif_machdep): Likewise.
2623 * src/x86/unix64.S: Add unwind info.
2624
4b5f402e
AJ		 26252003-01-14 Andreas Jaeger <aj@suse.de>
2626
2627 * src/ffitest.c (main): Only use ffi_closures if those are
2628 supported.
2629
7b5102af
TT		 26302003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2631
2632 * libffi/src/ffitest.c
c6900272 2633 add closure testcases
7b5102af
TT		 2634
26352003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2636
2637 * libffi/src/powerpc/ffi.c
c6900272 2638 fix alignment bug for float (4 byte aligned iso 8 byte)
7b5102af 2639
5f7c78d8
GK		 26402003-01-09 Geoffrey Keating <geoffk@apple.com>
2641
2642 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2643 * src/powerpc/darwin.S: Remove RCS version string.
2644
c75c7793
JS		 26452003-01-03 Jeff Sturm <jsturm@one-point.com>
2646
2647 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2648 * src/ffitest.c (main): Use static storage for closure.
2649 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2650 * src/sparc/v8.S (ffi_closure_v8): New.
2651 * src/sparc/v9.S (ffi_closure_v9): New.
2652
eb3c46a1
RM		 26532002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2654
2655 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2656 enumeration for X86_WIN32.
2657 * src/x86/win32.S: Added ffi_call_STDCALL function
2658 definition.
2659 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2660 switch cases for recognising FFI_STDCALL and
2661 calling ffi_call_STDCALL if target is X86_WIN32.
2662 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2663 stdcall versions of the "my_strlen" and "many"
2664 test functions (for X86_WIN32).
2665 Added test cases to test stdcall invocation using
2666 these functions.
2667
4977bab6
ZW		 26682002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2669
2670 * src/sh/sysv.S: Add DWARF2 unwind info.
2671
67333e84
UW		 26722002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2673
2674 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2675
bc3cf760
JW		 26762002-11-26 Jim Wilson <wilson@redhat.com>
2677
2678 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2679
4977bab6
ZW		 26802002-11-23 H.J. Lu <hjl@gnu.org>
2681
2682 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2683 Include ../config/accross.m4.
2684 * aclocal.m4; Rebuild.
2685 * configure: Likewise.
2686
18789f4e
UW		 26872002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2688
2689 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2690
6bee5372
DD		 26912002-11-11 DJ Delorie <dj@redhat.com>
2692
2693 * configure.in: Look for common files in the right place.
2694
8177895b
UW		 26952002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2696
2697 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2698 raw data as _Jv_word values, not ffi_raw.
2699 (ffi_java_ptrarray_to_raw): Likewise.
2700 (ffi_java_rvalue_to_raw): New function.
2701 (ffi_java_raw_call): Call it.
2702 (ffi_java_raw_to_rvalue): New function.
2703 (ffi_java_translate_args): Call it.
2704 * src/ffitest.c (closure_test_fn): Interpret return value
2705 as ffi_arg, not int.
2706 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2707 FFI_TYPE_POINTER case.
2708 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2709 values extended to word size.
2710
af50f4c8
AJ		 27112002-10-02 Andreas Jaeger <aj@suse.de>
2712
2713 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2714
5d297584
BT		 27152002-10-01 Bo Thorsen <bo@smetana.suse.de>
2716
2717 * include/ffi.h.in: Fix i386 win32 compilation.
2718
c4f17c6f
UW		 27192002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2720
2721 * configure.in: Add s390x-*-linux-* target.
2722 * configure: Regenerate.
2723 * include/ffi.h.in: Define S390X for s390x targets.
2724 (FFI_CLOSURES): Define for s390/s390x.
2725 (FFI_TRAMPOLINE_SIZE): Likewise.
2726 (FFI_NATIVE_RAW_API): Likewise.
2727 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2728 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
4977bab6 2729 * src/s390/ffi.c: Major rework of existing code. Add support for
c4f17c6f
UW		 2730 s390x targets. Add closure support.
2731 * src/s390/sysv.S: Likewise.
2732
d7b52625
RE		 27332002-09-29 Richard Earnshaw <rearnsha@arm.com>
2734
2735 * src/arm/sysv.S: Fix typo.
2736
28165e22
RE		 27372002-09-28 Richard Earnshaw <rearnsha@arm.com>
2738
2739 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2740 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2741 (ffi_call_SYSV): Handle soft-float.
2742
0b0f4163
BT		 27432002-09-27 Bo Thorsen <bo@suse.de>
2744
2745 * include/ffi.h.in: Fix multilib x86-64 support.
2746
6c5d742e
KG		 27472002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2748
2749 * Makefile.am (all-multi): Fix multilib parallel build.
2750
f4b2b0d6
KK		 27512002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2752
2753 * configure.in (sh[34]*-*-linux*): Add brackets.
2754 * configure: Regenerate.
2755
ddebdfeb
KK		 27562002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2757
2758 * Makefile.am: Add SH support.
2759 * Makefile.in: Regenerate.
2760 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2761 * configure: Regenerate.
2762 * include/ffi.h.in: Add SH support.
2763 * src/sh/ffi.c: New file.
2764 * src/sh/sysv.S: New file.
2765 * src/types.c: Add SH support.
2766
c94974ca
BT		 27672002-07-16 Bo Thorsen <bo@suse.de>
2768
2769 * src/x86/ffi64.c: New file that adds x86-64 support.
2770 * src/x86/unix64.S: New file that handles argument setup for
2771 x86-64.
2772 * src/x86/sysv.S: Don't use this on x86-64.
2773 * src/x86/ffi.c: Don't use this on x86-64.
2774 Remove unused vars.
2775 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2776 for x86-64.
2777 * src/ffitest.c (struct6): New test that tests a special case in
2778 the x86-64 ABI.
2779 (struct7): Likewise.
2780 (struct8): Likewise.
2781 (struct9): Likewise.
2782 (closure_test_fn): Silence warning about this when it's not used.
2783 (main): Add the new tests.
2784 (main): Fix a couple of wrong casts and silence some compiler warnings.
2785 * include/ffi.h.in: Add x86-64 ABI definition.
2786 * fficonfig.h.in: Regenerate.
2787 * Makefile.am: Add x86-64 support.
2788 * configure.in: Likewise.
2789 * Makefile.in: Regenerate.
2790 * configure: Likewise.
2791
27922002-06-24 Bo Thorsen <bo@suse.de>
2793
2794 * src/types.c: Merge settings for similar architectures.
2795 Add x86-64 sizes and alignments.
2796
27972002-06-23 Bo Thorsen <bo@suse.de>
2798
2799 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
2800 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
2801 * src/mips/ffi.c (ffi_prep_args): Likewise.
2802 * src/m68k/ffi.c (ffi_prep_args): Likewise.
2803
5be36a8f
L		 28042002-07-18 H.J. Lu (hjl@gnu.org)
2805
2806 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
2807 (libffi_la_SOURCES): Support MIPS_LINUX.
2808 (libffi_convenience_la_SOURCES): Likewise.
2809 * Makefile.in: Regenerated.
2810
2811 * configure.in (mips64*-*): Skip.
2812 (mips*-*-linux*): New.
2813 * configure: Regenerated.
2814
2815 * src/mips/ffi.c: Include <sgidefs.h>.
2816
ad5e163c
UW		 28172002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
2818
2819 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
2820
ee4586c5
RS		 28212002-05-27 Roger Sayle <roger@eyesopen.com>
2822
2823 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
2824
b86aa7b0
BT		 28252002-05-27 Bo Thorsen <bo@suse.de>
2826
2827 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
2828 fix formatting.
2829
859f7aaf
AT		 28302002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
2831
2832 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
2833 beginning of function (for older apple cc).
2834
6706f116
AO		 28352002-05-08 Alexandre Oliva <aoliva@redhat.com>
2836
2837 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
2838 script entry, and set LD to it when configuring multilibs.
2839 * configure: Rebuilt.
2840
fad655c8
JT		 28412002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
2842
2843 * configure.in (sparc64-*-netbsd*): Add target.
2844 (sparc-*-netbsdelf*): Likewise.
2845 * configure: Regenerate.
2846
189ab118
DM		 28472002-04-28 David S. Miller <davem@redhat.com>
2848
2849 * configure.in, configure: Fix SPARC test in previous change.
2850
22bcf65c
GT		 28512002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
2852
2853 * Makefile.am: Add Linux for S/390 support.
2854 * Makefile.in: Regenerate.
2855 * configure.in: Add Linux for S/390 support.
2856 * configure: Regenerate.
2857 * include/ffi.h.in: Add Linux for S/390 support.
2858 * src/s390/ffi.c: New file from libffi CVS tree.
2859 * src/s390/sysv.S: New file from libffi CVS tree.
2860
5d84cf0b
JJ		 28612002-04-28 Jakub Jelinek <jakub@redhat.com>
2862
2863 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
2864 %r_disp32().
2865 * src/sparc/v8.S: Use it.
2866 * src/sparc/v9.S: Likewise.
2867 * fficonfig.h.in: Rebuilt.
2868 * configure: Rebuilt.
2869
cc382e5c
HB		 28702002-04-08 Hans Boehm <Hans_Boehm@hp.com>
2871
2872 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
2873 correctly.
2874 * src/ia64/unix.S: Add unwind information. Fix comments.
2875 Save sp in a way that's compatible with unwind info.
2876 (ffi_call_unix): Correctly restore sp in all cases.
2877 * src/ia64/ffi.c: Add, fix comments.
2878
dc798645
JJ		 28792002-04-08 Jakub Jelinek <jakub@redhat.com>
2880
2881 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
2882
da00eaab
JT		 28832002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
2884
2885 * configure.in (alpha*-*-netbsd*): Add target.
2886 * configure: Regenerate.
2887
0c0b87ad
JS		 28882002-04-04 Jeff Sturm <jsturm@one-point.com>
2889
2890 * src/sparc/v8.S: Add unwind info.
2891 * src/sparc/v9.S: Likewise.
2892
320d7e7a
KW		 28932002-03-30 Krister Walfridsson <cato@df.lth.se>
2894
2895 * configure.in: Enable i*86-*-netbsdelf*.
2896 * configure: Rebuilt.
4977bab6 2897
c75e51ed
DB		 28982002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
2899
2900 PR other/2620
2901 * src/mips/n32.s: Delete
2902 * src/mips/o32.s: Delete
2903
5bb366e0
LR		 29042002-03-21 Loren J. Rittle <ljrittle@acm.org>
2905
2906 * configure.in: Enable alpha*-*-freebsd*.
2907 * configure: Rebuilt.
2908
ce3b24a7
BM		 29092002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2910
2911 * Makefile.am: libfficonvenience -> libffi_convenience.
2912 * Makefile.in: Rebuilt.
4977bab6 2913
1eb7ba30
BM		 2914 * Makefile.am: Define ffitest_OBJECTS.
2915 * Makefile.in: Rebuilt.
ce3b24a7 2916
073ae293
AT		 29172002-03-07 Andreas Tobler <toa@pop.agri.ch>
2918 David Edelsohn <edelsohn@gnu.org>
2919
2920 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
2921 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
2922 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
2923 * Makefile.in: Regenerate.
2924 * include/ffi.h.in: Add AIX and Darwin closure definitions.
2925 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
2926 (flush_icache, flush_range): New functions.
2927 (ffi_closure_helper_DARWIN): New function.
2928 * src/powerpc/aix_closure.S: New file.
2929 * src/powerpc/darwin_closure.S: New file.
4977bab6 2930
02c521f4
JS		 29312002-02-24 Jeff Sturm <jsturm@one-point.com>
2932
2933 * include/ffi.h.in: Add typedef for ffi_arg.
2934 * src/ffitest.c (main): Declare rint with ffi_arg.
4977bab6 2935
287dd527
AT		 29362002-02-21 Andreas Tobler <toa@pop.agri.ch>
2937
2938 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
2939 number of GPRs for floating-point arguments.
2940
e2f2e70a
AG		 29412002-01-31 Anthony Green <green@redhat.com>
2942
2943 * configure: Rebuilt.
2944 * configure.in: Replace CHECK_SIZEOF and endian tests with
2945 cross-compiler friendly macros.
2946 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
2947 macros.
2948
15b9b32f
DE		 29492002-01-18 David Edelsohn <edelsohn@gnu.org>
2950
2951 * src/powerpc/darwin.S (_ffi_call_AIX): New.
1f7d5413 2952 * src/powerpc/aix.S (ffi_call_DARWIN): New.
15b9b32f 2953
2f4006f8
DE		 29542002-01-17 David Edelsohn <edelsohn@gnu.org>
2955
2956 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
2957 (TARGET_SRC_POWERPC_AIX): New.
2958 (POWERPC_AIX): New stanza.
2959 * Makefile.in: Regenerate.
2960 * configure.in: Add AIX case.
2961 * configure: Regenerate.
2962 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
2963 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
2964 size. Fix "long double" support.
2965 (ffi_call): Add FFI_AIX case.
2966 * src/powerpc/aix.S: New.
2967
3c017635
JH		 29682001-10-09 John Hornkvist <john@toastedmarshmallow.com>
2969
2970 Implement Darwin PowerPC ABI.
2971 * configure.in: Handle powerpc-*-darwin*.
2972 * Makefile.am: Set source files for POWERPC_DARWIN.
2973 * configure: Rebuilt.
2974 * Makefile.in: Rebuilt.
2975 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
2976 POWERPC_DARWIN.
2977 * src/powerpc/darwin.S: New file.
2978 * src/powerpc/ffi_darwin.c: New file.
2979
cc712abf
JM		 29802001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
2981
2982 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
2983
03a8de49
RO		 29842001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2985
2986 * src/x86/sysv.S: Avoid gas-only .balign directive.
2987 Use C style comments.
2988
92a0e6c6
RO		 29892001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2990
2991 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
2992 Fixes PR bootstrap/3563.
2993
57482f58
RO		 29942001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2995
2996 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
2997
d7e2dd2f
RO		 29982001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2999
3000 * configure.in: Recognize sparc*-sun-* host.
3001 * configure: Regenerate.
4977bab6 3002
be72207c
AH		 30032001-06-06 Andrew Haley <aph@redhat.com>
3004
4075db8f 3005 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
be72207c 3006
552cb45a
AH		 30072001-06-03 Andrew Haley <aph@redhat.com>
3008
4075db8f
BT		 3009 * src/alpha/osf.S: Add unwind info.
3010 * src/powerpc/sysv.S: Add unwind info.
3011 * src/powerpc/ppc_closure.S: Likewise.
552cb45a 3012
7d45b968
JS		 30132000-05-31 Jeff Sturm <jsturm@one-point.com>
3014
3015 * configure.in: Fix AC_ARG_ENABLE usage.
3016 * configure: Rebuilt.
3017
2a7f63c4
BM		 30182001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3019
3020 * configure.in: Remove warning about beta code.
3021 * configure: Rebuilt.
3022
30232001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3024
03a55252
HB		 3025 * src/ia64/unix.S: Restore stack pointer when returning from
3026 ffi_closure_UNIX.
3027 * src/ia64/ffi.c: Fix typo in comment.
3028
df5da1c4
JW		 30292001-04-18 Jim Wilson <wilson@redhat.com>
3030
3031 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3032 to eliminate RAW DV.
3033
52c28977
BM		 30342001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3035
3036 * Makefile.am: Make a libtool convenience library.
3037 * Makefile.in: Rebuilt.
3038
3b905056
BM		 30392001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3040
3041 * configure.in: Use different syntax for subdirectory creation.
3042 * configure: Rebuilt.
3043
d6c4152a
JB		 30442001-03-27 Jon Beniston <jon@beniston.com>
3045
3046 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3047 * configure: Rebuilt.
3048 * Makefile.am: Added X86_WIN32 target support.
3049 * Makefile.in: Rebuilt.
3050
3051 * include/ffi.h.in: Added X86_WIN32 target support.
3052
3053 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3054 * src/types.c: Added X86_WIN32 target support.
3055
3056 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3057 stuff removed and made to work with CygWin's gas.
3058
675cdab3
BM		 30592001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3060
3061 * configure.in: Make target subdirectory in build dir.
3062 * Makefile.am: Override suffix based rules to specify correct output
3063 subdirectory.
3064 * Makefile.in: Rebuilt.
3065 * configure: Rebuilt.
3066
e5f8858a 30672001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
cc4c8975
KH		 3068
3069 * src/powerpc/ppc_closure.S: New file.
3070 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3071 involving long long and register pairs.
3072 (ffi_prep_closure): New function.
3073 (flush_icache): Likewise.
3074 (ffi_closure_helper_SYSV): Likewise.
3075 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3076 (FFI_TRAMPOLINE_SIZE): Likewise.
3077 (FFI_NATIVE_RAW_API): Likewise.
3078 * Makefile.in: Rebuilt.
3079 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3080 (TARGET_SRC_POWERPC): Likewise.
3081
9450a929
TT		 30822001-03-19 Tom Tromey <tromey@redhat.com>
3083
3084 * Makefile.in: Rebuilt.
3085 * Makefile.am (ffitest_LDFLAGS): New macro.
3086
4d2cf715
NC		 30872001-03-02 Nick Clifton <nickc@redhat.com>
3088
3089 * include/ffi.h.in: Remove RCS ident string.
3090 * include/ffi_mips.h: Remove RCS ident string.
3091 * src/debug.c: Remove RCS ident string.
3092 * src/ffitest.c: Remove RCS ident string.
3093 * src/prep_cif.c: Remove RCS ident string.
3094 * src/types.c: Remove RCS ident string.
3095 * src/alpha/ffi.c: Remove RCS ident string.
3096 * src/alpha/osf.S: Remove RCS ident string.
3097 * src/arm/ffi.c: Remove RCS ident string.
3098 * src/arm/sysv.S: Remove RCS ident string.
3099 * src/mips/ffi.c: Remove RCS ident string.
3100 * src/mips/n32.S: Remove RCS ident string.
3101 * src/mips/o32.S: Remove RCS ident string.
3102 * src/sparc/ffi.c: Remove RCS ident string.
3103 * src/sparc/v8.S: Remove RCS ident string.
3104 * src/sparc/v9.S: Remove RCS ident string.
3105 * src/x86/ffi.c: Remove RCS ident string.
3106 * src/x86/sysv.S: Remove RCS ident string.
3107
7e5fd99f
JM		 31082001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3109
3110 * include/ffi.h.in: Change sourceware.cygnus.com references to
3111 gcc.gnu.org.
3112
3f67ba6e
RH		 31132000-12-09 Richard Henderson <rth@redhat.com>
3114
3115 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3116 (ffi_closure_osf_inner): Index rather than increment avalue
3117 and arg_types. Give ffi_closure_osf the raw return value type.
3118 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3119 promotion.
3120
31212000-12-07 Richard Henderson <rth@redhat.com>
29fe0479
RH		 3122
3123 * src/raw_api.c (ffi_translate_args): Fix typo.
3124 (ffi_prep_closure): Likewise.
3125
3126 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3127 FFI_TRAMPOLINE_SIZE.
3128 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3129 cif->bytes for new ffi_call_osf implementation.
3130 (ffi_prep_args): Absorb into ...
3131 (ffi_call): ... here. Do all stack allocation here and
3132 avoid a callback function.
3133 (ffi_prep_closure, ffi_closure_osf_inner): New.
3134 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3135 (ffi_closure_osf): New.
3136
2f7bf5b7
AO		 31372000-09-10 Alexandre Oliva <aoliva@redhat.com>
3138
ed1c1a1a
AO		 3139 * config.guess, config.sub, install-sh: Removed.
3140 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3141 * Makefile.in: Rebuilt.
3142
2f7bf5b7
AO		 3143 * acinclude.m4: Include libtool macros from the top level.
3144 * aclocal.m4, configure: Rebuilt.
3145
e15a68e7
AO		 31462000-08-22 Alexandre Oliva <aoliva@redhat.com>
3147
3148 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3149 * configure: Rebuilt.
3150
7f8c11d5
SB		 31512000-05-11 Scott Bambrough <scottb@netwinder.org>
3152
4977bab6
ZW		 3153 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3154 memory correctly. Use conditional instructions, not branches where
7f8c11d5
SB		 3155 possible.
3156
c5acf51f
TT		 31572000-05-04 Tom Tromey <tromey@cygnus.com>
3158
3159 * configure: Rebuilt.
3160 * configure.in: Match `arm*-*-linux-*'.
3161 From Chris Dornan <cdornan@arm.com>.
3162
91bffbe9
JJ		 31632000-04-28 Jakub Jelinek <jakub@redhat.com>
3164
3165 * Makefile.am (SUBDIRS): Define.
3166 (AM_MAKEFLAGS): Likewise.
3167 (Multilib support.): Add section.
3168 * Makefile.in: Rebuilt.
3169 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
4075db8f
BT		 3170 New variables. Set for gcc using -print-multi-lib. Export them
3171 to libtool.
3172 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3173 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3174 for -shared links.
3175 (extra_compiler_flags_value, extra_compiler_flags): Check these
3176 for extra compiler options which need to be passed down in
3177 compiler_flags.
91bffbe9 3178
bd66587e
AG		 31792000-04-16 Anthony Green <green@redhat.com>
3180
3181 * configure: Rebuilt.
3182 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3183
3791773c
JJ		 31842000-04-14 Jakub Jelinek <jakub@redhat.com>
3185
4075db8f 3186 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3791773c
JJ		 3187 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3188 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3189 Replace all void * sizeofs with sizeof(int).
3190 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3191 different than DOUBLE.
3192 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3193 (ffi_prep_args_v9): New function.
3194 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3195 (ffi_V9_return_struct): New function.
3196 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3197 32bit code (not yet cross-arch calls).
3198 * src/sparc/v8.S: Add struct return delay nop.
3199 Handle long long.
3200 * src/sparc/v9.S: New file.
3201 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3202 is used on sparc64 only for structures larger than 32 bytes.
3203 Pass by reference for structures is done for structure arguments
3204 larger than 16 bytes.
3205 * src/ffitest.c (main): Use 64bit rint on sparc64.
3206 Run long long tests on sparc.
3207 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3208 sparc64.
3209 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3210 on sparc64.
3211 * configure.in (sparc-*-linux*): New supported target.
3212 (sparc64-*-linux*): Likewise.
3213 * configure: Rebuilt.
3214 * Makefile.am: Add v9.S to SPARC files.
3215 * Makefile.in: Likewise.
3216 (LINK): Surround $(CCLD) into double quotes, so that multilib
3217 compiles work correctly.
3218
db459595
APB		 32192000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3220
3221 * configure: Rebuilt.
3222 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3223 proposed by Bryce McKinlay.
3224
c1f63902
TT		 32252000-03-20 Tom Tromey <tromey@cygnus.com>
3226
3227 * Makefile.in: Hand edit for java_raw_api.lo.
3228
0f72dc9e
BM		 32292000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3230
3231 * config.guess, config.sub: Update from the gcc tree.
3232 Fix for PR libgcj/168.
3233
3c18f2d1
TT		 32342000-03-03 Tom Tromey <tromey@cygnus.com>
3235
323eee8a
TT		 3236 * Makefile.in: Fixed ia64 by hand.
3237
3c18f2d1
TT		 3238 * configure: Rebuilt.
3239 * configure.in (--enable-multilib): New option.
3240 (libffi_basedir): New subst.
3241 (AC_OUTPUT): Added multilib code.
3242
a2b10a92
TT		 32432000-03-02 Tom Tromey <tromey@cygnus.com>
3244
3245 * Makefile.in: Rebuilt.
3246 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3247 directory name.
3248
dc5de370
HB		 32492000-02-25 Hans Boehm <boehm@acm.org>
3250
3251 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3252 files.
3253 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3254 list.
3255 (ffi_prep_raw_closure): Use ffi_translate_args, not
3256 ffi_closure_translate.
3257 * src/java_raw_api.c: New file.
3258 * src/ffitest.c (closure_test_fn): New function.
3259 (main): Define `rint' as long long on IA64. Added new test when
3260 FFI_CLOSURES is defined.
3261 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3262 (ffi_abi): Recognize IA64.
3263 (ffi_raw): Added `flt' field.
3264 Added "Java raw API" code.
3265 * configure.in: Recognize ia64.
3266 * Makefile.am (TARGET_SRC_IA64): New macro.
3267 (libffi_la_common_SOURCES): Added java_raw_api.c.
3268 (libffi_la_SOURCES): Define in IA64 case.
3269
d79229bd
TT		 32702000-01-04 Tom Tromey <tromey@cygnus.com>
3271
3272 * Makefile.in: Rebuilt with newer automake.
3273
2c33b220
TT		 32741999-12-31 Tom Tromey <tromey@cygnus.com>
3275
3276 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3277
6ba9e3b0
TT		 32781999-09-01 Tom Tromey <tromey@cygnus.com>
3279
3280 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3281 undefs.
3282 * fficonfig.h.in: Rebuilt.
3283 * configure: Rebuilt.
3284 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3285 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3286 * acconfig.h: Don't #undef PACKAGE or VERSION.
3287
63e5e3e0
AG		 32881999-08-09 Anthony Green <green@cygnus.com>
3289
3290 * include/ffi.h.in: Try to work around messy header problem
3291 with PACKAGE and VERSION.
3292
3293 * configure: Rebuilt.
3294 * configure.in: Change version to 2.00-beta.
3295
4075db8f 3296 * fficonfig.h.in: Rebuilt.
63e5e3e0
AG		 3297 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3298
3299 * src/x86/ffi.c (ffi_raw_call): Rename.
3300
33011999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3302
3303 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3304 (ffi_prep_incoming_args_SYSV): Ditto.
3305 (ffi_prep_closure): Ditto.
3306 (ffi_closure_raw_SYSV): Ditto.
3307 (ffi_prep_raw_closure): More ditto.
3308 (ffi_call_raw): Final ditto.
3309
3310 * include/ffi.h.in: Add definitions for closure and raw API.
3311
3312 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4977bab6 3313 FFI_TYPE_UINT64.
63e5e3e0
AG		 3314
3315 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3316
3317 * src/raw_api.c: New file.
3318
3319 * include/ffi.h.in (ffi_raw): New type.
3320 (UINT_ARG, SINT_ARG): New defines.
3321 (ffi_closure, ffi_raw_closure): New types.
3322 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4977bab6 3323
63e5e3e0
AG		 3324 * configure.in: Add check for endianness and sizeof void*.
3325
3326 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3327 instead of directly.
3328
3329 * configure: Rebuilt.
3330
3331Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3332
3333 * configure.in: Add x86 and powerpc BeOS configurations.
3334 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3335
33361999-05-09 Anthony Green <green@cygnus.com>
3337
3338 * configure.in: Add warning about this being beta code.
3339 Remove src/Makefile.am from the picture.
3340 * configure: Rebuilt.
3341
3342 * Makefile.am: Move logic from src/Makefile.am. Add changes
3343 to support libffi as a target library.
3344 * Makefile.in: Rebuilt.
3345
3346 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3347 Upgraded to new autoconf, automake, libtool.
3348
3349 * README: Tweaks.
3350
3351 * LICENSE: Update copyright date.
3352
3353 * src/Makefile.am, src/Makefile.in: Removed.
3354
33551998-11-29 Anthony Green <green@cygnus.com>
3356
3357 * include/ChangeLog: Removed.
3358 * src/ChangeLog: Removed.
3359 * src/mips/ChangeLog: Removed.
3360 * src/sparc/ChangeLog: Remboved.
3361 * src/x86/ChangeLog: Removed.
3362
3363 * ChangeLog.v1: Created.
Mirror of https://gcc.gnu.org/git/gcc.git