]> 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
Makefile.am (write_entries_to_file_split): New variable.
[thirdparty/gcc.git] / libffi / ChangeLog
CommitLineData
2e250fa2
BK		 12008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
2 Andreas Tobler <a.tobler@schweiz.org>
3
4 * configure.ac: Add amd64-*-freebsd* target.
5 * configure: Regenerate.
6
7f92e242
L		 72008-01-30 H.J. Lu <hongjiu.lu@intel.com>
8
9 PR libffi/34612
10 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
11 returning struct.
12
13 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
14 tests.
15
95d4f544
DE		 162008-01-24 David Edelsohn <edelsohn@gnu.org>
17
18 * configure: Regenerate.
19
79ca8fc0
AT		 202008-01-06 Andreas Tobler <a.tobler@schweiz.org>
21
22 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
23
d7b07cc7
AT		 242008-01-05 Andreas Tobler <a.tobler@schweiz.org>
25
26 PR testsuite/32843
27 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
28 signed/unsigned int8/16 for X86_DARWIN.
29 Updated copyright info.
30 Handle one and two byte structs with special cif->flags.
31 * src/x86/ffitarget.h: Add special types for one and two byte structs.
32 Updated copyright info.
33 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
34 sysv.S
35 Remove code to pop args from the stack after call.
36 Special-case signed/unsigned for int8/16, one and two byte structs.
37 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
38 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
39 FFI_TYPE_SINT32.
40 Updated copyright info.
41
c94ae2e4
DD		 422007-12-08 David Daney <ddaney@avtrex.com>
43
44 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
45 SUBU, add with ADDU and use smaller code sequences.
46
3875b6d7 472007-12-07 David Daney <ddaney@avtrex.com>
44be66a1 48
3875b6d7
DD		 49 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
50 type.
51
4c42b3d8
DD		 522007-12-06 David Daney <ddaney@avtrex.com>
53
54 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
55 defined.
56 (ffi_java_raw): New typedef.
57 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
58 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
59 ffi_java_raw.
60 (ffi_java_raw_closure) : Same.
61 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
62 parameter types.
63 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
64 FFI_SIZEOF_JAVA_RAW.
65 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
66 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
67 sizeof(ffi_java_raw) for alignment calculations.
68 (ffi_java_ptrarray_to_raw): Same.
69 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
70 if FFI_SIZEOF_JAVA_RAW == 4.
71 (ffi_java_raw_to_rvalue): Same.
72 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
73 (ffi_java_translate_args): Same.
74 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
75 parameter types.
76 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
77
3ef50b62
DD		 782007-12-06 David Daney <ddaney@avtrex.com>
79
80 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
81 pointer values.
82
16287148
AT		 832007-12-01 Andreas Tobler <a.tobler@schweiz.org>
84
85 PR libffi/31937
86 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
87 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
88 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
89 set the NUM_FPR_ARG_REGISTERS according to.
90 Add support for potential soft-float support under hard-float
91 architecture.
92 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
93 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
94 to the FFI_LINUX_SOFT_FLOAT ABI.
95 (ffi_prep_cif_machdep): Likewise.
96 (ffi_closure_helper_SYSV): Likewise.
97 * src/powerpc/ppc_closure.S: Make sure not to store float/double
98 on archs where __NO_FPRS__ is true.
99 Add FFI_TYPE_UINT128 support.
100 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
101 Adjust copyright notice.
102
a4b7c9cd
AT		 1032007-11-25 Andreas Tobler <a.tobler@schweiz.org>
104
105 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
106 * include/ffi_common.h: ... here.
107 Update copyright.
108
e84296c6
AT		 1092007-11-17 Andreas Tobler <a.tobler@schweiz.org>
110
111 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
112 * src/powerpc/linux64.S: Likewise.
113 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
114 * testsuite/libffi.call/return_ldl.c: New test.
115
cb642590
AH		 1162007-09-04 <aph@redhat.com>
117
118 * src/arm/sysv.S (UNWIND): New.
119 (Whole file): Conditionally compile unwinder directives.
120 * src/arm/sysv.S: Add unwinder directives.
121
122 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
123 Only treat r0 as a struct address if we're actually returning a
124 struct by address.
125 Only copy the bytes that are actually within a struct.
126 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
127 is returned in r0, not passed by address.
128 (ffi_call): Allocate a word-sized temporary for the case where
129 a composite is returned in r0.
130 (ffi_prep_incoming_args_SYSV): Align as necessary.
131
1322007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
133
134 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
135 directly using the sys_cacheflush syscall.
136
1372007-07-27 Andrew Haley <aph@redhat.com>
138
139 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
140
8d5c5e35
MR		 1412007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
142
143 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
144 * configure.ac: Likewise.
145 * Makefile.in: Regenerate.
146 * include/Makefile.in: Likewise.
147 * testsuite/Makefile.in: Likewise.
148 * configure: Likewise.
149
787b0d19
DD		 1502007-08-24 David Daney <ddaney@avtrex.com>
151
152 * testsuite/libffi.call/return_sl.c: New test.
153
dfbb3e36
DD		 1542007-08-10 David Daney <ddaney@avtrex.com>
155
156 * testsuite/libffi.call/cls_multi_ushort.c,
157 testsuite/libffi.call/cls_align_uint16.c,
158 testsuite/libffi.call/nested_struct1.c,
159 testsuite/libffi.call/nested_struct3.c,
160 testsuite/libffi.call/cls_7_1_byte.c,
161 testsuite/libffi.call/nested_struct5.c,
162 testsuite/libffi.call/cls_double.c,
163 testsuite/libffi.call/nested_struct7.c,
164 testsuite/libffi.call/cls_sint.c,
165 testsuite/libffi.call/nested_struct9.c,
166 testsuite/libffi.call/cls_20byte1.c,
167 testsuite/libffi.call/cls_multi_sshortchar.c,
168 testsuite/libffi.call/cls_align_sint64.c,
169 testsuite/libffi.call/cls_3byte2.c,
170 testsuite/libffi.call/cls_multi_schar.c,
171 testsuite/libffi.call/cls_multi_uchar.c,
172 testsuite/libffi.call/cls_19byte.c,
173 testsuite/libffi.call/cls_9byte1.c,
174 testsuite/libffi.call/cls_align_float.c,
175 testsuite/libffi.call/closure_fn1.c,
176 testsuite/libffi.call/problem1.c,
177 testsuite/libffi.call/closure_fn3.c,
178 testsuite/libffi.call/cls_sshort.c,
179 testsuite/libffi.call/closure_fn5.c,
180 testsuite/libffi.call/cls_align_double.c,
181 testsuite/libffi.call/nested_struct.c,
182 testsuite/libffi.call/cls_2byte.c,
183 testsuite/libffi.call/nested_struct10.c,
184 testsuite/libffi.call/cls_4byte.c,
185 testsuite/libffi.call/cls_6byte.c,
186 testsuite/libffi.call/cls_8byte.c,
187 testsuite/libffi.call/cls_multi_sshort.c,
188 testsuite/libffi.call/cls_align_sint16.c,
189 testsuite/libffi.call/cls_align_uint32.c,
190 testsuite/libffi.call/cls_20byte.c,
191 testsuite/libffi.call/cls_float.c,
192 testsuite/libffi.call/nested_struct2.c,
193 testsuite/libffi.call/cls_5_1_byte.c,
194 testsuite/libffi.call/nested_struct4.c,
195 testsuite/libffi.call/cls_24byte.c,
196 testsuite/libffi.call/nested_struct6.c,
197 testsuite/libffi.call/cls_64byte.c,
198 testsuite/libffi.call/nested_struct8.c,
199 testsuite/libffi.call/cls_uint.c,
200 testsuite/libffi.call/cls_multi_ushortchar.c,
201 testsuite/libffi.call/cls_schar.c,
202 testsuite/libffi.call/cls_uchar.c,
203 testsuite/libffi.call/cls_align_uint64.c,
204 testsuite/libffi.call/cls_ulonglong.c,
205 testsuite/libffi.call/cls_align_longdouble.c,
206 testsuite/libffi.call/cls_1_1byte.c,
207 testsuite/libffi.call/cls_12byte.c,
208 testsuite/libffi.call/cls_3_1byte.c,
209 testsuite/libffi.call/cls_3byte1.c,
210 testsuite/libffi.call/cls_4_1byte.c,
211 testsuite/libffi.call/cls_6_1_byte.c,
212 testsuite/libffi.call/cls_16byte.c,
213 testsuite/libffi.call/cls_18byte.c,
214 testsuite/libffi.call/closure_fn0.c,
215 testsuite/libffi.call/cls_9byte2.c,
216 testsuite/libffi.call/closure_fn2.c,
217 testsuite/libffi.call/closure_fn4.c,
218 testsuite/libffi.call/cls_ushort.c,
219 testsuite/libffi.call/closure_fn6.c,
220 testsuite/libffi.call/cls_5byte.c,
221 testsuite/libffi.call/cls_align_pointer.c,
222 testsuite/libffi.call/cls_7byte.c,
223 testsuite/libffi.call/cls_align_sint32.c,
224 testsuite/libffi.special/unwindtest_ffi_call.cc,
225 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
226
89d9d98a
DD		 2272007-08-10 David Daney <ddaney@avtrex.com>
228
229 PR libffi/28313
230 * configure.ac: Don't treat mips64 as a special case.
231 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
232 * configure: Regenerate
233 * Makefile.in: Ditto.
234 * fficonfig.h.in: Ditto.
235 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
236 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
237 (FFI_DEFAULT_ABI): Set for n64 case.
238 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
239 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
240 (ffi_closure_N32): New function.
241 (.eh_frame): New section
242 * src/mips/o32.S: Clean up comments.
243 (ffi_closure_O32): Pass ffi_closure parameter in $12.
244 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
245 _MIPS_SIM == _ABIN32 throughout.
246 (FFI_MIPS_STOP_HERE): New, use in place of
247 ffi_stop_here.
248 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
249 to support n32/n64 ABIs.
250 (calc_n32_struct_flags): Rewrite.
251 (calc_n32_return_struct_flags): Remove unused variable. Reverse
252 position of flag bits.
253 (ffi_prep_cif_machdep): Rewrite n32 portion.
254 (ffi_call): Enable for n64. Add special handling for small structure
255 return values.
256 (ffi_prep_closure_loc): Add n32 and n64 support.
257 (ffi_closure_mips_inner_O32): Add cast to silence warning.
258 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
259
d23869d4
DD		 2602007-08-08 David Daney <ddaney@avtrex.com>
261
262 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
263 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
264 specifiers.
265 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
266 * testsuite/libffi.call/cls_sint.c (main): Ditto.
267 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
268 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
269 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
270 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
271 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
272 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
273 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
274 * testsuite/libffi.call/return_fl3.c (main): Ditto.
275 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
276 * testsuite/libffi.call/nested_struct.c (main): Ditto.
277 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
278 * testsuite/libffi.call/return_ll1.c (main): Ditto.
279 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
280 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
281 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
282 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
283 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
284 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
285 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
286 * testsuite/libffi.call/cls_uint.c (main): Ditto.
287 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
288 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
289 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
290 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
291 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
292 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
293 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
294 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
295 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
296
536e0a8f
AH		 2972007-08-07 Andrew Haley <aph@redhat.com>
298
299 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
300 checkin.
301
7665c9be
AH		 3022007-08-06 Andrew Haley <aph@redhat.com>
303
304 PR testsuite/32843
305 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
306 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
307 FFI_TYPE_SINT32.
308
58e11497
DD		 3092007-08-02 David Daney <ddaney@avtrex.com>
310
311 * testsuite/libffi.call/return_ul.c (main): Define return type as
312 ffi_arg. Use proper printf conversion specifier.
313
387cfbba
AH		 3142007-07-30 Andrew Haley <aph@redhat.com>
315
316 PR testsuite/32843
317 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
318 signed/unsigned int8/16.
319 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
320 Use a jump table.
321 Remove code to pop args from the stack after call.
322 Special-case signed/unsigned int8/16.
323 * testsuite/libffi.call/return_sc.c (main): Revert.
324
0ee0208e
RG		 3252007-07-26 Richard Guenther <rguenther@suse.de>
326
327 PR testsuite/32843
328 * testsuite/libffi.call/return_sc.c (main): Verify call
329 result as signed char, not ffi_arg.
330
04a49983
RO		 3312007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
332
333 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
334 * configure: Regenerate.
335
433e6a8c
DD		 3362007-07-11 David Daney <ddaney@avtrex.com>
337
338 * src/mips/ffi.c: Don't include sys/cachectl.h.
339 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
340 cacheflush().
341
0959e2b8
AH		 3422007-05-18 Aurelien Jarno <aurelien@aurel32.net>
343
344 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
345 from (ffi_prep_closure): ... this.
346 (FFI_INIT_TRAMPOLINE): Adjust.
347
3482005-12-31 Phil Blundell <pb@reciva.com>
349
350 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
351 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
352 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
353 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
354 (FFI_CLOSURES): Enable closure support.
355
3562007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
357
358 * testsuite/libffi.call/cls_multi_ushort.c,
359 testsuite/libffi.call/cls_align_uint16.c,
360 testsuite/libffi.call/nested_struct1.c,
361 testsuite/libffi.call/nested_struct3.c,
362 testsuite/libffi.call/cls_7_1_byte.c,
363 testsuite/libffi.call/cls_double.c,
364 testsuite/libffi.call/nested_struct5.c,
365 testsuite/libffi.call/nested_struct7.c,
366 testsuite/libffi.call/cls_sint.c,
367 testsuite/libffi.call/nested_struct9.c,
368 testsuite/libffi.call/cls_20byte1.c,
369 testsuite/libffi.call/cls_multi_sshortchar.c,
370 testsuite/libffi.call/cls_align_sint64.c,
371 testsuite/libffi.call/cls_3byte2.c,
372 testsuite/libffi.call/cls_multi_schar.c,
373 testsuite/libffi.call/cls_multi_uchar.c,
374 testsuite/libffi.call/cls_19byte.c,
375 testsuite/libffi.call/cls_9byte1.c,
376 testsuite/libffi.call/cls_align_float.c,
377 testsuite/libffi.call/closure_fn1.c,
378 testsuite/libffi.call/problem1.c,
379 testsuite/libffi.call/closure_fn3.c,
380 testsuite/libffi.call/cls_sshort.c,
381 testsuite/libffi.call/closure_fn5.c,
382 testsuite/libffi.call/cls_align_double.c,
383 testsuite/libffi.call/cls_2byte.c,
384 testsuite/libffi.call/nested_struct.c,
385 testsuite/libffi.call/nested_struct10.c,
386 testsuite/libffi.call/cls_4byte.c,
387 testsuite/libffi.call/cls_6byte.c,
388 testsuite/libffi.call/cls_8byte.c,
389 testsuite/libffi.call/cls_multi_sshort.c,
390 testsuite/libffi.call/cls_align_uint32.c,
391 testsuite/libffi.call/cls_align_sint16.c,
392 testsuite/libffi.call/cls_float.c,
393 testsuite/libffi.call/cls_20byte.c,
394 testsuite/libffi.call/cls_5_1_byte.c,
395 testsuite/libffi.call/nested_struct2.c,
396 testsuite/libffi.call/cls_24byte.c,
397 testsuite/libffi.call/nested_struct4.c,
398 testsuite/libffi.call/nested_struct6.c,
399 testsuite/libffi.call/cls_64byte.c,
400 testsuite/libffi.call/nested_struct8.c,
401 testsuite/libffi.call/cls_uint.c,
402 testsuite/libffi.call/cls_multi_ushortchar.c,
403 testsuite/libffi.call/cls_schar.c,
404 testsuite/libffi.call/cls_uchar.c,
405 testsuite/libffi.call/cls_align_uint64.c,
406 testsuite/libffi.call/cls_ulonglong.c,
407 testsuite/libffi.call/cls_align_longdouble.c,
408 testsuite/libffi.call/cls_1_1byte.c,
409 testsuite/libffi.call/cls_12byte.c,
410 testsuite/libffi.call/cls_3_1byte.c,
411 testsuite/libffi.call/cls_3byte1.c,
412 testsuite/libffi.call/cls_4_1byte.c,
413 testsuite/libffi.call/cls_6_1_byte.c,
414 testsuite/libffi.call/cls_16byte.c,
415 testsuite/libffi.call/cls_18byte.c,
416 testsuite/libffi.call/closure_fn0.c,
417 testsuite/libffi.call/cls_9byte2.c,
418 testsuite/libffi.call/closure_fn2.c,
419 testsuite/libffi.call/closure_fn4.c,
420 testsuite/libffi.call/cls_ushort.c,
421 testsuite/libffi.call/closure_fn6.c,
422 testsuite/libffi.call/cls_5byte.c,
423 testsuite/libffi.call/cls_align_pointer.c,
424 testsuite/libffi.call/cls_7byte.c,
425 testsuite/libffi.call/cls_align_sint32.c,
426 testsuite/libffi.special/unwindtest_ffi_call.cc,
427 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
428
12a27363
L		 4292007-07-05 H.J. Lu <hongjiu.lu@intel.com>
430
431 * aclocal.m4: Regenerated.
432
82a6cadf
PB		 4332007-06-02 Paolo Bonzini <bonzini@gnu.org>
434
435 * configure: Regenerate.
436
1cf3d07d
SE		 4372007-05-23 Steve Ellcey <sje@cup.hp.com>
438
439 * Makefile.in: Regenerate.
440 * configure: Regenerate.
441 * aclocal.m4: Regenerate.
442 * include/Makefile.in: Regenerate.
443 * testsuite/Makefile.in: Regenerate.
444
5635c26e
RZ		 4452007-05-10 Roman Zippel <zippel@linux-m68k.org>
446
447 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
448 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
449 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
450 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
451 (FFI_CLOSURES): Enable closure support.
452
e6fbf387
RZ		 4532007-05-10 Roman Zippel <zippel@linux-m68k.org>
454
455 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
456 * configure: Regenerate.
457 * fficonfig.h.in: Regenerate.
458 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
459 CFI_OFFSET,CFI_DEF_CFA): New macros.
460 (ffi_call_SYSV): Add callframe annotation.
461
b5296e52
RZ		 4622007-05-10 Roman Zippel <zippel@linux-m68k.org>
463
464 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
465 numerous test suite failures.
466 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
467
bd61d708
PB		 4682007-04-11 Paolo Bonzini <bonzini@gnu.org>
469
470 * Makefile.am (EXTRA_DIST): Bring up to date.
471 * Makefile.in: Regenerate.
472 * src/frv/eabi.S: Remove RCS keyword.
473
bf8da5fc
RH		 4742007-04-06 Richard Henderson <rth@redhat.com>
475
476 * configure.ac: Tidy target case.
477 (HAVE_LONG_DOUBLE): Allow the target to override.
478 * configure: Regenerate.
479 * include/ffi.h.in: Don't define ffi_type_foo if
480 LIBFFI_HIDE_BASIC_TYPES is defined.
481 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
482 to ffi_type_double.
483 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
484 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
485 (ffi_type_longdouble): Special case for Alpha. Don't define
486 if long double == double.
487
488 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
489 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
490 (ffi_call, ffi_closure_osf_inner): Likewise.
491 (ffi_closure_osf_inner): Likewise. Mark hidden.
492 (ffi_call_osf, ffi_closure_osf): Mark hidden.
493 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
494 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
495 (load_table): Handle 128-bit long double.
496
497 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
498
8669fed1
TT		 4992007-04-06 Tom Tromey <tromey@redhat.com>
500
501 PR libffi/31491:
502 * README: Fixed bug in example.
503
f0920e6c
JJ		 5042007-04-03 Jakub Jelinek <jakub@redhat.com>
505
506 * src/closures.c: Include sys/statfs.h.
507 (_GNU_SOURCE): Define on Linux.
508 (FFI_MMAP_EXEC_SELINUX): Define.
509 (selinux_enabled): New variable.
510 (selinux_enabled_check): New function.
511 (is_selinux_enabled): Define.
512 (dlmmap): Use it.
513
34a6ccda
UB		 5142007-03-24 Uros Bizjak <ubizjak@gmail.com>
515
516 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
517 Use 'volatile float sum' to create sum of floats to avoid false
518 negative due to excess precision on ix86 targets.
519 (main): Ditto.
520
8b9cf39e
AO		 5212007-03-08 Alexandre Oliva <aoliva@redhat.com>
522
523 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
524 patch.
525 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
526
18fa3240
AO		 5272007-03-07 Alexandre Oliva <aoliva@redhat.com>
528
529 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
530 (ffi_prep_closure_loc): New.
531 (ffi_prep_raw_closure_loc): New.
532 (ffi_prep_java_raw_closure_loc): New.
533 * src/closures.c: New file.
534 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
535 Replace sflags with exec_offset.
536 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
537 sub_segment_exec_offset): New macros.
538 (get_segment_flags, set_segment_flags, check_segment_merge): New
539 macros.
540 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
541 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
542 destroy_mspace): Use new macros.
543 (sys_alloc): Silence warning.
544 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
545 * Makefile.in: Rebuilt.
546 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
547 terms of ffi_prep_closure_loc.
548 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
549 from...
550 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
551 renamed version.
552 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
553 adjusted from...
554 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
555 the renamed version.
556 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
557 (ffi_prep_closure): ... this.
558 * src/pa/ffi.c: Likewise.
559 * src/cris/ffi.c: Likewise. Adjust.
560 * src/frv/ffi.c: Likewise.
561 * src/ia64/ffi.c: Likewise.
562 * src/mips/ffi.c: Likewise.
563 * src/powerpc/ffi_darwin.c: Likewise.
564 * src/s390/ffi.c: Likewise.
565 * src/sh/ffi.c: Likewise.
566 * src/sh64/ffi.c: Likewise.
567 * src/sparc/ffi.c: Likewise.
568 * src/x86/ffi64.c: Likewise.
569 * src/x86/ffi.c: Likewise.
570 (FFI_INIT_TRAMPOLINE): Adjust.
571 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
572 (ffi_prep_raw_closure): ... this.
573 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
574 (ffi_prep_closure): ... this.
575 (flush_icache): Adjust.
576
dd778331
AO		 5772007-03-07 Alexandre Oliva <aoliva@redhat.com>
578
579 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
580 Lea's malloc.
581
b5f8d49b
BM		 5822007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
583
584 * Makefile.am: Add dummy install-pdf target.
585 * Makefile.in: Regenerate
586
7a23933b
AK		 5872007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
588
589 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
590 ffi_closure_helper_SYSV): Add long double handling.
591
8aaf4681
JJ		 5922007-02-02 Jakub Jelinek <jakub@redhat.com>
593
594 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
595 immediately after bctrl instruction.
596
0eb5e2b9
AO		 5972007-01-18 Alexandre Oliva <aoliva@redhat.com>
598
599 * Makefile.am (all-recursive, install-recursive,
600 mostlyclean-recursive, clean-recursive, distclean-recursive,
601 maintainer-clean-recursive): Add missing targets.
602 * Makefile.in: Rebuilt.
603
f4f7486b
AT		 6042006-12-14 Andreas Tobler <a.tobler@schweiz.org>
605
606 * configure.ac: Add TARGET for x86_64-*-darwin*.
607 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
608 for X86_DARWIN.
609 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
610 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
611 * configure: Regenerate.
612 * Makefile.in: Regenerate.
613 * include/Makefile.in: Regenerate.
614 * testsuite/Makefile.in: Regenerate.
615 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
616 ffi_call only.
617
3d5eafe2
AT		 6182006-12-13 Andreas Tobler <a.tobler@schweiz.org>
619
620 * aclocal.m4: Regenerate with aclocal -I .. as written in the
621 Makefile.am.
622
d5523ce1 6232006-10-31 Geoffrey Keating <geoffk@apple.com>
58182de3
GK		 624
625 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
626 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
627 Darwin.
628 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
629 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
630
a3bd50cc
PB		 6312006-10-10 Paolo Bonzini <bonzini@gnu.org>
632 Sandro Tolaini <tolaini@libero.it>
633
634 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
635 conditional.
636 * configure: Regenerated.
637 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
638 (EXTRA_DIST): Add src/x86/darwin.S.
639 * Makefile.in: Regenerated.
640 * include/Makefile.in: Regenerated.
641 * testsuite/Makefile.in: Regenerated.
642
643 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
644 X86_WIN32, and additionally align stack to 16 bytes.
645 * src/x86/darwin.S: New, based on sysv.S.
646 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
647
b095810f
DD		 6482006-09-12 David Daney <ddaney@avtrex.com>
649
650 PR libffi/23935
651 * include/Makefile.am: Install both ffi.h and ffitarget.h in
652 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
653 * aclocal.m4: Regenerated for automake 1.9.6.
654 * Makefile.in: Regenerated.
655 * include/Makefile.in: Regenerated.
656 * testsuite/Makefile.in: Regenerated.
657
110b2db4
AT		 6582006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
659
660 * include/ffi_common.h (struct): Revert accidental commit.
661
8e5fde87
AT		 6622006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
663
664 * include/ffi_common.h: Remove lint directives.
665 * include/ffi.h.in: Likewise.
666
cb4132fe
TS		 6672006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
668
669 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
670 for 32-bit architectures.
671 * testsuite/libffi.call/return_ul.c: New test case.
672
0d24c3fa
DD		 6732006-07-19 David Daney <ddaney@avtrex.com>
674
675 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
676 xfail remains for mips64.
677
8665824c
CD		 6782006-05-23 Carlos O'Donell <carlos@codesourcery.com>
679
680 * Makefile.am: Add install-html target. Add install-html to .PHONY
681 * Makefile.in: Regenerate.
682 * aclocal.m4: Regenerate.
683 * include/Makefile.in: Regenerate.
684 * testsuite/Makefile.in: Regenerate.
685
84e2e370
JDA		 6862006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
687
688 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
689 stack slot.
690
ac6ed182
AT		 6912006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
692
693 * README: Remove notice about 'Crazy Comments'.
694 * src/debug.c: Remove lint directives. Cleanup white spaces.
695 * src/java_raw_api.c: Likewise.
696 * src/prep_cif.c: Likewise.
697 * src/raw_api.c: Likewise.
698 * src/ffitest.c: Delete. No longer needed, all test cases migrated
699 to the testsuite.
700 * src/arm/ffi.c: Remove lint directives.
701 * src/m32r/ffi.c: Likewise.
702 * src/pa/ffi.c: Likewise.
703 * src/powerpc/ffi.c: Likewise.
704 * src/powerpc/ffi_darwin.c: Likewise.
705 * src/sh/ffi.c: Likewise.
706 * src/sh64/ffi.c: Likewise.
707 * src/x86/ffi.c: Likewise.
708 * testsuite/libffi.call/float2.c: Likewise.
709 * testsuite/libffi.call/promotion.c: Likewise.
710 * testsuite/libffi.call/struct1.c: Likewise.
711
b7211585
AT		 7122006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
713
714 * src/pa/hpux32.S: Correct unwind offset calculation for
715 ffi_closure_pa32.
716 * src/pa/linux.S: Likewise.
717
86066f9b
JW		 7182006-04-12 James E Wilson <wilson@specifix.com>
719
720 PR libgcj/26483
721 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
722 (hfa_type_load): Call stf_spill.
723 (hfa_type_store): Call ldf_fill.
724 (ffi_call): Adjust calls to above routines. Add local temps for
725 macro result.
b7211585 726
615baed7
MK		 7272006-04-10 Matthias Klose <doko@debian.org>
728
729 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
730 directory names containing underscores.
731
8e7a2b58
JW		 7322006-04-07 James E Wilson <wilson@specifix.com>
733
734 * testsuite/libffi.call/float4.c: New testcase.
735
e6c300ef
JDA		 7362006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
737 Andreas Tobler <a.tobler@schweiz.ch>
738
739 * Makefile.am: Add PA_HPUX port.
740 * Makefile.in: Regenerate.
741 * include/Makefile.in: Likewise.
742 * testsuite/Makefile.in: Likewise.
743 * configure.ac: Add PA_HPUX rules.
744 * configure: Regenerate.
745 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
746 Add PA_HPUX and PA64_HPUX.
747 Rename FFI_LINUX ABI to FFI_PA32 ABI.
748 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
749 (FFI_TYPE_SMALL_STRUCT2): Define.
750 (FFI_TYPE_SMALL_STRUCT4): Likewise.
751 (FFI_TYPE_SMALL_STRUCT8): Likewise.
752 (FFI_TYPE_SMALL_STRUCT3): Redefine.
753 (FFI_TYPE_SMALL_STRUCT5): Likewise.
754 (FFI_TYPE_SMALL_STRUCT6): Likewise.
755 (FFI_TYPE_SMALL_STRUCT7): Likewise.
756 * src/pa/ffi.c (ROUND_DOWN): Delete.
757 (fldw, fstw, fldd, fstd): Use '__asm__'.
758 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
759 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
760 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
761 Simplify incrementing of stack slot variable. Change type of local
762 'n' to unsigned int.
763 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
764 double on PA_HPUX.
765 (ffi_prep_cif_machdep): Likewise.
766 (ffi_call): Likewise.
767 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
768 return type to ffi_status. Simplify incrementing of stack slot
769 variable. Only copy floating point argument registers when PA_LINUX
770 is true. Reformat debug statement.
771 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
772 FFI_TYPE_SMALL_STRUCT8.
773 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
774 declaration.
775 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
776 Add nops to cache flush. Add trampoline for PA_HPUX.
777 * src/pa/hpux32.S: New file.
778 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
779 ffi_prep_args_LINUX to ffi_prep_args_pa32.
780 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
781 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
782 argument type checks so that common argument types appear first.
783 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
784 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
785
75b8b1be
AM		 7862006-03-24 Alan Modra <amodra@bigpond.net.au>
787
788 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
789 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
790 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
791 FFI_TYPE_LONGDOUBLE.
792 (ffi_prep_args64): Assert using IBM extended double.
793 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
794 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
795 (ffi_call): Handle FFI_LINUX.
796 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
797 gpr3 return pointer as for struct return. Handle FFI_LINUX
798 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
799 unnecessarily.
800 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
801 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
802 Don't use r6 as pointer to results, instead use sp offset. Don't
803 make a special call to load lr with case table address, instead
804 use offset from previous call.
805 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
806 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
807 return.
808
697701ad
KK		 8092006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
810
811 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
812 passed with FP registers correctly.
813 (ffi_closure_helper_SYSV): Likewise.
814 * src/sh64/sysv.S: Likewise.
815
b295c471
AT		 8162006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
817
818 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
819 args and userdata unused.
820 (closure_test_fn1): Mark cif and userdata unused.
821 (main): Remove unused res.
822
9fdeb13b
AT		 8232006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
824
825 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
826 -O2, -O3, -Os and the warning flags -W -Wall.
827 * testsuite/libffi.special/special.exp: Likewise.
828 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
829 unused parameter unused for gcc or else do nothing.
830 * testsuite/libffi.special/ffitestcxx.h: Likewise.
831 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
832 and userdata unused.
833 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
834 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
835 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
836 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
837 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
838 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
839 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
840 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
841 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
842 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
843 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
844 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
845 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
846 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
847 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
848 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
849 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
850 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
851 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
852 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
853 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
854 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
855 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
856 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
857 Likewise.
858 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
859 Likewise.
860 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
861 Likewise.
862 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
863 void* to avoid compiler warning.
864 (main): Likewise.
865 (cls_struct_align_gn): Mark cif and userdata unused.
866 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
867 Likewise.
868 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
869 Likewise.
870 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
871 Likewise.
872 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
873 Likewise.
874 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
875 Likewise.
876 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
877 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
878 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
879 data unused.
880 (main): Cast res_call to silence gcc.
881 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
882 data unused.
883 (main): Cast res_call to silence gcc.
884 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
885 and data unused.
886 (main): Cast res_call to silence gcc.
887 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
888 data unused.
889 (main): Cast res_call to silence gcc.
890 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
891 data unused.
892 (main): Cast res_call to silence gcc.
893 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
894 and data unused.
895 (main): Cast res_call to silence gcc.
896 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
897 userdata unused.
898 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
899 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
900 userdata unused.
901 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
902 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
903 userdata unused.
904 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
905 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
906 userdata unused.
907 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
908 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
909 userdata unused.
910 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
911 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
912 and userdata unused.
913 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
914 userdata unused.
915 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
916 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
917 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
918 Cleanup white spaces.
919 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
920 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
921 cif and userdata unused.
922 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
923 Likewise.
924 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
925 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
926 formatters to silence gcc.
927 (B_gn): Mark cif and userdata unused.
928 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
929 unused.
930 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
931 (B_gn): Mark cif and userdata unused.
932 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
933 unused.
934 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
935 (B_gn): Mark cif and userdata unused.
936 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
937 unused.
938 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
939 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
940 * testsuite/libffi.call/problem1.c (stub): Likewise.
941 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
942 gcc.
943 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
944 in the last commit for this test case in the test case itself.
945 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
946 unused.
947 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
948 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
949 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
950 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
951 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
952 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
953
0a4e2bb5
KK		 9542006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
955
956 * src/sh/sysv.S: Fix register numbers in the FDE for
957 ffi_closure_SYSV.
958
723e5d5a
AT		 9592006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
960
961 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
962 declaration to avoid a false negative on ix86. See PR323.
963
e097f887
KK		 9642006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
965
966 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
967 and cast integer to void * if needed. Update the pointer to
968 the FP register saved area correctly.
969
1a4878f7
AT		 9702006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
971
972 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
973 is fixed.
974 * testsuite/libffi.call/nested_struct4.c: Likewise.
975
b47b5e08
AT		 9762006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
977
6429d58f
AT		 978 * testsuite/libffi.call/return_dbl.c: New test case.
979 * testsuite/libffi.call/return_dbl1.c: Likewise.
980 * testsuite/libffi.call/return_dbl2.c: Likewise.
981 * testsuite/libffi.call/return_fl.c: Likewise.
982 * testsuite/libffi.call/return_fl1.c: Likewise.
983 * testsuite/libffi.call/return_fl2.c: Likewise.
984 * testsuite/libffi.call/return_fl3.c: Likewise.
985 * testsuite/libffi.call/closure_fn6.c: Likewise.
986
b47b5e08
AT		 987 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
988 definition.
989 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
990 here to be used by other test cases too.
991
992 * testsuite/libffi.call/nested_struct10.c: New test case.
993 * testsuite/libffi.call/nested_struct9.c: Likewise.
994 * testsuite/libffi.call/nested_struct8.c: Likewise.
995 * testsuite/libffi.call/nested_struct7.c: Likewise.
996 * testsuite/libffi.call/nested_struct6.c: Likewise.
997 * testsuite/libffi.call/nested_struct5.c: Likewise.
998 * testsuite/libffi.call/nested_struct4.c: Likewise.
999
9abe31d4
AT		 10002006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1001
1002 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1003 * configure: Rebuilt.
1004
71aa1a70
JJ		 10052006-01-18 Jakub Jelinek <jakub@redhat.com>
1006
1007 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1008 instead do the shifting inline.
1009 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1010 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1011 and 8 byte structs, for the remaining struct sizes don't call
1012 __lshrdi3, instead do the shifting inline.
1013
8a39029d
TS		 10142005-12-07 Thiemo Seufer <ths@networkno.de>
1015
1016 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1017 missing parentheses.
1018 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1019 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1020 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1021 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1022 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1023 FA_0_0_OFF2.
1024 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1025 endianness bugs.
1026 (ffi_prep_closure): Improve trampoline instruction scheduling.
1027 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1028
cd4241aa
AM		 10292005-12-03 Alan Modra <amodra@bigpond.net.au>
1030
1031 * src/powerpc/ffi.c: Formatting.
1032 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1033 (ffi_prep_args64): Likewise.
1034
0363db46
GK		 10352005-09-30 Geoffrey Keating <geoffk@apple.com>
1036
1037 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1038 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1039
d674eb2f
TT		 10402005-09-26 Tom Tromey <tromey@redhat.com>
1041
1042 * testsuite/libffi.call/float1.c (value_type): New typedef.
1043 (CANARY): New define.
1044 (main): Check for result buffer overflow.
1045 * src/powerpc/linux64.S: Handle linux64 long double returns.
1046 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1047 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1048
2d9c5743
AM		 10492005-08-25 Alan Modra <amodra@bigpond.net.au>
1050
1051 PR target/23404
1052 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1053 homed fp args.
1054 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1055
8a42356f
JJ		 10562005-08-11 Jakub Jelinek <jakub@redhat.com>
1057
1058 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1059 (AH_BOTTOM): Add FFI_HIDDEN definition.
1060 * configure: Rebuilt.
1061 * fficonfig.h.in: Rebuilt.
1062 * src/powerpc/ffi.c (hidden): Remove.
1063 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1064 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1065 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1066 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1067 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1068 add FFI_HIDDEN to its prototype.
1069 (ffi_closure_SYSV_inner): New.
1070 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1071 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1072
863c5508
AS		 10732005-08-10 Alfred M. Szmidt <ams@gnu.org>
1074
1075 PR libffi/21819:
1076 * configure: Rebuilt.
1077 * configure.ac: Handle i*86-*-gnu*.
1078
6a26ea58
JJ		 10792005-08-09 Jakub Jelinek <jakub@redhat.com>
1080
1081 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1082 DW_CFA_offset_extended_sf rather than
1083 DW_CFA_GNU_negative_offset_extended.
1084 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1085
583a92e8
ST		 10862005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1087
1088 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1089 on sh3.
1090 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1091 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1092 partially on register.
1093 (ffi_closure_helper_SYSV): Likewise.
1094 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1095
5d0317fc
KK		 10962005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1097
1098 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1099 Remove empty line.
1100 * src/sh64/ffi.c (simple_type): Remove.
1101 (return_type): Handle small structures correctly.
1102 (ffi_prep_args): Likewise.
1103 (ffi_call): Likewise.
1104 (ffi_closure_helper_SYSV): Likewise.
1105 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1106 Emit position independent code if PIC and remove wrong datalabel
1107 prefixes from EH data.
1108
8d9fd5f5
AT		 11092005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1110
1111 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1112 * Makefile.in: Regenerate.
1113 * include/Makefile.in: Likewise.
1114 * testsuite/Makefile.in: Likewise.
1115 * configure.ac: Add POWERPC_FREEBSD rules.
1116 * configure: Regenerate.
1117 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1118 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1119 * src/powerpc/ffi.c: Add flags to handle small structure returns
1120 in ffi_call_SYSV.
1121 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1122 Aka FFI_SYSV.
1123 (ffi_closure_helper_SYSV): Likewise.
1124 * src/powerpc/ppc_closure.S: Add return types for small structures.
1125 * src/powerpc/sysv.S: Add bits to handle small structures for
1126 final SYSV 4 ABI.
1127
93c76f81
AT		 11282005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1129
1130 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1131 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1132 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1133
b65da304
RC		 11342005-07-05 Randolph Chung <tausq@debian.org>
1135
1136 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1137 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1138 structures. Kill compilation warnings.
1139 (ffi_closure_inner_LINUX): Print return values as hex in debug
1140 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1141 Properly handle 5-7 byte structure returns.
1142 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1143 (FFI_TYPE_SMALL_STRUCT2): Remove.
1144 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1145 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1146 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1147 (checksmst1, checksmst2): Remove.
1148 (checksmst3): Optimize handling of 3-byte struct returns.
1149 (checksmst567): Properly handle 5-7 byte struct returns.
1150
2572f1c6
RO		 11512005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1152
1153 PR libgcj/21943
1154 * src/mips/n32.S: Enforce PIC code.
1155 * src/mips/o32.S: Likewise.
1156
cc786582
RO		 11572005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1158
1159 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1160 * configure: Regenerate.
1161
6d0d376c
AM		 11622005-06-01 Alan Modra <amodra@bigpond.net.au>
1163
1164 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1165 to call ffi_closure_helper_SYSV. Append @local instead.
1166 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1167
a21946f5 11682005-05-17 Kelley Cook <kcook@gcc.gnu.org>
a097ef55
KC		 1169
1170 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1171 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1172 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1173 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1174 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1175
f4c767fd
MS		 11762005-05-09 Mike Stump <mrs@apple.com>
1177
1178 * configure: Regenerate.
1179
7e15b5fe
RH		 11802005-05-08 Richard Henderson <rth@redhat.com>
1181
1182 PR libffi/21285
1183 * src/alpha/osf.S: Update unwind into to match code.
1184
d56ea8d9
RH		 11852005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1186 Richard Henderson <rth@redhat.com>
1187
1188 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1189 bit 11 of flags.
1190 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1191 (ffi_prep_closure): Set carry bit if sse-used flag set.
1192 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1193 Only load sse registers if ssecount non-zero.
1194 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1195
11962005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
d948c201
RC		 1197
1198 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1199 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1200 * configure: Regenerate.
1201
fc0ad8d9
HPN		 12022005-04-20 Hans-Peter Nilsson <hp@axis.com>
1203
1204 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1205 have Tcl8.3-compatible intermediate variable.
1206
614348bb
SP		 12072005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1208 Hans-Peter Nilsson <hp@axis.com>
1209
1210 * Makefile.am: Add CRIS support.
1211 * configure.ac: Likewise.
29ede6db
HPN		 1212 * Makefile.in, configure, testsuite/Makefile.in,
1213 include/Makefile.in: Regenerate.
614348bb
SP		 1214 * src/cris: New directory.
1215 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1216 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
727e729b
HPN		 1217
1218 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1219 \r?\n in output tests.
1220
6dab073b
MS		 12212005-04-12 Mike Stump <mrs@apple.com>
1222
1223 * configure: Regenerate.
1224
51f55d7c
HB		 12252005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1226
1227 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1228
7efa52a3
SE		 12292005-03-30 Steve Ellcey <sje@cup.hp.com>
1230
1231 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1232 (ffi_sarg) Ditto.
1233 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1234 to 64 bits in ILP32 mode.
1235 Load 64 bits even for short data.
1236
1488177a
MS		 12372005-03-23 Mike Stump <mrs@apple.com>
1238
1239 * src/powerpc/darwin.S: Update for -m64 multilib.
1240 * src/powerpc/darwin_closure.S: Likewise.
1241
c6900272
ZW		 12422005-03-21 Zack Weinberg <zack@codesourcery.com>
1243
1244 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1245 Do not set tool_include_dir.
1246 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1247 Regenerate.
1248 * include/Makefile.am: Set gcc_version and toollibffidir.
1249 * include/Makefile.in: Regenerate.
1250
962eea0a
AH		 12512005-02-22 Andrew Haley <aph@redhat.com>
1252
1253 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1254 odd-numbered register pairs for 64-bit integer types.
1255
1b05e8bf
AT		 12562005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1257
1258 PR libffi/20104
1259 * testsuite/libffi.call/return_ll1.c: New test case.
1260
6027770b
JJ		 12612005-02-11 Janis Johnson <janis187@us.ibm.com>
1262
1263 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1264 * testsuite/libffi.call/float.c: Ditto.
1265 * testsuite/libffi.call/float2.c: Ditto.
1266 * testsuite/libffi.call/float3.c: Ditto.
1267
ee31ea68
AT		 12682005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1269
1270 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1271
c16be7d5
EB		 12722005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1273
1274 * testsuite/libffi.special/special.exp (cxx_options): Add
1275 -shared-libgcc.
1276
5df07255
RH		 12772004-12-31 Richard Henderson <rth@redhat.com>
1278
1279 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1280 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1281 offset parameters with a type parameter; deduce size and structure
1282 alignment. Update all users.
1283
81a69b13
RH		 12842004-12-31 Richard Henderson <rth@redhat.com>
1285
1286 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1287 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
c6900272 1288 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
81a69b13
RH		 1289 into ffi_prep_closure.
1290 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1291 from scratch.
1292
d833cfb0
RH		 12932004-12-27 Richard Henderson <rth@redhat.com>
1294
1295 * src/x86/unix64.S: Fix typo in unwind info.
1296
1a0f488c
RH		 12972004-12-25 Richard Henderson <rth@redhat.com>
1298
1299 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1300 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1301 (merge_classes): Check for it.
1302 (SSE_CLASS_P): New.
1303 (classify_argument): Pass byte_offset by value; perform all updates
1304 inside struct case.
1305 (examine_argument): Add classes argument; handle
1306 X86_64_COMPLEX_X87_CLASS.
1307 (ffi_prep_args): Merge into ...
1308 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1309 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1310 (ffi_fill_return_value): Remove.
1311 (ffi_prep_closure): Remove dead assert.
1312 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1313 Rewrite to use struct register_args instead of va_list. Create
1314 flags for handling structure returns.
1315 * src/x86/unix64.S: Remove dead strings.
1316 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1317 stack frame with ffi_call. Handle structure returns properly.
1318 (float2sse, floatfloat2sse, double2sse): Remove.
1319 (sse2float, sse2double, sse2floatfloat): Remove.
1320 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1321 to handle structure returns properly.
1322
7994d50c
DE		 13232004-12-08 David Edelsohn <edelsohn@gnu.org>
1324
1325 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1326 PICFLAG.
1327 * Makefile.in: Regenerated.
1328
afabd8e6
RS		 13292004-12-02 Richard Sandiford <rsandifo@redhat.com>
1330
1331 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1332 * configure, aclocal.m4, Makefile.in: Regenerate.
1333 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1334
2539bffe
KC		 13352004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1336
1337 * configure: Regenerate for libtool change.
1338
767f726e
KC		 13392004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1340
1341 * configure: Regenerate for libtool reversion.
1342
2c92e90f
KC		 13432004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1344
1345 * configure: Regenerate for libtool change.
1346
1df8e834
JDA		 13472004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1348
1349 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1350
13512004-11-23 Richard Sandiford <rsandifo@redhat.com>
1352
1353 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1354 of jal. Use an absolute encoding for the frame information.
1355
13562004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1357
1358 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1359 * acinclude.m4: Delete logic for sincludes.
1360 * aclocal.m4, Makefile.in, configure: Regenerate.
1361 * include/Makefile: Likewise.
1362 * testsuite/Makefile: Likewise.
1363
bf02e7fa
EB		 13642004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1365
1366 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1367 on a 8-byte boundary.
1368 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1369
f20459f1
RE		 13702004-10-27 Richard Earnshaw <rearnsha@arm.com>
1371
1372 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1373 long long values. Round stack allocation to a multiple of 8 bytes
1374 for ATPCS compatibility.
1375 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1376 names. Handle returning long long types. Add Thumb and interworking
1377 support. Improve soft-float code.
1378
fdc69597
RE		 13792004-10-27 Richard Earnshaw <rearnsha@arm.com>
1380
1381 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1382 (libffi_exit): New function.
1383 (libffi_init): Build the testglue wrapper if needed.
1384
1e02510f
EB		 13852004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1386
1387 PR other/18138
1388 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1389
d9a193b2
KI		 13902004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1391
1392 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1393
1fddb058
KK		 13942004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1395
1396 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1397 * testsuite/libffi.call/float3.c: New test case.
1398
e0d3b452
KK		 13992004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1400
1401 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1402 the function returning a structure pointed with R2.
1403 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1404 the structure return value if T bit set. Emit position
1405 independent code and EH data if PIC.
1406
a472e9d4
KI		 14072004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1408
1409 * Makefile.am: Add m32r support.
1410 * configure.ac: Likewise.
1411 * Makefile.in: Regenerate.
1412 * confiugre: Regenerate.
1413 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1414 (uint64, sint64, double, longdouble)
1415 * src/m32r: New directory.
1416 * src/m32r/ffi.c: New file.
1417 * src/m32r/sysv.S: Likewise.
1418 * src/m32r/ffitarget.h: Likewise.
1419
3d47e9ca
KK		 14202004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1421
1422 * testsuite/libffi.call/negint.c: New test case.
1423
fc9051dd
L		 14242004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1425
1426 PR libgcj/17465
1427 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1428 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1429 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1430 DYLD_LIBRARY_PATH.
1431
2010caea
AT		 14322004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1433
1434 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1435 * testsuite/libffi.call/promotion.c: Likewise.
1436 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1437 whitespaces.
1438 * testsuite/libffi.call/return_sc.c: Likewise.
1439 * testsuite/libffi.call/return_uc.c: Likewise.
1440
894be816
AT		 14412004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1442
1443 * src/powerpc/darwin.S: Fix comments and identation.
1444 * src/powerpc/darwin_closure.S: Likewise.
1445
2fe7404a
AT		 14462004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1447
1448 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1449 (ffi_prep_args): Handle longdouble arguments.
1450 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1451 longdouble.
1452 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1453 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1454 values.
1455 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1456 * src/types.c: Defined longdouble size and alignment for darwin.
1457
16070e45
AT		 14582004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1459
1460 * src/powerpc/aix.S: Remove whitespaces.
1461 * src/powerpc/aix_closure.S: Likewise.
1462 * src/powerpc/asm.h: Likewise.
1463 * src/powerpc/ffi.c: Likewise.
1464 * src/powerpc/ffitarget.h: Likewise.
1465 * src/powerpc/linux64.S: Likewise.
1466 * src/powerpc/linux64_closure.S: Likewise.
1467 * src/powerpc/ppc_closure.S: Likewise.
1468 * src/powerpc/sysv.S: Likewise.
1469
88d8c771
AG		 14702004-08-30 Anthony Green <green@redhat.com>
1471
1472 * Makefile.am: Add frv support.
1473 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1474 * configure.ac: Read configure.host.
1475 * configure.in: Read configure.host.
1476 * configure.host: New file. frv-elf needs libgloss.
1477 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
a23adb4d 1478 alignment. This is needed to frv and shouldn't harm the others.
88d8c771
AG		 1479 * include/ffi_common.h (ALIGN_DOWN): New macro.
1480 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1481
4c5f37d5
DD		 14822004-08-24 David Daney <daney@avtrex.com>
1483
1484 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1485 * testsuite/libffi.call/closure_fn1.c: Likewise.
1486 * testsuite/libffi.call/closure_fn2.c Likewise.
1487 * testsuite/libffi.call/closure_fn3.c: Likewise.
1488 * testsuite/libffi.call/closure_fn4.c: Likewise.
1489 * testsuite/libffi.call/closure_fn5.c: Likewise.
1490 * testsuite/libffi.call/cls_18byte.c: Likewise.
1491 * testsuite/libffi.call/cls_19byte.c: Likewise.
1492 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1493 * testsuite/libffi.call/cls_20byte.c: Likewise.
1494 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1495 * testsuite/libffi.call/cls_24byte.c: Likewise.
1496 * testsuite/libffi.call/cls_2byte.c: Likewise.
1497 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1498 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1499 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1500 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1501 * testsuite/libffi.call/cls_4byte.c: Likewise.
1502 * testsuite/libffi.call/cls_64byte.c: Likewise.
1503 * testsuite/libffi.call/cls_6byte.c: Likewise.
1504 * testsuite/libffi.call/cls_7byte.c: Likewise.
1505 * testsuite/libffi.call/cls_8byte.c: Likewise.
1506 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1507 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1508 * testsuite/libffi.call/cls_align_double.c: Likewise.
1509 * testsuite/libffi.call/cls_align_float.c: Likewise.
1510 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1511 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1512 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1513 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1514 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1515 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1516 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1517 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1518 * testsuite/libffi.call/cls_double.c: Likewise.
1519 * testsuite/libffi.call/cls_float.c: Likewise.
1520 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1521 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1522 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1523 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1524 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1525 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1526 * testsuite/libffi.call/cls_schar.c: Likewise.
1527 * testsuite/libffi.call/cls_sint.c: Likewise.
1528 * testsuite/libffi.call/cls_sshort.c: Likewise.
1529 * testsuite/libffi.call/cls_uchar.c: Likewise.
1530 * testsuite/libffi.call/cls_uint.c: Likewise.
1531 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1532 * testsuite/libffi.call/cls_ushort.c: Likewise.
1533 * testsuite/libffi.call/nested_struct.c: Likewise.
1534 * testsuite/libffi.call/nested_struct1.c: Likewise.
1535 * testsuite/libffi.call/nested_struct2.c: Likewise.
1536 * testsuite/libffi.call/nested_struct3.c: Likewise.
1537 * testsuite/libffi.call/problem1.c: Likewise.
1538 * testsuite/libffi.special/unwindtest.cc: Likewise.
1539 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1540 to zero.
1541 * testsuite/libffi.call/cls_16byte.c: Likewise.
1542 * testsuite/libffi.call/cls_5byte.c: Likewise.
1543
39dca114
DD		 15442004-08-23 David Daney <daney@avtrex.com>
1545
1546 PR libgcj/13141
1547 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1548 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1549 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1550 parameters and return types.
1551 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1552 (ffi_prep_closure): Ditto.
1553 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1554 alignment calculations.
1555 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1556 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1557
b790003a
CM		 15582004-08-14 Casey Marshall <csm@gnu.org>
1559
1560 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1561 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1562 integer (O32 ABI only).
1563 (ffi_prep_closure): new function.
1564 (ffi_closure_mips_inner_O32): new function.
c6900272 1565 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
b790003a
CM		 1566 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1567 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1568 64 bit integers correctly.
1569 (ffi_closure_O32): new function.
1570 Added DWARF-2 unwind info for both functions.
1571
547d71a1
AH		 15722004-08-10 Andrew Haley <aph@redhat.com>
1573
1574 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1575
98a5bb3b
RM		 15762004-08-01 Robert Millan <robertmh@gnu.org>
1577
1578 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1579 * configure: Regenerate.
1580
4d447364
MR		 15812004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1582
1583 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1584 and mmap() explicitly instead of relying on preset autoconf cache
1585 variables.
1586 * aclocal.m4: Regenerate.
1587 * configure: Regenerate.
1588
e5dce82f
UW		 15892004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1590
1591 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1592 (ffi_check_float_struct): Remove unused prototype.
1593
e6a0022b
GK		 15942004-06-30 Geoffrey Keating <geoffk@apple.com>
1595
1596 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1597 character on Darwin, use '\n\t' instead.
1598
5a2d52ca
MK		 15992004-06-26 Matthias Klose <doko@debian.org>
1600
1601 * libtool-version: Fix typo in revision/age.
1602
401844ef
MK		 16032004-06-17 Matthias Klose <doko@debian.org>
1604
1605 * libtool-version: New.
1606 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1607 * Makefile.in: Regenerate.
1608
94632047 16092004-06-15 Paolo Bonzini <bonzini@gnu.org>
076e5443
PB		 1610
1611 * Makefile.am: Remove useless multilib rules.
1612 * Makefile.in: Regenerate.
1613 * aclocal.m4: Regenerate with automake 1.8.5.
1614 * configure.ac: Remove useless multilib configury.
1615 * configure: Regenerate.
1616
94632047 16172004-06-15 Paolo Bonzini <bonzini@gnu.org>
9ac926b1
PB		 1618
1619 * .cvsignore: New file.
1620
37c849e4
JJ		 16212004-06-10 Jakub Jelinek <jakub@redhat.com>
1622
1623 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1624 fp_done.
1625 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1626 changed from 8.
1627
cb623172
SM		 16282004-06-06 Sean McNeil <sean@mcneil.com>
1629
1630 * configure.ac: Add x86_64-*-freebsd* support.
1631 * configure: Regenerate.
1632
c9b93377
JB		 16332004-04-26 Joe Buck <jbuck@welsh-buck.org>
1634
1635 Bug 15093
1636 * configure.ac: Test for existence of mmap and sys/mman.h before
1637 checking blacklist. Fix suggested by Jim Wilson.
1638 * configure: Regenerate.
1639
cf90fd1b
MA		 16402004-04-26 Matt Austern <austern@apple.com>
1641
1642 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1643 FDE location.
1644 * src/powerpc/darwin_closure.S: Likewise.
c6900272 1645
67781b44
AT		 16462004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1647
1648 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1649 error. Reported by Thomas Heller <theller@python.net>.
1650 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1651 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1652
14b078a8
MK		 16532004-03-20 Matthias Klose <doko@debian.org>
1654
1655 * src/pa/linux.S: Fix typo.
1656
d65a810f
MK		 16572004-03-19 Matthias Klose <doko@debian.org>
1658
67781b44 1659 * Makefile.am: Update.
d65a810f
MK		 1660 * Makefile.in: Regenerate.
1661 * src/pa/ffi.h.in: Remove.
1662 * src/pa/ffitarget.h: New file.
1663
16642004-02-10 Randolph Chung <tausq@debian.org>
1665
1666 * Makefile.am: Add PA support.
1667 * Makefile.in: Regenerate.
1668 * include/Makefile.in: Regenerate.
1669 * configure.ac: Add PA target.
1670 * configure: Regenerate.
1671 * src/pa/ffi.c: New file.
1672 * src/pa/ffi.h.in: Add PA support.
1673 * src/pa/linux.S: New file.
1674 * prep_cif.c: Add PA support.
1675
09cababc
HY		 16762004-03-16 Hosaka Yuji <hos@tamanegi.org>
1677
1678 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
1679 double.
1680 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
1681 with ecif->cif->flags.
1682 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
1683 with cif->flags.
1684 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
1685 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
1686 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
1687 sc_retstruct2b): Add for 1 or 2-bytes struct case.
1688
0e8d0c5f
KC		 16892004-03-15 Kelley Cook <kcook@gcc.gnu.org>
1690
1691 * configure.in: Rename file to ...
1692 * configure.ac: ... this.
1693 * fficonfig.h.in: Regenerate.
1694 * Makefile.in: Regenerate.
1695 * include/Makefile.in: Regenerate.
1696 * testsuite/Makefile.in: Regenerate.
1697
b8d50739
MA		 16982004-03-12 Matt Austern <austern@apple.com>
1699
1700 * src/powerpc/darwin.S: Fix EH information so it corresponds to
1701 changes in EH format resulting from addition of linkonce support.
1702 * src/powerpc/darwin_closure.S: Likewise.
09cababc 1703
3ed8e5b6
AT		 17042004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
1705 Paolo Bonzini <bonzini@gnu.org>
1706
1707 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
1708 Remove VPATH. Remove rules for object files. Remove multilib support.
1709 (AM_CCASFLAGS): Add.
1710 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
1711 (AC_PREREQ): Bump version to 2.59.
1712 (AC_INIT): Fill with version info and bug address.
1713 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
1714 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
1715 De-precious CC so that the right flags are passed down to multilibs.
1716 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
1717 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
1718 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
1719 * configure: Rebuilt.
1720 * aclocal.m4: Likewise.
1721 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
1722 * fficonfig.h.in: Likewise.
1723
254ea477
AS		 17242004-03-11 Andreas Schwab <schwab@suse.de>
1725
1726 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
1727 arguments from fp registers only for the first 8 parameter slots.
1728 Don't convert a float parameter when passed in memory.
1729
3beb1315
HPN		 17302004-03-09 Hans-Peter Nilsson <hp@axis.com>
1731
1732 * configure: Regenerate for config/accross.m4 correction.
1733
fecf735c
MK		 17342004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
1735
1736 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
1737 ecif->cif->bytes to bytes.
1738 (ffi_prep_cif_machdep): Add braces around nested if statement.
1739
4243752c
AM		 17402004-02-09 Alan Modra <amodra@bigpond.net.au>
1741
12001e0f
AM		 1742 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
1743
4243752c
AM		 1744 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
1745 (ffi_closure_helper_LINUX64): Fix typo.
1746 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
1747 for powerpc64-*-*.
1748 * testsuite/libffi.call/float.c: Likewise.
1749 * testsuite/libffi.call/float2.c: Likewise.
1750
bf310028
AM		 17512004-02-08 Alan Modra <amodra@bigpond.net.au>
1752
1753 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
1754 long double function return and long double arg handling.
1755 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
1756 Use "end_pfr" instead of "nf". Correct long double handling.
1757 Localise "temp".
1758 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
1759 return value.
1760 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
1761 space for long double return value. Adjust stack frame and offsets.
1762 Load f2 long double return.
1763
dd5f9ea2
AM		 17642004-02-07 Alan Modra <amodra@bigpond.net.au>
1765
1766 * src/types.c: Use 16 byte long double for POWERPC64.
1767
0ce78f01
EB		 17682004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1769
1770 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
1771 when the structure return address is passed in %o0.
1772 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
1773 (ffi_v9_layout_struct): Align the field following a nested structure
1774 on a word boundary. Use memmove instead of memcpy.
1775 (ffi_call): Update call to ffi_V9_return_struct.
1776 (ffi_prep_closure): Define 'ctx' only for V8.
1777 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
1778 and ffi_closure_sparc_inner_v9.
1779 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
1780 Always skip the structure return address. For structures and long
1781 doubles, copy the argument directly.
1782 (ffi_closure_sparc_inner_v9): Skip the structure return address only
1783 if required. Shift the maximum floating-point slot accordingly. For
1784 big structures, copy the argument directly; otherwise, left-justify the
1785 argument and call ffi_v9_layout_struct to lay out the structure on
1786 the stack.
1787 * src/sparc/v8.S: Undef STACKFRAME before defining it.
1788 (ffi_closure_v8): Pass the structure return address. Update call to
1789 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
1790 Skip the 'unimp' insn when returning long doubles and structures.
1791 * src/sparc/v9.S: Undef STACKFRAME before defining it.
1792 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
1793 FFI_TYPE_INT handling. Load structures both in integers and
1794 floating-point registers on return.
1795 * README: Update status of the SPARC port.
1796
7fcc1d9b
AT		 17972004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
1798
1799 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
1800 as of type ffi_arg.
1801 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
1802
6ed0ecaf
UW		 18032004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
1804
1805 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
1806 value as of type ffi_arg, not unsigned int.
1807
c930d8a5
MR		 18082004-01-21 Michael Ritzert <ritzert@t-online.de>
1809
1810 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
1811 of the LHS.
1812
7b2e3e60
AT		 18132004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
1814
1815 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
1816 Solaris.
1817
2e89756b
RO		 18182004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1819
1820 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
1821 to void *.
1822
628b26f8
RH		 18232003-12-10 Richard Henderson <rth@redhat.com>
1824
1825 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
1826 size_t instead of int.
1827
0ee646d6
HY		 18282003-12-04 Hosaka Yuji <hos@tamanegi.org>
1829
1830 * testsuite/libffi.call/many_win32.c: Include <float.h>.
1831 * testsuite/libffi.call/many_win32.c (main): Replace variable
1832 int i with unsigned long ul.
1833
1834 * testsuite/libffi.call/cls_align_uint64.c: New test case.
1835 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1836 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1837 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1838 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1839 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1840 * testsuite/libffi.call/cls_align_float.c: Likewise.
1841 * testsuite/libffi.call/cls_align_double.c: Likewise.
1842 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1843 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1844
2c4ed456
HY		 18452003-12-02 Hosaka Yuji <hos@tamanegi.org>
1846
1847 PR other/13221
1848 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
1849 Align arguments to 32 bits.
1850
d18c3386
AT		 18512003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
1852
1853 PR other/13221
1854 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
1855 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1856 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1857 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1858 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1859 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1860
1861 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
1862
88096b2a
KG		 18632003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
1864
1865 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
1866 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1867
3eee102e
AT		 18682003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
1869
1870 * Makefile.in: Rebuilt.
1871 * configure: Likewise.
1872 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
1873 the right type.
1874
5196736e
AJ		 18752003-11-21 Andreas Jaeger <aj@suse.de>
1876 Andreas Tobler <a.tobler@schweiz.ch>
1877
1878 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
1879 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
1880 * Makefile.in: Rebuilt.
1881 * aclocal.m4: Likewise.
1882 * configure: Likewise.
1883 * fficonfig.h.in: Likewise.
1884 * testsuite/lib/libffi-dg.exp: Add include dir.
1885 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
1886 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1887 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
1888 for ffi_closure if available.
1889 * testsuite/libffi.call/closure_fn1.c: Likewise.
1890 * testsuite/libffi.call/closure_fn2.c: Likewise.
1891 * testsuite/libffi.call/closure_fn3.c: Likewise.
1892 * testsuite/libffi.call/closure_fn4.c: Likewise.
1893 * testsuite/libffi.call/closure_fn5.c: Likewise.
1894 * testsuite/libffi.call/cls_12byte.c: Likewise.
1895 * testsuite/libffi.call/cls_16byte.c: Likewise.
1896 * testsuite/libffi.call/cls_18byte.c: Likewise.
1897 * testsuite/libffi.call/cls_19byte.c: Likewise.
1898 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1899 * testsuite/libffi.call/cls_20byte.c: Likewise.
1900 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1901 * testsuite/libffi.call/cls_24byte.c: Likewise.
1902 * testsuite/libffi.call/cls_2byte.c: Likewise.
1903 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1904 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1905 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1906 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1907 * testsuite/libffi.call/cls_4byte.c: Likewise.
1908 * testsuite/libffi.call/cls_5byte.c: Likewise.
1909 * testsuite/libffi.call/cls_64byte.c: Likewise.
1910 * testsuite/libffi.call/cls_6byte.c: Likewise.
1911 * testsuite/libffi.call/cls_7byte.c: Likewise.
1912 * testsuite/libffi.call/cls_8byte.c: Likewise.
1913 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1914 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1915 * testsuite/libffi.call/cls_double.c: Likewise.
1916 * testsuite/libffi.call/cls_float.c: Likewise.
1917 * testsuite/libffi.call/cls_schar.c: Likewise.
1918 * testsuite/libffi.call/cls_sint.c: Likewise.
1919 * testsuite/libffi.call/cls_sshort.c: Likewise.
1920 * testsuite/libffi.call/cls_uchar.c: Likewise.
1921 * testsuite/libffi.call/cls_uint.c: Likewise.
1922 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1923 * testsuite/libffi.call/cls_ushort.c: Likewise.
1924 * testsuite/libffi.call/nested_struct.c: Likewise.
1925 * testsuite/libffi.call/nested_struct1.c: Likewise.
1926 * testsuite/libffi.call/nested_struct2.c: Likewise.
1927 * testsuite/libffi.call/nested_struct3.c: Likewise.
1928 * testsuite/libffi.call/problem1.c: Likewise.
1929 * testsuite/libffi.special/unwindtest.cc: Likewise.
1930
9e6c3ecb
AT		 19312003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
1932
1933 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
1934
70d8f2a1
AT		 19352003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
1936
1937 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
1938 Add -lgcc_s to additional flags.
1939
ea81d2a3
AT		 19402003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
1941
1942 * configure.in, include/Makefile.am: PR libgcj/11147, install
1943 the ffitarget.h header file in a gcc versioned and target
1944 dependent place.
1945 * configure: Regenerated.
1946 * Makefile.in, include/Makefile.in: Likewise.
1947 * testsuite/Makefile.in: Likewise.
1948
647af4a9
AT		 19492003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
1950
1951 * testsuite/libffi.call/closure_fn0.c: Print result and check
1952 with dg-output to make debugging easier.
1953 * testsuite/libffi.call/closure_fn1.c: Likewise.
1954 * testsuite/libffi.call/closure_fn2.c: Likewise.
1955 * testsuite/libffi.call/closure_fn3.c: Likewise.
1956 * testsuite/libffi.call/closure_fn4.c: Likewise.
1957 * testsuite/libffi.call/closure_fn5.c: Likewise.
1958 * testsuite/libffi.call/cls_12byte.c: Likewise.
1959 * testsuite/libffi.call/cls_16byte.c: Likewise.
1960 * testsuite/libffi.call/cls_18byte.c: Likewise.
1961 * testsuite/libffi.call/cls_19byte.c: Likewise.
1962 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1963 * testsuite/libffi.call/cls_20byte.c: Likewise.
1964 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1965 * testsuite/libffi.call/cls_24byte.c: Likewise.
1966 * testsuite/libffi.call/cls_2byte.c: Likewise.
1967 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1968 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1969 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1970 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1971 * testsuite/libffi.call/cls_4byte.c: Likewise.
1972 * testsuite/libffi.call/cls_5byte.c: Likewise.
1973 * testsuite/libffi.call/cls_64byte.c: Likewise.
1974 * testsuite/libffi.call/cls_6byte.c: Likewise.
1975 * testsuite/libffi.call/cls_7byte.c: Likewise.
1976 * testsuite/libffi.call/cls_8byte.c: Likewise.
1977 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1978 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1979 * testsuite/libffi.call/cls_double.c: Likewise.
1980 * testsuite/libffi.call/cls_float.c: Likewise.
1981 * testsuite/libffi.call/cls_schar.c: Likewise.
1982 * testsuite/libffi.call/cls_sint.c: Likewise.
1983 * testsuite/libffi.call/cls_sshort.c: Likewise.
1984 * testsuite/libffi.call/cls_uchar.c: Likewise.
1985 * testsuite/libffi.call/cls_uint.c: Likewise.
1986 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1987 * testsuite/libffi.call/cls_ushort.c: Likewise.
1988 * testsuite/libffi.call/problem1.c: Likewise.
1989
1990 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
1991 static.
1992
ac47cc13
AT		 19932003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
1994
1995 * testsuite/libffi.call/cls_9byte2.c: New test case.
1996 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1997 * testsuite/libffi.call/cls_64byte.c: Likewise.
1998 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1999 * testsuite/libffi.call/cls_19byte.c: Likewise.
2000 * testsuite/libffi.call/cls_18byte.c: Likewise.
2001 * testsuite/libffi.call/closure_fn4.c: Likewise.
2002 * testsuite/libffi.call/closure_fn5.c: Likewise.
2003 * testsuite/libffi.call/cls_schar.c: Likewise.
2004 * testsuite/libffi.call/cls_sint.c: Likewise.
2005 * testsuite/libffi.call/cls_sshort.c: Likewise.
2006 * testsuite/libffi.call/nested_struct2.c: Likewise.
2007 * testsuite/libffi.call/nested_struct3.c: Likewise.
2008
f145ed6b
AT		 20092003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2010
2011 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2012 * testsuite/libffi.call/cls_uchar.c: Likewise.
2013 * testsuite/libffi.call/cls_uint.c: Likewise.
2014 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2015 * testsuite/libffi.call/cls_ushort.c: Likewise.
2016 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2017
2b04d015
AT		 20182003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2019
2020 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2021 the initialization.
2022
57f542b3
AT		 20232003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2024
2b04d015 2025 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
57f542b3
AT		 2026 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2027
e61acf72
DD		 20282003-10-22 David Daney <ddaney@avtrex.com>
2029
2030 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2031 __attribute__((__mode__(__SI__))) and friends.
2032
2461c08b
AS		 20332003-10-22 Andreas Schwab <schwab@suse.de>
2034
2035 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2036
1450eb7a
AT		 20372003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2038
2039 * configure.in: AC_LINK_FILES(ffitarget.h).
2040 * configure: Regenerate.
2041 * Makefile.in: Likewise.
2042 * include/Makefile.in: Likewise.
2043 * testsuite/Makefile.in: Likewise.
2044 * fficonfig.h.in: Likewise.
2045
20462003-10-21 Paolo Bonzini <bonzini@gnu.org>
2047 Richard Henderson <rth@redhat.com>
2048
2049 Avoid that ffi.h includes fficonfig.h.
2050
2051 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2052 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2053 (TARGET_SRC_MIPS_SGI): Removed.
2054 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2055 (MIPS_SGI): Removed.
2056 (CLEANFILES): Removed.
2057 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2058 targets.
2059 * acconfig.h: Removed.
2060 * configure.in: Compute sizeofs only for double and long double.
2061 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2062 into AC_DEFINE instead of using acconfig.h. Create
2063 include/ffitarget.h instead of include/fficonfig.h. Rename
2064 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2065 AC_DEFINE EH_FRAME_FLAGS.
2066 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2067 (hack_DATA): Add ffitarget.h.
2068 * include/ffi.h.in: Remove all system specific definitions.
2069 Declare raw API even if it is not installed, why bother?
2070 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2071 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2072 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2073 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2074 * include/ffi_common.h (bool): Do not define.
2075 (ffi_assert): Accept failed assertion.
2076 (ffi_type_test): Return void and accept file/line.
2077 (FFI_ASSERT): Pass stringized failed assertion.
2078 (FFI_ASSERT_AT): New macro.
2079 (FFI_ASSERT_VALID_TYPE): New macro.
2080 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2081 UINT64, SINT64): Define here with gcc's __attribute__ macro
2082 instead of in ffi.h
2083 (FLOAT32, ALIGN): Define here instead of in ffi.h
2084 * include/ffi-mips.h: Removed. Its content moved to
2085 src/mips/ffitarget.h after separating assembly and C sections.
2086 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2087 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2088 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2089 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2090 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2091 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2092 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2093 (ffi_type_test): Rewritten.
2094 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2095 FFI_ASSERT_VALID_TYPE.
2096 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2097 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2098 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2099 src/s390/ffitarget.h, src/sh/ffitarget.h,
2100 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2101 src/x86/ffitarget.h: New files.
2102 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2103 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2104 src/powerpc/aix.S, src/powerpc/darwin.S,
2105 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2106 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2107 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2108 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2109 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2110 include fficonfig.h
2111
27d54b2a
RO		 21122003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2113
2114 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2115 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2116
0df6970d
AT		 21172003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2118
2119 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2120 Used when FFI_DEBUG = 1.
2121
2612ffce
AM		 21222003-10-14 Alan Modra <amodra@bigpond.net.au>
2123
2124 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2125 and align.
2126
d54f7354
RO		 21272003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2128
2129 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2130 FFI_MIPS_O32 for O32 ABI.
2131
7a54c850
AT		 21322003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2133
2134 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2135 SPARC64. Cleanup whitespaces.
2136
9e3bfa9b
AT		 21372003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2138
2139 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2140 strongarm, xscale. Cleanup whitespaces.
2141 * testsuite/libffi.call/closure_fn1.c: Likewise.
2142 * testsuite/libffi.call/closure_fn2.c: Likewise.
2143 * testsuite/libffi.call/closure_fn3.c: Likewise.
2144 * testsuite/libffi.call/cls_12byte.c: Likewise.
2145 * testsuite/libffi.call/cls_16byte.c: Likewise.
2146 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2147 * testsuite/libffi.call/cls_20byte.c: Likewise.
2148 * testsuite/libffi.call/cls_24byte.c: Likewise.
2149 * testsuite/libffi.call/cls_2byte.c: Likewise.
2150 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2151 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2152 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2153 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2154 * testsuite/libffi.call/cls_4byte.c: Likewise.
2155 * testsuite/libffi.call/cls_5byte.c: Likewise.
2156 * testsuite/libffi.call/cls_6byte.c: Likewise.
2157 * testsuite/libffi.call/cls_7byte.c: Likewise.
2158 * testsuite/libffi.call/cls_8byte.c: Likewise.
2159 * testsuite/libffi.call/cls_double.c: Likewise.
2160 * testsuite/libffi.call/cls_float.c: Likewise.
2161 * testsuite/libffi.call/cls_uchar.c: Likewise.
2162 * testsuite/libffi.call/cls_uint.c: Likewise.
2163 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2164 * testsuite/libffi.call/cls_ushort.c: Likewise.
2165 * testsuite/libffi.call/nested_struct.c: Likewise.
2166 * testsuite/libffi.call/nested_struct1.c: Likewise.
2167 * testsuite/libffi.call/problem1.c: Likewise.
2168 * testsuite/libffi.special/unwindtest.cc: Likewise.
2169 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2170
3e86b595
DE		 21712003-09-18 David Edelsohn <edelsohn@gnu.org>
2172
2173 * src/powerpc/aix.S: Cleanup whitespaces.
2174 * src/powerpc/aix_closure.S: Likewise.
2175
8d8369f8
AT		 21762003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2177
2178 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2179 * src/powerpc/darwin_closure.S: Likewise.
2180 * src/powerpc/ffi_darwin.c: Likewise.
2181
69b182fd
AT		 21822003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2183 David Edelsohn <edelsohn@gnu.org>
2184
2185 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2186 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2187 parameter stack.
2188 * src/powerpc/darwin_closure.S: Likewise.
2189 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2190 according to the Darwin/AIX ABI.
2191 (ffi_prep_cif_machdep): Likewise.
2192 (ffi_closure_helper_DARWIN): Likewise.
2193 Remove the outgoing parameter stack logic. Simplify the evaluation
2194 of the different CASE types.
2195 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2196 statement in the trampoline code.
2197
df805cdc
KK		 21982003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2199
2200 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2201 for the register size.
2202 (ffi_closure_helper_SYSV): Handle the structure return value
2203 address correctly.
2204 (ffi_closure_helper_SYSV): Return the appropriate type when
2205 the registers are used for the structure return value.
2206 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2207 the 64-bit return value. Update copyright years.
2208
eb26c76c
RO		 22092003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2210
2211 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2212 srcdir for ffi_mips.h.
2213
6350c02c
AM		 22142003-09-12 Alan Modra <amodra@bigpond.net.au>
2215
2216 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2217 structure size.
2218 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2219 placement of float result.
2220 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2221 cast of "resp" for big-endian 64 bit machines.
2222
5af1c806
AM		 22232003-09-11 Alan Modra <amodra@bigpond.net.au>
2224
2225 * src/types.c (double, longdouble): Merge identical SH and ARM
2226 typedefs, and add POWERPC64.
2227 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2228 struct split over gpr and rest.
2229 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2230 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2231
086a4bd7
AT		 22322003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2233
2234 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2235 passing correctly.
c930d8a5 2236
e517f636
AM		 22372003-09-09 Alan Modra <amodra@bigpond.net.au>
2238
2239 * configure: Regenerate.
2240
6f20d627
AT		 22412003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2242
2243 * Makefile.am: Remove build rules for ffitest.
2244 * Makefile.in: Rebuilt.
c930d8a5 2245
6f20d627 22462003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
226660af
AT		 2247
2248 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2249 about implicit declaration of abort().
2250
bbf3057b
AT		 22512003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2252
2253 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2254 * Makefile.in: Rebuilt.
2255 * configure.in: Add dejagnu test framework.
2256 * configure: Rebuilt.
2257
2258 * testsuite/Makefile.am: New file.
2259 * testsuite/Makefile.in: Built
c930d8a5 2260 * testsuite/lib/libffi-dg.exp: New file.
bbf3057b
AT		 2261 * testsuite/config/default.exp: Likewise.
2262 * testsuite/libffi.call/call.exp: Likewise.
2263 * testsuite/libffi.call/ffitest.h: Likewise.
2264 * testsuite/libffi.call/closure_fn0.c: Likewise.
2265 * testsuite/libffi.call/closure_fn1.c: Likewise.
2266 * testsuite/libffi.call/closure_fn2.c: Likewise.
2267 * testsuite/libffi.call/closure_fn3.c: Likewise.
2268 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2269 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2270 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2271 * testsuite/libffi.call/cls_2byte.c: Likewise.
2272 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2273 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2274 * testsuite/libffi.call/cls_4byte.c: Likewise.
2275 * testsuite/libffi.call/cls_5byte.c: Likewise.
2276 * testsuite/libffi.call/cls_6byte.c: Likewise.
2277 * testsuite/libffi.call/cls_7byte.c: Likewise.
2278 * testsuite/libffi.call/cls_8byte.c: Likewise.
2279 * testsuite/libffi.call/cls_12byte.c: Likewise.
2280 * testsuite/libffi.call/cls_16byte.c: Likewise.
2281 * testsuite/libffi.call/cls_20byte.c: Likewise.
2282 * testsuite/libffi.call/cls_24byte.c: Likewise.
2283 * testsuite/libffi.call/cls_double.c: Likewise.
2284 * testsuite/libffi.call/cls_float.c: Likewise.
2285 * testsuite/libffi.call/cls_uchar.c: Likewise.
2286 * testsuite/libffi.call/cls_uint.c: Likewise.
2287 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2288 * testsuite/libffi.call/cls_ushort.c: Likewise.
2289 * testsuite/libffi.call/float.c: Likewise.
2290 * testsuite/libffi.call/float1.c: Likewise.
2291 * testsuite/libffi.call/float2.c: Likewise.
2292 * testsuite/libffi.call/many.c: Likewise.
2293 * testsuite/libffi.call/many_win32.c: Likewise.
2294 * testsuite/libffi.call/nested_struct.c: Likewise.
2295 * testsuite/libffi.call/nested_struct1.c: Likewise.
2296 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2297 * testsuite/libffi.call/problem1.c: Likewise.
2298 * testsuite/libffi.call/promotion.c: Likewise.
2299 * testsuite/libffi.call/return_ll.c: Likewise.
2300 * testsuite/libffi.call/return_sc.c: Likewise.
2301 * testsuite/libffi.call/return_uc.c: Likewise.
2302 * testsuite/libffi.call/strlen.c: Likewise.
2303 * testsuite/libffi.call/strlen_win32.c: Likewise.
2304 * testsuite/libffi.call/struct1.c: Likewise.
2305 * testsuite/libffi.call/struct2.c: Likewise.
2306 * testsuite/libffi.call/struct3.c: Likewise.
2307 * testsuite/libffi.call/struct4.c: Likewise.
2308 * testsuite/libffi.call/struct5.c: Likewise.
2309 * testsuite/libffi.call/struct6.c: Likewise.
2310 * testsuite/libffi.call/struct7.c: Likewise.
2311 * testsuite/libffi.call/struct8.c: Likewise.
c930d8a5 2312 * testsuite/libffi.call/struct9.c: Likewise.
bbf3057b
AT		 2313 * testsuite/libffi.special/special.exp: New file.
2314 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2315 * testsuite/libffi.special/unwindtest.cc: Likewise.
2316
c930d8a5 2317
996b4f8f
KK		 23182003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2319
2320 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2321 copyright years.
2322
b00badcd
AM		 23232003-08-02 Alan Modra <amodra@bigpond.net.au>
2324
2325 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2326 structure passing.
2327 (ffi_closure_helper_LINUX64): Likewise.
2328 * src/powerpc/linux64.S: Remove code writing to parm save area.
2329 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2330 address in lr from ffi_closure_helper_LINUX64 call to calculate
2331 table address. Optimize function tail.
2332
c1516eae
AT		 23332003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2334
2335 * src/sparc/ffi.c: Handle all floating point registers.
2336 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2337
fc0e6222
GP		 23382003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2339
2340 * README: Note that libffi is not part of GCC. Update the project
2341 URL and status.
2342
a1569df8
FS		 23432003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2344
2345 * src/powerpc/ppc_closure.S: Include ffi.h.
2346
12e816bf
RO		 23472003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2348
2349 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2350 Use C style comments.
2351
e860ed6d
KK		 23522003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2353
2354 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2355 * Makefile.in: Regenerate.
2356 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2357 * configure: Regenerate.
2358 * include/ffi.h.in: Add SHmedia support.
2359 * src/sh64/ffi.c: New file.
2360 * src/sh64/sysv.S: New file.
2361
a733b15e
JJ		 23622003-05-16 Jakub Jelinek <jakub@redhat.com>
2363
2364 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2365 should be read-only.
2366 * configure: Rebuilt.
2367 * fficonfig.h.in: Rebuilt.
2368 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2369 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2370 * src/powerpc/linux64.S: Likewise.
2371 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2372 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2373 if -fpic/-fPIC/-mrelocatable.
2374 * src/powerpc/powerpc_closure.S: Likewise.
2375 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2376 #write in .eh_frame flags.
2377 * src/sparc/v9.S: Likewise.
2378 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2379 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2380 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2381
bd19c9f4
JS		 23822003-05-07 Jeff Sturm <jsturm@one-point.com>
2383
2384 Fixes PR bootstrap/10656
2385 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2386 support for .register pseudo-op.
2387 * src/sparc/v8.S: Use it.
2388 * fficonfig.h.in: Rebuilt.
2389 * configure: Rebuilt.
2390
e9b84181
JJ		 23912003-04-18 Jakub Jelinek <jakub@redhat.com>
2392
2393 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2394 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2395 Make it the default on POWERPC64.
2396 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2397 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2398 * configure: Rebuilt.
2399 * src/powerpc/ffi.c (hidden): Define.
2400 (ffi_prep_args_SYSV): Renamed from
2401 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2402 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2403 ASM_NEEDS_REGISTERS64): New.
2404 (ffi_prep_args64): New function.
2405 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2406 (ffi_call): Likewise.
2407 (ffi_prep_closure): Likewise.
2408 (flush_icache): Surround by #ifndef POWERPC64.
2409 (ffi_dblfl): New union type.
2410 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2411 (ffi_closure_helper_LINUX64): New function.
2412 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2413 __powerpc64__.
2414 * src/powerpc/sysv.S: Likewise.
2415 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2416 * src/powerpc/linux64.S: New file.
2417 * src/powerpc/linux64_closure.S: New file.
2418 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2419 src/powerpc/linux64_closure.S.
2420 (TARGET_SRC_POWERPC): Likewise.
2421
2422 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2423 closure_test_fn3): Fix result printing on big-endian 64-bit
2424 machines.
2425 (main): Print tst2_arg instead of uninitialized tst2_result.
2426
2427 * src/ffitest.c (main): Hide what closure pointer really points to
2428 from the compiler.
2429
908a31e1
RE		 24302003-04-16 Richard Earnshaw <rearnsha@arm.com>
2431
2432 * configure.in (arm-*-netbsdelf*): Add configuration.
2433 (configure): Regenerated.
2434
41a73c4c
LR		 24352003-04-04 Loren J. Rittle <ljrittle@acm.org>
2436
2437 * include/Makefile.in: Regenerate.
2438
cd41c847
ZD		 24392003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2440
2441 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2442 bit mode.
2443 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2444 Receive closure pointer through parameter, read args using
2445 __builtin_dwarf_cfa.
2446 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2447
6db450f9
AS		 24482003-03-12 Andreas Schwab <schwab@suse.de>
2449
2450 * configure.in: Avoid trailing /. in toolexeclibdir.
2451 * configure: Rebuilt.
2452
70144e9c
AT		 24532003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2454
c6900272 2455 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
70144e9c 2456
79e43f89
AT		 24572003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2458
2459 * libffi/src/powerpc/darwin_closure.S:
2460 Fix alignement bug, allocate 8 bytes for the result.
c930d8a5 2461 * libffi/src/powerpc/aix_closure.S:
79e43f89
AT		 2462 Likewise.
2463 * libffi/src/powerpc/ffi_darwin.c:
2464 Update stackframe description for aix/darwin_closure.S.
c930d8a5 2465
0fcc9e55
JJ		 24662003-02-06 Jakub Jelinek <jakub@redhat.com>
2467
2468 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2469 attribute.
2470
63b87c2e
CC		 24712003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2472 Andreas Schwab <schwab@suse.de>
2473
c6900272 2474 * configure.in: Adjust command to source config-ml.in to account
63b87c2e
CC		 2475 for changes to the libffi_basedir definition.
2476 (libffi_basedir): Remove ${srcdir} from value and include trailing
2477 slash if nonempty.
c930d8a5 2478
c6900272 2479 * configure: Regenerate.
63b87c2e 2480
c33402e2
FS		 24812003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2482
2483 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2484
fd2e49b2
AH		 24852003-01-28 Andrew Haley <aph@redhat.com>
2486
2487 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2488 * src/x86/ffi64.c (ffi_prep_closure): New.
2489 (ffi_closure_UNIX64_inner): New.
2490 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2491
0e292c01
AO		 24922003-01-27 Alexandre Oliva <aoliva@redhat.com>
2493
2494 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2495 Remove USE_LIBDIR conditional.
2496 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2497 * Makefile.in, configure: Rebuilt.
2498
acabc220
DE		 24992003-01027 David Edelsohn <edelsohn@gnu.org>
2500
2501 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2502 * Makefile.in: Regenerate.
2503
be0acd42
AH		 25042003-01-22 Andrew Haley <aph@redhat.com>
2505
2506 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2507 unwind info.
2508
ed695c8e
AT		 25092003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2510
2511 * src/powerpc/darwin.S: Add unwind info.
2512 * src/powerpc/darwin_closure.S: Likewise.
2513
45ba15f4
AH		 25142003-01-14 Andrew Haley <aph@redhat.com>
2515
2516 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2517 (ffi_prep_cif_machdep): Likewise.
2518 * src/x86/unix64.S: Add unwind info.
2519
4b5f402e
AJ		 25202003-01-14 Andreas Jaeger <aj@suse.de>
2521
2522 * src/ffitest.c (main): Only use ffi_closures if those are
2523 supported.
2524
7b5102af
TT		 25252003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2526
2527 * libffi/src/ffitest.c
c6900272 2528 add closure testcases
7b5102af
TT		 2529
25302003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2531
2532 * libffi/src/powerpc/ffi.c
c6900272 2533 fix alignment bug for float (4 byte aligned iso 8 byte)
7b5102af 2534
5f7c78d8
GK		 25352003-01-09 Geoffrey Keating <geoffk@apple.com>
2536
2537 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2538 * src/powerpc/darwin.S: Remove RCS version string.
2539
c75c7793
JS		 25402003-01-03 Jeff Sturm <jsturm@one-point.com>
2541
2542 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2543 * src/ffitest.c (main): Use static storage for closure.
2544 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2545 * src/sparc/v8.S (ffi_closure_v8): New.
2546 * src/sparc/v9.S (ffi_closure_v9): New.
2547
eb3c46a1
RM		 25482002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2549
2550 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2551 enumeration for X86_WIN32.
2552 * src/x86/win32.S: Added ffi_call_STDCALL function
2553 definition.
2554 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2555 switch cases for recognising FFI_STDCALL and
2556 calling ffi_call_STDCALL if target is X86_WIN32.
2557 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2558 stdcall versions of the "my_strlen" and "many"
2559 test functions (for X86_WIN32).
2560 Added test cases to test stdcall invocation using
2561 these functions.
2562
4977bab6
ZW		 25632002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2564
2565 * src/sh/sysv.S: Add DWARF2 unwind info.
2566
67333e84
UW		 25672002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2568
2569 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2570
bc3cf760
JW		 25712002-11-26 Jim Wilson <wilson@redhat.com>
2572
2573 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2574
4977bab6
ZW		 25752002-11-23 H.J. Lu <hjl@gnu.org>
2576
2577 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2578 Include ../config/accross.m4.
2579 * aclocal.m4; Rebuild.
2580 * configure: Likewise.
2581
18789f4e
UW		 25822002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2583
2584 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2585
6bee5372
DD		 25862002-11-11 DJ Delorie <dj@redhat.com>
2587
2588 * configure.in: Look for common files in the right place.
2589
8177895b
UW		 25902002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2591
2592 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2593 raw data as _Jv_word values, not ffi_raw.
2594 (ffi_java_ptrarray_to_raw): Likewise.
2595 (ffi_java_rvalue_to_raw): New function.
2596 (ffi_java_raw_call): Call it.
2597 (ffi_java_raw_to_rvalue): New function.
2598 (ffi_java_translate_args): Call it.
2599 * src/ffitest.c (closure_test_fn): Interpret return value
2600 as ffi_arg, not int.
2601 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2602 FFI_TYPE_POINTER case.
2603 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2604 values extended to word size.
2605
af50f4c8
AJ		 26062002-10-02 Andreas Jaeger <aj@suse.de>
2607
2608 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2609
5d297584
BT		 26102002-10-01 Bo Thorsen <bo@smetana.suse.de>
2611
2612 * include/ffi.h.in: Fix i386 win32 compilation.
2613
c4f17c6f
UW		 26142002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2615
2616 * configure.in: Add s390x-*-linux-* target.
2617 * configure: Regenerate.
2618 * include/ffi.h.in: Define S390X for s390x targets.
2619 (FFI_CLOSURES): Define for s390/s390x.
2620 (FFI_TRAMPOLINE_SIZE): Likewise.
2621 (FFI_NATIVE_RAW_API): Likewise.
2622 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2623 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
4977bab6 2624 * src/s390/ffi.c: Major rework of existing code. Add support for
c4f17c6f
UW		 2625 s390x targets. Add closure support.
2626 * src/s390/sysv.S: Likewise.
2627
d7b52625
RE		 26282002-09-29 Richard Earnshaw <rearnsha@arm.com>
2629
2630 * src/arm/sysv.S: Fix typo.
2631
28165e22
RE		 26322002-09-28 Richard Earnshaw <rearnsha@arm.com>
2633
2634 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2635 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2636 (ffi_call_SYSV): Handle soft-float.
2637
0b0f4163
BT		 26382002-09-27 Bo Thorsen <bo@suse.de>
2639
2640 * include/ffi.h.in: Fix multilib x86-64 support.
2641
6c5d742e
KG		 26422002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2643
2644 * Makefile.am (all-multi): Fix multilib parallel build.
2645
f4b2b0d6
KK		 26462002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2647
2648 * configure.in (sh[34]*-*-linux*): Add brackets.
2649 * configure: Regenerate.
2650
ddebdfeb
KK		 26512002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2652
2653 * Makefile.am: Add SH support.
2654 * Makefile.in: Regenerate.
2655 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2656 * configure: Regenerate.
2657 * include/ffi.h.in: Add SH support.
2658 * src/sh/ffi.c: New file.
2659 * src/sh/sysv.S: New file.
2660 * src/types.c: Add SH support.
2661
c94974ca
BT		 26622002-07-16 Bo Thorsen <bo@suse.de>
2663
2664 * src/x86/ffi64.c: New file that adds x86-64 support.
2665 * src/x86/unix64.S: New file that handles argument setup for
2666 x86-64.
2667 * src/x86/sysv.S: Don't use this on x86-64.
2668 * src/x86/ffi.c: Don't use this on x86-64.
2669 Remove unused vars.
2670 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2671 for x86-64.
2672 * src/ffitest.c (struct6): New test that tests a special case in
2673 the x86-64 ABI.
2674 (struct7): Likewise.
2675 (struct8): Likewise.
2676 (struct9): Likewise.
2677 (closure_test_fn): Silence warning about this when it's not used.
2678 (main): Add the new tests.
2679 (main): Fix a couple of wrong casts and silence some compiler warnings.
2680 * include/ffi.h.in: Add x86-64 ABI definition.
2681 * fficonfig.h.in: Regenerate.
2682 * Makefile.am: Add x86-64 support.
2683 * configure.in: Likewise.
2684 * Makefile.in: Regenerate.
2685 * configure: Likewise.
2686
26872002-06-24 Bo Thorsen <bo@suse.de>
2688
2689 * src/types.c: Merge settings for similar architectures.
2690 Add x86-64 sizes and alignments.
2691
26922002-06-23 Bo Thorsen <bo@suse.de>
2693
2694 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
2695 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
2696 * src/mips/ffi.c (ffi_prep_args): Likewise.
2697 * src/m68k/ffi.c (ffi_prep_args): Likewise.
2698
5be36a8f
L		 26992002-07-18 H.J. Lu (hjl@gnu.org)
2700
2701 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
2702 (libffi_la_SOURCES): Support MIPS_LINUX.
2703 (libffi_convenience_la_SOURCES): Likewise.
2704 * Makefile.in: Regenerated.
2705
2706 * configure.in (mips64*-*): Skip.
2707 (mips*-*-linux*): New.
2708 * configure: Regenerated.
2709
2710 * src/mips/ffi.c: Include <sgidefs.h>.
2711
ad5e163c
UW		 27122002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
2713
2714 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
2715
ee4586c5
RS		 27162002-05-27 Roger Sayle <roger@eyesopen.com>
2717
2718 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
2719
b86aa7b0
BT		 27202002-05-27 Bo Thorsen <bo@suse.de>
2721
2722 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
2723 fix formatting.
2724
859f7aaf
AT		 27252002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
2726
2727 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
2728 beginning of function (for older apple cc).
2729
6706f116
AO		 27302002-05-08 Alexandre Oliva <aoliva@redhat.com>
2731
2732 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
2733 script entry, and set LD to it when configuring multilibs.
2734 * configure: Rebuilt.
2735
fad655c8
JT		 27362002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
2737
2738 * configure.in (sparc64-*-netbsd*): Add target.
2739 (sparc-*-netbsdelf*): Likewise.
2740 * configure: Regenerate.
2741
189ab118
DM		 27422002-04-28 David S. Miller <davem@redhat.com>
2743
2744 * configure.in, configure: Fix SPARC test in previous change.
2745
22bcf65c
GT		 27462002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
2747
2748 * Makefile.am: Add Linux for S/390 support.
2749 * Makefile.in: Regenerate.
2750 * configure.in: Add Linux for S/390 support.
2751 * configure: Regenerate.
2752 * include/ffi.h.in: Add Linux for S/390 support.
2753 * src/s390/ffi.c: New file from libffi CVS tree.
2754 * src/s390/sysv.S: New file from libffi CVS tree.
2755
5d84cf0b
JJ		 27562002-04-28 Jakub Jelinek <jakub@redhat.com>
2757
2758 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
2759 %r_disp32().
2760 * src/sparc/v8.S: Use it.
2761 * src/sparc/v9.S: Likewise.
2762 * fficonfig.h.in: Rebuilt.
2763 * configure: Rebuilt.
2764
cc382e5c
HB		 27652002-04-08 Hans Boehm <Hans_Boehm@hp.com>
2766
2767 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
2768 correctly.
2769 * src/ia64/unix.S: Add unwind information. Fix comments.
2770 Save sp in a way that's compatible with unwind info.
2771 (ffi_call_unix): Correctly restore sp in all cases.
2772 * src/ia64/ffi.c: Add, fix comments.
2773
dc798645
JJ		 27742002-04-08 Jakub Jelinek <jakub@redhat.com>
2775
2776 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
2777
da00eaab
JT		 27782002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
2779
2780 * configure.in (alpha*-*-netbsd*): Add target.
2781 * configure: Regenerate.
2782
0c0b87ad
JS		 27832002-04-04 Jeff Sturm <jsturm@one-point.com>
2784
2785 * src/sparc/v8.S: Add unwind info.
2786 * src/sparc/v9.S: Likewise.
2787
320d7e7a
KW		 27882002-03-30 Krister Walfridsson <cato@df.lth.se>
2789
2790 * configure.in: Enable i*86-*-netbsdelf*.
2791 * configure: Rebuilt.
4977bab6 2792
c75e51ed
DB		 27932002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
2794
2795 PR other/2620
2796 * src/mips/n32.s: Delete
2797 * src/mips/o32.s: Delete
2798
5bb366e0
LR		 27992002-03-21 Loren J. Rittle <ljrittle@acm.org>
2800
2801 * configure.in: Enable alpha*-*-freebsd*.
2802 * configure: Rebuilt.
2803
ce3b24a7
BM		 28042002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2805
2806 * Makefile.am: libfficonvenience -> libffi_convenience.
2807 * Makefile.in: Rebuilt.
4977bab6 2808
1eb7ba30
BM		 2809 * Makefile.am: Define ffitest_OBJECTS.
2810 * Makefile.in: Rebuilt.
ce3b24a7 2811
073ae293
AT		 28122002-03-07 Andreas Tobler <toa@pop.agri.ch>
2813 David Edelsohn <edelsohn@gnu.org>
2814
2815 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
2816 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
2817 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
2818 * Makefile.in: Regenerate.
2819 * include/ffi.h.in: Add AIX and Darwin closure definitions.
2820 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
2821 (flush_icache, flush_range): New functions.
2822 (ffi_closure_helper_DARWIN): New function.
2823 * src/powerpc/aix_closure.S: New file.
2824 * src/powerpc/darwin_closure.S: New file.
4977bab6 2825
02c521f4
JS		 28262002-02-24 Jeff Sturm <jsturm@one-point.com>
2827
2828 * include/ffi.h.in: Add typedef for ffi_arg.
2829 * src/ffitest.c (main): Declare rint with ffi_arg.
4977bab6 2830
287dd527
AT		 28312002-02-21 Andreas Tobler <toa@pop.agri.ch>
2832
2833 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
2834 number of GPRs for floating-point arguments.
2835
e2f2e70a
AG		 28362002-01-31 Anthony Green <green@redhat.com>
2837
2838 * configure: Rebuilt.
2839 * configure.in: Replace CHECK_SIZEOF and endian tests with
2840 cross-compiler friendly macros.
2841 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
2842 macros.
2843
15b9b32f
DE		 28442002-01-18 David Edelsohn <edelsohn@gnu.org>
2845
2846 * src/powerpc/darwin.S (_ffi_call_AIX): New.
1f7d5413 2847 * src/powerpc/aix.S (ffi_call_DARWIN): New.
15b9b32f 2848
2f4006f8
DE		 28492002-01-17 David Edelsohn <edelsohn@gnu.org>
2850
2851 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
2852 (TARGET_SRC_POWERPC_AIX): New.
2853 (POWERPC_AIX): New stanza.
2854 * Makefile.in: Regenerate.
2855 * configure.in: Add AIX case.
2856 * configure: Regenerate.
2857 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
2858 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
2859 size. Fix "long double" support.
2860 (ffi_call): Add FFI_AIX case.
2861 * src/powerpc/aix.S: New.
2862
3c017635
JH		 28632001-10-09 John Hornkvist <john@toastedmarshmallow.com>
2864
2865 Implement Darwin PowerPC ABI.
2866 * configure.in: Handle powerpc-*-darwin*.
2867 * Makefile.am: Set source files for POWERPC_DARWIN.
2868 * configure: Rebuilt.
2869 * Makefile.in: Rebuilt.
2870 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
2871 POWERPC_DARWIN.
2872 * src/powerpc/darwin.S: New file.
2873 * src/powerpc/ffi_darwin.c: New file.
2874
cc712abf
JM		 28752001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
2876
2877 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
2878
03a8de49
RO		 28792001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2880
2881 * src/x86/sysv.S: Avoid gas-only .balign directive.
2882 Use C style comments.
2883
92a0e6c6
RO		 28842001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2885
2886 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
2887 Fixes PR bootstrap/3563.
2888
57482f58
RO		 28892001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2890
2891 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
2892
d7e2dd2f
RO		 28932001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2894
2895 * configure.in: Recognize sparc*-sun-* host.
2896 * configure: Regenerate.
4977bab6 2897
be72207c
AH		 28982001-06-06 Andrew Haley <aph@redhat.com>
2899
4075db8f 2900 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
be72207c 2901
552cb45a
AH		 29022001-06-03 Andrew Haley <aph@redhat.com>
2903
4075db8f
BT		 2904 * src/alpha/osf.S: Add unwind info.
2905 * src/powerpc/sysv.S: Add unwind info.
2906 * src/powerpc/ppc_closure.S: Likewise.
552cb45a 2907
7d45b968
JS		 29082000-05-31 Jeff Sturm <jsturm@one-point.com>
2909
2910 * configure.in: Fix AC_ARG_ENABLE usage.
2911 * configure: Rebuilt.
2912
2a7f63c4
BM		 29132001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2914
2915 * configure.in: Remove warning about beta code.
2916 * configure: Rebuilt.
2917
29182001-04-25 Hans Boehm <Hans_Boehm@hp.com>
2919
03a55252
HB		 2920 * src/ia64/unix.S: Restore stack pointer when returning from
2921 ffi_closure_UNIX.
2922 * src/ia64/ffi.c: Fix typo in comment.
2923
df5da1c4
JW		 29242001-04-18 Jim Wilson <wilson@redhat.com>
2925
2926 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
2927 to eliminate RAW DV.
2928
52c28977
BM		 29292001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
2930
2931 * Makefile.am: Make a libtool convenience library.
2932 * Makefile.in: Rebuilt.
2933
3b905056
BM		 29342001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
2935
2936 * configure.in: Use different syntax for subdirectory creation.
2937 * configure: Rebuilt.
2938
d6c4152a
JB		 29392001-03-27 Jon Beniston <jon@beniston.com>
2940
2941 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
2942 * configure: Rebuilt.
2943 * Makefile.am: Added X86_WIN32 target support.
2944 * Makefile.in: Rebuilt.
2945
2946 * include/ffi.h.in: Added X86_WIN32 target support.
2947
2948 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
2949 * src/types.c: Added X86_WIN32 target support.
2950
2951 * src/x86/win32.S: New file. Based on sysv.S, but with EH
2952 stuff removed and made to work with CygWin's gas.
2953
675cdab3
BM		 29542001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
2955
2956 * configure.in: Make target subdirectory in build dir.
2957 * Makefile.am: Override suffix based rules to specify correct output
2958 subdirectory.
2959 * Makefile.in: Rebuilt.
2960 * configure: Rebuilt.
2961
e5f8858a 29622001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
cc4c8975
KH		 2963
2964 * src/powerpc/ppc_closure.S: New file.
2965 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
2966 involving long long and register pairs.
2967 (ffi_prep_closure): New function.
2968 (flush_icache): Likewise.
2969 (ffi_closure_helper_SYSV): Likewise.
2970 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
2971 (FFI_TRAMPOLINE_SIZE): Likewise.
2972 (FFI_NATIVE_RAW_API): Likewise.
2973 * Makefile.in: Rebuilt.
2974 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
2975 (TARGET_SRC_POWERPC): Likewise.
2976
9450a929
TT		 29772001-03-19 Tom Tromey <tromey@redhat.com>
2978
2979 * Makefile.in: Rebuilt.
2980 * Makefile.am (ffitest_LDFLAGS): New macro.
2981
4d2cf715
NC		 29822001-03-02 Nick Clifton <nickc@redhat.com>
2983
2984 * include/ffi.h.in: Remove RCS ident string.
2985 * include/ffi_mips.h: Remove RCS ident string.
2986 * src/debug.c: Remove RCS ident string.
2987 * src/ffitest.c: Remove RCS ident string.
2988 * src/prep_cif.c: Remove RCS ident string.
2989 * src/types.c: Remove RCS ident string.
2990 * src/alpha/ffi.c: Remove RCS ident string.
2991 * src/alpha/osf.S: Remove RCS ident string.
2992 * src/arm/ffi.c: Remove RCS ident string.
2993 * src/arm/sysv.S: Remove RCS ident string.
2994 * src/mips/ffi.c: Remove RCS ident string.
2995 * src/mips/n32.S: Remove RCS ident string.
2996 * src/mips/o32.S: Remove RCS ident string.
2997 * src/sparc/ffi.c: Remove RCS ident string.
2998 * src/sparc/v8.S: Remove RCS ident string.
2999 * src/sparc/v9.S: Remove RCS ident string.
3000 * src/x86/ffi.c: Remove RCS ident string.
3001 * src/x86/sysv.S: Remove RCS ident string.
3002
7e5fd99f
JM		 30032001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3004
3005 * include/ffi.h.in: Change sourceware.cygnus.com references to
3006 gcc.gnu.org.
3007
3f67ba6e
RH		 30082000-12-09 Richard Henderson <rth@redhat.com>
3009
3010 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3011 (ffi_closure_osf_inner): Index rather than increment avalue
3012 and arg_types. Give ffi_closure_osf the raw return value type.
3013 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3014 promotion.
3015
30162000-12-07 Richard Henderson <rth@redhat.com>
29fe0479
RH		 3017
3018 * src/raw_api.c (ffi_translate_args): Fix typo.
3019 (ffi_prep_closure): Likewise.
3020
3021 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3022 FFI_TRAMPOLINE_SIZE.
3023 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3024 cif->bytes for new ffi_call_osf implementation.
3025 (ffi_prep_args): Absorb into ...
3026 (ffi_call): ... here. Do all stack allocation here and
3027 avoid a callback function.
3028 (ffi_prep_closure, ffi_closure_osf_inner): New.
3029 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3030 (ffi_closure_osf): New.
3031
2f7bf5b7
AO		 30322000-09-10 Alexandre Oliva <aoliva@redhat.com>
3033
ed1c1a1a
AO		 3034 * config.guess, config.sub, install-sh: Removed.
3035 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3036 * Makefile.in: Rebuilt.
3037
2f7bf5b7
AO		 3038 * acinclude.m4: Include libtool macros from the top level.
3039 * aclocal.m4, configure: Rebuilt.
3040
e15a68e7
AO		 30412000-08-22 Alexandre Oliva <aoliva@redhat.com>
3042
3043 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3044 * configure: Rebuilt.
3045
7f8c11d5
SB		 30462000-05-11 Scott Bambrough <scottb@netwinder.org>
3047
4977bab6
ZW		 3048 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3049 memory correctly. Use conditional instructions, not branches where
7f8c11d5
SB		 3050 possible.
3051
c5acf51f
TT		 30522000-05-04 Tom Tromey <tromey@cygnus.com>
3053
3054 * configure: Rebuilt.
3055 * configure.in: Match `arm*-*-linux-*'.
3056 From Chris Dornan <cdornan@arm.com>.
3057
91bffbe9
JJ		 30582000-04-28 Jakub Jelinek <jakub@redhat.com>
3059
3060 * Makefile.am (SUBDIRS): Define.
3061 (AM_MAKEFLAGS): Likewise.
3062 (Multilib support.): Add section.
3063 * Makefile.in: Rebuilt.
3064 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
4075db8f
BT		 3065 New variables. Set for gcc using -print-multi-lib. Export them
3066 to libtool.
3067 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3068 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3069 for -shared links.
3070 (extra_compiler_flags_value, extra_compiler_flags): Check these
3071 for extra compiler options which need to be passed down in
3072 compiler_flags.
91bffbe9 3073
bd66587e
AG		 30742000-04-16 Anthony Green <green@redhat.com>
3075
3076 * configure: Rebuilt.
3077 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3078
3791773c
JJ		 30792000-04-14 Jakub Jelinek <jakub@redhat.com>
3080
4075db8f 3081 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3791773c
JJ		 3082 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3083 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3084 Replace all void * sizeofs with sizeof(int).
3085 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3086 different than DOUBLE.
3087 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3088 (ffi_prep_args_v9): New function.
3089 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3090 (ffi_V9_return_struct): New function.
3091 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3092 32bit code (not yet cross-arch calls).
3093 * src/sparc/v8.S: Add struct return delay nop.
3094 Handle long long.
3095 * src/sparc/v9.S: New file.
3096 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3097 is used on sparc64 only for structures larger than 32 bytes.
3098 Pass by reference for structures is done for structure arguments
3099 larger than 16 bytes.
3100 * src/ffitest.c (main): Use 64bit rint on sparc64.
3101 Run long long tests on sparc.
3102 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3103 sparc64.
3104 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3105 on sparc64.
3106 * configure.in (sparc-*-linux*): New supported target.
3107 (sparc64-*-linux*): Likewise.
3108 * configure: Rebuilt.
3109 * Makefile.am: Add v9.S to SPARC files.
3110 * Makefile.in: Likewise.
3111 (LINK): Surround $(CCLD) into double quotes, so that multilib
3112 compiles work correctly.
3113
db459595
APB		 31142000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3115
3116 * configure: Rebuilt.
3117 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3118 proposed by Bryce McKinlay.
3119
c1f63902
TT		 31202000-03-20 Tom Tromey <tromey@cygnus.com>
3121
3122 * Makefile.in: Hand edit for java_raw_api.lo.
3123
0f72dc9e
BM		 31242000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3125
3126 * config.guess, config.sub: Update from the gcc tree.
3127 Fix for PR libgcj/168.
3128
3c18f2d1
TT		 31292000-03-03 Tom Tromey <tromey@cygnus.com>
3130
323eee8a
TT		 3131 * Makefile.in: Fixed ia64 by hand.
3132
3c18f2d1
TT		 3133 * configure: Rebuilt.
3134 * configure.in (--enable-multilib): New option.
3135 (libffi_basedir): New subst.
3136 (AC_OUTPUT): Added multilib code.
3137
a2b10a92
TT		 31382000-03-02 Tom Tromey <tromey@cygnus.com>
3139
3140 * Makefile.in: Rebuilt.
3141 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3142 directory name.
3143
dc5de370
HB		 31442000-02-25 Hans Boehm <boehm@acm.org>
3145
3146 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3147 files.
3148 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3149 list.
3150 (ffi_prep_raw_closure): Use ffi_translate_args, not
3151 ffi_closure_translate.
3152 * src/java_raw_api.c: New file.
3153 * src/ffitest.c (closure_test_fn): New function.
3154 (main): Define `rint' as long long on IA64. Added new test when
3155 FFI_CLOSURES is defined.
3156 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3157 (ffi_abi): Recognize IA64.
3158 (ffi_raw): Added `flt' field.
3159 Added "Java raw API" code.
3160 * configure.in: Recognize ia64.
3161 * Makefile.am (TARGET_SRC_IA64): New macro.
3162 (libffi_la_common_SOURCES): Added java_raw_api.c.
3163 (libffi_la_SOURCES): Define in IA64 case.
3164
d79229bd
TT		 31652000-01-04 Tom Tromey <tromey@cygnus.com>
3166
3167 * Makefile.in: Rebuilt with newer automake.
3168
2c33b220
TT		 31691999-12-31 Tom Tromey <tromey@cygnus.com>
3170
3171 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3172
6ba9e3b0
TT		 31731999-09-01 Tom Tromey <tromey@cygnus.com>
3174
3175 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3176 undefs.
3177 * fficonfig.h.in: Rebuilt.
3178 * configure: Rebuilt.
3179 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3180 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3181 * acconfig.h: Don't #undef PACKAGE or VERSION.
3182
63e5e3e0
AG		 31831999-08-09 Anthony Green <green@cygnus.com>
3184
3185 * include/ffi.h.in: Try to work around messy header problem
3186 with PACKAGE and VERSION.
3187
3188 * configure: Rebuilt.
3189 * configure.in: Change version to 2.00-beta.
3190
4075db8f 3191 * fficonfig.h.in: Rebuilt.
63e5e3e0
AG		 3192 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3193
3194 * src/x86/ffi.c (ffi_raw_call): Rename.
3195
31961999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3197
3198 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3199 (ffi_prep_incoming_args_SYSV): Ditto.
3200 (ffi_prep_closure): Ditto.
3201 (ffi_closure_raw_SYSV): Ditto.
3202 (ffi_prep_raw_closure): More ditto.
3203 (ffi_call_raw): Final ditto.
3204
3205 * include/ffi.h.in: Add definitions for closure and raw API.
3206
3207 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4977bab6 3208 FFI_TYPE_UINT64.
63e5e3e0
AG		 3209
3210 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3211
3212 * src/raw_api.c: New file.
3213
3214 * include/ffi.h.in (ffi_raw): New type.
3215 (UINT_ARG, SINT_ARG): New defines.
3216 (ffi_closure, ffi_raw_closure): New types.
3217 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4977bab6 3218
63e5e3e0
AG		 3219 * configure.in: Add check for endianness and sizeof void*.
3220
3221 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3222 instead of directly.
3223
3224 * configure: Rebuilt.
3225
3226Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3227
3228 * configure.in: Add x86 and powerpc BeOS configurations.
3229 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3230
32311999-05-09 Anthony Green <green@cygnus.com>
3232
3233 * configure.in: Add warning about this being beta code.
3234 Remove src/Makefile.am from the picture.
3235 * configure: Rebuilt.
3236
3237 * Makefile.am: Move logic from src/Makefile.am. Add changes
3238 to support libffi as a target library.
3239 * Makefile.in: Rebuilt.
3240
3241 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3242 Upgraded to new autoconf, automake, libtool.
3243
3244 * README: Tweaks.
3245
3246 * LICENSE: Update copyright date.
3247
3248 * src/Makefile.am, src/Makefile.in: Removed.
3249
32501998-11-29 Anthony Green <green@cygnus.com>
3251
3252 * include/ChangeLog: Removed.
3253 * src/ChangeLog: Removed.
3254 * src/mips/ChangeLog: Removed.
3255 * src/sparc/ChangeLog: Remboved.
3256 * src/x86/ChangeLog: Removed.
3257
3258 * ChangeLog.v1: Created.
Mirror of https://gcc.gnu.org/git/gcc.git