1 2007-09-04 <aph@redhat.com>
3 * src/arm/sysv.S (UNWIND): New.
4 (Whole file): Conditionally compile unwinder directives.
5 * src/arm/sysv.S: Add unwinder directives.
7 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
8 Only treat r0 as a struct address if we're actually returning a
10 Only copy the bytes that are actually within a struct.
11 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
12 is returned in r0, not passed by address.
13 (ffi_call): Allocate a word-sized temporary for the case where
14 a composite is returned in r0.
15 (ffi_prep_incoming_args_SYSV): Align as necessary.
17 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
19 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
20 directly using the sys_cacheflush syscall.
22 2007-07-27 Andrew Haley <aph@redhat.com>
24 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
26 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
28 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
29 * configure.ac: Likewise.
30 * Makefile.in: Regenerate.
31 * include/Makefile.in: Likewise.
32 * testsuite/Makefile.in: Likewise.
33 * configure: Likewise.
35 2007-08-24 David Daney <ddaney@avtrex.com>
37 * testsuite/libffi.call/return_sl.c: New test.
39 2007-08-10 David Daney <ddaney@avtrex.com>
41 * testsuite/libffi.call/cls_multi_ushort.c,
42 testsuite/libffi.call/cls_align_uint16.c,
43 testsuite/libffi.call/nested_struct1.c,
44 testsuite/libffi.call/nested_struct3.c,
45 testsuite/libffi.call/cls_7_1_byte.c,
46 testsuite/libffi.call/nested_struct5.c,
47 testsuite/libffi.call/cls_double.c,
48 testsuite/libffi.call/nested_struct7.c,
49 testsuite/libffi.call/cls_sint.c,
50 testsuite/libffi.call/nested_struct9.c,
51 testsuite/libffi.call/cls_20byte1.c,
52 testsuite/libffi.call/cls_multi_sshortchar.c,
53 testsuite/libffi.call/cls_align_sint64.c,
54 testsuite/libffi.call/cls_3byte2.c,
55 testsuite/libffi.call/cls_multi_schar.c,
56 testsuite/libffi.call/cls_multi_uchar.c,
57 testsuite/libffi.call/cls_19byte.c,
58 testsuite/libffi.call/cls_9byte1.c,
59 testsuite/libffi.call/cls_align_float.c,
60 testsuite/libffi.call/closure_fn1.c,
61 testsuite/libffi.call/problem1.c,
62 testsuite/libffi.call/closure_fn3.c,
63 testsuite/libffi.call/cls_sshort.c,
64 testsuite/libffi.call/closure_fn5.c,
65 testsuite/libffi.call/cls_align_double.c,
66 testsuite/libffi.call/nested_struct.c,
67 testsuite/libffi.call/cls_2byte.c,
68 testsuite/libffi.call/nested_struct10.c,
69 testsuite/libffi.call/cls_4byte.c,
70 testsuite/libffi.call/cls_6byte.c,
71 testsuite/libffi.call/cls_8byte.c,
72 testsuite/libffi.call/cls_multi_sshort.c,
73 testsuite/libffi.call/cls_align_sint16.c,
74 testsuite/libffi.call/cls_align_uint32.c,
75 testsuite/libffi.call/cls_20byte.c,
76 testsuite/libffi.call/cls_float.c,
77 testsuite/libffi.call/nested_struct2.c,
78 testsuite/libffi.call/cls_5_1_byte.c,
79 testsuite/libffi.call/nested_struct4.c,
80 testsuite/libffi.call/cls_24byte.c,
81 testsuite/libffi.call/nested_struct6.c,
82 testsuite/libffi.call/cls_64byte.c,
83 testsuite/libffi.call/nested_struct8.c,
84 testsuite/libffi.call/cls_uint.c,
85 testsuite/libffi.call/cls_multi_ushortchar.c,
86 testsuite/libffi.call/cls_schar.c,
87 testsuite/libffi.call/cls_uchar.c,
88 testsuite/libffi.call/cls_align_uint64.c,
89 testsuite/libffi.call/cls_ulonglong.c,
90 testsuite/libffi.call/cls_align_longdouble.c,
91 testsuite/libffi.call/cls_1_1byte.c,
92 testsuite/libffi.call/cls_12byte.c,
93 testsuite/libffi.call/cls_3_1byte.c,
94 testsuite/libffi.call/cls_3byte1.c,
95 testsuite/libffi.call/cls_4_1byte.c,
96 testsuite/libffi.call/cls_6_1_byte.c,
97 testsuite/libffi.call/cls_16byte.c,
98 testsuite/libffi.call/cls_18byte.c,
99 testsuite/libffi.call/closure_fn0.c,
100 testsuite/libffi.call/cls_9byte2.c,
101 testsuite/libffi.call/closure_fn2.c,
102 testsuite/libffi.call/closure_fn4.c,
103 testsuite/libffi.call/cls_ushort.c,
104 testsuite/libffi.call/closure_fn6.c,
105 testsuite/libffi.call/cls_5byte.c,
106 testsuite/libffi.call/cls_align_pointer.c,
107 testsuite/libffi.call/cls_7byte.c,
108 testsuite/libffi.call/cls_align_sint32.c,
109 testsuite/libffi.special/unwindtest_ffi_call.cc,
110 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
112 2007-08-10 David Daney <ddaney@avtrex.com>
115 * configure.ac: Don't treat mips64 as a special case.
116 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
117 * configure: Regenerate
118 * Makefile.in: Ditto.
119 * fficonfig.h.in: Ditto.
120 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
121 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
122 (FFI_DEFAULT_ABI): Set for n64 case.
123 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
124 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
125 (ffi_closure_N32): New function.
126 (.eh_frame): New section
127 * src/mips/o32.S: Clean up comments.
128 (ffi_closure_O32): Pass ffi_closure parameter in $12.
129 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
130 _MIPS_SIM == _ABIN32 throughout.
131 (FFI_MIPS_STOP_HERE): New, use in place of
133 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
134 to support n32/n64 ABIs.
135 (calc_n32_struct_flags): Rewrite.
136 (calc_n32_return_struct_flags): Remove unused variable. Reverse
137 position of flag bits.
138 (ffi_prep_cif_machdep): Rewrite n32 portion.
139 (ffi_call): Enable for n64. Add special handling for small structure
141 (ffi_prep_closure_loc): Add n32 and n64 support.
142 (ffi_closure_mips_inner_O32): Add cast to silence warning.
143 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
145 2007-08-08 David Daney <ddaney@avtrex.com>
147 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
148 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
150 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
151 * testsuite/libffi.call/cls_sint.c (main): Ditto.
152 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
153 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
154 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
155 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
156 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
157 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
158 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
159 * testsuite/libffi.call/return_fl3.c (main): Ditto.
160 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
161 * testsuite/libffi.call/nested_struct.c (main): Ditto.
162 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
163 * testsuite/libffi.call/return_ll1.c (main): Ditto.
164 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
165 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
166 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
167 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
168 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
169 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
170 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
171 * testsuite/libffi.call/cls_uint.c (main): Ditto.
172 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
173 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
174 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
175 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
176 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
177 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
178 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
179 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
180 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
182 2007-08-07 Andrew Haley <aph@redhat.com>
184 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
187 2007-08-06 Andrew Haley <aph@redhat.com>
190 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
191 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
194 2007-08-02 David Daney <ddaney@avtrex.com>
196 * testsuite/libffi.call/return_ul.c (main): Define return type as
197 ffi_arg. Use proper printf conversion specifier.
199 2007-07-30 Andrew Haley <aph@redhat.com>
202 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
203 signed/unsigned int8/16.
204 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
206 Remove code to pop args from the stack after call.
207 Special-case signed/unsigned int8/16.
208 * testsuite/libffi.call/return_sc.c (main): Revert.
210 2007-07-26 Richard Guenther <rguenther@suse.de>
213 * testsuite/libffi.call/return_sc.c (main): Verify call
214 result as signed char, not ffi_arg.
216 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
218 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
219 * configure: Regenerate.
221 2007-07-11 David Daney <ddaney@avtrex.com>
223 * src/mips/ffi.c: Don't include sys/cachectl.h.
224 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
227 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
229 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
230 from (ffi_prep_closure): ... this.
231 (FFI_INIT_TRAMPOLINE): Adjust.
233 2005-12-31 Phil Blundell <pb@reciva.com>
235 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
236 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
237 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
238 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
239 (FFI_CLOSURES): Enable closure support.
241 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
243 * testsuite/libffi.call/cls_multi_ushort.c,
244 testsuite/libffi.call/cls_align_uint16.c,
245 testsuite/libffi.call/nested_struct1.c,
246 testsuite/libffi.call/nested_struct3.c,
247 testsuite/libffi.call/cls_7_1_byte.c,
248 testsuite/libffi.call/cls_double.c,
249 testsuite/libffi.call/nested_struct5.c,
250 testsuite/libffi.call/nested_struct7.c,
251 testsuite/libffi.call/cls_sint.c,
252 testsuite/libffi.call/nested_struct9.c,
253 testsuite/libffi.call/cls_20byte1.c,
254 testsuite/libffi.call/cls_multi_sshortchar.c,
255 testsuite/libffi.call/cls_align_sint64.c,
256 testsuite/libffi.call/cls_3byte2.c,
257 testsuite/libffi.call/cls_multi_schar.c,
258 testsuite/libffi.call/cls_multi_uchar.c,
259 testsuite/libffi.call/cls_19byte.c,
260 testsuite/libffi.call/cls_9byte1.c,
261 testsuite/libffi.call/cls_align_float.c,
262 testsuite/libffi.call/closure_fn1.c,
263 testsuite/libffi.call/problem1.c,
264 testsuite/libffi.call/closure_fn3.c,
265 testsuite/libffi.call/cls_sshort.c,
266 testsuite/libffi.call/closure_fn5.c,
267 testsuite/libffi.call/cls_align_double.c,
268 testsuite/libffi.call/cls_2byte.c,
269 testsuite/libffi.call/nested_struct.c,
270 testsuite/libffi.call/nested_struct10.c,
271 testsuite/libffi.call/cls_4byte.c,
272 testsuite/libffi.call/cls_6byte.c,
273 testsuite/libffi.call/cls_8byte.c,
274 testsuite/libffi.call/cls_multi_sshort.c,
275 testsuite/libffi.call/cls_align_uint32.c,
276 testsuite/libffi.call/cls_align_sint16.c,
277 testsuite/libffi.call/cls_float.c,
278 testsuite/libffi.call/cls_20byte.c,
279 testsuite/libffi.call/cls_5_1_byte.c,
280 testsuite/libffi.call/nested_struct2.c,
281 testsuite/libffi.call/cls_24byte.c,
282 testsuite/libffi.call/nested_struct4.c,
283 testsuite/libffi.call/nested_struct6.c,
284 testsuite/libffi.call/cls_64byte.c,
285 testsuite/libffi.call/nested_struct8.c,
286 testsuite/libffi.call/cls_uint.c,
287 testsuite/libffi.call/cls_multi_ushortchar.c,
288 testsuite/libffi.call/cls_schar.c,
289 testsuite/libffi.call/cls_uchar.c,
290 testsuite/libffi.call/cls_align_uint64.c,
291 testsuite/libffi.call/cls_ulonglong.c,
292 testsuite/libffi.call/cls_align_longdouble.c,
293 testsuite/libffi.call/cls_1_1byte.c,
294 testsuite/libffi.call/cls_12byte.c,
295 testsuite/libffi.call/cls_3_1byte.c,
296 testsuite/libffi.call/cls_3byte1.c,
297 testsuite/libffi.call/cls_4_1byte.c,
298 testsuite/libffi.call/cls_6_1_byte.c,
299 testsuite/libffi.call/cls_16byte.c,
300 testsuite/libffi.call/cls_18byte.c,
301 testsuite/libffi.call/closure_fn0.c,
302 testsuite/libffi.call/cls_9byte2.c,
303 testsuite/libffi.call/closure_fn2.c,
304 testsuite/libffi.call/closure_fn4.c,
305 testsuite/libffi.call/cls_ushort.c,
306 testsuite/libffi.call/closure_fn6.c,
307 testsuite/libffi.call/cls_5byte.c,
308 testsuite/libffi.call/cls_align_pointer.c,
309 testsuite/libffi.call/cls_7byte.c,
310 testsuite/libffi.call/cls_align_sint32.c,
311 testsuite/libffi.special/unwindtest_ffi_call.cc,
312 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
314 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
316 * aclocal.m4: Regenerated.
318 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
320 * configure: Regenerate.
322 2007-05-23 Steve Ellcey <sje@cup.hp.com>
324 * Makefile.in: Regenerate.
325 * configure: Regenerate.
326 * aclocal.m4: Regenerate.
327 * include/Makefile.in: Regenerate.
328 * testsuite/Makefile.in: Regenerate.
330 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
332 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
333 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
334 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
335 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
336 (FFI_CLOSURES): Enable closure support.
338 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
340 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
341 * configure: Regenerate.
342 * fficonfig.h.in: Regenerate.
343 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
344 CFI_OFFSET,CFI_DEF_CFA): New macros.
345 (ffi_call_SYSV): Add callframe annotation.
347 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
349 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
350 numerous test suite failures.
351 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
353 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
355 * Makefile.am (EXTRA_DIST): Bring up to date.
356 * Makefile.in: Regenerate.
357 * src/frv/eabi.S: Remove RCS keyword.
359 2007-04-06 Richard Henderson <rth@redhat.com>
361 * configure.ac: Tidy target case.
362 (HAVE_LONG_DOUBLE): Allow the target to override.
363 * configure: Regenerate.
364 * include/ffi.h.in: Don't define ffi_type_foo if
365 LIBFFI_HIDE_BASIC_TYPES is defined.
366 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
368 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
369 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
370 (ffi_type_longdouble): Special case for Alpha. Don't define
371 if long double == double.
373 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
374 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
375 (ffi_call, ffi_closure_osf_inner): Likewise.
376 (ffi_closure_osf_inner): Likewise. Mark hidden.
377 (ffi_call_osf, ffi_closure_osf): Mark hidden.
378 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
379 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
380 (load_table): Handle 128-bit long double.
382 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
384 2007-04-06 Tom Tromey <tromey@redhat.com>
387 * README: Fixed bug in example.
389 2007-04-03 Jakub Jelinek <jakub@redhat.com>
391 * src/closures.c: Include sys/statfs.h.
392 (_GNU_SOURCE): Define on Linux.
393 (FFI_MMAP_EXEC_SELINUX): Define.
394 (selinux_enabled): New variable.
395 (selinux_enabled_check): New function.
396 (is_selinux_enabled): Define.
399 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
401 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
402 Use 'volatile float sum' to create sum of floats to avoid false
403 negative due to excess precision on ix86 targets.
406 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
408 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
410 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
412 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
414 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
415 (ffi_prep_closure_loc): New.
416 (ffi_prep_raw_closure_loc): New.
417 (ffi_prep_java_raw_closure_loc): New.
418 * src/closures.c: New file.
419 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
420 Replace sflags with exec_offset.
421 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
422 sub_segment_exec_offset): New macros.
423 (get_segment_flags, set_segment_flags, check_segment_merge): New
425 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
426 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
427 destroy_mspace): Use new macros.
428 (sys_alloc): Silence warning.
429 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
430 * Makefile.in: Rebuilt.
431 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
432 terms of ffi_prep_closure_loc.
433 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
435 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
437 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
439 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
441 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
442 (ffi_prep_closure): ... this.
443 * src/pa/ffi.c: Likewise.
444 * src/cris/ffi.c: Likewise. Adjust.
445 * src/frv/ffi.c: Likewise.
446 * src/ia64/ffi.c: Likewise.
447 * src/mips/ffi.c: Likewise.
448 * src/powerpc/ffi_darwin.c: Likewise.
449 * src/s390/ffi.c: Likewise.
450 * src/sh/ffi.c: Likewise.
451 * src/sh64/ffi.c: Likewise.
452 * src/sparc/ffi.c: Likewise.
453 * src/x86/ffi64.c: Likewise.
454 * src/x86/ffi.c: Likewise.
455 (FFI_INIT_TRAMPOLINE): Adjust.
456 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
457 (ffi_prep_raw_closure): ... this.
458 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
459 (ffi_prep_closure): ... this.
460 (flush_icache): Adjust.
462 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
464 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
467 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
469 * Makefile.am: Add dummy install-pdf target.
470 * Makefile.in: Regenerate
472 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
474 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
475 ffi_closure_helper_SYSV): Add long double handling.
477 2007-02-02 Jakub Jelinek <jakub@redhat.com>
479 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
480 immediately after bctrl instruction.
482 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
484 * Makefile.am (all-recursive, install-recursive,
485 mostlyclean-recursive, clean-recursive, distclean-recursive,
486 maintainer-clean-recursive): Add missing targets.
487 * Makefile.in: Rebuilt.
489 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
491 * configure.ac: Add TARGET for x86_64-*-darwin*.
492 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
494 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
495 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
496 * configure: Regenerate.
497 * Makefile.in: Regenerate.
498 * include/Makefile.in: Regenerate.
499 * testsuite/Makefile.in: Regenerate.
500 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
503 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
505 * aclocal.m4: Regenerate with aclocal -I .. as written in the
508 2006-10-31 Geoffrey Keating <geoffk@apple.com>
510 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
511 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
513 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
514 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
516 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
517 Sandro Tolaini <tolaini@libero.it>
519 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
521 * configure: Regenerated.
522 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
523 (EXTRA_DIST): Add src/x86/darwin.S.
524 * Makefile.in: Regenerated.
525 * include/Makefile.in: Regenerated.
526 * testsuite/Makefile.in: Regenerated.
528 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
529 X86_WIN32, and additionally align stack to 16 bytes.
530 * src/x86/darwin.S: New, based on sysv.S.
531 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
533 2006-09-12 David Daney <ddaney@avtrex.com>
536 * include/Makefile.am: Install both ffi.h and ffitarget.h in
537 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
538 * aclocal.m4: Regenerated for automake 1.9.6.
539 * Makefile.in: Regenerated.
540 * include/Makefile.in: Regenerated.
541 * testsuite/Makefile.in: Regenerated.
543 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
545 * include/ffi_common.h (struct): Revert accidental commit.
547 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
549 * include/ffi_common.h: Remove lint directives.
550 * include/ffi.h.in: Likewise.
552 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
554 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
555 for 32-bit architectures.
556 * testsuite/libffi.call/return_ul.c: New test case.
558 2006-07-19 David Daney <ddaney@avtrex.com>
560 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
561 xfail remains for mips64.
563 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
565 * Makefile.am: Add install-html target. Add install-html to .PHONY
566 * Makefile.in: Regenerate.
567 * aclocal.m4: Regenerate.
568 * include/Makefile.in: Regenerate.
569 * testsuite/Makefile.in: Regenerate.
571 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
573 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
576 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
578 * README: Remove notice about 'Crazy Comments'.
579 * src/debug.c: Remove lint directives. Cleanup white spaces.
580 * src/java_raw_api.c: Likewise.
581 * src/prep_cif.c: Likewise.
582 * src/raw_api.c: Likewise.
583 * src/ffitest.c: Delete. No longer needed, all test cases migrated
585 * src/arm/ffi.c: Remove lint directives.
586 * src/m32r/ffi.c: Likewise.
587 * src/pa/ffi.c: Likewise.
588 * src/powerpc/ffi.c: Likewise.
589 * src/powerpc/ffi_darwin.c: Likewise.
590 * src/sh/ffi.c: Likewise.
591 * src/sh64/ffi.c: Likewise.
592 * src/x86/ffi.c: Likewise.
593 * testsuite/libffi.call/float2.c: Likewise.
594 * testsuite/libffi.call/promotion.c: Likewise.
595 * testsuite/libffi.call/struct1.c: Likewise.
597 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
599 * src/pa/hpux32.S: Correct unwind offset calculation for
601 * src/pa/linux.S: Likewise.
603 2006-04-12 James E Wilson <wilson@specifix.com>
606 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
607 (hfa_type_load): Call stf_spill.
608 (hfa_type_store): Call ldf_fill.
609 (ffi_call): Adjust calls to above routines. Add local temps for
612 2006-04-10 Matthias Klose <doko@debian.org>
614 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
615 directory names containing underscores.
617 2006-04-07 James E Wilson <wilson@specifix.com>
619 * testsuite/libffi.call/float4.c: New testcase.
621 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
622 Andreas Tobler <a.tobler@schweiz.ch>
624 * Makefile.am: Add PA_HPUX port.
625 * Makefile.in: Regenerate.
626 * include/Makefile.in: Likewise.
627 * testsuite/Makefile.in: Likewise.
628 * configure.ac: Add PA_HPUX rules.
629 * configure: Regenerate.
630 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
631 Add PA_HPUX and PA64_HPUX.
632 Rename FFI_LINUX ABI to FFI_PA32 ABI.
633 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
634 (FFI_TYPE_SMALL_STRUCT2): Define.
635 (FFI_TYPE_SMALL_STRUCT4): Likewise.
636 (FFI_TYPE_SMALL_STRUCT8): Likewise.
637 (FFI_TYPE_SMALL_STRUCT3): Redefine.
638 (FFI_TYPE_SMALL_STRUCT5): Likewise.
639 (FFI_TYPE_SMALL_STRUCT6): Likewise.
640 (FFI_TYPE_SMALL_STRUCT7): Likewise.
641 * src/pa/ffi.c (ROUND_DOWN): Delete.
642 (fldw, fstw, fldd, fstd): Use '__asm__'.
643 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
644 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
645 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
646 Simplify incrementing of stack slot variable. Change type of local
648 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
650 (ffi_prep_cif_machdep): Likewise.
651 (ffi_call): Likewise.
652 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
653 return type to ffi_status. Simplify incrementing of stack slot
654 variable. Only copy floating point argument registers when PA_LINUX
655 is true. Reformat debug statement.
656 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
657 FFI_TYPE_SMALL_STRUCT8.
658 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
660 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
661 Add nops to cache flush. Add trampoline for PA_HPUX.
662 * src/pa/hpux32.S: New file.
663 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
664 ffi_prep_args_LINUX to ffi_prep_args_pa32.
665 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
666 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
667 argument type checks so that common argument types appear first.
668 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
669 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
671 2006-03-24 Alan Modra <amodra@bigpond.net.au>
673 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
674 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
675 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
677 (ffi_prep_args64): Assert using IBM extended double.
678 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
679 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
680 (ffi_call): Handle FFI_LINUX.
681 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
682 gpr3 return pointer as for struct return. Handle FFI_LINUX
683 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
685 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
686 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
687 Don't use r6 as pointer to results, instead use sp offset. Don't
688 make a special call to load lr with case table address, instead
689 use offset from previous call.
690 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
691 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
694 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
696 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
697 passed with FP registers correctly.
698 (ffi_closure_helper_SYSV): Likewise.
699 * src/sh64/sysv.S: Likewise.
701 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
703 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
704 args and userdata unused.
705 (closure_test_fn1): Mark cif and userdata unused.
706 (main): Remove unused res.
708 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
710 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
711 -O2, -O3, -Os and the warning flags -W -Wall.
712 * testsuite/libffi.special/special.exp: Likewise.
713 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
714 unused parameter unused for gcc or else do nothing.
715 * testsuite/libffi.special/ffitestcxx.h: Likewise.
716 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
718 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
719 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
720 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
721 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
722 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
723 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
724 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
725 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
726 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
727 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
728 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
729 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
730 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
731 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
732 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
733 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
734 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
735 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
736 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
737 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
738 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
739 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
740 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
741 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
743 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
745 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
747 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
748 void* to avoid compiler warning.
750 (cls_struct_align_gn): Mark cif and userdata unused.
751 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
753 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
755 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
757 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
759 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
761 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
762 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
763 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
765 (main): Cast res_call to silence gcc.
766 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
768 (main): Cast res_call to silence gcc.
769 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
771 (main): Cast res_call to silence gcc.
772 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
774 (main): Cast res_call to silence gcc.
775 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
777 (main): Cast res_call to silence gcc.
778 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
780 (main): Cast res_call to silence gcc.
781 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
783 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
784 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
786 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
787 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
789 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
790 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
792 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
793 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
795 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
796 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
798 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
800 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
801 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
802 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
803 Cleanup white spaces.
804 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
805 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
806 cif and userdata unused.
807 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
809 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
810 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
811 formatters to silence gcc.
812 (B_gn): Mark cif and userdata unused.
813 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
815 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
816 (B_gn): Mark cif and userdata unused.
817 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
819 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
820 (B_gn): Mark cif and userdata unused.
821 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
823 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
824 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
825 * testsuite/libffi.call/problem1.c (stub): Likewise.
826 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
828 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
829 in the last commit for this test case in the test case itself.
830 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
832 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
833 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
834 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
835 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
836 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
837 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
839 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
841 * src/sh/sysv.S: Fix register numbers in the FDE for
844 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
846 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
847 declaration to avoid a false negative on ix86. See PR323.
849 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
851 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
852 and cast integer to void * if needed. Update the pointer to
853 the FP register saved area correctly.
855 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
857 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
859 * testsuite/libffi.call/nested_struct4.c: Likewise.
861 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
863 * testsuite/libffi.call/return_dbl.c: New test case.
864 * testsuite/libffi.call/return_dbl1.c: Likewise.
865 * testsuite/libffi.call/return_dbl2.c: Likewise.
866 * testsuite/libffi.call/return_fl.c: Likewise.
867 * testsuite/libffi.call/return_fl1.c: Likewise.
868 * testsuite/libffi.call/return_fl2.c: Likewise.
869 * testsuite/libffi.call/return_fl3.c: Likewise.
870 * testsuite/libffi.call/closure_fn6.c: Likewise.
872 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
874 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
875 here to be used by other test cases too.
877 * testsuite/libffi.call/nested_struct10.c: New test case.
878 * testsuite/libffi.call/nested_struct9.c: Likewise.
879 * testsuite/libffi.call/nested_struct8.c: Likewise.
880 * testsuite/libffi.call/nested_struct7.c: Likewise.
881 * testsuite/libffi.call/nested_struct6.c: Likewise.
882 * testsuite/libffi.call/nested_struct5.c: Likewise.
883 * testsuite/libffi.call/nested_struct4.c: Likewise.
885 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
887 * configure.ac: Enable libffi for sparc64-*-freebsd*.
888 * configure: Rebuilt.
890 2006-01-18 Jakub Jelinek <jakub@redhat.com>
892 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
893 instead do the shifting inline.
894 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
895 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
896 and 8 byte structs, for the remaining struct sizes don't call
897 __lshrdi3, instead do the shifting inline.
899 2005-12-07 Thiemo Seufer <ths@networkno.de>
901 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
903 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
904 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
905 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
906 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
907 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
909 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
911 (ffi_prep_closure): Improve trampoline instruction scheduling.
912 (ffi_closure_mips_inner_O32): Fix endianness bugs.
914 2005-12-03 Alan Modra <amodra@bigpond.net.au>
916 * src/powerpc/ffi.c: Formatting.
917 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
918 (ffi_prep_args64): Likewise.
920 2005-09-30 Geoffrey Keating <geoffk@apple.com>
922 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
923 darwin, use -shared-libgcc not -lgcc_s, and explain why.
925 2005-09-26 Tom Tromey <tromey@redhat.com>
927 * testsuite/libffi.call/float1.c (value_type): New typedef.
928 (CANARY): New define.
929 (main): Check for result buffer overflow.
930 * src/powerpc/linux64.S: Handle linux64 long double returns.
931 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
932 (ffi_prep_cif_machdep): Handle linux64 long double returns.
934 2005-08-25 Alan Modra <amodra@bigpond.net.au>
937 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
939 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
941 2005-08-11 Jakub Jelinek <jakub@redhat.com>
943 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
944 (AH_BOTTOM): Add FFI_HIDDEN definition.
945 * configure: Rebuilt.
946 * fficonfig.h.in: Rebuilt.
947 * src/powerpc/ffi.c (hidden): Remove.
948 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
949 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
950 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
951 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
952 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
953 add FFI_HIDDEN to its prototype.
954 (ffi_closure_SYSV_inner): New.
955 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
956 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
958 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
961 * configure: Rebuilt.
962 * configure.ac: Handle i*86-*-gnu*.
964 2005-08-09 Jakub Jelinek <jakub@redhat.com>
966 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
967 DW_CFA_offset_extended_sf rather than
968 DW_CFA_GNU_negative_offset_extended.
969 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
971 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
973 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
975 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
976 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
977 partially on register.
978 (ffi_closure_helper_SYSV): Likewise.
979 (ffi_prep_cif_machdep): Don't set too many cif->flags.
981 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
983 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
985 * src/sh64/ffi.c (simple_type): Remove.
986 (return_type): Handle small structures correctly.
987 (ffi_prep_args): Likewise.
988 (ffi_call): Likewise.
989 (ffi_closure_helper_SYSV): Likewise.
990 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
991 Emit position independent code if PIC and remove wrong datalabel
992 prefixes from EH data.
994 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
996 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
997 * Makefile.in: Regenerate.
998 * include/Makefile.in: Likewise.
999 * testsuite/Makefile.in: Likewise.
1000 * configure.ac: Add POWERPC_FREEBSD rules.
1001 * configure: Regenerate.
1002 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1003 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1004 * src/powerpc/ffi.c: Add flags to handle small structure returns
1006 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1008 (ffi_closure_helper_SYSV): Likewise.
1009 * src/powerpc/ppc_closure.S: Add return types for small structures.
1010 * src/powerpc/sysv.S: Add bits to handle small structures for
1013 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1015 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1016 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1017 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1019 2005-07-05 Randolph Chung <tausq@debian.org>
1021 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1022 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1023 structures. Kill compilation warnings.
1024 (ffi_closure_inner_LINUX): Print return values as hex in debug
1025 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1026 Properly handle 5-7 byte structure returns.
1027 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1028 (FFI_TYPE_SMALL_STRUCT2): Remove.
1029 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1030 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1031 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1032 (checksmst1, checksmst2): Remove.
1033 (checksmst3): Optimize handling of 3-byte struct returns.
1034 (checksmst567): Properly handle 5-7 byte struct returns.
1036 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1039 * src/mips/n32.S: Enforce PIC code.
1040 * src/mips/o32.S: Likewise.
1042 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1044 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1045 * configure: Regenerate.
1047 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1049 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1050 to call ffi_closure_helper_SYSV. Append @local instead.
1051 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1053 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1055 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1056 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1057 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1058 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1059 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1061 2005-05-09 Mike Stump <mrs@apple.com>
1063 * configure: Regenerate.
1065 2005-05-08 Richard Henderson <rth@redhat.com>
1068 * src/alpha/osf.S: Update unwind into to match code.
1070 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1071 Richard Henderson <rth@redhat.com>
1073 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1075 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1076 (ffi_prep_closure): Set carry bit if sse-used flag set.
1077 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1078 Only load sse registers if ssecount non-zero.
1079 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1081 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1083 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1084 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1085 * configure: Regenerate.
1087 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1089 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1090 have Tcl8.3-compatible intermediate variable.
1092 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1093 Hans-Peter Nilsson <hp@axis.com>
1095 * Makefile.am: Add CRIS support.
1096 * configure.ac: Likewise.
1097 * Makefile.in, configure, testsuite/Makefile.in,
1098 include/Makefile.in: Regenerate.
1099 * src/cris: New directory.
1100 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1101 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1103 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1104 \r?\n in output tests.
1106 2005-04-12 Mike Stump <mrs@apple.com>
1108 * configure: Regenerate.
1110 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1112 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1114 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1116 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1118 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1119 to 64 bits in ILP32 mode.
1120 Load 64 bits even for short data.
1122 2005-03-23 Mike Stump <mrs@apple.com>
1124 * src/powerpc/darwin.S: Update for -m64 multilib.
1125 * src/powerpc/darwin_closure.S: Likewise.
1127 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1129 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1130 Do not set tool_include_dir.
1131 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1133 * include/Makefile.am: Set gcc_version and toollibffidir.
1134 * include/Makefile.in: Regenerate.
1136 2005-02-22 Andrew Haley <aph@redhat.com>
1138 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1139 odd-numbered register pairs for 64-bit integer types.
1141 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1144 * testsuite/libffi.call/return_ll1.c: New test case.
1146 2005-02-11 Janis Johnson <janis187@us.ibm.com>
1148 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1149 * testsuite/libffi.call/float.c: Ditto.
1150 * testsuite/libffi.call/float2.c: Ditto.
1151 * testsuite/libffi.call/float3.c: Ditto.
1153 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1155 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1157 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1159 * testsuite/libffi.special/special.exp (cxx_options): Add
1162 2004-12-31 Richard Henderson <rth@redhat.com>
1164 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1165 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1166 offset parameters with a type parameter; deduce size and structure
1167 alignment. Update all users.
1169 2004-12-31 Richard Henderson <rth@redhat.com>
1171 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1172 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
1173 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
1174 into ffi_prep_closure.
1175 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1178 2004-12-27 Richard Henderson <rth@redhat.com>
1180 * src/x86/unix64.S: Fix typo in unwind info.
1182 2004-12-25 Richard Henderson <rth@redhat.com>
1184 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1185 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1186 (merge_classes): Check for it.
1188 (classify_argument): Pass byte_offset by value; perform all updates
1190 (examine_argument): Add classes argument; handle
1191 X86_64_COMPLEX_X87_CLASS.
1192 (ffi_prep_args): Merge into ...
1193 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1194 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1195 (ffi_fill_return_value): Remove.
1196 (ffi_prep_closure): Remove dead assert.
1197 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1198 Rewrite to use struct register_args instead of va_list. Create
1199 flags for handling structure returns.
1200 * src/x86/unix64.S: Remove dead strings.
1201 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1202 stack frame with ffi_call. Handle structure returns properly.
1203 (float2sse, floatfloat2sse, double2sse): Remove.
1204 (sse2float, sse2double, sse2floatfloat): Remove.
1205 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1206 to handle structure returns properly.
1208 2004-12-08 David Edelsohn <edelsohn@gnu.org>
1210 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1212 * Makefile.in: Regenerated.
1214 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
1216 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1217 * configure, aclocal.m4, Makefile.in: Regenerate.
1218 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1220 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1222 * configure: Regenerate for libtool change.
1224 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1226 * configure: Regenerate for libtool reversion.
1228 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1230 * configure: Regenerate for libtool change.
1232 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1234 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1236 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
1238 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1239 of jal. Use an absolute encoding for the frame information.
1241 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1243 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1244 * acinclude.m4: Delete logic for sincludes.
1245 * aclocal.m4, Makefile.in, configure: Regenerate.
1246 * include/Makefile: Likewise.
1247 * testsuite/Makefile: Likewise.
1249 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1251 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1252 on a 8-byte boundary.
1253 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1255 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1257 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1258 long long values. Round stack allocation to a multiple of 8 bytes
1259 for ATPCS compatibility.
1260 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1261 names. Handle returning long long types. Add Thumb and interworking
1262 support. Improve soft-float code.
1264 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1266 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1267 (libffi_exit): New function.
1268 (libffi_init): Build the testglue wrapper if needed.
1270 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1273 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1275 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1277 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1279 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1281 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1282 * testsuite/libffi.call/float3.c: New test case.
1284 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1286 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1287 the function returning a structure pointed with R2.
1288 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1289 the structure return value if T bit set. Emit position
1290 independent code and EH data if PIC.
1292 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1294 * Makefile.am: Add m32r support.
1295 * configure.ac: Likewise.
1296 * Makefile.in: Regenerate.
1297 * confiugre: Regenerate.
1298 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1299 (uint64, sint64, double, longdouble)
1300 * src/m32r: New directory.
1301 * src/m32r/ffi.c: New file.
1302 * src/m32r/sysv.S: Likewise.
1303 * src/m32r/ffitarget.h: Likewise.
1305 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1307 * testsuite/libffi.call/negint.c: New test case.
1309 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1312 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1313 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1314 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1317 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1319 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1320 * testsuite/libffi.call/promotion.c: Likewise.
1321 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1323 * testsuite/libffi.call/return_sc.c: Likewise.
1324 * testsuite/libffi.call/return_uc.c: Likewise.
1326 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1328 * src/powerpc/darwin.S: Fix comments and identation.
1329 * src/powerpc/darwin_closure.S: Likewise.
1331 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1333 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1334 (ffi_prep_args): Handle longdouble arguments.
1335 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1337 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1338 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1340 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1341 * src/types.c: Defined longdouble size and alignment for darwin.
1343 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1345 * src/powerpc/aix.S: Remove whitespaces.
1346 * src/powerpc/aix_closure.S: Likewise.
1347 * src/powerpc/asm.h: Likewise.
1348 * src/powerpc/ffi.c: Likewise.
1349 * src/powerpc/ffitarget.h: Likewise.
1350 * src/powerpc/linux64.S: Likewise.
1351 * src/powerpc/linux64_closure.S: Likewise.
1352 * src/powerpc/ppc_closure.S: Likewise.
1353 * src/powerpc/sysv.S: Likewise.
1355 2004-08-30 Anthony Green <green@redhat.com>
1357 * Makefile.am: Add frv support.
1358 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1359 * configure.ac: Read configure.host.
1360 * configure.in: Read configure.host.
1361 * configure.host: New file. frv-elf needs libgloss.
1362 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
1363 alignment. This is needed to frv and shouldn't harm the others.
1364 * include/ffi_common.h (ALIGN_DOWN): New macro.
1365 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1367 2004-08-24 David Daney <daney@avtrex.com>
1369 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1370 * testsuite/libffi.call/closure_fn1.c: Likewise.
1371 * testsuite/libffi.call/closure_fn2.c Likewise.
1372 * testsuite/libffi.call/closure_fn3.c: Likewise.
1373 * testsuite/libffi.call/closure_fn4.c: Likewise.
1374 * testsuite/libffi.call/closure_fn5.c: Likewise.
1375 * testsuite/libffi.call/cls_18byte.c: Likewise.
1376 * testsuite/libffi.call/cls_19byte.c: Likewise.
1377 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1378 * testsuite/libffi.call/cls_20byte.c: Likewise.
1379 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1380 * testsuite/libffi.call/cls_24byte.c: Likewise.
1381 * testsuite/libffi.call/cls_2byte.c: Likewise.
1382 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1383 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1384 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1385 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1386 * testsuite/libffi.call/cls_4byte.c: Likewise.
1387 * testsuite/libffi.call/cls_64byte.c: Likewise.
1388 * testsuite/libffi.call/cls_6byte.c: Likewise.
1389 * testsuite/libffi.call/cls_7byte.c: Likewise.
1390 * testsuite/libffi.call/cls_8byte.c: Likewise.
1391 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1392 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1393 * testsuite/libffi.call/cls_align_double.c: Likewise.
1394 * testsuite/libffi.call/cls_align_float.c: Likewise.
1395 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1396 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1397 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1398 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1399 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1400 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1401 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1402 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1403 * testsuite/libffi.call/cls_double.c: Likewise.
1404 * testsuite/libffi.call/cls_float.c: Likewise.
1405 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1406 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1407 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1408 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1409 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1410 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1411 * testsuite/libffi.call/cls_schar.c: Likewise.
1412 * testsuite/libffi.call/cls_sint.c: Likewise.
1413 * testsuite/libffi.call/cls_sshort.c: Likewise.
1414 * testsuite/libffi.call/cls_uchar.c: Likewise.
1415 * testsuite/libffi.call/cls_uint.c: Likewise.
1416 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1417 * testsuite/libffi.call/cls_ushort.c: Likewise.
1418 * testsuite/libffi.call/nested_struct.c: Likewise.
1419 * testsuite/libffi.call/nested_struct1.c: Likewise.
1420 * testsuite/libffi.call/nested_struct2.c: Likewise.
1421 * testsuite/libffi.call/nested_struct3.c: Likewise.
1422 * testsuite/libffi.call/problem1.c: Likewise.
1423 * testsuite/libffi.special/unwindtest.cc: Likewise.
1424 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1426 * testsuite/libffi.call/cls_16byte.c: Likewise.
1427 * testsuite/libffi.call/cls_5byte.c: Likewise.
1429 2004-08-23 David Daney <daney@avtrex.com>
1432 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1433 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1434 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1435 parameters and return types.
1436 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1437 (ffi_prep_closure): Ditto.
1438 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1439 alignment calculations.
1440 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1441 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1443 2004-08-14 Casey Marshall <csm@gnu.org>
1445 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1446 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1447 integer (O32 ABI only).
1448 (ffi_prep_closure): new function.
1449 (ffi_closure_mips_inner_O32): new function.
1450 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
1451 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1452 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1453 64 bit integers correctly.
1454 (ffi_closure_O32): new function.
1455 Added DWARF-2 unwind info for both functions.
1457 2004-08-10 Andrew Haley <aph@redhat.com>
1459 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1461 2004-08-01 Robert Millan <robertmh@gnu.org>
1463 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1464 * configure: Regenerate.
1466 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1468 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1469 and mmap() explicitly instead of relying on preset autoconf cache
1471 * aclocal.m4: Regenerate.
1472 * configure: Regenerate.
1474 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1476 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1477 (ffi_check_float_struct): Remove unused prototype.
1479 2004-06-30 Geoffrey Keating <geoffk@apple.com>
1481 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1482 character on Darwin, use '\n\t' instead.
1484 2004-06-26 Matthias Klose <doko@debian.org>
1486 * libtool-version: Fix typo in revision/age.
1488 2004-06-17 Matthias Klose <doko@debian.org>
1490 * libtool-version: New.
1491 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1492 * Makefile.in: Regenerate.
1494 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1496 * Makefile.am: Remove useless multilib rules.
1497 * Makefile.in: Regenerate.
1498 * aclocal.m4: Regenerate with automake 1.8.5.
1499 * configure.ac: Remove useless multilib configury.
1500 * configure: Regenerate.
1502 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1504 * .cvsignore: New file.
1506 2004-06-10 Jakub Jelinek <jakub@redhat.com>
1508 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1510 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1513 2004-06-06 Sean McNeil <sean@mcneil.com>
1515 * configure.ac: Add x86_64-*-freebsd* support.
1516 * configure: Regenerate.
1518 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
1521 * configure.ac: Test for existence of mmap and sys/mman.h before
1522 checking blacklist. Fix suggested by Jim Wilson.
1523 * configure: Regenerate.
1525 2004-04-26 Matt Austern <austern@apple.com>
1527 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1529 * src/powerpc/darwin_closure.S: Likewise.
1531 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1533 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1534 error. Reported by Thomas Heller <theller@python.net>.
1535 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1536 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1538 2004-03-20 Matthias Klose <doko@debian.org>
1540 * src/pa/linux.S: Fix typo.
1542 2004-03-19 Matthias Klose <doko@debian.org>
1544 * Makefile.am: Update.
1545 * Makefile.in: Regenerate.
1546 * src/pa/ffi.h.in: Remove.
1547 * src/pa/ffitarget.h: New file.
1549 2004-02-10 Randolph Chung <tausq@debian.org>
1551 * Makefile.am: Add PA support.
1552 * Makefile.in: Regenerate.
1553 * include/Makefile.in: Regenerate.
1554 * configure.ac: Add PA target.
1555 * configure: Regenerate.
1556 * src/pa/ffi.c: New file.
1557 * src/pa/ffi.h.in: Add PA support.
1558 * src/pa/linux.S: New file.
1559 * prep_cif.c: Add PA support.
1561 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
1563 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
1565 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
1566 with ecif->cif->flags.
1567 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
1569 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
1570 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
1571 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
1572 sc_retstruct2b): Add for 1 or 2-bytes struct case.
1574 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
1576 * configure.in: Rename file to ...
1577 * configure.ac: ... this.
1578 * fficonfig.h.in: Regenerate.
1579 * Makefile.in: Regenerate.
1580 * include/Makefile.in: Regenerate.
1581 * testsuite/Makefile.in: Regenerate.
1583 2004-03-12 Matt Austern <austern@apple.com>
1585 * src/powerpc/darwin.S: Fix EH information so it corresponds to
1586 changes in EH format resulting from addition of linkonce support.
1587 * src/powerpc/darwin_closure.S: Likewise.
1589 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
1590 Paolo Bonzini <bonzini@gnu.org>
1592 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
1593 Remove VPATH. Remove rules for object files. Remove multilib support.
1594 (AM_CCASFLAGS): Add.
1595 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
1596 (AC_PREREQ): Bump version to 2.59.
1597 (AC_INIT): Fill with version info and bug address.
1598 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
1599 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
1600 De-precious CC so that the right flags are passed down to multilibs.
1601 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
1602 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
1603 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
1604 * configure: Rebuilt.
1605 * aclocal.m4: Likewise.
1606 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
1607 * fficonfig.h.in: Likewise.
1609 2004-03-11 Andreas Schwab <schwab@suse.de>
1611 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
1612 arguments from fp registers only for the first 8 parameter slots.
1613 Don't convert a float parameter when passed in memory.
1615 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
1617 * configure: Regenerate for config/accross.m4 correction.
1619 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
1621 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
1622 ecif->cif->bytes to bytes.
1623 (ffi_prep_cif_machdep): Add braces around nested if statement.
1625 2004-02-09 Alan Modra <amodra@bigpond.net.au>
1627 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
1629 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
1630 (ffi_closure_helper_LINUX64): Fix typo.
1631 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
1633 * testsuite/libffi.call/float.c: Likewise.
1634 * testsuite/libffi.call/float2.c: Likewise.
1636 2004-02-08 Alan Modra <amodra@bigpond.net.au>
1638 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
1639 long double function return and long double arg handling.
1640 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
1641 Use "end_pfr" instead of "nf". Correct long double handling.
1643 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
1645 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
1646 space for long double return value. Adjust stack frame and offsets.
1647 Load f2 long double return.
1649 2004-02-07 Alan Modra <amodra@bigpond.net.au>
1651 * src/types.c: Use 16 byte long double for POWERPC64.
1653 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1655 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
1656 when the structure return address is passed in %o0.
1657 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
1658 (ffi_v9_layout_struct): Align the field following a nested structure
1659 on a word boundary. Use memmove instead of memcpy.
1660 (ffi_call): Update call to ffi_V9_return_struct.
1661 (ffi_prep_closure): Define 'ctx' only for V8.
1662 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
1663 and ffi_closure_sparc_inner_v9.
1664 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
1665 Always skip the structure return address. For structures and long
1666 doubles, copy the argument directly.
1667 (ffi_closure_sparc_inner_v9): Skip the structure return address only
1668 if required. Shift the maximum floating-point slot accordingly. For
1669 big structures, copy the argument directly; otherwise, left-justify the
1670 argument and call ffi_v9_layout_struct to lay out the structure on
1672 * src/sparc/v8.S: Undef STACKFRAME before defining it.
1673 (ffi_closure_v8): Pass the structure return address. Update call to
1674 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
1675 Skip the 'unimp' insn when returning long doubles and structures.
1676 * src/sparc/v9.S: Undef STACKFRAME before defining it.
1677 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
1678 FFI_TYPE_INT handling. Load structures both in integers and
1679 floating-point registers on return.
1680 * README: Update status of the SPARC port.
1682 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
1684 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
1686 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
1688 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
1690 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
1691 value as of type ffi_arg, not unsigned int.
1693 2004-01-21 Michael Ritzert <ritzert@t-online.de>
1695 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
1698 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
1700 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
1703 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1705 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
1708 2003-12-10 Richard Henderson <rth@redhat.com>
1710 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
1711 size_t instead of int.
1713 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
1715 * testsuite/libffi.call/many_win32.c: Include <float.h>.
1716 * testsuite/libffi.call/many_win32.c (main): Replace variable
1717 int i with unsigned long ul.
1719 * testsuite/libffi.call/cls_align_uint64.c: New test case.
1720 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1721 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1722 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1723 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1724 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1725 * testsuite/libffi.call/cls_align_float.c: Likewise.
1726 * testsuite/libffi.call/cls_align_double.c: Likewise.
1727 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1728 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1730 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
1733 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
1734 Align arguments to 32 bits.
1736 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
1739 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
1740 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1741 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1742 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1743 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1744 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1746 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
1748 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
1750 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
1751 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1753 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
1755 * Makefile.in: Rebuilt.
1756 * configure: Likewise.
1757 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
1760 2003-11-21 Andreas Jaeger <aj@suse.de>
1761 Andreas Tobler <a.tobler@schweiz.ch>
1763 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
1764 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
1765 * Makefile.in: Rebuilt.
1766 * aclocal.m4: Likewise.
1767 * configure: Likewise.
1768 * fficonfig.h.in: Likewise.
1769 * testsuite/lib/libffi-dg.exp: Add include dir.
1770 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
1771 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1772 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
1773 for ffi_closure if available.
1774 * testsuite/libffi.call/closure_fn1.c: Likewise.
1775 * testsuite/libffi.call/closure_fn2.c: Likewise.
1776 * testsuite/libffi.call/closure_fn3.c: Likewise.
1777 * testsuite/libffi.call/closure_fn4.c: Likewise.
1778 * testsuite/libffi.call/closure_fn5.c: Likewise.
1779 * testsuite/libffi.call/cls_12byte.c: Likewise.
1780 * testsuite/libffi.call/cls_16byte.c: Likewise.
1781 * testsuite/libffi.call/cls_18byte.c: Likewise.
1782 * testsuite/libffi.call/cls_19byte.c: Likewise.
1783 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1784 * testsuite/libffi.call/cls_20byte.c: Likewise.
1785 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1786 * testsuite/libffi.call/cls_24byte.c: Likewise.
1787 * testsuite/libffi.call/cls_2byte.c: Likewise.
1788 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1789 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1790 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1791 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1792 * testsuite/libffi.call/cls_4byte.c: Likewise.
1793 * testsuite/libffi.call/cls_5byte.c: Likewise.
1794 * testsuite/libffi.call/cls_64byte.c: Likewise.
1795 * testsuite/libffi.call/cls_6byte.c: Likewise.
1796 * testsuite/libffi.call/cls_7byte.c: Likewise.
1797 * testsuite/libffi.call/cls_8byte.c: Likewise.
1798 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1799 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1800 * testsuite/libffi.call/cls_double.c: Likewise.
1801 * testsuite/libffi.call/cls_float.c: Likewise.
1802 * testsuite/libffi.call/cls_schar.c: Likewise.
1803 * testsuite/libffi.call/cls_sint.c: Likewise.
1804 * testsuite/libffi.call/cls_sshort.c: Likewise.
1805 * testsuite/libffi.call/cls_uchar.c: Likewise.
1806 * testsuite/libffi.call/cls_uint.c: Likewise.
1807 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1808 * testsuite/libffi.call/cls_ushort.c: Likewise.
1809 * testsuite/libffi.call/nested_struct.c: Likewise.
1810 * testsuite/libffi.call/nested_struct1.c: Likewise.
1811 * testsuite/libffi.call/nested_struct2.c: Likewise.
1812 * testsuite/libffi.call/nested_struct3.c: Likewise.
1813 * testsuite/libffi.call/problem1.c: Likewise.
1814 * testsuite/libffi.special/unwindtest.cc: Likewise.
1816 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
1818 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
1820 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
1822 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
1823 Add -lgcc_s to additional flags.
1825 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
1827 * configure.in, include/Makefile.am: PR libgcj/11147, install
1828 the ffitarget.h header file in a gcc versioned and target
1830 * configure: Regenerated.
1831 * Makefile.in, include/Makefile.in: Likewise.
1832 * testsuite/Makefile.in: Likewise.
1834 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
1836 * testsuite/libffi.call/closure_fn0.c: Print result and check
1837 with dg-output to make debugging easier.
1838 * testsuite/libffi.call/closure_fn1.c: Likewise.
1839 * testsuite/libffi.call/closure_fn2.c: Likewise.
1840 * testsuite/libffi.call/closure_fn3.c: Likewise.
1841 * testsuite/libffi.call/closure_fn4.c: Likewise.
1842 * testsuite/libffi.call/closure_fn5.c: Likewise.
1843 * testsuite/libffi.call/cls_12byte.c: Likewise.
1844 * testsuite/libffi.call/cls_16byte.c: Likewise.
1845 * testsuite/libffi.call/cls_18byte.c: Likewise.
1846 * testsuite/libffi.call/cls_19byte.c: Likewise.
1847 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1848 * testsuite/libffi.call/cls_20byte.c: Likewise.
1849 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1850 * testsuite/libffi.call/cls_24byte.c: Likewise.
1851 * testsuite/libffi.call/cls_2byte.c: Likewise.
1852 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1853 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1854 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1855 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1856 * testsuite/libffi.call/cls_4byte.c: Likewise.
1857 * testsuite/libffi.call/cls_5byte.c: Likewise.
1858 * testsuite/libffi.call/cls_64byte.c: Likewise.
1859 * testsuite/libffi.call/cls_6byte.c: Likewise.
1860 * testsuite/libffi.call/cls_7byte.c: Likewise.
1861 * testsuite/libffi.call/cls_8byte.c: Likewise.
1862 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1863 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1864 * testsuite/libffi.call/cls_double.c: Likewise.
1865 * testsuite/libffi.call/cls_float.c: Likewise.
1866 * testsuite/libffi.call/cls_schar.c: Likewise.
1867 * testsuite/libffi.call/cls_sint.c: Likewise.
1868 * testsuite/libffi.call/cls_sshort.c: Likewise.
1869 * testsuite/libffi.call/cls_uchar.c: Likewise.
1870 * testsuite/libffi.call/cls_uint.c: Likewise.
1871 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1872 * testsuite/libffi.call/cls_ushort.c: Likewise.
1873 * testsuite/libffi.call/problem1.c: Likewise.
1875 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
1878 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
1880 * testsuite/libffi.call/cls_9byte2.c: New test case.
1881 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1882 * testsuite/libffi.call/cls_64byte.c: Likewise.
1883 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1884 * testsuite/libffi.call/cls_19byte.c: Likewise.
1885 * testsuite/libffi.call/cls_18byte.c: Likewise.
1886 * testsuite/libffi.call/closure_fn4.c: Likewise.
1887 * testsuite/libffi.call/closure_fn5.c: Likewise.
1888 * testsuite/libffi.call/cls_schar.c: Likewise.
1889 * testsuite/libffi.call/cls_sint.c: Likewise.
1890 * testsuite/libffi.call/cls_sshort.c: Likewise.
1891 * testsuite/libffi.call/nested_struct2.c: Likewise.
1892 * testsuite/libffi.call/nested_struct3.c: Likewise.
1894 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
1896 * testsuite/libffi.call/cls_double.c: Do a check on the result.
1897 * testsuite/libffi.call/cls_uchar.c: Likewise.
1898 * testsuite/libffi.call/cls_uint.c: Likewise.
1899 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1900 * testsuite/libffi.call/cls_ushort.c: Likewise.
1901 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
1903 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
1905 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
1908 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
1910 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
1911 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
1913 2003-10-22 David Daney <ddaney@avtrex.com>
1915 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
1916 __attribute__((__mode__(__SI__))) and friends.
1918 2003-10-22 Andreas Schwab <schwab@suse.de>
1920 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
1922 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
1924 * configure.in: AC_LINK_FILES(ffitarget.h).
1925 * configure: Regenerate.
1926 * Makefile.in: Likewise.
1927 * include/Makefile.in: Likewise.
1928 * testsuite/Makefile.in: Likewise.
1929 * fficonfig.h.in: Likewise.
1931 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
1932 Richard Henderson <rth@redhat.com>
1934 Avoid that ffi.h includes fficonfig.h.
1936 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
1937 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
1938 (TARGET_SRC_MIPS_SGI): Removed.
1939 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
1940 (MIPS_SGI): Removed.
1941 (CLEANFILES): Removed.
1942 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
1944 * acconfig.h: Removed.
1945 * configure.in: Compute sizeofs only for double and long double.
1946 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
1947 into AC_DEFINE instead of using acconfig.h. Create
1948 include/ffitarget.h instead of include/fficonfig.h. Rename
1949 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
1950 AC_DEFINE EH_FRAME_FLAGS.
1951 * include/Makefile.am (DISTCLEANFILES): New automake macro.
1952 (hack_DATA): Add ffitarget.h.
1953 * include/ffi.h.in: Remove all system specific definitions.
1954 Declare raw API even if it is not installed, why bother?
1955 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
1956 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
1957 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
1958 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
1959 * include/ffi_common.h (bool): Do not define.
1960 (ffi_assert): Accept failed assertion.
1961 (ffi_type_test): Return void and accept file/line.
1962 (FFI_ASSERT): Pass stringized failed assertion.
1963 (FFI_ASSERT_AT): New macro.
1964 (FFI_ASSERT_VALID_TYPE): New macro.
1965 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
1966 UINT64, SINT64): Define here with gcc's __attribute__ macro
1968 (FLOAT32, ALIGN): Define here instead of in ffi.h
1969 * include/ffi-mips.h: Removed. Its content moved to
1970 src/mips/ffitarget.h after separating assembly and C sections.
1971 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
1972 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
1973 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
1974 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
1975 SIZEOF_ARG -> FFI_SIZEOF_ARG.
1976 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
1977 * src/debug.c (ffi_assert): Accept stringized failed assertion.
1978 (ffi_type_test): Rewritten.
1979 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
1980 FFI_ASSERT_VALID_TYPE.
1981 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
1982 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
1983 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
1984 src/s390/ffitarget.h, src/sh/ffitarget.h,
1985 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
1986 src/x86/ffitarget.h: New files.
1987 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
1988 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
1989 src/powerpc/aix.S, src/powerpc/darwin.S,
1990 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
1991 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
1992 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
1993 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
1994 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
1997 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1999 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2000 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2002 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2004 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2005 Used when FFI_DEBUG = 1.
2007 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2009 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2012 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2014 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2015 FFI_MIPS_O32 for O32 ABI.
2017 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2019 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2020 SPARC64. Cleanup whitespaces.
2022 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2024 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2025 strongarm, xscale. Cleanup whitespaces.
2026 * testsuite/libffi.call/closure_fn1.c: Likewise.
2027 * testsuite/libffi.call/closure_fn2.c: Likewise.
2028 * testsuite/libffi.call/closure_fn3.c: Likewise.
2029 * testsuite/libffi.call/cls_12byte.c: Likewise.
2030 * testsuite/libffi.call/cls_16byte.c: Likewise.
2031 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2032 * testsuite/libffi.call/cls_20byte.c: Likewise.
2033 * testsuite/libffi.call/cls_24byte.c: Likewise.
2034 * testsuite/libffi.call/cls_2byte.c: Likewise.
2035 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2036 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2037 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2038 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2039 * testsuite/libffi.call/cls_4byte.c: Likewise.
2040 * testsuite/libffi.call/cls_5byte.c: Likewise.
2041 * testsuite/libffi.call/cls_6byte.c: Likewise.
2042 * testsuite/libffi.call/cls_7byte.c: Likewise.
2043 * testsuite/libffi.call/cls_8byte.c: Likewise.
2044 * testsuite/libffi.call/cls_double.c: Likewise.
2045 * testsuite/libffi.call/cls_float.c: Likewise.
2046 * testsuite/libffi.call/cls_uchar.c: Likewise.
2047 * testsuite/libffi.call/cls_uint.c: Likewise.
2048 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2049 * testsuite/libffi.call/cls_ushort.c: Likewise.
2050 * testsuite/libffi.call/nested_struct.c: Likewise.
2051 * testsuite/libffi.call/nested_struct1.c: Likewise.
2052 * testsuite/libffi.call/problem1.c: Likewise.
2053 * testsuite/libffi.special/unwindtest.cc: Likewise.
2054 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2056 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2058 * src/powerpc/aix.S: Cleanup whitespaces.
2059 * src/powerpc/aix_closure.S: Likewise.
2061 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2063 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2064 * src/powerpc/darwin_closure.S: Likewise.
2065 * src/powerpc/ffi_darwin.c: Likewise.
2067 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2068 David Edelsohn <edelsohn@gnu.org>
2070 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2071 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2073 * src/powerpc/darwin_closure.S: Likewise.
2074 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2075 according to the Darwin/AIX ABI.
2076 (ffi_prep_cif_machdep): Likewise.
2077 (ffi_closure_helper_DARWIN): Likewise.
2078 Remove the outgoing parameter stack logic. Simplify the evaluation
2079 of the different CASE types.
2080 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2081 statement in the trampoline code.
2083 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2085 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2086 for the register size.
2087 (ffi_closure_helper_SYSV): Handle the structure return value
2089 (ffi_closure_helper_SYSV): Return the appropriate type when
2090 the registers are used for the structure return value.
2091 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2092 the 64-bit return value. Update copyright years.
2094 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2096 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2097 srcdir for ffi_mips.h.
2099 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2101 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2103 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2104 placement of float result.
2105 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2106 cast of "resp" for big-endian 64 bit machines.
2108 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2110 * src/types.c (double, longdouble): Merge identical SH and ARM
2111 typedefs, and add POWERPC64.
2112 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2113 struct split over gpr and rest.
2114 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2115 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2117 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2119 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2122 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2124 * configure: Regenerate.
2126 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2128 * Makefile.am: Remove build rules for ffitest.
2129 * Makefile.in: Rebuilt.
2131 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2133 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2134 about implicit declaration of abort().
2136 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2138 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2139 * Makefile.in: Rebuilt.
2140 * configure.in: Add dejagnu test framework.
2141 * configure: Rebuilt.
2143 * testsuite/Makefile.am: New file.
2144 * testsuite/Makefile.in: Built
2145 * testsuite/lib/libffi-dg.exp: New file.
2146 * testsuite/config/default.exp: Likewise.
2147 * testsuite/libffi.call/call.exp: Likewise.
2148 * testsuite/libffi.call/ffitest.h: Likewise.
2149 * testsuite/libffi.call/closure_fn0.c: Likewise.
2150 * testsuite/libffi.call/closure_fn1.c: Likewise.
2151 * testsuite/libffi.call/closure_fn2.c: Likewise.
2152 * testsuite/libffi.call/closure_fn3.c: Likewise.
2153 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2154 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2155 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2156 * testsuite/libffi.call/cls_2byte.c: Likewise.
2157 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2158 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2159 * testsuite/libffi.call/cls_4byte.c: Likewise.
2160 * testsuite/libffi.call/cls_5byte.c: Likewise.
2161 * testsuite/libffi.call/cls_6byte.c: Likewise.
2162 * testsuite/libffi.call/cls_7byte.c: Likewise.
2163 * testsuite/libffi.call/cls_8byte.c: Likewise.
2164 * testsuite/libffi.call/cls_12byte.c: Likewise.
2165 * testsuite/libffi.call/cls_16byte.c: Likewise.
2166 * testsuite/libffi.call/cls_20byte.c: Likewise.
2167 * testsuite/libffi.call/cls_24byte.c: Likewise.
2168 * testsuite/libffi.call/cls_double.c: Likewise.
2169 * testsuite/libffi.call/cls_float.c: Likewise.
2170 * testsuite/libffi.call/cls_uchar.c: Likewise.
2171 * testsuite/libffi.call/cls_uint.c: Likewise.
2172 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2173 * testsuite/libffi.call/cls_ushort.c: Likewise.
2174 * testsuite/libffi.call/float.c: Likewise.
2175 * testsuite/libffi.call/float1.c: Likewise.
2176 * testsuite/libffi.call/float2.c: Likewise.
2177 * testsuite/libffi.call/many.c: Likewise.
2178 * testsuite/libffi.call/many_win32.c: Likewise.
2179 * testsuite/libffi.call/nested_struct.c: Likewise.
2180 * testsuite/libffi.call/nested_struct1.c: Likewise.
2181 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2182 * testsuite/libffi.call/problem1.c: Likewise.
2183 * testsuite/libffi.call/promotion.c: Likewise.
2184 * testsuite/libffi.call/return_ll.c: Likewise.
2185 * testsuite/libffi.call/return_sc.c: Likewise.
2186 * testsuite/libffi.call/return_uc.c: Likewise.
2187 * testsuite/libffi.call/strlen.c: Likewise.
2188 * testsuite/libffi.call/strlen_win32.c: Likewise.
2189 * testsuite/libffi.call/struct1.c: Likewise.
2190 * testsuite/libffi.call/struct2.c: Likewise.
2191 * testsuite/libffi.call/struct3.c: Likewise.
2192 * testsuite/libffi.call/struct4.c: Likewise.
2193 * testsuite/libffi.call/struct5.c: Likewise.
2194 * testsuite/libffi.call/struct6.c: Likewise.
2195 * testsuite/libffi.call/struct7.c: Likewise.
2196 * testsuite/libffi.call/struct8.c: Likewise.
2197 * testsuite/libffi.call/struct9.c: Likewise.
2198 * testsuite/libffi.special/special.exp: New file.
2199 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2200 * testsuite/libffi.special/unwindtest.cc: Likewise.
2203 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2205 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2208 2003-08-02 Alan Modra <amodra@bigpond.net.au>
2210 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2212 (ffi_closure_helper_LINUX64): Likewise.
2213 * src/powerpc/linux64.S: Remove code writing to parm save area.
2214 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2215 address in lr from ffi_closure_helper_LINUX64 call to calculate
2216 table address. Optimize function tail.
2218 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2220 * src/sparc/ffi.c: Handle all floating point registers.
2221 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2223 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2225 * README: Note that libffi is not part of GCC. Update the project
2228 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2230 * src/powerpc/ppc_closure.S: Include ffi.h.
2232 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2234 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2235 Use C style comments.
2237 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2239 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2240 * Makefile.in: Regenerate.
2241 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2242 * configure: Regenerate.
2243 * include/ffi.h.in: Add SHmedia support.
2244 * src/sh64/ffi.c: New file.
2245 * src/sh64/sysv.S: New file.
2247 2003-05-16 Jakub Jelinek <jakub@redhat.com>
2249 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2250 should be read-only.
2251 * configure: Rebuilt.
2252 * fficonfig.h.in: Rebuilt.
2253 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2254 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2255 * src/powerpc/linux64.S: Likewise.
2256 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2257 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2258 if -fpic/-fPIC/-mrelocatable.
2259 * src/powerpc/powerpc_closure.S: Likewise.
2260 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2261 #write in .eh_frame flags.
2262 * src/sparc/v9.S: Likewise.
2263 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2264 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2265 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2267 2003-05-07 Jeff Sturm <jsturm@one-point.com>
2269 Fixes PR bootstrap/10656
2270 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2271 support for .register pseudo-op.
2272 * src/sparc/v8.S: Use it.
2273 * fficonfig.h.in: Rebuilt.
2274 * configure: Rebuilt.
2276 2003-04-18 Jakub Jelinek <jakub@redhat.com>
2278 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2279 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2280 Make it the default on POWERPC64.
2281 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2282 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2283 * configure: Rebuilt.
2284 * src/powerpc/ffi.c (hidden): Define.
2285 (ffi_prep_args_SYSV): Renamed from
2286 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2287 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2288 ASM_NEEDS_REGISTERS64): New.
2289 (ffi_prep_args64): New function.
2290 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2291 (ffi_call): Likewise.
2292 (ffi_prep_closure): Likewise.
2293 (flush_icache): Surround by #ifndef POWERPC64.
2294 (ffi_dblfl): New union type.
2295 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2296 (ffi_closure_helper_LINUX64): New function.
2297 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2299 * src/powerpc/sysv.S: Likewise.
2300 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2301 * src/powerpc/linux64.S: New file.
2302 * src/powerpc/linux64_closure.S: New file.
2303 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2304 src/powerpc/linux64_closure.S.
2305 (TARGET_SRC_POWERPC): Likewise.
2307 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2308 closure_test_fn3): Fix result printing on big-endian 64-bit
2310 (main): Print tst2_arg instead of uninitialized tst2_result.
2312 * src/ffitest.c (main): Hide what closure pointer really points to
2315 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
2317 * configure.in (arm-*-netbsdelf*): Add configuration.
2318 (configure): Regenerated.
2320 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
2322 * include/Makefile.in: Regenerate.
2324 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2326 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2328 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2329 Receive closure pointer through parameter, read args using
2330 __builtin_dwarf_cfa.
2331 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2333 2003-03-12 Andreas Schwab <schwab@suse.de>
2335 * configure.in: Avoid trailing /. in toolexeclibdir.
2336 * configure: Rebuilt.
2338 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2340 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
2342 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2344 * libffi/src/powerpc/darwin_closure.S:
2345 Fix alignement bug, allocate 8 bytes for the result.
2346 * libffi/src/powerpc/aix_closure.S:
2348 * libffi/src/powerpc/ffi_darwin.c:
2349 Update stackframe description for aix/darwin_closure.S.
2351 2003-02-06 Jakub Jelinek <jakub@redhat.com>
2353 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2356 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2357 Andreas Schwab <schwab@suse.de>
2359 * configure.in: Adjust command to source config-ml.in to account
2360 for changes to the libffi_basedir definition.
2361 (libffi_basedir): Remove ${srcdir} from value and include trailing
2364 * configure: Regenerate.
2366 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2368 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2370 2003-01-28 Andrew Haley <aph@redhat.com>
2372 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2373 * src/x86/ffi64.c (ffi_prep_closure): New.
2374 (ffi_closure_UNIX64_inner): New.
2375 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2377 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
2379 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2380 Remove USE_LIBDIR conditional.
2381 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2382 * Makefile.in, configure: Rebuilt.
2384 2003-01027 David Edelsohn <edelsohn@gnu.org>
2386 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2387 * Makefile.in: Regenerate.
2389 2003-01-22 Andrew Haley <aph@redhat.com>
2391 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2394 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2396 * src/powerpc/darwin.S: Add unwind info.
2397 * src/powerpc/darwin_closure.S: Likewise.
2399 2003-01-14 Andrew Haley <aph@redhat.com>
2401 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2402 (ffi_prep_cif_machdep): Likewise.
2403 * src/x86/unix64.S: Add unwind info.
2405 2003-01-14 Andreas Jaeger <aj@suse.de>
2407 * src/ffitest.c (main): Only use ffi_closures if those are
2410 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2412 * libffi/src/ffitest.c
2413 add closure testcases
2415 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2417 * libffi/src/powerpc/ffi.c
2418 fix alignment bug for float (4 byte aligned iso 8 byte)
2420 2003-01-09 Geoffrey Keating <geoffk@apple.com>
2422 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2423 * src/powerpc/darwin.S: Remove RCS version string.
2425 2003-01-03 Jeff Sturm <jsturm@one-point.com>
2427 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2428 * src/ffitest.c (main): Use static storage for closure.
2429 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2430 * src/sparc/v8.S (ffi_closure_v8): New.
2431 * src/sparc/v9.S (ffi_closure_v9): New.
2433 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2435 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2436 enumeration for X86_WIN32.
2437 * src/x86/win32.S: Added ffi_call_STDCALL function
2439 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2440 switch cases for recognising FFI_STDCALL and
2441 calling ffi_call_STDCALL if target is X86_WIN32.
2442 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2443 stdcall versions of the "my_strlen" and "many"
2444 test functions (for X86_WIN32).
2445 Added test cases to test stdcall invocation using
2448 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2450 * src/sh/sysv.S: Add DWARF2 unwind info.
2452 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2454 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2456 2002-11-26 Jim Wilson <wilson@redhat.com>
2458 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2460 2002-11-23 H.J. Lu <hjl@gnu.org>
2462 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2463 Include ../config/accross.m4.
2464 * aclocal.m4; Rebuild.
2465 * configure: Likewise.
2467 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2469 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2471 2002-11-11 DJ Delorie <dj@redhat.com>
2473 * configure.in: Look for common files in the right place.
2475 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2477 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2478 raw data as _Jv_word values, not ffi_raw.
2479 (ffi_java_ptrarray_to_raw): Likewise.
2480 (ffi_java_rvalue_to_raw): New function.
2481 (ffi_java_raw_call): Call it.
2482 (ffi_java_raw_to_rvalue): New function.
2483 (ffi_java_translate_args): Call it.
2484 * src/ffitest.c (closure_test_fn): Interpret return value
2485 as ffi_arg, not int.
2486 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2487 FFI_TYPE_POINTER case.
2488 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2489 values extended to word size.
2491 2002-10-02 Andreas Jaeger <aj@suse.de>
2493 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2495 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
2497 * include/ffi.h.in: Fix i386 win32 compilation.
2499 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2501 * configure.in: Add s390x-*-linux-* target.
2502 * configure: Regenerate.
2503 * include/ffi.h.in: Define S390X for s390x targets.
2504 (FFI_CLOSURES): Define for s390/s390x.
2505 (FFI_TRAMPOLINE_SIZE): Likewise.
2506 (FFI_NATIVE_RAW_API): Likewise.
2507 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2508 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
2509 * src/s390/ffi.c: Major rework of existing code. Add support for
2510 s390x targets. Add closure support.
2511 * src/s390/sysv.S: Likewise.
2513 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
2515 * src/arm/sysv.S: Fix typo.
2517 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
2519 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2520 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2521 (ffi_call_SYSV): Handle soft-float.
2523 2002-09-27 Bo Thorsen <bo@suse.de>
2525 * include/ffi.h.in: Fix multilib x86-64 support.
2527 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2529 * Makefile.am (all-multi): Fix multilib parallel build.
2531 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2533 * configure.in (sh[34]*-*-linux*): Add brackets.
2534 * configure: Regenerate.
2536 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2538 * Makefile.am: Add SH support.
2539 * Makefile.in: Regenerate.
2540 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2541 * configure: Regenerate.
2542 * include/ffi.h.in: Add SH support.
2543 * src/sh/ffi.c: New file.
2544 * src/sh/sysv.S: New file.
2545 * src/types.c: Add SH support.
2547 2002-07-16 Bo Thorsen <bo@suse.de>
2549 * src/x86/ffi64.c: New file that adds x86-64 support.
2550 * src/x86/unix64.S: New file that handles argument setup for
2552 * src/x86/sysv.S: Don't use this on x86-64.
2553 * src/x86/ffi.c: Don't use this on x86-64.
2555 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2557 * src/ffitest.c (struct6): New test that tests a special case in
2559 (struct7): Likewise.
2560 (struct8): Likewise.
2561 (struct9): Likewise.
2562 (closure_test_fn): Silence warning about this when it's not used.
2563 (main): Add the new tests.
2564 (main): Fix a couple of wrong casts and silence some compiler warnings.
2565 * include/ffi.h.in: Add x86-64 ABI definition.
2566 * fficonfig.h.in: Regenerate.
2567 * Makefile.am: Add x86-64 support.
2568 * configure.in: Likewise.
2569 * Makefile.in: Regenerate.
2570 * configure: Likewise.
2572 2002-06-24 Bo Thorsen <bo@suse.de>
2574 * src/types.c: Merge settings for similar architectures.
2575 Add x86-64 sizes and alignments.
2577 2002-06-23 Bo Thorsen <bo@suse.de>
2579 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
2580 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
2581 * src/mips/ffi.c (ffi_prep_args): Likewise.
2582 * src/m68k/ffi.c (ffi_prep_args): Likewise.
2584 2002-07-18 H.J. Lu (hjl@gnu.org)
2586 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
2587 (libffi_la_SOURCES): Support MIPS_LINUX.
2588 (libffi_convenience_la_SOURCES): Likewise.
2589 * Makefile.in: Regenerated.
2591 * configure.in (mips64*-*): Skip.
2592 (mips*-*-linux*): New.
2593 * configure: Regenerated.
2595 * src/mips/ffi.c: Include <sgidefs.h>.
2597 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
2599 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
2601 2002-05-27 Roger Sayle <roger@eyesopen.com>
2603 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
2605 2002-05-27 Bo Thorsen <bo@suse.de>
2607 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
2610 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
2612 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
2613 beginning of function (for older apple cc).
2615 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
2617 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
2618 script entry, and set LD to it when configuring multilibs.
2619 * configure: Rebuilt.
2621 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
2623 * configure.in (sparc64-*-netbsd*): Add target.
2624 (sparc-*-netbsdelf*): Likewise.
2625 * configure: Regenerate.
2627 2002-04-28 David S. Miller <davem@redhat.com>
2629 * configure.in, configure: Fix SPARC test in previous change.
2631 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
2633 * Makefile.am: Add Linux for S/390 support.
2634 * Makefile.in: Regenerate.
2635 * configure.in: Add Linux for S/390 support.
2636 * configure: Regenerate.
2637 * include/ffi.h.in: Add Linux for S/390 support.
2638 * src/s390/ffi.c: New file from libffi CVS tree.
2639 * src/s390/sysv.S: New file from libffi CVS tree.
2641 2002-04-28 Jakub Jelinek <jakub@redhat.com>
2643 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
2645 * src/sparc/v8.S: Use it.
2646 * src/sparc/v9.S: Likewise.
2647 * fficonfig.h.in: Rebuilt.
2648 * configure: Rebuilt.
2650 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
2652 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
2654 * src/ia64/unix.S: Add unwind information. Fix comments.
2655 Save sp in a way that's compatible with unwind info.
2656 (ffi_call_unix): Correctly restore sp in all cases.
2657 * src/ia64/ffi.c: Add, fix comments.
2659 2002-04-08 Jakub Jelinek <jakub@redhat.com>
2661 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
2663 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
2665 * configure.in (alpha*-*-netbsd*): Add target.
2666 * configure: Regenerate.
2668 2002-04-04 Jeff Sturm <jsturm@one-point.com>
2670 * src/sparc/v8.S: Add unwind info.
2671 * src/sparc/v9.S: Likewise.
2673 2002-03-30 Krister Walfridsson <cato@df.lth.se>
2675 * configure.in: Enable i*86-*-netbsdelf*.
2676 * configure: Rebuilt.
2678 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
2681 * src/mips/n32.s: Delete
2682 * src/mips/o32.s: Delete
2684 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
2686 * configure.in: Enable alpha*-*-freebsd*.
2687 * configure: Rebuilt.
2689 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2691 * Makefile.am: libfficonvenience -> libffi_convenience.
2692 * Makefile.in: Rebuilt.
2694 * Makefile.am: Define ffitest_OBJECTS.
2695 * Makefile.in: Rebuilt.
2697 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
2698 David Edelsohn <edelsohn@gnu.org>
2700 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
2701 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
2702 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
2703 * Makefile.in: Regenerate.
2704 * include/ffi.h.in: Add AIX and Darwin closure definitions.
2705 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
2706 (flush_icache, flush_range): New functions.
2707 (ffi_closure_helper_DARWIN): New function.
2708 * src/powerpc/aix_closure.S: New file.
2709 * src/powerpc/darwin_closure.S: New file.
2711 2002-02-24 Jeff Sturm <jsturm@one-point.com>
2713 * include/ffi.h.in: Add typedef for ffi_arg.
2714 * src/ffitest.c (main): Declare rint with ffi_arg.
2716 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
2718 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
2719 number of GPRs for floating-point arguments.
2721 2002-01-31 Anthony Green <green@redhat.com>
2723 * configure: Rebuilt.
2724 * configure.in: Replace CHECK_SIZEOF and endian tests with
2725 cross-compiler friendly macros.
2726 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
2729 2002-01-18 David Edelsohn <edelsohn@gnu.org>
2731 * src/powerpc/darwin.S (_ffi_call_AIX): New.
2732 * src/powerpc/aix.S (ffi_call_DARWIN): New.
2734 2002-01-17 David Edelsohn <edelsohn@gnu.org>
2736 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
2737 (TARGET_SRC_POWERPC_AIX): New.
2738 (POWERPC_AIX): New stanza.
2739 * Makefile.in: Regenerate.
2740 * configure.in: Add AIX case.
2741 * configure: Regenerate.
2742 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
2743 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
2744 size. Fix "long double" support.
2745 (ffi_call): Add FFI_AIX case.
2746 * src/powerpc/aix.S: New.
2748 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
2750 Implement Darwin PowerPC ABI.
2751 * configure.in: Handle powerpc-*-darwin*.
2752 * Makefile.am: Set source files for POWERPC_DARWIN.
2753 * configure: Rebuilt.
2754 * Makefile.in: Rebuilt.
2755 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
2757 * src/powerpc/darwin.S: New file.
2758 * src/powerpc/ffi_darwin.c: New file.
2760 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
2762 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
2764 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2766 * src/x86/sysv.S: Avoid gas-only .balign directive.
2767 Use C style comments.
2769 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2771 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
2772 Fixes PR bootstrap/3563.
2774 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2776 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
2778 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2780 * configure.in: Recognize sparc*-sun-* host.
2781 * configure: Regenerate.
2783 2001-06-06 Andrew Haley <aph@redhat.com>
2785 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
2787 2001-06-03 Andrew Haley <aph@redhat.com>
2789 * src/alpha/osf.S: Add unwind info.
2790 * src/powerpc/sysv.S: Add unwind info.
2791 * src/powerpc/ppc_closure.S: Likewise.
2793 2000-05-31 Jeff Sturm <jsturm@one-point.com>
2795 * configure.in: Fix AC_ARG_ENABLE usage.
2796 * configure: Rebuilt.
2798 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2800 * configure.in: Remove warning about beta code.
2801 * configure: Rebuilt.
2803 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
2805 * src/ia64/unix.S: Restore stack pointer when returning from
2807 * src/ia64/ffi.c: Fix typo in comment.
2809 2001-04-18 Jim Wilson <wilson@redhat.com>
2811 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
2812 to eliminate RAW DV.
2814 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
2816 * Makefile.am: Make a libtool convenience library.
2817 * Makefile.in: Rebuilt.
2819 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
2821 * configure.in: Use different syntax for subdirectory creation.
2822 * configure: Rebuilt.
2824 2001-03-27 Jon Beniston <jon@beniston.com>
2826 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
2827 * configure: Rebuilt.
2828 * Makefile.am: Added X86_WIN32 target support.
2829 * Makefile.in: Rebuilt.
2831 * include/ffi.h.in: Added X86_WIN32 target support.
2833 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
2834 * src/types.c: Added X86_WIN32 target support.
2836 * src/x86/win32.S: New file. Based on sysv.S, but with EH
2837 stuff removed and made to work with CygWin's gas.
2839 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
2841 * configure.in: Make target subdirectory in build dir.
2842 * Makefile.am: Override suffix based rules to specify correct output
2844 * Makefile.in: Rebuilt.
2845 * configure: Rebuilt.
2847 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
2849 * src/powerpc/ppc_closure.S: New file.
2850 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
2851 involving long long and register pairs.
2852 (ffi_prep_closure): New function.
2853 (flush_icache): Likewise.
2854 (ffi_closure_helper_SYSV): Likewise.
2855 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
2856 (FFI_TRAMPOLINE_SIZE): Likewise.
2857 (FFI_NATIVE_RAW_API): Likewise.
2858 * Makefile.in: Rebuilt.
2859 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
2860 (TARGET_SRC_POWERPC): Likewise.
2862 2001-03-19 Tom Tromey <tromey@redhat.com>
2864 * Makefile.in: Rebuilt.
2865 * Makefile.am (ffitest_LDFLAGS): New macro.
2867 2001-03-02 Nick Clifton <nickc@redhat.com>
2869 * include/ffi.h.in: Remove RCS ident string.
2870 * include/ffi_mips.h: Remove RCS ident string.
2871 * src/debug.c: Remove RCS ident string.
2872 * src/ffitest.c: Remove RCS ident string.
2873 * src/prep_cif.c: Remove RCS ident string.
2874 * src/types.c: Remove RCS ident string.
2875 * src/alpha/ffi.c: Remove RCS ident string.
2876 * src/alpha/osf.S: Remove RCS ident string.
2877 * src/arm/ffi.c: Remove RCS ident string.
2878 * src/arm/sysv.S: Remove RCS ident string.
2879 * src/mips/ffi.c: Remove RCS ident string.
2880 * src/mips/n32.S: Remove RCS ident string.
2881 * src/mips/o32.S: Remove RCS ident string.
2882 * src/sparc/ffi.c: Remove RCS ident string.
2883 * src/sparc/v8.S: Remove RCS ident string.
2884 * src/sparc/v9.S: Remove RCS ident string.
2885 * src/x86/ffi.c: Remove RCS ident string.
2886 * src/x86/sysv.S: Remove RCS ident string.
2888 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
2890 * include/ffi.h.in: Change sourceware.cygnus.com references to
2893 2000-12-09 Richard Henderson <rth@redhat.com>
2895 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
2896 (ffi_closure_osf_inner): Index rather than increment avalue
2897 and arg_types. Give ffi_closure_osf the raw return value type.
2898 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
2901 2000-12-07 Richard Henderson <rth@redhat.com>
2903 * src/raw_api.c (ffi_translate_args): Fix typo.
2904 (ffi_prep_closure): Likewise.
2906 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
2907 FFI_TRAMPOLINE_SIZE.
2908 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
2909 cif->bytes for new ffi_call_osf implementation.
2910 (ffi_prep_args): Absorb into ...
2911 (ffi_call): ... here. Do all stack allocation here and
2912 avoid a callback function.
2913 (ffi_prep_closure, ffi_closure_osf_inner): New.
2914 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
2915 (ffi_closure_osf): New.
2917 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
2919 * config.guess, config.sub, install-sh: Removed.
2920 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
2921 * Makefile.in: Rebuilt.
2923 * acinclude.m4: Include libtool macros from the top level.
2924 * aclocal.m4, configure: Rebuilt.
2926 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
2928 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
2929 * configure: Rebuilt.
2931 2000-05-11 Scott Bambrough <scottb@netwinder.org>
2933 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
2934 memory correctly. Use conditional instructions, not branches where
2937 2000-05-04 Tom Tromey <tromey@cygnus.com>
2939 * configure: Rebuilt.
2940 * configure.in: Match `arm*-*-linux-*'.
2941 From Chris Dornan <cdornan@arm.com>.
2943 2000-04-28 Jakub Jelinek <jakub@redhat.com>
2945 * Makefile.am (SUBDIRS): Define.
2946 (AM_MAKEFLAGS): Likewise.
2947 (Multilib support.): Add section.
2948 * Makefile.in: Rebuilt.
2949 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
2950 New variables. Set for gcc using -print-multi-lib. Export them
2952 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
2953 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
2955 (extra_compiler_flags_value, extra_compiler_flags): Check these
2956 for extra compiler options which need to be passed down in
2959 2000-04-16 Anthony Green <green@redhat.com>
2961 * configure: Rebuilt.
2962 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
2964 2000-04-14 Jakub Jelinek <jakub@redhat.com>
2966 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
2967 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
2968 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
2969 Replace all void * sizeofs with sizeof(int).
2970 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
2971 different than DOUBLE.
2972 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
2973 (ffi_prep_args_v9): New function.
2974 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
2975 (ffi_V9_return_struct): New function.
2976 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
2977 32bit code (not yet cross-arch calls).
2978 * src/sparc/v8.S: Add struct return delay nop.
2980 * src/sparc/v9.S: New file.
2981 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
2982 is used on sparc64 only for structures larger than 32 bytes.
2983 Pass by reference for structures is done for structure arguments
2984 larger than 16 bytes.
2985 * src/ffitest.c (main): Use 64bit rint on sparc64.
2986 Run long long tests on sparc.
2987 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
2989 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
2991 * configure.in (sparc-*-linux*): New supported target.
2992 (sparc64-*-linux*): Likewise.
2993 * configure: Rebuilt.
2994 * Makefile.am: Add v9.S to SPARC files.
2995 * Makefile.in: Likewise.
2996 (LINK): Surround $(CCLD) into double quotes, so that multilib
2997 compiles work correctly.
2999 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3001 * configure: Rebuilt.
3002 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3003 proposed by Bryce McKinlay.
3005 2000-03-20 Tom Tromey <tromey@cygnus.com>
3007 * Makefile.in: Hand edit for java_raw_api.lo.
3009 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3011 * config.guess, config.sub: Update from the gcc tree.
3012 Fix for PR libgcj/168.
3014 2000-03-03 Tom Tromey <tromey@cygnus.com>
3016 * Makefile.in: Fixed ia64 by hand.
3018 * configure: Rebuilt.
3019 * configure.in (--enable-multilib): New option.
3020 (libffi_basedir): New subst.
3021 (AC_OUTPUT): Added multilib code.
3023 2000-03-02 Tom Tromey <tromey@cygnus.com>
3025 * Makefile.in: Rebuilt.
3026 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3029 2000-02-25 Hans Boehm <boehm@acm.org>
3031 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3033 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3035 (ffi_prep_raw_closure): Use ffi_translate_args, not
3036 ffi_closure_translate.
3037 * src/java_raw_api.c: New file.
3038 * src/ffitest.c (closure_test_fn): New function.
3039 (main): Define `rint' as long long on IA64. Added new test when
3040 FFI_CLOSURES is defined.
3041 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3042 (ffi_abi): Recognize IA64.
3043 (ffi_raw): Added `flt' field.
3044 Added "Java raw API" code.
3045 * configure.in: Recognize ia64.
3046 * Makefile.am (TARGET_SRC_IA64): New macro.
3047 (libffi_la_common_SOURCES): Added java_raw_api.c.
3048 (libffi_la_SOURCES): Define in IA64 case.
3050 2000-01-04 Tom Tromey <tromey@cygnus.com>
3052 * Makefile.in: Rebuilt with newer automake.
3054 1999-12-31 Tom Tromey <tromey@cygnus.com>
3056 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3058 1999-09-01 Tom Tromey <tromey@cygnus.com>
3060 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3062 * fficonfig.h.in: Rebuilt.
3063 * configure: Rebuilt.
3064 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3065 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3066 * acconfig.h: Don't #undef PACKAGE or VERSION.
3068 1999-08-09 Anthony Green <green@cygnus.com>
3070 * include/ffi.h.in: Try to work around messy header problem
3071 with PACKAGE and VERSION.
3073 * configure: Rebuilt.
3074 * configure.in: Change version to 2.00-beta.
3076 * fficonfig.h.in: Rebuilt.
3077 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3079 * src/x86/ffi.c (ffi_raw_call): Rename.
3081 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3083 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3084 (ffi_prep_incoming_args_SYSV): Ditto.
3085 (ffi_prep_closure): Ditto.
3086 (ffi_closure_raw_SYSV): Ditto.
3087 (ffi_prep_raw_closure): More ditto.
3088 (ffi_call_raw): Final ditto.
3090 * include/ffi.h.in: Add definitions for closure and raw API.
3092 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3095 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3097 * src/raw_api.c: New file.
3099 * include/ffi.h.in (ffi_raw): New type.
3100 (UINT_ARG, SINT_ARG): New defines.
3101 (ffi_closure, ffi_raw_closure): New types.
3102 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3104 * configure.in: Add check for endianness and sizeof void*.
3106 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3107 instead of directly.
3109 * configure: Rebuilt.
3111 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3113 * configure.in: Add x86 and powerpc BeOS configurations.
3114 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3116 1999-05-09 Anthony Green <green@cygnus.com>
3118 * configure.in: Add warning about this being beta code.
3119 Remove src/Makefile.am from the picture.
3120 * configure: Rebuilt.
3122 * Makefile.am: Move logic from src/Makefile.am. Add changes
3123 to support libffi as a target library.
3124 * Makefile.in: Rebuilt.
3126 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3127 Upgraded to new autoconf, automake, libtool.
3131 * LICENSE: Update copyright date.
3133 * src/Makefile.am, src/Makefile.in: Removed.
3135 1998-11-29 Anthony Green <green@cygnus.com>
3137 * include/ChangeLog: Removed.
3138 * src/ChangeLog: Removed.
3139 * src/mips/ChangeLog: Removed.
3140 * src/sparc/ChangeLog: Remboved.
3141 * src/x86/ChangeLog: Removed.
3143 * ChangeLog.v1: Created.