1 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
3 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
4 * src/powerpc/linux64.S: Likewise.
5 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
6 * testsuite/libffi.call/return_ldl.c: New test.
8 2007-09-04 <aph@redhat.com>
10 * src/arm/sysv.S (UNWIND): New.
11 (Whole file): Conditionally compile unwinder directives.
12 * src/arm/sysv.S: Add unwinder directives.
14 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
15 Only treat r0 as a struct address if we're actually returning a
17 Only copy the bytes that are actually within a struct.
18 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
19 is returned in r0, not passed by address.
20 (ffi_call): Allocate a word-sized temporary for the case where
21 a composite is returned in r0.
22 (ffi_prep_incoming_args_SYSV): Align as necessary.
24 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
26 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
27 directly using the sys_cacheflush syscall.
29 2007-07-27 Andrew Haley <aph@redhat.com>
31 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
33 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
35 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
36 * configure.ac: Likewise.
37 * Makefile.in: Regenerate.
38 * include/Makefile.in: Likewise.
39 * testsuite/Makefile.in: Likewise.
40 * configure: Likewise.
42 2007-08-24 David Daney <ddaney@avtrex.com>
44 * testsuite/libffi.call/return_sl.c: New test.
46 2007-08-10 David Daney <ddaney@avtrex.com>
48 * testsuite/libffi.call/cls_multi_ushort.c,
49 testsuite/libffi.call/cls_align_uint16.c,
50 testsuite/libffi.call/nested_struct1.c,
51 testsuite/libffi.call/nested_struct3.c,
52 testsuite/libffi.call/cls_7_1_byte.c,
53 testsuite/libffi.call/nested_struct5.c,
54 testsuite/libffi.call/cls_double.c,
55 testsuite/libffi.call/nested_struct7.c,
56 testsuite/libffi.call/cls_sint.c,
57 testsuite/libffi.call/nested_struct9.c,
58 testsuite/libffi.call/cls_20byte1.c,
59 testsuite/libffi.call/cls_multi_sshortchar.c,
60 testsuite/libffi.call/cls_align_sint64.c,
61 testsuite/libffi.call/cls_3byte2.c,
62 testsuite/libffi.call/cls_multi_schar.c,
63 testsuite/libffi.call/cls_multi_uchar.c,
64 testsuite/libffi.call/cls_19byte.c,
65 testsuite/libffi.call/cls_9byte1.c,
66 testsuite/libffi.call/cls_align_float.c,
67 testsuite/libffi.call/closure_fn1.c,
68 testsuite/libffi.call/problem1.c,
69 testsuite/libffi.call/closure_fn3.c,
70 testsuite/libffi.call/cls_sshort.c,
71 testsuite/libffi.call/closure_fn5.c,
72 testsuite/libffi.call/cls_align_double.c,
73 testsuite/libffi.call/nested_struct.c,
74 testsuite/libffi.call/cls_2byte.c,
75 testsuite/libffi.call/nested_struct10.c,
76 testsuite/libffi.call/cls_4byte.c,
77 testsuite/libffi.call/cls_6byte.c,
78 testsuite/libffi.call/cls_8byte.c,
79 testsuite/libffi.call/cls_multi_sshort.c,
80 testsuite/libffi.call/cls_align_sint16.c,
81 testsuite/libffi.call/cls_align_uint32.c,
82 testsuite/libffi.call/cls_20byte.c,
83 testsuite/libffi.call/cls_float.c,
84 testsuite/libffi.call/nested_struct2.c,
85 testsuite/libffi.call/cls_5_1_byte.c,
86 testsuite/libffi.call/nested_struct4.c,
87 testsuite/libffi.call/cls_24byte.c,
88 testsuite/libffi.call/nested_struct6.c,
89 testsuite/libffi.call/cls_64byte.c,
90 testsuite/libffi.call/nested_struct8.c,
91 testsuite/libffi.call/cls_uint.c,
92 testsuite/libffi.call/cls_multi_ushortchar.c,
93 testsuite/libffi.call/cls_schar.c,
94 testsuite/libffi.call/cls_uchar.c,
95 testsuite/libffi.call/cls_align_uint64.c,
96 testsuite/libffi.call/cls_ulonglong.c,
97 testsuite/libffi.call/cls_align_longdouble.c,
98 testsuite/libffi.call/cls_1_1byte.c,
99 testsuite/libffi.call/cls_12byte.c,
100 testsuite/libffi.call/cls_3_1byte.c,
101 testsuite/libffi.call/cls_3byte1.c,
102 testsuite/libffi.call/cls_4_1byte.c,
103 testsuite/libffi.call/cls_6_1_byte.c,
104 testsuite/libffi.call/cls_16byte.c,
105 testsuite/libffi.call/cls_18byte.c,
106 testsuite/libffi.call/closure_fn0.c,
107 testsuite/libffi.call/cls_9byte2.c,
108 testsuite/libffi.call/closure_fn2.c,
109 testsuite/libffi.call/closure_fn4.c,
110 testsuite/libffi.call/cls_ushort.c,
111 testsuite/libffi.call/closure_fn6.c,
112 testsuite/libffi.call/cls_5byte.c,
113 testsuite/libffi.call/cls_align_pointer.c,
114 testsuite/libffi.call/cls_7byte.c,
115 testsuite/libffi.call/cls_align_sint32.c,
116 testsuite/libffi.special/unwindtest_ffi_call.cc,
117 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
119 2007-08-10 David Daney <ddaney@avtrex.com>
122 * configure.ac: Don't treat mips64 as a special case.
123 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
124 * configure: Regenerate
125 * Makefile.in: Ditto.
126 * fficonfig.h.in: Ditto.
127 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
128 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
129 (FFI_DEFAULT_ABI): Set for n64 case.
130 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
131 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
132 (ffi_closure_N32): New function.
133 (.eh_frame): New section
134 * src/mips/o32.S: Clean up comments.
135 (ffi_closure_O32): Pass ffi_closure parameter in $12.
136 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
137 _MIPS_SIM == _ABIN32 throughout.
138 (FFI_MIPS_STOP_HERE): New, use in place of
140 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
141 to support n32/n64 ABIs.
142 (calc_n32_struct_flags): Rewrite.
143 (calc_n32_return_struct_flags): Remove unused variable. Reverse
144 position of flag bits.
145 (ffi_prep_cif_machdep): Rewrite n32 portion.
146 (ffi_call): Enable for n64. Add special handling for small structure
148 (ffi_prep_closure_loc): Add n32 and n64 support.
149 (ffi_closure_mips_inner_O32): Add cast to silence warning.
150 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
152 2007-08-08 David Daney <ddaney@avtrex.com>
154 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
155 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
157 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
158 * testsuite/libffi.call/cls_sint.c (main): Ditto.
159 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
160 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
161 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
162 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
163 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
164 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
165 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
166 * testsuite/libffi.call/return_fl3.c (main): Ditto.
167 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
168 * testsuite/libffi.call/nested_struct.c (main): Ditto.
169 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
170 * testsuite/libffi.call/return_ll1.c (main): Ditto.
171 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
172 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
173 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
174 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
175 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
176 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
177 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
178 * testsuite/libffi.call/cls_uint.c (main): Ditto.
179 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
180 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
181 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
182 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
183 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
184 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
185 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
186 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
187 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
189 2007-08-07 Andrew Haley <aph@redhat.com>
191 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
194 2007-08-06 Andrew Haley <aph@redhat.com>
197 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
198 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
201 2007-08-02 David Daney <ddaney@avtrex.com>
203 * testsuite/libffi.call/return_ul.c (main): Define return type as
204 ffi_arg. Use proper printf conversion specifier.
206 2007-07-30 Andrew Haley <aph@redhat.com>
209 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
210 signed/unsigned int8/16.
211 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
213 Remove code to pop args from the stack after call.
214 Special-case signed/unsigned int8/16.
215 * testsuite/libffi.call/return_sc.c (main): Revert.
217 2007-07-26 Richard Guenther <rguenther@suse.de>
220 * testsuite/libffi.call/return_sc.c (main): Verify call
221 result as signed char, not ffi_arg.
223 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
225 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
226 * configure: Regenerate.
228 2007-07-11 David Daney <ddaney@avtrex.com>
230 * src/mips/ffi.c: Don't include sys/cachectl.h.
231 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
234 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
236 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
237 from (ffi_prep_closure): ... this.
238 (FFI_INIT_TRAMPOLINE): Adjust.
240 2005-12-31 Phil Blundell <pb@reciva.com>
242 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
243 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
244 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
245 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
246 (FFI_CLOSURES): Enable closure support.
248 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
250 * testsuite/libffi.call/cls_multi_ushort.c,
251 testsuite/libffi.call/cls_align_uint16.c,
252 testsuite/libffi.call/nested_struct1.c,
253 testsuite/libffi.call/nested_struct3.c,
254 testsuite/libffi.call/cls_7_1_byte.c,
255 testsuite/libffi.call/cls_double.c,
256 testsuite/libffi.call/nested_struct5.c,
257 testsuite/libffi.call/nested_struct7.c,
258 testsuite/libffi.call/cls_sint.c,
259 testsuite/libffi.call/nested_struct9.c,
260 testsuite/libffi.call/cls_20byte1.c,
261 testsuite/libffi.call/cls_multi_sshortchar.c,
262 testsuite/libffi.call/cls_align_sint64.c,
263 testsuite/libffi.call/cls_3byte2.c,
264 testsuite/libffi.call/cls_multi_schar.c,
265 testsuite/libffi.call/cls_multi_uchar.c,
266 testsuite/libffi.call/cls_19byte.c,
267 testsuite/libffi.call/cls_9byte1.c,
268 testsuite/libffi.call/cls_align_float.c,
269 testsuite/libffi.call/closure_fn1.c,
270 testsuite/libffi.call/problem1.c,
271 testsuite/libffi.call/closure_fn3.c,
272 testsuite/libffi.call/cls_sshort.c,
273 testsuite/libffi.call/closure_fn5.c,
274 testsuite/libffi.call/cls_align_double.c,
275 testsuite/libffi.call/cls_2byte.c,
276 testsuite/libffi.call/nested_struct.c,
277 testsuite/libffi.call/nested_struct10.c,
278 testsuite/libffi.call/cls_4byte.c,
279 testsuite/libffi.call/cls_6byte.c,
280 testsuite/libffi.call/cls_8byte.c,
281 testsuite/libffi.call/cls_multi_sshort.c,
282 testsuite/libffi.call/cls_align_uint32.c,
283 testsuite/libffi.call/cls_align_sint16.c,
284 testsuite/libffi.call/cls_float.c,
285 testsuite/libffi.call/cls_20byte.c,
286 testsuite/libffi.call/cls_5_1_byte.c,
287 testsuite/libffi.call/nested_struct2.c,
288 testsuite/libffi.call/cls_24byte.c,
289 testsuite/libffi.call/nested_struct4.c,
290 testsuite/libffi.call/nested_struct6.c,
291 testsuite/libffi.call/cls_64byte.c,
292 testsuite/libffi.call/nested_struct8.c,
293 testsuite/libffi.call/cls_uint.c,
294 testsuite/libffi.call/cls_multi_ushortchar.c,
295 testsuite/libffi.call/cls_schar.c,
296 testsuite/libffi.call/cls_uchar.c,
297 testsuite/libffi.call/cls_align_uint64.c,
298 testsuite/libffi.call/cls_ulonglong.c,
299 testsuite/libffi.call/cls_align_longdouble.c,
300 testsuite/libffi.call/cls_1_1byte.c,
301 testsuite/libffi.call/cls_12byte.c,
302 testsuite/libffi.call/cls_3_1byte.c,
303 testsuite/libffi.call/cls_3byte1.c,
304 testsuite/libffi.call/cls_4_1byte.c,
305 testsuite/libffi.call/cls_6_1_byte.c,
306 testsuite/libffi.call/cls_16byte.c,
307 testsuite/libffi.call/cls_18byte.c,
308 testsuite/libffi.call/closure_fn0.c,
309 testsuite/libffi.call/cls_9byte2.c,
310 testsuite/libffi.call/closure_fn2.c,
311 testsuite/libffi.call/closure_fn4.c,
312 testsuite/libffi.call/cls_ushort.c,
313 testsuite/libffi.call/closure_fn6.c,
314 testsuite/libffi.call/cls_5byte.c,
315 testsuite/libffi.call/cls_align_pointer.c,
316 testsuite/libffi.call/cls_7byte.c,
317 testsuite/libffi.call/cls_align_sint32.c,
318 testsuite/libffi.special/unwindtest_ffi_call.cc,
319 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
321 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
323 * aclocal.m4: Regenerated.
325 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
327 * configure: Regenerate.
329 2007-05-23 Steve Ellcey <sje@cup.hp.com>
331 * Makefile.in: Regenerate.
332 * configure: Regenerate.
333 * aclocal.m4: Regenerate.
334 * include/Makefile.in: Regenerate.
335 * testsuite/Makefile.in: Regenerate.
337 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
339 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
340 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
341 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
342 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
343 (FFI_CLOSURES): Enable closure support.
345 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
347 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
348 * configure: Regenerate.
349 * fficonfig.h.in: Regenerate.
350 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
351 CFI_OFFSET,CFI_DEF_CFA): New macros.
352 (ffi_call_SYSV): Add callframe annotation.
354 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
356 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
357 numerous test suite failures.
358 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
360 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
362 * Makefile.am (EXTRA_DIST): Bring up to date.
363 * Makefile.in: Regenerate.
364 * src/frv/eabi.S: Remove RCS keyword.
366 2007-04-06 Richard Henderson <rth@redhat.com>
368 * configure.ac: Tidy target case.
369 (HAVE_LONG_DOUBLE): Allow the target to override.
370 * configure: Regenerate.
371 * include/ffi.h.in: Don't define ffi_type_foo if
372 LIBFFI_HIDE_BASIC_TYPES is defined.
373 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
375 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
376 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
377 (ffi_type_longdouble): Special case for Alpha. Don't define
378 if long double == double.
380 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
381 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
382 (ffi_call, ffi_closure_osf_inner): Likewise.
383 (ffi_closure_osf_inner): Likewise. Mark hidden.
384 (ffi_call_osf, ffi_closure_osf): Mark hidden.
385 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
386 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
387 (load_table): Handle 128-bit long double.
389 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
391 2007-04-06 Tom Tromey <tromey@redhat.com>
394 * README: Fixed bug in example.
396 2007-04-03 Jakub Jelinek <jakub@redhat.com>
398 * src/closures.c: Include sys/statfs.h.
399 (_GNU_SOURCE): Define on Linux.
400 (FFI_MMAP_EXEC_SELINUX): Define.
401 (selinux_enabled): New variable.
402 (selinux_enabled_check): New function.
403 (is_selinux_enabled): Define.
406 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
408 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
409 Use 'volatile float sum' to create sum of floats to avoid false
410 negative due to excess precision on ix86 targets.
413 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
415 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
417 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
419 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
421 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
422 (ffi_prep_closure_loc): New.
423 (ffi_prep_raw_closure_loc): New.
424 (ffi_prep_java_raw_closure_loc): New.
425 * src/closures.c: New file.
426 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
427 Replace sflags with exec_offset.
428 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
429 sub_segment_exec_offset): New macros.
430 (get_segment_flags, set_segment_flags, check_segment_merge): New
432 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
433 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
434 destroy_mspace): Use new macros.
435 (sys_alloc): Silence warning.
436 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
437 * Makefile.in: Rebuilt.
438 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
439 terms of ffi_prep_closure_loc.
440 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
442 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
444 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
446 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
448 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
449 (ffi_prep_closure): ... this.
450 * src/pa/ffi.c: Likewise.
451 * src/cris/ffi.c: Likewise. Adjust.
452 * src/frv/ffi.c: Likewise.
453 * src/ia64/ffi.c: Likewise.
454 * src/mips/ffi.c: Likewise.
455 * src/powerpc/ffi_darwin.c: Likewise.
456 * src/s390/ffi.c: Likewise.
457 * src/sh/ffi.c: Likewise.
458 * src/sh64/ffi.c: Likewise.
459 * src/sparc/ffi.c: Likewise.
460 * src/x86/ffi64.c: Likewise.
461 * src/x86/ffi.c: Likewise.
462 (FFI_INIT_TRAMPOLINE): Adjust.
463 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
464 (ffi_prep_raw_closure): ... this.
465 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
466 (ffi_prep_closure): ... this.
467 (flush_icache): Adjust.
469 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
471 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
474 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
476 * Makefile.am: Add dummy install-pdf target.
477 * Makefile.in: Regenerate
479 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
481 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
482 ffi_closure_helper_SYSV): Add long double handling.
484 2007-02-02 Jakub Jelinek <jakub@redhat.com>
486 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
487 immediately after bctrl instruction.
489 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
491 * Makefile.am (all-recursive, install-recursive,
492 mostlyclean-recursive, clean-recursive, distclean-recursive,
493 maintainer-clean-recursive): Add missing targets.
494 * Makefile.in: Rebuilt.
496 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
498 * configure.ac: Add TARGET for x86_64-*-darwin*.
499 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
501 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
502 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
503 * configure: Regenerate.
504 * Makefile.in: Regenerate.
505 * include/Makefile.in: Regenerate.
506 * testsuite/Makefile.in: Regenerate.
507 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
510 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
512 * aclocal.m4: Regenerate with aclocal -I .. as written in the
515 2006-10-31 Geoffrey Keating <geoffk@apple.com>
517 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
518 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
520 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
521 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
523 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
524 Sandro Tolaini <tolaini@libero.it>
526 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
528 * configure: Regenerated.
529 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
530 (EXTRA_DIST): Add src/x86/darwin.S.
531 * Makefile.in: Regenerated.
532 * include/Makefile.in: Regenerated.
533 * testsuite/Makefile.in: Regenerated.
535 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
536 X86_WIN32, and additionally align stack to 16 bytes.
537 * src/x86/darwin.S: New, based on sysv.S.
538 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
540 2006-09-12 David Daney <ddaney@avtrex.com>
543 * include/Makefile.am: Install both ffi.h and ffitarget.h in
544 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
545 * aclocal.m4: Regenerated for automake 1.9.6.
546 * Makefile.in: Regenerated.
547 * include/Makefile.in: Regenerated.
548 * testsuite/Makefile.in: Regenerated.
550 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
552 * include/ffi_common.h (struct): Revert accidental commit.
554 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
556 * include/ffi_common.h: Remove lint directives.
557 * include/ffi.h.in: Likewise.
559 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
561 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
562 for 32-bit architectures.
563 * testsuite/libffi.call/return_ul.c: New test case.
565 2006-07-19 David Daney <ddaney@avtrex.com>
567 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
568 xfail remains for mips64.
570 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
572 * Makefile.am: Add install-html target. Add install-html to .PHONY
573 * Makefile.in: Regenerate.
574 * aclocal.m4: Regenerate.
575 * include/Makefile.in: Regenerate.
576 * testsuite/Makefile.in: Regenerate.
578 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
580 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
583 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
585 * README: Remove notice about 'Crazy Comments'.
586 * src/debug.c: Remove lint directives. Cleanup white spaces.
587 * src/java_raw_api.c: Likewise.
588 * src/prep_cif.c: Likewise.
589 * src/raw_api.c: Likewise.
590 * src/ffitest.c: Delete. No longer needed, all test cases migrated
592 * src/arm/ffi.c: Remove lint directives.
593 * src/m32r/ffi.c: Likewise.
594 * src/pa/ffi.c: Likewise.
595 * src/powerpc/ffi.c: Likewise.
596 * src/powerpc/ffi_darwin.c: Likewise.
597 * src/sh/ffi.c: Likewise.
598 * src/sh64/ffi.c: Likewise.
599 * src/x86/ffi.c: Likewise.
600 * testsuite/libffi.call/float2.c: Likewise.
601 * testsuite/libffi.call/promotion.c: Likewise.
602 * testsuite/libffi.call/struct1.c: Likewise.
604 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
606 * src/pa/hpux32.S: Correct unwind offset calculation for
608 * src/pa/linux.S: Likewise.
610 2006-04-12 James E Wilson <wilson@specifix.com>
613 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
614 (hfa_type_load): Call stf_spill.
615 (hfa_type_store): Call ldf_fill.
616 (ffi_call): Adjust calls to above routines. Add local temps for
619 2006-04-10 Matthias Klose <doko@debian.org>
621 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
622 directory names containing underscores.
624 2006-04-07 James E Wilson <wilson@specifix.com>
626 * testsuite/libffi.call/float4.c: New testcase.
628 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
629 Andreas Tobler <a.tobler@schweiz.ch>
631 * Makefile.am: Add PA_HPUX port.
632 * Makefile.in: Regenerate.
633 * include/Makefile.in: Likewise.
634 * testsuite/Makefile.in: Likewise.
635 * configure.ac: Add PA_HPUX rules.
636 * configure: Regenerate.
637 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
638 Add PA_HPUX and PA64_HPUX.
639 Rename FFI_LINUX ABI to FFI_PA32 ABI.
640 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
641 (FFI_TYPE_SMALL_STRUCT2): Define.
642 (FFI_TYPE_SMALL_STRUCT4): Likewise.
643 (FFI_TYPE_SMALL_STRUCT8): Likewise.
644 (FFI_TYPE_SMALL_STRUCT3): Redefine.
645 (FFI_TYPE_SMALL_STRUCT5): Likewise.
646 (FFI_TYPE_SMALL_STRUCT6): Likewise.
647 (FFI_TYPE_SMALL_STRUCT7): Likewise.
648 * src/pa/ffi.c (ROUND_DOWN): Delete.
649 (fldw, fstw, fldd, fstd): Use '__asm__'.
650 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
651 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
652 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
653 Simplify incrementing of stack slot variable. Change type of local
655 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
657 (ffi_prep_cif_machdep): Likewise.
658 (ffi_call): Likewise.
659 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
660 return type to ffi_status. Simplify incrementing of stack slot
661 variable. Only copy floating point argument registers when PA_LINUX
662 is true. Reformat debug statement.
663 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
664 FFI_TYPE_SMALL_STRUCT8.
665 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
667 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
668 Add nops to cache flush. Add trampoline for PA_HPUX.
669 * src/pa/hpux32.S: New file.
670 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
671 ffi_prep_args_LINUX to ffi_prep_args_pa32.
672 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
673 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
674 argument type checks so that common argument types appear first.
675 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
676 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
678 2006-03-24 Alan Modra <amodra@bigpond.net.au>
680 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
681 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
682 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
684 (ffi_prep_args64): Assert using IBM extended double.
685 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
686 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
687 (ffi_call): Handle FFI_LINUX.
688 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
689 gpr3 return pointer as for struct return. Handle FFI_LINUX
690 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
692 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
693 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
694 Don't use r6 as pointer to results, instead use sp offset. Don't
695 make a special call to load lr with case table address, instead
696 use offset from previous call.
697 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
698 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
701 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
703 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
704 passed with FP registers correctly.
705 (ffi_closure_helper_SYSV): Likewise.
706 * src/sh64/sysv.S: Likewise.
708 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
710 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
711 args and userdata unused.
712 (closure_test_fn1): Mark cif and userdata unused.
713 (main): Remove unused res.
715 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
717 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
718 -O2, -O3, -Os and the warning flags -W -Wall.
719 * testsuite/libffi.special/special.exp: Likewise.
720 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
721 unused parameter unused for gcc or else do nothing.
722 * testsuite/libffi.special/ffitestcxx.h: Likewise.
723 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
725 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
726 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
727 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
728 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
729 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
730 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
731 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
732 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
733 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
734 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
735 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
736 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
737 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
738 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
739 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
740 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
741 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
742 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
743 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
744 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
745 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
746 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
747 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
748 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
750 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
752 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
754 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
755 void* to avoid compiler warning.
757 (cls_struct_align_gn): Mark cif and userdata unused.
758 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
760 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
762 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
764 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
766 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
768 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
769 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
770 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
772 (main): Cast res_call to silence gcc.
773 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
775 (main): Cast res_call to silence gcc.
776 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
778 (main): Cast res_call to silence gcc.
779 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
781 (main): Cast res_call to silence gcc.
782 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
784 (main): Cast res_call to silence gcc.
785 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
787 (main): Cast res_call to silence gcc.
788 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
790 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
791 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
793 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
794 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
796 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
797 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
799 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
800 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
802 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
803 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
805 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
807 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
808 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
809 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
810 Cleanup white spaces.
811 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
812 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
813 cif and userdata unused.
814 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
816 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
817 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
818 formatters to silence gcc.
819 (B_gn): Mark cif and userdata unused.
820 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
822 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
823 (B_gn): Mark cif and userdata unused.
824 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
826 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
827 (B_gn): Mark cif and userdata unused.
828 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
830 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
831 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
832 * testsuite/libffi.call/problem1.c (stub): Likewise.
833 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
835 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
836 in the last commit for this test case in the test case itself.
837 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
839 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
840 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
841 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
842 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
843 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
844 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
846 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
848 * src/sh/sysv.S: Fix register numbers in the FDE for
851 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
853 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
854 declaration to avoid a false negative on ix86. See PR323.
856 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
858 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
859 and cast integer to void * if needed. Update the pointer to
860 the FP register saved area correctly.
862 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
864 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
866 * testsuite/libffi.call/nested_struct4.c: Likewise.
868 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
870 * testsuite/libffi.call/return_dbl.c: New test case.
871 * testsuite/libffi.call/return_dbl1.c: Likewise.
872 * testsuite/libffi.call/return_dbl2.c: Likewise.
873 * testsuite/libffi.call/return_fl.c: Likewise.
874 * testsuite/libffi.call/return_fl1.c: Likewise.
875 * testsuite/libffi.call/return_fl2.c: Likewise.
876 * testsuite/libffi.call/return_fl3.c: Likewise.
877 * testsuite/libffi.call/closure_fn6.c: Likewise.
879 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
881 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
882 here to be used by other test cases too.
884 * testsuite/libffi.call/nested_struct10.c: New test case.
885 * testsuite/libffi.call/nested_struct9.c: Likewise.
886 * testsuite/libffi.call/nested_struct8.c: Likewise.
887 * testsuite/libffi.call/nested_struct7.c: Likewise.
888 * testsuite/libffi.call/nested_struct6.c: Likewise.
889 * testsuite/libffi.call/nested_struct5.c: Likewise.
890 * testsuite/libffi.call/nested_struct4.c: Likewise.
892 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
894 * configure.ac: Enable libffi for sparc64-*-freebsd*.
895 * configure: Rebuilt.
897 2006-01-18 Jakub Jelinek <jakub@redhat.com>
899 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
900 instead do the shifting inline.
901 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
902 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
903 and 8 byte structs, for the remaining struct sizes don't call
904 __lshrdi3, instead do the shifting inline.
906 2005-12-07 Thiemo Seufer <ths@networkno.de>
908 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
910 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
911 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
912 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
913 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
914 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
916 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
918 (ffi_prep_closure): Improve trampoline instruction scheduling.
919 (ffi_closure_mips_inner_O32): Fix endianness bugs.
921 2005-12-03 Alan Modra <amodra@bigpond.net.au>
923 * src/powerpc/ffi.c: Formatting.
924 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
925 (ffi_prep_args64): Likewise.
927 2005-09-30 Geoffrey Keating <geoffk@apple.com>
929 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
930 darwin, use -shared-libgcc not -lgcc_s, and explain why.
932 2005-09-26 Tom Tromey <tromey@redhat.com>
934 * testsuite/libffi.call/float1.c (value_type): New typedef.
935 (CANARY): New define.
936 (main): Check for result buffer overflow.
937 * src/powerpc/linux64.S: Handle linux64 long double returns.
938 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
939 (ffi_prep_cif_machdep): Handle linux64 long double returns.
941 2005-08-25 Alan Modra <amodra@bigpond.net.au>
944 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
946 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
948 2005-08-11 Jakub Jelinek <jakub@redhat.com>
950 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
951 (AH_BOTTOM): Add FFI_HIDDEN definition.
952 * configure: Rebuilt.
953 * fficonfig.h.in: Rebuilt.
954 * src/powerpc/ffi.c (hidden): Remove.
955 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
956 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
957 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
958 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
959 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
960 add FFI_HIDDEN to its prototype.
961 (ffi_closure_SYSV_inner): New.
962 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
963 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
965 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
968 * configure: Rebuilt.
969 * configure.ac: Handle i*86-*-gnu*.
971 2005-08-09 Jakub Jelinek <jakub@redhat.com>
973 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
974 DW_CFA_offset_extended_sf rather than
975 DW_CFA_GNU_negative_offset_extended.
976 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
978 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
980 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
982 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
983 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
984 partially on register.
985 (ffi_closure_helper_SYSV): Likewise.
986 (ffi_prep_cif_machdep): Don't set too many cif->flags.
988 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
990 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
992 * src/sh64/ffi.c (simple_type): Remove.
993 (return_type): Handle small structures correctly.
994 (ffi_prep_args): Likewise.
995 (ffi_call): Likewise.
996 (ffi_closure_helper_SYSV): Likewise.
997 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
998 Emit position independent code if PIC and remove wrong datalabel
999 prefixes from EH data.
1001 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1003 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1004 * Makefile.in: Regenerate.
1005 * include/Makefile.in: Likewise.
1006 * testsuite/Makefile.in: Likewise.
1007 * configure.ac: Add POWERPC_FREEBSD rules.
1008 * configure: Regenerate.
1009 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1010 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1011 * src/powerpc/ffi.c: Add flags to handle small structure returns
1013 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1015 (ffi_closure_helper_SYSV): Likewise.
1016 * src/powerpc/ppc_closure.S: Add return types for small structures.
1017 * src/powerpc/sysv.S: Add bits to handle small structures for
1020 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1022 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1023 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1024 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1026 2005-07-05 Randolph Chung <tausq@debian.org>
1028 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1029 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1030 structures. Kill compilation warnings.
1031 (ffi_closure_inner_LINUX): Print return values as hex in debug
1032 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1033 Properly handle 5-7 byte structure returns.
1034 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1035 (FFI_TYPE_SMALL_STRUCT2): Remove.
1036 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1037 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1038 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1039 (checksmst1, checksmst2): Remove.
1040 (checksmst3): Optimize handling of 3-byte struct returns.
1041 (checksmst567): Properly handle 5-7 byte struct returns.
1043 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1046 * src/mips/n32.S: Enforce PIC code.
1047 * src/mips/o32.S: Likewise.
1049 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1051 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1052 * configure: Regenerate.
1054 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1056 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1057 to call ffi_closure_helper_SYSV. Append @local instead.
1058 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1060 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1062 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1063 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1064 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1065 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1066 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1068 2005-05-09 Mike Stump <mrs@apple.com>
1070 * configure: Regenerate.
1072 2005-05-08 Richard Henderson <rth@redhat.com>
1075 * src/alpha/osf.S: Update unwind into to match code.
1077 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1078 Richard Henderson <rth@redhat.com>
1080 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1082 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1083 (ffi_prep_closure): Set carry bit if sse-used flag set.
1084 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1085 Only load sse registers if ssecount non-zero.
1086 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1088 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1090 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1091 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1092 * configure: Regenerate.
1094 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1096 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1097 have Tcl8.3-compatible intermediate variable.
1099 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1100 Hans-Peter Nilsson <hp@axis.com>
1102 * Makefile.am: Add CRIS support.
1103 * configure.ac: Likewise.
1104 * Makefile.in, configure, testsuite/Makefile.in,
1105 include/Makefile.in: Regenerate.
1106 * src/cris: New directory.
1107 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1108 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1110 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1111 \r?\n in output tests.
1113 2005-04-12 Mike Stump <mrs@apple.com>
1115 * configure: Regenerate.
1117 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1119 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1121 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1123 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1125 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1126 to 64 bits in ILP32 mode.
1127 Load 64 bits even for short data.
1129 2005-03-23 Mike Stump <mrs@apple.com>
1131 * src/powerpc/darwin.S: Update for -m64 multilib.
1132 * src/powerpc/darwin_closure.S: Likewise.
1134 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1136 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1137 Do not set tool_include_dir.
1138 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1140 * include/Makefile.am: Set gcc_version and toollibffidir.
1141 * include/Makefile.in: Regenerate.
1143 2005-02-22 Andrew Haley <aph@redhat.com>
1145 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1146 odd-numbered register pairs for 64-bit integer types.
1148 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1151 * testsuite/libffi.call/return_ll1.c: New test case.
1153 2005-02-11 Janis Johnson <janis187@us.ibm.com>
1155 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1156 * testsuite/libffi.call/float.c: Ditto.
1157 * testsuite/libffi.call/float2.c: Ditto.
1158 * testsuite/libffi.call/float3.c: Ditto.
1160 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1162 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1164 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1166 * testsuite/libffi.special/special.exp (cxx_options): Add
1169 2004-12-31 Richard Henderson <rth@redhat.com>
1171 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1172 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1173 offset parameters with a type parameter; deduce size and structure
1174 alignment. Update all users.
1176 2004-12-31 Richard Henderson <rth@redhat.com>
1178 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1179 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
1180 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
1181 into ffi_prep_closure.
1182 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1185 2004-12-27 Richard Henderson <rth@redhat.com>
1187 * src/x86/unix64.S: Fix typo in unwind info.
1189 2004-12-25 Richard Henderson <rth@redhat.com>
1191 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1192 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1193 (merge_classes): Check for it.
1195 (classify_argument): Pass byte_offset by value; perform all updates
1197 (examine_argument): Add classes argument; handle
1198 X86_64_COMPLEX_X87_CLASS.
1199 (ffi_prep_args): Merge into ...
1200 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1201 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1202 (ffi_fill_return_value): Remove.
1203 (ffi_prep_closure): Remove dead assert.
1204 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1205 Rewrite to use struct register_args instead of va_list. Create
1206 flags for handling structure returns.
1207 * src/x86/unix64.S: Remove dead strings.
1208 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1209 stack frame with ffi_call. Handle structure returns properly.
1210 (float2sse, floatfloat2sse, double2sse): Remove.
1211 (sse2float, sse2double, sse2floatfloat): Remove.
1212 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1213 to handle structure returns properly.
1215 2004-12-08 David Edelsohn <edelsohn@gnu.org>
1217 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1219 * Makefile.in: Regenerated.
1221 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
1223 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1224 * configure, aclocal.m4, Makefile.in: Regenerate.
1225 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1227 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1229 * configure: Regenerate for libtool change.
1231 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1233 * configure: Regenerate for libtool reversion.
1235 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1237 * configure: Regenerate for libtool change.
1239 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1241 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1243 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
1245 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1246 of jal. Use an absolute encoding for the frame information.
1248 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1250 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1251 * acinclude.m4: Delete logic for sincludes.
1252 * aclocal.m4, Makefile.in, configure: Regenerate.
1253 * include/Makefile: Likewise.
1254 * testsuite/Makefile: Likewise.
1256 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1258 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1259 on a 8-byte boundary.
1260 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1262 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1264 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1265 long long values. Round stack allocation to a multiple of 8 bytes
1266 for ATPCS compatibility.
1267 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1268 names. Handle returning long long types. Add Thumb and interworking
1269 support. Improve soft-float code.
1271 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1273 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1274 (libffi_exit): New function.
1275 (libffi_init): Build the testglue wrapper if needed.
1277 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1280 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1282 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1284 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1286 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1288 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1289 * testsuite/libffi.call/float3.c: New test case.
1291 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1293 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1294 the function returning a structure pointed with R2.
1295 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1296 the structure return value if T bit set. Emit position
1297 independent code and EH data if PIC.
1299 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1301 * Makefile.am: Add m32r support.
1302 * configure.ac: Likewise.
1303 * Makefile.in: Regenerate.
1304 * confiugre: Regenerate.
1305 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1306 (uint64, sint64, double, longdouble)
1307 * src/m32r: New directory.
1308 * src/m32r/ffi.c: New file.
1309 * src/m32r/sysv.S: Likewise.
1310 * src/m32r/ffitarget.h: Likewise.
1312 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1314 * testsuite/libffi.call/negint.c: New test case.
1316 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1319 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1320 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1321 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1324 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1326 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1327 * testsuite/libffi.call/promotion.c: Likewise.
1328 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1330 * testsuite/libffi.call/return_sc.c: Likewise.
1331 * testsuite/libffi.call/return_uc.c: Likewise.
1333 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1335 * src/powerpc/darwin.S: Fix comments and identation.
1336 * src/powerpc/darwin_closure.S: Likewise.
1338 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1340 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1341 (ffi_prep_args): Handle longdouble arguments.
1342 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1344 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1345 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1347 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1348 * src/types.c: Defined longdouble size and alignment for darwin.
1350 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1352 * src/powerpc/aix.S: Remove whitespaces.
1353 * src/powerpc/aix_closure.S: Likewise.
1354 * src/powerpc/asm.h: Likewise.
1355 * src/powerpc/ffi.c: Likewise.
1356 * src/powerpc/ffitarget.h: Likewise.
1357 * src/powerpc/linux64.S: Likewise.
1358 * src/powerpc/linux64_closure.S: Likewise.
1359 * src/powerpc/ppc_closure.S: Likewise.
1360 * src/powerpc/sysv.S: Likewise.
1362 2004-08-30 Anthony Green <green@redhat.com>
1364 * Makefile.am: Add frv support.
1365 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1366 * configure.ac: Read configure.host.
1367 * configure.in: Read configure.host.
1368 * configure.host: New file. frv-elf needs libgloss.
1369 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
1370 alignment. This is needed to frv and shouldn't harm the others.
1371 * include/ffi_common.h (ALIGN_DOWN): New macro.
1372 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1374 2004-08-24 David Daney <daney@avtrex.com>
1376 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1377 * testsuite/libffi.call/closure_fn1.c: Likewise.
1378 * testsuite/libffi.call/closure_fn2.c Likewise.
1379 * testsuite/libffi.call/closure_fn3.c: Likewise.
1380 * testsuite/libffi.call/closure_fn4.c: Likewise.
1381 * testsuite/libffi.call/closure_fn5.c: Likewise.
1382 * testsuite/libffi.call/cls_18byte.c: Likewise.
1383 * testsuite/libffi.call/cls_19byte.c: Likewise.
1384 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1385 * testsuite/libffi.call/cls_20byte.c: Likewise.
1386 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1387 * testsuite/libffi.call/cls_24byte.c: Likewise.
1388 * testsuite/libffi.call/cls_2byte.c: Likewise.
1389 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1390 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1391 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1392 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1393 * testsuite/libffi.call/cls_4byte.c: Likewise.
1394 * testsuite/libffi.call/cls_64byte.c: Likewise.
1395 * testsuite/libffi.call/cls_6byte.c: Likewise.
1396 * testsuite/libffi.call/cls_7byte.c: Likewise.
1397 * testsuite/libffi.call/cls_8byte.c: Likewise.
1398 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1399 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1400 * testsuite/libffi.call/cls_align_double.c: Likewise.
1401 * testsuite/libffi.call/cls_align_float.c: Likewise.
1402 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1403 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1404 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1405 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1406 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1407 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1408 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1409 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1410 * testsuite/libffi.call/cls_double.c: Likewise.
1411 * testsuite/libffi.call/cls_float.c: Likewise.
1412 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1413 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1414 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1415 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1416 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1417 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1418 * testsuite/libffi.call/cls_schar.c: Likewise.
1419 * testsuite/libffi.call/cls_sint.c: Likewise.
1420 * testsuite/libffi.call/cls_sshort.c: Likewise.
1421 * testsuite/libffi.call/cls_uchar.c: Likewise.
1422 * testsuite/libffi.call/cls_uint.c: Likewise.
1423 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1424 * testsuite/libffi.call/cls_ushort.c: Likewise.
1425 * testsuite/libffi.call/nested_struct.c: Likewise.
1426 * testsuite/libffi.call/nested_struct1.c: Likewise.
1427 * testsuite/libffi.call/nested_struct2.c: Likewise.
1428 * testsuite/libffi.call/nested_struct3.c: Likewise.
1429 * testsuite/libffi.call/problem1.c: Likewise.
1430 * testsuite/libffi.special/unwindtest.cc: Likewise.
1431 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1433 * testsuite/libffi.call/cls_16byte.c: Likewise.
1434 * testsuite/libffi.call/cls_5byte.c: Likewise.
1436 2004-08-23 David Daney <daney@avtrex.com>
1439 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1440 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1441 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1442 parameters and return types.
1443 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1444 (ffi_prep_closure): Ditto.
1445 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1446 alignment calculations.
1447 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1448 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1450 2004-08-14 Casey Marshall <csm@gnu.org>
1452 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1453 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1454 integer (O32 ABI only).
1455 (ffi_prep_closure): new function.
1456 (ffi_closure_mips_inner_O32): new function.
1457 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
1458 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1459 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1460 64 bit integers correctly.
1461 (ffi_closure_O32): new function.
1462 Added DWARF-2 unwind info for both functions.
1464 2004-08-10 Andrew Haley <aph@redhat.com>
1466 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1468 2004-08-01 Robert Millan <robertmh@gnu.org>
1470 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1471 * configure: Regenerate.
1473 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1475 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1476 and mmap() explicitly instead of relying on preset autoconf cache
1478 * aclocal.m4: Regenerate.
1479 * configure: Regenerate.
1481 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1483 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1484 (ffi_check_float_struct): Remove unused prototype.
1486 2004-06-30 Geoffrey Keating <geoffk@apple.com>
1488 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1489 character on Darwin, use '\n\t' instead.
1491 2004-06-26 Matthias Klose <doko@debian.org>
1493 * libtool-version: Fix typo in revision/age.
1495 2004-06-17 Matthias Klose <doko@debian.org>
1497 * libtool-version: New.
1498 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1499 * Makefile.in: Regenerate.
1501 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1503 * Makefile.am: Remove useless multilib rules.
1504 * Makefile.in: Regenerate.
1505 * aclocal.m4: Regenerate with automake 1.8.5.
1506 * configure.ac: Remove useless multilib configury.
1507 * configure: Regenerate.
1509 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1511 * .cvsignore: New file.
1513 2004-06-10 Jakub Jelinek <jakub@redhat.com>
1515 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1517 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1520 2004-06-06 Sean McNeil <sean@mcneil.com>
1522 * configure.ac: Add x86_64-*-freebsd* support.
1523 * configure: Regenerate.
1525 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
1528 * configure.ac: Test for existence of mmap and sys/mman.h before
1529 checking blacklist. Fix suggested by Jim Wilson.
1530 * configure: Regenerate.
1532 2004-04-26 Matt Austern <austern@apple.com>
1534 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1536 * src/powerpc/darwin_closure.S: Likewise.
1538 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1540 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1541 error. Reported by Thomas Heller <theller@python.net>.
1542 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1543 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1545 2004-03-20 Matthias Klose <doko@debian.org>
1547 * src/pa/linux.S: Fix typo.
1549 2004-03-19 Matthias Klose <doko@debian.org>
1551 * Makefile.am: Update.
1552 * Makefile.in: Regenerate.
1553 * src/pa/ffi.h.in: Remove.
1554 * src/pa/ffitarget.h: New file.
1556 2004-02-10 Randolph Chung <tausq@debian.org>
1558 * Makefile.am: Add PA support.
1559 * Makefile.in: Regenerate.
1560 * include/Makefile.in: Regenerate.
1561 * configure.ac: Add PA target.
1562 * configure: Regenerate.
1563 * src/pa/ffi.c: New file.
1564 * src/pa/ffi.h.in: Add PA support.
1565 * src/pa/linux.S: New file.
1566 * prep_cif.c: Add PA support.
1568 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
1570 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
1572 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
1573 with ecif->cif->flags.
1574 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
1576 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
1577 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
1578 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
1579 sc_retstruct2b): Add for 1 or 2-bytes struct case.
1581 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
1583 * configure.in: Rename file to ...
1584 * configure.ac: ... this.
1585 * fficonfig.h.in: Regenerate.
1586 * Makefile.in: Regenerate.
1587 * include/Makefile.in: Regenerate.
1588 * testsuite/Makefile.in: Regenerate.
1590 2004-03-12 Matt Austern <austern@apple.com>
1592 * src/powerpc/darwin.S: Fix EH information so it corresponds to
1593 changes in EH format resulting from addition of linkonce support.
1594 * src/powerpc/darwin_closure.S: Likewise.
1596 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
1597 Paolo Bonzini <bonzini@gnu.org>
1599 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
1600 Remove VPATH. Remove rules for object files. Remove multilib support.
1601 (AM_CCASFLAGS): Add.
1602 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
1603 (AC_PREREQ): Bump version to 2.59.
1604 (AC_INIT): Fill with version info and bug address.
1605 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
1606 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
1607 De-precious CC so that the right flags are passed down to multilibs.
1608 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
1609 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
1610 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
1611 * configure: Rebuilt.
1612 * aclocal.m4: Likewise.
1613 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
1614 * fficonfig.h.in: Likewise.
1616 2004-03-11 Andreas Schwab <schwab@suse.de>
1618 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
1619 arguments from fp registers only for the first 8 parameter slots.
1620 Don't convert a float parameter when passed in memory.
1622 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
1624 * configure: Regenerate for config/accross.m4 correction.
1626 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
1628 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
1629 ecif->cif->bytes to bytes.
1630 (ffi_prep_cif_machdep): Add braces around nested if statement.
1632 2004-02-09 Alan Modra <amodra@bigpond.net.au>
1634 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
1636 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
1637 (ffi_closure_helper_LINUX64): Fix typo.
1638 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
1640 * testsuite/libffi.call/float.c: Likewise.
1641 * testsuite/libffi.call/float2.c: Likewise.
1643 2004-02-08 Alan Modra <amodra@bigpond.net.au>
1645 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
1646 long double function return and long double arg handling.
1647 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
1648 Use "end_pfr" instead of "nf". Correct long double handling.
1650 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
1652 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
1653 space for long double return value. Adjust stack frame and offsets.
1654 Load f2 long double return.
1656 2004-02-07 Alan Modra <amodra@bigpond.net.au>
1658 * src/types.c: Use 16 byte long double for POWERPC64.
1660 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1662 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
1663 when the structure return address is passed in %o0.
1664 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
1665 (ffi_v9_layout_struct): Align the field following a nested structure
1666 on a word boundary. Use memmove instead of memcpy.
1667 (ffi_call): Update call to ffi_V9_return_struct.
1668 (ffi_prep_closure): Define 'ctx' only for V8.
1669 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
1670 and ffi_closure_sparc_inner_v9.
1671 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
1672 Always skip the structure return address. For structures and long
1673 doubles, copy the argument directly.
1674 (ffi_closure_sparc_inner_v9): Skip the structure return address only
1675 if required. Shift the maximum floating-point slot accordingly. For
1676 big structures, copy the argument directly; otherwise, left-justify the
1677 argument and call ffi_v9_layout_struct to lay out the structure on
1679 * src/sparc/v8.S: Undef STACKFRAME before defining it.
1680 (ffi_closure_v8): Pass the structure return address. Update call to
1681 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
1682 Skip the 'unimp' insn when returning long doubles and structures.
1683 * src/sparc/v9.S: Undef STACKFRAME before defining it.
1684 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
1685 FFI_TYPE_INT handling. Load structures both in integers and
1686 floating-point registers on return.
1687 * README: Update status of the SPARC port.
1689 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
1691 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
1693 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
1695 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
1697 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
1698 value as of type ffi_arg, not unsigned int.
1700 2004-01-21 Michael Ritzert <ritzert@t-online.de>
1702 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
1705 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
1707 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
1710 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1712 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
1715 2003-12-10 Richard Henderson <rth@redhat.com>
1717 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
1718 size_t instead of int.
1720 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
1722 * testsuite/libffi.call/many_win32.c: Include <float.h>.
1723 * testsuite/libffi.call/many_win32.c (main): Replace variable
1724 int i with unsigned long ul.
1726 * testsuite/libffi.call/cls_align_uint64.c: New test case.
1727 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1728 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1729 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1730 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1731 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1732 * testsuite/libffi.call/cls_align_float.c: Likewise.
1733 * testsuite/libffi.call/cls_align_double.c: Likewise.
1734 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1735 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1737 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
1740 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
1741 Align arguments to 32 bits.
1743 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
1746 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
1747 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1748 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1749 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1750 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1751 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1753 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
1755 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
1757 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
1758 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1760 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
1762 * Makefile.in: Rebuilt.
1763 * configure: Likewise.
1764 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
1767 2003-11-21 Andreas Jaeger <aj@suse.de>
1768 Andreas Tobler <a.tobler@schweiz.ch>
1770 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
1771 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
1772 * Makefile.in: Rebuilt.
1773 * aclocal.m4: Likewise.
1774 * configure: Likewise.
1775 * fficonfig.h.in: Likewise.
1776 * testsuite/lib/libffi-dg.exp: Add include dir.
1777 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
1778 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1779 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
1780 for ffi_closure if available.
1781 * testsuite/libffi.call/closure_fn1.c: Likewise.
1782 * testsuite/libffi.call/closure_fn2.c: Likewise.
1783 * testsuite/libffi.call/closure_fn3.c: Likewise.
1784 * testsuite/libffi.call/closure_fn4.c: Likewise.
1785 * testsuite/libffi.call/closure_fn5.c: Likewise.
1786 * testsuite/libffi.call/cls_12byte.c: Likewise.
1787 * testsuite/libffi.call/cls_16byte.c: Likewise.
1788 * testsuite/libffi.call/cls_18byte.c: Likewise.
1789 * testsuite/libffi.call/cls_19byte.c: Likewise.
1790 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1791 * testsuite/libffi.call/cls_20byte.c: Likewise.
1792 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1793 * testsuite/libffi.call/cls_24byte.c: Likewise.
1794 * testsuite/libffi.call/cls_2byte.c: Likewise.
1795 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1796 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1797 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1798 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1799 * testsuite/libffi.call/cls_4byte.c: Likewise.
1800 * testsuite/libffi.call/cls_5byte.c: Likewise.
1801 * testsuite/libffi.call/cls_64byte.c: Likewise.
1802 * testsuite/libffi.call/cls_6byte.c: Likewise.
1803 * testsuite/libffi.call/cls_7byte.c: Likewise.
1804 * testsuite/libffi.call/cls_8byte.c: Likewise.
1805 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1806 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1807 * testsuite/libffi.call/cls_double.c: Likewise.
1808 * testsuite/libffi.call/cls_float.c: Likewise.
1809 * testsuite/libffi.call/cls_schar.c: Likewise.
1810 * testsuite/libffi.call/cls_sint.c: Likewise.
1811 * testsuite/libffi.call/cls_sshort.c: Likewise.
1812 * testsuite/libffi.call/cls_uchar.c: Likewise.
1813 * testsuite/libffi.call/cls_uint.c: Likewise.
1814 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1815 * testsuite/libffi.call/cls_ushort.c: Likewise.
1816 * testsuite/libffi.call/nested_struct.c: Likewise.
1817 * testsuite/libffi.call/nested_struct1.c: Likewise.
1818 * testsuite/libffi.call/nested_struct2.c: Likewise.
1819 * testsuite/libffi.call/nested_struct3.c: Likewise.
1820 * testsuite/libffi.call/problem1.c: Likewise.
1821 * testsuite/libffi.special/unwindtest.cc: Likewise.
1823 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
1825 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
1827 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
1829 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
1830 Add -lgcc_s to additional flags.
1832 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
1834 * configure.in, include/Makefile.am: PR libgcj/11147, install
1835 the ffitarget.h header file in a gcc versioned and target
1837 * configure: Regenerated.
1838 * Makefile.in, include/Makefile.in: Likewise.
1839 * testsuite/Makefile.in: Likewise.
1841 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
1843 * testsuite/libffi.call/closure_fn0.c: Print result and check
1844 with dg-output to make debugging easier.
1845 * testsuite/libffi.call/closure_fn1.c: Likewise.
1846 * testsuite/libffi.call/closure_fn2.c: Likewise.
1847 * testsuite/libffi.call/closure_fn3.c: Likewise.
1848 * testsuite/libffi.call/closure_fn4.c: Likewise.
1849 * testsuite/libffi.call/closure_fn5.c: Likewise.
1850 * testsuite/libffi.call/cls_12byte.c: Likewise.
1851 * testsuite/libffi.call/cls_16byte.c: Likewise.
1852 * testsuite/libffi.call/cls_18byte.c: Likewise.
1853 * testsuite/libffi.call/cls_19byte.c: Likewise.
1854 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1855 * testsuite/libffi.call/cls_20byte.c: Likewise.
1856 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1857 * testsuite/libffi.call/cls_24byte.c: Likewise.
1858 * testsuite/libffi.call/cls_2byte.c: Likewise.
1859 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1860 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1861 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1862 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1863 * testsuite/libffi.call/cls_4byte.c: Likewise.
1864 * testsuite/libffi.call/cls_5byte.c: Likewise.
1865 * testsuite/libffi.call/cls_64byte.c: Likewise.
1866 * testsuite/libffi.call/cls_6byte.c: Likewise.
1867 * testsuite/libffi.call/cls_7byte.c: Likewise.
1868 * testsuite/libffi.call/cls_8byte.c: Likewise.
1869 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1870 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1871 * testsuite/libffi.call/cls_double.c: Likewise.
1872 * testsuite/libffi.call/cls_float.c: Likewise.
1873 * testsuite/libffi.call/cls_schar.c: Likewise.
1874 * testsuite/libffi.call/cls_sint.c: Likewise.
1875 * testsuite/libffi.call/cls_sshort.c: Likewise.
1876 * testsuite/libffi.call/cls_uchar.c: Likewise.
1877 * testsuite/libffi.call/cls_uint.c: Likewise.
1878 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1879 * testsuite/libffi.call/cls_ushort.c: Likewise.
1880 * testsuite/libffi.call/problem1.c: Likewise.
1882 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
1885 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
1887 * testsuite/libffi.call/cls_9byte2.c: New test case.
1888 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1889 * testsuite/libffi.call/cls_64byte.c: Likewise.
1890 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1891 * testsuite/libffi.call/cls_19byte.c: Likewise.
1892 * testsuite/libffi.call/cls_18byte.c: Likewise.
1893 * testsuite/libffi.call/closure_fn4.c: Likewise.
1894 * testsuite/libffi.call/closure_fn5.c: Likewise.
1895 * testsuite/libffi.call/cls_schar.c: Likewise.
1896 * testsuite/libffi.call/cls_sint.c: Likewise.
1897 * testsuite/libffi.call/cls_sshort.c: Likewise.
1898 * testsuite/libffi.call/nested_struct2.c: Likewise.
1899 * testsuite/libffi.call/nested_struct3.c: Likewise.
1901 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
1903 * testsuite/libffi.call/cls_double.c: Do a check on the result.
1904 * testsuite/libffi.call/cls_uchar.c: Likewise.
1905 * testsuite/libffi.call/cls_uint.c: Likewise.
1906 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1907 * testsuite/libffi.call/cls_ushort.c: Likewise.
1908 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
1910 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
1912 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
1915 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
1917 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
1918 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
1920 2003-10-22 David Daney <ddaney@avtrex.com>
1922 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
1923 __attribute__((__mode__(__SI__))) and friends.
1925 2003-10-22 Andreas Schwab <schwab@suse.de>
1927 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
1929 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
1931 * configure.in: AC_LINK_FILES(ffitarget.h).
1932 * configure: Regenerate.
1933 * Makefile.in: Likewise.
1934 * include/Makefile.in: Likewise.
1935 * testsuite/Makefile.in: Likewise.
1936 * fficonfig.h.in: Likewise.
1938 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
1939 Richard Henderson <rth@redhat.com>
1941 Avoid that ffi.h includes fficonfig.h.
1943 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
1944 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
1945 (TARGET_SRC_MIPS_SGI): Removed.
1946 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
1947 (MIPS_SGI): Removed.
1948 (CLEANFILES): Removed.
1949 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
1951 * acconfig.h: Removed.
1952 * configure.in: Compute sizeofs only for double and long double.
1953 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
1954 into AC_DEFINE instead of using acconfig.h. Create
1955 include/ffitarget.h instead of include/fficonfig.h. Rename
1956 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
1957 AC_DEFINE EH_FRAME_FLAGS.
1958 * include/Makefile.am (DISTCLEANFILES): New automake macro.
1959 (hack_DATA): Add ffitarget.h.
1960 * include/ffi.h.in: Remove all system specific definitions.
1961 Declare raw API even if it is not installed, why bother?
1962 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
1963 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
1964 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
1965 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
1966 * include/ffi_common.h (bool): Do not define.
1967 (ffi_assert): Accept failed assertion.
1968 (ffi_type_test): Return void and accept file/line.
1969 (FFI_ASSERT): Pass stringized failed assertion.
1970 (FFI_ASSERT_AT): New macro.
1971 (FFI_ASSERT_VALID_TYPE): New macro.
1972 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
1973 UINT64, SINT64): Define here with gcc's __attribute__ macro
1975 (FLOAT32, ALIGN): Define here instead of in ffi.h
1976 * include/ffi-mips.h: Removed. Its content moved to
1977 src/mips/ffitarget.h after separating assembly and C sections.
1978 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
1979 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
1980 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
1981 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
1982 SIZEOF_ARG -> FFI_SIZEOF_ARG.
1983 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
1984 * src/debug.c (ffi_assert): Accept stringized failed assertion.
1985 (ffi_type_test): Rewritten.
1986 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
1987 FFI_ASSERT_VALID_TYPE.
1988 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
1989 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
1990 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
1991 src/s390/ffitarget.h, src/sh/ffitarget.h,
1992 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
1993 src/x86/ffitarget.h: New files.
1994 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
1995 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
1996 src/powerpc/aix.S, src/powerpc/darwin.S,
1997 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
1998 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
1999 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2000 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2001 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2004 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2006 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2007 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2009 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2011 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2012 Used when FFI_DEBUG = 1.
2014 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2016 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2019 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2021 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2022 FFI_MIPS_O32 for O32 ABI.
2024 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2026 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2027 SPARC64. Cleanup whitespaces.
2029 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2031 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2032 strongarm, xscale. Cleanup whitespaces.
2033 * testsuite/libffi.call/closure_fn1.c: Likewise.
2034 * testsuite/libffi.call/closure_fn2.c: Likewise.
2035 * testsuite/libffi.call/closure_fn3.c: Likewise.
2036 * testsuite/libffi.call/cls_12byte.c: Likewise.
2037 * testsuite/libffi.call/cls_16byte.c: Likewise.
2038 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2039 * testsuite/libffi.call/cls_20byte.c: Likewise.
2040 * testsuite/libffi.call/cls_24byte.c: Likewise.
2041 * testsuite/libffi.call/cls_2byte.c: Likewise.
2042 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2043 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2044 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2045 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2046 * testsuite/libffi.call/cls_4byte.c: Likewise.
2047 * testsuite/libffi.call/cls_5byte.c: Likewise.
2048 * testsuite/libffi.call/cls_6byte.c: Likewise.
2049 * testsuite/libffi.call/cls_7byte.c: Likewise.
2050 * testsuite/libffi.call/cls_8byte.c: Likewise.
2051 * testsuite/libffi.call/cls_double.c: Likewise.
2052 * testsuite/libffi.call/cls_float.c: Likewise.
2053 * testsuite/libffi.call/cls_uchar.c: Likewise.
2054 * testsuite/libffi.call/cls_uint.c: Likewise.
2055 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2056 * testsuite/libffi.call/cls_ushort.c: Likewise.
2057 * testsuite/libffi.call/nested_struct.c: Likewise.
2058 * testsuite/libffi.call/nested_struct1.c: Likewise.
2059 * testsuite/libffi.call/problem1.c: Likewise.
2060 * testsuite/libffi.special/unwindtest.cc: Likewise.
2061 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2063 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2065 * src/powerpc/aix.S: Cleanup whitespaces.
2066 * src/powerpc/aix_closure.S: Likewise.
2068 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2070 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2071 * src/powerpc/darwin_closure.S: Likewise.
2072 * src/powerpc/ffi_darwin.c: Likewise.
2074 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2075 David Edelsohn <edelsohn@gnu.org>
2077 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2078 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2080 * src/powerpc/darwin_closure.S: Likewise.
2081 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2082 according to the Darwin/AIX ABI.
2083 (ffi_prep_cif_machdep): Likewise.
2084 (ffi_closure_helper_DARWIN): Likewise.
2085 Remove the outgoing parameter stack logic. Simplify the evaluation
2086 of the different CASE types.
2087 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2088 statement in the trampoline code.
2090 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2092 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2093 for the register size.
2094 (ffi_closure_helper_SYSV): Handle the structure return value
2096 (ffi_closure_helper_SYSV): Return the appropriate type when
2097 the registers are used for the structure return value.
2098 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2099 the 64-bit return value. Update copyright years.
2101 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2103 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2104 srcdir for ffi_mips.h.
2106 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2108 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2110 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2111 placement of float result.
2112 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2113 cast of "resp" for big-endian 64 bit machines.
2115 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2117 * src/types.c (double, longdouble): Merge identical SH and ARM
2118 typedefs, and add POWERPC64.
2119 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2120 struct split over gpr and rest.
2121 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2122 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2124 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2126 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2129 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2131 * configure: Regenerate.
2133 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2135 * Makefile.am: Remove build rules for ffitest.
2136 * Makefile.in: Rebuilt.
2138 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2140 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2141 about implicit declaration of abort().
2143 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2145 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2146 * Makefile.in: Rebuilt.
2147 * configure.in: Add dejagnu test framework.
2148 * configure: Rebuilt.
2150 * testsuite/Makefile.am: New file.
2151 * testsuite/Makefile.in: Built
2152 * testsuite/lib/libffi-dg.exp: New file.
2153 * testsuite/config/default.exp: Likewise.
2154 * testsuite/libffi.call/call.exp: Likewise.
2155 * testsuite/libffi.call/ffitest.h: Likewise.
2156 * testsuite/libffi.call/closure_fn0.c: Likewise.
2157 * testsuite/libffi.call/closure_fn1.c: Likewise.
2158 * testsuite/libffi.call/closure_fn2.c: Likewise.
2159 * testsuite/libffi.call/closure_fn3.c: Likewise.
2160 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2161 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2162 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2163 * testsuite/libffi.call/cls_2byte.c: Likewise.
2164 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2165 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2166 * testsuite/libffi.call/cls_4byte.c: Likewise.
2167 * testsuite/libffi.call/cls_5byte.c: Likewise.
2168 * testsuite/libffi.call/cls_6byte.c: Likewise.
2169 * testsuite/libffi.call/cls_7byte.c: Likewise.
2170 * testsuite/libffi.call/cls_8byte.c: Likewise.
2171 * testsuite/libffi.call/cls_12byte.c: Likewise.
2172 * testsuite/libffi.call/cls_16byte.c: Likewise.
2173 * testsuite/libffi.call/cls_20byte.c: Likewise.
2174 * testsuite/libffi.call/cls_24byte.c: Likewise.
2175 * testsuite/libffi.call/cls_double.c: Likewise.
2176 * testsuite/libffi.call/cls_float.c: Likewise.
2177 * testsuite/libffi.call/cls_uchar.c: Likewise.
2178 * testsuite/libffi.call/cls_uint.c: Likewise.
2179 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2180 * testsuite/libffi.call/cls_ushort.c: Likewise.
2181 * testsuite/libffi.call/float.c: Likewise.
2182 * testsuite/libffi.call/float1.c: Likewise.
2183 * testsuite/libffi.call/float2.c: Likewise.
2184 * testsuite/libffi.call/many.c: Likewise.
2185 * testsuite/libffi.call/many_win32.c: Likewise.
2186 * testsuite/libffi.call/nested_struct.c: Likewise.
2187 * testsuite/libffi.call/nested_struct1.c: Likewise.
2188 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2189 * testsuite/libffi.call/problem1.c: Likewise.
2190 * testsuite/libffi.call/promotion.c: Likewise.
2191 * testsuite/libffi.call/return_ll.c: Likewise.
2192 * testsuite/libffi.call/return_sc.c: Likewise.
2193 * testsuite/libffi.call/return_uc.c: Likewise.
2194 * testsuite/libffi.call/strlen.c: Likewise.
2195 * testsuite/libffi.call/strlen_win32.c: Likewise.
2196 * testsuite/libffi.call/struct1.c: Likewise.
2197 * testsuite/libffi.call/struct2.c: Likewise.
2198 * testsuite/libffi.call/struct3.c: Likewise.
2199 * testsuite/libffi.call/struct4.c: Likewise.
2200 * testsuite/libffi.call/struct5.c: Likewise.
2201 * testsuite/libffi.call/struct6.c: Likewise.
2202 * testsuite/libffi.call/struct7.c: Likewise.
2203 * testsuite/libffi.call/struct8.c: Likewise.
2204 * testsuite/libffi.call/struct9.c: Likewise.
2205 * testsuite/libffi.special/special.exp: New file.
2206 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2207 * testsuite/libffi.special/unwindtest.cc: Likewise.
2210 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2212 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2215 2003-08-02 Alan Modra <amodra@bigpond.net.au>
2217 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2219 (ffi_closure_helper_LINUX64): Likewise.
2220 * src/powerpc/linux64.S: Remove code writing to parm save area.
2221 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2222 address in lr from ffi_closure_helper_LINUX64 call to calculate
2223 table address. Optimize function tail.
2225 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2227 * src/sparc/ffi.c: Handle all floating point registers.
2228 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2230 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2232 * README: Note that libffi is not part of GCC. Update the project
2235 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2237 * src/powerpc/ppc_closure.S: Include ffi.h.
2239 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2241 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2242 Use C style comments.
2244 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2246 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2247 * Makefile.in: Regenerate.
2248 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2249 * configure: Regenerate.
2250 * include/ffi.h.in: Add SHmedia support.
2251 * src/sh64/ffi.c: New file.
2252 * src/sh64/sysv.S: New file.
2254 2003-05-16 Jakub Jelinek <jakub@redhat.com>
2256 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2257 should be read-only.
2258 * configure: Rebuilt.
2259 * fficonfig.h.in: Rebuilt.
2260 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2261 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2262 * src/powerpc/linux64.S: Likewise.
2263 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2264 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2265 if -fpic/-fPIC/-mrelocatable.
2266 * src/powerpc/powerpc_closure.S: Likewise.
2267 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2268 #write in .eh_frame flags.
2269 * src/sparc/v9.S: Likewise.
2270 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2271 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2272 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2274 2003-05-07 Jeff Sturm <jsturm@one-point.com>
2276 Fixes PR bootstrap/10656
2277 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2278 support for .register pseudo-op.
2279 * src/sparc/v8.S: Use it.
2280 * fficonfig.h.in: Rebuilt.
2281 * configure: Rebuilt.
2283 2003-04-18 Jakub Jelinek <jakub@redhat.com>
2285 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2286 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2287 Make it the default on POWERPC64.
2288 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2289 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2290 * configure: Rebuilt.
2291 * src/powerpc/ffi.c (hidden): Define.
2292 (ffi_prep_args_SYSV): Renamed from
2293 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2294 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2295 ASM_NEEDS_REGISTERS64): New.
2296 (ffi_prep_args64): New function.
2297 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2298 (ffi_call): Likewise.
2299 (ffi_prep_closure): Likewise.
2300 (flush_icache): Surround by #ifndef POWERPC64.
2301 (ffi_dblfl): New union type.
2302 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2303 (ffi_closure_helper_LINUX64): New function.
2304 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2306 * src/powerpc/sysv.S: Likewise.
2307 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2308 * src/powerpc/linux64.S: New file.
2309 * src/powerpc/linux64_closure.S: New file.
2310 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2311 src/powerpc/linux64_closure.S.
2312 (TARGET_SRC_POWERPC): Likewise.
2314 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2315 closure_test_fn3): Fix result printing on big-endian 64-bit
2317 (main): Print tst2_arg instead of uninitialized tst2_result.
2319 * src/ffitest.c (main): Hide what closure pointer really points to
2322 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
2324 * configure.in (arm-*-netbsdelf*): Add configuration.
2325 (configure): Regenerated.
2327 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
2329 * include/Makefile.in: Regenerate.
2331 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2333 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2335 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2336 Receive closure pointer through parameter, read args using
2337 __builtin_dwarf_cfa.
2338 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2340 2003-03-12 Andreas Schwab <schwab@suse.de>
2342 * configure.in: Avoid trailing /. in toolexeclibdir.
2343 * configure: Rebuilt.
2345 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2347 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
2349 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2351 * libffi/src/powerpc/darwin_closure.S:
2352 Fix alignement bug, allocate 8 bytes for the result.
2353 * libffi/src/powerpc/aix_closure.S:
2355 * libffi/src/powerpc/ffi_darwin.c:
2356 Update stackframe description for aix/darwin_closure.S.
2358 2003-02-06 Jakub Jelinek <jakub@redhat.com>
2360 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2363 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2364 Andreas Schwab <schwab@suse.de>
2366 * configure.in: Adjust command to source config-ml.in to account
2367 for changes to the libffi_basedir definition.
2368 (libffi_basedir): Remove ${srcdir} from value and include trailing
2371 * configure: Regenerate.
2373 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2375 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2377 2003-01-28 Andrew Haley <aph@redhat.com>
2379 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2380 * src/x86/ffi64.c (ffi_prep_closure): New.
2381 (ffi_closure_UNIX64_inner): New.
2382 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2384 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
2386 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2387 Remove USE_LIBDIR conditional.
2388 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2389 * Makefile.in, configure: Rebuilt.
2391 2003-01027 David Edelsohn <edelsohn@gnu.org>
2393 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2394 * Makefile.in: Regenerate.
2396 2003-01-22 Andrew Haley <aph@redhat.com>
2398 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2401 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2403 * src/powerpc/darwin.S: Add unwind info.
2404 * src/powerpc/darwin_closure.S: Likewise.
2406 2003-01-14 Andrew Haley <aph@redhat.com>
2408 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2409 (ffi_prep_cif_machdep): Likewise.
2410 * src/x86/unix64.S: Add unwind info.
2412 2003-01-14 Andreas Jaeger <aj@suse.de>
2414 * src/ffitest.c (main): Only use ffi_closures if those are
2417 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2419 * libffi/src/ffitest.c
2420 add closure testcases
2422 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2424 * libffi/src/powerpc/ffi.c
2425 fix alignment bug for float (4 byte aligned iso 8 byte)
2427 2003-01-09 Geoffrey Keating <geoffk@apple.com>
2429 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2430 * src/powerpc/darwin.S: Remove RCS version string.
2432 2003-01-03 Jeff Sturm <jsturm@one-point.com>
2434 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2435 * src/ffitest.c (main): Use static storage for closure.
2436 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2437 * src/sparc/v8.S (ffi_closure_v8): New.
2438 * src/sparc/v9.S (ffi_closure_v9): New.
2440 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2442 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2443 enumeration for X86_WIN32.
2444 * src/x86/win32.S: Added ffi_call_STDCALL function
2446 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2447 switch cases for recognising FFI_STDCALL and
2448 calling ffi_call_STDCALL if target is X86_WIN32.
2449 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2450 stdcall versions of the "my_strlen" and "many"
2451 test functions (for X86_WIN32).
2452 Added test cases to test stdcall invocation using
2455 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2457 * src/sh/sysv.S: Add DWARF2 unwind info.
2459 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2461 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2463 2002-11-26 Jim Wilson <wilson@redhat.com>
2465 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2467 2002-11-23 H.J. Lu <hjl@gnu.org>
2469 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2470 Include ../config/accross.m4.
2471 * aclocal.m4; Rebuild.
2472 * configure: Likewise.
2474 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2476 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2478 2002-11-11 DJ Delorie <dj@redhat.com>
2480 * configure.in: Look for common files in the right place.
2482 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2484 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2485 raw data as _Jv_word values, not ffi_raw.
2486 (ffi_java_ptrarray_to_raw): Likewise.
2487 (ffi_java_rvalue_to_raw): New function.
2488 (ffi_java_raw_call): Call it.
2489 (ffi_java_raw_to_rvalue): New function.
2490 (ffi_java_translate_args): Call it.
2491 * src/ffitest.c (closure_test_fn): Interpret return value
2492 as ffi_arg, not int.
2493 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2494 FFI_TYPE_POINTER case.
2495 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2496 values extended to word size.
2498 2002-10-02 Andreas Jaeger <aj@suse.de>
2500 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2502 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
2504 * include/ffi.h.in: Fix i386 win32 compilation.
2506 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2508 * configure.in: Add s390x-*-linux-* target.
2509 * configure: Regenerate.
2510 * include/ffi.h.in: Define S390X for s390x targets.
2511 (FFI_CLOSURES): Define for s390/s390x.
2512 (FFI_TRAMPOLINE_SIZE): Likewise.
2513 (FFI_NATIVE_RAW_API): Likewise.
2514 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2515 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
2516 * src/s390/ffi.c: Major rework of existing code. Add support for
2517 s390x targets. Add closure support.
2518 * src/s390/sysv.S: Likewise.
2520 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
2522 * src/arm/sysv.S: Fix typo.
2524 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
2526 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2527 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2528 (ffi_call_SYSV): Handle soft-float.
2530 2002-09-27 Bo Thorsen <bo@suse.de>
2532 * include/ffi.h.in: Fix multilib x86-64 support.
2534 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2536 * Makefile.am (all-multi): Fix multilib parallel build.
2538 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2540 * configure.in (sh[34]*-*-linux*): Add brackets.
2541 * configure: Regenerate.
2543 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2545 * Makefile.am: Add SH support.
2546 * Makefile.in: Regenerate.
2547 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2548 * configure: Regenerate.
2549 * include/ffi.h.in: Add SH support.
2550 * src/sh/ffi.c: New file.
2551 * src/sh/sysv.S: New file.
2552 * src/types.c: Add SH support.
2554 2002-07-16 Bo Thorsen <bo@suse.de>
2556 * src/x86/ffi64.c: New file that adds x86-64 support.
2557 * src/x86/unix64.S: New file that handles argument setup for
2559 * src/x86/sysv.S: Don't use this on x86-64.
2560 * src/x86/ffi.c: Don't use this on x86-64.
2562 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2564 * src/ffitest.c (struct6): New test that tests a special case in
2566 (struct7): Likewise.
2567 (struct8): Likewise.
2568 (struct9): Likewise.
2569 (closure_test_fn): Silence warning about this when it's not used.
2570 (main): Add the new tests.
2571 (main): Fix a couple of wrong casts and silence some compiler warnings.
2572 * include/ffi.h.in: Add x86-64 ABI definition.
2573 * fficonfig.h.in: Regenerate.
2574 * Makefile.am: Add x86-64 support.
2575 * configure.in: Likewise.
2576 * Makefile.in: Regenerate.
2577 * configure: Likewise.
2579 2002-06-24 Bo Thorsen <bo@suse.de>
2581 * src/types.c: Merge settings for similar architectures.
2582 Add x86-64 sizes and alignments.
2584 2002-06-23 Bo Thorsen <bo@suse.de>
2586 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
2587 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
2588 * src/mips/ffi.c (ffi_prep_args): Likewise.
2589 * src/m68k/ffi.c (ffi_prep_args): Likewise.
2591 2002-07-18 H.J. Lu (hjl@gnu.org)
2593 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
2594 (libffi_la_SOURCES): Support MIPS_LINUX.
2595 (libffi_convenience_la_SOURCES): Likewise.
2596 * Makefile.in: Regenerated.
2598 * configure.in (mips64*-*): Skip.
2599 (mips*-*-linux*): New.
2600 * configure: Regenerated.
2602 * src/mips/ffi.c: Include <sgidefs.h>.
2604 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
2606 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
2608 2002-05-27 Roger Sayle <roger@eyesopen.com>
2610 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
2612 2002-05-27 Bo Thorsen <bo@suse.de>
2614 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
2617 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
2619 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
2620 beginning of function (for older apple cc).
2622 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
2624 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
2625 script entry, and set LD to it when configuring multilibs.
2626 * configure: Rebuilt.
2628 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
2630 * configure.in (sparc64-*-netbsd*): Add target.
2631 (sparc-*-netbsdelf*): Likewise.
2632 * configure: Regenerate.
2634 2002-04-28 David S. Miller <davem@redhat.com>
2636 * configure.in, configure: Fix SPARC test in previous change.
2638 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
2640 * Makefile.am: Add Linux for S/390 support.
2641 * Makefile.in: Regenerate.
2642 * configure.in: Add Linux for S/390 support.
2643 * configure: Regenerate.
2644 * include/ffi.h.in: Add Linux for S/390 support.
2645 * src/s390/ffi.c: New file from libffi CVS tree.
2646 * src/s390/sysv.S: New file from libffi CVS tree.
2648 2002-04-28 Jakub Jelinek <jakub@redhat.com>
2650 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
2652 * src/sparc/v8.S: Use it.
2653 * src/sparc/v9.S: Likewise.
2654 * fficonfig.h.in: Rebuilt.
2655 * configure: Rebuilt.
2657 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
2659 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
2661 * src/ia64/unix.S: Add unwind information. Fix comments.
2662 Save sp in a way that's compatible with unwind info.
2663 (ffi_call_unix): Correctly restore sp in all cases.
2664 * src/ia64/ffi.c: Add, fix comments.
2666 2002-04-08 Jakub Jelinek <jakub@redhat.com>
2668 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
2670 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
2672 * configure.in (alpha*-*-netbsd*): Add target.
2673 * configure: Regenerate.
2675 2002-04-04 Jeff Sturm <jsturm@one-point.com>
2677 * src/sparc/v8.S: Add unwind info.
2678 * src/sparc/v9.S: Likewise.
2680 2002-03-30 Krister Walfridsson <cato@df.lth.se>
2682 * configure.in: Enable i*86-*-netbsdelf*.
2683 * configure: Rebuilt.
2685 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
2688 * src/mips/n32.s: Delete
2689 * src/mips/o32.s: Delete
2691 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
2693 * configure.in: Enable alpha*-*-freebsd*.
2694 * configure: Rebuilt.
2696 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2698 * Makefile.am: libfficonvenience -> libffi_convenience.
2699 * Makefile.in: Rebuilt.
2701 * Makefile.am: Define ffitest_OBJECTS.
2702 * Makefile.in: Rebuilt.
2704 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
2705 David Edelsohn <edelsohn@gnu.org>
2707 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
2708 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
2709 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
2710 * Makefile.in: Regenerate.
2711 * include/ffi.h.in: Add AIX and Darwin closure definitions.
2712 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
2713 (flush_icache, flush_range): New functions.
2714 (ffi_closure_helper_DARWIN): New function.
2715 * src/powerpc/aix_closure.S: New file.
2716 * src/powerpc/darwin_closure.S: New file.
2718 2002-02-24 Jeff Sturm <jsturm@one-point.com>
2720 * include/ffi.h.in: Add typedef for ffi_arg.
2721 * src/ffitest.c (main): Declare rint with ffi_arg.
2723 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
2725 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
2726 number of GPRs for floating-point arguments.
2728 2002-01-31 Anthony Green <green@redhat.com>
2730 * configure: Rebuilt.
2731 * configure.in: Replace CHECK_SIZEOF and endian tests with
2732 cross-compiler friendly macros.
2733 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
2736 2002-01-18 David Edelsohn <edelsohn@gnu.org>
2738 * src/powerpc/darwin.S (_ffi_call_AIX): New.
2739 * src/powerpc/aix.S (ffi_call_DARWIN): New.
2741 2002-01-17 David Edelsohn <edelsohn@gnu.org>
2743 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
2744 (TARGET_SRC_POWERPC_AIX): New.
2745 (POWERPC_AIX): New stanza.
2746 * Makefile.in: Regenerate.
2747 * configure.in: Add AIX case.
2748 * configure: Regenerate.
2749 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
2750 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
2751 size. Fix "long double" support.
2752 (ffi_call): Add FFI_AIX case.
2753 * src/powerpc/aix.S: New.
2755 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
2757 Implement Darwin PowerPC ABI.
2758 * configure.in: Handle powerpc-*-darwin*.
2759 * Makefile.am: Set source files for POWERPC_DARWIN.
2760 * configure: Rebuilt.
2761 * Makefile.in: Rebuilt.
2762 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
2764 * src/powerpc/darwin.S: New file.
2765 * src/powerpc/ffi_darwin.c: New file.
2767 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
2769 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
2771 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2773 * src/x86/sysv.S: Avoid gas-only .balign directive.
2774 Use C style comments.
2776 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2778 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
2779 Fixes PR bootstrap/3563.
2781 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2783 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
2785 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2787 * configure.in: Recognize sparc*-sun-* host.
2788 * configure: Regenerate.
2790 2001-06-06 Andrew Haley <aph@redhat.com>
2792 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
2794 2001-06-03 Andrew Haley <aph@redhat.com>
2796 * src/alpha/osf.S: Add unwind info.
2797 * src/powerpc/sysv.S: Add unwind info.
2798 * src/powerpc/ppc_closure.S: Likewise.
2800 2000-05-31 Jeff Sturm <jsturm@one-point.com>
2802 * configure.in: Fix AC_ARG_ENABLE usage.
2803 * configure: Rebuilt.
2805 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2807 * configure.in: Remove warning about beta code.
2808 * configure: Rebuilt.
2810 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
2812 * src/ia64/unix.S: Restore stack pointer when returning from
2814 * src/ia64/ffi.c: Fix typo in comment.
2816 2001-04-18 Jim Wilson <wilson@redhat.com>
2818 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
2819 to eliminate RAW DV.
2821 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
2823 * Makefile.am: Make a libtool convenience library.
2824 * Makefile.in: Rebuilt.
2826 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
2828 * configure.in: Use different syntax for subdirectory creation.
2829 * configure: Rebuilt.
2831 2001-03-27 Jon Beniston <jon@beniston.com>
2833 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
2834 * configure: Rebuilt.
2835 * Makefile.am: Added X86_WIN32 target support.
2836 * Makefile.in: Rebuilt.
2838 * include/ffi.h.in: Added X86_WIN32 target support.
2840 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
2841 * src/types.c: Added X86_WIN32 target support.
2843 * src/x86/win32.S: New file. Based on sysv.S, but with EH
2844 stuff removed and made to work with CygWin's gas.
2846 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
2848 * configure.in: Make target subdirectory in build dir.
2849 * Makefile.am: Override suffix based rules to specify correct output
2851 * Makefile.in: Rebuilt.
2852 * configure: Rebuilt.
2854 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
2856 * src/powerpc/ppc_closure.S: New file.
2857 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
2858 involving long long and register pairs.
2859 (ffi_prep_closure): New function.
2860 (flush_icache): Likewise.
2861 (ffi_closure_helper_SYSV): Likewise.
2862 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
2863 (FFI_TRAMPOLINE_SIZE): Likewise.
2864 (FFI_NATIVE_RAW_API): Likewise.
2865 * Makefile.in: Rebuilt.
2866 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
2867 (TARGET_SRC_POWERPC): Likewise.
2869 2001-03-19 Tom Tromey <tromey@redhat.com>
2871 * Makefile.in: Rebuilt.
2872 * Makefile.am (ffitest_LDFLAGS): New macro.
2874 2001-03-02 Nick Clifton <nickc@redhat.com>
2876 * include/ffi.h.in: Remove RCS ident string.
2877 * include/ffi_mips.h: Remove RCS ident string.
2878 * src/debug.c: Remove RCS ident string.
2879 * src/ffitest.c: Remove RCS ident string.
2880 * src/prep_cif.c: Remove RCS ident string.
2881 * src/types.c: Remove RCS ident string.
2882 * src/alpha/ffi.c: Remove RCS ident string.
2883 * src/alpha/osf.S: Remove RCS ident string.
2884 * src/arm/ffi.c: Remove RCS ident string.
2885 * src/arm/sysv.S: Remove RCS ident string.
2886 * src/mips/ffi.c: Remove RCS ident string.
2887 * src/mips/n32.S: Remove RCS ident string.
2888 * src/mips/o32.S: Remove RCS ident string.
2889 * src/sparc/ffi.c: Remove RCS ident string.
2890 * src/sparc/v8.S: Remove RCS ident string.
2891 * src/sparc/v9.S: Remove RCS ident string.
2892 * src/x86/ffi.c: Remove RCS ident string.
2893 * src/x86/sysv.S: Remove RCS ident string.
2895 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
2897 * include/ffi.h.in: Change sourceware.cygnus.com references to
2900 2000-12-09 Richard Henderson <rth@redhat.com>
2902 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
2903 (ffi_closure_osf_inner): Index rather than increment avalue
2904 and arg_types. Give ffi_closure_osf the raw return value type.
2905 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
2908 2000-12-07 Richard Henderson <rth@redhat.com>
2910 * src/raw_api.c (ffi_translate_args): Fix typo.
2911 (ffi_prep_closure): Likewise.
2913 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
2914 FFI_TRAMPOLINE_SIZE.
2915 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
2916 cif->bytes for new ffi_call_osf implementation.
2917 (ffi_prep_args): Absorb into ...
2918 (ffi_call): ... here. Do all stack allocation here and
2919 avoid a callback function.
2920 (ffi_prep_closure, ffi_closure_osf_inner): New.
2921 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
2922 (ffi_closure_osf): New.
2924 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
2926 * config.guess, config.sub, install-sh: Removed.
2927 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
2928 * Makefile.in: Rebuilt.
2930 * acinclude.m4: Include libtool macros from the top level.
2931 * aclocal.m4, configure: Rebuilt.
2933 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
2935 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
2936 * configure: Rebuilt.
2938 2000-05-11 Scott Bambrough <scottb@netwinder.org>
2940 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
2941 memory correctly. Use conditional instructions, not branches where
2944 2000-05-04 Tom Tromey <tromey@cygnus.com>
2946 * configure: Rebuilt.
2947 * configure.in: Match `arm*-*-linux-*'.
2948 From Chris Dornan <cdornan@arm.com>.
2950 2000-04-28 Jakub Jelinek <jakub@redhat.com>
2952 * Makefile.am (SUBDIRS): Define.
2953 (AM_MAKEFLAGS): Likewise.
2954 (Multilib support.): Add section.
2955 * Makefile.in: Rebuilt.
2956 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
2957 New variables. Set for gcc using -print-multi-lib. Export them
2959 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
2960 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
2962 (extra_compiler_flags_value, extra_compiler_flags): Check these
2963 for extra compiler options which need to be passed down in
2966 2000-04-16 Anthony Green <green@redhat.com>
2968 * configure: Rebuilt.
2969 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
2971 2000-04-14 Jakub Jelinek <jakub@redhat.com>
2973 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
2974 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
2975 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
2976 Replace all void * sizeofs with sizeof(int).
2977 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
2978 different than DOUBLE.
2979 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
2980 (ffi_prep_args_v9): New function.
2981 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
2982 (ffi_V9_return_struct): New function.
2983 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
2984 32bit code (not yet cross-arch calls).
2985 * src/sparc/v8.S: Add struct return delay nop.
2987 * src/sparc/v9.S: New file.
2988 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
2989 is used on sparc64 only for structures larger than 32 bytes.
2990 Pass by reference for structures is done for structure arguments
2991 larger than 16 bytes.
2992 * src/ffitest.c (main): Use 64bit rint on sparc64.
2993 Run long long tests on sparc.
2994 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
2996 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
2998 * configure.in (sparc-*-linux*): New supported target.
2999 (sparc64-*-linux*): Likewise.
3000 * configure: Rebuilt.
3001 * Makefile.am: Add v9.S to SPARC files.
3002 * Makefile.in: Likewise.
3003 (LINK): Surround $(CCLD) into double quotes, so that multilib
3004 compiles work correctly.
3006 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3008 * configure: Rebuilt.
3009 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3010 proposed by Bryce McKinlay.
3012 2000-03-20 Tom Tromey <tromey@cygnus.com>
3014 * Makefile.in: Hand edit for java_raw_api.lo.
3016 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3018 * config.guess, config.sub: Update from the gcc tree.
3019 Fix for PR libgcj/168.
3021 2000-03-03 Tom Tromey <tromey@cygnus.com>
3023 * Makefile.in: Fixed ia64 by hand.
3025 * configure: Rebuilt.
3026 * configure.in (--enable-multilib): New option.
3027 (libffi_basedir): New subst.
3028 (AC_OUTPUT): Added multilib code.
3030 2000-03-02 Tom Tromey <tromey@cygnus.com>
3032 * Makefile.in: Rebuilt.
3033 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3036 2000-02-25 Hans Boehm <boehm@acm.org>
3038 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3040 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3042 (ffi_prep_raw_closure): Use ffi_translate_args, not
3043 ffi_closure_translate.
3044 * src/java_raw_api.c: New file.
3045 * src/ffitest.c (closure_test_fn): New function.
3046 (main): Define `rint' as long long on IA64. Added new test when
3047 FFI_CLOSURES is defined.
3048 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3049 (ffi_abi): Recognize IA64.
3050 (ffi_raw): Added `flt' field.
3051 Added "Java raw API" code.
3052 * configure.in: Recognize ia64.
3053 * Makefile.am (TARGET_SRC_IA64): New macro.
3054 (libffi_la_common_SOURCES): Added java_raw_api.c.
3055 (libffi_la_SOURCES): Define in IA64 case.
3057 2000-01-04 Tom Tromey <tromey@cygnus.com>
3059 * Makefile.in: Rebuilt with newer automake.
3061 1999-12-31 Tom Tromey <tromey@cygnus.com>
3063 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3065 1999-09-01 Tom Tromey <tromey@cygnus.com>
3067 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3069 * fficonfig.h.in: Rebuilt.
3070 * configure: Rebuilt.
3071 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3072 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3073 * acconfig.h: Don't #undef PACKAGE or VERSION.
3075 1999-08-09 Anthony Green <green@cygnus.com>
3077 * include/ffi.h.in: Try to work around messy header problem
3078 with PACKAGE and VERSION.
3080 * configure: Rebuilt.
3081 * configure.in: Change version to 2.00-beta.
3083 * fficonfig.h.in: Rebuilt.
3084 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3086 * src/x86/ffi.c (ffi_raw_call): Rename.
3088 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3090 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3091 (ffi_prep_incoming_args_SYSV): Ditto.
3092 (ffi_prep_closure): Ditto.
3093 (ffi_closure_raw_SYSV): Ditto.
3094 (ffi_prep_raw_closure): More ditto.
3095 (ffi_call_raw): Final ditto.
3097 * include/ffi.h.in: Add definitions for closure and raw API.
3099 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3102 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3104 * src/raw_api.c: New file.
3106 * include/ffi.h.in (ffi_raw): New type.
3107 (UINT_ARG, SINT_ARG): New defines.
3108 (ffi_closure, ffi_raw_closure): New types.
3109 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3111 * configure.in: Add check for endianness and sizeof void*.
3113 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3114 instead of directly.
3116 * configure: Rebuilt.
3118 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3120 * configure.in: Add x86 and powerpc BeOS configurations.
3121 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3123 1999-05-09 Anthony Green <green@cygnus.com>
3125 * configure.in: Add warning about this being beta code.
3126 Remove src/Makefile.am from the picture.
3127 * configure: Rebuilt.
3129 * Makefile.am: Move logic from src/Makefile.am. Add changes
3130 to support libffi as a target library.
3131 * Makefile.in: Rebuilt.
3133 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3134 Upgraded to new autoconf, automake, libtool.
3138 * LICENSE: Update copyright date.
3140 * src/Makefile.am, src/Makefile.in: Removed.
3142 1998-11-29 Anthony Green <green@cygnus.com>
3144 * include/ChangeLog: Removed.
3145 * src/ChangeLog: Removed.
3146 * src/mips/ChangeLog: Removed.
3147 * src/sparc/ChangeLog: Remboved.
3148 * src/x86/ChangeLog: Removed.
3150 * ChangeLog.v1: Created.