]> git.ipfire.org Git - thirdparty/gcc.git/blame - libffi/ChangeLog
projects / thirdparty / gcc.git / blame
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | blame (incremental) | history | HEAD
re PR libffi/40700 (All amd64 libffi execution tests fail on Solaris 10/x86)
[thirdparty/gcc.git] / libffi / ChangeLog
CommitLineData
dcb00f2e
RO		 12009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
2
3 PR libffi/40700
4 * src/closures.c [X86_64 && __sun__ && __svr4__]
5 (FFI_MMAP_EXEC_WRIT): Define.
6
6c6859a1
DD		 72009-12-08 David Daney <ddaney@caviumnetworks.com>
8
9 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
10 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
11 * testsuite/libffi.call/stret_large.c: Same.
12 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
13 * testsuite/libffi.call/stret_large2.c: Same.
14 * testsuite/libffi.call/stret_medium2.c: Same.
15
9e7e65a0
DE		 162009-12-07 David Edelsohn <edelsohn@gnu.org>
17
18 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
19 typo.
20
5d7fc941
DE		 212009-12-05 David Edelsohn <edelsohn@gnu.org>
22
23 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
24 code.
9e7e65a0 25 * src/powerpc/aix_closure.S: Same.
6c6859a1 26
43e02a8a
RW		 272009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
28
29 * Makefile.in: Regenerate.
30 * configure: Regenerate.
31 * include/Makefile.in: Regenerate.
32 * man/Makefile.in: Regenerate.
33 * testsuite/Makefile.in: Regenerate.
34
2f2d7386
DE		 352009-12-04 David Edelsohn <edelsohn@gnu.org>
36
37 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
38 linux64_closure.S.
6c6859a1 39
6078f50a
UB		 402009-12-04 Uros Bizjak <ubizjak@gmail.com>
41
42 PR libffi/41908
43 * src/x86/ffi64.c (classify_argument): Update from
44 gcc/config/i386/i386.c.
45 (ffi_closure_unix64_inner): Do not use the address of two consecutive
46 SSE registers directly.
47 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
48 for x86_64 linux targets.
49
0c552622
DE		 502009-12-04 David Edelsohn <edelsohn@gnu.org>
51
52 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
53 pfr for long double split between fpr13 and stack.
54
afb5e622
DE		 552009-12-03 David Edelsohn <edelsohn@gnu.org>
56
57 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
58 fparg_count twice for long double.
59
be311914
DE		 602009-12-03 David Edelsohn <edelsohn@gnu.org>
61
62 PR libffi/42243
63 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
64
a27c4040
UB		 652009-12-03 Uros Bizjak <ubizjak@gmail.com>
66
67 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
ddf6c187 68 Remove xfails for x86 linux targets.
a27c4040 69
7837c247
DE		 702009-12-02 David Edelsohn <edelsohn@gnu.org>
71
72 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
73 case.
74
752009-12-01 David Edelsohn <edelsohn@gnu.org>
6d518d3b
DE		 76
77 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
78 register usage. Call ffi_prep_args directly. Add long double
79 return value support.
80 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
81 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
82 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
83 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
84 mode.
85 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
86 into case.
87 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
88 Allocate result area between params and FPRs.
89
5751cf6f
DE		 902009-11-30 David Edelsohn <edelsohn@gnu.org>
91
92 PR target/35484
93 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
94 AIX64.
95 * src/powerpc/aix.S: Implement AIX64 version.
96 * src/powerpc/aix_closure.S: Implement AIX64 version.
97 (ffi_closure_ASM): Use extsb, lha and displament addresses.
98 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
99 support.
100 (ffi_prep_cif_machdep): Same.
101 (ffi_call): Same.
102 (ffi_closure_helper_DARWIN): Same.
103
2265f646
AT		 1042009-11-02 Andreas Tobler <a.tobler@schweiz.org>
105
106 PR libffi/41908
107 * testsuite/libffi.call/testclosure.c: New test.
108
ef6ed66d
KT		 1092009-09-28 Kai Tietz <kai.tietz@onevision.com>
110
111 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
112 assembly version use of ___chkstk.
113
29d1370a
MK		 1142009-09-23 Matthias Klose <doko@ubuntu.com>
115
116 PR libffi/40242, PR libffi/41443
117 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
118 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
119 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
120 Change the conditionals to __SOFTFP__ || __ARM_EABI__
121 for -mfloat-abi=softfp to work.
122
987ce557
LR		 1232009-09-17 Loren J. Rittle <ljrittle@acm.org>
124
125 PR testsuite/32843 (strikes again)
126 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
127 enable proper extension on char and short.
128
5cbf8c8d
DD		 1292009-09-15 David Daney <ddaney@caviumnetworks.com>
130
131 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
132 handling for FFI_TYPE_POINTER.
133 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
134 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
135 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
136 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
137 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
138 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
139 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
140 structure and pointer returns.
141 (ffi_closure_N32): Add handling for pointer returns.
142 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
143 calc_n32_return_struct_flags): Handle soft-float.
144 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
145 (ffi_call_N32): Declare proper argument types.
146 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
147 soft-float.
148
df58e648
RW		 1492009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
150
151 * configure.ac (AC_PREREQ): Bump to 2.64.
152
5213506e
RW		 1532009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
154
70fa0efa
RW		 155 * Makefile.am (install-html, install-pdf): Remove.
156 * Makefile.in: Regenerate.
157
5213506e
RW		 158 * Makefile.in: Regenerate.
159 * aclocal.m4: Regenerate.
160 * configure: Regenerate.
161 * fficonfig.h.in: Regenerate.
162 * include/Makefile.in: Regenerate.
163 * man/Makefile.in: Regenerate.
164 * testsuite/Makefile.in: Regenerate.
165
1b3b24c2
RW		 1662009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
167
168 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
169
723512ba
DK		 1702009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
171
172 PR libffi/40807
173 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
174 return types for X86_WIN32.
175 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
176 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
177 _ffi_closure_STDCALL): Likewise.
178
179 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
180 (dlmmap, dlmunmap): Also use these functions on Cygwin.
181
33cde516
RS		 1822009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
183
184 PR testsuite/40699
185 PR testsuite/40707
186 PR testsuite/40709
187 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
188 2009-06-30 commits.
189
edd3a61c
RS		 1902009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
191
192 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
193 to "" before adding paths. (This reinstates an assignment that
194 was removed by my 2009-06-30 commit, but changes the initial
195 value from "." to "".)
196
33d1a550
L		 1972009-07-01 H.J. Lu <hongjiu.lu@intel.com>
198
199 PR testsuite/40601
200 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
201 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
202 empty.
203
b30bfd40
RS		 2042009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
205
206 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
207 to ld_library_path. Use add_path. Add just find_libgcc_s
208 to ld_library_path, not every libgcc multilib directory.
209
6975f17f
WL		 2102009-06-16 Wim Lewis <wiml@hhhh.org>
211
212 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
213 supposed to be callee-saved.
214 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
215 return buffer for odd-size structs.
216
7bbcc286
AT		 2172009-06-16 Andreas Tobler <a.tobler@schweiz.org>
218
219 PR libffi/40444
220 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
221 allow_stack_execute for Darwin.
222
fe673cc9
AH		 2232009-06-16 Andrew Haley <aph@redhat.com>
224
225 * configure.ac (TARGETDIR): Add missing blank lines.
226 * configure: Regenerate.
227
3bc462c2
AH		 2282009-06-16 Andrew Haley <aph@redhat.com>
229
230 * testsuite/libffi.call/cls_align_sint64.c,
231 testsuite/libffi.call/cls_align_uint64.c,
232 testsuite/libffi.call/cls_longdouble_va.c,
233 testsuite/libffi.call/cls_ulonglong.c,
234 testsuite/libffi.call/return_ll1.c,
235 testsuite/libffi.call/stret_medium2.c: Fix printf format
236 specifiers.
237 * testsuite/libffi.call/ffitest.h,
238 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
239
081b044b
AH		 2402009-06-15 Andrew Haley <aph@redhat.com>
241
242 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
243 * testsuite/libffi.call/err_bad_abi.c: Likewise.
244
062b8279
AH		 2452009-06-12 Andrew Haley <aph@redhat.com>
246
247 * Makefile.am: Remove info_TEXINFOS.
248
2492009-06-12 Andrew Haley <aph@redhat.com>
250
251 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
252 testsuite/libffi.call/cls_align_uint64.c,
253 testsuite/libffi.call/cls_ulonglong.c,
254 testsuite/libffi.call/return_ll1.c,
255 testsuite/libffi.call/stret_medium2.c: Fix printf format
256 specifiers.
257 testsuite/libffi.special/unwindtest.cc: include stdint.h.
258
2592009-06-11 Timothy Wall <twall@users.sf.net>
260
261 * Makefile.am,
262 configure.ac,
263 include/ffi.h.in,
264 include/ffi_common.h,
265 src/closures.c,
266 src/dlmalloc.c,
267 src/x86/ffi.c,
268 src/x86/ffitarget.h,
269 src/x86/win64.S (new),
270 README: Added win64 support (mingw or MSVC)
271 * Makefile.in,
272 include/Makefile.in,
273 man/Makefile.in,
274 testsuite/Makefile.in,
275 configure,
276 aclocal.m4: Regenerated
277 * ltcf-c.sh: properly escape cygwin/w32 path
278 * man/ffi_call.3: Clarify size requirements for return value.
279 * src/x86/ffi64.c: Fix filename in comment.
280 * src/x86/win32.S: Remove unused extern.
281
282 * testsuite/libffi.call/closure_fn0.c,
283 testsuite/libffi.call/closure_fn1.c,
284 testsuite/libffi.call/closure_fn2.c,
285 testsuite/libffi.call/closure_fn3.c,
286 testsuite/libffi.call/closure_fn4.c,
287 testsuite/libffi.call/closure_fn5.c,
288 testsuite/libffi.call/closure_fn6.c,
289 testsuite/libffi.call/closure_stdcall.c,
290 testsuite/libffi.call/cls_12byte.c,
291 testsuite/libffi.call/cls_16byte.c,
292 testsuite/libffi.call/cls_18byte.c,
293 testsuite/libffi.call/cls_19byte.c,
294 testsuite/libffi.call/cls_1_1byte.c,
295 testsuite/libffi.call/cls_20byte.c,
296 testsuite/libffi.call/cls_20byte1.c,
297 testsuite/libffi.call/cls_24byte.c,
298 testsuite/libffi.call/cls_2byte.c,
299 testsuite/libffi.call/cls_3_1byte.c,
300 testsuite/libffi.call/cls_3byte1.c,
301 testsuite/libffi.call/cls_3byte2.c,
302 testsuite/libffi.call/cls_4_1byte.c,
303 testsuite/libffi.call/cls_4byte.c,
304 testsuite/libffi.call/cls_5_1_byte.c,
305 testsuite/libffi.call/cls_5byte.c,
306 testsuite/libffi.call/cls_64byte.c,
307 testsuite/libffi.call/cls_6_1_byte.c,
308 testsuite/libffi.call/cls_6byte.c,
309 testsuite/libffi.call/cls_7_1_byte.c,
310 testsuite/libffi.call/cls_7byte.c,
311 testsuite/libffi.call/cls_8byte.c,
312 testsuite/libffi.call/cls_9byte1.c,
313 testsuite/libffi.call/cls_9byte2.c,
314 testsuite/libffi.call/cls_align_double.c,
315 testsuite/libffi.call/cls_align_float.c,
316 testsuite/libffi.call/cls_align_longdouble.c,
317 testsuite/libffi.call/cls_align_longdouble_split.c,
318 testsuite/libffi.call/cls_align_longdouble_split2.c,
319 testsuite/libffi.call/cls_align_pointer.c,
320 testsuite/libffi.call/cls_align_sint16.c,
321 testsuite/libffi.call/cls_align_sint32.c,
322 testsuite/libffi.call/cls_align_sint64.c,
323 testsuite/libffi.call/cls_align_uint16.c,
324 testsuite/libffi.call/cls_align_uint32.c,
325 testsuite/libffi.call/cls_align_uint64.c,
326 testsuite/libffi.call/cls_dbls_struct.c,
327 testsuite/libffi.call/cls_double.c,
328 testsuite/libffi.call/cls_double_va.c,
329 testsuite/libffi.call/cls_float.c,
330 testsuite/libffi.call/cls_longdouble.c,
331 testsuite/libffi.call/cls_longdouble_va.c,
332 testsuite/libffi.call/cls_multi_schar.c,
333 testsuite/libffi.call/cls_multi_sshort.c,
334 testsuite/libffi.call/cls_multi_sshortchar.c,
335 testsuite/libffi.call/cls_multi_uchar.c,
336 testsuite/libffi.call/cls_multi_ushort.c,
337 testsuite/libffi.call/cls_multi_ushortchar.c,
338 testsuite/libffi.call/cls_pointer.c,
339 testsuite/libffi.call/cls_pointer_stack.c,
340 testsuite/libffi.call/cls_schar.c,
341 testsuite/libffi.call/cls_sint.c,
342 testsuite/libffi.call/cls_sshort.c,
343 testsuite/libffi.call/cls_uchar.c,
344 testsuite/libffi.call/cls_uint.c,
345 testsuite/libffi.call/cls_ulonglong.c,
346 testsuite/libffi.call/cls_ushort.c,
347 testsuite/libffi.call/err_bad_abi.c,
348 testsuite/libffi.call/err_bad_typedef.c,
349 testsuite/libffi.call/float2.c,
350 testsuite/libffi.call/huge_struct.c,
351 testsuite/libffi.call/nested_struct.c,
352 testsuite/libffi.call/nested_struct1.c,
353 testsuite/libffi.call/nested_struct10.c,
354 testsuite/libffi.call/nested_struct2.c,
355 testsuite/libffi.call/nested_struct3.c,
356 testsuite/libffi.call/nested_struct4.c,
357 testsuite/libffi.call/nested_struct5.c,
358 testsuite/libffi.call/nested_struct6.c,
359 testsuite/libffi.call/nested_struct7.c,
360 testsuite/libffi.call/nested_struct8.c,
361 testsuite/libffi.call/nested_struct9.c,
362 testsuite/libffi.call/problem1.c,
363 testsuite/libffi.call/return_ldl.c,
364 testsuite/libffi.call/return_ll1.c,
365 testsuite/libffi.call/stret_large.c,
366 testsuite/libffi.call/stret_large2.c,
367 testsuite/libffi.call/stret_medium.c,
368 testsuite/libffi.call/stret_medium2.c,
369 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
370 of checking for MMAP. Use intptr_t instead of long casts.
371
f43def61
KK		 3722009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
373
374 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
375 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
376 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
377
6aa0b218
AH		 3782009-06-09 Andrew Haley <aph@redhat.com>
379
380 * src/x86/freebsd.S: Add missing file.
381
b7b857bb
AH		 3822009-06-08 Andrew Haley <aph@redhat.com>
383
384 Import from libffi 3.0.8:
385
386 * doc/libffi.texi: New file.
387 * doc/libffi.info: Likewise.
388 * doc/stamp-vti: Likewise.
389 * man/Makefile.am: New file.
390 * man/ffi_call.3: New file.
391
392 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
393 src/dlmalloc.c.
394 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
395
396 * configure.ac: Bump version to 3.0.8.
397 parisc*-*-linux*: Add.
398 i386-*-freebsd* | i386-*-openbsd*: Add.
399 powerpc-*-beos*: Add.
400 AM_CONDITIONAL X86_FREEBSD: Add.
401 AC_CONFIG_FILES: Add man/Makefile.
402
403 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
404
491e6ad1
AH		 4052009-06-08 Andrew Haley <aph@redhat.com>
406
407 * README: Import from libffi 3.0.8.
408
3b5b1a73
AH		 4092009-06-08 Andrew Haley <aph@redhat.com>
410
411 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
412 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
413 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
414 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
415
416 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
417 * testsuite/libffi.call/stret_medium.c: Likewise.
418 * testsuite/libffi.call/stret_large2.c: Likewise.
419 * testsuite/libffi.call/stret_large.c: Likewise.
420
4212008-12-26 Timothy Wall <twall@users.sf.net>
422
423 * testsuite/libffi.call/cls_longdouble.c,
424 testsuite/libffi.call/cls_longdouble_va.c,
425 testsuite/libffi.call/cls_align_longdouble.c,
426 testsuite/libffi.call/cls_align_longdouble_split.c,
427 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
428 failures on x86_64 cygwin/mingw.
429
4302008-12-22 Timothy Wall <twall@users.sf.net>
431
432 * testsuite/libffi.call/closure_fn0.c,
433 testsuite/libffi.call/closure_fn1.c,
434 testsuite/libffi.call/closure_fn2.c,
435 testsuite/libffi.call/closure_fn3.c,
436 testsuite/libffi.call/closure_fn4.c,
437 testsuite/libffi.call/closure_fn5.c,
438 testsuite/libffi.call/closure_fn6.c,
439 testsuite/libffi.call/closure_loc_fn0.c,
440 testsuite/libffi.call/closure_stdcall.c,
441 testsuite/libffi.call/cls_align_pointer.c,
442 testsuite/libffi.call/cls_pointer.c,
443 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
444 pointer to integer (intptr_t).
445 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
446
4472008-07-24 Anthony Green <green@redhat.com>
448
449 * testsuite/libffi.call/cls_dbls_struct.c,
450 testsuite/libffi.call/cls_double_va.c,
451 testsuite/libffi.call/cls_longdouble.c,
452 testsuite/libffi.call/cls_longdouble_va.c,
453 testsuite/libffi.call/cls_pointer.c,
454 testsuite/libffi.call/cls_pointer_stack.c,
455 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
456 compiler warnings.
457
4582008-03-04 Anthony Green <green@redhat.com>
459 Blake Chaffin
460 hos@tamanegi.org
461
462 * testsuite/libffi.call/cls_align_longdouble_split2.c
463 testsuite/libffi.call/cls_align_longdouble_split.c
464 testsuite/libffi.call/cls_dbls_struct.c
465 testsuite/libffi.call/cls_double_va.c
466 testsuite/libffi.call/cls_longdouble.c
467 testsuite/libffi.call/cls_longdouble_va.c
468 testsuite/libffi.call/cls_pointer.c
469 testsuite/libffi.call/cls_pointer_stack.c
470 testsuite/libffi.call/err_bad_abi.c
471 testsuite/libffi.call/err_bad_typedef.c
472 testsuite/libffi.call/stret_large2.c
473 testsuite/libffi.call/stret_large.c
474 testsuite/libffi.call/stret_medium2.c
475 testsuite/libffi.call/stret_medium.c: New tests from Apple.
476
56faec5e
AH		 4772009-06-05 Andrew Haley <aph@redhat.com>
478
479 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
480 libffi.
481
90409e4c
AH		 4822009-06-04 Andrew Haley <aph@redhat.com>
483
484 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
485 stdcall changes.
486
b1393750
AH		 4872008-02-26 Anthony Green <green@redhat.com>
488 Thomas Heller <theller@ctypes.org>
489
490 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
491 comment.
492
4932008-02-03 Timothy Wall <twall@users.sf.net>
494
495 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
496 offset based on code pointer, not data pointer.
497
4982008-01-31 Timothy Wall <twall@users.sf.net>
499
500 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
501 closures.
502 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
503 closures.
504 * src/x86/win32.S: Add assembly for stdcall closure.
505 * src/x86/ffi.c: Initialize stdcall closure trampoline.
506
96baa251
AH		 5072009-06-04 Andrew Haley <aph@redhat.com>
508
509 * include/ffi.h.in: Change void (*)() to void (*)(void).
510 * src/x86/ffi.c: Likewise.
511
7446546a
AH		 5122009-06-04 Andrew Haley <aph@redhat.com>
513
514 * src/powerpc/ppc_closure.S: Insert licence header.
515 * src/powerpc/linux64_closure.S: Likewise.
516 * src/m68k/sysv.S: Likewise.
517
518 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
519 * src/powerpc/ffi.c: Likewise.
520 * src/powerpc/ffi_darwin.c: Likewise.
521 * src/m32r/ffi.c: Likewise.
522 * src/sh64/ffi.c: Likewise.
523 * src/x86/ffi64.c: Likewise.
524 * src/alpha/ffi.c: Likewise.
525 * src/alpha/osf.S: Likewise.
526 * src/frv/ffi.c: Likewise.
527 * src/s390/ffi.c: Likewise.
528 * src/pa/ffi.c: Likewise.
529 * src/pa/hpux32.S: Likewise.
530 * src/ia64/unix.S: Likewise.
531 * src/ia64/ffi.c: Likewise.
532 * src/sparc/ffi.c: Likewise.
533 * src/mips/ffi.c: Likewise.
534 * src/sh/ffi.c: Likewise.
535
5362008-02-15 David Daney <ddaney@avtrex.com>
537
538 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
539 Define (conditionally), and use it to include cachectl.h.
540 (ffi_prep_closure_loc): Fix cache flushing.
541 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
542
5f933ef0
AH		 5432009-06-04 Andrew Haley <aph@redhat.com>
544
545 include/ffi.h.in,
546 src/arm/ffitarget.h,
547 src/arm/ffi.c,
548 src/arm/sysv.S,
549 src/powerpc/ffitarget.h,
550 src/closures.c,
551 src/sh64/ffitarget.h,
552 src/sh64/ffi.c,
553 src/sh64/sysv.S,
554 src/types.c,
555 src/x86/ffi64.c,
556 src/x86/ffitarget.h,
557 src/x86/win32.S,
558 src/x86/darwin.S,
559 src/x86/ffi.c,
560 src/x86/sysv.S,
561 src/x86/unix64.S,
562 src/alpha/ffitarget.h,
563 src/alpha/ffi.c,
564 src/alpha/osf.S,
565 src/m68k/ffitarget.h,
566 src/frv/ffitarget.h,
567 src/frv/ffi.c,
568 src/s390/ffitarget.h,
569 src/s390/sysv.S,
570 src/cris/ffitarget.h,
571 src/pa/linux.S,
572 src/pa/ffitarget.h,
573 src/pa/ffi.c,
574 src/raw_api.c,
575 src/ia64/ffitarget.h,
576 src/ia64/unix.S,
577 src/ia64/ffi.c,
578 src/ia64/ia64_flags.h,
579 src/java_raw_api.c,
580 src/debug.c,
581 src/sparc/v9.S,
582 src/sparc/ffitarget.h,
583 src/sparc/ffi.c,
584 src/sparc/v8.S,
585 src/mips/ffitarget.h,
586 src/mips/n32.S,
587 src/mips/o32.S,
588 src/mips/ffi.c,
589 src/prep_cif.c,
590 src/sh/ffitarget.h,
591 src/sh/ffi.c,
592 src/sh/sysv.S: Update license text.
593
84c40c4a
DK		 5942009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
595
596 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
597 (.eh_frame): Add FDE for it.
598
8353cd27
DK		 5992009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
600
601 * configure.ac: Also check if assembler supports pc-relative
602 relocs on X86_WIN32 targets.
603 * configure: Regenerate.
604 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
605 (_ffi_call_SYSV): Add missing function type symbol .def and
606 add EH markup labels.
607 (_ffi_call_STDCALL): Likewise.
608 (_ffi_closure_SYSV): Likewise.
609 (_ffi_closure_raw_SYSV): Likewise.
610 (.eh_frame): Add hand-crafted EH data.
611
748086b7
JJ		 6122009-04-09 Jakub Jelinek <jakub@redhat.com>
613
614 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
615 version 3 of the GNU General Public License and to point readers
616 at the COPYING3 file and the FSF's license web page.
617 * testsuite/libffi.call/call.exp: Likewise.
618 * testsuite/libffi.special/special.exp: Likewise.
619
7a9d3fe8
RW		 6202009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
621
622 * configure: Regenerate.
623
4d3d4078
RO		 6242008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
625
626 PR libffi/26048
627 * configure.ac (HAVE_AS_X86_PCREL): New test.
628 * configure: Regenerate.
629 * fficonfig.h.in: Regenerate.
630 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
631 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
632 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
633 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
634 * src/x86/unix64.S (.Lstore_table): Move to .text section.
635 (.Lload_table): Likewise.
636 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
637
baafc534
RW		 6382008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
639
640 * configure: Regenerate.
641
da17a98b
EB		 6422008-11-21 Eric Botcazou <ebotcazou@adacore.com>
643
644 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
645 signed/unsigned int8/16 return values.
646 * src/sparc/v8.S (ffi_call_v8): Likewise.
647 (ffi_closure_v8): Likewise.
648
80c6c2cb
PG		 6492008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
650 Steve Ellcey <sje@cup.hp.com>
651
652 * configure: Regenerate for new libtool.
653 * Makefile.in: Ditto.
654 * include/Makefile.in: Ditto.
655 * aclocal.m4: Ditto.
656
d7f68a67
AT		 6572008-08-25 Andreas Tobler <a.tobler@schweiz.org>
658
659 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
660 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
661 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
662 Adjust copyright notice.
663 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
664 register or two register to use for FFI_SYSV structs.
665 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
666 (ffi_closure_helper_SYSV): Fix the return type for
667 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
668 Adjust copyright notice.
669
130f4c50
KK		 6702008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
671
672 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
673 int.
674
18c04407
RW		 6752008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
676
677 * configure: Regenerate.
678 * include/Makefile.in: Regenerate.
679 * testsuite/Makefile.in: Regenerate.
680
cd985f66
JM		 6812008-06-07 Joseph Myers <joseph@codesourcery.com>
682
683 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
684 powerpc-*-beos*): Remove.
685 * configure: Regenerate.
686
7afa92c5
JB		 6872008-05-09 Julian Brown <julian@codesourcery.com>
688
689 * Makefile.am (LTLDFLAGS): New.
690 (libffi_la_LDFLAGS): Use above.
691 * Makefile.in: Regenerate.
692
deb984e6
PB		 6932008-04-18 Paolo Bonzini <bonzini@gnu.org>
694
695 PR bootstrap/35457
696 * aclocal.m4: Regenerate.
697 * configure: Regenerate.
698
3e9608a5
KK		 6992008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
700
701 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
702 * src/sh64/sysv.S: Likewise.
703
feb51530
MK		 7042008-03-26 Daniel Jacobowitz <dan@debian.org>
705
706 * src/arm/sysv.S: Fix ARM comment marker.
707
7082008-03-26 Jakub Jelinek <jakub@redhat.com>
709
710 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
711 * src/s390/sysv.S: Likewise.
712 * src/powerpc/ppc_closure.S: Likewise.
713 * src/powerpc/sysv.S: Likewise.
714 * src/x86/unix64.S: Likewise.
715 * src/x86/sysv.S: Likewise.
716 * src/sparc/v8.S: Likewise.
717 * src/sparc/v9.S: Likewise.
718 * src/m68k/sysv.S: Likewise.
719 * src/arm/sysv.S: Likewise.
720
38d24731
RW		 7212008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
722
723 * aclocal.m4: Regenerate.
724 * configure: Likewise.
725 * Makefile.in: Likewise.
726 * include/Makefile.in: Likewise.
727 * testsuite/Makefile.in: Likewise.
728
2e250fa2
BK		 7292008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
730 Andreas Tobler <a.tobler@schweiz.org>
731
732 * configure.ac: Add amd64-*-freebsd* target.
733 * configure: Regenerate.
734
7f92e242
L		 7352008-01-30 H.J. Lu <hongjiu.lu@intel.com>
736
737 PR libffi/34612
738 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
739 returning struct.
740
741 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
742 tests.
743
95d4f544
DE		 7442008-01-24 David Edelsohn <edelsohn@gnu.org>
745
746 * configure: Regenerate.
747
79ca8fc0
AT		 7482008-01-06 Andreas Tobler <a.tobler@schweiz.org>
749
750 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
751
d7b07cc7
AT		 7522008-01-05 Andreas Tobler <a.tobler@schweiz.org>
753
754 PR testsuite/32843
755 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
756 signed/unsigned int8/16 for X86_DARWIN.
757 Updated copyright info.
758 Handle one and two byte structs with special cif->flags.
759 * src/x86/ffitarget.h: Add special types for one and two byte structs.
760 Updated copyright info.
761 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
762 sysv.S
763 Remove code to pop args from the stack after call.
764 Special-case signed/unsigned for int8/16, one and two byte structs.
765 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
766 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
767 FFI_TYPE_SINT32.
768 Updated copyright info.
769
c94ae2e4
DD		 7702007-12-08 David Daney <ddaney@avtrex.com>
771
772 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
773 SUBU, add with ADDU and use smaller code sequences.
774
3875b6d7 7752007-12-07 David Daney <ddaney@avtrex.com>
44be66a1 776
3875b6d7
DD		 777 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
778 type.
779
4c42b3d8
DD		 7802007-12-06 David Daney <ddaney@avtrex.com>
781
782 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
783 defined.
784 (ffi_java_raw): New typedef.
785 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
786 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
787 ffi_java_raw.
788 (ffi_java_raw_closure) : Same.
789 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
790 parameter types.
791 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
792 FFI_SIZEOF_JAVA_RAW.
793 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
794 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
795 sizeof(ffi_java_raw) for alignment calculations.
796 (ffi_java_ptrarray_to_raw): Same.
797 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
798 if FFI_SIZEOF_JAVA_RAW == 4.
799 (ffi_java_raw_to_rvalue): Same.
800 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
801 (ffi_java_translate_args): Same.
802 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
803 parameter types.
804 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
805
3ef50b62
DD		 8062007-12-06 David Daney <ddaney@avtrex.com>
807
808 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
809 pointer values.
810
16287148
AT		 8112007-12-01 Andreas Tobler <a.tobler@schweiz.org>
812
813 PR libffi/31937
814 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
815 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
816 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
817 set the NUM_FPR_ARG_REGISTERS according to.
818 Add support for potential soft-float support under hard-float
819 architecture.
820 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
821 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
822 to the FFI_LINUX_SOFT_FLOAT ABI.
823 (ffi_prep_cif_machdep): Likewise.
824 (ffi_closure_helper_SYSV): Likewise.
825 * src/powerpc/ppc_closure.S: Make sure not to store float/double
826 on archs where __NO_FPRS__ is true.
827 Add FFI_TYPE_UINT128 support.
828 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
829 Adjust copyright notice.
830
a4b7c9cd
AT		 8312007-11-25 Andreas Tobler <a.tobler@schweiz.org>
832
833 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
834 * include/ffi_common.h: ... here.
835 Update copyright.
836
e84296c6
AT		 8372007-11-17 Andreas Tobler <a.tobler@schweiz.org>
838
839 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
840 * src/powerpc/linux64.S: Likewise.
841 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
842 * testsuite/libffi.call/return_ldl.c: New test.
843
cb642590
AH		 8442007-09-04 <aph@redhat.com>
845
846 * src/arm/sysv.S (UNWIND): New.
847 (Whole file): Conditionally compile unwinder directives.
848 * src/arm/sysv.S: Add unwinder directives.
849
850 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
851 Only treat r0 as a struct address if we're actually returning a
852 struct by address.
853 Only copy the bytes that are actually within a struct.
854 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
855 is returned in r0, not passed by address.
856 (ffi_call): Allocate a word-sized temporary for the case where
857 a composite is returned in r0.
858 (ffi_prep_incoming_args_SYSV): Align as necessary.
859
8602007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
861
862 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
863 directly using the sys_cacheflush syscall.
864
8652007-07-27 Andrew Haley <aph@redhat.com>
866
867 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
868
8d5c5e35
MR		 8692007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
870
871 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
872 * configure.ac: Likewise.
873 * Makefile.in: Regenerate.
874 * include/Makefile.in: Likewise.
875 * testsuite/Makefile.in: Likewise.
876 * configure: Likewise.
877
787b0d19
DD		 8782007-08-24 David Daney <ddaney@avtrex.com>
879
880 * testsuite/libffi.call/return_sl.c: New test.
881
dfbb3e36
DD		 8822007-08-10 David Daney <ddaney@avtrex.com>
883
884 * testsuite/libffi.call/cls_multi_ushort.c,
885 testsuite/libffi.call/cls_align_uint16.c,
886 testsuite/libffi.call/nested_struct1.c,
887 testsuite/libffi.call/nested_struct3.c,
888 testsuite/libffi.call/cls_7_1_byte.c,
889 testsuite/libffi.call/nested_struct5.c,
890 testsuite/libffi.call/cls_double.c,
891 testsuite/libffi.call/nested_struct7.c,
892 testsuite/libffi.call/cls_sint.c,
893 testsuite/libffi.call/nested_struct9.c,
894 testsuite/libffi.call/cls_20byte1.c,
895 testsuite/libffi.call/cls_multi_sshortchar.c,
896 testsuite/libffi.call/cls_align_sint64.c,
897 testsuite/libffi.call/cls_3byte2.c,
898 testsuite/libffi.call/cls_multi_schar.c,
899 testsuite/libffi.call/cls_multi_uchar.c,
900 testsuite/libffi.call/cls_19byte.c,
901 testsuite/libffi.call/cls_9byte1.c,
902 testsuite/libffi.call/cls_align_float.c,
903 testsuite/libffi.call/closure_fn1.c,
904 testsuite/libffi.call/problem1.c,
905 testsuite/libffi.call/closure_fn3.c,
906 testsuite/libffi.call/cls_sshort.c,
907 testsuite/libffi.call/closure_fn5.c,
908 testsuite/libffi.call/cls_align_double.c,
909 testsuite/libffi.call/nested_struct.c,
910 testsuite/libffi.call/cls_2byte.c,
911 testsuite/libffi.call/nested_struct10.c,
912 testsuite/libffi.call/cls_4byte.c,
913 testsuite/libffi.call/cls_6byte.c,
914 testsuite/libffi.call/cls_8byte.c,
915 testsuite/libffi.call/cls_multi_sshort.c,
916 testsuite/libffi.call/cls_align_sint16.c,
917 testsuite/libffi.call/cls_align_uint32.c,
918 testsuite/libffi.call/cls_20byte.c,
919 testsuite/libffi.call/cls_float.c,
920 testsuite/libffi.call/nested_struct2.c,
921 testsuite/libffi.call/cls_5_1_byte.c,
922 testsuite/libffi.call/nested_struct4.c,
923 testsuite/libffi.call/cls_24byte.c,
924 testsuite/libffi.call/nested_struct6.c,
925 testsuite/libffi.call/cls_64byte.c,
926 testsuite/libffi.call/nested_struct8.c,
927 testsuite/libffi.call/cls_uint.c,
928 testsuite/libffi.call/cls_multi_ushortchar.c,
929 testsuite/libffi.call/cls_schar.c,
930 testsuite/libffi.call/cls_uchar.c,
931 testsuite/libffi.call/cls_align_uint64.c,
932 testsuite/libffi.call/cls_ulonglong.c,
933 testsuite/libffi.call/cls_align_longdouble.c,
934 testsuite/libffi.call/cls_1_1byte.c,
935 testsuite/libffi.call/cls_12byte.c,
936 testsuite/libffi.call/cls_3_1byte.c,
937 testsuite/libffi.call/cls_3byte1.c,
938 testsuite/libffi.call/cls_4_1byte.c,
939 testsuite/libffi.call/cls_6_1_byte.c,
940 testsuite/libffi.call/cls_16byte.c,
941 testsuite/libffi.call/cls_18byte.c,
942 testsuite/libffi.call/closure_fn0.c,
943 testsuite/libffi.call/cls_9byte2.c,
944 testsuite/libffi.call/closure_fn2.c,
945 testsuite/libffi.call/closure_fn4.c,
946 testsuite/libffi.call/cls_ushort.c,
947 testsuite/libffi.call/closure_fn6.c,
948 testsuite/libffi.call/cls_5byte.c,
949 testsuite/libffi.call/cls_align_pointer.c,
950 testsuite/libffi.call/cls_7byte.c,
951 testsuite/libffi.call/cls_align_sint32.c,
952 testsuite/libffi.special/unwindtest_ffi_call.cc,
953 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
954
89d9d98a
DD		 9552007-08-10 David Daney <ddaney@avtrex.com>
956
957 PR libffi/28313
958 * configure.ac: Don't treat mips64 as a special case.
959 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
960 * configure: Regenerate
961 * Makefile.in: Ditto.
962 * fficonfig.h.in: Ditto.
963 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
964 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
965 (FFI_DEFAULT_ABI): Set for n64 case.
966 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
967 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
968 (ffi_closure_N32): New function.
969 (.eh_frame): New section
970 * src/mips/o32.S: Clean up comments.
971 (ffi_closure_O32): Pass ffi_closure parameter in $12.
972 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
973 _MIPS_SIM == _ABIN32 throughout.
974 (FFI_MIPS_STOP_HERE): New, use in place of
975 ffi_stop_here.
976 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
977 to support n32/n64 ABIs.
978 (calc_n32_struct_flags): Rewrite.
979 (calc_n32_return_struct_flags): Remove unused variable. Reverse
980 position of flag bits.
981 (ffi_prep_cif_machdep): Rewrite n32 portion.
982 (ffi_call): Enable for n64. Add special handling for small structure
983 return values.
984 (ffi_prep_closure_loc): Add n32 and n64 support.
985 (ffi_closure_mips_inner_O32): Add cast to silence warning.
986 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
987
d23869d4
DD		 9882007-08-08 David Daney <ddaney@avtrex.com>
989
990 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
991 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
992 specifiers.
993 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
994 * testsuite/libffi.call/cls_sint.c (main): Ditto.
995 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
996 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
997 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
998 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
999 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1000 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1001 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1002 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1003 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1004 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1005 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1006 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1007 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1008 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1009 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1010 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1011 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1012 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1013 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1014 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1015 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1016 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1017 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1018 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1019 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1020 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1021 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1022 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1023 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1024
536e0a8f
AH		 10252007-08-07 Andrew Haley <aph@redhat.com>
1026
1027 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1028 checkin.
1029
7665c9be
AH		 10302007-08-06 Andrew Haley <aph@redhat.com>
1031
1032 PR testsuite/32843
1033 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1034 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1035 FFI_TYPE_SINT32.
1036
58e11497
DD		 10372007-08-02 David Daney <ddaney@avtrex.com>
1038
1039 * testsuite/libffi.call/return_ul.c (main): Define return type as
1040 ffi_arg. Use proper printf conversion specifier.
1041
387cfbba
AH		 10422007-07-30 Andrew Haley <aph@redhat.com>
1043
1044 PR testsuite/32843
1045 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1046 signed/unsigned int8/16.
1047 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1048 Use a jump table.
1049 Remove code to pop args from the stack after call.
1050 Special-case signed/unsigned int8/16.
1051 * testsuite/libffi.call/return_sc.c (main): Revert.
1052
0ee0208e
RG		 10532007-07-26 Richard Guenther <rguenther@suse.de>
1054
1055 PR testsuite/32843
1056 * testsuite/libffi.call/return_sc.c (main): Verify call
1057 result as signed char, not ffi_arg.
1058
04a49983
RO		 10592007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1060
1061 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1062 * configure: Regenerate.
1063
433e6a8c
DD		 10642007-07-11 David Daney <ddaney@avtrex.com>
1065
1066 * src/mips/ffi.c: Don't include sys/cachectl.h.
1067 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1068 cacheflush().
1069
0959e2b8
AH		 10702007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1071
1072 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1073 from (ffi_prep_closure): ... this.
1074 (FFI_INIT_TRAMPOLINE): Adjust.
1075
10762005-12-31 Phil Blundell <pb@reciva.com>
1077
1078 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1079 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1080 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1081 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1082 (FFI_CLOSURES): Enable closure support.
1083
10842007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1085
1086 * testsuite/libffi.call/cls_multi_ushort.c,
1087 testsuite/libffi.call/cls_align_uint16.c,
1088 testsuite/libffi.call/nested_struct1.c,
1089 testsuite/libffi.call/nested_struct3.c,
1090 testsuite/libffi.call/cls_7_1_byte.c,
1091 testsuite/libffi.call/cls_double.c,
1092 testsuite/libffi.call/nested_struct5.c,
1093 testsuite/libffi.call/nested_struct7.c,
1094 testsuite/libffi.call/cls_sint.c,
1095 testsuite/libffi.call/nested_struct9.c,
1096 testsuite/libffi.call/cls_20byte1.c,
1097 testsuite/libffi.call/cls_multi_sshortchar.c,
1098 testsuite/libffi.call/cls_align_sint64.c,
1099 testsuite/libffi.call/cls_3byte2.c,
1100 testsuite/libffi.call/cls_multi_schar.c,
1101 testsuite/libffi.call/cls_multi_uchar.c,
1102 testsuite/libffi.call/cls_19byte.c,
1103 testsuite/libffi.call/cls_9byte1.c,
1104 testsuite/libffi.call/cls_align_float.c,
1105 testsuite/libffi.call/closure_fn1.c,
1106 testsuite/libffi.call/problem1.c,
1107 testsuite/libffi.call/closure_fn3.c,
1108 testsuite/libffi.call/cls_sshort.c,
1109 testsuite/libffi.call/closure_fn5.c,
1110 testsuite/libffi.call/cls_align_double.c,
1111 testsuite/libffi.call/cls_2byte.c,
1112 testsuite/libffi.call/nested_struct.c,
1113 testsuite/libffi.call/nested_struct10.c,
1114 testsuite/libffi.call/cls_4byte.c,
1115 testsuite/libffi.call/cls_6byte.c,
1116 testsuite/libffi.call/cls_8byte.c,
1117 testsuite/libffi.call/cls_multi_sshort.c,
1118 testsuite/libffi.call/cls_align_uint32.c,
1119 testsuite/libffi.call/cls_align_sint16.c,
1120 testsuite/libffi.call/cls_float.c,
1121 testsuite/libffi.call/cls_20byte.c,
1122 testsuite/libffi.call/cls_5_1_byte.c,
1123 testsuite/libffi.call/nested_struct2.c,
1124 testsuite/libffi.call/cls_24byte.c,
1125 testsuite/libffi.call/nested_struct4.c,
1126 testsuite/libffi.call/nested_struct6.c,
1127 testsuite/libffi.call/cls_64byte.c,
1128 testsuite/libffi.call/nested_struct8.c,
1129 testsuite/libffi.call/cls_uint.c,
1130 testsuite/libffi.call/cls_multi_ushortchar.c,
1131 testsuite/libffi.call/cls_schar.c,
1132 testsuite/libffi.call/cls_uchar.c,
1133 testsuite/libffi.call/cls_align_uint64.c,
1134 testsuite/libffi.call/cls_ulonglong.c,
1135 testsuite/libffi.call/cls_align_longdouble.c,
1136 testsuite/libffi.call/cls_1_1byte.c,
1137 testsuite/libffi.call/cls_12byte.c,
1138 testsuite/libffi.call/cls_3_1byte.c,
1139 testsuite/libffi.call/cls_3byte1.c,
1140 testsuite/libffi.call/cls_4_1byte.c,
1141 testsuite/libffi.call/cls_6_1_byte.c,
1142 testsuite/libffi.call/cls_16byte.c,
1143 testsuite/libffi.call/cls_18byte.c,
1144 testsuite/libffi.call/closure_fn0.c,
1145 testsuite/libffi.call/cls_9byte2.c,
1146 testsuite/libffi.call/closure_fn2.c,
1147 testsuite/libffi.call/closure_fn4.c,
1148 testsuite/libffi.call/cls_ushort.c,
1149 testsuite/libffi.call/closure_fn6.c,
1150 testsuite/libffi.call/cls_5byte.c,
1151 testsuite/libffi.call/cls_align_pointer.c,
1152 testsuite/libffi.call/cls_7byte.c,
1153 testsuite/libffi.call/cls_align_sint32.c,
1154 testsuite/libffi.special/unwindtest_ffi_call.cc,
1155 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1156
12a27363
L		 11572007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1158
1159 * aclocal.m4: Regenerated.
1160
82a6cadf
PB		 11612007-06-02 Paolo Bonzini <bonzini@gnu.org>
1162
1163 * configure: Regenerate.
1164
1cf3d07d
SE		 11652007-05-23 Steve Ellcey <sje@cup.hp.com>
1166
1167 * Makefile.in: Regenerate.
1168 * configure: Regenerate.
1169 * aclocal.m4: Regenerate.
1170 * include/Makefile.in: Regenerate.
1171 * testsuite/Makefile.in: Regenerate.
1172
5635c26e
RZ		 11732007-05-10 Roman Zippel <zippel@linux-m68k.org>
1174
1175 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1176 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1177 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1178 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1179 (FFI_CLOSURES): Enable closure support.
1180
e6fbf387
RZ		 11812007-05-10 Roman Zippel <zippel@linux-m68k.org>
1182
1183 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1184 * configure: Regenerate.
1185 * fficonfig.h.in: Regenerate.
1186 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1187 CFI_OFFSET,CFI_DEF_CFA): New macros.
1188 (ffi_call_SYSV): Add callframe annotation.
1189
b5296e52
RZ		 11902007-05-10 Roman Zippel <zippel@linux-m68k.org>
1191
1192 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1193 numerous test suite failures.
1194 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1195
bd61d708
PB		 11962007-04-11 Paolo Bonzini <bonzini@gnu.org>
1197
1198 * Makefile.am (EXTRA_DIST): Bring up to date.
1199 * Makefile.in: Regenerate.
1200 * src/frv/eabi.S: Remove RCS keyword.
1201
bf8da5fc
RH		 12022007-04-06 Richard Henderson <rth@redhat.com>
1203
1204 * configure.ac: Tidy target case.
1205 (HAVE_LONG_DOUBLE): Allow the target to override.
1206 * configure: Regenerate.
1207 * include/ffi.h.in: Don't define ffi_type_foo if
1208 LIBFFI_HIDE_BASIC_TYPES is defined.
1209 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1210 to ffi_type_double.
1211 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1212 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1213 (ffi_type_longdouble): Special case for Alpha. Don't define
1214 if long double == double.
1215
1216 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1217 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1218 (ffi_call, ffi_closure_osf_inner): Likewise.
1219 (ffi_closure_osf_inner): Likewise. Mark hidden.
1220 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1221 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1222 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1223 (load_table): Handle 128-bit long double.
1224
1225 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1226
8669fed1
TT		 12272007-04-06 Tom Tromey <tromey@redhat.com>
1228
1229 PR libffi/31491:
1230 * README: Fixed bug in example.
1231
f0920e6c
JJ		 12322007-04-03 Jakub Jelinek <jakub@redhat.com>
1233
1234 * src/closures.c: Include sys/statfs.h.
1235 (_GNU_SOURCE): Define on Linux.
1236 (FFI_MMAP_EXEC_SELINUX): Define.
1237 (selinux_enabled): New variable.
1238 (selinux_enabled_check): New function.
1239 (is_selinux_enabled): Define.
1240 (dlmmap): Use it.
1241
34a6ccda
UB		 12422007-03-24 Uros Bizjak <ubizjak@gmail.com>
1243
1244 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1245 Use 'volatile float sum' to create sum of floats to avoid false
1246 negative due to excess precision on ix86 targets.
1247 (main): Ditto.
1248
8b9cf39e
AO		 12492007-03-08 Alexandre Oliva <aoliva@redhat.com>
1250
1251 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1252 patch.
1253 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1254
18fa3240
AO		 12552007-03-07 Alexandre Oliva <aoliva@redhat.com>
1256
1257 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1258 (ffi_prep_closure_loc): New.
1259 (ffi_prep_raw_closure_loc): New.
1260 (ffi_prep_java_raw_closure_loc): New.
1261 * src/closures.c: New file.
1262 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1263 Replace sflags with exec_offset.
1264 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1265 sub_segment_exec_offset): New macros.
1266 (get_segment_flags, set_segment_flags, check_segment_merge): New
1267 macros.
1268 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1269 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1270 destroy_mspace): Use new macros.
1271 (sys_alloc): Silence warning.
1272 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1273 * Makefile.in: Rebuilt.
1274 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1275 terms of ffi_prep_closure_loc.
1276 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1277 from...
1278 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1279 renamed version.
1280 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1281 adjusted from...
1282 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1283 the renamed version.
1284 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1285 (ffi_prep_closure): ... this.
1286 * src/pa/ffi.c: Likewise.
1287 * src/cris/ffi.c: Likewise. Adjust.
1288 * src/frv/ffi.c: Likewise.
1289 * src/ia64/ffi.c: Likewise.
1290 * src/mips/ffi.c: Likewise.
1291 * src/powerpc/ffi_darwin.c: Likewise.
1292 * src/s390/ffi.c: Likewise.
1293 * src/sh/ffi.c: Likewise.
1294 * src/sh64/ffi.c: Likewise.
1295 * src/sparc/ffi.c: Likewise.
1296 * src/x86/ffi64.c: Likewise.
1297 * src/x86/ffi.c: Likewise.
1298 (FFI_INIT_TRAMPOLINE): Adjust.
1299 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1300 (ffi_prep_raw_closure): ... this.
1301 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1302 (ffi_prep_closure): ... this.
1303 (flush_icache): Adjust.
1304
dd778331
AO		 13052007-03-07 Alexandre Oliva <aoliva@redhat.com>
1306
1307 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1308 Lea's malloc.
1309
b5f8d49b
BM		 13102007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1311
1312 * Makefile.am: Add dummy install-pdf target.
1313 * Makefile.in: Regenerate
1314
7a23933b
AK		 13152007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1316
1317 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1318 ffi_closure_helper_SYSV): Add long double handling.
1319
8aaf4681
JJ		 13202007-02-02 Jakub Jelinek <jakub@redhat.com>
1321
1322 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1323 immediately after bctrl instruction.
1324
0eb5e2b9
AO		 13252007-01-18 Alexandre Oliva <aoliva@redhat.com>
1326
1327 * Makefile.am (all-recursive, install-recursive,
1328 mostlyclean-recursive, clean-recursive, distclean-recursive,
1329 maintainer-clean-recursive): Add missing targets.
1330 * Makefile.in: Rebuilt.
1331
f4f7486b
AT		 13322006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1333
1334 * configure.ac: Add TARGET for x86_64-*-darwin*.
1335 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1336 for X86_DARWIN.
1337 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1338 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1339 * configure: Regenerate.
1340 * Makefile.in: Regenerate.
1341 * include/Makefile.in: Regenerate.
1342 * testsuite/Makefile.in: Regenerate.
1343 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1344 ffi_call only.
1345
3d5eafe2
AT		 13462006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1347
1348 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1349 Makefile.am.
1350
d5523ce1 13512006-10-31 Geoffrey Keating <geoffk@apple.com>
58182de3
GK		 1352
1353 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1354 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1355 Darwin.
1356 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1357 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1358
a3bd50cc
PB		 13592006-10-10 Paolo Bonzini <bonzini@gnu.org>
1360 Sandro Tolaini <tolaini@libero.it>
1361
1362 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1363 conditional.
1364 * configure: Regenerated.
1365 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1366 (EXTRA_DIST): Add src/x86/darwin.S.
1367 * Makefile.in: Regenerated.
1368 * include/Makefile.in: Regenerated.
1369 * testsuite/Makefile.in: Regenerated.
1370
1371 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1372 X86_WIN32, and additionally align stack to 16 bytes.
1373 * src/x86/darwin.S: New, based on sysv.S.
1374 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1375
b095810f
DD		 13762006-09-12 David Daney <ddaney@avtrex.com>
1377
1378 PR libffi/23935
1379 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1380 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1381 * aclocal.m4: Regenerated for automake 1.9.6.
1382 * Makefile.in: Regenerated.
1383 * include/Makefile.in: Regenerated.
1384 * testsuite/Makefile.in: Regenerated.
1385
110b2db4
AT		 13862006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1387
1388 * include/ffi_common.h (struct): Revert accidental commit.
1389
8e5fde87
AT		 13902006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1391
1392 * include/ffi_common.h: Remove lint directives.
1393 * include/ffi.h.in: Likewise.
1394
cb4132fe
TS		 13952006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1396
1397 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1398 for 32-bit architectures.
1399 * testsuite/libffi.call/return_ul.c: New test case.
1400
0d24c3fa
DD		 14012006-07-19 David Daney <ddaney@avtrex.com>
1402
1403 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1404 xfail remains for mips64.
1405
8665824c
CD		 14062006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1407
1408 * Makefile.am: Add install-html target. Add install-html to .PHONY
1409 * Makefile.in: Regenerate.
1410 * aclocal.m4: Regenerate.
1411 * include/Makefile.in: Regenerate.
1412 * testsuite/Makefile.in: Regenerate.
1413
84e2e370
JDA		 14142006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1415
1416 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1417 stack slot.
1418
ac6ed182
AT		 14192006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1420
1421 * README: Remove notice about 'Crazy Comments'.
1422 * src/debug.c: Remove lint directives. Cleanup white spaces.
1423 * src/java_raw_api.c: Likewise.
1424 * src/prep_cif.c: Likewise.
1425 * src/raw_api.c: Likewise.
1426 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1427 to the testsuite.
1428 * src/arm/ffi.c: Remove lint directives.
1429 * src/m32r/ffi.c: Likewise.
1430 * src/pa/ffi.c: Likewise.
1431 * src/powerpc/ffi.c: Likewise.
1432 * src/powerpc/ffi_darwin.c: Likewise.
1433 * src/sh/ffi.c: Likewise.
1434 * src/sh64/ffi.c: Likewise.
1435 * src/x86/ffi.c: Likewise.
1436 * testsuite/libffi.call/float2.c: Likewise.
1437 * testsuite/libffi.call/promotion.c: Likewise.
1438 * testsuite/libffi.call/struct1.c: Likewise.
1439
b7211585
AT		 14402006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1441
1442 * src/pa/hpux32.S: Correct unwind offset calculation for
1443 ffi_closure_pa32.
1444 * src/pa/linux.S: Likewise.
1445
86066f9b
JW		 14462006-04-12 James E Wilson <wilson@specifix.com>
1447
1448 PR libgcj/26483
1449 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1450 (hfa_type_load): Call stf_spill.
1451 (hfa_type_store): Call ldf_fill.
1452 (ffi_call): Adjust calls to above routines. Add local temps for
1453 macro result.
b7211585 1454
615baed7
MK		 14552006-04-10 Matthias Klose <doko@debian.org>
1456
1457 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1458 directory names containing underscores.
1459
8e7a2b58
JW		 14602006-04-07 James E Wilson <wilson@specifix.com>
1461
1462 * testsuite/libffi.call/float4.c: New testcase.
1463
e6c300ef
JDA		 14642006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1465 Andreas Tobler <a.tobler@schweiz.ch>
1466
1467 * Makefile.am: Add PA_HPUX port.
1468 * Makefile.in: Regenerate.
1469 * include/Makefile.in: Likewise.
1470 * testsuite/Makefile.in: Likewise.
1471 * configure.ac: Add PA_HPUX rules.
1472 * configure: Regenerate.
1473 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1474 Add PA_HPUX and PA64_HPUX.
1475 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1476 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1477 (FFI_TYPE_SMALL_STRUCT2): Define.
1478 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1479 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1480 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1481 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1482 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1483 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1484 * src/pa/ffi.c (ROUND_DOWN): Delete.
1485 (fldw, fstw, fldd, fstd): Use '__asm__'.
1486 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1487 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1488 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1489 Simplify incrementing of stack slot variable. Change type of local
1490 'n' to unsigned int.
1491 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1492 double on PA_HPUX.
1493 (ffi_prep_cif_machdep): Likewise.
1494 (ffi_call): Likewise.
1495 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1496 return type to ffi_status. Simplify incrementing of stack slot
1497 variable. Only copy floating point argument registers when PA_LINUX
1498 is true. Reformat debug statement.
1499 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1500 FFI_TYPE_SMALL_STRUCT8.
1501 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1502 declaration.
1503 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1504 Add nops to cache flush. Add trampoline for PA_HPUX.
1505 * src/pa/hpux32.S: New file.
1506 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1507 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1508 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1509 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1510 argument type checks so that common argument types appear first.
1511 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1512 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1513
75b8b1be
AM		 15142006-03-24 Alan Modra <amodra@bigpond.net.au>
1515
1516 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1517 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1518 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1519 FFI_TYPE_LONGDOUBLE.
1520 (ffi_prep_args64): Assert using IBM extended double.
1521 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1522 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1523 (ffi_call): Handle FFI_LINUX.
1524 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1525 gpr3 return pointer as for struct return. Handle FFI_LINUX
1526 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1527 unnecessarily.
1528 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1529 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1530 Don't use r6 as pointer to results, instead use sp offset. Don't
1531 make a special call to load lr with case table address, instead
1532 use offset from previous call.
1533 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1534 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1535 return.
1536
697701ad
KK		 15372006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1538
1539 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1540 passed with FP registers correctly.
1541 (ffi_closure_helper_SYSV): Likewise.
1542 * src/sh64/sysv.S: Likewise.
1543
b295c471
AT		 15442006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1545
1546 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1547 args and userdata unused.
1548 (closure_test_fn1): Mark cif and userdata unused.
1549 (main): Remove unused res.
1550
9fdeb13b
AT		 15512006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1552
1553 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1554 -O2, -O3, -Os and the warning flags -W -Wall.
1555 * testsuite/libffi.special/special.exp: Likewise.
1556 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1557 unused parameter unused for gcc or else do nothing.
1558 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1559 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1560 and userdata unused.
1561 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1562 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1563 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1564 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1565 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1566 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1567 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1568 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1569 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1570 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1571 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1572 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1573 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1574 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1575 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1576 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1577 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1578 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1579 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1580 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1581 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1582 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1583 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1584 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1585 Likewise.
1586 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1587 Likewise.
1588 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1589 Likewise.
1590 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1591 void* to avoid compiler warning.
1592 (main): Likewise.
1593 (cls_struct_align_gn): Mark cif and userdata unused.
1594 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1595 Likewise.
1596 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1597 Likewise.
1598 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1599 Likewise.
1600 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1601 Likewise.
1602 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1603 Likewise.
1604 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1605 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1606 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1607 data unused.
1608 (main): Cast res_call to silence gcc.
1609 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1610 data unused.
1611 (main): Cast res_call to silence gcc.
1612 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1613 and data unused.
1614 (main): Cast res_call to silence gcc.
1615 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1616 data unused.
1617 (main): Cast res_call to silence gcc.
1618 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1619 data unused.
1620 (main): Cast res_call to silence gcc.
1621 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1622 and data unused.
1623 (main): Cast res_call to silence gcc.
1624 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1625 userdata unused.
1626 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1627 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1628 userdata unused.
1629 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1630 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1631 userdata unused.
1632 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1633 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1634 userdata unused.
1635 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1636 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1637 userdata unused.
1638 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1639 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1640 and userdata unused.
1641 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1642 userdata unused.
1643 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1644 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1645 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1646 Cleanup white spaces.
1647 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1648 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1649 cif and userdata unused.
1650 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1651 Likewise.
1652 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1653 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1654 formatters to silence gcc.
1655 (B_gn): Mark cif and userdata unused.
1656 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1657 unused.
1658 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1659 (B_gn): Mark cif and userdata unused.
1660 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1661 unused.
1662 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1663 (B_gn): Mark cif and userdata unused.
1664 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1665 unused.
1666 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1667 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1668 * testsuite/libffi.call/problem1.c (stub): Likewise.
1669 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1670 gcc.
1671 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1672 in the last commit for this test case in the test case itself.
1673 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1674 unused.
1675 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1676 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1677 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1678 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1679 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1680 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1681
0a4e2bb5
KK		 16822006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1683
1684 * src/sh/sysv.S: Fix register numbers in the FDE for
1685 ffi_closure_SYSV.
1686
723e5d5a
AT		 16872006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1688
1689 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1690 declaration to avoid a false negative on ix86. See PR323.
1691
e097f887
KK		 16922006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1693
1694 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1695 and cast integer to void * if needed. Update the pointer to
1696 the FP register saved area correctly.
1697
1a4878f7
AT		 16982006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1699
1700 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1701 is fixed.
1702 * testsuite/libffi.call/nested_struct4.c: Likewise.
1703
b47b5e08
AT		 17042006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1705
6429d58f
AT		 1706 * testsuite/libffi.call/return_dbl.c: New test case.
1707 * testsuite/libffi.call/return_dbl1.c: Likewise.
1708 * testsuite/libffi.call/return_dbl2.c: Likewise.
1709 * testsuite/libffi.call/return_fl.c: Likewise.
1710 * testsuite/libffi.call/return_fl1.c: Likewise.
1711 * testsuite/libffi.call/return_fl2.c: Likewise.
1712 * testsuite/libffi.call/return_fl3.c: Likewise.
1713 * testsuite/libffi.call/closure_fn6.c: Likewise.
1714
b47b5e08
AT		 1715 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1716 definition.
1717 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1718 here to be used by other test cases too.
1719
1720 * testsuite/libffi.call/nested_struct10.c: New test case.
1721 * testsuite/libffi.call/nested_struct9.c: Likewise.
1722 * testsuite/libffi.call/nested_struct8.c: Likewise.
1723 * testsuite/libffi.call/nested_struct7.c: Likewise.
1724 * testsuite/libffi.call/nested_struct6.c: Likewise.
1725 * testsuite/libffi.call/nested_struct5.c: Likewise.
1726 * testsuite/libffi.call/nested_struct4.c: Likewise.
1727
9abe31d4
AT		 17282006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1729
1730 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1731 * configure: Rebuilt.
1732
71aa1a70
JJ		 17332006-01-18 Jakub Jelinek <jakub@redhat.com>
1734
1735 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1736 instead do the shifting inline.
1737 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1738 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1739 and 8 byte structs, for the remaining struct sizes don't call
1740 __lshrdi3, instead do the shifting inline.
1741
8a39029d
TS		 17422005-12-07 Thiemo Seufer <ths@networkno.de>
1743
1744 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1745 missing parentheses.
1746 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1747 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1748 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1749 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1750 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1751 FA_0_0_OFF2.
1752 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1753 endianness bugs.
1754 (ffi_prep_closure): Improve trampoline instruction scheduling.
1755 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1756
cd4241aa
AM		 17572005-12-03 Alan Modra <amodra@bigpond.net.au>
1758
1759 * src/powerpc/ffi.c: Formatting.
1760 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1761 (ffi_prep_args64): Likewise.
1762
0363db46
GK		 17632005-09-30 Geoffrey Keating <geoffk@apple.com>
1764
1765 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1766 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1767
d674eb2f
TT		 17682005-09-26 Tom Tromey <tromey@redhat.com>
1769
1770 * testsuite/libffi.call/float1.c (value_type): New typedef.
1771 (CANARY): New define.
1772 (main): Check for result buffer overflow.
1773 * src/powerpc/linux64.S: Handle linux64 long double returns.
1774 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1775 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1776
2d9c5743
AM		 17772005-08-25 Alan Modra <amodra@bigpond.net.au>
1778
1779 PR target/23404
1780 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1781 homed fp args.
1782 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1783
8a42356f
JJ		 17842005-08-11 Jakub Jelinek <jakub@redhat.com>
1785
1786 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1787 (AH_BOTTOM): Add FFI_HIDDEN definition.
1788 * configure: Rebuilt.
1789 * fficonfig.h.in: Rebuilt.
1790 * src/powerpc/ffi.c (hidden): Remove.
1791 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1792 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1793 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1794 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1795 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1796 add FFI_HIDDEN to its prototype.
1797 (ffi_closure_SYSV_inner): New.
1798 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1799 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1800
863c5508
AS		 18012005-08-10 Alfred M. Szmidt <ams@gnu.org>
1802
1803 PR libffi/21819:
1804 * configure: Rebuilt.
1805 * configure.ac: Handle i*86-*-gnu*.
1806
6a26ea58
JJ		 18072005-08-09 Jakub Jelinek <jakub@redhat.com>
1808
1809 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1810 DW_CFA_offset_extended_sf rather than
1811 DW_CFA_GNU_negative_offset_extended.
1812 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1813
583a92e8
ST		 18142005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1815
1816 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1817 on sh3.
1818 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1819 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1820 partially on register.
1821 (ffi_closure_helper_SYSV): Likewise.
1822 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1823
5d0317fc
KK		 18242005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1825
1826 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1827 Remove empty line.
1828 * src/sh64/ffi.c (simple_type): Remove.
1829 (return_type): Handle small structures correctly.
1830 (ffi_prep_args): Likewise.
1831 (ffi_call): Likewise.
1832 (ffi_closure_helper_SYSV): Likewise.
1833 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1834 Emit position independent code if PIC and remove wrong datalabel
1835 prefixes from EH data.
1836
8d9fd5f5
AT		 18372005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1838
1839 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1840 * Makefile.in: Regenerate.
1841 * include/Makefile.in: Likewise.
1842 * testsuite/Makefile.in: Likewise.
1843 * configure.ac: Add POWERPC_FREEBSD rules.
1844 * configure: Regenerate.
1845 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1846 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1847 * src/powerpc/ffi.c: Add flags to handle small structure returns
1848 in ffi_call_SYSV.
1849 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1850 Aka FFI_SYSV.
1851 (ffi_closure_helper_SYSV): Likewise.
1852 * src/powerpc/ppc_closure.S: Add return types for small structures.
1853 * src/powerpc/sysv.S: Add bits to handle small structures for
1854 final SYSV 4 ABI.
1855
93c76f81
AT		 18562005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1857
1858 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1859 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1860 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1861
b65da304
RC		 18622005-07-05 Randolph Chung <tausq@debian.org>
1863
1864 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1865 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1866 structures. Kill compilation warnings.
1867 (ffi_closure_inner_LINUX): Print return values as hex in debug
1868 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1869 Properly handle 5-7 byte structure returns.
1870 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1871 (FFI_TYPE_SMALL_STRUCT2): Remove.
1872 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1873 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1874 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1875 (checksmst1, checksmst2): Remove.
1876 (checksmst3): Optimize handling of 3-byte struct returns.
1877 (checksmst567): Properly handle 5-7 byte struct returns.
1878
2572f1c6
RO		 18792005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1880
1881 PR libgcj/21943
1882 * src/mips/n32.S: Enforce PIC code.
1883 * src/mips/o32.S: Likewise.
1884
cc786582
RO		 18852005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1886
1887 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1888 * configure: Regenerate.
1889
6d0d376c
AM		 18902005-06-01 Alan Modra <amodra@bigpond.net.au>
1891
1892 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1893 to call ffi_closure_helper_SYSV. Append @local instead.
1894 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1895
a21946f5 18962005-05-17 Kelley Cook <kcook@gcc.gnu.org>
a097ef55
KC		 1897
1898 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1899 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1900 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1901 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1902 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1903
f4c767fd
MS		 19042005-05-09 Mike Stump <mrs@apple.com>
1905
1906 * configure: Regenerate.
1907
7e15b5fe
RH		 19082005-05-08 Richard Henderson <rth@redhat.com>
1909
1910 PR libffi/21285
1911 * src/alpha/osf.S: Update unwind into to match code.
1912
d56ea8d9
RH		 19132005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1914 Richard Henderson <rth@redhat.com>
1915
1916 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1917 bit 11 of flags.
1918 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1919 (ffi_prep_closure): Set carry bit if sse-used flag set.
1920 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1921 Only load sse registers if ssecount non-zero.
1922 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1923
19242005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
d948c201
RC		 1925
1926 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1927 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1928 * configure: Regenerate.
1929
fc0ad8d9
HPN		 19302005-04-20 Hans-Peter Nilsson <hp@axis.com>
1931
1932 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1933 have Tcl8.3-compatible intermediate variable.
1934
614348bb
SP		 19352005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1936 Hans-Peter Nilsson <hp@axis.com>
1937
1938 * Makefile.am: Add CRIS support.
1939 * configure.ac: Likewise.
29ede6db
HPN		 1940 * Makefile.in, configure, testsuite/Makefile.in,
1941 include/Makefile.in: Regenerate.
614348bb
SP		 1942 * src/cris: New directory.
1943 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1944 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
727e729b
HPN		 1945
1946 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1947 \r?\n in output tests.
1948
6dab073b
MS		 19492005-04-12 Mike Stump <mrs@apple.com>
1950
1951 * configure: Regenerate.
1952
51f55d7c
HB		 19532005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1954
1955 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1956
7efa52a3
SE		 19572005-03-30 Steve Ellcey <sje@cup.hp.com>
1958
1959 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1960 (ffi_sarg) Ditto.
1961 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1962 to 64 bits in ILP32 mode.
1963 Load 64 bits even for short data.
1964
1488177a
MS		 19652005-03-23 Mike Stump <mrs@apple.com>
1966
1967 * src/powerpc/darwin.S: Update for -m64 multilib.
1968 * src/powerpc/darwin_closure.S: Likewise.
1969
c6900272
ZW		 19702005-03-21 Zack Weinberg <zack@codesourcery.com>
1971
1972 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1973 Do not set tool_include_dir.
1974 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1975 Regenerate.
1976 * include/Makefile.am: Set gcc_version and toollibffidir.
1977 * include/Makefile.in: Regenerate.
1978
962eea0a
AH		 19792005-02-22 Andrew Haley <aph@redhat.com>
1980
1981 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1982 odd-numbered register pairs for 64-bit integer types.
1983
1b05e8bf
AT		 19842005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1985
1986 PR libffi/20104
1987 * testsuite/libffi.call/return_ll1.c: New test case.
1988
6027770b
JJ		 19892005-02-11 Janis Johnson <janis187@us.ibm.com>
1990
1991 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1992 * testsuite/libffi.call/float.c: Ditto.
1993 * testsuite/libffi.call/float2.c: Ditto.
1994 * testsuite/libffi.call/float3.c: Ditto.
1995
ee31ea68
AT		 19962005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1997
1998 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1999
c16be7d5
EB		 20002005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2001
2002 * testsuite/libffi.special/special.exp (cxx_options): Add
2003 -shared-libgcc.
2004
5df07255
RH		 20052004-12-31 Richard Henderson <rth@redhat.com>
2006
2007 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2008 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2009 offset parameters with a type parameter; deduce size and structure
2010 alignment. Update all users.
2011
81a69b13
RH		 20122004-12-31 Richard Henderson <rth@redhat.com>
2013
2014 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2015 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
c6900272 2016 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
81a69b13
RH		 2017 into ffi_prep_closure.
2018 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2019 from scratch.
2020
d833cfb0
RH		 20212004-12-27 Richard Henderson <rth@redhat.com>
2022
2023 * src/x86/unix64.S: Fix typo in unwind info.
2024
1a0f488c
RH		 20252004-12-25 Richard Henderson <rth@redhat.com>
2026
2027 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2028 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2029 (merge_classes): Check for it.
2030 (SSE_CLASS_P): New.
2031 (classify_argument): Pass byte_offset by value; perform all updates
2032 inside struct case.
2033 (examine_argument): Add classes argument; handle
2034 X86_64_COMPLEX_X87_CLASS.
2035 (ffi_prep_args): Merge into ...
2036 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2037 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2038 (ffi_fill_return_value): Remove.
2039 (ffi_prep_closure): Remove dead assert.
2040 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2041 Rewrite to use struct register_args instead of va_list. Create
2042 flags for handling structure returns.
2043 * src/x86/unix64.S: Remove dead strings.
2044 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2045 stack frame with ffi_call. Handle structure returns properly.
2046 (float2sse, floatfloat2sse, double2sse): Remove.
2047 (sse2float, sse2double, sse2floatfloat): Remove.
2048 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2049 to handle structure returns properly.
2050
7994d50c
DE		 20512004-12-08 David Edelsohn <edelsohn@gnu.org>
2052
2053 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2054 PICFLAG.
2055 * Makefile.in: Regenerated.
2056
afabd8e6
RS		 20572004-12-02 Richard Sandiford <rsandifo@redhat.com>
2058
2059 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2060 * configure, aclocal.m4, Makefile.in: Regenerate.
2061 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2062
2539bffe
KC		 20632004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2064
2065 * configure: Regenerate for libtool change.
2066
767f726e
KC		 20672004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2068
2069 * configure: Regenerate for libtool reversion.
2070
2c92e90f
KC		 20712004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2072
2073 * configure: Regenerate for libtool change.
2074
1df8e834
JDA		 20752004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2076
2077 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2078
20792004-11-23 Richard Sandiford <rsandifo@redhat.com>
2080
2081 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2082 of jal. Use an absolute encoding for the frame information.
2083
20842004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2085
2086 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2087 * acinclude.m4: Delete logic for sincludes.
2088 * aclocal.m4, Makefile.in, configure: Regenerate.
2089 * include/Makefile: Likewise.
2090 * testsuite/Makefile: Likewise.
2091
bf02e7fa
EB		 20922004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2093
2094 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2095 on a 8-byte boundary.
2096 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2097
f20459f1
RE		 20982004-10-27 Richard Earnshaw <rearnsha@arm.com>
2099
2100 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2101 long long values. Round stack allocation to a multiple of 8 bytes
2102 for ATPCS compatibility.
2103 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2104 names. Handle returning long long types. Add Thumb and interworking
2105 support. Improve soft-float code.
2106
fdc69597
RE		 21072004-10-27 Richard Earnshaw <rearnsha@arm.com>
2108
2109 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2110 (libffi_exit): New function.
2111 (libffi_init): Build the testglue wrapper if needed.
2112
1e02510f
EB		 21132004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2114
2115 PR other/18138
2116 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2117
d9a193b2
KI		 21182004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2119
2120 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2121
1fddb058
KK		 21222004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2123
2124 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2125 * testsuite/libffi.call/float3.c: New test case.
2126
e0d3b452
KK		 21272004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2128
2129 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2130 the function returning a structure pointed with R2.
2131 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2132 the structure return value if T bit set. Emit position
2133 independent code and EH data if PIC.
2134
a472e9d4
KI		 21352004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2136
2137 * Makefile.am: Add m32r support.
2138 * configure.ac: Likewise.
2139 * Makefile.in: Regenerate.
2140 * confiugre: Regenerate.
2141 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2142 (uint64, sint64, double, longdouble)
2143 * src/m32r: New directory.
2144 * src/m32r/ffi.c: New file.
2145 * src/m32r/sysv.S: Likewise.
2146 * src/m32r/ffitarget.h: Likewise.
2147
3d47e9ca
KK		 21482004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2149
2150 * testsuite/libffi.call/negint.c: New test case.
2151
fc9051dd
L		 21522004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2153
2154 PR libgcj/17465
2155 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2156 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2157 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2158 DYLD_LIBRARY_PATH.
2159
2010caea
AT		 21602004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2161
2162 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2163 * testsuite/libffi.call/promotion.c: Likewise.
2164 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2165 whitespaces.
2166 * testsuite/libffi.call/return_sc.c: Likewise.
2167 * testsuite/libffi.call/return_uc.c: Likewise.
2168
894be816
AT		 21692004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2170
2171 * src/powerpc/darwin.S: Fix comments and identation.
2172 * src/powerpc/darwin_closure.S: Likewise.
2173
2fe7404a
AT		 21742004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2175
2176 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2177 (ffi_prep_args): Handle longdouble arguments.
2178 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2179 longdouble.
2180 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2181 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2182 values.
2183 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2184 * src/types.c: Defined longdouble size and alignment for darwin.
2185
16070e45
AT		 21862004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2187
2188 * src/powerpc/aix.S: Remove whitespaces.
2189 * src/powerpc/aix_closure.S: Likewise.
2190 * src/powerpc/asm.h: Likewise.
2191 * src/powerpc/ffi.c: Likewise.
2192 * src/powerpc/ffitarget.h: Likewise.
2193 * src/powerpc/linux64.S: Likewise.
2194 * src/powerpc/linux64_closure.S: Likewise.
2195 * src/powerpc/ppc_closure.S: Likewise.
2196 * src/powerpc/sysv.S: Likewise.
2197
88d8c771
AG		 21982004-08-30 Anthony Green <green@redhat.com>
2199
2200 * Makefile.am: Add frv support.
2201 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2202 * configure.ac: Read configure.host.
2203 * configure.in: Read configure.host.
2204 * configure.host: New file. frv-elf needs libgloss.
2205 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
a23adb4d 2206 alignment. This is needed to frv and shouldn't harm the others.
88d8c771
AG		 2207 * include/ffi_common.h (ALIGN_DOWN): New macro.
2208 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2209
4c5f37d5
DD		 22102004-08-24 David Daney <daney@avtrex.com>
2211
2212 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2213 * testsuite/libffi.call/closure_fn1.c: Likewise.
2214 * testsuite/libffi.call/closure_fn2.c Likewise.
2215 * testsuite/libffi.call/closure_fn3.c: Likewise.
2216 * testsuite/libffi.call/closure_fn4.c: Likewise.
2217 * testsuite/libffi.call/closure_fn5.c: Likewise.
2218 * testsuite/libffi.call/cls_18byte.c: Likewise.
2219 * testsuite/libffi.call/cls_19byte.c: Likewise.
2220 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2221 * testsuite/libffi.call/cls_20byte.c: Likewise.
2222 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2223 * testsuite/libffi.call/cls_24byte.c: Likewise.
2224 * testsuite/libffi.call/cls_2byte.c: Likewise.
2225 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2226 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2227 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2228 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2229 * testsuite/libffi.call/cls_4byte.c: Likewise.
2230 * testsuite/libffi.call/cls_64byte.c: Likewise.
2231 * testsuite/libffi.call/cls_6byte.c: Likewise.
2232 * testsuite/libffi.call/cls_7byte.c: Likewise.
2233 * testsuite/libffi.call/cls_8byte.c: Likewise.
2234 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2235 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2236 * testsuite/libffi.call/cls_align_double.c: Likewise.
2237 * testsuite/libffi.call/cls_align_float.c: Likewise.
2238 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2239 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2240 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2241 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2242 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2243 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2244 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2245 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2246 * testsuite/libffi.call/cls_double.c: Likewise.
2247 * testsuite/libffi.call/cls_float.c: Likewise.
2248 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2249 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2250 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2251 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2252 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2253 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2254 * testsuite/libffi.call/cls_schar.c: Likewise.
2255 * testsuite/libffi.call/cls_sint.c: Likewise.
2256 * testsuite/libffi.call/cls_sshort.c: Likewise.
2257 * testsuite/libffi.call/cls_uchar.c: Likewise.
2258 * testsuite/libffi.call/cls_uint.c: Likewise.
2259 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2260 * testsuite/libffi.call/cls_ushort.c: Likewise.
2261 * testsuite/libffi.call/nested_struct.c: Likewise.
2262 * testsuite/libffi.call/nested_struct1.c: Likewise.
2263 * testsuite/libffi.call/nested_struct2.c: Likewise.
2264 * testsuite/libffi.call/nested_struct3.c: Likewise.
2265 * testsuite/libffi.call/problem1.c: Likewise.
2266 * testsuite/libffi.special/unwindtest.cc: Likewise.
2267 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2268 to zero.
2269 * testsuite/libffi.call/cls_16byte.c: Likewise.
2270 * testsuite/libffi.call/cls_5byte.c: Likewise.
2271
39dca114
DD		 22722004-08-23 David Daney <daney@avtrex.com>
2273
2274 PR libgcj/13141
2275 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2276 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2277 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2278 parameters and return types.
2279 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2280 (ffi_prep_closure): Ditto.
2281 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2282 alignment calculations.
2283 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2284 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2285
b790003a
CM		 22862004-08-14 Casey Marshall <csm@gnu.org>
2287
2288 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2289 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2290 integer (O32 ABI only).
2291 (ffi_prep_closure): new function.
2292 (ffi_closure_mips_inner_O32): new function.
c6900272 2293 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
b790003a
CM		 2294 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2295 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2296 64 bit integers correctly.
2297 (ffi_closure_O32): new function.
2298 Added DWARF-2 unwind info for both functions.
2299
547d71a1
AH		 23002004-08-10 Andrew Haley <aph@redhat.com>
2301
2302 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2303
98a5bb3b
RM		 23042004-08-01 Robert Millan <robertmh@gnu.org>
2305
2306 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2307 * configure: Regenerate.
2308
4d447364
MR		 23092004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2310
2311 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2312 and mmap() explicitly instead of relying on preset autoconf cache
2313 variables.
2314 * aclocal.m4: Regenerate.
2315 * configure: Regenerate.
2316
e5dce82f
UW		 23172004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2318
2319 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2320 (ffi_check_float_struct): Remove unused prototype.
2321
e6a0022b
GK		 23222004-06-30 Geoffrey Keating <geoffk@apple.com>
2323
2324 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2325 character on Darwin, use '\n\t' instead.
2326
5a2d52ca
MK		 23272004-06-26 Matthias Klose <doko@debian.org>
2328
2329 * libtool-version: Fix typo in revision/age.
2330
401844ef
MK		 23312004-06-17 Matthias Klose <doko@debian.org>
2332
2333 * libtool-version: New.
2334 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2335 * Makefile.in: Regenerate.
2336
94632047 23372004-06-15 Paolo Bonzini <bonzini@gnu.org>
076e5443
PB		 2338
2339 * Makefile.am: Remove useless multilib rules.
2340 * Makefile.in: Regenerate.
2341 * aclocal.m4: Regenerate with automake 1.8.5.
2342 * configure.ac: Remove useless multilib configury.
2343 * configure: Regenerate.
2344
94632047 23452004-06-15 Paolo Bonzini <bonzini@gnu.org>
9ac926b1
PB		 2346
2347 * .cvsignore: New file.
2348
37c849e4
JJ		 23492004-06-10 Jakub Jelinek <jakub@redhat.com>
2350
2351 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2352 fp_done.
2353 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2354 changed from 8.
2355
cb623172
SM		 23562004-06-06 Sean McNeil <sean@mcneil.com>
2357
2358 * configure.ac: Add x86_64-*-freebsd* support.
2359 * configure: Regenerate.
2360
c9b93377
JB		 23612004-04-26 Joe Buck <jbuck@welsh-buck.org>
2362
2363 Bug 15093
2364 * configure.ac: Test for existence of mmap and sys/mman.h before
2365 checking blacklist. Fix suggested by Jim Wilson.
2366 * configure: Regenerate.
2367
cf90fd1b
MA		 23682004-04-26 Matt Austern <austern@apple.com>
2369
2370 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2371 FDE location.
2372 * src/powerpc/darwin_closure.S: Likewise.
c6900272 2373
67781b44
AT		 23742004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2375
2376 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2377 error. Reported by Thomas Heller <theller@python.net>.
2378 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2379 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2380
14b078a8
MK		 23812004-03-20 Matthias Klose <doko@debian.org>
2382
2383 * src/pa/linux.S: Fix typo.
2384
d65a810f
MK		 23852004-03-19 Matthias Klose <doko@debian.org>
2386
67781b44 2387 * Makefile.am: Update.
d65a810f
MK		 2388 * Makefile.in: Regenerate.
2389 * src/pa/ffi.h.in: Remove.
2390 * src/pa/ffitarget.h: New file.
2391
23922004-02-10 Randolph Chung <tausq@debian.org>
2393
2394 * Makefile.am: Add PA support.
2395 * Makefile.in: Regenerate.
2396 * include/Makefile.in: Regenerate.
2397 * configure.ac: Add PA target.
2398 * configure: Regenerate.
2399 * src/pa/ffi.c: New file.
2400 * src/pa/ffi.h.in: Add PA support.
2401 * src/pa/linux.S: New file.
2402 * prep_cif.c: Add PA support.
2403
09cababc
HY		 24042004-03-16 Hosaka Yuji <hos@tamanegi.org>
2405
2406 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2407 double.
2408 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2409 with ecif->cif->flags.
2410 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2411 with cif->flags.
2412 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2413 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2414 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2415 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2416
0e8d0c5f
KC		 24172004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2418
2419 * configure.in: Rename file to ...
2420 * configure.ac: ... this.
2421 * fficonfig.h.in: Regenerate.
2422 * Makefile.in: Regenerate.
2423 * include/Makefile.in: Regenerate.
2424 * testsuite/Makefile.in: Regenerate.
2425
b8d50739
MA		 24262004-03-12 Matt Austern <austern@apple.com>
2427
2428 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2429 changes in EH format resulting from addition of linkonce support.
2430 * src/powerpc/darwin_closure.S: Likewise.
09cababc 2431
3ed8e5b6
AT		 24322004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2433 Paolo Bonzini <bonzini@gnu.org>
2434
2435 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2436 Remove VPATH. Remove rules for object files. Remove multilib support.
2437 (AM_CCASFLAGS): Add.
2438 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2439 (AC_PREREQ): Bump version to 2.59.
2440 (AC_INIT): Fill with version info and bug address.
2441 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2442 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2443 De-precious CC so that the right flags are passed down to multilibs.
2444 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2445 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2446 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2447 * configure: Rebuilt.
2448 * aclocal.m4: Likewise.
2449 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2450 * fficonfig.h.in: Likewise.
2451
254ea477
AS		 24522004-03-11 Andreas Schwab <schwab@suse.de>
2453
2454 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2455 arguments from fp registers only for the first 8 parameter slots.
2456 Don't convert a float parameter when passed in memory.
2457
3beb1315
HPN		 24582004-03-09 Hans-Peter Nilsson <hp@axis.com>
2459
2460 * configure: Regenerate for config/accross.m4 correction.
2461
fecf735c
MK		 24622004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2463
2464 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2465 ecif->cif->bytes to bytes.
2466 (ffi_prep_cif_machdep): Add braces around nested if statement.
2467
4243752c
AM		 24682004-02-09 Alan Modra <amodra@bigpond.net.au>
2469
12001e0f
AM		 2470 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2471
4243752c
AM		 2472 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2473 (ffi_closure_helper_LINUX64): Fix typo.
2474 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2475 for powerpc64-*-*.
2476 * testsuite/libffi.call/float.c: Likewise.
2477 * testsuite/libffi.call/float2.c: Likewise.
2478
bf310028
AM		 24792004-02-08 Alan Modra <amodra@bigpond.net.au>
2480
2481 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2482 long double function return and long double arg handling.
2483 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2484 Use "end_pfr" instead of "nf". Correct long double handling.
2485 Localise "temp".
2486 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2487 return value.
2488 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2489 space for long double return value. Adjust stack frame and offsets.
2490 Load f2 long double return.
2491
dd5f9ea2
AM		 24922004-02-07 Alan Modra <amodra@bigpond.net.au>
2493
2494 * src/types.c: Use 16 byte long double for POWERPC64.
2495
0ce78f01
EB		 24962004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2497
2498 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2499 when the structure return address is passed in %o0.
2500 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2501 (ffi_v9_layout_struct): Align the field following a nested structure
2502 on a word boundary. Use memmove instead of memcpy.
2503 (ffi_call): Update call to ffi_V9_return_struct.
2504 (ffi_prep_closure): Define 'ctx' only for V8.
2505 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2506 and ffi_closure_sparc_inner_v9.
2507 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2508 Always skip the structure return address. For structures and long
2509 doubles, copy the argument directly.
2510 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2511 if required. Shift the maximum floating-point slot accordingly. For
2512 big structures, copy the argument directly; otherwise, left-justify the
2513 argument and call ffi_v9_layout_struct to lay out the structure on
2514 the stack.
2515 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2516 (ffi_closure_v8): Pass the structure return address. Update call to
2517 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2518 Skip the 'unimp' insn when returning long doubles and structures.
2519 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2520 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2521 FFI_TYPE_INT handling. Load structures both in integers and
2522 floating-point registers on return.
2523 * README: Update status of the SPARC port.
2524
7fcc1d9b
AT		 25252004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2526
2527 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2528 as of type ffi_arg.
2529 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2530
6ed0ecaf
UW		 25312004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2532
2533 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2534 value as of type ffi_arg, not unsigned int.
2535
c930d8a5
MR		 25362004-01-21 Michael Ritzert <ritzert@t-online.de>
2537
2538 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2539 of the LHS.
2540
7b2e3e60
AT		 25412004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2542
2543 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2544 Solaris.
2545
2e89756b
RO		 25462004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2547
2548 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2549 to void *.
2550
628b26f8
RH		 25512003-12-10 Richard Henderson <rth@redhat.com>
2552
2553 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2554 size_t instead of int.
2555
0ee646d6
HY		 25562003-12-04 Hosaka Yuji <hos@tamanegi.org>
2557
2558 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2559 * testsuite/libffi.call/many_win32.c (main): Replace variable
2560 int i with unsigned long ul.
2561
2562 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2563 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2564 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2565 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2566 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2567 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2568 * testsuite/libffi.call/cls_align_float.c: Likewise.
2569 * testsuite/libffi.call/cls_align_double.c: Likewise.
2570 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2571 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2572
2c4ed456
HY		 25732003-12-02 Hosaka Yuji <hos@tamanegi.org>
2574
2575 PR other/13221
2576 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2577 Align arguments to 32 bits.
2578
d18c3386
AT		 25792003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2580
2581 PR other/13221
2582 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2583 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2584 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2585 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2586 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2587 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2588
2589 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2590
88096b2a
KG		 25912003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2592
2593 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2594 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2595
3eee102e
AT		 25962003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2597
2598 * Makefile.in: Rebuilt.
2599 * configure: Likewise.
2600 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2601 the right type.
2602
5196736e
AJ		 26032003-11-21 Andreas Jaeger <aj@suse.de>
2604 Andreas Tobler <a.tobler@schweiz.ch>
2605
2606 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2607 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2608 * Makefile.in: Rebuilt.
2609 * aclocal.m4: Likewise.
2610 * configure: Likewise.
2611 * fficonfig.h.in: Likewise.
2612 * testsuite/lib/libffi-dg.exp: Add include dir.
2613 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2614 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2615 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2616 for ffi_closure if available.
2617 * testsuite/libffi.call/closure_fn1.c: Likewise.
2618 * testsuite/libffi.call/closure_fn2.c: Likewise.
2619 * testsuite/libffi.call/closure_fn3.c: Likewise.
2620 * testsuite/libffi.call/closure_fn4.c: Likewise.
2621 * testsuite/libffi.call/closure_fn5.c: Likewise.
2622 * testsuite/libffi.call/cls_12byte.c: Likewise.
2623 * testsuite/libffi.call/cls_16byte.c: Likewise.
2624 * testsuite/libffi.call/cls_18byte.c: Likewise.
2625 * testsuite/libffi.call/cls_19byte.c: Likewise.
2626 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2627 * testsuite/libffi.call/cls_20byte.c: Likewise.
2628 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2629 * testsuite/libffi.call/cls_24byte.c: Likewise.
2630 * testsuite/libffi.call/cls_2byte.c: Likewise.
2631 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2632 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2633 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2634 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2635 * testsuite/libffi.call/cls_4byte.c: Likewise.
2636 * testsuite/libffi.call/cls_5byte.c: Likewise.
2637 * testsuite/libffi.call/cls_64byte.c: Likewise.
2638 * testsuite/libffi.call/cls_6byte.c: Likewise.
2639 * testsuite/libffi.call/cls_7byte.c: Likewise.
2640 * testsuite/libffi.call/cls_8byte.c: Likewise.
2641 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2642 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2643 * testsuite/libffi.call/cls_double.c: Likewise.
2644 * testsuite/libffi.call/cls_float.c: Likewise.
2645 * testsuite/libffi.call/cls_schar.c: Likewise.
2646 * testsuite/libffi.call/cls_sint.c: Likewise.
2647 * testsuite/libffi.call/cls_sshort.c: Likewise.
2648 * testsuite/libffi.call/cls_uchar.c: Likewise.
2649 * testsuite/libffi.call/cls_uint.c: Likewise.
2650 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2651 * testsuite/libffi.call/cls_ushort.c: Likewise.
2652 * testsuite/libffi.call/nested_struct.c: Likewise.
2653 * testsuite/libffi.call/nested_struct1.c: Likewise.
2654 * testsuite/libffi.call/nested_struct2.c: Likewise.
2655 * testsuite/libffi.call/nested_struct3.c: Likewise.
2656 * testsuite/libffi.call/problem1.c: Likewise.
2657 * testsuite/libffi.special/unwindtest.cc: Likewise.
2658
9e6c3ecb
AT		 26592003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2660
2661 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2662
70d8f2a1
AT		 26632003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2664
2665 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2666 Add -lgcc_s to additional flags.
2667
ea81d2a3
AT		 26682003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2669
2670 * configure.in, include/Makefile.am: PR libgcj/11147, install
2671 the ffitarget.h header file in a gcc versioned and target
2672 dependent place.
2673 * configure: Regenerated.
2674 * Makefile.in, include/Makefile.in: Likewise.
2675 * testsuite/Makefile.in: Likewise.
2676
647af4a9
AT		 26772003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2678
2679 * testsuite/libffi.call/closure_fn0.c: Print result and check
2680 with dg-output to make debugging easier.
2681 * testsuite/libffi.call/closure_fn1.c: Likewise.
2682 * testsuite/libffi.call/closure_fn2.c: Likewise.
2683 * testsuite/libffi.call/closure_fn3.c: Likewise.
2684 * testsuite/libffi.call/closure_fn4.c: Likewise.
2685 * testsuite/libffi.call/closure_fn5.c: Likewise.
2686 * testsuite/libffi.call/cls_12byte.c: Likewise.
2687 * testsuite/libffi.call/cls_16byte.c: Likewise.
2688 * testsuite/libffi.call/cls_18byte.c: Likewise.
2689 * testsuite/libffi.call/cls_19byte.c: Likewise.
2690 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2691 * testsuite/libffi.call/cls_20byte.c: Likewise.
2692 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2693 * testsuite/libffi.call/cls_24byte.c: Likewise.
2694 * testsuite/libffi.call/cls_2byte.c: Likewise.
2695 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2696 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2697 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2698 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2699 * testsuite/libffi.call/cls_4byte.c: Likewise.
2700 * testsuite/libffi.call/cls_5byte.c: Likewise.
2701 * testsuite/libffi.call/cls_64byte.c: Likewise.
2702 * testsuite/libffi.call/cls_6byte.c: Likewise.
2703 * testsuite/libffi.call/cls_7byte.c: Likewise.
2704 * testsuite/libffi.call/cls_8byte.c: Likewise.
2705 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2706 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2707 * testsuite/libffi.call/cls_double.c: Likewise.
2708 * testsuite/libffi.call/cls_float.c: Likewise.
2709 * testsuite/libffi.call/cls_schar.c: Likewise.
2710 * testsuite/libffi.call/cls_sint.c: Likewise.
2711 * testsuite/libffi.call/cls_sshort.c: Likewise.
2712 * testsuite/libffi.call/cls_uchar.c: Likewise.
2713 * testsuite/libffi.call/cls_uint.c: Likewise.
2714 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2715 * testsuite/libffi.call/cls_ushort.c: Likewise.
2716 * testsuite/libffi.call/problem1.c: Likewise.
2717
2718 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2719 static.
2720
ac47cc13
AT		 27212003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2722
2723 * testsuite/libffi.call/cls_9byte2.c: New test case.
2724 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2725 * testsuite/libffi.call/cls_64byte.c: Likewise.
2726 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2727 * testsuite/libffi.call/cls_19byte.c: Likewise.
2728 * testsuite/libffi.call/cls_18byte.c: Likewise.
2729 * testsuite/libffi.call/closure_fn4.c: Likewise.
2730 * testsuite/libffi.call/closure_fn5.c: Likewise.
2731 * testsuite/libffi.call/cls_schar.c: Likewise.
2732 * testsuite/libffi.call/cls_sint.c: Likewise.
2733 * testsuite/libffi.call/cls_sshort.c: Likewise.
2734 * testsuite/libffi.call/nested_struct2.c: Likewise.
2735 * testsuite/libffi.call/nested_struct3.c: Likewise.
2736
f145ed6b
AT		 27372003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2738
2739 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2740 * testsuite/libffi.call/cls_uchar.c: Likewise.
2741 * testsuite/libffi.call/cls_uint.c: Likewise.
2742 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2743 * testsuite/libffi.call/cls_ushort.c: Likewise.
2744 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2745
2b04d015
AT		 27462003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2747
2748 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2749 the initialization.
2750
57f542b3
AT		 27512003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2752
2b04d015 2753 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
57f542b3
AT		 2754 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2755
e61acf72
DD		 27562003-10-22 David Daney <ddaney@avtrex.com>
2757
2758 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2759 __attribute__((__mode__(__SI__))) and friends.
2760
2461c08b
AS		 27612003-10-22 Andreas Schwab <schwab@suse.de>
2762
2763 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2764
1450eb7a
AT		 27652003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2766
2767 * configure.in: AC_LINK_FILES(ffitarget.h).
2768 * configure: Regenerate.
2769 * Makefile.in: Likewise.
2770 * include/Makefile.in: Likewise.
2771 * testsuite/Makefile.in: Likewise.
2772 * fficonfig.h.in: Likewise.
2773
27742003-10-21 Paolo Bonzini <bonzini@gnu.org>
2775 Richard Henderson <rth@redhat.com>
2776
2777 Avoid that ffi.h includes fficonfig.h.
2778
2779 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2780 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2781 (TARGET_SRC_MIPS_SGI): Removed.
2782 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2783 (MIPS_SGI): Removed.
2784 (CLEANFILES): Removed.
2785 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2786 targets.
2787 * acconfig.h: Removed.
2788 * configure.in: Compute sizeofs only for double and long double.
2789 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2790 into AC_DEFINE instead of using acconfig.h. Create
2791 include/ffitarget.h instead of include/fficonfig.h. Rename
2792 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2793 AC_DEFINE EH_FRAME_FLAGS.
2794 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2795 (hack_DATA): Add ffitarget.h.
2796 * include/ffi.h.in: Remove all system specific definitions.
2797 Declare raw API even if it is not installed, why bother?
2798 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2799 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2800 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2801 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2802 * include/ffi_common.h (bool): Do not define.
2803 (ffi_assert): Accept failed assertion.
2804 (ffi_type_test): Return void and accept file/line.
2805 (FFI_ASSERT): Pass stringized failed assertion.
2806 (FFI_ASSERT_AT): New macro.
2807 (FFI_ASSERT_VALID_TYPE): New macro.
2808 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2809 UINT64, SINT64): Define here with gcc's __attribute__ macro
2810 instead of in ffi.h
2811 (FLOAT32, ALIGN): Define here instead of in ffi.h
2812 * include/ffi-mips.h: Removed. Its content moved to
2813 src/mips/ffitarget.h after separating assembly and C sections.
2814 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2815 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2816 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2817 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2818 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2819 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2820 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2821 (ffi_type_test): Rewritten.
2822 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2823 FFI_ASSERT_VALID_TYPE.
2824 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2825 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2826 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2827 src/s390/ffitarget.h, src/sh/ffitarget.h,
2828 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2829 src/x86/ffitarget.h: New files.
2830 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2831 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2832 src/powerpc/aix.S, src/powerpc/darwin.S,
2833 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2834 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2835 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2836 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2837 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2838 include fficonfig.h
2839
27d54b2a
RO		 28402003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2841
2842 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2843 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2844
0df6970d
AT		 28452003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2846
2847 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2848 Used when FFI_DEBUG = 1.
2849
2612ffce
AM		 28502003-10-14 Alan Modra <amodra@bigpond.net.au>
2851
2852 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2853 and align.
2854
d54f7354
RO		 28552003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2856
2857 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2858 FFI_MIPS_O32 for O32 ABI.
2859
7a54c850
AT		 28602003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2861
2862 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2863 SPARC64. Cleanup whitespaces.
2864
9e3bfa9b
AT		 28652003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2866
2867 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2868 strongarm, xscale. Cleanup whitespaces.
2869 * testsuite/libffi.call/closure_fn1.c: Likewise.
2870 * testsuite/libffi.call/closure_fn2.c: Likewise.
2871 * testsuite/libffi.call/closure_fn3.c: Likewise.
2872 * testsuite/libffi.call/cls_12byte.c: Likewise.
2873 * testsuite/libffi.call/cls_16byte.c: Likewise.
2874 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2875 * testsuite/libffi.call/cls_20byte.c: Likewise.
2876 * testsuite/libffi.call/cls_24byte.c: Likewise.
2877 * testsuite/libffi.call/cls_2byte.c: Likewise.
2878 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2879 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2880 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2881 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2882 * testsuite/libffi.call/cls_4byte.c: Likewise.
2883 * testsuite/libffi.call/cls_5byte.c: Likewise.
2884 * testsuite/libffi.call/cls_6byte.c: Likewise.
2885 * testsuite/libffi.call/cls_7byte.c: Likewise.
2886 * testsuite/libffi.call/cls_8byte.c: Likewise.
2887 * testsuite/libffi.call/cls_double.c: Likewise.
2888 * testsuite/libffi.call/cls_float.c: Likewise.
2889 * testsuite/libffi.call/cls_uchar.c: Likewise.
2890 * testsuite/libffi.call/cls_uint.c: Likewise.
2891 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2892 * testsuite/libffi.call/cls_ushort.c: Likewise.
2893 * testsuite/libffi.call/nested_struct.c: Likewise.
2894 * testsuite/libffi.call/nested_struct1.c: Likewise.
2895 * testsuite/libffi.call/problem1.c: Likewise.
2896 * testsuite/libffi.special/unwindtest.cc: Likewise.
2897 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2898
3e86b595
DE		 28992003-09-18 David Edelsohn <edelsohn@gnu.org>
2900
2901 * src/powerpc/aix.S: Cleanup whitespaces.
2902 * src/powerpc/aix_closure.S: Likewise.
2903
8d8369f8
AT		 29042003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2905
2906 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2907 * src/powerpc/darwin_closure.S: Likewise.
2908 * src/powerpc/ffi_darwin.c: Likewise.
2909
69b182fd
AT		 29102003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2911 David Edelsohn <edelsohn@gnu.org>
2912
2913 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2914 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2915 parameter stack.
2916 * src/powerpc/darwin_closure.S: Likewise.
2917 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2918 according to the Darwin/AIX ABI.
2919 (ffi_prep_cif_machdep): Likewise.
2920 (ffi_closure_helper_DARWIN): Likewise.
2921 Remove the outgoing parameter stack logic. Simplify the evaluation
2922 of the different CASE types.
2923 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2924 statement in the trampoline code.
2925
df805cdc
KK		 29262003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2927
2928 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2929 for the register size.
2930 (ffi_closure_helper_SYSV): Handle the structure return value
2931 address correctly.
2932 (ffi_closure_helper_SYSV): Return the appropriate type when
2933 the registers are used for the structure return value.
2934 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2935 the 64-bit return value. Update copyright years.
2936
eb26c76c
RO		 29372003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2938
2939 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2940 srcdir for ffi_mips.h.
2941
6350c02c
AM		 29422003-09-12 Alan Modra <amodra@bigpond.net.au>
2943
2944 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2945 structure size.
2946 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2947 placement of float result.
2948 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2949 cast of "resp" for big-endian 64 bit machines.
2950
5af1c806
AM		 29512003-09-11 Alan Modra <amodra@bigpond.net.au>
2952
2953 * src/types.c (double, longdouble): Merge identical SH and ARM
2954 typedefs, and add POWERPC64.
2955 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2956 struct split over gpr and rest.
2957 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2958 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2959
086a4bd7
AT		 29602003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2961
2962 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2963 passing correctly.
c930d8a5 2964
e517f636
AM		 29652003-09-09 Alan Modra <amodra@bigpond.net.au>
2966
2967 * configure: Regenerate.
2968
6f20d627
AT		 29692003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2970
2971 * Makefile.am: Remove build rules for ffitest.
2972 * Makefile.in: Rebuilt.
c930d8a5 2973
6f20d627 29742003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
226660af
AT		 2975
2976 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2977 about implicit declaration of abort().
2978
bbf3057b
AT		 29792003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2980
2981 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2982 * Makefile.in: Rebuilt.
2983 * configure.in: Add dejagnu test framework.
2984 * configure: Rebuilt.
2985
2986 * testsuite/Makefile.am: New file.
2987 * testsuite/Makefile.in: Built
c930d8a5 2988 * testsuite/lib/libffi-dg.exp: New file.
bbf3057b
AT		 2989 * testsuite/config/default.exp: Likewise.
2990 * testsuite/libffi.call/call.exp: Likewise.
2991 * testsuite/libffi.call/ffitest.h: Likewise.
2992 * testsuite/libffi.call/closure_fn0.c: Likewise.
2993 * testsuite/libffi.call/closure_fn1.c: Likewise.
2994 * testsuite/libffi.call/closure_fn2.c: Likewise.
2995 * testsuite/libffi.call/closure_fn3.c: Likewise.
2996 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2997 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2998 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2999 * testsuite/libffi.call/cls_2byte.c: Likewise.
3000 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3001 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3002 * testsuite/libffi.call/cls_4byte.c: Likewise.
3003 * testsuite/libffi.call/cls_5byte.c: Likewise.
3004 * testsuite/libffi.call/cls_6byte.c: Likewise.
3005 * testsuite/libffi.call/cls_7byte.c: Likewise.
3006 * testsuite/libffi.call/cls_8byte.c: Likewise.
3007 * testsuite/libffi.call/cls_12byte.c: Likewise.
3008 * testsuite/libffi.call/cls_16byte.c: Likewise.
3009 * testsuite/libffi.call/cls_20byte.c: Likewise.
3010 * testsuite/libffi.call/cls_24byte.c: Likewise.
3011 * testsuite/libffi.call/cls_double.c: Likewise.
3012 * testsuite/libffi.call/cls_float.c: Likewise.
3013 * testsuite/libffi.call/cls_uchar.c: Likewise.
3014 * testsuite/libffi.call/cls_uint.c: Likewise.
3015 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3016 * testsuite/libffi.call/cls_ushort.c: Likewise.
3017 * testsuite/libffi.call/float.c: Likewise.
3018 * testsuite/libffi.call/float1.c: Likewise.
3019 * testsuite/libffi.call/float2.c: Likewise.
3020 * testsuite/libffi.call/many.c: Likewise.
3021 * testsuite/libffi.call/many_win32.c: Likewise.
3022 * testsuite/libffi.call/nested_struct.c: Likewise.
3023 * testsuite/libffi.call/nested_struct1.c: Likewise.
3024 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3025 * testsuite/libffi.call/problem1.c: Likewise.
3026 * testsuite/libffi.call/promotion.c: Likewise.
3027 * testsuite/libffi.call/return_ll.c: Likewise.
3028 * testsuite/libffi.call/return_sc.c: Likewise.
3029 * testsuite/libffi.call/return_uc.c: Likewise.
3030 * testsuite/libffi.call/strlen.c: Likewise.
3031 * testsuite/libffi.call/strlen_win32.c: Likewise.
3032 * testsuite/libffi.call/struct1.c: Likewise.
3033 * testsuite/libffi.call/struct2.c: Likewise.
3034 * testsuite/libffi.call/struct3.c: Likewise.
3035 * testsuite/libffi.call/struct4.c: Likewise.
3036 * testsuite/libffi.call/struct5.c: Likewise.
3037 * testsuite/libffi.call/struct6.c: Likewise.
3038 * testsuite/libffi.call/struct7.c: Likewise.
3039 * testsuite/libffi.call/struct8.c: Likewise.
c930d8a5 3040 * testsuite/libffi.call/struct9.c: Likewise.
bbf3057b
AT		 3041 * testsuite/libffi.special/special.exp: New file.
3042 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3043 * testsuite/libffi.special/unwindtest.cc: Likewise.
3044
c930d8a5 3045
996b4f8f
KK		 30462003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3047
3048 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3049 copyright years.
3050
b00badcd
AM		 30512003-08-02 Alan Modra <amodra@bigpond.net.au>
3052
3053 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3054 structure passing.
3055 (ffi_closure_helper_LINUX64): Likewise.
3056 * src/powerpc/linux64.S: Remove code writing to parm save area.
3057 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3058 address in lr from ffi_closure_helper_LINUX64 call to calculate
3059 table address. Optimize function tail.
3060
c1516eae
AT		 30612003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3062
3063 * src/sparc/ffi.c: Handle all floating point registers.
3064 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3065
fc0e6222
GP		 30662003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3067
3068 * README: Note that libffi is not part of GCC. Update the project
3069 URL and status.
3070
a1569df8
FS		 30712003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3072
3073 * src/powerpc/ppc_closure.S: Include ffi.h.
3074
12e816bf
RO		 30752003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3076
3077 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3078 Use C style comments.
3079
e860ed6d
KK		 30802003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3081
3082 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3083 * Makefile.in: Regenerate.
3084 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3085 * configure: Regenerate.
3086 * include/ffi.h.in: Add SHmedia support.
3087 * src/sh64/ffi.c: New file.
3088 * src/sh64/sysv.S: New file.
3089
a733b15e
JJ		 30902003-05-16 Jakub Jelinek <jakub@redhat.com>
3091
3092 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3093 should be read-only.
3094 * configure: Rebuilt.
3095 * fficonfig.h.in: Rebuilt.
3096 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3097 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3098 * src/powerpc/linux64.S: Likewise.
3099 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3100 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3101 if -fpic/-fPIC/-mrelocatable.
3102 * src/powerpc/powerpc_closure.S: Likewise.
3103 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3104 #write in .eh_frame flags.
3105 * src/sparc/v9.S: Likewise.
3106 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3107 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3108 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3109
bd19c9f4
JS		 31102003-05-07 Jeff Sturm <jsturm@one-point.com>
3111
3112 Fixes PR bootstrap/10656
3113 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3114 support for .register pseudo-op.
3115 * src/sparc/v8.S: Use it.
3116 * fficonfig.h.in: Rebuilt.
3117 * configure: Rebuilt.
3118
e9b84181
JJ		 31192003-04-18 Jakub Jelinek <jakub@redhat.com>
3120
3121 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3122 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3123 Make it the default on POWERPC64.
3124 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3125 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3126 * configure: Rebuilt.
3127 * src/powerpc/ffi.c (hidden): Define.
3128 (ffi_prep_args_SYSV): Renamed from
3129 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3130 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3131 ASM_NEEDS_REGISTERS64): New.
3132 (ffi_prep_args64): New function.
3133 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3134 (ffi_call): Likewise.
3135 (ffi_prep_closure): Likewise.
3136 (flush_icache): Surround by #ifndef POWERPC64.
3137 (ffi_dblfl): New union type.
3138 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3139 (ffi_closure_helper_LINUX64): New function.
3140 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3141 __powerpc64__.
3142 * src/powerpc/sysv.S: Likewise.
3143 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3144 * src/powerpc/linux64.S: New file.
3145 * src/powerpc/linux64_closure.S: New file.
3146 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3147 src/powerpc/linux64_closure.S.
3148 (TARGET_SRC_POWERPC): Likewise.
3149
3150 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3151 closure_test_fn3): Fix result printing on big-endian 64-bit
3152 machines.
3153 (main): Print tst2_arg instead of uninitialized tst2_result.
3154
3155 * src/ffitest.c (main): Hide what closure pointer really points to
3156 from the compiler.
3157
908a31e1
RE		 31582003-04-16 Richard Earnshaw <rearnsha@arm.com>
3159
3160 * configure.in (arm-*-netbsdelf*): Add configuration.
3161 (configure): Regenerated.
3162
41a73c4c
LR		 31632003-04-04 Loren J. Rittle <ljrittle@acm.org>
3164
3165 * include/Makefile.in: Regenerate.
3166
cd41c847
ZD		 31672003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3168
3169 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3170 bit mode.
3171 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3172 Receive closure pointer through parameter, read args using
3173 __builtin_dwarf_cfa.
3174 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3175
6db450f9
AS		 31762003-03-12 Andreas Schwab <schwab@suse.de>
3177
3178 * configure.in: Avoid trailing /. in toolexeclibdir.
3179 * configure: Rebuilt.
3180
70144e9c
AT		 31812003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3182
c6900272 3183 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
70144e9c 3184
79e43f89
AT		 31852003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3186
3187 * libffi/src/powerpc/darwin_closure.S:
3188 Fix alignement bug, allocate 8 bytes for the result.
c930d8a5 3189 * libffi/src/powerpc/aix_closure.S:
79e43f89
AT		 3190 Likewise.
3191 * libffi/src/powerpc/ffi_darwin.c:
3192 Update stackframe description for aix/darwin_closure.S.
c930d8a5 3193
0fcc9e55
JJ		 31942003-02-06 Jakub Jelinek <jakub@redhat.com>
3195
3196 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3197 attribute.
3198
63b87c2e
CC		 31992003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3200 Andreas Schwab <schwab@suse.de>
3201
c6900272 3202 * configure.in: Adjust command to source config-ml.in to account
63b87c2e
CC		 3203 for changes to the libffi_basedir definition.
3204 (libffi_basedir): Remove ${srcdir} from value and include trailing
3205 slash if nonempty.
c930d8a5 3206
c6900272 3207 * configure: Regenerate.
63b87c2e 3208
c33402e2
FS		 32092003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3210
3211 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3212
fd2e49b2
AH		 32132003-01-28 Andrew Haley <aph@redhat.com>
3214
3215 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3216 * src/x86/ffi64.c (ffi_prep_closure): New.
3217 (ffi_closure_UNIX64_inner): New.
3218 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3219
0e292c01
AO		 32202003-01-27 Alexandre Oliva <aoliva@redhat.com>
3221
3222 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3223 Remove USE_LIBDIR conditional.
3224 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3225 * Makefile.in, configure: Rebuilt.
3226
acabc220
DE		 32272003-01027 David Edelsohn <edelsohn@gnu.org>
3228
3229 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3230 * Makefile.in: Regenerate.
3231
be0acd42
AH		 32322003-01-22 Andrew Haley <aph@redhat.com>
3233
3234 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3235 unwind info.
3236
ed695c8e
AT		 32372003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3238
3239 * src/powerpc/darwin.S: Add unwind info.
3240 * src/powerpc/darwin_closure.S: Likewise.
3241
45ba15f4
AH		 32422003-01-14 Andrew Haley <aph@redhat.com>
3243
3244 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3245 (ffi_prep_cif_machdep): Likewise.
3246 * src/x86/unix64.S: Add unwind info.
3247
4b5f402e
AJ		 32482003-01-14 Andreas Jaeger <aj@suse.de>
3249
3250 * src/ffitest.c (main): Only use ffi_closures if those are
3251 supported.
3252
7b5102af
TT		 32532003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3254
3255 * libffi/src/ffitest.c
c6900272 3256 add closure testcases
7b5102af
TT		 3257
32582003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3259
3260 * libffi/src/powerpc/ffi.c
c6900272 3261 fix alignment bug for float (4 byte aligned iso 8 byte)
7b5102af 3262
5f7c78d8
GK		 32632003-01-09 Geoffrey Keating <geoffk@apple.com>
3264
3265 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3266 * src/powerpc/darwin.S: Remove RCS version string.
3267
c75c7793
JS		 32682003-01-03 Jeff Sturm <jsturm@one-point.com>
3269
3270 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3271 * src/ffitest.c (main): Use static storage for closure.
3272 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3273 * src/sparc/v8.S (ffi_closure_v8): New.
3274 * src/sparc/v9.S (ffi_closure_v9): New.
3275
eb3c46a1
RM		 32762002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3277
3278 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3279 enumeration for X86_WIN32.
3280 * src/x86/win32.S: Added ffi_call_STDCALL function
3281 definition.
3282 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3283 switch cases for recognising FFI_STDCALL and
3284 calling ffi_call_STDCALL if target is X86_WIN32.
3285 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3286 stdcall versions of the "my_strlen" and "many"
3287 test functions (for X86_WIN32).
3288 Added test cases to test stdcall invocation using
3289 these functions.
3290
4977bab6
ZW		 32912002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3292
3293 * src/sh/sysv.S: Add DWARF2 unwind info.
3294
67333e84
UW		 32952002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3296
3297 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3298
bc3cf760
JW		 32992002-11-26 Jim Wilson <wilson@redhat.com>
3300
3301 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3302
4977bab6
ZW		 33032002-11-23 H.J. Lu <hjl@gnu.org>
3304
3305 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3306 Include ../config/accross.m4.
3307 * aclocal.m4; Rebuild.
3308 * configure: Likewise.
3309
18789f4e
UW		 33102002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3311
3312 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3313
6bee5372
DD		 33142002-11-11 DJ Delorie <dj@redhat.com>
3315
3316 * configure.in: Look for common files in the right place.
3317
8177895b
UW		 33182002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3319
3320 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3321 raw data as _Jv_word values, not ffi_raw.
3322 (ffi_java_ptrarray_to_raw): Likewise.
3323 (ffi_java_rvalue_to_raw): New function.
3324 (ffi_java_raw_call): Call it.
3325 (ffi_java_raw_to_rvalue): New function.
3326 (ffi_java_translate_args): Call it.
3327 * src/ffitest.c (closure_test_fn): Interpret return value
3328 as ffi_arg, not int.
3329 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3330 FFI_TYPE_POINTER case.
3331 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3332 values extended to word size.
3333
af50f4c8
AJ		 33342002-10-02 Andreas Jaeger <aj@suse.de>
3335
3336 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3337
5d297584
BT		 33382002-10-01 Bo Thorsen <bo@smetana.suse.de>
3339
3340 * include/ffi.h.in: Fix i386 win32 compilation.
3341
c4f17c6f
UW		 33422002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3343
3344 * configure.in: Add s390x-*-linux-* target.
3345 * configure: Regenerate.
3346 * include/ffi.h.in: Define S390X for s390x targets.
3347 (FFI_CLOSURES): Define for s390/s390x.
3348 (FFI_TRAMPOLINE_SIZE): Likewise.
3349 (FFI_NATIVE_RAW_API): Likewise.
3350 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3351 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
4977bab6 3352 * src/s390/ffi.c: Major rework of existing code. Add support for
c4f17c6f
UW		 3353 s390x targets. Add closure support.
3354 * src/s390/sysv.S: Likewise.
3355
d7b52625
RE		 33562002-09-29 Richard Earnshaw <rearnsha@arm.com>
3357
3358 * src/arm/sysv.S: Fix typo.
3359
28165e22
RE		 33602002-09-28 Richard Earnshaw <rearnsha@arm.com>
3361
3362 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3363 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3364 (ffi_call_SYSV): Handle soft-float.
3365
0b0f4163
BT		 33662002-09-27 Bo Thorsen <bo@suse.de>
3367
3368 * include/ffi.h.in: Fix multilib x86-64 support.
3369
6c5d742e
KG		 33702002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3371
3372 * Makefile.am (all-multi): Fix multilib parallel build.
3373
f4b2b0d6
KK		 33742002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3375
3376 * configure.in (sh[34]*-*-linux*): Add brackets.
3377 * configure: Regenerate.
3378
ddebdfeb
KK		 33792002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3380
3381 * Makefile.am: Add SH support.
3382 * Makefile.in: Regenerate.
3383 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3384 * configure: Regenerate.
3385 * include/ffi.h.in: Add SH support.
3386 * src/sh/ffi.c: New file.
3387 * src/sh/sysv.S: New file.
3388 * src/types.c: Add SH support.
3389
c94974ca
BT		 33902002-07-16 Bo Thorsen <bo@suse.de>
3391
3392 * src/x86/ffi64.c: New file that adds x86-64 support.
3393 * src/x86/unix64.S: New file that handles argument setup for
3394 x86-64.
3395 * src/x86/sysv.S: Don't use this on x86-64.
3396 * src/x86/ffi.c: Don't use this on x86-64.
3397 Remove unused vars.
3398 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3399 for x86-64.
3400 * src/ffitest.c (struct6): New test that tests a special case in
3401 the x86-64 ABI.
3402 (struct7): Likewise.
3403 (struct8): Likewise.
3404 (struct9): Likewise.
3405 (closure_test_fn): Silence warning about this when it's not used.
3406 (main): Add the new tests.
3407 (main): Fix a couple of wrong casts and silence some compiler warnings.
3408 * include/ffi.h.in: Add x86-64 ABI definition.
3409 * fficonfig.h.in: Regenerate.
3410 * Makefile.am: Add x86-64 support.
3411 * configure.in: Likewise.
3412 * Makefile.in: Regenerate.
3413 * configure: Likewise.
3414
34152002-06-24 Bo Thorsen <bo@suse.de>
3416
3417 * src/types.c: Merge settings for similar architectures.
3418 Add x86-64 sizes and alignments.
3419
34202002-06-23 Bo Thorsen <bo@suse.de>
3421
3422 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3423 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3424 * src/mips/ffi.c (ffi_prep_args): Likewise.
3425 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3426
5be36a8f
L		 34272002-07-18 H.J. Lu (hjl@gnu.org)
3428
3429 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3430 (libffi_la_SOURCES): Support MIPS_LINUX.
3431 (libffi_convenience_la_SOURCES): Likewise.
3432 * Makefile.in: Regenerated.
3433
3434 * configure.in (mips64*-*): Skip.
3435 (mips*-*-linux*): New.
3436 * configure: Regenerated.
3437
3438 * src/mips/ffi.c: Include <sgidefs.h>.
3439
ad5e163c
UW		 34402002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3441
3442 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3443
ee4586c5
RS		 34442002-05-27 Roger Sayle <roger@eyesopen.com>
3445
3446 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3447
b86aa7b0
BT		 34482002-05-27 Bo Thorsen <bo@suse.de>
3449
3450 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3451 fix formatting.
3452
859f7aaf
AT		 34532002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3454
3455 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3456 beginning of function (for older apple cc).
3457
6706f116
AO		 34582002-05-08 Alexandre Oliva <aoliva@redhat.com>
3459
3460 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3461 script entry, and set LD to it when configuring multilibs.
3462 * configure: Rebuilt.
3463
fad655c8
JT		 34642002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3465
3466 * configure.in (sparc64-*-netbsd*): Add target.
3467 (sparc-*-netbsdelf*): Likewise.
3468 * configure: Regenerate.
3469
189ab118
DM		 34702002-04-28 David S. Miller <davem@redhat.com>
3471
3472 * configure.in, configure: Fix SPARC test in previous change.
3473
22bcf65c
GT		 34742002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3475
3476 * Makefile.am: Add Linux for S/390 support.
3477 * Makefile.in: Regenerate.
3478 * configure.in: Add Linux for S/390 support.
3479 * configure: Regenerate.
3480 * include/ffi.h.in: Add Linux for S/390 support.
3481 * src/s390/ffi.c: New file from libffi CVS tree.
3482 * src/s390/sysv.S: New file from libffi CVS tree.
3483
5d84cf0b
JJ		 34842002-04-28 Jakub Jelinek <jakub@redhat.com>
3485
3486 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3487 %r_disp32().
3488 * src/sparc/v8.S: Use it.
3489 * src/sparc/v9.S: Likewise.
3490 * fficonfig.h.in: Rebuilt.
3491 * configure: Rebuilt.
3492
cc382e5c
HB		 34932002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3494
3495 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3496 correctly.
3497 * src/ia64/unix.S: Add unwind information. Fix comments.
3498 Save sp in a way that's compatible with unwind info.
3499 (ffi_call_unix): Correctly restore sp in all cases.
3500 * src/ia64/ffi.c: Add, fix comments.
3501
dc798645
JJ		 35022002-04-08 Jakub Jelinek <jakub@redhat.com>
3503
3504 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3505
da00eaab
JT		 35062002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3507
3508 * configure.in (alpha*-*-netbsd*): Add target.
3509 * configure: Regenerate.
3510
0c0b87ad
JS		 35112002-04-04 Jeff Sturm <jsturm@one-point.com>
3512
3513 * src/sparc/v8.S: Add unwind info.
3514 * src/sparc/v9.S: Likewise.
3515
320d7e7a
KW		 35162002-03-30 Krister Walfridsson <cato@df.lth.se>
3517
3518 * configure.in: Enable i*86-*-netbsdelf*.
3519 * configure: Rebuilt.
4977bab6 3520
c75e51ed
DB		 35212002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3522
3523 PR other/2620
3524 * src/mips/n32.s: Delete
3525 * src/mips/o32.s: Delete
3526
5bb366e0
LR		 35272002-03-21 Loren J. Rittle <ljrittle@acm.org>
3528
3529 * configure.in: Enable alpha*-*-freebsd*.
3530 * configure: Rebuilt.
3531
ce3b24a7
BM		 35322002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3533
3534 * Makefile.am: libfficonvenience -> libffi_convenience.
3535 * Makefile.in: Rebuilt.
4977bab6 3536
1eb7ba30
BM		 3537 * Makefile.am: Define ffitest_OBJECTS.
3538 * Makefile.in: Rebuilt.
ce3b24a7 3539
073ae293
AT		 35402002-03-07 Andreas Tobler <toa@pop.agri.ch>
3541 David Edelsohn <edelsohn@gnu.org>
3542
3543 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3544 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3545 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3546 * Makefile.in: Regenerate.
3547 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3548 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3549 (flush_icache, flush_range): New functions.
3550 (ffi_closure_helper_DARWIN): New function.
3551 * src/powerpc/aix_closure.S: New file.
3552 * src/powerpc/darwin_closure.S: New file.
4977bab6 3553
02c521f4
JS		 35542002-02-24 Jeff Sturm <jsturm@one-point.com>
3555
3556 * include/ffi.h.in: Add typedef for ffi_arg.
3557 * src/ffitest.c (main): Declare rint with ffi_arg.
4977bab6 3558
287dd527
AT		 35592002-02-21 Andreas Tobler <toa@pop.agri.ch>
3560
3561 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3562 number of GPRs for floating-point arguments.
3563
e2f2e70a
AG		 35642002-01-31 Anthony Green <green@redhat.com>
3565
3566 * configure: Rebuilt.
3567 * configure.in: Replace CHECK_SIZEOF and endian tests with
3568 cross-compiler friendly macros.
3569 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3570 macros.
3571
15b9b32f
DE		 35722002-01-18 David Edelsohn <edelsohn@gnu.org>
3573
3574 * src/powerpc/darwin.S (_ffi_call_AIX): New.
1f7d5413 3575 * src/powerpc/aix.S (ffi_call_DARWIN): New.
15b9b32f 3576
2f4006f8
DE		 35772002-01-17 David Edelsohn <edelsohn@gnu.org>
3578
3579 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3580 (TARGET_SRC_POWERPC_AIX): New.
3581 (POWERPC_AIX): New stanza.
3582 * Makefile.in: Regenerate.
3583 * configure.in: Add AIX case.
3584 * configure: Regenerate.
3585 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3586 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3587 size. Fix "long double" support.
3588 (ffi_call): Add FFI_AIX case.
3589 * src/powerpc/aix.S: New.
3590
3c017635
JH		 35912001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3592
3593 Implement Darwin PowerPC ABI.
3594 * configure.in: Handle powerpc-*-darwin*.
3595 * Makefile.am: Set source files for POWERPC_DARWIN.
3596 * configure: Rebuilt.
3597 * Makefile.in: Rebuilt.
3598 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3599 POWERPC_DARWIN.
3600 * src/powerpc/darwin.S: New file.
3601 * src/powerpc/ffi_darwin.c: New file.
3602
cc712abf
JM		 36032001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3604
3605 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3606
03a8de49
RO		 36072001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3608
3609 * src/x86/sysv.S: Avoid gas-only .balign directive.
3610 Use C style comments.
3611
92a0e6c6
RO		 36122001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3613
3614 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3615 Fixes PR bootstrap/3563.
3616
57482f58
RO		 36172001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3618
3619 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3620
d7e2dd2f
RO		 36212001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3622
3623 * configure.in: Recognize sparc*-sun-* host.
3624 * configure: Regenerate.
4977bab6 3625
be72207c
AH		 36262001-06-06 Andrew Haley <aph@redhat.com>
3627
4075db8f 3628 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
be72207c 3629
552cb45a
AH		 36302001-06-03 Andrew Haley <aph@redhat.com>
3631
4075db8f
BT		 3632 * src/alpha/osf.S: Add unwind info.
3633 * src/powerpc/sysv.S: Add unwind info.
3634 * src/powerpc/ppc_closure.S: Likewise.
552cb45a 3635
7d45b968
JS		 36362000-05-31 Jeff Sturm <jsturm@one-point.com>
3637
3638 * configure.in: Fix AC_ARG_ENABLE usage.
3639 * configure: Rebuilt.
3640
2a7f63c4
BM		 36412001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3642
3643 * configure.in: Remove warning about beta code.
3644 * configure: Rebuilt.
3645
36462001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3647
03a55252
HB		 3648 * src/ia64/unix.S: Restore stack pointer when returning from
3649 ffi_closure_UNIX.
3650 * src/ia64/ffi.c: Fix typo in comment.
3651
df5da1c4
JW		 36522001-04-18 Jim Wilson <wilson@redhat.com>
3653
3654 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3655 to eliminate RAW DV.
3656
52c28977
BM		 36572001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3658
3659 * Makefile.am: Make a libtool convenience library.
3660 * Makefile.in: Rebuilt.
3661
3b905056
BM		 36622001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3663
3664 * configure.in: Use different syntax for subdirectory creation.
3665 * configure: Rebuilt.
3666
d6c4152a
JB		 36672001-03-27 Jon Beniston <jon@beniston.com>
3668
3669 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3670 * configure: Rebuilt.
3671 * Makefile.am: Added X86_WIN32 target support.
3672 * Makefile.in: Rebuilt.
3673
3674 * include/ffi.h.in: Added X86_WIN32 target support.
3675
3676 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3677 * src/types.c: Added X86_WIN32 target support.
3678
3679 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3680 stuff removed and made to work with CygWin's gas.
3681
675cdab3
BM		 36822001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3683
3684 * configure.in: Make target subdirectory in build dir.
3685 * Makefile.am: Override suffix based rules to specify correct output
3686 subdirectory.
3687 * Makefile.in: Rebuilt.
3688 * configure: Rebuilt.
3689
e5f8858a 36902001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
cc4c8975
KH		 3691
3692 * src/powerpc/ppc_closure.S: New file.
3693 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3694 involving long long and register pairs.
3695 (ffi_prep_closure): New function.
3696 (flush_icache): Likewise.
3697 (ffi_closure_helper_SYSV): Likewise.
3698 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3699 (FFI_TRAMPOLINE_SIZE): Likewise.
3700 (FFI_NATIVE_RAW_API): Likewise.
3701 * Makefile.in: Rebuilt.
3702 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3703 (TARGET_SRC_POWERPC): Likewise.
3704
9450a929
TT		 37052001-03-19 Tom Tromey <tromey@redhat.com>
3706
3707 * Makefile.in: Rebuilt.
3708 * Makefile.am (ffitest_LDFLAGS): New macro.
3709
4d2cf715
NC		 37102001-03-02 Nick Clifton <nickc@redhat.com>
3711
3712 * include/ffi.h.in: Remove RCS ident string.
3713 * include/ffi_mips.h: Remove RCS ident string.
3714 * src/debug.c: Remove RCS ident string.
3715 * src/ffitest.c: Remove RCS ident string.
3716 * src/prep_cif.c: Remove RCS ident string.
3717 * src/types.c: Remove RCS ident string.
3718 * src/alpha/ffi.c: Remove RCS ident string.
3719 * src/alpha/osf.S: Remove RCS ident string.
3720 * src/arm/ffi.c: Remove RCS ident string.
3721 * src/arm/sysv.S: Remove RCS ident string.
3722 * src/mips/ffi.c: Remove RCS ident string.
3723 * src/mips/n32.S: Remove RCS ident string.
3724 * src/mips/o32.S: Remove RCS ident string.
3725 * src/sparc/ffi.c: Remove RCS ident string.
3726 * src/sparc/v8.S: Remove RCS ident string.
3727 * src/sparc/v9.S: Remove RCS ident string.
3728 * src/x86/ffi.c: Remove RCS ident string.
3729 * src/x86/sysv.S: Remove RCS ident string.
3730
7e5fd99f
JM		 37312001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3732
3733 * include/ffi.h.in: Change sourceware.cygnus.com references to
3734 gcc.gnu.org.
3735
3f67ba6e
RH		 37362000-12-09 Richard Henderson <rth@redhat.com>
3737
3738 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3739 (ffi_closure_osf_inner): Index rather than increment avalue
3740 and arg_types. Give ffi_closure_osf the raw return value type.
3741 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3742 promotion.
3743
37442000-12-07 Richard Henderson <rth@redhat.com>
29fe0479
RH		 3745
3746 * src/raw_api.c (ffi_translate_args): Fix typo.
3747 (ffi_prep_closure): Likewise.
3748
3749 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3750 FFI_TRAMPOLINE_SIZE.
3751 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3752 cif->bytes for new ffi_call_osf implementation.
3753 (ffi_prep_args): Absorb into ...
3754 (ffi_call): ... here. Do all stack allocation here and
3755 avoid a callback function.
3756 (ffi_prep_closure, ffi_closure_osf_inner): New.
3757 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3758 (ffi_closure_osf): New.
3759
2f7bf5b7
AO		 37602000-09-10 Alexandre Oliva <aoliva@redhat.com>
3761
ed1c1a1a
AO		 3762 * config.guess, config.sub, install-sh: Removed.
3763 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3764 * Makefile.in: Rebuilt.
3765
2f7bf5b7
AO		 3766 * acinclude.m4: Include libtool macros from the top level.
3767 * aclocal.m4, configure: Rebuilt.
3768
e15a68e7
AO		 37692000-08-22 Alexandre Oliva <aoliva@redhat.com>
3770
3771 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3772 * configure: Rebuilt.
3773
7f8c11d5
SB		 37742000-05-11 Scott Bambrough <scottb@netwinder.org>
3775
4977bab6
ZW		 3776 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3777 memory correctly. Use conditional instructions, not branches where
7f8c11d5
SB		 3778 possible.
3779
c5acf51f
TT		 37802000-05-04 Tom Tromey <tromey@cygnus.com>
3781
3782 * configure: Rebuilt.
3783 * configure.in: Match `arm*-*-linux-*'.
3784 From Chris Dornan <cdornan@arm.com>.
3785
91bffbe9
JJ		 37862000-04-28 Jakub Jelinek <jakub@redhat.com>
3787
3788 * Makefile.am (SUBDIRS): Define.
3789 (AM_MAKEFLAGS): Likewise.
3790 (Multilib support.): Add section.
3791 * Makefile.in: Rebuilt.
3792 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
4075db8f
BT		 3793 New variables. Set for gcc using -print-multi-lib. Export them
3794 to libtool.
3795 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3796 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3797 for -shared links.
3798 (extra_compiler_flags_value, extra_compiler_flags): Check these
3799 for extra compiler options which need to be passed down in
3800 compiler_flags.
91bffbe9 3801
bd66587e
AG		 38022000-04-16 Anthony Green <green@redhat.com>
3803
3804 * configure: Rebuilt.
3805 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3806
3791773c
JJ		 38072000-04-14 Jakub Jelinek <jakub@redhat.com>
3808
4075db8f 3809 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3791773c
JJ		 3810 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3811 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3812 Replace all void * sizeofs with sizeof(int).
3813 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3814 different than DOUBLE.
3815 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3816 (ffi_prep_args_v9): New function.
3817 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3818 (ffi_V9_return_struct): New function.
3819 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3820 32bit code (not yet cross-arch calls).
3821 * src/sparc/v8.S: Add struct return delay nop.
3822 Handle long long.
3823 * src/sparc/v9.S: New file.
3824 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3825 is used on sparc64 only for structures larger than 32 bytes.
3826 Pass by reference for structures is done for structure arguments
3827 larger than 16 bytes.
3828 * src/ffitest.c (main): Use 64bit rint on sparc64.
3829 Run long long tests on sparc.
3830 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3831 sparc64.
3832 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3833 on sparc64.
3834 * configure.in (sparc-*-linux*): New supported target.
3835 (sparc64-*-linux*): Likewise.
3836 * configure: Rebuilt.
3837 * Makefile.am: Add v9.S to SPARC files.
3838 * Makefile.in: Likewise.
3839 (LINK): Surround $(CCLD) into double quotes, so that multilib
3840 compiles work correctly.
3841
db459595
APB		 38422000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3843
3844 * configure: Rebuilt.
3845 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3846 proposed by Bryce McKinlay.
3847
c1f63902
TT		 38482000-03-20 Tom Tromey <tromey@cygnus.com>
3849
3850 * Makefile.in: Hand edit for java_raw_api.lo.
3851
0f72dc9e
BM		 38522000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3853
3854 * config.guess, config.sub: Update from the gcc tree.
3855 Fix for PR libgcj/168.
3856
3c18f2d1
TT		 38572000-03-03 Tom Tromey <tromey@cygnus.com>
3858
323eee8a
TT		 3859 * Makefile.in: Fixed ia64 by hand.
3860
3c18f2d1
TT		 3861 * configure: Rebuilt.
3862 * configure.in (--enable-multilib): New option.
3863 (libffi_basedir): New subst.
3864 (AC_OUTPUT): Added multilib code.
3865
a2b10a92
TT		 38662000-03-02 Tom Tromey <tromey@cygnus.com>
3867
3868 * Makefile.in: Rebuilt.
3869 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3870 directory name.
3871
dc5de370
HB		 38722000-02-25 Hans Boehm <boehm@acm.org>
3873
3874 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3875 files.
3876 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3877 list.
3878 (ffi_prep_raw_closure): Use ffi_translate_args, not
3879 ffi_closure_translate.
3880 * src/java_raw_api.c: New file.
3881 * src/ffitest.c (closure_test_fn): New function.
3882 (main): Define `rint' as long long on IA64. Added new test when
3883 FFI_CLOSURES is defined.
3884 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3885 (ffi_abi): Recognize IA64.
3886 (ffi_raw): Added `flt' field.
3887 Added "Java raw API" code.
3888 * configure.in: Recognize ia64.
3889 * Makefile.am (TARGET_SRC_IA64): New macro.
3890 (libffi_la_common_SOURCES): Added java_raw_api.c.
3891 (libffi_la_SOURCES): Define in IA64 case.
3892
d79229bd
TT		 38932000-01-04 Tom Tromey <tromey@cygnus.com>
3894
3895 * Makefile.in: Rebuilt with newer automake.
3896
2c33b220
TT		 38971999-12-31 Tom Tromey <tromey@cygnus.com>
3898
3899 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3900
6ba9e3b0
TT		 39011999-09-01 Tom Tromey <tromey@cygnus.com>
3902
3903 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3904 undefs.
3905 * fficonfig.h.in: Rebuilt.
3906 * configure: Rebuilt.
3907 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3908 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3909 * acconfig.h: Don't #undef PACKAGE or VERSION.
3910
63e5e3e0
AG		 39111999-08-09 Anthony Green <green@cygnus.com>
3912
3913 * include/ffi.h.in: Try to work around messy header problem
3914 with PACKAGE and VERSION.
3915
3916 * configure: Rebuilt.
3917 * configure.in: Change version to 2.00-beta.
3918
4075db8f 3919 * fficonfig.h.in: Rebuilt.
63e5e3e0
AG		 3920 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3921
3922 * src/x86/ffi.c (ffi_raw_call): Rename.
3923
39241999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3925
3926 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3927 (ffi_prep_incoming_args_SYSV): Ditto.
3928 (ffi_prep_closure): Ditto.
3929 (ffi_closure_raw_SYSV): Ditto.
3930 (ffi_prep_raw_closure): More ditto.
3931 (ffi_call_raw): Final ditto.
3932
3933 * include/ffi.h.in: Add definitions for closure and raw API.
3934
3935 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4977bab6 3936 FFI_TYPE_UINT64.
63e5e3e0
AG		 3937
3938 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3939
3940 * src/raw_api.c: New file.
3941
3942 * include/ffi.h.in (ffi_raw): New type.
3943 (UINT_ARG, SINT_ARG): New defines.
3944 (ffi_closure, ffi_raw_closure): New types.
3945 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4977bab6 3946
63e5e3e0
AG		 3947 * configure.in: Add check for endianness and sizeof void*.
3948
3949 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3950 instead of directly.
3951
3952 * configure: Rebuilt.
3953
3954Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3955
3956 * configure.in: Add x86 and powerpc BeOS configurations.
3957 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3958
39591999-05-09 Anthony Green <green@cygnus.com>
3960
3961 * configure.in: Add warning about this being beta code.
3962 Remove src/Makefile.am from the picture.
3963 * configure: Rebuilt.
3964
3965 * Makefile.am: Move logic from src/Makefile.am. Add changes
3966 to support libffi as a target library.
3967 * Makefile.in: Rebuilt.
3968
3969 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3970 Upgraded to new autoconf, automake, libtool.
3971
3972 * README: Tweaks.
3973
3974 * LICENSE: Update copyright date.
3975
3976 * src/Makefile.am, src/Makefile.in: Removed.
3977
39781998-11-29 Anthony Green <green@cygnus.com>
3979
3980 * include/ChangeLog: Removed.
3981 * src/ChangeLog: Removed.
3982 * src/mips/ChangeLog: Removed.
3983 * src/sparc/ChangeLog: Remboved.
3984 * src/x86/ChangeLog: Removed.
3985
3986 * ChangeLog.v1: Created.
Mirror of https://gcc.gnu.org/git/gcc.git