]> git.ipfire.org Git - thirdparty/gcc.git/blame - libffi/ChangeLog
ffi_darwin.c (ffi_closure_helper_DARWIN): Increment pfr for long double split between...
[thirdparty/gcc.git] / libffi / ChangeLog
CommitLineData
0c552622
DE
12009-12-04 David Edelsohn <edelsohn@gnu.org>
2
3 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
4 pfr for long double split between fpr13 and stack.
5
afb5e622
DE
62009-12-03 David Edelsohn <edelsohn@gnu.org>
7
8 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
9 fparg_count twice for long double.
10
be311914
DE
112009-12-03 David Edelsohn <edelsohn@gnu.org>
12
13 PR libffi/42243
14 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
15
a27c4040
UB
162009-12-03 Uros Bizjak <ubizjak@gmail.com>
17
18 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
ddf6c187 19 Remove xfails for x86 linux targets.
a27c4040 20
7837c247
DE
212009-12-02 David Edelsohn <edelsohn@gnu.org>
22
23 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
24 case.
25
262009-12-01 David Edelsohn <edelsohn@gnu.org>
6d518d3b
DE
27
28 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
29 register usage. Call ffi_prep_args directly. Add long double
30 return value support.
31 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
32 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
33 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
34 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
35 mode.
36 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
37 into case.
38 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
39 Allocate result area between params and FPRs.
40
5751cf6f
DE
412009-11-30 David Edelsohn <edelsohn@gnu.org>
42
43 PR target/35484
44 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
45 AIX64.
46 * src/powerpc/aix.S: Implement AIX64 version.
47 * src/powerpc/aix_closure.S: Implement AIX64 version.
48 (ffi_closure_ASM): Use extsb, lha and displament addresses.
49 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
50 support.
51 (ffi_prep_cif_machdep): Same.
52 (ffi_call): Same.
53 (ffi_closure_helper_DARWIN): Same.
54
2265f646
AT
552009-11-02 Andreas Tobler <a.tobler@schweiz.org>
56
57 PR libffi/41908
58 * testsuite/libffi.call/testclosure.c: New test.
59
ef6ed66d
KT
602009-09-28 Kai Tietz <kai.tietz@onevision.com>
61
62 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
63 assembly version use of ___chkstk.
64
29d1370a
MK
652009-09-23 Matthias Klose <doko@ubuntu.com>
66
67 PR libffi/40242, PR libffi/41443
68 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
69 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
70 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
71 Change the conditionals to __SOFTFP__ || __ARM_EABI__
72 for -mfloat-abi=softfp to work.
73
987ce557
LR
742009-09-17 Loren J. Rittle <ljrittle@acm.org>
75
76 PR testsuite/32843 (strikes again)
77 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
78 enable proper extension on char and short.
79
5cbf8c8d
DD
802009-09-15 David Daney <ddaney@caviumnetworks.com>
81
82 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
83 handling for FFI_TYPE_POINTER.
84 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
85 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
86 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
87 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
88 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
89 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
90 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
91 structure and pointer returns.
92 (ffi_closure_N32): Add handling for pointer returns.
93 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
94 calc_n32_return_struct_flags): Handle soft-float.
95 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
96 (ffi_call_N32): Declare proper argument types.
97 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
98 soft-float.
99
df58e648
RW
1002009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
101
102 * configure.ac (AC_PREREQ): Bump to 2.64.
103
5213506e
RW
1042009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
105
70fa0efa
RW
106 * Makefile.am (install-html, install-pdf): Remove.
107 * Makefile.in: Regenerate.
108
5213506e
RW
109 * Makefile.in: Regenerate.
110 * aclocal.m4: Regenerate.
111 * configure: Regenerate.
112 * fficonfig.h.in: Regenerate.
113 * include/Makefile.in: Regenerate.
114 * man/Makefile.in: Regenerate.
115 * testsuite/Makefile.in: Regenerate.
116
1b3b24c2
RW
1172009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
118
119 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
120
723512ba
DK
1212009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
122
123 PR libffi/40807
124 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
125 return types for X86_WIN32.
126 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
127 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
128 _ffi_closure_STDCALL): Likewise.
129
130 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
131 (dlmmap, dlmunmap): Also use these functions on Cygwin.
132
33cde516
RS
1332009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
134
135 PR testsuite/40699
136 PR testsuite/40707
137 PR testsuite/40709
138 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
139 2009-06-30 commits.
140
edd3a61c
RS
1412009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
142
143 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
144 to "" before adding paths. (This reinstates an assignment that
145 was removed by my 2009-06-30 commit, but changes the initial
146 value from "." to "".)
147
33d1a550
L
1482009-07-01 H.J. Lu <hongjiu.lu@intel.com>
149
150 PR testsuite/40601
151 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
152 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
153 empty.
154
b30bfd40
RS
1552009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
156
157 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
158 to ld_library_path. Use add_path. Add just find_libgcc_s
159 to ld_library_path, not every libgcc multilib directory.
160
6975f17f
WL
1612009-06-16 Wim Lewis <wiml@hhhh.org>
162
163 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
164 supposed to be callee-saved.
165 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
166 return buffer for odd-size structs.
167
7bbcc286
AT
1682009-06-16 Andreas Tobler <a.tobler@schweiz.org>
169
170 PR libffi/40444
171 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
172 allow_stack_execute for Darwin.
173
fe673cc9
AH
1742009-06-16 Andrew Haley <aph@redhat.com>
175
176 * configure.ac (TARGETDIR): Add missing blank lines.
177 * configure: Regenerate.
178
3bc462c2
AH
1792009-06-16 Andrew Haley <aph@redhat.com>
180
181 * testsuite/libffi.call/cls_align_sint64.c,
182 testsuite/libffi.call/cls_align_uint64.c,
183 testsuite/libffi.call/cls_longdouble_va.c,
184 testsuite/libffi.call/cls_ulonglong.c,
185 testsuite/libffi.call/return_ll1.c,
186 testsuite/libffi.call/stret_medium2.c: Fix printf format
187 specifiers.
188 * testsuite/libffi.call/ffitest.h,
189 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
190
081b044b
AH
1912009-06-15 Andrew Haley <aph@redhat.com>
192
193 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
194 * testsuite/libffi.call/err_bad_abi.c: Likewise.
195
062b8279
AH
1962009-06-12 Andrew Haley <aph@redhat.com>
197
198 * Makefile.am: Remove info_TEXINFOS.
199
2002009-06-12 Andrew Haley <aph@redhat.com>
201
202 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
203 testsuite/libffi.call/cls_align_uint64.c,
204 testsuite/libffi.call/cls_ulonglong.c,
205 testsuite/libffi.call/return_ll1.c,
206 testsuite/libffi.call/stret_medium2.c: Fix printf format
207 specifiers.
208 testsuite/libffi.special/unwindtest.cc: include stdint.h.
209
2102009-06-11 Timothy Wall <twall@users.sf.net>
211
212 * Makefile.am,
213 configure.ac,
214 include/ffi.h.in,
215 include/ffi_common.h,
216 src/closures.c,
217 src/dlmalloc.c,
218 src/x86/ffi.c,
219 src/x86/ffitarget.h,
220 src/x86/win64.S (new),
221 README: Added win64 support (mingw or MSVC)
222 * Makefile.in,
223 include/Makefile.in,
224 man/Makefile.in,
225 testsuite/Makefile.in,
226 configure,
227 aclocal.m4: Regenerated
228 * ltcf-c.sh: properly escape cygwin/w32 path
229 * man/ffi_call.3: Clarify size requirements for return value.
230 * src/x86/ffi64.c: Fix filename in comment.
231 * src/x86/win32.S: Remove unused extern.
232
233 * testsuite/libffi.call/closure_fn0.c,
234 testsuite/libffi.call/closure_fn1.c,
235 testsuite/libffi.call/closure_fn2.c,
236 testsuite/libffi.call/closure_fn3.c,
237 testsuite/libffi.call/closure_fn4.c,
238 testsuite/libffi.call/closure_fn5.c,
239 testsuite/libffi.call/closure_fn6.c,
240 testsuite/libffi.call/closure_stdcall.c,
241 testsuite/libffi.call/cls_12byte.c,
242 testsuite/libffi.call/cls_16byte.c,
243 testsuite/libffi.call/cls_18byte.c,
244 testsuite/libffi.call/cls_19byte.c,
245 testsuite/libffi.call/cls_1_1byte.c,
246 testsuite/libffi.call/cls_20byte.c,
247 testsuite/libffi.call/cls_20byte1.c,
248 testsuite/libffi.call/cls_24byte.c,
249 testsuite/libffi.call/cls_2byte.c,
250 testsuite/libffi.call/cls_3_1byte.c,
251 testsuite/libffi.call/cls_3byte1.c,
252 testsuite/libffi.call/cls_3byte2.c,
253 testsuite/libffi.call/cls_4_1byte.c,
254 testsuite/libffi.call/cls_4byte.c,
255 testsuite/libffi.call/cls_5_1_byte.c,
256 testsuite/libffi.call/cls_5byte.c,
257 testsuite/libffi.call/cls_64byte.c,
258 testsuite/libffi.call/cls_6_1_byte.c,
259 testsuite/libffi.call/cls_6byte.c,
260 testsuite/libffi.call/cls_7_1_byte.c,
261 testsuite/libffi.call/cls_7byte.c,
262 testsuite/libffi.call/cls_8byte.c,
263 testsuite/libffi.call/cls_9byte1.c,
264 testsuite/libffi.call/cls_9byte2.c,
265 testsuite/libffi.call/cls_align_double.c,
266 testsuite/libffi.call/cls_align_float.c,
267 testsuite/libffi.call/cls_align_longdouble.c,
268 testsuite/libffi.call/cls_align_longdouble_split.c,
269 testsuite/libffi.call/cls_align_longdouble_split2.c,
270 testsuite/libffi.call/cls_align_pointer.c,
271 testsuite/libffi.call/cls_align_sint16.c,
272 testsuite/libffi.call/cls_align_sint32.c,
273 testsuite/libffi.call/cls_align_sint64.c,
274 testsuite/libffi.call/cls_align_uint16.c,
275 testsuite/libffi.call/cls_align_uint32.c,
276 testsuite/libffi.call/cls_align_uint64.c,
277 testsuite/libffi.call/cls_dbls_struct.c,
278 testsuite/libffi.call/cls_double.c,
279 testsuite/libffi.call/cls_double_va.c,
280 testsuite/libffi.call/cls_float.c,
281 testsuite/libffi.call/cls_longdouble.c,
282 testsuite/libffi.call/cls_longdouble_va.c,
283 testsuite/libffi.call/cls_multi_schar.c,
284 testsuite/libffi.call/cls_multi_sshort.c,
285 testsuite/libffi.call/cls_multi_sshortchar.c,
286 testsuite/libffi.call/cls_multi_uchar.c,
287 testsuite/libffi.call/cls_multi_ushort.c,
288 testsuite/libffi.call/cls_multi_ushortchar.c,
289 testsuite/libffi.call/cls_pointer.c,
290 testsuite/libffi.call/cls_pointer_stack.c,
291 testsuite/libffi.call/cls_schar.c,
292 testsuite/libffi.call/cls_sint.c,
293 testsuite/libffi.call/cls_sshort.c,
294 testsuite/libffi.call/cls_uchar.c,
295 testsuite/libffi.call/cls_uint.c,
296 testsuite/libffi.call/cls_ulonglong.c,
297 testsuite/libffi.call/cls_ushort.c,
298 testsuite/libffi.call/err_bad_abi.c,
299 testsuite/libffi.call/err_bad_typedef.c,
300 testsuite/libffi.call/float2.c,
301 testsuite/libffi.call/huge_struct.c,
302 testsuite/libffi.call/nested_struct.c,
303 testsuite/libffi.call/nested_struct1.c,
304 testsuite/libffi.call/nested_struct10.c,
305 testsuite/libffi.call/nested_struct2.c,
306 testsuite/libffi.call/nested_struct3.c,
307 testsuite/libffi.call/nested_struct4.c,
308 testsuite/libffi.call/nested_struct5.c,
309 testsuite/libffi.call/nested_struct6.c,
310 testsuite/libffi.call/nested_struct7.c,
311 testsuite/libffi.call/nested_struct8.c,
312 testsuite/libffi.call/nested_struct9.c,
313 testsuite/libffi.call/problem1.c,
314 testsuite/libffi.call/return_ldl.c,
315 testsuite/libffi.call/return_ll1.c,
316 testsuite/libffi.call/stret_large.c,
317 testsuite/libffi.call/stret_large2.c,
318 testsuite/libffi.call/stret_medium.c,
319 testsuite/libffi.call/stret_medium2.c,
320 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
321 of checking for MMAP. Use intptr_t instead of long casts.
322
f43def61
KK
3232009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
324
325 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
326 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
327 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
328
6aa0b218
AH
3292009-06-09 Andrew Haley <aph@redhat.com>
330
331 * src/x86/freebsd.S: Add missing file.
332
b7b857bb
AH
3332009-06-08 Andrew Haley <aph@redhat.com>
334
335 Import from libffi 3.0.8:
336
337 * doc/libffi.texi: New file.
338 * doc/libffi.info: Likewise.
339 * doc/stamp-vti: Likewise.
340 * man/Makefile.am: New file.
341 * man/ffi_call.3: New file.
342
343 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
344 src/dlmalloc.c.
345 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
346
347 * configure.ac: Bump version to 3.0.8.
348 parisc*-*-linux*: Add.
349 i386-*-freebsd* | i386-*-openbsd*: Add.
350 powerpc-*-beos*: Add.
351 AM_CONDITIONAL X86_FREEBSD: Add.
352 AC_CONFIG_FILES: Add man/Makefile.
353
354 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
355
491e6ad1
AH
3562009-06-08 Andrew Haley <aph@redhat.com>
357
358 * README: Import from libffi 3.0.8.
359
3b5b1a73
AH
3602009-06-08 Andrew Haley <aph@redhat.com>
361
362 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
363 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
364 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
365 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
366
367 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
368 * testsuite/libffi.call/stret_medium.c: Likewise.
369 * testsuite/libffi.call/stret_large2.c: Likewise.
370 * testsuite/libffi.call/stret_large.c: Likewise.
371
3722008-12-26 Timothy Wall <twall@users.sf.net>
373
374 * testsuite/libffi.call/cls_longdouble.c,
375 testsuite/libffi.call/cls_longdouble_va.c,
376 testsuite/libffi.call/cls_align_longdouble.c,
377 testsuite/libffi.call/cls_align_longdouble_split.c,
378 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
379 failures on x86_64 cygwin/mingw.
380
3812008-12-22 Timothy Wall <twall@users.sf.net>
382
383 * testsuite/libffi.call/closure_fn0.c,
384 testsuite/libffi.call/closure_fn1.c,
385 testsuite/libffi.call/closure_fn2.c,
386 testsuite/libffi.call/closure_fn3.c,
387 testsuite/libffi.call/closure_fn4.c,
388 testsuite/libffi.call/closure_fn5.c,
389 testsuite/libffi.call/closure_fn6.c,
390 testsuite/libffi.call/closure_loc_fn0.c,
391 testsuite/libffi.call/closure_stdcall.c,
392 testsuite/libffi.call/cls_align_pointer.c,
393 testsuite/libffi.call/cls_pointer.c,
394 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
395 pointer to integer (intptr_t).
396 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
397
3982008-07-24 Anthony Green <green@redhat.com>
399
400 * testsuite/libffi.call/cls_dbls_struct.c,
401 testsuite/libffi.call/cls_double_va.c,
402 testsuite/libffi.call/cls_longdouble.c,
403 testsuite/libffi.call/cls_longdouble_va.c,
404 testsuite/libffi.call/cls_pointer.c,
405 testsuite/libffi.call/cls_pointer_stack.c,
406 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
407 compiler warnings.
408
4092008-03-04 Anthony Green <green@redhat.com>
410 Blake Chaffin
411 hos@tamanegi.org
412
413 * testsuite/libffi.call/cls_align_longdouble_split2.c
414 testsuite/libffi.call/cls_align_longdouble_split.c
415 testsuite/libffi.call/cls_dbls_struct.c
416 testsuite/libffi.call/cls_double_va.c
417 testsuite/libffi.call/cls_longdouble.c
418 testsuite/libffi.call/cls_longdouble_va.c
419 testsuite/libffi.call/cls_pointer.c
420 testsuite/libffi.call/cls_pointer_stack.c
421 testsuite/libffi.call/err_bad_abi.c
422 testsuite/libffi.call/err_bad_typedef.c
423 testsuite/libffi.call/stret_large2.c
424 testsuite/libffi.call/stret_large.c
425 testsuite/libffi.call/stret_medium2.c
426 testsuite/libffi.call/stret_medium.c: New tests from Apple.
427
56faec5e
AH
4282009-06-05 Andrew Haley <aph@redhat.com>
429
430 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
431 libffi.
432
90409e4c
AH
4332009-06-04 Andrew Haley <aph@redhat.com>
434
435 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
436 stdcall changes.
437
b1393750
AH
4382008-02-26 Anthony Green <green@redhat.com>
439 Thomas Heller <theller@ctypes.org>
440
441 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
442 comment.
443
4442008-02-03 Timothy Wall <twall@users.sf.net>
445
446 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
447 offset based on code pointer, not data pointer.
448
4492008-01-31 Timothy Wall <twall@users.sf.net>
450
451 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
452 closures.
453 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
454 closures.
455 * src/x86/win32.S: Add assembly for stdcall closure.
456 * src/x86/ffi.c: Initialize stdcall closure trampoline.
457
96baa251
AH
4582009-06-04 Andrew Haley <aph@redhat.com>
459
460 * include/ffi.h.in: Change void (*)() to void (*)(void).
461 * src/x86/ffi.c: Likewise.
462
7446546a
AH
4632009-06-04 Andrew Haley <aph@redhat.com>
464
465 * src/powerpc/ppc_closure.S: Insert licence header.
466 * src/powerpc/linux64_closure.S: Likewise.
467 * src/m68k/sysv.S: Likewise.
468
469 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
470 * src/powerpc/ffi.c: Likewise.
471 * src/powerpc/ffi_darwin.c: Likewise.
472 * src/m32r/ffi.c: Likewise.
473 * src/sh64/ffi.c: Likewise.
474 * src/x86/ffi64.c: Likewise.
475 * src/alpha/ffi.c: Likewise.
476 * src/alpha/osf.S: Likewise.
477 * src/frv/ffi.c: Likewise.
478 * src/s390/ffi.c: Likewise.
479 * src/pa/ffi.c: Likewise.
480 * src/pa/hpux32.S: Likewise.
481 * src/ia64/unix.S: Likewise.
482 * src/ia64/ffi.c: Likewise.
483 * src/sparc/ffi.c: Likewise.
484 * src/mips/ffi.c: Likewise.
485 * src/sh/ffi.c: Likewise.
486
4872008-02-15 David Daney <ddaney@avtrex.com>
488
489 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
490 Define (conditionally), and use it to include cachectl.h.
491 (ffi_prep_closure_loc): Fix cache flushing.
492 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
493
5f933ef0
AH
4942009-06-04 Andrew Haley <aph@redhat.com>
495
496 include/ffi.h.in,
497 src/arm/ffitarget.h,
498 src/arm/ffi.c,
499 src/arm/sysv.S,
500 src/powerpc/ffitarget.h,
501 src/closures.c,
502 src/sh64/ffitarget.h,
503 src/sh64/ffi.c,
504 src/sh64/sysv.S,
505 src/types.c,
506 src/x86/ffi64.c,
507 src/x86/ffitarget.h,
508 src/x86/win32.S,
509 src/x86/darwin.S,
510 src/x86/ffi.c,
511 src/x86/sysv.S,
512 src/x86/unix64.S,
513 src/alpha/ffitarget.h,
514 src/alpha/ffi.c,
515 src/alpha/osf.S,
516 src/m68k/ffitarget.h,
517 src/frv/ffitarget.h,
518 src/frv/ffi.c,
519 src/s390/ffitarget.h,
520 src/s390/sysv.S,
521 src/cris/ffitarget.h,
522 src/pa/linux.S,
523 src/pa/ffitarget.h,
524 src/pa/ffi.c,
525 src/raw_api.c,
526 src/ia64/ffitarget.h,
527 src/ia64/unix.S,
528 src/ia64/ffi.c,
529 src/ia64/ia64_flags.h,
530 src/java_raw_api.c,
531 src/debug.c,
532 src/sparc/v9.S,
533 src/sparc/ffitarget.h,
534 src/sparc/ffi.c,
535 src/sparc/v8.S,
536 src/mips/ffitarget.h,
537 src/mips/n32.S,
538 src/mips/o32.S,
539 src/mips/ffi.c,
540 src/prep_cif.c,
541 src/sh/ffitarget.h,
542 src/sh/ffi.c,
543 src/sh/sysv.S: Update license text.
544
84c40c4a
DK
5452009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
546
547 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
548 (.eh_frame): Add FDE for it.
549
8353cd27
DK
5502009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
551
552 * configure.ac: Also check if assembler supports pc-relative
553 relocs on X86_WIN32 targets.
554 * configure: Regenerate.
555 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
556 (_ffi_call_SYSV): Add missing function type symbol .def and
557 add EH markup labels.
558 (_ffi_call_STDCALL): Likewise.
559 (_ffi_closure_SYSV): Likewise.
560 (_ffi_closure_raw_SYSV): Likewise.
561 (.eh_frame): Add hand-crafted EH data.
562
748086b7
JJ
5632009-04-09 Jakub Jelinek <jakub@redhat.com>
564
565 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
566 version 3 of the GNU General Public License and to point readers
567 at the COPYING3 file and the FSF's license web page.
568 * testsuite/libffi.call/call.exp: Likewise.
569 * testsuite/libffi.special/special.exp: Likewise.
570
7a9d3fe8
RW
5712009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
572
573 * configure: Regenerate.
574
4d3d4078
RO
5752008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
576
577 PR libffi/26048
578 * configure.ac (HAVE_AS_X86_PCREL): New test.
579 * configure: Regenerate.
580 * fficonfig.h.in: Regenerate.
581 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
582 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
583 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
584 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
585 * src/x86/unix64.S (.Lstore_table): Move to .text section.
586 (.Lload_table): Likewise.
587 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
588
baafc534
RW
5892008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
590
591 * configure: Regenerate.
592
da17a98b
EB
5932008-11-21 Eric Botcazou <ebotcazou@adacore.com>
594
595 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
596 signed/unsigned int8/16 return values.
597 * src/sparc/v8.S (ffi_call_v8): Likewise.
598 (ffi_closure_v8): Likewise.
599
80c6c2cb
PG
6002008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
601 Steve Ellcey <sje@cup.hp.com>
602
603 * configure: Regenerate for new libtool.
604 * Makefile.in: Ditto.
605 * include/Makefile.in: Ditto.
606 * aclocal.m4: Ditto.
607
d7f68a67
AT
6082008-08-25 Andreas Tobler <a.tobler@schweiz.org>
609
610 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
611 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
612 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
613 Adjust copyright notice.
614 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
615 register or two register to use for FFI_SYSV structs.
616 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
617 (ffi_closure_helper_SYSV): Fix the return type for
618 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
619 Adjust copyright notice.
620
130f4c50
KK
6212008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
622
623 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
624 int.
625
18c04407
RW
6262008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
627
628 * configure: Regenerate.
629 * include/Makefile.in: Regenerate.
630 * testsuite/Makefile.in: Regenerate.
631
cd985f66
JM
6322008-06-07 Joseph Myers <joseph@codesourcery.com>
633
634 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
635 powerpc-*-beos*): Remove.
636 * configure: Regenerate.
637
7afa92c5
JB
6382008-05-09 Julian Brown <julian@codesourcery.com>
639
640 * Makefile.am (LTLDFLAGS): New.
641 (libffi_la_LDFLAGS): Use above.
642 * Makefile.in: Regenerate.
643
deb984e6
PB
6442008-04-18 Paolo Bonzini <bonzini@gnu.org>
645
646 PR bootstrap/35457
647 * aclocal.m4: Regenerate.
648 * configure: Regenerate.
649
3e9608a5
KK
6502008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
651
652 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
653 * src/sh64/sysv.S: Likewise.
654
feb51530
MK
6552008-03-26 Daniel Jacobowitz <dan@debian.org>
656
657 * src/arm/sysv.S: Fix ARM comment marker.
658
6592008-03-26 Jakub Jelinek <jakub@redhat.com>
660
661 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
662 * src/s390/sysv.S: Likewise.
663 * src/powerpc/ppc_closure.S: Likewise.
664 * src/powerpc/sysv.S: Likewise.
665 * src/x86/unix64.S: Likewise.
666 * src/x86/sysv.S: Likewise.
667 * src/sparc/v8.S: Likewise.
668 * src/sparc/v9.S: Likewise.
669 * src/m68k/sysv.S: Likewise.
670 * src/arm/sysv.S: Likewise.
671
38d24731
RW
6722008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
673
674 * aclocal.m4: Regenerate.
675 * configure: Likewise.
676 * Makefile.in: Likewise.
677 * include/Makefile.in: Likewise.
678 * testsuite/Makefile.in: Likewise.
679
2e250fa2
BK
6802008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
681 Andreas Tobler <a.tobler@schweiz.org>
682
683 * configure.ac: Add amd64-*-freebsd* target.
684 * configure: Regenerate.
685
7f92e242
L
6862008-01-30 H.J. Lu <hongjiu.lu@intel.com>
687
688 PR libffi/34612
689 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
690 returning struct.
691
692 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
693 tests.
694
95d4f544
DE
6952008-01-24 David Edelsohn <edelsohn@gnu.org>
696
697 * configure: Regenerate.
698
79ca8fc0
AT
6992008-01-06 Andreas Tobler <a.tobler@schweiz.org>
700
701 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
702
d7b07cc7
AT
7032008-01-05 Andreas Tobler <a.tobler@schweiz.org>
704
705 PR testsuite/32843
706 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
707 signed/unsigned int8/16 for X86_DARWIN.
708 Updated copyright info.
709 Handle one and two byte structs with special cif->flags.
710 * src/x86/ffitarget.h: Add special types for one and two byte structs.
711 Updated copyright info.
712 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
713 sysv.S
714 Remove code to pop args from the stack after call.
715 Special-case signed/unsigned for int8/16, one and two byte structs.
716 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
717 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
718 FFI_TYPE_SINT32.
719 Updated copyright info.
720
c94ae2e4
DD
7212007-12-08 David Daney <ddaney@avtrex.com>
722
723 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
724 SUBU, add with ADDU and use smaller code sequences.
725
3875b6d7 7262007-12-07 David Daney <ddaney@avtrex.com>
44be66a1 727
3875b6d7
DD
728 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
729 type.
730
4c42b3d8
DD
7312007-12-06 David Daney <ddaney@avtrex.com>
732
733 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
734 defined.
735 (ffi_java_raw): New typedef.
736 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
737 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
738 ffi_java_raw.
739 (ffi_java_raw_closure) : Same.
740 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
741 parameter types.
742 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
743 FFI_SIZEOF_JAVA_RAW.
744 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
745 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
746 sizeof(ffi_java_raw) for alignment calculations.
747 (ffi_java_ptrarray_to_raw): Same.
748 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
749 if FFI_SIZEOF_JAVA_RAW == 4.
750 (ffi_java_raw_to_rvalue): Same.
751 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
752 (ffi_java_translate_args): Same.
753 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
754 parameter types.
755 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
756
3ef50b62
DD
7572007-12-06 David Daney <ddaney@avtrex.com>
758
759 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
760 pointer values.
761
16287148
AT
7622007-12-01 Andreas Tobler <a.tobler@schweiz.org>
763
764 PR libffi/31937
765 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
766 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
767 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
768 set the NUM_FPR_ARG_REGISTERS according to.
769 Add support for potential soft-float support under hard-float
770 architecture.
771 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
772 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
773 to the FFI_LINUX_SOFT_FLOAT ABI.
774 (ffi_prep_cif_machdep): Likewise.
775 (ffi_closure_helper_SYSV): Likewise.
776 * src/powerpc/ppc_closure.S: Make sure not to store float/double
777 on archs where __NO_FPRS__ is true.
778 Add FFI_TYPE_UINT128 support.
779 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
780 Adjust copyright notice.
781
a4b7c9cd
AT
7822007-11-25 Andreas Tobler <a.tobler@schweiz.org>
783
784 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
785 * include/ffi_common.h: ... here.
786 Update copyright.
787
e84296c6
AT
7882007-11-17 Andreas Tobler <a.tobler@schweiz.org>
789
790 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
791 * src/powerpc/linux64.S: Likewise.
792 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
793 * testsuite/libffi.call/return_ldl.c: New test.
794
cb642590
AH
7952007-09-04 <aph@redhat.com>
796
797 * src/arm/sysv.S (UNWIND): New.
798 (Whole file): Conditionally compile unwinder directives.
799 * src/arm/sysv.S: Add unwinder directives.
800
801 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
802 Only treat r0 as a struct address if we're actually returning a
803 struct by address.
804 Only copy the bytes that are actually within a struct.
805 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
806 is returned in r0, not passed by address.
807 (ffi_call): Allocate a word-sized temporary for the case where
808 a composite is returned in r0.
809 (ffi_prep_incoming_args_SYSV): Align as necessary.
810
8112007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
812
813 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
814 directly using the sys_cacheflush syscall.
815
8162007-07-27 Andrew Haley <aph@redhat.com>
817
818 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
819
8d5c5e35
MR
8202007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
821
822 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
823 * configure.ac: Likewise.
824 * Makefile.in: Regenerate.
825 * include/Makefile.in: Likewise.
826 * testsuite/Makefile.in: Likewise.
827 * configure: Likewise.
828
787b0d19
DD
8292007-08-24 David Daney <ddaney@avtrex.com>
830
831 * testsuite/libffi.call/return_sl.c: New test.
832
dfbb3e36
DD
8332007-08-10 David Daney <ddaney@avtrex.com>
834
835 * testsuite/libffi.call/cls_multi_ushort.c,
836 testsuite/libffi.call/cls_align_uint16.c,
837 testsuite/libffi.call/nested_struct1.c,
838 testsuite/libffi.call/nested_struct3.c,
839 testsuite/libffi.call/cls_7_1_byte.c,
840 testsuite/libffi.call/nested_struct5.c,
841 testsuite/libffi.call/cls_double.c,
842 testsuite/libffi.call/nested_struct7.c,
843 testsuite/libffi.call/cls_sint.c,
844 testsuite/libffi.call/nested_struct9.c,
845 testsuite/libffi.call/cls_20byte1.c,
846 testsuite/libffi.call/cls_multi_sshortchar.c,
847 testsuite/libffi.call/cls_align_sint64.c,
848 testsuite/libffi.call/cls_3byte2.c,
849 testsuite/libffi.call/cls_multi_schar.c,
850 testsuite/libffi.call/cls_multi_uchar.c,
851 testsuite/libffi.call/cls_19byte.c,
852 testsuite/libffi.call/cls_9byte1.c,
853 testsuite/libffi.call/cls_align_float.c,
854 testsuite/libffi.call/closure_fn1.c,
855 testsuite/libffi.call/problem1.c,
856 testsuite/libffi.call/closure_fn3.c,
857 testsuite/libffi.call/cls_sshort.c,
858 testsuite/libffi.call/closure_fn5.c,
859 testsuite/libffi.call/cls_align_double.c,
860 testsuite/libffi.call/nested_struct.c,
861 testsuite/libffi.call/cls_2byte.c,
862 testsuite/libffi.call/nested_struct10.c,
863 testsuite/libffi.call/cls_4byte.c,
864 testsuite/libffi.call/cls_6byte.c,
865 testsuite/libffi.call/cls_8byte.c,
866 testsuite/libffi.call/cls_multi_sshort.c,
867 testsuite/libffi.call/cls_align_sint16.c,
868 testsuite/libffi.call/cls_align_uint32.c,
869 testsuite/libffi.call/cls_20byte.c,
870 testsuite/libffi.call/cls_float.c,
871 testsuite/libffi.call/nested_struct2.c,
872 testsuite/libffi.call/cls_5_1_byte.c,
873 testsuite/libffi.call/nested_struct4.c,
874 testsuite/libffi.call/cls_24byte.c,
875 testsuite/libffi.call/nested_struct6.c,
876 testsuite/libffi.call/cls_64byte.c,
877 testsuite/libffi.call/nested_struct8.c,
878 testsuite/libffi.call/cls_uint.c,
879 testsuite/libffi.call/cls_multi_ushortchar.c,
880 testsuite/libffi.call/cls_schar.c,
881 testsuite/libffi.call/cls_uchar.c,
882 testsuite/libffi.call/cls_align_uint64.c,
883 testsuite/libffi.call/cls_ulonglong.c,
884 testsuite/libffi.call/cls_align_longdouble.c,
885 testsuite/libffi.call/cls_1_1byte.c,
886 testsuite/libffi.call/cls_12byte.c,
887 testsuite/libffi.call/cls_3_1byte.c,
888 testsuite/libffi.call/cls_3byte1.c,
889 testsuite/libffi.call/cls_4_1byte.c,
890 testsuite/libffi.call/cls_6_1_byte.c,
891 testsuite/libffi.call/cls_16byte.c,
892 testsuite/libffi.call/cls_18byte.c,
893 testsuite/libffi.call/closure_fn0.c,
894 testsuite/libffi.call/cls_9byte2.c,
895 testsuite/libffi.call/closure_fn2.c,
896 testsuite/libffi.call/closure_fn4.c,
897 testsuite/libffi.call/cls_ushort.c,
898 testsuite/libffi.call/closure_fn6.c,
899 testsuite/libffi.call/cls_5byte.c,
900 testsuite/libffi.call/cls_align_pointer.c,
901 testsuite/libffi.call/cls_7byte.c,
902 testsuite/libffi.call/cls_align_sint32.c,
903 testsuite/libffi.special/unwindtest_ffi_call.cc,
904 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
905
89d9d98a
DD
9062007-08-10 David Daney <ddaney@avtrex.com>
907
908 PR libffi/28313
909 * configure.ac: Don't treat mips64 as a special case.
910 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
911 * configure: Regenerate
912 * Makefile.in: Ditto.
913 * fficonfig.h.in: Ditto.
914 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
915 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
916 (FFI_DEFAULT_ABI): Set for n64 case.
917 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
918 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
919 (ffi_closure_N32): New function.
920 (.eh_frame): New section
921 * src/mips/o32.S: Clean up comments.
922 (ffi_closure_O32): Pass ffi_closure parameter in $12.
923 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
924 _MIPS_SIM == _ABIN32 throughout.
925 (FFI_MIPS_STOP_HERE): New, use in place of
926 ffi_stop_here.
927 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
928 to support n32/n64 ABIs.
929 (calc_n32_struct_flags): Rewrite.
930 (calc_n32_return_struct_flags): Remove unused variable. Reverse
931 position of flag bits.
932 (ffi_prep_cif_machdep): Rewrite n32 portion.
933 (ffi_call): Enable for n64. Add special handling for small structure
934 return values.
935 (ffi_prep_closure_loc): Add n32 and n64 support.
936 (ffi_closure_mips_inner_O32): Add cast to silence warning.
937 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
938
d23869d4
DD
9392007-08-08 David Daney <ddaney@avtrex.com>
940
941 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
942 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
943 specifiers.
944 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
945 * testsuite/libffi.call/cls_sint.c (main): Ditto.
946 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
947 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
948 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
949 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
950 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
951 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
952 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
953 * testsuite/libffi.call/return_fl3.c (main): Ditto.
954 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
955 * testsuite/libffi.call/nested_struct.c (main): Ditto.
956 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
957 * testsuite/libffi.call/return_ll1.c (main): Ditto.
958 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
959 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
960 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
961 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
962 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
963 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
964 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
965 * testsuite/libffi.call/cls_uint.c (main): Ditto.
966 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
967 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
968 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
969 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
970 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
971 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
972 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
973 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
974 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
975
536e0a8f
AH
9762007-08-07 Andrew Haley <aph@redhat.com>
977
978 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
979 checkin.
980
7665c9be
AH
9812007-08-06 Andrew Haley <aph@redhat.com>
982
983 PR testsuite/32843
984 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
985 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
986 FFI_TYPE_SINT32.
987
58e11497
DD
9882007-08-02 David Daney <ddaney@avtrex.com>
989
990 * testsuite/libffi.call/return_ul.c (main): Define return type as
991 ffi_arg. Use proper printf conversion specifier.
992
387cfbba
AH
9932007-07-30 Andrew Haley <aph@redhat.com>
994
995 PR testsuite/32843
996 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
997 signed/unsigned int8/16.
998 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
999 Use a jump table.
1000 Remove code to pop args from the stack after call.
1001 Special-case signed/unsigned int8/16.
1002 * testsuite/libffi.call/return_sc.c (main): Revert.
1003
0ee0208e
RG
10042007-07-26 Richard Guenther <rguenther@suse.de>
1005
1006 PR testsuite/32843
1007 * testsuite/libffi.call/return_sc.c (main): Verify call
1008 result as signed char, not ffi_arg.
1009
04a49983
RO
10102007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1011
1012 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1013 * configure: Regenerate.
1014
433e6a8c
DD
10152007-07-11 David Daney <ddaney@avtrex.com>
1016
1017 * src/mips/ffi.c: Don't include sys/cachectl.h.
1018 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1019 cacheflush().
1020
0959e2b8
AH
10212007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1022
1023 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1024 from (ffi_prep_closure): ... this.
1025 (FFI_INIT_TRAMPOLINE): Adjust.
1026
10272005-12-31 Phil Blundell <pb@reciva.com>
1028
1029 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1030 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1031 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1032 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1033 (FFI_CLOSURES): Enable closure support.
1034
10352007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1036
1037 * testsuite/libffi.call/cls_multi_ushort.c,
1038 testsuite/libffi.call/cls_align_uint16.c,
1039 testsuite/libffi.call/nested_struct1.c,
1040 testsuite/libffi.call/nested_struct3.c,
1041 testsuite/libffi.call/cls_7_1_byte.c,
1042 testsuite/libffi.call/cls_double.c,
1043 testsuite/libffi.call/nested_struct5.c,
1044 testsuite/libffi.call/nested_struct7.c,
1045 testsuite/libffi.call/cls_sint.c,
1046 testsuite/libffi.call/nested_struct9.c,
1047 testsuite/libffi.call/cls_20byte1.c,
1048 testsuite/libffi.call/cls_multi_sshortchar.c,
1049 testsuite/libffi.call/cls_align_sint64.c,
1050 testsuite/libffi.call/cls_3byte2.c,
1051 testsuite/libffi.call/cls_multi_schar.c,
1052 testsuite/libffi.call/cls_multi_uchar.c,
1053 testsuite/libffi.call/cls_19byte.c,
1054 testsuite/libffi.call/cls_9byte1.c,
1055 testsuite/libffi.call/cls_align_float.c,
1056 testsuite/libffi.call/closure_fn1.c,
1057 testsuite/libffi.call/problem1.c,
1058 testsuite/libffi.call/closure_fn3.c,
1059 testsuite/libffi.call/cls_sshort.c,
1060 testsuite/libffi.call/closure_fn5.c,
1061 testsuite/libffi.call/cls_align_double.c,
1062 testsuite/libffi.call/cls_2byte.c,
1063 testsuite/libffi.call/nested_struct.c,
1064 testsuite/libffi.call/nested_struct10.c,
1065 testsuite/libffi.call/cls_4byte.c,
1066 testsuite/libffi.call/cls_6byte.c,
1067 testsuite/libffi.call/cls_8byte.c,
1068 testsuite/libffi.call/cls_multi_sshort.c,
1069 testsuite/libffi.call/cls_align_uint32.c,
1070 testsuite/libffi.call/cls_align_sint16.c,
1071 testsuite/libffi.call/cls_float.c,
1072 testsuite/libffi.call/cls_20byte.c,
1073 testsuite/libffi.call/cls_5_1_byte.c,
1074 testsuite/libffi.call/nested_struct2.c,
1075 testsuite/libffi.call/cls_24byte.c,
1076 testsuite/libffi.call/nested_struct4.c,
1077 testsuite/libffi.call/nested_struct6.c,
1078 testsuite/libffi.call/cls_64byte.c,
1079 testsuite/libffi.call/nested_struct8.c,
1080 testsuite/libffi.call/cls_uint.c,
1081 testsuite/libffi.call/cls_multi_ushortchar.c,
1082 testsuite/libffi.call/cls_schar.c,
1083 testsuite/libffi.call/cls_uchar.c,
1084 testsuite/libffi.call/cls_align_uint64.c,
1085 testsuite/libffi.call/cls_ulonglong.c,
1086 testsuite/libffi.call/cls_align_longdouble.c,
1087 testsuite/libffi.call/cls_1_1byte.c,
1088 testsuite/libffi.call/cls_12byte.c,
1089 testsuite/libffi.call/cls_3_1byte.c,
1090 testsuite/libffi.call/cls_3byte1.c,
1091 testsuite/libffi.call/cls_4_1byte.c,
1092 testsuite/libffi.call/cls_6_1_byte.c,
1093 testsuite/libffi.call/cls_16byte.c,
1094 testsuite/libffi.call/cls_18byte.c,
1095 testsuite/libffi.call/closure_fn0.c,
1096 testsuite/libffi.call/cls_9byte2.c,
1097 testsuite/libffi.call/closure_fn2.c,
1098 testsuite/libffi.call/closure_fn4.c,
1099 testsuite/libffi.call/cls_ushort.c,
1100 testsuite/libffi.call/closure_fn6.c,
1101 testsuite/libffi.call/cls_5byte.c,
1102 testsuite/libffi.call/cls_align_pointer.c,
1103 testsuite/libffi.call/cls_7byte.c,
1104 testsuite/libffi.call/cls_align_sint32.c,
1105 testsuite/libffi.special/unwindtest_ffi_call.cc,
1106 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1107
12a27363
L
11082007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1109
1110 * aclocal.m4: Regenerated.
1111
82a6cadf
PB
11122007-06-02 Paolo Bonzini <bonzini@gnu.org>
1113
1114 * configure: Regenerate.
1115
1cf3d07d
SE
11162007-05-23 Steve Ellcey <sje@cup.hp.com>
1117
1118 * Makefile.in: Regenerate.
1119 * configure: Regenerate.
1120 * aclocal.m4: Regenerate.
1121 * include/Makefile.in: Regenerate.
1122 * testsuite/Makefile.in: Regenerate.
1123
5635c26e
RZ
11242007-05-10 Roman Zippel <zippel@linux-m68k.org>
1125
1126 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1127 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1128 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1129 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1130 (FFI_CLOSURES): Enable closure support.
1131
e6fbf387
RZ
11322007-05-10 Roman Zippel <zippel@linux-m68k.org>
1133
1134 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1135 * configure: Regenerate.
1136 * fficonfig.h.in: Regenerate.
1137 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1138 CFI_OFFSET,CFI_DEF_CFA): New macros.
1139 (ffi_call_SYSV): Add callframe annotation.
1140
b5296e52
RZ
11412007-05-10 Roman Zippel <zippel@linux-m68k.org>
1142
1143 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1144 numerous test suite failures.
1145 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1146
bd61d708
PB
11472007-04-11 Paolo Bonzini <bonzini@gnu.org>
1148
1149 * Makefile.am (EXTRA_DIST): Bring up to date.
1150 * Makefile.in: Regenerate.
1151 * src/frv/eabi.S: Remove RCS keyword.
1152
bf8da5fc
RH
11532007-04-06 Richard Henderson <rth@redhat.com>
1154
1155 * configure.ac: Tidy target case.
1156 (HAVE_LONG_DOUBLE): Allow the target to override.
1157 * configure: Regenerate.
1158 * include/ffi.h.in: Don't define ffi_type_foo if
1159 LIBFFI_HIDE_BASIC_TYPES is defined.
1160 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1161 to ffi_type_double.
1162 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1163 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1164 (ffi_type_longdouble): Special case for Alpha. Don't define
1165 if long double == double.
1166
1167 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1168 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1169 (ffi_call, ffi_closure_osf_inner): Likewise.
1170 (ffi_closure_osf_inner): Likewise. Mark hidden.
1171 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1172 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1173 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1174 (load_table): Handle 128-bit long double.
1175
1176 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1177
8669fed1
TT
11782007-04-06 Tom Tromey <tromey@redhat.com>
1179
1180 PR libffi/31491:
1181 * README: Fixed bug in example.
1182
f0920e6c
JJ
11832007-04-03 Jakub Jelinek <jakub@redhat.com>
1184
1185 * src/closures.c: Include sys/statfs.h.
1186 (_GNU_SOURCE): Define on Linux.
1187 (FFI_MMAP_EXEC_SELINUX): Define.
1188 (selinux_enabled): New variable.
1189 (selinux_enabled_check): New function.
1190 (is_selinux_enabled): Define.
1191 (dlmmap): Use it.
1192
34a6ccda
UB
11932007-03-24 Uros Bizjak <ubizjak@gmail.com>
1194
1195 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1196 Use 'volatile float sum' to create sum of floats to avoid false
1197 negative due to excess precision on ix86 targets.
1198 (main): Ditto.
1199
8b9cf39e
AO
12002007-03-08 Alexandre Oliva <aoliva@redhat.com>
1201
1202 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1203 patch.
1204 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1205
18fa3240
AO
12062007-03-07 Alexandre Oliva <aoliva@redhat.com>
1207
1208 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1209 (ffi_prep_closure_loc): New.
1210 (ffi_prep_raw_closure_loc): New.
1211 (ffi_prep_java_raw_closure_loc): New.
1212 * src/closures.c: New file.
1213 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1214 Replace sflags with exec_offset.
1215 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1216 sub_segment_exec_offset): New macros.
1217 (get_segment_flags, set_segment_flags, check_segment_merge): New
1218 macros.
1219 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1220 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1221 destroy_mspace): Use new macros.
1222 (sys_alloc): Silence warning.
1223 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1224 * Makefile.in: Rebuilt.
1225 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1226 terms of ffi_prep_closure_loc.
1227 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1228 from...
1229 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1230 renamed version.
1231 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1232 adjusted from...
1233 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1234 the renamed version.
1235 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1236 (ffi_prep_closure): ... this.
1237 * src/pa/ffi.c: Likewise.
1238 * src/cris/ffi.c: Likewise. Adjust.
1239 * src/frv/ffi.c: Likewise.
1240 * src/ia64/ffi.c: Likewise.
1241 * src/mips/ffi.c: Likewise.
1242 * src/powerpc/ffi_darwin.c: Likewise.
1243 * src/s390/ffi.c: Likewise.
1244 * src/sh/ffi.c: Likewise.
1245 * src/sh64/ffi.c: Likewise.
1246 * src/sparc/ffi.c: Likewise.
1247 * src/x86/ffi64.c: Likewise.
1248 * src/x86/ffi.c: Likewise.
1249 (FFI_INIT_TRAMPOLINE): Adjust.
1250 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1251 (ffi_prep_raw_closure): ... this.
1252 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1253 (ffi_prep_closure): ... this.
1254 (flush_icache): Adjust.
1255
dd778331
AO
12562007-03-07 Alexandre Oliva <aoliva@redhat.com>
1257
1258 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1259 Lea's malloc.
1260
b5f8d49b
BM
12612007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1262
1263 * Makefile.am: Add dummy install-pdf target.
1264 * Makefile.in: Regenerate
1265
7a23933b
AK
12662007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1267
1268 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1269 ffi_closure_helper_SYSV): Add long double handling.
1270
8aaf4681
JJ
12712007-02-02 Jakub Jelinek <jakub@redhat.com>
1272
1273 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1274 immediately after bctrl instruction.
1275
0eb5e2b9
AO
12762007-01-18 Alexandre Oliva <aoliva@redhat.com>
1277
1278 * Makefile.am (all-recursive, install-recursive,
1279 mostlyclean-recursive, clean-recursive, distclean-recursive,
1280 maintainer-clean-recursive): Add missing targets.
1281 * Makefile.in: Rebuilt.
1282
f4f7486b
AT
12832006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1284
1285 * configure.ac: Add TARGET for x86_64-*-darwin*.
1286 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1287 for X86_DARWIN.
1288 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1289 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1290 * configure: Regenerate.
1291 * Makefile.in: Regenerate.
1292 * include/Makefile.in: Regenerate.
1293 * testsuite/Makefile.in: Regenerate.
1294 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1295 ffi_call only.
1296
3d5eafe2
AT
12972006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1298
1299 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1300 Makefile.am.
1301
d5523ce1 13022006-10-31 Geoffrey Keating <geoffk@apple.com>
58182de3
GK
1303
1304 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1305 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1306 Darwin.
1307 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1308 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1309
a3bd50cc
PB
13102006-10-10 Paolo Bonzini <bonzini@gnu.org>
1311 Sandro Tolaini <tolaini@libero.it>
1312
1313 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1314 conditional.
1315 * configure: Regenerated.
1316 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1317 (EXTRA_DIST): Add src/x86/darwin.S.
1318 * Makefile.in: Regenerated.
1319 * include/Makefile.in: Regenerated.
1320 * testsuite/Makefile.in: Regenerated.
1321
1322 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1323 X86_WIN32, and additionally align stack to 16 bytes.
1324 * src/x86/darwin.S: New, based on sysv.S.
1325 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1326
b095810f
DD
13272006-09-12 David Daney <ddaney@avtrex.com>
1328
1329 PR libffi/23935
1330 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1331 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1332 * aclocal.m4: Regenerated for automake 1.9.6.
1333 * Makefile.in: Regenerated.
1334 * include/Makefile.in: Regenerated.
1335 * testsuite/Makefile.in: Regenerated.
1336
110b2db4
AT
13372006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1338
1339 * include/ffi_common.h (struct): Revert accidental commit.
1340
8e5fde87
AT
13412006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1342
1343 * include/ffi_common.h: Remove lint directives.
1344 * include/ffi.h.in: Likewise.
1345
cb4132fe
TS
13462006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1347
1348 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1349 for 32-bit architectures.
1350 * testsuite/libffi.call/return_ul.c: New test case.
1351
0d24c3fa
DD
13522006-07-19 David Daney <ddaney@avtrex.com>
1353
1354 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1355 xfail remains for mips64.
1356
8665824c
CD
13572006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1358
1359 * Makefile.am: Add install-html target. Add install-html to .PHONY
1360 * Makefile.in: Regenerate.
1361 * aclocal.m4: Regenerate.
1362 * include/Makefile.in: Regenerate.
1363 * testsuite/Makefile.in: Regenerate.
1364
84e2e370
JDA
13652006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1366
1367 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1368 stack slot.
1369
ac6ed182
AT
13702006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1371
1372 * README: Remove notice about 'Crazy Comments'.
1373 * src/debug.c: Remove lint directives. Cleanup white spaces.
1374 * src/java_raw_api.c: Likewise.
1375 * src/prep_cif.c: Likewise.
1376 * src/raw_api.c: Likewise.
1377 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1378 to the testsuite.
1379 * src/arm/ffi.c: Remove lint directives.
1380 * src/m32r/ffi.c: Likewise.
1381 * src/pa/ffi.c: Likewise.
1382 * src/powerpc/ffi.c: Likewise.
1383 * src/powerpc/ffi_darwin.c: Likewise.
1384 * src/sh/ffi.c: Likewise.
1385 * src/sh64/ffi.c: Likewise.
1386 * src/x86/ffi.c: Likewise.
1387 * testsuite/libffi.call/float2.c: Likewise.
1388 * testsuite/libffi.call/promotion.c: Likewise.
1389 * testsuite/libffi.call/struct1.c: Likewise.
1390
b7211585
AT
13912006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1392
1393 * src/pa/hpux32.S: Correct unwind offset calculation for
1394 ffi_closure_pa32.
1395 * src/pa/linux.S: Likewise.
1396
86066f9b
JW
13972006-04-12 James E Wilson <wilson@specifix.com>
1398
1399 PR libgcj/26483
1400 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1401 (hfa_type_load): Call stf_spill.
1402 (hfa_type_store): Call ldf_fill.
1403 (ffi_call): Adjust calls to above routines. Add local temps for
1404 macro result.
b7211585 1405
615baed7
MK
14062006-04-10 Matthias Klose <doko@debian.org>
1407
1408 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1409 directory names containing underscores.
1410
8e7a2b58
JW
14112006-04-07 James E Wilson <wilson@specifix.com>
1412
1413 * testsuite/libffi.call/float4.c: New testcase.
1414
e6c300ef
JDA
14152006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1416 Andreas Tobler <a.tobler@schweiz.ch>
1417
1418 * Makefile.am: Add PA_HPUX port.
1419 * Makefile.in: Regenerate.
1420 * include/Makefile.in: Likewise.
1421 * testsuite/Makefile.in: Likewise.
1422 * configure.ac: Add PA_HPUX rules.
1423 * configure: Regenerate.
1424 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1425 Add PA_HPUX and PA64_HPUX.
1426 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1427 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1428 (FFI_TYPE_SMALL_STRUCT2): Define.
1429 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1430 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1431 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1432 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1433 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1434 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1435 * src/pa/ffi.c (ROUND_DOWN): Delete.
1436 (fldw, fstw, fldd, fstd): Use '__asm__'.
1437 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1438 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1439 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1440 Simplify incrementing of stack slot variable. Change type of local
1441 'n' to unsigned int.
1442 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1443 double on PA_HPUX.
1444 (ffi_prep_cif_machdep): Likewise.
1445 (ffi_call): Likewise.
1446 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1447 return type to ffi_status. Simplify incrementing of stack slot
1448 variable. Only copy floating point argument registers when PA_LINUX
1449 is true. Reformat debug statement.
1450 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1451 FFI_TYPE_SMALL_STRUCT8.
1452 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1453 declaration.
1454 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1455 Add nops to cache flush. Add trampoline for PA_HPUX.
1456 * src/pa/hpux32.S: New file.
1457 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1458 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1459 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1460 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1461 argument type checks so that common argument types appear first.
1462 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1463 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1464
75b8b1be
AM
14652006-03-24 Alan Modra <amodra@bigpond.net.au>
1466
1467 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1468 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1469 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1470 FFI_TYPE_LONGDOUBLE.
1471 (ffi_prep_args64): Assert using IBM extended double.
1472 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1473 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1474 (ffi_call): Handle FFI_LINUX.
1475 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1476 gpr3 return pointer as for struct return. Handle FFI_LINUX
1477 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1478 unnecessarily.
1479 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1480 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1481 Don't use r6 as pointer to results, instead use sp offset. Don't
1482 make a special call to load lr with case table address, instead
1483 use offset from previous call.
1484 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1485 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1486 return.
1487
697701ad
KK
14882006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1489
1490 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1491 passed with FP registers correctly.
1492 (ffi_closure_helper_SYSV): Likewise.
1493 * src/sh64/sysv.S: Likewise.
1494
b295c471
AT
14952006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1496
1497 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1498 args and userdata unused.
1499 (closure_test_fn1): Mark cif and userdata unused.
1500 (main): Remove unused res.
1501
9fdeb13b
AT
15022006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1503
1504 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1505 -O2, -O3, -Os and the warning flags -W -Wall.
1506 * testsuite/libffi.special/special.exp: Likewise.
1507 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1508 unused parameter unused for gcc or else do nothing.
1509 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1510 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1511 and userdata unused.
1512 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1513 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1514 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1515 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1516 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1517 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1518 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1519 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1520 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1521 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1522 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1523 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1524 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1525 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1526 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1527 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1528 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1529 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1530 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1531 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1532 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1533 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1534 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1535 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1536 Likewise.
1537 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1538 Likewise.
1539 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1540 Likewise.
1541 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1542 void* to avoid compiler warning.
1543 (main): Likewise.
1544 (cls_struct_align_gn): Mark cif and userdata unused.
1545 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1546 Likewise.
1547 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1548 Likewise.
1549 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1550 Likewise.
1551 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1552 Likewise.
1553 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1554 Likewise.
1555 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1556 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1557 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1558 data unused.
1559 (main): Cast res_call to silence gcc.
1560 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1561 data unused.
1562 (main): Cast res_call to silence gcc.
1563 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1564 and data unused.
1565 (main): Cast res_call to silence gcc.
1566 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1567 data unused.
1568 (main): Cast res_call to silence gcc.
1569 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1570 data unused.
1571 (main): Cast res_call to silence gcc.
1572 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1573 and data unused.
1574 (main): Cast res_call to silence gcc.
1575 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1576 userdata unused.
1577 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1578 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1579 userdata unused.
1580 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1581 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1582 userdata unused.
1583 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1584 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1585 userdata unused.
1586 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1587 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1588 userdata unused.
1589 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1590 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1591 and userdata unused.
1592 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1593 userdata unused.
1594 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1595 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1596 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1597 Cleanup white spaces.
1598 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1599 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1600 cif and userdata unused.
1601 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1602 Likewise.
1603 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1604 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1605 formatters to silence gcc.
1606 (B_gn): Mark cif and userdata unused.
1607 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1608 unused.
1609 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1610 (B_gn): Mark cif and userdata unused.
1611 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1612 unused.
1613 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1614 (B_gn): Mark cif and userdata unused.
1615 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1616 unused.
1617 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1618 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1619 * testsuite/libffi.call/problem1.c (stub): Likewise.
1620 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1621 gcc.
1622 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1623 in the last commit for this test case in the test case itself.
1624 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1625 unused.
1626 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1627 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1628 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1629 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1630 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1631 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1632
0a4e2bb5
KK
16332006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1634
1635 * src/sh/sysv.S: Fix register numbers in the FDE for
1636 ffi_closure_SYSV.
1637
723e5d5a
AT
16382006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1639
1640 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1641 declaration to avoid a false negative on ix86. See PR323.
1642
e097f887
KK
16432006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1644
1645 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1646 and cast integer to void * if needed. Update the pointer to
1647 the FP register saved area correctly.
1648
1a4878f7
AT
16492006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1650
1651 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1652 is fixed.
1653 * testsuite/libffi.call/nested_struct4.c: Likewise.
1654
b47b5e08
AT
16552006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1656
6429d58f
AT
1657 * testsuite/libffi.call/return_dbl.c: New test case.
1658 * testsuite/libffi.call/return_dbl1.c: Likewise.
1659 * testsuite/libffi.call/return_dbl2.c: Likewise.
1660 * testsuite/libffi.call/return_fl.c: Likewise.
1661 * testsuite/libffi.call/return_fl1.c: Likewise.
1662 * testsuite/libffi.call/return_fl2.c: Likewise.
1663 * testsuite/libffi.call/return_fl3.c: Likewise.
1664 * testsuite/libffi.call/closure_fn6.c: Likewise.
1665
b47b5e08
AT
1666 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1667 definition.
1668 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1669 here to be used by other test cases too.
1670
1671 * testsuite/libffi.call/nested_struct10.c: New test case.
1672 * testsuite/libffi.call/nested_struct9.c: Likewise.
1673 * testsuite/libffi.call/nested_struct8.c: Likewise.
1674 * testsuite/libffi.call/nested_struct7.c: Likewise.
1675 * testsuite/libffi.call/nested_struct6.c: Likewise.
1676 * testsuite/libffi.call/nested_struct5.c: Likewise.
1677 * testsuite/libffi.call/nested_struct4.c: Likewise.
1678
9abe31d4
AT
16792006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1680
1681 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1682 * configure: Rebuilt.
1683
71aa1a70
JJ
16842006-01-18 Jakub Jelinek <jakub@redhat.com>
1685
1686 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1687 instead do the shifting inline.
1688 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1689 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1690 and 8 byte structs, for the remaining struct sizes don't call
1691 __lshrdi3, instead do the shifting inline.
1692
8a39029d
TS
16932005-12-07 Thiemo Seufer <ths@networkno.de>
1694
1695 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1696 missing parentheses.
1697 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1698 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1699 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1700 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1701 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1702 FA_0_0_OFF2.
1703 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1704 endianness bugs.
1705 (ffi_prep_closure): Improve trampoline instruction scheduling.
1706 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1707
cd4241aa
AM
17082005-12-03 Alan Modra <amodra@bigpond.net.au>
1709
1710 * src/powerpc/ffi.c: Formatting.
1711 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1712 (ffi_prep_args64): Likewise.
1713
0363db46
GK
17142005-09-30 Geoffrey Keating <geoffk@apple.com>
1715
1716 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1717 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1718
d674eb2f
TT
17192005-09-26 Tom Tromey <tromey@redhat.com>
1720
1721 * testsuite/libffi.call/float1.c (value_type): New typedef.
1722 (CANARY): New define.
1723 (main): Check for result buffer overflow.
1724 * src/powerpc/linux64.S: Handle linux64 long double returns.
1725 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1726 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1727
2d9c5743
AM
17282005-08-25 Alan Modra <amodra@bigpond.net.au>
1729
1730 PR target/23404
1731 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1732 homed fp args.
1733 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1734
8a42356f
JJ
17352005-08-11 Jakub Jelinek <jakub@redhat.com>
1736
1737 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1738 (AH_BOTTOM): Add FFI_HIDDEN definition.
1739 * configure: Rebuilt.
1740 * fficonfig.h.in: Rebuilt.
1741 * src/powerpc/ffi.c (hidden): Remove.
1742 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1743 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1744 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1745 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1746 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1747 add FFI_HIDDEN to its prototype.
1748 (ffi_closure_SYSV_inner): New.
1749 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1750 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1751
863c5508
AS
17522005-08-10 Alfred M. Szmidt <ams@gnu.org>
1753
1754 PR libffi/21819:
1755 * configure: Rebuilt.
1756 * configure.ac: Handle i*86-*-gnu*.
1757
6a26ea58
JJ
17582005-08-09 Jakub Jelinek <jakub@redhat.com>
1759
1760 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1761 DW_CFA_offset_extended_sf rather than
1762 DW_CFA_GNU_negative_offset_extended.
1763 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1764
583a92e8
ST
17652005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1766
1767 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1768 on sh3.
1769 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1770 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1771 partially on register.
1772 (ffi_closure_helper_SYSV): Likewise.
1773 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1774
5d0317fc
KK
17752005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1776
1777 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1778 Remove empty line.
1779 * src/sh64/ffi.c (simple_type): Remove.
1780 (return_type): Handle small structures correctly.
1781 (ffi_prep_args): Likewise.
1782 (ffi_call): Likewise.
1783 (ffi_closure_helper_SYSV): Likewise.
1784 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1785 Emit position independent code if PIC and remove wrong datalabel
1786 prefixes from EH data.
1787
8d9fd5f5
AT
17882005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1789
1790 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1791 * Makefile.in: Regenerate.
1792 * include/Makefile.in: Likewise.
1793 * testsuite/Makefile.in: Likewise.
1794 * configure.ac: Add POWERPC_FREEBSD rules.
1795 * configure: Regenerate.
1796 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1797 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1798 * src/powerpc/ffi.c: Add flags to handle small structure returns
1799 in ffi_call_SYSV.
1800 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1801 Aka FFI_SYSV.
1802 (ffi_closure_helper_SYSV): Likewise.
1803 * src/powerpc/ppc_closure.S: Add return types for small structures.
1804 * src/powerpc/sysv.S: Add bits to handle small structures for
1805 final SYSV 4 ABI.
1806
93c76f81
AT
18072005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1808
1809 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1810 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1811 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1812
b65da304
RC
18132005-07-05 Randolph Chung <tausq@debian.org>
1814
1815 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1816 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1817 structures. Kill compilation warnings.
1818 (ffi_closure_inner_LINUX): Print return values as hex in debug
1819 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1820 Properly handle 5-7 byte structure returns.
1821 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1822 (FFI_TYPE_SMALL_STRUCT2): Remove.
1823 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1824 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1825 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1826 (checksmst1, checksmst2): Remove.
1827 (checksmst3): Optimize handling of 3-byte struct returns.
1828 (checksmst567): Properly handle 5-7 byte struct returns.
1829
2572f1c6
RO
18302005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1831
1832 PR libgcj/21943
1833 * src/mips/n32.S: Enforce PIC code.
1834 * src/mips/o32.S: Likewise.
1835
cc786582
RO
18362005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1837
1838 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1839 * configure: Regenerate.
1840
6d0d376c
AM
18412005-06-01 Alan Modra <amodra@bigpond.net.au>
1842
1843 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1844 to call ffi_closure_helper_SYSV. Append @local instead.
1845 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1846
a21946f5 18472005-05-17 Kelley Cook <kcook@gcc.gnu.org>
a097ef55
KC
1848
1849 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1850 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1851 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1852 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1853 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1854
f4c767fd
MS
18552005-05-09 Mike Stump <mrs@apple.com>
1856
1857 * configure: Regenerate.
1858
7e15b5fe
RH
18592005-05-08 Richard Henderson <rth@redhat.com>
1860
1861 PR libffi/21285
1862 * src/alpha/osf.S: Update unwind into to match code.
1863
d56ea8d9
RH
18642005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1865 Richard Henderson <rth@redhat.com>
1866
1867 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1868 bit 11 of flags.
1869 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1870 (ffi_prep_closure): Set carry bit if sse-used flag set.
1871 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1872 Only load sse registers if ssecount non-zero.
1873 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1874
18752005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
d948c201
RC
1876
1877 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1878 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1879 * configure: Regenerate.
1880
fc0ad8d9
HPN
18812005-04-20 Hans-Peter Nilsson <hp@axis.com>
1882
1883 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1884 have Tcl8.3-compatible intermediate variable.
1885
614348bb
SP
18862005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1887 Hans-Peter Nilsson <hp@axis.com>
1888
1889 * Makefile.am: Add CRIS support.
1890 * configure.ac: Likewise.
29ede6db
HPN
1891 * Makefile.in, configure, testsuite/Makefile.in,
1892 include/Makefile.in: Regenerate.
614348bb
SP
1893 * src/cris: New directory.
1894 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1895 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
727e729b
HPN
1896
1897 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1898 \r?\n in output tests.
1899
6dab073b
MS
19002005-04-12 Mike Stump <mrs@apple.com>
1901
1902 * configure: Regenerate.
1903
51f55d7c
HB
19042005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1905
1906 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1907
7efa52a3
SE
19082005-03-30 Steve Ellcey <sje@cup.hp.com>
1909
1910 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1911 (ffi_sarg) Ditto.
1912 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1913 to 64 bits in ILP32 mode.
1914 Load 64 bits even for short data.
1915
1488177a
MS
19162005-03-23 Mike Stump <mrs@apple.com>
1917
1918 * src/powerpc/darwin.S: Update for -m64 multilib.
1919 * src/powerpc/darwin_closure.S: Likewise.
1920
c6900272
ZW
19212005-03-21 Zack Weinberg <zack@codesourcery.com>
1922
1923 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1924 Do not set tool_include_dir.
1925 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1926 Regenerate.
1927 * include/Makefile.am: Set gcc_version and toollibffidir.
1928 * include/Makefile.in: Regenerate.
1929
962eea0a
AH
19302005-02-22 Andrew Haley <aph@redhat.com>
1931
1932 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1933 odd-numbered register pairs for 64-bit integer types.
1934
1b05e8bf
AT
19352005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1936
1937 PR libffi/20104
1938 * testsuite/libffi.call/return_ll1.c: New test case.
1939
6027770b
JJ
19402005-02-11 Janis Johnson <janis187@us.ibm.com>
1941
1942 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1943 * testsuite/libffi.call/float.c: Ditto.
1944 * testsuite/libffi.call/float2.c: Ditto.
1945 * testsuite/libffi.call/float3.c: Ditto.
1946
ee31ea68
AT
19472005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1948
1949 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1950
c16be7d5
EB
19512005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1952
1953 * testsuite/libffi.special/special.exp (cxx_options): Add
1954 -shared-libgcc.
1955
5df07255
RH
19562004-12-31 Richard Henderson <rth@redhat.com>
1957
1958 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1959 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1960 offset parameters with a type parameter; deduce size and structure
1961 alignment. Update all users.
1962
81a69b13
RH
19632004-12-31 Richard Henderson <rth@redhat.com>
1964
1965 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1966 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
c6900272 1967 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
81a69b13
RH
1968 into ffi_prep_closure.
1969 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1970 from scratch.
1971
d833cfb0
RH
19722004-12-27 Richard Henderson <rth@redhat.com>
1973
1974 * src/x86/unix64.S: Fix typo in unwind info.
1975
1a0f488c
RH
19762004-12-25 Richard Henderson <rth@redhat.com>
1977
1978 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1979 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1980 (merge_classes): Check for it.
1981 (SSE_CLASS_P): New.
1982 (classify_argument): Pass byte_offset by value; perform all updates
1983 inside struct case.
1984 (examine_argument): Add classes argument; handle
1985 X86_64_COMPLEX_X87_CLASS.
1986 (ffi_prep_args): Merge into ...
1987 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1988 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1989 (ffi_fill_return_value): Remove.
1990 (ffi_prep_closure): Remove dead assert.
1991 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1992 Rewrite to use struct register_args instead of va_list. Create
1993 flags for handling structure returns.
1994 * src/x86/unix64.S: Remove dead strings.
1995 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1996 stack frame with ffi_call. Handle structure returns properly.
1997 (float2sse, floatfloat2sse, double2sse): Remove.
1998 (sse2float, sse2double, sse2floatfloat): Remove.
1999 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2000 to handle structure returns properly.
2001
7994d50c
DE
20022004-12-08 David Edelsohn <edelsohn@gnu.org>
2003
2004 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2005 PICFLAG.
2006 * Makefile.in: Regenerated.
2007
afabd8e6
RS
20082004-12-02 Richard Sandiford <rsandifo@redhat.com>
2009
2010 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2011 * configure, aclocal.m4, Makefile.in: Regenerate.
2012 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2013
2539bffe
KC
20142004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2015
2016 * configure: Regenerate for libtool change.
2017
767f726e
KC
20182004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2019
2020 * configure: Regenerate for libtool reversion.
2021
2c92e90f
KC
20222004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2023
2024 * configure: Regenerate for libtool change.
2025
1df8e834
JDA
20262004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2027
2028 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2029
20302004-11-23 Richard Sandiford <rsandifo@redhat.com>
2031
2032 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2033 of jal. Use an absolute encoding for the frame information.
2034
20352004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2036
2037 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2038 * acinclude.m4: Delete logic for sincludes.
2039 * aclocal.m4, Makefile.in, configure: Regenerate.
2040 * include/Makefile: Likewise.
2041 * testsuite/Makefile: Likewise.
2042
bf02e7fa
EB
20432004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2044
2045 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2046 on a 8-byte boundary.
2047 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2048
f20459f1
RE
20492004-10-27 Richard Earnshaw <rearnsha@arm.com>
2050
2051 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2052 long long values. Round stack allocation to a multiple of 8 bytes
2053 for ATPCS compatibility.
2054 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2055 names. Handle returning long long types. Add Thumb and interworking
2056 support. Improve soft-float code.
2057
fdc69597
RE
20582004-10-27 Richard Earnshaw <rearnsha@arm.com>
2059
2060 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2061 (libffi_exit): New function.
2062 (libffi_init): Build the testglue wrapper if needed.
2063
1e02510f
EB
20642004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2065
2066 PR other/18138
2067 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2068
d9a193b2
KI
20692004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2070
2071 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2072
1fddb058
KK
20732004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2074
2075 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2076 * testsuite/libffi.call/float3.c: New test case.
2077
e0d3b452
KK
20782004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2079
2080 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2081 the function returning a structure pointed with R2.
2082 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2083 the structure return value if T bit set. Emit position
2084 independent code and EH data if PIC.
2085
a472e9d4
KI
20862004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2087
2088 * Makefile.am: Add m32r support.
2089 * configure.ac: Likewise.
2090 * Makefile.in: Regenerate.
2091 * confiugre: Regenerate.
2092 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2093 (uint64, sint64, double, longdouble)
2094 * src/m32r: New directory.
2095 * src/m32r/ffi.c: New file.
2096 * src/m32r/sysv.S: Likewise.
2097 * src/m32r/ffitarget.h: Likewise.
2098
3d47e9ca
KK
20992004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2100
2101 * testsuite/libffi.call/negint.c: New test case.
2102
fc9051dd
L
21032004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2104
2105 PR libgcj/17465
2106 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2107 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2108 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2109 DYLD_LIBRARY_PATH.
2110
2010caea
AT
21112004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2112
2113 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2114 * testsuite/libffi.call/promotion.c: Likewise.
2115 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2116 whitespaces.
2117 * testsuite/libffi.call/return_sc.c: Likewise.
2118 * testsuite/libffi.call/return_uc.c: Likewise.
2119
894be816
AT
21202004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2121
2122 * src/powerpc/darwin.S: Fix comments and identation.
2123 * src/powerpc/darwin_closure.S: Likewise.
2124
2fe7404a
AT
21252004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2126
2127 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2128 (ffi_prep_args): Handle longdouble arguments.
2129 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2130 longdouble.
2131 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2132 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2133 values.
2134 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2135 * src/types.c: Defined longdouble size and alignment for darwin.
2136
16070e45
AT
21372004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2138
2139 * src/powerpc/aix.S: Remove whitespaces.
2140 * src/powerpc/aix_closure.S: Likewise.
2141 * src/powerpc/asm.h: Likewise.
2142 * src/powerpc/ffi.c: Likewise.
2143 * src/powerpc/ffitarget.h: Likewise.
2144 * src/powerpc/linux64.S: Likewise.
2145 * src/powerpc/linux64_closure.S: Likewise.
2146 * src/powerpc/ppc_closure.S: Likewise.
2147 * src/powerpc/sysv.S: Likewise.
2148
88d8c771
AG
21492004-08-30 Anthony Green <green@redhat.com>
2150
2151 * Makefile.am: Add frv support.
2152 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2153 * configure.ac: Read configure.host.
2154 * configure.in: Read configure.host.
2155 * configure.host: New file. frv-elf needs libgloss.
2156 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
a23adb4d 2157 alignment. This is needed to frv and shouldn't harm the others.
88d8c771
AG
2158 * include/ffi_common.h (ALIGN_DOWN): New macro.
2159 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2160
4c5f37d5
DD
21612004-08-24 David Daney <daney@avtrex.com>
2162
2163 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2164 * testsuite/libffi.call/closure_fn1.c: Likewise.
2165 * testsuite/libffi.call/closure_fn2.c Likewise.
2166 * testsuite/libffi.call/closure_fn3.c: Likewise.
2167 * testsuite/libffi.call/closure_fn4.c: Likewise.
2168 * testsuite/libffi.call/closure_fn5.c: Likewise.
2169 * testsuite/libffi.call/cls_18byte.c: Likewise.
2170 * testsuite/libffi.call/cls_19byte.c: Likewise.
2171 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2172 * testsuite/libffi.call/cls_20byte.c: Likewise.
2173 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2174 * testsuite/libffi.call/cls_24byte.c: Likewise.
2175 * testsuite/libffi.call/cls_2byte.c: Likewise.
2176 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2177 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2178 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2179 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2180 * testsuite/libffi.call/cls_4byte.c: Likewise.
2181 * testsuite/libffi.call/cls_64byte.c: Likewise.
2182 * testsuite/libffi.call/cls_6byte.c: Likewise.
2183 * testsuite/libffi.call/cls_7byte.c: Likewise.
2184 * testsuite/libffi.call/cls_8byte.c: Likewise.
2185 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2186 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2187 * testsuite/libffi.call/cls_align_double.c: Likewise.
2188 * testsuite/libffi.call/cls_align_float.c: Likewise.
2189 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2190 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2191 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2192 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2193 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2194 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2195 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2196 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2197 * testsuite/libffi.call/cls_double.c: Likewise.
2198 * testsuite/libffi.call/cls_float.c: Likewise.
2199 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2200 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2201 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2202 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2203 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2204 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2205 * testsuite/libffi.call/cls_schar.c: Likewise.
2206 * testsuite/libffi.call/cls_sint.c: Likewise.
2207 * testsuite/libffi.call/cls_sshort.c: Likewise.
2208 * testsuite/libffi.call/cls_uchar.c: Likewise.
2209 * testsuite/libffi.call/cls_uint.c: Likewise.
2210 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2211 * testsuite/libffi.call/cls_ushort.c: Likewise.
2212 * testsuite/libffi.call/nested_struct.c: Likewise.
2213 * testsuite/libffi.call/nested_struct1.c: Likewise.
2214 * testsuite/libffi.call/nested_struct2.c: Likewise.
2215 * testsuite/libffi.call/nested_struct3.c: Likewise.
2216 * testsuite/libffi.call/problem1.c: Likewise.
2217 * testsuite/libffi.special/unwindtest.cc: Likewise.
2218 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2219 to zero.
2220 * testsuite/libffi.call/cls_16byte.c: Likewise.
2221 * testsuite/libffi.call/cls_5byte.c: Likewise.
2222
39dca114
DD
22232004-08-23 David Daney <daney@avtrex.com>
2224
2225 PR libgcj/13141
2226 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2227 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2228 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2229 parameters and return types.
2230 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2231 (ffi_prep_closure): Ditto.
2232 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2233 alignment calculations.
2234 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2235 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2236
b790003a
CM
22372004-08-14 Casey Marshall <csm@gnu.org>
2238
2239 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2240 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2241 integer (O32 ABI only).
2242 (ffi_prep_closure): new function.
2243 (ffi_closure_mips_inner_O32): new function.
c6900272 2244 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
b790003a
CM
2245 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2246 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2247 64 bit integers correctly.
2248 (ffi_closure_O32): new function.
2249 Added DWARF-2 unwind info for both functions.
2250
547d71a1
AH
22512004-08-10 Andrew Haley <aph@redhat.com>
2252
2253 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2254
98a5bb3b
RM
22552004-08-01 Robert Millan <robertmh@gnu.org>
2256
2257 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2258 * configure: Regenerate.
2259
4d447364
MR
22602004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2261
2262 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2263 and mmap() explicitly instead of relying on preset autoconf cache
2264 variables.
2265 * aclocal.m4: Regenerate.
2266 * configure: Regenerate.
2267
e5dce82f
UW
22682004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2269
2270 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2271 (ffi_check_float_struct): Remove unused prototype.
2272
e6a0022b
GK
22732004-06-30 Geoffrey Keating <geoffk@apple.com>
2274
2275 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2276 character on Darwin, use '\n\t' instead.
2277
5a2d52ca
MK
22782004-06-26 Matthias Klose <doko@debian.org>
2279
2280 * libtool-version: Fix typo in revision/age.
2281
401844ef
MK
22822004-06-17 Matthias Klose <doko@debian.org>
2283
2284 * libtool-version: New.
2285 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2286 * Makefile.in: Regenerate.
2287
94632047 22882004-06-15 Paolo Bonzini <bonzini@gnu.org>
076e5443
PB
2289
2290 * Makefile.am: Remove useless multilib rules.
2291 * Makefile.in: Regenerate.
2292 * aclocal.m4: Regenerate with automake 1.8.5.
2293 * configure.ac: Remove useless multilib configury.
2294 * configure: Regenerate.
2295
94632047 22962004-06-15 Paolo Bonzini <bonzini@gnu.org>
9ac926b1
PB
2297
2298 * .cvsignore: New file.
2299
37c849e4
JJ
23002004-06-10 Jakub Jelinek <jakub@redhat.com>
2301
2302 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2303 fp_done.
2304 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2305 changed from 8.
2306
cb623172
SM
23072004-06-06 Sean McNeil <sean@mcneil.com>
2308
2309 * configure.ac: Add x86_64-*-freebsd* support.
2310 * configure: Regenerate.
2311
c9b93377
JB
23122004-04-26 Joe Buck <jbuck@welsh-buck.org>
2313
2314 Bug 15093
2315 * configure.ac: Test for existence of mmap and sys/mman.h before
2316 checking blacklist. Fix suggested by Jim Wilson.
2317 * configure: Regenerate.
2318
cf90fd1b
MA
23192004-04-26 Matt Austern <austern@apple.com>
2320
2321 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2322 FDE location.
2323 * src/powerpc/darwin_closure.S: Likewise.
c6900272 2324
67781b44
AT
23252004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2326
2327 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2328 error. Reported by Thomas Heller <theller@python.net>.
2329 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2330 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2331
14b078a8
MK
23322004-03-20 Matthias Klose <doko@debian.org>
2333
2334 * src/pa/linux.S: Fix typo.
2335
d65a810f
MK
23362004-03-19 Matthias Klose <doko@debian.org>
2337
67781b44 2338 * Makefile.am: Update.
d65a810f
MK
2339 * Makefile.in: Regenerate.
2340 * src/pa/ffi.h.in: Remove.
2341 * src/pa/ffitarget.h: New file.
2342
23432004-02-10 Randolph Chung <tausq@debian.org>
2344
2345 * Makefile.am: Add PA support.
2346 * Makefile.in: Regenerate.
2347 * include/Makefile.in: Regenerate.
2348 * configure.ac: Add PA target.
2349 * configure: Regenerate.
2350 * src/pa/ffi.c: New file.
2351 * src/pa/ffi.h.in: Add PA support.
2352 * src/pa/linux.S: New file.
2353 * prep_cif.c: Add PA support.
2354
09cababc
HY
23552004-03-16 Hosaka Yuji <hos@tamanegi.org>
2356
2357 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2358 double.
2359 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2360 with ecif->cif->flags.
2361 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2362 with cif->flags.
2363 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2364 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2365 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2366 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2367
0e8d0c5f
KC
23682004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2369
2370 * configure.in: Rename file to ...
2371 * configure.ac: ... this.
2372 * fficonfig.h.in: Regenerate.
2373 * Makefile.in: Regenerate.
2374 * include/Makefile.in: Regenerate.
2375 * testsuite/Makefile.in: Regenerate.
2376
b8d50739
MA
23772004-03-12 Matt Austern <austern@apple.com>
2378
2379 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2380 changes in EH format resulting from addition of linkonce support.
2381 * src/powerpc/darwin_closure.S: Likewise.
09cababc 2382
3ed8e5b6
AT
23832004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2384 Paolo Bonzini <bonzini@gnu.org>
2385
2386 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2387 Remove VPATH. Remove rules for object files. Remove multilib support.
2388 (AM_CCASFLAGS): Add.
2389 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2390 (AC_PREREQ): Bump version to 2.59.
2391 (AC_INIT): Fill with version info and bug address.
2392 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2393 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2394 De-precious CC so that the right flags are passed down to multilibs.
2395 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2396 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2397 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2398 * configure: Rebuilt.
2399 * aclocal.m4: Likewise.
2400 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2401 * fficonfig.h.in: Likewise.
2402
254ea477
AS
24032004-03-11 Andreas Schwab <schwab@suse.de>
2404
2405 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2406 arguments from fp registers only for the first 8 parameter slots.
2407 Don't convert a float parameter when passed in memory.
2408
3beb1315
HPN
24092004-03-09 Hans-Peter Nilsson <hp@axis.com>
2410
2411 * configure: Regenerate for config/accross.m4 correction.
2412
fecf735c
MK
24132004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2414
2415 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2416 ecif->cif->bytes to bytes.
2417 (ffi_prep_cif_machdep): Add braces around nested if statement.
2418
4243752c
AM
24192004-02-09 Alan Modra <amodra@bigpond.net.au>
2420
12001e0f
AM
2421 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2422
4243752c
AM
2423 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2424 (ffi_closure_helper_LINUX64): Fix typo.
2425 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2426 for powerpc64-*-*.
2427 * testsuite/libffi.call/float.c: Likewise.
2428 * testsuite/libffi.call/float2.c: Likewise.
2429
bf310028
AM
24302004-02-08 Alan Modra <amodra@bigpond.net.au>
2431
2432 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2433 long double function return and long double arg handling.
2434 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2435 Use "end_pfr" instead of "nf". Correct long double handling.
2436 Localise "temp".
2437 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2438 return value.
2439 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2440 space for long double return value. Adjust stack frame and offsets.
2441 Load f2 long double return.
2442
dd5f9ea2
AM
24432004-02-07 Alan Modra <amodra@bigpond.net.au>
2444
2445 * src/types.c: Use 16 byte long double for POWERPC64.
2446
0ce78f01
EB
24472004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2448
2449 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2450 when the structure return address is passed in %o0.
2451 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2452 (ffi_v9_layout_struct): Align the field following a nested structure
2453 on a word boundary. Use memmove instead of memcpy.
2454 (ffi_call): Update call to ffi_V9_return_struct.
2455 (ffi_prep_closure): Define 'ctx' only for V8.
2456 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2457 and ffi_closure_sparc_inner_v9.
2458 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2459 Always skip the structure return address. For structures and long
2460 doubles, copy the argument directly.
2461 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2462 if required. Shift the maximum floating-point slot accordingly. For
2463 big structures, copy the argument directly; otherwise, left-justify the
2464 argument and call ffi_v9_layout_struct to lay out the structure on
2465 the stack.
2466 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2467 (ffi_closure_v8): Pass the structure return address. Update call to
2468 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2469 Skip the 'unimp' insn when returning long doubles and structures.
2470 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2471 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2472 FFI_TYPE_INT handling. Load structures both in integers and
2473 floating-point registers on return.
2474 * README: Update status of the SPARC port.
2475
7fcc1d9b
AT
24762004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2477
2478 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2479 as of type ffi_arg.
2480 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2481
6ed0ecaf
UW
24822004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2483
2484 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2485 value as of type ffi_arg, not unsigned int.
2486
c930d8a5
MR
24872004-01-21 Michael Ritzert <ritzert@t-online.de>
2488
2489 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2490 of the LHS.
2491
7b2e3e60
AT
24922004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2493
2494 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2495 Solaris.
2496
2e89756b
RO
24972004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2498
2499 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2500 to void *.
2501
628b26f8
RH
25022003-12-10 Richard Henderson <rth@redhat.com>
2503
2504 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2505 size_t instead of int.
2506
0ee646d6
HY
25072003-12-04 Hosaka Yuji <hos@tamanegi.org>
2508
2509 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2510 * testsuite/libffi.call/many_win32.c (main): Replace variable
2511 int i with unsigned long ul.
2512
2513 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2514 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2515 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2516 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2517 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2518 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2519 * testsuite/libffi.call/cls_align_float.c: Likewise.
2520 * testsuite/libffi.call/cls_align_double.c: Likewise.
2521 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2522 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2523
2c4ed456
HY
25242003-12-02 Hosaka Yuji <hos@tamanegi.org>
2525
2526 PR other/13221
2527 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2528 Align arguments to 32 bits.
2529
d18c3386
AT
25302003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2531
2532 PR other/13221
2533 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2534 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2535 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2536 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2537 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2538 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2539
2540 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2541
88096b2a
KG
25422003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2543
2544 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2545 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2546
3eee102e
AT
25472003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2548
2549 * Makefile.in: Rebuilt.
2550 * configure: Likewise.
2551 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2552 the right type.
2553
5196736e
AJ
25542003-11-21 Andreas Jaeger <aj@suse.de>
2555 Andreas Tobler <a.tobler@schweiz.ch>
2556
2557 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2558 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2559 * Makefile.in: Rebuilt.
2560 * aclocal.m4: Likewise.
2561 * configure: Likewise.
2562 * fficonfig.h.in: Likewise.
2563 * testsuite/lib/libffi-dg.exp: Add include dir.
2564 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2565 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2566 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2567 for ffi_closure if available.
2568 * testsuite/libffi.call/closure_fn1.c: Likewise.
2569 * testsuite/libffi.call/closure_fn2.c: Likewise.
2570 * testsuite/libffi.call/closure_fn3.c: Likewise.
2571 * testsuite/libffi.call/closure_fn4.c: Likewise.
2572 * testsuite/libffi.call/closure_fn5.c: Likewise.
2573 * testsuite/libffi.call/cls_12byte.c: Likewise.
2574 * testsuite/libffi.call/cls_16byte.c: Likewise.
2575 * testsuite/libffi.call/cls_18byte.c: Likewise.
2576 * testsuite/libffi.call/cls_19byte.c: Likewise.
2577 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2578 * testsuite/libffi.call/cls_20byte.c: Likewise.
2579 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2580 * testsuite/libffi.call/cls_24byte.c: Likewise.
2581 * testsuite/libffi.call/cls_2byte.c: Likewise.
2582 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2583 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2584 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2585 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2586 * testsuite/libffi.call/cls_4byte.c: Likewise.
2587 * testsuite/libffi.call/cls_5byte.c: Likewise.
2588 * testsuite/libffi.call/cls_64byte.c: Likewise.
2589 * testsuite/libffi.call/cls_6byte.c: Likewise.
2590 * testsuite/libffi.call/cls_7byte.c: Likewise.
2591 * testsuite/libffi.call/cls_8byte.c: Likewise.
2592 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2593 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2594 * testsuite/libffi.call/cls_double.c: Likewise.
2595 * testsuite/libffi.call/cls_float.c: Likewise.
2596 * testsuite/libffi.call/cls_schar.c: Likewise.
2597 * testsuite/libffi.call/cls_sint.c: Likewise.
2598 * testsuite/libffi.call/cls_sshort.c: Likewise.
2599 * testsuite/libffi.call/cls_uchar.c: Likewise.
2600 * testsuite/libffi.call/cls_uint.c: Likewise.
2601 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2602 * testsuite/libffi.call/cls_ushort.c: Likewise.
2603 * testsuite/libffi.call/nested_struct.c: Likewise.
2604 * testsuite/libffi.call/nested_struct1.c: Likewise.
2605 * testsuite/libffi.call/nested_struct2.c: Likewise.
2606 * testsuite/libffi.call/nested_struct3.c: Likewise.
2607 * testsuite/libffi.call/problem1.c: Likewise.
2608 * testsuite/libffi.special/unwindtest.cc: Likewise.
2609
9e6c3ecb
AT
26102003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2611
2612 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2613
70d8f2a1
AT
26142003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2615
2616 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2617 Add -lgcc_s to additional flags.
2618
ea81d2a3
AT
26192003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2620
2621 * configure.in, include/Makefile.am: PR libgcj/11147, install
2622 the ffitarget.h header file in a gcc versioned and target
2623 dependent place.
2624 * configure: Regenerated.
2625 * Makefile.in, include/Makefile.in: Likewise.
2626 * testsuite/Makefile.in: Likewise.
2627
647af4a9
AT
26282003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2629
2630 * testsuite/libffi.call/closure_fn0.c: Print result and check
2631 with dg-output to make debugging easier.
2632 * testsuite/libffi.call/closure_fn1.c: Likewise.
2633 * testsuite/libffi.call/closure_fn2.c: Likewise.
2634 * testsuite/libffi.call/closure_fn3.c: Likewise.
2635 * testsuite/libffi.call/closure_fn4.c: Likewise.
2636 * testsuite/libffi.call/closure_fn5.c: Likewise.
2637 * testsuite/libffi.call/cls_12byte.c: Likewise.
2638 * testsuite/libffi.call/cls_16byte.c: Likewise.
2639 * testsuite/libffi.call/cls_18byte.c: Likewise.
2640 * testsuite/libffi.call/cls_19byte.c: Likewise.
2641 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2642 * testsuite/libffi.call/cls_20byte.c: Likewise.
2643 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2644 * testsuite/libffi.call/cls_24byte.c: Likewise.
2645 * testsuite/libffi.call/cls_2byte.c: Likewise.
2646 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2647 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2648 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2649 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2650 * testsuite/libffi.call/cls_4byte.c: Likewise.
2651 * testsuite/libffi.call/cls_5byte.c: Likewise.
2652 * testsuite/libffi.call/cls_64byte.c: Likewise.
2653 * testsuite/libffi.call/cls_6byte.c: Likewise.
2654 * testsuite/libffi.call/cls_7byte.c: Likewise.
2655 * testsuite/libffi.call/cls_8byte.c: Likewise.
2656 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2657 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2658 * testsuite/libffi.call/cls_double.c: Likewise.
2659 * testsuite/libffi.call/cls_float.c: Likewise.
2660 * testsuite/libffi.call/cls_schar.c: Likewise.
2661 * testsuite/libffi.call/cls_sint.c: Likewise.
2662 * testsuite/libffi.call/cls_sshort.c: Likewise.
2663 * testsuite/libffi.call/cls_uchar.c: Likewise.
2664 * testsuite/libffi.call/cls_uint.c: Likewise.
2665 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2666 * testsuite/libffi.call/cls_ushort.c: Likewise.
2667 * testsuite/libffi.call/problem1.c: Likewise.
2668
2669 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2670 static.
2671
ac47cc13
AT
26722003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2673
2674 * testsuite/libffi.call/cls_9byte2.c: New test case.
2675 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2676 * testsuite/libffi.call/cls_64byte.c: Likewise.
2677 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2678 * testsuite/libffi.call/cls_19byte.c: Likewise.
2679 * testsuite/libffi.call/cls_18byte.c: Likewise.
2680 * testsuite/libffi.call/closure_fn4.c: Likewise.
2681 * testsuite/libffi.call/closure_fn5.c: Likewise.
2682 * testsuite/libffi.call/cls_schar.c: Likewise.
2683 * testsuite/libffi.call/cls_sint.c: Likewise.
2684 * testsuite/libffi.call/cls_sshort.c: Likewise.
2685 * testsuite/libffi.call/nested_struct2.c: Likewise.
2686 * testsuite/libffi.call/nested_struct3.c: Likewise.
2687
f145ed6b
AT
26882003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2689
2690 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2691 * testsuite/libffi.call/cls_uchar.c: Likewise.
2692 * testsuite/libffi.call/cls_uint.c: Likewise.
2693 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2694 * testsuite/libffi.call/cls_ushort.c: Likewise.
2695 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2696
2b04d015
AT
26972003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2698
2699 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2700 the initialization.
2701
57f542b3
AT
27022003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2703
2b04d015 2704 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
57f542b3
AT
2705 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2706
e61acf72
DD
27072003-10-22 David Daney <ddaney@avtrex.com>
2708
2709 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2710 __attribute__((__mode__(__SI__))) and friends.
2711
2461c08b
AS
27122003-10-22 Andreas Schwab <schwab@suse.de>
2713
2714 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2715
1450eb7a
AT
27162003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2717
2718 * configure.in: AC_LINK_FILES(ffitarget.h).
2719 * configure: Regenerate.
2720 * Makefile.in: Likewise.
2721 * include/Makefile.in: Likewise.
2722 * testsuite/Makefile.in: Likewise.
2723 * fficonfig.h.in: Likewise.
2724
27252003-10-21 Paolo Bonzini <bonzini@gnu.org>
2726 Richard Henderson <rth@redhat.com>
2727
2728 Avoid that ffi.h includes fficonfig.h.
2729
2730 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2731 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2732 (TARGET_SRC_MIPS_SGI): Removed.
2733 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2734 (MIPS_SGI): Removed.
2735 (CLEANFILES): Removed.
2736 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2737 targets.
2738 * acconfig.h: Removed.
2739 * configure.in: Compute sizeofs only for double and long double.
2740 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2741 into AC_DEFINE instead of using acconfig.h. Create
2742 include/ffitarget.h instead of include/fficonfig.h. Rename
2743 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2744 AC_DEFINE EH_FRAME_FLAGS.
2745 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2746 (hack_DATA): Add ffitarget.h.
2747 * include/ffi.h.in: Remove all system specific definitions.
2748 Declare raw API even if it is not installed, why bother?
2749 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2750 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2751 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2752 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2753 * include/ffi_common.h (bool): Do not define.
2754 (ffi_assert): Accept failed assertion.
2755 (ffi_type_test): Return void and accept file/line.
2756 (FFI_ASSERT): Pass stringized failed assertion.
2757 (FFI_ASSERT_AT): New macro.
2758 (FFI_ASSERT_VALID_TYPE): New macro.
2759 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2760 UINT64, SINT64): Define here with gcc's __attribute__ macro
2761 instead of in ffi.h
2762 (FLOAT32, ALIGN): Define here instead of in ffi.h
2763 * include/ffi-mips.h: Removed. Its content moved to
2764 src/mips/ffitarget.h after separating assembly and C sections.
2765 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2766 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2767 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2768 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2769 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2770 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2771 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2772 (ffi_type_test): Rewritten.
2773 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2774 FFI_ASSERT_VALID_TYPE.
2775 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2776 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2777 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2778 src/s390/ffitarget.h, src/sh/ffitarget.h,
2779 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2780 src/x86/ffitarget.h: New files.
2781 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2782 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2783 src/powerpc/aix.S, src/powerpc/darwin.S,
2784 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2785 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2786 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2787 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2788 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2789 include fficonfig.h
2790
27d54b2a
RO
27912003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2792
2793 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2794 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2795
0df6970d
AT
27962003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2797
2798 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2799 Used when FFI_DEBUG = 1.
2800
2612ffce
AM
28012003-10-14 Alan Modra <amodra@bigpond.net.au>
2802
2803 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2804 and align.
2805
d54f7354
RO
28062003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2807
2808 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2809 FFI_MIPS_O32 for O32 ABI.
2810
7a54c850
AT
28112003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2812
2813 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2814 SPARC64. Cleanup whitespaces.
2815
9e3bfa9b
AT
28162003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2817
2818 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2819 strongarm, xscale. Cleanup whitespaces.
2820 * testsuite/libffi.call/closure_fn1.c: Likewise.
2821 * testsuite/libffi.call/closure_fn2.c: Likewise.
2822 * testsuite/libffi.call/closure_fn3.c: Likewise.
2823 * testsuite/libffi.call/cls_12byte.c: Likewise.
2824 * testsuite/libffi.call/cls_16byte.c: Likewise.
2825 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2826 * testsuite/libffi.call/cls_20byte.c: Likewise.
2827 * testsuite/libffi.call/cls_24byte.c: Likewise.
2828 * testsuite/libffi.call/cls_2byte.c: Likewise.
2829 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2830 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2831 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2832 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2833 * testsuite/libffi.call/cls_4byte.c: Likewise.
2834 * testsuite/libffi.call/cls_5byte.c: Likewise.
2835 * testsuite/libffi.call/cls_6byte.c: Likewise.
2836 * testsuite/libffi.call/cls_7byte.c: Likewise.
2837 * testsuite/libffi.call/cls_8byte.c: Likewise.
2838 * testsuite/libffi.call/cls_double.c: Likewise.
2839 * testsuite/libffi.call/cls_float.c: Likewise.
2840 * testsuite/libffi.call/cls_uchar.c: Likewise.
2841 * testsuite/libffi.call/cls_uint.c: Likewise.
2842 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2843 * testsuite/libffi.call/cls_ushort.c: Likewise.
2844 * testsuite/libffi.call/nested_struct.c: Likewise.
2845 * testsuite/libffi.call/nested_struct1.c: Likewise.
2846 * testsuite/libffi.call/problem1.c: Likewise.
2847 * testsuite/libffi.special/unwindtest.cc: Likewise.
2848 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2849
3e86b595
DE
28502003-09-18 David Edelsohn <edelsohn@gnu.org>
2851
2852 * src/powerpc/aix.S: Cleanup whitespaces.
2853 * src/powerpc/aix_closure.S: Likewise.
2854
8d8369f8
AT
28552003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2856
2857 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2858 * src/powerpc/darwin_closure.S: Likewise.
2859 * src/powerpc/ffi_darwin.c: Likewise.
2860
69b182fd
AT
28612003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2862 David Edelsohn <edelsohn@gnu.org>
2863
2864 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2865 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2866 parameter stack.
2867 * src/powerpc/darwin_closure.S: Likewise.
2868 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2869 according to the Darwin/AIX ABI.
2870 (ffi_prep_cif_machdep): Likewise.
2871 (ffi_closure_helper_DARWIN): Likewise.
2872 Remove the outgoing parameter stack logic. Simplify the evaluation
2873 of the different CASE types.
2874 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2875 statement in the trampoline code.
2876
df805cdc
KK
28772003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2878
2879 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2880 for the register size.
2881 (ffi_closure_helper_SYSV): Handle the structure return value
2882 address correctly.
2883 (ffi_closure_helper_SYSV): Return the appropriate type when
2884 the registers are used for the structure return value.
2885 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2886 the 64-bit return value. Update copyright years.
2887
eb26c76c
RO
28882003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2889
2890 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2891 srcdir for ffi_mips.h.
2892
6350c02c
AM
28932003-09-12 Alan Modra <amodra@bigpond.net.au>
2894
2895 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2896 structure size.
2897 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2898 placement of float result.
2899 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2900 cast of "resp" for big-endian 64 bit machines.
2901
5af1c806
AM
29022003-09-11 Alan Modra <amodra@bigpond.net.au>
2903
2904 * src/types.c (double, longdouble): Merge identical SH and ARM
2905 typedefs, and add POWERPC64.
2906 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2907 struct split over gpr and rest.
2908 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2909 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2910
086a4bd7
AT
29112003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2912
2913 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2914 passing correctly.
c930d8a5 2915
e517f636
AM
29162003-09-09 Alan Modra <amodra@bigpond.net.au>
2917
2918 * configure: Regenerate.
2919
6f20d627
AT
29202003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2921
2922 * Makefile.am: Remove build rules for ffitest.
2923 * Makefile.in: Rebuilt.
c930d8a5 2924
6f20d627 29252003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
226660af
AT
2926
2927 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2928 about implicit declaration of abort().
2929
bbf3057b
AT
29302003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2931
2932 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2933 * Makefile.in: Rebuilt.
2934 * configure.in: Add dejagnu test framework.
2935 * configure: Rebuilt.
2936
2937 * testsuite/Makefile.am: New file.
2938 * testsuite/Makefile.in: Built
c930d8a5 2939 * testsuite/lib/libffi-dg.exp: New file.
bbf3057b
AT
2940 * testsuite/config/default.exp: Likewise.
2941 * testsuite/libffi.call/call.exp: Likewise.
2942 * testsuite/libffi.call/ffitest.h: Likewise.
2943 * testsuite/libffi.call/closure_fn0.c: Likewise.
2944 * testsuite/libffi.call/closure_fn1.c: Likewise.
2945 * testsuite/libffi.call/closure_fn2.c: Likewise.
2946 * testsuite/libffi.call/closure_fn3.c: Likewise.
2947 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2948 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2949 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2950 * testsuite/libffi.call/cls_2byte.c: Likewise.
2951 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2952 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2953 * testsuite/libffi.call/cls_4byte.c: Likewise.
2954 * testsuite/libffi.call/cls_5byte.c: Likewise.
2955 * testsuite/libffi.call/cls_6byte.c: Likewise.
2956 * testsuite/libffi.call/cls_7byte.c: Likewise.
2957 * testsuite/libffi.call/cls_8byte.c: Likewise.
2958 * testsuite/libffi.call/cls_12byte.c: Likewise.
2959 * testsuite/libffi.call/cls_16byte.c: Likewise.
2960 * testsuite/libffi.call/cls_20byte.c: Likewise.
2961 * testsuite/libffi.call/cls_24byte.c: Likewise.
2962 * testsuite/libffi.call/cls_double.c: Likewise.
2963 * testsuite/libffi.call/cls_float.c: Likewise.
2964 * testsuite/libffi.call/cls_uchar.c: Likewise.
2965 * testsuite/libffi.call/cls_uint.c: Likewise.
2966 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2967 * testsuite/libffi.call/cls_ushort.c: Likewise.
2968 * testsuite/libffi.call/float.c: Likewise.
2969 * testsuite/libffi.call/float1.c: Likewise.
2970 * testsuite/libffi.call/float2.c: Likewise.
2971 * testsuite/libffi.call/many.c: Likewise.
2972 * testsuite/libffi.call/many_win32.c: Likewise.
2973 * testsuite/libffi.call/nested_struct.c: Likewise.
2974 * testsuite/libffi.call/nested_struct1.c: Likewise.
2975 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2976 * testsuite/libffi.call/problem1.c: Likewise.
2977 * testsuite/libffi.call/promotion.c: Likewise.
2978 * testsuite/libffi.call/return_ll.c: Likewise.
2979 * testsuite/libffi.call/return_sc.c: Likewise.
2980 * testsuite/libffi.call/return_uc.c: Likewise.
2981 * testsuite/libffi.call/strlen.c: Likewise.
2982 * testsuite/libffi.call/strlen_win32.c: Likewise.
2983 * testsuite/libffi.call/struct1.c: Likewise.
2984 * testsuite/libffi.call/struct2.c: Likewise.
2985 * testsuite/libffi.call/struct3.c: Likewise.
2986 * testsuite/libffi.call/struct4.c: Likewise.
2987 * testsuite/libffi.call/struct5.c: Likewise.
2988 * testsuite/libffi.call/struct6.c: Likewise.
2989 * testsuite/libffi.call/struct7.c: Likewise.
2990 * testsuite/libffi.call/struct8.c: Likewise.
c930d8a5 2991 * testsuite/libffi.call/struct9.c: Likewise.
bbf3057b
AT
2992 * testsuite/libffi.special/special.exp: New file.
2993 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2994 * testsuite/libffi.special/unwindtest.cc: Likewise.
2995
c930d8a5 2996
996b4f8f
KK
29972003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2998
2999 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3000 copyright years.
3001
b00badcd
AM
30022003-08-02 Alan Modra <amodra@bigpond.net.au>
3003
3004 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3005 structure passing.
3006 (ffi_closure_helper_LINUX64): Likewise.
3007 * src/powerpc/linux64.S: Remove code writing to parm save area.
3008 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3009 address in lr from ffi_closure_helper_LINUX64 call to calculate
3010 table address. Optimize function tail.
3011
c1516eae
AT
30122003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3013
3014 * src/sparc/ffi.c: Handle all floating point registers.
3015 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3016
fc0e6222
GP
30172003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3018
3019 * README: Note that libffi is not part of GCC. Update the project
3020 URL and status.
3021
a1569df8
FS
30222003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3023
3024 * src/powerpc/ppc_closure.S: Include ffi.h.
3025
12e816bf
RO
30262003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3027
3028 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3029 Use C style comments.
3030
e860ed6d
KK
30312003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3032
3033 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3034 * Makefile.in: Regenerate.
3035 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3036 * configure: Regenerate.
3037 * include/ffi.h.in: Add SHmedia support.
3038 * src/sh64/ffi.c: New file.
3039 * src/sh64/sysv.S: New file.
3040
a733b15e
JJ
30412003-05-16 Jakub Jelinek <jakub@redhat.com>
3042
3043 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3044 should be read-only.
3045 * configure: Rebuilt.
3046 * fficonfig.h.in: Rebuilt.
3047 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3048 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3049 * src/powerpc/linux64.S: Likewise.
3050 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3051 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3052 if -fpic/-fPIC/-mrelocatable.
3053 * src/powerpc/powerpc_closure.S: Likewise.
3054 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3055 #write in .eh_frame flags.
3056 * src/sparc/v9.S: Likewise.
3057 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3058 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3059 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3060
bd19c9f4
JS
30612003-05-07 Jeff Sturm <jsturm@one-point.com>
3062
3063 Fixes PR bootstrap/10656
3064 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3065 support for .register pseudo-op.
3066 * src/sparc/v8.S: Use it.
3067 * fficonfig.h.in: Rebuilt.
3068 * configure: Rebuilt.
3069
e9b84181
JJ
30702003-04-18 Jakub Jelinek <jakub@redhat.com>
3071
3072 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3073 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3074 Make it the default on POWERPC64.
3075 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3076 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3077 * configure: Rebuilt.
3078 * src/powerpc/ffi.c (hidden): Define.
3079 (ffi_prep_args_SYSV): Renamed from
3080 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3081 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3082 ASM_NEEDS_REGISTERS64): New.
3083 (ffi_prep_args64): New function.
3084 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3085 (ffi_call): Likewise.
3086 (ffi_prep_closure): Likewise.
3087 (flush_icache): Surround by #ifndef POWERPC64.
3088 (ffi_dblfl): New union type.
3089 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3090 (ffi_closure_helper_LINUX64): New function.
3091 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3092 __powerpc64__.
3093 * src/powerpc/sysv.S: Likewise.
3094 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3095 * src/powerpc/linux64.S: New file.
3096 * src/powerpc/linux64_closure.S: New file.
3097 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3098 src/powerpc/linux64_closure.S.
3099 (TARGET_SRC_POWERPC): Likewise.
3100
3101 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3102 closure_test_fn3): Fix result printing on big-endian 64-bit
3103 machines.
3104 (main): Print tst2_arg instead of uninitialized tst2_result.
3105
3106 * src/ffitest.c (main): Hide what closure pointer really points to
3107 from the compiler.
3108
908a31e1
RE
31092003-04-16 Richard Earnshaw <rearnsha@arm.com>
3110
3111 * configure.in (arm-*-netbsdelf*): Add configuration.
3112 (configure): Regenerated.
3113
41a73c4c
LR
31142003-04-04 Loren J. Rittle <ljrittle@acm.org>
3115
3116 * include/Makefile.in: Regenerate.
3117
cd41c847
ZD
31182003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3119
3120 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3121 bit mode.
3122 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3123 Receive closure pointer through parameter, read args using
3124 __builtin_dwarf_cfa.
3125 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3126
6db450f9
AS
31272003-03-12 Andreas Schwab <schwab@suse.de>
3128
3129 * configure.in: Avoid trailing /. in toolexeclibdir.
3130 * configure: Rebuilt.
3131
70144e9c
AT
31322003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3133
c6900272 3134 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
70144e9c 3135
79e43f89
AT
31362003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3137
3138 * libffi/src/powerpc/darwin_closure.S:
3139 Fix alignement bug, allocate 8 bytes for the result.
c930d8a5 3140 * libffi/src/powerpc/aix_closure.S:
79e43f89
AT
3141 Likewise.
3142 * libffi/src/powerpc/ffi_darwin.c:
3143 Update stackframe description for aix/darwin_closure.S.
c930d8a5 3144
0fcc9e55
JJ
31452003-02-06 Jakub Jelinek <jakub@redhat.com>
3146
3147 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3148 attribute.
3149
63b87c2e
CC
31502003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3151 Andreas Schwab <schwab@suse.de>
3152
c6900272 3153 * configure.in: Adjust command to source config-ml.in to account
63b87c2e
CC
3154 for changes to the libffi_basedir definition.
3155 (libffi_basedir): Remove ${srcdir} from value and include trailing
3156 slash if nonempty.
c930d8a5 3157
c6900272 3158 * configure: Regenerate.
63b87c2e 3159
c33402e2
FS
31602003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3161
3162 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3163
fd2e49b2
AH
31642003-01-28 Andrew Haley <aph@redhat.com>
3165
3166 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3167 * src/x86/ffi64.c (ffi_prep_closure): New.
3168 (ffi_closure_UNIX64_inner): New.
3169 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3170
0e292c01
AO
31712003-01-27 Alexandre Oliva <aoliva@redhat.com>
3172
3173 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3174 Remove USE_LIBDIR conditional.
3175 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3176 * Makefile.in, configure: Rebuilt.
3177
acabc220
DE
31782003-01027 David Edelsohn <edelsohn@gnu.org>
3179
3180 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3181 * Makefile.in: Regenerate.
3182
be0acd42
AH
31832003-01-22 Andrew Haley <aph@redhat.com>
3184
3185 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3186 unwind info.
3187
ed695c8e
AT
31882003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3189
3190 * src/powerpc/darwin.S: Add unwind info.
3191 * src/powerpc/darwin_closure.S: Likewise.
3192
45ba15f4
AH
31932003-01-14 Andrew Haley <aph@redhat.com>
3194
3195 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3196 (ffi_prep_cif_machdep): Likewise.
3197 * src/x86/unix64.S: Add unwind info.
3198
4b5f402e
AJ
31992003-01-14 Andreas Jaeger <aj@suse.de>
3200
3201 * src/ffitest.c (main): Only use ffi_closures if those are
3202 supported.
3203
7b5102af
TT
32042003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3205
3206 * libffi/src/ffitest.c
c6900272 3207 add closure testcases
7b5102af
TT
3208
32092003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3210
3211 * libffi/src/powerpc/ffi.c
c6900272 3212 fix alignment bug for float (4 byte aligned iso 8 byte)
7b5102af 3213
5f7c78d8
GK
32142003-01-09 Geoffrey Keating <geoffk@apple.com>
3215
3216 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3217 * src/powerpc/darwin.S: Remove RCS version string.
3218
c75c7793
JS
32192003-01-03 Jeff Sturm <jsturm@one-point.com>
3220
3221 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3222 * src/ffitest.c (main): Use static storage for closure.
3223 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3224 * src/sparc/v8.S (ffi_closure_v8): New.
3225 * src/sparc/v9.S (ffi_closure_v9): New.
3226
eb3c46a1
RM
32272002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3228
3229 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3230 enumeration for X86_WIN32.
3231 * src/x86/win32.S: Added ffi_call_STDCALL function
3232 definition.
3233 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3234 switch cases for recognising FFI_STDCALL and
3235 calling ffi_call_STDCALL if target is X86_WIN32.
3236 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3237 stdcall versions of the "my_strlen" and "many"
3238 test functions (for X86_WIN32).
3239 Added test cases to test stdcall invocation using
3240 these functions.
3241
4977bab6
ZW
32422002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3243
3244 * src/sh/sysv.S: Add DWARF2 unwind info.
3245
67333e84
UW
32462002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3247
3248 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3249
bc3cf760
JW
32502002-11-26 Jim Wilson <wilson@redhat.com>
3251
3252 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3253
4977bab6
ZW
32542002-11-23 H.J. Lu <hjl@gnu.org>
3255
3256 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3257 Include ../config/accross.m4.
3258 * aclocal.m4; Rebuild.
3259 * configure: Likewise.
3260
18789f4e
UW
32612002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3262
3263 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3264
6bee5372
DD
32652002-11-11 DJ Delorie <dj@redhat.com>
3266
3267 * configure.in: Look for common files in the right place.
3268
8177895b
UW
32692002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3270
3271 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3272 raw data as _Jv_word values, not ffi_raw.
3273 (ffi_java_ptrarray_to_raw): Likewise.
3274 (ffi_java_rvalue_to_raw): New function.
3275 (ffi_java_raw_call): Call it.
3276 (ffi_java_raw_to_rvalue): New function.
3277 (ffi_java_translate_args): Call it.
3278 * src/ffitest.c (closure_test_fn): Interpret return value
3279 as ffi_arg, not int.
3280 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3281 FFI_TYPE_POINTER case.
3282 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3283 values extended to word size.
3284
af50f4c8
AJ
32852002-10-02 Andreas Jaeger <aj@suse.de>
3286
3287 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3288
5d297584
BT
32892002-10-01 Bo Thorsen <bo@smetana.suse.de>
3290
3291 * include/ffi.h.in: Fix i386 win32 compilation.
3292
c4f17c6f
UW
32932002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3294
3295 * configure.in: Add s390x-*-linux-* target.
3296 * configure: Regenerate.
3297 * include/ffi.h.in: Define S390X for s390x targets.
3298 (FFI_CLOSURES): Define for s390/s390x.
3299 (FFI_TRAMPOLINE_SIZE): Likewise.
3300 (FFI_NATIVE_RAW_API): Likewise.
3301 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3302 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
4977bab6 3303 * src/s390/ffi.c: Major rework of existing code. Add support for
c4f17c6f
UW
3304 s390x targets. Add closure support.
3305 * src/s390/sysv.S: Likewise.
3306
d7b52625
RE
33072002-09-29 Richard Earnshaw <rearnsha@arm.com>
3308
3309 * src/arm/sysv.S: Fix typo.
3310
28165e22
RE
33112002-09-28 Richard Earnshaw <rearnsha@arm.com>
3312
3313 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3314 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3315 (ffi_call_SYSV): Handle soft-float.
3316
0b0f4163
BT
33172002-09-27 Bo Thorsen <bo@suse.de>
3318
3319 * include/ffi.h.in: Fix multilib x86-64 support.
3320
6c5d742e
KG
33212002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3322
3323 * Makefile.am (all-multi): Fix multilib parallel build.
3324
f4b2b0d6
KK
33252002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3326
3327 * configure.in (sh[34]*-*-linux*): Add brackets.
3328 * configure: Regenerate.
3329
ddebdfeb
KK
33302002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3331
3332 * Makefile.am: Add SH support.
3333 * Makefile.in: Regenerate.
3334 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3335 * configure: Regenerate.
3336 * include/ffi.h.in: Add SH support.
3337 * src/sh/ffi.c: New file.
3338 * src/sh/sysv.S: New file.
3339 * src/types.c: Add SH support.
3340
c94974ca
BT
33412002-07-16 Bo Thorsen <bo@suse.de>
3342
3343 * src/x86/ffi64.c: New file that adds x86-64 support.
3344 * src/x86/unix64.S: New file that handles argument setup for
3345 x86-64.
3346 * src/x86/sysv.S: Don't use this on x86-64.
3347 * src/x86/ffi.c: Don't use this on x86-64.
3348 Remove unused vars.
3349 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3350 for x86-64.
3351 * src/ffitest.c (struct6): New test that tests a special case in
3352 the x86-64 ABI.
3353 (struct7): Likewise.
3354 (struct8): Likewise.
3355 (struct9): Likewise.
3356 (closure_test_fn): Silence warning about this when it's not used.
3357 (main): Add the new tests.
3358 (main): Fix a couple of wrong casts and silence some compiler warnings.
3359 * include/ffi.h.in: Add x86-64 ABI definition.
3360 * fficonfig.h.in: Regenerate.
3361 * Makefile.am: Add x86-64 support.
3362 * configure.in: Likewise.
3363 * Makefile.in: Regenerate.
3364 * configure: Likewise.
3365
33662002-06-24 Bo Thorsen <bo@suse.de>
3367
3368 * src/types.c: Merge settings for similar architectures.
3369 Add x86-64 sizes and alignments.
3370
33712002-06-23 Bo Thorsen <bo@suse.de>
3372
3373 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3374 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3375 * src/mips/ffi.c (ffi_prep_args): Likewise.
3376 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3377
5be36a8f
L
33782002-07-18 H.J. Lu (hjl@gnu.org)
3379
3380 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3381 (libffi_la_SOURCES): Support MIPS_LINUX.
3382 (libffi_convenience_la_SOURCES): Likewise.
3383 * Makefile.in: Regenerated.
3384
3385 * configure.in (mips64*-*): Skip.
3386 (mips*-*-linux*): New.
3387 * configure: Regenerated.
3388
3389 * src/mips/ffi.c: Include <sgidefs.h>.
3390
ad5e163c
UW
33912002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3392
3393 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3394
ee4586c5
RS
33952002-05-27 Roger Sayle <roger@eyesopen.com>
3396
3397 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3398
b86aa7b0
BT
33992002-05-27 Bo Thorsen <bo@suse.de>
3400
3401 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3402 fix formatting.
3403
859f7aaf
AT
34042002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3405
3406 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3407 beginning of function (for older apple cc).
3408
6706f116
AO
34092002-05-08 Alexandre Oliva <aoliva@redhat.com>
3410
3411 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3412 script entry, and set LD to it when configuring multilibs.
3413 * configure: Rebuilt.
3414
fad655c8
JT
34152002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3416
3417 * configure.in (sparc64-*-netbsd*): Add target.
3418 (sparc-*-netbsdelf*): Likewise.
3419 * configure: Regenerate.
3420
189ab118
DM
34212002-04-28 David S. Miller <davem@redhat.com>
3422
3423 * configure.in, configure: Fix SPARC test in previous change.
3424
22bcf65c
GT
34252002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3426
3427 * Makefile.am: Add Linux for S/390 support.
3428 * Makefile.in: Regenerate.
3429 * configure.in: Add Linux for S/390 support.
3430 * configure: Regenerate.
3431 * include/ffi.h.in: Add Linux for S/390 support.
3432 * src/s390/ffi.c: New file from libffi CVS tree.
3433 * src/s390/sysv.S: New file from libffi CVS tree.
3434
5d84cf0b
JJ
34352002-04-28 Jakub Jelinek <jakub@redhat.com>
3436
3437 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3438 %r_disp32().
3439 * src/sparc/v8.S: Use it.
3440 * src/sparc/v9.S: Likewise.
3441 * fficonfig.h.in: Rebuilt.
3442 * configure: Rebuilt.
3443
cc382e5c
HB
34442002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3445
3446 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3447 correctly.
3448 * src/ia64/unix.S: Add unwind information. Fix comments.
3449 Save sp in a way that's compatible with unwind info.
3450 (ffi_call_unix): Correctly restore sp in all cases.
3451 * src/ia64/ffi.c: Add, fix comments.
3452
dc798645
JJ
34532002-04-08 Jakub Jelinek <jakub@redhat.com>
3454
3455 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3456
da00eaab
JT
34572002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3458
3459 * configure.in (alpha*-*-netbsd*): Add target.
3460 * configure: Regenerate.
3461
0c0b87ad
JS
34622002-04-04 Jeff Sturm <jsturm@one-point.com>
3463
3464 * src/sparc/v8.S: Add unwind info.
3465 * src/sparc/v9.S: Likewise.
3466
320d7e7a
KW
34672002-03-30 Krister Walfridsson <cato@df.lth.se>
3468
3469 * configure.in: Enable i*86-*-netbsdelf*.
3470 * configure: Rebuilt.
4977bab6 3471
c75e51ed
DB
34722002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3473
3474 PR other/2620
3475 * src/mips/n32.s: Delete
3476 * src/mips/o32.s: Delete
3477
5bb366e0
LR
34782002-03-21 Loren J. Rittle <ljrittle@acm.org>
3479
3480 * configure.in: Enable alpha*-*-freebsd*.
3481 * configure: Rebuilt.
3482
ce3b24a7
BM
34832002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3484
3485 * Makefile.am: libfficonvenience -> libffi_convenience.
3486 * Makefile.in: Rebuilt.
4977bab6 3487
1eb7ba30
BM
3488 * Makefile.am: Define ffitest_OBJECTS.
3489 * Makefile.in: Rebuilt.
ce3b24a7 3490
073ae293
AT
34912002-03-07 Andreas Tobler <toa@pop.agri.ch>
3492 David Edelsohn <edelsohn@gnu.org>
3493
3494 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3495 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3496 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3497 * Makefile.in: Regenerate.
3498 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3499 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3500 (flush_icache, flush_range): New functions.
3501 (ffi_closure_helper_DARWIN): New function.
3502 * src/powerpc/aix_closure.S: New file.
3503 * src/powerpc/darwin_closure.S: New file.
4977bab6 3504
02c521f4
JS
35052002-02-24 Jeff Sturm <jsturm@one-point.com>
3506
3507 * include/ffi.h.in: Add typedef for ffi_arg.
3508 * src/ffitest.c (main): Declare rint with ffi_arg.
4977bab6 3509
287dd527
AT
35102002-02-21 Andreas Tobler <toa@pop.agri.ch>
3511
3512 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3513 number of GPRs for floating-point arguments.
3514
e2f2e70a
AG
35152002-01-31 Anthony Green <green@redhat.com>
3516
3517 * configure: Rebuilt.
3518 * configure.in: Replace CHECK_SIZEOF and endian tests with
3519 cross-compiler friendly macros.
3520 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3521 macros.
3522
15b9b32f
DE
35232002-01-18 David Edelsohn <edelsohn@gnu.org>
3524
3525 * src/powerpc/darwin.S (_ffi_call_AIX): New.
1f7d5413 3526 * src/powerpc/aix.S (ffi_call_DARWIN): New.
15b9b32f 3527
2f4006f8
DE
35282002-01-17 David Edelsohn <edelsohn@gnu.org>
3529
3530 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3531 (TARGET_SRC_POWERPC_AIX): New.
3532 (POWERPC_AIX): New stanza.
3533 * Makefile.in: Regenerate.
3534 * configure.in: Add AIX case.
3535 * configure: Regenerate.
3536 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3537 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3538 size. Fix "long double" support.
3539 (ffi_call): Add FFI_AIX case.
3540 * src/powerpc/aix.S: New.
3541
3c017635
JH
35422001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3543
3544 Implement Darwin PowerPC ABI.
3545 * configure.in: Handle powerpc-*-darwin*.
3546 * Makefile.am: Set source files for POWERPC_DARWIN.
3547 * configure: Rebuilt.
3548 * Makefile.in: Rebuilt.
3549 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3550 POWERPC_DARWIN.
3551 * src/powerpc/darwin.S: New file.
3552 * src/powerpc/ffi_darwin.c: New file.
3553
cc712abf
JM
35542001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3555
3556 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3557
03a8de49
RO
35582001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3559
3560 * src/x86/sysv.S: Avoid gas-only .balign directive.
3561 Use C style comments.
3562
92a0e6c6
RO
35632001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3564
3565 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3566 Fixes PR bootstrap/3563.
3567
57482f58
RO
35682001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3569
3570 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3571
d7e2dd2f
RO
35722001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3573
3574 * configure.in: Recognize sparc*-sun-* host.
3575 * configure: Regenerate.
4977bab6 3576
be72207c
AH
35772001-06-06 Andrew Haley <aph@redhat.com>
3578
4075db8f 3579 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
be72207c 3580
552cb45a
AH
35812001-06-03 Andrew Haley <aph@redhat.com>
3582
4075db8f
BT
3583 * src/alpha/osf.S: Add unwind info.
3584 * src/powerpc/sysv.S: Add unwind info.
3585 * src/powerpc/ppc_closure.S: Likewise.
552cb45a 3586
7d45b968
JS
35872000-05-31 Jeff Sturm <jsturm@one-point.com>
3588
3589 * configure.in: Fix AC_ARG_ENABLE usage.
3590 * configure: Rebuilt.
3591
2a7f63c4
BM
35922001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3593
3594 * configure.in: Remove warning about beta code.
3595 * configure: Rebuilt.
3596
35972001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3598
03a55252
HB
3599 * src/ia64/unix.S: Restore stack pointer when returning from
3600 ffi_closure_UNIX.
3601 * src/ia64/ffi.c: Fix typo in comment.
3602
df5da1c4
JW
36032001-04-18 Jim Wilson <wilson@redhat.com>
3604
3605 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3606 to eliminate RAW DV.
3607
52c28977
BM
36082001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3609
3610 * Makefile.am: Make a libtool convenience library.
3611 * Makefile.in: Rebuilt.
3612
3b905056
BM
36132001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3614
3615 * configure.in: Use different syntax for subdirectory creation.
3616 * configure: Rebuilt.
3617
d6c4152a
JB
36182001-03-27 Jon Beniston <jon@beniston.com>
3619
3620 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3621 * configure: Rebuilt.
3622 * Makefile.am: Added X86_WIN32 target support.
3623 * Makefile.in: Rebuilt.
3624
3625 * include/ffi.h.in: Added X86_WIN32 target support.
3626
3627 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3628 * src/types.c: Added X86_WIN32 target support.
3629
3630 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3631 stuff removed and made to work with CygWin's gas.
3632
675cdab3
BM
36332001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3634
3635 * configure.in: Make target subdirectory in build dir.
3636 * Makefile.am: Override suffix based rules to specify correct output
3637 subdirectory.
3638 * Makefile.in: Rebuilt.
3639 * configure: Rebuilt.
3640
e5f8858a 36412001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
cc4c8975
KH
3642
3643 * src/powerpc/ppc_closure.S: New file.
3644 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3645 involving long long and register pairs.
3646 (ffi_prep_closure): New function.
3647 (flush_icache): Likewise.
3648 (ffi_closure_helper_SYSV): Likewise.
3649 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3650 (FFI_TRAMPOLINE_SIZE): Likewise.
3651 (FFI_NATIVE_RAW_API): Likewise.
3652 * Makefile.in: Rebuilt.
3653 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3654 (TARGET_SRC_POWERPC): Likewise.
3655
9450a929
TT
36562001-03-19 Tom Tromey <tromey@redhat.com>
3657
3658 * Makefile.in: Rebuilt.
3659 * Makefile.am (ffitest_LDFLAGS): New macro.
3660
4d2cf715
NC
36612001-03-02 Nick Clifton <nickc@redhat.com>
3662
3663 * include/ffi.h.in: Remove RCS ident string.
3664 * include/ffi_mips.h: Remove RCS ident string.
3665 * src/debug.c: Remove RCS ident string.
3666 * src/ffitest.c: Remove RCS ident string.
3667 * src/prep_cif.c: Remove RCS ident string.
3668 * src/types.c: Remove RCS ident string.
3669 * src/alpha/ffi.c: Remove RCS ident string.
3670 * src/alpha/osf.S: Remove RCS ident string.
3671 * src/arm/ffi.c: Remove RCS ident string.
3672 * src/arm/sysv.S: Remove RCS ident string.
3673 * src/mips/ffi.c: Remove RCS ident string.
3674 * src/mips/n32.S: Remove RCS ident string.
3675 * src/mips/o32.S: Remove RCS ident string.
3676 * src/sparc/ffi.c: Remove RCS ident string.
3677 * src/sparc/v8.S: Remove RCS ident string.
3678 * src/sparc/v9.S: Remove RCS ident string.
3679 * src/x86/ffi.c: Remove RCS ident string.
3680 * src/x86/sysv.S: Remove RCS ident string.
3681
7e5fd99f
JM
36822001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3683
3684 * include/ffi.h.in: Change sourceware.cygnus.com references to
3685 gcc.gnu.org.
3686
3f67ba6e
RH
36872000-12-09 Richard Henderson <rth@redhat.com>
3688
3689 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3690 (ffi_closure_osf_inner): Index rather than increment avalue
3691 and arg_types. Give ffi_closure_osf the raw return value type.
3692 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3693 promotion.
3694
36952000-12-07 Richard Henderson <rth@redhat.com>
29fe0479
RH
3696
3697 * src/raw_api.c (ffi_translate_args): Fix typo.
3698 (ffi_prep_closure): Likewise.
3699
3700 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3701 FFI_TRAMPOLINE_SIZE.
3702 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3703 cif->bytes for new ffi_call_osf implementation.
3704 (ffi_prep_args): Absorb into ...
3705 (ffi_call): ... here. Do all stack allocation here and
3706 avoid a callback function.
3707 (ffi_prep_closure, ffi_closure_osf_inner): New.
3708 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3709 (ffi_closure_osf): New.
3710
2f7bf5b7
AO
37112000-09-10 Alexandre Oliva <aoliva@redhat.com>
3712
ed1c1a1a
AO
3713 * config.guess, config.sub, install-sh: Removed.
3714 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3715 * Makefile.in: Rebuilt.
3716
2f7bf5b7
AO
3717 * acinclude.m4: Include libtool macros from the top level.
3718 * aclocal.m4, configure: Rebuilt.
3719
e15a68e7
AO
37202000-08-22 Alexandre Oliva <aoliva@redhat.com>
3721
3722 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3723 * configure: Rebuilt.
3724
7f8c11d5
SB
37252000-05-11 Scott Bambrough <scottb@netwinder.org>
3726
4977bab6
ZW
3727 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3728 memory correctly. Use conditional instructions, not branches where
7f8c11d5
SB
3729 possible.
3730
c5acf51f
TT
37312000-05-04 Tom Tromey <tromey@cygnus.com>
3732
3733 * configure: Rebuilt.
3734 * configure.in: Match `arm*-*-linux-*'.
3735 From Chris Dornan <cdornan@arm.com>.
3736
91bffbe9
JJ
37372000-04-28 Jakub Jelinek <jakub@redhat.com>
3738
3739 * Makefile.am (SUBDIRS): Define.
3740 (AM_MAKEFLAGS): Likewise.
3741 (Multilib support.): Add section.
3742 * Makefile.in: Rebuilt.
3743 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
4075db8f
BT
3744 New variables. Set for gcc using -print-multi-lib. Export them
3745 to libtool.
3746 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3747 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3748 for -shared links.
3749 (extra_compiler_flags_value, extra_compiler_flags): Check these
3750 for extra compiler options which need to be passed down in
3751 compiler_flags.
91bffbe9 3752
bd66587e
AG
37532000-04-16 Anthony Green <green@redhat.com>
3754
3755 * configure: Rebuilt.
3756 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3757
3791773c
JJ
37582000-04-14 Jakub Jelinek <jakub@redhat.com>
3759
4075db8f 3760 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3791773c
JJ
3761 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3762 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3763 Replace all void * sizeofs with sizeof(int).
3764 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3765 different than DOUBLE.
3766 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3767 (ffi_prep_args_v9): New function.
3768 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3769 (ffi_V9_return_struct): New function.
3770 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3771 32bit code (not yet cross-arch calls).
3772 * src/sparc/v8.S: Add struct return delay nop.
3773 Handle long long.
3774 * src/sparc/v9.S: New file.
3775 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3776 is used on sparc64 only for structures larger than 32 bytes.
3777 Pass by reference for structures is done for structure arguments
3778 larger than 16 bytes.
3779 * src/ffitest.c (main): Use 64bit rint on sparc64.
3780 Run long long tests on sparc.
3781 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3782 sparc64.
3783 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3784 on sparc64.
3785 * configure.in (sparc-*-linux*): New supported target.
3786 (sparc64-*-linux*): Likewise.
3787 * configure: Rebuilt.
3788 * Makefile.am: Add v9.S to SPARC files.
3789 * Makefile.in: Likewise.
3790 (LINK): Surround $(CCLD) into double quotes, so that multilib
3791 compiles work correctly.
3792
db459595
APB
37932000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3794
3795 * configure: Rebuilt.
3796 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3797 proposed by Bryce McKinlay.
3798
c1f63902
TT
37992000-03-20 Tom Tromey <tromey@cygnus.com>
3800
3801 * Makefile.in: Hand edit for java_raw_api.lo.
3802
0f72dc9e
BM
38032000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3804
3805 * config.guess, config.sub: Update from the gcc tree.
3806 Fix for PR libgcj/168.
3807
3c18f2d1
TT
38082000-03-03 Tom Tromey <tromey@cygnus.com>
3809
323eee8a
TT
3810 * Makefile.in: Fixed ia64 by hand.
3811
3c18f2d1
TT
3812 * configure: Rebuilt.
3813 * configure.in (--enable-multilib): New option.
3814 (libffi_basedir): New subst.
3815 (AC_OUTPUT): Added multilib code.
3816
a2b10a92
TT
38172000-03-02 Tom Tromey <tromey@cygnus.com>
3818
3819 * Makefile.in: Rebuilt.
3820 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3821 directory name.
3822
dc5de370
HB
38232000-02-25 Hans Boehm <boehm@acm.org>
3824
3825 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3826 files.
3827 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3828 list.
3829 (ffi_prep_raw_closure): Use ffi_translate_args, not
3830 ffi_closure_translate.
3831 * src/java_raw_api.c: New file.
3832 * src/ffitest.c (closure_test_fn): New function.
3833 (main): Define `rint' as long long on IA64. Added new test when
3834 FFI_CLOSURES is defined.
3835 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3836 (ffi_abi): Recognize IA64.
3837 (ffi_raw): Added `flt' field.
3838 Added "Java raw API" code.
3839 * configure.in: Recognize ia64.
3840 * Makefile.am (TARGET_SRC_IA64): New macro.
3841 (libffi_la_common_SOURCES): Added java_raw_api.c.
3842 (libffi_la_SOURCES): Define in IA64 case.
3843
d79229bd
TT
38442000-01-04 Tom Tromey <tromey@cygnus.com>
3845
3846 * Makefile.in: Rebuilt with newer automake.
3847
2c33b220
TT
38481999-12-31 Tom Tromey <tromey@cygnus.com>
3849
3850 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3851
6ba9e3b0
TT
38521999-09-01 Tom Tromey <tromey@cygnus.com>
3853
3854 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3855 undefs.
3856 * fficonfig.h.in: Rebuilt.
3857 * configure: Rebuilt.
3858 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3859 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3860 * acconfig.h: Don't #undef PACKAGE or VERSION.
3861
63e5e3e0
AG
38621999-08-09 Anthony Green <green@cygnus.com>
3863
3864 * include/ffi.h.in: Try to work around messy header problem
3865 with PACKAGE and VERSION.
3866
3867 * configure: Rebuilt.
3868 * configure.in: Change version to 2.00-beta.
3869
4075db8f 3870 * fficonfig.h.in: Rebuilt.
63e5e3e0
AG
3871 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3872
3873 * src/x86/ffi.c (ffi_raw_call): Rename.
3874
38751999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3876
3877 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3878 (ffi_prep_incoming_args_SYSV): Ditto.
3879 (ffi_prep_closure): Ditto.
3880 (ffi_closure_raw_SYSV): Ditto.
3881 (ffi_prep_raw_closure): More ditto.
3882 (ffi_call_raw): Final ditto.
3883
3884 * include/ffi.h.in: Add definitions for closure and raw API.
3885
3886 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4977bab6 3887 FFI_TYPE_UINT64.
63e5e3e0
AG
3888
3889 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3890
3891 * src/raw_api.c: New file.
3892
3893 * include/ffi.h.in (ffi_raw): New type.
3894 (UINT_ARG, SINT_ARG): New defines.
3895 (ffi_closure, ffi_raw_closure): New types.
3896 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4977bab6 3897
63e5e3e0
AG
3898 * configure.in: Add check for endianness and sizeof void*.
3899
3900 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3901 instead of directly.
3902
3903 * configure: Rebuilt.
3904
3905Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3906
3907 * configure.in: Add x86 and powerpc BeOS configurations.
3908 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3909
39101999-05-09 Anthony Green <green@cygnus.com>
3911
3912 * configure.in: Add warning about this being beta code.
3913 Remove src/Makefile.am from the picture.
3914 * configure: Rebuilt.
3915
3916 * Makefile.am: Move logic from src/Makefile.am. Add changes
3917 to support libffi as a target library.
3918 * Makefile.in: Rebuilt.
3919
3920 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3921 Upgraded to new autoconf, automake, libtool.
3922
3923 * README: Tweaks.
3924
3925 * LICENSE: Update copyright date.
3926
3927 * src/Makefile.am, src/Makefile.in: Removed.
3928
39291998-11-29 Anthony Green <green@cygnus.com>
3930
3931 * include/ChangeLog: Removed.
3932 * src/ChangeLog: Removed.
3933 * src/mips/ChangeLog: Removed.
3934 * src/sparc/ChangeLog: Remboved.
3935 * src/x86/ChangeLog: Removed.
3936
3937 * ChangeLog.v1: Created.