]> git.ipfire.org Git - thirdparty/gcc.git/blame - libffi/ChangeLog
projects / thirdparty / gcc.git / blame
? search:
summary | shortlog | log | commit | commitdiff | tree
blob | blame (incremental) | history | HEAD
IRIX support
[thirdparty/gcc.git] / libffi / ChangeLog
CommitLineData
da4f8cce 12009-12-25 Frank Everdij <f.p.x.everdij@tudelft.nl>
2
3 * include/ffi.h.in: Placed '__GNUC__' ifdef around
4 '__attribute__((aligned(8)))' in ffi_closure, fixes compile for
5 IRIX MIPSPro c99.
6 * include/ffi_common.h: Added '__sgi' define to non
7 '__attribute__((__mode__()))' integer typedefs.
8 * src/mips/ffi.c (ffi_call, ffi_closure_mips_inner_O32,
9 ffi_closure_mips_inner_N32): Added 'defined(_MIPSEB)' to BE check.
10 (ffi_closure_mips_inner_O32, ffi_closure_mips_inner_N32): Added
11 FFI_LONGDOUBLE support and alignment(N32 only).
12 * src/mips/ffitarget.h: Corrected '#include <sgidefs.h>' for IRIX and
13 fixed non '__attribute__((__mode__()))' integer typedefs.
14 * src/mips/n32.S: Put '#ifdef linux' around '.abicalls' and '.eh_frame'
15 since they are Linux/GNU Assembler specific.
16
2a06dfe6 172009-12-25 Bradley Smith <brad@brad-smith.co.uk>
18
19 * configure.ac, Makefile.am, src/avr32/ffi.c,
20 src/avr32/ffitarget.h,
21 src/avr32/sysv.S: Add AVR32 port.
22 * configure, Makefile.in: Rebuilt.
23
cc6aa710 242009-12-21 Andreas Tobler <a.tobler@schweiz.org>
25
26 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
27 * configure: Regenerate.
28
2fb9cd2f 292009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
30
31 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
32
c64cba58 332009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
34
35 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
36 type on HP-UX.
37
c221d0e8 382009-12-11 Eric Botcazou <ebotcazou@adacore.com>
39
40 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
41 double' arguments.
42
db0ed941 432009-12-11 Eric Botcazou <ebotcazou@adacore.com>
44
45 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
46
4f2e543b 472009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
48
49 PR libffi/40700
50 * src/closures.c [X86_64 && __sun__ && __svr4__]
51 (FFI_MMAP_EXEC_WRIT): Define.
52
2e5e20e1 532009-12-08 David Daney <ddaney@caviumnetworks.com>
54
55 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
56 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
57 * testsuite/libffi.call/stret_large.c: Same.
58 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
59 * testsuite/libffi.call/stret_large2.c: Same.
60 * testsuite/libffi.call/stret_medium2.c: Same.
61
514d705d 622009-12-07 David Edelsohn <edelsohn@gnu.org>
63
64 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
65 typo.
66
82d1d66a 672009-12-05 David Edelsohn <edelsohn@gnu.org>
68
69 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
70 code.
514d705d 71 * src/powerpc/aix_closure.S: Same.
2e5e20e1 72
e6408286 732009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
74
75 * Makefile.in: Regenerate.
76 * configure: Regenerate.
77 * include/Makefile.in: Regenerate.
78 * man/Makefile.in: Regenerate.
79 * testsuite/Makefile.in: Regenerate.
80
029645af 812009-12-04 David Edelsohn <edelsohn@gnu.org>
82
83 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
84 linux64_closure.S.
2e5e20e1 85
f89c6c9d 862009-12-04 Uros Bizjak <ubizjak@gmail.com>
87
88 PR libffi/41908
89 * src/x86/ffi64.c (classify_argument): Update from
90 gcc/config/i386/i386.c.
91 (ffi_closure_unix64_inner): Do not use the address of two consecutive
92 SSE registers directly.
93 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
94 for x86_64 linux targets.
95
d66a47ee 962009-12-04 David Edelsohn <edelsohn@gnu.org>
97
98 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
99 pfr for long double split between fpr13 and stack.
100
3a95c696 1012009-12-03 David Edelsohn <edelsohn@gnu.org>
102
103 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
104 fparg_count twice for long double.
105
e3791522 1062009-12-03 David Edelsohn <edelsohn@gnu.org>
107
108 PR libffi/42243
109 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
110
0481262e 1112009-12-03 Uros Bizjak <ubizjak@gmail.com>
112
113 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
c485f998 114 Remove xfails for x86 linux targets.
0481262e 115
e5fdfcfa 1162009-12-02 David Edelsohn <edelsohn@gnu.org>
117
118 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
119 case.
120
1212009-12-01 David Edelsohn <edelsohn@gnu.org>
40e397ce 122
123 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
124 register usage. Call ffi_prep_args directly. Add long double
125 return value support.
126 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
127 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
128 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
129 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
130 mode.
131 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
132 into case.
133 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
134 Allocate result area between params and FPRs.
135
6e94d4ca 1362009-11-30 David Edelsohn <edelsohn@gnu.org>
137
138 PR target/35484
139 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
140 AIX64.
141 * src/powerpc/aix.S: Implement AIX64 version.
142 * src/powerpc/aix_closure.S: Implement AIX64 version.
143 (ffi_closure_ASM): Use extsb, lha and displament addresses.
144 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
145 support.
146 (ffi_prep_cif_machdep): Same.
147 (ffi_call): Same.
148 (ffi_closure_helper_DARWIN): Same.
149
2f9618b6 1502009-11-02 Andreas Tobler <a.tobler@schweiz.org>
151
152 PR libffi/41908
153 * testsuite/libffi.call/testclosure.c: New test.
154
80b483b5 1552009-09-28 Kai Tietz <kai.tietz@onevision.com>
156
157 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
158 assembly version use of ___chkstk.
159
520462c5 1602009-09-23 Matthias Klose <doko@ubuntu.com>
161
162 PR libffi/40242, PR libffi/41443
163 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
164 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
165 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
166 Change the conditionals to __SOFTFP__ || __ARM_EABI__
167 for -mfloat-abi=softfp to work.
168
513e3043 1692009-09-17 Loren J. Rittle <ljrittle@acm.org>
170
171 PR testsuite/32843 (strikes again)
172 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
173 enable proper extension on char and short.
174
841f7ca5 1752009-09-15 David Daney <ddaney@caviumnetworks.com>
176
177 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
178 handling for FFI_TYPE_POINTER.
179 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
180 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
181 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
182 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
183 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
184 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
185 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
186 structure and pointer returns.
187 (ffi_closure_N32): Add handling for pointer returns.
188 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
189 calc_n32_return_struct_flags): Handle soft-float.
190 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
191 (ffi_call_N32): Declare proper argument types.
192 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
193 soft-float.
194
81a71e2b 1952009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
196
197 * configure.ac (AC_PREREQ): Bump to 2.64.
198
cd9a469c 1992009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
200
115d57c5 201 * Makefile.am (install-html, install-pdf): Remove.
202 * Makefile.in: Regenerate.
203
cd9a469c 204 * Makefile.in: Regenerate.
205 * aclocal.m4: Regenerate.
206 * configure: Regenerate.
207 * fficonfig.h.in: Regenerate.
208 * include/Makefile.in: Regenerate.
209 * man/Makefile.in: Regenerate.
210 * testsuite/Makefile.in: Regenerate.
211
9abe4a66 2122009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
213
214 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
215
339e4d43 2162009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
217
218 PR libffi/40807
219 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
220 return types for X86_WIN32.
221 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
222 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
223 _ffi_closure_STDCALL): Likewise.
224
225 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
226 (dlmmap, dlmunmap): Also use these functions on Cygwin.
227
f09e0522 2282009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
229
230 PR testsuite/40699
231 PR testsuite/40707
232 PR testsuite/40709
233 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
234 2009-06-30 commits.
235
aa588440 2362009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
237
238 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
239 to "" before adding paths. (This reinstates an assignment that
240 was removed by my 2009-06-30 commit, but changes the initial
241 value from "." to "".)
242
0378a83e 2432009-07-01 H.J. Lu <hongjiu.lu@intel.com>
244
245 PR testsuite/40601
246 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
247 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
248 empty.
249
2f0e2680 2502009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
251
252 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
253 to ld_library_path. Use add_path. Add just find_libgcc_s
254 to ld_library_path, not every libgcc multilib directory.
255
a05d695d 2562009-06-16 Wim Lewis <wiml@hhhh.org>
257
258 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
259 supposed to be callee-saved.
260 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
261 return buffer for odd-size structs.
262
87f6899d 2632009-06-16 Andreas Tobler <a.tobler@schweiz.org>
264
265 PR libffi/40444
266 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
267 allow_stack_execute for Darwin.
268
fffc53bb 2692009-06-16 Andrew Haley <aph@redhat.com>
270
271 * configure.ac (TARGETDIR): Add missing blank lines.
272 * configure: Regenerate.
273
75c53eb0 2742009-06-16 Andrew Haley <aph@redhat.com>
275
276 * testsuite/libffi.call/cls_align_sint64.c,
277 testsuite/libffi.call/cls_align_uint64.c,
278 testsuite/libffi.call/cls_longdouble_va.c,
279 testsuite/libffi.call/cls_ulonglong.c,
280 testsuite/libffi.call/return_ll1.c,
281 testsuite/libffi.call/stret_medium2.c: Fix printf format
282 specifiers.
283 * testsuite/libffi.call/ffitest.h,
284 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
285
bb07cfdf 2862009-06-15 Andrew Haley <aph@redhat.com>
287
288 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
289 * testsuite/libffi.call/err_bad_abi.c: Likewise.
290
00581c89 2912009-06-12 Andrew Haley <aph@redhat.com>
292
293 * Makefile.am: Remove info_TEXINFOS.
294
2952009-06-12 Andrew Haley <aph@redhat.com>
296
297 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
298 testsuite/libffi.call/cls_align_uint64.c,
299 testsuite/libffi.call/cls_ulonglong.c,
300 testsuite/libffi.call/return_ll1.c,
301 testsuite/libffi.call/stret_medium2.c: Fix printf format
302 specifiers.
303 testsuite/libffi.special/unwindtest.cc: include stdint.h.
304
3052009-06-11 Timothy Wall <twall@users.sf.net>
306
307 * Makefile.am,
308 configure.ac,
309 include/ffi.h.in,
310 include/ffi_common.h,
311 src/closures.c,
312 src/dlmalloc.c,
313 src/x86/ffi.c,
314 src/x86/ffitarget.h,
315 src/x86/win64.S (new),
316 README: Added win64 support (mingw or MSVC)
317 * Makefile.in,
318 include/Makefile.in,
319 man/Makefile.in,
320 testsuite/Makefile.in,
321 configure,
322 aclocal.m4: Regenerated
323 * ltcf-c.sh: properly escape cygwin/w32 path
324 * man/ffi_call.3: Clarify size requirements for return value.
325 * src/x86/ffi64.c: Fix filename in comment.
326 * src/x86/win32.S: Remove unused extern.
327
328 * testsuite/libffi.call/closure_fn0.c,
329 testsuite/libffi.call/closure_fn1.c,
330 testsuite/libffi.call/closure_fn2.c,
331 testsuite/libffi.call/closure_fn3.c,
332 testsuite/libffi.call/closure_fn4.c,
333 testsuite/libffi.call/closure_fn5.c,
334 testsuite/libffi.call/closure_fn6.c,
335 testsuite/libffi.call/closure_stdcall.c,
336 testsuite/libffi.call/cls_12byte.c,
337 testsuite/libffi.call/cls_16byte.c,
338 testsuite/libffi.call/cls_18byte.c,
339 testsuite/libffi.call/cls_19byte.c,
340 testsuite/libffi.call/cls_1_1byte.c,
341 testsuite/libffi.call/cls_20byte.c,
342 testsuite/libffi.call/cls_20byte1.c,
343 testsuite/libffi.call/cls_24byte.c,
344 testsuite/libffi.call/cls_2byte.c,
345 testsuite/libffi.call/cls_3_1byte.c,
346 testsuite/libffi.call/cls_3byte1.c,
347 testsuite/libffi.call/cls_3byte2.c,
348 testsuite/libffi.call/cls_4_1byte.c,
349 testsuite/libffi.call/cls_4byte.c,
350 testsuite/libffi.call/cls_5_1_byte.c,
351 testsuite/libffi.call/cls_5byte.c,
352 testsuite/libffi.call/cls_64byte.c,
353 testsuite/libffi.call/cls_6_1_byte.c,
354 testsuite/libffi.call/cls_6byte.c,
355 testsuite/libffi.call/cls_7_1_byte.c,
356 testsuite/libffi.call/cls_7byte.c,
357 testsuite/libffi.call/cls_8byte.c,
358 testsuite/libffi.call/cls_9byte1.c,
359 testsuite/libffi.call/cls_9byte2.c,
360 testsuite/libffi.call/cls_align_double.c,
361 testsuite/libffi.call/cls_align_float.c,
362 testsuite/libffi.call/cls_align_longdouble.c,
363 testsuite/libffi.call/cls_align_longdouble_split.c,
364 testsuite/libffi.call/cls_align_longdouble_split2.c,
365 testsuite/libffi.call/cls_align_pointer.c,
366 testsuite/libffi.call/cls_align_sint16.c,
367 testsuite/libffi.call/cls_align_sint32.c,
368 testsuite/libffi.call/cls_align_sint64.c,
369 testsuite/libffi.call/cls_align_uint16.c,
370 testsuite/libffi.call/cls_align_uint32.c,
371 testsuite/libffi.call/cls_align_uint64.c,
372 testsuite/libffi.call/cls_dbls_struct.c,
373 testsuite/libffi.call/cls_double.c,
374 testsuite/libffi.call/cls_double_va.c,
375 testsuite/libffi.call/cls_float.c,
376 testsuite/libffi.call/cls_longdouble.c,
377 testsuite/libffi.call/cls_longdouble_va.c,
378 testsuite/libffi.call/cls_multi_schar.c,
379 testsuite/libffi.call/cls_multi_sshort.c,
380 testsuite/libffi.call/cls_multi_sshortchar.c,
381 testsuite/libffi.call/cls_multi_uchar.c,
382 testsuite/libffi.call/cls_multi_ushort.c,
383 testsuite/libffi.call/cls_multi_ushortchar.c,
384 testsuite/libffi.call/cls_pointer.c,
385 testsuite/libffi.call/cls_pointer_stack.c,
386 testsuite/libffi.call/cls_schar.c,
387 testsuite/libffi.call/cls_sint.c,
388 testsuite/libffi.call/cls_sshort.c,
389 testsuite/libffi.call/cls_uchar.c,
390 testsuite/libffi.call/cls_uint.c,
391 testsuite/libffi.call/cls_ulonglong.c,
392 testsuite/libffi.call/cls_ushort.c,
393 testsuite/libffi.call/err_bad_abi.c,
394 testsuite/libffi.call/err_bad_typedef.c,
395 testsuite/libffi.call/float2.c,
396 testsuite/libffi.call/huge_struct.c,
397 testsuite/libffi.call/nested_struct.c,
398 testsuite/libffi.call/nested_struct1.c,
399 testsuite/libffi.call/nested_struct10.c,
400 testsuite/libffi.call/nested_struct2.c,
401 testsuite/libffi.call/nested_struct3.c,
402 testsuite/libffi.call/nested_struct4.c,
403 testsuite/libffi.call/nested_struct5.c,
404 testsuite/libffi.call/nested_struct6.c,
405 testsuite/libffi.call/nested_struct7.c,
406 testsuite/libffi.call/nested_struct8.c,
407 testsuite/libffi.call/nested_struct9.c,
408 testsuite/libffi.call/problem1.c,
409 testsuite/libffi.call/return_ldl.c,
410 testsuite/libffi.call/return_ll1.c,
411 testsuite/libffi.call/stret_large.c,
412 testsuite/libffi.call/stret_large2.c,
413 testsuite/libffi.call/stret_medium.c,
414 testsuite/libffi.call/stret_medium2.c,
415 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
416 of checking for MMAP. Use intptr_t instead of long casts.
417
f6e79f69 4182009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
419
420 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
421 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
422 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
423
e5ad3c0e 4242009-06-09 Andrew Haley <aph@redhat.com>
425
426 * src/x86/freebsd.S: Add missing file.
427
1f429ea0 4282009-06-08 Andrew Haley <aph@redhat.com>
429
430 Import from libffi 3.0.8:
431
432 * doc/libffi.texi: New file.
433 * doc/libffi.info: Likewise.
434 * doc/stamp-vti: Likewise.
435 * man/Makefile.am: New file.
436 * man/ffi_call.3: New file.
437
438 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
439 src/dlmalloc.c.
440 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
441
442 * configure.ac: Bump version to 3.0.8.
443 parisc*-*-linux*: Add.
444 i386-*-freebsd* | i386-*-openbsd*: Add.
445 powerpc-*-beos*: Add.
446 AM_CONDITIONAL X86_FREEBSD: Add.
447 AC_CONFIG_FILES: Add man/Makefile.
448
449 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
450
7302546d 4512009-06-08 Andrew Haley <aph@redhat.com>
452
453 * README: Import from libffi 3.0.8.
454
07d9c635 4552009-06-08 Andrew Haley <aph@redhat.com>
456
457 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
458 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
459 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
460 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
461
462 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
463 * testsuite/libffi.call/stret_medium.c: Likewise.
464 * testsuite/libffi.call/stret_large2.c: Likewise.
465 * testsuite/libffi.call/stret_large.c: Likewise.
466
4672008-12-26 Timothy Wall <twall@users.sf.net>
468
469 * testsuite/libffi.call/cls_longdouble.c,
470 testsuite/libffi.call/cls_longdouble_va.c,
471 testsuite/libffi.call/cls_align_longdouble.c,
472 testsuite/libffi.call/cls_align_longdouble_split.c,
473 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
474 failures on x86_64 cygwin/mingw.
475
4762008-12-22 Timothy Wall <twall@users.sf.net>
477
478 * testsuite/libffi.call/closure_fn0.c,
479 testsuite/libffi.call/closure_fn1.c,
480 testsuite/libffi.call/closure_fn2.c,
481 testsuite/libffi.call/closure_fn3.c,
482 testsuite/libffi.call/closure_fn4.c,
483 testsuite/libffi.call/closure_fn5.c,
484 testsuite/libffi.call/closure_fn6.c,
485 testsuite/libffi.call/closure_loc_fn0.c,
486 testsuite/libffi.call/closure_stdcall.c,
487 testsuite/libffi.call/cls_align_pointer.c,
488 testsuite/libffi.call/cls_pointer.c,
489 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
490 pointer to integer (intptr_t).
491 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
492
4932008-07-24 Anthony Green <green@redhat.com>
494
495 * testsuite/libffi.call/cls_dbls_struct.c,
496 testsuite/libffi.call/cls_double_va.c,
497 testsuite/libffi.call/cls_longdouble.c,
498 testsuite/libffi.call/cls_longdouble_va.c,
499 testsuite/libffi.call/cls_pointer.c,
500 testsuite/libffi.call/cls_pointer_stack.c,
501 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
502 compiler warnings.
503
5042008-03-04 Anthony Green <green@redhat.com>
505 Blake Chaffin
506 hos@tamanegi.org
507
508 * testsuite/libffi.call/cls_align_longdouble_split2.c
509 testsuite/libffi.call/cls_align_longdouble_split.c
510 testsuite/libffi.call/cls_dbls_struct.c
511 testsuite/libffi.call/cls_double_va.c
512 testsuite/libffi.call/cls_longdouble.c
513 testsuite/libffi.call/cls_longdouble_va.c
514 testsuite/libffi.call/cls_pointer.c
515 testsuite/libffi.call/cls_pointer_stack.c
516 testsuite/libffi.call/err_bad_abi.c
517 testsuite/libffi.call/err_bad_typedef.c
518 testsuite/libffi.call/stret_large2.c
519 testsuite/libffi.call/stret_large.c
520 testsuite/libffi.call/stret_medium2.c
521 testsuite/libffi.call/stret_medium.c: New tests from Apple.
522
c720ee9b 5232009-06-05 Andrew Haley <aph@redhat.com>
524
525 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
526 libffi.
527
91e95000 5282009-06-04 Andrew Haley <aph@redhat.com>
529
530 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
531 stdcall changes.
532
1f9ea1fc 5332008-02-26 Anthony Green <green@redhat.com>
534 Thomas Heller <theller@ctypes.org>
535
536 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
537 comment.
538
5392008-02-03 Timothy Wall <twall@users.sf.net>
540
541 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
542 offset based on code pointer, not data pointer.
543
5442008-01-31 Timothy Wall <twall@users.sf.net>
545
546 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
547 closures.
548 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
549 closures.
550 * src/x86/win32.S: Add assembly for stdcall closure.
551 * src/x86/ffi.c: Initialize stdcall closure trampoline.
552
e01722f1 5532009-06-04 Andrew Haley <aph@redhat.com>
554
555 * include/ffi.h.in: Change void (*)() to void (*)(void).
556 * src/x86/ffi.c: Likewise.
557
6e59f66f 5582009-06-04 Andrew Haley <aph@redhat.com>
559
560 * src/powerpc/ppc_closure.S: Insert licence header.
561 * src/powerpc/linux64_closure.S: Likewise.
562 * src/m68k/sysv.S: Likewise.
563
564 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
565 * src/powerpc/ffi.c: Likewise.
566 * src/powerpc/ffi_darwin.c: Likewise.
567 * src/m32r/ffi.c: Likewise.
568 * src/sh64/ffi.c: Likewise.
569 * src/x86/ffi64.c: Likewise.
570 * src/alpha/ffi.c: Likewise.
571 * src/alpha/osf.S: Likewise.
572 * src/frv/ffi.c: Likewise.
573 * src/s390/ffi.c: Likewise.
574 * src/pa/ffi.c: Likewise.
575 * src/pa/hpux32.S: Likewise.
576 * src/ia64/unix.S: Likewise.
577 * src/ia64/ffi.c: Likewise.
578 * src/sparc/ffi.c: Likewise.
579 * src/mips/ffi.c: Likewise.
580 * src/sh/ffi.c: Likewise.
581
5822008-02-15 David Daney <ddaney@avtrex.com>
583
584 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
585 Define (conditionally), and use it to include cachectl.h.
586 (ffi_prep_closure_loc): Fix cache flushing.
587 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
588
9d98ec1b 5892009-06-04 Andrew Haley <aph@redhat.com>
590
591 include/ffi.h.in,
592 src/arm/ffitarget.h,
593 src/arm/ffi.c,
594 src/arm/sysv.S,
595 src/powerpc/ffitarget.h,
596 src/closures.c,
597 src/sh64/ffitarget.h,
598 src/sh64/ffi.c,
599 src/sh64/sysv.S,
600 src/types.c,
601 src/x86/ffi64.c,
602 src/x86/ffitarget.h,
603 src/x86/win32.S,
604 src/x86/darwin.S,
605 src/x86/ffi.c,
606 src/x86/sysv.S,
607 src/x86/unix64.S,
608 src/alpha/ffitarget.h,
609 src/alpha/ffi.c,
610 src/alpha/osf.S,
611 src/m68k/ffitarget.h,
612 src/frv/ffitarget.h,
613 src/frv/ffi.c,
614 src/s390/ffitarget.h,
615 src/s390/sysv.S,
616 src/cris/ffitarget.h,
617 src/pa/linux.S,
618 src/pa/ffitarget.h,
619 src/pa/ffi.c,
620 src/raw_api.c,
621 src/ia64/ffitarget.h,
622 src/ia64/unix.S,
623 src/ia64/ffi.c,
624 src/ia64/ia64_flags.h,
625 src/java_raw_api.c,
626 src/debug.c,
627 src/sparc/v9.S,
628 src/sparc/ffitarget.h,
629 src/sparc/ffi.c,
630 src/sparc/v8.S,
631 src/mips/ffitarget.h,
632 src/mips/n32.S,
633 src/mips/o32.S,
634 src/mips/ffi.c,
635 src/prep_cif.c,
636 src/sh/ffitarget.h,
637 src/sh/ffi.c,
638 src/sh/sysv.S: Update license text.
639
ad81e87b 6402009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
641
642 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
643 (.eh_frame): Add FDE for it.
644
1cde303d 6452009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
646
647 * configure.ac: Also check if assembler supports pc-relative
648 relocs on X86_WIN32 targets.
649 * configure: Regenerate.
650 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
651 (_ffi_call_SYSV): Add missing function type symbol .def and
652 add EH markup labels.
653 (_ffi_call_STDCALL): Likewise.
654 (_ffi_closure_SYSV): Likewise.
655 (_ffi_closure_raw_SYSV): Likewise.
656 (.eh_frame): Add hand-crafted EH data.
657
6bc9506f 6582009-04-09 Jakub Jelinek <jakub@redhat.com>
659
660 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
661 version 3 of the GNU General Public License and to point readers
662 at the COPYING3 file and the FSF's license web page.
663 * testsuite/libffi.call/call.exp: Likewise.
664 * testsuite/libffi.special/special.exp: Likewise.
665
9af964d4 6662009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
667
668 * configure: Regenerate.
669
95ff4d7d 6702008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
671
672 PR libffi/26048
673 * configure.ac (HAVE_AS_X86_PCREL): New test.
674 * configure: Regenerate.
675 * fficonfig.h.in: Regenerate.
676 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
677 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
678 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
679 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
680 * src/x86/unix64.S (.Lstore_table): Move to .text section.
681 (.Lload_table): Likewise.
682 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
683
9e6fd619 6842008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
685
686 * configure: Regenerate.
687
95c3349b 6882008-11-21 Eric Botcazou <ebotcazou@adacore.com>
689
690 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
691 signed/unsigned int8/16 return values.
692 * src/sparc/v8.S (ffi_call_v8): Likewise.
693 (ffi_closure_v8): Likewise.
694
4bc2b84f 6952008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
696 Steve Ellcey <sje@cup.hp.com>
697
698 * configure: Regenerate for new libtool.
699 * Makefile.in: Ditto.
700 * include/Makefile.in: Ditto.
701 * aclocal.m4: Ditto.
702
a2c1e278 7032008-08-25 Andreas Tobler <a.tobler@schweiz.org>
704
705 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
706 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
707 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
708 Adjust copyright notice.
709 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
710 register or two register to use for FFI_SYSV structs.
711 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
712 (ffi_closure_helper_SYSV): Fix the return type for
713 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
714 Adjust copyright notice.
715
0733c867 7162008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
717
718 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
719 int.
720
5b2c7553 7212008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
722
723 * configure: Regenerate.
724 * include/Makefile.in: Regenerate.
725 * testsuite/Makefile.in: Regenerate.
726
6af9f7ea 7272008-06-07 Joseph Myers <joseph@codesourcery.com>
728
729 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
730 powerpc-*-beos*): Remove.
731 * configure: Regenerate.
732
ac051731 7332008-05-09 Julian Brown <julian@codesourcery.com>
734
735 * Makefile.am (LTLDFLAGS): New.
736 (libffi_la_LDFLAGS): Use above.
737 * Makefile.in: Regenerate.
738
6a88da51 7392008-04-18 Paolo Bonzini <bonzini@gnu.org>
740
741 PR bootstrap/35457
742 * aclocal.m4: Regenerate.
743 * configure: Regenerate.
744
644af649 7452008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
746
747 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
748 * src/sh64/sysv.S: Likewise.
749
595c2b68 7502008-03-26 Daniel Jacobowitz <dan@debian.org>
751
752 * src/arm/sysv.S: Fix ARM comment marker.
753
7542008-03-26 Jakub Jelinek <jakub@redhat.com>
755
756 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
757 * src/s390/sysv.S: Likewise.
758 * src/powerpc/ppc_closure.S: Likewise.
759 * src/powerpc/sysv.S: Likewise.
760 * src/x86/unix64.S: Likewise.
761 * src/x86/sysv.S: Likewise.
762 * src/sparc/v8.S: Likewise.
763 * src/sparc/v9.S: Likewise.
764 * src/m68k/sysv.S: Likewise.
765 * src/arm/sysv.S: Likewise.
766
82eaf560 7672008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
768
769 * aclocal.m4: Regenerate.
770 * configure: Likewise.
771 * Makefile.in: Likewise.
772 * include/Makefile.in: Likewise.
773 * testsuite/Makefile.in: Likewise.
774
55e813ce 7752008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
776 Andreas Tobler <a.tobler@schweiz.org>
777
778 * configure.ac: Add amd64-*-freebsd* target.
779 * configure: Regenerate.
780
00eae123 7812008-01-30 H.J. Lu <hongjiu.lu@intel.com>
782
783 PR libffi/34612
784 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
785 returning struct.
786
787 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
788 tests.
789
ff3fb914 7902008-01-24 David Edelsohn <edelsohn@gnu.org>
791
792 * configure: Regenerate.
793
8f1f10d3 7942008-01-06 Andreas Tobler <a.tobler@schweiz.org>
795
796 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
797
ad73551f 7982008-01-05 Andreas Tobler <a.tobler@schweiz.org>
799
800 PR testsuite/32843
801 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
802 signed/unsigned int8/16 for X86_DARWIN.
803 Updated copyright info.
804 Handle one and two byte structs with special cif->flags.
805 * src/x86/ffitarget.h: Add special types for one and two byte structs.
806 Updated copyright info.
807 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
808 sysv.S
809 Remove code to pop args from the stack after call.
810 Special-case signed/unsigned for int8/16, one and two byte structs.
811 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
812 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
813 FFI_TYPE_SINT32.
814 Updated copyright info.
815
aaf16623 8162007-12-08 David Daney <ddaney@avtrex.com>
817
818 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
819 SUBU, add with ADDU and use smaller code sequences.
820
2c3f7870 8212007-12-07 David Daney <ddaney@avtrex.com>
2af77f57 822
2c3f7870 823 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
824 type.
825
5be4032c 8262007-12-06 David Daney <ddaney@avtrex.com>
827
828 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
829 defined.
830 (ffi_java_raw): New typedef.
831 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
832 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
833 ffi_java_raw.
834 (ffi_java_raw_closure) : Same.
835 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
836 parameter types.
837 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
838 FFI_SIZEOF_JAVA_RAW.
839 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
840 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
841 sizeof(ffi_java_raw) for alignment calculations.
842 (ffi_java_ptrarray_to_raw): Same.
843 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
844 if FFI_SIZEOF_JAVA_RAW == 4.
845 (ffi_java_raw_to_rvalue): Same.
846 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
847 (ffi_java_translate_args): Same.
848 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
849 parameter types.
850 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
851
79369c38 8522007-12-06 David Daney <ddaney@avtrex.com>
853
854 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
855 pointer values.
856
7cf0754c 8572007-12-01 Andreas Tobler <a.tobler@schweiz.org>
858
859 PR libffi/31937
860 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
861 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
862 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
863 set the NUM_FPR_ARG_REGISTERS according to.
864 Add support for potential soft-float support under hard-float
865 architecture.
866 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
867 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
868 to the FFI_LINUX_SOFT_FLOAT ABI.
869 (ffi_prep_cif_machdep): Likewise.
870 (ffi_closure_helper_SYSV): Likewise.
871 * src/powerpc/ppc_closure.S: Make sure not to store float/double
872 on archs where __NO_FPRS__ is true.
873 Add FFI_TYPE_UINT128 support.
874 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
875 Adjust copyright notice.
876
5036bf25 8772007-11-25 Andreas Tobler <a.tobler@schweiz.org>
878
879 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
880 * include/ffi_common.h: ... here.
881 Update copyright.
882
c3f75f11 8832007-11-17 Andreas Tobler <a.tobler@schweiz.org>
884
885 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
886 * src/powerpc/linux64.S: Likewise.
887 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
888 * testsuite/libffi.call/return_ldl.c: New test.
889
244c3f18 8902007-09-04 <aph@redhat.com>
891
892 * src/arm/sysv.S (UNWIND): New.
893 (Whole file): Conditionally compile unwinder directives.
894 * src/arm/sysv.S: Add unwinder directives.
895
896 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
897 Only treat r0 as a struct address if we're actually returning a
898 struct by address.
899 Only copy the bytes that are actually within a struct.
900 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
901 is returned in r0, not passed by address.
902 (ffi_call): Allocate a word-sized temporary for the case where
903 a composite is returned in r0.
904 (ffi_prep_incoming_args_SYSV): Align as necessary.
905
9062007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
907
908 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
909 directly using the sys_cacheflush syscall.
910
9112007-07-27 Andrew Haley <aph@redhat.com>
912
913 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
914
7a659e46 9152007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
916
917 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
918 * configure.ac: Likewise.
919 * Makefile.in: Regenerate.
920 * include/Makefile.in: Likewise.
921 * testsuite/Makefile.in: Likewise.
922 * configure: Likewise.
923
99b65499 9242007-08-24 David Daney <ddaney@avtrex.com>
925
926 * testsuite/libffi.call/return_sl.c: New test.
927
b5c438f4 9282007-08-10 David Daney <ddaney@avtrex.com>
929
930 * testsuite/libffi.call/cls_multi_ushort.c,
931 testsuite/libffi.call/cls_align_uint16.c,
932 testsuite/libffi.call/nested_struct1.c,
933 testsuite/libffi.call/nested_struct3.c,
934 testsuite/libffi.call/cls_7_1_byte.c,
935 testsuite/libffi.call/nested_struct5.c,
936 testsuite/libffi.call/cls_double.c,
937 testsuite/libffi.call/nested_struct7.c,
938 testsuite/libffi.call/cls_sint.c,
939 testsuite/libffi.call/nested_struct9.c,
940 testsuite/libffi.call/cls_20byte1.c,
941 testsuite/libffi.call/cls_multi_sshortchar.c,
942 testsuite/libffi.call/cls_align_sint64.c,
943 testsuite/libffi.call/cls_3byte2.c,
944 testsuite/libffi.call/cls_multi_schar.c,
945 testsuite/libffi.call/cls_multi_uchar.c,
946 testsuite/libffi.call/cls_19byte.c,
947 testsuite/libffi.call/cls_9byte1.c,
948 testsuite/libffi.call/cls_align_float.c,
949 testsuite/libffi.call/closure_fn1.c,
950 testsuite/libffi.call/problem1.c,
951 testsuite/libffi.call/closure_fn3.c,
952 testsuite/libffi.call/cls_sshort.c,
953 testsuite/libffi.call/closure_fn5.c,
954 testsuite/libffi.call/cls_align_double.c,
955 testsuite/libffi.call/nested_struct.c,
956 testsuite/libffi.call/cls_2byte.c,
957 testsuite/libffi.call/nested_struct10.c,
958 testsuite/libffi.call/cls_4byte.c,
959 testsuite/libffi.call/cls_6byte.c,
960 testsuite/libffi.call/cls_8byte.c,
961 testsuite/libffi.call/cls_multi_sshort.c,
962 testsuite/libffi.call/cls_align_sint16.c,
963 testsuite/libffi.call/cls_align_uint32.c,
964 testsuite/libffi.call/cls_20byte.c,
965 testsuite/libffi.call/cls_float.c,
966 testsuite/libffi.call/nested_struct2.c,
967 testsuite/libffi.call/cls_5_1_byte.c,
968 testsuite/libffi.call/nested_struct4.c,
969 testsuite/libffi.call/cls_24byte.c,
970 testsuite/libffi.call/nested_struct6.c,
971 testsuite/libffi.call/cls_64byte.c,
972 testsuite/libffi.call/nested_struct8.c,
973 testsuite/libffi.call/cls_uint.c,
974 testsuite/libffi.call/cls_multi_ushortchar.c,
975 testsuite/libffi.call/cls_schar.c,
976 testsuite/libffi.call/cls_uchar.c,
977 testsuite/libffi.call/cls_align_uint64.c,
978 testsuite/libffi.call/cls_ulonglong.c,
979 testsuite/libffi.call/cls_align_longdouble.c,
980 testsuite/libffi.call/cls_1_1byte.c,
981 testsuite/libffi.call/cls_12byte.c,
982 testsuite/libffi.call/cls_3_1byte.c,
983 testsuite/libffi.call/cls_3byte1.c,
984 testsuite/libffi.call/cls_4_1byte.c,
985 testsuite/libffi.call/cls_6_1_byte.c,
986 testsuite/libffi.call/cls_16byte.c,
987 testsuite/libffi.call/cls_18byte.c,
988 testsuite/libffi.call/closure_fn0.c,
989 testsuite/libffi.call/cls_9byte2.c,
990 testsuite/libffi.call/closure_fn2.c,
991 testsuite/libffi.call/closure_fn4.c,
992 testsuite/libffi.call/cls_ushort.c,
993 testsuite/libffi.call/closure_fn6.c,
994 testsuite/libffi.call/cls_5byte.c,
995 testsuite/libffi.call/cls_align_pointer.c,
996 testsuite/libffi.call/cls_7byte.c,
997 testsuite/libffi.call/cls_align_sint32.c,
998 testsuite/libffi.special/unwindtest_ffi_call.cc,
999 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
1000
edd57a0a 10012007-08-10 David Daney <ddaney@avtrex.com>
1002
1003 PR libffi/28313
1004 * configure.ac: Don't treat mips64 as a special case.
1005 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
1006 * configure: Regenerate
1007 * Makefile.in: Ditto.
1008 * fficonfig.h.in: Ditto.
1009 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
1010 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
1011 (FFI_DEFAULT_ABI): Set for n64 case.
1012 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
1013 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
1014 (ffi_closure_N32): New function.
1015 (.eh_frame): New section
1016 * src/mips/o32.S: Clean up comments.
1017 (ffi_closure_O32): Pass ffi_closure parameter in $12.
1018 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
1019 _MIPS_SIM == _ABIN32 throughout.
1020 (FFI_MIPS_STOP_HERE): New, use in place of
1021 ffi_stop_here.
1022 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1023 to support n32/n64 ABIs.
1024 (calc_n32_struct_flags): Rewrite.
1025 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1026 position of flag bits.
1027 (ffi_prep_cif_machdep): Rewrite n32 portion.
1028 (ffi_call): Enable for n64. Add special handling for small structure
1029 return values.
1030 (ffi_prep_closure_loc): Add n32 and n64 support.
1031 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1032 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1033
3471c5eb 10342007-08-08 David Daney <ddaney@avtrex.com>
1035
1036 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1037 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1038 specifiers.
1039 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1040 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1041 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1042 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1043 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1044 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1045 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1046 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1047 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1048 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1049 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1050 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1051 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1052 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1053 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1054 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1055 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1056 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1057 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1058 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1059 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1060 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1061 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1062 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1063 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1064 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1065 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1066 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1067 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1068 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1069 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1070
ad7bf8e7 10712007-08-07 Andrew Haley <aph@redhat.com>
1072
1073 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1074 checkin.
1075
6cc1286e 10762007-08-06 Andrew Haley <aph@redhat.com>
1077
1078 PR testsuite/32843
1079 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1080 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1081 FFI_TYPE_SINT32.
1082
c6f5483d 10832007-08-02 David Daney <ddaney@avtrex.com>
1084
1085 * testsuite/libffi.call/return_ul.c (main): Define return type as
1086 ffi_arg. Use proper printf conversion specifier.
1087
f6179f0e 10882007-07-30 Andrew Haley <aph@redhat.com>
1089
1090 PR testsuite/32843
1091 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1092 signed/unsigned int8/16.
1093 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1094 Use a jump table.
1095 Remove code to pop args from the stack after call.
1096 Special-case signed/unsigned int8/16.
1097 * testsuite/libffi.call/return_sc.c (main): Revert.
1098
7babed79 10992007-07-26 Richard Guenther <rguenther@suse.de>
1100
1101 PR testsuite/32843
1102 * testsuite/libffi.call/return_sc.c (main): Verify call
1103 result as signed char, not ffi_arg.
1104
f54a4242 11052007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1106
1107 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1108 * configure: Regenerate.
1109
e96a69da 11102007-07-11 David Daney <ddaney@avtrex.com>
1111
1112 * src/mips/ffi.c: Don't include sys/cachectl.h.
1113 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1114 cacheflush().
1115
187f6543 11162007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1117
1118 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1119 from (ffi_prep_closure): ... this.
1120 (FFI_INIT_TRAMPOLINE): Adjust.
1121
11222005-12-31 Phil Blundell <pb@reciva.com>
1123
1124 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1125 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1126 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1127 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1128 (FFI_CLOSURES): Enable closure support.
1129
11302007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1131
1132 * testsuite/libffi.call/cls_multi_ushort.c,
1133 testsuite/libffi.call/cls_align_uint16.c,
1134 testsuite/libffi.call/nested_struct1.c,
1135 testsuite/libffi.call/nested_struct3.c,
1136 testsuite/libffi.call/cls_7_1_byte.c,
1137 testsuite/libffi.call/cls_double.c,
1138 testsuite/libffi.call/nested_struct5.c,
1139 testsuite/libffi.call/nested_struct7.c,
1140 testsuite/libffi.call/cls_sint.c,
1141 testsuite/libffi.call/nested_struct9.c,
1142 testsuite/libffi.call/cls_20byte1.c,
1143 testsuite/libffi.call/cls_multi_sshortchar.c,
1144 testsuite/libffi.call/cls_align_sint64.c,
1145 testsuite/libffi.call/cls_3byte2.c,
1146 testsuite/libffi.call/cls_multi_schar.c,
1147 testsuite/libffi.call/cls_multi_uchar.c,
1148 testsuite/libffi.call/cls_19byte.c,
1149 testsuite/libffi.call/cls_9byte1.c,
1150 testsuite/libffi.call/cls_align_float.c,
1151 testsuite/libffi.call/closure_fn1.c,
1152 testsuite/libffi.call/problem1.c,
1153 testsuite/libffi.call/closure_fn3.c,
1154 testsuite/libffi.call/cls_sshort.c,
1155 testsuite/libffi.call/closure_fn5.c,
1156 testsuite/libffi.call/cls_align_double.c,
1157 testsuite/libffi.call/cls_2byte.c,
1158 testsuite/libffi.call/nested_struct.c,
1159 testsuite/libffi.call/nested_struct10.c,
1160 testsuite/libffi.call/cls_4byte.c,
1161 testsuite/libffi.call/cls_6byte.c,
1162 testsuite/libffi.call/cls_8byte.c,
1163 testsuite/libffi.call/cls_multi_sshort.c,
1164 testsuite/libffi.call/cls_align_uint32.c,
1165 testsuite/libffi.call/cls_align_sint16.c,
1166 testsuite/libffi.call/cls_float.c,
1167 testsuite/libffi.call/cls_20byte.c,
1168 testsuite/libffi.call/cls_5_1_byte.c,
1169 testsuite/libffi.call/nested_struct2.c,
1170 testsuite/libffi.call/cls_24byte.c,
1171 testsuite/libffi.call/nested_struct4.c,
1172 testsuite/libffi.call/nested_struct6.c,
1173 testsuite/libffi.call/cls_64byte.c,
1174 testsuite/libffi.call/nested_struct8.c,
1175 testsuite/libffi.call/cls_uint.c,
1176 testsuite/libffi.call/cls_multi_ushortchar.c,
1177 testsuite/libffi.call/cls_schar.c,
1178 testsuite/libffi.call/cls_uchar.c,
1179 testsuite/libffi.call/cls_align_uint64.c,
1180 testsuite/libffi.call/cls_ulonglong.c,
1181 testsuite/libffi.call/cls_align_longdouble.c,
1182 testsuite/libffi.call/cls_1_1byte.c,
1183 testsuite/libffi.call/cls_12byte.c,
1184 testsuite/libffi.call/cls_3_1byte.c,
1185 testsuite/libffi.call/cls_3byte1.c,
1186 testsuite/libffi.call/cls_4_1byte.c,
1187 testsuite/libffi.call/cls_6_1_byte.c,
1188 testsuite/libffi.call/cls_16byte.c,
1189 testsuite/libffi.call/cls_18byte.c,
1190 testsuite/libffi.call/closure_fn0.c,
1191 testsuite/libffi.call/cls_9byte2.c,
1192 testsuite/libffi.call/closure_fn2.c,
1193 testsuite/libffi.call/closure_fn4.c,
1194 testsuite/libffi.call/cls_ushort.c,
1195 testsuite/libffi.call/closure_fn6.c,
1196 testsuite/libffi.call/cls_5byte.c,
1197 testsuite/libffi.call/cls_align_pointer.c,
1198 testsuite/libffi.call/cls_7byte.c,
1199 testsuite/libffi.call/cls_align_sint32.c,
1200 testsuite/libffi.special/unwindtest_ffi_call.cc,
1201 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1202
cecc75a9 12032007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1204
1205 * aclocal.m4: Regenerated.
1206
07c46636 12072007-06-02 Paolo Bonzini <bonzini@gnu.org>
1208
1209 * configure: Regenerate.
1210
542f9784 12112007-05-23 Steve Ellcey <sje@cup.hp.com>
1212
1213 * Makefile.in: Regenerate.
1214 * configure: Regenerate.
1215 * aclocal.m4: Regenerate.
1216 * include/Makefile.in: Regenerate.
1217 * testsuite/Makefile.in: Regenerate.
1218
0e3a3bdb 12192007-05-10 Roman Zippel <zippel@linux-m68k.org>
1220
1221 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1222 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1223 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1224 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1225 (FFI_CLOSURES): Enable closure support.
1226
a53add80 12272007-05-10 Roman Zippel <zippel@linux-m68k.org>
1228
1229 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1230 * configure: Regenerate.
1231 * fficonfig.h.in: Regenerate.
1232 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1233 CFI_OFFSET,CFI_DEF_CFA): New macros.
1234 (ffi_call_SYSV): Add callframe annotation.
1235
bb3218e9 12362007-05-10 Roman Zippel <zippel@linux-m68k.org>
1237
1238 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1239 numerous test suite failures.
1240 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1241
5d0df87f 12422007-04-11 Paolo Bonzini <bonzini@gnu.org>
1243
1244 * Makefile.am (EXTRA_DIST): Bring up to date.
1245 * Makefile.in: Regenerate.
1246 * src/frv/eabi.S: Remove RCS keyword.
1247
ea4d299a 12482007-04-06 Richard Henderson <rth@redhat.com>
1249
1250 * configure.ac: Tidy target case.
1251 (HAVE_LONG_DOUBLE): Allow the target to override.
1252 * configure: Regenerate.
1253 * include/ffi.h.in: Don't define ffi_type_foo if
1254 LIBFFI_HIDE_BASIC_TYPES is defined.
1255 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1256 to ffi_type_double.
1257 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1258 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1259 (ffi_type_longdouble): Special case for Alpha. Don't define
1260 if long double == double.
1261
1262 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1263 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1264 (ffi_call, ffi_closure_osf_inner): Likewise.
1265 (ffi_closure_osf_inner): Likewise. Mark hidden.
1266 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1267 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1268 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1269 (load_table): Handle 128-bit long double.
1270
1271 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1272
770e494d 12732007-04-06 Tom Tromey <tromey@redhat.com>
1274
1275 PR libffi/31491:
1276 * README: Fixed bug in example.
1277
58606959 12782007-04-03 Jakub Jelinek <jakub@redhat.com>
1279
1280 * src/closures.c: Include sys/statfs.h.
1281 (_GNU_SOURCE): Define on Linux.
1282 (FFI_MMAP_EXEC_SELINUX): Define.
1283 (selinux_enabled): New variable.
1284 (selinux_enabled_check): New function.
1285 (is_selinux_enabled): Define.
1286 (dlmmap): Use it.
1287
7fed151e 12882007-03-24 Uros Bizjak <ubizjak@gmail.com>
1289
1290 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1291 Use 'volatile float sum' to create sum of floats to avoid false
1292 negative due to excess precision on ix86 targets.
1293 (main): Ditto.
1294
8fb4628d 12952007-03-08 Alexandre Oliva <aoliva@redhat.com>
1296
1297 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1298 patch.
1299 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1300
f20fe604 13012007-03-07 Alexandre Oliva <aoliva@redhat.com>
1302
1303 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1304 (ffi_prep_closure_loc): New.
1305 (ffi_prep_raw_closure_loc): New.
1306 (ffi_prep_java_raw_closure_loc): New.
1307 * src/closures.c: New file.
1308 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1309 Replace sflags with exec_offset.
1310 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1311 sub_segment_exec_offset): New macros.
1312 (get_segment_flags, set_segment_flags, check_segment_merge): New
1313 macros.
1314 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1315 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1316 destroy_mspace): Use new macros.
1317 (sys_alloc): Silence warning.
1318 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1319 * Makefile.in: Rebuilt.
1320 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1321 terms of ffi_prep_closure_loc.
1322 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1323 from...
1324 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1325 renamed version.
1326 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1327 adjusted from...
1328 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1329 the renamed version.
1330 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1331 (ffi_prep_closure): ... this.
1332 * src/pa/ffi.c: Likewise.
1333 * src/cris/ffi.c: Likewise. Adjust.
1334 * src/frv/ffi.c: Likewise.
1335 * src/ia64/ffi.c: Likewise.
1336 * src/mips/ffi.c: Likewise.
1337 * src/powerpc/ffi_darwin.c: Likewise.
1338 * src/s390/ffi.c: Likewise.
1339 * src/sh/ffi.c: Likewise.
1340 * src/sh64/ffi.c: Likewise.
1341 * src/sparc/ffi.c: Likewise.
1342 * src/x86/ffi64.c: Likewise.
1343 * src/x86/ffi.c: Likewise.
1344 (FFI_INIT_TRAMPOLINE): Adjust.
1345 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1346 (ffi_prep_raw_closure): ... this.
1347 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1348 (ffi_prep_closure): ... this.
1349 (flush_icache): Adjust.
1350
c3265eac 13512007-03-07 Alexandre Oliva <aoliva@redhat.com>
1352
1353 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1354 Lea's malloc.
1355
0ce4a5ce 13562007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1357
1358 * Makefile.am: Add dummy install-pdf target.
1359 * Makefile.in: Regenerate
1360
45444175 13612007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1362
1363 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1364 ffi_closure_helper_SYSV): Add long double handling.
1365
1df875cc 13662007-02-02 Jakub Jelinek <jakub@redhat.com>
1367
1368 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1369 immediately after bctrl instruction.
1370
27c4194f 13712007-01-18 Alexandre Oliva <aoliva@redhat.com>
1372
1373 * Makefile.am (all-recursive, install-recursive,
1374 mostlyclean-recursive, clean-recursive, distclean-recursive,
1375 maintainer-clean-recursive): Add missing targets.
1376 * Makefile.in: Rebuilt.
1377
eb6316a2 13782006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1379
1380 * configure.ac: Add TARGET for x86_64-*-darwin*.
1381 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1382 for X86_DARWIN.
1383 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1384 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1385 * configure: Regenerate.
1386 * Makefile.in: Regenerate.
1387 * include/Makefile.in: Regenerate.
1388 * testsuite/Makefile.in: Regenerate.
1389 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1390 ffi_call only.
1391
75a99c5e 13922006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1393
1394 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1395 Makefile.am.
1396
a39e926c 13972006-10-31 Geoffrey Keating <geoffk@apple.com>
c8dfb8ef 1398
1399 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1400 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1401 Darwin.
1402 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1403 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1404
eb819ecd 14052006-10-10 Paolo Bonzini <bonzini@gnu.org>
1406 Sandro Tolaini <tolaini@libero.it>
1407
1408 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1409 conditional.
1410 * configure: Regenerated.
1411 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1412 (EXTRA_DIST): Add src/x86/darwin.S.
1413 * Makefile.in: Regenerated.
1414 * include/Makefile.in: Regenerated.
1415 * testsuite/Makefile.in: Regenerated.
1416
1417 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1418 X86_WIN32, and additionally align stack to 16 bytes.
1419 * src/x86/darwin.S: New, based on sysv.S.
1420 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1421
23d9d417 14222006-09-12 David Daney <ddaney@avtrex.com>
1423
1424 PR libffi/23935
1425 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1426 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1427 * aclocal.m4: Regenerated for automake 1.9.6.
1428 * Makefile.in: Regenerated.
1429 * include/Makefile.in: Regenerated.
1430 * testsuite/Makefile.in: Regenerated.
1431
ec045626 14322006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1433
1434 * include/ffi_common.h (struct): Revert accidental commit.
1435
b9463909 14362006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1437
1438 * include/ffi_common.h: Remove lint directives.
1439 * include/ffi.h.in: Likewise.
1440
c695c024 14412006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1442
1443 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1444 for 32-bit architectures.
1445 * testsuite/libffi.call/return_ul.c: New test case.
1446
3a9d2a82 14472006-07-19 David Daney <ddaney@avtrex.com>
1448
1449 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1450 xfail remains for mips64.
1451
8f16e3ad 14522006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1453
1454 * Makefile.am: Add install-html target. Add install-html to .PHONY
1455 * Makefile.in: Regenerate.
1456 * aclocal.m4: Regenerate.
1457 * include/Makefile.in: Regenerate.
1458 * testsuite/Makefile.in: Regenerate.
1459
db951627 14602006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1461
1462 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1463 stack slot.
1464
9a333c86 14652006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1466
1467 * README: Remove notice about 'Crazy Comments'.
1468 * src/debug.c: Remove lint directives. Cleanup white spaces.
1469 * src/java_raw_api.c: Likewise.
1470 * src/prep_cif.c: Likewise.
1471 * src/raw_api.c: Likewise.
1472 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1473 to the testsuite.
1474 * src/arm/ffi.c: Remove lint directives.
1475 * src/m32r/ffi.c: Likewise.
1476 * src/pa/ffi.c: Likewise.
1477 * src/powerpc/ffi.c: Likewise.
1478 * src/powerpc/ffi_darwin.c: Likewise.
1479 * src/sh/ffi.c: Likewise.
1480 * src/sh64/ffi.c: Likewise.
1481 * src/x86/ffi.c: Likewise.
1482 * testsuite/libffi.call/float2.c: Likewise.
1483 * testsuite/libffi.call/promotion.c: Likewise.
1484 * testsuite/libffi.call/struct1.c: Likewise.
1485
7610fe2b 14862006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1487
1488 * src/pa/hpux32.S: Correct unwind offset calculation for
1489 ffi_closure_pa32.
1490 * src/pa/linux.S: Likewise.
1491
5adc6320 14922006-04-12 James E Wilson <wilson@specifix.com>
1493
1494 PR libgcj/26483
1495 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1496 (hfa_type_load): Call stf_spill.
1497 (hfa_type_store): Call ldf_fill.
1498 (ffi_call): Adjust calls to above routines. Add local temps for
1499 macro result.
7610fe2b 1500
d35f63a9 15012006-04-10 Matthias Klose <doko@debian.org>
1502
1503 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1504 directory names containing underscores.
1505
2d3dcfaa 15062006-04-07 James E Wilson <wilson@specifix.com>
1507
1508 * testsuite/libffi.call/float4.c: New testcase.
1509
1a82d6f6 15102006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1511 Andreas Tobler <a.tobler@schweiz.ch>
1512
1513 * Makefile.am: Add PA_HPUX port.
1514 * Makefile.in: Regenerate.
1515 * include/Makefile.in: Likewise.
1516 * testsuite/Makefile.in: Likewise.
1517 * configure.ac: Add PA_HPUX rules.
1518 * configure: Regenerate.
1519 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1520 Add PA_HPUX and PA64_HPUX.
1521 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1522 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1523 (FFI_TYPE_SMALL_STRUCT2): Define.
1524 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1525 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1526 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1527 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1528 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1529 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1530 * src/pa/ffi.c (ROUND_DOWN): Delete.
1531 (fldw, fstw, fldd, fstd): Use '__asm__'.
1532 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1533 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1534 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1535 Simplify incrementing of stack slot variable. Change type of local
1536 'n' to unsigned int.
1537 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1538 double on PA_HPUX.
1539 (ffi_prep_cif_machdep): Likewise.
1540 (ffi_call): Likewise.
1541 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1542 return type to ffi_status. Simplify incrementing of stack slot
1543 variable. Only copy floating point argument registers when PA_LINUX
1544 is true. Reformat debug statement.
1545 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1546 FFI_TYPE_SMALL_STRUCT8.
1547 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1548 declaration.
1549 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1550 Add nops to cache flush. Add trampoline for PA_HPUX.
1551 * src/pa/hpux32.S: New file.
1552 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1553 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1554 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1555 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1556 argument type checks so that common argument types appear first.
1557 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1558 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1559
7f5ddc97 15602006-03-24 Alan Modra <amodra@bigpond.net.au>
1561
1562 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1563 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1564 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1565 FFI_TYPE_LONGDOUBLE.
1566 (ffi_prep_args64): Assert using IBM extended double.
1567 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1568 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1569 (ffi_call): Handle FFI_LINUX.
1570 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1571 gpr3 return pointer as for struct return. Handle FFI_LINUX
1572 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1573 unnecessarily.
1574 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1575 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1576 Don't use r6 as pointer to results, instead use sp offset. Don't
1577 make a special call to load lr with case table address, instead
1578 use offset from previous call.
1579 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1580 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1581 return.
1582
cd10a5a7 15832006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1584
1585 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1586 passed with FP registers correctly.
1587 (ffi_closure_helper_SYSV): Likewise.
1588 * src/sh64/sysv.S: Likewise.
1589
a07533b6 15902006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1591
1592 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1593 args and userdata unused.
1594 (closure_test_fn1): Mark cif and userdata unused.
1595 (main): Remove unused res.
1596
bdc6f488 15972006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1598
1599 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1600 -O2, -O3, -Os and the warning flags -W -Wall.
1601 * testsuite/libffi.special/special.exp: Likewise.
1602 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1603 unused parameter unused for gcc or else do nothing.
1604 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1605 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1606 and userdata unused.
1607 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1608 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1609 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1610 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1611 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1612 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1613 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1614 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1615 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1616 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1617 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1618 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1619 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1620 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1621 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1622 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1623 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1624 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1625 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1626 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1627 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1628 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1629 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1630 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1631 Likewise.
1632 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1633 Likewise.
1634 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1635 Likewise.
1636 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1637 void* to avoid compiler warning.
1638 (main): Likewise.
1639 (cls_struct_align_gn): Mark cif and userdata unused.
1640 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1641 Likewise.
1642 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1643 Likewise.
1644 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1645 Likewise.
1646 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1647 Likewise.
1648 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1649 Likewise.
1650 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1651 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1652 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1653 data unused.
1654 (main): Cast res_call to silence gcc.
1655 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1656 data unused.
1657 (main): Cast res_call to silence gcc.
1658 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1659 and data unused.
1660 (main): Cast res_call to silence gcc.
1661 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1662 data unused.
1663 (main): Cast res_call to silence gcc.
1664 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1665 data unused.
1666 (main): Cast res_call to silence gcc.
1667 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1668 and data unused.
1669 (main): Cast res_call to silence gcc.
1670 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1671 userdata unused.
1672 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1673 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1674 userdata unused.
1675 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1676 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1677 userdata unused.
1678 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1679 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1680 userdata unused.
1681 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1682 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1683 userdata unused.
1684 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1685 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1686 and userdata unused.
1687 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1688 userdata unused.
1689 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1690 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1691 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1692 Cleanup white spaces.
1693 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1694 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1695 cif and userdata unused.
1696 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1697 Likewise.
1698 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1699 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1700 formatters to silence gcc.
1701 (B_gn): Mark cif and userdata unused.
1702 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1703 unused.
1704 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1705 (B_gn): Mark cif and userdata unused.
1706 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1707 unused.
1708 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1709 (B_gn): Mark cif and userdata unused.
1710 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1711 unused.
1712 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1713 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1714 * testsuite/libffi.call/problem1.c (stub): Likewise.
1715 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1716 gcc.
1717 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1718 in the last commit for this test case in the test case itself.
1719 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1720 unused.
1721 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1722 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1723 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1724 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1725 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1726 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1727
5ab2e6e4 17282006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1729
1730 * src/sh/sysv.S: Fix register numbers in the FDE for
1731 ffi_closure_SYSV.
1732
849b9f91 17332006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1734
1735 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1736 declaration to avoid a false negative on ix86. See PR323.
1737
003f258f 17382006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1739
1740 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1741 and cast integer to void * if needed. Update the pointer to
1742 the FP register saved area correctly.
1743
9121151b 17442006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1745
1746 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1747 is fixed.
1748 * testsuite/libffi.call/nested_struct4.c: Likewise.
1749
019a3b64 17502006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1751
6cbfba8c 1752 * testsuite/libffi.call/return_dbl.c: New test case.
1753 * testsuite/libffi.call/return_dbl1.c: Likewise.
1754 * testsuite/libffi.call/return_dbl2.c: Likewise.
1755 * testsuite/libffi.call/return_fl.c: Likewise.
1756 * testsuite/libffi.call/return_fl1.c: Likewise.
1757 * testsuite/libffi.call/return_fl2.c: Likewise.
1758 * testsuite/libffi.call/return_fl3.c: Likewise.
1759 * testsuite/libffi.call/closure_fn6.c: Likewise.
1760
019a3b64 1761 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1762 definition.
1763 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1764 here to be used by other test cases too.
1765
1766 * testsuite/libffi.call/nested_struct10.c: New test case.
1767 * testsuite/libffi.call/nested_struct9.c: Likewise.
1768 * testsuite/libffi.call/nested_struct8.c: Likewise.
1769 * testsuite/libffi.call/nested_struct7.c: Likewise.
1770 * testsuite/libffi.call/nested_struct6.c: Likewise.
1771 * testsuite/libffi.call/nested_struct5.c: Likewise.
1772 * testsuite/libffi.call/nested_struct4.c: Likewise.
1773
946a8d62 17742006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1775
1776 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1777 * configure: Rebuilt.
1778
cb99d24e 17792006-01-18 Jakub Jelinek <jakub@redhat.com>
1780
1781 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1782 instead do the shifting inline.
1783 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1784 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1785 and 8 byte structs, for the remaining struct sizes don't call
1786 __lshrdi3, instead do the shifting inline.
1787
06bf2366 17882005-12-07 Thiemo Seufer <ths@networkno.de>
1789
1790 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1791 missing parentheses.
1792 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1793 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1794 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1795 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1796 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1797 FA_0_0_OFF2.
1798 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1799 endianness bugs.
1800 (ffi_prep_closure): Improve trampoline instruction scheduling.
1801 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1802
9a385484 18032005-12-03 Alan Modra <amodra@bigpond.net.au>
1804
1805 * src/powerpc/ffi.c: Formatting.
1806 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1807 (ffi_prep_args64): Likewise.
1808
193c2942 18092005-09-30 Geoffrey Keating <geoffk@apple.com>
1810
1811 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1812 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1813
d3eb23fb 18142005-09-26 Tom Tromey <tromey@redhat.com>
1815
1816 * testsuite/libffi.call/float1.c (value_type): New typedef.
1817 (CANARY): New define.
1818 (main): Check for result buffer overflow.
1819 * src/powerpc/linux64.S: Handle linux64 long double returns.
1820 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1821 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1822
5497bbdf 18232005-08-25 Alan Modra <amodra@bigpond.net.au>
1824
1825 PR target/23404
1826 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1827 homed fp args.
1828 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1829
33b4ca90 18302005-08-11 Jakub Jelinek <jakub@redhat.com>
1831
1832 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1833 (AH_BOTTOM): Add FFI_HIDDEN definition.
1834 * configure: Rebuilt.
1835 * fficonfig.h.in: Rebuilt.
1836 * src/powerpc/ffi.c (hidden): Remove.
1837 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1838 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1839 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1840 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1841 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1842 add FFI_HIDDEN to its prototype.
1843 (ffi_closure_SYSV_inner): New.
1844 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1845 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1846
967ec51b 18472005-08-10 Alfred M. Szmidt <ams@gnu.org>
1848
1849 PR libffi/21819:
1850 * configure: Rebuilt.
1851 * configure.ac: Handle i*86-*-gnu*.
1852
80207f81 18532005-08-09 Jakub Jelinek <jakub@redhat.com>
1854
1855 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1856 DW_CFA_offset_extended_sf rather than
1857 DW_CFA_GNU_negative_offset_extended.
1858 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1859
6a0e74dc 18602005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1861
1862 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1863 on sh3.
1864 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1865 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1866 partially on register.
1867 (ffi_closure_helper_SYSV): Likewise.
1868 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1869
cef3f52b 18702005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1871
1872 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1873 Remove empty line.
1874 * src/sh64/ffi.c (simple_type): Remove.
1875 (return_type): Handle small structures correctly.
1876 (ffi_prep_args): Likewise.
1877 (ffi_call): Likewise.
1878 (ffi_closure_helper_SYSV): Likewise.
1879 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1880 Emit position independent code if PIC and remove wrong datalabel
1881 prefixes from EH data.
1882
66c44239 18832005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1884
1885 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1886 * Makefile.in: Regenerate.
1887 * include/Makefile.in: Likewise.
1888 * testsuite/Makefile.in: Likewise.
1889 * configure.ac: Add POWERPC_FREEBSD rules.
1890 * configure: Regenerate.
1891 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1892 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1893 * src/powerpc/ffi.c: Add flags to handle small structure returns
1894 in ffi_call_SYSV.
1895 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1896 Aka FFI_SYSV.
1897 (ffi_closure_helper_SYSV): Likewise.
1898 * src/powerpc/ppc_closure.S: Add return types for small structures.
1899 * src/powerpc/sysv.S: Add bits to handle small structures for
1900 final SYSV 4 ABI.
1901
a21bd97f 19022005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1903
1904 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1905 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1906 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1907
c350327d 19082005-07-05 Randolph Chung <tausq@debian.org>
1909
1910 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1911 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1912 structures. Kill compilation warnings.
1913 (ffi_closure_inner_LINUX): Print return values as hex in debug
1914 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1915 Properly handle 5-7 byte structure returns.
1916 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1917 (FFI_TYPE_SMALL_STRUCT2): Remove.
1918 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1919 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1920 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1921 (checksmst1, checksmst2): Remove.
1922 (checksmst3): Optimize handling of 3-byte struct returns.
1923 (checksmst567): Properly handle 5-7 byte struct returns.
1924
20750e69 19252005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1926
1927 PR libgcj/21943
1928 * src/mips/n32.S: Enforce PIC code.
1929 * src/mips/o32.S: Likewise.
1930
58841df8 19312005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1932
1933 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1934 * configure: Regenerate.
1935
cc46fb30 19362005-06-01 Alan Modra <amodra@bigpond.net.au>
1937
1938 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1939 to call ffi_closure_helper_SYSV. Append @local instead.
1940 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1941
6e912c1e 19422005-05-17 Kelley Cook <kcook@gcc.gnu.org>
69738b31 1943
1944 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1945 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1946 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1947 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1948 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1949
5f0a087e 19502005-05-09 Mike Stump <mrs@apple.com>
1951
1952 * configure: Regenerate.
1953
81734468 19542005-05-08 Richard Henderson <rth@redhat.com>
1955
1956 PR libffi/21285
1957 * src/alpha/osf.S: Update unwind into to match code.
1958
4e9df534 19592005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1960 Richard Henderson <rth@redhat.com>
1961
1962 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1963 bit 11 of flags.
1964 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1965 (ffi_prep_closure): Set carry bit if sse-used flag set.
1966 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1967 Only load sse registers if ssecount non-zero.
1968 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1969
19702005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
6155bf60 1971
1972 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1973 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1974 * configure: Regenerate.
1975
643e4a19 19762005-04-20 Hans-Peter Nilsson <hp@axis.com>
1977
1978 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1979 have Tcl8.3-compatible intermediate variable.
1980
7e5823cf 19812005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1982 Hans-Peter Nilsson <hp@axis.com>
1983
1984 * Makefile.am: Add CRIS support.
1985 * configure.ac: Likewise.
942132a7 1986 * Makefile.in, configure, testsuite/Makefile.in,
1987 include/Makefile.in: Regenerate.
7e5823cf 1988 * src/cris: New directory.
1989 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1990 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
f0a2eecf 1991
1992 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1993 \r?\n in output tests.
1994
3bcf2cd5 19952005-04-12 Mike Stump <mrs@apple.com>
1996
1997 * configure: Regenerate.
1998
b7292ee2 19992005-03-30 Hans Boehm <Hans.Boehm@hp.com>
2000
2001 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
2002
3bdc9b0e 20032005-03-30 Steve Ellcey <sje@cup.hp.com>
2004
2005 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
2006 (ffi_sarg) Ditto.
2007 * src/ia64/unix.S (ffi_closure_unix): Extend gp
2008 to 64 bits in ILP32 mode.
2009 Load 64 bits even for short data.
2010
7abd8367 20112005-03-23 Mike Stump <mrs@apple.com>
2012
2013 * src/powerpc/darwin.S: Update for -m64 multilib.
2014 * src/powerpc/darwin_closure.S: Likewise.
2015
bbb07d35 20162005-03-21 Zack Weinberg <zack@codesourcery.com>
2017
2018 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
2019 Do not set tool_include_dir.
2020 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
2021 Regenerate.
2022 * include/Makefile.am: Set gcc_version and toollibffidir.
2023 * include/Makefile.in: Regenerate.
2024
e6fd470b 20252005-02-22 Andrew Haley <aph@redhat.com>
2026
2027 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2028 odd-numbered register pairs for 64-bit integer types.
2029
4f373876 20302005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2031
2032 PR libffi/20104
2033 * testsuite/libffi.call/return_ll1.c: New test case.
2034
c2cf854f 20352005-02-11 Janis Johnson <janis187@us.ibm.com>
2036
2037 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2038 * testsuite/libffi.call/float.c: Ditto.
2039 * testsuite/libffi.call/float2.c: Ditto.
2040 * testsuite/libffi.call/float3.c: Ditto.
2041
9e71f2c8 20422005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2043
2044 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2045
45e075b4 20462005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2047
2048 * testsuite/libffi.special/special.exp (cxx_options): Add
2049 -shared-libgcc.
2050
f587fca2 20512004-12-31 Richard Henderson <rth@redhat.com>
2052
2053 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2054 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2055 offset parameters with a type parameter; deduce size and structure
2056 alignment. Update all users.
2057
717c502e 20582004-12-31 Richard Henderson <rth@redhat.com>
2059
2060 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2061 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
bbb07d35 2062 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
717c502e 2063 into ffi_prep_closure.
2064 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2065 from scratch.
2066
4a60643d 20672004-12-27 Richard Henderson <rth@redhat.com>
2068
2069 * src/x86/unix64.S: Fix typo in unwind info.
2070
66d912b3 20712004-12-25 Richard Henderson <rth@redhat.com>
2072
2073 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2074 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2075 (merge_classes): Check for it.
2076 (SSE_CLASS_P): New.
2077 (classify_argument): Pass byte_offset by value; perform all updates
2078 inside struct case.
2079 (examine_argument): Add classes argument; handle
2080 X86_64_COMPLEX_X87_CLASS.
2081 (ffi_prep_args): Merge into ...
2082 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2083 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2084 (ffi_fill_return_value): Remove.
2085 (ffi_prep_closure): Remove dead assert.
2086 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2087 Rewrite to use struct register_args instead of va_list. Create
2088 flags for handling structure returns.
2089 * src/x86/unix64.S: Remove dead strings.
2090 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2091 stack frame with ffi_call. Handle structure returns properly.
2092 (float2sse, floatfloat2sse, double2sse): Remove.
2093 (sse2float, sse2double, sse2floatfloat): Remove.
2094 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2095 to handle structure returns properly.
2096
a8594486 20972004-12-08 David Edelsohn <edelsohn@gnu.org>
2098
2099 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2100 PICFLAG.
2101 * Makefile.in: Regenerated.
2102
14ac9144 21032004-12-02 Richard Sandiford <rsandifo@redhat.com>
2104
2105 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2106 * configure, aclocal.m4, Makefile.in: Regenerate.
2107 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2108
d081dd29 21092004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2110
2111 * configure: Regenerate for libtool change.
2112
a234d196 21132004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2114
2115 * configure: Regenerate for libtool reversion.
2116
7f93fe7a 21172004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2118
2119 * configure: Regenerate for libtool change.
2120
c0e31427 21212004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2122
2123 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2124
21252004-11-23 Richard Sandiford <rsandifo@redhat.com>
2126
2127 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2128 of jal. Use an absolute encoding for the frame information.
2129
21302004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2131
2132 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2133 * acinclude.m4: Delete logic for sincludes.
2134 * aclocal.m4, Makefile.in, configure: Regenerate.
2135 * include/Makefile: Likewise.
2136 * testsuite/Makefile: Likewise.
2137
68fa8e89 21382004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2139
2140 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2141 on a 8-byte boundary.
2142 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2143
c11820d9 21442004-10-27 Richard Earnshaw <rearnsha@arm.com>
2145
2146 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2147 long long values. Round stack allocation to a multiple of 8 bytes
2148 for ATPCS compatibility.
2149 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2150 names. Handle returning long long types. Add Thumb and interworking
2151 support. Improve soft-float code.
2152
03b2b968 21532004-10-27 Richard Earnshaw <rearnsha@arm.com>
2154
2155 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2156 (libffi_exit): New function.
2157 (libffi_init): Build the testglue wrapper if needed.
2158
9f234538 21592004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2160
2161 PR other/18138
2162 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2163
61dc6540 21642004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2165
2166 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2167
3b77fb00 21682004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2169
2170 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2171 * testsuite/libffi.call/float3.c: New test case.
2172
7c2fd652 21732004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2174
2175 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2176 the function returning a structure pointed with R2.
2177 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2178 the structure return value if T bit set. Emit position
2179 independent code and EH data if PIC.
2180
7e74b6d4 21812004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2182
2183 * Makefile.am: Add m32r support.
2184 * configure.ac: Likewise.
2185 * Makefile.in: Regenerate.
2186 * confiugre: Regenerate.
2187 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2188 (uint64, sint64, double, longdouble)
2189 * src/m32r: New directory.
2190 * src/m32r/ffi.c: New file.
2191 * src/m32r/sysv.S: Likewise.
2192 * src/m32r/ffitarget.h: Likewise.
2193
d601026d 21942004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2195
2196 * testsuite/libffi.call/negint.c: New test case.
2197
ec1bc4ab 21982004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2199
2200 PR libgcj/17465
2201 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2202 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2203 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2204 DYLD_LIBRARY_PATH.
2205
41067a17 22062004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2207
2208 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2209 * testsuite/libffi.call/promotion.c: Likewise.
2210 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2211 whitespaces.
2212 * testsuite/libffi.call/return_sc.c: Likewise.
2213 * testsuite/libffi.call/return_uc.c: Likewise.
2214
acbdb39a 22152004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2216
2217 * src/powerpc/darwin.S: Fix comments and identation.
2218 * src/powerpc/darwin_closure.S: Likewise.
2219
de7fad36 22202004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2221
2222 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2223 (ffi_prep_args): Handle longdouble arguments.
2224 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2225 longdouble.
2226 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2227 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2228 values.
2229 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2230 * src/types.c: Defined longdouble size and alignment for darwin.
2231
8bf7bfd1 22322004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2233
2234 * src/powerpc/aix.S: Remove whitespaces.
2235 * src/powerpc/aix_closure.S: Likewise.
2236 * src/powerpc/asm.h: Likewise.
2237 * src/powerpc/ffi.c: Likewise.
2238 * src/powerpc/ffitarget.h: Likewise.
2239 * src/powerpc/linux64.S: Likewise.
2240 * src/powerpc/linux64_closure.S: Likewise.
2241 * src/powerpc/ppc_closure.S: Likewise.
2242 * src/powerpc/sysv.S: Likewise.
2243
2e818f50 22442004-08-30 Anthony Green <green@redhat.com>
2245
2246 * Makefile.am: Add frv support.
2247 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2248 * configure.ac: Read configure.host.
2249 * configure.in: Read configure.host.
2250 * configure.host: New file. frv-elf needs libgloss.
2251 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
752da755 2252 alignment. This is needed to frv and shouldn't harm the others.
2e818f50 2253 * include/ffi_common.h (ALIGN_DOWN): New macro.
2254 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2255
d36998ee 22562004-08-24 David Daney <daney@avtrex.com>
2257
2258 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2259 * testsuite/libffi.call/closure_fn1.c: Likewise.
2260 * testsuite/libffi.call/closure_fn2.c Likewise.
2261 * testsuite/libffi.call/closure_fn3.c: Likewise.
2262 * testsuite/libffi.call/closure_fn4.c: Likewise.
2263 * testsuite/libffi.call/closure_fn5.c: Likewise.
2264 * testsuite/libffi.call/cls_18byte.c: Likewise.
2265 * testsuite/libffi.call/cls_19byte.c: Likewise.
2266 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2267 * testsuite/libffi.call/cls_20byte.c: Likewise.
2268 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2269 * testsuite/libffi.call/cls_24byte.c: Likewise.
2270 * testsuite/libffi.call/cls_2byte.c: Likewise.
2271 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2272 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2273 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2274 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2275 * testsuite/libffi.call/cls_4byte.c: Likewise.
2276 * testsuite/libffi.call/cls_64byte.c: Likewise.
2277 * testsuite/libffi.call/cls_6byte.c: Likewise.
2278 * testsuite/libffi.call/cls_7byte.c: Likewise.
2279 * testsuite/libffi.call/cls_8byte.c: Likewise.
2280 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2281 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2282 * testsuite/libffi.call/cls_align_double.c: Likewise.
2283 * testsuite/libffi.call/cls_align_float.c: Likewise.
2284 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2285 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2286 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2287 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2288 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2289 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2290 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2291 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2292 * testsuite/libffi.call/cls_double.c: Likewise.
2293 * testsuite/libffi.call/cls_float.c: Likewise.
2294 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2295 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2296 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2297 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2298 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2299 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2300 * testsuite/libffi.call/cls_schar.c: Likewise.
2301 * testsuite/libffi.call/cls_sint.c: Likewise.
2302 * testsuite/libffi.call/cls_sshort.c: Likewise.
2303 * testsuite/libffi.call/cls_uchar.c: Likewise.
2304 * testsuite/libffi.call/cls_uint.c: Likewise.
2305 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2306 * testsuite/libffi.call/cls_ushort.c: Likewise.
2307 * testsuite/libffi.call/nested_struct.c: Likewise.
2308 * testsuite/libffi.call/nested_struct1.c: Likewise.
2309 * testsuite/libffi.call/nested_struct2.c: Likewise.
2310 * testsuite/libffi.call/nested_struct3.c: Likewise.
2311 * testsuite/libffi.call/problem1.c: Likewise.
2312 * testsuite/libffi.special/unwindtest.cc: Likewise.
2313 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2314 to zero.
2315 * testsuite/libffi.call/cls_16byte.c: Likewise.
2316 * testsuite/libffi.call/cls_5byte.c: Likewise.
2317
29ac6c83 23182004-08-23 David Daney <daney@avtrex.com>
2319
2320 PR libgcj/13141
2321 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2322 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2323 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2324 parameters and return types.
2325 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2326 (ffi_prep_closure): Ditto.
2327 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2328 alignment calculations.
2329 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2330 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2331
4487a7e8 23322004-08-14 Casey Marshall <csm@gnu.org>
2333
2334 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2335 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2336 integer (O32 ABI only).
2337 (ffi_prep_closure): new function.
2338 (ffi_closure_mips_inner_O32): new function.
bbb07d35 2339 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
4487a7e8 2340 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2341 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2342 64 bit integers correctly.
2343 (ffi_closure_O32): new function.
2344 Added DWARF-2 unwind info for both functions.
2345
bb2d2363 23462004-08-10 Andrew Haley <aph@redhat.com>
2347
2348 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2349
cb8a7007 23502004-08-01 Robert Millan <robertmh@gnu.org>
2351
2352 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2353 * configure: Regenerate.
2354
e2f66690 23552004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2356
2357 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2358 and mmap() explicitly instead of relying on preset autoconf cache
2359 variables.
2360 * aclocal.m4: Regenerate.
2361 * configure: Regenerate.
2362
e395522d 23632004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2364
2365 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2366 (ffi_check_float_struct): Remove unused prototype.
2367
b589ba75 23682004-06-30 Geoffrey Keating <geoffk@apple.com>
2369
2370 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2371 character on Darwin, use '\n\t' instead.
2372
7b709c61 23732004-06-26 Matthias Klose <doko@debian.org>
2374
2375 * libtool-version: Fix typo in revision/age.
2376
06e2077b 23772004-06-17 Matthias Klose <doko@debian.org>
2378
2379 * libtool-version: New.
2380 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2381 * Makefile.in: Regenerate.
2382
4889e098 23832004-06-15 Paolo Bonzini <bonzini@gnu.org>
d947df39 2384
2385 * Makefile.am: Remove useless multilib rules.
2386 * Makefile.in: Regenerate.
2387 * aclocal.m4: Regenerate with automake 1.8.5.
2388 * configure.ac: Remove useless multilib configury.
2389 * configure: Regenerate.
2390
4889e098 23912004-06-15 Paolo Bonzini <bonzini@gnu.org>
b2191698 2392
2393 * .cvsignore: New file.
2394
6bc491fa 23952004-06-10 Jakub Jelinek <jakub@redhat.com>
2396
2397 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2398 fp_done.
2399 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2400 changed from 8.
2401
e1c1d65f 24022004-06-06 Sean McNeil <sean@mcneil.com>
2403
2404 * configure.ac: Add x86_64-*-freebsd* support.
2405 * configure: Regenerate.
2406
2be99030 24072004-04-26 Joe Buck <jbuck@welsh-buck.org>
2408
2409 Bug 15093
2410 * configure.ac: Test for existence of mmap and sys/mman.h before
2411 checking blacklist. Fix suggested by Jim Wilson.
2412 * configure: Regenerate.
2413
efefde65 24142004-04-26 Matt Austern <austern@apple.com>
2415
2416 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2417 FDE location.
2418 * src/powerpc/darwin_closure.S: Likewise.
bbb07d35 2419
5766842b 24202004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2421
2422 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2423 error. Reported by Thomas Heller <theller@python.net>.
2424 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2425 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2426
43935228 24272004-03-20 Matthias Klose <doko@debian.org>
2428
2429 * src/pa/linux.S: Fix typo.
2430
f35c338f 24312004-03-19 Matthias Klose <doko@debian.org>
2432
5766842b 2433 * Makefile.am: Update.
f35c338f 2434 * Makefile.in: Regenerate.
2435 * src/pa/ffi.h.in: Remove.
2436 * src/pa/ffitarget.h: New file.
2437
24382004-02-10 Randolph Chung <tausq@debian.org>
2439
2440 * Makefile.am: Add PA support.
2441 * Makefile.in: Regenerate.
2442 * include/Makefile.in: Regenerate.
2443 * configure.ac: Add PA target.
2444 * configure: Regenerate.
2445 * src/pa/ffi.c: New file.
2446 * src/pa/ffi.h.in: Add PA support.
2447 * src/pa/linux.S: New file.
2448 * prep_cif.c: Add PA support.
2449
021bf237 24502004-03-16 Hosaka Yuji <hos@tamanegi.org>
2451
2452 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2453 double.
2454 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2455 with ecif->cif->flags.
2456 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2457 with cif->flags.
2458 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2459 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2460 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2461 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2462
4a900809 24632004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2464
2465 * configure.in: Rename file to ...
2466 * configure.ac: ... this.
2467 * fficonfig.h.in: Regenerate.
2468 * Makefile.in: Regenerate.
2469 * include/Makefile.in: Regenerate.
2470 * testsuite/Makefile.in: Regenerate.
2471
d128584a 24722004-03-12 Matt Austern <austern@apple.com>
2473
2474 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2475 changes in EH format resulting from addition of linkonce support.
2476 * src/powerpc/darwin_closure.S: Likewise.
021bf237 2477
b375837d 24782004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2479 Paolo Bonzini <bonzini@gnu.org>
2480
2481 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2482 Remove VPATH. Remove rules for object files. Remove multilib support.
2483 (AM_CCASFLAGS): Add.
2484 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2485 (AC_PREREQ): Bump version to 2.59.
2486 (AC_INIT): Fill with version info and bug address.
2487 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2488 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2489 De-precious CC so that the right flags are passed down to multilibs.
2490 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2491 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2492 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2493 * configure: Rebuilt.
2494 * aclocal.m4: Likewise.
2495 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2496 * fficonfig.h.in: Likewise.
2497
52cee448 24982004-03-11 Andreas Schwab <schwab@suse.de>
2499
2500 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2501 arguments from fp registers only for the first 8 parameter slots.
2502 Don't convert a float parameter when passed in memory.
2503
ac09b42b 25042004-03-09 Hans-Peter Nilsson <hp@axis.com>
2505
2506 * configure: Regenerate for config/accross.m4 correction.
2507
497c7020 25082004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2509
2510 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2511 ecif->cif->bytes to bytes.
2512 (ffi_prep_cif_machdep): Add braces around nested if statement.
2513
0ad895fb 25142004-02-09 Alan Modra <amodra@bigpond.net.au>
2515
ad98ab60 2516 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2517
0ad895fb 2518 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2519 (ffi_closure_helper_LINUX64): Fix typo.
2520 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2521 for powerpc64-*-*.
2522 * testsuite/libffi.call/float.c: Likewise.
2523 * testsuite/libffi.call/float2.c: Likewise.
2524
de9c7178 25252004-02-08 Alan Modra <amodra@bigpond.net.au>
2526
2527 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2528 long double function return and long double arg handling.
2529 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2530 Use "end_pfr" instead of "nf". Correct long double handling.
2531 Localise "temp".
2532 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2533 return value.
2534 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2535 space for long double return value. Adjust stack frame and offsets.
2536 Load f2 long double return.
2537
3474a085 25382004-02-07 Alan Modra <amodra@bigpond.net.au>
2539
2540 * src/types.c: Use 16 byte long double for POWERPC64.
2541
071ff5d6 25422004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2543
2544 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2545 when the structure return address is passed in %o0.
2546 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2547 (ffi_v9_layout_struct): Align the field following a nested structure
2548 on a word boundary. Use memmove instead of memcpy.
2549 (ffi_call): Update call to ffi_V9_return_struct.
2550 (ffi_prep_closure): Define 'ctx' only for V8.
2551 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2552 and ffi_closure_sparc_inner_v9.
2553 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2554 Always skip the structure return address. For structures and long
2555 doubles, copy the argument directly.
2556 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2557 if required. Shift the maximum floating-point slot accordingly. For
2558 big structures, copy the argument directly; otherwise, left-justify the
2559 argument and call ffi_v9_layout_struct to lay out the structure on
2560 the stack.
2561 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2562 (ffi_closure_v8): Pass the structure return address. Update call to
2563 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2564 Skip the 'unimp' insn when returning long doubles and structures.
2565 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2566 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2567 FFI_TYPE_INT handling. Load structures both in integers and
2568 floating-point registers on return.
2569 * README: Update status of the SPARC port.
2570
24ee5f14 25712004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2572
2573 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2574 as of type ffi_arg.
2575 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2576
1e0ee4d8 25772004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2578
2579 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2580 value as of type ffi_arg, not unsigned int.
2581
c886ae3f 25822004-01-21 Michael Ritzert <ritzert@t-online.de>
2583
2584 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2585 of the LHS.
2586
fcd85db3 25872004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2588
2589 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2590 Solaris.
2591
dcf910e2 25922004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2593
2594 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2595 to void *.
2596
3d3042b8 25972003-12-10 Richard Henderson <rth@redhat.com>
2598
2599 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2600 size_t instead of int.
2601
dc36d4d1 26022003-12-04 Hosaka Yuji <hos@tamanegi.org>
2603
2604 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2605 * testsuite/libffi.call/many_win32.c (main): Replace variable
2606 int i with unsigned long ul.
2607
2608 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2609 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2610 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2611 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2612 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2613 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2614 * testsuite/libffi.call/cls_align_float.c: Likewise.
2615 * testsuite/libffi.call/cls_align_double.c: Likewise.
2616 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2617 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2618
f90351a5 26192003-12-02 Hosaka Yuji <hos@tamanegi.org>
2620
2621 PR other/13221
2622 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2623 Align arguments to 32 bits.
2624
1c4a2a48 26252003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2626
2627 PR other/13221
2628 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2629 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2630 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2631 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2632 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2633 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2634
2635 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2636
8c899267 26372003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2638
2639 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2640 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2641
857794c9 26422003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2643
2644 * Makefile.in: Rebuilt.
2645 * configure: Likewise.
2646 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2647 the right type.
2648
e7f352cc 26492003-11-21 Andreas Jaeger <aj@suse.de>
2650 Andreas Tobler <a.tobler@schweiz.ch>
2651
2652 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2653 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2654 * Makefile.in: Rebuilt.
2655 * aclocal.m4: Likewise.
2656 * configure: Likewise.
2657 * fficonfig.h.in: Likewise.
2658 * testsuite/lib/libffi-dg.exp: Add include dir.
2659 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2660 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2661 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2662 for ffi_closure if available.
2663 * testsuite/libffi.call/closure_fn1.c: Likewise.
2664 * testsuite/libffi.call/closure_fn2.c: Likewise.
2665 * testsuite/libffi.call/closure_fn3.c: Likewise.
2666 * testsuite/libffi.call/closure_fn4.c: Likewise.
2667 * testsuite/libffi.call/closure_fn5.c: Likewise.
2668 * testsuite/libffi.call/cls_12byte.c: Likewise.
2669 * testsuite/libffi.call/cls_16byte.c: Likewise.
2670 * testsuite/libffi.call/cls_18byte.c: Likewise.
2671 * testsuite/libffi.call/cls_19byte.c: Likewise.
2672 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2673 * testsuite/libffi.call/cls_20byte.c: Likewise.
2674 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2675 * testsuite/libffi.call/cls_24byte.c: Likewise.
2676 * testsuite/libffi.call/cls_2byte.c: Likewise.
2677 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2678 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2679 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2680 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2681 * testsuite/libffi.call/cls_4byte.c: Likewise.
2682 * testsuite/libffi.call/cls_5byte.c: Likewise.
2683 * testsuite/libffi.call/cls_64byte.c: Likewise.
2684 * testsuite/libffi.call/cls_6byte.c: Likewise.
2685 * testsuite/libffi.call/cls_7byte.c: Likewise.
2686 * testsuite/libffi.call/cls_8byte.c: Likewise.
2687 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2688 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2689 * testsuite/libffi.call/cls_double.c: Likewise.
2690 * testsuite/libffi.call/cls_float.c: Likewise.
2691 * testsuite/libffi.call/cls_schar.c: Likewise.
2692 * testsuite/libffi.call/cls_sint.c: Likewise.
2693 * testsuite/libffi.call/cls_sshort.c: Likewise.
2694 * testsuite/libffi.call/cls_uchar.c: Likewise.
2695 * testsuite/libffi.call/cls_uint.c: Likewise.
2696 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2697 * testsuite/libffi.call/cls_ushort.c: Likewise.
2698 * testsuite/libffi.call/nested_struct.c: Likewise.
2699 * testsuite/libffi.call/nested_struct1.c: Likewise.
2700 * testsuite/libffi.call/nested_struct2.c: Likewise.
2701 * testsuite/libffi.call/nested_struct3.c: Likewise.
2702 * testsuite/libffi.call/problem1.c: Likewise.
2703 * testsuite/libffi.special/unwindtest.cc: Likewise.
2704
71e96ac5 27052003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2706
2707 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2708
fd0a337b 27092003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2710
2711 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2712 Add -lgcc_s to additional flags.
2713
c2dfd8f6 27142003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2715
2716 * configure.in, include/Makefile.am: PR libgcj/11147, install
2717 the ffitarget.h header file in a gcc versioned and target
2718 dependent place.
2719 * configure: Regenerated.
2720 * Makefile.in, include/Makefile.in: Likewise.
2721 * testsuite/Makefile.in: Likewise.
2722
17c9ab63 27232003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2724
2725 * testsuite/libffi.call/closure_fn0.c: Print result and check
2726 with dg-output to make debugging easier.
2727 * testsuite/libffi.call/closure_fn1.c: Likewise.
2728 * testsuite/libffi.call/closure_fn2.c: Likewise.
2729 * testsuite/libffi.call/closure_fn3.c: Likewise.
2730 * testsuite/libffi.call/closure_fn4.c: Likewise.
2731 * testsuite/libffi.call/closure_fn5.c: Likewise.
2732 * testsuite/libffi.call/cls_12byte.c: Likewise.
2733 * testsuite/libffi.call/cls_16byte.c: Likewise.
2734 * testsuite/libffi.call/cls_18byte.c: Likewise.
2735 * testsuite/libffi.call/cls_19byte.c: Likewise.
2736 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2737 * testsuite/libffi.call/cls_20byte.c: Likewise.
2738 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2739 * testsuite/libffi.call/cls_24byte.c: Likewise.
2740 * testsuite/libffi.call/cls_2byte.c: Likewise.
2741 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2742 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2743 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2744 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2745 * testsuite/libffi.call/cls_4byte.c: Likewise.
2746 * testsuite/libffi.call/cls_5byte.c: Likewise.
2747 * testsuite/libffi.call/cls_64byte.c: Likewise.
2748 * testsuite/libffi.call/cls_6byte.c: Likewise.
2749 * testsuite/libffi.call/cls_7byte.c: Likewise.
2750 * testsuite/libffi.call/cls_8byte.c: Likewise.
2751 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2752 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2753 * testsuite/libffi.call/cls_double.c: Likewise.
2754 * testsuite/libffi.call/cls_float.c: Likewise.
2755 * testsuite/libffi.call/cls_schar.c: Likewise.
2756 * testsuite/libffi.call/cls_sint.c: Likewise.
2757 * testsuite/libffi.call/cls_sshort.c: Likewise.
2758 * testsuite/libffi.call/cls_uchar.c: Likewise.
2759 * testsuite/libffi.call/cls_uint.c: Likewise.
2760 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2761 * testsuite/libffi.call/cls_ushort.c: Likewise.
2762 * testsuite/libffi.call/problem1.c: Likewise.
2763
2764 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2765 static.
2766
a0b53ce0 27672003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2768
2769 * testsuite/libffi.call/cls_9byte2.c: New test case.
2770 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2771 * testsuite/libffi.call/cls_64byte.c: Likewise.
2772 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2773 * testsuite/libffi.call/cls_19byte.c: Likewise.
2774 * testsuite/libffi.call/cls_18byte.c: Likewise.
2775 * testsuite/libffi.call/closure_fn4.c: Likewise.
2776 * testsuite/libffi.call/closure_fn5.c: Likewise.
2777 * testsuite/libffi.call/cls_schar.c: Likewise.
2778 * testsuite/libffi.call/cls_sint.c: Likewise.
2779 * testsuite/libffi.call/cls_sshort.c: Likewise.
2780 * testsuite/libffi.call/nested_struct2.c: Likewise.
2781 * testsuite/libffi.call/nested_struct3.c: Likewise.
2782
ed0ed2e9 27832003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2784
2785 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2786 * testsuite/libffi.call/cls_uchar.c: Likewise.
2787 * testsuite/libffi.call/cls_uint.c: Likewise.
2788 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2789 * testsuite/libffi.call/cls_ushort.c: Likewise.
2790 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2791
61641cc0 27922003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2793
2794 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2795 the initialization.
2796
6a759884 27972003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2798
61641cc0 2799 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
6a759884 2800 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2801
4f388831 28022003-10-22 David Daney <ddaney@avtrex.com>
2803
2804 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2805 __attribute__((__mode__(__SI__))) and friends.
2806
b4891aa6 28072003-10-22 Andreas Schwab <schwab@suse.de>
2808
2809 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2810
3d37e05c 28112003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2812
2813 * configure.in: AC_LINK_FILES(ffitarget.h).
2814 * configure: Regenerate.
2815 * Makefile.in: Likewise.
2816 * include/Makefile.in: Likewise.
2817 * testsuite/Makefile.in: Likewise.
2818 * fficonfig.h.in: Likewise.
2819
28202003-10-21 Paolo Bonzini <bonzini@gnu.org>
2821 Richard Henderson <rth@redhat.com>
2822
2823 Avoid that ffi.h includes fficonfig.h.
2824
2825 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2826 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2827 (TARGET_SRC_MIPS_SGI): Removed.
2828 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2829 (MIPS_SGI): Removed.
2830 (CLEANFILES): Removed.
2831 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2832 targets.
2833 * acconfig.h: Removed.
2834 * configure.in: Compute sizeofs only for double and long double.
2835 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2836 into AC_DEFINE instead of using acconfig.h. Create
2837 include/ffitarget.h instead of include/fficonfig.h. Rename
2838 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2839 AC_DEFINE EH_FRAME_FLAGS.
2840 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2841 (hack_DATA): Add ffitarget.h.
2842 * include/ffi.h.in: Remove all system specific definitions.
2843 Declare raw API even if it is not installed, why bother?
2844 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2845 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2846 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2847 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2848 * include/ffi_common.h (bool): Do not define.
2849 (ffi_assert): Accept failed assertion.
2850 (ffi_type_test): Return void and accept file/line.
2851 (FFI_ASSERT): Pass stringized failed assertion.
2852 (FFI_ASSERT_AT): New macro.
2853 (FFI_ASSERT_VALID_TYPE): New macro.
2854 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2855 UINT64, SINT64): Define here with gcc's __attribute__ macro
2856 instead of in ffi.h
2857 (FLOAT32, ALIGN): Define here instead of in ffi.h
2858 * include/ffi-mips.h: Removed. Its content moved to
2859 src/mips/ffitarget.h after separating assembly and C sections.
2860 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2861 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2862 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2863 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2864 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2865 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2866 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2867 (ffi_type_test): Rewritten.
2868 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2869 FFI_ASSERT_VALID_TYPE.
2870 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2871 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2872 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2873 src/s390/ffitarget.h, src/sh/ffitarget.h,
2874 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2875 src/x86/ffitarget.h: New files.
2876 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2877 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2878 src/powerpc/aix.S, src/powerpc/darwin.S,
2879 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2880 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2881 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2882 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2883 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2884 include fficonfig.h
2885
2a5f9ffc 28862003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2887
2888 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2889 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2890
51c5f96c 28912003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2892
2893 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2894 Used when FFI_DEBUG = 1.
2895
ee46e78f 28962003-10-14 Alan Modra <amodra@bigpond.net.au>
2897
2898 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2899 and align.
2900
a3abe7fe 29012003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2902
2903 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2904 FFI_MIPS_O32 for O32 ABI.
2905
1ee4561e 29062003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2907
2908 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2909 SPARC64. Cleanup whitespaces.
2910
93421fe6 29112003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2912
2913 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2914 strongarm, xscale. Cleanup whitespaces.
2915 * testsuite/libffi.call/closure_fn1.c: Likewise.
2916 * testsuite/libffi.call/closure_fn2.c: Likewise.
2917 * testsuite/libffi.call/closure_fn3.c: Likewise.
2918 * testsuite/libffi.call/cls_12byte.c: Likewise.
2919 * testsuite/libffi.call/cls_16byte.c: Likewise.
2920 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2921 * testsuite/libffi.call/cls_20byte.c: Likewise.
2922 * testsuite/libffi.call/cls_24byte.c: Likewise.
2923 * testsuite/libffi.call/cls_2byte.c: Likewise.
2924 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2925 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2926 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2927 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2928 * testsuite/libffi.call/cls_4byte.c: Likewise.
2929 * testsuite/libffi.call/cls_5byte.c: Likewise.
2930 * testsuite/libffi.call/cls_6byte.c: Likewise.
2931 * testsuite/libffi.call/cls_7byte.c: Likewise.
2932 * testsuite/libffi.call/cls_8byte.c: Likewise.
2933 * testsuite/libffi.call/cls_double.c: Likewise.
2934 * testsuite/libffi.call/cls_float.c: Likewise.
2935 * testsuite/libffi.call/cls_uchar.c: Likewise.
2936 * testsuite/libffi.call/cls_uint.c: Likewise.
2937 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2938 * testsuite/libffi.call/cls_ushort.c: Likewise.
2939 * testsuite/libffi.call/nested_struct.c: Likewise.
2940 * testsuite/libffi.call/nested_struct1.c: Likewise.
2941 * testsuite/libffi.call/problem1.c: Likewise.
2942 * testsuite/libffi.special/unwindtest.cc: Likewise.
2943 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2944
8a94dd75 29452003-09-18 David Edelsohn <edelsohn@gnu.org>
2946
2947 * src/powerpc/aix.S: Cleanup whitespaces.
2948 * src/powerpc/aix_closure.S: Likewise.
2949
bd1f3533 29502003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2951
2952 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2953 * src/powerpc/darwin_closure.S: Likewise.
2954 * src/powerpc/ffi_darwin.c: Likewise.
2955
7a142533 29562003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2957 David Edelsohn <edelsohn@gnu.org>
2958
2959 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2960 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2961 parameter stack.
2962 * src/powerpc/darwin_closure.S: Likewise.
2963 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2964 according to the Darwin/AIX ABI.
2965 (ffi_prep_cif_machdep): Likewise.
2966 (ffi_closure_helper_DARWIN): Likewise.
2967 Remove the outgoing parameter stack logic. Simplify the evaluation
2968 of the different CASE types.
2969 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2970 statement in the trampoline code.
2971
d80727ae 29722003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2973
2974 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2975 for the register size.
2976 (ffi_closure_helper_SYSV): Handle the structure return value
2977 address correctly.
2978 (ffi_closure_helper_SYSV): Return the appropriate type when
2979 the registers are used for the structure return value.
2980 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2981 the 64-bit return value. Update copyright years.
2982
47fab6c3 29832003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2984
2985 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2986 srcdir for ffi_mips.h.
2987
f397026a 29882003-09-12 Alan Modra <amodra@bigpond.net.au>
2989
2990 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2991 structure size.
2992 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2993 placement of float result.
2994 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2995 cast of "resp" for big-endian 64 bit machines.
2996
341b7dcf 29972003-09-11 Alan Modra <amodra@bigpond.net.au>
2998
2999 * src/types.c (double, longdouble): Merge identical SH and ARM
3000 typedefs, and add POWERPC64.
3001 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
3002 struct split over gpr and rest.
3003 (ffi_prep_cif_machdep): Correct intarg_count for structures.
3004 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
3005
06855f3a 30062003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
3007
3008 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
3009 passing correctly.
c886ae3f 3010
9165b0af 30112003-09-09 Alan Modra <amodra@bigpond.net.au>
3012
3013 * configure: Regenerate.
3014
98d58335 30152003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3016
3017 * Makefile.am: Remove build rules for ffitest.
3018 * Makefile.in: Rebuilt.
c886ae3f 3019
98d58335 30202003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
0262ee1f 3021
3022 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3023 about implicit declaration of abort().
3024
3743c332 30252003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3026
3027 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3028 * Makefile.in: Rebuilt.
3029 * configure.in: Add dejagnu test framework.
3030 * configure: Rebuilt.
3031
3032 * testsuite/Makefile.am: New file.
3033 * testsuite/Makefile.in: Built
c886ae3f 3034 * testsuite/lib/libffi-dg.exp: New file.
3743c332 3035 * testsuite/config/default.exp: Likewise.
3036 * testsuite/libffi.call/call.exp: Likewise.
3037 * testsuite/libffi.call/ffitest.h: Likewise.
3038 * testsuite/libffi.call/closure_fn0.c: Likewise.
3039 * testsuite/libffi.call/closure_fn1.c: Likewise.
3040 * testsuite/libffi.call/closure_fn2.c: Likewise.
3041 * testsuite/libffi.call/closure_fn3.c: Likewise.
3042 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3043 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3044 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3045 * testsuite/libffi.call/cls_2byte.c: Likewise.
3046 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3047 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3048 * testsuite/libffi.call/cls_4byte.c: Likewise.
3049 * testsuite/libffi.call/cls_5byte.c: Likewise.
3050 * testsuite/libffi.call/cls_6byte.c: Likewise.
3051 * testsuite/libffi.call/cls_7byte.c: Likewise.
3052 * testsuite/libffi.call/cls_8byte.c: Likewise.
3053 * testsuite/libffi.call/cls_12byte.c: Likewise.
3054 * testsuite/libffi.call/cls_16byte.c: Likewise.
3055 * testsuite/libffi.call/cls_20byte.c: Likewise.
3056 * testsuite/libffi.call/cls_24byte.c: Likewise.
3057 * testsuite/libffi.call/cls_double.c: Likewise.
3058 * testsuite/libffi.call/cls_float.c: Likewise.
3059 * testsuite/libffi.call/cls_uchar.c: Likewise.
3060 * testsuite/libffi.call/cls_uint.c: Likewise.
3061 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3062 * testsuite/libffi.call/cls_ushort.c: Likewise.
3063 * testsuite/libffi.call/float.c: Likewise.
3064 * testsuite/libffi.call/float1.c: Likewise.
3065 * testsuite/libffi.call/float2.c: Likewise.
3066 * testsuite/libffi.call/many.c: Likewise.
3067 * testsuite/libffi.call/many_win32.c: Likewise.
3068 * testsuite/libffi.call/nested_struct.c: Likewise.
3069 * testsuite/libffi.call/nested_struct1.c: Likewise.
3070 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3071 * testsuite/libffi.call/problem1.c: Likewise.
3072 * testsuite/libffi.call/promotion.c: Likewise.
3073 * testsuite/libffi.call/return_ll.c: Likewise.
3074 * testsuite/libffi.call/return_sc.c: Likewise.
3075 * testsuite/libffi.call/return_uc.c: Likewise.
3076 * testsuite/libffi.call/strlen.c: Likewise.
3077 * testsuite/libffi.call/strlen_win32.c: Likewise.
3078 * testsuite/libffi.call/struct1.c: Likewise.
3079 * testsuite/libffi.call/struct2.c: Likewise.
3080 * testsuite/libffi.call/struct3.c: Likewise.
3081 * testsuite/libffi.call/struct4.c: Likewise.
3082 * testsuite/libffi.call/struct5.c: Likewise.
3083 * testsuite/libffi.call/struct6.c: Likewise.
3084 * testsuite/libffi.call/struct7.c: Likewise.
3085 * testsuite/libffi.call/struct8.c: Likewise.
c886ae3f 3086 * testsuite/libffi.call/struct9.c: Likewise.
3743c332 3087 * testsuite/libffi.special/special.exp: New file.
3088 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3089 * testsuite/libffi.special/unwindtest.cc: Likewise.
3090
c886ae3f 3091
3a845f75 30922003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3093
3094 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3095 copyright years.
3096
4637dfa6 30972003-08-02 Alan Modra <amodra@bigpond.net.au>
3098
3099 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3100 structure passing.
3101 (ffi_closure_helper_LINUX64): Likewise.
3102 * src/powerpc/linux64.S: Remove code writing to parm save area.
3103 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3104 address in lr from ffi_closure_helper_LINUX64 call to calculate
3105 table address. Optimize function tail.
3106
8bcdc7f0 31072003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3108
3109 * src/sparc/ffi.c: Handle all floating point registers.
3110 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3111
3d6cc549 31122003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3113
3114 * README: Note that libffi is not part of GCC. Update the project
3115 URL and status.
3116
9b770b5a 31172003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3118
3119 * src/powerpc/ppc_closure.S: Include ffi.h.
3120
d46ab9f3 31212003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3122
3123 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3124 Use C style comments.
3125
81f536a3 31262003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3127
3128 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3129 * Makefile.in: Regenerate.
3130 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3131 * configure: Regenerate.
3132 * include/ffi.h.in: Add SHmedia support.
3133 * src/sh64/ffi.c: New file.
3134 * src/sh64/sysv.S: New file.
3135
1da888ec 31362003-05-16 Jakub Jelinek <jakub@redhat.com>
3137
3138 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3139 should be read-only.
3140 * configure: Rebuilt.
3141 * fficonfig.h.in: Rebuilt.
3142 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3143 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3144 * src/powerpc/linux64.S: Likewise.
3145 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3146 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3147 if -fpic/-fPIC/-mrelocatable.
3148 * src/powerpc/powerpc_closure.S: Likewise.
3149 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3150 #write in .eh_frame flags.
3151 * src/sparc/v9.S: Likewise.
3152 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3153 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3154 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3155
f3bd7a2c 31562003-05-07 Jeff Sturm <jsturm@one-point.com>
3157
3158 Fixes PR bootstrap/10656
3159 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3160 support for .register pseudo-op.
3161 * src/sparc/v8.S: Use it.
3162 * fficonfig.h.in: Rebuilt.
3163 * configure: Rebuilt.
3164
3647950d 31652003-04-18 Jakub Jelinek <jakub@redhat.com>
3166
3167 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3168 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3169 Make it the default on POWERPC64.
3170 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3171 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3172 * configure: Rebuilt.
3173 * src/powerpc/ffi.c (hidden): Define.
3174 (ffi_prep_args_SYSV): Renamed from
3175 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3176 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3177 ASM_NEEDS_REGISTERS64): New.
3178 (ffi_prep_args64): New function.
3179 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3180 (ffi_call): Likewise.
3181 (ffi_prep_closure): Likewise.
3182 (flush_icache): Surround by #ifndef POWERPC64.
3183 (ffi_dblfl): New union type.
3184 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3185 (ffi_closure_helper_LINUX64): New function.
3186 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3187 __powerpc64__.
3188 * src/powerpc/sysv.S: Likewise.
3189 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3190 * src/powerpc/linux64.S: New file.
3191 * src/powerpc/linux64_closure.S: New file.
3192 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3193 src/powerpc/linux64_closure.S.
3194 (TARGET_SRC_POWERPC): Likewise.
3195
3196 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3197 closure_test_fn3): Fix result printing on big-endian 64-bit
3198 machines.
3199 (main): Print tst2_arg instead of uninitialized tst2_result.
3200
3201 * src/ffitest.c (main): Hide what closure pointer really points to
3202 from the compiler.
3203
1636e920 32042003-04-16 Richard Earnshaw <rearnsha@arm.com>
3205
3206 * configure.in (arm-*-netbsdelf*): Add configuration.
3207 (configure): Regenerated.
3208
e663564a 32092003-04-04 Loren J. Rittle <ljrittle@acm.org>
3210
3211 * include/Makefile.in: Regenerate.
3212
02f5d4c9 32132003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3214
3215 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3216 bit mode.
3217 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3218 Receive closure pointer through parameter, read args using
3219 __builtin_dwarf_cfa.
3220 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3221
b8e27232 32222003-03-12 Andreas Schwab <schwab@suse.de>
3223
3224 * configure.in: Avoid trailing /. in toolexeclibdir.
3225 * configure: Rebuilt.
3226
5fee68c6 32272003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3228
bbb07d35 3229 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
5fee68c6 3230
ecadab3f 32312003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3232
3233 * libffi/src/powerpc/darwin_closure.S:
3234 Fix alignement bug, allocate 8 bytes for the result.
c886ae3f 3235 * libffi/src/powerpc/aix_closure.S:
ecadab3f 3236 Likewise.
3237 * libffi/src/powerpc/ffi_darwin.c:
3238 Update stackframe description for aix/darwin_closure.S.
c886ae3f 3239
a0294641 32402003-02-06 Jakub Jelinek <jakub@redhat.com>
3241
3242 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3243 attribute.
3244
624476ed 32452003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3246 Andreas Schwab <schwab@suse.de>
3247
bbb07d35 3248 * configure.in: Adjust command to source config-ml.in to account
624476ed 3249 for changes to the libffi_basedir definition.
3250 (libffi_basedir): Remove ${srcdir} from value and include trailing
3251 slash if nonempty.
c886ae3f 3252
bbb07d35 3253 * configure: Regenerate.
624476ed 3254
1143cc9b 32552003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3256
3257 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3258
967ae8fb 32592003-01-28 Andrew Haley <aph@redhat.com>
3260
3261 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3262 * src/x86/ffi64.c (ffi_prep_closure): New.
3263 (ffi_closure_UNIX64_inner): New.
3264 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3265
c8d4fa8c 32662003-01-27 Alexandre Oliva <aoliva@redhat.com>
3267
3268 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3269 Remove USE_LIBDIR conditional.
3270 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3271 * Makefile.in, configure: Rebuilt.
3272
2dd7852c 32732003-01027 David Edelsohn <edelsohn@gnu.org>
3274
3275 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3276 * Makefile.in: Regenerate.
3277
008eadcd 32782003-01-22 Andrew Haley <aph@redhat.com>
3279
3280 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3281 unwind info.
3282
cb440058 32832003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3284
3285 * src/powerpc/darwin.S: Add unwind info.
3286 * src/powerpc/darwin_closure.S: Likewise.
3287
149417bc 32882003-01-14 Andrew Haley <aph@redhat.com>
3289
3290 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3291 (ffi_prep_cif_machdep): Likewise.
3292 * src/x86/unix64.S: Add unwind info.
3293
d4dcd973 32942003-01-14 Andreas Jaeger <aj@suse.de>
3295
3296 * src/ffitest.c (main): Only use ffi_closures if those are
3297 supported.
3298
a7898d0c 32992003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3300
3301 * libffi/src/ffitest.c
bbb07d35 3302 add closure testcases
a7898d0c 3303
33042003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3305
3306 * libffi/src/powerpc/ffi.c
bbb07d35 3307 fix alignment bug for float (4 byte aligned iso 8 byte)
a7898d0c 3308
4e844289 33092003-01-09 Geoffrey Keating <geoffk@apple.com>
3310
3311 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3312 * src/powerpc/darwin.S: Remove RCS version string.
3313
0ef3e9c5 33142003-01-03 Jeff Sturm <jsturm@one-point.com>
3315
3316 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3317 * src/ffitest.c (main): Use static storage for closure.
3318 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3319 * src/sparc/v8.S (ffi_closure_v8): New.
3320 * src/sparc/v9.S (ffi_closure_v9): New.
3321
acdc3fc3 33222002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3323
3324 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3325 enumeration for X86_WIN32.
3326 * src/x86/win32.S: Added ffi_call_STDCALL function
3327 definition.
3328 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3329 switch cases for recognising FFI_STDCALL and
3330 calling ffi_call_STDCALL if target is X86_WIN32.
3331 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3332 stdcall versions of the "my_strlen" and "many"
3333 test functions (for X86_WIN32).
3334 Added test cases to test stdcall invocation using
3335 these functions.
3336
805e22b2 33372002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3338
3339 * src/sh/sysv.S: Add DWARF2 unwind info.
3340
6d219176 33412002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3342
3343 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3344
4827afc9 33452002-11-26 Jim Wilson <wilson@redhat.com>
3346
3347 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3348
805e22b2 33492002-11-23 H.J. Lu <hjl@gnu.org>
3350
3351 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3352 Include ../config/accross.m4.
3353 * aclocal.m4; Rebuild.
3354 * configure: Likewise.
3355
e3a04290 33562002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3357
3358 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3359
be36f77d 33602002-11-11 DJ Delorie <dj@redhat.com>
3361
3362 * configure.in: Look for common files in the right place.
3363
dd985e54 33642002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3365
3366 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3367 raw data as _Jv_word values, not ffi_raw.
3368 (ffi_java_ptrarray_to_raw): Likewise.
3369 (ffi_java_rvalue_to_raw): New function.
3370 (ffi_java_raw_call): Call it.
3371 (ffi_java_raw_to_rvalue): New function.
3372 (ffi_java_translate_args): Call it.
3373 * src/ffitest.c (closure_test_fn): Interpret return value
3374 as ffi_arg, not int.
3375 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3376 FFI_TYPE_POINTER case.
3377 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3378 values extended to word size.
3379
4a18e5fa 33802002-10-02 Andreas Jaeger <aj@suse.de>
3381
3382 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3383
3d8ac088 33842002-10-01 Bo Thorsen <bo@smetana.suse.de>
3385
3386 * include/ffi.h.in: Fix i386 win32 compilation.
3387
e0aaecf3 33882002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3389
3390 * configure.in: Add s390x-*-linux-* target.
3391 * configure: Regenerate.
3392 * include/ffi.h.in: Define S390X for s390x targets.
3393 (FFI_CLOSURES): Define for s390/s390x.
3394 (FFI_TRAMPOLINE_SIZE): Likewise.
3395 (FFI_NATIVE_RAW_API): Likewise.
3396 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3397 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
805e22b2 3398 * src/s390/ffi.c: Major rework of existing code. Add support for
e0aaecf3 3399 s390x targets. Add closure support.
3400 * src/s390/sysv.S: Likewise.
3401
c3e9c301 34022002-09-29 Richard Earnshaw <rearnsha@arm.com>
3403
3404 * src/arm/sysv.S: Fix typo.
3405
d13886be 34062002-09-28 Richard Earnshaw <rearnsha@arm.com>
3407
3408 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3409 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3410 (ffi_call_SYSV): Handle soft-float.
3411
aaff811e 34122002-09-27 Bo Thorsen <bo@suse.de>
3413
3414 * include/ffi.h.in: Fix multilib x86-64 support.
3415
5f7912d8 34162002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3417
3418 * Makefile.am (all-multi): Fix multilib parallel build.
3419
ccda5e55 34202002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3421
3422 * configure.in (sh[34]*-*-linux*): Add brackets.
3423 * configure: Regenerate.
3424
b1e5d71d 34252002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3426
3427 * Makefile.am: Add SH support.
3428 * Makefile.in: Regenerate.
3429 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3430 * configure: Regenerate.
3431 * include/ffi.h.in: Add SH support.
3432 * src/sh/ffi.c: New file.
3433 * src/sh/sysv.S: New file.
3434 * src/types.c: Add SH support.
3435
42c71854 34362002-07-16 Bo Thorsen <bo@suse.de>
3437
3438 * src/x86/ffi64.c: New file that adds x86-64 support.
3439 * src/x86/unix64.S: New file that handles argument setup for
3440 x86-64.
3441 * src/x86/sysv.S: Don't use this on x86-64.
3442 * src/x86/ffi.c: Don't use this on x86-64.
3443 Remove unused vars.
3444 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3445 for x86-64.
3446 * src/ffitest.c (struct6): New test that tests a special case in
3447 the x86-64 ABI.
3448 (struct7): Likewise.
3449 (struct8): Likewise.
3450 (struct9): Likewise.
3451 (closure_test_fn): Silence warning about this when it's not used.
3452 (main): Add the new tests.
3453 (main): Fix a couple of wrong casts and silence some compiler warnings.
3454 * include/ffi.h.in: Add x86-64 ABI definition.
3455 * fficonfig.h.in: Regenerate.
3456 * Makefile.am: Add x86-64 support.
3457 * configure.in: Likewise.
3458 * Makefile.in: Regenerate.
3459 * configure: Likewise.
3460
34612002-06-24 Bo Thorsen <bo@suse.de>
3462
3463 * src/types.c: Merge settings for similar architectures.
3464 Add x86-64 sizes and alignments.
3465
34662002-06-23 Bo Thorsen <bo@suse.de>
3467
3468 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3469 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3470 * src/mips/ffi.c (ffi_prep_args): Likewise.
3471 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3472
72726889 34732002-07-18 H.J. Lu (hjl@gnu.org)
3474
3475 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3476 (libffi_la_SOURCES): Support MIPS_LINUX.
3477 (libffi_convenience_la_SOURCES): Likewise.
3478 * Makefile.in: Regenerated.
3479
3480 * configure.in (mips64*-*): Skip.
3481 (mips*-*-linux*): New.
3482 * configure: Regenerated.
3483
3484 * src/mips/ffi.c: Include <sgidefs.h>.
3485
e17e640b 34862002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3487
3488 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3489
b5d69d52 34902002-05-27 Roger Sayle <roger@eyesopen.com>
3491
3492 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3493
eadca1d5 34942002-05-27 Bo Thorsen <bo@suse.de>
3495
3496 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3497 fix formatting.
3498
2abdadc9 34992002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3500
3501 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3502 beginning of function (for older apple cc).
3503
0b045cf7 35042002-05-08 Alexandre Oliva <aoliva@redhat.com>
3505
3506 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3507 script entry, and set LD to it when configuring multilibs.
3508 * configure: Rebuilt.
3509
7f0e056d 35102002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3511
3512 * configure.in (sparc64-*-netbsd*): Add target.
3513 (sparc-*-netbsdelf*): Likewise.
3514 * configure: Regenerate.
3515
b19313f4 35162002-04-28 David S. Miller <davem@redhat.com>
3517
3518 * configure.in, configure: Fix SPARC test in previous change.
3519
ea1f83c5 35202002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3521
3522 * Makefile.am: Add Linux for S/390 support.
3523 * Makefile.in: Regenerate.
3524 * configure.in: Add Linux for S/390 support.
3525 * configure: Regenerate.
3526 * include/ffi.h.in: Add Linux for S/390 support.
3527 * src/s390/ffi.c: New file from libffi CVS tree.
3528 * src/s390/sysv.S: New file from libffi CVS tree.
3529
b40c762b 35302002-04-28 Jakub Jelinek <jakub@redhat.com>
3531
3532 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3533 %r_disp32().
3534 * src/sparc/v8.S: Use it.
3535 * src/sparc/v9.S: Likewise.
3536 * fficonfig.h.in: Rebuilt.
3537 * configure: Rebuilt.
3538
00f87a50 35392002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3540
3541 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3542 correctly.
3543 * src/ia64/unix.S: Add unwind information. Fix comments.
3544 Save sp in a way that's compatible with unwind info.
3545 (ffi_call_unix): Correctly restore sp in all cases.
3546 * src/ia64/ffi.c: Add, fix comments.
3547
a622ae9d 35482002-04-08 Jakub Jelinek <jakub@redhat.com>
3549
3550 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3551
c1d86bcd 35522002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3553
3554 * configure.in (alpha*-*-netbsd*): Add target.
3555 * configure: Regenerate.
3556
3faa4340 35572002-04-04 Jeff Sturm <jsturm@one-point.com>
3558
3559 * src/sparc/v8.S: Add unwind info.
3560 * src/sparc/v9.S: Likewise.
3561
ec66e65b 35622002-03-30 Krister Walfridsson <cato@df.lth.se>
3563
3564 * configure.in: Enable i*86-*-netbsdelf*.
3565 * configure: Rebuilt.
805e22b2 3566
9ee9948b 35672002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3568
3569 PR other/2620
3570 * src/mips/n32.s: Delete
3571 * src/mips/o32.s: Delete
3572
6c74b4f2 35732002-03-21 Loren J. Rittle <ljrittle@acm.org>
3574
3575 * configure.in: Enable alpha*-*-freebsd*.
3576 * configure: Rebuilt.
3577
9d0a2dd5 35782002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3579
3580 * Makefile.am: libfficonvenience -> libffi_convenience.
3581 * Makefile.in: Rebuilt.
805e22b2 3582
ee7750c2 3583 * Makefile.am: Define ffitest_OBJECTS.
3584 * Makefile.in: Rebuilt.
9d0a2dd5 3585
f6a16bd8 35862002-03-07 Andreas Tobler <toa@pop.agri.ch>
3587 David Edelsohn <edelsohn@gnu.org>
3588
3589 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3590 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3591 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3592 * Makefile.in: Regenerate.
3593 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3594 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3595 (flush_icache, flush_range): New functions.
3596 (ffi_closure_helper_DARWIN): New function.
3597 * src/powerpc/aix_closure.S: New file.
3598 * src/powerpc/darwin_closure.S: New file.
805e22b2 3599
fed8d18d 36002002-02-24 Jeff Sturm <jsturm@one-point.com>
3601
3602 * include/ffi.h.in: Add typedef for ffi_arg.
3603 * src/ffitest.c (main): Declare rint with ffi_arg.
805e22b2 3604
ddac7e16 36052002-02-21 Andreas Tobler <toa@pop.agri.ch>
3606
3607 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3608 number of GPRs for floating-point arguments.
3609
ed35b400 36102002-01-31 Anthony Green <green@redhat.com>
3611
3612 * configure: Rebuilt.
3613 * configure.in: Replace CHECK_SIZEOF and endian tests with
3614 cross-compiler friendly macros.
3615 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3616 macros.
3617
bf270768 36182002-01-18 David Edelsohn <edelsohn@gnu.org>
3619
3620 * src/powerpc/darwin.S (_ffi_call_AIX): New.
dc155f85 3621 * src/powerpc/aix.S (ffi_call_DARWIN): New.
bf270768 3622
faae36ce 36232002-01-17 David Edelsohn <edelsohn@gnu.org>
3624
3625 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3626 (TARGET_SRC_POWERPC_AIX): New.
3627 (POWERPC_AIX): New stanza.
3628 * Makefile.in: Regenerate.
3629 * configure.in: Add AIX case.
3630 * configure: Regenerate.
3631 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3632 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3633 size. Fix "long double" support.
3634 (ffi_call): Add FFI_AIX case.
3635 * src/powerpc/aix.S: New.
3636
b81d6fe1 36372001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3638
3639 Implement Darwin PowerPC ABI.
3640 * configure.in: Handle powerpc-*-darwin*.
3641 * Makefile.am: Set source files for POWERPC_DARWIN.
3642 * configure: Rebuilt.
3643 * Makefile.in: Rebuilt.
3644 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3645 POWERPC_DARWIN.
3646 * src/powerpc/darwin.S: New file.
3647 * src/powerpc/ffi_darwin.c: New file.
3648
9bccb7a3 36492001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3650
3651 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3652
ac1f371e 36532001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3654
3655 * src/x86/sysv.S: Avoid gas-only .balign directive.
3656 Use C style comments.
3657
e1e36390 36582001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3659
3660 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3661 Fixes PR bootstrap/3563.
3662
deb29e7a 36632001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3664
3665 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3666
8008de24 36672001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3668
3669 * configure.in: Recognize sparc*-sun-* host.
3670 * configure: Regenerate.
805e22b2 3671
5fc46419 36722001-06-06 Andrew Haley <aph@redhat.com>
3673
968b0d1e 3674 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
5fc46419 3675
49ff2ded 36762001-06-03 Andrew Haley <aph@redhat.com>
3677
968b0d1e 3678 * src/alpha/osf.S: Add unwind info.
3679 * src/powerpc/sysv.S: Add unwind info.
3680 * src/powerpc/ppc_closure.S: Likewise.
49ff2ded 3681
070883fe 36822000-05-31 Jeff Sturm <jsturm@one-point.com>
3683
3684 * configure.in: Fix AC_ARG_ENABLE usage.
3685 * configure: Rebuilt.
3686
09da9506 36872001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3688
3689 * configure.in: Remove warning about beta code.
3690 * configure: Rebuilt.
3691
36922001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3693
6c1b002f 3694 * src/ia64/unix.S: Restore stack pointer when returning from
3695 ffi_closure_UNIX.
3696 * src/ia64/ffi.c: Fix typo in comment.
3697
c4312026 36982001-04-18 Jim Wilson <wilson@redhat.com>
3699
3700 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3701 to eliminate RAW DV.
3702
db6ca0af 37032001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3704
3705 * Makefile.am: Make a libtool convenience library.
3706 * Makefile.in: Rebuilt.
3707
9461b350 37082001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3709
3710 * configure.in: Use different syntax for subdirectory creation.
3711 * configure: Rebuilt.
3712
a4cc90ca 37132001-03-27 Jon Beniston <jon@beniston.com>
3714
3715 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3716 * configure: Rebuilt.
3717 * Makefile.am: Added X86_WIN32 target support.
3718 * Makefile.in: Rebuilt.
3719
3720 * include/ffi.h.in: Added X86_WIN32 target support.
3721
3722 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3723 * src/types.c: Added X86_WIN32 target support.
3724
3725 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3726 stuff removed and made to work with CygWin's gas.
3727
21f4579d 37282001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3729
3730 * configure.in: Make target subdirectory in build dir.
3731 * Makefile.am: Override suffix based rules to specify correct output
3732 subdirectory.
3733 * Makefile.in: Rebuilt.
3734 * configure: Rebuilt.
3735
29183c57 37362001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
384eea1f 3737
3738 * src/powerpc/ppc_closure.S: New file.
3739 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3740 involving long long and register pairs.
3741 (ffi_prep_closure): New function.
3742 (flush_icache): Likewise.
3743 (ffi_closure_helper_SYSV): Likewise.
3744 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3745 (FFI_TRAMPOLINE_SIZE): Likewise.
3746 (FFI_NATIVE_RAW_API): Likewise.
3747 * Makefile.in: Rebuilt.
3748 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3749 (TARGET_SRC_POWERPC): Likewise.
3750
49604b5b 37512001-03-19 Tom Tromey <tromey@redhat.com>
3752
3753 * Makefile.in: Rebuilt.
3754 * Makefile.am (ffitest_LDFLAGS): New macro.
3755
c8fdcc07 37562001-03-02 Nick Clifton <nickc@redhat.com>
3757
3758 * include/ffi.h.in: Remove RCS ident string.
3759 * include/ffi_mips.h: Remove RCS ident string.
3760 * src/debug.c: Remove RCS ident string.
3761 * src/ffitest.c: Remove RCS ident string.
3762 * src/prep_cif.c: Remove RCS ident string.
3763 * src/types.c: Remove RCS ident string.
3764 * src/alpha/ffi.c: Remove RCS ident string.
3765 * src/alpha/osf.S: Remove RCS ident string.
3766 * src/arm/ffi.c: Remove RCS ident string.
3767 * src/arm/sysv.S: Remove RCS ident string.
3768 * src/mips/ffi.c: Remove RCS ident string.
3769 * src/mips/n32.S: Remove RCS ident string.
3770 * src/mips/o32.S: Remove RCS ident string.
3771 * src/sparc/ffi.c: Remove RCS ident string.
3772 * src/sparc/v8.S: Remove RCS ident string.
3773 * src/sparc/v9.S: Remove RCS ident string.
3774 * src/x86/ffi.c: Remove RCS ident string.
3775 * src/x86/sysv.S: Remove RCS ident string.
3776
8cb0f8ea 37772001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3778
3779 * include/ffi.h.in: Change sourceware.cygnus.com references to
3780 gcc.gnu.org.
3781
18f704a9 37822000-12-09 Richard Henderson <rth@redhat.com>
3783
3784 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3785 (ffi_closure_osf_inner): Index rather than increment avalue
3786 and arg_types. Give ffi_closure_osf the raw return value type.
3787 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3788 promotion.
3789
37902000-12-07 Richard Henderson <rth@redhat.com>
fdf04128 3791
3792 * src/raw_api.c (ffi_translate_args): Fix typo.
3793 (ffi_prep_closure): Likewise.
3794
3795 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3796 FFI_TRAMPOLINE_SIZE.
3797 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3798 cif->bytes for new ffi_call_osf implementation.
3799 (ffi_prep_args): Absorb into ...
3800 (ffi_call): ... here. Do all stack allocation here and
3801 avoid a callback function.
3802 (ffi_prep_closure, ffi_closure_osf_inner): New.
3803 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3804 (ffi_closure_osf): New.
3805
652f285b 38062000-09-10 Alexandre Oliva <aoliva@redhat.com>
3807
4bfa769b 3808 * config.guess, config.sub, install-sh: Removed.
3809 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3810 * Makefile.in: Rebuilt.
3811
652f285b 3812 * acinclude.m4: Include libtool macros from the top level.
3813 * aclocal.m4, configure: Rebuilt.
3814
6f52caf7 38152000-08-22 Alexandre Oliva <aoliva@redhat.com>
3816
3817 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3818 * configure: Rebuilt.
3819
d03e800c 38202000-05-11 Scott Bambrough <scottb@netwinder.org>
3821
805e22b2 3822 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3823 memory correctly. Use conditional instructions, not branches where
d03e800c 3824 possible.
3825
f3e32e0b 38262000-05-04 Tom Tromey <tromey@cygnus.com>
3827
3828 * configure: Rebuilt.
3829 * configure.in: Match `arm*-*-linux-*'.
3830 From Chris Dornan <cdornan@arm.com>.
3831
b23e0d76 38322000-04-28 Jakub Jelinek <jakub@redhat.com>
3833
3834 * Makefile.am (SUBDIRS): Define.
3835 (AM_MAKEFLAGS): Likewise.
3836 (Multilib support.): Add section.
3837 * Makefile.in: Rebuilt.
3838 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
968b0d1e 3839 New variables. Set for gcc using -print-multi-lib. Export them
3840 to libtool.
3841 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3842 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3843 for -shared links.
3844 (extra_compiler_flags_value, extra_compiler_flags): Check these
3845 for extra compiler options which need to be passed down in
3846 compiler_flags.
b23e0d76 3847
2c9b297c 38482000-04-16 Anthony Green <green@redhat.com>
3849
3850 * configure: Rebuilt.
3851 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3852
d7a35c6c 38532000-04-14 Jakub Jelinek <jakub@redhat.com>
3854
968b0d1e 3855 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
d7a35c6c 3856 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3857 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3858 Replace all void * sizeofs with sizeof(int).
3859 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3860 different than DOUBLE.
3861 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3862 (ffi_prep_args_v9): New function.
3863 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3864 (ffi_V9_return_struct): New function.
3865 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3866 32bit code (not yet cross-arch calls).
3867 * src/sparc/v8.S: Add struct return delay nop.
3868 Handle long long.
3869 * src/sparc/v9.S: New file.
3870 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3871 is used on sparc64 only for structures larger than 32 bytes.
3872 Pass by reference for structures is done for structure arguments
3873 larger than 16 bytes.
3874 * src/ffitest.c (main): Use 64bit rint on sparc64.
3875 Run long long tests on sparc.
3876 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3877 sparc64.
3878 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3879 on sparc64.
3880 * configure.in (sparc-*-linux*): New supported target.
3881 (sparc64-*-linux*): Likewise.
3882 * configure: Rebuilt.
3883 * Makefile.am: Add v9.S to SPARC files.
3884 * Makefile.in: Likewise.
3885 (LINK): Surround $(CCLD) into double quotes, so that multilib
3886 compiles work correctly.
3887
800d12c7 38882000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3889
3890 * configure: Rebuilt.
3891 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3892 proposed by Bryce McKinlay.
3893
c44e8808 38942000-03-20 Tom Tromey <tromey@cygnus.com>
3895
3896 * Makefile.in: Hand edit for java_raw_api.lo.
3897
f764cb60 38982000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3899
3900 * config.guess, config.sub: Update from the gcc tree.
3901 Fix for PR libgcj/168.
3902
3b43a26e 39032000-03-03 Tom Tromey <tromey@cygnus.com>
3904
96215ab3 3905 * Makefile.in: Fixed ia64 by hand.
3906
3b43a26e 3907 * configure: Rebuilt.
3908 * configure.in (--enable-multilib): New option.
3909 (libffi_basedir): New subst.
3910 (AC_OUTPUT): Added multilib code.
3911
03ae52bb 39122000-03-02 Tom Tromey <tromey@cygnus.com>
3913
3914 * Makefile.in: Rebuilt.
3915 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3916 directory name.
3917
28c07da5 39182000-02-25 Hans Boehm <boehm@acm.org>
3919
3920 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3921 files.
3922 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3923 list.
3924 (ffi_prep_raw_closure): Use ffi_translate_args, not
3925 ffi_closure_translate.
3926 * src/java_raw_api.c: New file.
3927 * src/ffitest.c (closure_test_fn): New function.
3928 (main): Define `rint' as long long on IA64. Added new test when
3929 FFI_CLOSURES is defined.
3930 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3931 (ffi_abi): Recognize IA64.
3932 (ffi_raw): Added `flt' field.
3933 Added "Java raw API" code.
3934 * configure.in: Recognize ia64.
3935 * Makefile.am (TARGET_SRC_IA64): New macro.
3936 (libffi_la_common_SOURCES): Added java_raw_api.c.
3937 (libffi_la_SOURCES): Define in IA64 case.
3938
0bcdf5c5 39392000-01-04 Tom Tromey <tromey@cygnus.com>
3940
3941 * Makefile.in: Rebuilt with newer automake.
3942
e9de8c8b 39431999-12-31 Tom Tromey <tromey@cygnus.com>
3944
3945 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3946
20e58c02 39471999-09-01 Tom Tromey <tromey@cygnus.com>
3948
3949 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3950 undefs.
3951 * fficonfig.h.in: Rebuilt.
3952 * configure: Rebuilt.
3953 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3954 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3955 * acconfig.h: Don't #undef PACKAGE or VERSION.
3956
47c5fab6 39571999-08-09 Anthony Green <green@cygnus.com>
3958
3959 * include/ffi.h.in: Try to work around messy header problem
3960 with PACKAGE and VERSION.
3961
3962 * configure: Rebuilt.
3963 * configure.in: Change version to 2.00-beta.
3964
968b0d1e 3965 * fficonfig.h.in: Rebuilt.
47c5fab6 3966 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3967
3968 * src/x86/ffi.c (ffi_raw_call): Rename.
3969
39701999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3971
3972 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3973 (ffi_prep_incoming_args_SYSV): Ditto.
3974 (ffi_prep_closure): Ditto.
3975 (ffi_closure_raw_SYSV): Ditto.
3976 (ffi_prep_raw_closure): More ditto.
3977 (ffi_call_raw): Final ditto.
3978
3979 * include/ffi.h.in: Add definitions for closure and raw API.
3980
3981 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
805e22b2 3982 FFI_TYPE_UINT64.
47c5fab6 3983
3984 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3985
3986 * src/raw_api.c: New file.
3987
3988 * include/ffi.h.in (ffi_raw): New type.
3989 (UINT_ARG, SINT_ARG): New defines.
3990 (ffi_closure, ffi_raw_closure): New types.
3991 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
805e22b2 3992
47c5fab6 3993 * configure.in: Add check for endianness and sizeof void*.
3994
3995 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3996 instead of directly.
3997
3998 * configure: Rebuilt.
3999
4000Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
4001
4002 * configure.in: Add x86 and powerpc BeOS configurations.
4003 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
4004
40051999-05-09 Anthony Green <green@cygnus.com>
4006
4007 * configure.in: Add warning about this being beta code.
4008 Remove src/Makefile.am from the picture.
4009 * configure: Rebuilt.
4010
4011 * Makefile.am: Move logic from src/Makefile.am. Add changes
4012 to support libffi as a target library.
4013 * Makefile.in: Rebuilt.
4014
4015 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
4016 Upgraded to new autoconf, automake, libtool.
4017
4018 * README: Tweaks.
4019
4020 * LICENSE: Update copyright date.
4021
4022 * src/Makefile.am, src/Makefile.in: Removed.
4023
40241998-11-29 Anthony Green <green@cygnus.com>
4025
4026 * include/ChangeLog: Removed.
4027 * src/ChangeLog: Removed.
4028 * src/mips/ChangeLog: Removed.
4029 * src/sparc/ChangeLog: Remboved.
4030 * src/x86/ChangeLog: Removed.
4031
4032 * ChangeLog.v1: Created.
Mirror of https://gcc.gnu.org/git/gcc.git