]> git.ipfire.org Git - thirdparty/gcc.git/blob - libffi/ChangeLog
re PR libffi/40701 (Many libffi tests fail to compile on Tru64 UNIX)
[thirdparty/gcc.git] / libffi / ChangeLog
1 2010-01-07 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
2
3 PR libffi/40701
4 * testsuite/libffi.call/ffitest.h [__alpha__ && __osf__] (PRIdLL,
5 PRIuLL, PRId64, PRIu64, PRIuPTR): Define.
6 * testsuite/libffi.call/cls_align_sint64.c: Add -Wno-format on
7 alpha*-dec-osf*.
8 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
9 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
10 * testsuite/libffi.call/return_ll1.c: Likewise.
11 * testsuite/libffi.call/stret_medium2.c: Likewise.
12 * testsuite/libffi.special/ffitestcxx.h (allocate_mmap): Cast
13 MAP_FAILED to char *.
14
15 2010-01-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
16
17 * src/mips/n32.S: Use .abicalls and .eh_frame with __GNUC__.
18
19 2009-12-31 Anthony Green <green@redhat.com>
20
21 * README: Update for libffi 3.0.9.
22
23 2009-12-27 Matthias Klose <doko@ubuntu.com>
24
25 * configure.ac (HAVE_LONG_DOUBLE): Define for mips when
26 appropriate.
27 * configure: Rebuilt.
28
29 2009-12-26 Anthony Green <green@redhat.com>
30
31 * testsuite/libffi.call/cls_longdouble_va.c: Mark as xfail for
32 avr32*-*-*.
33 * testsuite/libffi.call/cls_double_va.c: Ditto.
34
35 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
36
37 * testsuite/libffi.call/ffitest.h: Conditionally include stdint.h
38 and inttypes.h.
39 * testsuite/libffi.special/unwindtest.cc: Ditto.
40
41 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
42
43 * configure.ac: Add amd64-*-openbsd*.
44 * configure: Rebuilt.
45 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Link
46 openbsd programs with -lpthread.
47
48 2009-12-26 Anthony Green <green@redhat.com>
49
50 * testsuite/libffi.call/cls_double_va.c,
51 testsuite/libffi.call/cls_longdouble.c,
52 testsuite/libffi.call/cls_longdouble_va.c,
53 testsuite/libffi.call/cls_pointer.c,
54 testsuite/libffi.call/cls_pointer_stack.c: Remove xfail for
55 mips*-*-* and arm*-*-*.
56 * testsuite/libffi.call/cls_align_longdouble_split.c,
57 testsuite/libffi.call/cls_align_longdouble_split2.c,
58 testsuite/libffi.call/stret_medium2.c,
59 testsuite/libffi.call/stret_medium.c,
60 testsuite/libffi.call/stret_large.c,
61 testsuite/libffi.call/stret_large2.c: Remove xfail for arm*-*-*.
62
63 2009-12-31 Kay Tietz <ktietz70@googlemail.com>
64
65 * testsuite/libffi.call/ffitest.h,
66 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRuLL): Fix
67 definitions.
68
69 2009-12-31 Carlo Bramini <carlo.bramix@libero.it>
70
71 * configure.ac (AM_LTLDFLAGS): Define for windows hosts.
72 * Makefile.am (libffi_la_LDFLAGS): Add AM_LTLDFLAGS.
73 * configure: Rebuilt.
74 * Makefile.in: Rebuilt.
75
76 2009-12-31 Anthony Green <green@redhat.com>
77 Blake Chaffin.
78
79 * testsuite/libffi.call/huge_struct.c: New test case from Blake
80 Chaffin @ Apple.
81
82 2009-12-28 David Edelsohn <edelsohn@gnu.org>
83
84 * src/powerpc/ffi_darwin.c (ffi_prep_args): Copy abi and nargs to
85 local variables.
86 (aix_adjust_aggregate_sizes): New function.
87 (ffi_prep_cif_machdep): Call it.
88
89 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
90
91 * configure.ac: Define FFI_MMAP_EXEC_WRIT for the given targets.
92 * configure: Regenerate.
93 * fficonfig.h.in: Likewise.
94 * src/closures.c: Remove the FFI_MMAP_EXEC_WRIT definition for
95 Solaris/x86.
96
97 2009-12-26 Andreas Schwab <schwab@linux-m68k.org>
98
99 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Advance intarg_count
100 when a float arguments is passed in memory.
101 (ffi_closure_helper_SYSV): Mark general registers as used up when
102 a 64bit or soft-float long double argument is passed in memory.
103
104 2009-12-25 Matthias Klose <doko@ubuntu.com>
105
106 * man/ffi_call.3: Fix #include in examples.
107 * doc/libffi.texi: Add dircategory.
108
109 2009-12-25 Frank Everdij <f.p.x.everdij@tudelft.nl>
110
111 * include/ffi.h.in: Placed '__GNUC__' ifdef around
112 '__attribute__((aligned(8)))' in ffi_closure, fixes compile for
113 IRIX MIPSPro c99.
114 * include/ffi_common.h: Added '__sgi' define to non
115 '__attribute__((__mode__()))' integer typedefs.
116 * src/mips/ffi.c (ffi_call, ffi_closure_mips_inner_O32,
117 ffi_closure_mips_inner_N32): Added 'defined(_MIPSEB)' to BE check.
118 (ffi_closure_mips_inner_O32, ffi_closure_mips_inner_N32): Added
119 FFI_LONGDOUBLE support and alignment(N32 only).
120 * src/mips/ffitarget.h: Corrected '#include <sgidefs.h>' for IRIX and
121 fixed non '__attribute__((__mode__()))' integer typedefs.
122 * src/mips/n32.S: Put '#ifdef linux' around '.abicalls' and '.eh_frame'
123 since they are Linux/GNU Assembler specific.
124
125 2009-12-25 Bradley Smith <brad@brad-smith.co.uk>
126
127 * configure.ac, Makefile.am, src/avr32/ffi.c,
128 src/avr32/ffitarget.h,
129 src/avr32/sysv.S: Add AVR32 port.
130 * configure, Makefile.in: Rebuilt.
131
132 2009-12-21 Andreas Tobler <a.tobler@schweiz.org>
133
134 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
135 * configure: Regenerate.
136
137 2009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
138
139 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
140
141 2009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
142
143 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
144 type on HP-UX.
145
146 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
147
148 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
149 double' arguments.
150
151 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
152
153 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
154
155 2009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
156
157 PR libffi/40700
158 * src/closures.c [X86_64 && __sun__ && __svr4__]
159 (FFI_MMAP_EXEC_WRIT): Define.
160
161 2009-12-08 David Daney <ddaney@caviumnetworks.com>
162
163 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
164 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
165 * testsuite/libffi.call/stret_large.c: Same.
166 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
167 * testsuite/libffi.call/stret_large2.c: Same.
168 * testsuite/libffi.call/stret_medium2.c: Same.
169
170 2009-12-07 David Edelsohn <edelsohn@gnu.org>
171
172 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
173 typo.
174
175 2009-12-05 David Edelsohn <edelsohn@gnu.org>
176
177 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
178 code.
179 * src/powerpc/aix_closure.S: Same.
180
181 2009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
182
183 * Makefile.in: Regenerate.
184 * configure: Regenerate.
185 * include/Makefile.in: Regenerate.
186 * man/Makefile.in: Regenerate.
187 * testsuite/Makefile.in: Regenerate.
188
189 2009-12-04 David Edelsohn <edelsohn@gnu.org>
190
191 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
192 linux64_closure.S.
193
194 2009-12-04 Uros Bizjak <ubizjak@gmail.com>
195
196 PR libffi/41908
197 * src/x86/ffi64.c (classify_argument): Update from
198 gcc/config/i386/i386.c.
199 (ffi_closure_unix64_inner): Do not use the address of two consecutive
200 SSE registers directly.
201 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
202 for x86_64 linux targets.
203
204 2009-12-04 David Edelsohn <edelsohn@gnu.org>
205
206 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
207 pfr for long double split between fpr13 and stack.
208
209 2009-12-03 David Edelsohn <edelsohn@gnu.org>
210
211 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
212 fparg_count twice for long double.
213
214 2009-12-03 David Edelsohn <edelsohn@gnu.org>
215
216 PR libffi/42243
217 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
218
219 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
220
221 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
222 Remove xfails for x86 linux targets.
223
224 2009-12-02 David Edelsohn <edelsohn@gnu.org>
225
226 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
227 case.
228
229 2009-12-01 David Edelsohn <edelsohn@gnu.org>
230
231 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
232 register usage. Call ffi_prep_args directly. Add long double
233 return value support.
234 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
235 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
236 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
237 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
238 mode.
239 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
240 into case.
241 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
242 Allocate result area between params and FPRs.
243
244 2009-11-30 David Edelsohn <edelsohn@gnu.org>
245
246 PR target/35484
247 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
248 AIX64.
249 * src/powerpc/aix.S: Implement AIX64 version.
250 * src/powerpc/aix_closure.S: Implement AIX64 version.
251 (ffi_closure_ASM): Use extsb, lha and displament addresses.
252 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
253 support.
254 (ffi_prep_cif_machdep): Same.
255 (ffi_call): Same.
256 (ffi_closure_helper_DARWIN): Same.
257
258 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
259
260 PR libffi/41908
261 * testsuite/libffi.call/testclosure.c: New test.
262
263 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
264
265 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
266 assembly version use of ___chkstk.
267
268 2009-09-23 Matthias Klose <doko@ubuntu.com>
269
270 PR libffi/40242, PR libffi/41443
271 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
272 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
273 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
274 Change the conditionals to __SOFTFP__ || __ARM_EABI__
275 for -mfloat-abi=softfp to work.
276
277 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
278
279 PR testsuite/32843 (strikes again)
280 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
281 enable proper extension on char and short.
282
283 2009-09-15 David Daney <ddaney@caviumnetworks.com>
284
285 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
286 handling for FFI_TYPE_POINTER.
287 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
288 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
289 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
290 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
291 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
292 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
293 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
294 structure and pointer returns.
295 (ffi_closure_N32): Add handling for pointer returns.
296 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
297 calc_n32_return_struct_flags): Handle soft-float.
298 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
299 (ffi_call_N32): Declare proper argument types.
300 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
301 soft-float.
302
303 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
304
305 * configure.ac (AC_PREREQ): Bump to 2.64.
306
307 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
308
309 * Makefile.am (install-html, install-pdf): Remove.
310 * Makefile.in: Regenerate.
311
312 * Makefile.in: Regenerate.
313 * aclocal.m4: Regenerate.
314 * configure: Regenerate.
315 * fficonfig.h.in: Regenerate.
316 * include/Makefile.in: Regenerate.
317 * man/Makefile.in: Regenerate.
318 * testsuite/Makefile.in: Regenerate.
319
320 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
321
322 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
323
324 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
325
326 PR libffi/40807
327 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
328 return types for X86_WIN32.
329 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
330 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
331 _ffi_closure_STDCALL): Likewise.
332
333 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
334 (dlmmap, dlmunmap): Also use these functions on Cygwin.
335
336 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
337
338 PR testsuite/40699
339 PR testsuite/40707
340 PR testsuite/40709
341 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
342 2009-06-30 commits.
343
344 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
345
346 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
347 to "" before adding paths. (This reinstates an assignment that
348 was removed by my 2009-06-30 commit, but changes the initial
349 value from "." to "".)
350
351 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
352
353 PR testsuite/40601
354 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
355 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
356 empty.
357
358 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
359
360 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
361 to ld_library_path. Use add_path. Add just find_libgcc_s
362 to ld_library_path, not every libgcc multilib directory.
363
364 2009-06-16 Wim Lewis <wiml@hhhh.org>
365
366 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
367 supposed to be callee-saved.
368 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
369 return buffer for odd-size structs.
370
371 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
372
373 PR libffi/40444
374 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
375 allow_stack_execute for Darwin.
376
377 2009-06-16 Andrew Haley <aph@redhat.com>
378
379 * configure.ac (TARGETDIR): Add missing blank lines.
380 * configure: Regenerate.
381
382 2009-06-16 Andrew Haley <aph@redhat.com>
383
384 * testsuite/libffi.call/cls_align_sint64.c,
385 testsuite/libffi.call/cls_align_uint64.c,
386 testsuite/libffi.call/cls_longdouble_va.c,
387 testsuite/libffi.call/cls_ulonglong.c,
388 testsuite/libffi.call/return_ll1.c,
389 testsuite/libffi.call/stret_medium2.c: Fix printf format
390 specifiers.
391 * testsuite/libffi.call/ffitest.h,
392 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
393
394 2009-06-15 Andrew Haley <aph@redhat.com>
395
396 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
397 * testsuite/libffi.call/err_bad_abi.c: Likewise.
398
399 2009-06-12 Andrew Haley <aph@redhat.com>
400
401 * Makefile.am: Remove info_TEXINFOS.
402
403 2009-06-12 Andrew Haley <aph@redhat.com>
404
405 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
406 testsuite/libffi.call/cls_align_uint64.c,
407 testsuite/libffi.call/cls_ulonglong.c,
408 testsuite/libffi.call/return_ll1.c,
409 testsuite/libffi.call/stret_medium2.c: Fix printf format
410 specifiers.
411 testsuite/libffi.special/unwindtest.cc: include stdint.h.
412
413 2009-06-11 Timothy Wall <twall@users.sf.net>
414
415 * Makefile.am,
416 configure.ac,
417 include/ffi.h.in,
418 include/ffi_common.h,
419 src/closures.c,
420 src/dlmalloc.c,
421 src/x86/ffi.c,
422 src/x86/ffitarget.h,
423 src/x86/win64.S (new),
424 README: Added win64 support (mingw or MSVC)
425 * Makefile.in,
426 include/Makefile.in,
427 man/Makefile.in,
428 testsuite/Makefile.in,
429 configure,
430 aclocal.m4: Regenerated
431 * ltcf-c.sh: properly escape cygwin/w32 path
432 * man/ffi_call.3: Clarify size requirements for return value.
433 * src/x86/ffi64.c: Fix filename in comment.
434 * src/x86/win32.S: Remove unused extern.
435
436 * testsuite/libffi.call/closure_fn0.c,
437 testsuite/libffi.call/closure_fn1.c,
438 testsuite/libffi.call/closure_fn2.c,
439 testsuite/libffi.call/closure_fn3.c,
440 testsuite/libffi.call/closure_fn4.c,
441 testsuite/libffi.call/closure_fn5.c,
442 testsuite/libffi.call/closure_fn6.c,
443 testsuite/libffi.call/closure_stdcall.c,
444 testsuite/libffi.call/cls_12byte.c,
445 testsuite/libffi.call/cls_16byte.c,
446 testsuite/libffi.call/cls_18byte.c,
447 testsuite/libffi.call/cls_19byte.c,
448 testsuite/libffi.call/cls_1_1byte.c,
449 testsuite/libffi.call/cls_20byte.c,
450 testsuite/libffi.call/cls_20byte1.c,
451 testsuite/libffi.call/cls_24byte.c,
452 testsuite/libffi.call/cls_2byte.c,
453 testsuite/libffi.call/cls_3_1byte.c,
454 testsuite/libffi.call/cls_3byte1.c,
455 testsuite/libffi.call/cls_3byte2.c,
456 testsuite/libffi.call/cls_4_1byte.c,
457 testsuite/libffi.call/cls_4byte.c,
458 testsuite/libffi.call/cls_5_1_byte.c,
459 testsuite/libffi.call/cls_5byte.c,
460 testsuite/libffi.call/cls_64byte.c,
461 testsuite/libffi.call/cls_6_1_byte.c,
462 testsuite/libffi.call/cls_6byte.c,
463 testsuite/libffi.call/cls_7_1_byte.c,
464 testsuite/libffi.call/cls_7byte.c,
465 testsuite/libffi.call/cls_8byte.c,
466 testsuite/libffi.call/cls_9byte1.c,
467 testsuite/libffi.call/cls_9byte2.c,
468 testsuite/libffi.call/cls_align_double.c,
469 testsuite/libffi.call/cls_align_float.c,
470 testsuite/libffi.call/cls_align_longdouble.c,
471 testsuite/libffi.call/cls_align_longdouble_split.c,
472 testsuite/libffi.call/cls_align_longdouble_split2.c,
473 testsuite/libffi.call/cls_align_pointer.c,
474 testsuite/libffi.call/cls_align_sint16.c,
475 testsuite/libffi.call/cls_align_sint32.c,
476 testsuite/libffi.call/cls_align_sint64.c,
477 testsuite/libffi.call/cls_align_uint16.c,
478 testsuite/libffi.call/cls_align_uint32.c,
479 testsuite/libffi.call/cls_align_uint64.c,
480 testsuite/libffi.call/cls_dbls_struct.c,
481 testsuite/libffi.call/cls_double.c,
482 testsuite/libffi.call/cls_double_va.c,
483 testsuite/libffi.call/cls_float.c,
484 testsuite/libffi.call/cls_longdouble.c,
485 testsuite/libffi.call/cls_longdouble_va.c,
486 testsuite/libffi.call/cls_multi_schar.c,
487 testsuite/libffi.call/cls_multi_sshort.c,
488 testsuite/libffi.call/cls_multi_sshortchar.c,
489 testsuite/libffi.call/cls_multi_uchar.c,
490 testsuite/libffi.call/cls_multi_ushort.c,
491 testsuite/libffi.call/cls_multi_ushortchar.c,
492 testsuite/libffi.call/cls_pointer.c,
493 testsuite/libffi.call/cls_pointer_stack.c,
494 testsuite/libffi.call/cls_schar.c,
495 testsuite/libffi.call/cls_sint.c,
496 testsuite/libffi.call/cls_sshort.c,
497 testsuite/libffi.call/cls_uchar.c,
498 testsuite/libffi.call/cls_uint.c,
499 testsuite/libffi.call/cls_ulonglong.c,
500 testsuite/libffi.call/cls_ushort.c,
501 testsuite/libffi.call/err_bad_abi.c,
502 testsuite/libffi.call/err_bad_typedef.c,
503 testsuite/libffi.call/float2.c,
504 testsuite/libffi.call/huge_struct.c,
505 testsuite/libffi.call/nested_struct.c,
506 testsuite/libffi.call/nested_struct1.c,
507 testsuite/libffi.call/nested_struct10.c,
508 testsuite/libffi.call/nested_struct2.c,
509 testsuite/libffi.call/nested_struct3.c,
510 testsuite/libffi.call/nested_struct4.c,
511 testsuite/libffi.call/nested_struct5.c,
512 testsuite/libffi.call/nested_struct6.c,
513 testsuite/libffi.call/nested_struct7.c,
514 testsuite/libffi.call/nested_struct8.c,
515 testsuite/libffi.call/nested_struct9.c,
516 testsuite/libffi.call/problem1.c,
517 testsuite/libffi.call/return_ldl.c,
518 testsuite/libffi.call/return_ll1.c,
519 testsuite/libffi.call/stret_large.c,
520 testsuite/libffi.call/stret_large2.c,
521 testsuite/libffi.call/stret_medium.c,
522 testsuite/libffi.call/stret_medium2.c,
523 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
524 of checking for MMAP. Use intptr_t instead of long casts.
525
526 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
527
528 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
529 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
530 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
531
532 2009-06-09 Andrew Haley <aph@redhat.com>
533
534 * src/x86/freebsd.S: Add missing file.
535
536 2009-06-08 Andrew Haley <aph@redhat.com>
537
538 Import from libffi 3.0.8:
539
540 * doc/libffi.texi: New file.
541 * doc/libffi.info: Likewise.
542 * doc/stamp-vti: Likewise.
543 * man/Makefile.am: New file.
544 * man/ffi_call.3: New file.
545
546 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
547 src/dlmalloc.c.
548 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
549
550 * configure.ac: Bump version to 3.0.8.
551 parisc*-*-linux*: Add.
552 i386-*-freebsd* | i386-*-openbsd*: Add.
553 powerpc-*-beos*: Add.
554 AM_CONDITIONAL X86_FREEBSD: Add.
555 AC_CONFIG_FILES: Add man/Makefile.
556
557 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
558
559 2009-06-08 Andrew Haley <aph@redhat.com>
560
561 * README: Import from libffi 3.0.8.
562
563 2009-06-08 Andrew Haley <aph@redhat.com>
564
565 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
566 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
567 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
568 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
569
570 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
571 * testsuite/libffi.call/stret_medium.c: Likewise.
572 * testsuite/libffi.call/stret_large2.c: Likewise.
573 * testsuite/libffi.call/stret_large.c: Likewise.
574
575 2008-12-26 Timothy Wall <twall@users.sf.net>
576
577 * testsuite/libffi.call/cls_longdouble.c,
578 testsuite/libffi.call/cls_longdouble_va.c,
579 testsuite/libffi.call/cls_align_longdouble.c,
580 testsuite/libffi.call/cls_align_longdouble_split.c,
581 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
582 failures on x86_64 cygwin/mingw.
583
584 2008-12-22 Timothy Wall <twall@users.sf.net>
585
586 * testsuite/libffi.call/closure_fn0.c,
587 testsuite/libffi.call/closure_fn1.c,
588 testsuite/libffi.call/closure_fn2.c,
589 testsuite/libffi.call/closure_fn3.c,
590 testsuite/libffi.call/closure_fn4.c,
591 testsuite/libffi.call/closure_fn5.c,
592 testsuite/libffi.call/closure_fn6.c,
593 testsuite/libffi.call/closure_loc_fn0.c,
594 testsuite/libffi.call/closure_stdcall.c,
595 testsuite/libffi.call/cls_align_pointer.c,
596 testsuite/libffi.call/cls_pointer.c,
597 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
598 pointer to integer (intptr_t).
599 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
600
601 2008-07-24 Anthony Green <green@redhat.com>
602
603 * testsuite/libffi.call/cls_dbls_struct.c,
604 testsuite/libffi.call/cls_double_va.c,
605 testsuite/libffi.call/cls_longdouble.c,
606 testsuite/libffi.call/cls_longdouble_va.c,
607 testsuite/libffi.call/cls_pointer.c,
608 testsuite/libffi.call/cls_pointer_stack.c,
609 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
610 compiler warnings.
611
612 2008-03-04 Anthony Green <green@redhat.com>
613 Blake Chaffin
614 hos@tamanegi.org
615
616 * testsuite/libffi.call/cls_align_longdouble_split2.c
617 testsuite/libffi.call/cls_align_longdouble_split.c
618 testsuite/libffi.call/cls_dbls_struct.c
619 testsuite/libffi.call/cls_double_va.c
620 testsuite/libffi.call/cls_longdouble.c
621 testsuite/libffi.call/cls_longdouble_va.c
622 testsuite/libffi.call/cls_pointer.c
623 testsuite/libffi.call/cls_pointer_stack.c
624 testsuite/libffi.call/err_bad_abi.c
625 testsuite/libffi.call/err_bad_typedef.c
626 testsuite/libffi.call/stret_large2.c
627 testsuite/libffi.call/stret_large.c
628 testsuite/libffi.call/stret_medium2.c
629 testsuite/libffi.call/stret_medium.c: New tests from Apple.
630
631 2009-06-05 Andrew Haley <aph@redhat.com>
632
633 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
634 libffi.
635
636 2009-06-04 Andrew Haley <aph@redhat.com>
637
638 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
639 stdcall changes.
640
641 2008-02-26 Anthony Green <green@redhat.com>
642 Thomas Heller <theller@ctypes.org>
643
644 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
645 comment.
646
647 2008-02-03 Timothy Wall <twall@users.sf.net>
648
649 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
650 offset based on code pointer, not data pointer.
651
652 2008-01-31 Timothy Wall <twall@users.sf.net>
653
654 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
655 closures.
656 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
657 closures.
658 * src/x86/win32.S: Add assembly for stdcall closure.
659 * src/x86/ffi.c: Initialize stdcall closure trampoline.
660
661 2009-06-04 Andrew Haley <aph@redhat.com>
662
663 * include/ffi.h.in: Change void (*)() to void (*)(void).
664 * src/x86/ffi.c: Likewise.
665
666 2009-06-04 Andrew Haley <aph@redhat.com>
667
668 * src/powerpc/ppc_closure.S: Insert licence header.
669 * src/powerpc/linux64_closure.S: Likewise.
670 * src/m68k/sysv.S: Likewise.
671
672 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
673 * src/powerpc/ffi.c: Likewise.
674 * src/powerpc/ffi_darwin.c: Likewise.
675 * src/m32r/ffi.c: Likewise.
676 * src/sh64/ffi.c: Likewise.
677 * src/x86/ffi64.c: Likewise.
678 * src/alpha/ffi.c: Likewise.
679 * src/alpha/osf.S: Likewise.
680 * src/frv/ffi.c: Likewise.
681 * src/s390/ffi.c: Likewise.
682 * src/pa/ffi.c: Likewise.
683 * src/pa/hpux32.S: Likewise.
684 * src/ia64/unix.S: Likewise.
685 * src/ia64/ffi.c: Likewise.
686 * src/sparc/ffi.c: Likewise.
687 * src/mips/ffi.c: Likewise.
688 * src/sh/ffi.c: Likewise.
689
690 2008-02-15 David Daney <ddaney@avtrex.com>
691
692 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
693 Define (conditionally), and use it to include cachectl.h.
694 (ffi_prep_closure_loc): Fix cache flushing.
695 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
696
697 2009-06-04 Andrew Haley <aph@redhat.com>
698
699 include/ffi.h.in,
700 src/arm/ffitarget.h,
701 src/arm/ffi.c,
702 src/arm/sysv.S,
703 src/powerpc/ffitarget.h,
704 src/closures.c,
705 src/sh64/ffitarget.h,
706 src/sh64/ffi.c,
707 src/sh64/sysv.S,
708 src/types.c,
709 src/x86/ffi64.c,
710 src/x86/ffitarget.h,
711 src/x86/win32.S,
712 src/x86/darwin.S,
713 src/x86/ffi.c,
714 src/x86/sysv.S,
715 src/x86/unix64.S,
716 src/alpha/ffitarget.h,
717 src/alpha/ffi.c,
718 src/alpha/osf.S,
719 src/m68k/ffitarget.h,
720 src/frv/ffitarget.h,
721 src/frv/ffi.c,
722 src/s390/ffitarget.h,
723 src/s390/sysv.S,
724 src/cris/ffitarget.h,
725 src/pa/linux.S,
726 src/pa/ffitarget.h,
727 src/pa/ffi.c,
728 src/raw_api.c,
729 src/ia64/ffitarget.h,
730 src/ia64/unix.S,
731 src/ia64/ffi.c,
732 src/ia64/ia64_flags.h,
733 src/java_raw_api.c,
734 src/debug.c,
735 src/sparc/v9.S,
736 src/sparc/ffitarget.h,
737 src/sparc/ffi.c,
738 src/sparc/v8.S,
739 src/mips/ffitarget.h,
740 src/mips/n32.S,
741 src/mips/o32.S,
742 src/mips/ffi.c,
743 src/prep_cif.c,
744 src/sh/ffitarget.h,
745 src/sh/ffi.c,
746 src/sh/sysv.S: Update license text.
747
748 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
749
750 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
751 (.eh_frame): Add FDE for it.
752
753 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
754
755 * configure.ac: Also check if assembler supports pc-relative
756 relocs on X86_WIN32 targets.
757 * configure: Regenerate.
758 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
759 (_ffi_call_SYSV): Add missing function type symbol .def and
760 add EH markup labels.
761 (_ffi_call_STDCALL): Likewise.
762 (_ffi_closure_SYSV): Likewise.
763 (_ffi_closure_raw_SYSV): Likewise.
764 (.eh_frame): Add hand-crafted EH data.
765
766 2009-04-09 Jakub Jelinek <jakub@redhat.com>
767
768 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
769 version 3 of the GNU General Public License and to point readers
770 at the COPYING3 file and the FSF's license web page.
771 * testsuite/libffi.call/call.exp: Likewise.
772 * testsuite/libffi.special/special.exp: Likewise.
773
774 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
775
776 * configure: Regenerate.
777
778 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
779
780 PR libffi/26048
781 * configure.ac (HAVE_AS_X86_PCREL): New test.
782 * configure: Regenerate.
783 * fficonfig.h.in: Regenerate.
784 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
785 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
786 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
787 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
788 * src/x86/unix64.S (.Lstore_table): Move to .text section.
789 (.Lload_table): Likewise.
790 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
791
792 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
793
794 * configure: Regenerate.
795
796 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
797
798 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
799 signed/unsigned int8/16 return values.
800 * src/sparc/v8.S (ffi_call_v8): Likewise.
801 (ffi_closure_v8): Likewise.
802
803 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
804 Steve Ellcey <sje@cup.hp.com>
805
806 * configure: Regenerate for new libtool.
807 * Makefile.in: Ditto.
808 * include/Makefile.in: Ditto.
809 * aclocal.m4: Ditto.
810
811 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
812
813 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
814 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
815 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
816 Adjust copyright notice.
817 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
818 register or two register to use for FFI_SYSV structs.
819 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
820 (ffi_closure_helper_SYSV): Fix the return type for
821 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
822 Adjust copyright notice.
823
824 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
825
826 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
827 int.
828
829 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
830
831 * configure: Regenerate.
832 * include/Makefile.in: Regenerate.
833 * testsuite/Makefile.in: Regenerate.
834
835 2008-06-07 Joseph Myers <joseph@codesourcery.com>
836
837 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
838 powerpc-*-beos*): Remove.
839 * configure: Regenerate.
840
841 2008-05-09 Julian Brown <julian@codesourcery.com>
842
843 * Makefile.am (LTLDFLAGS): New.
844 (libffi_la_LDFLAGS): Use above.
845 * Makefile.in: Regenerate.
846
847 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
848
849 PR bootstrap/35457
850 * aclocal.m4: Regenerate.
851 * configure: Regenerate.
852
853 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
854
855 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
856 * src/sh64/sysv.S: Likewise.
857
858 2008-03-26 Daniel Jacobowitz <dan@debian.org>
859
860 * src/arm/sysv.S: Fix ARM comment marker.
861
862 2008-03-26 Jakub Jelinek <jakub@redhat.com>
863
864 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
865 * src/s390/sysv.S: Likewise.
866 * src/powerpc/ppc_closure.S: Likewise.
867 * src/powerpc/sysv.S: Likewise.
868 * src/x86/unix64.S: Likewise.
869 * src/x86/sysv.S: Likewise.
870 * src/sparc/v8.S: Likewise.
871 * src/sparc/v9.S: Likewise.
872 * src/m68k/sysv.S: Likewise.
873 * src/arm/sysv.S: Likewise.
874
875 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
876
877 * aclocal.m4: Regenerate.
878 * configure: Likewise.
879 * Makefile.in: Likewise.
880 * include/Makefile.in: Likewise.
881 * testsuite/Makefile.in: Likewise.
882
883 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
884 Andreas Tobler <a.tobler@schweiz.org>
885
886 * configure.ac: Add amd64-*-freebsd* target.
887 * configure: Regenerate.
888
889 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
890
891 PR libffi/34612
892 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
893 returning struct.
894
895 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
896 tests.
897
898 2008-01-24 David Edelsohn <edelsohn@gnu.org>
899
900 * configure: Regenerate.
901
902 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
903
904 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
905
906 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
907
908 PR testsuite/32843
909 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
910 signed/unsigned int8/16 for X86_DARWIN.
911 Updated copyright info.
912 Handle one and two byte structs with special cif->flags.
913 * src/x86/ffitarget.h: Add special types for one and two byte structs.
914 Updated copyright info.
915 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
916 sysv.S
917 Remove code to pop args from the stack after call.
918 Special-case signed/unsigned for int8/16, one and two byte structs.
919 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
920 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
921 FFI_TYPE_SINT32.
922 Updated copyright info.
923
924 2007-12-08 David Daney <ddaney@avtrex.com>
925
926 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
927 SUBU, add with ADDU and use smaller code sequences.
928
929 2007-12-07 David Daney <ddaney@avtrex.com>
930
931 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
932 type.
933
934 2007-12-06 David Daney <ddaney@avtrex.com>
935
936 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
937 defined.
938 (ffi_java_raw): New typedef.
939 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
940 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
941 ffi_java_raw.
942 (ffi_java_raw_closure) : Same.
943 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
944 parameter types.
945 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
946 FFI_SIZEOF_JAVA_RAW.
947 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
948 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
949 sizeof(ffi_java_raw) for alignment calculations.
950 (ffi_java_ptrarray_to_raw): Same.
951 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
952 if FFI_SIZEOF_JAVA_RAW == 4.
953 (ffi_java_raw_to_rvalue): Same.
954 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
955 (ffi_java_translate_args): Same.
956 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
957 parameter types.
958 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
959
960 2007-12-06 David Daney <ddaney@avtrex.com>
961
962 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
963 pointer values.
964
965 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
966
967 PR libffi/31937
968 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
969 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
970 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
971 set the NUM_FPR_ARG_REGISTERS according to.
972 Add support for potential soft-float support under hard-float
973 architecture.
974 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
975 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
976 to the FFI_LINUX_SOFT_FLOAT ABI.
977 (ffi_prep_cif_machdep): Likewise.
978 (ffi_closure_helper_SYSV): Likewise.
979 * src/powerpc/ppc_closure.S: Make sure not to store float/double
980 on archs where __NO_FPRS__ is true.
981 Add FFI_TYPE_UINT128 support.
982 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
983 Adjust copyright notice.
984
985 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
986
987 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
988 * include/ffi_common.h: ... here.
989 Update copyright.
990
991 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
992
993 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
994 * src/powerpc/linux64.S: Likewise.
995 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
996 * testsuite/libffi.call/return_ldl.c: New test.
997
998 2007-09-04 <aph@redhat.com>
999
1000 * src/arm/sysv.S (UNWIND): New.
1001 (Whole file): Conditionally compile unwinder directives.
1002 * src/arm/sysv.S: Add unwinder directives.
1003
1004 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
1005 Only treat r0 as a struct address if we're actually returning a
1006 struct by address.
1007 Only copy the bytes that are actually within a struct.
1008 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
1009 is returned in r0, not passed by address.
1010 (ffi_call): Allocate a word-sized temporary for the case where
1011 a composite is returned in r0.
1012 (ffi_prep_incoming_args_SYSV): Align as necessary.
1013
1014 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
1015
1016 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
1017 directly using the sys_cacheflush syscall.
1018
1019 2007-07-27 Andrew Haley <aph@redhat.com>
1020
1021 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
1022
1023 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
1024
1025 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
1026 * configure.ac: Likewise.
1027 * Makefile.in: Regenerate.
1028 * include/Makefile.in: Likewise.
1029 * testsuite/Makefile.in: Likewise.
1030 * configure: Likewise.
1031
1032 2007-08-24 David Daney <ddaney@avtrex.com>
1033
1034 * testsuite/libffi.call/return_sl.c: New test.
1035
1036 2007-08-10 David Daney <ddaney@avtrex.com>
1037
1038 * testsuite/libffi.call/cls_multi_ushort.c,
1039 testsuite/libffi.call/cls_align_uint16.c,
1040 testsuite/libffi.call/nested_struct1.c,
1041 testsuite/libffi.call/nested_struct3.c,
1042 testsuite/libffi.call/cls_7_1_byte.c,
1043 testsuite/libffi.call/nested_struct5.c,
1044 testsuite/libffi.call/cls_double.c,
1045 testsuite/libffi.call/nested_struct7.c,
1046 testsuite/libffi.call/cls_sint.c,
1047 testsuite/libffi.call/nested_struct9.c,
1048 testsuite/libffi.call/cls_20byte1.c,
1049 testsuite/libffi.call/cls_multi_sshortchar.c,
1050 testsuite/libffi.call/cls_align_sint64.c,
1051 testsuite/libffi.call/cls_3byte2.c,
1052 testsuite/libffi.call/cls_multi_schar.c,
1053 testsuite/libffi.call/cls_multi_uchar.c,
1054 testsuite/libffi.call/cls_19byte.c,
1055 testsuite/libffi.call/cls_9byte1.c,
1056 testsuite/libffi.call/cls_align_float.c,
1057 testsuite/libffi.call/closure_fn1.c,
1058 testsuite/libffi.call/problem1.c,
1059 testsuite/libffi.call/closure_fn3.c,
1060 testsuite/libffi.call/cls_sshort.c,
1061 testsuite/libffi.call/closure_fn5.c,
1062 testsuite/libffi.call/cls_align_double.c,
1063 testsuite/libffi.call/nested_struct.c,
1064 testsuite/libffi.call/cls_2byte.c,
1065 testsuite/libffi.call/nested_struct10.c,
1066 testsuite/libffi.call/cls_4byte.c,
1067 testsuite/libffi.call/cls_6byte.c,
1068 testsuite/libffi.call/cls_8byte.c,
1069 testsuite/libffi.call/cls_multi_sshort.c,
1070 testsuite/libffi.call/cls_align_sint16.c,
1071 testsuite/libffi.call/cls_align_uint32.c,
1072 testsuite/libffi.call/cls_20byte.c,
1073 testsuite/libffi.call/cls_float.c,
1074 testsuite/libffi.call/nested_struct2.c,
1075 testsuite/libffi.call/cls_5_1_byte.c,
1076 testsuite/libffi.call/nested_struct4.c,
1077 testsuite/libffi.call/cls_24byte.c,
1078 testsuite/libffi.call/nested_struct6.c,
1079 testsuite/libffi.call/cls_64byte.c,
1080 testsuite/libffi.call/nested_struct8.c,
1081 testsuite/libffi.call/cls_uint.c,
1082 testsuite/libffi.call/cls_multi_ushortchar.c,
1083 testsuite/libffi.call/cls_schar.c,
1084 testsuite/libffi.call/cls_uchar.c,
1085 testsuite/libffi.call/cls_align_uint64.c,
1086 testsuite/libffi.call/cls_ulonglong.c,
1087 testsuite/libffi.call/cls_align_longdouble.c,
1088 testsuite/libffi.call/cls_1_1byte.c,
1089 testsuite/libffi.call/cls_12byte.c,
1090 testsuite/libffi.call/cls_3_1byte.c,
1091 testsuite/libffi.call/cls_3byte1.c,
1092 testsuite/libffi.call/cls_4_1byte.c,
1093 testsuite/libffi.call/cls_6_1_byte.c,
1094 testsuite/libffi.call/cls_16byte.c,
1095 testsuite/libffi.call/cls_18byte.c,
1096 testsuite/libffi.call/closure_fn0.c,
1097 testsuite/libffi.call/cls_9byte2.c,
1098 testsuite/libffi.call/closure_fn2.c,
1099 testsuite/libffi.call/closure_fn4.c,
1100 testsuite/libffi.call/cls_ushort.c,
1101 testsuite/libffi.call/closure_fn6.c,
1102 testsuite/libffi.call/cls_5byte.c,
1103 testsuite/libffi.call/cls_align_pointer.c,
1104 testsuite/libffi.call/cls_7byte.c,
1105 testsuite/libffi.call/cls_align_sint32.c,
1106 testsuite/libffi.special/unwindtest_ffi_call.cc,
1107 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
1108
1109 2007-08-10 David Daney <ddaney@avtrex.com>
1110
1111 PR libffi/28313
1112 * configure.ac: Don't treat mips64 as a special case.
1113 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
1114 * configure: Regenerate
1115 * Makefile.in: Ditto.
1116 * fficonfig.h.in: Ditto.
1117 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
1118 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
1119 (FFI_DEFAULT_ABI): Set for n64 case.
1120 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
1121 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
1122 (ffi_closure_N32): New function.
1123 (.eh_frame): New section
1124 * src/mips/o32.S: Clean up comments.
1125 (ffi_closure_O32): Pass ffi_closure parameter in $12.
1126 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
1127 _MIPS_SIM == _ABIN32 throughout.
1128 (FFI_MIPS_STOP_HERE): New, use in place of
1129 ffi_stop_here.
1130 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1131 to support n32/n64 ABIs.
1132 (calc_n32_struct_flags): Rewrite.
1133 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1134 position of flag bits.
1135 (ffi_prep_cif_machdep): Rewrite n32 portion.
1136 (ffi_call): Enable for n64. Add special handling for small structure
1137 return values.
1138 (ffi_prep_closure_loc): Add n32 and n64 support.
1139 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1140 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1141
1142 2007-08-08 David Daney <ddaney@avtrex.com>
1143
1144 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1145 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1146 specifiers.
1147 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1148 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1149 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1150 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1151 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1152 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1153 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1154 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1155 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1156 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1157 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1158 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1159 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1160 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1161 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1162 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1163 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1164 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1165 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1166 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1167 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1168 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1169 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1170 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1171 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1172 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1173 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1174 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1175 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1176 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1177 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1178
1179 2007-08-07 Andrew Haley <aph@redhat.com>
1180
1181 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1182 checkin.
1183
1184 2007-08-06 Andrew Haley <aph@redhat.com>
1185
1186 PR testsuite/32843
1187 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1188 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1189 FFI_TYPE_SINT32.
1190
1191 2007-08-02 David Daney <ddaney@avtrex.com>
1192
1193 * testsuite/libffi.call/return_ul.c (main): Define return type as
1194 ffi_arg. Use proper printf conversion specifier.
1195
1196 2007-07-30 Andrew Haley <aph@redhat.com>
1197
1198 PR testsuite/32843
1199 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1200 signed/unsigned int8/16.
1201 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1202 Use a jump table.
1203 Remove code to pop args from the stack after call.
1204 Special-case signed/unsigned int8/16.
1205 * testsuite/libffi.call/return_sc.c (main): Revert.
1206
1207 2007-07-26 Richard Guenther <rguenther@suse.de>
1208
1209 PR testsuite/32843
1210 * testsuite/libffi.call/return_sc.c (main): Verify call
1211 result as signed char, not ffi_arg.
1212
1213 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1214
1215 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1216 * configure: Regenerate.
1217
1218 2007-07-11 David Daney <ddaney@avtrex.com>
1219
1220 * src/mips/ffi.c: Don't include sys/cachectl.h.
1221 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1222 cacheflush().
1223
1224 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1225
1226 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1227 from (ffi_prep_closure): ... this.
1228 (FFI_INIT_TRAMPOLINE): Adjust.
1229
1230 2005-12-31 Phil Blundell <pb@reciva.com>
1231
1232 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1233 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1234 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1235 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1236 (FFI_CLOSURES): Enable closure support.
1237
1238 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1239
1240 * testsuite/libffi.call/cls_multi_ushort.c,
1241 testsuite/libffi.call/cls_align_uint16.c,
1242 testsuite/libffi.call/nested_struct1.c,
1243 testsuite/libffi.call/nested_struct3.c,
1244 testsuite/libffi.call/cls_7_1_byte.c,
1245 testsuite/libffi.call/cls_double.c,
1246 testsuite/libffi.call/nested_struct5.c,
1247 testsuite/libffi.call/nested_struct7.c,
1248 testsuite/libffi.call/cls_sint.c,
1249 testsuite/libffi.call/nested_struct9.c,
1250 testsuite/libffi.call/cls_20byte1.c,
1251 testsuite/libffi.call/cls_multi_sshortchar.c,
1252 testsuite/libffi.call/cls_align_sint64.c,
1253 testsuite/libffi.call/cls_3byte2.c,
1254 testsuite/libffi.call/cls_multi_schar.c,
1255 testsuite/libffi.call/cls_multi_uchar.c,
1256 testsuite/libffi.call/cls_19byte.c,
1257 testsuite/libffi.call/cls_9byte1.c,
1258 testsuite/libffi.call/cls_align_float.c,
1259 testsuite/libffi.call/closure_fn1.c,
1260 testsuite/libffi.call/problem1.c,
1261 testsuite/libffi.call/closure_fn3.c,
1262 testsuite/libffi.call/cls_sshort.c,
1263 testsuite/libffi.call/closure_fn5.c,
1264 testsuite/libffi.call/cls_align_double.c,
1265 testsuite/libffi.call/cls_2byte.c,
1266 testsuite/libffi.call/nested_struct.c,
1267 testsuite/libffi.call/nested_struct10.c,
1268 testsuite/libffi.call/cls_4byte.c,
1269 testsuite/libffi.call/cls_6byte.c,
1270 testsuite/libffi.call/cls_8byte.c,
1271 testsuite/libffi.call/cls_multi_sshort.c,
1272 testsuite/libffi.call/cls_align_uint32.c,
1273 testsuite/libffi.call/cls_align_sint16.c,
1274 testsuite/libffi.call/cls_float.c,
1275 testsuite/libffi.call/cls_20byte.c,
1276 testsuite/libffi.call/cls_5_1_byte.c,
1277 testsuite/libffi.call/nested_struct2.c,
1278 testsuite/libffi.call/cls_24byte.c,
1279 testsuite/libffi.call/nested_struct4.c,
1280 testsuite/libffi.call/nested_struct6.c,
1281 testsuite/libffi.call/cls_64byte.c,
1282 testsuite/libffi.call/nested_struct8.c,
1283 testsuite/libffi.call/cls_uint.c,
1284 testsuite/libffi.call/cls_multi_ushortchar.c,
1285 testsuite/libffi.call/cls_schar.c,
1286 testsuite/libffi.call/cls_uchar.c,
1287 testsuite/libffi.call/cls_align_uint64.c,
1288 testsuite/libffi.call/cls_ulonglong.c,
1289 testsuite/libffi.call/cls_align_longdouble.c,
1290 testsuite/libffi.call/cls_1_1byte.c,
1291 testsuite/libffi.call/cls_12byte.c,
1292 testsuite/libffi.call/cls_3_1byte.c,
1293 testsuite/libffi.call/cls_3byte1.c,
1294 testsuite/libffi.call/cls_4_1byte.c,
1295 testsuite/libffi.call/cls_6_1_byte.c,
1296 testsuite/libffi.call/cls_16byte.c,
1297 testsuite/libffi.call/cls_18byte.c,
1298 testsuite/libffi.call/closure_fn0.c,
1299 testsuite/libffi.call/cls_9byte2.c,
1300 testsuite/libffi.call/closure_fn2.c,
1301 testsuite/libffi.call/closure_fn4.c,
1302 testsuite/libffi.call/cls_ushort.c,
1303 testsuite/libffi.call/closure_fn6.c,
1304 testsuite/libffi.call/cls_5byte.c,
1305 testsuite/libffi.call/cls_align_pointer.c,
1306 testsuite/libffi.call/cls_7byte.c,
1307 testsuite/libffi.call/cls_align_sint32.c,
1308 testsuite/libffi.special/unwindtest_ffi_call.cc,
1309 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1310
1311 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1312
1313 * aclocal.m4: Regenerated.
1314
1315 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1316
1317 * configure: Regenerate.
1318
1319 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1320
1321 * Makefile.in: Regenerate.
1322 * configure: Regenerate.
1323 * aclocal.m4: Regenerate.
1324 * include/Makefile.in: Regenerate.
1325 * testsuite/Makefile.in: Regenerate.
1326
1327 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1328
1329 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1330 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1331 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1332 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1333 (FFI_CLOSURES): Enable closure support.
1334
1335 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1336
1337 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1338 * configure: Regenerate.
1339 * fficonfig.h.in: Regenerate.
1340 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1341 CFI_OFFSET,CFI_DEF_CFA): New macros.
1342 (ffi_call_SYSV): Add callframe annotation.
1343
1344 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1345
1346 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1347 numerous test suite failures.
1348 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1349
1350 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1351
1352 * Makefile.am (EXTRA_DIST): Bring up to date.
1353 * Makefile.in: Regenerate.
1354 * src/frv/eabi.S: Remove RCS keyword.
1355
1356 2007-04-06 Richard Henderson <rth@redhat.com>
1357
1358 * configure.ac: Tidy target case.
1359 (HAVE_LONG_DOUBLE): Allow the target to override.
1360 * configure: Regenerate.
1361 * include/ffi.h.in: Don't define ffi_type_foo if
1362 LIBFFI_HIDE_BASIC_TYPES is defined.
1363 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1364 to ffi_type_double.
1365 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1366 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1367 (ffi_type_longdouble): Special case for Alpha. Don't define
1368 if long double == double.
1369
1370 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1371 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1372 (ffi_call, ffi_closure_osf_inner): Likewise.
1373 (ffi_closure_osf_inner): Likewise. Mark hidden.
1374 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1375 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1376 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1377 (load_table): Handle 128-bit long double.
1378
1379 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1380
1381 2007-04-06 Tom Tromey <tromey@redhat.com>
1382
1383 PR libffi/31491:
1384 * README: Fixed bug in example.
1385
1386 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1387
1388 * src/closures.c: Include sys/statfs.h.
1389 (_GNU_SOURCE): Define on Linux.
1390 (FFI_MMAP_EXEC_SELINUX): Define.
1391 (selinux_enabled): New variable.
1392 (selinux_enabled_check): New function.
1393 (is_selinux_enabled): Define.
1394 (dlmmap): Use it.
1395
1396 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1397
1398 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1399 Use 'volatile float sum' to create sum of floats to avoid false
1400 negative due to excess precision on ix86 targets.
1401 (main): Ditto.
1402
1403 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1404
1405 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1406 patch.
1407 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1408
1409 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1410
1411 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1412 (ffi_prep_closure_loc): New.
1413 (ffi_prep_raw_closure_loc): New.
1414 (ffi_prep_java_raw_closure_loc): New.
1415 * src/closures.c: New file.
1416 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1417 Replace sflags with exec_offset.
1418 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1419 sub_segment_exec_offset): New macros.
1420 (get_segment_flags, set_segment_flags, check_segment_merge): New
1421 macros.
1422 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1423 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1424 destroy_mspace): Use new macros.
1425 (sys_alloc): Silence warning.
1426 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1427 * Makefile.in: Rebuilt.
1428 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1429 terms of ffi_prep_closure_loc.
1430 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1431 from...
1432 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1433 renamed version.
1434 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1435 adjusted from...
1436 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1437 the renamed version.
1438 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1439 (ffi_prep_closure): ... this.
1440 * src/pa/ffi.c: Likewise.
1441 * src/cris/ffi.c: Likewise. Adjust.
1442 * src/frv/ffi.c: Likewise.
1443 * src/ia64/ffi.c: Likewise.
1444 * src/mips/ffi.c: Likewise.
1445 * src/powerpc/ffi_darwin.c: Likewise.
1446 * src/s390/ffi.c: Likewise.
1447 * src/sh/ffi.c: Likewise.
1448 * src/sh64/ffi.c: Likewise.
1449 * src/sparc/ffi.c: Likewise.
1450 * src/x86/ffi64.c: Likewise.
1451 * src/x86/ffi.c: Likewise.
1452 (FFI_INIT_TRAMPOLINE): Adjust.
1453 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1454 (ffi_prep_raw_closure): ... this.
1455 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1456 (ffi_prep_closure): ... this.
1457 (flush_icache): Adjust.
1458
1459 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1460
1461 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1462 Lea's malloc.
1463
1464 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1465
1466 * Makefile.am: Add dummy install-pdf target.
1467 * Makefile.in: Regenerate
1468
1469 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1470
1471 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1472 ffi_closure_helper_SYSV): Add long double handling.
1473
1474 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1475
1476 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1477 immediately after bctrl instruction.
1478
1479 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1480
1481 * Makefile.am (all-recursive, install-recursive,
1482 mostlyclean-recursive, clean-recursive, distclean-recursive,
1483 maintainer-clean-recursive): Add missing targets.
1484 * Makefile.in: Rebuilt.
1485
1486 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1487
1488 * configure.ac: Add TARGET for x86_64-*-darwin*.
1489 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1490 for X86_DARWIN.
1491 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1492 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1493 * configure: Regenerate.
1494 * Makefile.in: Regenerate.
1495 * include/Makefile.in: Regenerate.
1496 * testsuite/Makefile.in: Regenerate.
1497 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1498 ffi_call only.
1499
1500 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1501
1502 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1503 Makefile.am.
1504
1505 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1506
1507 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1508 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1509 Darwin.
1510 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1511 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1512
1513 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1514 Sandro Tolaini <tolaini@libero.it>
1515
1516 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1517 conditional.
1518 * configure: Regenerated.
1519 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1520 (EXTRA_DIST): Add src/x86/darwin.S.
1521 * Makefile.in: Regenerated.
1522 * include/Makefile.in: Regenerated.
1523 * testsuite/Makefile.in: Regenerated.
1524
1525 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1526 X86_WIN32, and additionally align stack to 16 bytes.
1527 * src/x86/darwin.S: New, based on sysv.S.
1528 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1529
1530 2006-09-12 David Daney <ddaney@avtrex.com>
1531
1532 PR libffi/23935
1533 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1534 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1535 * aclocal.m4: Regenerated for automake 1.9.6.
1536 * Makefile.in: Regenerated.
1537 * include/Makefile.in: Regenerated.
1538 * testsuite/Makefile.in: Regenerated.
1539
1540 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1541
1542 * include/ffi_common.h (struct): Revert accidental commit.
1543
1544 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1545
1546 * include/ffi_common.h: Remove lint directives.
1547 * include/ffi.h.in: Likewise.
1548
1549 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1550
1551 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1552 for 32-bit architectures.
1553 * testsuite/libffi.call/return_ul.c: New test case.
1554
1555 2006-07-19 David Daney <ddaney@avtrex.com>
1556
1557 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1558 xfail remains for mips64.
1559
1560 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1561
1562 * Makefile.am: Add install-html target. Add install-html to .PHONY
1563 * Makefile.in: Regenerate.
1564 * aclocal.m4: Regenerate.
1565 * include/Makefile.in: Regenerate.
1566 * testsuite/Makefile.in: Regenerate.
1567
1568 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1569
1570 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1571 stack slot.
1572
1573 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1574
1575 * README: Remove notice about 'Crazy Comments'.
1576 * src/debug.c: Remove lint directives. Cleanup white spaces.
1577 * src/java_raw_api.c: Likewise.
1578 * src/prep_cif.c: Likewise.
1579 * src/raw_api.c: Likewise.
1580 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1581 to the testsuite.
1582 * src/arm/ffi.c: Remove lint directives.
1583 * src/m32r/ffi.c: Likewise.
1584 * src/pa/ffi.c: Likewise.
1585 * src/powerpc/ffi.c: Likewise.
1586 * src/powerpc/ffi_darwin.c: Likewise.
1587 * src/sh/ffi.c: Likewise.
1588 * src/sh64/ffi.c: Likewise.
1589 * src/x86/ffi.c: Likewise.
1590 * testsuite/libffi.call/float2.c: Likewise.
1591 * testsuite/libffi.call/promotion.c: Likewise.
1592 * testsuite/libffi.call/struct1.c: Likewise.
1593
1594 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1595
1596 * src/pa/hpux32.S: Correct unwind offset calculation for
1597 ffi_closure_pa32.
1598 * src/pa/linux.S: Likewise.
1599
1600 2006-04-12 James E Wilson <wilson@specifix.com>
1601
1602 PR libgcj/26483
1603 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1604 (hfa_type_load): Call stf_spill.
1605 (hfa_type_store): Call ldf_fill.
1606 (ffi_call): Adjust calls to above routines. Add local temps for
1607 macro result.
1608
1609 2006-04-10 Matthias Klose <doko@debian.org>
1610
1611 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1612 directory names containing underscores.
1613
1614 2006-04-07 James E Wilson <wilson@specifix.com>
1615
1616 * testsuite/libffi.call/float4.c: New testcase.
1617
1618 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1619 Andreas Tobler <a.tobler@schweiz.ch>
1620
1621 * Makefile.am: Add PA_HPUX port.
1622 * Makefile.in: Regenerate.
1623 * include/Makefile.in: Likewise.
1624 * testsuite/Makefile.in: Likewise.
1625 * configure.ac: Add PA_HPUX rules.
1626 * configure: Regenerate.
1627 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1628 Add PA_HPUX and PA64_HPUX.
1629 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1630 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1631 (FFI_TYPE_SMALL_STRUCT2): Define.
1632 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1633 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1634 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1635 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1636 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1637 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1638 * src/pa/ffi.c (ROUND_DOWN): Delete.
1639 (fldw, fstw, fldd, fstd): Use '__asm__'.
1640 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1641 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1642 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1643 Simplify incrementing of stack slot variable. Change type of local
1644 'n' to unsigned int.
1645 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1646 double on PA_HPUX.
1647 (ffi_prep_cif_machdep): Likewise.
1648 (ffi_call): Likewise.
1649 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1650 return type to ffi_status. Simplify incrementing of stack slot
1651 variable. Only copy floating point argument registers when PA_LINUX
1652 is true. Reformat debug statement.
1653 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1654 FFI_TYPE_SMALL_STRUCT8.
1655 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1656 declaration.
1657 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1658 Add nops to cache flush. Add trampoline for PA_HPUX.
1659 * src/pa/hpux32.S: New file.
1660 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1661 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1662 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1663 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1664 argument type checks so that common argument types appear first.
1665 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1666 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1667
1668 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1669
1670 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1671 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1672 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1673 FFI_TYPE_LONGDOUBLE.
1674 (ffi_prep_args64): Assert using IBM extended double.
1675 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1676 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1677 (ffi_call): Handle FFI_LINUX.
1678 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1679 gpr3 return pointer as for struct return. Handle FFI_LINUX
1680 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1681 unnecessarily.
1682 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1683 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1684 Don't use r6 as pointer to results, instead use sp offset. Don't
1685 make a special call to load lr with case table address, instead
1686 use offset from previous call.
1687 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1688 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1689 return.
1690
1691 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1692
1693 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1694 passed with FP registers correctly.
1695 (ffi_closure_helper_SYSV): Likewise.
1696 * src/sh64/sysv.S: Likewise.
1697
1698 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1699
1700 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1701 args and userdata unused.
1702 (closure_test_fn1): Mark cif and userdata unused.
1703 (main): Remove unused res.
1704
1705 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1706
1707 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1708 -O2, -O3, -Os and the warning flags -W -Wall.
1709 * testsuite/libffi.special/special.exp: Likewise.
1710 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1711 unused parameter unused for gcc or else do nothing.
1712 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1713 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1714 and userdata unused.
1715 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1716 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1717 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1718 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1719 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1720 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1721 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1722 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1723 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1724 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1725 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1726 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1727 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1728 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1729 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1730 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1731 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1732 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1733 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1734 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1735 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1736 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1737 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1738 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1739 Likewise.
1740 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1741 Likewise.
1742 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1743 Likewise.
1744 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1745 void* to avoid compiler warning.
1746 (main): Likewise.
1747 (cls_struct_align_gn): Mark cif and userdata unused.
1748 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1749 Likewise.
1750 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1751 Likewise.
1752 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1753 Likewise.
1754 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1755 Likewise.
1756 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1757 Likewise.
1758 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1759 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1760 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1761 data unused.
1762 (main): Cast res_call to silence gcc.
1763 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1764 data unused.
1765 (main): Cast res_call to silence gcc.
1766 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1767 and data unused.
1768 (main): Cast res_call to silence gcc.
1769 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1770 data unused.
1771 (main): Cast res_call to silence gcc.
1772 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1773 data unused.
1774 (main): Cast res_call to silence gcc.
1775 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1776 and data unused.
1777 (main): Cast res_call to silence gcc.
1778 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1779 userdata unused.
1780 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1781 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1782 userdata unused.
1783 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1784 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1785 userdata unused.
1786 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1787 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1788 userdata unused.
1789 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1790 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1791 userdata unused.
1792 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1793 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1794 and userdata unused.
1795 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1796 userdata unused.
1797 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1798 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1799 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1800 Cleanup white spaces.
1801 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1802 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1803 cif and userdata unused.
1804 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1805 Likewise.
1806 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1807 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1808 formatters to silence gcc.
1809 (B_gn): Mark cif and userdata unused.
1810 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1811 unused.
1812 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1813 (B_gn): Mark cif and userdata unused.
1814 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1815 unused.
1816 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1817 (B_gn): Mark cif and userdata unused.
1818 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1819 unused.
1820 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1821 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1822 * testsuite/libffi.call/problem1.c (stub): Likewise.
1823 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1824 gcc.
1825 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1826 in the last commit for this test case in the test case itself.
1827 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1828 unused.
1829 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1830 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1831 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1832 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1833 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1834 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1835
1836 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1837
1838 * src/sh/sysv.S: Fix register numbers in the FDE for
1839 ffi_closure_SYSV.
1840
1841 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1842
1843 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1844 declaration to avoid a false negative on ix86. See PR323.
1845
1846 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1847
1848 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1849 and cast integer to void * if needed. Update the pointer to
1850 the FP register saved area correctly.
1851
1852 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1853
1854 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1855 is fixed.
1856 * testsuite/libffi.call/nested_struct4.c: Likewise.
1857
1858 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1859
1860 * testsuite/libffi.call/return_dbl.c: New test case.
1861 * testsuite/libffi.call/return_dbl1.c: Likewise.
1862 * testsuite/libffi.call/return_dbl2.c: Likewise.
1863 * testsuite/libffi.call/return_fl.c: Likewise.
1864 * testsuite/libffi.call/return_fl1.c: Likewise.
1865 * testsuite/libffi.call/return_fl2.c: Likewise.
1866 * testsuite/libffi.call/return_fl3.c: Likewise.
1867 * testsuite/libffi.call/closure_fn6.c: Likewise.
1868
1869 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1870 definition.
1871 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1872 here to be used by other test cases too.
1873
1874 * testsuite/libffi.call/nested_struct10.c: New test case.
1875 * testsuite/libffi.call/nested_struct9.c: Likewise.
1876 * testsuite/libffi.call/nested_struct8.c: Likewise.
1877 * testsuite/libffi.call/nested_struct7.c: Likewise.
1878 * testsuite/libffi.call/nested_struct6.c: Likewise.
1879 * testsuite/libffi.call/nested_struct5.c: Likewise.
1880 * testsuite/libffi.call/nested_struct4.c: Likewise.
1881
1882 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1883
1884 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1885 * configure: Rebuilt.
1886
1887 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1888
1889 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1890 instead do the shifting inline.
1891 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1892 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1893 and 8 byte structs, for the remaining struct sizes don't call
1894 __lshrdi3, instead do the shifting inline.
1895
1896 2005-12-07 Thiemo Seufer <ths@networkno.de>
1897
1898 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1899 missing parentheses.
1900 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1901 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1902 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1903 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1904 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1905 FA_0_0_OFF2.
1906 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1907 endianness bugs.
1908 (ffi_prep_closure): Improve trampoline instruction scheduling.
1909 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1910
1911 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1912
1913 * src/powerpc/ffi.c: Formatting.
1914 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1915 (ffi_prep_args64): Likewise.
1916
1917 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1918
1919 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1920 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1921
1922 2005-09-26 Tom Tromey <tromey@redhat.com>
1923
1924 * testsuite/libffi.call/float1.c (value_type): New typedef.
1925 (CANARY): New define.
1926 (main): Check for result buffer overflow.
1927 * src/powerpc/linux64.S: Handle linux64 long double returns.
1928 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1929 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1930
1931 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1932
1933 PR target/23404
1934 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1935 homed fp args.
1936 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1937
1938 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1939
1940 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1941 (AH_BOTTOM): Add FFI_HIDDEN definition.
1942 * configure: Rebuilt.
1943 * fficonfig.h.in: Rebuilt.
1944 * src/powerpc/ffi.c (hidden): Remove.
1945 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1946 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1947 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1948 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1949 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1950 add FFI_HIDDEN to its prototype.
1951 (ffi_closure_SYSV_inner): New.
1952 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1953 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1954
1955 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1956
1957 PR libffi/21819:
1958 * configure: Rebuilt.
1959 * configure.ac: Handle i*86-*-gnu*.
1960
1961 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1962
1963 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1964 DW_CFA_offset_extended_sf rather than
1965 DW_CFA_GNU_negative_offset_extended.
1966 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1967
1968 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1969
1970 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1971 on sh3.
1972 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1973 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1974 partially on register.
1975 (ffi_closure_helper_SYSV): Likewise.
1976 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1977
1978 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1979
1980 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1981 Remove empty line.
1982 * src/sh64/ffi.c (simple_type): Remove.
1983 (return_type): Handle small structures correctly.
1984 (ffi_prep_args): Likewise.
1985 (ffi_call): Likewise.
1986 (ffi_closure_helper_SYSV): Likewise.
1987 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1988 Emit position independent code if PIC and remove wrong datalabel
1989 prefixes from EH data.
1990
1991 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1992
1993 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1994 * Makefile.in: Regenerate.
1995 * include/Makefile.in: Likewise.
1996 * testsuite/Makefile.in: Likewise.
1997 * configure.ac: Add POWERPC_FREEBSD rules.
1998 * configure: Regenerate.
1999 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
2000 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
2001 * src/powerpc/ffi.c: Add flags to handle small structure returns
2002 in ffi_call_SYSV.
2003 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
2004 Aka FFI_SYSV.
2005 (ffi_closure_helper_SYSV): Likewise.
2006 * src/powerpc/ppc_closure.S: Add return types for small structures.
2007 * src/powerpc/sysv.S: Add bits to handle small structures for
2008 final SYSV 4 ABI.
2009
2010 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
2011
2012 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
2013 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
2014 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
2015
2016 2005-07-05 Randolph Chung <tausq@debian.org>
2017
2018 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
2019 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
2020 structures. Kill compilation warnings.
2021 (ffi_closure_inner_LINUX): Print return values as hex in debug
2022 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
2023 Properly handle 5-7 byte structure returns.
2024 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
2025 (FFI_TYPE_SMALL_STRUCT2): Remove.
2026 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
2027 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
2028 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
2029 (checksmst1, checksmst2): Remove.
2030 (checksmst3): Optimize handling of 3-byte struct returns.
2031 (checksmst567): Properly handle 5-7 byte struct returns.
2032
2033 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2034
2035 PR libgcj/21943
2036 * src/mips/n32.S: Enforce PIC code.
2037 * src/mips/o32.S: Likewise.
2038
2039 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2040
2041 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
2042 * configure: Regenerate.
2043
2044 2005-06-01 Alan Modra <amodra@bigpond.net.au>
2045
2046 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
2047 to call ffi_closure_helper_SYSV. Append @local instead.
2048 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
2049
2050 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
2051
2052 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
2053 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
2054 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
2055 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
2056 include/Makefile.in, testsuite/Makefile.in: Regenerate.
2057
2058 2005-05-09 Mike Stump <mrs@apple.com>
2059
2060 * configure: Regenerate.
2061
2062 2005-05-08 Richard Henderson <rth@redhat.com>
2063
2064 PR libffi/21285
2065 * src/alpha/osf.S: Update unwind into to match code.
2066
2067 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
2068 Richard Henderson <rth@redhat.com>
2069
2070 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
2071 bit 11 of flags.
2072 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
2073 (ffi_prep_closure): Set carry bit if sse-used flag set.
2074 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
2075 Only load sse registers if ssecount non-zero.
2076 (ffi_closure_unix64): Only save sse registers if carry set on entry.
2077
2078 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
2079
2080 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
2081 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
2082 * configure: Regenerate.
2083
2084 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
2085
2086 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
2087 have Tcl8.3-compatible intermediate variable.
2088
2089 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
2090 Hans-Peter Nilsson <hp@axis.com>
2091
2092 * Makefile.am: Add CRIS support.
2093 * configure.ac: Likewise.
2094 * Makefile.in, configure, testsuite/Makefile.in,
2095 include/Makefile.in: Regenerate.
2096 * src/cris: New directory.
2097 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
2098 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
2099
2100 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
2101 \r?\n in output tests.
2102
2103 2005-04-12 Mike Stump <mrs@apple.com>
2104
2105 * configure: Regenerate.
2106
2107 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
2108
2109 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
2110
2111 2005-03-30 Steve Ellcey <sje@cup.hp.com>
2112
2113 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
2114 (ffi_sarg) Ditto.
2115 * src/ia64/unix.S (ffi_closure_unix): Extend gp
2116 to 64 bits in ILP32 mode.
2117 Load 64 bits even for short data.
2118
2119 2005-03-23 Mike Stump <mrs@apple.com>
2120
2121 * src/powerpc/darwin.S: Update for -m64 multilib.
2122 * src/powerpc/darwin_closure.S: Likewise.
2123
2124 2005-03-21 Zack Weinberg <zack@codesourcery.com>
2125
2126 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
2127 Do not set tool_include_dir.
2128 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
2129 Regenerate.
2130 * include/Makefile.am: Set gcc_version and toollibffidir.
2131 * include/Makefile.in: Regenerate.
2132
2133 2005-02-22 Andrew Haley <aph@redhat.com>
2134
2135 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2136 odd-numbered register pairs for 64-bit integer types.
2137
2138 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2139
2140 PR libffi/20104
2141 * testsuite/libffi.call/return_ll1.c: New test case.
2142
2143 2005-02-11 Janis Johnson <janis187@us.ibm.com>
2144
2145 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2146 * testsuite/libffi.call/float.c: Ditto.
2147 * testsuite/libffi.call/float2.c: Ditto.
2148 * testsuite/libffi.call/float3.c: Ditto.
2149
2150 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2151
2152 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2153
2154 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2155
2156 * testsuite/libffi.special/special.exp (cxx_options): Add
2157 -shared-libgcc.
2158
2159 2004-12-31 Richard Henderson <rth@redhat.com>
2160
2161 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2162 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2163 offset parameters with a type parameter; deduce size and structure
2164 alignment. Update all users.
2165
2166 2004-12-31 Richard Henderson <rth@redhat.com>
2167
2168 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2169 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
2170 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
2171 into ffi_prep_closure.
2172 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2173 from scratch.
2174
2175 2004-12-27 Richard Henderson <rth@redhat.com>
2176
2177 * src/x86/unix64.S: Fix typo in unwind info.
2178
2179 2004-12-25 Richard Henderson <rth@redhat.com>
2180
2181 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2182 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2183 (merge_classes): Check for it.
2184 (SSE_CLASS_P): New.
2185 (classify_argument): Pass byte_offset by value; perform all updates
2186 inside struct case.
2187 (examine_argument): Add classes argument; handle
2188 X86_64_COMPLEX_X87_CLASS.
2189 (ffi_prep_args): Merge into ...
2190 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2191 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2192 (ffi_fill_return_value): Remove.
2193 (ffi_prep_closure): Remove dead assert.
2194 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2195 Rewrite to use struct register_args instead of va_list. Create
2196 flags for handling structure returns.
2197 * src/x86/unix64.S: Remove dead strings.
2198 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2199 stack frame with ffi_call. Handle structure returns properly.
2200 (float2sse, floatfloat2sse, double2sse): Remove.
2201 (sse2float, sse2double, sse2floatfloat): Remove.
2202 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2203 to handle structure returns properly.
2204
2205 2004-12-08 David Edelsohn <edelsohn@gnu.org>
2206
2207 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2208 PICFLAG.
2209 * Makefile.in: Regenerated.
2210
2211 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2212
2213 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2214 * configure, aclocal.m4, Makefile.in: Regenerate.
2215 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2216
2217 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2218
2219 * configure: Regenerate for libtool change.
2220
2221 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2222
2223 * configure: Regenerate for libtool reversion.
2224
2225 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2226
2227 * configure: Regenerate for libtool change.
2228
2229 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2230
2231 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2232
2233 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2234
2235 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2236 of jal. Use an absolute encoding for the frame information.
2237
2238 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2239
2240 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2241 * acinclude.m4: Delete logic for sincludes.
2242 * aclocal.m4, Makefile.in, configure: Regenerate.
2243 * include/Makefile: Likewise.
2244 * testsuite/Makefile: Likewise.
2245
2246 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2247
2248 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2249 on a 8-byte boundary.
2250 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2251
2252 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2253
2254 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2255 long long values. Round stack allocation to a multiple of 8 bytes
2256 for ATPCS compatibility.
2257 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2258 names. Handle returning long long types. Add Thumb and interworking
2259 support. Improve soft-float code.
2260
2261 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2262
2263 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2264 (libffi_exit): New function.
2265 (libffi_init): Build the testglue wrapper if needed.
2266
2267 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2268
2269 PR other/18138
2270 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2271
2272 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2273
2274 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2275
2276 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2277
2278 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2279 * testsuite/libffi.call/float3.c: New test case.
2280
2281 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2282
2283 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2284 the function returning a structure pointed with R2.
2285 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2286 the structure return value if T bit set. Emit position
2287 independent code and EH data if PIC.
2288
2289 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2290
2291 * Makefile.am: Add m32r support.
2292 * configure.ac: Likewise.
2293 * Makefile.in: Regenerate.
2294 * confiugre: Regenerate.
2295 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2296 (uint64, sint64, double, longdouble)
2297 * src/m32r: New directory.
2298 * src/m32r/ffi.c: New file.
2299 * src/m32r/sysv.S: Likewise.
2300 * src/m32r/ffitarget.h: Likewise.
2301
2302 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2303
2304 * testsuite/libffi.call/negint.c: New test case.
2305
2306 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2307
2308 PR libgcj/17465
2309 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2310 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2311 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2312 DYLD_LIBRARY_PATH.
2313
2314 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2315
2316 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2317 * testsuite/libffi.call/promotion.c: Likewise.
2318 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2319 whitespaces.
2320 * testsuite/libffi.call/return_sc.c: Likewise.
2321 * testsuite/libffi.call/return_uc.c: Likewise.
2322
2323 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2324
2325 * src/powerpc/darwin.S: Fix comments and identation.
2326 * src/powerpc/darwin_closure.S: Likewise.
2327
2328 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2329
2330 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2331 (ffi_prep_args): Handle longdouble arguments.
2332 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2333 longdouble.
2334 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2335 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2336 values.
2337 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2338 * src/types.c: Defined longdouble size and alignment for darwin.
2339
2340 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2341
2342 * src/powerpc/aix.S: Remove whitespaces.
2343 * src/powerpc/aix_closure.S: Likewise.
2344 * src/powerpc/asm.h: Likewise.
2345 * src/powerpc/ffi.c: Likewise.
2346 * src/powerpc/ffitarget.h: Likewise.
2347 * src/powerpc/linux64.S: Likewise.
2348 * src/powerpc/linux64_closure.S: Likewise.
2349 * src/powerpc/ppc_closure.S: Likewise.
2350 * src/powerpc/sysv.S: Likewise.
2351
2352 2004-08-30 Anthony Green <green@redhat.com>
2353
2354 * Makefile.am: Add frv support.
2355 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2356 * configure.ac: Read configure.host.
2357 * configure.in: Read configure.host.
2358 * configure.host: New file. frv-elf needs libgloss.
2359 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2360 alignment. This is needed to frv and shouldn't harm the others.
2361 * include/ffi_common.h (ALIGN_DOWN): New macro.
2362 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2363
2364 2004-08-24 David Daney <daney@avtrex.com>
2365
2366 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2367 * testsuite/libffi.call/closure_fn1.c: Likewise.
2368 * testsuite/libffi.call/closure_fn2.c Likewise.
2369 * testsuite/libffi.call/closure_fn3.c: Likewise.
2370 * testsuite/libffi.call/closure_fn4.c: Likewise.
2371 * testsuite/libffi.call/closure_fn5.c: Likewise.
2372 * testsuite/libffi.call/cls_18byte.c: Likewise.
2373 * testsuite/libffi.call/cls_19byte.c: Likewise.
2374 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2375 * testsuite/libffi.call/cls_20byte.c: Likewise.
2376 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2377 * testsuite/libffi.call/cls_24byte.c: Likewise.
2378 * testsuite/libffi.call/cls_2byte.c: Likewise.
2379 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2380 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2381 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2382 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2383 * testsuite/libffi.call/cls_4byte.c: Likewise.
2384 * testsuite/libffi.call/cls_64byte.c: Likewise.
2385 * testsuite/libffi.call/cls_6byte.c: Likewise.
2386 * testsuite/libffi.call/cls_7byte.c: Likewise.
2387 * testsuite/libffi.call/cls_8byte.c: Likewise.
2388 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2389 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2390 * testsuite/libffi.call/cls_align_double.c: Likewise.
2391 * testsuite/libffi.call/cls_align_float.c: Likewise.
2392 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2393 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2394 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2395 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2396 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2397 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2398 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2399 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2400 * testsuite/libffi.call/cls_double.c: Likewise.
2401 * testsuite/libffi.call/cls_float.c: Likewise.
2402 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2403 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2404 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2405 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2406 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2407 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2408 * testsuite/libffi.call/cls_schar.c: Likewise.
2409 * testsuite/libffi.call/cls_sint.c: Likewise.
2410 * testsuite/libffi.call/cls_sshort.c: Likewise.
2411 * testsuite/libffi.call/cls_uchar.c: Likewise.
2412 * testsuite/libffi.call/cls_uint.c: Likewise.
2413 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2414 * testsuite/libffi.call/cls_ushort.c: Likewise.
2415 * testsuite/libffi.call/nested_struct.c: Likewise.
2416 * testsuite/libffi.call/nested_struct1.c: Likewise.
2417 * testsuite/libffi.call/nested_struct2.c: Likewise.
2418 * testsuite/libffi.call/nested_struct3.c: Likewise.
2419 * testsuite/libffi.call/problem1.c: Likewise.
2420 * testsuite/libffi.special/unwindtest.cc: Likewise.
2421 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2422 to zero.
2423 * testsuite/libffi.call/cls_16byte.c: Likewise.
2424 * testsuite/libffi.call/cls_5byte.c: Likewise.
2425
2426 2004-08-23 David Daney <daney@avtrex.com>
2427
2428 PR libgcj/13141
2429 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2430 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2431 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2432 parameters and return types.
2433 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2434 (ffi_prep_closure): Ditto.
2435 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2436 alignment calculations.
2437 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2438 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2439
2440 2004-08-14 Casey Marshall <csm@gnu.org>
2441
2442 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2443 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2444 integer (O32 ABI only).
2445 (ffi_prep_closure): new function.
2446 (ffi_closure_mips_inner_O32): new function.
2447 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2448 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2449 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2450 64 bit integers correctly.
2451 (ffi_closure_O32): new function.
2452 Added DWARF-2 unwind info for both functions.
2453
2454 2004-08-10 Andrew Haley <aph@redhat.com>
2455
2456 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2457
2458 2004-08-01 Robert Millan <robertmh@gnu.org>
2459
2460 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2461 * configure: Regenerate.
2462
2463 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2464
2465 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2466 and mmap() explicitly instead of relying on preset autoconf cache
2467 variables.
2468 * aclocal.m4: Regenerate.
2469 * configure: Regenerate.
2470
2471 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2472
2473 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2474 (ffi_check_float_struct): Remove unused prototype.
2475
2476 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2477
2478 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2479 character on Darwin, use '\n\t' instead.
2480
2481 2004-06-26 Matthias Klose <doko@debian.org>
2482
2483 * libtool-version: Fix typo in revision/age.
2484
2485 2004-06-17 Matthias Klose <doko@debian.org>
2486
2487 * libtool-version: New.
2488 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2489 * Makefile.in: Regenerate.
2490
2491 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2492
2493 * Makefile.am: Remove useless multilib rules.
2494 * Makefile.in: Regenerate.
2495 * aclocal.m4: Regenerate with automake 1.8.5.
2496 * configure.ac: Remove useless multilib configury.
2497 * configure: Regenerate.
2498
2499 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2500
2501 * .cvsignore: New file.
2502
2503 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2504
2505 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2506 fp_done.
2507 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2508 changed from 8.
2509
2510 2004-06-06 Sean McNeil <sean@mcneil.com>
2511
2512 * configure.ac: Add x86_64-*-freebsd* support.
2513 * configure: Regenerate.
2514
2515 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2516
2517 Bug 15093
2518 * configure.ac: Test for existence of mmap and sys/mman.h before
2519 checking blacklist. Fix suggested by Jim Wilson.
2520 * configure: Regenerate.
2521
2522 2004-04-26 Matt Austern <austern@apple.com>
2523
2524 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2525 FDE location.
2526 * src/powerpc/darwin_closure.S: Likewise.
2527
2528 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2529
2530 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2531 error. Reported by Thomas Heller <theller@python.net>.
2532 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2533 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2534
2535 2004-03-20 Matthias Klose <doko@debian.org>
2536
2537 * src/pa/linux.S: Fix typo.
2538
2539 2004-03-19 Matthias Klose <doko@debian.org>
2540
2541 * Makefile.am: Update.
2542 * Makefile.in: Regenerate.
2543 * src/pa/ffi.h.in: Remove.
2544 * src/pa/ffitarget.h: New file.
2545
2546 2004-02-10 Randolph Chung <tausq@debian.org>
2547
2548 * Makefile.am: Add PA support.
2549 * Makefile.in: Regenerate.
2550 * include/Makefile.in: Regenerate.
2551 * configure.ac: Add PA target.
2552 * configure: Regenerate.
2553 * src/pa/ffi.c: New file.
2554 * src/pa/ffi.h.in: Add PA support.
2555 * src/pa/linux.S: New file.
2556 * prep_cif.c: Add PA support.
2557
2558 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2559
2560 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2561 double.
2562 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2563 with ecif->cif->flags.
2564 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2565 with cif->flags.
2566 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2567 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2568 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2569 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2570
2571 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2572
2573 * configure.in: Rename file to ...
2574 * configure.ac: ... this.
2575 * fficonfig.h.in: Regenerate.
2576 * Makefile.in: Regenerate.
2577 * include/Makefile.in: Regenerate.
2578 * testsuite/Makefile.in: Regenerate.
2579
2580 2004-03-12 Matt Austern <austern@apple.com>
2581
2582 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2583 changes in EH format resulting from addition of linkonce support.
2584 * src/powerpc/darwin_closure.S: Likewise.
2585
2586 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2587 Paolo Bonzini <bonzini@gnu.org>
2588
2589 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2590 Remove VPATH. Remove rules for object files. Remove multilib support.
2591 (AM_CCASFLAGS): Add.
2592 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2593 (AC_PREREQ): Bump version to 2.59.
2594 (AC_INIT): Fill with version info and bug address.
2595 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2596 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2597 De-precious CC so that the right flags are passed down to multilibs.
2598 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2599 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2600 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2601 * configure: Rebuilt.
2602 * aclocal.m4: Likewise.
2603 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2604 * fficonfig.h.in: Likewise.
2605
2606 2004-03-11 Andreas Schwab <schwab@suse.de>
2607
2608 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2609 arguments from fp registers only for the first 8 parameter slots.
2610 Don't convert a float parameter when passed in memory.
2611
2612 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2613
2614 * configure: Regenerate for config/accross.m4 correction.
2615
2616 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2617
2618 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2619 ecif->cif->bytes to bytes.
2620 (ffi_prep_cif_machdep): Add braces around nested if statement.
2621
2622 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2623
2624 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2625
2626 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2627 (ffi_closure_helper_LINUX64): Fix typo.
2628 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2629 for powerpc64-*-*.
2630 * testsuite/libffi.call/float.c: Likewise.
2631 * testsuite/libffi.call/float2.c: Likewise.
2632
2633 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2634
2635 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2636 long double function return and long double arg handling.
2637 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2638 Use "end_pfr" instead of "nf". Correct long double handling.
2639 Localise "temp".
2640 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2641 return value.
2642 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2643 space for long double return value. Adjust stack frame and offsets.
2644 Load f2 long double return.
2645
2646 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2647
2648 * src/types.c: Use 16 byte long double for POWERPC64.
2649
2650 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2651
2652 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2653 when the structure return address is passed in %o0.
2654 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2655 (ffi_v9_layout_struct): Align the field following a nested structure
2656 on a word boundary. Use memmove instead of memcpy.
2657 (ffi_call): Update call to ffi_V9_return_struct.
2658 (ffi_prep_closure): Define 'ctx' only for V8.
2659 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2660 and ffi_closure_sparc_inner_v9.
2661 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2662 Always skip the structure return address. For structures and long
2663 doubles, copy the argument directly.
2664 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2665 if required. Shift the maximum floating-point slot accordingly. For
2666 big structures, copy the argument directly; otherwise, left-justify the
2667 argument and call ffi_v9_layout_struct to lay out the structure on
2668 the stack.
2669 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2670 (ffi_closure_v8): Pass the structure return address. Update call to
2671 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2672 Skip the 'unimp' insn when returning long doubles and structures.
2673 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2674 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2675 FFI_TYPE_INT handling. Load structures both in integers and
2676 floating-point registers on return.
2677 * README: Update status of the SPARC port.
2678
2679 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2680
2681 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2682 as of type ffi_arg.
2683 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2684
2685 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2686
2687 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2688 value as of type ffi_arg, not unsigned int.
2689
2690 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2691
2692 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2693 of the LHS.
2694
2695 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2696
2697 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2698 Solaris.
2699
2700 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2701
2702 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2703 to void *.
2704
2705 2003-12-10 Richard Henderson <rth@redhat.com>
2706
2707 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2708 size_t instead of int.
2709
2710 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2711
2712 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2713 * testsuite/libffi.call/many_win32.c (main): Replace variable
2714 int i with unsigned long ul.
2715
2716 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2717 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2718 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2719 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2720 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2721 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2722 * testsuite/libffi.call/cls_align_float.c: Likewise.
2723 * testsuite/libffi.call/cls_align_double.c: Likewise.
2724 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2725 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2726
2727 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2728
2729 PR other/13221
2730 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2731 Align arguments to 32 bits.
2732
2733 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2734
2735 PR other/13221
2736 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2737 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2738 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2739 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2740 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2741 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2742
2743 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2744
2745 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2746
2747 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2748 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2749
2750 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2751
2752 * Makefile.in: Rebuilt.
2753 * configure: Likewise.
2754 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2755 the right type.
2756
2757 2003-11-21 Andreas Jaeger <aj@suse.de>
2758 Andreas Tobler <a.tobler@schweiz.ch>
2759
2760 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2761 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2762 * Makefile.in: Rebuilt.
2763 * aclocal.m4: Likewise.
2764 * configure: Likewise.
2765 * fficonfig.h.in: Likewise.
2766 * testsuite/lib/libffi-dg.exp: Add include dir.
2767 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2768 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2769 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2770 for ffi_closure if available.
2771 * testsuite/libffi.call/closure_fn1.c: Likewise.
2772 * testsuite/libffi.call/closure_fn2.c: Likewise.
2773 * testsuite/libffi.call/closure_fn3.c: Likewise.
2774 * testsuite/libffi.call/closure_fn4.c: Likewise.
2775 * testsuite/libffi.call/closure_fn5.c: Likewise.
2776 * testsuite/libffi.call/cls_12byte.c: Likewise.
2777 * testsuite/libffi.call/cls_16byte.c: Likewise.
2778 * testsuite/libffi.call/cls_18byte.c: Likewise.
2779 * testsuite/libffi.call/cls_19byte.c: Likewise.
2780 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2781 * testsuite/libffi.call/cls_20byte.c: Likewise.
2782 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2783 * testsuite/libffi.call/cls_24byte.c: Likewise.
2784 * testsuite/libffi.call/cls_2byte.c: Likewise.
2785 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2786 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2787 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2788 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2789 * testsuite/libffi.call/cls_4byte.c: Likewise.
2790 * testsuite/libffi.call/cls_5byte.c: Likewise.
2791 * testsuite/libffi.call/cls_64byte.c: Likewise.
2792 * testsuite/libffi.call/cls_6byte.c: Likewise.
2793 * testsuite/libffi.call/cls_7byte.c: Likewise.
2794 * testsuite/libffi.call/cls_8byte.c: Likewise.
2795 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2796 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2797 * testsuite/libffi.call/cls_double.c: Likewise.
2798 * testsuite/libffi.call/cls_float.c: Likewise.
2799 * testsuite/libffi.call/cls_schar.c: Likewise.
2800 * testsuite/libffi.call/cls_sint.c: Likewise.
2801 * testsuite/libffi.call/cls_sshort.c: Likewise.
2802 * testsuite/libffi.call/cls_uchar.c: Likewise.
2803 * testsuite/libffi.call/cls_uint.c: Likewise.
2804 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2805 * testsuite/libffi.call/cls_ushort.c: Likewise.
2806 * testsuite/libffi.call/nested_struct.c: Likewise.
2807 * testsuite/libffi.call/nested_struct1.c: Likewise.
2808 * testsuite/libffi.call/nested_struct2.c: Likewise.
2809 * testsuite/libffi.call/nested_struct3.c: Likewise.
2810 * testsuite/libffi.call/problem1.c: Likewise.
2811 * testsuite/libffi.special/unwindtest.cc: Likewise.
2812
2813 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2814
2815 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2816
2817 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2818
2819 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2820 Add -lgcc_s to additional flags.
2821
2822 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2823
2824 * configure.in, include/Makefile.am: PR libgcj/11147, install
2825 the ffitarget.h header file in a gcc versioned and target
2826 dependent place.
2827 * configure: Regenerated.
2828 * Makefile.in, include/Makefile.in: Likewise.
2829 * testsuite/Makefile.in: Likewise.
2830
2831 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2832
2833 * testsuite/libffi.call/closure_fn0.c: Print result and check
2834 with dg-output to make debugging easier.
2835 * testsuite/libffi.call/closure_fn1.c: Likewise.
2836 * testsuite/libffi.call/closure_fn2.c: Likewise.
2837 * testsuite/libffi.call/closure_fn3.c: Likewise.
2838 * testsuite/libffi.call/closure_fn4.c: Likewise.
2839 * testsuite/libffi.call/closure_fn5.c: Likewise.
2840 * testsuite/libffi.call/cls_12byte.c: Likewise.
2841 * testsuite/libffi.call/cls_16byte.c: Likewise.
2842 * testsuite/libffi.call/cls_18byte.c: Likewise.
2843 * testsuite/libffi.call/cls_19byte.c: Likewise.
2844 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2845 * testsuite/libffi.call/cls_20byte.c: Likewise.
2846 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2847 * testsuite/libffi.call/cls_24byte.c: Likewise.
2848 * testsuite/libffi.call/cls_2byte.c: Likewise.
2849 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2850 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2851 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2852 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2853 * testsuite/libffi.call/cls_4byte.c: Likewise.
2854 * testsuite/libffi.call/cls_5byte.c: Likewise.
2855 * testsuite/libffi.call/cls_64byte.c: Likewise.
2856 * testsuite/libffi.call/cls_6byte.c: Likewise.
2857 * testsuite/libffi.call/cls_7byte.c: Likewise.
2858 * testsuite/libffi.call/cls_8byte.c: Likewise.
2859 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2860 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2861 * testsuite/libffi.call/cls_double.c: Likewise.
2862 * testsuite/libffi.call/cls_float.c: Likewise.
2863 * testsuite/libffi.call/cls_schar.c: Likewise.
2864 * testsuite/libffi.call/cls_sint.c: Likewise.
2865 * testsuite/libffi.call/cls_sshort.c: Likewise.
2866 * testsuite/libffi.call/cls_uchar.c: Likewise.
2867 * testsuite/libffi.call/cls_uint.c: Likewise.
2868 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2869 * testsuite/libffi.call/cls_ushort.c: Likewise.
2870 * testsuite/libffi.call/problem1.c: Likewise.
2871
2872 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2873 static.
2874
2875 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2876
2877 * testsuite/libffi.call/cls_9byte2.c: New test case.
2878 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2879 * testsuite/libffi.call/cls_64byte.c: Likewise.
2880 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2881 * testsuite/libffi.call/cls_19byte.c: Likewise.
2882 * testsuite/libffi.call/cls_18byte.c: Likewise.
2883 * testsuite/libffi.call/closure_fn4.c: Likewise.
2884 * testsuite/libffi.call/closure_fn5.c: Likewise.
2885 * testsuite/libffi.call/cls_schar.c: Likewise.
2886 * testsuite/libffi.call/cls_sint.c: Likewise.
2887 * testsuite/libffi.call/cls_sshort.c: Likewise.
2888 * testsuite/libffi.call/nested_struct2.c: Likewise.
2889 * testsuite/libffi.call/nested_struct3.c: Likewise.
2890
2891 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2892
2893 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2894 * testsuite/libffi.call/cls_uchar.c: Likewise.
2895 * testsuite/libffi.call/cls_uint.c: Likewise.
2896 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2897 * testsuite/libffi.call/cls_ushort.c: Likewise.
2898 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2899
2900 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2901
2902 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2903 the initialization.
2904
2905 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2906
2907 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2908 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2909
2910 2003-10-22 David Daney <ddaney@avtrex.com>
2911
2912 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2913 __attribute__((__mode__(__SI__))) and friends.
2914
2915 2003-10-22 Andreas Schwab <schwab@suse.de>
2916
2917 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2918
2919 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2920
2921 * configure.in: AC_LINK_FILES(ffitarget.h).
2922 * configure: Regenerate.
2923 * Makefile.in: Likewise.
2924 * include/Makefile.in: Likewise.
2925 * testsuite/Makefile.in: Likewise.
2926 * fficonfig.h.in: Likewise.
2927
2928 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2929 Richard Henderson <rth@redhat.com>
2930
2931 Avoid that ffi.h includes fficonfig.h.
2932
2933 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2934 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2935 (TARGET_SRC_MIPS_SGI): Removed.
2936 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2937 (MIPS_SGI): Removed.
2938 (CLEANFILES): Removed.
2939 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2940 targets.
2941 * acconfig.h: Removed.
2942 * configure.in: Compute sizeofs only for double and long double.
2943 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2944 into AC_DEFINE instead of using acconfig.h. Create
2945 include/ffitarget.h instead of include/fficonfig.h. Rename
2946 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2947 AC_DEFINE EH_FRAME_FLAGS.
2948 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2949 (hack_DATA): Add ffitarget.h.
2950 * include/ffi.h.in: Remove all system specific definitions.
2951 Declare raw API even if it is not installed, why bother?
2952 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2953 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2954 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2955 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2956 * include/ffi_common.h (bool): Do not define.
2957 (ffi_assert): Accept failed assertion.
2958 (ffi_type_test): Return void and accept file/line.
2959 (FFI_ASSERT): Pass stringized failed assertion.
2960 (FFI_ASSERT_AT): New macro.
2961 (FFI_ASSERT_VALID_TYPE): New macro.
2962 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2963 UINT64, SINT64): Define here with gcc's __attribute__ macro
2964 instead of in ffi.h
2965 (FLOAT32, ALIGN): Define here instead of in ffi.h
2966 * include/ffi-mips.h: Removed. Its content moved to
2967 src/mips/ffitarget.h after separating assembly and C sections.
2968 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2969 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2970 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2971 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2972 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2973 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2974 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2975 (ffi_type_test): Rewritten.
2976 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2977 FFI_ASSERT_VALID_TYPE.
2978 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2979 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2980 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2981 src/s390/ffitarget.h, src/sh/ffitarget.h,
2982 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2983 src/x86/ffitarget.h: New files.
2984 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2985 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2986 src/powerpc/aix.S, src/powerpc/darwin.S,
2987 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2988 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2989 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2990 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2991 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2992 include fficonfig.h
2993
2994 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2995
2996 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2997 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2998
2999 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
3000
3001 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
3002 Used when FFI_DEBUG = 1.
3003
3004 2003-10-14 Alan Modra <amodra@bigpond.net.au>
3005
3006 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
3007 and align.
3008
3009 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3010
3011 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
3012 FFI_MIPS_O32 for O32 ABI.
3013
3014 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
3015
3016 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
3017 SPARC64. Cleanup whitespaces.
3018
3019 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
3020
3021 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
3022 strongarm, xscale. Cleanup whitespaces.
3023 * testsuite/libffi.call/closure_fn1.c: Likewise.
3024 * testsuite/libffi.call/closure_fn2.c: Likewise.
3025 * testsuite/libffi.call/closure_fn3.c: Likewise.
3026 * testsuite/libffi.call/cls_12byte.c: Likewise.
3027 * testsuite/libffi.call/cls_16byte.c: Likewise.
3028 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3029 * testsuite/libffi.call/cls_20byte.c: Likewise.
3030 * testsuite/libffi.call/cls_24byte.c: Likewise.
3031 * testsuite/libffi.call/cls_2byte.c: Likewise.
3032 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3033 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3034 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3035 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3036 * testsuite/libffi.call/cls_4byte.c: Likewise.
3037 * testsuite/libffi.call/cls_5byte.c: Likewise.
3038 * testsuite/libffi.call/cls_6byte.c: Likewise.
3039 * testsuite/libffi.call/cls_7byte.c: Likewise.
3040 * testsuite/libffi.call/cls_8byte.c: Likewise.
3041 * testsuite/libffi.call/cls_double.c: Likewise.
3042 * testsuite/libffi.call/cls_float.c: Likewise.
3043 * testsuite/libffi.call/cls_uchar.c: Likewise.
3044 * testsuite/libffi.call/cls_uint.c: Likewise.
3045 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3046 * testsuite/libffi.call/cls_ushort.c: Likewise.
3047 * testsuite/libffi.call/nested_struct.c: Likewise.
3048 * testsuite/libffi.call/nested_struct1.c: Likewise.
3049 * testsuite/libffi.call/problem1.c: Likewise.
3050 * testsuite/libffi.special/unwindtest.cc: Likewise.
3051 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
3052
3053 2003-09-18 David Edelsohn <edelsohn@gnu.org>
3054
3055 * src/powerpc/aix.S: Cleanup whitespaces.
3056 * src/powerpc/aix_closure.S: Likewise.
3057
3058 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3059
3060 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
3061 * src/powerpc/darwin_closure.S: Likewise.
3062 * src/powerpc/ffi_darwin.c: Likewise.
3063
3064 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3065 David Edelsohn <edelsohn@gnu.org>
3066
3067 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
3068 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
3069 parameter stack.
3070 * src/powerpc/darwin_closure.S: Likewise.
3071 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
3072 according to the Darwin/AIX ABI.
3073 (ffi_prep_cif_machdep): Likewise.
3074 (ffi_closure_helper_DARWIN): Likewise.
3075 Remove the outgoing parameter stack logic. Simplify the evaluation
3076 of the different CASE types.
3077 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
3078 statement in the trampoline code.
3079
3080 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
3081
3082 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
3083 for the register size.
3084 (ffi_closure_helper_SYSV): Handle the structure return value
3085 address correctly.
3086 (ffi_closure_helper_SYSV): Return the appropriate type when
3087 the registers are used for the structure return value.
3088 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
3089 the 64-bit return value. Update copyright years.
3090
3091 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3092
3093 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
3094 srcdir for ffi_mips.h.
3095
3096 2003-09-12 Alan Modra <amodra@bigpond.net.au>
3097
3098 * src/prep_cif.c (initialize_aggregate): Include tail padding in
3099 structure size.
3100 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
3101 placement of float result.
3102 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
3103 cast of "resp" for big-endian 64 bit machines.
3104
3105 2003-09-11 Alan Modra <amodra@bigpond.net.au>
3106
3107 * src/types.c (double, longdouble): Merge identical SH and ARM
3108 typedefs, and add POWERPC64.
3109 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
3110 struct split over gpr and rest.
3111 (ffi_prep_cif_machdep): Correct intarg_count for structures.
3112 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
3113
3114 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
3115
3116 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
3117 passing correctly.
3118
3119 2003-09-09 Alan Modra <amodra@bigpond.net.au>
3120
3121 * configure: Regenerate.
3122
3123 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3124
3125 * Makefile.am: Remove build rules for ffitest.
3126 * Makefile.in: Rebuilt.
3127
3128 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3129
3130 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3131 about implicit declaration of abort().
3132
3133 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3134
3135 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3136 * Makefile.in: Rebuilt.
3137 * configure.in: Add dejagnu test framework.
3138 * configure: Rebuilt.
3139
3140 * testsuite/Makefile.am: New file.
3141 * testsuite/Makefile.in: Built
3142 * testsuite/lib/libffi-dg.exp: New file.
3143 * testsuite/config/default.exp: Likewise.
3144 * testsuite/libffi.call/call.exp: Likewise.
3145 * testsuite/libffi.call/ffitest.h: Likewise.
3146 * testsuite/libffi.call/closure_fn0.c: Likewise.
3147 * testsuite/libffi.call/closure_fn1.c: Likewise.
3148 * testsuite/libffi.call/closure_fn2.c: Likewise.
3149 * testsuite/libffi.call/closure_fn3.c: Likewise.
3150 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3151 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3152 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3153 * testsuite/libffi.call/cls_2byte.c: Likewise.
3154 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3155 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3156 * testsuite/libffi.call/cls_4byte.c: Likewise.
3157 * testsuite/libffi.call/cls_5byte.c: Likewise.
3158 * testsuite/libffi.call/cls_6byte.c: Likewise.
3159 * testsuite/libffi.call/cls_7byte.c: Likewise.
3160 * testsuite/libffi.call/cls_8byte.c: Likewise.
3161 * testsuite/libffi.call/cls_12byte.c: Likewise.
3162 * testsuite/libffi.call/cls_16byte.c: Likewise.
3163 * testsuite/libffi.call/cls_20byte.c: Likewise.
3164 * testsuite/libffi.call/cls_24byte.c: Likewise.
3165 * testsuite/libffi.call/cls_double.c: Likewise.
3166 * testsuite/libffi.call/cls_float.c: Likewise.
3167 * testsuite/libffi.call/cls_uchar.c: Likewise.
3168 * testsuite/libffi.call/cls_uint.c: Likewise.
3169 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3170 * testsuite/libffi.call/cls_ushort.c: Likewise.
3171 * testsuite/libffi.call/float.c: Likewise.
3172 * testsuite/libffi.call/float1.c: Likewise.
3173 * testsuite/libffi.call/float2.c: Likewise.
3174 * testsuite/libffi.call/many.c: Likewise.
3175 * testsuite/libffi.call/many_win32.c: Likewise.
3176 * testsuite/libffi.call/nested_struct.c: Likewise.
3177 * testsuite/libffi.call/nested_struct1.c: Likewise.
3178 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3179 * testsuite/libffi.call/problem1.c: Likewise.
3180 * testsuite/libffi.call/promotion.c: Likewise.
3181 * testsuite/libffi.call/return_ll.c: Likewise.
3182 * testsuite/libffi.call/return_sc.c: Likewise.
3183 * testsuite/libffi.call/return_uc.c: Likewise.
3184 * testsuite/libffi.call/strlen.c: Likewise.
3185 * testsuite/libffi.call/strlen_win32.c: Likewise.
3186 * testsuite/libffi.call/struct1.c: Likewise.
3187 * testsuite/libffi.call/struct2.c: Likewise.
3188 * testsuite/libffi.call/struct3.c: Likewise.
3189 * testsuite/libffi.call/struct4.c: Likewise.
3190 * testsuite/libffi.call/struct5.c: Likewise.
3191 * testsuite/libffi.call/struct6.c: Likewise.
3192 * testsuite/libffi.call/struct7.c: Likewise.
3193 * testsuite/libffi.call/struct8.c: Likewise.
3194 * testsuite/libffi.call/struct9.c: Likewise.
3195 * testsuite/libffi.special/special.exp: New file.
3196 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3197 * testsuite/libffi.special/unwindtest.cc: Likewise.
3198
3199
3200 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3201
3202 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3203 copyright years.
3204
3205 2003-08-02 Alan Modra <amodra@bigpond.net.au>
3206
3207 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3208 structure passing.
3209 (ffi_closure_helper_LINUX64): Likewise.
3210 * src/powerpc/linux64.S: Remove code writing to parm save area.
3211 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3212 address in lr from ffi_closure_helper_LINUX64 call to calculate
3213 table address. Optimize function tail.
3214
3215 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3216
3217 * src/sparc/ffi.c: Handle all floating point registers.
3218 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3219
3220 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3221
3222 * README: Note that libffi is not part of GCC. Update the project
3223 URL and status.
3224
3225 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3226
3227 * src/powerpc/ppc_closure.S: Include ffi.h.
3228
3229 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3230
3231 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3232 Use C style comments.
3233
3234 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3235
3236 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3237 * Makefile.in: Regenerate.
3238 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3239 * configure: Regenerate.
3240 * include/ffi.h.in: Add SHmedia support.
3241 * src/sh64/ffi.c: New file.
3242 * src/sh64/sysv.S: New file.
3243
3244 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3245
3246 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3247 should be read-only.
3248 * configure: Rebuilt.
3249 * fficonfig.h.in: Rebuilt.
3250 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3251 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3252 * src/powerpc/linux64.S: Likewise.
3253 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3254 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3255 if -fpic/-fPIC/-mrelocatable.
3256 * src/powerpc/powerpc_closure.S: Likewise.
3257 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3258 #write in .eh_frame flags.
3259 * src/sparc/v9.S: Likewise.
3260 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3261 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3262 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3263
3264 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3265
3266 Fixes PR bootstrap/10656
3267 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3268 support for .register pseudo-op.
3269 * src/sparc/v8.S: Use it.
3270 * fficonfig.h.in: Rebuilt.
3271 * configure: Rebuilt.
3272
3273 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3274
3275 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3276 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3277 Make it the default on POWERPC64.
3278 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3279 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3280 * configure: Rebuilt.
3281 * src/powerpc/ffi.c (hidden): Define.
3282 (ffi_prep_args_SYSV): Renamed from
3283 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3284 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3285 ASM_NEEDS_REGISTERS64): New.
3286 (ffi_prep_args64): New function.
3287 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3288 (ffi_call): Likewise.
3289 (ffi_prep_closure): Likewise.
3290 (flush_icache): Surround by #ifndef POWERPC64.
3291 (ffi_dblfl): New union type.
3292 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3293 (ffi_closure_helper_LINUX64): New function.
3294 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3295 __powerpc64__.
3296 * src/powerpc/sysv.S: Likewise.
3297 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3298 * src/powerpc/linux64.S: New file.
3299 * src/powerpc/linux64_closure.S: New file.
3300 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3301 src/powerpc/linux64_closure.S.
3302 (TARGET_SRC_POWERPC): Likewise.
3303
3304 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3305 closure_test_fn3): Fix result printing on big-endian 64-bit
3306 machines.
3307 (main): Print tst2_arg instead of uninitialized tst2_result.
3308
3309 * src/ffitest.c (main): Hide what closure pointer really points to
3310 from the compiler.
3311
3312 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3313
3314 * configure.in (arm-*-netbsdelf*): Add configuration.
3315 (configure): Regenerated.
3316
3317 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3318
3319 * include/Makefile.in: Regenerate.
3320
3321 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3322
3323 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3324 bit mode.
3325 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3326 Receive closure pointer through parameter, read args using
3327 __builtin_dwarf_cfa.
3328 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3329
3330 2003-03-12 Andreas Schwab <schwab@suse.de>
3331
3332 * configure.in: Avoid trailing /. in toolexeclibdir.
3333 * configure: Rebuilt.
3334
3335 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3336
3337 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3338
3339 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3340
3341 * libffi/src/powerpc/darwin_closure.S:
3342 Fix alignement bug, allocate 8 bytes for the result.
3343 * libffi/src/powerpc/aix_closure.S:
3344 Likewise.
3345 * libffi/src/powerpc/ffi_darwin.c:
3346 Update stackframe description for aix/darwin_closure.S.
3347
3348 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3349
3350 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3351 attribute.
3352
3353 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3354 Andreas Schwab <schwab@suse.de>
3355
3356 * configure.in: Adjust command to source config-ml.in to account
3357 for changes to the libffi_basedir definition.
3358 (libffi_basedir): Remove ${srcdir} from value and include trailing
3359 slash if nonempty.
3360
3361 * configure: Regenerate.
3362
3363 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3364
3365 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3366
3367 2003-01-28 Andrew Haley <aph@redhat.com>
3368
3369 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3370 * src/x86/ffi64.c (ffi_prep_closure): New.
3371 (ffi_closure_UNIX64_inner): New.
3372 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3373
3374 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
3375
3376 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3377 Remove USE_LIBDIR conditional.
3378 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3379 * Makefile.in, configure: Rebuilt.
3380
3381 2003-01027 David Edelsohn <edelsohn@gnu.org>
3382
3383 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3384 * Makefile.in: Regenerate.
3385
3386 2003-01-22 Andrew Haley <aph@redhat.com>
3387
3388 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3389 unwind info.
3390
3391 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3392
3393 * src/powerpc/darwin.S: Add unwind info.
3394 * src/powerpc/darwin_closure.S: Likewise.
3395
3396 2003-01-14 Andrew Haley <aph@redhat.com>
3397
3398 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3399 (ffi_prep_cif_machdep): Likewise.
3400 * src/x86/unix64.S: Add unwind info.
3401
3402 2003-01-14 Andreas Jaeger <aj@suse.de>
3403
3404 * src/ffitest.c (main): Only use ffi_closures if those are
3405 supported.
3406
3407 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3408
3409 * libffi/src/ffitest.c
3410 add closure testcases
3411
3412 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3413
3414 * libffi/src/powerpc/ffi.c
3415 fix alignment bug for float (4 byte aligned iso 8 byte)
3416
3417 2003-01-09 Geoffrey Keating <geoffk@apple.com>
3418
3419 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3420 * src/powerpc/darwin.S: Remove RCS version string.
3421
3422 2003-01-03 Jeff Sturm <jsturm@one-point.com>
3423
3424 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3425 * src/ffitest.c (main): Use static storage for closure.
3426 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3427 * src/sparc/v8.S (ffi_closure_v8): New.
3428 * src/sparc/v9.S (ffi_closure_v9): New.
3429
3430 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3431
3432 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3433 enumeration for X86_WIN32.
3434 * src/x86/win32.S: Added ffi_call_STDCALL function
3435 definition.
3436 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3437 switch cases for recognising FFI_STDCALL and
3438 calling ffi_call_STDCALL if target is X86_WIN32.
3439 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3440 stdcall versions of the "my_strlen" and "many"
3441 test functions (for X86_WIN32).
3442 Added test cases to test stdcall invocation using
3443 these functions.
3444
3445 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3446
3447 * src/sh/sysv.S: Add DWARF2 unwind info.
3448
3449 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3450
3451 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3452
3453 2002-11-26 Jim Wilson <wilson@redhat.com>
3454
3455 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3456
3457 2002-11-23 H.J. Lu <hjl@gnu.org>
3458
3459 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3460 Include ../config/accross.m4.
3461 * aclocal.m4; Rebuild.
3462 * configure: Likewise.
3463
3464 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3465
3466 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3467
3468 2002-11-11 DJ Delorie <dj@redhat.com>
3469
3470 * configure.in: Look for common files in the right place.
3471
3472 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3473
3474 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3475 raw data as _Jv_word values, not ffi_raw.
3476 (ffi_java_ptrarray_to_raw): Likewise.
3477 (ffi_java_rvalue_to_raw): New function.
3478 (ffi_java_raw_call): Call it.
3479 (ffi_java_raw_to_rvalue): New function.
3480 (ffi_java_translate_args): Call it.
3481 * src/ffitest.c (closure_test_fn): Interpret return value
3482 as ffi_arg, not int.
3483 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3484 FFI_TYPE_POINTER case.
3485 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3486 values extended to word size.
3487
3488 2002-10-02 Andreas Jaeger <aj@suse.de>
3489
3490 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3491
3492 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
3493
3494 * include/ffi.h.in: Fix i386 win32 compilation.
3495
3496 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3497
3498 * configure.in: Add s390x-*-linux-* target.
3499 * configure: Regenerate.
3500 * include/ffi.h.in: Define S390X for s390x targets.
3501 (FFI_CLOSURES): Define for s390/s390x.
3502 (FFI_TRAMPOLINE_SIZE): Likewise.
3503 (FFI_NATIVE_RAW_API): Likewise.
3504 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3505 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
3506 * src/s390/ffi.c: Major rework of existing code. Add support for
3507 s390x targets. Add closure support.
3508 * src/s390/sysv.S: Likewise.
3509
3510 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
3511
3512 * src/arm/sysv.S: Fix typo.
3513
3514 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
3515
3516 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3517 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3518 (ffi_call_SYSV): Handle soft-float.
3519
3520 2002-09-27 Bo Thorsen <bo@suse.de>
3521
3522 * include/ffi.h.in: Fix multilib x86-64 support.
3523
3524 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3525
3526 * Makefile.am (all-multi): Fix multilib parallel build.
3527
3528 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3529
3530 * configure.in (sh[34]*-*-linux*): Add brackets.
3531 * configure: Regenerate.
3532
3533 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3534
3535 * Makefile.am: Add SH support.
3536 * Makefile.in: Regenerate.
3537 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3538 * configure: Regenerate.
3539 * include/ffi.h.in: Add SH support.
3540 * src/sh/ffi.c: New file.
3541 * src/sh/sysv.S: New file.
3542 * src/types.c: Add SH support.
3543
3544 2002-07-16 Bo Thorsen <bo@suse.de>
3545
3546 * src/x86/ffi64.c: New file that adds x86-64 support.
3547 * src/x86/unix64.S: New file that handles argument setup for
3548 x86-64.
3549 * src/x86/sysv.S: Don't use this on x86-64.
3550 * src/x86/ffi.c: Don't use this on x86-64.
3551 Remove unused vars.
3552 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3553 for x86-64.
3554 * src/ffitest.c (struct6): New test that tests a special case in
3555 the x86-64 ABI.
3556 (struct7): Likewise.
3557 (struct8): Likewise.
3558 (struct9): Likewise.
3559 (closure_test_fn): Silence warning about this when it's not used.
3560 (main): Add the new tests.
3561 (main): Fix a couple of wrong casts and silence some compiler warnings.
3562 * include/ffi.h.in: Add x86-64 ABI definition.
3563 * fficonfig.h.in: Regenerate.
3564 * Makefile.am: Add x86-64 support.
3565 * configure.in: Likewise.
3566 * Makefile.in: Regenerate.
3567 * configure: Likewise.
3568
3569 2002-06-24 Bo Thorsen <bo@suse.de>
3570
3571 * src/types.c: Merge settings for similar architectures.
3572 Add x86-64 sizes and alignments.
3573
3574 2002-06-23 Bo Thorsen <bo@suse.de>
3575
3576 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3577 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3578 * src/mips/ffi.c (ffi_prep_args): Likewise.
3579 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3580
3581 2002-07-18 H.J. Lu (hjl@gnu.org)
3582
3583 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3584 (libffi_la_SOURCES): Support MIPS_LINUX.
3585 (libffi_convenience_la_SOURCES): Likewise.
3586 * Makefile.in: Regenerated.
3587
3588 * configure.in (mips64*-*): Skip.
3589 (mips*-*-linux*): New.
3590 * configure: Regenerated.
3591
3592 * src/mips/ffi.c: Include <sgidefs.h>.
3593
3594 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3595
3596 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3597
3598 2002-05-27 Roger Sayle <roger@eyesopen.com>
3599
3600 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3601
3602 2002-05-27 Bo Thorsen <bo@suse.de>
3603
3604 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3605 fix formatting.
3606
3607 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3608
3609 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3610 beginning of function (for older apple cc).
3611
3612 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3613
3614 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3615 script entry, and set LD to it when configuring multilibs.
3616 * configure: Rebuilt.
3617
3618 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3619
3620 * configure.in (sparc64-*-netbsd*): Add target.
3621 (sparc-*-netbsdelf*): Likewise.
3622 * configure: Regenerate.
3623
3624 2002-04-28 David S. Miller <davem@redhat.com>
3625
3626 * configure.in, configure: Fix SPARC test in previous change.
3627
3628 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3629
3630 * Makefile.am: Add Linux for S/390 support.
3631 * Makefile.in: Regenerate.
3632 * configure.in: Add Linux for S/390 support.
3633 * configure: Regenerate.
3634 * include/ffi.h.in: Add Linux for S/390 support.
3635 * src/s390/ffi.c: New file from libffi CVS tree.
3636 * src/s390/sysv.S: New file from libffi CVS tree.
3637
3638 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3639
3640 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3641 %r_disp32().
3642 * src/sparc/v8.S: Use it.
3643 * src/sparc/v9.S: Likewise.
3644 * fficonfig.h.in: Rebuilt.
3645 * configure: Rebuilt.
3646
3647 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3648
3649 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3650 correctly.
3651 * src/ia64/unix.S: Add unwind information. Fix comments.
3652 Save sp in a way that's compatible with unwind info.
3653 (ffi_call_unix): Correctly restore sp in all cases.
3654 * src/ia64/ffi.c: Add, fix comments.
3655
3656 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3657
3658 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3659
3660 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3661
3662 * configure.in (alpha*-*-netbsd*): Add target.
3663 * configure: Regenerate.
3664
3665 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3666
3667 * src/sparc/v8.S: Add unwind info.
3668 * src/sparc/v9.S: Likewise.
3669
3670 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3671
3672 * configure.in: Enable i*86-*-netbsdelf*.
3673 * configure: Rebuilt.
3674
3675 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3676
3677 PR other/2620
3678 * src/mips/n32.s: Delete
3679 * src/mips/o32.s: Delete
3680
3681 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3682
3683 * configure.in: Enable alpha*-*-freebsd*.
3684 * configure: Rebuilt.
3685
3686 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3687
3688 * Makefile.am: libfficonvenience -> libffi_convenience.
3689 * Makefile.in: Rebuilt.
3690
3691 * Makefile.am: Define ffitest_OBJECTS.
3692 * Makefile.in: Rebuilt.
3693
3694 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3695 David Edelsohn <edelsohn@gnu.org>
3696
3697 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3698 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3699 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3700 * Makefile.in: Regenerate.
3701 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3702 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3703 (flush_icache, flush_range): New functions.
3704 (ffi_closure_helper_DARWIN): New function.
3705 * src/powerpc/aix_closure.S: New file.
3706 * src/powerpc/darwin_closure.S: New file.
3707
3708 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3709
3710 * include/ffi.h.in: Add typedef for ffi_arg.
3711 * src/ffitest.c (main): Declare rint with ffi_arg.
3712
3713 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3714
3715 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3716 number of GPRs for floating-point arguments.
3717
3718 2002-01-31 Anthony Green <green@redhat.com>
3719
3720 * configure: Rebuilt.
3721 * configure.in: Replace CHECK_SIZEOF and endian tests with
3722 cross-compiler friendly macros.
3723 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3724 macros.
3725
3726 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3727
3728 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3729 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3730
3731 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3732
3733 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3734 (TARGET_SRC_POWERPC_AIX): New.
3735 (POWERPC_AIX): New stanza.
3736 * Makefile.in: Regenerate.
3737 * configure.in: Add AIX case.
3738 * configure: Regenerate.
3739 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3740 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3741 size. Fix "long double" support.
3742 (ffi_call): Add FFI_AIX case.
3743 * src/powerpc/aix.S: New.
3744
3745 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3746
3747 Implement Darwin PowerPC ABI.
3748 * configure.in: Handle powerpc-*-darwin*.
3749 * Makefile.am: Set source files for POWERPC_DARWIN.
3750 * configure: Rebuilt.
3751 * Makefile.in: Rebuilt.
3752 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3753 POWERPC_DARWIN.
3754 * src/powerpc/darwin.S: New file.
3755 * src/powerpc/ffi_darwin.c: New file.
3756
3757 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3758
3759 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3760
3761 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3762
3763 * src/x86/sysv.S: Avoid gas-only .balign directive.
3764 Use C style comments.
3765
3766 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3767
3768 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3769 Fixes PR bootstrap/3563.
3770
3771 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3772
3773 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3774
3775 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3776
3777 * configure.in: Recognize sparc*-sun-* host.
3778 * configure: Regenerate.
3779
3780 2001-06-06 Andrew Haley <aph@redhat.com>
3781
3782 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
3783
3784 2001-06-03 Andrew Haley <aph@redhat.com>
3785
3786 * src/alpha/osf.S: Add unwind info.
3787 * src/powerpc/sysv.S: Add unwind info.
3788 * src/powerpc/ppc_closure.S: Likewise.
3789
3790 2000-05-31 Jeff Sturm <jsturm@one-point.com>
3791
3792 * configure.in: Fix AC_ARG_ENABLE usage.
3793 * configure: Rebuilt.
3794
3795 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3796
3797 * configure.in: Remove warning about beta code.
3798 * configure: Rebuilt.
3799
3800 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3801
3802 * src/ia64/unix.S: Restore stack pointer when returning from
3803 ffi_closure_UNIX.
3804 * src/ia64/ffi.c: Fix typo in comment.
3805
3806 2001-04-18 Jim Wilson <wilson@redhat.com>
3807
3808 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3809 to eliminate RAW DV.
3810
3811 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3812
3813 * Makefile.am: Make a libtool convenience library.
3814 * Makefile.in: Rebuilt.
3815
3816 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3817
3818 * configure.in: Use different syntax for subdirectory creation.
3819 * configure: Rebuilt.
3820
3821 2001-03-27 Jon Beniston <jon@beniston.com>
3822
3823 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3824 * configure: Rebuilt.
3825 * Makefile.am: Added X86_WIN32 target support.
3826 * Makefile.in: Rebuilt.
3827
3828 * include/ffi.h.in: Added X86_WIN32 target support.
3829
3830 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3831 * src/types.c: Added X86_WIN32 target support.
3832
3833 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3834 stuff removed and made to work with CygWin's gas.
3835
3836 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3837
3838 * configure.in: Make target subdirectory in build dir.
3839 * Makefile.am: Override suffix based rules to specify correct output
3840 subdirectory.
3841 * Makefile.in: Rebuilt.
3842 * configure: Rebuilt.
3843
3844 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3845
3846 * src/powerpc/ppc_closure.S: New file.
3847 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3848 involving long long and register pairs.
3849 (ffi_prep_closure): New function.
3850 (flush_icache): Likewise.
3851 (ffi_closure_helper_SYSV): Likewise.
3852 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3853 (FFI_TRAMPOLINE_SIZE): Likewise.
3854 (FFI_NATIVE_RAW_API): Likewise.
3855 * Makefile.in: Rebuilt.
3856 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3857 (TARGET_SRC_POWERPC): Likewise.
3858
3859 2001-03-19 Tom Tromey <tromey@redhat.com>
3860
3861 * Makefile.in: Rebuilt.
3862 * Makefile.am (ffitest_LDFLAGS): New macro.
3863
3864 2001-03-02 Nick Clifton <nickc@redhat.com>
3865
3866 * include/ffi.h.in: Remove RCS ident string.
3867 * include/ffi_mips.h: Remove RCS ident string.
3868 * src/debug.c: Remove RCS ident string.
3869 * src/ffitest.c: Remove RCS ident string.
3870 * src/prep_cif.c: Remove RCS ident string.
3871 * src/types.c: Remove RCS ident string.
3872 * src/alpha/ffi.c: Remove RCS ident string.
3873 * src/alpha/osf.S: Remove RCS ident string.
3874 * src/arm/ffi.c: Remove RCS ident string.
3875 * src/arm/sysv.S: Remove RCS ident string.
3876 * src/mips/ffi.c: Remove RCS ident string.
3877 * src/mips/n32.S: Remove RCS ident string.
3878 * src/mips/o32.S: Remove RCS ident string.
3879 * src/sparc/ffi.c: Remove RCS ident string.
3880 * src/sparc/v8.S: Remove RCS ident string.
3881 * src/sparc/v9.S: Remove RCS ident string.
3882 * src/x86/ffi.c: Remove RCS ident string.
3883 * src/x86/sysv.S: Remove RCS ident string.
3884
3885 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3886
3887 * include/ffi.h.in: Change sourceware.cygnus.com references to
3888 gcc.gnu.org.
3889
3890 2000-12-09 Richard Henderson <rth@redhat.com>
3891
3892 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3893 (ffi_closure_osf_inner): Index rather than increment avalue
3894 and arg_types. Give ffi_closure_osf the raw return value type.
3895 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3896 promotion.
3897
3898 2000-12-07 Richard Henderson <rth@redhat.com>
3899
3900 * src/raw_api.c (ffi_translate_args): Fix typo.
3901 (ffi_prep_closure): Likewise.
3902
3903 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3904 FFI_TRAMPOLINE_SIZE.
3905 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3906 cif->bytes for new ffi_call_osf implementation.
3907 (ffi_prep_args): Absorb into ...
3908 (ffi_call): ... here. Do all stack allocation here and
3909 avoid a callback function.
3910 (ffi_prep_closure, ffi_closure_osf_inner): New.
3911 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3912 (ffi_closure_osf): New.
3913
3914 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3915
3916 * config.guess, config.sub, install-sh: Removed.
3917 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3918 * Makefile.in: Rebuilt.
3919
3920 * acinclude.m4: Include libtool macros from the top level.
3921 * aclocal.m4, configure: Rebuilt.
3922
3923 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3924
3925 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3926 * configure: Rebuilt.
3927
3928 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3929
3930 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3931 memory correctly. Use conditional instructions, not branches where
3932 possible.
3933
3934 2000-05-04 Tom Tromey <tromey@cygnus.com>
3935
3936 * configure: Rebuilt.
3937 * configure.in: Match `arm*-*-linux-*'.
3938 From Chris Dornan <cdornan@arm.com>.
3939
3940 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3941
3942 * Makefile.am (SUBDIRS): Define.
3943 (AM_MAKEFLAGS): Likewise.
3944 (Multilib support.): Add section.
3945 * Makefile.in: Rebuilt.
3946 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3947 New variables. Set for gcc using -print-multi-lib. Export them
3948 to libtool.
3949 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3950 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3951 for -shared links.
3952 (extra_compiler_flags_value, extra_compiler_flags): Check these
3953 for extra compiler options which need to be passed down in
3954 compiler_flags.
3955
3956 2000-04-16 Anthony Green <green@redhat.com>
3957
3958 * configure: Rebuilt.
3959 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3960
3961 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3962
3963 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3964 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3965 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3966 Replace all void * sizeofs with sizeof(int).
3967 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3968 different than DOUBLE.
3969 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3970 (ffi_prep_args_v9): New function.
3971 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3972 (ffi_V9_return_struct): New function.
3973 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3974 32bit code (not yet cross-arch calls).
3975 * src/sparc/v8.S: Add struct return delay nop.
3976 Handle long long.
3977 * src/sparc/v9.S: New file.
3978 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3979 is used on sparc64 only for structures larger than 32 bytes.
3980 Pass by reference for structures is done for structure arguments
3981 larger than 16 bytes.
3982 * src/ffitest.c (main): Use 64bit rint on sparc64.
3983 Run long long tests on sparc.
3984 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3985 sparc64.
3986 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3987 on sparc64.
3988 * configure.in (sparc-*-linux*): New supported target.
3989 (sparc64-*-linux*): Likewise.
3990 * configure: Rebuilt.
3991 * Makefile.am: Add v9.S to SPARC files.
3992 * Makefile.in: Likewise.
3993 (LINK): Surround $(CCLD) into double quotes, so that multilib
3994 compiles work correctly.
3995
3996 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3997
3998 * configure: Rebuilt.
3999 * configure.in: (i*86-*-solaris*): New libffi target. Patch
4000 proposed by Bryce McKinlay.
4001
4002 2000-03-20 Tom Tromey <tromey@cygnus.com>
4003
4004 * Makefile.in: Hand edit for java_raw_api.lo.
4005
4006 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
4007
4008 * config.guess, config.sub: Update from the gcc tree.
4009 Fix for PR libgcj/168.
4010
4011 2000-03-03 Tom Tromey <tromey@cygnus.com>
4012
4013 * Makefile.in: Fixed ia64 by hand.
4014
4015 * configure: Rebuilt.
4016 * configure.in (--enable-multilib): New option.
4017 (libffi_basedir): New subst.
4018 (AC_OUTPUT): Added multilib code.
4019
4020 2000-03-02 Tom Tromey <tromey@cygnus.com>
4021
4022 * Makefile.in: Rebuilt.
4023 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
4024 directory name.
4025
4026 2000-02-25 Hans Boehm <boehm@acm.org>
4027
4028 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
4029 files.
4030 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
4031 list.
4032 (ffi_prep_raw_closure): Use ffi_translate_args, not
4033 ffi_closure_translate.
4034 * src/java_raw_api.c: New file.
4035 * src/ffitest.c (closure_test_fn): New function.
4036 (main): Define `rint' as long long on IA64. Added new test when
4037 FFI_CLOSURES is defined.
4038 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
4039 (ffi_abi): Recognize IA64.
4040 (ffi_raw): Added `flt' field.
4041 Added "Java raw API" code.
4042 * configure.in: Recognize ia64.
4043 * Makefile.am (TARGET_SRC_IA64): New macro.
4044 (libffi_la_common_SOURCES): Added java_raw_api.c.
4045 (libffi_la_SOURCES): Define in IA64 case.
4046
4047 2000-01-04 Tom Tromey <tromey@cygnus.com>
4048
4049 * Makefile.in: Rebuilt with newer automake.
4050
4051 1999-12-31 Tom Tromey <tromey@cygnus.com>
4052
4053 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
4054
4055 1999-09-01 Tom Tromey <tromey@cygnus.com>
4056
4057 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
4058 undefs.
4059 * fficonfig.h.in: Rebuilt.
4060 * configure: Rebuilt.
4061 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
4062 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
4063 * acconfig.h: Don't #undef PACKAGE or VERSION.
4064
4065 1999-08-09 Anthony Green <green@cygnus.com>
4066
4067 * include/ffi.h.in: Try to work around messy header problem
4068 with PACKAGE and VERSION.
4069
4070 * configure: Rebuilt.
4071 * configure.in: Change version to 2.00-beta.
4072
4073 * fficonfig.h.in: Rebuilt.
4074 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
4075
4076 * src/x86/ffi.c (ffi_raw_call): Rename.
4077
4078 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
4079
4080 * src/x86/ffi.c (ffi_closure_SYSV): New function.
4081 (ffi_prep_incoming_args_SYSV): Ditto.
4082 (ffi_prep_closure): Ditto.
4083 (ffi_closure_raw_SYSV): Ditto.
4084 (ffi_prep_raw_closure): More ditto.
4085 (ffi_call_raw): Final ditto.
4086
4087 * include/ffi.h.in: Add definitions for closure and raw API.
4088
4089 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4090 FFI_TYPE_UINT64.
4091
4092 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
4093
4094 * src/raw_api.c: New file.
4095
4096 * include/ffi.h.in (ffi_raw): New type.
4097 (UINT_ARG, SINT_ARG): New defines.
4098 (ffi_closure, ffi_raw_closure): New types.
4099 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4100
4101 * configure.in: Add check for endianness and sizeof void*.
4102
4103 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
4104 instead of directly.
4105
4106 * configure: Rebuilt.
4107
4108 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
4109
4110 * configure.in: Add x86 and powerpc BeOS configurations.
4111 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
4112
4113 1999-05-09 Anthony Green <green@cygnus.com>
4114
4115 * configure.in: Add warning about this being beta code.
4116 Remove src/Makefile.am from the picture.
4117 * configure: Rebuilt.
4118
4119 * Makefile.am: Move logic from src/Makefile.am. Add changes
4120 to support libffi as a target library.
4121 * Makefile.in: Rebuilt.
4122
4123 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
4124 Upgraded to new autoconf, automake, libtool.
4125
4126 * README: Tweaks.
4127
4128 * LICENSE: Update copyright date.
4129
4130 * src/Makefile.am, src/Makefile.in: Removed.
4131
4132 1998-11-29 Anthony Green <green@cygnus.com>
4133
4134 * include/ChangeLog: Removed.
4135 * src/ChangeLog: Removed.
4136 * src/mips/ChangeLog: Removed.
4137 * src/sparc/ChangeLog: Remboved.
4138 * src/x86/ChangeLog: Removed.
4139
4140 * ChangeLog.v1: Created.