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