]> git.ipfire.org Git - thirdparty/gcc.git/blame - libffi/ChangeLog
* ifcvt.c (if_convert): Output slim multiple dumps with TDF_SLIM.
[thirdparty/gcc.git] / libffi / ChangeLog
CommitLineData
b107942a 12009-12-31 Anthony Green <green@redhat.com>
2
3 * README: Update for libffi 3.0.9.
4
08fbebd8 52009-12-27 Matthias Klose <doko@ubuntu.com>
6
7 * configure.ac (HAVE_LONG_DOUBLE): Define for mips when
8 appropriate.
9 * configure: Rebuilt.
10
cc9d0711 112009-12-26 Anthony Green <green@redhat.com>
12
13 * testsuite/libffi.call/cls_longdouble_va.c: Mark as xfail for
14 avr32*-*-*.
15 * testsuite/libffi.call/cls_double_va.c: Ditto.
16
78d986dd 172009-12-26 Andreas Tobler <a.tobler@schweiz.org>
18
19 * testsuite/libffi.call/ffitest.h: Conditionally include stdint.h
20 and inttypes.h.
21 * testsuite/libffi.special/unwindtest.cc: Ditto.
22
f2ed5930 232009-12-26 Andreas Tobler <a.tobler@schweiz.org>
24
25 * configure.ac: Add amd64-*-openbsd*.
26 * configure: Rebuilt.
27 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Link
28 openbsd programs with -lpthread.
29
e1d6e63c 302009-12-26 Anthony Green <green@redhat.com>
31
32 * testsuite/libffi.call/cls_double_va.c,
33 testsuite/libffi.call/cls_longdouble.c,
34 testsuite/libffi.call/cls_longdouble_va.c,
35 testsuite/libffi.call/cls_pointer.c,
36 testsuite/libffi.call/cls_pointer_stack.c: Remove xfail for
37 mips*-*-* and arm*-*-*.
38 * testsuite/libffi.call/cls_align_longdouble_split.c,
39 testsuite/libffi.call/cls_align_longdouble_split2.c,
40 testsuite/libffi.call/stret_medium2.c,
41 testsuite/libffi.call/stret_medium.c,
42 testsuite/libffi.call/stret_large.c,
43 testsuite/libffi.call/stret_large2.c: Remove xfail for arm*-*-*.
44
1a42df5d 452009-12-31 Kay Tietz <ktietz70@googlemail.com>
46
47 * testsuite/libffi.call/ffitest.h,
48 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRuLL): Fix
49 definitions.
50
512009-12-31 Carlo Bramini <carlo.bramix@libero.it>
52
53 * configure.ac (AM_LTLDFLAGS): Define for windows hosts.
54 * Makefile.am (libffi_la_LDFLAGS): Add AM_LTLDFLAGS.
55 * configure: Rebuilt.
56 * Makefile.in: Rebuilt.
57
7d374786 582009-12-31 Anthony Green <green@redhat.com>
59 Blake Chaffin.
60
61 * testsuite/libffi.call/huge_struct.c: New test case from Blake
62 Chaffin @ Apple.
63
0d0fdfce 642009-12-28 David Edelsohn <edelsohn@gnu.org>
65
66 * src/powerpc/ffi_darwin.c (ffi_prep_args): Copy abi and nargs to
67 local variables.
68 (aix_adjust_aggregate_sizes): New function.
69 (ffi_prep_cif_machdep): Call it.
70
3531d81f 712009-12-26 Andreas Tobler <a.tobler@schweiz.org>
72
73 * configure.ac: Define FFI_MMAP_EXEC_WRIT for the given targets.
74 * configure: Regenerate.
75 * fficonfig.h.in: Likewise.
76 * src/closures.c: Remove the FFI_MMAP_EXEC_WRIT definition for
77 Solaris/x86.
78
81b376ae 792009-12-26 Andreas Schwab <schwab@linux-m68k.org>
80
81 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Advance intarg_count
82 when a float arguments is passed in memory.
83 (ffi_closure_helper_SYSV): Mark general registers as used up when
84 a 64bit or soft-float long double argument is passed in memory.
85
cac80d00 862009-12-25 Matthias Klose <doko@ubuntu.com>
657f62fe 87
88 * man/ffi_call.3: Fix #include in examples.
89 * doc/libffi.texi: Add dircategory.
90
da4f8cce 912009-12-25 Frank Everdij <f.p.x.everdij@tudelft.nl>
92
93 * include/ffi.h.in: Placed '__GNUC__' ifdef around
94 '__attribute__((aligned(8)))' in ffi_closure, fixes compile for
95 IRIX MIPSPro c99.
96 * include/ffi_common.h: Added '__sgi' define to non
97 '__attribute__((__mode__()))' integer typedefs.
81b376ae 98 * src/mips/ffi.c (ffi_call, ffi_closure_mips_inner_O32,
da4f8cce 99 ffi_closure_mips_inner_N32): Added 'defined(_MIPSEB)' to BE check.
100 (ffi_closure_mips_inner_O32, ffi_closure_mips_inner_N32): Added
101 FFI_LONGDOUBLE support and alignment(N32 only).
102 * src/mips/ffitarget.h: Corrected '#include <sgidefs.h>' for IRIX and
103 fixed non '__attribute__((__mode__()))' integer typedefs.
104 * src/mips/n32.S: Put '#ifdef linux' around '.abicalls' and '.eh_frame'
105 since they are Linux/GNU Assembler specific.
106
2a06dfe6 1072009-12-25 Bradley Smith <brad@brad-smith.co.uk>
108
109 * configure.ac, Makefile.am, src/avr32/ffi.c,
110 src/avr32/ffitarget.h,
111 src/avr32/sysv.S: Add AVR32 port.
112 * configure, Makefile.in: Rebuilt.
113
cc6aa710 1142009-12-21 Andreas Tobler <a.tobler@schweiz.org>
115
116 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
117 * configure: Regenerate.
118
2fb9cd2f 1192009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
120
121 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
122
c64cba58 1232009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
124
125 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
126 type on HP-UX.
127
c221d0e8 1282009-12-11 Eric Botcazou <ebotcazou@adacore.com>
129
130 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
131 double' arguments.
132
db0ed941 1332009-12-11 Eric Botcazou <ebotcazou@adacore.com>
134
135 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
136
4f2e543b 1372009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
138
139 PR libffi/40700
140 * src/closures.c [X86_64 && __sun__ && __svr4__]
141 (FFI_MMAP_EXEC_WRIT): Define.
142
2e5e20e1 1432009-12-08 David Daney <ddaney@caviumnetworks.com>
144
145 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
146 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
147 * testsuite/libffi.call/stret_large.c: Same.
148 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
149 * testsuite/libffi.call/stret_large2.c: Same.
150 * testsuite/libffi.call/stret_medium2.c: Same.
151
514d705d 1522009-12-07 David Edelsohn <edelsohn@gnu.org>
153
154 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
155 typo.
156
82d1d66a 1572009-12-05 David Edelsohn <edelsohn@gnu.org>
158
159 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
160 code.
514d705d 161 * src/powerpc/aix_closure.S: Same.
2e5e20e1 162
e6408286 1632009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
164
165 * Makefile.in: Regenerate.
166 * configure: Regenerate.
167 * include/Makefile.in: Regenerate.
168 * man/Makefile.in: Regenerate.
169 * testsuite/Makefile.in: Regenerate.
170
029645af 1712009-12-04 David Edelsohn <edelsohn@gnu.org>
172
173 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
81b376ae 174 linux64_closure.S.
2e5e20e1 175
f89c6c9d 1762009-12-04 Uros Bizjak <ubizjak@gmail.com>
177
178 PR libffi/41908
179 * src/x86/ffi64.c (classify_argument): Update from
180 gcc/config/i386/i386.c.
181 (ffi_closure_unix64_inner): Do not use the address of two consecutive
182 SSE registers directly.
183 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
184 for x86_64 linux targets.
185
d66a47ee 1862009-12-04 David Edelsohn <edelsohn@gnu.org>
187
188 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
189 pfr for long double split between fpr13 and stack.
190
3a95c696 1912009-12-03 David Edelsohn <edelsohn@gnu.org>
192
193 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
194 fparg_count twice for long double.
195
e3791522 1962009-12-03 David Edelsohn <edelsohn@gnu.org>
197
198 PR libffi/42243
199 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
200
0481262e 2012009-12-03 Uros Bizjak <ubizjak@gmail.com>
202
203 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
c485f998 204 Remove xfails for x86 linux targets.
0481262e 205
e5fdfcfa 2062009-12-02 David Edelsohn <edelsohn@gnu.org>
207
208 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
209 case.
210
2112009-12-01 David Edelsohn <edelsohn@gnu.org>
40e397ce 212
213 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
214 register usage. Call ffi_prep_args directly. Add long double
215 return value support.
216 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
217 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
218 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
219 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
220 mode.
221 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
222 into case.
223 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
224 Allocate result area between params and FPRs.
225
6e94d4ca 2262009-11-30 David Edelsohn <edelsohn@gnu.org>
227
228 PR target/35484
229 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
230 AIX64.
231 * src/powerpc/aix.S: Implement AIX64 version.
232 * src/powerpc/aix_closure.S: Implement AIX64 version.
233 (ffi_closure_ASM): Use extsb, lha and displament addresses.
234 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
235 support.
236 (ffi_prep_cif_machdep): Same.
237 (ffi_call): Same.
238 (ffi_closure_helper_DARWIN): Same.
239
2f9618b6 2402009-11-02 Andreas Tobler <a.tobler@schweiz.org>
241
242 PR libffi/41908
243 * testsuite/libffi.call/testclosure.c: New test.
244
80b483b5 2452009-09-28 Kai Tietz <kai.tietz@onevision.com>
246
247 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
248 assembly version use of ___chkstk.
249
520462c5 2502009-09-23 Matthias Klose <doko@ubuntu.com>
251
252 PR libffi/40242, PR libffi/41443
253 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
254 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
255 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
256 Change the conditionals to __SOFTFP__ || __ARM_EABI__
257 for -mfloat-abi=softfp to work.
258
513e3043 2592009-09-17 Loren J. Rittle <ljrittle@acm.org>
260
261 PR testsuite/32843 (strikes again)
262 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
263 enable proper extension on char and short.
264
841f7ca5 2652009-09-15 David Daney <ddaney@caviumnetworks.com>
266
267 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
268 handling for FFI_TYPE_POINTER.
269 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
270 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
271 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
272 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
273 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
274 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
275 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
276 structure and pointer returns.
277 (ffi_closure_N32): Add handling for pointer returns.
278 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
279 calc_n32_return_struct_flags): Handle soft-float.
280 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
281 (ffi_call_N32): Declare proper argument types.
282 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
283 soft-float.
284
81a71e2b 2852009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
286
287 * configure.ac (AC_PREREQ): Bump to 2.64.
288
cd9a469c 2892009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
290
115d57c5 291 * Makefile.am (install-html, install-pdf): Remove.
292 * Makefile.in: Regenerate.
293
cd9a469c 294 * Makefile.in: Regenerate.
295 * aclocal.m4: Regenerate.
296 * configure: Regenerate.
297 * fficonfig.h.in: Regenerate.
298 * include/Makefile.in: Regenerate.
299 * man/Makefile.in: Regenerate.
300 * testsuite/Makefile.in: Regenerate.
301
9abe4a66 3022009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
303
304 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
305
339e4d43 3062009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
307
308 PR libffi/40807
309 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
310 return types for X86_WIN32.
311 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
312 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
313 _ffi_closure_STDCALL): Likewise.
314
315 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
316 (dlmmap, dlmunmap): Also use these functions on Cygwin.
317
f09e0522 3182009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
319
320 PR testsuite/40699
321 PR testsuite/40707
322 PR testsuite/40709
323 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
324 2009-06-30 commits.
325
aa588440 3262009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
327
328 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
329 to "" before adding paths. (This reinstates an assignment that
330 was removed by my 2009-06-30 commit, but changes the initial
331 value from "." to "".)
332
0378a83e 3332009-07-01 H.J. Lu <hongjiu.lu@intel.com>
334
335 PR testsuite/40601
336 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
337 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
338 empty.
339
2f0e2680 3402009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
341
342 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
343 to ld_library_path. Use add_path. Add just find_libgcc_s
344 to ld_library_path, not every libgcc multilib directory.
345
a05d695d 3462009-06-16 Wim Lewis <wiml@hhhh.org>
347
348 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
349 supposed to be callee-saved.
350 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
351 return buffer for odd-size structs.
352
87f6899d 3532009-06-16 Andreas Tobler <a.tobler@schweiz.org>
354
355 PR libffi/40444
356 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
357 allow_stack_execute for Darwin.
358
fffc53bb 3592009-06-16 Andrew Haley <aph@redhat.com>
360
361 * configure.ac (TARGETDIR): Add missing blank lines.
362 * configure: Regenerate.
363
75c53eb0 3642009-06-16 Andrew Haley <aph@redhat.com>
365
366 * testsuite/libffi.call/cls_align_sint64.c,
367 testsuite/libffi.call/cls_align_uint64.c,
368 testsuite/libffi.call/cls_longdouble_va.c,
369 testsuite/libffi.call/cls_ulonglong.c,
370 testsuite/libffi.call/return_ll1.c,
371 testsuite/libffi.call/stret_medium2.c: Fix printf format
372 specifiers.
373 * testsuite/libffi.call/ffitest.h,
374 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
375
bb07cfdf 3762009-06-15 Andrew Haley <aph@redhat.com>
377
378 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
379 * testsuite/libffi.call/err_bad_abi.c: Likewise.
81b376ae 380
00581c89 3812009-06-12 Andrew Haley <aph@redhat.com>
382
383 * Makefile.am: Remove info_TEXINFOS.
384
3852009-06-12 Andrew Haley <aph@redhat.com>
386
387 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
388 testsuite/libffi.call/cls_align_uint64.c,
389 testsuite/libffi.call/cls_ulonglong.c,
390 testsuite/libffi.call/return_ll1.c,
391 testsuite/libffi.call/stret_medium2.c: Fix printf format
392 specifiers.
393 testsuite/libffi.special/unwindtest.cc: include stdint.h.
394
3952009-06-11 Timothy Wall <twall@users.sf.net>
396
397 * Makefile.am,
81b376ae 398 configure.ac,
399 include/ffi.h.in,
400 include/ffi_common.h,
401 src/closures.c,
402 src/dlmalloc.c,
403 src/x86/ffi.c,
404 src/x86/ffitarget.h,
405 src/x86/win64.S (new),
00581c89 406 README: Added win64 support (mingw or MSVC)
81b376ae 407 * Makefile.in,
408 include/Makefile.in,
409 man/Makefile.in,
410 testsuite/Makefile.in,
411 configure,
412 aclocal.m4: Regenerated
413 * ltcf-c.sh: properly escape cygwin/w32 path
414 * man/ffi_call.3: Clarify size requirements for return value.
415 * src/x86/ffi64.c: Fix filename in comment.
416 * src/x86/win32.S: Remove unused extern.
417
418 * testsuite/libffi.call/closure_fn0.c,
419 testsuite/libffi.call/closure_fn1.c,
420 testsuite/libffi.call/closure_fn2.c,
421 testsuite/libffi.call/closure_fn3.c,
422 testsuite/libffi.call/closure_fn4.c,
423 testsuite/libffi.call/closure_fn5.c,
424 testsuite/libffi.call/closure_fn6.c,
00581c89 425 testsuite/libffi.call/closure_stdcall.c,
426 testsuite/libffi.call/cls_12byte.c,
427 testsuite/libffi.call/cls_16byte.c,
428 testsuite/libffi.call/cls_18byte.c,
429 testsuite/libffi.call/cls_19byte.c,
430 testsuite/libffi.call/cls_1_1byte.c,
431 testsuite/libffi.call/cls_20byte.c,
432 testsuite/libffi.call/cls_20byte1.c,
433 testsuite/libffi.call/cls_24byte.c,
434 testsuite/libffi.call/cls_2byte.c,
435 testsuite/libffi.call/cls_3_1byte.c,
436 testsuite/libffi.call/cls_3byte1.c,
437 testsuite/libffi.call/cls_3byte2.c,
438 testsuite/libffi.call/cls_4_1byte.c,
439 testsuite/libffi.call/cls_4byte.c,
440 testsuite/libffi.call/cls_5_1_byte.c,
441 testsuite/libffi.call/cls_5byte.c,
442 testsuite/libffi.call/cls_64byte.c,
443 testsuite/libffi.call/cls_6_1_byte.c,
444 testsuite/libffi.call/cls_6byte.c,
445 testsuite/libffi.call/cls_7_1_byte.c,
446 testsuite/libffi.call/cls_7byte.c,
447 testsuite/libffi.call/cls_8byte.c,
448 testsuite/libffi.call/cls_9byte1.c,
449 testsuite/libffi.call/cls_9byte2.c,
450 testsuite/libffi.call/cls_align_double.c,
451 testsuite/libffi.call/cls_align_float.c,
452 testsuite/libffi.call/cls_align_longdouble.c,
453 testsuite/libffi.call/cls_align_longdouble_split.c,
454 testsuite/libffi.call/cls_align_longdouble_split2.c,
455 testsuite/libffi.call/cls_align_pointer.c,
456 testsuite/libffi.call/cls_align_sint16.c,
457 testsuite/libffi.call/cls_align_sint32.c,
458 testsuite/libffi.call/cls_align_sint64.c,
459 testsuite/libffi.call/cls_align_uint16.c,
460 testsuite/libffi.call/cls_align_uint32.c,
461 testsuite/libffi.call/cls_align_uint64.c,
462 testsuite/libffi.call/cls_dbls_struct.c,
463 testsuite/libffi.call/cls_double.c,
464 testsuite/libffi.call/cls_double_va.c,
465 testsuite/libffi.call/cls_float.c,
466 testsuite/libffi.call/cls_longdouble.c,
467 testsuite/libffi.call/cls_longdouble_va.c,
468 testsuite/libffi.call/cls_multi_schar.c,
469 testsuite/libffi.call/cls_multi_sshort.c,
470 testsuite/libffi.call/cls_multi_sshortchar.c,
471 testsuite/libffi.call/cls_multi_uchar.c,
472 testsuite/libffi.call/cls_multi_ushort.c,
473 testsuite/libffi.call/cls_multi_ushortchar.c,
474 testsuite/libffi.call/cls_pointer.c,
475 testsuite/libffi.call/cls_pointer_stack.c,
476 testsuite/libffi.call/cls_schar.c,
477 testsuite/libffi.call/cls_sint.c,
478 testsuite/libffi.call/cls_sshort.c,
479 testsuite/libffi.call/cls_uchar.c,
480 testsuite/libffi.call/cls_uint.c,
481 testsuite/libffi.call/cls_ulonglong.c,
482 testsuite/libffi.call/cls_ushort.c,
483 testsuite/libffi.call/err_bad_abi.c,
484 testsuite/libffi.call/err_bad_typedef.c,
485 testsuite/libffi.call/float2.c,
486 testsuite/libffi.call/huge_struct.c,
487 testsuite/libffi.call/nested_struct.c,
488 testsuite/libffi.call/nested_struct1.c,
489 testsuite/libffi.call/nested_struct10.c,
490 testsuite/libffi.call/nested_struct2.c,
491 testsuite/libffi.call/nested_struct3.c,
492 testsuite/libffi.call/nested_struct4.c,
493 testsuite/libffi.call/nested_struct5.c,
494 testsuite/libffi.call/nested_struct6.c,
495 testsuite/libffi.call/nested_struct7.c,
496 testsuite/libffi.call/nested_struct8.c,
497 testsuite/libffi.call/nested_struct9.c,
498 testsuite/libffi.call/problem1.c,
499 testsuite/libffi.call/return_ldl.c,
500 testsuite/libffi.call/return_ll1.c,
501 testsuite/libffi.call/stret_large.c,
502 testsuite/libffi.call/stret_large2.c,
503 testsuite/libffi.call/stret_medium.c,
504 testsuite/libffi.call/stret_medium2.c,
81b376ae 505 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
506 of checking for MMAP. Use intptr_t instead of long casts.
00581c89 507
f6e79f69 5082009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
509
510 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
511 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
512 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
513
e5ad3c0e 5142009-06-09 Andrew Haley <aph@redhat.com>
515
516 * src/x86/freebsd.S: Add missing file.
517
1f429ea0 5182009-06-08 Andrew Haley <aph@redhat.com>
519
520 Import from libffi 3.0.8:
521
522 * doc/libffi.texi: New file.
523 * doc/libffi.info: Likewise.
524 * doc/stamp-vti: Likewise.
525 * man/Makefile.am: New file.
526 * man/ffi_call.3: New file.
527
528 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
529 src/dlmalloc.c.
530 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
531
532 * configure.ac: Bump version to 3.0.8.
533 parisc*-*-linux*: Add.
534 i386-*-freebsd* | i386-*-openbsd*: Add.
535 powerpc-*-beos*: Add.
536 AM_CONDITIONAL X86_FREEBSD: Add.
537 AC_CONFIG_FILES: Add man/Makefile.
538
539 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
540
7302546d 5412009-06-08 Andrew Haley <aph@redhat.com>
542
543 * README: Import from libffi 3.0.8.
544
07d9c635 5452009-06-08 Andrew Haley <aph@redhat.com>
546
547 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
548 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
549 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
550 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
551
552 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
553 * testsuite/libffi.call/stret_medium.c: Likewise.
554 * testsuite/libffi.call/stret_large2.c: Likewise.
555 * testsuite/libffi.call/stret_large.c: Likewise.
556
5572008-12-26 Timothy Wall <twall@users.sf.net>
558
81b376ae 559 * testsuite/libffi.call/cls_longdouble.c,
560 testsuite/libffi.call/cls_longdouble_va.c,
561 testsuite/libffi.call/cls_align_longdouble.c,
562 testsuite/libffi.call/cls_align_longdouble_split.c,
563 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
564 failures on x86_64 cygwin/mingw.
07d9c635 565
5662008-12-22 Timothy Wall <twall@users.sf.net>
567
81b376ae 568 * testsuite/libffi.call/closure_fn0.c,
569 testsuite/libffi.call/closure_fn1.c,
570 testsuite/libffi.call/closure_fn2.c,
571 testsuite/libffi.call/closure_fn3.c,
572 testsuite/libffi.call/closure_fn4.c,
573 testsuite/libffi.call/closure_fn5.c,
574 testsuite/libffi.call/closure_fn6.c,
575 testsuite/libffi.call/closure_loc_fn0.c,
576 testsuite/libffi.call/closure_stdcall.c,
577 testsuite/libffi.call/cls_align_pointer.c,
578 testsuite/libffi.call/cls_pointer.c,
579 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
580 pointer to integer (intptr_t).
581 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
07d9c635 582
5832008-07-24 Anthony Green <green@redhat.com>
584
81b376ae 585 * testsuite/libffi.call/cls_dbls_struct.c,
586 testsuite/libffi.call/cls_double_va.c,
587 testsuite/libffi.call/cls_longdouble.c,
588 testsuite/libffi.call/cls_longdouble_va.c,
589 testsuite/libffi.call/cls_pointer.c,
590 testsuite/libffi.call/cls_pointer_stack.c,
591 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
592 compiler warnings.
07d9c635 593
5942008-03-04 Anthony Green <green@redhat.com>
81b376ae 595 Blake Chaffin
596 hos@tamanegi.org
597
598 * testsuite/libffi.call/cls_align_longdouble_split2.c
599 testsuite/libffi.call/cls_align_longdouble_split.c
600 testsuite/libffi.call/cls_dbls_struct.c
601 testsuite/libffi.call/cls_double_va.c
602 testsuite/libffi.call/cls_longdouble.c
603 testsuite/libffi.call/cls_longdouble_va.c
604 testsuite/libffi.call/cls_pointer.c
605 testsuite/libffi.call/cls_pointer_stack.c
606 testsuite/libffi.call/err_bad_abi.c
607 testsuite/libffi.call/err_bad_typedef.c
608 testsuite/libffi.call/stret_large2.c
609 testsuite/libffi.call/stret_large.c
610 testsuite/libffi.call/stret_medium2.c
611 testsuite/libffi.call/stret_medium.c: New tests from Apple.
07d9c635 612
c720ee9b 6132009-06-05 Andrew Haley <aph@redhat.com>
614
615 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
616 libffi.
617
91e95000 6182009-06-04 Andrew Haley <aph@redhat.com>
619
620 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
621 stdcall changes.
622
1f9ea1fc 6232008-02-26 Anthony Green <green@redhat.com>
81b376ae 624 Thomas Heller <theller@ctypes.org>
1f9ea1fc 625
81b376ae 626 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
627 comment.
1f9ea1fc 628
6292008-02-03 Timothy Wall <twall@users.sf.net>
630
81b376ae 631 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
632 offset based on code pointer, not data pointer.
1f9ea1fc 633
6342008-01-31 Timothy Wall <twall@users.sf.net>
635
81b376ae 636 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
637 closures.
638 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
639 closures.
640 * src/x86/win32.S: Add assembly for stdcall closure.
641 * src/x86/ffi.c: Initialize stdcall closure trampoline.
1f9ea1fc 642
e01722f1 6432009-06-04 Andrew Haley <aph@redhat.com>
644
645 * include/ffi.h.in: Change void (*)() to void (*)(void).
646 * src/x86/ffi.c: Likewise.
647
6e59f66f 6482009-06-04 Andrew Haley <aph@redhat.com>
649
650 * src/powerpc/ppc_closure.S: Insert licence header.
651 * src/powerpc/linux64_closure.S: Likewise.
652 * src/m68k/sysv.S: Likewise.
653
654 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
655 * src/powerpc/ffi.c: Likewise.
656 * src/powerpc/ffi_darwin.c: Likewise.
657 * src/m32r/ffi.c: Likewise.
658 * src/sh64/ffi.c: Likewise.
659 * src/x86/ffi64.c: Likewise.
660 * src/alpha/ffi.c: Likewise.
661 * src/alpha/osf.S: Likewise.
662 * src/frv/ffi.c: Likewise.
663 * src/s390/ffi.c: Likewise.
664 * src/pa/ffi.c: Likewise.
665 * src/pa/hpux32.S: Likewise.
666 * src/ia64/unix.S: Likewise.
667 * src/ia64/ffi.c: Likewise.
668 * src/sparc/ffi.c: Likewise.
669 * src/mips/ffi.c: Likewise.
670 * src/sh/ffi.c: Likewise.
671
6722008-02-15 David Daney <ddaney@avtrex.com>
673
81b376ae 674 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
675 Define (conditionally), and use it to include cachectl.h.
676 (ffi_prep_closure_loc): Fix cache flushing.
677 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
6e59f66f 678
9d98ec1b 6792009-06-04 Andrew Haley <aph@redhat.com>
680
681 include/ffi.h.in,
682 src/arm/ffitarget.h,
683 src/arm/ffi.c,
684 src/arm/sysv.S,
685 src/powerpc/ffitarget.h,
686 src/closures.c,
687 src/sh64/ffitarget.h,
688 src/sh64/ffi.c,
689 src/sh64/sysv.S,
690 src/types.c,
691 src/x86/ffi64.c,
692 src/x86/ffitarget.h,
693 src/x86/win32.S,
694 src/x86/darwin.S,
695 src/x86/ffi.c,
696 src/x86/sysv.S,
697 src/x86/unix64.S,
698 src/alpha/ffitarget.h,
699 src/alpha/ffi.c,
700 src/alpha/osf.S,
701 src/m68k/ffitarget.h,
702 src/frv/ffitarget.h,
703 src/frv/ffi.c,
704 src/s390/ffitarget.h,
705 src/s390/sysv.S,
706 src/cris/ffitarget.h,
707 src/pa/linux.S,
708 src/pa/ffitarget.h,
709 src/pa/ffi.c,
710 src/raw_api.c,
711 src/ia64/ffitarget.h,
712 src/ia64/unix.S,
713 src/ia64/ffi.c,
714 src/ia64/ia64_flags.h,
715 src/java_raw_api.c,
716 src/debug.c,
717 src/sparc/v9.S,
718 src/sparc/ffitarget.h,
719 src/sparc/ffi.c,
720 src/sparc/v8.S,
721 src/mips/ffitarget.h,
722 src/mips/n32.S,
723 src/mips/o32.S,
724 src/mips/ffi.c,
725 src/prep_cif.c,
726 src/sh/ffitarget.h,
727 src/sh/ffi.c,
728 src/sh/sysv.S: Update license text.
729
ad81e87b 7302009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
731
732 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
733 (.eh_frame): Add FDE for it.
734
1cde303d 7352009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
736
737 * configure.ac: Also check if assembler supports pc-relative
738 relocs on X86_WIN32 targets.
739 * configure: Regenerate.
740 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
741 (_ffi_call_SYSV): Add missing function type symbol .def and
742 add EH markup labels.
743 (_ffi_call_STDCALL): Likewise.
744 (_ffi_closure_SYSV): Likewise.
745 (_ffi_closure_raw_SYSV): Likewise.
746 (.eh_frame): Add hand-crafted EH data.
747
6bc9506f 7482009-04-09 Jakub Jelinek <jakub@redhat.com>
749
750 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
751 version 3 of the GNU General Public License and to point readers
752 at the COPYING3 file and the FSF's license web page.
753 * testsuite/libffi.call/call.exp: Likewise.
754 * testsuite/libffi.special/special.exp: Likewise.
755
9af964d4 7562009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
757
758 * configure: Regenerate.
759
95ff4d7d 7602008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
761
762 PR libffi/26048
763 * configure.ac (HAVE_AS_X86_PCREL): New test.
764 * configure: Regenerate.
765 * fficonfig.h.in: Regenerate.
766 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
767 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
768 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
769 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
770 * src/x86/unix64.S (.Lstore_table): Move to .text section.
771 (.Lload_table): Likewise.
772 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
773
9e6fd619 7742008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
775
776 * configure: Regenerate.
777
95c3349b 7782008-11-21 Eric Botcazou <ebotcazou@adacore.com>
779
780 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
781 signed/unsigned int8/16 return values.
782 * src/sparc/v8.S (ffi_call_v8): Likewise.
783 (ffi_closure_v8): Likewise.
784
4bc2b84f 7852008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
81b376ae 786 Steve Ellcey <sje@cup.hp.com>
4bc2b84f 787
788 * configure: Regenerate for new libtool.
789 * Makefile.in: Ditto.
790 * include/Makefile.in: Ditto.
791 * aclocal.m4: Ditto.
792
a2c1e278 7932008-08-25 Andreas Tobler <a.tobler@schweiz.org>
794
795 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
796 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
797 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
798 Adjust copyright notice.
799 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
800 register or two register to use for FFI_SYSV structs.
801 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
802 (ffi_closure_helper_SYSV): Fix the return type for
803 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
804 Adjust copyright notice.
805
0733c867 8062008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
807
808 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
809 int.
810
5b2c7553 8112008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
812
813 * configure: Regenerate.
814 * include/Makefile.in: Regenerate.
815 * testsuite/Makefile.in: Regenerate.
816
6af9f7ea 8172008-06-07 Joseph Myers <joseph@codesourcery.com>
818
819 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
820 powerpc-*-beos*): Remove.
821 * configure: Regenerate.
822
ac051731 8232008-05-09 Julian Brown <julian@codesourcery.com>
824
825 * Makefile.am (LTLDFLAGS): New.
826 (libffi_la_LDFLAGS): Use above.
827 * Makefile.in: Regenerate.
828
6a88da51 8292008-04-18 Paolo Bonzini <bonzini@gnu.org>
830
831 PR bootstrap/35457
832 * aclocal.m4: Regenerate.
833 * configure: Regenerate.
834
644af649 8352008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
836
837 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
838 * src/sh64/sysv.S: Likewise.
839
595c2b68 8402008-03-26 Daniel Jacobowitz <dan@debian.org>
841
842 * src/arm/sysv.S: Fix ARM comment marker.
843
8442008-03-26 Jakub Jelinek <jakub@redhat.com>
845
846 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
847 * src/s390/sysv.S: Likewise.
848 * src/powerpc/ppc_closure.S: Likewise.
849 * src/powerpc/sysv.S: Likewise.
850 * src/x86/unix64.S: Likewise.
851 * src/x86/sysv.S: Likewise.
852 * src/sparc/v8.S: Likewise.
853 * src/sparc/v9.S: Likewise.
854 * src/m68k/sysv.S: Likewise.
855 * src/arm/sysv.S: Likewise.
856
82eaf560 8572008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
858
859 * aclocal.m4: Regenerate.
860 * configure: Likewise.
861 * Makefile.in: Likewise.
862 * include/Makefile.in: Likewise.
863 * testsuite/Makefile.in: Likewise.
864
55e813ce 8652008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
866 Andreas Tobler <a.tobler@schweiz.org>
867
868 * configure.ac: Add amd64-*-freebsd* target.
869 * configure: Regenerate.
870
00eae123 8712008-01-30 H.J. Lu <hongjiu.lu@intel.com>
872
873 PR libffi/34612
874 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
875 returning struct.
876
877 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
878 tests.
879
ff3fb914 8802008-01-24 David Edelsohn <edelsohn@gnu.org>
881
882 * configure: Regenerate.
883
8f1f10d3 8842008-01-06 Andreas Tobler <a.tobler@schweiz.org>
885
886 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
887
ad73551f 8882008-01-05 Andreas Tobler <a.tobler@schweiz.org>
889
890 PR testsuite/32843
891 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
892 signed/unsigned int8/16 for X86_DARWIN.
893 Updated copyright info.
894 Handle one and two byte structs with special cif->flags.
895 * src/x86/ffitarget.h: Add special types for one and two byte structs.
896 Updated copyright info.
897 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
898 sysv.S
899 Remove code to pop args from the stack after call.
900 Special-case signed/unsigned for int8/16, one and two byte structs.
901 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
902 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
903 FFI_TYPE_SINT32.
904 Updated copyright info.
905
aaf16623 9062007-12-08 David Daney <ddaney@avtrex.com>
907
81b376ae 908 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
aaf16623 909 SUBU, add with ADDU and use smaller code sequences.
910
2c3f7870 9112007-12-07 David Daney <ddaney@avtrex.com>
2af77f57 912
2c3f7870 913 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
914 type.
915
5be4032c 9162007-12-06 David Daney <ddaney@avtrex.com>
917
918 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
919 defined.
920 (ffi_java_raw): New typedef.
921 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
922 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
923 ffi_java_raw.
924 (ffi_java_raw_closure) : Same.
925 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
926 parameter types.
927 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
928 FFI_SIZEOF_JAVA_RAW.
929 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
930 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
931 sizeof(ffi_java_raw) for alignment calculations.
932 (ffi_java_ptrarray_to_raw): Same.
933 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
81b376ae 934 if FFI_SIZEOF_JAVA_RAW == 4.
5be4032c 935 (ffi_java_raw_to_rvalue): Same.
936 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
937 (ffi_java_translate_args): Same.
938 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
939 parameter types.
940 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
941
79369c38 9422007-12-06 David Daney <ddaney@avtrex.com>
943
81b376ae 944 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
79369c38 945 pointer values.
946
7cf0754c 9472007-12-01 Andreas Tobler <a.tobler@schweiz.org>
948
949 PR libffi/31937
950 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
951 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
952 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
953 set the NUM_FPR_ARG_REGISTERS according to.
954 Add support for potential soft-float support under hard-float
955 architecture.
956 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
957 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
958 to the FFI_LINUX_SOFT_FLOAT ABI.
959 (ffi_prep_cif_machdep): Likewise.
960 (ffi_closure_helper_SYSV): Likewise.
961 * src/powerpc/ppc_closure.S: Make sure not to store float/double
962 on archs where __NO_FPRS__ is true.
963 Add FFI_TYPE_UINT128 support.
964 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
965 Adjust copyright notice.
966
5036bf25 9672007-11-25 Andreas Tobler <a.tobler@schweiz.org>
968
969 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
970 * include/ffi_common.h: ... here.
971 Update copyright.
972
c3f75f11 9732007-11-17 Andreas Tobler <a.tobler@schweiz.org>
974
975 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
976 * src/powerpc/linux64.S: Likewise.
977 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
978 * testsuite/libffi.call/return_ldl.c: New test.
979
244c3f18 9802007-09-04 <aph@redhat.com>
981
982 * src/arm/sysv.S (UNWIND): New.
983 (Whole file): Conditionally compile unwinder directives.
984 * src/arm/sysv.S: Add unwinder directives.
985
986 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
987 Only treat r0 as a struct address if we're actually returning a
988 struct by address.
989 Only copy the bytes that are actually within a struct.
990 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
991 is returned in r0, not passed by address.
992 (ffi_call): Allocate a word-sized temporary for the case where
993 a composite is returned in r0.
994 (ffi_prep_incoming_args_SYSV): Align as necessary.
995
9962007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
997
81b376ae 998 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
244c3f18 999 directly using the sys_cacheflush syscall.
1000
10012007-07-27 Andrew Haley <aph@redhat.com>
1002
1003 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
1004
7a659e46 10052007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
1006
1007 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
1008 * configure.ac: Likewise.
1009 * Makefile.in: Regenerate.
1010 * include/Makefile.in: Likewise.
1011 * testsuite/Makefile.in: Likewise.
1012 * configure: Likewise.
1013
99b65499 10142007-08-24 David Daney <ddaney@avtrex.com>
1015
1016 * testsuite/libffi.call/return_sl.c: New test.
1017
b5c438f4 10182007-08-10 David Daney <ddaney@avtrex.com>
1019
1020 * testsuite/libffi.call/cls_multi_ushort.c,
1021 testsuite/libffi.call/cls_align_uint16.c,
1022 testsuite/libffi.call/nested_struct1.c,
1023 testsuite/libffi.call/nested_struct3.c,
1024 testsuite/libffi.call/cls_7_1_byte.c,
1025 testsuite/libffi.call/nested_struct5.c,
1026 testsuite/libffi.call/cls_double.c,
1027 testsuite/libffi.call/nested_struct7.c,
1028 testsuite/libffi.call/cls_sint.c,
1029 testsuite/libffi.call/nested_struct9.c,
1030 testsuite/libffi.call/cls_20byte1.c,
1031 testsuite/libffi.call/cls_multi_sshortchar.c,
1032 testsuite/libffi.call/cls_align_sint64.c,
1033 testsuite/libffi.call/cls_3byte2.c,
1034 testsuite/libffi.call/cls_multi_schar.c,
1035 testsuite/libffi.call/cls_multi_uchar.c,
1036 testsuite/libffi.call/cls_19byte.c,
1037 testsuite/libffi.call/cls_9byte1.c,
1038 testsuite/libffi.call/cls_align_float.c,
1039 testsuite/libffi.call/closure_fn1.c,
1040 testsuite/libffi.call/problem1.c,
1041 testsuite/libffi.call/closure_fn3.c,
1042 testsuite/libffi.call/cls_sshort.c,
1043 testsuite/libffi.call/closure_fn5.c,
1044 testsuite/libffi.call/cls_align_double.c,
1045 testsuite/libffi.call/nested_struct.c,
1046 testsuite/libffi.call/cls_2byte.c,
1047 testsuite/libffi.call/nested_struct10.c,
1048 testsuite/libffi.call/cls_4byte.c,
1049 testsuite/libffi.call/cls_6byte.c,
1050 testsuite/libffi.call/cls_8byte.c,
1051 testsuite/libffi.call/cls_multi_sshort.c,
1052 testsuite/libffi.call/cls_align_sint16.c,
1053 testsuite/libffi.call/cls_align_uint32.c,
1054 testsuite/libffi.call/cls_20byte.c,
1055 testsuite/libffi.call/cls_float.c,
1056 testsuite/libffi.call/nested_struct2.c,
1057 testsuite/libffi.call/cls_5_1_byte.c,
1058 testsuite/libffi.call/nested_struct4.c,
1059 testsuite/libffi.call/cls_24byte.c,
1060 testsuite/libffi.call/nested_struct6.c,
1061 testsuite/libffi.call/cls_64byte.c,
1062 testsuite/libffi.call/nested_struct8.c,
1063 testsuite/libffi.call/cls_uint.c,
1064 testsuite/libffi.call/cls_multi_ushortchar.c,
1065 testsuite/libffi.call/cls_schar.c,
1066 testsuite/libffi.call/cls_uchar.c,
1067 testsuite/libffi.call/cls_align_uint64.c,
1068 testsuite/libffi.call/cls_ulonglong.c,
1069 testsuite/libffi.call/cls_align_longdouble.c,
1070 testsuite/libffi.call/cls_1_1byte.c,
1071 testsuite/libffi.call/cls_12byte.c,
1072 testsuite/libffi.call/cls_3_1byte.c,
1073 testsuite/libffi.call/cls_3byte1.c,
1074 testsuite/libffi.call/cls_4_1byte.c,
1075 testsuite/libffi.call/cls_6_1_byte.c,
1076 testsuite/libffi.call/cls_16byte.c,
1077 testsuite/libffi.call/cls_18byte.c,
1078 testsuite/libffi.call/closure_fn0.c,
1079 testsuite/libffi.call/cls_9byte2.c,
1080 testsuite/libffi.call/closure_fn2.c,
1081 testsuite/libffi.call/closure_fn4.c,
1082 testsuite/libffi.call/cls_ushort.c,
1083 testsuite/libffi.call/closure_fn6.c,
1084 testsuite/libffi.call/cls_5byte.c,
1085 testsuite/libffi.call/cls_align_pointer.c,
1086 testsuite/libffi.call/cls_7byte.c,
1087 testsuite/libffi.call/cls_align_sint32.c,
1088 testsuite/libffi.special/unwindtest_ffi_call.cc,
1089 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
1090
edd57a0a 10912007-08-10 David Daney <ddaney@avtrex.com>
1092
1093 PR libffi/28313
1094 * configure.ac: Don't treat mips64 as a special case.
1095 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
1096 * configure: Regenerate
1097 * Makefile.in: Ditto.
1098 * fficonfig.h.in: Ditto.
1099 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
1100 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
1101 (FFI_DEFAULT_ABI): Set for n64 case.
1102 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
1103 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
1104 (ffi_closure_N32): New function.
1105 (.eh_frame): New section
1106 * src/mips/o32.S: Clean up comments.
1107 (ffi_closure_O32): Pass ffi_closure parameter in $12.
1108 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
1109 _MIPS_SIM == _ABIN32 throughout.
1110 (FFI_MIPS_STOP_HERE): New, use in place of
1111 ffi_stop_here.
1112 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1113 to support n32/n64 ABIs.
1114 (calc_n32_struct_flags): Rewrite.
1115 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1116 position of flag bits.
1117 (ffi_prep_cif_machdep): Rewrite n32 portion.
1118 (ffi_call): Enable for n64. Add special handling for small structure
1119 return values.
1120 (ffi_prep_closure_loc): Add n32 and n64 support.
1121 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1122 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1123
3471c5eb 11242007-08-08 David Daney <ddaney@avtrex.com>
1125
1126 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1127 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1128 specifiers.
1129 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1130 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1131 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1132 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1133 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1134 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1135 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1136 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1137 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1138 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1139 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1140 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1141 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1142 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1143 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1144 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1145 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1146 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1147 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1148 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1149 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1150 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1151 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1152 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1153 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1154 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1155 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1156 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1157 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1158 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1159 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1160
ad7bf8e7 11612007-08-07 Andrew Haley <aph@redhat.com>
1162
1163 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1164 checkin.
1165
6cc1286e 11662007-08-06 Andrew Haley <aph@redhat.com>
1167
1168 PR testsuite/32843
1169 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1170 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1171 FFI_TYPE_SINT32.
1172
c6f5483d 11732007-08-02 David Daney <ddaney@avtrex.com>
1174
1175 * testsuite/libffi.call/return_ul.c (main): Define return type as
1176 ffi_arg. Use proper printf conversion specifier.
81b376ae 1177
f6179f0e 11782007-07-30 Andrew Haley <aph@redhat.com>
1179
1180 PR testsuite/32843
1181 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1182 signed/unsigned int8/16.
1183 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1184 Use a jump table.
1185 Remove code to pop args from the stack after call.
1186 Special-case signed/unsigned int8/16.
1187 * testsuite/libffi.call/return_sc.c (main): Revert.
1188
7babed79 11892007-07-26 Richard Guenther <rguenther@suse.de>
1190
1191 PR testsuite/32843
1192 * testsuite/libffi.call/return_sc.c (main): Verify call
1193 result as signed char, not ffi_arg.
1194
f54a4242 11952007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1196
1197 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1198 * configure: Regenerate.
1199
e96a69da 12002007-07-11 David Daney <ddaney@avtrex.com>
1201
1202 * src/mips/ffi.c: Don't include sys/cachectl.h.
1203 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1204 cacheflush().
1205
187f6543 12062007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1207
1208 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1209 from (ffi_prep_closure): ... this.
1210 (FFI_INIT_TRAMPOLINE): Adjust.
1211
12122005-12-31 Phil Blundell <pb@reciva.com>
1213
81b376ae 1214 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1215 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
187f6543 1216 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1217 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1218 (FFI_CLOSURES): Enable closure support.
1219
12202007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1221
81b376ae 1222 * testsuite/libffi.call/cls_multi_ushort.c,
1223 testsuite/libffi.call/cls_align_uint16.c,
1224 testsuite/libffi.call/nested_struct1.c,
1225 testsuite/libffi.call/nested_struct3.c,
1226 testsuite/libffi.call/cls_7_1_byte.c,
1227 testsuite/libffi.call/cls_double.c,
1228 testsuite/libffi.call/nested_struct5.c,
1229 testsuite/libffi.call/nested_struct7.c,
1230 testsuite/libffi.call/cls_sint.c,
1231 testsuite/libffi.call/nested_struct9.c,
1232 testsuite/libffi.call/cls_20byte1.c,
1233 testsuite/libffi.call/cls_multi_sshortchar.c,
1234 testsuite/libffi.call/cls_align_sint64.c,
1235 testsuite/libffi.call/cls_3byte2.c,
1236 testsuite/libffi.call/cls_multi_schar.c,
1237 testsuite/libffi.call/cls_multi_uchar.c,
1238 testsuite/libffi.call/cls_19byte.c,
1239 testsuite/libffi.call/cls_9byte1.c,
1240 testsuite/libffi.call/cls_align_float.c,
1241 testsuite/libffi.call/closure_fn1.c,
1242 testsuite/libffi.call/problem1.c,
1243 testsuite/libffi.call/closure_fn3.c,
1244 testsuite/libffi.call/cls_sshort.c,
1245 testsuite/libffi.call/closure_fn5.c,
1246 testsuite/libffi.call/cls_align_double.c,
1247 testsuite/libffi.call/cls_2byte.c,
1248 testsuite/libffi.call/nested_struct.c,
1249 testsuite/libffi.call/nested_struct10.c,
1250 testsuite/libffi.call/cls_4byte.c,
1251 testsuite/libffi.call/cls_6byte.c,
1252 testsuite/libffi.call/cls_8byte.c,
1253 testsuite/libffi.call/cls_multi_sshort.c,
1254 testsuite/libffi.call/cls_align_uint32.c,
1255 testsuite/libffi.call/cls_align_sint16.c,
1256 testsuite/libffi.call/cls_float.c,
1257 testsuite/libffi.call/cls_20byte.c,
1258 testsuite/libffi.call/cls_5_1_byte.c,
1259 testsuite/libffi.call/nested_struct2.c,
1260 testsuite/libffi.call/cls_24byte.c,
1261 testsuite/libffi.call/nested_struct4.c,
1262 testsuite/libffi.call/nested_struct6.c,
1263 testsuite/libffi.call/cls_64byte.c,
1264 testsuite/libffi.call/nested_struct8.c,
1265 testsuite/libffi.call/cls_uint.c,
1266 testsuite/libffi.call/cls_multi_ushortchar.c,
1267 testsuite/libffi.call/cls_schar.c,
1268 testsuite/libffi.call/cls_uchar.c,
1269 testsuite/libffi.call/cls_align_uint64.c,
1270 testsuite/libffi.call/cls_ulonglong.c,
1271 testsuite/libffi.call/cls_align_longdouble.c,
1272 testsuite/libffi.call/cls_1_1byte.c,
1273 testsuite/libffi.call/cls_12byte.c,
1274 testsuite/libffi.call/cls_3_1byte.c,
1275 testsuite/libffi.call/cls_3byte1.c,
1276 testsuite/libffi.call/cls_4_1byte.c,
1277 testsuite/libffi.call/cls_6_1_byte.c,
1278 testsuite/libffi.call/cls_16byte.c,
1279 testsuite/libffi.call/cls_18byte.c,
1280 testsuite/libffi.call/closure_fn0.c,
1281 testsuite/libffi.call/cls_9byte2.c,
1282 testsuite/libffi.call/closure_fn2.c,
1283 testsuite/libffi.call/closure_fn4.c,
1284 testsuite/libffi.call/cls_ushort.c,
1285 testsuite/libffi.call/closure_fn6.c,
1286 testsuite/libffi.call/cls_5byte.c,
1287 testsuite/libffi.call/cls_align_pointer.c,
1288 testsuite/libffi.call/cls_7byte.c,
1289 testsuite/libffi.call/cls_align_sint32.c,
1290 testsuite/libffi.special/unwindtest_ffi_call.cc,
1291 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
187f6543 1292
cecc75a9 12932007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1294
1295 * aclocal.m4: Regenerated.
1296
07c46636 12972007-06-02 Paolo Bonzini <bonzini@gnu.org>
1298
1299 * configure: Regenerate.
1300
542f9784 13012007-05-23 Steve Ellcey <sje@cup.hp.com>
1302
1303 * Makefile.in: Regenerate.
1304 * configure: Regenerate.
1305 * aclocal.m4: Regenerate.
1306 * include/Makefile.in: Regenerate.
1307 * testsuite/Makefile.in: Regenerate.
1308
0e3a3bdb 13092007-05-10 Roman Zippel <zippel@linux-m68k.org>
1310
1311 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1312 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1313 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1314 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1315 (FFI_CLOSURES): Enable closure support.
1316
a53add80 13172007-05-10 Roman Zippel <zippel@linux-m68k.org>
1318
1319 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1320 * configure: Regenerate.
1321 * fficonfig.h.in: Regenerate.
1322 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1323 CFI_OFFSET,CFI_DEF_CFA): New macros.
1324 (ffi_call_SYSV): Add callframe annotation.
1325
bb3218e9 13262007-05-10 Roman Zippel <zippel@linux-m68k.org>
1327
1328 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1329 numerous test suite failures.
1330 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1331
5d0df87f 13322007-04-11 Paolo Bonzini <bonzini@gnu.org>
1333
1334 * Makefile.am (EXTRA_DIST): Bring up to date.
1335 * Makefile.in: Regenerate.
1336 * src/frv/eabi.S: Remove RCS keyword.
1337
ea4d299a 13382007-04-06 Richard Henderson <rth@redhat.com>
1339
1340 * configure.ac: Tidy target case.
1341 (HAVE_LONG_DOUBLE): Allow the target to override.
1342 * configure: Regenerate.
1343 * include/ffi.h.in: Don't define ffi_type_foo if
1344 LIBFFI_HIDE_BASIC_TYPES is defined.
1345 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1346 to ffi_type_double.
1347 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1348 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1349 (ffi_type_longdouble): Special case for Alpha. Don't define
1350 if long double == double.
1351
1352 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1353 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1354 (ffi_call, ffi_closure_osf_inner): Likewise.
1355 (ffi_closure_osf_inner): Likewise. Mark hidden.
1356 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1357 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1358 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1359 (load_table): Handle 128-bit long double.
1360
1361 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1362
770e494d 13632007-04-06 Tom Tromey <tromey@redhat.com>
1364
1365 PR libffi/31491:
1366 * README: Fixed bug in example.
1367
58606959 13682007-04-03 Jakub Jelinek <jakub@redhat.com>
1369
1370 * src/closures.c: Include sys/statfs.h.
1371 (_GNU_SOURCE): Define on Linux.
1372 (FFI_MMAP_EXEC_SELINUX): Define.
1373 (selinux_enabled): New variable.
1374 (selinux_enabled_check): New function.
1375 (is_selinux_enabled): Define.
1376 (dlmmap): Use it.
1377
7fed151e 13782007-03-24 Uros Bizjak <ubizjak@gmail.com>
1379
1380 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1381 Use 'volatile float sum' to create sum of floats to avoid false
1382 negative due to excess precision on ix86 targets.
81b376ae 1383 (main): Ditto.
7fed151e 1384
8fb4628d 13852007-03-08 Alexandre Oliva <aoliva@redhat.com>
1386
1387 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1388 patch.
1389 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1390
f20fe604 13912007-03-07 Alexandre Oliva <aoliva@redhat.com>
1392
1393 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1394 (ffi_prep_closure_loc): New.
1395 (ffi_prep_raw_closure_loc): New.
1396 (ffi_prep_java_raw_closure_loc): New.
1397 * src/closures.c: New file.
1398 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1399 Replace sflags with exec_offset.
1400 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1401 sub_segment_exec_offset): New macros.
1402 (get_segment_flags, set_segment_flags, check_segment_merge): New
1403 macros.
1404 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1405 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1406 destroy_mspace): Use new macros.
1407 (sys_alloc): Silence warning.
1408 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1409 * Makefile.in: Rebuilt.
1410 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1411 terms of ffi_prep_closure_loc.
1412 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1413 from...
1414 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1415 renamed version.
1416 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1417 adjusted from...
1418 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1419 the renamed version.
1420 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1421 (ffi_prep_closure): ... this.
1422 * src/pa/ffi.c: Likewise.
1423 * src/cris/ffi.c: Likewise. Adjust.
1424 * src/frv/ffi.c: Likewise.
1425 * src/ia64/ffi.c: Likewise.
1426 * src/mips/ffi.c: Likewise.
1427 * src/powerpc/ffi_darwin.c: Likewise.
1428 * src/s390/ffi.c: Likewise.
1429 * src/sh/ffi.c: Likewise.
1430 * src/sh64/ffi.c: Likewise.
1431 * src/sparc/ffi.c: Likewise.
1432 * src/x86/ffi64.c: Likewise.
1433 * src/x86/ffi.c: Likewise.
1434 (FFI_INIT_TRAMPOLINE): Adjust.
1435 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1436 (ffi_prep_raw_closure): ... this.
1437 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1438 (ffi_prep_closure): ... this.
1439 (flush_icache): Adjust.
1440
c3265eac 14412007-03-07 Alexandre Oliva <aoliva@redhat.com>
1442
1443 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1444 Lea's malloc.
1445
0ce4a5ce 14462007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1447
1448 * Makefile.am: Add dummy install-pdf target.
1449 * Makefile.in: Regenerate
1450
45444175 14512007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1452
1453 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1454 ffi_closure_helper_SYSV): Add long double handling.
1455
1df875cc 14562007-02-02 Jakub Jelinek <jakub@redhat.com>
1457
1458 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1459 immediately after bctrl instruction.
1460
27c4194f 14612007-01-18 Alexandre Oliva <aoliva@redhat.com>
1462
1463 * Makefile.am (all-recursive, install-recursive,
1464 mostlyclean-recursive, clean-recursive, distclean-recursive,
1465 maintainer-clean-recursive): Add missing targets.
1466 * Makefile.in: Rebuilt.
1467
eb6316a2 14682006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1469
1470 * configure.ac: Add TARGET for x86_64-*-darwin*.
1471 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1472 for X86_DARWIN.
1473 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1474 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1475 * configure: Regenerate.
1476 * Makefile.in: Regenerate.
1477 * include/Makefile.in: Regenerate.
1478 * testsuite/Makefile.in: Regenerate.
1479 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1480 ffi_call only.
1481
75a99c5e 14822006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1483
1484 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1485 Makefile.am.
1486
a39e926c 14872006-10-31 Geoffrey Keating <geoffk@apple.com>
c8dfb8ef 1488
1489 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1490 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1491 Darwin.
1492 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1493 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1494
eb819ecd 14952006-10-10 Paolo Bonzini <bonzini@gnu.org>
1496 Sandro Tolaini <tolaini@libero.it>
1497
81b376ae 1498 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
eb819ecd 1499 conditional.
1500 * configure: Regenerated.
1501 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1502 (EXTRA_DIST): Add src/x86/darwin.S.
1503 * Makefile.in: Regenerated.
1504 * include/Makefile.in: Regenerated.
1505 * testsuite/Makefile.in: Regenerated.
1506
1507 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1508 X86_WIN32, and additionally align stack to 16 bytes.
1509 * src/x86/darwin.S: New, based on sysv.S.
1510 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1511
23d9d417 15122006-09-12 David Daney <ddaney@avtrex.com>
1513
1514 PR libffi/23935
1515 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1516 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1517 * aclocal.m4: Regenerated for automake 1.9.6.
1518 * Makefile.in: Regenerated.
1519 * include/Makefile.in: Regenerated.
1520 * testsuite/Makefile.in: Regenerated.
1521
ec045626 15222006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1523
1524 * include/ffi_common.h (struct): Revert accidental commit.
1525
b9463909 15262006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1527
1528 * include/ffi_common.h: Remove lint directives.
1529 * include/ffi.h.in: Likewise.
1530
c695c024 15312006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1532
1533 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1534 for 32-bit architectures.
1535 * testsuite/libffi.call/return_ul.c: New test case.
1536
3a9d2a82 15372006-07-19 David Daney <ddaney@avtrex.com>
1538
1539 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1540 xfail remains for mips64.
1541
8f16e3ad 15422006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1543
1544 * Makefile.am: Add install-html target. Add install-html to .PHONY
1545 * Makefile.in: Regenerate.
1546 * aclocal.m4: Regenerate.
1547 * include/Makefile.in: Regenerate.
1548 * testsuite/Makefile.in: Regenerate.
1549
db951627 15502006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1551
1552 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1553 stack slot.
1554
9a333c86 15552006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1556
1557 * README: Remove notice about 'Crazy Comments'.
1558 * src/debug.c: Remove lint directives. Cleanup white spaces.
1559 * src/java_raw_api.c: Likewise.
1560 * src/prep_cif.c: Likewise.
1561 * src/raw_api.c: Likewise.
1562 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1563 to the testsuite.
1564 * src/arm/ffi.c: Remove lint directives.
1565 * src/m32r/ffi.c: Likewise.
1566 * src/pa/ffi.c: Likewise.
1567 * src/powerpc/ffi.c: Likewise.
1568 * src/powerpc/ffi_darwin.c: Likewise.
1569 * src/sh/ffi.c: Likewise.
1570 * src/sh64/ffi.c: Likewise.
1571 * src/x86/ffi.c: Likewise.
1572 * testsuite/libffi.call/float2.c: Likewise.
1573 * testsuite/libffi.call/promotion.c: Likewise.
1574 * testsuite/libffi.call/struct1.c: Likewise.
1575
7610fe2b 15762006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1577
1578 * src/pa/hpux32.S: Correct unwind offset calculation for
1579 ffi_closure_pa32.
1580 * src/pa/linux.S: Likewise.
1581
5adc6320 15822006-04-12 James E Wilson <wilson@specifix.com>
1583
1584 PR libgcj/26483
1585 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1586 (hfa_type_load): Call stf_spill.
1587 (hfa_type_store): Call ldf_fill.
1588 (ffi_call): Adjust calls to above routines. Add local temps for
1589 macro result.
7610fe2b 1590
d35f63a9 15912006-04-10 Matthias Klose <doko@debian.org>
1592
1593 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1594 directory names containing underscores.
1595
2d3dcfaa 15962006-04-07 James E Wilson <wilson@specifix.com>
1597
1598 * testsuite/libffi.call/float4.c: New testcase.
1599
1a82d6f6 16002006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1601 Andreas Tobler <a.tobler@schweiz.ch>
1602
1603 * Makefile.am: Add PA_HPUX port.
1604 * Makefile.in: Regenerate.
1605 * include/Makefile.in: Likewise.
1606 * testsuite/Makefile.in: Likewise.
1607 * configure.ac: Add PA_HPUX rules.
1608 * configure: Regenerate.
1609 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1610 Add PA_HPUX and PA64_HPUX.
1611 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1612 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1613 (FFI_TYPE_SMALL_STRUCT2): Define.
1614 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1615 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1616 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1617 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1618 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1619 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1620 * src/pa/ffi.c (ROUND_DOWN): Delete.
1621 (fldw, fstw, fldd, fstd): Use '__asm__'.
1622 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1623 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1624 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1625 Simplify incrementing of stack slot variable. Change type of local
1626 'n' to unsigned int.
1627 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1628 double on PA_HPUX.
1629 (ffi_prep_cif_machdep): Likewise.
1630 (ffi_call): Likewise.
1631 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1632 return type to ffi_status. Simplify incrementing of stack slot
1633 variable. Only copy floating point argument registers when PA_LINUX
1634 is true. Reformat debug statement.
1635 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1636 FFI_TYPE_SMALL_STRUCT8.
1637 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1638 declaration.
1639 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1640 Add nops to cache flush. Add trampoline for PA_HPUX.
1641 * src/pa/hpux32.S: New file.
1642 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1643 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1644 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1645 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1646 argument type checks so that common argument types appear first.
1647 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1648 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1649
7f5ddc97 16502006-03-24 Alan Modra <amodra@bigpond.net.au>
1651
1652 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1653 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1654 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1655 FFI_TYPE_LONGDOUBLE.
1656 (ffi_prep_args64): Assert using IBM extended double.
1657 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1658 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1659 (ffi_call): Handle FFI_LINUX.
1660 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1661 gpr3 return pointer as for struct return. Handle FFI_LINUX
1662 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1663 unnecessarily.
1664 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1665 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1666 Don't use r6 as pointer to results, instead use sp offset. Don't
1667 make a special call to load lr with case table address, instead
1668 use offset from previous call.
1669 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1670 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1671 return.
1672
cd10a5a7 16732006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1674
1675 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1676 passed with FP registers correctly.
1677 (ffi_closure_helper_SYSV): Likewise.
1678 * src/sh64/sysv.S: Likewise.
1679
a07533b6 16802006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1681
1682 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1683 args and userdata unused.
1684 (closure_test_fn1): Mark cif and userdata unused.
1685 (main): Remove unused res.
1686
bdc6f488 16872006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1688
1689 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1690 -O2, -O3, -Os and the warning flags -W -Wall.
1691 * testsuite/libffi.special/special.exp: Likewise.
1692 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1693 unused parameter unused for gcc or else do nothing.
1694 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1695 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1696 and userdata unused.
1697 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1698 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1699 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1700 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1701 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1702 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1703 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1704 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1705 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1706 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1707 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1708 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1709 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1710 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1711 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1712 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1713 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1714 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1715 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1716 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1717 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1718 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1719 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1720 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1721 Likewise.
1722 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1723 Likewise.
1724 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1725 Likewise.
1726 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1727 void* to avoid compiler warning.
1728 (main): Likewise.
1729 (cls_struct_align_gn): Mark cif and userdata unused.
1730 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1731 Likewise.
1732 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1733 Likewise.
1734 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1735 Likewise.
1736 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1737 Likewise.
1738 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1739 Likewise.
1740 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1741 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1742 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1743 data unused.
1744 (main): Cast res_call to silence gcc.
1745 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1746 data unused.
1747 (main): Cast res_call to silence gcc.
1748 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1749 and data unused.
1750 (main): Cast res_call to silence gcc.
1751 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1752 data unused.
1753 (main): Cast res_call to silence gcc.
1754 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1755 data unused.
1756 (main): Cast res_call to silence gcc.
1757 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1758 and data unused.
1759 (main): Cast res_call to silence gcc.
1760 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1761 userdata unused.
1762 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1763 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1764 userdata unused.
1765 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1766 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1767 userdata unused.
1768 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1769 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1770 userdata unused.
1771 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1772 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1773 userdata unused.
1774 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1775 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1776 and userdata unused.
1777 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1778 userdata unused.
1779 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1780 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1781 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1782 Cleanup white spaces.
1783 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1784 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1785 cif and userdata unused.
1786 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1787 Likewise.
1788 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1789 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1790 formatters to silence gcc.
1791 (B_gn): Mark cif and userdata unused.
1792 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1793 unused.
1794 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1795 (B_gn): Mark cif and userdata unused.
1796 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1797 unused.
1798 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1799 (B_gn): Mark cif and userdata unused.
1800 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1801 unused.
1802 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1803 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1804 * testsuite/libffi.call/problem1.c (stub): Likewise.
1805 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1806 gcc.
1807 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1808 in the last commit for this test case in the test case itself.
1809 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1810 unused.
1811 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1812 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1813 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1814 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1815 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1816 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1817
5ab2e6e4 18182006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1819
1820 * src/sh/sysv.S: Fix register numbers in the FDE for
1821 ffi_closure_SYSV.
1822
849b9f91 18232006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1824
1825 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1826 declaration to avoid a false negative on ix86. See PR323.
1827
003f258f 18282006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1829
1830 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1831 and cast integer to void * if needed. Update the pointer to
1832 the FP register saved area correctly.
81b376ae 1833
9121151b 18342006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1835
1836 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1837 is fixed.
1838 * testsuite/libffi.call/nested_struct4.c: Likewise.
1839
019a3b64 18402006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1841
6cbfba8c 1842 * testsuite/libffi.call/return_dbl.c: New test case.
1843 * testsuite/libffi.call/return_dbl1.c: Likewise.
1844 * testsuite/libffi.call/return_dbl2.c: Likewise.
1845 * testsuite/libffi.call/return_fl.c: Likewise.
1846 * testsuite/libffi.call/return_fl1.c: Likewise.
1847 * testsuite/libffi.call/return_fl2.c: Likewise.
1848 * testsuite/libffi.call/return_fl3.c: Likewise.
1849 * testsuite/libffi.call/closure_fn6.c: Likewise.
1850
019a3b64 1851 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1852 definition.
1853 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1854 here to be used by other test cases too.
1855
1856 * testsuite/libffi.call/nested_struct10.c: New test case.
1857 * testsuite/libffi.call/nested_struct9.c: Likewise.
1858 * testsuite/libffi.call/nested_struct8.c: Likewise.
1859 * testsuite/libffi.call/nested_struct7.c: Likewise.
1860 * testsuite/libffi.call/nested_struct6.c: Likewise.
1861 * testsuite/libffi.call/nested_struct5.c: Likewise.
1862 * testsuite/libffi.call/nested_struct4.c: Likewise.
1863
946a8d62 18642006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1865
1866 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1867 * configure: Rebuilt.
1868
cb99d24e 18692006-01-18 Jakub Jelinek <jakub@redhat.com>
1870
1871 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1872 instead do the shifting inline.
1873 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1874 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1875 and 8 byte structs, for the remaining struct sizes don't call
1876 __lshrdi3, instead do the shifting inline.
1877
06bf2366 18782005-12-07 Thiemo Seufer <ths@networkno.de>
1879
1880 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1881 missing parentheses.
1882 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1883 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1884 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1885 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1886 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1887 FA_0_0_OFF2.
1888 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1889 endianness bugs.
1890 (ffi_prep_closure): Improve trampoline instruction scheduling.
1891 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1892
9a385484 18932005-12-03 Alan Modra <amodra@bigpond.net.au>
1894
1895 * src/powerpc/ffi.c: Formatting.
1896 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1897 (ffi_prep_args64): Likewise.
1898
193c2942 18992005-09-30 Geoffrey Keating <geoffk@apple.com>
1900
1901 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1902 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1903
d3eb23fb 19042005-09-26 Tom Tromey <tromey@redhat.com>
1905
1906 * testsuite/libffi.call/float1.c (value_type): New typedef.
1907 (CANARY): New define.
1908 (main): Check for result buffer overflow.
1909 * src/powerpc/linux64.S: Handle linux64 long double returns.
1910 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1911 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1912
5497bbdf 19132005-08-25 Alan Modra <amodra@bigpond.net.au>
1914
1915 PR target/23404
1916 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1917 homed fp args.
1918 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1919
33b4ca90 19202005-08-11 Jakub Jelinek <jakub@redhat.com>
1921
1922 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1923 (AH_BOTTOM): Add FFI_HIDDEN definition.
1924 * configure: Rebuilt.
1925 * fficonfig.h.in: Rebuilt.
1926 * src/powerpc/ffi.c (hidden): Remove.
1927 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1928 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1929 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1930 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1931 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1932 add FFI_HIDDEN to its prototype.
1933 (ffi_closure_SYSV_inner): New.
1934 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1935 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1936
967ec51b 19372005-08-10 Alfred M. Szmidt <ams@gnu.org>
1938
1939 PR libffi/21819:
1940 * configure: Rebuilt.
1941 * configure.ac: Handle i*86-*-gnu*.
1942
80207f81 19432005-08-09 Jakub Jelinek <jakub@redhat.com>
1944
1945 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1946 DW_CFA_offset_extended_sf rather than
1947 DW_CFA_GNU_negative_offset_extended.
1948 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1949
6a0e74dc 19502005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1951
1952 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1953 on sh3.
1954 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1955 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1956 partially on register.
1957 (ffi_closure_helper_SYSV): Likewise.
1958 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1959
cef3f52b 19602005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1961
1962 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1963 Remove empty line.
1964 * src/sh64/ffi.c (simple_type): Remove.
1965 (return_type): Handle small structures correctly.
1966 (ffi_prep_args): Likewise.
1967 (ffi_call): Likewise.
1968 (ffi_closure_helper_SYSV): Likewise.
1969 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1970 Emit position independent code if PIC and remove wrong datalabel
1971 prefixes from EH data.
1972
66c44239 19732005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1974
1975 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1976 * Makefile.in: Regenerate.
1977 * include/Makefile.in: Likewise.
1978 * testsuite/Makefile.in: Likewise.
1979 * configure.ac: Add POWERPC_FREEBSD rules.
1980 * configure: Regenerate.
1981 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1982 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1983 * src/powerpc/ffi.c: Add flags to handle small structure returns
1984 in ffi_call_SYSV.
1985 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1986 Aka FFI_SYSV.
1987 (ffi_closure_helper_SYSV): Likewise.
1988 * src/powerpc/ppc_closure.S: Add return types for small structures.
1989 * src/powerpc/sysv.S: Add bits to handle small structures for
1990 final SYSV 4 ABI.
1991
a21bd97f 19922005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1993
1994 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1995 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1996 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1997
c350327d 19982005-07-05 Randolph Chung <tausq@debian.org>
1999
2000 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
2001 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
2002 structures. Kill compilation warnings.
2003 (ffi_closure_inner_LINUX): Print return values as hex in debug
2004 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
2005 Properly handle 5-7 byte structure returns.
2006 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
2007 (FFI_TYPE_SMALL_STRUCT2): Remove.
2008 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
2009 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
2010 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
2011 (checksmst1, checksmst2): Remove.
2012 (checksmst3): Optimize handling of 3-byte struct returns.
2013 (checksmst567): Properly handle 5-7 byte struct returns.
2014
20750e69 20152005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2016
2017 PR libgcj/21943
2018 * src/mips/n32.S: Enforce PIC code.
2019 * src/mips/o32.S: Likewise.
2020
58841df8 20212005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2022
2023 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
2024 * configure: Regenerate.
81b376ae 2025
cc46fb30 20262005-06-01 Alan Modra <amodra@bigpond.net.au>
2027
2028 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
2029 to call ffi_closure_helper_SYSV. Append @local instead.
2030 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
2031
6e912c1e 20322005-05-17 Kelley Cook <kcook@gcc.gnu.org>
69738b31 2033
2034 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
2035 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
2036 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
2037 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
2038 include/Makefile.in, testsuite/Makefile.in: Regenerate.
81b376ae 2039
5f0a087e 20402005-05-09 Mike Stump <mrs@apple.com>
2041
2042 * configure: Regenerate.
2043
81734468 20442005-05-08 Richard Henderson <rth@redhat.com>
2045
2046 PR libffi/21285
2047 * src/alpha/osf.S: Update unwind into to match code.
2048
4e9df534 20492005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
2050 Richard Henderson <rth@redhat.com>
2051
2052 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
2053 bit 11 of flags.
2054 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
2055 (ffi_prep_closure): Set carry bit if sse-used flag set.
2056 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
2057 Only load sse registers if ssecount non-zero.
2058 (ffi_closure_unix64): Only save sse registers if carry set on entry.
2059
20602005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
6155bf60 2061
2062 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
2063 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
2064 * configure: Regenerate.
2065
643e4a19 20662005-04-20 Hans-Peter Nilsson <hp@axis.com>
2067
2068 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
2069 have Tcl8.3-compatible intermediate variable.
2070
81b376ae 20712005-04-18 Simon Posnjak <simon.posnjak@siol.net>
7e5823cf 2072 Hans-Peter Nilsson <hp@axis.com>
2073
2074 * Makefile.am: Add CRIS support.
2075 * configure.ac: Likewise.
942132a7 2076 * Makefile.in, configure, testsuite/Makefile.in,
2077 include/Makefile.in: Regenerate.
7e5823cf 2078 * src/cris: New directory.
2079 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
2080 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
f0a2eecf 2081
2082 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
2083 \r?\n in output tests.
2084
3bcf2cd5 20852005-04-12 Mike Stump <mrs@apple.com>
2086
2087 * configure: Regenerate.
2088
b7292ee2 20892005-03-30 Hans Boehm <Hans.Boehm@hp.com>
2090
2091 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
81b376ae 2092
3bdc9b0e 20932005-03-30 Steve Ellcey <sje@cup.hp.com>
2094
2095 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
2096 (ffi_sarg) Ditto.
2097 * src/ia64/unix.S (ffi_closure_unix): Extend gp
2098 to 64 bits in ILP32 mode.
2099 Load 64 bits even for short data.
2100
7abd8367 21012005-03-23 Mike Stump <mrs@apple.com>
2102
2103 * src/powerpc/darwin.S: Update for -m64 multilib.
2104 * src/powerpc/darwin_closure.S: Likewise.
2105
bbb07d35 21062005-03-21 Zack Weinberg <zack@codesourcery.com>
2107
2108 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
2109 Do not set tool_include_dir.
2110 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
2111 Regenerate.
2112 * include/Makefile.am: Set gcc_version and toollibffidir.
2113 * include/Makefile.in: Regenerate.
2114
e6fd470b 21152005-02-22 Andrew Haley <aph@redhat.com>
2116
2117 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2118 odd-numbered register pairs for 64-bit integer types.
2119
4f373876 21202005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2121
2122 PR libffi/20104
2123 * testsuite/libffi.call/return_ll1.c: New test case.
2124
c2cf854f 21252005-02-11 Janis Johnson <janis187@us.ibm.com>
2126
2127 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2128 * testsuite/libffi.call/float.c: Ditto.
2129 * testsuite/libffi.call/float2.c: Ditto.
2130 * testsuite/libffi.call/float3.c: Ditto.
2131
9e71f2c8 21322005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2133
2134 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2135
45e075b4 21362005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2137
2138 * testsuite/libffi.special/special.exp (cxx_options): Add
2139 -shared-libgcc.
2140
f587fca2 21412004-12-31 Richard Henderson <rth@redhat.com>
2142
2143 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2144 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2145 offset parameters with a type parameter; deduce size and structure
2146 alignment. Update all users.
2147
717c502e 21482004-12-31 Richard Henderson <rth@redhat.com>
2149
2150 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2151 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
bbb07d35 2152 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
717c502e 2153 into ffi_prep_closure.
2154 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2155 from scratch.
2156
4a60643d 21572004-12-27 Richard Henderson <rth@redhat.com>
2158
2159 * src/x86/unix64.S: Fix typo in unwind info.
2160
66d912b3 21612004-12-25 Richard Henderson <rth@redhat.com>
2162
2163 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2164 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2165 (merge_classes): Check for it.
2166 (SSE_CLASS_P): New.
2167 (classify_argument): Pass byte_offset by value; perform all updates
2168 inside struct case.
2169 (examine_argument): Add classes argument; handle
2170 X86_64_COMPLEX_X87_CLASS.
2171 (ffi_prep_args): Merge into ...
2172 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2173 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2174 (ffi_fill_return_value): Remove.
2175 (ffi_prep_closure): Remove dead assert.
2176 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2177 Rewrite to use struct register_args instead of va_list. Create
2178 flags for handling structure returns.
2179 * src/x86/unix64.S: Remove dead strings.
2180 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2181 stack frame with ffi_call. Handle structure returns properly.
2182 (float2sse, floatfloat2sse, double2sse): Remove.
2183 (sse2float, sse2double, sse2floatfloat): Remove.
2184 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2185 to handle structure returns properly.
2186
a8594486 21872004-12-08 David Edelsohn <edelsohn@gnu.org>
2188
2189 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2190 PICFLAG.
2191 * Makefile.in: Regenerated.
2192
14ac9144 21932004-12-02 Richard Sandiford <rsandifo@redhat.com>
2194
2195 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2196 * configure, aclocal.m4, Makefile.in: Regenerate.
2197 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2198
d081dd29 21992004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2200
2201 * configure: Regenerate for libtool change.
2202
a234d196 22032004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2204
2205 * configure: Regenerate for libtool reversion.
2206
7f93fe7a 22072004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2208
2209 * configure: Regenerate for libtool change.
2210
c0e31427 22112004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2212
2213 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2214
22152004-11-23 Richard Sandiford <rsandifo@redhat.com>
2216
2217 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2218 of jal. Use an absolute encoding for the frame information.
2219
22202004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2221
2222 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2223 * acinclude.m4: Delete logic for sincludes.
2224 * aclocal.m4, Makefile.in, configure: Regenerate.
2225 * include/Makefile: Likewise.
2226 * testsuite/Makefile: Likewise.
2227
68fa8e89 22282004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2229
2230 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2231 on a 8-byte boundary.
2232 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2233
c11820d9 22342004-10-27 Richard Earnshaw <rearnsha@arm.com>
2235
2236 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2237 long long values. Round stack allocation to a multiple of 8 bytes
2238 for ATPCS compatibility.
2239 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2240 names. Handle returning long long types. Add Thumb and interworking
2241 support. Improve soft-float code.
2242
03b2b968 22432004-10-27 Richard Earnshaw <rearnsha@arm.com>
2244
2245 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2246 (libffi_exit): New function.
2247 (libffi_init): Build the testglue wrapper if needed.
2248
9f234538 22492004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2250
2251 PR other/18138
2252 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2253
61dc6540 22542004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2255
2256 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2257
3b77fb00 22582004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2259
2260 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2261 * testsuite/libffi.call/float3.c: New test case.
2262
7c2fd652 22632004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2264
2265 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2266 the function returning a structure pointed with R2.
2267 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2268 the structure return value if T bit set. Emit position
2269 independent code and EH data if PIC.
2270
7e74b6d4 22712004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2272
2273 * Makefile.am: Add m32r support.
2274 * configure.ac: Likewise.
2275 * Makefile.in: Regenerate.
2276 * confiugre: Regenerate.
2277 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2278 (uint64, sint64, double, longdouble)
2279 * src/m32r: New directory.
2280 * src/m32r/ffi.c: New file.
2281 * src/m32r/sysv.S: Likewise.
2282 * src/m32r/ffitarget.h: Likewise.
2283
d601026d 22842004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2285
2286 * testsuite/libffi.call/negint.c: New test case.
2287
ec1bc4ab 22882004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2289
2290 PR libgcj/17465
2291 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2292 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2293 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2294 DYLD_LIBRARY_PATH.
2295
41067a17 22962004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2297
2298 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2299 * testsuite/libffi.call/promotion.c: Likewise.
2300 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2301 whitespaces.
2302 * testsuite/libffi.call/return_sc.c: Likewise.
2303 * testsuite/libffi.call/return_uc.c: Likewise.
2304
acbdb39a 23052004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2306
2307 * src/powerpc/darwin.S: Fix comments and identation.
2308 * src/powerpc/darwin_closure.S: Likewise.
2309
de7fad36 23102004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2311
2312 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2313 (ffi_prep_args): Handle longdouble arguments.
2314 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2315 longdouble.
2316 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2317 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2318 values.
2319 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2320 * src/types.c: Defined longdouble size and alignment for darwin.
2321
8bf7bfd1 23222004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2323
2324 * src/powerpc/aix.S: Remove whitespaces.
2325 * src/powerpc/aix_closure.S: Likewise.
2326 * src/powerpc/asm.h: Likewise.
2327 * src/powerpc/ffi.c: Likewise.
2328 * src/powerpc/ffitarget.h: Likewise.
2329 * src/powerpc/linux64.S: Likewise.
2330 * src/powerpc/linux64_closure.S: Likewise.
2331 * src/powerpc/ppc_closure.S: Likewise.
2332 * src/powerpc/sysv.S: Likewise.
2333
2e818f50 23342004-08-30 Anthony Green <green@redhat.com>
2335
2336 * Makefile.am: Add frv support.
2337 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2338 * configure.ac: Read configure.host.
2339 * configure.in: Read configure.host.
2340 * configure.host: New file. frv-elf needs libgloss.
2341 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
752da755 2342 alignment. This is needed to frv and shouldn't harm the others.
2e818f50 2343 * include/ffi_common.h (ALIGN_DOWN): New macro.
2344 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2345
d36998ee 23462004-08-24 David Daney <daney@avtrex.com>
2347
2348 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2349 * testsuite/libffi.call/closure_fn1.c: Likewise.
2350 * testsuite/libffi.call/closure_fn2.c Likewise.
2351 * testsuite/libffi.call/closure_fn3.c: Likewise.
2352 * testsuite/libffi.call/closure_fn4.c: Likewise.
2353 * testsuite/libffi.call/closure_fn5.c: Likewise.
2354 * testsuite/libffi.call/cls_18byte.c: Likewise.
2355 * testsuite/libffi.call/cls_19byte.c: Likewise.
2356 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2357 * testsuite/libffi.call/cls_20byte.c: Likewise.
2358 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2359 * testsuite/libffi.call/cls_24byte.c: Likewise.
2360 * testsuite/libffi.call/cls_2byte.c: Likewise.
2361 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2362 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2363 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2364 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2365 * testsuite/libffi.call/cls_4byte.c: Likewise.
2366 * testsuite/libffi.call/cls_64byte.c: Likewise.
2367 * testsuite/libffi.call/cls_6byte.c: Likewise.
2368 * testsuite/libffi.call/cls_7byte.c: Likewise.
2369 * testsuite/libffi.call/cls_8byte.c: Likewise.
2370 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2371 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2372 * testsuite/libffi.call/cls_align_double.c: Likewise.
2373 * testsuite/libffi.call/cls_align_float.c: Likewise.
2374 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2375 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2376 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2377 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2378 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2379 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2380 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2381 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2382 * testsuite/libffi.call/cls_double.c: Likewise.
2383 * testsuite/libffi.call/cls_float.c: Likewise.
2384 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2385 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2386 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2387 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2388 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2389 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2390 * testsuite/libffi.call/cls_schar.c: Likewise.
2391 * testsuite/libffi.call/cls_sint.c: Likewise.
2392 * testsuite/libffi.call/cls_sshort.c: Likewise.
2393 * testsuite/libffi.call/cls_uchar.c: Likewise.
2394 * testsuite/libffi.call/cls_uint.c: Likewise.
2395 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2396 * testsuite/libffi.call/cls_ushort.c: Likewise.
2397 * testsuite/libffi.call/nested_struct.c: Likewise.
2398 * testsuite/libffi.call/nested_struct1.c: Likewise.
2399 * testsuite/libffi.call/nested_struct2.c: Likewise.
2400 * testsuite/libffi.call/nested_struct3.c: Likewise.
2401 * testsuite/libffi.call/problem1.c: Likewise.
2402 * testsuite/libffi.special/unwindtest.cc: Likewise.
2403 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2404 to zero.
2405 * testsuite/libffi.call/cls_16byte.c: Likewise.
2406 * testsuite/libffi.call/cls_5byte.c: Likewise.
2407
29ac6c83 24082004-08-23 David Daney <daney@avtrex.com>
2409
2410 PR libgcj/13141
2411 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2412 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2413 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2414 parameters and return types.
2415 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2416 (ffi_prep_closure): Ditto.
2417 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2418 alignment calculations.
2419 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2420 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2421
4487a7e8 24222004-08-14 Casey Marshall <csm@gnu.org>
2423
2424 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2425 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2426 integer (O32 ABI only).
2427 (ffi_prep_closure): new function.
2428 (ffi_closure_mips_inner_O32): new function.
bbb07d35 2429 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
4487a7e8 2430 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2431 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2432 64 bit integers correctly.
2433 (ffi_closure_O32): new function.
2434 Added DWARF-2 unwind info for both functions.
2435
bb2d2363 24362004-08-10 Andrew Haley <aph@redhat.com>
2437
2438 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2439
cb8a7007 24402004-08-01 Robert Millan <robertmh@gnu.org>
2441
2442 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2443 * configure: Regenerate.
2444
e2f66690 24452004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2446
2447 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2448 and mmap() explicitly instead of relying on preset autoconf cache
2449 variables.
2450 * aclocal.m4: Regenerate.
2451 * configure: Regenerate.
2452
e395522d 24532004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2454
2455 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2456 (ffi_check_float_struct): Remove unused prototype.
2457
b589ba75 24582004-06-30 Geoffrey Keating <geoffk@apple.com>
2459
2460 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2461 character on Darwin, use '\n\t' instead.
2462
7b709c61 24632004-06-26 Matthias Klose <doko@debian.org>
2464
2465 * libtool-version: Fix typo in revision/age.
2466
06e2077b 24672004-06-17 Matthias Klose <doko@debian.org>
2468
2469 * libtool-version: New.
2470 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2471 * Makefile.in: Regenerate.
2472
4889e098 24732004-06-15 Paolo Bonzini <bonzini@gnu.org>
d947df39 2474
2475 * Makefile.am: Remove useless multilib rules.
2476 * Makefile.in: Regenerate.
2477 * aclocal.m4: Regenerate with automake 1.8.5.
2478 * configure.ac: Remove useless multilib configury.
2479 * configure: Regenerate.
2480
4889e098 24812004-06-15 Paolo Bonzini <bonzini@gnu.org>
b2191698 2482
2483 * .cvsignore: New file.
2484
6bc491fa 24852004-06-10 Jakub Jelinek <jakub@redhat.com>
2486
2487 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2488 fp_done.
2489 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2490 changed from 8.
2491
e1c1d65f 24922004-06-06 Sean McNeil <sean@mcneil.com>
2493
2494 * configure.ac: Add x86_64-*-freebsd* support.
2495 * configure: Regenerate.
2496
2be99030 24972004-04-26 Joe Buck <jbuck@welsh-buck.org>
2498
2499 Bug 15093
2500 * configure.ac: Test for existence of mmap and sys/mman.h before
2501 checking blacklist. Fix suggested by Jim Wilson.
2502 * configure: Regenerate.
2503
efefde65 25042004-04-26 Matt Austern <austern@apple.com>
2505
2506 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2507 FDE location.
2508 * src/powerpc/darwin_closure.S: Likewise.
bbb07d35 2509
5766842b 25102004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2511
2512 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2513 error. Reported by Thomas Heller <theller@python.net>.
2514 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2515 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2516
43935228 25172004-03-20 Matthias Klose <doko@debian.org>
2518
2519 * src/pa/linux.S: Fix typo.
2520
f35c338f 25212004-03-19 Matthias Klose <doko@debian.org>
2522
5766842b 2523 * Makefile.am: Update.
f35c338f 2524 * Makefile.in: Regenerate.
2525 * src/pa/ffi.h.in: Remove.
2526 * src/pa/ffitarget.h: New file.
2527
25282004-02-10 Randolph Chung <tausq@debian.org>
2529
2530 * Makefile.am: Add PA support.
2531 * Makefile.in: Regenerate.
2532 * include/Makefile.in: Regenerate.
2533 * configure.ac: Add PA target.
2534 * configure: Regenerate.
2535 * src/pa/ffi.c: New file.
2536 * src/pa/ffi.h.in: Add PA support.
2537 * src/pa/linux.S: New file.
2538 * prep_cif.c: Add PA support.
2539
021bf237 25402004-03-16 Hosaka Yuji <hos@tamanegi.org>
2541
2542 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2543 double.
2544 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2545 with ecif->cif->flags.
2546 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2547 with cif->flags.
2548 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2549 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2550 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2551 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2552
4a900809 25532004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2554
2555 * configure.in: Rename file to ...
2556 * configure.ac: ... this.
2557 * fficonfig.h.in: Regenerate.
2558 * Makefile.in: Regenerate.
2559 * include/Makefile.in: Regenerate.
2560 * testsuite/Makefile.in: Regenerate.
2561
d128584a 25622004-03-12 Matt Austern <austern@apple.com>
2563
2564 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2565 changes in EH format resulting from addition of linkonce support.
2566 * src/powerpc/darwin_closure.S: Likewise.
021bf237 2567
b375837d 25682004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2569 Paolo Bonzini <bonzini@gnu.org>
2570
2571 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2572 Remove VPATH. Remove rules for object files. Remove multilib support.
2573 (AM_CCASFLAGS): Add.
2574 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2575 (AC_PREREQ): Bump version to 2.59.
2576 (AC_INIT): Fill with version info and bug address.
2577 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2578 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2579 De-precious CC so that the right flags are passed down to multilibs.
2580 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2581 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2582 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2583 * configure: Rebuilt.
2584 * aclocal.m4: Likewise.
2585 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2586 * fficonfig.h.in: Likewise.
2587
52cee448 25882004-03-11 Andreas Schwab <schwab@suse.de>
2589
2590 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2591 arguments from fp registers only for the first 8 parameter slots.
2592 Don't convert a float parameter when passed in memory.
2593
ac09b42b 25942004-03-09 Hans-Peter Nilsson <hp@axis.com>
2595
2596 * configure: Regenerate for config/accross.m4 correction.
2597
497c7020 25982004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2599
2600 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2601 ecif->cif->bytes to bytes.
2602 (ffi_prep_cif_machdep): Add braces around nested if statement.
2603
0ad895fb 26042004-02-09 Alan Modra <amodra@bigpond.net.au>
2605
ad98ab60 2606 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2607
0ad895fb 2608 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2609 (ffi_closure_helper_LINUX64): Fix typo.
2610 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2611 for powerpc64-*-*.
2612 * testsuite/libffi.call/float.c: Likewise.
2613 * testsuite/libffi.call/float2.c: Likewise.
2614
de9c7178 26152004-02-08 Alan Modra <amodra@bigpond.net.au>
2616
2617 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2618 long double function return and long double arg handling.
2619 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2620 Use "end_pfr" instead of "nf". Correct long double handling.
2621 Localise "temp".
2622 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2623 return value.
2624 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2625 space for long double return value. Adjust stack frame and offsets.
2626 Load f2 long double return.
2627
3474a085 26282004-02-07 Alan Modra <amodra@bigpond.net.au>
2629
2630 * src/types.c: Use 16 byte long double for POWERPC64.
2631
071ff5d6 26322004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2633
2634 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2635 when the structure return address is passed in %o0.
2636 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2637 (ffi_v9_layout_struct): Align the field following a nested structure
2638 on a word boundary. Use memmove instead of memcpy.
2639 (ffi_call): Update call to ffi_V9_return_struct.
2640 (ffi_prep_closure): Define 'ctx' only for V8.
2641 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2642 and ffi_closure_sparc_inner_v9.
2643 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2644 Always skip the structure return address. For structures and long
2645 doubles, copy the argument directly.
2646 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2647 if required. Shift the maximum floating-point slot accordingly. For
2648 big structures, copy the argument directly; otherwise, left-justify the
2649 argument and call ffi_v9_layout_struct to lay out the structure on
2650 the stack.
2651 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2652 (ffi_closure_v8): Pass the structure return address. Update call to
2653 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2654 Skip the 'unimp' insn when returning long doubles and structures.
2655 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2656 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2657 FFI_TYPE_INT handling. Load structures both in integers and
2658 floating-point registers on return.
2659 * README: Update status of the SPARC port.
2660
24ee5f14 26612004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2662
2663 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2664 as of type ffi_arg.
2665 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2666
1e0ee4d8 26672004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2668
2669 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2670 value as of type ffi_arg, not unsigned int.
2671
c886ae3f 26722004-01-21 Michael Ritzert <ritzert@t-online.de>
2673
2674 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2675 of the LHS.
2676
fcd85db3 26772004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2678
2679 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2680 Solaris.
2681
dcf910e2 26822004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2683
2684 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2685 to void *.
2686
3d3042b8 26872003-12-10 Richard Henderson <rth@redhat.com>
2688
2689 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2690 size_t instead of int.
2691
dc36d4d1 26922003-12-04 Hosaka Yuji <hos@tamanegi.org>
2693
2694 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2695 * testsuite/libffi.call/many_win32.c (main): Replace variable
2696 int i with unsigned long ul.
2697
2698 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2699 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2700 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2701 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2702 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2703 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2704 * testsuite/libffi.call/cls_align_float.c: Likewise.
2705 * testsuite/libffi.call/cls_align_double.c: Likewise.
2706 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2707 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2708
f90351a5 27092003-12-02 Hosaka Yuji <hos@tamanegi.org>
2710
2711 PR other/13221
2712 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2713 Align arguments to 32 bits.
2714
1c4a2a48 27152003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2716
2717 PR other/13221
2718 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2719 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2720 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2721 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2722 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2723 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2724
2725 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2726
8c899267 27272003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2728
2729 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2730 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2731
857794c9 27322003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2733
2734 * Makefile.in: Rebuilt.
2735 * configure: Likewise.
2736 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2737 the right type.
2738
e7f352cc 27392003-11-21 Andreas Jaeger <aj@suse.de>
2740 Andreas Tobler <a.tobler@schweiz.ch>
2741
2742 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2743 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2744 * Makefile.in: Rebuilt.
2745 * aclocal.m4: Likewise.
2746 * configure: Likewise.
2747 * fficonfig.h.in: Likewise.
2748 * testsuite/lib/libffi-dg.exp: Add include dir.
2749 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2750 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2751 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2752 for ffi_closure if available.
2753 * testsuite/libffi.call/closure_fn1.c: Likewise.
2754 * testsuite/libffi.call/closure_fn2.c: Likewise.
2755 * testsuite/libffi.call/closure_fn3.c: Likewise.
2756 * testsuite/libffi.call/closure_fn4.c: Likewise.
2757 * testsuite/libffi.call/closure_fn5.c: Likewise.
2758 * testsuite/libffi.call/cls_12byte.c: Likewise.
2759 * testsuite/libffi.call/cls_16byte.c: Likewise.
2760 * testsuite/libffi.call/cls_18byte.c: Likewise.
2761 * testsuite/libffi.call/cls_19byte.c: Likewise.
2762 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2763 * testsuite/libffi.call/cls_20byte.c: Likewise.
2764 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2765 * testsuite/libffi.call/cls_24byte.c: Likewise.
2766 * testsuite/libffi.call/cls_2byte.c: Likewise.
2767 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2768 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2769 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2770 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2771 * testsuite/libffi.call/cls_4byte.c: Likewise.
2772 * testsuite/libffi.call/cls_5byte.c: Likewise.
2773 * testsuite/libffi.call/cls_64byte.c: Likewise.
2774 * testsuite/libffi.call/cls_6byte.c: Likewise.
2775 * testsuite/libffi.call/cls_7byte.c: Likewise.
2776 * testsuite/libffi.call/cls_8byte.c: Likewise.
2777 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2778 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2779 * testsuite/libffi.call/cls_double.c: Likewise.
2780 * testsuite/libffi.call/cls_float.c: Likewise.
2781 * testsuite/libffi.call/cls_schar.c: Likewise.
2782 * testsuite/libffi.call/cls_sint.c: Likewise.
2783 * testsuite/libffi.call/cls_sshort.c: Likewise.
2784 * testsuite/libffi.call/cls_uchar.c: Likewise.
2785 * testsuite/libffi.call/cls_uint.c: Likewise.
2786 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2787 * testsuite/libffi.call/cls_ushort.c: Likewise.
2788 * testsuite/libffi.call/nested_struct.c: Likewise.
2789 * testsuite/libffi.call/nested_struct1.c: Likewise.
2790 * testsuite/libffi.call/nested_struct2.c: Likewise.
2791 * testsuite/libffi.call/nested_struct3.c: Likewise.
2792 * testsuite/libffi.call/problem1.c: Likewise.
2793 * testsuite/libffi.special/unwindtest.cc: Likewise.
2794
71e96ac5 27952003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2796
2797 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2798
fd0a337b 27992003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2800
2801 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2802 Add -lgcc_s to additional flags.
2803
c2dfd8f6 28042003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2805
2806 * configure.in, include/Makefile.am: PR libgcj/11147, install
2807 the ffitarget.h header file in a gcc versioned and target
2808 dependent place.
2809 * configure: Regenerated.
2810 * Makefile.in, include/Makefile.in: Likewise.
2811 * testsuite/Makefile.in: Likewise.
2812
17c9ab63 28132003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2814
2815 * testsuite/libffi.call/closure_fn0.c: Print result and check
2816 with dg-output to make debugging easier.
2817 * testsuite/libffi.call/closure_fn1.c: Likewise.
2818 * testsuite/libffi.call/closure_fn2.c: Likewise.
2819 * testsuite/libffi.call/closure_fn3.c: Likewise.
2820 * testsuite/libffi.call/closure_fn4.c: Likewise.
2821 * testsuite/libffi.call/closure_fn5.c: Likewise.
2822 * testsuite/libffi.call/cls_12byte.c: Likewise.
2823 * testsuite/libffi.call/cls_16byte.c: Likewise.
2824 * testsuite/libffi.call/cls_18byte.c: Likewise.
2825 * testsuite/libffi.call/cls_19byte.c: Likewise.
2826 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2827 * testsuite/libffi.call/cls_20byte.c: Likewise.
2828 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2829 * testsuite/libffi.call/cls_24byte.c: Likewise.
2830 * testsuite/libffi.call/cls_2byte.c: Likewise.
2831 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2832 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2833 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2834 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2835 * testsuite/libffi.call/cls_4byte.c: Likewise.
2836 * testsuite/libffi.call/cls_5byte.c: Likewise.
2837 * testsuite/libffi.call/cls_64byte.c: Likewise.
2838 * testsuite/libffi.call/cls_6byte.c: Likewise.
2839 * testsuite/libffi.call/cls_7byte.c: Likewise.
2840 * testsuite/libffi.call/cls_8byte.c: Likewise.
2841 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2842 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2843 * testsuite/libffi.call/cls_double.c: Likewise.
2844 * testsuite/libffi.call/cls_float.c: Likewise.
2845 * testsuite/libffi.call/cls_schar.c: Likewise.
2846 * testsuite/libffi.call/cls_sint.c: Likewise.
2847 * testsuite/libffi.call/cls_sshort.c: Likewise.
2848 * testsuite/libffi.call/cls_uchar.c: Likewise.
2849 * testsuite/libffi.call/cls_uint.c: Likewise.
2850 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2851 * testsuite/libffi.call/cls_ushort.c: Likewise.
2852 * testsuite/libffi.call/problem1.c: Likewise.
2853
2854 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2855 static.
2856
a0b53ce0 28572003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2858
2859 * testsuite/libffi.call/cls_9byte2.c: New test case.
2860 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2861 * testsuite/libffi.call/cls_64byte.c: Likewise.
2862 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2863 * testsuite/libffi.call/cls_19byte.c: Likewise.
2864 * testsuite/libffi.call/cls_18byte.c: Likewise.
2865 * testsuite/libffi.call/closure_fn4.c: Likewise.
2866 * testsuite/libffi.call/closure_fn5.c: Likewise.
2867 * testsuite/libffi.call/cls_schar.c: Likewise.
2868 * testsuite/libffi.call/cls_sint.c: Likewise.
2869 * testsuite/libffi.call/cls_sshort.c: Likewise.
2870 * testsuite/libffi.call/nested_struct2.c: Likewise.
2871 * testsuite/libffi.call/nested_struct3.c: Likewise.
2872
ed0ed2e9 28732003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2874
2875 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2876 * testsuite/libffi.call/cls_uchar.c: Likewise.
2877 * testsuite/libffi.call/cls_uint.c: Likewise.
2878 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2879 * testsuite/libffi.call/cls_ushort.c: Likewise.
2880 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2881
61641cc0 28822003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2883
2884 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2885 the initialization.
2886
6a759884 28872003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2888
61641cc0 2889 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
6a759884 2890 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2891
4f388831 28922003-10-22 David Daney <ddaney@avtrex.com>
2893
2894 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2895 __attribute__((__mode__(__SI__))) and friends.
2896
b4891aa6 28972003-10-22 Andreas Schwab <schwab@suse.de>
2898
2899 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2900
3d37e05c 29012003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2902
2903 * configure.in: AC_LINK_FILES(ffitarget.h).
2904 * configure: Regenerate.
2905 * Makefile.in: Likewise.
2906 * include/Makefile.in: Likewise.
2907 * testsuite/Makefile.in: Likewise.
2908 * fficonfig.h.in: Likewise.
2909
29102003-10-21 Paolo Bonzini <bonzini@gnu.org>
2911 Richard Henderson <rth@redhat.com>
2912
2913 Avoid that ffi.h includes fficonfig.h.
2914
2915 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2916 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2917 (TARGET_SRC_MIPS_SGI): Removed.
2918 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2919 (MIPS_SGI): Removed.
2920 (CLEANFILES): Removed.
2921 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2922 targets.
2923 * acconfig.h: Removed.
2924 * configure.in: Compute sizeofs only for double and long double.
2925 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2926 into AC_DEFINE instead of using acconfig.h. Create
2927 include/ffitarget.h instead of include/fficonfig.h. Rename
2928 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2929 AC_DEFINE EH_FRAME_FLAGS.
2930 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2931 (hack_DATA): Add ffitarget.h.
2932 * include/ffi.h.in: Remove all system specific definitions.
2933 Declare raw API even if it is not installed, why bother?
2934 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2935 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2936 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2937 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2938 * include/ffi_common.h (bool): Do not define.
2939 (ffi_assert): Accept failed assertion.
2940 (ffi_type_test): Return void and accept file/line.
2941 (FFI_ASSERT): Pass stringized failed assertion.
2942 (FFI_ASSERT_AT): New macro.
2943 (FFI_ASSERT_VALID_TYPE): New macro.
2944 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2945 UINT64, SINT64): Define here with gcc's __attribute__ macro
2946 instead of in ffi.h
2947 (FLOAT32, ALIGN): Define here instead of in ffi.h
2948 * include/ffi-mips.h: Removed. Its content moved to
2949 src/mips/ffitarget.h after separating assembly and C sections.
2950 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2951 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2952 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2953 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2954 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2955 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2956 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2957 (ffi_type_test): Rewritten.
2958 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2959 FFI_ASSERT_VALID_TYPE.
2960 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2961 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2962 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2963 src/s390/ffitarget.h, src/sh/ffitarget.h,
2964 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2965 src/x86/ffitarget.h: New files.
2966 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2967 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2968 src/powerpc/aix.S, src/powerpc/darwin.S,
2969 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2970 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2971 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2972 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2973 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2974 include fficonfig.h
2975
2a5f9ffc 29762003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2977
2978 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2979 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2980
51c5f96c 29812003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2982
2983 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2984 Used when FFI_DEBUG = 1.
2985
ee46e78f 29862003-10-14 Alan Modra <amodra@bigpond.net.au>
2987
2988 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2989 and align.
2990
a3abe7fe 29912003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2992
2993 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2994 FFI_MIPS_O32 for O32 ABI.
2995
1ee4561e 29962003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2997
2998 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2999 SPARC64. Cleanup whitespaces.
3000
93421fe6 30012003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
3002
3003 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
3004 strongarm, xscale. Cleanup whitespaces.
3005 * testsuite/libffi.call/closure_fn1.c: Likewise.
3006 * testsuite/libffi.call/closure_fn2.c: Likewise.
3007 * testsuite/libffi.call/closure_fn3.c: Likewise.
3008 * testsuite/libffi.call/cls_12byte.c: Likewise.
3009 * testsuite/libffi.call/cls_16byte.c: Likewise.
3010 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3011 * testsuite/libffi.call/cls_20byte.c: Likewise.
3012 * testsuite/libffi.call/cls_24byte.c: Likewise.
3013 * testsuite/libffi.call/cls_2byte.c: Likewise.
3014 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3015 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3016 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3017 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3018 * testsuite/libffi.call/cls_4byte.c: Likewise.
3019 * testsuite/libffi.call/cls_5byte.c: Likewise.
3020 * testsuite/libffi.call/cls_6byte.c: Likewise.
3021 * testsuite/libffi.call/cls_7byte.c: Likewise.
3022 * testsuite/libffi.call/cls_8byte.c: Likewise.
3023 * testsuite/libffi.call/cls_double.c: Likewise.
3024 * testsuite/libffi.call/cls_float.c: Likewise.
3025 * testsuite/libffi.call/cls_uchar.c: Likewise.
3026 * testsuite/libffi.call/cls_uint.c: Likewise.
3027 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3028 * testsuite/libffi.call/cls_ushort.c: Likewise.
3029 * testsuite/libffi.call/nested_struct.c: Likewise.
3030 * testsuite/libffi.call/nested_struct1.c: Likewise.
3031 * testsuite/libffi.call/problem1.c: Likewise.
3032 * testsuite/libffi.special/unwindtest.cc: Likewise.
3033 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
3034
8a94dd75 30352003-09-18 David Edelsohn <edelsohn@gnu.org>
3036
3037 * src/powerpc/aix.S: Cleanup whitespaces.
3038 * src/powerpc/aix_closure.S: Likewise.
3039
bd1f3533 30402003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3041
3042 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
3043 * src/powerpc/darwin_closure.S: Likewise.
3044 * src/powerpc/ffi_darwin.c: Likewise.
3045
7a142533 30462003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3047 David Edelsohn <edelsohn@gnu.org>
3048
3049 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
3050 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
3051 parameter stack.
3052 * src/powerpc/darwin_closure.S: Likewise.
3053 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
3054 according to the Darwin/AIX ABI.
3055 (ffi_prep_cif_machdep): Likewise.
3056 (ffi_closure_helper_DARWIN): Likewise.
3057 Remove the outgoing parameter stack logic. Simplify the evaluation
3058 of the different CASE types.
3059 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
3060 statement in the trampoline code.
3061
d80727ae 30622003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
3063
3064 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
3065 for the register size.
3066 (ffi_closure_helper_SYSV): Handle the structure return value
3067 address correctly.
3068 (ffi_closure_helper_SYSV): Return the appropriate type when
3069 the registers are used for the structure return value.
3070 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
3071 the 64-bit return value. Update copyright years.
3072
47fab6c3 30732003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3074
3075 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
3076 srcdir for ffi_mips.h.
3077
f397026a 30782003-09-12 Alan Modra <amodra@bigpond.net.au>
3079
3080 * src/prep_cif.c (initialize_aggregate): Include tail padding in
3081 structure size.
3082 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
3083 placement of float result.
3084 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
3085 cast of "resp" for big-endian 64 bit machines.
3086
341b7dcf 30872003-09-11 Alan Modra <amodra@bigpond.net.au>
3088
3089 * src/types.c (double, longdouble): Merge identical SH and ARM
3090 typedefs, and add POWERPC64.
3091 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
3092 struct split over gpr and rest.
3093 (ffi_prep_cif_machdep): Correct intarg_count for structures.
3094 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
3095
06855f3a 30962003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
3097
3098 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
3099 passing correctly.
c886ae3f 3100
9165b0af 31012003-09-09 Alan Modra <amodra@bigpond.net.au>
3102
3103 * configure: Regenerate.
3104
98d58335 31052003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3106
3107 * Makefile.am: Remove build rules for ffitest.
3108 * Makefile.in: Rebuilt.
c886ae3f 3109
98d58335 31102003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
0262ee1f 3111
3112 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3113 about implicit declaration of abort().
3114
3743c332 31152003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3116
3117 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3118 * Makefile.in: Rebuilt.
3119 * configure.in: Add dejagnu test framework.
3120 * configure: Rebuilt.
3121
3122 * testsuite/Makefile.am: New file.
3123 * testsuite/Makefile.in: Built
c886ae3f 3124 * testsuite/lib/libffi-dg.exp: New file.
3743c332 3125 * testsuite/config/default.exp: Likewise.
3126 * testsuite/libffi.call/call.exp: Likewise.
3127 * testsuite/libffi.call/ffitest.h: Likewise.
3128 * testsuite/libffi.call/closure_fn0.c: Likewise.
3129 * testsuite/libffi.call/closure_fn1.c: Likewise.
3130 * testsuite/libffi.call/closure_fn2.c: Likewise.
3131 * testsuite/libffi.call/closure_fn3.c: Likewise.
3132 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3133 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3134 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3135 * testsuite/libffi.call/cls_2byte.c: Likewise.
3136 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3137 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3138 * testsuite/libffi.call/cls_4byte.c: Likewise.
3139 * testsuite/libffi.call/cls_5byte.c: Likewise.
3140 * testsuite/libffi.call/cls_6byte.c: Likewise.
3141 * testsuite/libffi.call/cls_7byte.c: Likewise.
3142 * testsuite/libffi.call/cls_8byte.c: Likewise.
3143 * testsuite/libffi.call/cls_12byte.c: Likewise.
3144 * testsuite/libffi.call/cls_16byte.c: Likewise.
3145 * testsuite/libffi.call/cls_20byte.c: Likewise.
3146 * testsuite/libffi.call/cls_24byte.c: Likewise.
3147 * testsuite/libffi.call/cls_double.c: Likewise.
3148 * testsuite/libffi.call/cls_float.c: Likewise.
3149 * testsuite/libffi.call/cls_uchar.c: Likewise.
3150 * testsuite/libffi.call/cls_uint.c: Likewise.
3151 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3152 * testsuite/libffi.call/cls_ushort.c: Likewise.
3153 * testsuite/libffi.call/float.c: Likewise.
3154 * testsuite/libffi.call/float1.c: Likewise.
3155 * testsuite/libffi.call/float2.c: Likewise.
3156 * testsuite/libffi.call/many.c: Likewise.
3157 * testsuite/libffi.call/many_win32.c: Likewise.
3158 * testsuite/libffi.call/nested_struct.c: Likewise.
3159 * testsuite/libffi.call/nested_struct1.c: Likewise.
3160 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3161 * testsuite/libffi.call/problem1.c: Likewise.
3162 * testsuite/libffi.call/promotion.c: Likewise.
3163 * testsuite/libffi.call/return_ll.c: Likewise.
3164 * testsuite/libffi.call/return_sc.c: Likewise.
3165 * testsuite/libffi.call/return_uc.c: Likewise.
3166 * testsuite/libffi.call/strlen.c: Likewise.
3167 * testsuite/libffi.call/strlen_win32.c: Likewise.
3168 * testsuite/libffi.call/struct1.c: Likewise.
3169 * testsuite/libffi.call/struct2.c: Likewise.
3170 * testsuite/libffi.call/struct3.c: Likewise.
3171 * testsuite/libffi.call/struct4.c: Likewise.
3172 * testsuite/libffi.call/struct5.c: Likewise.
3173 * testsuite/libffi.call/struct6.c: Likewise.
3174 * testsuite/libffi.call/struct7.c: Likewise.
3175 * testsuite/libffi.call/struct8.c: Likewise.
c886ae3f 3176 * testsuite/libffi.call/struct9.c: Likewise.
3743c332 3177 * testsuite/libffi.special/special.exp: New file.
3178 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3179 * testsuite/libffi.special/unwindtest.cc: Likewise.
3180
c886ae3f 3181
3a845f75 31822003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3183
3184 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3185 copyright years.
3186
4637dfa6 31872003-08-02 Alan Modra <amodra@bigpond.net.au>
3188
3189 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3190 structure passing.
3191 (ffi_closure_helper_LINUX64): Likewise.
3192 * src/powerpc/linux64.S: Remove code writing to parm save area.
3193 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3194 address in lr from ffi_closure_helper_LINUX64 call to calculate
3195 table address. Optimize function tail.
3196
8bcdc7f0 31972003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3198
3199 * src/sparc/ffi.c: Handle all floating point registers.
3200 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3201
3d6cc549 32022003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3203
3204 * README: Note that libffi is not part of GCC. Update the project
3205 URL and status.
3206
9b770b5a 32072003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3208
3209 * src/powerpc/ppc_closure.S: Include ffi.h.
3210
d46ab9f3 32112003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3212
3213 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3214 Use C style comments.
3215
81f536a3 32162003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3217
3218 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3219 * Makefile.in: Regenerate.
3220 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3221 * configure: Regenerate.
3222 * include/ffi.h.in: Add SHmedia support.
3223 * src/sh64/ffi.c: New file.
3224 * src/sh64/sysv.S: New file.
3225
1da888ec 32262003-05-16 Jakub Jelinek <jakub@redhat.com>
3227
3228 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3229 should be read-only.
3230 * configure: Rebuilt.
3231 * fficonfig.h.in: Rebuilt.
3232 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3233 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3234 * src/powerpc/linux64.S: Likewise.
3235 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3236 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3237 if -fpic/-fPIC/-mrelocatable.
3238 * src/powerpc/powerpc_closure.S: Likewise.
3239 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3240 #write in .eh_frame flags.
3241 * src/sparc/v9.S: Likewise.
3242 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3243 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3244 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3245
f3bd7a2c 32462003-05-07 Jeff Sturm <jsturm@one-point.com>
3247
3248 Fixes PR bootstrap/10656
3249 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3250 support for .register pseudo-op.
3251 * src/sparc/v8.S: Use it.
3252 * fficonfig.h.in: Rebuilt.
3253 * configure: Rebuilt.
3254
3647950d 32552003-04-18 Jakub Jelinek <jakub@redhat.com>
3256
3257 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3258 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3259 Make it the default on POWERPC64.
3260 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3261 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3262 * configure: Rebuilt.
3263 * src/powerpc/ffi.c (hidden): Define.
3264 (ffi_prep_args_SYSV): Renamed from
3265 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3266 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3267 ASM_NEEDS_REGISTERS64): New.
3268 (ffi_prep_args64): New function.
3269 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3270 (ffi_call): Likewise.
3271 (ffi_prep_closure): Likewise.
3272 (flush_icache): Surround by #ifndef POWERPC64.
3273 (ffi_dblfl): New union type.
3274 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3275 (ffi_closure_helper_LINUX64): New function.
3276 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3277 __powerpc64__.
3278 * src/powerpc/sysv.S: Likewise.
3279 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3280 * src/powerpc/linux64.S: New file.
3281 * src/powerpc/linux64_closure.S: New file.
3282 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3283 src/powerpc/linux64_closure.S.
3284 (TARGET_SRC_POWERPC): Likewise.
3285
3286 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3287 closure_test_fn3): Fix result printing on big-endian 64-bit
3288 machines.
3289 (main): Print tst2_arg instead of uninitialized tst2_result.
3290
3291 * src/ffitest.c (main): Hide what closure pointer really points to
3292 from the compiler.
3293
1636e920 32942003-04-16 Richard Earnshaw <rearnsha@arm.com>
3295
3296 * configure.in (arm-*-netbsdelf*): Add configuration.
3297 (configure): Regenerated.
3298
e663564a 32992003-04-04 Loren J. Rittle <ljrittle@acm.org>
3300
3301 * include/Makefile.in: Regenerate.
3302
02f5d4c9 33032003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3304
3305 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3306 bit mode.
3307 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3308 Receive closure pointer through parameter, read args using
3309 __builtin_dwarf_cfa.
3310 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3311
b8e27232 33122003-03-12 Andreas Schwab <schwab@suse.de>
3313
3314 * configure.in: Avoid trailing /. in toolexeclibdir.
3315 * configure: Rebuilt.
3316
5fee68c6 33172003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3318
bbb07d35 3319 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
5fee68c6 3320
ecadab3f 33212003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3322
3323 * libffi/src/powerpc/darwin_closure.S:
3324 Fix alignement bug, allocate 8 bytes for the result.
c886ae3f 3325 * libffi/src/powerpc/aix_closure.S:
ecadab3f 3326 Likewise.
3327 * libffi/src/powerpc/ffi_darwin.c:
3328 Update stackframe description for aix/darwin_closure.S.
c886ae3f 3329
a0294641 33302003-02-06 Jakub Jelinek <jakub@redhat.com>
3331
3332 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3333 attribute.
3334
624476ed 33352003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3336 Andreas Schwab <schwab@suse.de>
3337
bbb07d35 3338 * configure.in: Adjust command to source config-ml.in to account
624476ed 3339 for changes to the libffi_basedir definition.
3340 (libffi_basedir): Remove ${srcdir} from value and include trailing
3341 slash if nonempty.
c886ae3f 3342
bbb07d35 3343 * configure: Regenerate.
624476ed 3344
1143cc9b 33452003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3346
3347 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3348
967ae8fb 33492003-01-28 Andrew Haley <aph@redhat.com>
3350
3351 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3352 * src/x86/ffi64.c (ffi_prep_closure): New.
3353 (ffi_closure_UNIX64_inner): New.
3354 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3355
c8d4fa8c 33562003-01-27 Alexandre Oliva <aoliva@redhat.com>
3357
3358 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3359 Remove USE_LIBDIR conditional.
3360 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3361 * Makefile.in, configure: Rebuilt.
3362
2dd7852c 33632003-01027 David Edelsohn <edelsohn@gnu.org>
3364
3365 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3366 * Makefile.in: Regenerate.
3367
008eadcd 33682003-01-22 Andrew Haley <aph@redhat.com>
3369
3370 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3371 unwind info.
3372
cb440058 33732003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3374
3375 * src/powerpc/darwin.S: Add unwind info.
3376 * src/powerpc/darwin_closure.S: Likewise.
3377
149417bc 33782003-01-14 Andrew Haley <aph@redhat.com>
3379
3380 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3381 (ffi_prep_cif_machdep): Likewise.
3382 * src/x86/unix64.S: Add unwind info.
3383
d4dcd973 33842003-01-14 Andreas Jaeger <aj@suse.de>
3385
3386 * src/ffitest.c (main): Only use ffi_closures if those are
3387 supported.
3388
a7898d0c 33892003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3390
3391 * libffi/src/ffitest.c
bbb07d35 3392 add closure testcases
a7898d0c 3393
33942003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3395
3396 * libffi/src/powerpc/ffi.c
bbb07d35 3397 fix alignment bug for float (4 byte aligned iso 8 byte)
a7898d0c 3398
4e844289 33992003-01-09 Geoffrey Keating <geoffk@apple.com>
3400
3401 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3402 * src/powerpc/darwin.S: Remove RCS version string.
3403
0ef3e9c5 34042003-01-03 Jeff Sturm <jsturm@one-point.com>
3405
3406 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3407 * src/ffitest.c (main): Use static storage for closure.
3408 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3409 * src/sparc/v8.S (ffi_closure_v8): New.
3410 * src/sparc/v9.S (ffi_closure_v9): New.
3411
acdc3fc3 34122002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3413
3414 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3415 enumeration for X86_WIN32.
3416 * src/x86/win32.S: Added ffi_call_STDCALL function
3417 definition.
3418 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3419 switch cases for recognising FFI_STDCALL and
3420 calling ffi_call_STDCALL if target is X86_WIN32.
3421 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3422 stdcall versions of the "my_strlen" and "many"
3423 test functions (for X86_WIN32).
3424 Added test cases to test stdcall invocation using
3425 these functions.
3426
805e22b2 34272002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3428
3429 * src/sh/sysv.S: Add DWARF2 unwind info.
3430
6d219176 34312002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3432
3433 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3434
4827afc9 34352002-11-26 Jim Wilson <wilson@redhat.com>
3436
3437 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3438
805e22b2 34392002-11-23 H.J. Lu <hjl@gnu.org>
3440
3441 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3442 Include ../config/accross.m4.
3443 * aclocal.m4; Rebuild.
3444 * configure: Likewise.
3445
e3a04290 34462002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3447
3448 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3449
be36f77d 34502002-11-11 DJ Delorie <dj@redhat.com>
3451
3452 * configure.in: Look for common files in the right place.
3453
dd985e54 34542002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3455
3456 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3457 raw data as _Jv_word values, not ffi_raw.
3458 (ffi_java_ptrarray_to_raw): Likewise.
3459 (ffi_java_rvalue_to_raw): New function.
3460 (ffi_java_raw_call): Call it.
3461 (ffi_java_raw_to_rvalue): New function.
3462 (ffi_java_translate_args): Call it.
3463 * src/ffitest.c (closure_test_fn): Interpret return value
3464 as ffi_arg, not int.
3465 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3466 FFI_TYPE_POINTER case.
3467 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3468 values extended to word size.
3469
4a18e5fa 34702002-10-02 Andreas Jaeger <aj@suse.de>
3471
3472 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3473
3d8ac088 34742002-10-01 Bo Thorsen <bo@smetana.suse.de>
3475
3476 * include/ffi.h.in: Fix i386 win32 compilation.
3477
e0aaecf3 34782002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3479
3480 * configure.in: Add s390x-*-linux-* target.
3481 * configure: Regenerate.
3482 * include/ffi.h.in: Define S390X for s390x targets.
3483 (FFI_CLOSURES): Define for s390/s390x.
3484 (FFI_TRAMPOLINE_SIZE): Likewise.
3485 (FFI_NATIVE_RAW_API): Likewise.
3486 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3487 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
805e22b2 3488 * src/s390/ffi.c: Major rework of existing code. Add support for
e0aaecf3 3489 s390x targets. Add closure support.
3490 * src/s390/sysv.S: Likewise.
3491
c3e9c301 34922002-09-29 Richard Earnshaw <rearnsha@arm.com>
3493
3494 * src/arm/sysv.S: Fix typo.
3495
d13886be 34962002-09-28 Richard Earnshaw <rearnsha@arm.com>
3497
3498 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3499 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3500 (ffi_call_SYSV): Handle soft-float.
3501
aaff811e 35022002-09-27 Bo Thorsen <bo@suse.de>
3503
3504 * include/ffi.h.in: Fix multilib x86-64 support.
3505
5f7912d8 35062002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3507
3508 * Makefile.am (all-multi): Fix multilib parallel build.
3509
ccda5e55 35102002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3511
3512 * configure.in (sh[34]*-*-linux*): Add brackets.
3513 * configure: Regenerate.
3514
b1e5d71d 35152002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3516
3517 * Makefile.am: Add SH support.
3518 * Makefile.in: Regenerate.
3519 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3520 * configure: Regenerate.
3521 * include/ffi.h.in: Add SH support.
3522 * src/sh/ffi.c: New file.
3523 * src/sh/sysv.S: New file.
3524 * src/types.c: Add SH support.
3525
42c71854 35262002-07-16 Bo Thorsen <bo@suse.de>
3527
3528 * src/x86/ffi64.c: New file that adds x86-64 support.
3529 * src/x86/unix64.S: New file that handles argument setup for
3530 x86-64.
3531 * src/x86/sysv.S: Don't use this on x86-64.
3532 * src/x86/ffi.c: Don't use this on x86-64.
3533 Remove unused vars.
3534 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3535 for x86-64.
3536 * src/ffitest.c (struct6): New test that tests a special case in
3537 the x86-64 ABI.
3538 (struct7): Likewise.
3539 (struct8): Likewise.
3540 (struct9): Likewise.
3541 (closure_test_fn): Silence warning about this when it's not used.
3542 (main): Add the new tests.
3543 (main): Fix a couple of wrong casts and silence some compiler warnings.
3544 * include/ffi.h.in: Add x86-64 ABI definition.
3545 * fficonfig.h.in: Regenerate.
3546 * Makefile.am: Add x86-64 support.
3547 * configure.in: Likewise.
3548 * Makefile.in: Regenerate.
3549 * configure: Likewise.
3550
35512002-06-24 Bo Thorsen <bo@suse.de>
3552
3553 * src/types.c: Merge settings for similar architectures.
3554 Add x86-64 sizes and alignments.
3555
35562002-06-23 Bo Thorsen <bo@suse.de>
3557
3558 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3559 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3560 * src/mips/ffi.c (ffi_prep_args): Likewise.
3561 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3562
72726889 35632002-07-18 H.J. Lu (hjl@gnu.org)
3564
3565 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3566 (libffi_la_SOURCES): Support MIPS_LINUX.
3567 (libffi_convenience_la_SOURCES): Likewise.
3568 * Makefile.in: Regenerated.
3569
3570 * configure.in (mips64*-*): Skip.
3571 (mips*-*-linux*): New.
3572 * configure: Regenerated.
3573
3574 * src/mips/ffi.c: Include <sgidefs.h>.
3575
e17e640b 35762002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3577
3578 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3579
b5d69d52 35802002-05-27 Roger Sayle <roger@eyesopen.com>
3581
3582 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3583
eadca1d5 35842002-05-27 Bo Thorsen <bo@suse.de>
3585
3586 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3587 fix formatting.
3588
2abdadc9 35892002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3590
3591 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3592 beginning of function (for older apple cc).
3593
0b045cf7 35942002-05-08 Alexandre Oliva <aoliva@redhat.com>
3595
3596 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3597 script entry, and set LD to it when configuring multilibs.
3598 * configure: Rebuilt.
3599
7f0e056d 36002002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3601
3602 * configure.in (sparc64-*-netbsd*): Add target.
3603 (sparc-*-netbsdelf*): Likewise.
3604 * configure: Regenerate.
3605
b19313f4 36062002-04-28 David S. Miller <davem@redhat.com>
3607
3608 * configure.in, configure: Fix SPARC test in previous change.
3609
ea1f83c5 36102002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3611
3612 * Makefile.am: Add Linux for S/390 support.
3613 * Makefile.in: Regenerate.
3614 * configure.in: Add Linux for S/390 support.
3615 * configure: Regenerate.
3616 * include/ffi.h.in: Add Linux for S/390 support.
3617 * src/s390/ffi.c: New file from libffi CVS tree.
3618 * src/s390/sysv.S: New file from libffi CVS tree.
3619
b40c762b 36202002-04-28 Jakub Jelinek <jakub@redhat.com>
3621
3622 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3623 %r_disp32().
3624 * src/sparc/v8.S: Use it.
3625 * src/sparc/v9.S: Likewise.
3626 * fficonfig.h.in: Rebuilt.
3627 * configure: Rebuilt.
3628
00f87a50 36292002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3630
3631 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3632 correctly.
3633 * src/ia64/unix.S: Add unwind information. Fix comments.
3634 Save sp in a way that's compatible with unwind info.
3635 (ffi_call_unix): Correctly restore sp in all cases.
3636 * src/ia64/ffi.c: Add, fix comments.
3637
a622ae9d 36382002-04-08 Jakub Jelinek <jakub@redhat.com>
3639
3640 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3641
c1d86bcd 36422002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3643
3644 * configure.in (alpha*-*-netbsd*): Add target.
3645 * configure: Regenerate.
3646
3faa4340 36472002-04-04 Jeff Sturm <jsturm@one-point.com>
3648
3649 * src/sparc/v8.S: Add unwind info.
3650 * src/sparc/v9.S: Likewise.
3651
ec66e65b 36522002-03-30 Krister Walfridsson <cato@df.lth.se>
3653
3654 * configure.in: Enable i*86-*-netbsdelf*.
3655 * configure: Rebuilt.
805e22b2 3656
9ee9948b 36572002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3658
3659 PR other/2620
3660 * src/mips/n32.s: Delete
3661 * src/mips/o32.s: Delete
3662
6c74b4f2 36632002-03-21 Loren J. Rittle <ljrittle@acm.org>
3664
3665 * configure.in: Enable alpha*-*-freebsd*.
3666 * configure: Rebuilt.
3667
9d0a2dd5 36682002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3669
3670 * Makefile.am: libfficonvenience -> libffi_convenience.
3671 * Makefile.in: Rebuilt.
805e22b2 3672
ee7750c2 3673 * Makefile.am: Define ffitest_OBJECTS.
3674 * Makefile.in: Rebuilt.
9d0a2dd5 3675
f6a16bd8 36762002-03-07 Andreas Tobler <toa@pop.agri.ch>
3677 David Edelsohn <edelsohn@gnu.org>
3678
3679 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3680 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3681 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3682 * Makefile.in: Regenerate.
3683 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3684 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3685 (flush_icache, flush_range): New functions.
3686 (ffi_closure_helper_DARWIN): New function.
3687 * src/powerpc/aix_closure.S: New file.
3688 * src/powerpc/darwin_closure.S: New file.
805e22b2 3689
fed8d18d 36902002-02-24 Jeff Sturm <jsturm@one-point.com>
3691
3692 * include/ffi.h.in: Add typedef for ffi_arg.
3693 * src/ffitest.c (main): Declare rint with ffi_arg.
805e22b2 3694
ddac7e16 36952002-02-21 Andreas Tobler <toa@pop.agri.ch>
3696
3697 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3698 number of GPRs for floating-point arguments.
3699
ed35b400 37002002-01-31 Anthony Green <green@redhat.com>
3701
3702 * configure: Rebuilt.
3703 * configure.in: Replace CHECK_SIZEOF and endian tests with
3704 cross-compiler friendly macros.
3705 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3706 macros.
3707
bf270768 37082002-01-18 David Edelsohn <edelsohn@gnu.org>
3709
3710 * src/powerpc/darwin.S (_ffi_call_AIX): New.
dc155f85 3711 * src/powerpc/aix.S (ffi_call_DARWIN): New.
bf270768 3712
faae36ce 37132002-01-17 David Edelsohn <edelsohn@gnu.org>
3714
3715 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3716 (TARGET_SRC_POWERPC_AIX): New.
3717 (POWERPC_AIX): New stanza.
3718 * Makefile.in: Regenerate.
3719 * configure.in: Add AIX case.
3720 * configure: Regenerate.
3721 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3722 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3723 size. Fix "long double" support.
3724 (ffi_call): Add FFI_AIX case.
3725 * src/powerpc/aix.S: New.
3726
b81d6fe1 37272001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3728
3729 Implement Darwin PowerPC ABI.
3730 * configure.in: Handle powerpc-*-darwin*.
3731 * Makefile.am: Set source files for POWERPC_DARWIN.
3732 * configure: Rebuilt.
3733 * Makefile.in: Rebuilt.
3734 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3735 POWERPC_DARWIN.
3736 * src/powerpc/darwin.S: New file.
3737 * src/powerpc/ffi_darwin.c: New file.
3738
9bccb7a3 37392001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3740
3741 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3742
ac1f371e 37432001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3744
3745 * src/x86/sysv.S: Avoid gas-only .balign directive.
3746 Use C style comments.
3747
e1e36390 37482001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3749
3750 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3751 Fixes PR bootstrap/3563.
3752
deb29e7a 37532001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3754
3755 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3756
8008de24 37572001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3758
3759 * configure.in: Recognize sparc*-sun-* host.
3760 * configure: Regenerate.
805e22b2 3761
5fc46419 37622001-06-06 Andrew Haley <aph@redhat.com>
3763
968b0d1e 3764 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
5fc46419 3765
49ff2ded 37662001-06-03 Andrew Haley <aph@redhat.com>
3767
968b0d1e 3768 * src/alpha/osf.S: Add unwind info.
3769 * src/powerpc/sysv.S: Add unwind info.
3770 * src/powerpc/ppc_closure.S: Likewise.
49ff2ded 3771
070883fe 37722000-05-31 Jeff Sturm <jsturm@one-point.com>
3773
3774 * configure.in: Fix AC_ARG_ENABLE usage.
3775 * configure: Rebuilt.
3776
09da9506 37772001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3778
3779 * configure.in: Remove warning about beta code.
3780 * configure: Rebuilt.
3781
37822001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3783
6c1b002f 3784 * src/ia64/unix.S: Restore stack pointer when returning from
3785 ffi_closure_UNIX.
3786 * src/ia64/ffi.c: Fix typo in comment.
3787
c4312026 37882001-04-18 Jim Wilson <wilson@redhat.com>
3789
3790 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3791 to eliminate RAW DV.
3792
db6ca0af 37932001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3794
3795 * Makefile.am: Make a libtool convenience library.
3796 * Makefile.in: Rebuilt.
3797
9461b350 37982001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3799
3800 * configure.in: Use different syntax for subdirectory creation.
3801 * configure: Rebuilt.
3802
a4cc90ca 38032001-03-27 Jon Beniston <jon@beniston.com>
3804
3805 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3806 * configure: Rebuilt.
3807 * Makefile.am: Added X86_WIN32 target support.
3808 * Makefile.in: Rebuilt.
3809
3810 * include/ffi.h.in: Added X86_WIN32 target support.
3811
3812 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3813 * src/types.c: Added X86_WIN32 target support.
3814
3815 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3816 stuff removed and made to work with CygWin's gas.
3817
21f4579d 38182001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3819
3820 * configure.in: Make target subdirectory in build dir.
3821 * Makefile.am: Override suffix based rules to specify correct output
3822 subdirectory.
3823 * Makefile.in: Rebuilt.
3824 * configure: Rebuilt.
3825
29183c57 38262001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
384eea1f 3827
3828 * src/powerpc/ppc_closure.S: New file.
3829 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3830 involving long long and register pairs.
3831 (ffi_prep_closure): New function.
3832 (flush_icache): Likewise.
3833 (ffi_closure_helper_SYSV): Likewise.
3834 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3835 (FFI_TRAMPOLINE_SIZE): Likewise.
3836 (FFI_NATIVE_RAW_API): Likewise.
3837 * Makefile.in: Rebuilt.
3838 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3839 (TARGET_SRC_POWERPC): Likewise.
3840
49604b5b 38412001-03-19 Tom Tromey <tromey@redhat.com>
3842
3843 * Makefile.in: Rebuilt.
3844 * Makefile.am (ffitest_LDFLAGS): New macro.
3845
c8fdcc07 38462001-03-02 Nick Clifton <nickc@redhat.com>
3847
3848 * include/ffi.h.in: Remove RCS ident string.
3849 * include/ffi_mips.h: Remove RCS ident string.
3850 * src/debug.c: Remove RCS ident string.
3851 * src/ffitest.c: Remove RCS ident string.
3852 * src/prep_cif.c: Remove RCS ident string.
3853 * src/types.c: Remove RCS ident string.
3854 * src/alpha/ffi.c: Remove RCS ident string.
3855 * src/alpha/osf.S: Remove RCS ident string.
3856 * src/arm/ffi.c: Remove RCS ident string.
3857 * src/arm/sysv.S: Remove RCS ident string.
3858 * src/mips/ffi.c: Remove RCS ident string.
3859 * src/mips/n32.S: Remove RCS ident string.
3860 * src/mips/o32.S: Remove RCS ident string.
3861 * src/sparc/ffi.c: Remove RCS ident string.
3862 * src/sparc/v8.S: Remove RCS ident string.
3863 * src/sparc/v9.S: Remove RCS ident string.
3864 * src/x86/ffi.c: Remove RCS ident string.
3865 * src/x86/sysv.S: Remove RCS ident string.
3866
8cb0f8ea 38672001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3868
3869 * include/ffi.h.in: Change sourceware.cygnus.com references to
3870 gcc.gnu.org.
3871
18f704a9 38722000-12-09 Richard Henderson <rth@redhat.com>
3873
3874 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3875 (ffi_closure_osf_inner): Index rather than increment avalue
3876 and arg_types. Give ffi_closure_osf the raw return value type.
3877 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3878 promotion.
3879
38802000-12-07 Richard Henderson <rth@redhat.com>
fdf04128 3881
3882 * src/raw_api.c (ffi_translate_args): Fix typo.
3883 (ffi_prep_closure): Likewise.
3884
3885 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3886 FFI_TRAMPOLINE_SIZE.
3887 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3888 cif->bytes for new ffi_call_osf implementation.
3889 (ffi_prep_args): Absorb into ...
3890 (ffi_call): ... here. Do all stack allocation here and
3891 avoid a callback function.
3892 (ffi_prep_closure, ffi_closure_osf_inner): New.
3893 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3894 (ffi_closure_osf): New.
3895
652f285b 38962000-09-10 Alexandre Oliva <aoliva@redhat.com>
3897
4bfa769b 3898 * config.guess, config.sub, install-sh: Removed.
3899 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3900 * Makefile.in: Rebuilt.
3901
652f285b 3902 * acinclude.m4: Include libtool macros from the top level.
3903 * aclocal.m4, configure: Rebuilt.
3904
6f52caf7 39052000-08-22 Alexandre Oliva <aoliva@redhat.com>
3906
3907 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3908 * configure: Rebuilt.
3909
d03e800c 39102000-05-11 Scott Bambrough <scottb@netwinder.org>
3911
805e22b2 3912 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3913 memory correctly. Use conditional instructions, not branches where
d03e800c 3914 possible.
3915
f3e32e0b 39162000-05-04 Tom Tromey <tromey@cygnus.com>
3917
3918 * configure: Rebuilt.
3919 * configure.in: Match `arm*-*-linux-*'.
3920 From Chris Dornan <cdornan@arm.com>.
3921
b23e0d76 39222000-04-28 Jakub Jelinek <jakub@redhat.com>
3923
3924 * Makefile.am (SUBDIRS): Define.
3925 (AM_MAKEFLAGS): Likewise.
3926 (Multilib support.): Add section.
3927 * Makefile.in: Rebuilt.
3928 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
968b0d1e 3929 New variables. Set for gcc using -print-multi-lib. Export them
3930 to libtool.
3931 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3932 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3933 for -shared links.
3934 (extra_compiler_flags_value, extra_compiler_flags): Check these
3935 for extra compiler options which need to be passed down in
3936 compiler_flags.
b23e0d76 3937
2c9b297c 39382000-04-16 Anthony Green <green@redhat.com>
3939
3940 * configure: Rebuilt.
3941 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3942
d7a35c6c 39432000-04-14 Jakub Jelinek <jakub@redhat.com>
3944
968b0d1e 3945 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
d7a35c6c 3946 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3947 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3948 Replace all void * sizeofs with sizeof(int).
3949 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3950 different than DOUBLE.
3951 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3952 (ffi_prep_args_v9): New function.
3953 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3954 (ffi_V9_return_struct): New function.
3955 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3956 32bit code (not yet cross-arch calls).
3957 * src/sparc/v8.S: Add struct return delay nop.
3958 Handle long long.
3959 * src/sparc/v9.S: New file.
3960 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3961 is used on sparc64 only for structures larger than 32 bytes.
3962 Pass by reference for structures is done for structure arguments
3963 larger than 16 bytes.
3964 * src/ffitest.c (main): Use 64bit rint on sparc64.
3965 Run long long tests on sparc.
3966 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3967 sparc64.
3968 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3969 on sparc64.
3970 * configure.in (sparc-*-linux*): New supported target.
3971 (sparc64-*-linux*): Likewise.
3972 * configure: Rebuilt.
3973 * Makefile.am: Add v9.S to SPARC files.
3974 * Makefile.in: Likewise.
3975 (LINK): Surround $(CCLD) into double quotes, so that multilib
3976 compiles work correctly.
3977
800d12c7 39782000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3979
3980 * configure: Rebuilt.
3981 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3982 proposed by Bryce McKinlay.
3983
c44e8808 39842000-03-20 Tom Tromey <tromey@cygnus.com>
3985
3986 * Makefile.in: Hand edit for java_raw_api.lo.
3987
f764cb60 39882000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3989
3990 * config.guess, config.sub: Update from the gcc tree.
3991 Fix for PR libgcj/168.
3992
3b43a26e 39932000-03-03 Tom Tromey <tromey@cygnus.com>
3994
96215ab3 3995 * Makefile.in: Fixed ia64 by hand.
3996
3b43a26e 3997 * configure: Rebuilt.
3998 * configure.in (--enable-multilib): New option.
3999 (libffi_basedir): New subst.
4000 (AC_OUTPUT): Added multilib code.
4001
03ae52bb 40022000-03-02 Tom Tromey <tromey@cygnus.com>
4003
4004 * Makefile.in: Rebuilt.
4005 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
4006 directory name.
4007
28c07da5 40082000-02-25 Hans Boehm <boehm@acm.org>
4009
4010 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
4011 files.
4012 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
4013 list.
4014 (ffi_prep_raw_closure): Use ffi_translate_args, not
4015 ffi_closure_translate.
4016 * src/java_raw_api.c: New file.
4017 * src/ffitest.c (closure_test_fn): New function.
4018 (main): Define `rint' as long long on IA64. Added new test when
4019 FFI_CLOSURES is defined.
4020 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
4021 (ffi_abi): Recognize IA64.
4022 (ffi_raw): Added `flt' field.
4023 Added "Java raw API" code.
4024 * configure.in: Recognize ia64.
4025 * Makefile.am (TARGET_SRC_IA64): New macro.
4026 (libffi_la_common_SOURCES): Added java_raw_api.c.
4027 (libffi_la_SOURCES): Define in IA64 case.
4028
0bcdf5c5 40292000-01-04 Tom Tromey <tromey@cygnus.com>
4030
4031 * Makefile.in: Rebuilt with newer automake.
4032
e9de8c8b 40331999-12-31 Tom Tromey <tromey@cygnus.com>
4034
4035 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
4036
20e58c02 40371999-09-01 Tom Tromey <tromey@cygnus.com>
4038
4039 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
4040 undefs.
4041 * fficonfig.h.in: Rebuilt.
4042 * configure: Rebuilt.
4043 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
4044 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
4045 * acconfig.h: Don't #undef PACKAGE or VERSION.
4046
47c5fab6 40471999-08-09 Anthony Green <green@cygnus.com>
4048
4049 * include/ffi.h.in: Try to work around messy header problem
4050 with PACKAGE and VERSION.
4051
4052 * configure: Rebuilt.
4053 * configure.in: Change version to 2.00-beta.
4054
968b0d1e 4055 * fficonfig.h.in: Rebuilt.
47c5fab6 4056 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
4057
4058 * src/x86/ffi.c (ffi_raw_call): Rename.
4059
40601999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
4061
4062 * src/x86/ffi.c (ffi_closure_SYSV): New function.
4063 (ffi_prep_incoming_args_SYSV): Ditto.
4064 (ffi_prep_closure): Ditto.
4065 (ffi_closure_raw_SYSV): Ditto.
4066 (ffi_prep_raw_closure): More ditto.
4067 (ffi_call_raw): Final ditto.
4068
4069 * include/ffi.h.in: Add definitions for closure and raw API.
4070
4071 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
805e22b2 4072 FFI_TYPE_UINT64.
47c5fab6 4073
4074 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
4075
4076 * src/raw_api.c: New file.
4077
4078 * include/ffi.h.in (ffi_raw): New type.
4079 (UINT_ARG, SINT_ARG): New defines.
4080 (ffi_closure, ffi_raw_closure): New types.
4081 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
805e22b2 4082
47c5fab6 4083 * configure.in: Add check for endianness and sizeof void*.
4084
4085 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
4086 instead of directly.
4087
4088 * configure: Rebuilt.
4089
4090Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
4091
4092 * configure.in: Add x86 and powerpc BeOS configurations.
4093 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
4094
40951999-05-09 Anthony Green <green@cygnus.com>
4096
4097 * configure.in: Add warning about this being beta code.
4098 Remove src/Makefile.am from the picture.
4099 * configure: Rebuilt.
4100
4101 * Makefile.am: Move logic from src/Makefile.am. Add changes
4102 to support libffi as a target library.
4103 * Makefile.in: Rebuilt.
4104
4105 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
4106 Upgraded to new autoconf, automake, libtool.
4107
4108 * README: Tweaks.
4109
4110 * LICENSE: Update copyright date.
4111
4112 * src/Makefile.am, src/Makefile.in: Removed.
4113
41141998-11-29 Anthony Green <green@cygnus.com>
4115
4116 * include/ChangeLog: Removed.
4117 * src/ChangeLog: Removed.
4118 * src/mips/ChangeLog: Removed.
4119 * src/sparc/ChangeLog: Remboved.
4120 * src/x86/ChangeLog: Removed.
4121
4122 * ChangeLog.v1: Created.