1 2009-12-03 David Edelsohn <edelsohn@gnu.org>
4 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
6 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
8 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
9 Remove xfails for x86 linux targets.
11 2009-12-02 David Edelsohn <edelsohn@gnu.org>
13 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
16 2009-12-01 David Edelsohn <edelsohn@gnu.org>
18 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
19 register usage. Call ffi_prep_args directly. Add long double
21 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
22 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
23 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
24 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
26 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
28 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
29 Allocate result area between params and FPRs.
31 2009-11-30 David Edelsohn <edelsohn@gnu.org>
34 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
36 * src/powerpc/aix.S: Implement AIX64 version.
37 * src/powerpc/aix_closure.S: Implement AIX64 version.
38 (ffi_closure_ASM): Use extsb, lha and displament addresses.
39 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
41 (ffi_prep_cif_machdep): Same.
43 (ffi_closure_helper_DARWIN): Same.
45 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
48 * testsuite/libffi.call/testclosure.c: New test.
50 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
52 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
53 assembly version use of ___chkstk.
55 2009-09-23 Matthias Klose <doko@ubuntu.com>
57 PR libffi/40242, PR libffi/41443
58 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
59 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
60 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
61 Change the conditionals to __SOFTFP__ || __ARM_EABI__
62 for -mfloat-abi=softfp to work.
64 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
66 PR testsuite/32843 (strikes again)
67 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
68 enable proper extension on char and short.
70 2009-09-15 David Daney <ddaney@caviumnetworks.com>
72 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
73 handling for FFI_TYPE_POINTER.
74 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
75 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
76 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
77 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
78 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
79 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
80 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
81 structure and pointer returns.
82 (ffi_closure_N32): Add handling for pointer returns.
83 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
84 calc_n32_return_struct_flags): Handle soft-float.
85 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
86 (ffi_call_N32): Declare proper argument types.
87 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
90 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
92 * configure.ac (AC_PREREQ): Bump to 2.64.
94 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
96 * Makefile.am (install-html, install-pdf): Remove.
97 * Makefile.in: Regenerate.
99 * Makefile.in: Regenerate.
100 * aclocal.m4: Regenerate.
101 * configure: Regenerate.
102 * fficonfig.h.in: Regenerate.
103 * include/Makefile.in: Regenerate.
104 * man/Makefile.in: Regenerate.
105 * testsuite/Makefile.in: Regenerate.
107 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
109 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
111 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
114 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
115 return types for X86_WIN32.
116 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
117 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
118 _ffi_closure_STDCALL): Likewise.
120 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
121 (dlmmap, dlmunmap): Also use these functions on Cygwin.
123 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
128 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
131 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
133 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
134 to "" before adding paths. (This reinstates an assignment that
135 was removed by my 2009-06-30 commit, but changes the initial
136 value from "." to "".)
138 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
141 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
142 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
145 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
147 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
148 to ld_library_path. Use add_path. Add just find_libgcc_s
149 to ld_library_path, not every libgcc multilib directory.
151 2009-06-16 Wim Lewis <wiml@hhhh.org>
153 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
154 supposed to be callee-saved.
155 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
156 return buffer for odd-size structs.
158 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
161 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
162 allow_stack_execute for Darwin.
164 2009-06-16 Andrew Haley <aph@redhat.com>
166 * configure.ac (TARGETDIR): Add missing blank lines.
167 * configure: Regenerate.
169 2009-06-16 Andrew Haley <aph@redhat.com>
171 * testsuite/libffi.call/cls_align_sint64.c,
172 testsuite/libffi.call/cls_align_uint64.c,
173 testsuite/libffi.call/cls_longdouble_va.c,
174 testsuite/libffi.call/cls_ulonglong.c,
175 testsuite/libffi.call/return_ll1.c,
176 testsuite/libffi.call/stret_medium2.c: Fix printf format
178 * testsuite/libffi.call/ffitest.h,
179 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
181 2009-06-15 Andrew Haley <aph@redhat.com>
183 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
184 * testsuite/libffi.call/err_bad_abi.c: Likewise.
186 2009-06-12 Andrew Haley <aph@redhat.com>
188 * Makefile.am: Remove info_TEXINFOS.
190 2009-06-12 Andrew Haley <aph@redhat.com>
192 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
193 testsuite/libffi.call/cls_align_uint64.c,
194 testsuite/libffi.call/cls_ulonglong.c,
195 testsuite/libffi.call/return_ll1.c,
196 testsuite/libffi.call/stret_medium2.c: Fix printf format
198 testsuite/libffi.special/unwindtest.cc: include stdint.h.
200 2009-06-11 Timothy Wall <twall@users.sf.net>
205 include/ffi_common.h,
210 src/x86/win64.S (new),
211 README: Added win64 support (mingw or MSVC)
215 testsuite/Makefile.in,
217 aclocal.m4: Regenerated
218 * ltcf-c.sh: properly escape cygwin/w32 path
219 * man/ffi_call.3: Clarify size requirements for return value.
220 * src/x86/ffi64.c: Fix filename in comment.
221 * src/x86/win32.S: Remove unused extern.
223 * testsuite/libffi.call/closure_fn0.c,
224 testsuite/libffi.call/closure_fn1.c,
225 testsuite/libffi.call/closure_fn2.c,
226 testsuite/libffi.call/closure_fn3.c,
227 testsuite/libffi.call/closure_fn4.c,
228 testsuite/libffi.call/closure_fn5.c,
229 testsuite/libffi.call/closure_fn6.c,
230 testsuite/libffi.call/closure_stdcall.c,
231 testsuite/libffi.call/cls_12byte.c,
232 testsuite/libffi.call/cls_16byte.c,
233 testsuite/libffi.call/cls_18byte.c,
234 testsuite/libffi.call/cls_19byte.c,
235 testsuite/libffi.call/cls_1_1byte.c,
236 testsuite/libffi.call/cls_20byte.c,
237 testsuite/libffi.call/cls_20byte1.c,
238 testsuite/libffi.call/cls_24byte.c,
239 testsuite/libffi.call/cls_2byte.c,
240 testsuite/libffi.call/cls_3_1byte.c,
241 testsuite/libffi.call/cls_3byte1.c,
242 testsuite/libffi.call/cls_3byte2.c,
243 testsuite/libffi.call/cls_4_1byte.c,
244 testsuite/libffi.call/cls_4byte.c,
245 testsuite/libffi.call/cls_5_1_byte.c,
246 testsuite/libffi.call/cls_5byte.c,
247 testsuite/libffi.call/cls_64byte.c,
248 testsuite/libffi.call/cls_6_1_byte.c,
249 testsuite/libffi.call/cls_6byte.c,
250 testsuite/libffi.call/cls_7_1_byte.c,
251 testsuite/libffi.call/cls_7byte.c,
252 testsuite/libffi.call/cls_8byte.c,
253 testsuite/libffi.call/cls_9byte1.c,
254 testsuite/libffi.call/cls_9byte2.c,
255 testsuite/libffi.call/cls_align_double.c,
256 testsuite/libffi.call/cls_align_float.c,
257 testsuite/libffi.call/cls_align_longdouble.c,
258 testsuite/libffi.call/cls_align_longdouble_split.c,
259 testsuite/libffi.call/cls_align_longdouble_split2.c,
260 testsuite/libffi.call/cls_align_pointer.c,
261 testsuite/libffi.call/cls_align_sint16.c,
262 testsuite/libffi.call/cls_align_sint32.c,
263 testsuite/libffi.call/cls_align_sint64.c,
264 testsuite/libffi.call/cls_align_uint16.c,
265 testsuite/libffi.call/cls_align_uint32.c,
266 testsuite/libffi.call/cls_align_uint64.c,
267 testsuite/libffi.call/cls_dbls_struct.c,
268 testsuite/libffi.call/cls_double.c,
269 testsuite/libffi.call/cls_double_va.c,
270 testsuite/libffi.call/cls_float.c,
271 testsuite/libffi.call/cls_longdouble.c,
272 testsuite/libffi.call/cls_longdouble_va.c,
273 testsuite/libffi.call/cls_multi_schar.c,
274 testsuite/libffi.call/cls_multi_sshort.c,
275 testsuite/libffi.call/cls_multi_sshortchar.c,
276 testsuite/libffi.call/cls_multi_uchar.c,
277 testsuite/libffi.call/cls_multi_ushort.c,
278 testsuite/libffi.call/cls_multi_ushortchar.c,
279 testsuite/libffi.call/cls_pointer.c,
280 testsuite/libffi.call/cls_pointer_stack.c,
281 testsuite/libffi.call/cls_schar.c,
282 testsuite/libffi.call/cls_sint.c,
283 testsuite/libffi.call/cls_sshort.c,
284 testsuite/libffi.call/cls_uchar.c,
285 testsuite/libffi.call/cls_uint.c,
286 testsuite/libffi.call/cls_ulonglong.c,
287 testsuite/libffi.call/cls_ushort.c,
288 testsuite/libffi.call/err_bad_abi.c,
289 testsuite/libffi.call/err_bad_typedef.c,
290 testsuite/libffi.call/float2.c,
291 testsuite/libffi.call/huge_struct.c,
292 testsuite/libffi.call/nested_struct.c,
293 testsuite/libffi.call/nested_struct1.c,
294 testsuite/libffi.call/nested_struct10.c,
295 testsuite/libffi.call/nested_struct2.c,
296 testsuite/libffi.call/nested_struct3.c,
297 testsuite/libffi.call/nested_struct4.c,
298 testsuite/libffi.call/nested_struct5.c,
299 testsuite/libffi.call/nested_struct6.c,
300 testsuite/libffi.call/nested_struct7.c,
301 testsuite/libffi.call/nested_struct8.c,
302 testsuite/libffi.call/nested_struct9.c,
303 testsuite/libffi.call/problem1.c,
304 testsuite/libffi.call/return_ldl.c,
305 testsuite/libffi.call/return_ll1.c,
306 testsuite/libffi.call/stret_large.c,
307 testsuite/libffi.call/stret_large2.c,
308 testsuite/libffi.call/stret_medium.c,
309 testsuite/libffi.call/stret_medium2.c,
310 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
311 of checking for MMAP. Use intptr_t instead of long casts.
313 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
315 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
316 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
317 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
319 2009-06-09 Andrew Haley <aph@redhat.com>
321 * src/x86/freebsd.S: Add missing file.
323 2009-06-08 Andrew Haley <aph@redhat.com>
325 Import from libffi 3.0.8:
327 * doc/libffi.texi: New file.
328 * doc/libffi.info: Likewise.
329 * doc/stamp-vti: Likewise.
330 * man/Makefile.am: New file.
331 * man/ffi_call.3: New file.
333 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
335 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
337 * configure.ac: Bump version to 3.0.8.
338 parisc*-*-linux*: Add.
339 i386-*-freebsd* | i386-*-openbsd*: Add.
340 powerpc-*-beos*: Add.
341 AM_CONDITIONAL X86_FREEBSD: Add.
342 AC_CONFIG_FILES: Add man/Makefile.
344 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
346 2009-06-08 Andrew Haley <aph@redhat.com>
348 * README: Import from libffi 3.0.8.
350 2009-06-08 Andrew Haley <aph@redhat.com>
352 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
353 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
354 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
355 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
357 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
358 * testsuite/libffi.call/stret_medium.c: Likewise.
359 * testsuite/libffi.call/stret_large2.c: Likewise.
360 * testsuite/libffi.call/stret_large.c: Likewise.
362 2008-12-26 Timothy Wall <twall@users.sf.net>
364 * testsuite/libffi.call/cls_longdouble.c,
365 testsuite/libffi.call/cls_longdouble_va.c,
366 testsuite/libffi.call/cls_align_longdouble.c,
367 testsuite/libffi.call/cls_align_longdouble_split.c,
368 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
369 failures on x86_64 cygwin/mingw.
371 2008-12-22 Timothy Wall <twall@users.sf.net>
373 * testsuite/libffi.call/closure_fn0.c,
374 testsuite/libffi.call/closure_fn1.c,
375 testsuite/libffi.call/closure_fn2.c,
376 testsuite/libffi.call/closure_fn3.c,
377 testsuite/libffi.call/closure_fn4.c,
378 testsuite/libffi.call/closure_fn5.c,
379 testsuite/libffi.call/closure_fn6.c,
380 testsuite/libffi.call/closure_loc_fn0.c,
381 testsuite/libffi.call/closure_stdcall.c,
382 testsuite/libffi.call/cls_align_pointer.c,
383 testsuite/libffi.call/cls_pointer.c,
384 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
385 pointer to integer (intptr_t).
386 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
388 2008-07-24 Anthony Green <green@redhat.com>
390 * testsuite/libffi.call/cls_dbls_struct.c,
391 testsuite/libffi.call/cls_double_va.c,
392 testsuite/libffi.call/cls_longdouble.c,
393 testsuite/libffi.call/cls_longdouble_va.c,
394 testsuite/libffi.call/cls_pointer.c,
395 testsuite/libffi.call/cls_pointer_stack.c,
396 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
399 2008-03-04 Anthony Green <green@redhat.com>
403 * testsuite/libffi.call/cls_align_longdouble_split2.c
404 testsuite/libffi.call/cls_align_longdouble_split.c
405 testsuite/libffi.call/cls_dbls_struct.c
406 testsuite/libffi.call/cls_double_va.c
407 testsuite/libffi.call/cls_longdouble.c
408 testsuite/libffi.call/cls_longdouble_va.c
409 testsuite/libffi.call/cls_pointer.c
410 testsuite/libffi.call/cls_pointer_stack.c
411 testsuite/libffi.call/err_bad_abi.c
412 testsuite/libffi.call/err_bad_typedef.c
413 testsuite/libffi.call/stret_large2.c
414 testsuite/libffi.call/stret_large.c
415 testsuite/libffi.call/stret_medium2.c
416 testsuite/libffi.call/stret_medium.c: New tests from Apple.
418 2009-06-05 Andrew Haley <aph@redhat.com>
420 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
423 2009-06-04 Andrew Haley <aph@redhat.com>
425 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
428 2008-02-26 Anthony Green <green@redhat.com>
429 Thomas Heller <theller@ctypes.org>
431 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
434 2008-02-03 Timothy Wall <twall@users.sf.net>
436 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
437 offset based on code pointer, not data pointer.
439 2008-01-31 Timothy Wall <twall@users.sf.net>
441 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
443 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
445 * src/x86/win32.S: Add assembly for stdcall closure.
446 * src/x86/ffi.c: Initialize stdcall closure trampoline.
448 2009-06-04 Andrew Haley <aph@redhat.com>
450 * include/ffi.h.in: Change void (*)() to void (*)(void).
451 * src/x86/ffi.c: Likewise.
453 2009-06-04 Andrew Haley <aph@redhat.com>
455 * src/powerpc/ppc_closure.S: Insert licence header.
456 * src/powerpc/linux64_closure.S: Likewise.
457 * src/m68k/sysv.S: Likewise.
459 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
460 * src/powerpc/ffi.c: Likewise.
461 * src/powerpc/ffi_darwin.c: Likewise.
462 * src/m32r/ffi.c: Likewise.
463 * src/sh64/ffi.c: Likewise.
464 * src/x86/ffi64.c: Likewise.
465 * src/alpha/ffi.c: Likewise.
466 * src/alpha/osf.S: Likewise.
467 * src/frv/ffi.c: Likewise.
468 * src/s390/ffi.c: Likewise.
469 * src/pa/ffi.c: Likewise.
470 * src/pa/hpux32.S: Likewise.
471 * src/ia64/unix.S: Likewise.
472 * src/ia64/ffi.c: Likewise.
473 * src/sparc/ffi.c: Likewise.
474 * src/mips/ffi.c: Likewise.
475 * src/sh/ffi.c: Likewise.
477 2008-02-15 David Daney <ddaney@avtrex.com>
479 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
480 Define (conditionally), and use it to include cachectl.h.
481 (ffi_prep_closure_loc): Fix cache flushing.
482 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
484 2009-06-04 Andrew Haley <aph@redhat.com>
490 src/powerpc/ffitarget.h,
492 src/sh64/ffitarget.h,
503 src/alpha/ffitarget.h,
506 src/m68k/ffitarget.h,
509 src/s390/ffitarget.h,
511 src/cris/ffitarget.h,
516 src/ia64/ffitarget.h,
519 src/ia64/ia64_flags.h,
523 src/sparc/ffitarget.h,
526 src/mips/ffitarget.h,
533 src/sh/sysv.S: Update license text.
535 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
537 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
538 (.eh_frame): Add FDE for it.
540 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
542 * configure.ac: Also check if assembler supports pc-relative
543 relocs on X86_WIN32 targets.
544 * configure: Regenerate.
545 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
546 (_ffi_call_SYSV): Add missing function type symbol .def and
547 add EH markup labels.
548 (_ffi_call_STDCALL): Likewise.
549 (_ffi_closure_SYSV): Likewise.
550 (_ffi_closure_raw_SYSV): Likewise.
551 (.eh_frame): Add hand-crafted EH data.
553 2009-04-09 Jakub Jelinek <jakub@redhat.com>
555 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
556 version 3 of the GNU General Public License and to point readers
557 at the COPYING3 file and the FSF's license web page.
558 * testsuite/libffi.call/call.exp: Likewise.
559 * testsuite/libffi.special/special.exp: Likewise.
561 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
563 * configure: Regenerate.
565 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
568 * configure.ac (HAVE_AS_X86_PCREL): New test.
569 * configure: Regenerate.
570 * fficonfig.h.in: Regenerate.
571 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
572 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
573 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
574 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
575 * src/x86/unix64.S (.Lstore_table): Move to .text section.
576 (.Lload_table): Likewise.
577 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
579 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
581 * configure: Regenerate.
583 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
585 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
586 signed/unsigned int8/16 return values.
587 * src/sparc/v8.S (ffi_call_v8): Likewise.
588 (ffi_closure_v8): Likewise.
590 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
591 Steve Ellcey <sje@cup.hp.com>
593 * configure: Regenerate for new libtool.
594 * Makefile.in: Ditto.
595 * include/Makefile.in: Ditto.
598 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
600 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
601 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
602 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
603 Adjust copyright notice.
604 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
605 register or two register to use for FFI_SYSV structs.
606 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
607 (ffi_closure_helper_SYSV): Fix the return type for
608 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
609 Adjust copyright notice.
611 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
613 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
616 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
618 * configure: Regenerate.
619 * include/Makefile.in: Regenerate.
620 * testsuite/Makefile.in: Regenerate.
622 2008-06-07 Joseph Myers <joseph@codesourcery.com>
624 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
625 powerpc-*-beos*): Remove.
626 * configure: Regenerate.
628 2008-05-09 Julian Brown <julian@codesourcery.com>
630 * Makefile.am (LTLDFLAGS): New.
631 (libffi_la_LDFLAGS): Use above.
632 * Makefile.in: Regenerate.
634 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
637 * aclocal.m4: Regenerate.
638 * configure: Regenerate.
640 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
642 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
643 * src/sh64/sysv.S: Likewise.
645 2008-03-26 Daniel Jacobowitz <dan@debian.org>
647 * src/arm/sysv.S: Fix ARM comment marker.
649 2008-03-26 Jakub Jelinek <jakub@redhat.com>
651 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
652 * src/s390/sysv.S: Likewise.
653 * src/powerpc/ppc_closure.S: Likewise.
654 * src/powerpc/sysv.S: Likewise.
655 * src/x86/unix64.S: Likewise.
656 * src/x86/sysv.S: Likewise.
657 * src/sparc/v8.S: Likewise.
658 * src/sparc/v9.S: Likewise.
659 * src/m68k/sysv.S: Likewise.
660 * src/arm/sysv.S: Likewise.
662 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
664 * aclocal.m4: Regenerate.
665 * configure: Likewise.
666 * Makefile.in: Likewise.
667 * include/Makefile.in: Likewise.
668 * testsuite/Makefile.in: Likewise.
670 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
671 Andreas Tobler <a.tobler@schweiz.org>
673 * configure.ac: Add amd64-*-freebsd* target.
674 * configure: Regenerate.
676 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
679 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
682 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
685 2008-01-24 David Edelsohn <edelsohn@gnu.org>
687 * configure: Regenerate.
689 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
691 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
693 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
696 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
697 signed/unsigned int8/16 for X86_DARWIN.
698 Updated copyright info.
699 Handle one and two byte structs with special cif->flags.
700 * src/x86/ffitarget.h: Add special types for one and two byte structs.
701 Updated copyright info.
702 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
704 Remove code to pop args from the stack after call.
705 Special-case signed/unsigned for int8/16, one and two byte structs.
706 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
707 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
709 Updated copyright info.
711 2007-12-08 David Daney <ddaney@avtrex.com>
713 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
714 SUBU, add with ADDU and use smaller code sequences.
716 2007-12-07 David Daney <ddaney@avtrex.com>
718 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
721 2007-12-06 David Daney <ddaney@avtrex.com>
723 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
725 (ffi_java_raw): New typedef.
726 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
727 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
729 (ffi_java_raw_closure) : Same.
730 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
732 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
734 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
735 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
736 sizeof(ffi_java_raw) for alignment calculations.
737 (ffi_java_ptrarray_to_raw): Same.
738 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
739 if FFI_SIZEOF_JAVA_RAW == 4.
740 (ffi_java_raw_to_rvalue): Same.
741 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
742 (ffi_java_translate_args): Same.
743 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
745 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
747 2007-12-06 David Daney <ddaney@avtrex.com>
749 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
752 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
755 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
756 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
757 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
758 set the NUM_FPR_ARG_REGISTERS according to.
759 Add support for potential soft-float support under hard-float
761 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
762 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
763 to the FFI_LINUX_SOFT_FLOAT ABI.
764 (ffi_prep_cif_machdep): Likewise.
765 (ffi_closure_helper_SYSV): Likewise.
766 * src/powerpc/ppc_closure.S: Make sure not to store float/double
767 on archs where __NO_FPRS__ is true.
768 Add FFI_TYPE_UINT128 support.
769 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
770 Adjust copyright notice.
772 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
774 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
775 * include/ffi_common.h: ... here.
778 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
780 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
781 * src/powerpc/linux64.S: Likewise.
782 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
783 * testsuite/libffi.call/return_ldl.c: New test.
785 2007-09-04 <aph@redhat.com>
787 * src/arm/sysv.S (UNWIND): New.
788 (Whole file): Conditionally compile unwinder directives.
789 * src/arm/sysv.S: Add unwinder directives.
791 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
792 Only treat r0 as a struct address if we're actually returning a
794 Only copy the bytes that are actually within a struct.
795 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
796 is returned in r0, not passed by address.
797 (ffi_call): Allocate a word-sized temporary for the case where
798 a composite is returned in r0.
799 (ffi_prep_incoming_args_SYSV): Align as necessary.
801 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
803 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
804 directly using the sys_cacheflush syscall.
806 2007-07-27 Andrew Haley <aph@redhat.com>
808 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
810 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
812 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
813 * configure.ac: Likewise.
814 * Makefile.in: Regenerate.
815 * include/Makefile.in: Likewise.
816 * testsuite/Makefile.in: Likewise.
817 * configure: Likewise.
819 2007-08-24 David Daney <ddaney@avtrex.com>
821 * testsuite/libffi.call/return_sl.c: New test.
823 2007-08-10 David Daney <ddaney@avtrex.com>
825 * testsuite/libffi.call/cls_multi_ushort.c,
826 testsuite/libffi.call/cls_align_uint16.c,
827 testsuite/libffi.call/nested_struct1.c,
828 testsuite/libffi.call/nested_struct3.c,
829 testsuite/libffi.call/cls_7_1_byte.c,
830 testsuite/libffi.call/nested_struct5.c,
831 testsuite/libffi.call/cls_double.c,
832 testsuite/libffi.call/nested_struct7.c,
833 testsuite/libffi.call/cls_sint.c,
834 testsuite/libffi.call/nested_struct9.c,
835 testsuite/libffi.call/cls_20byte1.c,
836 testsuite/libffi.call/cls_multi_sshortchar.c,
837 testsuite/libffi.call/cls_align_sint64.c,
838 testsuite/libffi.call/cls_3byte2.c,
839 testsuite/libffi.call/cls_multi_schar.c,
840 testsuite/libffi.call/cls_multi_uchar.c,
841 testsuite/libffi.call/cls_19byte.c,
842 testsuite/libffi.call/cls_9byte1.c,
843 testsuite/libffi.call/cls_align_float.c,
844 testsuite/libffi.call/closure_fn1.c,
845 testsuite/libffi.call/problem1.c,
846 testsuite/libffi.call/closure_fn3.c,
847 testsuite/libffi.call/cls_sshort.c,
848 testsuite/libffi.call/closure_fn5.c,
849 testsuite/libffi.call/cls_align_double.c,
850 testsuite/libffi.call/nested_struct.c,
851 testsuite/libffi.call/cls_2byte.c,
852 testsuite/libffi.call/nested_struct10.c,
853 testsuite/libffi.call/cls_4byte.c,
854 testsuite/libffi.call/cls_6byte.c,
855 testsuite/libffi.call/cls_8byte.c,
856 testsuite/libffi.call/cls_multi_sshort.c,
857 testsuite/libffi.call/cls_align_sint16.c,
858 testsuite/libffi.call/cls_align_uint32.c,
859 testsuite/libffi.call/cls_20byte.c,
860 testsuite/libffi.call/cls_float.c,
861 testsuite/libffi.call/nested_struct2.c,
862 testsuite/libffi.call/cls_5_1_byte.c,
863 testsuite/libffi.call/nested_struct4.c,
864 testsuite/libffi.call/cls_24byte.c,
865 testsuite/libffi.call/nested_struct6.c,
866 testsuite/libffi.call/cls_64byte.c,
867 testsuite/libffi.call/nested_struct8.c,
868 testsuite/libffi.call/cls_uint.c,
869 testsuite/libffi.call/cls_multi_ushortchar.c,
870 testsuite/libffi.call/cls_schar.c,
871 testsuite/libffi.call/cls_uchar.c,
872 testsuite/libffi.call/cls_align_uint64.c,
873 testsuite/libffi.call/cls_ulonglong.c,
874 testsuite/libffi.call/cls_align_longdouble.c,
875 testsuite/libffi.call/cls_1_1byte.c,
876 testsuite/libffi.call/cls_12byte.c,
877 testsuite/libffi.call/cls_3_1byte.c,
878 testsuite/libffi.call/cls_3byte1.c,
879 testsuite/libffi.call/cls_4_1byte.c,
880 testsuite/libffi.call/cls_6_1_byte.c,
881 testsuite/libffi.call/cls_16byte.c,
882 testsuite/libffi.call/cls_18byte.c,
883 testsuite/libffi.call/closure_fn0.c,
884 testsuite/libffi.call/cls_9byte2.c,
885 testsuite/libffi.call/closure_fn2.c,
886 testsuite/libffi.call/closure_fn4.c,
887 testsuite/libffi.call/cls_ushort.c,
888 testsuite/libffi.call/closure_fn6.c,
889 testsuite/libffi.call/cls_5byte.c,
890 testsuite/libffi.call/cls_align_pointer.c,
891 testsuite/libffi.call/cls_7byte.c,
892 testsuite/libffi.call/cls_align_sint32.c,
893 testsuite/libffi.special/unwindtest_ffi_call.cc,
894 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
896 2007-08-10 David Daney <ddaney@avtrex.com>
899 * configure.ac: Don't treat mips64 as a special case.
900 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
901 * configure: Regenerate
902 * Makefile.in: Ditto.
903 * fficonfig.h.in: Ditto.
904 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
905 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
906 (FFI_DEFAULT_ABI): Set for n64 case.
907 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
908 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
909 (ffi_closure_N32): New function.
910 (.eh_frame): New section
911 * src/mips/o32.S: Clean up comments.
912 (ffi_closure_O32): Pass ffi_closure parameter in $12.
913 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
914 _MIPS_SIM == _ABIN32 throughout.
915 (FFI_MIPS_STOP_HERE): New, use in place of
917 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
918 to support n32/n64 ABIs.
919 (calc_n32_struct_flags): Rewrite.
920 (calc_n32_return_struct_flags): Remove unused variable. Reverse
921 position of flag bits.
922 (ffi_prep_cif_machdep): Rewrite n32 portion.
923 (ffi_call): Enable for n64. Add special handling for small structure
925 (ffi_prep_closure_loc): Add n32 and n64 support.
926 (ffi_closure_mips_inner_O32): Add cast to silence warning.
927 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
929 2007-08-08 David Daney <ddaney@avtrex.com>
931 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
932 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
934 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
935 * testsuite/libffi.call/cls_sint.c (main): Ditto.
936 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
937 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
938 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
939 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
940 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
941 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
942 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
943 * testsuite/libffi.call/return_fl3.c (main): Ditto.
944 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
945 * testsuite/libffi.call/nested_struct.c (main): Ditto.
946 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
947 * testsuite/libffi.call/return_ll1.c (main): Ditto.
948 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
949 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
950 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
951 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
952 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
953 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
954 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
955 * testsuite/libffi.call/cls_uint.c (main): Ditto.
956 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
957 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
958 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
959 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
960 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
961 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
962 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
963 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
964 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
966 2007-08-07 Andrew Haley <aph@redhat.com>
968 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
971 2007-08-06 Andrew Haley <aph@redhat.com>
974 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
975 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
978 2007-08-02 David Daney <ddaney@avtrex.com>
980 * testsuite/libffi.call/return_ul.c (main): Define return type as
981 ffi_arg. Use proper printf conversion specifier.
983 2007-07-30 Andrew Haley <aph@redhat.com>
986 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
987 signed/unsigned int8/16.
988 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
990 Remove code to pop args from the stack after call.
991 Special-case signed/unsigned int8/16.
992 * testsuite/libffi.call/return_sc.c (main): Revert.
994 2007-07-26 Richard Guenther <rguenther@suse.de>
997 * testsuite/libffi.call/return_sc.c (main): Verify call
998 result as signed char, not ffi_arg.
1000 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1002 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1003 * configure: Regenerate.
1005 2007-07-11 David Daney <ddaney@avtrex.com>
1007 * src/mips/ffi.c: Don't include sys/cachectl.h.
1008 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1011 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1013 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1014 from (ffi_prep_closure): ... this.
1015 (FFI_INIT_TRAMPOLINE): Adjust.
1017 2005-12-31 Phil Blundell <pb@reciva.com>
1019 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1020 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1021 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1022 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1023 (FFI_CLOSURES): Enable closure support.
1025 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1027 * testsuite/libffi.call/cls_multi_ushort.c,
1028 testsuite/libffi.call/cls_align_uint16.c,
1029 testsuite/libffi.call/nested_struct1.c,
1030 testsuite/libffi.call/nested_struct3.c,
1031 testsuite/libffi.call/cls_7_1_byte.c,
1032 testsuite/libffi.call/cls_double.c,
1033 testsuite/libffi.call/nested_struct5.c,
1034 testsuite/libffi.call/nested_struct7.c,
1035 testsuite/libffi.call/cls_sint.c,
1036 testsuite/libffi.call/nested_struct9.c,
1037 testsuite/libffi.call/cls_20byte1.c,
1038 testsuite/libffi.call/cls_multi_sshortchar.c,
1039 testsuite/libffi.call/cls_align_sint64.c,
1040 testsuite/libffi.call/cls_3byte2.c,
1041 testsuite/libffi.call/cls_multi_schar.c,
1042 testsuite/libffi.call/cls_multi_uchar.c,
1043 testsuite/libffi.call/cls_19byte.c,
1044 testsuite/libffi.call/cls_9byte1.c,
1045 testsuite/libffi.call/cls_align_float.c,
1046 testsuite/libffi.call/closure_fn1.c,
1047 testsuite/libffi.call/problem1.c,
1048 testsuite/libffi.call/closure_fn3.c,
1049 testsuite/libffi.call/cls_sshort.c,
1050 testsuite/libffi.call/closure_fn5.c,
1051 testsuite/libffi.call/cls_align_double.c,
1052 testsuite/libffi.call/cls_2byte.c,
1053 testsuite/libffi.call/nested_struct.c,
1054 testsuite/libffi.call/nested_struct10.c,
1055 testsuite/libffi.call/cls_4byte.c,
1056 testsuite/libffi.call/cls_6byte.c,
1057 testsuite/libffi.call/cls_8byte.c,
1058 testsuite/libffi.call/cls_multi_sshort.c,
1059 testsuite/libffi.call/cls_align_uint32.c,
1060 testsuite/libffi.call/cls_align_sint16.c,
1061 testsuite/libffi.call/cls_float.c,
1062 testsuite/libffi.call/cls_20byte.c,
1063 testsuite/libffi.call/cls_5_1_byte.c,
1064 testsuite/libffi.call/nested_struct2.c,
1065 testsuite/libffi.call/cls_24byte.c,
1066 testsuite/libffi.call/nested_struct4.c,
1067 testsuite/libffi.call/nested_struct6.c,
1068 testsuite/libffi.call/cls_64byte.c,
1069 testsuite/libffi.call/nested_struct8.c,
1070 testsuite/libffi.call/cls_uint.c,
1071 testsuite/libffi.call/cls_multi_ushortchar.c,
1072 testsuite/libffi.call/cls_schar.c,
1073 testsuite/libffi.call/cls_uchar.c,
1074 testsuite/libffi.call/cls_align_uint64.c,
1075 testsuite/libffi.call/cls_ulonglong.c,
1076 testsuite/libffi.call/cls_align_longdouble.c,
1077 testsuite/libffi.call/cls_1_1byte.c,
1078 testsuite/libffi.call/cls_12byte.c,
1079 testsuite/libffi.call/cls_3_1byte.c,
1080 testsuite/libffi.call/cls_3byte1.c,
1081 testsuite/libffi.call/cls_4_1byte.c,
1082 testsuite/libffi.call/cls_6_1_byte.c,
1083 testsuite/libffi.call/cls_16byte.c,
1084 testsuite/libffi.call/cls_18byte.c,
1085 testsuite/libffi.call/closure_fn0.c,
1086 testsuite/libffi.call/cls_9byte2.c,
1087 testsuite/libffi.call/closure_fn2.c,
1088 testsuite/libffi.call/closure_fn4.c,
1089 testsuite/libffi.call/cls_ushort.c,
1090 testsuite/libffi.call/closure_fn6.c,
1091 testsuite/libffi.call/cls_5byte.c,
1092 testsuite/libffi.call/cls_align_pointer.c,
1093 testsuite/libffi.call/cls_7byte.c,
1094 testsuite/libffi.call/cls_align_sint32.c,
1095 testsuite/libffi.special/unwindtest_ffi_call.cc,
1096 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1098 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1100 * aclocal.m4: Regenerated.
1102 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1104 * configure: Regenerate.
1106 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1108 * Makefile.in: Regenerate.
1109 * configure: Regenerate.
1110 * aclocal.m4: Regenerate.
1111 * include/Makefile.in: Regenerate.
1112 * testsuite/Makefile.in: Regenerate.
1114 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1116 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1117 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1118 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1119 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1120 (FFI_CLOSURES): Enable closure support.
1122 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1124 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1125 * configure: Regenerate.
1126 * fficonfig.h.in: Regenerate.
1127 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1128 CFI_OFFSET,CFI_DEF_CFA): New macros.
1129 (ffi_call_SYSV): Add callframe annotation.
1131 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1133 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1134 numerous test suite failures.
1135 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1137 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1139 * Makefile.am (EXTRA_DIST): Bring up to date.
1140 * Makefile.in: Regenerate.
1141 * src/frv/eabi.S: Remove RCS keyword.
1143 2007-04-06 Richard Henderson <rth@redhat.com>
1145 * configure.ac: Tidy target case.
1146 (HAVE_LONG_DOUBLE): Allow the target to override.
1147 * configure: Regenerate.
1148 * include/ffi.h.in: Don't define ffi_type_foo if
1149 LIBFFI_HIDE_BASIC_TYPES is defined.
1150 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1152 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1153 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1154 (ffi_type_longdouble): Special case for Alpha. Don't define
1155 if long double == double.
1157 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1158 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1159 (ffi_call, ffi_closure_osf_inner): Likewise.
1160 (ffi_closure_osf_inner): Likewise. Mark hidden.
1161 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1162 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1163 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1164 (load_table): Handle 128-bit long double.
1166 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1168 2007-04-06 Tom Tromey <tromey@redhat.com>
1171 * README: Fixed bug in example.
1173 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1175 * src/closures.c: Include sys/statfs.h.
1176 (_GNU_SOURCE): Define on Linux.
1177 (FFI_MMAP_EXEC_SELINUX): Define.
1178 (selinux_enabled): New variable.
1179 (selinux_enabled_check): New function.
1180 (is_selinux_enabled): Define.
1183 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1185 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1186 Use 'volatile float sum' to create sum of floats to avoid false
1187 negative due to excess precision on ix86 targets.
1190 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1192 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1194 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1196 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1198 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1199 (ffi_prep_closure_loc): New.
1200 (ffi_prep_raw_closure_loc): New.
1201 (ffi_prep_java_raw_closure_loc): New.
1202 * src/closures.c: New file.
1203 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1204 Replace sflags with exec_offset.
1205 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1206 sub_segment_exec_offset): New macros.
1207 (get_segment_flags, set_segment_flags, check_segment_merge): New
1209 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1210 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1211 destroy_mspace): Use new macros.
1212 (sys_alloc): Silence warning.
1213 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1214 * Makefile.in: Rebuilt.
1215 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1216 terms of ffi_prep_closure_loc.
1217 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1219 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1221 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1223 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1224 the renamed version.
1225 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1226 (ffi_prep_closure): ... this.
1227 * src/pa/ffi.c: Likewise.
1228 * src/cris/ffi.c: Likewise. Adjust.
1229 * src/frv/ffi.c: Likewise.
1230 * src/ia64/ffi.c: Likewise.
1231 * src/mips/ffi.c: Likewise.
1232 * src/powerpc/ffi_darwin.c: Likewise.
1233 * src/s390/ffi.c: Likewise.
1234 * src/sh/ffi.c: Likewise.
1235 * src/sh64/ffi.c: Likewise.
1236 * src/sparc/ffi.c: Likewise.
1237 * src/x86/ffi64.c: Likewise.
1238 * src/x86/ffi.c: Likewise.
1239 (FFI_INIT_TRAMPOLINE): Adjust.
1240 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1241 (ffi_prep_raw_closure): ... this.
1242 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1243 (ffi_prep_closure): ... this.
1244 (flush_icache): Adjust.
1246 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1248 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1251 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1253 * Makefile.am: Add dummy install-pdf target.
1254 * Makefile.in: Regenerate
1256 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1258 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1259 ffi_closure_helper_SYSV): Add long double handling.
1261 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1263 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1264 immediately after bctrl instruction.
1266 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1268 * Makefile.am (all-recursive, install-recursive,
1269 mostlyclean-recursive, clean-recursive, distclean-recursive,
1270 maintainer-clean-recursive): Add missing targets.
1271 * Makefile.in: Rebuilt.
1273 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1275 * configure.ac: Add TARGET for x86_64-*-darwin*.
1276 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1278 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1279 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1280 * configure: Regenerate.
1281 * Makefile.in: Regenerate.
1282 * include/Makefile.in: Regenerate.
1283 * testsuite/Makefile.in: Regenerate.
1284 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1287 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1289 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1292 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1294 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1295 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1297 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1298 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1300 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1301 Sandro Tolaini <tolaini@libero.it>
1303 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1305 * configure: Regenerated.
1306 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1307 (EXTRA_DIST): Add src/x86/darwin.S.
1308 * Makefile.in: Regenerated.
1309 * include/Makefile.in: Regenerated.
1310 * testsuite/Makefile.in: Regenerated.
1312 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1313 X86_WIN32, and additionally align stack to 16 bytes.
1314 * src/x86/darwin.S: New, based on sysv.S.
1315 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1317 2006-09-12 David Daney <ddaney@avtrex.com>
1320 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1321 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1322 * aclocal.m4: Regenerated for automake 1.9.6.
1323 * Makefile.in: Regenerated.
1324 * include/Makefile.in: Regenerated.
1325 * testsuite/Makefile.in: Regenerated.
1327 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1329 * include/ffi_common.h (struct): Revert accidental commit.
1331 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1333 * include/ffi_common.h: Remove lint directives.
1334 * include/ffi.h.in: Likewise.
1336 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1338 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1339 for 32-bit architectures.
1340 * testsuite/libffi.call/return_ul.c: New test case.
1342 2006-07-19 David Daney <ddaney@avtrex.com>
1344 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1345 xfail remains for mips64.
1347 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1349 * Makefile.am: Add install-html target. Add install-html to .PHONY
1350 * Makefile.in: Regenerate.
1351 * aclocal.m4: Regenerate.
1352 * include/Makefile.in: Regenerate.
1353 * testsuite/Makefile.in: Regenerate.
1355 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1357 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1360 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1362 * README: Remove notice about 'Crazy Comments'.
1363 * src/debug.c: Remove lint directives. Cleanup white spaces.
1364 * src/java_raw_api.c: Likewise.
1365 * src/prep_cif.c: Likewise.
1366 * src/raw_api.c: Likewise.
1367 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1369 * src/arm/ffi.c: Remove lint directives.
1370 * src/m32r/ffi.c: Likewise.
1371 * src/pa/ffi.c: Likewise.
1372 * src/powerpc/ffi.c: Likewise.
1373 * src/powerpc/ffi_darwin.c: Likewise.
1374 * src/sh/ffi.c: Likewise.
1375 * src/sh64/ffi.c: Likewise.
1376 * src/x86/ffi.c: Likewise.
1377 * testsuite/libffi.call/float2.c: Likewise.
1378 * testsuite/libffi.call/promotion.c: Likewise.
1379 * testsuite/libffi.call/struct1.c: Likewise.
1381 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1383 * src/pa/hpux32.S: Correct unwind offset calculation for
1385 * src/pa/linux.S: Likewise.
1387 2006-04-12 James E Wilson <wilson@specifix.com>
1390 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1391 (hfa_type_load): Call stf_spill.
1392 (hfa_type_store): Call ldf_fill.
1393 (ffi_call): Adjust calls to above routines. Add local temps for
1396 2006-04-10 Matthias Klose <doko@debian.org>
1398 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1399 directory names containing underscores.
1401 2006-04-07 James E Wilson <wilson@specifix.com>
1403 * testsuite/libffi.call/float4.c: New testcase.
1405 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1406 Andreas Tobler <a.tobler@schweiz.ch>
1408 * Makefile.am: Add PA_HPUX port.
1409 * Makefile.in: Regenerate.
1410 * include/Makefile.in: Likewise.
1411 * testsuite/Makefile.in: Likewise.
1412 * configure.ac: Add PA_HPUX rules.
1413 * configure: Regenerate.
1414 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1415 Add PA_HPUX and PA64_HPUX.
1416 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1417 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1418 (FFI_TYPE_SMALL_STRUCT2): Define.
1419 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1420 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1421 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1422 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1423 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1424 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1425 * src/pa/ffi.c (ROUND_DOWN): Delete.
1426 (fldw, fstw, fldd, fstd): Use '__asm__'.
1427 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1428 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1429 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1430 Simplify incrementing of stack slot variable. Change type of local
1431 'n' to unsigned int.
1432 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1434 (ffi_prep_cif_machdep): Likewise.
1435 (ffi_call): Likewise.
1436 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1437 return type to ffi_status. Simplify incrementing of stack slot
1438 variable. Only copy floating point argument registers when PA_LINUX
1439 is true. Reformat debug statement.
1440 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1441 FFI_TYPE_SMALL_STRUCT8.
1442 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1444 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1445 Add nops to cache flush. Add trampoline for PA_HPUX.
1446 * src/pa/hpux32.S: New file.
1447 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1448 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1449 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1450 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1451 argument type checks so that common argument types appear first.
1452 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1453 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1455 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1457 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1458 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1459 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1460 FFI_TYPE_LONGDOUBLE.
1461 (ffi_prep_args64): Assert using IBM extended double.
1462 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1463 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1464 (ffi_call): Handle FFI_LINUX.
1465 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1466 gpr3 return pointer as for struct return. Handle FFI_LINUX
1467 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1469 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1470 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1471 Don't use r6 as pointer to results, instead use sp offset. Don't
1472 make a special call to load lr with case table address, instead
1473 use offset from previous call.
1474 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1475 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1478 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1480 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1481 passed with FP registers correctly.
1482 (ffi_closure_helper_SYSV): Likewise.
1483 * src/sh64/sysv.S: Likewise.
1485 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1487 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1488 args and userdata unused.
1489 (closure_test_fn1): Mark cif and userdata unused.
1490 (main): Remove unused res.
1492 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1494 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1495 -O2, -O3, -Os and the warning flags -W -Wall.
1496 * testsuite/libffi.special/special.exp: Likewise.
1497 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1498 unused parameter unused for gcc or else do nothing.
1499 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1500 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1501 and userdata unused.
1502 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1503 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1504 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1505 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1506 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1507 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1508 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1509 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1510 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1511 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1512 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1513 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1514 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1515 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1516 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1517 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1518 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1519 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1520 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1521 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1522 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1523 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1524 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1525 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1527 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1529 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1531 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1532 void* to avoid compiler warning.
1534 (cls_struct_align_gn): Mark cif and userdata unused.
1535 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1537 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1539 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1541 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1543 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1545 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1546 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1547 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1549 (main): Cast res_call to silence gcc.
1550 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1552 (main): Cast res_call to silence gcc.
1553 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1555 (main): Cast res_call to silence gcc.
1556 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1558 (main): Cast res_call to silence gcc.
1559 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1561 (main): Cast res_call to silence gcc.
1562 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1564 (main): Cast res_call to silence gcc.
1565 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1567 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1568 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1570 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1571 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1573 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1574 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1576 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1577 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1579 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1580 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1581 and userdata unused.
1582 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1584 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1585 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1586 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1587 Cleanup white spaces.
1588 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1589 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1590 cif and userdata unused.
1591 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1593 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1594 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1595 formatters to silence gcc.
1596 (B_gn): Mark cif and userdata unused.
1597 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1599 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1600 (B_gn): Mark cif and userdata unused.
1601 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1603 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1604 (B_gn): Mark cif and userdata unused.
1605 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1607 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1608 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1609 * testsuite/libffi.call/problem1.c (stub): Likewise.
1610 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1612 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1613 in the last commit for this test case in the test case itself.
1614 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1616 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1617 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1618 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1619 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1620 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1621 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1623 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1625 * src/sh/sysv.S: Fix register numbers in the FDE for
1628 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1630 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1631 declaration to avoid a false negative on ix86. See PR323.
1633 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1635 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1636 and cast integer to void * if needed. Update the pointer to
1637 the FP register saved area correctly.
1639 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1641 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1643 * testsuite/libffi.call/nested_struct4.c: Likewise.
1645 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1647 * testsuite/libffi.call/return_dbl.c: New test case.
1648 * testsuite/libffi.call/return_dbl1.c: Likewise.
1649 * testsuite/libffi.call/return_dbl2.c: Likewise.
1650 * testsuite/libffi.call/return_fl.c: Likewise.
1651 * testsuite/libffi.call/return_fl1.c: Likewise.
1652 * testsuite/libffi.call/return_fl2.c: Likewise.
1653 * testsuite/libffi.call/return_fl3.c: Likewise.
1654 * testsuite/libffi.call/closure_fn6.c: Likewise.
1656 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1658 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1659 here to be used by other test cases too.
1661 * testsuite/libffi.call/nested_struct10.c: New test case.
1662 * testsuite/libffi.call/nested_struct9.c: Likewise.
1663 * testsuite/libffi.call/nested_struct8.c: Likewise.
1664 * testsuite/libffi.call/nested_struct7.c: Likewise.
1665 * testsuite/libffi.call/nested_struct6.c: Likewise.
1666 * testsuite/libffi.call/nested_struct5.c: Likewise.
1667 * testsuite/libffi.call/nested_struct4.c: Likewise.
1669 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1671 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1672 * configure: Rebuilt.
1674 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1676 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1677 instead do the shifting inline.
1678 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1679 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1680 and 8 byte structs, for the remaining struct sizes don't call
1681 __lshrdi3, instead do the shifting inline.
1683 2005-12-07 Thiemo Seufer <ths@networkno.de>
1685 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1686 missing parentheses.
1687 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1688 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1689 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1690 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1691 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1693 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1695 (ffi_prep_closure): Improve trampoline instruction scheduling.
1696 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1698 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1700 * src/powerpc/ffi.c: Formatting.
1701 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1702 (ffi_prep_args64): Likewise.
1704 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1706 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1707 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1709 2005-09-26 Tom Tromey <tromey@redhat.com>
1711 * testsuite/libffi.call/float1.c (value_type): New typedef.
1712 (CANARY): New define.
1713 (main): Check for result buffer overflow.
1714 * src/powerpc/linux64.S: Handle linux64 long double returns.
1715 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1716 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1718 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1721 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1723 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1725 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1727 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1728 (AH_BOTTOM): Add FFI_HIDDEN definition.
1729 * configure: Rebuilt.
1730 * fficonfig.h.in: Rebuilt.
1731 * src/powerpc/ffi.c (hidden): Remove.
1732 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1733 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1734 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1735 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1736 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1737 add FFI_HIDDEN to its prototype.
1738 (ffi_closure_SYSV_inner): New.
1739 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1740 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1742 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1745 * configure: Rebuilt.
1746 * configure.ac: Handle i*86-*-gnu*.
1748 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1750 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1751 DW_CFA_offset_extended_sf rather than
1752 DW_CFA_GNU_negative_offset_extended.
1753 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1755 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1757 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1759 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1760 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1761 partially on register.
1762 (ffi_closure_helper_SYSV): Likewise.
1763 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1765 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1767 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1769 * src/sh64/ffi.c (simple_type): Remove.
1770 (return_type): Handle small structures correctly.
1771 (ffi_prep_args): Likewise.
1772 (ffi_call): Likewise.
1773 (ffi_closure_helper_SYSV): Likewise.
1774 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1775 Emit position independent code if PIC and remove wrong datalabel
1776 prefixes from EH data.
1778 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1780 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1781 * Makefile.in: Regenerate.
1782 * include/Makefile.in: Likewise.
1783 * testsuite/Makefile.in: Likewise.
1784 * configure.ac: Add POWERPC_FREEBSD rules.
1785 * configure: Regenerate.
1786 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1787 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1788 * src/powerpc/ffi.c: Add flags to handle small structure returns
1790 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1792 (ffi_closure_helper_SYSV): Likewise.
1793 * src/powerpc/ppc_closure.S: Add return types for small structures.
1794 * src/powerpc/sysv.S: Add bits to handle small structures for
1797 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1799 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1800 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1801 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1803 2005-07-05 Randolph Chung <tausq@debian.org>
1805 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1806 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1807 structures. Kill compilation warnings.
1808 (ffi_closure_inner_LINUX): Print return values as hex in debug
1809 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1810 Properly handle 5-7 byte structure returns.
1811 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1812 (FFI_TYPE_SMALL_STRUCT2): Remove.
1813 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1814 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1815 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1816 (checksmst1, checksmst2): Remove.
1817 (checksmst3): Optimize handling of 3-byte struct returns.
1818 (checksmst567): Properly handle 5-7 byte struct returns.
1820 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1823 * src/mips/n32.S: Enforce PIC code.
1824 * src/mips/o32.S: Likewise.
1826 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1828 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1829 * configure: Regenerate.
1831 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1833 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1834 to call ffi_closure_helper_SYSV. Append @local instead.
1835 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1837 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1839 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1840 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1841 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1842 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1843 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1845 2005-05-09 Mike Stump <mrs@apple.com>
1847 * configure: Regenerate.
1849 2005-05-08 Richard Henderson <rth@redhat.com>
1852 * src/alpha/osf.S: Update unwind into to match code.
1854 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1855 Richard Henderson <rth@redhat.com>
1857 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1859 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1860 (ffi_prep_closure): Set carry bit if sse-used flag set.
1861 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1862 Only load sse registers if ssecount non-zero.
1863 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1865 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1867 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1868 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1869 * configure: Regenerate.
1871 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1873 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1874 have Tcl8.3-compatible intermediate variable.
1876 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1877 Hans-Peter Nilsson <hp@axis.com>
1879 * Makefile.am: Add CRIS support.
1880 * configure.ac: Likewise.
1881 * Makefile.in, configure, testsuite/Makefile.in,
1882 include/Makefile.in: Regenerate.
1883 * src/cris: New directory.
1884 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1885 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1887 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1888 \r?\n in output tests.
1890 2005-04-12 Mike Stump <mrs@apple.com>
1892 * configure: Regenerate.
1894 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1896 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1898 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1900 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1902 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1903 to 64 bits in ILP32 mode.
1904 Load 64 bits even for short data.
1906 2005-03-23 Mike Stump <mrs@apple.com>
1908 * src/powerpc/darwin.S: Update for -m64 multilib.
1909 * src/powerpc/darwin_closure.S: Likewise.
1911 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1913 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1914 Do not set tool_include_dir.
1915 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1917 * include/Makefile.am: Set gcc_version and toollibffidir.
1918 * include/Makefile.in: Regenerate.
1920 2005-02-22 Andrew Haley <aph@redhat.com>
1922 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1923 odd-numbered register pairs for 64-bit integer types.
1925 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1928 * testsuite/libffi.call/return_ll1.c: New test case.
1930 2005-02-11 Janis Johnson <janis187@us.ibm.com>
1932 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1933 * testsuite/libffi.call/float.c: Ditto.
1934 * testsuite/libffi.call/float2.c: Ditto.
1935 * testsuite/libffi.call/float3.c: Ditto.
1937 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1939 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1941 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1943 * testsuite/libffi.special/special.exp (cxx_options): Add
1946 2004-12-31 Richard Henderson <rth@redhat.com>
1948 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1949 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1950 offset parameters with a type parameter; deduce size and structure
1951 alignment. Update all users.
1953 2004-12-31 Richard Henderson <rth@redhat.com>
1955 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1956 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
1957 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
1958 into ffi_prep_closure.
1959 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1962 2004-12-27 Richard Henderson <rth@redhat.com>
1964 * src/x86/unix64.S: Fix typo in unwind info.
1966 2004-12-25 Richard Henderson <rth@redhat.com>
1968 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1969 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1970 (merge_classes): Check for it.
1972 (classify_argument): Pass byte_offset by value; perform all updates
1974 (examine_argument): Add classes argument; handle
1975 X86_64_COMPLEX_X87_CLASS.
1976 (ffi_prep_args): Merge into ...
1977 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1978 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1979 (ffi_fill_return_value): Remove.
1980 (ffi_prep_closure): Remove dead assert.
1981 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1982 Rewrite to use struct register_args instead of va_list. Create
1983 flags for handling structure returns.
1984 * src/x86/unix64.S: Remove dead strings.
1985 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1986 stack frame with ffi_call. Handle structure returns properly.
1987 (float2sse, floatfloat2sse, double2sse): Remove.
1988 (sse2float, sse2double, sse2floatfloat): Remove.
1989 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1990 to handle structure returns properly.
1992 2004-12-08 David Edelsohn <edelsohn@gnu.org>
1994 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1996 * Makefile.in: Regenerated.
1998 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2000 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2001 * configure, aclocal.m4, Makefile.in: Regenerate.
2002 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2004 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2006 * configure: Regenerate for libtool change.
2008 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2010 * configure: Regenerate for libtool reversion.
2012 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2014 * configure: Regenerate for libtool change.
2016 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2018 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2020 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2022 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2023 of jal. Use an absolute encoding for the frame information.
2025 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2027 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2028 * acinclude.m4: Delete logic for sincludes.
2029 * aclocal.m4, Makefile.in, configure: Regenerate.
2030 * include/Makefile: Likewise.
2031 * testsuite/Makefile: Likewise.
2033 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2035 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2036 on a 8-byte boundary.
2037 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2039 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2041 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2042 long long values. Round stack allocation to a multiple of 8 bytes
2043 for ATPCS compatibility.
2044 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2045 names. Handle returning long long types. Add Thumb and interworking
2046 support. Improve soft-float code.
2048 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2050 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2051 (libffi_exit): New function.
2052 (libffi_init): Build the testglue wrapper if needed.
2054 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2057 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2059 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2061 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2063 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2065 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2066 * testsuite/libffi.call/float3.c: New test case.
2068 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2070 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2071 the function returning a structure pointed with R2.
2072 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2073 the structure return value if T bit set. Emit position
2074 independent code and EH data if PIC.
2076 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2078 * Makefile.am: Add m32r support.
2079 * configure.ac: Likewise.
2080 * Makefile.in: Regenerate.
2081 * confiugre: Regenerate.
2082 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2083 (uint64, sint64, double, longdouble)
2084 * src/m32r: New directory.
2085 * src/m32r/ffi.c: New file.
2086 * src/m32r/sysv.S: Likewise.
2087 * src/m32r/ffitarget.h: Likewise.
2089 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2091 * testsuite/libffi.call/negint.c: New test case.
2093 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2096 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2097 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2098 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2101 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2103 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2104 * testsuite/libffi.call/promotion.c: Likewise.
2105 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2107 * testsuite/libffi.call/return_sc.c: Likewise.
2108 * testsuite/libffi.call/return_uc.c: Likewise.
2110 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2112 * src/powerpc/darwin.S: Fix comments and identation.
2113 * src/powerpc/darwin_closure.S: Likewise.
2115 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2117 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2118 (ffi_prep_args): Handle longdouble arguments.
2119 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2121 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2122 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2124 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2125 * src/types.c: Defined longdouble size and alignment for darwin.
2127 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2129 * src/powerpc/aix.S: Remove whitespaces.
2130 * src/powerpc/aix_closure.S: Likewise.
2131 * src/powerpc/asm.h: Likewise.
2132 * src/powerpc/ffi.c: Likewise.
2133 * src/powerpc/ffitarget.h: Likewise.
2134 * src/powerpc/linux64.S: Likewise.
2135 * src/powerpc/linux64_closure.S: Likewise.
2136 * src/powerpc/ppc_closure.S: Likewise.
2137 * src/powerpc/sysv.S: Likewise.
2139 2004-08-30 Anthony Green <green@redhat.com>
2141 * Makefile.am: Add frv support.
2142 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2143 * configure.ac: Read configure.host.
2144 * configure.in: Read configure.host.
2145 * configure.host: New file. frv-elf needs libgloss.
2146 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2147 alignment. This is needed to frv and shouldn't harm the others.
2148 * include/ffi_common.h (ALIGN_DOWN): New macro.
2149 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2151 2004-08-24 David Daney <daney@avtrex.com>
2153 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2154 * testsuite/libffi.call/closure_fn1.c: Likewise.
2155 * testsuite/libffi.call/closure_fn2.c Likewise.
2156 * testsuite/libffi.call/closure_fn3.c: Likewise.
2157 * testsuite/libffi.call/closure_fn4.c: Likewise.
2158 * testsuite/libffi.call/closure_fn5.c: Likewise.
2159 * testsuite/libffi.call/cls_18byte.c: Likewise.
2160 * testsuite/libffi.call/cls_19byte.c: Likewise.
2161 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2162 * testsuite/libffi.call/cls_20byte.c: Likewise.
2163 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2164 * testsuite/libffi.call/cls_24byte.c: Likewise.
2165 * testsuite/libffi.call/cls_2byte.c: Likewise.
2166 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2167 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2168 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2169 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2170 * testsuite/libffi.call/cls_4byte.c: Likewise.
2171 * testsuite/libffi.call/cls_64byte.c: Likewise.
2172 * testsuite/libffi.call/cls_6byte.c: Likewise.
2173 * testsuite/libffi.call/cls_7byte.c: Likewise.
2174 * testsuite/libffi.call/cls_8byte.c: Likewise.
2175 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2176 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2177 * testsuite/libffi.call/cls_align_double.c: Likewise.
2178 * testsuite/libffi.call/cls_align_float.c: Likewise.
2179 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2180 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2181 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2182 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2183 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2184 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2185 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2186 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2187 * testsuite/libffi.call/cls_double.c: Likewise.
2188 * testsuite/libffi.call/cls_float.c: Likewise.
2189 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2190 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2191 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2192 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2193 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2194 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2195 * testsuite/libffi.call/cls_schar.c: Likewise.
2196 * testsuite/libffi.call/cls_sint.c: Likewise.
2197 * testsuite/libffi.call/cls_sshort.c: Likewise.
2198 * testsuite/libffi.call/cls_uchar.c: Likewise.
2199 * testsuite/libffi.call/cls_uint.c: Likewise.
2200 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2201 * testsuite/libffi.call/cls_ushort.c: Likewise.
2202 * testsuite/libffi.call/nested_struct.c: Likewise.
2203 * testsuite/libffi.call/nested_struct1.c: Likewise.
2204 * testsuite/libffi.call/nested_struct2.c: Likewise.
2205 * testsuite/libffi.call/nested_struct3.c: Likewise.
2206 * testsuite/libffi.call/problem1.c: Likewise.
2207 * testsuite/libffi.special/unwindtest.cc: Likewise.
2208 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2210 * testsuite/libffi.call/cls_16byte.c: Likewise.
2211 * testsuite/libffi.call/cls_5byte.c: Likewise.
2213 2004-08-23 David Daney <daney@avtrex.com>
2216 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2217 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2218 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2219 parameters and return types.
2220 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2221 (ffi_prep_closure): Ditto.
2222 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2223 alignment calculations.
2224 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2225 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2227 2004-08-14 Casey Marshall <csm@gnu.org>
2229 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2230 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2231 integer (O32 ABI only).
2232 (ffi_prep_closure): new function.
2233 (ffi_closure_mips_inner_O32): new function.
2234 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2235 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2236 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2237 64 bit integers correctly.
2238 (ffi_closure_O32): new function.
2239 Added DWARF-2 unwind info for both functions.
2241 2004-08-10 Andrew Haley <aph@redhat.com>
2243 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2245 2004-08-01 Robert Millan <robertmh@gnu.org>
2247 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2248 * configure: Regenerate.
2250 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2252 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2253 and mmap() explicitly instead of relying on preset autoconf cache
2255 * aclocal.m4: Regenerate.
2256 * configure: Regenerate.
2258 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2260 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2261 (ffi_check_float_struct): Remove unused prototype.
2263 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2265 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2266 character on Darwin, use '\n\t' instead.
2268 2004-06-26 Matthias Klose <doko@debian.org>
2270 * libtool-version: Fix typo in revision/age.
2272 2004-06-17 Matthias Klose <doko@debian.org>
2274 * libtool-version: New.
2275 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2276 * Makefile.in: Regenerate.
2278 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2280 * Makefile.am: Remove useless multilib rules.
2281 * Makefile.in: Regenerate.
2282 * aclocal.m4: Regenerate with automake 1.8.5.
2283 * configure.ac: Remove useless multilib configury.
2284 * configure: Regenerate.
2286 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2288 * .cvsignore: New file.
2290 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2292 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2294 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2297 2004-06-06 Sean McNeil <sean@mcneil.com>
2299 * configure.ac: Add x86_64-*-freebsd* support.
2300 * configure: Regenerate.
2302 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2305 * configure.ac: Test for existence of mmap and sys/mman.h before
2306 checking blacklist. Fix suggested by Jim Wilson.
2307 * configure: Regenerate.
2309 2004-04-26 Matt Austern <austern@apple.com>
2311 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2313 * src/powerpc/darwin_closure.S: Likewise.
2315 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2317 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2318 error. Reported by Thomas Heller <theller@python.net>.
2319 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2320 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2322 2004-03-20 Matthias Klose <doko@debian.org>
2324 * src/pa/linux.S: Fix typo.
2326 2004-03-19 Matthias Klose <doko@debian.org>
2328 * Makefile.am: Update.
2329 * Makefile.in: Regenerate.
2330 * src/pa/ffi.h.in: Remove.
2331 * src/pa/ffitarget.h: New file.
2333 2004-02-10 Randolph Chung <tausq@debian.org>
2335 * Makefile.am: Add PA support.
2336 * Makefile.in: Regenerate.
2337 * include/Makefile.in: Regenerate.
2338 * configure.ac: Add PA target.
2339 * configure: Regenerate.
2340 * src/pa/ffi.c: New file.
2341 * src/pa/ffi.h.in: Add PA support.
2342 * src/pa/linux.S: New file.
2343 * prep_cif.c: Add PA support.
2345 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2347 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2349 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2350 with ecif->cif->flags.
2351 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2353 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2354 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2355 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2356 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2358 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2360 * configure.in: Rename file to ...
2361 * configure.ac: ... this.
2362 * fficonfig.h.in: Regenerate.
2363 * Makefile.in: Regenerate.
2364 * include/Makefile.in: Regenerate.
2365 * testsuite/Makefile.in: Regenerate.
2367 2004-03-12 Matt Austern <austern@apple.com>
2369 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2370 changes in EH format resulting from addition of linkonce support.
2371 * src/powerpc/darwin_closure.S: Likewise.
2373 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2374 Paolo Bonzini <bonzini@gnu.org>
2376 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2377 Remove VPATH. Remove rules for object files. Remove multilib support.
2378 (AM_CCASFLAGS): Add.
2379 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2380 (AC_PREREQ): Bump version to 2.59.
2381 (AC_INIT): Fill with version info and bug address.
2382 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2383 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2384 De-precious CC so that the right flags are passed down to multilibs.
2385 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2386 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2387 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2388 * configure: Rebuilt.
2389 * aclocal.m4: Likewise.
2390 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2391 * fficonfig.h.in: Likewise.
2393 2004-03-11 Andreas Schwab <schwab@suse.de>
2395 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2396 arguments from fp registers only for the first 8 parameter slots.
2397 Don't convert a float parameter when passed in memory.
2399 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2401 * configure: Regenerate for config/accross.m4 correction.
2403 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2405 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2406 ecif->cif->bytes to bytes.
2407 (ffi_prep_cif_machdep): Add braces around nested if statement.
2409 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2411 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2413 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2414 (ffi_closure_helper_LINUX64): Fix typo.
2415 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2417 * testsuite/libffi.call/float.c: Likewise.
2418 * testsuite/libffi.call/float2.c: Likewise.
2420 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2422 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2423 long double function return and long double arg handling.
2424 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2425 Use "end_pfr" instead of "nf". Correct long double handling.
2427 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2429 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2430 space for long double return value. Adjust stack frame and offsets.
2431 Load f2 long double return.
2433 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2435 * src/types.c: Use 16 byte long double for POWERPC64.
2437 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2439 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2440 when the structure return address is passed in %o0.
2441 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2442 (ffi_v9_layout_struct): Align the field following a nested structure
2443 on a word boundary. Use memmove instead of memcpy.
2444 (ffi_call): Update call to ffi_V9_return_struct.
2445 (ffi_prep_closure): Define 'ctx' only for V8.
2446 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2447 and ffi_closure_sparc_inner_v9.
2448 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2449 Always skip the structure return address. For structures and long
2450 doubles, copy the argument directly.
2451 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2452 if required. Shift the maximum floating-point slot accordingly. For
2453 big structures, copy the argument directly; otherwise, left-justify the
2454 argument and call ffi_v9_layout_struct to lay out the structure on
2456 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2457 (ffi_closure_v8): Pass the structure return address. Update call to
2458 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2459 Skip the 'unimp' insn when returning long doubles and structures.
2460 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2461 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2462 FFI_TYPE_INT handling. Load structures both in integers and
2463 floating-point registers on return.
2464 * README: Update status of the SPARC port.
2466 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2468 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2470 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2472 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2474 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2475 value as of type ffi_arg, not unsigned int.
2477 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2479 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2482 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2484 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2487 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2489 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2492 2003-12-10 Richard Henderson <rth@redhat.com>
2494 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2495 size_t instead of int.
2497 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2499 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2500 * testsuite/libffi.call/many_win32.c (main): Replace variable
2501 int i with unsigned long ul.
2503 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2504 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2505 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2506 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2507 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2508 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2509 * testsuite/libffi.call/cls_align_float.c: Likewise.
2510 * testsuite/libffi.call/cls_align_double.c: Likewise.
2511 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2512 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2514 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2517 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2518 Align arguments to 32 bits.
2520 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2523 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2524 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2525 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2526 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2527 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2528 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2530 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2532 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2534 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2535 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2537 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2539 * Makefile.in: Rebuilt.
2540 * configure: Likewise.
2541 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2544 2003-11-21 Andreas Jaeger <aj@suse.de>
2545 Andreas Tobler <a.tobler@schweiz.ch>
2547 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2548 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2549 * Makefile.in: Rebuilt.
2550 * aclocal.m4: Likewise.
2551 * configure: Likewise.
2552 * fficonfig.h.in: Likewise.
2553 * testsuite/lib/libffi-dg.exp: Add include dir.
2554 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2555 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2556 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2557 for ffi_closure if available.
2558 * testsuite/libffi.call/closure_fn1.c: Likewise.
2559 * testsuite/libffi.call/closure_fn2.c: Likewise.
2560 * testsuite/libffi.call/closure_fn3.c: Likewise.
2561 * testsuite/libffi.call/closure_fn4.c: Likewise.
2562 * testsuite/libffi.call/closure_fn5.c: Likewise.
2563 * testsuite/libffi.call/cls_12byte.c: Likewise.
2564 * testsuite/libffi.call/cls_16byte.c: Likewise.
2565 * testsuite/libffi.call/cls_18byte.c: Likewise.
2566 * testsuite/libffi.call/cls_19byte.c: Likewise.
2567 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2568 * testsuite/libffi.call/cls_20byte.c: Likewise.
2569 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2570 * testsuite/libffi.call/cls_24byte.c: Likewise.
2571 * testsuite/libffi.call/cls_2byte.c: Likewise.
2572 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2573 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2574 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2575 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2576 * testsuite/libffi.call/cls_4byte.c: Likewise.
2577 * testsuite/libffi.call/cls_5byte.c: Likewise.
2578 * testsuite/libffi.call/cls_64byte.c: Likewise.
2579 * testsuite/libffi.call/cls_6byte.c: Likewise.
2580 * testsuite/libffi.call/cls_7byte.c: Likewise.
2581 * testsuite/libffi.call/cls_8byte.c: Likewise.
2582 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2583 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2584 * testsuite/libffi.call/cls_double.c: Likewise.
2585 * testsuite/libffi.call/cls_float.c: Likewise.
2586 * testsuite/libffi.call/cls_schar.c: Likewise.
2587 * testsuite/libffi.call/cls_sint.c: Likewise.
2588 * testsuite/libffi.call/cls_sshort.c: Likewise.
2589 * testsuite/libffi.call/cls_uchar.c: Likewise.
2590 * testsuite/libffi.call/cls_uint.c: Likewise.
2591 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2592 * testsuite/libffi.call/cls_ushort.c: Likewise.
2593 * testsuite/libffi.call/nested_struct.c: Likewise.
2594 * testsuite/libffi.call/nested_struct1.c: Likewise.
2595 * testsuite/libffi.call/nested_struct2.c: Likewise.
2596 * testsuite/libffi.call/nested_struct3.c: Likewise.
2597 * testsuite/libffi.call/problem1.c: Likewise.
2598 * testsuite/libffi.special/unwindtest.cc: Likewise.
2600 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2602 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2604 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2606 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2607 Add -lgcc_s to additional flags.
2609 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2611 * configure.in, include/Makefile.am: PR libgcj/11147, install
2612 the ffitarget.h header file in a gcc versioned and target
2614 * configure: Regenerated.
2615 * Makefile.in, include/Makefile.in: Likewise.
2616 * testsuite/Makefile.in: Likewise.
2618 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2620 * testsuite/libffi.call/closure_fn0.c: Print result and check
2621 with dg-output to make debugging easier.
2622 * testsuite/libffi.call/closure_fn1.c: Likewise.
2623 * testsuite/libffi.call/closure_fn2.c: Likewise.
2624 * testsuite/libffi.call/closure_fn3.c: Likewise.
2625 * testsuite/libffi.call/closure_fn4.c: Likewise.
2626 * testsuite/libffi.call/closure_fn5.c: Likewise.
2627 * testsuite/libffi.call/cls_12byte.c: Likewise.
2628 * testsuite/libffi.call/cls_16byte.c: Likewise.
2629 * testsuite/libffi.call/cls_18byte.c: Likewise.
2630 * testsuite/libffi.call/cls_19byte.c: Likewise.
2631 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2632 * testsuite/libffi.call/cls_20byte.c: Likewise.
2633 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2634 * testsuite/libffi.call/cls_24byte.c: Likewise.
2635 * testsuite/libffi.call/cls_2byte.c: Likewise.
2636 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2637 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2638 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2639 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2640 * testsuite/libffi.call/cls_4byte.c: Likewise.
2641 * testsuite/libffi.call/cls_5byte.c: Likewise.
2642 * testsuite/libffi.call/cls_64byte.c: Likewise.
2643 * testsuite/libffi.call/cls_6byte.c: Likewise.
2644 * testsuite/libffi.call/cls_7byte.c: Likewise.
2645 * testsuite/libffi.call/cls_8byte.c: Likewise.
2646 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2647 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2648 * testsuite/libffi.call/cls_double.c: Likewise.
2649 * testsuite/libffi.call/cls_float.c: Likewise.
2650 * testsuite/libffi.call/cls_schar.c: Likewise.
2651 * testsuite/libffi.call/cls_sint.c: Likewise.
2652 * testsuite/libffi.call/cls_sshort.c: Likewise.
2653 * testsuite/libffi.call/cls_uchar.c: Likewise.
2654 * testsuite/libffi.call/cls_uint.c: Likewise.
2655 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2656 * testsuite/libffi.call/cls_ushort.c: Likewise.
2657 * testsuite/libffi.call/problem1.c: Likewise.
2659 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2662 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2664 * testsuite/libffi.call/cls_9byte2.c: New test case.
2665 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2666 * testsuite/libffi.call/cls_64byte.c: Likewise.
2667 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2668 * testsuite/libffi.call/cls_19byte.c: Likewise.
2669 * testsuite/libffi.call/cls_18byte.c: Likewise.
2670 * testsuite/libffi.call/closure_fn4.c: Likewise.
2671 * testsuite/libffi.call/closure_fn5.c: Likewise.
2672 * testsuite/libffi.call/cls_schar.c: Likewise.
2673 * testsuite/libffi.call/cls_sint.c: Likewise.
2674 * testsuite/libffi.call/cls_sshort.c: Likewise.
2675 * testsuite/libffi.call/nested_struct2.c: Likewise.
2676 * testsuite/libffi.call/nested_struct3.c: Likewise.
2678 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2680 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2681 * testsuite/libffi.call/cls_uchar.c: Likewise.
2682 * testsuite/libffi.call/cls_uint.c: Likewise.
2683 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2684 * testsuite/libffi.call/cls_ushort.c: Likewise.
2685 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2687 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2689 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2692 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2694 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2695 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2697 2003-10-22 David Daney <ddaney@avtrex.com>
2699 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2700 __attribute__((__mode__(__SI__))) and friends.
2702 2003-10-22 Andreas Schwab <schwab@suse.de>
2704 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2706 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2708 * configure.in: AC_LINK_FILES(ffitarget.h).
2709 * configure: Regenerate.
2710 * Makefile.in: Likewise.
2711 * include/Makefile.in: Likewise.
2712 * testsuite/Makefile.in: Likewise.
2713 * fficonfig.h.in: Likewise.
2715 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2716 Richard Henderson <rth@redhat.com>
2718 Avoid that ffi.h includes fficonfig.h.
2720 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2721 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2722 (TARGET_SRC_MIPS_SGI): Removed.
2723 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2724 (MIPS_SGI): Removed.
2725 (CLEANFILES): Removed.
2726 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2728 * acconfig.h: Removed.
2729 * configure.in: Compute sizeofs only for double and long double.
2730 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2731 into AC_DEFINE instead of using acconfig.h. Create
2732 include/ffitarget.h instead of include/fficonfig.h. Rename
2733 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2734 AC_DEFINE EH_FRAME_FLAGS.
2735 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2736 (hack_DATA): Add ffitarget.h.
2737 * include/ffi.h.in: Remove all system specific definitions.
2738 Declare raw API even if it is not installed, why bother?
2739 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2740 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2741 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2742 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2743 * include/ffi_common.h (bool): Do not define.
2744 (ffi_assert): Accept failed assertion.
2745 (ffi_type_test): Return void and accept file/line.
2746 (FFI_ASSERT): Pass stringized failed assertion.
2747 (FFI_ASSERT_AT): New macro.
2748 (FFI_ASSERT_VALID_TYPE): New macro.
2749 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2750 UINT64, SINT64): Define here with gcc's __attribute__ macro
2752 (FLOAT32, ALIGN): Define here instead of in ffi.h
2753 * include/ffi-mips.h: Removed. Its content moved to
2754 src/mips/ffitarget.h after separating assembly and C sections.
2755 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2756 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2757 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2758 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2759 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2760 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2761 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2762 (ffi_type_test): Rewritten.
2763 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2764 FFI_ASSERT_VALID_TYPE.
2765 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2766 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2767 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2768 src/s390/ffitarget.h, src/sh/ffitarget.h,
2769 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2770 src/x86/ffitarget.h: New files.
2771 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2772 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2773 src/powerpc/aix.S, src/powerpc/darwin.S,
2774 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2775 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2776 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2777 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2778 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2781 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2783 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2784 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2786 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2788 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2789 Used when FFI_DEBUG = 1.
2791 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2793 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2796 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2798 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2799 FFI_MIPS_O32 for O32 ABI.
2801 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2803 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2804 SPARC64. Cleanup whitespaces.
2806 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2808 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2809 strongarm, xscale. Cleanup whitespaces.
2810 * testsuite/libffi.call/closure_fn1.c: Likewise.
2811 * testsuite/libffi.call/closure_fn2.c: Likewise.
2812 * testsuite/libffi.call/closure_fn3.c: Likewise.
2813 * testsuite/libffi.call/cls_12byte.c: Likewise.
2814 * testsuite/libffi.call/cls_16byte.c: Likewise.
2815 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2816 * testsuite/libffi.call/cls_20byte.c: Likewise.
2817 * testsuite/libffi.call/cls_24byte.c: Likewise.
2818 * testsuite/libffi.call/cls_2byte.c: Likewise.
2819 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2820 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2821 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2822 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2823 * testsuite/libffi.call/cls_4byte.c: Likewise.
2824 * testsuite/libffi.call/cls_5byte.c: Likewise.
2825 * testsuite/libffi.call/cls_6byte.c: Likewise.
2826 * testsuite/libffi.call/cls_7byte.c: Likewise.
2827 * testsuite/libffi.call/cls_8byte.c: Likewise.
2828 * testsuite/libffi.call/cls_double.c: Likewise.
2829 * testsuite/libffi.call/cls_float.c: Likewise.
2830 * testsuite/libffi.call/cls_uchar.c: Likewise.
2831 * testsuite/libffi.call/cls_uint.c: Likewise.
2832 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2833 * testsuite/libffi.call/cls_ushort.c: Likewise.
2834 * testsuite/libffi.call/nested_struct.c: Likewise.
2835 * testsuite/libffi.call/nested_struct1.c: Likewise.
2836 * testsuite/libffi.call/problem1.c: Likewise.
2837 * testsuite/libffi.special/unwindtest.cc: Likewise.
2838 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2840 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2842 * src/powerpc/aix.S: Cleanup whitespaces.
2843 * src/powerpc/aix_closure.S: Likewise.
2845 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2847 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2848 * src/powerpc/darwin_closure.S: Likewise.
2849 * src/powerpc/ffi_darwin.c: Likewise.
2851 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2852 David Edelsohn <edelsohn@gnu.org>
2854 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2855 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2857 * src/powerpc/darwin_closure.S: Likewise.
2858 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2859 according to the Darwin/AIX ABI.
2860 (ffi_prep_cif_machdep): Likewise.
2861 (ffi_closure_helper_DARWIN): Likewise.
2862 Remove the outgoing parameter stack logic. Simplify the evaluation
2863 of the different CASE types.
2864 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2865 statement in the trampoline code.
2867 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2869 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2870 for the register size.
2871 (ffi_closure_helper_SYSV): Handle the structure return value
2873 (ffi_closure_helper_SYSV): Return the appropriate type when
2874 the registers are used for the structure return value.
2875 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2876 the 64-bit return value. Update copyright years.
2878 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2880 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2881 srcdir for ffi_mips.h.
2883 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2885 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2887 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2888 placement of float result.
2889 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2890 cast of "resp" for big-endian 64 bit machines.
2892 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2894 * src/types.c (double, longdouble): Merge identical SH and ARM
2895 typedefs, and add POWERPC64.
2896 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2897 struct split over gpr and rest.
2898 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2899 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2901 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2903 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2906 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2908 * configure: Regenerate.
2910 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2912 * Makefile.am: Remove build rules for ffitest.
2913 * Makefile.in: Rebuilt.
2915 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2917 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2918 about implicit declaration of abort().
2920 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2922 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2923 * Makefile.in: Rebuilt.
2924 * configure.in: Add dejagnu test framework.
2925 * configure: Rebuilt.
2927 * testsuite/Makefile.am: New file.
2928 * testsuite/Makefile.in: Built
2929 * testsuite/lib/libffi-dg.exp: New file.
2930 * testsuite/config/default.exp: Likewise.
2931 * testsuite/libffi.call/call.exp: Likewise.
2932 * testsuite/libffi.call/ffitest.h: Likewise.
2933 * testsuite/libffi.call/closure_fn0.c: Likewise.
2934 * testsuite/libffi.call/closure_fn1.c: Likewise.
2935 * testsuite/libffi.call/closure_fn2.c: Likewise.
2936 * testsuite/libffi.call/closure_fn3.c: Likewise.
2937 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2938 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2939 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2940 * testsuite/libffi.call/cls_2byte.c: Likewise.
2941 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2942 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2943 * testsuite/libffi.call/cls_4byte.c: Likewise.
2944 * testsuite/libffi.call/cls_5byte.c: Likewise.
2945 * testsuite/libffi.call/cls_6byte.c: Likewise.
2946 * testsuite/libffi.call/cls_7byte.c: Likewise.
2947 * testsuite/libffi.call/cls_8byte.c: Likewise.
2948 * testsuite/libffi.call/cls_12byte.c: Likewise.
2949 * testsuite/libffi.call/cls_16byte.c: Likewise.
2950 * testsuite/libffi.call/cls_20byte.c: Likewise.
2951 * testsuite/libffi.call/cls_24byte.c: Likewise.
2952 * testsuite/libffi.call/cls_double.c: Likewise.
2953 * testsuite/libffi.call/cls_float.c: Likewise.
2954 * testsuite/libffi.call/cls_uchar.c: Likewise.
2955 * testsuite/libffi.call/cls_uint.c: Likewise.
2956 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2957 * testsuite/libffi.call/cls_ushort.c: Likewise.
2958 * testsuite/libffi.call/float.c: Likewise.
2959 * testsuite/libffi.call/float1.c: Likewise.
2960 * testsuite/libffi.call/float2.c: Likewise.
2961 * testsuite/libffi.call/many.c: Likewise.
2962 * testsuite/libffi.call/many_win32.c: Likewise.
2963 * testsuite/libffi.call/nested_struct.c: Likewise.
2964 * testsuite/libffi.call/nested_struct1.c: Likewise.
2965 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2966 * testsuite/libffi.call/problem1.c: Likewise.
2967 * testsuite/libffi.call/promotion.c: Likewise.
2968 * testsuite/libffi.call/return_ll.c: Likewise.
2969 * testsuite/libffi.call/return_sc.c: Likewise.
2970 * testsuite/libffi.call/return_uc.c: Likewise.
2971 * testsuite/libffi.call/strlen.c: Likewise.
2972 * testsuite/libffi.call/strlen_win32.c: Likewise.
2973 * testsuite/libffi.call/struct1.c: Likewise.
2974 * testsuite/libffi.call/struct2.c: Likewise.
2975 * testsuite/libffi.call/struct3.c: Likewise.
2976 * testsuite/libffi.call/struct4.c: Likewise.
2977 * testsuite/libffi.call/struct5.c: Likewise.
2978 * testsuite/libffi.call/struct6.c: Likewise.
2979 * testsuite/libffi.call/struct7.c: Likewise.
2980 * testsuite/libffi.call/struct8.c: Likewise.
2981 * testsuite/libffi.call/struct9.c: Likewise.
2982 * testsuite/libffi.special/special.exp: New file.
2983 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2984 * testsuite/libffi.special/unwindtest.cc: Likewise.
2987 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2989 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2992 2003-08-02 Alan Modra <amodra@bigpond.net.au>
2994 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2996 (ffi_closure_helper_LINUX64): Likewise.
2997 * src/powerpc/linux64.S: Remove code writing to parm save area.
2998 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2999 address in lr from ffi_closure_helper_LINUX64 call to calculate
3000 table address. Optimize function tail.
3002 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3004 * src/sparc/ffi.c: Handle all floating point registers.
3005 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3007 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3009 * README: Note that libffi is not part of GCC. Update the project
3012 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3014 * src/powerpc/ppc_closure.S: Include ffi.h.
3016 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3018 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3019 Use C style comments.
3021 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3023 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3024 * Makefile.in: Regenerate.
3025 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3026 * configure: Regenerate.
3027 * include/ffi.h.in: Add SHmedia support.
3028 * src/sh64/ffi.c: New file.
3029 * src/sh64/sysv.S: New file.
3031 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3033 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3034 should be read-only.
3035 * configure: Rebuilt.
3036 * fficonfig.h.in: Rebuilt.
3037 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3038 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3039 * src/powerpc/linux64.S: Likewise.
3040 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3041 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3042 if -fpic/-fPIC/-mrelocatable.
3043 * src/powerpc/powerpc_closure.S: Likewise.
3044 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3045 #write in .eh_frame flags.
3046 * src/sparc/v9.S: Likewise.
3047 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3048 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3049 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3051 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3053 Fixes PR bootstrap/10656
3054 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3055 support for .register pseudo-op.
3056 * src/sparc/v8.S: Use it.
3057 * fficonfig.h.in: Rebuilt.
3058 * configure: Rebuilt.
3060 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3062 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3063 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3064 Make it the default on POWERPC64.
3065 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3066 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3067 * configure: Rebuilt.
3068 * src/powerpc/ffi.c (hidden): Define.
3069 (ffi_prep_args_SYSV): Renamed from
3070 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3071 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3072 ASM_NEEDS_REGISTERS64): New.
3073 (ffi_prep_args64): New function.
3074 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3075 (ffi_call): Likewise.
3076 (ffi_prep_closure): Likewise.
3077 (flush_icache): Surround by #ifndef POWERPC64.
3078 (ffi_dblfl): New union type.
3079 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3080 (ffi_closure_helper_LINUX64): New function.
3081 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3083 * src/powerpc/sysv.S: Likewise.
3084 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3085 * src/powerpc/linux64.S: New file.
3086 * src/powerpc/linux64_closure.S: New file.
3087 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3088 src/powerpc/linux64_closure.S.
3089 (TARGET_SRC_POWERPC): Likewise.
3091 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3092 closure_test_fn3): Fix result printing on big-endian 64-bit
3094 (main): Print tst2_arg instead of uninitialized tst2_result.
3096 * src/ffitest.c (main): Hide what closure pointer really points to
3099 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3101 * configure.in (arm-*-netbsdelf*): Add configuration.
3102 (configure): Regenerated.
3104 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3106 * include/Makefile.in: Regenerate.
3108 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3110 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3112 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3113 Receive closure pointer through parameter, read args using
3114 __builtin_dwarf_cfa.
3115 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3117 2003-03-12 Andreas Schwab <schwab@suse.de>
3119 * configure.in: Avoid trailing /. in toolexeclibdir.
3120 * configure: Rebuilt.
3122 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3124 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3126 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3128 * libffi/src/powerpc/darwin_closure.S:
3129 Fix alignement bug, allocate 8 bytes for the result.
3130 * libffi/src/powerpc/aix_closure.S:
3132 * libffi/src/powerpc/ffi_darwin.c:
3133 Update stackframe description for aix/darwin_closure.S.
3135 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3137 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3140 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3141 Andreas Schwab <schwab@suse.de>
3143 * configure.in: Adjust command to source config-ml.in to account
3144 for changes to the libffi_basedir definition.
3145 (libffi_basedir): Remove ${srcdir} from value and include trailing
3148 * configure: Regenerate.
3150 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3152 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3154 2003-01-28 Andrew Haley <aph@redhat.com>
3156 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3157 * src/x86/ffi64.c (ffi_prep_closure): New.
3158 (ffi_closure_UNIX64_inner): New.
3159 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3161 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
3163 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3164 Remove USE_LIBDIR conditional.
3165 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3166 * Makefile.in, configure: Rebuilt.
3168 2003-01027 David Edelsohn <edelsohn@gnu.org>
3170 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3171 * Makefile.in: Regenerate.
3173 2003-01-22 Andrew Haley <aph@redhat.com>
3175 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3178 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3180 * src/powerpc/darwin.S: Add unwind info.
3181 * src/powerpc/darwin_closure.S: Likewise.
3183 2003-01-14 Andrew Haley <aph@redhat.com>
3185 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3186 (ffi_prep_cif_machdep): Likewise.
3187 * src/x86/unix64.S: Add unwind info.
3189 2003-01-14 Andreas Jaeger <aj@suse.de>
3191 * src/ffitest.c (main): Only use ffi_closures if those are
3194 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3196 * libffi/src/ffitest.c
3197 add closure testcases
3199 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3201 * libffi/src/powerpc/ffi.c
3202 fix alignment bug for float (4 byte aligned iso 8 byte)
3204 2003-01-09 Geoffrey Keating <geoffk@apple.com>
3206 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3207 * src/powerpc/darwin.S: Remove RCS version string.
3209 2003-01-03 Jeff Sturm <jsturm@one-point.com>
3211 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3212 * src/ffitest.c (main): Use static storage for closure.
3213 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3214 * src/sparc/v8.S (ffi_closure_v8): New.
3215 * src/sparc/v9.S (ffi_closure_v9): New.
3217 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3219 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3220 enumeration for X86_WIN32.
3221 * src/x86/win32.S: Added ffi_call_STDCALL function
3223 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3224 switch cases for recognising FFI_STDCALL and
3225 calling ffi_call_STDCALL if target is X86_WIN32.
3226 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3227 stdcall versions of the "my_strlen" and "many"
3228 test functions (for X86_WIN32).
3229 Added test cases to test stdcall invocation using
3232 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3234 * src/sh/sysv.S: Add DWARF2 unwind info.
3236 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3238 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3240 2002-11-26 Jim Wilson <wilson@redhat.com>
3242 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3244 2002-11-23 H.J. Lu <hjl@gnu.org>
3246 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3247 Include ../config/accross.m4.
3248 * aclocal.m4; Rebuild.
3249 * configure: Likewise.
3251 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3253 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3255 2002-11-11 DJ Delorie <dj@redhat.com>
3257 * configure.in: Look for common files in the right place.
3259 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3261 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3262 raw data as _Jv_word values, not ffi_raw.
3263 (ffi_java_ptrarray_to_raw): Likewise.
3264 (ffi_java_rvalue_to_raw): New function.
3265 (ffi_java_raw_call): Call it.
3266 (ffi_java_raw_to_rvalue): New function.
3267 (ffi_java_translate_args): Call it.
3268 * src/ffitest.c (closure_test_fn): Interpret return value
3269 as ffi_arg, not int.
3270 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3271 FFI_TYPE_POINTER case.
3272 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3273 values extended to word size.
3275 2002-10-02 Andreas Jaeger <aj@suse.de>
3277 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3279 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
3281 * include/ffi.h.in: Fix i386 win32 compilation.
3283 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3285 * configure.in: Add s390x-*-linux-* target.
3286 * configure: Regenerate.
3287 * include/ffi.h.in: Define S390X for s390x targets.
3288 (FFI_CLOSURES): Define for s390/s390x.
3289 (FFI_TRAMPOLINE_SIZE): Likewise.
3290 (FFI_NATIVE_RAW_API): Likewise.
3291 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3292 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
3293 * src/s390/ffi.c: Major rework of existing code. Add support for
3294 s390x targets. Add closure support.
3295 * src/s390/sysv.S: Likewise.
3297 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
3299 * src/arm/sysv.S: Fix typo.
3301 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
3303 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3304 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3305 (ffi_call_SYSV): Handle soft-float.
3307 2002-09-27 Bo Thorsen <bo@suse.de>
3309 * include/ffi.h.in: Fix multilib x86-64 support.
3311 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3313 * Makefile.am (all-multi): Fix multilib parallel build.
3315 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3317 * configure.in (sh[34]*-*-linux*): Add brackets.
3318 * configure: Regenerate.
3320 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3322 * Makefile.am: Add SH support.
3323 * Makefile.in: Regenerate.
3324 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3325 * configure: Regenerate.
3326 * include/ffi.h.in: Add SH support.
3327 * src/sh/ffi.c: New file.
3328 * src/sh/sysv.S: New file.
3329 * src/types.c: Add SH support.
3331 2002-07-16 Bo Thorsen <bo@suse.de>
3333 * src/x86/ffi64.c: New file that adds x86-64 support.
3334 * src/x86/unix64.S: New file that handles argument setup for
3336 * src/x86/sysv.S: Don't use this on x86-64.
3337 * src/x86/ffi.c: Don't use this on x86-64.
3339 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3341 * src/ffitest.c (struct6): New test that tests a special case in
3343 (struct7): Likewise.
3344 (struct8): Likewise.
3345 (struct9): Likewise.
3346 (closure_test_fn): Silence warning about this when it's not used.
3347 (main): Add the new tests.
3348 (main): Fix a couple of wrong casts and silence some compiler warnings.
3349 * include/ffi.h.in: Add x86-64 ABI definition.
3350 * fficonfig.h.in: Regenerate.
3351 * Makefile.am: Add x86-64 support.
3352 * configure.in: Likewise.
3353 * Makefile.in: Regenerate.
3354 * configure: Likewise.
3356 2002-06-24 Bo Thorsen <bo@suse.de>
3358 * src/types.c: Merge settings for similar architectures.
3359 Add x86-64 sizes and alignments.
3361 2002-06-23 Bo Thorsen <bo@suse.de>
3363 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3364 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3365 * src/mips/ffi.c (ffi_prep_args): Likewise.
3366 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3368 2002-07-18 H.J. Lu (hjl@gnu.org)
3370 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3371 (libffi_la_SOURCES): Support MIPS_LINUX.
3372 (libffi_convenience_la_SOURCES): Likewise.
3373 * Makefile.in: Regenerated.
3375 * configure.in (mips64*-*): Skip.
3376 (mips*-*-linux*): New.
3377 * configure: Regenerated.
3379 * src/mips/ffi.c: Include <sgidefs.h>.
3381 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3383 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3385 2002-05-27 Roger Sayle <roger@eyesopen.com>
3387 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3389 2002-05-27 Bo Thorsen <bo@suse.de>
3391 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3394 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3396 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3397 beginning of function (for older apple cc).
3399 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3401 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3402 script entry, and set LD to it when configuring multilibs.
3403 * configure: Rebuilt.
3405 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3407 * configure.in (sparc64-*-netbsd*): Add target.
3408 (sparc-*-netbsdelf*): Likewise.
3409 * configure: Regenerate.
3411 2002-04-28 David S. Miller <davem@redhat.com>
3413 * configure.in, configure: Fix SPARC test in previous change.
3415 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3417 * Makefile.am: Add Linux for S/390 support.
3418 * Makefile.in: Regenerate.
3419 * configure.in: Add Linux for S/390 support.
3420 * configure: Regenerate.
3421 * include/ffi.h.in: Add Linux for S/390 support.
3422 * src/s390/ffi.c: New file from libffi CVS tree.
3423 * src/s390/sysv.S: New file from libffi CVS tree.
3425 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3427 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3429 * src/sparc/v8.S: Use it.
3430 * src/sparc/v9.S: Likewise.
3431 * fficonfig.h.in: Rebuilt.
3432 * configure: Rebuilt.
3434 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3436 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3438 * src/ia64/unix.S: Add unwind information. Fix comments.
3439 Save sp in a way that's compatible with unwind info.
3440 (ffi_call_unix): Correctly restore sp in all cases.
3441 * src/ia64/ffi.c: Add, fix comments.
3443 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3445 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3447 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3449 * configure.in (alpha*-*-netbsd*): Add target.
3450 * configure: Regenerate.
3452 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3454 * src/sparc/v8.S: Add unwind info.
3455 * src/sparc/v9.S: Likewise.
3457 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3459 * configure.in: Enable i*86-*-netbsdelf*.
3460 * configure: Rebuilt.
3462 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3465 * src/mips/n32.s: Delete
3466 * src/mips/o32.s: Delete
3468 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3470 * configure.in: Enable alpha*-*-freebsd*.
3471 * configure: Rebuilt.
3473 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3475 * Makefile.am: libfficonvenience -> libffi_convenience.
3476 * Makefile.in: Rebuilt.
3478 * Makefile.am: Define ffitest_OBJECTS.
3479 * Makefile.in: Rebuilt.
3481 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3482 David Edelsohn <edelsohn@gnu.org>
3484 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3485 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3486 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3487 * Makefile.in: Regenerate.
3488 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3489 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3490 (flush_icache, flush_range): New functions.
3491 (ffi_closure_helper_DARWIN): New function.
3492 * src/powerpc/aix_closure.S: New file.
3493 * src/powerpc/darwin_closure.S: New file.
3495 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3497 * include/ffi.h.in: Add typedef for ffi_arg.
3498 * src/ffitest.c (main): Declare rint with ffi_arg.
3500 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3502 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3503 number of GPRs for floating-point arguments.
3505 2002-01-31 Anthony Green <green@redhat.com>
3507 * configure: Rebuilt.
3508 * configure.in: Replace CHECK_SIZEOF and endian tests with
3509 cross-compiler friendly macros.
3510 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3513 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3515 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3516 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3518 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3520 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3521 (TARGET_SRC_POWERPC_AIX): New.
3522 (POWERPC_AIX): New stanza.
3523 * Makefile.in: Regenerate.
3524 * configure.in: Add AIX case.
3525 * configure: Regenerate.
3526 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3527 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3528 size. Fix "long double" support.
3529 (ffi_call): Add FFI_AIX case.
3530 * src/powerpc/aix.S: New.
3532 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3534 Implement Darwin PowerPC ABI.
3535 * configure.in: Handle powerpc-*-darwin*.
3536 * Makefile.am: Set source files for POWERPC_DARWIN.
3537 * configure: Rebuilt.
3538 * Makefile.in: Rebuilt.
3539 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3541 * src/powerpc/darwin.S: New file.
3542 * src/powerpc/ffi_darwin.c: New file.
3544 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3546 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3548 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3550 * src/x86/sysv.S: Avoid gas-only .balign directive.
3551 Use C style comments.
3553 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3555 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3556 Fixes PR bootstrap/3563.
3558 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3560 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3562 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3564 * configure.in: Recognize sparc*-sun-* host.
3565 * configure: Regenerate.
3567 2001-06-06 Andrew Haley <aph@redhat.com>
3569 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
3571 2001-06-03 Andrew Haley <aph@redhat.com>
3573 * src/alpha/osf.S: Add unwind info.
3574 * src/powerpc/sysv.S: Add unwind info.
3575 * src/powerpc/ppc_closure.S: Likewise.
3577 2000-05-31 Jeff Sturm <jsturm@one-point.com>
3579 * configure.in: Fix AC_ARG_ENABLE usage.
3580 * configure: Rebuilt.
3582 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3584 * configure.in: Remove warning about beta code.
3585 * configure: Rebuilt.
3587 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3589 * src/ia64/unix.S: Restore stack pointer when returning from
3591 * src/ia64/ffi.c: Fix typo in comment.
3593 2001-04-18 Jim Wilson <wilson@redhat.com>
3595 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3596 to eliminate RAW DV.
3598 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3600 * Makefile.am: Make a libtool convenience library.
3601 * Makefile.in: Rebuilt.
3603 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3605 * configure.in: Use different syntax for subdirectory creation.
3606 * configure: Rebuilt.
3608 2001-03-27 Jon Beniston <jon@beniston.com>
3610 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3611 * configure: Rebuilt.
3612 * Makefile.am: Added X86_WIN32 target support.
3613 * Makefile.in: Rebuilt.
3615 * include/ffi.h.in: Added X86_WIN32 target support.
3617 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3618 * src/types.c: Added X86_WIN32 target support.
3620 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3621 stuff removed and made to work with CygWin's gas.
3623 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3625 * configure.in: Make target subdirectory in build dir.
3626 * Makefile.am: Override suffix based rules to specify correct output
3628 * Makefile.in: Rebuilt.
3629 * configure: Rebuilt.
3631 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3633 * src/powerpc/ppc_closure.S: New file.
3634 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3635 involving long long and register pairs.
3636 (ffi_prep_closure): New function.
3637 (flush_icache): Likewise.
3638 (ffi_closure_helper_SYSV): Likewise.
3639 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3640 (FFI_TRAMPOLINE_SIZE): Likewise.
3641 (FFI_NATIVE_RAW_API): Likewise.
3642 * Makefile.in: Rebuilt.
3643 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3644 (TARGET_SRC_POWERPC): Likewise.
3646 2001-03-19 Tom Tromey <tromey@redhat.com>
3648 * Makefile.in: Rebuilt.
3649 * Makefile.am (ffitest_LDFLAGS): New macro.
3651 2001-03-02 Nick Clifton <nickc@redhat.com>
3653 * include/ffi.h.in: Remove RCS ident string.
3654 * include/ffi_mips.h: Remove RCS ident string.
3655 * src/debug.c: Remove RCS ident string.
3656 * src/ffitest.c: Remove RCS ident string.
3657 * src/prep_cif.c: Remove RCS ident string.
3658 * src/types.c: Remove RCS ident string.
3659 * src/alpha/ffi.c: Remove RCS ident string.
3660 * src/alpha/osf.S: Remove RCS ident string.
3661 * src/arm/ffi.c: Remove RCS ident string.
3662 * src/arm/sysv.S: Remove RCS ident string.
3663 * src/mips/ffi.c: Remove RCS ident string.
3664 * src/mips/n32.S: Remove RCS ident string.
3665 * src/mips/o32.S: Remove RCS ident string.
3666 * src/sparc/ffi.c: Remove RCS ident string.
3667 * src/sparc/v8.S: Remove RCS ident string.
3668 * src/sparc/v9.S: Remove RCS ident string.
3669 * src/x86/ffi.c: Remove RCS ident string.
3670 * src/x86/sysv.S: Remove RCS ident string.
3672 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3674 * include/ffi.h.in: Change sourceware.cygnus.com references to
3677 2000-12-09 Richard Henderson <rth@redhat.com>
3679 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3680 (ffi_closure_osf_inner): Index rather than increment avalue
3681 and arg_types. Give ffi_closure_osf the raw return value type.
3682 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3685 2000-12-07 Richard Henderson <rth@redhat.com>
3687 * src/raw_api.c (ffi_translate_args): Fix typo.
3688 (ffi_prep_closure): Likewise.
3690 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3691 FFI_TRAMPOLINE_SIZE.
3692 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3693 cif->bytes for new ffi_call_osf implementation.
3694 (ffi_prep_args): Absorb into ...
3695 (ffi_call): ... here. Do all stack allocation here and
3696 avoid a callback function.
3697 (ffi_prep_closure, ffi_closure_osf_inner): New.
3698 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3699 (ffi_closure_osf): New.
3701 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3703 * config.guess, config.sub, install-sh: Removed.
3704 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3705 * Makefile.in: Rebuilt.
3707 * acinclude.m4: Include libtool macros from the top level.
3708 * aclocal.m4, configure: Rebuilt.
3710 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3712 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3713 * configure: Rebuilt.
3715 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3717 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3718 memory correctly. Use conditional instructions, not branches where
3721 2000-05-04 Tom Tromey <tromey@cygnus.com>
3723 * configure: Rebuilt.
3724 * configure.in: Match `arm*-*-linux-*'.
3725 From Chris Dornan <cdornan@arm.com>.
3727 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3729 * Makefile.am (SUBDIRS): Define.
3730 (AM_MAKEFLAGS): Likewise.
3731 (Multilib support.): Add section.
3732 * Makefile.in: Rebuilt.
3733 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3734 New variables. Set for gcc using -print-multi-lib. Export them
3736 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3737 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3739 (extra_compiler_flags_value, extra_compiler_flags): Check these
3740 for extra compiler options which need to be passed down in
3743 2000-04-16 Anthony Green <green@redhat.com>
3745 * configure: Rebuilt.
3746 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3748 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3750 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3751 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3752 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3753 Replace all void * sizeofs with sizeof(int).
3754 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3755 different than DOUBLE.
3756 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3757 (ffi_prep_args_v9): New function.
3758 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3759 (ffi_V9_return_struct): New function.
3760 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3761 32bit code (not yet cross-arch calls).
3762 * src/sparc/v8.S: Add struct return delay nop.
3764 * src/sparc/v9.S: New file.
3765 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3766 is used on sparc64 only for structures larger than 32 bytes.
3767 Pass by reference for structures is done for structure arguments
3768 larger than 16 bytes.
3769 * src/ffitest.c (main): Use 64bit rint on sparc64.
3770 Run long long tests on sparc.
3771 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3773 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3775 * configure.in (sparc-*-linux*): New supported target.
3776 (sparc64-*-linux*): Likewise.
3777 * configure: Rebuilt.
3778 * Makefile.am: Add v9.S to SPARC files.
3779 * Makefile.in: Likewise.
3780 (LINK): Surround $(CCLD) into double quotes, so that multilib
3781 compiles work correctly.
3783 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3785 * configure: Rebuilt.
3786 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3787 proposed by Bryce McKinlay.
3789 2000-03-20 Tom Tromey <tromey@cygnus.com>
3791 * Makefile.in: Hand edit for java_raw_api.lo.
3793 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3795 * config.guess, config.sub: Update from the gcc tree.
3796 Fix for PR libgcj/168.
3798 2000-03-03 Tom Tromey <tromey@cygnus.com>
3800 * Makefile.in: Fixed ia64 by hand.
3802 * configure: Rebuilt.
3803 * configure.in (--enable-multilib): New option.
3804 (libffi_basedir): New subst.
3805 (AC_OUTPUT): Added multilib code.
3807 2000-03-02 Tom Tromey <tromey@cygnus.com>
3809 * Makefile.in: Rebuilt.
3810 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3813 2000-02-25 Hans Boehm <boehm@acm.org>
3815 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3817 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3819 (ffi_prep_raw_closure): Use ffi_translate_args, not
3820 ffi_closure_translate.
3821 * src/java_raw_api.c: New file.
3822 * src/ffitest.c (closure_test_fn): New function.
3823 (main): Define `rint' as long long on IA64. Added new test when
3824 FFI_CLOSURES is defined.
3825 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3826 (ffi_abi): Recognize IA64.
3827 (ffi_raw): Added `flt' field.
3828 Added "Java raw API" code.
3829 * configure.in: Recognize ia64.
3830 * Makefile.am (TARGET_SRC_IA64): New macro.
3831 (libffi_la_common_SOURCES): Added java_raw_api.c.
3832 (libffi_la_SOURCES): Define in IA64 case.
3834 2000-01-04 Tom Tromey <tromey@cygnus.com>
3836 * Makefile.in: Rebuilt with newer automake.
3838 1999-12-31 Tom Tromey <tromey@cygnus.com>
3840 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3842 1999-09-01 Tom Tromey <tromey@cygnus.com>
3844 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3846 * fficonfig.h.in: Rebuilt.
3847 * configure: Rebuilt.
3848 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3849 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3850 * acconfig.h: Don't #undef PACKAGE or VERSION.
3852 1999-08-09 Anthony Green <green@cygnus.com>
3854 * include/ffi.h.in: Try to work around messy header problem
3855 with PACKAGE and VERSION.
3857 * configure: Rebuilt.
3858 * configure.in: Change version to 2.00-beta.
3860 * fficonfig.h.in: Rebuilt.
3861 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3863 * src/x86/ffi.c (ffi_raw_call): Rename.
3865 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3867 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3868 (ffi_prep_incoming_args_SYSV): Ditto.
3869 (ffi_prep_closure): Ditto.
3870 (ffi_closure_raw_SYSV): Ditto.
3871 (ffi_prep_raw_closure): More ditto.
3872 (ffi_call_raw): Final ditto.
3874 * include/ffi.h.in: Add definitions for closure and raw API.
3876 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3879 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3881 * src/raw_api.c: New file.
3883 * include/ffi.h.in (ffi_raw): New type.
3884 (UINT_ARG, SINT_ARG): New defines.
3885 (ffi_closure, ffi_raw_closure): New types.
3886 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3888 * configure.in: Add check for endianness and sizeof void*.
3890 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3891 instead of directly.
3893 * configure: Rebuilt.
3895 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3897 * configure.in: Add x86 and powerpc BeOS configurations.
3898 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3900 1999-05-09 Anthony Green <green@cygnus.com>
3902 * configure.in: Add warning about this being beta code.
3903 Remove src/Makefile.am from the picture.
3904 * configure: Rebuilt.
3906 * Makefile.am: Move logic from src/Makefile.am. Add changes
3907 to support libffi as a target library.
3908 * Makefile.in: Rebuilt.
3910 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3911 Upgraded to new autoconf, automake, libtool.
3915 * LICENSE: Update copyright date.
3917 * src/Makefile.am, src/Makefile.in: Removed.
3919 1998-11-29 Anthony Green <green@cygnus.com>
3921 * include/ChangeLog: Removed.
3922 * src/ChangeLog: Removed.
3923 * src/mips/ChangeLog: Removed.
3924 * src/sparc/ChangeLog: Remboved.
3925 * src/x86/ChangeLog: Removed.
3927 * ChangeLog.v1: Created.