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