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