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