1 2009-12-04 David Edelsohn <edelsohn@gnu.org>
3 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
6 2009-12-04 Uros Bizjak <ubizjak@gmail.com>
9 * src/x86/ffi64.c (classify_argument): Update from
10 gcc/config/i386/i386.c.
11 (ffi_closure_unix64_inner): Do not use the address of two consecutive
12 SSE registers directly.
13 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
14 for x86_64 linux targets.
16 2009-12-04 David Edelsohn <edelsohn@gnu.org>
18 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
19 pfr for long double split between fpr13 and stack.
21 2009-12-03 David Edelsohn <edelsohn@gnu.org>
23 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
24 fparg_count twice for long double.
26 2009-12-03 David Edelsohn <edelsohn@gnu.org>
29 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
31 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
33 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
34 Remove xfails for x86 linux targets.
36 2009-12-02 David Edelsohn <edelsohn@gnu.org>
38 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
41 2009-12-01 David Edelsohn <edelsohn@gnu.org>
43 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
44 register usage. Call ffi_prep_args directly. Add long double
46 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
47 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
48 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
49 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
51 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
53 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
54 Allocate result area between params and FPRs.
56 2009-11-30 David Edelsohn <edelsohn@gnu.org>
59 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
61 * src/powerpc/aix.S: Implement AIX64 version.
62 * src/powerpc/aix_closure.S: Implement AIX64 version.
63 (ffi_closure_ASM): Use extsb, lha and displament addresses.
64 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
66 (ffi_prep_cif_machdep): Same.
68 (ffi_closure_helper_DARWIN): Same.
70 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
73 * testsuite/libffi.call/testclosure.c: New test.
75 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
77 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
78 assembly version use of ___chkstk.
80 2009-09-23 Matthias Klose <doko@ubuntu.com>
82 PR libffi/40242, PR libffi/41443
83 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
84 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
85 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
86 Change the conditionals to __SOFTFP__ || __ARM_EABI__
87 for -mfloat-abi=softfp to work.
89 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
91 PR testsuite/32843 (strikes again)
92 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
93 enable proper extension on char and short.
95 2009-09-15 David Daney <ddaney@caviumnetworks.com>
97 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
98 handling for FFI_TYPE_POINTER.
99 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
100 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
101 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
102 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
103 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
104 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
105 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
106 structure and pointer returns.
107 (ffi_closure_N32): Add handling for pointer returns.
108 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
109 calc_n32_return_struct_flags): Handle soft-float.
110 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
111 (ffi_call_N32): Declare proper argument types.
112 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
115 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
117 * configure.ac (AC_PREREQ): Bump to 2.64.
119 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
121 * Makefile.am (install-html, install-pdf): Remove.
122 * Makefile.in: Regenerate.
124 * Makefile.in: Regenerate.
125 * aclocal.m4: Regenerate.
126 * configure: Regenerate.
127 * fficonfig.h.in: Regenerate.
128 * include/Makefile.in: Regenerate.
129 * man/Makefile.in: Regenerate.
130 * testsuite/Makefile.in: Regenerate.
132 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
134 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
136 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
139 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
140 return types for X86_WIN32.
141 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
142 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
143 _ffi_closure_STDCALL): Likewise.
145 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
146 (dlmmap, dlmunmap): Also use these functions on Cygwin.
148 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
153 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
156 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
158 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
159 to "" before adding paths. (This reinstates an assignment that
160 was removed by my 2009-06-30 commit, but changes the initial
161 value from "." to "".)
163 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
166 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
167 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
170 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
172 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
173 to ld_library_path. Use add_path. Add just find_libgcc_s
174 to ld_library_path, not every libgcc multilib directory.
176 2009-06-16 Wim Lewis <wiml@hhhh.org>
178 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
179 supposed to be callee-saved.
180 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
181 return buffer for odd-size structs.
183 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
186 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
187 allow_stack_execute for Darwin.
189 2009-06-16 Andrew Haley <aph@redhat.com>
191 * configure.ac (TARGETDIR): Add missing blank lines.
192 * configure: Regenerate.
194 2009-06-16 Andrew Haley <aph@redhat.com>
196 * testsuite/libffi.call/cls_align_sint64.c,
197 testsuite/libffi.call/cls_align_uint64.c,
198 testsuite/libffi.call/cls_longdouble_va.c,
199 testsuite/libffi.call/cls_ulonglong.c,
200 testsuite/libffi.call/return_ll1.c,
201 testsuite/libffi.call/stret_medium2.c: Fix printf format
203 * testsuite/libffi.call/ffitest.h,
204 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
206 2009-06-15 Andrew Haley <aph@redhat.com>
208 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
209 * testsuite/libffi.call/err_bad_abi.c: Likewise.
211 2009-06-12 Andrew Haley <aph@redhat.com>
213 * Makefile.am: Remove info_TEXINFOS.
215 2009-06-12 Andrew Haley <aph@redhat.com>
217 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
218 testsuite/libffi.call/cls_align_uint64.c,
219 testsuite/libffi.call/cls_ulonglong.c,
220 testsuite/libffi.call/return_ll1.c,
221 testsuite/libffi.call/stret_medium2.c: Fix printf format
223 testsuite/libffi.special/unwindtest.cc: include stdint.h.
225 2009-06-11 Timothy Wall <twall@users.sf.net>
230 include/ffi_common.h,
235 src/x86/win64.S (new),
236 README: Added win64 support (mingw or MSVC)
240 testsuite/Makefile.in,
242 aclocal.m4: Regenerated
243 * ltcf-c.sh: properly escape cygwin/w32 path
244 * man/ffi_call.3: Clarify size requirements for return value.
245 * src/x86/ffi64.c: Fix filename in comment.
246 * src/x86/win32.S: Remove unused extern.
248 * testsuite/libffi.call/closure_fn0.c,
249 testsuite/libffi.call/closure_fn1.c,
250 testsuite/libffi.call/closure_fn2.c,
251 testsuite/libffi.call/closure_fn3.c,
252 testsuite/libffi.call/closure_fn4.c,
253 testsuite/libffi.call/closure_fn5.c,
254 testsuite/libffi.call/closure_fn6.c,
255 testsuite/libffi.call/closure_stdcall.c,
256 testsuite/libffi.call/cls_12byte.c,
257 testsuite/libffi.call/cls_16byte.c,
258 testsuite/libffi.call/cls_18byte.c,
259 testsuite/libffi.call/cls_19byte.c,
260 testsuite/libffi.call/cls_1_1byte.c,
261 testsuite/libffi.call/cls_20byte.c,
262 testsuite/libffi.call/cls_20byte1.c,
263 testsuite/libffi.call/cls_24byte.c,
264 testsuite/libffi.call/cls_2byte.c,
265 testsuite/libffi.call/cls_3_1byte.c,
266 testsuite/libffi.call/cls_3byte1.c,
267 testsuite/libffi.call/cls_3byte2.c,
268 testsuite/libffi.call/cls_4_1byte.c,
269 testsuite/libffi.call/cls_4byte.c,
270 testsuite/libffi.call/cls_5_1_byte.c,
271 testsuite/libffi.call/cls_5byte.c,
272 testsuite/libffi.call/cls_64byte.c,
273 testsuite/libffi.call/cls_6_1_byte.c,
274 testsuite/libffi.call/cls_6byte.c,
275 testsuite/libffi.call/cls_7_1_byte.c,
276 testsuite/libffi.call/cls_7byte.c,
277 testsuite/libffi.call/cls_8byte.c,
278 testsuite/libffi.call/cls_9byte1.c,
279 testsuite/libffi.call/cls_9byte2.c,
280 testsuite/libffi.call/cls_align_double.c,
281 testsuite/libffi.call/cls_align_float.c,
282 testsuite/libffi.call/cls_align_longdouble.c,
283 testsuite/libffi.call/cls_align_longdouble_split.c,
284 testsuite/libffi.call/cls_align_longdouble_split2.c,
285 testsuite/libffi.call/cls_align_pointer.c,
286 testsuite/libffi.call/cls_align_sint16.c,
287 testsuite/libffi.call/cls_align_sint32.c,
288 testsuite/libffi.call/cls_align_sint64.c,
289 testsuite/libffi.call/cls_align_uint16.c,
290 testsuite/libffi.call/cls_align_uint32.c,
291 testsuite/libffi.call/cls_align_uint64.c,
292 testsuite/libffi.call/cls_dbls_struct.c,
293 testsuite/libffi.call/cls_double.c,
294 testsuite/libffi.call/cls_double_va.c,
295 testsuite/libffi.call/cls_float.c,
296 testsuite/libffi.call/cls_longdouble.c,
297 testsuite/libffi.call/cls_longdouble_va.c,
298 testsuite/libffi.call/cls_multi_schar.c,
299 testsuite/libffi.call/cls_multi_sshort.c,
300 testsuite/libffi.call/cls_multi_sshortchar.c,
301 testsuite/libffi.call/cls_multi_uchar.c,
302 testsuite/libffi.call/cls_multi_ushort.c,
303 testsuite/libffi.call/cls_multi_ushortchar.c,
304 testsuite/libffi.call/cls_pointer.c,
305 testsuite/libffi.call/cls_pointer_stack.c,
306 testsuite/libffi.call/cls_schar.c,
307 testsuite/libffi.call/cls_sint.c,
308 testsuite/libffi.call/cls_sshort.c,
309 testsuite/libffi.call/cls_uchar.c,
310 testsuite/libffi.call/cls_uint.c,
311 testsuite/libffi.call/cls_ulonglong.c,
312 testsuite/libffi.call/cls_ushort.c,
313 testsuite/libffi.call/err_bad_abi.c,
314 testsuite/libffi.call/err_bad_typedef.c,
315 testsuite/libffi.call/float2.c,
316 testsuite/libffi.call/huge_struct.c,
317 testsuite/libffi.call/nested_struct.c,
318 testsuite/libffi.call/nested_struct1.c,
319 testsuite/libffi.call/nested_struct10.c,
320 testsuite/libffi.call/nested_struct2.c,
321 testsuite/libffi.call/nested_struct3.c,
322 testsuite/libffi.call/nested_struct4.c,
323 testsuite/libffi.call/nested_struct5.c,
324 testsuite/libffi.call/nested_struct6.c,
325 testsuite/libffi.call/nested_struct7.c,
326 testsuite/libffi.call/nested_struct8.c,
327 testsuite/libffi.call/nested_struct9.c,
328 testsuite/libffi.call/problem1.c,
329 testsuite/libffi.call/return_ldl.c,
330 testsuite/libffi.call/return_ll1.c,
331 testsuite/libffi.call/stret_large.c,
332 testsuite/libffi.call/stret_large2.c,
333 testsuite/libffi.call/stret_medium.c,
334 testsuite/libffi.call/stret_medium2.c,
335 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
336 of checking for MMAP. Use intptr_t instead of long casts.
338 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
340 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
341 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
342 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
344 2009-06-09 Andrew Haley <aph@redhat.com>
346 * src/x86/freebsd.S: Add missing file.
348 2009-06-08 Andrew Haley <aph@redhat.com>
350 Import from libffi 3.0.8:
352 * doc/libffi.texi: New file.
353 * doc/libffi.info: Likewise.
354 * doc/stamp-vti: Likewise.
355 * man/Makefile.am: New file.
356 * man/ffi_call.3: New file.
358 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
360 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
362 * configure.ac: Bump version to 3.0.8.
363 parisc*-*-linux*: Add.
364 i386-*-freebsd* | i386-*-openbsd*: Add.
365 powerpc-*-beos*: Add.
366 AM_CONDITIONAL X86_FREEBSD: Add.
367 AC_CONFIG_FILES: Add man/Makefile.
369 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
371 2009-06-08 Andrew Haley <aph@redhat.com>
373 * README: Import from libffi 3.0.8.
375 2009-06-08 Andrew Haley <aph@redhat.com>
377 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
378 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
379 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
380 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
382 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
383 * testsuite/libffi.call/stret_medium.c: Likewise.
384 * testsuite/libffi.call/stret_large2.c: Likewise.
385 * testsuite/libffi.call/stret_large.c: Likewise.
387 2008-12-26 Timothy Wall <twall@users.sf.net>
389 * testsuite/libffi.call/cls_longdouble.c,
390 testsuite/libffi.call/cls_longdouble_va.c,
391 testsuite/libffi.call/cls_align_longdouble.c,
392 testsuite/libffi.call/cls_align_longdouble_split.c,
393 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
394 failures on x86_64 cygwin/mingw.
396 2008-12-22 Timothy Wall <twall@users.sf.net>
398 * testsuite/libffi.call/closure_fn0.c,
399 testsuite/libffi.call/closure_fn1.c,
400 testsuite/libffi.call/closure_fn2.c,
401 testsuite/libffi.call/closure_fn3.c,
402 testsuite/libffi.call/closure_fn4.c,
403 testsuite/libffi.call/closure_fn5.c,
404 testsuite/libffi.call/closure_fn6.c,
405 testsuite/libffi.call/closure_loc_fn0.c,
406 testsuite/libffi.call/closure_stdcall.c,
407 testsuite/libffi.call/cls_align_pointer.c,
408 testsuite/libffi.call/cls_pointer.c,
409 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
410 pointer to integer (intptr_t).
411 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
413 2008-07-24 Anthony Green <green@redhat.com>
415 * testsuite/libffi.call/cls_dbls_struct.c,
416 testsuite/libffi.call/cls_double_va.c,
417 testsuite/libffi.call/cls_longdouble.c,
418 testsuite/libffi.call/cls_longdouble_va.c,
419 testsuite/libffi.call/cls_pointer.c,
420 testsuite/libffi.call/cls_pointer_stack.c,
421 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
424 2008-03-04 Anthony Green <green@redhat.com>
428 * testsuite/libffi.call/cls_align_longdouble_split2.c
429 testsuite/libffi.call/cls_align_longdouble_split.c
430 testsuite/libffi.call/cls_dbls_struct.c
431 testsuite/libffi.call/cls_double_va.c
432 testsuite/libffi.call/cls_longdouble.c
433 testsuite/libffi.call/cls_longdouble_va.c
434 testsuite/libffi.call/cls_pointer.c
435 testsuite/libffi.call/cls_pointer_stack.c
436 testsuite/libffi.call/err_bad_abi.c
437 testsuite/libffi.call/err_bad_typedef.c
438 testsuite/libffi.call/stret_large2.c
439 testsuite/libffi.call/stret_large.c
440 testsuite/libffi.call/stret_medium2.c
441 testsuite/libffi.call/stret_medium.c: New tests from Apple.
443 2009-06-05 Andrew Haley <aph@redhat.com>
445 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
448 2009-06-04 Andrew Haley <aph@redhat.com>
450 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
453 2008-02-26 Anthony Green <green@redhat.com>
454 Thomas Heller <theller@ctypes.org>
456 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
459 2008-02-03 Timothy Wall <twall@users.sf.net>
461 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
462 offset based on code pointer, not data pointer.
464 2008-01-31 Timothy Wall <twall@users.sf.net>
466 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
468 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
470 * src/x86/win32.S: Add assembly for stdcall closure.
471 * src/x86/ffi.c: Initialize stdcall closure trampoline.
473 2009-06-04 Andrew Haley <aph@redhat.com>
475 * include/ffi.h.in: Change void (*)() to void (*)(void).
476 * src/x86/ffi.c: Likewise.
478 2009-06-04 Andrew Haley <aph@redhat.com>
480 * src/powerpc/ppc_closure.S: Insert licence header.
481 * src/powerpc/linux64_closure.S: Likewise.
482 * src/m68k/sysv.S: Likewise.
484 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
485 * src/powerpc/ffi.c: Likewise.
486 * src/powerpc/ffi_darwin.c: Likewise.
487 * src/m32r/ffi.c: Likewise.
488 * src/sh64/ffi.c: Likewise.
489 * src/x86/ffi64.c: Likewise.
490 * src/alpha/ffi.c: Likewise.
491 * src/alpha/osf.S: Likewise.
492 * src/frv/ffi.c: Likewise.
493 * src/s390/ffi.c: Likewise.
494 * src/pa/ffi.c: Likewise.
495 * src/pa/hpux32.S: Likewise.
496 * src/ia64/unix.S: Likewise.
497 * src/ia64/ffi.c: Likewise.
498 * src/sparc/ffi.c: Likewise.
499 * src/mips/ffi.c: Likewise.
500 * src/sh/ffi.c: Likewise.
502 2008-02-15 David Daney <ddaney@avtrex.com>
504 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
505 Define (conditionally), and use it to include cachectl.h.
506 (ffi_prep_closure_loc): Fix cache flushing.
507 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
509 2009-06-04 Andrew Haley <aph@redhat.com>
515 src/powerpc/ffitarget.h,
517 src/sh64/ffitarget.h,
528 src/alpha/ffitarget.h,
531 src/m68k/ffitarget.h,
534 src/s390/ffitarget.h,
536 src/cris/ffitarget.h,
541 src/ia64/ffitarget.h,
544 src/ia64/ia64_flags.h,
548 src/sparc/ffitarget.h,
551 src/mips/ffitarget.h,
558 src/sh/sysv.S: Update license text.
560 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
562 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
563 (.eh_frame): Add FDE for it.
565 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
567 * configure.ac: Also check if assembler supports pc-relative
568 relocs on X86_WIN32 targets.
569 * configure: Regenerate.
570 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
571 (_ffi_call_SYSV): Add missing function type symbol .def and
572 add EH markup labels.
573 (_ffi_call_STDCALL): Likewise.
574 (_ffi_closure_SYSV): Likewise.
575 (_ffi_closure_raw_SYSV): Likewise.
576 (.eh_frame): Add hand-crafted EH data.
578 2009-04-09 Jakub Jelinek <jakub@redhat.com>
580 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
581 version 3 of the GNU General Public License and to point readers
582 at the COPYING3 file and the FSF's license web page.
583 * testsuite/libffi.call/call.exp: Likewise.
584 * testsuite/libffi.special/special.exp: Likewise.
586 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
588 * configure: Regenerate.
590 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
593 * configure.ac (HAVE_AS_X86_PCREL): New test.
594 * configure: Regenerate.
595 * fficonfig.h.in: Regenerate.
596 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
597 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
598 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
599 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
600 * src/x86/unix64.S (.Lstore_table): Move to .text section.
601 (.Lload_table): Likewise.
602 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
604 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
606 * configure: Regenerate.
608 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
610 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
611 signed/unsigned int8/16 return values.
612 * src/sparc/v8.S (ffi_call_v8): Likewise.
613 (ffi_closure_v8): Likewise.
615 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
616 Steve Ellcey <sje@cup.hp.com>
618 * configure: Regenerate for new libtool.
619 * Makefile.in: Ditto.
620 * include/Makefile.in: Ditto.
623 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
625 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
626 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
627 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
628 Adjust copyright notice.
629 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
630 register or two register to use for FFI_SYSV structs.
631 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
632 (ffi_closure_helper_SYSV): Fix the return type for
633 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
634 Adjust copyright notice.
636 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
638 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
641 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
643 * configure: Regenerate.
644 * include/Makefile.in: Regenerate.
645 * testsuite/Makefile.in: Regenerate.
647 2008-06-07 Joseph Myers <joseph@codesourcery.com>
649 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
650 powerpc-*-beos*): Remove.
651 * configure: Regenerate.
653 2008-05-09 Julian Brown <julian@codesourcery.com>
655 * Makefile.am (LTLDFLAGS): New.
656 (libffi_la_LDFLAGS): Use above.
657 * Makefile.in: Regenerate.
659 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
662 * aclocal.m4: Regenerate.
663 * configure: Regenerate.
665 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
667 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
668 * src/sh64/sysv.S: Likewise.
670 2008-03-26 Daniel Jacobowitz <dan@debian.org>
672 * src/arm/sysv.S: Fix ARM comment marker.
674 2008-03-26 Jakub Jelinek <jakub@redhat.com>
676 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
677 * src/s390/sysv.S: Likewise.
678 * src/powerpc/ppc_closure.S: Likewise.
679 * src/powerpc/sysv.S: Likewise.
680 * src/x86/unix64.S: Likewise.
681 * src/x86/sysv.S: Likewise.
682 * src/sparc/v8.S: Likewise.
683 * src/sparc/v9.S: Likewise.
684 * src/m68k/sysv.S: Likewise.
685 * src/arm/sysv.S: Likewise.
687 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
689 * aclocal.m4: Regenerate.
690 * configure: Likewise.
691 * Makefile.in: Likewise.
692 * include/Makefile.in: Likewise.
693 * testsuite/Makefile.in: Likewise.
695 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
696 Andreas Tobler <a.tobler@schweiz.org>
698 * configure.ac: Add amd64-*-freebsd* target.
699 * configure: Regenerate.
701 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
704 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
707 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
710 2008-01-24 David Edelsohn <edelsohn@gnu.org>
712 * configure: Regenerate.
714 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
716 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
718 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
721 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
722 signed/unsigned int8/16 for X86_DARWIN.
723 Updated copyright info.
724 Handle one and two byte structs with special cif->flags.
725 * src/x86/ffitarget.h: Add special types for one and two byte structs.
726 Updated copyright info.
727 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
729 Remove code to pop args from the stack after call.
730 Special-case signed/unsigned for int8/16, one and two byte structs.
731 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
732 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
734 Updated copyright info.
736 2007-12-08 David Daney <ddaney@avtrex.com>
738 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
739 SUBU, add with ADDU and use smaller code sequences.
741 2007-12-07 David Daney <ddaney@avtrex.com>
743 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
746 2007-12-06 David Daney <ddaney@avtrex.com>
748 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
750 (ffi_java_raw): New typedef.
751 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
752 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
754 (ffi_java_raw_closure) : Same.
755 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
757 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
759 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
760 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
761 sizeof(ffi_java_raw) for alignment calculations.
762 (ffi_java_ptrarray_to_raw): Same.
763 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
764 if FFI_SIZEOF_JAVA_RAW == 4.
765 (ffi_java_raw_to_rvalue): Same.
766 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
767 (ffi_java_translate_args): Same.
768 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
770 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
772 2007-12-06 David Daney <ddaney@avtrex.com>
774 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
777 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
780 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
781 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
782 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
783 set the NUM_FPR_ARG_REGISTERS according to.
784 Add support for potential soft-float support under hard-float
786 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
787 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
788 to the FFI_LINUX_SOFT_FLOAT ABI.
789 (ffi_prep_cif_machdep): Likewise.
790 (ffi_closure_helper_SYSV): Likewise.
791 * src/powerpc/ppc_closure.S: Make sure not to store float/double
792 on archs where __NO_FPRS__ is true.
793 Add FFI_TYPE_UINT128 support.
794 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
795 Adjust copyright notice.
797 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
799 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
800 * include/ffi_common.h: ... here.
803 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
805 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
806 * src/powerpc/linux64.S: Likewise.
807 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
808 * testsuite/libffi.call/return_ldl.c: New test.
810 2007-09-04 <aph@redhat.com>
812 * src/arm/sysv.S (UNWIND): New.
813 (Whole file): Conditionally compile unwinder directives.
814 * src/arm/sysv.S: Add unwinder directives.
816 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
817 Only treat r0 as a struct address if we're actually returning a
819 Only copy the bytes that are actually within a struct.
820 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
821 is returned in r0, not passed by address.
822 (ffi_call): Allocate a word-sized temporary for the case where
823 a composite is returned in r0.
824 (ffi_prep_incoming_args_SYSV): Align as necessary.
826 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
828 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
829 directly using the sys_cacheflush syscall.
831 2007-07-27 Andrew Haley <aph@redhat.com>
833 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
835 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
837 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
838 * configure.ac: Likewise.
839 * Makefile.in: Regenerate.
840 * include/Makefile.in: Likewise.
841 * testsuite/Makefile.in: Likewise.
842 * configure: Likewise.
844 2007-08-24 David Daney <ddaney@avtrex.com>
846 * testsuite/libffi.call/return_sl.c: New test.
848 2007-08-10 David Daney <ddaney@avtrex.com>
850 * testsuite/libffi.call/cls_multi_ushort.c,
851 testsuite/libffi.call/cls_align_uint16.c,
852 testsuite/libffi.call/nested_struct1.c,
853 testsuite/libffi.call/nested_struct3.c,
854 testsuite/libffi.call/cls_7_1_byte.c,
855 testsuite/libffi.call/nested_struct5.c,
856 testsuite/libffi.call/cls_double.c,
857 testsuite/libffi.call/nested_struct7.c,
858 testsuite/libffi.call/cls_sint.c,
859 testsuite/libffi.call/nested_struct9.c,
860 testsuite/libffi.call/cls_20byte1.c,
861 testsuite/libffi.call/cls_multi_sshortchar.c,
862 testsuite/libffi.call/cls_align_sint64.c,
863 testsuite/libffi.call/cls_3byte2.c,
864 testsuite/libffi.call/cls_multi_schar.c,
865 testsuite/libffi.call/cls_multi_uchar.c,
866 testsuite/libffi.call/cls_19byte.c,
867 testsuite/libffi.call/cls_9byte1.c,
868 testsuite/libffi.call/cls_align_float.c,
869 testsuite/libffi.call/closure_fn1.c,
870 testsuite/libffi.call/problem1.c,
871 testsuite/libffi.call/closure_fn3.c,
872 testsuite/libffi.call/cls_sshort.c,
873 testsuite/libffi.call/closure_fn5.c,
874 testsuite/libffi.call/cls_align_double.c,
875 testsuite/libffi.call/nested_struct.c,
876 testsuite/libffi.call/cls_2byte.c,
877 testsuite/libffi.call/nested_struct10.c,
878 testsuite/libffi.call/cls_4byte.c,
879 testsuite/libffi.call/cls_6byte.c,
880 testsuite/libffi.call/cls_8byte.c,
881 testsuite/libffi.call/cls_multi_sshort.c,
882 testsuite/libffi.call/cls_align_sint16.c,
883 testsuite/libffi.call/cls_align_uint32.c,
884 testsuite/libffi.call/cls_20byte.c,
885 testsuite/libffi.call/cls_float.c,
886 testsuite/libffi.call/nested_struct2.c,
887 testsuite/libffi.call/cls_5_1_byte.c,
888 testsuite/libffi.call/nested_struct4.c,
889 testsuite/libffi.call/cls_24byte.c,
890 testsuite/libffi.call/nested_struct6.c,
891 testsuite/libffi.call/cls_64byte.c,
892 testsuite/libffi.call/nested_struct8.c,
893 testsuite/libffi.call/cls_uint.c,
894 testsuite/libffi.call/cls_multi_ushortchar.c,
895 testsuite/libffi.call/cls_schar.c,
896 testsuite/libffi.call/cls_uchar.c,
897 testsuite/libffi.call/cls_align_uint64.c,
898 testsuite/libffi.call/cls_ulonglong.c,
899 testsuite/libffi.call/cls_align_longdouble.c,
900 testsuite/libffi.call/cls_1_1byte.c,
901 testsuite/libffi.call/cls_12byte.c,
902 testsuite/libffi.call/cls_3_1byte.c,
903 testsuite/libffi.call/cls_3byte1.c,
904 testsuite/libffi.call/cls_4_1byte.c,
905 testsuite/libffi.call/cls_6_1_byte.c,
906 testsuite/libffi.call/cls_16byte.c,
907 testsuite/libffi.call/cls_18byte.c,
908 testsuite/libffi.call/closure_fn0.c,
909 testsuite/libffi.call/cls_9byte2.c,
910 testsuite/libffi.call/closure_fn2.c,
911 testsuite/libffi.call/closure_fn4.c,
912 testsuite/libffi.call/cls_ushort.c,
913 testsuite/libffi.call/closure_fn6.c,
914 testsuite/libffi.call/cls_5byte.c,
915 testsuite/libffi.call/cls_align_pointer.c,
916 testsuite/libffi.call/cls_7byte.c,
917 testsuite/libffi.call/cls_align_sint32.c,
918 testsuite/libffi.special/unwindtest_ffi_call.cc,
919 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
921 2007-08-10 David Daney <ddaney@avtrex.com>
924 * configure.ac: Don't treat mips64 as a special case.
925 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
926 * configure: Regenerate
927 * Makefile.in: Ditto.
928 * fficonfig.h.in: Ditto.
929 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
930 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
931 (FFI_DEFAULT_ABI): Set for n64 case.
932 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
933 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
934 (ffi_closure_N32): New function.
935 (.eh_frame): New section
936 * src/mips/o32.S: Clean up comments.
937 (ffi_closure_O32): Pass ffi_closure parameter in $12.
938 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
939 _MIPS_SIM == _ABIN32 throughout.
940 (FFI_MIPS_STOP_HERE): New, use in place of
942 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
943 to support n32/n64 ABIs.
944 (calc_n32_struct_flags): Rewrite.
945 (calc_n32_return_struct_flags): Remove unused variable. Reverse
946 position of flag bits.
947 (ffi_prep_cif_machdep): Rewrite n32 portion.
948 (ffi_call): Enable for n64. Add special handling for small structure
950 (ffi_prep_closure_loc): Add n32 and n64 support.
951 (ffi_closure_mips_inner_O32): Add cast to silence warning.
952 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
954 2007-08-08 David Daney <ddaney@avtrex.com>
956 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
957 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
959 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
960 * testsuite/libffi.call/cls_sint.c (main): Ditto.
961 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
962 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
963 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
964 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
965 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
966 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
967 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
968 * testsuite/libffi.call/return_fl3.c (main): Ditto.
969 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
970 * testsuite/libffi.call/nested_struct.c (main): Ditto.
971 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
972 * testsuite/libffi.call/return_ll1.c (main): Ditto.
973 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
974 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
975 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
976 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
977 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
978 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
979 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
980 * testsuite/libffi.call/cls_uint.c (main): Ditto.
981 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
982 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
983 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
984 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
985 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
986 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
987 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
988 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
989 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
991 2007-08-07 Andrew Haley <aph@redhat.com>
993 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
996 2007-08-06 Andrew Haley <aph@redhat.com>
999 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1000 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1003 2007-08-02 David Daney <ddaney@avtrex.com>
1005 * testsuite/libffi.call/return_ul.c (main): Define return type as
1006 ffi_arg. Use proper printf conversion specifier.
1008 2007-07-30 Andrew Haley <aph@redhat.com>
1011 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1012 signed/unsigned int8/16.
1013 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1015 Remove code to pop args from the stack after call.
1016 Special-case signed/unsigned int8/16.
1017 * testsuite/libffi.call/return_sc.c (main): Revert.
1019 2007-07-26 Richard Guenther <rguenther@suse.de>
1022 * testsuite/libffi.call/return_sc.c (main): Verify call
1023 result as signed char, not ffi_arg.
1025 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1027 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1028 * configure: Regenerate.
1030 2007-07-11 David Daney <ddaney@avtrex.com>
1032 * src/mips/ffi.c: Don't include sys/cachectl.h.
1033 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1036 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1038 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1039 from (ffi_prep_closure): ... this.
1040 (FFI_INIT_TRAMPOLINE): Adjust.
1042 2005-12-31 Phil Blundell <pb@reciva.com>
1044 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1045 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1046 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1047 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1048 (FFI_CLOSURES): Enable closure support.
1050 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1052 * testsuite/libffi.call/cls_multi_ushort.c,
1053 testsuite/libffi.call/cls_align_uint16.c,
1054 testsuite/libffi.call/nested_struct1.c,
1055 testsuite/libffi.call/nested_struct3.c,
1056 testsuite/libffi.call/cls_7_1_byte.c,
1057 testsuite/libffi.call/cls_double.c,
1058 testsuite/libffi.call/nested_struct5.c,
1059 testsuite/libffi.call/nested_struct7.c,
1060 testsuite/libffi.call/cls_sint.c,
1061 testsuite/libffi.call/nested_struct9.c,
1062 testsuite/libffi.call/cls_20byte1.c,
1063 testsuite/libffi.call/cls_multi_sshortchar.c,
1064 testsuite/libffi.call/cls_align_sint64.c,
1065 testsuite/libffi.call/cls_3byte2.c,
1066 testsuite/libffi.call/cls_multi_schar.c,
1067 testsuite/libffi.call/cls_multi_uchar.c,
1068 testsuite/libffi.call/cls_19byte.c,
1069 testsuite/libffi.call/cls_9byte1.c,
1070 testsuite/libffi.call/cls_align_float.c,
1071 testsuite/libffi.call/closure_fn1.c,
1072 testsuite/libffi.call/problem1.c,
1073 testsuite/libffi.call/closure_fn3.c,
1074 testsuite/libffi.call/cls_sshort.c,
1075 testsuite/libffi.call/closure_fn5.c,
1076 testsuite/libffi.call/cls_align_double.c,
1077 testsuite/libffi.call/cls_2byte.c,
1078 testsuite/libffi.call/nested_struct.c,
1079 testsuite/libffi.call/nested_struct10.c,
1080 testsuite/libffi.call/cls_4byte.c,
1081 testsuite/libffi.call/cls_6byte.c,
1082 testsuite/libffi.call/cls_8byte.c,
1083 testsuite/libffi.call/cls_multi_sshort.c,
1084 testsuite/libffi.call/cls_align_uint32.c,
1085 testsuite/libffi.call/cls_align_sint16.c,
1086 testsuite/libffi.call/cls_float.c,
1087 testsuite/libffi.call/cls_20byte.c,
1088 testsuite/libffi.call/cls_5_1_byte.c,
1089 testsuite/libffi.call/nested_struct2.c,
1090 testsuite/libffi.call/cls_24byte.c,
1091 testsuite/libffi.call/nested_struct4.c,
1092 testsuite/libffi.call/nested_struct6.c,
1093 testsuite/libffi.call/cls_64byte.c,
1094 testsuite/libffi.call/nested_struct8.c,
1095 testsuite/libffi.call/cls_uint.c,
1096 testsuite/libffi.call/cls_multi_ushortchar.c,
1097 testsuite/libffi.call/cls_schar.c,
1098 testsuite/libffi.call/cls_uchar.c,
1099 testsuite/libffi.call/cls_align_uint64.c,
1100 testsuite/libffi.call/cls_ulonglong.c,
1101 testsuite/libffi.call/cls_align_longdouble.c,
1102 testsuite/libffi.call/cls_1_1byte.c,
1103 testsuite/libffi.call/cls_12byte.c,
1104 testsuite/libffi.call/cls_3_1byte.c,
1105 testsuite/libffi.call/cls_3byte1.c,
1106 testsuite/libffi.call/cls_4_1byte.c,
1107 testsuite/libffi.call/cls_6_1_byte.c,
1108 testsuite/libffi.call/cls_16byte.c,
1109 testsuite/libffi.call/cls_18byte.c,
1110 testsuite/libffi.call/closure_fn0.c,
1111 testsuite/libffi.call/cls_9byte2.c,
1112 testsuite/libffi.call/closure_fn2.c,
1113 testsuite/libffi.call/closure_fn4.c,
1114 testsuite/libffi.call/cls_ushort.c,
1115 testsuite/libffi.call/closure_fn6.c,
1116 testsuite/libffi.call/cls_5byte.c,
1117 testsuite/libffi.call/cls_align_pointer.c,
1118 testsuite/libffi.call/cls_7byte.c,
1119 testsuite/libffi.call/cls_align_sint32.c,
1120 testsuite/libffi.special/unwindtest_ffi_call.cc,
1121 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1123 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1125 * aclocal.m4: Regenerated.
1127 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1129 * configure: Regenerate.
1131 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1133 * Makefile.in: Regenerate.
1134 * configure: Regenerate.
1135 * aclocal.m4: Regenerate.
1136 * include/Makefile.in: Regenerate.
1137 * testsuite/Makefile.in: Regenerate.
1139 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1141 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1142 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1143 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1144 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1145 (FFI_CLOSURES): Enable closure support.
1147 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1149 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1150 * configure: Regenerate.
1151 * fficonfig.h.in: Regenerate.
1152 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1153 CFI_OFFSET,CFI_DEF_CFA): New macros.
1154 (ffi_call_SYSV): Add callframe annotation.
1156 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1158 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1159 numerous test suite failures.
1160 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1162 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1164 * Makefile.am (EXTRA_DIST): Bring up to date.
1165 * Makefile.in: Regenerate.
1166 * src/frv/eabi.S: Remove RCS keyword.
1168 2007-04-06 Richard Henderson <rth@redhat.com>
1170 * configure.ac: Tidy target case.
1171 (HAVE_LONG_DOUBLE): Allow the target to override.
1172 * configure: Regenerate.
1173 * include/ffi.h.in: Don't define ffi_type_foo if
1174 LIBFFI_HIDE_BASIC_TYPES is defined.
1175 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1177 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1178 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1179 (ffi_type_longdouble): Special case for Alpha. Don't define
1180 if long double == double.
1182 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1183 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1184 (ffi_call, ffi_closure_osf_inner): Likewise.
1185 (ffi_closure_osf_inner): Likewise. Mark hidden.
1186 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1187 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1188 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1189 (load_table): Handle 128-bit long double.
1191 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1193 2007-04-06 Tom Tromey <tromey@redhat.com>
1196 * README: Fixed bug in example.
1198 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1200 * src/closures.c: Include sys/statfs.h.
1201 (_GNU_SOURCE): Define on Linux.
1202 (FFI_MMAP_EXEC_SELINUX): Define.
1203 (selinux_enabled): New variable.
1204 (selinux_enabled_check): New function.
1205 (is_selinux_enabled): Define.
1208 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1210 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1211 Use 'volatile float sum' to create sum of floats to avoid false
1212 negative due to excess precision on ix86 targets.
1215 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1217 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1219 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1221 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1223 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1224 (ffi_prep_closure_loc): New.
1225 (ffi_prep_raw_closure_loc): New.
1226 (ffi_prep_java_raw_closure_loc): New.
1227 * src/closures.c: New file.
1228 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1229 Replace sflags with exec_offset.
1230 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1231 sub_segment_exec_offset): New macros.
1232 (get_segment_flags, set_segment_flags, check_segment_merge): New
1234 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1235 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1236 destroy_mspace): Use new macros.
1237 (sys_alloc): Silence warning.
1238 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1239 * Makefile.in: Rebuilt.
1240 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1241 terms of ffi_prep_closure_loc.
1242 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1244 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1246 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1248 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1249 the renamed version.
1250 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1251 (ffi_prep_closure): ... this.
1252 * src/pa/ffi.c: Likewise.
1253 * src/cris/ffi.c: Likewise. Adjust.
1254 * src/frv/ffi.c: Likewise.
1255 * src/ia64/ffi.c: Likewise.
1256 * src/mips/ffi.c: Likewise.
1257 * src/powerpc/ffi_darwin.c: Likewise.
1258 * src/s390/ffi.c: Likewise.
1259 * src/sh/ffi.c: Likewise.
1260 * src/sh64/ffi.c: Likewise.
1261 * src/sparc/ffi.c: Likewise.
1262 * src/x86/ffi64.c: Likewise.
1263 * src/x86/ffi.c: Likewise.
1264 (FFI_INIT_TRAMPOLINE): Adjust.
1265 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1266 (ffi_prep_raw_closure): ... this.
1267 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1268 (ffi_prep_closure): ... this.
1269 (flush_icache): Adjust.
1271 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1273 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1276 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1278 * Makefile.am: Add dummy install-pdf target.
1279 * Makefile.in: Regenerate
1281 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1283 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1284 ffi_closure_helper_SYSV): Add long double handling.
1286 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1288 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1289 immediately after bctrl instruction.
1291 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1293 * Makefile.am (all-recursive, install-recursive,
1294 mostlyclean-recursive, clean-recursive, distclean-recursive,
1295 maintainer-clean-recursive): Add missing targets.
1296 * Makefile.in: Rebuilt.
1298 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1300 * configure.ac: Add TARGET for x86_64-*-darwin*.
1301 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1303 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1304 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1305 * configure: Regenerate.
1306 * Makefile.in: Regenerate.
1307 * include/Makefile.in: Regenerate.
1308 * testsuite/Makefile.in: Regenerate.
1309 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1312 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1314 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1317 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1319 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1320 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1322 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1323 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1325 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1326 Sandro Tolaini <tolaini@libero.it>
1328 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1330 * configure: Regenerated.
1331 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1332 (EXTRA_DIST): Add src/x86/darwin.S.
1333 * Makefile.in: Regenerated.
1334 * include/Makefile.in: Regenerated.
1335 * testsuite/Makefile.in: Regenerated.
1337 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1338 X86_WIN32, and additionally align stack to 16 bytes.
1339 * src/x86/darwin.S: New, based on sysv.S.
1340 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1342 2006-09-12 David Daney <ddaney@avtrex.com>
1345 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1346 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1347 * aclocal.m4: Regenerated for automake 1.9.6.
1348 * Makefile.in: Regenerated.
1349 * include/Makefile.in: Regenerated.
1350 * testsuite/Makefile.in: Regenerated.
1352 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1354 * include/ffi_common.h (struct): Revert accidental commit.
1356 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1358 * include/ffi_common.h: Remove lint directives.
1359 * include/ffi.h.in: Likewise.
1361 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1363 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1364 for 32-bit architectures.
1365 * testsuite/libffi.call/return_ul.c: New test case.
1367 2006-07-19 David Daney <ddaney@avtrex.com>
1369 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1370 xfail remains for mips64.
1372 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1374 * Makefile.am: Add install-html target. Add install-html to .PHONY
1375 * Makefile.in: Regenerate.
1376 * aclocal.m4: Regenerate.
1377 * include/Makefile.in: Regenerate.
1378 * testsuite/Makefile.in: Regenerate.
1380 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1382 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1385 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1387 * README: Remove notice about 'Crazy Comments'.
1388 * src/debug.c: Remove lint directives. Cleanup white spaces.
1389 * src/java_raw_api.c: Likewise.
1390 * src/prep_cif.c: Likewise.
1391 * src/raw_api.c: Likewise.
1392 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1394 * src/arm/ffi.c: Remove lint directives.
1395 * src/m32r/ffi.c: Likewise.
1396 * src/pa/ffi.c: Likewise.
1397 * src/powerpc/ffi.c: Likewise.
1398 * src/powerpc/ffi_darwin.c: Likewise.
1399 * src/sh/ffi.c: Likewise.
1400 * src/sh64/ffi.c: Likewise.
1401 * src/x86/ffi.c: Likewise.
1402 * testsuite/libffi.call/float2.c: Likewise.
1403 * testsuite/libffi.call/promotion.c: Likewise.
1404 * testsuite/libffi.call/struct1.c: Likewise.
1406 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1408 * src/pa/hpux32.S: Correct unwind offset calculation for
1410 * src/pa/linux.S: Likewise.
1412 2006-04-12 James E Wilson <wilson@specifix.com>
1415 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1416 (hfa_type_load): Call stf_spill.
1417 (hfa_type_store): Call ldf_fill.
1418 (ffi_call): Adjust calls to above routines. Add local temps for
1421 2006-04-10 Matthias Klose <doko@debian.org>
1423 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1424 directory names containing underscores.
1426 2006-04-07 James E Wilson <wilson@specifix.com>
1428 * testsuite/libffi.call/float4.c: New testcase.
1430 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1431 Andreas Tobler <a.tobler@schweiz.ch>
1433 * Makefile.am: Add PA_HPUX port.
1434 * Makefile.in: Regenerate.
1435 * include/Makefile.in: Likewise.
1436 * testsuite/Makefile.in: Likewise.
1437 * configure.ac: Add PA_HPUX rules.
1438 * configure: Regenerate.
1439 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1440 Add PA_HPUX and PA64_HPUX.
1441 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1442 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1443 (FFI_TYPE_SMALL_STRUCT2): Define.
1444 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1445 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1446 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1447 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1448 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1449 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1450 * src/pa/ffi.c (ROUND_DOWN): Delete.
1451 (fldw, fstw, fldd, fstd): Use '__asm__'.
1452 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1453 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1454 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1455 Simplify incrementing of stack slot variable. Change type of local
1456 'n' to unsigned int.
1457 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1459 (ffi_prep_cif_machdep): Likewise.
1460 (ffi_call): Likewise.
1461 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1462 return type to ffi_status. Simplify incrementing of stack slot
1463 variable. Only copy floating point argument registers when PA_LINUX
1464 is true. Reformat debug statement.
1465 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1466 FFI_TYPE_SMALL_STRUCT8.
1467 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1469 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1470 Add nops to cache flush. Add trampoline for PA_HPUX.
1471 * src/pa/hpux32.S: New file.
1472 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1473 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1474 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1475 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1476 argument type checks so that common argument types appear first.
1477 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1478 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1480 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1482 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1483 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1484 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1485 FFI_TYPE_LONGDOUBLE.
1486 (ffi_prep_args64): Assert using IBM extended double.
1487 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1488 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1489 (ffi_call): Handle FFI_LINUX.
1490 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1491 gpr3 return pointer as for struct return. Handle FFI_LINUX
1492 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1494 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1495 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1496 Don't use r6 as pointer to results, instead use sp offset. Don't
1497 make a special call to load lr with case table address, instead
1498 use offset from previous call.
1499 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1500 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1503 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1505 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1506 passed with FP registers correctly.
1507 (ffi_closure_helper_SYSV): Likewise.
1508 * src/sh64/sysv.S: Likewise.
1510 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1512 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1513 args and userdata unused.
1514 (closure_test_fn1): Mark cif and userdata unused.
1515 (main): Remove unused res.
1517 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1519 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1520 -O2, -O3, -Os and the warning flags -W -Wall.
1521 * testsuite/libffi.special/special.exp: Likewise.
1522 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1523 unused parameter unused for gcc or else do nothing.
1524 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1525 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1526 and userdata unused.
1527 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1528 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1529 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1530 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1531 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1532 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1533 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1534 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1535 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1536 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1537 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1538 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1539 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1540 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1541 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1542 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1543 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1544 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1545 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1546 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1547 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1548 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1549 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1550 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1552 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1554 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1556 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1557 void* to avoid compiler warning.
1559 (cls_struct_align_gn): Mark cif and userdata unused.
1560 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1562 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1564 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1566 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1568 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1570 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1571 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1572 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1574 (main): Cast res_call to silence gcc.
1575 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1577 (main): Cast res_call to silence gcc.
1578 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1580 (main): Cast res_call to silence gcc.
1581 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1583 (main): Cast res_call to silence gcc.
1584 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1586 (main): Cast res_call to silence gcc.
1587 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1589 (main): Cast res_call to silence gcc.
1590 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1592 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1593 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1595 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1596 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1598 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1599 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1601 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1602 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1604 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1605 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1606 and userdata unused.
1607 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1609 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1610 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1611 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1612 Cleanup white spaces.
1613 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1614 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1615 cif and userdata unused.
1616 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1618 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1619 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1620 formatters to silence gcc.
1621 (B_gn): Mark cif and userdata unused.
1622 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1624 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1625 (B_gn): Mark cif and userdata unused.
1626 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1628 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1629 (B_gn): Mark cif and userdata unused.
1630 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1632 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1633 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1634 * testsuite/libffi.call/problem1.c (stub): Likewise.
1635 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1637 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1638 in the last commit for this test case in the test case itself.
1639 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1641 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1642 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1643 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1644 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1645 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1646 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1648 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1650 * src/sh/sysv.S: Fix register numbers in the FDE for
1653 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1655 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1656 declaration to avoid a false negative on ix86. See PR323.
1658 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1660 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1661 and cast integer to void * if needed. Update the pointer to
1662 the FP register saved area correctly.
1664 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1666 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1668 * testsuite/libffi.call/nested_struct4.c: Likewise.
1670 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1672 * testsuite/libffi.call/return_dbl.c: New test case.
1673 * testsuite/libffi.call/return_dbl1.c: Likewise.
1674 * testsuite/libffi.call/return_dbl2.c: Likewise.
1675 * testsuite/libffi.call/return_fl.c: Likewise.
1676 * testsuite/libffi.call/return_fl1.c: Likewise.
1677 * testsuite/libffi.call/return_fl2.c: Likewise.
1678 * testsuite/libffi.call/return_fl3.c: Likewise.
1679 * testsuite/libffi.call/closure_fn6.c: Likewise.
1681 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1683 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1684 here to be used by other test cases too.
1686 * testsuite/libffi.call/nested_struct10.c: New test case.
1687 * testsuite/libffi.call/nested_struct9.c: Likewise.
1688 * testsuite/libffi.call/nested_struct8.c: Likewise.
1689 * testsuite/libffi.call/nested_struct7.c: Likewise.
1690 * testsuite/libffi.call/nested_struct6.c: Likewise.
1691 * testsuite/libffi.call/nested_struct5.c: Likewise.
1692 * testsuite/libffi.call/nested_struct4.c: Likewise.
1694 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1696 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1697 * configure: Rebuilt.
1699 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1701 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1702 instead do the shifting inline.
1703 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1704 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1705 and 8 byte structs, for the remaining struct sizes don't call
1706 __lshrdi3, instead do the shifting inline.
1708 2005-12-07 Thiemo Seufer <ths@networkno.de>
1710 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1711 missing parentheses.
1712 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1713 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1714 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1715 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1716 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1718 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1720 (ffi_prep_closure): Improve trampoline instruction scheduling.
1721 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1723 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1725 * src/powerpc/ffi.c: Formatting.
1726 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1727 (ffi_prep_args64): Likewise.
1729 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1731 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1732 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1734 2005-09-26 Tom Tromey <tromey@redhat.com>
1736 * testsuite/libffi.call/float1.c (value_type): New typedef.
1737 (CANARY): New define.
1738 (main): Check for result buffer overflow.
1739 * src/powerpc/linux64.S: Handle linux64 long double returns.
1740 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1741 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1743 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1746 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1748 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1750 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1752 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1753 (AH_BOTTOM): Add FFI_HIDDEN definition.
1754 * configure: Rebuilt.
1755 * fficonfig.h.in: Rebuilt.
1756 * src/powerpc/ffi.c (hidden): Remove.
1757 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1758 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1759 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1760 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1761 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1762 add FFI_HIDDEN to its prototype.
1763 (ffi_closure_SYSV_inner): New.
1764 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1765 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1767 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1770 * configure: Rebuilt.
1771 * configure.ac: Handle i*86-*-gnu*.
1773 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1775 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1776 DW_CFA_offset_extended_sf rather than
1777 DW_CFA_GNU_negative_offset_extended.
1778 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1780 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1782 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1784 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1785 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1786 partially on register.
1787 (ffi_closure_helper_SYSV): Likewise.
1788 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1790 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1792 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1794 * src/sh64/ffi.c (simple_type): Remove.
1795 (return_type): Handle small structures correctly.
1796 (ffi_prep_args): Likewise.
1797 (ffi_call): Likewise.
1798 (ffi_closure_helper_SYSV): Likewise.
1799 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1800 Emit position independent code if PIC and remove wrong datalabel
1801 prefixes from EH data.
1803 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1805 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1806 * Makefile.in: Regenerate.
1807 * include/Makefile.in: Likewise.
1808 * testsuite/Makefile.in: Likewise.
1809 * configure.ac: Add POWERPC_FREEBSD rules.
1810 * configure: Regenerate.
1811 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1812 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1813 * src/powerpc/ffi.c: Add flags to handle small structure returns
1815 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1817 (ffi_closure_helper_SYSV): Likewise.
1818 * src/powerpc/ppc_closure.S: Add return types for small structures.
1819 * src/powerpc/sysv.S: Add bits to handle small structures for
1822 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1824 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1825 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1826 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1828 2005-07-05 Randolph Chung <tausq@debian.org>
1830 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1831 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1832 structures. Kill compilation warnings.
1833 (ffi_closure_inner_LINUX): Print return values as hex in debug
1834 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1835 Properly handle 5-7 byte structure returns.
1836 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1837 (FFI_TYPE_SMALL_STRUCT2): Remove.
1838 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1839 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1840 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1841 (checksmst1, checksmst2): Remove.
1842 (checksmst3): Optimize handling of 3-byte struct returns.
1843 (checksmst567): Properly handle 5-7 byte struct returns.
1845 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1848 * src/mips/n32.S: Enforce PIC code.
1849 * src/mips/o32.S: Likewise.
1851 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1853 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1854 * configure: Regenerate.
1856 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1858 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1859 to call ffi_closure_helper_SYSV. Append @local instead.
1860 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1862 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1864 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1865 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1866 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1867 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1868 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1870 2005-05-09 Mike Stump <mrs@apple.com>
1872 * configure: Regenerate.
1874 2005-05-08 Richard Henderson <rth@redhat.com>
1877 * src/alpha/osf.S: Update unwind into to match code.
1879 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1880 Richard Henderson <rth@redhat.com>
1882 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1884 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1885 (ffi_prep_closure): Set carry bit if sse-used flag set.
1886 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1887 Only load sse registers if ssecount non-zero.
1888 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1890 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1892 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1893 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1894 * configure: Regenerate.
1896 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1898 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1899 have Tcl8.3-compatible intermediate variable.
1901 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1902 Hans-Peter Nilsson <hp@axis.com>
1904 * Makefile.am: Add CRIS support.
1905 * configure.ac: Likewise.
1906 * Makefile.in, configure, testsuite/Makefile.in,
1907 include/Makefile.in: Regenerate.
1908 * src/cris: New directory.
1909 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1910 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1912 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1913 \r?\n in output tests.
1915 2005-04-12 Mike Stump <mrs@apple.com>
1917 * configure: Regenerate.
1919 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1921 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1923 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1925 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1927 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1928 to 64 bits in ILP32 mode.
1929 Load 64 bits even for short data.
1931 2005-03-23 Mike Stump <mrs@apple.com>
1933 * src/powerpc/darwin.S: Update for -m64 multilib.
1934 * src/powerpc/darwin_closure.S: Likewise.
1936 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1938 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1939 Do not set tool_include_dir.
1940 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1942 * include/Makefile.am: Set gcc_version and toollibffidir.
1943 * include/Makefile.in: Regenerate.
1945 2005-02-22 Andrew Haley <aph@redhat.com>
1947 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1948 odd-numbered register pairs for 64-bit integer types.
1950 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1953 * testsuite/libffi.call/return_ll1.c: New test case.
1955 2005-02-11 Janis Johnson <janis187@us.ibm.com>
1957 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1958 * testsuite/libffi.call/float.c: Ditto.
1959 * testsuite/libffi.call/float2.c: Ditto.
1960 * testsuite/libffi.call/float3.c: Ditto.
1962 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1964 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1966 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1968 * testsuite/libffi.special/special.exp (cxx_options): Add
1971 2004-12-31 Richard Henderson <rth@redhat.com>
1973 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1974 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1975 offset parameters with a type parameter; deduce size and structure
1976 alignment. Update all users.
1978 2004-12-31 Richard Henderson <rth@redhat.com>
1980 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1981 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
1982 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
1983 into ffi_prep_closure.
1984 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1987 2004-12-27 Richard Henderson <rth@redhat.com>
1989 * src/x86/unix64.S: Fix typo in unwind info.
1991 2004-12-25 Richard Henderson <rth@redhat.com>
1993 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1994 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1995 (merge_classes): Check for it.
1997 (classify_argument): Pass byte_offset by value; perform all updates
1999 (examine_argument): Add classes argument; handle
2000 X86_64_COMPLEX_X87_CLASS.
2001 (ffi_prep_args): Merge into ...
2002 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2003 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2004 (ffi_fill_return_value): Remove.
2005 (ffi_prep_closure): Remove dead assert.
2006 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2007 Rewrite to use struct register_args instead of va_list. Create
2008 flags for handling structure returns.
2009 * src/x86/unix64.S: Remove dead strings.
2010 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2011 stack frame with ffi_call. Handle structure returns properly.
2012 (float2sse, floatfloat2sse, double2sse): Remove.
2013 (sse2float, sse2double, sse2floatfloat): Remove.
2014 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2015 to handle structure returns properly.
2017 2004-12-08 David Edelsohn <edelsohn@gnu.org>
2019 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2021 * Makefile.in: Regenerated.
2023 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2025 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2026 * configure, aclocal.m4, Makefile.in: Regenerate.
2027 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2029 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2031 * configure: Regenerate for libtool change.
2033 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2035 * configure: Regenerate for libtool reversion.
2037 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2039 * configure: Regenerate for libtool change.
2041 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2043 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2045 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2047 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2048 of jal. Use an absolute encoding for the frame information.
2050 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2052 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2053 * acinclude.m4: Delete logic for sincludes.
2054 * aclocal.m4, Makefile.in, configure: Regenerate.
2055 * include/Makefile: Likewise.
2056 * testsuite/Makefile: Likewise.
2058 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2060 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2061 on a 8-byte boundary.
2062 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2064 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2066 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2067 long long values. Round stack allocation to a multiple of 8 bytes
2068 for ATPCS compatibility.
2069 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2070 names. Handle returning long long types. Add Thumb and interworking
2071 support. Improve soft-float code.
2073 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2075 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2076 (libffi_exit): New function.
2077 (libffi_init): Build the testglue wrapper if needed.
2079 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2082 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2084 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2086 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2088 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2090 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2091 * testsuite/libffi.call/float3.c: New test case.
2093 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2095 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2096 the function returning a structure pointed with R2.
2097 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2098 the structure return value if T bit set. Emit position
2099 independent code and EH data if PIC.
2101 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2103 * Makefile.am: Add m32r support.
2104 * configure.ac: Likewise.
2105 * Makefile.in: Regenerate.
2106 * confiugre: Regenerate.
2107 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2108 (uint64, sint64, double, longdouble)
2109 * src/m32r: New directory.
2110 * src/m32r/ffi.c: New file.
2111 * src/m32r/sysv.S: Likewise.
2112 * src/m32r/ffitarget.h: Likewise.
2114 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2116 * testsuite/libffi.call/negint.c: New test case.
2118 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2121 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2122 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2123 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2126 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2128 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2129 * testsuite/libffi.call/promotion.c: Likewise.
2130 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2132 * testsuite/libffi.call/return_sc.c: Likewise.
2133 * testsuite/libffi.call/return_uc.c: Likewise.
2135 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2137 * src/powerpc/darwin.S: Fix comments and identation.
2138 * src/powerpc/darwin_closure.S: Likewise.
2140 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2142 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2143 (ffi_prep_args): Handle longdouble arguments.
2144 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2146 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2147 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2149 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2150 * src/types.c: Defined longdouble size and alignment for darwin.
2152 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2154 * src/powerpc/aix.S: Remove whitespaces.
2155 * src/powerpc/aix_closure.S: Likewise.
2156 * src/powerpc/asm.h: Likewise.
2157 * src/powerpc/ffi.c: Likewise.
2158 * src/powerpc/ffitarget.h: Likewise.
2159 * src/powerpc/linux64.S: Likewise.
2160 * src/powerpc/linux64_closure.S: Likewise.
2161 * src/powerpc/ppc_closure.S: Likewise.
2162 * src/powerpc/sysv.S: Likewise.
2164 2004-08-30 Anthony Green <green@redhat.com>
2166 * Makefile.am: Add frv support.
2167 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2168 * configure.ac: Read configure.host.
2169 * configure.in: Read configure.host.
2170 * configure.host: New file. frv-elf needs libgloss.
2171 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2172 alignment. This is needed to frv and shouldn't harm the others.
2173 * include/ffi_common.h (ALIGN_DOWN): New macro.
2174 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2176 2004-08-24 David Daney <daney@avtrex.com>
2178 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2179 * testsuite/libffi.call/closure_fn1.c: Likewise.
2180 * testsuite/libffi.call/closure_fn2.c Likewise.
2181 * testsuite/libffi.call/closure_fn3.c: Likewise.
2182 * testsuite/libffi.call/closure_fn4.c: Likewise.
2183 * testsuite/libffi.call/closure_fn5.c: Likewise.
2184 * testsuite/libffi.call/cls_18byte.c: Likewise.
2185 * testsuite/libffi.call/cls_19byte.c: Likewise.
2186 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2187 * testsuite/libffi.call/cls_20byte.c: Likewise.
2188 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2189 * testsuite/libffi.call/cls_24byte.c: Likewise.
2190 * testsuite/libffi.call/cls_2byte.c: Likewise.
2191 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2192 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2193 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2194 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2195 * testsuite/libffi.call/cls_4byte.c: Likewise.
2196 * testsuite/libffi.call/cls_64byte.c: Likewise.
2197 * testsuite/libffi.call/cls_6byte.c: Likewise.
2198 * testsuite/libffi.call/cls_7byte.c: Likewise.
2199 * testsuite/libffi.call/cls_8byte.c: Likewise.
2200 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2201 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2202 * testsuite/libffi.call/cls_align_double.c: Likewise.
2203 * testsuite/libffi.call/cls_align_float.c: Likewise.
2204 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2205 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2206 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2207 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2208 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2209 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2210 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2211 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2212 * testsuite/libffi.call/cls_double.c: Likewise.
2213 * testsuite/libffi.call/cls_float.c: Likewise.
2214 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2215 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2216 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2217 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2218 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2219 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2220 * testsuite/libffi.call/cls_schar.c: Likewise.
2221 * testsuite/libffi.call/cls_sint.c: Likewise.
2222 * testsuite/libffi.call/cls_sshort.c: Likewise.
2223 * testsuite/libffi.call/cls_uchar.c: Likewise.
2224 * testsuite/libffi.call/cls_uint.c: Likewise.
2225 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2226 * testsuite/libffi.call/cls_ushort.c: Likewise.
2227 * testsuite/libffi.call/nested_struct.c: Likewise.
2228 * testsuite/libffi.call/nested_struct1.c: Likewise.
2229 * testsuite/libffi.call/nested_struct2.c: Likewise.
2230 * testsuite/libffi.call/nested_struct3.c: Likewise.
2231 * testsuite/libffi.call/problem1.c: Likewise.
2232 * testsuite/libffi.special/unwindtest.cc: Likewise.
2233 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2235 * testsuite/libffi.call/cls_16byte.c: Likewise.
2236 * testsuite/libffi.call/cls_5byte.c: Likewise.
2238 2004-08-23 David Daney <daney@avtrex.com>
2241 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2242 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2243 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2244 parameters and return types.
2245 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2246 (ffi_prep_closure): Ditto.
2247 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2248 alignment calculations.
2249 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2250 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2252 2004-08-14 Casey Marshall <csm@gnu.org>
2254 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2255 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2256 integer (O32 ABI only).
2257 (ffi_prep_closure): new function.
2258 (ffi_closure_mips_inner_O32): new function.
2259 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2260 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2261 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2262 64 bit integers correctly.
2263 (ffi_closure_O32): new function.
2264 Added DWARF-2 unwind info for both functions.
2266 2004-08-10 Andrew Haley <aph@redhat.com>
2268 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2270 2004-08-01 Robert Millan <robertmh@gnu.org>
2272 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2273 * configure: Regenerate.
2275 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2277 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2278 and mmap() explicitly instead of relying on preset autoconf cache
2280 * aclocal.m4: Regenerate.
2281 * configure: Regenerate.
2283 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2285 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2286 (ffi_check_float_struct): Remove unused prototype.
2288 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2290 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2291 character on Darwin, use '\n\t' instead.
2293 2004-06-26 Matthias Klose <doko@debian.org>
2295 * libtool-version: Fix typo in revision/age.
2297 2004-06-17 Matthias Klose <doko@debian.org>
2299 * libtool-version: New.
2300 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2301 * Makefile.in: Regenerate.
2303 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2305 * Makefile.am: Remove useless multilib rules.
2306 * Makefile.in: Regenerate.
2307 * aclocal.m4: Regenerate with automake 1.8.5.
2308 * configure.ac: Remove useless multilib configury.
2309 * configure: Regenerate.
2311 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2313 * .cvsignore: New file.
2315 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2317 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2319 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2322 2004-06-06 Sean McNeil <sean@mcneil.com>
2324 * configure.ac: Add x86_64-*-freebsd* support.
2325 * configure: Regenerate.
2327 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2330 * configure.ac: Test for existence of mmap and sys/mman.h before
2331 checking blacklist. Fix suggested by Jim Wilson.
2332 * configure: Regenerate.
2334 2004-04-26 Matt Austern <austern@apple.com>
2336 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2338 * src/powerpc/darwin_closure.S: Likewise.
2340 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2342 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2343 error. Reported by Thomas Heller <theller@python.net>.
2344 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2345 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2347 2004-03-20 Matthias Klose <doko@debian.org>
2349 * src/pa/linux.S: Fix typo.
2351 2004-03-19 Matthias Klose <doko@debian.org>
2353 * Makefile.am: Update.
2354 * Makefile.in: Regenerate.
2355 * src/pa/ffi.h.in: Remove.
2356 * src/pa/ffitarget.h: New file.
2358 2004-02-10 Randolph Chung <tausq@debian.org>
2360 * Makefile.am: Add PA support.
2361 * Makefile.in: Regenerate.
2362 * include/Makefile.in: Regenerate.
2363 * configure.ac: Add PA target.
2364 * configure: Regenerate.
2365 * src/pa/ffi.c: New file.
2366 * src/pa/ffi.h.in: Add PA support.
2367 * src/pa/linux.S: New file.
2368 * prep_cif.c: Add PA support.
2370 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2372 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2374 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2375 with ecif->cif->flags.
2376 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2378 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2379 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2380 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2381 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2383 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2385 * configure.in: Rename file to ...
2386 * configure.ac: ... this.
2387 * fficonfig.h.in: Regenerate.
2388 * Makefile.in: Regenerate.
2389 * include/Makefile.in: Regenerate.
2390 * testsuite/Makefile.in: Regenerate.
2392 2004-03-12 Matt Austern <austern@apple.com>
2394 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2395 changes in EH format resulting from addition of linkonce support.
2396 * src/powerpc/darwin_closure.S: Likewise.
2398 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2399 Paolo Bonzini <bonzini@gnu.org>
2401 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2402 Remove VPATH. Remove rules for object files. Remove multilib support.
2403 (AM_CCASFLAGS): Add.
2404 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2405 (AC_PREREQ): Bump version to 2.59.
2406 (AC_INIT): Fill with version info and bug address.
2407 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2408 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2409 De-precious CC so that the right flags are passed down to multilibs.
2410 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2411 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2412 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2413 * configure: Rebuilt.
2414 * aclocal.m4: Likewise.
2415 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2416 * fficonfig.h.in: Likewise.
2418 2004-03-11 Andreas Schwab <schwab@suse.de>
2420 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2421 arguments from fp registers only for the first 8 parameter slots.
2422 Don't convert a float parameter when passed in memory.
2424 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2426 * configure: Regenerate for config/accross.m4 correction.
2428 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2430 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2431 ecif->cif->bytes to bytes.
2432 (ffi_prep_cif_machdep): Add braces around nested if statement.
2434 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2436 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2438 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2439 (ffi_closure_helper_LINUX64): Fix typo.
2440 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2442 * testsuite/libffi.call/float.c: Likewise.
2443 * testsuite/libffi.call/float2.c: Likewise.
2445 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2447 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2448 long double function return and long double arg handling.
2449 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2450 Use "end_pfr" instead of "nf". Correct long double handling.
2452 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2454 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2455 space for long double return value. Adjust stack frame and offsets.
2456 Load f2 long double return.
2458 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2460 * src/types.c: Use 16 byte long double for POWERPC64.
2462 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2464 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2465 when the structure return address is passed in %o0.
2466 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2467 (ffi_v9_layout_struct): Align the field following a nested structure
2468 on a word boundary. Use memmove instead of memcpy.
2469 (ffi_call): Update call to ffi_V9_return_struct.
2470 (ffi_prep_closure): Define 'ctx' only for V8.
2471 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2472 and ffi_closure_sparc_inner_v9.
2473 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2474 Always skip the structure return address. For structures and long
2475 doubles, copy the argument directly.
2476 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2477 if required. Shift the maximum floating-point slot accordingly. For
2478 big structures, copy the argument directly; otherwise, left-justify the
2479 argument and call ffi_v9_layout_struct to lay out the structure on
2481 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2482 (ffi_closure_v8): Pass the structure return address. Update call to
2483 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2484 Skip the 'unimp' insn when returning long doubles and structures.
2485 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2486 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2487 FFI_TYPE_INT handling. Load structures both in integers and
2488 floating-point registers on return.
2489 * README: Update status of the SPARC port.
2491 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2493 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2495 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2497 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2499 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2500 value as of type ffi_arg, not unsigned int.
2502 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2504 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2507 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2509 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2512 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2514 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2517 2003-12-10 Richard Henderson <rth@redhat.com>
2519 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2520 size_t instead of int.
2522 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2524 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2525 * testsuite/libffi.call/many_win32.c (main): Replace variable
2526 int i with unsigned long ul.
2528 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2529 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2530 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2531 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2532 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2533 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2534 * testsuite/libffi.call/cls_align_float.c: Likewise.
2535 * testsuite/libffi.call/cls_align_double.c: Likewise.
2536 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2537 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2539 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2542 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2543 Align arguments to 32 bits.
2545 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2548 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2549 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2550 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2551 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2552 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2553 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2555 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2557 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2559 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2560 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2562 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2564 * Makefile.in: Rebuilt.
2565 * configure: Likewise.
2566 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2569 2003-11-21 Andreas Jaeger <aj@suse.de>
2570 Andreas Tobler <a.tobler@schweiz.ch>
2572 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2573 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2574 * Makefile.in: Rebuilt.
2575 * aclocal.m4: Likewise.
2576 * configure: Likewise.
2577 * fficonfig.h.in: Likewise.
2578 * testsuite/lib/libffi-dg.exp: Add include dir.
2579 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2580 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2581 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2582 for ffi_closure if available.
2583 * testsuite/libffi.call/closure_fn1.c: Likewise.
2584 * testsuite/libffi.call/closure_fn2.c: Likewise.
2585 * testsuite/libffi.call/closure_fn3.c: Likewise.
2586 * testsuite/libffi.call/closure_fn4.c: Likewise.
2587 * testsuite/libffi.call/closure_fn5.c: Likewise.
2588 * testsuite/libffi.call/cls_12byte.c: Likewise.
2589 * testsuite/libffi.call/cls_16byte.c: Likewise.
2590 * testsuite/libffi.call/cls_18byte.c: Likewise.
2591 * testsuite/libffi.call/cls_19byte.c: Likewise.
2592 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2593 * testsuite/libffi.call/cls_20byte.c: Likewise.
2594 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2595 * testsuite/libffi.call/cls_24byte.c: Likewise.
2596 * testsuite/libffi.call/cls_2byte.c: Likewise.
2597 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2598 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2599 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2600 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2601 * testsuite/libffi.call/cls_4byte.c: Likewise.
2602 * testsuite/libffi.call/cls_5byte.c: Likewise.
2603 * testsuite/libffi.call/cls_64byte.c: Likewise.
2604 * testsuite/libffi.call/cls_6byte.c: Likewise.
2605 * testsuite/libffi.call/cls_7byte.c: Likewise.
2606 * testsuite/libffi.call/cls_8byte.c: Likewise.
2607 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2608 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2609 * testsuite/libffi.call/cls_double.c: Likewise.
2610 * testsuite/libffi.call/cls_float.c: Likewise.
2611 * testsuite/libffi.call/cls_schar.c: Likewise.
2612 * testsuite/libffi.call/cls_sint.c: Likewise.
2613 * testsuite/libffi.call/cls_sshort.c: Likewise.
2614 * testsuite/libffi.call/cls_uchar.c: Likewise.
2615 * testsuite/libffi.call/cls_uint.c: Likewise.
2616 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2617 * testsuite/libffi.call/cls_ushort.c: Likewise.
2618 * testsuite/libffi.call/nested_struct.c: Likewise.
2619 * testsuite/libffi.call/nested_struct1.c: Likewise.
2620 * testsuite/libffi.call/nested_struct2.c: Likewise.
2621 * testsuite/libffi.call/nested_struct3.c: Likewise.
2622 * testsuite/libffi.call/problem1.c: Likewise.
2623 * testsuite/libffi.special/unwindtest.cc: Likewise.
2625 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2627 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2629 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2631 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2632 Add -lgcc_s to additional flags.
2634 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2636 * configure.in, include/Makefile.am: PR libgcj/11147, install
2637 the ffitarget.h header file in a gcc versioned and target
2639 * configure: Regenerated.
2640 * Makefile.in, include/Makefile.in: Likewise.
2641 * testsuite/Makefile.in: Likewise.
2643 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2645 * testsuite/libffi.call/closure_fn0.c: Print result and check
2646 with dg-output to make debugging easier.
2647 * testsuite/libffi.call/closure_fn1.c: Likewise.
2648 * testsuite/libffi.call/closure_fn2.c: Likewise.
2649 * testsuite/libffi.call/closure_fn3.c: Likewise.
2650 * testsuite/libffi.call/closure_fn4.c: Likewise.
2651 * testsuite/libffi.call/closure_fn5.c: Likewise.
2652 * testsuite/libffi.call/cls_12byte.c: Likewise.
2653 * testsuite/libffi.call/cls_16byte.c: Likewise.
2654 * testsuite/libffi.call/cls_18byte.c: Likewise.
2655 * testsuite/libffi.call/cls_19byte.c: Likewise.
2656 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2657 * testsuite/libffi.call/cls_20byte.c: Likewise.
2658 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2659 * testsuite/libffi.call/cls_24byte.c: Likewise.
2660 * testsuite/libffi.call/cls_2byte.c: Likewise.
2661 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2662 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2663 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2664 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2665 * testsuite/libffi.call/cls_4byte.c: Likewise.
2666 * testsuite/libffi.call/cls_5byte.c: Likewise.
2667 * testsuite/libffi.call/cls_64byte.c: Likewise.
2668 * testsuite/libffi.call/cls_6byte.c: Likewise.
2669 * testsuite/libffi.call/cls_7byte.c: Likewise.
2670 * testsuite/libffi.call/cls_8byte.c: Likewise.
2671 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2672 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2673 * testsuite/libffi.call/cls_double.c: Likewise.
2674 * testsuite/libffi.call/cls_float.c: Likewise.
2675 * testsuite/libffi.call/cls_schar.c: Likewise.
2676 * testsuite/libffi.call/cls_sint.c: Likewise.
2677 * testsuite/libffi.call/cls_sshort.c: Likewise.
2678 * testsuite/libffi.call/cls_uchar.c: Likewise.
2679 * testsuite/libffi.call/cls_uint.c: Likewise.
2680 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2681 * testsuite/libffi.call/cls_ushort.c: Likewise.
2682 * testsuite/libffi.call/problem1.c: Likewise.
2684 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2687 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2689 * testsuite/libffi.call/cls_9byte2.c: New test case.
2690 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2691 * testsuite/libffi.call/cls_64byte.c: Likewise.
2692 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2693 * testsuite/libffi.call/cls_19byte.c: Likewise.
2694 * testsuite/libffi.call/cls_18byte.c: Likewise.
2695 * testsuite/libffi.call/closure_fn4.c: Likewise.
2696 * testsuite/libffi.call/closure_fn5.c: Likewise.
2697 * testsuite/libffi.call/cls_schar.c: Likewise.
2698 * testsuite/libffi.call/cls_sint.c: Likewise.
2699 * testsuite/libffi.call/cls_sshort.c: Likewise.
2700 * testsuite/libffi.call/nested_struct2.c: Likewise.
2701 * testsuite/libffi.call/nested_struct3.c: Likewise.
2703 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2705 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2706 * testsuite/libffi.call/cls_uchar.c: Likewise.
2707 * testsuite/libffi.call/cls_uint.c: Likewise.
2708 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2709 * testsuite/libffi.call/cls_ushort.c: Likewise.
2710 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2712 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2714 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2717 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2719 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2720 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2722 2003-10-22 David Daney <ddaney@avtrex.com>
2724 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2725 __attribute__((__mode__(__SI__))) and friends.
2727 2003-10-22 Andreas Schwab <schwab@suse.de>
2729 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2731 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2733 * configure.in: AC_LINK_FILES(ffitarget.h).
2734 * configure: Regenerate.
2735 * Makefile.in: Likewise.
2736 * include/Makefile.in: Likewise.
2737 * testsuite/Makefile.in: Likewise.
2738 * fficonfig.h.in: Likewise.
2740 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2741 Richard Henderson <rth@redhat.com>
2743 Avoid that ffi.h includes fficonfig.h.
2745 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2746 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2747 (TARGET_SRC_MIPS_SGI): Removed.
2748 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2749 (MIPS_SGI): Removed.
2750 (CLEANFILES): Removed.
2751 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2753 * acconfig.h: Removed.
2754 * configure.in: Compute sizeofs only for double and long double.
2755 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2756 into AC_DEFINE instead of using acconfig.h. Create
2757 include/ffitarget.h instead of include/fficonfig.h. Rename
2758 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2759 AC_DEFINE EH_FRAME_FLAGS.
2760 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2761 (hack_DATA): Add ffitarget.h.
2762 * include/ffi.h.in: Remove all system specific definitions.
2763 Declare raw API even if it is not installed, why bother?
2764 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2765 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2766 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2767 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2768 * include/ffi_common.h (bool): Do not define.
2769 (ffi_assert): Accept failed assertion.
2770 (ffi_type_test): Return void and accept file/line.
2771 (FFI_ASSERT): Pass stringized failed assertion.
2772 (FFI_ASSERT_AT): New macro.
2773 (FFI_ASSERT_VALID_TYPE): New macro.
2774 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2775 UINT64, SINT64): Define here with gcc's __attribute__ macro
2777 (FLOAT32, ALIGN): Define here instead of in ffi.h
2778 * include/ffi-mips.h: Removed. Its content moved to
2779 src/mips/ffitarget.h after separating assembly and C sections.
2780 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2781 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2782 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2783 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2784 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2785 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2786 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2787 (ffi_type_test): Rewritten.
2788 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2789 FFI_ASSERT_VALID_TYPE.
2790 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2791 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2792 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2793 src/s390/ffitarget.h, src/sh/ffitarget.h,
2794 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2795 src/x86/ffitarget.h: New files.
2796 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2797 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2798 src/powerpc/aix.S, src/powerpc/darwin.S,
2799 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2800 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2801 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2802 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2803 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2806 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2808 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2809 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2811 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2813 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2814 Used when FFI_DEBUG = 1.
2816 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2818 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2821 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2823 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2824 FFI_MIPS_O32 for O32 ABI.
2826 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2828 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2829 SPARC64. Cleanup whitespaces.
2831 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2833 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2834 strongarm, xscale. Cleanup whitespaces.
2835 * testsuite/libffi.call/closure_fn1.c: Likewise.
2836 * testsuite/libffi.call/closure_fn2.c: Likewise.
2837 * testsuite/libffi.call/closure_fn3.c: Likewise.
2838 * testsuite/libffi.call/cls_12byte.c: Likewise.
2839 * testsuite/libffi.call/cls_16byte.c: Likewise.
2840 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2841 * testsuite/libffi.call/cls_20byte.c: Likewise.
2842 * testsuite/libffi.call/cls_24byte.c: Likewise.
2843 * testsuite/libffi.call/cls_2byte.c: Likewise.
2844 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2845 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2846 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2847 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2848 * testsuite/libffi.call/cls_4byte.c: Likewise.
2849 * testsuite/libffi.call/cls_5byte.c: Likewise.
2850 * testsuite/libffi.call/cls_6byte.c: Likewise.
2851 * testsuite/libffi.call/cls_7byte.c: Likewise.
2852 * testsuite/libffi.call/cls_8byte.c: Likewise.
2853 * testsuite/libffi.call/cls_double.c: Likewise.
2854 * testsuite/libffi.call/cls_float.c: Likewise.
2855 * testsuite/libffi.call/cls_uchar.c: Likewise.
2856 * testsuite/libffi.call/cls_uint.c: Likewise.
2857 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2858 * testsuite/libffi.call/cls_ushort.c: Likewise.
2859 * testsuite/libffi.call/nested_struct.c: Likewise.
2860 * testsuite/libffi.call/nested_struct1.c: Likewise.
2861 * testsuite/libffi.call/problem1.c: Likewise.
2862 * testsuite/libffi.special/unwindtest.cc: Likewise.
2863 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2865 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2867 * src/powerpc/aix.S: Cleanup whitespaces.
2868 * src/powerpc/aix_closure.S: Likewise.
2870 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2872 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2873 * src/powerpc/darwin_closure.S: Likewise.
2874 * src/powerpc/ffi_darwin.c: Likewise.
2876 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2877 David Edelsohn <edelsohn@gnu.org>
2879 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2880 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2882 * src/powerpc/darwin_closure.S: Likewise.
2883 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2884 according to the Darwin/AIX ABI.
2885 (ffi_prep_cif_machdep): Likewise.
2886 (ffi_closure_helper_DARWIN): Likewise.
2887 Remove the outgoing parameter stack logic. Simplify the evaluation
2888 of the different CASE types.
2889 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2890 statement in the trampoline code.
2892 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2894 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2895 for the register size.
2896 (ffi_closure_helper_SYSV): Handle the structure return value
2898 (ffi_closure_helper_SYSV): Return the appropriate type when
2899 the registers are used for the structure return value.
2900 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2901 the 64-bit return value. Update copyright years.
2903 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2905 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2906 srcdir for ffi_mips.h.
2908 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2910 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2912 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2913 placement of float result.
2914 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2915 cast of "resp" for big-endian 64 bit machines.
2917 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2919 * src/types.c (double, longdouble): Merge identical SH and ARM
2920 typedefs, and add POWERPC64.
2921 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2922 struct split over gpr and rest.
2923 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2924 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2926 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2928 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2931 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2933 * configure: Regenerate.
2935 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2937 * Makefile.am: Remove build rules for ffitest.
2938 * Makefile.in: Rebuilt.
2940 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2942 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2943 about implicit declaration of abort().
2945 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2947 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2948 * Makefile.in: Rebuilt.
2949 * configure.in: Add dejagnu test framework.
2950 * configure: Rebuilt.
2952 * testsuite/Makefile.am: New file.
2953 * testsuite/Makefile.in: Built
2954 * testsuite/lib/libffi-dg.exp: New file.
2955 * testsuite/config/default.exp: Likewise.
2956 * testsuite/libffi.call/call.exp: Likewise.
2957 * testsuite/libffi.call/ffitest.h: Likewise.
2958 * testsuite/libffi.call/closure_fn0.c: Likewise.
2959 * testsuite/libffi.call/closure_fn1.c: Likewise.
2960 * testsuite/libffi.call/closure_fn2.c: Likewise.
2961 * testsuite/libffi.call/closure_fn3.c: Likewise.
2962 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2963 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2964 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2965 * testsuite/libffi.call/cls_2byte.c: Likewise.
2966 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2967 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2968 * testsuite/libffi.call/cls_4byte.c: Likewise.
2969 * testsuite/libffi.call/cls_5byte.c: Likewise.
2970 * testsuite/libffi.call/cls_6byte.c: Likewise.
2971 * testsuite/libffi.call/cls_7byte.c: Likewise.
2972 * testsuite/libffi.call/cls_8byte.c: Likewise.
2973 * testsuite/libffi.call/cls_12byte.c: Likewise.
2974 * testsuite/libffi.call/cls_16byte.c: Likewise.
2975 * testsuite/libffi.call/cls_20byte.c: Likewise.
2976 * testsuite/libffi.call/cls_24byte.c: Likewise.
2977 * testsuite/libffi.call/cls_double.c: Likewise.
2978 * testsuite/libffi.call/cls_float.c: Likewise.
2979 * testsuite/libffi.call/cls_uchar.c: Likewise.
2980 * testsuite/libffi.call/cls_uint.c: Likewise.
2981 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2982 * testsuite/libffi.call/cls_ushort.c: Likewise.
2983 * testsuite/libffi.call/float.c: Likewise.
2984 * testsuite/libffi.call/float1.c: Likewise.
2985 * testsuite/libffi.call/float2.c: Likewise.
2986 * testsuite/libffi.call/many.c: Likewise.
2987 * testsuite/libffi.call/many_win32.c: Likewise.
2988 * testsuite/libffi.call/nested_struct.c: Likewise.
2989 * testsuite/libffi.call/nested_struct1.c: Likewise.
2990 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2991 * testsuite/libffi.call/problem1.c: Likewise.
2992 * testsuite/libffi.call/promotion.c: Likewise.
2993 * testsuite/libffi.call/return_ll.c: Likewise.
2994 * testsuite/libffi.call/return_sc.c: Likewise.
2995 * testsuite/libffi.call/return_uc.c: Likewise.
2996 * testsuite/libffi.call/strlen.c: Likewise.
2997 * testsuite/libffi.call/strlen_win32.c: Likewise.
2998 * testsuite/libffi.call/struct1.c: Likewise.
2999 * testsuite/libffi.call/struct2.c: Likewise.
3000 * testsuite/libffi.call/struct3.c: Likewise.
3001 * testsuite/libffi.call/struct4.c: Likewise.
3002 * testsuite/libffi.call/struct5.c: Likewise.
3003 * testsuite/libffi.call/struct6.c: Likewise.
3004 * testsuite/libffi.call/struct7.c: Likewise.
3005 * testsuite/libffi.call/struct8.c: Likewise.
3006 * testsuite/libffi.call/struct9.c: Likewise.
3007 * testsuite/libffi.special/special.exp: New file.
3008 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3009 * testsuite/libffi.special/unwindtest.cc: Likewise.
3012 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3014 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3017 2003-08-02 Alan Modra <amodra@bigpond.net.au>
3019 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3021 (ffi_closure_helper_LINUX64): Likewise.
3022 * src/powerpc/linux64.S: Remove code writing to parm save area.
3023 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3024 address in lr from ffi_closure_helper_LINUX64 call to calculate
3025 table address. Optimize function tail.
3027 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3029 * src/sparc/ffi.c: Handle all floating point registers.
3030 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3032 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3034 * README: Note that libffi is not part of GCC. Update the project
3037 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3039 * src/powerpc/ppc_closure.S: Include ffi.h.
3041 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3043 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3044 Use C style comments.
3046 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3048 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3049 * Makefile.in: Regenerate.
3050 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3051 * configure: Regenerate.
3052 * include/ffi.h.in: Add SHmedia support.
3053 * src/sh64/ffi.c: New file.
3054 * src/sh64/sysv.S: New file.
3056 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3058 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3059 should be read-only.
3060 * configure: Rebuilt.
3061 * fficonfig.h.in: Rebuilt.
3062 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3063 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3064 * src/powerpc/linux64.S: Likewise.
3065 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3066 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3067 if -fpic/-fPIC/-mrelocatable.
3068 * src/powerpc/powerpc_closure.S: Likewise.
3069 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3070 #write in .eh_frame flags.
3071 * src/sparc/v9.S: Likewise.
3072 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3073 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3074 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3076 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3078 Fixes PR bootstrap/10656
3079 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3080 support for .register pseudo-op.
3081 * src/sparc/v8.S: Use it.
3082 * fficonfig.h.in: Rebuilt.
3083 * configure: Rebuilt.
3085 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3087 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3088 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3089 Make it the default on POWERPC64.
3090 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3091 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3092 * configure: Rebuilt.
3093 * src/powerpc/ffi.c (hidden): Define.
3094 (ffi_prep_args_SYSV): Renamed from
3095 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3096 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3097 ASM_NEEDS_REGISTERS64): New.
3098 (ffi_prep_args64): New function.
3099 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3100 (ffi_call): Likewise.
3101 (ffi_prep_closure): Likewise.
3102 (flush_icache): Surround by #ifndef POWERPC64.
3103 (ffi_dblfl): New union type.
3104 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3105 (ffi_closure_helper_LINUX64): New function.
3106 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3108 * src/powerpc/sysv.S: Likewise.
3109 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3110 * src/powerpc/linux64.S: New file.
3111 * src/powerpc/linux64_closure.S: New file.
3112 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3113 src/powerpc/linux64_closure.S.
3114 (TARGET_SRC_POWERPC): Likewise.
3116 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3117 closure_test_fn3): Fix result printing on big-endian 64-bit
3119 (main): Print tst2_arg instead of uninitialized tst2_result.
3121 * src/ffitest.c (main): Hide what closure pointer really points to
3124 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3126 * configure.in (arm-*-netbsdelf*): Add configuration.
3127 (configure): Regenerated.
3129 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3131 * include/Makefile.in: Regenerate.
3133 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3135 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3137 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3138 Receive closure pointer through parameter, read args using
3139 __builtin_dwarf_cfa.
3140 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3142 2003-03-12 Andreas Schwab <schwab@suse.de>
3144 * configure.in: Avoid trailing /. in toolexeclibdir.
3145 * configure: Rebuilt.
3147 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3149 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3151 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3153 * libffi/src/powerpc/darwin_closure.S:
3154 Fix alignement bug, allocate 8 bytes for the result.
3155 * libffi/src/powerpc/aix_closure.S:
3157 * libffi/src/powerpc/ffi_darwin.c:
3158 Update stackframe description for aix/darwin_closure.S.
3160 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3162 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3165 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3166 Andreas Schwab <schwab@suse.de>
3168 * configure.in: Adjust command to source config-ml.in to account
3169 for changes to the libffi_basedir definition.
3170 (libffi_basedir): Remove ${srcdir} from value and include trailing
3173 * configure: Regenerate.
3175 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3177 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3179 2003-01-28 Andrew Haley <aph@redhat.com>
3181 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3182 * src/x86/ffi64.c (ffi_prep_closure): New.
3183 (ffi_closure_UNIX64_inner): New.
3184 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3186 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
3188 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3189 Remove USE_LIBDIR conditional.
3190 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3191 * Makefile.in, configure: Rebuilt.
3193 2003-01027 David Edelsohn <edelsohn@gnu.org>
3195 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3196 * Makefile.in: Regenerate.
3198 2003-01-22 Andrew Haley <aph@redhat.com>
3200 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3203 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3205 * src/powerpc/darwin.S: Add unwind info.
3206 * src/powerpc/darwin_closure.S: Likewise.
3208 2003-01-14 Andrew Haley <aph@redhat.com>
3210 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3211 (ffi_prep_cif_machdep): Likewise.
3212 * src/x86/unix64.S: Add unwind info.
3214 2003-01-14 Andreas Jaeger <aj@suse.de>
3216 * src/ffitest.c (main): Only use ffi_closures if those are
3219 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3221 * libffi/src/ffitest.c
3222 add closure testcases
3224 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3226 * libffi/src/powerpc/ffi.c
3227 fix alignment bug for float (4 byte aligned iso 8 byte)
3229 2003-01-09 Geoffrey Keating <geoffk@apple.com>
3231 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3232 * src/powerpc/darwin.S: Remove RCS version string.
3234 2003-01-03 Jeff Sturm <jsturm@one-point.com>
3236 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3237 * src/ffitest.c (main): Use static storage for closure.
3238 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3239 * src/sparc/v8.S (ffi_closure_v8): New.
3240 * src/sparc/v9.S (ffi_closure_v9): New.
3242 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3244 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3245 enumeration for X86_WIN32.
3246 * src/x86/win32.S: Added ffi_call_STDCALL function
3248 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3249 switch cases for recognising FFI_STDCALL and
3250 calling ffi_call_STDCALL if target is X86_WIN32.
3251 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3252 stdcall versions of the "my_strlen" and "many"
3253 test functions (for X86_WIN32).
3254 Added test cases to test stdcall invocation using
3257 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3259 * src/sh/sysv.S: Add DWARF2 unwind info.
3261 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3263 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3265 2002-11-26 Jim Wilson <wilson@redhat.com>
3267 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3269 2002-11-23 H.J. Lu <hjl@gnu.org>
3271 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3272 Include ../config/accross.m4.
3273 * aclocal.m4; Rebuild.
3274 * configure: Likewise.
3276 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3278 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3280 2002-11-11 DJ Delorie <dj@redhat.com>
3282 * configure.in: Look for common files in the right place.
3284 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3286 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3287 raw data as _Jv_word values, not ffi_raw.
3288 (ffi_java_ptrarray_to_raw): Likewise.
3289 (ffi_java_rvalue_to_raw): New function.
3290 (ffi_java_raw_call): Call it.
3291 (ffi_java_raw_to_rvalue): New function.
3292 (ffi_java_translate_args): Call it.
3293 * src/ffitest.c (closure_test_fn): Interpret return value
3294 as ffi_arg, not int.
3295 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3296 FFI_TYPE_POINTER case.
3297 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3298 values extended to word size.
3300 2002-10-02 Andreas Jaeger <aj@suse.de>
3302 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3304 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
3306 * include/ffi.h.in: Fix i386 win32 compilation.
3308 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3310 * configure.in: Add s390x-*-linux-* target.
3311 * configure: Regenerate.
3312 * include/ffi.h.in: Define S390X for s390x targets.
3313 (FFI_CLOSURES): Define for s390/s390x.
3314 (FFI_TRAMPOLINE_SIZE): Likewise.
3315 (FFI_NATIVE_RAW_API): Likewise.
3316 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3317 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
3318 * src/s390/ffi.c: Major rework of existing code. Add support for
3319 s390x targets. Add closure support.
3320 * src/s390/sysv.S: Likewise.
3322 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
3324 * src/arm/sysv.S: Fix typo.
3326 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
3328 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3329 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3330 (ffi_call_SYSV): Handle soft-float.
3332 2002-09-27 Bo Thorsen <bo@suse.de>
3334 * include/ffi.h.in: Fix multilib x86-64 support.
3336 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3338 * Makefile.am (all-multi): Fix multilib parallel build.
3340 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3342 * configure.in (sh[34]*-*-linux*): Add brackets.
3343 * configure: Regenerate.
3345 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3347 * Makefile.am: Add SH support.
3348 * Makefile.in: Regenerate.
3349 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3350 * configure: Regenerate.
3351 * include/ffi.h.in: Add SH support.
3352 * src/sh/ffi.c: New file.
3353 * src/sh/sysv.S: New file.
3354 * src/types.c: Add SH support.
3356 2002-07-16 Bo Thorsen <bo@suse.de>
3358 * src/x86/ffi64.c: New file that adds x86-64 support.
3359 * src/x86/unix64.S: New file that handles argument setup for
3361 * src/x86/sysv.S: Don't use this on x86-64.
3362 * src/x86/ffi.c: Don't use this on x86-64.
3364 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3366 * src/ffitest.c (struct6): New test that tests a special case in
3368 (struct7): Likewise.
3369 (struct8): Likewise.
3370 (struct9): Likewise.
3371 (closure_test_fn): Silence warning about this when it's not used.
3372 (main): Add the new tests.
3373 (main): Fix a couple of wrong casts and silence some compiler warnings.
3374 * include/ffi.h.in: Add x86-64 ABI definition.
3375 * fficonfig.h.in: Regenerate.
3376 * Makefile.am: Add x86-64 support.
3377 * configure.in: Likewise.
3378 * Makefile.in: Regenerate.
3379 * configure: Likewise.
3381 2002-06-24 Bo Thorsen <bo@suse.de>
3383 * src/types.c: Merge settings for similar architectures.
3384 Add x86-64 sizes and alignments.
3386 2002-06-23 Bo Thorsen <bo@suse.de>
3388 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3389 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3390 * src/mips/ffi.c (ffi_prep_args): Likewise.
3391 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3393 2002-07-18 H.J. Lu (hjl@gnu.org)
3395 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3396 (libffi_la_SOURCES): Support MIPS_LINUX.
3397 (libffi_convenience_la_SOURCES): Likewise.
3398 * Makefile.in: Regenerated.
3400 * configure.in (mips64*-*): Skip.
3401 (mips*-*-linux*): New.
3402 * configure: Regenerated.
3404 * src/mips/ffi.c: Include <sgidefs.h>.
3406 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3408 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3410 2002-05-27 Roger Sayle <roger@eyesopen.com>
3412 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3414 2002-05-27 Bo Thorsen <bo@suse.de>
3416 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3419 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3421 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3422 beginning of function (for older apple cc).
3424 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3426 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3427 script entry, and set LD to it when configuring multilibs.
3428 * configure: Rebuilt.
3430 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3432 * configure.in (sparc64-*-netbsd*): Add target.
3433 (sparc-*-netbsdelf*): Likewise.
3434 * configure: Regenerate.
3436 2002-04-28 David S. Miller <davem@redhat.com>
3438 * configure.in, configure: Fix SPARC test in previous change.
3440 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3442 * Makefile.am: Add Linux for S/390 support.
3443 * Makefile.in: Regenerate.
3444 * configure.in: Add Linux for S/390 support.
3445 * configure: Regenerate.
3446 * include/ffi.h.in: Add Linux for S/390 support.
3447 * src/s390/ffi.c: New file from libffi CVS tree.
3448 * src/s390/sysv.S: New file from libffi CVS tree.
3450 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3452 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3454 * src/sparc/v8.S: Use it.
3455 * src/sparc/v9.S: Likewise.
3456 * fficonfig.h.in: Rebuilt.
3457 * configure: Rebuilt.
3459 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3461 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3463 * src/ia64/unix.S: Add unwind information. Fix comments.
3464 Save sp in a way that's compatible with unwind info.
3465 (ffi_call_unix): Correctly restore sp in all cases.
3466 * src/ia64/ffi.c: Add, fix comments.
3468 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3470 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3472 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3474 * configure.in (alpha*-*-netbsd*): Add target.
3475 * configure: Regenerate.
3477 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3479 * src/sparc/v8.S: Add unwind info.
3480 * src/sparc/v9.S: Likewise.
3482 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3484 * configure.in: Enable i*86-*-netbsdelf*.
3485 * configure: Rebuilt.
3487 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3490 * src/mips/n32.s: Delete
3491 * src/mips/o32.s: Delete
3493 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3495 * configure.in: Enable alpha*-*-freebsd*.
3496 * configure: Rebuilt.
3498 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3500 * Makefile.am: libfficonvenience -> libffi_convenience.
3501 * Makefile.in: Rebuilt.
3503 * Makefile.am: Define ffitest_OBJECTS.
3504 * Makefile.in: Rebuilt.
3506 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3507 David Edelsohn <edelsohn@gnu.org>
3509 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3510 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3511 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3512 * Makefile.in: Regenerate.
3513 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3514 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3515 (flush_icache, flush_range): New functions.
3516 (ffi_closure_helper_DARWIN): New function.
3517 * src/powerpc/aix_closure.S: New file.
3518 * src/powerpc/darwin_closure.S: New file.
3520 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3522 * include/ffi.h.in: Add typedef for ffi_arg.
3523 * src/ffitest.c (main): Declare rint with ffi_arg.
3525 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3527 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3528 number of GPRs for floating-point arguments.
3530 2002-01-31 Anthony Green <green@redhat.com>
3532 * configure: Rebuilt.
3533 * configure.in: Replace CHECK_SIZEOF and endian tests with
3534 cross-compiler friendly macros.
3535 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3538 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3540 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3541 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3543 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3545 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3546 (TARGET_SRC_POWERPC_AIX): New.
3547 (POWERPC_AIX): New stanza.
3548 * Makefile.in: Regenerate.
3549 * configure.in: Add AIX case.
3550 * configure: Regenerate.
3551 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3552 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3553 size. Fix "long double" support.
3554 (ffi_call): Add FFI_AIX case.
3555 * src/powerpc/aix.S: New.
3557 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3559 Implement Darwin PowerPC ABI.
3560 * configure.in: Handle powerpc-*-darwin*.
3561 * Makefile.am: Set source files for POWERPC_DARWIN.
3562 * configure: Rebuilt.
3563 * Makefile.in: Rebuilt.
3564 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3566 * src/powerpc/darwin.S: New file.
3567 * src/powerpc/ffi_darwin.c: New file.
3569 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3571 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3573 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3575 * src/x86/sysv.S: Avoid gas-only .balign directive.
3576 Use C style comments.
3578 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3580 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3581 Fixes PR bootstrap/3563.
3583 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3585 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3587 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3589 * configure.in: Recognize sparc*-sun-* host.
3590 * configure: Regenerate.
3592 2001-06-06 Andrew Haley <aph@redhat.com>
3594 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
3596 2001-06-03 Andrew Haley <aph@redhat.com>
3598 * src/alpha/osf.S: Add unwind info.
3599 * src/powerpc/sysv.S: Add unwind info.
3600 * src/powerpc/ppc_closure.S: Likewise.
3602 2000-05-31 Jeff Sturm <jsturm@one-point.com>
3604 * configure.in: Fix AC_ARG_ENABLE usage.
3605 * configure: Rebuilt.
3607 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3609 * configure.in: Remove warning about beta code.
3610 * configure: Rebuilt.
3612 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3614 * src/ia64/unix.S: Restore stack pointer when returning from
3616 * src/ia64/ffi.c: Fix typo in comment.
3618 2001-04-18 Jim Wilson <wilson@redhat.com>
3620 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3621 to eliminate RAW DV.
3623 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3625 * Makefile.am: Make a libtool convenience library.
3626 * Makefile.in: Rebuilt.
3628 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3630 * configure.in: Use different syntax for subdirectory creation.
3631 * configure: Rebuilt.
3633 2001-03-27 Jon Beniston <jon@beniston.com>
3635 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3636 * configure: Rebuilt.
3637 * Makefile.am: Added X86_WIN32 target support.
3638 * Makefile.in: Rebuilt.
3640 * include/ffi.h.in: Added X86_WIN32 target support.
3642 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3643 * src/types.c: Added X86_WIN32 target support.
3645 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3646 stuff removed and made to work with CygWin's gas.
3648 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3650 * configure.in: Make target subdirectory in build dir.
3651 * Makefile.am: Override suffix based rules to specify correct output
3653 * Makefile.in: Rebuilt.
3654 * configure: Rebuilt.
3656 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3658 * src/powerpc/ppc_closure.S: New file.
3659 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3660 involving long long and register pairs.
3661 (ffi_prep_closure): New function.
3662 (flush_icache): Likewise.
3663 (ffi_closure_helper_SYSV): Likewise.
3664 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3665 (FFI_TRAMPOLINE_SIZE): Likewise.
3666 (FFI_NATIVE_RAW_API): Likewise.
3667 * Makefile.in: Rebuilt.
3668 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3669 (TARGET_SRC_POWERPC): Likewise.
3671 2001-03-19 Tom Tromey <tromey@redhat.com>
3673 * Makefile.in: Rebuilt.
3674 * Makefile.am (ffitest_LDFLAGS): New macro.
3676 2001-03-02 Nick Clifton <nickc@redhat.com>
3678 * include/ffi.h.in: Remove RCS ident string.
3679 * include/ffi_mips.h: Remove RCS ident string.
3680 * src/debug.c: Remove RCS ident string.
3681 * src/ffitest.c: Remove RCS ident string.
3682 * src/prep_cif.c: Remove RCS ident string.
3683 * src/types.c: Remove RCS ident string.
3684 * src/alpha/ffi.c: Remove RCS ident string.
3685 * src/alpha/osf.S: Remove RCS ident string.
3686 * src/arm/ffi.c: Remove RCS ident string.
3687 * src/arm/sysv.S: Remove RCS ident string.
3688 * src/mips/ffi.c: Remove RCS ident string.
3689 * src/mips/n32.S: Remove RCS ident string.
3690 * src/mips/o32.S: Remove RCS ident string.
3691 * src/sparc/ffi.c: Remove RCS ident string.
3692 * src/sparc/v8.S: Remove RCS ident string.
3693 * src/sparc/v9.S: Remove RCS ident string.
3694 * src/x86/ffi.c: Remove RCS ident string.
3695 * src/x86/sysv.S: Remove RCS ident string.
3697 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3699 * include/ffi.h.in: Change sourceware.cygnus.com references to
3702 2000-12-09 Richard Henderson <rth@redhat.com>
3704 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3705 (ffi_closure_osf_inner): Index rather than increment avalue
3706 and arg_types. Give ffi_closure_osf the raw return value type.
3707 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3710 2000-12-07 Richard Henderson <rth@redhat.com>
3712 * src/raw_api.c (ffi_translate_args): Fix typo.
3713 (ffi_prep_closure): Likewise.
3715 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3716 FFI_TRAMPOLINE_SIZE.
3717 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3718 cif->bytes for new ffi_call_osf implementation.
3719 (ffi_prep_args): Absorb into ...
3720 (ffi_call): ... here. Do all stack allocation here and
3721 avoid a callback function.
3722 (ffi_prep_closure, ffi_closure_osf_inner): New.
3723 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3724 (ffi_closure_osf): New.
3726 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3728 * config.guess, config.sub, install-sh: Removed.
3729 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3730 * Makefile.in: Rebuilt.
3732 * acinclude.m4: Include libtool macros from the top level.
3733 * aclocal.m4, configure: Rebuilt.
3735 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3737 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3738 * configure: Rebuilt.
3740 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3742 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3743 memory correctly. Use conditional instructions, not branches where
3746 2000-05-04 Tom Tromey <tromey@cygnus.com>
3748 * configure: Rebuilt.
3749 * configure.in: Match `arm*-*-linux-*'.
3750 From Chris Dornan <cdornan@arm.com>.
3752 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3754 * Makefile.am (SUBDIRS): Define.
3755 (AM_MAKEFLAGS): Likewise.
3756 (Multilib support.): Add section.
3757 * Makefile.in: Rebuilt.
3758 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3759 New variables. Set for gcc using -print-multi-lib. Export them
3761 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3762 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3764 (extra_compiler_flags_value, extra_compiler_flags): Check these
3765 for extra compiler options which need to be passed down in
3768 2000-04-16 Anthony Green <green@redhat.com>
3770 * configure: Rebuilt.
3771 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3773 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3775 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3776 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3777 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3778 Replace all void * sizeofs with sizeof(int).
3779 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3780 different than DOUBLE.
3781 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3782 (ffi_prep_args_v9): New function.
3783 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3784 (ffi_V9_return_struct): New function.
3785 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3786 32bit code (not yet cross-arch calls).
3787 * src/sparc/v8.S: Add struct return delay nop.
3789 * src/sparc/v9.S: New file.
3790 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3791 is used on sparc64 only for structures larger than 32 bytes.
3792 Pass by reference for structures is done for structure arguments
3793 larger than 16 bytes.
3794 * src/ffitest.c (main): Use 64bit rint on sparc64.
3795 Run long long tests on sparc.
3796 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3798 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3800 * configure.in (sparc-*-linux*): New supported target.
3801 (sparc64-*-linux*): Likewise.
3802 * configure: Rebuilt.
3803 * Makefile.am: Add v9.S to SPARC files.
3804 * Makefile.in: Likewise.
3805 (LINK): Surround $(CCLD) into double quotes, so that multilib
3806 compiles work correctly.
3808 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3810 * configure: Rebuilt.
3811 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3812 proposed by Bryce McKinlay.
3814 2000-03-20 Tom Tromey <tromey@cygnus.com>
3816 * Makefile.in: Hand edit for java_raw_api.lo.
3818 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3820 * config.guess, config.sub: Update from the gcc tree.
3821 Fix for PR libgcj/168.
3823 2000-03-03 Tom Tromey <tromey@cygnus.com>
3825 * Makefile.in: Fixed ia64 by hand.
3827 * configure: Rebuilt.
3828 * configure.in (--enable-multilib): New option.
3829 (libffi_basedir): New subst.
3830 (AC_OUTPUT): Added multilib code.
3832 2000-03-02 Tom Tromey <tromey@cygnus.com>
3834 * Makefile.in: Rebuilt.
3835 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3838 2000-02-25 Hans Boehm <boehm@acm.org>
3840 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3842 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3844 (ffi_prep_raw_closure): Use ffi_translate_args, not
3845 ffi_closure_translate.
3846 * src/java_raw_api.c: New file.
3847 * src/ffitest.c (closure_test_fn): New function.
3848 (main): Define `rint' as long long on IA64. Added new test when
3849 FFI_CLOSURES is defined.
3850 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3851 (ffi_abi): Recognize IA64.
3852 (ffi_raw): Added `flt' field.
3853 Added "Java raw API" code.
3854 * configure.in: Recognize ia64.
3855 * Makefile.am (TARGET_SRC_IA64): New macro.
3856 (libffi_la_common_SOURCES): Added java_raw_api.c.
3857 (libffi_la_SOURCES): Define in IA64 case.
3859 2000-01-04 Tom Tromey <tromey@cygnus.com>
3861 * Makefile.in: Rebuilt with newer automake.
3863 1999-12-31 Tom Tromey <tromey@cygnus.com>
3865 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3867 1999-09-01 Tom Tromey <tromey@cygnus.com>
3869 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3871 * fficonfig.h.in: Rebuilt.
3872 * configure: Rebuilt.
3873 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3874 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3875 * acconfig.h: Don't #undef PACKAGE or VERSION.
3877 1999-08-09 Anthony Green <green@cygnus.com>
3879 * include/ffi.h.in: Try to work around messy header problem
3880 with PACKAGE and VERSION.
3882 * configure: Rebuilt.
3883 * configure.in: Change version to 2.00-beta.
3885 * fficonfig.h.in: Rebuilt.
3886 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3888 * src/x86/ffi.c (ffi_raw_call): Rename.
3890 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3892 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3893 (ffi_prep_incoming_args_SYSV): Ditto.
3894 (ffi_prep_closure): Ditto.
3895 (ffi_closure_raw_SYSV): Ditto.
3896 (ffi_prep_raw_closure): More ditto.
3897 (ffi_call_raw): Final ditto.
3899 * include/ffi.h.in: Add definitions for closure and raw API.
3901 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3904 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3906 * src/raw_api.c: New file.
3908 * include/ffi.h.in (ffi_raw): New type.
3909 (UINT_ARG, SINT_ARG): New defines.
3910 (ffi_closure, ffi_raw_closure): New types.
3911 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3913 * configure.in: Add check for endianness and sizeof void*.
3915 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3916 instead of directly.
3918 * configure: Rebuilt.
3920 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3922 * configure.in: Add x86 and powerpc BeOS configurations.
3923 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3925 1999-05-09 Anthony Green <green@cygnus.com>
3927 * configure.in: Add warning about this being beta code.
3928 Remove src/Makefile.am from the picture.
3929 * configure: Rebuilt.
3931 * Makefile.am: Move logic from src/Makefile.am. Add changes
3932 to support libffi as a target library.
3933 * Makefile.in: Rebuilt.
3935 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3936 Upgraded to new autoconf, automake, libtool.
3940 * LICENSE: Update copyright date.
3942 * src/Makefile.am, src/Makefile.in: Removed.
3944 1998-11-29 Anthony Green <green@cygnus.com>
3946 * include/ChangeLog: Removed.
3947 * src/ChangeLog: Removed.
3948 * src/mips/ChangeLog: Removed.
3949 * src/sparc/ChangeLog: Remboved.
3950 * src/x86/ChangeLog: Removed.
3952 * ChangeLog.v1: Created.