1 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
3 * configure.ac: Also check if assembler supports pc-relative
4 relocs on X86_WIN32 targets.
5 * configure: Regenerate.
6 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
7 (_ffi_call_SYSV): Add missing function type symbol .def and
9 (_ffi_call_STDCALL): Likewise.
10 (_ffi_closure_SYSV): Likewise.
11 (_ffi_closure_raw_SYSV): Likewise.
12 (.eh_frame): Add hand-crafted EH data.
14 2009-04-09 Jakub Jelinek <jakub@redhat.com>
16 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
17 version 3 of the GNU General Public License and to point readers
18 at the COPYING3 file and the FSF's license web page.
19 * testsuite/libffi.call/call.exp: Likewise.
20 * testsuite/libffi.special/special.exp: Likewise.
22 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
24 * configure: Regenerate.
26 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
29 * configure.ac (HAVE_AS_X86_PCREL): New test.
30 * configure: Regenerate.
31 * fficonfig.h.in: Regenerate.
32 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
33 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
34 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
35 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
36 * src/x86/unix64.S (.Lstore_table): Move to .text section.
37 (.Lload_table): Likewise.
38 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
40 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
42 * configure: Regenerate.
44 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
46 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
47 signed/unsigned int8/16 return values.
48 * src/sparc/v8.S (ffi_call_v8): Likewise.
49 (ffi_closure_v8): Likewise.
51 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
52 Steve Ellcey <sje@cup.hp.com>
54 * configure: Regenerate for new libtool.
56 * include/Makefile.in: Ditto.
59 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
61 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
62 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
63 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
64 Adjust copyright notice.
65 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
66 register or two register to use for FFI_SYSV structs.
67 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
68 (ffi_closure_helper_SYSV): Fix the return type for
69 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
70 Adjust copyright notice.
72 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
74 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
77 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
79 * configure: Regenerate.
80 * include/Makefile.in: Regenerate.
81 * testsuite/Makefile.in: Regenerate.
83 2008-06-07 Joseph Myers <joseph@codesourcery.com>
85 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
86 powerpc-*-beos*): Remove.
87 * configure: Regenerate.
89 2008-05-09 Julian Brown <julian@codesourcery.com>
91 * Makefile.am (LTLDFLAGS): New.
92 (libffi_la_LDFLAGS): Use above.
93 * Makefile.in: Regenerate.
95 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
98 * aclocal.m4: Regenerate.
99 * configure: Regenerate.
101 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
103 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
104 * src/sh64/sysv.S: Likewise.
106 2008-03-26 Daniel Jacobowitz <dan@debian.org>
108 * src/arm/sysv.S: Fix ARM comment marker.
110 2008-03-26 Jakub Jelinek <jakub@redhat.com>
112 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
113 * src/s390/sysv.S: Likewise.
114 * src/powerpc/ppc_closure.S: Likewise.
115 * src/powerpc/sysv.S: Likewise.
116 * src/x86/unix64.S: Likewise.
117 * src/x86/sysv.S: Likewise.
118 * src/sparc/v8.S: Likewise.
119 * src/sparc/v9.S: Likewise.
120 * src/m68k/sysv.S: Likewise.
121 * src/arm/sysv.S: Likewise.
123 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
125 * aclocal.m4: Regenerate.
126 * configure: Likewise.
127 * Makefile.in: Likewise.
128 * include/Makefile.in: Likewise.
129 * testsuite/Makefile.in: Likewise.
131 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
132 Andreas Tobler <a.tobler@schweiz.org>
134 * configure.ac: Add amd64-*-freebsd* target.
135 * configure: Regenerate.
137 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
140 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
143 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
146 2008-01-24 David Edelsohn <edelsohn@gnu.org>
148 * configure: Regenerate.
150 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
152 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
154 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
157 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
158 signed/unsigned int8/16 for X86_DARWIN.
159 Updated copyright info.
160 Handle one and two byte structs with special cif->flags.
161 * src/x86/ffitarget.h: Add special types for one and two byte structs.
162 Updated copyright info.
163 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
165 Remove code to pop args from the stack after call.
166 Special-case signed/unsigned for int8/16, one and two byte structs.
167 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
168 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
170 Updated copyright info.
172 2007-12-08 David Daney <ddaney@avtrex.com>
174 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
175 SUBU, add with ADDU and use smaller code sequences.
177 2007-12-07 David Daney <ddaney@avtrex.com>
179 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
182 2007-12-06 David Daney <ddaney@avtrex.com>
184 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
186 (ffi_java_raw): New typedef.
187 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
188 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
190 (ffi_java_raw_closure) : Same.
191 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
193 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
195 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
196 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
197 sizeof(ffi_java_raw) for alignment calculations.
198 (ffi_java_ptrarray_to_raw): Same.
199 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
200 if FFI_SIZEOF_JAVA_RAW == 4.
201 (ffi_java_raw_to_rvalue): Same.
202 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
203 (ffi_java_translate_args): Same.
204 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
206 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
208 2007-12-06 David Daney <ddaney@avtrex.com>
210 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
213 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
216 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
217 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
218 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
219 set the NUM_FPR_ARG_REGISTERS according to.
220 Add support for potential soft-float support under hard-float
222 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
223 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
224 to the FFI_LINUX_SOFT_FLOAT ABI.
225 (ffi_prep_cif_machdep): Likewise.
226 (ffi_closure_helper_SYSV): Likewise.
227 * src/powerpc/ppc_closure.S: Make sure not to store float/double
228 on archs where __NO_FPRS__ is true.
229 Add FFI_TYPE_UINT128 support.
230 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
231 Adjust copyright notice.
233 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
235 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
236 * include/ffi_common.h: ... here.
239 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
241 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
242 * src/powerpc/linux64.S: Likewise.
243 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
244 * testsuite/libffi.call/return_ldl.c: New test.
246 2007-09-04 <aph@redhat.com>
248 * src/arm/sysv.S (UNWIND): New.
249 (Whole file): Conditionally compile unwinder directives.
250 * src/arm/sysv.S: Add unwinder directives.
252 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
253 Only treat r0 as a struct address if we're actually returning a
255 Only copy the bytes that are actually within a struct.
256 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
257 is returned in r0, not passed by address.
258 (ffi_call): Allocate a word-sized temporary for the case where
259 a composite is returned in r0.
260 (ffi_prep_incoming_args_SYSV): Align as necessary.
262 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
264 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
265 directly using the sys_cacheflush syscall.
267 2007-07-27 Andrew Haley <aph@redhat.com>
269 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
271 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
273 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
274 * configure.ac: Likewise.
275 * Makefile.in: Regenerate.
276 * include/Makefile.in: Likewise.
277 * testsuite/Makefile.in: Likewise.
278 * configure: Likewise.
280 2007-08-24 David Daney <ddaney@avtrex.com>
282 * testsuite/libffi.call/return_sl.c: New test.
284 2007-08-10 David Daney <ddaney@avtrex.com>
286 * testsuite/libffi.call/cls_multi_ushort.c,
287 testsuite/libffi.call/cls_align_uint16.c,
288 testsuite/libffi.call/nested_struct1.c,
289 testsuite/libffi.call/nested_struct3.c,
290 testsuite/libffi.call/cls_7_1_byte.c,
291 testsuite/libffi.call/nested_struct5.c,
292 testsuite/libffi.call/cls_double.c,
293 testsuite/libffi.call/nested_struct7.c,
294 testsuite/libffi.call/cls_sint.c,
295 testsuite/libffi.call/nested_struct9.c,
296 testsuite/libffi.call/cls_20byte1.c,
297 testsuite/libffi.call/cls_multi_sshortchar.c,
298 testsuite/libffi.call/cls_align_sint64.c,
299 testsuite/libffi.call/cls_3byte2.c,
300 testsuite/libffi.call/cls_multi_schar.c,
301 testsuite/libffi.call/cls_multi_uchar.c,
302 testsuite/libffi.call/cls_19byte.c,
303 testsuite/libffi.call/cls_9byte1.c,
304 testsuite/libffi.call/cls_align_float.c,
305 testsuite/libffi.call/closure_fn1.c,
306 testsuite/libffi.call/problem1.c,
307 testsuite/libffi.call/closure_fn3.c,
308 testsuite/libffi.call/cls_sshort.c,
309 testsuite/libffi.call/closure_fn5.c,
310 testsuite/libffi.call/cls_align_double.c,
311 testsuite/libffi.call/nested_struct.c,
312 testsuite/libffi.call/cls_2byte.c,
313 testsuite/libffi.call/nested_struct10.c,
314 testsuite/libffi.call/cls_4byte.c,
315 testsuite/libffi.call/cls_6byte.c,
316 testsuite/libffi.call/cls_8byte.c,
317 testsuite/libffi.call/cls_multi_sshort.c,
318 testsuite/libffi.call/cls_align_sint16.c,
319 testsuite/libffi.call/cls_align_uint32.c,
320 testsuite/libffi.call/cls_20byte.c,
321 testsuite/libffi.call/cls_float.c,
322 testsuite/libffi.call/nested_struct2.c,
323 testsuite/libffi.call/cls_5_1_byte.c,
324 testsuite/libffi.call/nested_struct4.c,
325 testsuite/libffi.call/cls_24byte.c,
326 testsuite/libffi.call/nested_struct6.c,
327 testsuite/libffi.call/cls_64byte.c,
328 testsuite/libffi.call/nested_struct8.c,
329 testsuite/libffi.call/cls_uint.c,
330 testsuite/libffi.call/cls_multi_ushortchar.c,
331 testsuite/libffi.call/cls_schar.c,
332 testsuite/libffi.call/cls_uchar.c,
333 testsuite/libffi.call/cls_align_uint64.c,
334 testsuite/libffi.call/cls_ulonglong.c,
335 testsuite/libffi.call/cls_align_longdouble.c,
336 testsuite/libffi.call/cls_1_1byte.c,
337 testsuite/libffi.call/cls_12byte.c,
338 testsuite/libffi.call/cls_3_1byte.c,
339 testsuite/libffi.call/cls_3byte1.c,
340 testsuite/libffi.call/cls_4_1byte.c,
341 testsuite/libffi.call/cls_6_1_byte.c,
342 testsuite/libffi.call/cls_16byte.c,
343 testsuite/libffi.call/cls_18byte.c,
344 testsuite/libffi.call/closure_fn0.c,
345 testsuite/libffi.call/cls_9byte2.c,
346 testsuite/libffi.call/closure_fn2.c,
347 testsuite/libffi.call/closure_fn4.c,
348 testsuite/libffi.call/cls_ushort.c,
349 testsuite/libffi.call/closure_fn6.c,
350 testsuite/libffi.call/cls_5byte.c,
351 testsuite/libffi.call/cls_align_pointer.c,
352 testsuite/libffi.call/cls_7byte.c,
353 testsuite/libffi.call/cls_align_sint32.c,
354 testsuite/libffi.special/unwindtest_ffi_call.cc,
355 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
357 2007-08-10 David Daney <ddaney@avtrex.com>
360 * configure.ac: Don't treat mips64 as a special case.
361 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
362 * configure: Regenerate
363 * Makefile.in: Ditto.
364 * fficonfig.h.in: Ditto.
365 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
366 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
367 (FFI_DEFAULT_ABI): Set for n64 case.
368 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
369 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
370 (ffi_closure_N32): New function.
371 (.eh_frame): New section
372 * src/mips/o32.S: Clean up comments.
373 (ffi_closure_O32): Pass ffi_closure parameter in $12.
374 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
375 _MIPS_SIM == _ABIN32 throughout.
376 (FFI_MIPS_STOP_HERE): New, use in place of
378 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
379 to support n32/n64 ABIs.
380 (calc_n32_struct_flags): Rewrite.
381 (calc_n32_return_struct_flags): Remove unused variable. Reverse
382 position of flag bits.
383 (ffi_prep_cif_machdep): Rewrite n32 portion.
384 (ffi_call): Enable for n64. Add special handling for small structure
386 (ffi_prep_closure_loc): Add n32 and n64 support.
387 (ffi_closure_mips_inner_O32): Add cast to silence warning.
388 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
390 2007-08-08 David Daney <ddaney@avtrex.com>
392 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
393 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
395 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
396 * testsuite/libffi.call/cls_sint.c (main): Ditto.
397 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
398 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
399 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
400 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
401 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
402 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
403 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
404 * testsuite/libffi.call/return_fl3.c (main): Ditto.
405 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
406 * testsuite/libffi.call/nested_struct.c (main): Ditto.
407 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
408 * testsuite/libffi.call/return_ll1.c (main): Ditto.
409 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
410 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
411 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
412 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
413 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
414 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
415 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
416 * testsuite/libffi.call/cls_uint.c (main): Ditto.
417 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
418 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
419 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
420 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
421 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
422 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
423 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
424 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
425 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
427 2007-08-07 Andrew Haley <aph@redhat.com>
429 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
432 2007-08-06 Andrew Haley <aph@redhat.com>
435 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
436 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
439 2007-08-02 David Daney <ddaney@avtrex.com>
441 * testsuite/libffi.call/return_ul.c (main): Define return type as
442 ffi_arg. Use proper printf conversion specifier.
444 2007-07-30 Andrew Haley <aph@redhat.com>
447 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
448 signed/unsigned int8/16.
449 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
451 Remove code to pop args from the stack after call.
452 Special-case signed/unsigned int8/16.
453 * testsuite/libffi.call/return_sc.c (main): Revert.
455 2007-07-26 Richard Guenther <rguenther@suse.de>
458 * testsuite/libffi.call/return_sc.c (main): Verify call
459 result as signed char, not ffi_arg.
461 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
463 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
464 * configure: Regenerate.
466 2007-07-11 David Daney <ddaney@avtrex.com>
468 * src/mips/ffi.c: Don't include sys/cachectl.h.
469 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
472 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
474 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
475 from (ffi_prep_closure): ... this.
476 (FFI_INIT_TRAMPOLINE): Adjust.
478 2005-12-31 Phil Blundell <pb@reciva.com>
480 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
481 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
482 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
483 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
484 (FFI_CLOSURES): Enable closure support.
486 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
488 * testsuite/libffi.call/cls_multi_ushort.c,
489 testsuite/libffi.call/cls_align_uint16.c,
490 testsuite/libffi.call/nested_struct1.c,
491 testsuite/libffi.call/nested_struct3.c,
492 testsuite/libffi.call/cls_7_1_byte.c,
493 testsuite/libffi.call/cls_double.c,
494 testsuite/libffi.call/nested_struct5.c,
495 testsuite/libffi.call/nested_struct7.c,
496 testsuite/libffi.call/cls_sint.c,
497 testsuite/libffi.call/nested_struct9.c,
498 testsuite/libffi.call/cls_20byte1.c,
499 testsuite/libffi.call/cls_multi_sshortchar.c,
500 testsuite/libffi.call/cls_align_sint64.c,
501 testsuite/libffi.call/cls_3byte2.c,
502 testsuite/libffi.call/cls_multi_schar.c,
503 testsuite/libffi.call/cls_multi_uchar.c,
504 testsuite/libffi.call/cls_19byte.c,
505 testsuite/libffi.call/cls_9byte1.c,
506 testsuite/libffi.call/cls_align_float.c,
507 testsuite/libffi.call/closure_fn1.c,
508 testsuite/libffi.call/problem1.c,
509 testsuite/libffi.call/closure_fn3.c,
510 testsuite/libffi.call/cls_sshort.c,
511 testsuite/libffi.call/closure_fn5.c,
512 testsuite/libffi.call/cls_align_double.c,
513 testsuite/libffi.call/cls_2byte.c,
514 testsuite/libffi.call/nested_struct.c,
515 testsuite/libffi.call/nested_struct10.c,
516 testsuite/libffi.call/cls_4byte.c,
517 testsuite/libffi.call/cls_6byte.c,
518 testsuite/libffi.call/cls_8byte.c,
519 testsuite/libffi.call/cls_multi_sshort.c,
520 testsuite/libffi.call/cls_align_uint32.c,
521 testsuite/libffi.call/cls_align_sint16.c,
522 testsuite/libffi.call/cls_float.c,
523 testsuite/libffi.call/cls_20byte.c,
524 testsuite/libffi.call/cls_5_1_byte.c,
525 testsuite/libffi.call/nested_struct2.c,
526 testsuite/libffi.call/cls_24byte.c,
527 testsuite/libffi.call/nested_struct4.c,
528 testsuite/libffi.call/nested_struct6.c,
529 testsuite/libffi.call/cls_64byte.c,
530 testsuite/libffi.call/nested_struct8.c,
531 testsuite/libffi.call/cls_uint.c,
532 testsuite/libffi.call/cls_multi_ushortchar.c,
533 testsuite/libffi.call/cls_schar.c,
534 testsuite/libffi.call/cls_uchar.c,
535 testsuite/libffi.call/cls_align_uint64.c,
536 testsuite/libffi.call/cls_ulonglong.c,
537 testsuite/libffi.call/cls_align_longdouble.c,
538 testsuite/libffi.call/cls_1_1byte.c,
539 testsuite/libffi.call/cls_12byte.c,
540 testsuite/libffi.call/cls_3_1byte.c,
541 testsuite/libffi.call/cls_3byte1.c,
542 testsuite/libffi.call/cls_4_1byte.c,
543 testsuite/libffi.call/cls_6_1_byte.c,
544 testsuite/libffi.call/cls_16byte.c,
545 testsuite/libffi.call/cls_18byte.c,
546 testsuite/libffi.call/closure_fn0.c,
547 testsuite/libffi.call/cls_9byte2.c,
548 testsuite/libffi.call/closure_fn2.c,
549 testsuite/libffi.call/closure_fn4.c,
550 testsuite/libffi.call/cls_ushort.c,
551 testsuite/libffi.call/closure_fn6.c,
552 testsuite/libffi.call/cls_5byte.c,
553 testsuite/libffi.call/cls_align_pointer.c,
554 testsuite/libffi.call/cls_7byte.c,
555 testsuite/libffi.call/cls_align_sint32.c,
556 testsuite/libffi.special/unwindtest_ffi_call.cc,
557 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
559 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
561 * aclocal.m4: Regenerated.
563 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
565 * configure: Regenerate.
567 2007-05-23 Steve Ellcey <sje@cup.hp.com>
569 * Makefile.in: Regenerate.
570 * configure: Regenerate.
571 * aclocal.m4: Regenerate.
572 * include/Makefile.in: Regenerate.
573 * testsuite/Makefile.in: Regenerate.
575 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
577 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
578 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
579 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
580 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
581 (FFI_CLOSURES): Enable closure support.
583 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
585 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
586 * configure: Regenerate.
587 * fficonfig.h.in: Regenerate.
588 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
589 CFI_OFFSET,CFI_DEF_CFA): New macros.
590 (ffi_call_SYSV): Add callframe annotation.
592 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
594 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
595 numerous test suite failures.
596 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
598 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
600 * Makefile.am (EXTRA_DIST): Bring up to date.
601 * Makefile.in: Regenerate.
602 * src/frv/eabi.S: Remove RCS keyword.
604 2007-04-06 Richard Henderson <rth@redhat.com>
606 * configure.ac: Tidy target case.
607 (HAVE_LONG_DOUBLE): Allow the target to override.
608 * configure: Regenerate.
609 * include/ffi.h.in: Don't define ffi_type_foo if
610 LIBFFI_HIDE_BASIC_TYPES is defined.
611 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
613 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
614 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
615 (ffi_type_longdouble): Special case for Alpha. Don't define
616 if long double == double.
618 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
619 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
620 (ffi_call, ffi_closure_osf_inner): Likewise.
621 (ffi_closure_osf_inner): Likewise. Mark hidden.
622 (ffi_call_osf, ffi_closure_osf): Mark hidden.
623 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
624 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
625 (load_table): Handle 128-bit long double.
627 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
629 2007-04-06 Tom Tromey <tromey@redhat.com>
632 * README: Fixed bug in example.
634 2007-04-03 Jakub Jelinek <jakub@redhat.com>
636 * src/closures.c: Include sys/statfs.h.
637 (_GNU_SOURCE): Define on Linux.
638 (FFI_MMAP_EXEC_SELINUX): Define.
639 (selinux_enabled): New variable.
640 (selinux_enabled_check): New function.
641 (is_selinux_enabled): Define.
644 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
646 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
647 Use 'volatile float sum' to create sum of floats to avoid false
648 negative due to excess precision on ix86 targets.
651 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
653 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
655 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
657 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
659 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
660 (ffi_prep_closure_loc): New.
661 (ffi_prep_raw_closure_loc): New.
662 (ffi_prep_java_raw_closure_loc): New.
663 * src/closures.c: New file.
664 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
665 Replace sflags with exec_offset.
666 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
667 sub_segment_exec_offset): New macros.
668 (get_segment_flags, set_segment_flags, check_segment_merge): New
670 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
671 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
672 destroy_mspace): Use new macros.
673 (sys_alloc): Silence warning.
674 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
675 * Makefile.in: Rebuilt.
676 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
677 terms of ffi_prep_closure_loc.
678 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
680 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
682 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
684 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
686 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
687 (ffi_prep_closure): ... this.
688 * src/pa/ffi.c: Likewise.
689 * src/cris/ffi.c: Likewise. Adjust.
690 * src/frv/ffi.c: Likewise.
691 * src/ia64/ffi.c: Likewise.
692 * src/mips/ffi.c: Likewise.
693 * src/powerpc/ffi_darwin.c: Likewise.
694 * src/s390/ffi.c: Likewise.
695 * src/sh/ffi.c: Likewise.
696 * src/sh64/ffi.c: Likewise.
697 * src/sparc/ffi.c: Likewise.
698 * src/x86/ffi64.c: Likewise.
699 * src/x86/ffi.c: Likewise.
700 (FFI_INIT_TRAMPOLINE): Adjust.
701 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
702 (ffi_prep_raw_closure): ... this.
703 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
704 (ffi_prep_closure): ... this.
705 (flush_icache): Adjust.
707 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
709 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
712 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
714 * Makefile.am: Add dummy install-pdf target.
715 * Makefile.in: Regenerate
717 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
719 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
720 ffi_closure_helper_SYSV): Add long double handling.
722 2007-02-02 Jakub Jelinek <jakub@redhat.com>
724 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
725 immediately after bctrl instruction.
727 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
729 * Makefile.am (all-recursive, install-recursive,
730 mostlyclean-recursive, clean-recursive, distclean-recursive,
731 maintainer-clean-recursive): Add missing targets.
732 * Makefile.in: Rebuilt.
734 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
736 * configure.ac: Add TARGET for x86_64-*-darwin*.
737 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
739 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
740 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
741 * configure: Regenerate.
742 * Makefile.in: Regenerate.
743 * include/Makefile.in: Regenerate.
744 * testsuite/Makefile.in: Regenerate.
745 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
748 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
750 * aclocal.m4: Regenerate with aclocal -I .. as written in the
753 2006-10-31 Geoffrey Keating <geoffk@apple.com>
755 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
756 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
758 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
759 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
761 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
762 Sandro Tolaini <tolaini@libero.it>
764 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
766 * configure: Regenerated.
767 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
768 (EXTRA_DIST): Add src/x86/darwin.S.
769 * Makefile.in: Regenerated.
770 * include/Makefile.in: Regenerated.
771 * testsuite/Makefile.in: Regenerated.
773 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
774 X86_WIN32, and additionally align stack to 16 bytes.
775 * src/x86/darwin.S: New, based on sysv.S.
776 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
778 2006-09-12 David Daney <ddaney@avtrex.com>
781 * include/Makefile.am: Install both ffi.h and ffitarget.h in
782 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
783 * aclocal.m4: Regenerated for automake 1.9.6.
784 * Makefile.in: Regenerated.
785 * include/Makefile.in: Regenerated.
786 * testsuite/Makefile.in: Regenerated.
788 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
790 * include/ffi_common.h (struct): Revert accidental commit.
792 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
794 * include/ffi_common.h: Remove lint directives.
795 * include/ffi.h.in: Likewise.
797 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
799 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
800 for 32-bit architectures.
801 * testsuite/libffi.call/return_ul.c: New test case.
803 2006-07-19 David Daney <ddaney@avtrex.com>
805 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
806 xfail remains for mips64.
808 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
810 * Makefile.am: Add install-html target. Add install-html to .PHONY
811 * Makefile.in: Regenerate.
812 * aclocal.m4: Regenerate.
813 * include/Makefile.in: Regenerate.
814 * testsuite/Makefile.in: Regenerate.
816 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
818 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
821 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
823 * README: Remove notice about 'Crazy Comments'.
824 * src/debug.c: Remove lint directives. Cleanup white spaces.
825 * src/java_raw_api.c: Likewise.
826 * src/prep_cif.c: Likewise.
827 * src/raw_api.c: Likewise.
828 * src/ffitest.c: Delete. No longer needed, all test cases migrated
830 * src/arm/ffi.c: Remove lint directives.
831 * src/m32r/ffi.c: Likewise.
832 * src/pa/ffi.c: Likewise.
833 * src/powerpc/ffi.c: Likewise.
834 * src/powerpc/ffi_darwin.c: Likewise.
835 * src/sh/ffi.c: Likewise.
836 * src/sh64/ffi.c: Likewise.
837 * src/x86/ffi.c: Likewise.
838 * testsuite/libffi.call/float2.c: Likewise.
839 * testsuite/libffi.call/promotion.c: Likewise.
840 * testsuite/libffi.call/struct1.c: Likewise.
842 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
844 * src/pa/hpux32.S: Correct unwind offset calculation for
846 * src/pa/linux.S: Likewise.
848 2006-04-12 James E Wilson <wilson@specifix.com>
851 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
852 (hfa_type_load): Call stf_spill.
853 (hfa_type_store): Call ldf_fill.
854 (ffi_call): Adjust calls to above routines. Add local temps for
857 2006-04-10 Matthias Klose <doko@debian.org>
859 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
860 directory names containing underscores.
862 2006-04-07 James E Wilson <wilson@specifix.com>
864 * testsuite/libffi.call/float4.c: New testcase.
866 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
867 Andreas Tobler <a.tobler@schweiz.ch>
869 * Makefile.am: Add PA_HPUX port.
870 * Makefile.in: Regenerate.
871 * include/Makefile.in: Likewise.
872 * testsuite/Makefile.in: Likewise.
873 * configure.ac: Add PA_HPUX rules.
874 * configure: Regenerate.
875 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
876 Add PA_HPUX and PA64_HPUX.
877 Rename FFI_LINUX ABI to FFI_PA32 ABI.
878 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
879 (FFI_TYPE_SMALL_STRUCT2): Define.
880 (FFI_TYPE_SMALL_STRUCT4): Likewise.
881 (FFI_TYPE_SMALL_STRUCT8): Likewise.
882 (FFI_TYPE_SMALL_STRUCT3): Redefine.
883 (FFI_TYPE_SMALL_STRUCT5): Likewise.
884 (FFI_TYPE_SMALL_STRUCT6): Likewise.
885 (FFI_TYPE_SMALL_STRUCT7): Likewise.
886 * src/pa/ffi.c (ROUND_DOWN): Delete.
887 (fldw, fstw, fldd, fstd): Use '__asm__'.
888 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
889 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
890 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
891 Simplify incrementing of stack slot variable. Change type of local
893 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
895 (ffi_prep_cif_machdep): Likewise.
896 (ffi_call): Likewise.
897 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
898 return type to ffi_status. Simplify incrementing of stack slot
899 variable. Only copy floating point argument registers when PA_LINUX
900 is true. Reformat debug statement.
901 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
902 FFI_TYPE_SMALL_STRUCT8.
903 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
905 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
906 Add nops to cache flush. Add trampoline for PA_HPUX.
907 * src/pa/hpux32.S: New file.
908 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
909 ffi_prep_args_LINUX to ffi_prep_args_pa32.
910 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
911 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
912 argument type checks so that common argument types appear first.
913 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
914 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
916 2006-03-24 Alan Modra <amodra@bigpond.net.au>
918 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
919 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
920 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
922 (ffi_prep_args64): Assert using IBM extended double.
923 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
924 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
925 (ffi_call): Handle FFI_LINUX.
926 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
927 gpr3 return pointer as for struct return. Handle FFI_LINUX
928 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
930 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
931 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
932 Don't use r6 as pointer to results, instead use sp offset. Don't
933 make a special call to load lr with case table address, instead
934 use offset from previous call.
935 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
936 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
939 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
941 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
942 passed with FP registers correctly.
943 (ffi_closure_helper_SYSV): Likewise.
944 * src/sh64/sysv.S: Likewise.
946 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
948 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
949 args and userdata unused.
950 (closure_test_fn1): Mark cif and userdata unused.
951 (main): Remove unused res.
953 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
955 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
956 -O2, -O3, -Os and the warning flags -W -Wall.
957 * testsuite/libffi.special/special.exp: Likewise.
958 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
959 unused parameter unused for gcc or else do nothing.
960 * testsuite/libffi.special/ffitestcxx.h: Likewise.
961 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
963 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
964 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
965 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
966 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
967 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
968 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
969 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
970 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
971 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
972 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
973 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
974 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
975 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
976 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
977 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
978 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
979 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
980 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
981 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
982 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
983 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
984 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
985 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
986 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
988 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
990 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
992 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
993 void* to avoid compiler warning.
995 (cls_struct_align_gn): Mark cif and userdata unused.
996 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
998 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1000 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1002 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1004 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1006 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1007 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1008 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1010 (main): Cast res_call to silence gcc.
1011 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1013 (main): Cast res_call to silence gcc.
1014 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1016 (main): Cast res_call to silence gcc.
1017 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1019 (main): Cast res_call to silence gcc.
1020 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1022 (main): Cast res_call to silence gcc.
1023 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1025 (main): Cast res_call to silence gcc.
1026 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1028 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1029 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1031 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1032 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1034 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1035 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1037 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1038 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1040 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1041 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1042 and userdata unused.
1043 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1045 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1046 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1047 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1048 Cleanup white spaces.
1049 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1050 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1051 cif and userdata unused.
1052 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1054 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1055 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1056 formatters to silence gcc.
1057 (B_gn): Mark cif and userdata unused.
1058 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1060 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1061 (B_gn): Mark cif and userdata unused.
1062 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1064 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1065 (B_gn): Mark cif and userdata unused.
1066 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1068 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1069 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1070 * testsuite/libffi.call/problem1.c (stub): Likewise.
1071 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1073 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1074 in the last commit for this test case in the test case itself.
1075 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1077 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1078 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1079 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1080 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1081 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1082 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1084 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1086 * src/sh/sysv.S: Fix register numbers in the FDE for
1089 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1091 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1092 declaration to avoid a false negative on ix86. See PR323.
1094 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1096 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1097 and cast integer to void * if needed. Update the pointer to
1098 the FP register saved area correctly.
1100 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1102 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1104 * testsuite/libffi.call/nested_struct4.c: Likewise.
1106 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1108 * testsuite/libffi.call/return_dbl.c: New test case.
1109 * testsuite/libffi.call/return_dbl1.c: Likewise.
1110 * testsuite/libffi.call/return_dbl2.c: Likewise.
1111 * testsuite/libffi.call/return_fl.c: Likewise.
1112 * testsuite/libffi.call/return_fl1.c: Likewise.
1113 * testsuite/libffi.call/return_fl2.c: Likewise.
1114 * testsuite/libffi.call/return_fl3.c: Likewise.
1115 * testsuite/libffi.call/closure_fn6.c: Likewise.
1117 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1119 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1120 here to be used by other test cases too.
1122 * testsuite/libffi.call/nested_struct10.c: New test case.
1123 * testsuite/libffi.call/nested_struct9.c: Likewise.
1124 * testsuite/libffi.call/nested_struct8.c: Likewise.
1125 * testsuite/libffi.call/nested_struct7.c: Likewise.
1126 * testsuite/libffi.call/nested_struct6.c: Likewise.
1127 * testsuite/libffi.call/nested_struct5.c: Likewise.
1128 * testsuite/libffi.call/nested_struct4.c: Likewise.
1130 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1132 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1133 * configure: Rebuilt.
1135 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1137 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1138 instead do the shifting inline.
1139 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1140 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1141 and 8 byte structs, for the remaining struct sizes don't call
1142 __lshrdi3, instead do the shifting inline.
1144 2005-12-07 Thiemo Seufer <ths@networkno.de>
1146 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1147 missing parentheses.
1148 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1149 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1150 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1151 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1152 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1154 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1156 (ffi_prep_closure): Improve trampoline instruction scheduling.
1157 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1159 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1161 * src/powerpc/ffi.c: Formatting.
1162 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1163 (ffi_prep_args64): Likewise.
1165 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1167 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1168 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1170 2005-09-26 Tom Tromey <tromey@redhat.com>
1172 * testsuite/libffi.call/float1.c (value_type): New typedef.
1173 (CANARY): New define.
1174 (main): Check for result buffer overflow.
1175 * src/powerpc/linux64.S: Handle linux64 long double returns.
1176 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1177 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1179 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1182 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1184 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1186 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1188 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1189 (AH_BOTTOM): Add FFI_HIDDEN definition.
1190 * configure: Rebuilt.
1191 * fficonfig.h.in: Rebuilt.
1192 * src/powerpc/ffi.c (hidden): Remove.
1193 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1194 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1195 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1196 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1197 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1198 add FFI_HIDDEN to its prototype.
1199 (ffi_closure_SYSV_inner): New.
1200 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1201 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1203 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1206 * configure: Rebuilt.
1207 * configure.ac: Handle i*86-*-gnu*.
1209 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1211 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1212 DW_CFA_offset_extended_sf rather than
1213 DW_CFA_GNU_negative_offset_extended.
1214 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1216 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1218 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1220 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1221 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1222 partially on register.
1223 (ffi_closure_helper_SYSV): Likewise.
1224 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1226 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1228 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1230 * src/sh64/ffi.c (simple_type): Remove.
1231 (return_type): Handle small structures correctly.
1232 (ffi_prep_args): Likewise.
1233 (ffi_call): Likewise.
1234 (ffi_closure_helper_SYSV): Likewise.
1235 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1236 Emit position independent code if PIC and remove wrong datalabel
1237 prefixes from EH data.
1239 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1241 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1242 * Makefile.in: Regenerate.
1243 * include/Makefile.in: Likewise.
1244 * testsuite/Makefile.in: Likewise.
1245 * configure.ac: Add POWERPC_FREEBSD rules.
1246 * configure: Regenerate.
1247 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1248 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1249 * src/powerpc/ffi.c: Add flags to handle small structure returns
1251 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1253 (ffi_closure_helper_SYSV): Likewise.
1254 * src/powerpc/ppc_closure.S: Add return types for small structures.
1255 * src/powerpc/sysv.S: Add bits to handle small structures for
1258 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1260 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1261 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1262 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1264 2005-07-05 Randolph Chung <tausq@debian.org>
1266 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1267 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1268 structures. Kill compilation warnings.
1269 (ffi_closure_inner_LINUX): Print return values as hex in debug
1270 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1271 Properly handle 5-7 byte structure returns.
1272 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1273 (FFI_TYPE_SMALL_STRUCT2): Remove.
1274 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1275 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1276 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1277 (checksmst1, checksmst2): Remove.
1278 (checksmst3): Optimize handling of 3-byte struct returns.
1279 (checksmst567): Properly handle 5-7 byte struct returns.
1281 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1284 * src/mips/n32.S: Enforce PIC code.
1285 * src/mips/o32.S: Likewise.
1287 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1289 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1290 * configure: Regenerate.
1292 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1294 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1295 to call ffi_closure_helper_SYSV. Append @local instead.
1296 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1298 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1300 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1301 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1302 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1303 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1304 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1306 2005-05-09 Mike Stump <mrs@apple.com>
1308 * configure: Regenerate.
1310 2005-05-08 Richard Henderson <rth@redhat.com>
1313 * src/alpha/osf.S: Update unwind into to match code.
1315 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1316 Richard Henderson <rth@redhat.com>
1318 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1320 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1321 (ffi_prep_closure): Set carry bit if sse-used flag set.
1322 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1323 Only load sse registers if ssecount non-zero.
1324 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1326 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1328 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1329 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1330 * configure: Regenerate.
1332 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1334 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1335 have Tcl8.3-compatible intermediate variable.
1337 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1338 Hans-Peter Nilsson <hp@axis.com>
1340 * Makefile.am: Add CRIS support.
1341 * configure.ac: Likewise.
1342 * Makefile.in, configure, testsuite/Makefile.in,
1343 include/Makefile.in: Regenerate.
1344 * src/cris: New directory.
1345 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1346 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1348 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1349 \r?\n in output tests.
1351 2005-04-12 Mike Stump <mrs@apple.com>
1353 * configure: Regenerate.
1355 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1357 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1359 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1361 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1363 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1364 to 64 bits in ILP32 mode.
1365 Load 64 bits even for short data.
1367 2005-03-23 Mike Stump <mrs@apple.com>
1369 * src/powerpc/darwin.S: Update for -m64 multilib.
1370 * src/powerpc/darwin_closure.S: Likewise.
1372 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1374 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1375 Do not set tool_include_dir.
1376 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1378 * include/Makefile.am: Set gcc_version and toollibffidir.
1379 * include/Makefile.in: Regenerate.
1381 2005-02-22 Andrew Haley <aph@redhat.com>
1383 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1384 odd-numbered register pairs for 64-bit integer types.
1386 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1389 * testsuite/libffi.call/return_ll1.c: New test case.
1391 2005-02-11 Janis Johnson <janis187@us.ibm.com>
1393 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1394 * testsuite/libffi.call/float.c: Ditto.
1395 * testsuite/libffi.call/float2.c: Ditto.
1396 * testsuite/libffi.call/float3.c: Ditto.
1398 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1400 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1402 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1404 * testsuite/libffi.special/special.exp (cxx_options): Add
1407 2004-12-31 Richard Henderson <rth@redhat.com>
1409 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1410 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1411 offset parameters with a type parameter; deduce size and structure
1412 alignment. Update all users.
1414 2004-12-31 Richard Henderson <rth@redhat.com>
1416 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1417 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
1418 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
1419 into ffi_prep_closure.
1420 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1423 2004-12-27 Richard Henderson <rth@redhat.com>
1425 * src/x86/unix64.S: Fix typo in unwind info.
1427 2004-12-25 Richard Henderson <rth@redhat.com>
1429 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1430 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1431 (merge_classes): Check for it.
1433 (classify_argument): Pass byte_offset by value; perform all updates
1435 (examine_argument): Add classes argument; handle
1436 X86_64_COMPLEX_X87_CLASS.
1437 (ffi_prep_args): Merge into ...
1438 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1439 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1440 (ffi_fill_return_value): Remove.
1441 (ffi_prep_closure): Remove dead assert.
1442 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1443 Rewrite to use struct register_args instead of va_list. Create
1444 flags for handling structure returns.
1445 * src/x86/unix64.S: Remove dead strings.
1446 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1447 stack frame with ffi_call. Handle structure returns properly.
1448 (float2sse, floatfloat2sse, double2sse): Remove.
1449 (sse2float, sse2double, sse2floatfloat): Remove.
1450 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1451 to handle structure returns properly.
1453 2004-12-08 David Edelsohn <edelsohn@gnu.org>
1455 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1457 * Makefile.in: Regenerated.
1459 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
1461 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1462 * configure, aclocal.m4, Makefile.in: Regenerate.
1463 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1465 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1467 * configure: Regenerate for libtool change.
1469 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1471 * configure: Regenerate for libtool reversion.
1473 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1475 * configure: Regenerate for libtool change.
1477 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1479 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1481 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
1483 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1484 of jal. Use an absolute encoding for the frame information.
1486 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1488 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1489 * acinclude.m4: Delete logic for sincludes.
1490 * aclocal.m4, Makefile.in, configure: Regenerate.
1491 * include/Makefile: Likewise.
1492 * testsuite/Makefile: Likewise.
1494 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1496 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1497 on a 8-byte boundary.
1498 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1500 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1502 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1503 long long values. Round stack allocation to a multiple of 8 bytes
1504 for ATPCS compatibility.
1505 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1506 names. Handle returning long long types. Add Thumb and interworking
1507 support. Improve soft-float code.
1509 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1511 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1512 (libffi_exit): New function.
1513 (libffi_init): Build the testglue wrapper if needed.
1515 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1518 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1520 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1522 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1524 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1526 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1527 * testsuite/libffi.call/float3.c: New test case.
1529 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1531 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1532 the function returning a structure pointed with R2.
1533 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1534 the structure return value if T bit set. Emit position
1535 independent code and EH data if PIC.
1537 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1539 * Makefile.am: Add m32r support.
1540 * configure.ac: Likewise.
1541 * Makefile.in: Regenerate.
1542 * confiugre: Regenerate.
1543 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1544 (uint64, sint64, double, longdouble)
1545 * src/m32r: New directory.
1546 * src/m32r/ffi.c: New file.
1547 * src/m32r/sysv.S: Likewise.
1548 * src/m32r/ffitarget.h: Likewise.
1550 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1552 * testsuite/libffi.call/negint.c: New test case.
1554 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1557 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1558 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1559 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1562 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1564 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1565 * testsuite/libffi.call/promotion.c: Likewise.
1566 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1568 * testsuite/libffi.call/return_sc.c: Likewise.
1569 * testsuite/libffi.call/return_uc.c: Likewise.
1571 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1573 * src/powerpc/darwin.S: Fix comments and identation.
1574 * src/powerpc/darwin_closure.S: Likewise.
1576 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1578 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1579 (ffi_prep_args): Handle longdouble arguments.
1580 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1582 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1583 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1585 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1586 * src/types.c: Defined longdouble size and alignment for darwin.
1588 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1590 * src/powerpc/aix.S: Remove whitespaces.
1591 * src/powerpc/aix_closure.S: Likewise.
1592 * src/powerpc/asm.h: Likewise.
1593 * src/powerpc/ffi.c: Likewise.
1594 * src/powerpc/ffitarget.h: Likewise.
1595 * src/powerpc/linux64.S: Likewise.
1596 * src/powerpc/linux64_closure.S: Likewise.
1597 * src/powerpc/ppc_closure.S: Likewise.
1598 * src/powerpc/sysv.S: Likewise.
1600 2004-08-30 Anthony Green <green@redhat.com>
1602 * Makefile.am: Add frv support.
1603 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1604 * configure.ac: Read configure.host.
1605 * configure.in: Read configure.host.
1606 * configure.host: New file. frv-elf needs libgloss.
1607 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
1608 alignment. This is needed to frv and shouldn't harm the others.
1609 * include/ffi_common.h (ALIGN_DOWN): New macro.
1610 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1612 2004-08-24 David Daney <daney@avtrex.com>
1614 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1615 * testsuite/libffi.call/closure_fn1.c: Likewise.
1616 * testsuite/libffi.call/closure_fn2.c Likewise.
1617 * testsuite/libffi.call/closure_fn3.c: Likewise.
1618 * testsuite/libffi.call/closure_fn4.c: Likewise.
1619 * testsuite/libffi.call/closure_fn5.c: Likewise.
1620 * testsuite/libffi.call/cls_18byte.c: Likewise.
1621 * testsuite/libffi.call/cls_19byte.c: Likewise.
1622 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1623 * testsuite/libffi.call/cls_20byte.c: Likewise.
1624 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1625 * testsuite/libffi.call/cls_24byte.c: Likewise.
1626 * testsuite/libffi.call/cls_2byte.c: Likewise.
1627 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1628 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1629 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1630 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1631 * testsuite/libffi.call/cls_4byte.c: Likewise.
1632 * testsuite/libffi.call/cls_64byte.c: Likewise.
1633 * testsuite/libffi.call/cls_6byte.c: Likewise.
1634 * testsuite/libffi.call/cls_7byte.c: Likewise.
1635 * testsuite/libffi.call/cls_8byte.c: Likewise.
1636 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1637 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1638 * testsuite/libffi.call/cls_align_double.c: Likewise.
1639 * testsuite/libffi.call/cls_align_float.c: Likewise.
1640 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1641 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1642 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1643 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1644 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1645 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1646 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1647 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1648 * testsuite/libffi.call/cls_double.c: Likewise.
1649 * testsuite/libffi.call/cls_float.c: Likewise.
1650 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1651 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1652 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1653 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1654 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1655 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1656 * testsuite/libffi.call/cls_schar.c: Likewise.
1657 * testsuite/libffi.call/cls_sint.c: Likewise.
1658 * testsuite/libffi.call/cls_sshort.c: Likewise.
1659 * testsuite/libffi.call/cls_uchar.c: Likewise.
1660 * testsuite/libffi.call/cls_uint.c: Likewise.
1661 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1662 * testsuite/libffi.call/cls_ushort.c: Likewise.
1663 * testsuite/libffi.call/nested_struct.c: Likewise.
1664 * testsuite/libffi.call/nested_struct1.c: Likewise.
1665 * testsuite/libffi.call/nested_struct2.c: Likewise.
1666 * testsuite/libffi.call/nested_struct3.c: Likewise.
1667 * testsuite/libffi.call/problem1.c: Likewise.
1668 * testsuite/libffi.special/unwindtest.cc: Likewise.
1669 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1671 * testsuite/libffi.call/cls_16byte.c: Likewise.
1672 * testsuite/libffi.call/cls_5byte.c: Likewise.
1674 2004-08-23 David Daney <daney@avtrex.com>
1677 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1678 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1679 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1680 parameters and return types.
1681 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1682 (ffi_prep_closure): Ditto.
1683 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1684 alignment calculations.
1685 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1686 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1688 2004-08-14 Casey Marshall <csm@gnu.org>
1690 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1691 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1692 integer (O32 ABI only).
1693 (ffi_prep_closure): new function.
1694 (ffi_closure_mips_inner_O32): new function.
1695 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
1696 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1697 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1698 64 bit integers correctly.
1699 (ffi_closure_O32): new function.
1700 Added DWARF-2 unwind info for both functions.
1702 2004-08-10 Andrew Haley <aph@redhat.com>
1704 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1706 2004-08-01 Robert Millan <robertmh@gnu.org>
1708 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1709 * configure: Regenerate.
1711 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1713 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1714 and mmap() explicitly instead of relying on preset autoconf cache
1716 * aclocal.m4: Regenerate.
1717 * configure: Regenerate.
1719 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1721 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1722 (ffi_check_float_struct): Remove unused prototype.
1724 2004-06-30 Geoffrey Keating <geoffk@apple.com>
1726 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1727 character on Darwin, use '\n\t' instead.
1729 2004-06-26 Matthias Klose <doko@debian.org>
1731 * libtool-version: Fix typo in revision/age.
1733 2004-06-17 Matthias Klose <doko@debian.org>
1735 * libtool-version: New.
1736 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1737 * Makefile.in: Regenerate.
1739 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1741 * Makefile.am: Remove useless multilib rules.
1742 * Makefile.in: Regenerate.
1743 * aclocal.m4: Regenerate with automake 1.8.5.
1744 * configure.ac: Remove useless multilib configury.
1745 * configure: Regenerate.
1747 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1749 * .cvsignore: New file.
1751 2004-06-10 Jakub Jelinek <jakub@redhat.com>
1753 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1755 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1758 2004-06-06 Sean McNeil <sean@mcneil.com>
1760 * configure.ac: Add x86_64-*-freebsd* support.
1761 * configure: Regenerate.
1763 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
1766 * configure.ac: Test for existence of mmap and sys/mman.h before
1767 checking blacklist. Fix suggested by Jim Wilson.
1768 * configure: Regenerate.
1770 2004-04-26 Matt Austern <austern@apple.com>
1772 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1774 * src/powerpc/darwin_closure.S: Likewise.
1776 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1778 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1779 error. Reported by Thomas Heller <theller@python.net>.
1780 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1781 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1783 2004-03-20 Matthias Klose <doko@debian.org>
1785 * src/pa/linux.S: Fix typo.
1787 2004-03-19 Matthias Klose <doko@debian.org>
1789 * Makefile.am: Update.
1790 * Makefile.in: Regenerate.
1791 * src/pa/ffi.h.in: Remove.
1792 * src/pa/ffitarget.h: New file.
1794 2004-02-10 Randolph Chung <tausq@debian.org>
1796 * Makefile.am: Add PA support.
1797 * Makefile.in: Regenerate.
1798 * include/Makefile.in: Regenerate.
1799 * configure.ac: Add PA target.
1800 * configure: Regenerate.
1801 * src/pa/ffi.c: New file.
1802 * src/pa/ffi.h.in: Add PA support.
1803 * src/pa/linux.S: New file.
1804 * prep_cif.c: Add PA support.
1806 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
1808 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
1810 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
1811 with ecif->cif->flags.
1812 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
1814 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
1815 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
1816 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
1817 sc_retstruct2b): Add for 1 or 2-bytes struct case.
1819 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
1821 * configure.in: Rename file to ...
1822 * configure.ac: ... this.
1823 * fficonfig.h.in: Regenerate.
1824 * Makefile.in: Regenerate.
1825 * include/Makefile.in: Regenerate.
1826 * testsuite/Makefile.in: Regenerate.
1828 2004-03-12 Matt Austern <austern@apple.com>
1830 * src/powerpc/darwin.S: Fix EH information so it corresponds to
1831 changes in EH format resulting from addition of linkonce support.
1832 * src/powerpc/darwin_closure.S: Likewise.
1834 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
1835 Paolo Bonzini <bonzini@gnu.org>
1837 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
1838 Remove VPATH. Remove rules for object files. Remove multilib support.
1839 (AM_CCASFLAGS): Add.
1840 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
1841 (AC_PREREQ): Bump version to 2.59.
1842 (AC_INIT): Fill with version info and bug address.
1843 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
1844 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
1845 De-precious CC so that the right flags are passed down to multilibs.
1846 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
1847 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
1848 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
1849 * configure: Rebuilt.
1850 * aclocal.m4: Likewise.
1851 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
1852 * fficonfig.h.in: Likewise.
1854 2004-03-11 Andreas Schwab <schwab@suse.de>
1856 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
1857 arguments from fp registers only for the first 8 parameter slots.
1858 Don't convert a float parameter when passed in memory.
1860 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
1862 * configure: Regenerate for config/accross.m4 correction.
1864 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
1866 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
1867 ecif->cif->bytes to bytes.
1868 (ffi_prep_cif_machdep): Add braces around nested if statement.
1870 2004-02-09 Alan Modra <amodra@bigpond.net.au>
1872 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
1874 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
1875 (ffi_closure_helper_LINUX64): Fix typo.
1876 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
1878 * testsuite/libffi.call/float.c: Likewise.
1879 * testsuite/libffi.call/float2.c: Likewise.
1881 2004-02-08 Alan Modra <amodra@bigpond.net.au>
1883 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
1884 long double function return and long double arg handling.
1885 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
1886 Use "end_pfr" instead of "nf". Correct long double handling.
1888 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
1890 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
1891 space for long double return value. Adjust stack frame and offsets.
1892 Load f2 long double return.
1894 2004-02-07 Alan Modra <amodra@bigpond.net.au>
1896 * src/types.c: Use 16 byte long double for POWERPC64.
1898 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1900 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
1901 when the structure return address is passed in %o0.
1902 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
1903 (ffi_v9_layout_struct): Align the field following a nested structure
1904 on a word boundary. Use memmove instead of memcpy.
1905 (ffi_call): Update call to ffi_V9_return_struct.
1906 (ffi_prep_closure): Define 'ctx' only for V8.
1907 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
1908 and ffi_closure_sparc_inner_v9.
1909 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
1910 Always skip the structure return address. For structures and long
1911 doubles, copy the argument directly.
1912 (ffi_closure_sparc_inner_v9): Skip the structure return address only
1913 if required. Shift the maximum floating-point slot accordingly. For
1914 big structures, copy the argument directly; otherwise, left-justify the
1915 argument and call ffi_v9_layout_struct to lay out the structure on
1917 * src/sparc/v8.S: Undef STACKFRAME before defining it.
1918 (ffi_closure_v8): Pass the structure return address. Update call to
1919 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
1920 Skip the 'unimp' insn when returning long doubles and structures.
1921 * src/sparc/v9.S: Undef STACKFRAME before defining it.
1922 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
1923 FFI_TYPE_INT handling. Load structures both in integers and
1924 floating-point registers on return.
1925 * README: Update status of the SPARC port.
1927 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
1929 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
1931 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
1933 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
1935 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
1936 value as of type ffi_arg, not unsigned int.
1938 2004-01-21 Michael Ritzert <ritzert@t-online.de>
1940 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
1943 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
1945 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
1948 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1950 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
1953 2003-12-10 Richard Henderson <rth@redhat.com>
1955 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
1956 size_t instead of int.
1958 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
1960 * testsuite/libffi.call/many_win32.c: Include <float.h>.
1961 * testsuite/libffi.call/many_win32.c (main): Replace variable
1962 int i with unsigned long ul.
1964 * testsuite/libffi.call/cls_align_uint64.c: New test case.
1965 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1966 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1967 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1968 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1969 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1970 * testsuite/libffi.call/cls_align_float.c: Likewise.
1971 * testsuite/libffi.call/cls_align_double.c: Likewise.
1972 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1973 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1975 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
1978 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
1979 Align arguments to 32 bits.
1981 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
1984 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
1985 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1986 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1987 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1988 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1989 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1991 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
1993 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
1995 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
1996 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1998 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2000 * Makefile.in: Rebuilt.
2001 * configure: Likewise.
2002 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2005 2003-11-21 Andreas Jaeger <aj@suse.de>
2006 Andreas Tobler <a.tobler@schweiz.ch>
2008 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2009 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2010 * Makefile.in: Rebuilt.
2011 * aclocal.m4: Likewise.
2012 * configure: Likewise.
2013 * fficonfig.h.in: Likewise.
2014 * testsuite/lib/libffi-dg.exp: Add include dir.
2015 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2016 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2017 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2018 for ffi_closure if available.
2019 * testsuite/libffi.call/closure_fn1.c: Likewise.
2020 * testsuite/libffi.call/closure_fn2.c: Likewise.
2021 * testsuite/libffi.call/closure_fn3.c: Likewise.
2022 * testsuite/libffi.call/closure_fn4.c: Likewise.
2023 * testsuite/libffi.call/closure_fn5.c: Likewise.
2024 * testsuite/libffi.call/cls_12byte.c: Likewise.
2025 * testsuite/libffi.call/cls_16byte.c: Likewise.
2026 * testsuite/libffi.call/cls_18byte.c: Likewise.
2027 * testsuite/libffi.call/cls_19byte.c: Likewise.
2028 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2029 * testsuite/libffi.call/cls_20byte.c: Likewise.
2030 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2031 * testsuite/libffi.call/cls_24byte.c: Likewise.
2032 * testsuite/libffi.call/cls_2byte.c: Likewise.
2033 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2034 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2035 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2036 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2037 * testsuite/libffi.call/cls_4byte.c: Likewise.
2038 * testsuite/libffi.call/cls_5byte.c: Likewise.
2039 * testsuite/libffi.call/cls_64byte.c: Likewise.
2040 * testsuite/libffi.call/cls_6byte.c: Likewise.
2041 * testsuite/libffi.call/cls_7byte.c: Likewise.
2042 * testsuite/libffi.call/cls_8byte.c: Likewise.
2043 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2044 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2045 * testsuite/libffi.call/cls_double.c: Likewise.
2046 * testsuite/libffi.call/cls_float.c: Likewise.
2047 * testsuite/libffi.call/cls_schar.c: Likewise.
2048 * testsuite/libffi.call/cls_sint.c: Likewise.
2049 * testsuite/libffi.call/cls_sshort.c: Likewise.
2050 * testsuite/libffi.call/cls_uchar.c: Likewise.
2051 * testsuite/libffi.call/cls_uint.c: Likewise.
2052 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2053 * testsuite/libffi.call/cls_ushort.c: Likewise.
2054 * testsuite/libffi.call/nested_struct.c: Likewise.
2055 * testsuite/libffi.call/nested_struct1.c: Likewise.
2056 * testsuite/libffi.call/nested_struct2.c: Likewise.
2057 * testsuite/libffi.call/nested_struct3.c: Likewise.
2058 * testsuite/libffi.call/problem1.c: Likewise.
2059 * testsuite/libffi.special/unwindtest.cc: Likewise.
2061 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2063 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2065 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2067 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2068 Add -lgcc_s to additional flags.
2070 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2072 * configure.in, include/Makefile.am: PR libgcj/11147, install
2073 the ffitarget.h header file in a gcc versioned and target
2075 * configure: Regenerated.
2076 * Makefile.in, include/Makefile.in: Likewise.
2077 * testsuite/Makefile.in: Likewise.
2079 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2081 * testsuite/libffi.call/closure_fn0.c: Print result and check
2082 with dg-output to make debugging easier.
2083 * testsuite/libffi.call/closure_fn1.c: Likewise.
2084 * testsuite/libffi.call/closure_fn2.c: Likewise.
2085 * testsuite/libffi.call/closure_fn3.c: Likewise.
2086 * testsuite/libffi.call/closure_fn4.c: Likewise.
2087 * testsuite/libffi.call/closure_fn5.c: Likewise.
2088 * testsuite/libffi.call/cls_12byte.c: Likewise.
2089 * testsuite/libffi.call/cls_16byte.c: Likewise.
2090 * testsuite/libffi.call/cls_18byte.c: Likewise.
2091 * testsuite/libffi.call/cls_19byte.c: Likewise.
2092 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2093 * testsuite/libffi.call/cls_20byte.c: Likewise.
2094 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2095 * testsuite/libffi.call/cls_24byte.c: Likewise.
2096 * testsuite/libffi.call/cls_2byte.c: Likewise.
2097 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2098 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2099 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2100 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2101 * testsuite/libffi.call/cls_4byte.c: Likewise.
2102 * testsuite/libffi.call/cls_5byte.c: Likewise.
2103 * testsuite/libffi.call/cls_64byte.c: Likewise.
2104 * testsuite/libffi.call/cls_6byte.c: Likewise.
2105 * testsuite/libffi.call/cls_7byte.c: Likewise.
2106 * testsuite/libffi.call/cls_8byte.c: Likewise.
2107 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2108 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2109 * testsuite/libffi.call/cls_double.c: Likewise.
2110 * testsuite/libffi.call/cls_float.c: Likewise.
2111 * testsuite/libffi.call/cls_schar.c: Likewise.
2112 * testsuite/libffi.call/cls_sint.c: Likewise.
2113 * testsuite/libffi.call/cls_sshort.c: Likewise.
2114 * testsuite/libffi.call/cls_uchar.c: Likewise.
2115 * testsuite/libffi.call/cls_uint.c: Likewise.
2116 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2117 * testsuite/libffi.call/cls_ushort.c: Likewise.
2118 * testsuite/libffi.call/problem1.c: Likewise.
2120 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2123 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2125 * testsuite/libffi.call/cls_9byte2.c: New test case.
2126 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2127 * testsuite/libffi.call/cls_64byte.c: Likewise.
2128 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2129 * testsuite/libffi.call/cls_19byte.c: Likewise.
2130 * testsuite/libffi.call/cls_18byte.c: Likewise.
2131 * testsuite/libffi.call/closure_fn4.c: Likewise.
2132 * testsuite/libffi.call/closure_fn5.c: Likewise.
2133 * testsuite/libffi.call/cls_schar.c: Likewise.
2134 * testsuite/libffi.call/cls_sint.c: Likewise.
2135 * testsuite/libffi.call/cls_sshort.c: Likewise.
2136 * testsuite/libffi.call/nested_struct2.c: Likewise.
2137 * testsuite/libffi.call/nested_struct3.c: Likewise.
2139 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2141 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2142 * testsuite/libffi.call/cls_uchar.c: Likewise.
2143 * testsuite/libffi.call/cls_uint.c: Likewise.
2144 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2145 * testsuite/libffi.call/cls_ushort.c: Likewise.
2146 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2148 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2150 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2153 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2155 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2156 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2158 2003-10-22 David Daney <ddaney@avtrex.com>
2160 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2161 __attribute__((__mode__(__SI__))) and friends.
2163 2003-10-22 Andreas Schwab <schwab@suse.de>
2165 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2167 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2169 * configure.in: AC_LINK_FILES(ffitarget.h).
2170 * configure: Regenerate.
2171 * Makefile.in: Likewise.
2172 * include/Makefile.in: Likewise.
2173 * testsuite/Makefile.in: Likewise.
2174 * fficonfig.h.in: Likewise.
2176 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2177 Richard Henderson <rth@redhat.com>
2179 Avoid that ffi.h includes fficonfig.h.
2181 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2182 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2183 (TARGET_SRC_MIPS_SGI): Removed.
2184 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2185 (MIPS_SGI): Removed.
2186 (CLEANFILES): Removed.
2187 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2189 * acconfig.h: Removed.
2190 * configure.in: Compute sizeofs only for double and long double.
2191 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2192 into AC_DEFINE instead of using acconfig.h. Create
2193 include/ffitarget.h instead of include/fficonfig.h. Rename
2194 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2195 AC_DEFINE EH_FRAME_FLAGS.
2196 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2197 (hack_DATA): Add ffitarget.h.
2198 * include/ffi.h.in: Remove all system specific definitions.
2199 Declare raw API even if it is not installed, why bother?
2200 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2201 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2202 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2203 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2204 * include/ffi_common.h (bool): Do not define.
2205 (ffi_assert): Accept failed assertion.
2206 (ffi_type_test): Return void and accept file/line.
2207 (FFI_ASSERT): Pass stringized failed assertion.
2208 (FFI_ASSERT_AT): New macro.
2209 (FFI_ASSERT_VALID_TYPE): New macro.
2210 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2211 UINT64, SINT64): Define here with gcc's __attribute__ macro
2213 (FLOAT32, ALIGN): Define here instead of in ffi.h
2214 * include/ffi-mips.h: Removed. Its content moved to
2215 src/mips/ffitarget.h after separating assembly and C sections.
2216 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2217 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2218 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2219 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2220 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2221 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2222 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2223 (ffi_type_test): Rewritten.
2224 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2225 FFI_ASSERT_VALID_TYPE.
2226 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2227 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2228 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2229 src/s390/ffitarget.h, src/sh/ffitarget.h,
2230 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2231 src/x86/ffitarget.h: New files.
2232 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2233 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2234 src/powerpc/aix.S, src/powerpc/darwin.S,
2235 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2236 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2237 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2238 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2239 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2242 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2244 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2245 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2247 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2249 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2250 Used when FFI_DEBUG = 1.
2252 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2254 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2257 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2259 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2260 FFI_MIPS_O32 for O32 ABI.
2262 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2264 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2265 SPARC64. Cleanup whitespaces.
2267 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2269 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2270 strongarm, xscale. Cleanup whitespaces.
2271 * testsuite/libffi.call/closure_fn1.c: Likewise.
2272 * testsuite/libffi.call/closure_fn2.c: Likewise.
2273 * testsuite/libffi.call/closure_fn3.c: Likewise.
2274 * testsuite/libffi.call/cls_12byte.c: Likewise.
2275 * testsuite/libffi.call/cls_16byte.c: Likewise.
2276 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2277 * testsuite/libffi.call/cls_20byte.c: Likewise.
2278 * testsuite/libffi.call/cls_24byte.c: Likewise.
2279 * testsuite/libffi.call/cls_2byte.c: Likewise.
2280 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2281 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2282 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2283 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2284 * testsuite/libffi.call/cls_4byte.c: Likewise.
2285 * testsuite/libffi.call/cls_5byte.c: Likewise.
2286 * testsuite/libffi.call/cls_6byte.c: Likewise.
2287 * testsuite/libffi.call/cls_7byte.c: Likewise.
2288 * testsuite/libffi.call/cls_8byte.c: Likewise.
2289 * testsuite/libffi.call/cls_double.c: Likewise.
2290 * testsuite/libffi.call/cls_float.c: Likewise.
2291 * testsuite/libffi.call/cls_uchar.c: Likewise.
2292 * testsuite/libffi.call/cls_uint.c: Likewise.
2293 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2294 * testsuite/libffi.call/cls_ushort.c: Likewise.
2295 * testsuite/libffi.call/nested_struct.c: Likewise.
2296 * testsuite/libffi.call/nested_struct1.c: Likewise.
2297 * testsuite/libffi.call/problem1.c: Likewise.
2298 * testsuite/libffi.special/unwindtest.cc: Likewise.
2299 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2301 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2303 * src/powerpc/aix.S: Cleanup whitespaces.
2304 * src/powerpc/aix_closure.S: Likewise.
2306 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2308 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2309 * src/powerpc/darwin_closure.S: Likewise.
2310 * src/powerpc/ffi_darwin.c: Likewise.
2312 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2313 David Edelsohn <edelsohn@gnu.org>
2315 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2316 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2318 * src/powerpc/darwin_closure.S: Likewise.
2319 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2320 according to the Darwin/AIX ABI.
2321 (ffi_prep_cif_machdep): Likewise.
2322 (ffi_closure_helper_DARWIN): Likewise.
2323 Remove the outgoing parameter stack logic. Simplify the evaluation
2324 of the different CASE types.
2325 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2326 statement in the trampoline code.
2328 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2330 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2331 for the register size.
2332 (ffi_closure_helper_SYSV): Handle the structure return value
2334 (ffi_closure_helper_SYSV): Return the appropriate type when
2335 the registers are used for the structure return value.
2336 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2337 the 64-bit return value. Update copyright years.
2339 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2341 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2342 srcdir for ffi_mips.h.
2344 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2346 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2348 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2349 placement of float result.
2350 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2351 cast of "resp" for big-endian 64 bit machines.
2353 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2355 * src/types.c (double, longdouble): Merge identical SH and ARM
2356 typedefs, and add POWERPC64.
2357 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2358 struct split over gpr and rest.
2359 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2360 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2362 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2364 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2367 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2369 * configure: Regenerate.
2371 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2373 * Makefile.am: Remove build rules for ffitest.
2374 * Makefile.in: Rebuilt.
2376 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2378 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2379 about implicit declaration of abort().
2381 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2383 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2384 * Makefile.in: Rebuilt.
2385 * configure.in: Add dejagnu test framework.
2386 * configure: Rebuilt.
2388 * testsuite/Makefile.am: New file.
2389 * testsuite/Makefile.in: Built
2390 * testsuite/lib/libffi-dg.exp: New file.
2391 * testsuite/config/default.exp: Likewise.
2392 * testsuite/libffi.call/call.exp: Likewise.
2393 * testsuite/libffi.call/ffitest.h: Likewise.
2394 * testsuite/libffi.call/closure_fn0.c: Likewise.
2395 * testsuite/libffi.call/closure_fn1.c: Likewise.
2396 * testsuite/libffi.call/closure_fn2.c: Likewise.
2397 * testsuite/libffi.call/closure_fn3.c: Likewise.
2398 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2399 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2400 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2401 * testsuite/libffi.call/cls_2byte.c: Likewise.
2402 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2403 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2404 * testsuite/libffi.call/cls_4byte.c: Likewise.
2405 * testsuite/libffi.call/cls_5byte.c: Likewise.
2406 * testsuite/libffi.call/cls_6byte.c: Likewise.
2407 * testsuite/libffi.call/cls_7byte.c: Likewise.
2408 * testsuite/libffi.call/cls_8byte.c: Likewise.
2409 * testsuite/libffi.call/cls_12byte.c: Likewise.
2410 * testsuite/libffi.call/cls_16byte.c: Likewise.
2411 * testsuite/libffi.call/cls_20byte.c: Likewise.
2412 * testsuite/libffi.call/cls_24byte.c: Likewise.
2413 * testsuite/libffi.call/cls_double.c: Likewise.
2414 * testsuite/libffi.call/cls_float.c: Likewise.
2415 * testsuite/libffi.call/cls_uchar.c: Likewise.
2416 * testsuite/libffi.call/cls_uint.c: Likewise.
2417 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2418 * testsuite/libffi.call/cls_ushort.c: Likewise.
2419 * testsuite/libffi.call/float.c: Likewise.
2420 * testsuite/libffi.call/float1.c: Likewise.
2421 * testsuite/libffi.call/float2.c: Likewise.
2422 * testsuite/libffi.call/many.c: Likewise.
2423 * testsuite/libffi.call/many_win32.c: Likewise.
2424 * testsuite/libffi.call/nested_struct.c: Likewise.
2425 * testsuite/libffi.call/nested_struct1.c: Likewise.
2426 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2427 * testsuite/libffi.call/problem1.c: Likewise.
2428 * testsuite/libffi.call/promotion.c: Likewise.
2429 * testsuite/libffi.call/return_ll.c: Likewise.
2430 * testsuite/libffi.call/return_sc.c: Likewise.
2431 * testsuite/libffi.call/return_uc.c: Likewise.
2432 * testsuite/libffi.call/strlen.c: Likewise.
2433 * testsuite/libffi.call/strlen_win32.c: Likewise.
2434 * testsuite/libffi.call/struct1.c: Likewise.
2435 * testsuite/libffi.call/struct2.c: Likewise.
2436 * testsuite/libffi.call/struct3.c: Likewise.
2437 * testsuite/libffi.call/struct4.c: Likewise.
2438 * testsuite/libffi.call/struct5.c: Likewise.
2439 * testsuite/libffi.call/struct6.c: Likewise.
2440 * testsuite/libffi.call/struct7.c: Likewise.
2441 * testsuite/libffi.call/struct8.c: Likewise.
2442 * testsuite/libffi.call/struct9.c: Likewise.
2443 * testsuite/libffi.special/special.exp: New file.
2444 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2445 * testsuite/libffi.special/unwindtest.cc: Likewise.
2448 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2450 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2453 2003-08-02 Alan Modra <amodra@bigpond.net.au>
2455 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2457 (ffi_closure_helper_LINUX64): Likewise.
2458 * src/powerpc/linux64.S: Remove code writing to parm save area.
2459 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2460 address in lr from ffi_closure_helper_LINUX64 call to calculate
2461 table address. Optimize function tail.
2463 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2465 * src/sparc/ffi.c: Handle all floating point registers.
2466 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2468 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2470 * README: Note that libffi is not part of GCC. Update the project
2473 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2475 * src/powerpc/ppc_closure.S: Include ffi.h.
2477 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2479 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2480 Use C style comments.
2482 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2484 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2485 * Makefile.in: Regenerate.
2486 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2487 * configure: Regenerate.
2488 * include/ffi.h.in: Add SHmedia support.
2489 * src/sh64/ffi.c: New file.
2490 * src/sh64/sysv.S: New file.
2492 2003-05-16 Jakub Jelinek <jakub@redhat.com>
2494 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2495 should be read-only.
2496 * configure: Rebuilt.
2497 * fficonfig.h.in: Rebuilt.
2498 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2499 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2500 * src/powerpc/linux64.S: Likewise.
2501 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2502 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2503 if -fpic/-fPIC/-mrelocatable.
2504 * src/powerpc/powerpc_closure.S: Likewise.
2505 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2506 #write in .eh_frame flags.
2507 * src/sparc/v9.S: Likewise.
2508 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2509 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2510 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2512 2003-05-07 Jeff Sturm <jsturm@one-point.com>
2514 Fixes PR bootstrap/10656
2515 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2516 support for .register pseudo-op.
2517 * src/sparc/v8.S: Use it.
2518 * fficonfig.h.in: Rebuilt.
2519 * configure: Rebuilt.
2521 2003-04-18 Jakub Jelinek <jakub@redhat.com>
2523 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2524 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2525 Make it the default on POWERPC64.
2526 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2527 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2528 * configure: Rebuilt.
2529 * src/powerpc/ffi.c (hidden): Define.
2530 (ffi_prep_args_SYSV): Renamed from
2531 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2532 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2533 ASM_NEEDS_REGISTERS64): New.
2534 (ffi_prep_args64): New function.
2535 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2536 (ffi_call): Likewise.
2537 (ffi_prep_closure): Likewise.
2538 (flush_icache): Surround by #ifndef POWERPC64.
2539 (ffi_dblfl): New union type.
2540 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2541 (ffi_closure_helper_LINUX64): New function.
2542 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2544 * src/powerpc/sysv.S: Likewise.
2545 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2546 * src/powerpc/linux64.S: New file.
2547 * src/powerpc/linux64_closure.S: New file.
2548 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2549 src/powerpc/linux64_closure.S.
2550 (TARGET_SRC_POWERPC): Likewise.
2552 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2553 closure_test_fn3): Fix result printing on big-endian 64-bit
2555 (main): Print tst2_arg instead of uninitialized tst2_result.
2557 * src/ffitest.c (main): Hide what closure pointer really points to
2560 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
2562 * configure.in (arm-*-netbsdelf*): Add configuration.
2563 (configure): Regenerated.
2565 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
2567 * include/Makefile.in: Regenerate.
2569 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2571 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2573 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2574 Receive closure pointer through parameter, read args using
2575 __builtin_dwarf_cfa.
2576 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2578 2003-03-12 Andreas Schwab <schwab@suse.de>
2580 * configure.in: Avoid trailing /. in toolexeclibdir.
2581 * configure: Rebuilt.
2583 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2585 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
2587 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2589 * libffi/src/powerpc/darwin_closure.S:
2590 Fix alignement bug, allocate 8 bytes for the result.
2591 * libffi/src/powerpc/aix_closure.S:
2593 * libffi/src/powerpc/ffi_darwin.c:
2594 Update stackframe description for aix/darwin_closure.S.
2596 2003-02-06 Jakub Jelinek <jakub@redhat.com>
2598 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2601 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2602 Andreas Schwab <schwab@suse.de>
2604 * configure.in: Adjust command to source config-ml.in to account
2605 for changes to the libffi_basedir definition.
2606 (libffi_basedir): Remove ${srcdir} from value and include trailing
2609 * configure: Regenerate.
2611 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2613 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2615 2003-01-28 Andrew Haley <aph@redhat.com>
2617 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2618 * src/x86/ffi64.c (ffi_prep_closure): New.
2619 (ffi_closure_UNIX64_inner): New.
2620 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2622 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
2624 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2625 Remove USE_LIBDIR conditional.
2626 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2627 * Makefile.in, configure: Rebuilt.
2629 2003-01027 David Edelsohn <edelsohn@gnu.org>
2631 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2632 * Makefile.in: Regenerate.
2634 2003-01-22 Andrew Haley <aph@redhat.com>
2636 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2639 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2641 * src/powerpc/darwin.S: Add unwind info.
2642 * src/powerpc/darwin_closure.S: Likewise.
2644 2003-01-14 Andrew Haley <aph@redhat.com>
2646 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2647 (ffi_prep_cif_machdep): Likewise.
2648 * src/x86/unix64.S: Add unwind info.
2650 2003-01-14 Andreas Jaeger <aj@suse.de>
2652 * src/ffitest.c (main): Only use ffi_closures if those are
2655 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2657 * libffi/src/ffitest.c
2658 add closure testcases
2660 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2662 * libffi/src/powerpc/ffi.c
2663 fix alignment bug for float (4 byte aligned iso 8 byte)
2665 2003-01-09 Geoffrey Keating <geoffk@apple.com>
2667 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2668 * src/powerpc/darwin.S: Remove RCS version string.
2670 2003-01-03 Jeff Sturm <jsturm@one-point.com>
2672 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2673 * src/ffitest.c (main): Use static storage for closure.
2674 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2675 * src/sparc/v8.S (ffi_closure_v8): New.
2676 * src/sparc/v9.S (ffi_closure_v9): New.
2678 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2680 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2681 enumeration for X86_WIN32.
2682 * src/x86/win32.S: Added ffi_call_STDCALL function
2684 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2685 switch cases for recognising FFI_STDCALL and
2686 calling ffi_call_STDCALL if target is X86_WIN32.
2687 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2688 stdcall versions of the "my_strlen" and "many"
2689 test functions (for X86_WIN32).
2690 Added test cases to test stdcall invocation using
2693 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2695 * src/sh/sysv.S: Add DWARF2 unwind info.
2697 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2699 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2701 2002-11-26 Jim Wilson <wilson@redhat.com>
2703 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2705 2002-11-23 H.J. Lu <hjl@gnu.org>
2707 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2708 Include ../config/accross.m4.
2709 * aclocal.m4; Rebuild.
2710 * configure: Likewise.
2712 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2714 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2716 2002-11-11 DJ Delorie <dj@redhat.com>
2718 * configure.in: Look for common files in the right place.
2720 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2722 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2723 raw data as _Jv_word values, not ffi_raw.
2724 (ffi_java_ptrarray_to_raw): Likewise.
2725 (ffi_java_rvalue_to_raw): New function.
2726 (ffi_java_raw_call): Call it.
2727 (ffi_java_raw_to_rvalue): New function.
2728 (ffi_java_translate_args): Call it.
2729 * src/ffitest.c (closure_test_fn): Interpret return value
2730 as ffi_arg, not int.
2731 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2732 FFI_TYPE_POINTER case.
2733 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2734 values extended to word size.
2736 2002-10-02 Andreas Jaeger <aj@suse.de>
2738 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2740 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
2742 * include/ffi.h.in: Fix i386 win32 compilation.
2744 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2746 * configure.in: Add s390x-*-linux-* target.
2747 * configure: Regenerate.
2748 * include/ffi.h.in: Define S390X for s390x targets.
2749 (FFI_CLOSURES): Define for s390/s390x.
2750 (FFI_TRAMPOLINE_SIZE): Likewise.
2751 (FFI_NATIVE_RAW_API): Likewise.
2752 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2753 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
2754 * src/s390/ffi.c: Major rework of existing code. Add support for
2755 s390x targets. Add closure support.
2756 * src/s390/sysv.S: Likewise.
2758 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
2760 * src/arm/sysv.S: Fix typo.
2762 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
2764 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2765 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2766 (ffi_call_SYSV): Handle soft-float.
2768 2002-09-27 Bo Thorsen <bo@suse.de>
2770 * include/ffi.h.in: Fix multilib x86-64 support.
2772 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2774 * Makefile.am (all-multi): Fix multilib parallel build.
2776 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2778 * configure.in (sh[34]*-*-linux*): Add brackets.
2779 * configure: Regenerate.
2781 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2783 * Makefile.am: Add SH support.
2784 * Makefile.in: Regenerate.
2785 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2786 * configure: Regenerate.
2787 * include/ffi.h.in: Add SH support.
2788 * src/sh/ffi.c: New file.
2789 * src/sh/sysv.S: New file.
2790 * src/types.c: Add SH support.
2792 2002-07-16 Bo Thorsen <bo@suse.de>
2794 * src/x86/ffi64.c: New file that adds x86-64 support.
2795 * src/x86/unix64.S: New file that handles argument setup for
2797 * src/x86/sysv.S: Don't use this on x86-64.
2798 * src/x86/ffi.c: Don't use this on x86-64.
2800 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2802 * src/ffitest.c (struct6): New test that tests a special case in
2804 (struct7): Likewise.
2805 (struct8): Likewise.
2806 (struct9): Likewise.
2807 (closure_test_fn): Silence warning about this when it's not used.
2808 (main): Add the new tests.
2809 (main): Fix a couple of wrong casts and silence some compiler warnings.
2810 * include/ffi.h.in: Add x86-64 ABI definition.
2811 * fficonfig.h.in: Regenerate.
2812 * Makefile.am: Add x86-64 support.
2813 * configure.in: Likewise.
2814 * Makefile.in: Regenerate.
2815 * configure: Likewise.
2817 2002-06-24 Bo Thorsen <bo@suse.de>
2819 * src/types.c: Merge settings for similar architectures.
2820 Add x86-64 sizes and alignments.
2822 2002-06-23 Bo Thorsen <bo@suse.de>
2824 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
2825 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
2826 * src/mips/ffi.c (ffi_prep_args): Likewise.
2827 * src/m68k/ffi.c (ffi_prep_args): Likewise.
2829 2002-07-18 H.J. Lu (hjl@gnu.org)
2831 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
2832 (libffi_la_SOURCES): Support MIPS_LINUX.
2833 (libffi_convenience_la_SOURCES): Likewise.
2834 * Makefile.in: Regenerated.
2836 * configure.in (mips64*-*): Skip.
2837 (mips*-*-linux*): New.
2838 * configure: Regenerated.
2840 * src/mips/ffi.c: Include <sgidefs.h>.
2842 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
2844 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
2846 2002-05-27 Roger Sayle <roger@eyesopen.com>
2848 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
2850 2002-05-27 Bo Thorsen <bo@suse.de>
2852 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
2855 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
2857 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
2858 beginning of function (for older apple cc).
2860 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
2862 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
2863 script entry, and set LD to it when configuring multilibs.
2864 * configure: Rebuilt.
2866 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
2868 * configure.in (sparc64-*-netbsd*): Add target.
2869 (sparc-*-netbsdelf*): Likewise.
2870 * configure: Regenerate.
2872 2002-04-28 David S. Miller <davem@redhat.com>
2874 * configure.in, configure: Fix SPARC test in previous change.
2876 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
2878 * Makefile.am: Add Linux for S/390 support.
2879 * Makefile.in: Regenerate.
2880 * configure.in: Add Linux for S/390 support.
2881 * configure: Regenerate.
2882 * include/ffi.h.in: Add Linux for S/390 support.
2883 * src/s390/ffi.c: New file from libffi CVS tree.
2884 * src/s390/sysv.S: New file from libffi CVS tree.
2886 2002-04-28 Jakub Jelinek <jakub@redhat.com>
2888 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
2890 * src/sparc/v8.S: Use it.
2891 * src/sparc/v9.S: Likewise.
2892 * fficonfig.h.in: Rebuilt.
2893 * configure: Rebuilt.
2895 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
2897 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
2899 * src/ia64/unix.S: Add unwind information. Fix comments.
2900 Save sp in a way that's compatible with unwind info.
2901 (ffi_call_unix): Correctly restore sp in all cases.
2902 * src/ia64/ffi.c: Add, fix comments.
2904 2002-04-08 Jakub Jelinek <jakub@redhat.com>
2906 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
2908 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
2910 * configure.in (alpha*-*-netbsd*): Add target.
2911 * configure: Regenerate.
2913 2002-04-04 Jeff Sturm <jsturm@one-point.com>
2915 * src/sparc/v8.S: Add unwind info.
2916 * src/sparc/v9.S: Likewise.
2918 2002-03-30 Krister Walfridsson <cato@df.lth.se>
2920 * configure.in: Enable i*86-*-netbsdelf*.
2921 * configure: Rebuilt.
2923 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
2926 * src/mips/n32.s: Delete
2927 * src/mips/o32.s: Delete
2929 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
2931 * configure.in: Enable alpha*-*-freebsd*.
2932 * configure: Rebuilt.
2934 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2936 * Makefile.am: libfficonvenience -> libffi_convenience.
2937 * Makefile.in: Rebuilt.
2939 * Makefile.am: Define ffitest_OBJECTS.
2940 * Makefile.in: Rebuilt.
2942 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
2943 David Edelsohn <edelsohn@gnu.org>
2945 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
2946 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
2947 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
2948 * Makefile.in: Regenerate.
2949 * include/ffi.h.in: Add AIX and Darwin closure definitions.
2950 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
2951 (flush_icache, flush_range): New functions.
2952 (ffi_closure_helper_DARWIN): New function.
2953 * src/powerpc/aix_closure.S: New file.
2954 * src/powerpc/darwin_closure.S: New file.
2956 2002-02-24 Jeff Sturm <jsturm@one-point.com>
2958 * include/ffi.h.in: Add typedef for ffi_arg.
2959 * src/ffitest.c (main): Declare rint with ffi_arg.
2961 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
2963 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
2964 number of GPRs for floating-point arguments.
2966 2002-01-31 Anthony Green <green@redhat.com>
2968 * configure: Rebuilt.
2969 * configure.in: Replace CHECK_SIZEOF and endian tests with
2970 cross-compiler friendly macros.
2971 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
2974 2002-01-18 David Edelsohn <edelsohn@gnu.org>
2976 * src/powerpc/darwin.S (_ffi_call_AIX): New.
2977 * src/powerpc/aix.S (ffi_call_DARWIN): New.
2979 2002-01-17 David Edelsohn <edelsohn@gnu.org>
2981 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
2982 (TARGET_SRC_POWERPC_AIX): New.
2983 (POWERPC_AIX): New stanza.
2984 * Makefile.in: Regenerate.
2985 * configure.in: Add AIX case.
2986 * configure: Regenerate.
2987 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
2988 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
2989 size. Fix "long double" support.
2990 (ffi_call): Add FFI_AIX case.
2991 * src/powerpc/aix.S: New.
2993 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
2995 Implement Darwin PowerPC ABI.
2996 * configure.in: Handle powerpc-*-darwin*.
2997 * Makefile.am: Set source files for POWERPC_DARWIN.
2998 * configure: Rebuilt.
2999 * Makefile.in: Rebuilt.
3000 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3002 * src/powerpc/darwin.S: New file.
3003 * src/powerpc/ffi_darwin.c: New file.
3005 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3007 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3009 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3011 * src/x86/sysv.S: Avoid gas-only .balign directive.
3012 Use C style comments.
3014 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3016 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3017 Fixes PR bootstrap/3563.
3019 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3021 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3023 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3025 * configure.in: Recognize sparc*-sun-* host.
3026 * configure: Regenerate.
3028 2001-06-06 Andrew Haley <aph@redhat.com>
3030 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
3032 2001-06-03 Andrew Haley <aph@redhat.com>
3034 * src/alpha/osf.S: Add unwind info.
3035 * src/powerpc/sysv.S: Add unwind info.
3036 * src/powerpc/ppc_closure.S: Likewise.
3038 2000-05-31 Jeff Sturm <jsturm@one-point.com>
3040 * configure.in: Fix AC_ARG_ENABLE usage.
3041 * configure: Rebuilt.
3043 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3045 * configure.in: Remove warning about beta code.
3046 * configure: Rebuilt.
3048 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3050 * src/ia64/unix.S: Restore stack pointer when returning from
3052 * src/ia64/ffi.c: Fix typo in comment.
3054 2001-04-18 Jim Wilson <wilson@redhat.com>
3056 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3057 to eliminate RAW DV.
3059 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3061 * Makefile.am: Make a libtool convenience library.
3062 * Makefile.in: Rebuilt.
3064 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3066 * configure.in: Use different syntax for subdirectory creation.
3067 * configure: Rebuilt.
3069 2001-03-27 Jon Beniston <jon@beniston.com>
3071 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3072 * configure: Rebuilt.
3073 * Makefile.am: Added X86_WIN32 target support.
3074 * Makefile.in: Rebuilt.
3076 * include/ffi.h.in: Added X86_WIN32 target support.
3078 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3079 * src/types.c: Added X86_WIN32 target support.
3081 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3082 stuff removed and made to work with CygWin's gas.
3084 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3086 * configure.in: Make target subdirectory in build dir.
3087 * Makefile.am: Override suffix based rules to specify correct output
3089 * Makefile.in: Rebuilt.
3090 * configure: Rebuilt.
3092 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3094 * src/powerpc/ppc_closure.S: New file.
3095 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3096 involving long long and register pairs.
3097 (ffi_prep_closure): New function.
3098 (flush_icache): Likewise.
3099 (ffi_closure_helper_SYSV): Likewise.
3100 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3101 (FFI_TRAMPOLINE_SIZE): Likewise.
3102 (FFI_NATIVE_RAW_API): Likewise.
3103 * Makefile.in: Rebuilt.
3104 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3105 (TARGET_SRC_POWERPC): Likewise.
3107 2001-03-19 Tom Tromey <tromey@redhat.com>
3109 * Makefile.in: Rebuilt.
3110 * Makefile.am (ffitest_LDFLAGS): New macro.
3112 2001-03-02 Nick Clifton <nickc@redhat.com>
3114 * include/ffi.h.in: Remove RCS ident string.
3115 * include/ffi_mips.h: Remove RCS ident string.
3116 * src/debug.c: Remove RCS ident string.
3117 * src/ffitest.c: Remove RCS ident string.
3118 * src/prep_cif.c: Remove RCS ident string.
3119 * src/types.c: Remove RCS ident string.
3120 * src/alpha/ffi.c: Remove RCS ident string.
3121 * src/alpha/osf.S: Remove RCS ident string.
3122 * src/arm/ffi.c: Remove RCS ident string.
3123 * src/arm/sysv.S: Remove RCS ident string.
3124 * src/mips/ffi.c: Remove RCS ident string.
3125 * src/mips/n32.S: Remove RCS ident string.
3126 * src/mips/o32.S: Remove RCS ident string.
3127 * src/sparc/ffi.c: Remove RCS ident string.
3128 * src/sparc/v8.S: Remove RCS ident string.
3129 * src/sparc/v9.S: Remove RCS ident string.
3130 * src/x86/ffi.c: Remove RCS ident string.
3131 * src/x86/sysv.S: Remove RCS ident string.
3133 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3135 * include/ffi.h.in: Change sourceware.cygnus.com references to
3138 2000-12-09 Richard Henderson <rth@redhat.com>
3140 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3141 (ffi_closure_osf_inner): Index rather than increment avalue
3142 and arg_types. Give ffi_closure_osf the raw return value type.
3143 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3146 2000-12-07 Richard Henderson <rth@redhat.com>
3148 * src/raw_api.c (ffi_translate_args): Fix typo.
3149 (ffi_prep_closure): Likewise.
3151 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3152 FFI_TRAMPOLINE_SIZE.
3153 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3154 cif->bytes for new ffi_call_osf implementation.
3155 (ffi_prep_args): Absorb into ...
3156 (ffi_call): ... here. Do all stack allocation here and
3157 avoid a callback function.
3158 (ffi_prep_closure, ffi_closure_osf_inner): New.
3159 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3160 (ffi_closure_osf): New.
3162 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3164 * config.guess, config.sub, install-sh: Removed.
3165 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3166 * Makefile.in: Rebuilt.
3168 * acinclude.m4: Include libtool macros from the top level.
3169 * aclocal.m4, configure: Rebuilt.
3171 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3173 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3174 * configure: Rebuilt.
3176 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3178 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3179 memory correctly. Use conditional instructions, not branches where
3182 2000-05-04 Tom Tromey <tromey@cygnus.com>
3184 * configure: Rebuilt.
3185 * configure.in: Match `arm*-*-linux-*'.
3186 From Chris Dornan <cdornan@arm.com>.
3188 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3190 * Makefile.am (SUBDIRS): Define.
3191 (AM_MAKEFLAGS): Likewise.
3192 (Multilib support.): Add section.
3193 * Makefile.in: Rebuilt.
3194 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3195 New variables. Set for gcc using -print-multi-lib. Export them
3197 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3198 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3200 (extra_compiler_flags_value, extra_compiler_flags): Check these
3201 for extra compiler options which need to be passed down in
3204 2000-04-16 Anthony Green <green@redhat.com>
3206 * configure: Rebuilt.
3207 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3209 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3211 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3212 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3213 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3214 Replace all void * sizeofs with sizeof(int).
3215 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3216 different than DOUBLE.
3217 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3218 (ffi_prep_args_v9): New function.
3219 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3220 (ffi_V9_return_struct): New function.
3221 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3222 32bit code (not yet cross-arch calls).
3223 * src/sparc/v8.S: Add struct return delay nop.
3225 * src/sparc/v9.S: New file.
3226 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3227 is used on sparc64 only for structures larger than 32 bytes.
3228 Pass by reference for structures is done for structure arguments
3229 larger than 16 bytes.
3230 * src/ffitest.c (main): Use 64bit rint on sparc64.
3231 Run long long tests on sparc.
3232 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3234 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3236 * configure.in (sparc-*-linux*): New supported target.
3237 (sparc64-*-linux*): Likewise.
3238 * configure: Rebuilt.
3239 * Makefile.am: Add v9.S to SPARC files.
3240 * Makefile.in: Likewise.
3241 (LINK): Surround $(CCLD) into double quotes, so that multilib
3242 compiles work correctly.
3244 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3246 * configure: Rebuilt.
3247 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3248 proposed by Bryce McKinlay.
3250 2000-03-20 Tom Tromey <tromey@cygnus.com>
3252 * Makefile.in: Hand edit for java_raw_api.lo.
3254 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3256 * config.guess, config.sub: Update from the gcc tree.
3257 Fix for PR libgcj/168.
3259 2000-03-03 Tom Tromey <tromey@cygnus.com>
3261 * Makefile.in: Fixed ia64 by hand.
3263 * configure: Rebuilt.
3264 * configure.in (--enable-multilib): New option.
3265 (libffi_basedir): New subst.
3266 (AC_OUTPUT): Added multilib code.
3268 2000-03-02 Tom Tromey <tromey@cygnus.com>
3270 * Makefile.in: Rebuilt.
3271 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3274 2000-02-25 Hans Boehm <boehm@acm.org>
3276 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3278 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3280 (ffi_prep_raw_closure): Use ffi_translate_args, not
3281 ffi_closure_translate.
3282 * src/java_raw_api.c: New file.
3283 * src/ffitest.c (closure_test_fn): New function.
3284 (main): Define `rint' as long long on IA64. Added new test when
3285 FFI_CLOSURES is defined.
3286 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3287 (ffi_abi): Recognize IA64.
3288 (ffi_raw): Added `flt' field.
3289 Added "Java raw API" code.
3290 * configure.in: Recognize ia64.
3291 * Makefile.am (TARGET_SRC_IA64): New macro.
3292 (libffi_la_common_SOURCES): Added java_raw_api.c.
3293 (libffi_la_SOURCES): Define in IA64 case.
3295 2000-01-04 Tom Tromey <tromey@cygnus.com>
3297 * Makefile.in: Rebuilt with newer automake.
3299 1999-12-31 Tom Tromey <tromey@cygnus.com>
3301 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3303 1999-09-01 Tom Tromey <tromey@cygnus.com>
3305 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3307 * fficonfig.h.in: Rebuilt.
3308 * configure: Rebuilt.
3309 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3310 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3311 * acconfig.h: Don't #undef PACKAGE or VERSION.
3313 1999-08-09 Anthony Green <green@cygnus.com>
3315 * include/ffi.h.in: Try to work around messy header problem
3316 with PACKAGE and VERSION.
3318 * configure: Rebuilt.
3319 * configure.in: Change version to 2.00-beta.
3321 * fficonfig.h.in: Rebuilt.
3322 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3324 * src/x86/ffi.c (ffi_raw_call): Rename.
3326 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3328 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3329 (ffi_prep_incoming_args_SYSV): Ditto.
3330 (ffi_prep_closure): Ditto.
3331 (ffi_closure_raw_SYSV): Ditto.
3332 (ffi_prep_raw_closure): More ditto.
3333 (ffi_call_raw): Final ditto.
3335 * include/ffi.h.in: Add definitions for closure and raw API.
3337 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3340 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3342 * src/raw_api.c: New file.
3344 * include/ffi.h.in (ffi_raw): New type.
3345 (UINT_ARG, SINT_ARG): New defines.
3346 (ffi_closure, ffi_raw_closure): New types.
3347 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3349 * configure.in: Add check for endianness and sizeof void*.
3351 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3352 instead of directly.
3354 * configure: Rebuilt.
3356 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3358 * configure.in: Add x86 and powerpc BeOS configurations.
3359 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3361 1999-05-09 Anthony Green <green@cygnus.com>
3363 * configure.in: Add warning about this being beta code.
3364 Remove src/Makefile.am from the picture.
3365 * configure: Rebuilt.
3367 * Makefile.am: Move logic from src/Makefile.am. Add changes
3368 to support libffi as a target library.
3369 * Makefile.in: Rebuilt.
3371 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3372 Upgraded to new autoconf, automake, libtool.
3376 * LICENSE: Update copyright date.
3378 * src/Makefile.am, src/Makefile.in: Removed.
3380 1998-11-29 Anthony Green <green@cygnus.com>
3382 * include/ChangeLog: Removed.
3383 * src/ChangeLog: Removed.
3384 * src/mips/ChangeLog: Removed.
3385 * src/sparc/ChangeLog: Remboved.
3386 * src/x86/ChangeLog: Removed.
3388 * ChangeLog.v1: Created.