1 2009-04-21 Release Manager
5 2009-04-09 Jakub Jelinek <jakub@redhat.com>
7 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
8 version 3 of the GNU General Public License and to point readers
9 at the COPYING3 file and the FSF's license web page.
10 * testsuite/libffi.call/call.exp: Likewise.
11 * testsuite/libffi.special/special.exp: Likewise.
13 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
15 * configure: Regenerate.
17 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
20 * configure.ac (HAVE_AS_X86_PCREL): New test.
21 * configure: Regenerate.
22 * fficonfig.h.in: Regenerate.
23 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
24 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
25 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
26 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
27 * src/x86/unix64.S (.Lstore_table): Move to .text section.
28 (.Lload_table): Likewise.
29 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
31 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
33 * configure: Regenerate.
35 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
37 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
38 signed/unsigned int8/16 return values.
39 * src/sparc/v8.S (ffi_call_v8): Likewise.
40 (ffi_closure_v8): Likewise.
42 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
43 Steve Ellcey <sje@cup.hp.com>
45 * configure: Regenerate for new libtool.
47 * include/Makefile.in: Ditto.
50 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
52 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
53 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
54 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
55 Adjust copyright notice.
56 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
57 register or two register to use for FFI_SYSV structs.
58 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
59 (ffi_closure_helper_SYSV): Fix the return type for
60 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
61 Adjust copyright notice.
63 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
65 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
68 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
70 * configure: Regenerate.
71 * include/Makefile.in: Regenerate.
72 * testsuite/Makefile.in: Regenerate.
74 2008-06-07 Joseph Myers <joseph@codesourcery.com>
76 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
77 powerpc-*-beos*): Remove.
78 * configure: Regenerate.
80 2008-05-09 Julian Brown <julian@codesourcery.com>
82 * Makefile.am (LTLDFLAGS): New.
83 (libffi_la_LDFLAGS): Use above.
84 * Makefile.in: Regenerate.
86 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
89 * aclocal.m4: Regenerate.
90 * configure: Regenerate.
92 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
94 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
95 * src/sh64/sysv.S: Likewise.
97 2008-03-26 Daniel Jacobowitz <dan@debian.org>
99 * src/arm/sysv.S: Fix ARM comment marker.
101 2008-03-26 Jakub Jelinek <jakub@redhat.com>
103 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
104 * src/s390/sysv.S: Likewise.
105 * src/powerpc/ppc_closure.S: Likewise.
106 * src/powerpc/sysv.S: Likewise.
107 * src/x86/unix64.S: Likewise.
108 * src/x86/sysv.S: Likewise.
109 * src/sparc/v8.S: Likewise.
110 * src/sparc/v9.S: Likewise.
111 * src/m68k/sysv.S: Likewise.
112 * src/arm/sysv.S: Likewise.
114 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
116 * aclocal.m4: Regenerate.
117 * configure: Likewise.
118 * Makefile.in: Likewise.
119 * include/Makefile.in: Likewise.
120 * testsuite/Makefile.in: Likewise.
122 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
123 Andreas Tobler <a.tobler@schweiz.org>
125 * configure.ac: Add amd64-*-freebsd* target.
126 * configure: Regenerate.
128 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
131 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
134 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
137 2008-01-24 David Edelsohn <edelsohn@gnu.org>
139 * configure: Regenerate.
141 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
143 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
145 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
148 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
149 signed/unsigned int8/16 for X86_DARWIN.
150 Updated copyright info.
151 Handle one and two byte structs with special cif->flags.
152 * src/x86/ffitarget.h: Add special types for one and two byte structs.
153 Updated copyright info.
154 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
156 Remove code to pop args from the stack after call.
157 Special-case signed/unsigned for int8/16, one and two byte structs.
158 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
159 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
161 Updated copyright info.
163 2007-12-08 David Daney <ddaney@avtrex.com>
165 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
166 SUBU, add with ADDU and use smaller code sequences.
168 2007-12-07 David Daney <ddaney@avtrex.com>
170 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
173 2007-12-06 David Daney <ddaney@avtrex.com>
175 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
177 (ffi_java_raw): New typedef.
178 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
179 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
181 (ffi_java_raw_closure) : Same.
182 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
184 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
186 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
187 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
188 sizeof(ffi_java_raw) for alignment calculations.
189 (ffi_java_ptrarray_to_raw): Same.
190 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
191 if FFI_SIZEOF_JAVA_RAW == 4.
192 (ffi_java_raw_to_rvalue): Same.
193 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
194 (ffi_java_translate_args): Same.
195 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
197 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
199 2007-12-06 David Daney <ddaney@avtrex.com>
201 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
204 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
207 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
208 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
209 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
210 set the NUM_FPR_ARG_REGISTERS according to.
211 Add support for potential soft-float support under hard-float
213 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
214 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
215 to the FFI_LINUX_SOFT_FLOAT ABI.
216 (ffi_prep_cif_machdep): Likewise.
217 (ffi_closure_helper_SYSV): Likewise.
218 * src/powerpc/ppc_closure.S: Make sure not to store float/double
219 on archs where __NO_FPRS__ is true.
220 Add FFI_TYPE_UINT128 support.
221 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
222 Adjust copyright notice.
224 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
226 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
227 * include/ffi_common.h: ... here.
230 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
232 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
233 * src/powerpc/linux64.S: Likewise.
234 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
235 * testsuite/libffi.call/return_ldl.c: New test.
237 2007-09-04 <aph@redhat.com>
239 * src/arm/sysv.S (UNWIND): New.
240 (Whole file): Conditionally compile unwinder directives.
241 * src/arm/sysv.S: Add unwinder directives.
243 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
244 Only treat r0 as a struct address if we're actually returning a
246 Only copy the bytes that are actually within a struct.
247 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
248 is returned in r0, not passed by address.
249 (ffi_call): Allocate a word-sized temporary for the case where
250 a composite is returned in r0.
251 (ffi_prep_incoming_args_SYSV): Align as necessary.
253 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
255 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
256 directly using the sys_cacheflush syscall.
258 2007-07-27 Andrew Haley <aph@redhat.com>
260 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
262 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
264 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
265 * configure.ac: Likewise.
266 * Makefile.in: Regenerate.
267 * include/Makefile.in: Likewise.
268 * testsuite/Makefile.in: Likewise.
269 * configure: Likewise.
271 2007-08-24 David Daney <ddaney@avtrex.com>
273 * testsuite/libffi.call/return_sl.c: New test.
275 2007-08-10 David Daney <ddaney@avtrex.com>
277 * testsuite/libffi.call/cls_multi_ushort.c,
278 testsuite/libffi.call/cls_align_uint16.c,
279 testsuite/libffi.call/nested_struct1.c,
280 testsuite/libffi.call/nested_struct3.c,
281 testsuite/libffi.call/cls_7_1_byte.c,
282 testsuite/libffi.call/nested_struct5.c,
283 testsuite/libffi.call/cls_double.c,
284 testsuite/libffi.call/nested_struct7.c,
285 testsuite/libffi.call/cls_sint.c,
286 testsuite/libffi.call/nested_struct9.c,
287 testsuite/libffi.call/cls_20byte1.c,
288 testsuite/libffi.call/cls_multi_sshortchar.c,
289 testsuite/libffi.call/cls_align_sint64.c,
290 testsuite/libffi.call/cls_3byte2.c,
291 testsuite/libffi.call/cls_multi_schar.c,
292 testsuite/libffi.call/cls_multi_uchar.c,
293 testsuite/libffi.call/cls_19byte.c,
294 testsuite/libffi.call/cls_9byte1.c,
295 testsuite/libffi.call/cls_align_float.c,
296 testsuite/libffi.call/closure_fn1.c,
297 testsuite/libffi.call/problem1.c,
298 testsuite/libffi.call/closure_fn3.c,
299 testsuite/libffi.call/cls_sshort.c,
300 testsuite/libffi.call/closure_fn5.c,
301 testsuite/libffi.call/cls_align_double.c,
302 testsuite/libffi.call/nested_struct.c,
303 testsuite/libffi.call/cls_2byte.c,
304 testsuite/libffi.call/nested_struct10.c,
305 testsuite/libffi.call/cls_4byte.c,
306 testsuite/libffi.call/cls_6byte.c,
307 testsuite/libffi.call/cls_8byte.c,
308 testsuite/libffi.call/cls_multi_sshort.c,
309 testsuite/libffi.call/cls_align_sint16.c,
310 testsuite/libffi.call/cls_align_uint32.c,
311 testsuite/libffi.call/cls_20byte.c,
312 testsuite/libffi.call/cls_float.c,
313 testsuite/libffi.call/nested_struct2.c,
314 testsuite/libffi.call/cls_5_1_byte.c,
315 testsuite/libffi.call/nested_struct4.c,
316 testsuite/libffi.call/cls_24byte.c,
317 testsuite/libffi.call/nested_struct6.c,
318 testsuite/libffi.call/cls_64byte.c,
319 testsuite/libffi.call/nested_struct8.c,
320 testsuite/libffi.call/cls_uint.c,
321 testsuite/libffi.call/cls_multi_ushortchar.c,
322 testsuite/libffi.call/cls_schar.c,
323 testsuite/libffi.call/cls_uchar.c,
324 testsuite/libffi.call/cls_align_uint64.c,
325 testsuite/libffi.call/cls_ulonglong.c,
326 testsuite/libffi.call/cls_align_longdouble.c,
327 testsuite/libffi.call/cls_1_1byte.c,
328 testsuite/libffi.call/cls_12byte.c,
329 testsuite/libffi.call/cls_3_1byte.c,
330 testsuite/libffi.call/cls_3byte1.c,
331 testsuite/libffi.call/cls_4_1byte.c,
332 testsuite/libffi.call/cls_6_1_byte.c,
333 testsuite/libffi.call/cls_16byte.c,
334 testsuite/libffi.call/cls_18byte.c,
335 testsuite/libffi.call/closure_fn0.c,
336 testsuite/libffi.call/cls_9byte2.c,
337 testsuite/libffi.call/closure_fn2.c,
338 testsuite/libffi.call/closure_fn4.c,
339 testsuite/libffi.call/cls_ushort.c,
340 testsuite/libffi.call/closure_fn6.c,
341 testsuite/libffi.call/cls_5byte.c,
342 testsuite/libffi.call/cls_align_pointer.c,
343 testsuite/libffi.call/cls_7byte.c,
344 testsuite/libffi.call/cls_align_sint32.c,
345 testsuite/libffi.special/unwindtest_ffi_call.cc,
346 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
348 2007-08-10 David Daney <ddaney@avtrex.com>
351 * configure.ac: Don't treat mips64 as a special case.
352 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
353 * configure: Regenerate
354 * Makefile.in: Ditto.
355 * fficonfig.h.in: Ditto.
356 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
357 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
358 (FFI_DEFAULT_ABI): Set for n64 case.
359 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
360 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
361 (ffi_closure_N32): New function.
362 (.eh_frame): New section
363 * src/mips/o32.S: Clean up comments.
364 (ffi_closure_O32): Pass ffi_closure parameter in $12.
365 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
366 _MIPS_SIM == _ABIN32 throughout.
367 (FFI_MIPS_STOP_HERE): New, use in place of
369 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
370 to support n32/n64 ABIs.
371 (calc_n32_struct_flags): Rewrite.
372 (calc_n32_return_struct_flags): Remove unused variable. Reverse
373 position of flag bits.
374 (ffi_prep_cif_machdep): Rewrite n32 portion.
375 (ffi_call): Enable for n64. Add special handling for small structure
377 (ffi_prep_closure_loc): Add n32 and n64 support.
378 (ffi_closure_mips_inner_O32): Add cast to silence warning.
379 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
381 2007-08-08 David Daney <ddaney@avtrex.com>
383 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
384 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
386 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
387 * testsuite/libffi.call/cls_sint.c (main): Ditto.
388 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
389 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
390 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
391 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
392 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
393 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
394 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
395 * testsuite/libffi.call/return_fl3.c (main): Ditto.
396 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
397 * testsuite/libffi.call/nested_struct.c (main): Ditto.
398 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
399 * testsuite/libffi.call/return_ll1.c (main): Ditto.
400 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
401 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
402 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
403 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
404 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
405 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
406 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
407 * testsuite/libffi.call/cls_uint.c (main): Ditto.
408 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
409 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
410 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
411 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
412 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
413 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
414 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
415 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
416 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
418 2007-08-07 Andrew Haley <aph@redhat.com>
420 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
423 2007-08-06 Andrew Haley <aph@redhat.com>
426 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
427 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
430 2007-08-02 David Daney <ddaney@avtrex.com>
432 * testsuite/libffi.call/return_ul.c (main): Define return type as
433 ffi_arg. Use proper printf conversion specifier.
435 2007-07-30 Andrew Haley <aph@redhat.com>
438 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
439 signed/unsigned int8/16.
440 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
442 Remove code to pop args from the stack after call.
443 Special-case signed/unsigned int8/16.
444 * testsuite/libffi.call/return_sc.c (main): Revert.
446 2007-07-26 Richard Guenther <rguenther@suse.de>
449 * testsuite/libffi.call/return_sc.c (main): Verify call
450 result as signed char, not ffi_arg.
452 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
454 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
455 * configure: Regenerate.
457 2007-07-11 David Daney <ddaney@avtrex.com>
459 * src/mips/ffi.c: Don't include sys/cachectl.h.
460 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
463 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
465 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
466 from (ffi_prep_closure): ... this.
467 (FFI_INIT_TRAMPOLINE): Adjust.
469 2005-12-31 Phil Blundell <pb@reciva.com>
471 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
472 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
473 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
474 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
475 (FFI_CLOSURES): Enable closure support.
477 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
479 * testsuite/libffi.call/cls_multi_ushort.c,
480 testsuite/libffi.call/cls_align_uint16.c,
481 testsuite/libffi.call/nested_struct1.c,
482 testsuite/libffi.call/nested_struct3.c,
483 testsuite/libffi.call/cls_7_1_byte.c,
484 testsuite/libffi.call/cls_double.c,
485 testsuite/libffi.call/nested_struct5.c,
486 testsuite/libffi.call/nested_struct7.c,
487 testsuite/libffi.call/cls_sint.c,
488 testsuite/libffi.call/nested_struct9.c,
489 testsuite/libffi.call/cls_20byte1.c,
490 testsuite/libffi.call/cls_multi_sshortchar.c,
491 testsuite/libffi.call/cls_align_sint64.c,
492 testsuite/libffi.call/cls_3byte2.c,
493 testsuite/libffi.call/cls_multi_schar.c,
494 testsuite/libffi.call/cls_multi_uchar.c,
495 testsuite/libffi.call/cls_19byte.c,
496 testsuite/libffi.call/cls_9byte1.c,
497 testsuite/libffi.call/cls_align_float.c,
498 testsuite/libffi.call/closure_fn1.c,
499 testsuite/libffi.call/problem1.c,
500 testsuite/libffi.call/closure_fn3.c,
501 testsuite/libffi.call/cls_sshort.c,
502 testsuite/libffi.call/closure_fn5.c,
503 testsuite/libffi.call/cls_align_double.c,
504 testsuite/libffi.call/cls_2byte.c,
505 testsuite/libffi.call/nested_struct.c,
506 testsuite/libffi.call/nested_struct10.c,
507 testsuite/libffi.call/cls_4byte.c,
508 testsuite/libffi.call/cls_6byte.c,
509 testsuite/libffi.call/cls_8byte.c,
510 testsuite/libffi.call/cls_multi_sshort.c,
511 testsuite/libffi.call/cls_align_uint32.c,
512 testsuite/libffi.call/cls_align_sint16.c,
513 testsuite/libffi.call/cls_float.c,
514 testsuite/libffi.call/cls_20byte.c,
515 testsuite/libffi.call/cls_5_1_byte.c,
516 testsuite/libffi.call/nested_struct2.c,
517 testsuite/libffi.call/cls_24byte.c,
518 testsuite/libffi.call/nested_struct4.c,
519 testsuite/libffi.call/nested_struct6.c,
520 testsuite/libffi.call/cls_64byte.c,
521 testsuite/libffi.call/nested_struct8.c,
522 testsuite/libffi.call/cls_uint.c,
523 testsuite/libffi.call/cls_multi_ushortchar.c,
524 testsuite/libffi.call/cls_schar.c,
525 testsuite/libffi.call/cls_uchar.c,
526 testsuite/libffi.call/cls_align_uint64.c,
527 testsuite/libffi.call/cls_ulonglong.c,
528 testsuite/libffi.call/cls_align_longdouble.c,
529 testsuite/libffi.call/cls_1_1byte.c,
530 testsuite/libffi.call/cls_12byte.c,
531 testsuite/libffi.call/cls_3_1byte.c,
532 testsuite/libffi.call/cls_3byte1.c,
533 testsuite/libffi.call/cls_4_1byte.c,
534 testsuite/libffi.call/cls_6_1_byte.c,
535 testsuite/libffi.call/cls_16byte.c,
536 testsuite/libffi.call/cls_18byte.c,
537 testsuite/libffi.call/closure_fn0.c,
538 testsuite/libffi.call/cls_9byte2.c,
539 testsuite/libffi.call/closure_fn2.c,
540 testsuite/libffi.call/closure_fn4.c,
541 testsuite/libffi.call/cls_ushort.c,
542 testsuite/libffi.call/closure_fn6.c,
543 testsuite/libffi.call/cls_5byte.c,
544 testsuite/libffi.call/cls_align_pointer.c,
545 testsuite/libffi.call/cls_7byte.c,
546 testsuite/libffi.call/cls_align_sint32.c,
547 testsuite/libffi.special/unwindtest_ffi_call.cc,
548 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
550 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
552 * aclocal.m4: Regenerated.
554 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
556 * configure: Regenerate.
558 2007-05-23 Steve Ellcey <sje@cup.hp.com>
560 * Makefile.in: Regenerate.
561 * configure: Regenerate.
562 * aclocal.m4: Regenerate.
563 * include/Makefile.in: Regenerate.
564 * testsuite/Makefile.in: Regenerate.
566 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
568 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
569 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
570 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
571 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
572 (FFI_CLOSURES): Enable closure support.
574 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
576 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
577 * configure: Regenerate.
578 * fficonfig.h.in: Regenerate.
579 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
580 CFI_OFFSET,CFI_DEF_CFA): New macros.
581 (ffi_call_SYSV): Add callframe annotation.
583 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
585 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
586 numerous test suite failures.
587 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
589 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
591 * Makefile.am (EXTRA_DIST): Bring up to date.
592 * Makefile.in: Regenerate.
593 * src/frv/eabi.S: Remove RCS keyword.
595 2007-04-06 Richard Henderson <rth@redhat.com>
597 * configure.ac: Tidy target case.
598 (HAVE_LONG_DOUBLE): Allow the target to override.
599 * configure: Regenerate.
600 * include/ffi.h.in: Don't define ffi_type_foo if
601 LIBFFI_HIDE_BASIC_TYPES is defined.
602 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
604 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
605 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
606 (ffi_type_longdouble): Special case for Alpha. Don't define
607 if long double == double.
609 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
610 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
611 (ffi_call, ffi_closure_osf_inner): Likewise.
612 (ffi_closure_osf_inner): Likewise. Mark hidden.
613 (ffi_call_osf, ffi_closure_osf): Mark hidden.
614 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
615 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
616 (load_table): Handle 128-bit long double.
618 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
620 2007-04-06 Tom Tromey <tromey@redhat.com>
623 * README: Fixed bug in example.
625 2007-04-03 Jakub Jelinek <jakub@redhat.com>
627 * src/closures.c: Include sys/statfs.h.
628 (_GNU_SOURCE): Define on Linux.
629 (FFI_MMAP_EXEC_SELINUX): Define.
630 (selinux_enabled): New variable.
631 (selinux_enabled_check): New function.
632 (is_selinux_enabled): Define.
635 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
637 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
638 Use 'volatile float sum' to create sum of floats to avoid false
639 negative due to excess precision on ix86 targets.
642 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
644 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
646 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
648 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
650 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
651 (ffi_prep_closure_loc): New.
652 (ffi_prep_raw_closure_loc): New.
653 (ffi_prep_java_raw_closure_loc): New.
654 * src/closures.c: New file.
655 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
656 Replace sflags with exec_offset.
657 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
658 sub_segment_exec_offset): New macros.
659 (get_segment_flags, set_segment_flags, check_segment_merge): New
661 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
662 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
663 destroy_mspace): Use new macros.
664 (sys_alloc): Silence warning.
665 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
666 * Makefile.in: Rebuilt.
667 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
668 terms of ffi_prep_closure_loc.
669 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
671 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
673 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
675 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
677 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
678 (ffi_prep_closure): ... this.
679 * src/pa/ffi.c: Likewise.
680 * src/cris/ffi.c: Likewise. Adjust.
681 * src/frv/ffi.c: Likewise.
682 * src/ia64/ffi.c: Likewise.
683 * src/mips/ffi.c: Likewise.
684 * src/powerpc/ffi_darwin.c: Likewise.
685 * src/s390/ffi.c: Likewise.
686 * src/sh/ffi.c: Likewise.
687 * src/sh64/ffi.c: Likewise.
688 * src/sparc/ffi.c: Likewise.
689 * src/x86/ffi64.c: Likewise.
690 * src/x86/ffi.c: Likewise.
691 (FFI_INIT_TRAMPOLINE): Adjust.
692 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
693 (ffi_prep_raw_closure): ... this.
694 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
695 (ffi_prep_closure): ... this.
696 (flush_icache): Adjust.
698 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
700 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
703 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
705 * Makefile.am: Add dummy install-pdf target.
706 * Makefile.in: Regenerate
708 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
710 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
711 ffi_closure_helper_SYSV): Add long double handling.
713 2007-02-02 Jakub Jelinek <jakub@redhat.com>
715 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
716 immediately after bctrl instruction.
718 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
720 * Makefile.am (all-recursive, install-recursive,
721 mostlyclean-recursive, clean-recursive, distclean-recursive,
722 maintainer-clean-recursive): Add missing targets.
723 * Makefile.in: Rebuilt.
725 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
727 * configure.ac: Add TARGET for x86_64-*-darwin*.
728 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
730 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
731 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
732 * configure: Regenerate.
733 * Makefile.in: Regenerate.
734 * include/Makefile.in: Regenerate.
735 * testsuite/Makefile.in: Regenerate.
736 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
739 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
741 * aclocal.m4: Regenerate with aclocal -I .. as written in the
744 2006-10-31 Geoffrey Keating <geoffk@apple.com>
746 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
747 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
749 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
750 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
752 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
753 Sandro Tolaini <tolaini@libero.it>
755 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
757 * configure: Regenerated.
758 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
759 (EXTRA_DIST): Add src/x86/darwin.S.
760 * Makefile.in: Regenerated.
761 * include/Makefile.in: Regenerated.
762 * testsuite/Makefile.in: Regenerated.
764 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
765 X86_WIN32, and additionally align stack to 16 bytes.
766 * src/x86/darwin.S: New, based on sysv.S.
767 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
769 2006-09-12 David Daney <ddaney@avtrex.com>
772 * include/Makefile.am: Install both ffi.h and ffitarget.h in
773 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
774 * aclocal.m4: Regenerated for automake 1.9.6.
775 * Makefile.in: Regenerated.
776 * include/Makefile.in: Regenerated.
777 * testsuite/Makefile.in: Regenerated.
779 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
781 * include/ffi_common.h (struct): Revert accidental commit.
783 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
785 * include/ffi_common.h: Remove lint directives.
786 * include/ffi.h.in: Likewise.
788 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
790 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
791 for 32-bit architectures.
792 * testsuite/libffi.call/return_ul.c: New test case.
794 2006-07-19 David Daney <ddaney@avtrex.com>
796 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
797 xfail remains for mips64.
799 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
801 * Makefile.am: Add install-html target. Add install-html to .PHONY
802 * Makefile.in: Regenerate.
803 * aclocal.m4: Regenerate.
804 * include/Makefile.in: Regenerate.
805 * testsuite/Makefile.in: Regenerate.
807 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
809 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
812 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
814 * README: Remove notice about 'Crazy Comments'.
815 * src/debug.c: Remove lint directives. Cleanup white spaces.
816 * src/java_raw_api.c: Likewise.
817 * src/prep_cif.c: Likewise.
818 * src/raw_api.c: Likewise.
819 * src/ffitest.c: Delete. No longer needed, all test cases migrated
821 * src/arm/ffi.c: Remove lint directives.
822 * src/m32r/ffi.c: Likewise.
823 * src/pa/ffi.c: Likewise.
824 * src/powerpc/ffi.c: Likewise.
825 * src/powerpc/ffi_darwin.c: Likewise.
826 * src/sh/ffi.c: Likewise.
827 * src/sh64/ffi.c: Likewise.
828 * src/x86/ffi.c: Likewise.
829 * testsuite/libffi.call/float2.c: Likewise.
830 * testsuite/libffi.call/promotion.c: Likewise.
831 * testsuite/libffi.call/struct1.c: Likewise.
833 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
835 * src/pa/hpux32.S: Correct unwind offset calculation for
837 * src/pa/linux.S: Likewise.
839 2006-04-12 James E Wilson <wilson@specifix.com>
842 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
843 (hfa_type_load): Call stf_spill.
844 (hfa_type_store): Call ldf_fill.
845 (ffi_call): Adjust calls to above routines. Add local temps for
848 2006-04-10 Matthias Klose <doko@debian.org>
850 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
851 directory names containing underscores.
853 2006-04-07 James E Wilson <wilson@specifix.com>
855 * testsuite/libffi.call/float4.c: New testcase.
857 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
858 Andreas Tobler <a.tobler@schweiz.ch>
860 * Makefile.am: Add PA_HPUX port.
861 * Makefile.in: Regenerate.
862 * include/Makefile.in: Likewise.
863 * testsuite/Makefile.in: Likewise.
864 * configure.ac: Add PA_HPUX rules.
865 * configure: Regenerate.
866 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
867 Add PA_HPUX and PA64_HPUX.
868 Rename FFI_LINUX ABI to FFI_PA32 ABI.
869 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
870 (FFI_TYPE_SMALL_STRUCT2): Define.
871 (FFI_TYPE_SMALL_STRUCT4): Likewise.
872 (FFI_TYPE_SMALL_STRUCT8): Likewise.
873 (FFI_TYPE_SMALL_STRUCT3): Redefine.
874 (FFI_TYPE_SMALL_STRUCT5): Likewise.
875 (FFI_TYPE_SMALL_STRUCT6): Likewise.
876 (FFI_TYPE_SMALL_STRUCT7): Likewise.
877 * src/pa/ffi.c (ROUND_DOWN): Delete.
878 (fldw, fstw, fldd, fstd): Use '__asm__'.
879 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
880 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
881 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
882 Simplify incrementing of stack slot variable. Change type of local
884 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
886 (ffi_prep_cif_machdep): Likewise.
887 (ffi_call): Likewise.
888 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
889 return type to ffi_status. Simplify incrementing of stack slot
890 variable. Only copy floating point argument registers when PA_LINUX
891 is true. Reformat debug statement.
892 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
893 FFI_TYPE_SMALL_STRUCT8.
894 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
896 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
897 Add nops to cache flush. Add trampoline for PA_HPUX.
898 * src/pa/hpux32.S: New file.
899 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
900 ffi_prep_args_LINUX to ffi_prep_args_pa32.
901 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
902 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
903 argument type checks so that common argument types appear first.
904 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
905 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
907 2006-03-24 Alan Modra <amodra@bigpond.net.au>
909 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
910 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
911 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
913 (ffi_prep_args64): Assert using IBM extended double.
914 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
915 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
916 (ffi_call): Handle FFI_LINUX.
917 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
918 gpr3 return pointer as for struct return. Handle FFI_LINUX
919 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
921 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
922 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
923 Don't use r6 as pointer to results, instead use sp offset. Don't
924 make a special call to load lr with case table address, instead
925 use offset from previous call.
926 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
927 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
930 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
932 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
933 passed with FP registers correctly.
934 (ffi_closure_helper_SYSV): Likewise.
935 * src/sh64/sysv.S: Likewise.
937 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
939 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
940 args and userdata unused.
941 (closure_test_fn1): Mark cif and userdata unused.
942 (main): Remove unused res.
944 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
946 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
947 -O2, -O3, -Os and the warning flags -W -Wall.
948 * testsuite/libffi.special/special.exp: Likewise.
949 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
950 unused parameter unused for gcc or else do nothing.
951 * testsuite/libffi.special/ffitestcxx.h: Likewise.
952 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
954 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
955 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
956 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
957 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
958 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
959 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
960 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
961 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
962 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
963 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
964 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
965 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
966 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
967 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
968 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
969 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
970 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
971 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
972 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
973 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
974 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
975 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
976 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
977 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
979 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
981 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
983 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
984 void* to avoid compiler warning.
986 (cls_struct_align_gn): Mark cif and userdata unused.
987 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
989 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
991 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
993 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
995 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
997 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
998 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
999 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1001 (main): Cast res_call to silence gcc.
1002 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1004 (main): Cast res_call to silence gcc.
1005 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1007 (main): Cast res_call to silence gcc.
1008 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1010 (main): Cast res_call to silence gcc.
1011 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1013 (main): Cast res_call to silence gcc.
1014 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1016 (main): Cast res_call to silence gcc.
1017 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1019 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1020 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1022 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1023 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1025 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1026 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1028 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1029 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1031 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1032 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1033 and userdata unused.
1034 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1036 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1037 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1038 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1039 Cleanup white spaces.
1040 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1041 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1042 cif and userdata unused.
1043 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1045 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1046 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1047 formatters to silence gcc.
1048 (B_gn): Mark cif and userdata unused.
1049 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1051 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1052 (B_gn): Mark cif and userdata unused.
1053 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1055 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1056 (B_gn): Mark cif and userdata unused.
1057 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1059 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1060 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1061 * testsuite/libffi.call/problem1.c (stub): Likewise.
1062 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1064 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1065 in the last commit for this test case in the test case itself.
1066 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1068 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1069 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1070 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1071 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1072 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1073 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1075 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1077 * src/sh/sysv.S: Fix register numbers in the FDE for
1080 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1082 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1083 declaration to avoid a false negative on ix86. See PR323.
1085 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1087 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1088 and cast integer to void * if needed. Update the pointer to
1089 the FP register saved area correctly.
1091 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1093 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1095 * testsuite/libffi.call/nested_struct4.c: Likewise.
1097 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1099 * testsuite/libffi.call/return_dbl.c: New test case.
1100 * testsuite/libffi.call/return_dbl1.c: Likewise.
1101 * testsuite/libffi.call/return_dbl2.c: Likewise.
1102 * testsuite/libffi.call/return_fl.c: Likewise.
1103 * testsuite/libffi.call/return_fl1.c: Likewise.
1104 * testsuite/libffi.call/return_fl2.c: Likewise.
1105 * testsuite/libffi.call/return_fl3.c: Likewise.
1106 * testsuite/libffi.call/closure_fn6.c: Likewise.
1108 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1110 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1111 here to be used by other test cases too.
1113 * testsuite/libffi.call/nested_struct10.c: New test case.
1114 * testsuite/libffi.call/nested_struct9.c: Likewise.
1115 * testsuite/libffi.call/nested_struct8.c: Likewise.
1116 * testsuite/libffi.call/nested_struct7.c: Likewise.
1117 * testsuite/libffi.call/nested_struct6.c: Likewise.
1118 * testsuite/libffi.call/nested_struct5.c: Likewise.
1119 * testsuite/libffi.call/nested_struct4.c: Likewise.
1121 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1123 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1124 * configure: Rebuilt.
1126 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1128 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1129 instead do the shifting inline.
1130 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1131 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1132 and 8 byte structs, for the remaining struct sizes don't call
1133 __lshrdi3, instead do the shifting inline.
1135 2005-12-07 Thiemo Seufer <ths@networkno.de>
1137 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1138 missing parentheses.
1139 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1140 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1141 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1142 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1143 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1145 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1147 (ffi_prep_closure): Improve trampoline instruction scheduling.
1148 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1150 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1152 * src/powerpc/ffi.c: Formatting.
1153 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1154 (ffi_prep_args64): Likewise.
1156 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1158 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1159 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1161 2005-09-26 Tom Tromey <tromey@redhat.com>
1163 * testsuite/libffi.call/float1.c (value_type): New typedef.
1164 (CANARY): New define.
1165 (main): Check for result buffer overflow.
1166 * src/powerpc/linux64.S: Handle linux64 long double returns.
1167 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1168 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1170 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1173 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1175 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1177 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1179 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1180 (AH_BOTTOM): Add FFI_HIDDEN definition.
1181 * configure: Rebuilt.
1182 * fficonfig.h.in: Rebuilt.
1183 * src/powerpc/ffi.c (hidden): Remove.
1184 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1185 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1186 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1187 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1188 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1189 add FFI_HIDDEN to its prototype.
1190 (ffi_closure_SYSV_inner): New.
1191 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1192 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1194 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1197 * configure: Rebuilt.
1198 * configure.ac: Handle i*86-*-gnu*.
1200 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1202 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1203 DW_CFA_offset_extended_sf rather than
1204 DW_CFA_GNU_negative_offset_extended.
1205 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1207 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1209 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1211 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1212 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1213 partially on register.
1214 (ffi_closure_helper_SYSV): Likewise.
1215 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1217 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1219 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1221 * src/sh64/ffi.c (simple_type): Remove.
1222 (return_type): Handle small structures correctly.
1223 (ffi_prep_args): Likewise.
1224 (ffi_call): Likewise.
1225 (ffi_closure_helper_SYSV): Likewise.
1226 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1227 Emit position independent code if PIC and remove wrong datalabel
1228 prefixes from EH data.
1230 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1232 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1233 * Makefile.in: Regenerate.
1234 * include/Makefile.in: Likewise.
1235 * testsuite/Makefile.in: Likewise.
1236 * configure.ac: Add POWERPC_FREEBSD rules.
1237 * configure: Regenerate.
1238 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1239 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1240 * src/powerpc/ffi.c: Add flags to handle small structure returns
1242 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1244 (ffi_closure_helper_SYSV): Likewise.
1245 * src/powerpc/ppc_closure.S: Add return types for small structures.
1246 * src/powerpc/sysv.S: Add bits to handle small structures for
1249 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1251 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1252 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1253 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1255 2005-07-05 Randolph Chung <tausq@debian.org>
1257 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1258 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1259 structures. Kill compilation warnings.
1260 (ffi_closure_inner_LINUX): Print return values as hex in debug
1261 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1262 Properly handle 5-7 byte structure returns.
1263 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1264 (FFI_TYPE_SMALL_STRUCT2): Remove.
1265 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1266 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1267 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1268 (checksmst1, checksmst2): Remove.
1269 (checksmst3): Optimize handling of 3-byte struct returns.
1270 (checksmst567): Properly handle 5-7 byte struct returns.
1272 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1275 * src/mips/n32.S: Enforce PIC code.
1276 * src/mips/o32.S: Likewise.
1278 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1280 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1281 * configure: Regenerate.
1283 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1285 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1286 to call ffi_closure_helper_SYSV. Append @local instead.
1287 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1289 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1291 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1292 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1293 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1294 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1295 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1297 2005-05-09 Mike Stump <mrs@apple.com>
1299 * configure: Regenerate.
1301 2005-05-08 Richard Henderson <rth@redhat.com>
1304 * src/alpha/osf.S: Update unwind into to match code.
1306 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1307 Richard Henderson <rth@redhat.com>
1309 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1311 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1312 (ffi_prep_closure): Set carry bit if sse-used flag set.
1313 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1314 Only load sse registers if ssecount non-zero.
1315 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1317 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1319 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1320 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1321 * configure: Regenerate.
1323 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1325 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1326 have Tcl8.3-compatible intermediate variable.
1328 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1329 Hans-Peter Nilsson <hp@axis.com>
1331 * Makefile.am: Add CRIS support.
1332 * configure.ac: Likewise.
1333 * Makefile.in, configure, testsuite/Makefile.in,
1334 include/Makefile.in: Regenerate.
1335 * src/cris: New directory.
1336 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1337 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1339 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1340 \r?\n in output tests.
1342 2005-04-12 Mike Stump <mrs@apple.com>
1344 * configure: Regenerate.
1346 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1348 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1350 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1352 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1354 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1355 to 64 bits in ILP32 mode.
1356 Load 64 bits even for short data.
1358 2005-03-23 Mike Stump <mrs@apple.com>
1360 * src/powerpc/darwin.S: Update for -m64 multilib.
1361 * src/powerpc/darwin_closure.S: Likewise.
1363 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1365 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1366 Do not set tool_include_dir.
1367 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1369 * include/Makefile.am: Set gcc_version and toollibffidir.
1370 * include/Makefile.in: Regenerate.
1372 2005-02-22 Andrew Haley <aph@redhat.com>
1374 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1375 odd-numbered register pairs for 64-bit integer types.
1377 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1380 * testsuite/libffi.call/return_ll1.c: New test case.
1382 2005-02-11 Janis Johnson <janis187@us.ibm.com>
1384 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1385 * testsuite/libffi.call/float.c: Ditto.
1386 * testsuite/libffi.call/float2.c: Ditto.
1387 * testsuite/libffi.call/float3.c: Ditto.
1389 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1391 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1393 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1395 * testsuite/libffi.special/special.exp (cxx_options): Add
1398 2004-12-31 Richard Henderson <rth@redhat.com>
1400 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1401 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1402 offset parameters with a type parameter; deduce size and structure
1403 alignment. Update all users.
1405 2004-12-31 Richard Henderson <rth@redhat.com>
1407 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1408 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
1409 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
1410 into ffi_prep_closure.
1411 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1414 2004-12-27 Richard Henderson <rth@redhat.com>
1416 * src/x86/unix64.S: Fix typo in unwind info.
1418 2004-12-25 Richard Henderson <rth@redhat.com>
1420 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1421 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1422 (merge_classes): Check for it.
1424 (classify_argument): Pass byte_offset by value; perform all updates
1426 (examine_argument): Add classes argument; handle
1427 X86_64_COMPLEX_X87_CLASS.
1428 (ffi_prep_args): Merge into ...
1429 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1430 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1431 (ffi_fill_return_value): Remove.
1432 (ffi_prep_closure): Remove dead assert.
1433 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1434 Rewrite to use struct register_args instead of va_list. Create
1435 flags for handling structure returns.
1436 * src/x86/unix64.S: Remove dead strings.
1437 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1438 stack frame with ffi_call. Handle structure returns properly.
1439 (float2sse, floatfloat2sse, double2sse): Remove.
1440 (sse2float, sse2double, sse2floatfloat): Remove.
1441 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1442 to handle structure returns properly.
1444 2004-12-08 David Edelsohn <edelsohn@gnu.org>
1446 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1448 * Makefile.in: Regenerated.
1450 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
1452 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1453 * configure, aclocal.m4, Makefile.in: Regenerate.
1454 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1456 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1458 * configure: Regenerate for libtool change.
1460 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1462 * configure: Regenerate for libtool reversion.
1464 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1466 * configure: Regenerate for libtool change.
1468 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1470 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1472 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
1474 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1475 of jal. Use an absolute encoding for the frame information.
1477 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1479 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1480 * acinclude.m4: Delete logic for sincludes.
1481 * aclocal.m4, Makefile.in, configure: Regenerate.
1482 * include/Makefile: Likewise.
1483 * testsuite/Makefile: Likewise.
1485 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1487 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1488 on a 8-byte boundary.
1489 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1491 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1493 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1494 long long values. Round stack allocation to a multiple of 8 bytes
1495 for ATPCS compatibility.
1496 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1497 names. Handle returning long long types. Add Thumb and interworking
1498 support. Improve soft-float code.
1500 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1502 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1503 (libffi_exit): New function.
1504 (libffi_init): Build the testglue wrapper if needed.
1506 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1509 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1511 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1513 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1515 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1517 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1518 * testsuite/libffi.call/float3.c: New test case.
1520 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1522 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1523 the function returning a structure pointed with R2.
1524 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1525 the structure return value if T bit set. Emit position
1526 independent code and EH data if PIC.
1528 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1530 * Makefile.am: Add m32r support.
1531 * configure.ac: Likewise.
1532 * Makefile.in: Regenerate.
1533 * confiugre: Regenerate.
1534 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1535 (uint64, sint64, double, longdouble)
1536 * src/m32r: New directory.
1537 * src/m32r/ffi.c: New file.
1538 * src/m32r/sysv.S: Likewise.
1539 * src/m32r/ffitarget.h: Likewise.
1541 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1543 * testsuite/libffi.call/negint.c: New test case.
1545 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1548 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1549 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1550 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1553 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1555 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1556 * testsuite/libffi.call/promotion.c: Likewise.
1557 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1559 * testsuite/libffi.call/return_sc.c: Likewise.
1560 * testsuite/libffi.call/return_uc.c: Likewise.
1562 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1564 * src/powerpc/darwin.S: Fix comments and identation.
1565 * src/powerpc/darwin_closure.S: Likewise.
1567 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1569 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1570 (ffi_prep_args): Handle longdouble arguments.
1571 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1573 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1574 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1576 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1577 * src/types.c: Defined longdouble size and alignment for darwin.
1579 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1581 * src/powerpc/aix.S: Remove whitespaces.
1582 * src/powerpc/aix_closure.S: Likewise.
1583 * src/powerpc/asm.h: Likewise.
1584 * src/powerpc/ffi.c: Likewise.
1585 * src/powerpc/ffitarget.h: Likewise.
1586 * src/powerpc/linux64.S: Likewise.
1587 * src/powerpc/linux64_closure.S: Likewise.
1588 * src/powerpc/ppc_closure.S: Likewise.
1589 * src/powerpc/sysv.S: Likewise.
1591 2004-08-30 Anthony Green <green@redhat.com>
1593 * Makefile.am: Add frv support.
1594 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1595 * configure.ac: Read configure.host.
1596 * configure.in: Read configure.host.
1597 * configure.host: New file. frv-elf needs libgloss.
1598 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
1599 alignment. This is needed to frv and shouldn't harm the others.
1600 * include/ffi_common.h (ALIGN_DOWN): New macro.
1601 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1603 2004-08-24 David Daney <daney@avtrex.com>
1605 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1606 * testsuite/libffi.call/closure_fn1.c: Likewise.
1607 * testsuite/libffi.call/closure_fn2.c Likewise.
1608 * testsuite/libffi.call/closure_fn3.c: Likewise.
1609 * testsuite/libffi.call/closure_fn4.c: Likewise.
1610 * testsuite/libffi.call/closure_fn5.c: Likewise.
1611 * testsuite/libffi.call/cls_18byte.c: Likewise.
1612 * testsuite/libffi.call/cls_19byte.c: Likewise.
1613 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1614 * testsuite/libffi.call/cls_20byte.c: Likewise.
1615 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1616 * testsuite/libffi.call/cls_24byte.c: Likewise.
1617 * testsuite/libffi.call/cls_2byte.c: Likewise.
1618 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1619 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1620 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1621 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1622 * testsuite/libffi.call/cls_4byte.c: Likewise.
1623 * testsuite/libffi.call/cls_64byte.c: Likewise.
1624 * testsuite/libffi.call/cls_6byte.c: Likewise.
1625 * testsuite/libffi.call/cls_7byte.c: Likewise.
1626 * testsuite/libffi.call/cls_8byte.c: Likewise.
1627 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1628 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1629 * testsuite/libffi.call/cls_align_double.c: Likewise.
1630 * testsuite/libffi.call/cls_align_float.c: Likewise.
1631 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1632 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1633 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1634 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1635 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1636 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1637 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1638 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1639 * testsuite/libffi.call/cls_double.c: Likewise.
1640 * testsuite/libffi.call/cls_float.c: Likewise.
1641 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1642 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1643 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1644 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1645 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1646 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1647 * testsuite/libffi.call/cls_schar.c: Likewise.
1648 * testsuite/libffi.call/cls_sint.c: Likewise.
1649 * testsuite/libffi.call/cls_sshort.c: Likewise.
1650 * testsuite/libffi.call/cls_uchar.c: Likewise.
1651 * testsuite/libffi.call/cls_uint.c: Likewise.
1652 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1653 * testsuite/libffi.call/cls_ushort.c: Likewise.
1654 * testsuite/libffi.call/nested_struct.c: Likewise.
1655 * testsuite/libffi.call/nested_struct1.c: Likewise.
1656 * testsuite/libffi.call/nested_struct2.c: Likewise.
1657 * testsuite/libffi.call/nested_struct3.c: Likewise.
1658 * testsuite/libffi.call/problem1.c: Likewise.
1659 * testsuite/libffi.special/unwindtest.cc: Likewise.
1660 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1662 * testsuite/libffi.call/cls_16byte.c: Likewise.
1663 * testsuite/libffi.call/cls_5byte.c: Likewise.
1665 2004-08-23 David Daney <daney@avtrex.com>
1668 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1669 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1670 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1671 parameters and return types.
1672 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1673 (ffi_prep_closure): Ditto.
1674 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1675 alignment calculations.
1676 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1677 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1679 2004-08-14 Casey Marshall <csm@gnu.org>
1681 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1682 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1683 integer (O32 ABI only).
1684 (ffi_prep_closure): new function.
1685 (ffi_closure_mips_inner_O32): new function.
1686 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
1687 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1688 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1689 64 bit integers correctly.
1690 (ffi_closure_O32): new function.
1691 Added DWARF-2 unwind info for both functions.
1693 2004-08-10 Andrew Haley <aph@redhat.com>
1695 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1697 2004-08-01 Robert Millan <robertmh@gnu.org>
1699 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1700 * configure: Regenerate.
1702 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1704 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1705 and mmap() explicitly instead of relying on preset autoconf cache
1707 * aclocal.m4: Regenerate.
1708 * configure: Regenerate.
1710 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1712 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1713 (ffi_check_float_struct): Remove unused prototype.
1715 2004-06-30 Geoffrey Keating <geoffk@apple.com>
1717 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1718 character on Darwin, use '\n\t' instead.
1720 2004-06-26 Matthias Klose <doko@debian.org>
1722 * libtool-version: Fix typo in revision/age.
1724 2004-06-17 Matthias Klose <doko@debian.org>
1726 * libtool-version: New.
1727 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1728 * Makefile.in: Regenerate.
1730 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1732 * Makefile.am: Remove useless multilib rules.
1733 * Makefile.in: Regenerate.
1734 * aclocal.m4: Regenerate with automake 1.8.5.
1735 * configure.ac: Remove useless multilib configury.
1736 * configure: Regenerate.
1738 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1740 * .cvsignore: New file.
1742 2004-06-10 Jakub Jelinek <jakub@redhat.com>
1744 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1746 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1749 2004-06-06 Sean McNeil <sean@mcneil.com>
1751 * configure.ac: Add x86_64-*-freebsd* support.
1752 * configure: Regenerate.
1754 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
1757 * configure.ac: Test for existence of mmap and sys/mman.h before
1758 checking blacklist. Fix suggested by Jim Wilson.
1759 * configure: Regenerate.
1761 2004-04-26 Matt Austern <austern@apple.com>
1763 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1765 * src/powerpc/darwin_closure.S: Likewise.
1767 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1769 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1770 error. Reported by Thomas Heller <theller@python.net>.
1771 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1772 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1774 2004-03-20 Matthias Klose <doko@debian.org>
1776 * src/pa/linux.S: Fix typo.
1778 2004-03-19 Matthias Klose <doko@debian.org>
1780 * Makefile.am: Update.
1781 * Makefile.in: Regenerate.
1782 * src/pa/ffi.h.in: Remove.
1783 * src/pa/ffitarget.h: New file.
1785 2004-02-10 Randolph Chung <tausq@debian.org>
1787 * Makefile.am: Add PA support.
1788 * Makefile.in: Regenerate.
1789 * include/Makefile.in: Regenerate.
1790 * configure.ac: Add PA target.
1791 * configure: Regenerate.
1792 * src/pa/ffi.c: New file.
1793 * src/pa/ffi.h.in: Add PA support.
1794 * src/pa/linux.S: New file.
1795 * prep_cif.c: Add PA support.
1797 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
1799 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
1801 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
1802 with ecif->cif->flags.
1803 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
1805 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
1806 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
1807 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
1808 sc_retstruct2b): Add for 1 or 2-bytes struct case.
1810 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
1812 * configure.in: Rename file to ...
1813 * configure.ac: ... this.
1814 * fficonfig.h.in: Regenerate.
1815 * Makefile.in: Regenerate.
1816 * include/Makefile.in: Regenerate.
1817 * testsuite/Makefile.in: Regenerate.
1819 2004-03-12 Matt Austern <austern@apple.com>
1821 * src/powerpc/darwin.S: Fix EH information so it corresponds to
1822 changes in EH format resulting from addition of linkonce support.
1823 * src/powerpc/darwin_closure.S: Likewise.
1825 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
1826 Paolo Bonzini <bonzini@gnu.org>
1828 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
1829 Remove VPATH. Remove rules for object files. Remove multilib support.
1830 (AM_CCASFLAGS): Add.
1831 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
1832 (AC_PREREQ): Bump version to 2.59.
1833 (AC_INIT): Fill with version info and bug address.
1834 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
1835 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
1836 De-precious CC so that the right flags are passed down to multilibs.
1837 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
1838 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
1839 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
1840 * configure: Rebuilt.
1841 * aclocal.m4: Likewise.
1842 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
1843 * fficonfig.h.in: Likewise.
1845 2004-03-11 Andreas Schwab <schwab@suse.de>
1847 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
1848 arguments from fp registers only for the first 8 parameter slots.
1849 Don't convert a float parameter when passed in memory.
1851 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
1853 * configure: Regenerate for config/accross.m4 correction.
1855 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
1857 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
1858 ecif->cif->bytes to bytes.
1859 (ffi_prep_cif_machdep): Add braces around nested if statement.
1861 2004-02-09 Alan Modra <amodra@bigpond.net.au>
1863 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
1865 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
1866 (ffi_closure_helper_LINUX64): Fix typo.
1867 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
1869 * testsuite/libffi.call/float.c: Likewise.
1870 * testsuite/libffi.call/float2.c: Likewise.
1872 2004-02-08 Alan Modra <amodra@bigpond.net.au>
1874 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
1875 long double function return and long double arg handling.
1876 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
1877 Use "end_pfr" instead of "nf". Correct long double handling.
1879 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
1881 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
1882 space for long double return value. Adjust stack frame and offsets.
1883 Load f2 long double return.
1885 2004-02-07 Alan Modra <amodra@bigpond.net.au>
1887 * src/types.c: Use 16 byte long double for POWERPC64.
1889 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1891 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
1892 when the structure return address is passed in %o0.
1893 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
1894 (ffi_v9_layout_struct): Align the field following a nested structure
1895 on a word boundary. Use memmove instead of memcpy.
1896 (ffi_call): Update call to ffi_V9_return_struct.
1897 (ffi_prep_closure): Define 'ctx' only for V8.
1898 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
1899 and ffi_closure_sparc_inner_v9.
1900 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
1901 Always skip the structure return address. For structures and long
1902 doubles, copy the argument directly.
1903 (ffi_closure_sparc_inner_v9): Skip the structure return address only
1904 if required. Shift the maximum floating-point slot accordingly. For
1905 big structures, copy the argument directly; otherwise, left-justify the
1906 argument and call ffi_v9_layout_struct to lay out the structure on
1908 * src/sparc/v8.S: Undef STACKFRAME before defining it.
1909 (ffi_closure_v8): Pass the structure return address. Update call to
1910 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
1911 Skip the 'unimp' insn when returning long doubles and structures.
1912 * src/sparc/v9.S: Undef STACKFRAME before defining it.
1913 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
1914 FFI_TYPE_INT handling. Load structures both in integers and
1915 floating-point registers on return.
1916 * README: Update status of the SPARC port.
1918 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
1920 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
1922 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
1924 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
1926 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
1927 value as of type ffi_arg, not unsigned int.
1929 2004-01-21 Michael Ritzert <ritzert@t-online.de>
1931 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
1934 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
1936 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
1939 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1941 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
1944 2003-12-10 Richard Henderson <rth@redhat.com>
1946 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
1947 size_t instead of int.
1949 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
1951 * testsuite/libffi.call/many_win32.c: Include <float.h>.
1952 * testsuite/libffi.call/many_win32.c (main): Replace variable
1953 int i with unsigned long ul.
1955 * testsuite/libffi.call/cls_align_uint64.c: New test case.
1956 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1957 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1958 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1959 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1960 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1961 * testsuite/libffi.call/cls_align_float.c: Likewise.
1962 * testsuite/libffi.call/cls_align_double.c: Likewise.
1963 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1964 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1966 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
1969 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
1970 Align arguments to 32 bits.
1972 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
1975 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
1976 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1977 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1978 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1979 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1980 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1982 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
1984 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
1986 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
1987 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1989 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
1991 * Makefile.in: Rebuilt.
1992 * configure: Likewise.
1993 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
1996 2003-11-21 Andreas Jaeger <aj@suse.de>
1997 Andreas Tobler <a.tobler@schweiz.ch>
1999 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2000 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2001 * Makefile.in: Rebuilt.
2002 * aclocal.m4: Likewise.
2003 * configure: Likewise.
2004 * fficonfig.h.in: Likewise.
2005 * testsuite/lib/libffi-dg.exp: Add include dir.
2006 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2007 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2008 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2009 for ffi_closure if available.
2010 * testsuite/libffi.call/closure_fn1.c: Likewise.
2011 * testsuite/libffi.call/closure_fn2.c: Likewise.
2012 * testsuite/libffi.call/closure_fn3.c: Likewise.
2013 * testsuite/libffi.call/closure_fn4.c: Likewise.
2014 * testsuite/libffi.call/closure_fn5.c: Likewise.
2015 * testsuite/libffi.call/cls_12byte.c: Likewise.
2016 * testsuite/libffi.call/cls_16byte.c: Likewise.
2017 * testsuite/libffi.call/cls_18byte.c: Likewise.
2018 * testsuite/libffi.call/cls_19byte.c: Likewise.
2019 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2020 * testsuite/libffi.call/cls_20byte.c: Likewise.
2021 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2022 * testsuite/libffi.call/cls_24byte.c: Likewise.
2023 * testsuite/libffi.call/cls_2byte.c: Likewise.
2024 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2025 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2026 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2027 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2028 * testsuite/libffi.call/cls_4byte.c: Likewise.
2029 * testsuite/libffi.call/cls_5byte.c: Likewise.
2030 * testsuite/libffi.call/cls_64byte.c: Likewise.
2031 * testsuite/libffi.call/cls_6byte.c: Likewise.
2032 * testsuite/libffi.call/cls_7byte.c: Likewise.
2033 * testsuite/libffi.call/cls_8byte.c: Likewise.
2034 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2035 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2036 * testsuite/libffi.call/cls_double.c: Likewise.
2037 * testsuite/libffi.call/cls_float.c: Likewise.
2038 * testsuite/libffi.call/cls_schar.c: Likewise.
2039 * testsuite/libffi.call/cls_sint.c: Likewise.
2040 * testsuite/libffi.call/cls_sshort.c: Likewise.
2041 * testsuite/libffi.call/cls_uchar.c: Likewise.
2042 * testsuite/libffi.call/cls_uint.c: Likewise.
2043 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2044 * testsuite/libffi.call/cls_ushort.c: Likewise.
2045 * testsuite/libffi.call/nested_struct.c: Likewise.
2046 * testsuite/libffi.call/nested_struct1.c: Likewise.
2047 * testsuite/libffi.call/nested_struct2.c: Likewise.
2048 * testsuite/libffi.call/nested_struct3.c: Likewise.
2049 * testsuite/libffi.call/problem1.c: Likewise.
2050 * testsuite/libffi.special/unwindtest.cc: Likewise.
2052 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2054 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2056 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2058 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2059 Add -lgcc_s to additional flags.
2061 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2063 * configure.in, include/Makefile.am: PR libgcj/11147, install
2064 the ffitarget.h header file in a gcc versioned and target
2066 * configure: Regenerated.
2067 * Makefile.in, include/Makefile.in: Likewise.
2068 * testsuite/Makefile.in: Likewise.
2070 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2072 * testsuite/libffi.call/closure_fn0.c: Print result and check
2073 with dg-output to make debugging easier.
2074 * testsuite/libffi.call/closure_fn1.c: Likewise.
2075 * testsuite/libffi.call/closure_fn2.c: Likewise.
2076 * testsuite/libffi.call/closure_fn3.c: Likewise.
2077 * testsuite/libffi.call/closure_fn4.c: Likewise.
2078 * testsuite/libffi.call/closure_fn5.c: Likewise.
2079 * testsuite/libffi.call/cls_12byte.c: Likewise.
2080 * testsuite/libffi.call/cls_16byte.c: Likewise.
2081 * testsuite/libffi.call/cls_18byte.c: Likewise.
2082 * testsuite/libffi.call/cls_19byte.c: Likewise.
2083 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2084 * testsuite/libffi.call/cls_20byte.c: Likewise.
2085 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2086 * testsuite/libffi.call/cls_24byte.c: Likewise.
2087 * testsuite/libffi.call/cls_2byte.c: Likewise.
2088 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2089 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2090 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2091 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2092 * testsuite/libffi.call/cls_4byte.c: Likewise.
2093 * testsuite/libffi.call/cls_5byte.c: Likewise.
2094 * testsuite/libffi.call/cls_64byte.c: Likewise.
2095 * testsuite/libffi.call/cls_6byte.c: Likewise.
2096 * testsuite/libffi.call/cls_7byte.c: Likewise.
2097 * testsuite/libffi.call/cls_8byte.c: Likewise.
2098 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2099 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2100 * testsuite/libffi.call/cls_double.c: Likewise.
2101 * testsuite/libffi.call/cls_float.c: Likewise.
2102 * testsuite/libffi.call/cls_schar.c: Likewise.
2103 * testsuite/libffi.call/cls_sint.c: Likewise.
2104 * testsuite/libffi.call/cls_sshort.c: Likewise.
2105 * testsuite/libffi.call/cls_uchar.c: Likewise.
2106 * testsuite/libffi.call/cls_uint.c: Likewise.
2107 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2108 * testsuite/libffi.call/cls_ushort.c: Likewise.
2109 * testsuite/libffi.call/problem1.c: Likewise.
2111 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2114 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2116 * testsuite/libffi.call/cls_9byte2.c: New test case.
2117 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2118 * testsuite/libffi.call/cls_64byte.c: Likewise.
2119 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2120 * testsuite/libffi.call/cls_19byte.c: Likewise.
2121 * testsuite/libffi.call/cls_18byte.c: Likewise.
2122 * testsuite/libffi.call/closure_fn4.c: Likewise.
2123 * testsuite/libffi.call/closure_fn5.c: Likewise.
2124 * testsuite/libffi.call/cls_schar.c: Likewise.
2125 * testsuite/libffi.call/cls_sint.c: Likewise.
2126 * testsuite/libffi.call/cls_sshort.c: Likewise.
2127 * testsuite/libffi.call/nested_struct2.c: Likewise.
2128 * testsuite/libffi.call/nested_struct3.c: Likewise.
2130 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2132 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2133 * testsuite/libffi.call/cls_uchar.c: Likewise.
2134 * testsuite/libffi.call/cls_uint.c: Likewise.
2135 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2136 * testsuite/libffi.call/cls_ushort.c: Likewise.
2137 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2139 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2141 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2144 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2146 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2147 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2149 2003-10-22 David Daney <ddaney@avtrex.com>
2151 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2152 __attribute__((__mode__(__SI__))) and friends.
2154 2003-10-22 Andreas Schwab <schwab@suse.de>
2156 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2158 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2160 * configure.in: AC_LINK_FILES(ffitarget.h).
2161 * configure: Regenerate.
2162 * Makefile.in: Likewise.
2163 * include/Makefile.in: Likewise.
2164 * testsuite/Makefile.in: Likewise.
2165 * fficonfig.h.in: Likewise.
2167 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2168 Richard Henderson <rth@redhat.com>
2170 Avoid that ffi.h includes fficonfig.h.
2172 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2173 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2174 (TARGET_SRC_MIPS_SGI): Removed.
2175 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2176 (MIPS_SGI): Removed.
2177 (CLEANFILES): Removed.
2178 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2180 * acconfig.h: Removed.
2181 * configure.in: Compute sizeofs only for double and long double.
2182 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2183 into AC_DEFINE instead of using acconfig.h. Create
2184 include/ffitarget.h instead of include/fficonfig.h. Rename
2185 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2186 AC_DEFINE EH_FRAME_FLAGS.
2187 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2188 (hack_DATA): Add ffitarget.h.
2189 * include/ffi.h.in: Remove all system specific definitions.
2190 Declare raw API even if it is not installed, why bother?
2191 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2192 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2193 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2194 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2195 * include/ffi_common.h (bool): Do not define.
2196 (ffi_assert): Accept failed assertion.
2197 (ffi_type_test): Return void and accept file/line.
2198 (FFI_ASSERT): Pass stringized failed assertion.
2199 (FFI_ASSERT_AT): New macro.
2200 (FFI_ASSERT_VALID_TYPE): New macro.
2201 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2202 UINT64, SINT64): Define here with gcc's __attribute__ macro
2204 (FLOAT32, ALIGN): Define here instead of in ffi.h
2205 * include/ffi-mips.h: Removed. Its content moved to
2206 src/mips/ffitarget.h after separating assembly and C sections.
2207 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2208 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2209 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2210 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2211 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2212 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2213 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2214 (ffi_type_test): Rewritten.
2215 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2216 FFI_ASSERT_VALID_TYPE.
2217 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2218 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2219 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2220 src/s390/ffitarget.h, src/sh/ffitarget.h,
2221 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2222 src/x86/ffitarget.h: New files.
2223 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2224 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2225 src/powerpc/aix.S, src/powerpc/darwin.S,
2226 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2227 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2228 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2229 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2230 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2233 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2235 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2236 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2238 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2240 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2241 Used when FFI_DEBUG = 1.
2243 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2245 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2248 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2250 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2251 FFI_MIPS_O32 for O32 ABI.
2253 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2255 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2256 SPARC64. Cleanup whitespaces.
2258 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2260 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2261 strongarm, xscale. Cleanup whitespaces.
2262 * testsuite/libffi.call/closure_fn1.c: Likewise.
2263 * testsuite/libffi.call/closure_fn2.c: Likewise.
2264 * testsuite/libffi.call/closure_fn3.c: Likewise.
2265 * testsuite/libffi.call/cls_12byte.c: Likewise.
2266 * testsuite/libffi.call/cls_16byte.c: Likewise.
2267 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2268 * testsuite/libffi.call/cls_20byte.c: Likewise.
2269 * testsuite/libffi.call/cls_24byte.c: Likewise.
2270 * testsuite/libffi.call/cls_2byte.c: Likewise.
2271 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2272 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2273 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2274 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2275 * testsuite/libffi.call/cls_4byte.c: Likewise.
2276 * testsuite/libffi.call/cls_5byte.c: Likewise.
2277 * testsuite/libffi.call/cls_6byte.c: Likewise.
2278 * testsuite/libffi.call/cls_7byte.c: Likewise.
2279 * testsuite/libffi.call/cls_8byte.c: Likewise.
2280 * testsuite/libffi.call/cls_double.c: Likewise.
2281 * testsuite/libffi.call/cls_float.c: Likewise.
2282 * testsuite/libffi.call/cls_uchar.c: Likewise.
2283 * testsuite/libffi.call/cls_uint.c: Likewise.
2284 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2285 * testsuite/libffi.call/cls_ushort.c: Likewise.
2286 * testsuite/libffi.call/nested_struct.c: Likewise.
2287 * testsuite/libffi.call/nested_struct1.c: Likewise.
2288 * testsuite/libffi.call/problem1.c: Likewise.
2289 * testsuite/libffi.special/unwindtest.cc: Likewise.
2290 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2292 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2294 * src/powerpc/aix.S: Cleanup whitespaces.
2295 * src/powerpc/aix_closure.S: Likewise.
2297 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2299 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2300 * src/powerpc/darwin_closure.S: Likewise.
2301 * src/powerpc/ffi_darwin.c: Likewise.
2303 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2304 David Edelsohn <edelsohn@gnu.org>
2306 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2307 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2309 * src/powerpc/darwin_closure.S: Likewise.
2310 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2311 according to the Darwin/AIX ABI.
2312 (ffi_prep_cif_machdep): Likewise.
2313 (ffi_closure_helper_DARWIN): Likewise.
2314 Remove the outgoing parameter stack logic. Simplify the evaluation
2315 of the different CASE types.
2316 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2317 statement in the trampoline code.
2319 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2321 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2322 for the register size.
2323 (ffi_closure_helper_SYSV): Handle the structure return value
2325 (ffi_closure_helper_SYSV): Return the appropriate type when
2326 the registers are used for the structure return value.
2327 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2328 the 64-bit return value. Update copyright years.
2330 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2332 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2333 srcdir for ffi_mips.h.
2335 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2337 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2339 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2340 placement of float result.
2341 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2342 cast of "resp" for big-endian 64 bit machines.
2344 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2346 * src/types.c (double, longdouble): Merge identical SH and ARM
2347 typedefs, and add POWERPC64.
2348 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2349 struct split over gpr and rest.
2350 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2351 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2353 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2355 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2358 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2360 * configure: Regenerate.
2362 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2364 * Makefile.am: Remove build rules for ffitest.
2365 * Makefile.in: Rebuilt.
2367 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2369 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2370 about implicit declaration of abort().
2372 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2374 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2375 * Makefile.in: Rebuilt.
2376 * configure.in: Add dejagnu test framework.
2377 * configure: Rebuilt.
2379 * testsuite/Makefile.am: New file.
2380 * testsuite/Makefile.in: Built
2381 * testsuite/lib/libffi-dg.exp: New file.
2382 * testsuite/config/default.exp: Likewise.
2383 * testsuite/libffi.call/call.exp: Likewise.
2384 * testsuite/libffi.call/ffitest.h: Likewise.
2385 * testsuite/libffi.call/closure_fn0.c: Likewise.
2386 * testsuite/libffi.call/closure_fn1.c: Likewise.
2387 * testsuite/libffi.call/closure_fn2.c: Likewise.
2388 * testsuite/libffi.call/closure_fn3.c: Likewise.
2389 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2390 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2391 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2392 * testsuite/libffi.call/cls_2byte.c: Likewise.
2393 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2394 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2395 * testsuite/libffi.call/cls_4byte.c: Likewise.
2396 * testsuite/libffi.call/cls_5byte.c: Likewise.
2397 * testsuite/libffi.call/cls_6byte.c: Likewise.
2398 * testsuite/libffi.call/cls_7byte.c: Likewise.
2399 * testsuite/libffi.call/cls_8byte.c: Likewise.
2400 * testsuite/libffi.call/cls_12byte.c: Likewise.
2401 * testsuite/libffi.call/cls_16byte.c: Likewise.
2402 * testsuite/libffi.call/cls_20byte.c: Likewise.
2403 * testsuite/libffi.call/cls_24byte.c: Likewise.
2404 * testsuite/libffi.call/cls_double.c: Likewise.
2405 * testsuite/libffi.call/cls_float.c: Likewise.
2406 * testsuite/libffi.call/cls_uchar.c: Likewise.
2407 * testsuite/libffi.call/cls_uint.c: Likewise.
2408 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2409 * testsuite/libffi.call/cls_ushort.c: Likewise.
2410 * testsuite/libffi.call/float.c: Likewise.
2411 * testsuite/libffi.call/float1.c: Likewise.
2412 * testsuite/libffi.call/float2.c: Likewise.
2413 * testsuite/libffi.call/many.c: Likewise.
2414 * testsuite/libffi.call/many_win32.c: Likewise.
2415 * testsuite/libffi.call/nested_struct.c: Likewise.
2416 * testsuite/libffi.call/nested_struct1.c: Likewise.
2417 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2418 * testsuite/libffi.call/problem1.c: Likewise.
2419 * testsuite/libffi.call/promotion.c: Likewise.
2420 * testsuite/libffi.call/return_ll.c: Likewise.
2421 * testsuite/libffi.call/return_sc.c: Likewise.
2422 * testsuite/libffi.call/return_uc.c: Likewise.
2423 * testsuite/libffi.call/strlen.c: Likewise.
2424 * testsuite/libffi.call/strlen_win32.c: Likewise.
2425 * testsuite/libffi.call/struct1.c: Likewise.
2426 * testsuite/libffi.call/struct2.c: Likewise.
2427 * testsuite/libffi.call/struct3.c: Likewise.
2428 * testsuite/libffi.call/struct4.c: Likewise.
2429 * testsuite/libffi.call/struct5.c: Likewise.
2430 * testsuite/libffi.call/struct6.c: Likewise.
2431 * testsuite/libffi.call/struct7.c: Likewise.
2432 * testsuite/libffi.call/struct8.c: Likewise.
2433 * testsuite/libffi.call/struct9.c: Likewise.
2434 * testsuite/libffi.special/special.exp: New file.
2435 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2436 * testsuite/libffi.special/unwindtest.cc: Likewise.
2439 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2441 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2444 2003-08-02 Alan Modra <amodra@bigpond.net.au>
2446 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2448 (ffi_closure_helper_LINUX64): Likewise.
2449 * src/powerpc/linux64.S: Remove code writing to parm save area.
2450 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2451 address in lr from ffi_closure_helper_LINUX64 call to calculate
2452 table address. Optimize function tail.
2454 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2456 * src/sparc/ffi.c: Handle all floating point registers.
2457 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2459 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2461 * README: Note that libffi is not part of GCC. Update the project
2464 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2466 * src/powerpc/ppc_closure.S: Include ffi.h.
2468 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2470 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2471 Use C style comments.
2473 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2475 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2476 * Makefile.in: Regenerate.
2477 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2478 * configure: Regenerate.
2479 * include/ffi.h.in: Add SHmedia support.
2480 * src/sh64/ffi.c: New file.
2481 * src/sh64/sysv.S: New file.
2483 2003-05-16 Jakub Jelinek <jakub@redhat.com>
2485 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2486 should be read-only.
2487 * configure: Rebuilt.
2488 * fficonfig.h.in: Rebuilt.
2489 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2490 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2491 * src/powerpc/linux64.S: Likewise.
2492 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2493 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2494 if -fpic/-fPIC/-mrelocatable.
2495 * src/powerpc/powerpc_closure.S: Likewise.
2496 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2497 #write in .eh_frame flags.
2498 * src/sparc/v9.S: Likewise.
2499 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2500 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2501 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2503 2003-05-07 Jeff Sturm <jsturm@one-point.com>
2505 Fixes PR bootstrap/10656
2506 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2507 support for .register pseudo-op.
2508 * src/sparc/v8.S: Use it.
2509 * fficonfig.h.in: Rebuilt.
2510 * configure: Rebuilt.
2512 2003-04-18 Jakub Jelinek <jakub@redhat.com>
2514 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2515 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2516 Make it the default on POWERPC64.
2517 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2518 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2519 * configure: Rebuilt.
2520 * src/powerpc/ffi.c (hidden): Define.
2521 (ffi_prep_args_SYSV): Renamed from
2522 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2523 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2524 ASM_NEEDS_REGISTERS64): New.
2525 (ffi_prep_args64): New function.
2526 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2527 (ffi_call): Likewise.
2528 (ffi_prep_closure): Likewise.
2529 (flush_icache): Surround by #ifndef POWERPC64.
2530 (ffi_dblfl): New union type.
2531 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2532 (ffi_closure_helper_LINUX64): New function.
2533 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2535 * src/powerpc/sysv.S: Likewise.
2536 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2537 * src/powerpc/linux64.S: New file.
2538 * src/powerpc/linux64_closure.S: New file.
2539 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2540 src/powerpc/linux64_closure.S.
2541 (TARGET_SRC_POWERPC): Likewise.
2543 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2544 closure_test_fn3): Fix result printing on big-endian 64-bit
2546 (main): Print tst2_arg instead of uninitialized tst2_result.
2548 * src/ffitest.c (main): Hide what closure pointer really points to
2551 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
2553 * configure.in (arm-*-netbsdelf*): Add configuration.
2554 (configure): Regenerated.
2556 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
2558 * include/Makefile.in: Regenerate.
2560 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2562 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2564 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2565 Receive closure pointer through parameter, read args using
2566 __builtin_dwarf_cfa.
2567 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2569 2003-03-12 Andreas Schwab <schwab@suse.de>
2571 * configure.in: Avoid trailing /. in toolexeclibdir.
2572 * configure: Rebuilt.
2574 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2576 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
2578 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2580 * libffi/src/powerpc/darwin_closure.S:
2581 Fix alignement bug, allocate 8 bytes for the result.
2582 * libffi/src/powerpc/aix_closure.S:
2584 * libffi/src/powerpc/ffi_darwin.c:
2585 Update stackframe description for aix/darwin_closure.S.
2587 2003-02-06 Jakub Jelinek <jakub@redhat.com>
2589 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2592 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2593 Andreas Schwab <schwab@suse.de>
2595 * configure.in: Adjust command to source config-ml.in to account
2596 for changes to the libffi_basedir definition.
2597 (libffi_basedir): Remove ${srcdir} from value and include trailing
2600 * configure: Regenerate.
2602 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2604 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2606 2003-01-28 Andrew Haley <aph@redhat.com>
2608 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2609 * src/x86/ffi64.c (ffi_prep_closure): New.
2610 (ffi_closure_UNIX64_inner): New.
2611 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2613 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
2615 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2616 Remove USE_LIBDIR conditional.
2617 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2618 * Makefile.in, configure: Rebuilt.
2620 2003-01027 David Edelsohn <edelsohn@gnu.org>
2622 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2623 * Makefile.in: Regenerate.
2625 2003-01-22 Andrew Haley <aph@redhat.com>
2627 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2630 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2632 * src/powerpc/darwin.S: Add unwind info.
2633 * src/powerpc/darwin_closure.S: Likewise.
2635 2003-01-14 Andrew Haley <aph@redhat.com>
2637 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2638 (ffi_prep_cif_machdep): Likewise.
2639 * src/x86/unix64.S: Add unwind info.
2641 2003-01-14 Andreas Jaeger <aj@suse.de>
2643 * src/ffitest.c (main): Only use ffi_closures if those are
2646 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2648 * libffi/src/ffitest.c
2649 add closure testcases
2651 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2653 * libffi/src/powerpc/ffi.c
2654 fix alignment bug for float (4 byte aligned iso 8 byte)
2656 2003-01-09 Geoffrey Keating <geoffk@apple.com>
2658 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2659 * src/powerpc/darwin.S: Remove RCS version string.
2661 2003-01-03 Jeff Sturm <jsturm@one-point.com>
2663 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2664 * src/ffitest.c (main): Use static storage for closure.
2665 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2666 * src/sparc/v8.S (ffi_closure_v8): New.
2667 * src/sparc/v9.S (ffi_closure_v9): New.
2669 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2671 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2672 enumeration for X86_WIN32.
2673 * src/x86/win32.S: Added ffi_call_STDCALL function
2675 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2676 switch cases for recognising FFI_STDCALL and
2677 calling ffi_call_STDCALL if target is X86_WIN32.
2678 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2679 stdcall versions of the "my_strlen" and "many"
2680 test functions (for X86_WIN32).
2681 Added test cases to test stdcall invocation using
2684 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2686 * src/sh/sysv.S: Add DWARF2 unwind info.
2688 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2690 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2692 2002-11-26 Jim Wilson <wilson@redhat.com>
2694 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2696 2002-11-23 H.J. Lu <hjl@gnu.org>
2698 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2699 Include ../config/accross.m4.
2700 * aclocal.m4; Rebuild.
2701 * configure: Likewise.
2703 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2705 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2707 2002-11-11 DJ Delorie <dj@redhat.com>
2709 * configure.in: Look for common files in the right place.
2711 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2713 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2714 raw data as _Jv_word values, not ffi_raw.
2715 (ffi_java_ptrarray_to_raw): Likewise.
2716 (ffi_java_rvalue_to_raw): New function.
2717 (ffi_java_raw_call): Call it.
2718 (ffi_java_raw_to_rvalue): New function.
2719 (ffi_java_translate_args): Call it.
2720 * src/ffitest.c (closure_test_fn): Interpret return value
2721 as ffi_arg, not int.
2722 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2723 FFI_TYPE_POINTER case.
2724 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2725 values extended to word size.
2727 2002-10-02 Andreas Jaeger <aj@suse.de>
2729 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2731 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
2733 * include/ffi.h.in: Fix i386 win32 compilation.
2735 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2737 * configure.in: Add s390x-*-linux-* target.
2738 * configure: Regenerate.
2739 * include/ffi.h.in: Define S390X for s390x targets.
2740 (FFI_CLOSURES): Define for s390/s390x.
2741 (FFI_TRAMPOLINE_SIZE): Likewise.
2742 (FFI_NATIVE_RAW_API): Likewise.
2743 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2744 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
2745 * src/s390/ffi.c: Major rework of existing code. Add support for
2746 s390x targets. Add closure support.
2747 * src/s390/sysv.S: Likewise.
2749 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
2751 * src/arm/sysv.S: Fix typo.
2753 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
2755 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2756 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2757 (ffi_call_SYSV): Handle soft-float.
2759 2002-09-27 Bo Thorsen <bo@suse.de>
2761 * include/ffi.h.in: Fix multilib x86-64 support.
2763 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2765 * Makefile.am (all-multi): Fix multilib parallel build.
2767 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2769 * configure.in (sh[34]*-*-linux*): Add brackets.
2770 * configure: Regenerate.
2772 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2774 * Makefile.am: Add SH support.
2775 * Makefile.in: Regenerate.
2776 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2777 * configure: Regenerate.
2778 * include/ffi.h.in: Add SH support.
2779 * src/sh/ffi.c: New file.
2780 * src/sh/sysv.S: New file.
2781 * src/types.c: Add SH support.
2783 2002-07-16 Bo Thorsen <bo@suse.de>
2785 * src/x86/ffi64.c: New file that adds x86-64 support.
2786 * src/x86/unix64.S: New file that handles argument setup for
2788 * src/x86/sysv.S: Don't use this on x86-64.
2789 * src/x86/ffi.c: Don't use this on x86-64.
2791 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2793 * src/ffitest.c (struct6): New test that tests a special case in
2795 (struct7): Likewise.
2796 (struct8): Likewise.
2797 (struct9): Likewise.
2798 (closure_test_fn): Silence warning about this when it's not used.
2799 (main): Add the new tests.
2800 (main): Fix a couple of wrong casts and silence some compiler warnings.
2801 * include/ffi.h.in: Add x86-64 ABI definition.
2802 * fficonfig.h.in: Regenerate.
2803 * Makefile.am: Add x86-64 support.
2804 * configure.in: Likewise.
2805 * Makefile.in: Regenerate.
2806 * configure: Likewise.
2808 2002-06-24 Bo Thorsen <bo@suse.de>
2810 * src/types.c: Merge settings for similar architectures.
2811 Add x86-64 sizes and alignments.
2813 2002-06-23 Bo Thorsen <bo@suse.de>
2815 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
2816 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
2817 * src/mips/ffi.c (ffi_prep_args): Likewise.
2818 * src/m68k/ffi.c (ffi_prep_args): Likewise.
2820 2002-07-18 H.J. Lu (hjl@gnu.org)
2822 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
2823 (libffi_la_SOURCES): Support MIPS_LINUX.
2824 (libffi_convenience_la_SOURCES): Likewise.
2825 * Makefile.in: Regenerated.
2827 * configure.in (mips64*-*): Skip.
2828 (mips*-*-linux*): New.
2829 * configure: Regenerated.
2831 * src/mips/ffi.c: Include <sgidefs.h>.
2833 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
2835 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
2837 2002-05-27 Roger Sayle <roger@eyesopen.com>
2839 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
2841 2002-05-27 Bo Thorsen <bo@suse.de>
2843 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
2846 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
2848 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
2849 beginning of function (for older apple cc).
2851 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
2853 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
2854 script entry, and set LD to it when configuring multilibs.
2855 * configure: Rebuilt.
2857 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
2859 * configure.in (sparc64-*-netbsd*): Add target.
2860 (sparc-*-netbsdelf*): Likewise.
2861 * configure: Regenerate.
2863 2002-04-28 David S. Miller <davem@redhat.com>
2865 * configure.in, configure: Fix SPARC test in previous change.
2867 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
2869 * Makefile.am: Add Linux for S/390 support.
2870 * Makefile.in: Regenerate.
2871 * configure.in: Add Linux for S/390 support.
2872 * configure: Regenerate.
2873 * include/ffi.h.in: Add Linux for S/390 support.
2874 * src/s390/ffi.c: New file from libffi CVS tree.
2875 * src/s390/sysv.S: New file from libffi CVS tree.
2877 2002-04-28 Jakub Jelinek <jakub@redhat.com>
2879 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
2881 * src/sparc/v8.S: Use it.
2882 * src/sparc/v9.S: Likewise.
2883 * fficonfig.h.in: Rebuilt.
2884 * configure: Rebuilt.
2886 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
2888 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
2890 * src/ia64/unix.S: Add unwind information. Fix comments.
2891 Save sp in a way that's compatible with unwind info.
2892 (ffi_call_unix): Correctly restore sp in all cases.
2893 * src/ia64/ffi.c: Add, fix comments.
2895 2002-04-08 Jakub Jelinek <jakub@redhat.com>
2897 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
2899 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
2901 * configure.in (alpha*-*-netbsd*): Add target.
2902 * configure: Regenerate.
2904 2002-04-04 Jeff Sturm <jsturm@one-point.com>
2906 * src/sparc/v8.S: Add unwind info.
2907 * src/sparc/v9.S: Likewise.
2909 2002-03-30 Krister Walfridsson <cato@df.lth.se>
2911 * configure.in: Enable i*86-*-netbsdelf*.
2912 * configure: Rebuilt.
2914 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
2917 * src/mips/n32.s: Delete
2918 * src/mips/o32.s: Delete
2920 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
2922 * configure.in: Enable alpha*-*-freebsd*.
2923 * configure: Rebuilt.
2925 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
2927 * Makefile.am: libfficonvenience -> libffi_convenience.
2928 * Makefile.in: Rebuilt.
2930 * Makefile.am: Define ffitest_OBJECTS.
2931 * Makefile.in: Rebuilt.
2933 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
2934 David Edelsohn <edelsohn@gnu.org>
2936 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
2937 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
2938 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
2939 * Makefile.in: Regenerate.
2940 * include/ffi.h.in: Add AIX and Darwin closure definitions.
2941 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
2942 (flush_icache, flush_range): New functions.
2943 (ffi_closure_helper_DARWIN): New function.
2944 * src/powerpc/aix_closure.S: New file.
2945 * src/powerpc/darwin_closure.S: New file.
2947 2002-02-24 Jeff Sturm <jsturm@one-point.com>
2949 * include/ffi.h.in: Add typedef for ffi_arg.
2950 * src/ffitest.c (main): Declare rint with ffi_arg.
2952 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
2954 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
2955 number of GPRs for floating-point arguments.
2957 2002-01-31 Anthony Green <green@redhat.com>
2959 * configure: Rebuilt.
2960 * configure.in: Replace CHECK_SIZEOF and endian tests with
2961 cross-compiler friendly macros.
2962 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
2965 2002-01-18 David Edelsohn <edelsohn@gnu.org>
2967 * src/powerpc/darwin.S (_ffi_call_AIX): New.
2968 * src/powerpc/aix.S (ffi_call_DARWIN): New.
2970 2002-01-17 David Edelsohn <edelsohn@gnu.org>
2972 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
2973 (TARGET_SRC_POWERPC_AIX): New.
2974 (POWERPC_AIX): New stanza.
2975 * Makefile.in: Regenerate.
2976 * configure.in: Add AIX case.
2977 * configure: Regenerate.
2978 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
2979 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
2980 size. Fix "long double" support.
2981 (ffi_call): Add FFI_AIX case.
2982 * src/powerpc/aix.S: New.
2984 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
2986 Implement Darwin PowerPC ABI.
2987 * configure.in: Handle powerpc-*-darwin*.
2988 * Makefile.am: Set source files for POWERPC_DARWIN.
2989 * configure: Rebuilt.
2990 * Makefile.in: Rebuilt.
2991 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
2993 * src/powerpc/darwin.S: New file.
2994 * src/powerpc/ffi_darwin.c: New file.
2996 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
2998 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3000 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3002 * src/x86/sysv.S: Avoid gas-only .balign directive.
3003 Use C style comments.
3005 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3007 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3008 Fixes PR bootstrap/3563.
3010 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3012 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3014 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3016 * configure.in: Recognize sparc*-sun-* host.
3017 * configure: Regenerate.
3019 2001-06-06 Andrew Haley <aph@redhat.com>
3021 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
3023 2001-06-03 Andrew Haley <aph@redhat.com>
3025 * src/alpha/osf.S: Add unwind info.
3026 * src/powerpc/sysv.S: Add unwind info.
3027 * src/powerpc/ppc_closure.S: Likewise.
3029 2000-05-31 Jeff Sturm <jsturm@one-point.com>
3031 * configure.in: Fix AC_ARG_ENABLE usage.
3032 * configure: Rebuilt.
3034 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3036 * configure.in: Remove warning about beta code.
3037 * configure: Rebuilt.
3039 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3041 * src/ia64/unix.S: Restore stack pointer when returning from
3043 * src/ia64/ffi.c: Fix typo in comment.
3045 2001-04-18 Jim Wilson <wilson@redhat.com>
3047 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3048 to eliminate RAW DV.
3050 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3052 * Makefile.am: Make a libtool convenience library.
3053 * Makefile.in: Rebuilt.
3055 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3057 * configure.in: Use different syntax for subdirectory creation.
3058 * configure: Rebuilt.
3060 2001-03-27 Jon Beniston <jon@beniston.com>
3062 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3063 * configure: Rebuilt.
3064 * Makefile.am: Added X86_WIN32 target support.
3065 * Makefile.in: Rebuilt.
3067 * include/ffi.h.in: Added X86_WIN32 target support.
3069 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3070 * src/types.c: Added X86_WIN32 target support.
3072 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3073 stuff removed and made to work with CygWin's gas.
3075 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3077 * configure.in: Make target subdirectory in build dir.
3078 * Makefile.am: Override suffix based rules to specify correct output
3080 * Makefile.in: Rebuilt.
3081 * configure: Rebuilt.
3083 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3085 * src/powerpc/ppc_closure.S: New file.
3086 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3087 involving long long and register pairs.
3088 (ffi_prep_closure): New function.
3089 (flush_icache): Likewise.
3090 (ffi_closure_helper_SYSV): Likewise.
3091 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3092 (FFI_TRAMPOLINE_SIZE): Likewise.
3093 (FFI_NATIVE_RAW_API): Likewise.
3094 * Makefile.in: Rebuilt.
3095 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3096 (TARGET_SRC_POWERPC): Likewise.
3098 2001-03-19 Tom Tromey <tromey@redhat.com>
3100 * Makefile.in: Rebuilt.
3101 * Makefile.am (ffitest_LDFLAGS): New macro.
3103 2001-03-02 Nick Clifton <nickc@redhat.com>
3105 * include/ffi.h.in: Remove RCS ident string.
3106 * include/ffi_mips.h: Remove RCS ident string.
3107 * src/debug.c: Remove RCS ident string.
3108 * src/ffitest.c: Remove RCS ident string.
3109 * src/prep_cif.c: Remove RCS ident string.
3110 * src/types.c: Remove RCS ident string.
3111 * src/alpha/ffi.c: Remove RCS ident string.
3112 * src/alpha/osf.S: Remove RCS ident string.
3113 * src/arm/ffi.c: Remove RCS ident string.
3114 * src/arm/sysv.S: Remove RCS ident string.
3115 * src/mips/ffi.c: Remove RCS ident string.
3116 * src/mips/n32.S: Remove RCS ident string.
3117 * src/mips/o32.S: Remove RCS ident string.
3118 * src/sparc/ffi.c: Remove RCS ident string.
3119 * src/sparc/v8.S: Remove RCS ident string.
3120 * src/sparc/v9.S: Remove RCS ident string.
3121 * src/x86/ffi.c: Remove RCS ident string.
3122 * src/x86/sysv.S: Remove RCS ident string.
3124 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3126 * include/ffi.h.in: Change sourceware.cygnus.com references to
3129 2000-12-09 Richard Henderson <rth@redhat.com>
3131 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3132 (ffi_closure_osf_inner): Index rather than increment avalue
3133 and arg_types. Give ffi_closure_osf the raw return value type.
3134 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3137 2000-12-07 Richard Henderson <rth@redhat.com>
3139 * src/raw_api.c (ffi_translate_args): Fix typo.
3140 (ffi_prep_closure): Likewise.
3142 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3143 FFI_TRAMPOLINE_SIZE.
3144 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3145 cif->bytes for new ffi_call_osf implementation.
3146 (ffi_prep_args): Absorb into ...
3147 (ffi_call): ... here. Do all stack allocation here and
3148 avoid a callback function.
3149 (ffi_prep_closure, ffi_closure_osf_inner): New.
3150 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3151 (ffi_closure_osf): New.
3153 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3155 * config.guess, config.sub, install-sh: Removed.
3156 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3157 * Makefile.in: Rebuilt.
3159 * acinclude.m4: Include libtool macros from the top level.
3160 * aclocal.m4, configure: Rebuilt.
3162 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3164 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3165 * configure: Rebuilt.
3167 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3169 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3170 memory correctly. Use conditional instructions, not branches where
3173 2000-05-04 Tom Tromey <tromey@cygnus.com>
3175 * configure: Rebuilt.
3176 * configure.in: Match `arm*-*-linux-*'.
3177 From Chris Dornan <cdornan@arm.com>.
3179 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3181 * Makefile.am (SUBDIRS): Define.
3182 (AM_MAKEFLAGS): Likewise.
3183 (Multilib support.): Add section.
3184 * Makefile.in: Rebuilt.
3185 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3186 New variables. Set for gcc using -print-multi-lib. Export them
3188 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3189 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3191 (extra_compiler_flags_value, extra_compiler_flags): Check these
3192 for extra compiler options which need to be passed down in
3195 2000-04-16 Anthony Green <green@redhat.com>
3197 * configure: Rebuilt.
3198 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3200 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3202 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3203 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3204 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3205 Replace all void * sizeofs with sizeof(int).
3206 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3207 different than DOUBLE.
3208 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3209 (ffi_prep_args_v9): New function.
3210 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3211 (ffi_V9_return_struct): New function.
3212 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3213 32bit code (not yet cross-arch calls).
3214 * src/sparc/v8.S: Add struct return delay nop.
3216 * src/sparc/v9.S: New file.
3217 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3218 is used on sparc64 only for structures larger than 32 bytes.
3219 Pass by reference for structures is done for structure arguments
3220 larger than 16 bytes.
3221 * src/ffitest.c (main): Use 64bit rint on sparc64.
3222 Run long long tests on sparc.
3223 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3225 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3227 * configure.in (sparc-*-linux*): New supported target.
3228 (sparc64-*-linux*): Likewise.
3229 * configure: Rebuilt.
3230 * Makefile.am: Add v9.S to SPARC files.
3231 * Makefile.in: Likewise.
3232 (LINK): Surround $(CCLD) into double quotes, so that multilib
3233 compiles work correctly.
3235 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3237 * configure: Rebuilt.
3238 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3239 proposed by Bryce McKinlay.
3241 2000-03-20 Tom Tromey <tromey@cygnus.com>
3243 * Makefile.in: Hand edit for java_raw_api.lo.
3245 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3247 * config.guess, config.sub: Update from the gcc tree.
3248 Fix for PR libgcj/168.
3250 2000-03-03 Tom Tromey <tromey@cygnus.com>
3252 * Makefile.in: Fixed ia64 by hand.
3254 * configure: Rebuilt.
3255 * configure.in (--enable-multilib): New option.
3256 (libffi_basedir): New subst.
3257 (AC_OUTPUT): Added multilib code.
3259 2000-03-02 Tom Tromey <tromey@cygnus.com>
3261 * Makefile.in: Rebuilt.
3262 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3265 2000-02-25 Hans Boehm <boehm@acm.org>
3267 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3269 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3271 (ffi_prep_raw_closure): Use ffi_translate_args, not
3272 ffi_closure_translate.
3273 * src/java_raw_api.c: New file.
3274 * src/ffitest.c (closure_test_fn): New function.
3275 (main): Define `rint' as long long on IA64. Added new test when
3276 FFI_CLOSURES is defined.
3277 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3278 (ffi_abi): Recognize IA64.
3279 (ffi_raw): Added `flt' field.
3280 Added "Java raw API" code.
3281 * configure.in: Recognize ia64.
3282 * Makefile.am (TARGET_SRC_IA64): New macro.
3283 (libffi_la_common_SOURCES): Added java_raw_api.c.
3284 (libffi_la_SOURCES): Define in IA64 case.
3286 2000-01-04 Tom Tromey <tromey@cygnus.com>
3288 * Makefile.in: Rebuilt with newer automake.
3290 1999-12-31 Tom Tromey <tromey@cygnus.com>
3292 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3294 1999-09-01 Tom Tromey <tromey@cygnus.com>
3296 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3298 * fficonfig.h.in: Rebuilt.
3299 * configure: Rebuilt.
3300 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3301 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3302 * acconfig.h: Don't #undef PACKAGE or VERSION.
3304 1999-08-09 Anthony Green <green@cygnus.com>
3306 * include/ffi.h.in: Try to work around messy header problem
3307 with PACKAGE and VERSION.
3309 * configure: Rebuilt.
3310 * configure.in: Change version to 2.00-beta.
3312 * fficonfig.h.in: Rebuilt.
3313 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3315 * src/x86/ffi.c (ffi_raw_call): Rename.
3317 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3319 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3320 (ffi_prep_incoming_args_SYSV): Ditto.
3321 (ffi_prep_closure): Ditto.
3322 (ffi_closure_raw_SYSV): Ditto.
3323 (ffi_prep_raw_closure): More ditto.
3324 (ffi_call_raw): Final ditto.
3326 * include/ffi.h.in: Add definitions for closure and raw API.
3328 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3331 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3333 * src/raw_api.c: New file.
3335 * include/ffi.h.in (ffi_raw): New type.
3336 (UINT_ARG, SINT_ARG): New defines.
3337 (ffi_closure, ffi_raw_closure): New types.
3338 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3340 * configure.in: Add check for endianness and sizeof void*.
3342 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3343 instead of directly.
3345 * configure: Rebuilt.
3347 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3349 * configure.in: Add x86 and powerpc BeOS configurations.
3350 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3352 1999-05-09 Anthony Green <green@cygnus.com>
3354 * configure.in: Add warning about this being beta code.
3355 Remove src/Makefile.am from the picture.
3356 * configure: Rebuilt.
3358 * Makefile.am: Move logic from src/Makefile.am. Add changes
3359 to support libffi as a target library.
3360 * Makefile.in: Rebuilt.
3362 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3363 Upgraded to new autoconf, automake, libtool.
3367 * LICENSE: Update copyright date.
3369 * src/Makefile.am, src/Makefile.in: Removed.
3371 1998-11-29 Anthony Green <green@cygnus.com>
3373 * include/ChangeLog: Removed.
3374 * src/ChangeLog: Removed.
3375 * src/mips/ChangeLog: Removed.
3376 * src/sparc/ChangeLog: Remboved.
3377 * src/x86/ChangeLog: Removed.
3379 * ChangeLog.v1: Created.