1 2009-06-04 Andrew Haley <aph@redhat.com>
3 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
6 2008-02-26 Anthony Green <green@redhat.com>
7 Thomas Heller <theller@ctypes.org>
9 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
12 2008-02-03 Timothy Wall <twall@users.sf.net>
14 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
15 offset based on code pointer, not data pointer.
17 2008-01-31 Timothy Wall <twall@users.sf.net>
19 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
21 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
23 * src/x86/win32.S: Add assembly for stdcall closure.
24 * src/x86/ffi.c: Initialize stdcall closure trampoline.
26 2009-06-04 Andrew Haley <aph@redhat.com>
28 * include/ffi.h.in: Change void (*)() to void (*)(void).
29 * src/x86/ffi.c: Likewise.
31 2009-06-04 Andrew Haley <aph@redhat.com>
33 * src/powerpc/ppc_closure.S: Insert licence header.
34 * src/powerpc/linux64_closure.S: Likewise.
35 * src/m68k/sysv.S: Likewise.
37 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
38 * src/powerpc/ffi.c: Likewise.
39 * src/powerpc/ffi_darwin.c: Likewise.
40 * src/m32r/ffi.c: Likewise.
41 * src/sh64/ffi.c: Likewise.
42 * src/x86/ffi64.c: Likewise.
43 * src/alpha/ffi.c: Likewise.
44 * src/alpha/osf.S: Likewise.
45 * src/frv/ffi.c: Likewise.
46 * src/s390/ffi.c: Likewise.
47 * src/pa/ffi.c: Likewise.
48 * src/pa/hpux32.S: Likewise.
49 * src/ia64/unix.S: Likewise.
50 * src/ia64/ffi.c: Likewise.
51 * src/sparc/ffi.c: Likewise.
52 * src/mips/ffi.c: Likewise.
53 * src/sh/ffi.c: Likewise.
55 2008-02-15 David Daney <ddaney@avtrex.com>
57 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
58 Define (conditionally), and use it to include cachectl.h.
59 (ffi_prep_closure_loc): Fix cache flushing.
60 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
62 2009-06-04 Andrew Haley <aph@redhat.com>
68 src/powerpc/ffitarget.h,
81 src/alpha/ffitarget.h,
97 src/ia64/ia64_flags.h,
101 src/sparc/ffitarget.h,
104 src/mips/ffitarget.h,
111 src/sh/sysv.S: Update license text.
113 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
115 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
116 (.eh_frame): Add FDE for it.
118 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
120 * configure.ac: Also check if assembler supports pc-relative
121 relocs on X86_WIN32 targets.
122 * configure: Regenerate.
123 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
124 (_ffi_call_SYSV): Add missing function type symbol .def and
125 add EH markup labels.
126 (_ffi_call_STDCALL): Likewise.
127 (_ffi_closure_SYSV): Likewise.
128 (_ffi_closure_raw_SYSV): Likewise.
129 (.eh_frame): Add hand-crafted EH data.
131 2009-04-09 Jakub Jelinek <jakub@redhat.com>
133 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
134 version 3 of the GNU General Public License and to point readers
135 at the COPYING3 file and the FSF's license web page.
136 * testsuite/libffi.call/call.exp: Likewise.
137 * testsuite/libffi.special/special.exp: Likewise.
139 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
141 * configure: Regenerate.
143 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
146 * configure.ac (HAVE_AS_X86_PCREL): New test.
147 * configure: Regenerate.
148 * fficonfig.h.in: Regenerate.
149 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
150 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
151 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
152 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
153 * src/x86/unix64.S (.Lstore_table): Move to .text section.
154 (.Lload_table): Likewise.
155 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
157 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
159 * configure: Regenerate.
161 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
163 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
164 signed/unsigned int8/16 return values.
165 * src/sparc/v8.S (ffi_call_v8): Likewise.
166 (ffi_closure_v8): Likewise.
168 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
169 Steve Ellcey <sje@cup.hp.com>
171 * configure: Regenerate for new libtool.
172 * Makefile.in: Ditto.
173 * include/Makefile.in: Ditto.
176 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
178 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
179 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
180 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
181 Adjust copyright notice.
182 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
183 register or two register to use for FFI_SYSV structs.
184 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
185 (ffi_closure_helper_SYSV): Fix the return type for
186 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
187 Adjust copyright notice.
189 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
191 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
194 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
196 * configure: Regenerate.
197 * include/Makefile.in: Regenerate.
198 * testsuite/Makefile.in: Regenerate.
200 2008-06-07 Joseph Myers <joseph@codesourcery.com>
202 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
203 powerpc-*-beos*): Remove.
204 * configure: Regenerate.
206 2008-05-09 Julian Brown <julian@codesourcery.com>
208 * Makefile.am (LTLDFLAGS): New.
209 (libffi_la_LDFLAGS): Use above.
210 * Makefile.in: Regenerate.
212 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
215 * aclocal.m4: Regenerate.
216 * configure: Regenerate.
218 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
220 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
221 * src/sh64/sysv.S: Likewise.
223 2008-03-26 Daniel Jacobowitz <dan@debian.org>
225 * src/arm/sysv.S: Fix ARM comment marker.
227 2008-03-26 Jakub Jelinek <jakub@redhat.com>
229 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
230 * src/s390/sysv.S: Likewise.
231 * src/powerpc/ppc_closure.S: Likewise.
232 * src/powerpc/sysv.S: Likewise.
233 * src/x86/unix64.S: Likewise.
234 * src/x86/sysv.S: Likewise.
235 * src/sparc/v8.S: Likewise.
236 * src/sparc/v9.S: Likewise.
237 * src/m68k/sysv.S: Likewise.
238 * src/arm/sysv.S: Likewise.
240 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
242 * aclocal.m4: Regenerate.
243 * configure: Likewise.
244 * Makefile.in: Likewise.
245 * include/Makefile.in: Likewise.
246 * testsuite/Makefile.in: Likewise.
248 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
249 Andreas Tobler <a.tobler@schweiz.org>
251 * configure.ac: Add amd64-*-freebsd* target.
252 * configure: Regenerate.
254 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
257 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
260 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
263 2008-01-24 David Edelsohn <edelsohn@gnu.org>
265 * configure: Regenerate.
267 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
269 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
271 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
274 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
275 signed/unsigned int8/16 for X86_DARWIN.
276 Updated copyright info.
277 Handle one and two byte structs with special cif->flags.
278 * src/x86/ffitarget.h: Add special types for one and two byte structs.
279 Updated copyright info.
280 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
282 Remove code to pop args from the stack after call.
283 Special-case signed/unsigned for int8/16, one and two byte structs.
284 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
285 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
287 Updated copyright info.
289 2007-12-08 David Daney <ddaney@avtrex.com>
291 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
292 SUBU, add with ADDU and use smaller code sequences.
294 2007-12-07 David Daney <ddaney@avtrex.com>
296 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
299 2007-12-06 David Daney <ddaney@avtrex.com>
301 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
303 (ffi_java_raw): New typedef.
304 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
305 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
307 (ffi_java_raw_closure) : Same.
308 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
310 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
312 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
313 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
314 sizeof(ffi_java_raw) for alignment calculations.
315 (ffi_java_ptrarray_to_raw): Same.
316 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
317 if FFI_SIZEOF_JAVA_RAW == 4.
318 (ffi_java_raw_to_rvalue): Same.
319 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
320 (ffi_java_translate_args): Same.
321 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
323 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
325 2007-12-06 David Daney <ddaney@avtrex.com>
327 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
330 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
333 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
334 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
335 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
336 set the NUM_FPR_ARG_REGISTERS according to.
337 Add support for potential soft-float support under hard-float
339 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
340 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
341 to the FFI_LINUX_SOFT_FLOAT ABI.
342 (ffi_prep_cif_machdep): Likewise.
343 (ffi_closure_helper_SYSV): Likewise.
344 * src/powerpc/ppc_closure.S: Make sure not to store float/double
345 on archs where __NO_FPRS__ is true.
346 Add FFI_TYPE_UINT128 support.
347 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
348 Adjust copyright notice.
350 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
352 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
353 * include/ffi_common.h: ... here.
356 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
358 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
359 * src/powerpc/linux64.S: Likewise.
360 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
361 * testsuite/libffi.call/return_ldl.c: New test.
363 2007-09-04 <aph@redhat.com>
365 * src/arm/sysv.S (UNWIND): New.
366 (Whole file): Conditionally compile unwinder directives.
367 * src/arm/sysv.S: Add unwinder directives.
369 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
370 Only treat r0 as a struct address if we're actually returning a
372 Only copy the bytes that are actually within a struct.
373 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
374 is returned in r0, not passed by address.
375 (ffi_call): Allocate a word-sized temporary for the case where
376 a composite is returned in r0.
377 (ffi_prep_incoming_args_SYSV): Align as necessary.
379 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
381 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
382 directly using the sys_cacheflush syscall.
384 2007-07-27 Andrew Haley <aph@redhat.com>
386 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
388 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
390 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
391 * configure.ac: Likewise.
392 * Makefile.in: Regenerate.
393 * include/Makefile.in: Likewise.
394 * testsuite/Makefile.in: Likewise.
395 * configure: Likewise.
397 2007-08-24 David Daney <ddaney@avtrex.com>
399 * testsuite/libffi.call/return_sl.c: New test.
401 2007-08-10 David Daney <ddaney@avtrex.com>
403 * testsuite/libffi.call/cls_multi_ushort.c,
404 testsuite/libffi.call/cls_align_uint16.c,
405 testsuite/libffi.call/nested_struct1.c,
406 testsuite/libffi.call/nested_struct3.c,
407 testsuite/libffi.call/cls_7_1_byte.c,
408 testsuite/libffi.call/nested_struct5.c,
409 testsuite/libffi.call/cls_double.c,
410 testsuite/libffi.call/nested_struct7.c,
411 testsuite/libffi.call/cls_sint.c,
412 testsuite/libffi.call/nested_struct9.c,
413 testsuite/libffi.call/cls_20byte1.c,
414 testsuite/libffi.call/cls_multi_sshortchar.c,
415 testsuite/libffi.call/cls_align_sint64.c,
416 testsuite/libffi.call/cls_3byte2.c,
417 testsuite/libffi.call/cls_multi_schar.c,
418 testsuite/libffi.call/cls_multi_uchar.c,
419 testsuite/libffi.call/cls_19byte.c,
420 testsuite/libffi.call/cls_9byte1.c,
421 testsuite/libffi.call/cls_align_float.c,
422 testsuite/libffi.call/closure_fn1.c,
423 testsuite/libffi.call/problem1.c,
424 testsuite/libffi.call/closure_fn3.c,
425 testsuite/libffi.call/cls_sshort.c,
426 testsuite/libffi.call/closure_fn5.c,
427 testsuite/libffi.call/cls_align_double.c,
428 testsuite/libffi.call/nested_struct.c,
429 testsuite/libffi.call/cls_2byte.c,
430 testsuite/libffi.call/nested_struct10.c,
431 testsuite/libffi.call/cls_4byte.c,
432 testsuite/libffi.call/cls_6byte.c,
433 testsuite/libffi.call/cls_8byte.c,
434 testsuite/libffi.call/cls_multi_sshort.c,
435 testsuite/libffi.call/cls_align_sint16.c,
436 testsuite/libffi.call/cls_align_uint32.c,
437 testsuite/libffi.call/cls_20byte.c,
438 testsuite/libffi.call/cls_float.c,
439 testsuite/libffi.call/nested_struct2.c,
440 testsuite/libffi.call/cls_5_1_byte.c,
441 testsuite/libffi.call/nested_struct4.c,
442 testsuite/libffi.call/cls_24byte.c,
443 testsuite/libffi.call/nested_struct6.c,
444 testsuite/libffi.call/cls_64byte.c,
445 testsuite/libffi.call/nested_struct8.c,
446 testsuite/libffi.call/cls_uint.c,
447 testsuite/libffi.call/cls_multi_ushortchar.c,
448 testsuite/libffi.call/cls_schar.c,
449 testsuite/libffi.call/cls_uchar.c,
450 testsuite/libffi.call/cls_align_uint64.c,
451 testsuite/libffi.call/cls_ulonglong.c,
452 testsuite/libffi.call/cls_align_longdouble.c,
453 testsuite/libffi.call/cls_1_1byte.c,
454 testsuite/libffi.call/cls_12byte.c,
455 testsuite/libffi.call/cls_3_1byte.c,
456 testsuite/libffi.call/cls_3byte1.c,
457 testsuite/libffi.call/cls_4_1byte.c,
458 testsuite/libffi.call/cls_6_1_byte.c,
459 testsuite/libffi.call/cls_16byte.c,
460 testsuite/libffi.call/cls_18byte.c,
461 testsuite/libffi.call/closure_fn0.c,
462 testsuite/libffi.call/cls_9byte2.c,
463 testsuite/libffi.call/closure_fn2.c,
464 testsuite/libffi.call/closure_fn4.c,
465 testsuite/libffi.call/cls_ushort.c,
466 testsuite/libffi.call/closure_fn6.c,
467 testsuite/libffi.call/cls_5byte.c,
468 testsuite/libffi.call/cls_align_pointer.c,
469 testsuite/libffi.call/cls_7byte.c,
470 testsuite/libffi.call/cls_align_sint32.c,
471 testsuite/libffi.special/unwindtest_ffi_call.cc,
472 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
474 2007-08-10 David Daney <ddaney@avtrex.com>
477 * configure.ac: Don't treat mips64 as a special case.
478 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
479 * configure: Regenerate
480 * Makefile.in: Ditto.
481 * fficonfig.h.in: Ditto.
482 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
483 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
484 (FFI_DEFAULT_ABI): Set for n64 case.
485 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
486 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
487 (ffi_closure_N32): New function.
488 (.eh_frame): New section
489 * src/mips/o32.S: Clean up comments.
490 (ffi_closure_O32): Pass ffi_closure parameter in $12.
491 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
492 _MIPS_SIM == _ABIN32 throughout.
493 (FFI_MIPS_STOP_HERE): New, use in place of
495 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
496 to support n32/n64 ABIs.
497 (calc_n32_struct_flags): Rewrite.
498 (calc_n32_return_struct_flags): Remove unused variable. Reverse
499 position of flag bits.
500 (ffi_prep_cif_machdep): Rewrite n32 portion.
501 (ffi_call): Enable for n64. Add special handling for small structure
503 (ffi_prep_closure_loc): Add n32 and n64 support.
504 (ffi_closure_mips_inner_O32): Add cast to silence warning.
505 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
507 2007-08-08 David Daney <ddaney@avtrex.com>
509 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
510 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
512 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
513 * testsuite/libffi.call/cls_sint.c (main): Ditto.
514 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
515 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
516 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
517 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
518 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
519 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
520 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
521 * testsuite/libffi.call/return_fl3.c (main): Ditto.
522 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
523 * testsuite/libffi.call/nested_struct.c (main): Ditto.
524 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
525 * testsuite/libffi.call/return_ll1.c (main): Ditto.
526 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
527 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
528 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
529 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
530 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
531 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
532 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
533 * testsuite/libffi.call/cls_uint.c (main): Ditto.
534 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
535 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
536 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
537 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
538 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
539 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
540 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
541 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
542 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
544 2007-08-07 Andrew Haley <aph@redhat.com>
546 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
549 2007-08-06 Andrew Haley <aph@redhat.com>
552 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
553 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
556 2007-08-02 David Daney <ddaney@avtrex.com>
558 * testsuite/libffi.call/return_ul.c (main): Define return type as
559 ffi_arg. Use proper printf conversion specifier.
561 2007-07-30 Andrew Haley <aph@redhat.com>
564 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
565 signed/unsigned int8/16.
566 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
568 Remove code to pop args from the stack after call.
569 Special-case signed/unsigned int8/16.
570 * testsuite/libffi.call/return_sc.c (main): Revert.
572 2007-07-26 Richard Guenther <rguenther@suse.de>
575 * testsuite/libffi.call/return_sc.c (main): Verify call
576 result as signed char, not ffi_arg.
578 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
580 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
581 * configure: Regenerate.
583 2007-07-11 David Daney <ddaney@avtrex.com>
585 * src/mips/ffi.c: Don't include sys/cachectl.h.
586 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
589 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
591 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
592 from (ffi_prep_closure): ... this.
593 (FFI_INIT_TRAMPOLINE): Adjust.
595 2005-12-31 Phil Blundell <pb@reciva.com>
597 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
598 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
599 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
600 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
601 (FFI_CLOSURES): Enable closure support.
603 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
605 * testsuite/libffi.call/cls_multi_ushort.c,
606 testsuite/libffi.call/cls_align_uint16.c,
607 testsuite/libffi.call/nested_struct1.c,
608 testsuite/libffi.call/nested_struct3.c,
609 testsuite/libffi.call/cls_7_1_byte.c,
610 testsuite/libffi.call/cls_double.c,
611 testsuite/libffi.call/nested_struct5.c,
612 testsuite/libffi.call/nested_struct7.c,
613 testsuite/libffi.call/cls_sint.c,
614 testsuite/libffi.call/nested_struct9.c,
615 testsuite/libffi.call/cls_20byte1.c,
616 testsuite/libffi.call/cls_multi_sshortchar.c,
617 testsuite/libffi.call/cls_align_sint64.c,
618 testsuite/libffi.call/cls_3byte2.c,
619 testsuite/libffi.call/cls_multi_schar.c,
620 testsuite/libffi.call/cls_multi_uchar.c,
621 testsuite/libffi.call/cls_19byte.c,
622 testsuite/libffi.call/cls_9byte1.c,
623 testsuite/libffi.call/cls_align_float.c,
624 testsuite/libffi.call/closure_fn1.c,
625 testsuite/libffi.call/problem1.c,
626 testsuite/libffi.call/closure_fn3.c,
627 testsuite/libffi.call/cls_sshort.c,
628 testsuite/libffi.call/closure_fn5.c,
629 testsuite/libffi.call/cls_align_double.c,
630 testsuite/libffi.call/cls_2byte.c,
631 testsuite/libffi.call/nested_struct.c,
632 testsuite/libffi.call/nested_struct10.c,
633 testsuite/libffi.call/cls_4byte.c,
634 testsuite/libffi.call/cls_6byte.c,
635 testsuite/libffi.call/cls_8byte.c,
636 testsuite/libffi.call/cls_multi_sshort.c,
637 testsuite/libffi.call/cls_align_uint32.c,
638 testsuite/libffi.call/cls_align_sint16.c,
639 testsuite/libffi.call/cls_float.c,
640 testsuite/libffi.call/cls_20byte.c,
641 testsuite/libffi.call/cls_5_1_byte.c,
642 testsuite/libffi.call/nested_struct2.c,
643 testsuite/libffi.call/cls_24byte.c,
644 testsuite/libffi.call/nested_struct4.c,
645 testsuite/libffi.call/nested_struct6.c,
646 testsuite/libffi.call/cls_64byte.c,
647 testsuite/libffi.call/nested_struct8.c,
648 testsuite/libffi.call/cls_uint.c,
649 testsuite/libffi.call/cls_multi_ushortchar.c,
650 testsuite/libffi.call/cls_schar.c,
651 testsuite/libffi.call/cls_uchar.c,
652 testsuite/libffi.call/cls_align_uint64.c,
653 testsuite/libffi.call/cls_ulonglong.c,
654 testsuite/libffi.call/cls_align_longdouble.c,
655 testsuite/libffi.call/cls_1_1byte.c,
656 testsuite/libffi.call/cls_12byte.c,
657 testsuite/libffi.call/cls_3_1byte.c,
658 testsuite/libffi.call/cls_3byte1.c,
659 testsuite/libffi.call/cls_4_1byte.c,
660 testsuite/libffi.call/cls_6_1_byte.c,
661 testsuite/libffi.call/cls_16byte.c,
662 testsuite/libffi.call/cls_18byte.c,
663 testsuite/libffi.call/closure_fn0.c,
664 testsuite/libffi.call/cls_9byte2.c,
665 testsuite/libffi.call/closure_fn2.c,
666 testsuite/libffi.call/closure_fn4.c,
667 testsuite/libffi.call/cls_ushort.c,
668 testsuite/libffi.call/closure_fn6.c,
669 testsuite/libffi.call/cls_5byte.c,
670 testsuite/libffi.call/cls_align_pointer.c,
671 testsuite/libffi.call/cls_7byte.c,
672 testsuite/libffi.call/cls_align_sint32.c,
673 testsuite/libffi.special/unwindtest_ffi_call.cc,
674 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
676 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
678 * aclocal.m4: Regenerated.
680 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
682 * configure: Regenerate.
684 2007-05-23 Steve Ellcey <sje@cup.hp.com>
686 * Makefile.in: Regenerate.
687 * configure: Regenerate.
688 * aclocal.m4: Regenerate.
689 * include/Makefile.in: Regenerate.
690 * testsuite/Makefile.in: Regenerate.
692 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
694 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
695 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
696 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
697 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
698 (FFI_CLOSURES): Enable closure support.
700 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
702 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
703 * configure: Regenerate.
704 * fficonfig.h.in: Regenerate.
705 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
706 CFI_OFFSET,CFI_DEF_CFA): New macros.
707 (ffi_call_SYSV): Add callframe annotation.
709 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
711 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
712 numerous test suite failures.
713 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
715 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
717 * Makefile.am (EXTRA_DIST): Bring up to date.
718 * Makefile.in: Regenerate.
719 * src/frv/eabi.S: Remove RCS keyword.
721 2007-04-06 Richard Henderson <rth@redhat.com>
723 * configure.ac: Tidy target case.
724 (HAVE_LONG_DOUBLE): Allow the target to override.
725 * configure: Regenerate.
726 * include/ffi.h.in: Don't define ffi_type_foo if
727 LIBFFI_HIDE_BASIC_TYPES is defined.
728 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
730 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
731 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
732 (ffi_type_longdouble): Special case for Alpha. Don't define
733 if long double == double.
735 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
736 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
737 (ffi_call, ffi_closure_osf_inner): Likewise.
738 (ffi_closure_osf_inner): Likewise. Mark hidden.
739 (ffi_call_osf, ffi_closure_osf): Mark hidden.
740 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
741 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
742 (load_table): Handle 128-bit long double.
744 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
746 2007-04-06 Tom Tromey <tromey@redhat.com>
749 * README: Fixed bug in example.
751 2007-04-03 Jakub Jelinek <jakub@redhat.com>
753 * src/closures.c: Include sys/statfs.h.
754 (_GNU_SOURCE): Define on Linux.
755 (FFI_MMAP_EXEC_SELINUX): Define.
756 (selinux_enabled): New variable.
757 (selinux_enabled_check): New function.
758 (is_selinux_enabled): Define.
761 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
763 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
764 Use 'volatile float sum' to create sum of floats to avoid false
765 negative due to excess precision on ix86 targets.
768 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
770 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
772 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
774 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
776 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
777 (ffi_prep_closure_loc): New.
778 (ffi_prep_raw_closure_loc): New.
779 (ffi_prep_java_raw_closure_loc): New.
780 * src/closures.c: New file.
781 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
782 Replace sflags with exec_offset.
783 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
784 sub_segment_exec_offset): New macros.
785 (get_segment_flags, set_segment_flags, check_segment_merge): New
787 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
788 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
789 destroy_mspace): Use new macros.
790 (sys_alloc): Silence warning.
791 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
792 * Makefile.in: Rebuilt.
793 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
794 terms of ffi_prep_closure_loc.
795 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
797 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
799 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
801 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
803 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
804 (ffi_prep_closure): ... this.
805 * src/pa/ffi.c: Likewise.
806 * src/cris/ffi.c: Likewise. Adjust.
807 * src/frv/ffi.c: Likewise.
808 * src/ia64/ffi.c: Likewise.
809 * src/mips/ffi.c: Likewise.
810 * src/powerpc/ffi_darwin.c: Likewise.
811 * src/s390/ffi.c: Likewise.
812 * src/sh/ffi.c: Likewise.
813 * src/sh64/ffi.c: Likewise.
814 * src/sparc/ffi.c: Likewise.
815 * src/x86/ffi64.c: Likewise.
816 * src/x86/ffi.c: Likewise.
817 (FFI_INIT_TRAMPOLINE): Adjust.
818 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
819 (ffi_prep_raw_closure): ... this.
820 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
821 (ffi_prep_closure): ... this.
822 (flush_icache): Adjust.
824 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
826 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
829 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
831 * Makefile.am: Add dummy install-pdf target.
832 * Makefile.in: Regenerate
834 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
836 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
837 ffi_closure_helper_SYSV): Add long double handling.
839 2007-02-02 Jakub Jelinek <jakub@redhat.com>
841 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
842 immediately after bctrl instruction.
844 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
846 * Makefile.am (all-recursive, install-recursive,
847 mostlyclean-recursive, clean-recursive, distclean-recursive,
848 maintainer-clean-recursive): Add missing targets.
849 * Makefile.in: Rebuilt.
851 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
853 * configure.ac: Add TARGET for x86_64-*-darwin*.
854 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
856 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
857 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
858 * configure: Regenerate.
859 * Makefile.in: Regenerate.
860 * include/Makefile.in: Regenerate.
861 * testsuite/Makefile.in: Regenerate.
862 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
865 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
867 * aclocal.m4: Regenerate with aclocal -I .. as written in the
870 2006-10-31 Geoffrey Keating <geoffk@apple.com>
872 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
873 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
875 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
876 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
878 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
879 Sandro Tolaini <tolaini@libero.it>
881 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
883 * configure: Regenerated.
884 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
885 (EXTRA_DIST): Add src/x86/darwin.S.
886 * Makefile.in: Regenerated.
887 * include/Makefile.in: Regenerated.
888 * testsuite/Makefile.in: Regenerated.
890 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
891 X86_WIN32, and additionally align stack to 16 bytes.
892 * src/x86/darwin.S: New, based on sysv.S.
893 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
895 2006-09-12 David Daney <ddaney@avtrex.com>
898 * include/Makefile.am: Install both ffi.h and ffitarget.h in
899 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
900 * aclocal.m4: Regenerated for automake 1.9.6.
901 * Makefile.in: Regenerated.
902 * include/Makefile.in: Regenerated.
903 * testsuite/Makefile.in: Regenerated.
905 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
907 * include/ffi_common.h (struct): Revert accidental commit.
909 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
911 * include/ffi_common.h: Remove lint directives.
912 * include/ffi.h.in: Likewise.
914 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
916 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
917 for 32-bit architectures.
918 * testsuite/libffi.call/return_ul.c: New test case.
920 2006-07-19 David Daney <ddaney@avtrex.com>
922 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
923 xfail remains for mips64.
925 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
927 * Makefile.am: Add install-html target. Add install-html to .PHONY
928 * Makefile.in: Regenerate.
929 * aclocal.m4: Regenerate.
930 * include/Makefile.in: Regenerate.
931 * testsuite/Makefile.in: Regenerate.
933 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
935 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
938 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
940 * README: Remove notice about 'Crazy Comments'.
941 * src/debug.c: Remove lint directives. Cleanup white spaces.
942 * src/java_raw_api.c: Likewise.
943 * src/prep_cif.c: Likewise.
944 * src/raw_api.c: Likewise.
945 * src/ffitest.c: Delete. No longer needed, all test cases migrated
947 * src/arm/ffi.c: Remove lint directives.
948 * src/m32r/ffi.c: Likewise.
949 * src/pa/ffi.c: Likewise.
950 * src/powerpc/ffi.c: Likewise.
951 * src/powerpc/ffi_darwin.c: Likewise.
952 * src/sh/ffi.c: Likewise.
953 * src/sh64/ffi.c: Likewise.
954 * src/x86/ffi.c: Likewise.
955 * testsuite/libffi.call/float2.c: Likewise.
956 * testsuite/libffi.call/promotion.c: Likewise.
957 * testsuite/libffi.call/struct1.c: Likewise.
959 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
961 * src/pa/hpux32.S: Correct unwind offset calculation for
963 * src/pa/linux.S: Likewise.
965 2006-04-12 James E Wilson <wilson@specifix.com>
968 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
969 (hfa_type_load): Call stf_spill.
970 (hfa_type_store): Call ldf_fill.
971 (ffi_call): Adjust calls to above routines. Add local temps for
974 2006-04-10 Matthias Klose <doko@debian.org>
976 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
977 directory names containing underscores.
979 2006-04-07 James E Wilson <wilson@specifix.com>
981 * testsuite/libffi.call/float4.c: New testcase.
983 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
984 Andreas Tobler <a.tobler@schweiz.ch>
986 * Makefile.am: Add PA_HPUX port.
987 * Makefile.in: Regenerate.
988 * include/Makefile.in: Likewise.
989 * testsuite/Makefile.in: Likewise.
990 * configure.ac: Add PA_HPUX rules.
991 * configure: Regenerate.
992 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
993 Add PA_HPUX and PA64_HPUX.
994 Rename FFI_LINUX ABI to FFI_PA32 ABI.
995 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
996 (FFI_TYPE_SMALL_STRUCT2): Define.
997 (FFI_TYPE_SMALL_STRUCT4): Likewise.
998 (FFI_TYPE_SMALL_STRUCT8): Likewise.
999 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1000 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1001 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1002 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1003 * src/pa/ffi.c (ROUND_DOWN): Delete.
1004 (fldw, fstw, fldd, fstd): Use '__asm__'.
1005 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1006 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1007 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1008 Simplify incrementing of stack slot variable. Change type of local
1009 'n' to unsigned int.
1010 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1012 (ffi_prep_cif_machdep): Likewise.
1013 (ffi_call): Likewise.
1014 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1015 return type to ffi_status. Simplify incrementing of stack slot
1016 variable. Only copy floating point argument registers when PA_LINUX
1017 is true. Reformat debug statement.
1018 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1019 FFI_TYPE_SMALL_STRUCT8.
1020 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1022 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1023 Add nops to cache flush. Add trampoline for PA_HPUX.
1024 * src/pa/hpux32.S: New file.
1025 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1026 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1027 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1028 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1029 argument type checks so that common argument types appear first.
1030 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1031 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1033 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1035 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1036 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1037 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1038 FFI_TYPE_LONGDOUBLE.
1039 (ffi_prep_args64): Assert using IBM extended double.
1040 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1041 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1042 (ffi_call): Handle FFI_LINUX.
1043 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1044 gpr3 return pointer as for struct return. Handle FFI_LINUX
1045 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1047 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1048 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1049 Don't use r6 as pointer to results, instead use sp offset. Don't
1050 make a special call to load lr with case table address, instead
1051 use offset from previous call.
1052 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1053 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1056 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1058 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1059 passed with FP registers correctly.
1060 (ffi_closure_helper_SYSV): Likewise.
1061 * src/sh64/sysv.S: Likewise.
1063 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1065 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1066 args and userdata unused.
1067 (closure_test_fn1): Mark cif and userdata unused.
1068 (main): Remove unused res.
1070 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1072 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1073 -O2, -O3, -Os and the warning flags -W -Wall.
1074 * testsuite/libffi.special/special.exp: Likewise.
1075 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1076 unused parameter unused for gcc or else do nothing.
1077 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1078 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1079 and userdata unused.
1080 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1081 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1082 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1083 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1084 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1085 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1086 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1087 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1088 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1089 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1090 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1091 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1092 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1093 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1094 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1095 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1096 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1097 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1098 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1099 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1100 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1101 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1102 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1103 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1105 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1107 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1109 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1110 void* to avoid compiler warning.
1112 (cls_struct_align_gn): Mark cif and userdata unused.
1113 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1115 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1117 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1119 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1121 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1123 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1124 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1125 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1127 (main): Cast res_call to silence gcc.
1128 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1130 (main): Cast res_call to silence gcc.
1131 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1133 (main): Cast res_call to silence gcc.
1134 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1136 (main): Cast res_call to silence gcc.
1137 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1139 (main): Cast res_call to silence gcc.
1140 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1142 (main): Cast res_call to silence gcc.
1143 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1145 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1146 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1148 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1149 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1151 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1152 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1154 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1155 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1157 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1158 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1159 and userdata unused.
1160 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1162 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1163 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1164 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1165 Cleanup white spaces.
1166 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1167 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1168 cif and userdata unused.
1169 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1171 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1172 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1173 formatters to silence gcc.
1174 (B_gn): Mark cif and userdata unused.
1175 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1177 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1178 (B_gn): Mark cif and userdata unused.
1179 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1181 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1182 (B_gn): Mark cif and userdata unused.
1183 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1185 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1186 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1187 * testsuite/libffi.call/problem1.c (stub): Likewise.
1188 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1190 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1191 in the last commit for this test case in the test case itself.
1192 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1194 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1195 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1196 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1197 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1198 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1199 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1201 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1203 * src/sh/sysv.S: Fix register numbers in the FDE for
1206 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1208 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1209 declaration to avoid a false negative on ix86. See PR323.
1211 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1213 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1214 and cast integer to void * if needed. Update the pointer to
1215 the FP register saved area correctly.
1217 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1219 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1221 * testsuite/libffi.call/nested_struct4.c: Likewise.
1223 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1225 * testsuite/libffi.call/return_dbl.c: New test case.
1226 * testsuite/libffi.call/return_dbl1.c: Likewise.
1227 * testsuite/libffi.call/return_dbl2.c: Likewise.
1228 * testsuite/libffi.call/return_fl.c: Likewise.
1229 * testsuite/libffi.call/return_fl1.c: Likewise.
1230 * testsuite/libffi.call/return_fl2.c: Likewise.
1231 * testsuite/libffi.call/return_fl3.c: Likewise.
1232 * testsuite/libffi.call/closure_fn6.c: Likewise.
1234 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1236 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1237 here to be used by other test cases too.
1239 * testsuite/libffi.call/nested_struct10.c: New test case.
1240 * testsuite/libffi.call/nested_struct9.c: Likewise.
1241 * testsuite/libffi.call/nested_struct8.c: Likewise.
1242 * testsuite/libffi.call/nested_struct7.c: Likewise.
1243 * testsuite/libffi.call/nested_struct6.c: Likewise.
1244 * testsuite/libffi.call/nested_struct5.c: Likewise.
1245 * testsuite/libffi.call/nested_struct4.c: Likewise.
1247 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1249 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1250 * configure: Rebuilt.
1252 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1254 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1255 instead do the shifting inline.
1256 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1257 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1258 and 8 byte structs, for the remaining struct sizes don't call
1259 __lshrdi3, instead do the shifting inline.
1261 2005-12-07 Thiemo Seufer <ths@networkno.de>
1263 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1264 missing parentheses.
1265 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1266 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1267 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1268 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1269 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1271 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1273 (ffi_prep_closure): Improve trampoline instruction scheduling.
1274 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1276 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1278 * src/powerpc/ffi.c: Formatting.
1279 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1280 (ffi_prep_args64): Likewise.
1282 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1284 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1285 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1287 2005-09-26 Tom Tromey <tromey@redhat.com>
1289 * testsuite/libffi.call/float1.c (value_type): New typedef.
1290 (CANARY): New define.
1291 (main): Check for result buffer overflow.
1292 * src/powerpc/linux64.S: Handle linux64 long double returns.
1293 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1294 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1296 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1299 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1301 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1303 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1305 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1306 (AH_BOTTOM): Add FFI_HIDDEN definition.
1307 * configure: Rebuilt.
1308 * fficonfig.h.in: Rebuilt.
1309 * src/powerpc/ffi.c (hidden): Remove.
1310 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1311 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1312 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1313 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1314 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1315 add FFI_HIDDEN to its prototype.
1316 (ffi_closure_SYSV_inner): New.
1317 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1318 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1320 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1323 * configure: Rebuilt.
1324 * configure.ac: Handle i*86-*-gnu*.
1326 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1328 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1329 DW_CFA_offset_extended_sf rather than
1330 DW_CFA_GNU_negative_offset_extended.
1331 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1333 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1335 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1337 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1338 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1339 partially on register.
1340 (ffi_closure_helper_SYSV): Likewise.
1341 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1343 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1345 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1347 * src/sh64/ffi.c (simple_type): Remove.
1348 (return_type): Handle small structures correctly.
1349 (ffi_prep_args): Likewise.
1350 (ffi_call): Likewise.
1351 (ffi_closure_helper_SYSV): Likewise.
1352 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1353 Emit position independent code if PIC and remove wrong datalabel
1354 prefixes from EH data.
1356 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1358 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1359 * Makefile.in: Regenerate.
1360 * include/Makefile.in: Likewise.
1361 * testsuite/Makefile.in: Likewise.
1362 * configure.ac: Add POWERPC_FREEBSD rules.
1363 * configure: Regenerate.
1364 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1365 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1366 * src/powerpc/ffi.c: Add flags to handle small structure returns
1368 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1370 (ffi_closure_helper_SYSV): Likewise.
1371 * src/powerpc/ppc_closure.S: Add return types for small structures.
1372 * src/powerpc/sysv.S: Add bits to handle small structures for
1375 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1377 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1378 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1379 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1381 2005-07-05 Randolph Chung <tausq@debian.org>
1383 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1384 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1385 structures. Kill compilation warnings.
1386 (ffi_closure_inner_LINUX): Print return values as hex in debug
1387 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1388 Properly handle 5-7 byte structure returns.
1389 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1390 (FFI_TYPE_SMALL_STRUCT2): Remove.
1391 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1392 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1393 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1394 (checksmst1, checksmst2): Remove.
1395 (checksmst3): Optimize handling of 3-byte struct returns.
1396 (checksmst567): Properly handle 5-7 byte struct returns.
1398 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1401 * src/mips/n32.S: Enforce PIC code.
1402 * src/mips/o32.S: Likewise.
1404 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1406 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1407 * configure: Regenerate.
1409 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1411 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1412 to call ffi_closure_helper_SYSV. Append @local instead.
1413 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1415 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1417 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1418 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1419 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1420 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1421 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1423 2005-05-09 Mike Stump <mrs@apple.com>
1425 * configure: Regenerate.
1427 2005-05-08 Richard Henderson <rth@redhat.com>
1430 * src/alpha/osf.S: Update unwind into to match code.
1432 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1433 Richard Henderson <rth@redhat.com>
1435 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1437 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1438 (ffi_prep_closure): Set carry bit if sse-used flag set.
1439 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1440 Only load sse registers if ssecount non-zero.
1441 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1443 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1445 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1446 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1447 * configure: Regenerate.
1449 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1451 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1452 have Tcl8.3-compatible intermediate variable.
1454 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1455 Hans-Peter Nilsson <hp@axis.com>
1457 * Makefile.am: Add CRIS support.
1458 * configure.ac: Likewise.
1459 * Makefile.in, configure, testsuite/Makefile.in,
1460 include/Makefile.in: Regenerate.
1461 * src/cris: New directory.
1462 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1463 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1465 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1466 \r?\n in output tests.
1468 2005-04-12 Mike Stump <mrs@apple.com>
1470 * configure: Regenerate.
1472 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1474 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1476 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1478 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1480 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1481 to 64 bits in ILP32 mode.
1482 Load 64 bits even for short data.
1484 2005-03-23 Mike Stump <mrs@apple.com>
1486 * src/powerpc/darwin.S: Update for -m64 multilib.
1487 * src/powerpc/darwin_closure.S: Likewise.
1489 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1491 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1492 Do not set tool_include_dir.
1493 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1495 * include/Makefile.am: Set gcc_version and toollibffidir.
1496 * include/Makefile.in: Regenerate.
1498 2005-02-22 Andrew Haley <aph@redhat.com>
1500 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1501 odd-numbered register pairs for 64-bit integer types.
1503 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1506 * testsuite/libffi.call/return_ll1.c: New test case.
1508 2005-02-11 Janis Johnson <janis187@us.ibm.com>
1510 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1511 * testsuite/libffi.call/float.c: Ditto.
1512 * testsuite/libffi.call/float2.c: Ditto.
1513 * testsuite/libffi.call/float3.c: Ditto.
1515 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1517 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1519 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1521 * testsuite/libffi.special/special.exp (cxx_options): Add
1524 2004-12-31 Richard Henderson <rth@redhat.com>
1526 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1527 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1528 offset parameters with a type parameter; deduce size and structure
1529 alignment. Update all users.
1531 2004-12-31 Richard Henderson <rth@redhat.com>
1533 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1534 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
1535 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
1536 into ffi_prep_closure.
1537 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1540 2004-12-27 Richard Henderson <rth@redhat.com>
1542 * src/x86/unix64.S: Fix typo in unwind info.
1544 2004-12-25 Richard Henderson <rth@redhat.com>
1546 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1547 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1548 (merge_classes): Check for it.
1550 (classify_argument): Pass byte_offset by value; perform all updates
1552 (examine_argument): Add classes argument; handle
1553 X86_64_COMPLEX_X87_CLASS.
1554 (ffi_prep_args): Merge into ...
1555 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1556 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1557 (ffi_fill_return_value): Remove.
1558 (ffi_prep_closure): Remove dead assert.
1559 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1560 Rewrite to use struct register_args instead of va_list. Create
1561 flags for handling structure returns.
1562 * src/x86/unix64.S: Remove dead strings.
1563 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1564 stack frame with ffi_call. Handle structure returns properly.
1565 (float2sse, floatfloat2sse, double2sse): Remove.
1566 (sse2float, sse2double, sse2floatfloat): Remove.
1567 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1568 to handle structure returns properly.
1570 2004-12-08 David Edelsohn <edelsohn@gnu.org>
1572 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1574 * Makefile.in: Regenerated.
1576 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
1578 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1579 * configure, aclocal.m4, Makefile.in: Regenerate.
1580 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1582 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1584 * configure: Regenerate for libtool change.
1586 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1588 * configure: Regenerate for libtool reversion.
1590 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1592 * configure: Regenerate for libtool change.
1594 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1596 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1598 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
1600 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1601 of jal. Use an absolute encoding for the frame information.
1603 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1605 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1606 * acinclude.m4: Delete logic for sincludes.
1607 * aclocal.m4, Makefile.in, configure: Regenerate.
1608 * include/Makefile: Likewise.
1609 * testsuite/Makefile: Likewise.
1611 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1613 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1614 on a 8-byte boundary.
1615 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1617 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1619 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1620 long long values. Round stack allocation to a multiple of 8 bytes
1621 for ATPCS compatibility.
1622 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1623 names. Handle returning long long types. Add Thumb and interworking
1624 support. Improve soft-float code.
1626 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1628 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1629 (libffi_exit): New function.
1630 (libffi_init): Build the testglue wrapper if needed.
1632 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1635 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1637 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1639 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1641 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1643 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1644 * testsuite/libffi.call/float3.c: New test case.
1646 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1648 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1649 the function returning a structure pointed with R2.
1650 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1651 the structure return value if T bit set. Emit position
1652 independent code and EH data if PIC.
1654 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1656 * Makefile.am: Add m32r support.
1657 * configure.ac: Likewise.
1658 * Makefile.in: Regenerate.
1659 * confiugre: Regenerate.
1660 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1661 (uint64, sint64, double, longdouble)
1662 * src/m32r: New directory.
1663 * src/m32r/ffi.c: New file.
1664 * src/m32r/sysv.S: Likewise.
1665 * src/m32r/ffitarget.h: Likewise.
1667 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1669 * testsuite/libffi.call/negint.c: New test case.
1671 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1674 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1675 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1676 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1679 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1681 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1682 * testsuite/libffi.call/promotion.c: Likewise.
1683 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1685 * testsuite/libffi.call/return_sc.c: Likewise.
1686 * testsuite/libffi.call/return_uc.c: Likewise.
1688 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1690 * src/powerpc/darwin.S: Fix comments and identation.
1691 * src/powerpc/darwin_closure.S: Likewise.
1693 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1695 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1696 (ffi_prep_args): Handle longdouble arguments.
1697 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1699 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1700 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1702 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1703 * src/types.c: Defined longdouble size and alignment for darwin.
1705 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1707 * src/powerpc/aix.S: Remove whitespaces.
1708 * src/powerpc/aix_closure.S: Likewise.
1709 * src/powerpc/asm.h: Likewise.
1710 * src/powerpc/ffi.c: Likewise.
1711 * src/powerpc/ffitarget.h: Likewise.
1712 * src/powerpc/linux64.S: Likewise.
1713 * src/powerpc/linux64_closure.S: Likewise.
1714 * src/powerpc/ppc_closure.S: Likewise.
1715 * src/powerpc/sysv.S: Likewise.
1717 2004-08-30 Anthony Green <green@redhat.com>
1719 * Makefile.am: Add frv support.
1720 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1721 * configure.ac: Read configure.host.
1722 * configure.in: Read configure.host.
1723 * configure.host: New file. frv-elf needs libgloss.
1724 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
1725 alignment. This is needed to frv and shouldn't harm the others.
1726 * include/ffi_common.h (ALIGN_DOWN): New macro.
1727 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1729 2004-08-24 David Daney <daney@avtrex.com>
1731 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1732 * testsuite/libffi.call/closure_fn1.c: Likewise.
1733 * testsuite/libffi.call/closure_fn2.c Likewise.
1734 * testsuite/libffi.call/closure_fn3.c: Likewise.
1735 * testsuite/libffi.call/closure_fn4.c: Likewise.
1736 * testsuite/libffi.call/closure_fn5.c: Likewise.
1737 * testsuite/libffi.call/cls_18byte.c: Likewise.
1738 * testsuite/libffi.call/cls_19byte.c: Likewise.
1739 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1740 * testsuite/libffi.call/cls_20byte.c: Likewise.
1741 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1742 * testsuite/libffi.call/cls_24byte.c: Likewise.
1743 * testsuite/libffi.call/cls_2byte.c: Likewise.
1744 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1745 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1746 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1747 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1748 * testsuite/libffi.call/cls_4byte.c: Likewise.
1749 * testsuite/libffi.call/cls_64byte.c: Likewise.
1750 * testsuite/libffi.call/cls_6byte.c: Likewise.
1751 * testsuite/libffi.call/cls_7byte.c: Likewise.
1752 * testsuite/libffi.call/cls_8byte.c: Likewise.
1753 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1754 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1755 * testsuite/libffi.call/cls_align_double.c: Likewise.
1756 * testsuite/libffi.call/cls_align_float.c: Likewise.
1757 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1758 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1759 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1760 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1761 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1762 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1763 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1764 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1765 * testsuite/libffi.call/cls_double.c: Likewise.
1766 * testsuite/libffi.call/cls_float.c: Likewise.
1767 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1768 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1769 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1770 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1771 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1772 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1773 * testsuite/libffi.call/cls_schar.c: Likewise.
1774 * testsuite/libffi.call/cls_sint.c: Likewise.
1775 * testsuite/libffi.call/cls_sshort.c: Likewise.
1776 * testsuite/libffi.call/cls_uchar.c: Likewise.
1777 * testsuite/libffi.call/cls_uint.c: Likewise.
1778 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1779 * testsuite/libffi.call/cls_ushort.c: Likewise.
1780 * testsuite/libffi.call/nested_struct.c: Likewise.
1781 * testsuite/libffi.call/nested_struct1.c: Likewise.
1782 * testsuite/libffi.call/nested_struct2.c: Likewise.
1783 * testsuite/libffi.call/nested_struct3.c: Likewise.
1784 * testsuite/libffi.call/problem1.c: Likewise.
1785 * testsuite/libffi.special/unwindtest.cc: Likewise.
1786 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1788 * testsuite/libffi.call/cls_16byte.c: Likewise.
1789 * testsuite/libffi.call/cls_5byte.c: Likewise.
1791 2004-08-23 David Daney <daney@avtrex.com>
1794 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1795 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1796 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1797 parameters and return types.
1798 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1799 (ffi_prep_closure): Ditto.
1800 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1801 alignment calculations.
1802 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1803 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1805 2004-08-14 Casey Marshall <csm@gnu.org>
1807 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1808 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1809 integer (O32 ABI only).
1810 (ffi_prep_closure): new function.
1811 (ffi_closure_mips_inner_O32): new function.
1812 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
1813 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1814 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1815 64 bit integers correctly.
1816 (ffi_closure_O32): new function.
1817 Added DWARF-2 unwind info for both functions.
1819 2004-08-10 Andrew Haley <aph@redhat.com>
1821 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1823 2004-08-01 Robert Millan <robertmh@gnu.org>
1825 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1826 * configure: Regenerate.
1828 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1830 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1831 and mmap() explicitly instead of relying on preset autoconf cache
1833 * aclocal.m4: Regenerate.
1834 * configure: Regenerate.
1836 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1838 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1839 (ffi_check_float_struct): Remove unused prototype.
1841 2004-06-30 Geoffrey Keating <geoffk@apple.com>
1843 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1844 character on Darwin, use '\n\t' instead.
1846 2004-06-26 Matthias Klose <doko@debian.org>
1848 * libtool-version: Fix typo in revision/age.
1850 2004-06-17 Matthias Klose <doko@debian.org>
1852 * libtool-version: New.
1853 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1854 * Makefile.in: Regenerate.
1856 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1858 * Makefile.am: Remove useless multilib rules.
1859 * Makefile.in: Regenerate.
1860 * aclocal.m4: Regenerate with automake 1.8.5.
1861 * configure.ac: Remove useless multilib configury.
1862 * configure: Regenerate.
1864 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1866 * .cvsignore: New file.
1868 2004-06-10 Jakub Jelinek <jakub@redhat.com>
1870 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1872 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1875 2004-06-06 Sean McNeil <sean@mcneil.com>
1877 * configure.ac: Add x86_64-*-freebsd* support.
1878 * configure: Regenerate.
1880 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
1883 * configure.ac: Test for existence of mmap and sys/mman.h before
1884 checking blacklist. Fix suggested by Jim Wilson.
1885 * configure: Regenerate.
1887 2004-04-26 Matt Austern <austern@apple.com>
1889 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1891 * src/powerpc/darwin_closure.S: Likewise.
1893 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1895 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1896 error. Reported by Thomas Heller <theller@python.net>.
1897 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1898 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1900 2004-03-20 Matthias Klose <doko@debian.org>
1902 * src/pa/linux.S: Fix typo.
1904 2004-03-19 Matthias Klose <doko@debian.org>
1906 * Makefile.am: Update.
1907 * Makefile.in: Regenerate.
1908 * src/pa/ffi.h.in: Remove.
1909 * src/pa/ffitarget.h: New file.
1911 2004-02-10 Randolph Chung <tausq@debian.org>
1913 * Makefile.am: Add PA support.
1914 * Makefile.in: Regenerate.
1915 * include/Makefile.in: Regenerate.
1916 * configure.ac: Add PA target.
1917 * configure: Regenerate.
1918 * src/pa/ffi.c: New file.
1919 * src/pa/ffi.h.in: Add PA support.
1920 * src/pa/linux.S: New file.
1921 * prep_cif.c: Add PA support.
1923 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
1925 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
1927 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
1928 with ecif->cif->flags.
1929 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
1931 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
1932 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
1933 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
1934 sc_retstruct2b): Add for 1 or 2-bytes struct case.
1936 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
1938 * configure.in: Rename file to ...
1939 * configure.ac: ... this.
1940 * fficonfig.h.in: Regenerate.
1941 * Makefile.in: Regenerate.
1942 * include/Makefile.in: Regenerate.
1943 * testsuite/Makefile.in: Regenerate.
1945 2004-03-12 Matt Austern <austern@apple.com>
1947 * src/powerpc/darwin.S: Fix EH information so it corresponds to
1948 changes in EH format resulting from addition of linkonce support.
1949 * src/powerpc/darwin_closure.S: Likewise.
1951 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
1952 Paolo Bonzini <bonzini@gnu.org>
1954 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
1955 Remove VPATH. Remove rules for object files. Remove multilib support.
1956 (AM_CCASFLAGS): Add.
1957 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
1958 (AC_PREREQ): Bump version to 2.59.
1959 (AC_INIT): Fill with version info and bug address.
1960 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
1961 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
1962 De-precious CC so that the right flags are passed down to multilibs.
1963 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
1964 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
1965 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
1966 * configure: Rebuilt.
1967 * aclocal.m4: Likewise.
1968 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
1969 * fficonfig.h.in: Likewise.
1971 2004-03-11 Andreas Schwab <schwab@suse.de>
1973 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
1974 arguments from fp registers only for the first 8 parameter slots.
1975 Don't convert a float parameter when passed in memory.
1977 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
1979 * configure: Regenerate for config/accross.m4 correction.
1981 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
1983 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
1984 ecif->cif->bytes to bytes.
1985 (ffi_prep_cif_machdep): Add braces around nested if statement.
1987 2004-02-09 Alan Modra <amodra@bigpond.net.au>
1989 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
1991 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
1992 (ffi_closure_helper_LINUX64): Fix typo.
1993 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
1995 * testsuite/libffi.call/float.c: Likewise.
1996 * testsuite/libffi.call/float2.c: Likewise.
1998 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2000 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2001 long double function return and long double arg handling.
2002 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2003 Use "end_pfr" instead of "nf". Correct long double handling.
2005 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2007 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2008 space for long double return value. Adjust stack frame and offsets.
2009 Load f2 long double return.
2011 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2013 * src/types.c: Use 16 byte long double for POWERPC64.
2015 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2017 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2018 when the structure return address is passed in %o0.
2019 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2020 (ffi_v9_layout_struct): Align the field following a nested structure
2021 on a word boundary. Use memmove instead of memcpy.
2022 (ffi_call): Update call to ffi_V9_return_struct.
2023 (ffi_prep_closure): Define 'ctx' only for V8.
2024 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2025 and ffi_closure_sparc_inner_v9.
2026 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2027 Always skip the structure return address. For structures and long
2028 doubles, copy the argument directly.
2029 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2030 if required. Shift the maximum floating-point slot accordingly. For
2031 big structures, copy the argument directly; otherwise, left-justify the
2032 argument and call ffi_v9_layout_struct to lay out the structure on
2034 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2035 (ffi_closure_v8): Pass the structure return address. Update call to
2036 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2037 Skip the 'unimp' insn when returning long doubles and structures.
2038 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2039 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2040 FFI_TYPE_INT handling. Load structures both in integers and
2041 floating-point registers on return.
2042 * README: Update status of the SPARC port.
2044 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2046 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2048 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2050 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2052 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2053 value as of type ffi_arg, not unsigned int.
2055 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2057 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2060 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2062 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2065 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2067 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2070 2003-12-10 Richard Henderson <rth@redhat.com>
2072 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2073 size_t instead of int.
2075 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2077 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2078 * testsuite/libffi.call/many_win32.c (main): Replace variable
2079 int i with unsigned long ul.
2081 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2082 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2083 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2084 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2085 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2086 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2087 * testsuite/libffi.call/cls_align_float.c: Likewise.
2088 * testsuite/libffi.call/cls_align_double.c: Likewise.
2089 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2090 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2092 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2095 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2096 Align arguments to 32 bits.
2098 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2101 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2102 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2103 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2104 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2105 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2106 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2108 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2110 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2112 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2113 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2115 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2117 * Makefile.in: Rebuilt.
2118 * configure: Likewise.
2119 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2122 2003-11-21 Andreas Jaeger <aj@suse.de>
2123 Andreas Tobler <a.tobler@schweiz.ch>
2125 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2126 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2127 * Makefile.in: Rebuilt.
2128 * aclocal.m4: Likewise.
2129 * configure: Likewise.
2130 * fficonfig.h.in: Likewise.
2131 * testsuite/lib/libffi-dg.exp: Add include dir.
2132 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2133 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2134 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2135 for ffi_closure if available.
2136 * testsuite/libffi.call/closure_fn1.c: Likewise.
2137 * testsuite/libffi.call/closure_fn2.c: Likewise.
2138 * testsuite/libffi.call/closure_fn3.c: Likewise.
2139 * testsuite/libffi.call/closure_fn4.c: Likewise.
2140 * testsuite/libffi.call/closure_fn5.c: Likewise.
2141 * testsuite/libffi.call/cls_12byte.c: Likewise.
2142 * testsuite/libffi.call/cls_16byte.c: Likewise.
2143 * testsuite/libffi.call/cls_18byte.c: Likewise.
2144 * testsuite/libffi.call/cls_19byte.c: Likewise.
2145 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2146 * testsuite/libffi.call/cls_20byte.c: Likewise.
2147 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2148 * testsuite/libffi.call/cls_24byte.c: Likewise.
2149 * testsuite/libffi.call/cls_2byte.c: Likewise.
2150 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2151 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2152 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2153 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2154 * testsuite/libffi.call/cls_4byte.c: Likewise.
2155 * testsuite/libffi.call/cls_5byte.c: Likewise.
2156 * testsuite/libffi.call/cls_64byte.c: Likewise.
2157 * testsuite/libffi.call/cls_6byte.c: Likewise.
2158 * testsuite/libffi.call/cls_7byte.c: Likewise.
2159 * testsuite/libffi.call/cls_8byte.c: Likewise.
2160 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2161 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2162 * testsuite/libffi.call/cls_double.c: Likewise.
2163 * testsuite/libffi.call/cls_float.c: Likewise.
2164 * testsuite/libffi.call/cls_schar.c: Likewise.
2165 * testsuite/libffi.call/cls_sint.c: Likewise.
2166 * testsuite/libffi.call/cls_sshort.c: Likewise.
2167 * testsuite/libffi.call/cls_uchar.c: Likewise.
2168 * testsuite/libffi.call/cls_uint.c: Likewise.
2169 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2170 * testsuite/libffi.call/cls_ushort.c: Likewise.
2171 * testsuite/libffi.call/nested_struct.c: Likewise.
2172 * testsuite/libffi.call/nested_struct1.c: Likewise.
2173 * testsuite/libffi.call/nested_struct2.c: Likewise.
2174 * testsuite/libffi.call/nested_struct3.c: Likewise.
2175 * testsuite/libffi.call/problem1.c: Likewise.
2176 * testsuite/libffi.special/unwindtest.cc: Likewise.
2178 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2180 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2182 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2184 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2185 Add -lgcc_s to additional flags.
2187 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2189 * configure.in, include/Makefile.am: PR libgcj/11147, install
2190 the ffitarget.h header file in a gcc versioned and target
2192 * configure: Regenerated.
2193 * Makefile.in, include/Makefile.in: Likewise.
2194 * testsuite/Makefile.in: Likewise.
2196 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2198 * testsuite/libffi.call/closure_fn0.c: Print result and check
2199 with dg-output to make debugging easier.
2200 * testsuite/libffi.call/closure_fn1.c: Likewise.
2201 * testsuite/libffi.call/closure_fn2.c: Likewise.
2202 * testsuite/libffi.call/closure_fn3.c: Likewise.
2203 * testsuite/libffi.call/closure_fn4.c: Likewise.
2204 * testsuite/libffi.call/closure_fn5.c: Likewise.
2205 * testsuite/libffi.call/cls_12byte.c: Likewise.
2206 * testsuite/libffi.call/cls_16byte.c: Likewise.
2207 * testsuite/libffi.call/cls_18byte.c: Likewise.
2208 * testsuite/libffi.call/cls_19byte.c: Likewise.
2209 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2210 * testsuite/libffi.call/cls_20byte.c: Likewise.
2211 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2212 * testsuite/libffi.call/cls_24byte.c: Likewise.
2213 * testsuite/libffi.call/cls_2byte.c: Likewise.
2214 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2215 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2216 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2217 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2218 * testsuite/libffi.call/cls_4byte.c: Likewise.
2219 * testsuite/libffi.call/cls_5byte.c: Likewise.
2220 * testsuite/libffi.call/cls_64byte.c: Likewise.
2221 * testsuite/libffi.call/cls_6byte.c: Likewise.
2222 * testsuite/libffi.call/cls_7byte.c: Likewise.
2223 * testsuite/libffi.call/cls_8byte.c: Likewise.
2224 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2225 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2226 * testsuite/libffi.call/cls_double.c: Likewise.
2227 * testsuite/libffi.call/cls_float.c: Likewise.
2228 * testsuite/libffi.call/cls_schar.c: Likewise.
2229 * testsuite/libffi.call/cls_sint.c: Likewise.
2230 * testsuite/libffi.call/cls_sshort.c: Likewise.
2231 * testsuite/libffi.call/cls_uchar.c: Likewise.
2232 * testsuite/libffi.call/cls_uint.c: Likewise.
2233 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2234 * testsuite/libffi.call/cls_ushort.c: Likewise.
2235 * testsuite/libffi.call/problem1.c: Likewise.
2237 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2240 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2242 * testsuite/libffi.call/cls_9byte2.c: New test case.
2243 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2244 * testsuite/libffi.call/cls_64byte.c: Likewise.
2245 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2246 * testsuite/libffi.call/cls_19byte.c: Likewise.
2247 * testsuite/libffi.call/cls_18byte.c: Likewise.
2248 * testsuite/libffi.call/closure_fn4.c: Likewise.
2249 * testsuite/libffi.call/closure_fn5.c: Likewise.
2250 * testsuite/libffi.call/cls_schar.c: Likewise.
2251 * testsuite/libffi.call/cls_sint.c: Likewise.
2252 * testsuite/libffi.call/cls_sshort.c: Likewise.
2253 * testsuite/libffi.call/nested_struct2.c: Likewise.
2254 * testsuite/libffi.call/nested_struct3.c: Likewise.
2256 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2258 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2259 * testsuite/libffi.call/cls_uchar.c: Likewise.
2260 * testsuite/libffi.call/cls_uint.c: Likewise.
2261 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2262 * testsuite/libffi.call/cls_ushort.c: Likewise.
2263 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2265 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2267 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2270 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2272 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2273 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2275 2003-10-22 David Daney <ddaney@avtrex.com>
2277 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2278 __attribute__((__mode__(__SI__))) and friends.
2280 2003-10-22 Andreas Schwab <schwab@suse.de>
2282 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2284 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2286 * configure.in: AC_LINK_FILES(ffitarget.h).
2287 * configure: Regenerate.
2288 * Makefile.in: Likewise.
2289 * include/Makefile.in: Likewise.
2290 * testsuite/Makefile.in: Likewise.
2291 * fficonfig.h.in: Likewise.
2293 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2294 Richard Henderson <rth@redhat.com>
2296 Avoid that ffi.h includes fficonfig.h.
2298 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2299 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2300 (TARGET_SRC_MIPS_SGI): Removed.
2301 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2302 (MIPS_SGI): Removed.
2303 (CLEANFILES): Removed.
2304 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2306 * acconfig.h: Removed.
2307 * configure.in: Compute sizeofs only for double and long double.
2308 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2309 into AC_DEFINE instead of using acconfig.h. Create
2310 include/ffitarget.h instead of include/fficonfig.h. Rename
2311 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2312 AC_DEFINE EH_FRAME_FLAGS.
2313 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2314 (hack_DATA): Add ffitarget.h.
2315 * include/ffi.h.in: Remove all system specific definitions.
2316 Declare raw API even if it is not installed, why bother?
2317 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2318 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2319 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2320 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2321 * include/ffi_common.h (bool): Do not define.
2322 (ffi_assert): Accept failed assertion.
2323 (ffi_type_test): Return void and accept file/line.
2324 (FFI_ASSERT): Pass stringized failed assertion.
2325 (FFI_ASSERT_AT): New macro.
2326 (FFI_ASSERT_VALID_TYPE): New macro.
2327 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2328 UINT64, SINT64): Define here with gcc's __attribute__ macro
2330 (FLOAT32, ALIGN): Define here instead of in ffi.h
2331 * include/ffi-mips.h: Removed. Its content moved to
2332 src/mips/ffitarget.h after separating assembly and C sections.
2333 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2334 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2335 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2336 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2337 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2338 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2339 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2340 (ffi_type_test): Rewritten.
2341 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2342 FFI_ASSERT_VALID_TYPE.
2343 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2344 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2345 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2346 src/s390/ffitarget.h, src/sh/ffitarget.h,
2347 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2348 src/x86/ffitarget.h: New files.
2349 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2350 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2351 src/powerpc/aix.S, src/powerpc/darwin.S,
2352 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2353 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2354 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2355 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2356 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2359 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2361 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2362 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2364 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2366 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2367 Used when FFI_DEBUG = 1.
2369 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2371 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2374 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2376 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2377 FFI_MIPS_O32 for O32 ABI.
2379 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2381 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2382 SPARC64. Cleanup whitespaces.
2384 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2386 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2387 strongarm, xscale. Cleanup whitespaces.
2388 * testsuite/libffi.call/closure_fn1.c: Likewise.
2389 * testsuite/libffi.call/closure_fn2.c: Likewise.
2390 * testsuite/libffi.call/closure_fn3.c: Likewise.
2391 * testsuite/libffi.call/cls_12byte.c: Likewise.
2392 * testsuite/libffi.call/cls_16byte.c: Likewise.
2393 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2394 * testsuite/libffi.call/cls_20byte.c: Likewise.
2395 * testsuite/libffi.call/cls_24byte.c: Likewise.
2396 * testsuite/libffi.call/cls_2byte.c: Likewise.
2397 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2398 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2399 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2400 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2401 * testsuite/libffi.call/cls_4byte.c: Likewise.
2402 * testsuite/libffi.call/cls_5byte.c: Likewise.
2403 * testsuite/libffi.call/cls_6byte.c: Likewise.
2404 * testsuite/libffi.call/cls_7byte.c: Likewise.
2405 * testsuite/libffi.call/cls_8byte.c: Likewise.
2406 * testsuite/libffi.call/cls_double.c: Likewise.
2407 * testsuite/libffi.call/cls_float.c: Likewise.
2408 * testsuite/libffi.call/cls_uchar.c: Likewise.
2409 * testsuite/libffi.call/cls_uint.c: Likewise.
2410 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2411 * testsuite/libffi.call/cls_ushort.c: Likewise.
2412 * testsuite/libffi.call/nested_struct.c: Likewise.
2413 * testsuite/libffi.call/nested_struct1.c: Likewise.
2414 * testsuite/libffi.call/problem1.c: Likewise.
2415 * testsuite/libffi.special/unwindtest.cc: Likewise.
2416 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2418 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2420 * src/powerpc/aix.S: Cleanup whitespaces.
2421 * src/powerpc/aix_closure.S: Likewise.
2423 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2425 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2426 * src/powerpc/darwin_closure.S: Likewise.
2427 * src/powerpc/ffi_darwin.c: Likewise.
2429 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2430 David Edelsohn <edelsohn@gnu.org>
2432 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2433 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2435 * src/powerpc/darwin_closure.S: Likewise.
2436 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2437 according to the Darwin/AIX ABI.
2438 (ffi_prep_cif_machdep): Likewise.
2439 (ffi_closure_helper_DARWIN): Likewise.
2440 Remove the outgoing parameter stack logic. Simplify the evaluation
2441 of the different CASE types.
2442 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2443 statement in the trampoline code.
2445 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2447 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2448 for the register size.
2449 (ffi_closure_helper_SYSV): Handle the structure return value
2451 (ffi_closure_helper_SYSV): Return the appropriate type when
2452 the registers are used for the structure return value.
2453 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2454 the 64-bit return value. Update copyright years.
2456 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2458 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2459 srcdir for ffi_mips.h.
2461 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2463 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2465 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2466 placement of float result.
2467 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2468 cast of "resp" for big-endian 64 bit machines.
2470 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2472 * src/types.c (double, longdouble): Merge identical SH and ARM
2473 typedefs, and add POWERPC64.
2474 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2475 struct split over gpr and rest.
2476 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2477 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2479 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2481 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2484 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2486 * configure: Regenerate.
2488 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2490 * Makefile.am: Remove build rules for ffitest.
2491 * Makefile.in: Rebuilt.
2493 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2495 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2496 about implicit declaration of abort().
2498 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2500 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2501 * Makefile.in: Rebuilt.
2502 * configure.in: Add dejagnu test framework.
2503 * configure: Rebuilt.
2505 * testsuite/Makefile.am: New file.
2506 * testsuite/Makefile.in: Built
2507 * testsuite/lib/libffi-dg.exp: New file.
2508 * testsuite/config/default.exp: Likewise.
2509 * testsuite/libffi.call/call.exp: Likewise.
2510 * testsuite/libffi.call/ffitest.h: Likewise.
2511 * testsuite/libffi.call/closure_fn0.c: Likewise.
2512 * testsuite/libffi.call/closure_fn1.c: Likewise.
2513 * testsuite/libffi.call/closure_fn2.c: Likewise.
2514 * testsuite/libffi.call/closure_fn3.c: Likewise.
2515 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2516 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2517 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2518 * testsuite/libffi.call/cls_2byte.c: Likewise.
2519 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2520 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2521 * testsuite/libffi.call/cls_4byte.c: Likewise.
2522 * testsuite/libffi.call/cls_5byte.c: Likewise.
2523 * testsuite/libffi.call/cls_6byte.c: Likewise.
2524 * testsuite/libffi.call/cls_7byte.c: Likewise.
2525 * testsuite/libffi.call/cls_8byte.c: Likewise.
2526 * testsuite/libffi.call/cls_12byte.c: Likewise.
2527 * testsuite/libffi.call/cls_16byte.c: Likewise.
2528 * testsuite/libffi.call/cls_20byte.c: Likewise.
2529 * testsuite/libffi.call/cls_24byte.c: Likewise.
2530 * testsuite/libffi.call/cls_double.c: Likewise.
2531 * testsuite/libffi.call/cls_float.c: Likewise.
2532 * testsuite/libffi.call/cls_uchar.c: Likewise.
2533 * testsuite/libffi.call/cls_uint.c: Likewise.
2534 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2535 * testsuite/libffi.call/cls_ushort.c: Likewise.
2536 * testsuite/libffi.call/float.c: Likewise.
2537 * testsuite/libffi.call/float1.c: Likewise.
2538 * testsuite/libffi.call/float2.c: Likewise.
2539 * testsuite/libffi.call/many.c: Likewise.
2540 * testsuite/libffi.call/many_win32.c: Likewise.
2541 * testsuite/libffi.call/nested_struct.c: Likewise.
2542 * testsuite/libffi.call/nested_struct1.c: Likewise.
2543 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2544 * testsuite/libffi.call/problem1.c: Likewise.
2545 * testsuite/libffi.call/promotion.c: Likewise.
2546 * testsuite/libffi.call/return_ll.c: Likewise.
2547 * testsuite/libffi.call/return_sc.c: Likewise.
2548 * testsuite/libffi.call/return_uc.c: Likewise.
2549 * testsuite/libffi.call/strlen.c: Likewise.
2550 * testsuite/libffi.call/strlen_win32.c: Likewise.
2551 * testsuite/libffi.call/struct1.c: Likewise.
2552 * testsuite/libffi.call/struct2.c: Likewise.
2553 * testsuite/libffi.call/struct3.c: Likewise.
2554 * testsuite/libffi.call/struct4.c: Likewise.
2555 * testsuite/libffi.call/struct5.c: Likewise.
2556 * testsuite/libffi.call/struct6.c: Likewise.
2557 * testsuite/libffi.call/struct7.c: Likewise.
2558 * testsuite/libffi.call/struct8.c: Likewise.
2559 * testsuite/libffi.call/struct9.c: Likewise.
2560 * testsuite/libffi.special/special.exp: New file.
2561 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2562 * testsuite/libffi.special/unwindtest.cc: Likewise.
2565 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2567 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2570 2003-08-02 Alan Modra <amodra@bigpond.net.au>
2572 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2574 (ffi_closure_helper_LINUX64): Likewise.
2575 * src/powerpc/linux64.S: Remove code writing to parm save area.
2576 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2577 address in lr from ffi_closure_helper_LINUX64 call to calculate
2578 table address. Optimize function tail.
2580 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2582 * src/sparc/ffi.c: Handle all floating point registers.
2583 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2585 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2587 * README: Note that libffi is not part of GCC. Update the project
2590 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2592 * src/powerpc/ppc_closure.S: Include ffi.h.
2594 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2596 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2597 Use C style comments.
2599 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2601 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2602 * Makefile.in: Regenerate.
2603 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2604 * configure: Regenerate.
2605 * include/ffi.h.in: Add SHmedia support.
2606 * src/sh64/ffi.c: New file.
2607 * src/sh64/sysv.S: New file.
2609 2003-05-16 Jakub Jelinek <jakub@redhat.com>
2611 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2612 should be read-only.
2613 * configure: Rebuilt.
2614 * fficonfig.h.in: Rebuilt.
2615 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2616 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2617 * src/powerpc/linux64.S: Likewise.
2618 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2619 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2620 if -fpic/-fPIC/-mrelocatable.
2621 * src/powerpc/powerpc_closure.S: Likewise.
2622 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2623 #write in .eh_frame flags.
2624 * src/sparc/v9.S: Likewise.
2625 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2626 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2627 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2629 2003-05-07 Jeff Sturm <jsturm@one-point.com>
2631 Fixes PR bootstrap/10656
2632 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2633 support for .register pseudo-op.
2634 * src/sparc/v8.S: Use it.
2635 * fficonfig.h.in: Rebuilt.
2636 * configure: Rebuilt.
2638 2003-04-18 Jakub Jelinek <jakub@redhat.com>
2640 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2641 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2642 Make it the default on POWERPC64.
2643 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2644 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2645 * configure: Rebuilt.
2646 * src/powerpc/ffi.c (hidden): Define.
2647 (ffi_prep_args_SYSV): Renamed from
2648 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2649 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2650 ASM_NEEDS_REGISTERS64): New.
2651 (ffi_prep_args64): New function.
2652 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2653 (ffi_call): Likewise.
2654 (ffi_prep_closure): Likewise.
2655 (flush_icache): Surround by #ifndef POWERPC64.
2656 (ffi_dblfl): New union type.
2657 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2658 (ffi_closure_helper_LINUX64): New function.
2659 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2661 * src/powerpc/sysv.S: Likewise.
2662 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2663 * src/powerpc/linux64.S: New file.
2664 * src/powerpc/linux64_closure.S: New file.
2665 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2666 src/powerpc/linux64_closure.S.
2667 (TARGET_SRC_POWERPC): Likewise.
2669 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2670 closure_test_fn3): Fix result printing on big-endian 64-bit
2672 (main): Print tst2_arg instead of uninitialized tst2_result.
2674 * src/ffitest.c (main): Hide what closure pointer really points to
2677 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
2679 * configure.in (arm-*-netbsdelf*): Add configuration.
2680 (configure): Regenerated.
2682 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
2684 * include/Makefile.in: Regenerate.
2686 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2688 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2690 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2691 Receive closure pointer through parameter, read args using
2692 __builtin_dwarf_cfa.
2693 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2695 2003-03-12 Andreas Schwab <schwab@suse.de>
2697 * configure.in: Avoid trailing /. in toolexeclibdir.
2698 * configure: Rebuilt.
2700 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2702 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
2704 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2706 * libffi/src/powerpc/darwin_closure.S:
2707 Fix alignement bug, allocate 8 bytes for the result.
2708 * libffi/src/powerpc/aix_closure.S:
2710 * libffi/src/powerpc/ffi_darwin.c:
2711 Update stackframe description for aix/darwin_closure.S.
2713 2003-02-06 Jakub Jelinek <jakub@redhat.com>
2715 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2718 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2719 Andreas Schwab <schwab@suse.de>
2721 * configure.in: Adjust command to source config-ml.in to account
2722 for changes to the libffi_basedir definition.
2723 (libffi_basedir): Remove ${srcdir} from value and include trailing
2726 * configure: Regenerate.
2728 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2730 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2732 2003-01-28 Andrew Haley <aph@redhat.com>
2734 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2735 * src/x86/ffi64.c (ffi_prep_closure): New.
2736 (ffi_closure_UNIX64_inner): New.
2737 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2739 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
2741 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2742 Remove USE_LIBDIR conditional.
2743 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2744 * Makefile.in, configure: Rebuilt.
2746 2003-01027 David Edelsohn <edelsohn@gnu.org>
2748 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2749 * Makefile.in: Regenerate.
2751 2003-01-22 Andrew Haley <aph@redhat.com>
2753 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2756 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2758 * src/powerpc/darwin.S: Add unwind info.
2759 * src/powerpc/darwin_closure.S: Likewise.
2761 2003-01-14 Andrew Haley <aph@redhat.com>
2763 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2764 (ffi_prep_cif_machdep): Likewise.
2765 * src/x86/unix64.S: Add unwind info.
2767 2003-01-14 Andreas Jaeger <aj@suse.de>
2769 * src/ffitest.c (main): Only use ffi_closures if those are
2772 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2774 * libffi/src/ffitest.c
2775 add closure testcases
2777 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2779 * libffi/src/powerpc/ffi.c
2780 fix alignment bug for float (4 byte aligned iso 8 byte)
2782 2003-01-09 Geoffrey Keating <geoffk@apple.com>
2784 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2785 * src/powerpc/darwin.S: Remove RCS version string.
2787 2003-01-03 Jeff Sturm <jsturm@one-point.com>
2789 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2790 * src/ffitest.c (main): Use static storage for closure.
2791 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2792 * src/sparc/v8.S (ffi_closure_v8): New.
2793 * src/sparc/v9.S (ffi_closure_v9): New.
2795 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2797 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2798 enumeration for X86_WIN32.
2799 * src/x86/win32.S: Added ffi_call_STDCALL function
2801 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2802 switch cases for recognising FFI_STDCALL and
2803 calling ffi_call_STDCALL if target is X86_WIN32.
2804 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2805 stdcall versions of the "my_strlen" and "many"
2806 test functions (for X86_WIN32).
2807 Added test cases to test stdcall invocation using
2810 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2812 * src/sh/sysv.S: Add DWARF2 unwind info.
2814 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2816 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2818 2002-11-26 Jim Wilson <wilson@redhat.com>
2820 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2822 2002-11-23 H.J. Lu <hjl@gnu.org>
2824 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2825 Include ../config/accross.m4.
2826 * aclocal.m4; Rebuild.
2827 * configure: Likewise.
2829 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2831 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2833 2002-11-11 DJ Delorie <dj@redhat.com>
2835 * configure.in: Look for common files in the right place.
2837 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2839 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2840 raw data as _Jv_word values, not ffi_raw.
2841 (ffi_java_ptrarray_to_raw): Likewise.
2842 (ffi_java_rvalue_to_raw): New function.
2843 (ffi_java_raw_call): Call it.
2844 (ffi_java_raw_to_rvalue): New function.
2845 (ffi_java_translate_args): Call it.
2846 * src/ffitest.c (closure_test_fn): Interpret return value
2847 as ffi_arg, not int.
2848 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2849 FFI_TYPE_POINTER case.
2850 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2851 values extended to word size.
2853 2002-10-02 Andreas Jaeger <aj@suse.de>
2855 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2857 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
2859 * include/ffi.h.in: Fix i386 win32 compilation.
2861 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2863 * configure.in: Add s390x-*-linux-* target.
2864 * configure: Regenerate.
2865 * include/ffi.h.in: Define S390X for s390x targets.
2866 (FFI_CLOSURES): Define for s390/s390x.
2867 (FFI_TRAMPOLINE_SIZE): Likewise.
2868 (FFI_NATIVE_RAW_API): Likewise.
2869 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2870 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
2871 * src/s390/ffi.c: Major rework of existing code. Add support for
2872 s390x targets. Add closure support.
2873 * src/s390/sysv.S: Likewise.
2875 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
2877 * src/arm/sysv.S: Fix typo.
2879 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
2881 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2882 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2883 (ffi_call_SYSV): Handle soft-float.
2885 2002-09-27 Bo Thorsen <bo@suse.de>
2887 * include/ffi.h.in: Fix multilib x86-64 support.
2889 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2891 * Makefile.am (all-multi): Fix multilib parallel build.
2893 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2895 * configure.in (sh[34]*-*-linux*): Add brackets.
2896 * configure: Regenerate.
2898 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2900 * Makefile.am: Add SH support.
2901 * Makefile.in: Regenerate.
2902 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2903 * configure: Regenerate.
2904 * include/ffi.h.in: Add SH support.
2905 * src/sh/ffi.c: New file.
2906 * src/sh/sysv.S: New file.
2907 * src/types.c: Add SH support.
2909 2002-07-16 Bo Thorsen <bo@suse.de>
2911 * src/x86/ffi64.c: New file that adds x86-64 support.
2912 * src/x86/unix64.S: New file that handles argument setup for
2914 * src/x86/sysv.S: Don't use this on x86-64.
2915 * src/x86/ffi.c: Don't use this on x86-64.
2917 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2919 * src/ffitest.c (struct6): New test that tests a special case in
2921 (struct7): Likewise.
2922 (struct8): Likewise.
2923 (struct9): Likewise.
2924 (closure_test_fn): Silence warning about this when it's not used.
2925 (main): Add the new tests.
2926 (main): Fix a couple of wrong casts and silence some compiler warnings.
2927 * include/ffi.h.in: Add x86-64 ABI definition.
2928 * fficonfig.h.in: Regenerate.
2929 * Makefile.am: Add x86-64 support.
2930 * configure.in: Likewise.
2931 * Makefile.in: Regenerate.
2932 * configure: Likewise.
2934 2002-06-24 Bo Thorsen <bo@suse.de>
2936 * src/types.c: Merge settings for similar architectures.
2937 Add x86-64 sizes and alignments.
2939 2002-06-23 Bo Thorsen <bo@suse.de>
2941 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
2942 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
2943 * src/mips/ffi.c (ffi_prep_args): Likewise.
2944 * src/m68k/ffi.c (ffi_prep_args): Likewise.
2946 2002-07-18 H.J. Lu (hjl@gnu.org)
2948 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
2949 (libffi_la_SOURCES): Support MIPS_LINUX.
2950 (libffi_convenience_la_SOURCES): Likewise.
2951 * Makefile.in: Regenerated.
2953 * configure.in (mips64*-*): Skip.
2954 (mips*-*-linux*): New.
2955 * configure: Regenerated.
2957 * src/mips/ffi.c: Include <sgidefs.h>.
2959 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
2961 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
2963 2002-05-27 Roger Sayle <roger@eyesopen.com>
2965 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
2967 2002-05-27 Bo Thorsen <bo@suse.de>
2969 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
2972 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
2974 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
2975 beginning of function (for older apple cc).
2977 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
2979 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
2980 script entry, and set LD to it when configuring multilibs.
2981 * configure: Rebuilt.
2983 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
2985 * configure.in (sparc64-*-netbsd*): Add target.
2986 (sparc-*-netbsdelf*): Likewise.
2987 * configure: Regenerate.
2989 2002-04-28 David S. Miller <davem@redhat.com>
2991 * configure.in, configure: Fix SPARC test in previous change.
2993 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
2995 * Makefile.am: Add Linux for S/390 support.
2996 * Makefile.in: Regenerate.
2997 * configure.in: Add Linux for S/390 support.
2998 * configure: Regenerate.
2999 * include/ffi.h.in: Add Linux for S/390 support.
3000 * src/s390/ffi.c: New file from libffi CVS tree.
3001 * src/s390/sysv.S: New file from libffi CVS tree.
3003 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3005 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3007 * src/sparc/v8.S: Use it.
3008 * src/sparc/v9.S: Likewise.
3009 * fficonfig.h.in: Rebuilt.
3010 * configure: Rebuilt.
3012 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3014 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3016 * src/ia64/unix.S: Add unwind information. Fix comments.
3017 Save sp in a way that's compatible with unwind info.
3018 (ffi_call_unix): Correctly restore sp in all cases.
3019 * src/ia64/ffi.c: Add, fix comments.
3021 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3023 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3025 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3027 * configure.in (alpha*-*-netbsd*): Add target.
3028 * configure: Regenerate.
3030 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3032 * src/sparc/v8.S: Add unwind info.
3033 * src/sparc/v9.S: Likewise.
3035 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3037 * configure.in: Enable i*86-*-netbsdelf*.
3038 * configure: Rebuilt.
3040 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3043 * src/mips/n32.s: Delete
3044 * src/mips/o32.s: Delete
3046 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3048 * configure.in: Enable alpha*-*-freebsd*.
3049 * configure: Rebuilt.
3051 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3053 * Makefile.am: libfficonvenience -> libffi_convenience.
3054 * Makefile.in: Rebuilt.
3056 * Makefile.am: Define ffitest_OBJECTS.
3057 * Makefile.in: Rebuilt.
3059 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3060 David Edelsohn <edelsohn@gnu.org>
3062 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3063 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3064 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3065 * Makefile.in: Regenerate.
3066 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3067 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3068 (flush_icache, flush_range): New functions.
3069 (ffi_closure_helper_DARWIN): New function.
3070 * src/powerpc/aix_closure.S: New file.
3071 * src/powerpc/darwin_closure.S: New file.
3073 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3075 * include/ffi.h.in: Add typedef for ffi_arg.
3076 * src/ffitest.c (main): Declare rint with ffi_arg.
3078 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3080 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3081 number of GPRs for floating-point arguments.
3083 2002-01-31 Anthony Green <green@redhat.com>
3085 * configure: Rebuilt.
3086 * configure.in: Replace CHECK_SIZEOF and endian tests with
3087 cross-compiler friendly macros.
3088 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3091 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3093 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3094 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3096 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3098 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3099 (TARGET_SRC_POWERPC_AIX): New.
3100 (POWERPC_AIX): New stanza.
3101 * Makefile.in: Regenerate.
3102 * configure.in: Add AIX case.
3103 * configure: Regenerate.
3104 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3105 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3106 size. Fix "long double" support.
3107 (ffi_call): Add FFI_AIX case.
3108 * src/powerpc/aix.S: New.
3110 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3112 Implement Darwin PowerPC ABI.
3113 * configure.in: Handle powerpc-*-darwin*.
3114 * Makefile.am: Set source files for POWERPC_DARWIN.
3115 * configure: Rebuilt.
3116 * Makefile.in: Rebuilt.
3117 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3119 * src/powerpc/darwin.S: New file.
3120 * src/powerpc/ffi_darwin.c: New file.
3122 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3124 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3126 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3128 * src/x86/sysv.S: Avoid gas-only .balign directive.
3129 Use C style comments.
3131 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3133 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3134 Fixes PR bootstrap/3563.
3136 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3138 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3140 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3142 * configure.in: Recognize sparc*-sun-* host.
3143 * configure: Regenerate.
3145 2001-06-06 Andrew Haley <aph@redhat.com>
3147 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
3149 2001-06-03 Andrew Haley <aph@redhat.com>
3151 * src/alpha/osf.S: Add unwind info.
3152 * src/powerpc/sysv.S: Add unwind info.
3153 * src/powerpc/ppc_closure.S: Likewise.
3155 2000-05-31 Jeff Sturm <jsturm@one-point.com>
3157 * configure.in: Fix AC_ARG_ENABLE usage.
3158 * configure: Rebuilt.
3160 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3162 * configure.in: Remove warning about beta code.
3163 * configure: Rebuilt.
3165 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3167 * src/ia64/unix.S: Restore stack pointer when returning from
3169 * src/ia64/ffi.c: Fix typo in comment.
3171 2001-04-18 Jim Wilson <wilson@redhat.com>
3173 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3174 to eliminate RAW DV.
3176 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3178 * Makefile.am: Make a libtool convenience library.
3179 * Makefile.in: Rebuilt.
3181 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3183 * configure.in: Use different syntax for subdirectory creation.
3184 * configure: Rebuilt.
3186 2001-03-27 Jon Beniston <jon@beniston.com>
3188 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3189 * configure: Rebuilt.
3190 * Makefile.am: Added X86_WIN32 target support.
3191 * Makefile.in: Rebuilt.
3193 * include/ffi.h.in: Added X86_WIN32 target support.
3195 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3196 * src/types.c: Added X86_WIN32 target support.
3198 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3199 stuff removed and made to work with CygWin's gas.
3201 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3203 * configure.in: Make target subdirectory in build dir.
3204 * Makefile.am: Override suffix based rules to specify correct output
3206 * Makefile.in: Rebuilt.
3207 * configure: Rebuilt.
3209 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3211 * src/powerpc/ppc_closure.S: New file.
3212 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3213 involving long long and register pairs.
3214 (ffi_prep_closure): New function.
3215 (flush_icache): Likewise.
3216 (ffi_closure_helper_SYSV): Likewise.
3217 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3218 (FFI_TRAMPOLINE_SIZE): Likewise.
3219 (FFI_NATIVE_RAW_API): Likewise.
3220 * Makefile.in: Rebuilt.
3221 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3222 (TARGET_SRC_POWERPC): Likewise.
3224 2001-03-19 Tom Tromey <tromey@redhat.com>
3226 * Makefile.in: Rebuilt.
3227 * Makefile.am (ffitest_LDFLAGS): New macro.
3229 2001-03-02 Nick Clifton <nickc@redhat.com>
3231 * include/ffi.h.in: Remove RCS ident string.
3232 * include/ffi_mips.h: Remove RCS ident string.
3233 * src/debug.c: Remove RCS ident string.
3234 * src/ffitest.c: Remove RCS ident string.
3235 * src/prep_cif.c: Remove RCS ident string.
3236 * src/types.c: Remove RCS ident string.
3237 * src/alpha/ffi.c: Remove RCS ident string.
3238 * src/alpha/osf.S: Remove RCS ident string.
3239 * src/arm/ffi.c: Remove RCS ident string.
3240 * src/arm/sysv.S: Remove RCS ident string.
3241 * src/mips/ffi.c: Remove RCS ident string.
3242 * src/mips/n32.S: Remove RCS ident string.
3243 * src/mips/o32.S: Remove RCS ident string.
3244 * src/sparc/ffi.c: Remove RCS ident string.
3245 * src/sparc/v8.S: Remove RCS ident string.
3246 * src/sparc/v9.S: Remove RCS ident string.
3247 * src/x86/ffi.c: Remove RCS ident string.
3248 * src/x86/sysv.S: Remove RCS ident string.
3250 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3252 * include/ffi.h.in: Change sourceware.cygnus.com references to
3255 2000-12-09 Richard Henderson <rth@redhat.com>
3257 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3258 (ffi_closure_osf_inner): Index rather than increment avalue
3259 and arg_types. Give ffi_closure_osf the raw return value type.
3260 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3263 2000-12-07 Richard Henderson <rth@redhat.com>
3265 * src/raw_api.c (ffi_translate_args): Fix typo.
3266 (ffi_prep_closure): Likewise.
3268 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3269 FFI_TRAMPOLINE_SIZE.
3270 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3271 cif->bytes for new ffi_call_osf implementation.
3272 (ffi_prep_args): Absorb into ...
3273 (ffi_call): ... here. Do all stack allocation here and
3274 avoid a callback function.
3275 (ffi_prep_closure, ffi_closure_osf_inner): New.
3276 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3277 (ffi_closure_osf): New.
3279 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3281 * config.guess, config.sub, install-sh: Removed.
3282 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3283 * Makefile.in: Rebuilt.
3285 * acinclude.m4: Include libtool macros from the top level.
3286 * aclocal.m4, configure: Rebuilt.
3288 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3290 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3291 * configure: Rebuilt.
3293 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3295 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3296 memory correctly. Use conditional instructions, not branches where
3299 2000-05-04 Tom Tromey <tromey@cygnus.com>
3301 * configure: Rebuilt.
3302 * configure.in: Match `arm*-*-linux-*'.
3303 From Chris Dornan <cdornan@arm.com>.
3305 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3307 * Makefile.am (SUBDIRS): Define.
3308 (AM_MAKEFLAGS): Likewise.
3309 (Multilib support.): Add section.
3310 * Makefile.in: Rebuilt.
3311 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3312 New variables. Set for gcc using -print-multi-lib. Export them
3314 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3315 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3317 (extra_compiler_flags_value, extra_compiler_flags): Check these
3318 for extra compiler options which need to be passed down in
3321 2000-04-16 Anthony Green <green@redhat.com>
3323 * configure: Rebuilt.
3324 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3326 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3328 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3329 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3330 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3331 Replace all void * sizeofs with sizeof(int).
3332 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3333 different than DOUBLE.
3334 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3335 (ffi_prep_args_v9): New function.
3336 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3337 (ffi_V9_return_struct): New function.
3338 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3339 32bit code (not yet cross-arch calls).
3340 * src/sparc/v8.S: Add struct return delay nop.
3342 * src/sparc/v9.S: New file.
3343 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3344 is used on sparc64 only for structures larger than 32 bytes.
3345 Pass by reference for structures is done for structure arguments
3346 larger than 16 bytes.
3347 * src/ffitest.c (main): Use 64bit rint on sparc64.
3348 Run long long tests on sparc.
3349 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3351 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3353 * configure.in (sparc-*-linux*): New supported target.
3354 (sparc64-*-linux*): Likewise.
3355 * configure: Rebuilt.
3356 * Makefile.am: Add v9.S to SPARC files.
3357 * Makefile.in: Likewise.
3358 (LINK): Surround $(CCLD) into double quotes, so that multilib
3359 compiles work correctly.
3361 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3363 * configure: Rebuilt.
3364 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3365 proposed by Bryce McKinlay.
3367 2000-03-20 Tom Tromey <tromey@cygnus.com>
3369 * Makefile.in: Hand edit for java_raw_api.lo.
3371 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3373 * config.guess, config.sub: Update from the gcc tree.
3374 Fix for PR libgcj/168.
3376 2000-03-03 Tom Tromey <tromey@cygnus.com>
3378 * Makefile.in: Fixed ia64 by hand.
3380 * configure: Rebuilt.
3381 * configure.in (--enable-multilib): New option.
3382 (libffi_basedir): New subst.
3383 (AC_OUTPUT): Added multilib code.
3385 2000-03-02 Tom Tromey <tromey@cygnus.com>
3387 * Makefile.in: Rebuilt.
3388 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3391 2000-02-25 Hans Boehm <boehm@acm.org>
3393 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3395 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3397 (ffi_prep_raw_closure): Use ffi_translate_args, not
3398 ffi_closure_translate.
3399 * src/java_raw_api.c: New file.
3400 * src/ffitest.c (closure_test_fn): New function.
3401 (main): Define `rint' as long long on IA64. Added new test when
3402 FFI_CLOSURES is defined.
3403 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3404 (ffi_abi): Recognize IA64.
3405 (ffi_raw): Added `flt' field.
3406 Added "Java raw API" code.
3407 * configure.in: Recognize ia64.
3408 * Makefile.am (TARGET_SRC_IA64): New macro.
3409 (libffi_la_common_SOURCES): Added java_raw_api.c.
3410 (libffi_la_SOURCES): Define in IA64 case.
3412 2000-01-04 Tom Tromey <tromey@cygnus.com>
3414 * Makefile.in: Rebuilt with newer automake.
3416 1999-12-31 Tom Tromey <tromey@cygnus.com>
3418 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3420 1999-09-01 Tom Tromey <tromey@cygnus.com>
3422 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3424 * fficonfig.h.in: Rebuilt.
3425 * configure: Rebuilt.
3426 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3427 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3428 * acconfig.h: Don't #undef PACKAGE or VERSION.
3430 1999-08-09 Anthony Green <green@cygnus.com>
3432 * include/ffi.h.in: Try to work around messy header problem
3433 with PACKAGE and VERSION.
3435 * configure: Rebuilt.
3436 * configure.in: Change version to 2.00-beta.
3438 * fficonfig.h.in: Rebuilt.
3439 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3441 * src/x86/ffi.c (ffi_raw_call): Rename.
3443 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3445 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3446 (ffi_prep_incoming_args_SYSV): Ditto.
3447 (ffi_prep_closure): Ditto.
3448 (ffi_closure_raw_SYSV): Ditto.
3449 (ffi_prep_raw_closure): More ditto.
3450 (ffi_call_raw): Final ditto.
3452 * include/ffi.h.in: Add definitions for closure and raw API.
3454 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3457 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3459 * src/raw_api.c: New file.
3461 * include/ffi.h.in (ffi_raw): New type.
3462 (UINT_ARG, SINT_ARG): New defines.
3463 (ffi_closure, ffi_raw_closure): New types.
3464 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3466 * configure.in: Add check for endianness and sizeof void*.
3468 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3469 instead of directly.
3471 * configure: Rebuilt.
3473 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3475 * configure.in: Add x86 and powerpc BeOS configurations.
3476 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3478 1999-05-09 Anthony Green <green@cygnus.com>
3480 * configure.in: Add warning about this being beta code.
3481 Remove src/Makefile.am from the picture.
3482 * configure: Rebuilt.
3484 * Makefile.am: Move logic from src/Makefile.am. Add changes
3485 to support libffi as a target library.
3486 * Makefile.in: Rebuilt.
3488 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3489 Upgraded to new autoconf, automake, libtool.
3493 * LICENSE: Update copyright date.
3495 * src/Makefile.am, src/Makefile.in: Removed.
3497 1998-11-29 Anthony Green <green@cygnus.com>
3499 * include/ChangeLog: Removed.
3500 * src/ChangeLog: Removed.
3501 * src/mips/ChangeLog: Removed.
3502 * src/sparc/ChangeLog: Remboved.
3503 * src/x86/ChangeLog: Removed.
3505 * ChangeLog.v1: Created.