1 2008-02-26 Anthony Green <green@redhat.com>
2 Thomas Heller <theller@ctypes.org>
4 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
7 2008-02-03 Timothy Wall <twall@users.sf.net>
9 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
10 offset based on code pointer, not data pointer.
12 2008-01-31 Timothy Wall <twall@users.sf.net>
14 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
16 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
18 * src/x86/win32.S: Add assembly for stdcall closure.
19 * src/x86/ffi.c: Initialize stdcall closure trampoline.
21 2009-06-04 Andrew Haley <aph@redhat.com>
23 * include/ffi.h.in: Change void (*)() to void (*)(void).
24 * src/x86/ffi.c: Likewise.
26 2009-06-04 Andrew Haley <aph@redhat.com>
28 * src/powerpc/ppc_closure.S: Insert licence header.
29 * src/powerpc/linux64_closure.S: Likewise.
30 * src/m68k/sysv.S: Likewise.
32 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
33 * src/powerpc/ffi.c: Likewise.
34 * src/powerpc/ffi_darwin.c: Likewise.
35 * src/m32r/ffi.c: Likewise.
36 * src/sh64/ffi.c: Likewise.
37 * src/x86/ffi64.c: Likewise.
38 * src/alpha/ffi.c: Likewise.
39 * src/alpha/osf.S: Likewise.
40 * src/frv/ffi.c: Likewise.
41 * src/s390/ffi.c: Likewise.
42 * src/pa/ffi.c: Likewise.
43 * src/pa/hpux32.S: Likewise.
44 * src/ia64/unix.S: Likewise.
45 * src/ia64/ffi.c: Likewise.
46 * src/sparc/ffi.c: Likewise.
47 * src/mips/ffi.c: Likewise.
48 * src/sh/ffi.c: Likewise.
50 2008-02-15 David Daney <ddaney@avtrex.com>
52 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
53 Define (conditionally), and use it to include cachectl.h.
54 (ffi_prep_closure_loc): Fix cache flushing.
55 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
57 2009-06-04 Andrew Haley <aph@redhat.com>
63 src/powerpc/ffitarget.h,
76 src/alpha/ffitarget.h,
92 src/ia64/ia64_flags.h,
96 src/sparc/ffitarget.h,
106 src/sh/sysv.S: Update license text.
108 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
110 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
111 (.eh_frame): Add FDE for it.
113 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
115 * configure.ac: Also check if assembler supports pc-relative
116 relocs on X86_WIN32 targets.
117 * configure: Regenerate.
118 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
119 (_ffi_call_SYSV): Add missing function type symbol .def and
120 add EH markup labels.
121 (_ffi_call_STDCALL): Likewise.
122 (_ffi_closure_SYSV): Likewise.
123 (_ffi_closure_raw_SYSV): Likewise.
124 (.eh_frame): Add hand-crafted EH data.
126 2009-04-09 Jakub Jelinek <jakub@redhat.com>
128 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
129 version 3 of the GNU General Public License and to point readers
130 at the COPYING3 file and the FSF's license web page.
131 * testsuite/libffi.call/call.exp: Likewise.
132 * testsuite/libffi.special/special.exp: Likewise.
134 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
136 * configure: Regenerate.
138 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
141 * configure.ac (HAVE_AS_X86_PCREL): New test.
142 * configure: Regenerate.
143 * fficonfig.h.in: Regenerate.
144 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
145 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
146 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
147 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
148 * src/x86/unix64.S (.Lstore_table): Move to .text section.
149 (.Lload_table): Likewise.
150 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
152 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
154 * configure: Regenerate.
156 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
158 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
159 signed/unsigned int8/16 return values.
160 * src/sparc/v8.S (ffi_call_v8): Likewise.
161 (ffi_closure_v8): Likewise.
163 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
164 Steve Ellcey <sje@cup.hp.com>
166 * configure: Regenerate for new libtool.
167 * Makefile.in: Ditto.
168 * include/Makefile.in: Ditto.
171 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
173 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
174 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
175 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
176 Adjust copyright notice.
177 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
178 register or two register to use for FFI_SYSV structs.
179 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
180 (ffi_closure_helper_SYSV): Fix the return type for
181 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
182 Adjust copyright notice.
184 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
186 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
189 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
191 * configure: Regenerate.
192 * include/Makefile.in: Regenerate.
193 * testsuite/Makefile.in: Regenerate.
195 2008-06-07 Joseph Myers <joseph@codesourcery.com>
197 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
198 powerpc-*-beos*): Remove.
199 * configure: Regenerate.
201 2008-05-09 Julian Brown <julian@codesourcery.com>
203 * Makefile.am (LTLDFLAGS): New.
204 (libffi_la_LDFLAGS): Use above.
205 * Makefile.in: Regenerate.
207 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
210 * aclocal.m4: Regenerate.
211 * configure: Regenerate.
213 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
215 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
216 * src/sh64/sysv.S: Likewise.
218 2008-03-26 Daniel Jacobowitz <dan@debian.org>
220 * src/arm/sysv.S: Fix ARM comment marker.
222 2008-03-26 Jakub Jelinek <jakub@redhat.com>
224 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
225 * src/s390/sysv.S: Likewise.
226 * src/powerpc/ppc_closure.S: Likewise.
227 * src/powerpc/sysv.S: Likewise.
228 * src/x86/unix64.S: Likewise.
229 * src/x86/sysv.S: Likewise.
230 * src/sparc/v8.S: Likewise.
231 * src/sparc/v9.S: Likewise.
232 * src/m68k/sysv.S: Likewise.
233 * src/arm/sysv.S: Likewise.
235 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
237 * aclocal.m4: Regenerate.
238 * configure: Likewise.
239 * Makefile.in: Likewise.
240 * include/Makefile.in: Likewise.
241 * testsuite/Makefile.in: Likewise.
243 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
244 Andreas Tobler <a.tobler@schweiz.org>
246 * configure.ac: Add amd64-*-freebsd* target.
247 * configure: Regenerate.
249 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
252 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
255 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
258 2008-01-24 David Edelsohn <edelsohn@gnu.org>
260 * configure: Regenerate.
262 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
264 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
266 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
269 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
270 signed/unsigned int8/16 for X86_DARWIN.
271 Updated copyright info.
272 Handle one and two byte structs with special cif->flags.
273 * src/x86/ffitarget.h: Add special types for one and two byte structs.
274 Updated copyright info.
275 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
277 Remove code to pop args from the stack after call.
278 Special-case signed/unsigned for int8/16, one and two byte structs.
279 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
280 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
282 Updated copyright info.
284 2007-12-08 David Daney <ddaney@avtrex.com>
286 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
287 SUBU, add with ADDU and use smaller code sequences.
289 2007-12-07 David Daney <ddaney@avtrex.com>
291 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
294 2007-12-06 David Daney <ddaney@avtrex.com>
296 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
298 (ffi_java_raw): New typedef.
299 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
300 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
302 (ffi_java_raw_closure) : Same.
303 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
305 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
307 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
308 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
309 sizeof(ffi_java_raw) for alignment calculations.
310 (ffi_java_ptrarray_to_raw): Same.
311 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
312 if FFI_SIZEOF_JAVA_RAW == 4.
313 (ffi_java_raw_to_rvalue): Same.
314 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
315 (ffi_java_translate_args): Same.
316 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
318 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
320 2007-12-06 David Daney <ddaney@avtrex.com>
322 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
325 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
328 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
329 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
330 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
331 set the NUM_FPR_ARG_REGISTERS according to.
332 Add support for potential soft-float support under hard-float
334 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
335 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
336 to the FFI_LINUX_SOFT_FLOAT ABI.
337 (ffi_prep_cif_machdep): Likewise.
338 (ffi_closure_helper_SYSV): Likewise.
339 * src/powerpc/ppc_closure.S: Make sure not to store float/double
340 on archs where __NO_FPRS__ is true.
341 Add FFI_TYPE_UINT128 support.
342 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
343 Adjust copyright notice.
345 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
347 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
348 * include/ffi_common.h: ... here.
351 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
353 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
354 * src/powerpc/linux64.S: Likewise.
355 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
356 * testsuite/libffi.call/return_ldl.c: New test.
358 2007-09-04 <aph@redhat.com>
360 * src/arm/sysv.S (UNWIND): New.
361 (Whole file): Conditionally compile unwinder directives.
362 * src/arm/sysv.S: Add unwinder directives.
364 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
365 Only treat r0 as a struct address if we're actually returning a
367 Only copy the bytes that are actually within a struct.
368 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
369 is returned in r0, not passed by address.
370 (ffi_call): Allocate a word-sized temporary for the case where
371 a composite is returned in r0.
372 (ffi_prep_incoming_args_SYSV): Align as necessary.
374 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
376 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
377 directly using the sys_cacheflush syscall.
379 2007-07-27 Andrew Haley <aph@redhat.com>
381 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
383 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
385 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
386 * configure.ac: Likewise.
387 * Makefile.in: Regenerate.
388 * include/Makefile.in: Likewise.
389 * testsuite/Makefile.in: Likewise.
390 * configure: Likewise.
392 2007-08-24 David Daney <ddaney@avtrex.com>
394 * testsuite/libffi.call/return_sl.c: New test.
396 2007-08-10 David Daney <ddaney@avtrex.com>
398 * testsuite/libffi.call/cls_multi_ushort.c,
399 testsuite/libffi.call/cls_align_uint16.c,
400 testsuite/libffi.call/nested_struct1.c,
401 testsuite/libffi.call/nested_struct3.c,
402 testsuite/libffi.call/cls_7_1_byte.c,
403 testsuite/libffi.call/nested_struct5.c,
404 testsuite/libffi.call/cls_double.c,
405 testsuite/libffi.call/nested_struct7.c,
406 testsuite/libffi.call/cls_sint.c,
407 testsuite/libffi.call/nested_struct9.c,
408 testsuite/libffi.call/cls_20byte1.c,
409 testsuite/libffi.call/cls_multi_sshortchar.c,
410 testsuite/libffi.call/cls_align_sint64.c,
411 testsuite/libffi.call/cls_3byte2.c,
412 testsuite/libffi.call/cls_multi_schar.c,
413 testsuite/libffi.call/cls_multi_uchar.c,
414 testsuite/libffi.call/cls_19byte.c,
415 testsuite/libffi.call/cls_9byte1.c,
416 testsuite/libffi.call/cls_align_float.c,
417 testsuite/libffi.call/closure_fn1.c,
418 testsuite/libffi.call/problem1.c,
419 testsuite/libffi.call/closure_fn3.c,
420 testsuite/libffi.call/cls_sshort.c,
421 testsuite/libffi.call/closure_fn5.c,
422 testsuite/libffi.call/cls_align_double.c,
423 testsuite/libffi.call/nested_struct.c,
424 testsuite/libffi.call/cls_2byte.c,
425 testsuite/libffi.call/nested_struct10.c,
426 testsuite/libffi.call/cls_4byte.c,
427 testsuite/libffi.call/cls_6byte.c,
428 testsuite/libffi.call/cls_8byte.c,
429 testsuite/libffi.call/cls_multi_sshort.c,
430 testsuite/libffi.call/cls_align_sint16.c,
431 testsuite/libffi.call/cls_align_uint32.c,
432 testsuite/libffi.call/cls_20byte.c,
433 testsuite/libffi.call/cls_float.c,
434 testsuite/libffi.call/nested_struct2.c,
435 testsuite/libffi.call/cls_5_1_byte.c,
436 testsuite/libffi.call/nested_struct4.c,
437 testsuite/libffi.call/cls_24byte.c,
438 testsuite/libffi.call/nested_struct6.c,
439 testsuite/libffi.call/cls_64byte.c,
440 testsuite/libffi.call/nested_struct8.c,
441 testsuite/libffi.call/cls_uint.c,
442 testsuite/libffi.call/cls_multi_ushortchar.c,
443 testsuite/libffi.call/cls_schar.c,
444 testsuite/libffi.call/cls_uchar.c,
445 testsuite/libffi.call/cls_align_uint64.c,
446 testsuite/libffi.call/cls_ulonglong.c,
447 testsuite/libffi.call/cls_align_longdouble.c,
448 testsuite/libffi.call/cls_1_1byte.c,
449 testsuite/libffi.call/cls_12byte.c,
450 testsuite/libffi.call/cls_3_1byte.c,
451 testsuite/libffi.call/cls_3byte1.c,
452 testsuite/libffi.call/cls_4_1byte.c,
453 testsuite/libffi.call/cls_6_1_byte.c,
454 testsuite/libffi.call/cls_16byte.c,
455 testsuite/libffi.call/cls_18byte.c,
456 testsuite/libffi.call/closure_fn0.c,
457 testsuite/libffi.call/cls_9byte2.c,
458 testsuite/libffi.call/closure_fn2.c,
459 testsuite/libffi.call/closure_fn4.c,
460 testsuite/libffi.call/cls_ushort.c,
461 testsuite/libffi.call/closure_fn6.c,
462 testsuite/libffi.call/cls_5byte.c,
463 testsuite/libffi.call/cls_align_pointer.c,
464 testsuite/libffi.call/cls_7byte.c,
465 testsuite/libffi.call/cls_align_sint32.c,
466 testsuite/libffi.special/unwindtest_ffi_call.cc,
467 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
469 2007-08-10 David Daney <ddaney@avtrex.com>
472 * configure.ac: Don't treat mips64 as a special case.
473 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
474 * configure: Regenerate
475 * Makefile.in: Ditto.
476 * fficonfig.h.in: Ditto.
477 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
478 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
479 (FFI_DEFAULT_ABI): Set for n64 case.
480 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
481 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
482 (ffi_closure_N32): New function.
483 (.eh_frame): New section
484 * src/mips/o32.S: Clean up comments.
485 (ffi_closure_O32): Pass ffi_closure parameter in $12.
486 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
487 _MIPS_SIM == _ABIN32 throughout.
488 (FFI_MIPS_STOP_HERE): New, use in place of
490 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
491 to support n32/n64 ABIs.
492 (calc_n32_struct_flags): Rewrite.
493 (calc_n32_return_struct_flags): Remove unused variable. Reverse
494 position of flag bits.
495 (ffi_prep_cif_machdep): Rewrite n32 portion.
496 (ffi_call): Enable for n64. Add special handling for small structure
498 (ffi_prep_closure_loc): Add n32 and n64 support.
499 (ffi_closure_mips_inner_O32): Add cast to silence warning.
500 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
502 2007-08-08 David Daney <ddaney@avtrex.com>
504 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
505 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
507 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
508 * testsuite/libffi.call/cls_sint.c (main): Ditto.
509 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
510 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
511 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
512 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
513 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
514 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
515 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
516 * testsuite/libffi.call/return_fl3.c (main): Ditto.
517 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
518 * testsuite/libffi.call/nested_struct.c (main): Ditto.
519 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
520 * testsuite/libffi.call/return_ll1.c (main): Ditto.
521 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
522 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
523 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
524 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
525 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
526 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
527 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
528 * testsuite/libffi.call/cls_uint.c (main): Ditto.
529 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
530 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
531 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
532 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
533 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
534 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
535 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
536 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
537 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
539 2007-08-07 Andrew Haley <aph@redhat.com>
541 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
544 2007-08-06 Andrew Haley <aph@redhat.com>
547 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
548 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
551 2007-08-02 David Daney <ddaney@avtrex.com>
553 * testsuite/libffi.call/return_ul.c (main): Define return type as
554 ffi_arg. Use proper printf conversion specifier.
556 2007-07-30 Andrew Haley <aph@redhat.com>
559 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
560 signed/unsigned int8/16.
561 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
563 Remove code to pop args from the stack after call.
564 Special-case signed/unsigned int8/16.
565 * testsuite/libffi.call/return_sc.c (main): Revert.
567 2007-07-26 Richard Guenther <rguenther@suse.de>
570 * testsuite/libffi.call/return_sc.c (main): Verify call
571 result as signed char, not ffi_arg.
573 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
575 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
576 * configure: Regenerate.
578 2007-07-11 David Daney <ddaney@avtrex.com>
580 * src/mips/ffi.c: Don't include sys/cachectl.h.
581 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
584 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
586 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
587 from (ffi_prep_closure): ... this.
588 (FFI_INIT_TRAMPOLINE): Adjust.
590 2005-12-31 Phil Blundell <pb@reciva.com>
592 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
593 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
594 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
595 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
596 (FFI_CLOSURES): Enable closure support.
598 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
600 * testsuite/libffi.call/cls_multi_ushort.c,
601 testsuite/libffi.call/cls_align_uint16.c,
602 testsuite/libffi.call/nested_struct1.c,
603 testsuite/libffi.call/nested_struct3.c,
604 testsuite/libffi.call/cls_7_1_byte.c,
605 testsuite/libffi.call/cls_double.c,
606 testsuite/libffi.call/nested_struct5.c,
607 testsuite/libffi.call/nested_struct7.c,
608 testsuite/libffi.call/cls_sint.c,
609 testsuite/libffi.call/nested_struct9.c,
610 testsuite/libffi.call/cls_20byte1.c,
611 testsuite/libffi.call/cls_multi_sshortchar.c,
612 testsuite/libffi.call/cls_align_sint64.c,
613 testsuite/libffi.call/cls_3byte2.c,
614 testsuite/libffi.call/cls_multi_schar.c,
615 testsuite/libffi.call/cls_multi_uchar.c,
616 testsuite/libffi.call/cls_19byte.c,
617 testsuite/libffi.call/cls_9byte1.c,
618 testsuite/libffi.call/cls_align_float.c,
619 testsuite/libffi.call/closure_fn1.c,
620 testsuite/libffi.call/problem1.c,
621 testsuite/libffi.call/closure_fn3.c,
622 testsuite/libffi.call/cls_sshort.c,
623 testsuite/libffi.call/closure_fn5.c,
624 testsuite/libffi.call/cls_align_double.c,
625 testsuite/libffi.call/cls_2byte.c,
626 testsuite/libffi.call/nested_struct.c,
627 testsuite/libffi.call/nested_struct10.c,
628 testsuite/libffi.call/cls_4byte.c,
629 testsuite/libffi.call/cls_6byte.c,
630 testsuite/libffi.call/cls_8byte.c,
631 testsuite/libffi.call/cls_multi_sshort.c,
632 testsuite/libffi.call/cls_align_uint32.c,
633 testsuite/libffi.call/cls_align_sint16.c,
634 testsuite/libffi.call/cls_float.c,
635 testsuite/libffi.call/cls_20byte.c,
636 testsuite/libffi.call/cls_5_1_byte.c,
637 testsuite/libffi.call/nested_struct2.c,
638 testsuite/libffi.call/cls_24byte.c,
639 testsuite/libffi.call/nested_struct4.c,
640 testsuite/libffi.call/nested_struct6.c,
641 testsuite/libffi.call/cls_64byte.c,
642 testsuite/libffi.call/nested_struct8.c,
643 testsuite/libffi.call/cls_uint.c,
644 testsuite/libffi.call/cls_multi_ushortchar.c,
645 testsuite/libffi.call/cls_schar.c,
646 testsuite/libffi.call/cls_uchar.c,
647 testsuite/libffi.call/cls_align_uint64.c,
648 testsuite/libffi.call/cls_ulonglong.c,
649 testsuite/libffi.call/cls_align_longdouble.c,
650 testsuite/libffi.call/cls_1_1byte.c,
651 testsuite/libffi.call/cls_12byte.c,
652 testsuite/libffi.call/cls_3_1byte.c,
653 testsuite/libffi.call/cls_3byte1.c,
654 testsuite/libffi.call/cls_4_1byte.c,
655 testsuite/libffi.call/cls_6_1_byte.c,
656 testsuite/libffi.call/cls_16byte.c,
657 testsuite/libffi.call/cls_18byte.c,
658 testsuite/libffi.call/closure_fn0.c,
659 testsuite/libffi.call/cls_9byte2.c,
660 testsuite/libffi.call/closure_fn2.c,
661 testsuite/libffi.call/closure_fn4.c,
662 testsuite/libffi.call/cls_ushort.c,
663 testsuite/libffi.call/closure_fn6.c,
664 testsuite/libffi.call/cls_5byte.c,
665 testsuite/libffi.call/cls_align_pointer.c,
666 testsuite/libffi.call/cls_7byte.c,
667 testsuite/libffi.call/cls_align_sint32.c,
668 testsuite/libffi.special/unwindtest_ffi_call.cc,
669 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
671 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
673 * aclocal.m4: Regenerated.
675 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
677 * configure: Regenerate.
679 2007-05-23 Steve Ellcey <sje@cup.hp.com>
681 * Makefile.in: Regenerate.
682 * configure: Regenerate.
683 * aclocal.m4: Regenerate.
684 * include/Makefile.in: Regenerate.
685 * testsuite/Makefile.in: Regenerate.
687 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
689 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
690 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
691 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
692 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
693 (FFI_CLOSURES): Enable closure support.
695 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
697 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
698 * configure: Regenerate.
699 * fficonfig.h.in: Regenerate.
700 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
701 CFI_OFFSET,CFI_DEF_CFA): New macros.
702 (ffi_call_SYSV): Add callframe annotation.
704 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
706 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
707 numerous test suite failures.
708 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
710 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
712 * Makefile.am (EXTRA_DIST): Bring up to date.
713 * Makefile.in: Regenerate.
714 * src/frv/eabi.S: Remove RCS keyword.
716 2007-04-06 Richard Henderson <rth@redhat.com>
718 * configure.ac: Tidy target case.
719 (HAVE_LONG_DOUBLE): Allow the target to override.
720 * configure: Regenerate.
721 * include/ffi.h.in: Don't define ffi_type_foo if
722 LIBFFI_HIDE_BASIC_TYPES is defined.
723 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
725 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
726 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
727 (ffi_type_longdouble): Special case for Alpha. Don't define
728 if long double == double.
730 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
731 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
732 (ffi_call, ffi_closure_osf_inner): Likewise.
733 (ffi_closure_osf_inner): Likewise. Mark hidden.
734 (ffi_call_osf, ffi_closure_osf): Mark hidden.
735 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
736 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
737 (load_table): Handle 128-bit long double.
739 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
741 2007-04-06 Tom Tromey <tromey@redhat.com>
744 * README: Fixed bug in example.
746 2007-04-03 Jakub Jelinek <jakub@redhat.com>
748 * src/closures.c: Include sys/statfs.h.
749 (_GNU_SOURCE): Define on Linux.
750 (FFI_MMAP_EXEC_SELINUX): Define.
751 (selinux_enabled): New variable.
752 (selinux_enabled_check): New function.
753 (is_selinux_enabled): Define.
756 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
758 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
759 Use 'volatile float sum' to create sum of floats to avoid false
760 negative due to excess precision on ix86 targets.
763 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
765 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
767 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
769 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
771 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
772 (ffi_prep_closure_loc): New.
773 (ffi_prep_raw_closure_loc): New.
774 (ffi_prep_java_raw_closure_loc): New.
775 * src/closures.c: New file.
776 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
777 Replace sflags with exec_offset.
778 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
779 sub_segment_exec_offset): New macros.
780 (get_segment_flags, set_segment_flags, check_segment_merge): New
782 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
783 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
784 destroy_mspace): Use new macros.
785 (sys_alloc): Silence warning.
786 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
787 * Makefile.in: Rebuilt.
788 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
789 terms of ffi_prep_closure_loc.
790 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
792 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
794 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
796 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
798 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
799 (ffi_prep_closure): ... this.
800 * src/pa/ffi.c: Likewise.
801 * src/cris/ffi.c: Likewise. Adjust.
802 * src/frv/ffi.c: Likewise.
803 * src/ia64/ffi.c: Likewise.
804 * src/mips/ffi.c: Likewise.
805 * src/powerpc/ffi_darwin.c: Likewise.
806 * src/s390/ffi.c: Likewise.
807 * src/sh/ffi.c: Likewise.
808 * src/sh64/ffi.c: Likewise.
809 * src/sparc/ffi.c: Likewise.
810 * src/x86/ffi64.c: Likewise.
811 * src/x86/ffi.c: Likewise.
812 (FFI_INIT_TRAMPOLINE): Adjust.
813 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
814 (ffi_prep_raw_closure): ... this.
815 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
816 (ffi_prep_closure): ... this.
817 (flush_icache): Adjust.
819 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
821 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
824 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
826 * Makefile.am: Add dummy install-pdf target.
827 * Makefile.in: Regenerate
829 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
831 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
832 ffi_closure_helper_SYSV): Add long double handling.
834 2007-02-02 Jakub Jelinek <jakub@redhat.com>
836 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
837 immediately after bctrl instruction.
839 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
841 * Makefile.am (all-recursive, install-recursive,
842 mostlyclean-recursive, clean-recursive, distclean-recursive,
843 maintainer-clean-recursive): Add missing targets.
844 * Makefile.in: Rebuilt.
846 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
848 * configure.ac: Add TARGET for x86_64-*-darwin*.
849 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
851 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
852 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
853 * configure: Regenerate.
854 * Makefile.in: Regenerate.
855 * include/Makefile.in: Regenerate.
856 * testsuite/Makefile.in: Regenerate.
857 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
860 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
862 * aclocal.m4: Regenerate with aclocal -I .. as written in the
865 2006-10-31 Geoffrey Keating <geoffk@apple.com>
867 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
868 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
870 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
871 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
873 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
874 Sandro Tolaini <tolaini@libero.it>
876 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
878 * configure: Regenerated.
879 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
880 (EXTRA_DIST): Add src/x86/darwin.S.
881 * Makefile.in: Regenerated.
882 * include/Makefile.in: Regenerated.
883 * testsuite/Makefile.in: Regenerated.
885 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
886 X86_WIN32, and additionally align stack to 16 bytes.
887 * src/x86/darwin.S: New, based on sysv.S.
888 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
890 2006-09-12 David Daney <ddaney@avtrex.com>
893 * include/Makefile.am: Install both ffi.h and ffitarget.h in
894 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
895 * aclocal.m4: Regenerated for automake 1.9.6.
896 * Makefile.in: Regenerated.
897 * include/Makefile.in: Regenerated.
898 * testsuite/Makefile.in: Regenerated.
900 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
902 * include/ffi_common.h (struct): Revert accidental commit.
904 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
906 * include/ffi_common.h: Remove lint directives.
907 * include/ffi.h.in: Likewise.
909 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
911 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
912 for 32-bit architectures.
913 * testsuite/libffi.call/return_ul.c: New test case.
915 2006-07-19 David Daney <ddaney@avtrex.com>
917 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
918 xfail remains for mips64.
920 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
922 * Makefile.am: Add install-html target. Add install-html to .PHONY
923 * Makefile.in: Regenerate.
924 * aclocal.m4: Regenerate.
925 * include/Makefile.in: Regenerate.
926 * testsuite/Makefile.in: Regenerate.
928 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
930 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
933 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
935 * README: Remove notice about 'Crazy Comments'.
936 * src/debug.c: Remove lint directives. Cleanup white spaces.
937 * src/java_raw_api.c: Likewise.
938 * src/prep_cif.c: Likewise.
939 * src/raw_api.c: Likewise.
940 * src/ffitest.c: Delete. No longer needed, all test cases migrated
942 * src/arm/ffi.c: Remove lint directives.
943 * src/m32r/ffi.c: Likewise.
944 * src/pa/ffi.c: Likewise.
945 * src/powerpc/ffi.c: Likewise.
946 * src/powerpc/ffi_darwin.c: Likewise.
947 * src/sh/ffi.c: Likewise.
948 * src/sh64/ffi.c: Likewise.
949 * src/x86/ffi.c: Likewise.
950 * testsuite/libffi.call/float2.c: Likewise.
951 * testsuite/libffi.call/promotion.c: Likewise.
952 * testsuite/libffi.call/struct1.c: Likewise.
954 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
956 * src/pa/hpux32.S: Correct unwind offset calculation for
958 * src/pa/linux.S: Likewise.
960 2006-04-12 James E Wilson <wilson@specifix.com>
963 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
964 (hfa_type_load): Call stf_spill.
965 (hfa_type_store): Call ldf_fill.
966 (ffi_call): Adjust calls to above routines. Add local temps for
969 2006-04-10 Matthias Klose <doko@debian.org>
971 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
972 directory names containing underscores.
974 2006-04-07 James E Wilson <wilson@specifix.com>
976 * testsuite/libffi.call/float4.c: New testcase.
978 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
979 Andreas Tobler <a.tobler@schweiz.ch>
981 * Makefile.am: Add PA_HPUX port.
982 * Makefile.in: Regenerate.
983 * include/Makefile.in: Likewise.
984 * testsuite/Makefile.in: Likewise.
985 * configure.ac: Add PA_HPUX rules.
986 * configure: Regenerate.
987 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
988 Add PA_HPUX and PA64_HPUX.
989 Rename FFI_LINUX ABI to FFI_PA32 ABI.
990 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
991 (FFI_TYPE_SMALL_STRUCT2): Define.
992 (FFI_TYPE_SMALL_STRUCT4): Likewise.
993 (FFI_TYPE_SMALL_STRUCT8): Likewise.
994 (FFI_TYPE_SMALL_STRUCT3): Redefine.
995 (FFI_TYPE_SMALL_STRUCT5): Likewise.
996 (FFI_TYPE_SMALL_STRUCT6): Likewise.
997 (FFI_TYPE_SMALL_STRUCT7): Likewise.
998 * src/pa/ffi.c (ROUND_DOWN): Delete.
999 (fldw, fstw, fldd, fstd): Use '__asm__'.
1000 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1001 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1002 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1003 Simplify incrementing of stack slot variable. Change type of local
1004 'n' to unsigned int.
1005 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1007 (ffi_prep_cif_machdep): Likewise.
1008 (ffi_call): Likewise.
1009 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1010 return type to ffi_status. Simplify incrementing of stack slot
1011 variable. Only copy floating point argument registers when PA_LINUX
1012 is true. Reformat debug statement.
1013 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1014 FFI_TYPE_SMALL_STRUCT8.
1015 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1017 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1018 Add nops to cache flush. Add trampoline for PA_HPUX.
1019 * src/pa/hpux32.S: New file.
1020 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1021 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1022 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1023 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1024 argument type checks so that common argument types appear first.
1025 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1026 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1028 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1030 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1031 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1032 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1033 FFI_TYPE_LONGDOUBLE.
1034 (ffi_prep_args64): Assert using IBM extended double.
1035 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1036 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1037 (ffi_call): Handle FFI_LINUX.
1038 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1039 gpr3 return pointer as for struct return. Handle FFI_LINUX
1040 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1042 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1043 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1044 Don't use r6 as pointer to results, instead use sp offset. Don't
1045 make a special call to load lr with case table address, instead
1046 use offset from previous call.
1047 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1048 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1051 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1053 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1054 passed with FP registers correctly.
1055 (ffi_closure_helper_SYSV): Likewise.
1056 * src/sh64/sysv.S: Likewise.
1058 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1060 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1061 args and userdata unused.
1062 (closure_test_fn1): Mark cif and userdata unused.
1063 (main): Remove unused res.
1065 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1067 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1068 -O2, -O3, -Os and the warning flags -W -Wall.
1069 * testsuite/libffi.special/special.exp: Likewise.
1070 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1071 unused parameter unused for gcc or else do nothing.
1072 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1073 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1074 and userdata unused.
1075 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1076 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1077 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1078 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1079 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1080 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1081 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1082 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1083 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1084 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1085 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1086 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1087 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1088 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1089 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1090 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1091 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1092 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1093 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1094 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1095 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1096 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1097 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1098 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1100 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1102 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1104 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1105 void* to avoid compiler warning.
1107 (cls_struct_align_gn): Mark cif and userdata unused.
1108 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1110 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1112 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1114 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1116 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1118 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1119 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1120 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1122 (main): Cast res_call to silence gcc.
1123 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1125 (main): Cast res_call to silence gcc.
1126 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1128 (main): Cast res_call to silence gcc.
1129 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1131 (main): Cast res_call to silence gcc.
1132 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1134 (main): Cast res_call to silence gcc.
1135 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1137 (main): Cast res_call to silence gcc.
1138 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1140 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1141 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1143 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1144 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1146 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1147 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1149 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1150 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1152 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1153 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1154 and userdata unused.
1155 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1157 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1158 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1159 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1160 Cleanup white spaces.
1161 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1162 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1163 cif and userdata unused.
1164 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1166 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1167 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1168 formatters to silence gcc.
1169 (B_gn): Mark cif and userdata unused.
1170 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1172 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1173 (B_gn): Mark cif and userdata unused.
1174 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1176 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1177 (B_gn): Mark cif and userdata unused.
1178 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1180 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1181 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1182 * testsuite/libffi.call/problem1.c (stub): Likewise.
1183 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1185 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1186 in the last commit for this test case in the test case itself.
1187 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1189 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1190 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1191 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1192 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1193 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1194 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1196 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1198 * src/sh/sysv.S: Fix register numbers in the FDE for
1201 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1203 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1204 declaration to avoid a false negative on ix86. See PR323.
1206 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1208 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1209 and cast integer to void * if needed. Update the pointer to
1210 the FP register saved area correctly.
1212 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1214 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1216 * testsuite/libffi.call/nested_struct4.c: Likewise.
1218 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1220 * testsuite/libffi.call/return_dbl.c: New test case.
1221 * testsuite/libffi.call/return_dbl1.c: Likewise.
1222 * testsuite/libffi.call/return_dbl2.c: Likewise.
1223 * testsuite/libffi.call/return_fl.c: Likewise.
1224 * testsuite/libffi.call/return_fl1.c: Likewise.
1225 * testsuite/libffi.call/return_fl2.c: Likewise.
1226 * testsuite/libffi.call/return_fl3.c: Likewise.
1227 * testsuite/libffi.call/closure_fn6.c: Likewise.
1229 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1231 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1232 here to be used by other test cases too.
1234 * testsuite/libffi.call/nested_struct10.c: New test case.
1235 * testsuite/libffi.call/nested_struct9.c: Likewise.
1236 * testsuite/libffi.call/nested_struct8.c: Likewise.
1237 * testsuite/libffi.call/nested_struct7.c: Likewise.
1238 * testsuite/libffi.call/nested_struct6.c: Likewise.
1239 * testsuite/libffi.call/nested_struct5.c: Likewise.
1240 * testsuite/libffi.call/nested_struct4.c: Likewise.
1242 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1244 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1245 * configure: Rebuilt.
1247 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1249 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1250 instead do the shifting inline.
1251 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1252 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1253 and 8 byte structs, for the remaining struct sizes don't call
1254 __lshrdi3, instead do the shifting inline.
1256 2005-12-07 Thiemo Seufer <ths@networkno.de>
1258 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1259 missing parentheses.
1260 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1261 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1262 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1263 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1264 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1266 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1268 (ffi_prep_closure): Improve trampoline instruction scheduling.
1269 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1271 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1273 * src/powerpc/ffi.c: Formatting.
1274 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1275 (ffi_prep_args64): Likewise.
1277 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1279 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1280 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1282 2005-09-26 Tom Tromey <tromey@redhat.com>
1284 * testsuite/libffi.call/float1.c (value_type): New typedef.
1285 (CANARY): New define.
1286 (main): Check for result buffer overflow.
1287 * src/powerpc/linux64.S: Handle linux64 long double returns.
1288 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1289 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1291 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1294 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1296 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1298 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1300 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1301 (AH_BOTTOM): Add FFI_HIDDEN definition.
1302 * configure: Rebuilt.
1303 * fficonfig.h.in: Rebuilt.
1304 * src/powerpc/ffi.c (hidden): Remove.
1305 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1306 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1307 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1308 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1309 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1310 add FFI_HIDDEN to its prototype.
1311 (ffi_closure_SYSV_inner): New.
1312 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1313 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1315 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1318 * configure: Rebuilt.
1319 * configure.ac: Handle i*86-*-gnu*.
1321 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1323 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1324 DW_CFA_offset_extended_sf rather than
1325 DW_CFA_GNU_negative_offset_extended.
1326 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1328 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1330 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1332 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1333 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1334 partially on register.
1335 (ffi_closure_helper_SYSV): Likewise.
1336 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1338 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1340 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1342 * src/sh64/ffi.c (simple_type): Remove.
1343 (return_type): Handle small structures correctly.
1344 (ffi_prep_args): Likewise.
1345 (ffi_call): Likewise.
1346 (ffi_closure_helper_SYSV): Likewise.
1347 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1348 Emit position independent code if PIC and remove wrong datalabel
1349 prefixes from EH data.
1351 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1353 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1354 * Makefile.in: Regenerate.
1355 * include/Makefile.in: Likewise.
1356 * testsuite/Makefile.in: Likewise.
1357 * configure.ac: Add POWERPC_FREEBSD rules.
1358 * configure: Regenerate.
1359 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1360 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1361 * src/powerpc/ffi.c: Add flags to handle small structure returns
1363 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1365 (ffi_closure_helper_SYSV): Likewise.
1366 * src/powerpc/ppc_closure.S: Add return types for small structures.
1367 * src/powerpc/sysv.S: Add bits to handle small structures for
1370 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1372 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1373 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
1374 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
1376 2005-07-05 Randolph Chung <tausq@debian.org>
1378 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
1379 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
1380 structures. Kill compilation warnings.
1381 (ffi_closure_inner_LINUX): Print return values as hex in debug
1382 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
1383 Properly handle 5-7 byte structure returns.
1384 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
1385 (FFI_TYPE_SMALL_STRUCT2): Remove.
1386 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
1387 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
1388 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
1389 (checksmst1, checksmst2): Remove.
1390 (checksmst3): Optimize handling of 3-byte struct returns.
1391 (checksmst567): Properly handle 5-7 byte struct returns.
1393 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1396 * src/mips/n32.S: Enforce PIC code.
1397 * src/mips/o32.S: Likewise.
1399 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1401 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
1402 * configure: Regenerate.
1404 2005-06-01 Alan Modra <amodra@bigpond.net.au>
1406 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
1407 to call ffi_closure_helper_SYSV. Append @local instead.
1408 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
1410 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
1412 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
1413 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
1414 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
1415 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
1416 include/Makefile.in, testsuite/Makefile.in: Regenerate.
1418 2005-05-09 Mike Stump <mrs@apple.com>
1420 * configure: Regenerate.
1422 2005-05-08 Richard Henderson <rth@redhat.com>
1425 * src/alpha/osf.S: Update unwind into to match code.
1427 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
1428 Richard Henderson <rth@redhat.com>
1430 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
1432 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
1433 (ffi_prep_closure): Set carry bit if sse-used flag set.
1434 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
1435 Only load sse registers if ssecount non-zero.
1436 (ffi_closure_unix64): Only save sse registers if carry set on entry.
1438 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
1440 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
1441 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
1442 * configure: Regenerate.
1444 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
1446 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
1447 have Tcl8.3-compatible intermediate variable.
1449 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
1450 Hans-Peter Nilsson <hp@axis.com>
1452 * Makefile.am: Add CRIS support.
1453 * configure.ac: Likewise.
1454 * Makefile.in, configure, testsuite/Makefile.in,
1455 include/Makefile.in: Regenerate.
1456 * src/cris: New directory.
1457 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
1458 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
1460 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
1461 \r?\n in output tests.
1463 2005-04-12 Mike Stump <mrs@apple.com>
1465 * configure: Regenerate.
1467 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
1469 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
1471 2005-03-30 Steve Ellcey <sje@cup.hp.com>
1473 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
1475 * src/ia64/unix.S (ffi_closure_unix): Extend gp
1476 to 64 bits in ILP32 mode.
1477 Load 64 bits even for short data.
1479 2005-03-23 Mike Stump <mrs@apple.com>
1481 * src/powerpc/darwin.S: Update for -m64 multilib.
1482 * src/powerpc/darwin_closure.S: Likewise.
1484 2005-03-21 Zack Weinberg <zack@codesourcery.com>
1486 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
1487 Do not set tool_include_dir.
1488 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
1490 * include/Makefile.am: Set gcc_version and toollibffidir.
1491 * include/Makefile.in: Regenerate.
1493 2005-02-22 Andrew Haley <aph@redhat.com>
1495 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
1496 odd-numbered register pairs for 64-bit integer types.
1498 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
1501 * testsuite/libffi.call/return_ll1.c: New test case.
1503 2005-02-11 Janis Johnson <janis187@us.ibm.com>
1505 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
1506 * testsuite/libffi.call/float.c: Ditto.
1507 * testsuite/libffi.call/float2.c: Ditto.
1508 * testsuite/libffi.call/float3.c: Ditto.
1510 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
1512 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
1514 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
1516 * testsuite/libffi.special/special.exp (cxx_options): Add
1519 2004-12-31 Richard Henderson <rth@redhat.com>
1521 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
1522 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
1523 offset parameters with a type parameter; deduce size and structure
1524 alignment. Update all users.
1526 2004-12-31 Richard Henderson <rth@redhat.com>
1528 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
1529 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
1530 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
1531 into ffi_prep_closure.
1532 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
1535 2004-12-27 Richard Henderson <rth@redhat.com>
1537 * src/x86/unix64.S: Fix typo in unwind info.
1539 2004-12-25 Richard Henderson <rth@redhat.com>
1541 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
1542 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
1543 (merge_classes): Check for it.
1545 (classify_argument): Pass byte_offset by value; perform all updates
1547 (examine_argument): Add classes argument; handle
1548 X86_64_COMPLEX_X87_CLASS.
1549 (ffi_prep_args): Merge into ...
1550 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
1551 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
1552 (ffi_fill_return_value): Remove.
1553 (ffi_prep_closure): Remove dead assert.
1554 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
1555 Rewrite to use struct register_args instead of va_list. Create
1556 flags for handling structure returns.
1557 * src/x86/unix64.S: Remove dead strings.
1558 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
1559 stack frame with ffi_call. Handle structure returns properly.
1560 (float2sse, floatfloat2sse, double2sse): Remove.
1561 (sse2float, sse2double, sse2floatfloat): Remove.
1562 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
1563 to handle structure returns properly.
1565 2004-12-08 David Edelsohn <edelsohn@gnu.org>
1567 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
1569 * Makefile.in: Regenerated.
1571 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
1573 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
1574 * configure, aclocal.m4, Makefile.in: Regenerate.
1575 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
1577 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
1579 * configure: Regenerate for libtool change.
1581 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
1583 * configure: Regenerate for libtool reversion.
1585 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
1587 * configure: Regenerate for libtool change.
1589 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1591 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
1593 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
1595 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
1596 of jal. Use an absolute encoding for the frame information.
1598 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
1600 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
1601 * acinclude.m4: Delete logic for sincludes.
1602 * aclocal.m4, Makefile.in, configure: Regenerate.
1603 * include/Makefile: Likewise.
1604 * testsuite/Makefile: Likewise.
1606 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
1608 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
1609 on a 8-byte boundary.
1610 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
1612 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1614 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
1615 long long values. Round stack allocation to a multiple of 8 bytes
1616 for ATPCS compatibility.
1617 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
1618 names. Handle returning long long types. Add Thumb and interworking
1619 support. Improve soft-float code.
1621 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
1623 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
1624 (libffi_exit): New function.
1625 (libffi_init): Build the testglue wrapper if needed.
1627 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
1630 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
1632 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1634 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
1636 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
1638 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
1639 * testsuite/libffi.call/float3.c: New test case.
1641 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
1643 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
1644 the function returning a structure pointed with R2.
1645 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
1646 the structure return value if T bit set. Emit position
1647 independent code and EH data if PIC.
1649 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
1651 * Makefile.am: Add m32r support.
1652 * configure.ac: Likewise.
1653 * Makefile.in: Regenerate.
1654 * confiugre: Regenerate.
1655 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
1656 (uint64, sint64, double, longdouble)
1657 * src/m32r: New directory.
1658 * src/m32r/ffi.c: New file.
1659 * src/m32r/sysv.S: Likewise.
1660 * src/m32r/ffitarget.h: Likewise.
1662 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
1664 * testsuite/libffi.call/negint.c: New test case.
1666 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
1669 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
1670 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
1671 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
1674 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1676 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
1677 * testsuite/libffi.call/promotion.c: Likewise.
1678 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
1680 * testsuite/libffi.call/return_sc.c: Likewise.
1681 * testsuite/libffi.call/return_uc.c: Likewise.
1683 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
1685 * src/powerpc/darwin.S: Fix comments and identation.
1686 * src/powerpc/darwin_closure.S: Likewise.
1688 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1690 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
1691 (ffi_prep_args): Handle longdouble arguments.
1692 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
1694 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
1695 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
1697 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
1698 * src/types.c: Defined longdouble size and alignment for darwin.
1700 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
1702 * src/powerpc/aix.S: Remove whitespaces.
1703 * src/powerpc/aix_closure.S: Likewise.
1704 * src/powerpc/asm.h: Likewise.
1705 * src/powerpc/ffi.c: Likewise.
1706 * src/powerpc/ffitarget.h: Likewise.
1707 * src/powerpc/linux64.S: Likewise.
1708 * src/powerpc/linux64_closure.S: Likewise.
1709 * src/powerpc/ppc_closure.S: Likewise.
1710 * src/powerpc/sysv.S: Likewise.
1712 2004-08-30 Anthony Green <green@redhat.com>
1714 * Makefile.am: Add frv support.
1715 * Makefile.in, testsuite/Makefile.in: Rebuilt.
1716 * configure.ac: Read configure.host.
1717 * configure.in: Read configure.host.
1718 * configure.host: New file. frv-elf needs libgloss.
1719 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
1720 alignment. This is needed to frv and shouldn't harm the others.
1721 * include/ffi_common.h (ALIGN_DOWN): New macro.
1722 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
1724 2004-08-24 David Daney <daney@avtrex.com>
1726 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
1727 * testsuite/libffi.call/closure_fn1.c: Likewise.
1728 * testsuite/libffi.call/closure_fn2.c Likewise.
1729 * testsuite/libffi.call/closure_fn3.c: Likewise.
1730 * testsuite/libffi.call/closure_fn4.c: Likewise.
1731 * testsuite/libffi.call/closure_fn5.c: Likewise.
1732 * testsuite/libffi.call/cls_18byte.c: Likewise.
1733 * testsuite/libffi.call/cls_19byte.c: Likewise.
1734 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
1735 * testsuite/libffi.call/cls_20byte.c: Likewise.
1736 * testsuite/libffi.call/cls_20byte1.c: Likewise.
1737 * testsuite/libffi.call/cls_24byte.c: Likewise.
1738 * testsuite/libffi.call/cls_2byte.c: Likewise.
1739 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
1740 * testsuite/libffi.call/cls_3byte1.c: Likewise.
1741 * testsuite/libffi.call/cls_3byte2.c: Likewise.
1742 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
1743 * testsuite/libffi.call/cls_4byte.c: Likewise.
1744 * testsuite/libffi.call/cls_64byte.c: Likewise.
1745 * testsuite/libffi.call/cls_6byte.c: Likewise.
1746 * testsuite/libffi.call/cls_7byte.c: Likewise.
1747 * testsuite/libffi.call/cls_8byte.c: Likewise.
1748 * testsuite/libffi.call/cls_9byte1.c: Likewise.
1749 * testsuite/libffi.call/cls_9byte2.c: Likewise.
1750 * testsuite/libffi.call/cls_align_double.c: Likewise.
1751 * testsuite/libffi.call/cls_align_float.c: Likewise.
1752 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
1753 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
1754 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
1755 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
1756 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
1757 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
1758 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
1759 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
1760 * testsuite/libffi.call/cls_double.c: Likewise.
1761 * testsuite/libffi.call/cls_float.c: Likewise.
1762 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
1763 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
1764 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
1765 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
1766 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
1767 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
1768 * testsuite/libffi.call/cls_schar.c: Likewise.
1769 * testsuite/libffi.call/cls_sint.c: Likewise.
1770 * testsuite/libffi.call/cls_sshort.c: Likewise.
1771 * testsuite/libffi.call/cls_uchar.c: Likewise.
1772 * testsuite/libffi.call/cls_uint.c: Likewise.
1773 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
1774 * testsuite/libffi.call/cls_ushort.c: Likewise.
1775 * testsuite/libffi.call/nested_struct.c: Likewise.
1776 * testsuite/libffi.call/nested_struct1.c: Likewise.
1777 * testsuite/libffi.call/nested_struct2.c: Likewise.
1778 * testsuite/libffi.call/nested_struct3.c: Likewise.
1779 * testsuite/libffi.call/problem1.c: Likewise.
1780 * testsuite/libffi.special/unwindtest.cc: Likewise.
1781 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
1783 * testsuite/libffi.call/cls_16byte.c: Likewise.
1784 * testsuite/libffi.call/cls_5byte.c: Likewise.
1786 2004-08-23 David Daney <daney@avtrex.com>
1789 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
1790 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
1791 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
1792 parameters and return types.
1793 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
1794 (ffi_prep_closure): Ditto.
1795 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
1796 alignment calculations.
1797 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
1798 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
1800 2004-08-14 Casey Marshall <csm@gnu.org>
1802 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
1803 contain `FFI_TYPE_UINT64' as return type for any 64-bit
1804 integer (O32 ABI only).
1805 (ffi_prep_closure): new function.
1806 (ffi_closure_mips_inner_O32): new function.
1807 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
1808 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
1809 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
1810 64 bit integers correctly.
1811 (ffi_closure_O32): new function.
1812 Added DWARF-2 unwind info for both functions.
1814 2004-08-10 Andrew Haley <aph@redhat.com>
1816 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
1818 2004-08-01 Robert Millan <robertmh@gnu.org>
1820 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
1821 * configure: Regenerate.
1823 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
1825 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
1826 and mmap() explicitly instead of relying on preset autoconf cache
1828 * aclocal.m4: Regenerate.
1829 * configure: Regenerate.
1831 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
1833 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
1834 (ffi_check_float_struct): Remove unused prototype.
1836 2004-06-30 Geoffrey Keating <geoffk@apple.com>
1838 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
1839 character on Darwin, use '\n\t' instead.
1841 2004-06-26 Matthias Klose <doko@debian.org>
1843 * libtool-version: Fix typo in revision/age.
1845 2004-06-17 Matthias Klose <doko@debian.org>
1847 * libtool-version: New.
1848 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
1849 * Makefile.in: Regenerate.
1851 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1853 * Makefile.am: Remove useless multilib rules.
1854 * Makefile.in: Regenerate.
1855 * aclocal.m4: Regenerate with automake 1.8.5.
1856 * configure.ac: Remove useless multilib configury.
1857 * configure: Regenerate.
1859 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
1861 * .cvsignore: New file.
1863 2004-06-10 Jakub Jelinek <jakub@redhat.com>
1865 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
1867 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
1870 2004-06-06 Sean McNeil <sean@mcneil.com>
1872 * configure.ac: Add x86_64-*-freebsd* support.
1873 * configure: Regenerate.
1875 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
1878 * configure.ac: Test for existence of mmap and sys/mman.h before
1879 checking blacklist. Fix suggested by Jim Wilson.
1880 * configure: Regenerate.
1882 2004-04-26 Matt Austern <austern@apple.com>
1884 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
1886 * src/powerpc/darwin_closure.S: Likewise.
1888 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
1890 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
1891 error. Reported by Thomas Heller <theller@python.net>.
1892 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
1893 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
1895 2004-03-20 Matthias Klose <doko@debian.org>
1897 * src/pa/linux.S: Fix typo.
1899 2004-03-19 Matthias Klose <doko@debian.org>
1901 * Makefile.am: Update.
1902 * Makefile.in: Regenerate.
1903 * src/pa/ffi.h.in: Remove.
1904 * src/pa/ffitarget.h: New file.
1906 2004-02-10 Randolph Chung <tausq@debian.org>
1908 * Makefile.am: Add PA support.
1909 * Makefile.in: Regenerate.
1910 * include/Makefile.in: Regenerate.
1911 * configure.ac: Add PA target.
1912 * configure: Regenerate.
1913 * src/pa/ffi.c: New file.
1914 * src/pa/ffi.h.in: Add PA support.
1915 * src/pa/linux.S: New file.
1916 * prep_cif.c: Add PA support.
1918 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
1920 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
1922 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
1923 with ecif->cif->flags.
1924 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
1926 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
1927 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
1928 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
1929 sc_retstruct2b): Add for 1 or 2-bytes struct case.
1931 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
1933 * configure.in: Rename file to ...
1934 * configure.ac: ... this.
1935 * fficonfig.h.in: Regenerate.
1936 * Makefile.in: Regenerate.
1937 * include/Makefile.in: Regenerate.
1938 * testsuite/Makefile.in: Regenerate.
1940 2004-03-12 Matt Austern <austern@apple.com>
1942 * src/powerpc/darwin.S: Fix EH information so it corresponds to
1943 changes in EH format resulting from addition of linkonce support.
1944 * src/powerpc/darwin_closure.S: Likewise.
1946 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
1947 Paolo Bonzini <bonzini@gnu.org>
1949 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
1950 Remove VPATH. Remove rules for object files. Remove multilib support.
1951 (AM_CCASFLAGS): Add.
1952 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
1953 (AC_PREREQ): Bump version to 2.59.
1954 (AC_INIT): Fill with version info and bug address.
1955 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
1956 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
1957 De-precious CC so that the right flags are passed down to multilibs.
1958 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
1959 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
1960 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
1961 * configure: Rebuilt.
1962 * aclocal.m4: Likewise.
1963 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
1964 * fficonfig.h.in: Likewise.
1966 2004-03-11 Andreas Schwab <schwab@suse.de>
1968 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
1969 arguments from fp registers only for the first 8 parameter slots.
1970 Don't convert a float parameter when passed in memory.
1972 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
1974 * configure: Regenerate for config/accross.m4 correction.
1976 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
1978 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
1979 ecif->cif->bytes to bytes.
1980 (ffi_prep_cif_machdep): Add braces around nested if statement.
1982 2004-02-09 Alan Modra <amodra@bigpond.net.au>
1984 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
1986 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
1987 (ffi_closure_helper_LINUX64): Fix typo.
1988 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
1990 * testsuite/libffi.call/float.c: Likewise.
1991 * testsuite/libffi.call/float2.c: Likewise.
1993 2004-02-08 Alan Modra <amodra@bigpond.net.au>
1995 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
1996 long double function return and long double arg handling.
1997 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
1998 Use "end_pfr" instead of "nf". Correct long double handling.
2000 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2002 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2003 space for long double return value. Adjust stack frame and offsets.
2004 Load f2 long double return.
2006 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2008 * src/types.c: Use 16 byte long double for POWERPC64.
2010 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2012 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2013 when the structure return address is passed in %o0.
2014 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2015 (ffi_v9_layout_struct): Align the field following a nested structure
2016 on a word boundary. Use memmove instead of memcpy.
2017 (ffi_call): Update call to ffi_V9_return_struct.
2018 (ffi_prep_closure): Define 'ctx' only for V8.
2019 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2020 and ffi_closure_sparc_inner_v9.
2021 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2022 Always skip the structure return address. For structures and long
2023 doubles, copy the argument directly.
2024 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2025 if required. Shift the maximum floating-point slot accordingly. For
2026 big structures, copy the argument directly; otherwise, left-justify the
2027 argument and call ffi_v9_layout_struct to lay out the structure on
2029 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2030 (ffi_closure_v8): Pass the structure return address. Update call to
2031 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2032 Skip the 'unimp' insn when returning long doubles and structures.
2033 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2034 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2035 FFI_TYPE_INT handling. Load structures both in integers and
2036 floating-point registers on return.
2037 * README: Update status of the SPARC port.
2039 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2041 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2043 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2045 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2047 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2048 value as of type ffi_arg, not unsigned int.
2050 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2052 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2055 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2057 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2060 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2062 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2065 2003-12-10 Richard Henderson <rth@redhat.com>
2067 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2068 size_t instead of int.
2070 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2072 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2073 * testsuite/libffi.call/many_win32.c (main): Replace variable
2074 int i with unsigned long ul.
2076 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2077 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2078 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2079 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2080 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2081 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2082 * testsuite/libffi.call/cls_align_float.c: Likewise.
2083 * testsuite/libffi.call/cls_align_double.c: Likewise.
2084 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2085 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2087 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2090 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2091 Align arguments to 32 bits.
2093 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2096 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2097 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2098 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2099 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2100 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2101 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2103 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2105 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2107 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2108 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2110 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2112 * Makefile.in: Rebuilt.
2113 * configure: Likewise.
2114 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2117 2003-11-21 Andreas Jaeger <aj@suse.de>
2118 Andreas Tobler <a.tobler@schweiz.ch>
2120 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2121 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2122 * Makefile.in: Rebuilt.
2123 * aclocal.m4: Likewise.
2124 * configure: Likewise.
2125 * fficonfig.h.in: Likewise.
2126 * testsuite/lib/libffi-dg.exp: Add include dir.
2127 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2128 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2129 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2130 for ffi_closure if available.
2131 * testsuite/libffi.call/closure_fn1.c: Likewise.
2132 * testsuite/libffi.call/closure_fn2.c: Likewise.
2133 * testsuite/libffi.call/closure_fn3.c: Likewise.
2134 * testsuite/libffi.call/closure_fn4.c: Likewise.
2135 * testsuite/libffi.call/closure_fn5.c: Likewise.
2136 * testsuite/libffi.call/cls_12byte.c: Likewise.
2137 * testsuite/libffi.call/cls_16byte.c: Likewise.
2138 * testsuite/libffi.call/cls_18byte.c: Likewise.
2139 * testsuite/libffi.call/cls_19byte.c: Likewise.
2140 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2141 * testsuite/libffi.call/cls_20byte.c: Likewise.
2142 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2143 * testsuite/libffi.call/cls_24byte.c: Likewise.
2144 * testsuite/libffi.call/cls_2byte.c: Likewise.
2145 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2146 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2147 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2148 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2149 * testsuite/libffi.call/cls_4byte.c: Likewise.
2150 * testsuite/libffi.call/cls_5byte.c: Likewise.
2151 * testsuite/libffi.call/cls_64byte.c: Likewise.
2152 * testsuite/libffi.call/cls_6byte.c: Likewise.
2153 * testsuite/libffi.call/cls_7byte.c: Likewise.
2154 * testsuite/libffi.call/cls_8byte.c: Likewise.
2155 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2156 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2157 * testsuite/libffi.call/cls_double.c: Likewise.
2158 * testsuite/libffi.call/cls_float.c: Likewise.
2159 * testsuite/libffi.call/cls_schar.c: Likewise.
2160 * testsuite/libffi.call/cls_sint.c: Likewise.
2161 * testsuite/libffi.call/cls_sshort.c: Likewise.
2162 * testsuite/libffi.call/cls_uchar.c: Likewise.
2163 * testsuite/libffi.call/cls_uint.c: Likewise.
2164 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2165 * testsuite/libffi.call/cls_ushort.c: Likewise.
2166 * testsuite/libffi.call/nested_struct.c: Likewise.
2167 * testsuite/libffi.call/nested_struct1.c: Likewise.
2168 * testsuite/libffi.call/nested_struct2.c: Likewise.
2169 * testsuite/libffi.call/nested_struct3.c: Likewise.
2170 * testsuite/libffi.call/problem1.c: Likewise.
2171 * testsuite/libffi.special/unwindtest.cc: Likewise.
2173 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2175 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2177 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2179 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2180 Add -lgcc_s to additional flags.
2182 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2184 * configure.in, include/Makefile.am: PR libgcj/11147, install
2185 the ffitarget.h header file in a gcc versioned and target
2187 * configure: Regenerated.
2188 * Makefile.in, include/Makefile.in: Likewise.
2189 * testsuite/Makefile.in: Likewise.
2191 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2193 * testsuite/libffi.call/closure_fn0.c: Print result and check
2194 with dg-output to make debugging easier.
2195 * testsuite/libffi.call/closure_fn1.c: Likewise.
2196 * testsuite/libffi.call/closure_fn2.c: Likewise.
2197 * testsuite/libffi.call/closure_fn3.c: Likewise.
2198 * testsuite/libffi.call/closure_fn4.c: Likewise.
2199 * testsuite/libffi.call/closure_fn5.c: Likewise.
2200 * testsuite/libffi.call/cls_12byte.c: Likewise.
2201 * testsuite/libffi.call/cls_16byte.c: Likewise.
2202 * testsuite/libffi.call/cls_18byte.c: Likewise.
2203 * testsuite/libffi.call/cls_19byte.c: Likewise.
2204 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2205 * testsuite/libffi.call/cls_20byte.c: Likewise.
2206 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2207 * testsuite/libffi.call/cls_24byte.c: Likewise.
2208 * testsuite/libffi.call/cls_2byte.c: Likewise.
2209 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2210 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2211 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2212 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2213 * testsuite/libffi.call/cls_4byte.c: Likewise.
2214 * testsuite/libffi.call/cls_5byte.c: Likewise.
2215 * testsuite/libffi.call/cls_64byte.c: Likewise.
2216 * testsuite/libffi.call/cls_6byte.c: Likewise.
2217 * testsuite/libffi.call/cls_7byte.c: Likewise.
2218 * testsuite/libffi.call/cls_8byte.c: Likewise.
2219 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2220 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2221 * testsuite/libffi.call/cls_double.c: Likewise.
2222 * testsuite/libffi.call/cls_float.c: Likewise.
2223 * testsuite/libffi.call/cls_schar.c: Likewise.
2224 * testsuite/libffi.call/cls_sint.c: Likewise.
2225 * testsuite/libffi.call/cls_sshort.c: Likewise.
2226 * testsuite/libffi.call/cls_uchar.c: Likewise.
2227 * testsuite/libffi.call/cls_uint.c: Likewise.
2228 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2229 * testsuite/libffi.call/cls_ushort.c: Likewise.
2230 * testsuite/libffi.call/problem1.c: Likewise.
2232 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2235 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2237 * testsuite/libffi.call/cls_9byte2.c: New test case.
2238 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2239 * testsuite/libffi.call/cls_64byte.c: Likewise.
2240 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2241 * testsuite/libffi.call/cls_19byte.c: Likewise.
2242 * testsuite/libffi.call/cls_18byte.c: Likewise.
2243 * testsuite/libffi.call/closure_fn4.c: Likewise.
2244 * testsuite/libffi.call/closure_fn5.c: Likewise.
2245 * testsuite/libffi.call/cls_schar.c: Likewise.
2246 * testsuite/libffi.call/cls_sint.c: Likewise.
2247 * testsuite/libffi.call/cls_sshort.c: Likewise.
2248 * testsuite/libffi.call/nested_struct2.c: Likewise.
2249 * testsuite/libffi.call/nested_struct3.c: Likewise.
2251 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2253 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2254 * testsuite/libffi.call/cls_uchar.c: Likewise.
2255 * testsuite/libffi.call/cls_uint.c: Likewise.
2256 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2257 * testsuite/libffi.call/cls_ushort.c: Likewise.
2258 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2260 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2262 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2265 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2267 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2268 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2270 2003-10-22 David Daney <ddaney@avtrex.com>
2272 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2273 __attribute__((__mode__(__SI__))) and friends.
2275 2003-10-22 Andreas Schwab <schwab@suse.de>
2277 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2279 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2281 * configure.in: AC_LINK_FILES(ffitarget.h).
2282 * configure: Regenerate.
2283 * Makefile.in: Likewise.
2284 * include/Makefile.in: Likewise.
2285 * testsuite/Makefile.in: Likewise.
2286 * fficonfig.h.in: Likewise.
2288 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2289 Richard Henderson <rth@redhat.com>
2291 Avoid that ffi.h includes fficonfig.h.
2293 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2294 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2295 (TARGET_SRC_MIPS_SGI): Removed.
2296 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2297 (MIPS_SGI): Removed.
2298 (CLEANFILES): Removed.
2299 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2301 * acconfig.h: Removed.
2302 * configure.in: Compute sizeofs only for double and long double.
2303 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2304 into AC_DEFINE instead of using acconfig.h. Create
2305 include/ffitarget.h instead of include/fficonfig.h. Rename
2306 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2307 AC_DEFINE EH_FRAME_FLAGS.
2308 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2309 (hack_DATA): Add ffitarget.h.
2310 * include/ffi.h.in: Remove all system specific definitions.
2311 Declare raw API even if it is not installed, why bother?
2312 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2313 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2314 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2315 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2316 * include/ffi_common.h (bool): Do not define.
2317 (ffi_assert): Accept failed assertion.
2318 (ffi_type_test): Return void and accept file/line.
2319 (FFI_ASSERT): Pass stringized failed assertion.
2320 (FFI_ASSERT_AT): New macro.
2321 (FFI_ASSERT_VALID_TYPE): New macro.
2322 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2323 UINT64, SINT64): Define here with gcc's __attribute__ macro
2325 (FLOAT32, ALIGN): Define here instead of in ffi.h
2326 * include/ffi-mips.h: Removed. Its content moved to
2327 src/mips/ffitarget.h after separating assembly and C sections.
2328 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2329 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2330 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2331 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2332 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2333 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2334 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2335 (ffi_type_test): Rewritten.
2336 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2337 FFI_ASSERT_VALID_TYPE.
2338 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2339 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2340 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2341 src/s390/ffitarget.h, src/sh/ffitarget.h,
2342 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2343 src/x86/ffitarget.h: New files.
2344 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2345 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2346 src/powerpc/aix.S, src/powerpc/darwin.S,
2347 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2348 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2349 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2350 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2351 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2354 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2356 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2357 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2359 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2361 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2362 Used when FFI_DEBUG = 1.
2364 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2366 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2369 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2371 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2372 FFI_MIPS_O32 for O32 ABI.
2374 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
2376 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
2377 SPARC64. Cleanup whitespaces.
2379 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
2381 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
2382 strongarm, xscale. Cleanup whitespaces.
2383 * testsuite/libffi.call/closure_fn1.c: Likewise.
2384 * testsuite/libffi.call/closure_fn2.c: Likewise.
2385 * testsuite/libffi.call/closure_fn3.c: Likewise.
2386 * testsuite/libffi.call/cls_12byte.c: Likewise.
2387 * testsuite/libffi.call/cls_16byte.c: Likewise.
2388 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2389 * testsuite/libffi.call/cls_20byte.c: Likewise.
2390 * testsuite/libffi.call/cls_24byte.c: Likewise.
2391 * testsuite/libffi.call/cls_2byte.c: Likewise.
2392 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2393 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2394 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2395 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2396 * testsuite/libffi.call/cls_4byte.c: Likewise.
2397 * testsuite/libffi.call/cls_5byte.c: Likewise.
2398 * testsuite/libffi.call/cls_6byte.c: Likewise.
2399 * testsuite/libffi.call/cls_7byte.c: Likewise.
2400 * testsuite/libffi.call/cls_8byte.c: Likewise.
2401 * testsuite/libffi.call/cls_double.c: Likewise.
2402 * testsuite/libffi.call/cls_float.c: Likewise.
2403 * testsuite/libffi.call/cls_uchar.c: Likewise.
2404 * testsuite/libffi.call/cls_uint.c: Likewise.
2405 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2406 * testsuite/libffi.call/cls_ushort.c: Likewise.
2407 * testsuite/libffi.call/nested_struct.c: Likewise.
2408 * testsuite/libffi.call/nested_struct1.c: Likewise.
2409 * testsuite/libffi.call/problem1.c: Likewise.
2410 * testsuite/libffi.special/unwindtest.cc: Likewise.
2411 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
2413 2003-09-18 David Edelsohn <edelsohn@gnu.org>
2415 * src/powerpc/aix.S: Cleanup whitespaces.
2416 * src/powerpc/aix_closure.S: Likewise.
2418 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2420 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
2421 * src/powerpc/darwin_closure.S: Likewise.
2422 * src/powerpc/ffi_darwin.c: Likewise.
2424 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
2425 David Edelsohn <edelsohn@gnu.org>
2427 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
2428 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
2430 * src/powerpc/darwin_closure.S: Likewise.
2431 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
2432 according to the Darwin/AIX ABI.
2433 (ffi_prep_cif_machdep): Likewise.
2434 (ffi_closure_helper_DARWIN): Likewise.
2435 Remove the outgoing parameter stack logic. Simplify the evaluation
2436 of the different CASE types.
2437 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
2438 statement in the trampoline code.
2440 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
2442 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
2443 for the register size.
2444 (ffi_closure_helper_SYSV): Handle the structure return value
2446 (ffi_closure_helper_SYSV): Return the appropriate type when
2447 the registers are used for the structure return value.
2448 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
2449 the 64-bit return value. Update copyright years.
2451 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2453 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
2454 srcdir for ffi_mips.h.
2456 2003-09-12 Alan Modra <amodra@bigpond.net.au>
2458 * src/prep_cif.c (initialize_aggregate): Include tail padding in
2460 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
2461 placement of float result.
2462 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
2463 cast of "resp" for big-endian 64 bit machines.
2465 2003-09-11 Alan Modra <amodra@bigpond.net.au>
2467 * src/types.c (double, longdouble): Merge identical SH and ARM
2468 typedefs, and add POWERPC64.
2469 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
2470 struct split over gpr and rest.
2471 (ffi_prep_cif_machdep): Correct intarg_count for structures.
2472 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
2474 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
2476 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
2479 2003-09-09 Alan Modra <amodra@bigpond.net.au>
2481 * configure: Regenerate.
2483 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2485 * Makefile.am: Remove build rules for ffitest.
2486 * Makefile.in: Rebuilt.
2488 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2490 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
2491 about implicit declaration of abort().
2493 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
2495 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
2496 * Makefile.in: Rebuilt.
2497 * configure.in: Add dejagnu test framework.
2498 * configure: Rebuilt.
2500 * testsuite/Makefile.am: New file.
2501 * testsuite/Makefile.in: Built
2502 * testsuite/lib/libffi-dg.exp: New file.
2503 * testsuite/config/default.exp: Likewise.
2504 * testsuite/libffi.call/call.exp: Likewise.
2505 * testsuite/libffi.call/ffitest.h: Likewise.
2506 * testsuite/libffi.call/closure_fn0.c: Likewise.
2507 * testsuite/libffi.call/closure_fn1.c: Likewise.
2508 * testsuite/libffi.call/closure_fn2.c: Likewise.
2509 * testsuite/libffi.call/closure_fn3.c: Likewise.
2510 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2511 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2512 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2513 * testsuite/libffi.call/cls_2byte.c: Likewise.
2514 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2515 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2516 * testsuite/libffi.call/cls_4byte.c: Likewise.
2517 * testsuite/libffi.call/cls_5byte.c: Likewise.
2518 * testsuite/libffi.call/cls_6byte.c: Likewise.
2519 * testsuite/libffi.call/cls_7byte.c: Likewise.
2520 * testsuite/libffi.call/cls_8byte.c: Likewise.
2521 * testsuite/libffi.call/cls_12byte.c: Likewise.
2522 * testsuite/libffi.call/cls_16byte.c: Likewise.
2523 * testsuite/libffi.call/cls_20byte.c: Likewise.
2524 * testsuite/libffi.call/cls_24byte.c: Likewise.
2525 * testsuite/libffi.call/cls_double.c: Likewise.
2526 * testsuite/libffi.call/cls_float.c: Likewise.
2527 * testsuite/libffi.call/cls_uchar.c: Likewise.
2528 * testsuite/libffi.call/cls_uint.c: Likewise.
2529 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2530 * testsuite/libffi.call/cls_ushort.c: Likewise.
2531 * testsuite/libffi.call/float.c: Likewise.
2532 * testsuite/libffi.call/float1.c: Likewise.
2533 * testsuite/libffi.call/float2.c: Likewise.
2534 * testsuite/libffi.call/many.c: Likewise.
2535 * testsuite/libffi.call/many_win32.c: Likewise.
2536 * testsuite/libffi.call/nested_struct.c: Likewise.
2537 * testsuite/libffi.call/nested_struct1.c: Likewise.
2538 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
2539 * testsuite/libffi.call/problem1.c: Likewise.
2540 * testsuite/libffi.call/promotion.c: Likewise.
2541 * testsuite/libffi.call/return_ll.c: Likewise.
2542 * testsuite/libffi.call/return_sc.c: Likewise.
2543 * testsuite/libffi.call/return_uc.c: Likewise.
2544 * testsuite/libffi.call/strlen.c: Likewise.
2545 * testsuite/libffi.call/strlen_win32.c: Likewise.
2546 * testsuite/libffi.call/struct1.c: Likewise.
2547 * testsuite/libffi.call/struct2.c: Likewise.
2548 * testsuite/libffi.call/struct3.c: Likewise.
2549 * testsuite/libffi.call/struct4.c: Likewise.
2550 * testsuite/libffi.call/struct5.c: Likewise.
2551 * testsuite/libffi.call/struct6.c: Likewise.
2552 * testsuite/libffi.call/struct7.c: Likewise.
2553 * testsuite/libffi.call/struct8.c: Likewise.
2554 * testsuite/libffi.call/struct9.c: Likewise.
2555 * testsuite/libffi.special/special.exp: New file.
2556 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2557 * testsuite/libffi.special/unwindtest.cc: Likewise.
2560 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
2562 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
2565 2003-08-02 Alan Modra <amodra@bigpond.net.au>
2567 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
2569 (ffi_closure_helper_LINUX64): Likewise.
2570 * src/powerpc/linux64.S: Remove code writing to parm save area.
2571 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
2572 address in lr from ffi_closure_helper_LINUX64 call to calculate
2573 table address. Optimize function tail.
2575 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
2577 * src/sparc/ffi.c: Handle all floating point registers.
2578 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
2580 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
2582 * README: Note that libffi is not part of GCC. Update the project
2585 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2587 * src/powerpc/ppc_closure.S: Include ffi.h.
2589 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2591 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
2592 Use C style comments.
2594 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
2596 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
2597 * Makefile.in: Regenerate.
2598 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
2599 * configure: Regenerate.
2600 * include/ffi.h.in: Add SHmedia support.
2601 * src/sh64/ffi.c: New file.
2602 * src/sh64/sysv.S: New file.
2604 2003-05-16 Jakub Jelinek <jakub@redhat.com>
2606 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
2607 should be read-only.
2608 * configure: Rebuilt.
2609 * fficonfig.h.in: Rebuilt.
2610 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
2611 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
2612 * src/powerpc/linux64.S: Likewise.
2613 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
2614 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
2615 if -fpic/-fPIC/-mrelocatable.
2616 * src/powerpc/powerpc_closure.S: Likewise.
2617 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
2618 #write in .eh_frame flags.
2619 * src/sparc/v9.S: Likewise.
2620 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
2621 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
2622 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
2624 2003-05-07 Jeff Sturm <jsturm@one-point.com>
2626 Fixes PR bootstrap/10656
2627 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
2628 support for .register pseudo-op.
2629 * src/sparc/v8.S: Use it.
2630 * fficonfig.h.in: Rebuilt.
2631 * configure: Rebuilt.
2633 2003-04-18 Jakub Jelinek <jakub@redhat.com>
2635 * include/ffi.h.in (POWERPC64): Define if 64-bit.
2636 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
2637 Make it the default on POWERPC64.
2638 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
2639 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
2640 * configure: Rebuilt.
2641 * src/powerpc/ffi.c (hidden): Define.
2642 (ffi_prep_args_SYSV): Renamed from
2643 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
2644 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
2645 ASM_NEEDS_REGISTERS64): New.
2646 (ffi_prep_args64): New function.
2647 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
2648 (ffi_call): Likewise.
2649 (ffi_prep_closure): Likewise.
2650 (flush_icache): Surround by #ifndef POWERPC64.
2651 (ffi_dblfl): New union type.
2652 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
2653 (ffi_closure_helper_LINUX64): New function.
2654 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
2656 * src/powerpc/sysv.S: Likewise.
2657 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
2658 * src/powerpc/linux64.S: New file.
2659 * src/powerpc/linux64_closure.S: New file.
2660 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
2661 src/powerpc/linux64_closure.S.
2662 (TARGET_SRC_POWERPC): Likewise.
2664 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
2665 closure_test_fn3): Fix result printing on big-endian 64-bit
2667 (main): Print tst2_arg instead of uninitialized tst2_result.
2669 * src/ffitest.c (main): Hide what closure pointer really points to
2672 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
2674 * configure.in (arm-*-netbsdelf*): Add configuration.
2675 (configure): Regenerated.
2677 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
2679 * include/Makefile.in: Regenerate.
2681 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
2683 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
2685 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
2686 Receive closure pointer through parameter, read args using
2687 __builtin_dwarf_cfa.
2688 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
2690 2003-03-12 Andreas Schwab <schwab@suse.de>
2692 * configure.in: Avoid trailing /. in toolexeclibdir.
2693 * configure: Rebuilt.
2695 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
2697 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
2699 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2701 * libffi/src/powerpc/darwin_closure.S:
2702 Fix alignement bug, allocate 8 bytes for the result.
2703 * libffi/src/powerpc/aix_closure.S:
2705 * libffi/src/powerpc/ffi_darwin.c:
2706 Update stackframe description for aix/darwin_closure.S.
2708 2003-02-06 Jakub Jelinek <jakub@redhat.com>
2710 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
2713 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
2714 Andreas Schwab <schwab@suse.de>
2716 * configure.in: Adjust command to source config-ml.in to account
2717 for changes to the libffi_basedir definition.
2718 (libffi_basedir): Remove ${srcdir} from value and include trailing
2721 * configure: Regenerate.
2723 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2725 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
2727 2003-01-28 Andrew Haley <aph@redhat.com>
2729 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
2730 * src/x86/ffi64.c (ffi_prep_closure): New.
2731 (ffi_closure_UNIX64_inner): New.
2732 * src/x86/unix64.S (ffi_closure_UNIX64): New.
2734 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
2736 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
2737 Remove USE_LIBDIR conditional.
2738 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
2739 * Makefile.in, configure: Rebuilt.
2741 2003-01027 David Edelsohn <edelsohn@gnu.org>
2743 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
2744 * Makefile.in: Regenerate.
2746 2003-01-22 Andrew Haley <aph@redhat.com>
2748 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
2751 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
2753 * src/powerpc/darwin.S: Add unwind info.
2754 * src/powerpc/darwin_closure.S: Likewise.
2756 2003-01-14 Andrew Haley <aph@redhat.com>
2758 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
2759 (ffi_prep_cif_machdep): Likewise.
2760 * src/x86/unix64.S: Add unwind info.
2762 2003-01-14 Andreas Jaeger <aj@suse.de>
2764 * src/ffitest.c (main): Only use ffi_closures if those are
2767 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
2769 * libffi/src/ffitest.c
2770 add closure testcases
2772 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
2774 * libffi/src/powerpc/ffi.c
2775 fix alignment bug for float (4 byte aligned iso 8 byte)
2777 2003-01-09 Geoffrey Keating <geoffk@apple.com>
2779 * src/powerpc/ffi_darwin.c: Remove RCS version string.
2780 * src/powerpc/darwin.S: Remove RCS version string.
2782 2003-01-03 Jeff Sturm <jsturm@one-point.com>
2784 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
2785 * src/ffitest.c (main): Use static storage for closure.
2786 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
2787 * src/sparc/v8.S (ffi_closure_v8): New.
2788 * src/sparc/v9.S (ffi_closure_v9): New.
2790 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
2792 * include/ffi.h.in: Added FFI_STDCALL ffi_type
2793 enumeration for X86_WIN32.
2794 * src/x86/win32.S: Added ffi_call_STDCALL function
2796 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
2797 switch cases for recognising FFI_STDCALL and
2798 calling ffi_call_STDCALL if target is X86_WIN32.
2799 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
2800 stdcall versions of the "my_strlen" and "many"
2801 test functions (for X86_WIN32).
2802 Added test cases to test stdcall invocation using
2805 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
2807 * src/sh/sysv.S: Add DWARF2 unwind info.
2809 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
2811 * src/s390/sysv.S (.eh_frame section): Make section read-only.
2813 2002-11-26 Jim Wilson <wilson@redhat.com>
2815 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
2817 2002-11-23 H.J. Lu <hjl@gnu.org>
2819 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
2820 Include ../config/accross.m4.
2821 * aclocal.m4; Rebuild.
2822 * configure: Likewise.
2824 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
2826 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
2828 2002-11-11 DJ Delorie <dj@redhat.com>
2830 * configure.in: Look for common files in the right place.
2832 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
2834 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
2835 raw data as _Jv_word values, not ffi_raw.
2836 (ffi_java_ptrarray_to_raw): Likewise.
2837 (ffi_java_rvalue_to_raw): New function.
2838 (ffi_java_raw_call): Call it.
2839 (ffi_java_raw_to_rvalue): New function.
2840 (ffi_java_translate_args): Call it.
2841 * src/ffitest.c (closure_test_fn): Interpret return value
2842 as ffi_arg, not int.
2843 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
2844 FFI_TYPE_POINTER case.
2845 (ffi_closure_helper_SYSV): Likewise. Also, assume return
2846 values extended to word size.
2848 2002-10-02 Andreas Jaeger <aj@suse.de>
2850 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
2852 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
2854 * include/ffi.h.in: Fix i386 win32 compilation.
2856 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
2858 * configure.in: Add s390x-*-linux-* target.
2859 * configure: Regenerate.
2860 * include/ffi.h.in: Define S390X for s390x targets.
2861 (FFI_CLOSURES): Define for s390/s390x.
2862 (FFI_TRAMPOLINE_SIZE): Likewise.
2863 (FFI_NATIVE_RAW_API): Likewise.
2864 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
2865 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
2866 * src/s390/ffi.c: Major rework of existing code. Add support for
2867 s390x targets. Add closure support.
2868 * src/s390/sysv.S: Likewise.
2870 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
2872 * src/arm/sysv.S: Fix typo.
2874 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
2876 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
2877 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
2878 (ffi_call_SYSV): Handle soft-float.
2880 2002-09-27 Bo Thorsen <bo@suse.de>
2882 * include/ffi.h.in: Fix multilib x86-64 support.
2884 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2886 * Makefile.am (all-multi): Fix multilib parallel build.
2888 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
2890 * configure.in (sh[34]*-*-linux*): Add brackets.
2891 * configure: Regenerate.
2893 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
2895 * Makefile.am: Add SH support.
2896 * Makefile.in: Regenerate.
2897 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
2898 * configure: Regenerate.
2899 * include/ffi.h.in: Add SH support.
2900 * src/sh/ffi.c: New file.
2901 * src/sh/sysv.S: New file.
2902 * src/types.c: Add SH support.
2904 2002-07-16 Bo Thorsen <bo@suse.de>
2906 * src/x86/ffi64.c: New file that adds x86-64 support.
2907 * src/x86/unix64.S: New file that handles argument setup for
2909 * src/x86/sysv.S: Don't use this on x86-64.
2910 * src/x86/ffi.c: Don't use this on x86-64.
2912 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
2914 * src/ffitest.c (struct6): New test that tests a special case in
2916 (struct7): Likewise.
2917 (struct8): Likewise.
2918 (struct9): Likewise.
2919 (closure_test_fn): Silence warning about this when it's not used.
2920 (main): Add the new tests.
2921 (main): Fix a couple of wrong casts and silence some compiler warnings.
2922 * include/ffi.h.in: Add x86-64 ABI definition.
2923 * fficonfig.h.in: Regenerate.
2924 * Makefile.am: Add x86-64 support.
2925 * configure.in: Likewise.
2926 * Makefile.in: Regenerate.
2927 * configure: Likewise.
2929 2002-06-24 Bo Thorsen <bo@suse.de>
2931 * src/types.c: Merge settings for similar architectures.
2932 Add x86-64 sizes and alignments.
2934 2002-06-23 Bo Thorsen <bo@suse.de>
2936 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
2937 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
2938 * src/mips/ffi.c (ffi_prep_args): Likewise.
2939 * src/m68k/ffi.c (ffi_prep_args): Likewise.
2941 2002-07-18 H.J. Lu (hjl@gnu.org)
2943 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
2944 (libffi_la_SOURCES): Support MIPS_LINUX.
2945 (libffi_convenience_la_SOURCES): Likewise.
2946 * Makefile.in: Regenerated.
2948 * configure.in (mips64*-*): Skip.
2949 (mips*-*-linux*): New.
2950 * configure: Regenerated.
2952 * src/mips/ffi.c: Include <sgidefs.h>.
2954 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
2956 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
2958 2002-05-27 Roger Sayle <roger@eyesopen.com>
2960 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
2962 2002-05-27 Bo Thorsen <bo@suse.de>
2964 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
2967 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
2969 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
2970 beginning of function (for older apple cc).
2972 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
2974 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
2975 script entry, and set LD to it when configuring multilibs.
2976 * configure: Rebuilt.
2978 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
2980 * configure.in (sparc64-*-netbsd*): Add target.
2981 (sparc-*-netbsdelf*): Likewise.
2982 * configure: Regenerate.
2984 2002-04-28 David S. Miller <davem@redhat.com>
2986 * configure.in, configure: Fix SPARC test in previous change.
2988 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
2990 * Makefile.am: Add Linux for S/390 support.
2991 * Makefile.in: Regenerate.
2992 * configure.in: Add Linux for S/390 support.
2993 * configure: Regenerate.
2994 * include/ffi.h.in: Add Linux for S/390 support.
2995 * src/s390/ffi.c: New file from libffi CVS tree.
2996 * src/s390/sysv.S: New file from libffi CVS tree.
2998 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3000 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3002 * src/sparc/v8.S: Use it.
3003 * src/sparc/v9.S: Likewise.
3004 * fficonfig.h.in: Rebuilt.
3005 * configure: Rebuilt.
3007 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3009 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3011 * src/ia64/unix.S: Add unwind information. Fix comments.
3012 Save sp in a way that's compatible with unwind info.
3013 (ffi_call_unix): Correctly restore sp in all cases.
3014 * src/ia64/ffi.c: Add, fix comments.
3016 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3018 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3020 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3022 * configure.in (alpha*-*-netbsd*): Add target.
3023 * configure: Regenerate.
3025 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3027 * src/sparc/v8.S: Add unwind info.
3028 * src/sparc/v9.S: Likewise.
3030 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3032 * configure.in: Enable i*86-*-netbsdelf*.
3033 * configure: Rebuilt.
3035 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3038 * src/mips/n32.s: Delete
3039 * src/mips/o32.s: Delete
3041 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3043 * configure.in: Enable alpha*-*-freebsd*.
3044 * configure: Rebuilt.
3046 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3048 * Makefile.am: libfficonvenience -> libffi_convenience.
3049 * Makefile.in: Rebuilt.
3051 * Makefile.am: Define ffitest_OBJECTS.
3052 * Makefile.in: Rebuilt.
3054 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3055 David Edelsohn <edelsohn@gnu.org>
3057 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3058 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3059 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3060 * Makefile.in: Regenerate.
3061 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3062 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3063 (flush_icache, flush_range): New functions.
3064 (ffi_closure_helper_DARWIN): New function.
3065 * src/powerpc/aix_closure.S: New file.
3066 * src/powerpc/darwin_closure.S: New file.
3068 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3070 * include/ffi.h.in: Add typedef for ffi_arg.
3071 * src/ffitest.c (main): Declare rint with ffi_arg.
3073 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3075 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3076 number of GPRs for floating-point arguments.
3078 2002-01-31 Anthony Green <green@redhat.com>
3080 * configure: Rebuilt.
3081 * configure.in: Replace CHECK_SIZEOF and endian tests with
3082 cross-compiler friendly macros.
3083 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3086 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3088 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3089 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3091 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3093 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3094 (TARGET_SRC_POWERPC_AIX): New.
3095 (POWERPC_AIX): New stanza.
3096 * Makefile.in: Regenerate.
3097 * configure.in: Add AIX case.
3098 * configure: Regenerate.
3099 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3100 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3101 size. Fix "long double" support.
3102 (ffi_call): Add FFI_AIX case.
3103 * src/powerpc/aix.S: New.
3105 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3107 Implement Darwin PowerPC ABI.
3108 * configure.in: Handle powerpc-*-darwin*.
3109 * Makefile.am: Set source files for POWERPC_DARWIN.
3110 * configure: Rebuilt.
3111 * Makefile.in: Rebuilt.
3112 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3114 * src/powerpc/darwin.S: New file.
3115 * src/powerpc/ffi_darwin.c: New file.
3117 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3119 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3121 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3123 * src/x86/sysv.S: Avoid gas-only .balign directive.
3124 Use C style comments.
3126 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3128 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3129 Fixes PR bootstrap/3563.
3131 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3133 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3135 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3137 * configure.in: Recognize sparc*-sun-* host.
3138 * configure: Regenerate.
3140 2001-06-06 Andrew Haley <aph@redhat.com>
3142 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
3144 2001-06-03 Andrew Haley <aph@redhat.com>
3146 * src/alpha/osf.S: Add unwind info.
3147 * src/powerpc/sysv.S: Add unwind info.
3148 * src/powerpc/ppc_closure.S: Likewise.
3150 2000-05-31 Jeff Sturm <jsturm@one-point.com>
3152 * configure.in: Fix AC_ARG_ENABLE usage.
3153 * configure: Rebuilt.
3155 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3157 * configure.in: Remove warning about beta code.
3158 * configure: Rebuilt.
3160 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3162 * src/ia64/unix.S: Restore stack pointer when returning from
3164 * src/ia64/ffi.c: Fix typo in comment.
3166 2001-04-18 Jim Wilson <wilson@redhat.com>
3168 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3169 to eliminate RAW DV.
3171 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3173 * Makefile.am: Make a libtool convenience library.
3174 * Makefile.in: Rebuilt.
3176 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3178 * configure.in: Use different syntax for subdirectory creation.
3179 * configure: Rebuilt.
3181 2001-03-27 Jon Beniston <jon@beniston.com>
3183 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3184 * configure: Rebuilt.
3185 * Makefile.am: Added X86_WIN32 target support.
3186 * Makefile.in: Rebuilt.
3188 * include/ffi.h.in: Added X86_WIN32 target support.
3190 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3191 * src/types.c: Added X86_WIN32 target support.
3193 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3194 stuff removed and made to work with CygWin's gas.
3196 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3198 * configure.in: Make target subdirectory in build dir.
3199 * Makefile.am: Override suffix based rules to specify correct output
3201 * Makefile.in: Rebuilt.
3202 * configure: Rebuilt.
3204 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3206 * src/powerpc/ppc_closure.S: New file.
3207 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3208 involving long long and register pairs.
3209 (ffi_prep_closure): New function.
3210 (flush_icache): Likewise.
3211 (ffi_closure_helper_SYSV): Likewise.
3212 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3213 (FFI_TRAMPOLINE_SIZE): Likewise.
3214 (FFI_NATIVE_RAW_API): Likewise.
3215 * Makefile.in: Rebuilt.
3216 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3217 (TARGET_SRC_POWERPC): Likewise.
3219 2001-03-19 Tom Tromey <tromey@redhat.com>
3221 * Makefile.in: Rebuilt.
3222 * Makefile.am (ffitest_LDFLAGS): New macro.
3224 2001-03-02 Nick Clifton <nickc@redhat.com>
3226 * include/ffi.h.in: Remove RCS ident string.
3227 * include/ffi_mips.h: Remove RCS ident string.
3228 * src/debug.c: Remove RCS ident string.
3229 * src/ffitest.c: Remove RCS ident string.
3230 * src/prep_cif.c: Remove RCS ident string.
3231 * src/types.c: Remove RCS ident string.
3232 * src/alpha/ffi.c: Remove RCS ident string.
3233 * src/alpha/osf.S: Remove RCS ident string.
3234 * src/arm/ffi.c: Remove RCS ident string.
3235 * src/arm/sysv.S: Remove RCS ident string.
3236 * src/mips/ffi.c: Remove RCS ident string.
3237 * src/mips/n32.S: Remove RCS ident string.
3238 * src/mips/o32.S: Remove RCS ident string.
3239 * src/sparc/ffi.c: Remove RCS ident string.
3240 * src/sparc/v8.S: Remove RCS ident string.
3241 * src/sparc/v9.S: Remove RCS ident string.
3242 * src/x86/ffi.c: Remove RCS ident string.
3243 * src/x86/sysv.S: Remove RCS ident string.
3245 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3247 * include/ffi.h.in: Change sourceware.cygnus.com references to
3250 2000-12-09 Richard Henderson <rth@redhat.com>
3252 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3253 (ffi_closure_osf_inner): Index rather than increment avalue
3254 and arg_types. Give ffi_closure_osf the raw return value type.
3255 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3258 2000-12-07 Richard Henderson <rth@redhat.com>
3260 * src/raw_api.c (ffi_translate_args): Fix typo.
3261 (ffi_prep_closure): Likewise.
3263 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3264 FFI_TRAMPOLINE_SIZE.
3265 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3266 cif->bytes for new ffi_call_osf implementation.
3267 (ffi_prep_args): Absorb into ...
3268 (ffi_call): ... here. Do all stack allocation here and
3269 avoid a callback function.
3270 (ffi_prep_closure, ffi_closure_osf_inner): New.
3271 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3272 (ffi_closure_osf): New.
3274 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3276 * config.guess, config.sub, install-sh: Removed.
3277 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3278 * Makefile.in: Rebuilt.
3280 * acinclude.m4: Include libtool macros from the top level.
3281 * aclocal.m4, configure: Rebuilt.
3283 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3285 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3286 * configure: Rebuilt.
3288 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3290 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3291 memory correctly. Use conditional instructions, not branches where
3294 2000-05-04 Tom Tromey <tromey@cygnus.com>
3296 * configure: Rebuilt.
3297 * configure.in: Match `arm*-*-linux-*'.
3298 From Chris Dornan <cdornan@arm.com>.
3300 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3302 * Makefile.am (SUBDIRS): Define.
3303 (AM_MAKEFLAGS): Likewise.
3304 (Multilib support.): Add section.
3305 * Makefile.in: Rebuilt.
3306 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3307 New variables. Set for gcc using -print-multi-lib. Export them
3309 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3310 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3312 (extra_compiler_flags_value, extra_compiler_flags): Check these
3313 for extra compiler options which need to be passed down in
3316 2000-04-16 Anthony Green <green@redhat.com>
3318 * configure: Rebuilt.
3319 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3321 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3323 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3324 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3325 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3326 Replace all void * sizeofs with sizeof(int).
3327 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3328 different than DOUBLE.
3329 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3330 (ffi_prep_args_v9): New function.
3331 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3332 (ffi_V9_return_struct): New function.
3333 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3334 32bit code (not yet cross-arch calls).
3335 * src/sparc/v8.S: Add struct return delay nop.
3337 * src/sparc/v9.S: New file.
3338 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3339 is used on sparc64 only for structures larger than 32 bytes.
3340 Pass by reference for structures is done for structure arguments
3341 larger than 16 bytes.
3342 * src/ffitest.c (main): Use 64bit rint on sparc64.
3343 Run long long tests on sparc.
3344 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3346 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3348 * configure.in (sparc-*-linux*): New supported target.
3349 (sparc64-*-linux*): Likewise.
3350 * configure: Rebuilt.
3351 * Makefile.am: Add v9.S to SPARC files.
3352 * Makefile.in: Likewise.
3353 (LINK): Surround $(CCLD) into double quotes, so that multilib
3354 compiles work correctly.
3356 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3358 * configure: Rebuilt.
3359 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3360 proposed by Bryce McKinlay.
3362 2000-03-20 Tom Tromey <tromey@cygnus.com>
3364 * Makefile.in: Hand edit for java_raw_api.lo.
3366 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3368 * config.guess, config.sub: Update from the gcc tree.
3369 Fix for PR libgcj/168.
3371 2000-03-03 Tom Tromey <tromey@cygnus.com>
3373 * Makefile.in: Fixed ia64 by hand.
3375 * configure: Rebuilt.
3376 * configure.in (--enable-multilib): New option.
3377 (libffi_basedir): New subst.
3378 (AC_OUTPUT): Added multilib code.
3380 2000-03-02 Tom Tromey <tromey@cygnus.com>
3382 * Makefile.in: Rebuilt.
3383 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
3386 2000-02-25 Hans Boehm <boehm@acm.org>
3388 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
3390 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
3392 (ffi_prep_raw_closure): Use ffi_translate_args, not
3393 ffi_closure_translate.
3394 * src/java_raw_api.c: New file.
3395 * src/ffitest.c (closure_test_fn): New function.
3396 (main): Define `rint' as long long on IA64. Added new test when
3397 FFI_CLOSURES is defined.
3398 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
3399 (ffi_abi): Recognize IA64.
3400 (ffi_raw): Added `flt' field.
3401 Added "Java raw API" code.
3402 * configure.in: Recognize ia64.
3403 * Makefile.am (TARGET_SRC_IA64): New macro.
3404 (libffi_la_common_SOURCES): Added java_raw_api.c.
3405 (libffi_la_SOURCES): Define in IA64 case.
3407 2000-01-04 Tom Tromey <tromey@cygnus.com>
3409 * Makefile.in: Rebuilt with newer automake.
3411 1999-12-31 Tom Tromey <tromey@cygnus.com>
3413 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
3415 1999-09-01 Tom Tromey <tromey@cygnus.com>
3417 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
3419 * fficonfig.h.in: Rebuilt.
3420 * configure: Rebuilt.
3421 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
3422 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
3423 * acconfig.h: Don't #undef PACKAGE or VERSION.
3425 1999-08-09 Anthony Green <green@cygnus.com>
3427 * include/ffi.h.in: Try to work around messy header problem
3428 with PACKAGE and VERSION.
3430 * configure: Rebuilt.
3431 * configure.in: Change version to 2.00-beta.
3433 * fficonfig.h.in: Rebuilt.
3434 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
3436 * src/x86/ffi.c (ffi_raw_call): Rename.
3438 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
3440 * src/x86/ffi.c (ffi_closure_SYSV): New function.
3441 (ffi_prep_incoming_args_SYSV): Ditto.
3442 (ffi_prep_closure): Ditto.
3443 (ffi_closure_raw_SYSV): Ditto.
3444 (ffi_prep_raw_closure): More ditto.
3445 (ffi_call_raw): Final ditto.
3447 * include/ffi.h.in: Add definitions for closure and raw API.
3449 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
3452 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
3454 * src/raw_api.c: New file.
3456 * include/ffi.h.in (ffi_raw): New type.
3457 (UINT_ARG, SINT_ARG): New defines.
3458 (ffi_closure, ffi_raw_closure): New types.
3459 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
3461 * configure.in: Add check for endianness and sizeof void*.
3463 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
3464 instead of directly.
3466 * configure: Rebuilt.
3468 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
3470 * configure.in: Add x86 and powerpc BeOS configurations.
3471 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
3473 1999-05-09 Anthony Green <green@cygnus.com>
3475 * configure.in: Add warning about this being beta code.
3476 Remove src/Makefile.am from the picture.
3477 * configure: Rebuilt.
3479 * Makefile.am: Move logic from src/Makefile.am. Add changes
3480 to support libffi as a target library.
3481 * Makefile.in: Rebuilt.
3483 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
3484 Upgraded to new autoconf, automake, libtool.
3488 * LICENSE: Update copyright date.
3490 * src/Makefile.am, src/Makefile.in: Removed.
3492 1998-11-29 Anthony Green <green@cygnus.com>
3494 * include/ChangeLog: Removed.
3495 * src/ChangeLog: Removed.
3496 * src/mips/ChangeLog: Removed.
3497 * src/sparc/ChangeLog: Remboved.
3498 * src/x86/ChangeLog: Removed.
3500 * ChangeLog.v1: Created.