1 2010-01-06 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
3 * src/mips/n32.S: Use .abicalls and .eh_frame with __GNUC__.
5 2009-12-31 Anthony Green <green@redhat.com>
7 * README: Update for libffi 3.0.9.
9 2009-12-27 Matthias Klose <doko@ubuntu.com>
11 * configure.ac (HAVE_LONG_DOUBLE): Define for mips when
15 2009-12-26 Anthony Green <green@redhat.com>
17 * testsuite/libffi.call/cls_longdouble_va.c: Mark as xfail for
19 * testsuite/libffi.call/cls_double_va.c: Ditto.
21 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
23 * testsuite/libffi.call/ffitest.h: Conditionally include stdint.h
25 * testsuite/libffi.special/unwindtest.cc: Ditto.
27 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
29 * configure.ac: Add amd64-*-openbsd*.
31 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Link
32 openbsd programs with -lpthread.
34 2009-12-26 Anthony Green <green@redhat.com>
36 * testsuite/libffi.call/cls_double_va.c,
37 testsuite/libffi.call/cls_longdouble.c,
38 testsuite/libffi.call/cls_longdouble_va.c,
39 testsuite/libffi.call/cls_pointer.c,
40 testsuite/libffi.call/cls_pointer_stack.c: Remove xfail for
41 mips*-*-* and arm*-*-*.
42 * testsuite/libffi.call/cls_align_longdouble_split.c,
43 testsuite/libffi.call/cls_align_longdouble_split2.c,
44 testsuite/libffi.call/stret_medium2.c,
45 testsuite/libffi.call/stret_medium.c,
46 testsuite/libffi.call/stret_large.c,
47 testsuite/libffi.call/stret_large2.c: Remove xfail for arm*-*-*.
49 2009-12-31 Kay Tietz <ktietz70@googlemail.com>
51 * testsuite/libffi.call/ffitest.h,
52 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRuLL): Fix
55 2009-12-31 Carlo Bramini <carlo.bramix@libero.it>
57 * configure.ac (AM_LTLDFLAGS): Define for windows hosts.
58 * Makefile.am (libffi_la_LDFLAGS): Add AM_LTLDFLAGS.
60 * Makefile.in: Rebuilt.
62 2009-12-31 Anthony Green <green@redhat.com>
65 * testsuite/libffi.call/huge_struct.c: New test case from Blake
68 2009-12-28 David Edelsohn <edelsohn@gnu.org>
70 * src/powerpc/ffi_darwin.c (ffi_prep_args): Copy abi and nargs to
72 (aix_adjust_aggregate_sizes): New function.
73 (ffi_prep_cif_machdep): Call it.
75 2009-12-26 Andreas Tobler <a.tobler@schweiz.org>
77 * configure.ac: Define FFI_MMAP_EXEC_WRIT for the given targets.
78 * configure: Regenerate.
79 * fficonfig.h.in: Likewise.
80 * src/closures.c: Remove the FFI_MMAP_EXEC_WRIT definition for
83 2009-12-26 Andreas Schwab <schwab@linux-m68k.org>
85 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Advance intarg_count
86 when a float arguments is passed in memory.
87 (ffi_closure_helper_SYSV): Mark general registers as used up when
88 a 64bit or soft-float long double argument is passed in memory.
90 2009-12-25 Matthias Klose <doko@ubuntu.com>
92 * man/ffi_call.3: Fix #include in examples.
93 * doc/libffi.texi: Add dircategory.
95 2009-12-25 Frank Everdij <f.p.x.everdij@tudelft.nl>
97 * include/ffi.h.in: Placed '__GNUC__' ifdef around
98 '__attribute__((aligned(8)))' in ffi_closure, fixes compile for
100 * include/ffi_common.h: Added '__sgi' define to non
101 '__attribute__((__mode__()))' integer typedefs.
102 * src/mips/ffi.c (ffi_call, ffi_closure_mips_inner_O32,
103 ffi_closure_mips_inner_N32): Added 'defined(_MIPSEB)' to BE check.
104 (ffi_closure_mips_inner_O32, ffi_closure_mips_inner_N32): Added
105 FFI_LONGDOUBLE support and alignment(N32 only).
106 * src/mips/ffitarget.h: Corrected '#include <sgidefs.h>' for IRIX and
107 fixed non '__attribute__((__mode__()))' integer typedefs.
108 * src/mips/n32.S: Put '#ifdef linux' around '.abicalls' and '.eh_frame'
109 since they are Linux/GNU Assembler specific.
111 2009-12-25 Bradley Smith <brad@brad-smith.co.uk>
113 * configure.ac, Makefile.am, src/avr32/ffi.c,
114 src/avr32/ffitarget.h,
115 src/avr32/sysv.S: Add AVR32 port.
116 * configure, Makefile.in: Rebuilt.
118 2009-12-21 Andreas Tobler <a.tobler@schweiz.org>
120 * configure.ac: Make i?86 build on FreeBSD and OpenBSD.
121 * configure: Regenerate.
123 2009-12-15 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
125 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on PA HP-UX.
127 2009-12-13 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
129 * src/pa/ffi.c (ffi_closure_inner_pa32): Handle FFI_TYPE_LONGDOUBLE
132 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
134 * src/sparc/ffi.c (ffi_closure_sparc_inner_v9): Properly align 'long
137 2009-12-11 Eric Botcazou <ebotcazou@adacore.com>
139 * testsuite/libffi.call/ffitest.h: Define PRIuPTR on Solaris < 10.
141 2009-12-10 Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
144 * src/closures.c [X86_64 && __sun__ && __svr4__]
145 (FFI_MMAP_EXEC_WRIT): Define.
147 2009-12-08 David Daney <ddaney@caviumnetworks.com>
149 * testsuite/libffi.call/stret_medium.c: Remove xfail for mips*-*-*
150 * testsuite/libffi.call/cls_align_longdouble_split2.c: Same.
151 * testsuite/libffi.call/stret_large.c: Same.
152 * testsuite/libffi.call/cls_align_longdouble_split.c: Same.
153 * testsuite/libffi.call/stret_large2.c: Same.
154 * testsuite/libffi.call/stret_medium2.c: Same.
156 2009-12-07 David Edelsohn <edelsohn@gnu.org>
158 * src/powerpc/aix_closure.S (libffi_closure_ASM): Fix tablejump
161 2009-12-05 David Edelsohn <edelsohn@gnu.org>
163 * src/powerpc/aix.S: Update AIX32 code to be consistent with AIX64
165 * src/powerpc/aix_closure.S: Same.
167 2009-12-05 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
169 * Makefile.in: Regenerate.
170 * configure: Regenerate.
171 * include/Makefile.in: Regenerate.
172 * man/Makefile.in: Regenerate.
173 * testsuite/Makefile.in: Regenerate.
175 2009-12-04 David Edelsohn <edelsohn@gnu.org>
177 * src/powerpc/aix_closure.S: Reorganize 64-bit code to match
180 2009-12-04 Uros Bizjak <ubizjak@gmail.com>
183 * src/x86/ffi64.c (classify_argument): Update from
184 gcc/config/i386/i386.c.
185 (ffi_closure_unix64_inner): Do not use the address of two consecutive
186 SSE registers directly.
187 * testsuite/libffi.call/cls_dbls_struct.c (main): Remove xfail
188 for x86_64 linux targets.
190 2009-12-04 David Edelsohn <edelsohn@gnu.org>
192 * src/powerpc/ffi_darwin.c (ffi_closure_helper_DARWIN): Increment
193 pfr for long double split between fpr13 and stack.
195 2009-12-03 David Edelsohn <edelsohn@gnu.org>
197 * src/powerpc/ffi_darwin.c (ffi_prep_args): Increment next_arg and
198 fparg_count twice for long double.
200 2009-12-03 David Edelsohn <edelsohn@gnu.org>
203 * src/powerpc/ffi_darwin.c (ffi_prep_args): Remove extra parentheses.
205 2009-12-03 Uros Bizjak <ubizjak@gmail.com>
207 * testsuite/libffi.call/cls_longdouble_va.c (main): Fix format string.
208 Remove xfails for x86 linux targets.
210 2009-12-02 David Edelsohn <edelsohn@gnu.org>
212 * src/powerpc/ffi_darwin.c (ffi_prep_args): Fix typo in INT64
215 2009-12-01 David Edelsohn <edelsohn@gnu.org>
217 * src/powerpc/aix.S (ffi_call_AIX): Convert to more standard
218 register usage. Call ffi_prep_args directly. Add long double
219 return value support.
220 * src/powerpc/ffi_darwin.c (ffi_prep_args): Double arg increment
221 applies to FFI_TYPE_DOUBLE. Correct fpr_base increment typo.
222 Separate FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases.
223 (ffi_prep_cif_machdep): Only 16 byte stack alignment in 64 bit
225 (ffi_closure_helper_DARWIN): Remove nf and ng counters. Move temp
227 * src/powerpc/aix_closure.S: Maintain 16 byte stack alignment.
228 Allocate result area between params and FPRs.
230 2009-11-30 David Edelsohn <edelsohn@gnu.org>
233 * src/powerpc/ffitarget.h (POWERPC64): Define for PPC64 Linux and
235 * src/powerpc/aix.S: Implement AIX64 version.
236 * src/powerpc/aix_closure.S: Implement AIX64 version.
237 (ffi_closure_ASM): Use extsb, lha and displament addresses.
238 * src/powerpc/ffi_darwin.c (ffi_prep_args): Implement AIX64
240 (ffi_prep_cif_machdep): Same.
242 (ffi_closure_helper_DARWIN): Same.
244 2009-11-02 Andreas Tobler <a.tobler@schweiz.org>
247 * testsuite/libffi.call/testclosure.c: New test.
249 2009-09-28 Kai Tietz <kai.tietz@onevision.com>
251 * src/x86/win64.S (_ffi_call_win64 stack): Remove for gnu
252 assembly version use of ___chkstk.
254 2009-09-23 Matthias Klose <doko@ubuntu.com>
256 PR libffi/40242, PR libffi/41443
257 * src/arm/sysv.S (__ARM_ARCH__): Define for processors
258 __ARM_ARCH_6T2__, __ARM_ARCH_6M__, __ARM_ARCH_7__,
259 __ARM_ARCH_7A__, __ARM_ARCH_7R__, __ARM_ARCH_7M__.
260 Change the conditionals to __SOFTFP__ || __ARM_EABI__
261 for -mfloat-abi=softfp to work.
263 2009-09-17 Loren J. Rittle <ljrittle@acm.org>
265 PR testsuite/32843 (strikes again)
266 * src/x86/ffi.c (ffi_prep_cif_machdep): Add X86_FREEBSD to
267 enable proper extension on char and short.
269 2009-09-15 David Daney <ddaney@caviumnetworks.com>
271 * src/java_raw_api.c (ffi_java_raw_to_rvalue): Remove special
272 handling for FFI_TYPE_POINTER.
273 * src/mips/ffitarget.h (FFI_TYPE_STRUCT_D_SOFT,
274 FFI_TYPE_STRUCT_F_SOFT, FFI_TYPE_STRUCT_DD_SOFT,
275 FFI_TYPE_STRUCT_FF_SOFT, FFI_TYPE_STRUCT_FD_SOFT,
276 FFI_TYPE_STRUCT_DF_SOFT, FFI_TYPE_STRUCT_SOFT): New defines.
277 (FFI_N32_SOFT_FLOAT, FFI_N64_SOFT_FLOAT): New ffi_abi enumerations.
278 (enum ffi_abi): Set FFI_DEFAULT_ABI for soft-float.
279 * src/mips/n32.S (ffi_call_N32): Add handling for soft-float
280 structure and pointer returns.
281 (ffi_closure_N32): Add handling for pointer returns.
282 * src/mips/ffi.c (ffi_prep_args, calc_n32_struct_flags,
283 calc_n32_return_struct_flags): Handle soft-float.
284 (ffi_prep_cif_machdep): Handle soft-float, fix pointer handling.
285 (ffi_call_N32): Declare proper argument types.
286 (ffi_call, copy_struct_N32, ffi_closure_mips_inner_N32): Handle
289 2009-08-24 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
291 * configure.ac (AC_PREREQ): Bump to 2.64.
293 2009-08-22 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
295 * Makefile.am (install-html, install-pdf): Remove.
296 * Makefile.in: Regenerate.
298 * Makefile.in: Regenerate.
299 * aclocal.m4: Regenerate.
300 * configure: Regenerate.
301 * fficonfig.h.in: Regenerate.
302 * include/Makefile.in: Regenerate.
303 * man/Makefile.in: Regenerate.
304 * testsuite/Makefile.in: Regenerate.
306 2009-07-30 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
308 * configure.ac (_AC_ARG_VAR_PRECIOUS): Use m4_rename_force.
310 2009-07-24 Dave Korn <dave.korn.cygwin@gmail.com>
313 * src/x86/ffi.c (ffi_prep_cif_machdep): Also use sign/zero-extending
314 return types for X86_WIN32.
315 * src/x86/win32.S (_ffi_call_SYSV): Handle omitted return types.
316 (_ffi_call_STDCALL, _ffi_closure_SYSV, _ffi_closure_raw_SYSV,
317 _ffi_closure_STDCALL): Likewise.
319 * src/closures.c (is_selinux_enabled): Define to const 0 for Cygwin.
320 (dlmmap, dlmunmap): Also use these functions on Cygwin.
322 2009-07-11 Richard Sandiford <rdsandiford@googlemail.com>
327 * testsuite/lib/libffi-dg.exp: Revert 2009-07-02, 2009-07-01 and
330 2009-07-01 Richard Sandiford <r.sandiford@uk.ibm.com>
332 * testsuite/lib/libffi-dg.exp (libffi-init): Set ld_library_path
333 to "" before adding paths. (This reinstates an assignment that
334 was removed by my 2009-06-30 commit, but changes the initial
335 value from "." to "".)
337 2009-07-01 H.J. Lu <hongjiu.lu@intel.com>
340 * testsuite/lib/libffi-dg.exp (libffi-init): Properly set
341 gccdir. Adjust ld_library_path for gcc only if gccdir isn't
344 2009-06-30 Richard Sandiford <r.sandiford@uk.ibm.com>
346 * testsuite/lib/libffi-dg.exp (libffi-init): Don't add "."
347 to ld_library_path. Use add_path. Add just find_libgcc_s
348 to ld_library_path, not every libgcc multilib directory.
350 2009-06-16 Wim Lewis <wiml@hhhh.org>
352 * src/powerpc/ffi.c: Avoid clobbering cr3 and cr4, which are
353 supposed to be callee-saved.
354 * src/powerpc/sysv.S (small_struct_return_value): Fix overrun of
355 return buffer for odd-size structs.
357 2009-06-16 Andreas Tobler <a.tobler@schweiz.org>
360 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Add
361 allow_stack_execute for Darwin.
363 2009-06-16 Andrew Haley <aph@redhat.com>
365 * configure.ac (TARGETDIR): Add missing blank lines.
366 * configure: Regenerate.
368 2009-06-16 Andrew Haley <aph@redhat.com>
370 * testsuite/libffi.call/cls_align_sint64.c,
371 testsuite/libffi.call/cls_align_uint64.c,
372 testsuite/libffi.call/cls_longdouble_va.c,
373 testsuite/libffi.call/cls_ulonglong.c,
374 testsuite/libffi.call/return_ll1.c,
375 testsuite/libffi.call/stret_medium2.c: Fix printf format
377 * testsuite/libffi.call/ffitest.h,
378 testsuite/libffi.special/ffitestcxx.h (PRIdLL, PRIuLL): Define.
380 2009-06-15 Andrew Haley <aph@redhat.com>
382 * testsuite/libffi.call/err_bad_typedef.c: xfail everywhere.
383 * testsuite/libffi.call/err_bad_abi.c: Likewise.
385 2009-06-12 Andrew Haley <aph@redhat.com>
387 * Makefile.am: Remove info_TEXINFOS.
389 2009-06-12 Andrew Haley <aph@redhat.com>
391 * ChangeLog.libffi: testsuite/libffi.call/cls_align_sint64.c,
392 testsuite/libffi.call/cls_align_uint64.c,
393 testsuite/libffi.call/cls_ulonglong.c,
394 testsuite/libffi.call/return_ll1.c,
395 testsuite/libffi.call/stret_medium2.c: Fix printf format
397 testsuite/libffi.special/unwindtest.cc: include stdint.h.
399 2009-06-11 Timothy Wall <twall@users.sf.net>
404 include/ffi_common.h,
409 src/x86/win64.S (new),
410 README: Added win64 support (mingw or MSVC)
414 testsuite/Makefile.in,
416 aclocal.m4: Regenerated
417 * ltcf-c.sh: properly escape cygwin/w32 path
418 * man/ffi_call.3: Clarify size requirements for return value.
419 * src/x86/ffi64.c: Fix filename in comment.
420 * src/x86/win32.S: Remove unused extern.
422 * testsuite/libffi.call/closure_fn0.c,
423 testsuite/libffi.call/closure_fn1.c,
424 testsuite/libffi.call/closure_fn2.c,
425 testsuite/libffi.call/closure_fn3.c,
426 testsuite/libffi.call/closure_fn4.c,
427 testsuite/libffi.call/closure_fn5.c,
428 testsuite/libffi.call/closure_fn6.c,
429 testsuite/libffi.call/closure_stdcall.c,
430 testsuite/libffi.call/cls_12byte.c,
431 testsuite/libffi.call/cls_16byte.c,
432 testsuite/libffi.call/cls_18byte.c,
433 testsuite/libffi.call/cls_19byte.c,
434 testsuite/libffi.call/cls_1_1byte.c,
435 testsuite/libffi.call/cls_20byte.c,
436 testsuite/libffi.call/cls_20byte1.c,
437 testsuite/libffi.call/cls_24byte.c,
438 testsuite/libffi.call/cls_2byte.c,
439 testsuite/libffi.call/cls_3_1byte.c,
440 testsuite/libffi.call/cls_3byte1.c,
441 testsuite/libffi.call/cls_3byte2.c,
442 testsuite/libffi.call/cls_4_1byte.c,
443 testsuite/libffi.call/cls_4byte.c,
444 testsuite/libffi.call/cls_5_1_byte.c,
445 testsuite/libffi.call/cls_5byte.c,
446 testsuite/libffi.call/cls_64byte.c,
447 testsuite/libffi.call/cls_6_1_byte.c,
448 testsuite/libffi.call/cls_6byte.c,
449 testsuite/libffi.call/cls_7_1_byte.c,
450 testsuite/libffi.call/cls_7byte.c,
451 testsuite/libffi.call/cls_8byte.c,
452 testsuite/libffi.call/cls_9byte1.c,
453 testsuite/libffi.call/cls_9byte2.c,
454 testsuite/libffi.call/cls_align_double.c,
455 testsuite/libffi.call/cls_align_float.c,
456 testsuite/libffi.call/cls_align_longdouble.c,
457 testsuite/libffi.call/cls_align_longdouble_split.c,
458 testsuite/libffi.call/cls_align_longdouble_split2.c,
459 testsuite/libffi.call/cls_align_pointer.c,
460 testsuite/libffi.call/cls_align_sint16.c,
461 testsuite/libffi.call/cls_align_sint32.c,
462 testsuite/libffi.call/cls_align_sint64.c,
463 testsuite/libffi.call/cls_align_uint16.c,
464 testsuite/libffi.call/cls_align_uint32.c,
465 testsuite/libffi.call/cls_align_uint64.c,
466 testsuite/libffi.call/cls_dbls_struct.c,
467 testsuite/libffi.call/cls_double.c,
468 testsuite/libffi.call/cls_double_va.c,
469 testsuite/libffi.call/cls_float.c,
470 testsuite/libffi.call/cls_longdouble.c,
471 testsuite/libffi.call/cls_longdouble_va.c,
472 testsuite/libffi.call/cls_multi_schar.c,
473 testsuite/libffi.call/cls_multi_sshort.c,
474 testsuite/libffi.call/cls_multi_sshortchar.c,
475 testsuite/libffi.call/cls_multi_uchar.c,
476 testsuite/libffi.call/cls_multi_ushort.c,
477 testsuite/libffi.call/cls_multi_ushortchar.c,
478 testsuite/libffi.call/cls_pointer.c,
479 testsuite/libffi.call/cls_pointer_stack.c,
480 testsuite/libffi.call/cls_schar.c,
481 testsuite/libffi.call/cls_sint.c,
482 testsuite/libffi.call/cls_sshort.c,
483 testsuite/libffi.call/cls_uchar.c,
484 testsuite/libffi.call/cls_uint.c,
485 testsuite/libffi.call/cls_ulonglong.c,
486 testsuite/libffi.call/cls_ushort.c,
487 testsuite/libffi.call/err_bad_abi.c,
488 testsuite/libffi.call/err_bad_typedef.c,
489 testsuite/libffi.call/float2.c,
490 testsuite/libffi.call/huge_struct.c,
491 testsuite/libffi.call/nested_struct.c,
492 testsuite/libffi.call/nested_struct1.c,
493 testsuite/libffi.call/nested_struct10.c,
494 testsuite/libffi.call/nested_struct2.c,
495 testsuite/libffi.call/nested_struct3.c,
496 testsuite/libffi.call/nested_struct4.c,
497 testsuite/libffi.call/nested_struct5.c,
498 testsuite/libffi.call/nested_struct6.c,
499 testsuite/libffi.call/nested_struct7.c,
500 testsuite/libffi.call/nested_struct8.c,
501 testsuite/libffi.call/nested_struct9.c,
502 testsuite/libffi.call/problem1.c,
503 testsuite/libffi.call/return_ldl.c,
504 testsuite/libffi.call/return_ll1.c,
505 testsuite/libffi.call/stret_large.c,
506 testsuite/libffi.call/stret_large2.c,
507 testsuite/libffi.call/stret_medium.c,
508 testsuite/libffi.call/stret_medium2.c,
509 testsuite/libffi.special/unwindtest.cc: use ffi_closure_alloc instead
510 of checking for MMAP. Use intptr_t instead of long casts.
512 2009-06-11 Kaz Kojima <kkojima@gcc.gnu.org>
514 * testsuite/libffi.call/cls_longdouble_va.c: Add xfail sh*-*-linux-*.
515 * testsuite/libffi.call/err_bad_abi.c: Add xfail sh*-*-*.
516 * testsuite/libffi.call/err_bad_typedef.c: Likewise.
518 2009-06-09 Andrew Haley <aph@redhat.com>
520 * src/x86/freebsd.S: Add missing file.
522 2009-06-08 Andrew Haley <aph@redhat.com>
524 Import from libffi 3.0.8:
526 * doc/libffi.texi: New file.
527 * doc/libffi.info: Likewise.
528 * doc/stamp-vti: Likewise.
529 * man/Makefile.am: New file.
530 * man/ffi_call.3: New file.
532 * Makefile.am (EXTRA_DIST): Add src/x86/darwin64.S,
534 (nodist_libffi_la_SOURCES): Add X86_FREEBSD.
536 * configure.ac: Bump version to 3.0.8.
537 parisc*-*-linux*: Add.
538 i386-*-freebsd* | i386-*-openbsd*: Add.
539 powerpc-*-beos*: Add.
540 AM_CONDITIONAL X86_FREEBSD: Add.
541 AC_CONFIG_FILES: Add man/Makefile.
543 * include/ffi.h.in (FFI_FN): Change void (*)() to void (*)(void).
545 2009-06-08 Andrew Haley <aph@redhat.com>
547 * README: Import from libffi 3.0.8.
549 2009-06-08 Andrew Haley <aph@redhat.com>
551 * testsuite/libffi.call/err_bad_abi.c: Add xfails.
552 * testsuite/libffi.call/cls_longdouble_va.c: Add xfails.
553 * testsuite/libffi.call/cls_dbls_struct.c: Add xfail x86_64-*-linux-*.
554 * testsuite/libffi.call/err_bad_typedef.c: Add xfails.
556 * testsuite/libffi.call/stret_medium2.c: Add __UNUSED__ to args.
557 * testsuite/libffi.call/stret_medium.c: Likewise.
558 * testsuite/libffi.call/stret_large2.c: Likewise.
559 * testsuite/libffi.call/stret_large.c: Likewise.
561 2008-12-26 Timothy Wall <twall@users.sf.net>
563 * testsuite/libffi.call/cls_longdouble.c,
564 testsuite/libffi.call/cls_longdouble_va.c,
565 testsuite/libffi.call/cls_align_longdouble.c,
566 testsuite/libffi.call/cls_align_longdouble_split.c,
567 testsuite/libffi.call/cls_align_longdouble_split2.c: mark expected
568 failures on x86_64 cygwin/mingw.
570 2008-12-22 Timothy Wall <twall@users.sf.net>
572 * testsuite/libffi.call/closure_fn0.c,
573 testsuite/libffi.call/closure_fn1.c,
574 testsuite/libffi.call/closure_fn2.c,
575 testsuite/libffi.call/closure_fn3.c,
576 testsuite/libffi.call/closure_fn4.c,
577 testsuite/libffi.call/closure_fn5.c,
578 testsuite/libffi.call/closure_fn6.c,
579 testsuite/libffi.call/closure_loc_fn0.c,
580 testsuite/libffi.call/closure_stdcall.c,
581 testsuite/libffi.call/cls_align_pointer.c,
582 testsuite/libffi.call/cls_pointer.c,
583 testsuite/libffi.call/cls_pointer_stack.c: use portable cast from
584 pointer to integer (intptr_t).
585 * testsuite/libffi.call/cls_longdouble.c: disable for win64.
587 2008-07-24 Anthony Green <green@redhat.com>
589 * testsuite/libffi.call/cls_dbls_struct.c,
590 testsuite/libffi.call/cls_double_va.c,
591 testsuite/libffi.call/cls_longdouble.c,
592 testsuite/libffi.call/cls_longdouble_va.c,
593 testsuite/libffi.call/cls_pointer.c,
594 testsuite/libffi.call/cls_pointer_stack.c,
595 testsuite/libffi.call/err_bad_abi.c: Clean up failures from
598 2008-03-04 Anthony Green <green@redhat.com>
602 * testsuite/libffi.call/cls_align_longdouble_split2.c
603 testsuite/libffi.call/cls_align_longdouble_split.c
604 testsuite/libffi.call/cls_dbls_struct.c
605 testsuite/libffi.call/cls_double_va.c
606 testsuite/libffi.call/cls_longdouble.c
607 testsuite/libffi.call/cls_longdouble_va.c
608 testsuite/libffi.call/cls_pointer.c
609 testsuite/libffi.call/cls_pointer_stack.c
610 testsuite/libffi.call/err_bad_abi.c
611 testsuite/libffi.call/err_bad_typedef.c
612 testsuite/libffi.call/stret_large2.c
613 testsuite/libffi.call/stret_large.c
614 testsuite/libffi.call/stret_medium2.c
615 testsuite/libffi.call/stret_medium.c: New tests from Apple.
617 2009-06-05 Andrew Haley <aph@redhat.com>
619 * src/x86/ffitarget.h, src/x86/ffi.c: Merge stdcall changes from
622 2009-06-04 Andrew Haley <aph@redhat.com>
624 * src/x86/ffitarget.h, src/x86/win32.S, src/x86/ffi.c: Back out
627 2008-02-26 Anthony Green <green@redhat.com>
628 Thomas Heller <theller@ctypes.org>
630 * src/x86/ffi.c (ffi_closure_SYSV_inner): Change C++ comment to C
633 2008-02-03 Timothy Wall <twall@users.sf.net>
635 * src/x86/ffi.c (FFI_INIT_TRAMPOLINE_STDCALL): Calculate jump return
636 offset based on code pointer, not data pointer.
638 2008-01-31 Timothy Wall <twall@users.sf.net>
640 * testsuite/libffi.call/closure_stdcall.c: Add test for stdcall
642 * src/x86/ffitarget.h: Increase size of trampoline for stdcall
644 * src/x86/win32.S: Add assembly for stdcall closure.
645 * src/x86/ffi.c: Initialize stdcall closure trampoline.
647 2009-06-04 Andrew Haley <aph@redhat.com>
649 * include/ffi.h.in: Change void (*)() to void (*)(void).
650 * src/x86/ffi.c: Likewise.
652 2009-06-04 Andrew Haley <aph@redhat.com>
654 * src/powerpc/ppc_closure.S: Insert licence header.
655 * src/powerpc/linux64_closure.S: Likewise.
656 * src/m68k/sysv.S: Likewise.
658 * src/sh64/ffi.c: Change void (*)() to void (*)(void).
659 * src/powerpc/ffi.c: Likewise.
660 * src/powerpc/ffi_darwin.c: Likewise.
661 * src/m32r/ffi.c: Likewise.
662 * src/sh64/ffi.c: Likewise.
663 * src/x86/ffi64.c: Likewise.
664 * src/alpha/ffi.c: Likewise.
665 * src/alpha/osf.S: Likewise.
666 * src/frv/ffi.c: Likewise.
667 * src/s390/ffi.c: Likewise.
668 * src/pa/ffi.c: Likewise.
669 * src/pa/hpux32.S: Likewise.
670 * src/ia64/unix.S: Likewise.
671 * src/ia64/ffi.c: Likewise.
672 * src/sparc/ffi.c: Likewise.
673 * src/mips/ffi.c: Likewise.
674 * src/sh/ffi.c: Likewise.
676 2008-02-15 David Daney <ddaney@avtrex.com>
678 * src/mips/ffi.c (USE__BUILTIN___CLEAR_CACHE):
679 Define (conditionally), and use it to include cachectl.h.
680 (ffi_prep_closure_loc): Fix cache flushing.
681 * src/mips/ffitarget.h (_ABIN32, _ABI64, _ABIO32): Define.
683 2009-06-04 Andrew Haley <aph@redhat.com>
689 src/powerpc/ffitarget.h,
691 src/sh64/ffitarget.h,
702 src/alpha/ffitarget.h,
705 src/m68k/ffitarget.h,
708 src/s390/ffitarget.h,
710 src/cris/ffitarget.h,
715 src/ia64/ffitarget.h,
718 src/ia64/ia64_flags.h,
722 src/sparc/ffitarget.h,
725 src/mips/ffitarget.h,
732 src/sh/sysv.S: Update license text.
734 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
736 * src/x86/win32.S (_ffi_closure_STDCALL): New function.
737 (.eh_frame): Add FDE for it.
739 2009-05-22 Dave Korn <dave.korn.cygwin@gmail.com>
741 * configure.ac: Also check if assembler supports pc-relative
742 relocs on X86_WIN32 targets.
743 * configure: Regenerate.
744 * src/x86/win32.S (ffi_prep_args): Declare extern, not global.
745 (_ffi_call_SYSV): Add missing function type symbol .def and
746 add EH markup labels.
747 (_ffi_call_STDCALL): Likewise.
748 (_ffi_closure_SYSV): Likewise.
749 (_ffi_closure_raw_SYSV): Likewise.
750 (.eh_frame): Add hand-crafted EH data.
752 2009-04-09 Jakub Jelinek <jakub@redhat.com>
754 * testsuite/lib/libffi-dg.exp: Change copyright header to refer to
755 version 3 of the GNU General Public License and to point readers
756 at the COPYING3 file and the FSF's license web page.
757 * testsuite/libffi.call/call.exp: Likewise.
758 * testsuite/libffi.special/special.exp: Likewise.
760 2009-03-01 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
762 * configure: Regenerate.
764 2008-12-18 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
767 * configure.ac (HAVE_AS_X86_PCREL): New test.
768 * configure: Regenerate.
769 * fficonfig.h.in: Regenerate.
770 * src/x86/sysv.S [!FFI_NO_RAW_API]: Precalculate
771 RAW_CLOSURE_CIF_OFFSET, RAW_CLOSURE_FUN_OFFSET,
772 RAW_CLOSURE_USER_DATA_OFFSET for the Solaris 10/x86 assembler.
773 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
774 * src/x86/unix64.S (.Lstore_table): Move to .text section.
775 (.Lload_table): Likewise.
776 (.eh_frame): Only use SYMBOL-. iff HAVE_AS_X86_PCREL.
778 2008-12-18 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
780 * configure: Regenerate.
782 2008-11-21 Eric Botcazou <ebotcazou@adacore.com>
784 * src/sparc/ffi.c (ffi_prep_cif_machdep): Add support for
785 signed/unsigned int8/16 return values.
786 * src/sparc/v8.S (ffi_call_v8): Likewise.
787 (ffi_closure_v8): Likewise.
789 2008-09-26 Peter O'Gorman <pogma@thewrittenword.com>
790 Steve Ellcey <sje@cup.hp.com>
792 * configure: Regenerate for new libtool.
793 * Makefile.in: Ditto.
794 * include/Makefile.in: Ditto.
797 2008-08-25 Andreas Tobler <a.tobler@schweiz.org>
799 * src/powerpc/ffitarget.h (ffi_abi): Add FFI_LINUX and
800 FFI_LINUX_SOFT_FLOAT to the POWERPC_FREEBSD enum.
801 Add note about flag bits used for FFI_SYSV_TYPE_SMALL_STRUCT.
802 Adjust copyright notice.
803 * src/powerpc/ffi.c: Add two new flags to indicate if we have one
804 register or two register to use for FFI_SYSV structs.
805 (ffi_prep_cif_machdep): Pass the right register flag introduced above.
806 (ffi_closure_helper_SYSV): Fix the return type for
807 FFI_SYSV_TYPE_SMALL_STRUCT. Comment.
808 Adjust copyright notice.
810 2008-07-16 Kaz Kojima <kkojima@gcc.gnu.org>
812 * src/sh/ffi.c (ffi_prep_closure_loc): Turn INSN into an unsigned
815 2008-06-17 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
817 * configure: Regenerate.
818 * include/Makefile.in: Regenerate.
819 * testsuite/Makefile.in: Regenerate.
821 2008-06-07 Joseph Myers <joseph@codesourcery.com>
823 * configure.ac (parisc*-*-linux*, powerpc-*-sysv*,
824 powerpc-*-beos*): Remove.
825 * configure: Regenerate.
827 2008-05-09 Julian Brown <julian@codesourcery.com>
829 * Makefile.am (LTLDFLAGS): New.
830 (libffi_la_LDFLAGS): Use above.
831 * Makefile.in: Regenerate.
833 2008-04-18 Paolo Bonzini <bonzini@gnu.org>
836 * aclocal.m4: Regenerate.
837 * configure: Regenerate.
839 2008-03-26 Kaz Kojima <kkojima@gcc.gnu.org>
841 * src/sh/sysv.S: Add .note.GNU-stack on Linux.
842 * src/sh64/sysv.S: Likewise.
844 2008-03-26 Daniel Jacobowitz <dan@debian.org>
846 * src/arm/sysv.S: Fix ARM comment marker.
848 2008-03-26 Jakub Jelinek <jakub@redhat.com>
850 * src/alpha/osf.S: Add .note.GNU-stack on Linux.
851 * src/s390/sysv.S: Likewise.
852 * src/powerpc/ppc_closure.S: Likewise.
853 * src/powerpc/sysv.S: Likewise.
854 * src/x86/unix64.S: Likewise.
855 * src/x86/sysv.S: Likewise.
856 * src/sparc/v8.S: Likewise.
857 * src/sparc/v9.S: Likewise.
858 * src/m68k/sysv.S: Likewise.
859 * src/arm/sysv.S: Likewise.
861 2008-03-16 Ralf Wildenhues <Ralf.Wildenhues@gmx.de>
863 * aclocal.m4: Regenerate.
864 * configure: Likewise.
865 * Makefile.in: Likewise.
866 * include/Makefile.in: Likewise.
867 * testsuite/Makefile.in: Likewise.
869 2008-02-12 Bjoern Koenig <bkoenig@alpha-tierchen.de>
870 Andreas Tobler <a.tobler@schweiz.org>
872 * configure.ac: Add amd64-*-freebsd* target.
873 * configure: Regenerate.
875 2008-01-30 H.J. Lu <hongjiu.lu@intel.com>
878 * src/x86/sysv.S (ffi_closure_SYSV): Pop 4 byte from stack when
881 * testsuite/libffi.call/call.exp: Add "-O2 -fomit-frame-pointer"
884 2008-01-24 David Edelsohn <edelsohn@gnu.org>
886 * configure: Regenerate.
888 2008-01-06 Andreas Tobler <a.tobler@schweiz.org>
890 * src/x86/ffi.c (ffi_prep_cif_machdep): Fix thinko.
892 2008-01-05 Andreas Tobler <a.tobler@schweiz.org>
895 * src/x86/ffi.c (ffi_prep_cif_machdep): Add code for
896 signed/unsigned int8/16 for X86_DARWIN.
897 Updated copyright info.
898 Handle one and two byte structs with special cif->flags.
899 * src/x86/ffitarget.h: Add special types for one and two byte structs.
900 Updated copyright info.
901 * src/x86/darwin.S (ffi_call_SYSV): Rewrite to use a jump table like
903 Remove code to pop args from the stack after call.
904 Special-case signed/unsigned for int8/16, one and two byte structs.
905 (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
906 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
908 Updated copyright info.
910 2007-12-08 David Daney <ddaney@avtrex.com>
912 * src/mips/n32.S (ffi_call_N32): Replace dadd with ADDU, dsub with
913 SUBU, add with ADDU and use smaller code sequences.
915 2007-12-07 David Daney <ddaney@avtrex.com>
917 * src/mips/ffi.c (ffi_prep_cif_machdep): Handle long double return
920 2007-12-06 David Daney <ddaney@avtrex.com>
922 * include/ffi.h.in (FFI_SIZEOF_JAVA_RAW): Define if not already
924 (ffi_java_raw): New typedef.
925 (ffi_java_raw_call, ffi_java_ptrarray_to_raw,
926 ffi_java_raw_to_ptrarray): Change parameter types from ffi_raw to
928 (ffi_java_raw_closure) : Same.
929 (ffi_prep_java_raw_closure, ffi_prep_java_raw_closure_loc): Change
931 * src/java_raw_api.c (ffi_java_raw_size): Replace FFI_SIZEOF_ARG with
933 (ffi_java_raw_to_ptrarray): Change type of raw to ffi_java_raw.
934 Replace FFI_SIZEOF_ARG with FFI_SIZEOF_JAVA_RAW. Use
935 sizeof(ffi_java_raw) for alignment calculations.
936 (ffi_java_ptrarray_to_raw): Same.
937 (ffi_java_rvalue_to_raw): Add special handling for FFI_TYPE_POINTER
938 if FFI_SIZEOF_JAVA_RAW == 4.
939 (ffi_java_raw_to_rvalue): Same.
940 (ffi_java_raw_call): Change type of raw to ffi_java_raw.
941 (ffi_java_translate_args): Same.
942 (ffi_prep_java_raw_closure_loc, ffi_prep_java_raw_closure): Change
944 * src/mips/ffitarget.h (FFI_SIZEOF_JAVA_RAW): Define for N32 ABI.
946 2007-12-06 David Daney <ddaney@avtrex.com>
948 * src/mips/n32.S (ffi_closure_N32): Use 64-bit add instruction on
951 2007-12-01 Andreas Tobler <a.tobler@schweiz.org>
954 * src/powerpc/ffitarget.h: Introduce new ABI FFI_LINUX_SOFT_FLOAT.
955 Add local FFI_TYPE_UINT128 to handle soft-float long-double-128.
956 * src/powerpc/ffi.c: Distinguish between __NO_FPRS__ and not and
957 set the NUM_FPR_ARG_REGISTERS according to.
958 Add support for potential soft-float support under hard-float
960 (ffi_prep_args_SYSV): Set NUM_FPR_ARG_REGISTERS to 0 in case of
961 FFI_LINUX_SOFT_FLOAT, handle float, doubles and long-doubles according
962 to the FFI_LINUX_SOFT_FLOAT ABI.
963 (ffi_prep_cif_machdep): Likewise.
964 (ffi_closure_helper_SYSV): Likewise.
965 * src/powerpc/ppc_closure.S: Make sure not to store float/double
966 on archs where __NO_FPRS__ is true.
967 Add FFI_TYPE_UINT128 support.
968 * src/powerpc/sysv.S: Add support for soft-float long-double-128.
969 Adjust copyright notice.
971 2007-11-25 Andreas Tobler <a.tobler@schweiz.org>
973 * src/closures.c: Move defintion of MAYBE_UNUSED from here to ...
974 * include/ffi_common.h: ... here.
977 2007-11-17 Andreas Tobler <a.tobler@schweiz.org>
979 * src/powerpc/sysv.S: Load correct cr to compare if we have long double.
980 * src/powerpc/linux64.S: Likewise.
981 * src/powerpc/ffi.c: Add a comment to show which part goes into cr6.
982 * testsuite/libffi.call/return_ldl.c: New test.
984 2007-09-04 <aph@redhat.com>
986 * src/arm/sysv.S (UNWIND): New.
987 (Whole file): Conditionally compile unwinder directives.
988 * src/arm/sysv.S: Add unwinder directives.
990 * src/arm/ffi.c (ffi_prep_args): Align structs by at least 4 bytes.
991 Only treat r0 as a struct address if we're actually returning a
993 Only copy the bytes that are actually within a struct.
994 (ffi_prep_cif_machdep): A Composite Type not larger than 4 bytes
995 is returned in r0, not passed by address.
996 (ffi_call): Allocate a word-sized temporary for the case where
997 a composite is returned in r0.
998 (ffi_prep_incoming_args_SYSV): Align as necessary.
1000 2007-08-05 Steven Newbury <s_j_newbury@yahoo.co.uk>
1002 * src/arm/ffi.c (FFI_INIT_TRAMPOLINE): Use __clear_cache instead of
1003 directly using the sys_cacheflush syscall.
1005 2007-07-27 Andrew Haley <aph@redhat.com>
1007 * src/arm/sysv.S (ffi_closure_SYSV): Add soft-float.
1009 2007-09-03 Maciej W. Rozycki <macro@linux-mips.org>
1011 * Makefile.am: Unify MIPS_IRIX and MIPS_LINUX into MIPS.
1012 * configure.ac: Likewise.
1013 * Makefile.in: Regenerate.
1014 * include/Makefile.in: Likewise.
1015 * testsuite/Makefile.in: Likewise.
1016 * configure: Likewise.
1018 2007-08-24 David Daney <ddaney@avtrex.com>
1020 * testsuite/libffi.call/return_sl.c: New test.
1022 2007-08-10 David Daney <ddaney@avtrex.com>
1024 * testsuite/libffi.call/cls_multi_ushort.c,
1025 testsuite/libffi.call/cls_align_uint16.c,
1026 testsuite/libffi.call/nested_struct1.c,
1027 testsuite/libffi.call/nested_struct3.c,
1028 testsuite/libffi.call/cls_7_1_byte.c,
1029 testsuite/libffi.call/nested_struct5.c,
1030 testsuite/libffi.call/cls_double.c,
1031 testsuite/libffi.call/nested_struct7.c,
1032 testsuite/libffi.call/cls_sint.c,
1033 testsuite/libffi.call/nested_struct9.c,
1034 testsuite/libffi.call/cls_20byte1.c,
1035 testsuite/libffi.call/cls_multi_sshortchar.c,
1036 testsuite/libffi.call/cls_align_sint64.c,
1037 testsuite/libffi.call/cls_3byte2.c,
1038 testsuite/libffi.call/cls_multi_schar.c,
1039 testsuite/libffi.call/cls_multi_uchar.c,
1040 testsuite/libffi.call/cls_19byte.c,
1041 testsuite/libffi.call/cls_9byte1.c,
1042 testsuite/libffi.call/cls_align_float.c,
1043 testsuite/libffi.call/closure_fn1.c,
1044 testsuite/libffi.call/problem1.c,
1045 testsuite/libffi.call/closure_fn3.c,
1046 testsuite/libffi.call/cls_sshort.c,
1047 testsuite/libffi.call/closure_fn5.c,
1048 testsuite/libffi.call/cls_align_double.c,
1049 testsuite/libffi.call/nested_struct.c,
1050 testsuite/libffi.call/cls_2byte.c,
1051 testsuite/libffi.call/nested_struct10.c,
1052 testsuite/libffi.call/cls_4byte.c,
1053 testsuite/libffi.call/cls_6byte.c,
1054 testsuite/libffi.call/cls_8byte.c,
1055 testsuite/libffi.call/cls_multi_sshort.c,
1056 testsuite/libffi.call/cls_align_sint16.c,
1057 testsuite/libffi.call/cls_align_uint32.c,
1058 testsuite/libffi.call/cls_20byte.c,
1059 testsuite/libffi.call/cls_float.c,
1060 testsuite/libffi.call/nested_struct2.c,
1061 testsuite/libffi.call/cls_5_1_byte.c,
1062 testsuite/libffi.call/nested_struct4.c,
1063 testsuite/libffi.call/cls_24byte.c,
1064 testsuite/libffi.call/nested_struct6.c,
1065 testsuite/libffi.call/cls_64byte.c,
1066 testsuite/libffi.call/nested_struct8.c,
1067 testsuite/libffi.call/cls_uint.c,
1068 testsuite/libffi.call/cls_multi_ushortchar.c,
1069 testsuite/libffi.call/cls_schar.c,
1070 testsuite/libffi.call/cls_uchar.c,
1071 testsuite/libffi.call/cls_align_uint64.c,
1072 testsuite/libffi.call/cls_ulonglong.c,
1073 testsuite/libffi.call/cls_align_longdouble.c,
1074 testsuite/libffi.call/cls_1_1byte.c,
1075 testsuite/libffi.call/cls_12byte.c,
1076 testsuite/libffi.call/cls_3_1byte.c,
1077 testsuite/libffi.call/cls_3byte1.c,
1078 testsuite/libffi.call/cls_4_1byte.c,
1079 testsuite/libffi.call/cls_6_1_byte.c,
1080 testsuite/libffi.call/cls_16byte.c,
1081 testsuite/libffi.call/cls_18byte.c,
1082 testsuite/libffi.call/closure_fn0.c,
1083 testsuite/libffi.call/cls_9byte2.c,
1084 testsuite/libffi.call/closure_fn2.c,
1085 testsuite/libffi.call/closure_fn4.c,
1086 testsuite/libffi.call/cls_ushort.c,
1087 testsuite/libffi.call/closure_fn6.c,
1088 testsuite/libffi.call/cls_5byte.c,
1089 testsuite/libffi.call/cls_align_pointer.c,
1090 testsuite/libffi.call/cls_7byte.c,
1091 testsuite/libffi.call/cls_align_sint32.c,
1092 testsuite/libffi.special/unwindtest_ffi_call.cc,
1093 testsuite/libffi.special/unwindtest.cc: Remove xfail for mips64*-*-*.
1095 2007-08-10 David Daney <ddaney@avtrex.com>
1098 * configure.ac: Don't treat mips64 as a special case.
1099 * Makefile.am (nodist_libffi_la_SOURCES): Add n32.S.
1100 * configure: Regenerate
1101 * Makefile.in: Ditto.
1102 * fficonfig.h.in: Ditto.
1103 * src/mips/ffitarget.h (REG_L, REG_S, SUBU, ADDU, SRL, LI): Indent.
1104 (LA, EH_FRAME_ALIGN, FDE_ADDR_BYTES): New preprocessor macros.
1105 (FFI_DEFAULT_ABI): Set for n64 case.
1106 (FFI_CLOSURES, FFI_TRAMPOLINE_SIZE): Define for n32 and n64 cases.
1107 * src/mips/n32.S (ffi_call_N32): Add debug macros and labels for FDE.
1108 (ffi_closure_N32): New function.
1109 (.eh_frame): New section
1110 * src/mips/o32.S: Clean up comments.
1111 (ffi_closure_O32): Pass ffi_closure parameter in $12.
1112 * src/mips/ffi.c: Use FFI_MIPS_N32 instead of
1113 _MIPS_SIM == _ABIN32 throughout.
1114 (FFI_MIPS_STOP_HERE): New, use in place of
1116 (ffi_prep_args): Use unsigned long to hold pointer values. Rewrite
1117 to support n32/n64 ABIs.
1118 (calc_n32_struct_flags): Rewrite.
1119 (calc_n32_return_struct_flags): Remove unused variable. Reverse
1120 position of flag bits.
1121 (ffi_prep_cif_machdep): Rewrite n32 portion.
1122 (ffi_call): Enable for n64. Add special handling for small structure
1124 (ffi_prep_closure_loc): Add n32 and n64 support.
1125 (ffi_closure_mips_inner_O32): Add cast to silence warning.
1126 (copy_struct_N32, ffi_closure_mips_inner_N32): New functions.
1128 2007-08-08 David Daney <ddaney@avtrex.com>
1130 * testsuite/libffi.call/ffitest.h (ffi_type_mylong): Remove definition.
1131 * testsuite/libffi.call/cls_align_uint16.c (main): Use correct type
1133 * testsuite/libffi.call/nested_struct1.c (main): Ditto.
1134 * testsuite/libffi.call/cls_sint.c (main): Ditto.
1135 * testsuite/libffi.call/nested_struct9.c (main): Ditto.
1136 * testsuite/libffi.call/cls_20byte1.c (main): Ditto.
1137 * testsuite/libffi.call/cls_9byte1.c (main): Ditto.
1138 * testsuite/libffi.call/closure_fn1.c (main): Ditto.
1139 * testsuite/libffi.call/closure_fn3.c (main): Ditto.
1140 * testsuite/libffi.call/return_dbl2.c (main): Ditto.
1141 * testsuite/libffi.call/cls_sshort.c (main): Ditto.
1142 * testsuite/libffi.call/return_fl3.c (main): Ditto.
1143 * testsuite/libffi.call/closure_fn5.c (main): Ditto.
1144 * testsuite/libffi.call/nested_struct.c (main): Ditto.
1145 * testsuite/libffi.call/nested_struct10.c (main): Ditto.
1146 * testsuite/libffi.call/return_ll1.c (main): Ditto.
1147 * testsuite/libffi.call/cls_8byte.c (main): Ditto.
1148 * testsuite/libffi.call/cls_align_uint32.c (main): Ditto.
1149 * testsuite/libffi.call/cls_align_sint16.c (main): Ditto.
1150 * testsuite/libffi.call/cls_20byte.c (main): Ditto.
1151 * testsuite/libffi.call/nested_struct2.c (main): Ditto.
1152 * testsuite/libffi.call/cls_24byte.c (main): Ditto.
1153 * testsuite/libffi.call/nested_struct6.c (main): Ditto.
1154 * testsuite/libffi.call/cls_uint.c (main): Ditto.
1155 * testsuite/libffi.call/cls_12byte.c (main): Ditto.
1156 * testsuite/libffi.call/cls_16byte.c (main): Ditto.
1157 * testsuite/libffi.call/closure_fn0.c (main): Ditto.
1158 * testsuite/libffi.call/cls_9byte2.c (main): Ditto.
1159 * testsuite/libffi.call/closure_fn2.c (main): Ditto.
1160 * testsuite/libffi.call/return_dbl1.c (main): Ditto.
1161 * testsuite/libffi.call/closure_fn4.c (main): Ditto.
1162 * testsuite/libffi.call/closure_fn6.c (main): Ditto.
1163 * testsuite/libffi.call/cls_align_sint32.c (main): Ditto.
1165 2007-08-07 Andrew Haley <aph@redhat.com>
1167 * src/x86/sysv.S (ffi_closure_raw_SYSV): Fix typo in previous
1170 2007-08-06 Andrew Haley <aph@redhat.com>
1173 * src/x86/sysv.S (ffi_closure_raw_SYSV): Handle FFI_TYPE_UINT8,
1174 FFI_TYPE_SINT8, FFI_TYPE_UINT16, FFI_TYPE_SINT16, FFI_TYPE_UINT32,
1177 2007-08-02 David Daney <ddaney@avtrex.com>
1179 * testsuite/libffi.call/return_ul.c (main): Define return type as
1180 ffi_arg. Use proper printf conversion specifier.
1182 2007-07-30 Andrew Haley <aph@redhat.com>
1185 * src/x86/ffi.c (ffi_prep_cif_machdep): in x86 case, add code for
1186 signed/unsigned int8/16.
1187 * src/x86/sysv.S (ffi_call_SYSV): Rewrite to:
1189 Remove code to pop args from the stack after call.
1190 Special-case signed/unsigned int8/16.
1191 * testsuite/libffi.call/return_sc.c (main): Revert.
1193 2007-07-26 Richard Guenther <rguenther@suse.de>
1196 * testsuite/libffi.call/return_sc.c (main): Verify call
1197 result as signed char, not ffi_arg.
1199 2007-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
1201 * configure.ac (i?86-*-solaris2.1[0-9]): Set TARGET to X86_64.
1202 * configure: Regenerate.
1204 2007-07-11 David Daney <ddaney@avtrex.com>
1206 * src/mips/ffi.c: Don't include sys/cachectl.h.
1207 (ffi_prep_closure_loc): Use __builtin___clear_cache() instead of
1210 2007-05-18 Aurelien Jarno <aurelien@aurel32.net>
1212 * src/arm/ffi.c (ffi_prep_closure_loc): Renamed and ajusted
1213 from (ffi_prep_closure): ... this.
1214 (FFI_INIT_TRAMPOLINE): Adjust.
1216 2005-12-31 Phil Blundell <pb@reciva.com>
1218 * src/arm/ffi.c (ffi_prep_incoming_args_SYSV,
1219 ffi_closure_SYSV_inner, ffi_prep_closure): New, add closure support.
1220 * src/arm/sysv.S(ffi_closure_SYSV): Likewise.
1221 * src/arm/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1222 (FFI_CLOSURES): Enable closure support.
1224 2007-07-03 Andrew Haley <aph@hedges.billgatliff.com>
1226 * testsuite/libffi.call/cls_multi_ushort.c,
1227 testsuite/libffi.call/cls_align_uint16.c,
1228 testsuite/libffi.call/nested_struct1.c,
1229 testsuite/libffi.call/nested_struct3.c,
1230 testsuite/libffi.call/cls_7_1_byte.c,
1231 testsuite/libffi.call/cls_double.c,
1232 testsuite/libffi.call/nested_struct5.c,
1233 testsuite/libffi.call/nested_struct7.c,
1234 testsuite/libffi.call/cls_sint.c,
1235 testsuite/libffi.call/nested_struct9.c,
1236 testsuite/libffi.call/cls_20byte1.c,
1237 testsuite/libffi.call/cls_multi_sshortchar.c,
1238 testsuite/libffi.call/cls_align_sint64.c,
1239 testsuite/libffi.call/cls_3byte2.c,
1240 testsuite/libffi.call/cls_multi_schar.c,
1241 testsuite/libffi.call/cls_multi_uchar.c,
1242 testsuite/libffi.call/cls_19byte.c,
1243 testsuite/libffi.call/cls_9byte1.c,
1244 testsuite/libffi.call/cls_align_float.c,
1245 testsuite/libffi.call/closure_fn1.c,
1246 testsuite/libffi.call/problem1.c,
1247 testsuite/libffi.call/closure_fn3.c,
1248 testsuite/libffi.call/cls_sshort.c,
1249 testsuite/libffi.call/closure_fn5.c,
1250 testsuite/libffi.call/cls_align_double.c,
1251 testsuite/libffi.call/cls_2byte.c,
1252 testsuite/libffi.call/nested_struct.c,
1253 testsuite/libffi.call/nested_struct10.c,
1254 testsuite/libffi.call/cls_4byte.c,
1255 testsuite/libffi.call/cls_6byte.c,
1256 testsuite/libffi.call/cls_8byte.c,
1257 testsuite/libffi.call/cls_multi_sshort.c,
1258 testsuite/libffi.call/cls_align_uint32.c,
1259 testsuite/libffi.call/cls_align_sint16.c,
1260 testsuite/libffi.call/cls_float.c,
1261 testsuite/libffi.call/cls_20byte.c,
1262 testsuite/libffi.call/cls_5_1_byte.c,
1263 testsuite/libffi.call/nested_struct2.c,
1264 testsuite/libffi.call/cls_24byte.c,
1265 testsuite/libffi.call/nested_struct4.c,
1266 testsuite/libffi.call/nested_struct6.c,
1267 testsuite/libffi.call/cls_64byte.c,
1268 testsuite/libffi.call/nested_struct8.c,
1269 testsuite/libffi.call/cls_uint.c,
1270 testsuite/libffi.call/cls_multi_ushortchar.c,
1271 testsuite/libffi.call/cls_schar.c,
1272 testsuite/libffi.call/cls_uchar.c,
1273 testsuite/libffi.call/cls_align_uint64.c,
1274 testsuite/libffi.call/cls_ulonglong.c,
1275 testsuite/libffi.call/cls_align_longdouble.c,
1276 testsuite/libffi.call/cls_1_1byte.c,
1277 testsuite/libffi.call/cls_12byte.c,
1278 testsuite/libffi.call/cls_3_1byte.c,
1279 testsuite/libffi.call/cls_3byte1.c,
1280 testsuite/libffi.call/cls_4_1byte.c,
1281 testsuite/libffi.call/cls_6_1_byte.c,
1282 testsuite/libffi.call/cls_16byte.c,
1283 testsuite/libffi.call/cls_18byte.c,
1284 testsuite/libffi.call/closure_fn0.c,
1285 testsuite/libffi.call/cls_9byte2.c,
1286 testsuite/libffi.call/closure_fn2.c,
1287 testsuite/libffi.call/closure_fn4.c,
1288 testsuite/libffi.call/cls_ushort.c,
1289 testsuite/libffi.call/closure_fn6.c,
1290 testsuite/libffi.call/cls_5byte.c,
1291 testsuite/libffi.call/cls_align_pointer.c,
1292 testsuite/libffi.call/cls_7byte.c,
1293 testsuite/libffi.call/cls_align_sint32.c,
1294 testsuite/libffi.special/unwindtest_ffi_call.cc,
1295 testsuite/libffi.special/unwindtest.cc: Enable for ARM.
1297 2007-07-05 H.J. Lu <hongjiu.lu@intel.com>
1299 * aclocal.m4: Regenerated.
1301 2007-06-02 Paolo Bonzini <bonzini@gnu.org>
1303 * configure: Regenerate.
1305 2007-05-23 Steve Ellcey <sje@cup.hp.com>
1307 * Makefile.in: Regenerate.
1308 * configure: Regenerate.
1309 * aclocal.m4: Regenerate.
1310 * include/Makefile.in: Regenerate.
1311 * testsuite/Makefile.in: Regenerate.
1313 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1315 * src/m68k/ffi.c (ffi_prep_incoming_args_SYSV,
1316 ffi_closure_SYSV_inner,ffi_prep_closure): New, add closure support.
1317 * src/m68k/sysv.S(ffi_closure_SYSV,ffi_closure_struct_SYSV): Likewise.
1318 * src/m68k/ffitarget.h (FFI_TRAMPOLINE_SIZE): Likewise.
1319 (FFI_CLOSURES): Enable closure support.
1321 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1323 * configure.ac (HAVE_AS_CFI_PSEUDO_OP): New test.
1324 * configure: Regenerate.
1325 * fficonfig.h.in: Regenerate.
1326 * src/m68k/sysv.S (CFI_STARTPROC,CFI_ENDPROC,
1327 CFI_OFFSET,CFI_DEF_CFA): New macros.
1328 (ffi_call_SYSV): Add callframe annotation.
1330 2007-05-10 Roman Zippel <zippel@linux-m68k.org>
1332 * src/m68k/ffi.c (ffi_prep_args,ffi_prep_cif_machdep): Fix
1333 numerous test suite failures.
1334 * src/m68k/sysv.S (ffi_call_SYSV): Likewise.
1336 2007-04-11 Paolo Bonzini <bonzini@gnu.org>
1338 * Makefile.am (EXTRA_DIST): Bring up to date.
1339 * Makefile.in: Regenerate.
1340 * src/frv/eabi.S: Remove RCS keyword.
1342 2007-04-06 Richard Henderson <rth@redhat.com>
1344 * configure.ac: Tidy target case.
1345 (HAVE_LONG_DOUBLE): Allow the target to override.
1346 * configure: Regenerate.
1347 * include/ffi.h.in: Don't define ffi_type_foo if
1348 LIBFFI_HIDE_BASIC_TYPES is defined.
1349 (ffi_type_longdouble): If not HAVE_LONG_DOUBLE, define
1351 * types.c (LIBFFI_HIDE_BASIC_TYPES): Define.
1352 (FFI_TYPEDEF, ffi_type_void): Mark the data const.
1353 (ffi_type_longdouble): Special case for Alpha. Don't define
1354 if long double == double.
1356 * src/alpha/ffi.c (FFI_TYPE_LONGDOUBLE): Assert unique value.
1357 (ffi_prep_cif_machdep): Handle it as the 128-bit type.
1358 (ffi_call, ffi_closure_osf_inner): Likewise.
1359 (ffi_closure_osf_inner): Likewise. Mark hidden.
1360 (ffi_call_osf, ffi_closure_osf): Mark hidden.
1361 * src/alpha/ffitarget.h (FFI_LAST_ABI): Tidy definition.
1362 * src/alpha/osf.S (ffi_call_osf, ffi_closure_osf): Mark hidden.
1363 (load_table): Handle 128-bit long double.
1365 * testsuite/libffi.call/float4.c: Add -mieee for alpha.
1367 2007-04-06 Tom Tromey <tromey@redhat.com>
1370 * README: Fixed bug in example.
1372 2007-04-03 Jakub Jelinek <jakub@redhat.com>
1374 * src/closures.c: Include sys/statfs.h.
1375 (_GNU_SOURCE): Define on Linux.
1376 (FFI_MMAP_EXEC_SELINUX): Define.
1377 (selinux_enabled): New variable.
1378 (selinux_enabled_check): New function.
1379 (is_selinux_enabled): Define.
1382 2007-03-24 Uros Bizjak <ubizjak@gmail.com>
1384 * testsuite/libffi.call/return_fl2.c (return_fl): Mark as static.
1385 Use 'volatile float sum' to create sum of floats to avoid false
1386 negative due to excess precision on ix86 targets.
1389 2007-03-08 Alexandre Oliva <aoliva@redhat.com>
1391 * src/powerpc/ffi.c (flush_icache): Fix left-over from previous
1393 (ffi_prep_closure_loc): Remove unneeded casts. Add needed ones.
1395 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1397 * include/ffi.h.in (ffi_closure_alloc, ffi_closure_free): New.
1398 (ffi_prep_closure_loc): New.
1399 (ffi_prep_raw_closure_loc): New.
1400 (ffi_prep_java_raw_closure_loc): New.
1401 * src/closures.c: New file.
1402 * src/dlmalloc.c [FFI_MMAP_EXEC_WRIT] (struct malloc_segment):
1403 Replace sflags with exec_offset.
1404 [FFI_MMAP_EXEC_WRIT] (mmap_exec_offset, add_segment_exec_offset,
1405 sub_segment_exec_offset): New macros.
1406 (get_segment_flags, set_segment_flags, check_segment_merge): New
1408 (is_mmapped_segment, is_extern_segment): Use get_segment_flags.
1409 (add_segment, sys_alloc, create_mspace, create_mspace_with_base,
1410 destroy_mspace): Use new macros.
1411 (sys_alloc): Silence warning.
1412 * Makefile.am (libffi_la_SOURCES): Add src/closures.c.
1413 * Makefile.in: Rebuilt.
1414 * src/prep_cif [FFI_CLOSURES] (ffi_prep_closure): Implement in
1415 terms of ffi_prep_closure_loc.
1416 * src/raw_api.c (ffi_prep_raw_closure_loc): Renamed and adjusted
1418 (ffi_prep_raw_closure): ... this. Re-implement in terms of the
1420 * src/java_raw_api (ffi_prep_java_raw_closure_loc): Renamed and
1422 (ffi_prep_java_raw_closure): ... this. Re-implement in terms of
1423 the renamed version.
1424 * src/alpha/ffi.c (ffi_prep_closure_loc): Renamed from
1425 (ffi_prep_closure): ... this.
1426 * src/pa/ffi.c: Likewise.
1427 * src/cris/ffi.c: Likewise. Adjust.
1428 * src/frv/ffi.c: Likewise.
1429 * src/ia64/ffi.c: Likewise.
1430 * src/mips/ffi.c: Likewise.
1431 * src/powerpc/ffi_darwin.c: Likewise.
1432 * src/s390/ffi.c: Likewise.
1433 * src/sh/ffi.c: Likewise.
1434 * src/sh64/ffi.c: Likewise.
1435 * src/sparc/ffi.c: Likewise.
1436 * src/x86/ffi64.c: Likewise.
1437 * src/x86/ffi.c: Likewise.
1438 (FFI_INIT_TRAMPOLINE): Adjust.
1439 (ffi_prep_raw_closure_loc): Renamed and adjusted from...
1440 (ffi_prep_raw_closure): ... this.
1441 * src/powerpc/ffi.c (ffi_prep_closure_loc): Renamed from
1442 (ffi_prep_closure): ... this.
1443 (flush_icache): Adjust.
1445 2007-03-07 Alexandre Oliva <aoliva@redhat.com>
1447 * src/dlmalloc.c: New file, imported version 2.8.3 of Doug
1450 2007-03-01 Brooks Moses <brooks.moses@codesourcery.com>
1452 * Makefile.am: Add dummy install-pdf target.
1453 * Makefile.in: Regenerate
1455 2007-02-13 Andreas Krebbel <krebbel1@de.ibm.com>
1457 * src/s390/ffi.c (ffi_prep_args, ffi_prep_cif_machdep,
1458 ffi_closure_helper_SYSV): Add long double handling.
1460 2007-02-02 Jakub Jelinek <jakub@redhat.com>
1462 * src/powerpc/linux64.S (ffi_call_LINUX64): Move restore of r2
1463 immediately after bctrl instruction.
1465 2007-01-18 Alexandre Oliva <aoliva@redhat.com>
1467 * Makefile.am (all-recursive, install-recursive,
1468 mostlyclean-recursive, clean-recursive, distclean-recursive,
1469 maintainer-clean-recursive): Add missing targets.
1470 * Makefile.in: Rebuilt.
1472 2006-12-14 Andreas Tobler <a.tobler@schweiz.org>
1474 * configure.ac: Add TARGET for x86_64-*-darwin*.
1475 * Makefile.am (nodist_libffi_la_SOURCES): Add rules for 64-bit sources
1477 * src/x86/ffitarget.h: Set trampoline size for x86_64-*-darwin*.
1478 * src/x86/darwin64.S: New file for x86_64-*-darwin* support.
1479 * configure: Regenerate.
1480 * Makefile.in: Regenerate.
1481 * include/Makefile.in: Regenerate.
1482 * testsuite/Makefile.in: Regenerate.
1483 * testsuite/libffi.special/unwindtest_ffi_call.cc: New test case for
1486 2006-12-13 Andreas Tobler <a.tobler@schweiz.org>
1488 * aclocal.m4: Regenerate with aclocal -I .. as written in the
1491 2006-10-31 Geoffrey Keating <geoffk@apple.com>
1493 * src/powerpc/ffi_darwin.c (darwin_adjust_aggregate_sizes): New.
1494 (ffi_prep_cif_machdep): Call darwin_adjust_aggregate_sizes for
1496 * testsuite/libffi.call/nested_struct4.c: Remove Darwin XFAIL.
1497 * testsuite/libffi.call/nested_struct6.c: Remove Darwin XFAIL.
1499 2006-10-10 Paolo Bonzini <bonzini@gnu.org>
1500 Sandro Tolaini <tolaini@libero.it>
1502 * configure.ac [i*86-*-darwin*]: Set X86_DARWIN symbol and
1504 * configure: Regenerated.
1505 * Makefile.am (nodist_libffi_la_SOURCES) [X86_DARWIN]: New case.
1506 (EXTRA_DIST): Add src/x86/darwin.S.
1507 * Makefile.in: Regenerated.
1508 * include/Makefile.in: Regenerated.
1509 * testsuite/Makefile.in: Regenerated.
1511 * src/x86/ffi.c (ffi_prep_cif_machdep) [X86_DARWIN]: Treat like
1512 X86_WIN32, and additionally align stack to 16 bytes.
1513 * src/x86/darwin.S: New, based on sysv.S.
1514 * src/prep_cif.c (ffi_prep_cif) [X86_DARWIN]: Align > 8-byte structs.
1516 2006-09-12 David Daney <ddaney@avtrex.com>
1519 * include/Makefile.am: Install both ffi.h and ffitarget.h in
1520 $(libdir)/gcc/$(target_alias)/$(gcc_version)/include.
1521 * aclocal.m4: Regenerated for automake 1.9.6.
1522 * Makefile.in: Regenerated.
1523 * include/Makefile.in: Regenerated.
1524 * testsuite/Makefile.in: Regenerated.
1526 2006-08-17 Andreas Tobler <a.tobler@schweiz.ch>
1528 * include/ffi_common.h (struct): Revert accidental commit.
1530 2006-08-15 Andreas Tobler <a.tobler@schweiz.ch>
1532 * include/ffi_common.h: Remove lint directives.
1533 * include/ffi.h.in: Likewise.
1535 2006-07-25 Torsten Schoenfeld <kaffeetisch@gmx.de>
1537 * include/ffi.h.in (ffi_type_ulong, ffi_type_slong): Define correctly
1538 for 32-bit architectures.
1539 * testsuite/libffi.call/return_ul.c: New test case.
1541 2006-07-19 David Daney <ddaney@avtrex.com>
1543 * testsuite/libffi.call/closure_fn6.c: Remove xfail for mips,
1544 xfail remains for mips64.
1546 2006-05-23 Carlos O'Donell <carlos@codesourcery.com>
1548 * Makefile.am: Add install-html target. Add install-html to .PHONY
1549 * Makefile.in: Regenerate.
1550 * aclocal.m4: Regenerate.
1551 * include/Makefile.in: Regenerate.
1552 * testsuite/Makefile.in: Regenerate.
1554 2006-05-18 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1556 * pa/ffi.c (ffi_prep_args_pa32): Load floating point arguments from
1559 2006-04-22 Andreas Tobler <a.tobler@schweiz.ch>
1561 * README: Remove notice about 'Crazy Comments'.
1562 * src/debug.c: Remove lint directives. Cleanup white spaces.
1563 * src/java_raw_api.c: Likewise.
1564 * src/prep_cif.c: Likewise.
1565 * src/raw_api.c: Likewise.
1566 * src/ffitest.c: Delete. No longer needed, all test cases migrated
1568 * src/arm/ffi.c: Remove lint directives.
1569 * src/m32r/ffi.c: Likewise.
1570 * src/pa/ffi.c: Likewise.
1571 * src/powerpc/ffi.c: Likewise.
1572 * src/powerpc/ffi_darwin.c: Likewise.
1573 * src/sh/ffi.c: Likewise.
1574 * src/sh64/ffi.c: Likewise.
1575 * src/x86/ffi.c: Likewise.
1576 * testsuite/libffi.call/float2.c: Likewise.
1577 * testsuite/libffi.call/promotion.c: Likewise.
1578 * testsuite/libffi.call/struct1.c: Likewise.
1580 2006-04-13 Andreas Tobler <a.tobler@schweiz.ch>
1582 * src/pa/hpux32.S: Correct unwind offset calculation for
1584 * src/pa/linux.S: Likewise.
1586 2006-04-12 James E Wilson <wilson@specifix.com>
1589 * src/ia64/ffi.c (stf_spill, ldf_fill): Rewrite as macros.
1590 (hfa_type_load): Call stf_spill.
1591 (hfa_type_store): Call ldf_fill.
1592 (ffi_call): Adjust calls to above routines. Add local temps for
1595 2006-04-10 Matthias Klose <doko@debian.org>
1597 * testsuite/lib/libffi-dg.exp (libffi-init): Recognize multilib
1598 directory names containing underscores.
1600 2006-04-07 James E Wilson <wilson@specifix.com>
1602 * testsuite/libffi.call/float4.c: New testcase.
1604 2006-04-05 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
1605 Andreas Tobler <a.tobler@schweiz.ch>
1607 * Makefile.am: Add PA_HPUX port.
1608 * Makefile.in: Regenerate.
1609 * include/Makefile.in: Likewise.
1610 * testsuite/Makefile.in: Likewise.
1611 * configure.ac: Add PA_HPUX rules.
1612 * configure: Regenerate.
1613 * src/pa/ffitarget.h: Rename linux target to PA_LINUX.
1614 Add PA_HPUX and PA64_HPUX.
1615 Rename FFI_LINUX ABI to FFI_PA32 ABI.
1616 (FFI_TRAMPOLINE_SIZE): Define for 32-bit HP-UX targets.
1617 (FFI_TYPE_SMALL_STRUCT2): Define.
1618 (FFI_TYPE_SMALL_STRUCT4): Likewise.
1619 (FFI_TYPE_SMALL_STRUCT8): Likewise.
1620 (FFI_TYPE_SMALL_STRUCT3): Redefine.
1621 (FFI_TYPE_SMALL_STRUCT5): Likewise.
1622 (FFI_TYPE_SMALL_STRUCT6): Likewise.
1623 (FFI_TYPE_SMALL_STRUCT7): Likewise.
1624 * src/pa/ffi.c (ROUND_DOWN): Delete.
1625 (fldw, fstw, fldd, fstd): Use '__asm__'.
1626 (ffi_struct_type): Add support for FFI_TYPE_SMALL_STRUCT2,
1627 FFI_TYPE_SMALL_STRUCT4 and FFI_TYPE_SMALL_STRUCT8.
1628 (ffi_prep_args_LINUX): Rename to ffi_prep_args_pa32. Update comment.
1629 Simplify incrementing of stack slot variable. Change type of local
1630 'n' to unsigned int.
1631 (ffi_size_stack_LINUX): Rename to ffi_size_stack_pa32. Handle long
1633 (ffi_prep_cif_machdep): Likewise.
1634 (ffi_call): Likewise.
1635 (ffi_closure_inner_LINUX): Rename to ffi_closure_inner_pa32. Change
1636 return type to ffi_status. Simplify incrementing of stack slot
1637 variable. Only copy floating point argument registers when PA_LINUX
1638 is true. Reformat debug statement.
1639 Add support for FFI_TYPE_SMALL_STRUCT2, FFI_TYPE_SMALL_STRUCT4 and
1640 FFI_TYPE_SMALL_STRUCT8.
1641 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Add 'extern' to
1643 (ffi_prep_closure): Make linux trampoline conditional on PA_LINUX.
1644 Add nops to cache flush. Add trampoline for PA_HPUX.
1645 * src/pa/hpux32.S: New file.
1646 * src/pa/linux.S (ffi_call_LINUX): Rename to ffi_call_pa32. Rename
1647 ffi_prep_args_LINUX to ffi_prep_args_pa32.
1648 Localize labels. Add support for 2, 4 and 8-byte small structs. Handle
1649 unaligned destinations in 3, 5, 6 and 7-byte small structs. Order
1650 argument type checks so that common argument types appear first.
1651 (ffi_closure_LINUX): Rename to ffi_closure_pa32. Rename
1652 ffi_closure_inner_LINUX to ffi_closure_inner_pa32.
1654 2006-03-24 Alan Modra <amodra@bigpond.net.au>
1656 * src/powerpc/ffitarget.h (enum ffi_abi): Add FFI_LINUX. Default
1657 for 32-bit using IBM extended double format. Fix FFI_LAST_ABI.
1658 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Handle linux variant of
1659 FFI_TYPE_LONGDOUBLE.
1660 (ffi_prep_args64): Assert using IBM extended double.
1661 (ffi_prep_cif_machdep): Don't munge FFI_TYPE_LONGDOUBLE type.
1662 Handle FFI_LINUX FFI_TYPE_LONGDOUBLE return and args.
1663 (ffi_call): Handle FFI_LINUX.
1664 (ffi_closure_helper_SYSV): Non FFI_LINUX long double return needs
1665 gpr3 return pointer as for struct return. Handle FFI_LINUX
1666 FFI_TYPE_LONGDOUBLE return and args. Don't increment "nf"
1668 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Load both f1 and f2
1669 for FFI_TYPE_LONGDOUBLE. Move epilogue insns into case table.
1670 Don't use r6 as pointer to results, instead use sp offset. Don't
1671 make a special call to load lr with case table address, instead
1672 use offset from previous call.
1673 * src/powerpc/sysv.S (ffi_call_SYSV): Save long double return.
1674 * src/powerpc/linux64.S (ffi_call_LINUX64): Simplify long double
1677 2006-03-15 Kaz Kojima <kkojima@gcc.gnu.org>
1679 * src/sh64/ffi.c (ffi_prep_cif_machdep): Handle float arguments
1680 passed with FP registers correctly.
1681 (ffi_closure_helper_SYSV): Likewise.
1682 * src/sh64/sysv.S: Likewise.
1684 2006-03-01 Andreas Tobler <a.tobler@schweiz.ch>
1686 * testsuite/libffi.special/unwindtest.cc (closure_test_fn): Mark cif,
1687 args and userdata unused.
1688 (closure_test_fn1): Mark cif and userdata unused.
1689 (main): Remove unused res.
1691 2006-02-28 Andreas Tobler <a.tobler@schweiz.ch>
1693 * testsuite/libffi.call/call.exp: Adjust FSF address. Add test runs for
1694 -O2, -O3, -Os and the warning flags -W -Wall.
1695 * testsuite/libffi.special/special.exp: Likewise.
1696 * testsuite/libffi.call/ffitest.h: Add an __UNUSED__ macro to mark
1697 unused parameter unused for gcc or else do nothing.
1698 * testsuite/libffi.special/ffitestcxx.h: Likewise.
1699 * testsuite/libffi.call/cls_12byte.c (cls_struct_12byte_gn): Mark cif
1700 and userdata unused.
1701 * testsuite/libffi.call/cls_16byte.c (cls_struct_16byte_gn): Likewise.
1702 * testsuite/libffi.call/cls_18byte.c (cls_struct_18byte_gn): Likewise.
1703 * testsuite/libffi.call/cls_19byte.c (cls_struct_19byte_gn): Likewise.
1704 * testsuite/libffi.call/cls_1_1byte.c (cls_struct_1_1byte_gn): Likewise.
1705 * testsuite/libffi.call/cls_20byte.c (cls_struct_20byte_gn): Likewise.
1706 * testsuite/libffi.call/cls_20byte1.c (cls_struct_20byte_gn): Likewise.
1707 * testsuite/libffi.call/cls_24byte.c (cls_struct_24byte_gn): Likewise.
1708 * testsuite/libffi.call/cls_2byte.c (cls_struct_2byte_gn): Likewise.
1709 * testsuite/libffi.call/cls_3_1byte.c (cls_struct_3_1byte_gn): Likewise.
1710 * testsuite/libffi.call/cls_3byte1.c (cls_struct_3byte_gn): Likewise.
1711 * testsuite/libffi.call/cls_3byte2.c (cls_struct_3byte_gn1): Likewise.
1712 * testsuite/libffi.call/cls_4_1byte.c (cls_struct_4_1byte_gn): Likewise.
1713 * testsuite/libffi.call/cls_4byte.c (cls_struct_4byte_gn): Likewise.
1714 * testsuite/libffi.call/cls_5_1_byte.c (cls_struct_5byte_gn): Likewise.
1715 * testsuite/libffi.call/cls_5byte.c (cls_struct_5byte_gn): Likewise.
1716 * testsuite/libffi.call/cls_64byte.c (cls_struct_64byte_gn): Likewise.
1717 * testsuite/libffi.call/cls_6_1_byte.c (cls_struct_6byte_gn): Likewise.
1718 * testsuite/libffi.call/cls_6byte.c (cls_struct_6byte_gn): Likewise.
1719 * testsuite/libffi.call/cls_7_1_byte.c (cls_struct_7byte_gn): Likewise.
1720 * testsuite/libffi.call/cls_7byte.c (cls_struct_7byte_gn): Likewise.
1721 * testsuite/libffi.call/cls_8byte.c (cls_struct_8byte_gn): Likewise.
1722 * testsuite/libffi.call/cls_9byte1.c (cls_struct_9byte_gn): Likewise.
1723 * testsuite/libffi.call/cls_9byte2.c (cls_struct_9byte_gn): Likewise.
1724 * testsuite/libffi.call/cls_align_double.c (cls_struct_align_gn):
1726 * testsuite/libffi.call/cls_align_float.c (cls_struct_align_gn):
1728 * testsuite/libffi.call/cls_align_longdouble.c (cls_struct_align_gn):
1730 * testsuite/libffi.call/cls_align_pointer.c (cls_struct_align_fn): Cast
1731 void* to avoid compiler warning.
1733 (cls_struct_align_gn): Mark cif and userdata unused.
1734 * testsuite/libffi.call/cls_align_sint16.c (cls_struct_align_gn):
1736 * testsuite/libffi.call/cls_align_sint32.c (cls_struct_align_gn):
1738 * testsuite/libffi.call/cls_align_sint64.c (cls_struct_align_gn):
1740 * testsuite/libffi.call/cls_align_uint16.c (cls_struct_align_gn):
1742 * testsuite/libffi.call/cls_align_uint32.c (cls_struct_align_gn):
1744 * testsuite/libffi.call/cls_double.c (cls_ret_double_fn): Likewise.
1745 * testsuite/libffi.call/cls_float.c (cls_ret_float_fn): Likewise.
1746 * testsuite/libffi.call/cls_multi_schar.c (test_func_gn): Mark cif and
1748 (main): Cast res_call to silence gcc.
1749 * testsuite/libffi.call/cls_multi_sshort.c (test_func_gn): Mark cif and
1751 (main): Cast res_call to silence gcc.
1752 * testsuite/libffi.call/cls_multi_sshortchar.c (test_func_gn): Mark cif
1754 (main): Cast res_call to silence gcc.
1755 * testsuite/libffi.call/cls_multi_uchar.c (test_func_gn): Mark cif and
1757 (main): Cast res_call to silence gcc.
1758 * testsuite/libffi.call/cls_multi_ushort.c (test_func_gn): Mark cif and
1760 (main): Cast res_call to silence gcc.
1761 * testsuite/libffi.call/cls_multi_ushortchar.c (test_func_gn): Mark cif
1763 (main): Cast res_call to silence gcc.
1764 * testsuite/libffi.call/cls_schar.c (cls_ret_schar_fn): Mark cif and
1766 (cls_ret_schar_fn): Cast printf parameter to silence gcc.
1767 * testsuite/libffi.call/cls_sint.c (cls_ret_sint_fn): Mark cif and
1769 (cls_ret_sint_fn): Cast printf parameter to silence gcc.
1770 * testsuite/libffi.call/cls_sshort.c (cls_ret_sshort_fn): Mark cif and
1772 (cls_ret_sshort_fn): Cast printf parameter to silence gcc.
1773 * testsuite/libffi.call/cls_uchar.c (cls_ret_uchar_fn): Mark cif and
1775 (cls_ret_uchar_fn): Cast printf parameter to silence gcc.
1776 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Mark cif and
1778 (cls_ret_uint_fn): Cast printf parameter to silence gcc.
1779 * testsuite/libffi.call/cls_ulonglong.c (cls_ret_ulonglong_fn): Mark cif
1780 and userdata unused.
1781 * testsuite/libffi.call/cls_ushort.c (cls_ret_ushort_fn): Mark cif and
1783 (cls_ret_ushort_fn): Cast printf parameter to silence gcc.
1784 * testsuite/libffi.call/float.c (floating): Remove unused parameter e.
1785 * testsuite/libffi.call/float1.c (main): Remove unused variable i.
1786 Cleanup white spaces.
1787 * testsuite/libffi.call/negint.c (checking): Remove unused variable i.
1788 * testsuite/libffi.call/nested_struct.c (cls_struct_combined_gn): Mark
1789 cif and userdata unused.
1790 * testsuite/libffi.call/nested_struct1.c (cls_struct_combined_gn):
1792 * testsuite/libffi.call/nested_struct10.c (B_gn): Likewise.
1793 * testsuite/libffi.call/nested_struct2.c (B_fn): Adjust printf
1794 formatters to silence gcc.
1795 (B_gn): Mark cif and userdata unused.
1796 * testsuite/libffi.call/nested_struct3.c (B_gn): Mark cif and userdata
1798 * testsuite/libffi.call/nested_struct4.c: Mention related PR.
1799 (B_gn): Mark cif and userdata unused.
1800 * testsuite/libffi.call/nested_struct5.c (B_gn): Mark cif and userdata
1802 * testsuite/libffi.call/nested_struct6.c: Mention related PR.
1803 (B_gn): Mark cif and userdata unused.
1804 * testsuite/libffi.call/nested_struct7.c (B_gn): Mark cif and userdata
1806 * testsuite/libffi.call/nested_struct8.c (B_gn): Likewise.
1807 * testsuite/libffi.call/nested_struct9.c (B_gn): Likewise.
1808 * testsuite/libffi.call/problem1.c (stub): Likewise.
1809 * testsuite/libffi.call/pyobjc-tc.c (main): Cast the result to silence
1811 * testsuite/libffi.call/return_fl2.c (return_fl): Add the note mentioned
1812 in the last commit for this test case in the test case itself.
1813 * testsuite/libffi.call/closure_fn0.c (closure_test_fn0): Mark cif as
1815 * testsuite/libffi.call/closure_fn1.c (closure_test_fn1): Likewise.
1816 * testsuite/libffi.call/closure_fn2.c (closure_test_fn2): Likewise.
1817 * testsuite/libffi.call/closure_fn3.c (closure_test_fn3): Likewise.
1818 * testsuite/libffi.call/closure_fn4.c (closure_test_fn0): Likewise.
1819 * testsuite/libffi.call/closure_fn5.c (closure_test_fn5): Likewise.
1820 * testsuite/libffi.call/closure_fn6.c (closure_test_fn0): Likewise.
1822 2006-02-22 Kaz Kojima <kkojima@gcc.gnu.org>
1824 * src/sh/sysv.S: Fix register numbers in the FDE for
1827 2006-02-20 Andreas Tobler <a.tobler@schweiz.ch>
1829 * testsuite/libffi.call/return_fl2.c (return_fl): Remove static
1830 declaration to avoid a false negative on ix86. See PR323.
1832 2006-02-18 Kaz Kojima <kkojima@gcc.gnu.org>
1834 * src/sh/ffi.c (ffi_closure_helper_SYSV): Remove unused variable
1835 and cast integer to void * if needed. Update the pointer to
1836 the FP register saved area correctly.
1838 2006-02-17 Andreas Tobler <a.tobler@schweiz.ch>
1840 * testsuite/libffi.call/nested_struct6.c: XFAIL this test until PR25630
1842 * testsuite/libffi.call/nested_struct4.c: Likewise.
1844 2006-02-16 Andreas Tobler <a.tobler@schweiz.ch>
1846 * testsuite/libffi.call/return_dbl.c: New test case.
1847 * testsuite/libffi.call/return_dbl1.c: Likewise.
1848 * testsuite/libffi.call/return_dbl2.c: Likewise.
1849 * testsuite/libffi.call/return_fl.c: Likewise.
1850 * testsuite/libffi.call/return_fl1.c: Likewise.
1851 * testsuite/libffi.call/return_fl2.c: Likewise.
1852 * testsuite/libffi.call/return_fl3.c: Likewise.
1853 * testsuite/libffi.call/closure_fn6.c: Likewise.
1855 * testsuite/libffi.call/nested_struct2.c: Remove ffi_type_mylong
1857 * testsuite/libffi.call/ffitest.h: Add ffi_type_mylong definition
1858 here to be used by other test cases too.
1860 * testsuite/libffi.call/nested_struct10.c: New test case.
1861 * testsuite/libffi.call/nested_struct9.c: Likewise.
1862 * testsuite/libffi.call/nested_struct8.c: Likewise.
1863 * testsuite/libffi.call/nested_struct7.c: Likewise.
1864 * testsuite/libffi.call/nested_struct6.c: Likewise.
1865 * testsuite/libffi.call/nested_struct5.c: Likewise.
1866 * testsuite/libffi.call/nested_struct4.c: Likewise.
1868 2006-01-21 Andreas Tobler <a.tobler@schweiz.ch>
1870 * configure.ac: Enable libffi for sparc64-*-freebsd*.
1871 * configure: Rebuilt.
1873 2006-01-18 Jakub Jelinek <jakub@redhat.com>
1875 * src/powerpc/sysv.S (smst_two_register): Don't call __ashldi3,
1876 instead do the shifting inline.
1877 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't compute %r5
1878 shift count unconditionally. Simplify load sequences for 1, 2, 3, 4
1879 and 8 byte structs, for the remaining struct sizes don't call
1880 __lshrdi3, instead do the shifting inline.
1882 2005-12-07 Thiemo Seufer <ths@networkno.de>
1884 * src/mips/ffitarget.h: Remove obsolete sgidefs.h include. Add
1885 missing parentheses.
1886 * src/mips/o32.S (ffi_call_O32): Code formatting. Define
1887 and use A3_OFF, FP_OFF, RA_OFF. Micro-optimizations.
1888 (ffi_closure_O32): Likewise, but with newly defined A3_OFF2,
1889 A2_OFF2, A1_OFF2, A0_OFF2, RA_OFF2, FP_OFF2, S0_OFF2, GP_OFF2,
1890 V1_OFF2, V0_OFF2, FA_1_1_OFF2, FA_1_0_OFF2, FA_0_1_OFF2,
1892 * src/mips/ffi.c (ffi_prep_args): Code formatting. Fix
1894 (ffi_prep_closure): Improve trampoline instruction scheduling.
1895 (ffi_closure_mips_inner_O32): Fix endianness bugs.
1897 2005-12-03 Alan Modra <amodra@bigpond.net.au>
1899 * src/powerpc/ffi.c: Formatting.
1900 (ffi_prep_args_SYSV): Avoid possible aliasing problems by using unions.
1901 (ffi_prep_args64): Likewise.
1903 2005-09-30 Geoffrey Keating <geoffk@apple.com>
1905 * testsuite/lib/libffi-dg.exp (libffi_target_compile): For
1906 darwin, use -shared-libgcc not -lgcc_s, and explain why.
1908 2005-09-26 Tom Tromey <tromey@redhat.com>
1910 * testsuite/libffi.call/float1.c (value_type): New typedef.
1911 (CANARY): New define.
1912 (main): Check for result buffer overflow.
1913 * src/powerpc/linux64.S: Handle linux64 long double returns.
1914 * src/powerpc/ffi.c (FLAG_RETURNS_128BITS): New constant.
1915 (ffi_prep_cif_machdep): Handle linux64 long double returns.
1917 2005-08-25 Alan Modra <amodra@bigpond.net.au>
1920 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Correct placement of stack
1922 (ffi_status ffi_prep_cif_machdep): Correct stack sizing for same.
1924 2005-08-11 Jakub Jelinek <jakub@redhat.com>
1926 * configure.ac (HAVE_HIDDEN_VISIBILITY_ATTRIBUTE): New test.
1927 (AH_BOTTOM): Add FFI_HIDDEN definition.
1928 * configure: Rebuilt.
1929 * fficonfig.h.in: Rebuilt.
1930 * src/powerpc/ffi.c (hidden): Remove.
1931 (ffi_closure_LINUX64, ffi_prep_args64, ffi_call_LINUX64,
1932 ffi_closure_helper_LINUX64): Use FFI_HIDDEN instead of hidden.
1933 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64,
1934 .ffi_closure_LINUX64): Use FFI_HIDDEN instead of .hidden.
1935 * src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV): Remove,
1936 add FFI_HIDDEN to its prototype.
1937 (ffi_closure_SYSV_inner): New.
1938 * src/x86/sysv.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1939 * src/x86/win32.S (ffi_closure_SYSV, ffi_closure_raw_SYSV): New.
1941 2005-08-10 Alfred M. Szmidt <ams@gnu.org>
1944 * configure: Rebuilt.
1945 * configure.ac: Handle i*86-*-gnu*.
1947 2005-08-09 Jakub Jelinek <jakub@redhat.com>
1949 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Use
1950 DW_CFA_offset_extended_sf rather than
1951 DW_CFA_GNU_negative_offset_extended.
1952 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise.
1954 2005-07-22 SUGIOKA Toshinobu <sugioka@itonet.co.jp>
1956 * src/sh/sysv.S (ffi_call_SYSV): Stop argument popping correctly
1958 (ffi_closure_SYSV): Change the stack layout for sh3 struct argument.
1959 * src/sh/ffi.c (ffi_prep_args): Fix sh3 argument copy, when it is
1960 partially on register.
1961 (ffi_closure_helper_SYSV): Likewise.
1962 (ffi_prep_cif_machdep): Don't set too many cif->flags.
1964 2005-07-20 Kaz Kojima <kkojima@gcc.gnu.org>
1966 * src/sh/ffi.c (ffi_call): Handle small structures correctly.
1968 * src/sh64/ffi.c (simple_type): Remove.
1969 (return_type): Handle small structures correctly.
1970 (ffi_prep_args): Likewise.
1971 (ffi_call): Likewise.
1972 (ffi_closure_helper_SYSV): Likewise.
1973 * src/sh64/sysv.S (ffi_call_SYSV): Handle 1, 2 and 4-byte return.
1974 Emit position independent code if PIC and remove wrong datalabel
1975 prefixes from EH data.
1977 2005-07-19 Andreas Tobler <a.tobler@schweiz.ch>
1979 * Makefile.am (nodist_libffi_la_SOURCES): Add POWERPC_FREEBSD.
1980 * Makefile.in: Regenerate.
1981 * include/Makefile.in: Likewise.
1982 * testsuite/Makefile.in: Likewise.
1983 * configure.ac: Add POWERPC_FREEBSD rules.
1984 * configure: Regenerate.
1985 * src/powerpc/ffitarget.h: Add POWERPC_FREEBSD rules.
1986 (FFI_SYSV_TYPE_SMALL_STRUCT): Define.
1987 * src/powerpc/ffi.c: Add flags to handle small structure returns
1989 (ffi_prep_cif_machdep): Handle small structures for SYSV 4 ABI.
1991 (ffi_closure_helper_SYSV): Likewise.
1992 * src/powerpc/ppc_closure.S: Add return types for small structures.
1993 * src/powerpc/sysv.S: Add bits to handle small structures for
1996 2005-07-10 Andreas Tobler <a.tobler@schweiz.ch>
1998 * testsuite/libffi.call/cls_5_1_byte.c: New test file.
1999 * testsuite/libffi.call/cls_6_1_byte.c: Likewise.
2000 * testsuite/libffi.call/cls_7_1_byte.c: Likewise.
2002 2005-07-05 Randolph Chung <tausq@debian.org>
2004 * src/pa/ffi.c (ffi_struct_type): Rename FFI_TYPE_SMALL_STRUCT1
2005 as FFI_TYPE_SMALL_STRUCT3. Break out handling for 5-7 byte
2006 structures. Kill compilation warnings.
2007 (ffi_closure_inner_LINUX): Print return values as hex in debug
2008 message. Rename FFI_TYPE_SMALL_STRUCT1 as FFI_TYPE_SMALL_STRUCT3.
2009 Properly handle 5-7 byte structure returns.
2010 * src/pa/ffitarget.h (FFI_TYPE_SMALL_STRUCT1)
2011 (FFI_TYPE_SMALL_STRUCT2): Remove.
2012 (FFI_TYPE_SMALL_STRUCT3, FFI_TYPE_SMALL_STRUCT5)
2013 (FFI_TYPE_SMALL_STRUCT6, FFI_TYPE_SMALL_STRUCT7): Define.
2014 * src/pa/linux.S: Mark source file as using PA1.1 assembly.
2015 (checksmst1, checksmst2): Remove.
2016 (checksmst3): Optimize handling of 3-byte struct returns.
2017 (checksmst567): Properly handle 5-7 byte struct returns.
2019 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2022 * src/mips/n32.S: Enforce PIC code.
2023 * src/mips/o32.S: Likewise.
2025 2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2027 * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
2028 * configure: Regenerate.
2030 2005-06-01 Alan Modra <amodra@bigpond.net.au>
2032 * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
2033 to call ffi_closure_helper_SYSV. Append @local instead.
2034 * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
2036 2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
2038 * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
2039 Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
2040 * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
2041 * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
2042 include/Makefile.in, testsuite/Makefile.in: Regenerate.
2044 2005-05-09 Mike Stump <mrs@apple.com>
2046 * configure: Regenerate.
2048 2005-05-08 Richard Henderson <rth@redhat.com>
2051 * src/alpha/osf.S: Update unwind into to match code.
2053 2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
2054 Richard Henderson <rth@redhat.com>
2056 * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
2058 (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
2059 (ffi_prep_closure): Set carry bit if sse-used flag set.
2060 * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
2061 Only load sse registers if ssecount non-zero.
2062 (ffi_closure_unix64): Only save sse registers if carry set on entry.
2064 2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
2066 * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
2067 powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
2068 * configure: Regenerate.
2070 2005-04-20 Hans-Peter Nilsson <hp@axis.com>
2072 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
2073 have Tcl8.3-compatible intermediate variable.
2075 2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
2076 Hans-Peter Nilsson <hp@axis.com>
2078 * Makefile.am: Add CRIS support.
2079 * configure.ac: Likewise.
2080 * Makefile.in, configure, testsuite/Makefile.in,
2081 include/Makefile.in: Regenerate.
2082 * src/cris: New directory.
2083 * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
2084 * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
2086 * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
2087 \r?\n in output tests.
2089 2005-04-12 Mike Stump <mrs@apple.com>
2091 * configure: Regenerate.
2093 2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
2095 * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
2097 2005-03-30 Steve Ellcey <sje@cup.hp.com>
2099 * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
2101 * src/ia64/unix.S (ffi_closure_unix): Extend gp
2102 to 64 bits in ILP32 mode.
2103 Load 64 bits even for short data.
2105 2005-03-23 Mike Stump <mrs@apple.com>
2107 * src/powerpc/darwin.S: Update for -m64 multilib.
2108 * src/powerpc/darwin_closure.S: Likewise.
2110 2005-03-21 Zack Weinberg <zack@codesourcery.com>
2112 * configure.ac: Do not invoke TL_AC_GCC_VERSION.
2113 Do not set tool_include_dir.
2114 * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
2116 * include/Makefile.am: Set gcc_version and toollibffidir.
2117 * include/Makefile.in: Regenerate.
2119 2005-02-22 Andrew Haley <aph@redhat.com>
2121 * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
2122 odd-numbered register pairs for 64-bit integer types.
2124 2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
2127 * testsuite/libffi.call/return_ll1.c: New test case.
2129 2005-02-11 Janis Johnson <janis187@us.ibm.com>
2131 * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
2132 * testsuite/libffi.call/float.c: Ditto.
2133 * testsuite/libffi.call/float2.c: Ditto.
2134 * testsuite/libffi.call/float3.c: Ditto.
2136 2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
2138 * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
2140 2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
2142 * testsuite/libffi.special/special.exp (cxx_options): Add
2145 2004-12-31 Richard Henderson <rth@redhat.com>
2147 * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
2148 (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
2149 offset parameters with a type parameter; deduce size and structure
2150 alignment. Update all users.
2152 2004-12-31 Richard Henderson <rth@redhat.com>
2154 * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
2155 (FFI_TYPE_LONGDOUBLE): Fix for ia64.
2156 * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
2157 into ffi_prep_closure.
2158 * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
2161 2004-12-27 Richard Henderson <rth@redhat.com>
2163 * src/x86/unix64.S: Fix typo in unwind info.
2165 2004-12-25 Richard Henderson <rth@redhat.com>
2167 * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
2168 (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
2169 (merge_classes): Check for it.
2171 (classify_argument): Pass byte_offset by value; perform all updates
2173 (examine_argument): Add classes argument; handle
2174 X86_64_COMPLEX_X87_CLASS.
2175 (ffi_prep_args): Merge into ...
2176 (ffi_call): ... here. Share stack frame with ffi_call_unix64.
2177 (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
2178 (ffi_fill_return_value): Remove.
2179 (ffi_prep_closure): Remove dead assert.
2180 (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
2181 Rewrite to use struct register_args instead of va_list. Create
2182 flags for handling structure returns.
2183 * src/x86/unix64.S: Remove dead strings.
2184 (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
2185 stack frame with ffi_call. Handle structure returns properly.
2186 (float2sse, floatfloat2sse, double2sse): Remove.
2187 (sse2float, sse2double, sse2floatfloat): Remove.
2188 (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
2189 to handle structure returns properly.
2191 2004-12-08 David Edelsohn <edelsohn@gnu.org>
2193 * Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
2195 * Makefile.in: Regenerated.
2197 2004-12-02 Richard Sandiford <rsandifo@redhat.com>
2199 * configure.ac: Use TL_AC_GCC_VERSION to set gcc_version.
2200 * configure, aclocal.m4, Makefile.in: Regenerate.
2201 * include/Makefile.in, testsuite/Makefile.in: Regenerate.
2203 2004-11-29 Kelley Cook <kcook@gcc.gnu.org>
2205 * configure: Regenerate for libtool change.
2207 2004-11-25 Kelley Cook <kcook@gcc.gnu.org>
2209 * configure: Regenerate for libtool reversion.
2211 2004-11-24 Kelley Cook <kcook@gcc.gnu.org>
2213 * configure: Regenerate for libtool change.
2215 2004-11-23 John David Anglin <dave.anglin@nrc-cnrc.gc.ca>
2217 * testsuite/lib/libffi-dg.exp: Use new procs in target-libpath.exp.
2219 2004-11-23 Richard Sandiford <rsandifo@redhat.com>
2221 * src/mips/o32.S (ffi_call_O32, ffi_closure_O32): Use jalr instead
2222 of jal. Use an absolute encoding for the frame information.
2224 2004-11-23 Kelley Cook <kcook@gcc.gnu.org>
2226 * Makefile.am: Remove no-dependencies. Add ACLOCAL_AMFLAGS.
2227 * acinclude.m4: Delete logic for sincludes.
2228 * aclocal.m4, Makefile.in, configure: Regenerate.
2229 * include/Makefile: Likewise.
2230 * testsuite/Makefile: Likewise.
2232 2004-11-22 Eric Botcazou <ebotcazou@libertysurf.fr>
2234 * src/sparc/ffi.c (ffi_prep_closure): Align doubles and 64-bit integers
2235 on a 8-byte boundary.
2236 * src/sparc/v8.S (ffi_closure_v8): Reserve frame space for arguments.
2238 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2240 * src/arm/ffi.c (ffi_prep_cif_machdep): Handle functions that return
2241 long long values. Round stack allocation to a multiple of 8 bytes
2242 for ATPCS compatibility.
2243 * src/arm/sysv.S (ffi_call_SYSV): Rework to avoid use of APCS register
2244 names. Handle returning long long types. Add Thumb and interworking
2245 support. Improve soft-float code.
2247 2004-10-27 Richard Earnshaw <rearnsha@arm.com>
2249 * testsuite/lib/libffi-db.exp (load_gcc_lib): New function.
2250 (libffi_exit): New function.
2251 (libffi_init): Build the testglue wrapper if needed.
2253 2004-10-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2256 * testsuite/lib/libffi-dg.exp: Accept more than one multilib libgcc.
2258 2004-10-25 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2260 * src/m32r/libffitarget.h (FFI_CLOSURES): Set to 0.
2262 2004-10-20 Kaz Kojima <kkojima@gcc.gnu.org>
2264 * src/sh/sysv.S (ffi_call_SYSV): Don't align for double data.
2265 * testsuite/libffi.call/float3.c: New test case.
2267 2004-10-18 Kaz Kojima <kkojima@gcc.gnu.org>
2269 * src/sh/ffi.c (ffi_prep_closure): Set T bit in trampoline for
2270 the function returning a structure pointed with R2.
2271 * src/sh/sysv.S (ffi_closure_SYSV): Use R2 as the pointer to
2272 the structure return value if T bit set. Emit position
2273 independent code and EH data if PIC.
2275 2004-10-13 Kazuhiro Inaoka <inaoka.kazuhiro@renesas.com>
2277 * Makefile.am: Add m32r support.
2278 * configure.ac: Likewise.
2279 * Makefile.in: Regenerate.
2280 * confiugre: Regenerate.
2281 * src/types.c: Add m32r port to FFI_INTERNAL_TYPEDEF
2282 (uint64, sint64, double, longdouble)
2283 * src/m32r: New directory.
2284 * src/m32r/ffi.c: New file.
2285 * src/m32r/sysv.S: Likewise.
2286 * src/m32r/ffitarget.h: Likewise.
2288 2004-10-02 Kaz Kojima <kkojima@gcc.gnu.org>
2290 * testsuite/libffi.call/negint.c: New test case.
2292 2004-09-14 H.J. Lu <hongjiu.lu@intel.com>
2295 * testsuite/lib/libffi-dg.exp: Don't use global ld_library_path.
2296 Set up LD_LIBRARY_PATH, SHLIB_PATH, LD_LIBRARYN32_PATH,
2297 LD_LIBRARY64_PATH, LD_LIBRARY_PATH_32, LD_LIBRARY_PATH_64 and
2300 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2302 * testsuite/libffi.call/many_win32.c: Remove whitespaces.
2303 * testsuite/libffi.call/promotion.c: Likewise.
2304 * testsuite/libffi.call/return_ll.c: Remove unused var. Cleanup
2306 * testsuite/libffi.call/return_sc.c: Likewise.
2307 * testsuite/libffi.call/return_uc.c: Likewise.
2309 2004-09-05 Andreas Tobler <a.tobler@schweiz.ch>
2311 * src/powerpc/darwin.S: Fix comments and identation.
2312 * src/powerpc/darwin_closure.S: Likewise.
2314 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2316 * src/powerpc/ffi_darwin.c: Add flag for longdouble return values.
2317 (ffi_prep_args): Handle longdouble arguments.
2318 (ffi_prep_cif_machdep): Set flags for longdouble. Calculate space for
2320 (ffi_closure_helper_DARWIN): Add closure handling for longdouble.
2321 * src/powerpc/darwin.S (_ffi_call_DARWIN): Add handling of longdouble
2323 * src/powerpc/darwin_closure.S (_ffi_closure_ASM): Likewise.
2324 * src/types.c: Defined longdouble size and alignment for darwin.
2326 2004-09-02 Andreas Tobler <a.tobler@schweiz.ch>
2328 * src/powerpc/aix.S: Remove whitespaces.
2329 * src/powerpc/aix_closure.S: Likewise.
2330 * src/powerpc/asm.h: Likewise.
2331 * src/powerpc/ffi.c: Likewise.
2332 * src/powerpc/ffitarget.h: Likewise.
2333 * src/powerpc/linux64.S: Likewise.
2334 * src/powerpc/linux64_closure.S: Likewise.
2335 * src/powerpc/ppc_closure.S: Likewise.
2336 * src/powerpc/sysv.S: Likewise.
2338 2004-08-30 Anthony Green <green@redhat.com>
2340 * Makefile.am: Add frv support.
2341 * Makefile.in, testsuite/Makefile.in: Rebuilt.
2342 * configure.ac: Read configure.host.
2343 * configure.in: Read configure.host.
2344 * configure.host: New file. frv-elf needs libgloss.
2345 * include/ffi.h.in: Force ffi_closure to have a nice big (8)
2346 alignment. This is needed to frv and shouldn't harm the others.
2347 * include/ffi_common.h (ALIGN_DOWN): New macro.
2348 * src/frv/ffi.c, src/frv/ffitarget.h, src/frv/eabi.S: New files.
2350 2004-08-24 David Daney <daney@avtrex.com>
2352 * testsuite/libffi.call/closure_fn0.c: Xfail mips64* instead of mips*.
2353 * testsuite/libffi.call/closure_fn1.c: Likewise.
2354 * testsuite/libffi.call/closure_fn2.c Likewise.
2355 * testsuite/libffi.call/closure_fn3.c: Likewise.
2356 * testsuite/libffi.call/closure_fn4.c: Likewise.
2357 * testsuite/libffi.call/closure_fn5.c: Likewise.
2358 * testsuite/libffi.call/cls_18byte.c: Likewise.
2359 * testsuite/libffi.call/cls_19byte.c: Likewise.
2360 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2361 * testsuite/libffi.call/cls_20byte.c: Likewise.
2362 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2363 * testsuite/libffi.call/cls_24byte.c: Likewise.
2364 * testsuite/libffi.call/cls_2byte.c: Likewise.
2365 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2366 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2367 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2368 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2369 * testsuite/libffi.call/cls_4byte.c: Likewise.
2370 * testsuite/libffi.call/cls_64byte.c: Likewise.
2371 * testsuite/libffi.call/cls_6byte.c: Likewise.
2372 * testsuite/libffi.call/cls_7byte.c: Likewise.
2373 * testsuite/libffi.call/cls_8byte.c: Likewise.
2374 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2375 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2376 * testsuite/libffi.call/cls_align_double.c: Likewise.
2377 * testsuite/libffi.call/cls_align_float.c: Likewise.
2378 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2379 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2380 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2381 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2382 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2383 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2384 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2385 * testsuite/libffi.call/cls_align_uint64.c: Likewise.
2386 * testsuite/libffi.call/cls_double.c: Likewise.
2387 * testsuite/libffi.call/cls_float.c: Likewise.
2388 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2389 * testsuite/libffi.call/cls_multi_sshort.c: Likewise.
2390 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2391 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2392 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2393 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2394 * testsuite/libffi.call/cls_schar.c: Likewise.
2395 * testsuite/libffi.call/cls_sint.c: Likewise.
2396 * testsuite/libffi.call/cls_sshort.c: Likewise.
2397 * testsuite/libffi.call/cls_uchar.c: Likewise.
2398 * testsuite/libffi.call/cls_uint.c: Likewise.
2399 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2400 * testsuite/libffi.call/cls_ushort.c: Likewise.
2401 * testsuite/libffi.call/nested_struct.c: Likewise.
2402 * testsuite/libffi.call/nested_struct1.c: Likewise.
2403 * testsuite/libffi.call/nested_struct2.c: Likewise.
2404 * testsuite/libffi.call/nested_struct3.c: Likewise.
2405 * testsuite/libffi.call/problem1.c: Likewise.
2406 * testsuite/libffi.special/unwindtest.cc: Likewise.
2407 * testsuite/libffi.call/cls_12byte.c: Likewise and set return value
2409 * testsuite/libffi.call/cls_16byte.c: Likewise.
2410 * testsuite/libffi.call/cls_5byte.c: Likewise.
2412 2004-08-23 David Daney <daney@avtrex.com>
2415 * src/mips/ffitarget.h (FFI_O32_SOFT_FLOAT): New ABI.
2416 * src/mips/ffi.c (ffi_prep_args): Fix alignment calculation.
2417 (ffi_prep_cif_machdep): Handle FFI_O32_SOFT_FLOAT floating point
2418 parameters and return types.
2419 (ffi_call): Handle FFI_O32_SOFT_FLOAT ABI.
2420 (ffi_prep_closure): Ditto.
2421 (ffi_closure_mips_inner_O32): Handle FFI_O32_SOFT_FLOAT ABI, fix
2422 alignment calculations.
2423 * src/mips/o32.S (ffi_closure_O32): Don't use floating point
2424 instructions if FFI_O32_SOFT_FLOAT, make stack frame ABI compliant.
2426 2004-08-14 Casey Marshall <csm@gnu.org>
2428 * src/mips/ffi.c (ffi_pref_cif_machdep): set `cif->flags' to
2429 contain `FFI_TYPE_UINT64' as return type for any 64-bit
2430 integer (O32 ABI only).
2431 (ffi_prep_closure): new function.
2432 (ffi_closure_mips_inner_O32): new function.
2433 * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
2434 `FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
2435 * src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
2436 64 bit integers correctly.
2437 (ffi_closure_O32): new function.
2438 Added DWARF-2 unwind info for both functions.
2440 2004-08-10 Andrew Haley <aph@redhat.com>
2442 * src/x86/ffi64.c (ffi_prep_args ): 8-align all stack arguments.
2444 2004-08-01 Robert Millan <robertmh@gnu.org>
2446 * configure.ac: Detect knetbsd-gnu and kfreebsd-gnu.
2447 * configure: Regenerate.
2449 2004-07-30 Maciej W. Rozycki <macro@linux-mips.org>
2451 * acinclude.m4 (AC_FUNC_MMAP_BLACKLIST): Check for <sys/mman.h>
2452 and mmap() explicitly instead of relying on preset autoconf cache
2454 * aclocal.m4: Regenerate.
2455 * configure: Regenerate.
2457 2004-07-11 Ulrich Weigand <uweigand@de.ibm.com>
2459 * src/s390/ffi.c (ffi_prep_args): Fix C aliasing violation.
2460 (ffi_check_float_struct): Remove unused prototype.
2462 2004-06-30 Geoffrey Keating <geoffk@apple.com>
2464 * src/powerpc/ffi_darwin.c (flush_icache): ';' is a comment
2465 character on Darwin, use '\n\t' instead.
2467 2004-06-26 Matthias Klose <doko@debian.org>
2469 * libtool-version: Fix typo in revision/age.
2471 2004-06-17 Matthias Klose <doko@debian.org>
2473 * libtool-version: New.
2474 * Makefile.am (libffi_la_LDFLAGS): Use -version-info for soname.
2475 * Makefile.in: Regenerate.
2477 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2479 * Makefile.am: Remove useless multilib rules.
2480 * Makefile.in: Regenerate.
2481 * aclocal.m4: Regenerate with automake 1.8.5.
2482 * configure.ac: Remove useless multilib configury.
2483 * configure: Regenerate.
2485 2004-06-15 Paolo Bonzini <bonzini@gnu.org>
2487 * .cvsignore: New file.
2489 2004-06-10 Jakub Jelinek <jakub@redhat.com>
2491 * src/ia64/unix.S (ffi_call_unix): Insert group barrier break
2493 (ffi_closure_UNIX): Fix f14/f15 adjustment if FLOAT_SZ is ever
2496 2004-06-06 Sean McNeil <sean@mcneil.com>
2498 * configure.ac: Add x86_64-*-freebsd* support.
2499 * configure: Regenerate.
2501 2004-04-26 Joe Buck <jbuck@welsh-buck.org>
2504 * configure.ac: Test for existence of mmap and sys/mman.h before
2505 checking blacklist. Fix suggested by Jim Wilson.
2506 * configure: Regenerate.
2508 2004-04-26 Matt Austern <austern@apple.com>
2510 * src/powerpc/darwin.S: Go through a non-lazy pointer for initial
2512 * src/powerpc/darwin_closure.S: Likewise.
2514 2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
2516 * testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2517 error. Reported by Thomas Heller <theller@python.net>.
2518 * testsuite/libffi.call/cls_multi_sshort.c (main): Likewise.
2519 * testsuite/libffi.call/cls_multi_ushort.c (main): Likewise.
2521 2004-03-20 Matthias Klose <doko@debian.org>
2523 * src/pa/linux.S: Fix typo.
2525 2004-03-19 Matthias Klose <doko@debian.org>
2527 * Makefile.am: Update.
2528 * Makefile.in: Regenerate.
2529 * src/pa/ffi.h.in: Remove.
2530 * src/pa/ffitarget.h: New file.
2532 2004-02-10 Randolph Chung <tausq@debian.org>
2534 * Makefile.am: Add PA support.
2535 * Makefile.in: Regenerate.
2536 * include/Makefile.in: Regenerate.
2537 * configure.ac: Add PA target.
2538 * configure: Regenerate.
2539 * src/pa/ffi.c: New file.
2540 * src/pa/ffi.h.in: Add PA support.
2541 * src/pa/linux.S: New file.
2542 * prep_cif.c: Add PA support.
2544 2004-03-16 Hosaka Yuji <hos@tamanegi.org>
2546 * src/types.c: Fix alignment size of X86_WIN32 case int64 and
2548 * src/x86/ffi.c (ffi_prep_args): Replace ecif->cif->rtype->type
2549 with ecif->cif->flags.
2550 (ffi_call, ffi_prep_incoming_args_SYSV): Replace cif->rtype->type
2552 (ffi_prep_cif_machdep): Add X86_WIN32 struct case.
2553 (ffi_closure_SYSV): Add 1 or 2-bytes struct case for X86_WIN32.
2554 * src/x86/win32.S (retstruct1b, retstruct2b, sc_retstruct1b,
2555 sc_retstruct2b): Add for 1 or 2-bytes struct case.
2557 2004-03-15 Kelley Cook <kcook@gcc.gnu.org>
2559 * configure.in: Rename file to ...
2560 * configure.ac: ... this.
2561 * fficonfig.h.in: Regenerate.
2562 * Makefile.in: Regenerate.
2563 * include/Makefile.in: Regenerate.
2564 * testsuite/Makefile.in: Regenerate.
2566 2004-03-12 Matt Austern <austern@apple.com>
2568 * src/powerpc/darwin.S: Fix EH information so it corresponds to
2569 changes in EH format resulting from addition of linkonce support.
2570 * src/powerpc/darwin_closure.S: Likewise.
2572 2004-03-11 Andreas Tobler <a.tobler@schweiz.ch>
2573 Paolo Bonzini <bonzini@gnu.org>
2575 * Makefile.am (AUTOMAKE_OPTIONS): Set them.
2576 Remove VPATH. Remove rules for object files. Remove multilib support.
2577 (AM_CCASFLAGS): Add.
2578 * configure.in (AC_CONFIG_HEADERS): Relace AM_CONFIG_HEADER.
2579 (AC_PREREQ): Bump version to 2.59.
2580 (AC_INIT): Fill with version info and bug address.
2581 (ORIGINAL_LD_FOR_MULTILIBS): Remove.
2582 (AM_ENABLE_MULTILIB): Use this instead of AC_ARG_ENABLE.
2583 De-precious CC so that the right flags are passed down to multilibs.
2584 (AC_MSG_ERROR): Replace obsolete macro AC_ERROR.
2585 (AC_CONFIG_FILES): Replace obsolete macro AC_LINK_FILES.
2586 (AC_OUTPUT): Reorganize the output with AC_CONFIG_COMMANDS.
2587 * configure: Rebuilt.
2588 * aclocal.m4: Likewise.
2589 * Makefile.in, include/Makefile.in, testsuite/Makefile.in: Likewise.
2590 * fficonfig.h.in: Likewise.
2592 2004-03-11 Andreas Schwab <schwab@suse.de>
2594 * src/ia64/ffi.c (ffi_prep_incoming_args_UNIX): Get floating point
2595 arguments from fp registers only for the first 8 parameter slots.
2596 Don't convert a float parameter when passed in memory.
2598 2004-03-09 Hans-Peter Nilsson <hp@axis.com>
2600 * configure: Regenerate for config/accross.m4 correction.
2602 2004-02-25 Matt Kraai <kraai@alumni.cmu.edu>
2604 * src/powerpc/ffi.c (ffi_prep_args_SYSV): Change
2605 ecif->cif->bytes to bytes.
2606 (ffi_prep_cif_machdep): Add braces around nested if statement.
2608 2004-02-09 Alan Modra <amodra@bigpond.net.au>
2610 * src/types.c (pointer): POWERPC64 has 8 byte pointers.
2612 * src/powerpc/ffi.c (ffi_prep_args64): Correct long double handling.
2613 (ffi_closure_helper_LINUX64): Fix typo.
2614 * testsuite/libffi.call/cls_align_longdouble.c: Pass -mlong-double-128
2616 * testsuite/libffi.call/float.c: Likewise.
2617 * testsuite/libffi.call/float2.c: Likewise.
2619 2004-02-08 Alan Modra <amodra@bigpond.net.au>
2621 * src/powerpc/ffi.c (ffi_prep_cif_machdep <FFI_LINUX64>): Correct
2622 long double function return and long double arg handling.
2623 (ffi_closure_helper_LINUX64): Formatting. Delete unused "ng" var.
2624 Use "end_pfr" instead of "nf". Correct long double handling.
2626 * src/powerpc/linux64.S (ffi_call_LINUX64): Save f2 long double
2628 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Allocate
2629 space for long double return value. Adjust stack frame and offsets.
2630 Load f2 long double return.
2632 2004-02-07 Alan Modra <amodra@bigpond.net.au>
2634 * src/types.c: Use 16 byte long double for POWERPC64.
2636 2004-01-25 Eric Botcazou <ebotcazou@libertysurf.fr>
2638 * src/sparc/ffi.c (ffi_prep_args_v9): Shift the parameter array
2639 when the structure return address is passed in %o0.
2640 (ffi_V9_return_struct): Rename into ffi_v9_layout_struct.
2641 (ffi_v9_layout_struct): Align the field following a nested structure
2642 on a word boundary. Use memmove instead of memcpy.
2643 (ffi_call): Update call to ffi_V9_return_struct.
2644 (ffi_prep_closure): Define 'ctx' only for V8.
2645 (ffi_closure_sparc_inner): Clone into ffi_closure_sparc_inner_v8
2646 and ffi_closure_sparc_inner_v9.
2647 (ffi_closure_sparc_inner_v8): Return long doubles by reference.
2648 Always skip the structure return address. For structures and long
2649 doubles, copy the argument directly.
2650 (ffi_closure_sparc_inner_v9): Skip the structure return address only
2651 if required. Shift the maximum floating-point slot accordingly. For
2652 big structures, copy the argument directly; otherwise, left-justify the
2653 argument and call ffi_v9_layout_struct to lay out the structure on
2655 * src/sparc/v8.S: Undef STACKFRAME before defining it.
2656 (ffi_closure_v8): Pass the structure return address. Update call to
2657 ffi_closure_sparc_inner_v8. Short-circuit FFI_TYPE_INT handling.
2658 Skip the 'unimp' insn when returning long doubles and structures.
2659 * src/sparc/v9.S: Undef STACKFRAME before defining it.
2660 (ffi_closure_v9): Increase the frame size by 2 words. Short-circuit
2661 FFI_TYPE_INT handling. Load structures both in integers and
2662 floating-point registers on return.
2663 * README: Update status of the SPARC port.
2665 2004-01-24 Andreas Tobler <a.tobler@schweiz.ch>
2667 * testsuite/libffi.call/pyobjc-tc.c (main): Treat result value
2669 * testsuite/libffi.call/struct3.c (main): Fix CHECK.
2671 2004-01-22 Ulrich Weigand <uweigand@de.ibm.com>
2673 * testsuite/libffi.call/cls_uint.c (cls_ret_uint_fn): Treat result
2674 value as of type ffi_arg, not unsigned int.
2676 2004-01-21 Michael Ritzert <ritzert@t-online.de>
2678 * ffi64.c (ffi_prep_args): Cast the RHS of an assignment instead
2681 2004-01-12 Andreas Tobler <a.tobler@schweiz.ch>
2683 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_32 for
2686 2004-01-08 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2688 * testsuite/libffi.call/ffitest.h (allocate_mmap): Cast MAP_FAILED
2691 2003-12-10 Richard Henderson <rth@redhat.com>
2693 * testsuite/libffi.call/cls_align_pointer.c: Cast pointers to
2694 size_t instead of int.
2696 2003-12-04 Hosaka Yuji <hos@tamanegi.org>
2698 * testsuite/libffi.call/many_win32.c: Include <float.h>.
2699 * testsuite/libffi.call/many_win32.c (main): Replace variable
2700 int i with unsigned long ul.
2702 * testsuite/libffi.call/cls_align_uint64.c: New test case.
2703 * testsuite/libffi.call/cls_align_sint64.c: Likewise.
2704 * testsuite/libffi.call/cls_align_uint32.c: Likewise.
2705 * testsuite/libffi.call/cls_align_sint32.c: Likewise.
2706 * testsuite/libffi.call/cls_align_uint16.c: Likewise.
2707 * testsuite/libffi.call/cls_align_sint16.c: Likewise.
2708 * testsuite/libffi.call/cls_align_float.c: Likewise.
2709 * testsuite/libffi.call/cls_align_double.c: Likewise.
2710 * testsuite/libffi.call/cls_align_longdouble.c: Likewise.
2711 * testsuite/libffi.call/cls_align_pointer.c: Likewise.
2713 2003-12-02 Hosaka Yuji <hos@tamanegi.org>
2716 * src/x86/ffi.c (ffi_prep_args, ffi_prep_incoming_args_SYSV):
2717 Align arguments to 32 bits.
2719 2003-12-01 Andreas Tobler <a.tobler@schweiz.ch>
2722 * testsuite/libffi.call/cls_multi_sshort.c: New test case.
2723 * testsuite/libffi.call/cls_multi_sshortchar.c: Likewise.
2724 * testsuite/libffi.call/cls_multi_uchar.c: Likewise.
2725 * testsuite/libffi.call/cls_multi_schar.c: Likewise.
2726 * testsuite/libffi.call/cls_multi_ushortchar.c: Likewise.
2727 * testsuite/libffi.call/cls_multi_ushort.c: Likewise.
2729 * testsuite/libffi.special/unwindtest.cc: Cosmetics.
2731 2003-11-26 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
2733 * testsuite/libffi.call/ffitest.h: Include <fcntl.h>.
2734 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2736 2003-11-22 Andreas Tobler <a.tobler@schweiz.ch>
2738 * Makefile.in: Rebuilt.
2739 * configure: Likewise.
2740 * testsuite/libffi.special/unwindtest.cc: Convert the mmap to
2743 2003-11-21 Andreas Jaeger <aj@suse.de>
2744 Andreas Tobler <a.tobler@schweiz.ch>
2746 * acinclude.m4: Add AC_FUNC_MMAP_BLACKLIST.
2747 * configure.in: Call AC_FUNC_MMAP_BLACKLIST.
2748 * Makefile.in: Rebuilt.
2749 * aclocal.m4: Likewise.
2750 * configure: Likewise.
2751 * fficonfig.h.in: Likewise.
2752 * testsuite/lib/libffi-dg.exp: Add include dir.
2753 * testsuite/libffi.call/ffitest.h: Add MMAP definitions.
2754 * testsuite/libffi.special/ffitestcxx.h: Likewise.
2755 * testsuite/libffi.call/closure_fn0.c: Use MMAP functionality
2756 for ffi_closure if available.
2757 * testsuite/libffi.call/closure_fn1.c: Likewise.
2758 * testsuite/libffi.call/closure_fn2.c: Likewise.
2759 * testsuite/libffi.call/closure_fn3.c: Likewise.
2760 * testsuite/libffi.call/closure_fn4.c: Likewise.
2761 * testsuite/libffi.call/closure_fn5.c: Likewise.
2762 * testsuite/libffi.call/cls_12byte.c: Likewise.
2763 * testsuite/libffi.call/cls_16byte.c: Likewise.
2764 * testsuite/libffi.call/cls_18byte.c: Likewise.
2765 * testsuite/libffi.call/cls_19byte.c: Likewise.
2766 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2767 * testsuite/libffi.call/cls_20byte.c: Likewise.
2768 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2769 * testsuite/libffi.call/cls_24byte.c: Likewise.
2770 * testsuite/libffi.call/cls_2byte.c: Likewise.
2771 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2772 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2773 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2774 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2775 * testsuite/libffi.call/cls_4byte.c: Likewise.
2776 * testsuite/libffi.call/cls_5byte.c: Likewise.
2777 * testsuite/libffi.call/cls_64byte.c: Likewise.
2778 * testsuite/libffi.call/cls_6byte.c: Likewise.
2779 * testsuite/libffi.call/cls_7byte.c: Likewise.
2780 * testsuite/libffi.call/cls_8byte.c: Likewise.
2781 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2782 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2783 * testsuite/libffi.call/cls_double.c: Likewise.
2784 * testsuite/libffi.call/cls_float.c: Likewise.
2785 * testsuite/libffi.call/cls_schar.c: Likewise.
2786 * testsuite/libffi.call/cls_sint.c: Likewise.
2787 * testsuite/libffi.call/cls_sshort.c: Likewise.
2788 * testsuite/libffi.call/cls_uchar.c: Likewise.
2789 * testsuite/libffi.call/cls_uint.c: Likewise.
2790 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2791 * testsuite/libffi.call/cls_ushort.c: Likewise.
2792 * testsuite/libffi.call/nested_struct.c: Likewise.
2793 * testsuite/libffi.call/nested_struct1.c: Likewise.
2794 * testsuite/libffi.call/nested_struct2.c: Likewise.
2795 * testsuite/libffi.call/nested_struct3.c: Likewise.
2796 * testsuite/libffi.call/problem1.c: Likewise.
2797 * testsuite/libffi.special/unwindtest.cc: Likewise.
2799 2003-11-20 Andreas Tobler <a.tobler@schweiz.ch>
2801 * testsuite/lib/libffi-dg.exp: Make the -lgcc_s conditional.
2803 2003-11-19 Andreas Tobler <a.tobler@schweiz.ch>
2805 * testsuite/lib/libffi-dg.exp: Add DYLD_LIBRARY_PATH for darwin.
2806 Add -lgcc_s to additional flags.
2808 2003-11-12 Andreas Tobler <a.tobler@schweiz.ch>
2810 * configure.in, include/Makefile.am: PR libgcj/11147, install
2811 the ffitarget.h header file in a gcc versioned and target
2813 * configure: Regenerated.
2814 * Makefile.in, include/Makefile.in: Likewise.
2815 * testsuite/Makefile.in: Likewise.
2817 2003-11-09 Andreas Tobler <a.tobler@schweiz.ch>
2819 * testsuite/libffi.call/closure_fn0.c: Print result and check
2820 with dg-output to make debugging easier.
2821 * testsuite/libffi.call/closure_fn1.c: Likewise.
2822 * testsuite/libffi.call/closure_fn2.c: Likewise.
2823 * testsuite/libffi.call/closure_fn3.c: Likewise.
2824 * testsuite/libffi.call/closure_fn4.c: Likewise.
2825 * testsuite/libffi.call/closure_fn5.c: Likewise.
2826 * testsuite/libffi.call/cls_12byte.c: Likewise.
2827 * testsuite/libffi.call/cls_16byte.c: Likewise.
2828 * testsuite/libffi.call/cls_18byte.c: Likewise.
2829 * testsuite/libffi.call/cls_19byte.c: Likewise.
2830 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
2831 * testsuite/libffi.call/cls_20byte.c: Likewise.
2832 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2833 * testsuite/libffi.call/cls_24byte.c: Likewise.
2834 * testsuite/libffi.call/cls_2byte.c: Likewise.
2835 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
2836 * testsuite/libffi.call/cls_3byte1.c: Likewise.
2837 * testsuite/libffi.call/cls_3byte2.c: Likewise.
2838 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
2839 * testsuite/libffi.call/cls_4byte.c: Likewise.
2840 * testsuite/libffi.call/cls_5byte.c: Likewise.
2841 * testsuite/libffi.call/cls_64byte.c: Likewise.
2842 * testsuite/libffi.call/cls_6byte.c: Likewise.
2843 * testsuite/libffi.call/cls_7byte.c: Likewise.
2844 * testsuite/libffi.call/cls_8byte.c: Likewise.
2845 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2846 * testsuite/libffi.call/cls_9byte2.c: Likewise.
2847 * testsuite/libffi.call/cls_double.c: Likewise.
2848 * testsuite/libffi.call/cls_float.c: Likewise.
2849 * testsuite/libffi.call/cls_schar.c: Likewise.
2850 * testsuite/libffi.call/cls_sint.c: Likewise.
2851 * testsuite/libffi.call/cls_sshort.c: Likewise.
2852 * testsuite/libffi.call/cls_uchar.c: Likewise.
2853 * testsuite/libffi.call/cls_uint.c: Likewise.
2854 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2855 * testsuite/libffi.call/cls_ushort.c: Likewise.
2856 * testsuite/libffi.call/problem1.c: Likewise.
2858 * testsuite/libffi.special/unwindtest.cc: Make ffi_closure
2861 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2863 * testsuite/libffi.call/cls_9byte2.c: New test case.
2864 * testsuite/libffi.call/cls_9byte1.c: Likewise.
2865 * testsuite/libffi.call/cls_64byte.c: Likewise.
2866 * testsuite/libffi.call/cls_20byte1.c: Likewise.
2867 * testsuite/libffi.call/cls_19byte.c: Likewise.
2868 * testsuite/libffi.call/cls_18byte.c: Likewise.
2869 * testsuite/libffi.call/closure_fn4.c: Likewise.
2870 * testsuite/libffi.call/closure_fn5.c: Likewise.
2871 * testsuite/libffi.call/cls_schar.c: Likewise.
2872 * testsuite/libffi.call/cls_sint.c: Likewise.
2873 * testsuite/libffi.call/cls_sshort.c: Likewise.
2874 * testsuite/libffi.call/nested_struct2.c: Likewise.
2875 * testsuite/libffi.call/nested_struct3.c: Likewise.
2877 2003-11-08 Andreas Tobler <a.tobler@schweiz.ch>
2879 * testsuite/libffi.call/cls_double.c: Do a check on the result.
2880 * testsuite/libffi.call/cls_uchar.c: Likewise.
2881 * testsuite/libffi.call/cls_uint.c: Likewise.
2882 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
2883 * testsuite/libffi.call/cls_ushort.c: Likewise.
2884 * testsuite/libffi.call/return_sc.c: Cleanup whitespaces.
2886 2003-11-06 Andreas Tobler <a.tobler@schweiz.ch>
2888 * src/prep_cif.c (ffi_prep_cif): Move the validity check after
2891 2003-10-23 Andreas Tobler <a.tobler@schweiz.ch>
2893 * src/java_raw_api.c (ffi_java_ptrarray_to_raw): Replace
2894 FFI_ASSERT(FALSE) with FFI_ASSERT(0).
2896 2003-10-22 David Daney <ddaney@avtrex.com>
2898 * src/mips/ffitarget.h: Replace undefined UINT32 and friends with
2899 __attribute__((__mode__(__SI__))) and friends.
2901 2003-10-22 Andreas Schwab <schwab@suse.de>
2903 * src/ia64/ffi.c: Replace FALSE/TRUE with false/true.
2905 2003-10-21 Andreas Tobler <a.tobler@schweiz.ch>
2907 * configure.in: AC_LINK_FILES(ffitarget.h).
2908 * configure: Regenerate.
2909 * Makefile.in: Likewise.
2910 * include/Makefile.in: Likewise.
2911 * testsuite/Makefile.in: Likewise.
2912 * fficonfig.h.in: Likewise.
2914 2003-10-21 Paolo Bonzini <bonzini@gnu.org>
2915 Richard Henderson <rth@redhat.com>
2917 Avoid that ffi.h includes fficonfig.h.
2919 * Makefile.am (EXTRA_DIST): Include ffitarget.h files
2920 (TARGET_SRC_MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2921 (TARGET_SRC_MIPS_SGI): Removed.
2922 (MIPS_GCC): Renamed to TARGET_SRC_MIPS_IRIX.
2923 (MIPS_SGI): Removed.
2924 (CLEANFILES): Removed.
2925 (mostlyclean-am, clean-am, mostlyclean-sub, clean-sub): New
2927 * acconfig.h: Removed.
2928 * configure.in: Compute sizeofs only for double and long double.
2929 Use them to define and subst HAVE_LONG_DOUBLE. Include comments
2930 into AC_DEFINE instead of using acconfig.h. Create
2931 include/ffitarget.h instead of include/fficonfig.h. Rename
2932 MIPS_GCC to MIPS_IRIX, drop MIPS_SGI since we are in gcc's tree.
2933 AC_DEFINE EH_FRAME_FLAGS.
2934 * include/Makefile.am (DISTCLEANFILES): New automake macro.
2935 (hack_DATA): Add ffitarget.h.
2936 * include/ffi.h.in: Remove all system specific definitions.
2937 Declare raw API even if it is not installed, why bother?
2938 Use limits.h instead of SIZEOF_* to define ffi_type_*. Do
2939 not define EH_FRAME_FLAGS, it is in fficonfig.h now. Include
2940 ffitarget.h instead of fficonfig.h. Remove ALIGN macro.
2941 (UINT_ARG, INT_ARG): Removed, use ffi_arg and ffi_sarg instead.
2942 * include/ffi_common.h (bool): Do not define.
2943 (ffi_assert): Accept failed assertion.
2944 (ffi_type_test): Return void and accept file/line.
2945 (FFI_ASSERT): Pass stringized failed assertion.
2946 (FFI_ASSERT_AT): New macro.
2947 (FFI_ASSERT_VALID_TYPE): New macro.
2948 (UINT8, SINT8, UINT16, SINT16, UINT32, SINT32,
2949 UINT64, SINT64): Define here with gcc's __attribute__ macro
2951 (FLOAT32, ALIGN): Define here instead of in ffi.h
2952 * include/ffi-mips.h: Removed. Its content moved to
2953 src/mips/ffitarget.h after separating assembly and C sections.
2954 * src/alpha/ffi.c, src/alpha/ffi.c, src/java_raw_api.c
2955 src/prep_cif.c, src/raw_api.c, src/ia64/ffi.c,
2956 src/mips/ffi.c, src/mips/n32.S, src/mips/o32.S,
2957 src/mips/ffitarget.h, src/sparc/ffi.c, src/x86/ffi64.c:
2958 SIZEOF_ARG -> FFI_SIZEOF_ARG.
2959 * src/ia64/ffi.c: Include stdbool.h (provided by GCC 2.95+).
2960 * src/debug.c (ffi_assert): Accept stringized failed assertion.
2961 (ffi_type_test): Rewritten.
2962 * src/prep-cif.c (initialize_aggregate, ffi_prep_cif): Call
2963 FFI_ASSERT_VALID_TYPE.
2964 * src/alpha/ffitarget.h, src/arm/ffitarget.h,
2965 src/ia64/ffitarget.h, src/m68k/ffitarget.h,
2966 src/mips/ffitarget.h, src/powerpc/ffitarget.h,
2967 src/s390/ffitarget.h, src/sh/ffitarget.h,
2968 src/sh64/ffitarget.h, src/sparc/ffitarget.h,
2969 src/x86/ffitarget.h: New files.
2970 * src/alpha/osf.S, src/arm/sysv.S, src/ia64/unix.S,
2971 src/m68k/sysv.S, src/mips/n32.S, src/mips/o32.S,
2972 src/powerpc/aix.S, src/powerpc/darwin.S,
2973 src/powerpc/ffi_darwin.c, src/powerpc/linux64.S,
2974 src/powerpc/linux64_closure.S, src/powerpc/ppc_closure.S,
2975 src/powerpc/sysv.S, src/s390/sysv.S, src/sh/sysv.S,
2976 src/sh64/sysv.S, src/sparc/v8.S, src/sparc/v9.S,
2977 src/x86/sysv.S, src/x86/unix64.S, src/x86/win32.S:
2980 2003-10-20 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2982 * src/mips/ffi.c: Use _ABIN32, _ABIO32 instead of external
2983 _MIPS_SIM_NABI32, _MIPS_SIM_ABI32.
2985 2003-10-19 Andreas Tobler <a.tobler@schweiz.ch>
2987 * src/powerpc/ffi_darwin.c (ffi_prep_args): Declare bytes again.
2988 Used when FFI_DEBUG = 1.
2990 2003-10-14 Alan Modra <amodra@bigpond.net.au>
2992 * src/types.c (double, longdouble): Default POWERPC64 to 8 byte size
2995 2003-10-06 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
2997 * include/ffi_mips.h: Define FFI_MIPS_N32 for N32/N64 ABIs,
2998 FFI_MIPS_O32 for O32 ABI.
3000 2003-10-01 Andreas Tobler <a.tobler@schweiz.ch>
3002 * testsuite/lib/libffi-dg.exp: Set LD_LIBRARY_PATH_64 for
3003 SPARC64. Cleanup whitespaces.
3005 2003-09-19 Andreas Tobler <a.tobler@schweiz.ch>
3007 * testsuite/libffi.call/closure_fn0.c: Xfail mips, arm,
3008 strongarm, xscale. Cleanup whitespaces.
3009 * testsuite/libffi.call/closure_fn1.c: Likewise.
3010 * testsuite/libffi.call/closure_fn2.c: Likewise.
3011 * testsuite/libffi.call/closure_fn3.c: Likewise.
3012 * testsuite/libffi.call/cls_12byte.c: Likewise.
3013 * testsuite/libffi.call/cls_16byte.c: Likewise.
3014 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3015 * testsuite/libffi.call/cls_20byte.c: Likewise.
3016 * testsuite/libffi.call/cls_24byte.c: Likewise.
3017 * testsuite/libffi.call/cls_2byte.c: Likewise.
3018 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3019 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3020 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3021 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3022 * testsuite/libffi.call/cls_4byte.c: Likewise.
3023 * testsuite/libffi.call/cls_5byte.c: Likewise.
3024 * testsuite/libffi.call/cls_6byte.c: Likewise.
3025 * testsuite/libffi.call/cls_7byte.c: Likewise.
3026 * testsuite/libffi.call/cls_8byte.c: Likewise.
3027 * testsuite/libffi.call/cls_double.c: Likewise.
3028 * testsuite/libffi.call/cls_float.c: Likewise.
3029 * testsuite/libffi.call/cls_uchar.c: Likewise.
3030 * testsuite/libffi.call/cls_uint.c: Likewise.
3031 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3032 * testsuite/libffi.call/cls_ushort.c: Likewise.
3033 * testsuite/libffi.call/nested_struct.c: Likewise.
3034 * testsuite/libffi.call/nested_struct1.c: Likewise.
3035 * testsuite/libffi.call/problem1.c: Likewise.
3036 * testsuite/libffi.special/unwindtest.cc: Likewise.
3037 * testsuite/libffi.call/pyobjc-tc.c: Cleanup whitespaces.
3039 2003-09-18 David Edelsohn <edelsohn@gnu.org>
3041 * src/powerpc/aix.S: Cleanup whitespaces.
3042 * src/powerpc/aix_closure.S: Likewise.
3044 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3046 * src/powerpc/darwin.S: Cleanup whitespaces, comment formatting.
3047 * src/powerpc/darwin_closure.S: Likewise.
3048 * src/powerpc/ffi_darwin.c: Likewise.
3050 2003-09-18 Andreas Tobler <a.tobler@schweiz.ch>
3051 David Edelsohn <edelsohn@gnu.org>
3053 * src/types.c (double): Add AIX and Darwin to the right TYPEDEF.
3054 * src/powerpc/aix_closure.S: Remove the pointer to the outgoing
3056 * src/powerpc/darwin_closure.S: Likewise.
3057 * src/powerpc/ffi_darwin.c (ffi_prep_args): Handle structures
3058 according to the Darwin/AIX ABI.
3059 (ffi_prep_cif_machdep): Likewise.
3060 (ffi_closure_helper_DARWIN): Likewise.
3061 Remove the outgoing parameter stack logic. Simplify the evaluation
3062 of the different CASE types.
3063 (ffi_prep_clousure): Avoid the casts on lvalues. Change the branch
3064 statement in the trampoline code.
3066 2003-09-18 Kaz Kojima <kkojima@gcc.gnu.org>
3068 * src/sh/ffi.c (ffi_prep_args): Take account into the alignement
3069 for the register size.
3070 (ffi_closure_helper_SYSV): Handle the structure return value
3072 (ffi_closure_helper_SYSV): Return the appropriate type when
3073 the registers are used for the structure return value.
3074 * src/sh/sysv.S (ffi_closure_SYSV): Fix the stack layout for
3075 the 64-bit return value. Update copyright years.
3077 2003-09-17 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3079 * testsuite/lib/libffi-dg.exp (libffi_target_compile): Search in
3080 srcdir for ffi_mips.h.
3082 2003-09-12 Alan Modra <amodra@bigpond.net.au>
3084 * src/prep_cif.c (initialize_aggregate): Include tail padding in
3086 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Correct
3087 placement of float result.
3088 * testsuite/libffi.special/unwindtest.cc (closure_test_fn1): Correct
3089 cast of "resp" for big-endian 64 bit machines.
3091 2003-09-11 Alan Modra <amodra@bigpond.net.au>
3093 * src/types.c (double, longdouble): Merge identical SH and ARM
3094 typedefs, and add POWERPC64.
3095 * src/powerpc/ffi.c (ffi_prep_args64): Correct next_arg calc for
3096 struct split over gpr and rest.
3097 (ffi_prep_cif_machdep): Correct intarg_count for structures.
3098 * src/powerpc/linux64.S (ffi_call_LINUX64): Fix gpr offsets.
3100 2003-09-09 Andreas Tobler <a.tobler@schweiz.ch>
3102 * src/powerpc/ffi.c (ffi_closure_helper_SYSV) Handle struct
3105 2003-09-09 Alan Modra <amodra@bigpond.net.au>
3107 * configure: Regenerate.
3109 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3111 * Makefile.am: Remove build rules for ffitest.
3112 * Makefile.in: Rebuilt.
3114 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3116 * src/java_raw_api.c: Include <stdlib.h> to fix compiler warning
3117 about implicit declaration of abort().
3119 2003-09-04 Andreas Tobler <a.tobler@schweiz.ch>
3121 * Makefile.am: Add dejagnu test framework. Fixes PR other/11411.
3122 * Makefile.in: Rebuilt.
3123 * configure.in: Add dejagnu test framework.
3124 * configure: Rebuilt.
3126 * testsuite/Makefile.am: New file.
3127 * testsuite/Makefile.in: Built
3128 * testsuite/lib/libffi-dg.exp: New file.
3129 * testsuite/config/default.exp: Likewise.
3130 * testsuite/libffi.call/call.exp: Likewise.
3131 * testsuite/libffi.call/ffitest.h: Likewise.
3132 * testsuite/libffi.call/closure_fn0.c: Likewise.
3133 * testsuite/libffi.call/closure_fn1.c: Likewise.
3134 * testsuite/libffi.call/closure_fn2.c: Likewise.
3135 * testsuite/libffi.call/closure_fn3.c: Likewise.
3136 * testsuite/libffi.call/cls_1_1byte.c: Likewise.
3137 * testsuite/libffi.call/cls_3_1byte.c: Likewise.
3138 * testsuite/libffi.call/cls_4_1byte.c: Likewise.
3139 * testsuite/libffi.call/cls_2byte.c: Likewise.
3140 * testsuite/libffi.call/cls_3byte1.c: Likewise.
3141 * testsuite/libffi.call/cls_3byte2.c: Likewise.
3142 * testsuite/libffi.call/cls_4byte.c: Likewise.
3143 * testsuite/libffi.call/cls_5byte.c: Likewise.
3144 * testsuite/libffi.call/cls_6byte.c: Likewise.
3145 * testsuite/libffi.call/cls_7byte.c: Likewise.
3146 * testsuite/libffi.call/cls_8byte.c: Likewise.
3147 * testsuite/libffi.call/cls_12byte.c: Likewise.
3148 * testsuite/libffi.call/cls_16byte.c: Likewise.
3149 * testsuite/libffi.call/cls_20byte.c: Likewise.
3150 * testsuite/libffi.call/cls_24byte.c: Likewise.
3151 * testsuite/libffi.call/cls_double.c: Likewise.
3152 * testsuite/libffi.call/cls_float.c: Likewise.
3153 * testsuite/libffi.call/cls_uchar.c: Likewise.
3154 * testsuite/libffi.call/cls_uint.c: Likewise.
3155 * testsuite/libffi.call/cls_ulonglong.c: Likewise.
3156 * testsuite/libffi.call/cls_ushort.c: Likewise.
3157 * testsuite/libffi.call/float.c: Likewise.
3158 * testsuite/libffi.call/float1.c: Likewise.
3159 * testsuite/libffi.call/float2.c: Likewise.
3160 * testsuite/libffi.call/many.c: Likewise.
3161 * testsuite/libffi.call/many_win32.c: Likewise.
3162 * testsuite/libffi.call/nested_struct.c: Likewise.
3163 * testsuite/libffi.call/nested_struct1.c: Likewise.
3164 * testsuite/libffi.call/pyobjc-tc.c: Likewise.
3165 * testsuite/libffi.call/problem1.c: Likewise.
3166 * testsuite/libffi.call/promotion.c: Likewise.
3167 * testsuite/libffi.call/return_ll.c: Likewise.
3168 * testsuite/libffi.call/return_sc.c: Likewise.
3169 * testsuite/libffi.call/return_uc.c: Likewise.
3170 * testsuite/libffi.call/strlen.c: Likewise.
3171 * testsuite/libffi.call/strlen_win32.c: Likewise.
3172 * testsuite/libffi.call/struct1.c: Likewise.
3173 * testsuite/libffi.call/struct2.c: Likewise.
3174 * testsuite/libffi.call/struct3.c: Likewise.
3175 * testsuite/libffi.call/struct4.c: Likewise.
3176 * testsuite/libffi.call/struct5.c: Likewise.
3177 * testsuite/libffi.call/struct6.c: Likewise.
3178 * testsuite/libffi.call/struct7.c: Likewise.
3179 * testsuite/libffi.call/struct8.c: Likewise.
3180 * testsuite/libffi.call/struct9.c: Likewise.
3181 * testsuite/libffi.special/special.exp: New file.
3182 * testsuite/libffi.special/ffitestcxx.h: Likewise.
3183 * testsuite/libffi.special/unwindtest.cc: Likewise.
3186 2003-08-13 Kaz Kojima <kkojima@gcc.gnu.org>
3188 * src/sh/ffi.c (OFS_INT16): Set 0 for little endian case. Update
3191 2003-08-02 Alan Modra <amodra@bigpond.net.au>
3193 * src/powerpc/ffi.c (ffi_prep_args64): Modify for changed gcc
3195 (ffi_closure_helper_LINUX64): Likewise.
3196 * src/powerpc/linux64.S: Remove code writing to parm save area.
3197 * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Use return
3198 address in lr from ffi_closure_helper_LINUX64 call to calculate
3199 table address. Optimize function tail.
3201 2003-07-28 Andreas Tobler <a.tobler@schweiz.ch>
3203 * src/sparc/ffi.c: Handle all floating point registers.
3204 * src/sparc/v9.S: Likewise. Fixes second part of PR target/11410.
3206 2003-07-11 Gerald Pfeifer <pfeifer@dbai.tuwien.ac.at>
3208 * README: Note that libffi is not part of GCC. Update the project
3211 2003-06-19 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3213 * src/powerpc/ppc_closure.S: Include ffi.h.
3215 2003-06-13 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3217 * src/x86/sysv.S: Avoid gas-only .uleb128/.sleb128 directives.
3218 Use C style comments.
3220 2003-06-13 Kaz Kojima <kkojima@rr.iij4u.or.jp>
3222 * Makefile.am: Add SHmedia support. Fix a typo of SH support.
3223 * Makefile.in: Regenerate.
3224 * configure.in (sh64-*-linux*, sh5*-*-linux*): Add target.
3225 * configure: Regenerate.
3226 * include/ffi.h.in: Add SHmedia support.
3227 * src/sh64/ffi.c: New file.
3228 * src/sh64/sysv.S: New file.
3230 2003-05-16 Jakub Jelinek <jakub@redhat.com>
3232 * configure.in (HAVE_RO_EH_FRAME): Check whether .eh_frame section
3233 should be read-only.
3234 * configure: Rebuilt.
3235 * fficonfig.h.in: Rebuilt.
3236 * include/ffi.h.in (EH_FRAME_FLAGS): Define.
3237 * src/alpha/osf.S: Use EH_FRAME_FLAGS.
3238 * src/powerpc/linux64.S: Likewise.
3239 * src/powerpc/linux64_closure.S: Likewise. Include ffi.h.
3240 * src/powerpc/sysv.S: Use EH_FRAME_FLAGS. Use pcrel encoding
3241 if -fpic/-fPIC/-mrelocatable.
3242 * src/powerpc/powerpc_closure.S: Likewise.
3243 * src/sparc/v8.S: If HAVE_RO_EH_FRAME is defined, don't include
3244 #write in .eh_frame flags.
3245 * src/sparc/v9.S: Likewise.
3246 * src/x86/unix64.S: Use EH_FRAME_FLAGS.
3247 * src/x86/sysv.S: Likewise. Use pcrel encoding if -fpic/-fPIC.
3248 * src/s390/sysv.S: Use EH_FRAME_FLAGS. Include ffi.h.
3250 2003-05-07 Jeff Sturm <jsturm@one-point.com>
3252 Fixes PR bootstrap/10656
3253 * configure.in (HAVE_AS_REGISTER_PSEUDO_OP): Test assembler
3254 support for .register pseudo-op.
3255 * src/sparc/v8.S: Use it.
3256 * fficonfig.h.in: Rebuilt.
3257 * configure: Rebuilt.
3259 2003-04-18 Jakub Jelinek <jakub@redhat.com>
3261 * include/ffi.h.in (POWERPC64): Define if 64-bit.
3262 (enum ffi_abi): Add FFI_LINUX64 on POWERPC.
3263 Make it the default on POWERPC64.
3264 (FFI_TRAMPOLINE_SIZE): Define to 24 on POWERPC64.
3265 * configure.in: Change powerpc-*-linux* into powerpc*-*-linux*.
3266 * configure: Rebuilt.
3267 * src/powerpc/ffi.c (hidden): Define.
3268 (ffi_prep_args_SYSV): Renamed from
3269 ffi_prep_args. Cast pointers to unsigned long to shut up warnings.
3270 (NUM_GPR_ARG_REGISTERS64, NUM_FPR_ARG_REGISTERS64,
3271 ASM_NEEDS_REGISTERS64): New.
3272 (ffi_prep_args64): New function.
3273 (ffi_prep_cif_machdep): Handle FFI_LINUX64 ABI.
3274 (ffi_call): Likewise.
3275 (ffi_prep_closure): Likewise.
3276 (flush_icache): Surround by #ifndef POWERPC64.
3277 (ffi_dblfl): New union type.
3278 (ffi_closure_helper_SYSV): Use it to avoid aliasing problems.
3279 (ffi_closure_helper_LINUX64): New function.
3280 * src/powerpc/ppc_closure.S: Surround whole file by #ifndef
3282 * src/powerpc/sysv.S: Likewise.
3283 (ffi_call_SYSV): Rename ffi_prep_args to ffi_prep_args_SYSV.
3284 * src/powerpc/linux64.S: New file.
3285 * src/powerpc/linux64_closure.S: New file.
3286 * Makefile.am (EXTRA_DIST): Add src/powerpc/linux64.S and
3287 src/powerpc/linux64_closure.S.
3288 (TARGET_SRC_POWERPC): Likewise.
3290 * src/ffitest.c (closure_test_fn, closure_test_fn1, closure_test_fn2,
3291 closure_test_fn3): Fix result printing on big-endian 64-bit
3293 (main): Print tst2_arg instead of uninitialized tst2_result.
3295 * src/ffitest.c (main): Hide what closure pointer really points to
3298 2003-04-16 Richard Earnshaw <rearnsha@arm.com>
3300 * configure.in (arm-*-netbsdelf*): Add configuration.
3301 (configure): Regenerated.
3303 2003-04-04 Loren J. Rittle <ljrittle@acm.org>
3305 * include/Makefile.in: Regenerate.
3307 2003-03-21 Zdenek Dvorak <rakdver@atrey.karlin.mff.cuni.cz>
3309 * libffi/include/ffi.h.in: Define X86 instead of X86_64 in 32
3311 * libffi/src/x86/ffi.c (ffi_closure_SYSV, ffi_closure_raw_SYSV):
3312 Receive closure pointer through parameter, read args using
3313 __builtin_dwarf_cfa.
3314 (FFI_INIT_TRAMPOLINE): Send closure reference through eax.
3316 2003-03-12 Andreas Schwab <schwab@suse.de>
3318 * configure.in: Avoid trailing /. in toolexeclibdir.
3319 * configure: Rebuilt.
3321 2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
3323 * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
3325 2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
3327 * libffi/src/powerpc/darwin_closure.S:
3328 Fix alignement bug, allocate 8 bytes for the result.
3329 * libffi/src/powerpc/aix_closure.S:
3331 * libffi/src/powerpc/ffi_darwin.c:
3332 Update stackframe description for aix/darwin_closure.S.
3334 2003-02-06 Jakub Jelinek <jakub@redhat.com>
3336 * src/s390/ffi.c (ffi_closure_helper_SYSV): Add hidden visibility
3339 2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
3340 Andreas Schwab <schwab@suse.de>
3342 * configure.in: Adjust command to source config-ml.in to account
3343 for changes to the libffi_basedir definition.
3344 (libffi_basedir): Remove ${srcdir} from value and include trailing
3347 * configure: Regenerate.
3349 2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
3351 * src/powerpc/ppc_closure.S: Recode to fit shared libs.
3353 2003-01-28 Andrew Haley <aph@redhat.com>
3355 * include/ffi.h.in: Enable FFI_CLOSURES for x86_64.
3356 * src/x86/ffi64.c (ffi_prep_closure): New.
3357 (ffi_closure_UNIX64_inner): New.
3358 * src/x86/unix64.S (ffi_closure_UNIX64): New.
3360 2003-01-27 Alexandre Oliva <aoliva@redhat.com>
3362 * configure.in (toolexecdir, toolexeclibdir): Set and AC_SUBST.
3363 Remove USE_LIBDIR conditional.
3364 * Makefile.am (toolexecdir, toolexeclibdir): Don't override.
3365 * Makefile.in, configure: Rebuilt.
3367 2003-01027 David Edelsohn <edelsohn@gnu.org>
3369 * Makefile.am (TARGET_SRC_POWERPC_AIX): Fix typo.
3370 * Makefile.in: Regenerate.
3372 2003-01-22 Andrew Haley <aph@redhat.com>
3374 * src/powerpc/darwin.S (_ffi_call_AIX): Add Augmentation size to
3377 2003-01-21 Andreas Tobler <a.tobler@schweiz.ch>
3379 * src/powerpc/darwin.S: Add unwind info.
3380 * src/powerpc/darwin_closure.S: Likewise.
3382 2003-01-14 Andrew Haley <aph@redhat.com>
3384 * src/x86/ffi64.c (ffi_prep_args): Check for void retval.
3385 (ffi_prep_cif_machdep): Likewise.
3386 * src/x86/unix64.S: Add unwind info.
3388 2003-01-14 Andreas Jaeger <aj@suse.de>
3390 * src/ffitest.c (main): Only use ffi_closures if those are
3393 2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
3395 * libffi/src/ffitest.c
3396 add closure testcases
3398 2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
3400 * libffi/src/powerpc/ffi.c
3401 fix alignment bug for float (4 byte aligned iso 8 byte)
3403 2003-01-09 Geoffrey Keating <geoffk@apple.com>
3405 * src/powerpc/ffi_darwin.c: Remove RCS version string.
3406 * src/powerpc/darwin.S: Remove RCS version string.
3408 2003-01-03 Jeff Sturm <jsturm@one-point.com>
3410 * include/ffi.h.in: Add closure defines for SPARC, SPARC64.
3411 * src/ffitest.c (main): Use static storage for closure.
3412 * src/sparc/ffi.c (ffi_prep_closure, ffi_closure_sparc_inner): New.
3413 * src/sparc/v8.S (ffi_closure_v8): New.
3414 * src/sparc/v9.S (ffi_closure_v9): New.
3416 2002-11-10 Ranjit Mathew <rmathew@hotmail.com>
3418 * include/ffi.h.in: Added FFI_STDCALL ffi_type
3419 enumeration for X86_WIN32.
3420 * src/x86/win32.S: Added ffi_call_STDCALL function
3422 * src/x86/ffi.c (ffi_call/ffi_raw_call): Added
3423 switch cases for recognising FFI_STDCALL and
3424 calling ffi_call_STDCALL if target is X86_WIN32.
3425 * src/ffitest.c (my_stdcall_strlen/stdcall_many):
3426 stdcall versions of the "my_strlen" and "many"
3427 test functions (for X86_WIN32).
3428 Added test cases to test stdcall invocation using
3431 2002-12-02 Kaz Kojima <kkojima@gcc.gnu.org>
3433 * src/sh/sysv.S: Add DWARF2 unwind info.
3435 2002-11-27 Ulrich Weigand <uweigand@de.ibm.com>
3437 * src/s390/sysv.S (.eh_frame section): Make section read-only.
3439 2002-11-26 Jim Wilson <wilson@redhat.com>
3441 * src/types.c (FFI_TYPE_POINTER): Has size 8 on IA64.
3443 2002-11-23 H.J. Lu <hjl@gnu.org>
3445 * acinclude.m4: Add dummy AM_PROG_LIBTOOL.
3446 Include ../config/accross.m4.
3447 * aclocal.m4; Rebuild.
3448 * configure: Likewise.
3450 2002-11-15 Ulrich Weigand <uweigand@de.ibm.com>
3452 * src/s390/sysv.S (.eh_frame section): Adapt to pcrel FDE encoding.
3454 2002-11-11 DJ Delorie <dj@redhat.com>
3456 * configure.in: Look for common files in the right place.
3458 2002-10-08 Ulrich Weigand <uweigand@de.ibm.com>
3460 * src/java_raw_api.c (ffi_java_raw_to_ptrarray): Interpret
3461 raw data as _Jv_word values, not ffi_raw.
3462 (ffi_java_ptrarray_to_raw): Likewise.
3463 (ffi_java_rvalue_to_raw): New function.
3464 (ffi_java_raw_call): Call it.
3465 (ffi_java_raw_to_rvalue): New function.
3466 (ffi_java_translate_args): Call it.
3467 * src/ffitest.c (closure_test_fn): Interpret return value
3468 as ffi_arg, not int.
3469 * src/s390/ffi.c (ffi_prep_cif_machdep): Add missing
3470 FFI_TYPE_POINTER case.
3471 (ffi_closure_helper_SYSV): Likewise. Also, assume return
3472 values extended to word size.
3474 2002-10-02 Andreas Jaeger <aj@suse.de>
3476 * src/x86/ffi64.c (ffi_prep_cif_machdep): Remove debug output.
3478 2002-10-01 Bo Thorsen <bo@smetana.suse.de>
3480 * include/ffi.h.in: Fix i386 win32 compilation.
3482 2002-09-30 Ulrich Weigand <uweigand@de.ibm.com>
3484 * configure.in: Add s390x-*-linux-* target.
3485 * configure: Regenerate.
3486 * include/ffi.h.in: Define S390X for s390x targets.
3487 (FFI_CLOSURES): Define for s390/s390x.
3488 (FFI_TRAMPOLINE_SIZE): Likewise.
3489 (FFI_NATIVE_RAW_API): Likewise.
3490 * src/prep_cif.c (ffi_prep_cif): Do not compute stack space for s390.
3491 * src/types.c (FFI_TYPE_POINTER): Use 8-byte pointers on s390x.
3492 * src/s390/ffi.c: Major rework of existing code. Add support for
3493 s390x targets. Add closure support.
3494 * src/s390/sysv.S: Likewise.
3496 2002-09-29 Richard Earnshaw <rearnsha@arm.com>
3498 * src/arm/sysv.S: Fix typo.
3500 2002-09-28 Richard Earnshaw <rearnsha@arm.com>
3502 * src/arm/sysv.S: If we don't have machine/asm.h and the pre-processor
3503 has defined __USER_LABEL_PREFIX__, then use it in CNAME.
3504 (ffi_call_SYSV): Handle soft-float.
3506 2002-09-27 Bo Thorsen <bo@suse.de>
3508 * include/ffi.h.in: Fix multilib x86-64 support.
3510 2002-09-22 Kaveh R. Ghazi <ghazi@caip.rutgers.edu>
3512 * Makefile.am (all-multi): Fix multilib parallel build.
3514 2002-07-19 Kaz Kojima <kkojima@gcc.gnu.org>
3516 * configure.in (sh[34]*-*-linux*): Add brackets.
3517 * configure: Regenerate.
3519 2002-07-18 Kaz Kojima <kkojima@gcc.gnu.org>
3521 * Makefile.am: Add SH support.
3522 * Makefile.in: Regenerate.
3523 * configure.in (sh-*-linux*, sh[34]*-*-linux*): Add target.
3524 * configure: Regenerate.
3525 * include/ffi.h.in: Add SH support.
3526 * src/sh/ffi.c: New file.
3527 * src/sh/sysv.S: New file.
3528 * src/types.c: Add SH support.
3530 2002-07-16 Bo Thorsen <bo@suse.de>
3532 * src/x86/ffi64.c: New file that adds x86-64 support.
3533 * src/x86/unix64.S: New file that handles argument setup for
3535 * src/x86/sysv.S: Don't use this on x86-64.
3536 * src/x86/ffi.c: Don't use this on x86-64.
3538 * src/prep_cif.c (ffi_prep_cif): Don't do stack size calculation
3540 * src/ffitest.c (struct6): New test that tests a special case in
3542 (struct7): Likewise.
3543 (struct8): Likewise.
3544 (struct9): Likewise.
3545 (closure_test_fn): Silence warning about this when it's not used.
3546 (main): Add the new tests.
3547 (main): Fix a couple of wrong casts and silence some compiler warnings.
3548 * include/ffi.h.in: Add x86-64 ABI definition.
3549 * fficonfig.h.in: Regenerate.
3550 * Makefile.am: Add x86-64 support.
3551 * configure.in: Likewise.
3552 * Makefile.in: Regenerate.
3553 * configure: Likewise.
3555 2002-06-24 Bo Thorsen <bo@suse.de>
3557 * src/types.c: Merge settings for similar architectures.
3558 Add x86-64 sizes and alignments.
3560 2002-06-23 Bo Thorsen <bo@suse.de>
3562 * src/arm/ffi.c (ffi_prep_args): Remove unused vars.
3563 * src/sparc/ffi.c (ffi_prep_args_v8): Likewise.
3564 * src/mips/ffi.c (ffi_prep_args): Likewise.
3565 * src/m68k/ffi.c (ffi_prep_args): Likewise.
3567 2002-07-18 H.J. Lu (hjl@gnu.org)
3569 * Makefile.am (TARGET_SRC_MIPS_LINUX): New.
3570 (libffi_la_SOURCES): Support MIPS_LINUX.
3571 (libffi_convenience_la_SOURCES): Likewise.
3572 * Makefile.in: Regenerated.
3574 * configure.in (mips64*-*): Skip.
3575 (mips*-*-linux*): New.
3576 * configure: Regenerated.
3578 * src/mips/ffi.c: Include <sgidefs.h>.
3580 2002-06-06 Ulrich Weigand <uweigand@de.ibm.com>
3582 * src/s390/sysv.S: Save/restore %r6. Add DWARF-2 unwind info.
3584 2002-05-27 Roger Sayle <roger@eyesopen.com>
3586 * src/x86/ffi.c (ffi_prep_args): Remove reference to avn.
3588 2002-05-27 Bo Thorsen <bo@suse.de>
3590 * src/x86/ffi.c (ffi_prep_args): Remove unused variable and
3593 2002-05-13 Andreas Tobler <a.tobler@schweiz.ch>
3595 * src/powerpc/ffi_darwin.c (ffi_prep_closure): Declare fd at
3596 beginning of function (for older apple cc).
3598 2002-05-08 Alexandre Oliva <aoliva@redhat.com>
3600 * configure.in (ORIGINAL_LD_FOR_MULTILIBS): Preserve LD at
3601 script entry, and set LD to it when configuring multilibs.
3602 * configure: Rebuilt.
3604 2002-05-05 Jason Thorpe <thorpej@wasabisystems.com>
3606 * configure.in (sparc64-*-netbsd*): Add target.
3607 (sparc-*-netbsdelf*): Likewise.
3608 * configure: Regenerate.
3610 2002-04-28 David S. Miller <davem@redhat.com>
3612 * configure.in, configure: Fix SPARC test in previous change.
3614 2002-04-29 Gerhard Tonn <GerhardTonn@swol.de>
3616 * Makefile.am: Add Linux for S/390 support.
3617 * Makefile.in: Regenerate.
3618 * configure.in: Add Linux for S/390 support.
3619 * configure: Regenerate.
3620 * include/ffi.h.in: Add Linux for S/390 support.
3621 * src/s390/ffi.c: New file from libffi CVS tree.
3622 * src/s390/sysv.S: New file from libffi CVS tree.
3624 2002-04-28 Jakub Jelinek <jakub@redhat.com>
3626 * configure.in (HAVE_AS_SPARC_UA_PCREL): Check for working
3628 * src/sparc/v8.S: Use it.
3629 * src/sparc/v9.S: Likewise.
3630 * fficonfig.h.in: Rebuilt.
3631 * configure: Rebuilt.
3633 2002-04-08 Hans Boehm <Hans_Boehm@hp.com>
3635 * src/java_raw_api.c (ffi_java_raw_size): Handle FFI_TYPE_DOUBLE
3637 * src/ia64/unix.S: Add unwind information. Fix comments.
3638 Save sp in a way that's compatible with unwind info.
3639 (ffi_call_unix): Correctly restore sp in all cases.
3640 * src/ia64/ffi.c: Add, fix comments.
3642 2002-04-08 Jakub Jelinek <jakub@redhat.com>
3644 * src/sparc/v8.S: Make .eh_frame dependent on target word size.
3646 2002-04-06 Jason Thorpe <thorpej@wasabisystems.com>
3648 * configure.in (alpha*-*-netbsd*): Add target.
3649 * configure: Regenerate.
3651 2002-04-04 Jeff Sturm <jsturm@one-point.com>
3653 * src/sparc/v8.S: Add unwind info.
3654 * src/sparc/v9.S: Likewise.
3656 2002-03-30 Krister Walfridsson <cato@df.lth.se>
3658 * configure.in: Enable i*86-*-netbsdelf*.
3659 * configure: Rebuilt.
3661 2002-03-29 David Billinghurst <David.Billinghurst@riotinto.com>
3664 * src/mips/n32.s: Delete
3665 * src/mips/o32.s: Delete
3667 2002-03-21 Loren J. Rittle <ljrittle@acm.org>
3669 * configure.in: Enable alpha*-*-freebsd*.
3670 * configure: Rebuilt.
3672 2002-03-17 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3674 * Makefile.am: libfficonvenience -> libffi_convenience.
3675 * Makefile.in: Rebuilt.
3677 * Makefile.am: Define ffitest_OBJECTS.
3678 * Makefile.in: Rebuilt.
3680 2002-03-07 Andreas Tobler <toa@pop.agri.ch>
3681 David Edelsohn <edelsohn@gnu.org>
3683 * Makefile.am (EXTRA_DIST): Add Darwin and AIX closure files.
3684 (TARGET_SRC_POWERPC_AIX): Add aix_closure.S.
3685 (TARGET_SRC_POWERPC_DARWIN): Add darwin_closure.S.
3686 * Makefile.in: Regenerate.
3687 * include/ffi.h.in: Add AIX and Darwin closure definitions.
3688 * src/powerpc/ffi_darwin.c (ffi_prep_closure): New function.
3689 (flush_icache, flush_range): New functions.
3690 (ffi_closure_helper_DARWIN): New function.
3691 * src/powerpc/aix_closure.S: New file.
3692 * src/powerpc/darwin_closure.S: New file.
3694 2002-02-24 Jeff Sturm <jsturm@one-point.com>
3696 * include/ffi.h.in: Add typedef for ffi_arg.
3697 * src/ffitest.c (main): Declare rint with ffi_arg.
3699 2002-02-21 Andreas Tobler <toa@pop.agri.ch>
3701 * src/powerpc/ffi_darwin.c (ffi_prep_args): Skip appropriate
3702 number of GPRs for floating-point arguments.
3704 2002-01-31 Anthony Green <green@redhat.com>
3706 * configure: Rebuilt.
3707 * configure.in: Replace CHECK_SIZEOF and endian tests with
3708 cross-compiler friendly macros.
3709 * aclocal.m4 (AC_COMPILE_CHECK_SIZEOF, AC_C_BIGENDIAN_CROSS): New
3712 2002-01-18 David Edelsohn <edelsohn@gnu.org>
3714 * src/powerpc/darwin.S (_ffi_call_AIX): New.
3715 * src/powerpc/aix.S (ffi_call_DARWIN): New.
3717 2002-01-17 David Edelsohn <edelsohn@gnu.org>
3719 * Makefile.am (EXTRA_DIST): Add Darwin and AIX files.
3720 (TARGET_SRC_POWERPC_AIX): New.
3721 (POWERPC_AIX): New stanza.
3722 * Makefile.in: Regenerate.
3723 * configure.in: Add AIX case.
3724 * configure: Regenerate.
3725 * include/ffi.h.in (ffi_abi): Add FFI_AIX.
3726 * src/powerpc/ffi_darwin.c (ffi_status): Use "long" to scale frame
3727 size. Fix "long double" support.
3728 (ffi_call): Add FFI_AIX case.
3729 * src/powerpc/aix.S: New.
3731 2001-10-09 John Hornkvist <john@toastedmarshmallow.com>
3733 Implement Darwin PowerPC ABI.
3734 * configure.in: Handle powerpc-*-darwin*.
3735 * Makefile.am: Set source files for POWERPC_DARWIN.
3736 * configure: Rebuilt.
3737 * Makefile.in: Rebuilt.
3738 * include/ffi.h.in: Define FFI_DARWIN and FFI_DEFAULT_ABI for
3740 * src/powerpc/darwin.S: New file.
3741 * src/powerpc/ffi_darwin.c: New file.
3743 2001-10-07 Joseph S. Myers <jsm28@cam.ac.uk>
3745 * src/x86/ffi.c: Fix spelling error of "separate" as "seperate".
3747 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3749 * src/x86/sysv.S: Avoid gas-only .balign directive.
3750 Use C style comments.
3752 2001-07-16 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3754 * src/alpha/ffi.c (ffi_prep_closure): Avoid gas-only mnemonic.
3755 Fixes PR bootstrap/3563.
3757 2001-06-26 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3759 * src/alpha/osf.S (ffi_closure_osf): Use .rdata for ECOFF.
3761 2001-06-25 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
3763 * configure.in: Recognize sparc*-sun-* host.
3764 * configure: Regenerate.
3766 2001-06-06 Andrew Haley <aph@redhat.com>
3768 * src/alpha/osf.S (__FRAME_BEGIN__): Conditionalize for ELF.
3770 2001-06-03 Andrew Haley <aph@redhat.com>
3772 * src/alpha/osf.S: Add unwind info.
3773 * src/powerpc/sysv.S: Add unwind info.
3774 * src/powerpc/ppc_closure.S: Likewise.
3776 2000-05-31 Jeff Sturm <jsturm@one-point.com>
3778 * configure.in: Fix AC_ARG_ENABLE usage.
3779 * configure: Rebuilt.
3781 2001-05-06 Bryce McKinlay <bryce@waitaki.otago.ac.nz>
3783 * configure.in: Remove warning about beta code.
3784 * configure: Rebuilt.
3786 2001-04-25 Hans Boehm <Hans_Boehm@hp.com>
3788 * src/ia64/unix.S: Restore stack pointer when returning from
3790 * src/ia64/ffi.c: Fix typo in comment.
3792 2001-04-18 Jim Wilson <wilson@redhat.com>
3794 * src/ia64/unix.S: Delete unnecessary increment and decrement of loc2
3795 to eliminate RAW DV.
3797 2001-04-12 Bryce McKinlay <bryce@albatross.co.nz>
3799 * Makefile.am: Make a libtool convenience library.
3800 * Makefile.in: Rebuilt.
3802 2001-03-29 Bryce McKinlay <bryce@albatross.co.nz>
3804 * configure.in: Use different syntax for subdirectory creation.
3805 * configure: Rebuilt.
3807 2001-03-27 Jon Beniston <jon@beniston.com>
3809 * configure.in: Added X86_WIN32 target (Win32, CygWin, MingW).
3810 * configure: Rebuilt.
3811 * Makefile.am: Added X86_WIN32 target support.
3812 * Makefile.in: Rebuilt.
3814 * include/ffi.h.in: Added X86_WIN32 target support.
3816 * src/ffitest.c: Doesn't run structure tests for X86_WIN32 targets.
3817 * src/types.c: Added X86_WIN32 target support.
3819 * src/x86/win32.S: New file. Based on sysv.S, but with EH
3820 stuff removed and made to work with CygWin's gas.
3822 2001-03-26 Bryce McKinlay <bryce@albatross.co.nz>
3824 * configure.in: Make target subdirectory in build dir.
3825 * Makefile.am: Override suffix based rules to specify correct output
3827 * Makefile.in: Rebuilt.
3828 * configure: Rebuilt.
3830 2001-03-23 Kevin B Hendricks <khendricks@ivey.uwo.ca>
3832 * src/powerpc/ppc_closure.S: New file.
3833 * src/powerpc/ffi.c (ffi_prep_args): Fixed ABI compatibility bug
3834 involving long long and register pairs.
3835 (ffi_prep_closure): New function.
3836 (flush_icache): Likewise.
3837 (ffi_closure_helper_SYSV): Likewise.
3838 * include/ffi.h.in (FFI_CLOSURES): Define on PPC.
3839 (FFI_TRAMPOLINE_SIZE): Likewise.
3840 (FFI_NATIVE_RAW_API): Likewise.
3841 * Makefile.in: Rebuilt.
3842 * Makefile.am (EXTRA_DIST): Added src/powerpc/ppc_closure.S.
3843 (TARGET_SRC_POWERPC): Likewise.
3845 2001-03-19 Tom Tromey <tromey@redhat.com>
3847 * Makefile.in: Rebuilt.
3848 * Makefile.am (ffitest_LDFLAGS): New macro.
3850 2001-03-02 Nick Clifton <nickc@redhat.com>
3852 * include/ffi.h.in: Remove RCS ident string.
3853 * include/ffi_mips.h: Remove RCS ident string.
3854 * src/debug.c: Remove RCS ident string.
3855 * src/ffitest.c: Remove RCS ident string.
3856 * src/prep_cif.c: Remove RCS ident string.
3857 * src/types.c: Remove RCS ident string.
3858 * src/alpha/ffi.c: Remove RCS ident string.
3859 * src/alpha/osf.S: Remove RCS ident string.
3860 * src/arm/ffi.c: Remove RCS ident string.
3861 * src/arm/sysv.S: Remove RCS ident string.
3862 * src/mips/ffi.c: Remove RCS ident string.
3863 * src/mips/n32.S: Remove RCS ident string.
3864 * src/mips/o32.S: Remove RCS ident string.
3865 * src/sparc/ffi.c: Remove RCS ident string.
3866 * src/sparc/v8.S: Remove RCS ident string.
3867 * src/sparc/v9.S: Remove RCS ident string.
3868 * src/x86/ffi.c: Remove RCS ident string.
3869 * src/x86/sysv.S: Remove RCS ident string.
3871 2001-02-08 Joseph S. Myers <jsm28@cam.ac.uk>
3873 * include/ffi.h.in: Change sourceware.cygnus.com references to
3876 2000-12-09 Richard Henderson <rth@redhat.com>
3878 * src/alpha/ffi.c (ffi_call): Simplify struct return test.
3879 (ffi_closure_osf_inner): Index rather than increment avalue
3880 and arg_types. Give ffi_closure_osf the raw return value type.
3881 * src/alpha/osf.S (ffi_closure_osf): Handle return value type
3884 2000-12-07 Richard Henderson <rth@redhat.com>
3886 * src/raw_api.c (ffi_translate_args): Fix typo.
3887 (ffi_prep_closure): Likewise.
3889 * include/ffi.h.in [ALPHA]: Define FFI_CLOSURES and
3890 FFI_TRAMPOLINE_SIZE.
3891 * src/alpha/ffi.c (ffi_prep_cif_machdep): Adjust minimal
3892 cif->bytes for new ffi_call_osf implementation.
3893 (ffi_prep_args): Absorb into ...
3894 (ffi_call): ... here. Do all stack allocation here and
3895 avoid a callback function.
3896 (ffi_prep_closure, ffi_closure_osf_inner): New.
3897 * src/alpha/osf.S (ffi_call_osf): Reimplement with no callback.
3898 (ffi_closure_osf): New.
3900 2000-09-10 Alexandre Oliva <aoliva@redhat.com>
3902 * config.guess, config.sub, install-sh: Removed.
3903 * ltconfig, ltmain.sh, missing, mkinstalldirs: Likewise.
3904 * Makefile.in: Rebuilt.
3906 * acinclude.m4: Include libtool macros from the top level.
3907 * aclocal.m4, configure: Rebuilt.
3909 2000-08-22 Alexandre Oliva <aoliva@redhat.com>
3911 * configure.in [i*86-*-freebsd*] (TARGET, TARGETDIR): Set.
3912 * configure: Rebuilt.
3914 2000-05-11 Scott Bambrough <scottb@netwinder.org>
3916 * libffi/src/arm/sysv.S (ffi_call_SYSV): Doubles are not saved to
3917 memory correctly. Use conditional instructions, not branches where
3920 2000-05-04 Tom Tromey <tromey@cygnus.com>
3922 * configure: Rebuilt.
3923 * configure.in: Match `arm*-*-linux-*'.
3924 From Chris Dornan <cdornan@arm.com>.
3926 2000-04-28 Jakub Jelinek <jakub@redhat.com>
3928 * Makefile.am (SUBDIRS): Define.
3929 (AM_MAKEFLAGS): Likewise.
3930 (Multilib support.): Add section.
3931 * Makefile.in: Rebuilt.
3932 * ltconfig (extra_compiler_flags, extra_compiler_flags_value):
3933 New variables. Set for gcc using -print-multi-lib. Export them
3935 (sparc64-*-linux-gnu*): Use libsuff 64 for search paths.
3936 * ltmain.sh (B|b|V): Don't throw away gcc's -B, -b and -V options
3938 (extra_compiler_flags_value, extra_compiler_flags): Check these
3939 for extra compiler options which need to be passed down in
3942 2000-04-16 Anthony Green <green@redhat.com>
3944 * configure: Rebuilt.
3945 * configure.in: Change i*86-pc-linux* to i*86-*-linux*.
3947 2000-04-14 Jakub Jelinek <jakub@redhat.com>
3949 * include/ffi.h.in (SPARC64): Define for 64bit SPARC builds.
3950 Set SPARC FFI_DEFAULT_ABI based on SPARC64 define.
3951 * src/sparc/ffi.c (ffi_prep_args_v8): Renamed from ffi_prep_args.
3952 Replace all void * sizeofs with sizeof(int).
3953 Only compare type with FFI_TYPE_LONGDOUBLE if LONGDOUBLE is
3954 different than DOUBLE.
3955 Remove FFI_TYPE_SINT32 and FFI_TYPE_UINT32 cases (handled elsewhere).
3956 (ffi_prep_args_v9): New function.
3957 (ffi_prep_cif_machdep): Handle V9 ABI and long long on V8.
3958 (ffi_V9_return_struct): New function.
3959 (ffi_call): Handle FFI_V9 ABI from 64bit code and FFI_V8 ABI from
3960 32bit code (not yet cross-arch calls).
3961 * src/sparc/v8.S: Add struct return delay nop.
3963 * src/sparc/v9.S: New file.
3964 * src/prep_cif.c (ffi_prep_cif): Return structure pointer
3965 is used on sparc64 only for structures larger than 32 bytes.
3966 Pass by reference for structures is done for structure arguments
3967 larger than 16 bytes.
3968 * src/ffitest.c (main): Use 64bit rint on sparc64.
3969 Run long long tests on sparc.
3970 * src/types.c (FFI_TYPE_POINTER): Pointer is 64bit on alpha and
3972 (FFI_TYPE_LONGDOUBLE): long double is 128 bit aligned to 128 bits
3974 * configure.in (sparc-*-linux*): New supported target.
3975 (sparc64-*-linux*): Likewise.
3976 * configure: Rebuilt.
3977 * Makefile.am: Add v9.S to SPARC files.
3978 * Makefile.in: Likewise.
3979 (LINK): Surround $(CCLD) into double quotes, so that multilib
3980 compiles work correctly.
3982 2000-04-04 Alexandre Petit-Bianco <apbianco@cygnus.com>
3984 * configure: Rebuilt.
3985 * configure.in: (i*86-*-solaris*): New libffi target. Patch
3986 proposed by Bryce McKinlay.
3988 2000-03-20 Tom Tromey <tromey@cygnus.com>
3990 * Makefile.in: Hand edit for java_raw_api.lo.
3992 2000-03-08 Bryce McKinlay <bryce@albatross.co.nz>
3994 * config.guess, config.sub: Update from the gcc tree.
3995 Fix for PR libgcj/168.
3997 2000-03-03 Tom Tromey <tromey@cygnus.com>
3999 * Makefile.in: Fixed ia64 by hand.
4001 * configure: Rebuilt.
4002 * configure.in (--enable-multilib): New option.
4003 (libffi_basedir): New subst.
4004 (AC_OUTPUT): Added multilib code.
4006 2000-03-02 Tom Tromey <tromey@cygnus.com>
4008 * Makefile.in: Rebuilt.
4009 * Makefile.am (TARGET_SRC_IA64): Use `ia64', not `alpha', as
4012 2000-02-25 Hans Boehm <boehm@acm.org>
4014 * src/ia64/ffi.c, src/ia64/ia64_flags.h, src/ia64/unix.S: New
4016 * src/raw_api.c (ffi_translate_args): Fixed typo in argument
4018 (ffi_prep_raw_closure): Use ffi_translate_args, not
4019 ffi_closure_translate.
4020 * src/java_raw_api.c: New file.
4021 * src/ffitest.c (closure_test_fn): New function.
4022 (main): Define `rint' as long long on IA64. Added new test when
4023 FFI_CLOSURES is defined.
4024 * include/ffi.h.in (ALIGN): Use size_t, not unsigned.
4025 (ffi_abi): Recognize IA64.
4026 (ffi_raw): Added `flt' field.
4027 Added "Java raw API" code.
4028 * configure.in: Recognize ia64.
4029 * Makefile.am (TARGET_SRC_IA64): New macro.
4030 (libffi_la_common_SOURCES): Added java_raw_api.c.
4031 (libffi_la_SOURCES): Define in IA64 case.
4033 2000-01-04 Tom Tromey <tromey@cygnus.com>
4035 * Makefile.in: Rebuilt with newer automake.
4037 1999-12-31 Tom Tromey <tromey@cygnus.com>
4039 * Makefile.am (INCLUDES): Added -I$(top_srcdir)/src.
4041 1999-09-01 Tom Tromey <tromey@cygnus.com>
4043 * include/ffi.h.in: Removed PACKAGE and VERSION defines and
4045 * fficonfig.h.in: Rebuilt.
4046 * configure: Rebuilt.
4047 * configure.in: Pass 3rd argument to AM_INIT_AUTOMAKE.
4048 Use AM_PROG_LIBTOOL (automake 1.4 compatibility).
4049 * acconfig.h: Don't #undef PACKAGE or VERSION.
4051 1999-08-09 Anthony Green <green@cygnus.com>
4053 * include/ffi.h.in: Try to work around messy header problem
4054 with PACKAGE and VERSION.
4056 * configure: Rebuilt.
4057 * configure.in: Change version to 2.00-beta.
4059 * fficonfig.h.in: Rebuilt.
4060 * acconfig.h (FFI_NO_STRUCTS, FFI_NO_RAW_API): Define.
4062 * src/x86/ffi.c (ffi_raw_call): Rename.
4064 1999-08-02 Kresten Krab Thorup <krab@dominiq.is.s.u-tokyo.ac.jp>
4066 * src/x86/ffi.c (ffi_closure_SYSV): New function.
4067 (ffi_prep_incoming_args_SYSV): Ditto.
4068 (ffi_prep_closure): Ditto.
4069 (ffi_closure_raw_SYSV): Ditto.
4070 (ffi_prep_raw_closure): More ditto.
4071 (ffi_call_raw): Final ditto.
4073 * include/ffi.h.in: Add definitions for closure and raw API.
4075 * src/x86/ffi.c (ffi_prep_cif_machdep): Added case for
4078 * Makefile.am (libffi_la_common_SOURCES): Added raw_api.c
4080 * src/raw_api.c: New file.
4082 * include/ffi.h.in (ffi_raw): New type.
4083 (UINT_ARG, SINT_ARG): New defines.
4084 (ffi_closure, ffi_raw_closure): New types.
4085 (ffi_prep_closure, ffi_prep_raw_closure): New declarations.
4087 * configure.in: Add check for endianness and sizeof void*.
4089 * src/x86/sysv.S (ffi_call_SYSV): Call fixup routine via argument,
4090 instead of directly.
4092 * configure: Rebuilt.
4094 Thu Jul 8 14:28:42 1999 Anthony Green <green@cygnus.com>
4096 * configure.in: Add x86 and powerpc BeOS configurations.
4097 From Makoto Kato <m_kato@ga2.so-net.ne.jp>.
4099 1999-05-09 Anthony Green <green@cygnus.com>
4101 * configure.in: Add warning about this being beta code.
4102 Remove src/Makefile.am from the picture.
4103 * configure: Rebuilt.
4105 * Makefile.am: Move logic from src/Makefile.am. Add changes
4106 to support libffi as a target library.
4107 * Makefile.in: Rebuilt.
4109 * aclocal.m4, config.guess, config.sub, ltconfig, ltmain.sh:
4110 Upgraded to new autoconf, automake, libtool.
4114 * LICENSE: Update copyright date.
4116 * src/Makefile.am, src/Makefile.in: Removed.
4118 1998-11-29 Anthony Green <green@cygnus.com>
4120 * include/ChangeLog: Removed.
4121 * src/ChangeLog: Removed.
4122 * src/mips/ChangeLog: Removed.
4123 * src/sparc/ChangeLog: Remboved.
4124 * src/x86/ChangeLog: Removed.
4126 * ChangeLog.v1: Created.