+2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
+
+ PR libgcj/21943
+ * src/mips/n32.S: Enforce PIC code.
+ * src/mips/o32.S: Likewise.
+
+2005-06-15 Rainer Orth <ro@TechFak.Uni-Bielefeld.DE>
+
+ * configure.ac: Treat i*86-*-solaris2.10 and up as X86_64.
+ * configure: Regenerate.
+
+2005-06-01 Alan Modra <amodra@bigpond.net.au>
+
+ * src/powerpc/ppc_closure.S (ffi_closure_SYSV): Don't use JUMPTARGET
+ to call ffi_closure_helper_SYSV. Append @local instead.
+ * src/powerpc/sysv.S (ffi_call_SYSV): Likewise for ffi_prep_args_SYSV.
+
+2005-05-17 Kelley Cook <kcook@gcc.gnu.org>
+
+ * configure.ac: Use AC_C_BIGENDIAN instead of AC_C_BIGENDIAN_CROSS.
+ Use AC_CHECK_SIZEOF instead of AC_COMPILE_CHECK_SIZEOF.
+ * Makefile.am (ACLOCAL_AMFLAGS): Remove -I ../config.
+ * aclocal.m4, configure, fficonfig.h.in, Makefile.in,
+ include/Makefile.in, testsuite/Makefile.in: Regenerate.
+
+2005-05-09 Mike Stump <mrs@apple.com>
+
+ * configure: Regenerate.
+
+2005-05-08 Richard Henderson <rth@redhat.com>
+
+ PR libffi/21285
+ * src/alpha/osf.S: Update unwind into to match code.
+
+2005-05-04 Andreas Degert <ad@papyrus-gmbh.de>
+ Richard Henderson <rth@redhat.com>
+
+ * src/x86/ffi64.c (ffi_prep_cif_machdep): Save sse-used flag in
+ bit 11 of flags.
+ (ffi_call): Mask return type field. Pass ssecount to ffi_call_unix64.
+ (ffi_prep_closure): Set carry bit if sse-used flag set.
+ * src/x86/unix64.S (ffi_call_unix64): Add ssecount argument.
+ Only load sse registers if ssecount non-zero.
+ (ffi_closure_unix64): Only save sse registers if carry set on entry.
+
+2005-04-29 Ralf Corsepius <ralf.corsepius@rtems.org>
+
+ * configure.ac: Add i*86-*-rtems*, sparc*-*-rtems*,
+ powerpc-*rtems*, arm*-*-rtems*, sh-*-rtems*.
+ * configure: Regenerate.
+
+2005-04-20 Hans-Peter Nilsson <hp@axis.com>
+
+ * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): In regsub use,
+ have Tcl8.3-compatible intermediate variable.
+
+2005-04-18 Simon Posnjak <simon.posnjak@siol.net>
+ Hans-Peter Nilsson <hp@axis.com>
+
+ * Makefile.am: Add CRIS support.
+ * configure.ac: Likewise.
+ * Makefile.in, configure, testsuite/Makefile.in,
+ include/Makefile.in: Regenerate.
+ * src/cris: New directory.
+ * src/cris/ffi.c, src/cris/sysv.S, src/cris/ffitarget.h: New files.
+ * src/prep_cif.c (ffi_prep_cif): Wrap in #ifndef __CRIS__.
+
+ * testsuite/lib/libffi-dg.exp (libffi-dg-test-1): Replace \n with
+ \r?\n in output tests.
+
+2005-04-12 Mike Stump <mrs@apple.com>
+
+ * configure: Regenerate.
+
+2005-03-30 Hans Boehm <Hans.Boehm@hp.com>
+
+ * src/ia64/ffitarget.h (ffi_arg): Use long long instead of DI.
+
+2005-03-30 Steve Ellcey <sje@cup.hp.com>
+
+ * src/ia64/ffitarget.h (ffi_arg) ADD DI attribute.
+ (ffi_sarg) Ditto.
+ * src/ia64/unix.S (ffi_closure_unix): Extend gp
+ to 64 bits in ILP32 mode.
+ Load 64 bits even for short data.
+
+2005-03-23 Mike Stump <mrs@apple.com>
+
+ * src/powerpc/darwin.S: Update for -m64 multilib.
+ * src/powerpc/darwin_closure.S: Likewise.
+
+2005-03-21 Zack Weinberg <zack@codesourcery.com>
+
+ * configure.ac: Do not invoke TL_AC_GCC_VERSION.
+ Do not set tool_include_dir.
+ * aclocal.m4, configure, Makefile.in, testsuite/Makefile.in:
+ Regenerate.
+ * include/Makefile.am: Set gcc_version and toollibffidir.
+ * include/Makefile.in: Regenerate.
+
+2005-02-22 Andrew Haley <aph@redhat.com>
+
+ * src/powerpc/ffi.c (ffi_prep_cif_machdep): Bump alignment to
+ odd-numbered register pairs for 64-bit integer types.
+
+2005-02-23 Andreas Tobler <a.tobler@schweiz.ch>
+
+ PR libffi/20104
+ * testsuite/libffi.call/return_ll1.c: New test case.
+
+2005-02-11 Janis Johnson <janis187@us.ibm.com>
+
+ * testsuite/libffi.call/cls_align_longdouble.c: Remove dg-options.
+ * testsuite/libffi.call/float.c: Ditto.
+ * testsuite/libffi.call/float2.c: Ditto.
+ * testsuite/libffi.call/float3.c: Ditto.
+
+2005-02-08 Andreas Tobler <a.tobler@schweiz.ch>
+
+ * src/frv/ffitarget.h: Remove PPC stuff which does not belong to frv.
+
+2005-01-12 Eric Botcazou <ebotcazou@libertysurf.fr>
+
+ * testsuite/libffi.special/special.exp (cxx_options): Add
+ -shared-libgcc.
+
+2004-12-31 Richard Henderson <rth@redhat.com>
+
+ * src/types.c (FFI_AGGREGATE_TYPEDEF): Remove.
+ (FFI_TYPEDEF): Rename from FFI_INTEGRAL_TYPEDEF. Replace size and
+ offset parameters with a type parameter; deduce size and structure
+ alignment. Update all users.
+
+2004-12-31 Richard Henderson <rth@redhat.com>
+
+ * src/types.c (FFI_TYPE_POINTER): Define with sizeof.
+ (FFI_TYPE_LONGDOUBLE): Fix for ia64.
+ * src/ia64/ffitarget.h (struct ffi_ia64_trampoline_struct): Move
+ into ffi_prep_closure.
+ * src/ia64/ia64_flags.h, src/ia64/ffi.c, src/ia64/unix.S: Rewrite
+ from scratch.
+
+2004-12-27 Richard Henderson <rth@redhat.com>
+
+ * src/x86/unix64.S: Fix typo in unwind info.
+
+2004-12-25 Richard Henderson <rth@redhat.com>
+
+ * src/x86/ffi64.c (struct register_args): Rename from stackLayout.
+ (enum x86_64_reg_class): Add X86_64_COMPLEX_X87_CLASS.
+ (merge_classes): Check for it.
+ (SSE_CLASS_P): New.
+ (classify_argument): Pass byte_offset by value; perform all updates
+ inside struct case.
+ (examine_argument): Add classes argument; handle
+ X86_64_COMPLEX_X87_CLASS.
+ (ffi_prep_args): Merge into ...
+ (ffi_call): ... here. Share stack frame with ffi_call_unix64.
+ (ffi_prep_cif_machdep): Setup cif->flags for proper structure return.
+ (ffi_fill_return_value): Remove.
+ (ffi_prep_closure): Remove dead assert.
+ (ffi_closure_unix64_inner): Rename from ffi_closure_UNIX64_inner.
+ Rewrite to use struct register_args instead of va_list. Create
+ flags for handling structure returns.
+ * src/x86/unix64.S: Remove dead strings.
+ (ffi_call_unix64): Rename from ffi_call_UNIX64. Rewrite to share
+ stack frame with ffi_call. Handle structure returns properly.
+ (float2sse, floatfloat2sse, double2sse): Remove.
+ (sse2float, sse2double, sse2floatfloat): Remove.
+ (ffi_closure_unix64): Rename from ffi_closure_UNIX64. Rewrite
+ to handle structure returns properly.
+
2004-12-08 David Edelsohn <edelsohn@gnu.org>
* Makefile.am (AM_MAKEFLAGS): Remove duplicate LIBCFLAGS and
integer (O32 ABI only).
(ffi_prep_closure): new function.
(ffi_closure_mips_inner_O32): new function.
- * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
+ * src/mips/ffitarget.h: Define `FFI_CLOSURES' and
`FFI_TRAMPOLINE_SIZE' appropriately if the ABI is o32.
* src/mips/o32.S (ffi_call_O32): add labels for .eh_frame. Return
64 bit integers correctly.
* src/powerpc/darwin.S: Go through a non-lazy pointer for initial
FDE location.
* src/powerpc/darwin_closure.S: Likewise.
-
+
2004-04-24 Andreas Tobler <a.tobler@schweiz.ch>
* testsuite/libffi.call/cls_multi_schar.c (main): Fix initialization
2003-03-03 Andreas Tobler <a.tobler@schweiz.ch>
- * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
+ * src/powerpc/darwin_closure.S: Recode to fit dynamic libraries.
2003-02-06 Andreas Tobler <a.tobler@schweiz.ch>
2003-01-31 Christian Cornelssen <ccorn@cs.tu-berlin.de>,
Andreas Schwab <schwab@suse.de>
- * configure.in: Adjust command to source config-ml.in to account
+ * configure.in: Adjust command to source config-ml.in to account
for changes to the libffi_basedir definition.
(libffi_basedir): Remove ${srcdir} from value and include trailing
slash if nonempty.
- * configure: Regenerate.
+ * configure: Regenerate.
2003-01-29 Franz Sirl <Franz.Sirl-kernel@lauterbach.com>
2003-01-13 Andreas Tobler <a.tobler@schweiz.ch>
* libffi/src/ffitest.c
- add closure testcases
+ add closure testcases
2003-01-13 Kevin B. Hendricks <khendricks@ivey.uwo.ca>
* libffi/src/powerpc/ffi.c
- fix alignment bug for float (4 byte aligned iso 8 byte)
+ fix alignment bug for float (4 byte aligned iso 8 byte)
2003-01-09 Geoffrey Keating <geoffk@apple.com>