+2014-03-10 Mike Hommey <mh+mozilla@glandium.org>
+
+ * configure.ac: Allow building for mipsel with Android NDK r8.
+ * Makefile.am (AM_MAKEFLAGS): Replace double quotes with single quotes.
+
+2014-03-01 Anthony Green <green@moxielogic.com>
+
+ * Makefile.am (EXTRA_DIST): Replace old scripts with
+ generate-darwin-source-and-headers.py.
+ * Makefile.in: Rebuilt.
+
+2014-02-28 Anthony Green <green@moxielogic.com>
+
+ * Makefile.am (AM_CFLAGS): Reintroduce missing -DFFI_DEBUG for
+ --enable-debug builds.
+ * Makefile.in: Rebuilt.
+
+2014-02-28 Makoto Kato <m_kato@ga2.so-net.ne.jp>
+
+ * src/closures.c: Fix build failure when using clang for Android.
+
+2014-02-28 Marcin Wojdyr <wojdyr@gmail.com>
+
+ * libffi.pc.in (toolexeclibdir): use -L${toolexeclibdir} instead
+ of -L${libdir}.
+
+2014-02-28 Paulo Pizarro <paulo.pizarro@gmail.com>
+
+ * src/bfin/sysv.S: Calling functions in shared libraries requires
+ considering the GOT.
+
+2014-02-28 Josh Triplett <josh@joshtriplett.org>
+
+ * src/x86/ffi64.c (classify_argument): Handle case where
+ FFI_TYPE_LONGDOUBLE == FFI_TYPE_DOUBLE.
+
+2014-02-28 Anthony Green <green@moxielogic.com>
+
+ * ltmain.sh: Generate with libtool-2.4.2.418.
+ * m4/libtool.m4, m4/ltoptions.m4, m4/ltversion.m4: Ditto.
+ * configure: Rebuilt.
+
+2014-02-28 Dominik Vogt <vogt@linux.vnet.ibm.com>
+
+ * configure.ac (AC_ARG_ENABLE struct): Fix typo in help
+ message.
+ (AC_ARG_ENABLE raw_api): Ditto.
+ * configure, fficonfig.h.in: Rebuilt.
+
+2014-02-28 Will Newton <will.newton@linaro.org>
+
+ * src/arm/sysv.S: Initialize IP register with FP.
+
+2014-02-28 Yufeng Zhang <yufeng.zhang@arm.com>
+
+ * src/aarch64/sysv.S (ffi_closure_SYSV): Use x29 as the
+ main CFA reg; update cfi_rel_offset.
+
+2014-02-09 Heiher <r@hev.cc>
+
+ * src/mips/n32.S: Fix call floating point va function.
+
+2014-01-21 Zachary Waldowski <zach@waldowski.me>
+
+ * src/aarch64/ffi.c: Fix missing semicolons on assertions under
+ debug mode.
+
+2013-12-30 Zachary Waldowski <zach@waldowski.me>
+
+ * .gitignore: Exclude darwin_* generated source and build_* trees.
+ * src/aarch64/ffi.c, src/arm/ffi.c, src/x86/ffi.c: Inhibit Clang
+ previous prototype warnings.
+ * src/arm/ffi.c: Prevent NULL dereference, fix short type warning
+ * src/dlmalloc.c: Fix warnings from set_segment_flags return type,
+ and the native use of size_t for malloc on platforms
+ * src/arm/sysv.S: Use unified syntax. Clang clean-ups for
+ ARM_FUNC_START.
+ * generate-osx-source-and-headers.py: Remove.
+ * build-ios.sh: Remove.
+ * libffi.xcodeproj/project.pbxproj: Rebuild targets. Include
+ x86_64+aarch64 pieces in library. Export headers properly.
+ * src/x86/ffi64.c: More Clang warning clean-ups.
+ * src/closures.c (open_temp_exec_file_dir): Use size_t.
+ * src/prep_cif.c (ffi_prep_cif_core): Cast ALIGN result.
+ * src/aarch64/sysv.S: Use CNAME for global symbols. Only use
+ .size for ELF targets.
+ * src/aarch64/ffi.c: Clean up for double == long double. Clean up
+ from Clang warnings. Use Clang cache invalidation builtin. Use
+ size_t in place of unsigned in many places. Accommodate for
+ differences in Apple AArch64 ABI.
+
+2013-12-02 Daniel RodrÃguez Troitiño <drodrigueztroitino@yahoo.es>
+
+ * generate-darwin-source-and-headers.py: Clean up, modernize,
+ merged version of previous scripts.
+
+2013-11-21 Anthony Green <green@moxielogic.com>
+
+ * configure, Makefile.in, include/Makefile.in, include/ffi.h.in,
+ man/Makefile.in, testsuite/Makefile.in, fficonfig.h.in: Rebuilt.
+
+2013-11-21 Alan Modra <amodra@gmail.com>
+
+ * Makefile.am (EXTRA_DIST): Add new src/powerpc files.
+ (nodist_libffi_la_SOURCES <POWERPC, POWERPC_FREEBSD>): Likewise.
+ * configure.ac (HAVE_LONG_DOUBLE_VARIANT): Define for powerpc.
+ * include/ffi.h.in (ffi_prep_types): Declare.
+ * src/prep_cif.c (ffi_prep_cif_core): Call ffi_prep_types.
+ * src/types.c (FFI_NONCONST_TYPEDEF): Define and use for
+ HAVE_LONG_DOUBLE_VARIANT.
+ * src/powerpc/ffi_powerpc.h: New file.
+ * src/powerpc/ffi.c: Split into..
+ * src/powerpc/ffi_sysv.c: ..new file, and..
+ * src/powerpc/ffi_linux64.c: ..new file, rewriting parts.
+ * src/powerpc/ffitarget.h (enum ffi_abi): Rewrite powerpc ABI
+ selection as bits controlling features.
+ * src/powerpc/linux64.S: For consistency, use POWERPC64 rather
+ than __powerpc64__.
+ * src/powerpc/linux64_closure.S: Likewise.
+ * src/powerpc/ppc_closure.S: Likewise. Move .note.FNU-stack
+ inside guard.
+ * src/powerpc/sysv.S: Likewise.
+ * configure: Regenerate.
+ * fficonfig.h.in: Regenerate.
+ * Makefile.in: Regenerate.
+
+2013-11-20 Alan Modra <amodra@gmail.com>
+
+ * src/powerpc/ffi.c (ffi_prep_cif_machdep_core): Use
+ NUM_FPR_ARG_REGISTERS64 and NUM_GPR_ARG_REGISTERS64 not their
+ 32-bit versions for 64-bit code.
+ * src/powerpc/linux64_closure.S: Don't use the return value area
+ as a parameter save area on ELFv2.
+
+2013-11-18 Iain Sandoe <iain@codesourcery.com>
+
+ * src/powerpc/darwin.S (EH): Correct use of pcrel FDE encoding.
+ * src/powerpc/darwin_closure.S (EH): Likewise. Modernise picbase
+ labels.
+
+2013-11-18 Anthony Green <green@moxielogic.com>
+
+ * src/arm/ffi.c (ffi_call): Hoist declaration of temp to top of
+ function.
+ * src/arm/ffi.c (ffi_closure_inner): Moderize function declaration
+ to appease compiler.
+ Thanks for Gregory P. Smith <greg@krypto.org>.
+
+2013-11-18 Anthony Green <green@moxielogic.com>
+
+ * README (tested): Mention PowerPC ELFv2.
+
+2013-11-16 Alan Modra <amodra@gmail.com>
+
+ * src/powerpc/ppc_closure.S: Move errant #endif to where it belongs.
+ Don't bl .Luint128.
+
+2013-11-16 Alan Modra <amodra@gmail.com>
+
+ * src/powerpc/ffi.c (ffi_prep_cif_machdep_core): Use #if _CALL_ELF
+ test to select parameter save sizing for ELFv2 vs. ELFv1.
+ * src/powerpc/ffitarget.h (FFI_V2_TYPE_FLOAT_HOMOG,
+ FFI_V2_TYPE_DOUBLE_HOMOG, FFI_V2_TYPE_SMALL_STRUCT): Define.
+ (FFI_TRAMPOLINE_SIZE): Define variant for ELFv2.
+ * src/powerpc/ffi.c (FLAG_ARG_NEEDS_PSAVE): Define.
+ (discover_homogeneous_aggregate): New function.
+ (ffi_prep_args64): Adjust start of param save area for ELFv2.
+ Handle homogenous floating point struct parms.
+ (ffi_prep_cif_machdep_core): Adjust space calculation for ELFv2.
+ Handle ELFv2 return values. Set FLAG_ARG_NEEDS_PSAVE. Handle
+ homogenous floating point structs.
+ (ffi_call): Increase size of smst_buffer for ELFv2. Handle ELFv2.
+ (flush_icache): Compile for ELFv2.
+ (ffi_prep_closure_loc): Set up ELFv2 trampoline.
+ (ffi_closure_helper_LINUX64): Don't return all structs directly
+ to caller. Handle homogenous floating point structs. Handle
+ ELFv2 struct return values.
+ * src/powerpc/linux64.S (ffi_call_LINUX64): Set up r2 for
+ ELFv2. Adjust toc save location. Call function pointer using
+ r12. Handle FLAG_RETURNS_SMST. Don't predict branches.
+ * src/powerpc/linux64_closure.S (ffi_closure_LINUX64): Set up r2
+ for ELFv2. Define ELFv2 versions of STACKFRAME, PARMSAVE, and
+ RETVAL. Handle possibly missing parameter save area. Handle
+ ELFv2 return values.
+ (.note.GNU-stack): Move inside outer #ifdef.
+
+2013-11-16 Alan Modra <amodra@gmail.com>
+
+ * src/powerpc/ffi.c (ffi_prep_cif_machdep): Revert 2013-02-08
+ change. Do not consume an int arg when returning a small struct
+ for FFI_SYSV ABI.
+ (ffi_call): Only use bounce buffer when FLAG_RETURNS_SMST.
+ Properly copy bounce buffer to destination.
+ * src/powerpc/sysv.S: Revert 2013-02-08 change.
+ * src/powerpc/ppc_closure.S: Remove stray '+'.
+
+2013-11-16 Alan Modra <amodra@gmail.com>
+
+ * src/powerpc/ffi.c (ffi_prep_args64): Align struct parameters
+ according to __STRUCT_PARM_ALIGN__.
+ (ffi_prep_cif_machdep_core): Likewise.
+ (ffi_closure_helper_LINUX64): Likewise.
+
+2013-11-16 Alan Modra <amodra@gmail.com>
+
+ * src/powerpc/linux64.S (ffi_call_LINUX64): Tweak restore of r28.
+ (.note.GNU-stack): Move inside outer #ifdef.
+ * src/powerpc/linux64_closure.S (STACKFRAME, PARMSAVE,
+ RETVAL): Define and use throughout.
+ (ffi_closure_LINUX64): Save fprs before buying stack.
+ (.note.GNU-stack): Move inside outer #ifdef.
+
+2013-11-16 Alan Modra <amodra@gmail.com>
+
+ * src/powerpc/ffitarget.h (FFI_TARGET_SPECIFIC_VARIADIC): Define.
+ (FFI_EXTRA_CIF_FIELDS): Define.
+ * src/powerpc/ffi.c (ffi_prep_args64): Save fprs as per the
+ ABI, not to both fpr and param save area.
+ (ffi_prep_cif_machdep_core): Renamed from ffi_prep_cif_machdep.
+ Keep initial flags. Formatting. Remove dead FFI_LINUX_SOFT_FLOAT
+ code.
+ (ffi_prep_cif_machdep, ffi_prep_cif_machdep_var): New functions.
+ (ffi_closure_helper_LINUX64): Pass floating point as per ABI,
+ not to both fpr and parameter save areas.
+
+ * libffi/testsuite/libffi.call/cls_double_va.c (main): Correct
+ function cast and don't call ffi_prep_cif.
+ * libffi/testsuite/libffi.call/cls_longdouble_va.c (main): Likewise.
+
+2013-11-15 Andrew Haley <aph@redhat.com>
+
+ * doc/libffi.texi (Closure Example): Fix the sample code.
+ * doc/libffi.info, doc/stamp-vti, doc/version.texi: Rebuilt.
+
+2013-11-15 Andrew Haley <aph@redhat.com>
+
+ * testsuite/libffi.call/va_struct1.c (main): Fix broken test.
+ * testsuite/libffi.call/cls_uint_va.c (cls_ret_T_fn): Likewise
+ * testsuite/libffi.call/cls_struct_va1.c (test_fn): Likewise.
+ * testsuite/libffi.call/va_1.c (main): Likewise.
+
+2013-11-14 David Schneider <david.schneider@bivab.de>
+
+ * src/arm/ffi.c: Fix register allocation for mixed float and
+ doubles.
+ * testsuite/libffi.call/cls_many_mixed_float_double.c: Testcase
+ for many mixed float and double arguments.
+
+2013-11-13 Alan Modra <amodra@gmail.com>
+
+ * doc/libffi.texi (Simple Example): Correct example code.
+ * doc/libffi.info, doc/stamp-vti, doc/version.texi: Rebuilt.
+
+2013-11-13 Anthony Green <green@moxielogic.com>
+
+ * include/ffi_common.h: Respect HAVE_ALLOCA_H for GNU compiler
+ based build. (Thanks to tmr111116 on github)
+
+2013-11-09 Anthony Green <green@moxielogic.com>
+
+ * m4/libtool.m4: Refresh.
+ * configure, Makefile.in: Rebuilt.
+ * README: Add more notes about next release.
+
+2013-11-09 Shigeharu TAKENO <shige@iee.niit.ac.jp>
+
+ * m4/ax_gcc_archflag.m4 (ax_gcc_arch): Don't recognize
+ UltraSPARC-IIi as ultrasparc3.
+
+2013-11-06 Mark Kettenis <kettenis@gnu.org>
+
+ * src/x86/freebsd.S (ffi_call_SYSV): Align the stack pointer to
+ 16-bytes.
+
+2013-11-06 Konstantin Belousov <kib@freebsd.org>
+
+ * src/x86/freebsd.S (ffi_closure_raw_SYSV): Mark the assembler
+ source as not requiring executable stack.
+
+2013-11-02 Anthony Green <green@moxielogic.com>
+
+ * doc/libffi.texi (The Basics): Clarify return value buffer size
+ requirements. Also, NULL result buffer pointers are no longer
+ supported.
+ * doc/libffi.info: Rebuilt.
+
+2013-11-02 Mischa Jonker <mjonker@synopsys.com>
+
+ * Makefile.am (nodist_libffi_la_SOURCES): Fix build error.
+ * Makefile.in: Rebuilt.
+
+2013-11-02 David Schneider <david.schneider@bivab.de>
+
+ * src/arm/ffi.c: more robust argument handling for closures on arm hardfloat
+ * testsuite/libffi.call/many_mixed.c: New file.
+ * testsuite/libffi.call/cls_many_mixed_args.c: More tests.
+
+2013-11-02 Vitaly Budovski
+
+ * src/x86/ffi.c (ffi_prep_cif_machdep): Don't align stack for win32.
+
2013-10-23 Mark H Weaver <mhw@netris.org>
* src/mips/ffi.c: Fix handling of uint32_t arguments on the
git.haskell.org / libffi.git / blobdiff