gcc: introduce version 4.5.2
PATCH 1/1 gcc: introduce version 4.5.2
- based on gcc-4.5.2-RC-20101208 (svn 167585)
- without linaro patches (which seem to be the root of
problems when compiling at least on armv4 and armv5) :
- samba 3.2.15 or 3.3.9 : http://tinderbox.openembedded.net/public/logs/task/8666826.txt
- samba 3.5.6 : http://pastebin.com/yuiYX2CM
--- recipes/gcc/gcc-4.5.2.inc | 65 +++++++ recipes/gcc/gcc-4.5.2/100-uclibc-conf.patch | 37 ++++ recipes/gcc/gcc-4.5.2/Makefile.in.patch | 30 ++++ recipes/gcc/gcc-4.5.2/arm-bswapsi2.patch | 13 ++ recipes/gcc/gcc-4.5.2/cache-amnesia.patch | 31 ++++ .../gcc/gcc-4.5.2/fortran-cross-compile-hack.patch | 30 ++++ .../gcc-4.3.1-ARCH_FLAGS_FOR_TARGET.patch | 31 ++++ .../gcc-4.5.2/gcc-armv4-pass-fix-v4bx-to-ld.patch | 29 ++++ recipes/gcc/gcc-4.5.2/gcc-flags-for-build.patch | 178 ++++++++++++++++++++ .../gcc-4.5.2/gcc-scalar-widening-pr45847.patch | 63 +++++++ .../gcc-uclibc-locale-ctype_touplow_t.patch | 67 ++++++++ .../libstdc++-emit-__cxa_end_cleanup-in-text.patch | 40 +++++ recipes/gcc/gcc-4.5.2/mips64-nomultilib.patch | 52 ++++++ recipes/gcc/gcc-4.5.2/zecke-no-host-includes.patch | 31 ++++ recipes/gcc/gcc-4.5.2/zecke-xgcc-cpp.patch | 28 +++ recipes/gcc/gcc-cross-initial_4.5.2.bb | 2 + recipes/gcc/gcc-cross-intermediate_4.5.2.bb | 2 + recipes/gcc/gcc-cross-sdk_4.5.2.bb | 8 + recipes/gcc/gcc-cross_4.5.2.bb | 12 ++ recipes/gcc/gcc_4.5.2.bb | 15 ++ 20 files changed, 764 insertions(+), 0 deletions(-) create mode 100644 recipes/gcc/gcc-4.5.2.inc create mode 100644 recipes/gcc/gcc-4.5.2/100-uclibc-conf.patch create mode 100644 recipes/gcc/gcc-4.5.2/Makefile.in.patch create mode 100644 recipes/gcc/gcc-4.5.2/arm-bswapsi2.patch create mode 100644 recipes/gcc/gcc-4.5.2/cache-amnesia.patch create mode 100644 recipes/gcc/gcc-4.5.2/fortran-cross-compile-hack.patch create mode 100644 recipes/gcc/gcc-4.5.2/gcc-4.3.1-ARCH_FLAGS_FOR_TARGET.patch create mode 100644 recipes/gcc/gcc-4.5.2/gcc-armv4-pass-fix-v4bx-to-ld.patch create mode 100644 recipes/gcc/gcc-4.5.2/gcc-flags-for-build.patch create mode 100644 recipes/gcc/gcc-4.5.2/gcc-scalar-widening-pr45847.patch create mode 100644 recipes/gcc/gcc-4.5.2/gcc-uclibc-locale-ctype_touplow_t.patch create mode 100644 recipes/gcc/gcc-4.5.2/libstdc++-emit-__cxa_end_cleanup-in-text.patch create mode 100644 recipes/gcc/gcc-4.5.2/mips64-nomultilib.patch create mode 100644 recipes/gcc/gcc-4.5.2/zecke-no-host-includes.patch create mode 100644 recipes/gcc/gcc-4.5.2/zecke-xgcc-cpp.patch create mode 100644 recipes/gcc/gcc-cross-initial_4.5.2.bb create mode 100644 recipes/gcc/gcc-cross-intermediate_4.5.2.bb create mode 100644 recipes/gcc/gcc-cross-sdk_4.5.2.bb create mode 100644 recipes/gcc/gcc-cross_4.5.2.bb create mode 100644 recipes/gcc/gcc_4.5.2.bb
recipes/gcc/gcc-4.5.2.inc new file +65 -0
@@ -0,0 +1,65 @@ +DEFAULT_PREFERENCE = "-999" +ARM_INSTRUCTION_SET = "arm" + +require gcc-common.inc +LICENSE = "GPLv3" + +DEPENDS = "mpfr gmp libmpc libelf" +NATIVEDEPS = "mpfr-native gmp-native libmpc-native" + + +INC_PR = "r1" + +SRCREV = "167585" +PV = "4.5.2" +# BINV should be incremented after updating to a revision +# after a minor gcc release (e.g. 4.5.1 or 4.5.2) has been made +# the value will be minor-release+1 e.g. if minor release was +# 4.5.1 then the value below will be 2 which will mean 4.5.2 +# which will be next minor release and so on. + +BINV = "${PV}.2" +BRANCH = "gcc-4_5-branch" +PR_append = "+svnr${SRCPV}" + +SRC_URI = "svn://gcc.gnu.org/svn/gcc/branches;module=${BRANCH} \ + file://gcc-4.3.1-ARCH_FLAGS_FOR_TARGET.patch \ + file://100-uclibc-conf.patch \ + file://gcc-uclibc-locale-ctype_touplow_t.patch \ + file://cache-amnesia.patch \ + file://gcc-flags-for-build.patch \ + file://libstdc++-emit-__cxa_end_cleanup-in-text.patch \ + file://arm-bswapsi2.patch \ + file://Makefile.in.patch \ + file://gcc-armv4-pass-fix-v4bx-to-ld.patch \ + file://gcc-scalar-widening-pr45847.patch \ + " + +SRC_URI_append_mips64 = " file://mips64-nomultilib.patch " +SRC_URI_append_mips64el = " file://mips64-nomultilib.patch " + +# Language Overrides +FORTRAN = "" +JAVA = "" +rename_srcdir (){ + mv ${WORKDIR}/${BRANCH} ${WORKDIR}/gcc-${PV} +} + +do_unpack_append() { + bb.build.exec_func('rename_srcdir', d) +} + +EXTRA_OECONF_BASE = " --enable-lto \ + --enable-libssp \ + --disable-bootstrap --disable-libgomp \ + --disable-libmudflap" + +EXTRA_OECONF_INITIAL = "--disable-libmudflap --disable-libgomp --disable-libssp --enable-decimal-float=no" +EXTRA_OECONF_INTERMEDIATE = "--disable-libmudflap --disable-libgomp --disable-libssp" + +EXTRA_OECONF_append_linux-uclibc = " --disable-decimal-float " +EXTRA_OECONF_append_linux-uclibceabi = " --disable-decimal-float " +EXTRA_OECONF_append_linux-uclibcspe = " --disable-decimal-float " + +EXTRA_OECONF_append_mips64 = " --with-arch-64=mips64 --with-tune-64=mips64" +EXTRA_OECONF_append_mips64el = " --with-arch-64=mips64 --with-tune-64=mips64"
recipes/gcc/gcc-4.5.2/100-uclibc-conf.patch new file +37 -0
@@ -0,0 +1,37 @@ +Index: gcc-4.3.1/contrib/regression/objs-gcc.sh +=================================================================== +--- gcc-4.3.1.orig/contrib/regression/objs-gcc.sh 2007-12-24 15:18:57.000000000 -0800 ++++ gcc-4.3.1/contrib/regression/objs-gcc.sh 2008-08-16 01:15:12.000000000 -0700 +@@ -105,6 +105,10 @@ + then + make all-gdb all-dejagnu all-ld || exit 1 + make install-gdb install-dejagnu install-ld || exit 1 ++elif [ $H_REAL_TARGET = $H_REAL_HOST -a $H_REAL_TARGET = i686-pc-linux-uclibc ] ++ then ++ make all-gdb all-dejagnu all-ld || exit 1 ++ make install-gdb install-dejagnu install-ld || exit 1 + elif [ $H_REAL_TARGET = $H_REAL_HOST ] ; then + make bootstrap || exit 1 + make install || exit 1 +Index: gcc-4.3.1/libjava/classpath/ltconfig +=================================================================== +--- gcc-4.3.1.orig/libjava/classpath/ltconfig 2007-06-03 16:18:43.000000000 -0700 ++++ gcc-4.3.1/libjava/classpath/ltconfig 2008-08-16 01:15:12.000000000 -0700 +@@ -603,7 +603,7 @@ + + # Transform linux* to *-*-linux-gnu*, to support old configure scripts. + case $host_os in +-linux-gnu*) ;; ++linux-gnu*|linux-uclibc*) ;; + linux*) host=`echo $host | sed 's/^\(.*-.*-linux\)\(.*\)$/\1-gnu\2/'` + esac + +@@ -1251,7 +1251,7 @@ + ;; + + # This must be Linux ELF. +-linux-gnu*) ++linux*) + version_type=linux + need_lib_prefix=no + need_version=no
recipes/gcc/gcc-4.5.2/Makefile.in.patch new file +30 -0
@@ -0,0 +1,30 @@ +Index: gcc-4.5/gcc/Makefile.in +=================================================================== +--- gcc-4.5.orig/gcc/Makefile.in ++++ gcc-4.5/gcc/Makefile.in +@@ -656,7 +656,7 @@ LIBGCC2_INCLUDES = + TARGET_LIBGCC2_CFLAGS = + + # Options to use when compiling crtbegin/end. +-CRTSTUFF_CFLAGS = -O2 $(GCC_CFLAGS) $(INCLUDES) $(MULTILIB_CFLAGS) -g0 \ ++CRTSTUFF_CFLAGS = -O2 $(GCC_CFLAGS) $(TARGET_INCLUDES) $(MULTILIB_CFLAGS) -g0 \ + -finhibit-size-directive -fno-inline -fno-exceptions \ + -fno-zero-initialized-in-bss -fno-toplevel-reorder -fno-tree-vectorize \ + $(INHIBIT_LIBC_CFLAGS) +@@ -1038,10 +1038,14 @@ BUILD_ERRORS = build/errors.o + # -I$(@D) and -I$(srcdir)/$(@D) cause the subdirectory of the file + # currently being compiled, in both source trees, to be examined as well. + # libintl.h will be found in ../intl if we are using the included libintl. +-INCLUDES = -I. -I$(@D) -I$(srcdir) -I$(srcdir)/$(@D) \ ++# ++# TARGET_INCLUDES is added to avoid that GMPINC (which points to the host ++# include dir) is used for compiling libgcc.a ++TARGET_INCLUDES = -I. -I$(@D) -I$(srcdir) -I$(srcdir)/$(@D) \ + -I$(srcdir)/../include @INCINTL@ \ +- $(CPPINC) $(GMPINC) $(DECNUMINC) \ ++ $(CPPINC) $(DECNUMINC) \ + $(PPLINC) $(CLOOGINC) $(LIBELFINC) ++INCLUDES = $(TARGET_INCLUDES) $(GMPINC) + + .c.o: + $(COMPILER) -c $(ALL_COMPILERFLAGS) $(ALL_CPPFLAGS) $< $(OUTPUT_OPTION)
recipes/gcc/gcc-4.5.2/arm-bswapsi2.patch new file +13 -0
@@ -0,0 +1,13 @@ +Index: gcc-4.5/gcc/config/arm/arm.md +=================================================================== +--- gcc-4.5.orig/gcc/config/arm/arm.md 2010-06-17 09:13:07.000000000 -0700 ++++ gcc-4.5/gcc/config/arm/arm.md 2010-06-22 08:08:45.397212002 -0700 +@@ -11267,7 +11267,7 @@ + (define_expand "bswapsi2" + [(set (match_operand:SI 0 "s_register_operand" "=r") + (bswap:SI (match_operand:SI 1 "s_register_operand" "r")))] +-"TARGET_EITHER" ++"TARGET_EITHER && (arm_arch6 && !optimize_size)" + " + if (!arm_arch6) + {
recipes/gcc/gcc-4.5.2/cache-amnesia.patch new file +31 -0
@@ -0,0 +1,31 @@ +--- + gcc/configure | 2 +- + gcc/configure.ac | 2 +- + 2 files changed, 2 insertions(+), 2 deletions(-) + +Index: gcc-4.5+svnr155514/gcc/configure +=================================================================== +--- gcc-4.5+svnr155514.orig/gcc/configure 2009-12-29 22:00:40.000000000 -0800 ++++ gcc-4.5+svnr155514/gcc/configure 2009-12-29 23:52:43.381592113 -0800 +@@ -10467,7 +10467,7 @@ else + saved_CFLAGS="${CFLAGS}" + CC="${CC_FOR_BUILD}" CFLAGS="${CFLAGS_FOR_BUILD}" \ + LDFLAGS="${LDFLAGS_FOR_BUILD}" \ +- ${realsrcdir}/configure \ ++ CONFIG_SITE= ${realsrcdir}/configure --cache-file=./other.cache \ + --enable-languages=${enable_languages-all} \ + --target=$target_alias --host=$build_alias --build=$build_alias + CFLAGS="${saved_CFLAGS}" +Index: gcc-4.5+svnr155514/gcc/configure.ac +=================================================================== +--- gcc-4.5+svnr155514.orig/gcc/configure.ac 2009-12-29 22:00:40.000000000 -0800 ++++ gcc-4.5+svnr155514/gcc/configure.ac 2009-12-29 23:51:54.589091778 -0800 +@@ -1458,7 +1458,7 @@ else + saved_CFLAGS="${CFLAGS}" + CC="${CC_FOR_BUILD}" CFLAGS="${CFLAGS_FOR_BUILD}" \ + LDFLAGS="${LDFLAGS_FOR_BUILD}" \ +- ${realsrcdir}/configure \ ++ CONFIG_SITE= ${realsrcdir}/configure --cache-file=./other.cache \ + --enable-languages=${enable_languages-all} \ + --target=$target_alias --host=$build_alias --build=$build_alias + CFLAGS="${saved_CFLAGS}"
recipes/gcc/gcc-4.5.2/fortran-cross-compile-hack.patch new file +30 -0
@@ -0,0 +1,30 @@ +* Fortran would have searched for arm-angstrom-gnueabi-gfortran but would have used + used gfortan. For gcc_4.2.2.bb we want to use the gfortran compiler from our cross + directory. + +Index: gcc-4.5+svnr155514/libgfortran/configure +=================================================================== +--- gcc-4.5+svnr155514.orig/libgfortran/configure 2009-12-29 22:02:01.000000000 -0800 ++++ gcc-4.5+svnr155514/libgfortran/configure 2009-12-30 08:12:40.889091657 -0800 +@@ -11655,7 +11655,7 @@ CC="$lt_save_CC" + + # We need gfortran to compile parts of the library + #AC_PROG_FC(gfortran) +-FC="$GFORTRAN" ++#FC="$GFORTRAN" + ac_ext=${ac_fc_srcext-f} + ac_compile='$FC -c $FCFLAGS $ac_fcflags_srcext conftest.$ac_ext >&5' + ac_link='$FC -o conftest$ac_exeext $FCFLAGS $LDFLAGS $ac_fcflags_srcext conftest.$ac_ext $LIBS >&5' +Index: gcc-4.5+svnr155514/libgfortran/configure.ac +=================================================================== +--- gcc-4.5+svnr155514.orig/libgfortran/configure.ac 2009-12-29 22:02:01.000000000 -0800 ++++ gcc-4.5+svnr155514/libgfortran/configure.ac 2009-12-30 08:12:13.453094218 -0800 +@@ -187,7 +187,7 @@ AC_SUBST(enable_static) + + # We need gfortran to compile parts of the library + #AC_PROG_FC(gfortran) +-FC="$GFORTRAN" ++#FC="$GFORTRAN" + AC_PROG_FC(gfortran) + + # extra LD Flags which are required for targets
recipes/gcc/gcc-4.5.2/gcc-4.3.1-ARCH_FLAGS_FOR_TARGET.patch new file +31 -0
@@ -0,0 +1,31 @@ +--- + configure | 2 +- + configure.ac | 2 +- + 2 files changed, 2 insertions(+), 2 deletions(-) + +Index: gcc-4.3.1/configure.ac +=================================================================== +--- gcc-4.3.1.orig/configure.ac 2008-07-21 12:29:18.000000000 -0700 ++++ gcc-4.3.1/configure.ac 2008-07-21 12:29:35.000000000 -0700 +@@ -2352,7 +2352,7 @@ fi + # for target_alias and gcc doesn't manage it consistently. + target_configargs="--cache-file=./config.cache ${target_configargs}" + +-FLAGS_FOR_TARGET= ++FLAGS_FOR_TARGET="$ARCH_FLAGS_FOR_TARGET" + case " $target_configdirs " in + *" newlib "*) + case " $target_configargs " in +Index: gcc-4.3.1/configure +=================================================================== +--- gcc-4.3.1.orig/configure 2008-07-21 12:29:48.000000000 -0700 ++++ gcc-4.3.1/configure 2008-07-21 12:29:59.000000000 -0700 +@@ -5841,7 +5841,7 @@ fi + # for target_alias and gcc doesn't manage it consistently. + target_configargs="--cache-file=./config.cache ${target_configargs}" + +-FLAGS_FOR_TARGET= ++FLAGS_FOR_TARGET="$ARCH_FLAGS_FOR_TARGET" + case " $target_configdirs " in + *" newlib "*) + case " $target_configargs " in
recipes/gcc/gcc-4.5.2/gcc-armv4-pass-fix-v4bx-to-ld.patch new file +29 -0
@@ -0,0 +1,29 @@ +The LINK_SPEC for linux gets overwritten by linux-eabi.h which +means the value of TARGET_FIX_V4BX_SPEC gets lost and as a result +the option is not passed to linker when chosing march=armv4 +This patch redefines this in linux-eabi.h and reinserts it +for eabi defaulting toolchains. + +We might want to send it upstream + +-Khem +Index: gcc-4.5/gcc/config/arm/linux-eabi.h +=================================================================== +--- gcc-4.5.orig/gcc/config/arm/linux-eabi.h ++++ gcc-4.5/gcc/config/arm/linux-eabi.h +@@ -63,10 +63,14 @@ + #undef GLIBC_DYNAMIC_LINKER + #define GLIBC_DYNAMIC_LINKER "/lib/ld-linux.so.3" + ++/* For armv4 we pass --fix-v4bx to linker to support EABI */ ++#undef TARGET_FIX_V4BX_SPEC ++#define TARGET_FIX_V4BX_SPEC " %{mcpu=arm8|mcpu=arm810|mcpu=strongarm*|march=armv4:--fix-v4bx}" ++ + /* At this point, bpabi.h will have clobbered LINK_SPEC. We want to + use the GNU/Linux version, not the generic BPABI version. */ + #undef LINK_SPEC +-#define LINK_SPEC LINUX_TARGET_LINK_SPEC BE8_LINK_SPEC ++#define LINK_SPEC LINUX_TARGET_LINK_SPEC BE8_LINK_SPEC TARGET_FIX_V4BX_SPEC + + /* Use the default LIBGCC_SPEC, not the version in linux-elf.h, as we + do not use -lfloat. */
recipes/gcc/gcc-4.5.2/gcc-flags-for-build.patch new file +178 -0
178 lines not on this page · raw message
recipes/gcc/gcc-4.5.2/gcc-scalar-widening-pr45847.patch new file +63 -0
63 lines not on this page · raw message
recipes/gcc/gcc-4.5.2/gcc-uclibc-locale-ctype_touplow_t.patch new file +67 -0
67 lines not on this page · raw message
recipes/gcc/gcc-4.5.2/libstdc++-emit-__cxa_end_cleanup-in-text.patch new file +40 -0
@@ -0,0 +1,40 @@ +2010-06-07 Khem Raj <[email protected]> + + * libsupc++/eh_arm.cc (__cxa_end_cleanup): Use .pushsection/.popsection + to emit inline assembly into .text section. + +Index: gcc-4.5/libstdc++-v3/libsupc++/eh_arm.cc +=================================================================== +--- gcc-4.5.orig/libstdc++-v3/libsupc++/eh_arm.cc 2010-06-04 23:20:18.000000000 -0700 ++++ gcc-4.5/libstdc++-v3/libsupc++/eh_arm.cc 2010-06-08 11:27:34.247541722 -0700 +@@ -157,22 +157,26 @@ __gnu_end_cleanup(void) + // Assembly wrapper to call __gnu_end_cleanup without clobbering r1-r3. + // Also push r4 to preserve stack alignment. + #ifdef __thumb__ +-asm (".global __cxa_end_cleanup\n" ++asm (" .pushsection .text.__cxa_end_cleanup\n" ++" .global __cxa_end_cleanup\n" + " .type __cxa_end_cleanup, \"function\"\n" + " .thumb_func\n" + "__cxa_end_cleanup:\n" + " push\t{r1, r2, r3, r4}\n" + " bl\t__gnu_end_cleanup\n" + " pop\t{r1, r2, r3, r4}\n" +-" bl\t_Unwind_Resume @ Never returns\n"); ++" bl\t_Unwind_Resume @ Never returns\n" ++" .popsection\n"); + #else +-asm (".global __cxa_end_cleanup\n" ++asm (" .pushsection .text.__cxa_end_cleanup\n" ++" .global __cxa_end_cleanup\n" + " .type __cxa_end_cleanup, \"function\"\n" + "__cxa_end_cleanup:\n" + " stmfd\tsp!, {r1, r2, r3, r4}\n" + " bl\t__gnu_end_cleanup\n" + " ldmfd\tsp!, {r1, r2, r3, r4}\n" +-" bl\t_Unwind_Resume @ Never returns\n"); ++" bl\t_Unwind_Resume @ Never returns\n" ++" .popsection\n"); + #endif + + #endif
recipes/gcc/gcc-4.5.2/mips64-nomultilib.patch new file +52 -0
52 lines not on this page · raw message
recipes/gcc/gcc-4.5.2/zecke-no-host-includes.patch new file +31 -0
@@ -0,0 +1,31 @@ +Index: gcc-4.4+svnr145550/gcc/incpath.c +=================================================================== +--- gcc-4.4+svnr145550.orig/gcc/incpath.c 2009-04-04 13:48:31.000000000 -0700 ++++ gcc-4.4+svnr145550/gcc/incpath.c 2009-04-04 14:49:29.000000000 -0700 +@@ -417,6 +417,26 @@ + p->construct = 0; + p->user_supplied_p = user_supplied_p; + ++#ifdef CROSS_COMPILE ++ /* A common error when cross compiling is including ++ host headers. This code below will try to fail fast ++ for cross compiling. Currently we consider /usr/include, ++ /opt/include and /sw/include as harmful. */ ++ { ++ /* printf("Adding Path: %s\n", p->name ); */ ++ if( strstr(p->name, "/usr/include" ) == p->name ) { ++ fprintf(stderr, _("CROSS COMPILE Badness: /usr/include in INCLUDEPATH: %s\n"), p->name); ++ abort(); ++ } else if( strstr(p->name, "/sw/include") == p->name ) { ++ fprintf(stderr, _("CROSS COMPILE Badness: /sw/include in INCLUDEPATH: %s\n"), p->name); ++ abort(); ++ } else if( strstr(p->name, "/opt/include") == p->name ) { ++ fprintf(stderr, _("CROSS COMPILE Badness: /opt/include in INCLUDEPATH: %s\n"), p->name); ++ abort(); ++ } ++ } ++#endif ++ + add_cpp_dir_path (p, chain); + } +
recipes/gcc/gcc-4.5.2/zecke-xgcc-cpp.patch new file +28 -0
@@ -0,0 +1,28 @@ +upstream: n/a +comment: Use the preprocessor we have just compiled instead the one of +the system. There might be incompabilities between us and them. + +Index: gcc-4.3.1/Makefile.in +=================================================================== +--- gcc-4.3.1.orig/Makefile.in 2008-08-19 01:09:56.000000000 -0700 ++++ gcc-4.3.1/Makefile.in 2008-08-19 01:13:27.000000000 -0700 +@@ -204,6 +204,7 @@ + AR="$(AR_FOR_TARGET)"; export AR; \ + AS="$(COMPILER_AS_FOR_TARGET)"; export AS; \ + CC="$(CC_FOR_TARGET)"; export CC; \ ++ CPP="$(CC_FOR_TARGET) -E"; export CPP; \ + CFLAGS="$(CFLAGS_FOR_TARGET) $(SYSROOT_CFLAGS_FOR_TARGET) $(DEBUG_PREFIX_CFLAGS_FOR_TARGET)"; export CFLAGS; \ + CONFIG_SHELL="$(SHELL)"; export CONFIG_SHELL; \ + CPPFLAGS="$(CPPFLAGS_FOR_TARGET)"; export CPPFLAGS; \ +Index: gcc-4.3.1/Makefile.tpl +=================================================================== +--- gcc-4.3.1.orig/Makefile.tpl 2008-08-21 00:07:58.000000000 -0700 ++++ gcc-4.3.1/Makefile.tpl 2008-08-21 00:09:52.000000000 -0700 +@@ -223,6 +223,7 @@ + AR="$(AR_FOR_TARGET)"; export AR; \ + AS="$(COMPILER_AS_FOR_TARGET)"; export AS; \ + CC="$(CC_FOR_TARGET)"; export CC; \ ++ CPP="$(CC_FOR_TARGET) -E"; export CPP; \ + CFLAGS="$(CFLAGS_FOR_TARGET) $(SYSROOT_CFLAGS_FOR_TARGET) $(DEBUG_PREFIX_CFLAGS_FOR_TARGET)"; export CFLAGS; \ + CONFIG_SHELL="$(SHELL)"; export CONFIG_SHELL; \ + CPPFLAGS="$(CPPFLAGS_FOR_TARGET)"; export CPPFLAGS; \
recipes/gcc/gcc-cross-initial_4.5.2.bb new file +2 -0
@@ -0,0 +1,2 @@ +require gcc-cross_${PV}.bb +require gcc-cross-initial.inc
recipes/gcc/gcc-cross-intermediate_4.5.2.bb new file +2 -0
@@ -0,0 +1,2 @@ +require gcc-cross_${PV}.bb +require gcc-cross-intermediate.inc
recipes/gcc/gcc-cross-sdk_4.5.2.bb new file +8 -0
@@ -0,0 +1,8 @@ +inherit sdk + +require gcc-${PV}.inc +require gcc-cross-sdk.inc +PR = "${INC_PR}.1" + +EXTRA_OECONF += " --disable-libunwind-exceptions --with-mpfr=${STAGING_DIR_NATIVE}${prefix_native} --with-system-zlib" +
recipes/gcc/gcc-cross_4.5.2.bb new file +12 -0
@@ -0,0 +1,12 @@ +require gcc-${PV}.inc +require gcc-cross4.inc + +PR = "${INC_PR}.0" + +NATIVEDEPS += "libmpc-native libelf-native" +SRC_URI_append_fail-fast = " file://zecke-no-host-includes.patch " + +EXTRA_OECONF += " --disable-libunwind-exceptions --with-mpfr=${STAGING_DIR_NATIVE}${prefix_native} \ + -with-libelf=${STAGING_DIR_NATIVE}${prefix_native} --with-system-zlib" + +ARCH_FLAGS_FOR_TARGET += "-isystem${STAGING_DIR_TARGET}${target_includedir}"
recipes/gcc/gcc_4.5.2.bb new file +15 -0
@@ -0,0 +1,15 @@ +PR = "${INC_PR}.1" +require gcc-${PV}.inc +require gcc-configure-target.inc +require gcc-package-target.inc + +# Gcc 4.3.3 installs crt* in a '4.3.1' dir.... +FILES_${PN} += "\ + ${gcclibdir}/${TARGET_SYS}/*/*.o \ + ${libexecdir}/gcc/${TARGET_SYS}/${BINV}/lto1 \ + ${libexecdir}/gcc/${TARGET_SYS}/${BINV}/lto-wrapper \ +" + +SRC_URI_append = "file://fortran-cross-compile-hack.patch" + +ARCH_FLAGS_FOR_TARGET += "-isystem${STAGING_INCDIR}"
Signature · 1 line
1.6.3.3
Quoted 6 lines · Eric Bénard
- based on gcc-4.5.2-RC-20101208 (svn 167585)
- without linaro patches (which seem to be the root of
problems when compiling at least on armv4 and armv5) :
- samba 3.2.15 or 3.3.9 : http://tinderbox.openembedded.net/public/logs/task/8666826.txt
- samba 3.5.6 : http://pastebin.com/yuiYX2CM
as discussed on IRC may be we should try to find the faulty patch in current gcc 4.5 linaro patches we have.
Quoted 937 lines
Signed-off-by Eric Bénard <[email protected]>---
recipes/gcc/gcc-4.5.2.inc | 65 +++++++recipes/gcc/gcc-4.5.2/100-uclibc-conf.patch| 37 ++++recipes/gcc/gcc-4.5.2/Makefile.in.patch| 30 ++++recipes/gcc/gcc-4.5.2/arm-bswapsi2.patch| 13 ++recipes/gcc/gcc-4.5.2/cache-amnesia.patch| 31 ++++.../gcc/gcc-4.5.2/fortran-cross-compile-hack.patch| 30 ++++.../gcc-4.3.1-ARCH_FLAGS_FOR_TARGET.patch| 31 ++++.../gcc-4.5.2/gcc-armv4-pass-fix-v4bx-to-ld.patch| 29 ++++recipes/gcc/gcc-4.5.2/gcc-flags-for-build.patch| 178 ++++++++++++++++++++.../gcc-4.5.2/gcc-scalar-widening-pr45847.patch| 63 +++++++.../gcc-uclibc-locale-ctype_touplow_t.patch| 67 ++++++++.../libstdc++-emit-__cxa_end_cleanup-in-text.patch| 40 +++++recipes/gcc/gcc-4.5.2/mips64-nomultilib.patch| 52 ++++++recipes/gcc/gcc-4.5.2/zecke-no-host-includes.patch| 31 ++++recipes/gcc/gcc-4.5.2/zecke-xgcc-cpp.patch| 28 +++ recipes/gcc/gcc-cross-initial_4.5.2.bb | 2 + recipes/gcc/gcc-cross-intermediate_4.5.2.bb | 2 + recipes/gcc/gcc-cross-sdk_4.5.2.bb | 8 + recipes/gcc/gcc-cross_4.5.2.bb | 12 ++ recipes/gcc/gcc_4.5.2.bb | 15 ++20 files changed, 764 insertions(+), 0 deletions(-)create mode 100644 recipes/gcc/gcc-4.5.2.inc
create mode 100644recipes/gcc/gcc-4.5.2/100-uclibc-conf.patch
create mode 100644recipes/gcc/gcc-4.5.2/Makefile.in.patch
create mode 100644recipes/gcc/gcc-4.5.2/arm-bswapsi2.patch
create mode 100644recipes/gcc/gcc-4.5.2/cache-amnesia.patch
create mode 100644recipes/gcc/gcc-4.5.2/fortran-cross-compile-hack.patchcreate mode 100644recipes/gcc/gcc-4.5.2/gcc-4.3.1-ARCH_FLAGS_FOR_TARGET.patchcreate mode 100644recipes/gcc/gcc-4.5.2/gcc-armv4-pass-fix-v4bx-to-ld.patchcreate mode 100644recipes/gcc/gcc-4.5.2/gcc-flags-for-build.patch
create mode 100644recipes/gcc/gcc-4.5.2/gcc-scalar-widening-pr45847.patchcreate mode 100644recipes/gcc/gcc-4.5.2/gcc-uclibc-locale-ctype_touplow_t.patchcreate mode 100644recipes/gcc/gcc-4.5.2/libstdc++-emit-__cxa_end_cleanup-in-text.patchcreate mode 100644recipes/gcc/gcc-4.5.2/mips64-nomultilib.patch
create mode 100644recipes/gcc/gcc-4.5.2/zecke-no-host-includes.patchcreate mode 100644recipes/gcc/gcc-4.5.2/zecke-xgcc-cpp.patch
create mode 100644 recipes/gcc/gcc-cross-initial_4.5.2.bb
create mode 100644 recipes/gcc/gcc-cross-intermediate_4.5.2.bb
create mode 100644 recipes/gcc/gcc-cross-sdk_4.5.2.bb
create mode 100644 recipes/gcc/gcc-cross_4.5.2.bb
create mode 100644 recipes/gcc/gcc_4.5.2.bbrecipes/gcc/gcc-4.5.2.inc new file +65
-0@@ -0,0 +1,65 @@ +DEFAULT_PREFERENCE = "-999" +ARM_INSTRUCTION_SET = "arm" + +require gcc-common.inc +LICENSE = "GPLv3" + +DEPENDS = "mpfr gmp libmpc libelf" +NATIVEDEPS = "mpfr-native gmp-native libmpc-native" + + +INC_PR = "r1" + +SRCREV = "167585" +PV = "4.5.2" +# BINV should be incremented after updating to a revision +# after a minor gcc release (e.g. 4.5.1 or 4.5.2) has been made +# the value will be minor-release+1 e.g. if minor release was +# 4.5.1 then the value below will be 2 which will mean 4.5.2 +# which will be next minor release and so on. + +BINV = "${PV}.2" +BRANCH = "gcc-4_5-branch" +PR_append = "+svnr${SRCPV}" + +SRC_URI = "svn://gcc.gnu.org/svn/gcc/branches;module=${BRANCH} \ + file://gcc-4.3.1-ARCH_FLAGS_FOR_TARGET.patch \ + file://100-uclibc-conf.patch \ + file://gcc-uclibc-locale-ctype_touplow_t.patch \ + file://cache-amnesia.patch \ + file://gcc-flags-for-build.patch \ + file://libstdc++-emit-__cxa_end_cleanup-in-text.patch \ + file://arm-bswapsi2.patch \ + file://Makefile.in.patch \ + file://gcc-armv4-pass-fix-v4bx-to-ld.patch \ + file://gcc-scalar-widening-pr45847.patch \ + " + +SRC_URI_append_mips64 = " file://mips64-nomultilib.patch " +SRC_URI_append_mips64el = " file://mips64-nomultilib.patch " + +# Language Overrides +FORTRAN = "" +JAVA = "" +rename_srcdir (){ + mv ${WORKDIR}/${BRANCH} ${WORKDIR}/gcc-${PV} +} + +do_unpack_append() { + bb.build.exec_func('rename_srcdir', d) +} + +EXTRA_OECONF_BASE = " --enable-lto \ + --enable-libssp \ + --disable-bootstrap --disable-libgomp \ + --disable-libmudflap" + +EXTRA_OECONF_INITIAL = "--disable-libmudflap --disable-libgomp --disable-libssp --enable-decimal-float=no" +EXTRA_OECONF_INTERMEDIATE = "--disable-libmudflap --disable-libgomp --disable-libssp" + +EXTRA_OECONF_append_linux-uclibc = " --disable-decimal-float " +EXTRA_OECONF_append_linux-uclibceabi = " --disable-decimal-float " +EXTRA_OECONF_append_linux-uclibcspe = " --disable-decimal-float " + +EXTRA_OECONF_append_mips64 = " --with-arch-64=mips64 --with-tune-64=mips64" +EXTRA_OECONF_append_mips64el = " --with-arch-64=mips64 --with-tune-64=mips64"recipes/gcc/gcc-4.5.2/100-uclibc-conf.patch new file +37
-0@@ -0,0 +1,37 @@ +Index: gcc-4.3.1/contrib/regression/objs-gcc.sh +=================================================================== +--- gcc-4.3.1.orig/contrib/regression/objs-gcc.sh 2007-12-24 15:18:57.000000000 -0800 ++++ gcc-4.3.1/contrib/regression/objs-gcc.sh 2008-08-16 01:15:12.000000000 -0700 +@@ -105,6 +105,10 @@ + then + make all-gdb all-dejagnu all-ld || exit 1 + make install-gdb install-dejagnu install-ld || exit 1 ++elif [ $H_REAL_TARGET = $H_REAL_HOST -a $H_REAL_TARGET = i686-pc-linux-uclibc ] ++ then ++ make all-gdb all-dejagnu all-ld || exit 1 ++ make install-gdb install-dejagnu install-ld || exit 1 + elif [ $H_REAL_TARGET = $H_REAL_HOST ] ; then + make bootstrap || exit 1 + make install || exit 1 +Index: gcc-4.3.1/libjava/classpath/ltconfig +=================================================================== +--- gcc-4.3.1.orig/libjava/classpath/ltconfig 2007-06-03 16:18:43.000000000 -0700 ++++ gcc-4.3.1/libjava/classpath/ltconfig 2008-08-16 01:15:12.000000000 -0700 +@@ -603,7 +603,7 @@ + + # Transform linux* to *-*-linux-gnu*, to support old configure scripts. + case $host_os in +-linux-gnu*) ;; ++linux-gnu*|linux-uclibc*) ;; + linux*) host=`echo $host | sed 's/^\(.*-.*-linux\)\(.*\)$/\1-gnu\2/'` + esac + +@@ -1251,7 +1251,7 @@ + ;; + + # This must be Linux ELF. +-linux-gnu*) ++linux*) + version_type=linux + need_lib_prefix=no + need_version=norecipes/gcc/gcc-4.5.2/Makefile.in.patch new file +30
-0@@ -0,0 +1,30 @@ +Index: gcc-4.5/gcc/Makefile.in +=================================================================== +--- gcc-4.5.orig/gcc/Makefile.in ++++ gcc-4.5/gcc/Makefile.in +@@ -656,7 +656,7 @@ LIBGCC2_INCLUDES = + TARGET_LIBGCC2_CFLAGS = + + # Options to use when compiling crtbegin/end. +-CRTSTUFF_CFLAGS = -O2 $(GCC_CFLAGS) $(INCLUDES) $(MULTILIB_CFLAGS) -g0 \ ++CRTSTUFF_CFLAGS = -O2 $(GCC_CFLAGS) $(TARGET_INCLUDES) $(MULTILIB_CFLAGS) -g0 \ + -finhibit-size-directive -fno-inline -fno-exceptions \ + -fno-zero-initialized-in-bss -fno-toplevel-reorder -fno-tree-vectorize \ + $(INHIBIT_LIBC_CFLAGS) +@@ -1038,10 +1038,14 @@ BUILD_ERRORS = build/errors.o + # -I$(@D) and -I$(srcdir)/$(@D) cause the subdirectory of the file + # currently being compiled, in both source trees, to be examined as well. + # libintl.h will be found in ../intl if we are using the included libintl. +-INCLUDES = -I. -I$(@D) -I$(srcdir) -I$(srcdir)/$(@D) \ ++# ++# TARGET_INCLUDES is added to avoid that GMPINC (which points to the host ++# include dir) is used for compiling libgcc.a ++TARGET_INCLUDES = -I. -I$(@D) -I$(srcdir) -I$(srcdir)/$(@D) \ + -I$(srcdir)/../include @INCINTL@ \ +- $(CPPINC) $(GMPINC) $(DECNUMINC) \ ++ $(CPPINC) $(DECNUMINC) \ + $(PPLINC) $(CLOOGINC) $(LIBELFINC) ++INCLUDES = $(TARGET_INCLUDES) $(GMPINC) + + .c.o: + $(COMPILER) -c $(ALL_COMPILERFLAGS) $(ALL_CPPFLAGS) $< $(OUTPUT_OPTION)recipes/gcc/gcc-4.5.2/arm-bswapsi2.patch new file +13
-0@@ -0,0 +1,13 @@ +Index: gcc-4.5/gcc/config/arm/arm.md +=================================================================== +--- gcc-4.5.orig/gcc/config/arm/arm.md 2010-06-17 09:13:07.000000000 -0700 ++++ gcc-4.5/gcc/config/arm/arm.md 2010-06-22 08:08:45.397212002 -0700 +@@ -11267,7 +11267,7 @@ + (define_expand "bswapsi2" + [(set (match_operand:SI 0 "s_register_operand" "=r") + (bswap:SI (match_operand:SI 1 "s_register_operand" "r")))] +-"TARGET_EITHER" ++"TARGET_EITHER && (arm_arch6 && !optimize_size)" + " + if (!arm_arch6) + {recipes/gcc/gcc-4.5.2/cache-amnesia.patch new file +31
-0@@ -0,0 +1,31 @@ +--- + gcc/configure | 2 +- + gcc/configure.ac | 2 +- + 2 files changed, 2 insertions(+), 2 deletions(-) + +Index: gcc-4.5+svnr155514/gcc/configure +=================================================================== +--- gcc-4.5+svnr155514.orig/gcc/configure 2009-12-29 22:00:40.000000000 -0800 ++++ gcc-4.5+svnr155514/gcc/configure 2009-12-29 23:52:43.381592113 -0800 +@@ -10467,7 +10467,7 @@ else + saved_CFLAGS="${CFLAGS}" + CC="${CC_FOR_BUILD}" CFLAGS="${CFLAGS_FOR_BUILD}" \ + LDFLAGS="${LDFLAGS_FOR_BUILD}" \ +- ${realsrcdir}/configure \ ++ CONFIG_SITE= ${realsrcdir}/configure --cache-file=./other.cache \ + --enable-languages=${enable_languages-all} \ + --target=$target_alias --host=$build_alias --build=$build_alias + CFLAGS="${saved_CFLAGS}" +Index: gcc-4.5+svnr155514/gcc/configure.ac +=================================================================== +--- gcc-4.5+svnr155514.orig/gcc/configure.ac 2009-12-29 22:00:40.000000000 -0800 ++++ gcc-4.5+svnr155514/gcc/configure.ac 2009-12-29 23:51:54.589091778 -0800 +@@ -1458,7 +1458,7 @@ else + saved_CFLAGS="${CFLAGS}" + CC="${CC_FOR_BUILD}" CFLAGS="${CFLAGS_FOR_BUILD}" \ + LDFLAGS="${LDFLAGS_FOR_BUILD}" \ +- ${realsrcdir}/configure \ ++ CONFIG_SITE= ${realsrcdir}/configure --cache-file=./other.cache \ + --enable-languages=${enable_languages-all} \ + --target=$target_alias --host=$build_alias --build=$build_alias + CFLAGS="${saved_CFLAGS}"recipes/gcc/gcc-4.5.2/fortran-cross-compile-hack.patch new file +30
-0@@ -0,0 +1,30 @@ +* Fortran would have searched for arm-angstrom-gnueabi-gfortran but would have used + used gfortan. For gcc_4.2.2.bb we want to use the gfortran compiler from our cross + directory. + +Index: gcc-4.5+svnr155514/libgfortran/configure +=================================================================== +--- gcc-4.5+svnr155514.orig/libgfortran/configure 2009-12-29 22:02:01.000000000 -0800 ++++ gcc-4.5+svnr155514/libgfortran/configure 2009-12-30 08:12:40.889091657 -0800 +@@ -11655,7 +11655,7 @@ CC="$lt_save_CC" + + # We need gfortran to compile parts of the library + #AC_PROG_FC(gfortran) +-FC="$GFORTRAN" ++#FC="$GFORTRAN" + ac_ext=${ac_fc_srcext-f} + ac_compile='$FC -c $FCFLAGS $ac_fcflags_srcext conftest.$ac_ext >&5' + ac_link='$FC -o conftest$ac_exeext $FCFLAGS $LDFLAGS $ac_fcflags_srcext conftest.$ac_ext $LIBS >&5' +Index: gcc-4.5+svnr155514/libgfortran/configure.ac +=================================================================== +--- gcc-4.5+svnr155514.orig/libgfortran/configure.ac 2009-12-29 22:02:01.000000000 -0800 ++++ gcc-4.5+svnr155514/libgfortran/configure.ac 2009-12-30 08:12:13.453094218 -0800 +@@ -187,7 +187,7 @@ AC_SUBST(enable_static) + + # We need gfortran to compile parts of the library + #AC_PROG_FC(gfortran) +-FC="$GFORTRAN" ++#FC="$GFORTRAN" + AC_PROG_FC(gfortran) + + # extra LD Flags which are required for targetsrecipes/gcc/gcc-4.5.2/gcc-4.3.1-ARCH_FLAGS_FOR_TARGET.patch new file +31
-0@@ -0,0 +1,31 @@ +--- + configure | 2 +- + configure.ac | 2 +- + 2 files changed, 2 insertions(+), 2 deletions(-) + +Index: gcc-4.3.1/configure.ac +=================================================================== +--- gcc-4.3.1.orig/configure.ac 2008-07-21 12:29:18.000000000 -0700 ++++ gcc-4.3.1/configure.ac 2008-07-21 12:29:35.000000000 -0700 +@@ -2352,7 +2352,7 @@ fi + # for target_alias and gcc doesn't manage it consistently. + target_configargs="--cache-file=./config.cache ${target_configargs}" + +-FLAGS_FOR_TARGET= ++FLAGS_FOR_TARGET="$ARCH_FLAGS_FOR_TARGET" + case " $target_configdirs " in + *" newlib "*) + case " $target_configargs " in +Index: gcc-4.3.1/configure +=================================================================== +--- gcc-4.3.1.orig/configure 2008-07-21 12:29:48.000000000 -0700 ++++ gcc-4.3.1/configure 2008-07-21 12:29:59.000000000 -0700 +@@ -5841,7 +5841,7 @@ fi + # for target_alias and gcc doesn't manage it consistently. + target_configargs="--cache-file=./config.cache ${target_configargs}" + +-FLAGS_FOR_TARGET= ++FLAGS_FOR_TARGET="$ARCH_FLAGS_FOR_TARGET" + case " $target_configdirs " in + *" newlib "*) + case " $target_configargs " inrecipes/gcc/gcc-4.5.2/gcc-armv4-pass-fix-v4bx-to-ld.patch new file +29
-0@@ -0,0 +1,29 @@ +The LINK_SPEC for linux gets overwritten by linux-eabi.h which +means the value of TARGET_FIX_V4BX_SPEC gets lost and as a result +the option is not passed to linker when chosing march=armv4 +This patch redefines this in linux-eabi.h and reinserts it +for eabi defaulting toolchains. + +We might want to send it upstream + +-Khem +Index: gcc-4.5/gcc/config/arm/linux-eabi.h +=================================================================== +--- gcc-4.5.orig/gcc/config/arm/linux-eabi.h ++++ gcc-4.5/gcc/config/arm/linux-eabi.h +@@ -63,10 +63,14 @@ + #undef GLIBC_DYNAMIC_LINKER + #define GLIBC_DYNAMIC_LINKER "/lib/ld-linux.so.3" + ++/* For armv4 we pass --fix-v4bx to linker to support EABI */ ++#undef TARGET_FIX_V4BX_SPEC ++#define TARGET_FIX_V4BX_SPEC " %{mcpu=arm8|mcpu=arm810|mcpu=strongarm*|march=armv4:--fix-v4bx}" ++ + /* At this point, bpabi.h will have clobbered LINK_SPEC. We want to + use the GNU/Linux version, not the generic BPABI version. */ + #undef LINK_SPEC +-#define LINK_SPEC LINUX_TARGET_LINK_SPEC BE8_LINK_SPEC ++#define LINK_SPEC LINUX_TARGET_LINK_SPEC BE8_LINK_SPEC TARGET_FIX_V4BX_SPEC + + /* Use the default LIBGCC_SPEC, not the version in linux-elf.h, as we + do not use -lfloat. */recipes/gcc/gcc-4.5.2/gcc-flags-for-build.patch new file +178
-0@@ -0,0 +1,178 @@ +Index: gcc-4.5/Makefile.def +=================================================================== +--- gcc-4.5.orig/Makefile.def ++++ gcc-4.5/Makefile.def +@@ -240,6 +240,7 @@ flags_to_pass = { flag= AWK ; }; + flags_to_pass = { flag= BISON ; }; + flags_to_pass = { flag= CC_FOR_BUILD ; }; + flags_to_pass = { flag= CFLAGS_FOR_BUILD ; }; ++flags_to_pass = { flag= CPPFLAGS_FOR_BUILD ; }; + flags_to_pass = { flag= CXX_FOR_BUILD ; }; + flags_to_pass = { flag= EXPECT ; }; + flags_to_pass = { flag= FLEX ; }; +Index: gcc-4.5/gcc/Makefile.in +=================================================================== +--- gcc-4.5.orig/gcc/Makefile.in ++++ gcc-4.5/gcc/Makefile.in +@@ -766,7 +766,7 @@ BUILD_LINKERFLAGS = $(BUILD_CFLAGS) + + # Native linker and preprocessor flags. For x-fragment overrides. + BUILD_LDFLAGS=@BUILD_LDFLAGS@ +-BUILD_CPPFLAGS=$(ALL_CPPFLAGS) ++BUILD_CPPFLAGS=$(INCLUDES) @BUILD_CPPFLAGS@ $(X_CPPFLAGS) + + # Actual name to use when installing a native compiler. + GCC_INSTALL_NAME := $(shell echo gcc|sed '$(program_transform_name)') +Index: gcc-4.5/gcc/configure.ac +=================================================================== +--- gcc-4.5.orig/gcc/configure.ac ++++ gcc-4.5/gcc/configure.ac +@@ -1798,16 +1798,18 @@ AC_SUBST(inhibit_libc) + # Also, we cannot run fixincludes. + + # These are the normal (build=host) settings: +-CC_FOR_BUILD='$(CC)' AC_SUBST(CC_FOR_BUILD) +-BUILD_CFLAGS='$(ALL_CFLAGS)' AC_SUBST(BUILD_CFLAGS) +-BUILD_LDFLAGS='$(LDFLAGS)' AC_SUBST(BUILD_LDFLAGS) +-STMP_FIXINC=stmp-fixinc AC_SUBST(STMP_FIXINC) ++CC_FOR_BUILD='$(CC)' AC_SUBST(CC_FOR_BUILD) ++BUILD_CFLAGS='$(ALL_CFLAGS)' AC_SUBST(BUILD_CFLAGS) ++BUILD_LDFLAGS='$(LDFLAGS)' AC_SUBST(BUILD_LDFLAGS) ++BUILD_CPPFLAGS='$(ALL_CPPFLAGS)' AC_SUBST(BUILD_CPPFLAGS) ++STMP_FIXINC=stmp-fixinc AC_SUBST(STMP_FIXINC) + + # And these apply if build != host, or we are generating coverage data + if test x$build != x$host || test "x$coverage_flags" != x + then + BUILD_CFLAGS='$(INTERNAL_CFLAGS) $(T_CFLAGS) $(CFLAGS_FOR_BUILD)' + BUILD_LDFLAGS='$(LDFLAGS_FOR_BUILD)' ++ BUILD_CPPFLAGS='$(CPPFLAGS_FOR_BUILD)' + fi + + # Expand extra_headers to include complete path. +Index: gcc-4.5/Makefile.in +=================================================================== +--- gcc-4.5.orig/Makefile.in ++++ gcc-4.5/Makefile.in +@@ -333,6 +333,7 @@ AR_FOR_BUILD = @AR_FOR_BUILD@ + AS_FOR_BUILD = @AS_FOR_BUILD@ + CC_FOR_BUILD = @CC_FOR_BUILD@ + CFLAGS_FOR_BUILD = @CFLAGS_FOR_BUILD@ ++CPPFLAGS_FOR_BUILD = @CPPFLAGS_FOR_BUILD@ + CXXFLAGS_FOR_BUILD = @CXXFLAGS_FOR_BUILD@ + CXX_FOR_BUILD = @CXX_FOR_BUILD@ + DLLTOOL_FOR_BUILD = @DLLTOOL_FOR_BUILD@ +@@ -662,6 +663,7 @@ BASE_FLAGS_TO_PASS = \ + "BISON=$(BISON)" \ + "CC_FOR_BUILD=$(CC_FOR_BUILD)" \ + "CFLAGS_FOR_BUILD=$(CFLAGS_FOR_BUILD)" \ ++ "CPPFLAGS_FOR_BUILD=$(CPPFLAGS_FOR_BUILD)" \ + "CXX_FOR_BUILD=$(CXX_FOR_BUILD)" \ + "EXPECT=$(EXPECT)" \ + "FLEX=$(FLEX)" \ +Index: gcc-4.5/gcc/configure +=================================================================== +--- gcc-4.5.orig/gcc/configure ++++ gcc-4.5/gcc/configure +@@ -707,6 +707,7 @@ SED + LIBTOOL + collect2 + STMP_FIXINC ++BUILD_CPPFLAGS + BUILD_LDFLAGS + BUILD_CFLAGS + CC_FOR_BUILD +@@ -10982,6 +10983,7 @@ fi + CC_FOR_BUILD='$(CC)' + BUILD_CFLAGS='$(ALL_CFLAGS)' + BUILD_LDFLAGS='$(LDFLAGS)' ++BUILD_CPPFLAGS='$(ALL_CPPFLAGS)' + STMP_FIXINC=stmp-fixinc + + # And these apply if build != host, or we are generating coverage data +@@ -10989,6 +10991,7 @@ if test x$build != x$host || test "x$cov + then + BUILD_CFLAGS='$(INTERNAL_CFLAGS) $(T_CFLAGS) $(CFLAGS_FOR_BUILD)' + BUILD_LDFLAGS='$(LDFLAGS_FOR_BUILD)' ++ BUILD_CPPFLAGS='$(CPPFLAGS_FOR_BUILD)' + fi + + # Expand extra_headers to include complete path. +@@ -17108,7 +17111,7 @@ else + lt_dlunknown=0; lt_dlno_uscore=1; lt_dlneed_uscore=2 + lt_status=$lt_dlunknown + cat > conftest.$ac_ext <<_LT_EOF +-#line 17111 "configure" ++#line 17114 "configure" + #include "confdefs.h" + + #if HAVE_DLFCN_H +@@ -17214,7 +17217,7 @@ else + lt_dlunknown=0; lt_dlno_uscore=1; lt_dlneed_uscore=2 + lt_status=$lt_dlunknown + cat > conftest.$ac_ext <<_LT_EOF +-#line 17217 "configure" ++#line 17220 "configure" + #include "confdefs.h" + + #if HAVE_DLFCN_H +Index: gcc-4.5/Makefile.tpl +=================================================================== +--- gcc-4.5.orig/Makefile.tpl ++++ gcc-4.5/Makefile.tpl +@@ -336,6 +336,7 @@ AR_FOR_BUILD = @AR_FOR_BUILD@ + AS_FOR_BUILD = @AS_FOR_BUILD@ + CC_FOR_BUILD = @CC_FOR_BUILD@ + CFLAGS_FOR_BUILD = @CFLAGS_FOR_BUILD@ ++CPPFLAGS_FOR_BUILD = @CPPFLAGS_FOR_BUILD@ + CXXFLAGS_FOR_BUILD = @CXXFLAGS_FOR_BUILD@ + CXX_FOR_BUILD = @CXX_FOR_BUILD@ + DLLTOOL_FOR_BUILD = @DLLTOOL_FOR_BUILD@ +Index: gcc-4.5/configure +=================================================================== +--- gcc-4.5.orig/configure ++++ gcc-4.5/configure +@@ -651,6 +651,7 @@ GCJ_FOR_BUILD + DLLTOOL_FOR_BUILD + CXX_FOR_BUILD + CXXFLAGS_FOR_BUILD ++CPPFLAGS_FOR_BUILD + CFLAGS_FOR_BUILD + CC_FOR_BUILD + AS_FOR_BUILD +@@ -8036,6 +8037,7 @@ esac + # our build compiler if desired. + if test x"${build}" = x"${host}" ; then + CFLAGS_FOR_BUILD=${CFLAGS_FOR_BUILD-${CFLAGS}} ++ CPPFLAGS_FOR_BUILD=${CPPFLAGS_FOR_BUILD-${CPPFLAGS}} + CXXFLAGS_FOR_BUILD=${CXXFLAGS_FOR_BUILD-${CXXFLAGS}} + LDFLAGS_FOR_BUILD=${LDFLAGS_FOR_BUILD-${LDFLAGS}} + fi +@@ -8101,6 +8103,7 @@ done + + + ++ + + + +Index: gcc-4.5/configure.ac +=================================================================== +--- gcc-4.5.orig/configure.ac ++++ gcc-4.5/configure.ac +@@ -3089,6 +3089,7 @@ esac + # our build compiler if desired. + if test x"${build}" = x"${host}" ; then + CFLAGS_FOR_BUILD=${CFLAGS_FOR_BUILD-${CFLAGS}} ++ CPPFLAGS_FOR_BUILD=${CPPFLAGS_FOR_BUILD-${CPPFLAGS}} + CXXFLAGS_FOR_BUILD=${CXXFLAGS_FOR_BUILD-${CXXFLAGS}} + LDFLAGS_FOR_BUILD=${LDFLAGS_FOR_BUILD-${LDFLAGS}} + fi +@@ -3155,6 +3156,7 @@ AC_SUBST(AR_FOR_BUILD) + AC_SUBST(AS_FOR_BUILD) + AC_SUBST(CC_FOR_BUILD) + AC_SUBST(CFLAGS_FOR_BUILD) ++AC_SUBST(CPPFLAGS_FOR_BUILD) + AC_SUBST(CXXFLAGS_FOR_BUILD) + AC_SUBST(CXX_FOR_BUILD) + AC_SUBST(DLLTOOL_FOR_BUILD)recipes/gcc/gcc-4.5.2/gcc-scalar-widening-pr45847.patch new file +63
-0@@ -0,0 +1,63 @@ +Hi, + +The attached patch fixes Bugzilla 45847 +(http://gcc.gnu.org/bugzilla/show_bug.cgi?id=45847). When compiling +without -mvectorize-with-neon-quad and vectorizing scalar widening +operations that widen words to double words, there are no corresponding +vector types for DI scalar types. For this scenario, a call to +get_vect_type_for_scalar_type() returns NULL and an absent NULL-check +caused this segfault. The attached patch adds this NULL-check. Also, +this is consistent with all the other places where a NULL-check follows +a call to get_vect_type_for_scalar_type() in tree-vect-patterns.c. + +Regression tested with arm-linux-gnueabi. OK? + +-- +Tejas Belagod +ARM. + +gcc/ + +2010-10-05 Tejas Belagod <[email protected]> + + * tree-vect-patterns.c (vect_recog_widen_mult_pattern): Add NULL + check for vectype_out returned by get_vectype_for_scalar_type(). + +testsuite/ + +2010-10-05 Tejas Belagod <[email protected]> + + * gcc.dg/vect/pr45847.c: New test. + +Index: gcc-4.5/gcc/testsuite/gcc.dg/vect/pr45847.c +=================================================================== +--- /dev/null ++++ gcc-4.5/gcc/testsuite/gcc.dg/vect/pr45847.c +@@ -0,0 +1,15 @@ ++/* { dg-do compile } */ ++ ++ ++long long foo (long long *__restrict a, int *__restrict b, int *__restrict c ) ++{ ++ int i; ++ long long sum=0; ++ for (i=0;i<256;i++) ++ sum += (long long)b[i] * c[i]; ++ ++ return sum; ++} ++ ++/* { dg-final { cleanup-tree-dump "vect" } } */ ++ +Index: gcc-4.5/gcc/tree-vect-patterns.c +=================================================================== +--- gcc-4.5.orig/gcc/tree-vect-patterns.c ++++ gcc-4.5/gcc/tree-vect-patterns.c +@@ -411,6 +411,7 @@ vect_recog_widen_mult_pattern (gimple la + /* Check target support */ + vectype = get_vectype_for_scalar_type (half_type0); + if (!vectype ++ || !get_vectype_for_scalar_type (type) + || !supportable_widening_operation (WIDEN_MULT_EXPR, last_stmt, vectype, + &dummy, &dummy, &dummy_code, + &dummy_code, &dummy_int, &dummy_vec))recipes/gcc/gcc-4.5.2/gcc-uclibc-locale-ctype_touplow_t.patch new file +67
-0@@ -0,0 +1,67 @@ +Index: gcc-4.5/libstdc++-v3/config/locale/generic/c_locale.h +=================================================================== +--- gcc-4.5.orig/libstdc++-v3/config/locale/generic/c_locale.h 2010-06-30 22:30:53.993316002 -0700 ++++ gcc-4.5/libstdc++-v3/config/locale/generic/c_locale.h 2010-06-30 22:31:26.043316001 -0700 +@@ -41,12 +41,17 @@ + + #include <clocale> + #include <cstddef> ++#include <features.h> ++#include <ctype.h> + + #define _GLIBCXX_NUM_CATEGORIES 0 + + _GLIBCXX_BEGIN_NAMESPACE(std) +- +- typedef int* __c_locale; ++#ifdef __UCLIBC__ ++ typedef __ctype_touplow_t* __c_locale; ++#else ++ typedef int* __c_locale; ++#endif + + // Convert numeric value of type double and long double to string and + // return length of string. If vsnprintf is available use it, otherwise +Index: gcc-4.5/libstdc++-v3/config/os/gnu-linux/ctype_base.h +=================================================================== +--- gcc-4.5.orig/libstdc++-v3/config/os/gnu-linux/ctype_base.h 2010-06-30 22:30:54.013316002 -0700 ++++ gcc-4.5/libstdc++-v3/config/os/gnu-linux/ctype_base.h 2010-06-30 22:31:26.053316001 -0700 +@@ -33,14 +33,21 @@ + */ + + // Information as gleaned from /usr/include/ctype.h +- ++ ++#include <features.h> ++#include <ctype.h> ++ + _GLIBCXX_BEGIN_NAMESPACE(std) + + /// @brief Base class for ctype. + struct ctype_base + { + // Non-standard typedefs. +- typedef const int* __to_type; ++#ifdef __UCLIBC__ ++ typedef const __ctype_touplow_t* __to_type; ++#else ++ typedef const int* __to_type; ++#endif + + // NB: Offsets into ctype<char>::_M_table force a particular size + // on the mask type. Because of this, we don't use an enum. +Index: gcc-4.5/libstdc++-v3/config/locale/generic/c_locale.cc +=================================================================== +--- gcc-4.5.orig/libstdc++-v3/config/locale/generic/c_locale.cc 2010-06-28 12:12:42.000000000 -0700 ++++ gcc-4.5/libstdc++-v3/config/locale/generic/c_locale.cc 2010-06-30 22:31:26.063316001 -0700 +@@ -256,5 +256,10 @@ _GLIBCXX_END_NAMESPACE + #ifdef _GLIBCXX_LONG_DOUBLE_COMPAT + #define _GLIBCXX_LDBL_COMPAT(dbl, ldbl) \ + extern "C" void ldbl (void) __attribute__ ((alias (#dbl))) ++#ifdef __UCLIBC__ ++// This is because __c_locale is of type __ctype_touplow_t* which is short on uclibc. for glibc its int* ++_GLIBCXX_LDBL_COMPAT(_ZSt14__convert_to_vIdEvPKcRT_RSt12_Ios_IostateRKPs, _ZSt14__convert_to_vIeEvPKcRT_RSt12_Ios_IostateRKPs); ++#else + _GLIBCXX_LDBL_COMPAT(_ZSt14__convert_to_vIdEvPKcRT_RSt12_Ios_IostateRKPi, _ZSt14__convert_to_vIeEvPKcRT_RSt12_Ios_IostateRKPi); ++#endif + #endif // _GLIBCXX_LONG_DOUBLE_COMPATrecipes/gcc/gcc-4.5.2/libstdc++-emit-__cxa_end_cleanup-in-text.patch new file +40
-0@@ -0,0 +1,40 @@ +2010-06-07 Khem Raj <[email protected]> + + * libsupc++/eh_arm.cc (__cxa_end_cleanup): Use .pushsection/.popsection + to emit inline assembly into .text section. + +Index: gcc-4.5/libstdc++-v3/libsupc++/eh_arm.cc +=================================================================== +--- gcc-4.5.orig/libstdc++-v3/libsupc++/eh_arm.cc 2010-06-04 23:20:18.000000000 -0700 ++++ gcc-4.5/libstdc++-v3/libsupc++/eh_arm.cc 2010-06-08 11:27:34.247541722 -0700 +@@ -157,22 +157,26 @@ __gnu_end_cleanup(void) + // Assembly wrapper to call __gnu_end_cleanup without clobbering r1-r3. + // Also push r4 to preserve stack alignment. + #ifdef __thumb__ +-asm (".global __cxa_end_cleanup\n" ++asm (" .pushsection .text.__cxa_end_cleanup\n" ++" .global __cxa_end_cleanup\n" + " .type __cxa_end_cleanup, \"function\"\n" + " .thumb_func\n" + "__cxa_end_cleanup:\n" + " push\t{r1, r2, r3, r4}\n" + " bl\t__gnu_end_cleanup\n" + " pop\t{r1, r2, r3, r4}\n" +-" bl\t_Unwind_Resume @ Never returns\n"); ++" bl\t_Unwind_Resume @ Never returns\n" ++" .popsection\n"); + #else +-asm (".global __cxa_end_cleanup\n" ++asm (" .pushsection .text.__cxa_end_cleanup\n" ++" .global __cxa_end_cleanup\n" + " .type __cxa_end_cleanup, \"function\"\n" + "__cxa_end_cleanup:\n" + " stmfd\tsp!, {r1, r2, r3, r4}\n" + " bl\t__gnu_end_cleanup\n" + " ldmfd\tsp!, {r1, r2, r3, r4}\n" +-" bl\t_Unwind_Resume @ Never returns\n"); ++" bl\t_Unwind_Resume @ Never returns\n" ++" .popsection\n"); + #endif + + #endifrecipes/gcc/gcc-4.5.2/mips64-nomultilib.patch new file +52
-0@@ -0,0 +1,52 @@ +Index: gcc-4.5/gcc/config/mips/linux64.h +=================================================================== +--- gcc-4.5.orig/gcc/config/mips/linux64.h 2010-09-25 02:05:05.484423095 -0700 ++++ gcc-4.5/gcc/config/mips/linux64.h 2010-09-25 02:31:18.524931014 -0700 +@@ -26,7 +26,7 @@ + BASE_DRIVER_SELF_SPECS, \ + LINUX_DRIVER_SELF_SPECS \ + " %{!EB:%{!EL:%(endian_spec)}}" \ +- " %{!mabi=*: -mabi=n32}" ++ " %{!mabi=*: -mabi=64}" + + #undef LIB_SPEC + #define LIB_SPEC "\ +@@ -35,9 +35,9 @@ + %{!shared: \ + %{profile:-lc_p} %{!profile:-lc}}" + +-#define GLIBC_DYNAMIC_LINKER32 "/lib/ld.so.1" +-#define GLIBC_DYNAMIC_LINKER64 "/lib64/ld.so.1" +-#define GLIBC_DYNAMIC_LINKERN32 "/lib32/ld.so.1" ++#define GLIBC_DYNAMIC_LINKER32 "/lib32/ld.so.1" ++#define GLIBC_DYNAMIC_LINKER64 "/lib/ld.so.1" ++#define GLIBC_DYNAMIC_LINKERN32 "/lib64/ld.so.1" + #define UCLIBC_DYNAMIC_LINKERN32 "/lib32/ld-uClibc.so.0" + #define LINUX_DYNAMIC_LINKERN32 \ + CHOOSE_DYNAMIC_LINKER (GLIBC_DYNAMIC_LINKERN32, UCLIBC_DYNAMIC_LINKERN32) +Index: gcc-4.5/gcc/config.gcc +=================================================================== +--- gcc-4.5.orig/gcc/config.gcc 2010-07-22 16:37:17.000000000 -0700 ++++ gcc-4.5/gcc/config.gcc 2010-09-25 02:25:41.412414136 -0700 +@@ -1707,7 +1707,7 @@ + *-*-irix6*) + tm_file="${tm_file} mips/iris6.h" + tmake_file="${tmake_file} mips/t-iris6" +- tm_defines="${tm_defines} MIPS_ISA_DEFAULT=3 MIPS_ABI_DEFAULT=ABI_N32" ++ tm_defines="${tm_defines} MIPS_ISA_DEFAULT=3 MIPS_ABI_DEFAULT=ABI_64" + case ${target} in + *-*-irix6.[0-4]*) + use_gcc_stdint=provide +Index: gcc-4.5/gcc/config/mips/t-linux64 +=================================================================== +--- gcc-4.5.orig/gcc/config/mips/t-linux64 2010-07-11 16:14:42.000000000 -0700 ++++ gcc-4.5/gcc/config/mips/t-linux64 2010-09-25 02:29:52.758708250 -0700 +@@ -18,7 +18,7 @@ + + MULTILIB_OPTIONS = mabi=n32/mabi=32/mabi=64 + MULTILIB_DIRNAMES = n32 32 64 +-MULTILIB_OSDIRNAMES = ../lib32 ../lib ../lib64 ++MULTILIB_OSDIRNAMES = ../lib64 ../lib32 ../lib + + EXTRA_MULTILIB_PARTS=crtbegin.o crtend.o crtbeginS.o crtendS.o crtbeginT.o +recipes/gcc/gcc-4.5.2/zecke-no-host-includes.patch new file +31
-0@@ -0,0 +1,31 @@ +Index: gcc-4.4+svnr145550/gcc/incpath.c +=================================================================== +--- gcc-4.4+svnr145550.orig/gcc/incpath.c 2009-04-04 13:48:31.000000000 -0700 ++++ gcc-4.4+svnr145550/gcc/incpath.c 2009-04-04 14:49:29.000000000 -0700 +@@ -417,6 +417,26 @@ + p->construct = 0; + p->user_supplied_p = user_supplied_p; + ++#ifdef CROSS_COMPILE ++ /* A common error when cross compiling is including ++ host headers. This code below will try to fail fast ++ for cross compiling. Currently we consider /usr/include, ++ /opt/include and /sw/include as harmful. */ ++ { ++ /* printf("Adding Path: %s\n", p->name ); */ ++ if( strstr(p->name, "/usr/include" ) == p->name ) { ++ fprintf(stderr, _("CROSS COMPILE Badness: /usr/include in INCLUDEPATH: %s\n"), p->name); ++ abort(); ++ } else if( strstr(p->name, "/sw/include") == p->name ) { ++ fprintf(stderr, _("CROSS COMPILE Badness: /sw/include in INCLUDEPATH: %s\n"), p->name); ++ abort(); ++ } else if( strstr(p->name, "/opt/include") == p->name ) { ++ fprintf(stderr, _("CROSS COMPILE Badness: /opt/include in INCLUDEPATH: %s\n"), p->name); ++ abort(); ++ } ++ } ++#endif ++ + add_cpp_dir_path (p, chain); + } +recipes/gcc/gcc-4.5.2/zecke-xgcc-cpp.patch new file +28
-0@@ -0,0 +1,28 @@ +upstream: n/a +comment: Use the preprocessor we have just compiled instead the one of +the system. There might be incompabilities between us and them. + +Index: gcc-4.3.1/Makefile.in +=================================================================== +--- gcc-4.3.1.orig/Makefile.in 2008-08-19 01:09:56.000000000 -0700 ++++ gcc-4.3.1/Makefile.in 2008-08-19 01:13:27.000000000 -0700 +@@ -204,6 +204,7 @@ + AR="$(AR_FOR_TARGET)"; export AR; \ + AS="$(COMPILER_AS_FOR_TARGET)"; export AS; \ + CC="$(CC_FOR_TARGET)"; export CC; \ ++ CPP="$(CC_FOR_TARGET) -E"; export CPP; \ + CFLAGS="$(CFLAGS_FOR_TARGET) $(SYSROOT_CFLAGS_FOR_TARGET) $(DEBUG_PREFIX_CFLAGS_FOR_TARGET)"; export CFLAGS; \ + CONFIG_SHELL="$(SHELL)"; export CONFIG_SHELL; \ + CPPFLAGS="$(CPPFLAGS_FOR_TARGET)"; export CPPFLAGS; \ +Index: gcc-4.3.1/Makefile.tpl +=================================================================== +--- gcc-4.3.1.orig/Makefile.tpl 2008-08-21 00:07:58.000000000 -0700 ++++ gcc-4.3.1/Makefile.tpl 2008-08-21 00:09:52.000000000 -0700 +@@ -223,6 +223,7 @@ + AR="$(AR_FOR_TARGET)"; export AR; \ + AS="$(COMPILER_AS_FOR_TARGET)"; export AS; \ + CC="$(CC_FOR_TARGET)"; export CC; \ ++ CPP="$(CC_FOR_TARGET) -E"; export CPP; \ + CFLAGS="$(CFLAGS_FOR_TARGET) $(SYSROOT_CFLAGS_FOR_TARGET) $(DEBUG_PREFIX_CFLAGS_FOR_TARGET)"; export CFLAGS; \ + CONFIG_SHELL="$(SHELL)"; export CONFIG_SHELL; \ + CPPFLAGS="$(CPPFLAGS_FOR_TARGET)"; export CPPFLAGS; \recipes/gcc/gcc-cross-initial_4.5.2.bb new file +2
-0@@ -0,0 +1,2 @@ +require gcc-cross_${PV}.bb +require gcc-cross-initial.increcipes/gcc/gcc-cross-intermediate_4.5.2.bb new file +2
-0@@ -0,0 +1,2 @@ +require gcc-cross_${PV}.bb +require gcc-cross-intermediate.increcipes/gcc/gcc-cross-sdk_4.5.2.bb new file +8
-0@@ -0,0 +1,8 @@ +inherit sdk + +require gcc-${PV}.inc +require gcc-cross-sdk.inc +PR = "${INC_PR}.1" + +EXTRA_OECONF += " --disable-libunwind-exceptions --with-mpfr=${STAGING_DIR_NATIVE}${prefix_native} --with-system-zlib" +recipes/gcc/gcc-cross_4.5.2.bb new file +12
-0@@ -0,0 +1,12 @@ +require gcc-${PV}.inc +require gcc-cross4.inc + +PR = "${INC_PR}.0" + +NATIVEDEPS += "libmpc-native libelf-native" +SRC_URI_append_fail-fast = " file://zecke-no-host-includes.patch " + +EXTRA_OECONF += " --disable-libunwind-exceptions --with-mpfr=${STAGING_DIR_NATIVE}${prefix_native} \ + -with-libelf=${STAGING_DIR_NATIVE}${prefix_native} --with-system-zlib" + +ARCH_FLAGS_FOR_TARGET += "-isystem${STAGING_DIR_TARGET}${target_includedir}"recipes/gcc/gcc_4.5.2.bb new file +15
-1@@ -0,0 +1,15 @@ +PR = "${INC_PR}.1" +require gcc-${PV}.inc +require gcc-configure-target.inc +require gcc-package-target.inc + +# Gcc 4.3.3 installs crt* in a '4.3.1' dir.... +FILES_${PN} += "\ + ${gcclibdir}/${TARGET_SYS}/*/*.o \ + ${libexecdir}/gcc/${TARGET_SYS}/${BINV}/lto1 \ + ${libexecdir}/gcc/${TARGET_SYS}/${BINV}/lto-wrapper \ +" + +SRC_URI_append = "file://fortran-cross-compile-hack.patch" + +ARCH_FLAGS_FOR_TARGET += "-isystem${STAGING_INCDIR}" --1.6.3.3
_______________________________________________
Openembedded-devel mailing list
[email protected]
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
Quoted 10 lines · Khem Raj
Quoted 6 lines · Eric Bénard
- based on gcc-4.5.2-RC-20101208 (svn 167585)
- without linaro patches (which seem to be the root of
problems when compiling at least on armv4 and armv5) :
- samba 3.2.15 or 3.3.9 : http://tinderbox.openembedded.net/public/logs/task/8666826.txt
- samba 3.5.6 : http://pastebin.com/yuiYX2CMas discussed on IRC may be we should try to find the faulty patch in current gcc 4.5 linaro patches we have.
I don't think that we need another svn-based pre-release recipe for gcc. But I would still prefer to have a recipe for gcc 4.5.2 once it's released, because I'd like to be able to pin a stable version with only few required patches and Linaro only cares about ARM while I do not.
Regards,
Andreas
Quoted 9 lines · Andreas Oberritter
Quoted 8 lines · Khem Raj
Quoted 6 lines · Eric Bénard
- based on gcc-4.5.2-RC-20101208 (svn 167585)
- without linaro patches (which seem to be the root of
problems when compiling at least on armv4 and armv5) :
- samba 3.2.15 or 3.3.9 : http://tinderbox.openembedded.net/public/logs/task/8666826.txt
- samba 3.5.6 : http://pastebin.com/yuiYX2CM
Quoted 8 lines
Quoted 2 lines
as discussed on IRC may be we should try to find the faulty patch in current gcc 4.5 linaro patches we have.
I don't think that we need another svn-based pre-release recipe for gcc. But I would still prefer to have a recipe for gcc 4.5.2 once it's released, because I'd like to be able to pin a stable version with only few required patches and Linaro only cares about ARM while I do not.
We should avoid creating recipes for every minor release. As we see minor release only contain bug fixes.
agreed. If you dont care about linaro thats fine. Would you prefer a gcc which is widely used or just used by you
hence supported only by you. My point it more common we use better we have the support for it and community as a whole gets benefitted. If linaro patches dont
harm your case negatively would you still be averse to them ? Do you have some cases where they created regressions compile time or run time or degraded the performance ?
that will be interesting.
why do you think gcc-4.5.2 will be more stable then say gcc-4.5.2 + patches from upstream 4.5 branch
What advantage do you see of having a vanilla gcc without linaro patches ?
Quoted 8 lines
Regards,
Andreas_______________________________________________
Openembedded-devel mailing list
[email protected]
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
Quoted 14 lines · Khem Raj
Quoted 12 lines · Andreas Oberritter
Quoted 10 lines · Khem Raj
Quoted 6 lines · Eric Bénard
- based on gcc-4.5.2-RC-20101208 (svn 167585)
- without linaro patches (which seem to be the root of
problems when compiling at least on armv4 and armv5) :
- samba 3.2.15 or 3.3.9 : http://tinderbox.openembedded.net/public/logs/task/8666826.txt
- samba 3.5.6 : http://pastebin.com/yuiYX2CMas discussed on IRC may be we should try to find the faulty patch in current gcc 4.5 linaro patches we have.
Quoted 8 lines
Quoted 5 lines
I don't think that we need another svn-based pre-release recipe for gcc. But I would still prefer to have a recipe for gcc 4.5.2 once it's released, because I'd like to be able to pin a stable version with only few required patches and Linaro only cares about ARM while I do not.
We should avoid creating recipes for every minor release. As we see minor release only contain bug fixes.
What's wrong with bug fixes in minor releases? Btw., I have no problem moving from one minor release to another, which means that I would be happy with only one recipe for every major release.
Quoted 3 lines
agreed. If you dont care about linaro thats fine. Would you prefer a gcc which is widely used or just used by you
hence supported only by you.
I'd prefer a gcc which is widely used. We just seem to have different opinions on what that means. IMO, a released version will surely be used by many more people globally (i.e. outside OE) than a random svn version of a linaro gcc. Also, since Eric already sent this patch, I'm sure that I'm not the only one who'd prefer to have a recipe for a current, released version in OE.
While gcc*_4.5.bb is widely in use (well, because it's the default recipe and there is no alternative for 4.5 yet), its revision and linaro patch level changes every other day. It went from r10 to r25 during the last 10 weeks. This wastes a lot of time rebuilding the compiler, or - if you don't want to mix different gcc's output - rebuilding the whole system. In comparison, gcc-4.4.4 is still at r7.
Quoted 7 lines
My point it more common we use better we
have the support for it and community as a whole gets benefitted. If linaro patches dont
harm your case negatively would you still be averse to them ? Do you have some cases where they created regressions compile time or run time or degraded the performance ?
that will be interesting.
I haven't discovered any regressions by the compiler itself, but I also haven't discovered any benefits.
My main concern is that a weekly changing compiler can hardly ever become a good foundation of a stable build environment. Of course I could pin a random SRCREV, but then again I'm likely to become the only user of this specific SRCREV globally, some day. Oh, wait, I can not pin a SRCREV, because the patches keep getting added as local files.
Quoted 2 lines
why do you think gcc-4.5.2 will be more stable then say gcc-4.5.2 + patches from upstream 4.5 branch
It depends on the patches. Most of them need testing. Releases (and their candidates) already would have been tested and/or reviewed by some people before hitting OE.
As already stated in my previous mail, I'm not against applying patches from upstream, provided they are _required_ to make it work in some configuration.
Quoted 1 line
What advantage do you see of having a vanilla gcc without linaro patches ?
See above.
In the (hopefully unlikely) event that a random svn version of linaro gcc introduces a regression, say on the MIPS platform, who am I going to report that to? Would I have to bisect all linaro patches manually? Surely, the upstream developers wouldn't care about reports about heavily patched gccs, and linaro developers wouldn't care about MIPS.
Regards,
Andreas
Quoted 28 lines · Andreas Oberritter
Quoted 14 lines · Khem Raj
Quoted 12 lines · Andreas Oberritter
Quoted 10 lines · Khem Raj
Quoted 6 lines · Eric Bénard
- based on gcc-4.5.2-RC-20101208 (svn 167585)
- without linaro patches (which seem to be the root of
problems when compiling at least on armv4 and armv5) :
- samba 3.2.15 or 3.3.9 : http://tinderbox.openembedded.net/public/logs/task/8666826.txt
- samba 3.5.6 : http://pastebin.com/yuiYX2CMas discussed on IRC may be we should try to find the faulty patch in current gcc 4.5 linaro patches we have.
Quoted 8 lines
Quoted 5 lines
I don't think that we need another svn-based pre-release recipe for gcc. But I would still prefer to have a recipe for gcc 4.5.2 once it's released, because I'd like to be able to pin a stable version with only few required patches and Linaro only cares about ARM while I do not.
We should avoid creating recipes for every minor release. As we see minor release only contain bug fixes.
What's wrong with bug fixes in minor releases? Btw., I have no problem moving from one minor release to another, which means that I would be happy with only one recipe for every major release.
Quoted 8 lines
Quoted 3 lines
agreed. If you dont care about linaro thats fine. Would you prefer a gcc which is widely used or just used by you
hence supported only by you.I'd prefer a gcc which is widely used. We just seem to have different opinions on what that means. IMO, a released version will surely be used by many more people globally (i.e. outside OE) than a random svn version of a linaro gcc.
its not random svn version and certainly not of linaro.
Its a revision from release branch which means
its the latest release say 4.5.1 + fixes to this branch up until that revision. Also note that I rely upon others from community to test and verify patches. Its impossible for me or anybody to test the patches for every combination but important is that we react to regressions and we fix them. If it was ignored then I would be worried.
Also, since Eric already sent this patch, I'm sure that
Quoted 9 lines
I'm not the only one who'd prefer to have a recipe for a current, released version in OE.
While gcc*_4.5.bb is widely in use (well, because it's the default recipe and there is no alternative for 4.5 yet), its revision and linaro patch level changes every other day. It went from r10 to r25 during the last 10 weeks. This wastes a lot of time rebuilding the compiler, or - if you don't want to mix different gcc's output - rebuilding the whole system. In comparison, gcc-4.4.4 is still at r7.
While its a big feature pull so its expected a bit of developement. and since we have a release and testing tags, that could be used. and gives you more no churn at all and you can happily use it and then move forward at longer intervals and you dont have to rebuild often but if you just hate to build gcc and therefore would like to stop its developement of these recipes is probably not achievable. I am not averse to adding whatever version you assume will be stable for you its just that it adds to maintenance burden and we already have a lot of recipes for toolchain.
Quoted 9 lines
Quoted 8 lines
My point it more common we use better we
have the support for it and community as a whole gets benefitted. If linaro patches dont
harm your case negatively would you still be averse to them ? Do you have some cases where they created regressions compile time or run time or degraded the performance ?
that will be interesting.
Quoted 6 lines
I haven't discovered any regressions by the compiler itself, but I also haven't discovered any benefits.
My main concern is that a weekly changing compiler can hardly ever become a good foundation of a stable build environment.
yes it wont and I think the churn should reduce as we have stabilize after all its a .dev branch and even compilers are softwares.
Of course I
Quoted 3 lines
could pin a random SRCREV, but then again I'm likely to become the only user of this specific SRCREV globally, some day. Oh, wait, I can not pin a SRCREV, because the patches keep getting added as local files.
yes they are local because
Quoted 8 lines
Quoted 3 lines
why do you think gcc-4.5.2 will be more stable then say gcc-4.5.2 + patches from upstream 4.5 branch
It depends on the patches. Most of them need testing. Releases (and their candidates) already would have been tested and/or reviewed by some people before hitting OE.
in my experience OE finds bugs in any release. The amount of software we compile is huge and way we compile is unique. Thats no different than adding patches from Linaro or any body
Quoted 6 lines
As already stated in my previous mail, I'm not against applying patches from upstream, provided they are _required_ to make it work in some configuration.
Quoted 1 line
What advantage do you see of having a vanilla gcc without linaro patches ?
Quoted 8 lines
See above.
In the (hopefully unlikely) event that a random svn version of linaro gcc introduces a regression, say on the MIPS platform, who am I going to report that to? Would I have to bisect all linaro patches manually? Surely, the upstream developers wouldn't care about reports about heavily patched gccs, and linaro developers wouldn't care about MIPS.
Quite a lot of OE users are right now using armv7 in some form so I think linaro patches are very helpful to them.
Well as a humble community member I try my best and so does many of us and we find the problem and fix them. If we find regressions then we cant fix them then we backout the faulty.
Quoted 7 lines
Regards,
Andreas_______________________________________________
Openembedded-devel mailing list
[email protected]
http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-devel
Hi Khem, hi Andreas,
Quoted 9 lines · Khem Raj
Quoted 6 lines · Andreas Oberritter
In the (hopefully unlikely) event that a random svn version of linaro gcc introduces a regression, say on the MIPS platform, who am I going to report that to? Would I have to bisect all linaro patches manually? Surely, the upstream developers wouldn't care about reports about heavily patched gccs, and linaro developers wouldn't care about MIPS.
Quite a lot of OE users are right now using armv7 in some form so I think linaro patches are very helpful to them.
Linaro is having an interesting discussion on the linaro-toolchain list about this :
http://lists.linaro.org/pipermail/linaro-toolchain/2010-December/000650.html
http://lists.linaro.org/pipermail/linaro-toolchain/2010-December/000651.html
Eric