gcc: Add patch to resolve i*86 tune configuration overrides
PATCH 1/1 gcc: Add patch to resolve i*86 tune configuration overrides
When compiling gcc-runtime for i686 libatomic, libgomp and libitm are compiled with additional '-march' and '-mtune' overrides. This typically does not cause build errors (e.g. core2-32 tune), and results in less optimally tuned outputs. However with specific TUNE_CCARGS (e.g. has '-mfpmath=sse' but not '-msse*'), the build errors with conflicting '-mfpmath' options.
This is caused by gcc-runtime specific configure scripts and Makefile options that override by adding additional '-march'/'-mtune' args. OE builds gcc-runtime differently to other build systems, and provides target specific arch/tune options directly. As such, this change adds a patch for gcc to remove the specific cases where the args are added.
--- meta/recipes-devtools/gcc/gcc-10.2.inc | 1 + .../gcc/0001-lib-Remove-i-86-march-overrides.patch | 113 +++++++++++++++++++++ 2 files changed, 114 insertions(+) create mode 100644 meta/recipes-devtools/gcc/gcc/0001-lib-Remove-i-86-march-overrides.patch
meta/recipes-devtools/gcc/gcc-10.2.inc +1 -0
@@ -69,6 +69,7 @@ SRC_URI = "\ file://0002-aarch64-Introduce-SLS-mitigation-for-RET-and-BR-inst.patch \ file://0003-aarch64-Mitigate-SLS-for-BLR-instruction.patch \ file://0001-aarch64-Fix-up-__aarch64_cas16_acq_rel-fallback.patch \ + file://0001-lib-Remove-i-86-march-overrides.patch \ " SRC_URI[sha256sum] = "b8dd4368bb9c7f0b98188317ee0254dd8cc99d1e3a18d0ff146c855fe16c1d8c"
meta/recipes-devtools/gcc/gcc/0001-lib-Remove-i-86-march-overrides.patch new file +113 -0
113 lines not on this page · raw message
---
2.29.2
Quoted 7 lines · Nathan Rossi
When compiling gcc-runtime for i686 libatomic, libgomp and libitm are compiled with additional '-march' and '-mtune' overrides. This typically does not cause build errors (e.g. core2-32 tune), and results in less optimally tuned outputs. However with specific
TUNE_CCARGS(e.g. has '-mfpmath=sse' but not '-msse*'), the build errors with conflicting '-mfpmath' options.
Quoted 6 lines
This is caused by gcc-runtime specific configure scripts and
Makefileoptions that override by adding additional '-march'/'-mtune' args. OE builds gcc-runtime differently to other build systems, and provides target specificarch/tuneoptions directly. As such, this change adds a patch for gcc to remove the specific cases where the args are added.
looks fine to me.
Quoted 145 lines
Signed-off-by Nathan Rossi <[email protected]>--- meta/recipes-devtools/gcc/gcc-10.2.inc | 1 + .../gcc/0001-lib-Remove-i-86-march-overrides.patch | 113 +++++++++++++++++++++ 2 files changed, 114 insertions(+)create mode 100644
meta/recipes-devtools/gcc/gcc/0001-lib-Remove-i-86-march-overrides.patchmeta/recipes-devtools/gcc/gcc-10.2.inc +1
-0@@ -69,6 +69,7 @@ SRC_URI = "\ file://0002-aarch64-Introduce-SLS-mitigation-for-RET-and-BR-inst.patch \ file://0003-aarch64-Mitigate-SLS-for-BLR-instruction.patch \ file://0001-aarch64-Fix-up-__aarch64_cas16_acq_rel-fallback.patch \ + file://0001-lib-Remove-i-86-march-overrides.patch \ " SRC_URI[sha256sum] = "b8dd4368bb9c7f0b98188317ee0254dd8cc99d1e3a18d0ff146c855fe16c1d8c"meta/recipes-devtools/gcc/gcc/0001-lib-Remove-i-86-march-overrides.patch new file +113
-1@@ -0,0 +1,113 @@ +From 3776789cde95916f95deef48acf9532b0ebf91ae Mon Sep 17 00:00:00 2001 +From: Nathan Rossi <[email protected]> +Date: Tue, 15 Dec 2020 11:43:36 +1000 +Subject: [PATCH] lib*: Remove i*86 march overrides + +OE does not pass the '--with-arch' option to gccs configure, as such +some gcc-runtime libraries try to override the value of '-march' and +'-mtune' which OE already provides. This causes conflicts with other +i*86 instruction/architecture flags (e.g. -msse* and -mfpmath=*). +Additionally this caused the following libraries to be built with less +optimized tune configurations. + +Upstream suggests that this should be detecting or otherwise checking if +the target supports the desired functionality before overriding. + + https://gcc.gnu.org/legacy-ml/gcc-patches/2016-04/msg01344.html + +libatomic also overrides the '-march' specifically for IFUNC objects. +OE already supplies the correct march flag, so remove setting. + +Upstream-Status: Inappropriate [OE Specific] +Signed-off-by: Nathan Rossi <[email protected]> +--- + libatomic/Makefile.am | 1 - + libatomic/Makefile.in | 1 - + libatomic/configure.tgt | 9 --------- + libgomp/configure.tgt | 8 -------- + libitm/configure.tgt | 9 --------- + 5 files changed, 28 deletions(-) + +diff --git a/libatomic/Makefile.am b/libatomic/Makefile.am +index ac1ca64587..5aa16e0699 100644 +--- a/libatomic/Makefile.am ++++ b/libatomic/Makefile.am +@@ -133,7 +133,6 @@ libatomic_la_LIBADD += $(foreach s,$(SIZES),$(addsuffix _$(s)_1_.lo,$(SIZEOBJS)) + libatomic_la_LIBADD += $(addsuffix _8_2_.lo,$(SIZEOBJS)) + endif + if ARCH_I386 +-IFUNC_OPTIONS = -march=i586 + libatomic_la_LIBADD += $(addsuffix _8_1_.lo,$(SIZEOBJS)) + endif + if ARCH_X86_64 +diff --git a/libatomic/Makefile.in b/libatomic/Makefile.in +index 97df2d7ff0..82c511d420 100644 +--- a/libatomic/Makefile.in ++++ b/libatomic/Makefile.in +@@ -432,7 +432,6 @@ libatomic_la_LIBADD = $(foreach s,$(SIZES),$(addsuffix \ + _$(s)_.lo,$(SIZEOBJS))) $(am__append_1) $(am__append_2) \ + $(am__append_3) $(am__append_4) + @ARCH_ARM_LINUX_TRUE@@HAVE_IFUNC_TRUE@IFUNC_OPTIONS = -march=armv7-a+fp -DHAVE_KERNEL64 +-@ARCH_I386_TRUE@@HAVE_IFUNC_TRUE@IFUNC_OPTIONS = -march=i586 + @ARCH_X86_64_TRUE@@HAVE_IFUNC_TRUE@IFUNC_OPTIONS = -mcx16 + libatomic_convenience_la_SOURCES = $(libatomic_la_SOURCES) + libatomic_convenience_la_LIBADD = $(libatomic_la_LIBADD) +diff --git a/libatomic/configure.tgt b/libatomic/configure.tgt +index 5dd0926d20..3cc41773c3 100644 +--- a/libatomic/configure.tgt ++++ b/libatomic/configure.tgt +@@ -82,15 +82,6 @@ case "${target_cpu}" in + ;; + + i[3456]86) +- case " ${CC} ${CFLAGS} " in +- *" -m64 "*|*" -mx32 "*) +- ;; +- *) +- if test -z "$with_arch"; then +- XCFLAGS="${XCFLAGS} -march=i486 -mtune=${target_cpu}" +- XCFLAGS="${XCFLAGS} -fomit-frame-pointer" +- fi +- esac + ARCH=x86 + # ??? Detect when -march=i686 is already enabled. + try_ifunc=yes +diff --git a/libgomp/configure.tgt b/libgomp/configure.tgt +index 4790a31e39..46c4c958e6 100644 +--- a/libgomp/configure.tgt ++++ b/libgomp/configure.tgt +@@ -72,14 +72,6 @@ if test x$enable_linux_futex = xyes; then + # Note that bare i386 is not included here. We need cmpxchg. + i[456]86-*-linux*) + config_path="linux/x86 linux posix" +- case " ${CC} ${CFLAGS} " in +- *" -m64 "*|*" -mx32 "*) +- ;; +- *) +- if test -z "$with_arch"; then +- XCFLAGS="${XCFLAGS} -march=i486 -mtune=${target_cpu}" +- fi +- esac + ;; + + # Similar jiggery-pokery for x86_64 multilibs, except here we +diff --git a/libitm/configure.tgt b/libitm/configure.tgt +index 04109160e9..3d78ea609d 100644 +--- a/libitm/configure.tgt ++++ b/libitm/configure.tgt +@@ -59,15 +59,6 @@ case "${target_cpu}" in + arm*) ARCH=arm ;; + + i[3456]86) +- case " ${CC} ${CFLAGS} " in +- *" -m64 "*|*" -mx32 "*) +- ;; +- *) +- if test -z "$with_arch"; then +- XCFLAGS="${XCFLAGS} -march=i486 -mtune=${target_cpu}" +- XCFLAGS="${XCFLAGS} -fomit-frame-pointer" +- fi +- esac + XCFLAGS="${XCFLAGS} -mrtm" + ARCH=x86 + ;; ---2.29.2
Nathan
Can you check if your patch is still needed after
https://github.com/kraj/gcc/commit/0411ae7f08e0f5a8b02ff313d26d27a0f6d1bb34
Quoted 157 lines · Nathan Rossi
When compiling gcc-runtime for i686 libatomic, libgomp and libitm are compiled with additional '-march' and '-mtune' overrides. This typically does not cause build errors (e.g. core2-32 tune), and results in less optimally tuned outputs. However with specific
TUNE_CCARGS(e.g. has '-mfpmath=sse' but not '-msse*'), the build errors with conflicting '-mfpmath' options.This is caused by gcc-runtime specific configure scripts and
Makefileoptions that override by adding additional '-march'/'-mtune' args. OE builds gcc-runtime differently to other build systems, and provides target specificarch/tuneoptions directly. As such, this change adds a patch for gcc to remove the specific cases where the args are added.Signed-off-by Nathan Rossi <[email protected]>--- meta/recipes-devtools/gcc/gcc-10.2.inc | 1 + .../gcc/0001-lib-Remove-i-86-march-overrides.patch | 113 +++++++++++++++++++++ 2 files changed, 114 insertions(+) create mode 100644 meta/recipes-devtools/gcc/gcc/0001-lib-Remove-i-86-march-overrides.patchmeta/recipes-devtools/gcc/gcc-10.2.inc +1
-0@@ -69,6 +69,7 @@ SRC_URI = "\ file://0002-aarch64-Introduce-SLS-mitigation-for-RET-and-BR-inst.patch \ file://0003-aarch64-Mitigate-SLS-for-BLR-instruction.patch \ file://0001-aarch64-Fix-up-__aarch64_cas16_acq_rel-fallback.patch \ + file://0001-lib-Remove-i-86-march-overrides.patch \ " SRC_URI[sha256sum] = "b8dd4368bb9c7f0b98188317ee0254dd8cc99d1e3a18d0ff146c855fe16c1d8c"meta/recipes-devtools/gcc/gcc/0001-lib-Remove-i-86-march-overrides.patch new file +113
-1@@ -0,0 +1,113 @@ +From 3776789cde95916f95deef48acf9532b0ebf91ae Mon Sep 17 00:00:00 2001 +From: Nathan Rossi <[email protected]> +Date: Tue, 15 Dec 2020 11:43:36 +1000 +Subject: [PATCH] lib*: Remove i*86 march overrides + +OE does not pass the '--with-arch' option to gccs configure, as such +some gcc-runtime libraries try to override the value of '-march' and +'-mtune' which OE already provides. This causes conflicts with other +i*86 instruction/architecture flags (e.g. -msse* and -mfpmath=*). +Additionally this caused the following libraries to be built with less +optimized tune configurations. + +Upstream suggests that this should be detecting or otherwise checking if +the target supports the desired functionality before overriding. + + https://gcc.gnu.org/legacy-ml/gcc-patches/2016-04/msg01344.html + +libatomic also overrides the '-march' specifically for IFUNC objects. +OE already supplies the correct march flag, so remove setting. + +Upstream-Status: Inappropriate [OE Specific] +Signed-off-by: Nathan Rossi <[email protected]> +--- + libatomic/Makefile.am | 1 - + libatomic/Makefile.in | 1 - + libatomic/configure.tgt | 9 --------- + libgomp/configure.tgt | 8 -------- + libitm/configure.tgt | 9 --------- + 5 files changed, 28 deletions(-) + +diff --git a/libatomic/Makefile.am b/libatomic/Makefile.am +index ac1ca64587..5aa16e0699 100644 +--- a/libatomic/Makefile.am ++++ b/libatomic/Makefile.am +@@ -133,7 +133,6 @@ libatomic_la_LIBADD += $(foreach s,$(SIZES),$(addsuffix _$(s)_1_.lo,$(SIZEOBJS)) + libatomic_la_LIBADD += $(addsuffix _8_2_.lo,$(SIZEOBJS)) + endif + if ARCH_I386 +-IFUNC_OPTIONS = -march=i586 + libatomic_la_LIBADD += $(addsuffix _8_1_.lo,$(SIZEOBJS)) + endif + if ARCH_X86_64 +diff --git a/libatomic/Makefile.in b/libatomic/Makefile.in +index 97df2d7ff0..82c511d420 100644 +--- a/libatomic/Makefile.in ++++ b/libatomic/Makefile.in +@@ -432,7 +432,6 @@ libatomic_la_LIBADD = $(foreach s,$(SIZES),$(addsuffix \ + _$(s)_.lo,$(SIZEOBJS))) $(am__append_1) $(am__append_2) \ + $(am__append_3) $(am__append_4) + @ARCH_ARM_LINUX_TRUE@@HAVE_IFUNC_TRUE@IFUNC_OPTIONS = -march=armv7-a+fp -DHAVE_KERNEL64 +-@ARCH_I386_TRUE@@HAVE_IFUNC_TRUE@IFUNC_OPTIONS = -march=i586 + @ARCH_X86_64_TRUE@@HAVE_IFUNC_TRUE@IFUNC_OPTIONS = -mcx16 + libatomic_convenience_la_SOURCES = $(libatomic_la_SOURCES) + libatomic_convenience_la_LIBADD = $(libatomic_la_LIBADD) +diff --git a/libatomic/configure.tgt b/libatomic/configure.tgt +index 5dd0926d20..3cc41773c3 100644 +--- a/libatomic/configure.tgt ++++ b/libatomic/configure.tgt +@@ -82,15 +82,6 @@ case "${target_cpu}" in + ;; + + i[3456]86) +- case " ${CC} ${CFLAGS} " in +- *" -m64 "*|*" -mx32 "*) +- ;; +- *) +- if test -z "$with_arch"; then +- XCFLAGS="${XCFLAGS} -march=i486 -mtune=${target_cpu}" +- XCFLAGS="${XCFLAGS} -fomit-frame-pointer" +- fi +- esac + ARCH=x86 + # ??? Detect when -march=i686 is already enabled. + try_ifunc=yes +diff --git a/libgomp/configure.tgt b/libgomp/configure.tgt +index 4790a31e39..46c4c958e6 100644 +--- a/libgomp/configure.tgt ++++ b/libgomp/configure.tgt +@@ -72,14 +72,6 @@ if test x$enable_linux_futex = xyes; then + # Note that bare i386 is not included here. We need cmpxchg. + i[456]86-*-linux*) + config_path="linux/x86 linux posix" +- case " ${CC} ${CFLAGS} " in +- *" -m64 "*|*" -mx32 "*) +- ;; +- *) +- if test -z "$with_arch"; then +- XCFLAGS="${XCFLAGS} -march=i486 -mtune=${target_cpu}" +- fi +- esac + ;; + + # Similar jiggery-pokery for x86_64 multilibs, except here we +diff --git a/libitm/configure.tgt b/libitm/configure.tgt +index 04109160e9..3d78ea609d 100644 +--- a/libitm/configure.tgt ++++ b/libitm/configure.tgt +@@ -59,15 +59,6 @@ case "${target_cpu}" in + arm*) ARCH=arm ;; + + i[3456]86) +- case " ${CC} ${CFLAGS} " in +- *" -m64 "*|*" -mx32 "*) +- ;; +- *) +- if test -z "$with_arch"; then +- XCFLAGS="${XCFLAGS} -march=i486 -mtune=${target_cpu}" +- XCFLAGS="${XCFLAGS} -fomit-frame-pointer" +- fi +- esac + XCFLAGS="${XCFLAGS} -mrtm" + ARCH=x86 + ;; ---2.29.2
Quoted 6 lines · Khem Raj
Nathan
Can you check if your patch is still needed after
https://github.com/kraj/gcc/commit/0411ae7f08e0f5a8b02ff313d26d27a0f6d1bb34
Hi Khem,
I tested this change and it resolves the configuration issue. Did you want the above change backported to replace the oe specific patch? Since that oe specific patch has already been backported to a number of releases.
Thanks,
Nathan
Quoted 159 lines · Nathan Rossi
Quoted 157 lines
When compiling gcc-runtime for i686 libatomic, libgomp and libitm are compiled with additional '-march' and '-mtune' overrides. This typically does not cause build errors (e.g. core2-32 tune), and results in less optimally tuned outputs. However with specific
TUNE_CCARGS(e.g. has '-mfpmath=sse' but not '-msse*'), the build errors with conflicting '-mfpmath' options.This is caused by gcc-runtime specific configure scripts and
Makefileoptions that override by adding additional '-march'/'-mtune' args. OE builds gcc-runtime differently to other build systems, and provides target specificarch/tuneoptions directly. As such, this change adds a patch for gcc to remove the specific cases where the args are added.Signed-off-by Nathan Rossi <[email protected]>--- meta/recipes-devtools/gcc/gcc-10.2.inc | 1 + .../gcc/0001-lib-Remove-i-86-march-overrides.patch | 113 +++++++++++++++++++++ 2 files changed, 114 insertions(+) create mode 100644 meta/recipes-devtools/gcc/gcc/0001-lib-Remove-i-86-march-overrides.patchmeta/recipes-devtools/gcc/gcc-10.2.inc +1
-0@@ -69,6 +69,7 @@ SRC_URI = "\ file://0002-aarch64-Introduce-SLS-mitigation-for-RET-and-BR-inst.patch \ file://0003-aarch64-Mitigate-SLS-for-BLR-instruction.patch \ file://0001-aarch64-Fix-up-__aarch64_cas16_acq_rel-fallback.patch \ + file://0001-lib-Remove-i-86-march-overrides.patch \ " SRC_URI[sha256sum] = "b8dd4368bb9c7f0b98188317ee0254dd8cc99d1e3a18d0ff146c855fe16c1d8c"meta/recipes-devtools/gcc/gcc/0001-lib-Remove-i-86-march-overrides.patch new file +113
-1@@ -0,0 +1,113 @@ +From 3776789cde95916f95deef48acf9532b0ebf91ae Mon Sep 17 00:00:00 2001 +From: Nathan Rossi <[email protected]> +Date: Tue, 15 Dec 2020 11:43:36 +1000 +Subject: [PATCH] lib*: Remove i*86 march overrides + +OE does not pass the '--with-arch' option to gccs configure, as such +some gcc-runtime libraries try to override the value of '-march' and +'-mtune' which OE already provides. This causes conflicts with other +i*86 instruction/architecture flags (e.g. -msse* and -mfpmath=*). +Additionally this caused the following libraries to be built with less +optimized tune configurations. + +Upstream suggests that this should be detecting or otherwise checking if +the target supports the desired functionality before overriding. + + https://gcc.gnu.org/legacy-ml/gcc-patches/2016-04/msg01344.html + +libatomic also overrides the '-march' specifically for IFUNC objects. +OE already supplies the correct march flag, so remove setting. + +Upstream-Status: Inappropriate [OE Specific] +Signed-off-by: Nathan Rossi <[email protected]> +--- + libatomic/Makefile.am | 1 - + libatomic/Makefile.in | 1 - + libatomic/configure.tgt | 9 --------- + libgomp/configure.tgt | 8 -------- + libitm/configure.tgt | 9 --------- + 5 files changed, 28 deletions(-) + +diff --git a/libatomic/Makefile.am b/libatomic/Makefile.am +index ac1ca64587..5aa16e0699 100644 +--- a/libatomic/Makefile.am ++++ b/libatomic/Makefile.am +@@ -133,7 +133,6 @@ libatomic_la_LIBADD += $(foreach s,$(SIZES),$(addsuffix _$(s)_1_.lo,$(SIZEOBJS)) + libatomic_la_LIBADD += $(addsuffix _8_2_.lo,$(SIZEOBJS)) + endif + if ARCH_I386 +-IFUNC_OPTIONS = -march=i586 + libatomic_la_LIBADD += $(addsuffix _8_1_.lo,$(SIZEOBJS)) + endif + if ARCH_X86_64 +diff --git a/libatomic/Makefile.in b/libatomic/Makefile.in +index 97df2d7ff0..82c511d420 100644 +--- a/libatomic/Makefile.in ++++ b/libatomic/Makefile.in +@@ -432,7 +432,6 @@ libatomic_la_LIBADD = $(foreach s,$(SIZES),$(addsuffix \ + _$(s)_.lo,$(SIZEOBJS))) $(am__append_1) $(am__append_2) \ + $(am__append_3) $(am__append_4) + @ARCH_ARM_LINUX_TRUE@@HAVE_IFUNC_TRUE@IFUNC_OPTIONS = -march=armv7-a+fp -DHAVE_KERNEL64 +-@ARCH_I386_TRUE@@HAVE_IFUNC_TRUE@IFUNC_OPTIONS = -march=i586 + @ARCH_X86_64_TRUE@@HAVE_IFUNC_TRUE@IFUNC_OPTIONS = -mcx16 + libatomic_convenience_la_SOURCES = $(libatomic_la_SOURCES) + libatomic_convenience_la_LIBADD = $(libatomic_la_LIBADD) +diff --git a/libatomic/configure.tgt b/libatomic/configure.tgt +index 5dd0926d20..3cc41773c3 100644 +--- a/libatomic/configure.tgt ++++ b/libatomic/configure.tgt +@@ -82,15 +82,6 @@ case "${target_cpu}" in + ;; + + i[3456]86) +- case " ${CC} ${CFLAGS} " in +- *" -m64 "*|*" -mx32 "*) +- ;; +- *) +- if test -z "$with_arch"; then +- XCFLAGS="${XCFLAGS} -march=i486 -mtune=${target_cpu}" +- XCFLAGS="${XCFLAGS} -fomit-frame-pointer" +- fi +- esac + ARCH=x86 + # ??? Detect when -march=i686 is already enabled. + try_ifunc=yes +diff --git a/libgomp/configure.tgt b/libgomp/configure.tgt +index 4790a31e39..46c4c958e6 100644 +--- a/libgomp/configure.tgt ++++ b/libgomp/configure.tgt +@@ -72,14 +72,6 @@ if test x$enable_linux_futex = xyes; then + # Note that bare i386 is not included here. We need cmpxchg. + i[456]86-*-linux*) + config_path="linux/x86 linux posix" +- case " ${CC} ${CFLAGS} " in +- *" -m64 "*|*" -mx32 "*) +- ;; +- *) +- if test -z "$with_arch"; then +- XCFLAGS="${XCFLAGS} -march=i486 -mtune=${target_cpu}" +- fi +- esac + ;; + + # Similar jiggery-pokery for x86_64 multilibs, except here we +diff --git a/libitm/configure.tgt b/libitm/configure.tgt +index 04109160e9..3d78ea609d 100644 +--- a/libitm/configure.tgt ++++ b/libitm/configure.tgt +@@ -59,15 +59,6 @@ case "${target_cpu}" in + arm*) ARCH=arm ;; + + i[3456]86) +- case " ${CC} ${CFLAGS} " in +- *" -m64 "*|*" -mx32 "*) +- ;; +- *) +- if test -z "$with_arch"; then +- XCFLAGS="${XCFLAGS} -march=i486 -mtune=${target_cpu}" +- XCFLAGS="${XCFLAGS} -fomit-frame-pointer" +- fi +- esac + XCFLAGS="${XCFLAGS} -mrtm" + ARCH=x86 + ;; ---2.29.2
I think it will be good to do so
Quoted 195 lines · Nathan Rossi
Quoted 6 lines · Khem Raj
Nathan
Can you check if your patch is still needed after
https://github.com/kraj/gcc/commit/0411ae7f08e0f5a8b02ff313d26d27a0f6d1bb34
Hi Khem,
I tested this change and it resolves the configuration issue. Did you want the above change backported to replace the oe specific patch? Since that oe specific patch has already been backported to a number of releases.
Thanks,
NathanQuoted 2 lines
On Tue, Dec 15, 2020 at 12:35 AM Nathan Rossi <[email protected]>
Quoted 3 lines · wrote:
Quoted 3 lines
When compiling gcc-runtime for i686 libatomic, libgomp and libitm are compiled with additional '-march' and '-mtune' overrides. This
typically
Quoted 11 lines
Quoted 11 lines
does not cause build errors (e.g. core2-32 tune), and results in less optimally tuned outputs. However with specific
TUNE_CCARGS(e.g. has '-mfpmath=sse' but not '-msse*'), the build errors with conflicting '-mfpmath' options.This is caused by gcc-runtime specific configure scripts and
Makefileoptions that override by adding additional '-march'/'-mtune' args. OE builds gcc-runtime differently to other build systems, and provides target specificarch/tuneoptions directly. As such, this change adds a patch for gcc to remove the specific cases where the args are added.Quoted 4 lines
Quoted 4 lines
Signed-off-by Nathan Rossi <[email protected]>--- meta/recipes-devtools/gcc/gcc-10.2.inc | 1 + .../gcc/0001-lib-Remove-i-86-march-overrides.patch | 113+++++++++++++++++++++
Quoted 2 lines
Quoted 2 lines
2 files changed, 114 insertions(+) create mode 100644
meta/recipes-devtools/gcc/gcc/0001-lib-Remove-i-86-march-overrides.patchQuoted 2 lines
Review
Quoted 2 lines
(quoted diff) +0
-0b/meta/recipes-devtools/gcc/gcc-10.2.inc
Quoted 5 lines
meta/recipes-devtools/gcc/gcc-10.2.inc SRC_URI = "\ · line 69
Quoted 5 lines
index
e88f2ade6d..1e69c91311100644meta/recipes-devtools/gcc/gcc-10.2.inc +0
-0@@ -69,6 +69,7 @@ SRC_URI = "\file://0002-aarch64-Introduce-SLS-mitigation-for-RET-and-BR-inst.patch \
Quoted 1 line
meta/recipes-devtools/gcc/gcc-10.2.inc SRC_URI = "\ · line 70
Quoted 1 line
meta/recipes-devtools/gcc/gcc-10.2.inc +0
-0file://0003-aarch64-Mitigate-SLS-for-BLR-instruction.patch\
Quoted 1 line
Quoted 1 line
file://0001-aarch64-Fix-up-__aarch64_cas16_acq_rel-fallback.patch \
Quoted 3 lines
Quoted 3 lines
+ file://0001-lib-Remove-i-86-march-overrides.patch \
"
SRC_URI[sha256sum] ="
b8dd4368bb9c7f0b98188317ee0254dd8cc99d1e3a18d0ff146c855fe16c1d8c"Quoted 2 lines
Quoted 2 lines
diff --git
a/meta/recipes-devtools/gcc/gcc/0001-lib-Remove-i-86-march-overrides.patchb/meta/recipes-devtools/gcc/gcc/0001-lib-Remove-i-86-march-overrides.patchQuoted 4 lines
Quoted 4 lines
new file mode 100644
index 0000000000..3f04e121ba
---/dev/null
+++
b/meta/recipes-devtools/gcc/gcc/0001-lib-Remove-i-86-march-overrides.patchQuoted 6 lines
Quoted 6 lines
meta/recipes-devtools/gcc/gcc-10.2.inc +5
-0@@ -0,0 +1,113 @@ +From 3776789cde95916f95deef48acf9532b0ebf91ae Mon Sep 17 00:00:00 2001 +From: Nathan Rossi <[email protected]> +Date: Tue, 15 Dec 2020 11:43:36 +1000 +Subject: [PATCH] lib*: Remove i*86 march overrides +Quoted 8 lines
meta/recipes-devtools/gcc/gcc-10.2.inc line 13
Quoted 8 lines
meta/recipes-devtools/gcc/gcc-10.2.inc +8
-0+OE does not pass the '--with-arch' option to gccs configure, as such +some gcc-runtime libraries try to override the value of '-march' and +'-mtune' which OE already provides. This causes conflicts with other +i*86 instruction/architecture flags (e.g. -msse* and -mfpmath=*). +Additionally this caused the following libraries to be built with less +optimized tune configurations. + +Upstream suggests that this should be detecting or otherwise checkingif
Quoted 14 lines
Quoted 14 lines
meta/recipes-devtools/gcc/gcc-10.2.inc +14
-0+the target supports the desired functionality before overriding. + + https://gcc.gnu.org/legacy-ml/gcc-patches/2016-04/msg01344.html + +libatomic also overrides the '-march' specifically for IFUNC objects. +OE already supplies the correct march flag, so remove setting. + +Upstream-Status: Inappropriate [OE Specific] +Signed-off-by: Nathan Rossi <[email protected]> +--- + libatomic/Makefile.am | 1 - + libatomic/Makefile.in | 1 - + libatomic/configure.tgt | 9 --------- + libgomp/configure.tgt | 8 --------Quoted 8 lines
meta/recipes-devtools/gcc/gcc-10.2.inc line 35
Quoted 8 lines
meta/recipes-devtools/gcc/gcc-10.2.inc +8
-0+ libitm/configure.tgt | 9 --------- + 5 files changed, 28 deletions(-) + +diff --git a/libatomic/Makefile.am b/libatomic/Makefile.am +index ac1ca64587..5aa16e0699 100644 +--- a/libatomic/Makefile.am ++++ b/libatomic/Makefile.am +@@ -133,7 +133,6 @@ libatomic_la_LIBADD += $(foreachs,$(SIZES),$(addsuffix _$(s)_1_.lo,$(SIZEOBJS))
Quoted 4 lines
Quoted 4 lines
meta/recipes-devtools/gcc/gcc-10.2.inc +4
-0+ libatomic_la_LIBADD += $(addsuffix _8_2_.lo,$(SIZEOBJS)) + endif + if ARCH_I386 +-IFUNC_OPTIONS = -march=i586Quoted 8 lines
meta/recipes-devtools/gcc/gcc-10.2.inc line 47
Quoted 8 lines
meta/recipes-devtools/gcc/gcc-10.2.inc +8
-0+ libatomic_la_LIBADD += $(addsuffix _8_1_.lo,$(SIZEOBJS)) + endif + if ARCH_X86_64 +diff --git a/libatomic/Makefile.in b/libatomic/Makefile.in +index 97df2d7ff0..82c511d420 100644 +--- a/libatomic/Makefile.in ++++ b/libatomic/Makefile.in +@@ -432,7 +432,6 @@ libatomic_la_LIBADD = $(foreachs,$(SIZES),$(addsuffix \
Quoted 3 lines
meta/recipes-devtools/gcc/gcc-10.2.inc line 50
Quoted 3 lines
meta/recipes-devtools/gcc/gcc-10.2.inc +3
-0+ _$(s)_.lo,$(SIZEOBJS))) $(am__append_1) $(am__append_2) \ + $(am__append_3) $(am__append_4) + @ARCH_ARM_LINUX_TRUE@@HAVE_IFUNC_TRUE@IFUNC_OPTIONS =-march=armv7-a+fp -DHAVE_KERNEL64
Quoted 68 lines
Quoted 68 lines
meta/recipes-devtools/gcc/gcc-10.2.inc +63
-1+-@ARCH_I386_TRUE@@HAVE_IFUNC_TRUE@IFUNC_OPTIONS = -march=i586 + @ARCH_X86_64_TRUE@@HAVE_IFUNC_TRUE@IFUNC_OPTIONS = -mcx16 + libatomic_convenience_la_SOURCES = $(libatomic_la_SOURCES) + libatomic_convenience_la_LIBADD = $(libatomic_la_LIBADD) +diff --git a/libatomic/configure.tgt b/libatomic/configure.tgt +index 5dd0926d20..3cc41773c3 100644 +--- a/libatomic/configure.tgt ++++ b/libatomic/configure.tgt +@@ -82,15 +82,6 @@ case "${target_cpu}" in + ;; + + i[3456]86) +- case " ${CC} ${CFLAGS} " in +- *" -m64 "*|*" -mx32 "*) +- ;; +- *) +- if test -z "$with_arch"; then +- XCFLAGS="${XCFLAGS} -march=i486 -mtune=${target_cpu}" +- XCFLAGS="${XCFLAGS} -fomit-frame-pointer" +- fi +- esac + ARCH=x86 + # ??? Detect when -march=i686 is already enabled. + try_ifunc=yes +diff --git a/libgomp/configure.tgt b/libgomp/configure.tgt +index 4790a31e39..46c4c958e6 100644 +--- a/libgomp/configure.tgt ++++ b/libgomp/configure.tgt +@@ -72,14 +72,6 @@ if test x$enable_linux_futex = xyes; then + # Note that bare i386 is not included here. We need cmpxchg. + i[456]86-*-linux*) + config_path="linux/x86 linux posix" +- case " ${CC} ${CFLAGS} " in +- *" -m64 "*|*" -mx32 "*) +- ;; +- *) +- if test -z "$with_arch"; then +- XCFLAGS="${XCFLAGS} -march=i486 -mtune=${target_cpu}" +- fi +- esac + ;; + + # Similar jiggery-pokery for x86_64 multilibs, except here we +diff --git a/libitm/configure.tgt b/libitm/configure.tgt +index 04109160e9..3d78ea609d 100644 +--- a/libitm/configure.tgt ++++ b/libitm/configure.tgt +@@ -59,15 +59,6 @@ case "${target_cpu}" in + arm*) ARCH=arm ;; + + i[3456]86) +- case " ${CC} ${CFLAGS} " in +- *" -m64 "*|*" -mx32 "*) +- ;; +- *) +- if test -z "$with_arch"; then +- XCFLAGS="${XCFLAGS} -march=i486 -mtune=${target_cpu}" +- XCFLAGS="${XCFLAGS} -fomit-frame-pointer" +- fi +- esac + XCFLAGS="${XCFLAGS} -mrtm" + ARCH=x86 + ;; ---2.29.2