[PATCH 0/4] gcc fixes and eglibc update

[PATCH 0/4] gcc fixes and eglibc update

[Khem Raj]

This patchset updates eglibc 2.15 to latest (all qemu boot into sato image)
fix unpackaged file warning in libgcc
Remove USE_NLS from gcc configure ( depends on a patch from my previous pull )

All uclibc/eglibc images and meta-toolchains build ok and boot ok too.

The following changes since commit f4b5afe46ff018570c1949653eb6995428827f6f:

  sanity.bbclass: Increase LAYER_CONF_VERSION to match bblayers change (2012-06-25 17:20:06 +0100)

are available in the git repository at:

  git://git.openembedded.org/openembedded-core-contrib kraj/toolchain
  http://cgit.openembedded.org/cgit.cgi/openembedded-core-contrib/log/?h=kraj/toolchain

Dennis Lan (1):
  eglibc: support mipsel little endian, filter out -march=mips32

Khem Raj (3):
  libgcc: Remove unpackage unwind.h
  gcc-configure-sdk,gcc-configure-cross: Dont recompute USE_NLS
  eglibc-2.15: Update to latest on 2.15 branch

 meta/recipes-core/eglibc/eglibc_2.15.bb           |    3 ++-
 meta/recipes-devtools/gcc/gcc-configure-cross.inc |    2 --
 meta/recipes-devtools/gcc/gcc-configure-sdk.inc   |    4 ----
 meta/recipes-devtools/gcc/libgcc_4.7.bb           |    4 ++++
 4 files changed, 6 insertions(+), 7 deletions(-)

-- 
1.7.9.5

[PATCH 1/4] libgcc: Remove unpackage unwind.h

[Khem Raj]

Fixes build warnings like below

WARNING: For recipe libgcc, the following files/directories were installed but not shipped in any package:
WARNING:   /usr/lib/arm-oe-linux-gnueabi/4.7.2/include
WARNING:   /usr/lib/arm-oe-linux-gnueabi/4.7.2/include/unwind.h

Signed-off-by: Khem Raj <raj.khem@gmail.com>
---
 meta/recipes-devtools/gcc/libgcc_4.7.bb |    4 ++++
 1 file changed, 4 insertions(+)

diff --git a/meta/recipes-devtools/gcc/libgcc_4.7.bb b/meta/recipes-devtools/gcc/libgcc_4.7.bb
index 891c03e..55aa03c 100644
--- a/meta/recipes-devtools/gcc/libgcc_4.7.bb
+++ b/meta/recipes-devtools/gcc/libgcc_4.7.bb
@@ -58,6 +58,10 @@ do_install () {
 
 	mv ${D}${libdir}/gcc/* ${D}${libdir}
 	rm -rf ${D}${libdir}/gcc/
+	# unwind.h is installed here which is shipped in gcc-cross
+	# as well as target gcc and they are identical so we dont
+	# ship one with libgcc here
+	rm -rf ${D}${libdir}/${TARGET_SYS}/${BINV}/include
 }
 
 do_package_write_ipk[depends] += "virtual/${MLPREFIX}libc:do_package"
-- 
1.7.9.5

[PATCH 2/4] gcc-configure-sdk, gcc-configure-cross: Dont recompute USE_NLS

[Khem Raj]

This was needed because on eglibc based builds we did not
define USE_NLS but now we define it in default config to have
a valid value so we do not need this here. Moreover it was
wrong and was not covering all cases of uclibc triplets

Signed-off-by: Khem Raj <raj.khem@gmail.com>
---
 meta/recipes-devtools/gcc/gcc-configure-cross.inc |    2 --
 meta/recipes-devtools/gcc/gcc-configure-sdk.inc   |    4 ----
 2 files changed, 6 deletions(-)

diff --git a/meta/recipes-devtools/gcc/gcc-configure-cross.inc b/meta/recipes-devtools/gcc/gcc-configure-cross.inc
index 3cb3452..243ee55d 100644
--- a/meta/recipes-devtools/gcc/gcc-configure-cross.inc
+++ b/meta/recipes-devtools/gcc/gcc-configure-cross.inc
@@ -1,7 +1,5 @@
 require gcc-configure-common.inc
 
-USE_NLS = '${@base_conditional( "TARGET_OS", "linux-uclibc", "no", "", d )}'
-
 EXTRA_OECONF += " --enable-poison-system-directories \
 		"
 
diff --git a/meta/recipes-devtools/gcc/gcc-configure-sdk.inc b/meta/recipes-devtools/gcc/gcc-configure-sdk.inc
index 90e20f5..51c0217 100644
--- a/meta/recipes-devtools/gcc/gcc-configure-sdk.inc
+++ b/meta/recipes-devtools/gcc/gcc-configure-sdk.inc
@@ -1,9 +1,5 @@
 require gcc-configure-common.inc
 
-# The two lines below conflict, this needs fixing - RP
-USE_NLS = '${@base_conditional( "TARGET_OS", "linux-uclibc", "no", "", d )}'
-USE_NLS = '${@base_conditional( "TARGET_OS", "linux-uclibceabi", "no", "", d )}'
-
 EXTRA_OECONF_PATHS = "--with-gxx-include-dir=${SDKPATH}/sysroots/${TUNE_PKGARCH}${TARGET_VENDOR}-${TARGET_OS}${target_includedir}/c++ \
                       --with-build-time-tools=${STAGING_DIR_NATIVE}${prefix_native}/${TARGET_SYS}/bin \
                       --with-sysroot=${SDKPATH}/sysroots/${TUNE_PKGARCH}${TARGET_VENDOR}-${TARGET_OS} \
-- 
1.7.9.5

[PATCH 3/4] eglibc: support mipsel little endian, filter out -march=mips32

[Khem Raj]

From: Dennis Lan <dennis.yxun@gmail.com>

Quote for eglibc source:
 for mips eglibc now builds syscall tables for all abi's
 so we make sure that we choose right march option which is
 compatible with o32,n32 and n64 abi's
 e.g. -march=mips32 is not compatible with n32 and n64 therefore
 we filter it out in such case -march=from-abi which will be
 mips1 when using o32 and mips3 when using n32/n64

Signed-off-by: Dennis Lan <dennis.yxun@gmail.com>
---
 meta/recipes-core/eglibc/eglibc_2.15.bb |    1 +
 1 file changed, 1 insertion(+)

diff --git a/meta/recipes-core/eglibc/eglibc_2.15.bb b/meta/recipes-core/eglibc/eglibc_2.15.bb
index e57a112..b2922b4 100644
--- a/meta/recipes-core/eglibc/eglibc_2.15.bb
+++ b/meta/recipes-core/eglibc/eglibc_2.15.bb
@@ -101,6 +101,7 @@ do_patch_append() {
 # mips1 when using o32 and mips3 when using n32/n64
 
 TUNE_CCARGS_mips := "${@oe_filter_out('-march=mips32', '${TUNE_CCARGS}', d)}"
+TUNE_CCARGS_mipsel := "${@oe_filter_out('-march=mips32', '${TUNE_CCARGS}', d)}"
 
 # We need to ensure that all of the i386 and x86_64 headers are identical
 # to support the multilib case.  We do this by copying headers from x86_64
-- 
1.7.9.5

[PATCH 4/4] eglibc-2.15: Update to latest on 2.15 branch

[Khem Raj]

Signed-off-by: Khem Raj <raj.khem@gmail.com>
---
 meta/recipes-core/eglibc/eglibc_2.15.bb |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/meta/recipes-core/eglibc/eglibc_2.15.bb b/meta/recipes-core/eglibc/eglibc_2.15.bb
index b2922b4..7dd9810 100644
--- a/meta/recipes-core/eglibc/eglibc_2.15.bb
+++ b/meta/recipes-core/eglibc/eglibc_2.15.bb
@@ -1,6 +1,6 @@
 require eglibc.inc
 
-SRCREV = "17386"
+SRCREV = "19294"
 
 DEPENDS += "gperf-native"
 PR = "r12"
-- 
1.7.9.5

Re: [PATCH 3/4] eglibc: support mipsel little endian, filter out -march=mips32

[Dennis.Yxun]

HI Raj:
   So this will eventually goes into OE-core, right?
   As I see Bruce has no response for the "override scheme"
so mips -> for big endian mips o32 (could be mipseb better?, less confusion)
and mipsel -> for little endian mips o32, they in same level, not of
one's subset

Dennis

On Thu, Jun 28, 2012 at 9:10 AM, Khem Raj <raj.khem@gmail.com> wrote:
> From: Dennis Lan <dennis.yxun@gmail.com>
>
> Quote for eglibc source:
>  for mips eglibc now builds syscall tables for all abi's
>  so we make sure that we choose right march option which is
>  compatible with o32,n32 and n64 abi's
>  e.g. -march=mips32 is not compatible with n32 and n64 therefore
>  we filter it out in such case -march=from-abi which will be
>  mips1 when using o32 and mips3 when using n32/n64
>
> Signed-off-by: Dennis Lan <dennis.yxun@gmail.com>
> ---
>  meta/recipes-core/eglibc/eglibc_2.15.bb |    1 +
>  1 file changed, 1 insertion(+)
>
> diff --git a/meta/recipes-core/eglibc/eglibc_2.15.bb b/meta/recipes-core/eglibc/eglibc_2.15.bb
> index e57a112..b2922b4 100644
> --- a/meta/recipes-core/eglibc/eglibc_2.15.bb
> +++ b/meta/recipes-core/eglibc/eglibc_2.15.bb
> @@ -101,6 +101,7 @@ do_patch_append() {
>  # mips1 when using o32 and mips3 when using n32/n64
>
>  TUNE_CCARGS_mips := "${@oe_filter_out('-march=mips32', '${TUNE_CCARGS}', d)}"
> +TUNE_CCARGS_mipsel := "${@oe_filter_out('-march=mips32', '${TUNE_CCARGS}', d)}"
>
>  # We need to ensure that all of the i386 and x86_64 headers are identical
>  # to support the multilib case.  We do this by copying headers from x86_64
> --
> 1.7.9.5
>
>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core

Re: [PATCH 3/4] eglibc: support mipsel little endian, filter out -march=mips32

[Khem Raj]

[-- Attachment #1: Type: text/plain, Size: 2232 bytes --]

On Thursday, June 28, 2012, Dennis.Yxun <dennis.yxun@gmail.com> wrote:
> HI Raj:
>   So this will eventually goes into OE-core, right?
>   As I see Bruce has no response for the "override scheme"
> so mips -> for big endian mips o32 (could be mipseb better?, less
confusion)

Mips has always meant big endian no need to change there


> and mipsel -> for little endian mips o32, they in same level, not of
> one's subset
>
> Dennis
>
> On Thu, Jun 28, 2012 at 9:10 AM, Khem Raj <raj.khem@gmail.com> wrote:
>> From: Dennis Lan <dennis.yxun@gmail.com>
>>
>> Quote for eglibc source:
>>  for mips eglibc now builds syscall tables for all abi's
>>  so we make sure that we choose right march option which is
>>  compatible with o32,n32 and n64 abi's
>>  e.g. -march=mips32 is not compatible with n32 and n64 therefore
>>  we filter it out in such case -march=from-abi which will be
>>  mips1 when using o32 and mips3 when using n32/n64
>>
>> Signed-off-by: Dennis Lan <dennis.yxun@gmail.com>
>> ---
>>  meta/recipes-core/eglibc/eglibc_2.15.bb |    1 +
>>  1 file changed, 1 insertion(+)
>>
>> diff --git a/meta/recipes-core/eglibc/eglibc_2.15.bbb/meta/recipes-core/eglibc/
eglibc_2.15.bb
>> index e57a112..b2922b4 100644
>> --- a/meta/recipes-core/eglibc/eglibc_2.15.bb
>> +++ b/meta/recipes-core/eglibc/eglibc_2.15.bb
>> @@ -101,6 +101,7 @@ do_patch_append() {
>>  # mips1 when using o32 and mips3 when using n32/n64
>>
>>  TUNE_CCARGS_mips := "${@oe_filter_out('-march=mips32',
'${TUNE_CCARGS}', d)}"
>> +TUNE_CCARGS_mipsel := "${@oe_filter_out('-march=mips32',
'${TUNE_CCARGS}', d)}"
>>
>>  # We need to ensure that all of the i386 and x86_64 headers are
identical
>>  # to support the multilib case.  We do this by copying headers from
x86_64
>> --
>> 1.7.9.5
>>
>>
>> _______________________________________________
>> Openembedded-core mailing list
>> Openembedded-core@lists.openembedded.org
>> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>

[-- Attachment #2: Type: text/html, Size: 3391 bytes --]