* [PATCH 1/6] Fix mdadm for powerpc64
@ 2011-10-04 22:03 Matthew McClintock
2011-10-04 22:03 ` [PATCH 2/6] Fix perl for powerpc64 build Matthew McClintock
` (4 more replies)
0 siblings, 5 replies; 25+ messages in thread
From: Matthew McClintock @ 2011-10-04 22:03 UTC (permalink / raw)
To: openembedded-core
This fixes build failures from build warnings on powerpc64
Signed-off-by: Matthew McClintock <msm@freescale.com>
---
Replied to feedback on messsage about compilation errors seen.
Not sure why x86_64 is not seeing this currently
meta/recipes-extended/mdadm/mdadm_3.2.2.bb | 3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/meta/recipes-extended/mdadm/mdadm_3.2.2.bb b/meta/recipes-extended/mdadm/mdadm_3.2.2.bb
index 5d29ae7..1c83f89 100644
--- a/meta/recipes-extended/mdadm/mdadm_3.2.2.bb
+++ b/meta/recipes-extended/mdadm/mdadm_3.2.2.bb
@@ -8,7 +8,7 @@ LIC_FILES_CHKSUM = "file://COPYING;md5=0636e73ff0215e8d672dc4c32c317bb3 \
file://mdmon.c;beginline=4;endline=18;md5=af7d8444d9c4d3e5c7caac0d9d34039d \
file://mdadm.h;beglinlne=4;endline=22;md5=462bc9936ac0d3da110191a3f9994161"
-PR = "r0"
+PR = "r1"
SRC_URI = "${KERNELORG_MIRROR}/linux/utils/raid/mdadm/${BPN}-${PV}.tar.bz2"
@@ -16,6 +16,7 @@ SRC_URI[md5sum] = "12ee2fbf3beddb60601fb7a4c4905651"
SRC_URI[sha256sum] = "0d1a04e688b082bc88846e3f524abd50bc782b6ffc06123140f7d358c8f9b906"
CFLAGS += "-fno-strict-aliasing"
+EXTRA_OEMAKE_append_powerpc64 = "CXFLAGS=-Wno-error=format"
inherit autotools
--
1.7.6.1
^ permalink raw reply related [flat|nested] 25+ messages in thread
* [PATCH 2/6] Fix perl for powerpc64 build
2011-10-04 22:03 [PATCH 1/6] Fix mdadm for powerpc64 Matthew McClintock
@ 2011-10-04 22:03 ` Matthew McClintock
2011-10-05 11:46 ` Richard Purdie
2011-10-04 22:03 ` [PATCH 3/6] Add new IMAGE_CLASSES variable for classes for image generation Matthew McClintock
` (3 subsequent siblings)
4 siblings, 1 reply; 25+ messages in thread
From: Matthew McClintock @ 2011-10-04 22:03 UTC (permalink / raw)
To: openembedded-core
Build environment for ARCH was set to powerpc on a powerpc64
build, this adds an override to set this to powerpc64 when
doing this type of build
Signed-off-by: Matthew McClintock <msm@freescale.com>
---
No comments on original patch:
http://lists.linuxtogo.org/pipermail/openembedded-core/2011-September/010340.html
meta/recipes-devtools/perl/perl_5.12.3.bb | 6 ++++--
1 files changed, 4 insertions(+), 2 deletions(-)
diff --git a/meta/recipes-devtools/perl/perl_5.12.3.bb b/meta/recipes-devtools/perl/perl_5.12.3.bb
index e4769a4..d1ac0e8 100644
--- a/meta/recipes-devtools/perl/perl_5.12.3.bb
+++ b/meta/recipes-devtools/perl/perl_5.12.3.bb
@@ -7,7 +7,7 @@ LIC_FILES_CHKSUM = "file://Copying;md5=2b4c6ffbcfcbdee469f02565f253d81a \
# We need gnugrep (for -I)
DEPENDS = "virtual/db grep-native"
DEPENDS += "gdbm zlib"
-PR = "r5"
+PR = "r6"
# 5.10.1 has Module::Build built-in
PROVIDES += "libmodule-build-perl"
@@ -95,6 +95,8 @@ export LDDLFLAGS = "${LDFLAGS} -shared"
# We're almost Debian, aren't we?
CFLAGS += "-DDEBIAN"
+export ARCH_powerpc64 = "powerpc64"
+
do_nolargefile() {
sed -i -e "s,\(uselargefiles=\)'define',\1'undef',g" \
-e "s,\(d_readdir64_r=\)'define',\1'undef',g" \
@@ -156,7 +158,7 @@ do_configure() {
config.sh-${TARGET_ARCH}-${TARGET_OS}
case "${TARGET_ARCH}" in
- x86_64 | powerpc | s390)
+ x86_64 | powerpc | powerpc64 | s390)
sed -i -e "s,\(need_va_copy=\)'undef',\1'define',g" \
config.sh-${TARGET_ARCH}-${TARGET_OS}
;;
--
1.7.6.1
^ permalink raw reply related [flat|nested] 25+ messages in thread
* [PATCH 3/6] Add new IMAGE_CLASSES variable for classes for image generation
2011-10-04 22:03 [PATCH 1/6] Fix mdadm for powerpc64 Matthew McClintock
2011-10-04 22:03 ` [PATCH 2/6] Fix perl for powerpc64 build Matthew McClintock
@ 2011-10-04 22:03 ` Matthew McClintock
2011-10-05 22:52 ` McClintock Matthew-B29882
2011-10-04 22:03 ` [PATCH 4/6] scripts/oe-buildenv-internal: Add SOCKS5_{USER, PASSWD} to BB_ENV_EXTRAWHITE Matthew McClintock
` (2 subsequent siblings)
4 siblings, 1 reply; 25+ messages in thread
From: Matthew McClintock @ 2011-10-04 22:03 UTC (permalink / raw)
To: openembedded-core
Allows us to import classes only for images and not to the global
namespace
Signed-off-by: Matthew McClintock <msm@freescale.com>
---
Work around to solve issue from this thread:
https://lists.yoctoproject.org/pipermail/yocto/2011-September/005059.html
meta/classes/image.bbclass | 3 ++-
1 files changed, 2 insertions(+), 1 deletions(-)
diff --git a/meta/classes/image.bbclass b/meta/classes/image.bbclass
index f17e989..f03df5b 100644
--- a/meta/classes/image.bbclass
+++ b/meta/classes/image.bbclass
@@ -111,7 +111,8 @@ def get_devtable_list(d):
str += " %s" % bb.which(bb.data.getVar('BBPATH', d, 1), devtable)
return str
-inherit image_types
+IMAGE_CLASESS ??= ""
+inherit image_types ${IMAGE_CLASSES}
IMAGE_POSTPROCESS_COMMAND ?= ""
MACHINE_POSTPROCESS_COMMAND ?= ""
--
1.7.6.1
^ permalink raw reply related [flat|nested] 25+ messages in thread
* [PATCH 4/6] scripts/oe-buildenv-internal: Add SOCKS5_{USER, PASSWD} to BB_ENV_EXTRAWHITE
2011-10-04 22:03 [PATCH 1/6] Fix mdadm for powerpc64 Matthew McClintock
2011-10-04 22:03 ` [PATCH 2/6] Fix perl for powerpc64 build Matthew McClintock
2011-10-04 22:03 ` [PATCH 3/6] Add new IMAGE_CLASSES variable for classes for image generation Matthew McClintock
@ 2011-10-04 22:03 ` Matthew McClintock
2011-10-05 13:25 ` Richard Purdie
2011-10-04 22:03 ` [PATCH 5/6] Fix sysprof for powerpc64 Matthew McClintock
2011-10-04 22:03 ` [PATCH 6/6] Bring libnl up to date with meta-openebedded Matthew McClintock
4 siblings, 1 reply; 25+ messages in thread
From: Matthew McClintock @ 2011-10-04 22:03 UTC (permalink / raw)
To: openembedded-core
From: Kumar Gala <galak@kernel.crashing.org>
If a SOCKS5 gateway is needed for a proxy access like git it might also
require authentication to the proxy via a password and username. Adding
SOCKS5_USER & SOCKS5_PASSWD to BB_ENV_EXTRAWHITE allow for automation
of the authentication request to occur when something like a git fetch
is going through the proxy.
This patch requires the bitbake patch to add extra exportvars so
these variables get passed from Env -> bitbake -> fetcher
Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
Signed-off-by: Matthew McClintock <msm@freescale.com>
---
This requires a bitbake patch that was just sent:
http://lists.linuxtogo.org/pipermail/bitbake-devel/2011-October/001453.html
scripts/oe-buildenv-internal | 3 +--
1 files changed, 1 insertions(+), 2 deletions(-)
diff --git a/scripts/oe-buildenv-internal b/scripts/oe-buildenv-internal
index 61ac18c..21b92b0 100755
--- a/scripts/oe-buildenv-internal
+++ b/scripts/oe-buildenv-internal
@@ -70,5 +70,4 @@ unset BITBAKEDIR
# Used by the runqemu script
export BUILDDIR
export PATH
-
-export BB_ENV_EXTRAWHITE="MACHINE DISTRO TCMODE TCLIBC http_proxy ftp_proxy https_proxy all_proxy ALL_PROXY no_proxy SSH_AGENT_PID SSH_AUTH_SOCK BB_SRCREV_POLICY SDKMACHINE BB_NUMBER_THREADS PARALLEL_MAKE GIT_PROXY_COMMAND"
+export BB_ENV_EXTRAWHITE="MACHINE DISTRO TCMODE TCLIBC http_proxy ftp_proxy https_proxy all_proxy ALL_PROXY no_proxy SSH_AGENT_PID SSH_AUTH_SOCK BB_SRCREV_POLICY SDKMACHINE BB_NUMBER_THREADS PARALLEL_MAKE GIT_PROXY_COMMAND GIT_PROXY_IGNORE SOCKS5_PASSWD SOCKS5_USER"
--
1.7.6.1
^ permalink raw reply related [flat|nested] 25+ messages in thread
* [PATCH 5/6] Fix sysprof for powerpc64
2011-10-04 22:03 [PATCH 1/6] Fix mdadm for powerpc64 Matthew McClintock
` (2 preceding siblings ...)
2011-10-04 22:03 ` [PATCH 4/6] scripts/oe-buildenv-internal: Add SOCKS5_{USER, PASSWD} to BB_ENV_EXTRAWHITE Matthew McClintock
@ 2011-10-04 22:03 ` Matthew McClintock
2011-10-05 0:05 ` Khem Raj
2011-10-04 22:03 ` [PATCH 6/6] Bring libnl up to date with meta-openebedded Matthew McClintock
4 siblings, 1 reply; 25+ messages in thread
From: Matthew McClintock @ 2011-10-04 22:03 UTC (permalink / raw)
To: openembedded-core
sysprof will not build properly without this defined
Signed-off-by: Matthew McClintock <msm@freescale.com>
---
No comments on original patch sent
meta/recipes-kernel/sysprof/sysprof_git.bb | 2 ++
1 files changed, 2 insertions(+), 0 deletions(-)
diff --git a/meta/recipes-kernel/sysprof/sysprof_git.bb b/meta/recipes-kernel/sysprof/sysprof_git.bb
index 10bde04..271b5d8 100644
--- a/meta/recipes-kernel/sysprof/sysprof_git.bb
+++ b/meta/recipes-kernel/sysprof/sysprof_git.bb
@@ -16,6 +16,8 @@ SRC_URI_append_arm = " file://rmb-arm.patch"
SRC_URI_append_mips = " file://rmb-mips.patch"
SRC_URI_append_powerpc = " file://ppc-macro-fix.patch"
+export CFLAGS_append_powerpc64 = " -D__ppc64__"
+
S = "${WORKDIR}/git"
inherit autotools
--
1.7.6.1
^ permalink raw reply related [flat|nested] 25+ messages in thread
* [PATCH 6/6] Bring libnl up to date with meta-openebedded
2011-10-04 22:03 [PATCH 1/6] Fix mdadm for powerpc64 Matthew McClintock
` (3 preceding siblings ...)
2011-10-04 22:03 ` [PATCH 5/6] Fix sysprof for powerpc64 Matthew McClintock
@ 2011-10-04 22:03 ` Matthew McClintock
4 siblings, 0 replies; 25+ messages in thread
From: Matthew McClintock @ 2011-10-04 22:03 UTC (permalink / raw)
To: openembedded-core
Signed-off-by: Matthew McClintock <msm@freescale.com>
---
This is a blind copy of libnl-2.0 from meta-openembedded
Not sure how to completly verify this is working, feel free
to squash the previous patch with this one
.../libnl/libnl-2.0/fix-makefile.patch | 32 --------------------
meta/recipes-support/libnl/libnl.inc | 9 +++++
.../recipes-support/libnl/libnl/fix-makefile.patch | 32 ++++++++++++++++++++
meta/recipes-support/libnl/libnl/fix-pc-file.patch | 11 +++++++
.../libnl/libnl/fix-pktloc-dep-race.patch | 20 ++++++++++++
meta/recipes-support/libnl/libnl_2.0.bb | 24 ++++++++------
6 files changed, 86 insertions(+), 42 deletions(-)
delete mode 100644 meta/recipes-support/libnl/libnl-2.0/fix-makefile.patch
create mode 100644 meta/recipes-support/libnl/libnl.inc
create mode 100644 meta/recipes-support/libnl/libnl/fix-makefile.patch
create mode 100644 meta/recipes-support/libnl/libnl/fix-pc-file.patch
create mode 100644 meta/recipes-support/libnl/libnl/fix-pktloc-dep-race.patch
diff --git a/meta/recipes-support/libnl/libnl-2.0/fix-makefile.patch b/meta/recipes-support/libnl/libnl-2.0/fix-makefile.patch
deleted file mode 100644
index 3e88fbd..0000000
--- a/meta/recipes-support/libnl/libnl-2.0/fix-makefile.patch
+++ /dev/null
@@ -1,32 +0,0 @@
-Upstream-Status: Pending
-
-12/03/2010
-
-add explicit rules for header files generated by lex and yacc,
-otherwise the build of lib/route/pktloc.c may fail in a parallel
-environment.
-
-Signed-off-by: Qing He <qing.he@intel.com>
-
-12/06/2010
-
-the dependency rule should really read pktloc.lo instead of
-pktloc.$(OBJEXT), since it's in a libtool setup.
-
-Signed-off-by: Qing He <qing.he@intel.com>
-
-diff --git a/lib/Makefile.am b/lib/Makefile.am
-index 92a916e..e8b8ef3 100644
---- a/lib/Makefile.am
-+++ b/lib/Makefile.am
-@@ -35,6 +35,10 @@ route/pktloc_grammar.c: route/pktloc_grammar.l
- route/pktloc_syntax.c: route/pktloc_syntax.y
- $(YACC) -d $(YFLAGS) -o $@ $^
-
-+route/pktloc.lo: route/pktloc_syntax.h route/pktloc_grammar.h
-+route/pktloc_syntax.h: route/pktloc_syntax.c
-+route/pktloc_grammar.h: route/pktloc_grammar.c
-+
- libnl_route_la_LDFLAGS = -version-info 2:0:0
- libnl_route_la_LIBADD = libnl.la
- libnl_route_la_SOURCES = \
diff --git a/meta/recipes-support/libnl/libnl.inc b/meta/recipes-support/libnl/libnl.inc
new file mode 100644
index 0000000..6f502b1
--- /dev/null
+++ b/meta/recipes-support/libnl/libnl.inc
@@ -0,0 +1,9 @@
+DESCRIPTION = "libnl is a library for applications dealing with netlink sockets"
+SECTION = "libs/network"
+LICENSE = "LGPL"
+HOMEPAGE = "http://www.infradead.org/~tgr/libnl/"
+
+INC_PR = "r5"
+
+inherit autotools pkgconfig
+
diff --git a/meta/recipes-support/libnl/libnl/fix-makefile.patch b/meta/recipes-support/libnl/libnl/fix-makefile.patch
new file mode 100644
index 0000000..3e88fbd
--- /dev/null
+++ b/meta/recipes-support/libnl/libnl/fix-makefile.patch
@@ -0,0 +1,32 @@
+Upstream-Status: Pending
+
+12/03/2010
+
+add explicit rules for header files generated by lex and yacc,
+otherwise the build of lib/route/pktloc.c may fail in a parallel
+environment.
+
+Signed-off-by: Qing He <qing.he@intel.com>
+
+12/06/2010
+
+the dependency rule should really read pktloc.lo instead of
+pktloc.$(OBJEXT), since it's in a libtool setup.
+
+Signed-off-by: Qing He <qing.he@intel.com>
+
+diff --git a/lib/Makefile.am b/lib/Makefile.am
+index 92a916e..e8b8ef3 100644
+--- a/lib/Makefile.am
++++ b/lib/Makefile.am
+@@ -35,6 +35,10 @@ route/pktloc_grammar.c: route/pktloc_grammar.l
+ route/pktloc_syntax.c: route/pktloc_syntax.y
+ $(YACC) -d $(YFLAGS) -o $@ $^
+
++route/pktloc.lo: route/pktloc_syntax.h route/pktloc_grammar.h
++route/pktloc_syntax.h: route/pktloc_syntax.c
++route/pktloc_grammar.h: route/pktloc_grammar.c
++
+ libnl_route_la_LDFLAGS = -version-info 2:0:0
+ libnl_route_la_LIBADD = libnl.la
+ libnl_route_la_SOURCES = \
diff --git a/meta/recipes-support/libnl/libnl/fix-pc-file.patch b/meta/recipes-support/libnl/libnl/fix-pc-file.patch
new file mode 100644
index 0000000..17666fb
--- /dev/null
+++ b/meta/recipes-support/libnl/libnl/fix-pc-file.patch
@@ -0,0 +1,11 @@
+Index: libnl-2.0/libnl-2.0.pc.in
+===================================================================
+--- libnl-2.0.orig/libnl-2.0.pc.in
++++ libnl-2.0/libnl-2.0.pc.in
+@@ -6,5 +6,5 @@
+ Name: libnl
+ Description: Convenience library for netlink sockets
+ Version: @PACKAGE_VERSION@
+-Libs: -L${libdir} -lnl
++Libs: -L${libdir} -lnl -lnl-genl -lnl-nf -lnl-route
+ Cflags: -I${includedir}
diff --git a/meta/recipes-support/libnl/libnl/fix-pktloc-dep-race.patch b/meta/recipes-support/libnl/libnl/fix-pktloc-dep-race.patch
new file mode 100644
index 0000000..ee21744
--- /dev/null
+++ b/meta/recipes-support/libnl/libnl/fix-pktloc-dep-race.patch
@@ -0,0 +1,20 @@
+Index: libnl-2.0/lib/Makefile.am
+===================================================================
+--- libnl-2.0.orig/lib/Makefile.am
++++ libnl-2.0/lib/Makefile.am
+@@ -27,11 +27,15 @@ CLEANFILES = \
+ route/pktloc_grammar.c route/pktloc_grammar.h \
+ route/pktloc_syntax.c route/pktloc_syntax.h
+
++BUILT_SOURCES = route/pktloc_syntax.h route/pktloc_grammar.h
++
+ # Hack to avoid using ylwrap. It does not function correctly in combination
+ # with --header-file=
++route/pktloc_grammar.h: route/pktloc_grammar.c
+ route/pktloc_grammar.c: route/pktloc_grammar.l
+ $(LEX) --header-file=route/pktloc_grammar.h $(LFLAGS) -o $@ $^
+
++route/pktloc_syntax.h: route/pktloc_syntax.c
+ route/pktloc_syntax.c: route/pktloc_syntax.y
+ $(YACC) -d $(YFLAGS) -o $@ $^
+
diff --git a/meta/recipes-support/libnl/libnl_2.0.bb b/meta/recipes-support/libnl/libnl_2.0.bb
index 0dfcaf6..f2c88aa 100644
--- a/meta/recipes-support/libnl/libnl_2.0.bb
+++ b/meta/recipes-support/libnl/libnl_2.0.bb
@@ -1,20 +1,24 @@
-DESCRIPTION = "libnl is a library for applications dealing with netlink sockets."
-HOMEPAGE = "http://www.infradead.org/~tgr/libnl/"
-SECTION = "libs/network"
+require libnl.inc
+PE = "1"
+PR = "${INC_PR}.0"
LICENSE = "LGPLv2.1"
LIC_FILES_CHKSUM = "file://COPYING;md5=2b41e13261a330ee784153ecbb6a82bc"
DEPENDS = "flex-native bison-native"
-PR = "r2"
-SRC_URI= "http://www.infradead.org/~tgr/libnl/files/${BPN}-${PV}.tar.gz \
- file://fix-makefile.patch \
- "
+SRC_URI = "\
+ http://www.infradead.org/~tgr/libnl/files/${BPN}-${PV}.tar.gz \
+ file://fix-pc-file.patch \
+ file://fix-pktloc-dep-race.patch \
+ file://fix-makefile.patch \
+"
SRC_URI[md5sum] = "6aaf1e9802a17a7d702bb0638044ffa7"
SRC_URI[sha256sum] = "5a40dc903d3ca1074da7424b908bec8ff16936484798c7e46e53e9db8bc87a9c"
-inherit autotools pkgconfig
-
-LEAD_SONAME = "libnl.so"
+PACKAGES =+ "${PN}-route ${PN}-nf ${PN}-genl ${PN}-cli"
+FILES_${PN}-route = "${libdir}/libnl-route.so.*"
+FILES_${PN}-nf = "${libdir}/libnl-nf.so.*"
+FILES_${PN}-genl = "${libdir}/libnl-genl.so.*"
+FILES_${PN}-cli = "${libdir}/libnl-cli.so.*"
--
1.7.6.1
^ permalink raw reply related [flat|nested] 25+ messages in thread
* Re: [PATCH 5/6] Fix sysprof for powerpc64
2011-10-04 22:03 ` [PATCH 5/6] Fix sysprof for powerpc64 Matthew McClintock
@ 2011-10-05 0:05 ` Khem Raj
2011-10-05 6:44 ` McClintock Matthew-B29882
2011-10-05 11:35 ` Richard Purdie
0 siblings, 2 replies; 25+ messages in thread
From: Khem Raj @ 2011-10-05 0:05 UTC (permalink / raw)
To: openembedded-core
On 10/4/2011 3:03 PM, Matthew McClintock wrote:
> sysprof will not build properly without this defined
>
> Signed-off-by: Matthew McClintock<msm@freescale.com>
> ---
> No comments on original patch sent
>
> meta/recipes-kernel/sysprof/sysprof_git.bb | 2 ++
> 1 files changed, 2 insertions(+), 0 deletions(-)
>
> diff --git a/meta/recipes-kernel/sysprof/sysprof_git.bb b/meta/recipes-kernel/sysprof/sysprof_git.bb
> index 10bde04..271b5d8 100644
> --- a/meta/recipes-kernel/sysprof/sysprof_git.bb
> +++ b/meta/recipes-kernel/sysprof/sysprof_git.bb
> @@ -16,6 +16,8 @@ SRC_URI_append_arm = " file://rmb-arm.patch"
> SRC_URI_append_mips = " file://rmb-mips.patch"
> SRC_URI_append_powerpc = " file://ppc-macro-fix.patch"
>
> +export CFLAGS_append_powerpc64 = " -D__ppc64__"
> +
this is a gcc built-in define I wonder why you need to add it explicitly
> S = "${WORKDIR}/git"
>
> inherit autotools
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [PATCH 5/6] Fix sysprof for powerpc64
2011-10-05 0:05 ` Khem Raj
@ 2011-10-05 6:44 ` McClintock Matthew-B29882
2011-10-05 11:35 ` Richard Purdie
1 sibling, 0 replies; 25+ messages in thread
From: McClintock Matthew-B29882 @ 2011-10-05 6:44 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer
On Tue, Oct 4, 2011 at 7:05 PM, Khem Raj <raj.khem@gmail.com> wrote:
>> diff --git a/meta/recipes-kernel/sysprof/sysprof_git.bb
>> b/meta/recipes-kernel/sysprof/sysprof_git.bb
>> index 10bde04..271b5d8 100644
>> --- a/meta/recipes-kernel/sysprof/sysprof_git.bb
>> +++ b/meta/recipes-kernel/sysprof/sysprof_git.bb
>> @@ -16,6 +16,8 @@ SRC_URI_append_arm = " file://rmb-arm.patch"
>> SRC_URI_append_mips = " file://rmb-mips.patch"
>> SRC_URI_append_powerpc = " file://ppc-macro-fix.patch"
>>
>> +export CFLAGS_append_powerpc64 = " -D__ppc64__"
>> +
>
> this is a gcc built-in define I wonder why you need to add it explicitly
The file in question is compiled like:
powerpc64-fsl-linux-gcc -m64 -mcpu=e5500
--sysroot=/local/home/mattsm/git/poky/build_p5020ds-64b_release/tmp/sysroots/p5020ds-64b
-DHAVE_CONFIG_H -I.
-I/local/home/mattsm/git/poky/build_p5020ds-64b_release/tmp/sysroots/p5020ds-64b/usr/include/glib-2.0
-I/local/home/mattsm/git/poky/build_p5020ds-64b_release/tmp/sysroots/p5020ds-64b/usr/lib64/glib-2.0/include
-O2 -pipe -g -feliminate-unused-debug-types -Wall -MT
sysprof_cli-collector.o -MD -MP -MF .deps/sysprof_cli-collector.Tpo -c
-o sysprof_cli-collector.o `test -f 'collector.c' || echo
'./'`collector.c
Without the -D__ppc64__ it gives the following:
collector.c: In function ‘sysprof_perf_counter_open’:
collector.c:193:21: error: ‘__NR_perf_counter_open’ undeclared (first
use in this function)
collector.c:193:21: note: each undeclared identifier is reported only
once for each function it appears in
collector.c: In function ‘process_event’:
collector.c:654:11: warning: variable ‘name’ set but not used
[-Wunused-but-set-variable]
collector.c: In function ‘sysprof_perf_counter_open’:
collector.c:194:1: warning: control reaches end of non-void function
[-Wreturn-type]
Looking at collector.c it's very simple:
#ifndef __NR_perf_counter_open
#if defined(__i386__)
#define __NR_perf_counter_open 336
#elif defined(__x86_64__)
#define __NR_perf_counter_open 298
[...snip...]
#elif defined(__hppa__)
#define __NR_perf_counter_open 318
#elif defined(__ppc__) || defined(__ppc64__)
#define __NR_perf_counter_open 319
#elif defined(__s390__)
[...snip...]
#endif
#endif
Am I missing something obvious?
-M
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [PATCH 5/6] Fix sysprof for powerpc64
2011-10-05 0:05 ` Khem Raj
2011-10-05 6:44 ` McClintock Matthew-B29882
@ 2011-10-05 11:35 ` Richard Purdie
2011-10-05 16:46 ` Khem Raj
` (2 more replies)
1 sibling, 3 replies; 25+ messages in thread
From: Richard Purdie @ 2011-10-05 11:35 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer
On Tue, 2011-10-04 at 17:05 -0700, Khem Raj wrote:
> On 10/4/2011 3:03 PM, Matthew McClintock wrote:
> > sysprof will not build properly without this defined
> >
> > Signed-off-by: Matthew McClintock<msm@freescale.com>
> > ---
> > No comments on original patch sent
> >
> > meta/recipes-kernel/sysprof/sysprof_git.bb | 2 ++
> > 1 files changed, 2 insertions(+), 0 deletions(-)
> >
> > diff --git a/meta/recipes-kernel/sysprof/sysprof_git.bb b/meta/recipes-kernel/sysprof/sysprof_git.bb
> > index 10bde04..271b5d8 100644
> > --- a/meta/recipes-kernel/sysprof/sysprof_git.bb
> > +++ b/meta/recipes-kernel/sysprof/sysprof_git.bb
> > @@ -16,6 +16,8 @@ SRC_URI_append_arm = " file://rmb-arm.patch"
> > SRC_URI_append_mips = " file://rmb-mips.patch"
> > SRC_URI_append_powerpc = " file://ppc-macro-fix.patch"
> >
> > +export CFLAGS_append_powerpc64 = " -D__ppc64__"
> > +
>
> this is a gcc built-in define I wonder why you need to add it explicitly
Are you sure? Is the built-in define not __powerpc64__ ?
Cheers,
Richard
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [PATCH 2/6] Fix perl for powerpc64 build
2011-10-04 22:03 ` [PATCH 2/6] Fix perl for powerpc64 build Matthew McClintock
@ 2011-10-05 11:46 ` Richard Purdie
2011-10-05 19:04 ` McClintock Matthew-B29882
0 siblings, 1 reply; 25+ messages in thread
From: Richard Purdie @ 2011-10-05 11:46 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer
On Tue, 2011-10-04 at 17:03 -0500, Matthew McClintock wrote:
> Build environment for ARCH was set to powerpc on a powerpc64
> build, this adds an override to set this to powerpc64 when
> doing this type of build
>
> Signed-off-by: Matthew McClintock <msm@freescale.com>
> ---
> No comments on original patch:
>
> http://lists.linuxtogo.org/pipermail/openembedded-core/2011-September/010340.html
>
> meta/recipes-devtools/perl/perl_5.12.3.bb | 6 ++++--
> 1 files changed, 4 insertions(+), 2 deletions(-)
>
> diff --git a/meta/recipes-devtools/perl/perl_5.12.3.bb b/meta/recipes-devtools/perl/perl_5.12.3.bb
> index e4769a4..d1ac0e8 100644
> --- a/meta/recipes-devtools/perl/perl_5.12.3.bb
> +++ b/meta/recipes-devtools/perl/perl_5.12.3.bb
> @@ -7,7 +7,7 @@ LIC_FILES_CHKSUM = "file://Copying;md5=2b4c6ffbcfcbdee469f02565f253d81a \
> # We need gnugrep (for -I)
> DEPENDS = "virtual/db grep-native"
> DEPENDS += "gdbm zlib"
> -PR = "r5"
> +PR = "r6"
>
> # 5.10.1 has Module::Build built-in
> PROVIDES += "libmodule-build-perl"
> @@ -95,6 +95,8 @@ export LDDLFLAGS = "${LDFLAGS} -shared"
> # We're almost Debian, aren't we?
> CFLAGS += "-DDEBIAN"
>
> +export ARCH_powerpc64 = "powerpc64"
> +
I'm finding it really hard to believe this makes any difference. I just
tried a powerpc64 build here and perl built fine. I can't find any
evidence for any part of the perl build using an ARCH variable
(TARGET_ARCH is used but that isn't ARCH).
Can you give some further information about how/where this is used?
> do_nolargefile() {
> sed -i -e "s,\(uselargefiles=\)'define',\1'undef',g" \
> -e "s,\(d_readdir64_r=\)'define',\1'undef',g" \
> @@ -156,7 +158,7 @@ do_configure() {
> config.sh-${TARGET_ARCH}-${TARGET_OS}
>
> case "${TARGET_ARCH}" in
> - x86_64 | powerpc | s390)
> + x86_64 | powerpc | powerpc64 | s390)
> sed -i -e "s,\(need_va_copy=\)'undef',\1'define',g" \
> config.sh-${TARGET_ARCH}-${TARGET_OS}
> ;;
This bit is fine...
Cheers,
Richard
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [PATCH 4/6] scripts/oe-buildenv-internal: Add SOCKS5_{USER, PASSWD} to BB_ENV_EXTRAWHITE
2011-10-04 22:03 ` [PATCH 4/6] scripts/oe-buildenv-internal: Add SOCKS5_{USER, PASSWD} to BB_ENV_EXTRAWHITE Matthew McClintock
@ 2011-10-05 13:25 ` Richard Purdie
0 siblings, 0 replies; 25+ messages in thread
From: Richard Purdie @ 2011-10-05 13:25 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer
On Tue, 2011-10-04 at 17:03 -0500, Matthew McClintock wrote:
> From: Kumar Gala <galak@kernel.crashing.org>
>
> If a SOCKS5 gateway is needed for a proxy access like git it might also
> require authentication to the proxy via a password and username. Adding
> SOCKS5_USER & SOCKS5_PASSWD to BB_ENV_EXTRAWHITE allow for automation
> of the authentication request to occur when something like a git fetch
> is going through the proxy.
>
> This patch requires the bitbake patch to add extra exportvars so
> these variables get passed from Env -> bitbake -> fetcher
>
> Signed-off-by: Kumar Gala <galak@kernel.crashing.org>
> Signed-off-by: Matthew McClintock <msm@freescale.com>
> ---
> This requires a bitbake patch that was just sent:
>
> http://lists.linuxtogo.org/pipermail/bitbake-devel/2011-October/001453.html
Merged to master, thanks.
Richard
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [PATCH 5/6] Fix sysprof for powerpc64
2011-10-05 11:35 ` Richard Purdie
@ 2011-10-05 16:46 ` Khem Raj
2011-10-05 16:48 ` McClintock Matthew-B29882
2011-10-05 18:30 ` Phil Blundell
2 siblings, 0 replies; 25+ messages in thread
From: Khem Raj @ 2011-10-05 16:46 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer
On Wed, Oct 5, 2011 at 4:35 AM, Richard Purdie
<richard.purdie@linuxfoundation.org> wrote:
> On Tue, 2011-10-04 at 17:05 -0700, Khem Raj wrote:
>> On 10/4/2011 3:03 PM, Matthew McClintock wrote:
>> > sysprof will not build properly without this defined
>> >
>> > Signed-off-by: Matthew McClintock<msm@freescale.com>
>> > ---
>> > No comments on original patch sent
>> >
>> > meta/recipes-kernel/sysprof/sysprof_git.bb | 2 ++
>> > 1 files changed, 2 insertions(+), 0 deletions(-)
>> >
>> > diff --git a/meta/recipes-kernel/sysprof/sysprof_git.bb b/meta/recipes-kernel/sysprof/sysprof_git.bb
>> > index 10bde04..271b5d8 100644
>> > --- a/meta/recipes-kernel/sysprof/sysprof_git.bb
>> > +++ b/meta/recipes-kernel/sysprof/sysprof_git.bb
>> > @@ -16,6 +16,8 @@ SRC_URI_append_arm = " file://rmb-arm.patch"
>> > SRC_URI_append_mips = " file://rmb-mips.patch"
>> > SRC_URI_append_powerpc = " file://ppc-macro-fix.patch"
>> >
>> > +export CFLAGS_append_powerpc64 = " -D__ppc64__"
>> > +
>>
>> this is a gcc built-in define I wonder why you need to add it explicitly
>
> Are you sure? Is the built-in define not __powerpc64__ ?
yes that one too is a builtin define. I don't have the OE/ppc64 build
around to explore more
>
> Cheers,
>
> Richard
>
>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [PATCH 5/6] Fix sysprof for powerpc64
2011-10-05 11:35 ` Richard Purdie
2011-10-05 16:46 ` Khem Raj
@ 2011-10-05 16:48 ` McClintock Matthew-B29882
2011-10-05 17:06 ` Khem Raj
2011-10-05 18:30 ` Phil Blundell
2 siblings, 1 reply; 25+ messages in thread
From: McClintock Matthew-B29882 @ 2011-10-05 16:48 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer
On Wed, Oct 5, 2011 at 6:35 AM, Richard Purdie
<richard.purdie@linuxfoundation.org> wrote:
>> this is a gcc built-in define I wonder why you need to add it explicitly
>
> Are you sure? Is the built-in define not __powerpc64__ ?
Further investigation:
[mattsm@right build_p5020ds-64b_release (testing $)]$
./tmp/sysroots/x86_64-linux/usr/bin/ppc64e5500-fsl-linux/powerpc64-fsl-linux-gcc
-c -E -dM empty.c | grep __p
#define __powerpc64__ 1
#define __powerpc__ 1
[mattsm@right build_p5020ds-64b_release (testing $)]$
-M
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [PATCH 5/6] Fix sysprof for powerpc64
2011-10-05 16:48 ` McClintock Matthew-B29882
@ 2011-10-05 17:06 ` Khem Raj
2011-10-05 17:10 ` Saul Wold
0 siblings, 1 reply; 25+ messages in thread
From: Khem Raj @ 2011-10-05 17:06 UTC (permalink / raw)
To: McClintock Matthew-B29882,
Patches and discussions about the oe-core layer
On Wed, Oct 5, 2011 at 9:48 AM, McClintock Matthew-B29882
<B29882@freescale.com> wrote:
> On Wed, Oct 5, 2011 at 6:35 AM, Richard Purdie
> <richard.purdie@linuxfoundation.org> wrote:
>>> this is a gcc built-in define I wonder why you need to add it explicitly
>>
>> Are you sure? Is the built-in define not __powerpc64__ ?
>
> Further investigation:
>
> [mattsm@right build_p5020ds-64b_release (testing $)]$
> ./tmp/sysroots/x86_64-linux/usr/bin/ppc64e5500-fsl-linux/powerpc64-fsl-linux-gcc
> -c -E -dM empty.c | grep __p
> #define __powerpc64__ 1
> #define __powerpc__ 1
> [mattsm@right build_p5020ds-64b_release (testing $)]$
yes I was trying it with a differently configured gcc. For linux gcc
defines __powerpc64__
and for darwin it has __ppc64__ as builtin define.
I think adding check for __powerpc64__ and __powerpc__
would be nice and can be submitted to sysprof upstream IMO
>
> -M
>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [PATCH 5/6] Fix sysprof for powerpc64
2011-10-05 17:06 ` Khem Raj
@ 2011-10-05 17:10 ` Saul Wold
0 siblings, 0 replies; 25+ messages in thread
From: Saul Wold @ 2011-10-05 17:10 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer; +Cc: McClintock Matthew-B29882
On 10/05/2011 10:06 AM, Khem Raj wrote:
> On Wed, Oct 5, 2011 at 9:48 AM, McClintock Matthew-B29882
> <B29882@freescale.com> wrote:
>> On Wed, Oct 5, 2011 at 6:35 AM, Richard Purdie
>> <richard.purdie@linuxfoundation.org> wrote:
>>>> this is a gcc built-in define I wonder why you need to add it explicitly
>>>
>>> Are you sure? Is the built-in define not __powerpc64__ ?
>>
>> Further investigation:
>>
>> [mattsm@right build_p5020ds-64b_release (testing $)]$
>> ./tmp/sysroots/x86_64-linux/usr/bin/ppc64e5500-fsl-linux/powerpc64-fsl-linux-gcc
>> -c -E -dM empty.c | grep __p
>> #define __powerpc64__ 1
>> #define __powerpc__ 1
>> [mattsm@right build_p5020ds-64b_release (testing $)]$
>
> yes I was trying it with a differently configured gcc. For linux gcc
> defines __powerpc64__
> and for darwin it has __ppc64__ as builtin define.
>
> I think adding check for __powerpc64__ and __powerpc__
> would be nice and can be submitted to sysprof upstream IMO
>
I think this would be a good fix to have in the upstream sysprof and it
would remove the CFLAGS_append from the recipe.
Sau!
>>
>> -M
>>
>> _______________________________________________
>> 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
>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [PATCH 5/6] Fix sysprof for powerpc64
2011-10-05 11:35 ` Richard Purdie
2011-10-05 16:46 ` Khem Raj
2011-10-05 16:48 ` McClintock Matthew-B29882
@ 2011-10-05 18:30 ` Phil Blundell
2011-10-05 18:41 ` McClintock Matthew-B29882
2011-10-05 19:13 ` Khem Raj
2 siblings, 2 replies; 25+ messages in thread
From: Phil Blundell @ 2011-10-05 18:30 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer
On Wed, 2011-10-05 at 12:35 +0100, Richard Purdie wrote:
> On Tue, 2011-10-04 at 17:05 -0700, Khem Raj wrote:
> > On 10/4/2011 3:03 PM, Matthew McClintock wrote:
> > > sysprof will not build properly without this defined
> > >
> > > Signed-off-by: Matthew McClintock<msm@freescale.com>
> > > ---
> > > No comments on original patch sent
> > >
> > > meta/recipes-kernel/sysprof/sysprof_git.bb | 2 ++
> > > 1 files changed, 2 insertions(+), 0 deletions(-)
> > >
> > > diff --git a/meta/recipes-kernel/sysprof/sysprof_git.bb b/meta/recipes-kernel/sysprof/sysprof_git.bb
> > > index 10bde04..271b5d8 100644
> > > --- a/meta/recipes-kernel/sysprof/sysprof_git.bb
> > > +++ b/meta/recipes-kernel/sysprof/sysprof_git.bb
> > > @@ -16,6 +16,8 @@ SRC_URI_append_arm = " file://rmb-arm.patch"
> > > SRC_URI_append_mips = " file://rmb-mips.patch"
> > > SRC_URI_append_powerpc = " file://ppc-macro-fix.patch"
> > >
> > > +export CFLAGS_append_powerpc64 = " -D__ppc64__"
> > > +
> >
> > this is a gcc built-in define I wonder why you need to add it explicitly
>
> Are you sure? Is the built-in define not __powerpc64__ ?
Yes, __powerpc64__ is the generic one. I think __ppc64__ is only a
predefine on Darwin.
p.
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [PATCH 5/6] Fix sysprof for powerpc64
2011-10-05 18:30 ` Phil Blundell
@ 2011-10-05 18:41 ` McClintock Matthew-B29882
2011-10-05 19:13 ` Khem Raj
1 sibling, 0 replies; 25+ messages in thread
From: McClintock Matthew-B29882 @ 2011-10-05 18:41 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer
On Wed, Oct 5, 2011 at 1:30 PM, Phil Blundell <philb@gnu.org> wrote:
> On Wed, 2011-10-05 at 12:35 +0100, Richard Purdie wrote:
>> On Tue, 2011-10-04 at 17:05 -0700, Khem Raj wrote:
>> > On 10/4/2011 3:03 PM, Matthew McClintock wrote:
>> > > sysprof will not build properly without this defined
>> > >
>> > > Signed-off-by: Matthew McClintock<msm@freescale.com>
>> > > ---
>> > > No comments on original patch sent
>> > >
>> > > meta/recipes-kernel/sysprof/sysprof_git.bb | 2 ++
>> > > 1 files changed, 2 insertions(+), 0 deletions(-)
>> > >
>> > > diff --git a/meta/recipes-kernel/sysprof/sysprof_git.bb b/meta/recipes-kernel/sysprof/sysprof_git.bb
>> > > index 10bde04..271b5d8 100644
>> > > --- a/meta/recipes-kernel/sysprof/sysprof_git.bb
>> > > +++ b/meta/recipes-kernel/sysprof/sysprof_git.bb
>> > > @@ -16,6 +16,8 @@ SRC_URI_append_arm = " file://rmb-arm.patch"
>> > > SRC_URI_append_mips = " file://rmb-mips.patch"
>> > > SRC_URI_append_powerpc = " file://ppc-macro-fix.patch"
>> > >
>> > > +export CFLAGS_append_powerpc64 = " -D__ppc64__"
>> > > +
>> >
>> > this is a gcc built-in define I wonder why you need to add it explicitly
>>
>> Are you sure? Is the built-in define not __powerpc64__ ?
>
> Yes, __powerpc64__ is the generic one. I think __ppc64__ is only a
> predefine on Darwin.
This seems to be recently fixed upstream. I will just add a patch for
this, it's not even in 1.1.8 which we just updated too in oe-core.
-M
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [PATCH 2/6] Fix perl for powerpc64 build
2011-10-05 11:46 ` Richard Purdie
@ 2011-10-05 19:04 ` McClintock Matthew-B29882
2011-10-07 15:52 ` Richard Purdie
0 siblings, 1 reply; 25+ messages in thread
From: McClintock Matthew-B29882 @ 2011-10-05 19:04 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer
On Wed, Oct 5, 2011 at 6:46 AM, Richard Purdie
<richard.purdie@linuxfoundation.org> wrote:
>> --- a/meta/recipes-devtools/perl/perl_5.12.3.bb
>> +++ b/meta/recipes-devtools/perl/perl_5.12.3.bb
>> @@ -7,7 +7,7 @@ LIC_FILES_CHKSUM = "file://Copying;md5=2b4c6ffbcfcbdee469f02565f253d81a \
>> # We need gnugrep (for -I)
>> DEPENDS = "virtual/db grep-native"
>> DEPENDS += "gdbm zlib"
>> -PR = "r5"
>> +PR = "r6"
>>
>> # 5.10.1 has Module::Build built-in
>> PROVIDES += "libmodule-build-perl"
>> @@ -95,6 +95,8 @@ export LDDLFLAGS = "${LDFLAGS} -shared"
>> # We're almost Debian, aren't we?
>> CFLAGS += "-DDEBIAN"
>>
>> +export ARCH_powerpc64 = "powerpc64"
>> +
>
> I'm finding it really hard to believe this makes any difference. I just
> tried a powerpc64 build here and perl built fine. I can't find any
> evidence for any part of the perl build using an ARCH variable
> (TARGET_ARCH is used but that isn't ARCH).
>
> Can you give some further information about how/where this is used?
I'm trying to reproduce this again on 'edison' as well and it does not
appear to be occurring anymore! The problem before had to do with a
configuration script error:
config.sh-powerpc64-linux
This script was getting invoked as 'config.sh-powerpc-linux' on the
64bit build, then the file did not exist and the configure step failed
out. Not sure if this was somehow fixed in a different manner.
Please drop this patch for now. I'll bring it up again if I see issues.
-M
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [PATCH 5/6] Fix sysprof for powerpc64
2011-10-05 18:30 ` Phil Blundell
2011-10-05 18:41 ` McClintock Matthew-B29882
@ 2011-10-05 19:13 ` Khem Raj
1 sibling, 0 replies; 25+ messages in thread
From: Khem Raj @ 2011-10-05 19:13 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer
On Wed, Oct 5, 2011 at 11:30 AM, Phil Blundell <philb@gnu.org> wrote:
> On Wed, 2011-10-05 at 12:35 +0100, Richard Purdie wrote:
>> On Tue, 2011-10-04 at 17:05 -0700, Khem Raj wrote:
>> > On 10/4/2011 3:03 PM, Matthew McClintock wrote:
>> > > sysprof will not build properly without this defined
>> > >
>> > > Signed-off-by: Matthew McClintock<msm@freescale.com>
>> > > ---
>> > > No comments on original patch sent
>> > >
>> > > meta/recipes-kernel/sysprof/sysprof_git.bb | 2 ++
>> > > 1 files changed, 2 insertions(+), 0 deletions(-)
>> > >
>> > > diff --git a/meta/recipes-kernel/sysprof/sysprof_git.bb b/meta/recipes-kernel/sysprof/sysprof_git.bb
>> > > index 10bde04..271b5d8 100644
>> > > --- a/meta/recipes-kernel/sysprof/sysprof_git.bb
>> > > +++ b/meta/recipes-kernel/sysprof/sysprof_git.bb
>> > > @@ -16,6 +16,8 @@ SRC_URI_append_arm = " file://rmb-arm.patch"
>> > > SRC_URI_append_mips = " file://rmb-mips.patch"
>> > > SRC_URI_append_powerpc = " file://ppc-macro-fix.patch"
>> > >
>> > > +export CFLAGS_append_powerpc64 = " -D__ppc64__"
>> > > +
>> >
>> > this is a gcc built-in define I wonder why you need to add it explicitly
>>
>> Are you sure? Is the built-in define not __powerpc64__ ?
>
> Yes, __powerpc64__ is the generic one. I think __ppc64__ is only a
> predefine on Darwin.
__powerpc64__ is not generic. Its Linux specific.
>
> p.
>
>
>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [PATCH 3/6] Add new IMAGE_CLASSES variable for classes for image generation
2011-10-04 22:03 ` [PATCH 3/6] Add new IMAGE_CLASSES variable for classes for image generation Matthew McClintock
@ 2011-10-05 22:52 ` McClintock Matthew-B29882
2011-10-05 23:53 ` Richard Purdie
0 siblings, 1 reply; 25+ messages in thread
From: McClintock Matthew-B29882 @ 2011-10-05 22:52 UTC (permalink / raw)
To: openembedded-core@lists.openembedded.org
On Tue, Oct 4, 2011 at 5:03 PM, Matthew McClintock <msm@freescale.com> wrote:
> -inherit image_types
> +IMAGE_CLASESS ??= ""
> +inherit image_types ${IMAGE_CLASSES}
So this does not work properly, I thought I tested it but... need to
do something like this:
IMAGE_CLASSES = "image_types"
inherit ${IMAGE_CLASSES}
then let the user added via
IMAGE_CLASESS += "image_types_uboot"
then there will always be something that is expanded to avoid this issue:
ERROR: Could not inherit file classes/$IMAGE_CLASSES.bbclass
| ETA: --:--:--
ERROR: Command execution failed: Exited with 1
or
ERROR: Could not inherit file classes/.bbclass
| ETA: --:--:--
ERROR: Command execution failed: Exited with 1
-M
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [PATCH 3/6] Add new IMAGE_CLASSES variable for classes for image generation
2011-10-05 22:52 ` McClintock Matthew-B29882
@ 2011-10-05 23:53 ` Richard Purdie
2011-10-06 0:25 ` McClintock Matthew-B29882
0 siblings, 1 reply; 25+ messages in thread
From: Richard Purdie @ 2011-10-05 23:53 UTC (permalink / raw)
To: McClintock Matthew-B29882,
Patches and discussions about the oe-core layer
On Wed, 2011-10-05 at 22:52 +0000, McClintock Matthew-B29882 wrote:
> On Tue, Oct 4, 2011 at 5:03 PM, Matthew McClintock <msm@freescale.com> wrote:
> > -inherit image_types
> > +IMAGE_CLASESS ??= ""
> > +inherit image_types ${IMAGE_CLASSES}
>
> So this does not work properly, I thought I tested it but... need to
> do something like this:
>
> IMAGE_CLASSES = "image_types"
> inherit ${IMAGE_CLASSES}
>
> then let the user added via
>
> IMAGE_CLASESS += "image_types_uboot"
>
> then there will always be something that is expanded to avoid this issue:
>
> ERROR: Could not inherit file classes/$IMAGE_CLASSES.bbclass
> | ETA: --:--:--
> ERROR: Command execution failed: Exited with 1
Its not expanding the variables? :/ This happened even when
IMAGE_CLASSES had a value before the inherit?
> or
>
> ERROR: Could not inherit file classes/.bbclass
> | ETA: --:--:--
> ERROR: Command execution failed: Exited with 1
I'd be prepared to call this last one a bitbake bug...
Cheers,
Richard
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [PATCH 3/6] Add new IMAGE_CLASSES variable for classes for image generation
2011-10-05 23:53 ` Richard Purdie
@ 2011-10-06 0:25 ` McClintock Matthew-B29882
2011-10-11 16:44 ` McClintock Matthew-B29882
0 siblings, 1 reply; 25+ messages in thread
From: McClintock Matthew-B29882 @ 2011-10-06 0:25 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer
On Wed, Oct 5, 2011 at 6:53 PM, Richard Purdie
<richard.purdie@linuxfoundation.org> wrote:
>> So this does not work properly, I thought I tested it but... need to
>> do something like this:
>>
>> IMAGE_CLASSES = "image_types"
>> inherit ${IMAGE_CLASSES}
>>
>> then let the user added via
>>
>> IMAGE_CLASESS += "image_types_uboot"
>>
>> then there will always be something that is expanded to avoid this issue:
>>
>> ERROR: Could not inherit file classes/$IMAGE_CLASSES.bbclass
>> | ETA: --:--:--
>> ERROR: Command execution failed: Exited with 1
>
> Its not expanding the variables? :/ This happened even when
> IMAGE_CLASSES had a value before the inherit?
If I added a IMAGE_CLASSES to something other than "" it worked fine.
If I left it as "" it tried to expand the blank var and look in the
BBPATHS or wherever it looks for classes for an unnamed ".bbclass"
file.
-M
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [PATCH 2/6] Fix perl for powerpc64 build
2011-10-05 19:04 ` McClintock Matthew-B29882
@ 2011-10-07 15:52 ` Richard Purdie
2011-10-07 16:07 ` McClintock Matthew-B29882
0 siblings, 1 reply; 25+ messages in thread
From: Richard Purdie @ 2011-10-07 15:52 UTC (permalink / raw)
To: McClintock Matthew-B29882,
Patches and discussions about the oe-core layer
On Wed, 2011-10-05 at 19:04 +0000, McClintock Matthew-B29882 wrote:
> On Wed, Oct 5, 2011 at 6:46 AM, Richard Purdie
> <richard.purdie@linuxfoundation.org> wrote:
> >> --- a/meta/recipes-devtools/perl/perl_5.12.3.bb
> >> +++ b/meta/recipes-devtools/perl/perl_5.12.3.bb
> >> @@ -7,7 +7,7 @@ LIC_FILES_CHKSUM = "file://Copying;md5=2b4c6ffbcfcbdee469f02565f253d81a \
> >> # We need gnugrep (for -I)
> >> DEPENDS = "virtual/db grep-native"
> >> DEPENDS += "gdbm zlib"
> >> -PR = "r5"
> >> +PR = "r6"
> >>
> >> # 5.10.1 has Module::Build built-in
> >> PROVIDES += "libmodule-build-perl"
> >> @@ -95,6 +95,8 @@ export LDDLFLAGS = "${LDFLAGS} -shared"
> >> # We're almost Debian, aren't we?
> >> CFLAGS += "-DDEBIAN"
> >>
> >> +export ARCH_powerpc64 = "powerpc64"
> >> +
> >
> > I'm finding it really hard to believe this makes any difference. I just
> > tried a powerpc64 build here and perl built fine. I can't find any
> > evidence for any part of the perl build using an ARCH variable
> > (TARGET_ARCH is used but that isn't ARCH).
> >
> > Can you give some further information about how/where this is used?
>
> I'm trying to reproduce this again on 'edison' as well and it does not
> appear to be occurring anymore! The problem before had to do with a
> configuration script error:
>
> config.sh-powerpc64-linux
>
> This script was getting invoked as 'config.sh-powerpc-linux' on the
> 64bit build, then the file did not exist and the configure step failed
> out. Not sure if this was somehow fixed in a different manner.
>
> Please drop this patch for now. I'll bring it up again if I see issues.
Do you still need this bit:
case "${TARGET_ARCH}" in
- x86_64 | powerpc | s390)
+ x86_64 | powerpc | powerpc64 | s390)
sed -i -e "s,\(need_va_copy=\)'undef',\1'define',g" \
config.sh-${TARGET_ARCH}-${TARGET_OS}
;;
?
Cheers,
Richard
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [PATCH 2/6] Fix perl for powerpc64 build
2011-10-07 15:52 ` Richard Purdie
@ 2011-10-07 16:07 ` McClintock Matthew-B29882
0 siblings, 0 replies; 25+ messages in thread
From: McClintock Matthew-B29882 @ 2011-10-07 16:07 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer
On Fri, Oct 7, 2011 at 10:52 AM, Richard Purdie
<richard.purdie@linuxfoundation.org> wrote:
> Do you still need this bit:
>
> case "${TARGET_ARCH}" in
> - x86_64 | powerpc | s390)
> + x86_64 | powerpc | powerpc64 | s390)
> sed -i -e "s,\(need_va_copy=\)'undef',\1'define',g" \
> config.sh-${TARGET_ARCH}-${TARGET_OS}
> ;;
Missing this does not seem to affect things negatively so far, I've
been build testing mostly though - will revisit this later if
something comes up.
-M
^ permalink raw reply [flat|nested] 25+ messages in thread
* Re: [PATCH 3/6] Add new IMAGE_CLASSES variable for classes for image generation
2011-10-06 0:25 ` McClintock Matthew-B29882
@ 2011-10-11 16:44 ` McClintock Matthew-B29882
0 siblings, 0 replies; 25+ messages in thread
From: McClintock Matthew-B29882 @ 2011-10-11 16:44 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer
On Wed, Oct 5, 2011 at 7:24 PM, Matthew McClintock <msm@freescale.com> wrote:
>> Its not expanding the variables? :/ This happened even when
>> IMAGE_CLASSES had a value before the inherit?
>
> If I added a IMAGE_CLASSES to something other than "" it worked fine.
> If I left it as "" it tried to expand the blank var and look in the
> BBPATHS or wherever it looks for classes for an unnamed ".bbclass"
> file.
>> IMAGE_CLASSES = "image_types"
>> inherit ${IMAGE_CLASSES}
Should I update this patch to look like the above? Tracking down the
bitbake issue will be a little more involved for me =)
-M
^ permalink raw reply [flat|nested] 25+ messages in thread
end of thread, other threads:[~2011-10-11 16:53 UTC | newest]
Thread overview: 25+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-10-04 22:03 [PATCH 1/6] Fix mdadm for powerpc64 Matthew McClintock
2011-10-04 22:03 ` [PATCH 2/6] Fix perl for powerpc64 build Matthew McClintock
2011-10-05 11:46 ` Richard Purdie
2011-10-05 19:04 ` McClintock Matthew-B29882
2011-10-07 15:52 ` Richard Purdie
2011-10-07 16:07 ` McClintock Matthew-B29882
2011-10-04 22:03 ` [PATCH 3/6] Add new IMAGE_CLASSES variable for classes for image generation Matthew McClintock
2011-10-05 22:52 ` McClintock Matthew-B29882
2011-10-05 23:53 ` Richard Purdie
2011-10-06 0:25 ` McClintock Matthew-B29882
2011-10-11 16:44 ` McClintock Matthew-B29882
2011-10-04 22:03 ` [PATCH 4/6] scripts/oe-buildenv-internal: Add SOCKS5_{USER, PASSWD} to BB_ENV_EXTRAWHITE Matthew McClintock
2011-10-05 13:25 ` Richard Purdie
2011-10-04 22:03 ` [PATCH 5/6] Fix sysprof for powerpc64 Matthew McClintock
2011-10-05 0:05 ` Khem Raj
2011-10-05 6:44 ` McClintock Matthew-B29882
2011-10-05 11:35 ` Richard Purdie
2011-10-05 16:46 ` Khem Raj
2011-10-05 16:48 ` McClintock Matthew-B29882
2011-10-05 17:06 ` Khem Raj
2011-10-05 17:10 ` Saul Wold
2011-10-05 18:30 ` Phil Blundell
2011-10-05 18:41 ` McClintock Matthew-B29882
2011-10-05 19:13 ` Khem Raj
2011-10-04 22:03 ` [PATCH 6/6] Bring libnl up to date with meta-openebedded Matthew McClintock
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox