* [PATCH 1/3] perf: add perf.inc
2012-07-06 20:29 [PATCH 0/3] perf-scripting/perf-tui features, v2 tom.zanussi
@ 2012-07-06 20:29 ` tom.zanussi
2012-07-06 20:29 ` [PATCH 2/3] perf: add arch-specific perf-scripting feature tom.zanussi
` (2 subsequent siblings)
3 siblings, 0 replies; 10+ messages in thread
From: tom.zanussi @ 2012-07-06 20:29 UTC (permalink / raw)
To: openembedded-core, richard.purdie
From: Tom Zanussi <tom.zanussi@intel.com>
Add a perf.inc to contain utility functions and definitions and to
avoid cluttering up the main recipe.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
---
meta/recipes-kernel/perf/perf.inc | 30 ++++++++++++++++++++++++++++++
meta/recipes-kernel/perf/perf_3.4.bb | 2 ++
2 files changed, 32 insertions(+), 0 deletions(-)
create mode 100644 meta/recipes-kernel/perf/perf.inc
diff --git a/meta/recipes-kernel/perf/perf.inc b/meta/recipes-kernel/perf/perf.inc
new file mode 100644
index 0000000..9b46311
--- /dev/null
+++ b/meta/recipes-kernel/perf/perf.inc
@@ -0,0 +1,30 @@
+def perf_arch_supports_feature(feature, trueval, falseval, d):
+ """
+ Check which perf features a particular architecture supports.
+
+ The perf_features_by_arch table below lists the perf features
+ supported by each arch. If you want a particular arch to support
+ a given set of perf features, add an entry for the arch to this
+ table. Otherwise the arch won't support any features (which may
+ be exactly what you want, just a barebones perf without any extra
+ baggage, which is what you get if you don't add any features
+ here.)
+
+ Available perf features:
+ perf-scripting: enable support for Perl and Python bindings
+ perf-tui: enable support for the perf TUI (via libnewt)
+
+ """
+ perf_features_by_arch = {
+ "core2": "perf-scripting perf-tui",
+ "core2-64": "perf-scripting perf-tui",
+ "x86": "perf-scripting perf-tui",
+ "x86-64": "perf-scripting perf-tui",
+ "i586": "perf-scripting perf-tui",
+ }
+
+ tune = d.getVar("DEFAULTTUNE", True) or ""
+ if perf_features_by_arch.has_key(tune):
+ if feature in perf_features_by_arch[tune]:
+ return trueval
+ return falseval
diff --git a/meta/recipes-kernel/perf/perf_3.4.bb b/meta/recipes-kernel/perf/perf_3.4.bb
index 381332e..d494243 100644
--- a/meta/recipes-kernel/perf/perf_3.4.bb
+++ b/meta/recipes-kernel/perf/perf_3.4.bb
@@ -11,6 +11,8 @@ LIC_FILES_CHKSUM = "file://COPYING;md5=d7810fab7487fb0aad327b76f1be7cd7"
PR = "r1"
+require perf.inc
+
BUILDPERF_libc-uclibc = "no"
DEPENDS = "virtual/kernel \
--
1.7.0.4
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH 2/3] perf: add arch-specific perf-scripting feature
2012-07-06 20:29 [PATCH 0/3] perf-scripting/perf-tui features, v2 tom.zanussi
2012-07-06 20:29 ` [PATCH 1/3] perf: add perf.inc tom.zanussi
@ 2012-07-06 20:29 ` tom.zanussi
2012-07-08 22:00 ` Saul Wold
2012-07-06 20:29 ` [PATCH 3/3] perf: add arch-specific perf-tui feature tom.zanussi
2012-07-08 5:18 ` [PATCH 0/3] perf-scripting/perf-tui features, v2 Tom Zanussi
3 siblings, 1 reply; 10+ messages in thread
From: tom.zanussi @ 2012-07-06 20:29 UTC (permalink / raw)
To: openembedded-core, richard.purdie
From: Tom Zanussi <tom.zanussi@intel.com>
Add a new architecture-specific feature named 'perf-scripting'.
Adding this into the perf_features_by_arch table in perf.inc for a
given arch will enable perf scripting on a target built for that arch,
which will turn on all the language bindings currently aavailable in
perf (Perl and Python), if perf is included in an image.
If 'perf-scripting' isn't named as a feature (the default), all perf
language bindings will be disabled and unavailable.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
---
meta/recipes-kernel/perf/perf_3.4.bb | 12 +++++++++---
1 files changed, 9 insertions(+), 3 deletions(-)
diff --git a/meta/recipes-kernel/perf/perf_3.4.bb b/meta/recipes-kernel/perf/perf_3.4.bb
index d494243..a79e9e6 100644
--- a/meta/recipes-kernel/perf/perf_3.4.bb
+++ b/meta/recipes-kernel/perf/perf_3.4.bb
@@ -9,7 +9,7 @@ as well."
LICENSE = "GPLv2"
LIC_FILES_CHKSUM = "file://COPYING;md5=d7810fab7487fb0aad327b76f1be7cd7"
-PR = "r1"
+PR = "r2"
require perf.inc
@@ -21,7 +21,8 @@ DEPENDS = "virtual/kernel \
${MLPREFIX}binutils \
"
-RDEPENDS_${PN} += "elfutils perl perl-modules python"
+SCRIPTING_RDEPENDS = "${@perf_arch_supports_feature('perf-scripting', 'perl perl-modules python', '',d)}"
+RDEPENDS_${PN} += "elfutils ${SCRIPTING_RDEPENDS}"
PROVIDES = "virtual/perf"
@@ -45,6 +46,8 @@ export PERL_ARCHLIB = "${STAGING_LIBDIR}${PERL_OWN_DIR}/perl/${@get_perl_version
S = "${STAGING_KERNEL_DIR}"
B = "${WORKDIR}/${BPN}-${PV}"
+SCRIPTING_DEFINES = "${@perf_arch_supports_feature('perf-scripting', '', 'NO_LIBPERL=1 NO_LIBPYTHON=1',d)}"
+
EXTRA_OEMAKE = \
'-C ${S}/tools/perf \
O=${B} \
@@ -53,7 +56,7 @@ EXTRA_OEMAKE = \
CC="${CC}" \
AR="${AR}" \
prefix=/usr \
- NO_GTK2=1 NO_NEWT=1 NO_DWARF=1 \
+ NO_GTK2=1 NO_NEWT=1 NO_DWARF=1 ${SCRIPTING_DEFINES} \
'
do_compile() {
@@ -62,6 +65,9 @@ do_compile() {
do_install() {
oe_runmake DESTDIR=${D} install
+ if [ "${@perf_arch_supports_feature('perf-scripting', 1, 0, d)}" = "1" ]; then
+ oe_runmake DESTDIR=${D} install-python_ext
+ fi
}
PACKAGE_ARCH = "${MACHINE_ARCH}"
--
1.7.0.4
^ permalink raw reply related [flat|nested] 10+ messages in thread* Re: [PATCH 2/3] perf: add arch-specific perf-scripting feature
2012-07-06 20:29 ` [PATCH 2/3] perf: add arch-specific perf-scripting feature tom.zanussi
@ 2012-07-08 22:00 ` Saul Wold
2012-07-09 4:02 ` Tom Zanussi
0 siblings, 1 reply; 10+ messages in thread
From: Saul Wold @ 2012-07-08 22:00 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer
On 07/06/2012 01:29 PM, tom.zanussi@intel.com wrote:
> From: Tom Zanussi <tom.zanussi@intel.com>
>
> Add a new architecture-specific feature named 'perf-scripting'.
> Adding this into the perf_features_by_arch table in perf.inc for a
> given arch will enable perf scripting on a target built for that arch,
> which will turn on all the language bindings currently aavailable in
> perf (Perl and Python), if perf is included in an image.
>
> If 'perf-scripting' isn't named as a feature (the default), all perf
> language bindings will be disabled and unavailable.
>
> Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
> ---
> meta/recipes-kernel/perf/perf_3.4.bb | 12 +++++++++---
> 1 files changed, 9 insertions(+), 3 deletions(-)
>
> diff --git a/meta/recipes-kernel/perf/perf_3.4.bb b/meta/recipes-kernel/perf/perf_3.4.bb
> index d494243..a79e9e6 100644
> --- a/meta/recipes-kernel/perf/perf_3.4.bb
> +++ b/meta/recipes-kernel/perf/perf_3.4.bb
> @@ -9,7 +9,7 @@ as well."
> LICENSE = "GPLv2"
> LIC_FILES_CHKSUM = "file://COPYING;md5=d7810fab7487fb0aad327b76f1be7cd7"
>
> -PR = "r1"
> +PR = "r2"
>
> require perf.inc
>
> @@ -21,7 +21,8 @@ DEPENDS = "virtual/kernel \
> ${MLPREFIX}binutils \
> "
>
> -RDEPENDS_${PN} += "elfutils perl perl-modules python"
> +SCRIPTING_RDEPENDS = "${@perf_arch_supports_feature('perf-scripting', 'perl perl-modules python', '',d)}"
> +RDEPENDS_${PN} += "elfutils ${SCRIPTING_RDEPENDS}"
>
> PROVIDES = "virtual/perf"
>
> @@ -45,6 +46,8 @@ export PERL_ARCHLIB = "${STAGING_LIBDIR}${PERL_OWN_DIR}/perl/${@get_perl_version
> S = "${STAGING_KERNEL_DIR}"
> B = "${WORKDIR}/${BPN}-${PV}"
>
> +SCRIPTING_DEFINES = "${@perf_arch_supports_feature('perf-scripting', '', 'NO_LIBPERL=1 NO_LIBPYTHON=1',d)}"
> +
> EXTRA_OEMAKE = \
> '-C ${S}/tools/perf \
> O=${B} \
> @@ -53,7 +56,7 @@ EXTRA_OEMAKE = \
> CC="${CC}" \
> AR="${AR}" \
> prefix=/usr \
> - NO_GTK2=1 NO_NEWT=1 NO_DWARF=1 \
> + NO_GTK2=1 NO_NEWT=1 NO_DWARF=1 ${SCRIPTING_DEFINES} \
> '
>
> do_compile() {
> @@ -62,6 +65,9 @@ do_compile() {
>
> do_install() {
> oe_runmake DESTDIR=${D} install
> + if [ "${@perf_arch_supports_feature('perf-scripting', 1, 0, d)}" = "1" ]; then
> + oe_runmake DESTDIR=${D} install-python_ext
> + fi
> }
>
> PACKAGE_ARCH = "${MACHINE_ARCH}"
>
Tom,
I am not sure this one is working correctly, I this set of 3 in a build
and it failed, so I bicested down to this patch. I am still seeing the
setup.py: Permission denied problem:
> | install scripts/python/bin/* -t '/srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/work/qemux86_64-poky-linux/perf-3.4-r3/image/usr/libexec/perf-core/scripts/python/bin'
> | make: Leaving directory `/srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/sysroots/qemux86-64/usr/src/kernel/tools/perf'
> | NOTE: make -j 16 -C /srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/sysroots/qemux86-64/usr/src/kernel/tools/perf O=/srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/work/qemux86_64-poky-linux/perf-3.4-r3/perf-3.4 CROSS_COMPILE=x86_64-poky-linux- ARCH=x86 CC=x86_64-poky-linux-gcc -m64 --sysroot=/srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/sysroots/qemux86-64 AR=x86_64-poky-linux-ar prefix=/usr NO_GTK2=1 NO_DWARF=1 DESTDIR=/srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/work/qemux86_64-poky-linux/perf-3.4-r3/image install-python_ext
> | make: Entering directory `/srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/sysroots/qemux86-64/usr/src/kernel/tools/perf'
> | util/setup.py --quiet install --root='//srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/work/qemux86_64-poky-linux/perf-3.4-r3/image'
> | make: execvp: util/setup.py: Permission denied
> | make: *** [install-python_ext] Error 127
> | make: Leaving directory `/srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/sysroots/qemux86-64/usr/src/kernel/tools/perf'
> | ERROR: oe_runmake failed
> | ERROR: Function failed: do_install (see /srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/work/qemux86_64-poky-linux/perf-3.4-r3/temp/log.do_install.6037 for further information)
Sau!
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [PATCH 2/3] perf: add arch-specific perf-scripting feature
2012-07-08 22:00 ` Saul Wold
@ 2012-07-09 4:02 ` Tom Zanussi
0 siblings, 0 replies; 10+ messages in thread
From: Tom Zanussi @ 2012-07-09 4:02 UTC (permalink / raw)
To: Saul Wold; +Cc: Patches and discussions about the oe-core layer
On Sun, 2012-07-08 at 15:00 -0700, Saul Wold wrote:
> On 07/06/2012 01:29 PM, tom.zanussi@intel.com wrote:
> > From: Tom Zanussi <tom.zanussi@intel.com>
> >
> > Add a new architecture-specific feature named 'perf-scripting'.
> > Adding this into the perf_features_by_arch table in perf.inc for a
> > given arch will enable perf scripting on a target built for that arch,
> > which will turn on all the language bindings currently aavailable in
> > perf (Perl and Python), if perf is included in an image.
> >
> > If 'perf-scripting' isn't named as a feature (the default), all perf
> > language bindings will be disabled and unavailable.
> >
> > Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
> > ---
> > meta/recipes-kernel/perf/perf_3.4.bb | 12 +++++++++---
> > 1 files changed, 9 insertions(+), 3 deletions(-)
> >
> > diff --git a/meta/recipes-kernel/perf/perf_3.4.bb b/meta/recipes-kernel/perf/perf_3.4.bb
> > index d494243..a79e9e6 100644
> > --- a/meta/recipes-kernel/perf/perf_3.4.bb
> > +++ b/meta/recipes-kernel/perf/perf_3.4.bb
> > @@ -9,7 +9,7 @@ as well."
> > LICENSE = "GPLv2"
> > LIC_FILES_CHKSUM = "file://COPYING;md5=d7810fab7487fb0aad327b76f1be7cd7"
> >
> > -PR = "r1"
> > +PR = "r2"
> >
> > require perf.inc
> >
> > @@ -21,7 +21,8 @@ DEPENDS = "virtual/kernel \
> > ${MLPREFIX}binutils \
> > "
> >
> > -RDEPENDS_${PN} += "elfutils perl perl-modules python"
> > +SCRIPTING_RDEPENDS = "${@perf_arch_supports_feature('perf-scripting', 'perl perl-modules python', '',d)}"
> > +RDEPENDS_${PN} += "elfutils ${SCRIPTING_RDEPENDS}"
> >
> > PROVIDES = "virtual/perf"
> >
> > @@ -45,6 +46,8 @@ export PERL_ARCHLIB = "${STAGING_LIBDIR}${PERL_OWN_DIR}/perl/${@get_perl_version
> > S = "${STAGING_KERNEL_DIR}"
> > B = "${WORKDIR}/${BPN}-${PV}"
> >
> > +SCRIPTING_DEFINES = "${@perf_arch_supports_feature('perf-scripting', '', 'NO_LIBPERL=1 NO_LIBPYTHON=1',d)}"
> > +
> > EXTRA_OEMAKE = \
> > '-C ${S}/tools/perf \
> > O=${B} \
> > @@ -53,7 +56,7 @@ EXTRA_OEMAKE = \
> > CC="${CC}" \
> > AR="${AR}" \
> > prefix=/usr \
> > - NO_GTK2=1 NO_NEWT=1 NO_DWARF=1 \
> > + NO_GTK2=1 NO_NEWT=1 NO_DWARF=1 ${SCRIPTING_DEFINES} \
> > '
> >
> > do_compile() {
> > @@ -62,6 +65,9 @@ do_compile() {
> >
> > do_install() {
> > oe_runmake DESTDIR=${D} install
> > + if [ "${@perf_arch_supports_feature('perf-scripting', 1, 0, d)}" = "1" ]; then
> > + oe_runmake DESTDIR=${D} install-python_ext
> > + fi
> > }
> >
> > PACKAGE_ARCH = "${MACHINE_ARCH}"
> >
>
> Tom,
>
> I am not sure this one is working correctly, I this set of 3 in a build
> and it failed, so I bicested down to this patch. I am still seeing the
> setup.py: Permission denied problem:
Looks like your build isn't getting the kernel changes needed for this
patchset. They were posted at the same time and have actually actually
already been pulled in - you just need to get the kernel in your build
to use them...
Tom
> > | install scripts/python/bin/* -t '/srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/work/qemux86_64-poky-linux/perf-3.4-r3/image/usr/libexec/perf-core/scripts/python/bin'
> > | make: Leaving directory `/srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/sysroots/qemux86-64/usr/src/kernel/tools/perf'
> > | NOTE: make -j 16 -C /srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/sysroots/qemux86-64/usr/src/kernel/tools/perf O=/srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/work/qemux86_64-poky-linux/perf-3.4-r3/perf-3.4 CROSS_COMPILE=x86_64-poky-linux- ARCH=x86 CC=x86_64-poky-linux-gcc -m64 --sysroot=/srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/sysroots/qemux86-64 AR=x86_64-poky-linux-ar prefix=/usr NO_GTK2=1 NO_DWARF=1 DESTDIR=/srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/work/qemux86_64-poky-linux/perf-3.4-r3/image install-python_ext
> > | make: Entering directory `/srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/sysroots/qemux86-64/usr/src/kernel/tools/perf'
> > | util/setup.py --quiet install --root='//srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/work/qemux86_64-poky-linux/perf-3.4-r3/image'
> > | make: execvp: util/setup.py: Permission denied
> > | make: *** [install-python_ext] Error 127
> > | make: Leaving directory `/srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/sysroots/qemux86-64/usr/src/kernel/tools/perf'
> > | ERROR: oe_runmake failed
> > | ERROR: Function failed: do_install (see /srv/ssd/sgw_ab/yocto-autobuilder/yocto-slave/nightly-x86-64/build/build/tmp/work/qemux86_64-poky-linux/perf-3.4-r3/temp/log.do_install.6037 for further information)
>
>
> Sau!
>
^ permalink raw reply [flat|nested] 10+ messages in thread
* [PATCH 3/3] perf: add arch-specific perf-tui feature
2012-07-06 20:29 [PATCH 0/3] perf-scripting/perf-tui features, v2 tom.zanussi
2012-07-06 20:29 ` [PATCH 1/3] perf: add perf.inc tom.zanussi
2012-07-06 20:29 ` [PATCH 2/3] perf: add arch-specific perf-scripting feature tom.zanussi
@ 2012-07-06 20:29 ` tom.zanussi
2012-07-08 5:18 ` [PATCH 0/3] perf-scripting/perf-tui features, v2 Tom Zanussi
3 siblings, 0 replies; 10+ messages in thread
From: tom.zanussi @ 2012-07-06 20:29 UTC (permalink / raw)
To: openembedded-core, richard.purdie
From: Tom Zanussi <tom.zanussi@intel.com>
Add a new architecture-specific feature named 'perf-tui'. Adding this
into the perf_features_by_arch table in perf.inc for a given arch will
enable the perf TUI (Text-base UI) user interface on a target built
for that arch, which adds libnewt and turns on the perf text UI
options in perf, if perf is included in an image.
If 'perf-tui' isn't named as a feature (the default), the perf TUI
will be disabled and unavailable.
Signed-off-by: Tom Zanussi <tom.zanussi@intel.com>
---
meta/recipes-kernel/perf/perf_3.4.bb | 8 ++++++--
1 files changed, 6 insertions(+), 2 deletions(-)
diff --git a/meta/recipes-kernel/perf/perf_3.4.bb b/meta/recipes-kernel/perf/perf_3.4.bb
index a79e9e6..aa9dd6c 100644
--- a/meta/recipes-kernel/perf/perf_3.4.bb
+++ b/meta/recipes-kernel/perf/perf_3.4.bb
@@ -9,16 +9,19 @@ as well."
LICENSE = "GPLv2"
LIC_FILES_CHKSUM = "file://COPYING;md5=d7810fab7487fb0aad327b76f1be7cd7"
-PR = "r2"
+PR = "r3"
require perf.inc
BUILDPERF_libc-uclibc = "no"
+TUI_DEPENDS = "${@perf_arch_supports_feature('perf-tui', 'libnewt', '',d)}"
+
DEPENDS = "virtual/kernel \
virtual/${MLPREFIX}libc \
${MLPREFIX}elfutils \
${MLPREFIX}binutils \
+ ${TUI_DEPENDS} \
"
SCRIPTING_RDEPENDS = "${@perf_arch_supports_feature('perf-scripting', 'perl perl-modules python', '',d)}"
@@ -47,6 +50,7 @@ S = "${STAGING_KERNEL_DIR}"
B = "${WORKDIR}/${BPN}-${PV}"
SCRIPTING_DEFINES = "${@perf_arch_supports_feature('perf-scripting', '', 'NO_LIBPERL=1 NO_LIBPYTHON=1',d)}"
+TUI_DEFINES = "${@perf_arch_supports_feature('perf-tui', '', 'NO_NEWT=1',d)}"
EXTRA_OEMAKE = \
'-C ${S}/tools/perf \
@@ -56,7 +60,7 @@ EXTRA_OEMAKE = \
CC="${CC}" \
AR="${AR}" \
prefix=/usr \
- NO_GTK2=1 NO_NEWT=1 NO_DWARF=1 ${SCRIPTING_DEFINES} \
+ NO_GTK2=1 ${TUI_DEFINES} NO_DWARF=1 ${SCRIPTING_DEFINES} \
'
do_compile() {
--
1.7.0.4
^ permalink raw reply related [flat|nested] 10+ messages in thread* Re: [PATCH 0/3] perf-scripting/perf-tui features, v2
2012-07-06 20:29 [PATCH 0/3] perf-scripting/perf-tui features, v2 tom.zanussi
` (2 preceding siblings ...)
2012-07-06 20:29 ` [PATCH 3/3] perf: add arch-specific perf-tui feature tom.zanussi
@ 2012-07-08 5:18 ` Tom Zanussi
2012-07-09 7:55 ` Saul Wold
3 siblings, 1 reply; 10+ messages in thread
From: Tom Zanussi @ 2012-07-08 5:18 UTC (permalink / raw)
To: openembedded-core
On Fri, 2012-07-06 at 15:29 -0500, tom.zanussi@intel.com wrote:
> From: Tom Zanussi <tom.zanussi@intel.com>
>
> This patchset adds perf-scriping and perf-tui features as
> a per-arch thing rather than per-machine, as suggested by
> Richard Purdie.
>
> I just started testing this, it works so far, but thought
> I'd put it out there now in case there were comments before
> putting too much effort into that, and possibly giving other
> people a chance to try it out.
>
Looks fine after more build-testing: qemux86, qemux86-64, crownbay,
sugarbay, qemuarm and qemuppc both with and without perf-scripting and
perf-tui, and builds without perf as well all build fine here.
Tom
> The following changes since commit 5051e9837fa698e03d0a7a8a918ee7aa98409ce1:
> Robert P. J. Day (1):
> bitbake: usermanual: Fix missing markup
>
> are available in the git repository at:
>
> git://git.yoctoproject.org/poky-contrib.git tzanussi/perf-scripting-v2
> http://git.yoctoproject.org/cgit/cgit.cgi/poky-contrib/log/?h=tzanussi/perf-scripting-v2
>
> Tom Zanussi (3):
> perf: add perf.inc
> perf: add arch-specific perf-scripting feature
> perf: add arch-specific perf-tui feature
>
> meta/recipes-kernel/perf/perf.inc | 30 ++++++++++++++++++++++++++++++
> meta/recipes-kernel/perf/perf_3.4.bb | 18 +++++++++++++++---
> 2 files changed, 45 insertions(+), 3 deletions(-)
> create mode 100644 meta/recipes-kernel/perf/perf.inc
>
^ permalink raw reply [flat|nested] 10+ messages in thread* Re: [PATCH 0/3] perf-scripting/perf-tui features, v2
2012-07-08 5:18 ` [PATCH 0/3] perf-scripting/perf-tui features, v2 Tom Zanussi
@ 2012-07-09 7:55 ` Saul Wold
2012-07-09 13:47 ` Tom Zanussi
0 siblings, 1 reply; 10+ messages in thread
From: Saul Wold @ 2012-07-09 7:55 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer
On 07/07/2012 10:18 PM, Tom Zanussi wrote:
> On Fri, 2012-07-06 at 15:29 -0500, tom.zanussi@intel.com wrote:
>> From: Tom Zanussi <tom.zanussi@intel.com>
>>
>> This patchset adds perf-scriping and perf-tui features as
>> a per-arch thing rather than per-machine, as suggested by
>> Richard Purdie.
>>
>> I just started testing this, it works so far, but thought
>> I'd put it out there now in case there were comments before
>> putting too much effort into that, and possibly giving other
>> people a chance to try it out.
>>
>
> Looks fine after more build-testing: qemux86, qemux86-64, crownbay,
> sugarbay, qemuarm and qemuppc both with and without perf-scripting and
> perf-tui, and builds without perf as well all build fine here.
>
Tom, I am sorry, I am not sure how else to say this, I have tested
master with the only the following change and I am still getting the
the setup.py failure mentioned.
I am not sure what else to suspect at this point, this is Master plus 3
commits, please help me understand where I have gone wrong.
This is the branch I have tested, which is 3 commits pasted Master:
http://git.yoctoproject.org/cgit/cgit.cgi/poky-contrib/log/?h=sgw/fail
Thanks
Sau!
> Tom
>
>> The following changes since commit 5051e9837fa698e03d0a7a8a918ee7aa98409ce1:
>> Robert P. J. Day (1):
>> bitbake: usermanual: Fix missing markup
>>
>> are available in the git repository at:
>>
>> git://git.yoctoproject.org/poky-contrib.git tzanussi/perf-scripting-v2
>> http://git.yoctoproject.org/cgit/cgit.cgi/poky-contrib/log/?h=tzanussi/perf-scripting-v2
>>
>> Tom Zanussi (3):
>> perf: add perf.inc
>> perf: add arch-specific perf-scripting feature
>> perf: add arch-specific perf-tui feature
>>
>> meta/recipes-kernel/perf/perf.inc | 30 ++++++++++++++++++++++++++++++
>> meta/recipes-kernel/perf/perf_3.4.bb | 18 +++++++++++++++---
>> 2 files changed, 45 insertions(+), 3 deletions(-)
>> create mode 100644 meta/recipes-kernel/perf/perf.inc
>>
>
>
>
> _______________________________________________
> Openembedded-core mailing list
> Openembedded-core@lists.openembedded.org
> http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
>
>
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 0/3] perf-scripting/perf-tui features, v2
2012-07-09 7:55 ` Saul Wold
@ 2012-07-09 13:47 ` Tom Zanussi
2012-07-09 13:51 ` Bruce Ashfield
0 siblings, 1 reply; 10+ messages in thread
From: Tom Zanussi @ 2012-07-09 13:47 UTC (permalink / raw)
To: Saul Wold; +Cc: Patches and discussions about the oe-core layer
On Mon, 2012-07-09 at 00:55 -0700, Saul Wold wrote:
> On 07/07/2012 10:18 PM, Tom Zanussi wrote:
> > On Fri, 2012-07-06 at 15:29 -0500, tom.zanussi@intel.com wrote:
> >> From: Tom Zanussi <tom.zanussi@intel.com>
> >>
> >> This patchset adds perf-scriping and perf-tui features as
> >> a per-arch thing rather than per-machine, as suggested by
> >> Richard Purdie.
> >>
> >> I just started testing this, it works so far, but thought
> >> I'd put it out there now in case there were comments before
> >> putting too much effort into that, and possibly giving other
> >> people a chance to try it out.
> >>
> >
> > Looks fine after more build-testing: qemux86, qemux86-64, crownbay,
> > sugarbay, qemuarm and qemuppc both with and without perf-scripting and
> > perf-tui, and builds without perf as well all build fine here.
> >
> Tom, I am sorry, I am not sure how else to say this, I have tested
> master with the only the following change and I am still getting the
> the setup.py failure mentioned.
>
> I am not sure what else to suspect at this point, this is Master plus 3
> commits, please help me understand where I have gone wrong.
>
> This is the branch I have tested, which is 3 commits pasted Master:
>
> http://git.yoctoproject.org/cgit/cgit.cgi/poky-contrib/log/?h=sgw/fail
>
Saul,
It's because your kernel hasn't advanced the SRCREVs to include the new
commits - the above patches won't work without the associated kernel
changes. So if you're build-testing qemux86-64 for instance, you'd want
to change the following line in linux_yocto_3.4.bb from:
SRCREV_machine_qemux86-64 ?= "d345461f0d52d1a82553c2edd845a8bee908923f"
to include the last commit of the associated kernel patch series i.e.the
top of the common-pc-64 branch:
SRCREV_machine_qemux86-64 ?= "49f1599b2ce0288599805b5364b895f5a9febb3e"
That should get the kernel changes associated with the patchset into
your kernel and everything should build nicely...
Tom
> Thanks
> Sau!
>
> > Tom
> >
> >> The following changes since commit 5051e9837fa698e03d0a7a8a918ee7aa98409ce1:
> >> Robert P. J. Day (1):
> >> bitbake: usermanual: Fix missing markup
> >>
> >> are available in the git repository at:
> >>
> >> git://git.yoctoproject.org/poky-contrib.git tzanussi/perf-scripting-v2
> >> http://git.yoctoproject.org/cgit/cgit.cgi/poky-contrib/log/?h=tzanussi/perf-scripting-v2
> >>
> >> Tom Zanussi (3):
> >> perf: add perf.inc
> >> perf: add arch-specific perf-scripting feature
> >> perf: add arch-specific perf-tui feature
> >>
> >> meta/recipes-kernel/perf/perf.inc | 30 ++++++++++++++++++++++++++++++
> >> meta/recipes-kernel/perf/perf_3.4.bb | 18 +++++++++++++++---
> >> 2 files changed, 45 insertions(+), 3 deletions(-)
> >> create mode 100644 meta/recipes-kernel/perf/perf.inc
> >>
> >
> >
> >
> > _______________________________________________
> > Openembedded-core mailing list
> > Openembedded-core@lists.openembedded.org
> > http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core
> >
> >
>
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 0/3] perf-scripting/perf-tui features, v2
2012-07-09 13:47 ` Tom Zanussi
@ 2012-07-09 13:51 ` Bruce Ashfield
0 siblings, 0 replies; 10+ messages in thread
From: Bruce Ashfield @ 2012-07-09 13:51 UTC (permalink / raw)
To: Patches and discussions about the oe-core layer
On Mon, Jul 9, 2012 at 9:47 AM, Tom Zanussi <tom.zanussi@intel.com> wrote:
> On Mon, 2012-07-09 at 00:55 -0700, Saul Wold wrote:
>> On 07/07/2012 10:18 PM, Tom Zanussi wrote:
>> > On Fri, 2012-07-06 at 15:29 -0500, tom.zanussi@intel.com wrote:
>> >> From: Tom Zanussi <tom.zanussi@intel.com>
>> >>
>> >> This patchset adds perf-scriping and perf-tui features as
>> >> a per-arch thing rather than per-machine, as suggested by
>> >> Richard Purdie.
>> >>
>> >> I just started testing this, it works so far, but thought
>> >> I'd put it out there now in case there were comments before
>> >> putting too much effort into that, and possibly giving other
>> >> people a chance to try it out.
>> >>
>> >
>> > Looks fine after more build-testing: qemux86, qemux86-64, crownbay,
>> > sugarbay, qemuarm and qemuppc both with and without perf-scripting and
>> > perf-tui, and builds without perf as well all build fine here.
>> >
>> Tom, I am sorry, I am not sure how else to say this, I have tested
>> master with the only the following change and I am still getting the
>> the setup.py failure mentioned.
>>
>> I am not sure what else to suspect at this point, this is Master plus 3
>> commits, please help me understand where I have gone wrong.
>>
>> This is the branch I have tested, which is 3 commits pasted Master:
>>
>> http://git.yoctoproject.org/cgit/cgit.cgi/poky-contrib/log/?h=sgw/fail
>>
>
> Saul,
>
> It's because your kernel hasn't advanced the SRCREVs to include the new
> commits - the above patches won't work without the associated kernel
> changes. So if you're build-testing qemux86-64 for instance, you'd want
> to change the following line in linux_yocto_3.4.bb from:
>
> SRCREV_machine_qemux86-64 ?= "d345461f0d52d1a82553c2edd845a8bee908923f"
>
> to include the last commit of the associated kernel patch series i.e.the
> top of the common-pc-64 branch:
>
> SRCREV_machine_qemux86-64 ?= "49f1599b2ce0288599805b5364b895f5a9febb3e"
.. and I'm about to send a new set of SRCREVs that have this and
some more. I was just waiting to see how these conversations played out.
Cheers,
Bruce
>
> That should get the kernel changes associated with the patchset into
> your kernel and everything should build nicely...
>
> Tom
>
>> Thanks
>> Sau!
>>
>> > Tom
>> >
>> >> The following changes since commit 5051e9837fa698e03d0a7a8a918ee7aa98409ce1:
>> >> Robert P. J. Day (1):
>> >> bitbake: usermanual: Fix missing markup
>> >>
>> >> are available in the git repository at:
>> >>
>> >> git://git.yoctoproject.org/poky-contrib.git tzanussi/perf-scripting-v2
>> >> http://git.yoctoproject.org/cgit/cgit.cgi/poky-contrib/log/?h=tzanussi/perf-scripting-v2
>> >>
>> >> Tom Zanussi (3):
>> >> perf: add perf.inc
>> >> perf: add arch-specific perf-scripting feature
>> >> perf: add arch-specific perf-tui feature
>> >>
>> >> meta/recipes-kernel/perf/perf.inc | 30 ++++++++++++++++++++++++++++++
>> >> meta/recipes-kernel/perf/perf_3.4.bb | 18 +++++++++++++++---
>> >> 2 files changed, 45 insertions(+), 3 deletions(-)
>> >> create mode 100644 meta/recipes-kernel/perf/perf.inc
>> >>
>> >
>> >
>> >
>> > _______________________________________________
>> > 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
--
"Thou shalt not follow the NULL pointer, for chaos and madness await
thee at its end"
^ permalink raw reply [flat|nested] 10+ messages in thread