kernel July 2016

kernel@lists.fedoraproject.org

10 participants
16 discussions

[PATCH] Removes unused-kernel-patches in Makefile
by Miguel A. Flores Silverio 19 Jul '16

19 Jul '16

After discussing with my mentor [ Laura Abbott ]. Looks like the following is not longer needed. Signed-off-by: Miguel Flores Silverio <floresmigu3l(a)gmail.com> --- Makefile | 3 --- 1 file changed, 3 deletions(-) diff --git a/Makefile b/Makefile index 5354035..f20475f 100644 --- a/Makefile +++ b/Makefile @@ -111,9 +111,6 @@ release: config-release include Makefile.release -unused-kernel-patches: - @for f in *.patch; do if [ -e $$f ]; then (egrep -q "^Patch[[:digit:]]+:[[:space:]]+$$f" $(SPECFILE) || echo "Unused: $$f") && egrep -q "^ApplyPatch[[:space:]]+$$f|^ApplyOptionalPatch[[:space:]]+$$f" $(SPECFILE) || echo "Unapplied: $$f"; fi; done - ifeq ($(MAKECMDGOALS),me a sandwich) .PHONY: me a sandwich me a: -- 2.7.4

2 1

[PATCHv4] Script to generate perf-man tarball.
by Miguel A. Flores Silverio 19 Jul '16

19 Jul '16

This script helps generate the perf-man tarball. By default it uses the kernel version of the upstream tree set in LINUX_GIT. In the given case the the upstream tree and the current tree do not have the same version use the --version=x.y Signed-off-by: Miguel Flores Silverio <floresmigu3l(a)gmail.com> --- * v1 - use LINUX_GIT * v2 - Fix: quotes around $LINUX - moves to scripts directory * v3 - Add prompt to specify kernel version and patchlevel. * v4 - Use flag --version=x.y instead of prompt - Remove debuggin comments scripts/generate-perf-man.sh | 44 ++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 44 insertions(+) create mode 100755 scripts/generate-perf-man.sh diff --git a/scripts/generate-perf-man.sh b/scripts/generate-perf-man.sh new file mode 100755 index 0000000..6f51820 --- /dev/null +++ b/scripts/generate-perf-man.sh @@ -0,0 +1,44 @@ +#! /bin/sh +# Small script to generate the perf-man tarball. The script relies on having +# LINUX_GIT set in your local .bashrc. By default the script will use the +# the kernel version of the upstream tree set in LINUX_GIT. Use --version=x.y +# to set a specific version. + +# [Default] eg. ./scritps/generate-perf-man +# eg. ./scripts/generate-perf-man --version=4.8 + +if [ -f ~/.bashrc ]; then + source ~/.bashrc +fi + +if [ ! -d "$LINUX_GIT" ]; then + echo "Error: \$LINUX_GIT is not set to the upstream git tree." + exit 1 +fi + +BASEDIR=$(dirname "$(cd $(dirname $BASH_SOURCE[0]) && pwd)") +pushd "$LINUX_GIT" +KERNEL_VERSION=$( awk '/^VERSION =/ {print $3}' Makefile ) +KERNEL_PATCHLEVEL=$( awk '/^PATCHLEVEL =/ {print $3}' Makefile ) + +if [ ! -z "$@" ]; then + for opt in "$@"; do + case $opt in + --version=*.*) + version="${opt#*=}" + KERNEL_VERSION=$( awk -F. '{print $1}' <<< $version ) + KERNEL_PATCHLEVEL=$( awk -F. '{print $2}' <<< $version ) + ;; + -h | --help) + usage + ;; + *) + ;; + esac + done +fi +cd tools/perf/Documentation/ +make +tar -czvf $BASEDIR/perf-man-${KERNEL_VERSION}.${KERNEL_PATCHLEVEL}.tar.gz *.1 +make clean +popd -- 2.7.4

2 1

Fwd: [PATCH] drm/i915/skl: Fix redundant cursor update, fix cursor underruns
by Hans de Goede 16 Jul '16

16 Jul '16

And another skylake laptop external monitor fix which would be good to add to the Fedora kernels. (note it has a Cc stable) -------- Forwarded Message -------- Subject: [PATCH] drm/i915/skl: Fix redundant cursor update, fix cursor underruns Date: Fri, 15 Jul 2016 18:13:56 -0400 From: Lyude <cpaul(a)redhat.com> To: intel-gfx(a)lists.freedesktop.org CC: Lyude <cpaul(a)redhat.com>, Radhakrishna Sripada <radhakrishna.sripada(a)intel.com>, Hans de Goede <hdegoede(a)redhat.com>, stable(a)vger.kernel.org, Daniel Vetter <daniel.vetter(a)intel.com>, Jani Nikula <jani.nikula(a)linux.intel.com>, David Airlie <airlied(a)linux.ie>, open list:INTEL DRM DRIVERS (excluding Poulsbo, Moorestow...), linux-kernel(a)vger.kernel.org (open list) <dri-devel(a)lists.freedesktop.org> At long last, the time has finally come for Skylake users to plug their external displays back in. During intel_atomic_commit() on Skylake, we've actually been arming the registers to update the cursor information twice instead of just once. Once in i9xx_update_cursor(), and once in skl_wm_flush_pipe(). This isn't actually necessary, and removing the later update in skl_wm_flush_pipe() has completely stopped the underruns on this T460p from occurring when moving the mouse cursor from one monitor to another. Signed-off-by: Lyude <cpaul(a)redhat.com> Cc: Radhakrishna Sripada <radhakrishna.sripada(a)intel.com> Cc: Hans de Goede <hdegoede(a)redhat.com> Cc: stable(a)vger.kernel.org --- drivers/gpu/drm/i915/intel_pm.c | 1 - 1 file changed, 1 deletion(-) diff --git a/drivers/gpu/drm/i915/intel_pm.c b/drivers/gpu/drm/i915/intel_pm.c index 7ac71ec..4771a03 100644 --- a/drivers/gpu/drm/i915/intel_pm.c +++ b/drivers/gpu/drm/i915/intel_pm.c @@ -3860,7 +3860,6 @@ skl_wm_flush_pipe(struct drm_i915_private *dev_priv, enum pipe pipe, int pass) I915_WRITE(PLANE_SURF(pipe, plane), I915_READ(PLANE_SURF(pipe, plane))); } - I915_WRITE(CURBASE(pipe), I915_READ(CURBASE(pipe))); } static bool -- 2.7.4

1 0

unsubscribe
by Robert M. Albrecht 15 Jul '16

15 Jul '16

1 0

[PATCHv2] Script to add a new patch newpatch.sh
by Miguel A. Flores Silverio 15 Jul '16

15 Jul '16

Script helps add a new patch to the soruce tree. It enables the builid macro and uses the name of the patch as label instead of using rpmdev-bumpspec. Signed-off-by: Miguel Flores Silverio <floresmigu3l(a)gmail.com> --- *v1 - Fixed: Fail adding a patch with no description - Fixed: Fail to add a patch outside the tree - Fixed: Fail when adding a patch already in the tree directory - Fixed: Fail to add to git when running script outside tree scripts/newpatch.sh | 41 +++++++++++++++++++++++++++++++++++++++++ 1 file changed, 41 insertions(+) create mode 100755 scripts/newpatch.sh diff --git a/scripts/newpatch.sh b/scripts/newpatch.sh new file mode 100755 index 0000000..2153fc8 --- /dev/null +++ b/scripts/newpatch.sh @@ -0,0 +1,41 @@ +#! /bin/sh + +# Facilitates the addition of a new patch to the source tree. +# -- Moves patch to tree +# -- Adds patch to kernel.spec list of patches +# -- Adds patch to git +# -- change buildid macro to the name of the patch being added + +# Base directory is relative to where the script is. +BASEDIR="$(dirname "$(cd $(dirname $BASH_SOURCE[0]) && pwd)")" +pushd $BASEDIR > /dev/null +# Check for at least patch +if [ "$#" -lt 1 ]; then + echo "usage: $0 [ /path/to/patch/ ] [ description ]" + exit 1 +fi +PATCHDIR=$1 +DESC=$2 +PATCH="$(basename "$PATCHDIR")" +# Kernel.spec file in the current tree +SPECFILE="$BASEDIR/kernel.spec" +# If adding patch from outside the source tree move it to the source tree +if [ -z "$(ls | grep $PATCH)" ]; then + mv $PATCHDIR $BASEDIR/ +fi + +if [ ! -z "$(grep $PATCH $SPECFILE)" ] +then + echo "$PATCH already in kernel.spec" + exit 1 +fi +# ID number of the last patch in kernel.spec +LPATCH_ID=$(grep ^Patch $SPECFILE | tail -n1 | awk '{ print $1 }' | sed s/Patch// | sed s/://) +# ID of the next patch to be added to kernel.spec +NPATCH_ID=$(($LPATCH_ID + 1 )) +# Add patch with new id at the end of the list of patches +sed -i "/^Patch$LPATCH_ID:\ /a#\ $DESC\nPatch$NPATCH_ID:\ $PATCH" $SPECFILE +# Add it to git +git add $PATCH +sed -i "s/^.*define buildid .*$/%define buildid .$PATCH/" $SPECFILE +popd > /dev/null -- 2.7.4 Some tests I performed: # running script within tree and patch also withing tree, no description - $ ./scripts/newpatch.sh some_patch.patch # running script outside tree and patch outside tree, no description - $ bash ~/kernel/scripts/newpatch.sh ~/Documents/fix_all_the_things.patch # running within tree and patch outside tree, with decription - $ ./scripts/newpatch.sh ~/Documents/fix_all_the_things.patch "It is magic!"

1 0

rpm hang when building latest kernel ...
by Prarit Bhargava 12 Jul '16

12 Jul '16

Hi labbott, I'm trying to build the latest kernel rpm [1] on an updated F24 and it is getting stuck here: + /root/rpmbuild/SOURCES/mod-sign.sh certs/signing_key.pem.sign certs/signing_key.x509.sign /root/rpmbuild/BUILDROOT/kernel-4.7.0-0.rc6.git2.2.fc24.x86_64/lib/modules/4.7.0-0.rc6.git2.2.fc24prarit.x86_64/ find: '/root/rpmbuild/BUILDROOT/kernel-4.7.0-0.rc6.git2.2.fc24.x86_64/lib/modules/4.7.0-0.rc6.git2.2.fc24prarit.x86_64/': No such file or directory find: '/root/rpmbuild/BUILDROOT/kernel-4.7.0-0.rc6.git2.2.fc24.x86_64/lib/modules/4.7.0-0.rc6.git2.2.fc24prarit.x86_64/': No such file or directory ps aux shows: root 8710 0.0 0.0 9660 2500 pts/0 S+ 20:02 0:00 /bin/bash /root/rpmbuild/SOURCES/mod-sign.sh certs/signing_key.pem.sign certs/signing_key.x509.sign /root/rpmbuild/BUILDROOT/kernel-4.7.0-0.rc6.git2.2.fc24.x86_64/lib/modules/4.7.0-0.rc6.git2.2.fc24prarit.x86_64/ root 8716 0.0 0.0 4376 788 pts/0 S+ 20:02 0:00 tail -c 28 Any ideas? Have you seen this? Just want to know before I start debugging ;) P. [1] http://koji.fedoraproject.org/koji/buildinfo?buildID=778835

3 3

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

kernel July 2016