kernel May 2016

kernel@lists.fedoraproject.org

15 participants
11 discussions

[PATCH] Script to generate perf-man tarball.
by Miguel A. Flores Silverio 02 Jun '16

02 Jun '16

This is a small script that semi automates the creation of the perf-man pages. This version of the script assumes LINUX_GIT is set in you local .bashrc file. Signed-off-by: Miguel Flores Silverio <floresmigu3l(a)gmail.com> --- scripts/generate-perf-man.sh | 19 +++++++++++++++++++ 1 file changed, 19 insertions(+) create mode 100644 scripts/generate-perf-man.sh diff --git a/scripts/generate-perf-man.sh b/scripts/generate-perf-man.sh new file mode 100644 index 0000000..ef11b53 --- /dev/null +++ b/scripts/generate-perf-man.sh @@ -0,0 +1,19 @@ +#! /bin/sh +source ~/.bashrc # LINUX_GIT is set + +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 ) + +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

[PATCH] config: x86: stop pretending to be an Lguest guest
by Paul Bolle 31 May '16

31 May '16

Ever since v2.6.36 there's a conflict between OLPC support and Lguest guest support: a kernel image can't both support booting OLPC x86 hardware and booting as an Lguest guest. Booting a kernel image that tries to do both as an Lguest guest will fail with lguest: Reinjecting trap 13 for fault at 0x1000062: Invalid argument This means that people wanting to boot an Lguest guest on 32 bits x86 needed to build their own kernel image without OLPC support since, it seems, Fedora 15. So let's stop pretending to support being an Lguest guest. People wanting to have that will now not only have to disable CONFIG_OLPC but also enable CONFIG_LGUEST_GUEST. (A second order effect is that both CONFIG_VIRTIO and CONFIG_VIRTIO_CONSOLE will change from 'y' to 'm' in our builds. Assuming a sane build system no one should notice. We'll see.) Signed-off-by: Paul Bolle <pebolle(a)tiscali.nl> --- config-x86-generic | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) Tested on F22. Applies cleanly to master (F25?) but I have not even tried building that. The conflict between these two options is known for years now. See https://lists.ozlabs.org/pipermail/lguest/2013-April/001975.html . (There's also a conflict with early microcode loading, see https://lists.ozlabs.org/pipermail/lguest/2013-May/002001.html . That one can actually be worked around by booting the guest with the dis_ucode_ldr kernel parameter.) diff --git a/config-x86-generic b/config-x86-generic index 13de9462e556..c6c02779399f 100644 --- a/config-x86-generic +++ b/config-x86-generic @@ -283,7 +283,7 @@ CONFIG_KVM_INTEL=m CONFIG_KVM_AMD=m CONFIG_KVM_DEVICE_ASSIGNMENT=y CONFIG_LGUEST=m -CONFIG_LGUEST_GUEST=y +# CONFIG_LGUEST_GUEST is not set CONFIG_HYPERVISOR_GUEST=y CONFIG_PARAVIRT=y -- 2.4.11

3 5

[PATCHv2] Script to generate perf-man tarball.
by Miguel A. Flores Silverio 30 May '16

30 May '16

1 0

possible debug config extension
by Jiri Pirko 29 May '16

29 May '16

Hi. There could be really handy to have more debug options enabled in debug kernel. Namely: CONFIG_PREEMPT=y CONFIG_DEBUG_PREEMPT=y CONFIG_SLUB_DEBUG=y CONFIG_SLUB_DEBUG_ON=y CONFIG_DEBUG_PAGEALLOC=y CONFIG_KASAN=y CONFIG_PANIC_ON_OOPS=y CONFIG_UBSAN=y CONFIG_UBSAN_SANITIZE_ALL=y KASan is especially useful. However the overhead of KASan for example is quite significant. So the question is, would it make sense to add these to the debug config? Another option would be to introduce a "deeper-debug" config and do an extra rpm for that. That would let user to choose between ordinary debug kernel and deeper debug kernel. What do you think? Thanks! Jiri

2 1

CONFIG_INTEL_POWERCLAMP not enabled in fedora
by Stefan Assmann 19 May '16

19 May '16

I recently wanted to experiment with Intel idle injection but noticed that CONFIG_INTEL_POWERCLAMP is not enabled in f23/f24. # cat /boot/config-4.5.4-300.fc24.x86_64 | grep POWERCLAMP # CONFIG_INTEL_POWERCLAMP is not set Is this is an oversight? Any chance to still get this enabled in f24? Thanks! Stefan

3 3

[PATCH] config: arm64: Enable generic PCI host bridge
by Jeremy Linton 17 May '16

17 May '16

The ARM Juno development platform's PCIe is supported by the generic PCI host driver. Enable it so that PCIe attached peripherals work. Signed-off-by: Jeremy Linton <jeremy.linton(a)arm.com> --- config-arm64 | 3 +++ 1 file changed, 3 insertions(+) diff --git a/config-arm64 b/config-arm64 index 77dde02..2db4e9a 100644 --- a/config-arm64 +++ b/config-arm64 @@ -110,6 +110,9 @@ CONFIG_CRYPTO_DEV_CCP=y CONFIG_CRYPTO_DEV_CCP_DD=m CONFIG_CRYPTO_DEV_CCP_CRYPTO=m +# ARM Juno +CONFIG_PCI_HOST_GENERIC=y + # APM Xgene CONFIG_POWER_RESET_XGENE=y CONFIG_COMMON_CLK_XGENE=y -- 2.5.5

4 7

Add support for mipsel and mips64el
by Michal Toman 11 May '16

11 May '16

Hello, as a part of the effort to revive Fedora MIPS, kernel needs to build properly on MIPS. I have been testing various configurations in the past 6 months and finally arrived to a state where I think it is good/stable enough to be proposed. Attached is a patch that I have used to build working kernels since 4.6.0-0.rc1. It enables build on two architectures - mipsel and mips64el. Both of them build images for the malta board (evaluation board by MIPS Technologies with good support in QEMU). mips64el builds an extra flavor - octeon - for octeon based boards. These are used in high performance network appliances and we also plan to use them for our Fedora builders. Given the MIPS kernel needs to be built for a specific board I was not able to find a better way than adding an extra flavor. There are 4 patches required - 3 of them are waiting in the queue to be upstreamed, I have added a fourth one that softens Werror so the build does not die on a harmless warning about 32-bit VDSO on 64-bit octeons. The only thing I was not able to understand is the magic around kernel-debuginfo-common. My debugfiles.list is always empty and I need to remove it in order to make the build succeed (with an empty kernel-debuginfo-common). It would be great if you could point me in the right direction for this one. Please let me know if I can do anything else to smoothen the process. Thanks, Michal

1 1

Re: Performance for kernel-4.6.0-0.rc5.git0.1.wastedcores
by Jirka Hladky 04 May '16

04 May '16

Hi Jeremy, please have a look at Linpack and Stream results http://perf-desktop.brq.redhat.com/LinpackAndStream/Red-Hat-Enterprise-Linu… and http://perf-desktop.brq.redhat.com/LinpackAndStream/Red-Hat-Enterprise-Linu… We have collected /proc/sched_debug every 10 seconds - there are ""*scheddebug" files for every run. We are also collecting mpstat statistics every 2 seconds. For example during a test when 64 stream jobs were running in parallel there are corresponding /proc/sched_debug logs: 4.6.0-0.rc5.git0.1.el7.x86_64/throughput-performance/Results-2016-Apr-28_14h01m32s/RawData/stream.affinityRun.064streams.loop01.scheddebug <http://perf-desktop.brq.redhat.com/LinpackAndStream/Red-Hat-Enterprise-Linu…> 4.6.0-0.rc5.git0.1.wastedcores.el7.x86_64/throughput-performance/Results-2016-Apr-29_15h37m17s/RawData/stream.affinityRun.064streams.loop01.scheddebug <http://perf-desktop.brq.redhat.com/LinpackAndStream/Red-Hat-Enterprise-Linu…> However, we are not sure how to process these log files. Is there any tool to do so or could you share some hints what to look for? Thanks a lot! Jirka On Tue, May 3, 2016 at 4:18 PM, Jeremy Eder <jeder(a)redhat.com> wrote: > Yikes. 20% off...why? Do you have any system stats? Particularly > /proc/sched_debug collected at some interval. > > On Mon, May 2, 2016 at 12:39 PM, Jirka Hladky <jhladky(a)redhat.com> wrote: > >> Hi all, >> >> we have compared scheduler performance >> for kernel-4.6.0-0.rc5.git0.1.wastedcores >> against kernel-4.6.0-0.rc5.git0.1.el7. >> >> *Summary: *against all expectations* kernel-4.6.0-0.rc5.git0.1.wastedcores performs on >> average 20% WORSE than the regular kernel. * >> >> Results for different servers: >> >> SPECjbb2005: >> >> - karkulka-02.lab.eng.brq.redhat.com-htON >> <http://perf-desktop.brq.redhat.com/SPECjbb/.vs./karkulka-02.lab.eng.brq.red…> >> - kiff-01.cluster-qe.lab.eng.brq.redhat.com-htON >> <http://perf-desktop.brq.redhat.com/SPECjbb/.vs./kiff-01.cluster-qe.lab.eng.…> >> >> SPECjvm2008: >> >> - karkulka-02.lab.eng.brq.redhat.com-htON >> <http://perf-desktop.brq.redhat.com/SPECjvm/.vs./karkulka-02.lab.eng.brq.red…> >> - kiff-01.cluster-qe.lab.eng.brq.redhat.com-htON >> <http://perf-desktop.brq.redhat.com/SPECjvm/.vs./kiff-01.cluster-qe.lab.eng.…> >> >> Linpack and Stream: >> >> - karkulka-02.lab.eng.brq.redhat.com-htON >> <http://perf-desktop.brq.redhat.com/LinpackAndStream/.vs./karkulka-02.lab.en…> >> - kiff-01.cluster-qe.lab.eng.brq.redhat.com-htON >> <http://perf-desktop.brq.redhat.com/LinpackAndStream/.vs./kiff-01.cluster-qe…> >> >> Any comments, questions, suggestions? >> >> Jirka and Kamil >> >> On Thu, Apr 28, 2016 at 12:25 AM, Jirka Hladky <jhladky(a)redhat.com> >> wrote: >> >>> Hi Jeremy, >>> >>> I have just read >>> >>> http://www.i3s.unice.fr/~jplozi/wastedcores/files/extended_talk.pdf >>> http://www.ece.ubc.ca/~sasha/papers/eurosys16-final29.pdf >>> >>> We will definitely test it and share the results here. >>> >>> Jirka >>> >>> >>> > > > -- > > -- Jeremy Eder >

1 2

Performance for kernel-4.6.0-0.rc5.git0.1.wastedcores
by Jirka Hladky 02 May '16

02 May '16

Hi all, we have compared scheduler performance for kernel-4.6.0-0.rc5.git0.1.wastedcores against kernel-4.6.0-0.rc5.git0.1.el7. *Summary: *against all expectations* kernel-4.6.0-0.rc5.git0.1.wastedcores performs on average 20% WORSE than the regular kernel. * Results for different servers: SPECjbb2005: - karkulka-02.lab.eng.brq.redhat.com-htON <http://perf-desktop.brq.redhat.com/SPECjbb/.vs./karkulka-02.lab.eng.brq.red…> - kiff-01.cluster-qe.lab.eng.brq.redhat.com-htON <http://perf-desktop.brq.redhat.com/SPECjbb/.vs./kiff-01.cluster-qe.lab.eng.…> SPECjvm2008: - karkulka-02.lab.eng.brq.redhat.com-htON <http://perf-desktop.brq.redhat.com/SPECjvm/.vs./karkulka-02.lab.eng.brq.red…> - kiff-01.cluster-qe.lab.eng.brq.redhat.com-htON <http://perf-desktop.brq.redhat.com/SPECjvm/.vs./kiff-01.cluster-qe.lab.eng.…> Linpack and Stream: - karkulka-02.lab.eng.brq.redhat.com-htON <http://perf-desktop.brq.redhat.com/LinpackAndStream/.vs./karkulka-02.lab.en…> - kiff-01.cluster-qe.lab.eng.brq.redhat.com-htON <http://perf-desktop.brq.redhat.com/LinpackAndStream/.vs./kiff-01.cluster-qe…> Any comments, questions, suggestions? Jirka and Kamil On Thu, Apr 28, 2016 at 12:25 AM, Jirka Hladky <jhladky(a)redhat.com> wrote: > Hi Jeremy, > > I have just read > > http://www.i3s.unice.fr/~jplozi/wastedcores/files/extended_talk.pdf > http://www.ece.ubc.ca/~sasha/papers/eurosys16-final29.pdf > > We will definitely test it and share the results here. > > Jirka > > >

1 0

[PATCH 1/2] Enable SOC_TI81XX used on hp-t410
by Nicolas Chauvet 02 May '16

02 May '16

This will enable support for consumer devices such as: http://www8.hp.com/us/en/thin-clients/t410.html Multi-platform support was introduced by Tony Lindgren Please see : http://muru.com/linux/t410/ Working and tested with v4.5 kernel is: - serial - ethernet This can be enabled on kernel v4.5+ --- config-armv7 | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/config-armv7 b/config-armv7 index bcf054c..1723c5a 100644 --- a/config-armv7 +++ b/config-armv7 @@ -21,7 +21,7 @@ CONFIG_ARCH_OMAP2PLUS_TYPICAL=y CONFIG_SOC_OMAP5=y # CONFIG_SOC_DRA7XX is not set CONFIG_SOC_OMAP3430=y -# CONFIG_SOC_TI81XX is not set +CONFIG_SOC_TI81XX=y # CONFIG_MACH_NOKIA_RX51 is not set # CONFIG_MACH_OMAP_LDP is not set # CONFIG_MACH_OMAP3517EVM is not set -- 1.7.2.1

1 1

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

kernel May 2016