On Wed, Apr 23, 2014 at 5:11 PM, Sanjay Singh Rawat <sanjay.ra...@linaro.org > wrote:
> Signed-off-by: Sanjay Singh Rawat <sanjay.ra...@linaro.org> > --- > - This will help in catching issue like > https://lkml.org/lkml/2014/3/25/122 > > - Tested on my x86 dev m/c > Output: > running hotplug stress on cpu : cpu1 > pass: cpu hotplug count = 200 duration = 16.98 seconds > running hotplug stress on cpu : cpu2 > pass: cpu hotplug count = 200 duration = 16.36 seconds > running hotplug stress on cpu : cpu3 > pass: cpu hotplug count = 200 duration = 15.05 seconds > > - Issue: For a hotplug count of 500 the system just hangs. Need to > manually restart. > Is it safe to simply plugout the cpu? What will happend to the tasks > in the cpu runqueue. > --- > cpuhotplug/cpuhotplug_09.sh | 70 > ++++++++++++++++++++++++++++++++++++++++++ > cpuhotplug/cpuhotplug_09.txt | 1 + > 2 files changed, 71 insertions(+) > create mode 100755 cpuhotplug/cpuhotplug_09.sh > create mode 100644 cpuhotplug/cpuhotplug_09.txt > > diff --git a/cpuhotplug/cpuhotplug_09.sh b/cpuhotplug/cpuhotplug_09.sh > new file mode 100755 > index 0000000..be0c91c > --- /dev/null > +++ b/cpuhotplug/cpuhotplug_09.sh > @@ -0,0 +1,70 @@ > +#!/bin/bash > +# > +# PM-QA validation test suite for the power management on Linux > +# > +# Copyright (C) 2014, Linaro Limited. > +# > +# This program is free software; you can redistribute it and/or > +# modify it under the terms of the GNU General Public License > +# as published by the Free Software Foundation; either version 2 > +# of the License, or (at your option) any later version. > +# > +# This program is distributed in the hope that it will be useful, > +# but WITHOUT ANY WARRANTY; without even the implied warranty of > +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > +# GNU General Public License for more details. > +# > +# You should have received a copy of the GNU General Public License > +# along with this program; if not, write to the Free Software > +# Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA > 02110-1301, USA. > +# > +# Contributors: > +# Sanjay Singh Rawat <sanjay.ra...@linaro.org> > +# - initial API and implementation > +# > + > +# URL : > https://wiki.linaro.org/WorkingGroups/PowerManagement/Resources/TestSuite/PmQaSpecification#cpuhotplug_09 > + > +source ../include/functions.sh > +STRESS_COUNT=200 > + > +cpuhotplug_stress() { > + local cpu=$1 > + local dirpath=$CPU_PATH/$1 > + local state= > + local start= > + local end= > + shift 1 > + > + if [ "$cpu" == "cpu0" ]; then > + # skipping main cpu > + return 0 > + fi > + > + echo "running hotplug stress on cpu :" $cpu > + start=$(echo $(awk '{print $1}' /proc/uptime)) > + > + for i in $(seq 1 $STRESS_COUNT); > + do > + set_offline $cpu > + state=$(get_online $cpu) > + if [ $state -ne 0 ]; then > + echo "fail: cpu offline failed at " $i > + return 1 > + fi > + > + set_online $cpu > + state=$(get_online $cpu) > + if [ $state -ne 1 ]; then > + echo "fail: cpu online failed at " $i > + return 1 > + fi > + done > + > + end=$(echo $(awk '{print $1}' /proc/uptime)) > + echo "pass: cpu hotplug in/out count =" $i "duration =" $(echo > "$end-$start" | bc) "seconds" > + return 0 > +} > + > +for_each_cpu cpuhotplug_stress > +test_status_show > diff --git a/cpuhotplug/cpuhotplug_09.txt b/cpuhotplug/cpuhotplug_09.txt > new file mode 100644 > index 0000000..5b0f8ac > --- /dev/null > +++ b/cpuhotplug/cpuhotplug_09.txt > @@ -0,0 +1 @@ > +test to stress cpu hotplug feature > -- > 1.7.10.4 > > This is a good start, but too regular. I would like more randomness to the hotplug process in terms of order of cpus being hotplugged in and out. I tried to come up with something long ago[1] but didn't quite finish it. The script should start with a list of cpus, then maintain a list of online and offline cpus (can be probed repeatedly at runtime too) and then at random select either the online or offline operation and then a random cpu from that list to online or offline. Regards, Amit [1] https://bugs.launchpad.net/linaro-big-little-system/+bug/1097213/comments/2
_______________________________________________ linaro-dev mailing list linaro-dev@lists.linaro.org http://lists.linaro.org/mailman/listinfo/linaro-dev
