controllers/pids: Add new testcases
PATCH v2 1/1 controllers/pids: Add new testcases
Created new testcase to test the pids controller which is used to stop any new tasks from being fork()'d after a certain limit is reached.
--- runtest/controllers | 2 + testcases/kernel/controllers/pids/.gitignore | 1 + testcases/kernel/controllers/pids/Makefile | 28 +++++++ testcases/kernel/controllers/pids/pids.sh | 109 ++++++++++++++++++++++++++ testcases/kernel/controllers/pids/pids_task.c | 60 ++++++++++++++ 5 files changed, 200 insertions(+) create mode 100644 testcases/kernel/controllers/pids/.gitignore create mode 100644 testcases/kernel/controllers/pids/Makefile create mode 100644 testcases/kernel/controllers/pids/pids.sh create mode 100644 testcases/kernel/controllers/pids/pids_task.c
runtest/controllers +2 -0
@@ -277,3 +277,5 @@ cpuset_memory_spread cpuset_memory_spread_testset.sh cpuset_regression_test cpuset_regression_test.sh cgroup_xattr cgroup_xattr + +pids pids.sh
testcases/kernel/controllers/pids/.gitignore new file +1 -0
@@ -0,0 +1 @@ +/pids_task
testcases/kernel/controllers/pids/Makefile new file +28 -0
@@ -0,0 +1,28 @@ +# +# Copyright (C) 2015, SUSE +# +# 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. +# +# Author: Cedric Hnyda <[email protected]> +# + +top_srcdir ?= ../../../.. + +include $(top_srcdir)/include/mk/testcases.mk +include $(abs_srcdir)/../Makefile.inc + +INSTALL_TARGETS := *.sh + +include $(top_srcdir)/include/mk/generic_leaf_target.mk
testcases/kernel/controllers/pids/pids.sh new file +109 -0
109 lines not on this page · raw message
testcases/kernel/controllers/pids/pids_task.c new file +60 -0
60 lines not on this page · raw message
Signature · 1 line
2.1.4
PATCH v2 1/1 controllers/pids: Add new testcases
Hi!
Quoted 26 lines
testcases/kernel/controllers/pids/pids.sh new file +23
-0@@ -0,0 +1,109 @@ +#!/bin/sh + +################################################################################ +## ## +## Copyright (c) 2015 SUSE ## +## ## +## 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 ## +## ## +## Author: Cedric Hnyda <[email protected]> ## +## ##
testcases/kernel/controllers/pids/pids.sh line 33
Quoted 10 lines
testcases/kernel/controllers/pids/pids.sh new file +10
-0+################################################################################ + +# Usage +# ./pids.sh +# +# Description +# +# 1) Find if pids is mounted, if not mounted, pids will be mounted +# 2) Limit the number of pid avalaible and try to fork with pid_task +#
Can we do this a bit more versatile? I.e. pass the limit to test to the pids.sh script and change the pids_task to expect failure after N - 1 forks?
I would like to test this with limit set to 1, 2, 10, 50 and make sure that we can fork limit-1 tasks. That way we will do only one test in the script and call the script several times instead.
And actually what happens when the limit is reached and we try to attach process to it, does it fail too? And what happens when we set it to 0? (these are candidates for subsequent tests)
Quoted 35 lines
testcases/kernel/controllers/pids/pids.sh new file +35
-0+export TCID="pids" +export TST_TOTAL=2 +mounted=1 + +. test.sh + +cleanup() +{ + tst_resm TINFO "removing created directories" + rmdir $testpath + if [ "$mounted" -ne "1" ]; then + tst_resm TINFO "Umounting pids" + umount $mount_point + rmdir $mount_point + fi +} + +setup() +{ + tst_require_root + + exist=`grep -w pids /proc/cgroups | cut -f1`; + if [ "$exist" = "" ]; then + tst_brkm TCONF NULL "pids not supported" + fi + + mount_point=`grep -w pids /proc/mounts | cut -f 2 | cut -d " " -f2` + + if [ "$mount_point" = "" ]; then + mounted=0 + mount_point=/dev/cgroup + fi + + TST_CLEANUP=cleanup +
testcases/kernel/controllers/pids/pids.sh line 79
Quoted 10 lines
testcases/kernel/controllers/pids/pids.sh new file +10
-0+ testpath=$mount_point/ltp_$TCID + + if [ "$mounted" -eq "0" ]; then + ROD mkdir -p $mount_point + ROD mount -t cgroup -o pids none $mount_point + fi + ROD mkdir -p $testpath +} + +setup;
^ Useless semicolon
testcases/kernel/controllers/pids/pids.sh line 92
Quoted 12 lines
testcases/kernel/controllers/pids/pids.sh new file +12
-0+tst_resm TINFO "Restrict the number of pid to 1" +ROD echo 1 > $testpath/pids.max + +pids_task "$testpath/tasks" +ret=$? + +if [ "$ret" -eq "2" ]; then + tst_resm TPASS "fork failed as expected" +elif [ "$ret" -eq "0" ]; then + tst_resm TFAIL "fork didn't fail despite the limit" +else + tst_resm TBROK "failed to open $testpath/tasks"
Actually we may get here when the fork() failed unexpectedly as well.
It would be better to do just tst_resm TBROK "pids_task failed" here and print more detailed description of the failure from the pids_task instead.
Otherwise it looks good.
Signature · 2 lines
Cyril Hrubis [email protected]