[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20140809143034.GA28345@localhost>
Date: Sat, 9 Aug 2014 22:30:34 +0800
From: Fengguang Wu <fengguang.wu@...el.com>
To: Peter Zijlstra <peterz@...radead.org>
Cc: Dave Hansen <dave.hansen@...el.com>,
LKML <linux-kernel@...r.kernel.org>, lkp@...org
Subject: [sched] Out of memory: Kill process 2999 (rc) score 9 or sacrifice
child
Hi Peter,
We noticed the below changes on
git://git.kernel.org/pub/scm/linux/kernel/git/peterz/queue.git sched/wait
commit d58d631b474c552dce72da2dce9dd276d731b79a ("sched: Fix finish_task_switch vs prev_state")
test case: vm-vp-quantal-x86_64/boot/1
9e6e6179961e8dd d58d631b474c552dce72da2dc testbox/testcase/testparams
--------------- ------------------------- ---------------------------
0 +Inf% 1 ± 0% TOTAL dmesg.Out_of_memory:Kill_process
0 +Inf% 0 ±50% TOTAL dmesg.Kernel_panic-not_syncing:Out_of_memory_and_no_killable_processes
In commit 9e6e6179961e8dd, the boot dmesg is
[ 7.537598] Freeing unused kernel memory: 3352K (ffffffff82230000 - ffffffff82576000)
[ 7.558273] random: init urandom read with 11 bits of entropy available
[ 7.687132] init: Failed to create pty - disabling logging for job
[ 7.688578] init: Temporary process spawn error: No space left on device
[ 68.298970] reboot: Restarting system
In d58d631b474c552dce72da2dc, the OOM occurred immediately after the
"No space left on device" line. The qemu has mem=320M and the initrds
are 24M in total. What's interesting is, in the 5 boot tests for
d58d631b47 and its parent commit, this OOM message is 100%
reproducible on commit d58d631b47, while its parent boots all fine.
Thanks,
Fengguang
View attachment ".dmesg" of type "text/plain" (70809 bytes)
Powered by blists - more mailing lists