On Wed, 8 Dec 2010, CAI Qian wrote: > Bisect indicated that this is the first bad commit, > > commit 696d3cd5fb318c070dc757fe109e04e398138172 > Author: David Rientjes <rientjes@xxxxxxxxxx> > Date: Fri Jun 11 22:45:17 2010 +0200 > > __out_of_memory() only has a single caller, so fold it into > out_of_memory() and add a comment about locking for its call to > oom_kill_process(). > > Signed-off-by: David Rientjes <rientjes@xxxxxxxxxx> > Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@xxxxxxxxxxxxxx> > Cc: KOSAKI Motohiro <kosaki.motohiro@xxxxxxxxxxxxxx> > Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> > This commit dropped the releasing of tasklist_lock when the oom killer chooses not to act because it finds another task that has already been killed but has yet to exit. That's fixed by b52723c5, so this bisect isn't the source of your problem. You didn't report the specific mmotm kernel that this was happening on, so trying to diagnose or reproduce it is diffcult. Could you try 2.6.37-rc5 with your test case? If it works fine, could you try mmotm-2010-12-02-16-34? -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxxx For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom policy in Canada: sign http://dissolvethecrtc.ca/ Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>