On 2011.04.29 at 12:58 -0700, Christian Kujau wrote: > On Fri, 29 Apr 2011 at 09:37, Dave Chinner wrote: > > Did you read the email that pointed to CONFIG_TINY_RCU as the > > ptotential source of the problem? Can you change to CONFIG_TREE_RCU > > and see if the problem goes away? > > Tried with TREE_RCU, unfortunately the problem persists. However, the OOM > messages did not make it to the disk this time. Syslog continued to work > until 11:51 (local time), the slabinfo-8-TREE_RCU.txt file was written to > until 11:57, the machine was already unusable at that point. At 12:07 I > took a "screenshot" and then powercycled the machine: > > http://nerdbynature.de/bits/2.6.39-rc4/oom/ > (The files marked with -8 are the current ones) > > Btw, the xfs filesystem I'm du'ing over is still mounted with "noatime". > > Any ideas? Patches to try? I could be the hrtimer bug again. Would you try to reproduce the issue with this patch applied? http://git.us.kernel.org/?p=linux/kernel/git/tip/linux-2.6-tip.git;a=commit;h=ce31332d3c77532d6ea97ddcb475a2b02dd358b4 -- Markus _______________________________________________ xfs mailing list xfs@xxxxxxxxxxx http://oss.sgi.com/mailman/listinfo/xfs
