On Wed, 10 Mar 2010 02:41:08 -0800 (PST) David Rientjes <rientjes@xxxxxxxxxx> wrote: > This patchset is a rewrite of the out of memory killer to address several > issues that have been raised recently. The most notable change is a > complete rewrite of the badness heuristic that determines which task is > killed; the goal was to make it as simple and predictable as possible > while still addressing issues that plague the VM. > > Changes from version 2: > > - updated to mmotm-2010-03-09-19-15 > > - schedule a timeout for current if it was not selected for oom kill > when it has returned VM_FAULT_OOM so memory can freed to prevent > needlessly recalling the oom killer and looping. > > To apply, download the -mm tree from > http://userweb.kernel.org/~akpm/mmotm/broken-out.tar.gz first. > > This patchset is also available for each kernel release from: > > http://www.kernel.org/pub/linux/kernel/people/rientjes/oom-killer-rewrite/ > One question. Assume a host A and B. A has 4G memory, B has 8G memory. Here, an applicaton which consumes 2G memory. Then, this application's oom_score will be 500 on A, 250 on B. How admin detemine the best oom_score_adj value ? Does it depend on envrionment even if runnning the same application ? Is it bad to use bare value as echo 1G > /proc/<pid>/oom_score_adj for getting 1G bytes of excuse to this application ? Thanks, -Kame -- 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/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>