On Tue, 6 Mar 2012 23:21:52 -0800 (PST) David Rientjes <rientjes@xxxxxxxxxx> wrote: > Nope, all patches I've ever proposed for the oom killer have been merged > in some form or another. > > > When exiting a process which have plenty threads, this patch allow to eat all > > of reserve memory > > and bring us new serious failure. > > > > It closes the risk of livelock if an oom killed thread, thread A, cannot > exit because it's blocked on another thread, thread B, which cannot exit > because it requires memory in the exit path and doesn't have access to > memory reserves. So this patch makes it more likely that an oom killed > thread will be able to exit without livelocking. But it also "allow to eat all of reserve memory and bring us new serious failure". In theory, at least. And afaict the proposed patch is a theoretical thing as well. Has anyone sat down and created tests to demonstrate either problem? This patch is either two-steps-forward-and-one-back or it is one-step-forward-and-two-steps-back. How are we to determine which of these it is? -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/ Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>