> YAMAMOTO Takashi wrote: > > hi, > > > >> Daisuke Nishimura wrote: > >>> Hi, Yamamoto-san. > >>> > >>> I'm reviewing and testing your patch now. > >>> > >> In building kernel infinitely(in a cgroup of > >> memory.limit=64M and swap.limit=128M, with swappiness=100), > >> almost all of the swap (1GB) is consumed as swap cache > >> after a day or so. > >> As a result, processes are occasionally OOM-killed even when > >> the swap.usage of the group doesn't exceed the limit. > >> > >> I don't know why the swap cache uses up swap space. > >> I will test whether a similar issue happens without your patch. > >> Do you have any thoughts? > > > > my patch tends to yield more swap cache because it makes try_to_unmap > > fail and shrink_page_list leaves swap cache in that case. > > i'm not sure how it causes 1GB swap cache, tho. > > > > Agree. > > I suspected that the cause of this problem was the behavior > of shrink_page_list as you said, so I thought one of Rik's > split-lru patchset: > > http://lkml.org/lkml/2008/3/4/492 > [patch 04/20] free swap space on swap-in/activation > > would reduce the usage of swap cache to half of the total swap. > But it didn't help, so I think there may be some other causes. do you mean you tested with the patch in the url? i don't think remove_exclusive_swap_page works for us because our page has more references than it expects. ie. ptes, cache, isolate_page (unless your tree has another change for remove_exclusive_swap_page. i haven't checked other patches in the patchset.) YAMAMOTO Takashi > > > Thanks, > Daisuke Nishimura. _______________________________________________ Containers mailing list Containers@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linux-foundation.org/mailman/listinfo/containers