Re: [PATCH] mm:vmscan: fix inaccurate reclaim during proactive reclaim

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



Thank you for your reply. shrink_lruvec() is nested in deep loop. Reclaimer may have already reclaimed part of requested memory in one loop, but before adjust sc->nr_to_reclaim in outer loop, call shrink_lruvec() again will still follow the current sc->nr_to_reclaim to work. It will eventually lead to overreclaim. 
Problematic case is easy to be constructed. Allocate lots of anonymous memory(e.g. 20G) in a memcg, then swapping by writing memory.reclaim and there is a certain probability of overreclaim.
(Sorry to find out that there was a problem with the format of the previous reply. Thanks again.)

[Index of Archives]     [Linux ARM Kernel]     [Linux ARM]     [Linux Omap]     [Fedora ARM]     [IETF Annouce]     [Bugtraq]     [Linux OMAP]     [Linux MIPS]     [eCos]     [Asterisk Internet PBX]     [Linux API]

  Powered by Linux