Re: 2.6.14 kswapd eating too much CPU

From: Nick Piggin (nickpiggin_at_yahoo.com.au)
Date: 11/23/05

  • Next message: Rohit Seth: "RE: [PATCH 5/5] Light fragmentation avoidance without usemap: 005_drainpercpu" 
    Date:	Wed, 23 Nov 2005 13:19:56 +1100
To: Jan Kasprzak <kas@fi.muni.cz>

    Jan Kasprzak wrote:
    > Andrew Morton wrote:
    > : Jan Kasprzak <kas@fi.muni.cz> wrote:
    > : >
    > : > I have noticed that on my system kswapd eats too much CPU time every two
    > : > hours or so. This started when I upgraded this server to 2.6.14.2 (was 2.6.13.2
    > : > before), and added another 4 GB of memory (to the total of 8GB).
    > :
    > : Next time it happens, please gather some memory info (while it's happening):
    > :
    > : cat /proc/meminfo
    >
    [snip]

    > Hope this helps,
    >

    Can't see anything yet. Sysrq-M would be good. cat /proc/zoneinfo gets you
    most of the way there though.

    A couple of samples would be handy, especially from /proc/vmstat.

    cat /proc/vmstat > vmstat.out ; sleep 10 ; cat /proc/vmstat >> vmstat.out

    The same for /proc/zoneinfo would be a good idea as well.

    Also - when you say "too much cpu time", what does this mean? Does
    performance noticably drop compared with 2.6.13 performing the same cron
    job? Because kswapd is supposed to unburden other processes from page
    reclaim work, so if it is working *properly*, then the more CPU it uses
    the better.

    Thanks,
    Nick 

    -- 
SUSE Labs, Novell Inc.
Send instant messages to your online friends http://au.messenger.yahoo.com 
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/
  • Next message: Rohit Seth: "RE: [PATCH 5/5] Light fragmentation avoidance without usemap: 005_drainpercpu"

    Relevant Pages

    • [patch 0/6] remove PageReserved
      ... Most of the arch code is just reserved memory reporting, ... Hugh has some code which handles this case but he suggested that we ... Send instant messages to your online friends http://au.messenger.yahoo.com ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: [patch 2] mm: speculative get_page
      ... I'm in the process of recovering one of those with an earlier set ... need to be an atomic flag though, so it can be an "impossible" ... Send instant messages to your online friends http://au.messenger.yahoo.com ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: [patch 2.6.13-rc4] fix get_user_pages bug
      ... Please all hack it around as you see fit, ... pass back a special code VM_FAULT_WRITE to say it has done its full job ... Send instant messages to your online friends http://au.messenger.yahoo.com ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • [patch 0/7] lockless pagecache 2
      ... Code clarity and commenting improvement. ... RCU attempt - completely wrong?). ... Send instant messages to your online friends http://au.messenger.yahoo.com ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: RT patch acceptance
      ... > I think where we violently disagree is that in your earlier posts you ... If you look at some of my earlier posts, ... Send instant messages to your online friends http://au.messenger.yahoo.com ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    Loading