Re: 2.6.0-test4-mm1 - kswap hogs cpu OO takes ages to start!

From: Rik van Riel (riel_at_redhat.com)
Date: 08/31/03

  • Next message: Petr Baudis: "IDE DMA breakage w/ 2.4.21+ and 2.6.0-test4(-mm4)" 
    Date:	Sun, 31 Aug 2003 12:06:59 -0400 (EDT)
To: Con Kolivas <kernel@kolivas.org>

    On Thu, 28 Aug 2003, Con Kolivas wrote:

    > > Does this make a difference?
    >
    > Tried it. No change.
    >
    > kswapd0 can hit 90% cpu at times unless the swappiness is increased.

    Looks like the problem is that cache and process pages are on
    the same lists, forcing kswapd to scan the lists endlessly.

    One thing you could try is splitting the lists, at least the
    active list, like done in 2.4-rmap15... 

    -- 
"Debugging is twice as hard as writing the code in the first place.
Therefore, if you write the code as cleverly as possible, you are,
by definition, not smart enough to debug it." - Brian W. Kernighan
-
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: Petr Baudis: "IDE DMA breakage w/ 2.4.21+ and 2.6.0-test4(-mm4)"

    Relevant Pages

    • Re: [PATCH] Split futex global spinlock futex_lock
      ... multiple lists in each one. ... There isn't any speed penalty for having a lock per list_head when you ... because the cache line has to bounce anyway for the ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: large files unnecessary trashing filesystem cache?
      ... > or system wide) you can have in filesystem cache. ... into a small number of lists ... And the other way is to make it so complicated that there are no obvious deficiencies. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: Problems with kernel mmap (failing tst-mmap-eofsync in glibc on parisc)
      ... > This last read fails to get the data on parisc. ... > CPU cache is virtually indexed, and the page the write is storing the ... flush_dcache_pagechecks both the shared and non-shared mmap lists, ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Oops on APM wakeup on 2.4.22/23-pre, works with 2.6.0-test
      ... use 2.6.0 due to other problems such as unstable PCMCIA ... right now (have to solder null-modem cable first etc), ... the followups to me, I don't read the lists ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)
    • Re: dev->release = (void (*)(struct device *))kfree;
      ... will still be unregistered (removed from the various lists) but the ... releasecallback won't be called until the last user. ... send the line "unsubscribe linux-kernel" in ...
      (Linux-Kernel)