Re: lowmem_reserve (replaces protection)

From: Andrew Morton (akpm_at_osdl.org)
Date: 10/27/04

  • Next message: K.R. Foley: "Re: [patch] Real-Time Preemption, -RT-2.6.9-mm1-V0" 
    Date:	Tue, 26 Oct 2004 20:43:08 -0700
To: Nick Piggin <nickpiggin@yahoo.com.au>

    Nick Piggin <nickpiggin@yahoo.com.au> wrote:
    >
    > It actually can overscan lower zones a little bit, because
    > whenever any higher zone in the pgdat is low on memory, then
    > it and all zones below it get scanned too.

    Because we know that all of the eligible zones are below pages_low. kswapd
    will then work to bring all the relevant zones back to pages_high.

    When working on this code it is very very easy to break the zone levelling:
    you *have* to run a workload mix and monitor the numbers in /proc/vmstat to
    ensure that all zones are undergoing page scanning at frequencies which are
    proportional to their sizes. It's easy to screw up the zone levelling so
    all allocations end up coming from ZONE_NORMAL and pagecache pages in, say,
    ZONE_DMA end up just sitting there.

    -
    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: K.R. Foley: "Re: [patch] Real-Time Preemption, -RT-2.6.9-mm1-V0"

    Relevant Pages