Re: [RFC PATCH 0/3] Avoid the use of congestion_wait under zone pressure

Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]
From: Andrew Morton
Date: Thursday, March 11, 2010 - 4:41 pm

On Mon,  8 Mar 2010 11:48:20 +0000
Mel Gorman <mel@csn.ul.ie> wrote:


clear_bdi_congested() is called each time a write completes and the
queue is below the congestion threshold.

So if the page allocator or kswapd call congestion_wait() against a
non-congested queue, they'll wake up on the very next write completion.

Hence the above-quoted claim seems to me to be a significant mis-analysis and
perhaps explains why the patchset didn't seem to help anything?
--
Previous message: [thread] [date] [author]
Next message: [thread] [date] [author]

Messages in current thread:
Re: [PATCH 1/3] page-allocator: Under memory pressure, wai ..., Christian Ehrhardt, (Tue Mar 9, 8:42 am)
Re: [PATCH 1/3] page-allocator: Under memory pressure, wai ..., Christian Ehrhardt, (Tue Mar 9, 8:56 am)
Re: [PATCH 1/3] page-allocator: Under memory pressure, wai ..., Christoph Lameter, (Tue Mar 9, 10:11 am)
Re: [RFC PATCH 0/3] Avoid the use of congestion_wait under ..., Andrew Morton, (Thu Mar 11, 4:41 pm)
Re: [RFC PATCH 0/3] Avoid the use of congestion_wait under ..., Christian Ehrhardt, (Thu Mar 11, 11:39 pm)
Re: [RFC PATCH 0/3] Avoid the use of congestion_wait under ..., Christian Ehrhardt, (Fri Mar 12, 5:15 am)
Re: [RFC PATCH 0/3] Avoid the use of congestion_wait under ..., Christian Ehrhardt, (Mon Mar 15, 5:34 am)
Re: [RFC PATCH 0/3] Avoid the use of congestion_wait under ..., Christian Ehrhardt, (Mon Mar 15, 7:45 am)