> For what it's worth, I've had no reports of anything like this on
> x86 cpus. The VM in 2.1 looks rock solid, with the single exception
> of the kswapd looping resource hog issue and the eternal debate
> about oom behaviour.
A patch for this will be out later this week. Probably
tomorrow, but maybe even tonight...
The looping behaviour is easily fixed. If kswapd is looping,
it means that we're out of memory and that try_to_free_page()
is failing. That means that we can easily test for OOM when
kswapd starts failing --> looping solved.
Then there's the question of whether we should kill a random
process or a non-random one. When looking at it strictly,
there's hardly an argument for the "I want a random process
killed" folks, since a non-random one is just one of the
choices the random process killing routine could have made...
The patch will be available from my home page, just like
the scheduler bigpatch I posted today.
Rik.
+-------------------------------------------------------------------+
| Linux memory management tour guide. H.H.vanRiel@phys.uu.nl |
| Scouting Vries cubscout leader. http://www.phys.uu.nl/~riel/ |
+-------------------------------------------------------------------+
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/