Re: [PATCH] fix spurious OOM kills

[Thomas Gleixner]

Andrea,

On Wed, 2004-12-15 at 00:55 +0100, Andrea Arcangeli wrote:
> On Wed, Dec 15, 2004 at 10:30:18AM +1100, Nick Piggin wrote:
> > Was there some swap-token (or can anyone think of any relevant)
> > changes recently?
> 
> The only thing I know about recent swap-token changes, is that Andrew
> fixed a bug by ignoring the token thing at priority 0, this the
> early-oom failures in my queue (the VM bug was introduced sometime
> before or during 2.6.9-rc). The fix is already in mainline according to
> my out of sync copy of kernel CVS.

cset 1.2055.40.21
vmscan: ignore swap token when in trouble 

It solves one of the problems, but your fix is really the only complete
fix I have in hands since this thread(s) started. + my simple changes to
the whom to kill selection :)

Also your modification slow down the machine in case of low memory. This
makes sense and does not hurt anybody. But it's still better than the
crappy oom kills which I still have with rc3 even in "normal" desktop
usage.

<RANT exclude="andrea">
We are discussing this since nearly two month. Is there a final fixup in
sight before christmas ? And I mean christmas 2004.

I'm just waiting for somebody who advises me to put more memory into my
boxes.
</RANT>

tglx


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/