VM deadlock avoidance

VM deadlock avoidance

[Peter Zijlstra]

Hi Dave,

I have a new version of these patches; I'm still using SOCK_VMIO socket
tagging and skb->emergency marks, since I have not come up with another
approach that might work and my RFC to netdev has so far been ignored.

Other than this though, it changed quite a bit;
 - I now use the regular allocation paths and cover all allocations
needed to process a skb (although the RX pool sizing might need more
variables)
 - The emergency RX pool size is based on ip[46]frag_high_thresh and
ip[46]_rt_max_size so that fragment assembly and dst route cache
allocations cannot exhaust the memory. (more paths need analysis xfrm,
conntrack?)
 - skb->emergency packets skip taps
 - skb->emergency packets warn about and ignores NF_QUEUE targets

http://programming.kicks-ass.net/kernel-patches/vm_deadlock/v9/

The patches definitely need more work but would you agree with the
general direction I'm working in or would you suggest yet another
direction?

Kind regards,

Peter

Re: VM deadlock avoidance

[David Miller]

From: Peter Zijlstra <a.p.zijlstra@chello.nl>
Date: Mon, 27 Nov 2006 15:06:25 +0100

> The patches definitely need more work but would you agree with the
> general direction I'm working in or would you suggest yet another
> direction?

You're definitely going to have to wait at least a week or so
before I can review anything serious, and in any event it needs
to be posted here so people can reply to the patch posting for
commentary, not stuck up on some random web site which makes
review terribly difficult.