[openib-general] Re: [PATCH]: was Re: LLTX and netif_stop_queue

[David S. Miller]

On Thu, 27 Jan 2005 08:16:45 +0100
Andi Kleen <ak at suse.de> wrote:

> > Yes, that's the idea.
> 
> per CPU only works in preemptive kernel if you have preemption
> disabled all the time. Do you? 
> 
> Seems not likely to me.

BH is disabled in these code paths (specifically we're talking
about ->hard_start_xmit()), as that is where the recursion
check goes.

Otherwise, loopback's LLTX is broken as it relies on this property
as well.