Re: [RFC 1/2] softirq: Defer net rx/tx processing to ksoftirqd context

[Peter Zijlstra]

On Thu, Jan 11, 2018 at 11:38:40AM -0500, David Miller wrote:
> From: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> Date: Thu, 11 Jan 2018 17:32:04 +0100
> 
> > You just introduced a 2ms preempt-disable region I think, that's not
> > cool for PREEMPT and a plain bug on PREEMPT_RT.
> 
> I'd just like to know what the difference is between ksoftirqd running
> softirq work for 2ms and the real softirq context doing it.  The
> latter already has this "2ms preempt-disable region" of sorts, right?

True I suppose.. and PREEMPT_RT does the whole softirq business
differently anyway.

I'll go back to staring at PTI crud..