Re: [RFC PATCH 1/5] x86: introduce preemption disable prefix

From: Peter Zijlstra
Date: Thu Oct 18 2018 - 03:54:27 EST

On Wed, Oct 17, 2018 at 06:22:48PM -0700, Andy Lutomirski wrote:
> > On Oct 17, 2018, at 5:54 PM, Nadav Amit <namit@xxxxxxxxxx> wrote:
> >
> > It is sometimes beneficial to prevent preemption for very few
> > instructions, or prevent preemption for some instructions that precede
> > a branch (this latter case will be introduced in the next patches).
> >
> > To provide such functionality on x86-64, we use an empty REX-prefix
> > (opcode 0x40) as an indication that preemption is disabled for the
> > following instruction.
> Nifty!
> That being said, I think you have a few bugs.

> First, you canât just ignore a rescheduling interrupt, as you
> introduce unbounded latency when this happens â youâre effectively
> emulating preempt_enable_no_resched(), which is not a drop-in
> replacement for preempt_enable().

> To fix this, you may need to jump to a slow-path trampoline that calls
> schedule() at the end or consider rewinding one instruction instead.
> Or use TF, which is only a little bit terrifying...

At which point we're very close to in-kernel rseq.