Re: [RFC PATCH] sys_membarrier(): system/process-wide memory barrier (x86) (v12)

[Steven Rostedt <rostedt@goodmis.org>]

On Tue, 17 Mar 2015 12:46:41 +0000 (UTC)
Mathieu Desnoyers <mathieu.desnoyers@efficios.com> wrote:


> > Would that help?
> > 
> > 	this_cpu_write(saved_next, next);
> > 	rq = context_switch(rq, prev, next);
> > 	rq->curr = this_cpu_read(saved_next);
> 
> Assuming there is a full memory barrier (e.g. load_cr3) within
> context_switch, it would help for ordering memory accesses that
> are performed prior to the preemption, but not for memory accesses
> to be performed immediately after return to userspace from preemption.

Hmm, I was thinking that there was a spin_unlock(rq->lock) after that,
but it appears that context_switch() does the unlock. If there was an
spin_unlock() after this code, then it could work. There's always
setting the rq->curr in the context_switch() call itself.

-- Steve