Re: Possible lock-less list race in scheduler_ipi()

[Mathieu Desnoyers <mathieu.desnoyers@efficios.com>]

----- Original Message -----
> From: "Steven Rostedt" <rostedt@goodmis.org>
> To: "Mathieu Desnoyers" <mathieu.desnoyers@efficios.com>
> Cc: "Linus Torvalds" <torvalds@linux-foundation.org>, "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>, "Huang Ying"
> <ying.huang@intel.com>, "Lai Jiangshan" <laijs@cn.fujitsu.com>, "Lai Jiangshan" <eag0628@gmail.com>, "Peter
> Zijlstra" <peterz@infradead.org>, "LKML" <linux-kernel@vger.kernel.org>, "Ingo Molnar" <mingo@kernel.org>
> Sent: Friday, March 6, 2015 2:03:47 PM
> Subject: Re: Possible lock-less list race in scheduler_ipi()
> 
> On Fri, 6 Mar 2015 14:35:23 +0000 (UTC)
> Mathieu Desnoyers <mathieu.desnoyers@efficios.com> wrote:
> 
> 
> > Indeed, the compiler should never reorder loads/stores from/to
> > same memory location from a program order POV. What I had in mind
> > is a bit more far-fetched though: it would involve having the compiler
> > reorder this load after a store to another memory location, which
> > would in turn allow another execution context (interrupt or thread)
> > to corrupt the list.
> 
> You mean on another CPU? Because the code you are worried about has
> interrupts disabled.

I'm worried that another CPU could issue try_to_wake_up() on a
task concurrently with the llist iteration within sched_ttwu_pending().

AFAIU, ttwu_queue_remote() is called from ttwu_queue() without holding
the rq lock. So I'm wondering what prevents corruption of the wake_list
in this situation.

Thanks,

Mathieu

-- 
Mathieu Desnoyers
EfficiOS Inc.
http://www.efficios.com