Messages in this thread | | | Date | Tue, 29 Sep 2015 13:41:50 -0400 | Subject | Re: [RFC][PATCH] sched: Fix TASK_DEAD race in finish_task_switch() | From | Linus Torvalds <> |
| |
On Tue, Sep 29, 2015 at 12:50 PM, Peter Zijlstra <peterz@infradead.org> wrote: > > The control dependency creates a LOAD->STORE order, that is, no STOREs > can happen until we observe !p->on_cpu.
Fair enough.
> Right, but wmb isn't sufficient as it doesn't order the prev->state LOAD > vs the prev->on_cpu = 0 STORE. If those happen in the wrong order the > described race can happen and we get a use-after-free.
.. and you convinced me. The patch is good. Ack.
Linus
| |