On Wed, 2015-04-08 at 15:12 +0100, David Woodhouse wrote:
> On Wed, 2015-04-08 at 15:08 +0200, Johannes Berg wrote:
> > Additionally, the failure mode of this was the process running out of
> > memory due to receiving the same results over and over again - does that
> > happen for you? It seems it was stuck in recvmsg(), but that may just be
> > a side effect of happening to interrupt at that point?
> 
> No, strace shows it's just sitting in recvmsg().
> 
> As I said, I'm not *sure* it's caused by the same commit; bisecting is
> distinctly non-trivial. It just seemed likely.

FWIW I went back to the Fedora 3.19 kernel for a week and it didn't
show up again. After rebooting to 4.0 earlier today, it's happened
already.

I'll see if I can find a more reliable way of reproducing it, which
will make it slightly saner to try bisecting.


-- 
dwmw2