Re: bad page state in 3.13-rc4

[Dave Jones]

On Thu, Dec 19, 2013 at 08:11:58PM +0000, Mel Gorman wrote:
 
 > Dave, was this a NUMA machine?

It's a dual core i5-4670T with hyperthreading.

 > If yes, was CONFIG_NUMA_BALANCING set?

no.

 > Dave, when this this
 > bug start triggering? If it's due to a recent change in trinity, can you
 > check if 3.12 is also affected? If not, can you check if the bug started
 > happening somewhere around these commits?
 
Right now it can take hours for it to reproduce. Until I can narrow it down to
something repeatable, bisecting and trying old builds is going to be really time-consuming.

Given the other VM bugs that have Sasha and I have been finding since I added
the mmap reuse code to trinity, this is probably something else that has been
there for a while.

 > A few bad state bugs have shown up on linux-mm recently but my impression
 > was that they were related to rmap_walk changes currently in next. The
 > initial log indicated that this was 3.13-rc4 but is it really 3.13-rc4 or
 > are there any -next patches applied?

no, just rc4 (plus a handful of small patches to fix oopses etc that I've already
diagnosed).  I'm glad Sasha spends time running this stuff on -next, because 
there aren't enough hours in the day for me to look at the stuff I find
in Linus' tree without looking at what's coming next.

	Dave

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/