On Sun, Oct 23, 2016 at 05:32:21PM -0400, Chris Mason wrote:
>
>
> On 10/22/2016 11:20 AM, Dave Jones wrote:
> > On Fri, Oct 21, 2016 at 04:02:45PM -0400, Dave Jones wrote:
> >
> > > > It could be worth trying this, too:
> > > >
> > > > https://git.kernel.org/cgit/linux/kernel/git/luto/linux.git/commit/?h=x86/vmap_stack&id=174531fef4e8
> > > >
> > > > It occurred to me that the current code is a little bit fragile.
> > >
> > > It's been nearly 24hrs with the above changes, and it's been pretty much
> > > silent the whole time.
> > >
> > > The only thing of note over that time period has been a btrfs lockdep
> > > warning that's been around for a while, and occasional btrfs checksum
> > > failures, which I've been seeing for a while, but seem to have gotten
> > > worse since 4.8.
> > >
> > > I'm pretty confident in the disk being ok in this machine, so I think
> > > the checksum warnings are bogus. Chris suggested they may be the result
> > > of memory corruption, but there's little else going on.
> >
> > The only interesting thing last nights run was this..
> >
> > BUG: Bad page state in process kworker/u8:1 pfn:4e2b70
> > page:ffffea00138adc00 count:0 mapcount:0 mapping:ffff88046e9fc2e0 index:0xdf0
> > flags: 0x400000000000000c(referenced|uptodate)
> > page dumped because: non-NULL mapping
> > CPU: 3 PID: 24234 Comm: kworker/u8:1 Not tainted 4.9.0-rc1-think+ #11
> > Workqueue: writeback wb_workfn (flush-btrfs-2)
>
> Well crud, we're back to wondering if this is Btrfs or the stack
> corruption. Since the pagevecs are on the stack and this is a new
> crash, my guess is you'll be able to trigger it on xfs/ext4 too. But we
> should make sure.
Here's an interesting one from today, pointing the finger at xattrs again.
[69943.450108] Oops: 0003 [#1] PREEMPT SMP DEBUG_PAGEALLOC
[69943.454452] CPU: 1 PID: 21558 Comm: trinity-c60 Not tainted 4.9.0-rc1-think+ #11
[69943.463510] task: ffff8804f8dd3740 task.stack: ffffc9000b108000
[69943.468077] RIP: 0010:[<ffffffff810c3f6b>]