On Mon 25-07-16 15:30:59, Ross Zwisler wrote:
> On Fri, Jul 22, 2016 at 02:19:40PM +0200, Jan Kara wrote:
> > Currently PTE gets updated in wp_pfn_shared() after dax_pfn_mkwrite()
> > has released corresponding radix tree entry lock. When we want to
> > writeprotect PTE on cache flush, we need PTE modification to happen
> > under radix tree entry lock to ensure consisten updates of PTE and radix
> > tree (standard faults use page lock to ensure this consistency). So move
> > update of PTE bit into dax_pfn_mkwrite().
> >
> > Signed-off-by: Jan Kara <[email protected]>
>
> After applying the whole series to a v4.7 baseline I was hitting a deadlock in
> my testing, and it bisected to this commit. This deadlock happens in my QEMU
> guest with generic/068, ext4 and DAX. It reproduces 100% of the time after
> this commit.
>
> Here is the lockdep info, passed through kasan_symbolize.py:
Thanks! I've checked why I didn't see this and apparently I've run last
round of testing on the wrong branch. Drat.
I've fixed the bug you've spotted (we need to release tree_lock earlier)
but xfstests are triggering some more issues now for me so I'm debugging
those.
Honza
--
Jan Kara <[email protected]>
SUSE Labs, CR
_______________________________________________
Linux-nvdimm mailing list
[email protected]
https://lists.01.org/mailman/listinfo/linux-nvdimm
