Re: fuse: kernel BUG at mm/truncate.c:763!

[Matthew Wilcox]

On Mon, Mar 29, 2021 at 10:01:58AM +0100, Luis Henriques wrote:
> On Fri, Mar 19, 2021 at 09:02:33AM +0000, Luis Henriques wrote:
> > On Thu, Mar 18, 2021 at 11:55:43AM +0000, Matthew Wilcox wrote:
> > > On Thu, Mar 18, 2021 at 11:29:28AM +0000, Luis Henriques wrote:
> > > > On Thu, Mar 18, 2021 at 02:03:02PM +0300, Kirill A. Shutemov wrote:
> > > > > On Thu, Mar 18, 2021 at 11:59:59AM +0100, Miklos Szeredi wrote:
> > > > > > > [16247.536348] page:00000000dfe36ab1 refcount:673 mapcount:0 mapping:00000000f982a7f8 index:0x1400 pfn:0x4c65e00
> > > > > > > [16247.536359] head:00000000dfe36ab1 order:9 compound_mapcount:0 compound_pincount:0
> > > > > > 
> > > > > > This is a compound page alright.   Have no idea how it got into fuse's
> > > > > > pagecache.
> > > > > 
> > > > > 
> > > > > Luis, do you have CONFIG_READ_ONLY_THP_FOR_FS enabled?
> > > > 
> > > > Yes, it looks like Tumbleweed kernels have that config option enabled by
> > > > default.  And it this feature was introduced in 5.4 (the bug doesn't seem
> > > > to be reproducible in 5.3).
> > > 
> > > Can you try adding this patch?
> > > 
> > > https://git.infradead.org/users/willy/pagecache.git/commitdiff/369a4fcd78369b7a026bdef465af9669bde98ef4
> > 
> > Good news, looks like this patch fixes the issue[1].  Thanks a lot
> > everyone.  Is this already queued somewhere for 5.12?  Also, it would be
> > nice to have it Cc'ed for stable kernels >= 5.4.
> 
> Ping.  Are you planning to push this for 5.12, or is that queued for the
> 5.13 merged window?  Or "none of the above"? :)

Sorry, dropped the ball on this one.  This patch is good for that point
in the patch series, but I'm not sure it works against upstream in all
cases.  I need to spend some time evaluating it.  Thanks for the reminder.