Re: [BUG] loop.c oopses

[Jens Axboe]

On Tue, Jul 16 2002, Andrew Morton wrote:
> Jens Axboe wrote:
> > 
> > On Tue, Jul 16 2002, Andrew Morton wrote:
> > > William Lee Irwin III wrote:
> > > >
> > > > loop.c oopses when bio_copy() returns NULL. This was encountered while
> > > > running dbench 16 on a loopback-mounted reiserfs filesystem.
> > >
> > > ugh.  GFP_NOIO is evil.  I guess it's better to add __GFP_HIGH
> > > there, but it's not a happy solution.
> > 
> > GFP_NOIO has __GFP_WAIT set, so bio_copy -> bio_alloc -> mempool_alloc
> > should never fail. Puzzled.
> > 
> 
> Presumably the loop driver was called from within shrink_cache(),
> as PF_MEMALLOC.  Those allocations can fail.

Maybe I'm being dense, but I don't see how PF_MEMALLOC would prevent
mempool_alloc() from doing the right thing still. In the end we'll just
end up stalling on our own pool.

> That's maybe wrong - if there are a decent number of pages
> under writeback then we should be able to just wait it out.
> But it gets tricky with the loop driver...

Indeed

-- 
Jens Axboe