Re: [RFC PATCH 0/6] Do not call ->writepage[s] from direct reclaimand use a_ops->writepages() where possible

[Christoph Hellwig]

On Tue, Jun 15, 2010 at 06:14:19PM +0200, Andrea Arcangeli wrote:
> On Tue, Jun 15, 2010 at 04:38:38PM +0100, Mel Gorman wrote:
> > That is pretty much what Dave is claiming here at
> > http://lkml.org/lkml/2010/4/13/121 where if mempool_alloc_slab() needed
> 
> This stack trace shows writepage called by shrink_page_list... that
> contradict Christoph's claim that xfs already won't writepage if
> invoked by direct reclaim.

We only recently did that - before that we tried to get the VM fixed
multiple times but finally had to bite the bullet and follow ext4 and
btrfs in that regard.

> Again not what looks like from the stack trace. Also grepping for
> PF_MEMALLOC in fs/xfs shows nothing. In fact it's ext4_write_inode
> that skips the write if PF_MEMALLOC is set, not writepage apparently
> (only did a quick grep so I might be wrong). I suspect
> ext4_write_inode is the case I just mentioned about slab shrink, not
> ->writepage ;).

ext4 in fact does not check PF_MEMALLOC but simply refuses to write
out anything in ->writepage in most cases.  There is a corner case
when the page doesn't have any buffers attached where it wouldn't
have write out data, without actually calling the allocator.  I
suspect this code actually is a leftover as we don't normally strip
buffers from a page that had them before.

> inodes are small, it's no big deal to keep an inode pinned and not
> slab-reclaimable because dirty, while skipping real writepage in
> memory pressure could really open a regression in oom false positives!
> One pagecache much bigger than one inode and there can be plenty more
> dirty pagecache than inodes.

At least for XFS ->write_inode is really simple these days.  If it's
a synchronous writeout, which won't happen from these path it logs the
inode, which is far less harmless than the whole allocator code, and
for write = 0 it only adds it to the delayed write queue, which doesn't
call into the I/O stack at all.

--
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/