xfs
[Top] [All Lists]

Re: [PATCH] xfs: allow writeback from kswapd

To: xfs@xxxxxxxxxxx
Subject: Re: [PATCH] xfs: allow writeback from kswapd
From: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Fri, 9 Jul 2010 12:13:29 -0400
In-reply-to: <20100628143444.GB5473@xxxxxxxxxxxxx>
References: <20100628143444.GB5473@xxxxxxxxxxxxx>
User-agent: Mutt/1.5.20 (2009-08-17)
ping?

On Mon, Jun 28, 2010 at 10:34:44AM -0400, Christoph Hellwig wrote:
> We only need disable I/O from direct or memcg reclaim.
> 
> Signed-off-by: Christoph Hellwig <hch@xxxxxx>
> 
> Index: xfs-dev/fs/xfs/linux-2.6/xfs_aops.c
> ===================================================================
> --- xfs-dev.orig/fs/xfs/linux-2.6/xfs_aops.c  2010-06-28 11:57:06.652261386 
> +0200
> +++ xfs-dev/fs/xfs/linux-2.6/xfs_aops.c       2010-06-28 11:59:17.846068204 
> +0200
> @@ -1049,16 +1049,15 @@ xfs_vm_writepage(
>       /*
>        * Refuse to write the page out if we are called from reclaim context.
>        *
> -      * This is primarily to avoid stack overflows when called from deep
> -      * used stacks in random callers for direct reclaim, but disabling
> -      * reclaim for kswap is a nice side-effect as kswapd causes rather
> -      * suboptimal I/O patters, too.
> +      * This avoids stack overflows when called from deeply used stacks in
> +      * random callers for direct reclaim or memcg reclaim.  We explicitly
> +      * allow reclaim from kswapd as the stack usage there is relatively low.
>        *
>        * This should really be done by the core VM, but until that happens
>        * filesystems like XFS, btrfs and ext4 have to take care of this
>        * by themselves.
>        */
> -     if (current->flags & PF_MEMALLOC)
> +     if ((current->flags & (PF_MEMALLOC|PF_KSWAPD)) == PF_MEMALLOC)
>               goto out_fail;
>  
>       /*
> 
> _______________________________________________
> xfs mailing list
> xfs@xxxxxxxxxxx
> http://oss.sgi.com/mailman/listinfo/xfs
---end quoted text---

<Prev in Thread] Current Thread [Next in Thread>