xfs
[Top] [All Lists]

Re: [PATCH 2/2] xfs: reclaim all inodes by background tree walks

To: Dave Chinner <david@xxxxxxxxxxxxx>
Subject: Re: [PATCH 2/2] xfs: reclaim all inodes by background tree walks
From: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Fri, 8 Jan 2010 05:24:08 -0500
Cc: xfs@xxxxxxxxxxx
In-reply-to: <1262819125-27083-3-git-send-email-david@xxxxxxxxxxxxx>
References: <1262819125-27083-1-git-send-email-david@xxxxxxxxxxxxx> <1262819125-27083-3-git-send-email-david@xxxxxxxxxxxxx>
User-agent: Mutt/1.5.19 (2009-01-05)
Looks safe to me.  I wonder whaimpact leaving the inodes around for
longer has to memory usage for inode heavy workloads, though.

>  unlock_and_requeue:
> +     /*
> +      * We could return EAGAIN here to make reclaim rescan the inode tree in
> +      * a short while. However, this just burns CPU time scanning the tree
> +      * waiting for IO to complete and xfssyncd never goes back to the idle
> +      * state. Instead, return 0 to let the next scheduled background reclaim
> +      * attempt to reclaim the inode again.
> +      */
>       xfs_iflags_clear(ip, XFS_IRECLAIM);
>       xfs_iunlock(ip, XFS_ILOCK_EXCL);
> -     return EAGAIN;
> +     return 0;

This is an unrelated change and should be a patch of it's own.

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