xfs
[Top] [All Lists]

Re: [PATCH 3/5] [XFS] Block callers of xfs_flush_inodes() correctly.

To: Dave Chinner <david@xxxxxxxxxxxxx>
Subject: Re: [PATCH 3/5] [XFS] Block callers of xfs_flush_inodes() correctly.
From: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date: Mon, 16 Mar 2009 05:13:31 -0400
Cc: xfs@xxxxxxxxxxx, mpatocka@xxxxxxxxxx
In-reply-to: <1237116707-25793-4-git-send-email-david@xxxxxxxxxxxxx>
References: <1237116707-25793-1-git-send-email-david@xxxxxxxxxxxxx> <1237116707-25793-4-git-send-email-david@xxxxxxxxxxxxx>
User-agent: Mutt/1.5.18 (2008-05-17)
On Sun, Mar 15, 2009 at 10:31:45PM +1100, Dave Chinner wrote:
> xfs_flush_inodes() currently uses a magic timeout to wait for
> some inodes to be flushed before returning. This isn't
> really reliable but used to be the best that could be done
> due to deadlock potential of waiting for the entire flush.
> 
> Now the inode flush is safe to execute while we hold page
> and inode locks, we can wait for all the inodes to flush
> synchronously. Convert the wait mechanism to a completion
> to do this efficiently. This should remove all remaining
> spurious ENOSPC errors from the delayed allocation reservation
> path.

Why do we queue it up to a different thread if we synchronously wait
for it anyway?

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