xfs
[Top] [All Lists]

Re: [PATCH 4/5] xfs: use xfs_ilock_map_shared in xfs_attr_get

To: Christoph Hellwig <hch@xxxxxxxxxxxxx>
Subject: Re: [PATCH 4/5] xfs: use xfs_ilock_map_shared in xfs_attr_get
From: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Fri, 6 Dec 2013 08:05:57 +1100
Cc: xfs@xxxxxxxxxxx
Delivered-to: xfs@xxxxxxxxxxx
In-reply-to: <20131205210159.GA30318@xxxxxxxxxxxxx>
References: <20131205155830.620826868@xxxxxxxxxxxxxxxxxxxxxx> <20131205155951.679310054@xxxxxxxxxxxxxxxxxxxxxx> <20131205205910.GD29897@dastard> <20131205210159.GA30318@xxxxxxxxxxxxx>
User-agent: Mutt/1.5.21 (2010-09-15)
On Thu, Dec 05, 2013 at 01:01:59PM -0800, Christoph Hellwig wrote:
> On Fri, Dec 06, 2013 at 07:59:10AM +1100, Dave Chinner wrote:
> > I think the locking here should be moved inside xfs_attr_get_int()
> 
> Or we could just kill xfs_attr_get_int..
> 
> > so that it uses the same locking pattern as xfs_attr_set() and
> > xfs_attr_remove().
> > 
> > Also, xfs_attr_list() needs this treatment (the attr version of
> > readdir) as well (and it has the locking inside xfs_attr_list_int(),
> > too ;).
> >
> > It looks like xfs_readlink needs fixing, too.
> 
> Haven't really done an in-depth audit, mostly just looking at
> where the asserts kick in..

Right - I just did a scan with cscope on the users of
XFS_ILOCK_SHARED, and those two were the only ones that stuck out
that weren't handled correctly....

Cheers,

Dave.
-- 
Dave Chinner
david@xxxxxxxxxxxxx

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