Re: [REVIEW] Invalidate dentry on unlink/rmdir if using CI mode

xfs

Re: [REVIEW] Invalidate dentry on unlink/rmdir if using CI mode

To:	Barry Naujok <bnaujok@xxxxxxx>
Subject:	Re: [REVIEW] Invalidate dentry on unlink/rmdir if using CI mode
From:	Christoph Hellwig <hch@xxxxxxxxxxxxx>
Date:	Fri, 13 Jun 2008 06:25:55 -0400
Cc:	"xfs@xxxxxxxxxxx" <xfs@xxxxxxxxxxx>
In-reply-to:	<op.ucoa9vjb3jf8g2@pc-bnaujok.melbourne.sgi.com>
References:	<op.ucoa9vjb3jf8g2@pc-bnaujok.melbourne.sgi.com>
Sender:	xfs-bounce@xxxxxxxxxxx
User-agent:	Mutt/1.5.17 (2007-11-01)

On Fri, Jun 13, 2008 at 04:06:57PM +1000, Barry Naujok wrote:
> Once of the interesting features of vfs_unlink/d_delete functionality
> is that if it's the only inode being referenced, it makes the dentry
> negative. CI doesn't work with negative dentries, so the following code
> invalidate the dentry in CI mode.

Please add a comment like this patch description near the the calls
you added, otherwise ok.

[More with this subject...]

<Prev in Thread]	Current Thread	[Next in Thread>
[REVIEW] Invalidate dentry on unlink/rmdir if using CI mode, Barry Naujok Re: [REVIEW] Invalidate dentry on unlink/rmdir if using CI mode, Christoph Hellwig <=

Previous by Date:	[PATCH] Prevent extent btree block allocation failures, Lachlan McIlroy
Next by Date:	Re: [PATCH] fix extent corruption in xfs_iext_irec_compact_full(), Christoph Hellwig
Previous by Thread:	[REVIEW] Invalidate dentry on unlink/rmdir if using CI mode, Barry Naujok
Next by Thread:	[PATCH] fix extent corruption in xfs_iext_irec_compact_full(), Lachlan McIlroy
Indexes:	[Date] [Thread] [Top] [All Lists]