linux-kernel - Re: [patch] aio: invalidate async directio writes

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20080618194823.GA7237@infradead.org>
Date:	Wed, 18 Jun 2008 15:48:23 -0400
From:	Christoph Hellwig <hch@...radead.org>
To:	Jeff Moyer <jmoyer@...hat.com>
Cc:	Christoph Hellwig <hch@...radead.org>, akpm@...ux-foundation.org,
	zach.brown@...cle.com, linux-aio@...ck.org,
	linux-kernel@...r.kernel.org
Subject: Re: [patch] aio: invalidate async directio writes

On Wed, Jun 18, 2008 at 03:45:28PM -0400, Jeff Moyer wrote:
> > Can we please move all these aio_complete calls to user context?  Having
> > AIO contexts completing from irq context is a major pain for complex
> > filesystems like XFS.
> 
> Can you help me understand why this is a pain?  I'm having trouble
> making the connection.

With your patch we complete aio dio write request in user context, which
is great for filesystems that need to do more complex activity in the
completion handler, e.g. XFS for the unwritten extent conversion.  But
only doing this for the write case is only very partially useful, we
should be doing this for the read case, too.

See fs/xfs/linux-2.6/xfs_aops.c:xfs_end_io_direct() for what I mean.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/