Date: Fri, 18 May 2007 16:12:28 +0100
From: Alan Cox <alan@lxorguk.ukuu.org.uk>
To: Jeff Garzik <jeff@garzik.org>
Cc: Robert Hancock <hancockr@shaw.ca>, Zoltan Boszormenyi <zboszor@dunaweb.hu>,
       Peer Chen <pchen@nvidia.com>,
       linux-kernel <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] drivers/ata: Add the SW NCQ support to sata_nv for
 MCP51/MCP55/MCP61
Message-ID: <20070518161228.7e89b6dc@the-village.bc.nu>
In-Reply-To: <464DB97B.8030204@garzik.org>
References: <fa.iernIz8Xcuvw0SBwi4A99Ju0g2o@ifi.uio.no>
	<464D0855.20703@shaw.ca>
	<20070518133456.5f1392e9@the-village.bc.nu>
	<464DB97B.8030204@garzik.org>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 1182
Lines: 30

On Fri, 18 May 2007 10:34:35 -0400
Jeff Garzik <jeff@garzik.org> wrote:

> Alan Cox wrote:
> >> That shouldn't be a problem, libata default DMA mask is 32 bits (which 
> >> isn't overridden with this controller) and so the block layer will 
> >> bounce any data being read/written above that point with IOMMU or 
> >> swiotlb. The comment is a bit unnecessarily scary.
> > 
> > Adding a BUG_ON for this would be wise. Its trivial to check and a BUG
> > rather than corruption if this assumption ever changes would be far
> > preferable
> 
> The default DMA mask -everywhere- is 32 bits.
> 
> A lot of code will break if this assumption ever changes, not just libata.

Little lesson from history..

Over ten years ago someone (Eric Youngdale I guess) stuck a panic check
for DMA over 16MBytes in the AHA 1542 ISA SCSI driver. Last year it
triggered. The panic probably saved someone from corruption and meant the
bug could be fixed.

Alan
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/