linux-kernel - Re: [REPOST PATCH] sata_promise: use TF interface for polling NODATA commands

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] [day] [month] [year] [list]

Message-ID: <20070607173027.GC11882@havoc.gtf.org>
Date:	Thu, 7 Jun 2007 13:30:27 -0400
From:	Jeff Garzik <jeff@...zik.org>
To:	Chuck Ebbert <cebbert@...hat.com>
Cc:	Tejun Heo <htejun@...il.com>, Mikael Pettersson <mikpe@...uu.se>,
	akpm@...ux-foundation.org, david@...eaves.com,
	jean.luc.coulon@...il.com, jgarzik@...ox.com,
	michal.k.k.piotrowski@...il.com, torvalds@...ux-foundation.org,
	linux-kernel@...r.kernel.org
Subject: Re: [REPOST PATCH] sata_promise: use TF interface for polling NODATA commands

On Thu, Jun 07, 2007 at 01:25:07PM -0400, Jeff Garzik wrote:
> On Thu, Jun 07, 2007 at 01:20:13PM -0400, Chuck Ebbert wrote:
> > On 06/06/2007 12:05 PM, Jeff Garzik wrote:
> > > FYI to all -- 
> > > 
> > > As a reminder. the Promise hardware programs registers when it receives
> > > a SET FEATURES - XFER MODE.
> > > 
> > > If data transfer is occurring on OTHER ports at the time this is issued,
> > > then data corruption is guaranteed to occur.  Polling will not fix this
> > > problem -- all ports need to be inactive, when a SET FEATURES - XFER
> > > MODE command is issued for any port.
> 
> > So is this patch OK but yet more work needs to be done, or does
> > this patch cause new problems?
> 
> Causes no /new/ problems...  :)  The existing problem described above
> remains.

Just to expand...  this problem doesn't really affect a lot of users in
the majority case, since we do speed tuning before data transfer starts.

The main problem that remains is the rare (but no less important) case
where libata-EH will tune speed during operation in respond to certain
classes of errors.

	Jeff



-
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/