Re: [PATCH #upstream] sata_sis: convert to slave_link

[Jeff Garzik <jeff@garzik.org>]

On 09/01/2009 10:19 AM, Tejun Heo wrote:
> During introduction of slave_link, sata_sis slipped through the crack
> and left with ad-hoc merged SCR access.  As SCR status was shared for
> both the master and slave devices, when only one of the device is
> online, libata EH would think both are online but would only get valid
> device signature for the actually present one, which in turn trigger
> the probing safety net mechanism and make EH retry causing large delay
> during boot.  This patch converts sata_sis to slave_link mechanism.
>
> This bug was reported by TAXI in bko#14075.
>
>   http://bugzilla.kernel.org/show_bug.cgi?id=14075
>
> Signed-off-by: Tejun Heo<tj@kernel.org>
> Reported-by: TAXI<taxi@a-city.de>
> Cc: Uwe Koziolek<uwe.koziolek@gmx.net>
> ---
> This is fairly straight-forward fix but I couldn't test it extensively
> and it's way too late in the rc cycle.  I think it would be best to
> put it in #upstream and let it see some testing in linux-next and then
> backport them through -stable.
>
> Thanks.
>
>   drivers/ata/sata_sis.c |   75 ++++++++++++++++---------------------------------
>   1 file changed, 25 insertions(+), 50 deletions(-)

applied