Re: [PATCH 1/1]mmc: set timeout for SDHCI host before sending busycmd

From: Wolfram Sang
Date: Thu Feb 24 2011 - 15:44:10 EST

On Thu, Feb 24, 2011 at 08:15:04PM +0000, Chris Ball wrote:
> On Thu, Feb 24, 2011 at 10:32:30AM -0800, Philip Rakity wrote:
> > proposed this a while ago and strongly support just removing the quirk for broken timeout and setting the timeout value to maximum of 0xE.
> >
> > This also handles the case of the sd device having a timeout value too low. In my testing I have come across SD cards that do not provide the correct value.
> > We force our pxa168, pxa910, and mmp2 controllers to have 0xE.
> Yeah, OLPC's CaFe controller -- which might be the same hardware as yours,
> actually -- has the same problem.
> Does anyone know of a reason (beyond strict spec-compliance, I suppose)
> for honoring the timeout value rather than using 0xE everywhere? If not,
> I'm willing to try out Philip's suggestion.

+1. A full cycle in linux-next might an idea to be on the safe side? That would
be 2.6.40-material then. Or too slow?


Pengutronix e.K. | Wolfram Sang |
Industrial Linux Solutions | |

Attachment: signature.asc
Description: Digital signature