Re: SDHCI: timeout during data transfer

["Luca Tettamanti" <kronos.it@gmail.com>]

On Tue, Oct 14, 2008 at 11:34 AM, Luca Tettamanti <kronos.it@gmail.com> wrote:
> On Sun, Oct 12, 2008 at 10:52 AM, Pierre Ossman <drzeus-mmc@drzeus.cx> wrote:
>> On Fri, 3 Oct 2008 13:27:52 +0200
>> "Luca Tettamanti" <kronos.it@gmail.com> wrote:
>>>
>>> Hum, cannot reproduce (but it was consistently failing when I tested
>>> the patch... the only difference is a mkfs in between). I just got a
>>> few retries:
>>
>> A silly question, but did you also try disabling the debugging? Since
>> this is a timing issue, the debugging output could be just enough to
>> make the problem go away.
>
> Ok, I managed to reproduce without the debugging, but this time it
> took over a couple of GB before it started failing. Block numbers
> differ from the previous failures (so maybe it's not a cluster of
> broken blocks).
> Will go back and test with debugging enabled.

I finally managed to capture a failure with debug enabled, I'm
attaching the log.
To recap:
- heavy write loads sometimes result in a timeout
- reads seem unaffected (I read the card multiple times without errors)
- doubling the timeout (as per my first patch) makes the timeout disappear
- windows doesn't suffer from the issue

I've tested two other cards that do not show the problem, but they are
not "high speed":

mmc0: new high speed SD card at address 0002 (troubles)
vs:
mmc0: new SD card at address e624 (OK)

Luca