[PATCH 1/2] mmc: core: use card pointer as the first parameter of execute_tuning()

Doug Anderson dianders at chromium.org
Fri Jan 30 17:08:36 PST 2015


On Thu, Jan 29, 2015 at 4:13 PM, Doug Anderson <dianders at chromium.org> wrote:
> I still don't exactly have a warm fuzzy about using the send_status()
> command like this, but it seems to work (actually, I should verify
> that myself--I've been taking Addy's word that his solution works).

So I finally started up a stress test.  I picked Addy's patch and
reverted my big fat delay.  Result was that Addy's patch doesn't seem
to hurt but once I revert my big fat delay then I seem to start
getting failures.  :(  I think it's actually on a different SD card
than the one that was originally causing failures, FWIW.

As usual with these kinds of tests, I'd like to test this a whole lot
more since it only reproduces sometimes, but unfortunately I'm going
to be unavailable for a few weeks starting from today, so I'm not sure
I'm not sure I can do more debugging / testing than what I've already

...but unless using send_status allows us to remove the delay (or
unless there's some other justification for it) then it seems like we
should just drop this series...

More information about the Linux-rockchip mailing list