2.6.35.6 fails to suspend (pxa2xx-mci.0)

[Sven Neumann]

On Sat, 2010-10-09 at 03:07 +0200, Ohad Ben-Cohen wrote:
> On Fri, Oct 8, 2010 at 10:08 PM, Rafael J. Wysocki <rjw at sisk.pl> wrote:
> > There is a problem with the pxa2xx-mci.0 suspend that has been introduced
> > some time after 2.6.34, but it is not related to commit 152e1d5920.
> 
> Sven, what type of card do you have in that slot ? is it SDIO ? what's
> the function driver ?
> 
> I have a hunch about this, but it may be a wild guess as I don't know
> what's your setup exactly.
> 
> Anyway, can you please try this and see if anything changes ?
> 
> diff --git a/drivers/mmc/core/core.c b/drivers/mmc/core/core.c
> index c94565d..515ff39 100644
> --- a/drivers/mmc/core/core.c
> +++ b/drivers/mmc/core/core.c
> @@ -1682,6 +1682,19 @@ int mmc_suspend_host(struct mmc_host *host)
>         if (host->bus_ops && !host->bus_dead) {
>                 if (host->bus_ops->suspend)
>                         err = host->bus_ops->suspend(host);
> +               if (err == -ENOSYS || !host->bus_ops->resume) {
> +                       /*
> +                        * We simply "remove" the card in this case.
> +                        * It will be redetected on resume.
> +                        */
> +                       if (host->bus_ops->remove)
> +                               host->bus_ops->remove(host);
> +                       mmc_claim_host(host);
> +                       mmc_detach_bus(host);
> +                       mmc_release_host(host);
> +                       host->pm_flags = 0;
> +                       err = 0;
> +               }
>         }
>         mmc_bus_put(host);
> 
> The reason I'm asking this is because it seems like commit
> 4c2ef25fe0b847d2ae818f74758ddb0be1c27d8e "mmc: fix all hangs related
> to mmc/sd card insert/removal during suspend/resume" has broken
> suspending SDIO function drivers which don't have a suspend handler
> (or do have one, but for some reason it returns -ENOSYS, like
> libertas_sdio's does).

Yes, this patch fixes the problem. Tested with v2.6.35-rc7. With above
patch applied, suspend (and resume!) work again with and without
pm_async=0.

Thanks a lot to all of you for your help with this problem.


Regards,
Sven