Re: [PATCH] mmc: Allow non-sleeping GPIO cd

[Evan Green]

On Tue, May 22, 2018 at 5:08 AM Ulf Hansson <ulf.hansson@xxxxxxxxxx> wrote:

> First, I thought I preferred this option, as it becomes clear of what
> goes on. However I then realize, that it may not be worth it, because
> in the end I guess the caller (sdhci), will not be able to deal with
> error codes. For example, what would it do if it receives  a -ENOTSUPP
> from mmc_gpio_get_cd_nosleep()?


Uffe,
Yeah, that would end up looking more like my original RFC patch, where we
just kind of assume the card is there (and therefore ignore the quirk) if
we're not in a preemptible state. Sounds like I should make another spin
this way, at least for consideration.

Andy,
Thanks for the review. I will incorporate your feedback into the next spin.
-Evan