| Message ID | 20190618153448.27145-6-ulf.hansson@linaro.org (mailing list archive) |
|---|---|
| State | Not Applicable |
| Delegated to: | Johannes Berg |
| Headers | show |
| Series | mmc: sdio: Various fixes/improvements for SDIO PM | expand |
Hi, On Tue, Jun 18, 2019 at 8:35 AM Ulf Hansson <ulf.hansson@linaro.org> wrote: > > It looks like the original idea behind always doing a re-initialization of > a removable SDIO card during system resume in mmc_sdio_resume(), is to try > to play safe to detect whether the card has been removed. > > However, this seems like a really a bad idea as it will most likely screw > things up, especially when the card is expected to remain powered on during > system suspend by the SDIO func driver. > > Let's fix this, simply by trusting that the detect work checks if the card > is alive and inserted, which is being scheduled at the PM_POST_SUSPEND > notification anyway. > > Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org> > --- > drivers/mmc/core/sdio.c | 14 +++++++------- > 1 file changed, 7 insertions(+), 7 deletions(-) I'm not sure if it's even worth mentioning since the device I tested was rk3288-veyron (both WiFi variants) and we have a non-removable, powered-in-suspend, wakeup-disabled card. That means it isn't affected at all by your change. ...but I suppose I can at least confirm that your change didn't break me if that's worth anything. :-P Tested-by: Douglas Anderson <dianders@chromium.org> I would also say that, though I don't have any history here, your patch seems reasonable to me. So you can add a: Reviewed-by: Douglas Anderson <dianders@chromium.org>
diff --git a/drivers/mmc/core/sdio.c b/drivers/mmc/core/sdio.c index a9bfcae8db5b..945416c53b56 100644 --- a/drivers/mmc/core/sdio.c +++ b/drivers/mmc/core/sdio.c @@ -982,7 +982,11 @@ static int mmc_sdio_resume(struct mmc_host *host) /* Basic card reinitialization. */ mmc_claim_host(host); - /* Restore power if needed */ + /* + * Restore power and reinitialize the card when needed. Note that a + * removable card is checked from a detect work later on in the resume + * process. + */ if (!mmc_card_keep_power(host)) { mmc_power_up(host, host->card->ocr); /* @@ -996,12 +1000,8 @@ static int mmc_sdio_resume(struct mmc_host *host) pm_runtime_set_active(&host->card->dev); pm_runtime_enable(&host->card->dev); } - } - - /* No need to reinitialize powered-resumed nonremovable cards */ - if (mmc_card_is_removable(host) || !mmc_card_keep_power(host)) { - err = mmc_sdio_reinit_card(host, mmc_card_keep_power(host)); - } else if (mmc_card_keep_power(host) && mmc_card_wake_sdio_irq(host)) { + err = mmc_sdio_reinit_card(host, 0); + } else if (mmc_card_wake_sdio_irq(host)) { /* We may have switched to 1-bit mode during suspend */ err = sdio_enable_4bit_bus(host->card); }
It looks like the original idea behind always doing a re-initialization of a removable SDIO card during system resume in mmc_sdio_resume(), is to try to play safe to detect whether the card has been removed. However, this seems like a really a bad idea as it will most likely screw things up, especially when the card is expected to remain powered on during system suspend by the SDIO func driver. Let's fix this, simply by trusting that the detect work checks if the card is alive and inserted, which is being scheduled at the PM_POST_SUSPEND notification anyway. Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org> --- drivers/mmc/core/sdio.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-)