Re: [PATCH] mmc: sprd: Fix vqmmc not shutting down after the card was pulled

[Ulf Hansson]

On Thu, 9 Nov 2023 at 07:37, Wenchao Chen <wenchao.chen@xxxxxxxxxx> wrote:
>
> With cat regulator_summary, we found that vqmmc was not shutting
> down after the card was pulled.
>
> cat /sys/kernel/debug/regulator/regulator_summary
> 1.before fix
> 1)Insert SD card
>  vddsdio                1    1  0 unknown  3500mV 0mA  1200mV  3750mV
>     71100000.mmc-vqmmc  1                         0mA  3500mV  3600mV
>
> 2)Pull out the SD card
>  vddsdio                1    1  0 unknown  3500mV 0mA  1200mV  3750mV
>     71100000.mmc-vqmmc  1                         0mA  3500mV  3600mV
>
> 2.after fix
> 1)Insert SD cardt
>  vddsdio                1    1  0 unknown  3500mV 0mA  1200mV  3750mV
>     71100000.mmc-vqmmc  1                         0mA  3500mV  3600mV
>
> 2)Pull out the SD card
>  vddsdio                0    1  0 unknown  3500mV 0mA  1200mV  3750mV
>     71100000.mmc-vqmmc  0                         0mA  3500mV  3600mV
>
> Fixes: fb8bd90f83c4 ("mmc: sdhci-sprd: Add Spreadtrum's initial host controller")
> Signed-off-by: Wenchao Chen <wenchao.chen@xxxxxxxxxx>
> ---
>  drivers/mmc/host/sdhci-sprd.c | 62 +++++++++++++++++++++++++++++++++++
>  1 file changed, 62 insertions(+)
>
> diff --git a/drivers/mmc/host/sdhci-sprd.c b/drivers/mmc/host/sdhci-sprd.c
> index 6b84ba27e6ab..b5f22e48e571 100644
> --- a/drivers/mmc/host/sdhci-sprd.c
> +++ b/drivers/mmc/host/sdhci-sprd.c
> @@ -90,6 +90,7 @@ struct sdhci_sprd_host {
>         u32 base_rate;
>         int flags; /* backup of host attribute */
>         u32 phy_delay[MMC_TIMING_MMC_HS400 + 2];
> +       u8 power_mode;
>  };
>
>  enum sdhci_sprd_tuning_type {
> @@ -416,12 +417,71 @@ static void sdhci_sprd_request_done(struct sdhci_host *host,
>         mmc_request_done(host->mmc, mrq);
>  }
>
> +static void sdhci_sprd_signal_voltage_switch(struct sdhci_host *host, bool enable)
> +{
> +       const char *name = mmc_hostname(host->mmc);
> +       bool on;
> +
> +       if (IS_ERR(host->mmc->supply.vqmmc))
> +               return;
> +
> +       on = regulator_is_enabled(host->mmc->supply.vqmmc);
> +       if (!(on ^ enable))
> +               return;
> +
> +       if (enable) {
> +               if (regulator_enable(host->mmc->supply.vqmmc))
> +                       pr_err("%s: signal voltage enable fail!\n", name);
> +               else if (regulator_is_enabled(host->mmc->supply.vqmmc))
> +                       pr_debug("%s: signal voltage enable success!\n", name);
> +               else
> +                       pr_err("%s: signal voltage enable hw fail!\n", name);
> +       } else {
> +               if (regulator_disable(host->mmc->supply.vqmmc))
> +                       pr_err("%s: signal voltage disable fail\n", name);
> +               else if (!regulator_is_enabled(host->mmc->supply.vqmmc))
> +                       pr_debug("%s: signal voltage disable success!\n", name);
> +               else
> +                       pr_err("%s: signal voltage disable hw fail\n", name);
> +       }

This looks rather noisy to me.

How about converting into using mmc_regulator_enable|disable_vqmmc()
instead - and then simply drop the entire function above.

> +}
> +
> +static void sdhci_sprd_set_power(struct sdhci_host *host, unsigned char mode,
> +                                unsigned short vdd)
> +{
> +       struct sdhci_sprd_host *sprd_host = TO_SPRD_HOST(host);
> +       struct mmc_host *mmc = host->mmc;
> +
> +       if (sprd_host->power_mode == mmc->ios.power_mode)
> +               return;
> +
> +       switch (mode) {
> +       case MMC_POWER_OFF:
> +               sdhci_sprd_signal_voltage_switch(host, false);
> +               if (!IS_ERR(mmc->supply.vmmc))
> +                       mmc_regulator_set_ocr(host->mmc, mmc->supply.vmmc, 0);
> +               break;
> +       case MMC_POWER_ON:
> +       case MMC_POWER_UP:

This means that a usleep_range() below is being executed for both
MMC_POWER_ON and MMC_POWER_UP, even if we actually didn't change the
regulators. I assume that isn't needed and nor the intent, right?

> +               if (!IS_ERR(mmc->supply.vmmc))
> +                       mmc_regulator_set_ocr(host->mmc, mmc->supply.vmmc, vdd);
> +
> +               /* waiting for mmc->supply.vmmc to stabilize */
> +               usleep_range(200, 250);
> +               sdhci_sprd_signal_voltage_switch(host, true);
> +               break;
> +       }
> +
> +       sprd_host->power_mode = mmc->ios.power_mode;

If we convert to use the helper functions
mmc_regulator_enable|disable_vqmmc(), you shouldn't need to use
sprd_host->power_mode, as that is already taken care of by the
helpers.

> +}
> +
>  static struct sdhci_ops sdhci_sprd_ops = {
>         .read_l = sdhci_sprd_readl,
>         .write_l = sdhci_sprd_writel,
>         .write_w = sdhci_sprd_writew,
>         .write_b = sdhci_sprd_writeb,
>         .set_clock = sdhci_sprd_set_clock,
> +       .set_power = sdhci_sprd_set_power,
>         .get_max_clock = sdhci_sprd_get_max_clock,
>         .get_min_clock = sdhci_sprd_get_min_clock,
>         .set_bus_width = sdhci_set_bus_width,
> @@ -805,6 +865,8 @@ static int sdhci_sprd_probe(struct platform_device *pdev)
>         sprd_host->version = ((host->version & SDHCI_VENDOR_VER_MASK) >>
>                                SDHCI_VENDOR_VER_SHIFT);
>
> +       sprd_host->power_mode = MMC_POWER_OFF;
> +
>         pm_runtime_get_noresume(&pdev->dev);
>         pm_runtime_set_active(&pdev->dev);
>         pm_runtime_enable(&pdev->dev);

Kind regards
Uffe