Re: [PATCH V12 12/23] mmc: sdhci-uhs2: skip signal_voltage_switch()

[Victor Shih]

On Tue, Oct 3, 2023 at 5:58 PM Ulf Hansson <ulf.hansson@xxxxxxxxxx> wrote:
>
> On Fri, 15 Sept 2023 at 11:44, Victor Shih <victorshihgli@xxxxxxxxx> wrote:
> >
> > From: Victor Shih <victor.shih@xxxxxxxxxxxxxxxxxxx>
> >
> > For UHS2, the signal voltage is supplied by vdd2 which is already 1.8v,
> > so no voltage switch required.
>
> Can you please elaborate on this? I don't get anything of the above, sorry.
>
> >
> > Signed-off-by: Ben Chuang <ben.chuang@xxxxxxxxxxxxxxxxxxx>
> > Signed-off-by: AKASHI Takahiro <takahiro.akashi@xxxxxxxxxx>
> > Signed-off-by: Victor Shih <victor.shih@xxxxxxxxxxxxxxxxxxx>
> > Acked-by: Adrian Hunter <adrian.hunter@xxxxxxxxx>
> > ---
> >
> > Updates in V5:
> >  - Use sdhci_uhs2_mode() to simplify code in
> >    sdhci_uhs2_start_signal_voltage_switch().
> >
> > ---
> >
> >  drivers/mmc/host/sdhci-uhs2.c | 24 ++++++++++++++++++++++++
> >  1 file changed, 24 insertions(+)
> >
> > diff --git a/drivers/mmc/host/sdhci-uhs2.c b/drivers/mmc/host/sdhci-uhs2.c
> > index fc37a34629c2..92fb69b7e209 100644
> > --- a/drivers/mmc/host/sdhci-uhs2.c
> > +++ b/drivers/mmc/host/sdhci-uhs2.c
> > @@ -142,6 +142,27 @@ static void sdhci_uhs2_set_power(struct sdhci_host *host, unsigned char mode, un
> >         }
> >  }
> >
> > +/*****************************************************************************\
> > + *                                                                           *
> > + * MMC callbacks                                                             *
> > + *                                                                           *
> > +\*****************************************************************************/
> > +
> > +static int sdhci_uhs2_start_signal_voltage_switch(struct mmc_host *mmc,
> > +                                                 struct mmc_ios *ios)
> > +{
> > +       struct sdhci_host *host = mmc_priv(mmc);
> > +
> > +       /*
> > +        * For UHS2, the signal voltage is supplied by vdd2 which is
> > +        * already 1.8v so no voltage switch required.
> > +        */
> > +       if (sdhci_uhs2_mode(host))
> > +               return 0;
>
> This is just wrong. If we are initializing a uhs2 card, we certainly
> should call ->start_signal_voltage_switch() callback at all. This is
> for UHS-I cards, right?
>

Hi, Ulf

UHS-II does not need single_voltage.
I will modify the commit message in the next version.
sdhci_uhs2_start_signal_voltage_switch() is under
mmc_host_ops.start_signal_voltage_switch host ops,
therefore, we need to keep the UHS-I path here.

Thanks, Victor Shih

> > +
> > +       return sdhci_start_signal_voltage_switch(mmc, ios);
> > +}
> > +
> >  /*****************************************************************************\
> >   *                                                                           *
> >   * Driver init/exit                                                          *
> > @@ -150,6 +171,9 @@ static void sdhci_uhs2_set_power(struct sdhci_host *host, unsigned char mode, un
> >
> >  static int sdhci_uhs2_host_ops_init(struct sdhci_host *host)
> >  {
> > +       host->mmc_host_ops.start_signal_voltage_switch =
> > +               sdhci_uhs2_start_signal_voltage_switch;
> > +
> >         return 0;
> >  }
> >
>
> Kind regards
> Uffe