Re: [PATCH v5 3/3] mmc: host: meson-mx-sdhc: new driver for the Amlogic Meson SDHC host

[Martin Blumenstingl]

Hi Ulf,

thank you for looking into this!

On Mon, Apr 27, 2020 at 9:20 PM Ulf Hansson <ulf.hansson@xxxxxxxxxx> wrote:
[...]
> > +static void meson_mx_sdhc_wait_cmd_ready(struct mmc_host *mmc)
> > +{
> > +       struct meson_mx_sdhc_host *host = mmc_priv(mmc);
> > +       u32 stat, esta;
> > +       int ret;
> > +
> > +       ret = regmap_read_poll_timeout(host->regmap, MESON_SDHC_STAT, stat,
> > +                                      !(stat & MESON_SDHC_STAT_CMD_BUSY), 1,
> > +                                      100000);
>
> Please use defines for timeout values.
I'll take care of this here and all other places which you have found

[...]
> > +       if (cmd->data)
> > +               host->platform->set_pdma(mmc);
> > +
> > +       if (host->platform->wait_before_send)
> > +               host->platform->wait_before_send(mmc);
> > +
> > +       regmap_write(host->regmap, MESON_SDHC_SEND, send);
>
> Isn't there a configurable timeout to set for the command?
>
> I mean the driver sets mmc->max_busy_timeout to 30s in ->probe(), but
> can the timeout be configured to a lower value?
there's MESON_SDHC_CTRL_RX_TIMEOUT and MESON_SDHC_CTRL_RX_PERIOD
here's what the datasheet has to say about them:
- rx_timeout(cmd or wcrc Receiving Timeout, default 64)
- rc_period(Period between response/cmd and default next cmd,default
8) - I'm not even sure if this is related somehow

if you have a specific test-case for me to provoke these timeouts I
can try playing around with these values
otherwise we have to ask Jianxin and see whether he can get some
information about this from the internal team at Amlogic

[...]
> > +       mmc->caps |= MMC_CAP_ERASE | MMC_CAP_HW_RESET;
>
> Should you also set MMC_CAP_WAIT_WHILE_BUSY? It sounded like the
> driver supported this.
I can try setting it.
>From our previous discussion (on the meson-mx-sdio driver) I have
learned that eMMC will be a good test-case for it ;-)

[...]
> FYI: I left out all comments related to the clock provider
> initialization. I think it makes better sense to review that code,
> after you have converted to use the devm_clk_hw_register() and avoid
> registering a separate driver for it.
yes, that makes sense
I expect the code to be easier since it'll be one big driver with the
next version (so no more platform device allocation, etc.)

> Other than the minor comments, this looks good to me.
great - it would be great if this could finally make it into v5.8


Martin