On Fri, Nov 15, 2019 at 4:08 AM Stephen Boyd <swboyd@xxxxxxxxxxxx> wrote:
>
> Quoting Amit Kucheria (2019-11-11 11:21:28)
> > TSENS IP v2.3 onwards adds support for a watchdog to detect if the TSENS
> > HW FSM is frozen. Add support to detect and restart the FSM in the
>
> Maybe 'frozen' is an ambiguous term? Maybe 'stuck' or 'has stopped
> making progress'?
Alright, let's keep Disney out of this. Will use 'stuck'.
> > driver. The watchdog is configured by the bootloader, we just enable the
> > feature in the kernel.
>
> Does it work to enable it if we don't configure it in the bootloader?
TBH, I don't know. Getting modified firmware to test this will be a
bit of a challenge.
> >
> > Signed-off-by: Amit Kucheria <amit.kucheria@xxxxxxxxxx>
> > ---
> > drivers/thermal/qcom/tsens-common.c | 41 +++++++++++++++++++++++++++++
> > drivers/thermal/qcom/tsens-v2.c | 10 +++++++
> > drivers/thermal/qcom/tsens.h | 12 +++++++++
> > 3 files changed, 63 insertions(+)
> >
> > diff --git a/drivers/thermal/qcom/tsens-common.c b/drivers/thermal/qcom/tsens-common.c
> > index 2989cb952cdb..9432518502a7 100644
> > --- a/drivers/thermal/qcom/tsens-common.c
> > +++ b/drivers/thermal/qcom/tsens-common.c
> > @@ -794,6 +820,21 @@ int __init init_common(struct tsens_priv *priv)
> > }
> > }
> >
> > + if (tsens_version(priv) > VER_1_X && ver_minor > 2) {
> > + /* Watchdog is present only on v2.3+ */
> > + for (i = 0, j = WDOG_BARK_STATUS; j <= CC_MON_MASK; i++, j++) {
>
> The variable 'i' is not actually used in this loop. What's going on?
Sorry, left over from a botched copy-paste job from the loop above
this. Will fix.
>
> > + priv->rf[j] = devm_regmap_field_alloc(dev, priv->tm_map,
> > + priv->fields[j]);
> > + if (IS_ERR(priv->rf[j])) {
> > + ret = PTR_ERR(priv->rf[j]);
> > + goto err_put_device;
> > + }
> > + }
> > + /* Enable WDOG and disable cycle completion monitoring */
> > + regmap_field_write(priv->rf[WDOG_BARK_MASK], 0);
> > + regmap_field_write(priv->rf[CC_MON_MASK], 1);
> > + }
