On Mon, Nov 29, 2021 at 12:53:39PM +0530, Maulik Shah wrote:
> Hi Shawn,
Hi Maulik,
Thanks for the comment!
>
> On 11/26/2021 3:05 PM, Shawn Guo wrote:
> > +static int __maybe_unused qcom_mpm_suspend_late(struct device *dev)
>
> why maybe unused?
>
> > +{
> > + struct qcom_mpm_priv *priv = dev_get_drvdata(dev);
> > + int i, ret;
> > +
> > + for (i = 0; i < priv->reg_stride; i++)
> > + qcom_mpm_write(priv, MPM_REG_STATUS, i, 0);
> > +
> > + /* Notify RPM to write vMPM into HW */
> > + ret = mbox_send_message(priv->mbox_chan, NULL);
> > + if (ret < 0)
> > + return ret;
> > +
> > + return 0;
> > +}
> > +
> > +static int __maybe_unused qcom_mpm_resume_early(struct device *dev)
> > +{
> > + /* Nothing to do for now */
> > + return 0;
> > +}
> > +
> > +static const struct dev_pm_ops qcom_mpm_pm_ops = {
> > + SET_LATE_SYSTEM_SLEEP_PM_OPS(qcom_mpm_suspend_late,
> > + qcom_mpm_resume_early)
> > +};
>
> This is not limited to suspend, you will need to notify RPM during deepest
> cpu idle state entry as well, since MPM may be monitoring interrupts in that
> case too.
Yeah, I was trying to test this MPM driver with cpuidle, but failed to
see the SoC get into vlow/vmin state from cpuidle. Do you have any
suggestion how I should test it properly?
> This may be handled for both suspend/CPUidle using cpu pm notifications
> where in last cpu entering deepest idle may notify RPM (something similar to
> what rpmh-rsc.c does)
Thanks for the pointer! I will take a look.
Shawn
