Re: [PATCH v2 1/2] phy: qcom: pcie: Determine has_nocsr_reset dynamically

[Wenbin Yao (Consultant)]

On 2/12/2025 8:31 PM, Konrad Dybcio wrote:
> On 11.02.2025 10:53 AM, Philipp Zabel wrote:
> > On Di, 2025-02-11 at 17:42 +0800, Wenbin Yao wrote:
> > > From: Konrad Dybcio <konrad.dybcio@xxxxxxxxxxxxxxxx>
> > >
> > > Decide the in-driver logic based on whether the nocsr reset is present
> > > and defer checking the appropriateness of that to dt-bindings to save
> > > on boilerplate.
> > >
> > > Reset controller APIs are fine consuming a nullptr, so no additional
> > > checks are necessary there.
> > >
> > > Signed-off-by: Konrad Dybcio <konrad.dybcio@xxxxxxxxxxxxxxxx>
> > > Signed-off-by: Wenbin Yao <quic_wenbyao@xxxxxxxxxxx>
> > > Reviewed-by: Abel Vesa <abel.vesa@xxxxxxxxxx>
> > > Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@xxxxxxxxxx>
> > > ---
> [...]
>
> > >   static void qmp_pcie_init_port_b(struct qmp_pcie *qmp, const struct qmp_phy_cfg_tbls *tbls)
> > > @@ -4203,11 +4196,14 @@ static int qmp_pcie_reset_init(struct qmp_pcie *qmp)
> > >   	if (ret)
> > >   		return dev_err_probe(dev, ret, "failed to get resets\n");
> > >   
> > > -	if (cfg->has_nocsr_reset) {
> > > -		qmp->nocsr_reset = devm_reset_control_get_exclusive(dev, "phy_nocsr");
> > > -		if (IS_ERR(qmp->nocsr_reset))
> > > +	qmp->nocsr_reset = devm_reset_control_get_exclusive(dev, "phy_nocsr");
> > > +	if (IS_ERR(qmp->nocsr_reset)) {
> > > +		if (PTR_ERR(qmp->nocsr_reset) == -ENOENT ||
> > > +		    PTR_ERR(qmp->nocsr_reset) == -EINVAL)
> > Why is -EINVAL ignored here?
> If the NOCSR (partial) reset is missing, we can still assert the "full" reset
> and program the hardware from the ground up. It's also needed for backwards
> dt compat as not all platforms described it when originally added.

Seems like we really can't ignore -EINVAL. If no_csr reset is missing in
dts, it will return -ENOENT, which is turned into NULL by
devm_reset_control_get_optional_exclusive. -EINVAL indicates something
wrong in args we passed to the function and the reset property that need to
be fixed.

> > Without this you could just use
> > devm_reset_control_get_optional_exclusive(), which already turns -
> > ENOENT into NULL. That seems to me the correct thing to do, as from
> > driver point-of-view, this reset control is optional.
> Good point, I forgot _optional_ was a thing in the reset framework
>
> Konrad

Will use devm_reset_control_get_optional_exclusive function in patch v3.

--
With best wishes
Wenbin