Re: [PATCH v2 02/13] arm64: dts: ti: k3-am642-phyboard-electra-rdk: fix USB clocking for compliance

[Siddharth Vadapalli]

On 07/05/26 13:55, Wadim Egorov wrote:
> Hi,
>
> On 5/6/26 5:09 PM, Siddharth Vadapalli wrote:
> > According to section "6.5.3 Normative Spread Spectrum Clocking (SSC)" of
> > the USB 3.2 Specification, SSC should be enabled by default. This protects
> > against EMI violations. Hence, enable internal SSC for USB SuperSpeed.
> >
> > Fixes: c48ac0efe6d7 ("arm64: dts: ti: Add support for phyBOARD-Electra-AM642")
> > Cc: <stable@xxxxxxxxxxxxxxx>
> > Signed-off-by: Siddharth Vadapalli <s-vadapalli@xxxxxx>
> > ---
> >
> > v1:
> > https://lore.kernel.org/r/20260505110631.1144200-3-s-vadapalli@xxxxxx/
> > No changes since v1.
> >
> >   arch/arm64/boot/dts/ti/k3-am642-phyboard-electra-rdk.dts | 9 +++++++++
> >   1 file changed, 9 insertions(+)
> >
> > diff --git a/arch/arm64/boot/dts/ti/k3-am642-phyboard-electra-rdk.dts b/arch/arm64/boot/dts/ti/k3-am642-phyboard-electra-rdk.dts
> > index 793538f94942..a85d7d08bd1b 100644
> > --- a/arch/arm64/boot/dts/ti/k3-am642-phyboard-electra-rdk.dts
> > +++ b/arch/arm64/boot/dts/ti/k3-am642-phyboard-electra-rdk.dts
> > @@ -439,12 +439,21 @@ &sdhci1 {
> >   	status = "okay";
> >   };
> >   
> > +&serdes_wiz0 {
> > +	ti,core-clk-sel = <1>;  /* Select internal reference clock */
> > +	ti,ssc-enable; /* Enable SSC */
> > +	ti,ssc-type = <1>; /* 1 for Downspread */
> > +	ti,ssc-frequency-hz = <33000>; /* 33 KHz */
> > +	ti,ssc-depth-per-mil = <5>; /* 0.5% depth */
>
> I don't think the comments are very helpful. The property names already give a meaning.

The comments have been added for three reasons:
1. The meaning of the following properties isn't obvious:
	ti,core-clk-sel = <1>
	ti,ssc-type = <1>
2. For ease of 'grepping'. Grepping for '33 KHz' for example based on the 
USB 3.2 Specification's modulation rate will not show '33000' in the results.
3. Completeness / Consistency. Since some of the less obvious properties 
have been described via comments, the remaining have also been commented 
on, although it is obvious what it means (ti,ssc-enable for example).

Unless you have a strong objection to removing the comments, I would prefer 
retaining them. Please let me know.

Regards,
Siddharth.