Re: [PATCH 4/5] phy: qcom: snps-femto-v2: Move PHY init registers to config tables

From: Varadarajan Narayanan <varadarajan.narayanan@oss.qualcomm.com>
To: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com>
Cc: Vinod Koul <vkoul@kernel.org>,
	Neil Armstrong <neil.armstrong@linaro.org>,
	Manivannan Sadhasivam <mani@kernel.org>,
	Rob Herring <robh@kernel.org>,
	Krzysztof Kozlowski <krzk+dt@kernel.org>,
	Conor Dooley <conor+dt@kernel.org>,
	Wesley Cheng <quic_wcheng@quicinc.com>,
	linux-arm-msm@vger.kernel.org, linux-phy@lists.infradead.org,
	devicetree@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH 4/5] phy: qcom: snps-femto-v2: Move PHY init registers to config tables
Date: Thu, 27 Aug 2026 09:59:08 +0530	[thread overview]
Message-ID: <ao+9FOzw6d+hjDyB@hu-varada-blr.qualcomm.com> (raw)
In-Reply-To: <d3d075ac-598a-4b79-8d59-556657edb23b@oss.qualcomm.com>

On Tue, Aug 25, 2026 at 10:54:58AM +0200, Konrad Dybcio wrote:
> On 8/24/26 8:37 AM, Varadarajan Narayanan wrote:
> > The driver currently programs the common PHY register sequence directly in
> > qcom_snps_hsphy_init(), with the fixed register writes split around the
> > override parameter handling. This makes it hard to support SoCs that need a
> > different sequence or different values.
> >
> > Move the fixed pre- and post-override register programming into per-
> > compatible configuration tables, and keep the 7 nm override parameters in
> > the same match-data structure. This allows the driver to share the common
> > sequence while adding SoC-specific sequences for future platforms such as
> > IPQ9650.
> >
> > Signed-off-by: Varadarajan Narayanan <varadarajan.narayanan@oss.qualcomm.com>
> > ---
>
> [...]
>
> > +	for (tmp = data->pre; tmp->offset != UINT_MAX; tmp++)
>
> "pre"/"post" alone are not descriptive - perhaps pre_tuning/post_tuning?

Will change.

> Or do we maybe know whether the order matters at all?

I believe so. This seems to follow the order given in the hardware
programming guide.

> > +		qcom_snps_hsphy_write_mask(hsphy->base, tmp->offset, tmp->mask, tmp->val);
>
> Custom termination is both error-prone and memory-inefficient, add
> another int to size the arrays

Ok.

Thanks
Varada

-- 
linux-phy mailing list
linux-phy@lists.infradead.org
https://lists.infradead.org/mailman/listinfo/linux-phy