Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Wed, 27 Feb 2019 10:09:52 -0800
From:   Bjorn Andersson <bjorn.andersson@linaro.org>
To:     Marc Gonzalez <marc.w.gonzalez@free.fr>
Cc:     Kishon Vijay Abraham I <kishon@ti.com>,
        Rob Herring <robh+dt@kernel.org>,
        Mark Rutland <mark.rutland@arm.com>,
        MSM <linux-arm-msm@vger.kernel.org>,
        LKML <linux-kernel@vger.kernel.org>,
        Can Guo <cang@codeaurora.org>,
        Vivek Gautam <vivek.gautam@codeaurora.org>,
        Jeffrey Hugo <jhugo@codeaurora.org>,
        Nicolas Dechesne <nicolas.dechesne@linaro.org>
Subject: Re: [PATCH] phy: qcom: qmp: Add SDM845 PCIe QMP PHY support
Message-ID: <20190227180952.GA1803@tuxbook-pro>
References: <20190226065919.22218-1-bjorn.andersson@linaro.org>
 <b17f2c12-fd2c-549e-92b9-96a422e5f07c@free.fr>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <b17f2c12-fd2c-549e-92b9-96a422e5f07c@free.fr>
User-Agent: Mutt/1.11.3 (2019-02-01)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On Wed 27 Feb 01:27 PST 2019, Marc Gonzalez wrote:

> On 26/02/2019 07:59, Bjorn Andersson wrote:
> 
> > qcom_qmp_phy_init() is extended to support the additional register
> > writes needed in PCS MISC and the appropriate sequences and resources
> > are defined for SDM845.
> > 
> > Signed-off-by: Bjorn Andersson <bjorn.andersson@linaro.org>
> > ---
> >  .../devicetree/bindings/phy/qcom-qmp-phy.txt  |   7 +
> >  drivers/phy/qualcomm/phy-qcom-qmp.c           | 160 ++++++++++++++++++
> >  drivers/phy/qualcomm/phy-qcom-qmp.h           |  12 ++
> >  3 files changed, 179 insertions(+)
> > 
> > diff --git a/Documentation/devicetree/bindings/phy/qcom-qmp-phy.txt b/Documentation/devicetree/bindings/phy/qcom-qmp-phy.txt
> > index 5d181fc3cc18..dd2725a9d3f7 100644
> > --- a/Documentation/devicetree/bindings/phy/qcom-qmp-phy.txt
> > +++ b/Documentation/devicetree/bindings/phy/qcom-qmp-phy.txt
> > @@ -11,6 +11,7 @@ Required properties:
> >  	       "qcom,msm8996-qmp-usb3-phy" for 14nm USB3 phy on msm8996,
> >  	       "qcom,msm8998-qmp-usb3-phy" for USB3 QMP V3 phy on msm8998,
> >  	       "qcom,msm8998-qmp-ufs-phy" for UFS QMP phy on msm8998,
> > +	       "qcom,sdm845-qmp-pcie-phy" for PCIe phy on sdm845,
> >  	       "qcom,sdm845-qmp-usb3-phy" for USB3 QMP V3 phy on sdm845,
> >  	       "qcom,sdm845-qmp-usb3-uni-phy" for USB3 QMP V3 UNI phy on sdm845,
> >  	       "qcom,sdm845-qmp-ufs-phy" for UFS QMP phy on sdm845.
> > @@ -48,6 +49,10 @@ Required properties:
> >  			"aux", "cfg_ahb", "ref".
> >  		For "qcom,msm8998-qmp-ufs-phy" must contain:
> >  			"ref", "ref_aux".
> > +		For "qcom,sdm845-qmp-usb3-phy" must contain:
> > +			"aux", "cfg_ahb", "ref", "refgen".
> 
> qcom,sdm845-qmp-usb3-phy in a PCIe patch?
> 

Must have forgotten to fix it up after copy pasting the other line,
thanks for spotting.

> > +		For "qcom,sdm845-qmp-usb3-phy" must contain:
> > +			"aux", "cfg_ahb", "ref", "com_aux".
> 
> qcom,sdm845-qmp-usb3-phy again in a PCIe patch?
> 

Ditto.

> 
> > +static const struct qmp_phy_init_tbl sdm845_pcie_pcs_misc_tbl[] = {
> > +	QMP_PHY_INIT_CFG(QPHY_V3_PCS_MISC_OSC_DTCT_CONFIG2, 0x52),
> > +	QMP_PHY_INIT_CFG(QPHY_V3_PCS_MISC_OSC_DTCT_MODE2_CONFIG2, 0x10),
> > +	QMP_PHY_INIT_CFG(QPHY_V3_PCS_MISC_OSC_DTCT_MODE2_CONFIG4, 0x1a),
> > +	QMP_PHY_INIT_CFG(QPHY_V3_PCS_MISC_OSC_DTCT_MODE2_CONFIG5, 0x06),
> > +	QMP_PHY_INIT_CFG(QPHY_V3_PCS_MISC_PCIE_INT_AUX_CLK_CONFIG1, 0x00),
> > +};
> 
> I was thinking I might need to do the same for msm8998, since downstream
> tweaks pcs_misc + 0x2c = 0x52
> (dunno if that's QPHY_V3_PCS_MISC_OSC_DTCT_CONFIG2)
> 

Yeah, we probably should add the 8998 sequences as well. I haven't been
able to compare the two side by side, but it's expected that they will
be slightly different.

> The thing is, phy driver writes unconditionally to pcs_misc + 0x0c
> (QPHY_V3_PCS_MISC_CLAMP_ENABLE). Should that be moved elsewhere?
> 

I spotted this as the only difference in the initialization sequence,
but concluded that it seems to work fine as it is. So I left that part
intact.

Regards,
Bjorn