Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.37 as permitted sender) client-ip=23.128.96.37;
MIME-Version: 1.0
References: <20231023155013.512999-1-romain.gantois@bootlin.com>
 <20231023155013.512999-5-romain.gantois@bootlin.com> <df71bbe5-fec0-45cc-adb4-acfbcc356ba3@lunn.ch>
 <9d447cb5-c8f9-06d2-0909-2870d57f0f7a@bootlin.com> <4bde9110-c88d-4a22-8e67-e788db4673f3@lunn.ch>
In-Reply-To: <4bde9110-c88d-4a22-8e67-e788db4673f3@lunn.ch>
From:   Robert Marko <robert.marko@sartura.hr>
Date:   Tue, 24 Oct 2023 18:49:55 +0200
Message-ID: <CA+HBbNEMvijVFMKYTUopScwcMRwPFpdZ99GwY-Gb=hTLfgJx5Q@mail.gmail.com>
Subject: Re: [PATCH net-next 4/5] net: ipqess: add a PSGMII calibration
 procedure to the IPQESS driver
To:     Andrew Lunn <andrew@lunn.ch>
Cc:     Romain Gantois <romain.gantois@bootlin.com>, davem@davemloft.net,
        Rob Herring <robh+dt@kernel.org>,
        Krzysztof Kozlowski <krzysztof.kozlowski+dt@linaro.org>,
        Jakub Kicinski <kuba@kernel.org>,
        Eric Dumazet <edumazet@google.com>,
        Paolo Abeni <pabeni@redhat.com>, netdev@vger.kernel.org,
        linux-kernel@vger.kernel.org, devicetree@vger.kernel.org,
        thomas.petazzoni@bootlin.com,
        Florian Fainelli <f.fainelli@gmail.com>,
        Heiner Kallweit <hkallweit1@gmail.com>,
        Russell King <linux@armlinux.org.uk>,
        linux-arm-kernel@lists.infradead.org,
        Vladimir Oltean <vladimir.oltean@nxp.com>,
        Luka Perkov <luka.perkov@sartura.hr>,
        Andy Gross <agross@kernel.org>,
        Bjorn Andersson <andersson@kernel.org>,
        Konrad Dybcio <konrad.dybcio@somainline.org>,
        Maxime Chevallier <maxime.chevallier@bootlin.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Tue, Oct 24, 2023 at 4:13=E2=80=AFPM Andrew Lunn <andrew@lunn.ch> wrote:
>
> > Yes, I'll add more detailed comments to the code in the v2. The calibra=
tion
> > procedure itself targets the PSGMII device, which is internal to the So=
C and can
> > be logically accessed as a PHY device on the MDIO bus. This component i=
s a
> > little opaque and has some nonstandard MII register definitions.
> >
> > The "testing" phase that follows the calibration accesses both the inte=
rnal
> > QCA8K switch ports and the external QCA8075 PHY. For example, it puts b=
oth the
> > switch ports and the PHY ports in loopback before starting packet gener=
ation on
> > the external PHYs. This is done to verify that the PSGMII link works co=
rrectly
> > after being calibrated.
> >
> > So this code interacts with both internal ESS devices and external PHYs=
, but
> > mostly the former, which is why I kept everything in the MAC/switch dri=
ver.
>
> Accessing the external PHYs i would suggest go over the normal phylib
> API. Somebody might build a board using a different PHY, with
> different registers. If all you need is loopback, there is a phylib
> call for that.
>
> Directly accessing the internal ESS is fine, it cannot be changed, but
> if there are phylib helpers which do the same thing, consider using
> them.

Hi,
This SoC is a bit special as it only supports using RGMII and PSGMII.
PSGMII is Qualcomm-s modification of SGMII with 5 SGMII lines to provide
1G of bandwidth from the switch to PHY-s.

However, PSGMII is also weird in the sense that it requires
calibration to be carried
on each boot, as otherwise packet loss will start to occur.
But for calibration to work, you must enable loopback on the switch ports a=
nd
on the PHY-s, both loopback and CRC verification must be enabled.
Then you can actually enable the PSGMII serdes calibration in the SoC but i=
t
must occur on all of the PHYs at the same time, hence why broadcast is used=
.

As far as the PHY-s go, there are only 2 PHY models supporting PSGMII,
QCA8072 and QCA8075, both from Qualcomm, and differing only in the
number of ports.
QCA8072 has 2 ports while QCA8075 has 5 ports.
Each of them also has a serdes PHY exposed over MDIO.
These PHY-s are still being used in IPQ8074 and IPQ6018 802.11ax SoC-s.

So in a nutshell, this is how stuff is connected (To the best of my knowled=
ge):
https://asciiflow.com/#/share/eJzVVktuwjAQvYrlFUggwk9p2fGHBVVoVAmkbAJxRSTjo=
GDUIMQtKg5T9TScpG4T8iEOARLaYs3iZTx%2B9oxnMt5Aos4RrJEVxjmI1TUyYQ1uFGgpsPZYKe=
UUuGaoJIoMUWRR9qFA4I79%2B%2BeloigksB606QyZBFFXE7JoGoSaBsbIjLIIDp5Fpi8N27Kcj=
baI5UjkKY%2FT0fUJZQFQMW%2Bu2J3olDfRYTcGWqsFRta5%2B8TqAufffVzh8q0l8tI8fBodLw=
oRSoZJgeDDRYZ%2BQAkcUNmbrXiwai8aGJr%2BqiMt9aOCxkrHNK%2BTcG43pZfsYUGxG4scZg8=
PVyqeGvN5%2BmceNusP5bL4lLSWflmA%2FKbT6SxUMbzoxEfkH9NcEJXdjfURUzH%2FaQAkuTvo=
94HUGxfk9nOrLXPDE0tTtWye6UwlBOElP8qxNJk6XhqAcWgYLcHzN2X2CpozRoKr%2FCPhJHCk1=
rlWd%2FJ8chtflnj3ILZT3LTZX%2FN7PoqWW0j%2BNrh3Ol7RRSe6IBuHRhgklpGpsWLwt519fK=
vyQ9BbT0xdC524N3bdahqLBXsgMnga%2BQw5WefrhOkEmzkxtTc0TLcjCmKpUE1vC0EQGuoS5Ue=
Foxc1i0%2F6u3RGgTtITNzRJ%2FbD3tHwi%2FfORYFbuP0C5MQClA%3D%3D)

Sorry for the external link, but I cannot get the ASCII diagram to
show properly via plain-text.

Regards,
Robert
>      Andrew


--=20
Robert Marko
Staff Embedded Linux Engineer
Sartura Ltd.
Lendavska ulica 16a
10000 Zagreb, Croatia
Email: robert.marko@sartura.hr
Web: www.sartura.hr