linux-kernel - Re: [PATCH 3/6] arm64: dts: qcom: ipq5332: Add MDIO device tree

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <efe398ff-825b-4959-bcfc-ff0870d1932b@quicinc.com>
Date: Wed, 17 Jan 2024 23:10:07 +0800
From: Jie Luo <quic_luoj@...cinc.com>
To: Andrew Lunn <andrew@...n.ch>
CC: <andersson@...nel.org>, <konrad.dybcio@...aro.org>, <robh+dt@...nel.org>,
        <krzysztof.kozlowski+dt@...aro.org>, <conor+dt@...nel.org>,
        <linux-arm-msm@...r.kernel.org>, <devicetree@...r.kernel.org>,
        <linux-kernel@...r.kernel.org>, <netdev@...r.kernel.org>,
        <quic_kkumarcs@...cinc.com>, <quic_suruchia@...cinc.com>,
        <quic_soni@...cinc.com>, <quic_pavir@...cinc.com>,
        <quic_souravp@...cinc.com>, <quic_linchen@...cinc.com>,
        <quic_leiwei@...cinc.com>
Subject: Re: [PATCH 3/6] arm64: dts: qcom: ipq5332: Add MDIO device tree

On 1/17/2024 6:56 AM, Andrew Lunn wrote:
>> Another one is the MDIO slave(gpio25, 26), which is dedicated
>> for receiving the back pressure signal from the connected Ethernet switch
>> device QCA8386.
>>
>> There is a MDIO master block integrated in QCA8386 switch device, this
>> integrated MDIO master is dedicated for generating the back
>> pressure signal to IPQ5332 SoC.
>>
>> This MDIO slave block of IPQ5322 just needs to configure these PIN
>> mux for MDC and MDIO PINs. No additional driver is needed for this MDIO
>> slave block of IPQ5332.
> 
> So there is a proprietary protocol running over the MDIO bus? And its
> completely implemented in hardware in the slave block? Is this even
> MDIO? Does it use c22 or c45 bus transactions? How is the slave
> address configured, or is that also hard coded?
> 
> 	Andrew
> 

Hi Andrew,
Yes, this is a custom HW mechanism using the MDIO C22 frame, to enable 
back pressure from the QCA8386 switch to the IPQ5332 SoC. The slave 
block in the IPQ5332 SoC implements the back pressure function. There is 
no configuration for the MDIO slave address of IPQ5332 required, since 
the connection is one to one between slave and master.

However upon further review, we believe this node definition belongs to
the board DTS file, since the switch configuration is a board property.
We will move out this MDIO slave config from the patch series to
avoid the confusion. We will also rename the node from 'mdio0-state' to 
'backpressure-state' to make this clear.

Thanks.