linux-kernel - Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for multipd model

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 for Android: free password hash cracker in your pocket

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

Message-ID: <04f5e3cb-d2f5-747c-1fd0-4b61d845e2c5@quicinc.com>
Date:   Wed, 14 Jun 2023 17:13:27 +0530
From:   Manikanta Mylavarapu <quic_mmanikan@...cinc.com>
To:     Krzysztof Kozlowski <krzysztof.kozlowski@...aro.org>,
        Kalle Valo <kvalo@...nel.org>
CC:     <agross@...nel.org>, <andersson@...nel.org>,
        <konrad.dybcio@...aro.org>, <robh+dt@...nel.org>,
        <krzysztof.kozlowski+dt@...aro.org>, <conor+dt@...nel.org>,
        <jassisinghbrar@...il.com>, <mathieu.poirier@...aro.org>,
        <mturquette@...libre.com>, <sboyd@...nel.org>,
        <quic_eberman@...cinc.com>, <quic_mojha@...cinc.com>,
        <loic.poulain@...aro.org>, <linux-arm-msm@...r.kernel.org>,
        <devicetree@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
        <linux-remoteproc@...r.kernel.org>, <linux-clk@...r.kernel.org>,
        <quic_srichara@...cinc.com>, <quic_sjaganat@...cinc.com>,
        <quic_kathirav@...cinc.com>, <quic_anusha@...cinc.com>,
        <quic_poovendh@...cinc.com>, <quic_varada@...cinc.com>,
        <quic_devipriy@...cinc.com>
Subject: Re: [PATCH V2 01/13] dt-bindings: remoteproc: qcom: Add support for
 multipd model



On 6/7/2023 1:57 PM, Krzysztof Kozlowski wrote:
> On 07/06/2023 10:10, Manikanta Mylavarapu wrote:
>>
>>
>> On 6/6/2023 7:19 PM, Kalle Valo wrote:
>>> Manikanta Mylavarapu <quic_mmanikan@...cinc.com> writes:
>>>
>>>>>>>> +
>>>>>>>> +    properties:
>>>>>>>> +      compatible:
>>>>>>>> +        enum:
>>>>>>>> +          - qcom,ipq5018-wcss-ahb-mpd
>>>>>>>> +          - qcom,ipq9574-wcss-ahb-mpd
>>>>>>>> +          - qcom,ipq5018-wcss-pcie-mpd
>>>>>>>
>>>>>>> Keep rather alphabetical order (so both 5018 together).
>>>>>>>
>>>>>>> I also do not understand these at all. Why adding bus type to
>>>>>>> compatible? This rarely is allowed (unless it is PCIe controller within
>>>>>>> soc).
>>>>>>>
>>>>>> IPQ5018 SOC has in-built PCIE controller. Here QDSP6 will bring up
>>>>>> external(PCIE) and internal (AHB) wifi radio's. To separate AHB, PCIE
>>>>>> radio's properties, i have added bus type to compatible.
>>>>>
>>>>> It's the same device - WCSS - right? We do not create multiple nodes and
>>>>> compatibles for the same devices. Bus suffixes are almost never parts of
>>>>> compatibles.
>>>>
>>>>
>>>> No it's not the same device. WCSS on inside IPQ5018 and WCSS attached
>>>> via pcie to IPQ5018. Here QDSP6 managing both WCSS's.
>>>>
>>>> So for better clarity i will use attached SOC ID in compatible.
>>>> Below are the new compatible's.
>>>>
>>>> - qcom,ipq5018-wcss-mpd //IPQ5018 internal radio
>>>> - qcom,ipq9574-wcss-mpd	//IPQ9574 internal radio
>>>> - qcom,qcn6122-wcss-mpd //IPQ5018 attached radio
>>>
>>> What mandates that there's just one QCN6122 device attached to PCI?
>>> Assuming fixed PCI configurations like that makes me worried.
>>>
>>
>> IPQ5018 always has one internal radio, attached pcie radio's depends on
>> no of pcie ports. IPQ5018 has 2 pcie ports, so it supports max two
>> qcn6122 devices. One compatible (qcom,qcn6122-wcss-mpd) itself support's
>> number of pcie devices controlled by QDSP6.
> 
> So this is hot-pluggable (or at least board-pluggable), then should not
> be a part of static DTS.
> 
> Some concepts of virtual-processes is anyway far away from hardware
> description, thus does not fit into DTS. Adding now to the equation PCIe
> with variable number of such processes, brings us even further.
> 
> This is not a DT property. Remember - DT describes hardware.
> 
> Best regards,
> Krzysztof
> 

In the multipd architecture based Socs, There is one Q6 DSP which runs 
the OS/kernel and there are one or more instances of WCSS radios
(It can be either internal or pcie attached).
These WCSS cores are controlled by the Q6 (Q6 DSP brings wcss radios out 
of reset/ shuts it down etc). Q6 forms the 'root Protection domain' and 
the wcss radios are termed as the 'user protection domain'.
The compatible's that is being added here are to manage the 'root 
domain' and 'user domain'.
Not sure if using the words 'pcie'/'ahb' made it confusing.
So, 'qcom,ipq5018-q6-mpd' and 'qcom,ipq5018-wcss-mpd'.

There will be multiple instances of 'qcom,ipq5018-wcss-mpd' in DT based 
on number of wcss radios connected on that board and only one instance 
of 'qcom,ipq5018-q6-mpd'.

Is this approach ok ?

Thanks & Regards,
Manikanta.