linux-kernel - Re: [PATCH v1] arm64: dts: qcom: monaco-evk: Add firmware-name to QUPv3 nodes

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] [day] [month] [year] [list]

Message-ID: <f2ea3da2-4582-a2d7-9a90-8050b3e8a27f@oss.qualcomm.com>
Date: Fri, 24 Oct 2025 17:07:20 +0530
From: Viken Dadhaniya <viken.dadhaniya@....qualcomm.com>
To: Konrad Dybcio <konrad.dybcio@....qualcomm.com>, andersson@...nel.org,
        konradybcio@...nel.org, robh@...nel.org, krzk+dt@...nel.org,
        conor+dt@...nel.org, linux-arm-msm@...r.kernel.org,
        devicetree@...r.kernel.org, linux-kernel@...r.kernel.org
Cc: mukesh.savaliya@....qualcomm.com, anup.kulkarni@....qualcomm.com
Subject: Re: [PATCH v1] arm64: dts: qcom: monaco-evk: Add firmware-name to
 QUPv3 nodes



On 10/22/2025 10:30 PM, Konrad Dybcio wrote:
> On 9/26/25 8:43 AM, Viken Dadhaniya wrote:
>>
>>
>> On 9/25/2025 2:22 PM, Konrad Dybcio wrote:
>>> On 9/25/25 6:26 AM, Viken Dadhaniya wrote:
>>>> Traditionally, firmware loading for Serial Engines (SE) in the QUP hardware
>>>> of Qualcomm SoCs has been managed by TrustZone (TZ). While this approach
>>>> ensures secure SE assignment and access control, it limits flexibility for
>>>> developers who need to enable various protocols on different SEs.
>>>>
>>>> Add the firmware-name property to QUPv3 nodes in the device tree to enable
>>>> firmware loading from the Linux environment. Handle SE assignments and
>>>> access control permissions directly within Linux, removing the dependency
>>>> on TrustZone.
>>>>
>>>> Signed-off-by: Viken Dadhaniya <viken.dadhaniya@....qualcomm.com>
>>>> ---
>>>
>>> Viken, you've posted a lot of patches like these lately for
>>> specific boards.
>>>
>>> Should we instead push this to the SoC DTSI, to limit the boilerplate?
>>
>> The firmware load from Linux feature is applicable only to devkit boards
>> (RB3 Gen2 and EVK), and therefore, it is being added exclusively to the
>> board-specific device tree files.
> 
> Do we expect GENI_SE_INVALID_PROTO to ever show up on hardware that
> can't make use of fw loading though?

If TZ (TrustZone) does not load firmware for any SE, the SE protocol status will
remain GENI_SE_INVALID_PROTO. In such cases, Linux can load the firmware to
initialize the SE and set the required protocol.

> 
>>> If I recall correctly, you said these ELFs are not vendor-signed, so
>>> it should be OK to have them be assigned globally
>>>
>>
>> This feature is intended solely for use with the development kit and is not
>> intended for commercialization. Therefore, vendor sign-off is currently not
>> applicable.
> 
> Well you conveniently never really mentioned this in the driver patchset..
> 
> Konrad