linux-kernel - Re: [PATCH 3/4] ARM: dts: qcom: msm8960: expressatt: Add Magnetometer

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

Message-ID: <5clxuhy7tprkfxkonzmucfm3d74brk37zjw7lbdj3m2kqail6z@gjoy4pdtaotw>
Date: Fri, 21 Nov 2025 15:11:14 +0200
From: Dmitry Baryshkov <dmitry.baryshkov@....qualcomm.com>
To: Konrad Dybcio <konrad.dybcio@....qualcomm.com>
Cc: Rudraksha Gupta <guptarud@...il.com>,
        Bjorn Andersson <andersson@...nel.org>,
        Konrad Dybcio <konradybcio@...nel.org>, Rob Herring <robh@...nel.org>,
        Krzysztof Kozlowski <krzk+dt@...nel.org>,
        Conor Dooley <conor+dt@...nel.org>, linux-arm-msm@...r.kernel.org,
        devicetree@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH 3/4] ARM: dts: qcom: msm8960: expressatt: Add Magnetometer

On Fri, Nov 21, 2025 at 01:54:22PM +0100, Konrad Dybcio wrote:
> On 11/21/25 12:35 PM, Rudraksha Gupta wrote:
> >>> +        iovdd-supply = <&pm8921_lvs4>;
> >>> +
> >>> +        // TODO: Figure out Mount Matrix
> >> pmOS wiki has this:
> >>
> >> https://github.com/SiddharthManthan/accelerometer-mount-matrix-generator/blob/master/rotation_matrix.py
> >>
> >> but for accelerometers.. shouldn't be too different, I would assume
> > 
> > Unfortunately this seems to generate a broken matrix for me for the accelerometer. Downstream is also useless here as I believe Android does something slightly different.
> > 
> > I have been trying to upstream the pinephone pro's accelerometer and magnetometer as well ( https://lore.kernel.org/all/20250929-ppp_light_accel_mag_vol-down-v4-0-6598f22d3451@gmail.com/ ), but I'm just not sure if what I'm doing is right. I think the easiest solution for me right now is to get the screen working and use GUI programs to figure out if the mount matrix is right. If you have other suggestions, please let me know. I'm all ears.
> > 
> > But since I have your attention for now, I had a couple questions on trying to get the screen to work:
> > 
> > - The pipeline is MDP4 <-> DSI <-> Panel, right? Any information you can give me so that I can try to get this working faster?
> 
> Yes that's the pipeline
> 
> Unless the MDP4 driver is missing something (which it may as well, all
> hw since ~2013 has moved on to MDP5/DPU), you should just be required to
> come up with a panel driver (DSI init sequences, poking at regulators,
> reset GPIO(s) etc.)

For MDP4 I mostly test LVDS and HDMI output, my DSI4 devices, Nexus7 is
not that usable for testing. Last time I checked it, it was working.

> 
> > - Is IOMMU broken? It seems like I can use APQ8064's implementation, but that's broken since at least 5.10. Should this be reported, and if so where?
> 
> I don't know, this platform uses its own arcane driver.. +Dmitry?
> If you have a bug report, please send it to linux-arm-msm

It seemed to work for the minimal test cases (mainly around display).
Please send bug reports to this ML, cc iommu@...ts.linux.dev

> 
> > - I see Adreno 2xx, but seems like there is not Adreno 225 DTS compatible? Any insight here?
> 
> Adreno is a separate HW block, you can/should get software rendering to
> work first

Jonathan Marek had, if I'm not mistaken, A225 implemented, but likely
not contributed upstream. You can try asking him.

-- 
With best wishes
Dmitry