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: <3genyggxae5ejlpi2k2zflliaujdov6f2nd5nppzxtg7fmerff@52dac4oh2c3z>
Date: Sat, 1 Nov 2025 10:43:08 +0200
From: Dmitry Baryshkov <dmitry.baryshkov@....qualcomm.com>
To: Tessolve Upstream <tessolveupstream@...il.com>
Cc: Bjorn Andersson <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
Subject: Re: [PATCH v1 1/1] arm64: dts: qcom: talos-evk: Add support for
 dual-channel LVDS panel

On Fri, Oct 31, 2025 at 02:42:05PM +0530, Tessolve Upstream wrote:
> 
> 
> On 29/10/25 21:18, Bjorn Andersson wrote:
> > On Tue, Oct 28, 2025 at 11:46:36AM +0530, Sudarshan Shetty wrote:
> >> This patch introduces a new device tree for the QCS615 Talos
> > 
> > "This patch" doesn't make sense when you look at the git log once the
> > patch has been accepted, please avoid it.
> > 
> > Please read https://docs.kernel.org/process/submitting-patches.html#describe-your-changes
> > 
> > Start your commit message with a "problem description", describe what
> > this LVDS talos is, why it should have it's own dts file etc.
> 
> Okay, will update in next patch.
> > 
> >> EVK platform with dual-channel LVDS display support.
> >>
> >> The new DTS file (`talos-evk-lvds.dts`) is based on the existing
> >> `talos-evk.dts` and extends it to enable a dual-channel LVDS display
> >> configuration using the TI SN65DSI84 DSI-to-LVDS bridge.
> >>
> >> where channel-A carries odd pixel and channel-B carries even pixel
> >> on the QCS615 talos evk platform.
> >>
> >> Signed-off-by: Sudarshan Shetty <tessolveupstream@...il.com>
> >> ---
> >>  arch/arm64/boot/dts/qcom/Makefile           |   1 +
> >>  arch/arm64/boot/dts/qcom/talos-evk-lvds.dts | 128 ++++++++++++++++++++
> >>  2 files changed, 129 insertions(+)
> >>  create mode 100644 arch/arm64/boot/dts/qcom/talos-evk-lvds.dts
> >>
> >> diff --git a/arch/arm64/boot/dts/qcom/Makefile b/arch/arm64/boot/dts/qcom/Makefile
> >> index d5a3dd98137d..6e7b04e67287 100644
> >> --- a/arch/arm64/boot/dts/qcom/Makefile
> >> +++ b/arch/arm64/boot/dts/qcom/Makefile
> >> @@ -307,6 +307,7 @@ dtb-$(CONFIG_ARCH_QCOM)	+= sm8750-mtp.dtb
> >>  dtb-$(CONFIG_ARCH_QCOM)	+= sm8750-qrd.dtb
> >>  dtb-$(CONFIG_ARCH_QCOM)	+= talos-evk.dtb
> >>  dtb-$(CONFIG_ARCH_QCOM)	+= talos-evk-dsi.dtb
> >> +dtb-$(CONFIG_ARCH_QCOM)	+= talos-evk-lvds.dtb
> >>  x1e001de-devkit-el2-dtbs	:= x1e001de-devkit.dtb x1-el2.dtbo
> >>  dtb-$(CONFIG_ARCH_QCOM)	+= x1e001de-devkit.dtb x1e001de-devkit-el2.dtb
> >>  x1e78100-lenovo-thinkpad-t14s-el2-dtbs	:= x1e78100-lenovo-thinkpad-t14s.dtb x1-el2.dtbo
> >> diff --git a/arch/arm64/boot/dts/qcom/talos-evk-lvds.dts b/arch/arm64/boot/dts/qcom/talos-evk-lvds.dts
> >> new file mode 100644
> >> index 000000000000..7ba4ab96ada6
> >> --- /dev/null
> >> +++ b/arch/arm64/boot/dts/qcom/talos-evk-lvds.dts
> >> @@ -0,0 +1,128 @@
> >> +// SPDX-License-Identifier: BSD-3-Clause
> >> +/*
> >> + * Copyright (c) Qualcomm Technologies, Inc. and/or its subsidiaries.
> >> + */
> >> +/dts-v1/;
> >> +#include "talos-evk.dts"
> > 
> > We don't include .dts files, split the existing one in a dtsi and dts
> > file and then include the dtsi here. Or provide provide this as a dtso
> > overlay on top of the dts.
> > 
> > It's not clear to me which is the correct way, because you didn't
> > adequately described how the SN65DSI84 enter the picture. Is it always
> > there, but not part of the standard dip-switch configuration? Or is this
> > some mezzanine?
> 
> Thanks for the feedback.

You didn't answer the question. It was about the hardware, not about
DTs.

> Currently, the Talos device tree hierarchy is organized as follows:
> 
> talos-som.dtsi — defines SoM-specific interfaces
> talos-evk.dts — adds carrier board (CB) interfaces such as MicroSD, power
> button, and HDMI
> talos-evk-lvds.dts — enables the LVDS display (includes SoM + CB +
> LVDS + disables HDMI)

So, is LVDS a part of the standard board or is it a mezzanine?

> 
> The LVDS and HDMI displays share a common DSI output, so only one 
> interface can be active at a time. At present, talos-evk-lvds.dts 
> includes talos-evk.dts directly so that the base SoM and carrier 
> interfaces (e.g., MicroSD, power button) remain available.
> 
> However, as you pointed out, including a .dts file directly is not
> recommended upstream. To address this, I am considering the following 
> restructuring options:
> 
> Option 1: Introduce a talos-cb.dtsi
> 
> talos-som.dtsi: SoM-specific interfaces
> talos-cb.dtsi: common carrier board interfaces (MicroSD, power button, etc.)
> talos-evk.dts: includes talos-som.dtsi + talos-cb.dtsi + HDMI
> talos-evk-lvds.dts: includes talos-som.dtsi + talos-cb.dtsi + LVDS
> 
> This approach avoids including .dts files directly and keeps the carrier
> board interfaces centralized and reusable.It also cleanly separates SoM
> and CB content and is consistent with how other Qualcomm platforms 
> structure their EVK variants.

This sounds okay, but please respond to previous questions.

> 
> Option 2: Move CB interfaces to talos-som.dtsi (disabled by default)

Why? What happens when somebody reuses the SoM with some other base
board?

> 
> Move MicroSD, power button, etc., to talos-som.dtsi with status = "disabled";
> Enable these interfaces in each top-level DTS (e.g., talos-evk.dts,
> talos-evk-lvds.dts)
> While this also avoids .dts inclusion, it may make the SoM DTS 
> unnecessarily complex and less reusable, as those CB-specific 
> peripherals don’t belong to the SoM hardware.
> 
> Let me know your prepared approach here.

If you yourself don't like the second option, why do you propose it?

-- 
With best wishes
Dmitry

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ