From: Wasim Nazir <quic_wasimn@quicinc.com>
To: Krzysztof Kozlowski <krzk@kernel.org>
Cc: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>,
Bjorn Andersson <andersson@kernel.org>,
Konrad Dybcio <konradybcio@kernel.org>,
Rob Herring <robh@kernel.org>,
Krzysztof Kozlowski <krzk+dt@kernel.org>,
Conor Dooley <conor+dt@kernel.org>,
<linux-arm-msm@vger.kernel.org>, <devicetree@vger.kernel.org>,
<linux-kernel@vger.kernel.org>, <kernel@quicinc.com>
Subject: Re: [PATCH v5 5/6] arm64: dts: qcom: Add support for QCS9075 Ride & Ride-r3
Date: Thu, 6 Mar 2025 13:55:23 +0530 [thread overview]
Message-ID: <Z8lb889QrqluPXXl@hu-wasimn-hyd.qualcomm.com> (raw)
In-Reply-To: <Z8laCxtHOdNm3rRu@hu-wasimn-hyd.qualcomm.com>
On Thu, Mar 06, 2025 at 01:47:15PM +0530, Wasim Nazir wrote:
> On Mon, Mar 03, 2025 at 08:46:55AM +0100, Krzysztof Kozlowski wrote:
> > On 27/02/2025 08:37, Wasim Nazir wrote:
> > > On Wed, Jan 15, 2025 at 09:35:34AM +0100, Krzysztof Kozlowski wrote:
> > >> On 15/01/2025 06:48, Wasim Nazir wrote:
> > >>>> The the SoC, I am asking about the board. Why each of them is for
> > >>>> example r3?
> > >>>>
> > >>>> So this is not sufficient explanation, nothing about the board, and
> > >>>> again just look Renesas and NXP.
> > >>>>
> > >>>
> > >>> Hi Krzysztof,
> > >>>
> > >>> sa8775p(AUTO), qcs9100(IOT), qcs9075(IOT) are different SoCs based on
> > >>> safety capabilities and memory map, serving different purpose.
> > >>> Ride & Ride-r3 are different boards based on ethernet capabilities and
> > >>> are compatible with all the SoCs mentioned.
> > >>
> > >
> > > Hi Krzysztof,
> > >
> > >> Compatible? What does it mean for a board?
> > >>
> > >
> > > Ride board is based on multiple daughter cards (SOC-card, display,
> > > camera, ethernet, pcie, sensor, etc.).
> > >
> > > The SOC is not directly soldered to Ride board, instead SOC is soldered
> > > on SIP (System in Package) card which can be mounted on SOC-daughter card of
> > > Ride board.
> > > - SoC => SIP-card => SOC-daughter-card (Ride)
> >
> >
> > So basically pretty like other designs using SoM.
> >
> > >
> > > Together with SIP cards and other daughter cards we are creating different
> > > <soc>-Ride Variants with differences in memory map & thermal mitigations.
> > >
> > > The SIP card consists of SOC, PMIC & DDR and it is pin compatible to the
> > > SOC daughter card of <soc>-Ride board. Only SOC is changing accross SIP
> > > cards, except an additional third party SIL-PMIC for SAIL, which is not
> > > present in QCS9075 Ride.
> >
> > Just like every SoM
> >
> > >
> > > Other daughter cards remains same for <soc>-Ride variants, except
> > > ethernet card which is different for <soc>-Ride rev3 variants.
> > >
> > > So the Ride board (combination of daughter cards) is same across the SIP,
> > > while SOC on SIP card is changing which can be sa8775p, qcs9100 or qcs9075.
> > >
> > >> Third time: did you look how other vendors do it?
> > >>
> > >
> > > Yes, we have reviewed other vendors. However, please feel free to share
> > > any specific reference you would like us to follow.
> > >
> > > Here are few reference files we found from other vendors where similar
> > > tasks are performed which includes code refactoring and HW modularity:
> > > - Freescale: fsl-ls208xa.dtsi, fsl-ls2088a.dtsi, fsl-ls2081a-rdb.dts
> >
> > That's an unexpected choice - I would rather look at dozen of SoMs for
> > iMX platforms.
> >
> > > - Renesas: white-hawk-common.dtsi, r8a779g0-white-hawk.dts
> > > - Rockchip: px30-engicam-common.dtsi, px30-engicam-ctouch2.dtsi,
> > > px30-engicam-px30-core-ctouch2.dts
> > >
> > > In our case along with describing the HW, code refactoring is also done
> > > which might be causing confusion, but we are ready for any inputs for
> > > correction.
> >
> > I don't understand why this was not properly described since beginning.
> > You had the hardware in your hands and went with incomplete or even
> > incorrect hardware description.
> >
> > >
> > > Putting this pictorial diagram for updated DT structure depicting our HW.
> > > - qcs9xxx-module.dtsi specifying QCS9xxx based SIP card/module having
> > > SoC, PMICs, Memory-map updates.
> > > - qcom-ride-common.dtsi specifying ride daughter boards, here we are
> > > doing code refactoring also as this is common for all ride boards.
> > > - qcom-ride-ethernet-aqr115c.dtso specifying ethernet overlay board which
> > > uses 2.5G phy and can be overlayed to ride boards to get ride-r3.
> > > By default ride uses 1G phy.
> > > - qcs9075-iq-9075-evk.dts is the new name for RB8 as per new product
> > > name. We will be changing this in next patch series.
> > >
> > > +-----------------------------------------------------------------------------------------------------------------------------------------------+
> > > | |
> > > | sa8775p.dtsi |
> > > | | |
> > > | +-------------------------+-----------------------+ |
> > > | | | | |
> > > | v | v |
> > > | qcs9075-module.dtsi | qcs9100-module.dtsi |
> >
> > So this is the SoM?
>
> Yes this is SoM.
>
> >
> > > | | | | |
> > > | v v v |
> > > | (IOT) (AUTO) (IOT) |
> > > | | | | |
> > > | +----------------------+ | | |
> > > | | | | | |
> > > | | | +-------------------------+-----------------------+-------------------< qcom-ride-common.dtsi |
> >
> > Which piece of actual hardware is represented in qcom-ride-common?
> >
>
> All daughter cards like SOC-card, display, camera, ethernet, pcie, sensor, etc.
>
> > > | | | | | | | | |
> > > | v v v v v v v |
> > > | qcs9075-iq-9075-evk.dts qcs9075-ride.dts sa8775p-ride.dts qcs9100-ride.dts |
> > > | | | | |
> > > | | +-------------------------+-----------------------+-------------------< qcom-ride-ethernet-aqr115c.dtso |
> > > | | | | | | | |
> > > | v v v v v v |
> > > | qcs9075-ride-r3.dts sa8775p-ride-r3.dts qcs9100-ride-r3.dts |
> >
> > I think I gave already few times that answer: No. You cannot reference
> > from a module.c another .c file. You cannot reference DTS from DTS.
> >
> > Strictly speaking you can, of course, but you must not. That's not how
> > source code is done to be manageable and readable.
>
> Ah the arrow is leading to confusion.
>
> Actually we are not including dts here instead *.dtso file will be
> overlayed to *-ride.dts to generate *-ride-r3.dts.
>
> Below is the correct arrow sequence.
>
> | qcs9075-iq-9075-evk.dts qcs9075-ride.dts sa8775p-ride.dts qcs9100-ride.dts |
> | | | | |
> | +-------------------------+-----------------------+---------------------< qcom-ride-ethernet-aqr115c.dtso |
> | | | | |
> | v v v |
> | qcs9075-ride-r3.dts sa8775p-ride-r3.dts qcs9100-ride-r3.dts |
>
> >
> > > | |
> > > +-----------------------------------------------------------------------------------------------------------------------------------------------+
> > >
> > >>>
> > >>> With the combination of these 3 SoCs and 2 boards, we have 6 platforms,
> > >>> all of which we need.
> > >>> - sa8775p-ride.dts is auto grade Ride platform with safety feature.
> > >>> - qcs9100-ride.dts is IOT grade Ride platform with safety feature.
> > >>> - qcs9075-ride.dts is IOT grade Ride platform without safety feature.
> > >>>
> > >>> Since the Ride-r3 boards are essentially Ride boards with Ethernet
> > >>> modifications, we can convert the Ride-r3 DTS to overlays.
> > >> How one board can be with multiple SoCs? If it is soldered, it's close
> > >> to impossible - that's just not the same board. If it is not soldered,
> > >> why you are not explaining it? What is Ride board? What is there? What
> > >> can go there? How it can be used in other SoCs? Or for which SoCs? Is
> > >> there a datasheet available?
> > >>
> > >
> > > As our SoC is based on SIP card and SIP card is compatible with Ride
> > > board, we could able to use same Ride board (which is combination of
> > > multiple daughter cards) with multiple SIP cards.
> > > These SIP cards can be of sa8775p, qcs9100 or qcs9075 SOC.
> >
> > Describe properly the hardware - if you have a module or SIP if you
> > decide not to use industry-standard naming (but why...), then describe
> > it in DTSI.
>
> We refer to it as ‘module’ in our datasheet, so I use the same term
> here. Thanks for pointing it out; we can proceed with the SoM name.
>
> Below is the updated diagram:
> +-----------------------------------------------------------------------------------------------------------------------------------------------+
> | |
> | sa8775p.dtsi |
> | | |
> | +-------------------------+-----------------------+ |
> | | | | |
> | v | v |
> | qcs9075-som.dtsi | qcs9100-som.dtsi |
> | | | | |
> | v v v |
> | (IOT) (AUTO) (IOT) |
> | | | | |
> | +----------------------+ | | |
> | | | | | |
> | | | +-------------------------+-----------------------+-----------------< sa8775p-ride-common.dtsi |
> | | | | | | | | |
> | v v v v v v v |
> | qcs9075-iq-9075-evk.dts qcs9075-ride.dts sa8775p-ride.dts qcs9100-ride.dts |
> | | | | |
> | +-------------------------+-----------------------+-------------------< sa8775p-ride-ethernet-aqr115c.dtso |
> | | | | |
> | v v v |
> | qcs9075-ride-r3.dts sa8775p-ride-r3.dts qcs9100-ride-r3.dts |
> | |
> +-----------------------------------------------------------------------------------------------------------------------------------------------+
>
Updating typo: qcom-ride-* changed to sa8775p-ride-*.
> >
> > Best regards,
> > Krzysztof
>
> Thanks & regards,
> Wasim
next prev parent reply other threads:[~2025-03-06 8:25 UTC|newest]
Thread overview: 59+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-12-29 15:23 [PATCH v5 0/6] arm64: qcom: Add support for QCS9075 boards Wasim Nazir
2024-12-29 15:23 ` [PATCH v5 1/6] dt-bindings: arm: qcom,ids: add SoC ID for QCS9075 Wasim Nazir
2024-12-29 15:23 ` [PATCH v5 2/6] soc: qcom: socinfo: add QCS9075 SoC ID Wasim Nazir
2024-12-29 15:23 ` [PATCH v5 3/6] dt-bindings: arm: qcom: Document rb8/ride/ride-r3 on QCS9075 Wasim Nazir
2024-12-29 15:23 ` [PATCH v5 4/6] arm64: dts: qcom: Add support for QCS9075 RB8 Wasim Nazir
2025-05-06 12:08 ` Dmitry Baryshkov
2025-05-06 12:29 ` Wasim Nazir
2025-05-06 12:30 ` Dmitry Baryshkov
2025-05-06 13:14 ` Wasim Nazir
2025-05-09 22:56 ` Konrad Dybcio
2024-12-29 15:23 ` [PATCH v5 5/6] arm64: dts: qcom: Add support for QCS9075 Ride & Ride-r3 Wasim Nazir
2024-12-30 15:32 ` Konrad Dybcio
2024-12-31 5:10 ` Andrew Lunn
2025-01-02 9:13 ` Wasim Nazir
2024-12-30 15:45 ` Dmitry Baryshkov
2025-01-02 9:07 ` Wasim Nazir
2025-01-03 5:50 ` Dmitry Baryshkov
2025-01-03 7:07 ` Wasim Nazir
2025-01-03 10:31 ` Dmitry Baryshkov
2025-01-03 18:59 ` Wasim Nazir
2025-01-03 19:58 ` Dmitry Baryshkov
2025-01-08 14:09 ` Krzysztof Kozlowski
2025-01-09 14:47 ` Wasim Nazir
2025-01-09 15:03 ` Wasim Nazir
2025-01-09 16:16 ` Krzysztof Kozlowski
2025-01-15 5:48 ` Wasim Nazir
2025-01-15 8:35 ` Krzysztof Kozlowski
2025-02-27 7:37 ` Wasim Nazir
2025-03-03 7:46 ` Krzysztof Kozlowski
2025-03-06 8:17 ` Wasim Nazir
2025-03-06 8:25 ` Wasim Nazir [this message]
2025-03-11 8:02 ` Krzysztof Kozlowski
2025-03-20 11:45 ` Wasim Nazir
2025-03-29 4:48 ` Krzysztof Kozlowski
2025-04-02 6:48 ` Wasim Nazir
2025-03-11 7:59 ` Krzysztof Kozlowski
2025-01-09 13:52 ` Wasim Nazir
2025-01-06 23:59 ` Bjorn Andersson
2025-01-09 13:36 ` Wasim Nazir
2024-12-29 15:23 ` [PATCH v5 6/6] arm64: dts: qcom: Enable cpu cooling devices for QCS9075 platforms Wasim Nazir
2024-12-30 6:02 ` Aiqun(Maria) Yu
2024-12-30 15:35 ` Konrad Dybcio
2024-12-31 11:05 ` Manaf Meethalavalappu Pallikunhi
2024-12-31 16:21 ` Konrad Dybcio
2025-01-08 12:10 ` Manaf Meethalavalappu Pallikunhi
2024-12-30 15:40 ` Dmitry Baryshkov
2024-12-31 12:01 ` Manaf Meethalavalappu Pallikunhi
2025-01-03 5:51 ` Dmitry Baryshkov
2025-01-08 12:27 ` Manaf Meethalavalappu Pallikunhi
2025-01-08 12:46 ` Dmitry Baryshkov
2025-01-08 16:08 ` Manaf Meethalavalappu Pallikunhi
2025-01-09 14:30 ` Konrad Dybcio
2025-01-09 23:54 ` Dmitry Baryshkov
2025-01-10 12:31 ` Konrad Dybcio
2025-01-13 8:43 ` Dmitry Baryshkov
2025-01-14 19:16 ` Manaf Meethalavalappu Pallikunhi
2025-01-15 8:54 ` Dmitry Baryshkov
2024-12-30 18:17 ` [PATCH v5 0/6] arm64: qcom: Add support for QCS9075 boards Rob Herring (Arm)
2025-01-07 16:38 ` (subset) " Bjorn Andersson
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=Z8lb889QrqluPXXl@hu-wasimn-hyd.qualcomm.com \
--to=quic_wasimn@quicinc.com \
--cc=andersson@kernel.org \
--cc=conor+dt@kernel.org \
--cc=devicetree@vger.kernel.org \
--cc=dmitry.baryshkov@linaro.org \
--cc=kernel@quicinc.com \
--cc=konradybcio@kernel.org \
--cc=krzk+dt@kernel.org \
--cc=krzk@kernel.org \
--cc=linux-arm-msm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=robh@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox