linux-kernel - Re: [PATCH V3 0/3] X1E001DE Snapdragon Devkit for Windows

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

Message-ID: <86v7x81mmj.wl-maz@kernel.org>
Date: Thu, 31 Oct 2024 12:29:40 +0000
From: Marc Zyngier <maz@...nel.org>
To: Abel Vesa <abel.vesa@...aro.org>, Jens Glathe <jens.glathe@...schoolsolutions.biz>
Cc: 	Sibi Sankar <quic_sibis@...cinc.com>,
	andersson@...nel.org,
	konradybcio@...nel.org,
	krzk+dt@...nel.org,
	robh+dt@...nel.org,
	dmitry.baryshkov@...aro.org,
	linux-kernel@...r.kernel.org,
	linux-arm-msm@...r.kernel.org,
	devicetree@...r.kernel.org,
	conor+dt@...nel.org,
	srinivas.kandagatla@...aro.org,
	quic_jjohnson@...cinc.com
Subject: Re: [PATCH V3 0/3] X1E001DE Snapdragon Devkit for Windows

On Thu, 31 Oct 2024 10:00:40 +0000,
Jens Glathe <jens.glathe@...schoolsolutions.biz> wrote:
> 
> On 31.10.24 10:46, Abel Vesa wrote:
> > On 24-10-30 17:02:32, Marc Zyngier wrote:
> >> On Fri, 25 Oct 2024 13:32:24 +0100,
> >> Sibi Sankar <quic_sibis@...cinc.com> wrote:
> >>> Add initial support for X1E001DE Snapdragon Devkit for Windows. X1E001DE
> >>> is the speed binned variant of X1E80100 that supports turbo boost up to
> >>> 4.3 Ghz. The initial support includes the following:
> >>> 
> >>> -DSPs
> >>> -Ethernet (RTL8125BG) over the pcie 5 instance.
> >>> -NVme
> >>> -Wifi
> >>> -USB-C ports
> >>> 
> >>> V3:
> >>> * Asked around and looked at the firmware, couldn't find a codename so
> >>>    will keep it as DEVKIT. Will update it if someone from the community
> >>>    finds something else.
> >> My machine has the following information as part of its DMI tables:
> >> 
> >> Handle 0x0005, DMI type 1, 27 bytes
> >> System Information
> >> 	Manufacturer: Qualcomm
> >> 	Product Name: Snapdragon-Devkit
> >> 	Version: 2.1
> >> 	Serial Number: 5
> >> 	UUID: 63b5fc8b-9c50-89aa-fd0f-3fcef93dc291
> >> 	Wake-up Type: Power Switch
> >> 	SKU Number: 6
> >> 	Family: SCP_HAMOA
> >> 
> >> So I guess that Snapdragon-Devkit is another possible name. But given
> >> that it is a bit of a mouthful, devkit, Devkit, or any other variation
> >> on the case would work for me.
> > The point was to have something unique A codename would be unique.
> > Naming it Snapdragon-Devkit (or just devkit) will be confusing since
> > there was already a 2023 devkit (from Microsoft) with the Snapdragon
> > 8cx Gen 3, and probably the next compute platform will also have a devkit
> > as well. So probably "X Elite devkit" could be the right option..

Odd, I didn't get that email.

My point was the the HW already comes with a full description as part
of the existing tables. If you really want something that is truly
unique to that platform and that can be used by a tool (be it
firmware, kernel or userspace) to understand what it is running on,
then you cannot have *less* information.

At the very least, you would need Manufacturer, Product Name, Version
and Family.

But does it really matter? I don't think it is *that* crucial. At the
end of the day, this is only used to pick the correct DT out of a set
for a given SoC, or worse case a family of SoCs that are closely
related.

> As for The Windows Dev Kit 2023, dmidecode says this:
> 
> Handle 0x0009, DMI type 1, 27 bytes
> System Information
>     Manufacturer: Microsoft Corporation
>     Product Name: Windows Dev Kit 2023
>     Version: 124I:00097T:000M:0200000B:07
>     Serial Number: 0F01C4F22373F6
>     UUID: e4a4662c-8367-75d0-a54f-1d04bd404860
>     Wake-up Type: Unknown
>     SKU Number: 2043
>     Family: Surface
> 
> That's also really a mouthful. In my patchset for it there were some
> name / path changes, microsoft/blackrock it is now. Would be cool to
> have short and unique names. In the end, whatever works and is unique.
> Like those UUIDs?

Are those actually per platform? or per unit? On my box, the serial
number is probably a dud. What does the UUID reports on your X1E box?

Thanks,

	M.

-- 
Without deviation from the norm, progress is not possible.