linux-kernel - Re: [PATCH] arm64: tegra: Bump #address-cells and #size-cells on Tegra186

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: <CALHNRZ8N=NnirL_vBYjsUt_w8hSXzu5z7H7ditFQTjuHH2Zs2A@mail.gmail.com>
Date: Wed, 7 May 2025 11:26:57 -0500
From: Aaron Kling <webgeek1234@...il.com>
To: Thierry Reding <thierry.reding@...il.com>
Cc: Rob Herring <robh@...nel.org>, Krzysztof Kozlowski <krzk+dt@...nel.org>, 
	Conor Dooley <conor+dt@...nel.org>, Jonathan Hunter <jonathanh@...dia.com>, devicetree@...r.kernel.org, 
	linux-tegra@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] arm64: tegra: Bump #address-cells and #size-cells on Tegra186

On Wed, May 7, 2025 at 10:37 AM Thierry Reding <thierry.reding@...il.com> wrote:
>
> On Sat, Apr 19, 2025 at 10:30:31PM -0500, Aaron Kling via B4 Relay wrote:
> > From: Aaron Kling <webgeek1234@...il.com>
> >
> > This was done for Tegra194 and Tegra234 in 2838cfd, but Tegra186 was not
> > part of that change. The same reasoning for that commit also applies to
> > Tegra186, plus keeping the archs as close to each other as possible makes
> > it easier to compare between them and support features concurrently.
>
> As explained in the commit that you referenced, the reason for making
> these changes for Tegra194 and Tegra234 was so that the PCI and GPU
> nodes could move back into the bus@0 node. This doesn't exist on
> Tegra186, and the top-level already has #address-cells = <2> and
> #size-cells = <2>.

This isn't recursive, though. I had thought it was, but kept having
issues. Then I found docs that say:

The #address-cells and #size-cells properties are not inherited from
ancestors in the devicetree. They shall be explicitly defined. [0]

> Does this actually fix a bug? Just making this look more similar to
> Tegra194/234 doesn't seem like the best of justifications for bloating
> the DT.

Tegra132 and Tegra210 also have size 2 on all these nodes. I probably
should have mentioned that in the message too. But having Tegra186 as
the only odd out tegra arm64 arch is confusing and makes for extra
work when trying to implement things across all archs.

What made me sit down and and type all this out was an attempt to get
simplefb working for seamless display handoff. And I could not get the
reserved-memory nodes and iommu-addresses and all to line up. Not
until I made everything connected to that have #address-cells = <2>
and #size-cells = <2>. Which happened to line up with every arm64
tegra arch except t186, so I submitted this.

Sincerely,
Aaron Kling

[0] https://devicetree-specification.readthedocs.io/en/stable/devicetree-basics.html#address-cells-and-size-cells