linux-kernel - Re: [PATCH] arm64: dts: imx8dxl-ss-conn: Disable USB3 nodes

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: <20250611-unread-other-a559c3b98494@thorsis.com>
Date: Wed, 11 Jun 2025 09:00:51 +0200
From: Alexander Dahl <ada@...rsis.com>
To: Frank Li <Frank.li@....com>
Cc: Alexander Dahl <ada@...rsis.com>, devicetree@...r.kernel.org,
	Rob Herring <robh@...nel.org>,
	Krzysztof Kozlowski <krzk+dt@...nel.org>,
	Conor Dooley <conor+dt@...nel.org>, Shawn Guo <shawnguo@...nel.org>,
	Sascha Hauer <s.hauer@...gutronix.de>,
	Pengutronix Kernel Team <kernel@...gutronix.de>,
	Fabio Estevam <festevam@...il.com>,
	"open list:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE" <imx@...ts.linux.dev>,
	"moderated list:ARM/FREESCALE IMX / MXC ARM ARCHITECTURE" <linux-arm-kernel@...ts.infradead.org>,
	open list <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] arm64: dts: imx8dxl-ss-conn: Disable USB3 nodes

Hello Frank,

thanks for reviewing.

Am Tue, Jun 10, 2025 at 11:34:22AM -0400 schrieb Frank Li:
> On Tue, Jun 10, 2025 at 08:07:56AM +0200, Alexander Dahl wrote:
> > The i.MX 8DualXLite/8SoloXLite has a different connectivity memory map
> > than the generic i.MX8 has.  One conflicting resource is usb, where the
> > imx8dxl has a second usb2 phy @5b110000, while the generic imx8 dtsi has
> > one usb2 phy and one usb3 phy, and the usb3otg @5b110000.  When
> > including both imx8dxl-ss-conn.dtsi and imx8-ss-conn.dtsi as done in
> > imx8dxl.dtsi this leads to a duplicate unit-address warning.
> >
> > The usb3otg node was introduced after the initial imx8dxl support with
> > commit a8bd7f155126 ("arm64: dts: imx8qxp: add cadence usb3 support")
> > and since then leads to warnings like this (when building with ptxdist):
> >
> >     CPP generic/platform-v8a/build-target/linux-6.12/arch/arm64/boot/dts/freescale/imx8dxl-evk.dts
> >     DTC generic/platform-v8a/packages/linux-6.12/boot/imx8dxl-evk.dtb
> >     /home/adahl/Work/bsp/thorsis/generic/platform-v8a/build-target/linux-6.12/arch/arm64/boot/dts/freescale/imx8-ss-conn.dtsi:148.24-182.4: Warning (unique_unit_address): /bus@...00000/usb@...10000: duplicate unit-address (also used in node /bus@...00000/usbphy@...10000)
> 
> why linux-6.12? maybe just your directory name. trim down
> "generic/platform-v8a/build-target/linux-6.12/" just keep kernel's related
> patch here.

Yes, just my directory name.  Sorry about that.
Same happens when building recent v6.16-rc1 or master.

> what command do you use? I have not met this warning.

This is ptxdist [1] building dtb files with its own recipe, not going
through kernel's make dtbs.  I could reproduce it on a clean kernel
tree however like this (with PATH and ARCH=arm64 set):

    make defconfig
    make W=2 dtbs

Output (stripped) is this in that case:

      DTC     arch/arm64/boot/dts/freescale/imx8dxl-evk.dtb
    …/arch/arm64/boot/dts/freescale/imx8-ss-conn.dtsi:148.24-182.4: Warning (unique_unit_address): /bus@...00000/usb@...10000: duplicate unit-address (also used in node /bus@...00000/usbphy@...10000)
    …/arch/arm64/boot/dts/freescale/imx8-ss-hsio.dtsi:52.23-81.4: Warning (unique_unit_address): /bus@...00000/pcie@...10000: duplicate unit-address (also used in node /bus@...00000/pcie-ep@...10000)
      also defined at …/arch/arm64/boot/dts/freescale/imx8dxl-ss-hsio.dtsi:41.23-50.4
      also defined at …/arch/arm64/boot/dts/freescale/imx8dxl-evk.dts:645.8-653.3

These warnings do not show up with W=1.

So why do I get this?  kernel build calls dtc with
-Wno-unique_unit_address or not depending on warning level.
ptxdist does not call dtc with that option, so those warnings are not
silenced (details for the curious at [2]).

Do we fix issues coming up with W=2?
If yes I would adapt the commit message and send a v2.

Greets
Alex

[1] https://www.ptxdist.org/
[2] https://git.pengutronix.de/cgit/ptxdist/tree/scripts/lib/ptxd_make_world_dtb.sh