linux-kernel - Re: [PATCH v2 2/2] arm64: dts: rockchip: add RK3588 DP carrier from Theobroma Systems

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 for Android: free password hash cracker in your pocket

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

Message-ID: <10574731.0AQdONaE2F@diego>
Date: Mon, 11 Aug 2025 09:19:29 +0200
From: Heiko Stübner <heiko@...ech.de>
To: Quentin Schulz <quentin.schulz@...rry.de>
Cc: robh@...nel.org, krzk+dt@...nel.org, conor+dt@...nel.org,
 devicetree@...r.kernel.org, linux-arm-kernel@...ts.infradead.org,
 linux-rockchip@...ts.infradead.org, linux-kernel@...r.kernel.org,
 damon.ding@...k-chips.com, Heiko Stuebner <heiko.stuebner@...rry.de>
Subject:
 Re: [PATCH v2 2/2] arm64: dts: rockchip: add RK3588 DP carrier from Theobroma
 Systems

Am Freitag, 25. Juli 2025, 14:29:12 Mitteleuropäische Sommerzeit schrieb Quentin Schulz:
> Hi Heiko,
> 
> On 7/23/25 9:09 PM, Heiko Stuebner wrote:
> > From: Heiko Stuebner <heiko.stuebner@...rry.de>
> > 
> > The DisplayPort carrier is a very simple baseboard only providing serial,
> > ethernet and a displayport output.
> > 
> > But its main functionality is that it routes the Analogix eDP controller
> > to this DisplayPort output, which allows to test that controller simply
> > by hooking it up to a suitable monitor.
> > 
> > The Analogix-DP controller supports eDP 1.3 and DP 1.2, so can drive
> > both eDP displays as well as full DP monitors. It does not support DP+
> > so passive DP-to-HDMI adapters won't work.
> > 
> 
> I tested this on master (2942242dde896) + v2 of the eDP driver as listed 
> in the cover letter, or with v3 of the eDP driver + 
> 48f05c3b4b701ae7687fd44d462c88b7ac67e952 and in both cases I have weird 
> behaviors.
> 
> First, `reboot` is stuck for a very long time before actually rebooting. 
> I think you have a stacktrace when you tried yourself, I don't so cannot 
> send one.

I tracked this down last week. The hang is easily reproduceable and the
fault of the PCIe SMMU. See
  https://lore.kernel.org/all/4400329.mogB4TqSGs@diego/
for the full detail.

TL;DR is, SMMU without users (and thus clocks/power-domains/or whatever)
hangs on reboot.


> Also, I tested on two different DP displays, one has a green tint, the 
> other one purple. When trying out other resolutions with modetest, the 
> kernel would often crash (no trace, just nothing on console, SSH dead). 
> Note that the same HW setup with downstream kernel from 
> https://git.theobroma-systems.com/tiger-linux.git/log/?h=linux-6.1-stan-rkr3.2-tiger 
> works just fine. I would assume this has nothing to do with this Device 
> Tree patch here but rather the eDP patches missing some bits maybe?
> 
> @Damon do you have some idea?
> 
> The display with green tint is a Iiyama ProLite XU2294HSU and the one 
> with the purple tint is a Dell P2319H if that helps.
> 
> modetest output: https://paste.debian.net/1387797/.
> modetest output with downstream kernel: https://paste.debian.net/1387798/
> 
> The Device Tree looks good to me otherwise.
> 
> Cheers,
> Quentin
>