Re: [PATCH v2 0/2] drm/i915/cx0: TCSS power for the non-TC C20 PHY on PTL port B

[Imre Deak <imre.deak@intel.com>]

Hi Burak,

thanks for the report, root-causing and reasoning. The actual issue is
misdetecting the PHY type for port B on some platforms with a second eDP
panel connected. That misdetection should be fixed by

https://lore.kernel.org/all/20261029113017.618286-7-imre.deak@intel.com

Could you please give a try to the latest drm-tip kernel, where the
above fix is merged, and provide a dmesg log with that, if possible
attaching the log to the

https://gitlab.freedesktop.org/drm/xe/kernel/-/work_items/9196

ticket?

Thanks,
Imre

On Thu, Oct 15, 2026 at 01:26:39AM +0300, Burak Göncü via B4 Relay wrote:
> On PTL, port B is driven by a C20 PHY that sits in the Type-C
> subsystem, but nothing in the driver requests TCSS power for it; only
> the firmware does, when it lights a display on the port at boot. The
> ASUS Zenbook Duo UX8407AA drives its second internal panel from port B.
> After a power-on with the detachable keyboard lying on that panel, the
> first enable of the panel fails and can hang the machine (gitlab issue
> 9196).
> 
> Patch 2 requests TCSS power for such a PHY around its PLL enable and
> disable. Patch 1 first serializes the TCSS mailbox handshake of
> Wa_14020908590, which patch 2 reuses and which the TC ports could
> already race on.
> 
> Two questions from v1 are still open:
> 
> - Does port B's C20 PHY on PTL always need the TCSS power request when
>   it is used as a non-TC port, as this series assumes, or does this
>   belong in the port's power well / AUX handling, or in a board quirk?
> - Should the request be held for as long as the port is in use rather
>   than per PLL enable/disable? Releasing it while the panel is off puts
>   the hardware back into the state an unpatched kernel has; holding it
>   showed no measurable difference in package power or PC10/LPI
>   residency on this machine.
> 
> EDITME: v2 test results on the hardware.
> 
> Compile-tested on drm-tip with W=1 for i915 and xe (clang). HDMI/DP on
> port B could not be tested, as this board only drives eDP there.
> 
> An LLM-based assistant helped with the investigation and with writing
> the patches; I tested them on the hardware and reviewed them.
> 
> ---
> Changes in v2:
> - New patch 1: serialize the TCSS mailbox with a display-wide mutex; it
>   was covered only by each TC port's own lock (Sashiko review)
> - Patch 2: also wait for the TCSS power state to clear when the request
>   is released, as the TC PHY code does (Sashiko review)
> - Drop the RFC prefix
> - Link to v1: https://patch.msgid.link/20261006-ptl-port-b-tcss-v1-1-240670342cd3@gmail.com
> 
> ---
> Burak Göncü (2):
>       drm/i915/tc: serialize the TCSS power request mailbox
>       drm/i915/cx0: request TCSS power for the non-TC C20 PHY on PTL port B
> 
>  drivers/gpu/drm/i915/display/intel_cx0_phy.c       | 37 ++++++++++++++++++++++
>  drivers/gpu/drm/i915/display/intel_display_core.h  |  5 +++
>  .../gpu/drm/i915/display/intel_display_driver.c    |  1 +
>  drivers/gpu/drm/i915/display/intel_tc.c            | 33 +++++++++++++++++--
>  drivers/gpu/drm/i915/display/intel_tc.h            |  1 +
>  5 files changed, 74 insertions(+), 3 deletions(-)
> ---
> base-commit: daeb062fc40f727fafefed080c6d12a5ad1d0d12
> change-id: 20261006-ptl-port-b-tcss-8b1cbc5d98e0
> 
> Best regards,
> --  
> Burak Göncü <burakgon12@gmail.com>
> 
>