Re: [PATCH 2/2] drm/i915/display: Don't require a 1:1 pipe to port mapping for DC3co

[sashiko-bot@kernel.org]

Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] drm/i915/display: DC3CO accidentally disabled for DISPLAY_VER > 35
--

commit fe8edcc01c82523ff83cda52b4cdaab761f57f66
Author: Suraj Kandpal <suraj.kandpal@intel.com>

drm/i915/display: Don't require a 1:1 pipe to port mapping for DC3co

This commit relaxes the strict 1:1 pipe-to-port mapping requirement for DC3co
on Xe3p_LPD (DISPLAY_VER 35). It replaces the previous if/else conditional
with a simplified check to allow DC3co on eDP panels wired to non-matching
ports.

> diff --git a/drivers/gpu/drm/i915/display/intel_display_power.c b/drivers/gpu/drm/i915/display/intel_display_power.c
> index 2e534ddb73479..d475e62e8f40b 100644
> --- a/drivers/gpu/drm/i915/display/intel_display_power.c
> +++ b/drivers/gpu/drm/i915/display/intel_display_power.c
> @@ -428,18 +428,14 @@ void intel_display_power_dc3co_update(struct intel_display *display, u32 trigger
>  static bool intel_dc3co_port_pipe_compatible(struct intel_dp *intel_dp,
>  					     const struct intel_crtc_state *crtc_state)
>  {
> +	struct intel_display *display = to_intel_display(intel_dp);
>  	struct intel_digital_port *dig_port = dp_to_dig_port(intel_dp);
>  	enum pipe pipe = to_intel_crtc(crtc_state->uapi.crtc)->pipe;
>  	enum port port = dig_port->base.port;
>  	int num_pipes = intel_crtc_num_joined_pipes(crtc_state);
>  
> -	/* Need to follow 1:1 mapping because of CMTG restriction */
> -	if (DISPLAY_VER(to_intel_display(crtc_state)) == 35)
> -		return num_pipes == 1 &&
> -		       ((pipe == PIPE_A && port == PORT_A) ||
> -			(pipe == PIPE_B && port == PORT_B));
> -	else
> -		return num_pipes == 1 && pipe <= PIPE_B && port <= PORT_B;
> +	return DISPLAY_VER(display) == 35 && num_pipes == 1 && pipe <= PIPE_B &&
> +		port <= PORT_B;

[Severity: Medium]
Will this inadvertently disable DC3co for platforms where DISPLAY_VER is
greater than 35?

By replacing the entire if/else block with a single return statement that
explicitly checks DISPLAY_VER(display) == 35, it appears
intel_dc3co_port_pipe_compatible() will now unconditionally return false for
any future hardware with a version > 35.

Since the HAS_DC3CO() macro usually guards entry into DC3co paths for
platforms where DISPLAY_VER >= 35, does it make sense to drop the strict
version equality check here so that future platforms aren't unintentionally
restricted?

>  }

-- 
Sashiko AI review · https://sashiko.dev/#/patchset/20260924055734.2138787-1-suraj.kandpal@intel.com?part=2