On Mon, Jan 12, 2026 at 03:25:05PM +0800, Xilin Wu wrote:
> On 1/12/2026 11:11 AM, Dmitry Baryshkov wrote:
> > From: Jessica Zhang <[email protected]>
> > 
> > LM block doesn't have a hardware buffer (unlike PINGPONG and DSC
> > encoders). As such, don't use ephemeral max_mixer_width and
> > MAX_HDISPLAY_SPLIT to validate requested modes. Instead use PP and DSC
> > buffer widths.
> > 
> > While on the DPU 8.x+ supports a max linewidth of 8960 for PINGPONG_0,
> > there is some additional logic that needs to be added to the resource
> > manager to specifically try and reserve PINGPONG_0 for modes that are
> > greater than 5k.
> > 
> > Signed-off-by: Jessica Zhang <[email protected]>
> > Tested-by: Xilin Wu <[email protected]> # qcs6490-radxa-dragon-q6a
> > [DB: reworked to drop catalog changes, updated commit message]
> > Signed-off-by: Dmitry Baryshkov <[email protected]>
> 
> Sorry for the late reply, my colleagues are still testing the new series on
> qcs6490.
> 
> However, this completely breaks 4K 120Hz display on SC8280XP CRD, which was
> previously functional (albeit with the clock check bypassed [1]). The
> display now shows a solid blue screen. Kernel logs indicate that only one
> layer mixer is being used, instead of the two that were used previously.

Could you please post debugfs/dri/0/state and debugfs/dri/0/debug/core_perf/*?

> 
> [1] 
> https://lore.kernel.org/all/[email protected]/
> 
> -- 
> Best regards,
> Xilin Wu <[email protected]>

-- 
With best wishes
Dmitry

Reply via email to