[PATCH v6 03/14] drm/msm/dpu: switch RM to use crtc_id rather than enc_id for allocation

[Dmitry Baryshkov]

On Fri, Feb 14, 2025 at 04:14:26PM -0800, Jessica Zhang wrote:
> From: Dmitry Baryshkov <dmitry.baryshkov at linaro.org>
> 
> Up to now the driver has been using encoder to allocate hardware
> resources. Switch it to use CRTC id in preparation for the next step.
> 
> Reviewed-by: Abhinav Kumar <quic_abhinavk at quicinc.com>
> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov at linaro.org>
> Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov at linaro.org>
> Signed-off-by: Jessica Zhang <quic_jesszhan at quicinc.com>
> ---
> Changes in v6:
> - Drop duplicate cstate initialization code and unnecessary memset
> Changes in v5:
> - Reordered to prevent breaking CI and upon partial application
> 
> Changes in v4 (due to rebase):
> - moved *_get_assigned_resources() changes for DSPP and LM from
>   encoder *_virt_atomic_mode_set() to *_assign_crtc_resources()
> ---
>  drivers/gpu/drm/msm/disp/dpu1/dpu_crtc.c    |  18 +--
>  drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c |  10 +-
>  drivers/gpu/drm/msm/disp/dpu1/dpu_kms.h     |  12 +-
>  drivers/gpu/drm/msm/disp/dpu1/dpu_rm.c      | 189 ++++++++++++++--------------
>  drivers/gpu/drm/msm/disp/dpu1/dpu_rm.h      |   7 +-
>  5 files changed, 110 insertions(+), 126 deletions(-)

This commit breaks several tests in CI:
- sc7180-trogdor-kingoftown:
  - kms_cursor_crc at cursor-dpms
  - kms_pipe_crc_basic at disable-crc-after-crtc
- sc7180-trogdor-lazor-limozeen
  - kms_cursor_crc at cursor-dpms
  - kms_pipe_crc_basic at disable-crc-after-crtc

Corresponding pipeline is available at [1]

As I had to rebase your changes on top of msm-next, corresponding tree
is available at [2]. It might be possible that the regression is
introduced by my rebase.

[1] https://gitlab.freedesktop.org/drm/msm/-/pipelines/1374165

[2] https://gitlab.freedesktop.org/lumag/msm/-/commits/msm-next-lumag-cwb

-- 
With best wishes
Dmitry