Quoting Kuogee Hsieh (2022-08-24 10:22:31)
> At current implementation there is an extra 0 at 1.62G link rate which cause
> no correct pixel_div selected for 1.62G link rate to calculate mvid and nvid.
> This patch delete the extra 0 to have mvid and nvid be calculated correctly.
>
> Changes in v2:
> -- fix Fixes tag's text
>
> Fixes: 937f941ca06f  ("drm/msm/dp: Use qmp phy for DP PLL and PHY")
> Signed-off-by: Kuogee Hsieh <quic_khs...@quicinc.com>
>
> Reviewed-by: Stephen Boyd <swb...@chromium.org>
> eviewed-by: Abhinav Kumar <quic_abhin...@quicinc.com>

Missing an R.

>
> diff --git a/drivers/gpu/drm/msm/dp/dp_catalog.c 
> b/drivers/gpu/drm/msm/dp/dp_catalog.c
> index 7257515..676279d 100644
> --- a/drivers/gpu/drm/msm/dp/dp_catalog.c
> +++ b/drivers/gpu/drm/msm/dp/dp_catalog.c
> @@ -431,7 +431,7 @@ void dp_catalog_ctrl_config_msa(struct dp_catalog 
> *dp_catalog,
>
>         if (rate == link_rate_hbr3)
>                 pixel_div = 6;
> -       else if (rate == 1620000 || rate == 270000)
> +       else if (rate == 162000 || rate == 270000)

BTW

  $ git grep 162000 -- drivers/gpu/drm/msm/dp/
  drivers/gpu/drm/msm/dp/dp_catalog.c:    else if (rate == 1620000 ||
rate == 270000)
  drivers/gpu/drm/msm/dp/dp_ctrl.c:
ctrl->link->link_params.rate = 162000;
  drivers/gpu/drm/msm/dp/dp_ctrl.c:       case 162000:

We really should have a #define for these things. Can you send a
different followup patch that makes something like HBR3_KHZ, HBR2_KHZ,
etc. and then replaces these bare numbers throughout the driver? That
would prevent this from happening again. This one liner can and should
be backported to stable trees.

Reply via email to