On Fri, 12 Jul 2024 at 22:41, Abhinav Kumar <quic_abhinavk@xxxxxxxxxxx> wrote:
> On 6/24/2024 2:13 PM, Dmitry Baryshkov wrote:
> > The commit b954fa6baaca ("drm/msm/dpu: Refactor rm iterator") removed
> > zero-init of the hw_ctl array, but didn't change the error condition,
> > that checked for hw_ctl[i] being NULL. Use indices check instead.
> >
> > Fixes: b954fa6baaca ("drm/msm/dpu: Refactor rm iterator")
> > Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@xxxxxxxxxx>
> > ---
> > drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
> > index 5d205e09cf45..7613005fbfea 100644
> > --- a/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
> > +++ b/drivers/gpu/drm/msm/disp/dpu1/dpu_encoder.c
> > @@ -1186,7 +1186,7 @@ static void :tag(struct drm_encoder *drm_enc,
> > return;
> > }
> >
> > - if (!hw_ctl[i]) {
> > + if (i >= num_ctl) {
>
> This is not very clear to me.
>
> How will we hit this condition? I dont see i going beyond 1 in this loop
> and neither should num_ctl
Why? the driver doesn't support flushing through a single CTL, so
num_ctl = num_intf.
>
> Will it be just easier to bring back the NULL assignment at the top?
>
> struct dpu_hw_blk *hw_ctl[MAX_CHANNELS_PER_ENC] = { NULL };
>
> I also see the same issue for other blocks such as hw_dsc, hw_lm
Other blocks loop properly up to the num_resource. I'd prefer to drop
the NULL init from the DSPP init and use num_dspp instead.
>
> > DPU_ERROR_ENC(dpu_enc,
> > "no ctl block assigned at idx: %d\n", i);
> > return;
> >
--
With best wishes
Dmitry
