On 2023-06-22 22:47:04, Abhinav Kumar wrote:
> On 6/22/2023 6:37 PM, Dmitry Baryshkov wrote:
> > All DSC_BLK_1_2 declarations incorrectly pass 0x29c as the block length.
> > This includes the common block itself, enc subblocks and some empty
> > space around. Change that to pass 0x4 instead, the length of common
> > register block itself.
> >
> > Fixes: 0d1b10c63346 ("drm/msm/dpu: add DSC 1.2 hw blocks for relevant chipsets")
> > Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@xxxxxxxxxx>
>
> There is no need of a fixes tag for this.
>
> This is not a bug but was intentional.
>
> Till we added sub-block parsing support we had to dump the whole block.
>
> And hence I would suggest this change should be merged after the
> sub-block parsing change otherwise we wont have full register dumps for DSC.
This was indeed intentional, we discussed it in [1].
In fact I asked to make it 0xf00 + 0x10 or 0xf80 + 0x10 to also cover
the CTL registers, but that change didn't make it through. 0x29c is an
arbitrary number that I have no clue what it was based on.
[1]: https://lore.kernel.org/linux-arm-msm/y2whfntyo2rbrg3taazjdw5sijle6k6swzl4uutcxm6tmuayh4@uxdur74uasua/
- Marijn
