Re: [PATCH v6 05/11] drm/msm/a6xx: Implement preemption for a7xx targets

[Antonino Maniscalco]

On 9/27/24 6:29 PM, Rob Clark wrote:
> On Thu, Sep 26, 2024 at 2:17 PM Antonino Maniscalco
> <antomani103@xxxxxxxxx> wrote:
> > This patch implements preemption feature for A6xx targets, this allows
> > the GPU to switch to a higher priority ringbuffer if one is ready. A6XX
> > hardware as such supports multiple levels of preemption granularities,
> > ranging from coarse grained(ringbuffer level) to a more fine grained
> > such as draw-call level or a bin boundary level preemption. This patch
> > enables the basic preemption level, with more fine grained preemption
> > support to follow.
> >
> > Reviewed-by: Akhil P Oommen <quic_akhilpo@xxxxxxxxxxx>
> > Tested-by: Neil Armstrong <neil.armstrong@xxxxxxxxxx> # on SM8650-QRD
> > Tested-by: Neil Armstrong <neil.armstrong@xxxxxxxxxx> # on SM8550-QRD
> > Tested-by: Neil Armstrong <neil.armstrong@xxxxxxxxxx> # on SM8450-HDK
> > Signed-off-by: Sharat Masetty <smasetty@xxxxxxxxxxxxxx>
> > Signed-off-by: Antonino Maniscalco <antomani103@xxxxxxxxx>
> > ---
> >   drivers/gpu/drm/msm/Makefile              |   1 +
> >   drivers/gpu/drm/msm/adreno/a6xx_gpu.c     | 283 +++++++++++++++++++++-
> >   drivers/gpu/drm/msm/adreno/a6xx_gpu.h     | 168 +++++++++++++
> >   drivers/gpu/drm/msm/adreno/a6xx_preempt.c | 377 ++++++++++++++++++++++++++++++
> >   drivers/gpu/drm/msm/msm_ringbuffer.h      |   7 +
> >   5 files changed, 825 insertions(+), 11 deletions(-)
>
> [snip]
>
> > diff --git a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
> > index 6e065500b64d6d95599d89c33e6703c92f210047..355a3e210335d60a5bed0ee287912271c353402a 100644
> > --- a/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
> > +++ b/drivers/gpu/drm/msm/adreno/a6xx_gpu.c
>
> [snip]
>
> > @@ -784,6 +1015,16 @@ static int a6xx_ucode_load(struct msm_gpu *gpu)
> >                  msm_gem_object_set_name(a6xx_gpu->shadow_bo, "shadow");
> >          }
> >
> > +       a6xx_gpu->pwrup_reglist_ptr = msm_gem_kernel_new(gpu->dev, PAGE_SIZE,
> > +                                                        MSM_BO_WC  | MSM_BO_MAP_PRIV,
> > +                                                        gpu->aspace, &a6xx_gpu->pwrup_reglist_bo,
> > +                                                        &a6xx_gpu->pwrup_reglist_iova);
>
> Couldn't this also be MSM_BO_GPU_READONLY?  And same for the
> preempt_postamble_ptr, I assume?

So for pwrup_reglist_ptr, it contains the cpu_gpu_lock struct that needs 
to be written by the gpu. The list itself is part of that struct.

> Also, it looks like we are only writing the smmu_info from the
> kernel.. is there any way that could be split out into a RO buffer?

For preempt_postamble_ptr, and smmu_info I think it should be possible 
so I'll work on that.

> BR,
> -R
>
> > +
> > +       if (IS_ERR(a6xx_gpu->pwrup_reglist_ptr))
> > +               return PTR_ERR(a6xx_gpu->pwrup_reglist_ptr);
> > +
> > +       msm_gem_object_set_name(a6xx_gpu->pwrup_reglist_bo, "pwrup_reglist");
> > +
> >          return 0;
> >   }


Best regards,
--
Antonino Maniscalco <antomani103@xxxxxxxxx>