On Fri, Jul 07, 2023 at 01:22:56AM +0200, Konrad Dybcio wrote:
>
> On 6.07.2023 23:10, Rob Clark wrote:
> > From: Rob Clark <robdclark@xxxxxxxxxxxx>
> >
> > Even in the ocmem case, the allocated ocmem buffer size should match the
> > requested size.
> >
> > Signed-off-by: Rob Clark <robdclark@xxxxxxxxxxxx>
> > ---
> [...]
>
> > +
> > + WARN_ON(ocmem_hdl->len != adreno_gpu->info->gmem);
> I believe this should be an error condition. If the sizes are mismatched,
> best case scenario you get suboptimal perf and worst case scenario your
> system explodes.
No, the worst case scenarios are subtle bugs like random corruptions,
pagefaults etc which you debug for months. ;)
-Akhil.
>
> Very nice cleanup though!
>
> Konrad
> >
> > return 0;
> > }
> > @@ -1097,7 +1098,6 @@ int adreno_gpu_init(struct drm_device *drm, struct platform_device *pdev,
> >
> > adreno_gpu->funcs = funcs;
> > adreno_gpu->info = adreno_info(config->rev);
> > - adreno_gpu->gmem = adreno_gpu->info->gmem;
> > adreno_gpu->revn = adreno_gpu->info->revn;
> > adreno_gpu->rev = *rev;
> >
> > diff --git a/drivers/gpu/drm/msm/adreno/adreno_gpu.h b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
> > index 6830c3776c2d..aaf09c642dc6 100644
> > --- a/drivers/gpu/drm/msm/adreno/adreno_gpu.h
> > +++ b/drivers/gpu/drm/msm/adreno/adreno_gpu.h
> > @@ -77,7 +77,6 @@ struct adreno_gpu {
> > struct msm_gpu base;
> > struct adreno_rev rev;
> > const struct adreno_info *info;
> > - uint32_t gmem; /* actual gmem size */
> > uint32_t revn; /* numeric revision name */
> > uint16_t speedbin;
> > const struct adreno_gpu_funcs *funcs;
