[Intel-gfx] [PATCH 3/7] drm/i915/fb: Factor out functions to remap contiguous FB obj pages

[Matthew Auld]

On Tue, 26 Oct 2021 at 23:51, Imre Deak <imre.deak at intel.com> wrote:
>
> Factor out functions needed to map contiguous FB obj pages to a GTT/DPT
> VMA view in the next patch.
>
> No functional changes.
>
> Signed-off-by: Imre Deak <imre.deak at intel.com>
> ---
>  drivers/gpu/drm/i915/gt/intel_ggtt.c | 110 +++++++++++++++------------
>  1 file changed, 60 insertions(+), 50 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/gt/intel_ggtt.c b/drivers/gpu/drm/i915/gt/intel_ggtt.c
> index 57c97554393b9..a30366d4965b6 100644
> --- a/drivers/gpu/drm/i915/gt/intel_ggtt.c
> +++ b/drivers/gpu/drm/i915/gt/intel_ggtt.c
> @@ -1387,6 +1387,25 @@ intel_rotate_pages(struct intel_rotation_info *rot_info,
>         return ERR_PTR(ret);
>  }
>
> +static struct scatterlist *
> +add_padding_pages(unsigned int count,
> +                 struct sg_table *st, struct scatterlist *sg)
> +{
> +       st->nents++;
> +
> +       /*
> +        * The DE ignores the PTEs for the padding tiles, the sg entry
> +        * here is just a convenience to indicate how many padding PTEs
> +        * to insert at this spot.
> +        */
> +       sg_set_page(sg, NULL, count * 4096, 0);

s/4096/I915_GTT_PAGE_SIZE ?

> +       sg_dma_address(sg) = 0;

I guess maybe a little bit scary, since that might be a valid address.
Using the vma->vm scratch might be annoying though, since it could be
a different type than the object, plus this is only the GGTT.

Looks fine I think,
Reviewed-by: Matthew Auld <matthew.auld at intel.com>