Thank you for your contribution! Sashiko AI review found 1 potential issue(s) 
to consider:
- [Low] The patch removes the validation from the integrated graphics path, 
rendering the `has_lmembar()` logic completely dead and meaningless.
--

commit b290f7e32abe43d35ac4f5def3b10e9bcf641a88
Author: Andrew Zhou <[email protected]>

drm/xe/display: Remove invalid initial plane PTE check for iGPU

This commit removes the initial plane PTE local memory validation for
integrated graphics, as Pantherlake can have the XE_GGTT_PTEDM bit set.
The check was causing a failure to create the initial framebuffer.

> diff --git a/drivers/gpu/drm/xe/display/xe_initial_plane.c 
> b/drivers/gpu/drm/xe/display/xe_initial_plane.c
> index 0f86b73036d03..8c95dc4e6f2ef 100644
> --- a/drivers/gpu/drm/xe/display/xe_initial_plane.c
> +++ b/drivers/gpu/drm/xe/display/xe_initial_plane.c

[ ... ]

> @@ -93,13 +92,6 @@ initial_plane_bo(struct xe_device *xe,
>                       return NULL;
>               }
>  
> -             pte = xe_ggtt_read_pte(tile0->mem.ggtt, base);
> -
> -             if (is_pte_local(pte) != need_pte_local(xe)) {
> -                     drm_err(&xe->drm, "Initial plane PTE has bad local 
> memory bit\n");
> -                     return NULL;
> -             }
> -

[Severity: Low]
Does this leave need_pte_local() with dead logic?

By removing this validation from the integrated graphics path (!IS_DGFX(xe)),
the only remaining caller of need_pte_local() is inside the earlier 
if (IS_DGFX(xe)) branch in initial_plane_bo().

Because need_pte_local() is defined as:

static bool need_pte_local(struct xe_device *xe)
{
        return IS_DGFX(xe) || has_lmembar(xe);
}

It will now unconditionally evaluate to true when called, rendering the
has_lmembar(xe) check completely meaningless. 

Should need_pte_local() and its underlying logic be simplified or removed 
as well to prevent confusion?

>               phys_base = base;
>               flags |= XE_BO_FLAG_STOLEN;
>

-- 
Sashiko AI review ยท 
https://sashiko.dev/#/patchset/[email protected]?part=1

Reply via email to