Quoting Sebastian Brzezinka (2025-07-18 12:28:11)
> The value 4095 is intended to represent PAGE_SIZE - 1, as stated
> in the commit that introduced this change:
> commit 5f22cc0b134a ("drm/i915: Fix mismatch between misplaced
>                       vma check and vma insert")

No, that is exactly the misconception I wanted to avoid.

The core kernel's system PAGE_SIZE is not necessarily the same as the
paging boundary used by the GTT.
-Chris

Reply via email to