We can't mix 64K and 4K pte's in the same page-table, so for now we
align 64K objects to 2M to avoid any potential mixing. This is
potentially wasteful but in reality shouldn't be too bad since this only
applies to the virtual address space of a 48b PPGTT.
Suggested-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
Signed-off-by: Matthew Auld <matthew.auld@xxxxxxxxx>
Cc: Joonas Lahtinen <joonas.lahtinen@xxxxxxxxxxxxxxx>
---
drivers/gpu/drm/i915/i915_vma.c | 10 +++++++++-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/i915/i915_vma.c b/drivers/gpu/drm/i915/i915_vma.c
index cee1d00dc085..596269172cd2 100644
--- a/drivers/gpu/drm/i915/i915_vma.c
+++ b/drivers/gpu/drm/i915/i915_vma.c
@@ -495,7 +495,15 @@ i915_vma_insert(struct i915_vma *vma, u64 size, u64 alignment, u64 flags)
if (end > (1ULL << 32) &&
vma->page_sizes.sg > I915_GTT_PAGE_SIZE) {
u64 page_alignment =
- rounddown_pow_of_two(vma->page_sizes.sg);
+ rounddown_pow_of_two(vma->page_sizes.sg |
+ I915_GTT_PAGE_SIZE_2M);
+
+ /* We can't mix 64K and 4K PTEs in the same page-table (2M
+ * block), and so to avoid the ugliness and complexity of
+ * coloring we opt for just aligning 64K objects to 2M.
+ */
+ if (vma->page_sizes.sg & I915_GTT_PAGE_SIZE_64K)
+ size = round_up(size, I915_GTT_PAGE_SIZE_2M);
alignment = max(alignment, page_alignment);
}
--
2.9.4
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
