[Intel-gfx] [PATCH] drm/i915: Avoid overflow in computing pot_hole loop termination

[Chris Wilson]

When using the mock_ppgtt selftest, the GTT is large enough to cause an
overflow in pot_hole() when adding 2 pages to the address. Avoid the
overflow by computing the final valid address and iterating up to that
address.

Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
Cc: Matthew Auld <matthew.auld at intel.com>
---
 drivers/gpu/drm/i915/selftests/i915_gem_gtt.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
index f1056faa9a6a..3d54e7d81b9d 100644
--- a/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
+++ b/drivers/gpu/drm/i915/selftests/i915_gem_gtt.c
@@ -615,7 +615,7 @@ static int pot_hole(struct drm_i915_private *i915,
 		u64 addr;
 
 		for (addr = round_up(hole_start + I915_GTT_PAGE_SIZE, step) - I915_GTT_PAGE_SIZE;
-		     addr + vma->size <= hole_end;
+		     addr <= round_down(hole_end - 2*I915_GTT_PAGE_SIZE, step) - I915_GTT_PAGE_SIZE;
 		     addr += step) {
 			err = i915_vma_pin(vma, 0, 0, addr | flags);
 			if (err) {
-- 
2.11.0