[PATCH i-g-t v8 29/33] lib/intel_batchbuffer: Don't decanonize address, to be squashed.

Zbigniew Kempczyński zbigniew.kempczynski at intel.com
Mon Oct 19 15:59:29 UTC 2020


To be consistent with intel_buf address returning behavior.

Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski at intel.com>
---
 lib/intel_batchbuffer.c | 8 +-------
 1 file changed, 1 insertion(+), 7 deletions(-)

diff --git a/lib/intel_batchbuffer.c b/lib/intel_batchbuffer.c
index 41e82cde..8311fec8 100644
--- a/lib/intel_batchbuffer.c
+++ b/lib/intel_batchbuffer.c
@@ -2465,7 +2465,6 @@ uint64_t intel_bb_get_object_offset(struct intel_bb *ibb, uint32_t handle)
 {
 	struct drm_i915_gem_exec_object2 object = { .handle = handle };
 	struct drm_i915_gem_exec_object2 **found;
-	uint64_t address;
 
 	igt_assert(ibb);
 
@@ -2473,12 +2472,7 @@ uint64_t intel_bb_get_object_offset(struct intel_bb *ibb, uint32_t handle)
 	if (!found)
 		return INTEL_BUF_INVALID_ADDRESS;
 
-	address = (*found)->offset;
-
-	if (address == INTEL_BUF_INVALID_ADDRESS)
-		return address;
-
-	return gen8_canonical_addr(address & (ibb->gtt_size - 1));
+	return (*found)->offset;
 }
 
 /**
-- 
2.26.0



More information about the Intel-gfx-trybot mailing list