[Beignet] [PATCH 1/3] return 32 could gain 0.2% performance on opencv optical flow case.
xionghu.luo at intel.com
xionghu.luo at intel.com
Wed Jul 22 19:54:12 PDT 2015
From: Yang Rong <rong.r.yang at intel.com>
Signed-off-by: Luo Xionghu<xionghu.luo at intel.com>
---
src/cl_gt_device.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/src/cl_gt_device.h b/src/cl_gt_device.h
index 4b43c20..26bb8e8 100644
--- a/src/cl_gt_device.h
+++ b/src/cl_gt_device.h
@@ -38,7 +38,7 @@
.native_vector_width_float = 4,
.native_vector_width_double = 2,
.native_vector_width_half = 8,
-.preferred_wg_sz_mul = 16,
+.preferred_wg_sz_mul = 32,
.address_bits = 32,
.max_mem_alloc_size = 512 * 1024 * 1024,
.image_support = CL_TRUE,
--
1.9.1
More information about the Beignet
mailing list