[Piglit] [PATCH] cl: Add test opencv-merge-hist

Tom Stellard tom at stellard.net
Fri Oct 11 19:09:30 CEST 2013


From: Tom Stellard <thomas.stellard at amd.com>

---
 tests/cl/program/execute/opencv-merge-hist.cl | 97 +++++++++++++++++++++++++++
 1 file changed, 97 insertions(+)
 create mode 100644 tests/cl/program/execute/opencv-merge-hist.cl

diff --git a/tests/cl/program/execute/opencv-merge-hist.cl b/tests/cl/program/execute/opencv-merge-hist.cl
new file mode 100644
index 0000000..b96b1b0
--- /dev/null
+++ b/tests/cl/program/execute/opencv-merge-hist.cl
@@ -0,0 +1,97 @@
+// The kernel in this test is taken from the opencv library (opencv.org) and
+// has the following license:
+//
+//                           License Agreement
+//                For Open Source Computer Vision Library
+//
+// Copyright (C) 2010-2012, Institute Of Software Chinese Academy Of Science, all rights reserved.
+// Copyright (C) 2010-2012, Advanced Micro Devices, Inc., all rights reserved.
+// Copyright (C) 2010-2012, Multicoreware, Inc., all rights reserved.
+// Third party copyrights are property of their respective owners.
+//
+// @Authors
+//    Niko Li, newlife20080214 at gmail.com
+//    Jia Haipeng, jiahaipeng95 at gmail.com
+//    Xu Pang, pangxu010 at 163.com
+//    Wenju He, wenju at multicorewareinc.com
+// Redistribution and use in source and binary forms, with or without modification,
+// are permitted provided that the following conditions are met:
+//
+//   * Redistribution's of source code must retain the above copyright notice,
+//     this list of conditions and the following disclaimer.
+//
+//   * Redistribution's in binary form must reproduce the above copyright notice,
+//     this list of conditions and the following disclaimer in the documentation
+//     and/or other GpuMaterials provided with the distribution.
+//
+//   * The name of the copyright holders may not be used to endorse or promote products
+//     derived from this software without specific prior written permission.
+//
+// This software is provided by the copyright holders and contributors as is and
+// any express or implied warranties, including, but not limited to, the implied
+// warranties of merchantability and fitness for a particular purpose are disclaimed.
+// In no event shall the Intel Corporation or contributors be liable for any direct,
+// indirect, incidental, special, exemplary, or consequential damages
+// (including, but not limited to, procurement of substitute goods or services;
+// loss of use, data, or profits; or business interruption) however caused
+// and on any theory of liability, whether in contract, strict liability,
+// or tort (including negligence or otherwise) arising in any way out of
+// the use of this software, even if advised of the possibility of such damage.
+//
+//
+
+// This test checks that the loop:
+//  for(int stride = HISTOGRAM256_WORK_GROUP_SIZE /2; stride > 0; stride >>= 1)
+// in this kernel is unrolled correctly or not unrolled at all.  The Clang
+// OpenCL frontend was unrolling this in way that created illegal uses of the
+// barrier() builtin which resulted in GPU hangs on r600g.
+/*!
+[config]
+name: OpenCV merge-hist
+kernel_name: merge_hist
+dimensions: 1
+global_size: 65536 1 1
+local_size: 256 1 1
+
+[test]
+arg_in:  0 buffer int[65536] repeat 0
+arg_out: 1 buffer int[256]   repeat 0
+arg_in:  2 int 256
+
+!*/
+
+#define PARTIAL_HISTOGRAM256_COUNT     (256)
+#define HISTOGRAM256_BIN_COUNT         (256)
+
+#define HISTOGRAM256_WORK_GROUP_SIZE     (256)
+#define HISTOGRAM256_LOCAL_MEM_SIZE      (HISTOGRAM256_BIN_COUNT)
+
+#define NBANKS (16)
+#define NBANKS_BIT (4)
+
+
+__kernel __attribute__((reqd_work_group_size(256,1,1)))void merge_hist(__global int* buf,
+                __global int* hist,
+                int src_step)
+{
+    int lx = get_local_id(0);
+    int gx = get_group_id(0);
+
+    int sum = 0;
+
+    for(int i = lx; i < PARTIAL_HISTOGRAM256_COUNT; i += HISTOGRAM256_WORK_GROUP_SIZE)
+        sum += buf[ mad24(i, src_step, gx)];
+
+    __local int data[HISTOGRAM256_WORK_GROUP_SIZE];
+    data[lx] = sum;
+
+    for(int stride = HISTOGRAM256_WORK_GROUP_SIZE /2; stride > 0; stride >>= 1)
+    {
+        barrier(CLK_LOCAL_MEM_FENCE);
+        if(lx < stride)
+            data[lx] += data[lx + stride];
+    }
+
+    if(lx == 0)
+        hist[gx] = data[0];
+}
-- 
1.7.11.4



More information about the Piglit mailing list