[Beignet] [PATCH] Add workgroup reduce add optimization

[Grigore Lupescu]

Signed-off-by: Grigore Lupescu <grigore.lupescu at intel.com>
---
 backend/src/backend/gen_context.cpp | 33 +++++++++++++++++++++------------
 1 file changed, 21 insertions(+), 12 deletions(-)

diff --git a/backend/src/backend/gen_context.cpp b/backend/src/backend/gen_context.cpp
index 0ea0dd0..3370a7d 100644
--- a/backend/src/backend/gen_context.cpp
+++ b/backend/src/backend/gen_context.cpp
@@ -2943,21 +2943,31 @@ namespace gbe
           }
         }
       }
-    } else if (wg_op == ir::WORKGROUP_OP_REDUCE_ADD) {
+    }
+    else if (wg_op == ir::WORKGROUP_OP_REDUCE_ADD){
+      tmp.hstride = GEN_HORIZONTAL_STRIDE_1;
+      tmp.vstride = GEN_VERTICAL_STRIDE_4;
+      tmp.width = GEN_WIDTH_4;
+
       GBE_ASSERT(tmp.type == theVal.type);
-      GenRegister v = GenRegister::toUniform(tmp, theVal.type);
-      for (uint32_t i = 0; i < simd; i++) {
-        p->ADD(threadData, threadData, v);
-        v.subnr += typeSize(theVal.type);
-        if (v.subnr == 32) {
-          v.subnr = 0;
-          v.nr++;
-        }
+      GenRegister partialSum = tmp;
+
+      /* adjust offset, compute add with ADD4/ADD */
+      for (uint32_t i = 1; i < simd/4; i++){
+        tmp = tmp.suboffset(tmp, 4);
+        p->push();
+        p->curr.execWidth = GEN_WIDTH_16;
+        p->ADD(partialSum, partialSum, tmp);
+        p->pop();
       }
-    }
 
+      for (uint32_t i = 0; i < 4; i++){
+        p->ADD(threadData, threadData, partialSum);
+        partialSum = GenRegister::suboffset(partialSum, 1);
+      }
+    }
     p->pop();
-  }
+}
 
 #define SEND_RESULT_MSG() \
 do { \
@@ -3123,7 +3133,6 @@ do { \
         p->curr.predicate = GEN_PREDICATE_NONE;
         p->WAIT(2);
         p->patchJMPI(jip, (p->n_instruction() - jip), 0);
-    
         /* Do something when get the msg. */
         p->curr.execWidth = simd;
         p->MOV(dst, msgData);
-- 
2.1.4