Bug fixed.

1. Make link for ReduceSumAll function.
2. Fix the memory leak issue.

source/network/XBackwardMath.cpp

@@ -1398,11 +1398,32 @@ void XMathGrad::GradReduceSum(XTensor * node, bool isEfficient)
 /*
 gradient for reduceSumAll
 for
-TODO
+c = reduceSumAll(a)
+we have
+dE/da = dE/dc * 1
+
+>> node - the node (c) for backward computation
+>> isEfficient - indicates whether the computation is in
+                 an efficient manner
 */
 void XMathGrad::GradReduceSumAll(XTensor * node, bool isEfficient)
 {
-    ShowNTErrors("TODO!");
+    XLink &income = node->income;
+    CheckNTErrors(income.tailNum == 1, "Wrong input tensor number for Reduce!");
+
+    XTensor * a = income.tails[0];
+    XTensor * b = NewTensorBufV2(a, a->devID, a->mem);
+
+    XNoder::MakeGrad(a);
+
+    DTYPE value = node->grad->Get0D();
+    _SetDataFixed(b, (void*)&value);
+
+    _Sum(a->grad, b, a->grad);
+
+    DelTensorBuf(b);
+
+    node->visitMark = NODE_FINISHED;
 }
 
 /*

source/tensor/core/reduce/ReduceSumAll.cpp

@@ -79,6 +79,8 @@ void _ReduceSumAll(const XTensor * source, DTYPE * value)
     /* call _ReduceSum function */
     _ReduceSumAll(source, target);
     *value = target->Get0D();
+
+    DelTensorBuf(target);
 }
 
 /*

source/tensor/test/TReduceSumAll.cpp

@@ -71,7 +71,7 @@ bool TestReduceSumAll1()
     sGPU->SetData(sData, sUnitNum);
 
     /* call ReduceSumAll function */
-    summation = _ReduceSumAll(sGPU);
+    summation = ReduceSumAllValue(*sGPU);
 
     /* check results */
     gpuTest = (fabs(answer - summation) < 1e-4F);