COMPMID-417: Fix assert in GEMMTranspose

The assert was checking the wrong thing. Only if the window over the
input is smaller than the number of processed elements, the output shape
would be empty.

However, the valid region will be empty if the input's first dimension
is less than the number of elements processed. That required the changes
in TensorShape.

Change-Id: I36fed7893dfd502e26c5c776c9a2d774d6cd91c6
Reviewed-on: http://mpd-gerrit.cambridge.arm.com/79813
Tested-by: Kaizen <jeremy.johnson+kaizengerrit@arm.com>
Reviewed-by: Gian Marco Iodice <gianmarco.iodice@arm.com>

3 files changed, 15 insertions, 6 deletions

diff --git a/arm_compute/core/TensorShape.h b/arm_compute/core/TensorShape.h
index f8b3181686..17348e4fc6 100644
--- a/arm_compute/core/TensorShape.h
+++ b/arm_compute/core/TensorShape.h
@@ -73,7 +73,13 @@ public:
      */
     void set(size_t dimension, size_t value)
     {
-        ARM_COMPUTE_ERROR_ON(value < 1);
+        // Clear entire shape if one dimension is zero
+        if(value == 0)
+        {
+            _num_dimensions = 0;
+            std::fill(_id.begin(), _id.end(), 0);
+            return;
+        }
 
         // Make sure all empty dimensions are filled with 1
         std::fill(_id.begin() + _num_dimensions, _id.end(), 1);
diff --git a/src/core/CL/kernels/CLGEMMTranspose1xWKernel.cpp b/src/core/CL/kernels/CLGEMMTranspose1xWKernel.cpp
index 27b215f2c8..5057c8ff21 100644
--- a/src/core/CL/kernels/CLGEMMTranspose1xWKernel.cpp
+++ b/src/core/CL/kernels/CLGEMMTranspose1xWKernel.cpp
@@ -57,8 +57,7 @@ void CLGEMMTranspose1xWKernel::configure(const ICLTensor *input, ICLTensor *outp
     ARM_COMPUTE_ERROR_ON_MISMATCHING_FIXED_POINT(input, output);
 
     const unsigned int num_elems_processed_per_iteration = 16 / input->info()->element_size();
-    const float        scale_x                           = num_elems_processed_per_iteration;
-    ARM_COMPUTE_ERROR_ON((0 == static_cast<int>(input->info()->dimension(0) * (1.f / scale_x))));
+    const int          scale_x                           = num_elems_processed_per_iteration;
 
     _input  = input;
     _output = output;
@@ -80,6 +79,8 @@ void CLGEMMTranspose1xWKernel::configure(const ICLTensor *input, ICLTensor *outp
     // Configure window
     Window win = calculate_max_window(*input->info(), Steps(num_elems_processed_per_iteration));
 
+    ARM_COMPUTE_ERROR_ON_MSG((win.x().end() / scale_x) == 0, "Transposed shape would be 0 in the second dimension");
+
     AccessWindowHorizontal input_access(input->info(), 0, num_elems_processed_per_iteration);
     AccessWindowTranspose  output_access(output->info(), 0, 0, num_elems_processed_per_iteration, 1, scale_x, 1.f / scale_x);
 
diff --git a/src/core/NEON/kernels/NEGEMMTranspose1xWKernel.cpp b/src/core/NEON/kernels/NEGEMMTranspose1xWKernel.cpp
index 881ef122a1..95063a7875 100644
--- a/src/core/NEON/kernels/NEGEMMTranspose1xWKernel.cpp
+++ b/src/core/NEON/kernels/NEGEMMTranspose1xWKernel.cpp
@@ -60,14 +60,16 @@ void NEGEMMTranspose1xWKernel::configure(const ITensor *input, ITensor *output)
     ARM_COMPUTE_ERROR_ON_MISMATCHING_FIXED_POINT(input, output);
 
     const unsigned int num_elems_processed_per_iteration = 16 / input->info()->element_size();
-    const float        scale_x                           = num_elems_processed_per_iteration;
-    ARM_COMPUTE_ERROR_ON((0 == static_cast<int>(input->info()->dimension(0) * (1.f / scale_x))));
+    const int          scale_x                           = num_elems_processed_per_iteration;
 
     _input  = input;
     _output = output;
 
     // Configure kernel window
-    Window                win = calculate_max_window(*input->info(), Steps(num_elems_processed_per_iteration));
+    Window win = calculate_max_window(*input->info(), Steps(num_elems_processed_per_iteration));
+
+    ARM_COMPUTE_ERROR_ON_MSG((win.x().end() / scale_x) == 0, "Transposed shape would be 0 in the second dimension");
+
     AccessWindowTranspose output_access(output->info(), 0, 0, num_elems_processed_per_iteration, 1, scale_x, 1.f / scale_x);
 
     update_window_and_padding(win,