Merge branch 'develop' into uif2-initial

src/targets/gpu/kernels/include/migraphx/kernels/pad.hpp

@@ -28,6 +28,7 @@
 #include <migraphx/kernels/index.hpp>
 #include <migraphx/kernels/algorithm.hpp>
 #include <migraphx/kernels/ranges.hpp>
+#include <migraphx/kernels/vec.hpp>
 
 namespace migraphx {
 
@@ -53,9 +54,9 @@ __device__ void pad(const index& idx,
         if(any_of(range_multi.begin(), range_multi.end(), [&](auto j) {
                return multi[j] < offsets[j] or input_idx[j] >= input_bounds[j];
            }))
-            output[multi] = pad_val;
+            output[multi] = implicit_conversion(pad_val);
         else
-            output[multi] = input[input_idx];
+            output[multi] = implicit_conversion(input[input_idx]);
     });
 }
 

src/targets/gpu/kernels/include/migraphx/kernels/tensor_view.hpp

@@ -27,6 +27,7 @@
 #include <migraphx/kernels/shape.hpp>
 #include <migraphx/kernels/debug.hpp>
 #include <migraphx/kernels/iota_iterator.hpp>
+#include <migraphx/kernels/float8.hpp>
 
 namespace migraphx {