launch bounds (#257)

Co-authored-by: jluitjens <jluitjens@nvidia.com>
NVIDIA · Aug 26, 2022 · beaa582 · beaa582
1 parent 3b71580
commit beaa582
Showing 1 changed file with 2 additions and 1 deletion.
diff --git a/include/matx/kernels/conv.cuh b/include/matx/kernels/conv.cuh
@@ -28,8 +28,9 @@ typedef enum {
   MATX_C_METHOD_AUTO,
 } matxConvCorrMethod_t;
 
-#ifdef __CUDACC__  
+#ifdef __CUDACC__ 
 template <typename OutType, typename InType, typename FilterType>
+__launch_bounds__(1024)
 __global__ void Conv1D(OutType d_out, InType d_in, FilterType d_filter,
                        index_t signal_len,
                        matxConvCorrMode_t mode)