1024 threads

Author: nicolas-chaulet

cuda/include/cuda_utils.h

@@ -10,7 +10,7 @@
 
 #include <vector>
 
-#define TOTAL_THREADS 512
+#define TOTAL_THREADS 1024
 
 inline int opt_n_threads(int work_size) {
   const int pow_2 = std::log(static_cast<double>(work_size)) / std::log(2.0);