'typo'

Author: tchaton

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);

cuda/src/ball_query_gpu.cu

@@ -110,7 +110,7 @@ void query_ball_point_kernel_partial_wrapper(long batch_size,
 						int64_t *idx_out,
 						float *dist_out) {
 
-	query_ball_point_kernel_partial_dense<<<batch_size, THREADS>>>(
+	query_ball_point_kernel_partial_dense<<<batch_size, TOTAL_THREADS>>>(
 		size_x, size_y, radius, nsample, x, y,
 		batch_x, batch_y, idx_out, dist_out);