|
|
@@ -2,7 +2,7 @@
|
|
|
*
|
|
|
* Copyright (C) 2011-2013,2015,2017 Inria
|
|
|
* Copyright (C) 2010-2019 CNRS
|
|
|
- * Copyright (C) 2009-2011,2013-2018 Université de Bordeaux
|
|
|
+ * Copyright (C) 2009-2011,2013-2019 Université de Bordeaux
|
|
|
*
|
|
|
* StarPU is free software; you can redistribute it and/or modify
|
|
|
* it under the terms of the GNU Lesser General Public License as published by
|
|
|
@@ -96,8 +96,11 @@ func <<<grid,block,0,starpu_cuda_get_local_stream()>>> (foo, bar);
|
|
|
cudaStreamSynchronize(starpu_cuda_get_local_stream());
|
|
|
\endcode
|
|
|
|
|
|
+as well as the use of cudaMemcpyAsync(), etc. for each CUDA operation one needs
|
|
|
+to use a version that takes the a stream parameter.
|
|
|
+
|
|
|
Unfortunately, some CUDA libraries do not have stream variants of
|
|
|
-kernels. This will lower the potential for overlapping.
|
|
|
+kernels. This will seriously lower the potential for overlapping.
|
|
|
|
|
|
Calling starpu_cublas_init() makes StarPU already do appropriate calls for the
|
|
|
CUBLAS library. Some libraries like Magma may however change the current stream of CUBLAS v1,
|
