Skip to content

Commit

Permalink
PR tensorflow#8696: [NVIDIA] Fix XLA build with the new cuDNN
Browse files Browse the repository at this point in the history
Imported from GitHub PR openxla/xla#8696

This PR fixes a couple of minor issues to support the XLA build against the cuDNN v9.

cc. @reedwm
Copybara import of the project:

--
fb0ae743eafea727423dd02736214fc6f31364ee by Kaixi Hou <kaixih@nvidia.com>:

Fix support to cudnn v9

Merging this change closes tensorflow#8696

PiperOrigin-RevId: 600734575
  • Loading branch information
kaixih authored and tensorflower-gardener committed Jan 23, 2024
1 parent 4bfbeff commit 7670a94
Showing 1 changed file with 4 additions and 0 deletions.
4 changes: 4 additions & 0 deletions third_party/xla/xla/service/gpu/conv_algorithm_picker.cc
Original file line number Diff line number Diff line change
Expand Up @@ -72,7 +72,11 @@ limitations under the License.

#if (defined(GOOGLE_CUDA) && GOOGLE_CUDA)
#include "third_party/gpus/cudnn/cudnn.h" // IWYU pragma: keep
#if CUDNN_VERSION >= 9000
#include "third_party/gpus/cudnn/cudnn_ops.h"
#else
#include "third_party/gpus/cudnn/cudnn_ops_infer.h"
#endif // CUDNN_VERSION >= 9000
#include "xla/service/gpu/buffer_comparator.h"
#include "xla/stream_executor/gpu/redzone_allocator.h"
#endif
Expand Down

0 comments on commit 7670a94

Please sign in to comment.