diff options
| author |  Benjamin Kramer <kramerb@google.com> | 2018-08-03 10:29:38 -0700 |
|---|---|---|
| committer |  TensorFlower Gardener <gardener@tensorflow.org> | 2018-08-03 10:33:04 -0700 |
| commit | 1e55f802fd4bc30a3e68726e2914dba0c08ffbd8 (patch) | |
| tree | e7d7234c073301ad56cdaf098e93502fc0ab2e72 /BUILD | |
| parent | e36c16c6720dd64ae5d8a1f8555102a1323af9ae (diff) | |
[XLA:GPU] Add a fast version of gemmStridedBatched for cuda 9.1
It's unfortunate that this was only added in 9.1, but I haven't found a good
way of emulating the behavior on 9.0 without falling back to non-batched gemms.
PiperOrigin-RevId: 207286575
Diffstat (limited to 'BUILD')
0 files changed, 0 insertions, 0 deletions