csrc/multi_tensor_l2norm_kernel.cu · 17ee854eaa737fd6769125d7bb1e3d4c42b1d707 · OpenDAS / apex · GitLab

"vscode:/vscode.git/clone" did not exist on "cd4b39a900232c51e1b8e678681b57d6274e9e14"

Find file Blame History Permalink

enable wider load/store for multi_tensor_apply kernels (#763) · 17ee854e
Deyu Fu authored Apr 30, 2020
```
* modify MTA axpby for wider load/store

* Make scale/axpby/l2/adam/lamb multi_tensor uses wider load
```
17ee854e

multi_tensor_l2norm_kernel.cu 12 KB