Commit 7ffeb34
[XPU] [Feature] [2/3] add fp8 scaled_mm_v2 implementation for XPU (pytorch#167518)
This PR implements `scaled_mm_v2` for XPU follows the work in pytorch#164141 .
## PR stack:
- pytorch#165978 : implementation of XPU scaled_mm and oneDNN kernel
- -> pytorch#167518 : implementation of XPU scaled_mm_v2
- pytorch#166056 : Op registration
Pull Request resolved: pytorch#167518
Approved by: https://github.com/EikanWang, https://github.com/liangan11 parent 63b012a commit 7ffeb34
File tree
3 files changed
+613
-0
lines changed- aten/src/ATen
- native/mkldnn/xpu
- xpu
3 files changed
+613
-0
lines changed
0 commit comments