Commit 2834998
Make FP8 BMM output contiguous (#3270)
Summary:
Pull Request resolved: #3270
X-link: facebookresearch/FBGEMM#370
Make fp8 bmm output contiguous as [silu_mul](https://fburl.com/code/sa1faq0w) requests output tensor of fp8 bmm stride(-1) to be 1. This Diff fixes the issue
Reviewed By: jspark1105
Differential Revision: D64811808
fbshipit-source-id: e0f213f24fbf8bf989576371af1e2ada4cafbfb11 parent a954965 commit 2834998
File tree
1 file changed
+1
-1
lines changed- fbgemm_gpu/experimental/gen_ai/src/quantize/cutlass_extensions
1 file changed
+1
-1
lines changedLines changed: 1 addition & 1 deletion
| Original file line number | Diff line number | Diff line change | |
|---|---|---|---|
| |||
448 | 448 | | |
449 | 449 | | |
450 | 450 | | |
451 | | - | |
| 451 | + | |
452 | 452 | | |
453 | 453 | | |
454 | 454 | | |
| |||
0 commit comments