- 28 Aug, 2023 1 commit
-
-
dan_the_3rd authored
When seqlen=8136, `smem_sz = 48840`, and apparently starting the kernel returns an `invalid argument` CUDA error. `48840 < 48 * 1024` but apparently it's still above the limit somehow..? Tested on A100
-
- 23 Jul, 2023 1 commit
-
-
Tri Dao authored
-
- 06 Jul, 2023 1 commit
-
-
Tri Dao authored
-
- 03 Jul, 2023 1 commit
-
-
Tri Dao authored
-
- 02 Jul, 2023 1 commit
-
-
Tri Dao authored
-
- 30 May, 2023 1 commit
-
-
Tri Dao authored
-
- 21 Apr, 2023 1 commit
-
-
Tri Dao authored
-
- 29 Mar, 2023 1 commit
-
-
Tri Dao authored
-
- 15 Mar, 2023 1 commit
-
-
Tri Dao authored
-
- 15 Jan, 2023 2 commits
- 04 Jan, 2023 3 commits