This repository has been archived by the owner on Nov 25, 2024. It is now read-only.
gather/scatter optimizations: use warp as basic working unit; use mem… #142
Job | Run time |
---|---|
50s | |
1m 10s | |
2s | |
2s | |
13m 21s | |
13m 15s | |
12m 1s | |
11m 50s | |
4m 32s | |
4m 27s | |
4m 9s | |
4m 4s | |
4m 23s | |
4m 26s | |
4m 5s | |
4m 8s | |
9s | |
24m 29s | |
31m 56s | |
7s | |
4s | |
12m 4s | |
12m 39s | |
12m 38s | |
8m 52s | |
7m 11s | |
7m 5s | |
6m 55s | |
6m 18s | |
6m 13s | |
5m 56s | |
6m 8s | |
6m 1s | |
3s | |
3m 18s | |
16m 12s | |
25m 16s | |
1m 22s | |
1m 4s | |
0s | |
4h 48m 45s |