[Triton] Add tl.gather
with a naive codegen implementation
#13096
Job | Run time |
---|---|
3s | |
26s | |
10m 28s | |
8m 54s | |
15m 44s | |
10m 24s | |
10m 35s | |
56m 34s |
tl.gather
with a naive codegen implementation
#13096
Job | Run time |
---|---|
3s | |
26s | |
10m 28s | |
8m 54s | |
15m 44s | |
10m 24s | |
10m 35s | |
56m 34s |