Keren Zhou 307b5caa49 [BACKEND] Fix scan issues on repetitive warps and improve perf when there's a single warp on the axis (#2330) ...

1. On the axis, using `getAxisNumWarpsWithUniqueData` instead of getting
the raw number of warps to avoid communication among warps that handle
the same piece of data.
2. When there's a single warp on the axis, using warp Intrinsics for
communication and skip shared memory.

Need a follow up PR for code clean up.

2023-09-18 17:45:05 -04:00

..

examples

[FORMAT] autopep8 now uses max-line-length=88 (#1410)

2023-03-25 15:46:50 -07:00

src

[FRONTEND] interpreter rewrite (#2321)

2023-09-17 14:58:50 -07:00

test

[BACKEND] Fix scan issues on repetitive warps and improve perf when there's a single warp on the axis (#2330)

2023-09-18 17:45:05 -04:00

triton

[RUNTIME][INTERPRETER] now also override __str__ method for tensors (#2325)

2023-09-17 16:49:30 -07:00

tutorials

[OPS] Remove unnecessary perf bug workaround (#2240)

2023-09-04 21:30:54 -07:00

MANIFEST.in

[TOOLS] improved ahead-of-time compiler (#1805)

2023-06-21 01:02:58 -07:00

pyproject.toml

[BUILD] use ninja (#2318)

2023-09-17 02:08:04 -07:00

setup.py

[FRONTEND] interpreter rewrite (#2321)

2023-09-17 14:58:50 -07:00