xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-10 06:35:00 +08:00
Code Issues Packages Projects Releases Wiki Activity

[ROCm][AMD][Bugfix] adding a missing triton autotune config (#4845)

Browse Source
This commit is contained in:
Hongxia Yang 2024-05-16 13:46:52 -04:00 committed by GitHub
parent f09edd8a25
commit b5853f9963
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 10 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
vllm/attention/ops/triton_flash_attention.py
View File

@ -239,6 +239,16 @@ def _attn_fwd_inner(
num_stages=1,
num_warps=8,
),
triton.Config(
{
"BLOCK_M": 128,
"BLOCK_N": 64,
"waves_per_eu": 1,
"PRE_LOAD_V": False,
},
num_stages=1,
num_warps=4,
),
triton.Config(
{
"BLOCK_M": 128,