biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

Add flake8-implicit-str-concat rules to Ruff (#33191)

Browse Source 
Signed-off-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
This commit is contained in:
Harry Mellor
2026-01-28 04:56:10 +00:00
committed by GitHub
parent a97b5e206d
commit 2eb673a088
16 changed files with 34 additions and 33 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
vllm/model_executor/layers/quantization/kernels/scaled_mm/aiter.py
View File

@@ -32,8 +32,8 @@ class AiterInt8ScaledMMLinearKernel(CutlassInt8ScaledMMLinearKernel):
return (
False,
"requires setting `VLLM_ROCM_USE_AITER=1` "
+ "and `VLLM_ROCM_USE_AITER_LINEAR=1`. "
+ "`VLLM_ROCM_USE_AITER_LINEAR` default is True.",
"and `VLLM_ROCM_USE_AITER_LINEAR=1`. "
"`VLLM_ROCM_USE_AITER_LINEAR` default is True.",
)
return True, None
@@ -97,9 +97,9 @@ class AiterInt8ScaledMMLinearKernel(CutlassInt8ScaledMMLinearKernel):
per_token_scale_a and per_channel_scale_b
), (
"Currently only support per-tensor-per-tensor GEMM "
+ " and per-token-per-channel GEMM through AITER"
" and per-token-per-channel GEMM through AITER"
" w8a8 scaled gemm. `AiterInt8ScaledMMLinearKernel` "
+ "does not support AITER block scaled GEMM."
"does not support AITER block scaled GEMM."
)
# gemm_a8w8_CK(a, b, scale_a, scale_b, bias) expects