[ROCM] Fix ROCm warnings, environment flag access, and GEMM kernel naming for consistency in _aiter_ops.py (#28464)
Signed-off-by: vllmellm <vllm.ellm@embeddedllm.com>
This commit is contained in:
@@ -117,4 +117,4 @@ class AiterScaledMMLinearKernel(CutlassScaledMMLinearKernel):
|
||||
# a to be [M, K]
|
||||
# b to be [N, K]
|
||||
# CutlassScaledMMLinearKernel prepare weight `w_q` in [K, N] format
|
||||
return rocm_aiter_ops.gemm_w8a8(x_q, w_q.t(), x_s, w_s, bias, out_dtype)
|
||||
return rocm_aiter_ops.gemm_a8w8(x_q, w_q.t(), x_s, w_s, bias, out_dtype)
|
||||
|
||||
Reference in New Issue
Block a user