[CPU] Refactor CPU W8A8 scaled_mm (#23071)
Signed-off-by: jiang1.li <jiang1.li@intel.com>
This commit is contained in:
@@ -142,6 +142,12 @@ direct_register_custom_op(
|
||||
)
|
||||
|
||||
|
||||
def check_cpu_sgl_kernel(n: int, k: int, dtype: torch.dtype):
|
||||
return (torch._C._cpu._is_amx_tile_supported()
|
||||
and (dtype in (torch.bfloat16, torch.int8)) and k % 32 == 0
|
||||
and n % 16 == 0)
|
||||
|
||||
|
||||
def cpu_unquantized_gemm(layer: torch.nn.Module,
|
||||
x: torch.Tensor,
|
||||
weight: torch.Tensor,
|
||||
|
||||
Reference in New Issue
Block a user