github
/
ColossalAI
mirror of https://github.com/hpcaitech/ColossalAI
1
0
Fork 0
Code Issues Projects Releases Wiki Activity

[fp8] linear perf enhancement

pull/6009/head
botbw 2024-08-15 03:12:08 +00:00
parent 88fa096d78
commit 1a2e90dcc1
1 changed files with 1 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
colossalai/quantization/fp8.py
View File

@ -728,14 +728,11 @@ class _LinearFp8(torch.autograd.Function):
return x_grad.reshape(ctx.x_shape), w_grad, bias_grad
@torch.compile(mode="reduce-overhead", disable=not SUPPORT_TORCH_COMPILE)
@torch.compile(mode="max-autotune-no-cudagraphs", disable=not SUPPORT_TORCH_COMPILE, dynamic=False)
def _linear_fp8(input: torch.Tensor, weight: torch.Tensor, bias: Optional[torch.Tensor] = None) -> torch.Tensor:
return _LinearFp8.apply(input, weight, bias)
def linear_fp8(input: torch.Tensor, weight: torch.Tensor, bias: Optional[torch.Tensor] = None) -> torch.Tensor:
out = _linear_fp8(input, weight, bias)
if SUPPORT_TORCH_COMPILE:
# avoid modifying the tensor created from cuda graph
out = out.clone()
return out