Formatting

.github/workflows/scripts/build.sh

@@ -12,7 +12,7 @@ export MAX_JOBS=1
 # Make sure release wheels are built for the following architectures
 export PYTORCH_ROCM_ARCH="gfx90a;gfx942"
 
-rm -f $(which sccache)
+rm -f "$(which sccache)"
 
 export MAX_JOBS=32
 

vllm/_custom_ops.py

@@ -1,7 +1,7 @@
 import contextlib
 import functools
 import importlib
-from typing import TYPE_CHECKING, List, Optional, Tuple, Type, Union
+from typing import TYPE_CHECKING, List, Optional, Tuple, Union
 
 import torch
 import torch.library
@@ -242,8 +242,8 @@ def scaled_rms_norm(out: torch.Tensor, input: torch.Tensor,
 def scaled_fused_add_rms_norm(out: torch.Tensor, input: torch.Tensor,
                               residual: torch.Tensor, weight: torch.Tensor,
                               scale: torch.Tensor, epsilon: float) -> None:
-    torch.ops._C.fused_add_rms_norm_static_fp8_quant(out, input, residual, weight, scale,
-                                           epsilon)
+    torch.ops._C.fused_add_rms_norm_static_fp8_quant(out, input, residual,
+                                                     weight, scale, epsilon)
 
 
 def advance_step_flashattn(num_seqs: int, num_queries: int, block_size: int,

vllm/attention/backends/hpu_attn.py

@@ -141,6 +141,7 @@ def forward(
         k_scale: float = 1.0,
         v_scale: float = 1.0,
         attn_type: AttentionType = AttentionType.DECODER,
+        fp8_out_scale: Optional[torch.Tensor] = None,
     ) -> torch.Tensor:
         """Forward pass with xFormers and PagedAttention.
 

vllm/utils.py

@@ -158,7 +158,7 @@ class _Sentinel:
 ALL_PINNED_SENTINEL = _Sentinel()
 
 
-class rpd_trace():
+class rpd_trace:
 
     def __init__(self,
                  filename=None,
@@ -244,7 +244,7 @@ def is_hipScopedMarker_available():
     return hipScopedMarker is not None
 
 
-class rpd_mark():
+class rpd_mark:
 
     def __init__(self, name=None):
         self.name = name