From 92049756cc698c9c1cdf99fff387b8fb610c39ff Mon Sep 17 00:00:00 2001
From: Konrad Zawora <kzawora@habana.ai>
Date: Fri, 6 Dec 2024 17:47:53 +0200
Subject: [PATCH] fix (hopefully) all linter errors

---
 vllm/config.py                                      | 2 +-
 vllm/executor/ray_hpu_executor.py                   | 3 ++-
 vllm/lora/layers.py                                 | 3 ---
 vllm/model_executor/layers/quantization/__init__.py | 2 +-
 vllm/platforms/hpu.py                               | 3 ++-
 vllm/utils.py                                       | 2 --
 6 files changed, 6 insertions(+), 9 deletions(-)

diff --git a/vllm/config.py b/vllm/config.py
index 901b360b2864d..d41f8752704ad 100644
--- a/vllm/config.py
+++ b/vllm/config.py
@@ -1130,7 +1130,7 @@ class SchedulerConfig:
     chunked_prefill_enabled: bool = field(init=False)
 
     # Maximum number of prefill sequences to be
-    # processed in a single iteration. Used only with padding-aware 
+    # processed in a single iteration. Used only with padding-aware
     # scheduling.
     max_num_prefill_seqs: Optional[int] = None
 
diff --git a/vllm/executor/ray_hpu_executor.py b/vllm/executor/ray_hpu_executor.py
index 8a6e193c13da7..1c6825d8d365b 100644
--- a/vllm/executor/ray_hpu_executor.py
+++ b/vllm/executor/ray_hpu_executor.py
@@ -2,7 +2,8 @@
 import os
 from collections import defaultdict
 from itertools import islice, repeat
-from typing import TYPE_CHECKING, Any, Callable, Dict, List, Optional, Tuple, Type
+from typing import (TYPE_CHECKING, Any, Callable, Dict, List, Optional, Tuple,
+                    Type)
 
 import msgspec
 
diff --git a/vllm/lora/layers.py b/vllm/lora/layers.py
index 13c15adc78e99..bec904c30a660 100644
--- a/vllm/lora/layers.py
+++ b/vllm/lora/layers.py
@@ -33,9 +33,6 @@
     VocabParallelEmbedding)
 from vllm.platforms import current_platform
 
-if current_platform.is_hpu():
-    from vllm_hpu_extension.punica_hpu import GaudiPunicaWrapper
-
 if TYPE_CHECKING:
     pass
 
diff --git a/vllm/model_executor/layers/quantization/__init__.py b/vllm/model_executor/layers/quantization/__init__.py
index cc9e45cc1d232..6f2b9f3be322c 100644
--- a/vllm/model_executor/layers/quantization/__init__.py
+++ b/vllm/model_executor/layers/quantization/__init__.py
@@ -50,13 +50,13 @@ def get_quantization_config(quantization: str) -> Type[QuantizationConfig]:
     from .gptq_marlin import GPTQMarlinConfig
     from .gptq_marlin_24 import GPTQMarlin24Config
     from .hqq_marlin import HQQMarlinConfig
+    from .inc import INCConfig
     from .ipex_quant import IPEXConfig
     from .marlin import MarlinConfig
     from .modelopt import ModelOptFp8Config
     from .neuron_quant import NeuronQuantConfig
     from .qqq import QQQConfig
     from .tpu_int8 import Int8TpuConfig
-    from .inc import INCConfig
 
     method_to_config: Dict[str, Type[QuantizationConfig]] = {
         "aqlm": AQLMConfig,
diff --git a/vllm/platforms/hpu.py b/vllm/platforms/hpu.py
index 2d77f3e0c2fb5..2b1a1dad95753 100644
--- a/vllm/platforms/hpu.py
+++ b/vllm/platforms/hpu.py
@@ -32,7 +32,8 @@ def check_and_update_config(cls, vllm_config: VllmConfig) -> None:
         parallel_config = vllm_config.parallel_config
         if parallel_config.worker_cls == "auto":
             if scheduler_config.is_multi_step:
-                parallel_config.worker_cls = "vllm.worker.multi_step_hpu_worker.MultiStepHPUWorker"
+                parallel_config.worker_cls = \
+                    "vllm.worker.multi_step_hpu_worker.MultiStepHPUWorker"
             elif vllm_config.speculative_config:
                 parallel_config.worker_cls = \
                     "vllm.spec_decode.spec_decode_worker.create_spec_worker"
diff --git a/vllm/utils.py b/vllm/utils.py
index c426e58a49e55..320b14653854b 100644
--- a/vllm/utils.py
+++ b/vllm/utils.py
@@ -773,8 +773,6 @@ def is_pin_memory_available() -> bool:
         return False
     elif current_platform.is_cpu() or current_platform.is_openvino():
         return False
-    elif (current_platform.is_hpu()):
-        return False
     return True