Gahow Wang
445e491123
third_party/vllm/ now tracked in git for direct patch management.
Based on vLLM v0.18.1 release with one patch applied:
vllm/v1/core/sched/scheduler.py:
Replace fatal assert with graceful skip when KV transfer callback
arrives for an already-aborted request during PD disaggregated serving.
Future vLLM modifications should be made directly in third_party/vllm/
and committed normally. The patches/ directory is kept as documentation
of what changed from upstream.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
67 lines
1.7 KiB
Python
67 lines
1.7 KiB
Python
# SPDX-License-Identifier: Apache-2.0
|
|
# SPDX-FileCopyrightText: Copyright contributors to the vLLM project
|
|
|
|
"""Custom exceptions for vLLM."""
|
|
|
|
from typing import Any
|
|
|
|
|
|
class VLLMValidationError(ValueError):
|
|
"""vLLM-specific validation error for request validation failures.
|
|
|
|
Args:
|
|
message: The error message describing the validation failure.
|
|
parameter: Optional parameter name that failed validation.
|
|
value: Optional value that was rejected during validation.
|
|
"""
|
|
|
|
def __init__(
|
|
self,
|
|
message: str,
|
|
*,
|
|
parameter: str | None = None,
|
|
value: Any = None,
|
|
) -> None:
|
|
super().__init__(message)
|
|
self.parameter = parameter
|
|
self.value = value
|
|
|
|
def __str__(self):
|
|
base = super().__str__()
|
|
extras = []
|
|
if self.parameter is not None:
|
|
extras.append(f"parameter={self.parameter}")
|
|
if self.value is not None:
|
|
extras.append(f"value={self.value}")
|
|
return f"{base} ({', '.join(extras)})" if extras else base
|
|
|
|
|
|
class VLLMNotFoundError(Exception):
|
|
"""vLLM-specific NotFoundError"""
|
|
|
|
pass
|
|
|
|
|
|
class LoRAAdapterNotFoundError(VLLMNotFoundError):
|
|
"""Exception raised when a LoRA adapter is not found.
|
|
|
|
This exception is thrown when a requested LoRA adapter does not exist
|
|
in the system.
|
|
|
|
Attributes:
|
|
message: The error message string describing the exception
|
|
"""
|
|
|
|
message: str
|
|
|
|
def __init__(
|
|
self,
|
|
lora_name: str,
|
|
lora_path: str,
|
|
) -> None:
|
|
message = f"Loading lora {lora_name} failed: No adapter found for {lora_path}"
|
|
self.message = message
|
|
|
|
def __str__(self):
|
|
return self.message
|