[core] port pynvml into vllm codebase (#12963)

Signed-off-by: youkaichao <youkaichao@gmail.com>

.pre-commit-config.yaml

@@ -8,25 +8,28 @@ repos:
   - id: yapf
     args: [--in-place, --verbose]
     additional_dependencies: [toml] # TODO: Remove when yapf is upgraded
+    exclude: 'vllm/third_party/.*'
 - repo: https://github.com/astral-sh/ruff-pre-commit
   rev: v0.9.3
   hooks:
   - id: ruff
     args: [--output-format, github]
+    exclude: 'vllm/third_party/.*'
 - repo: https://github.com/codespell-project/codespell
   rev: v2.4.0
   hooks:
   - id: codespell
-    exclude: 'benchmarks/sonnet.txt|(build|tests/(lora/data|models/fixtures|prompts))/.*'
+    exclude: 'benchmarks/sonnet.txt|(build|tests/(lora/data|models/fixtures|prompts))/.*|vllm/third_party/.*'
 - repo: https://github.com/PyCQA/isort
   rev: 5.13.2
   hooks:
   - id: isort
+    exclude: 'vllm/third_party/.*'
 - repo: https://github.com/pre-commit/mirrors-clang-format
   rev: v19.1.7
   hooks:
   - id: clang-format
-    exclude: 'csrc/(moe/topk_softmax_kernels.cu|quantization/gguf/(ggml-common.h|dequantize.cuh|vecdotq.cuh|mmq.cuh|mmvq.cuh))'
+    exclude: 'csrc/(moe/topk_softmax_kernels.cu|quantization/gguf/(ggml-common.h|dequantize.cuh|vecdotq.cuh|mmq.cuh|mmvq.cuh))|vllm/third_party/.*'
     types_or: [c++, cuda]
     args: [--style=file, --verbose]
 - repo: https://github.com/jackdewinter/pymarkdown
@@ -34,10 +37,12 @@ repos:
   hooks:
   - id: pymarkdown
     args: [fix]
+    exclude: 'vllm/third_party/.*'
 - repo: https://github.com/rhysd/actionlint
   rev: v1.7.7
   hooks:
   - id: actionlint
+    exclude: 'vllm/third_party/.*'
 - repo: local
   hooks:
   - id: mypy-local
@@ -47,6 +52,7 @@ repos:
     types: [python]
     additional_dependencies: &mypy_deps [mypy==1.11.1, types-setuptools, types-PyYAML, types-requests]
     stages: [pre-commit] # Don't run in CI
+    exclude: 'vllm/third_party/.*'
   - id: mypy-3.9 # TODO: Use https://github.com/pre-commit/mirrors-mypy when mypy setup is less awkward
     name: Run mypy for Python 3.9
     entry: tools/mypy.sh 1 "3.9"
@@ -54,6 +60,7 @@ repos:
     types: [python]
     additional_dependencies: *mypy_deps
     stages: [manual] # Only run in CI
+    exclude: 'vllm/third_party/.*'
   - id: mypy-3.10 # TODO: Use https://github.com/pre-commit/mirrors-mypy when mypy setup is less awkward
     name: Run mypy for Python 3.10
     entry: tools/mypy.sh 1 "3.10"
@@ -61,6 +68,7 @@ repos:
     types: [python]
     additional_dependencies: *mypy_deps
     stages: [manual] # Only run in CI
+    exclude: 'vllm/third_party/.*'
   - id: mypy-3.11 # TODO: Use https://github.com/pre-commit/mirrors-mypy when mypy setup is less awkward
     name: Run mypy for Python 3.11
     entry: tools/mypy.sh 1 "3.11"
@@ -68,6 +76,7 @@ repos:
     types: [python]
     additional_dependencies: *mypy_deps
     stages: [manual] # Only run in CI
+    exclude: 'vllm/third_party/.*'
   - id: mypy-3.12 # TODO: Use https://github.com/pre-commit/mirrors-mypy when mypy setup is less awkward
     name: Run mypy for Python 3.12
     entry: tools/mypy.sh 1 "3.12"
@@ -75,16 +84,19 @@ repos:
     types: [python]
     additional_dependencies: *mypy_deps
     stages: [manual] # Only run in CI
+    exclude: 'vllm/third_party/.*'
   - id: shellcheck
     name: Lint shell scripts
     entry: tools/shellcheck.sh
     language: script
     types: [shell]
+    exclude: 'vllm/third_party/.*'
   - id: png-lint
     name: Lint PNG exports from excalidraw
     entry: tools/png-lint.sh
     language: script
     types: [png]
+    exclude: 'vllm/third_party/.*'
   - id: signoff-commit
     name: Sign-off Commit
     entry: bash
@@ -97,17 +109,20 @@ repos:
     language: system
     verbose: true
     stages: [commit-msg]
+    exclude: 'vllm/third_party/.*'
   - id: check-spdx-header
     name: Check SPDX headers
     entry: python tools/check_spdx_header.py
     language: python
     types: [python]
+    exclude: 'vllm/third_party/.*'
   - id: suggestion
     name: Suggestion
     entry: bash -c 'echo "To bypass pre-commit hooks, add --no-verify to git commit."'
     language: system
     verbose: true
     pass_filenames: false
+    exclude: 'vllm/third_party/.*'
   - id: check-filenames
     name: Check for spaces in all filenames
     entry: bash
@@ -117,3 +132,4 @@ repos:
     language: system
     always_run: true
     pass_filenames: false
+    exclude: 'vllm/third_party/.*'

requirements-cuda.txt

@@ -3,7 +3,6 @@
 
 # Dependencies for NVIDIA GPUs
 ray[default] >= 2.9
-nvidia-ml-py >= 12.560.30 # for pynvml package
 torch == 2.5.1
 torchaudio==2.5.1
 # These must be updated alongside torch

tests/utils.py

@@ -46,8 +46,9 @@ if current_platform.is_rocm():
         finally:
             amdsmi_shut_down()
 elif current_platform.is_cuda():
-    from pynvml import (nvmlDeviceGetHandleByIndex, nvmlDeviceGetMemoryInfo,
-                        nvmlInit, nvmlShutdown)
+    from vllm.third_party.pynvml import (nvmlDeviceGetHandleByIndex,
+                                         nvmlDeviceGetMemoryInfo, nvmlInit,
+                                         nvmlShutdown)
 
     @contextmanager
     def _nvml():

vllm/utils.py

@@ -2239,34 +2239,13 @@ def import_pynvml():
         This causes errors when both of them are installed.
         Starting from version 12.0, it migrates to a new module
         named `pynvml_utils` to avoid the conflict.
-    
-    TL;DR: if users have pynvml<12.0 installed, it will cause problems.
-    Otherwise, `import pynvml` will import the correct module.
-    We take the safest approach here, to manually import the correct
-    `pynvml.py` module from the `nvidia-ml-py` package.
+    It is so confusing that many packages in the community use the
+    unofficial one by mistake, and we have to handle this case.
+    For example, `nvcr.io/nvidia/pytorch:24.12-py3` uses the unofficial
+    one, and it will cause errors, see the issue
+    https://github.com/vllm-project/vllm/issues/12847 for example.
+    After all the troubles, we decide to copy the official `pynvml`
+    module to our codebase, and use it directly.
     """
-    if TYPE_CHECKING:
-        import pynvml
-        return pynvml
-    if "pynvml" in sys.modules:
-        import pynvml
-        if pynvml.__file__.endswith("__init__.py"):
-            # this is pynvml < 12.0
-            raise RuntimeError(
-                "You are using a deprecated `pynvml` package. "
-                "Please uninstall `pynvml` or upgrade to at least"
-                " version 12.0. See https://pypi.org/project/pynvml "
-                "for more information.")
-        return sys.modules["pynvml"]
-    import importlib.util
-    import os
-    import site
-    for site_dir in site.getsitepackages():
-        pynvml_path = os.path.join(site_dir, "pynvml.py")
-        if os.path.exists(pynvml_path):
-            spec = importlib.util.spec_from_file_location(
-                "pynvml", pynvml_path)
-            pynvml = importlib.util.module_from_spec(spec)
-            sys.modules["pynvml"] = pynvml
-            spec.loader.exec_module(pynvml)
-            return pynvml
+    import vllm.third_party.pynvml as pynvml
+    return pynvml