From b5c6a30d4e1c01f3f98c43dc75ceae39a1c15b05 Mon Sep 17 00:00:00 2001 From: tzhouam Date: Sat, 4 Jul 2026 15:43:12 +0800 Subject: [PATCH] ci: fix numpy/numba, libnvJitLink.so.13, and gr00t AutoProcessor.register on align branch Three regressions surfaced only on dev/vllm-align in Buildkite build 2520 (26 failing jobs vs 5 on main); ~15 of the 19 align-unique failures trace to these: 1. numpy 2.5 vs numba: Dockerfile.ci pinned numpy==2.2.6 before the flashinfer and torch cu130 reinstalls, which resolved numpy back to 2.5 and broke 'import numba' (needs numpy <= 2.4). Re-pin numpy as the final dependency op. 2. libnvJitLink.so.13 not found: the CUDA-13 JIT linker lives in the nvidia-cuda-nvjitlink-cu13 wheel, not /usr/local/cuda (CUDA 12) as the old LD_LIBRARY_PATH assumed; bitsandbytes/torch.compile subprocesses could not dlopen it (broke Entrypoint L4, a multi-GPU diffusion job, both Quantization jobs via hang/timeout). Install the wheel and register its lib dir via ldconfig. 3. transformers >= 5.5 AutoProcessor.register now requires the config class, not a model_type string; processing_gr00t_n1d7.py passed "Gr00tN1d7" -> AttributeError 'str' object has no attribute '__module__' -> pytest collection errors (exit 2). Pass Gr00tN1d7Config, matching AutoModel.register in modeling/gr00t_n1d7.py. Co-Authored-By: Claude Opus 4.8 --- docker/Dockerfile.ci | 31 ++++++++++++++----- .../gr00t/modeling/processing_gr00t_n1d7.py | 8 ++++- 2 files changed, 31 insertions(+), 8 deletions(-) diff --git a/docker/Dockerfile.ci b/docker/Dockerfile.ci index 70848371c..6471ee44b 100644 --- a/docker/Dockerfile.ci +++ b/docker/Dockerfile.ci @@ -21,11 +21,10 @@ RUN apt-get update && \ # The cu130 wheel index lives at //cu130/vllm/ but the actual wheel files # are stored at the top level //. ENV VLLM_PRECOMPILED_WHEEL_COMMIT=2dfaae752b4db0d43cfc0715c780e33be030d0f1 -# Ensure CUDA library path is in LD_LIBRARY_PATH so bitsandbytes' precompiled -# C extension can find libnvJitLink.so.13 (CUDA 13.x JIT linker) at import time. -# The base vLLM image has bitsandbytes pre-installed; without this env var, -# worker subprocesses fail to load bitsandbytes' C extension, producing -# OSError tracebacks that pollute CI logs. +# Keep the base image's CUDA-12 toolkit libs on the loader path. NOTE: this does +# NOT resolve libnvJitLink.so.13 (that CUDA-13 lib is not in /usr/local/cuda here) — +# the .so.13 JIT linker is handled separately via ldconfig after the cu130 wheels +# are installed (see the nvidia-cuda-nvjitlink-cu13 step below). ENV LD_LIBRARY_PATH=/usr/local/cuda/targets/x86_64-linux/lib:${LD_LIBRARY_PATH} RUN VLLM_WHEEL_URL=$(python3 -c "import urllib.request,re; \ from urllib.parse import urljoin, quote; \ @@ -43,8 +42,7 @@ RUN uv pip install --system ".[dev]" RUN uv pip install --system --upgrade \ "flashinfer-python==0.6.13" \ - "flashinfer-cubin==0.6.13" \ - "numpy==2.2.6" + "flashinfer-cubin==0.6.13" RUN uv pip install --system --upgrade \ "flashinfer-jit-cache==0.6.13" \ @@ -57,6 +55,25 @@ RUN uv pip install --system --force-reinstall \ "torch==2.11.0" "torchaudio==2.11.0" "torchvision==0.26.0" \ --torch-backend cu130 +# CUDA 13 JIT linker: libnvJitLink.so.13 ships only in the nvidia-cuda-nvjitlink-cu13 +# pip wheel (the v0.24.0 base image's /usr/local/cuda toolkit is CUDA 12, so the +# LD_LIBRARY_PATH above never resolves the .so.13). The cu130 torch/vllm wheels pull +# this package into site-packages, but its lib dir is not on the dynamic-loader path, +# so bitsandbytes / torch.compile worker subprocesses fail to dlopen it. Register the +# actual lib dir with ldconfig (system-wide, survives subprocess env resets). +RUN uv pip install --system "nvidia-cuda-nvjitlink-cu13" && \ + NVJITLINK_LIB="$(find / -name 'libnvJitLink.so.13*' 2>/dev/null | head -1)" && \ + test -n "$NVJITLINK_LIB" && \ + dirname "$NVJITLINK_LIB" > /etc/ld.so.conf.d/nvjitlink-cu13.conf && \ + ldconfig && \ + echo "Registered ${NVJITLINK_LIB} with ldconfig" + +# Re-pin numpy LAST. numba requires numpy <= 2.4, but the flashinfer and torch cu130 +# reinstalls above resolve numpy back up to 2.5, which breaks `import numba` at test +# time ("Numba needs NumPy 2.4 or less"). This must be the final dependency op so +# nothing can clobber it again. +RUN uv pip install --system "numpy==2.2.6" + RUN ln -sf /usr/bin/python3 /usr/bin/python ENTRYPOINT [] diff --git a/vllm_omni/diffusion/models/gr00t/modeling/processing_gr00t_n1d7.py b/vllm_omni/diffusion/models/gr00t/modeling/processing_gr00t_n1d7.py index ba128f450..ade5eac83 100644 --- a/vllm_omni/diffusion/models/gr00t/modeling/processing_gr00t_n1d7.py +++ b/vllm_omni/diffusion/models/gr00t/modeling/processing_gr00t_n1d7.py @@ -19,6 +19,7 @@ from transformers.utils import cached_file from vllm.logger import init_logger from vllm_omni.diffusion.models.gr00t.configs.embodiment_configs import ModalityConfig +from vllm_omni.diffusion.models.gr00t.configs.gr00t_n1d7 import Gr00tN1d7Config from vllm_omni.diffusion.models.gr00t.dataio.embodiment_tags import EmbodimentTag from vllm_omni.diffusion.models.gr00t.dataio.state_action.state_action_processor import StateActionProcessor from vllm_omni.diffusion.models.gr00t.dataio.utils import parse_modality_configs, to_json_serializable @@ -552,4 +553,9 @@ class Gr00tN1d7Processor(ProcessorMixin): return cls(**processor_kwargs, transformers_loading_kwargs=transformers_loading_kwargs) -AutoProcessor.register("Gr00tN1d7", Gr00tN1d7Processor) +# transformers >= 5.5 requires the config *class* (not a model_type string) as the +# first arg to AutoProcessor.register — it does `key.__module__` internally. Passing +# the string "Gr00tN1d7" raises AttributeError: 'str' object has no attribute +# '__module__'. Match the AutoModel.register(Gr00tN1d7Config, ...) call in +# modeling/gr00t_n1d7.py. +AutoProcessor.register(Gr00tN1d7Config, Gr00tN1d7Processor)