Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[Installation]: Cannot compile flash attention when building from source #8878

Closed
1 task done
KevinRSX opened this issue Sep 27, 2024 · 7 comments · Fixed by #8931
Closed
1 task done

[Installation]: Cannot compile flash attention when building from source #8878

KevinRSX opened this issue Sep 27, 2024 · 7 comments · Fixed by #8931
Labels
installation Installation problems

Comments

@KevinRSX
Copy link

KevinRSX commented Sep 27, 2024 •
edited
Loading

Your current environment

The doc did not specify what CUDA version is required, but the recommended Docker image nvcr.io/nvidia/pytorch:23.10-py3 uses 12.2.

The output of python collect_env.py, running from the NVIDIA PyTorch 23.10 Docker image.

Collecting environment information...
PyTorch version: 2.1.0a0+32f93b1
Is debug build: False
CUDA used to build PyTorch: 12.2
ROCM used to build PyTorch: N/A

OS: Ubuntu 22.04.3 LTS (x86_64)
GCC version: (Ubuntu 11.4.0-1ubuntu1~22.04) 11.4.0
Clang version: Could not collect
CMake version: version 3.27.6
Libc version: glibc-2.35

Python version: 3.10.12 (main, Jun 11 2023, 05:26:28) [GCC 11.4.0] (64-bit runtime)
Python platform: Linux-5.15.0-122-generic-x86_64-with-glibc2.35
Is CUDA available: True
CUDA runtime version: 12.2.140
CUDA_MODULE_LOADING set to: LAZY
GPU models and configuration: GPU 0: NVIDIA A30
Nvidia driver version: 535.183.01
cuDNN version: Probably one of the following:
/usr/lib/x86_64-linux-gnu/libcudnn.so.8.9.5
/usr/lib/x86_64-linux-gnu/libcudnn_adv_infer.so.8.9.5
/usr/lib/x86_64-linux-gnu/libcudnn_adv_train.so.8.9.5
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_infer.so.8.9.5
/usr/lib/x86_64-linux-gnu/libcudnn_cnn_train.so.8.9.5
/usr/lib/x86_64-linux-gnu/libcudnn_ops_infer.so.8.9.5
/usr/lib/x86_64-linux-gnu/libcudnn_ops_train.so.8.9.5
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

CPU:
Architecture:                         x86_64
CPU op-mode(s):                       32-bit, 64-bit
Address sizes:                        43 bits physical, 48 bits virtual
Byte Order:                           Little Endian
CPU(s):                               64
On-line CPU(s) list:                  0-63
Vendor ID:                            AuthenticAMD
Model name:                           AMD EPYC 7302 16-Core Processor
CPU family:                           23
Model:                                49
Thread(s) per core:                   2
Core(s) per socket:                   16
Socket(s):                            2
Stepping:                             0
BogoMIPS:                             5988.55
Flags:                                fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl nonstop_tsc cpuid extd_apicid aperfmperf rapl pni pclmulqdq monitor ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt aes xsave avx f16c rdrand lahf_lm cmp_legacy svm extapic cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ibs skinit wdt tce topoext perfctr_core perfctr_nb bpext perfctr_llc mwaitx cpb cat_l3 cdp_l3 hw_pstate ssbd mba ibrs ibpb stibp vmmcall fsgsbase bmi1 avx2 smep bmi2 cqm rdt_a rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local clzero irperf xsaveerptr rdpru wbnoinvd amd_ppin arat npt lbrv svm_lock nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold avic v_vmsave_vmload vgif v_spec_ctrl umip rdpid overflow_recov succor smca sme sev sev_es
Virtualization:                       AMD-V
L1d cache:                            1 MiB (32 instances)
L1i cache:                            1 MiB (32 instances)
L2 cache:                             16 MiB (32 instances)
L3 cache:                             256 MiB (16 instances)
NUMA node(s):                         2
NUMA node0 CPU(s):                    0-15,32-47
NUMA node1 CPU(s):                    16-31,48-63
Vulnerability Gather data sampling:   Not affected
Vulnerability Itlb multihit:          Not affected
Vulnerability L1tf:                   Not affected
Vulnerability Mds:                    Not affected
Vulnerability Meltdown:               Not affected
Vulnerability Mmio stale data:        Not affected
Vulnerability Reg file data sampling: Not affected
Vulnerability Retbleed:               Mitigation; untrained return thunk; SMT enabled with STIBP protection
Vulnerability Spec rstack overflow:   Mitigation; safe RET
Vulnerability Spec store bypass:      Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1:             Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:             Mitigation; Retpolines; IBPB conditional; STIBP always-on; RSB filling; PBRSB-eIBRS Not affected; BHI Not affected
Vulnerability Srbds:                  Not affected
Vulnerability Tsx async abort:        Not affected

Versions of relevant libraries:
[pip3] numpy==1.22.2
[pip3] nvidia-dali-cuda120==1.30.0
[pip3] nvidia-pyindex==1.0.9
[pip3] onnx==1.14.0
[pip3] pynvml==11.4.1
[pip3] pytorch-quantization==2.1.2
[pip3] pyzmq==25.1.1
[pip3] torch==2.1.0a0+32f93b1
[pip3] torch-tensorrt==0.0.0
[pip3] torchdata==0.7.0a0
[pip3] torchtext==0.16.0a0
[pip3] torchvision==0.16.0a0
[pip3] triton==2.1.0+e621604
[conda] Could not collect
ROCM Version: Could not collect
Neuron SDK Version: N/A
vLLM Version: N/A
vLLM Build Flags:
CUDA Archs: 5.2 6.0 6.1 7.0 7.2 7.5 8.0 8.6 8.7 9.0+PTX; ROCm: Disabled; Neuron: Disabled
GPU Topology:
GPU0    NIC0    NIC1    NIC2    CPU Affinity    NUMA Affinity   GPU NUMA ID
GPU0     X      NODE    SYS     SYS     0-15,32-47      0               N/A
NIC0    NODE     X      SYS     SYS
NIC1    SYS     SYS      X      PIX
NIC2    SYS     SYS     PIX      X

Legend:

  X    = Self
  SYS  = Connection traversing PCIe as well as the SMP interconnect between NUMA nodes (e.g., QPI/UPI)
  NODE = Connection traversing PCIe as well as the interconnect between PCIe Host Bridges within a NUMA node
  PHB  = Connection traversing PCIe as well as a PCIe Host Bridge (typically the CPU)
  PXB  = Connection traversing multiple PCIe bridges (without traversing the PCIe Host Bridge)
  PIX  = Connection traversing at most a single PCIe bridge
  NV#  = Connection traversing a bonded set of # NVLinks

NIC Legend:

  NIC0: mlx5_0
  NIC1: mlx5_1
  NIC2: mlx5_2

How you are installing vllm

Command (ran from the NVIDIA PyTorch 23.10 Docker image):

pip install -e .
Toggle to see view output (excluding installation of dependencies) 
Building wheels for collected packages: vllm
  Building editable for vllm (pyproject.toml) ... error
  error: subprocess-exited-with-error

  × Building editable for vllm (pyproject.toml) did not run successfully.
  │ exit code: 1
  ╰─> [305 lines of output]
      /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/_distutils_hack/__init__.py:54: UserWarning: Reliance on distutils from stdlib is deprecated. Users must rely on setuptools to provide the distutils module. Avoid importing distutils or import setuptools first, and avoid setting SETUPTOOLS_USE_DISTUTILS=stdlib. Register concerns at https://github.com/pypa/setuptools/issues/new?template=distutils-deprecation.yml
        warnings.warn(
      /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/_subclasses/functional_tensor.py:258: UserWarning: Failed to initialize NumPy: No module named 'numpy' (Triggered internally at ../torch/csrc/utils/tensor_numpy.cpp:84.)
        cpu = _conversion_method_template(device=torch.device("cpu"))
      running editable_wheel
      creating /tmp/pip-wheel-ul85b39i/.tmp-tf3dh1m_/vllm.egg-info
      writing /tmp/pip-wheel-ul85b39i/.tmp-tf3dh1m_/vllm.egg-info/PKG-INFO
      writing dependency_links to /tmp/pip-wheel-ul85b39i/.tmp-tf3dh1m_/vllm.egg-info/dependency_links.txt
      writing entry points to /tmp/pip-wheel-ul85b39i/.tmp-tf3dh1m_/vllm.egg-info/entry_points.txt
      writing requirements to /tmp/pip-wheel-ul85b39i/.tmp-tf3dh1m_/vllm.egg-info/requires.txt
      writing top-level names to /tmp/pip-wheel-ul85b39i/.tmp-tf3dh1m_/vllm.egg-info/top_level.txt
      writing manifest file '/tmp/pip-wheel-ul85b39i/.tmp-tf3dh1m_/vllm.egg-info/SOURCES.txt'
      ERROR setuptools_scm._file_finders.git listing git files failed - pretending there aren't any
      reading manifest file '/tmp/pip-wheel-ul85b39i/.tmp-tf3dh1m_/vllm.egg-info/SOURCES.txt'
      reading manifest template 'MANIFEST.in'
      adding license file 'LICENSE'
      writing manifest file '/tmp/pip-wheel-ul85b39i/.tmp-tf3dh1m_/vllm.egg-info/SOURCES.txt'
      creating '/tmp/pip-wheel-ul85b39i/.tmp-tf3dh1m_/vllm-0.6.1.post3.dev138+g7193774b.cu122.dist-info'
      creating /tmp/pip-wheel-ul85b39i/.tmp-tf3dh1m_/vllm-0.6.1.post3.dev138+g7193774b.cu122.dist-info/WHEEL
      running build_py
      running build_ext
      -- The CXX compiler identification is GNU 11.4.0
      -- Detecting CXX compiler ABI info
      -- Detecting CXX compiler ABI info - done
      -- Check for working CXX compiler: /usr/bin/c++ - skipped
      -- Detecting CXX compile features
      -- Detecting CXX compile features - done
      -- Build type: RelWithDebInfo
      -- Target device: cuda
      -- Found Python: /usr/bin/python (found version "3.10.12") found components: Interpreter Development.Module Development.SABIModule
      -- Found python matching: /usr/bin/python.
      -- Found CUDA: /usr/local/cuda (found version "12.2")
      -- The CUDA compiler identification is NVIDIA 12.2.140
      -- Detecting CUDA compiler ABI info
      -- Detecting CUDA compiler ABI info - done
      -- Check for working CUDA compiler: /usr/local/cuda/bin/nvcc - skipped
      -- Detecting CUDA compile features
      -- Detecting CUDA compile features - done
      -- Found CUDAToolkit: /usr/local/cuda/include (found version "12.2.140")
      -- Performing Test CMAKE_HAVE_LIBC_PTHREAD
      -- Performing Test CMAKE_HAVE_LIBC_PTHREAD - Success
      -- Found Threads: TRUE
      -- Caffe2: CUDA detected: 12.2
      -- Caffe2: CUDA nvcc is: /usr/local/cuda/bin/nvcc
      -- Caffe2: CUDA toolkit directory: /usr/local/cuda
      -- Caffe2: Header version is: 12.2
      -- /usr/local/cuda/lib64/libnvrtc.so shorthash is 000ca627
      -- USE_CUDNN is set to 0. Compiling without cuDNN support
      -- USE_CUSPARSELT is set to 0. Compiling without cuSPARSELt support
      CMake Warning at /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/share/cmake/Caffe2/public/utils.cmake:382 (message):
        In the future we will require one to explicitly pass TORCH_CUDA_ARCH_LIST
        to cmake instead of implicitly setting it as an env variable.  This will
        become a FATAL_ERROR in future version of pytorch.
      Call Stack (most recent call first):
        /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/share/cmake/Caffe2/public/cuda.cmake:300 (torch_cuda_get_nvcc_gencode_flag)
        /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/share/cmake/Caffe2/Caffe2Config.cmake:86 (include)
        /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/share/cmake/Torch/TorchConfig.cmake:68 (find_package)
        CMakeLists.txt:84 (find_package)


      -- Added CUDA NVCC flags for: -gencode;arch=compute_52,code=sm_52;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_72,code=sm_72;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_80,code=sm_80;-gencode;arch=compute_86,code=sm_86;-gencode;arch=compute_87,code=sm_87;-gencode;arch=compute_90,code=sm_90;-gencode;arch=compute_90,code=compute_90
      CMake Warning at /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/share/cmake/Torch/TorchConfig.cmake:22 (message):
        static library kineto_LIBRARY-NOTFOUND not found.
      Call Stack (most recent call first):
        /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/share/cmake/Torch/TorchConfig.cmake:120 (append_torchlib_if_found)
        CMakeLists.txt:84 (find_package)


      -- Found Torch: /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/lib/libtorch.so
      -- Enabling core extension.
      -- CUDA supported arches: 7.0;7.5;8.0;8.6;8.9;9.0
      -- discarding unsupported CUDA arch 3.10.
      -- discarding unsupported CUDA arch 3.10.
      -- discarding unsupported CUDA arch 3.10.
      -- discarding unsupported CUDA arch 3.10.
      -- discarding unsupported CUDA arch 3.10.
      -- CUDA target arches: 70-real;75-real;80-real;86-real;90-real;90-virtual
      -- CMake Version: 3.30.3
      -- CUTLASS 3.5.1
      -- CUDART: /usr/local/cuda/lib64/libcudart.so
      -- CUDA Driver: /usr/local/cuda/lib64/stubs/libcuda.so
      -- NVRTC: /usr/local/cuda/lib64/libnvrtc.so
      -- Default Install Location: install
      -- Found Python3: /usr/bin/python3.10 (found suitable version "3.10.12", minimum required is "3.5") found components: Interpreter
      -- Make cute::tuple be the new standard-layout tuple type
      -- CUDA Compilation Architectures: 70;72;75;80;86;87;89;90;90a
      -- Enable caching of reference results in conv unit tests
      -- Enable rigorous conv problem sizes in conv unit tests
      -- Using NVCC flags: --expt-relaxed-constexpr;-DCUTE_USE_PACKED_TUPLE=1;-DCUTLASS_TEST_LEVEL=0;-DCUTLASS_TEST_ENABLE_CACHED_RESULTS=1;-DCUTLASS_CONV_UNIT_TEST_RIGOROUS_SIZE_ENABLED=1;-DCUTLASS_DEBUG_TRACE_LEVEL=0;-Xcompiler=-Wconversion;-Xcompiler=-fno-strict-aliasing;-lineinfo
      -- Configuring cublas ...
      -- cuBLAS Disabled.
      -- Configuring cuBLAS ... done.
      -- Machete generation completed successfully.
      -- Machete generated sources: /workspace/vllm/csrc/quantization/machete/generated/machete_mm_bf16u4.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_bf16u4_impl_part0.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_bf16u4_impl_part1.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_bf16u4b8.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_bf16u4b8_impl_part0.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_bf16u4b8_impl_part1.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_bf16u8.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_bf16u8_impl_part0.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_bf16u8_impl_part1.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_bf16u8b128.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_bf16u8b128_impl_part0.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_bf16u8b128_impl_part1.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_f16u4.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_f16u4_impl_part0.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_f16u4_impl_part1.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_f16u4b8.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_f16u4b8_impl_part0.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_f16u4b8_impl_part1.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_f16u8.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_f16u8_impl_part0.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_f16u8_impl_part1.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_f16u8b128.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_f16u8b128_impl_part0.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_mm_f16u8b128_impl_part1.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_prepack_bf16u4.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_prepack_bf16u4b8.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_prepack_bf16u8.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_prepack_bf16u8b128.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_prepack_f16u4.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_prepack_f16u4b8.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_prepack_f16u8.cu;/workspace/vllm/csrc/quantization/machete/generated/machete_prepack_f16u8b128.cu
      -- Enabling C extension.
      -- Enabling moe extension.
      -- Build type: RelWithDebInfo
      -- Target device: cuda
      -- Building vllm-flash-attn inside vLLM. Skipping flag detection and relying on parent build.
      -- vllm-flash-attn is available at /tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src
      -- Configuring done (15.1s)
      -- Generating done (0.1s)
      -- Build files have been written to: /tmp/tmp7hhhhz5w.build-temp
      [1/137] Building CXX object CMakeFiles/_core_C.dir/csrc/core/torch_bindings.cpp.o
      [2/137] Linking CXX shared module _core_C.abi3.so
      [3/137] Building CXX object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/flash_api.cpp.o
      [4/137] Building CXX object CMakeFiles/_moe_C.dir/csrc/moe/torch_bindings.cpp.o
      [5/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_fp16_sm80.cu.o
      FAILED: _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_fp16_sm80.cu.o
      /usr/local/cuda/bin/nvcc -forward-unknown-to-host-compiler -DFLASHATTENTION_DISABLE_DROPOUT -DFLASHATTENTION_DISABLE_UNEVEN_K -DPy_LIMITED_API=3 -DTORCH_EXTENSION_NAME=vllm_flash_attn_c -DUSE_C10D_GLOO -DUSE_C10D_NCCL -DUSE_DISTRIBUTED -DUSE_RPC -DUSE_TENSORPIPE -Dvllm_flash_attn_c_EXPORTS -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/cutlass/include -isystem /usr/include/python3.10 -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include/torch/csrc/api/include -isystem /usr/local/cuda/include -DONNX_NAMESPACE=onnx_c2 -Xcudafe --diag_suppress=cc_clobber_ignored,--diag_suppress=field_without_dll_interface,--diag_suppress=base_class_has_different_dll_interface,--diag_suppress=dll_interface_conflict_none_assumed,--diag_suppress=dll_interface_conflict_dllexport_assumed,--diag_suppress=bad_friend_decl --expt-relaxed-constexpr --expt-extended-lambda -O3 -g -DNDEBUG -std=c++17 "--generate-code=arch=compute_70,code=[sm_70]" "--generate-code=arch=compute_75,code=[sm_75]" "--generate-code=arch=compute_80,code=[sm_80]" "--generate-code=arch=compute_86,code=[sm_86]" "--generate-code=arch=compute_90,code=[sm_90]" "--generate-code=arch=compute_90,code=[compute_90]" -Xcompiler=-fPIC --expt-relaxed-constexpr -DENABLE_FP8 --threads=1 --expt-extended-lambda --use_fast_math -D_GLIBCXX_USE_CXX11_ABI=0 -MD -MT _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_fp16_sm80.cu.o -MF _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_fp16_sm80.cu.o.d -x cu -c /tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src/flash_fwd_hdim160_fp16_sm80.cu -o _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_fp16_sm80.cu.o
      Killed
      [6/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_bf16_sm80.cu.o
      FAILED: _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_bf16_sm80.cu.o
      /usr/local/cuda/bin/nvcc -forward-unknown-to-host-compiler -DFLASHATTENTION_DISABLE_DROPOUT -DFLASHATTENTION_DISABLE_UNEVEN_K -DPy_LIMITED_API=3 -DTORCH_EXTENSION_NAME=vllm_flash_attn_c -DUSE_C10D_GLOO -DUSE_C10D_NCCL -DUSE_DISTRIBUTED -DUSE_RPC -DUSE_TENSORPIPE -Dvllm_flash_attn_c_EXPORTS -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/cutlass/include -isystem /usr/include/python3.10 -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include/torch/csrc/api/include -isystem /usr/local/cuda/include -DONNX_NAMESPACE=onnx_c2 -Xcudafe --diag_suppress=cc_clobber_ignored,--diag_suppress=field_without_dll_interface,--diag_suppress=base_class_has_different_dll_interface,--diag_suppress=dll_interface_conflict_none_assumed,--diag_suppress=dll_interface_conflict_dllexport_assumed,--diag_suppress=bad_friend_decl --expt-relaxed-constexpr --expt-extended-lambda -O3 -g -DNDEBUG -std=c++17 "--generate-code=arch=compute_70,code=[sm_70]" "--generate-code=arch=compute_75,code=[sm_75]" "--generate-code=arch=compute_80,code=[sm_80]" "--generate-code=arch=compute_86,code=[sm_86]" "--generate-code=arch=compute_90,code=[sm_90]" "--generate-code=arch=compute_90,code=[compute_90]" -Xcompiler=-fPIC --expt-relaxed-constexpr -DENABLE_FP8 --threads=1 --expt-extended-lambda --use_fast_math -D_GLIBCXX_USE_CXX11_ABI=0 -MD -MT _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_bf16_sm80.cu.o -MF _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_bf16_sm80.cu.o.d -x cu -c /tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src/flash_fwd_hdim160_bf16_sm80.cu -o _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_bf16_sm80.cu.o
      Killed
      [7/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_bf16_causal_sm80.cu.o
      FAILED: _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_bf16_causal_sm80.cu.o
      /usr/local/cuda/bin/nvcc -forward-unknown-to-host-compiler -DFLASHATTENTION_DISABLE_DROPOUT -DFLASHATTENTION_DISABLE_UNEVEN_K -DPy_LIMITED_API=3 -DTORCH_EXTENSION_NAME=vllm_flash_attn_c -DUSE_C10D_GLOO -DUSE_C10D_NCCL -DUSE_DISTRIBUTED -DUSE_RPC -DUSE_TENSORPIPE -Dvllm_flash_attn_c_EXPORTS -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/cutlass/include -isystem /usr/include/python3.10 -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include/torch/csrc/api/include -isystem /usr/local/cuda/include -DONNX_NAMESPACE=onnx_c2 -Xcudafe --diag_suppress=cc_clobber_ignored,--diag_suppress=field_without_dll_interface,--diag_suppress=base_class_has_different_dll_interface,--diag_suppress=dll_interface_conflict_none_assumed,--diag_suppress=dll_interface_conflict_dllexport_assumed,--diag_suppress=bad_friend_decl --expt-relaxed-constexpr --expt-extended-lambda -O3 -g -DNDEBUG -std=c++17 "--generate-code=arch=compute_70,code=[sm_70]" "--generate-code=arch=compute_75,code=[sm_75]" "--generate-code=arch=compute_80,code=[sm_80]" "--generate-code=arch=compute_86,code=[sm_86]" "--generate-code=arch=compute_90,code=[sm_90]" "--generate-code=arch=compute_90,code=[compute_90]" -Xcompiler=-fPIC --expt-relaxed-constexpr -DENABLE_FP8 --threads=1 --expt-extended-lambda --use_fast_math -D_GLIBCXX_USE_CXX11_ABI=0 -MD -MT _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_bf16_causal_sm80.cu.o -MF _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_bf16_causal_sm80.cu.o.d -x cu -c /tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src/flash_fwd_hdim160_bf16_causal_sm80.cu -o _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_bf16_causal_sm80.cu.o
      Killed
      [8/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim256_fp16_sm80.cu.o
      FAILED: _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim256_fp16_sm80.cu.o
      /usr/local/cuda/bin/nvcc -forward-unknown-to-host-compiler -DFLASHATTENTION_DISABLE_DROPOUT -DFLASHATTENTION_DISABLE_UNEVEN_K -DPy_LIMITED_API=3 -DTORCH_EXTENSION_NAME=vllm_flash_attn_c -DUSE_C10D_GLOO -DUSE_C10D_NCCL -DUSE_DISTRIBUTED -DUSE_RPC -DUSE_TENSORPIPE -Dvllm_flash_attn_c_EXPORTS -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/cutlass/include -isystem /usr/include/python3.10 -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include/torch/csrc/api/include -isystem /usr/local/cuda/include -DONNX_NAMESPACE=onnx_c2 -Xcudafe --diag_suppress=cc_clobber_ignored,--diag_suppress=field_without_dll_interface,--diag_suppress=base_class_has_different_dll_interface,--diag_suppress=dll_interface_conflict_none_assumed,--diag_suppress=dll_interface_conflict_dllexport_assumed,--diag_suppress=bad_friend_decl --expt-relaxed-constexpr --expt-extended-lambda -O3 -g -DNDEBUG -std=c++17 "--generate-code=arch=compute_70,code=[sm_70]" "--generate-code=arch=compute_75,code=[sm_75]" "--generate-code=arch=compute_80,code=[sm_80]" "--generate-code=arch=compute_86,code=[sm_86]" "--generate-code=arch=compute_90,code=[sm_90]" "--generate-code=arch=compute_90,code=[compute_90]" -Xcompiler=-fPIC --expt-relaxed-constexpr -DENABLE_FP8 --threads=1 --expt-extended-lambda --use_fast_math -D_GLIBCXX_USE_CXX11_ABI=0 -MD -MT _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim256_fp16_sm80.cu.o -MF _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim256_fp16_sm80.cu.o.d -x cu -c /tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src/flash_fwd_hdim256_fp16_sm80.cu -o _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim256_fp16_sm80.cu.o
      Killed
      [9/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim224_bf16_causal_sm80.cu.o
      FAILED: _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim224_bf16_causal_sm80.cu.o
      /usr/local/cuda/bin/nvcc -forward-unknown-to-host-compiler -DFLASHATTENTION_DISABLE_DROPOUT -DFLASHATTENTION_DISABLE_UNEVEN_K -DPy_LIMITED_API=3 -DTORCH_EXTENSION_NAME=vllm_flash_attn_c -DUSE_C10D_GLOO -DUSE_C10D_NCCL -DUSE_DISTRIBUTED -DUSE_RPC -DUSE_TENSORPIPE -Dvllm_flash_attn_c_EXPORTS -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/cutlass/include -isystem /usr/include/python3.10 -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include/torch/csrc/api/include -isystem /usr/local/cuda/include -DONNX_NAMESPACE=onnx_c2 -Xcudafe --diag_suppress=cc_clobber_ignored,--diag_suppress=field_without_dll_interface,--diag_suppress=base_class_has_different_dll_interface,--diag_suppress=dll_interface_conflict_none_assumed,--diag_suppress=dll_interface_conflict_dllexport_assumed,--diag_suppress=bad_friend_decl --expt-relaxed-constexpr --expt-extended-lambda -O3 -g -DNDEBUG -std=c++17 "--generate-code=arch=compute_70,code=[sm_70]" "--generate-code=arch=compute_75,code=[sm_75]" "--generate-code=arch=compute_80,code=[sm_80]" "--generate-code=arch=compute_86,code=[sm_86]" "--generate-code=arch=compute_90,code=[sm_90]" "--generate-code=arch=compute_90,code=[compute_90]" -Xcompiler=-fPIC --expt-relaxed-constexpr -DENABLE_FP8 --threads=1 --expt-extended-lambda --use_fast_math -D_GLIBCXX_USE_CXX11_ABI=0 -MD -MT _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim224_bf16_causal_sm80.cu.o -MF _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim224_bf16_causal_sm80.cu.o.d -x cu -c /tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src/flash_fwd_hdim224_bf16_causal_sm80.cu -o _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim224_bf16_causal_sm80.cu.o
      Killed
      [10/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim96_bf16_causal_sm80.cu.o
      FAILED: _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim96_bf16_causal_sm80.cu.o
      /usr/local/cuda/bin/nvcc -forward-unknown-to-host-compiler -DFLASHATTENTION_DISABLE_DROPOUT -DFLASHATTENTION_DISABLE_UNEVEN_K -DPy_LIMITED_API=3 -DTORCH_EXTENSION_NAME=vllm_flash_attn_c -DUSE_C10D_GLOO -DUSE_C10D_NCCL -DUSE_DISTRIBUTED -DUSE_RPC -DUSE_TENSORPIPE -Dvllm_flash_attn_c_EXPORTS -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/cutlass/include -isystem /usr/include/python3.10 -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include/torch/csrc/api/include -isystem /usr/local/cuda/include -DONNX_NAMESPACE=onnx_c2 -Xcudafe --diag_suppress=cc_clobber_ignored,--diag_suppress=field_without_dll_interface,--diag_suppress=base_class_has_different_dll_interface,--diag_suppress=dll_interface_conflict_none_assumed,--diag_suppress=dll_interface_conflict_dllexport_assumed,--diag_suppress=bad_friend_decl --expt-relaxed-constexpr --expt-extended-lambda -O3 -g -DNDEBUG -std=c++17 "--generate-code=arch=compute_70,code=[sm_70]" "--generate-code=arch=compute_75,code=[sm_75]" "--generate-code=arch=compute_80,code=[sm_80]" "--generate-code=arch=compute_86,code=[sm_86]" "--generate-code=arch=compute_90,code=[sm_90]" "--generate-code=arch=compute_90,code=[compute_90]" -Xcompiler=-fPIC --expt-relaxed-constexpr -DENABLE_FP8 --threads=1 --expt-extended-lambda --use_fast_math -D_GLIBCXX_USE_CXX11_ABI=0 -MD -MT _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim96_bf16_causal_sm80.cu.o -MF _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim96_bf16_causal_sm80.cu.o.d -x cu -c /tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src/flash_fwd_hdim96_bf16_causal_sm80.cu -o _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim96_bf16_causal_sm80.cu.o
      Killed
      [11/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim96_fp16_causal_sm80.cu.o
      FAILED: _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim96_fp16_causal_sm80.cu.o
      /usr/local/cuda/bin/nvcc -forward-unknown-to-host-compiler -DFLASHATTENTION_DISABLE_DROPOUT -DFLASHATTENTION_DISABLE_UNEVEN_K -DPy_LIMITED_API=3 -DTORCH_EXTENSION_NAME=vllm_flash_attn_c -DUSE_C10D_GLOO -DUSE_C10D_NCCL -DUSE_DISTRIBUTED -DUSE_RPC -DUSE_TENSORPIPE -Dvllm_flash_attn_c_EXPORTS -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/cutlass/include -isystem /usr/include/python3.10 -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include/torch/csrc/api/include -isystem /usr/local/cuda/include -DONNX_NAMESPACE=onnx_c2 -Xcudafe --diag_suppress=cc_clobber_ignored,--diag_suppress=field_without_dll_interface,--diag_suppress=base_class_has_different_dll_interface,--diag_suppress=dll_interface_conflict_none_assumed,--diag_suppress=dll_interface_conflict_dllexport_assumed,--diag_suppress=bad_friend_decl --expt-relaxed-constexpr --expt-extended-lambda -O3 -g -DNDEBUG -std=c++17 "--generate-code=arch=compute_70,code=[sm_70]" "--generate-code=arch=compute_75,code=[sm_75]" "--generate-code=arch=compute_80,code=[sm_80]" "--generate-code=arch=compute_86,code=[sm_86]" "--generate-code=arch=compute_90,code=[sm_90]" "--generate-code=arch=compute_90,code=[compute_90]" -Xcompiler=-fPIC --expt-relaxed-constexpr -DENABLE_FP8 --threads=1 --expt-extended-lambda --use_fast_math -D_GLIBCXX_USE_CXX11_ABI=0 -MD -MT _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim96_fp16_causal_sm80.cu.o -MF _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim96_fp16_causal_sm80.cu.o.d -x cu -c /tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src/flash_fwd_hdim96_fp16_causal_sm80.cu -o _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim96_fp16_causal_sm80.cu.o
      Killed
      [12/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim128_fp16_sm80.cu.o
      FAILED: _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim128_fp16_sm80.cu.o
      /usr/local/cuda/bin/nvcc -forward-unknown-to-host-compiler -DFLASHATTENTION_DISABLE_DROPOUT -DFLASHATTENTION_DISABLE_UNEVEN_K -DPy_LIMITED_API=3 -DTORCH_EXTENSION_NAME=vllm_flash_attn_c -DUSE_C10D_GLOO -DUSE_C10D_NCCL -DUSE_DISTRIBUTED -DUSE_RPC -DUSE_TENSORPIPE -Dvllm_flash_attn_c_EXPORTS -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/cutlass/include -isystem /usr/include/python3.10 -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include/torch/csrc/api/include -isystem /usr/local/cuda/include -DONNX_NAMESPACE=onnx_c2 -Xcudafe --diag_suppress=cc_clobber_ignored,--diag_suppress=field_without_dll_interface,--diag_suppress=base_class_has_different_dll_interface,--diag_suppress=dll_interface_conflict_none_assumed,--diag_suppress=dll_interface_conflict_dllexport_assumed,--diag_suppress=bad_friend_decl --expt-relaxed-constexpr --expt-extended-lambda -O3 -g -DNDEBUG -std=c++17 "--generate-code=arch=compute_70,code=[sm_70]" "--generate-code=arch=compute_75,code=[sm_75]" "--generate-code=arch=compute_80,code=[sm_80]" "--generate-code=arch=compute_86,code=[sm_86]" "--generate-code=arch=compute_90,code=[sm_90]" "--generate-code=arch=compute_90,code=[compute_90]" -Xcompiler=-fPIC --expt-relaxed-constexpr -DENABLE_FP8 --threads=1 --expt-extended-lambda --use_fast_math -D_GLIBCXX_USE_CXX11_ABI=0 -MD -MT _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim128_fp16_sm80.cu.o -MF _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim128_fp16_sm80.cu.o.d -x cu -c /tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src/flash_fwd_hdim128_fp16_sm80.cu -o _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim128_fp16_sm80.cu.o
      Killed
      [13/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim256_bf16_sm80.cu.o
      FAILED: _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim256_bf16_sm80.cu.o
      /usr/local/cuda/bin/nvcc -forward-unknown-to-host-compiler -DFLASHATTENTION_DISABLE_DROPOUT -DFLASHATTENTION_DISABLE_UNEVEN_K -DPy_LIMITED_API=3 -DTORCH_EXTENSION_NAME=vllm_flash_attn_c -DUSE_C10D_GLOO -DUSE_C10D_NCCL -DUSE_DISTRIBUTED -DUSE_RPC -DUSE_TENSORPIPE -Dvllm_flash_attn_c_EXPORTS -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/cutlass/include -isystem /usr/include/python3.10 -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include/torch/csrc/api/include -isystem /usr/local/cuda/include -DONNX_NAMESPACE=onnx_c2 -Xcudafe --diag_suppress=cc_clobber_ignored,--diag_suppress=field_without_dll_interface,--diag_suppress=base_class_has_different_dll_interface,--diag_suppress=dll_interface_conflict_none_assumed,--diag_suppress=dll_interface_conflict_dllexport_assumed,--diag_suppress=bad_friend_decl --expt-relaxed-constexpr --expt-extended-lambda -O3 -g -DNDEBUG -std=c++17 "--generate-code=arch=compute_70,code=[sm_70]" "--generate-code=arch=compute_75,code=[sm_75]" "--generate-code=arch=compute_80,code=[sm_80]" "--generate-code=arch=compute_86,code=[sm_86]" "--generate-code=arch=compute_90,code=[sm_90]" "--generate-code=arch=compute_90,code=[compute_90]" -Xcompiler=-fPIC --expt-relaxed-constexpr -DENABLE_FP8 --threads=1 --expt-extended-lambda --use_fast_math -D_GLIBCXX_USE_CXX11_ABI=0 -MD -MT _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim256_bf16_sm80.cu.o -MF _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim256_bf16_sm80.cu.o.d -x cu -c /tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src/flash_fwd_hdim256_bf16_sm80.cu -o _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim256_bf16_sm80.cu.o
      Killed
      [14/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim256_bf16_causal_sm80.cu.o
      FAILED: _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim256_bf16_causal_sm80.cu.o
      /usr/local/cuda/bin/nvcc -forward-unknown-to-host-compiler -DFLASHATTENTION_DISABLE_DROPOUT -DFLASHATTENTION_DISABLE_UNEVEN_K -DPy_LIMITED_API=3 -DTORCH_EXTENSION_NAME=vllm_flash_attn_c -DUSE_C10D_GLOO -DUSE_C10D_NCCL -DUSE_DISTRIBUTED -DUSE_RPC -DUSE_TENSORPIPE -Dvllm_flash_attn_c_EXPORTS -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/cutlass/include -isystem /usr/include/python3.10 -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include/torch/csrc/api/include -isystem /usr/local/cuda/include -DONNX_NAMESPACE=onnx_c2 -Xcudafe --diag_suppress=cc_clobber_ignored,--diag_suppress=field_without_dll_interface,--diag_suppress=base_class_has_different_dll_interface,--diag_suppress=dll_interface_conflict_none_assumed,--diag_suppress=dll_interface_conflict_dllexport_assumed,--diag_suppress=bad_friend_decl --expt-relaxed-constexpr --expt-extended-lambda -O3 -g -DNDEBUG -std=c++17 "--generate-code=arch=compute_70,code=[sm_70]" "--generate-code=arch=compute_75,code=[sm_75]" "--generate-code=arch=compute_80,code=[sm_80]" "--generate-code=arch=compute_86,code=[sm_86]" "--generate-code=arch=compute_90,code=[sm_90]" "--generate-code=arch=compute_90,code=[compute_90]" -Xcompiler=-fPIC --expt-relaxed-constexpr -DENABLE_FP8 --threads=1 --expt-extended-lambda --use_fast_math -D_GLIBCXX_USE_CXX11_ABI=0 -MD -MT _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim256_bf16_causal_sm80.cu.o -MF _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim256_bf16_causal_sm80.cu.o.d -x cu -c /tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src/flash_fwd_hdim256_bf16_causal_sm80.cu -o _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim256_bf16_causal_sm80.cu.o
      Killed
      [15/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_fp16_causal_sm80.cu.o
      FAILED: _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_fp16_causal_sm80.cu.o
      /usr/local/cuda/bin/nvcc -forward-unknown-to-host-compiler -DFLASHATTENTION_DISABLE_DROPOUT -DFLASHATTENTION_DISABLE_UNEVEN_K -DPy_LIMITED_API=3 -DTORCH_EXTENSION_NAME=vllm_flash_attn_c -DUSE_C10D_GLOO -DUSE_C10D_NCCL -DUSE_DISTRIBUTED -DUSE_RPC -DUSE_TENSORPIPE -Dvllm_flash_attn_c_EXPORTS -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src -I/tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/cutlass/include -isystem /usr/include/python3.10 -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include -isystem /tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/torch/include/torch/csrc/api/include -isystem /usr/local/cuda/include -DONNX_NAMESPACE=onnx_c2 -Xcudafe --diag_suppress=cc_clobber_ignored,--diag_suppress=field_without_dll_interface,--diag_suppress=base_class_has_different_dll_interface,--diag_suppress=dll_interface_conflict_none_assumed,--diag_suppress=dll_interface_conflict_dllexport_assumed,--diag_suppress=bad_friend_decl --expt-relaxed-constexpr --expt-extended-lambda -O3 -g -DNDEBUG -std=c++17 "--generate-code=arch=compute_70,code=[sm_70]" "--generate-code=arch=compute_75,code=[sm_75]" "--generate-code=arch=compute_80,code=[sm_80]" "--generate-code=arch=compute_86,code=[sm_86]" "--generate-code=arch=compute_90,code=[sm_90]" "--generate-code=arch=compute_90,code=[compute_90]" -Xcompiler=-fPIC --expt-relaxed-constexpr -DENABLE_FP8 --threads=1 --expt-extended-lambda --use_fast_math -D_GLIBCXX_USE_CXX11_ABI=0 -MD -MT _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_fp16_causal_sm80.cu.o -MF _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_fp16_causal_sm80.cu.o.d -x cu -c /tmp/tmp7hhhhz5w.build-temp/_deps/vllm-flash-attn-src/csrc/flash_attn/src/flash_fwd_hdim160_fp16_causal_sm80.cu -o _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim160_fp16_causal_sm80.cu.o
      Killed
      [16/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim192_bf16_causal_sm80.cu.o
      [17/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim192_fp16_causal_sm80.cu.o
      [18/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim32_bf16_causal_sm80.cu.o
      [19/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim32_fp16_causal_sm80.cu.o
      [20/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim192_fp16_sm80.cu.o
      [21/137] Building CUDA object CMakeFiles/_moe_C.dir/csrc/moe/marlin_moe_ops.cu.o
      [22/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim64_fp16_causal_sm80.cu.o
      [23/137] Building CUDA object CMakeFiles/_C.dir/csrc/cache_kernels.cu.o
      [24/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim64_bf16_causal_sm80.cu.o
      [25/137] Building CUDA object CMakeFiles/_moe_C.dir/csrc/moe/topk_softmax_kernels.cu.o
      [26/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim192_bf16_sm80.cu.o
      [27/137] Building CUDA object CMakeFiles/_C.dir/csrc/pos_encoding_kernels.cu.o
      [28/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim32_bf16_sm80.cu.o
      [29/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim96_bf16_sm80.cu.o
      [30/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim224_fp16_causal_sm80.cu.o
      [31/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim32_fp16_sm80.cu.o
      [32/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim64_bf16_sm80.cu.o
      [33/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim256_fp16_causal_sm80.cu.o
      [34/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim96_fp16_sm80.cu.o
      [35/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim64_fp16_sm80.cu.o
      [36/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim128_fp16_causal_sm80.cu.o
      [37/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim160_fp16_causal_sm80.cu.o
      [38/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim224_bf16_sm80.cu.o
      [39/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim224_fp16_sm80.cu.o
      [40/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim128_bf16_causal_sm80.cu.o
      [41/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim160_bf16_causal_sm80.cu.o
      [42/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim192_bf16_causal_sm80.cu.o
      [43/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim224_bf16_causal_sm80.cu.o
      [44/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim224_fp16_causal_sm80.cu.o
      [45/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim32_bf16_causal_sm80.cu.o
      [46/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim256_bf16_causal_sm80.cu.o
      [47/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_hdim128_bf16_sm80.cu.o
      [48/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim192_fp16_causal_sm80.cu.o
      [49/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim32_fp16_causal_sm80.cu.o
      [50/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim256_fp16_causal_sm80.cu.o
      [51/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim128_bf16_causal_sm80.cu.o
      [52/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim128_fp16_causal_sm80.cu.o
      [53/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim160_fp16_sm80.cu.o
      [54/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim64_bf16_causal_sm80.cu.o
      [55/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim160_bf16_sm80.cu.o
      [56/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim224_bf16_sm80.cu.o
      [57/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim192_bf16_sm80.cu.o
      [58/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim192_fp16_sm80.cu.o
      [59/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim256_bf16_sm80.cu.o
      [60/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim32_bf16_sm80.cu.o
      [61/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim224_fp16_sm80.cu.o
      [62/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim32_fp16_sm80.cu.o
      [63/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim256_fp16_sm80.cu.o
      [64/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim128_bf16_sm80.cu.o
      [65/137] Building CUDA object CMakeFiles/_moe_C.dir/csrc/moe/marlin_kernels/marlin_moe_kernel_ku4b8.cu.o
      [66/137] Building CUDA object CMakeFiles/_moe_C.dir/csrc/moe/marlin_kernels/marlin_moe_kernel_ku8b128.cu.o
      [67/137] Building CUDA object _deps/vllm-flash-attn-build/CMakeFiles/vllm_flash_attn_c.dir/csrc/flash_attn/src/flash_fwd_split_hdim128_fp16_sm80.cu.o
      [68/137] Building CUDA object CMakeFiles/_C.dir/csrc/attention/attention_kernels.cu.o
      ninja: build stopped: subcommand failed.
      Traceback (most recent call last):
        File "/tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/setuptools/command/editable_wheel.py", line 138, in run
          self._create_wheel_file(bdist_wheel)
        File "/tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/setuptools/command/editable_wheel.py", line 341, in _create_wheel_file
          files, mapping = self._run_build_commands(dist_name, unpacked, lib, tmp)
        File "/tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/setuptools/command/editable_wheel.py", line 264, in _run_build_commands
          self._run_build_subcommands()
        File "/tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/setuptools/command/editable_wheel.py", line 291, in _run_build_subcommands
          self.run_command(name)
        File "/usr/lib/python3.10/distutils/cmd.py", line 313, in run_command
          self.distribution.run_command(command)
        File "/tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/setuptools/dist.py", line 950, in run_command
          super().run_command(command)
        File "/usr/lib/python3.10/distutils/dist.py", line 985, in run_command
          cmd_obj.run()
        File "<string>", line 235, in run
        File "/tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/setuptools/command/build_ext.py", line 98, in run
          _build_ext.run(self)
        File "/usr/lib/python3.10/distutils/command/build_ext.py", line 340, in run
          self.build_extensions()
        File "<string>", line 209, in build_extensions
        File "/usr/lib/python3.10/subprocess.py", line 369, in check_call
          raise CalledProcessError(retcode, cmd)
      subprocess.CalledProcessError: Command '['cmake', '--build', '.', '-j=64', '--target=_core_C', '--target=_moe_C', '--target=vllm_flash_attn_c', '--target=_C']' returned non-zero exit status 1.
      /usr/lib/python3.10/distutils/dist.py:985: _DebuggingTips: Problem in editable installation.
      !!

              ********************************************************************************
              An error happened while installing `vllm` in editable mode.

              The following steps are recommended to help debug this problem:

              - Try to install the project normally, without using the editable mode.
                Does the error still persist?
                (If it does, try fixing the problem before attempting the editable mode).
              - If you are using binary extensions, make sure you have all OS-level
                dependencies installed (e.g. compilers, toolchains, binary libraries, ...).
              - Try the latest version of setuptools (maybe the error was already fixed).
              - If you (or your project dependencies) are using any setuptools extension
                or customization, make sure they support the editable mode.

              After following the steps above, if the problem still persists and
              you think this is related to how setuptools handles editable installations,
              please submit a reproducible example
              (see https://stackoverflow.com/help/minimal-reproducible-example) to:

                  https://github.com/pypa/setuptools/issues

              See https://setuptools.pypa.io/en/latest/userguide/development_mode.html for details.
              ********************************************************************************

      !!
        cmd_obj.run()
      Traceback (most recent call last):
        File "/usr/local/lib/python3.10/dist-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 353, in <module>
          main()
        File "/usr/local/lib/python3.10/dist-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 335, in main
          json_out['return_val'] = hook(**hook_input['kwargs'])
        File "/usr/local/lib/python3.10/dist-packages/pip/_vendor/pyproject_hooks/_in_process/_in_process.py", line 273, in build_editable
          return hook(wheel_directory, config_settings, metadata_directory)
        File "/tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/setuptools/build_meta.py", line 459, in build_editable
          return self._build_with_temp_dir(
        File "/tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/setuptools/build_meta.py", line 403, in _build_with_temp_dir
          self.run_setup()
        File "/tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/setuptools/build_meta.py", line 318, in run_setup
          exec(code, locals())
        File "<string>", line 483, in <module>
        File "/tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/setuptools/__init__.py", line 117, in setup
          return distutils.core.setup(**attrs)
        File "/usr/lib/python3.10/distutils/core.py", line 148, in setup
          dist.run_commands()
        File "/usr/lib/python3.10/distutils/dist.py", line 966, in run_commands
          self.run_command(cmd)
        File "/tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/setuptools/dist.py", line 950, in run_command
          super().run_command(command)
        File "/usr/lib/python3.10/distutils/dist.py", line 985, in run_command
          cmd_obj.run()
        File "/tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/setuptools/command/editable_wheel.py", line 138, in run
          self._create_wheel_file(bdist_wheel)
        File "/tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/setuptools/command/editable_wheel.py", line 341, in _create_wheel_file
          files, mapping = self._run_build_commands(dist_name, unpacked, lib, tmp)
        File "/tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/setuptools/command/editable_wheel.py", line 264, in _run_build_commands
          self._run_build_subcommands()
        File "/tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/setuptools/command/editable_wheel.py", line 291, in _run_build_subcommands
          self.run_command(name)
        File "/usr/lib/python3.10/distutils/cmd.py", line 313, in run_command
          self.distribution.run_command(command)
        File "/tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/setuptools/dist.py", line 950, in run_command
          super().run_command(command)
        File "/usr/lib/python3.10/distutils/dist.py", line 985, in run_command
          cmd_obj.run()
        File "<string>", line 235, in run
        File "/tmp/pip-build-env-jlatwit1/overlay/local/lib/python3.10/dist-packages/setuptools/command/build_ext.py", line 98, in run
          _build_ext.run(self)
        File "/usr/lib/python3.10/distutils/command/build_ext.py", line 340, in run
          self.build_extensions()
        File "<string>", line 209, in build_extensions
        File "/usr/lib/python3.10/subprocess.py", line 369, in check_call
          raise CalledProcessError(retcode, cmd)
      subprocess.CalledProcessError: Command '['cmake', '--build', '.', '-j=64', '--target=_core_C', '--target=_moe_C', '--target=vllm_flash_attn_c', '--target=_C']' returned non-zero exit status 1.
      [end of output]

  note: This error originates from a subprocess, and is likely not a problem with pip.
  ERROR: Failed building editable for vllm
Failed to build vllm
ERROR: Could not build wheels for vllm, which is required to install pyproject.toml-based projects

Before submitting a new issue...

  • Make sure you already searched for relevant issues, and asked the chatbot living at the bottom right corner of the documentation page, which can answer lots of frequently asked questions.
@KevinRSX KevinRSX added the installation Installation problems label Sep 27, 2024
@dengminhao
Copy link

dengminhao commented Sep 27, 2024 •
edited
Loading

That is OOM-kill.
export MAX_JOBS=[half-your-cpu-cores]
and retry. If you free memory is very low, you can set it to 1, but it will take a lot of time to compile.

@Imss27
Copy link
Contributor

Imss27 commented Sep 27, 2024

If it is OOM issue, this might help #8532
Also, #8818 and #8550 updated the documentation for similar issues.

@KevinRSX
Copy link
Author

Yeah, this seems like an OOM to me. The machine I used to report this issue has 128 GB memory, and then I switched to a cluster with 1 TB and I could see the peak usage when compiling the flash attention kernel was around 300.

The doc update looks very helpful. I'll definitely try it out.

Btw, I'm just curious as to why the compilation guide says it can be done in 5-10 minutes, while I use a powerful cluster and it took me 2 hours? Did I do something terribly wrong?

@KevinRSX
Copy link
Author

Popping this up - any idea why the compilation takes much longer than what was noted on the Doc? Thanks!

@youkaichao
Copy link
Member

5-10 minutes

it was an estimation long time ago. many kernels are added recently.

if you don't touch kernels, I would suggest using the python-only build from source approach, see https://docs.vllm.ai/en/latest/getting_started/installation.html#build-from-source-without-compilation

@youkaichao youkaichao linked a pull request Sep 29, 2024 that will close this issue
[doc] organize installation doc and expose per-commit docker #8931
Merged
@youkaichao
Copy link
Member

I updated the description in #8931

@KevinRSX
Copy link
Author

Thanks!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
installation Installation problems
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

[doc] organize installation doc and expose per-commit docker youkaichao/vllm
4 participants
@youkaichao @KevinRSX @Imss27 @dengminhao