RuntimeError: CUDA error: invalid device function ROIAlign_forward_cuda

Attempting to forward inference the panoptic fpn model results in a CUDA error.

To Reproduce

Attempting to run a predictor using the model panoptic_fpn_R_101_dconv_cascade_gn_3x.yaml.

The following error is produced:

error in deformable_im2col: invalid device function
... < repeated ~30 times> ...
File "/opt/conda/lib/python3.6/site-packages/torch/autograd/grad_mode.py", line 49, in decorate_no_grad
    return func(*args, **kwargs)
  File "/app/detectron2/detectron2/engine/defaults.py", line 176, in __call__
    predictions = self.model([inputs])[0]
  File "/opt/conda/lib/python3.6/site-packages/torch/nn/modules/module.py", line 541, in __call__
    result = self.forward(*input, **kwargs)
  File "/app/detectron2/detectron2/modeling/meta_arch/panoptic_fpn.py", line 98, in forward
    images, features, proposals, gt_instances
  File "/opt/conda/lib/python3.6/site-packages/torch/nn/modules/module.py", line 541, in __call__
    result = self.forward(*input, **kwargs)
  File "/app/detectron2/detectron2/modeling/roi_heads/cascade_rcnn.py", line 97, in forward
    pred_instances = self._forward_box(features_list, proposals)
  File "/app/detectron2/detectron2/modeling/roi_heads/cascade_rcnn.py", line 112, in _forward_box
    head_outputs.append(self._run_stage(features, proposals, k))
  File "/app/detectron2/detectron2/modeling/roi_heads/cascade_rcnn.py", line 203, in _run_stage
    box_features = self.box_pooler(features, [x.proposal_boxes for x in proposals])
  File "/opt/conda/lib/python3.6/site-packages/torch/nn/modules/module.py", line 541, in __call__
    result = self.forward(*input, **kwargs)
  File "/app/detectron2/detectron2/modeling/poolers.py", line 192, in forward
    output[inds] = pooler(x_level, pooler_fmt_boxes_level)
  File "/opt/conda/lib/python3.6/site-packages/torch/nn/modules/module.py", line 541, in __call__
    result = self.forward(*input, **kwargs)
  File "/app/detectron2/detectron2/layers/roi_align.py", line 95, in forward
    input, rois, self.output_size, self.spatial_scale, self.sampling_ratio, self.aligned
  File "/app/detectron2/detectron2/layers/roi_align.py", line 20, in forward
    input, roi, spatial_scale, output_size[0], output_size[1], sampling_ratio, aligned
RuntimeError: CUDA error: invalid device function (ROIAlign_forward_cuda at /app/detectron2/detectron2/layers/csrc/ROIAlign/ROIAlign_cuda.cu:359)
frame #0: c10::Error::Error(c10::SourceLocation, std::string const&) + 0x47 (0x7ffa5c402687 in /opt/conda/lib/python3.6/site-packages/torch/lib/libc10.so)
frame #1: ROIAlign_forward_cuda(at::Tensor const&, at::Tensor const&, float, int, int, int, bool) + 0xa37 (0x7ffa0653b6f5 in /app/detectron2/detectron2/_C.cpython-36m-x86_64-linux-gnu.so)
frame #2: ROIAlign_forward(at::Tensor const&, at::Tensor const&, float, int, int, int, bool) + 0xbc (0x7ffa064c9fdc in /app/detectron2/detectron2/_C.cpython-36m-x86_64-linux-gnu.so)
frame #3: <unknown function> + 0x5961a (0x7ffa064db61a in /app/detectron2/detectron2/_C.cpython-36m-x86_64-linux-gnu.so)
frame #4: <unknown function> + 0x5971e (0x7ffa064db71e in /app/detectron2/detectron2/_C.cpython-36m-x86_64-linux-gnu.so)
frame #5: <unknown function> + 0x53ca0 (0x7ffa064d5ca0 in /app/detectron2/detectron2/_C.cpython-36m-x86_64-linux-gnu.so)
<omitting python frames>
frame #12: THPFunction_apply(_object*, _object*) + 0x9ff (0x7ffa5d63dacf in /opt/conda/lib/python3.6/site-packages/torch/lib/libtorch_python.so)

Environment

---------------------  -------------------------------------------------------------------
Python                 3.6.8 |Anaconda, Inc.| (default, Dec 30 2018, 01:22:34) [GCC 7.3.0]
Detectron2 Compiler    GCC 5.4
DETECTRON2_ENV_MODULE  <not set>
PyTorch                1.3.0
PyTorch Debug Build    False
CUDA available         True
GPU 0,1                GeForce GTX 1080 Ti
Pillow                 6.2.0
cv2                    3.4.4
---------------------  -------------------------------------------------------------------
PyTorch built with:
  - GCC 7.3
  - Intel(R) Math Kernel Library Version 2019.0.4 Product Build 20190411 for Intel(R) 64 architecture applications
  - Intel(R) MKL-DNN v0.20.5 (Git Hash 0125f28c61c1f822fd48570b4c1066f96fcb9b2e)
  - OpenMP 201511 (a.k.a. OpenMP 4.5)
  - NNPACK is enabled
  - CUDA Runtime 10.1
  - NVCC architecture flags: -gencode;arch=compute_35,code=sm_35;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_50,code=compute_50
  - CuDNN 7.6.3
  - Magma 2.5.1
  - Build settings: BLAS=MKL, BUILD_NAMEDTENSOR=OFF, BUILD_TYPE=Release, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -fopenmp -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -O2 -fPIC -Wno-narrowing -Wall -Wextra -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-variable -Wno-unused-function -Wno-unused-result -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Wno-stringop-overflow, DISABLE_NUMA=1, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, USE_CUDA=True, USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON, USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON, USE_STATIC_DISPATCH=OFF

[ppwwyyxx]

I was able to reproduce the same error when I use the wrong version of cuda.

What I did: I install pytorch from conda install pytorch torchvision cudatoolkit=10.1 -c pytorch, however my local cuda runtime and nvcc are in 10.0. In this case, I can observe the same error. Please check whether your cuda version is correct.

[ppwwyyxx]

It seems that mismatched NVCC vs CUDA Runtime version is the root cause. Closing but feel free to reopen if this does not solve your issue.