ERROR: SiftGPU not fully supported (Windows 11 + WSL2 + Docker) #1289
Description
Hi guys!
I'm trying to use colmap app from your docker image and I got an error when I run it.
My system:
Windows 11 21H2
Wsl 2
Docker v20.10.8
Nvidia driver v510.06 (as I know it is the version for CUDA, but I also tried v471.96)
GPU: Nvidia GeForce 1070 6gb
CUDA drivers v10 (I also tried v11)
Visual Studio 2019 Community
What I did:
- Installed VS2019
- Installed CUDA drivers
- Downloaded and run docker image.
- Run
colmap automatic_reconstructor --image_path ./images --workspace_path .inside the container and got
Processed file [1/24]
Name: 2020-05-15 14-27-34.JPG
Dimensions: 4608 x 3456
Camera: #2 - SIMPLE_RADIAL
Focal Length: 3291.43px (Prior)
GPS: LAT=53.409, LON=2.227, ALT=94.799
Features: 9106
Processed file [2/24]
Name: 2020-05-15 14-27-57.JPG
Dimensions: 4608 x 3456
Camera: #4 - SIMPLE_RADIAL
Focal Length: 3291.43px (Prior)
GPS: LAT=53.409, LON=2.227, ALT=94.799
Features: 8472
...
...
...
And then A lot of
colmap CUDA driver version is insufficient for CUDA runtime version
ERROR: SiftGPU not fully supported
What did I do wrong?
I also run one test from CUDA tests:
Seems that it's ok
Attaching to testcuda_test_1
test_1 | 2021-09-03 13:47:15.754615: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 FMA
test_1 | To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
test_1 | 2021-09-03 13:47:15.773321: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node
test_1 | Your kernel may have been built without NUMA support.
test_1 | 2021-09-03 13:47:15.777980: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node
test_1 | Your kernel may have been built without NUMA support.
test_1 | 2021-09-03 13:47:15.778364: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node
test_1 | Your kernel may have been built without NUMA support.
test_1 | 2021-09-03 13:47:16.188330: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node
test_1 | Your kernel may have been built without NUMA support.
test_1 | 2021-09-03 13:47:16.188826: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node
test_1 | Your kernel may have been built without NUMA support.
test_1 | 2021-09-03 13:47:16.188977: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1594] Could not identify NUMA node of platform GPU id 0, defaulting to 0. Your kernel may not have been built with NUMA support.
test_1 | 2021-09-03 13:47:16.189186: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node
test_1 | Your kernel may have been built without NUMA support.
test_1 | 2021-09-03 13:47:16.189352: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1510] Created device /device:GPU:0 with 6623 MB memory: -> device: 0, name: NVIDIA GeForce GTX 1070, pci bus id: 0000:01:00.0, compute capability: 6.1
test_1 | 2021-09-04 14:46:19.006939: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 FMA
test_1 | To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
test_1 | 2021-09-04 14:46:19.156315: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node
test_1 | Your kernel may have been built without NUMA support.
test_1 | 2021-09-04 14:46:19.181850: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node
test_1 | Your kernel may have been built without NUMA support.
test_1 | 2021-09-04 14:46:19.182056: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node
test_1 | Your kernel may have been built without NUMA support.
test_1 | 2021-09-04 14:46:20.109325: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node
test_1 | Your kernel may have been built without NUMA support.
test_1 | 2021-09-04 14:46:20.109600: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node
test_1 | Your kernel may have been built without NUMA support.
test_1 | 2021-09-04 14:46:20.109625: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1594] Could not identify NUMA node of platform GPU id 0, defaulting to 0. Your kernel may not have been built with NUMA support.
test_1 | 2021-09-04 14:46:20.109884: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node
test_1 | Your kernel may have been built without NUMA support.
test_1 | 2021-09-04 14:46:20.109934: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1510] Created device /device:GPU:0 with 6623 MB memory: -> device: 0, name: NVIDIA GeForce GTX 1070, pci bus id: 0000:01:00.0, compute capability: 6.1
test_1 | 2021-09-06 14:37:09.048992: I tensorflow/core/platform/cpu_feature_guard.cc:142] This TensorFlow binary is optimized with oneAPI Deep Neural Network Library (oneDNN) to use the following CPU instructions in performance-critical operations: AVX2 FMA
test_1 | To enable them in other operations, rebuild TensorFlow with the appropriate compiler flags.
test_1 | 2021-09-06 14:37:09.208695: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node
test_1 | Your kernel may have been built without NUMA support.
test_1 | 2021-09-06 14:37:09.245278: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node
test_1 | Your kernel may have been built without NUMA support.
test_1 | 2021-09-06 14:37:09.245532: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node
test_1 | Your kernel may have been built without NUMA support.
test_1 | 2021-09-06 14:37:10.565061: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node
test_1 | Your kernel may have been built without NUMA support.
test_1 | 2021-09-06 14:37:10.565253: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node
test_1 | Your kernel may have been built without NUMA support.
test_1 | 2021-09-06 14:37:10.565280: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1594] Could not identify NUMA node of platform GPU id 0, defaulting to 0. Your kernel may not have been built with NUMA support.
test_1 | 2021-09-06 14:37:10.565446: I tensorflow/stream_executor/cuda/cuda_gpu_executor.cc:923] could not open file to read NUMA node: /sys/bus/pci/devices/0000:01:00.0/numa_node
test_1 | Your kernel may have been built without NUMA support.
test_1 | 2021-09-06 14:37:10.565491: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1510] Created device /device:GPU:0 with 6623 MB memory: -> device: 0, name: NVIDIA GeForce GTX 1070, pci bus id: 0000:01:00.0, compute capability: 6.1
testcuda_test_1 exited with code 0