Open mihaibujanca opened 5 years ago
@Paul92 trace
Got sensor of type PointCloud
Got sensor of type Camera
Got sensor of type Depth
Got sensor of type Camera
Got sensor of type GroundTruthTrajectory
[OK] Initializing sensor
Performing sanity checks
SB sensor 0x5555558ed590
rgb_sensor 0x5555558639b0
Rgb sensor file format 1 ... [OK]
Depth sensor file format 1 ... [OK]
Rgb sensor pixel format 2 ... [OK]
Depth sensor pixel format 5 ... [OK]
Sensor size match ... [OK]
Initializing slam system
[New Thread 0x7fffeabc3700 (LWP 5437)]
GPUs found:
deviceId: 0: TITAN RTX
deviceId: 0: TITAN RTX (emulated)
Connecting to sensor...
sensor dimensions depth ( 640 / 480 )
sensor dimensions color ( 640 / 480 )
depth intrinsics
481.1999816895 0.0000000000 319.4998474121 0.0000000000
0.0000000000 480.0000000000 239.4999847412 0.0000000000
0.0000000000 0.0000000000 1.0000000000 0.0000000000
0.0000000000 0.0000000000 0.0000000000 1.0000000000
color intrinsics
481.1999816895 0.0000000000 319.4998474121 0.0000000000
0.0000000000 480.0000000000 239.4999847412 0.0000000000
0.0000000000 0.0000000000 1.0000000000 0.0000000000
0.0000000000 0.0000000000 0.0000000000 1.0000000000
creating image manager
[New Thread 0x7fffea295700 (LWP 5438)]
[New Thread 0x7fffe9911700 (LWP 5439)]
Running singlethreaded
Cuda API error detected: cudaMemcpyToSymbol returned (0xd)
(cuda-gdb) where
#0 0x00007ffff0f59010 in cudbgReportDriverApiError () from /usr/lib/x86_64-linux-gnu/libcuda.so.1
#1 0x00007ffff0f5beea in cudbgReportDriverInternalError () from /usr/lib/x86_64-linux-gnu/libcuda.so.1
#2 0x00007ffff0f5f263 in cudbgApiDetach () from /usr/lib/x86_64-linux-gnu/libcuda.so.1
#3 0x00007ffff118c5f3 in cudbgMain () from /usr/lib/x86_64-linux-gnu/libcuda.so.1
#4 0x00007ffff119ec33 in cudbgMain () from /usr/lib/x86_64-linux-gnu/libcuda.so.1
#5 0x00007ffff2acbd4c in cudaMemcpyToSymbol () from /home/mihai/Projects/slambench3/build/lib/libbundlefusion-original.so
#6 0x00007ffff2a83483 in cudaError cudaMemcpyToSymbol<float [30][33]>(float const (&) [30][33], void const*, unsigned long, unsigned long, cudaMemcpyKind) ()
from /home/mihai/Projects/slambench3/build/lib/libbundlefusion-original.so
#7 0x00007ffff2a79a80 in ProgramCU::InitFilterKernels(std::vector<float, std::allocator<float> > const&, std::vector<unsigned int, std::allocator<unsigned int> >&) ()
from /home/mihai/Projects/slambench3/build/lib/libbundlefusion-original.so
#8 0x00007ffff2a2d69d in SiftParam::ParseSiftParam (this=this@entry=0x5555560271f0) at /home/mihai/Projects/slambench3/benchmarks/bundlefusion/src/original/FriedLiver/Source/SiftGPU/SiftGPU.cpp:175
#9 0x00007ffff2a2d8ce in SiftGPU::InitSiftGPU (this=0x5555560271f0) at /home/mihai/Projects/slambench3/benchmarks/bundlefusion/src/original/FriedLiver/Source/SiftGPU/SiftGPU.cpp:71
#10 0x00007ffff29aed5e in Bundler::initSift (this=this@entry=0x55555601cb00, widthSift=640, heightSift=heightSift@entry=480, isLocal=isLocal@entry=true)
at /home/mihai/Projects/slambench3/benchmarks/bundlefusion/src/original/FriedLiver/Source/Bundler.cpp:62
#11 0x00007ffff29b1093 in Bundler::Bundler (this=0x55555601cb00, maxNumImages=11, maxNumKeysPerImage=1024, siftIntrinsicsInv=..., manager=0x55555590d0c0, isLocal=<optimised out>)
at /home/mihai/Projects/slambench3/benchmarks/bundlefusion/src/original/FriedLiver/Source/Bundler.cpp:23
#12 0x00007ffff29a86f8 in OnlineBundler::OnlineBundler (this=0x555556024c00, sensor=<optimised out>, imageManager=0x55555590d0c0)
at /home/mihai/Projects/slambench3/benchmarks/bundlefusion/src/original/FriedLiver/Source/OnlineBundler.cpp:41
#13 0x00007ffff29210fe in sb_init_slam_system (slam_settings=0x5555558845b0) at /home/mihai/Projects/slambench3/benchmarks/bundlefusion/src/original/bundlefusion_library.cpp:147
#14 0x000055555558d91d in SLAMBenchConfiguration::InitAlgorithms (this=<optimised out>) at /home/mihai/Projects/slambench3/framework/shared/src/SLAMBenchConfiguration.cpp:294
#15 0x0000555555586b98 in main (argc=9, argv=<optimised out>) at /home/mihai/Projects/slambench3/framework/tools/loaders/benchmark_loader.cpp:91
@Paul92, when attempting to run BundleFusion, the following error is encountered:
Attempted to run it on a laptop with an NVidia Geforce 960M, CUDA 9.0 and gcc-6. Also attempted running on a Titan RTX 24GB, CUDA 10 + gcc 8, as well as CUDA 9 + gcc-6.
Full log: