Closed zhuminjie closed 5 years ago
What is the output of nvidia-smi
(you need to run it from command line)?
Looks like error 2
is an 'out of memory error'. Try to reduce the problem size. Also, if the K20 is not the only Nvidia GPU on the machine, make sure you are using the correct one.
Thank you for the reply. The k20 is actually installed. I change cuda version to 10.0 and with -O3, it can run successfully. Not sure why, it seems working now!
Another question is that have you compared with Amgx from Nvidia ?
Another question is that have you compared with Amgx from Nvidia ?
Not really. If you manage to run the same tests though, please feel free to post the results here (or make a pull request with the testing code).
Sure, if I can run it, I will let you know.
Hi,
Thank you a lot for implementing this wonderful library. I really want to use it. The problem I have now is the benchmark file amgcl.cu in the shared memory for poisson equation. I have the K20m GPU installed and can compile this file. But When I run it, it gives,
terminate called after throwing an instance of 'std::runtime_error' what(): CUDA error 2 at "./amgcl/amgcl/backend/cuda.hpp:247 Abort (core dumped)
I didn't try anything in the source code but compile it with make
I really need your help to find anything I did wrong! Thank you so much!
Here is my makefile
CUDA_VERSION = 6.5.14 LINKFLAGS = -Xcompiler \"-Wl,-rpath,/usr/local/apps/cuda/cuda-$(CUDA_VERSION)/lib64\" LIBS = -L/usr/local/apps/cuda/cuda-$(CUDA_VERSION)/lib64 -lcudart -lcusparse
poisson_cu.o: poisson.cu $(NVCC) -std=c++11 -O2 $(INCLUDES) $(CUDA_INC) poisson.cu -c -o poisson_cu.o
poisson_cu: poisson_cu.o $(NVCC) $(LINKFLAGS) $(LIBS) poisson_cu.o -o poisson_cu