I am trying to run a simple CUDA kernel on my CUDA device, but even if it
compiles I cannot manage to get it run. The compilation output is the following:
~/x10-trunk_labsrv8/x10.dist/bin/x10c++ -O -classpath
/home/lsalucci/x10-trunk_labsrv8/x10.gml/lib/native_gml.jar -x10lib
/home/lsalucci/x10-trunk_labsrv8/x10.gml/native_gml.properties -post '# # # -L
/usr/lib/libblas -L /usr/lib/lapack -lblas -llapack' cuVector.x10
cuVectorMult.x10 cuSparseCSC.x10 -o cuGMLlib
x10c++: cuGML/cuVector.cu(44): Warning: Cannot tell what pointer points to,
assuming global memory space
cuGML/cuVector.cu(50): Warning: Cannot tell what pointer points to,
assuming global memory space
x10c++: cuGML/cuVector.cu(44): Warning: Cannot tell what pointer points to,
assuming global memory space
cuGML/cuVector.cu(50): Warning: Cannot tell what pointer points to,
assuming global memory space
x10c++: cuGML/cuVector.cu(44): Warning: Cannot tell what pointer points to,
assuming global memory space
cuGML/cuVector.cu(50): Warning: Cannot tell what pointer points to,
assuming global memory space
x10c++: cuGML/cuVector.cu(44): Warning: Cannot tell what pointer points to,
assuming global memory space
cuGML/cuVector.cu(50): Warning: Cannot tell what pointer points to,
assuming global memory space
Even though when I run it I get the following runtime error:
X10RT: async 78 is not a CUDA kernel.
Aborted (core dumped)
Any suggestion? Is the problem caused by my code or by the runtime?
Thanks,
Luca Salucci
------------------------------------------------------------------------------
Managing the Performance of Cloud-Based Applications
Take advantage of what the Cloud has to offer - Avoid Common Pitfalls.
Read the Whitepaper.
http://pubads.g.doubleclick.net/gampad/clk?id=121051231&iu=/4140/ostg.clktrk
_______________________________________________
X10-users mailing list
X10-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/x10-users