Trying to install on fedora f13 x86_64. did:
sh ~/Download/cudatoolkit_3.2.9_linux_64_fedora13.run (install into /opt/cuda) sh ~/Download/gpucomputingsdk_3.2_linux.run (install into ~/NVIDIA_GPU_Computing_SDK) Then cd ~/NVIDIA_GPU_Computing_SDK/C make gets as far as: g++ -fPIC -m64 -o ../../bin/linux/release/vectorAddDrv obj/x86_64/release/vectorAdd.cpp.o -L/opt/cuda/lib64 -L../../lib - L../../common/lib/linux -L../../../shared//lib -lcuda - L/opt/cuda/lib64 -L../../lib -L../../common/lib/linux -L../../../shared//lib -lcutil_x86_64 -lshrutil_x86_64 -ldl -lpthread /usr/bin/ld: cannot find -lcuda If I add -L/usr/lib64/nvidia: g++ -fPIC -m64 -o ../../bin/linux/release/vectorAddDrv obj/x86_64/release/vectorAdd.cpp.o -L/opt/cuda/lib64 -L../../lib - L../../common/lib/linux -L../../../shared//lib -lcuda - L/opt/cuda/lib64 -L/usr/lib64/nvidia -L../../lib -L../../common/lib/linux - L../../../shared//lib -lcutil_x86_64 -lshrutil_x86_64 -ldl -lpthread obj/x86_64/release/vectorAdd.cpp.o: In function `CleanupNoFailure()': vectorAdd.cpp:(.text+0x151): undefined reference to `cuMemFree_v2' vectorAdd.cpp:(.text+0x162): undefined reference to `cuMemFree_v2' vectorAdd.cpp:(.text+0x173): undefined reference to `cuMemFree_v2' obj/x86_64/release/vectorAdd.cpp.o: In function `main': vectorAdd.cpp:(.text+0x358): undefined reference to `cuCtxCreate_v2' vectorAdd.cpp:(.text+0x647): undefined reference to `cuMemAlloc_v2' vectorAdd.cpp:(.text+0x66b): undefined reference to `cuMemAlloc_v2' vectorAdd.cpp:(.text+0x68f): undefined reference to `cuMemAlloc_v2' vectorAdd.cpp:(.text+0x6bc): undefined reference to `cuMemcpyHtoD_v2' vectorAdd.cpp:(.text+0x6e9): undefined reference to `cuMemcpyHtoD_v2' vectorAdd.cpp:(.text+0x84f): undefined reference to `cuMemcpyDtoH_v2' collect2: ld returned 1 exit status _______________________________________________ PyCUDA mailing list PyCUDA@tiker.net http://lists.tiker.net/listinfo/pycuda
