Problem with running code with double precision values Double precision gives wrong result

[rmd22]

I am trying to run code with double precision on CUDA, but get wrong result. nvcc compiles the code correctly but while executing it on GPU there seems to be some problem.
The code for single precision works fine.
I am using Fedora 10 (Linux version 2.6.27.5-117.fc10.x86_64) on 64 bit AMD with Tesla c1060 GPU.
Feel free to discuss your views about this. Any help will be greatly appreciated.

[jph4599]

Do you have a device that’s compute capability 1.3?

Are you compiling for doubles?

You need to add the switch “-arch sm_13” to your nvcc command line, otherwise doubles will be silently demoted to floats. See the “Mandelbrot” sample in the CUDA SDK for an example of how to switch between different kernels based on the compute capability of the GPU.

You should also be careful to suffix all floating point literals with “f” (for example, “1.0f”) otherwise they will be interpreted as doubles by the compiler.

[rmd22]

Thanks for the reply. I’ll try it out. I am compiling my code for the double.