cuda 2.3 run on fermi

can the code compiled by cuda2.3 run on fermi

yes, but only via compiling PTX and having later drivers (that support Fermi) JIT that to native Fermi binaries

I think the binaries I compiled with CUDA SDK 2.3 ran just fine on my Fermi device (GTX 460). I only use the runtime API for my projects.

As far as I know nvcc by default embeds the generated PTX code into the executable when using the 2.3 SDK or later. This allows the CUDA 4.0 driver to translate the code as needed for the Fermi architecture - even when the binary is shipped with the CUDA 2.3 runtime library.

I confess I haven’t switched to CUDA 3.x or 4.0 yet. 2.3 is working fine for all of my projects.


like this i got it


thanks for your valuable experience

thanks for your valuable experience

as I have no cuda-enabled cards I have to code and debug with cuda2.3 emulation mode.