Hi!
I tried to find this information on internet but I didn’t success.
So can you please say me what is the size of a pointer in a GeForce GTX 580 (compute capability 2.0).
Thanks
Hi!
I tried to find this information on internet but I didn’t success.
So can you please say me what is the size of a pointer in a GeForce GTX 580 (compute capability 2.0).
Thanks
There is an option in NVCC that allows you compile to 32 or 64bit address space
I tried to use -m64 or -m32 but I can’t compile with -m32.
Is it by default compiled for 32bits and if I add -m64 it is done with 64 bits ?
The general CUDA approach is that the size of pointers in the device code match the size of pointers in the host code. So if you build a 32-bit application, host and device pointers will be 32 bit. If you build a 64-bit application, host and device pointers will be 64 bit.