Scott Gray <sgray(a)nervanasys.com> writes:
I'm curious if you noticed this little project:
With cuda 7, it seems like that could be leveraged to replace forking off
an instance of nvcc. I think compiling cuda-c in this way should be much
faster. I'm too busy to play with it right now, but I was wondering if you
had any plans to integrate this?
I wasn't aware of this, thanks for pointing it out! It'd be great to
integrate this. I personally can't spare the time right now, but I'd
love to take a patch.
Btw, thanks for all the work you've put into pycuda. It's truly a pleasure
to program in. I only wish I'd written my assembler in python to start out
with so I could integrate it with pycuda and get dynamically generated
assembly at run time. One of these days I'll port it over. I think my
perl days are over.
Glad to hear you're finding it useful!