Mark Blackman wrote:
O. Hartmann wrote:
well, my question may sound heretic, but since we use mostly Linux based
systems in our scientific environment and FreeBSD seems to lack in
severe support in GPGPU/CUDA capable graphics boards
Hmm, interesting requirement. Sounds like a nice Google SoC project or
even something funded.
This project is by far too much for a gsoc student not to mention who
would mentor it... People throw around the term "GPGPU/CUDA" too much
and don't realize that it breaks down into...
1) Kernel drivers (not ported to FreeBSD - see PSCNV.. There has other
related things like drm, firmware. etc)
2) Runtime (do you need CUDA API, OpenCL.. etc)
3) Assembler
4) Code generation (ptx, direct isa.. which can be nv50, nvc0. etc)
5) Programming model (front-end which is typically CUDA/OpenCL or lesser
known HMPP)
_______________________________________________
freebsd-questions@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-questions
To unsubscribe, send any mail to "freebsd-questions-unsubscr...@freebsd.org"
