Re: Enabling ROCm on Everything
On Tue, 2023-03-21 at 19:41 +0100, Christian Kastner wrote:
>
> I agree that a split of some sort will probably be necessary in the
> short-to-midterm, and your proposed solutions looks reasonable to me.
>
> One difficulty we will need to figure out one way or another is how to
> actually bring the user to the right package. What do we do when the
> user wants to `apt install pytorch-rocm`?
If possible, I suggest we stick to only one single binary package that
supports multiple selected architectures. The pytorch-rocm popcon
will not likely to be very large to deserve such a special treatment.
Neither does pytorch-cuda.
> Another difficulty we might need to consider is: what if the system in
> question contains multiple GPU architectures (e.g. 6800 XT and 7900 XT)?
We link multiple gfx architectures in a single fat binary. Then the
multi-GPU arch issue is alleviated. By the way, I suppose the
mixture of GPU architecture is a rare configuration among pytorch
users.
> The above aren't arguments against, I'm rather just starting to collect
> TODOs.
I've managed to build pytorch-cuda locally. Probably we can work
together for pytorch-rocm and its reverse depenedencies after
I finalized the bits for pytorch-cuda.
Reply to: