As the article says, cuBLAS is deterministic, but other CUDA primitives (eg. some of those in cudNN) are not.
Yes, the non-determinism is being introduced somewhere, but that is splitting hairs. The point is that the primitives that you work with on GPUs are non-deterministic by design.
I mostly take issue with you phrasing it as a bug and using it to insult the authors.
Yes, the non-determinism is being introduced somewhere, but that is splitting hairs. The point is that the primitives that you work with on GPUs are non-deterministic by design.
I mostly take issue with you phrasing it as a bug and using it to insult the authors.