News for hackers. Stuff that matters.
Re:How to Think About GPUs by einpoklum on Wednesday August 20, 2025 @01:17PM (#44963863)
> that can itself perform actual SIMD instructions?

Mostly, no; it can't really perform actual SIMD instructions itself. If you look at the SASS (the assembly language used on NVIDIA GPUs) I don't believe you'll see anything like that.

In high-level code, you do have expressions involving "vectorized types", which look like they would translate into SIMD instruction, but they 'serialize', at the single thread level.

There are exceptions to this though, like FP16 operations which might work on 2xFP16 32-bit registers, and other cases. But that is not the rule.

Re:How to Think About GPUs by pklausler on Wednesday August 20, 2025 @04:22PM (#44966014)
Re:How to Think About GPUs by einpoklum on Thursday August 21, 2025 @09:29AM (#44972530)
The "video instructions" are indeed another exception: Operations on sub-lanes of 32-bit values: 2x16 or 4x8. This is relevant for graphics/video work, where you often have Red, Green, Blue, Alpha channels of 8 bits each. Their use is uncommon (AFAICT) in CUDA compute work.
Re:How to Think About GPUs by shaklee3 on Thursday August 21, 2025 @12:43AM (#44969094)
not true; there are a lot of simd instructions on GPUs
Re:How to Think About GPUs by einpoklum on Thursday August 21, 2025 @09:26AM (#44972492)
Such as?
Re:How to Think About GPUs by shaklee3 on Thursday August 21, 2025 @09:53AM (#44972829)
dp4a, ldg. just Google it. there's a whole page of them