Ish, today I am on the fence because sometimes(half-half) its necessary, a year ago it was completely impossible, but I think its a question of context length, attention and task. Modern models can handle raw PTX pretty well. If you have below 20k lines, it's able to make its own abstractions in the thinking loop Nth pass when it needs to. The only reason we would have to make our own abstractions today would be to fill the limits of the current context context window attention trade-offs the companies make.
I think the article is calling for the redundancy of high-level tiling abstractions. There is simply no need to work at a higher level and give up performance today when code is free.
alt Hacker News
Ish, today I am on the fence because sometimes(half-half) its necessary, a year ago it was completely impossible, but I think its a question of context length, attention and task. Modern models can handle raw PTX pretty well. If you have below 20k lines, it's able to make its own abstractions in the thinking loop Nth pass when it needs to. The only reason we would have to make our own abstractions today would be to fill the limits of the current context context window attention trade-offs the companies make.
I think the article is calling for the redundancy of high-level tiling abstractions. There is simply no need to work at a higher level and give up performance today when code is free.