6 ms·
The article does make an example quite early; > GPUs achieve very high performance without any of this logic, at the expense of requiring explicitly parallel p
by aDyslecticCrow 10d ago
The article does make an example quite early;
> GPUs achieve very high performance without any of this logic, at the expense of requiring explicitly parallel programs.
GPU cores are in some ways closer to "PDP-11", they're either acting as thousands of parallel simple processors, or expose pretty raw instructions for very parallel use-cases.
- legobmw99 10d agoThat seems fair, CUDA kernels and shader code do feel like they're at a similar level of abstraction over the hardware as C was to the PDP-11. But I do think there isn't really an equivalent for modern CPU ISAs
- aDyslecticCrow 10d agoMabie hand-rolling LLVM IR representations would count.
- jpollock 10d agoDoesn't assembly allow devs to ignore speculative execution as well? You can place LFENCE(x86)/CSDB(arm) around code blocks, but you can do that in C too.