[CHY872]

I think in general, most CPU architectures pick branch not taken for forward branches and branch taken for backwards branches on the first try. So I feel like builtin_expect gives weighting to let the compiler shuffle code around to make it fit that pattern.

There are ways of doing it in hardware, I remember a supervisor discussing it with respect to MIPS. I also remember them saying they went through the entire code generation stage of GCC and found that every single point at which GCC would try to use it was somewhere where it would be actively unhelpful.

[astrange]

__builtin_expect is good for error handling code, because gcc can avoid size-increasing optimizations in that path, and it can even move all the unlikely code out into another section so it won't take up space in your caches.

But its code generation is better for a 99%/1% case than a 60%/40% case, because Intel doesn't listen to branch hints anymore nor really give advice on how to tune for them.