News for hackers. Stuff that matters.
Re:Push ifs up and fors down by titzer on Wednesday November 15, 2023 @07:42PM (#38284493)
It's not just the direct cost of a branch, it's the downstream costs as well. Removing (or automatically folding branches in a compiler) can lead to optimizations after the branch.

E.g.

    if (foo > 0) {
      x = 3;
    } else {
      x = 7;
    }
    return x * 9;
If the compiler (or programmer) knows foo is greater than zero (even if we don't know what it actually is), then the whole thing folds into:

    return 27;
That also means that foo is not even used, so it might get dead-code eliminated.

If that gets inlined, then the optimizations just keep stacking up.

(not that the article was about that, it's just one implication of removing branches: downstream code can be optimized knowing more).

So, in summary, compilers matter.

Re:Push ifs up and fors down by RenThraysk on Thursday November 16, 2023 @11:58AM (#38292023)
That code shouldn't generate any branching (unless on an odd cpu). A serious compiler should recognise it can be done with a conditional move.
Re:Push ifs up and fors down by titzer on Thursday November 16, 2023 @07:18PM (#38297562)
A conditional move is not guaranteed to be faster than a predictable branch. It varies a lot by microarchitecture. It was just an example. Imagine if you had a store in the else branch. Now optimizing away a store means that load elimination may kick in. Folding branches in the compiler unlocks tons of downstream benefits.
Re:Push ifs up and fors down by wyager on Thursday November 16, 2023 @05:03PM (#38296216)
"If the branch condition is not volatile, compilers will usually lift it."
Re:Push ifs up and fors down by titzer on Thursday November 16, 2023 @07:18PM (#38297574)
Lift it where?