korrents

korrents · piece

The ITTAGE indirect branch predictor

Nelson Elhage · 4 Jul 2025 · blog.nelhage.com ↗

3 korrents from this piece

Nelson Elhage did not write this page.

Every claim below was made in this piece, quoted word for word and numbered in the order the piece makes them, so you can read it there rather than take our word for it. The sentence above each quote is our reading of the claim, not their wording. Each quote was checked against a stored copy of the page at build time; where the two differ, the quote is the fact.

  1. Our reading

    Modern CPUs can now predict the indirect dispatch jump in a bytecode interpreter loop with high accuracy.

    Their wordsModern CPUs mostly no longer struggle to predict the bytecode-dispatch indirect jump inside a "conventional" bytecode interpreter loop.

  2. Our reading

    Most conditional branches in software are strongly biased toward one outcome rather than close to fifty-fifty.

    Their wordsmost branches are heavily "biased" one way or another - e.g. a "branch taken" rate of 10% or 90% is more common than 50%

  3. Our reading

    Using a neural network is likely to become easier than designing a custom-tuned prediction algorithm for many tasks.

    Their wordsin 2025 it might end up being easier to just throw a neural net at the problem, instead of a custom-designed-and-tuned prediction algorithm!