A new register allocator for ZJIT

(railsatscale.com)

55 points | by tenderlove 4 days ago ago

6 comments

  • ksec a day ago ago

    Glad to read news on Ruby ZJIT development. I wonder if ZJIT is being used in production at Shopify yet?

    • molf a day ago ago

      I have no idea; but I presume they don't, given that ZJIT today is still much slower than YJIT? [1]

      [1]: https://rubybench.github.io

    • kimos a day ago ago

      Used in production? Probably not yet. But likely tested in production on a small %. They’ve gotten practiced at using the Shopify Rails monolith as the biggest test suite and regression tests for Rails and Ruby.

      • ksec a day ago ago

        Yes that is what I meant, should have worded it as tested in production.

        Hopefully it is no longer experimental in Ruby 4.1.

  • nryoo a day ago ago

    How big is the lifetime holes thing in practice? On loops the contiguous-interval model spills way more than it should. Wondering if that alone explains most of the YJIT gap.

  • djwatson24 a day ago ago

    I’ve always wondered about linear scan vs SSA based spilling & regalloc, like libfirm & qbe use:

    ‘Register spilling and live-range splitting for SSA-form programs’ and ‘Preference-Guided Register Assignment’

    It’s much cleaner to code, since you don’t need to generate live ranges explicitly, and can break it up in two passes. Linear scan creates new inactive /active intervals, while the ssa form allocators just reuse the ssa graph. I’ve never benchmarked them back to back though.