Join the discussion

Write your take first — we'll ask for email only when you're ready to publish.

  • Hacker News
  • (2012)
    by froh
  • All well and good provided your opcodes are sequential/dense
    by nly
  • > Therefore, the standard forces the compiler to generate "safe" code for the switch. Safety, as usual, has cost, so the switch version ends up doing a bit more per loop iteration.

    Safety only has a cost in this case because the switch is fundamentally just operating on an integer. With an actual enumerated type (rather than C's primitive "enums as numeric aliases"), which even a basic type system could trivially enforce, there would be no need for this check, because the domain of the value would be guaranteed at compile-time.

  • I understood up to the comparison operations in this article, but I'm having trouble understanding branch prediction even after reading it. It seems like they used branch prediction optimization... Sometimes when I read articles like this, I start to question whether I can really call myself a programmer
  • You can probably eliminate the switch bounds check by making the default case execute __builtin_unreachable(). The switch version is safe for non-conforming opcodes where the computed goto version invokes UB.

Explore Birbla archives