News for hackers. Stuff that matters.
Re:Fast Sorting, Branchless by Design by jedisct1 on Friday February 20, 2026 @11:47AM (#47090393)
a_int and b_int are signed values.
Re:Fast Sorting, Branchless by Design by rep_lodsb on Friday February 20, 2026 @12:02PM (#47090615)
It makes no difference whether they're signed or unsigned. Unless the subtraction is checking for overflow, or using a wider integer type than the numbers being compared, the high bit will not in every case indicate which number is smaller.

e.g.

    0x8000_0000 < 0x0000_0001 for signed numbers
    0x8000_0000 - 0x0000_0001 = 0x7fff_ffff, high bit clear
Re:Fast Sorting, Branchless by Design by jedisct1 on Friday February 20, 2026 @12:20PM (#47090864)
They are using a wider type.
Re:Fast Sorting, Branchless by Design by rep_lodsb on Friday February 20, 2026 @12:28PM (#47090981)
Yes, looking at the source code on GitHub now cleared that up!

Didn't see it mentioned in the article though, maybe I missed it. If not, I think that this detail would be a good thing to include, both since it's a common mistake that others with less experience might make, and to get ahead of nitpicky comments like mine :)