Too much discussion of the XOR swap trick

Re:Too much discussion of the XOR swap trick by jcalvinowens on Thursday April 16, 2026 @08:27AM (#47791997)

The "party trick" is more general than just xor, for example:

    void swap(unsigned* a, unsigned* b)
    {
        *a = *a + *b;
        *b = *a - *b;
        *a = *a - *b;
    }

GCC still sees through it with restrict: https://godbolt.org/z/8n3bGha3e

[ Parent | on HN ]

Re:Too much discussion of the XOR swap trick by zahlman on Thursday April 16, 2026 @12:17PM (#47795649)

This is covered in TFA.

[ Parent | on HN ]

Re:Too much discussion of the XOR swap trick by sifar on Thursday April 16, 2026 @10:30AM (#47793518)

This has overflow issues unlike xor.

[ Parent | on HN ]

Re:Too much discussion of the XOR swap trick by deathanatos on Thursday April 16, 2026 @11:15AM (#47794427)

Depending on what you think the "issue" is, one of:

* wrapping is well-defined behavior for unsigned integers; signed integer wrapping is UB, but is not used here.

* the equations (that a & b cancel each other out, resulting in the swap) hold even when done in a mod N ring.

[ Parent | on HN ]

Re:Too much discussion of the XOR swap trick by warmwaffles on Thursday April 16, 2026 @10:31AM (#47793535)

Yea this is operating on the pointer to the values, but if you use just an `unsigned a` it uses xor

[ Parent | on HN ]