inc_loop_asm.c

= c/inc_loop_asm.c
{file}
{tag=CPU microbenchmark}

This is the only way that we've managed to reliably get a single `inc` instruction loop, by using <inline assembly>, e.g. on we do <x86>:
``
loop:
  inc %[i];
  cmp %[max], %[i];
  jb loop;
``

For 1s on <Ciro Santilli's hardware/P14s> <Ubuntu 25.04> GCC 14.2 -O0 x86_64 we need about 5 billion:
``
time ./inc_loop_asm.out 5000000000
``

 Back to article page