On Thu, Aug 12, 2021 at 9:33 AM David Rowley <
dgrowleyml@gmail.com> wrote:
> Something there that might cause confusion is we do a configure check
> to see if popcntq works and define HAVE_X86_64_POPCNTQ if it does.
> I'm still a bit confused at why we bother doing that. Surely it just
> means that if the build machine does not have popcntq that we'll
> always use pg_popcount64_slow, regardless if the machine that's
> actually running the code has popcntq.
Yeah, it's a bit strange, a configure check makes more sense if we have a way to specify we can build with a direct call (like x86-64-v2), but we don't right now. Maybe short-term we could always do the runtime check on x86-64.
