On Wed, 2024-06-05 at 17:23 -0700, Jeff Davis wrote:
> A brief test shows that there may be a performance regression for
> libc
> default collations. But if so, I'm not sure that's avoidable if the
> goal is to take away setlocale. I'll see if removing the extra
> branches
> mitigates it.
I redid the test and didn't see a difference, and then I ran a
standalone microbenchmark to compare strcoll() vs strcoll_l(), and
didn't see a difference there, either.
Another implementation may show a difference, but it doesn't seem to be
a problem for glibc.
I think this patch series is a nice cleanup, as well, making libc more
like the other providers and not dependent on global state.
Regards,
Jeff Davis