On Mon, Feb 9, 2026 at 12:25 PM Matthias Kretz <[email protected]>
wrote:

> Matthias Kretz [Monday, 9 February 2026 11:57:20 CET]:
> > Tomasz Kaminski [Monday, 9 February 2026 11:47:59 CET]:
> > > On Mon, Feb 9, 2026 at 11:38 AM Matthias Kretz <[email protected]>
> > >
> > > wrote:
> > > > clang-tidy "complained" about several "magic numbers" in the code. I
> > > > have
> > > > a
> > > > patch to resolve some of these.
> > >
> > > Where exactly does it complain? Can you use __int_traits::width instead
> > > (see <bit> header).
> >
> > constant expressions and constraints, so yes, a constant expression would
> > work. However, it's '__is_integer_nonstrict<long long>::__width' or
> > '__int_traits<unsigned long long>::__digits', no?
>
> Since I already need <limits> (because I need digits, max, and lowest for
> integers and floats) I guess I better use that instead?

Yes, makes sense.


> Or should I use (and
> move/copy) experimental/bits/numeric_traits.h? (P1841 come back please …)
>
> --
> ──────────────────────────────────────────────────────────────────────────
>  Dr. Matthias Kretz                           https://mattkretz.github.io
>  GSI Helmholtz Center for Heavy Ion Research               https://gsi.de
>  std::simd
> ──────────────────────────────────────────────────────────────────────────
>

Reply via email to