Re: POSIX and float rounding

[Rick Jones via tcpdump-workers <tcpdump-workers () lists tcpdump org>]

> --- Begin Message ---
>
>
> From: Rick Jones <jonesrick () google com>
>
> Date: Thu, 19 Mar 2026 12:58:46 -0700
>
> I was never really on the "CPU side" of SPEC for benchmarking in my past
> worklife, but I have a vague recollection that they had issues with
> comparing the results of benchmarks against the reference results which may
> have had allowance for a bit of "fuzziness."  It was called "specdiff" and
> seems to still exist:
> https://www.spec.org/cpu2017/Docs/utility.html#specdiff  although I do not
> know if they would be willing to release it outside of the licensed
> benchmark sources.   But I thought I'd mention it if another option were
> desired.
>
> On Thu, Mar 19, 2026 at 12:13 PM Guy Harris <gharris () sonic net> wrote:
>
> > In any case, given all this mishegoss, the right answer is probably to
> > have the script see how the floating-point value 0.5 is printed with %.0f,
> > and choose which test to perform based on that, in the hopes that whatever
> > rounding the scripting program does matches what tcpdump will do.
> > _______________________________________________
> > tcpdump-workers mailing list -- tcpdump-workers () lists tcpdump org
> > To unsubscribe send an email to tcpdump-workers-leave () lists tcpdump org
> > %(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s
>
> --- End Message ---
_______________________________________________
tcpdump-workers mailing list -- tcpdump-workers () lists tcpdump org
To unsubscribe send an email to tcpdump-workers-leave () lists tcpdump org
%(web_page_url)slistinfo%(cgiext)s/%(_internal_name)s