Date: Sun, 3 Feb 2019 09:42:59 -0800 From: Steve Kargl <sgk@troutmask.apl.washington.edu> To: freebsd-numerics@freebsd.org Subject: Re: Implement cexpl Message-ID: <20190203174259.GA42562@troutmask.apl.washington.edu> In-Reply-To: <20190203173056.GA42499@troutmask.apl.washington.edu> References: <20190203173056.GA42499@troutmask.apl.washington.edu>
next in thread | previous in thread | raw e-mail | index | archive | help
On Sun, Feb 03, 2019 at 09:30:56AM -0800, Steve Kargl wrote:
> The following patch implements cexpl() for ld80 and ld128 architectures.
>
> The kernel files for large x are named ld{80,128}/k_cexpl.c to avoid
> confusion with the ld{80,128}/k_expl.h files.
>
> I do not have access to ld128, so the ld128 does not use bit twiddling.
> I cannot test the ld128 implementation, but I believe that it works
> (for some definition of work).
>
I forgot to mention that the initial files under ld{80,128} where
created via
svn cp src/k_exp.c ld80/k_cexpl.c
svn cp src/s_cexp.c ld80/s_cexpl.c
svn cp src/k_exp.c ld128/k_cexpl.c
svn cp src/s_cexp.c ld128/s_cexpl.c
so the diff is not against an empty file; but against the
copied files.
--
Steve
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20190203174259.GA42562>