allow ffs & co. a binary search

Erich Dollansky erichsfreebsdlist at alogt.com
Fri Jun 12 00:11:03 UTC 2015


Hi,

On Thu, 11 Jun 2015 08:51:50 -0600
Ian Lepore <ian at freebsd.org> wrote:

> On Thu, 2015-06-11 at 22:43 +0800, Erich Dollansky wrote:
> > Hi,
> > 
> > On Thu, 11 Jun 2015 06:57:51 -0600
> > Ian Lepore <ian at freebsd.org> wrote:
> > 
> > > On Thu, 2015-06-11 at 15:21 +0800, Erich Dollansky wrote:
> > > > On Sun, 07 Jun 2015 08:53:18 +0200
> > > > Hans Petter Selasky <hps at selasky.org> wrote:
> > > > 
> > > Is there any reason that this whole change, userland and kernel,
> > > isn't just:
> > > 
> > > #define ffs(n) __builtin_ffs(n)
> > 
> > I did not see this. Where is it?
> > 
> > Of course, if this is already done, it does not have to be done
> > again.
> > 
> > Erich
> 
> I'm not quite sure what you're asking.  Both clang and gcc support
> __builtin_ffs(), so I'm saying why don't we just use it and call the
> whole problem solved?
> 
so, it is not in the sources yet. Yes, then use it and let the compiler
guys crack their heads.

Erich


More information about the freebsd-hackers mailing list