[PATCH] Add branch prediction favouring normal numbers

[Mike Frysinger]

On Thursday 27 December 2012 22:52:21 Siddhesh Poyarekar wrote:
> On Thu, Dec 27, 2012 at 04:38:56PM -0500, Mike Frysinger wrote:
> > > +  if (__glibc_likely (EX> -42 || (EX==-42 && X[1]>=TWO10)))
> > > +    norm(x,y,p);
> > > +  else
> > > +    denorm(x,y,p);
> > 
> > should we fix the style while we're modifying the file ?  there's quite a
> > number of spaces that are "missing" ...
> 
> I did that because the coding style in these files is at best
> inconsistent and at worst, unreadable.  I'm not making the coding
> style any more inconsistent with that change.

sure, i'm not saying you're making it worse.  in the case of files that are 
messed up (like much of the math files), i'm in the habit of fixing the lines 
i'm already modifying since the line is already being changed.

> Of course, if I'm mistaken and there is in fact a coding style to that
> file then I'd really like to know.

reformatting them is probably a good idea.  it's not like we have source 
material to worry about keeping in sync with.
-mike
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: This is a digitally signed message part.
URL: <http://sourceware.org/pipermail/libc-alpha/attachments/20121228/d9782fe6/attachment.sig>