Add extra optimizations: -Ox

Joseph S. Myers jsm28@cam.ac.uk
Fri Mar 9 15:35:00 GMT 2001


On Fri, 9 Mar 2001, Cosmin Truta wrote:

> > This patch fails to include changes to invoke.texi to document the new
> > flag.
> 
> That's because I wasn't sure whether you guys will incorporate it or not;
> it was rather a suggestion than a patch.

It certainly looked like a patch!  If submitting something intended as
just a suggestion, I think it could be appropriate to submit *only* the
documentation patch (which can be written before the actual code changes)
as a description for comments of what the proposed change would do.  
Documentation without code changes makes sense in this context (and in
some unusual cases can be a complete installable change in its own right -
e.g. when I added documentation of the formal syntax of attributes that
partly describes how they ought to be fixed to work as well as how they
currently work, without having made the changes to fix them to work
properly).  Code changes (to user-visible features) without documentation
make much less sense, since you should have a clear conception of what the
feature does and when it should be used when implementing it, and should
document it to transfer that understanding.  (I prefer the method of
writing documentation and testcases before the code changes; each person
may work in their own style, but documentation is still of fundamental
importance as part of a complete change or proposed change.)

-- 
Joseph S. Myers
jsm28@cam.ac.uk



More information about the Gcc-patches mailing list