Danek Duvall wrote:
> Roland Mainz wrote:
> > Danek Duvall wrote:
> > > Sriram Natarajan wrote:
> > > > Now, assuming this can be done, does any one have any objections with
> > > > the concept of delivering multiple version (regular and cpu optimized)
> > > > of some critical libraries that are under sfw consolidation ?
> > >
> > > I certainly don't, though generally greater optimization means a longer
> > > build time, which is eventually going to get painful.
> >
> > Why ?
> 
> I don't understand your question -- why is the build time going to get
> longer, or why is a longer build time more painful?  Aren't the answers
> obvious?

The question is "why are longer build times be painfull" ? For quick
development they are painfull but for production binaries performance is
IMO more important.
I remeber still the case of Mozilla/FireFox which ran with "-xO2" which
made the Sun Studio builds _much_ slower than gcc builds (e.g. the
builds were more or less unuseeable on an Ultra5 when lots of
JavaScript/DOM operations were used ; after the change to -xO4 and other
minor tweaks the builds were even useable (but slow) on a
SPARCstation5/110MHz).
After some time we switched the build over to -xO4 and that solved lots
of performance problems (and as a side-effect forced the compiler team
to start fixing all the bugs in the C++ compiler which popped-up when
higher optimisation levels were used (it took more than half a year to
hunt down all the bugs but IMO it was worth the trouble))

----

Bye,
Roland

-- 
  __ .  . __
 (o.\ \/ /.o) roland.mainz at nrubsig.org
  \__\/\/__/  MPEG specialist, C&&JAVA&&Sun&&Unix programmer
  /O /==\ O\  TEL +49 641 3992797
 (;O/ \/ \O;)

Reply via email to