Re: [found] gcc4 vs gcc3: a case (kerTeX)

[tlaronde%polynum.com@localhost]

On Wed, Mar 09, 2011 at 09:35:35PM +0100, tlaronde%polynum.com@localhost wrote:
> On Wed, Mar 09, 2011 at 12:08:51PM -0800, Chuck Swiger wrote:
> > On Mar 9, 2011, at 11:50 AM, tlaronde%polynum.com@localhost wrote:
> > >> Does "-O2 -fno-strict-aliasing" work better?
> > >> 
> > >> C99's type-punning assumptions (implied by -O2 in gcc-4.x including 
> > >> -fstrict-aliasing) aren't safe with older code which worked fine with 
> > >> prior versions of the compiler.
> > > 
> > > Thanks for the tip, but in this case no: it doesn't work.
> > 
> > OK.  Too bad, but it was the most likely candidate and worth a try.
> > 
> > If you wanted to triage the issue more carefully, you'd need to enable the 
> > various suboptions which are enabled by -O2 which are not in -O1 and figure 
> > out which one(s) are responsible.
> > 

The offending is : optimize-sibling-calls. 
-O2 -fno-optimize-sibling-calls passes)

Since D.E.K. uses artfully coroutines, my feeling is that this is not a
bug in the source, but a dangerous optimization.

Regards,
-- 
        Thierry Laronde <tlaronde +AT+ polynum +dot+ com>
                      http://www.kergis.com/
Key fingerprint = 0FF7 E906 FBAF FE95 FD89  250D 52B1 AE95 6006 F40C