[Bug c++/15837] Optimization (of templatized functions?)

[adruab at voria dot com]

------- Additional Comments From adruab at voria dot com  2004-06-06 19:25 -------
Heh, thanks, that makes sense.  I've never actually run into any issue like that
before, where problems are undiagnosable by the compiler :).

So it working correctly using a p = (type*)((char*)p + stride) method is purely
coincidental?  Does it violate the same aliasing rules?  I assume at this point
that you are talking about aliasing as in relation to memory boundaries (in
which case the other would still break the code).  Can you point me to spot in
the spec that discusses aliasing rules?

Thanks for the help,
I hope that this bug can help diagnose problems like this in the future.

Adruab

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15837