------- Comment #19 from steven at gcc dot gnu dot org 2010-01-30 22:20 -------
FWIW the tri-state solution is problematic because there are cases where type
punning doesn't lead to this kind of optimizations. For example if the casted
pointer is casted back to the correct type, or if there is never a dereference.
To disable strict aliasing in such cases leads to missed optimizations. That
would make you happy, but it would also make another crowd unhappy again.
Whatever GCC does, it's going to make people unhappy somewhere. This seems to
be the fate of compilers everywhere...
This is a closed discussion. *Please* stop re-opening this bug. You will *not*
get what you want. It's nothing personal, it's just that this has been
discussed many, many times before and a decision has been made. (And I just
read the Troll article on your blog... ;-)
--
steven at gcc dot gnu dot org changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|UNCONFIRMED |RESOLVED
Resolution| |INVALID
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=42907
