This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug middle-end/28417] suboptimal 'division by constant' optimization
- From: "vda dot linux at googlemail dot com" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 2 Aug 2006 19:05:42 -0000
- Subject: [Bug middle-end/28417] suboptimal 'division by constant' optimization
- References: <bug-28417-12956@http.gcc.gnu.org/bugzilla/>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Comment #9 from vda dot linux at googlemail dot com 2006-08-02 19:05 -------
Thanks for the link to .pdf! Who's Tege? And his email address is ...?
Ok. This new version of the patch is working for signed divisions too, and is
giving the results which are always same or better than "old" code.
If it would be deemed acceptable for mainline, it can be simply be integrated
into choose_multiplier().
Since new code needs to know maximum possible dividend instead of (n,prec)
tuple, and it can generate better code if dividend is known to be small enough
(example: (n & 0xffff) / 10 will need only a mul, not mul+shift), it will make
sense to propagate possible dividend range into choose_multiplier() intead of
(n,prec). TODO later...
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=28417