This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
[Bug middle-end/23181] [4.1 Regression] Slowdown of the bresenham line drawing by roughly 20%
- From: "law at redhat dot com" <gcc-bugzilla at gcc dot gnu dot org>
- To: gcc-bugs at gcc dot gnu dot org
- Date: 10 Nov 2005 18:26:09 -0000
- Subject: [Bug middle-end/23181] [4.1 Regression] Slowdown of the bresenham line drawing by roughly 20%
- References: <bug-23181-176@http.gcc.gnu.org/bugzilla/>
- Reply-to: gcc-bugzilla at gcc dot gnu dot org
------- Comment #16 from law at redhat dot com 2005-11-10 18:26 -------
Subject: Re: [4.1 Regression] Slowdown of the
bresenham line drawing by roughly 20%
On Wed, 2005-11-02 at 14:32 +0000, hubicka at ucw dot cz wrote:
> Hmm, perhaps restricting the reassociation + simplification into case
> where it kills last use of the intermediate result?
That's the only thing I had come up with too (see comment #14). I don't
particularly like that idea, probably because it's a hack.
But I like to be practical, and reality is the simple reassociation done
by DOM doesn't make nearly as much sense when the parent statement
doesn't die as a result of the optimization.
Jeff
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23181