------- Additional Comments From law at redhat dot com 2004-11-19 03:14 -------
Subject: Re: [4.0 Regression] jump threading
on trees is slow with switch statements with large # of cases
On Thu, 2004-11-18 at 18:23 +0000, pinskia at gcc dot gnu dot org wrote:
> So we are still twice as slow as 3.3 but faster than before.
> Using -ftime-report I get:
> dominator optimization: 11.69 (42%) usr 0.13 ( 4%) sys 12.28 (38%) wall
> rename registers : 3.32 (12%) usr 0.12 ( 4%) sys 3.62 (11%) wall
>
> So only DOM needs to be sped up.
>
> For DOM the problem comes from thread_across_edge when we go and update the
> profile of the BB.
That's part of the problem, but there's definitely more stuff going
on than just that annoying useless divide operation in the code to
update the profile. There's some issues in tree-ssa-threadupdate as
well.
> For renane registers, most of a profile of compiling this source comes from
> regrename.c the loop which
> is around line 1761.
Yes. I know.
Jeff
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=15524
