This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: 3.0 high priority: optimization/3083: C++ frontend consumesinacceptable amounts of CPU with -O3
- To: Gerald Pfeifer <pfeifer at dbai dot tuwien dot ac dot at>
- Subject: Re: 3.0 high priority: optimization/3083: C++ frontend consumesinacceptable amounts of CPU with -O3
- From: Mark Mitchell <mark at codesourcery dot com>
- Date: Wed, 13 Jun 2001 17:06:32 -0700
- cc: "nathan at codesourcery dot com" <nathan at codesourcery dot com>, "pcarlini at unitus dot it" <pcarlini at unitus dot it>, "gcc at gcc dot gnu dot org" <gcc at gcc dot gnu dot org>
--On Thursday, June 14, 2001 01:44:52 AM +0200 Gerald Pfeifer
<pfeifer@dbai.tuwien.ac.at> wrote:
> On Sat, 9 Jun 2001, Mark Mitchell wrote:
>>> http://gcc.gnu.org/ml/gcc/2001-05/msg01651.html
>> Yes -- although that's something a little different. It's not just
>> the order that's wrong with -O3 -- it's that we're not selective about
>> what it makes sense to inline and what it doesn't.
>>
>> But, yes, we're not touching this stuff for 3.0. No way, no how.
>
> Just to make sure: My PR is about a compile-time regression *on the
> GCC 3.0 branch*, that happend during the month of May.
>
> We now take more than twice as long than one month ago to compile the
> example in the PR.
Yup. There are two separate issues: what heuristic we should use,
and what has gone wrong with your code. I do not want to change the
heuristic; I would love to fix the problem in your code. I hope
to look at it tonight, but I am honestly not optimistic at this point.
Yours,
--
Mark Mitchell mark@codesourcery.com
CodeSourcery, LLC http://www.codesourcery.com