This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: branch predictions
- To: Jason Eckhardt <jle at cygnus dot com>
- Subject: Re: branch predictions
- From: Jeffrey A Law <law at cygnus dot com>
- Date: Thu, 04 May 2000 09:24:12 -0600
- cc: Mark Mitchell <mark at codesourcery dot com>, gcc at gcc dot gnu dot org
- Reply-To: law at cygnus dot com
In message <Pine.SOL.3.91.1000504075559.14034A-100000@cse.cygnus.com>you writ
e:
[ ... ]
> did get reasonable results empirically over their benchmarks, but
> many times real profiling feedback is better. This is why bb-reorder can
> work transparently with either. If profiling feedback is available, it
> will be used. Otherwise, we fallback to static predictions.
Right. I just want to reiterate Jason's point.
The code was designed from the start to work with *either* static or dynamic
branch predictions. It will prefer dynamic predictions if they are available.
jeff