This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: The new scheduler and x86 CPUs
- To: <dewar at gnat dot com>
- Subject: Re: The new scheduler and x86 CPUs
- From: Bernd Schmidt <bernds at redhat dot com>
- Date: Wed, 29 Aug 2001 10:49:07 +0100 (BST)
- cc: <vmakarov at toke dot toronto dot redhat dot com>, <gcc at gcc dot gnu dot org>, <jh at suse dot cz>
On Tue, 28 Aug 2001 dewar@gnat.com wrote:
> <<Why is OOO a dead-end approach? If there's parallelism to extract
> in a program, then if a compiler can find it, an OOO core can find it
> just as well.
> >>
>
> That's clearly false, since the OOO core has to do things on the fly in
> very limited time, while a compiler can do extensive analysis without
> such time constraints.
There are several cases where less extensive analysis is needed at run-time
(e.g. to determine memory aliases), or where certain aspects of the program
are only available at run-time, since they are variable (e.g. load
latencies).
> Now of course the question of whether EPIC architectures
> + clever compilers can in *fact* do better than OOO cores remains to be seen.
Don't forget that if your compiler can do something very clever, it can do
so for both EPIC and OOO>
Bernd