Can I review this change again before commit? Or can you separate the patch into inorder/o3 portions? If you give me through this weekend I can try to figure out why combining the code is causing stats differences. -Korey
On Tue, Jan 15, 2013 at 4:15 PM, Ali Saidi <[email protected]> wrote: > > On Jan 15, 2013, at 6:59 PM, Nilay Vaish <[email protected]> wrote: > > > On Tue, 15 Jan 2013, Ali Saidi wrote: > > > >> > >> > >>> On Jan. 8, 2013, 6:29 a.m., Ali Saidi wrote: > >>>> Hi Nilay, > >>>> > >>>> Thanks for resurrecting this. In principle I have no issues, I just > would like to make sure that the moved code includes all the latest fixes > to the branch predictor that numerous people have contributed to. If so and > you get no stats differences please commit. > >>>> > >>>> Thanks, > >>>> Ali > >>>> > >> > >> I'm still not sure why stats changes should occur. There may be some > differences, but since a lot more effort has been put in verifying the o3 > cpu branch predictor by people like Nathanael, it would be good to keep > that one around and make it work with the in-order cpu or add the methods > as appropriate. > >> > >> > > > > The stats would changes only for the inorder-cpu. For the o3 cpu, they > would remain the same. > Thanks Nilay, that sounds good. > Ali > > _______________________________________________ > gem5-dev mailing list > [email protected] > http://m5sim.org/mailman/listinfo/gem5-dev > -- - Korey _______________________________________________ gem5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/gem5-dev
