Yup, Lisa is correct. O3's for an of out of order design. If you want to approximate an in-order using O3 you can restrict some of the out of order queues (IQ,LSQ,etc.)...
However, the In-Order model in the m5-dev tree is ready to be tested and used. I have one set of patches that I haven't pushed yet that optimize memory usage by the model (but that should be coming soon). Otherwise, the InOrder model has currently been tested for the ALPHA ISA and a few of the spec2k benchmarks in SE mode (you can run the regression tests when you download the latest copy). As always, more tests and more "testers" would be helpful before I guess we would call it "production ready" but feel free to try as-is. Also, documentation on m5sim.org would also be nice :) On Tue, Jul 28, 2009 at 1:11 PM, Lisa Hsu <[email protected]> wrote: > Either InOrder or TimingCPU. > O3 stands for out-of-order, so you definitely don't want that if you want > an in-order core. > > TimingCPU is a 1CPI machine that can be attached to a timing memory model. > > InOrder is new, and Korey can speak more to it, but I believe it is a > mildly pipelined in-order CPU, more realistic than TimingCPU. However, I am > not positive, nor do I know if it is really production ready - Korey? > > Lisa > > On Tue, Jul 28, 2009 at 9:21 AM, soumyaroop roy <[email protected]> wrote: > >> Hi there, >> >> To be able to perform a detailed timing on a CPU with an inorder execution >> core, should I be using the DerivO3CPU model or the InOrderCPU model? >> >> regards, >> Soumyaroop. >> >> -- >> Soumyaroop Roy >> Ph.D. Candidate >> Department of Computer Science and Engineering >> University of South Florida, Tampa >> http://www.csee.usf.edu/~sroy <http://www.csee.usf.edu/%7Esroy> >> _______________________________________________ >> m5-users mailing list >> [email protected] >> http://m5sim.org/cgi-bin/mailman/listinfo/m5-users >> > > > _______________________________________________ > m5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/m5-users > -- - Korey
_______________________________________________ m5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/m5-users
