Hello, >I need to work with x86, the key idea is to have complex cores without any modification and basic cores, the basic cores will not have floating point unit, branch predictors and a smaller private cache than complex cores.
I'm not aware of x86 in gem5, but generally with the O3 model it seem to be not hard to do. I'm not sure if removing brach predictors can be easy, but you can mimic it with very small buffers. Regards, -- Fernando A. Endo, PhD student and researcher Université de Grenoble, UJF France 2015-04-30 22:17 GMT+02:00 Rodrigo Reynolds Ramírez < [email protected]>: > Hello everyone, > > I continue thinking about heterogeneous architectures, in the gem5 web > page I read something interesting: > > "Multiprocessor / multi-system capability. Thanks to gem5's object > orientation, instantiation of multiple CPU objects within a system is > trivial. Combined with the snooping coherence protocol supported by the > caches, gem5 can model symmetric and asymmetric multiprocessor systems." > > But the only real reference to AMP is a workshop from Antony Gutierrez, > http://www.m5sim.org/wiki/images/d/da/Gutierrez_gem5_workshop_2012.pdf. > But he showed how to define ARM processors. > > I would like to know how to do something similar with x86, I want to > simulate something like the Intel quickia. Do you think, is it possible? > > Thanks > Rodrigo > > ------------------------------ > From: [email protected] > To: [email protected] > Date: Wed, 29 Apr 2015 20:10:31 +0000 > Subject: [gem5-users] gem5 and AMP > > > Hello everyone, > > I need to work with asymmetric multiprocessors, reading different papers > and questions in the forums I conclude that people who have worked with AMP > and gem5 have done it using ARM. > > I need to work with x86, the key idea is to have complex cores without any > modification and basic cores, the basic cores will not have floating point > unit, branch predictors and a smaller private cache than complex cores. > > I would like to know if is it possible to do this in gem5? > > I really appreciate any suggestion and reply. > > Rodrigo > > _______________________________________________ gem5-users mailing list > [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users > > _______________________________________________ > gem5-users mailing list > [email protected] > http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users >
_______________________________________________ gem5-users mailing list [email protected] http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
