The regular detailed config is modeled after a very aggressive OoO core, 8 way, with very large predictors and caches. One thing to double check is that you are using the proper cache options (line size, size, assoc) because these are overridden by the defaults if you just specify --cpu-type=arm_detailed. But, as far as I can tell the arm_detailed config has predictor sizes, pipeline depth and width that are closer to the A15. Perhaps the O3 model is closer to an A9 but I am not sure about this.
-Tony On Mon, Oct 29, 2012 at 10:44 AM, Amin Farmahini <amin...@gmail.com> wrote: > I am not an ARM guy, but by looking at the spec I would say that > ARM_O3_V7a.py (arm_detailed) is modeled after A9. > > Thanks, > Amin > > > On Mon, Oct 29, 2012 at 6:15 AM, Fernando Endo > <fernando.en...@gmail.com>wrote: > >> Hi, >> >> I also wonder what cpu "arm_detailed" correspond to. If it simulates a >> real cpu, it can only be A9 or A15, the two only OOO processors within >> ARMv7-A arch. >> >> Regards, >> >> -- >> Fernando A. Endo, PhD student and researcher >> >> CEA Lab >> and >> Université de Grenoble, UJF >> France >> >> 2012/10/29 Robert PINSKER <robert.pins...@st.com> >> >>> Hi**** >>> >>> ** ** >>> >>> I note the response Jack Harvard gave in February stating that use of >>> ARM_O3_V7a.py should give A15-like simulation.**** >>> >>> ** ** >>> >>> Am I right in believing that “—cpu-type=arm_detailed” will select this >>> V7a config?**** >>> >>> ** ** >>> >>> If so, I am somewhat puzzled because my measurements of Coremark using >>> arm_detailed are within 2% of the result I expect for A9.**** >>> >>> ** ** >>> >>> If I use “—cpu-type=detailed”, I get Coremark figures considerably >>> higher than I get even with A15, i.e. apparently not representative of any >>> ARM core.**** >>> >>> ** ** >>> >>> Can anyone explain this, or tell me where I am going wrong?**** >>> >>> ** ** >>> >>> Thanks**** >>> >>> ** ** >>> >>> Robert.** >>> >>> ** ** >>> >>> _______________________________________________ >>> gem5-users mailing list >>> gem5-users@gem5.org >>> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users >>> >> >> _______________________________________________ >> gem5-users mailing list >> gem5-users@gem5.org >> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users >> > > > _______________________________________________ > gem5-users mailing list > gem5-users@gem5.org > http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users >
_______________________________________________ gem5-users mailing list gem5-users@gem5.org http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
