Hi, In O3CPU, there is a pipeline, with register renaming and some other things. so each instruction may take some cycles to complete. However in atomic model, all operations are done in one tick for an instruction.
For more information, read http://www.gem5.org/dist/tutorials/isca_pres_2011.pdf (section "CPU Modeling ") -- // Naderan *Mahmood; On Tue, Dec 20, 2011 at 10:52 AM, Jungwook Kim <[email protected]> wrote: > Hi, all > I'm wondering that some result values are different between o3 (detailed) > mode and normal (atomic) mode in full system simulation. > for example, in stats.txt file, > the value of "system.cpu0.icache.overall_miss_rate" is 0.012472 in atomic > cpu mode but in detailed mode, the value is 0.071092. > also, the value of "system.cpu0.numCycles" is 3646715099 in atomic mode > but the value is 2133833981 in detailed mode. > which value is correct ? > my simulation parameters are as follows. > build/ARM_FS/gem5.fast configs/example/fs.py -d -n 2 --caches --l2cache -b > bench_name ( detailed mode ) > build/ARM_FS/gem5.fast configs/example/fs.py -n 2 --caches --l2cache -b > bench_name ( normal mode ) > > > _______________________________________________ > 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
