Hello,

If the core configuration is multi-issue, that can happen. You can check
the issueWidth in the config.ini file, for example.

Regards,

--
Fernando A. Endo, PhD student and researcher

Université de Grenoble, UJF
France



2014-05-26 16:08 GMT+02:00 Anderson Sartor via gem5-users <
[email protected]>:

> Hi all,
> I'm running a single core pipelined version of MIPS and I'm having trouble
> to understand some of the simulation restuls.
>
> Running a linked list algorithm I have the following results:
> sim_ops                                     944501495
>   # Number of ops (including micro ops) simulated
> system.cpu.committedInsts_total             944501495
>   # Number of Instructions committed (Total)
> system.cpu.runCycles                       1213082009
>   # Number of cycles cpu stages are processed.
>
> Now, running a sequential matrix multiplication algorithm:
> sim_insts                                   118167524
>   # Number of instructions simulated
> system.cpu.committedInsts_total             118167524
>   # Number of Instructions committed (Total)
> system.cpu.runCycles                        117915425
>   # Number of cycles cpu stages are processed.
>
> and the command I'm using to simulate:
> ./build/MIPS/gem5.opt configs/example/se.py -n1 --cpu-type=inorder
> --caches -c tests/$1
>
> How can the number of runCycles be less than the number of executed
> instructions in a simple pipeline (matrix multiplication example)? (i.e.
> more than 1 instruction is executed in the same cycle)
>
> _______________________________________________
> 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

Reply via email to