Hello,

I think it depends on what memory model one is simulating. If you're using
the classic model, than number 2 may define the cache latency.

You can change them, run and see if the config.ini will reflect the changes.

Regards,

--
Fernando A. Endo, PhD student and researcher

Université de Grenoble, UJF
France



2013/6/19 Mann <[email protected]>

> Hi All
>
> Cache latencies are configurable at 2 places
>
> 1)  src/mem/protocol/**L1Cache.sm
>   machine(L1Cache, "MSI Directory L1 Cache CMP")
>   : Sequencer * sequencer,
>    int l1_request_latency = 2,
>    int l1_response_latency = 2,
> &
> 2) configs/common/Caches.py
>    class L1Cache(BaseCache):
>     assoc = 2
>     block_size = 64
>     latency = '1ns'
>     mshrs = 10
>     tgts_per_mshr = 5
>     is_top_level = True
>
> I am not defining anything in running script.
> Which one is finally used during simulation ?
>
> --
> Cheers...........
> Mann
>
> _______________________________________________
> 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