Hi Andreas,

  Thank you for your reply.

  I used MI protocol by using :
   scons build/ALPHA/gem5.opt PROTOCOL=MI_example RUBY=True
  I am using ruby_fs system and command used is:
 ./build/ALPHA/gem5.opt configs/example/ruby_fs.py -n 2 --l1i_size=32kB
--l1d_size=32kB --l2_size=8MB --num-l2caches=8 -b barnes
   Changed protocol from MI_example to MESI_Cmp_Directory and again
executed ALPHA ruby_fs using the same command.

  Only parameter I have changed is coherency protocol and I git the
difference in number of instructions simulated.

   Where I find more information about usage of LOCK/UNLOCK operations used
by cache coherency protocol in gem5.

Thank You,
Geeta




On Tue, Jun 10, 2014 at 1:09 PM, Andreas Hansson <[email protected]>
wrote:

>  Hi Geeta,
>
>  This has been discussed quite a few times, so I’d suggest you do a
> search on the archive.
>
>  In short, there are a number of reasons this happens, and without
> knowing more about your use-case (full system or not, etc), I would guess
> the reason is simply differences in acquiring/releasing locks.
>
>  Andreas
>
>   From: Geeta Patil via gem5-users <[email protected]>
> Reply-To: Geeta Patil <[email protected]>, gem5 users
> mailing list <[email protected]>
> Date: Tuesday, 10 June 2014 06:24
> To: "[email protected]" <[email protected]>
> Subject: [gem5-users] Cache Coherency Statistic Output
>
>   Hi All,
>
>  I have executed barens(splash2 benchmark program) using MI and MESI
> protocol.
>
>  I am finding number of instructions simulated. When I checked the
> statistic file, number of instructions simulated using MI protocol are
> 12184940560 and number of instructions simulated using MI protocol are
> 12174567882.
>
>  I am not able to understand when I am executing same program why is
> there difference of 10372678 between number of instructions simulated.
>
>  I want to understand why there is difference.
>
>  Can any body help me.
>
>  Thank You,
>  Geeta
>
>
>
> -- IMPORTANT NOTICE: The contents of this email and any attachments are
> confidential and may also be privileged. If you are not the intended
> recipient, please notify the sender immediately and do not disclose the
> contents to any other person, use it for any purpose, or store or copy the
> information in any medium. Thank you.
>
> ARM Limited, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ,
> Registered in England & Wales, Company No: 2557590
> ARM Holdings plc, Registered office 110 Fulbourn Road, Cambridge CB1 9NJ,
> Registered in England & Wales, Company No: 2548782
>
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Reply via email to