Nice discussion! thanks!

Regards,

--
Fernando A. Endo, PhD student and researcher

Université de Grenoble, UJF
France



2014-04-11 18:28 GMT+02:00 Ali Saidi <[email protected]>:

>  Yes, if you have a spin lock in some code, it's highly likely that a
> faster core will spin for more instructions on that lock and therefore the
> number if instructions will increase somewhat proportional to frequency.
>
>
>
> Ali
>
>
>
> On 11.04.2014 10:29, Heba Khdr wrote:
>
>  The number of committed instructions was increasing when I increase the
> frequency from 0.2 Ghz, 0.4, 0.6 , and so on till 1Ghz. However, after that
> it will not increase, but as I mentioned in the first email.
> #instructions in 1.2 Ghz will be the same liken #instructions in 0.6 Ghz
> and so on.
>
> Best,
> Heba
>
>
> On Fri, Apr 11, 2014 at 5:15 PM, narayanamoor <[email protected]>wrote:
>
>>  Also, in multi core simulations the number of committed instructions do
>> not represent work fully as some instructions may be used for
>> synchronization.
>>
>>
>> Thanks
>> Srini
>>
>>
>>
>>
>>
>>
>> -------- Original message --------
>> From: Andreas Hansson <[email protected]>
>> Date: 04/11/2014 9:11 AM (GMT-06:00)
>> To: gem5 users mailing list <[email protected]>
>> Subject: Re: [gem5-users] number of committed instructions
>>
>>
>> Hi Fernando,
>>
>> Something to keep in mind is that if it is a full system run then the
>> number of committed instructions could very well change.
>>
>> Andreas
>>
>> From: Fernando Endo <[email protected]>
>> Reply-To: gem5 users mailing list <[email protected]>
>> Date: Friday, 11 April 2014 15:08
>> To: gem5 users mailing list <[email protected]>
>> Subject: Re: [gem5-users] number of committed instructions
>>
>>  Hello,
>>
>> The number of committed instructions should be always the same! Because
>> the real work done can be represented by committed instruction, and it is
>> the same if one change the CPU frequency. What may change is the number of
>> executed instructions (iew stage), because the dynamic behavior, i.e. the
>> speculation, will be different with different CPU/cache freq ratio.
>>
>> Regards,
>>
>> --
>> Fernando A. Endo, PhD student and researcher
>>
>> Universit� de Grenoble, UJF
>> France
>>
>>
>>
>> 2014-04-11 10:42 GMT+02:00 Heba Khdr <
>> [email protected]>:
>>
>>>     Dear all,
>>>
>>> I have a question:
>>> I am running parsec applications using gem5 and for different
>>> freqeuencies in each experiment.
>>> When I increase the frequency, the number of committed instructions
>>> (which I got it from stats.txt) increase, till I exceed the frequency more
>>> than 1 Ghz, then the number of instructions will not increase.
>>> It will not saturated, but I will get the following:
>>> #instructions for 1.2 Ghz = #instructions for 0.6 Ghz.
>>> #instructions for 1.6 Ghz = #instructions for 0.8 Ghz.
>>>
>>> So, what do you think the problem is ?
>>>
>>>
>>> Thanks in advance.
>>>
>>>
>>> _______________________________________________
>>> gem5-users mailing list
>>> [email protected]
>>> http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users
>>
>>
>> -- 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
>
>
> _______________________________________________
> gem5-users mailing 
> [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
>
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Reply via email to