Dear Kim,
Your question is not clear. However, I will answer it according to what I 
understand.
cpu0.cycle-break: will tell force a job to stop after running this number of 
cycles.
flexus.set-region-interval: will tell Flexus when to reset the statistics it 
gathered when this region finishes. therefore, if you want to run of 100M 
cycles for a job and specify this parameter to 10M then you will run your 
single job and get 10 regions of statistics for that job.
if you use too many regions you will have statistics from various parts of your 
code. However, the stats will be collected over fewer number of instructions 
per region. You can use these parameters the way they fit your needs.

the number of flexpoints that you generate is determined by the number you 
specify in the run_job command you use to generate the flexpoints.

Regards,
Mutaz

________________________________________
From: Eun-Sung Kim [[email protected]]
Sent: Thursday, February 17, 2011 10:57 PM
To: [email protected]
Subject: Optimal number of flexpoints and regions

Hi all,

Assume that a workload is executed 100M cycles.
Then, parameter are set up as follows. As a result after run_job for creating 
flexpoints, generated flexpoint and regions are shown,
cpu0.cycle-break/  flexus.set-region-interval (array set 
flexus_commands_flexpoint in .run_job.rc.tcl)  -> generated flexpoints/  
generated regions
100M/ 100M -> 1/  1
100M/  10M -> 1/ 10
100M/    5M -> 1/ 20

10M/ 10M -> 10/ 1
10M/   5M -> 10/ 2
10M/   1M -> 10/10     ..... etc.

How can I choice the optimal value of each 
parameter(cpu0.cycle-break/flexus.set-region-interval)?
I'm not sure whether more slices would be better.

--Kim


Reply via email to