Okay thanks let me try internally. Thanks Brendan. 

--------------------------------------------
On Wed, 15/6/16, Brendan Fitzgerald <[email protected]> wrote:

 Subject: Re: [marss86-devel] strange IPC results with SPEC CPU2006
 To: "saravanan s" <[email protected]>
 Cc: "msc" <[email protected]>, "MARSS List" <[email protected]>
 Date: Wednesday, 15 June, 2016, 7:24 PM
 
 Because
 SPEC is a licensed piece of software, we cannot just give it
 out. You have to purchase a license yourself if you wish to
 use it.
 On Wed, Jun 15, 2016 at
 9:49 AM, saravanan s <[email protected]>
 wrote:
 Hello:
 Could you please share the steps to get the CPU 2006
 licensed version or image where i can download ? and I will
 follow the steps to integrate into ubunty.natty.qcow2 image.
 Thank you for your help!
 
 
 
 I have only bzip2 which comes with default. But I would like
 to run all other workload as well. Please help me with that.
 Thank you!
 
 
 
 I am able to run ParsecROI, Stream, Splash2..yet to get the
 SPEC CPU 2006. Please provide more details to me. Thank
 you!
 
 
 
 With Best Regards,
 
 SARAVANAN.S
 
 Senior Research Engineer,
 
 IBM India Ltd
 
 Systems and Technologies Group,
 
 Embassy Golf Links Business Park,
 
 Bangalore-560 071
 
 Tel(Direct):91-80-41777346
 
 
 
 --------------------------------------------
 
 On Tue, 31/5/16, Brendan Fitzgerald <[email protected]>
 wrote:
 
 
 
  Subject: Re: [marss86-devel] strange IPC results with SPEC
 CPU2006
 
  To: "msc" <[email protected]>
 
  Cc: "MARSS List" <[email protected]>
 
  Date: Tuesday, 31 May, 2016, 5:37 PM
 
 
 
  With only
 
  100M cycles, I can' be sure the program got to any
 
  meaningful portions. 
 
  There
 
  are a few options you can try:1) Increase the
 
  number of cycles to something much higher. Think over a
 
  billion2) Run the benchmark directly instead of
 
  using the runspec command.3) Insert the ptl_call
 
  to start the simulation within the spec
 
  benchmarks
 
  When I run
 
  spec benchmarks, I can see IPC differences from 0.27 to
 
  1.5
 
  On Tue,
 
  May 31, 2016 at 12:00 AM, msc <[email protected]>
 
  wrote:
 
  Hi
 
  Yesterday, I created a
 
  SPEC_CPU2006.qcow2 image disk based on
 ubuntu-natty.qcow2
 
  provided from MARSSx86 official website. I installed
 CPU
 
  2006 files on '/root/' directory to make
 simulations
 
  easy with 'start_sim, stop_sim,
 
  kill_sim'. After the installation
 
  was completed, I followed CPU2006 documentation to build
 and
 
  run the workloads. The build operations were completed
 
  without any errors, and I used 'runspec' command
 as
 
  follows.
 
  ./start_sim ;
 
  runspec --config=myconfig.cfg --size=ref --noreportable
 
  --iterations=1 bzip2; ./stop_sim ;./kill_sim
 
  also, I ran all workloads based on
 
  that command lines. (additionally, I specified
 stopcycle
 
  100M and single_core, )
 
  However, today, when I check the
 
  simulation results, I show some strange results. All
 IPC
 
  results in log files are similar. To show what I say, I
 give
 
  you some IPC results as follows.
 
                GemsFDTD  
 
   lbm             wrf          sphinx3    
  
 
  h264ref       gcc         libquantumuser
 
           0.97849     0.97876     1.01513    
  
 
  0.97197      0.96678     0.99265    
 
   1.01742kernel        0.57844     0.59495
 
      0.57975      0.57746       0.59781    
 0.59026
 
       0.60412total          0.86600    
 
   0.87202     0.88925      0.85880      0.86203
  
 
    0.87850      0.89806
 
  (other workloads also have similar
 
  results)
 
  Recently,
 
  when I used parsecROI.qcow2, all workloads show
 different
 
  IPC results. I give you PARSEC workloads' IPC results
 (I
 
  ran the workloads with 10B cycles and
 
  single_core).
 
       
 
          swaptions        fluidanimate     vips
  
 
         freqmineuser          1.14556  
 
          1.28826      1.71456      
 
   1.12002kernel        0.37523          
 
  0.66409      0.91947        0.79926total
 
           1.13923           1.27027    
  1.64984
 
         1.10204
 
 
 
  With the PARSEC results, I can't
 
  be sure that the SPEC CPU2006 disk I made is operated
 
  correctly, and I don't know what I missed if there
 are
 
  some problems. So, I want you to answer me about its
 
  correctness. If some problems are, please let me know how
 to
 
  solve the problems. 
 
  Thanks in advance.
 
 
 
 
 
 
 
 
 
 
 
  _______________________________________________
 
 
 
  http://www.marss86.org
 
 
 
  Marss86-Devel mailing list
 
 
 
  [email protected]
 
 
 
  https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
 
 
 
 
 
 
 
 
 
  -----Inline Attachment Follows-----
 
 
 
  _______________________________________________
 
  http://www.marss86.org
 
  Marss86-Devel mailing list
 
  [email protected]
 
  https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel
 
 

_______________________________________________
http://www.marss86.org
Marss86-Devel mailing list
[email protected]
https://www.cs.binghamton.edu/mailman/listinfo/marss86-devel

Reply via email to