Hi Liang,

Please don't email both gem5-dev and gem5-users as it creates two entries on 
the mailing list trackers.


I don't think there is any more testing besides what is described in that 
answer.


I do think it is not enough, and there are some efforts in progress, but fixing 
it is also hard, especially for stats correctness, since it is hard to make a 
golden reference.


So, I recommend that you select the content that you care about, and make sure 
that it runs and produces the expected functional results to start with.


As for SPEC2000, I have never run it myself, but I'm pretty certain it will 
work. Give it a try and describe what you did and how it failed in detail. I 
don't think the stat checks there are passing however.

The closest to master version that you use, the more likely it is that you will 
get help, so stick to master if you can.
________________________________
From: gem5-dev <[email protected]> on behalf of 梁政 
<[email protected]>
Sent: Wednesday, November 14, 2018 12:13:40 PM
To: [email protected]; [email protected]
Subject: [gem5-dev] How to Perform Regression Test

Hi




I am learning to use and want to do some modifications on gem5. I can make my 
version work. But I am not sure whether its' results are right. It seems I 
should use regression test to check it.




However, there is little information about how to perform a canonical 
regression test. Only 1 question 1 in StackOverflow 
(https://stackoverflow.com/questions/52279971/how-to-run-the-gem5-unit-tests). 
And I cannot get the point of the official 
website(http://gem5.org/Regression_Tests).




And I even don't know how to set paths for the example regression test with the 
gem5 official python script (http://gem5.org/Regression_Tests). I have 
SPEC2000, but I don't know how to get mcf and twolf defined in 
config/common/cpu2000.py work. I think it is too horrible.




I am a beginner. I wonder if there are some test scripts to test correctness? 
What do the gem5 developers use? Should I use the latest gem5-dev version to 
obtain some reference statistics?

Thanks very much.




Best Regards

Zheng Liang

EECS, Peking University








_______________________________________________
gem5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/gem5-dev
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.
_______________________________________________
gem5-users mailing list
[email protected]
http://m5sim.org/cgi-bin/mailman/listinfo/gem5-users

Reply via email to