Project

General

Profile

Edit History

Wiki » EMIR » PV Center (Case Utility Infrastructure) »

Failure verify SOP » History » Version 1

jun chen, 03/23/2025 12:39 AM

1 1 jun chen 
# Failure verify SOP
2 





    3
    
    
    
Mass failure can occur due to critical code change.





    4
    
    
    
Or, if RD tried to introduce some fundamental enhance, which may break existing regression suite.





    5
    
    
    





    6
    
    
    
## WARNING:





    7
    
    
    





    8
    
    
    
**A COMPLETE FAILURE REPORT IS NECESSARY





    9
    
    
    
Direct re-golden is prohibited**





    10
    
    
    





    11
    
    
    
## Procedure





    12
    
    
    





    13
    
    
    
Step 1: run regression and investigate log/reg.log





    14
    
    
    
Step 2: try to understand what command failure and why





    15
    
    
    
Step 3: if failure is caused by tiny variation, a correlation check is a must.





    16
    
    
    





    17
    
    
    
## Discussion Demo:





    18
    
    
    





    19
    
    
    
Fix sanity check and re-golden related case. attachment:"resistance_sanity_check_regolden.docx" (ChunYan) 





    20
    
    
    
Feedback and regression failure sharing from Junc: vsupply_current_difference.pptx"





    21
    
    
    





    22
    
    
    
## See below examples:





    23
    
    
    





    24
    
    
    
1. PG mesh change, cause 60+ regression fail, each case has resistance and ir comparison to prove tiny variation. attachment:"node_reduction_investigation.pptx" (XiaoHu, DanPing)





    25
    
    
    
2. twf, em code change cause 7+ regression fail, each case has log tcl investigation and solution attachment:"pow_twf_em_fail_case.pptx" (LangYun, Heng)





    26
    
    
    
3. PG pad change, caused 20+ case fail, each fail needs detailed report and discussion attachment:"use_pad_phypin_as_pg_with_feedback.pptx" (Min)