suspend/memory_after_suspend_auto fails with total amount of memory differing before/after the test

[pieqq]

Bug Description

Recent run exhibiting this problem:

https://certification.canonical.com/hardware/201711-25989/submission/409962/

This device is an NVIDIA DGX-Station running 20.04,with 251 GB of RAM.

CPU: Intel(R) Xeon(R) CPU E5-2698 v4 @ 2.20GHz
GPUs: NVIDIA - 10de:1db2 (x4)

The test fails with the following output:

2c2
< total: 270165549056
---
> total: 270165532672

Meaning there is a difference of 16384 Kbytes between the two runs.

We had an internal discussion, and it looks like the issue is related to some graphics tests being executed before the two memory gathering info, resulting in some RAM being used or released for the GPUs.

Cert-blocker Test Case

• cert-blocker

To Reproduce

Run the SRU test plan on that machine.

Expected Result

suspend/memory_after_suspend_auto job passes

Actual Result

suspend/memory_after_suspend_auto fails with output described above.

Environment

20.04, with the latest version of Checkbox (4.3.0dev44ubuntu20.04.1)

Relevant log output

Additional context

No response

[syncronize-issues-to-jira]

Thank you for reporting your feedback to us!

The internal ticket has been created: https://warthogs.atlassian.net/browse/CHECKBOX-1725.

This message was autogenerated