Merge ~alextu/plainbox-provider-pc-sanity:check-nvidia-tips-for-known-issue into plainbox-provider-pc-sanity:master

[Alex Tu]

after count is 0 gpu rc6
after count is 0 gpu rc6p
after count is 0 gpu rc6pp
[ERROR] failed:100.0%;PCI Device: NVIDIA Corporation Device 1f99;
[ERROR] nvidia devices not sleep deep to 0%. Check /var/tmp/checkbox-ng/sessions/checkbox-run-2020-12-28T09.20.33.session/session-share/ondemand_powertop.csv for detail.
[ERROR] There's a know issue LP: #1904762 that Nvidia driver not sleep with kernel 5.10
[ERROR] check-nvidia.sh testing failed.
------------------------------------------------------------------------- >8 ---
Outcome: job failed
Finalizing session that hasn't been submitted anywhere: checkbox-run-2020-12-28T09.20.33
==================================[ Results ]===================================
 ☑ : Generate an entry for each graphics card present in the system.
 ☒ : check if nvidia dgpu behave as we expected.

[Alex Tu (alextu)]

updated the printed information based on YC's suggestion:

--------------[ Running job 2 / 2. Estimated time left: unknown ]---------------
----------------[ check if nvidia dgpu behave as we expected. ]-----------------
ID: com.canonical.certification::miscellanea/check-nvidia
Category: com.canonical.plainbox::miscellanea
... 8< -------------------------------------------------------------------------
+ check-nvidia.sh --out /var/tmp/checkbox-ng/sessions/checkbox-run-2020-12-29T06.04.09.session/session-share
/usr/bin/prime-select
/usr/bin/gpu-manager
/usr/sbin/powertop
[INFO] current mode is on-demand mode.
modprobe cpufreq_stats failedLoaded 0 prior measurements
RAPL device for cpu 0
RAPL Using PowerCap Sysfs : Domain Mask d
RAPL device for cpu 0
RAPL Using PowerCap Sysfs : Domain Mask d
Devfreq not enabled
glob returned GLOB_ABORTED
Preparing to take measurements
Taking 1 measurement(s) for a duration of 3 second(s) each.
PowerTOP outputing using base filename /var/tmp/checkbox-ng/sessions/checkbox-run-2020-12-29T06.04.09.session/session-share/ondemand_powertop.csv
after count is 0 gpu c0
after count is 0 gpu rc6
after count is 0 gpu rc6p
after count is 0 gpu rc6pp
after count is 0 gpu c0
after count is 0 gpu rc6
after count is 0 gpu rc6p
after count is 0 gpu rc6pp
after count is 0 gpu c0
after count is 0 gpu rc6
after count is 0 gpu rc6p
after count is 0 gpu rc6pp
after count is 0 gpu c0
after count is 0 gpu rc6
after count is 0 gpu rc6p
after count is 0 gpu rc6pp
[ERROR] failed:100.0%;PCI Device: NVIDIA Corporation Device 1f99;
[ERROR] nvidia devices not sleep deep to 0%. Check /var/tmp/checkbox-ng/sessions/checkbox-run-2020-12-29T06.04.09.session/session-share/ondemand_powertop.csv for detail.
[INFO] There's a know issue LP: #1904762 that Nvidia driver not sleep with kernel 5.10
[ERROR] check-nvidia.sh testing failed.
------------------------------------------------------------------------- >8 ---
Outcome: job failed
Finalizing session that hasn't been submitted anywhere: checkbox-run-2020-12-29T06.04.09
==================================[ Results ]===================================
 ☑ : Generate an entry for each graphics card present in the system.
 ☒ : check if nvidia dgpu behave as we expected.

[Yuan-Chen Cheng (ycheng-twn)]

LGTM

review: Approve

diff --git a/bin/check-nvidia.sh b/bin/check-nvidia.sh
index 85b0a04..e2a67c7 100755
--- a/bin/check-nvidia.sh
+++ b/bin/check-nvidia.sh
@@ -78,6 +78,9 @@ check_nvidia_sleep() {
     if [ -n "$result" ]; then
        show_error "$result"
        show_error "nvidia devices not sleep deep to 0%. Check $OUTPUT_FOLDER/$filename for detail."
+       if uname -r | grep -q "5.10"; then
+        echo "[INFO] There's a know issue LP: #1904762 that Nvidia driver not sleep with kernel 5.10"
+       fi
        return $function_failed
     fi
 }