DCE, nvrng and nvethernet drivers happen errors when resume from suspend

Robotics & Edge Computing Jetson & Embedded Systems Jetson AGX Orin
1

Hello. I found issues about suspend/resume function with the reference document.

We get some driver error logs while doing resume from suspend on Orin generation. Those are not expected behavior. Could you please investigate and modify?

Error messages

  • DCE error

    DCE: FW Boot Done
[   60.754412] CPU:0, Error: dce-fabric@0xde00000, irq=37
[   60.754416] **************************************
[   60.754420] CPU:0, Error:dce-fabric, Errmon:4
[   60.754429]    Error Code            : SLAVE_ERR
[   60.754431]    Overflow              : Multiple SLAVE_ERR
[   60.754441]
[   60.754443]    Error Code            : SLAVE_ERR
[   60.754445]    MASTER_ID             : DCE
[   60.754447]    Address               : 0xdc9ed80
[   60.754451]    Cache                 : 0x3 -- Bufferable Modifiable
[   60.754455]    Protection            : 0x2 -- Unprivileged, Non-Secure, Data Access
[   60.754457]    Access_Type           : Read
[   60.754459]    Access_ID             : 0x0
[   60.754462]    Fabric                : dce-fabric
[   60.754463]    Slave_Id              : 0x5
[   60.754465]    Burst_length          : 0x7
[   60.754467]    Burst_type            : 0x1
[   60.754468]    Beat_size             : 0x3
[   60.754470]    VQC                   : 0x0
[   60.754472]    GRPSEC                : 0x3f
[   60.754474]    FALCONSEC             : 0x0
[   60.754478]  **************************************

  • nvrng error

    [   61.440872] PM: dpm_run_callback(): platform_pm_resume+0x0/0x80 returns -110
[   61.440877] PM: Device 3ae0000.nvrng failed to resume: error -110

  • nvethernet error

    [   71.871547] nvethernet 6810000.ethernet eth0: ether_get_wol: phydev is null check iface up status

  • susres-20230830r1.log (112.5 KB)

Reproduce steps

Step:

  1. Turn on power
  2. echo +20 > /sys/class/rtc/rtc1/wakealarm; sleep 3; systemctl suspend
  3. It fails to resume

Related information

3

Hi,

We don’t check various issue in one topic. I would suggest you can file 3 topic for each and share the detail steps to reproduce the error you saw.

Also, if those don’t affect the normal usecase, I would suggest you can check the release note document first to see if they are all known issue and could be ignored.

4

Thank you for quick response.

If you need to divide each drivers, I can create 2 issues. Do you need? I want nvidia’s bsp kernel team to check. So, those are all nvidia’s device drivers. DCE driver issue is known issue at release note. sorry.

I already gave detail steps for reproduce and error messages. In other words, some
device drivers happen error messages when we do resume.

It’s difficult to estimate whether to keep using normally when error messages in user side. It’s nvrng driver specially. I checked that eth0 can use after resume with the message.

I found about DCE driver issue from release note (Jetson_Linux_Release_Notes_r35.4.1.pdf)

  • 3905997 SC7(suspend/resume) does not work on Jetson devices that are connected to the Display in MST mode. SC7 works as expected with all other DP versions in SST mode.
  • 3420652 Display does not resume after SC7 suspend/resume cycle.

Thanks.

5

Hi,

Yes, please file 2 separates new topics for the remaining 2 issues. And share the detail reproduce steps again.

If that needs to try multiple times, please also let me know how many times it took on your side to reproduce issue.

7

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.