e1000e 0000:00:1f.6 eno1: Detected Hardware Unit Hang

lordnaikon · 2021-12-02 19:05:45

Hi,
i am running a Dell Latitude 5421 that has a

$ lspci
...
0000:00:1f.6 Ethernet controller: Intel Corporation Ethernet Connection (14) I219-LM (rev 11)

running with the e1000e driver (i think) without any problems the last few weeks (new machine). Only recently i encountered connection problems (was on wifi mostly so i can't say when this happened for sure). journalctl is hitting me with

Dez 02 19:56:57 dustin-dell ntpd[1538]: Deleting interface #260 enp0s31f6, fe80::f622:8d0a:ff29:8b79%2#123, interface stats: received=0, sent=0, dropped=0, active_time=11 secs
Dez 02 19:56:59 dustin-dell NetworkManager[1523]: <info>  [1638471419.4117] device (enp0s31f6): carrier: link connected
Dez 02 19:56:59 dustin-dell kernel: e1000e 0000:00:1f.6 enp0s31f6: NIC Link is Up 1000 Mbps Full Duplex, Flow Control: Rx/Tx
Dez 02 19:57:00 dustin-dell ntpd[1538]: Listen normally on 261 enp0s31f6 [fe80::f622:8d0a:ff29:8b79%2]:123
Dez 02 19:57:00 dustin-dell ntpd[1538]: new interface(s) found: waking up resolver
Dez 02 19:57:11 dustin-dell kernel: e1000e 0000:00:1f.6 enp0s31f6: Detected Hardware Unit Hang:
                                      TDH                  <3>
                                      TDT                  <4>
                                      next_to_use          <4>
                                      next_to_clean        <3>
                                    buffer_info[next_to_clean]:
                                      time_stamp           <1008cd457>
                                      next_to_watch        <3>
                                      jiffies              <1008cd780>
                                      next_to_watch.status <0>
                                    MAC Status             <40080283>
                                    PHY Status             <796d>
                                    PHY 1000BASE-T Status  <3800>
                                    PHY Extended Status    <3000>
                                    PCI Status             <10>

constantly. I found i few older Topics in the Forum but nothing conclusive. Looks like it es fairly recent. What i did found was some discussion here https://bugzilla.kernel.org/show_bug.cgi?id=213377 and here https://bugzilla.kernel.org/show_bug.cgi?id=213651. People in there talking about very similar if not the same hardware and wondering if these "fixed" recently reached the stable kernel and is causing the problems i have after i updated the system/kernel to the version where those "fixes" are introduced

$ uname -a
Linux dustin-dell 5.15.5-arch1-1 #1 SMP PREEMPT Thu, 25 Nov 2021 22:09:33 +0000 x86_64 GNU/Linux

lordnaikon · 2021-12-06 06:50:23

Looks like it appears mostly after returning from hibernate/suspend mode. After a clean reboot it looks like to be ok.

seth · 2021-12-06 09:24:45

https://forum.proxmox.com/threads/e1000 … 284/page-8

Try

ethtool -K enp0s31f6 gso off gro off tso off tx off rx off rxvlan off txvlan off sg off

Looks like it appears mostly after returning from hibernate/suspend mode.

ceterum censeo: do you have a parallel windows installation?

wondering if these "fixed" recently reached the stable kernel

https://git.kernel.org/pub/scm/linux/ke … .h?h=v5.15
No issues w/ the lts kernel? You can try to downgrade to 5.14 (don't forget out of tree modules like nvidia, but otherwise the kernel downgrade should be fine)

lordnaikon · 2021-12-18 11:37:28

Thanks @seth and sorry for the delay. I tried ethtool and it didn't help much and i don't really want to work with an "old" kernel. After fiddling around with it, it looks like it is mostly triggered by the suspend and a restart works almost every time and no i don't have a parallel windows installation or any other system parallel to my arch installation. I don't think there is anything left to do and this needs to be fixed upstream.

seth · 2021-12-18 14:20:44

The downgrade/lts thing is to

wondering if these "fixed" recently reached the stable kernel

identify/confirm the regression.

You'll likely be asked to https://wiki.archlinux.org/title/Bisect … s_with_Git it anyway (and catching it between two binary releases will lower the amount of required rebuilds)

Arch Linux

#1 2021-12-02 19:05:45

e1000e 0000:00:1f.6 eno1: Detected Hardware Unit Hang

#2 2021-12-06 06:50:23

Re: e1000e 0000:00:1f.6 eno1: Detected Hardware Unit Hang

#3 2021-12-06 09:24:45

Re: e1000e 0000:00:1f.6 eno1: Detected Hardware Unit Hang

#4 2021-12-18 11:37:28

Re: e1000e 0000:00:1f.6 eno1: Detected Hardware Unit Hang

#5 2021-12-18 14:20:44

Re: e1000e 0000:00:1f.6 eno1: Detected Hardware Unit Hang

Board footer