Bug 206219

Summary:	e1000e does not come up anymore
Product:	Drivers	Reporter:	nvaert1986 (nvaert1986)
Component:	Network	Assignee:	drivers_network (drivers_network)
Status:	RESOLVED DUPLICATE
Severity:	high
Priority:	P1
Hardware:	x86-64
OS:	Linux
Kernel Version:	5.4.12	Subsystem:
Regression:	Yes	Bisected commit-id:

Description nvaert1986 2020-01-16 09:48:48 UTC

Ever since I upgraded my kernel 5.4.x on my Lenovo ThinkPad P52 my Intel Onboard NIC stopped working. It tries to come online, but just doesn't and spams dmesg full with the message below approximately every 7 seconds.

e1000e: enp0s31f6 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None

This is definitely a new bug as it does not occur with previous kernel. So there has to be a regression somewhere.

Right now I'm temporarily using a Realtek USB nic (using r8152) which is working fine.

I've tried to connect the NIC to various switches, but has the same issue with all switches. It works fine on 5.3.18.

Kernel snippet:
[    1.499833] e1000e: Intel(R) PRO/1000 Network Driver - 3.2.6-k
[    1.499834] e1000e: Copyright(c) 1999 - 2015 Intel Corporation.
[    1.500182] e1000e 0000:00:1f.6: Interrupt Throttling Rate (ints/sec) set to dynamic conservative mode
[    1.686013] e1000e 0000:00:1f.6 0000:00:1f.6 (uninitialized): registered PHC clock
[    1.748777] e1000e 0000:00:1f.6 eth1: (PCI Express:2.5GT/s:Width x1) e8:6a:64:aa:7b:a1
[    1.748778] e1000e 0000:00:1f.6 eth1: Intel(R) PRO/1000 Network Connection
[    1.748835] e1000e 0000:00:1f.6 eth1: MAC: 13, PHY: 12, PBA No: FFFFFF-0FF
[    1.750721] e1000e 0000:00:1f.6 enp0s31f6: renamed from eth1
[   15.423202] e1000e: enp0s31f6 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
[   22.279151] e1000e: enp0s31f6 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None
[   29.005512] e1000e: enp0s31f6 NIC Link is Up 1000 Mbps Full Duplex, Flow Control: None

Comment 1 nvaert1986 2020-01-16 09:54:29 UTC

Update: I've tried e100e version 3.6.0 and 3.5.1 from the Intel website, but they do not compile with kernel 5.4.x unfortunately.

Comment 2 nvaert1986 2020-01-16 10:30:48 UTC

Update: Also found the issue here: http://lkml.iu.edu/hypermail/linux/kernel/1910.3/07624.html seems like the same issue and it's still present in the stable kernel.

Comment 3 nvaert1986 2020-01-16 10:36:30 UTC

Got the e1000e driver version 3.6.0 compiled by removing line 2799 from kcompat.h:2799, but the issue still occurs in my case.

Comment 4 nvaert1986 2020-01-16 11:36:37 UTC

Seems to be a duplicate of: https://bugzilla.kernel.org/show_bug.cgi?id=205047#c33.

Attachment: https://bugzilla.kernel.org/attachment.cgi?id=285979&action=diff from bug: https://bugzilla.kernel.org/show_bug.cgi?id=205047#c33 resolves the issue for me, please revert this patch in the mainline kernel.

*** This bug has been marked as a duplicate of bug 205047 ***