Re: [BUG]: WARNING: CPU: 5 PID: 0 at net/sched/sch_generic.c:442 dev_watchdog+0x24d/0x260

[Heiner Kallweit]

On 13.04.2021 22:59, Xose Vazquez Perez wrote:
> A non-recurring bug, on 5.11.12-300.fc34.x86_64 (Fedora kernel).
> 
> Thanks.
> 
> 
> 0c:00.0 Ethernet controller [0200]: Realtek Semiconductor Co., Ltd. RTL8111/8168/8411 PCI Express Gigabit Ethernet Controller [10ec:8168] (rev 06)
> 
> [    2.968280] libphy: r8169: probed
> [    2.968844] r8169 0000:0c:00.0 eth0: RTL8168e/8111e, 2c:41:38:9e:98:93, XID 2c2, IRQ 47
> [    2.968849] r8169 0000:0c:00.0 eth0: jumbo features [frames: 9194 bytes, tx checksumming: ko]
> [    4.071966] RTL8211DN Gigabit Ethernet r8169-c00:00: attached PHY driver (mii_bus:phy_addr=r8169-c00:00, irq=IGNORE)
> [    4.323834] r8169 0000:0c:00.0 eth0: Link is Down
> [    6.729111] r8169 0000:0c:00.0 eth0: Link is Up - 1Gbps/Full - flow control rx/tx
> 
> [106378.638739] ------------[ cut here ]------------
> [106378.638757] NETDEV WATCHDOG: eth0 (r8169): transmit queue 0 timed out

This is a standard tx timeout and can have very different reasons.
Few questions:

- Is this a regression? If yes, can you bisect?
- Can you reproduce it? If yes, which type of activity triggers it?