All of lore.kernel.org
 help / color / mirror / Atom feed
From: Nicolai Buchwitz <nb@tipi-net.de>
To: Mikkel Eskildsen <mke@visotia.com>
Cc: Oliver Neukum <oneukum@suse.com>,
	Breno Leitao <leitao@debian.org>,
	netdev@vger.kernel.org, linux-usb@vger.kernel.org,
	Jakub Kicinski <kuba@kernel.org>,
	Eric Dumazet <edumazet@kernel.org>,
	Paolo Abeni <pabeni@redhat.com>
Subject: Re: [BUG] cdc_ncm: soft lockup with netconsole over an AX88179B dock
Date: Mon, 19 Oct 2026 16:42:02 +0200	[thread overview]
Message-ID: <6aaea20ede0bd16a5138bc126f3ec897@tipi-net.de> (raw)
In-Reply-To: <PRAP189MB185019A5A8D411482BB6E0CDA9982@PRAP189MB1850.EURP189.PROD.OUTLOOK.COM>

Hi Mikkel

On 19.10.2026 15:48, Mikkel Eskildsen wrote:
> Hi,
> 
> My laptop panicked with a soft lockup on 2026-10-16 and again on 
> 2026-10-19 while netconsole was running over a USB dock NIC that uses 
> cdc_ncm. Both times the laptop was docked and netconsole was on that 
> interface. I have turned netconsole off for now.

I was able to reproduce the issue with a similar device.

Seems like netconsole holds the tx lock while the cdc_ncm driver
lets pending network work run in the middle of the send.
... which then starves on the lock netconsole still holds.

Below is a patch which hopefully fixes the issue.

> 
> The trace below is from pstore. The netconsole thread on CPU15 is in 
> netpoll_send_skb -> usbnet_start_xmit -> cdc_ncm_tx_fixup. The 
> spin_unlock_bh() in cdc_ncm_tx_fixup runs pending softirqs, TCP RX 
> sends a packet and it spins in sch_direct_xmit. My guess is that it 
> waits for the eth0 tx queue lock that netpoll already holds. The other 
> CPUs then wait on the socket lock until the watchdog panics.
> 
> The first panic looks the same, but pstore lost the part with the 
> netconsole thread.
> 

> [...]

> I can test a patch if that helps. Is this a known issue?

https://github.com/nbuchwitz/linux/commit/319cb0933e2eaf984b98906ebddb55befb4e2996

> 
> Mikkel

  parent reply	other threads:[~2026-10-19 14:42 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-19 13:48 [BUG] cdc_ncm: soft lockup with netconsole over an AX88179B dock Mikkel Eskildsen
2026-10-19 13:56 ` Mikkel Eskildsen
2026-10-19 14:42 ` Nicolai Buchwitz [this message]
2026-10-19 17:37   ` Mikkel Eskildsen

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=6aaea20ede0bd16a5138bc126f3ec897@tipi-net.de \
    --to=nb@tipi-net.de \
    --cc=edumazet@kernel.org \
    --cc=kuba@kernel.org \
    --cc=leitao@debian.org \
    --cc=linux-usb@vger.kernel.org \
    --cc=mke@visotia.com \
    --cc=netdev@vger.kernel.org \
    --cc=oneukum@suse.com \
    --cc=pabeni@redhat.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.