Re: [PATCH net] r8169: fix LED-related deadlock on module removal

[Lukas Wunner <lukas@wunner.de>]

On Wed, Apr 17, 2024 at 09:43:27AM +0200, Greg KH wrote:
> On Wed, Apr 17, 2024 at 09:16:04AM +0200, Heiner Kallweit wrote:
> > On 17.04.2024 09:04, Greg KH wrote:
> > > On Wed, Apr 17, 2024 at 08:02:31AM +0200, Heiner Kallweit wrote:
> > >> On 17.04.2024 04:34, Jakub Kicinski wrote:
> > >>> On Mon, 15 Apr 2024 13:57:17 +0200 Heiner Kallweit wrote:
> > >>>> Binding devm_led_classdev_register() to the netdev is problematic
> > >>>> because on module removal we get a RTNL-related deadlock. Fix this
> > >>>> by avoiding the device-managed LED functions.
> > >>>>
> > >>>> Note: We can safely call led_classdev_unregister() for a LED even
> > >>>> if registering it failed, because led_classdev_unregister() detects
> > >>>> this and is a no-op in this case.
> > >>>>
> > >>>> Fixes: 18764b883e15 ("r8169: add support for LED's on RTL8168/RTL8101")
> > >>>> Cc: <stable@vger.kernel.org> # 6.8.x
> > >>>> Reported-by: Lukas Wunner <lukas@wunner.de>
> > >>>> Signed-off-by: Heiner Kallweit <hkallweit1@gmail.com>
> > >> 
> > >> This is a version of the fix modified to apply on 6.8.
> > > 
> > > That was not obvious at all :(
> > > 
> > Stating "Cc: <stable@vger.kernel.org> # 6.8.x" isn't sufficient?
> 
> Without showing what commit id this is in Linus's tree, no.

The upstream commit id *is* called out in the patch, but it's buried
below the three dashes:

    The original change was introduced with 6.8, 6.9 added support for
    LEDs on RTL8125. Therefore the first version of the fix applied on
    6.9-rc only. This is the modified version for 6.8.
    Upstream commit: 19fa4f2a85d7
                     ^^^^^^^^^^^^

The proper way to do this is to prominently add ...

    commit 19fa4f2a85d777a8052e869c1b892a2f7556569d upstream.

... or ...

    [ Upstream commit 19fa4f2a85d777a8052e869c1b892a2f7556569d ]

... as the first line of the commit message, as per
Documentation/process/stable-kernel-rules.rst