[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Date: Tue, 03 Mar 2020 15:47:04 -0800 (PST)
From: David Miller <davem@...emloft.net>
To: rmk+kernel@...linux.org.uk
Cc: andrew@...n.ch, f.fainelli@...il.com, vivien.didelot@...il.com,
kuba@...nel.org, olteanv@...il.com, ioana.ciornei@....com,
netdev@...r.kernel.org
Subject: Re: [PATCH net v2] net: dsa: fix phylink_start()/phylink_stop()
calls
From: Russell King <rmk+kernel@...linux.org.uk>
Date: Tue, 03 Mar 2020 15:01:46 +0000
> Place phylink_start()/phylink_stop() inside dsa_port_enable() and
> dsa_port_disable(), which ensures that we call phylink_stop() before
> tearing down phylink - which is a documented requirement. Failure
> to do so can cause use-after-free bugs.
>
> Fixes: 0e27921816ad ("net: dsa: Use PHYLINK for the CPU/DSA ports")
> Signed-off-by: Russell King <rmk+kernel@...linux.org.uk>
> ---
> v2: reorder phylink_start() / phylink_stop() as per Andrew's comment.
Applied and queued up for -stable, thanks Russell.
The patch applied with some fuzz but still is correct, so I wonder
if you generated this against net-next or a tree with other local
changes?
Thanks again.
Powered by blists - more mailing lists