Re: [PATCH net-next] net: phylink: fix NULL pl->pcs dereference during phylink_pcs_poll_start

On Thu, Jun 30, 2022 at 03:46:54PM +0200, Nicolas Ferre wrote:

Vladimir, Russell,

On 29/06/2022 at 21:33, Vladimir Oltean wrote:

quoted

The current link mode of the phylink instance may not require an
attached PCS. However, phylink_major_config() unconditionally
dereferences this potentially NULL pointer when restarting the link poll
timer, which will panic the kernel.

Fix the problem by checking whether a PCS exists in phylink_pcs_poll_start(),
otherwise do nothing. The code prior to the blamed patch also only
looked at pcs->poll within an "if (pcs)" block.

Fixes: bfac8c490d60 ("net: phylink: disable PCS polling over major configuration")
Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
---
  drivers/net/phy/phylink.c | 2 +-
  1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/phy/phylink.c b/drivers/net/phy/phylink.c
index 1a7550f5fdf5..48f0b9b39491 100644
--- a/drivers/net/phy/phylink.c
+++ b/drivers/net/phy/phylink.c

@@ -766,7 +766,7 @@ static void phylink_pcs_poll_stop(struct phylink *pl)
  static void phylink_pcs_poll_start(struct phylink *pl)
  {
-	if (pl->pcs->poll && pl->cfg_link_an_mode == MLO_AN_INBAND)
+	if (pl->pcs && pl->pcs->poll && pl->cfg_link_an_mode == MLO_AN_INBAND)
  		mod_timer(&pl->link_poll, jiffies + HZ);
  }

Fixes the NULL pointer on my boards:
Tested-by: Nicolas Ferre <nicolas.ferre@microchip.com> # on sam9x60ek

Thanks all, hopefully it'll get applied to net-next soon.

Sadly, this slipped through my testing, as the only platform I have
access to at the moment always supplies a PCS (mvneta based) so there's
no way my testing would ever have caught this. Sorry for the problems.

-- 
RMK's Patch system: https://www.armlinux.org.uk/developer/patches/
FTTP is here! 40Mbps down 10Mbps up. Decent connectivity at last!