linux-kernel - Re: [PATCH net-next] net: phy: realtek: check validity of 10GbE link-partner advertisement

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <ZwBmycWDB6ui4Y7j@makrotopia.org>
Date: Fri, 4 Oct 2024 23:06:01 +0100
From: Daniel Golle <daniel@...rotopia.org>
To: Andrew Lunn <andrew@...n.ch>
Cc: Heiner Kallweit <hkallweit1@...il.com>,
	Russell King <linux@...linux.org.uk>,
	"David S. Miller" <davem@...emloft.net>,
	Eric Dumazet <edumazet@...gle.com>,
	Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>,
	netdev@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH net-next] net: phy: realtek: check validity of 10GbE
 link-partner advertisement

On Fri, Oct 04, 2024 at 11:17:28PM +0200, Andrew Lunn wrote:
> On Fri, Oct 04, 2024 at 04:50:36PM +0100, Daniel Golle wrote:
> > Only use link-partner advertisement bits for 10GbE modes if they are
> > actually valid. Check LOCALOK and REMOTEOK bits and clear 10GbE modes
> > unless both of them are set.
> > This prevents misinterpreting the stale 2500M link-partner advertisement
> > bit in case a subsequent linkpartner doesn't do any NBase-T
> > advertisement at all.
> > 
> > Signed-off-by: Daniel Golle <daniel@...rotopia.org>
> > ---
> >  drivers/net/phy/realtek.c | 4 ++++
> >  1 file changed, 4 insertions(+)
> > 
> > diff --git a/drivers/net/phy/realtek.c b/drivers/net/phy/realtek.c
> > index c4d0d93523ad..d276477cf511 100644
> > --- a/drivers/net/phy/realtek.c
> > +++ b/drivers/net/phy/realtek.c
> > @@ -927,6 +927,10 @@ static int rtl822x_read_status(struct phy_device *phydev)
> >  		if (lpadv < 0)
> >  			return lpadv;
> >  
> > +		if (!(lpadv & MDIO_AN_10GBT_STAT_REMOK) ||
> > +		    !(lpadv & MDIO_AN_10GBT_STAT_LOCOK))
> > +			lpadv = 0;
> > +
> >  		mii_10gbt_stat_mod_linkmode_lpa_t(phydev->lp_advertising,
> >  						  lpadv);
> 
> I know lpadv is coming from a vendor register, but does
> MDIO_AN_10GBT_STAT_LOCOK and MDIO_AN_10GBT_STAT_REMOK apply if it was
> also from the register defined in 802.3? I'm just wondering if this
> test should be inside mii_10gbt_stat_mod_linkmode_lpa_t()?

Yes, it does apply and I thought the same, but as
mii_10gbt_stat_mod_linkmode_lpa_t is used in various places without
checking those two bits we may break other PHYs which may not use
them (and apparently this is mostly a problem on RealTek PHYs where
all the other bits in the register persist in case of a non-NBase-T-
capable subsequent link-partner after initially being connected to
an NBase-T-capable one).

Maybe we could introduce a new function
mii_10gbt_stat_mod_linkmode_lpa_validate_t()
which calls mii_10gbt_stat_mod_linkmode_lpa_t() but checks LOCOK and
REMOK as a precondition?