linux-kernel - Re: [PATCH] net: phy: micrel: make *-skew-ps check more lenient

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]

Date:   Wed, 13 Oct 2021 10:46:19 +0200
From:   Matthias Schiffer <matthias.schiffer@...tq-group.com>
To:     Philippe Schenker <philippe.schenker@...adex.com>
Cc:     "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "davem@...emloft.net" <davem@...emloft.net>,
        "kuba@...nel.org" <kuba@...nel.org>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        "linux@...linux.org.uk" <linux@...linux.org.uk>,
        "hkallweit1@...il.com" <hkallweit1@...il.com>,
        "andrew@...n.ch" <andrew@...n.ch>
Subject: Re: [PATCH] net: phy: micrel: make *-skew-ps check more lenient

On Wed, 2021-10-13 at 08:25 +0000, Philippe Schenker wrote:
> On Tue, 2021-10-12 at 12:34 +0200, Matthias Schiffer wrote:
> > It seems reasonable to fine-tune only some of the skew values when
> > using
> > one of the rgmii-*id PHY modes, and even when all skew values are
> > specified, using the correct ID PHY mode makes sense for documentation
> > purposes. Such a configuration also appears in the binding docs in
> > Documentation/devicetree/bindings/net/micrel-ksz90x1.txt, so the
> > driver
> > should not warn about it.
> 
> I don't think your commit message is right. The rgmii-*id PHY modes are
> no longer just for documentation purposes on KSZ9031 PHY. They are used
> to set the skew-registers according to .

Yes, this was implemented in [1]. The commit message explicitly states
that fine-tuning is still possible using *-skew-ps.

> 
> The warning is there, that in case you override the skew registers of
> one of the modes rgmii-id, rgmii-txid, rgmii-rxid with *-skew-ps
> settings in DT.

The "rgmii" mode should not be handled differently from "rgmii-*id" in
my opinion. Otherwise for a device that is basically "rgmii-id", but
requires slight fine-tuning, you have to set the mode to the incorrect
value "rgmii" in the DTS to avoid this warning.


> 
> Therefore I also think the warning is valuable and should be kept. We
> may want to reword it though.
> 
> Philippe

[1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/drivers/net/phy/micrel.c?id=bcf3440c6dd78bfe5836ec0990fe36d7b4bb7d20


> 
> > 
> > Signed-off-by: Matthias Schiffer <matthias.schiffer@...tq-group.com>
> > ---
> >  drivers/net/phy/micrel.c | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/net/phy/micrel.c b/drivers/net/phy/micrel.c
> > index c330a5a9f665..03e58ebf68af 100644
> > --- a/drivers/net/phy/micrel.c
> > +++ b/drivers/net/phy/micrel.c
> > @@ -863,9 +863,9 @@ static int ksz9031_config_init(struct phy_device
> > *phydev)
> >                                 MII_KSZ9031RN_TX_DATA_PAD_SKEW, 4,
> >                                 tx_data_skews, 4, &update);
> >  
> > -               if (update && phydev->interface !=
> > PHY_INTERFACE_MODE_RGMII)
> > +               if (update && !phy_interface_is_rgmii(phydev))
> >                         phydev_warn(phydev,
> > -                                   "*-skew-ps values should be used
> > only with phy-mode = \"rgmii\"\n");
> > +                                   "*-skew-ps values should be used
> > only with RGMII PHY modes\n");
> >  
> >                 /* Silicon Errata Sheet (DS80000691D or DS80000692D):
> >                  * When the device links in the 1000BASE-T slave mode
> > only,
> 
>