From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754035AbdBGLmZ convert rfc822-to-8bit (ORCPT ); Tue, 7 Feb 2017 06:42:25 -0500 Received: from smtp-out6.electric.net ([192.162.217.190]:55341 "EHLO smtp-out6.electric.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753769AbdBGLmX (ORCPT ); Tue, 7 Feb 2017 06:42:23 -0500 From: David Laight To: "'Andrew Lunn'" , Florian Fainelli CC: "netdev@vger.kernel.org" , "zefir.kurtisi@neratec.com" , "Sebastian Hesselbarth" , Wingman Kwok , Murali Karicheri , Vivien Didelot , "David S. Miller" , Philippe Reynes , Sergei Shtylyov , Jisheng Zhang , Jarod Wilson , open list Subject: RE: [PATCH net-next 4/4] net: dsa: Do not clobber PHY link outside of state machine Thread-Topic: [PATCH net-next 4/4] net: dsa: Do not clobber PHY link outside of state machine Thread-Index: AQHSgNbrmSEQFOqtUk2BEIHPWfn/aKFdazsA Date: Tue, 7 Feb 2017 11:42:12 +0000 Message-ID: <063D6719AE5E284EB5DD2968C1650D6DB027DB40@AcuExch.aculab.com> References: <20170206235523.23216-1-f.fainelli@gmail.com> <20170206235523.23216-5-f.fainelli@gmail.com> <20170207001132.GB30169@lunn.ch> In-Reply-To: <20170207001132.GB30169@lunn.ch> Accept-Language: en-GB, en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.202.99.200] Content-Type: text/plain; charset="Windows-1252" Content-Transfer-Encoding: 8BIT MIME-Version: 1.0 X-Outbound-IP: 213.249.233.130 X-Env-From: David.Laight@ACULAB.COM X-Proto: esmtps X-Revdns: X-HELO: AcuExch.aculab.com X-TLS: TLSv1:AES128-SHA:128 X-Authenticated_ID: X-PolicySMART: 3396946, 3397078 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org From: Andrew Lunn > Sent: 07 February 2017 00:12 > On Mon, Feb 06, 2017 at 03:55:23PM -0800, Florian Fainelli wrote: > > Calling phy_read_status() means that we may call into > > genphy_read_status() which in turn will use genphy_update_link() which > > can make changes to phydev->link outside of the state machine's state > > transitions. This is an invalid behavior that is now caught as of > > 811a919135b9 ("phy state machine: failsafe leave invalid RUNNING state") > > > > Reported-by: Zefir Kurtisi > > Signed-off-by: Florian Fainelli > > --- > > net/dsa/slave.c | 10 +++------- > > 1 file changed, 3 insertions(+), 7 deletions(-) > > > > diff --git a/net/dsa/slave.c b/net/dsa/slave.c > > index 09fc3e9462c1..4b6fb6b14de4 100644 > > --- a/net/dsa/slave.c > > +++ b/net/dsa/slave.c > > @@ -651,14 +651,10 @@ dsa_slave_get_link_ksettings(struct net_device *dev, > > struct ethtool_link_ksettings *cmd) > > { > > struct dsa_slave_priv *p = netdev_priv(dev); > > - int err; > > + int err = -EOPNOTSUPP; > > > > - err = -EOPNOTSUPP; > > - if (p->phy != NULL) { > > - err = phy_read_status(p->phy); > > - if (err == 0) > > - err = phy_ethtool_ksettings_get(p->phy, cmd); > > - } > > + if (p->phy != NULL) > > + err = phy_ethtool_ksettings_get(p->phy, cmd); > > Hi Florian > > So what we are effectively doing is returning the state from the last > poll/interrupt. The poll information could be up to 1 second out of > date, but those PHYs using interrupts should give more fresh > information. Another option is to request the state machine re-read the phy status and wakeup the caller when it has the result. Related is that polling the phy status every second can be bad news for system latency. Typically it involves bit-bashing an interface with spin-delays to meet the slow timings (with extra delays to allow for posted writes). An alternative would be to do a very slow bit-bang on each timer tick, this would need to be sped up and finished before any other actions. David