From mboxrd@z Thu Jan 1 00:00:00 1970 From: Ivan Khoronzhuk Subject: Re: [PATCH 04/15] drivers: net: cpsw: ethtool: fix accessing to suspended device Date: Thu, 16 Jun 2016 16:32:27 +0300 Message-ID: <5762AA6B.9060003@linaro.org> References: <20160615115603.4897-1-grygorii.strashko@ti.com> <20160615115603.4897-5-grygorii.strashko@ti.com> <57617ECF.2010902@gmail.com> <5761860C.2020502@ti.com> <5762A010.2070201@linaro.org> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <5762A010.2070201@linaro.org> Sender: netdev-owner@vger.kernel.org To: Grygorii Strashko , Florian Fainelli , "David S. Miller" , netdev@vger.kernel.org, Sekhar Nori , Mugunthan V N Cc: linux-kernel@vger.kernel.org, linux-omap@vger.kernel.org, Rob Herring , devicetree@vger.kernel.org List-Id: devicetree@vger.kernel.org On 16.06.16 15:48, Ivan Khoronzhuk wrote: > > > On 15.06.16 19:45, Grygorii Strashko wrote: >> On 06/15/2016 07:14 PM, Florian Fainelli wrote: >>> On 06/15/2016 04:55 AM, Grygorii Strashko wrote: >>>> The CPSW might be suspended by RPM if all ethX interfaces are down, >>>> but it still could be accesible through ethtool interfce. In this case >>>> ethtool operations, requiring registers access, will cause L3 errors and >>>> CPSW crash. >>>> >>>> Hence, fix it by adding RPM get/put calls in ethtool callbcaks which >>>> can access CPSW registers: .set_coalesce(), .get_ethtool_stats(), >>>> .set_pauseparam(), .get_regs() >>> >>> Provided that you implement an ethtool_ops::begin, it will be called >>> before each ethtool operation runs, so that could allow you to eliminate >>> some of the duplication here. Conversely ethtool_ops::end terminates >>> each operation and can be used for that purpose. >> >> Ah. Thanks for the advice. (assume you've meant .complete()) > > Maybe better leave as is, I'm going to add two additional ehttool ops > containing RPM sensitive part and not, better to split code on those, > if possible...and hiding put/get in begin op can call rpm when it's not > needed at all.. But it's not a big deal, ethtool operations are rarelly anyway, and have impact only when interface is down that makes it even more rarely. > >> >> >> > -- Regards, Ivan Khoronzhuk