Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
MIME-Version: 1.0
References: <1540957223-30984-1-git-send-email-harini.katakam@xilinx.com>
 <1540957223-30984-4-git-send-email-harini.katakam@xilinx.com> <20181031145400.GH20889@lunn.ch>
In-Reply-To: <20181031145400.GH20889@lunn.ch>
From:   Harini Katakam <harinik@xilinx.com>
Date:   Wed, 31 Oct 2018 20:40:25 +0530
Message-ID: <CAFcVECKg4u5t9=TOJ1omVJ=8ves_yvzcDZZA_-bE4OAv96cymg@mail.gmail.com>
Subject: Re: [PATCH 3/4] net: macb: Add pm runtime support
To:     Andrew Lunn <andrew@lunn.ch>
Cc:     Harini Katakam <harini.katakam@xilinx.com>,
        Nicolas Ferre <nicolas.ferre@microchip.com>,
        David Miller <davem@davemloft.net>,
        Claudiu Beznea <claudiu.beznea@microchip.com>,
        netdev@vger.kernel.org, linux-kernel@vger.kernel.org,
        Michal Simek <michal.simek@xilinx.com>,
        Shubhrajyoti Datta <shubhrajyoti.datta@xilinx.com>
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

Hi Andrew,
On Wed, Oct 31, 2018 at 8:24 PM Andrew Lunn <andrew@lunn.ch> wrote:
>
> On Wed, Oct 31, 2018 at 09:10:22AM +0530, Harini Katakam wrote:
> > From: Harini Katakam <harinik@xilinx.com>
> >
> > Add runtime pm functions and move clock handling there.
> > If device is suspended and not a wake device, then return from
> > mdio read/write functions without performing any action because
> > the clocks are not active.
> >
> > Signed-off-by: Shubhrajyoti Datta <shubhrajyoti.datta@xilinx.com>
> > Signed-off-by: Harini Katakam <harinik@xilinx.com>
> > ---
> > Changes from RFC:
> > Updated pm get sync/put sync calls.
> > Removed unecessary clk up in mdio helpers.
>
> This last bit has me worried.
>
> The MDIO bus is a shared bus with a life of its own.  You can have
> multiple PHYs and switches on it. The PHYs can for a different
> Ethernet MAC. Switch drivers will expect to be able to address the
> switch when the interface is down.
>
> The FEC driver did something similar for a while. I had to make MDIO
> read/write runtime PM aware, otherwise i could not access the Ethernet
> switch hanging of its MDIO bus.

Yes, I understand. But I thought I'd handle it when we have a separate
MDIO bus driver. As of now, with macb, I do not see a way, for ex.,
for MAC1 to access MDIO0 when MAC0 is suspended.
However, I will work a bit more on this solution.

With the clk up, I've noticed that there is atleast one PHY status poll
that ends up bringing the clock up after the MAC has been suspended.
Of course phy is supended immediately after that but it just delays
this full power down process.

Regards,
Harini