[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <70350444-f05f-c57f-d4d8-0e059b4d896c@silicom-usa.com>
Date: Mon, 3 Dec 2018 17:02:40 +0000
From: Steve Douthit <stephend@...icom-usa.com>
To: Andrew Lunn <andrew@...n.ch>
CC: Jeff Kirsher <jeffrey.t.kirsher@...el.com>,
"David S. Miller" <davem@...emloft.net>,
"intel-wired-lan@...ts.osuosl.org" <intel-wired-lan@...ts.osuosl.org>,
"netdev@...r.kernel.org" <netdev@...r.kernel.org>,
Florian Fainelli <f.fainelli@...il.com>
Subject: Re: [PATCH net-next v2 1/2] ixgbe: register a mdiobus
On 12/3/18 11:54 AM, Andrew Lunn wrote:
>> +static s32 ixgbe_x550em_a_mii_bus_read(struct mii_bus *bus, int addr,
>> + int regnum)
>> +{
>> + struct ixgbe_adapter *adapter = (struct ixgbe_adapter *)bus->priv;
>> + struct ixgbe_hw *hw = &adapter->hw;
>> + u32 gssr = hw->phy.phy_semaphore_mask | IXGBE_GSSR_TOKEN_SM;
>> +
>> + if (hw->bus.lan_id)
>> + gssr |= IXGBE_GSSR_PHY1_SM;
>> + else
>> + gssr |= IXGBE_GSSR_PHY0_SM;
>
> Hi Steve
>
> If you only have one bus, do you still need this? One semaphore is all
> you need. And i'm not even sure you need that. The MDIO layer will
> perform locking, assuming everything goes through the MDIO layer.
AFAIK I still need part of it. There's a PHY polling unit in the card
that we need to sync with independent of the locking in the MDIO layer.
I can drop the hw->bus.lan_id check though and unconditionally OR in the
IXGBE_GSSR_PHY0_SM since we always register on function 0 now.
Powered by blists - more mailing lists