[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <573A02BA.3080709@gmail.com>
Date: Mon, 16 May 2016 10:26:18 -0700
From: Florian Fainelli <f.fainelli@...il.com>
To: Andrew Lunn <andrew@...n.ch>, Rabin Vincent <rabin@....in>
Cc: Rabin Vincent <rabin.vincent@...s.com>,
"David S. Miller" <davem@...emloft.net>, netdev@...r.kernel.org,
linux-m68k@...ts.linux-m68k.org, linux-mips@...ux-mips.org,
devicetree@...r.kernel.org, Rabin Vincent <rabinv@...s.com>
Subject: Re: [PATCH] phy: remove irq param to fix crash in fixed_phy_add()
On 05/16/2016 06:40 AM, Andrew Lunn wrote:
> On Mon, May 16, 2016 at 03:11:35PM +0200, Rabin Vincent wrote:
>> On Mon, May 16, 2016 at 02:29:03PM +0200, Andrew Lunn wrote:
>>> What i think is better is to make fixed_phy_add() return -EPROBE_DEFER
>>> if it is called before fixed_mdio_bus_init().
>>
>> I don't see how this will work for platforms such as ar7 and bcm47xx
>> which call fixed_phy_add() from platform code.
>
> Ah! Not good.
>
> fixed_phy_add() is the lower layer call. What we can do is only access
> fmb->mii_bus->irq[phy_addr] if irq != PHY_POLL. That should make ar7
> and bcm47xx work again.
>
> The higher level function fixed_phy_register() should return
> -EPROBE_DEFER if fixed_mdio_bus_init() has not been called yet.
ar7 and bcm47xx date back from when the fixed MDIO bus needed its fixed
PHYs to be registered before the MDIO bus driver had a chance to probe
them, otherwise, you would not be able to utilize them. Things have
changed now, and your suggestion makes sense.
--
Florian
Powered by blists - more mailing lists