Re: [PATCH 0/3] at803x: Add quirk to disable SmartEEE

From: Heiner Kallweit <hkallweit1@gmail.com>
To: Russell King - ARM Linux admin <linux@armlinux.org.uk>
Cc: mark.rutland@arm.com, andrew@lunn.ch, f.fainelli@gmail.com,
	devicetree@vger.kernel.org, s.hauer@pengutronix.de,
	Carlo Caione <ccaione@baylibre.com>,
	robh+dt@kernel.org, abailon@baylibre.com, kernel@pengutronix.de,
	fabio.estevam@nxp.com, shawnguo@kernel.org, aisheng.dong@nxp.com,
	linux-arm-kernel@lists.infradead.org, linux-imx@nxp.com
Subject: Re: [PATCH 0/3] at803x: Add quirk to disable SmartEEE
Date: Fri, 25 Jan 2019 20:27:16 +0100	[thread overview]
Message-ID: <3e09ef54-f817-2659-4e9e-74a88a80cbcc@gmail.com> (raw)
In-Reply-To: <20190125191947.um4w4e5jpsv2kcxc@e5254000004ec.dyn.armlinux.org.uk>

On 25.01.2019 20:19, Russell King - ARM Linux admin wrote:
> On Fri, Jan 25, 2019 at 08:07:56PM +0100, Heiner Kallweit wrote:
>> On 25.01.2019 19:48, Carlo Caione wrote:
>>> On 25/01/19 13:06, Russell King - ARM Linux admin wrote:
>>>> On Fri, Jan 25, 2019 at 12:55:10PM +0000, Carlo Caione wrote:
>>>>> SmartEEE, compatible with IEEE802.3az standard, is designed to include
>>>>> legacy MAC without EEE capability into the power saving system. SmartEEE
>>>>> is enabled by default configuration on AR8031 after power-on or hardware
>>>>> reset.
>>>>>
>>>>> Unfortunately this is proved causing issues for certain hw
>>>>> configurations.  We add a quirk to optionally disable SmartEEE.
>>>>
>>>> It may be a good idea to detail what "certain hw configurations" are
>>>> and/or what the "issues" are, so it's easier to find this if/when
>>>> others encounter the same issue.
>>>
>>> This fix was upstreamed from the imx downstream kernel so I do not have much informations on that but something is reported in the U-Boot fix for the same problem at [0] so I'll add more info in the next revision.
>>>
>>> Cheers,
>>>
>>> [0] https://patchwork.ozlabs.org/patch/1015783/
>>>
>> Thanks for the link. According to the description SmartEEE works only if
>> both link partners support it (hence are Atheros PHY's). I would assume
>> that this usually isn't the case, shouldn't the feature be an opt-in
>> therefore?
> 
> I'm not convinced about that - I've been using AtherOS SmartEEE with
> i.MX6 for quite some time (I have patches which allow me to read out
> the stats via ethtool, and configure it - they're specific to the
> FEC driver though.)  SmartEEE works just like EEE for me, and I
> notice no problems.
> 
> I have i.MX6 boards connected to Netgear GS116 and similar switches,
> and also Marvell DSA switches on other platforms.
> 
> This is why I said above about "more detail".  Then maybe folk would
> know what to look for to see whether there is a problem.
> 
OK, then the description of the referenced patch isn't fully correct
and SmartEEE and EEE are partially compatible, and the problem is
just that we don't know exactly to which extent.

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel