Re: [PATCH net v1] mlxbf-gige: Support workaround for MDIO GPIO degradation bug

[Andrew Lunn <andrew@lunn.ch>]

> Hi Andrew, 
> 

> I implemented and tested the changes you suggested and the
> runtime_resume/suspend work smoothly for MDIO.

> However, we have another issue. I noticed that even if
> mdio_read/write() functions are not being called,
> runtime_resume/suspend() are still called regularly. After
> investigation, I found out that this is due to ethtool being called
> regularly. Ethtool automatically triggers the resume/suspend even if
> we do no MDIO access. A different team wrote a script which monitors
> "ethtool -S eth0" every 60 seconds. So every minute, we are running
> resume/suspend and enabling/disabling the MDIO clock. Seems counter
> productive. That team said that it is a requirement that they
> collect these statistics about the mlxbf_gige interface.

> Is there any way to prevent ethtool from calling resume/suspend
> without changing core kernel code?

> If not, what would you suggest? 

You need to put the MDIO bus device into its own pm_domain. Try
calling dev_pm_domain_set() to separate the MDIO bus from the MAC
driver in terms of power domains. ethtool will then power on/off the
MAC but leave the MDIO bus alone.

	Andrew