From: Ben Hutchings <redacted>
Date: Fri, 17 Jun 2011 04:51:35 +0100
On Thu, 2011-06-16 at 23:10 -0400, David Miller wrote:
quoted
From: Stephen Hemminger <redacted>
Date: Tue, 14 Jun 2011 00:02:30 -0400
quoted
In my experience if phy reads once successfully, it is going
to read every time. If there is a problem it only happens on
the first access (powered off, bad timing, etc).
It also happens when the PHY can't get a response for a certain
register, for whatever reason, before internal hw timeouts trigger.
Please, check all MII accesses. That's what I do in every driver
I've written.
It doesn't help that the mii_if_info operations are defined to never
return errors. This doesn't prevent drivers from doing so internally,
but it does set a bad example.
I totally agree.