Re: [PATCH v2 2/2] net-sysfs: Ensure begin/complete are called in speed_show() and duplex_show()

[Andy Shevchenko <andriy.shevchenko@linux.intel.com>]

On Wed, Feb 05, 2020 at 10:23:45AM +0100, Michal Kubecek wrote:
> On Wed, Feb 05, 2020 at 11:06:38AM +0200, Andy Shevchenko wrote:
> > On Wed, Feb 05, 2020 at 04:16:16PM +0800, Kai-Heng Feng wrote:
> > > Device like igb gets runtime suspended when there's no link partner. We
> > > can't get correct speed under that state:
> > > $ cat /sys/class/net/enp3s0/speed
> > > 1000
> > > 
> > > In addition to that, an error can also be spotted in dmesg:
> > > [  385.991957] igb 0000:03:00.0 enp3s0: PCIe link lost
> > > 
> > > It's because the igb device doesn't get runtime resumed before calling
> > > get_link_ksettings().
> > > 
> > > So let's use a new helper to call begin() and complete() like what
> > > dev_ethtool() does, to runtime resume/suspend or power up/down the
> > > device properly.
> > > 
> > > Once this fix is in place, igb can show the speed correctly without link
> > > partner:
> > > $ cat /sys/class/net/enp3s0/speed
> > > -1
> > 
> > What is the meaning of -1? Does it tells us "Hey, something is bad in hardware
> > I can't tell you the speed" or does it imply anything else?
> 
> It's SPEED_UNKNOWN constant printed with "%d" template.
> 
> > Wouldn't be better to report 0?
> > 
> > Where is the documentation part of this ABI change?
> 
> It's not an ABI change, /sys/class/net/*/speed already shows -1 when the
> device reports SPEED_UNKNOWN. The only change is that after this patch,
> igb driver reports SPEED_UNKNOWN rather than an outdated value if there
> is no link.

Thanks for elaboration.
Perhaps add a couple of words to the commit message that there is no ABI
changes.

-- 
With Best Regards,
Andy Shevchenko