On Mon, 14 Sep 2020 09:25:44 -0700 Florian Fainelli wrote: > >> Can we consider using get_ethtool_stats and ETH_SS_PAUSE_STATS as a > >> stringset identifier? That way there is a single point within driver to > >> fetch stats. > > > > Can you say more? There are no strings reported in this patch set. > > What I am suggesting is that we have a central and unique method for > drivers to be called for all ethtool statisitcs to be obtained, and not > create another ethtool operation specifically for pause stats.
That won't work for statistics which correspond to a non-singleton object, like queue stats. > Today we have get_ethtool_stats and a stringset argument that tells you > which type of statistic to return. I am not suggesting that we return > strings or that it should be necessary for fetching pause stats. A multiplexer call or a call that dumps everything and then core picks out what it needs?
