Sun, Oct 29, 2017 at 12:17:14PM CET, j...@resnulli.us wrote: >Sun, Oct 29, 2017 at 10:36:32AM CET, da...@davemloft.net wrote: >>From: Jiri Pirko <j...@resnulli.us> >>Date: Wed, 25 Oct 2017 16:34:56 +0200 >> >>> This patchset does a bit of cleanup of leftovers after block callbacks >>> patchset. The main part is patch 2, which restores the original handling >>> of tc offload feature flag. >> >>Jiri, this series does not apply cleanly to net-next. > >I will respin. > > >> >>Also, you should really work out the block offload semantics >>with Jakub, continuing the discussion of patch #2. > >I have some ideas. I could let driver to enable/disable already >registered callback. Will look at it.
The cb enable disable according to NETIF_F_HW_TC ethtool flag could be done. It will require some additional work in many drivers. Most of the drivers do not effect the flag change. I will rebase and repost this patchset for now to fix the regression and work on NETIF_F_HW_TC handling in a follow-up.
