# Re: \[PATCH v3 net-next 10/12\] net: bonding: ensure .ndo\_get\_stats64 can sleep

- From: Saeed Mahameed \<saeed@kernel.org\>
- Date: 2021-01-08 19:39:57 
- List: netdev
- Page: https://ratatoskr.run/netdev/2021/01/11192179
- Thread: https://ratatoskr.run/netdev/2021/01/11191857/t.md

```text
On Fri, 2021-01-08 at 10:27 +0100, Eric Dumazet wrote:
> On Fri, Jan 8, 2021 at 10:21 AM Vladimir Oltean [ref]
> wrote:
> > On Fri, Jan 08, 2021 at 10:14:01AM +0100, Eric Dumazet wrote:
> > > If you disagree, repost a rebased patch series so that we can
> > > test/compare and choose the best solution.
> > 
> > I would rather use Saeed's time as a reviewer to my existing and
> > current
> > patch set.
> 
> Yes, same feeling here, but Saeed brought back his own old
> implementation, so maybe he does not feel the same way ?

Agreed, Vladimir's work is more complete than mine, my work was just a
RFC to show the concept, it was far from being ready or testable with
some use cases such as bonding/bridge/ovs/etc...

Let me take a look at the current series, and if I see that the
rcu/dev_hold approach is more lightweight then i will suggest it to
Vladimir and he can make the final decision.






```