On Tue, Nov 15, 2016 at 9:04 AM, Jiri Benc <jb...@redhat.com> wrote: > On Tue, 15 Nov 2016 08:40:58 -0800, Pravin Shelar wrote: >> On Tue, Nov 15, 2016 at 6:30 AM, Jiri Benc <jb...@redhat.com> wrote: >> > It would be a bit cleaner to do this assignment just after rt is >> > assigned (but after the IS_ERR(rt) condition), get rid of the added >> > ip_rt_put call above and move the existing ip_rt_put call in the bypass >> > case just before the vxlan_encap_bypass call... >> > >> Does it really matters given that next patches in this series moves >> this duplicate code and does pretty much what you are describing? > > Okay, right. I tried to look also at patches further in the series but > it seemed to me this will leave an instance of ip_rt_put that could be > avoided. But it will not. > > Acked-by: Jiri Benc <jb...@redhat.com> > > (It would make the reviewers' life easier if the individual patches > were more self contained. Ideally, each patch should be able to stand > on its own. This unrelated code shuffling makes it too easy to miss > things...) > I tried to keep one patch targeting single cleanup. But all patches targeted same code, that makes it dependent on each other.
> Anyway, thanks for the cleanup! > Thanks for all reviews.
