On Tue, 16 Jun 2020 at 04:00, David Miller <da...@davemloft.net> wrote: > > From: Vladimir Oltean <olte...@gmail.com> > Date: Tue, 16 Jun 2020 00:55:50 +0300 > > > On Mon, 15 Jun 2020 at 22:33, Davide Caratti <dcara...@redhat.com> wrote: > >> > >> assigning a dummy value of 'clock_id' to avoid cancellation of the cycle > >> timer before its initialization was a temporary solution, and we still > >> need to handle the case where act_gate timer parameters are changed by > >> commands like the following one: > >> > >> # tc action replace action gate <parameters> > >> > >> the fix consists in the following items: > >> > >> 1) remove the workaround assignment of 'clock_id', and init the list of > >> entries before the first error path after IDR atomic check/allocation > >> 2) validate 'clock_id' earlier: there is no need to do IDR atomic > >> check/allocation if we know that 'clock_id' is a bad value > >> 3) use a dedicated function, 'gate_setup_timer()', to ensure that the > >> timer is cancelled and re-initialized on action overwrite, and also > >> ensure we initialize the timer in the error path of tcf_gate_init() > >> > >> v2: avoid 'goto' in gate_setup_timer (thanks to Cong Wang) > >> > > > > The change log is put under the 3 '---' characters for a reason: it is > > relevant only to reviewers, and git automatically trims it when > > applying the patch. The way it is now, the commit message would > > contain this line about "v2 ...". > > I completely disagree and I ask submitters of networking changes to keep > the changelog in the commit message. > > Later people will look at this commit and ask "why didn't they do X > or Y" and if the changelog shows that the submitter was asked not to > do X or Y that is useful information.
Interesting, I didn't know that. If there are really relevant changes which might matter post-review (which I don't consider "avoid 'goto' in gate_setup_timer" to be), I usually try to integrate them better into the main commit message. Nonetheless, sorry, feel free to ignore me! -Vladimir
