Messages in this thread | | | Date | Sun, 5 Apr 2015 07:05:37 +0200 | From | Ingo Molnar <> | Subject | Re: [PATCH 2/2] nohz: make nohz_full imply isolcpus |
| |
* Chris Metcalf <cmetcalf@ezchip.com> wrote:
> On 4/4/2015 10:10 AM, Rik van Riel wrote: > >>Rik, what's the change you're proposing that's similar to this one? Thanks! > >I don't have this particular one, and I like it. > > > >I know there are use cases where isolcpus= without > >nohz_full= makes sense, but I cannot think of the > >reverse. > > > >Acked-by: Rik van Riel<riel@redhat.com> > > Thanks, I'll push it via the tile tree unless someone would prefer otherwise.
Yes, I'd prefer otherwise: please send the final, agreed upon patch to the timer tree.
> (The tick_nohz_full_set_cpus() and tick_nohz_full_clear_cpus() > routines are in earlier tile tree commits, the latter supporting a > change to the tile network driver.)
This is absolutely not OK, please push this through the timer tree. We don't do generic timer changes through architecture trees.
Thanks,
Ingo
| |