Messages in this thread | | | Date | Fri, 3 Apr 2015 15:20:20 -0400 | From | Chris Metcalf <> | Subject | Re: [PATCH 2/2] nohz: make nohz_full imply isolcpus |
| |
On 04/03/2015 01:42 PM, Frederic Weisbecker wrote: > On Fri, Apr 03, 2015 at 12:24:08PM -0400, cmetcalf@ezchip.com wrote: >> From: Chris Metcalf <cmetcalf@ezchip.com> >> >> It's not clear that nohz_full is useful without isolcpus also >> set, since otherwise the scheduler has to run periodically to >> try to determine whether to steal work from other cores. >> >> Signed-off-by: Chris Metcalf <cmetcalf@ezchip.com> > I think Rick has a similar patch.
I didn't see anything relevant in linux-next, though I did see cpu_isolated_map made into a public symbol in a recent commit by Rik.
Rik, what's the change you're proposing that's similar to this one? Thanks!
>> --- >> I am puzzled why this has not been done before, so I suspect >> there is some argument against it that I am missing, but I >> wasn't able to turn anything up by searching LKML. >> >> kernel/sched/core.c | 1 + >> 1 file changed, 1 insertion(+) >> >> diff --git a/kernel/sched/core.c b/kernel/sched/core.c >> index f0f831e8a345..275f12c608f2 100644 >> --- a/kernel/sched/core.c >> +++ b/kernel/sched/core.c >> @@ -6836,6 +6836,7 @@ static int init_sched_domains(const struct cpumask *cpu_map) >> doms_cur = alloc_sched_domains(ndoms_cur); >> if (!doms_cur) >> doms_cur = &fallback_doms; >> + tick_nohz_full_set_cpus(cpu_isolated_map); >> cpumask_andnot(doms_cur[0], cpu_map, cpu_isolated_map); >> err = build_sched_domains(doms_cur[0], NULL); >> register_sched_domain_sysctl(); >> -- >> 2.1.2 >>
-- Chris Metcalf, EZChip Semiconductor http://www.ezchip.com
| |