Messages in this thread | | | Date | Sat, 10 Oct 2015 20:52:55 +0200 | From | Oleg Nesterov <> | Subject | [PATCH 0/3] (Was: sched: start stopper early) |
| |
To avoid the confusion, this has nothing to do with "stop_machine" changes we discuss in another thread, but
On 10/09, Oleg Nesterov wrote: > > > case CPU_ONLINE: > > + stop_machine_unpark(cpu); > > /* > > * At this point a starting CPU has marked itself as online via > > * set_cpu_online(). But it might not yet have marked itself > > @@ -5337,7 +5340,7 @@ static int sched_cpu_active(struct notifier_block *nfb, > > * Thus, fall-through and help the starting CPU along. > > */ > > case CPU_DOWN_FAILED: > > - set_cpu_active((long)hcpu, true); > > + set_cpu_active(cpu, true); > > On a second thought, we can't do this (and your initial change has > the same problem). > > We can not wakeup it before set_cpu_active(). This can lead to the > same problem fixed by dd9d3843755da95f6 "sched: Fix cpu_active_mask/ > cpu_online_mask race".
OTOH, I don't understand why do we actually need this fix... Or, iow I don't really understand the cpu_active() checks in select_fallback_rq().
Looks like we have some strange arch/ which has the "unsafe" online && !active window, but then it is not clear why it is safe to mark it active in sched_cpu_active(CPU_ONLINE). Confused.
And I am even more confused by the fact that select_fallback_rq() checks cpu_active(), but select_task_rq() doesn't. This can't be right in any case.
Oleg.
kernel/sched/core.c | 41 +++++++++++++++++++++-------------------- 1 files changed, 21 insertions(+), 20 deletions(-)
| |