On Thu, Aug 23, 2018 at 05:57:06PM -0500, Grygorii Strashko wrote:
> Hi
>
> On 07/31/2018 05:52 PM, Frederic Weisbecker wrote:
> > Before updating the full nohz tick or the idle time on IRQ exit, we
> > check first if we are not in a nesting interrupt, whether the inner
> > interrupt is a hard or a soft IRQ.
> >
> > There is a historical reason for that: the dyntick idle mode used to
> > reprogram the tick on IRQ exit, after softirq processing, and there was
> > no point in doing that job in the outer nesting interrupt because the
> > tick update will be performed through the end of the inner interrupt
> > eventually, with even potential new timer updates.
> >
> > One corner case could show up though: if an idle tick interrupts a softirq
> > executing inline in the idle loop (through a call to local_bh_enable())
> > after we entered in dynticks mode, the IRQ won't reprogram the tick
> > because it assumes the softirq executes on an inner IRQ-tail. As a
> > result we might put the CPU in sleep mode with the tick completely
> > stopped whereas a timer can still be enqueued. Indeed there is no tick
> > reprogramming in local_bh_enable(). We probably asssumed there was no bh
> > disabled section in idle, although there didn't seem to be debug code
> > ensuring that.
> >
> > Nowadays the nesting interrupt optimization still stands but only concern
> > full dynticks. The tick is stopped on IRQ exit in full dynticks mode
> > and we want to wait for the end of the inner IRQ to reprogramm the tick.
> > But in_interrupt() doesn't make a difference between softirqs executing
> > on IRQ tail and those executing inline. What was to be considered a
> > corner case in dynticks-idle mode now becomes a serious opportunity for
> > a bug in full dynticks mode: if a tick interrupts a task executing
> > softirq inline, the tick reprogramming will be ignored and we may exit
> > to userspace after local_bh_enable() with an enqueued timer that will
> > never fire.
> >
> > To fix this, simply keep reprogramming the tick if we are in a hardirq
> > interrupting softirq. We can still figure out a way later to restore
> > this optimization while excluding inline softirq processing.
> >
> > Reported-by: Anna-Maria Gleixner <anna-maria@xxxxxxxxxxxxx>
> > Signed-off-by: Frederic Weisbecker <frederic@xxxxxxxxxx>
> > Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> > Cc: Ingo Molnar <mingo@xxxxxxxxxx>
> > Tested-by: Anna-Maria Gleixner <anna-maria@xxxxxxxxxxxxx>
> > ---
> > kernel/softirq.c | 2 +-
> > 1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/kernel/softirq.c b/kernel/softirq.c
> > index 900dcfe..0980a81 100644
> > --- a/kernel/softirq.c
> > +++ b/kernel/softirq.c
> > @@ -386,7 +386,7 @@ static inline void tick_irq_exit(void)
> >
> > /* Make sure that timer wheel updates are propagated */
> > if ((idle_cpu(cpu) && !need_resched()) || tick_nohz_full_cpu(cpu)) {
> > - if (!in_interrupt())
> > + if (!in_irq())
> > tick_nohz_irq_exit();
> > }
> > #endif
> >
>
> This patch was back ported to the Stable linux-4.14.y and It causes regression -
> flood of "NOHZ: local_softirq_pending" messages on all TI boards during boot (NFS boot):
>
> [ 4.179796] NOHZ: local_softirq_pending 2c2 in sirq 256
> [ 4.185051] NOHZ: local_softirq_pending 2c2 in sirq 256
>
> the same is not reproducible with LKML - seems due to changes in tick-sched.c
> __tick_nohz_idle_enter()/tick_nohz_irq_exit().
What changes do you think fixed this?
> I've generated backtrace from can_stop_idle_tick() (see below) and seems this
> patch makes tick_nohz_irq_exit() call unconditional in case of nested interrupt:
>
> gic_handle_irq
> |- irq_exit
> |- preempt_count_sub(HARDIRQ_OFFSET); <-- [1]
> |-__do_softirq
> <irqs enabled>
> |- gic_handle_irq()
> |- irq_exit()
> |- tick_irq_exit()
> if (!in_irq()) <-- My understanding is that this condition will be always true due to [1]
> tick_nohz_irq_exit();
> |-__tick_nohz_idle_enter()
> |- can_stop_idle_tick()
>
> Sry, not sure if my conclusion is right and how can it be fixed.
Any pointers to a patch that might need to be backported would be
appreciated.
thanks,
greg k-h
