Re: [patch V2 2/9] irqtime: Make accounting correct on RT

[Thomas Gleixner]

On Mon, Dec 07 2020 at 14:27, Frederic Weisbecker wrote:
> On Fri, Dec 04, 2020 at 06:01:53PM +0100, Thomas Gleixner wrote:
>> vtime_account_irq and irqtime_account_irq() base checks on preempt_count()
>> which fails on RT because preempt_count() does not contain the softirq
>> accounting which is seperate on RT.
>> 
>> These checks do not need the full preempt count as they only operate on the
>> hard and softirq sections.
>> 
>> Use irq_count() instead which provides the correct value on both RT and non
>> RT kernels. The compiler is clever enough to fold the masking for !RT:
>> 
>>        99b:	65 8b 05 00 00 00 00 	mov    %gs:0x0(%rip),%eax
>>  -     9a2:	25 ff ff ff 7f       	and    $0x7fffffff,%eax
>>  +     9a2:	25 00 ff ff 00       	and    $0xffff00,%eax
>> 
>> Reported-by: Sebastian Andrzej Siewior <bigeasy@xxxxxxxxxxxxx>
>> Signed-off-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
>
> Reviewed-by: Frederic Weisbecker <frederic@xxxxxxxxxx>
>
> Also I'm seeing a few other explicit SOFTIRQ_MASK checks on top
> of preempt_count(), especially on RCU:
>
>    $ git grep SOFTIRQ_MASK
>    arch/sh/kernel/irq.c:                   (irqctx->tinfo.preempt_count & ~SOFTIRQ_MASK) |
>    arch/sh/kernel/irq.c:                   (curctx->tinfo.preempt_count & SOFTIRQ_MASK);
>    kernel/rcu/rcutorture.c:                if (preempt_count() & (SOFTIRQ_MASK | HARDIRQ_MASK))
>    kernel/rcu/tree_exp.h:          if (!(preempt_count() & (PREEMPT_MASK | SOFTIRQ_MASK)) ||
>    kernel/rcu/tree_plugin.h:                       !!(preempt_count() & (PREEMPT_MASK | SOFTIRQ_MASK));
>    kernel/rcu/tree_plugin.h:           (preempt_count() &
>    (PREEMPT_MASK | SOFTIRQ_MASK))) {
>
> I guess some RT RCU config take care of that?

I ignored sh so far and RCU has some quirks for RT vs. softirqs.

> Also tracing:
>
>      kernel/trace/ring_buffer.c:     if (!(pc & (NMI_MASK | HARDIRQ_MASK | SOFTIRQ_OFFSET)))
>      kernel/trace/trace.c:           ((pc & SOFTIRQ_OFFSET) ? TRACE_FLAG_SOFTIRQ : 0)

Bah, I somehow lost the cure for that while rebasing.

Thanks,

        tglx