Re: [RFC PATCH] sched: Add missing memory barrier in switch_mm_cid

[Mathieu Desnoyers]

On 2024-03-11 04:45, Yeo Reum Yun wrote:
> Hi. Mathieu. Sorry to late answer.
>
> > diff --git a/arch/x86/include/asm/barrier.h b/arch/x86/include/asm/barrier.h
> > index 35389b2af88e..0d5e54201eb2 100644
> > --- a/arch/x86/include/asm/barrier.h
> > +++ b/arch/x86/include/asm/barrier.h
> > @@ -79,6 +79,9 @@ do {                                                                  \
> >   #define __smp_mb__before_atomic()      do { } while (0)
> >   #define __smp_mb__after_atomic()       do { } while (0)
>
> > +/* Writing to CR3 provides a full memory barrier in switch_mm(). */
> > +#define smp_mb__after_switch_mm()      do { } while (0)
> > +
> > #include <asm-generic/barrier.h>
>
> IIUC, ppc already does smp_mb() in switch_mm.
>
> Would it better to add the same macro which do nothing to pcc?\\

Does it ?

Based on arch/powerpc/include/asm/membarrier.h, it appears that
powerpc does _not_ have a guaranteed barrier in switch_mm():

static inline void membarrier_arch_switch_mm(struct mm_struct *prev,
                                             struct mm_struct *next,
                                             struct task_struct *tsk)
{
        /*
         * Only need the full barrier when switching between processes.
         * Barrier when switching from kernel to userspace is not
         * required here, given that it is implied by mmdrop(). Barrier
         * when switching from userspace to kernel is not needed after
         * store to rq->curr.
         */
        if (IS_ENABLED(CONFIG_SMP) &&
            likely(!(atomic_read(&next->membarrier_state) &
                     (MEMBARRIER_STATE_PRIVATE_EXPEDITED |
                      MEMBARRIER_STATE_GLOBAL_EXPEDITED)) || !prev))
                return;

        /*
         * The membarrier system call requires a full memory barrier
         * after storing to rq->curr, before going back to user-space.
         */
        smp_mb();
}

AFAIU the barrier provided in powerpc switch_mm_irqs_off() is only in the
"new_on_cpu" case. Am I missing something ?

Thanks,

Mathieu

> Thanks!
--
Mathieu Desnoyers
EfficiOS Inc.
https://www.efficios.com