Re: [PATCH tip/core/rcu 2/4] smp: Add source and destination CPUs to __call_single_data

[Paul E. McKenney]

On Fri, Sep 04, 2020 at 10:16:10AM +0800, Boqun Feng wrote:
> Hi Paul,
> 
> On Mon, Aug 31, 2020 at 11:14:15AM -0700, paulmck@xxxxxxxxxx wrote:
> > From: "Paul E. McKenney" <paulmck@xxxxxxxxxx>
> > 
> > This commit adds a destination CPU to __call_single_data, and is inspired
> > by an earlier commit by Peter Zijlstra.  This version adds #ifdef to
> > permit use by 32-bit systems and supplying the destination CPU for all
> > smp_call_function*() requests, not just smp_call_function_single().
> > 
> > If need be, 32-bit systems could be accommodated by shrinking the flags
> > field to 16 bits (the atomic_t variant is currently unused) and by
> > providing only eight bits for CPU on such systems.
> > 
> > It is not clear that the addition of the fields to __call_single_node
> > are really needed.
> > 
> > Link: https://lore.kernel.org/lkml/20200615164048.GC2531@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx/
> > Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> > Cc: Ingo Molnar <mingo@xxxxxxxxxx>
> > Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> > Cc: Sebastian Andrzej Siewior <bigeasy@xxxxxxxxxxxxx>
> > Cc: Frederic Weisbecker <frederic@xxxxxxxxxx>
> > Signed-off-by: Paul E. McKenney <paulmck@xxxxxxxxxx>
> > ---
> >  include/linux/smp.h       | 3 +++
> >  include/linux/smp_types.h | 3 +++
> >  kernel/smp.c              | 2 ++
> >  3 files changed, 8 insertions(+)
> > 
> > diff --git a/include/linux/smp.h b/include/linux/smp.h
> > index 80d557e..9f13966 100644
> > --- a/include/linux/smp.h
> > +++ b/include/linux/smp.h
> > @@ -26,6 +26,9 @@ struct __call_single_data {
> >  		struct {
> >  			struct llist_node llist;
> >  			unsigned int flags;
> > +#ifdef CONFIG_64BIT
> > +			u16 src, dst;
> > +#endif
> >  		};
> >  	};
> >  	smp_call_func_t func;
> > diff --git a/include/linux/smp_types.h b/include/linux/smp_types.h
> > index 364b3ae..2e8461a 100644
> > --- a/include/linux/smp_types.h
> > +++ b/include/linux/smp_types.h
> > @@ -61,6 +61,9 @@ struct __call_single_node {
> >  		unsigned int	u_flags;
> >  		atomic_t	a_flags;
> >  	};
> > +#ifdef CONFIG_64BIT
> > +	u16 src, dst;
> > +#endif
> >  };
> >  
> >  #endif /* __LINUX_SMP_TYPES_H */
> > diff --git a/kernel/smp.c b/kernel/smp.c
> > index d0ae8eb..a47382d 100644
> > --- a/kernel/smp.c
> > +++ b/kernel/smp.c
> > @@ -375,6 +375,7 @@ int smp_call_function_single(int cpu, smp_call_func_t func, void *info,
> >  
> >  	csd->func = func;
> >  	csd->info = info;
> > +	csd->dst = cpu;
> 
> Would this and the next modification cause compile errors with
> CONFIG_64BIT = n? I saw you add #ifdef guard in the next patch, so maybe
> move those two into next patch (of course, if they trigger compile
> errors)

Good point, will fix, thank you!

							Thanx, Paul

> Regards,
> Boqun
> 
> >  
> >  	err = generic_exec_single(cpu, csd);
> >  
> > @@ -540,6 +541,7 @@ static void smp_call_function_many_cond(const struct cpumask *mask,
> >  			csd->flags |= CSD_TYPE_SYNC;
> >  		csd->func = func;
> >  		csd->info = info;
> > +		csd->dst = cpu;
> >  		if (llist_add(&csd->llist, &per_cpu(call_single_queue, cpu)))
> >  			__cpumask_set_cpu(cpu, cfd->cpumask_ipi);
> >  	}
> > -- 
> > 2.9.5
> >