Re: [PATCH v3] rcu: Reduce synchronize_rcu() latency by reporting GP kthread's CPU QS early

[Boqun Feng <boqun.feng@gmail.com>]

On Sun, Jan 11, 2026 at 03:42:36PM +0530, Samir M wrote:
> 
> On 30/12/25 6:11 am, Joel Fernandes wrote:
> > The RCU grace period mechanism uses a two-phase FQS (Force Quiescent
> > State) design where the first FQS saves dyntick-idle snapshots and
> > the second FQS compares them. This results in long and unnecessary latency
> > for synchronize_rcu() on idle systems (two FQS waits of ~3ms each with
> > 1000HZ) whenever one FQS wait sufficed.
> > 
> > Some investigations showed that the GP kthread's CPU is the holdout CPU
> > a lot of times after the first FQS as - it cannot be detected as "idle"
> > because it's actively running the FQS scan in the GP kthread.
> > 
> > Therefore, at the end of rcu_gp_init(), immediately report a quiescent
> > state for the GP kthread's CPU using rcu_qs() + rcu_report_qs_rdp(). The
> > GP kthread cannot be in an RCU read-side critical section while running
> > GP initialization, so this is safe and results in significant latency
> > improvements.
> > 
> > I benchmarked 100 synchronize_rcu() calls with 32 CPUs, 10 runs each
> > showing significant latency improvements (default settings for fqs jiffies):
> > 
> > Baseline (without fix):
> > | Run | Mean      | Min      | Max       |
> > |-----|-----------|----------|-----------|
> > | 1   | 10.088 ms | 9.989 ms | 18.848 ms |
> > | 2   | 10.064 ms | 9.982 ms | 16.470 ms |
> > | 3   | 10.051 ms | 9.988 ms | 15.113 ms |
> > | 4   | 10.125 ms | 9.929 ms | 22.411 ms |
> > | 5   |  8.695 ms | 5.996 ms | 15.471 ms |
> > | 6   | 10.157 ms | 9.977 ms | 25.723 ms |
> > | 7   | 10.102 ms | 9.990 ms | 20.224 ms |
> > | 8   |  8.050 ms | 5.985 ms | 10.007 ms |
> > | 9   | 10.059 ms | 9.978 ms | 15.934 ms |
> > | 10  | 10.077 ms | 9.984 ms | 17.703 ms |
> > 
> > With fix:
> > | Run | Mean     | Min      | Max       |
> > |-----|----------|----------|-----------|
> > | 1   | 6.027 ms | 5.915 ms |  8.589 ms |
> > | 2   | 6.032 ms | 5.984 ms |  9.241 ms |
> > | 3   | 6.010 ms | 5.986 ms |  7.004 ms |
> > | 4   | 6.076 ms | 5.993 ms | 10.001 ms |
> > | 5   | 6.084 ms | 5.893 ms | 10.250 ms |
> > | 6   | 6.034 ms | 5.908 ms |  9.456 ms |
> > | 7   | 6.051 ms | 5.993 ms | 10.000 ms |
> > | 8   | 6.057 ms | 5.941 ms | 10.001 ms |
> > | 9   | 6.016 ms | 5.927 ms |  7.540 ms |
> > | 10  | 6.036 ms | 5.993 ms |  9.579 ms |
> > 
> > Summary:
> > - Mean latency: 9.75 ms -> 6.04 ms (38% improvement)
> > - Max latency:  25.72 ms -> 10.25 ms (60% improvement)
> > 
> > Additional bridge setup/teardown testing by Uladzislau Rezki on x86_64
> > with 64 CPUs (100 iterations of bridge add/configure/delete):
> > 
> >                                 real time
> > 1 - default:                   24.221s
> > 2 - this patch:                20.754s  (14% faster)
> > 3 - this patch + wake_from_gp: 15.895s  (34% faster)
> > 4 - wake_from_gp only:         18.947s  (22% faster)
> > 
> > Per-synchronize_rcu() latency (in usec):
> >                1         2         3       4
> > median: 37249.5   31540.5   15765   22480
> > min:    7881      7918      9803    7857
> > max:    63651     55639     31861   32040
> > 
> > This patch combined with rcu_normal_wake_from_gp reduces bridge
> > setup/teardown time from 24 seconds to 16 seconds.
> > 
> > Tested rcutorture TREE and SRCU configurations.
> > 
> > Reviewed-by: Paul E. McKenney <paulmck@kernel.org>
> > Tested-by: Uladzislau Rezki (Sony) <urezki@gmail.com>
> > Signed-off-by: Joel Fernandes <joelagnelf@nvidia.com>
> > ---
> >   kernel/rcu/tree.c | 12 ++++++++++++
> >   1 file changed, 12 insertions(+)
> > 
> > diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
> > index 78c045a5ef03..b7c818cabe44 100644
> > --- a/kernel/rcu/tree.c
> > +++ b/kernel/rcu/tree.c
> > @@ -160,6 +160,7 @@ static void rcu_report_qs_rnp(unsigned long mask, struct rcu_node *rnp,
> >   			      unsigned long gps, unsigned long flags);
> >   static void invoke_rcu_core(void);
> >   static void rcu_report_exp_rdp(struct rcu_data *rdp);
> > +static void rcu_report_qs_rdp(struct rcu_data *rdp);
> >   static void check_cb_ovld_locked(struct rcu_data *rdp, struct rcu_node *rnp);
> >   static bool rcu_rdp_is_offloaded(struct rcu_data *rdp);
> >   static bool rcu_rdp_cpu_online(struct rcu_data *rdp);
> > @@ -1983,6 +1984,17 @@ static noinline_for_stack bool rcu_gp_init(void)
> >   	if (IS_ENABLED(CONFIG_RCU_STRICT_GRACE_PERIOD))
> >   		on_each_cpu(rcu_strict_gp_boundary, NULL, 0);
> > +	/*
> > +	 * Immediately report QS for the GP kthread's CPU. The GP kthread
> > +	 * cannot be in an RCU read-side critical section while running
> > +	 * the FQS scan. This eliminates the need for a second FQS wait
> > +	 * when all CPUs are idle.
> > +	 */
> > +	preempt_disable();
> > +	rcu_qs();
> > +	rcu_report_qs_rdp(this_cpu_ptr(&rcu_data));
> > +	preempt_enable();
> > +
> >   	return true;
> >   }
> 
> Hi,
> 
> 
> I verified this patch on ppc64 systems and observed consistent performance
> improvements.
> 
> The testing was conducted on Power LPARs using 20 cores (160 CPUs) with SMT
> enabled and disabled. All tests were performed on the latest upstream kernel
> (v6.19.0-rc3+), and the patch showed measurable improvements in both SMT
> configurations.
> 
> SMT Mode   |      With Patch (s)    |    Without Patch (s)  | Improvement
> (%)
> ————————————————————————————————————
> SMT ON.      |    51.662               |      75.540  |       31.61% faster
> SMT OFF      |    44.246               |      59.933  |       26.18% faster
> 
> Please add below tag: 
Tested-by: Samir M <samir@linux.ibm.com>
> 

Applied, thanks!

Regards,
Boqun

> Thanks,
> Samir
> 
>