On 7/3/2025 9:05 PM, Uladzislau Rezki wrote:
> On Wed, Jul 02, 2025 at 04:59:36PM +0200, Uladzislau Rezki (Sony) wrote:
>> Automatically enable the rcu_normal_wake_from_gp parameter on
>> systems with a small number of CPUs. The activation threshold
>> is set to 16 CPUs.
>>
>> This helps to reduce a latency of normal synchronize_rcu() API
>> by waking up GP-waiters earlier and decoupling synchronize_rcu()
>> callers from regular callback handling.
>>
>> A benchmark running 64 parallel jobs(system with 64 CPUs) invoking
>> synchronize_rcu() demonstrates a notable latency reduction with the
>> setting enabled.
>>
>> Latency distribution (microseconds):
>>
>> <default>
>> 0 - 9999 : 1
>> 10000 - 19999 : 4
>> 20000 - 29999 : 399
>> 30000 - 39999 : 3197
>> 40000 - 49999 : 10428
>> 50000 - 59999 : 17363
>> 60000 - 69999 : 15529
>> 70000 - 79999 : 9287
>> 80000 - 89999 : 4249
>> 90000 - 99999 : 1915
>> 100000 - 109999 : 922
>> 110000 - 119999 : 390
>> 120000 - 129999 : 187
>> ...
>> <default>
>>
>> <rcu_normal_wake_from_gp>
>> 0 - 9999 : 1
>> 10000 - 19999 : 234
>> 20000 - 29999 : 6678
>> 30000 - 39999 : 33463
>> 40000 - 49999 : 20669
>> 50000 - 59999 : 2766
>> 60000 - 69999 : 183
>> ...
>> <rcu_normal_wake_from_gp>
>>
>> Reviewed-by: Joel Fernandes <[email protected]>
>> Signed-off-by: Uladzislau Rezki (Sony) <[email protected]>
>> ---
>> kernel/rcu/tree.c | 14 +++++++++++---
>> 1 file changed, 11 insertions(+), 3 deletions(-)
>>
>> diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
>> index e8a4b720d7d2..b88ceb35cebd 100644
>> --- a/kernel/rcu/tree.c
>> +++ b/kernel/rcu/tree.c
>> @@ -1625,8 +1625,10 @@ static void rcu_sr_put_wait_head(struct llist_node
>> *node)
>> atomic_set_release(&sr_wn->inuse, 0);
>> }
>>
>> -/* Disabled by default. */
>> -static int rcu_normal_wake_from_gp;
>> +/* Enable rcu_normal_wake_from_gp automatically on small systems. */
>> +#define WAKE_FROM_GP_CPU_THRESHOLD 16
>> +
>> +static int rcu_normal_wake_from_gp = -1;
>> module_param(rcu_normal_wake_from_gp, int, 0644);
>> static struct workqueue_struct *sync_wq;
>>
>> @@ -3239,7 +3241,7 @@ static void synchronize_rcu_normal(void)
>>
>> trace_rcu_sr_normal(rcu_state.name, &rs.head, TPS("request"));
>>
>> - if (!READ_ONCE(rcu_normal_wake_from_gp)) {
>> + if (READ_ONCE(rcu_normal_wake_from_gp) < 1) {
>> wait_rcu_gp(call_rcu_hurry);
>> goto trace_complete_out;
>> }
>> @@ -4843,6 +4845,12 @@ void __init rcu_init(void)
>> sync_wq = alloc_workqueue("sync_wq", WQ_MEM_RECLAIM, 0);
>> WARN_ON(!sync_wq);
>>
>> + /* Respect if explicitly disabled via a boot parameter. */
>> + if (rcu_normal_wake_from_gp < 0) {
>> + if (num_possible_cpus() <= WAKE_FROM_GP_CPU_THRESHOLD)
>> + rcu_normal_wake_from_gp = 1;
>> + }
>> +
>> /* Fill in default value for rcutree.qovld boot parameter. */
>> /* -After- the rcu_node ->lock fields are initialized! */
>> if (qovld < 0)
>> --
>> 2.39.5
>>
> Neeraj, are you planning to take this for next merge window?
>
Yes, I have queued v2 of these for testing here:
https://git.kernel.org/pub/scm/linux/kernel/git/neeraj.upadhyay/linux-rcu.git/log/?h=dev.01.07.2025
Will include them in PR for next merge window.
- Neeraj
> --
> Uladzislau Rezlo
