sched/fair: Prefer fully idle cores for NOHZ balancing
PATCH 1/1 sched/fair: Prefer fully idle cores for NOHZ balancing
find_new_ilb() selects the first idle housekeeping CPU without considering whether another thread is running on the same physical core. On an SMT system, the idle load balancer can therefore activate both siblings even when another housekeeping CPU has an entirely idle core.
On most SMT systems, this is not problematic because the idle load balancer is a short-lived activity and the transient wakeup of a sibling has negligible performance impact.
However, this can be particularly costly on NVIDIA Olympus cores used in Vera. Briefly activating an otherwise idle sibling can reduce the performance available to the other sibling and this effect does not necessarily end once the activated sibling becomes idle: after the ILB finishes and its CPU enters WFI, full single-thread performance is restored only after the sibling has remained idle for a qualification interval (10 Ki cycles on the tested Vera system). Repeated short sibling wakeups can therefore sustain the interference even with little actual overlap.
Prevent this by preferring an idle housekeeping CPU whose entire SMT core is idle. Retain the first idle CPU as a fallback when no fully idle core is available, so NOHZ balancing continues to make forward progress.
Tests performed using an ad hoc GEMM benchmark running one CPU-intensive task per SMT core within its CPU affinity mask improved from approximately 6.2 TFLOP/s to 9.4 TFLOP/s.
Note that this preference may wake a fully idle physical core instead of using an idle sibling of an active core, potentially increasing ILB wakeup latency or energy consumption on some architectures. It may also scan additional CPUs before selecting the one to run the ILB. The selection falls back to the first idle CPU when no fully idle SMT core is available. Non-SMT systems continue to select the first idle housekeeping CPU.
--- kernel/sched/fair.c | 23 ++++++++++++++++++----- 1 file changed, 18 insertions(+), 5 deletions(-)
kernel/sched/fair.c +18 -5
@@ -13966,7 +13966,7 @@ static inline int find_new_ilb(void) {int this_cpu =smp_processor_id ();const struct cpumask *hk_mask; -int ilb_cpu; +int ilb_cpu, fallback =-1 ; hk_mask =housekeeping_cpumask (HK_TYPE_KERNEL_NOISE );
@@ -13974,19 +13974,32 @@ static inline int find_new_ilb(void)if (ilb_cpu == this_cpu)continue ; -if (idle_cpu (ilb_cpu)) +if (!idle_cpu (ilb_cpu)) +continue ; + +/* +* Running the idle load balancer on an idle sibling of a busy +* SMT core can reduce the capacity available to its sibling. Prefer +* a CPU whose entire core is idle, but retain the first idle CPU as +* a fallback so idle balancing can still make progress when no fully +* idle core exists. +*/ +if (!sched_smt_active () ||is_core_idle (ilb_cpu))return ilb_cpu; + +if (fallback <0 ) + fallback = ilb_cpu; } -return -1 ; +return fallback; }/* * Kick a CPU to do the NOHZ balancing, if it is time for it, via a cross-CPU * SMP function call (IPI). * -* We pick the first idle CPU in the HK_TYPE_KERNEL_NOISE housekeeping set -* (if there is one). +* Prefer a CPU on a fully idle core in the HK_TYPE_KERNEL_NOISE housekeeping +* set. Fall back to the first idle CPU when no fully idle core exists. */ static void kick_ilb (unsigned int flags) {
Signature · 1 line
2.55.0
K Prateek Nayak
reviewerHello Andrea,
Quoted 6 lines · Andrea Righi
(quoted diff) +1
-1@@ -13974,19 +13974,32 @@ static inline int find_new_ilb(void) if (ilb_cpu == this_cpu) continue; - if (idle_cpu(ilb_cpu)) + if (!idle_cpu(ilb_cpu))
Review static inline int find_new_ilb(void) · line 13987
Quoted 10 lines
(quoted diff) +10
-0+ continue; + + /* + * Running the idle load balancer on an idle sibling of a busy + * SMT core can reduce the capacity available to its sibling. Prefer + * a CPU whose entire core is idle, but retain the first idle CPU as + * a fallback so idle balancing can still make progress when no fully + * idle core exists. + */ + if (!sched_smt_active() || is_core_idle(ilb_cpu))
nit.
is_core_idle() here would iterate all siblings and on systems with SMT-4 and SMT-8, that overhead is apparently visible when one thread per core is occupied based on past optimizations like f8858d96061f ("sched/fair:
Optimize should_we_balance() for large SMT systems").
Copying the same approach from that optimization, can we do:
(Only build tested)
kernel/sched/fair.c +21 -7
@@ -13849,21 +13849,35 @@ static inline int on_null_domain(struct rq *rq)*/ static inline int find_new_ilb (void ) { +struct cpumask *ilb_cpus =this_cpu_cpumask_var_ptr (select_rq_mask);int this_cpu =smp_processor_id (); -const struct cpumask *hk_mask; -int ilb_cpu; +int ilb_cpu, fallback =-1 ; - hk_mask =housekeeping_cpumask (HK_TYPE_KERNEL_NOISE ); +cpumask_and (ilb_cpus, nohz.idle_cpus_mask,housekeeping_cpumask (HK_TYPE_KERNEL_NOISE )); -for_each_cpu_and (ilb_cpu, nohz.idle_cpus_mask, hk_mask) { +for_each_cpu (ilb_cpu, ilb_cpus) {if (ilb_cpu == this_cpu)continue ; -if (idle_cpu (ilb_cpu)) -return ilb_cpu; +if (!idle_cpu (ilb_cpu)) +continue ; + +if (sched_smt_active () && !is_core_idle (ilb_cpu)) { +if (fallback ==-1 ) + fallback = ilb_cpu; +/* +* If the core is not idle, and first SMT sibling which is +* idle has been found, then its not needed to check other +* SMT siblings for idleness: +*/ +cpumask_andnot (ilb_cpus, ilb_cpus,cpu_smt_mask (ilb_cpu)); +continue ; + } + +return ilb_cpu; } -return -1 ; +return fallback; }/*
---
It is safe to use "select_rq_mask" here since this is the tick handler trying to find an ilb_cpu and "select_rq_mask" is only used in contexts with IRQs disabled. It can probably be renamed to suggest that it is safe to be used in any IRQ disabled context as a temporary mask.
Thoughts?
Signature · 2 lines
Thanks and Regards, Prateek
Hi Prateek,
Quoted 22 lines · K Prateek Nayak
Hello Andrea,
Quoted 6 lines · Andrea Righi
kernel/sched/fair.c +1
-1@@ -13974,19 +13974,32 @@ static inline int find_new_ilb(void)if (ilb_cpu == this_cpu)continue ; -if (idle_cpu (ilb_cpu)) +if (!idle_cpu (ilb_cpu))kernel/sched/fair.c static inline int find_new_ilb(void) · line 13987
Quoted 10 lines
kernel/sched/fair.c +10
-0+continue ; + +/* +* Running the idle load balancer on an idle sibling of a busy +* SMT core can reduce the capacity available to its sibling. Prefer +* a CPU whose entire core is idle, but retain the first idle CPU as +* a fallback so idle balancing can still make progress when no fully +* idle core exists. +*/ +if (!sched_smt_active () ||is_core_idle (ilb_cpu))nit.
Quoted 6 lines
is_core_idle()here would iterate all siblings and on systems with SMT-4 and SMT-8, that overhead is apparently visible when one thread per core is occupied based on past optimizations likef8858d96061f("sched/fair:
Optimizeshould_we_balance()for large SMT systems").Copying the same approach from that optimization, can we do:
Good point. Without removing the remaining siblings, we may call is_core_idle() repeatedly for the same partially busy SMT core. I'll repeat my tests on my Vera system and incorporate your suggestion in v2 if I don't see any regression.
Quoted 52 lines
(Only build tested)
kernel/sched/fair.c +21
-8@@ -13849,21 +13849,35 @@ static inline int on_null_domain(struct rq *rq)*/ static inline int find_new_ilb (void ) { +struct cpumask *ilb_cpus =this_cpu_cpumask_var_ptr (select_rq_mask);int this_cpu =smp_processor_id (); -const struct cpumask *hk_mask; -int ilb_cpu; +int ilb_cpu, fallback =-1 ; - hk_mask =housekeeping_cpumask (HK_TYPE_KERNEL_NOISE ); +cpumask_and (ilb_cpus, nohz.idle_cpus_mask,housekeeping_cpumask (HK_TYPE_KERNEL_NOISE )); -for_each_cpu_and (ilb_cpu, nohz.idle_cpus_mask, hk_mask) { +for_each_cpu (ilb_cpu, ilb_cpus) {if (ilb_cpu == this_cpu)continue ; -if (idle_cpu (ilb_cpu)) -return ilb_cpu; +if (!idle_cpu (ilb_cpu)) +continue ; + +if (sched_smt_active () && !is_core_idle (ilb_cpu)) { +if (fallback ==-1 ) + fallback = ilb_cpu; +/* +* If the core is not idle, and first SMT sibling which is +* idle has been found, then its not needed to check other +* SMT siblings for idleness: +*/ +cpumask_andnot (ilb_cpus, ilb_cpus,cpu_smt_mask (ilb_cpu)); +continue ; + } + +return ilb_cpu; } -return -1 ; +return fallback; }/* ---
Quoted 6 lines
It is safe to use "
select_rq_mask" here since this is the tick handler trying to find anilb_cpuand "select_rq_mask" is only used in contexts with IRQs disabled. It can probably be renamed to suggest that it is safe to be used in any IRQ disabled context as a temporary mask.Thoughts?
Agreed. We can also add lockdep_assert_irqs_disabled() to find_new_ilb() to better document and verify the condition that makes reusing select_rq_mask safe.
Speaking of that, instead of renaming it, would it be better to provide a helper to access select_rq_mask with lockdep_assert_irqs_disabled()?
Thanks,
-Andrea
K Prateek Nayak
reviewerHello Andrea,
Quoted 30 lines · Andrea Righi
Hi Prateek,
Quoted 27 lines · K Prateek Nayak
Hello Andrea,
Quoted 6 lines · Andrea Righi
kernel/sched/fair.c +1
-1@@ -13974,19 +13974,32 @@ static inline int find_new_ilb(void)if (ilb_cpu == this_cpu)continue ; -if (idle_cpu (ilb_cpu)) +if (!idle_cpu (ilb_cpu))kernel/sched/fair.c static inline int find_new_ilb(void) · line 13987
Quoted 10 lines
kernel/sched/fair.c +10
-0+continue ; + +/* +* Running the idle load balancer on an idle sibling of a busy +* SMT core can reduce the capacity available to its sibling. Prefer +* a CPU whose entire core is idle, but retain the first idle CPU as +* a fallback so idle balancing can still make progress when no fully +* idle core exists. +*/ +if (!sched_smt_active () ||is_core_idle (ilb_cpu))nit.
is_core_idle()here would iterate all siblings and on systems with SMT-4 and SMT-8, that overhead is apparently visible when one thread per core is occupied based on past optimizations likef8858d96061f("sched/fair:
Optimizeshould_we_balance()for large SMT systems").
Quoted 5 lines
Quoted 1 line
Copying the same approach from that optimization, can we do:
Good point. Without removing the remaining siblings, we may call
is_core_idle()repeatedly for the same partially busy SMT core. I'll repeat my tests on my Vera system and incorporate your suggestion in v2 if I don't see any regression.
Thank you! From my experience, it is not very visible on SMT-2 but Shrikanth can vouch for the overheads on SMT-4, SMT-8 systems.
Quoted 58 lines
Quoted 58 lines
(Only build tested)
kernel/sched/fair.c +21
-8@@ -13849,21 +13849,35 @@ static inline int on_null_domain(struct rq *rq)*/ static inline int find_new_ilb (void ) { +struct cpumask *ilb_cpus =this_cpu_cpumask_var_ptr (select_rq_mask);int this_cpu =smp_processor_id (); -const struct cpumask *hk_mask; -int ilb_cpu; +int ilb_cpu, fallback =-1 ; - hk_mask =housekeeping_cpumask (HK_TYPE_KERNEL_NOISE ); +cpumask_and (ilb_cpus, nohz.idle_cpus_mask,housekeeping_cpumask (HK_TYPE_KERNEL_NOISE )); -for_each_cpu_and (ilb_cpu, nohz.idle_cpus_mask, hk_mask) { +for_each_cpu (ilb_cpu, ilb_cpus) {if (ilb_cpu == this_cpu)continue ; -if (idle_cpu (ilb_cpu)) -return ilb_cpu; +if (!idle_cpu (ilb_cpu)) +continue ; + +if (sched_smt_active () && !is_core_idle (ilb_cpu)) { +if (fallback ==-1 ) + fallback = ilb_cpu; +/* +* If the core is not idle, and first SMT sibling which is +* idle has been found, then its not needed to check other +* SMT siblings for idleness: +*/ +cpumask_andnot (ilb_cpus, ilb_cpus,cpu_smt_mask (ilb_cpu)); +continue ; + } + +return ilb_cpu; } -return -1 ; +return fallback; }/* --- It is safe to use "
select_rq_mask" here since this is the tick handler trying to find anilb_cpuand "select_rq_mask" is only used in contexts with IRQs disabled. It can probably be renamed to suggest that it is safe to be used in any IRQ disabled context as a temporary mask.
Quoted 4 lines
Quoted 1 line
Thoughts?
Agreed. We can also add
lockdep_assert_irqs_disabled()tofind_new_ilb()to better document and verify the condition that makes reusingselect_rq_masksafe.
That works too but it just looks a bot odd to have the selectrq_mask in a load balancing function.
Quoted 3 lines
Speaking of that, instead of renaming it, would it be better to provide a helper to access
select_rq_maskwithlockdep_assert_irqs_disabled()?
I'll defer to Peter on that :-)
He had previously suggested renaming it when there were discussions to reuse it here
https://lore.kernel.org/lkml/[email protected]/
Signature · 2 lines
Thanks and Regards, Prateek
Quoted 33 lines · K Prateek Nayak
Hello Andrea,
Quoted 30 lines · Andrea Righi
Hi Prateek,
Quoted 27 lines · K Prateek Nayak
Hello Andrea,
Quoted 6 lines · Andrea Righi
kernel/sched/fair.c +1
-1@@ -13974,19 +13974,32 @@ static inline int find_new_ilb(void)if (ilb_cpu == this_cpu)continue ; -if (idle_cpu (ilb_cpu)) +if (!idle_cpu (ilb_cpu))kernel/sched/fair.c static inline int find_new_ilb(void) · line 13987
Quoted 10 lines
kernel/sched/fair.c +10
-0+continue ; + +/* +* Running the idle load balancer on an idle sibling of a busy +* SMT core can reduce the capacity available to its sibling. Prefer +* a CPU whose entire core is idle, but retain the first idle CPU as +* a fallback so idle balancing can still make progress when no fully +* idle core exists. +*/ +if (!sched_smt_active () ||is_core_idle (ilb_cpu))nit.
is_core_idle()here would iterate all siblings and on systems with SMT-4 and SMT-8, that overhead is apparently visible when one thread per core is occupied based on past optimizations likef8858d96061f("sched/fair:
Optimizeshould_we_balance()for large SMT systems").
Quoted 9 lines
Quoted 5 lines
Quoted 1 line
Copying the same approach from that optimization, can we do:
Good point. Without removing the remaining siblings, we may call
is_core_idle()repeatedly for the same partially busy SMT core. I'll repeat my tests on my Vera system and incorporate your suggestion in v2 if I don't see any regression.Thank you! From my experience, it is not very visible on SMT-2 but Shrikanth can vouch for the overheads on SMT-4, SMT-8 systems.
I had the same thought when I saw is_core_idle there. I will try to give the patch a try.
Quoted 76 lines
Quoted 58 lines
Quoted 58 lines
(Only build tested)
kernel/sched/fair.c +21
-8@@ -13849,21 +13849,35 @@ static inline int on_null_domain(struct rq *rq)*/ static inline int find_new_ilb (void ) { +struct cpumask *ilb_cpus =this_cpu_cpumask_var_ptr (select_rq_mask);int this_cpu =smp_processor_id (); -const struct cpumask *hk_mask; -int ilb_cpu; +int ilb_cpu, fallback =-1 ; - hk_mask =housekeeping_cpumask (HK_TYPE_KERNEL_NOISE ); +cpumask_and (ilb_cpus, nohz.idle_cpus_mask,housekeeping_cpumask (HK_TYPE_KERNEL_NOISE )); -for_each_cpu_and (ilb_cpu, nohz.idle_cpus_mask, hk_mask) { +for_each_cpu (ilb_cpu, ilb_cpus) {if (ilb_cpu == this_cpu)continue ; -if (idle_cpu (ilb_cpu)) -return ilb_cpu; +if (!idle_cpu (ilb_cpu)) +continue ; + +if (sched_smt_active () && !is_core_idle (ilb_cpu)) { +if (fallback ==-1 ) + fallback = ilb_cpu; +/* +* If the core is not idle, and first SMT sibling which is +* idle has been found, then its not needed to check other +* SMT siblings for idleness: +*/ +cpumask_andnot (ilb_cpus, ilb_cpus,cpu_smt_mask (ilb_cpu)); +continue ; + } + +return ilb_cpu; } -return -1 ; +return fallback; }/* --- It is safe to use "
select_rq_mask" here since this is the tick handler trying to find anilb_cpuand "select_rq_mask" is only used in contexts with IRQs disabled. It can probably be renamed to suggest that it is safe to be used in any IRQ disabled context as a temporary mask.Quoted 4 lines
Quoted 1 line
Thoughts?
Agreed. We can also add
lockdep_assert_irqs_disabled()tofind_new_ilb()to better document and verify the condition that makes reusingselect_rq_masksafe.That works too but it just looks a bot odd to have the
selectrq_maskin a load balancing function.Quoted 3 lines
Speaking of that, instead of renaming it, would it be better to provide a helper to access
select_rq_maskwithlockdep_assert_irqs_disabled()?I'll defer to Peter on that :-)
He had previously suggested renaming it when there were discussions to reuse it here
https://lore.kernel.org/lkml/[email protected]/
On Wed, Jul 29, 2026 at 04:09:05PM +0530, K Prateek Nayak wrote:
...
Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
It is safe to use "
select_rq_mask" here since this is the tick handler trying to find anilb_cpuand "select_rq_mask" is only used in contexts with IRQs disabled. It can probably be renamed to suggest that it is safe to be used in any IRQ disabled context as a temporary mask.
Quoted 7 lines
Quoted 4 lines
Quoted 1 line
Thoughts?
Agreed. We can also add
lockdep_assert_irqs_disabled()tofind_new_ilb()to better document and verify the condition that makes reusingselect_rq_masksafe.That works too but it just looks a bot odd to have the
selectrq_maskin a load balancing function.
Ack.
Quoted 10 lines
Quoted 4 lines
Speaking of that, instead of renaming it, would it be better to provide a helper to access
select_rq_maskwithlockdep_assert_irqs_disabled()?I'll defer to Peter on that :-)
He had previously suggested renaming it when there were discussions to reuse it here
https://lore.kernel.org/lkml/[email protected]/
Thanks for the pointer! Looking at that discussion, this seems broader than a simple mechanical rename and it doesn't appear to have converged on a concrete interface. It looks like a useful but independent scheduler cleanup and I'd prefer not to make it a prerequisite for this.
Maybe for v2 I can keep the existing select_rq_mask, add lockdep_assert_irqs_disabled() and implement your suggestion. Results are looking good so far on my side with your change.
Peter, do you have a preference on whether the mask renaming/accessor work should be addressed separately, or included as a preparatory patch for this?
Thanks,
-Andrea
Peter Zijlstra
maintainerQuoted 2 lines · Andrea Righi
Peter, do you have a preference on whether the mask renaming/accessor work should be addressed separately, or included as a preparatory patch for this?
Separate would be fine. Perhaps another way to do it is using the fancy new clang context analysis.