Adjust NUMA imbalance for multiple LLCs
v6 cover Adjust NUMA imbalance for multiple LLCs
Changelog sinve v5
o Fix off-by-one error
Changelog since V4
o Scale imbalance based on the top domain that prefers siblings
o Keep allowed imbalance as 2 up to the point where LLCs can be overloaded
Changelog since V3
o Calculate imb_numa_nr for multiple SD_NUMA domains
o Restore behaviour where communicating pairs remain on the same node
Commit 7d2b5dd0bcc4 ("sched/numa: Allow a floating imbalance between NUMA nodes") allowed an imbalance between NUMA nodes such that communicating tasks would not be pulled apart by the load balancer. This works fine when there is a 1:1 relationship between LLC and node but can be suboptimal for multiple LLCs if independent tasks prematurely use CPUs sharing cache.
The series addresses two problems -- inconsistent logic when allowing a NUMA imbalance and sub-optimal performance when there are many LLCs per NUMA node.
include/linux/sched/topology.h | 1 + kernel/sched/fair.c | 30 ++++++++++--------- kernel/sched/topology.c | 53 ++++++++++++++++++++++++++++++++++ 3 files changed, 71 insertions(+), 13 deletions(-)
Signature · 1 line
2.31.1
PATCH v6 1/2 sched/fair: Improve consistency of allowed NUMA balance calculations
There are inconsistencies when determining if a NUMA imbalance is allowed that should be corrected.
o allow_numa_imbalance changes types and is not always examining the destination group so both the type should be corrected as well as the naming.
o find_idlest_group uses the sched_domain's weight instead of the group weight which is different to find_busiest_group
o find_busiest_group uses the source group instead of the destination which is different to task_numa_find_cpu
o Both find_idlest_group and find_busiest_group should account for the number of running tasks if a move was allowed to be consistent with task_numa_find_cpu
--- kernel/sched/fair.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-)
kernel/sched/fair.c +10 -8
@@ -9003,9 +9003,10 @@ static bool update_pick_idlest(struct sched_group *idlest,* This is an approximation as the number of running tasks may not be * related to the number of busy CPUs due to sched_setaffinity. */ -static inline bool allow_numa_imbalance (int dst_running,int dst_weight) +static inline bool +allow_numa_imbalance (unsigned int running,unsigned int weight) { -return (dst_running < (dst_weight >>2 )); +return (running < (weight >>2 )); }/*
@@ -9139,12 +9140,13 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu)return idlest;#endif /* -* Otherwise, keep the task on this node to stay close -* its wakeup source and improve locality. If there is -* a real need of migration, periodic load balance will -* take care of it. +* Otherwise, keep the task close to the wakeup source +* and improve locality if the number of running tasks +* would remain below threshold where an imbalance is +* allowed. If there is a real need of migration, +* periodic load balance will take care of it. */ -if (allow_numa_imbalance (local_sgs.sum_nr_running, sd->span_weight)) +if (allow_numa_imbalance (local_sgs.sum_nr_running +1 , local_sgs.group_weight))return NULL ; }
@@ -9350,7 +9352,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s/* Consider allowing a small imbalance between NUMA groups */ if (env->sd->flags &SD_NUMA ) { env->imbalance =adjust_numa_imbalance (env->imbalance, - busiest->sum_nr_running, busiest->group_weight); + local->sum_nr_running +1 , local->group_weight); }return ;
Signature · 1 line
2.31.1
Re: [PATCH 1/2] sched/fair: Improve consistency of allowed NUMA balance calculations
Hello Mel,
Quoted 14 lines · Mel Gorman
There are inconsistencies when determining if a NUMA imbalance is allowed that should be corrected.
o
allow_numa_imbalancechanges types and is not always examining the destination group so both the type should be corrected as well as the naming.
ofind_idlest_groupuses the sched_domain's weight instead of the group weight which is different tofind_busiest_group
ofind_busiest_groupuses the source group instead of the destination which is different totask_numa_find_cpu
o Bothfind_idlest_groupandfind_busiest_groupshould account for the number of running tasks if a move was allowed to be consistent withtask_numa_find_cpu
This patch looks good to me. Using the local_sgs.group_weight instead of sd->span_weight is definitely helping improve the spread of the stream threads on AMD EPYC processors.
Quoted 54 lines
Fixes 7d2b5dd0bcc4 ("sched/numa: Allow a floating imbalance between NUMA nodes")Signed-off-by Mel Gorman <[email protected]>--- kernel/sched/fair.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-)kernel/sched/fair.c +10
-8@@ -9003,9 +9003,10 @@ static bool update_pick_idlest(struct sched_group *idlest,* This is an approximation as the number of running tasks may not be * related to the number of busy CPUs due to sched_setaffinity. */ -static inline bool allow_numa_imbalance (int dst_running,int dst_weight) +static inline bool +allow_numa_imbalance (unsigned int running,unsigned int weight) { -return (dst_running < (dst_weight >>2 )); +return (running < (weight >>2 )); }/* @@ -9139,12 +9140,13 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu)return idlest;#endif /* -* Otherwise, keep the task on this node to stay close -* its wakeup source and improve locality. If there is -* a real need of migration, periodic load balance will -* take care of it. +* Otherwise, keep the task close to the wakeup source +* and improve locality if the number of running tasks +* would remain below threshold where an imbalance is +* allowed. If there is a real need of migration, +* periodic load balance will take care of it. */ -if (allow_numa_imbalance (local_sgs.sum_nr_running, sd->span_weight)) +if (allow_numa_imbalance (local_sgs.sum_nr_running +1 , local_sgs.group_weight))return NULL ; }@@ -9350,7 +9352,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s/* Consider allowing a small imbalance between NUMA groups */ if (env->sd->flags &SD_NUMA ) { env->imbalance =adjust_numa_imbalance (env->imbalance, - busiest->sum_nr_running, busiest->group_weight); + local->sum_nr_running +1 , local->group_weight); }return ;Signature · 1 line
2.31.1
Re: [PATCH 1/2] sched/fair: Improve consistency of allowed NUMA balance calculations
In reply to Mel GormanQuoted 15 lines · Mel Gorman
There are inconsistencies when determining if a NUMA imbalance is allowed that should be corrected.
o
allow_numa_imbalancechanges types and is not always examining the destination group so both the type should be corrected as well as the naming.
ofind_idlest_groupuses the sched_domain's weight instead of the group weight which is different tofind_busiest_group
ofind_busiest_groupuses the source group instead of the destination which is different totask_numa_find_cpu
o Bothfind_idlest_groupandfind_busiest_groupshould account for the number of running tasks if a move was allowed to be consistent withtask_numa_find_cpu
Quoted 2 lines
Fixes 7d2b5dd0bcc4 ("sched/numa: Allow a floating imbalance between NUMA nodes")Signed-off-by Mel Gorman <[email protected]>
Quoted 52 lines
--- kernel/sched/fair.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-)kernel/sched/fair.c +10
-9@@ -9003,9 +9003,10 @@ static bool update_pick_idlest(struct sched_group *idlest,* This is an approximation as the number of running tasks may not be * related to the number of busy CPUs due to sched_setaffinity. */ -static inline bool allow_numa_imbalance (int dst_running,int dst_weight) +static inline bool +allow_numa_imbalance (unsigned int running,unsigned int weight) { -return (dst_running < (dst_weight >>2 )); +return (running < (weight >>2 )); }/* @@ -9139,12 +9140,13 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu)return idlest;#endif /* -* Otherwise, keep the task on this node to stay close -* its wakeup source and improve locality. If there is -* a real need of migration, periodic load balance will -* take care of it. +* Otherwise, keep the task close to the wakeup source +* and improve locality if the number of running tasks +* would remain below threshold where an imbalance is +* allowed. If there is a real need of migration, +* periodic load balance will take care of it. */ -if (allow_numa_imbalance (local_sgs.sum_nr_running, sd->span_weight)) +if (allow_numa_imbalance (local_sgs.sum_nr_running +1 , local_sgs.group_weight))return NULL ; }@@ -9350,7 +9352,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s/* Consider allowing a small imbalance between NUMA groups */ if (env->sd->flags &SD_NUMA ) { env->imbalance =adjust_numa_imbalance (env->imbalance, - busiest->sum_nr_running, busiest->group_weight); + local->sum_nr_running +1 , local->group_weight); }return ; --2.31.1
Re: [PATCH 1/2] sched/fair: Improve consistency of allowed NUMA balance calculations
In reply to Mel GormanQuoted 68 lines · Mel Gorman
There are inconsistencies when determining if a NUMA imbalance is allowed that should be corrected.
o
allow_numa_imbalancechanges types and is not always examining the destination group so both the type should be corrected as well as the naming.
ofind_idlest_groupuses the sched_domain's weight instead of the group weight which is different tofind_busiest_group
ofind_busiest_groupuses the source group instead of the destination which is different totask_numa_find_cpu
o Bothfind_idlest_groupandfind_busiest_groupshould account for the number of running tasks if a move was allowed to be consistent withtask_numa_find_cpuFixes 7d2b5dd0bcc4 ("sched/numa: Allow a floating imbalance between NUMA nodes")Signed-off-by Mel Gorman <[email protected]>--- kernel/sched/fair.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-)kernel/sched/fair.c +10
-8@@ -9003,9 +9003,10 @@ static bool update_pick_idlest(struct sched_group *idlest,* This is an approximation as the number of running tasks may not be * related to the number of busy CPUs due to sched_setaffinity. */ -static inline bool allow_numa_imbalance (int dst_running,int dst_weight) +static inline bool +allow_numa_imbalance (unsigned int running,unsigned int weight) { -return (dst_running < (dst_weight >>2 )); +return (running < (weight >>2 )); }/* @@ -9139,12 +9140,13 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu)return idlest;#endif /* -* Otherwise, keep the task on this node to stay close -* its wakeup source and improve locality. If there is -* a real need of migration, periodic load balance will -* take care of it. +* Otherwise, keep the task close to the wakeup source +* and improve locality if the number of running tasks +* would remain below threshold where an imbalance is +* allowed. If there is a real need of migration, +* periodic load balance will take care of it. */ -if (allow_numa_imbalance (local_sgs.sum_nr_running, sd->span_weight)) +if (allow_numa_imbalance (local_sgs.sum_nr_running +1 , local_sgs.group_weight))return NULL ; }@@ -9350,7 +9352,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s/* Consider allowing a small imbalance between NUMA groups */ if (env->sd->flags &SD_NUMA ) { env->imbalance =adjust_numa_imbalance (env->imbalance, - busiest->sum_nr_running, busiest->group_weight); + local->sum_nr_running +1 , local->group_weight); }return ;Signature · 1 line
2.31.1
Looks good to me.
Signature · 2 lines
Thanks and Regards Srikar Dronamraju
[tip: sched/core] sched/fair: Improve consistency of allowed NUMA balance calculations
In reply to Mel GormanThe following commit has been merged into the sched/core branch of tip:
The following commit has been merged into the sched/core branch of tip:
Commit-ID: 2cfb7a1b031b0e816af7a6ee0c6ab83b0acdf05a Gitweb: https://git.kernel.org/tip/2cfb7a1b031b0e816af7a6ee0c6ab83b0acdf05a Author: Mel Gorman <[email protected]> AuthorDate: Tue, 08 Feb 2022 09:43:33 Committer: Peter Zijlstra <[email protected]> CommitterDate: Fri, 11 Feb 2022 23:30:08 +01:00
sched/fair: Improve consistency of allowed NUMA balance calculations
There are inconsistencies when determining if a NUMA imbalance is allowed that should be corrected.
o allow_numa_imbalance changes types and is not always examining the destination group so both the type should be corrected as well as the naming.
o find_idlest_group uses the sched_domain's weight instead of the group weight which is different to find_busiest_group
o find_busiest_group uses the source group instead of the destination which is different to task_numa_find_cpu
o Both find_idlest_group and find_busiest_group should account for the number of running tasks if a move was allowed to be consistent with task_numa_find_cpu
--- kernel/sched/fair.c | 18 ++++++++++-------- 1 file changed, 10 insertions(+), 8 deletions(-)
kernel/sched/fair.c +10 -8
@@ -9005,9 +9005,10 @@ static bool update_pick_idlest(struct sched_group *idlest,* This is an approximation as the number of running tasks may not be * related to the number of busy CPUs due to sched_setaffinity. */ -static inline bool allow_numa_imbalance (int dst_running,int dst_weight) +static inline bool +allow_numa_imbalance (unsigned int running,unsigned int weight) { -return (dst_running < (dst_weight >>2 )); +return (running < (weight >>2 )); }/*
@@ -9141,12 +9142,13 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu)return idlest;#endif /* -* Otherwise, keep the task on this node to stay close -* its wakeup source and improve locality. If there is -* a real need of migration, periodic load balance will -* take care of it. +* Otherwise, keep the task close to the wakeup source +* and improve locality if the number of running tasks +* would remain below threshold where an imbalance is +* allowed. If there is a real need of migration, +* periodic load balance will take care of it. */ -if (allow_numa_imbalance (local_sgs.sum_nr_running, sd->span_weight)) +if (allow_numa_imbalance (local_sgs.sum_nr_running +1 , local_sgs.group_weight))return NULL ; }
@@ -9352,7 +9354,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s/* Consider allowing a small imbalance between NUMA groups */ if (env->sd->flags &SD_NUMA ) { env->imbalance =adjust_numa_imbalance (env->imbalance, - busiest->sum_nr_running, busiest->group_weight); + local->sum_nr_running +1 , local->group_weight); }return ;
PATCH v6 2/2 sched/fair: Adjust the allowed NUMA imbalance when SD_NUMA spans multiple LLCs
In reply to Mel GormanCommit 7d2b5dd0bcc4 ("sched/numa: Allow a floating imbalance between NUMA nodes") allowed an imbalance between NUMA nodes such that communicating tasks would not be pulled apart by the load balancer. This works fine when there is a 1:1 relationship between LLC and node but can be suboptimal for multiple LLCs if independent tasks prematurely use CPUs sharing cache.
Zen* has multiple LLCs per node with local memory channels and due to the allowed imbalance, it's far harder to tune some workloads to run optimally than it is on hardware that has 1 LLC per node. This patch allows an imbalance to exist up to the point where LLCs should be balanced between nodes.
On a Zen3 machine running STREAM parallelised with OMP to have on instance per LLC the results and without binding, the results are
5.17.0-rc0 5.17.0-rc0
vanilla sched-numaimb-v6
MB/sec copy-16 162596.94 ( 0.00%) 580559.74 ( 257.05%)
MB/sec scale-16 136901.28 ( 0.00%) 374450.52 ( 173.52%)
MB/sec add-16 157300.70 ( 0.00%) 564113.76 ( 258.62%)
MB/sec triad-16 151446.88 ( 0.00%) 564304.24 ( 272.61%)STREAM can use directives to force the spread if the OpenMP is new enough but that doesn't help if an application uses threads and it's not known in advance how many threads will be created.
Coremark is a CPU and cache intensive benchmark parallelised with threads. When running with 1 thread per core, the vanilla kernel allows threads to contend on cache. With the patch;
5.17.0-rc0 5.17.0-rc0
vanilla sched-numaimb-v5
Min Score-16 368239.36 ( 0.00%) 389816.06 ( 5.86%)
Hmean Score-16 388607.33 ( 0.00%) 427877.08 * 10.11%*
Max Score-16 408945.69 ( 0.00%) 481022.17 ( 17.62%)
Stddev Score-16 15247.04 ( 0.00%) 24966.82 ( -63.75%)
CoeffVar Score-16 3.92 ( 0.00%) 5.82 ( -48.48%)It can also make a big difference for semi-realistic workloads like specjbb which can execute arbitrary numbers of threads without advance knowledge of how they should be placed. Even in cases where the average performance is neutral, the results are more stable.
5.17.0-rc0 5.17.0-rc0
vanilla sched-numaimb-v6
Hmean tput-1 71631.55 ( 0.00%) 73065.57 ( 2.00%)
Hmean tput-8 582758.78 ( 0.00%) 556777.23 ( -4.46%)
Hmean tput-16 1020372.75 ( 0.00%) 1009995.26 ( -1.02%)
Hmean tput-24 1416430.67 ( 0.00%) 1398700.11 ( -1.25%)
Hmean tput-32 1687702.72 ( 0.00%) 1671357.04 ( -0.97%)
Hmean tput-40 1798094.90 ( 0.00%) 2015616.46 * 12.10%*
Hmean tput-48 1972731.77 ( 0.00%) 2333233.72 ( 18.27%)
Hmean tput-56 2386872.38 ( 0.00%) 2759483.38 ( 15.61%)
Hmean tput-64 2909475.33 ( 0.00%) 2925074.69 ( 0.54%)
Hmean tput-72 2585071.36 ( 0.00%) 2962443.97 ( 14.60%)
Hmean tput-80 2994387.24 ( 0.00%) 3015980.59 ( 0.72%)
Hmean tput-88 3061408.57 ( 0.00%) 3010296.16 ( -1.67%)
Hmean tput-96 3052394.82 ( 0.00%) 2784743.41 ( -8.77%)
Hmean tput-104 2997814.76 ( 0.00%) 2758184.50 ( -7.99%)
Hmean tput-112 2955353.29 ( 0.00%) 2859705.09 ( -3.24%)
Hmean tput-120 2889770.71 ( 0.00%) 2764478.46 ( -4.34%)
Hmean tput-128 2871713.84 ( 0.00%) 2750136.73 ( -4.23%)
Stddev tput-1 5325.93 ( 0.00%) 2002.53 ( 62.40%)
Stddev tput-8 6630.54 ( 0.00%) 10905.00 ( -64.47%)
Stddev tput-16 25608.58 ( 0.00%) 6851.16 ( 73.25%)
Stddev tput-24 12117.69 ( 0.00%) 4227.79 ( 65.11%)
Stddev tput-32 27577.16 ( 0.00%) 8761.05 ( 68.23%)
Stddev tput-40 59505.86 ( 0.00%) 2048.49 ( 96.56%)
Stddev tput-48 168330.30 ( 0.00%) 93058.08 ( 44.72%)
Stddev tput-56 219540.39 ( 0.00%) 30687.02 ( 86.02%)
Stddev tput-64 121750.35 ( 0.00%) 9617.36 ( 92.10%)
Stddev tput-72 223387.05 ( 0.00%) 34081.13 ( 84.74%)
Stddev tput-80 128198.46 ( 0.00%) 22565.19 ( 82.40%)
Stddev tput-88 136665.36 ( 0.00%) 27905.97 ( 79.58%)
Stddev tput-96 111925.81 ( 0.00%) 99615.79 ( 11.00%)
Stddev tput-104 146455.96 ( 0.00%) 28861.98 ( 80.29%)
Stddev tput-112 88740.49 ( 0.00%) 58288.23 ( 34.32%)
Stddev tput-120 186384.86 ( 0.00%) 45812.03 ( 75.42%)
Stddev tput-128 78761.09 ( 0.00%) 57418.48 ( 27.10%)Similarly, for embarassingly parallel problems like NPB-ep, there are improvements due to better spreading across LLC when the machine is not fully utilised.
vanilla sched-numaimb-v6 Min ep.D 31.79 ( 0.00%) 26.11 ( 17.87%) Amean ep.D 31.86 ( 0.00%) 26.17 * 17.86%* Stddev ep.D 0.07 ( 0.00%) 0.05 ( 24.41%) CoeffVar ep.D 0.22 ( 0.00%) 0.20 ( 7.97%) Max ep.D 31.93 ( 0.00%) 26.21 ( 17.91%)
--- include/linux/sched/topology.h | 1 + kernel/sched/fair.c | 22 +++++++------- kernel/sched/topology.c | 53 ++++++++++++++++++++++++++++++++++ 3 files changed, 66 insertions(+), 10 deletions(-)
include/linux/sched/topology.h +1 -0
@@ -93,6 +93,7 @@ struct sched_domain {unsigned int busy_factor;/* less balancing by factor if busy */ unsigned int imbalance_pct;/* No balance until over watermark */ unsigned int cache_nice_tries;/* Leave cache hot tasks for # tries */ +unsigned int imb_numa_nr;/* Nr running tasks that allows a NUMA imbalance */ int nohz_idle;/* NOHZ IDLE status */ int flags;/* See SD_* */
kernel/sched/fair.c +12 -10
@@ -1489,6 +1489,7 @@ struct task_numa_env {int src_cpu, src_nid;int dst_cpu, dst_nid; +int imb_numa_nr;struct numa_stats src_stats, dst_stats;
@@ -1503,7 +1504,7 @@ struct task_numa_env {static unsigned long cpu_load (struct rq *rq);static unsigned long cpu_runnable (struct rq *rq);static inline long adjust_numa_imbalance (int imbalance, -int dst_running,int dst_weight); +int dst_running,int imb_numa_nr); static inline enum numa_type numa_classify(unsigned int imbalance_pct,
@@ -1884,7 +1885,7 @@ static void task_numa_find_cpu(struct task_numa_env *env, dst_running = env->dst_stats.nr_running +1 ; imbalance =max (0 , dst_running - src_running); imbalance =adjust_numa_imbalance (imbalance, dst_running, - env->dst_stats.weight); + env->imb_numa_nr);/* Use idle CPU if there is no imbalance */ if (!imbalance) {
@@ -1949,8 +1950,10 @@ static int task_numa_migrate(struct task_struct *p)*/ rcu_read_lock (); sd =rcu_dereference (per_cpu (sd_numa, env.src_cpu)); -if (sd) +if (sd) { env.imbalance_pct =100 + (sd->imbalance_pct -100 ) /2 ; + env.imb_numa_nr = sd->imb_numa_nr; + }rcu_read_unlock ();/*
@@ -9003,10 +9006,9 @@ static bool update_pick_idlest(struct sched_group *idlest,* This is an approximation as the number of running tasks may not be * related to the number of busy CPUs due to sched_setaffinity. */ -static inline bool -allow_numa_imbalance (unsigned int running,unsigned int weight) +static inline bool allow_numa_imbalance (int running,int imb_numa_nr) { -return (running < (weight >>2 )); +return running <= imb_numa_nr; }/*
@@ -9146,7 +9148,7 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu)* allowed. If there is a real need of migration, * periodic load balance will take care of it. */ -if (allow_numa_imbalance (local_sgs.sum_nr_running +1 , local_sgs.group_weight)) +if (allow_numa_imbalance (local_sgs.sum_nr_running +1 , sd->imb_numa_nr))return NULL ; }
@@ -9238,9 +9240,9 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd#define NUMA_IMBALANCE_MIN 2static inline long adjust_numa_imbalance (int imbalance, -int dst_running,int dst_weight) +int dst_running,int imb_numa_nr) { -if (!allow_numa_imbalance (dst_running, dst_weight)) +if (!allow_numa_imbalance (dst_running, imb_numa_nr))return imbalance;/*
@@ -9352,7 +9354,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s/* Consider allowing a small imbalance between NUMA groups */ if (env->sd->flags &SD_NUMA ) { env->imbalance =adjust_numa_imbalance (env->imbalance, - local->sum_nr_running +1 , local->group_weight); + local->sum_nr_running +1 , env->sd->imb_numa_nr); }return ;
kernel/sched/topology.c +53 -0
59 lines not on this page · raw message
Signature · 1 line
2.31.1
Re: [PATCH 2/2] sched/fair: Adjust the allowed NUMA imbalance when SD_NUMA spans multiple LLCs
On Tue, Feb 08, 2022 at 09:43:34AM +0000, Mel Gorman wrote:
[..snip..]
Quoted 50 lines
kernel/sched/topology.c +42
-0@@ -2242,6 +2242,59 @@ build_sched_domains(const struct cpumask *cpu_map, struct sched_domain_attr *att } } +/* +* Calculate an allowed NUMA imbalance such that LLCs do not get +* imbalanced. +*/ +for_each_cpu (i, cpu_map) { +unsigned int imb =0 ; +unsigned int imb_span =1 ; + +for (sd = *per_cpu_ptr (d.sd, i); sd; sd = sd->parent) { +struct sched_domain *child = sd->child; + +if (!(sd->flags &SD_SHARE_PKG_RESOURCES ) && child && + (child->flags &SD_SHARE_PKG_RESOURCES )) { +struct sched_domain *top, *top_p; +unsigned int nr_llcs; + +/* +* For a single LLC per node, allow an +* imbalance up to 25% of the node. This is an +* arbitrary cutoff based on SMT-2 to balance +* between memory bandwidth and avoiding +* premature sharing of HT resources and SMT-4 +* or SMT-8 *may* benefit from a different +* cutoff. +* +* For multiple LLCs, allow an imbalance +* until multiple tasks would share an LLC +* on one node while LLCs on another node +* remain idle. +*/ + nr_llcs = sd->span_weight / child->span_weight; +if (nr_llcs ==1 ) + imb = sd->span_weight >>2 ; +else + imb = nr_llcs; + sd->imb_numa_nr = imb; + +/* Set span based on the first NUMA domain. */ + top = sd; + top_p = top->parent; +while (top_p && !(top_p->flags &SD_NUMA )) { + top = top->parent;
kernel/sched/topology.c build_sched_domains(const struct cpumask *cpu_map, struct sched_domain_attr *att · line 2296
Quoted 10 lines
kernel/sched/topology.c +10
-0+ top_p = top->parent; + } + imb_span = top_p ? top_p->span_weight : sd->span_weight; + }else { +int factor =max (1U , (sd->span_weight / imb_span)); + + sd->imb_numa_nr = imb * factor; + } + } + }
On a 2 Socket Zen3 servers with 64 cores per socket, the imb_numa_nr works out to be as follows for different Node Per Socket (NPS) modes
NPS = 1:
======
SMT(span = 2) -- > MC (span = 16) --> DIE (span = 128) --> NUMA (span = 256)
Parent of LLC is DIE. nr_llcs = 128/16 = 8. imb = 8. top_p = NUMA. imb_span = 256.
for NUMA doman, factor = max(1U, 256/256) = 1. Thus sd->imb_numa_nr = 8.
NPS = 2
========
SMT(span=2)--> MC(span=16)--> NODE(span=64)--> NUMA1(span=128)--> NUMA2(span=256)
Parent of LLC = NODE. nr_llcs = 64/16 = 4. imb = 4. top_p = NUMA1. imb_span = 128.
For NUMA1 domain, factor = 1. sd->imb_numa_nr = 4. For NUMA2 domain, factor = 2. sd->imb_numa_nr = 8
NPS = 4
========
SMT(span=2)--> MC(span=16)--> NODE(span=32)--> NUMA1(span=128)--> NUMA2(span=256)
Parent of LLC = NODE. nr_llcs = 32/16 = 2. imb = 2. top_p = NUMA1. imb_span = 128.
For NUMA1 domain, factor = 1. sd->imb_numa_nr = 2. For NUMA2 domain, factor = 2. sd->imb_numa_nr = 4
The imb_numa_nr looks good for all the NPS modes. Furthermore, running stream with 16 threads (equal to the number of LLCs in the system) yields good results on all the NPS modes with this imb_numa_nr.
--
Thanks and Regards
gautham.
Re: [PATCH 2/2] sched/fair: Adjust the allowed NUMA imbalance when SD_NUMA spans multiple LLCs
In reply to Mel GormanHello Mel,
On 2/8/2022 3:13 PM, Mel Gorman wrote:
[..snip..]
Quoted 9 lines
On a Zen3 machine running STREAM parallelised with OMP to have on instance per LLC the results and without binding, the results are
5.17.0-rc0 5.17.0-rc0 vanilla sched-numaimb-v6 MB/sec copy-16 162596.94 ( 0.00%) 580559.74 ( 257.05%) MB/sec scale-16 136901.28 ( 0.00%) 374450.52 ( 173.52%) MB/sec add-16 157300.70 ( 0.00%) 564113.76 ( 258.62%) MB/sec triad-16 151446.88 ( 0.00%) 564304.24 ( 272.61%)
I was able to test STREAM without binding on different NPS configurations of two socket Zen3 machine.
The results look good:
sched-tip - 5.17.0-rc1 tip sched/core
mel-v6 - 5.17.0-rc1 tip sched/core + this patch
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Stream with 16 threads.
built with -DSTREAM_ARRAY_SIZE=128000000, -DNTIMES=10 Zen3, 64C128T per socket, 2 sockets,
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
NPS1
Test: sched-tip mel-v6
Copy: 114470.18 (0.00 pct) 152806.94 (33.49 pct) Scale: 111575.12 (0.00 pct) 189784.57 (70.09 pct) Add: 125436.15 (0.00 pct) 213371.05 (70.10 pct) Triad: 123068.86 (0.00 pct) 209809.11 (70.48 pct)
NPS2
Test: sched-tip mel-v6
Copy: 57936.28 (0.00 pct) 155038.70 (167.60 pct) Scale: 55599.30 (0.00 pct) 192601.59 (246.41 pct) Add: 63096.96 (0.00 pct) 211462.58 (235.13 pct) Triad: 61983.39 (0.00 pct) 208909.34 (237.04 pct)
NPS4
Test: sched-tip mel-v6
Copy: 43946.42 (0.00 pct) 119583.69 (172.11 pct) Scale: 33750.96 (0.00 pct) 180130.83 (433.70 pct) Add: 39109.72 (0.00 pct) 170296.68 (335.43 pct) Triad: 36598.88 (0.00 pct) 169953.47 (364.36 pct)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Stream with 16 threads.
built with -DSTREAM_ARRAY_SIZE=128000000, -DNTIMES=100 Zen3, 64C128T per socket, 2 sockets,
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
NPS1
Test: sched-tip mel-v6
Copy: 132402.79 (0.00 pct) 225587.85 (70.37 pct) Scale: 126923.02 (0.00 pct) 214363.58 (68.89 pct) Add: 145596.55 (0.00 pct) 260901.92 (79.19 pct) Triad: 143092.91 (0.00 pct) 249081.79 (74.06 pct)
NPS 2
Test: sched-tip mel-v6
Copy: 107386.27 (0.00 pct) 227623.31 (111.96 pct) Scale: 100941.44 (0.00 pct) 218116.63 (116.08 pct) Add: 115854.52 (0.00 pct) 272756.95 (135.43 pct) Triad: 113369.96 (0.00 pct) 260235.32 (129.54 pct)
NPS4
Test: sched-tip mel-v6
Copy: 91083.07 (0.00 pct) 247163.90 (171.36 pct) Scale: 90352.54 (0.00 pct) 223914.31 (147.82 pct) Add: 101973.98 (0.00 pct) 272842.42 (167.56 pct) Triad: 99773.65 (0.00 pct) 258904.54 (159.49 pct)
There is a significant improvement throughout the board with v6 outperforming tip/sched/core in every case!
--
Thanks and Regards
Prateek
Re: [PATCH 2/2] sched/fair: Adjust the allowed NUMA imbalance when SD_NUMA spans multiple LLCs
Quoted 5 lines · K Prateek Nayak
There is a significant improvement throughout the board with v6 outperforming tip/sched/core in every case!
Tested-by K Prateek Nayak <[email protected]>
Thanks very much Prateek and Gautham for reviewing and testing!
Signature · 2 lines
Mel Gorman SUSE Labs
Jirka Hladky
Tested-byRe: [PATCH 2/2] sched/fair: Adjust the allowed NUMA imbalance when SD_NUMA spans multiple LLCs
Hi Mel,
we have tested 5.17.0_rc2 + v6 of your patch series and the results are very good!
We see performance gains up to 25% for the hackbench (32 processes). Even more importantly, we don't see any performance losses that we have experienced with the previous versions of the patch series for NAS and SPECjbb2005 workloads.
Thanks a lot
Jirka
Quoted 14 lines · Mel Gorman
Quoted 5 lines · K Prateek Nayak
There is a significant improvement throughout the board with v6 outperforming tip/sched/core in every case!
Tested-by K Prateek Nayak <[email protected]>Thanks very much Prateek and Gautham for reviewing and testing!
--
Mel Gorman
SUSE Labs
Signature · 1 line
-Jirka
Re: [PATCH 2/2] sched/fair: Adjust the allowed NUMA imbalance when SD_NUMA spans multiple LLCs
In reply to Mel GormanQuoted 12 lines · Mel Gorman
Commit
7d2b5dd0bcc4("sched/numa: Allow a floating imbalance between NUMA nodes") allowed an imbalance between NUMA nodes such that communicating tasks would not be pulled apart by the load balancer. This works fine when there is a 1:1 relationship between LLC and node but can be suboptimal for multiple LLCs if independent tasks prematurely use CPUs sharing cache.Zen* has multiple LLCs per node with local memory channels and due to the allowed imbalance, it's far harder to tune some workloads to run optimally than it is on hardware that has 1 LLC per node. This patch allows an imbalance to exist up to the point where LLCs should be balanced between nodes.
Quoted 5 lines
On a Zen3 machine running STREAM parallelised with OMP to have on instance per LLC the results and without binding, the results are
5.17.0-rc0 5.17.0-rc0
Looks good to me.
Signature · 2 lines
Thanks and Regards Srikar Dronamraju
[tip: sched/core] sched/fair: Adjust the allowed NUMA imbalance when SD_NUMA spans multiple LLCs
In reply to Mel GormanThe following commit has been merged into the sched/core branch of tip:
The following commit has been merged into the sched/core branch of tip:
Commit-ID: e496132ebedd870b67f1f6d2428f9bb9d7ae27fd Gitweb: https://git.kernel.org/tip/e496132ebedd870b67f1f6d2428f9bb9d7ae27fd Author: Mel Gorman <[email protected]> AuthorDate: Tue, 08 Feb 2022 09:43:34 Committer: Peter Zijlstra <[email protected]> CommitterDate: Fri, 11 Feb 2022 23:30:08 +01:00
sched/fair: Adjust the allowed NUMA imbalance when SD_NUMA spans multiple LLCs
Commit 7d2b5dd0bcc4 ("sched/numa: Allow a floating imbalance between NUMA nodes") allowed an imbalance between NUMA nodes such that communicating tasks would not be pulled apart by the load balancer. This works fine when there is a 1:1 relationship between LLC and node but can be suboptimal for multiple LLCs if independent tasks prematurely use CPUs sharing cache.
Zen* has multiple LLCs per node with local memory channels and due to the allowed imbalance, it's far harder to tune some workloads to run optimally than it is on hardware that has 1 LLC per node. This patch allows an imbalance to exist up to the point where LLCs should be balanced between nodes.
On a Zen3 machine running STREAM parallelised with OMP to have on instance per LLC the results and without binding, the results are
5.17.0-rc0 5.17.0-rc0
vanilla sched-numaimb-v6
MB/sec copy-16 162596.94 ( 0.00%) 580559.74 ( 257.05%)
MB/sec scale-16 136901.28 ( 0.00%) 374450.52 ( 173.52%)
MB/sec add-16 157300.70 ( 0.00%) 564113.76 ( 258.62%)
MB/sec triad-16 151446.88 ( 0.00%) 564304.24 ( 272.61%)STREAM can use directives to force the spread if the OpenMP is new enough but that doesn't help if an application uses threads and it's not known in advance how many threads will be created.
Coremark is a CPU and cache intensive benchmark parallelised with threads. When running with 1 thread per core, the vanilla kernel allows threads to contend on cache. With the patch;
5.17.0-rc0 5.17.0-rc0
vanilla sched-numaimb-v5
Min Score-16 368239.36 ( 0.00%) 389816.06 ( 5.86%)
Hmean Score-16 388607.33 ( 0.00%) 427877.08 * 10.11%*
Max Score-16 408945.69 ( 0.00%) 481022.17 ( 17.62%)
Stddev Score-16 15247.04 ( 0.00%) 24966.82 ( -63.75%)
CoeffVar Score-16 3.92 ( 0.00%) 5.82 ( -48.48%)It can also make a big difference for semi-realistic workloads like specjbb which can execute arbitrary numbers of threads without advance knowledge of how they should be placed. Even in cases where the average performance is neutral, the results are more stable.
5.17.0-rc0 5.17.0-rc0
vanilla sched-numaimb-v6
Hmean tput-1 71631.55 ( 0.00%) 73065.57 ( 2.00%)
Hmean tput-8 582758.78 ( 0.00%) 556777.23 ( -4.46%)
Hmean tput-16 1020372.75 ( 0.00%) 1009995.26 ( -1.02%)
Hmean tput-24 1416430.67 ( 0.00%) 1398700.11 ( -1.25%)
Hmean tput-32 1687702.72 ( 0.00%) 1671357.04 ( -0.97%)
Hmean tput-40 1798094.90 ( 0.00%) 2015616.46 * 12.10%*
Hmean tput-48 1972731.77 ( 0.00%) 2333233.72 ( 18.27%)
Hmean tput-56 2386872.38 ( 0.00%) 2759483.38 ( 15.61%)
Hmean tput-64 2909475.33 ( 0.00%) 2925074.69 ( 0.54%)
Hmean tput-72 2585071.36 ( 0.00%) 2962443.97 ( 14.60%)
Hmean tput-80 2994387.24 ( 0.00%) 3015980.59 ( 0.72%)
Hmean tput-88 3061408.57 ( 0.00%) 3010296.16 ( -1.67%)
Hmean tput-96 3052394.82 ( 0.00%) 2784743.41 ( -8.77%)
Hmean tput-104 2997814.76 ( 0.00%) 2758184.50 ( -7.99%)
Hmean tput-112 2955353.29 ( 0.00%) 2859705.09 ( -3.24%)
Hmean tput-120 2889770.71 ( 0.00%) 2764478.46 ( -4.34%)
Hmean tput-128 2871713.84 ( 0.00%) 2750136.73 ( -4.23%)
Stddev tput-1 5325.93 ( 0.00%) 2002.53 ( 62.40%)
Stddev tput-8 6630.54 ( 0.00%) 10905.00 ( -64.47%)
Stddev tput-16 25608.58 ( 0.00%) 6851.16 ( 73.25%)
Stddev tput-24 12117.69 ( 0.00%) 4227.79 ( 65.11%)
Stddev tput-32 27577.16 ( 0.00%) 8761.05 ( 68.23%)
Stddev tput-40 59505.86 ( 0.00%) 2048.49 ( 96.56%)
Stddev tput-48 168330.30 ( 0.00%) 93058.08 ( 44.72%)
Stddev tput-56 219540.39 ( 0.00%) 30687.02 ( 86.02%)
Stddev tput-64 121750.35 ( 0.00%) 9617.36 ( 92.10%)
Stddev tput-72 223387.05 ( 0.00%) 34081.13 ( 84.74%)
Stddev tput-80 128198.46 ( 0.00%) 22565.19 ( 82.40%)
Stddev tput-88 136665.36 ( 0.00%) 27905.97 ( 79.58%)
Stddev tput-96 111925.81 ( 0.00%) 99615.79 ( 11.00%)
Stddev tput-104 146455.96 ( 0.00%) 28861.98 ( 80.29%)
Stddev tput-112 88740.49 ( 0.00%) 58288.23 ( 34.32%)
Stddev tput-120 186384.86 ( 0.00%) 45812.03 ( 75.42%)
Stddev tput-128 78761.09 ( 0.00%) 57418.48 ( 27.10%)Similarly, for embarassingly parallel problems like NPB-ep, there are improvements due to better spreading across LLC when the machine is not fully utilised.
vanilla sched-numaimb-v6 Min ep.D 31.79 ( 0.00%) 26.11 ( 17.87%) Amean ep.D 31.86 ( 0.00%) 26.17 * 17.86%* Stddev ep.D 0.07 ( 0.00%) 0.05 ( 24.41%) CoeffVar ep.D 0.22 ( 0.00%) 0.20 ( 7.97%) Max ep.D 31.93 ( 0.00%) 26.21 ( 17.91%)
--- include/linux/sched/topology.h | 1 +- kernel/sched/fair.c | 22 +++++++------- kernel/sched/topology.c | 53 +++++++++++++++++++++++++++++++++- 3 files changed, 66 insertions(+), 10 deletions(-)
include/linux/sched/topology.h +1 -0
@@ -93,6 +93,7 @@ struct sched_domain {unsigned int busy_factor;/* less balancing by factor if busy */ unsigned int imbalance_pct;/* No balance until over watermark */ unsigned int cache_nice_tries;/* Leave cache hot tasks for # tries */ +unsigned int imb_numa_nr;/* Nr running tasks that allows a NUMA imbalance */ int nohz_idle;/* NOHZ IDLE status */ int flags;/* See SD_* */
kernel/sched/fair.c +12 -10
@@ -1489,6 +1489,7 @@ struct task_numa_env {int src_cpu, src_nid;int dst_cpu, dst_nid; +int imb_numa_nr;struct numa_stats src_stats, dst_stats;
@@ -1503,7 +1504,7 @@ struct task_numa_env {static unsigned long cpu_load (struct rq *rq);static unsigned long cpu_runnable (struct rq *rq);static inline long adjust_numa_imbalance (int imbalance, -int dst_running,int dst_weight); +int dst_running,int imb_numa_nr); static inline enum numa_type numa_classify(unsigned int imbalance_pct,
@@ -1884,7 +1885,7 @@ static void task_numa_find_cpu(struct task_numa_env *env, dst_running = env->dst_stats.nr_running +1 ; imbalance =max (0 , dst_running - src_running); imbalance =adjust_numa_imbalance (imbalance, dst_running, - env->dst_stats.weight); + env->imb_numa_nr);/* Use idle CPU if there is no imbalance */ if (!imbalance) {
@@ -1949,8 +1950,10 @@ static int task_numa_migrate(struct task_struct *p)*/ rcu_read_lock (); sd =rcu_dereference (per_cpu (sd_numa, env.src_cpu)); -if (sd) +if (sd) { env.imbalance_pct =100 + (sd->imbalance_pct -100 ) /2 ; + env.imb_numa_nr = sd->imb_numa_nr; + }rcu_read_unlock ();/*
@@ -9005,10 +9008,9 @@ static bool update_pick_idlest(struct sched_group *idlest,* This is an approximation as the number of running tasks may not be * related to the number of busy CPUs due to sched_setaffinity. */ -static inline bool -allow_numa_imbalance (unsigned int running,unsigned int weight) +static inline bool allow_numa_imbalance (int running,int imb_numa_nr) { -return (running < (weight >>2 )); +return running <= imb_numa_nr; }/*
@@ -9148,7 +9150,7 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu)* allowed. If there is a real need of migration, * periodic load balance will take care of it. */ -if (allow_numa_imbalance (local_sgs.sum_nr_running +1 , local_sgs.group_weight)) +if (allow_numa_imbalance (local_sgs.sum_nr_running +1 , sd->imb_numa_nr))return NULL ; }
@@ -9240,9 +9242,9 @@ next_group:#define NUMA_IMBALANCE_MIN 2static inline long adjust_numa_imbalance (int imbalance, -int dst_running,int dst_weight) +int dst_running,int imb_numa_nr) { -if (!allow_numa_imbalance (dst_running, dst_weight)) +if (!allow_numa_imbalance (dst_running, imb_numa_nr))return imbalance;/*
@@ -9354,7 +9356,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s/* Consider allowing a small imbalance between NUMA groups */ if (env->sd->flags &SD_NUMA ) { env->imbalance =adjust_numa_imbalance (env->imbalance, - local->sum_nr_running +1 , local->group_weight); + local->sum_nr_running +1 , env->sd->imb_numa_nr); }return ;
kernel/sched/topology.c +53 -0
59 lines not on this page · raw message
Re: [PATCH 2/2] sched/fair: Adjust the allowed NUMA imbalance when SD_NUMA spans multiple LLCs
In reply to Mel GormanQuoted 85 lines · Mel Gorman
Commit
7d2b5dd0bcc4("sched/numa: Allow a floating imbalance between NUMA nodes") allowed an imbalance between NUMA nodes such that communicating tasks would not be pulled apart by the load balancer. This works fine when there is a 1:1 relationship between LLC and node but can be suboptimal for multiple LLCs if independent tasks prematurely use CPUs sharing cache.Zen* has multiple LLCs per node with local memory channels and due to the allowed imbalance, it's far harder to tune some workloads to run optimally than it is on hardware that has 1 LLC per node. This patch allows an imbalance to exist up to the point where LLCs should be balanced between nodes.
On a Zen3 machine running STREAM parallelised with OMP to have on instance per LLC the results and without binding, the results are
5.17.0-rc0 5.17.0-rc0 vanilla sched-numaimb-v6 MB/sec copy-16 162596.94 ( 0.00%) 580559.74 ( 257.05%) MB/sec scale-16 136901.28 ( 0.00%) 374450.52 ( 173.52%) MB/sec add-16 157300.70 ( 0.00%) 564113.76 ( 258.62%) MB/sec triad-16 151446.88 ( 0.00%) 564304.24 ( 272.61%)STREAM can use directives to force the spread if the OpenMP is new enough but that doesn't help if an application uses threads and it's not known in advance how many threads will be created.
Coremark is a CPU and cache intensive benchmark parallelised with threads. When running with 1 thread per core, the vanilla kernel allows threads to contend on cache. With the patch;
5.17.0-rc0 5.17.0-rc0 vanilla sched-numaimb-v5 Min Score-16 368239.36 ( 0.00%) 389816.06 ( 5.86%) Hmean Score-16 388607.33 ( 0.00%) 427877.08 * 10.11%* Max Score-16 408945.69 ( 0.00%) 481022.17 ( 17.62%) Stddev Score-16 15247.04 ( 0.00%) 24966.82 ( -63.75%) CoeffVar Score-16 3.92 ( 0.00%) 5.82 ( -48.48%)It can also make a big difference for semi-realistic workloads like specjbb which can execute arbitrary numbers of threads without advance knowledge of how they should be placed. Even in cases where the average performance is neutral, the results are more stable.
5.17.0-rc0 5.17.0-rc0 vanilla sched-numaimb-v6 Hmean tput-1 71631.55 ( 0.00%) 73065.57 ( 2.00%) Hmean tput-8 582758.78 ( 0.00%) 556777.23 ( -4.46%) Hmean tput-16 1020372.75 ( 0.00%) 1009995.26 ( -1.02%) Hmean tput-24 1416430.67 ( 0.00%) 1398700.11 ( -1.25%) Hmean tput-32 1687702.72 ( 0.00%) 1671357.04 ( -0.97%) Hmean tput-40 1798094.90 ( 0.00%) 2015616.46 * 12.10%* Hmean tput-48 1972731.77 ( 0.00%) 2333233.72 ( 18.27%) Hmean tput-56 2386872.38 ( 0.00%) 2759483.38 ( 15.61%) Hmean tput-64 2909475.33 ( 0.00%) 2925074.69 ( 0.54%) Hmean tput-72 2585071.36 ( 0.00%) 2962443.97 ( 14.60%) Hmean tput-80 2994387.24 ( 0.00%) 3015980.59 ( 0.72%) Hmean tput-88 3061408.57 ( 0.00%) 3010296.16 ( -1.67%) Hmean tput-96 3052394.82 ( 0.00%) 2784743.41 ( -8.77%) Hmean tput-104 2997814.76 ( 0.00%) 2758184.50 ( -7.99%) Hmean tput-112 2955353.29 ( 0.00%) 2859705.09 ( -3.24%) Hmean tput-120 2889770.71 ( 0.00%) 2764478.46 ( -4.34%) Hmean tput-128 2871713.84 ( 0.00%) 2750136.73 ( -4.23%) Stddev tput-1 5325.93 ( 0.00%) 2002.53 ( 62.40%) Stddev tput-8 6630.54 ( 0.00%) 10905.00 ( -64.47%) Stddev tput-16 25608.58 ( 0.00%) 6851.16 ( 73.25%) Stddev tput-24 12117.69 ( 0.00%) 4227.79 ( 65.11%) Stddev tput-32 27577.16 ( 0.00%) 8761.05 ( 68.23%) Stddev tput-40 59505.86 ( 0.00%) 2048.49 ( 96.56%) Stddev tput-48 168330.30 ( 0.00%) 93058.08 ( 44.72%) Stddev tput-56 219540.39 ( 0.00%) 30687.02 ( 86.02%) Stddev tput-64 121750.35 ( 0.00%) 9617.36 ( 92.10%) Stddev tput-72 223387.05 ( 0.00%) 34081.13 ( 84.74%) Stddev tput-80 128198.46 ( 0.00%) 22565.19 ( 82.40%) Stddev tput-88 136665.36 ( 0.00%) 27905.97 ( 79.58%) Stddev tput-96 111925.81 ( 0.00%) 99615.79 ( 11.00%) Stddev tput-104 146455.96 ( 0.00%) 28861.98 ( 80.29%) Stddev tput-112 88740.49 ( 0.00%) 58288.23 ( 34.32%) Stddev tput-120 186384.86 ( 0.00%) 45812.03 ( 75.42%) Stddev tput-128 78761.09 ( 0.00%) 57418.48 ( 27.10%)Similarly, for embarassingly parallel problems like NPB-ep, there are improvements due to better spreading across LLC when the machine is not fully utilised.
Quoted 8 lines
vanilla sched-numaimb-v6 Min ep.D 31.79 ( 0.00%) 26.11 ( 17.87%) Amean ep.D 31.86 ( 0.00%) 26.17 * 17.86%* Stddev ep.D 0.07 ( 0.00%) 0.05 ( 24.41%) CoeffVar ep.D 0.22 ( 0.00%) 0.20 ( 7.97%) Max ep.D 31.93 ( 0.00%) 26.21 ( 17.91%)Signed-off-by Mel Gorman <[email protected]>
Good to see that you have been able to move on SD_NUMA instead of SD_PREFER_SIBLING.
The allowed imbalance looks also more consistent whatever the number of LLC
Quoted 170 lines
--- include/linux/sched/topology.h | 1 + kernel/sched/fair.c | 22 +++++++------- kernel/sched/topology.c | 53 ++++++++++++++++++++++++++++++++++ 3 files changed, 66 insertions(+), 10 deletions(-)include/linux/sched/topology.h +1
-0@@ -93,6 +93,7 @@ struct sched_domain {unsigned int busy_factor;/* less balancing by factor if busy */ unsigned int imbalance_pct;/* No balance until over watermark */ unsigned int cache_nice_tries;/* Leave cache hot tasks for # tries */ +unsigned int imb_numa_nr;/* Nr running tasks that allows a NUMA imbalance */ int nohz_idle;/* NOHZ IDLE status */ int flags;/* See SD_* */ kernel/sched/fair.c +12
-10@@ -1489,6 +1489,7 @@ struct task_numa_env {int src_cpu, src_nid;int dst_cpu, dst_nid; +int imb_numa_nr;struct numa_stats src_stats, dst_stats;@@ -1503,7 +1504,7 @@ struct task_numa_env {static unsigned long cpu_load (struct rq *rq);static unsigned long cpu_runnable (struct rq *rq);static inline long adjust_numa_imbalance (int imbalance, -int dst_running,int dst_weight); +int dst_running,int imb_numa_nr); static inline enum numa_type numa_classify(unsigned int imbalance_pct,@@ -1884,7 +1885,7 @@ static void task_numa_find_cpu(struct task_numa_env *env, dst_running = env->dst_stats.nr_running +1 ; imbalance =max (0 , dst_running - src_running); imbalance =adjust_numa_imbalance (imbalance, dst_running, - env->dst_stats.weight); + env->imb_numa_nr);/* Use idle CPU if there is no imbalance */ if (!imbalance) {@@ -1949,8 +1950,10 @@ static int task_numa_migrate(struct task_struct *p)*/ rcu_read_lock (); sd =rcu_dereference (per_cpu (sd_numa, env.src_cpu)); -if (sd) +if (sd) { env.imbalance_pct =100 + (sd->imbalance_pct -100 ) /2 ; + env.imb_numa_nr = sd->imb_numa_nr; + }rcu_read_unlock ();/* @@ -9003,10 +9006,9 @@ static bool update_pick_idlest(struct sched_group *idlest,* This is an approximation as the number of running tasks may not be * related to the number of busy CPUs due to sched_setaffinity. */ -static inline bool -allow_numa_imbalance (unsigned int running,unsigned int weight) +static inline bool allow_numa_imbalance (int running,int imb_numa_nr) { -return (running < (weight >>2 )); +return running <= imb_numa_nr; }/* @@ -9146,7 +9148,7 @@ find_idlest_group(struct sched_domain *sd, struct task_struct *p, int this_cpu)* allowed. If there is a real need of migration, * periodic load balance will take care of it. */ -if (allow_numa_imbalance (local_sgs.sum_nr_running +1 , local_sgs.group_weight)) +if (allow_numa_imbalance (local_sgs.sum_nr_running +1 , sd->imb_numa_nr))return NULL ; }@@ -9238,9 +9240,9 @@ static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sd#define NUMA_IMBALANCE_MIN 2static inline long adjust_numa_imbalance (int imbalance, -int dst_running,int dst_weight) +int dst_running,int imb_numa_nr) { -if (!allow_numa_imbalance (dst_running, dst_weight)) +if (!allow_numa_imbalance (dst_running, imb_numa_nr))return imbalance;/* @@ -9352,7 +9354,7 @@ static inline void calculate_imbalance(struct lb_env *env, struct sd_lb_stats *s/* Consider allowing a small imbalance between NUMA groups */ if (env->sd->flags &SD_NUMA ) { env->imbalance =adjust_numa_imbalance (env->imbalance, - local->sum_nr_running +1 , local->group_weight); + local->sum_nr_running +1 , env->sd->imb_numa_nr); }return ;kernel/sched/topology.c +53
-1@@ -2242,6 +2242,59 @@ build_sched_domains(const struct cpumask *cpu_map, struct sched_domain_attr *att } } +/* +* Calculate an allowed NUMA imbalance such that LLCs do not get +* imbalanced. +*/ +for_each_cpu (i, cpu_map) { +unsigned int imb =0 ; +unsigned int imb_span =1 ; + +for (sd = *per_cpu_ptr (d.sd, i); sd; sd = sd->parent) { +struct sched_domain *child = sd->child; + +if (!(sd->flags &SD_SHARE_PKG_RESOURCES ) && child && + (child->flags &SD_SHARE_PKG_RESOURCES )) { +struct sched_domain *top, *top_p; +unsigned int nr_llcs; + +/* +* For a single LLC per node, allow an +* imbalance up to 25% of the node. This is an +* arbitrary cutoff based on SMT-2 to balance +* between memory bandwidth and avoiding +* premature sharing of HT resources and SMT-4 +* or SMT-8 *may* benefit from a different +* cutoff. +* +* For multiple LLCs, allow an imbalance +* until multiple tasks would share an LLC +* on one node while LLCs on another node +* remain idle. +*/ + nr_llcs = sd->span_weight / child->span_weight; +if (nr_llcs ==1 ) + imb = sd->span_weight >>2 ; +else + imb = nr_llcs; + sd->imb_numa_nr = imb; + +/* Set span based on the first NUMA domain. */ + top = sd; + top_p = top->parent; +while (top_p && !(top_p->flags &SD_NUMA )) { + top = top->parent; + top_p = top->parent; + } + imb_span = top_p ? top_p->span_weight : sd->span_weight; + }else { +int factor =max (1U , (sd->span_weight / imb_span)); + + sd->imb_numa_nr = imb * factor; + } + } + } +/* Calculate CPU capacity for physical packages and nodes */ for (i = nr_cpumask_bits-1 ; i >=0 ; i--) {if (!cpumask_test_cpu (i, cpu_map)) --2.31.1
Peter Zijlstra
maintainerQuoted 17 lines · Mel Gorman
Changelog sinve v5
o Fix off-by-one errorChangelog since V4
o Scale imbalance based on the top domain that prefers siblings
o Keep allowed imbalance as 2 up to the point where LLCs can be overloadedChangelog since V3
o Calculateimb_numa_nrfor multipleSD_NUMAdomains
o Restore behaviour where communicating pairs remain on the same nodeCommit
7d2b5dd0bcc4("sched/numa: Allow a floating imbalance between NUMA nodes") allowed an imbalance between NUMA nodes such that communicating tasks would not be pulled apart by the load balancer. This works fine when there is a 1:1 relationship between LLC and node but can be suboptimal for multiple LLCs if independent tasks prematurely use CPUs sharing cache.
Quoted 8 lines
The series addresses two problems -- inconsistent logic when allowing a NUMA imbalance and sub-optimal performance when there are many LLCs per NUMA node.
include/linux/sched/topology.h | 1 + kernel/sched/fair.c | 30 ++++++++++--------- kernel/sched/topology.c | 53 ++++++++++++++++++++++++++++++++++ 3 files changed, 71 insertions(+), 13 deletions(-)
Thanks!