From 8b36d07f1d63de102d464f44a89704bc62d00811 Mon Sep 17 00:00:00 2001 From: Ricardo Neri Date: Thu, 6 Apr 2023 13:31:37 -0700 Subject: sched/fair: Move is_core_idle() out of CONFIG_NUMA asym_packing needs this function to determine whether an SMT core is a suitable destination for load balancing. Signed-off-by: Ricardo Neri Signed-off-by: Peter Zijlstra (Intel) Tested-by: Zhang Rui Link: https://lore.kernel.org/r/20230406203148.19182-2-ricardo.neri-calderon@linux.intel.com --- kernel/sched/fair.c | 34 +++++++++++++++++----------------- 1 file changed, 17 insertions(+), 17 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 373ff5f55884..a47208dbb42a 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -1064,6 +1064,23 @@ update_stats_curr_start(struct cfs_rq *cfs_rq, struct sched_entity *se) * Scheduling class queueing methods: */ +static inline bool is_core_idle(int cpu) +{ +#ifdef CONFIG_SCHED_SMT + int sibling; + + for_each_cpu(sibling, cpu_smt_mask(cpu)) { + if (cpu == sibling) + continue; + + if (!idle_cpu(sibling)) + return false; + } +#endif + + return true; +} + #ifdef CONFIG_NUMA #define NUMA_IMBALANCE_MIN 2 @@ -1700,23 +1717,6 @@ struct numa_stats { int idle_cpu; }; -static inline bool is_core_idle(int cpu) -{ -#ifdef CONFIG_SCHED_SMT - int sibling; - - for_each_cpu(sibling, cpu_smt_mask(cpu)) { - if (cpu == sibling) - continue; - - if (!idle_cpu(sibling)) - return false; - } -#endif - - return true; -} - struct task_numa_env { struct task_struct *p; -- cgit v1.2.3 From eefefa716c9fa6aa73159f09954b7eeba4cafd09 Mon Sep 17 00:00:00 2001 From: Ricardo Neri Date: Thu, 6 Apr 2023 13:31:38 -0700 Subject: sched/fair: Only do asym_packing load balancing from fully idle SMT cores When balancing load between cores, all the SMT siblings of the destination CPU, if any, must be idle. Otherwise, pulling new tasks degrades the throughput of the busy SMT siblings. The overall throughput of the system remains the same. When balancing load within an SMT core this consideration is not relevant. Follow the priorities that hardware indicates. Suggested-by: Valentin Schneider Signed-off-by: Ricardo Neri Signed-off-by: Peter Zijlstra (Intel) Tested-by: Zhang Rui Link: https://lore.kernel.org/r/20230406203148.19182-3-ricardo.neri-calderon@linux.intel.com --- kernel/sched/fair.c | 56 ++++++++++++++++++++++++++++++++++++++--------------- 1 file changed, 40 insertions(+), 16 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index a47208dbb42a..713d03e73978 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -9330,6 +9330,25 @@ group_type group_classify(unsigned int imbalance_pct, return group_has_spare; } +/** + * sched_use_asym_prio - Check whether asym_packing priority must be used + * @sd: The scheduling domain of the load balancing + * @cpu: A CPU + * + * Always use CPU priority when balancing load between SMT siblings. When + * balancing load between cores, it is not sufficient that @cpu is idle. Only + * use CPU priority if the whole core is idle. + * + * Returns: True if the priority of @cpu must be followed. False otherwise. + */ +static bool sched_use_asym_prio(struct sched_domain *sd, int cpu) +{ + if (!sched_smt_active()) + return true; + + return sd->flags & SD_SHARE_CPUCAPACITY || is_core_idle(cpu); +} + /** * asym_smt_can_pull_tasks - Check whether the load balancing CPU can pull tasks * @dst_cpu: Destination CPU of the load balancing @@ -9340,6 +9359,9 @@ group_type group_classify(unsigned int imbalance_pct, * Check the state of the SMT siblings of both @sds::local and @sg and decide * if @dst_cpu can pull tasks. * + * This function must be called only if all the SMT siblings of @dst_cpu are + * idle, if any. + * * If @dst_cpu does not have SMT siblings, it can pull tasks if two or more of * the SMT siblings of @sg are busy. If only one CPU in @sg is busy, pull tasks * only if @dst_cpu has higher priority. @@ -9349,8 +9371,7 @@ group_type group_classify(unsigned int imbalance_pct, * Bigger imbalances in the number of busy CPUs will be dealt with in * update_sd_pick_busiest(). * - * If @sg does not have SMT siblings, only pull tasks if all of the SMT siblings - * of @dst_cpu are idle and @sg has lower priority. + * If @sg does not have SMT siblings, only pull tasks if @sg has lower priority. * * Return: true if @dst_cpu can pull tasks, false otherwise. */ @@ -9398,15 +9419,8 @@ static bool asym_smt_can_pull_tasks(int dst_cpu, struct sd_lb_stats *sds, return false; } - /* - * @sg does not have SMT siblings. Ensure that @sds::local does not end - * up with more than one busy SMT sibling and only pull tasks if there - * are not busy CPUs (i.e., no CPU has running tasks). - */ - if (!sds->local_stat.sum_nr_running) - return sched_asym_prefer(dst_cpu, sg->asym_prefer_cpu); - - return false; + /* If we are here @dst_cpu has SMT siblings and are also idle. */ + return sched_asym_prefer(dst_cpu, sg->asym_prefer_cpu); #else /* Always return false so that callers deal with non-SMT cases. */ return false; @@ -9417,7 +9431,11 @@ static inline bool sched_asym(struct lb_env *env, struct sd_lb_stats *sds, struct sg_lb_stats *sgs, struct sched_group *group) { - /* Only do SMT checks if either local or candidate have SMT siblings */ + /* Ensure that the whole local core is idle, if applicable. */ + if (!sched_use_asym_prio(env->sd, env->dst_cpu)) + return false; + + /* Only do SMT checks if either local or candidate have SMT siblings. */ if ((sds->local->flags & SD_SHARE_CPUCAPACITY) || (group->flags & SD_SHARE_CPUCAPACITY)) return asym_smt_can_pull_tasks(env->dst_cpu, sds, sgs, group); @@ -10632,11 +10650,13 @@ static inline bool asym_active_balance(struct lb_env *env) { /* - * ASYM_PACKING needs to force migrate tasks from busy but - * lower priority CPUs in order to pack all tasks in the - * highest priority CPUs. + * ASYM_PACKING needs to force migrate tasks from busy but lower + * priority CPUs in order to pack all tasks in the highest priority + * CPUs. When done between cores, do it only if the whole core if the + * whole core is idle. */ return env->idle != CPU_NOT_IDLE && (env->sd->flags & SD_ASYM_PACKING) && + sched_use_asym_prio(env->sd, env->dst_cpu) && sched_asym_prefer(env->dst_cpu, env->src_cpu); } @@ -11371,9 +11391,13 @@ static void nohz_balancer_kick(struct rq *rq) * When ASYM_PACKING; see if there's a more preferred CPU * currently idle; in which case, kick the ILB to move tasks * around. + * + * When balancing betwen cores, all the SMT siblings of the + * preferred CPU must be idle. */ for_each_cpu_and(i, sched_domain_span(sd), nohz.idle_cpus_mask) { - if (sched_asym_prefer(i, cpu)) { + if (sched_use_asym_prio(sd, i) && + sched_asym_prefer(i, cpu)) { flags = NOHZ_STATS_KICK | NOHZ_BALANCE_KICK; goto unlock; } -- cgit v1.2.3 From ef7657d4d2d6a8456aa624010de456c32a135fe9 Mon Sep 17 00:00:00 2001 From: Ricardo Neri Date: Thu, 6 Apr 2023 13:31:39 -0700 Subject: sched/fair: Simplify asym_packing logic for SMT cores Callers of asym_smt_can_pull_tasks() check the idle state of the destination CPU and its SMT siblings, if any. No extra checks are needed in such function. Since SMT cores divide capacity among its siblings, priorities only really make sense if only one sibling is active. This is true for SMT2, SMT4, SMT8, etc. Do not use asym_packing load balance for this case. Instead, let find_busiest_group() handle imbalances. When balancing non-SMT cores or at higher scheduling domains (e.g., between MC scheduling groups), continue using priorities. Signed-off-by: Ricardo Neri Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Len Brown Tested-by: Zhang Rui Link: https://lore.kernel.org/r/20230406203148.19182-4-ricardo.neri-calderon@linux.intel.com --- kernel/sched/fair.c | 33 ++++++++++++--------------------- 1 file changed, 12 insertions(+), 21 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 713d03e73978..a8a02ae7d082 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -9366,12 +9366,9 @@ static bool sched_use_asym_prio(struct sched_domain *sd, int cpu) * the SMT siblings of @sg are busy. If only one CPU in @sg is busy, pull tasks * only if @dst_cpu has higher priority. * - * If both @dst_cpu and @sg have SMT siblings, and @sg has exactly one more - * busy CPU than @sds::local, let @dst_cpu pull tasks if it has higher priority. - * Bigger imbalances in the number of busy CPUs will be dealt with in - * update_sd_pick_busiest(). - * - * If @sg does not have SMT siblings, only pull tasks if @sg has lower priority. + * When dealing with SMT cores, only use priorities if the SMT core has exactly + * one busy sibling. find_busiest_group() will handle bigger imbalances in the + * number of busy CPUs. * * Return: true if @dst_cpu can pull tasks, false otherwise. */ @@ -9380,12 +9377,10 @@ static bool asym_smt_can_pull_tasks(int dst_cpu, struct sd_lb_stats *sds, struct sched_group *sg) { #ifdef CONFIG_SCHED_SMT - bool local_is_smt, sg_is_smt; + bool local_is_smt; int sg_busy_cpus; local_is_smt = sds->local->flags & SD_SHARE_CPUCAPACITY; - sg_is_smt = sg->flags & SD_SHARE_CPUCAPACITY; - sg_busy_cpus = sgs->group_weight - sgs->idle_cpus; if (!local_is_smt) { @@ -9406,21 +9401,17 @@ static bool asym_smt_can_pull_tasks(int dst_cpu, struct sd_lb_stats *sds, return sched_asym_prefer(dst_cpu, sg->asym_prefer_cpu); } - /* @dst_cpu has SMT siblings. */ - - if (sg_is_smt) { - int local_busy_cpus = sds->local->group_weight - - sds->local_stat.idle_cpus; - int busy_cpus_delta = sg_busy_cpus - local_busy_cpus; - - if (busy_cpus_delta == 1) - return sched_asym_prefer(dst_cpu, sg->asym_prefer_cpu); - + /* + * If we are here @dst_cpu has SMT siblings and are also idle. + * + * CPU priorities does not make sense for SMT cores with more than one + * busy sibling. + */ + if (group->flags & SD_SHARE_CPUCAPACITY && sg_busy_cpus != 1) return false; - } - /* If we are here @dst_cpu has SMT siblings and are also idle. */ return sched_asym_prefer(dst_cpu, sg->asym_prefer_cpu); + #else /* Always return false so that callers deal with non-SMT cases. */ return false; -- cgit v1.2.3 From 18ad34532755feb5b9f4284b07769b1bfec18ab3 Mon Sep 17 00:00:00 2001 From: Ricardo Neri Date: Thu, 6 Apr 2023 13:31:40 -0700 Subject: sched/fair: Let low-priority cores help high-priority busy SMT cores Using asym_packing priorities within an SMT core is straightforward. Just follow the priorities that hardware indicates. When balancing load from an SMT core, also consider the idle state of its siblings. Priorities do not reflect that an SMT core divides its throughput among all its busy siblings. They only makes sense when exactly one sibling is busy. Indicate that active balance is needed if the destination CPU has lower priority than the source CPU but the latter has busy SMT siblings. Make find_busiest_queue() not skip higher-priority SMT cores with more than busy sibling. Suggested-by: Valentin Schneider Signed-off-by: Ricardo Neri Signed-off-by: Peter Zijlstra (Intel) Tested-by: Zhang Rui Link: https://lore.kernel.org/r/20230406203148.19182-5-ricardo.neri-calderon@linux.intel.com --- kernel/sched/fair.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index a8a02ae7d082..85ce2494234d 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -10551,8 +10551,15 @@ static struct rq *find_busiest_queue(struct lb_env *env, nr_running == 1) continue; - /* Make sure we only pull tasks from a CPU of lower priority */ + /* + * Make sure we only pull tasks from a CPU of lower priority + * when balancing between SMT siblings. + * + * If balancing between cores, let lower priority CPUs help + * SMT cores with more than one busy sibling. + */ if ((env->sd->flags & SD_ASYM_PACKING) && + sched_use_asym_prio(env->sd, i) && sched_asym_prefer(i, env->dst_cpu) && nr_running == 1) continue; @@ -10645,10 +10652,15 @@ asym_active_balance(struct lb_env *env) * priority CPUs in order to pack all tasks in the highest priority * CPUs. When done between cores, do it only if the whole core if the * whole core is idle. + * + * If @env::src_cpu is an SMT core with busy siblings, let + * the lower priority @env::dst_cpu help it. Do not follow + * CPU priority. */ return env->idle != CPU_NOT_IDLE && (env->sd->flags & SD_ASYM_PACKING) && sched_use_asym_prio(env->sd, env->dst_cpu) && - sched_asym_prefer(env->dst_cpu, env->src_cpu); + (sched_asym_prefer(env->dst_cpu, env->src_cpu) || + !sched_use_asym_prio(env->sd, env->src_cpu)); } static inline bool -- cgit v1.2.3 From 5fd6d7f43958cb62da105c8413eac3e78480f09a Mon Sep 17 00:00:00 2001 From: Ricardo Neri Date: Thu, 6 Apr 2023 13:31:41 -0700 Subject: sched/fair: Keep a fully_busy SMT sched group as busiest When comparing two fully_busy scheduling groups, keep the current busiest group if it represents an SMT core. Tasks in such scheduling group share CPU resources and need more help than tasks in a non-SMT fully_busy group. Signed-off-by: Ricardo Neri Signed-off-by: Peter Zijlstra (Intel) Tested-by: Zhang Rui Link: https://lore.kernel.org/r/20230406203148.19182-6-ricardo.neri-calderon@linux.intel.com --- kernel/sched/fair.c | 16 ++++++++++++++-- 1 file changed, 14 insertions(+), 2 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 85ce2494234d..4a9f04095742 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -9619,10 +9619,22 @@ static bool update_sd_pick_busiest(struct lb_env *env, * contention when accessing shared HW resources. * * XXX for now avg_load is not computed and always 0 so we - * select the 1st one. + * select the 1st one, except if @sg is composed of SMT + * siblings. */ - if (sgs->avg_load <= busiest->avg_load) + + if (sgs->avg_load < busiest->avg_load) return false; + + if (sgs->avg_load == busiest->avg_load) { + /* + * SMT sched groups need more help than non-SMT groups. + * If @sg happens to also be SMT, either choice is good. + */ + if (sds->busiest->flags & SD_SHARE_CPUCAPACITY) + return false; + } + break; case group_has_spare: -- cgit v1.2.3 From 43726bdedd29797d8e1fee2e7300a6d2b9a74ba8 Mon Sep 17 00:00:00 2001 From: Ricardo Neri Date: Thu, 6 Apr 2023 13:31:42 -0700 Subject: sched/fair: Use the busiest group to set prefer_sibling The prefer_sibling setting acts on the busiest group to move excess tasks to the local group. This should be done as per request of the child of the busiest group's sched domain, not the local group's. Using the flags of the child domain of the local group works fortuitously if both groups have child domains. There are cases, however, in which the busiest group's sched domain has child but the local group's does not. Consider, for instance a non-SMT core (or an SMT core with only one online sibling) doing load balance with an SMT core at the MC level. SD_PREFER_SIBLING of the busiest group's child domain will not be honored. We are left with a fully busy SMT core and an idle non-SMT core. Suggested-by: Dietmar Eggemann Signed-off-by: Ricardo Neri Signed-off-by: Peter Zijlstra (Intel) Tested-by: Zhang Rui Link: https://lore.kernel.org/r/20230406203148.19182-7-ricardo.neri-calderon@linux.intel.com --- kernel/sched/fair.c | 15 +++++++++++---- 1 file changed, 11 insertions(+), 4 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 4a9f04095742..3bb89346dbb3 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -10109,7 +10109,6 @@ static void update_idle_cpu_scan(struct lb_env *env, static inline void update_sd_lb_stats(struct lb_env *env, struct sd_lb_stats *sds) { - struct sched_domain *child = env->sd->child; struct sched_group *sg = env->sd->groups; struct sg_lb_stats *local = &sds->local_stat; struct sg_lb_stats tmp_sgs; @@ -10150,8 +10149,13 @@ next_group: sg = sg->next; } while (sg != env->sd->groups); - /* Tag domain that child domain prefers tasks go to siblings first */ - sds->prefer_sibling = child && child->flags & SD_PREFER_SIBLING; + /* + * Indicate that the child domain of the busiest group prefers tasks + * go to a child's sibling domains first. NB the flags of a sched group + * are those of the child domain. + */ + if (sds->busiest) + sds->prefer_sibling = !!(sds->busiest->flags & SD_PREFER_SIBLING); if (env->sd->flags & SD_NUMA) @@ -10461,7 +10465,10 @@ static struct sched_group *find_busiest_group(struct lb_env *env) goto out_balanced; } - /* Try to move all excess tasks to child's sibling domain */ + /* + * Try to move all excess tasks to a sibling domain of the busiest + * group's child domain. + */ if (sds.prefer_sibling && local->group_type == group_has_spare && busiest->sum_nr_running > local->sum_nr_running + 1) goto force_balance; -- cgit v1.2.3 From c9ca07886aaa40225a29e5c1e46ac31d2e14f53a Mon Sep 17 00:00:00 2001 From: Ricardo Neri Date: Thu, 6 Apr 2023 13:31:43 -0700 Subject: sched/fair: Do not even the number of busy CPUs via asym_packing Now that find_busiest_group() triggers load balancing between a fully_ busy SMT2 core and an idle non-SMT core, it is no longer needed to force balancing via asym_packing. Use asym_packing only as intended: when there is high-priority CPU that is idle. After this change, the same logic apply to SMT and non-SMT local groups. It makes less sense having a separate function to deal specifically with SMT. Fold the logic in asym_smt_can_pull_tasks() into sched_asym(). Signed-off-by: Ricardo Neri Signed-off-by: Peter Zijlstra (Intel) Tested-by: Zhang Rui Link: https://lore.kernel.org/r/20230406203148.19182-8-ricardo.neri-calderon@linux.intel.com --- kernel/sched/fair.c | 86 +++++++++++++---------------------------------------- 1 file changed, 21 insertions(+), 65 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 3bb89346dbb3..48b6f0ca13ac 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -9350,74 +9350,26 @@ static bool sched_use_asym_prio(struct sched_domain *sd, int cpu) } /** - * asym_smt_can_pull_tasks - Check whether the load balancing CPU can pull tasks - * @dst_cpu: Destination CPU of the load balancing + * sched_asym - Check if the destination CPU can do asym_packing load balance + * @env: The load balancing environment * @sds: Load-balancing data with statistics of the local group * @sgs: Load-balancing statistics of the candidate busiest group - * @sg: The candidate busiest group + * @group: The candidate busiest group * - * Check the state of the SMT siblings of both @sds::local and @sg and decide - * if @dst_cpu can pull tasks. + * @env::dst_cpu can do asym_packing if it has higher priority than the + * preferred CPU of @group. * - * This function must be called only if all the SMT siblings of @dst_cpu are - * idle, if any. + * SMT is a special case. If we are balancing load between cores, @env::dst_cpu + * can do asym_packing balance only if all its SMT siblings are idle. Also, it + * can only do it if @group is an SMT group and has exactly on busy CPU. Larger + * imbalances in the number of CPUS are dealt with in find_busiest_group(). * - * If @dst_cpu does not have SMT siblings, it can pull tasks if two or more of - * the SMT siblings of @sg are busy. If only one CPU in @sg is busy, pull tasks - * only if @dst_cpu has higher priority. + * If we are balancing load within an SMT core, or at DIE domain level, always + * proceed. * - * When dealing with SMT cores, only use priorities if the SMT core has exactly - * one busy sibling. find_busiest_group() will handle bigger imbalances in the - * number of busy CPUs. - * - * Return: true if @dst_cpu can pull tasks, false otherwise. + * Return: true if @env::dst_cpu can do with asym_packing load balance. False + * otherwise. */ -static bool asym_smt_can_pull_tasks(int dst_cpu, struct sd_lb_stats *sds, - struct sg_lb_stats *sgs, - struct sched_group *sg) -{ -#ifdef CONFIG_SCHED_SMT - bool local_is_smt; - int sg_busy_cpus; - - local_is_smt = sds->local->flags & SD_SHARE_CPUCAPACITY; - sg_busy_cpus = sgs->group_weight - sgs->idle_cpus; - - if (!local_is_smt) { - /* - * If we are here, @dst_cpu is idle and does not have SMT - * siblings. Pull tasks if candidate group has two or more - * busy CPUs. - */ - if (sg_busy_cpus >= 2) /* implies sg_is_smt */ - return true; - - /* - * @dst_cpu does not have SMT siblings. @sg may have SMT - * siblings and only one is busy. In such case, @dst_cpu - * can help if it has higher priority and is idle (i.e., - * it has no running tasks). - */ - return sched_asym_prefer(dst_cpu, sg->asym_prefer_cpu); - } - - /* - * If we are here @dst_cpu has SMT siblings and are also idle. - * - * CPU priorities does not make sense for SMT cores with more than one - * busy sibling. - */ - if (group->flags & SD_SHARE_CPUCAPACITY && sg_busy_cpus != 1) - return false; - - return sched_asym_prefer(dst_cpu, sg->asym_prefer_cpu); - -#else - /* Always return false so that callers deal with non-SMT cases. */ - return false; -#endif -} - static inline bool sched_asym(struct lb_env *env, struct sd_lb_stats *sds, struct sg_lb_stats *sgs, struct sched_group *group) @@ -9426,10 +9378,14 @@ sched_asym(struct lb_env *env, struct sd_lb_stats *sds, struct sg_lb_stats *sgs if (!sched_use_asym_prio(env->sd, env->dst_cpu)) return false; - /* Only do SMT checks if either local or candidate have SMT siblings. */ - if ((sds->local->flags & SD_SHARE_CPUCAPACITY) || - (group->flags & SD_SHARE_CPUCAPACITY)) - return asym_smt_can_pull_tasks(env->dst_cpu, sds, sgs, group); + /* + * CPU priorities does not make sense for SMT cores with more than one + * busy sibling. + */ + if (group->flags & SD_SHARE_CPUCAPACITY) { + if (sgs->group_weight - sgs->idle_cpus != 1) + return false; + } return sched_asym_prefer(env->dst_cpu, group->asym_prefer_cpu); } -- cgit v1.2.3 From 40b4d3dc328265c8ec6688657d74813edf785c83 Mon Sep 17 00:00:00 2001 From: Ricardo Neri Date: Thu, 6 Apr 2023 13:31:44 -0700 Subject: sched/topology: Check SDF_SHARED_CHILD in highest_flag_domain() Do not assume that all the children of a scheduling domain have a given flag. Check whether it has the SDF_SHARED_CHILD meta flag. Suggested-by: Ionela Voinescu Signed-off-by: Ricardo Neri Signed-off-by: Peter Zijlstra (Intel) Link: https://lore.kernel.org/r/20230406203148.19182-9-ricardo.neri-calderon@linux.intel.com --- kernel/sched/sched.h | 22 +++++++++++++++++++--- 1 file changed, 19 insertions(+), 3 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index ec7b3e0a2b20..678446251c35 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -1772,6 +1772,13 @@ queue_balance_callback(struct rq *rq, for (__sd = rcu_dereference_check_sched_domain(cpu_rq(cpu)->sd); \ __sd; __sd = __sd->parent) +/* A mask of all the SD flags that have the SDF_SHARED_CHILD metaflag */ +#define SD_FLAG(name, mflags) (name * !!((mflags) & SDF_SHARED_CHILD)) | +static const unsigned int SD_SHARED_CHILD_MASK = +#include +0; +#undef SD_FLAG + /** * highest_flag_domain - Return highest sched_domain containing flag. * @cpu: The CPU whose highest level of sched domain is to @@ -1779,16 +1786,25 @@ queue_balance_callback(struct rq *rq, * @flag: The flag to check for the highest sched_domain * for the given CPU. * - * Returns the highest sched_domain of a CPU which contains the given flag. + * Returns the highest sched_domain of a CPU which contains @flag. If @flag has + * the SDF_SHARED_CHILD metaflag, all the children domains also have @flag. */ static inline struct sched_domain *highest_flag_domain(int cpu, int flag) { struct sched_domain *sd, *hsd = NULL; for_each_domain(cpu, sd) { - if (!(sd->flags & flag)) + if (sd->flags & flag) { + hsd = sd; + continue; + } + + /* + * Stop the search if @flag is known to be shared at lower + * levels. It will not be found further up. + */ + if (flag & SD_SHARED_CHILD_MASK) break; - hsd = sd; } return hsd; -- cgit v1.2.3 From 519fabc7aaba3f0847cf37d5f9a5740c370eb777 Mon Sep 17 00:00:00 2001 From: Suren Baghdasaryan Date: Thu, 2 Mar 2023 17:13:46 -0800 Subject: psi: remove 500ms min window size limitation for triggers Current 500ms min window size for psi triggers limits polling interval to 50ms to prevent polling threads from using too much cpu bandwidth by polling too frequently. However the number of cgroups with triggers is unlimited, so this protection can be defeated by creating multiple cgroups with psi triggers (triggers in each cgroup are served by a single "psimon" kernel thread). Instead of limiting min polling period, which also limits the latency of psi events, it's better to limit psi trigger creation to authorized users only, like we do for system-wide psi triggers (/proc/pressure/* files can be written only by processes with CAP_SYS_RESOURCE capability). This also makes access rules for cgroup psi files consistent with system-wide ones. Add a CAP_SYS_RESOURCE capability check for cgroup psi file writers and remove the psi window min size limitation. Suggested-by: Sudarshan Rajagopalan Signed-off-by: Suren Baghdasaryan Signed-off-by: Peter Zijlstra (Intel) Acked-by: Michal Hocko Acked-by: Johannes Weiner Link: https://lore.kernel.org/all/cover.1676067791.git.quic_sudaraja@quicinc.com/ --- kernel/cgroup/cgroup.c | 12 ++++++++++++ kernel/sched/psi.c | 4 +--- 2 files changed, 13 insertions(+), 3 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index 625d7483951c..b26ae200abef 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -3877,6 +3877,14 @@ static __poll_t cgroup_pressure_poll(struct kernfs_open_file *of, return psi_trigger_poll(&ctx->psi.trigger, of->file, pt); } +static int cgroup_pressure_open(struct kernfs_open_file *of) +{ + if (of->file->f_mode & FMODE_WRITE && !capable(CAP_SYS_RESOURCE)) + return -EPERM; + + return 0; +} + static void cgroup_pressure_release(struct kernfs_open_file *of) { struct cgroup_file_ctx *ctx = of->priv; @@ -5276,6 +5284,7 @@ static struct cftype cgroup_psi_files[] = { { .name = "io.pressure", .file_offset = offsetof(struct cgroup, psi_files[PSI_IO]), + .open = cgroup_pressure_open, .seq_show = cgroup_io_pressure_show, .write = cgroup_io_pressure_write, .poll = cgroup_pressure_poll, @@ -5284,6 +5293,7 @@ static struct cftype cgroup_psi_files[] = { { .name = "memory.pressure", .file_offset = offsetof(struct cgroup, psi_files[PSI_MEM]), + .open = cgroup_pressure_open, .seq_show = cgroup_memory_pressure_show, .write = cgroup_memory_pressure_write, .poll = cgroup_pressure_poll, @@ -5292,6 +5302,7 @@ static struct cftype cgroup_psi_files[] = { { .name = "cpu.pressure", .file_offset = offsetof(struct cgroup, psi_files[PSI_CPU]), + .open = cgroup_pressure_open, .seq_show = cgroup_cpu_pressure_show, .write = cgroup_cpu_pressure_write, .poll = cgroup_pressure_poll, @@ -5301,6 +5312,7 @@ static struct cftype cgroup_psi_files[] = { { .name = "irq.pressure", .file_offset = offsetof(struct cgroup, psi_files[PSI_IRQ]), + .open = cgroup_pressure_open, .seq_show = cgroup_irq_pressure_show, .write = cgroup_irq_pressure_write, .poll = cgroup_pressure_poll, diff --git a/kernel/sched/psi.c b/kernel/sched/psi.c index e072f6b31bf3..b49af594ad28 100644 --- a/kernel/sched/psi.c +++ b/kernel/sched/psi.c @@ -160,7 +160,6 @@ __setup("psi=", setup_psi); #define EXP_300s 2034 /* 1/exp(2s/300s) */ /* PSI trigger definitions */ -#define WINDOW_MIN_US 500000 /* Min window size is 500ms */ #define WINDOW_MAX_US 10000000 /* Max window size is 10s */ #define UPDATES_PER_WINDOW 10 /* 10 updates per window */ @@ -1305,8 +1304,7 @@ struct psi_trigger *psi_trigger_create(struct psi_group *group, if (state >= PSI_NONIDLE) return ERR_PTR(-EINVAL); - if (window_us < WINDOW_MIN_US || - window_us > WINDOW_MAX_US) + if (window_us == 0 || window_us > WINDOW_MAX_US) return ERR_PTR(-EINVAL); /* -- cgit v1.2.3 From bf2dc42d6beb890c995b8b09f881ef1b37259107 Mon Sep 17 00:00:00 2001 From: Tim C Chen Date: Thu, 4 May 2023 09:09:51 -0700 Subject: sched/topology: Propagate SMT flags when removing degenerate domain When a degenerate cluster domain for core with SMT CPUs is removed, the SD_SHARE_CPUCAPACITY flag in the local child sched group was not propagated to the new parent. We need this flag to properly determine whether the local sched group is SMT. Set the flag in the local child sched group of the new parent sched domain. Signed-off-by: Tim Chen Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Ricardo Neri Link: https://lkml.kernel.org/r/73cf0959eafa53c02e7ef6bf805d751d9190e55d.1683156492.git.tim.c.chen@linux.intel.com --- kernel/sched/topology.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'kernel/sched') diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c index 6682535e37c8..ca4472281c28 100644 --- a/kernel/sched/topology.c +++ b/kernel/sched/topology.c @@ -719,8 +719,13 @@ cpu_attach_domain(struct sched_domain *sd, struct root_domain *rd, int cpu) if (sd_parent_degenerate(tmp, parent)) { tmp->parent = parent->parent; - if (parent->parent) + + if (parent->parent) { parent->parent->child = tmp; + if (tmp->flags & SD_SHARE_CPUCAPACITY) + parent->parent->groups->flags |= SD_SHARE_CPUCAPACITY; + } + /* * Transfer SD_PREFER_SIBLING down in case of a * degenerate parent; the spans match for this -- cgit v1.2.3 From a6fcdd8d95f7486150b3faadfea119fc3dfc3b74 Mon Sep 17 00:00:00 2001 From: "晏艳(采苓)" Date: Sat, 6 May 2023 15:42:53 +0800 Subject: sched/debug: Correct printing for rq->nr_uninterruptible Commit e6fe3f422be1 ("sched: Make multiple runqueue task counters 32-bit") changed the type for rq->nr_uninterruptible from "unsigned long" to "unsigned int", but left wrong cast print to /sys/kernel/debug/sched/debug and to the console. For example, nr_uninterruptible's value is fffffff7 with type "unsigned int", (long)nr_uninterruptible shows 4294967287 while (int)nr_uninterruptible prints -9. So using int cast fixes wrong printing. Signed-off-by: Yan Yan Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20230506074253.44526-1-yanyan.yan@antgroup.com --- kernel/sched/debug.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/sched') diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c index 0b2340a79b65..066ff1c8ae4e 100644 --- a/kernel/sched/debug.c +++ b/kernel/sched/debug.c @@ -777,7 +777,7 @@ static void print_cpu(struct seq_file *m, int cpu) #define P(x) \ do { \ if (sizeof(rq->x) == 4) \ - SEQ_printf(m, " .%-30s: %ld\n", #x, (long)(rq->x)); \ + SEQ_printf(m, " .%-30s: %d\n", #x, (int)(rq->x)); \ else \ SEQ_printf(m, " .%-30s: %Ld\n", #x, (long long)(rq->x));\ } while (0) -- cgit v1.2.3 From 111cd11bbc54850f24191c52ff217da88a5e639b Mon Sep 17 00:00:00 2001 From: Juri Lelli Date: Mon, 8 May 2023 09:58:50 +0200 Subject: sched/cpuset: Bring back cpuset_mutex Turns out percpu_cpuset_rwsem - commit 1243dc518c9d ("cgroup/cpuset: Convert cpuset_mutex to percpu_rwsem") - wasn't such a brilliant idea, as it has been reported to cause slowdowns in workloads that need to change cpuset configuration frequently and it is also not implementing priority inheritance (which causes troubles with realtime workloads). Convert percpu_cpuset_rwsem back to regular cpuset_mutex. Also grab it only for SCHED_DEADLINE tasks (other policies don't care about stable cpusets anyway). Signed-off-by: Juri Lelli Reviewed-by: Waiman Long Signed-off-by: Tejun Heo --- include/linux/cpuset.h | 8 +-- kernel/cgroup/cpuset.c | 159 +++++++++++++++++++++++++------------------------ kernel/sched/core.c | 22 ++++--- 3 files changed, 99 insertions(+), 90 deletions(-) (limited to 'kernel/sched') diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h index 980b76a1237e..f90e6325d707 100644 --- a/include/linux/cpuset.h +++ b/include/linux/cpuset.h @@ -71,8 +71,8 @@ extern void cpuset_init_smp(void); extern void cpuset_force_rebuild(void); extern void cpuset_update_active_cpus(void); extern void cpuset_wait_for_hotplug(void); -extern void cpuset_read_lock(void); -extern void cpuset_read_unlock(void); +extern void cpuset_lock(void); +extern void cpuset_unlock(void); extern void cpuset_cpus_allowed(struct task_struct *p, struct cpumask *mask); extern bool cpuset_cpus_allowed_fallback(struct task_struct *p); extern nodemask_t cpuset_mems_allowed(struct task_struct *p); @@ -189,8 +189,8 @@ static inline void cpuset_update_active_cpus(void) static inline void cpuset_wait_for_hotplug(void) { } -static inline void cpuset_read_lock(void) { } -static inline void cpuset_read_unlock(void) { } +static inline void cpuset_lock(void) { } +static inline void cpuset_unlock(void) { } static inline void cpuset_cpus_allowed(struct task_struct *p, struct cpumask *mask) diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c index 428ab46291e2..041c0809adaf 100644 --- a/kernel/cgroup/cpuset.c +++ b/kernel/cgroup/cpuset.c @@ -366,22 +366,23 @@ static struct cpuset top_cpuset = { if (is_cpuset_online(((des_cs) = css_cs((pos_css))))) /* - * There are two global locks guarding cpuset structures - cpuset_rwsem and + * There are two global locks guarding cpuset structures - cpuset_mutex and * callback_lock. We also require taking task_lock() when dereferencing a * task's cpuset pointer. See "The task_lock() exception", at the end of this - * comment. The cpuset code uses only cpuset_rwsem write lock. Other - * kernel subsystems can use cpuset_read_lock()/cpuset_read_unlock() to - * prevent change to cpuset structures. + * comment. The cpuset code uses only cpuset_mutex. Other kernel subsystems + * can use cpuset_lock()/cpuset_unlock() to prevent change to cpuset + * structures. Note that cpuset_mutex needs to be a mutex as it is used in + * paths that rely on priority inheritance (e.g. scheduler - on RT) for + * correctness. * * A task must hold both locks to modify cpusets. If a task holds - * cpuset_rwsem, it blocks others wanting that rwsem, ensuring that it - * is the only task able to also acquire callback_lock and be able to - * modify cpusets. It can perform various checks on the cpuset structure - * first, knowing nothing will change. It can also allocate memory while - * just holding cpuset_rwsem. While it is performing these checks, various - * callback routines can briefly acquire callback_lock to query cpusets. - * Once it is ready to make the changes, it takes callback_lock, blocking - * everyone else. + * cpuset_mutex, it blocks others, ensuring that it is the only task able to + * also acquire callback_lock and be able to modify cpusets. It can perform + * various checks on the cpuset structure first, knowing nothing will change. + * It can also allocate memory while just holding cpuset_mutex. While it is + * performing these checks, various callback routines can briefly acquire + * callback_lock to query cpusets. Once it is ready to make the changes, it + * takes callback_lock, blocking everyone else. * * Calls to the kernel memory allocator can not be made while holding * callback_lock, as that would risk double tripping on callback_lock @@ -403,16 +404,16 @@ static struct cpuset top_cpuset = { * guidelines for accessing subsystem state in kernel/cgroup.c */ -DEFINE_STATIC_PERCPU_RWSEM(cpuset_rwsem); +static DEFINE_MUTEX(cpuset_mutex); -void cpuset_read_lock(void) +void cpuset_lock(void) { - percpu_down_read(&cpuset_rwsem); + mutex_lock(&cpuset_mutex); } -void cpuset_read_unlock(void) +void cpuset_unlock(void) { - percpu_up_read(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); } static DEFINE_SPINLOCK(callback_lock); @@ -496,7 +497,7 @@ static inline bool partition_is_populated(struct cpuset *cs, * One way or another, we guarantee to return some non-empty subset * of cpu_online_mask. * - * Call with callback_lock or cpuset_rwsem held. + * Call with callback_lock or cpuset_mutex held. */ static void guarantee_online_cpus(struct task_struct *tsk, struct cpumask *pmask) @@ -538,7 +539,7 @@ out_unlock: * One way or another, we guarantee to return some non-empty subset * of node_states[N_MEMORY]. * - * Call with callback_lock or cpuset_rwsem held. + * Call with callback_lock or cpuset_mutex held. */ static void guarantee_online_mems(struct cpuset *cs, nodemask_t *pmask) { @@ -550,7 +551,7 @@ static void guarantee_online_mems(struct cpuset *cs, nodemask_t *pmask) /* * update task's spread flag if cpuset's page/slab spread flag is set * - * Call with callback_lock or cpuset_rwsem held. The check can be skipped + * Call with callback_lock or cpuset_mutex held. The check can be skipped * if on default hierarchy. */ static void cpuset_update_task_spread_flags(struct cpuset *cs, @@ -575,7 +576,7 @@ static void cpuset_update_task_spread_flags(struct cpuset *cs, * * One cpuset is a subset of another if all its allowed CPUs and * Memory Nodes are a subset of the other, and its exclusive flags - * are only set if the other's are set. Call holding cpuset_rwsem. + * are only set if the other's are set. Call holding cpuset_mutex. */ static int is_cpuset_subset(const struct cpuset *p, const struct cpuset *q) @@ -713,7 +714,7 @@ out: * If we replaced the flag and mask values of the current cpuset * (cur) with those values in the trial cpuset (trial), would * our various subset and exclusive rules still be valid? Presumes - * cpuset_rwsem held. + * cpuset_mutex held. * * 'cur' is the address of an actual, in-use cpuset. Operations * such as list traversal that depend on the actual address of the @@ -829,7 +830,7 @@ static void update_domain_attr_tree(struct sched_domain_attr *dattr, rcu_read_unlock(); } -/* Must be called with cpuset_rwsem held. */ +/* Must be called with cpuset_mutex held. */ static inline int nr_cpusets(void) { /* jump label reference count + the top-level cpuset */ @@ -855,7 +856,7 @@ static inline int nr_cpusets(void) * domains when operating in the severe memory shortage situations * that could cause allocation failures below. * - * Must be called with cpuset_rwsem held. + * Must be called with cpuset_mutex held. * * The three key local variables below are: * cp - cpuset pointer, used (together with pos_css) to perform a @@ -1084,7 +1085,7 @@ static void dl_rebuild_rd_accounting(void) struct cpuset *cs = NULL; struct cgroup_subsys_state *pos_css; - percpu_rwsem_assert_held(&cpuset_rwsem); + lockdep_assert_held(&cpuset_mutex); lockdep_assert_cpus_held(); lockdep_assert_held(&sched_domains_mutex); @@ -1134,7 +1135,7 @@ partition_and_rebuild_sched_domains(int ndoms_new, cpumask_var_t doms_new[], * 'cpus' is removed, then call this routine to rebuild the * scheduler's dynamic sched domains. * - * Call with cpuset_rwsem held. Takes cpus_read_lock(). + * Call with cpuset_mutex held. Takes cpus_read_lock(). */ static void rebuild_sched_domains_locked(void) { @@ -1145,7 +1146,7 @@ static void rebuild_sched_domains_locked(void) int ndoms; lockdep_assert_cpus_held(); - percpu_rwsem_assert_held(&cpuset_rwsem); + lockdep_assert_held(&cpuset_mutex); /* * If we have raced with CPU hotplug, return early to avoid @@ -1196,9 +1197,9 @@ static void rebuild_sched_domains_locked(void) void rebuild_sched_domains(void) { cpus_read_lock(); - percpu_down_write(&cpuset_rwsem); + mutex_lock(&cpuset_mutex); rebuild_sched_domains_locked(); - percpu_up_write(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); cpus_read_unlock(); } @@ -1208,7 +1209,7 @@ void rebuild_sched_domains(void) * @new_cpus: the temp variable for the new effective_cpus mask * * Iterate through each task of @cs updating its cpus_allowed to the - * effective cpuset's. As this function is called with cpuset_rwsem held, + * effective cpuset's. As this function is called with cpuset_mutex held, * cpuset membership stays stable. For top_cpuset, task_cpu_possible_mask() * is used instead of effective_cpus to make sure all offline CPUs are also * included as hotplug code won't update cpumasks for tasks in top_cpuset. @@ -1322,7 +1323,7 @@ static int update_parent_subparts_cpumask(struct cpuset *cs, int cmd, int old_prs, new_prs; int part_error = PERR_NONE; /* Partition error? */ - percpu_rwsem_assert_held(&cpuset_rwsem); + lockdep_assert_held(&cpuset_mutex); /* * The parent must be a partition root. @@ -1545,7 +1546,7 @@ static int update_parent_subparts_cpumask(struct cpuset *cs, int cmd, * * On legacy hierarchy, effective_cpus will be the same with cpu_allowed. * - * Called with cpuset_rwsem held + * Called with cpuset_mutex held */ static void update_cpumasks_hier(struct cpuset *cs, struct tmpmasks *tmp, bool force) @@ -1705,7 +1706,7 @@ static void update_sibling_cpumasks(struct cpuset *parent, struct cpuset *cs, struct cpuset *sibling; struct cgroup_subsys_state *pos_css; - percpu_rwsem_assert_held(&cpuset_rwsem); + lockdep_assert_held(&cpuset_mutex); /* * Check all its siblings and call update_cpumasks_hier() @@ -1955,12 +1956,12 @@ static void *cpuset_being_rebound; * @cs: the cpuset in which each task's mems_allowed mask needs to be changed * * Iterate through each task of @cs updating its mems_allowed to the - * effective cpuset's. As this function is called with cpuset_rwsem held, + * effective cpuset's. As this function is called with cpuset_mutex held, * cpuset membership stays stable. */ static void update_tasks_nodemask(struct cpuset *cs) { - static nodemask_t newmems; /* protected by cpuset_rwsem */ + static nodemask_t newmems; /* protected by cpuset_mutex */ struct css_task_iter it; struct task_struct *task; @@ -1973,7 +1974,7 @@ static void update_tasks_nodemask(struct cpuset *cs) * take while holding tasklist_lock. Forks can happen - the * mpol_dup() cpuset_being_rebound check will catch such forks, * and rebind their vma mempolicies too. Because we still hold - * the global cpuset_rwsem, we know that no other rebind effort + * the global cpuset_mutex, we know that no other rebind effort * will be contending for the global variable cpuset_being_rebound. * It's ok if we rebind the same mm twice; mpol_rebind_mm() * is idempotent. Also migrate pages in each mm to new nodes. @@ -2019,7 +2020,7 @@ static void update_tasks_nodemask(struct cpuset *cs) * * On legacy hierarchy, effective_mems will be the same with mems_allowed. * - * Called with cpuset_rwsem held + * Called with cpuset_mutex held */ static void update_nodemasks_hier(struct cpuset *cs, nodemask_t *new_mems) { @@ -2072,7 +2073,7 @@ static void update_nodemasks_hier(struct cpuset *cs, nodemask_t *new_mems) * mempolicies and if the cpuset is marked 'memory_migrate', * migrate the tasks pages to the new memory. * - * Call with cpuset_rwsem held. May take callback_lock during call. + * Call with cpuset_mutex held. May take callback_lock during call. * Will take tasklist_lock, scan tasklist for tasks in cpuset cs, * lock each such tasks mm->mmap_lock, scan its vma's and rebind * their mempolicies to the cpusets new mems_allowed. @@ -2164,7 +2165,7 @@ static int update_relax_domain_level(struct cpuset *cs, s64 val) * @cs: the cpuset in which each task's spread flags needs to be changed * * Iterate through each task of @cs updating its spread flags. As this - * function is called with cpuset_rwsem held, cpuset membership stays + * function is called with cpuset_mutex held, cpuset membership stays * stable. */ static void update_tasks_flags(struct cpuset *cs) @@ -2184,7 +2185,7 @@ static void update_tasks_flags(struct cpuset *cs) * cs: the cpuset to update * turning_on: whether the flag is being set or cleared * - * Call with cpuset_rwsem held. + * Call with cpuset_mutex held. */ static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, @@ -2234,7 +2235,7 @@ out: * @new_prs: new partition root state * Return: 0 if successful, != 0 if error * - * Call with cpuset_rwsem held. + * Call with cpuset_mutex held. */ static int update_prstate(struct cpuset *cs, int new_prs) { @@ -2472,7 +2473,7 @@ static int cpuset_can_attach_check(struct cpuset *cs) return 0; } -/* Called by cgroups to determine if a cpuset is usable; cpuset_rwsem held */ +/* Called by cgroups to determine if a cpuset is usable; cpuset_mutex held */ static int cpuset_can_attach(struct cgroup_taskset *tset) { struct cgroup_subsys_state *css; @@ -2484,7 +2485,7 @@ static int cpuset_can_attach(struct cgroup_taskset *tset) cpuset_attach_old_cs = task_cs(cgroup_taskset_first(tset, &css)); cs = css_cs(css); - percpu_down_write(&cpuset_rwsem); + mutex_lock(&cpuset_mutex); /* Check to see if task is allowed in the cpuset */ ret = cpuset_can_attach_check(cs); @@ -2506,7 +2507,7 @@ static int cpuset_can_attach(struct cgroup_taskset *tset) */ cs->attach_in_progress++; out_unlock: - percpu_up_write(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); return ret; } @@ -2518,15 +2519,15 @@ static void cpuset_cancel_attach(struct cgroup_taskset *tset) cgroup_taskset_first(tset, &css); cs = css_cs(css); - percpu_down_write(&cpuset_rwsem); + mutex_lock(&cpuset_mutex); cs->attach_in_progress--; if (!cs->attach_in_progress) wake_up(&cpuset_attach_wq); - percpu_up_write(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); } /* - * Protected by cpuset_rwsem. cpus_attach is used only by cpuset_attach_task() + * Protected by cpuset_mutex. cpus_attach is used only by cpuset_attach_task() * but we can't allocate it dynamically there. Define it global and * allocate from cpuset_init(). */ @@ -2535,7 +2536,7 @@ static nodemask_t cpuset_attach_nodemask_to; static void cpuset_attach_task(struct cpuset *cs, struct task_struct *task) { - percpu_rwsem_assert_held(&cpuset_rwsem); + lockdep_assert_held(&cpuset_mutex); if (cs != &top_cpuset) guarantee_online_cpus(task, cpus_attach); @@ -2565,7 +2566,7 @@ static void cpuset_attach(struct cgroup_taskset *tset) cs = css_cs(css); lockdep_assert_cpus_held(); /* see cgroup_attach_lock() */ - percpu_down_write(&cpuset_rwsem); + mutex_lock(&cpuset_mutex); cpus_updated = !cpumask_equal(cs->effective_cpus, oldcs->effective_cpus); mems_updated = !nodes_equal(cs->effective_mems, oldcs->effective_mems); @@ -2626,7 +2627,7 @@ out: if (!cs->attach_in_progress) wake_up(&cpuset_attach_wq); - percpu_up_write(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); } /* The various types of files and directories in a cpuset file system */ @@ -2658,7 +2659,7 @@ static int cpuset_write_u64(struct cgroup_subsys_state *css, struct cftype *cft, int retval = 0; cpus_read_lock(); - percpu_down_write(&cpuset_rwsem); + mutex_lock(&cpuset_mutex); if (!is_cpuset_online(cs)) { retval = -ENODEV; goto out_unlock; @@ -2694,7 +2695,7 @@ static int cpuset_write_u64(struct cgroup_subsys_state *css, struct cftype *cft, break; } out_unlock: - percpu_up_write(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); cpus_read_unlock(); return retval; } @@ -2707,7 +2708,7 @@ static int cpuset_write_s64(struct cgroup_subsys_state *css, struct cftype *cft, int retval = -ENODEV; cpus_read_lock(); - percpu_down_write(&cpuset_rwsem); + mutex_lock(&cpuset_mutex); if (!is_cpuset_online(cs)) goto out_unlock; @@ -2720,7 +2721,7 @@ static int cpuset_write_s64(struct cgroup_subsys_state *css, struct cftype *cft, break; } out_unlock: - percpu_up_write(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); cpus_read_unlock(); return retval; } @@ -2753,7 +2754,7 @@ static ssize_t cpuset_write_resmask(struct kernfs_open_file *of, * operation like this one can lead to a deadlock through kernfs * active_ref protection. Let's break the protection. Losing the * protection is okay as we check whether @cs is online after - * grabbing cpuset_rwsem anyway. This only happens on the legacy + * grabbing cpuset_mutex anyway. This only happens on the legacy * hierarchies. */ css_get(&cs->css); @@ -2761,7 +2762,7 @@ static ssize_t cpuset_write_resmask(struct kernfs_open_file *of, flush_work(&cpuset_hotplug_work); cpus_read_lock(); - percpu_down_write(&cpuset_rwsem); + mutex_lock(&cpuset_mutex); if (!is_cpuset_online(cs)) goto out_unlock; @@ -2785,7 +2786,7 @@ static ssize_t cpuset_write_resmask(struct kernfs_open_file *of, free_cpuset(trialcs); out_unlock: - percpu_up_write(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); cpus_read_unlock(); kernfs_unbreak_active_protection(of->kn); css_put(&cs->css); @@ -2933,13 +2934,13 @@ static ssize_t sched_partition_write(struct kernfs_open_file *of, char *buf, css_get(&cs->css); cpus_read_lock(); - percpu_down_write(&cpuset_rwsem); + mutex_lock(&cpuset_mutex); if (!is_cpuset_online(cs)) goto out_unlock; retval = update_prstate(cs, val); out_unlock: - percpu_up_write(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); cpus_read_unlock(); css_put(&cs->css); return retval ?: nbytes; @@ -3156,7 +3157,7 @@ static int cpuset_css_online(struct cgroup_subsys_state *css) return 0; cpus_read_lock(); - percpu_down_write(&cpuset_rwsem); + mutex_lock(&cpuset_mutex); set_bit(CS_ONLINE, &cs->flags); if (is_spread_page(parent)) @@ -3207,7 +3208,7 @@ static int cpuset_css_online(struct cgroup_subsys_state *css) cpumask_copy(cs->effective_cpus, parent->cpus_allowed); spin_unlock_irq(&callback_lock); out_unlock: - percpu_up_write(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); cpus_read_unlock(); return 0; } @@ -3228,7 +3229,7 @@ static void cpuset_css_offline(struct cgroup_subsys_state *css) struct cpuset *cs = css_cs(css); cpus_read_lock(); - percpu_down_write(&cpuset_rwsem); + mutex_lock(&cpuset_mutex); if (is_partition_valid(cs)) update_prstate(cs, 0); @@ -3247,7 +3248,7 @@ static void cpuset_css_offline(struct cgroup_subsys_state *css) cpuset_dec(); clear_bit(CS_ONLINE, &cs->flags); - percpu_up_write(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); cpus_read_unlock(); } @@ -3260,7 +3261,7 @@ static void cpuset_css_free(struct cgroup_subsys_state *css) static void cpuset_bind(struct cgroup_subsys_state *root_css) { - percpu_down_write(&cpuset_rwsem); + mutex_lock(&cpuset_mutex); spin_lock_irq(&callback_lock); if (is_in_v2_mode()) { @@ -3273,7 +3274,7 @@ static void cpuset_bind(struct cgroup_subsys_state *root_css) } spin_unlock_irq(&callback_lock); - percpu_up_write(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); } /* @@ -3294,7 +3295,7 @@ static int cpuset_can_fork(struct task_struct *task, struct css_set *cset) return 0; lockdep_assert_held(&cgroup_mutex); - percpu_down_write(&cpuset_rwsem); + mutex_lock(&cpuset_mutex); /* Check to see if task is allowed in the cpuset */ ret = cpuset_can_attach_check(cs); @@ -3315,7 +3316,7 @@ static int cpuset_can_fork(struct task_struct *task, struct css_set *cset) */ cs->attach_in_progress++; out_unlock: - percpu_up_write(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); return ret; } @@ -3331,11 +3332,11 @@ static void cpuset_cancel_fork(struct task_struct *task, struct css_set *cset) if (same_cs) return; - percpu_down_write(&cpuset_rwsem); + mutex_lock(&cpuset_mutex); cs->attach_in_progress--; if (!cs->attach_in_progress) wake_up(&cpuset_attach_wq); - percpu_up_write(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); } /* @@ -3363,7 +3364,7 @@ static void cpuset_fork(struct task_struct *task) } /* CLONE_INTO_CGROUP */ - percpu_down_write(&cpuset_rwsem); + mutex_lock(&cpuset_mutex); guarantee_online_mems(cs, &cpuset_attach_nodemask_to); cpuset_attach_task(cs, task); @@ -3371,7 +3372,7 @@ static void cpuset_fork(struct task_struct *task) if (!cs->attach_in_progress) wake_up(&cpuset_attach_wq); - percpu_up_write(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); } struct cgroup_subsys cpuset_cgrp_subsys = { @@ -3472,7 +3473,7 @@ hotplug_update_tasks_legacy(struct cpuset *cs, is_empty = cpumask_empty(cs->cpus_allowed) || nodes_empty(cs->mems_allowed); - percpu_up_write(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); /* * Move tasks to the nearest ancestor with execution resources, @@ -3482,7 +3483,7 @@ hotplug_update_tasks_legacy(struct cpuset *cs, if (is_empty) remove_tasks_in_empty_cpuset(cs); - percpu_down_write(&cpuset_rwsem); + mutex_lock(&cpuset_mutex); } static void @@ -3533,14 +3534,14 @@ static void cpuset_hotplug_update_tasks(struct cpuset *cs, struct tmpmasks *tmp) retry: wait_event(cpuset_attach_wq, cs->attach_in_progress == 0); - percpu_down_write(&cpuset_rwsem); + mutex_lock(&cpuset_mutex); /* * We have raced with task attaching. We wait until attaching * is finished, so we won't attach a task to an empty cpuset. */ if (cs->attach_in_progress) { - percpu_up_write(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); goto retry; } @@ -3637,7 +3638,7 @@ update_tasks: cpus_updated, mems_updated); unlock: - percpu_up_write(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); } /** @@ -3667,7 +3668,7 @@ static void cpuset_hotplug_workfn(struct work_struct *work) if (on_dfl && !alloc_cpumasks(NULL, &tmp)) ptmp = &tmp; - percpu_down_write(&cpuset_rwsem); + mutex_lock(&cpuset_mutex); /* fetch the available cpus/mems and find out which changed how */ cpumask_copy(&new_cpus, cpu_active_mask); @@ -3724,7 +3725,7 @@ static void cpuset_hotplug_workfn(struct work_struct *work) update_tasks_nodemask(&top_cpuset); } - percpu_up_write(&cpuset_rwsem); + mutex_unlock(&cpuset_mutex); /* if cpus or mems changed, we need to propagate to descendants */ if (cpus_updated || mems_updated) { @@ -4155,7 +4156,7 @@ void __cpuset_memory_pressure_bump(void) * - Used for /proc//cpuset. * - No need to task_lock(tsk) on this tsk->cpuset reference, as it * doesn't really matter if tsk->cpuset changes after we read it, - * and we take cpuset_rwsem, keeping cpuset_attach() from changing it + * and we take cpuset_mutex, keeping cpuset_attach() from changing it * anyway. */ int proc_cpuset_show(struct seq_file *m, struct pid_namespace *ns, diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 944c3ae39861..d826bec1c522 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -7590,6 +7590,7 @@ static int __sched_setscheduler(struct task_struct *p, int reset_on_fork; int queue_flags = DEQUEUE_SAVE | DEQUEUE_MOVE | DEQUEUE_NOCLOCK; struct rq *rq; + bool cpuset_locked = false; /* The pi code expects interrupts enabled */ BUG_ON(pi && in_interrupt()); @@ -7639,8 +7640,14 @@ recheck: return retval; } - if (pi) - cpuset_read_lock(); + /* + * SCHED_DEADLINE bandwidth accounting relies on stable cpusets + * information. + */ + if (dl_policy(policy) || dl_policy(p->policy)) { + cpuset_locked = true; + cpuset_lock(); + } /* * Make sure no PI-waiters arrive (or leave) while we are @@ -7716,8 +7723,8 @@ change: if (unlikely(oldpolicy != -1 && oldpolicy != p->policy)) { policy = oldpolicy = -1; task_rq_unlock(rq, p, &rf); - if (pi) - cpuset_read_unlock(); + if (cpuset_locked) + cpuset_unlock(); goto recheck; } @@ -7784,7 +7791,8 @@ change: task_rq_unlock(rq, p, &rf); if (pi) { - cpuset_read_unlock(); + if (cpuset_locked) + cpuset_unlock(); rt_mutex_adjust_pi(p); } @@ -7796,8 +7804,8 @@ change: unlock: task_rq_unlock(rq, p, &rf); - if (pi) - cpuset_read_unlock(); + if (cpuset_locked) + cpuset_unlock(); return retval; } -- cgit v1.2.3 From 6c24849f5515e4966d94fa5279bdff4acf2e9489 Mon Sep 17 00:00:00 2001 From: Juri Lelli Date: Mon, 8 May 2023 09:58:51 +0200 Subject: sched/cpuset: Keep track of SCHED_DEADLINE task in cpusets Qais reported that iterating over all tasks when rebuilding root domains for finding out which ones are DEADLINE and need their bandwidth correctly restored on such root domains can be a costly operation (10+ ms delays on suspend-resume). To fix the problem keep track of the number of DEADLINE tasks belonging to each cpuset and then use this information (followup patch) to only perform the above iteration if DEADLINE tasks are actually present in the cpuset for which a corresponding root domain is being rebuilt. Reported-by: Qais Yousef Link: https://lore.kernel.org/lkml/20230206221428.2125324-1-qyousef@layalina.io/ Signed-off-by: Juri Lelli Reviewed-by: Waiman Long Signed-off-by: Tejun Heo --- include/linux/cpuset.h | 4 ++++ kernel/cgroup/cgroup.c | 4 ++++ kernel/cgroup/cpuset.c | 25 +++++++++++++++++++++++++ kernel/sched/deadline.c | 14 ++++++++++++++ 4 files changed, 47 insertions(+) (limited to 'kernel/sched') diff --git a/include/linux/cpuset.h b/include/linux/cpuset.h index f90e6325d707..d629094fac6e 100644 --- a/include/linux/cpuset.h +++ b/include/linux/cpuset.h @@ -71,6 +71,8 @@ extern void cpuset_init_smp(void); extern void cpuset_force_rebuild(void); extern void cpuset_update_active_cpus(void); extern void cpuset_wait_for_hotplug(void); +extern void inc_dl_tasks_cs(struct task_struct *task); +extern void dec_dl_tasks_cs(struct task_struct *task); extern void cpuset_lock(void); extern void cpuset_unlock(void); extern void cpuset_cpus_allowed(struct task_struct *p, struct cpumask *mask); @@ -189,6 +191,8 @@ static inline void cpuset_update_active_cpus(void) static inline void cpuset_wait_for_hotplug(void) { } +static inline void inc_dl_tasks_cs(struct task_struct *task) { } +static inline void dec_dl_tasks_cs(struct task_struct *task) { } static inline void cpuset_lock(void) { } static inline void cpuset_unlock(void) { } diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index 625d7483951c..9d809191a54f 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -57,6 +57,7 @@ #include #include #include +#include #include #include @@ -6683,6 +6684,9 @@ void cgroup_exit(struct task_struct *tsk) list_add_tail(&tsk->cg_list, &cset->dying_tasks); cset->nr_tasks--; + if (dl_task(tsk)) + dec_dl_tasks_cs(tsk); + WARN_ON_ONCE(cgroup_task_frozen(tsk)); if (unlikely(!(tsk->flags & PF_KTHREAD) && test_bit(CGRP_FREEZE, &task_dfl_cgroup(tsk)->flags))) diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c index 041c0809adaf..ca195ff8b298 100644 --- a/kernel/cgroup/cpuset.c +++ b/kernel/cgroup/cpuset.c @@ -193,6 +193,12 @@ struct cpuset { int use_parent_ecpus; int child_ecpus_count; + /* + * number of SCHED_DEADLINE tasks attached to this cpuset, so that we + * know when to rebuild associated root domain bandwidth information. + */ + int nr_deadline_tasks; + /* Invalid partition error code, not lock protected */ enum prs_errcode prs_err; @@ -245,6 +251,20 @@ static inline struct cpuset *parent_cs(struct cpuset *cs) return css_cs(cs->css.parent); } +void inc_dl_tasks_cs(struct task_struct *p) +{ + struct cpuset *cs = task_cs(p); + + cs->nr_deadline_tasks++; +} + +void dec_dl_tasks_cs(struct task_struct *p) +{ + struct cpuset *cs = task_cs(p); + + cs->nr_deadline_tasks--; +} + /* bits in struct cpuset flags field */ typedef enum { CS_ONLINE, @@ -2499,6 +2519,11 @@ static int cpuset_can_attach(struct cgroup_taskset *tset) ret = security_task_setscheduler(task); if (ret) goto out_unlock; + + if (dl_task(task)) { + cs->nr_deadline_tasks++; + cpuset_attach_old_cs->nr_deadline_tasks--; + } } /* diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index 5a9a4b81c972..e11de074a6fd 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -16,6 +16,8 @@ * Fabio Checconi */ +#include + /* * Default limits for DL period; on the top end we guard against small util * tasks still getting ridiculously long effective runtimes, on the bottom end we @@ -2596,6 +2598,12 @@ static void switched_from_dl(struct rq *rq, struct task_struct *p) if (task_on_rq_queued(p) && p->dl.dl_runtime) task_non_contending(p); + /* + * In case a task is setscheduled out from SCHED_DEADLINE we need to + * keep track of that on its cpuset (for correct bandwidth tracking). + */ + dec_dl_tasks_cs(p); + if (!task_on_rq_queued(p)) { /* * Inactive timer is armed. However, p is leaving DEADLINE and @@ -2636,6 +2644,12 @@ static void switched_to_dl(struct rq *rq, struct task_struct *p) if (hrtimer_try_to_cancel(&p->dl.inactive_timer) == 1) put_task_struct(p); + /* + * In case a task is setscheduled to SCHED_DEADLINE we need to keep + * track of that on its cpuset (for correct bandwidth tracking). + */ + inc_dl_tasks_cs(p); + /* If p is not queued we will update its parameters at next wakeup. */ if (!task_on_rq_queued(p)) { add_rq_bw(&p->dl, &rq->dl); -- cgit v1.2.3 From 85989106feb734437e2d598b639991b9185a43a6 Mon Sep 17 00:00:00 2001 From: Dietmar Eggemann Date: Mon, 8 May 2023 09:58:53 +0200 Subject: sched/deadline: Create DL BW alloc, free & check overflow interface While moving a set of tasks between exclusive cpusets, cpuset_can_attach() -> task_can_attach() calls dl_cpu_busy(..., p) for DL BW overflow checking and per-task DL BW allocation on the destination root_domain for the DL tasks in this set. This approach has the issue of not freeing already allocated DL BW in the following error cases: (1) The set of tasks includes multiple DL tasks and DL BW overflow checking fails for one of the subsequent DL tasks. (2) Another controller next to the cpuset controller which is attached to the same cgroup fails in its can_attach(). To address this problem rework dl_cpu_busy(): (1) Split it into dl_bw_check_overflow() & dl_bw_alloc() and add a dedicated dl_bw_free(). (2) dl_bw_alloc() & dl_bw_free() take a `u64 dl_bw` parameter instead of a `struct task_struct *p` used in dl_cpu_busy(). This allows to allocate DL BW for a set of tasks too rather than only for a single task. Signed-off-by: Dietmar Eggemann Signed-off-by: Juri Lelli Signed-off-by: Tejun Heo --- include/linux/sched.h | 2 ++ kernel/sched/core.c | 4 ++-- kernel/sched/deadline.c | 53 +++++++++++++++++++++++++++++++++++++------------ kernel/sched/sched.h | 2 +- 4 files changed, 45 insertions(+), 16 deletions(-) (limited to 'kernel/sched') diff --git a/include/linux/sched.h b/include/linux/sched.h index eed5d65b8d1f..0bee06542450 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1853,6 +1853,8 @@ current_restore_flags(unsigned long orig_flags, unsigned long flags) extern int cpuset_cpumask_can_shrink(const struct cpumask *cur, const struct cpumask *trial); extern int task_can_attach(struct task_struct *p, const struct cpumask *cs_effective_cpus); +extern int dl_bw_alloc(int cpu, u64 dl_bw); +extern void dl_bw_free(int cpu, u64 dl_bw); #ifdef CONFIG_SMP extern void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask); extern int set_cpus_allowed_ptr(struct task_struct *p, const struct cpumask *new_mask); diff --git a/kernel/sched/core.c b/kernel/sched/core.c index d826bec1c522..df659892d7d5 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -9319,7 +9319,7 @@ int task_can_attach(struct task_struct *p, if (unlikely(cpu >= nr_cpu_ids)) return -EINVAL; - ret = dl_cpu_busy(cpu, p); + ret = dl_bw_alloc(cpu, p->dl.dl_bw); } out: @@ -9604,7 +9604,7 @@ static void cpuset_cpu_active(void) static int cpuset_cpu_inactive(unsigned int cpu) { if (!cpuhp_tasks_frozen) { - int ret = dl_cpu_busy(cpu, NULL); + int ret = dl_bw_check_overflow(cpu); if (ret) return ret; diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index e11de074a6fd..166c3e6eae61 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -3058,26 +3058,38 @@ int dl_cpuset_cpumask_can_shrink(const struct cpumask *cur, return ret; } -int dl_cpu_busy(int cpu, struct task_struct *p) +enum dl_bw_request { + dl_bw_req_check_overflow = 0, + dl_bw_req_alloc, + dl_bw_req_free +}; + +static int dl_bw_manage(enum dl_bw_request req, int cpu, u64 dl_bw) { - unsigned long flags, cap; + unsigned long flags; struct dl_bw *dl_b; - bool overflow; + bool overflow = 0; rcu_read_lock_sched(); dl_b = dl_bw_of(cpu); raw_spin_lock_irqsave(&dl_b->lock, flags); - cap = dl_bw_capacity(cpu); - overflow = __dl_overflow(dl_b, cap, 0, p ? p->dl.dl_bw : 0); - if (!overflow && p) { - /* - * We reserve space for this task in the destination - * root_domain, as we can't fail after this point. - * We will free resources in the source root_domain - * later on (see set_cpus_allowed_dl()). - */ - __dl_add(dl_b, p->dl.dl_bw, dl_bw_cpus(cpu)); + if (req == dl_bw_req_free) { + __dl_sub(dl_b, dl_bw, dl_bw_cpus(cpu)); + } else { + unsigned long cap = dl_bw_capacity(cpu); + + overflow = __dl_overflow(dl_b, cap, 0, dl_bw); + + if (req == dl_bw_req_alloc && !overflow) { + /* + * We reserve space in the destination + * root_domain, as we can't fail after this point. + * We will free resources in the source root_domain + * later on (see set_cpus_allowed_dl()). + */ + __dl_add(dl_b, dl_bw, dl_bw_cpus(cpu)); + } } raw_spin_unlock_irqrestore(&dl_b->lock, flags); @@ -3085,6 +3097,21 @@ int dl_cpu_busy(int cpu, struct task_struct *p) return overflow ? -EBUSY : 0; } + +int dl_bw_check_overflow(int cpu) +{ + return dl_bw_manage(dl_bw_req_check_overflow, cpu, 0); +} + +int dl_bw_alloc(int cpu, u64 dl_bw) +{ + return dl_bw_manage(dl_bw_req_alloc, cpu, dl_bw); +} + +void dl_bw_free(int cpu, u64 dl_bw) +{ + dl_bw_manage(dl_bw_req_free, cpu, dl_bw); +} #endif #ifdef CONFIG_SCHED_DEBUG diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index ec7b3e0a2b20..0ad712811e35 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -330,7 +330,7 @@ extern void __getparam_dl(struct task_struct *p, struct sched_attr *attr); extern bool __checkparam_dl(const struct sched_attr *attr); extern bool dl_param_changed(struct task_struct *p, const struct sched_attr *attr); extern int dl_cpuset_cpumask_can_shrink(const struct cpumask *cur, const struct cpumask *trial); -extern int dl_cpu_busy(int cpu, struct task_struct *p); +extern int dl_bw_check_overflow(int cpu); #ifdef CONFIG_CGROUP_SCHED -- cgit v1.2.3 From 2ef269ef1ac006acf974793d975539244d77b28f Mon Sep 17 00:00:00 2001 From: Dietmar Eggemann Date: Mon, 8 May 2023 09:58:54 +0200 Subject: cgroup/cpuset: Free DL BW in case can_attach() fails cpuset_can_attach() can fail. Postpone DL BW allocation until all tasks have been checked. DL BW is not allocated per-task but as a sum over all DL tasks migrating. If multiple controllers are attached to the cgroup next to the cpuset controller a non-cpuset can_attach() can fail. In this case free DL BW in cpuset_cancel_attach(). Finally, update cpuset DL task count (nr_deadline_tasks) only in cpuset_attach(). Suggested-by: Waiman Long Signed-off-by: Dietmar Eggemann Signed-off-by: Juri Lelli Reviewed-by: Waiman Long Signed-off-by: Tejun Heo --- include/linux/sched.h | 2 +- kernel/cgroup/cpuset.c | 53 +++++++++++++++++++++++++++++++++++++++++++++----- kernel/sched/core.c | 17 ++-------------- 3 files changed, 51 insertions(+), 21 deletions(-) (limited to 'kernel/sched') diff --git a/include/linux/sched.h b/include/linux/sched.h index 0bee06542450..2553918f0b61 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -1852,7 +1852,7 @@ current_restore_flags(unsigned long orig_flags, unsigned long flags) } extern int cpuset_cpumask_can_shrink(const struct cpumask *cur, const struct cpumask *trial); -extern int task_can_attach(struct task_struct *p, const struct cpumask *cs_effective_cpus); +extern int task_can_attach(struct task_struct *p); extern int dl_bw_alloc(int cpu, u64 dl_bw); extern void dl_bw_free(int cpu, u64 dl_bw); #ifdef CONFIG_SMP diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c index b7168970fff2..2c76fcd9f0bc 100644 --- a/kernel/cgroup/cpuset.c +++ b/kernel/cgroup/cpuset.c @@ -198,6 +198,8 @@ struct cpuset { * know when to rebuild associated root domain bandwidth information. */ int nr_deadline_tasks; + int nr_migrate_dl_tasks; + u64 sum_migrate_dl_bw; /* Invalid partition error code, not lock protected */ enum prs_errcode prs_err; @@ -2496,16 +2498,23 @@ static int cpuset_can_attach_check(struct cpuset *cs) return 0; } +static void reset_migrate_dl_data(struct cpuset *cs) +{ + cs->nr_migrate_dl_tasks = 0; + cs->sum_migrate_dl_bw = 0; +} + /* Called by cgroups to determine if a cpuset is usable; cpuset_mutex held */ static int cpuset_can_attach(struct cgroup_taskset *tset) { struct cgroup_subsys_state *css; - struct cpuset *cs; + struct cpuset *cs, *oldcs; struct task_struct *task; int ret; /* used later by cpuset_attach() */ cpuset_attach_old_cs = task_cs(cgroup_taskset_first(tset, &css)); + oldcs = cpuset_attach_old_cs; cs = css_cs(css); mutex_lock(&cpuset_mutex); @@ -2516,7 +2525,7 @@ static int cpuset_can_attach(struct cgroup_taskset *tset) goto out_unlock; cgroup_taskset_for_each(task, css, tset) { - ret = task_can_attach(task, cs->effective_cpus); + ret = task_can_attach(task); if (ret) goto out_unlock; ret = security_task_setscheduler(task); @@ -2524,11 +2533,31 @@ static int cpuset_can_attach(struct cgroup_taskset *tset) goto out_unlock; if (dl_task(task)) { - cs->nr_deadline_tasks++; - cpuset_attach_old_cs->nr_deadline_tasks--; + cs->nr_migrate_dl_tasks++; + cs->sum_migrate_dl_bw += task->dl.dl_bw; } } + if (!cs->nr_migrate_dl_tasks) + goto out_success; + + if (!cpumask_intersects(oldcs->effective_cpus, cs->effective_cpus)) { + int cpu = cpumask_any_and(cpu_active_mask, cs->effective_cpus); + + if (unlikely(cpu >= nr_cpu_ids)) { + reset_migrate_dl_data(cs); + ret = -EINVAL; + goto out_unlock; + } + + ret = dl_bw_alloc(cpu, cs->sum_migrate_dl_bw); + if (ret) { + reset_migrate_dl_data(cs); + goto out_unlock; + } + } + +out_success: /* * Mark attach is in progress. This makes validate_change() fail * changes which zero cpus/mems_allowed. @@ -2551,6 +2580,14 @@ static void cpuset_cancel_attach(struct cgroup_taskset *tset) cs->attach_in_progress--; if (!cs->attach_in_progress) wake_up(&cpuset_attach_wq); + + if (cs->nr_migrate_dl_tasks) { + int cpu = cpumask_any(cs->effective_cpus); + + dl_bw_free(cpu, cs->sum_migrate_dl_bw); + reset_migrate_dl_data(cs); + } + mutex_unlock(&cpuset_mutex); } @@ -2651,6 +2688,12 @@ static void cpuset_attach(struct cgroup_taskset *tset) out: cs->old_mems_allowed = cpuset_attach_nodemask_to; + if (cs->nr_migrate_dl_tasks) { + cs->nr_deadline_tasks += cs->nr_migrate_dl_tasks; + oldcs->nr_deadline_tasks -= cs->nr_migrate_dl_tasks; + reset_migrate_dl_data(cs); + } + cs->attach_in_progress--; if (!cs->attach_in_progress) wake_up(&cpuset_attach_wq); @@ -3330,7 +3373,7 @@ static int cpuset_can_fork(struct task_struct *task, struct css_set *cset) if (ret) goto out_unlock; - ret = task_can_attach(task, cs->effective_cpus); + ret = task_can_attach(task); if (ret) goto out_unlock; diff --git a/kernel/sched/core.c b/kernel/sched/core.c index df659892d7d5..ed0d7381b2ec 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -9294,8 +9294,7 @@ int cpuset_cpumask_can_shrink(const struct cpumask *cur, return ret; } -int task_can_attach(struct task_struct *p, - const struct cpumask *cs_effective_cpus) +int task_can_attach(struct task_struct *p) { int ret = 0; @@ -9308,21 +9307,9 @@ int task_can_attach(struct task_struct *p, * success of set_cpus_allowed_ptr() on all attached tasks * before cpus_mask may be changed. */ - if (p->flags & PF_NO_SETAFFINITY) { + if (p->flags & PF_NO_SETAFFINITY) ret = -EINVAL; - goto out; - } - - if (dl_task(p) && !cpumask_intersects(task_rq(p)->rd->span, - cs_effective_cpus)) { - int cpu = cpumask_any_and(cpu_active_mask, cs_effective_cpus); - if (unlikely(cpu >= nr_cpu_ids)) - return -EINVAL; - ret = dl_bw_alloc(cpu, p->dl.dl_bw); - } - -out: return ret; } -- cgit v1.2.3 From 616db8779b1e3f93075df691432cccc5ef3c3ba0 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 17 May 2023 17:02:08 -1000 Subject: workqueue: Automatically mark CPU-hogging work items CPU_INTENSIVE If a per-cpu work item hogs the CPU, it can prevent other work items from starting through concurrency management. A per-cpu workqueue which intends to host such CPU-hogging work items can choose to not participate in concurrency management by setting %WQ_CPU_INTENSIVE; however, this can be error-prone and difficult to debug when missed. This patch adds an automatic CPU usage based detection. If a concurrency-managed work item consumes more CPU time than the threshold (10ms by default) continuously without intervening sleeps, wq_worker_tick() which is called from scheduler_tick() will detect the condition and automatically mark it CPU_INTENSIVE. The mechanism isn't foolproof: * Detection depends on tick hitting the work item. Getting preempted at the right timings may allow a violating work item to evade detection at least temporarily. * nohz_full CPUs may not be running ticks and thus can fail detection. * Even when detection is working, the 10ms detection delays can add up if many CPU-hogging work items are queued at the same time. However, in vast majority of cases, this should be able to detect violations reliably and provide reasonable protection with a small increase in code complexity. If some work items trigger this condition repeatedly, the bigger problem likely is the CPU being saturated with such per-cpu work items and the solution would be making them UNBOUND. The next patch will add a debug mechanism to help spot such cases. v4: Documentation for workqueue.cpu_intensive_thresh_us added to kernel-parameters.txt. v3: Switch to use wq_worker_tick() instead of hooking into preemptions as suggested by Peter. v2: Lai pointed out that wq_worker_stopping() also needs to be called from preemption and rtlock paths and an earlier patch was updated accordingly. This patch adds a comment describing the risk of infinte recursions and how they're avoided. Signed-off-by: Tejun Heo Acked-by: Peter Zijlstra Cc: Linus Torvalds Cc: Lai Jiangshan --- Documentation/admin-guide/kernel-parameters.txt | 7 +++ Documentation/core-api/workqueue.rst | 38 +++++++------- kernel/sched/core.c | 3 ++ kernel/workqueue.c | 68 +++++++++++++++++++++++-- kernel/workqueue_internal.h | 2 + tools/workqueue/wq_monitor.py | 13 ++++- 6 files changed, 106 insertions(+), 25 deletions(-) (limited to 'kernel/sched') diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt index 9e5bab29685f..1f2185cf2f0a 100644 --- a/Documentation/admin-guide/kernel-parameters.txt +++ b/Documentation/admin-guide/kernel-parameters.txt @@ -6931,6 +6931,13 @@ it can be updated at runtime by writing to the corresponding sysfs file. + workqueue.cpu_intensive_thresh_us= + Per-cpu work items which run for longer than this + threshold are automatically considered CPU intensive + and excluded from concurrency management to prevent + them from noticeably delaying other per-cpu work + items. Default is 10000 (10ms). + workqueue.disable_numa By default, all work items queued to unbound workqueues are affine to the NUMA nodes they're diff --git a/Documentation/core-api/workqueue.rst b/Documentation/core-api/workqueue.rst index 7e5c39310bbf..a389f31b025c 100644 --- a/Documentation/core-api/workqueue.rst +++ b/Documentation/core-api/workqueue.rst @@ -354,25 +354,25 @@ Monitoring Use tools/workqueue/wq_monitor.py to monitor workqueue operations: :: $ tools/workqueue/wq_monitor.py events - total infl CMwake mayday rescued - events 18545 0 5 - - - events_highpri 8 0 0 - - - events_long 3 0 0 - - - events_unbound 38306 0 - - - - events_freezable 0 0 0 - - - events_power_efficient 29598 0 0 - - - events_freezable_power_ 10 0 0 - - - sock_diag_events 0 0 0 - - - - total infl CMwake mayday rescued - events 18548 0 5 - - - events_highpri 8 0 0 - - - events_long 3 0 0 - - - events_unbound 38322 0 - - - - events_freezable 0 0 0 - - - events_power_efficient 29603 0 0 - - - events_freezable_power_ 10 0 0 - - - sock_diag_events 0 0 0 - - + total infl CPUitsv CMwake mayday rescued + events 18545 0 0 5 - - + events_highpri 8 0 0 0 - - + events_long 3 0 0 0 - - + events_unbound 38306 0 - - - - + events_freezable 0 0 0 0 - - + events_power_efficient 29598 0 0 0 - - + events_freezable_power_ 10 0 0 0 - - + sock_diag_events 0 0 0 0 - - + + total infl CPUitsv CMwake mayday rescued + events 18548 0 0 5 - - + events_highpri 8 0 0 0 - - + events_long 3 0 0 0 - - + events_unbound 38322 0 - - - - + events_freezable 0 0 0 0 - - + events_power_efficient 29603 0 0 0 - - + events_freezable_power_ 10 0 0 0 - - + sock_diag_events 0 0 0 0 - - ... diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 944c3ae39861..3484cada9a4a 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -5632,6 +5632,9 @@ void scheduler_tick(void) perf_event_task_tick(); + if (curr->flags & PF_WQ_WORKER) + wq_worker_tick(curr); + #ifdef CONFIG_SMP rq->idle_balance = idle_cpu(cpu); trigger_load_balance(rq); diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 942421443603..3dc83d5eba50 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -212,6 +212,7 @@ struct worker_pool { enum pool_workqueue_stats { PWQ_STAT_STARTED, /* work items started execution */ PWQ_STAT_COMPLETED, /* work items completed execution */ + PWQ_STAT_CPU_INTENSIVE, /* wq_cpu_intensive_thresh_us violations */ PWQ_STAT_CM_WAKEUP, /* concurrency-management worker wakeups */ PWQ_STAT_MAYDAY, /* maydays to rescuer */ PWQ_STAT_RESCUED, /* linked work items executed by rescuer */ @@ -332,6 +333,14 @@ static struct kmem_cache *pwq_cache; static cpumask_var_t *wq_numa_possible_cpumask; /* possible CPUs of each node */ +/* + * Per-cpu work items which run for longer than the following threshold are + * automatically considered CPU intensive and excluded from concurrency + * management to prevent them from noticeably delaying other per-cpu work items. + */ +static unsigned long wq_cpu_intensive_thresh_us = 10000; +module_param_named(cpu_intensive_thresh_us, wq_cpu_intensive_thresh_us, ulong, 0644); + static bool wq_disable_numa; module_param_named(disable_numa, wq_disable_numa, bool, 0444); @@ -962,6 +971,13 @@ void wq_worker_running(struct task_struct *task) if (!(worker->flags & WORKER_NOT_RUNNING)) worker->pool->nr_running++; preempt_enable(); + + /* + * CPU intensive auto-detection cares about how long a work item hogged + * CPU without sleeping. Reset the starting timestamp on wakeup. + */ + worker->current_at = worker->task->se.sum_exec_runtime; + worker->sleeping = 0; } @@ -1012,6 +1028,45 @@ void wq_worker_sleeping(struct task_struct *task) raw_spin_unlock_irq(&pool->lock); } +/** + * wq_worker_tick - a scheduler tick occurred while a kworker is running + * @task: task currently running + * + * Called from scheduler_tick(). We're in the IRQ context and the current + * worker's fields which follow the 'K' locking rule can be accessed safely. + */ +void wq_worker_tick(struct task_struct *task) +{ + struct worker *worker = kthread_data(task); + struct pool_workqueue *pwq = worker->current_pwq; + struct worker_pool *pool = worker->pool; + + if (!pwq) + return; + + /* + * If the current worker is concurrency managed and hogged the CPU for + * longer than wq_cpu_intensive_thresh_us, it's automatically marked + * CPU_INTENSIVE to avoid stalling other concurrency-managed work items. + */ + if ((worker->flags & WORKER_NOT_RUNNING) || + worker->task->se.sum_exec_runtime - worker->current_at < + wq_cpu_intensive_thresh_us * NSEC_PER_USEC) + return; + + raw_spin_lock(&pool->lock); + + worker_set_flags(worker, WORKER_CPU_INTENSIVE); + pwq->stats[PWQ_STAT_CPU_INTENSIVE]++; + + if (need_more_worker(pool)) { + pwq->stats[PWQ_STAT_CM_WAKEUP]++; + wake_up_worker(pool); + } + + raw_spin_unlock(&pool->lock); +} + /** * wq_worker_last_func - retrieve worker's last work function * @task: Task to retrieve last work function of. @@ -2327,7 +2382,6 @@ __acquires(&pool->lock) { struct pool_workqueue *pwq = get_work_pwq(work); struct worker_pool *pool = worker->pool; - bool cpu_intensive = pwq->wq->flags & WQ_CPU_INTENSIVE; unsigned long work_data; struct worker *collision; #ifdef CONFIG_LOCKDEP @@ -2364,6 +2418,7 @@ __acquires(&pool->lock) worker->current_work = work; worker->current_func = work->func; worker->current_pwq = pwq; + worker->current_at = worker->task->se.sum_exec_runtime; work_data = *work_data_bits(work); worker->current_color = get_work_color(work_data); @@ -2381,7 +2436,7 @@ __acquires(&pool->lock) * of concurrency management and the next code block will chain * execution of the pending work items. */ - if (unlikely(cpu_intensive)) + if (unlikely(pwq->wq->flags & WQ_CPU_INTENSIVE)) worker_set_flags(worker, WORKER_CPU_INTENSIVE); /* @@ -2461,9 +2516,12 @@ __acquires(&pool->lock) raw_spin_lock_irq(&pool->lock); - /* clear cpu intensive status */ - if (unlikely(cpu_intensive)) - worker_clr_flags(worker, WORKER_CPU_INTENSIVE); + /* + * In addition to %WQ_CPU_INTENSIVE, @worker may also have been marked + * CPU intensive by wq_worker_tick() if @work hogged CPU longer than + * wq_cpu_intensive_thresh_us. Clear it. + */ + worker_clr_flags(worker, WORKER_CPU_INTENSIVE); /* tag the worker for identification in schedule() */ worker->last_func = worker->current_func; diff --git a/kernel/workqueue_internal.h b/kernel/workqueue_internal.h index c2455be7b4c2..6b1d66e28269 100644 --- a/kernel/workqueue_internal.h +++ b/kernel/workqueue_internal.h @@ -31,6 +31,7 @@ struct worker { struct work_struct *current_work; /* K: work being processed and its */ work_func_t current_func; /* K: function */ struct pool_workqueue *current_pwq; /* K: pwq */ + u64 current_at; /* K: runtime at start or last wakeup */ unsigned int current_color; /* K: color */ int sleeping; /* S: is worker sleeping? */ @@ -76,6 +77,7 @@ static inline struct worker *current_wq_worker(void) */ void wq_worker_running(struct task_struct *task); void wq_worker_sleeping(struct task_struct *task); +void wq_worker_tick(struct task_struct *task); work_func_t wq_worker_last_func(struct task_struct *task); #endif /* _KERNEL_WORKQUEUE_INTERNAL_H */ diff --git a/tools/workqueue/wq_monitor.py b/tools/workqueue/wq_monitor.py index fc1643ba06b3..7c6f523b9164 100644 --- a/tools/workqueue/wq_monitor.py +++ b/tools/workqueue/wq_monitor.py @@ -11,6 +11,11 @@ https://github.com/osandov/drgn. infl The number of currently in-flight work items. + CPUitsv The number of times a concurrency-managed work item hogged CPU + longer than the threshold (workqueue.cpu_intensive_thresh_us) + and got excluded from concurrency management to avoid stalling + other work items. + CMwake The number of concurrency-management wake-ups while executing a work item of the workqueue. @@ -53,6 +58,7 @@ WQ_MEM_RECLAIM = prog['WQ_MEM_RECLAIM'] PWQ_STAT_STARTED = prog['PWQ_STAT_STARTED'] # work items started execution PWQ_STAT_COMPLETED = prog['PWQ_STAT_COMPLETED'] # work items completed execution +PWQ_STAT_CPU_INTENSIVE = prog['PWQ_STAT_CPU_INTENSIVE'] # wq_cpu_intensive_thresh_us violations PWQ_STAT_CM_WAKEUP = prog['PWQ_STAT_CM_WAKEUP'] # concurrency-management worker wakeups PWQ_STAT_MAYDAY = prog['PWQ_STAT_MAYDAY'] # maydays to rescuer PWQ_STAT_RESCUED = prog['PWQ_STAT_RESCUED'] # linked work items executed by rescuer @@ -75,19 +81,23 @@ class WqStats: 'mem_reclaim' : self.mem_reclaim, 'started' : self.stats[PWQ_STAT_STARTED], 'completed' : self.stats[PWQ_STAT_COMPLETED], + 'cpu_intensive' : self.stats[PWQ_STAT_CPU_INTENSIVE], 'cm_wakeup' : self.stats[PWQ_STAT_CM_WAKEUP], 'mayday' : self.stats[PWQ_STAT_MAYDAY], 'rescued' : self.stats[PWQ_STAT_RESCUED], } def table_header_str(): - return f'{"":>24} {"total":>8} {"infl":>5} {"CMwake":>7} {"mayday":>7} {"rescued":>7}' + return f'{"":>24} {"total":>8} {"infl":>5} '\ + f'{"CPUitsv":>7} {"CMwake":>7} {"mayday":>7} {"rescued":>7}' def table_row_str(self): + cpu_intensive = '-' cm_wakeup = '-' mayday = '-' rescued = '-' if not self.unbound: + cpu_intensive = str(self.stats[PWQ_STAT_CPU_INTENSIVE]) cm_wakeup = str(self.stats[PWQ_STAT_CM_WAKEUP]) if self.mem_reclaim: @@ -97,6 +107,7 @@ class WqStats: out = f'{self.name[-24:]:24} ' \ f'{self.stats[PWQ_STAT_STARTED]:8} ' \ f'{max(self.stats[PWQ_STAT_STARTED] - self.stats[PWQ_STAT_COMPLETED], 0):5} ' \ + f'{cpu_intensive:>7} ' \ f'{cm_wakeup:>7} ' \ f'{mayday:>7} ' \ f'{rescued:>7} ' -- cgit v1.2.3 From e2a1f85bf9f509afd09b5d3308e3489b65845c28 Mon Sep 17 00:00:00 2001 From: Yang Yang Date: Sun, 14 May 2023 09:33:38 -0700 Subject: sched/psi: Avoid resetting the min update period when it is unnecessary Psi_group's poll_min_period is determined by the minimum window size of psi_trigger when creating new triggers. While destroying a psi_trigger, there is no need to reset poll_min_period if the psi_trigger being destroyed did not have the minimum window size, since in this condition poll_min_period will remain the same as before. Signed-off-by: Yang Yang Signed-off-by: Peter Zijlstra (Intel) Acked-by: Suren Baghdasaryan Link: https://lkml.kernel.org/r/20230514163338.834345-1-surenb@google.com --- kernel/sched/psi.c | 15 ++++++++++----- 1 file changed, 10 insertions(+), 5 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/psi.c b/kernel/sched/psi.c index b49af594ad28..81fca77397f6 100644 --- a/kernel/sched/psi.c +++ b/kernel/sched/psi.c @@ -1407,11 +1407,16 @@ void psi_trigger_destroy(struct psi_trigger *t) group->rtpoll_nr_triggers[t->state]--; if (!group->rtpoll_nr_triggers[t->state]) group->rtpoll_states &= ~(1 << t->state); - /* reset min update period for the remaining triggers */ - list_for_each_entry(tmp, &group->rtpoll_triggers, node) - period = min(period, div_u64(tmp->win.size, - UPDATES_PER_WINDOW)); - group->rtpoll_min_period = period; + /* + * Reset min update period for the remaining triggers + * iff the destroying trigger had the min window size. + */ + if (group->rtpoll_min_period == div_u64(t->win.size, UPDATES_PER_WINDOW)) { + list_for_each_entry(tmp, &group->rtpoll_triggers, node) + period = min(period, div_u64(tmp->win.size, + UPDATES_PER_WINDOW)); + group->rtpoll_min_period = period; + } /* Destroy rtpoll_task when the last trigger is destroyed */ if (group->rtpoll_states == 0) { group->rtpoll_until = 0; -- cgit v1.2.3 From d55ebae3f3122b07689cc4c34043114e09ce904c Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 22 May 2023 21:50:17 +0200 Subject: sched: Hide unused sched_update_scaling() This function is only used when CONFIG_SMP is enabled, without that there is no caller and no prototype: kernel/sched/fair.c:688:5: error: no previous prototype for 'sched_update_scaling' [-Werror=missing-prototypes Hide the definition in the same #ifdef check as the declaration. Fixes: 8a99b6833c88 ("sched: Move SCHED_DEBUG sysctl to debugfs") Signed-off-by: Arnd Bergmann Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Vincent Guittot Link: https://lore.kernel.org/r/20230522195021.3456768-2-arnd@kernel.org --- kernel/sched/fair.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 48b6f0ca13ac..2c1b345c3b8d 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -684,7 +684,7 @@ struct sched_entity *__pick_last_entity(struct cfs_rq *cfs_rq) /************************************************************** * Scheduling class statistics methods: */ - +#ifdef CONFIG_SMP int sched_update_scaling(void) { unsigned int factor = get_update_sysctl_factor(); @@ -702,6 +702,7 @@ int sched_update_scaling(void) return 0; } #endif +#endif /* * delta /= w -- cgit v1.2.3 From 378be384e01f13fc44d0adc70873de525586ad74 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 22 May 2023 21:50:18 +0200 Subject: sched: Add schedule_user() declaration The schedule_user() function is used on powerpc and sparc architectures, but only ever called from assembler, so it has no prototype, causing a harmless W=1 warning: kernel/sched/core.c:6730:35: error: no previous prototype for 'schedule_user' [-Werror=missing-prototypes] Add a prototype in sched/sched.h to shut up the warning. Signed-off-by: Arnd Bergmann Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Vincent Guittot Link: https://lore.kernel.org/r/20230522195021.3456768-3-arnd@kernel.org --- kernel/sched/sched.h | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel/sched') diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 678446251c35..192e7816234e 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -2376,6 +2376,7 @@ static inline struct cpuidle_state *idle_get_state(struct rq *rq) #endif extern void schedule_idle(void); +asmlinkage void schedule_user(void); extern void sysrq_sched_debug_show(void); extern void sched_init_granularity(void); -- cgit v1.2.3 From c0bdfd72fbfb7319581bd5bb09b4f10979385bac Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 22 May 2023 21:50:19 +0200 Subject: sched/fair: Hide unused init_cfs_bandwidth() stub init_cfs_bandwidth() is only used when CONFIG_FAIR_GROUP_SCHED is enabled, and without this causes a W=1 warning for the missing prototype: kernel/sched/fair.c:6131:6: error: no previous prototype for 'init_cfs_bandwidth' The normal implementation is only defined for CONFIG_CFS_BANDWIDTH, so the stub exists when CFS_BANDWIDTH is disabled but FAIR_GROUP_SCHED is enabled. Signed-off-by: Arnd Bergmann Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Vincent Guittot Link: https://lore.kernel.org/r/20230522195021.3456768-4-arnd@kernel.org --- kernel/sched/fair.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 2c1b345c3b8d..a7a8ccde3bd7 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -6169,9 +6169,8 @@ static inline int throttled_lb_pair(struct task_group *tg, return 0; } -void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b) {} - #ifdef CONFIG_FAIR_GROUP_SCHED +void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b) {} static void init_cfs_rq_runtime(struct cfs_rq *cfs_rq) {} #endif -- cgit v1.2.3 From f7df852ad6dbb84644e75df7402d9a34f39f31bd Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 22 May 2023 21:50:20 +0200 Subject: sched: Make task_vruntime_update() prototype visible Having the prototype next to the caller but not visible to the callee causes a W=1 warning: kernel/sched/fair.c:11985:6: error: no previous prototype for 'task_vruntime_update' [-Werror=missing-prototypes] Move this to a header, as we do for all other function declarations. Signed-off-by: Arnd Bergmann Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Vincent Guittot Link: https://lore.kernel.org/r/20230522195021.3456768-5-arnd@kernel.org --- kernel/sched/sched.h | 1 + 1 file changed, 1 insertion(+) (limited to 'kernel/sched') diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 192e7816234e..ce07782f0f6c 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -1245,6 +1245,7 @@ static inline raw_spinlock_t *__rq_lockp(struct rq *rq) bool cfs_prio_less(const struct task_struct *a, const struct task_struct *b, bool fi); +void task_vruntime_update(struct rq *rq, struct task_struct *p, bool in_fi); /* * Helpers to check if the CPU's core cookie matches with the task's cookie -- cgit v1.2.3 From 7aa55f2a5902646a19db89dab9961867724b27b8 Mon Sep 17 00:00:00 2001 From: Arnd Bergmann Date: Mon, 22 May 2023 21:50:21 +0200 Subject: sched/fair: Move unused stub functions to header These four functions have a normal definition for CONFIG_FAIR_GROUP_SCHED, and empty one that is only referenced when FAIR_GROUP_SCHED is disabled but CGROUP_SCHED is still enabled. If both are turned off, the functions are still defined but the misisng prototype causes a W=1 warning: kernel/sched/fair.c:12544:6: error: no previous prototype for 'free_fair_sched_group' kernel/sched/fair.c:12546:5: error: no previous prototype for 'alloc_fair_sched_group' kernel/sched/fair.c:12553:6: error: no previous prototype for 'online_fair_sched_group' kernel/sched/fair.c:12555:6: error: no previous prototype for 'unregister_fair_sched_group' Move the alternatives into the header as static inline functions with the correct combination of #ifdef checks to avoid the warning without adding even more complexity. Signed-off-by: Arnd Bergmann Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Vincent Guittot Link: https://lore.kernel.org/r/20230522195021.3456768-6-arnd@kernel.org --- kernel/sched/fair.c | 6 +++--- kernel/sched/sched.h | 2 -- 2 files changed, 3 insertions(+), 5 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index a7a8ccde3bd7..48b6f0ca13ac 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -684,7 +684,7 @@ struct sched_entity *__pick_last_entity(struct cfs_rq *cfs_rq) /************************************************************** * Scheduling class statistics methods: */ -#ifdef CONFIG_SMP + int sched_update_scaling(void) { unsigned int factor = get_update_sysctl_factor(); @@ -702,7 +702,6 @@ int sched_update_scaling(void) return 0; } #endif -#endif /* * delta /= w @@ -6169,8 +6168,9 @@ static inline int throttled_lb_pair(struct task_group *tg, return 0; } -#ifdef CONFIG_FAIR_GROUP_SCHED void init_cfs_bandwidth(struct cfs_bandwidth *cfs_b) {} + +#ifdef CONFIG_FAIR_GROUP_SCHED static void init_cfs_rq_runtime(struct cfs_rq *cfs_rq) {} #endif diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index ce07782f0f6c..678446251c35 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -1245,7 +1245,6 @@ static inline raw_spinlock_t *__rq_lockp(struct rq *rq) bool cfs_prio_less(const struct task_struct *a, const struct task_struct *b, bool fi); -void task_vruntime_update(struct rq *rq, struct task_struct *p, bool in_fi); /* * Helpers to check if the CPU's core cookie matches with the task's cookie @@ -2377,7 +2376,6 @@ static inline struct cpuidle_state *idle_get_state(struct rq *rq) #endif extern void schedule_idle(void); -asmlinkage void schedule_user(void); extern void sysrq_sched_debug_show(void); extern void sched_init_granularity(void); -- cgit v1.2.3 From 3f4bf7aa315bf55b2a569bf77f61ff81c7e11fc1 Mon Sep 17 00:00:00 2001 From: Miaohe Lin Date: Wed, 24 May 2023 18:25:14 +0800 Subject: sched/deadline: remove unused dl_bandwidth The default deadline bandwidth control structure has been removed since commit eb77cf1c151c ("sched/deadline: Remove unused def_dl_bandwidth") leading to unused init_dl_bandwidth() and struct dl_bandwidth. Remove them to clean up the code. Signed-off-by: Miaohe Lin Signed-off-by: Peter Zijlstra (Intel) Acked-by: Juri Lelli Link: https://lore.kernel.org/r/20230524102514.407486-1-linmiaohe@huawei.com --- kernel/sched/deadline.c | 7 ------- kernel/sched/sched.h | 7 ------- 2 files changed, 14 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index 5a9a4b81c972..f827067ad03b 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -489,13 +489,6 @@ static inline int is_leftmost(struct task_struct *p, struct dl_rq *dl_rq) static void init_dl_rq_bw_ratio(struct dl_rq *dl_rq); -void init_dl_bandwidth(struct dl_bandwidth *dl_b, u64 period, u64 runtime) -{ - raw_spin_lock_init(&dl_b->dl_runtime_lock); - dl_b->dl_period = period; - dl_b->dl_runtime = runtime; -} - void init_dl_bw(struct dl_bw *dl_b) { raw_spin_lock_init(&dl_b->lock); diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 678446251c35..d8ba81c66579 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -286,12 +286,6 @@ struct rt_bandwidth { void __dl_clear_params(struct task_struct *p); -struct dl_bandwidth { - raw_spinlock_t dl_runtime_lock; - u64 dl_runtime; - u64 dl_period; -}; - static inline int dl_bandwidth_enabled(void) { return sysctl_sched_rt_runtime >= 0; @@ -2394,7 +2388,6 @@ extern struct rt_bandwidth def_rt_bandwidth; extern void init_rt_bandwidth(struct rt_bandwidth *rt_b, u64 period, u64 runtime); extern bool sched_rt_bandwidth_account(struct rt_rq *rt_rq); -extern void init_dl_bandwidth(struct dl_bandwidth *dl_b, u64 period, u64 runtime); extern void init_dl_task_timer(struct sched_dl_entity *dl_se); extern void init_dl_inactive_task_timer(struct sched_dl_entity *dl_se); -- cgit v1.2.3 From 0f613bfa8268a89be25f2b6b58fc6fe8ccd9a2ba Mon Sep 17 00:00:00 2001 From: Mark Rutland Date: Mon, 5 Jun 2023 08:01:15 +0100 Subject: locking/atomic: treewide: use raw_atomic*_() Now that we have raw_atomic*_() definitions, there's no need to use arch_atomic*_() definitions outside of the low-level atomic definitions. Move treewide users of arch_atomic*_() over to the equivalent raw_atomic*_(). There should be no functional change as a result of this patch. Signed-off-by: Mark Rutland Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Kees Cook Link: https://lore.kernel.org/r/20230605070124.3741859-19-mark.rutland@arm.com --- arch/powerpc/kernel/smp.c | 12 ++++++------ arch/x86/kernel/alternative.c | 4 ++-- arch/x86/kernel/cpu/mce/core.c | 16 ++++++++-------- arch/x86/kernel/nmi.c | 2 +- arch/x86/kernel/pvclock.c | 4 ++-- arch/x86/kvm/x86.c | 2 +- include/asm-generic/bitops/atomic.h | 12 ++++++------ include/asm-generic/bitops/lock.h | 8 ++++---- include/linux/context_tracking.h | 4 ++-- include/linux/context_tracking_state.h | 2 +- include/linux/cpumask.h | 2 +- include/linux/jump_label.h | 2 +- kernel/context_tracking.c | 12 ++++++------ kernel/sched/clock.c | 2 +- 14 files changed, 42 insertions(+), 42 deletions(-) (limited to 'kernel/sched') diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c index 265801a3e94c..e8965f18686f 100644 --- a/arch/powerpc/kernel/smp.c +++ b/arch/powerpc/kernel/smp.c @@ -417,9 +417,9 @@ noinstr static void nmi_ipi_lock_start(unsigned long *flags) { raw_local_irq_save(*flags); hard_irq_disable(); - while (arch_atomic_cmpxchg(&__nmi_ipi_lock, 0, 1) == 1) { + while (raw_atomic_cmpxchg(&__nmi_ipi_lock, 0, 1) == 1) { raw_local_irq_restore(*flags); - spin_until_cond(arch_atomic_read(&__nmi_ipi_lock) == 0); + spin_until_cond(raw_atomic_read(&__nmi_ipi_lock) == 0); raw_local_irq_save(*flags); hard_irq_disable(); } @@ -427,15 +427,15 @@ noinstr static void nmi_ipi_lock_start(unsigned long *flags) noinstr static void nmi_ipi_lock(void) { - while (arch_atomic_cmpxchg(&__nmi_ipi_lock, 0, 1) == 1) - spin_until_cond(arch_atomic_read(&__nmi_ipi_lock) == 0); + while (raw_atomic_cmpxchg(&__nmi_ipi_lock, 0, 1) == 1) + spin_until_cond(raw_atomic_read(&__nmi_ipi_lock) == 0); } noinstr static void nmi_ipi_unlock(void) { smp_mb(); - WARN_ON(arch_atomic_read(&__nmi_ipi_lock) != 1); - arch_atomic_set(&__nmi_ipi_lock, 0); + WARN_ON(raw_atomic_read(&__nmi_ipi_lock) != 1); + raw_atomic_set(&__nmi_ipi_lock, 0); } noinstr static void nmi_ipi_unlock_end(unsigned long *flags) diff --git a/arch/x86/kernel/alternative.c b/arch/x86/kernel/alternative.c index f615e0cb6d93..18f16e93838f 100644 --- a/arch/x86/kernel/alternative.c +++ b/arch/x86/kernel/alternative.c @@ -1799,7 +1799,7 @@ struct bp_patching_desc *try_get_desc(void) { struct bp_patching_desc *desc = &bp_desc; - if (!arch_atomic_inc_not_zero(&desc->refs)) + if (!raw_atomic_inc_not_zero(&desc->refs)) return NULL; return desc; @@ -1810,7 +1810,7 @@ static __always_inline void put_desc(void) struct bp_patching_desc *desc = &bp_desc; smp_mb__before_atomic(); - arch_atomic_dec(&desc->refs); + raw_atomic_dec(&desc->refs); } static __always_inline void *text_poke_addr(struct text_poke_loc *tp) diff --git a/arch/x86/kernel/cpu/mce/core.c b/arch/x86/kernel/cpu/mce/core.c index 2eec60f50057..ab156e6e7120 100644 --- a/arch/x86/kernel/cpu/mce/core.c +++ b/arch/x86/kernel/cpu/mce/core.c @@ -1022,12 +1022,12 @@ static noinstr int mce_start(int *no_way_out) if (!timeout) return ret; - arch_atomic_add(*no_way_out, &global_nwo); + raw_atomic_add(*no_way_out, &global_nwo); /* * Rely on the implied barrier below, such that global_nwo * is updated before mce_callin. */ - order = arch_atomic_inc_return(&mce_callin); + order = raw_atomic_inc_return(&mce_callin); arch_cpumask_clear_cpu(smp_processor_id(), &mce_missing_cpus); /* Enable instrumentation around calls to external facilities */ @@ -1036,10 +1036,10 @@ static noinstr int mce_start(int *no_way_out) /* * Wait for everyone. */ - while (arch_atomic_read(&mce_callin) != num_online_cpus()) { + while (raw_atomic_read(&mce_callin) != num_online_cpus()) { if (mce_timed_out(&timeout, "Timeout: Not all CPUs entered broadcast exception handler")) { - arch_atomic_set(&global_nwo, 0); + raw_atomic_set(&global_nwo, 0); goto out; } ndelay(SPINUNIT); @@ -1054,7 +1054,7 @@ static noinstr int mce_start(int *no_way_out) /* * Monarch: Starts executing now, the others wait. */ - arch_atomic_set(&mce_executing, 1); + raw_atomic_set(&mce_executing, 1); } else { /* * Subject: Now start the scanning loop one by one in @@ -1062,10 +1062,10 @@ static noinstr int mce_start(int *no_way_out) * This way when there are any shared banks it will be * only seen by one CPU before cleared, avoiding duplicates. */ - while (arch_atomic_read(&mce_executing) < order) { + while (raw_atomic_read(&mce_executing) < order) { if (mce_timed_out(&timeout, "Timeout: Subject CPUs unable to finish machine check processing")) { - arch_atomic_set(&global_nwo, 0); + raw_atomic_set(&global_nwo, 0); goto out; } ndelay(SPINUNIT); @@ -1075,7 +1075,7 @@ static noinstr int mce_start(int *no_way_out) /* * Cache the global no_way_out state. */ - *no_way_out = arch_atomic_read(&global_nwo); + *no_way_out = raw_atomic_read(&global_nwo); ret = order; diff --git a/arch/x86/kernel/nmi.c b/arch/x86/kernel/nmi.c index 776f4b1e395b..a0c551846b35 100644 --- a/arch/x86/kernel/nmi.c +++ b/arch/x86/kernel/nmi.c @@ -496,7 +496,7 @@ DEFINE_IDTENTRY_RAW(exc_nmi) */ sev_es_nmi_complete(); if (IS_ENABLED(CONFIG_NMI_CHECK_CPU)) - arch_atomic_long_inc(&nsp->idt_calls); + raw_atomic_long_inc(&nsp->idt_calls); if (IS_ENABLED(CONFIG_SMP) && arch_cpu_is_offline(smp_processor_id())) return; diff --git a/arch/x86/kernel/pvclock.c b/arch/x86/kernel/pvclock.c index 56acf53a782a..b3f81379c2fc 100644 --- a/arch/x86/kernel/pvclock.c +++ b/arch/x86/kernel/pvclock.c @@ -101,11 +101,11 @@ u64 __pvclock_clocksource_read(struct pvclock_vcpu_time_info *src, bool dowd) * updating at the same time, and one of them could be slightly behind, * making the assumption that last_value always go forward fail to hold. */ - last = arch_atomic64_read(&last_value); + last = raw_atomic64_read(&last_value); do { if (ret <= last) return last; - } while (!arch_atomic64_try_cmpxchg(&last_value, &last, ret)); + } while (!raw_atomic64_try_cmpxchg(&last_value, &last, ret)); return ret; } diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index ceb7c5e9cf9e..ac6f60906810 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -13155,7 +13155,7 @@ EXPORT_SYMBOL_GPL(kvm_arch_end_assignment); bool noinstr kvm_arch_has_assigned_device(struct kvm *kvm) { - return arch_atomic_read(&kvm->arch.assigned_device_count); + return raw_atomic_read(&kvm->arch.assigned_device_count); } EXPORT_SYMBOL_GPL(kvm_arch_has_assigned_device); diff --git a/include/asm-generic/bitops/atomic.h b/include/asm-generic/bitops/atomic.h index 71ab4ba9c25d..e076e079f6b2 100644 --- a/include/asm-generic/bitops/atomic.h +++ b/include/asm-generic/bitops/atomic.h @@ -15,21 +15,21 @@ static __always_inline void arch_set_bit(unsigned int nr, volatile unsigned long *p) { p += BIT_WORD(nr); - arch_atomic_long_or(BIT_MASK(nr), (atomic_long_t *)p); + raw_atomic_long_or(BIT_MASK(nr), (atomic_long_t *)p); } static __always_inline void arch_clear_bit(unsigned int nr, volatile unsigned long *p) { p += BIT_WORD(nr); - arch_atomic_long_andnot(BIT_MASK(nr), (atomic_long_t *)p); + raw_atomic_long_andnot(BIT_MASK(nr), (atomic_long_t *)p); } static __always_inline void arch_change_bit(unsigned int nr, volatile unsigned long *p) { p += BIT_WORD(nr); - arch_atomic_long_xor(BIT_MASK(nr), (atomic_long_t *)p); + raw_atomic_long_xor(BIT_MASK(nr), (atomic_long_t *)p); } static __always_inline int @@ -39,7 +39,7 @@ arch_test_and_set_bit(unsigned int nr, volatile unsigned long *p) unsigned long mask = BIT_MASK(nr); p += BIT_WORD(nr); - old = arch_atomic_long_fetch_or(mask, (atomic_long_t *)p); + old = raw_atomic_long_fetch_or(mask, (atomic_long_t *)p); return !!(old & mask); } @@ -50,7 +50,7 @@ arch_test_and_clear_bit(unsigned int nr, volatile unsigned long *p) unsigned long mask = BIT_MASK(nr); p += BIT_WORD(nr); - old = arch_atomic_long_fetch_andnot(mask, (atomic_long_t *)p); + old = raw_atomic_long_fetch_andnot(mask, (atomic_long_t *)p); return !!(old & mask); } @@ -61,7 +61,7 @@ arch_test_and_change_bit(unsigned int nr, volatile unsigned long *p) unsigned long mask = BIT_MASK(nr); p += BIT_WORD(nr); - old = arch_atomic_long_fetch_xor(mask, (atomic_long_t *)p); + old = raw_atomic_long_fetch_xor(mask, (atomic_long_t *)p); return !!(old & mask); } diff --git a/include/asm-generic/bitops/lock.h b/include/asm-generic/bitops/lock.h index 630f2f6b9595..40913516e654 100644 --- a/include/asm-generic/bitops/lock.h +++ b/include/asm-generic/bitops/lock.h @@ -25,7 +25,7 @@ arch_test_and_set_bit_lock(unsigned int nr, volatile unsigned long *p) if (READ_ONCE(*p) & mask) return 1; - old = arch_atomic_long_fetch_or_acquire(mask, (atomic_long_t *)p); + old = raw_atomic_long_fetch_or_acquire(mask, (atomic_long_t *)p); return !!(old & mask); } @@ -41,7 +41,7 @@ static __always_inline void arch_clear_bit_unlock(unsigned int nr, volatile unsigned long *p) { p += BIT_WORD(nr); - arch_atomic_long_fetch_andnot_release(BIT_MASK(nr), (atomic_long_t *)p); + raw_atomic_long_fetch_andnot_release(BIT_MASK(nr), (atomic_long_t *)p); } /** @@ -63,7 +63,7 @@ arch___clear_bit_unlock(unsigned int nr, volatile unsigned long *p) p += BIT_WORD(nr); old = READ_ONCE(*p); old &= ~BIT_MASK(nr); - arch_atomic_long_set_release((atomic_long_t *)p, old); + raw_atomic_long_set_release((atomic_long_t *)p, old); } /** @@ -83,7 +83,7 @@ static inline bool arch_clear_bit_unlock_is_negative_byte(unsigned int nr, unsigned long mask = BIT_MASK(nr); p += BIT_WORD(nr); - old = arch_atomic_long_fetch_andnot_release(mask, (atomic_long_t *)p); + old = raw_atomic_long_fetch_andnot_release(mask, (atomic_long_t *)p); return !!(old & BIT(7)); } #define arch_clear_bit_unlock_is_negative_byte arch_clear_bit_unlock_is_negative_byte diff --git a/include/linux/context_tracking.h b/include/linux/context_tracking.h index d3cbb6c16bab..6e76b9dba00e 100644 --- a/include/linux/context_tracking.h +++ b/include/linux/context_tracking.h @@ -119,7 +119,7 @@ extern void ct_idle_exit(void); */ static __always_inline bool rcu_dynticks_curr_cpu_in_eqs(void) { - return !(arch_atomic_read(this_cpu_ptr(&context_tracking.state)) & RCU_DYNTICKS_IDX); + return !(raw_atomic_read(this_cpu_ptr(&context_tracking.state)) & RCU_DYNTICKS_IDX); } /* @@ -128,7 +128,7 @@ static __always_inline bool rcu_dynticks_curr_cpu_in_eqs(void) */ static __always_inline unsigned long ct_state_inc(int incby) { - return arch_atomic_add_return(incby, this_cpu_ptr(&context_tracking.state)); + return raw_atomic_add_return(incby, this_cpu_ptr(&context_tracking.state)); } static __always_inline bool warn_rcu_enter(void) diff --git a/include/linux/context_tracking_state.h b/include/linux/context_tracking_state.h index fdd537ea513f..bbff5f7f8803 100644 --- a/include/linux/context_tracking_state.h +++ b/include/linux/context_tracking_state.h @@ -51,7 +51,7 @@ DECLARE_PER_CPU(struct context_tracking, context_tracking); #ifdef CONFIG_CONTEXT_TRACKING_USER static __always_inline int __ct_state(void) { - return arch_atomic_read(this_cpu_ptr(&context_tracking.state)) & CT_STATE_MASK; + return raw_atomic_read(this_cpu_ptr(&context_tracking.state)) & CT_STATE_MASK; } #endif diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h index ca736b05ec7b..0d2e2a38b92d 100644 --- a/include/linux/cpumask.h +++ b/include/linux/cpumask.h @@ -1071,7 +1071,7 @@ static inline const struct cpumask *get_cpu_mask(unsigned int cpu) */ static __always_inline unsigned int num_online_cpus(void) { - return arch_atomic_read(&__num_online_cpus); + return raw_atomic_read(&__num_online_cpus); } #define num_possible_cpus() cpumask_weight(cpu_possible_mask) #define num_present_cpus() cpumask_weight(cpu_present_mask) diff --git a/include/linux/jump_label.h b/include/linux/jump_label.h index 4e968ebadce6..f0a949b7c973 100644 --- a/include/linux/jump_label.h +++ b/include/linux/jump_label.h @@ -257,7 +257,7 @@ extern enum jump_label_type jump_label_init_type(struct jump_entry *entry); static __always_inline int static_key_count(struct static_key *key) { - return arch_atomic_read(&key->enabled); + return raw_atomic_read(&key->enabled); } static __always_inline void jump_label_init(void) diff --git a/kernel/context_tracking.c b/kernel/context_tracking.c index a09f1c19336a..6ef0b35fc28c 100644 --- a/kernel/context_tracking.c +++ b/kernel/context_tracking.c @@ -510,7 +510,7 @@ void noinstr __ct_user_enter(enum ctx_state state) * In this we case we don't care about any concurrency/ordering. */ if (!IS_ENABLED(CONFIG_CONTEXT_TRACKING_IDLE)) - arch_atomic_set(&ct->state, state); + raw_atomic_set(&ct->state, state); } else { /* * Even if context tracking is disabled on this CPU, because it's outside @@ -527,7 +527,7 @@ void noinstr __ct_user_enter(enum ctx_state state) */ if (!IS_ENABLED(CONFIG_CONTEXT_TRACKING_IDLE)) { /* Tracking for vtime only, no concurrent RCU EQS accounting */ - arch_atomic_set(&ct->state, state); + raw_atomic_set(&ct->state, state); } else { /* * Tracking for vtime and RCU EQS. Make sure we don't race @@ -535,7 +535,7 @@ void noinstr __ct_user_enter(enum ctx_state state) * RCU only requires RCU_DYNTICKS_IDX increments to be fully * ordered. */ - arch_atomic_add(state, &ct->state); + raw_atomic_add(state, &ct->state); } } } @@ -630,12 +630,12 @@ void noinstr __ct_user_exit(enum ctx_state state) * In this we case we don't care about any concurrency/ordering. */ if (!IS_ENABLED(CONFIG_CONTEXT_TRACKING_IDLE)) - arch_atomic_set(&ct->state, CONTEXT_KERNEL); + raw_atomic_set(&ct->state, CONTEXT_KERNEL); } else { if (!IS_ENABLED(CONFIG_CONTEXT_TRACKING_IDLE)) { /* Tracking for vtime only, no concurrent RCU EQS accounting */ - arch_atomic_set(&ct->state, CONTEXT_KERNEL); + raw_atomic_set(&ct->state, CONTEXT_KERNEL); } else { /* * Tracking for vtime and RCU EQS. Make sure we don't race @@ -643,7 +643,7 @@ void noinstr __ct_user_exit(enum ctx_state state) * RCU only requires RCU_DYNTICKS_IDX increments to be fully * ordered. */ - arch_atomic_sub(state, &ct->state); + raw_atomic_sub(state, &ct->state); } } } diff --git a/kernel/sched/clock.c b/kernel/sched/clock.c index b5cc2b53464d..71443cff31f0 100644 --- a/kernel/sched/clock.c +++ b/kernel/sched/clock.c @@ -287,7 +287,7 @@ again: clock = wrap_max(clock, min_clock); clock = wrap_min(clock, max_clock); - if (!arch_try_cmpxchg64(&scd->clock, &old_clock, clock)) + if (!raw_try_cmpxchg64(&scd->clock, &old_clock, clock)) goto again; return clock; -- cgit v1.2.3 From 0dd37d6dd33a9c23351e6115ae8cdac7863bc7de Mon Sep 17 00:00:00 2001 From: Yicong Yang Date: Tue, 30 May 2023 16:25:07 +0800 Subject: sched/fair: Don't balance task to its current running CPU We've run into the case that the balancer tries to balance a migration disabled task and trigger the warning in set_task_cpu() like below: ------------[ cut here ]------------ WARNING: CPU: 7 PID: 0 at kernel/sched/core.c:3115 set_task_cpu+0x188/0x240 Modules linked in: hclgevf xt_CHECKSUM ipt_REJECT nf_reject_ipv4 <...snip> CPU: 7 PID: 0 Comm: swapper/7 Kdump: loaded Tainted: G O 6.1.0-rc4+ #1 Hardware name: Huawei TaiShan 2280 V2/BC82AMDC, BIOS 2280-V2 CS V5.B221.01 12/09/2021 pstate: 604000c9 (nZCv daIF +PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : set_task_cpu+0x188/0x240 lr : load_balance+0x5d0/0xc60 sp : ffff80000803bc70 x29: ffff80000803bc70 x28: ffff004089e190e8 x27: ffff004089e19040 x26: ffff007effcabc38 x25: 0000000000000000 x24: 0000000000000001 x23: ffff80000803be84 x22: 000000000000000c x21: ffffb093e79e2a78 x20: 000000000000000c x19: ffff004089e19040 x18: 0000000000000000 x17: 0000000000001fad x16: 0000000000000030 x15: 0000000000000000 x14: 0000000000000003 x13: 0000000000000000 x12: 0000000000000000 x11: 0000000000000001 x10: 0000000000000400 x9 : ffffb093e4cee530 x8 : 00000000fffffffe x7 : 0000000000ce168a x6 : 000000000000013e x5 : 00000000ffffffe1 x4 : 0000000000000001 x3 : 0000000000000b2a x2 : 0000000000000b2a x1 : ffffb093e6d6c510 x0 : 0000000000000001 Call trace: set_task_cpu+0x188/0x240 load_balance+0x5d0/0xc60 rebalance_domains+0x26c/0x380 _nohz_idle_balance.isra.0+0x1e0/0x370 run_rebalance_domains+0x6c/0x80 __do_softirq+0x128/0x3d8 ____do_softirq+0x18/0x24 call_on_irq_stack+0x2c/0x38 do_softirq_own_stack+0x24/0x3c __irq_exit_rcu+0xcc/0xf4 irq_exit_rcu+0x18/0x24 el1_interrupt+0x4c/0xe4 el1h_64_irq_handler+0x18/0x2c el1h_64_irq+0x74/0x78 arch_cpu_idle+0x18/0x4c default_idle_call+0x58/0x194 do_idle+0x244/0x2b0 cpu_startup_entry+0x30/0x3c secondary_start_kernel+0x14c/0x190 __secondary_switched+0xb0/0xb4 ---[ end trace 0000000000000000 ]--- Further investigation shows that the warning is superfluous, the migration disabled task is just going to be migrated to its current running CPU. This is because that on load balance if the dst_cpu is not allowed by the task, we'll re-select a new_dst_cpu as a candidate. If no task can be balanced to dst_cpu we'll try to balance the task to the new_dst_cpu instead. In this case when the migration disabled task is not on CPU it only allows to run on its current CPU, load balance will select its current CPU as new_dst_cpu and later triggers the warning above. The new_dst_cpu is chosen from the env->dst_grpmask. Currently it contains CPUs in sched_group_span() and if we have overlapped groups it's possible to run into this case. This patch makes env->dst_grpmask of group_balance_mask() which exclude any CPUs from the busiest group and solve the issue. For balancing in a domain with no overlapped groups the behaviour keeps same as before. Suggested-by: Vincent Guittot Signed-off-by: Yicong Yang Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Vincent Guittot Link: https://lore.kernel.org/r/20230530082507.10444-1-yangyicong@huawei.com --- kernel/sched/fair.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 48b6f0ca13ac..df0ff905a4aa 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -10742,7 +10742,7 @@ static int load_balance(int this_cpu, struct rq *this_rq, .sd = sd, .dst_cpu = this_cpu, .dst_rq = this_rq, - .dst_grpmask = sched_group_span(sd->groups), + .dst_grpmask = group_balance_mask(sd->groups), .idle = idle, .loop_break = SCHED_NR_MIGRATE_BREAK, .cpus = cpus, -- cgit v1.2.3 From d5e1586617be7093ea3419e3fa9387ed833cdbb1 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 2 Jun 2023 10:42:53 +0200 Subject: sched: Unconditionally use full-fat wait_task_inactive() While modifying wait_task_inactive() for PREEMPT_RT; the build robot noted that UP got broken. This led to audit and consideration of the UP implementation of wait_task_inactive(). It looks like the UP implementation is also broken for PREEMPT; consider task_current_syscall() getting preempted between the two calls to wait_task_inactive(). Therefore move the wait_task_inactive() implementation out of CONFIG_SMP and unconditionally use it. Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20230602103731.GA630648%40hirez.programming.kicks-ass.net --- include/linux/sched.h | 7 +- kernel/sched/core.c | 216 +++++++++++++++++++++++++------------------------- 2 files changed, 110 insertions(+), 113 deletions(-) (limited to 'kernel/sched') diff --git a/include/linux/sched.h b/include/linux/sched.h index eed5d65b8d1f..1292d38d66cc 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -2006,15 +2006,12 @@ static __always_inline void scheduler_ipi(void) */ preempt_fold_need_resched(); } -extern unsigned long wait_task_inactive(struct task_struct *, unsigned int match_state); #else static inline void scheduler_ipi(void) { } -static inline unsigned long wait_task_inactive(struct task_struct *p, unsigned int match_state) -{ - return 1; -} #endif +extern unsigned long wait_task_inactive(struct task_struct *, unsigned int match_state); + /* * Set thread flags in other task's structures. * See asm/thread_info.h for TIF_xxxx flags available: diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 944c3ae39861..810cf7dc98cf 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -2213,6 +2213,114 @@ void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags) rq_clock_skip_update(rq); } +/* + * wait_task_inactive - wait for a thread to unschedule. + * + * Wait for the thread to block in any of the states set in @match_state. + * If it changes, i.e. @p might have woken up, then return zero. When we + * succeed in waiting for @p to be off its CPU, we return a positive number + * (its total switch count). If a second call a short while later returns the + * same number, the caller can be sure that @p has remained unscheduled the + * whole time. + * + * The caller must ensure that the task *will* unschedule sometime soon, + * else this function might spin for a *long* time. This function can't + * be called with interrupts off, or it may introduce deadlock with + * smp_call_function() if an IPI is sent by the same process we are + * waiting to become inactive. + */ +unsigned long wait_task_inactive(struct task_struct *p, unsigned int match_state) +{ + int running, queued; + struct rq_flags rf; + unsigned long ncsw; + struct rq *rq; + + for (;;) { + /* + * We do the initial early heuristics without holding + * any task-queue locks at all. We'll only try to get + * the runqueue lock when things look like they will + * work out! + */ + rq = task_rq(p); + + /* + * If the task is actively running on another CPU + * still, just relax and busy-wait without holding + * any locks. + * + * NOTE! Since we don't hold any locks, it's not + * even sure that "rq" stays as the right runqueue! + * But we don't care, since "task_on_cpu()" will + * return false if the runqueue has changed and p + * is actually now running somewhere else! + */ + while (task_on_cpu(rq, p)) { + if (!(READ_ONCE(p->__state) & match_state)) + return 0; + cpu_relax(); + } + + /* + * Ok, time to look more closely! We need the rq + * lock now, to be *sure*. If we're wrong, we'll + * just go back and repeat. + */ + rq = task_rq_lock(p, &rf); + trace_sched_wait_task(p); + running = task_on_cpu(rq, p); + queued = task_on_rq_queued(p); + ncsw = 0; + if (READ_ONCE(p->__state) & match_state) + ncsw = p->nvcsw | LONG_MIN; /* sets MSB */ + task_rq_unlock(rq, p, &rf); + + /* + * If it changed from the expected state, bail out now. + */ + if (unlikely(!ncsw)) + break; + + /* + * Was it really running after all now that we + * checked with the proper locks actually held? + * + * Oops. Go back and try again.. + */ + if (unlikely(running)) { + cpu_relax(); + continue; + } + + /* + * It's not enough that it's not actively running, + * it must be off the runqueue _entirely_, and not + * preempted! + * + * So if it was still runnable (but just not actively + * running right now), it's preempted, and we should + * yield - it could be a while. + */ + if (unlikely(queued)) { + ktime_t to = NSEC_PER_SEC / HZ; + + set_current_state(TASK_UNINTERRUPTIBLE); + schedule_hrtimeout(&to, HRTIMER_MODE_REL_HARD); + continue; + } + + /* + * Ahh, all good. It wasn't running, and it wasn't + * runnable, which means that it will never become + * running in the future either. We're all done! + */ + break; + } + + return ncsw; +} + #ifdef CONFIG_SMP static void @@ -3341,114 +3449,6 @@ out: } #endif /* CONFIG_NUMA_BALANCING */ -/* - * wait_task_inactive - wait for a thread to unschedule. - * - * Wait for the thread to block in any of the states set in @match_state. - * If it changes, i.e. @p might have woken up, then return zero. When we - * succeed in waiting for @p to be off its CPU, we return a positive number - * (its total switch count). If a second call a short while later returns the - * same number, the caller can be sure that @p has remained unscheduled the - * whole time. - * - * The caller must ensure that the task *will* unschedule sometime soon, - * else this function might spin for a *long* time. This function can't - * be called with interrupts off, or it may introduce deadlock with - * smp_call_function() if an IPI is sent by the same process we are - * waiting to become inactive. - */ -unsigned long wait_task_inactive(struct task_struct *p, unsigned int match_state) -{ - int running, queued; - struct rq_flags rf; - unsigned long ncsw; - struct rq *rq; - - for (;;) { - /* - * We do the initial early heuristics without holding - * any task-queue locks at all. We'll only try to get - * the runqueue lock when things look like they will - * work out! - */ - rq = task_rq(p); - - /* - * If the task is actively running on another CPU - * still, just relax and busy-wait without holding - * any locks. - * - * NOTE! Since we don't hold any locks, it's not - * even sure that "rq" stays as the right runqueue! - * But we don't care, since "task_on_cpu()" will - * return false if the runqueue has changed and p - * is actually now running somewhere else! - */ - while (task_on_cpu(rq, p)) { - if (!(READ_ONCE(p->__state) & match_state)) - return 0; - cpu_relax(); - } - - /* - * Ok, time to look more closely! We need the rq - * lock now, to be *sure*. If we're wrong, we'll - * just go back and repeat. - */ - rq = task_rq_lock(p, &rf); - trace_sched_wait_task(p); - running = task_on_cpu(rq, p); - queued = task_on_rq_queued(p); - ncsw = 0; - if (READ_ONCE(p->__state) & match_state) - ncsw = p->nvcsw | LONG_MIN; /* sets MSB */ - task_rq_unlock(rq, p, &rf); - - /* - * If it changed from the expected state, bail out now. - */ - if (unlikely(!ncsw)) - break; - - /* - * Was it really running after all now that we - * checked with the proper locks actually held? - * - * Oops. Go back and try again.. - */ - if (unlikely(running)) { - cpu_relax(); - continue; - } - - /* - * It's not enough that it's not actively running, - * it must be off the runqueue _entirely_, and not - * preempted! - * - * So if it was still runnable (but just not actively - * running right now), it's preempted, and we should - * yield - it could be a while. - */ - if (unlikely(queued)) { - ktime_t to = NSEC_PER_SEC / HZ; - - set_current_state(TASK_UNINTERRUPTIBLE); - schedule_hrtimeout(&to, HRTIMER_MODE_REL_HARD); - continue; - } - - /* - * Ahh, all good. It wasn't running, and it wasn't - * runnable, which means that it will never become - * running in the future either. We're all done! - */ - break; - } - - return ncsw; -} - /*** * kick_process - kick a running thread to enter/exit the kernel * @p: the to-be-kicked thread -- cgit v1.2.3 From 1c06918788e8ae6e69e4381a2806617312922524 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Wed, 31 May 2023 16:39:07 +0200 Subject: sched: Consider task_struct::saved_state in wait_task_inactive() With the introduction of task_struct::saved_state in commit 5f220be21418 ("sched/wakeup: Prepare for RT sleeping spin/rwlocks") matching the task state has gotten more complicated. That same commit changed try_to_wake_up() to consider both states, but wait_task_inactive() has been neglected. Sebastian noted that the wait_task_inactive() usage in ptrace_check_attach() can misbehave when ptrace_stop() is blocked on the tasklist_lock after it sets TASK_TRACED. Therefore extract a common helper from ttwu_state_match() and use that to teach wait_task_inactive() about the PREEMPT_RT locks. Originally-by: Sebastian Andrzej Siewior Signed-off-by: Peter Zijlstra (Intel) Tested-by: Sebastian Andrzej Siewior Link: https://lkml.kernel.org/r/20230601091234.GW83892@hirez.programming.kicks-ass.net --- kernel/sched/core.c | 59 +++++++++++++++++++++++++++++++++++++++++++---------- 1 file changed, 48 insertions(+), 11 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 810cf7dc98cf..ac38225e6d09 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -2213,6 +2213,39 @@ void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags) rq_clock_skip_update(rq); } +static __always_inline +int __task_state_match(struct task_struct *p, unsigned int state) +{ + if (READ_ONCE(p->__state) & state) + return 1; + +#ifdef CONFIG_PREEMPT_RT + if (READ_ONCE(p->saved_state) & state) + return -1; +#endif + return 0; +} + +static __always_inline +int task_state_match(struct task_struct *p, unsigned int state) +{ +#ifdef CONFIG_PREEMPT_RT + int match; + + /* + * Serialize against current_save_and_set_rtlock_wait_state() and + * current_restore_rtlock_saved_state(). + */ + raw_spin_lock_irq(&p->pi_lock); + match = __task_state_match(p, state); + raw_spin_unlock_irq(&p->pi_lock); + + return match; +#else + return __task_state_match(p, state); +#endif +} + /* * wait_task_inactive - wait for a thread to unschedule. * @@ -2231,7 +2264,7 @@ void check_preempt_curr(struct rq *rq, struct task_struct *p, int flags) */ unsigned long wait_task_inactive(struct task_struct *p, unsigned int match_state) { - int running, queued; + int running, queued, match; struct rq_flags rf; unsigned long ncsw; struct rq *rq; @@ -2257,7 +2290,7 @@ unsigned long wait_task_inactive(struct task_struct *p, unsigned int match_state * is actually now running somewhere else! */ while (task_on_cpu(rq, p)) { - if (!(READ_ONCE(p->__state) & match_state)) + if (!task_state_match(p, match_state)) return 0; cpu_relax(); } @@ -2272,8 +2305,15 @@ unsigned long wait_task_inactive(struct task_struct *p, unsigned int match_state running = task_on_cpu(rq, p); queued = task_on_rq_queued(p); ncsw = 0; - if (READ_ONCE(p->__state) & match_state) + if ((match = __task_state_match(p, match_state))) { + /* + * When matching on p->saved_state, consider this task + * still queued so it will wait. + */ + if (match < 0) + queued = 1; ncsw = p->nvcsw | LONG_MIN; /* sets MSB */ + } task_rq_unlock(rq, p, &rf); /* @@ -4003,15 +4043,14 @@ static void ttwu_queue(struct task_struct *p, int cpu, int wake_flags) static __always_inline bool ttwu_state_match(struct task_struct *p, unsigned int state, int *success) { + int match; + if (IS_ENABLED(CONFIG_DEBUG_PREEMPT)) { WARN_ON_ONCE((state & TASK_RTLOCK_WAIT) && state != TASK_RTLOCK_WAIT); } - if (READ_ONCE(p->__state) & state) { - *success = 1; - return true; - } + *success = !!(match = __task_state_match(p, state)); #ifdef CONFIG_PREEMPT_RT /* @@ -4027,12 +4066,10 @@ bool ttwu_state_match(struct task_struct *p, unsigned int state, int *success) * p::saved_state to TASK_RUNNING so any further tests will * not result in false positives vs. @success */ - if (p->saved_state & state) { + if (match < 0) p->saved_state = TASK_RUNNING; - *success = 1; - } #endif - return false; + return match > 0; } /* -- cgit v1.2.3 From fb7d4948c4da2dbd26da4b7ec76bbd2f19ff862a Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 19 May 2023 12:21:10 +0200 Subject: sched/clock: Provide local_clock_noinstr() Now that all ARCH_WANTS_NO_INSTR architectures (arm64, loongarch, s390, x86) provide sched_clock_noinstr(), use this to provide local_clock_noinstr(). This local_clock_noinstr() will be safe to use from noinstr code with the assumption that any such noinstr code is non-preemptible (it had better be, entry code will have IRQs disabled while __cpuidle must have preemption disabled). Specifically, preempt_enable_notrace(), a common part of many a sched_clock() implementation calls out to schedule() -- even though, per the above, it will never trigger -- which frustrates noinstr validation. vmlinux.o: warning: objtool: local_clock+0xb5: call to preempt_schedule_notrace_thunk() leaves .noinstr.text section Signed-off-by: Peter Zijlstra (Intel) Tested-by: Michael Kelley # Hyper-V Link: https://lore.kernel.org/r/20230519102715.978624636@infradead.org --- include/linux/sched/clock.h | 17 ++++++++++++++++- kernel/sched/clock.c | 19 +++++++++++++------ 2 files changed, 29 insertions(+), 7 deletions(-) (limited to 'kernel/sched') diff --git a/include/linux/sched/clock.h b/include/linux/sched/clock.h index ca008f7d3615..196f0ca351a2 100644 --- a/include/linux/sched/clock.h +++ b/include/linux/sched/clock.h @@ -12,7 +12,16 @@ * * Please use one of the three interfaces below. */ -extern unsigned long long notrace sched_clock(void); +extern u64 sched_clock(void); + +#if defined(CONFIG_ARCH_WANTS_NO_INSTR) || defined(CONFIG_GENERIC_SCHED_CLOCK) +extern u64 sched_clock_noinstr(void); +#else +static __always_inline u64 sched_clock_noinstr(void) +{ + return sched_clock(); +} +#endif /* * See the comment in kernel/sched/clock.c @@ -45,6 +54,11 @@ static inline u64 cpu_clock(int cpu) return sched_clock(); } +static __always_inline u64 local_clock_noinstr(void) +{ + return sched_clock_noinstr(); +} + static __always_inline u64 local_clock(void) { return sched_clock(); @@ -79,6 +93,7 @@ static inline u64 cpu_clock(int cpu) return sched_clock_cpu(cpu); } +extern u64 local_clock_noinstr(void); extern u64 local_clock(void); #endif diff --git a/kernel/sched/clock.c b/kernel/sched/clock.c index b5cc2b53464d..5a575a0ba4e6 100644 --- a/kernel/sched/clock.c +++ b/kernel/sched/clock.c @@ -266,7 +266,7 @@ static __always_inline u64 sched_clock_local(struct sched_clock_data *scd) s64 delta; again: - now = sched_clock(); + now = sched_clock_noinstr(); delta = now - scd->tick_raw; if (unlikely(delta < 0)) delta = 0; @@ -293,22 +293,29 @@ again: return clock; } -noinstr u64 local_clock(void) +noinstr u64 local_clock_noinstr(void) { u64 clock; if (static_branch_likely(&__sched_clock_stable)) - return sched_clock() + __sched_clock_offset; + return sched_clock_noinstr() + __sched_clock_offset; if (!static_branch_likely(&sched_clock_running)) - return sched_clock(); + return sched_clock_noinstr(); - preempt_disable_notrace(); clock = sched_clock_local(this_scd()); - preempt_enable_notrace(); return clock; } + +u64 local_clock(void) +{ + u64 now; + preempt_disable_notrace(); + now = local_clock_noinstr(); + preempt_enable_notrace(); + return now; +} EXPORT_SYMBOL_GPL(local_clock); static notrace u64 sched_clock_remote(struct sched_clock_data *scd) -- cgit v1.2.3 From 3eb6d6ececca2fd566d717b37ab467c246f66be7 Mon Sep 17 00:00:00 2001 From: Dietmar Eggemann Date: Mon, 15 May 2023 13:57:34 +0200 Subject: sched/fair: Refactor CPU utilization functions There is a lot of code duplication in cpu_util_next() & cpu_util_cfs(). Remove this by allowing cpu_util_next() to be called with p = NULL. Rename cpu_util_next() to cpu_util() since the '_next' suffix is no longer necessary to distinct cpu utilization related functions. Implement cpu_util_cfs(cpu) as cpu_util(cpu, p = NULL, -1). This will allow to code future related cpu util changes only in one place, namely in cpu_util(). Signed-off-by: Dietmar Eggemann Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Vincent Guittot Link: https://lore.kernel.org/r/20230515115735.296329-2-dietmar.eggemann@arm.com --- kernel/sched/fair.c | 63 ++++++++++++++++++++++++++++++++++++++++------------ kernel/sched/sched.h | 47 +-------------------------------------- 2 files changed, 50 insertions(+), 60 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index df0ff905a4aa..09e3be2e0464 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -7202,11 +7202,41 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) return target; } -/* - * Predicts what cpu_util(@cpu) would return if @p was removed from @cpu - * (@dst_cpu = -1) or migrated to @dst_cpu. - */ -static unsigned long cpu_util_next(int cpu, struct task_struct *p, int dst_cpu) +/** + * cpu_util() - Estimates the amount of CPU capacity used by CFS tasks. + * @cpu: the CPU to get the utilization for + * @p: task for which the CPU utilization should be predicted or NULL + * @dst_cpu: CPU @p migrates to, -1 if @p moves from @cpu or @p == NULL + * + * The unit of the return value must be the same as the one of CPU capacity + * so that CPU utilization can be compared with CPU capacity. + * + * CPU utilization is the sum of running time of runnable tasks plus the + * recent utilization of currently non-runnable tasks on that CPU. + * It represents the amount of CPU capacity currently used by CFS tasks in + * the range [0..max CPU capacity] with max CPU capacity being the CPU + * capacity at f_max. + * + * The estimated CPU utilization is defined as the maximum between CPU + * utilization and sum of the estimated utilization of the currently + * runnable tasks on that CPU. It preserves a utilization "snapshot" of + * previously-executed tasks, which helps better deduce how busy a CPU will + * be when a long-sleeping task wakes up. The contribution to CPU utilization + * of such a task would be significantly decayed at this point of time. + * + * CPU utilization can be higher than the current CPU capacity + * (f_curr/f_max * max CPU capacity) or even the max CPU capacity because + * of rounding errors as well as task migrations or wakeups of new tasks. + * CPU utilization has to be capped to fit into the [0..max CPU capacity] + * range. Otherwise a group of CPUs (CPU0 util = 121% + CPU1 util = 80%) + * could be seen as over-utilized even though CPU1 has 20% of spare CPU + * capacity. CPU utilization is allowed to overshoot current CPU capacity + * though since this is useful for predicting the CPU capacity required + * after task migrations (scheduler-driven DVFS). + * + * Return: (Estimated) utilization for the specified CPU. + */ +static unsigned long cpu_util(int cpu, struct task_struct *p, int dst_cpu) { struct cfs_rq *cfs_rq = &cpu_rq(cpu)->cfs; unsigned long util = READ_ONCE(cfs_rq->avg.util_avg); @@ -7217,9 +7247,9 @@ static unsigned long cpu_util_next(int cpu, struct task_struct *p, int dst_cpu) * contribution. In all the other cases @cpu is not impacted by the * migration so its util_avg is already correct. */ - if (task_cpu(p) == cpu && dst_cpu != cpu) + if (p && task_cpu(p) == cpu && dst_cpu != cpu) lsub_positive(&util, task_util(p)); - else if (task_cpu(p) != cpu && dst_cpu == cpu) + else if (p && task_cpu(p) != cpu && dst_cpu == cpu) util += task_util(p); if (sched_feat(UTIL_EST)) { @@ -7255,7 +7285,7 @@ static unsigned long cpu_util_next(int cpu, struct task_struct *p, int dst_cpu) */ if (dst_cpu == cpu) util_est += _task_util_est(p); - else if (unlikely(task_on_rq_queued(p) || current == p)) + else if (p && unlikely(task_on_rq_queued(p) || current == p)) lsub_positive(&util_est, _task_util_est(p)); util = max(util, util_est); @@ -7264,6 +7294,11 @@ static unsigned long cpu_util_next(int cpu, struct task_struct *p, int dst_cpu) return min(util, capacity_orig_of(cpu)); } +unsigned long cpu_util_cfs(int cpu) +{ + return cpu_util(cpu, NULL, -1); +} + /* * cpu_util_without: compute cpu utilization without any contributions from *p * @cpu: the CPU which utilization is requested @@ -7281,9 +7316,9 @@ static unsigned long cpu_util_without(int cpu, struct task_struct *p) { /* Task has no contribution or is new */ if (cpu != task_cpu(p) || !READ_ONCE(p->se.avg.last_update_time)) - return cpu_util_cfs(cpu); + p = NULL; - return cpu_util_next(cpu, p, -1); + return cpu_util(cpu, p, -1); } /* @@ -7330,7 +7365,7 @@ static inline void eenv_task_busy_time(struct energy_env *eenv, * cpu_capacity. * * The contribution of the task @p for which we want to estimate the - * energy cost is removed (by cpu_util_next()) and must be calculated + * energy cost is removed (by cpu_util()) and must be calculated * separately (see eenv_task_busy_time). This ensures: * * - A stable PD utilization, no matter which CPU of that PD we want to place @@ -7351,7 +7386,7 @@ static inline void eenv_pd_busy_time(struct energy_env *eenv, int cpu; for_each_cpu(cpu, pd_cpus) { - unsigned long util = cpu_util_next(cpu, p, -1); + unsigned long util = cpu_util(cpu, p, -1); busy_time += effective_cpu_util(cpu, util, ENERGY_UTIL, NULL); } @@ -7375,7 +7410,7 @@ eenv_pd_max_util(struct energy_env *eenv, struct cpumask *pd_cpus, for_each_cpu(cpu, pd_cpus) { struct task_struct *tsk = (cpu == dst_cpu) ? p : NULL; - unsigned long util = cpu_util_next(cpu, p, dst_cpu); + unsigned long util = cpu_util(cpu, p, dst_cpu); unsigned long cpu_util; /* @@ -7521,7 +7556,7 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu) if (!cpumask_test_cpu(cpu, p->cpus_ptr)) continue; - util = cpu_util_next(cpu, p, cpu); + util = cpu_util(cpu, p, cpu); cpu_cap = capacity_of(cpu); /* diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index d8ba81c66579..aaf6fc2df6ff 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -2955,53 +2955,8 @@ static inline unsigned long cpu_util_dl(struct rq *rq) return READ_ONCE(rq->avg_dl.util_avg); } -/** - * cpu_util_cfs() - Estimates the amount of CPU capacity used by CFS tasks. - * @cpu: the CPU to get the utilization for. - * - * The unit of the return value must be the same as the one of CPU capacity - * so that CPU utilization can be compared with CPU capacity. - * - * CPU utilization is the sum of running time of runnable tasks plus the - * recent utilization of currently non-runnable tasks on that CPU. - * It represents the amount of CPU capacity currently used by CFS tasks in - * the range [0..max CPU capacity] with max CPU capacity being the CPU - * capacity at f_max. - * - * The estimated CPU utilization is defined as the maximum between CPU - * utilization and sum of the estimated utilization of the currently - * runnable tasks on that CPU. It preserves a utilization "snapshot" of - * previously-executed tasks, which helps better deduce how busy a CPU will - * be when a long-sleeping task wakes up. The contribution to CPU utilization - * of such a task would be significantly decayed at this point of time. - * - * CPU utilization can be higher than the current CPU capacity - * (f_curr/f_max * max CPU capacity) or even the max CPU capacity because - * of rounding errors as well as task migrations or wakeups of new tasks. - * CPU utilization has to be capped to fit into the [0..max CPU capacity] - * range. Otherwise a group of CPUs (CPU0 util = 121% + CPU1 util = 80%) - * could be seen as over-utilized even though CPU1 has 20% of spare CPU - * capacity. CPU utilization is allowed to overshoot current CPU capacity - * though since this is useful for predicting the CPU capacity required - * after task migrations (scheduler-driven DVFS). - * - * Return: (Estimated) utilization for the specified CPU. - */ -static inline unsigned long cpu_util_cfs(int cpu) -{ - struct cfs_rq *cfs_rq; - unsigned long util; - - cfs_rq = &cpu_rq(cpu)->cfs; - util = READ_ONCE(cfs_rq->avg.util_avg); - if (sched_feat(UTIL_EST)) { - util = max_t(unsigned long, util, - READ_ONCE(cfs_rq->avg.util_est.enqueued)); - } - - return min(util, capacity_orig_of(cpu)); -} +extern unsigned long cpu_util_cfs(int cpu); static inline unsigned long cpu_util_rt(struct rq *rq) { -- cgit v1.2.3 From 7d0583cf9ec7bf8e5897dc7d3a7059e8fae5464a Mon Sep 17 00:00:00 2001 From: Dietmar Eggemann Date: Mon, 15 May 2023 13:57:35 +0200 Subject: sched/fair, cpufreq: Introduce 'runnable boosting' The responsiveness of the Per Entity Load Tracking (PELT) util_avg in mobile devices is still considered too low for utilization changes during task ramp-up. In Android this manifests in the fact that the first frames of a UI activity are very prone to be jankframes (a frame which doesn't meet the required frame rendering time, e.g. 16ms@60Hz) since the CPU frequency is normally low at this point and has to ramp up quickly. The beginning of an UI activity is also characterized by the occurrence of CPU contention, especially on little CPUs. Current little CPUs can have an original CPU capacity of only ~ 150 which means that the actual CPU capacity at lower frequency can even be much smaller. Schedutil maps CPU util_avg into CPU frequency request via: util = effective_cpu_util(..., cpu_util_cfs(cpu), ...) -> util = map_util_perf(util) -> freq = map_util_freq(util, ...) CPU contention for CFS tasks can be detected by 'CPU runnable > CPU utililization' in cpu_util_cfs_boost() -> cpu_util(..., boost = 1). Schedutil uses 'runnable boosting' by calling cpu_util_cfs_boost(). To be in sync with schedutil's CPU frequency selection, Energy Aware Scheduling (EAS) also calls cpu_util(..., boost = 1) during max util detection. Moreover, 'runnable boosting' is also used in load-balance for busiest CPU selection when the migration type is 'migrate_util', i.e. only at sched domains which don't have the SD_SHARE_PKG_RESOURCES flag set. Suggested-by: Vincent Guittot Signed-off-by: Dietmar Eggemann Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Vincent Guittot Link: https://lore.kernel.org/r/20230515115735.296329-3-dietmar.eggemann@arm.com --- kernel/sched/cpufreq_schedutil.c | 3 ++- kernel/sched/fair.c | 38 ++++++++++++++++++++++++++++++-------- kernel/sched/sched.h | 1 + 3 files changed, 33 insertions(+), 9 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c index e3211455b203..4492608b7d7f 100644 --- a/kernel/sched/cpufreq_schedutil.c +++ b/kernel/sched/cpufreq_schedutil.c @@ -155,10 +155,11 @@ static unsigned int get_next_freq(struct sugov_policy *sg_policy, static void sugov_get_util(struct sugov_cpu *sg_cpu) { + unsigned long util = cpu_util_cfs_boost(sg_cpu->cpu); struct rq *rq = cpu_rq(sg_cpu->cpu); sg_cpu->bw_dl = cpu_bw_dl(rq); - sg_cpu->util = effective_cpu_util(sg_cpu->cpu, cpu_util_cfs(sg_cpu->cpu), + sg_cpu->util = effective_cpu_util(sg_cpu->cpu, util, FREQUENCY_UTIL, NULL); } diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 09e3be2e0464..6189d1a45635 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -7207,6 +7207,7 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) * @cpu: the CPU to get the utilization for * @p: task for which the CPU utilization should be predicted or NULL * @dst_cpu: CPU @p migrates to, -1 if @p moves from @cpu or @p == NULL + * @boost: 1 to enable boosting, otherwise 0 * * The unit of the return value must be the same as the one of CPU capacity * so that CPU utilization can be compared with CPU capacity. @@ -7224,6 +7225,12 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) * be when a long-sleeping task wakes up. The contribution to CPU utilization * of such a task would be significantly decayed at this point of time. * + * Boosted CPU utilization is defined as max(CPU runnable, CPU utilization). + * CPU contention for CFS tasks can be detected by CPU runnable > CPU + * utilization. Boosting is implemented in cpu_util() so that internal + * users (e.g. EAS) can use it next to external users (e.g. schedutil), + * latter via cpu_util_cfs_boost(). + * * CPU utilization can be higher than the current CPU capacity * (f_curr/f_max * max CPU capacity) or even the max CPU capacity because * of rounding errors as well as task migrations or wakeups of new tasks. @@ -7234,12 +7241,19 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) * though since this is useful for predicting the CPU capacity required * after task migrations (scheduler-driven DVFS). * - * Return: (Estimated) utilization for the specified CPU. + * Return: (Boosted) (estimated) utilization for the specified CPU. */ -static unsigned long cpu_util(int cpu, struct task_struct *p, int dst_cpu) +static unsigned long +cpu_util(int cpu, struct task_struct *p, int dst_cpu, int boost) { struct cfs_rq *cfs_rq = &cpu_rq(cpu)->cfs; unsigned long util = READ_ONCE(cfs_rq->avg.util_avg); + unsigned long runnable; + + if (boost) { + runnable = READ_ONCE(cfs_rq->avg.runnable_avg); + util = max(util, runnable); + } /* * If @dst_cpu is -1 or @p migrates from @cpu to @dst_cpu remove its @@ -7257,6 +7271,9 @@ static unsigned long cpu_util(int cpu, struct task_struct *p, int dst_cpu) util_est = READ_ONCE(cfs_rq->avg.util_est.enqueued); + if (boost) + util_est = max(util_est, runnable); + /* * During wake-up @p isn't enqueued yet and doesn't contribute * to any cpu_rq(cpu)->cfs.avg.util_est.enqueued. @@ -7296,7 +7313,12 @@ static unsigned long cpu_util(int cpu, struct task_struct *p, int dst_cpu) unsigned long cpu_util_cfs(int cpu) { - return cpu_util(cpu, NULL, -1); + return cpu_util(cpu, NULL, -1, 0); +} + +unsigned long cpu_util_cfs_boost(int cpu) +{ + return cpu_util(cpu, NULL, -1, 1); } /* @@ -7318,7 +7340,7 @@ static unsigned long cpu_util_without(int cpu, struct task_struct *p) if (cpu != task_cpu(p) || !READ_ONCE(p->se.avg.last_update_time)) p = NULL; - return cpu_util(cpu, p, -1); + return cpu_util(cpu, p, -1, 0); } /* @@ -7386,7 +7408,7 @@ static inline void eenv_pd_busy_time(struct energy_env *eenv, int cpu; for_each_cpu(cpu, pd_cpus) { - unsigned long util = cpu_util(cpu, p, -1); + unsigned long util = cpu_util(cpu, p, -1, 0); busy_time += effective_cpu_util(cpu, util, ENERGY_UTIL, NULL); } @@ -7410,7 +7432,7 @@ eenv_pd_max_util(struct energy_env *eenv, struct cpumask *pd_cpus, for_each_cpu(cpu, pd_cpus) { struct task_struct *tsk = (cpu == dst_cpu) ? p : NULL; - unsigned long util = cpu_util(cpu, p, dst_cpu); + unsigned long util = cpu_util(cpu, p, dst_cpu, 1); unsigned long cpu_util; /* @@ -7556,7 +7578,7 @@ static int find_energy_efficient_cpu(struct task_struct *p, int prev_cpu) if (!cpumask_test_cpu(cpu, p->cpus_ptr)) continue; - util = cpu_util(cpu, p, cpu); + util = cpu_util(cpu, p, cpu, 0); cpu_cap = capacity_of(cpu); /* @@ -10607,7 +10629,7 @@ static struct rq *find_busiest_queue(struct lb_env *env, break; case migrate_util: - util = cpu_util_cfs(i); + util = cpu_util_cfs_boost(i); /* * Don't try to pull utilization from a CPU with one diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index aaf6fc2df6ff..556496c77dc2 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -2957,6 +2957,7 @@ static inline unsigned long cpu_util_dl(struct rq *rq) extern unsigned long cpu_util_cfs(int cpu); +extern unsigned long cpu_util_cfs_boost(int cpu); static inline unsigned long cpu_util_rt(struct rq *rq) { -- cgit v1.2.3 From a707df30c9438a9d4d0a43ae7f22b59b078f94c4 Mon Sep 17 00:00:00 2001 From: Tom Rix Date: Sun, 11 Jun 2023 08:25:35 -0400 Subject: sched/fair: Rename variable cpu_util eff_util cppcheck reports kernel/sched/fair.c:7436:17: style: Local variable 'cpu_util' shadows outer function [shadowFunction] unsigned long cpu_util; ^ Clean this up by renaming the variable to eff_util Signed-off-by: Tom Rix Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Valentin Schneider Reviewed-by: Dietmar Eggemann Link: https://lore.kernel.org/r/20230611122535.183654-1-trix@redhat.com --- kernel/sched/fair.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 6189d1a45635..7666dbc2b788 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -7433,7 +7433,7 @@ eenv_pd_max_util(struct energy_env *eenv, struct cpumask *pd_cpus, for_each_cpu(cpu, pd_cpus) { struct task_struct *tsk = (cpu == dst_cpu) ? p : NULL; unsigned long util = cpu_util(cpu, p, dst_cpu, 1); - unsigned long cpu_util; + unsigned long eff_util; /* * Performance domain frequency: utilization clamping @@ -7442,8 +7442,8 @@ eenv_pd_max_util(struct energy_env *eenv, struct cpumask *pd_cpus, * NOTE: in case RT tasks are running, by default the * FREQUENCY_UTIL's utilization can be max OPP. */ - cpu_util = effective_cpu_util(cpu, util, FREQUENCY_UTIL, tsk); - max_util = max(max_util, cpu_util); + eff_util = effective_cpu_util(cpu, util, FREQUENCY_UTIL, tsk); + max_util = max(max_util, eff_util); } return min(max_util, eenv->cpu_cap); -- cgit v1.2.3 From 0cce0fde499a92c726cd2e24f7763644f7c9f971 Mon Sep 17 00:00:00 2001 From: Miaohe Lin Date: Sat, 3 Jun 2023 15:36:45 +0800 Subject: sched/topology: Mark set_sched_topology() __init All callers of set_sched_topology() are within __init section. Mark it __init too. Signed-off-by: Miaohe Lin Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Valentin Schneider Link: https://lore.kernel.org/r/20230603073645.1173332-1-linmiaohe@huawei.com --- include/linux/sched/topology.h | 2 +- kernel/sched/topology.c | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) (limited to 'kernel/sched') diff --git a/include/linux/sched/topology.h b/include/linux/sched/topology.h index 816df6cc444e..67b573d5bf28 100644 --- a/include/linux/sched/topology.h +++ b/include/linux/sched/topology.h @@ -203,7 +203,7 @@ struct sched_domain_topology_level { #endif }; -extern void set_sched_topology(struct sched_domain_topology_level *tl); +extern void __init set_sched_topology(struct sched_domain_topology_level *tl); #ifdef CONFIG_SCHED_DEBUG # define SD_INIT_NAME(type) .name = #type diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c index ca4472281c28..cb92dc5f5646 100644 --- a/kernel/sched/topology.c +++ b/kernel/sched/topology.c @@ -1681,7 +1681,7 @@ static struct sched_domain_topology_level *sched_domain_topology_saved; #define for_each_sd_topology(tl) \ for (tl = sched_domain_topology; tl->mask; tl++) -void set_sched_topology(struct sched_domain_topology_level *tl) +void __init set_sched_topology(struct sched_domain_topology_level *tl) { if (WARN_ON_ONCE(sched_smp_initialized)) return; -- cgit v1.2.3 From ef73d6a4ef0b35524125c3cfc6deafc26a0c966a Mon Sep 17 00:00:00 2001 From: Arve Hjønnevåg Date: Fri, 2 Jun 2023 21:23:46 +0000 Subject: sched/wait: Fix a kthread_park race with wait_woken() MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit kthread_park and wait_woken have a similar race that kthread_stop and wait_woken used to have before it was fixed in commit cb6538e740d7 ("sched/wait: Fix a kthread race with wait_woken()"). Extend that fix to also cover kthread_park. [jstultz: Made changes suggested by Peter to optimize memory loads] Signed-off-by: Arve Hjønnevåg Signed-off-by: John Stultz Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Valentin Schneider Link: https://lore.kernel.org/r/20230602212350.535358-1-jstultz@google.com --- include/linux/kthread.h | 1 + kernel/kthread.c | 10 ++++++++++ kernel/sched/wait.c | 7 +------ 3 files changed, 12 insertions(+), 6 deletions(-) (limited to 'kernel/sched') diff --git a/include/linux/kthread.h b/include/linux/kthread.h index 30e5bec81d2b..f1f95a71a4bc 100644 --- a/include/linux/kthread.h +++ b/include/linux/kthread.h @@ -89,6 +89,7 @@ int kthread_stop(struct task_struct *k); bool kthread_should_stop(void); bool kthread_should_park(void); bool __kthread_should_park(struct task_struct *k); +bool kthread_should_stop_or_park(void); bool kthread_freezable_should_stop(bool *was_frozen); void *kthread_func(struct task_struct *k); void *kthread_data(struct task_struct *k); diff --git a/kernel/kthread.c b/kernel/kthread.c index 490792b1066e..07a057086d26 100644 --- a/kernel/kthread.c +++ b/kernel/kthread.c @@ -182,6 +182,16 @@ bool kthread_should_park(void) } EXPORT_SYMBOL_GPL(kthread_should_park); +bool kthread_should_stop_or_park(void) +{ + struct kthread *kthread = __to_kthread(current); + + if (!kthread) + return false; + + return kthread->flags & (BIT(KTHREAD_SHOULD_STOP) | BIT(KTHREAD_SHOULD_PARK)); +} + /** * kthread_freezable_should_stop - should this freezable kthread return now? * @was_frozen: optional out parameter, indicates whether %current was frozen diff --git a/kernel/sched/wait.c b/kernel/sched/wait.c index 133b74730738..48c53e4739ea 100644 --- a/kernel/sched/wait.c +++ b/kernel/sched/wait.c @@ -425,11 +425,6 @@ int autoremove_wake_function(struct wait_queue_entry *wq_entry, unsigned mode, i } EXPORT_SYMBOL(autoremove_wake_function); -static inline bool is_kthread_should_stop(void) -{ - return (current->flags & PF_KTHREAD) && kthread_should_stop(); -} - /* * DEFINE_WAIT_FUNC(wait, woken_wake_func); * @@ -459,7 +454,7 @@ long wait_woken(struct wait_queue_entry *wq_entry, unsigned mode, long timeout) * or woken_wake_function() sees our store to current->state. */ set_current_state(mode); /* A */ - if (!(wq_entry->flags & WQ_FLAG_WOKEN) && !is_kthread_should_stop()) + if (!(wq_entry->flags & WQ_FLAG_WOKEN) && !kthread_should_stop_or_park()) timeout = schedule_timeout(timeout); __set_current_state(TASK_RUNNING); -- cgit v1.2.3 From 6a9d623aad89539eca71eb264db6b9d538620ad5 Mon Sep 17 00:00:00 2001 From: Vineeth Pillai Date: Tue, 30 May 2023 09:55:25 -0400 Subject: sched/deadline: Fix bandwidth reclaim equation in GRUB According to the GRUB[1] rule, the runtime is depreciated as: "dq = -max{u, (1 - Uinact - Uextra)} dt" (1) To guarantee that deadline tasks doesn't starve lower class tasks, we do not allocate the full bandwidth of the cpu to deadline tasks. Maximum bandwidth usable by deadline tasks is denoted by "Umax". Considering Umax, equation (1) becomes: "dq = -(max{u, (Umax - Uinact - Uextra)} / Umax) dt" (2) Current implementation has a minor bug in equation (2), which this patch fixes. The reclamation logic is verified by a sample program which creates multiple deadline threads and observing their utilization. The tests were run on an isolated cpu(isolcpus=3) on a 4 cpu system. Tests on 6.3.0 ============== RUN 1: runtime=7ms, deadline=period=10ms, RT capacity = 95% TID[693]: RECLAIM=1, (r=7ms, d=10ms, p=10ms), Util: 93.33 TID[693]: RECLAIM=1, (r=7ms, d=10ms, p=10ms), Util: 93.35 RUN 2: runtime=1ms, deadline=period=100ms, RT capacity = 95% TID[708]: RECLAIM=1, (r=1ms, d=100ms, p=100ms), Util: 16.69 TID[708]: RECLAIM=1, (r=1ms, d=100ms, p=100ms), Util: 16.69 RUN 3: 2 tasks Task 1: runtime=1ms, deadline=period=10ms Task 2: runtime=1ms, deadline=period=100ms TID[631]: RECLAIM=1, (r=1ms, d=10ms, p=10ms), Util: 62.67 TID[632]: RECLAIM=1, (r=1ms, d=100ms, p=100ms), Util: 6.37 TID[631]: RECLAIM=1, (r=1ms, d=10ms, p=10ms), Util: 62.38 TID[632]: RECLAIM=1, (r=1ms, d=100ms, p=100ms), Util: 6.23 As seen above, the reclamation doesn't reclaim the maximum allowed bandwidth and as the bandwidth of tasks gets smaller, the reclaimed bandwidth also comes down. Tests with this patch applied ============================= RUN 1: runtime=7ms, deadline=period=10ms, RT capacity = 95% TID[608]: RECLAIM=1, (r=7ms, d=10ms, p=10ms), Util: 95.19 TID[608]: RECLAIM=1, (r=7ms, d=10ms, p=10ms), Util: 95.16 RUN 2: runtime=1ms, deadline=period=100ms, RT capacity = 95% TID[616]: RECLAIM=1, (r=1ms, d=100ms, p=100ms), Util: 95.27 TID[616]: RECLAIM=1, (r=1ms, d=100ms, p=100ms), Util: 95.21 RUN 3: 2 tasks Task 1: runtime=1ms, deadline=period=10ms Task 2: runtime=1ms, deadline=period=100ms TID[620]: RECLAIM=1, (r=1ms, d=10ms, p=10ms), Util: 86.64 TID[621]: RECLAIM=1, (r=1ms, d=100ms, p=100ms), Util: 8.66 TID[620]: RECLAIM=1, (r=1ms, d=10ms, p=10ms), Util: 86.45 TID[621]: RECLAIM=1, (r=1ms, d=100ms, p=100ms), Util: 8.73 Running tasks on all cpus allowing for migration also showed that the utilization is reclaimed to the maximum. Running 10 tasks on 3 cpus SCHED_FLAG_RECLAIM - top shows: %Cpu0 : 94.6 us, 0.0 sy, 0.0 ni, 5.4 id, 0.0 wa %Cpu1 : 95.2 us, 0.0 sy, 0.0 ni, 4.8 id, 0.0 wa %Cpu2 : 95.8 us, 0.0 sy, 0.0 ni, 4.2 id, 0.0 wa [1]: Abeni, Luca & Lipari, Giuseppe & Parri, Andrea & Sun, Youcheng. (2015). Parallel and sequential reclaiming in multicore real-time global scheduling. Signed-off-by: Vineeth Pillai (Google) Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Daniel Bristot de Oliveira Acked-by: Juri Lelli Link: https://lore.kernel.org/r/20230530135526.2385378-1-vineeth@bitbyteword.org --- kernel/sched/deadline.c | 50 +++++++++++++++++++++++-------------------------- kernel/sched/sched.h | 6 ++++++ 2 files changed, 29 insertions(+), 27 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c index f827067ad03b..e41a36bd66a6 100644 --- a/kernel/sched/deadline.c +++ b/kernel/sched/deadline.c @@ -1253,43 +1253,39 @@ int dl_runtime_exceeded(struct sched_dl_entity *dl_se) } /* - * This function implements the GRUB accounting rule: - * according to the GRUB reclaiming algorithm, the runtime is - * not decreased as "dq = -dt", but as - * "dq = -max{u / Umax, (1 - Uinact - Uextra)} dt", + * This function implements the GRUB accounting rule. According to the + * GRUB reclaiming algorithm, the runtime is not decreased as "dq = -dt", + * but as "dq = -(max{u, (Umax - Uinact - Uextra)} / Umax) dt", * where u is the utilization of the task, Umax is the maximum reclaimable * utilization, Uinact is the (per-runqueue) inactive utilization, computed * as the difference between the "total runqueue utilization" and the - * runqueue active utilization, and Uextra is the (per runqueue) extra + * "runqueue active utilization", and Uextra is the (per runqueue) extra * reclaimable utilization. - * Since rq->dl.running_bw and rq->dl.this_bw contain utilizations - * multiplied by 2^BW_SHIFT, the result has to be shifted right by - * BW_SHIFT. - * Since rq->dl.bw_ratio contains 1 / Umax multiplied by 2^RATIO_SHIFT, - * dl_bw is multiped by rq->dl.bw_ratio and shifted right by RATIO_SHIFT. - * Since delta is a 64 bit variable, to have an overflow its value - * should be larger than 2^(64 - 20 - 8), which is more than 64 seconds. - * So, overflow is not an issue here. + * Since rq->dl.running_bw and rq->dl.this_bw contain utilizations multiplied + * by 2^BW_SHIFT, the result has to be shifted right by BW_SHIFT. + * Since rq->dl.bw_ratio contains 1 / Umax multiplied by 2^RATIO_SHIFT, dl_bw + * is multiped by rq->dl.bw_ratio and shifted right by RATIO_SHIFT. + * Since delta is a 64 bit variable, to have an overflow its value should be + * larger than 2^(64 - 20 - 8), which is more than 64 seconds. So, overflow is + * not an issue here. */ static u64 grub_reclaim(u64 delta, struct rq *rq, struct sched_dl_entity *dl_se) { - u64 u_inact = rq->dl.this_bw - rq->dl.running_bw; /* Utot - Uact */ u64 u_act; - u64 u_act_min = (dl_se->dl_bw * rq->dl.bw_ratio) >> RATIO_SHIFT; + u64 u_inact = rq->dl.this_bw - rq->dl.running_bw; /* Utot - Uact */ /* - * Instead of computing max{u * bw_ratio, (1 - u_inact - u_extra)}, - * we compare u_inact + rq->dl.extra_bw with - * 1 - (u * rq->dl.bw_ratio >> RATIO_SHIFT), because - * u_inact + rq->dl.extra_bw can be larger than - * 1 * (so, 1 - u_inact - rq->dl.extra_bw would be negative - * leading to wrong results) + * Instead of computing max{u, (u_max - u_inact - u_extra)}, we + * compare u_inact + u_extra with u_max - u, because u_inact + u_extra + * can be larger than u_max. So, u_max - u_inact - u_extra would be + * negative leading to wrong results. */ - if (u_inact + rq->dl.extra_bw > BW_UNIT - u_act_min) - u_act = u_act_min; + if (u_inact + rq->dl.extra_bw > rq->dl.max_bw - dl_se->dl_bw) + u_act = dl_se->dl_bw; else - u_act = BW_UNIT - u_inact - rq->dl.extra_bw; + u_act = rq->dl.max_bw - u_inact - rq->dl.extra_bw; + u_act = (u_act * rq->dl.bw_ratio) >> RATIO_SHIFT; return (delta * u_act) >> BW_SHIFT; } @@ -2788,12 +2784,12 @@ static void init_dl_rq_bw_ratio(struct dl_rq *dl_rq) { if (global_rt_runtime() == RUNTIME_INF) { dl_rq->bw_ratio = 1 << RATIO_SHIFT; - dl_rq->extra_bw = 1 << BW_SHIFT; + dl_rq->max_bw = dl_rq->extra_bw = 1 << BW_SHIFT; } else { dl_rq->bw_ratio = to_ratio(global_rt_runtime(), global_rt_period()) >> (BW_SHIFT - RATIO_SHIFT); - dl_rq->extra_bw = to_ratio(global_rt_period(), - global_rt_runtime()); + dl_rq->max_bw = dl_rq->extra_bw = + to_ratio(global_rt_period(), global_rt_runtime()); } } diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 556496c77dc2..36e23e49817d 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -747,6 +747,12 @@ struct dl_rq { u64 this_bw; u64 extra_bw; + /* + * Maximum available bandwidth for reclaiming by SCHED_FLAG_RECLAIM + * tasks of this rq. Used in calculation of reclaimable bandwidth(GRUB). + */ + u64 max_bw; + /* * Inverse of the fraction of CPU utilization that can be reclaimed * by the GRUB algorithm. -- cgit v1.2.3 From cab3ecaed5cdcc9c36a96874b4c45056a46ece45 Mon Sep 17 00:00:00 2001 From: Hao Jia Date: Tue, 13 Jun 2023 16:20:09 +0800 Subject: sched/core: Fixed missing rq clock update before calling set_rq_offline() When using a cpufreq governor that uses cpufreq_add_update_util_hook(), it is possible to trigger a missing update_rq_clock() warning for the CPU hotplug path: rq_attach_root() set_rq_offline() rq_offline_rt() __disable_runtime() sched_rt_rq_enqueue() enqueue_top_rt_rq() cpufreq_update_util() data->func(data, rq_clock(rq), flags) Move update_rq_clock() from sched_cpu_deactivate() (one of it's callers) into set_rq_offline() such that it covers all set_rq_offline() usage. Additionally change rq_attach_root() to use rq_lock_irqsave() so that it will properly manage the runqueue clock flags. Suggested-by: Ben Segall Signed-off-by: Hao Jia Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Vincent Guittot Link: https://lkml.kernel.org/r/20230613082012.49615-2-jiahao.os@bytedance.com --- kernel/sched/core.c | 2 +- kernel/sched/topology.c | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index ac38225e6d09..442efe59d188 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -9585,6 +9585,7 @@ void set_rq_offline(struct rq *rq) if (rq->online) { const struct sched_class *class; + update_rq_clock(rq); for_each_class(class) { if (class->rq_offline) class->rq_offline(rq); @@ -9726,7 +9727,6 @@ int sched_cpu_deactivate(unsigned int cpu) rq_lock_irqsave(rq, &rf); if (rq->rd) { - update_rq_clock(rq); BUG_ON(!cpumask_test_cpu(cpu, rq->rd->span)); set_rq_offline(rq); } diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c index cb92dc5f5646..d3a3b2646ec4 100644 --- a/kernel/sched/topology.c +++ b/kernel/sched/topology.c @@ -487,9 +487,9 @@ static void free_rootdomain(struct rcu_head *rcu) void rq_attach_root(struct rq *rq, struct root_domain *rd) { struct root_domain *old_rd = NULL; - unsigned long flags; + struct rq_flags rf; - raw_spin_rq_lock_irqsave(rq, flags); + rq_lock_irqsave(rq, &rf); if (rq->rd) { old_rd = rq->rd; @@ -515,7 +515,7 @@ void rq_attach_root(struct rq *rq, struct root_domain *rd) if (cpumask_test_cpu(rq->cpu, cpu_active_mask)) set_rq_online(rq); - raw_spin_rq_unlock_irqrestore(rq, flags); + rq_unlock_irqrestore(rq, &rf); if (old_rd) call_rcu(&old_rd->rcu, free_rootdomain); -- cgit v1.2.3 From 96500560f0c73c71bca1b27536c6254fa0e8ce37 Mon Sep 17 00:00:00 2001 From: Hao Jia Date: Tue, 13 Jun 2023 16:20:10 +0800 Subject: sched/core: Avoid double calling update_rq_clock() in __balance_push_cpu_stop() There is a double update_rq_clock() invocation: __balance_push_cpu_stop() update_rq_clock() __migrate_task() update_rq_clock() Sadly select_fallback_rq() also needs update_rq_clock() for __do_set_cpus_allowed(), it is not possible to remove the update from __balance_push_cpu_stop(). So remove it from __migrate_task() and ensure all callers of this function call update_rq_clock() prior to calling it. Signed-off-by: Hao Jia Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Vincent Guittot Link: https://lkml.kernel.org/r/20230613082012.49615-3-jiahao.os@bytedance.com --- kernel/sched/core.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'kernel/sched') diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 442efe59d188..c7db597e8175 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -2546,7 +2546,6 @@ static struct rq *__migrate_task(struct rq *rq, struct rq_flags *rf, if (!is_cpu_allowed(p, dest_cpu)) return rq; - update_rq_clock(rq); rq = move_queued_task(rq, rf, p, dest_cpu); return rq; @@ -2604,10 +2603,12 @@ static int migration_cpu_stop(void *data) goto out; } - if (task_on_rq_queued(p)) + if (task_on_rq_queued(p)) { + update_rq_clock(rq); rq = __migrate_task(rq, &rf, p, arg->dest_cpu); - else + } else { p->wake_cpu = arg->dest_cpu; + } /* * XXX __migrate_task() can fail, at which point we might end -- cgit v1.2.3 From ebb83d84e49b54369b0db67136a5fe1087124dcc Mon Sep 17 00:00:00 2001 From: Hao Jia Date: Tue, 13 Jun 2023 16:20:11 +0800 Subject: sched/core: Avoid multiple calling update_rq_clock() in __cfsb_csd_unthrottle() After commit 8ad075c2eb1f ("sched: Async unthrottling for cfs bandwidth"), we may update the rq clock multiple times in the loop of __cfsb_csd_unthrottle(). A prior (although less common) instance of this problem exists in unthrottle_offline_cfs_rqs(). Cure both by ensuring update_rq_clock() is called before the loop and setting RQCF_ACT_SKIP during the loop, to supress further updates. The alternative would be pulling update_rq_clock() out of unthrottle_cfs_rq(), but that gives an even bigger mess. Fixes: 8ad075c2eb1f ("sched: Async unthrottling for cfs bandwidth") Reviewed-By: Ben Segall Suggested-by: Vincent Guittot Signed-off-by: Hao Jia Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Vincent Guittot Link: https://lkml.kernel.org/r/20230613082012.49615-4-jiahao.os@bytedance.com --- kernel/sched/fair.c | 18 ++++++++++++++++++ kernel/sched/sched.h | 22 ++++++++++++++++++++++ 2 files changed, 40 insertions(+) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 7666dbc2b788..a80a73909dc2 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5576,6 +5576,14 @@ static void __cfsb_csd_unthrottle(void *arg) rq_lock(rq, &rf); + /* + * Iterating over the list can trigger several call to + * update_rq_clock() in unthrottle_cfs_rq(). + * Do it once and skip the potential next ones. + */ + update_rq_clock(rq); + rq_clock_start_loop_update(rq); + /* * Since we hold rq lock we're safe from concurrent manipulation of * the CSD list. However, this RCU critical section annotates the @@ -5595,6 +5603,7 @@ static void __cfsb_csd_unthrottle(void *arg) rcu_read_unlock(); + rq_clock_stop_loop_update(rq); rq_unlock(rq, &rf); } @@ -6115,6 +6124,13 @@ static void __maybe_unused unthrottle_offline_cfs_rqs(struct rq *rq) lockdep_assert_rq_held(rq); + /* + * The rq clock has already been updated in the + * set_rq_offline(), so we should skip updating + * the rq clock again in unthrottle_cfs_rq(). + */ + rq_clock_start_loop_update(rq); + rcu_read_lock(); list_for_each_entry_rcu(tg, &task_groups, list) { struct cfs_rq *cfs_rq = tg->cfs_rq[cpu_of(rq)]; @@ -6137,6 +6153,8 @@ static void __maybe_unused unthrottle_offline_cfs_rqs(struct rq *rq) unthrottle_cfs_rq(cfs_rq); } rcu_read_unlock(); + + rq_clock_stop_loop_update(rq); } #else /* CONFIG_CFS_BANDWIDTH */ diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h index 36e23e49817d..50d4b61aef3a 100644 --- a/kernel/sched/sched.h +++ b/kernel/sched/sched.h @@ -1546,6 +1546,28 @@ static inline void rq_clock_cancel_skipupdate(struct rq *rq) rq->clock_update_flags &= ~RQCF_REQ_SKIP; } +/* + * During cpu offlining and rq wide unthrottling, we can trigger + * an update_rq_clock() for several cfs and rt runqueues (Typically + * when using list_for_each_entry_*) + * rq_clock_start_loop_update() can be called after updating the clock + * once and before iterating over the list to prevent multiple update. + * After the iterative traversal, we need to call rq_clock_stop_loop_update() + * to clear RQCF_ACT_SKIP of rq->clock_update_flags. + */ +static inline void rq_clock_start_loop_update(struct rq *rq) +{ + lockdep_assert_rq_held(rq); + SCHED_WARN_ON(rq->clock_update_flags & RQCF_ACT_SKIP); + rq->clock_update_flags |= RQCF_ACT_SKIP; +} + +static inline void rq_clock_stop_loop_update(struct rq *rq) +{ + lockdep_assert_rq_held(rq); + rq->clock_update_flags &= ~RQCF_ACT_SKIP; +} + struct rq_flags { unsigned long flags; struct pin_cookie cookie; -- cgit v1.2.3 From ae2ad293d6be143ad223f5f947cca07bcbe42595 Mon Sep 17 00:00:00 2001 From: Miaohe Lin Date: Tue, 20 Jun 2023 16:07:47 +0800 Subject: sched/fair: Use recent_used_cpu to test p->cpus_ptr When checking whether a recently used CPU can be a potential idle candidate, recent_used_cpu should be used to test p->cpus_ptr as p->recent_used_cpu is not equal to recent_used_cpu and candidate decision is made based on recent_used_cpu here. Fixes: 89aafd67f28c ("sched/fair: Use prev instead of new target as recent_used_cpu") Signed-off-by: Miaohe Lin Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Phil Auld Acked-by: Mel Gorman Link: https://lore.kernel.org/r/20230620080747.359122-1-linmiaohe@huawei.com --- kernel/sched/fair.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'kernel/sched') diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index a80a73909dc2..b3e25be58e2b 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -7174,7 +7174,7 @@ static int select_idle_sibling(struct task_struct *p, int prev, int target) recent_used_cpu != target && cpus_share_cache(recent_used_cpu, target) && (available_idle_cpu(recent_used_cpu) || sched_idle_cpu(recent_used_cpu)) && - cpumask_test_cpu(p->recent_used_cpu, p->cpus_ptr) && + cpumask_test_cpu(recent_used_cpu, p->cpus_ptr) && asym_fits_cpu(task_util, util_min, util_max, recent_used_cpu)) { return recent_used_cpu; } -- cgit v1.2.3 From aff037078ecaecf34a7c2afab1341815f90fba5e Mon Sep 17 00:00:00 2001 From: Suren Baghdasaryan Date: Thu, 29 Jun 2023 17:56:12 -0700 Subject: sched/psi: use kernfs polling functions for PSI trigger polling Destroying psi trigger in cgroup_file_release causes UAF issues when a cgroup is removed from under a polling process. This is happening because cgroup removal causes a call to cgroup_file_release while the actual file is still alive. Destroying the trigger at this point would also destroy its waitqueue head and if there is still a polling process on that file accessing the waitqueue, it will step on the freed pointer: do_select vfs_poll do_rmdir cgroup_rmdir kernfs_drain_open_files cgroup_file_release cgroup_pressure_release psi_trigger_destroy wake_up_pollfree(&t->event_wait) // vfs_poll is unblocked synchronize_rcu kfree(t) poll_freewait -> UAF access to the trigger's waitqueue head Patch [1] fixed this issue for epoll() case using wake_up_pollfree(), however the same issue exists for synchronous poll() case. The root cause of this issue is that the lifecycles of the psi trigger's waitqueue and of the file associated with the trigger are different. Fix this by using kernfs_generic_poll function when polling on cgroup-specific psi triggers. It internally uses kernfs_open_node->poll waitqueue head with its lifecycle tied to the file's lifecycle. This also renders the fix in [1] obsolete, so revert it. [1] commit c2dbe32d5db5 ("sched/psi: Fix use-after-free in ep_remove_wait_queue()") Fixes: 0e94682b73bf ("psi: introduce psi monitor") Closes: https://lore.kernel.org/all/20230613062306.101831-1-lujialin4@huawei.com/ Reported-by: Lu Jialin Signed-off-by: Suren Baghdasaryan Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/20230630005612.1014540-1-surenb@google.com --- include/linux/psi.h | 5 +++-- include/linux/psi_types.h | 3 +++ kernel/cgroup/cgroup.c | 2 +- kernel/sched/psi.c | 29 +++++++++++++++++++++-------- 4 files changed, 28 insertions(+), 11 deletions(-) (limited to 'kernel/sched') diff --git a/include/linux/psi.h b/include/linux/psi.h index ab26200c2803..e0745873e3f2 100644 --- a/include/linux/psi.h +++ b/include/linux/psi.h @@ -23,8 +23,9 @@ void psi_memstall_enter(unsigned long *flags); void psi_memstall_leave(unsigned long *flags); int psi_show(struct seq_file *s, struct psi_group *group, enum psi_res res); -struct psi_trigger *psi_trigger_create(struct psi_group *group, - char *buf, enum psi_res res, struct file *file); +struct psi_trigger *psi_trigger_create(struct psi_group *group, char *buf, + enum psi_res res, struct file *file, + struct kernfs_open_file *of); void psi_trigger_destroy(struct psi_trigger *t); __poll_t psi_trigger_poll(void **trigger_ptr, struct file *file, diff --git a/include/linux/psi_types.h b/include/linux/psi_types.h index 040c089581c6..f1fd3a8044e0 100644 --- a/include/linux/psi_types.h +++ b/include/linux/psi_types.h @@ -137,6 +137,9 @@ struct psi_trigger { /* Wait queue for polling */ wait_queue_head_t event_wait; + /* Kernfs file for cgroup triggers */ + struct kernfs_open_file *of; + /* Pending event flag */ int event; diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index bfe3cd8ccf36..f55a40db065f 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -3730,7 +3730,7 @@ static ssize_t pressure_write(struct kernfs_open_file *of, char *buf, } psi = cgroup_psi(cgrp); - new = psi_trigger_create(psi, buf, res, of->file); + new = psi_trigger_create(psi, buf, res, of->file, of); if (IS_ERR(new)) { cgroup_put(cgrp); return PTR_ERR(new); diff --git a/kernel/sched/psi.c b/kernel/sched/psi.c index 81fca77397f6..9bb3f2b3ccfc 100644 --- a/kernel/sched/psi.c +++ b/kernel/sched/psi.c @@ -493,8 +493,12 @@ static u64 update_triggers(struct psi_group *group, u64 now, bool *update_total, continue; /* Generate an event */ - if (cmpxchg(&t->event, 0, 1) == 0) - wake_up_interruptible(&t->event_wait); + if (cmpxchg(&t->event, 0, 1) == 0) { + if (t->of) + kernfs_notify(t->of->kn); + else + wake_up_interruptible(&t->event_wait); + } t->last_event_time = now; /* Reset threshold breach flag once event got generated */ t->pending_event = false; @@ -1271,8 +1275,9 @@ int psi_show(struct seq_file *m, struct psi_group *group, enum psi_res res) return 0; } -struct psi_trigger *psi_trigger_create(struct psi_group *group, - char *buf, enum psi_res res, struct file *file) +struct psi_trigger *psi_trigger_create(struct psi_group *group, char *buf, + enum psi_res res, struct file *file, + struct kernfs_open_file *of) { struct psi_trigger *t; enum psi_states state; @@ -1331,7 +1336,9 @@ struct psi_trigger *psi_trigger_create(struct psi_group *group, t->event = 0; t->last_event_time = 0; - init_waitqueue_head(&t->event_wait); + t->of = of; + if (!of) + init_waitqueue_head(&t->event_wait); t->pending_event = false; t->aggregator = privileged ? PSI_POLL : PSI_AVGS; @@ -1388,7 +1395,10 @@ void psi_trigger_destroy(struct psi_trigger *t) * being accessed later. Can happen if cgroup is deleted from under a * polling process. */ - wake_up_pollfree(&t->event_wait); + if (t->of) + kernfs_notify(t->of->kn); + else + wake_up_interruptible(&t->event_wait); if (t->aggregator == PSI_AVGS) { mutex_lock(&group->avgs_lock); @@ -1465,7 +1475,10 @@ __poll_t psi_trigger_poll(void **trigger_ptr, if (!t) return DEFAULT_POLLMASK | EPOLLERR | EPOLLPRI; - poll_wait(file, &t->event_wait, wait); + if (t->of) + kernfs_generic_poll(t->of, wait); + else + poll_wait(file, &t->event_wait, wait); if (cmpxchg(&t->event, 1, 0) == 1) ret |= EPOLLPRI; @@ -1535,7 +1548,7 @@ static ssize_t psi_write(struct file *file, const char __user *user_buf, return -EBUSY; } - new = psi_trigger_create(&psi_system, buf, res, file); + new = psi_trigger_create(&psi_system, buf, res, file, NULL); if (IS_ERR(new)) { mutex_unlock(&seq->lock); return PTR_ERR(new); -- cgit v1.2.3