From cac5cefbade90ff0bb0b393d301fa3b5234cf056 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Wed, 28 May 2025 10:09:01 +0200 Subject: sched/smp: Make SMP unconditional Simplify the scheduler by making CONFIG_SMP=y primitives and data structures unconditional. Introduce transitory wrappers for functionality not yet converted to SMP. Note that this patch is pretty large, because there's no clear separation between various aspects of the SMP scheduler, it's basically a huge block of #ifdef CONFIG_SMP. A fair amount of it has to be switched on for it to boot and work on UP systems. Signed-off-by: Ingo Molnar Acked-by: Peter Zijlstra Cc: Dietmar Eggemann Cc: Juri Lelli Cc: Linus Torvalds Cc: Mel Gorman Cc: Sebastian Andrzej Siewior Cc: Shrikanth Hegde Cc: Steven Rostedt Cc: Valentin Schneider Cc: Vincent Guittot Link: https://lore.kernel.org/r/20250528080924.2273858-21-mingo@kernel.org --- include/linux/sched/deadline.h | 4 ---- include/linux/sched/idle.h | 4 ---- include/linux/sched/nohz.h | 4 ++-- include/linux/sched/topology.h | 32 -------------------------------- 4 files changed, 2 insertions(+), 42 deletions(-) (limited to 'include/linux/sched') diff --git a/include/linux/sched/deadline.h b/include/linux/sched/deadline.h index f9aabbc9d22e..c40115d4e34d 100644 --- a/include/linux/sched/deadline.h +++ b/include/linux/sched/deadline.h @@ -29,15 +29,11 @@ static inline bool dl_time_before(u64 a, u64 b) return (s64)(a - b) < 0; } -#ifdef CONFIG_SMP - struct root_domain; extern void dl_add_task_root_domain(struct task_struct *p); extern void dl_clear_root_domain(struct root_domain *rd); extern void dl_clear_root_domain_cpu(int cpu); -#endif /* CONFIG_SMP */ - extern u64 dl_cookie; extern bool dl_bw_visited(int cpu, u64 cookie); diff --git a/include/linux/sched/idle.h b/include/linux/sched/idle.h index 439f6029d3b9..8465ff1f20d1 100644 --- a/include/linux/sched/idle.h +++ b/include/linux/sched/idle.h @@ -11,11 +11,7 @@ enum cpu_idle_type { CPU_MAX_IDLE_TYPES }; -#ifdef CONFIG_SMP extern void wake_up_if_idle(int cpu); -#else -static inline void wake_up_if_idle(int cpu) { } -#endif /* * Idle thread specific functions to determine the need_resched diff --git a/include/linux/sched/nohz.h b/include/linux/sched/nohz.h index 6d67e9a5af6b..0db7f67935fe 100644 --- a/include/linux/sched/nohz.h +++ b/include/linux/sched/nohz.h @@ -6,7 +6,7 @@ * This is the interface between the scheduler and nohz/dynticks: */ -#if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ_COMMON) +#ifdef CONFIG_NO_HZ_COMMON extern void nohz_balance_enter_idle(int cpu); extern int get_nohz_timer_target(void); #else @@ -23,7 +23,7 @@ static inline void calc_load_nohz_remote(struct rq *rq) { } static inline void calc_load_nohz_stop(void) { } #endif /* CONFIG_NO_HZ_COMMON */ -#if defined(CONFIG_NO_HZ_COMMON) && defined(CONFIG_SMP) +#ifdef CONFIG_NO_HZ_COMMON extern void wake_up_nohz_cpu(int cpu); #else static inline void wake_up_nohz_cpu(int cpu) { } diff --git a/include/linux/sched/topology.h b/include/linux/sched/topology.h index 198bb5cc1774..e54e7fa76ba6 100644 --- a/include/linux/sched/topology.h +++ b/include/linux/sched/topology.h @@ -9,7 +9,6 @@ /* * sched-domains (multiprocessor balancing) declarations: */ -#ifdef CONFIG_SMP /* Generate SD flag indexes */ #define SD_FLAG(name, mflags) __##name, @@ -200,37 +199,6 @@ extern void sched_update_asym_prefer_cpu(int cpu, int old_prio, int new_prio); # define SD_INIT_NAME(type) .name = #type -#else /* CONFIG_SMP */ - -struct sched_domain_attr; - -static inline void -partition_sched_domains(int ndoms_new, cpumask_var_t doms_new[], - struct sched_domain_attr *dattr_new) -{ -} - -static inline bool cpus_equal_capacity(int this_cpu, int that_cpu) -{ - return true; -} - -static inline bool cpus_share_cache(int this_cpu, int that_cpu) -{ - return true; -} - -static inline bool cpus_share_resources(int this_cpu, int that_cpu) -{ - return true; -} - -static inline void sched_update_asym_prefer_cpu(int cpu, int old_prio, int new_prio) -{ -} - -#endif /* !CONFIG_SMP */ - #if defined(CONFIG_ENERGY_MODEL) && defined(CONFIG_CPU_FREQ_GOV_SCHEDUTIL) extern void rebuild_sched_domains_energy(void); #else -- cgit v1.2.3 From 1f25730e5a780b33f78e3ea23e64d3f75e0b2042 Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Wed, 28 May 2025 10:09:07 +0200 Subject: sched/smp: Use the SMP version of sched_exec() Simplify the scheduler making CONFIG_SMP=y sched_exec() code unconditional. Signed-off-by: Ingo Molnar Acked-by: Peter Zijlstra Cc: Dietmar Eggemann Cc: Juri Lelli Cc: Linus Torvalds Cc: Mel Gorman Cc: Sebastian Andrzej Siewior Cc: Shrikanth Hegde Cc: Steven Rostedt Cc: Valentin Schneider Cc: Vincent Guittot Link: https://lore.kernel.org/r/20250528080924.2273858-27-mingo@kernel.org --- include/linux/sched/task.h | 4 ---- 1 file changed, 4 deletions(-) (limited to 'include/linux/sched') diff --git a/include/linux/sched/task.h b/include/linux/sched/task.h index ca1db4b92c32..c517dbc242f7 100644 --- a/include/linux/sched/task.h +++ b/include/linux/sched/task.h @@ -109,11 +109,7 @@ int kernel_wait(pid_t pid, int *stat); extern void free_task(struct task_struct *tsk); /* sched_exec is called by processes performing an exec */ -#ifdef CONFIG_SMP extern void sched_exec(void); -#else -#define sched_exec() {} -#endif static inline struct task_struct *get_task_struct(struct task_struct *t) { -- cgit v1.2.3 From e075f4360931263f5ec006ea5dadc065e5e98eb8 Mon Sep 17 00:00:00 2001 From: Li Chen Date: Thu, 10 Jul 2025 18:57:07 +0800 Subject: smpboot: introduce SDTL_INIT() helper to tidy sched topology setup Define a small SDTL_INIT(maskfn, flagsfn, name) macro and use it to build the sched_domain_topology_level array. Purely a cleanup; behaviour is unchanged. Suggested-by: Thomas Gleixner Signed-off-by: Li Chen Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: K Prateek Nayak Tested-by: K Prateek Nayak Link: https://lore.kernel.org/r/20250710105715.66594-2-me@linux.beauty --- include/linux/sched/topology.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux/sched') diff --git a/include/linux/sched/topology.h b/include/linux/sched/topology.h index e54e7fa76ba6..0d5daaa277b7 100644 --- a/include/linux/sched/topology.h +++ b/include/linux/sched/topology.h @@ -196,8 +196,8 @@ struct sched_domain_topology_level { extern void __init set_sched_topology(struct sched_domain_topology_level *tl); extern void sched_update_asym_prefer_cpu(int cpu, int old_prio, int new_prio); - -# define SD_INIT_NAME(type) .name = #type +#define SDTL_INIT(maskfn, flagsfn, dname) ((struct sched_domain_topology_level) \ + { .mask = maskfn, .sd_flags = flagsfn, .name = #dname }) #if defined(CONFIG_ENERGY_MODEL) && defined(CONFIG_CPU_FREQ_GOV_SCHEDUTIL) extern void rebuild_sched_domains_energy(void); -- cgit v1.2.3 From 1eec89a671413ce38df9fe9e70f5130a9eb79a59 Mon Sep 17 00:00:00 2001 From: K Prateek Nayak Date: Fri, 11 Jul 2025 11:20:30 +0530 Subject: sched/topology: Remove sched_domain_topology_level::flags Support for overlapping domains added in commit e3589f6c81e4 ("sched: Allow for overlapping sched_domain spans") also allowed forcefully setting SD_OVERLAP for !NUMA domains via FORCE_SD_OVERLAP sched_feat(). Since NUMA domains had to be presumed overlapping to ensure correct behavior, "sched_domain_topology_level::flags" was introduced. NUMA domains added the SDTL_OVERLAP flag would ensure SD_OVERLAP was always added during build_sched_domains() for these domains, even when FORCE_SD_OVERLAP was off. Condition for adding the SD_OVERLAP flag at the aforementioned commit was as follows: if (tl->flags & SDTL_OVERLAP || sched_feat(FORCE_SD_OVERLAP)) sd->flags |= SD_OVERLAP; The FORCE_SD_OVERLAP debug feature was removed in commit af85596c74de ("sched/topology: Remove FORCE_SD_OVERLAP") which left the NUMA domains as the exclusive users of SDTL_OVERLAP, SD_OVERLAP, and SD_NUMA flags. Get rid of SDTL_OVERLAP and SD_OVERLAP as they have become redundant and instead rely on SD_NUMA to detect the only overlapping domain currently supported. Since SDTL_OVERLAP was the only user of "tl->flags", get rid of "sched_domain_topology_level::flags" too. Signed-off-by: K Prateek Nayak Signed-off-by: Peter Zijlstra (Intel) Link: https://lkml.kernel.org/r/ba4dbdf8-bc37-493d-b2e0-2efb00ea3e19@amd.com --- include/linux/sched/sd_flags.h | 8 -------- include/linux/sched/topology.h | 3 --- 2 files changed, 11 deletions(-) (limited to 'include/linux/sched') diff --git a/include/linux/sched/sd_flags.h b/include/linux/sched/sd_flags.h index b04a5d04dee9..42839cfa2778 100644 --- a/include/linux/sched/sd_flags.h +++ b/include/linux/sched/sd_flags.h @@ -153,14 +153,6 @@ SD_FLAG(SD_ASYM_PACKING, SDF_NEEDS_GROUPS) */ SD_FLAG(SD_PREFER_SIBLING, SDF_NEEDS_GROUPS) -/* - * sched_groups of this level overlap - * - * SHARED_PARENT: Set for all NUMA levels above NODE. - * NEEDS_GROUPS: Overlaps can only exist with more than one group. - */ -SD_FLAG(SD_OVERLAP, SDF_SHARED_PARENT | SDF_NEEDS_GROUPS) - /* * Cross-node balancing * diff --git a/include/linux/sched/topology.h b/include/linux/sched/topology.h index 0d5daaa277b7..5263746b63e8 100644 --- a/include/linux/sched/topology.h +++ b/include/linux/sched/topology.h @@ -175,8 +175,6 @@ bool cpus_share_resources(int this_cpu, int that_cpu); typedef const struct cpumask *(*sched_domain_mask_f)(int cpu); typedef int (*sched_domain_flags_f)(void); -#define SDTL_OVERLAP 0x01 - struct sd_data { struct sched_domain *__percpu *sd; struct sched_domain_shared *__percpu *sds; @@ -187,7 +185,6 @@ struct sd_data { struct sched_domain_topology_level { sched_domain_mask_f mask; sched_domain_flags_f sd_flags; - int flags; int numa_level; struct sd_data data; char *name; -- cgit v1.2.3 From 8671bad873ebeb082afcf7b4501395c374da6023 Mon Sep 17 00:00:00 2001 From: "Luis Claudio R. Goncalves" Date: Mon, 7 Jul 2025 11:03:59 -0300 Subject: sched: Do not call __put_task_struct() on rt if pi_blocked_on is set With PREEMPT_RT enabled, some of the calls to put_task_struct() coming from rt_mutex_adjust_prio_chain() could happen in preemptible context and with a mutex enqueued. That could lead to this sequence: rt_mutex_adjust_prio_chain() put_task_struct() __put_task_struct() sched_ext_free() spin_lock_irqsave() rtlock_lock() ---> TRIGGERS lockdep_assert(!current->pi_blocked_on); This is not a SCHED_EXT bug. The first cleanup function called by __put_task_struct() is sched_ext_free() and it happens to take a (RT) spin_lock, which in the scenario described above, would trigger the lockdep assertion of "!current->pi_blocked_on". Crystal Wood was able to identify the problem as __put_task_struct() being called during rt_mutex_adjust_prio_chain(), in the context of a process with a mutex enqueued. Instead of adding more complex conditions to decide when to directly call __put_task_struct() and when to defer the call, unconditionally resort to the deferred call on PREEMPT_RT to simplify the code. Fixes: 893cdaaa3977 ("sched: avoid false lockdep splat in put_task_struct()") Suggested-by: Crystal Wood Signed-off-by: Luis Claudio R. Goncalves Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Wander Lairson Costa Reviewed-by: Valentin Schneider Reviewed-by: Sebastian Andrzej Siewior Link: https://lore.kernel.org/r/aGvTz5VaPFyj0pBV@uudg.org --- include/linux/sched/task.h | 27 ++++++++++----------------- 1 file changed, 10 insertions(+), 17 deletions(-) (limited to 'include/linux/sched') diff --git a/include/linux/sched/task.h b/include/linux/sched/task.h index c517dbc242f7..ea41795a352b 100644 --- a/include/linux/sched/task.h +++ b/include/linux/sched/task.h @@ -131,24 +131,17 @@ static inline void put_task_struct(struct task_struct *t) return; /* - * In !RT, it is always safe to call __put_task_struct(). - * Under RT, we can only call it in preemptible context. - */ - if (!IS_ENABLED(CONFIG_PREEMPT_RT) || preemptible()) { - static DEFINE_WAIT_OVERRIDE_MAP(put_task_map, LD_WAIT_SLEEP); - - lock_map_acquire_try(&put_task_map); - __put_task_struct(t); - lock_map_release(&put_task_map); - return; - } - - /* - * under PREEMPT_RT, we can't call put_task_struct + * Under PREEMPT_RT, we can't call __put_task_struct * in atomic context because it will indirectly - * acquire sleeping locks. + * acquire sleeping locks. The same is true if the + * current process has a mutex enqueued (blocked on + * a PI chain). + * + * In !RT, it is always safe to call __put_task_struct(). + * Though, in order to simplify the code, resort to the + * deferred call too. * - * call_rcu() will schedule delayed_put_task_struct_rcu() + * call_rcu() will schedule __put_task_struct_rcu_cb() * to be called in process context. * * __put_task_struct() is called when @@ -161,7 +154,7 @@ static inline void put_task_struct(struct task_struct *t) * * delayed_free_task() also uses ->rcu, but it is only called * when it fails to fork a process. Therefore, there is no - * way it can conflict with put_task_struct(). + * way it can conflict with __put_task_struct(). */ call_rcu(&t->rcu, __put_task_struct_rcu_cb); } -- cgit v1.2.3