From a8b62fd0850503cf1e557d7e5a98d3f1f5c25eef Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 21 Sep 2020 12:58:17 +0200 Subject: stop_machine: Add function and caller debug info Crashes in stop-machine are hard to connect to the calling code, add a little something to help with that. Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Valentin Schneider Reviewed-by: Daniel Bristot de Oliveira Link: https://lkml.kernel.org/r/20201023102346.116513635@infradead.org --- include/linux/stop_machine.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'include') diff --git a/include/linux/stop_machine.h b/include/linux/stop_machine.h index 76d8b09384a7..30577c3aecf8 100644 --- a/include/linux/stop_machine.h +++ b/include/linux/stop_machine.h @@ -24,6 +24,7 @@ typedef int (*cpu_stop_fn_t)(void *arg); struct cpu_stop_work { struct list_head list; /* cpu_stopper->works */ cpu_stop_fn_t fn; + unsigned long caller; void *arg; struct cpu_stop_done *done; }; @@ -36,6 +37,8 @@ void stop_machine_park(int cpu); void stop_machine_unpark(int cpu); void stop_machine_yield(const struct cpumask *cpumask); +extern void print_stop_info(const char *log_lvl, struct task_struct *task); + #else /* CONFIG_SMP */ #include @@ -80,6 +83,8 @@ static inline bool stop_one_cpu_nowait(unsigned int cpu, return false; } +static inline void print_stop_info(const char *log_lvl, struct task_struct *task) { } + #endif /* CONFIG_SMP */ /* -- cgit From 1cf12e08bc4d50a76b80c42a3109c53d8794a0c9 Mon Sep 17 00:00:00 2001 From: Thomas Gleixner Date: Wed, 16 Sep 2020 09:27:18 +0200 Subject: sched/hotplug: Consolidate task migration on CPU unplug With the new mechanism which kicks tasks off the outgoing CPU at the end of schedule() the situation on an outgoing CPU right before the stopper thread brings it down completely is: - All user tasks and all unbound kernel threads have either been migrated away or are not running and the next wakeup will move them to a online CPU. - All per CPU kernel threads, except cpu hotplug thread and the stopper thread have either been unbound or parked by the responsible CPU hotplug callback. That means that at the last step before the stopper thread is invoked the cpu hotplug thread is the last legitimate running task on the outgoing CPU. Add a final wait step right before the stopper thread is kicked which ensures that any still running tasks on the way to park or on the way to kick themself of the CPU are either sleeping or gone. This allows to remove the migrate_tasks() crutch in sched_cpu_dying(). If sched_cpu_dying() detects that there is still another running task aside of the stopper thread then it will explode with the appropriate fireworks. Signed-off-by: Thomas Gleixner Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Valentin Schneider Reviewed-by: Daniel Bristot de Oliveira Link: https://lkml.kernel.org/r/20201023102346.547163969@infradead.org --- include/linux/cpuhotplug.h | 1 + include/linux/sched/hotplug.h | 2 ++ 2 files changed, 3 insertions(+) (limited to 'include') diff --git a/include/linux/cpuhotplug.h b/include/linux/cpuhotplug.h index bc56287a1ed1..0042ef362511 100644 --- a/include/linux/cpuhotplug.h +++ b/include/linux/cpuhotplug.h @@ -152,6 +152,7 @@ enum cpuhp_state { CPUHP_AP_ONLINE, CPUHP_TEARDOWN_CPU, CPUHP_AP_ONLINE_IDLE, + CPUHP_AP_SCHED_WAIT_EMPTY, CPUHP_AP_SMPBOOT_THREADS, CPUHP_AP_X86_VDSO_VMA_ONLINE, CPUHP_AP_IRQ_AFFINITY_ONLINE, diff --git a/include/linux/sched/hotplug.h b/include/linux/sched/hotplug.h index 9a62ffdd296f..412cdaba33eb 100644 --- a/include/linux/sched/hotplug.h +++ b/include/linux/sched/hotplug.h @@ -11,8 +11,10 @@ extern int sched_cpu_activate(unsigned int cpu); extern int sched_cpu_deactivate(unsigned int cpu); #ifdef CONFIG_HOTPLUG_CPU +extern int sched_cpu_wait_empty(unsigned int cpu); extern int sched_cpu_dying(unsigned int cpu); #else +# define sched_cpu_wait_empty NULL # define sched_cpu_dying NULL #endif -- cgit From af449901b84c98cbd84a0113223ba3bcfcb12a26 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 17 Sep 2020 10:38:30 +0200 Subject: sched: Add migrate_disable() Add the base migrate_disable() support (under protest). While migrate_disable() is (currently) required for PREEMPT_RT, it is also one of the biggest flaws in the system. Notably this is just the base implementation, it is broken vs sched_setaffinity() and hotplug, both solved in additional patches for ease of review. Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Valentin Schneider Reviewed-by: Daniel Bristot de Oliveira Link: https://lkml.kernel.org/r/20201023102346.818170844@infradead.org --- include/linux/preempt.h | 65 +++++++++++++++++++++++++++++++++++++++++++++++++ include/linux/sched.h | 3 +++ 2 files changed, 68 insertions(+) (limited to 'include') diff --git a/include/linux/preempt.h b/include/linux/preempt.h index 7d9c1c0e149c..97ba7c920653 100644 --- a/include/linux/preempt.h +++ b/include/linux/preempt.h @@ -322,6 +322,69 @@ static inline void preempt_notifier_init(struct preempt_notifier *notifier, #endif +#if defined(CONFIG_SMP) && defined(CONFIG_PREEMPT_RT) + +/* + * Migrate-Disable and why it is (strongly) undesired. + * + * The premise of the Real-Time schedulers we have on Linux + * (SCHED_FIFO/SCHED_DEADLINE) is that M CPUs can/will run M tasks + * concurrently, provided there are sufficient runnable tasks, also known as + * work-conserving. For instance SCHED_DEADLINE tries to schedule the M + * earliest deadline threads, and SCHED_FIFO the M highest priority threads. + * + * The correctness of various scheduling models depends on this, but is it + * broken by migrate_disable() that doesn't imply preempt_disable(). Where + * preempt_disable() implies an immediate priority ceiling, preemptible + * migrate_disable() allows nesting. + * + * The worst case is that all tasks preempt one another in a migrate_disable() + * region and stack on a single CPU. This then reduces the available bandwidth + * to a single CPU. And since Real-Time schedulability theory considers the + * Worst-Case only, all Real-Time analysis shall revert to single-CPU + * (instantly solving the SMP analysis problem). + * + * + * The reason we have it anyway. + * + * PREEMPT_RT breaks a number of assumptions traditionally held. By forcing a + * number of primitives into becoming preemptible, they would also allow + * migration. This turns out to break a bunch of per-cpu usage. To this end, + * all these primitives employ migirate_disable() to restore this implicit + * assumption. + * + * This is a 'temporary' work-around at best. The correct solution is getting + * rid of the above assumptions and reworking the code to employ explicit + * per-cpu locking or short preempt-disable regions. + * + * The end goal must be to get rid of migrate_disable(), alternatively we need + * a schedulability theory that does not depend on abritrary migration. + * + * + * Notes on the implementation. + * + * The implementation is particularly tricky since existing code patterns + * dictate neither migrate_disable() nor migrate_enable() is allowed to block. + * This means that it cannot use cpus_read_lock() to serialize against hotplug, + * nor can it easily migrate itself into a pending affinity mask change on + * migrate_enable(). + * + * + * Note: even non-work-conserving schedulers like semi-partitioned depends on + * migration, so migrate_disable() is not only a problem for + * work-conserving schedulers. + * + */ +extern void migrate_disable(void); +extern void migrate_enable(void); + +#elif defined(CONFIG_PREEMPT_RT) + +static inline void migrate_disable(void) { } +static inline void migrate_enable(void) { } + +#else /* !CONFIG_PREEMPT_RT */ + /** * migrate_disable - Prevent migration of the current task * @@ -352,4 +415,6 @@ static __always_inline void migrate_enable(void) preempt_enable(); } +#endif /* CONFIG_SMP && CONFIG_PREEMPT_RT */ + #endif /* __LINUX_PREEMPT_H */ diff --git a/include/linux/sched.h b/include/linux/sched.h index 063cd120b459..0732356c0eca 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -714,6 +714,9 @@ struct task_struct { int nr_cpus_allowed; const cpumask_t *cpus_ptr; cpumask_t cpus_mask; +#if defined(CONFIG_SMP) && defined(CONFIG_PREEMPT_RT) + int migration_disabled; +#endif #ifdef CONFIG_PREEMPT_RCU int rcu_read_lock_nesting; -- cgit From 6d337eab041d56bb8f0e7794f39906c21054c512 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Fri, 18 Sep 2020 17:24:31 +0200 Subject: sched: Fix migrate_disable() vs set_cpus_allowed_ptr() Concurrent migrate_disable() and set_cpus_allowed_ptr() has interesting features. We rely on set_cpus_allowed_ptr() to not return until the task runs inside the provided mask. This expectation is exported to userspace. This means that any set_cpus_allowed_ptr() caller must wait until migrate_enable() allows migrations. At the same time, we don't want migrate_enable() to schedule, due to patterns like: preempt_disable(); migrate_disable(); ... migrate_enable(); preempt_enable(); And: raw_spin_lock(&B); spin_unlock(&A); this means that when migrate_enable() must restore the affinity mask, it cannot wait for completion thereof. Luck will have it that that is exactly the case where there is a pending set_cpus_allowed_ptr(), so let that provide storage for the async stop machine. Much thanks to Valentin who used TLA+ most effective and found lots of 'interesting' cases. Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Valentin Schneider Reviewed-by: Daniel Bristot de Oliveira Link: https://lkml.kernel.org/r/20201023102346.921768277@infradead.org --- include/linux/sched.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include') diff --git a/include/linux/sched.h b/include/linux/sched.h index 0732356c0eca..90a0c92741d7 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -714,6 +714,7 @@ struct task_struct { int nr_cpus_allowed; const cpumask_t *cpus_ptr; cpumask_t cpus_mask; + void *migration_pending; #if defined(CONFIG_SMP) && defined(CONFIG_PREEMPT_RT) int migration_disabled; #endif -- cgit From 14e292f8d45380c519a83d9b0f37089a17eedcdf Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Thu, 1 Oct 2020 15:54:14 +0200 Subject: sched,rt: Use cpumask_any*_distribute() Replace a bunch of cpumask_any*() instances with cpumask_any*_distribute(), by injecting this little bit of random in cpu selection, we reduce the chance two competing balance operations working off the same lowest_mask pick the same CPU. Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Valentin Schneider Reviewed-by: Daniel Bristot de Oliveira Link: https://lkml.kernel.org/r/20201023102347.190759694@infradead.org --- include/linux/cpumask.h | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'include') diff --git a/include/linux/cpumask.h b/include/linux/cpumask.h index f0d895d6ac39..383684e30f12 100644 --- a/include/linux/cpumask.h +++ b/include/linux/cpumask.h @@ -199,6 +199,11 @@ static inline int cpumask_any_and_distribute(const struct cpumask *src1p, return cpumask_next_and(-1, src1p, src2p); } +static inline int cpumask_any_distribute(const struct cpumask *srcp) +{ + return cpumask_first(srcp); +} + #define for_each_cpu(cpu, mask) \ for ((cpu) = 0; (cpu) < 1; (cpu)++, (void)mask) #define for_each_cpu_not(cpu, mask) \ @@ -252,6 +257,7 @@ int cpumask_any_but(const struct cpumask *mask, unsigned int cpu); unsigned int cpumask_local_spread(unsigned int i, int node); int cpumask_any_and_distribute(const struct cpumask *src1p, const struct cpumask *src2p); +int cpumask_any_distribute(const struct cpumask *srcp); /** * for_each_cpu - iterate over every cpu in a mask -- cgit From a7c81556ec4d341dfdbf2cc478ead89d73e474a7 Mon Sep 17 00:00:00 2001 From: Peter Zijlstra Date: Mon, 28 Sep 2020 17:06:07 +0200 Subject: sched: Fix migrate_disable() vs rt/dl balancing In order to minimize the interference of migrate_disable() on lower priority tasks, which can be deprived of runtime due to being stuck below a higher priority task. Teach the RT/DL balancers to push away these higher priority tasks when a lower priority task gets selected to run on a freshly demoted CPU (pull). This adds migration interference to the higher priority task, but restores bandwidth to system that would otherwise be irrevocably lost. Without this it would be possible to have all tasks on the system stuck on a single CPU, each task preempted in a migrate_disable() section with a single high priority task running. This way we can still approximate running the M highest priority tasks on the system. Migrating the top task away is (ofcourse) still subject to migrate_disable() too, which means the lower task is subject to an interference equivalent to the worst case migrate_disable() section. Signed-off-by: Peter Zijlstra (Intel) Reviewed-by: Daniel Bristot de Oliveira Link: https://lkml.kernel.org/r/20201023102347.499155098@infradead.org --- include/linux/preempt.h | 40 ++++++++++++++++++++++------------------ include/linux/sched.h | 3 ++- 2 files changed, 24 insertions(+), 19 deletions(-) (limited to 'include') diff --git a/include/linux/preempt.h b/include/linux/preempt.h index 97ba7c920653..8b43922e65df 100644 --- a/include/linux/preempt.h +++ b/include/linux/preempt.h @@ -325,24 +325,28 @@ static inline void preempt_notifier_init(struct preempt_notifier *notifier, #if defined(CONFIG_SMP) && defined(CONFIG_PREEMPT_RT) /* - * Migrate-Disable and why it is (strongly) undesired. - * - * The premise of the Real-Time schedulers we have on Linux - * (SCHED_FIFO/SCHED_DEADLINE) is that M CPUs can/will run M tasks - * concurrently, provided there are sufficient runnable tasks, also known as - * work-conserving. For instance SCHED_DEADLINE tries to schedule the M - * earliest deadline threads, and SCHED_FIFO the M highest priority threads. - * - * The correctness of various scheduling models depends on this, but is it - * broken by migrate_disable() that doesn't imply preempt_disable(). Where - * preempt_disable() implies an immediate priority ceiling, preemptible - * migrate_disable() allows nesting. - * - * The worst case is that all tasks preempt one another in a migrate_disable() - * region and stack on a single CPU. This then reduces the available bandwidth - * to a single CPU. And since Real-Time schedulability theory considers the - * Worst-Case only, all Real-Time analysis shall revert to single-CPU - * (instantly solving the SMP analysis problem). + * Migrate-Disable and why it is undesired. + * + * When a preempted task becomes elegible to run under the ideal model (IOW it + * becomes one of the M highest priority tasks), it might still have to wait + * for the preemptee's migrate_disable() section to complete. Thereby suffering + * a reduction in bandwidth in the exact duration of the migrate_disable() + * section. + * + * Per this argument, the change from preempt_disable() to migrate_disable() + * gets us: + * + * - a higher priority tasks gains reduced wake-up latency; with preempt_disable() + * it would have had to wait for the lower priority task. + * + * - a lower priority tasks; which under preempt_disable() could've instantly + * migrated away when another CPU becomes available, is now constrained + * by the ability to push the higher priority task away, which might itself be + * in a migrate_disable() section, reducing it's available bandwidth. + * + * IOW it trades latency / moves the interference term, but it stays in the + * system, and as long as it remains unbounded, the system is not fully + * deterministic. * * * The reason we have it anyway. diff --git a/include/linux/sched.h b/include/linux/sched.h index 90a0c92741d7..3af9d52fe093 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -716,8 +716,9 @@ struct task_struct { cpumask_t cpus_mask; void *migration_pending; #if defined(CONFIG_SMP) && defined(CONFIG_PREEMPT_RT) - int migration_disabled; + unsigned short migration_disabled; #endif + unsigned short migration_flags; #ifdef CONFIG_PREEMPT_RCU int rcu_read_lock_nesting; -- cgit