cpuset/isolation: Honour kthreads preferred affinity

Documentation/core-api/housekeeping.rst

@@ -0,0 +1,111 @@
+======================================
+Housekeeping
+======================================
+
+
+CPU Isolation moves away kernel work that may otherwise run on any CPU.
+The purpose of its related features is to reduce the OS jitter that some
+extreme workloads can't stand, such as in some DPDK usecases.
+
+The kernel work moved away by CPU isolation is commonly described as
+"housekeeping" because it includes ground work that performs cleanups,
+statistics maintainance and actions relying on them, memory release,
+various deferrals etc...
+
+Sometimes housekeeping is just some unbound work (unbound workqueues,
+unbound timers, ...) that gets easily assigned to non-isolated CPUs.
+But sometimes housekeeping is tied to a specific CPU and requires
+elaborated tricks to be offloaded to non-isolated CPUs (RCU_NOCB, remote
+scheduler tick, etc...).
+
+Thus, a housekeeping CPU can be considered as the reverse of an isolated
+CPU. It is simply a CPU that can execute housekeeping work. There must
+always be at least one online housekeeping CPU at any time. The CPUs that
+are not	isolated are automatically assigned as housekeeping.
+
+Housekeeping is currently divided in four features described
+by the ``enum hk_type type``:
+
+1.	HK_TYPE_DOMAIN matches the work moved away by scheduler domain
+	isolation performed through ``isolcpus=domain`` boot parameter or
+	isolated cpuset partitions in cgroup v2. This includes scheduler
+	load balancing, unbound workqueues and timers.
+
+2.	HK_TYPE_KERNEL_NOISE matches the work moved away by tick isolation
+	performed through ``nohz_full=`` or ``isolcpus=nohz`` boot
+	parameters. This includes remote scheduler tick, vmstat and lockup
+	watchdog.
+
+3.	HK_TYPE_MANAGED_IRQ matches the IRQ handlers moved away by managed
+	IRQ isolation performed through ``isolcpus=managed_irq``.
+
+4.	HK_TYPE_DOMAIN_BOOT matches the work moved away by scheduler domain
+	isolation performed through ``isolcpus=domain`` only. It is similar
+	to HK_TYPE_DOMAIN except it ignores the isolation performed by
+	cpusets.
+
+
+Housekeeping cpumasks
+=================================
+
+Housekeeping cpumasks include the CPUs that can execute the work moved
+away by the matching isolation feature. These cpumasks are returned by
+the following function::
+
+	const struct cpumask *housekeeping_cpumask(enum hk_type type)
+
+By default, if neither ``nohz_full=``, nor ``isolcpus``, nor cpuset's
+isolated partitions are used, which covers most usecases, this function
+returns the cpu_possible_mask.
+
+Otherwise the function returns the cpumask complement of the isolation
+feature. For example:
+
+With isolcpus=domain,7 the following will return a mask with all possible
+CPUs except 7::
+
+	housekeeping_cpumask(HK_TYPE_DOMAIN)
+
+Similarly with nohz_full=5,6 the following will return a mask with all
+possible CPUs except 5,6::
+
+	housekeeping_cpumask(HK_TYPE_KERNEL_NOISE)
+
+
+Synchronization against cpusets
+=================================
+
+Cpuset can modify the HK_TYPE_DOMAIN housekeeping cpumask while creating,
+modifying or deleting an isolated partition.
+
+The users of HK_TYPE_DOMAIN cpumask must then make sure to synchronize
+properly against cpuset in order to make sure that:
+
+1.	The cpumask snapshot stays coherent.
+
+2.	No housekeeping work is queued on a newly made isolated CPU.
+
+3.	Pending housekeeping work that was queued to a non isolated
+	CPU which just turned isolated through cpuset must be flushed
+	before the related created/modified isolated partition is made
+	available to userspace.
+
+This synchronization is maintained by an RCU based scheme. The cpuset update
+side waits for an RCU grace period after updating the HK_TYPE_DOMAIN
+cpumask and before flushing pending works. On the read side, care must be
+taken to gather the housekeeping target election and the work enqueue within
+the same RCU read side critical section.
+
+A typical layout example would look like this on the update side
+(``housekeeping_update()``)::
+
+	rcu_assign_pointer(housekeeping_cpumasks[type], trial);
+	synchronize_rcu();
+	flush_workqueue(example_workqueue);
+
+And then on the read side::
+
+	rcu_read_lock();
+	cpu = housekeeping_any_cpu(HK_TYPE_DOMAIN);
+	queue_work_on(cpu, example_workqueue, work);
+	rcu_read_unlock();

Documentation/core-api/index.rst

@@ -25,6 +25,7 @@ it.
    symbol-namespaces
    asm-annotations
    real-time/index
+   housekeeping.rst
 
 Data structures and low-level utilities
 =======================================

arch/arm64/kernel/cpufeature.c

@@ -1656,6 +1656,18 @@ has_cpuid_feature(const struct arm64_cpu_capabilities *entry, int scope)
 	return feature_matches(val, entry);
 }
 
+/*
+ * 32 bits support CPUs can't be isolated because tasks may be
+ * arbitrarily affine to them, defeating the purpose of isolation.
+ */
+bool arch_isolated_cpus_can_update(struct cpumask *new_cpus)
+{
+	if (static_branch_unlikely(&arm64_mismatched_32bit_el0))
+		return !cpumask_intersects(cpu_32bit_el0_mask, new_cpus);
+	else
+		return true;
+}
+
 const struct cpumask *system_32bit_el0_cpumask(void)
 {
 	if (!system_supports_32bit_el0())
@@ -1669,7 +1681,7 @@ const struct cpumask *system_32bit_el0_cpumask(void)
 
 const struct cpumask *task_cpu_fallback_mask(struct task_struct *p)
 {
-	return __task_cpu_possible_mask(p, housekeeping_cpumask(HK_TYPE_TICK));
+	return __task_cpu_possible_mask(p, housekeeping_cpumask(HK_TYPE_DOMAIN));
 }
 
 static int __init parse_32bit_el0_param(char *str)
@@ -3987,8 +3999,8 @@ static int enable_mismatched_32bit_el0(unsigned int cpu)
 	bool cpu_32bit = false;
 
 	if (id_aa64pfr0_32bit_el0(info->reg_id_aa64pfr0)) {
-		if (!housekeeping_cpu(cpu, HK_TYPE_TICK))
-			pr_info("Treating adaptive-ticks CPU %u as 64-bit only\n", cpu);
+		if (!housekeeping_cpu(cpu, HK_TYPE_DOMAIN))
+			pr_info("Treating domain isolated CPU %u as 64-bit only\n", cpu);
 		else
 			cpu_32bit = true;
 	}

block/blk-mq.c

@@ -4257,12 +4257,16 @@ static void blk_mq_map_swqueue(struct request_queue *q)
 
 		/*
 		 * Rule out isolated CPUs from hctx->cpumask to avoid
-		 * running block kworker on isolated CPUs
+		 * running block kworker on isolated CPUs.
+		 * FIXME: cpuset should propagate further changes to isolated CPUs
+		 * here.
 		 */
+		rcu_read_lock();
 		for_each_cpu(cpu, hctx->cpumask) {
 			if (cpu_is_isolated(cpu))
 				cpumask_clear_cpu(cpu, hctx->cpumask);
 		}
+		rcu_read_unlock();
 
 		/*
 		 * Initialize batch roundrobin counts

drivers/base/cpu.c

@@ -291,7 +291,7 @@ static ssize_t print_cpus_isolated(struct device *dev,
 		return -ENOMEM;
 
 	cpumask_andnot(isolated, cpu_possible_mask,
-		       housekeeping_cpumask(HK_TYPE_DOMAIN));
+		       housekeeping_cpumask(HK_TYPE_DOMAIN_BOOT));
 	len = sysfs_emit(buf, "%*pbl\n", cpumask_pr_args(isolated));
 
 	free_cpumask_var(isolated);

drivers/pci/pci-driver.c

@@ -302,9 +302,8 @@ struct drv_dev_and_id {
 	const struct pci_device_id *id;
 };
 
-static long local_pci_probe(void *_ddi)
+static int local_pci_probe(struct drv_dev_and_id *ddi)
 {
-	struct drv_dev_and_id *ddi = _ddi;
 	struct pci_dev *pci_dev = ddi->dev;
 	struct pci_driver *pci_drv = ddi->drv;
 	struct device *dev = &pci_dev->dev;
@@ -338,6 +337,21 @@ static long local_pci_probe(void *_ddi)
 	return 0;
 }
 
+static struct workqueue_struct *pci_probe_wq;
+
+struct pci_probe_arg {
+	struct drv_dev_and_id *ddi;
+	struct work_struct work;
+	int ret;
+};
+
+static void local_pci_probe_callback(struct work_struct *work)
+{
+	struct pci_probe_arg *arg = container_of(work, struct pci_probe_arg, work);
+
+	arg->ret = local_pci_probe(arg->ddi);
+}
+
 static bool pci_physfn_is_probed(struct pci_dev *dev)
 {
 #ifdef CONFIG_PCI_IOV
@@ -362,40 +376,55 @@ static int pci_call_probe(struct pci_driver *drv, struct pci_dev *dev,
 	dev->is_probed = 1;
 
 	cpu_hotplug_disable();
-
 	/*
 	 * Prevent nesting work_on_cpu() for the case where a Virtual Function
 	 * device is probed from work_on_cpu() of the Physical device.
 	 */
 	if (node < 0 || node >= MAX_NUMNODES || !node_online(node) ||
 	    pci_physfn_is_probed(dev)) {
-		cpu = nr_cpu_ids;
+		error = local_pci_probe(&ddi);
 	} else {
-		cpumask_var_t wq_domain_mask;
+		struct pci_probe_arg arg = { .ddi = &ddi };
 
-		if (!zalloc_cpumask_var(&wq_domain_mask, GFP_KERNEL)) {
-			error = -ENOMEM;
-			goto out;
+		INIT_WORK_ONSTACK(&arg.work, local_pci_probe_callback);
+		/*
+		 * The target election and the enqueue of the work must be within
+		 * the same RCU read side section so that when the workqueue pool
+		 * is flushed after a housekeeping cpumask update, further readers
+		 * are guaranteed to queue the probing work to the appropriate
+		 * targets.
+		 */
+		rcu_read_lock();
+		cpu = cpumask_any_and(cpumask_of_node(node),
+				      housekeeping_cpumask(HK_TYPE_DOMAIN));
+
+		if (cpu < nr_cpu_ids) {
+			struct workqueue_struct *wq = pci_probe_wq;
+
+			if (WARN_ON_ONCE(!wq))
+				wq = system_percpu_wq;
+			queue_work_on(cpu, wq, &arg.work);
+			rcu_read_unlock();
+			flush_work(&arg.work);
+			error = arg.ret;
+		} else {
+			rcu_read_unlock();
+			error = local_pci_probe(&ddi);
 		}
-		cpumask_and(wq_domain_mask,
-			    housekeeping_cpumask(HK_TYPE_WQ),
-			    housekeeping_cpumask(HK_TYPE_DOMAIN));
 
-		cpu = cpumask_any_and(cpumask_of_node(node),
-				      wq_domain_mask);
-		free_cpumask_var(wq_domain_mask);
+		destroy_work_on_stack(&arg.work);
 	}
 
-	if (cpu < nr_cpu_ids)
-		error = work_on_cpu(cpu, local_pci_probe, &ddi);
-	else
-		error = local_pci_probe(&ddi);
-out:
 	dev->is_probed = 0;
 	cpu_hotplug_enable();
 	return error;
 }
 
+void pci_probe_flush_workqueue(void)
+{
+	flush_workqueue(pci_probe_wq);
+}
+
 /**
  * __pci_device_probe - check if a driver wants to claim a specific PCI device
  * @drv: driver to call to check if it wants the PCI device
@@ -1733,6 +1762,10 @@ static int __init pci_driver_init(void)
 {
 	int ret;
 
+	pci_probe_wq = alloc_workqueue("sync_wq", WQ_PERCPU, 0);
+	if (!pci_probe_wq)
+		return -ENOMEM;
+
 	ret = bus_register(&pci_bus_type);
 	if (ret)
 		return ret;

include/linux/cpu.h

@@ -229,4 +229,8 @@ static inline bool cpu_attack_vector_mitigated(enum cpu_attack_vectors v)
 #define smt_mitigations SMT_MITIGATIONS_OFF
 #endif
 
+struct cpumask;
+
+bool arch_isolated_cpus_can_update(struct cpumask *new_cpus);
+
 #endif /* _LINUX_CPU_H_ */

include/linux/cpuhplock.h

@@ -13,6 +13,7 @@
 struct device;
 
 extern int lockdep_is_cpus_held(void);
+extern int lockdep_is_cpus_write_held(void);
 
 #ifdef CONFIG_HOTPLUG_CPU
 void cpus_write_lock(void);

include/linux/cpuset.h

@@ -18,6 +18,8 @@
 #include <linux/mmu_context.h>
 #include <linux/jump_label.h>
 
+extern bool lockdep_is_cpuset_held(void);
+
 #ifdef CONFIG_CPUSETS
 
 /*
@@ -77,7 +79,6 @@ extern void cpuset_unlock(void);
 extern void cpuset_cpus_allowed_locked(struct task_struct *p, struct cpumask *mask);
 extern void cpuset_cpus_allowed(struct task_struct *p, struct cpumask *mask);
 extern bool cpuset_cpus_allowed_fallback(struct task_struct *p);
-extern bool cpuset_cpu_is_isolated(int cpu);
 extern nodemask_t cpuset_mems_allowed(struct task_struct *p);
 #define cpuset_current_mems_allowed (current->mems_allowed)
 void cpuset_init_current_mems_allowed(void);
@@ -213,11 +214,6 @@ static inline bool cpuset_cpus_allowed_fallback(struct task_struct *p)
 	return false;
 }
 
-static inline bool cpuset_cpu_is_isolated(int cpu)
-{
-	return false;
-}
-
 static inline nodemask_t cpuset_mems_allowed(struct task_struct *p)
 {
 	return node_possible_map;

include/linux/kthread.h

@@ -100,6 +100,7 @@ void kthread_unpark(struct task_struct *k);
 void kthread_parkme(void);
 void kthread_exit(long result) __noreturn;
 void kthread_complete_and_exit(struct completion *, long) __noreturn;
+int kthreads_update_housekeeping(void);
 
 int kthreadd(void *unused);
 extern struct task_struct *kthreadd_task;

include/linux/memcontrol.h

@@ -1037,6 +1037,8 @@ static inline u64 cgroup_id_from_mm(struct mm_struct *mm)
 	return id;
 }
 
+void mem_cgroup_flush_workqueue(void);
+
 extern int mem_cgroup_init(void);
 #else /* CONFIG_MEMCG */
 
@@ -1436,6 +1438,8 @@ static inline u64 cgroup_id_from_mm(struct mm_struct *mm)
 	return 0;
 }
 
+static inline void mem_cgroup_flush_workqueue(void) { }
+
 static inline int mem_cgroup_init(void) { return 0; }
 #endif /* CONFIG_MEMCG */
 

include/linux/mmu_context.h

@@ -24,7 +24,7 @@ static inline void leave_mm(void) { }
 #ifndef task_cpu_possible_mask
 # define task_cpu_possible_mask(p)	cpu_possible_mask
 # define task_cpu_possible(cpu, p)	true
-# define task_cpu_fallback_mask(p)	housekeeping_cpumask(HK_TYPE_TICK)
+# define task_cpu_fallback_mask(p)	housekeeping_cpumask(HK_TYPE_DOMAIN)
 #else
 # define task_cpu_possible(cpu, p)	cpumask_test_cpu((cpu), task_cpu_possible_mask(p))
 #endif