X-Git-Url: http://ftp.safe.ca/?a=blobdiff_plain;f=kernel%2Fcpuset.c;h=02b9611eadde3ebe638b9c24ffbbb5ec8ada5c06;hb=9b8eb4d14767209c83087063352cd04266ecdfd1;hp=8b35fbd8292f2d5b53b613eab630c67af87ec6f8;hpb=1d3504fcf5606579d60b649d19f44b3871c1ddae;p=safe%2Fjmp%2Flinux-2.6 diff --git a/kernel/cpuset.c b/kernel/cpuset.c index 8b35fbd..02b9611 100644 --- a/kernel/cpuset.c +++ b/kernel/cpuset.c @@ -14,6 +14,8 @@ * 2003-10-22 Updates by Stephen Hemminger. * 2004 May-July Rework by Paul Jackson. * 2006 Rework by Paul Menage to use generic cgroups + * 2008 Rework of the scheduler domains and CPU hotplug handling + * by Max Krasnyansky * * This file is subject to the terms and conditions of the GNU General Public * License. See the file COPYING in the main directory of the Linux @@ -34,6 +36,7 @@ #include #include #include +#include #include #include #include @@ -54,11 +57,18 @@ #include #include #include -#include #include #include /* + * Workqueue for cpuset related tasks. + * + * Using kevent workqueue may cause deadlock when memory_migrate + * is set. So we create a separate workqueue thread for cpuset. + */ +static struct workqueue_struct *cpuset_wq; + +/* * Tracks how many cpusets are currently defined in system. * When there is only one cpuset (the root cpuset) we can * short circuit some hooks. @@ -82,17 +92,11 @@ struct cpuset { struct cgroup_subsys_state css; unsigned long flags; /* "unsigned long" so bitops work */ - cpumask_t cpus_allowed; /* CPUs allowed to tasks in cpuset */ + cpumask_var_t cpus_allowed; /* CPUs allowed to tasks in cpuset */ nodemask_t mems_allowed; /* Memory Nodes allowed to tasks */ struct cpuset *parent; /* my parent */ - /* - * Copy of global cpuset_mems_generation as of the most - * recent time this cpuset changed its mems_allowed. - */ - int mems_generation; - struct fmeter fmeter; /* memory_pressure filter */ /* partition number for rebuild_sched_domains() */ @@ -118,15 +122,12 @@ static inline struct cpuset *task_cs(struct task_struct *task) return container_of(task_subsys_state(task, cpuset_subsys_id), struct cpuset, css); } -struct cpuset_hotplug_scanner { - struct cgroup_scanner scan; - struct cgroup *to; -}; /* bits in struct cpuset flags field */ typedef enum { CS_CPU_EXCLUSIVE, CS_MEM_EXCLUSIVE, + CS_MEM_HARDWALL, CS_MEMORY_MIGRATE, CS_SCHED_LOAD_BALANCE, CS_SPREAD_PAGE, @@ -144,6 +145,11 @@ static inline int is_mem_exclusive(const struct cpuset *cs) return test_bit(CS_MEM_EXCLUSIVE, &cs->flags); } +static inline int is_mem_hardwall(const struct cpuset *cs) +{ + return test_bit(CS_MEM_HARDWALL, &cs->flags); +} + static inline int is_sched_load_balance(const struct cpuset *cs) { return test_bit(CS_SCHED_LOAD_BALANCE, &cs->flags); @@ -164,31 +170,8 @@ static inline int is_spread_slab(const struct cpuset *cs) return test_bit(CS_SPREAD_SLAB, &cs->flags); } -/* - * Increment this integer everytime any cpuset changes its - * mems_allowed value. Users of cpusets can track this generation - * number, and avoid having to lock and reload mems_allowed unless - * the cpuset they're using changes generation. - * - * A single, global generation is needed because cpuset_attach_task() could - * reattach a task to a different cpuset, which must not have its - * generation numbers aliased with those of that tasks previous cpuset. - * - * Generations are needed for mems_allowed because one task cannot - * modify another's memory placement. So we must enable every task, - * on every visit to __alloc_pages(), to efficiently check whether - * its current->cpuset->mems_allowed has changed, requiring an update - * of its current->mems_allowed. - * - * Since writes to cpuset_mems_generation are guarded by the cgroup lock - * there is no need to mark it atomic. - */ -static int cpuset_mems_generation; - static struct cpuset top_cpuset = { .flags = ((1 << CS_CPU_EXCLUSIVE) | (1 << CS_MEM_EXCLUSIVE)), - .cpus_allowed = CPU_MASK_ALL, - .mems_allowed = NODE_MASK_ALL, }; /* @@ -218,12 +201,9 @@ static struct cpuset top_cpuset = { * If a task is only holding callback_mutex, then it has read-only * access to cpusets. * - * The task_struct fields mems_allowed and mems_generation may only - * be accessed in the context of that task, so require no locks. - * - * The cpuset_common_file_write handler for operations that modify - * the cpuset hierarchy holds cgroup_mutex across the entire operation, - * single threading all such cpuset modifications across the system. + * Now, the task_struct fields mems_allowed and mempolicy may be changed + * by other task, we use alloc_lock in the task_struct fields to protect + * them. * * The cpuset_common_file_read() handlers only hold callback_mutex across * small pieces of code, such as when reading out possibly multi-word @@ -235,9 +215,22 @@ static struct cpuset top_cpuset = { static DEFINE_MUTEX(callback_mutex); -/* This is ugly, but preserves the userspace API for existing cpuset +/* + * cpuset_buffer_lock protects both the cpuset_name and cpuset_nodelist + * buffers. They are statically allocated to prevent using excess stack + * when calling cpuset_print_task_mems_allowed(). + */ +#define CPUSET_NAME_LEN (128) +#define CPUSET_NODELIST_LEN (256) +static char cpuset_name[CPUSET_NAME_LEN]; +static char cpuset_nodelist[CPUSET_NODELIST_LEN]; +static DEFINE_SPINLOCK(cpuset_buffer_lock); + +/* + * This is ugly, but preserves the userspace API for existing cpuset * users. If someone tries to mount the "cpuset" filesystem, we - * silently switch it to mount "cgroup" instead */ + * silently switch it to mount "cgroup" instead + */ static int cpuset_get_sb(struct file_system_type *fs_type, int flags, const char *unused_dev_name, void *data, struct vfsmount *mnt) @@ -261,7 +254,7 @@ static struct file_system_type cpuset_fs_type = { }; /* - * Return in *pmask the portion of a cpusets's cpus_allowed that + * Return in pmask the portion of a cpusets's cpus_allowed that * are online. If none are online, walk up the cpuset hierarchy * until we find one that does have some online cpus. If we get * all the way to the top and still haven't found any online cpus, @@ -274,15 +267,16 @@ static struct file_system_type cpuset_fs_type = { * Call with callback_mutex held. */ -static void guarantee_online_cpus(const struct cpuset *cs, cpumask_t *pmask) +static void guarantee_online_cpus(const struct cpuset *cs, + struct cpumask *pmask) { - while (cs && !cpus_intersects(cs->cpus_allowed, cpu_online_map)) + while (cs && !cpumask_intersects(cs->cpus_allowed, cpu_online_mask)) cs = cs->parent; if (cs) - cpus_and(*pmask, cs->cpus_allowed, cpu_online_map); + cpumask_and(pmask, cs->cpus_allowed, cpu_online_mask); else - *pmask = cpu_online_map; - BUG_ON(!cpus_intersects(*pmask, cpu_online_map)); + cpumask_copy(pmask, cpu_online_mask); + BUG_ON(!cpumask_intersects(pmask, cpu_online_mask)); } /* @@ -311,80 +305,22 @@ static void guarantee_online_mems(const struct cpuset *cs, nodemask_t *pmask) BUG_ON(!nodes_intersects(*pmask, node_states[N_HIGH_MEMORY])); } -/** - * cpuset_update_task_memory_state - update task memory placement - * - * If the current tasks cpusets mems_allowed changed behind our - * backs, update current->mems_allowed, mems_generation and task NUMA - * mempolicy to the new value. - * - * Task mempolicy is updated by rebinding it relative to the - * current->cpuset if a task has its memory placement changed. - * Do not call this routine if in_interrupt(). - * - * Call without callback_mutex or task_lock() held. May be - * called with or without cgroup_mutex held. Thanks in part to - * 'the_top_cpuset_hack', the task's cpuset pointer will never - * be NULL. This routine also might acquire callback_mutex during - * call. - * - * Reading current->cpuset->mems_generation doesn't need task_lock - * to guard the current->cpuset derefence, because it is guarded - * from concurrent freeing of current->cpuset using RCU. - * - * The rcu_dereference() is technically probably not needed, - * as I don't actually mind if I see a new cpuset pointer but - * an old value of mems_generation. However this really only - * matters on alpha systems using cpusets heavily. If I dropped - * that rcu_dereference(), it would save them a memory barrier. - * For all other arch's, rcu_dereference is a no-op anyway, and for - * alpha systems not using cpusets, another planned optimization, - * avoiding the rcu critical section for tasks in the root cpuset - * which is statically allocated, so can't vanish, will make this - * irrelevant. Better to use RCU as intended, than to engage in - * some cute trick to save a memory barrier that is impossible to - * test, for alpha systems using cpusets heavily, which might not - * even exist. - * - * This routine is needed to update the per-task mems_allowed data, - * within the tasks context, when it is trying to allocate memory - * (in various mm/mempolicy.c routines) and notices that some other - * task has been modifying its cpuset. +/* + * update task's spread flag if cpuset's page/slab spread flag is set + * + * Called with callback_mutex/cgroup_mutex held */ - -void cpuset_update_task_memory_state(void) +static void cpuset_update_task_spread_flag(struct cpuset *cs, + struct task_struct *tsk) { - int my_cpusets_mem_gen; - struct task_struct *tsk = current; - struct cpuset *cs; - - if (task_cs(tsk) == &top_cpuset) { - /* Don't need rcu for top_cpuset. It's never freed. */ - my_cpusets_mem_gen = top_cpuset.mems_generation; - } else { - rcu_read_lock(); - my_cpusets_mem_gen = task_cs(current)->mems_generation; - rcu_read_unlock(); - } - - if (my_cpusets_mem_gen != tsk->cpuset_mems_generation) { - mutex_lock(&callback_mutex); - task_lock(tsk); - cs = task_cs(tsk); /* Maybe changed when task not locked */ - guarantee_online_mems(cs, &tsk->mems_allowed); - tsk->cpuset_mems_generation = cs->mems_generation; - if (is_spread_page(cs)) - tsk->flags |= PF_SPREAD_PAGE; - else - tsk->flags &= ~PF_SPREAD_PAGE; - if (is_spread_slab(cs)) - tsk->flags |= PF_SPREAD_SLAB; - else - tsk->flags &= ~PF_SPREAD_SLAB; - task_unlock(tsk); - mutex_unlock(&callback_mutex); - mpol_rebind_task(tsk, &tsk->mems_allowed); - } + if (is_spread_page(cs)) + tsk->flags |= PF_SPREAD_PAGE; + else + tsk->flags &= ~PF_SPREAD_PAGE; + if (is_spread_slab(cs)) + tsk->flags |= PF_SPREAD_SLAB; + else + tsk->flags &= ~PF_SPREAD_SLAB; } /* @@ -397,12 +333,43 @@ void cpuset_update_task_memory_state(void) static int is_cpuset_subset(const struct cpuset *p, const struct cpuset *q) { - return cpus_subset(p->cpus_allowed, q->cpus_allowed) && + return cpumask_subset(p->cpus_allowed, q->cpus_allowed) && nodes_subset(p->mems_allowed, q->mems_allowed) && is_cpu_exclusive(p) <= is_cpu_exclusive(q) && is_mem_exclusive(p) <= is_mem_exclusive(q); } +/** + * alloc_trial_cpuset - allocate a trial cpuset + * @cs: the cpuset that the trial cpuset duplicates + */ +static struct cpuset *alloc_trial_cpuset(const struct cpuset *cs) +{ + struct cpuset *trial; + + trial = kmemdup(cs, sizeof(*cs), GFP_KERNEL); + if (!trial) + return NULL; + + if (!alloc_cpumask_var(&trial->cpus_allowed, GFP_KERNEL)) { + kfree(trial); + return NULL; + } + cpumask_copy(trial->cpus_allowed, cs->cpus_allowed); + + return trial; +} + +/** + * free_trial_cpuset - free the trial cpuset + * @trial: the trial cpuset to be freed + */ +static void free_trial_cpuset(struct cpuset *trial) +{ + free_cpumask_var(trial->cpus_allowed); + kfree(trial); +} + /* * validate_change() - Used to validate that any proposed cpuset change * follows the structural rules for cpusets. @@ -452,7 +419,7 @@ static int validate_change(const struct cpuset *cur, const struct cpuset *trial) c = cgroup_cs(cont); if ((is_cpu_exclusive(trial) || is_cpu_exclusive(c)) && c != cur && - cpus_intersects(trial->cpus_allowed, c->cpus_allowed)) + cpumask_intersects(trial->cpus_allowed, c->cpus_allowed)) return -EINVAL; if ((is_mem_exclusive(trial) || is_mem_exclusive(c)) && c != cur && @@ -462,7 +429,7 @@ static int validate_change(const struct cpuset *cur, const struct cpuset *trial) /* Cpusets with tasks can't have empty cpus_allowed or mems_allowed */ if (cgroup_task_count(cur->css.cgroup)) { - if (cpus_empty(trial->cpus_allowed) || + if (cpumask_empty(trial->cpus_allowed) || nodes_empty(trial->mems_allowed)) { return -ENOSPC; } @@ -471,44 +438,63 @@ static int validate_change(const struct cpuset *cur, const struct cpuset *trial) return 0; } +#ifdef CONFIG_SMP /* - * Helper routine for rebuild_sched_domains(). + * Helper routine for generate_sched_domains(). * Do cpusets a, b have overlapping cpus_allowed masks? */ - static int cpusets_overlap(struct cpuset *a, struct cpuset *b) { - return cpus_intersects(a->cpus_allowed, b->cpus_allowed); + return cpumask_intersects(a->cpus_allowed, b->cpus_allowed); } static void update_domain_attr(struct sched_domain_attr *dattr, struct cpuset *c) { - if (!dattr) - return; if (dattr->relax_domain_level < c->relax_domain_level) dattr->relax_domain_level = c->relax_domain_level; return; } +static void +update_domain_attr_tree(struct sched_domain_attr *dattr, struct cpuset *c) +{ + LIST_HEAD(q); + + list_add(&c->stack_list, &q); + while (!list_empty(&q)) { + struct cpuset *cp; + struct cgroup *cont; + struct cpuset *child; + + cp = list_first_entry(&q, struct cpuset, stack_list); + list_del(q.next); + + if (cpumask_empty(cp->cpus_allowed)) + continue; + + if (is_sched_load_balance(cp)) + update_domain_attr(dattr, cp); + + list_for_each_entry(cont, &cp->css.cgroup->children, sibling) { + child = cgroup_cs(cont); + list_add_tail(&child->stack_list, &q); + } + } +} + /* - * rebuild_sched_domains() + * generate_sched_domains() * - * If the flag 'sched_load_balance' of any cpuset with non-empty - * 'cpus' changes, or if the 'cpus' allowed changes in any cpuset - * which has that flag enabled, or if any cpuset with a non-empty - * 'cpus' is removed, then call this routine to rebuild the - * scheduler's dynamic sched domains. - * - * This routine builds a partial partition of the systems CPUs - * (the set of non-overlappping cpumask_t's in the array 'part' - * below), and passes that partial partition to the kernel/sched.c - * partition_sched_domains() routine, which will rebuild the - * schedulers load balancing domains (sched domains) as specified - * by that partial partition. A 'partial partition' is a set of - * non-overlapping subsets whose union is a subset of that set. + * This function builds a partial partition of the systems CPUs + * A 'partial partition' is a set of non-overlapping subsets whose + * union is a subset of that set. + * The output of this function needs to be passed to kernel/sched.c + * partition_sched_domains() routine, which will rebuild the scheduler's + * load balancing domains (sched domains) as specified by that partial + * partition. * - * See "What is sched_load_balance" in Documentation/cpusets.txt + * See "What is sched_load_balance" in Documentation/cgroups/cpusets.txt * for a background explanation of this. * * Does not return errors, on the theory that the callers of this @@ -516,16 +502,10 @@ update_domain_attr(struct sched_domain_attr *dattr, struct cpuset *c) * domains when operating in the severe memory shortage situations * that could cause allocation failures below. * - * Call with cgroup_mutex held. May take callback_mutex during - * call due to the kfifo_alloc() and kmalloc() calls. May nest - * a call to the get_online_cpus()/put_online_cpus() pair. - * Must not be called holding callback_mutex, because we must not - * call get_online_cpus() while holding callback_mutex. Elsewhere - * the kernel nests callback_mutex inside get_online_cpus() calls. - * So the reverse nesting would risk an ABBA deadlock. + * Must be called with cgroup_lock held. * * The three key local variables below are: - * q - a kfifo queue of cpuset pointers, used to implement a + * q - a linked-list queue of cpuset pointers, used to implement a * top-down scan of all cpusets. This scan loads a pointer * to each cpuset marked is_sched_load_balance into the * array 'csa'. For our purposes, rebuilding the schedulers @@ -557,57 +537,70 @@ update_domain_attr(struct sched_domain_attr *dattr, struct cpuset *c) * element of the partition (one sched domain) to be passed to * partition_sched_domains(). */ - -static void rebuild_sched_domains(void) +static int generate_sched_domains(cpumask_var_t **domains, + struct sched_domain_attr **attributes) { - struct kfifo *q; /* queue of cpusets to be scanned */ + LIST_HEAD(q); /* queue of cpusets to be scanned */ struct cpuset *cp; /* scans q */ struct cpuset **csa; /* array of all cpuset ptrs */ int csn; /* how many cpuset ptrs in csa so far */ int i, j, k; /* indices for partition finding loops */ - cpumask_t *doms; /* resulting partition; i.e. sched domains */ + cpumask_var_t *doms; /* resulting partition; i.e. sched domains */ struct sched_domain_attr *dattr; /* attributes for custom domains */ - int ndoms; /* number of sched domains in result */ - int nslot; /* next empty doms[] cpumask_t slot */ + int ndoms = 0; /* number of sched domains in result */ + int nslot; /* next empty doms[] struct cpumask slot */ - q = NULL; - csa = NULL; doms = NULL; dattr = NULL; + csa = NULL; /* Special case for the 99% of systems with one, full, sched domain */ if (is_sched_load_balance(&top_cpuset)) { ndoms = 1; - doms = kmalloc(sizeof(cpumask_t), GFP_KERNEL); + doms = alloc_sched_domains(ndoms); if (!doms) - goto rebuild; + goto done; + dattr = kmalloc(sizeof(struct sched_domain_attr), GFP_KERNEL); if (dattr) { *dattr = SD_ATTR_INIT; - update_domain_attr(dattr, &top_cpuset); + update_domain_attr_tree(dattr, &top_cpuset); } - *doms = top_cpuset.cpus_allowed; - goto rebuild; - } + cpumask_copy(doms[0], top_cpuset.cpus_allowed); - q = kfifo_alloc(number_of_cpusets * sizeof(cp), GFP_KERNEL, NULL); - if (IS_ERR(q)) goto done; + } + csa = kmalloc(number_of_cpusets * sizeof(cp), GFP_KERNEL); if (!csa) goto done; csn = 0; - cp = &top_cpuset; - __kfifo_put(q, (void *)&cp, sizeof(cp)); - while (__kfifo_get(q, (void *)&cp, sizeof(cp))) { + list_add(&top_cpuset.stack_list, &q); + while (!list_empty(&q)) { struct cgroup *cont; struct cpuset *child; /* scans child cpusets of cp */ - if (is_sched_load_balance(cp)) + + cp = list_first_entry(&q, struct cpuset, stack_list); + list_del(q.next); + + if (cpumask_empty(cp->cpus_allowed)) + continue; + + /* + * All child cpusets contain a subset of the parent's cpus, so + * just skip them, and then we call update_domain_attr_tree() + * to calc relax_domain_level of the corresponding sched + * domain. + */ + if (is_sched_load_balance(cp)) { csa[csn++] = cp; + continue; + } + list_for_each_entry(cont, &cp->css.cgroup->children, sibling) { child = cgroup_cs(cont); - __kfifo_put(q, (void *)&child, sizeof(cp)); + list_add_tail(&child->stack_list, &q); } } @@ -638,91 +631,158 @@ restart: } } - /* Convert to */ - doms = kmalloc(ndoms * sizeof(cpumask_t), GFP_KERNEL); + /* + * Now we know how many domains to create. + * Convert to and populate cpu masks. + */ + doms = alloc_sched_domains(ndoms); if (!doms) - goto rebuild; + goto done; + + /* + * The rest of the code, including the scheduler, can deal with + * dattr==NULL case. No need to abort if alloc fails. + */ dattr = kmalloc(ndoms * sizeof(struct sched_domain_attr), GFP_KERNEL); for (nslot = 0, i = 0; i < csn; i++) { struct cpuset *a = csa[i]; + struct cpumask *dp; int apn = a->pn; - if (apn >= 0) { - cpumask_t *dp = doms + nslot; - - if (nslot == ndoms) { - static int warnings = 10; - if (warnings) { - printk(KERN_WARNING - "rebuild_sched_domains confused:" - " nslot %d, ndoms %d, csn %d, i %d," - " apn %d\n", - nslot, ndoms, csn, i, apn); - warnings--; - } - continue; + if (apn < 0) { + /* Skip completed partitions */ + continue; + } + + dp = doms[nslot]; + + if (nslot == ndoms) { + static int warnings = 10; + if (warnings) { + printk(KERN_WARNING + "rebuild_sched_domains confused:" + " nslot %d, ndoms %d, csn %d, i %d," + " apn %d\n", + nslot, ndoms, csn, i, apn); + warnings--; } + continue; + } - cpus_clear(*dp); - if (dattr) - *(dattr + nslot) = SD_ATTR_INIT; - for (j = i; j < csn; j++) { - struct cpuset *b = csa[j]; + cpumask_clear(dp); + if (dattr) + *(dattr + nslot) = SD_ATTR_INIT; + for (j = i; j < csn; j++) { + struct cpuset *b = csa[j]; - if (apn == b->pn) { - cpus_or(*dp, *dp, b->cpus_allowed); - b->pn = -1; - update_domain_attr(dattr, b); - } + if (apn == b->pn) { + cpumask_or(dp, dp, b->cpus_allowed); + if (dattr) + update_domain_attr_tree(dattr + nslot, b); + + /* Done with this partition */ + b->pn = -1; } - nslot++; } + nslot++; } BUG_ON(nslot != ndoms); -rebuild: - /* Have scheduler rebuild sched domains */ +done: + kfree(csa); + + /* + * Fallback to the default domain if kmalloc() failed. + * See comments in partition_sched_domains(). + */ + if (doms == NULL) + ndoms = 1; + + *domains = doms; + *attributes = dattr; + return ndoms; +} + +/* + * Rebuild scheduler domains. + * + * Call with neither cgroup_mutex held nor within get_online_cpus(). + * Takes both cgroup_mutex and get_online_cpus(). + * + * Cannot be directly called from cpuset code handling changes + * to the cpuset pseudo-filesystem, because it cannot be called + * from code that already holds cgroup_mutex. + */ +static void do_rebuild_sched_domains(struct work_struct *unused) +{ + struct sched_domain_attr *attr; + cpumask_var_t *doms; + int ndoms; + get_online_cpus(); - partition_sched_domains(ndoms, doms, dattr); + + /* Generate domain masks and attrs */ + cgroup_lock(); + ndoms = generate_sched_domains(&doms, &attr); + cgroup_unlock(); + + /* Have scheduler rebuild the domains */ + partition_sched_domains(ndoms, doms, attr); + put_online_cpus(); +} +#else /* !CONFIG_SMP */ +static void do_rebuild_sched_domains(struct work_struct *unused) +{ +} -done: - if (q && !IS_ERR(q)) - kfifo_free(q); - kfree(csa); - /* Don't kfree(doms) -- partition_sched_domains() does that. */ - /* Don't kfree(dattr) -- partition_sched_domains() does that. */ +static int generate_sched_domains(cpumask_var_t **domains, + struct sched_domain_attr **attributes) +{ + *domains = NULL; + return 1; } +#endif /* CONFIG_SMP */ -static inline int started_after_time(struct task_struct *t1, - struct timespec *time, - struct task_struct *t2) +static DECLARE_WORK(rebuild_sched_domains_work, do_rebuild_sched_domains); + +/* + * Rebuild scheduler domains, asynchronously via workqueue. + * + * If the flag 'sched_load_balance' of any cpuset with non-empty + * 'cpus' changes, or if the 'cpus' allowed changes in any cpuset + * which has that flag enabled, or if any cpuset with a non-empty + * 'cpus' is removed, then call this routine to rebuild the + * scheduler's dynamic sched domains. + * + * The rebuild_sched_domains() and partition_sched_domains() + * routines must nest cgroup_lock() inside get_online_cpus(), + * but such cpuset changes as these must nest that locking the + * other way, holding cgroup_lock() for much of the code. + * + * So in order to avoid an ABBA deadlock, the cpuset code handling + * these user changes delegates the actual sched domain rebuilding + * to a separate workqueue thread, which ends up processing the + * above do_rebuild_sched_domains() function. + */ +static void async_rebuild_sched_domains(void) { - int start_diff = timespec_compare(&t1->start_time, time); - if (start_diff > 0) { - return 1; - } else if (start_diff < 0) { - return 0; - } else { - /* - * Arbitrarily, if two processes started at the same - * time, we'll say that the lower pointer value - * started first. Note that t2 may have exited by now - * so this may not be a valid pointer any longer, but - * that's fine - it still serves to distinguish - * between two tasks started (effectively) - * simultaneously. - */ - return t1 > t2; - } + queue_work(cpuset_wq, &rebuild_sched_domains_work); } -static inline int started_after(void *p1, void *p2) +/* + * Accomplishes the same scheduler domain rebuild as the above + * async_rebuild_sched_domains(), however it directly calls the + * rebuild routine synchronously rather than calling it via an + * asynchronous work thread. + * + * This can only be called from code that is not holding + * cgroup_mutex (not nested in a cgroup_lock() call.) + */ +void rebuild_sched_domains(void) { - struct task_struct *t1 = p1; - struct task_struct *t2 = p2; - return started_after_time(t1, &t2->start_time, t2); + do_rebuild_sched_domains(NULL); } /** @@ -735,9 +795,10 @@ static inline int started_after(void *p1, void *p2) * Return nonzero if this tasks's cpus_allowed mask should be changed (in other * words, if its mask is not equal to its cpuset's mask). */ -int cpuset_test_cpumask(struct task_struct *tsk, struct cgroup_scanner *scan) +static int cpuset_test_cpumask(struct task_struct *tsk, + struct cgroup_scanner *scan) { - return !cpus_equal(tsk->cpus_allowed, + return !cpumask_equal(&tsk->cpus_allowed, (cgroup_cs(scan->cg))->cpus_allowed); } @@ -752,9 +813,34 @@ int cpuset_test_cpumask(struct task_struct *tsk, struct cgroup_scanner *scan) * We don't need to re-check for the cgroup/cpuset membership, since we're * holding cgroup_lock() at this point. */ -void cpuset_change_cpumask(struct task_struct *tsk, struct cgroup_scanner *scan) +static void cpuset_change_cpumask(struct task_struct *tsk, + struct cgroup_scanner *scan) +{ + set_cpus_allowed_ptr(tsk, ((cgroup_cs(scan->cg))->cpus_allowed)); +} + +/** + * update_tasks_cpumask - Update the cpumasks of tasks in the cpuset. + * @cs: the cpuset in which each task's cpus_allowed mask needs to be changed + * @heap: if NULL, defer allocating heap memory to cgroup_scan_tasks() + * + * Called with cgroup_mutex held + * + * The cgroup_scan_tasks() function will scan all the tasks in a cgroup, + * calling callback functions for each. + * + * No return value. It's guaranteed that cgroup_scan_tasks() always returns 0 + * if @heap != NULL. + */ +static void update_tasks_cpumask(struct cpuset *cs, struct ptr_heap *heap) { - set_cpus_allowed_ptr(tsk, &((cgroup_cs(scan->cg))->cpus_allowed)); + struct cgroup_scanner scan; + + scan.cg = cs->css.cgroup; + scan.test_task = cpuset_test_cpumask; + scan.process_task = cpuset_change_cpumask; + scan.heap = heap; + cgroup_scan_tasks(&scan); } /** @@ -762,10 +848,9 @@ void cpuset_change_cpumask(struct task_struct *tsk, struct cgroup_scanner *scan) * @cs: the cpuset to consider * @buf: buffer of cpu numbers written to this cpuset */ -static int update_cpumask(struct cpuset *cs, char *buf) +static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs, + const char *buf) { - struct cpuset trialcs; - struct cgroup_scanner scan; struct ptr_heap heap; int retval; int is_load_balanced; @@ -774,54 +859,50 @@ static int update_cpumask(struct cpuset *cs, char *buf) if (cs == &top_cpuset) return -EACCES; - trialcs = *cs; - /* * An empty cpus_allowed is ok only if the cpuset has no tasks. * Since cpulist_parse() fails on an empty mask, we special case * that parsing. The validate_change() call ensures that cpusets * with tasks have cpus. */ - buf = strstrip(buf); if (!*buf) { - cpus_clear(trialcs.cpus_allowed); + cpumask_clear(trialcs->cpus_allowed); } else { - retval = cpulist_parse(buf, trialcs.cpus_allowed); + retval = cpulist_parse(buf, trialcs->cpus_allowed); if (retval < 0) return retval; + + if (!cpumask_subset(trialcs->cpus_allowed, cpu_active_mask)) + return -EINVAL; } - cpus_and(trialcs.cpus_allowed, trialcs.cpus_allowed, cpu_online_map); - retval = validate_change(cs, &trialcs); + retval = validate_change(cs, trialcs); if (retval < 0) return retval; /* Nothing to do if the cpus didn't change */ - if (cpus_equal(cs->cpus_allowed, trialcs.cpus_allowed)) + if (cpumask_equal(cs->cpus_allowed, trialcs->cpus_allowed)) return 0; - retval = heap_init(&heap, PAGE_SIZE, GFP_KERNEL, &started_after); + retval = heap_init(&heap, PAGE_SIZE, GFP_KERNEL, NULL); if (retval) return retval; - is_load_balanced = is_sched_load_balance(&trialcs); + is_load_balanced = is_sched_load_balance(trialcs); mutex_lock(&callback_mutex); - cs->cpus_allowed = trialcs.cpus_allowed; + cpumask_copy(cs->cpus_allowed, trialcs->cpus_allowed); mutex_unlock(&callback_mutex); /* * Scan tasks in the cpuset, and update the cpumasks of any * that need an update. */ - scan.cg = cs->css.cgroup; - scan.test_task = cpuset_test_cpumask; - scan.process_task = cpuset_change_cpumask; - scan.heap = &heap; - cgroup_scan_tasks(&scan); + update_tasks_cpumask(cs, &heap); + heap_free(&heap); if (is_load_balanced) - rebuild_sched_domains(); + async_rebuild_sched_domains(); return 0; } @@ -839,21 +920,10 @@ static int update_cpumask(struct cpuset *cs, char *buf) * call to guarantee_online_mems(), as we know no one is changing * our task's cpuset. * - * Hold callback_mutex around the two modifications of our tasks - * mems_allowed to synchronize with cpuset_mems_allowed(). - * * While the mm_struct we are migrating is typically from some * other task, the task_struct mems_allowed that we are hacking * is for our current task, which must allocate new pages for that * migrating memory region. - * - * We call cpuset_update_task_memory_state() before hacking - * our tasks mems_allowed, so that we are assured of being in - * sync with our tasks cpuset, and in particular, callbacks to - * cpuset_update_task_memory_state() from nested page allocations - * won't see any mismatch of our cpuset and task mems_generation - * values, so won't overwrite our hacked tasks mems_allowed - * nodemask. */ static void cpuset_migrate_mm(struct mm_struct *mm, const nodemask_t *from, @@ -861,55 +931,187 @@ static void cpuset_migrate_mm(struct mm_struct *mm, const nodemask_t *from, { struct task_struct *tsk = current; - cpuset_update_task_memory_state(); - - mutex_lock(&callback_mutex); tsk->mems_allowed = *to; - mutex_unlock(&callback_mutex); do_migrate_pages(mm, from, to, MPOL_MF_MOVE_ALL); - mutex_lock(&callback_mutex); guarantee_online_mems(task_cs(tsk),&tsk->mems_allowed); - mutex_unlock(&callback_mutex); +} + +/* + * cpuset_change_task_nodemask - change task's mems_allowed and mempolicy + * @tsk: the task to change + * @newmems: new nodes that the task will be set + * + * In order to avoid seeing no nodes if the old and new nodes are disjoint, + * we structure updates as setting all new allowed nodes, then clearing newly + * disallowed ones. + */ +static void cpuset_change_task_nodemask(struct task_struct *tsk, + nodemask_t *newmems) +{ +repeat: + /* + * Allow tasks that have access to memory reserves because they have + * been OOM killed to get memory anywhere. + */ + if (unlikely(test_thread_flag(TIF_MEMDIE))) + return; + if (current->flags & PF_EXITING) /* Let dying task have memory */ + return; + + task_lock(tsk); + nodes_or(tsk->mems_allowed, tsk->mems_allowed, *newmems); + mpol_rebind_task(tsk, newmems, MPOL_REBIND_STEP1); + + + /* + * ensure checking ->mems_allowed_change_disable after setting all new + * allowed nodes. + * + * the read-side task can see an nodemask with new allowed nodes and + * old allowed nodes. and if it allocates page when cpuset clears newly + * disallowed ones continuous, it can see the new allowed bits. + * + * And if setting all new allowed nodes is after the checking, setting + * all new allowed nodes and clearing newly disallowed ones will be done + * continuous, and the read-side task may find no node to alloc page. + */ + smp_mb(); + + /* + * Allocation of memory is very fast, we needn't sleep when waiting + * for the read-side. + */ + while (ACCESS_ONCE(tsk->mems_allowed_change_disable)) { + task_unlock(tsk); + if (!task_curr(tsk)) + yield(); + goto repeat; + } + + /* + * ensure checking ->mems_allowed_change_disable before clearing all new + * disallowed nodes. + * + * if clearing newly disallowed bits before the checking, the read-side + * task may find no node to alloc page. + */ + smp_mb(); + + mpol_rebind_task(tsk, newmems, MPOL_REBIND_STEP2); + tsk->mems_allowed = *newmems; + task_unlock(tsk); +} + +/* + * Update task's mems_allowed and rebind its mempolicy and vmas' mempolicy + * of it to cpuset's new mems_allowed, and migrate pages to new nodes if + * memory_migrate flag is set. Called with cgroup_mutex held. + */ +static void cpuset_change_nodemask(struct task_struct *p, + struct cgroup_scanner *scan) +{ + struct mm_struct *mm; + struct cpuset *cs; + int migrate; + const nodemask_t *oldmem = scan->data; + NODEMASK_ALLOC(nodemask_t, newmems, GFP_KERNEL); + + if (!newmems) + return; + + cs = cgroup_cs(scan->cg); + guarantee_online_mems(cs, newmems); + + cpuset_change_task_nodemask(p, newmems); + + NODEMASK_FREE(newmems); + + mm = get_task_mm(p); + if (!mm) + return; + + migrate = is_memory_migrate(cs); + + mpol_rebind_mm(mm, &cs->mems_allowed); + if (migrate) + cpuset_migrate_mm(mm, oldmem, &cs->mems_allowed); + mmput(mm); +} + +static void *cpuset_being_rebound; + +/** + * update_tasks_nodemask - Update the nodemasks of tasks in the cpuset. + * @cs: the cpuset in which each task's mems_allowed mask needs to be changed + * @oldmem: old mems_allowed of cpuset cs + * @heap: if NULL, defer allocating heap memory to cgroup_scan_tasks() + * + * Called with cgroup_mutex held + * No return value. It's guaranteed that cgroup_scan_tasks() always returns 0 + * if @heap != NULL. + */ +static void update_tasks_nodemask(struct cpuset *cs, const nodemask_t *oldmem, + struct ptr_heap *heap) +{ + struct cgroup_scanner scan; + + cpuset_being_rebound = cs; /* causes mpol_dup() rebind */ + + scan.cg = cs->css.cgroup; + scan.test_task = NULL; + scan.process_task = cpuset_change_nodemask; + scan.heap = heap; + scan.data = (nodemask_t *)oldmem; + + /* + * The mpol_rebind_mm() call takes mmap_sem, which we couldn't + * 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 cgroup_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. + */ + cgroup_scan_tasks(&scan); + + /* We're done rebinding vmas to this cpuset's new mems_allowed. */ + cpuset_being_rebound = NULL; } /* * Handle user request to change the 'mems' memory placement * of a cpuset. Needs to validate the request, update the - * cpusets mems_allowed and mems_generation, and for each - * task in the cpuset, rebind any vma mempolicies and if - * the cpuset is marked 'memory_migrate', migrate the tasks - * pages to the new memory. + * cpusets mems_allowed, and for each task in the cpuset, + * update mems_allowed and rebind task's mempolicy and any vma + * mempolicies and if the cpuset is marked 'memory_migrate', + * migrate the tasks pages to the new memory. * * Call with cgroup_mutex held. May take callback_mutex during call. * Will take tasklist_lock, scan tasklist for tasks in cpuset cs, * lock each such tasks mm->mmap_sem, scan its vma's and rebind * their mempolicies to the cpusets new mems_allowed. */ - -static void *cpuset_being_rebound; - -static int update_nodemask(struct cpuset *cs, char *buf) +static int update_nodemask(struct cpuset *cs, struct cpuset *trialcs, + const char *buf) { - struct cpuset trialcs; - nodemask_t oldmem; - struct task_struct *p; - struct mm_struct **mmarray; - int i, n, ntasks; - int migrate; - int fudge; + NODEMASK_ALLOC(nodemask_t, oldmem, GFP_KERNEL); int retval; - struct cgroup_iter it; + struct ptr_heap heap; + + if (!oldmem) + return -ENOMEM; /* * top_cpuset.mems_allowed tracks node_stats[N_HIGH_MEMORY]; * it's read-only */ - if (cs == &top_cpuset) - return -EACCES; - - trialcs = *cs; + if (cs == &top_cpuset) { + retval = -EACCES; + goto done; + } /* * An empty mems_allowed is ok iff there are no tasks in the cpuset. @@ -917,104 +1119,41 @@ static int update_nodemask(struct cpuset *cs, char *buf) * that parsing. The validate_change() call ensures that cpusets * with tasks have memory. */ - buf = strstrip(buf); if (!*buf) { - nodes_clear(trialcs.mems_allowed); + nodes_clear(trialcs->mems_allowed); } else { - retval = nodelist_parse(buf, trialcs.mems_allowed); + retval = nodelist_parse(buf, trialcs->mems_allowed); if (retval < 0) goto done; + + if (!nodes_subset(trialcs->mems_allowed, + node_states[N_HIGH_MEMORY])) { + retval = -EINVAL; + goto done; + } } - nodes_and(trialcs.mems_allowed, trialcs.mems_allowed, - node_states[N_HIGH_MEMORY]); - oldmem = cs->mems_allowed; - if (nodes_equal(oldmem, trialcs.mems_allowed)) { + *oldmem = cs->mems_allowed; + if (nodes_equal(*oldmem, trialcs->mems_allowed)) { retval = 0; /* Too easy - nothing to do */ goto done; } - retval = validate_change(cs, &trialcs); + retval = validate_change(cs, trialcs); + if (retval < 0) + goto done; + + retval = heap_init(&heap, PAGE_SIZE, GFP_KERNEL, NULL); if (retval < 0) goto done; mutex_lock(&callback_mutex); - cs->mems_allowed = trialcs.mems_allowed; - cs->mems_generation = cpuset_mems_generation++; + cs->mems_allowed = trialcs->mems_allowed; mutex_unlock(&callback_mutex); - cpuset_being_rebound = cs; /* causes mpol_copy() rebind */ - - fudge = 10; /* spare mmarray[] slots */ - fudge += cpus_weight(cs->cpus_allowed); /* imagine one fork-bomb/cpu */ - retval = -ENOMEM; - - /* - * Allocate mmarray[] to hold mm reference for each task - * in cpuset cs. Can't kmalloc GFP_KERNEL while holding - * tasklist_lock. We could use GFP_ATOMIC, but with a - * few more lines of code, we can retry until we get a big - * enough mmarray[] w/o using GFP_ATOMIC. - */ - while (1) { - ntasks = cgroup_task_count(cs->css.cgroup); /* guess */ - ntasks += fudge; - mmarray = kmalloc(ntasks * sizeof(*mmarray), GFP_KERNEL); - if (!mmarray) - goto done; - read_lock(&tasklist_lock); /* block fork */ - if (cgroup_task_count(cs->css.cgroup) <= ntasks) - break; /* got enough */ - read_unlock(&tasklist_lock); /* try again */ - kfree(mmarray); - } - - n = 0; - - /* Load up mmarray[] with mm reference for each task in cpuset. */ - cgroup_iter_start(cs->css.cgroup, &it); - while ((p = cgroup_iter_next(cs->css.cgroup, &it))) { - struct mm_struct *mm; - - if (n >= ntasks) { - printk(KERN_WARNING - "Cpuset mempolicy rebind incomplete.\n"); - break; - } - mm = get_task_mm(p); - if (!mm) - continue; - mmarray[n++] = mm; - } - cgroup_iter_end(cs->css.cgroup, &it); - read_unlock(&tasklist_lock); + update_tasks_nodemask(cs, oldmem, &heap); - /* - * Now that we've dropped the tasklist spinlock, we can - * rebind the vma mempolicies of each mm in mmarray[] to their - * new cpuset, and release that mm. The mpol_rebind_mm() - * call takes mmap_sem, which we couldn't take while holding - * tasklist_lock. Forks can happen again now - the mpol_copy() - * cpuset_being_rebound check will catch such forks, and rebind - * their vma mempolicies too. Because we still hold the global - * cgroup_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. - */ - migrate = is_memory_migrate(cs); - for (i = 0; i < n; i++) { - struct mm_struct *mm = mmarray[i]; - - mpol_rebind_mm(mm, &cs->mems_allowed); - if (migrate) - cpuset_migrate_mm(mm, &oldmem, &cs->mems_allowed); - mmput(mm); - } - - /* We're done rebinding vmas to this cpuset's new mems_allowed. */ - kfree(mmarray); - cpuset_being_rebound = NULL; - retval = 0; + heap_free(&heap); done: + NODEMASK_FREE(oldmem); return retval; } @@ -1023,77 +1162,117 @@ int current_cpuset_is_being_rebound(void) return task_cs(current) == cpuset_being_rebound; } -/* - * Call with cgroup_mutex held. - */ - -static int update_memory_pressure_enabled(struct cpuset *cs, char *buf) -{ - if (simple_strtoul(buf, NULL, 10) != 0) - cpuset_memory_pressure_enabled = 1; - else - cpuset_memory_pressure_enabled = 0; - return 0; -} - -static int update_relax_domain_level(struct cpuset *cs, char *buf) +static int update_relax_domain_level(struct cpuset *cs, s64 val) { - int val = simple_strtol(buf, NULL, 10); - - if (val < 0) - val = -1; +#ifdef CONFIG_SMP + if (val < -1 || val >= SD_LV_MAX) + return -EINVAL; +#endif if (val != cs->relax_domain_level) { cs->relax_domain_level = val; - rebuild_sched_domains(); + if (!cpumask_empty(cs->cpus_allowed) && + is_sched_load_balance(cs)) + async_rebuild_sched_domains(); } return 0; } /* + * cpuset_change_flag - make a task's spread flags the same as its cpuset's + * @tsk: task to be updated + * @scan: struct cgroup_scanner containing the cgroup of the task + * + * Called by cgroup_scan_tasks() for each task in a cgroup. + * + * We don't need to re-check for the cgroup/cpuset membership, since we're + * holding cgroup_lock() at this point. + */ +static void cpuset_change_flag(struct task_struct *tsk, + struct cgroup_scanner *scan) +{ + cpuset_update_task_spread_flag(cgroup_cs(scan->cg), tsk); +} + +/* + * update_tasks_flags - update the spread flags of tasks in the cpuset. + * @cs: the cpuset in which each task's spread flags needs to be changed + * @heap: if NULL, defer allocating heap memory to cgroup_scan_tasks() + * + * Called with cgroup_mutex held + * + * The cgroup_scan_tasks() function will scan all the tasks in a cgroup, + * calling callback functions for each. + * + * No return value. It's guaranteed that cgroup_scan_tasks() always returns 0 + * if @heap != NULL. + */ +static void update_tasks_flags(struct cpuset *cs, struct ptr_heap *heap) +{ + struct cgroup_scanner scan; + + scan.cg = cs->css.cgroup; + scan.test_task = NULL; + scan.process_task = cpuset_change_flag; + scan.heap = heap; + cgroup_scan_tasks(&scan); +} + +/* * update_flag - read a 0 or a 1 in a file and update associated flag - * bit: the bit to update (CS_CPU_EXCLUSIVE, CS_MEM_EXCLUSIVE, - * CS_SCHED_LOAD_BALANCE, - * CS_NOTIFY_ON_RELEASE, CS_MEMORY_MIGRATE, - * CS_SPREAD_PAGE, CS_SPREAD_SLAB) - * cs: the cpuset to update - * buf: the buffer where we read the 0 or 1 + * bit: the bit to update (see cpuset_flagbits_t) + * cs: the cpuset to update + * turning_on: whether the flag is being set or cleared * * Call with cgroup_mutex held. */ -static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, char *buf) +static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, + int turning_on) { - int turning_on; - struct cpuset trialcs; + struct cpuset *trialcs; + int balance_flag_changed; + int spread_flag_changed; + struct ptr_heap heap; int err; - int cpus_nonempty, balance_flag_changed; - turning_on = (simple_strtoul(buf, NULL, 10) != 0); + trialcs = alloc_trial_cpuset(cs); + if (!trialcs) + return -ENOMEM; - trialcs = *cs; if (turning_on) - set_bit(bit, &trialcs.flags); + set_bit(bit, &trialcs->flags); else - clear_bit(bit, &trialcs.flags); + clear_bit(bit, &trialcs->flags); - err = validate_change(cs, &trialcs); + err = validate_change(cs, trialcs); if (err < 0) - return err; + goto out; + + err = heap_init(&heap, PAGE_SIZE, GFP_KERNEL, NULL); + if (err < 0) + goto out; - cpus_nonempty = !cpus_empty(trialcs.cpus_allowed); balance_flag_changed = (is_sched_load_balance(cs) != - is_sched_load_balance(&trialcs)); + is_sched_load_balance(trialcs)); + + spread_flag_changed = ((is_spread_slab(cs) != is_spread_slab(trialcs)) + || (is_spread_page(cs) != is_spread_page(trialcs))); mutex_lock(&callback_mutex); - cs->flags = trialcs.flags; + cs->flags = trialcs->flags; mutex_unlock(&callback_mutex); - if (cpus_nonempty && balance_flag_changed) - rebuild_sched_domains(); + if (!cpumask_empty(trialcs->cpus_allowed) && balance_flag_changed) + async_rebuild_sched_domains(); - return 0; + if (spread_flag_changed) + update_tasks_flags(cs, &heap); + heap_free(&heap); +out: + free_trial_cpuset(trialcs); + return err; } /* @@ -1194,43 +1373,110 @@ static int fmeter_getrate(struct fmeter *fmp) return val; } +/* Protected by cgroup_lock */ +static cpumask_var_t cpus_attach; + /* Called by cgroups to determine if a cpuset is usable; cgroup_mutex held */ -static int cpuset_can_attach(struct cgroup_subsys *ss, - struct cgroup *cont, struct task_struct *tsk) +static int cpuset_can_attach(struct cgroup_subsys *ss, struct cgroup *cont, + struct task_struct *tsk, bool threadgroup) { + int ret; struct cpuset *cs = cgroup_cs(cont); - if (cpus_empty(cs->cpus_allowed) || nodes_empty(cs->mems_allowed)) + if (cpumask_empty(cs->cpus_allowed) || nodes_empty(cs->mems_allowed)) return -ENOSPC; - return security_task_setscheduler(tsk, 0, NULL); + /* + * Kthreads bound to specific cpus cannot be moved to a new cpuset; we + * cannot change their cpu affinity and isolating such threads by their + * set of allowed nodes is unnecessary. Thus, cpusets are not + * applicable for such threads. This prevents checking for success of + * set_cpus_allowed_ptr() on all attached tasks before cpus_allowed may + * be changed. + */ + if (tsk->flags & PF_THREAD_BOUND) + return -EINVAL; + + ret = security_task_setscheduler(tsk, 0, NULL); + if (ret) + return ret; + if (threadgroup) { + struct task_struct *c; + + rcu_read_lock(); + list_for_each_entry_rcu(c, &tsk->thread_group, thread_group) { + ret = security_task_setscheduler(c, 0, NULL); + if (ret) { + rcu_read_unlock(); + return ret; + } + } + rcu_read_unlock(); + } + return 0; } -static void cpuset_attach(struct cgroup_subsys *ss, - struct cgroup *cont, struct cgroup *oldcont, - struct task_struct *tsk) +static void cpuset_attach_task(struct task_struct *tsk, nodemask_t *to, + struct cpuset *cs) +{ + int err; + /* + * can_attach beforehand should guarantee that this doesn't fail. + * TODO: have a better way to handle failure here + */ + err = set_cpus_allowed_ptr(tsk, cpus_attach); + WARN_ON_ONCE(err); + + cpuset_change_task_nodemask(tsk, to); + cpuset_update_task_spread_flag(cs, tsk); + +} + +static void cpuset_attach(struct cgroup_subsys *ss, struct cgroup *cont, + struct cgroup *oldcont, struct task_struct *tsk, + bool threadgroup) { - cpumask_t cpus; - nodemask_t from, to; struct mm_struct *mm; struct cpuset *cs = cgroup_cs(cont); struct cpuset *oldcs = cgroup_cs(oldcont); + NODEMASK_ALLOC(nodemask_t, from, GFP_KERNEL); + NODEMASK_ALLOC(nodemask_t, to, GFP_KERNEL); + + if (from == NULL || to == NULL) + goto alloc_fail; + + if (cs == &top_cpuset) { + cpumask_copy(cpus_attach, cpu_possible_mask); + } else { + guarantee_online_cpus(cs, cpus_attach); + } + guarantee_online_mems(cs, to); - mutex_lock(&callback_mutex); - guarantee_online_cpus(cs, &cpus); - set_cpus_allowed_ptr(tsk, &cpus); - mutex_unlock(&callback_mutex); + /* do per-task migration stuff possibly for each in the threadgroup */ + cpuset_attach_task(tsk, to, cs); + if (threadgroup) { + struct task_struct *c; + rcu_read_lock(); + list_for_each_entry_rcu(c, &tsk->thread_group, thread_group) { + cpuset_attach_task(c, to, cs); + } + rcu_read_unlock(); + } - from = oldcs->mems_allowed; - to = cs->mems_allowed; + /* change mm; only needs to be done once even if threadgroup */ + *from = oldcs->mems_allowed; + *to = cs->mems_allowed; mm = get_task_mm(tsk); if (mm) { - mpol_rebind_mm(mm, &to); + mpol_rebind_mm(mm, to); if (is_memory_migrate(cs)) - cpuset_migrate_mm(mm, &from, &to); + cpuset_migrate_mm(mm, from, to); mmput(mm); } +alloc_fail: + NODEMASK_FREE(from); + NODEMASK_FREE(to); } /* The various types of files and directories in a cpuset file system */ @@ -1241,6 +1487,7 @@ typedef enum { FILE_MEMLIST, FILE_CPU_EXCLUSIVE, FILE_MEM_EXCLUSIVE, + FILE_MEM_HARDWALL, FILE_SCHED_LOAD_BALANCE, FILE_SCHED_RELAX_DOMAIN_LEVEL, FILE_MEMORY_PRESSURE_ENABLED, @@ -1249,85 +1496,103 @@ typedef enum { FILE_SPREAD_SLAB, } cpuset_filetype_t; -static ssize_t cpuset_common_file_write(struct cgroup *cont, - struct cftype *cft, - struct file *file, - const char __user *userbuf, - size_t nbytes, loff_t *unused_ppos) +static int cpuset_write_u64(struct cgroup *cgrp, struct cftype *cft, u64 val) { - struct cpuset *cs = cgroup_cs(cont); - cpuset_filetype_t type = cft->private; - char *buffer; int retval = 0; + struct cpuset *cs = cgroup_cs(cgrp); + cpuset_filetype_t type = cft->private; - /* Crude upper limit on largest legitimate cpulist user might write. */ - if (nbytes > 100U + 6 * max(NR_CPUS, MAX_NUMNODES)) - return -E2BIG; - - /* +1 for nul-terminator */ - if ((buffer = kmalloc(nbytes + 1, GFP_KERNEL)) == 0) - return -ENOMEM; - - if (copy_from_user(buffer, userbuf, nbytes)) { - retval = -EFAULT; - goto out1; - } - buffer[nbytes] = 0; /* nul-terminate */ - - cgroup_lock(); - - if (cgroup_is_removed(cont)) { - retval = -ENODEV; - goto out2; - } + if (!cgroup_lock_live_group(cgrp)) + return -ENODEV; switch (type) { - case FILE_CPULIST: - retval = update_cpumask(cs, buffer); - break; - case FILE_MEMLIST: - retval = update_nodemask(cs, buffer); - break; case FILE_CPU_EXCLUSIVE: - retval = update_flag(CS_CPU_EXCLUSIVE, cs, buffer); + retval = update_flag(CS_CPU_EXCLUSIVE, cs, val); break; case FILE_MEM_EXCLUSIVE: - retval = update_flag(CS_MEM_EXCLUSIVE, cs, buffer); + retval = update_flag(CS_MEM_EXCLUSIVE, cs, val); break; - case FILE_SCHED_LOAD_BALANCE: - retval = update_flag(CS_SCHED_LOAD_BALANCE, cs, buffer); + case FILE_MEM_HARDWALL: + retval = update_flag(CS_MEM_HARDWALL, cs, val); break; - case FILE_SCHED_RELAX_DOMAIN_LEVEL: - retval = update_relax_domain_level(cs, buffer); + case FILE_SCHED_LOAD_BALANCE: + retval = update_flag(CS_SCHED_LOAD_BALANCE, cs, val); break; case FILE_MEMORY_MIGRATE: - retval = update_flag(CS_MEMORY_MIGRATE, cs, buffer); + retval = update_flag(CS_MEMORY_MIGRATE, cs, val); break; case FILE_MEMORY_PRESSURE_ENABLED: - retval = update_memory_pressure_enabled(cs, buffer); + cpuset_memory_pressure_enabled = !!val; break; case FILE_MEMORY_PRESSURE: retval = -EACCES; break; case FILE_SPREAD_PAGE: - retval = update_flag(CS_SPREAD_PAGE, cs, buffer); - cs->mems_generation = cpuset_mems_generation++; + retval = update_flag(CS_SPREAD_PAGE, cs, val); break; case FILE_SPREAD_SLAB: - retval = update_flag(CS_SPREAD_SLAB, cs, buffer); - cs->mems_generation = cpuset_mems_generation++; + retval = update_flag(CS_SPREAD_SLAB, cs, val); + break; + default: + retval = -EINVAL; + break; + } + cgroup_unlock(); + return retval; +} + +static int cpuset_write_s64(struct cgroup *cgrp, struct cftype *cft, s64 val) +{ + int retval = 0; + struct cpuset *cs = cgroup_cs(cgrp); + cpuset_filetype_t type = cft->private; + + if (!cgroup_lock_live_group(cgrp)) + return -ENODEV; + + switch (type) { + case FILE_SCHED_RELAX_DOMAIN_LEVEL: + retval = update_relax_domain_level(cs, val); + break; + default: + retval = -EINVAL; + break; + } + cgroup_unlock(); + return retval; +} + +/* + * Common handling for a write to a "cpus" or "mems" file. + */ +static int cpuset_write_resmask(struct cgroup *cgrp, struct cftype *cft, + const char *buf) +{ + int retval = 0; + struct cpuset *cs = cgroup_cs(cgrp); + struct cpuset *trialcs; + + if (!cgroup_lock_live_group(cgrp)) + return -ENODEV; + + trialcs = alloc_trial_cpuset(cs); + if (!trialcs) + return -ENOMEM; + + switch (cft->private) { + case FILE_CPULIST: + retval = update_cpumask(cs, trialcs, buf); + break; + case FILE_MEMLIST: + retval = update_nodemask(cs, trialcs, buf); break; default: retval = -EINVAL; - goto out2; + break; } - if (retval == 0) - retval = nbytes; -out2: + free_trial_cpuset(trialcs); cgroup_unlock(); -out1: - kfree(buffer); return retval; } @@ -1345,24 +1610,32 @@ out1: static int cpuset_sprintf_cpulist(char *page, struct cpuset *cs) { - cpumask_t mask; + int ret; mutex_lock(&callback_mutex); - mask = cs->cpus_allowed; + ret = cpulist_scnprintf(page, PAGE_SIZE, cs->cpus_allowed); mutex_unlock(&callback_mutex); - return cpulist_scnprintf(page, PAGE_SIZE, mask); + return ret; } static int cpuset_sprintf_memlist(char *page, struct cpuset *cs) { - nodemask_t mask; + NODEMASK_ALLOC(nodemask_t, mask, GFP_KERNEL); + int retval; + + if (mask == NULL) + return -ENOMEM; mutex_lock(&callback_mutex); - mask = cs->mems_allowed; + *mask = cs->mems_allowed; mutex_unlock(&callback_mutex); - return nodelist_scnprintf(page, PAGE_SIZE, mask); + retval = nodelist_scnprintf(page, PAGE_SIZE, *mask); + + NODEMASK_FREE(mask); + + return retval; } static ssize_t cpuset_common_file_read(struct cgroup *cont, @@ -1389,33 +1662,6 @@ static ssize_t cpuset_common_file_read(struct cgroup *cont, case FILE_MEMLIST: s += cpuset_sprintf_memlist(s, cs); break; - case FILE_CPU_EXCLUSIVE: - *s++ = is_cpu_exclusive(cs) ? '1' : '0'; - break; - case FILE_MEM_EXCLUSIVE: - *s++ = is_mem_exclusive(cs) ? '1' : '0'; - break; - case FILE_SCHED_LOAD_BALANCE: - *s++ = is_sched_load_balance(cs) ? '1' : '0'; - break; - case FILE_SCHED_RELAX_DOMAIN_LEVEL: - s += sprintf(s, "%d", cs->relax_domain_level); - break; - case FILE_MEMORY_MIGRATE: - *s++ = is_memory_migrate(cs) ? '1' : '0'; - break; - case FILE_MEMORY_PRESSURE_ENABLED: - *s++ = cpuset_memory_pressure_enabled ? '1' : '0'; - break; - case FILE_MEMORY_PRESSURE: - s += sprintf(s, "%d", fmeter_getrate(&cs->fmeter)); - break; - case FILE_SPREAD_PAGE: - *s++ = is_spread_page(cs) ? '1' : '0'; - break; - case FILE_SPREAD_SLAB: - *s++ = is_spread_slab(cs) ? '1' : '0'; - break; default: retval = -EINVAL; goto out; @@ -1428,121 +1674,158 @@ out: return retval; } +static u64 cpuset_read_u64(struct cgroup *cont, struct cftype *cft) +{ + struct cpuset *cs = cgroup_cs(cont); + cpuset_filetype_t type = cft->private; + switch (type) { + case FILE_CPU_EXCLUSIVE: + return is_cpu_exclusive(cs); + case FILE_MEM_EXCLUSIVE: + return is_mem_exclusive(cs); + case FILE_MEM_HARDWALL: + return is_mem_hardwall(cs); + case FILE_SCHED_LOAD_BALANCE: + return is_sched_load_balance(cs); + case FILE_MEMORY_MIGRATE: + return is_memory_migrate(cs); + case FILE_MEMORY_PRESSURE_ENABLED: + return cpuset_memory_pressure_enabled; + case FILE_MEMORY_PRESSURE: + return fmeter_getrate(&cs->fmeter); + case FILE_SPREAD_PAGE: + return is_spread_page(cs); + case FILE_SPREAD_SLAB: + return is_spread_slab(cs); + default: + BUG(); + } + + /* Unreachable but makes gcc happy */ + return 0; +} +static s64 cpuset_read_s64(struct cgroup *cont, struct cftype *cft) +{ + struct cpuset *cs = cgroup_cs(cont); + cpuset_filetype_t type = cft->private; + switch (type) { + case FILE_SCHED_RELAX_DOMAIN_LEVEL: + return cs->relax_domain_level; + default: + BUG(); + } + /* Unrechable but makes gcc happy */ + return 0; +} /* * for the common functions, 'private' gives the type of file */ -static struct cftype cft_cpus = { - .name = "cpus", - .read = cpuset_common_file_read, - .write = cpuset_common_file_write, - .private = FILE_CPULIST, -}; - -static struct cftype cft_mems = { - .name = "mems", - .read = cpuset_common_file_read, - .write = cpuset_common_file_write, - .private = FILE_MEMLIST, -}; - -static struct cftype cft_cpu_exclusive = { - .name = "cpu_exclusive", - .read = cpuset_common_file_read, - .write = cpuset_common_file_write, - .private = FILE_CPU_EXCLUSIVE, -}; - -static struct cftype cft_mem_exclusive = { - .name = "mem_exclusive", - .read = cpuset_common_file_read, - .write = cpuset_common_file_write, - .private = FILE_MEM_EXCLUSIVE, -}; - -static struct cftype cft_sched_load_balance = { - .name = "sched_load_balance", - .read = cpuset_common_file_read, - .write = cpuset_common_file_write, - .private = FILE_SCHED_LOAD_BALANCE, -}; - -static struct cftype cft_sched_relax_domain_level = { - .name = "sched_relax_domain_level", - .read = cpuset_common_file_read, - .write = cpuset_common_file_write, - .private = FILE_SCHED_RELAX_DOMAIN_LEVEL, -}; - -static struct cftype cft_memory_migrate = { - .name = "memory_migrate", - .read = cpuset_common_file_read, - .write = cpuset_common_file_write, - .private = FILE_MEMORY_MIGRATE, +static struct cftype files[] = { + { + .name = "cpus", + .read = cpuset_common_file_read, + .write_string = cpuset_write_resmask, + .max_write_len = (100U + 6 * NR_CPUS), + .private = FILE_CPULIST, + }, + + { + .name = "mems", + .read = cpuset_common_file_read, + .write_string = cpuset_write_resmask, + .max_write_len = (100U + 6 * MAX_NUMNODES), + .private = FILE_MEMLIST, + }, + + { + .name = "cpu_exclusive", + .read_u64 = cpuset_read_u64, + .write_u64 = cpuset_write_u64, + .private = FILE_CPU_EXCLUSIVE, + }, + + { + .name = "mem_exclusive", + .read_u64 = cpuset_read_u64, + .write_u64 = cpuset_write_u64, + .private = FILE_MEM_EXCLUSIVE, + }, + + { + .name = "mem_hardwall", + .read_u64 = cpuset_read_u64, + .write_u64 = cpuset_write_u64, + .private = FILE_MEM_HARDWALL, + }, + + { + .name = "sched_load_balance", + .read_u64 = cpuset_read_u64, + .write_u64 = cpuset_write_u64, + .private = FILE_SCHED_LOAD_BALANCE, + }, + + { + .name = "sched_relax_domain_level", + .read_s64 = cpuset_read_s64, + .write_s64 = cpuset_write_s64, + .private = FILE_SCHED_RELAX_DOMAIN_LEVEL, + }, + + { + .name = "memory_migrate", + .read_u64 = cpuset_read_u64, + .write_u64 = cpuset_write_u64, + .private = FILE_MEMORY_MIGRATE, + }, + + { + .name = "memory_pressure", + .read_u64 = cpuset_read_u64, + .write_u64 = cpuset_write_u64, + .private = FILE_MEMORY_PRESSURE, + .mode = S_IRUGO, + }, + + { + .name = "memory_spread_page", + .read_u64 = cpuset_read_u64, + .write_u64 = cpuset_write_u64, + .private = FILE_SPREAD_PAGE, + }, + + { + .name = "memory_spread_slab", + .read_u64 = cpuset_read_u64, + .write_u64 = cpuset_write_u64, + .private = FILE_SPREAD_SLAB, + }, }; static struct cftype cft_memory_pressure_enabled = { .name = "memory_pressure_enabled", - .read = cpuset_common_file_read, - .write = cpuset_common_file_write, + .read_u64 = cpuset_read_u64, + .write_u64 = cpuset_write_u64, .private = FILE_MEMORY_PRESSURE_ENABLED, }; -static struct cftype cft_memory_pressure = { - .name = "memory_pressure", - .read = cpuset_common_file_read, - .write = cpuset_common_file_write, - .private = FILE_MEMORY_PRESSURE, -}; - -static struct cftype cft_spread_page = { - .name = "memory_spread_page", - .read = cpuset_common_file_read, - .write = cpuset_common_file_write, - .private = FILE_SPREAD_PAGE, -}; - -static struct cftype cft_spread_slab = { - .name = "memory_spread_slab", - .read = cpuset_common_file_read, - .write = cpuset_common_file_write, - .private = FILE_SPREAD_SLAB, -}; - static int cpuset_populate(struct cgroup_subsys *ss, struct cgroup *cont) { int err; - if ((err = cgroup_add_file(cont, ss, &cft_cpus)) < 0) - return err; - if ((err = cgroup_add_file(cont, ss, &cft_mems)) < 0) - return err; - if ((err = cgroup_add_file(cont, ss, &cft_cpu_exclusive)) < 0) - return err; - if ((err = cgroup_add_file(cont, ss, &cft_mem_exclusive)) < 0) - return err; - if ((err = cgroup_add_file(cont, ss, &cft_memory_migrate)) < 0) - return err; - if ((err = cgroup_add_file(cont, ss, &cft_sched_load_balance)) < 0) - return err; - if ((err = cgroup_add_file(cont, ss, - &cft_sched_relax_domain_level)) < 0) - return err; - if ((err = cgroup_add_file(cont, ss, &cft_memory_pressure)) < 0) - return err; - if ((err = cgroup_add_file(cont, ss, &cft_spread_page)) < 0) - return err; - if ((err = cgroup_add_file(cont, ss, &cft_spread_slab)) < 0) + err = cgroup_add_files(cont, ss, files, ARRAY_SIZE(files)); + if (err) return err; /* memory_pressure_enabled is in root cpuset only */ - if (err == 0 && !cont->parent) + if (!cont->parent) err = cgroup_add_file(cont, ss, - &cft_memory_pressure_enabled); - return 0; + &cft_memory_pressure_enabled); + return err; } /* @@ -1578,7 +1861,7 @@ static void cpuset_post_clone(struct cgroup_subsys *ss, parent_cs = cgroup_cs(parent); cs->mems_allowed = parent_cs->mems_allowed; - cs->cpus_allowed = parent_cs->cpus_allowed; + cpumask_copy(cs->cpus_allowed, parent_cs->cpus_allowed); return; } @@ -1596,25 +1879,25 @@ static struct cgroup_subsys_state *cpuset_create( struct cpuset *parent; if (!cont->parent) { - /* This is early initialization for the top cgroup */ - top_cpuset.mems_generation = cpuset_mems_generation++; return &top_cpuset.css; } parent = cgroup_cs(cont->parent); cs = kmalloc(sizeof(*cs), GFP_KERNEL); if (!cs) return ERR_PTR(-ENOMEM); + if (!alloc_cpumask_var(&cs->cpus_allowed, GFP_KERNEL)) { + kfree(cs); + return ERR_PTR(-ENOMEM); + } - cpuset_update_task_memory_state(); cs->flags = 0; if (is_spread_page(parent)) set_bit(CS_SPREAD_PAGE, &cs->flags); if (is_spread_slab(parent)) set_bit(CS_SPREAD_SLAB, &cs->flags); set_bit(CS_SCHED_LOAD_BALANCE, &cs->flags); - cpus_clear(cs->cpus_allowed); + cpumask_clear(cs->cpus_allowed); nodes_clear(cs->mems_allowed); - cs->mems_generation = cpuset_mems_generation++; fmeter_init(&cs->fmeter); cs->relax_domain_level = -1; @@ -1624,34 +1907,27 @@ static struct cgroup_subsys_state *cpuset_create( } /* - * Locking note on the strange update_flag() call below: - * * If the cpuset being removed has its flag 'sched_load_balance' * enabled, then simulate turning sched_load_balance off, which - * will call rebuild_sched_domains(). The get_online_cpus() - * call in rebuild_sched_domains() must not be made while holding - * callback_mutex. Elsewhere the kernel nests callback_mutex inside - * get_online_cpus() calls. So the reverse nesting would risk an - * ABBA deadlock. + * will call async_rebuild_sched_domains(). */ static void cpuset_destroy(struct cgroup_subsys *ss, struct cgroup *cont) { struct cpuset *cs = cgroup_cs(cont); - cpuset_update_task_memory_state(); - if (is_sched_load_balance(cs)) - update_flag(CS_SCHED_LOAD_BALANCE, cs, "0"); + update_flag(CS_SCHED_LOAD_BALANCE, cs, 0); number_of_cpusets--; + free_cpumask_var(cs->cpus_allowed); kfree(cs); } struct cgroup_subsys cpuset_subsys = { .name = "cpuset", .create = cpuset_create, - .destroy = cpuset_destroy, + .destroy = cpuset_destroy, .can_attach = cpuset_can_attach, .attach = cpuset_attach, .populate = cpuset_populate, @@ -1660,19 +1936,6 @@ struct cgroup_subsys cpuset_subsys = { .early_init = 1, }; -/* - * cpuset_init_early - just enough so that the calls to - * cpuset_update_task_memory_state() in early init code - * are harmless. - */ - -int __init cpuset_init_early(void) -{ - top_cpuset.mems_generation = cpuset_mems_generation++; - return 0; -} - - /** * cpuset_init - initialize cpusets at system boot * @@ -1683,11 +1946,13 @@ int __init cpuset_init(void) { int err = 0; - cpus_setall(top_cpuset.cpus_allowed); + if (!alloc_cpumask_var(&top_cpuset.cpus_allowed, GFP_KERNEL)) + BUG(); + + cpumask_setall(top_cpuset.cpus_allowed); nodes_setall(top_cpuset.mems_allowed); fmeter_init(&top_cpuset.fmeter); - top_cpuset.mems_generation = cpuset_mems_generation++; set_bit(CS_SCHED_LOAD_BALANCE, &top_cpuset.flags); top_cpuset.relax_domain_level = -1; @@ -1695,6 +1960,9 @@ int __init cpuset_init(void) if (err < 0) return err; + if (!alloc_cpumask_var(&cpus_attach, GFP_KERNEL)) + BUG(); + number_of_cpusets = 1; return 0; } @@ -1707,12 +1975,12 @@ int __init cpuset_init(void) * Called by cgroup_scan_tasks() for each task in a cgroup. * Return nonzero to stop the walk through the tasks. */ -void cpuset_do_move_task(struct task_struct *tsk, struct cgroup_scanner *scan) +static void cpuset_do_move_task(struct task_struct *tsk, + struct cgroup_scanner *scan) { - struct cpuset_hotplug_scanner *chsp; + struct cgroup *new_cgroup = scan->data; - chsp = container_of(scan, struct cpuset_hotplug_scanner, scan); - cgroup_attach_task(chsp->to, tsk); + cgroup_attach_task(new_cgroup, tsk); } /** @@ -1728,21 +1996,21 @@ void cpuset_do_move_task(struct task_struct *tsk, struct cgroup_scanner *scan) */ static void move_member_tasks_to_cpuset(struct cpuset *from, struct cpuset *to) { - struct cpuset_hotplug_scanner scan; + struct cgroup_scanner scan; - scan.scan.cg = from->css.cgroup; - scan.scan.test_task = NULL; /* select all tasks in cgroup */ - scan.scan.process_task = cpuset_do_move_task; - scan.scan.heap = NULL; - scan.to = to->css.cgroup; + scan.cg = from->css.cgroup; + scan.test_task = NULL; /* select all tasks in cgroup */ + scan.process_task = cpuset_do_move_task; + scan.heap = NULL; + scan.data = to->css.cgroup; - if (cgroup_scan_tasks((struct cgroup_scanner *)&scan)) + if (cgroup_scan_tasks(&scan)) printk(KERN_ERR "move_member_tasks_to_cpuset: " "cgroup_scan_tasks failed\n"); } /* - * If common_cpu_mem_hotplug_unplug(), below, unplugs any CPUs + * If CPU and/or memory hotplug handlers, below, unplug any CPUs * or memory nodes, we need to walk over the cpuset hierarchy, * removing that CPU or node from all cpusets. If this removes the * last CPU or node from a cpuset, then move the tasks in the empty @@ -1768,7 +2036,7 @@ static void remove_tasks_in_empty_cpuset(struct cpuset *cs) * has online cpus, so can't be empty). */ parent = cs->parent; - while (cpus_empty(parent->cpus_allowed) || + while (cpumask_empty(parent->cpus_allowed) || nodes_empty(parent->mems_allowed)) parent = parent->parent; @@ -1790,65 +2058,52 @@ static void remove_tasks_in_empty_cpuset(struct cpuset *cs) * that has tasks along with an empty 'mems'. But if we did see such * a cpuset, we'd handle it just like we do if its 'cpus' was empty. */ -static void scan_for_empty_cpusets(const struct cpuset *root) +static void scan_for_empty_cpusets(struct cpuset *root) { + LIST_HEAD(queue); struct cpuset *cp; /* scans cpusets being updated */ struct cpuset *child; /* scans child cpusets of cp */ - struct list_head queue; struct cgroup *cont; + NODEMASK_ALLOC(nodemask_t, oldmems, GFP_KERNEL); - INIT_LIST_HEAD(&queue); + if (oldmems == NULL) + return; list_add_tail((struct list_head *)&root->stack_list, &queue); while (!list_empty(&queue)) { - cp = container_of(queue.next, struct cpuset, stack_list); + cp = list_first_entry(&queue, struct cpuset, stack_list); list_del(queue.next); list_for_each_entry(cont, &cp->css.cgroup->children, sibling) { child = cgroup_cs(cont); list_add_tail(&child->stack_list, &queue); } - cont = cp->css.cgroup; /* Continue past cpusets with all cpus, mems online */ - if (cpus_subset(cp->cpus_allowed, cpu_online_map) && + if (cpumask_subset(cp->cpus_allowed, cpu_active_mask) && nodes_subset(cp->mems_allowed, node_states[N_HIGH_MEMORY])) continue; + *oldmems = cp->mems_allowed; + /* Remove offline cpus and mems from this cpuset. */ mutex_lock(&callback_mutex); - cpus_and(cp->cpus_allowed, cp->cpus_allowed, cpu_online_map); + cpumask_and(cp->cpus_allowed, cp->cpus_allowed, + cpu_active_mask); nodes_and(cp->mems_allowed, cp->mems_allowed, node_states[N_HIGH_MEMORY]); mutex_unlock(&callback_mutex); /* Move tasks from the empty cpuset to a parent */ - if (cpus_empty(cp->cpus_allowed) || + if (cpumask_empty(cp->cpus_allowed) || nodes_empty(cp->mems_allowed)) remove_tasks_in_empty_cpuset(cp); + else { + update_tasks_cpumask(cp, NULL); + update_tasks_nodemask(cp, oldmems, NULL); + } } -} - -/* - * The cpus_allowed and mems_allowed nodemasks in the top_cpuset track - * cpu_online_map and node_states[N_HIGH_MEMORY]. Force the top cpuset to - * track what's online after any CPU or memory node hotplug or unplug event. - * - * Since there are two callers of this routine, one for CPU hotplug - * events and one for memory node hotplug events, we could have coded - * two separate routines here. We code it as a single common routine - * in order to minimize text size. - */ - -static void common_cpu_mem_hotplug_unplug(void) -{ - cgroup_lock(); - - top_cpuset.cpus_allowed = cpu_online_map; - top_cpuset.mems_allowed = node_states[N_HIGH_MEMORY]; - scan_for_empty_cpusets(&top_cpuset); - - cgroup_unlock(); + NODEMASK_FREE(oldmems); } /* @@ -1859,29 +2114,81 @@ static void common_cpu_mem_hotplug_unplug(void) * * This routine ensures that top_cpuset.cpus_allowed tracks * cpu_online_map on each CPU hotplug (cpuhp) event. + * + * Called within get_online_cpus(). Needs to call cgroup_lock() + * before calling generate_sched_domains(). */ - -static int cpuset_handle_cpuhp(struct notifier_block *unused_nb, +static int cpuset_track_online_cpus(struct notifier_block *unused_nb, unsigned long phase, void *unused_cpu) { - if (phase == CPU_DYING || phase == CPU_DYING_FROZEN) + struct sched_domain_attr *attr; + cpumask_var_t *doms; + int ndoms; + + switch (phase) { + case CPU_ONLINE: + case CPU_ONLINE_FROZEN: + case CPU_DOWN_PREPARE: + case CPU_DOWN_PREPARE_FROZEN: + case CPU_DOWN_FAILED: + case CPU_DOWN_FAILED_FROZEN: + break; + + default: return NOTIFY_DONE; + } - common_cpu_mem_hotplug_unplug(); - return 0; + cgroup_lock(); + mutex_lock(&callback_mutex); + cpumask_copy(top_cpuset.cpus_allowed, cpu_active_mask); + mutex_unlock(&callback_mutex); + scan_for_empty_cpusets(&top_cpuset); + ndoms = generate_sched_domains(&doms, &attr); + cgroup_unlock(); + + /* Have scheduler rebuild the domains */ + partition_sched_domains(ndoms, doms, attr); + + return NOTIFY_OK; } #ifdef CONFIG_MEMORY_HOTPLUG /* * Keep top_cpuset.mems_allowed tracking node_states[N_HIGH_MEMORY]. - * Call this routine anytime after you change - * node_states[N_HIGH_MEMORY]. - * See also the previous routine cpuset_handle_cpuhp(). + * Call this routine anytime after node_states[N_HIGH_MEMORY] changes. + * See also the previous routine cpuset_track_online_cpus(). */ - -void cpuset_track_online_nodes(void) +static int cpuset_track_online_nodes(struct notifier_block *self, + unsigned long action, void *arg) { - common_cpu_mem_hotplug_unplug(); + NODEMASK_ALLOC(nodemask_t, oldmems, GFP_KERNEL); + + if (oldmems == NULL) + return NOTIFY_DONE; + + cgroup_lock(); + switch (action) { + case MEM_ONLINE: + *oldmems = top_cpuset.mems_allowed; + mutex_lock(&callback_mutex); + top_cpuset.mems_allowed = node_states[N_HIGH_MEMORY]; + mutex_unlock(&callback_mutex); + update_tasks_nodemask(&top_cpuset, oldmems, NULL); + break; + case MEM_OFFLINE: + /* + * needn't update top_cpuset.mems_allowed explicitly because + * scan_for_empty_cpusets() will update it. + */ + scan_for_empty_cpusets(&top_cpuset); + break; + default: + break; + } + cgroup_unlock(); + + NODEMASK_FREE(oldmems); + return NOTIFY_OK; } #endif @@ -1893,40 +2200,76 @@ void cpuset_track_online_nodes(void) void __init cpuset_init_smp(void) { - top_cpuset.cpus_allowed = cpu_online_map; + cpumask_copy(top_cpuset.cpus_allowed, cpu_active_mask); top_cpuset.mems_allowed = node_states[N_HIGH_MEMORY]; - hotcpu_notifier(cpuset_handle_cpuhp, 0); + hotcpu_notifier(cpuset_track_online_cpus, 0); + hotplug_memory_notifier(cpuset_track_online_nodes, 10); + + cpuset_wq = create_singlethread_workqueue("cpuset"); + BUG_ON(!cpuset_wq); } /** - * cpuset_cpus_allowed - return cpus_allowed mask from a tasks cpuset. * @tsk: pointer to task_struct from which to obtain cpuset->cpus_allowed. - * @pmask: pointer to cpumask_t variable to receive cpus_allowed set. + * @pmask: pointer to struct cpumask variable to receive cpus_allowed set. * - * Description: Returns the cpumask_t cpus_allowed of the cpuset + * Description: Returns the cpumask_var_t cpus_allowed of the cpuset * attached to the specified @tsk. Guaranteed to return some non-empty * subset of cpu_online_map, even if this means going outside the * tasks cpuset. **/ -void cpuset_cpus_allowed(struct task_struct *tsk, cpumask_t *pmask) +void cpuset_cpus_allowed(struct task_struct *tsk, struct cpumask *pmask) { mutex_lock(&callback_mutex); - cpuset_cpus_allowed_locked(tsk, pmask); + task_lock(tsk); + guarantee_online_cpus(task_cs(tsk), pmask); + task_unlock(tsk); mutex_unlock(&callback_mutex); } -/** - * cpuset_cpus_allowed_locked - return cpus_allowed mask from a tasks cpuset. - * Must be called with callback_mutex held. - **/ -void cpuset_cpus_allowed_locked(struct task_struct *tsk, cpumask_t *pmask) +int cpuset_cpus_allowed_fallback(struct task_struct *tsk) { - task_lock(tsk); - guarantee_online_cpus(task_cs(tsk), pmask); - task_unlock(tsk); + const struct cpuset *cs; + int cpu; + + rcu_read_lock(); + cs = task_cs(tsk); + if (cs) + cpumask_copy(&tsk->cpus_allowed, cs->cpus_allowed); + rcu_read_unlock(); + + /* + * We own tsk->cpus_allowed, nobody can change it under us. + * + * But we used cs && cs->cpus_allowed lockless and thus can + * race with cgroup_attach_task() or update_cpumask() and get + * the wrong tsk->cpus_allowed. However, both cases imply the + * subsequent cpuset_change_cpumask()->set_cpus_allowed_ptr() + * which takes task_rq_lock(). + * + * If we are called after it dropped the lock we must see all + * changes in tsk_cs()->cpus_allowed. Otherwise we can temporary + * set any mask even if it is not right from task_cs() pov, + * the pending set_cpus_allowed_ptr() will fix things. + */ + + cpu = cpumask_any_and(&tsk->cpus_allowed, cpu_active_mask); + if (cpu >= nr_cpu_ids) { + /* + * Either tsk->cpus_allowed is wrong (see above) or it + * is actually empty. The latter case is only possible + * if we are racing with remove_tasks_in_empty_cpuset(). + * Like above we can temporary set any mask and rely on + * set_cpus_allowed_ptr() as synchronization point. + */ + cpumask_copy(&tsk->cpus_allowed, cpu_possible_mask); + cpu = cpumask_any(cpu_active_mask); + } + + return cpu; } void cpuset_init_current_mems_allowed(void) @@ -1958,58 +2301,48 @@ nodemask_t cpuset_mems_allowed(struct task_struct *tsk) } /** - * cpuset_zonelist_valid_mems_allowed - check zonelist vs. curremt mems_allowed - * @zl: the zonelist to be checked + * cpuset_nodemask_valid_mems_allowed - check nodemask vs. curremt mems_allowed + * @nodemask: the nodemask to be checked * - * Are any of the nodes on zonelist zl allowed in current->mems_allowed? + * Are any of the nodes in the nodemask allowed in current->mems_allowed? */ -int cpuset_zonelist_valid_mems_allowed(struct zonelist *zl) +int cpuset_nodemask_valid_mems_allowed(nodemask_t *nodemask) { - int i; - - for (i = 0; zl->zones[i]; i++) { - int nid = zone_to_nid(zl->zones[i]); - - if (node_isset(nid, current->mems_allowed)) - return 1; - } - return 0; + return nodes_intersects(*nodemask, current->mems_allowed); } /* - * nearest_exclusive_ancestor() - Returns the nearest mem_exclusive - * ancestor to the specified cpuset. Call holding callback_mutex. - * If no ancestor is mem_exclusive (an unusual configuration), then - * returns the root cpuset. + * nearest_hardwall_ancestor() - Returns the nearest mem_exclusive or + * mem_hardwall ancestor to the specified cpuset. Call holding + * callback_mutex. If no ancestor is mem_exclusive or mem_hardwall + * (an unusual configuration), then returns the root cpuset. */ -static const struct cpuset *nearest_exclusive_ancestor(const struct cpuset *cs) +static const struct cpuset *nearest_hardwall_ancestor(const struct cpuset *cs) { - while (!is_mem_exclusive(cs) && cs->parent) + while (!(is_mem_exclusive(cs) || is_mem_hardwall(cs)) && cs->parent) cs = cs->parent; return cs; } /** - * cpuset_zone_allowed_softwall - Can we allocate on zone z's memory node? - * @z: is this zone on an allowed node? + * cpuset_node_allowed_softwall - Can we allocate on a memory node? + * @node: is this an allowed node? * @gfp_mask: memory allocation flags * - * If we're in interrupt, yes, we can always allocate. If - * __GFP_THISNODE is set, yes, we can always allocate. If zone - * z's node is in our tasks mems_allowed, yes. If it's not a - * __GFP_HARDWALL request and this zone's nodes is in the nearest - * mem_exclusive cpuset ancestor to this tasks cpuset, yes. - * If the task has been OOM killed and has access to memory reserves - * as specified by the TIF_MEMDIE flag, yes. + * If we're in interrupt, yes, we can always allocate. If __GFP_THISNODE is + * set, yes, we can always allocate. If node is in our task's mems_allowed, + * yes. If it's not a __GFP_HARDWALL request and this node is in the nearest + * hardwalled cpuset ancestor to this task's cpuset, yes. If the task has been + * OOM killed and has access to memory reserves as specified by the TIF_MEMDIE + * flag, yes. * Otherwise, no. * - * If __GFP_HARDWALL is set, cpuset_zone_allowed_softwall() - * reduces to cpuset_zone_allowed_hardwall(). Otherwise, - * cpuset_zone_allowed_softwall() might sleep, and might allow a zone - * from an enclosing cpuset. + * If __GFP_HARDWALL is set, cpuset_node_allowed_softwall() reduces to + * cpuset_node_allowed_hardwall(). Otherwise, cpuset_node_allowed_softwall() + * might sleep, and might allow a node from an enclosing cpuset. * - * cpuset_zone_allowed_hardwall() only handles the simpler case of - * hardwall cpusets, and never sleeps. + * cpuset_node_allowed_hardwall() only handles the simpler case of hardwall + * cpusets, and never sleeps. * * The __GFP_THISNODE placement logic is really handled elsewhere, * by forcibly using a zonelist starting at a specified node, and by @@ -2021,7 +2354,7 @@ static const struct cpuset *nearest_exclusive_ancestor(const struct cpuset *cs) * and do not allow allocations outside the current tasks cpuset * unless the task has been OOM killed as is marked TIF_MEMDIE. * GFP_KERNEL allocations are not so marked, so can escape to the - * nearest enclosing mem_exclusive ancestor cpuset. + * nearest enclosing hardwalled ancestor cpuset. * * Scanning up parent cpusets requires callback_mutex. The * __alloc_pages() routine only calls here with __GFP_HARDWALL bit @@ -2044,24 +2377,21 @@ static const struct cpuset *nearest_exclusive_ancestor(const struct cpuset *cs) * in_interrupt - any node ok (current task context irrelevant) * GFP_ATOMIC - any node ok * TIF_MEMDIE - any node ok - * GFP_KERNEL - any node in enclosing mem_exclusive cpuset ok + * GFP_KERNEL - any node in enclosing hardwalled cpuset ok * GFP_USER - only nodes in current tasks mems allowed ok. * * Rule: - * Don't call cpuset_zone_allowed_softwall if you can't sleep, unless you + * Don't call cpuset_node_allowed_softwall if you can't sleep, unless you * pass in the __GFP_HARDWALL flag set in gfp_flag, which disables * the code that might scan up ancestor cpusets and sleep. */ - -int __cpuset_zone_allowed_softwall(struct zone *z, gfp_t gfp_mask) +int __cpuset_node_allowed_softwall(int node, gfp_t gfp_mask) { - int node; /* node that zone z is on */ const struct cpuset *cs; /* current cpuset ancestors */ int allowed; /* is allocation in zone z allowed? */ if (in_interrupt() || (gfp_mask & __GFP_THISNODE)) return 1; - node = zone_to_nid(z); might_sleep_if(!(gfp_mask & __GFP_HARDWALL)); if (node_isset(node, current->mems_allowed)) return 1; @@ -2081,7 +2411,7 @@ int __cpuset_zone_allowed_softwall(struct zone *z, gfp_t gfp_mask) mutex_lock(&callback_mutex); task_lock(current); - cs = nearest_exclusive_ancestor(task_cs(current)); + cs = nearest_hardwall_ancestor(task_cs(current)); task_unlock(current); allowed = node_isset(node, cs->mems_allowed); @@ -2090,15 +2420,15 @@ int __cpuset_zone_allowed_softwall(struct zone *z, gfp_t gfp_mask) } /* - * cpuset_zone_allowed_hardwall - Can we allocate on zone z's memory node? - * @z: is this zone on an allowed node? + * cpuset_node_allowed_hardwall - Can we allocate on a memory node? + * @node: is this an allowed node? * @gfp_mask: memory allocation flags * - * If we're in interrupt, yes, we can always allocate. - * If __GFP_THISNODE is set, yes, we can always allocate. If zone - * z's node is in our tasks mems_allowed, yes. If the task has been - * OOM killed and has access to memory reserves as specified by the - * TIF_MEMDIE flag, yes. Otherwise, no. + * If we're in interrupt, yes, we can always allocate. If __GFP_THISNODE is + * set, yes, we can always allocate. If node is in our task's mems_allowed, + * yes. If the task has been OOM killed and has access to memory reserves as + * specified by the TIF_MEMDIE flag, yes. + * Otherwise, no. * * The __GFP_THISNODE placement logic is really handled elsewhere, * by forcibly using a zonelist starting at a specified node, and by @@ -2106,20 +2436,16 @@ int __cpuset_zone_allowed_softwall(struct zone *z, gfp_t gfp_mask) * any node on the zonelist except the first. By the time any such * calls get to this routine, we should just shut up and say 'yes'. * - * Unlike the cpuset_zone_allowed_softwall() variant, above, - * this variant requires that the zone be in the current tasks + * Unlike the cpuset_node_allowed_softwall() variant, above, + * this variant requires that the node be in the current task's * mems_allowed or that we're in interrupt. It does not scan up the * cpuset hierarchy for the nearest enclosing mem_exclusive cpuset. * It never sleeps. */ - -int __cpuset_zone_allowed_hardwall(struct zone *z, gfp_t gfp_mask) +int __cpuset_node_allowed_hardwall(int node, gfp_t gfp_mask) { - int node; /* node that zone z is on */ - if (in_interrupt() || (gfp_mask & __GFP_THISNODE)) return 1; - node = zone_to_nid(z); if (node_isset(node, current->mems_allowed)) return 1; /* @@ -2132,22 +2458,6 @@ int __cpuset_zone_allowed_hardwall(struct zone *z, gfp_t gfp_mask) } /** - * cpuset_lock - lock out any changes to cpuset structures - * - * The out of memory (oom) code needs to mutex_lock cpusets - * from being changed while it scans the tasklist looking for a - * task in an overlapping cpuset. Expose callback_mutex via this - * cpuset_lock() routine, so the oom code can lock it, before - * locking the task list. The tasklist_lock is a spinlock, so - * must be taken inside callback_mutex. - */ - -void cpuset_lock(void) -{ - mutex_lock(&callback_mutex); -} - -/** * cpuset_unlock - release lock on cpuset changes * * Undo the lock taken in a previous cpuset_lock() call. @@ -2159,7 +2469,8 @@ void cpuset_unlock(void) } /** - * cpuset_mem_spread_node() - On which node to begin search for a page + * cpuset_mem_spread_node() - On which node to begin search for a file page + * cpuset_slab_spread_node() - On which node to begin search for a slab page * * If a task is marked PF_SPREAD_PAGE or PF_SPREAD_SLAB (as for * tasks in a cpuset with is_spread_page or is_spread_slab set), @@ -2184,16 +2495,27 @@ void cpuset_unlock(void) * See kmem_cache_alloc_node(). */ -int cpuset_mem_spread_node(void) +static int cpuset_spread_node(int *rotor) { int node; - node = next_node(current->cpuset_mem_spread_rotor, current->mems_allowed); + node = next_node(*rotor, current->mems_allowed); if (node == MAX_NUMNODES) node = first_node(current->mems_allowed); - current->cpuset_mem_spread_rotor = node; + *rotor = node; return node; } + +int cpuset_mem_spread_node(void) +{ + return cpuset_spread_node(¤t->cpuset_mem_spread_rotor); +} + +int cpuset_slab_spread_node(void) +{ + return cpuset_spread_node(¤t->cpuset_slab_spread_rotor); +} + EXPORT_SYMBOL_GPL(cpuset_mem_spread_node); /** @@ -2213,6 +2535,29 @@ int cpuset_mems_allowed_intersects(const struct task_struct *tsk1, return nodes_intersects(tsk1->mems_allowed, tsk2->mems_allowed); } +/** + * cpuset_print_task_mems_allowed - prints task's cpuset and mems_allowed + * @task: pointer to task_struct of some task. + * + * Description: Prints @task's name, cpuset name, and cached copy of its + * mems_allowed to the kernel log. Must hold task_lock(task) to allow + * dereferencing task_cs(task). + */ +void cpuset_print_task_mems_allowed(struct task_struct *tsk) +{ + struct dentry *dentry; + + dentry = task_cs(tsk)->css.cgroup->dentry; + spin_lock(&cpuset_buffer_lock); + snprintf(cpuset_name, CPUSET_NAME_LEN, + dentry ? (const char *)dentry->d_name.name : "/"); + nodelist_scnprintf(cpuset_nodelist, CPUSET_NODELIST_LEN, + tsk->mems_allowed); + printk(KERN_INFO "%s cpuset=%s mems_allowed=%s\n", + tsk->comm, cpuset_name, cpuset_nodelist); + spin_unlock(&cpuset_buffer_lock); +} + /* * Collection of memory_pressure is suppressed unless * this flag is enabled by writing "1" to the special @@ -2306,23 +2651,13 @@ const struct file_operations proc_cpuset_operations = { }; #endif /* CONFIG_PROC_PID_CPUSET */ -/* Display task cpus_allowed, mems_allowed in /proc//status file. */ +/* Display task mems_allowed in /proc//status file. */ void cpuset_task_status_allowed(struct seq_file *m, struct task_struct *task) { - seq_printf(m, "Cpus_allowed:\t"); - m->count += cpumask_scnprintf(m->buf + m->count, m->size - m->count, - task->cpus_allowed); - seq_printf(m, "\n"); - seq_printf(m, "Cpus_allowed_list:\t"); - m->count += cpulist_scnprintf(m->buf + m->count, m->size - m->count, - task->cpus_allowed); - seq_printf(m, "\n"); seq_printf(m, "Mems_allowed:\t"); - m->count += nodemask_scnprintf(m->buf + m->count, m->size - m->count, - task->mems_allowed); + seq_nodemask(m, &task->mems_allowed); seq_printf(m, "\n"); seq_printf(m, "Mems_allowed_list:\t"); - m->count += nodelist_scnprintf(m->buf + m->count, m->size - m->count, - task->mems_allowed); + seq_nodemask_list(m, &task->mems_allowed); seq_printf(m, "\n"); }