#include <asm/uaccess.h>
#include <asm/atomic.h>
#include <linux/mutex.h>
-#include <linux/kfifo.h>
#include <linux/workqueue.h>
#include <linux/cgroup.h>
my_cpusets_mem_gen = top_cpuset.mems_generation;
} else {
rcu_read_lock();
- my_cpusets_mem_gen = task_cs(current)->mems_generation;
+ my_cpusets_mem_gen = task_cs(tsk)->mems_generation;
rcu_read_unlock();
}
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 (cpus_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()
*
- * 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 will be called to rebuild the scheduler's dynamic
+ * 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 the 'sched_relax_domain_level' of any cpuset which has
+ * that flag enabled and with non-empty 'cpus' changes,
+ * - or if any cpuset with non-empty 'cpus' is removed,
+ * - or if a cpu gets offlined.
*
* This routine builds a partial partition of the systems CPUs
* (the set of non-overlappping cpumask_t's in the array 'part'
* So the reverse nesting would risk an ABBA deadlock.
*
* 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
void rebuild_sched_domains(void)
{
- 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 ndoms; /* number of sched domains in result */
int nslot; /* next empty doms[] cpumask_t slot */
- q = NULL;
csa = NULL;
doms = NULL;
dattr = NULL;
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;
}
- 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 (cpus_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);
}
}
cpus_or(*dp, *dp, b->cpus_allowed);
b->pn = -1;
if (dattr)
- update_domain_attr(dattr
+ update_domain_attr_tree(dattr
+ nslot, b);
}
}
put_online_cpus();
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 inline int started_after_time(struct task_struct *t1,
- struct timespec *time,
- struct task_struct *t2)
-{
- 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;
- }
-}
-
-static inline int started_after(void *p1, void *p2)
-{
- struct task_struct *t1 = p1;
- struct task_struct *t2 = p2;
- return started_after_time(t1, &t2->start_time, t2);
-}
-
/**
* cpuset_test_cpumask - test a task's cpus_allowed versus its cpuset's
* @tsk: task to test
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
+ *
+ * Called with cgroup_mutex held
+ *
+ * The cgroup_scan_tasks() function will scan all the tasks in a cgroup,
+ * calling callback functions for each.
+ *
+ * Return 0 if successful, -errno if not.
+ */
+static int update_tasks_cpumask(struct cpuset *cs)
+{
+ struct cgroup_scanner scan;
+ struct ptr_heap heap;
+ int retval;
+
+ /*
+ * cgroup_scan_tasks() will initialize heap->gt for us.
+ * heap_init() is still needed here for we should not change
+ * cs->cpus_allowed when heap_init() fails.
+ */
+ retval = heap_init(&heap, PAGE_SIZE, GFP_KERNEL, NULL);
+ if (retval)
+ return retval;
+
+ scan.cg = cs->css.cgroup;
+ scan.test_task = cpuset_test_cpumask;
+ scan.process_task = cpuset_change_cpumask;
+ scan.heap = &heap;
+ retval = cgroup_scan_tasks(&scan);
+
+ heap_free(&heap);
+ return retval;
+}
+
/**
* update_cpumask - update the cpus_allowed mask of a cpuset and all tasks in it
* @cs: the cpuset to consider
static int update_cpumask(struct cpuset *cs, const char *buf)
{
struct cpuset trialcs;
- struct cgroup_scanner scan;
- struct ptr_heap heap;
int retval;
int is_load_balanced;
if (cpus_equal(cs->cpus_allowed, trialcs.cpus_allowed))
return 0;
- retval = heap_init(&heap, PAGE_SIZE, GFP_KERNEL, &started_after);
- if (retval)
- return retval;
-
is_load_balanced = is_sched_load_balance(&trialcs);
mutex_lock(&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);
- heap_free(&heap);
+ retval = update_tasks_cpumask(cs);
+ if (retval < 0)
+ return retval;
if (is_load_balanced)
rebuild_sched_domains();
mutex_unlock(&callback_mutex);
}
-/*
- * 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.
- *
- * 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, const char *buf)
+/**
+ * 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
+ *
+ * Called with cgroup_mutex held
+ * Return 0 if successful, -errno if not.
+ */
+static int update_tasks_nodemask(struct cpuset *cs, const nodemask_t *oldmem)
{
- struct cpuset trialcs;
- nodemask_t oldmem;
struct task_struct *p;
struct mm_struct **mmarray;
int i, n, ntasks;
int migrate;
int fudge;
- int retval;
struct cgroup_iter it;
-
- /*
- * top_cpuset.mems_allowed tracks node_stats[N_HIGH_MEMORY];
- * it's read-only
- */
- if (cs == &top_cpuset)
- return -EACCES;
-
- trialcs = *cs;
-
- /*
- * An empty mems_allowed is ok iff there are no tasks in the cpuset.
- * Since nodelist_parse() fails on an empty mask, we special case
- * that parsing. The validate_change() call ensures that cpusets
- * with tasks have memory.
- */
- if (!*buf) {
- nodes_clear(trialcs.mems_allowed);
- } else {
- retval = nodelist_parse(buf, trialcs.mems_allowed);
- if (retval < 0)
- goto done;
-
- if (!nodes_subset(trialcs.mems_allowed,
- node_states[N_HIGH_MEMORY]))
- return -EINVAL;
- }
- 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);
- if (retval < 0)
- goto done;
-
- mutex_lock(&callback_mutex);
- cs->mems_allowed = trialcs.mems_allowed;
- cs->mems_generation = cpuset_mems_generation++;
- mutex_unlock(&callback_mutex);
+ int retval;
cpuset_being_rebound = cs; /* causes mpol_dup() rebind */
mpol_rebind_mm(mm, &cs->mems_allowed);
if (migrate)
- cpuset_migrate_mm(mm, &oldmem, &cs->mems_allowed);
+ cpuset_migrate_mm(mm, oldmem, &cs->mems_allowed);
mmput(mm);
}
return retval;
}
+/*
+ * 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.
+ *
+ * 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 int update_nodemask(struct cpuset *cs, const char *buf)
+{
+ struct cpuset trialcs;
+ nodemask_t oldmem;
+ int retval;
+
+ /*
+ * top_cpuset.mems_allowed tracks node_stats[N_HIGH_MEMORY];
+ * it's read-only
+ */
+ if (cs == &top_cpuset)
+ return -EACCES;
+
+ trialcs = *cs;
+
+ /*
+ * An empty mems_allowed is ok iff there are no tasks in the cpuset.
+ * Since nodelist_parse() fails on an empty mask, we special case
+ * that parsing. The validate_change() call ensures that cpusets
+ * with tasks have memory.
+ */
+ if (!*buf) {
+ nodes_clear(trialcs.mems_allowed);
+ } else {
+ retval = nodelist_parse(buf, trialcs.mems_allowed);
+ if (retval < 0)
+ goto done;
+
+ if (!nodes_subset(trialcs.mems_allowed,
+ node_states[N_HIGH_MEMORY]))
+ return -EINVAL;
+ }
+ 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);
+ if (retval < 0)
+ goto done;
+
+ mutex_lock(&callback_mutex);
+ cs->mems_allowed = trialcs.mems_allowed;
+ cs->mems_generation = cpuset_mems_generation++;
+ mutex_unlock(&callback_mutex);
+
+ retval = update_tasks_nodemask(cs, &oldmem);
+done:
+ return retval;
+}
+
int current_cpuset_is_being_rebound(void)
{
return task_cs(current) == cpuset_being_rebound;
if (val != cs->relax_domain_level) {
cs->relax_domain_level = val;
- rebuild_sched_domains();
+ if (!cpus_empty(cs->cpus_allowed) && is_sched_load_balance(cs))
+ rebuild_sched_domains();
}
return 0;
scan.scan.heap = NULL;
scan.to = to->css.cgroup;
- if (cgroup_scan_tasks((struct cgroup_scanner *)&scan))
+ if (cgroup_scan_tasks(&scan.scan))
printk(KERN_ERR "move_member_tasks_to_cpuset: "
"cgroup_scan_tasks failed\n");
}
*/
static void scan_for_empty_cpusets(const 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;
-
- INIT_LIST_HEAD(&queue);
+ nodemask_t oldmems;
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) &&
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);
if (cpus_empty(cp->cpus_allowed) ||
nodes_empty(cp->mems_allowed))
remove_tasks_in_empty_cpuset(cp);
+ else {
+ update_tasks_cpumask(cp);
+ update_tasks_nodemask(cp, &oldmems);
+ }
}
}
}
/**
-
* 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.
git.openpandora.org / pandora-kernel.git / blobdiff