#include <linux/mount.h>
#include <linux/namei.h>
#include <linux/pagemap.h>
-#include <linux/prio_heap.h>
#include <linux/proc_fs.h>
#include <linux/rcupdate.h>
#include <linux/sched.h>
*/
int number_of_cpusets __read_mostly;
-/* Retrieve the cpuset from a cgroup */
+/* Forward declare cgroup structures */
struct cgroup_subsys cpuset_subsys;
struct cpuset;
/* partition number for rebuild_sched_domains() */
int pn;
+ /* for custom sched domain */
+ int relax_domain_level;
+
/* used for walking a cpuset heirarchy */
struct list_head stack_list;
};
typedef enum {
CS_CPU_EXCLUSIVE,
CS_MEM_EXCLUSIVE,
+ CS_MEM_HARDWALL,
CS_MEMORY_MIGRATE,
CS_SCHED_LOAD_BALANCE,
CS_SPREAD_PAGE,
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);
* 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 attach_task() could
+ * 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 anothers memory placement. So we must enable every task,
+ * 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 cpuset_mems_generation is guarded by manage_mutex,
+ * 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;
};
/*
- * We have two global cpuset mutexes below. They can nest.
- * It is ok to first take manage_mutex, then nest callback_mutex. We also
- * require taking task_lock() when dereferencing a tasks cpuset pointer.
- * See "The task_lock() exception", at the end of this comment.
+ * There are two global mutexes guarding cpuset structures. The first
+ * is the main control groups cgroup_mutex, accessed via
+ * cgroup_lock()/cgroup_unlock(). The second is the cpuset-specific
+ * callback_mutex, below. They can nest. It is ok to first take
+ * cgroup_mutex, then nest callback_mutex. We also require taking
+ * task_lock() when dereferencing a task's cpuset pointer. See "The
+ * task_lock() exception", at the end of this comment.
*
* A task must hold both mutexes to modify cpusets. If a task
- * holds manage_mutex, then it blocks others wanting that mutex,
+ * holds cgroup_mutex, then it blocks others wanting that mutex,
* ensuring that it is the only task able to also acquire callback_mutex
* and be able to modify cpusets. It can perform various checks on
* the cpuset structure first, knowing nothing will change. It can
- * also allocate memory while just holding manage_mutex. While it is
+ * also allocate memory while just holding cgroup_mutex. While it is
* performing these checks, various callback routines can briefly
* acquire callback_mutex to query cpusets. Once it is ready to make
* the changes, it takes callback_mutex, blocking everyone else.
* The task_struct fields mems_allowed and mems_generation may only
* be accessed in the context of that task, so require no locks.
*
- * Any task can increment and decrement the count field without lock.
- * So in general, code holding manage_mutex or callback_mutex can't rely
- * on the count field not changing. However, if the count goes to
- * zero, then only attach_task(), which holds both mutexes, can
- * increment it again. Because a count of zero means that no tasks
- * are currently attached, therefore there is no way a task attached
- * to that cpuset can fork (the other way to increment the count).
- * So code holding manage_mutex or callback_mutex can safely assume that
- * if the count is zero, it will stay zero. Similarly, if a task
- * holds manage_mutex or callback_mutex on a cpuset with zero count, it
- * knows that the cpuset won't be removed, as cpuset_rmdir() needs
- * both of those mutexes.
- *
* The cpuset_common_file_write handler for operations that modify
- * the cpuset hierarchy holds manage_mutex across the entire operation,
+ * the cpuset hierarchy holds cgroup_mutex across the entire operation,
* single threading all such cpuset modifications across the system.
*
* The cpuset_common_file_read() handlers only hold callback_mutex across
* small pieces of code, such as when reading out possibly multi-word
* cpumasks and nodemasks.
*
- * The fork and exit callbacks cpuset_fork() and cpuset_exit(), don't
- * (usually) take either mutex. These are the two most performance
- * critical pieces of code here. The exception occurs on cpuset_exit(),
- * when a task in a notify_on_release cpuset exits. Then manage_mutex
- * is taken, and if the cpuset count is zero, a usermode call made
- * to /sbin/cpuset_release_agent with the name of the cpuset (path
- * relative to the root of cpuset file system) as the argument.
- *
- * A cpuset can only be deleted if both its 'count' of using tasks
- * is zero, and its list of 'children' cpusets is empty. Since all
- * tasks in the system use _some_ cpuset, and since there is always at
- * least one task in the system (init), therefore, top_cpuset
- * always has either children cpusets and/or using tasks. So we don't
- * need a special hack to ensure that top_cpuset cannot be deleted.
- *
- * The above "Tale of Two Semaphores" would be complete, but for:
- *
- * The task_lock() exception
- *
- * The need for this exception arises from the action of attach_task(),
- * which overwrites one tasks cpuset pointer with another. It does
- * so using both mutexes, however there are several performance
- * critical places that need to reference task->cpuset without the
- * expense of grabbing a system global mutex. Therefore except as
- * noted below, when dereferencing or, as in attach_task(), modifying
- * a tasks cpuset pointer we use task_lock(), which acts on a spinlock
- * (task->alloc_lock) already in the task_struct routinely used for
- * such matters.
- *
- * P.S. One more locking exception. RCU is used to guard the
- * update of a tasks cpuset pointer by attach_task() and the
- * access of task->cpuset->mems_generation via that pointer in
- * the routine cpuset_update_task_memory_state().
+ * Accessing a task's cpuset should be done in accordance with the
+ * guidelines for accessing subsystem state in kernel/cgroup.c
*/
static DEFINE_MUTEX(callback_mutex);
* Do not call this routine if in_interrupt().
*
* Call without callback_mutex or task_lock() held. May be
- * called with or without manage_mutex held. Thanks in part to
- * 'the_top_cpuset_hack', the tasks cpuset pointer will never
- * be NULL. This routine also might acquire callback_mutex and
- * current->mm->mmap_sem during call.
+ * 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 by attach_task(),
- * using RCU.
+ * 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
*
* One cpuset is a subset of another if all its allowed CPUs and
* Memory Nodes are a subset of the other, and its exclusive flags
- * are only set if the other's are set. Call holding manage_mutex.
+ * are only set if the other's are set. Call holding cgroup_mutex.
*/
static int is_cpuset_subset(const struct cpuset *p, const struct cpuset *q)
* If we replaced the flag and mask values of the current cpuset
* (cur) with those values in the trial cpuset (trial), would
* our various subset and exclusive rules still be valid? Presumes
- * manage_mutex held.
+ * cgroup_mutex held.
*
* 'cur' is the address of an actual, in-use cpuset. Operations
* such as list traversal that depend on the actual address of the
if (!is_cpuset_subset(trial, par))
return -EACCES;
- /* If either I or some sibling (!= me) is exclusive, we can't overlap */
+ /*
+ * If either I or some sibling (!= me) is exclusive, we can't
+ * overlap
+ */
list_for_each_entry(cont, &par->css.cgroup->children, sibling) {
c = cgroup_cs(cont);
if ((is_cpu_exclusive(trial) || is_cpu_exclusive(c)) &&
return cpus_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;
+}
+
/*
* rebuild_sched_domains()
*
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 */
+ 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 */
q = NULL;
csa = NULL;
doms = NULL;
+ dattr = NULL;
/* Special case for the 99% of systems with one, full, sched domain */
if (is_sched_load_balance(&top_cpuset)) {
doms = kmalloc(sizeof(cpumask_t), GFP_KERNEL);
if (!doms)
goto rebuild;
+ dattr = kmalloc(sizeof(struct sched_domain_attr), GFP_KERNEL);
+ if (dattr) {
+ *dattr = SD_ATTR_INIT;
+ update_domain_attr(dattr, &top_cpuset);
+ }
*doms = top_cpuset.cpus_allowed;
goto rebuild;
}
doms = kmalloc(ndoms * sizeof(cpumask_t), GFP_KERNEL);
if (!doms)
goto rebuild;
+ dattr = kmalloc(ndoms * sizeof(struct sched_domain_attr), GFP_KERNEL);
for (nslot = 0, i = 0; i < csn; i++) {
struct cpuset *a = csa[i];
}
cpus_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);
}
}
nslot++;
rebuild:
/* Have scheduler rebuild sched domains */
get_online_cpus();
- partition_sched_domains(ndoms, doms);
+ partition_sched_domains(ndoms, doms, dattr);
put_online_cpus();
done:
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,
return started_after_time(t1, &t2->start_time, t2);
}
-/*
- * Call with manage_mutex held. May take callback_mutex during call.
+/**
+ * cpuset_test_cpumask - test a task's cpus_allowed versus its cpuset's
+ * @tsk: task to test
+ * @scan: struct cgroup_scanner contained in its struct cpuset_hotplug_scanner
+ *
+ * Call with cgroup_mutex held. May take callback_mutex during call.
+ * Called for each task in a cgroup by cgroup_scan_tasks().
+ * 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).
+ */
+static int cpuset_test_cpumask(struct task_struct *tsk,
+ struct cgroup_scanner *scan)
+{
+ return !cpus_equal(tsk->cpus_allowed,
+ (cgroup_cs(scan->cg))->cpus_allowed);
+}
+
+/**
+ * cpuset_change_cpumask - make a task's cpus_allowed the same as its cpuset's
+ * @tsk: task to test
+ * @scan: struct cgroup_scanner containing the cgroup of the task
+ *
+ * Called by cgroup_scan_tasks() for each task in a cgroup whose
+ * cpus_allowed mask needs to be changed.
+ *
+ * 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_cpumask(struct task_struct *tsk,
+ struct cgroup_scanner *scan)
+{
+ set_cpus_allowed_ptr(tsk, &((cgroup_cs(scan->cg))->cpus_allowed));
+}
+/**
+ * update_cpumask - update the cpus_allowed mask of a cpuset and all tasks in it
+ * @cs: the cpuset to consider
+ * @buf: buffer of cpu numbers written to this cpuset
+ */
static int update_cpumask(struct cpuset *cs, char *buf)
{
struct cpuset trialcs;
- int retval, i;
- int is_load_balanced;
- struct cgroup_iter it;
- struct cgroup *cgrp = cs->css.cgroup;
- struct task_struct *p, *dropped;
- /* Never dereference latest_task, since it's not refcounted */
- struct task_struct *latest_task = NULL;
+ struct cgroup_scanner scan;
struct ptr_heap heap;
- struct timespec latest_time = { 0, 0 };
+ int retval;
+ int is_load_balanced;
/* top_cpuset.cpus_allowed tracks cpu_online_map; it's read-only */
if (cs == &top_cpuset)
trialcs = *cs;
/*
- * An empty cpus_allowed is ok iff there are no tasks in the cpuset.
+ * 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.
/* Nothing to do if the cpus didn't change */
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;
cs->cpus_allowed = trialcs.cpus_allowed;
mutex_unlock(&callback_mutex);
- again:
/*
* Scan tasks in the cpuset, and update the cpumasks of any
- * that need an update. Since we can't call set_cpus_allowed()
- * while holding tasklist_lock, gather tasks to be processed
- * in a heap structure. If the statically-sized heap fills up,
- * overflow tasks that started later, and in future iterations
- * only consider tasks that started after the latest task in
- * the previous pass. This guarantees forward progress and
- * that we don't miss any tasks
+ * that need an update.
*/
- heap.size = 0;
- cgroup_iter_start(cgrp, &it);
- while ((p = cgroup_iter_next(cgrp, &it))) {
- /* Only affect tasks that don't have the right cpus_allowed */
- if (cpus_equal(p->cpus_allowed, cs->cpus_allowed))
- continue;
- /*
- * Only process tasks that started after the last task
- * we processed
- */
- if (!started_after_time(p, &latest_time, latest_task))
- continue;
- dropped = heap_insert(&heap, p);
- if (dropped == NULL) {
- get_task_struct(p);
- } else if (dropped != p) {
- get_task_struct(p);
- put_task_struct(dropped);
- }
- }
- cgroup_iter_end(cgrp, &it);
- if (heap.size) {
- for (i = 0; i < heap.size; i++) {
- struct task_struct *p = heap.ptrs[i];
- if (i == 0) {
- latest_time = p->start_time;
- latest_task = p;
- }
- set_cpus_allowed(p, cs->cpus_allowed);
- put_task_struct(p);
- }
- /*
- * If we had to process any tasks at all, scan again
- * in case some of them were in the middle of forking
- * children that didn't notice the new cpumask
- * restriction. Not the most efficient way to do it,
- * but it avoids having to take callback_mutex in the
- * fork path
- */
- goto again;
- }
+ 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);
+
if (is_load_balanced)
rebuild_sched_domains();
-
return 0;
}
* Temporarilly set tasks mems_allowed to target nodes of migration,
* so that the migration code can allocate pages on these nodes.
*
- * Call holding manage_mutex, so our current->cpuset won't change
- * during this call, as manage_mutex holds off any attach_task()
+ * Call holding cgroup_mutex, so current's cpuset won't change
+ * during this call, as manage_mutex holds off any cpuset_attach()
* calls. Therefore we don't need to take task_lock around the
* call to guarantee_online_mems(), as we know no one is changing
- * our tasks cpuset.
+ * our task's cpuset.
*
* Hold callback_mutex around the two modifications of our tasks
* mems_allowed to synchronize with cpuset_mems_allowed().
* the cpuset is marked 'memory_migrate', migrate the tasks
* pages to the new memory.
*
- * Call with manage_mutex held. May take callback_mutex during call.
+ * 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.
cs->mems_generation = cpuset_mems_generation++;
mutex_unlock(&callback_mutex);
- cpuset_being_rebound = cs; /* causes mpol_copy() rebind */
+ cpuset_being_rebound = cs; /* causes mpol_dup() rebind */
fudge = 10; /* spare mmarray[] slots */
fudge += cpus_weight(cs->cpus_allowed); /* imagine one fork-bomb/cpu */
* 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()
+ * tasklist_lock. Forks can happen again now - the mpol_dup()
* cpuset_being_rebound check will catch such forks, and rebind
* their vma mempolicies too. Because we still hold the global
- * cpuset manage_mutex, we know that no other rebind effort will
+ * 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.
mmput(mm);
}
- /* We're done rebinding vma's to this cpusets new mems_allowed. */
+ /* We're done rebinding vmas to this cpuset's new mems_allowed. */
kfree(mmarray);
cpuset_being_rebound = NULL;
retval = 0;
return task_cs(current) == cpuset_being_rebound;
}
-/*
- * Call with manage_mutex held.
- */
-
-static int update_memory_pressure_enabled(struct cpuset *cs, char *buf)
+static int update_relax_domain_level(struct cpuset *cs, s64 val)
{
- if (simple_strtoul(buf, NULL, 10) != 0)
- cpuset_memory_pressure_enabled = 1;
- else
- cpuset_memory_pressure_enabled = 0;
+ if ((int)val < 0)
+ val = -1;
+
+ if (val != cs->relax_domain_level) {
+ cs->relax_domain_level = val;
+ rebuild_sched_domains();
+ }
+
return 0;
}
/*
* 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
- *
- * Call with manage_mutex held.
+ * 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;
int err;
int cpus_nonempty, balance_flag_changed;
- turning_on = (simple_strtoul(buf, NULL, 10) != 0);
-
trialcs = *cs;
if (turning_on)
set_bit(bit, &trialcs.flags);
return val;
}
+/* 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)
{
mutex_lock(&callback_mutex);
guarantee_online_cpus(cs, &cpus);
- set_cpus_allowed(tsk, cpus);
+ set_cpus_allowed_ptr(tsk, &cpus);
mutex_unlock(&callback_mutex);
from = oldcs->mems_allowed;
FILE_MEMLIST,
FILE_CPU_EXCLUSIVE,
FILE_MEM_EXCLUSIVE,
+ FILE_MEM_HARDWALL,
FILE_SCHED_LOAD_BALANCE,
+ FILE_SCHED_RELAX_DOMAIN_LEVEL,
FILE_MEMORY_PRESSURE_ENABLED,
FILE_MEMORY_PRESSURE,
FILE_SPREAD_PAGE,
return -E2BIG;
/* +1 for nul-terminator */
- if ((buffer = kmalloc(nbytes + 1, GFP_KERNEL)) == 0)
+ buffer = kmalloc(nbytes + 1, GFP_KERNEL);
+ if (!buffer)
return -ENOMEM;
if (copy_from_user(buffer, userbuf, nbytes)) {
case FILE_MEMLIST:
retval = update_nodemask(cs, buffer);
break;
+ default:
+ retval = -EINVAL;
+ goto out2;
+ }
+
+ if (retval == 0)
+ retval = nbytes;
+out2:
+ cgroup_unlock();
+out1:
+ kfree(buffer);
+ return retval;
+}
+
+static int cpuset_write_u64(struct cgroup *cgrp, struct cftype *cft, u64 val)
+{
+ int retval = 0;
+ struct cpuset *cs = cgroup_cs(cgrp);
+ cpuset_filetype_t type = cft->private;
+
+ cgroup_lock();
+
+ if (cgroup_is_removed(cgrp)) {
+ cgroup_unlock();
+ return -ENODEV;
+ }
+
+ switch (type) {
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_MEM_HARDWALL:
+ retval = update_flag(CS_MEM_HARDWALL, cs, val);
break;
case FILE_SCHED_LOAD_BALANCE:
- retval = update_flag(CS_SCHED_LOAD_BALANCE, cs, buffer);
+ 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);
+ retval = update_flag(CS_SPREAD_PAGE, cs, val);
cs->mems_generation = cpuset_mems_generation++;
break;
case FILE_SPREAD_SLAB:
- retval = update_flag(CS_SPREAD_SLAB, cs, buffer);
+ retval = update_flag(CS_SPREAD_SLAB, cs, val);
cs->mems_generation = cpuset_mems_generation++;
break;
default:
retval = -EINVAL;
- goto out2;
+ break;
}
+ cgroup_unlock();
+ return retval;
+}
- if (retval == 0)
- retval = nbytes;
-out2:
+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;
+
+ cgroup_lock();
+
+ if (cgroup_is_removed(cgrp)) {
+ cgroup_unlock();
+ return -ENODEV;
+ }
+ switch (type) {
+ case FILE_SCHED_RELAX_DOMAIN_LEVEL:
+ retval = update_relax_domain_level(cs, val);
+ break;
+ default:
+ retval = -EINVAL;
+ break;
+ }
cgroup_unlock();
-out1:
- kfree(buffer);
return retval;
}
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_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;
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();
+ }
+}
-
+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();
+ }
+}
/*
* 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_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 = cpuset_common_file_write,
+ .private = FILE_CPULIST,
+ },
+
+ {
+ .name = "mems",
+ .read = cpuset_common_file_read,
+ .write = cpuset_common_file_write,
+ .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,
+ },
+
+ {
+ .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_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;
}
/*
* If this becomes a problem for some users who wish to
* allow that scenario, then cpuset_post_clone() could be
* changed to grant parent->cpus_allowed-sibling_cpus_exclusive
- * (and likewise for mems) to the new cgroup.
+ * (and likewise for mems) to the new cgroup. Called with cgroup_mutex
+ * held.
*/
static void cpuset_post_clone(struct cgroup_subsys *ss,
struct cgroup *cgroup)
/*
* cpuset_create - create a cpuset
- * parent: cpuset that will be parent of the new cpuset.
- * name: name of the new cpuset. Will be strcpy'ed.
- * mode: mode to set on new inode
- *
- * Must be called with the mutex on the parent inode held
+ * ss: cpuset cgroup subsystem
+ * cont: control group that the new cpuset will be part of
*/
static struct cgroup_subsys_state *cpuset_create(
if (is_spread_slab(parent))
set_bit(CS_SPREAD_SLAB, &cs->flags);
set_bit(CS_SCHED_LOAD_BALANCE, &cs->flags);
- cs->cpus_allowed = CPU_MASK_NONE;
- cs->mems_allowed = NODE_MASK_NONE;
+ cpus_clear(cs->cpus_allowed);
+ nodes_clear(cs->mems_allowed);
cs->mems_generation = cpuset_mems_generation++;
fmeter_init(&cs->fmeter);
+ cs->relax_domain_level = -1;
cs->parent = parent;
number_of_cpusets++;
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--;
kfree(cs);
{
int err = 0;
- top_cpuset.cpus_allowed = CPU_MASK_ALL;
- top_cpuset.mems_allowed = NODE_MASK_ALL;
+ cpus_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;
err = register_filesystem(&cpuset_fs_type);
if (err < 0)
* 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;
* @from: cpuset in which the tasks currently reside
* @to: cpuset to which the tasks will be moved
*
- * Called with manage_sem held
- * callback_mutex must not be held, as attach_task() will take it.
+ * Called with cgroup_mutex held
+ * callback_mutex must not be held, as cpuset_attach() will take it.
*
* The cgroup_scan_tasks() function will scan all the tasks in a cgroup,
* calling callback functions for each.
* last CPU or node from a cpuset, then move the tasks in the empty
* cpuset to its next-highest non-empty parent.
*
- * The parent cpuset has some superset of the 'mems' nodes that the
- * newly empty cpuset held, so no migration of memory is necessary.
- *
- * Called with both manage_sem and callback_sem held
+ * Called with cgroup_mutex held
+ * callback_mutex must not be held, as cpuset_attach() will take it.
*/
static void remove_tasks_in_empty_cpuset(struct cpuset *cs)
{
struct cpuset *parent;
- /* the cgroup's css_sets list is in use if there are tasks
- in the cpuset; the list is empty if there are none;
- the cs->css.refcnt seems always 0 */
+ /*
+ * The cgroup's css_sets list is in use if there are tasks
+ * in the cpuset; the list is empty if there are none;
+ * the cs->css.refcnt seems always 0.
+ */
if (list_empty(&cs->css.cgroup->css_sets))
return;
* has online cpus, so can't be empty).
*/
parent = cs->parent;
- while (cpus_empty(parent->cpus_allowed)) {
- /*
- * this empty cpuset should now be considered to
- * have been used, and therefore eligible for
- * release when empty (if it is notify_on_release)
- */
+ while (cpus_empty(parent->cpus_allowed) ||
+ nodes_empty(parent->mems_allowed))
parent = parent->parent;
- }
move_member_tasks_to_cpuset(cs, parent);
}
* Walk the specified cpuset subtree and look for empty cpusets.
* The tasks of such cpuset must be moved to a parent cpuset.
*
- * Note that such a notify_on_release cpuset must have had, at some time,
- * member tasks or cpuset descendants and cpus and memory, before it can
- * be a candidate for release.
- *
- * Called with manage_mutex held. We take callback_mutex to modify
+ * Called with cgroup_mutex held. We take callback_mutex to modify
* cpus_allowed and mems_allowed.
*
* This walk processes the tree from top to bottom, completing one layer
list_add_tail((struct list_head *)&root->stack_list, &queue);
- mutex_lock(&callback_mutex);
while (!list_empty(&queue)) {
cp = container_of(queue.next, struct cpuset, stack_list);
list_del(queue.next);
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;
+
/* Remove offline cpus and mems from this cpuset. */
+ mutex_lock(&callback_mutex);
cpus_and(cp->cpus_allowed, cp->cpus_allowed, cpu_online_map);
nodes_and(cp->mems_allowed, cp->mems_allowed,
node_states[N_HIGH_MEMORY]);
- if ((cpus_empty(cp->cpus_allowed) ||
- nodes_empty(cp->mems_allowed))) {
- /* Move tasks from the empty cpuset to a parent */
- mutex_unlock(&callback_mutex);
+ mutex_unlock(&callback_mutex);
+
+ /* Move tasks from the empty cpuset to a parent */
+ if (cpus_empty(cp->cpus_allowed) ||
+ nodes_empty(cp->mems_allowed))
remove_tasks_in_empty_cpuset(cp);
- mutex_lock(&callback_mutex);
- }
}
- mutex_unlock(&callback_mutex);
- return;
}
/*
* 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.
*
* Description: Returns the cpumask_t cpus_allowed of the cpuset
* attached to the specified @tsk. Guaranteed to return some non-empty
* tasks cpuset.
**/
-cpumask_t cpuset_cpus_allowed(struct task_struct *tsk)
+void cpuset_cpus_allowed(struct task_struct *tsk, cpumask_t *pmask)
{
- cpumask_t mask;
-
mutex_lock(&callback_mutex);
- mask = cpuset_cpus_allowed_locked(tsk);
+ cpuset_cpus_allowed_locked(tsk, pmask);
mutex_unlock(&callback_mutex);
-
- return mask;
}
/**
* cpuset_cpus_allowed_locked - return cpus_allowed mask from a tasks cpuset.
- * Must be called with callback_mutex held.
+ * Must be called with callback_mutex held.
**/
-cpumask_t cpuset_cpus_allowed_locked(struct task_struct *tsk)
+void cpuset_cpus_allowed_locked(struct task_struct *tsk, cpumask_t *pmask)
{
- cpumask_t mask;
-
task_lock(tsk);
- guarantee_online_cpus(task_cs(tsk), &mask);
+ guarantee_online_cpus(task_cs(tsk), pmask);
task_unlock(tsk);
-
- return mask;
}
void cpuset_init_current_mems_allowed(void)
{
- current->mems_allowed = NODE_MASK_ALL;
+ nodes_setall(current->mems_allowed);
}
/**
}
/**
- * 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;
}
* __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.
+ * hardwalled 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.
* Otherwise, no.
* 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
* 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:
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);
* - Used for /proc/<pid>/cpuset.
* - No need to task_lock(tsk) on this tsk->cpuset reference, as it
* doesn't really matter if tsk->cpuset changes after we read it,
- * and we take manage_mutex, keeping attach_task() from changing it
- * anyway. No need to check that tsk->cpuset != NULL, thanks to
- * the_top_cpuset_hack in cpuset_exit(), which sets an exiting tasks
- * cpuset to top_cpuset.
+ * and we take cgroup_mutex, keeping cpuset_attach() from changing it
+ * anyway.
*/
static int proc_cpuset_show(struct seq_file *m, void *unused_v)
{
#endif /* CONFIG_PROC_PID_CPUSET */
/* Display task cpus_allowed, mems_allowed in /proc/<pid>/status file. */
-char *cpuset_task_status_allowed(struct task_struct *task, char *buffer)
-{
- buffer += sprintf(buffer, "Cpus_allowed:\t");
- buffer += cpumask_scnprintf(buffer, PAGE_SIZE, task->cpus_allowed);
- buffer += sprintf(buffer, "\n");
- buffer += sprintf(buffer, "Mems_allowed:\t");
- buffer += nodemask_scnprintf(buffer, PAGE_SIZE, task->mems_allowed);
- buffer += sprintf(buffer, "\n");
- return buffer;
+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_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_printf(m, "\n");
}