2005-04-17 02:20:36 +04:00
|
|
|
/* Kernel thread helper functions.
|
|
|
|
* Copyright (C) 2004 IBM Corporation, Rusty Russell.
|
|
|
|
*
|
2007-05-09 13:34:32 +04:00
|
|
|
* Creation is done via kthreadd, so that we get a clean environment
|
2005-04-17 02:20:36 +04:00
|
|
|
* even if we're invoked from userspace (think modprobe, hotplug cpu,
|
|
|
|
* etc.).
|
|
|
|
*/
|
|
|
|
#include <linux/sched.h>
|
|
|
|
#include <linux/kthread.h>
|
|
|
|
#include <linux/completion.h>
|
|
|
|
#include <linux/err.h>
|
cpuset,mm: update tasks' mems_allowed in time
Fix allocating page cache/slab object on the unallowed node when memory
spread is set by updating tasks' mems_allowed after its cpuset's mems is
changed.
In order to update tasks' mems_allowed in time, we must modify the code of
memory policy. Because the memory policy is applied in the process's
context originally. After applying this patch, one task directly
manipulates anothers mems_allowed, and we use alloc_lock in the
task_struct to protect mems_allowed and memory policy of the task.
But in the fast path, we didn't use lock to protect them, because adding a
lock may lead to performance regression. But if we don't add a lock,the
task might see no nodes when changing cpuset's mems_allowed to some
non-overlapping set. In order to avoid it, we set all new allowed nodes,
then clear newly disallowed ones.
[lee.schermerhorn@hp.com:
The rework of mpol_new() to extract the adjusting of the node mask to
apply cpuset and mpol flags "context" breaks set_mempolicy() and mbind()
with MPOL_PREFERRED and a NULL nodemask--i.e., explicit local
allocation. Fix this by adding the check for MPOL_PREFERRED and empty
node mask to mpol_new_mpolicy().
Remove the now unneeded 'nodes = NULL' from mpol_new().
Note that mpol_new_mempolicy() is always called with a non-NULL
'nodes' parameter now that it has been removed from mpol_new().
Therefore, we don't need to test nodes for NULL before testing it for
'empty'. However, just to be extra paranoid, add a VM_BUG_ON() to
verify this assumption.]
[lee.schermerhorn@hp.com:
I don't think the function name 'mpol_new_mempolicy' is descriptive
enough to differentiate it from mpol_new().
This function applies cpuset set context, usually constraining nodes
to those allowed by the cpuset. However, when the 'RELATIVE_NODES flag
is set, it also translates the nodes. So I settled on
'mpol_set_nodemask()', because the comment block for mpol_new() mentions
that we need to call this function to "set nodes".
Some additional minor line length, whitespace and typo cleanup.]
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Christoph Lameter <cl@linux-foundation.org>
Cc: Paul Menage <menage@google.com>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Yasunori Goto <y-goto@jp.fujitsu.com>
Cc: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2009-06-17 02:31:49 +04:00
|
|
|
#include <linux/cpuset.h>
|
2005-04-17 02:20:36 +04:00
|
|
|
#include <linux/unistd.h>
|
|
|
|
#include <linux/file.h>
|
2011-05-23 22:51:41 +04:00
|
|
|
#include <linux/export.h>
|
2006-03-23 14:00:24 +03:00
|
|
|
#include <linux/mutex.h>
|
2010-06-29 12:07:09 +04:00
|
|
|
#include <linux/slab.h>
|
|
|
|
#include <linux/freezer.h>
|
2012-10-11 05:28:25 +04:00
|
|
|
#include <linux/ptrace.h>
|
2013-05-01 02:27:21 +04:00
|
|
|
#include <linux/uaccess.h>
|
2009-04-15 03:39:12 +04:00
|
|
|
#include <trace/events/sched.h>
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2007-05-09 13:34:32 +04:00
|
|
|
static DEFINE_SPINLOCK(kthread_create_lock);
|
|
|
|
static LIST_HEAD(kthread_create_list);
|
|
|
|
struct task_struct *kthreadd_task;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
|
|
|
struct kthread_create_info
|
|
|
|
{
|
2007-05-09 13:34:32 +04:00
|
|
|
/* Information passed to kthread() from kthreadd. */
|
2005-04-17 02:20:36 +04:00
|
|
|
int (*threadfn)(void *data);
|
|
|
|
void *data;
|
2011-03-23 02:30:44 +03:00
|
|
|
int node;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2007-05-09 13:34:32 +04:00
|
|
|
/* Result passed back to kthread_create() from kthreadd. */
|
2005-04-17 02:20:36 +04:00
|
|
|
struct task_struct *result;
|
2013-11-13 03:06:45 +04:00
|
|
|
struct completion *done;
|
2006-11-22 17:55:48 +03:00
|
|
|
|
2007-05-09 13:34:32 +04:00
|
|
|
struct list_head list;
|
2005-04-17 02:20:36 +04:00
|
|
|
};
|
|
|
|
|
2009-06-18 03:27:45 +04:00
|
|
|
struct kthread {
|
2012-07-16 14:42:36 +04:00
|
|
|
unsigned long flags;
|
|
|
|
unsigned int cpu;
|
2010-06-29 12:07:09 +04:00
|
|
|
void *data;
|
2012-07-16 14:42:36 +04:00
|
|
|
struct completion parked;
|
2009-06-18 03:27:45 +04:00
|
|
|
struct completion exited;
|
2005-04-17 02:20:36 +04:00
|
|
|
};
|
|
|
|
|
2012-07-16 14:42:36 +04:00
|
|
|
enum KTHREAD_BITS {
|
|
|
|
KTHREAD_IS_PER_CPU = 0,
|
|
|
|
KTHREAD_SHOULD_STOP,
|
|
|
|
KTHREAD_SHOULD_PARK,
|
|
|
|
KTHREAD_IS_PARKED,
|
|
|
|
};
|
|
|
|
|
2013-04-30 02:05:01 +04:00
|
|
|
#define __to_kthread(vfork) \
|
|
|
|
container_of(vfork, struct kthread, exited)
|
|
|
|
|
|
|
|
static inline struct kthread *to_kthread(struct task_struct *k)
|
|
|
|
{
|
|
|
|
return __to_kthread(k->vfork_done);
|
|
|
|
}
|
|
|
|
|
|
|
|
static struct kthread *to_live_kthread(struct task_struct *k)
|
|
|
|
{
|
|
|
|
struct completion *vfork = ACCESS_ONCE(k->vfork_done);
|
|
|
|
if (likely(vfork))
|
|
|
|
return __to_kthread(vfork);
|
|
|
|
return NULL;
|
|
|
|
}
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2006-06-25 16:49:19 +04:00
|
|
|
/**
|
|
|
|
* kthread_should_stop - should this kthread return now?
|
|
|
|
*
|
2007-02-10 12:45:59 +03:00
|
|
|
* When someone calls kthread_stop() on your kthread, it will be woken
|
2006-06-25 16:49:19 +04:00
|
|
|
* and this will return true. You should then return, and your return
|
|
|
|
* value will be passed through to kthread_stop().
|
|
|
|
*/
|
2012-07-16 14:42:36 +04:00
|
|
|
bool kthread_should_stop(void)
|
2005-04-17 02:20:36 +04:00
|
|
|
{
|
2012-07-16 14:42:36 +04:00
|
|
|
return test_bit(KTHREAD_SHOULD_STOP, &to_kthread(current)->flags);
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(kthread_should_stop);
|
|
|
|
|
2012-07-16 14:42:36 +04:00
|
|
|
/**
|
|
|
|
* kthread_should_park - should this kthread park now?
|
|
|
|
*
|
|
|
|
* When someone calls kthread_park() on your kthread, it will be woken
|
|
|
|
* and this will return true. You should then do the necessary
|
|
|
|
* cleanup and call kthread_parkme()
|
|
|
|
*
|
|
|
|
* Similar to kthread_should_stop(), but this keeps the thread alive
|
|
|
|
* and in a park position. kthread_unpark() "restarts" the thread and
|
|
|
|
* calls the thread function again.
|
|
|
|
*/
|
|
|
|
bool kthread_should_park(void)
|
|
|
|
{
|
|
|
|
return test_bit(KTHREAD_SHOULD_PARK, &to_kthread(current)->flags);
|
|
|
|
}
|
2015-08-07 01:46:45 +03:00
|
|
|
EXPORT_SYMBOL_GPL(kthread_should_park);
|
2012-07-16 14:42:36 +04:00
|
|
|
|
2011-11-22 00:32:23 +04:00
|
|
|
/**
|
|
|
|
* kthread_freezable_should_stop - should this freezable kthread return now?
|
|
|
|
* @was_frozen: optional out parameter, indicates whether %current was frozen
|
|
|
|
*
|
|
|
|
* kthread_should_stop() for freezable kthreads, which will enter
|
|
|
|
* refrigerator if necessary. This function is safe from kthread_stop() /
|
|
|
|
* freezer deadlock and freezable kthreads should use this function instead
|
|
|
|
* of calling try_to_freeze() directly.
|
|
|
|
*/
|
|
|
|
bool kthread_freezable_should_stop(bool *was_frozen)
|
|
|
|
{
|
|
|
|
bool frozen = false;
|
|
|
|
|
|
|
|
might_sleep();
|
|
|
|
|
|
|
|
if (unlikely(freezing(current)))
|
|
|
|
frozen = __refrigerator(true);
|
|
|
|
|
|
|
|
if (was_frozen)
|
|
|
|
*was_frozen = frozen;
|
|
|
|
|
|
|
|
return kthread_should_stop();
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(kthread_freezable_should_stop);
|
|
|
|
|
2010-06-29 12:07:09 +04:00
|
|
|
/**
|
|
|
|
* kthread_data - return data value specified on kthread creation
|
|
|
|
* @task: kthread task in question
|
|
|
|
*
|
|
|
|
* Return the data value specified when kthread @task was created.
|
|
|
|
* The caller is responsible for ensuring the validity of @task when
|
|
|
|
* calling this function.
|
|
|
|
*/
|
|
|
|
void *kthread_data(struct task_struct *task)
|
|
|
|
{
|
|
|
|
return to_kthread(task)->data;
|
|
|
|
}
|
|
|
|
|
2013-05-01 02:27:21 +04:00
|
|
|
/**
|
|
|
|
* probe_kthread_data - speculative version of kthread_data()
|
|
|
|
* @task: possible kthread task in question
|
|
|
|
*
|
|
|
|
* @task could be a kthread task. Return the data value specified when it
|
|
|
|
* was created if accessible. If @task isn't a kthread task or its data is
|
|
|
|
* inaccessible for any reason, %NULL is returned. This function requires
|
|
|
|
* that @task itself is safe to dereference.
|
|
|
|
*/
|
|
|
|
void *probe_kthread_data(struct task_struct *task)
|
|
|
|
{
|
|
|
|
struct kthread *kthread = to_kthread(task);
|
|
|
|
void *data = NULL;
|
|
|
|
|
|
|
|
probe_kernel_read(&data, &kthread->data, sizeof(data));
|
|
|
|
return data;
|
|
|
|
}
|
|
|
|
|
2012-07-16 14:42:36 +04:00
|
|
|
static void __kthread_parkme(struct kthread *self)
|
|
|
|
{
|
2013-04-09 11:33:34 +04:00
|
|
|
__set_current_state(TASK_PARKED);
|
2012-07-16 14:42:36 +04:00
|
|
|
while (test_bit(KTHREAD_SHOULD_PARK, &self->flags)) {
|
|
|
|
if (!test_and_set_bit(KTHREAD_IS_PARKED, &self->flags))
|
|
|
|
complete(&self->parked);
|
|
|
|
schedule();
|
2013-04-09 11:33:34 +04:00
|
|
|
__set_current_state(TASK_PARKED);
|
2012-07-16 14:42:36 +04:00
|
|
|
}
|
|
|
|
clear_bit(KTHREAD_IS_PARKED, &self->flags);
|
|
|
|
__set_current_state(TASK_RUNNING);
|
|
|
|
}
|
|
|
|
|
|
|
|
void kthread_parkme(void)
|
|
|
|
{
|
|
|
|
__kthread_parkme(to_kthread(current));
|
|
|
|
}
|
2015-08-07 01:46:45 +03:00
|
|
|
EXPORT_SYMBOL_GPL(kthread_parkme);
|
2012-07-16 14:42:36 +04:00
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
static int kthread(void *_create)
|
|
|
|
{
|
2009-06-18 03:27:45 +04:00
|
|
|
/* Copy data: it's on kthread's stack */
|
2005-04-17 02:20:36 +04:00
|
|
|
struct kthread_create_info *create = _create;
|
2009-06-18 03:27:45 +04:00
|
|
|
int (*threadfn)(void *data) = create->threadfn;
|
|
|
|
void *data = create->data;
|
2013-11-13 03:06:45 +04:00
|
|
|
struct completion *done;
|
2009-06-18 03:27:45 +04:00
|
|
|
struct kthread self;
|
|
|
|
int ret;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2012-07-16 14:42:36 +04:00
|
|
|
self.flags = 0;
|
2010-06-29 12:07:09 +04:00
|
|
|
self.data = data;
|
2009-06-18 03:27:45 +04:00
|
|
|
init_completion(&self.exited);
|
2012-07-16 14:42:36 +04:00
|
|
|
init_completion(&self.parked);
|
2009-06-18 03:27:45 +04:00
|
|
|
current->vfork_done = &self.exited;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2013-11-13 03:06:45 +04:00
|
|
|
/* If user was SIGKILLed, I release the structure. */
|
|
|
|
done = xchg(&create->done, NULL);
|
|
|
|
if (!done) {
|
|
|
|
kfree(create);
|
|
|
|
do_exit(-EINTR);
|
|
|
|
}
|
2005-04-17 02:20:36 +04:00
|
|
|
/* OK, tell user we're spawned, wait for stop or wakeup */
|
2007-05-24 00:57:27 +04:00
|
|
|
__set_current_state(TASK_UNINTERRUPTIBLE);
|
2009-04-09 19:50:35 +04:00
|
|
|
create->result = current;
|
2013-11-13 03:06:45 +04:00
|
|
|
complete(done);
|
2005-04-17 02:20:36 +04:00
|
|
|
schedule();
|
|
|
|
|
2009-06-18 03:27:45 +04:00
|
|
|
ret = -EINTR;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2012-07-16 14:42:36 +04:00
|
|
|
if (!test_bit(KTHREAD_SHOULD_STOP, &self.flags)) {
|
|
|
|
__kthread_parkme(&self);
|
|
|
|
ret = threadfn(data);
|
|
|
|
}
|
2009-06-18 03:27:45 +04:00
|
|
|
/* we can't just return, we must preserve "self" on stack */
|
|
|
|
do_exit(ret);
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
|
|
|
|
2011-03-23 02:30:44 +03:00
|
|
|
/* called from do_fork() to get node information for about to be created task */
|
|
|
|
int tsk_fork_get_node(struct task_struct *tsk)
|
|
|
|
{
|
|
|
|
#ifdef CONFIG_NUMA
|
|
|
|
if (tsk == kthreadd_task)
|
|
|
|
return tsk->pref_node_fork;
|
|
|
|
#endif
|
2014-04-04 01:46:25 +04:00
|
|
|
return NUMA_NO_NODE;
|
2011-03-23 02:30:44 +03:00
|
|
|
}
|
|
|
|
|
2007-05-09 13:34:32 +04:00
|
|
|
static void create_kthread(struct kthread_create_info *create)
|
2005-04-17 02:20:36 +04:00
|
|
|
{
|
|
|
|
int pid;
|
|
|
|
|
2011-03-23 02:30:44 +03:00
|
|
|
#ifdef CONFIG_NUMA
|
|
|
|
current->pref_node_fork = create->node;
|
|
|
|
#endif
|
2005-04-17 02:20:36 +04:00
|
|
|
/* We want our own signal handler (we take no signals by default). */
|
|
|
|
pid = kernel_thread(kthread, create, CLONE_FS | CLONE_FILES | SIGCHLD);
|
2009-06-18 03:27:43 +04:00
|
|
|
if (pid < 0) {
|
2013-11-13 03:06:45 +04:00
|
|
|
/* If user was SIGKILLed, I release the structure. */
|
|
|
|
struct completion *done = xchg(&create->done, NULL);
|
|
|
|
|
|
|
|
if (!done) {
|
|
|
|
kfree(create);
|
|
|
|
return;
|
|
|
|
}
|
2005-04-17 02:20:36 +04:00
|
|
|
create->result = ERR_PTR(pid);
|
2013-11-13 03:06:45 +04:00
|
|
|
complete(done);
|
2009-06-18 03:27:43 +04:00
|
|
|
}
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
|
|
|
|
2006-06-25 16:49:19 +04:00
|
|
|
/**
|
2011-03-23 02:30:44 +03:00
|
|
|
* kthread_create_on_node - create a kthread.
|
2006-06-25 16:49:19 +04:00
|
|
|
* @threadfn: the function to run until signal_pending(current).
|
|
|
|
* @data: data ptr for @threadfn.
|
2015-09-05 01:42:42 +03:00
|
|
|
* @node: task and thread structures for the thread are allocated on this node
|
2006-06-25 16:49:19 +04:00
|
|
|
* @namefmt: printf-style name for the thread.
|
|
|
|
*
|
|
|
|
* Description: This helper function creates and names a kernel
|
|
|
|
* thread. The thread will be stopped: use wake_up_process() to start
|
2015-09-05 01:42:42 +03:00
|
|
|
* it. See also kthread_run(). The new thread has SCHED_NORMAL policy and
|
|
|
|
* is affine to all CPUs.
|
2006-06-25 16:49:19 +04:00
|
|
|
*
|
2011-03-23 02:30:44 +03:00
|
|
|
* If thread is going to be bound on a particular cpu, give its node
|
2015-09-05 01:42:42 +03:00
|
|
|
* in @node, to get NUMA affinity for kthread stack, or else give NUMA_NO_NODE.
|
2006-06-25 16:49:19 +04:00
|
|
|
* When woken, the thread will run @threadfn() with @data as its
|
2007-02-10 12:45:59 +03:00
|
|
|
* argument. @threadfn() can either call do_exit() directly if it is a
|
2011-03-31 05:57:33 +04:00
|
|
|
* standalone thread for which no one will call kthread_stop(), or
|
2006-06-25 16:49:19 +04:00
|
|
|
* return when 'kthread_should_stop()' is true (which means
|
|
|
|
* kthread_stop() has been called). The return value should be zero
|
|
|
|
* or a negative error number; it will be passed to kthread_stop().
|
|
|
|
*
|
2014-06-05 03:05:36 +04:00
|
|
|
* Returns a task_struct or ERR_PTR(-ENOMEM) or ERR_PTR(-EINTR).
|
2006-06-25 16:49:19 +04:00
|
|
|
*/
|
2011-03-23 02:30:44 +03:00
|
|
|
struct task_struct *kthread_create_on_node(int (*threadfn)(void *data),
|
2012-07-16 14:42:36 +04:00
|
|
|
void *data, int node,
|
2011-03-23 02:30:44 +03:00
|
|
|
const char namefmt[],
|
|
|
|
...)
|
2005-04-17 02:20:36 +04:00
|
|
|
{
|
2013-11-13 03:06:45 +04:00
|
|
|
DECLARE_COMPLETION_ONSTACK(done);
|
|
|
|
struct task_struct *task;
|
|
|
|
struct kthread_create_info *create = kmalloc(sizeof(*create),
|
|
|
|
GFP_KERNEL);
|
|
|
|
|
|
|
|
if (!create)
|
|
|
|
return ERR_PTR(-ENOMEM);
|
|
|
|
create->threadfn = threadfn;
|
|
|
|
create->data = data;
|
|
|
|
create->node = node;
|
|
|
|
create->done = &done;
|
2007-05-09 13:34:32 +04:00
|
|
|
|
|
|
|
spin_lock(&kthread_create_lock);
|
2013-11-13 03:06:45 +04:00
|
|
|
list_add_tail(&create->list, &kthread_create_list);
|
2007-05-09 13:34:32 +04:00
|
|
|
spin_unlock(&kthread_create_lock);
|
|
|
|
|
2008-04-29 11:59:23 +04:00
|
|
|
wake_up_process(kthreadd_task);
|
2013-11-13 03:06:45 +04:00
|
|
|
/*
|
|
|
|
* Wait for completion in killable state, for I might be chosen by
|
|
|
|
* the OOM killer while kthreadd is trying to allocate memory for
|
|
|
|
* new kernel thread.
|
|
|
|
*/
|
|
|
|
if (unlikely(wait_for_completion_killable(&done))) {
|
|
|
|
/*
|
|
|
|
* If I was SIGKILLed before kthreadd (or new kernel thread)
|
|
|
|
* calls complete(), leave the cleanup of this structure to
|
|
|
|
* that thread.
|
|
|
|
*/
|
|
|
|
if (xchg(&create->done, NULL))
|
2014-06-05 03:05:36 +04:00
|
|
|
return ERR_PTR(-EINTR);
|
2013-11-13 03:06:45 +04:00
|
|
|
/*
|
|
|
|
* kthreadd (or new kernel thread) will call complete()
|
|
|
|
* shortly.
|
|
|
|
*/
|
|
|
|
wait_for_completion(&done);
|
|
|
|
}
|
|
|
|
task = create->result;
|
|
|
|
if (!IS_ERR(task)) {
|
2011-01-07 15:41:40 +03:00
|
|
|
static const struct sched_param param = { .sched_priority = 0 };
|
2005-04-17 02:20:36 +04:00
|
|
|
va_list args;
|
2009-04-09 19:50:36 +04:00
|
|
|
|
2005-04-17 02:20:36 +04:00
|
|
|
va_start(args, namefmt);
|
2013-11-13 03:06:45 +04:00
|
|
|
vsnprintf(task->comm, sizeof(task->comm), namefmt, args);
|
2005-04-17 02:20:36 +04:00
|
|
|
va_end(args);
|
2009-04-09 19:50:36 +04:00
|
|
|
/*
|
|
|
|
* root may have changed our (kthreadd's) priority or CPU mask.
|
|
|
|
* The kernel thread should not inherit these properties.
|
|
|
|
*/
|
2013-11-13 03:06:45 +04:00
|
|
|
sched_setscheduler_nocheck(task, SCHED_NORMAL, ¶m);
|
|
|
|
set_cpus_allowed_ptr(task, cpu_all_mask);
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
2013-11-13 03:06:45 +04:00
|
|
|
kfree(create);
|
|
|
|
return task;
|
2005-04-17 02:20:36 +04:00
|
|
|
}
|
2011-03-23 02:30:44 +03:00
|
|
|
EXPORT_SYMBOL(kthread_create_on_node);
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2015-05-15 18:43:34 +03:00
|
|
|
static void __kthread_bind_mask(struct task_struct *p, const struct cpumask *mask, long state)
|
2012-07-16 14:42:36 +04:00
|
|
|
{
|
2015-05-15 18:43:34 +03:00
|
|
|
unsigned long flags;
|
|
|
|
|
2013-04-09 11:33:34 +04:00
|
|
|
if (!wait_task_inactive(p, state)) {
|
|
|
|
WARN_ON(1);
|
|
|
|
return;
|
|
|
|
}
|
2015-05-15 18:43:34 +03:00
|
|
|
|
2012-07-16 14:42:36 +04:00
|
|
|
/* It's safe because the task is inactive. */
|
2015-05-15 18:43:34 +03:00
|
|
|
raw_spin_lock_irqsave(&p->pi_lock, flags);
|
|
|
|
do_set_cpus_allowed(p, mask);
|
2013-03-20 00:45:20 +04:00
|
|
|
p->flags |= PF_NO_SETAFFINITY;
|
2015-05-15 18:43:34 +03:00
|
|
|
raw_spin_unlock_irqrestore(&p->pi_lock, flags);
|
|
|
|
}
|
|
|
|
|
|
|
|
static void __kthread_bind(struct task_struct *p, unsigned int cpu, long state)
|
|
|
|
{
|
|
|
|
__kthread_bind_mask(p, cpumask_of(cpu), state);
|
|
|
|
}
|
|
|
|
|
|
|
|
void kthread_bind_mask(struct task_struct *p, const struct cpumask *mask)
|
|
|
|
{
|
|
|
|
__kthread_bind_mask(p, mask, TASK_UNINTERRUPTIBLE);
|
2012-07-16 14:42:36 +04:00
|
|
|
}
|
|
|
|
|
2009-12-16 20:04:39 +03:00
|
|
|
/**
|
|
|
|
* kthread_bind - bind a just-created kthread to a cpu.
|
|
|
|
* @p: thread created by kthread_create().
|
|
|
|
* @cpu: cpu (might not be online, must be possible) for @k to run on.
|
|
|
|
*
|
|
|
|
* Description: This function is equivalent to set_cpus_allowed(),
|
|
|
|
* except that @cpu doesn't need to be online, and the thread must be
|
|
|
|
* stopped (i.e., just returned from kthread_create()).
|
|
|
|
*/
|
|
|
|
void kthread_bind(struct task_struct *p, unsigned int cpu)
|
|
|
|
{
|
2013-04-09 11:33:34 +04:00
|
|
|
__kthread_bind(p, cpu, TASK_UNINTERRUPTIBLE);
|
2009-12-16 20:04:39 +03:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL(kthread_bind);
|
|
|
|
|
2012-07-16 14:42:36 +04:00
|
|
|
/**
|
|
|
|
* kthread_create_on_cpu - Create a cpu bound kthread
|
|
|
|
* @threadfn: the function to run until signal_pending(current).
|
|
|
|
* @data: data ptr for @threadfn.
|
|
|
|
* @cpu: The cpu on which the thread should be bound,
|
|
|
|
* @namefmt: printf-style name for the thread. Format is restricted
|
|
|
|
* to "name.*%u". Code fills in cpu number.
|
|
|
|
*
|
|
|
|
* Description: This helper function creates and names a kernel thread
|
|
|
|
* The thread will be woken and put into park mode.
|
|
|
|
*/
|
|
|
|
struct task_struct *kthread_create_on_cpu(int (*threadfn)(void *data),
|
|
|
|
void *data, unsigned int cpu,
|
|
|
|
const char *namefmt)
|
|
|
|
{
|
|
|
|
struct task_struct *p;
|
|
|
|
|
2014-10-10 02:26:18 +04:00
|
|
|
p = kthread_create_on_node(threadfn, data, cpu_to_node(cpu), namefmt,
|
2012-07-16 14:42:36 +04:00
|
|
|
cpu);
|
|
|
|
if (IS_ERR(p))
|
|
|
|
return p;
|
|
|
|
set_bit(KTHREAD_IS_PER_CPU, &to_kthread(p)->flags);
|
|
|
|
to_kthread(p)->cpu = cpu;
|
|
|
|
/* Park the thread to get it out of TASK_UNINTERRUPTIBLE state */
|
|
|
|
kthread_park(p);
|
|
|
|
return p;
|
|
|
|
}
|
|
|
|
|
2013-04-09 11:33:34 +04:00
|
|
|
static void __kthread_unpark(struct task_struct *k, struct kthread *kthread)
|
|
|
|
{
|
|
|
|
clear_bit(KTHREAD_SHOULD_PARK, &kthread->flags);
|
|
|
|
/*
|
|
|
|
* We clear the IS_PARKED bit here as we don't wait
|
|
|
|
* until the task has left the park code. So if we'd
|
|
|
|
* park before that happens we'd see the IS_PARKED bit
|
|
|
|
* which might be about to be cleared.
|
|
|
|
*/
|
|
|
|
if (test_and_clear_bit(KTHREAD_IS_PARKED, &kthread->flags)) {
|
|
|
|
if (test_bit(KTHREAD_IS_PER_CPU, &kthread->flags))
|
|
|
|
__kthread_bind(k, kthread->cpu, TASK_PARKED);
|
|
|
|
wake_up_state(k, TASK_PARKED);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
|
2012-07-16 14:42:36 +04:00
|
|
|
/**
|
|
|
|
* kthread_unpark - unpark a thread created by kthread_create().
|
|
|
|
* @k: thread created by kthread_create().
|
|
|
|
*
|
|
|
|
* Sets kthread_should_park() for @k to return false, wakes it, and
|
|
|
|
* waits for it to return. If the thread is marked percpu then its
|
|
|
|
* bound to the cpu again.
|
|
|
|
*/
|
|
|
|
void kthread_unpark(struct task_struct *k)
|
|
|
|
{
|
2013-04-30 02:05:12 +04:00
|
|
|
struct kthread *kthread = to_live_kthread(k);
|
2012-07-16 14:42:36 +04:00
|
|
|
|
2013-04-09 11:33:34 +04:00
|
|
|
if (kthread)
|
|
|
|
__kthread_unpark(k, kthread);
|
2012-07-16 14:42:36 +04:00
|
|
|
}
|
2015-08-07 01:46:45 +03:00
|
|
|
EXPORT_SYMBOL_GPL(kthread_unpark);
|
2012-07-16 14:42:36 +04:00
|
|
|
|
|
|
|
/**
|
|
|
|
* kthread_park - park a thread created by kthread_create().
|
|
|
|
* @k: thread created by kthread_create().
|
|
|
|
*
|
|
|
|
* Sets kthread_should_park() for @k to return true, wakes it, and
|
|
|
|
* waits for it to return. This can also be called after kthread_create()
|
|
|
|
* instead of calling wake_up_process(): the thread will park without
|
|
|
|
* calling threadfn().
|
|
|
|
*
|
|
|
|
* Returns 0 if the thread is parked, -ENOSYS if the thread exited.
|
|
|
|
* If called by the kthread itself just the park bit is set.
|
|
|
|
*/
|
|
|
|
int kthread_park(struct task_struct *k)
|
|
|
|
{
|
2013-04-30 02:05:12 +04:00
|
|
|
struct kthread *kthread = to_live_kthread(k);
|
2012-07-16 14:42:36 +04:00
|
|
|
int ret = -ENOSYS;
|
|
|
|
|
|
|
|
if (kthread) {
|
|
|
|
if (!test_bit(KTHREAD_IS_PARKED, &kthread->flags)) {
|
|
|
|
set_bit(KTHREAD_SHOULD_PARK, &kthread->flags);
|
|
|
|
if (k != current) {
|
|
|
|
wake_up_process(k);
|
|
|
|
wait_for_completion(&kthread->parked);
|
|
|
|
}
|
|
|
|
}
|
|
|
|
ret = 0;
|
|
|
|
}
|
|
|
|
return ret;
|
|
|
|
}
|
2015-08-07 01:46:45 +03:00
|
|
|
EXPORT_SYMBOL_GPL(kthread_park);
|
2012-07-16 14:42:36 +04:00
|
|
|
|
2006-06-25 16:49:19 +04:00
|
|
|
/**
|
|
|
|
* kthread_stop - stop a thread created by kthread_create().
|
|
|
|
* @k: thread created by kthread_create().
|
|
|
|
*
|
|
|
|
* Sets kthread_should_stop() for @k to return true, wakes it, and
|
2009-06-19 04:51:13 +04:00
|
|
|
* waits for it to exit. This can also be called after kthread_create()
|
|
|
|
* instead of calling wake_up_process(): the thread will exit without
|
|
|
|
* calling threadfn().
|
|
|
|
*
|
|
|
|
* If threadfn() may call do_exit() itself, the caller must ensure
|
|
|
|
* task_struct can't go away.
|
2006-06-25 16:49:19 +04:00
|
|
|
*
|
|
|
|
* Returns the result of threadfn(), or %-EINTR if wake_up_process()
|
|
|
|
* was never called.
|
|
|
|
*/
|
2005-04-17 02:20:36 +04:00
|
|
|
int kthread_stop(struct task_struct *k)
|
|
|
|
{
|
2013-04-30 02:05:12 +04:00
|
|
|
struct kthread *kthread;
|
2005-04-17 02:20:36 +04:00
|
|
|
int ret;
|
|
|
|
|
tracing, sched: LTTng instrumentation - scheduler
Instrument the scheduler activity (sched_switch, migration, wakeups,
wait for a task, signal delivery) and process/thread
creation/destruction (fork, exit, kthread stop). Actually, kthread
creation is not instrumented in this patch because it is architecture
dependent. It allows to connect tracers such as ftrace which detects
scheduling latencies, good/bad scheduler decisions. Tools like LTTng can
export this scheduler information along with instrumentation of the rest
of the kernel activity to perform post-mortem analysis on the scheduler
activity.
About the performance impact of tracepoints (which is comparable to
markers), even without immediate values optimizations, tests done by
Hideo Aoki on ia64 show no regression. His test case was using hackbench
on a kernel where scheduler instrumentation (about 5 events in code
scheduler code) was added. See the "Tracepoints" patch header for
performance result detail.
Changelog :
- Change instrumentation location and parameter to match ftrace
instrumentation, previously done with kernel markers.
[ mingo@elte.hu: conflict resolutions ]
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Acked-by: 'Peter Zijlstra' <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-07-18 20:16:17 +04:00
|
|
|
trace_sched_kthread_stop(k);
|
2013-04-30 02:05:12 +04:00
|
|
|
|
|
|
|
get_task_struct(k);
|
|
|
|
kthread = to_live_kthread(k);
|
2012-07-16 14:42:36 +04:00
|
|
|
if (kthread) {
|
|
|
|
set_bit(KTHREAD_SHOULD_STOP, &kthread->flags);
|
2013-04-09 11:33:34 +04:00
|
|
|
__kthread_unpark(k, kthread);
|
2009-06-18 03:27:45 +04:00
|
|
|
wake_up_process(k);
|
|
|
|
wait_for_completion(&kthread->exited);
|
|
|
|
}
|
|
|
|
ret = k->exit_code;
|
2005-04-17 02:20:36 +04:00
|
|
|
put_task_struct(k);
|
tracing, sched: LTTng instrumentation - scheduler
Instrument the scheduler activity (sched_switch, migration, wakeups,
wait for a task, signal delivery) and process/thread
creation/destruction (fork, exit, kthread stop). Actually, kthread
creation is not instrumented in this patch because it is architecture
dependent. It allows to connect tracers such as ftrace which detects
scheduling latencies, good/bad scheduler decisions. Tools like LTTng can
export this scheduler information along with instrumentation of the rest
of the kernel activity to perform post-mortem analysis on the scheduler
activity.
About the performance impact of tracepoints (which is comparable to
markers), even without immediate values optimizations, tests done by
Hideo Aoki on ia64 show no regression. His test case was using hackbench
on a kernel where scheduler instrumentation (about 5 events in code
scheduler code) was added. See the "Tracepoints" patch header for
performance result detail.
Changelog :
- Change instrumentation location and parameter to match ftrace
instrumentation, previously done with kernel markers.
[ mingo@elte.hu: conflict resolutions ]
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
Acked-by: 'Peter Zijlstra' <peterz@infradead.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
2008-07-18 20:16:17 +04:00
|
|
|
|
2013-04-30 02:05:12 +04:00
|
|
|
trace_sched_kthread_stop_ret(ret);
|
2005-04-17 02:20:36 +04:00
|
|
|
return ret;
|
|
|
|
}
|
2006-07-14 11:24:05 +04:00
|
|
|
EXPORT_SYMBOL(kthread_stop);
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2007-07-31 11:39:16 +04:00
|
|
|
int kthreadd(void *unused)
|
2005-04-17 02:20:36 +04:00
|
|
|
{
|
2007-05-09 13:34:32 +04:00
|
|
|
struct task_struct *tsk = current;
|
2005-04-17 02:20:36 +04:00
|
|
|
|
2007-07-31 11:39:16 +04:00
|
|
|
/* Setup a clean context for our children to inherit. */
|
2007-05-09 13:34:32 +04:00
|
|
|
set_task_comm(tsk, "kthreadd");
|
2007-05-09 13:34:37 +04:00
|
|
|
ignore_signals(tsk);
|
2009-03-31 08:05:10 +04:00
|
|
|
set_cpus_allowed_ptr(tsk, cpu_all_mask);
|
2012-12-13 01:51:39 +04:00
|
|
|
set_mems_allowed(node_states[N_MEMORY]);
|
2007-05-09 13:34:32 +04:00
|
|
|
|
2011-11-23 21:28:17 +04:00
|
|
|
current->flags |= PF_NOFREEZE;
|
2007-05-09 13:34:32 +04:00
|
|
|
|
|
|
|
for (;;) {
|
|
|
|
set_current_state(TASK_INTERRUPTIBLE);
|
|
|
|
if (list_empty(&kthread_create_list))
|
|
|
|
schedule();
|
|
|
|
__set_current_state(TASK_RUNNING);
|
|
|
|
|
|
|
|
spin_lock(&kthread_create_lock);
|
|
|
|
while (!list_empty(&kthread_create_list)) {
|
|
|
|
struct kthread_create_info *create;
|
|
|
|
|
|
|
|
create = list_entry(kthread_create_list.next,
|
|
|
|
struct kthread_create_info, list);
|
|
|
|
list_del_init(&create->list);
|
|
|
|
spin_unlock(&kthread_create_lock);
|
|
|
|
|
|
|
|
create_kthread(create);
|
|
|
|
|
|
|
|
spin_lock(&kthread_create_lock);
|
|
|
|
}
|
|
|
|
spin_unlock(&kthread_create_lock);
|
|
|
|
}
|
|
|
|
|
|
|
|
return 0;
|
|
|
|
}
|
2010-06-29 12:07:09 +04:00
|
|
|
|
2010-12-22 12:27:53 +03:00
|
|
|
void __init_kthread_worker(struct kthread_worker *worker,
|
|
|
|
const char *name,
|
|
|
|
struct lock_class_key *key)
|
|
|
|
{
|
|
|
|
spin_lock_init(&worker->lock);
|
|
|
|
lockdep_set_class_and_name(&worker->lock, key, name);
|
|
|
|
INIT_LIST_HEAD(&worker->work_list);
|
|
|
|
worker->task = NULL;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(__init_kthread_worker);
|
|
|
|
|
2010-06-29 12:07:09 +04:00
|
|
|
/**
|
|
|
|
* kthread_worker_fn - kthread function to process kthread_worker
|
|
|
|
* @worker_ptr: pointer to initialized kthread_worker
|
|
|
|
*
|
|
|
|
* This function can be used as @threadfn to kthread_create() or
|
|
|
|
* kthread_run() with @worker_ptr argument pointing to an initialized
|
|
|
|
* kthread_worker. The started kthread will process work_list until
|
|
|
|
* the it is stopped with kthread_stop(). A kthread can also call
|
|
|
|
* this function directly after extra initialization.
|
|
|
|
*
|
|
|
|
* Different kthreads can be used for the same kthread_worker as long
|
|
|
|
* as there's only one kthread attached to it at any given time. A
|
|
|
|
* kthread_worker without an attached kthread simply collects queued
|
|
|
|
* kthread_works.
|
|
|
|
*/
|
|
|
|
int kthread_worker_fn(void *worker_ptr)
|
|
|
|
{
|
|
|
|
struct kthread_worker *worker = worker_ptr;
|
|
|
|
struct kthread_work *work;
|
|
|
|
|
|
|
|
WARN_ON(worker->task);
|
|
|
|
worker->task = current;
|
|
|
|
repeat:
|
|
|
|
set_current_state(TASK_INTERRUPTIBLE); /* mb paired w/ kthread_stop */
|
|
|
|
|
|
|
|
if (kthread_should_stop()) {
|
|
|
|
__set_current_state(TASK_RUNNING);
|
|
|
|
spin_lock_irq(&worker->lock);
|
|
|
|
worker->task = NULL;
|
|
|
|
spin_unlock_irq(&worker->lock);
|
|
|
|
return 0;
|
|
|
|
}
|
|
|
|
|
|
|
|
work = NULL;
|
|
|
|
spin_lock_irq(&worker->lock);
|
|
|
|
if (!list_empty(&worker->work_list)) {
|
|
|
|
work = list_first_entry(&worker->work_list,
|
|
|
|
struct kthread_work, node);
|
|
|
|
list_del_init(&work->node);
|
|
|
|
}
|
2012-07-20 00:52:53 +04:00
|
|
|
worker->current_work = work;
|
2010-06-29 12:07:09 +04:00
|
|
|
spin_unlock_irq(&worker->lock);
|
|
|
|
|
|
|
|
if (work) {
|
|
|
|
__set_current_state(TASK_RUNNING);
|
|
|
|
work->func(work);
|
|
|
|
} else if (!freezing(current))
|
|
|
|
schedule();
|
|
|
|
|
|
|
|
try_to_freeze();
|
|
|
|
goto repeat;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(kthread_worker_fn);
|
|
|
|
|
2012-07-20 00:52:53 +04:00
|
|
|
/* insert @work before @pos in @worker */
|
|
|
|
static void insert_kthread_work(struct kthread_worker *worker,
|
|
|
|
struct kthread_work *work,
|
|
|
|
struct list_head *pos)
|
|
|
|
{
|
|
|
|
lockdep_assert_held(&worker->lock);
|
|
|
|
|
|
|
|
list_add_tail(&work->node, pos);
|
2012-07-20 00:52:53 +04:00
|
|
|
work->worker = worker;
|
2014-07-26 08:03:59 +04:00
|
|
|
if (!worker->current_work && likely(worker->task))
|
2012-07-20 00:52:53 +04:00
|
|
|
wake_up_process(worker->task);
|
|
|
|
}
|
|
|
|
|
2010-06-29 12:07:09 +04:00
|
|
|
/**
|
|
|
|
* queue_kthread_work - queue a kthread_work
|
|
|
|
* @worker: target kthread_worker
|
|
|
|
* @work: kthread_work to queue
|
|
|
|
*
|
|
|
|
* Queue @work to work processor @task for async execution. @task
|
|
|
|
* must have been created with kthread_worker_create(). Returns %true
|
|
|
|
* if @work was successfully queued, %false if it was already pending.
|
|
|
|
*/
|
|
|
|
bool queue_kthread_work(struct kthread_worker *worker,
|
|
|
|
struct kthread_work *work)
|
|
|
|
{
|
|
|
|
bool ret = false;
|
|
|
|
unsigned long flags;
|
|
|
|
|
|
|
|
spin_lock_irqsave(&worker->lock, flags);
|
|
|
|
if (list_empty(&work->node)) {
|
2012-07-20 00:52:53 +04:00
|
|
|
insert_kthread_work(worker, work, &worker->work_list);
|
2010-06-29 12:07:09 +04:00
|
|
|
ret = true;
|
|
|
|
}
|
|
|
|
spin_unlock_irqrestore(&worker->lock, flags);
|
|
|
|
return ret;
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(queue_kthread_work);
|
|
|
|
|
2012-07-20 00:52:53 +04:00
|
|
|
struct kthread_flush_work {
|
|
|
|
struct kthread_work work;
|
|
|
|
struct completion done;
|
|
|
|
};
|
|
|
|
|
|
|
|
static void kthread_flush_work_fn(struct kthread_work *work)
|
|
|
|
{
|
|
|
|
struct kthread_flush_work *fwork =
|
|
|
|
container_of(work, struct kthread_flush_work, work);
|
|
|
|
complete(&fwork->done);
|
|
|
|
}
|
|
|
|
|
2010-06-29 12:07:09 +04:00
|
|
|
/**
|
|
|
|
* flush_kthread_work - flush a kthread_work
|
|
|
|
* @work: work to flush
|
|
|
|
*
|
|
|
|
* If @work is queued or executing, wait for it to finish execution.
|
|
|
|
*/
|
|
|
|
void flush_kthread_work(struct kthread_work *work)
|
|
|
|
{
|
2012-07-20 00:52:53 +04:00
|
|
|
struct kthread_flush_work fwork = {
|
|
|
|
KTHREAD_WORK_INIT(fwork.work, kthread_flush_work_fn),
|
|
|
|
COMPLETION_INITIALIZER_ONSTACK(fwork.done),
|
|
|
|
};
|
|
|
|
struct kthread_worker *worker;
|
|
|
|
bool noop = false;
|
|
|
|
|
|
|
|
retry:
|
|
|
|
worker = work->worker;
|
|
|
|
if (!worker)
|
|
|
|
return;
|
2010-06-29 12:07:09 +04:00
|
|
|
|
2012-07-20 00:52:53 +04:00
|
|
|
spin_lock_irq(&worker->lock);
|
|
|
|
if (work->worker != worker) {
|
|
|
|
spin_unlock_irq(&worker->lock);
|
|
|
|
goto retry;
|
|
|
|
}
|
2010-06-29 12:07:09 +04:00
|
|
|
|
2012-07-20 00:52:53 +04:00
|
|
|
if (!list_empty(&work->node))
|
|
|
|
insert_kthread_work(worker, &fwork.work, work->node.next);
|
|
|
|
else if (worker->current_work == work)
|
|
|
|
insert_kthread_work(worker, &fwork.work, worker->work_list.next);
|
|
|
|
else
|
|
|
|
noop = true;
|
2010-06-29 12:07:09 +04:00
|
|
|
|
2012-07-20 00:52:53 +04:00
|
|
|
spin_unlock_irq(&worker->lock);
|
2010-06-29 12:07:09 +04:00
|
|
|
|
2012-07-20 00:52:53 +04:00
|
|
|
if (!noop)
|
|
|
|
wait_for_completion(&fwork.done);
|
2010-06-29 12:07:09 +04:00
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(flush_kthread_work);
|
|
|
|
|
|
|
|
/**
|
|
|
|
* flush_kthread_worker - flush all current works on a kthread_worker
|
|
|
|
* @worker: worker to flush
|
|
|
|
*
|
|
|
|
* Wait until all currently executing or pending works on @worker are
|
|
|
|
* finished.
|
|
|
|
*/
|
|
|
|
void flush_kthread_worker(struct kthread_worker *worker)
|
|
|
|
{
|
|
|
|
struct kthread_flush_work fwork = {
|
|
|
|
KTHREAD_WORK_INIT(fwork.work, kthread_flush_work_fn),
|
|
|
|
COMPLETION_INITIALIZER_ONSTACK(fwork.done),
|
|
|
|
};
|
|
|
|
|
|
|
|
queue_kthread_work(worker, &fwork.work);
|
|
|
|
wait_for_completion(&fwork.done);
|
|
|
|
}
|
|
|
|
EXPORT_SYMBOL_GPL(flush_kthread_worker);
|