Skip to content

L
linux
Commit 72ec7029 authored by Tejun Heo's avatar Tejun Heo
Browse files
Options

cgroup: make task iterators deal with cgroup_subsys_state instead of cgroup 

cgroup is in the process of converting to css (cgroup_subsys_state)
from cgroup as the principal subsystem interface handle.  This is
mostly to prepare for the unified hierarchy support where css's will
be created and destroyed dynamically but also helps cleaning up
subsystem implementations as css is usually what they are interested
in anyway.

This patch converts task iterators to deal with css instead of cgroup.
Note that under unified hierarchy, different sets of tasks will be
considered belonging to a given cgroup depending on the subsystem in
question and making the iterators deal with css instead cgroup
provides them with enough information about the iteration.

While at it, fix several function comment formats in cpuset.c.

This patch doesn't introduce any behavior differences.
Signed-off-by: default avatarTejun Heo <tj@kernel.org>
Acked-by: default avatarLi Zefan <lizefan@huawei.com>
Acked-by: default avatarMichal Hocko <mhocko@suse.cz>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Balbir Singh <bsingharora@gmail.com>
Cc: Matt Helsley <matthltc@us.ibm.com>
parent e535837b
Show whitespace changes
Inline Side-by-side
Showing with 104 additions and 107 deletions
include/linux/cgroup.h
View file @ 72ec7029
......@@ -880,18 +880,19 @@ css_next_descendant_post(struct cgroup_subsys_state *pos,
for ((pos) = css_next_descendant_post(NULL, (css)); (pos); \
(pos) = css_next_descendant_post((pos), (css)))
/* A cgroup_task_iter should be treated as an opaque object */
struct cgroup_task_iter {
struct cgroup *origin_cgrp;
/* A css_task_iter should be treated as an opaque object */
struct css_task_iter {
struct cgroup_subsys_state *origin_css;
struct list_head *cset_link;
struct list_head *task;
};
void cgroup_task_iter_start(struct cgroup *cgrp, struct cgroup_task_iter *it);
struct task_struct *cgroup_task_iter_next(struct cgroup_task_iter *it);
void cgroup_task_iter_end(struct cgroup_task_iter *it);
void css_task_iter_start(struct cgroup_subsys_state *css,
struct css_task_iter *it);
struct task_struct *css_task_iter_next(struct css_task_iter *it);
void css_task_iter_end(struct css_task_iter *it);
int cgroup_scan_tasks(struct cgroup *cgrp,
int css_scan_tasks(struct cgroup_subsys_state *css,
bool (*test)(struct task_struct *, void *),
void (*process)(struct task_struct *, void *),
void *data, struct ptr_heap *heap);
......
kernel/cgroup.c
View file @ 72ec7029
......@@ -370,7 +370,7 @@ static int cgroup_init_idr(struct cgroup_subsys *ss,
/*
* css_set_lock protects the list of css_set objects, and the chain of
* tasks off each css_set. Nests outside task->alloc_lock due to
* cgroup_task_iter_start().
* css_task_iter_start().
*/
static DEFINE_RWLOCK(css_set_lock);
static int css_set_count;
......@@ -398,9 +398,9 @@ static unsigned long css_set_hash(struct cgroup_subsys_state *css[])
/*
* We don't maintain the lists running through each css_set to its task
* until after the first call to cgroup_task_iter_start(). This reduces
* the fork()/exit() overhead for people who have cgroups compiled into
* their kernel but not actually in use.
* until after the first call to css_task_iter_start(). This reduces the
* fork()/exit() overhead for people who have cgroups compiled into their
* kernel but not actually in use.
*/
static int use_task_css_set_links __read_mostly;
......@@ -2989,7 +2989,7 @@ int cgroup_task_count(const struct cgroup *cgrp)
* To reduce the fork() overhead for systems that are not actually using
* their cgroups capability, we don't maintain the lists running through
* each css_set to its tasks until we see the list actually used - in other
* words after the first call to cgroup_task_iter_start().
* words after the first call to css_task_iter_start().
*/
static void cgroup_enable_task_cg_lists(void)
{
......@@ -3204,12 +3204,12 @@ css_next_descendant_post(struct cgroup_subsys_state *pos,
EXPORT_SYMBOL_GPL(css_next_descendant_post);
/**
* cgroup_advance_task_iter - advance a task itererator to the next css_set
* css_advance_task_iter - advance a task itererator to the next css_set
* @it: the iterator to advance
*
* Advance @it to the next css_set to walk.
*/
static void cgroup_advance_task_iter(struct cgroup_task_iter *it)
static void css_advance_task_iter(struct css_task_iter *it)
{
struct list_head *l = it->cset_link;
struct cgrp_cset_link *link;
......@@ -3218,7 +3218,7 @@ static void cgroup_advance_task_iter(struct cgroup_task_iter *it)
/* Advance to the next non-empty css_set */
do {
l = l->next;
if (l == &it->origin_cgrp->cset_links) {
if (l == &it->origin_css->cgroup->cset_links) {
it->cset_link = NULL;
return;
}
......@@ -3230,47 +3230,48 @@ static void cgroup_advance_task_iter(struct cgroup_task_iter *it)
}
/**
* cgroup_task_iter_start - initiate task iteration
* @cgrp: the cgroup to walk tasks of
* css_task_iter_start - initiate task iteration
* @css: the css to walk tasks of
* @it: the task iterator to use
*
* Initiate iteration through the tasks of @cgrp. The caller can call
* cgroup_task_iter_next() to walk through the tasks until the function
* returns NULL. On completion of iteration, cgroup_task_iter_end() must
* be called.
* Initiate iteration through the tasks of @css. The caller can call
* css_task_iter_next() to walk through the tasks until the function
* returns NULL. On completion of iteration, css_task_iter_end() must be
* called.
*
* Note that this function acquires a lock which is released when the
* iteration finishes. The caller can't sleep while iteration is in
* progress.
*/
void cgroup_task_iter_start(struct cgroup *cgrp, struct cgroup_task_iter *it)
void css_task_iter_start(struct cgroup_subsys_state *css,
struct css_task_iter *it)
__acquires(css_set_lock)
{
/*
* The first time anyone tries to iterate across a cgroup,
* we need to enable the list linking each css_set to its
* tasks, and fix up all existing tasks.
* The first time anyone tries to iterate across a css, we need to
* enable the list linking each css_set to its tasks, and fix up
* all existing tasks.
*/
if (!use_task_css_set_links)
cgroup_enable_task_cg_lists();
read_lock(&css_set_lock);
it->origin_cgrp = cgrp;
it->cset_link = &cgrp->cset_links;
it->origin_css = css;
it->cset_link = &css->cgroup->cset_links;
cgroup_advance_task_iter(it);
css_advance_task_iter(it);
}
/**
* cgroup_task_iter_next - return the next task for the iterator
* css_task_iter_next - return the next task for the iterator
* @it: the task iterator being iterated
*
* The "next" function for task iteration. @it should have been
* initialized via cgroup_task_iter_start(). Returns NULL when the
* iteration reaches the end.
* initialized via css_task_iter_start(). Returns NULL when the iteration
* reaches the end.
*/
struct task_struct *cgroup_task_iter_next(struct cgroup_task_iter *it)
struct task_struct *css_task_iter_next(struct css_task_iter *it)
{
struct task_struct *res;
struct list_head *l = it->task;
......@@ -3288,7 +3289,7 @@ struct task_struct *cgroup_task_iter_next(struct cgroup_task_iter *it)
* We reached the end of this task list - move on to the
* next cgrp_cset_link.
*/
cgroup_advance_task_iter(it);
css_advance_task_iter(it);
} else {
it->task = l;
}
......@@ -3296,12 +3297,12 @@ struct task_struct *cgroup_task_iter_next(struct cgroup_task_iter *it)
}
/**
* cgroup_task_iter_end - finish task iteration
* css_task_iter_end - finish task iteration
* @it: the task iterator to finish
*
* Finish task iteration started by cgroup_task_iter_start().
* Finish task iteration started by css_task_iter_start().
*/
void cgroup_task_iter_end(struct cgroup_task_iter *it)
void css_task_iter_end(struct css_task_iter *it)
__releases(css_set_lock)
{
read_unlock(&css_set_lock);
......@@ -3342,24 +3343,24 @@ static inline int started_after(void *p1, void *p2)
}
/**
* cgroup_scan_tasks - iterate though all the tasks in a cgroup
* @cgrp: the cgroup to iterate tasks of
* css_scan_tasks - iterate though all the tasks in a css
* @css: the css to iterate tasks of
* @test: optional test callback
* @process: process callback
* @data: data passed to @test and @process
* @heap: optional pre-allocated heap used for task iteration
*
* Iterate through all the tasks in a cgroup, calling @test for each, and
* if it returns %true, call @process for it also.
* Iterate through all the tasks in @css, calling @test for each, and if it
* returns %true, call @process for it also.
*
* @test may be NULL, meaning always true (select all tasks), which
* effectively duplicates cgroup_task_iter_{start,next,end}() but does not
* effectively duplicates css_task_iter_{start,next,end}() but does not
* lock css_set_lock for the call to @process.
*
* It is guaranteed that @process will act on every task that is a member
* of @cgrp for the duration of this call. This function may or may not
* call @process for tasks that exit or move to a different cgroup during
* the call, or are forked or move into the cgroup during the call.
* of @css for the duration of this call. This function may or may not
* call @process for tasks that exit or move to a different css during the
* call, or are forked or move into the css during the call.
*
* Note that @test may be called with locks held, and may in some
* situations be called multiple times for the same task, so it should be
......@@ -3370,13 +3371,13 @@ static inline int started_after(void *p1, void *p2)
* temporary heap will be used (allocation of which may cause this function
* to fail).
*/
int cgroup_scan_tasks(struct cgroup *cgrp,
int css_scan_tasks(struct cgroup_subsys_state *css,
bool (*test)(struct task_struct *, void *),
void (*process)(struct task_struct *, void *),
void *data, struct ptr_heap *heap)
{
int retval, i;
struct cgroup_task_iter it;
struct css_task_iter it;
struct task_struct *p, *dropped;
/* Never dereference latest_task, since it's not refcounted */
struct task_struct *latest_task = NULL;
......@@ -3397,7 +3398,7 @@ int cgroup_scan_tasks(struct cgroup *cgrp,
again:
/*
* Scan tasks in the cgroup, using the @test callback to determine
* Scan tasks in the css, using the @test callback to determine
* which are of interest, and invoking @process callback on the
* ones which need an update. Since we don't want to hold any
* locks during the task updates, gather tasks to be processed in a
......@@ -3408,8 +3409,8 @@ int cgroup_scan_tasks(struct cgroup *cgrp,
* guarantees forward progress and that we don't miss any tasks.
*/
heap->size = 0;
cgroup_task_iter_start(cgrp, &it);
while ((p = cgroup_task_iter_next(&it))) {
css_task_iter_start(css, &it);
while ((p = css_task_iter_next(&it))) {
/*
* Only affect tasks that qualify per the caller's callback,
* if he provided one
......@@ -3442,7 +3443,7 @@ int cgroup_scan_tasks(struct cgroup *cgrp,
* the heap and wasn't inserted
*/
}
cgroup_task_iter_end(&it);
css_task_iter_end(&it);
if (heap->size) {
for (i = 0; i < heap->size; i++) {
......@@ -3485,7 +3486,8 @@ static void cgroup_transfer_one_task(struct task_struct *task, void *data)
*/
int cgroup_transfer_tasks(struct cgroup *to, struct cgroup *from)
{
return cgroup_scan_tasks(from, NULL, cgroup_transfer_one_task, to, NULL);
return css_scan_tasks(&from->dummy_css, NULL, cgroup_transfer_one_task,
to, NULL);
}
/*
......@@ -3639,7 +3641,7 @@ static int pidlist_array_load(struct cgroup *cgrp, enum cgroup_filetype type,
pid_t *array;
int length;
int pid, n = 0; /* used for populating the array */
struct cgroup_task_iter it;
struct css_task_iter it;
struct task_struct *tsk;
struct cgroup_pidlist *l;
......@@ -3654,8 +3656,8 @@ static int pidlist_array_load(struct cgroup *cgrp, enum cgroup_filetype type,
if (!array)
return -ENOMEM;
/* now, populate the array */
cgroup_task_iter_start(cgrp, &it);
while ((tsk = cgroup_task_iter_next(&it))) {
css_task_iter_start(&cgrp->dummy_css, &it);
while ((tsk = css_task_iter_next(&it))) {
if (unlikely(n == length))
break;
/* get tgid or pid for procs or tasks file respectively */
......@@ -3666,7 +3668,7 @@ static int pidlist_array_load(struct cgroup *cgrp, enum cgroup_filetype type,
if (pid > 0) /* make sure to only use valid results */
array[n++] = pid;
}
cgroup_task_iter_end(&it);
css_task_iter_end(&it);
length = n;
/* now sort & (if procs) strip out duplicates */
sort(array, length, sizeof(pid_t), cmppid, NULL);
......@@ -3700,7 +3702,7 @@ int cgroupstats_build(struct cgroupstats *stats, struct dentry *dentry)
{
int ret = -EINVAL;
struct cgroup *cgrp;
struct cgroup_task_iter it;
struct css_task_iter it;
struct task_struct *tsk;
/*
......@@ -3714,8 +3716,8 @@ int cgroupstats_build(struct cgroupstats *stats, struct dentry *dentry)
ret = 0;
cgrp = dentry->d_fsdata;
cgroup_task_iter_start(cgrp, &it);
while ((tsk = cgroup_task_iter_next(&it))) {
css_task_iter_start(&cgrp->dummy_css, &it);
while ((tsk = css_task_iter_next(&it))) {
switch (tsk->state) {
case TASK_RUNNING:
stats->nr_running++;
......@@ -3735,7 +3737,7 @@ int cgroupstats_build(struct cgroupstats *stats, struct dentry *dentry)
break;
}
}
cgroup_task_iter_end(&it);
css_task_iter_end(&it);
err:
return ret;
......
kernel/cgroup_freezer.c
View file @ 72ec7029
......@@ -258,7 +258,7 @@ static void update_if_frozen(struct cgroup_subsys_state *css)
{
struct freezer *freezer = css_freezer(css);
struct cgroup_subsys_state *pos;
struct cgroup_task_iter it;
struct css_task_iter it;
struct task_struct *task;
WARN_ON_ONCE(!rcu_read_lock_held());
......@@ -279,9 +279,9 @@ static void update_if_frozen(struct cgroup_subsys_state *css)
}
/* are all tasks frozen? */
cgroup_task_iter_start(css->cgroup, &it);
css_task_iter_start(css, &it);
while ((task = cgroup_task_iter_next(&it))) {
while ((task = css_task_iter_next(&it))) {
if (freezing(task)) {
/*
* freezer_should_skip() indicates that the task
......@@ -296,7 +296,7 @@ static void update_if_frozen(struct cgroup_subsys_state *css)
freezer->state |= CGROUP_FROZEN;
out_iter_end:
cgroup_task_iter_end(&it);
css_task_iter_end(&it);
out_unlock:
spin_unlock_irq(&freezer->lock);
}
......@@ -322,26 +322,24 @@ static int freezer_read(struct cgroup_subsys_state *css, struct cftype *cft,
static void freeze_cgroup(struct freezer *freezer)
{
struct cgroup *cgroup = freezer->css.cgroup;
struct cgroup_task_iter it;
struct css_task_iter it;
struct task_struct *task;
cgroup_task_iter_start(cgroup, &it);
while ((task = cgroup_task_iter_next(&it)))
css_task_iter_start(&freezer->css, &it);
while ((task = css_task_iter_next(&it)))
freeze_task(task);
cgroup_task_iter_end(&it);
css_task_iter_end(&it);
}
static void unfreeze_cgroup(struct freezer *freezer)
{
struct cgroup *cgroup = freezer->css.cgroup;
struct cgroup_task_iter it;
struct css_task_iter it;
struct task_struct *task;
cgroup_task_iter_start(cgroup, &it);
while ((task = cgroup_task_iter_next(&it)))
css_task_iter_start(&freezer->css, &it);
while ((task = css_task_iter_next(&it)))
__thaw_task(task);
cgroup_task_iter_end(&it);
css_task_iter_end(&it);
}
/**
......
kernel/cpuset.c
View file @ 72ec7029
......@@ -832,8 +832,8 @@ static struct cpuset *effective_nodemask_cpuset(struct cpuset *cs)
* @tsk: task to test
* @data: cpuset to @tsk belongs to
*
* Called by cgroup_scan_tasks() for each task in a cgroup whose
* cpus_allowed mask needs to be changed.
* Called by css_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 cpuset_mutex at this point.
......@@ -849,27 +849,26 @@ static void cpuset_change_cpumask(struct task_struct *tsk, void *data)
/**
* 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()
* @heap: if NULL, defer allocating heap memory to css_scan_tasks()
*
* Called with cpuset_mutex held
*
* The cgroup_scan_tasks() function will scan all the tasks in a cgroup,
* The css_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
* No return value. It's guaranteed that css_scan_tasks() always returns 0
* if @heap != NULL.
*/
static void update_tasks_cpumask(struct cpuset *cs, struct ptr_heap *heap)
{
cgroup_scan_tasks(cs->css.cgroup, NULL, cpuset_change_cpumask, cs,
heap);
css_scan_tasks(&cs->css, NULL, cpuset_change_cpumask, cs, heap);
}
/*
* update_tasks_cpumask_hier - Update the cpumasks of tasks in the hierarchy.
* @root_cs: the root cpuset of the hierarchy
* @update_root: update root cpuset or not?
* @heap: the heap used by cgroup_scan_tasks()
* @heap: the heap used by css_scan_tasks()
*
* This will update cpumasks of tasks in @root_cs and all other empty cpusets
* which take on cpumask of @root_cs.
......@@ -1082,11 +1081,10 @@ 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
* @heap: if NULL, defer allocating heap memory to cgroup_scan_tasks()
* @heap: if NULL, defer allocating heap memory to css_scan_tasks()
*
* Called with cpuset_mutex held
* No return value. It's guaranteed that cgroup_scan_tasks() always returns 0
* if @heap != NULL.
* Called with cpuset_mutex held. No return value. It's guaranteed that
* css_scan_tasks() always returns 0 if @heap != NULL.
*/
static void update_tasks_nodemask(struct cpuset *cs, struct ptr_heap *heap)
{
......@@ -1109,8 +1107,7 @@ static void update_tasks_nodemask(struct cpuset *cs, struct ptr_heap *heap)
* 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(cs->css.cgroup, NULL, cpuset_change_nodemask, &arg,
heap);
css_scan_tasks(&cs->css, NULL, cpuset_change_nodemask, &arg, heap);
/*
* All the tasks' nodemasks have been updated, update
......@@ -1126,7 +1123,7 @@ static void update_tasks_nodemask(struct cpuset *cs, struct ptr_heap *heap)
* update_tasks_nodemask_hier - Update the nodemasks of tasks in the hierarchy.
* @cs: the root cpuset of the hierarchy
* @update_root: update the root cpuset or not?
* @heap: the heap used by cgroup_scan_tasks()
* @heap: the heap used by css_scan_tasks()
*
* This will update nodemasks of tasks in @root_cs and all other empty cpusets
* which take on nodemask of @root_cs.
......@@ -1254,12 +1251,12 @@ static int update_relax_domain_level(struct cpuset *cs, s64 val)
return 0;
}
/*
/**
* cpuset_change_flag - make a task's spread flags the same as its cpuset's
* @tsk: task to be updated
* @data: cpuset to @tsk belongs to
*
* Called by cgroup_scan_tasks() for each task in a cgroup.
* Called by css_scan_tasks() for each task in a cgroup.
*
* We don't need to re-check for the cgroup/cpuset membership, since we're
* holding cpuset_mutex at this point.
......@@ -1271,22 +1268,22 @@ static void cpuset_change_flag(struct task_struct *tsk, void *data)
cpuset_update_task_spread_flag(cs, 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()
* @heap: if NULL, defer allocating heap memory to css_scan_tasks()
*
* Called with cpuset_mutex held
*
* The cgroup_scan_tasks() function will scan all the tasks in a cgroup,
* The css_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
* No return value. It's guaranteed that css_scan_tasks() always returns 0
* if @heap != NULL.
*/
static void update_tasks_flags(struct cpuset *cs, struct ptr_heap *heap)
{
cgroup_scan_tasks(cs->css.cgroup, NULL, cpuset_change_flag, cs, heap);
css_scan_tasks(&cs->css, NULL, cpuset_change_flag, cs, heap);
}
/*
......
mm/memcontrol.c
View file @ 72ec7029
......@@ -1799,12 +1799,11 @@ static void mem_cgroup_out_of_memory(struct mem_cgroup *memcg, gfp_t gfp_mask,
check_panic_on_oom(CONSTRAINT_MEMCG, gfp_mask, order, NULL);
totalpages = mem_cgroup_get_limit(memcg) >> PAGE_SHIFT ? : 1;
for_each_mem_cgroup_tree(iter, memcg) {
struct cgroup *cgroup = iter->css.cgroup;
struct cgroup_task_iter it;
struct css_task_iter it;
struct task_struct *task;
cgroup_task_iter_start(cgroup, &it);
while ((task = cgroup_task_iter_next(&it))) {
css_task_iter_start(&iter->css, &it);
while ((task = css_task_iter_next(&it))) {
switch (oom_scan_process_thread(task, totalpages, NULL,
false)) {
case OOM_SCAN_SELECT:
......@@ -1817,7 +1816,7 @@ static void mem_cgroup_out_of_memory(struct mem_cgroup *memcg, gfp_t gfp_mask,
case OOM_SCAN_CONTINUE:
continue;
case OOM_SCAN_ABORT:
cgroup_task_iter_end(&it);
css_task_iter_end(&it);
mem_cgroup_iter_break(memcg, iter);
if (chosen)
put_task_struct(chosen);
......@@ -1834,7 +1833,7 @@ static void mem_cgroup_out_of_memory(struct mem_cgroup *memcg, gfp_t gfp_mask,
get_task_struct(chosen);
}
}
cgroup_task_iter_end(&it);
css_task_iter_end(&it);
}
if (!chosen)
......
Markdown is supported
0% or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment
GitLab Nexedi Edition | About GitLab | About Nexedi | 沪ICP备2021021310号-2 | 沪ICP备2021021310号-7