From: Waiman Long <longman-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
To: Tejun Heo <tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>,
Zefan Li <lizefan.x-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org>,
Johannes Weiner <hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org>,
Christian Brauner
<brauner-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
Cc: cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
"Juri Lelli" <juri.lelli-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
"Dietmar Eggemann"
<dietmar.eggemann-5wv7dgnIgG8@public.gmane.org>,
"Michal Koutný" <mkoutny-IBi9RG/b67k@public.gmane.org>,
"Giuseppe Scrivano"
<gscrivan-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
"Waiman Long" <longman-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
Subject: [PATCH v4 2/5] cgroup/cpuset: Make cpuset_fork() handle CLONE_INTO_CGROUP properly
Date: Tue, 11 Apr 2023 09:35:58 -0400 [thread overview]
Message-ID: <20230411133601.2969636-3-longman@redhat.com> (raw)
In-Reply-To: <20230411133601.2969636-1-longman-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
By default, the clone(2) syscall spawn a child process into the same
cgroup as its parent. With the use of the CLONE_INTO_CGROUP flag
introduced by commit ef2c41cf38a7 ("clone3: allow spawning processes
into cgroups"), the child will be spawned into a different cgroup which
is somewhat similar to writing the child's tid into "cgroup.threads".
The current cpuset_fork() method does not properly handle the
CLONE_INTO_CGROUP case where the cpuset of the child may be different
from that of its parent. Update the cpuset_fork() method to treat the
CLONE_INTO_CGROUP case similar to cpuset_attach().
Since the newly cloned task has not been running yet, its actual
memory usage isn't known. So it is not necessary to make change to mm
in cpuset_fork().
Fixes: ef2c41cf38a7 ("clone3: allow spawning processes into cgroups")
Reported-by: Giuseppe Scrivano <gscrivan-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
Signed-off-by: Waiman Long <longman-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
---
kernel/cgroup/cpuset.c | 62 ++++++++++++++++++++++++++++--------------
1 file changed, 42 insertions(+), 20 deletions(-)
diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c
index 066689a7dcc3..e954d5abb784 100644
--- a/kernel/cgroup/cpuset.c
+++ b/kernel/cgroup/cpuset.c
@@ -2520,16 +2520,33 @@ static void cpuset_cancel_attach(struct cgroup_taskset *tset)
}
/*
- * Protected by cpuset_rwsem. cpus_attach is used only by cpuset_attach()
+ * Protected by cpuset_rwsem. cpus_attach is used only by cpuset_attach_task()
* but we can't allocate it dynamically there. Define it global and
* allocate from cpuset_init().
*/
static cpumask_var_t cpus_attach;
+static nodemask_t cpuset_attach_nodemask_to;
+
+static void cpuset_attach_task(struct cpuset *cs, struct task_struct *task)
+{
+ percpu_rwsem_assert_held(&cpuset_rwsem);
+
+ if (cs != &top_cpuset)
+ guarantee_online_cpus(task, cpus_attach);
+ else
+ cpumask_copy(cpus_attach, task_cpu_possible_mask(task));
+ /*
+ * can_attach beforehand should guarantee that this doesn't
+ * fail. TODO: have a better way to handle failure here
+ */
+ WARN_ON_ONCE(set_cpus_allowed_ptr(task, cpus_attach));
+
+ cpuset_change_task_nodemask(task, &cpuset_attach_nodemask_to);
+ cpuset_update_task_spread_flags(cs, task);
+}
static void cpuset_attach(struct cgroup_taskset *tset)
{
- /* static buf protected by cpuset_rwsem */
- static nodemask_t cpuset_attach_nodemask_to;
struct task_struct *task;
struct task_struct *leader;
struct cgroup_subsys_state *css;
@@ -2560,20 +2577,8 @@ static void cpuset_attach(struct cgroup_taskset *tset)
guarantee_online_mems(cs, &cpuset_attach_nodemask_to);
- cgroup_taskset_for_each(task, css, tset) {
- if (cs != &top_cpuset)
- guarantee_online_cpus(task, cpus_attach);
- else
- cpumask_copy(cpus_attach, task_cpu_possible_mask(task));
- /*
- * can_attach beforehand should guarantee that this doesn't
- * fail. TODO: have a better way to handle failure here
- */
- WARN_ON_ONCE(set_cpus_allowed_ptr(task, cpus_attach));
-
- cpuset_change_task_nodemask(task, &cpuset_attach_nodemask_to);
- cpuset_update_task_spread_flags(cs, task);
- }
+ cgroup_taskset_for_each(task, css, tset)
+ cpuset_attach_task(cs, task);
/*
* Change mm for all threadgroup leaders. This is expensive and may
@@ -3271,11 +3276,28 @@ static void cpuset_bind(struct cgroup_subsys_state *root_css)
*/
static void cpuset_fork(struct task_struct *task)
{
- if (task_css_is_root(task, cpuset_cgrp_id))
+ struct cpuset *cs;
+ bool same_cs;
+
+ rcu_read_lock();
+ cs = task_cs(task);
+ same_cs = (cs == task_cs(current));
+ rcu_read_unlock();
+
+ if (same_cs) {
+ if (cs == &top_cpuset)
+ return;
+
+ set_cpus_allowed_ptr(task, current->cpus_ptr);
+ task->mems_allowed = current->mems_allowed;
return;
+ }
- set_cpus_allowed_ptr(task, current->cpus_ptr);
- task->mems_allowed = current->mems_allowed;
+ /* CLONE_INTO_CGROUP */
+ percpu_down_write(&cpuset_rwsem);
+ guarantee_online_mems(cs, &cpuset_attach_nodemask_to);
+ cpuset_attach_task(cs, task);
+ percpu_up_write(&cpuset_rwsem);
}
struct cgroup_subsys cpuset_cgrp_subsys = {
--
2.31.1
WARNING: multiple messages have this Message-ID (diff)
From: Waiman Long <longman@redhat.com>
To: Tejun Heo <tj@kernel.org>, Zefan Li <lizefan.x@bytedance.com>,
Johannes Weiner <hannes@cmpxchg.org>,
Christian Brauner <brauner@kernel.org>
Cc: cgroups@vger.kernel.org, linux-kernel@vger.kernel.org,
"Juri Lelli" <juri.lelli@redhat.com>,
"Dietmar Eggemann" <dietmar.eggemann@arm.com>,
"Michal Koutný" <mkoutny@suse.com>,
"Giuseppe Scrivano" <gscrivan@redhat.com>,
"Waiman Long" <longman@redhat.com>
Subject: [PATCH v4 2/5] cgroup/cpuset: Make cpuset_fork() handle CLONE_INTO_CGROUP properly
Date: Tue, 11 Apr 2023 09:35:58 -0400 [thread overview]
Message-ID: <20230411133601.2969636-3-longman@redhat.com> (raw)
In-Reply-To: <20230411133601.2969636-1-longman@redhat.com>
By default, the clone(2) syscall spawn a child process into the same
cgroup as its parent. With the use of the CLONE_INTO_CGROUP flag
introduced by commit ef2c41cf38a7 ("clone3: allow spawning processes
into cgroups"), the child will be spawned into a different cgroup which
is somewhat similar to writing the child's tid into "cgroup.threads".
The current cpuset_fork() method does not properly handle the
CLONE_INTO_CGROUP case where the cpuset of the child may be different
from that of its parent. Update the cpuset_fork() method to treat the
CLONE_INTO_CGROUP case similar to cpuset_attach().
Since the newly cloned task has not been running yet, its actual
memory usage isn't known. So it is not necessary to make change to mm
in cpuset_fork().
Fixes: ef2c41cf38a7 ("clone3: allow spawning processes into cgroups")
Reported-by: Giuseppe Scrivano <gscrivan@redhat.com>
Signed-off-by: Waiman Long <longman@redhat.com>
---
kernel/cgroup/cpuset.c | 62 ++++++++++++++++++++++++++++--------------
1 file changed, 42 insertions(+), 20 deletions(-)
diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c
index 066689a7dcc3..e954d5abb784 100644
--- a/kernel/cgroup/cpuset.c
+++ b/kernel/cgroup/cpuset.c
@@ -2520,16 +2520,33 @@ static void cpuset_cancel_attach(struct cgroup_taskset *tset)
}
/*
- * Protected by cpuset_rwsem. cpus_attach is used only by cpuset_attach()
+ * Protected by cpuset_rwsem. cpus_attach is used only by cpuset_attach_task()
* but we can't allocate it dynamically there. Define it global and
* allocate from cpuset_init().
*/
static cpumask_var_t cpus_attach;
+static nodemask_t cpuset_attach_nodemask_to;
+
+static void cpuset_attach_task(struct cpuset *cs, struct task_struct *task)
+{
+ percpu_rwsem_assert_held(&cpuset_rwsem);
+
+ if (cs != &top_cpuset)
+ guarantee_online_cpus(task, cpus_attach);
+ else
+ cpumask_copy(cpus_attach, task_cpu_possible_mask(task));
+ /*
+ * can_attach beforehand should guarantee that this doesn't
+ * fail. TODO: have a better way to handle failure here
+ */
+ WARN_ON_ONCE(set_cpus_allowed_ptr(task, cpus_attach));
+
+ cpuset_change_task_nodemask(task, &cpuset_attach_nodemask_to);
+ cpuset_update_task_spread_flags(cs, task);
+}
static void cpuset_attach(struct cgroup_taskset *tset)
{
- /* static buf protected by cpuset_rwsem */
- static nodemask_t cpuset_attach_nodemask_to;
struct task_struct *task;
struct task_struct *leader;
struct cgroup_subsys_state *css;
@@ -2560,20 +2577,8 @@ static void cpuset_attach(struct cgroup_taskset *tset)
guarantee_online_mems(cs, &cpuset_attach_nodemask_to);
- cgroup_taskset_for_each(task, css, tset) {
- if (cs != &top_cpuset)
- guarantee_online_cpus(task, cpus_attach);
- else
- cpumask_copy(cpus_attach, task_cpu_possible_mask(task));
- /*
- * can_attach beforehand should guarantee that this doesn't
- * fail. TODO: have a better way to handle failure here
- */
- WARN_ON_ONCE(set_cpus_allowed_ptr(task, cpus_attach));
-
- cpuset_change_task_nodemask(task, &cpuset_attach_nodemask_to);
- cpuset_update_task_spread_flags(cs, task);
- }
+ cgroup_taskset_for_each(task, css, tset)
+ cpuset_attach_task(cs, task);
/*
* Change mm for all threadgroup leaders. This is expensive and may
@@ -3271,11 +3276,28 @@ static void cpuset_bind(struct cgroup_subsys_state *root_css)
*/
static void cpuset_fork(struct task_struct *task)
{
- if (task_css_is_root(task, cpuset_cgrp_id))
+ struct cpuset *cs;
+ bool same_cs;
+
+ rcu_read_lock();
+ cs = task_cs(task);
+ same_cs = (cs == task_cs(current));
+ rcu_read_unlock();
+
+ if (same_cs) {
+ if (cs == &top_cpuset)
+ return;
+
+ set_cpus_allowed_ptr(task, current->cpus_ptr);
+ task->mems_allowed = current->mems_allowed;
return;
+ }
- set_cpus_allowed_ptr(task, current->cpus_ptr);
- task->mems_allowed = current->mems_allowed;
+ /* CLONE_INTO_CGROUP */
+ percpu_down_write(&cpuset_rwsem);
+ guarantee_online_mems(cs, &cpuset_attach_nodemask_to);
+ cpuset_attach_task(cs, task);
+ percpu_up_write(&cpuset_rwsem);
}
struct cgroup_subsys cpuset_cgrp_subsys = {
--
2.31.1
next prev parent reply other threads:[~2023-04-11 13:35 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-04-11 13:35 [PATCH v4 0/5] cgroup/cpuset: Fix CLONE_INTO_CGROUP problem & other issues Waiman Long
2023-04-11 13:35 ` Waiman Long
[not found] ` <20230411133601.2969636-1-longman-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2023-04-11 13:35 ` [PATCH v4 1/5] cgroup/cpuset: Wake up cpuset_attach_wq tasks in cpuset_cancel_attach() Waiman Long
2023-04-11 13:35 ` Waiman Long
2023-04-11 13:35 ` Waiman Long [this message]
2023-04-11 13:35 ` [PATCH v4 2/5] cgroup/cpuset: Make cpuset_fork() handle CLONE_INTO_CGROUP properly Waiman Long
2023-04-11 13:36 ` [PATCH v4 4/5] cgroup/cpuset: Make cpuset_attach_task() skip subpartitions CPUs for top_cpuset Waiman Long
2023-04-11 13:36 ` Waiman Long
2023-04-12 18:26 ` [PATCH v4 0/5] cgroup/cpuset: Fix CLONE_INTO_CGROUP problem & other issues Tejun Heo
2023-04-12 18:26 ` Tejun Heo
[not found] ` <ZDb32V9xcWOi2_CL-NiLfg/pYEd1N0TnZuCh8vA@public.gmane.org>
2023-04-12 18:29 ` Waiman Long
2023-04-12 18:29 ` Waiman Long
2023-04-11 13:35 ` [PATCH v4 3/5] cgroup/cpuset: Add cpuset_can_fork() and cpuset_cancel_fork() methods Waiman Long
2023-04-11 13:35 ` Waiman Long
2023-04-11 13:36 ` [PATCH v4 5/5] cgroup/cpuset: Optimize out unneeded cpuset_can_fork/cpuset_cancel_fork calls Waiman Long
[not found] ` <20230411133601.2969636-6-longman-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2023-04-12 18:27 ` Tejun Heo
2023-04-12 18:27 ` Tejun Heo
[not found] ` <ZDb4G2jgQFK8h8Ys-NiLfg/pYEd1N0TnZuCh8vA@public.gmane.org>
2023-04-12 18:40 ` Waiman Long
2023-04-12 18:40 ` Waiman Long
[not found] ` <90b7bc16-0673-02b7-dad1-f24bc956f1c5-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
2023-04-12 19:17 ` Tejun Heo
2023-04-12 19:17 ` Tejun Heo
[not found] ` <ZDcDxecF0Y5F0pV6-NiLfg/pYEd1N0TnZuCh8vA@public.gmane.org>
2023-04-12 19:23 ` Waiman Long
2023-04-12 19:23 ` Waiman Long
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230411133601.2969636-3-longman@redhat.com \
--to=longman-h+wxahxf7alqt0dzr+alfa@public.gmane.org \
--cc=brauner-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
--cc=cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=dietmar.eggemann-5wv7dgnIgG8@public.gmane.org \
--cc=gscrivan-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
--cc=hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org \
--cc=juri.lelli-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=lizefan.x-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org \
--cc=mkoutny-IBi9RG/b67k@public.gmane.org \
--cc=tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.