This is a note to let you know that I've just added the patch titled
cgroup/cpuset: Make cpuset_fork() handle CLONE_INTO_CGROUP properly
to the 5.10-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
cgroup-cpuset-make-cpuset_fork-handle-clone_into_cgroup-properly.patch
and it can be found in the queue-5.10 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@xxxxxxxxxxxxxxx> know about it.
>From stable-owner@xxxxxxxxxxxxxxx Mon Apr 17 19:21:27 2023
From: Waiman Long <longman@xxxxxxxxxx>
Date: Mon, 17 Apr 2023 13:19:57 -0400
Subject: cgroup/cpuset: Make cpuset_fork() handle CLONE_INTO_CGROUP properly
To: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>, stable@xxxxxxxxxxxxxxx
Cc: "Tejun Heo" <tj@xxxxxxxxxx>, "Michal Koutný" <mkoutny@xxxxxxxx>, "Giuseppe Scrivano" <gscrivan@xxxxxxxxxx>, "Waiman Long" <longman@xxxxxxxxxx>
Message-ID: <20230417171958.3389333-3-longman@xxxxxxxxxx>
From: Waiman Long <longman@xxxxxxxxxx>
commit 42a11bf5c5436e91b040aeb04063be1710bb9f9c upstream.
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@xxxxxxxxxx>
Signed-off-by: Waiman Long <longman@xxxxxxxxxx>
Cc: stable@xxxxxxxxxxxxxxx # v5.7+
Signed-off-by: Tejun Heo <tj@xxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
kernel/cgroup/cpuset.c | 60 +++++++++++++++++++++++++++++++++++--------------
1 file changed, 44 insertions(+), 16 deletions(-)
--- a/kernel/cgroup/cpuset.c
+++ b/kernel/cgroup/cpuset.c
@@ -2207,16 +2207,29 @@ static void cpuset_cancel_attach(struct
}
/*
- * 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);
+
+ /*
+ * 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;
@@ -2237,16 +2250,8 @@ static void cpuset_attach(struct cgroup_
guarantee_online_mems(cs, &cpuset_attach_nodemask_to);
- cgroup_taskset_for_each(task, css, tset) {
- /*
- * 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
@@ -2914,11 +2919,34 @@ static void cpuset_bind(struct cgroup_su
*/
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;
+ }
+
+ /* CLONE_INTO_CGROUP */
+ percpu_down_write(&cpuset_rwsem);
+ guarantee_online_mems(cs, &cpuset_attach_nodemask_to);
- set_cpus_allowed_ptr(task, current->cpus_ptr);
- task->mems_allowed = current->mems_allowed;
+ /* prepare for attach */
+ if (cs == &top_cpuset)
+ cpumask_copy(cpus_attach, cpu_possible_mask);
+ else
+ guarantee_online_cpus(cs, cpus_attach);
+ cpuset_attach_task(cs, task);
+ percpu_up_write(&cpuset_rwsem);
}
struct cgroup_subsys cpuset_cgrp_subsys = {
Patches currently in stable-queue which might be from stable-owner@xxxxxxxxxxxxxxx are
queue-5.10/kbuild-check-the-minimum-assembler-version-in-kconfig.patch
queue-5.10/cgroup-cpuset-make-cpuset_fork-handle-clone_into_cgroup-properly.patch
queue-5.10/kbuild-check-config_as_is_llvm-instead-of-llvm_ias.patch
queue-5.10/riscv-handle-zicsr-zifencei-issues-between-clang-and-binutils.patch
queue-5.10/kbuild-switch-to-f-variants-of-integrated-assembler-flag.patch
queue-5.10/cgroup-cpuset-add-cpuset_can_fork-and-cpuset_cancel_fork-methods.patch
queue-5.10/cgroup-cpuset-change-references-of-cpuset_mutex-to-cpuset_rwsem.patch
queue-5.10/cgroup-cpuset-skip-spread-flags-update-on-v2.patch
