lwn.git/kernel/cgroup, branch master

Merge tag 'cgroup-for-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup

2026-04-15T17:18:49+00:00

Pull cgroup updates from Tejun Heo: - cgroup_file_notify() locking converted from a global lock to per-cgroup_file spinlock with a lockless fast-path when no notification is needed - Misc changes including exposing cgroup helpers for sched_ext and minor fixes * tag 'cgroup-for-7.1' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup: cgroup/rdma: fix swapped arguments in pr_warn() format string cgroup/dmem: remove region parameter from dmemcg_parse_limit cgroup: replace global cgroup_file_kn_lock with per-cgroup_file lock cgroup: add lockless fast-path checks to cgroup_file_notify() cgroup: reduce cgroup_file_kn_lock hold time in cgroup_file_notify() cgroup: Expose some cgroup helpers

cgroup/rdma: fix swapped arguments in pr_warn() format string

2026-04-10T08:30:08+00:00

The format string says "device %p ... rdma cgroup %p" but the arguments were passed as (cg, device), printing them in the wrong order. Signed-off-by: cuitao Signed-off-by: Tejun Heo

cgroup/cpuset: Skip security check for hotplug induced v1 task migration

2026-03-31T19:14:13+00:00

When a CPU hot removal causes a v1 cpuset to lose all its CPUs, the cpuset hotplug handler will schedule a work function to migrate tasks in that cpuset with no CPU to its ancestor to enable those tasks to continue running. If a strict security policy is in place, however, the task migration may fail when security_task_setscheduler() call in cpuset_can_attach() returns a -EACCES error. That will mean that those tasks will have no CPU to run on. The system administrators will have to explicitly intervene to either add CPUs to that cpuset or move the tasks elsewhere if they are aware of it. This problem was found by a reported test failure in the LTP's cpuset_hotplug_test.sh. Fix this problem by treating this special case as an exception to skip the setsched security check in cpuset_can_attach() when a v1 cpuset with tasks have no CPU left. With that patch applied, the cpuset_hotplug_test.sh test can be run successfully without failure. Signed-off-by: Waiman Long Signed-off-by: Tejun Heo

cgroup/cpuset: Simplify setsched decision check in task iteration loop of cpuset_can_attach()

2026-03-31T19:14:13+00:00

Centralize the check required to run security_task_setscheduler() in the task iteration loop of cpuset_can_attach() outside of the loop as it has no dependency on the characteristics of the tasks themselves. There is no functional change. Signed-off-by: Waiman Long Signed-off-by: Tejun Heo

cgroup: Fix cgroup_drain_dying() testing the wrong condition

2026-03-26T00:08:04+00:00

cgroup_drain_dying() was using cgroup_is_populated() to test whether there are dying tasks to wait for. cgroup_is_populated() tests nr_populated_csets, nr_populated_domain_children and nr_populated_threaded_children, but cgroup_drain_dying() only needs to care about this cgroup's own tasks - whether there are children is cgroup_destroy_locked()'s concern. This caused hangs during shutdown. When systemd tried to rmdir a cgroup that had no direct tasks but had a populated child, cgroup_drain_dying() would enter its wait loop because cgroup_is_populated() was true from nr_populated_domain_children. The task iterator found nothing to wait for, yet the populated state never cleared because it was driven by live tasks in the child cgroup. Fix it by using cgroup_has_tasks() which only tests nr_populated_csets. v3: Fix cgroup_is_populated() -> cgroup_has_tasks() (Sebastian). v2: https://lore.kernel.org/r/20260323200205.1063629-1-tj@kernel.org Reported-by: Sebastian Andrzej Siewior Fixes: 1b164b876c36 ("cgroup: Wait for dying tasks to leave on rmdir") Signed-off-by: Tejun Heo Tested-by: Sebastian Andrzej Siewior

cgroup: Wait for dying tasks to leave on rmdir

2026-03-24T20:21:40+00:00

a72f73c4dd9b ("cgroup: Don't expose dead tasks in cgroup") hid PF_EXITING tasks from cgroup.procs so that systemd doesn't see tasks that have already been reaped via waitpid(). However, the populated counter (nr_populated_csets) is only decremented when the task later passes through cgroup_task_dead() in finish_task_switch(). This means cgroup.procs can appear empty while the cgroup is still populated, causing rmdir to fail with -EBUSY. Fix this by making cgroup_rmdir() wait for dying tasks to fully leave. If the cgroup is populated but all remaining tasks have PF_EXITING set (the task iterator returns none due to the existing filter), wait for a kick from cgroup_task_dead() and retry. The wait is brief as tasks are removed from the cgroup's css_set between PF_EXITING assertion in do_exit() and cgroup_task_dead() in finish_task_switch(). v2: cgroup_is_populated() true to false transition happens under css_set_lock not cgroup_mutex, so retest under css_set_lock before sleeping to avoid missed wakeups (Sebastian). Fixes: a72f73c4dd9b ("cgroup: Don't expose dead tasks in cgroup") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-lkp/202603222104.2c81684e-lkp@intel.com Reported-by: Sebastian Andrzej Siewior Signed-off-by: Tejun Heo Reviewed-by: Sebastian Andrzej Siewior Cc: Bert Karwatzki Cc: Michal Koutny Cc: cgroups@vger.kernel.org

cgroup/dmem: remove region parameter from dmemcg_parse_limit

2026-03-21T19:24:02+00:00

dmemcg_parse_limit does not use the region parameter. Remove it. Signed-off-by: Thadeu Lima de Souza Cascardo Signed-off-by: Tejun Heo

cgroup: replace global cgroup_file_kn_lock with per-cgroup_file lock

2026-03-11T22:16:21+00:00

Replace the global cgroup_file_kn_lock with a per-cgroup_file spinlock to eliminate cross-cgroup contention as it is not really protecting data shared between different cgroups. The lock is initialized in cgroup_add_file() alongside timer_setup(). No lock acquisition is needed during initialization since the cgroup directory is being populated under cgroup_mutex and no concurrent accessors exist at that point. Reported-by: Jakub Kicinski Signed-off-by: Shakeel Butt Signed-off-by: Tejun Heo

cgroup: add lockless fast-path checks to cgroup_file_notify()

2026-03-11T22:16:21+00:00

Add lockless checks before acquiring cgroup_file_kn_lock: 1. READ_ONCE(cfile->kn) NULL check to skip torn-down files. 2. READ_ONCE(cfile->notified_at) rate-limit check to skip when within the notification interval. If within the interval, arm the deferred timer via timer_reduce() and confirm it is pending before returning -- if the timer fired in between, fall through to the lock path so the notification is not lost. Both checks have safe error directions -- a stale read can only cause unnecessary lock acquisition, never a missed notification. The critical section is simplified to just taking a kernfs_get() reference and updating notified_at. Annotate cfile->kn and cfile->notified_at write sites with WRITE_ONCE() to pair with the lockless readers. Reported-by: Jakub Kicinski Signed-off-by: Shakeel Butt Signed-off-by: Tejun Heo

cgroup: reduce cgroup_file_kn_lock hold time in cgroup_file_notify()

2026-03-11T22:16:21+00:00

cgroup_file_notify() calls kernfs_notify() while holding the global cgroup_file_kn_lock. kernfs_notify() does non-trivial work including wake_up_interruptible() and acquisition of a second global spinlock (kernfs_notify_lock), inflating the hold time. Take a kernfs_get() reference under the lock and call kernfs_notify() after dropping it, following the pattern from cgroup_file_show(). Reported-by: Jakub Kicinski Signed-off-by: Shakeel Butt Signed-off-by: Tejun Heo