kernel_google_redbull

Evolution-X-Devices/kernel_google_redbull

Files

Alexander Kuznetsov 7e7ff4d058 cgroup1: don't allow '\n' in renaming

commit b7e24eb1caa5f8da20d405d262dba67943aedc42 upstream.

cgroup_mkdir() have restriction on newline usage in names:
$ mkdir $'/sys/fs/cgroup/cpu/test\ntest2'
mkdir: cannot create directory
'/sys/fs/cgroup/cpu/test\ntest2': Invalid argument

But in cgroup1_rename() such check is missed.
This allows us to make /proc/<pid>/cgroup unparsable:
$ mkdir /sys/fs/cgroup/cpu/test
$ mv /sys/fs/cgroup/cpu/test $'/sys/fs/cgroup/cpu/test\ntest2'
$ echo $$ > $'/sys/fs/cgroup/cpu/test\ntest2'
$ cat /proc/self/cgroup
11:pids:/
10:freezer:/
9:hugetlb:/
8:cpuset:/
7:blkio:/user.slice
6:memory:/user.slice
5:net_cls,net_prio:/
4:perf_event:/
3:devices:/user.slice
2:cpu,cpuacct:/test
test2
1:name=systemd:/
0::/

Signed-off-by: Alexander Kuznetsov <wwfq@yandex-team.ru>
Reported-by: Andrey Krasichkov <buglloc@yandex-team.ru>
Acked-by: Dmitry Yakunin <zeil@yandex-team.ru>
Cc: stable@vger.kernel.org
Signed-off-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

2021-06-16 11:54:59 +02:00

cgroup-internal.h

cgroup/tracing: Move taking of spin lock out of trace event handlers

2018-07-11 10:48:47 -07:00

cgroup-v1.c

cgroup1: don't allow '\n' in renaming

2021-06-16 11:54:59 +02:00

cgroup.c

cgroup: disable controllers at parse time

2021-06-16 11:54:57 +02:00

cpuset.c

cpuset: restore sanity to cpuset_cpus_allowed_fallback()