From: Tejun Heo tj@kernel.org
commit 4a7ba45b1a435e7097ca0f79a847d0949d0eb088 upstream.
memcg_write_event_control() accesses the dentry->d_name of the specified control fd to route the write call. As a cgroup interface file can't be renamed, it's safe to access d_name as long as the specified file is a regular cgroup file. Also, as these cgroup interface files can't be removed before the directory, it's safe to access the parent too.
Prior to 347c4a874710 ("memcg: remove cgroup_event->cft"), there was a call to __file_cft() which verified that the specified file is a regular cgroupfs file before further accesses. The cftype pointer returned from __file_cft() was no longer necessary and the commit inadvertently dropped the file type check with it allowing any file to slip through. With the invarients broken, the d_name and parent accesses can now race against renames and removals of arbitrary files and cause use-after-free's.
Fix the bug by resurrecting the file type check in __file_cft(). Now that cgroupfs is implemented through kernfs, checking the file operations needs to go through a layer of indirection. Instead, let's check the superblock and dentry type.
Link: https://lkml.kernel.org/r/Y5FRm/cfcKPGzWwl@slm.duckdns.org Fixes: 347c4a874710 ("memcg: remove cgroup_event->cft") Signed-off-by: Tejun Heo tj@kernel.org Reported-by: Jann Horn jannh@google.com Acked-by: Roman Gushchin roman.gushchin@linux.dev Acked-by: Johannes Weiner hannes@cmpxchg.org Cc: Linus Torvalds torvalds@linux-foundation.org Cc: Michal Hocko mhocko@kernel.org Cc: Muchun Song songmuchun@bytedance.com Cc: Shakeel Butt shakeelb@google.com Cc: stable@vger.kernel.org [3.14+] Signed-off-by: Andrew Morton akpm@linux-foundation.org Signed-off-by: Greg Kroah-Hartman gregkh@linuxfoundation.org --- include/linux/cgroup.h | 1 + kernel/cgroup/cgroup-internal.h | 1 - mm/memcontrol.c | 15 +++++++++++++-- 3 files changed, 14 insertions(+), 3 deletions(-)
--- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -68,6 +68,7 @@ struct css_task_iter { struct list_head iters_node; /* css_set->task_iters */ };
+extern struct file_system_type cgroup_fs_type; extern struct cgroup_root cgrp_dfl_root; extern struct css_set init_css_set;
--- a/kernel/cgroup/cgroup-internal.h +++ b/kernel/cgroup/cgroup-internal.h @@ -122,7 +122,6 @@ extern struct mutex cgroup_mutex; extern spinlock_t css_set_lock; extern struct cgroup_subsys *cgroup_subsys[]; extern struct list_head cgroup_roots; -extern struct file_system_type cgroup_fs_type;
/* iterate across the hierarchies */ #define for_each_root(root) \ --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -3878,6 +3878,7 @@ static ssize_t memcg_write_event_control unsigned int efd, cfd; struct fd efile; struct fd cfile; + struct dentry *cdentry; const char *name; char *endp; int ret; @@ -3929,6 +3930,16 @@ static ssize_t memcg_write_event_control goto out_put_cfile;
/* + * The control file must be a regular cgroup1 file. As a regular cgroup + * file can't be renamed, it's safe to access its name afterwards. + */ + cdentry = cfile.file->f_path.dentry; + if (cdentry->d_sb->s_type != &cgroup_fs_type || !d_is_reg(cdentry)) { + ret = -EINVAL; + goto out_put_cfile; + } + + /* * Determine the event callbacks and set them in @event. This used * to be done via struct cftype but cgroup core no longer knows * about these events. The following is crude but the whole thing @@ -3936,7 +3947,7 @@ static ssize_t memcg_write_event_control * * DO NOT ADD NEW FILES. */ - name = cfile.file->f_path.dentry->d_name.name; + name = cdentry->d_name.name;
if (!strcmp(name, "memory.usage_in_bytes")) { event->register_event = mem_cgroup_usage_register_event; @@ -3960,7 +3971,7 @@ static ssize_t memcg_write_event_control * automatically removed on cgroup destruction but the removal is * asynchronous, so take an extra ref on @css. */ - cfile_css = css_tryget_online_from_dir(cfile.file->f_path.dentry->d_parent, + cfile_css = css_tryget_online_from_dir(cdentry->d_parent, &memory_cgrp_subsys); ret = -EINVAL; if (IS_ERR(cfile_css))