diff options
| -rw-r--r-- | block/blk-cgroup-fc-appid.c | 4 | ||||
| -rw-r--r-- | include/linux/cgroup-defs.h | 16 | ||||
| -rw-r--r-- | include/linux/cgroup.h | 13 | ||||
| -rw-r--r-- | kernel/cgroup/cgroup.c | 32 | ||||
| -rw-r--r-- | mm/memcontrol.c | 4 | ||||
| -rw-r--r-- | net/netfilter/nft_socket.c | 9 | ||||
| -rw-r--r-- | tools/perf/util/bpf_skel/bperf_cgroup.bpf.c | 2 |
7 files changed, 46 insertions, 34 deletions
diff --git a/block/blk-cgroup-fc-appid.c b/block/blk-cgroup-fc-appid.c index 760a2e1878dd..842e5e1c0f3c 100644 --- a/block/blk-cgroup-fc-appid.c +++ b/block/blk-cgroup-fc-appid.c @@ -19,8 +19,8 @@ int blkcg_set_fc_appid(char *app_id, u64 cgrp_id, size_t app_id_len) return -EINVAL; cgrp = cgroup_get_from_id(cgrp_id); - if (!cgrp) - return -ENOENT; + if (IS_ERR(cgrp)) + return PTR_ERR(cgrp); css = cgroup_get_e_css(cgrp, &io_cgrp_subsys); if (!css) { ret = -ENOENT; diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h index 4bcf56b3491c..1283993d7ea8 100644 --- a/include/linux/cgroup-defs.h +++ b/include/linux/cgroup-defs.h @@ -384,7 +384,7 @@ struct cgroup { /* * The depth this cgroup is at. The root is at depth zero and each * step down the hierarchy increments the level. This along with - * ancestor_ids[] can determine whether a given cgroup is a + * ancestors[] can determine whether a given cgroup is a * descendant of another without traversing the hierarchy. */ int level; @@ -504,8 +504,8 @@ struct cgroup { /* Used to store internal freezer state */ struct cgroup_freezer_state freezer; - /* ids of the ancestors at each level including self */ - u64 ancestor_ids[]; + /* All ancestors including self */ + struct cgroup *ancestors[]; }; /* @@ -522,11 +522,15 @@ struct cgroup_root { /* Unique id for this hierarchy. */ int hierarchy_id; - /* The root cgroup. Root is destroyed on its release. */ + /* + * The root cgroup. The containing cgroup_root will be destroyed on its + * release. cgrp->ancestors[0] will be used overflowing into the + * following field. cgrp_ancestor_storage must immediately follow. + */ struct cgroup cgrp; - /* for cgrp->ancestor_ids[0] */ - u64 cgrp_ancestor_id_storage; + /* must follow cgrp for cgrp->ancestors[0], see above */ + struct cgroup *cgrp_ancestor_storage; /* Number of cgroups in the hierarchy, used only for /proc/cgroups */ atomic_t nr_cgrps; diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h index ac5d0515680e..54de466cc74b 100644 --- a/include/linux/cgroup.h +++ b/include/linux/cgroup.h @@ -574,7 +574,7 @@ static inline bool cgroup_is_descendant(struct cgroup *cgrp, { if (cgrp->root != ancestor->root || cgrp->level < ancestor->level) return false; - return cgrp->ancestor_ids[ancestor->level] == cgroup_id(ancestor); + return cgrp->ancestors[ancestor->level] == ancestor; } /** @@ -591,11 +591,9 @@ static inline bool cgroup_is_descendant(struct cgroup *cgrp, static inline struct cgroup *cgroup_ancestor(struct cgroup *cgrp, int ancestor_level) { - if (cgrp->level < ancestor_level) + if (ancestor_level < 0 || ancestor_level > cgrp->level) return NULL; - while (cgrp && cgrp->level > ancestor_level) - cgrp = cgroup_parent(cgrp); - return cgrp; + return cgrp->ancestors[ancestor_level]; } /** @@ -747,11 +745,6 @@ static inline bool task_under_cgroup_hierarchy(struct task_struct *task, static inline void cgroup_path_from_kernfs_id(u64 id, char *buf, size_t buflen) {} - -static inline struct cgroup *cgroup_get_from_id(u64 id) -{ - return NULL; -} #endif /* !CONFIG_CGROUPS */ #ifdef CONFIG_CGROUPS diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index e4bb5d57f4d1..65497a1a44fa 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -2050,7 +2050,7 @@ int cgroup_setup_root(struct cgroup_root *root, u16 ss_mask) } root_cgrp->kn = kernfs_root_to_node(root->kf_root); WARN_ON_ONCE(cgroup_ino(root_cgrp) != 1); - root_cgrp->ancestor_ids[0] = cgroup_id(root_cgrp); + root_cgrp->ancestors[0] = root_cgrp; ret = css_populate_dir(&root_cgrp->self); if (ret) @@ -5432,8 +5432,7 @@ static struct cgroup *cgroup_create(struct cgroup *parent, const char *name, int ret; /* allocate the cgroup and its ID, 0 is reserved for the root */ - cgrp = kzalloc(struct_size(cgrp, ancestor_ids, (level + 1)), - GFP_KERNEL); + cgrp = kzalloc(struct_size(cgrp, ancestors, (level + 1)), GFP_KERNEL); if (!cgrp) return ERR_PTR(-ENOMEM); @@ -5485,7 +5484,7 @@ static struct cgroup *cgroup_create(struct cgroup *parent, const char *name, spin_lock_irq(&css_set_lock); for (tcgrp = cgrp; tcgrp; tcgrp = cgroup_parent(tcgrp)) { - cgrp->ancestor_ids[tcgrp->level] = cgroup_id(tcgrp); + cgrp->ancestors[tcgrp->level] = tcgrp; if (tcgrp != cgrp) { tcgrp->nr_descendants++; @@ -6038,12 +6037,13 @@ void cgroup_path_from_kernfs_id(u64 id, char *buf, size_t buflen) /* * cgroup_get_from_id : get the cgroup associated with cgroup id * @id: cgroup id - * On success return the cgrp, on failure return NULL + * On success return the cgrp or ERR_PTR on failure + * Only cgroups within current task's cgroup NS are valid. */ struct cgroup *cgroup_get_from_id(u64 id) { struct kernfs_node *kn; - struct cgroup *cgrp = NULL; + struct cgroup *cgrp = NULL, *root_cgrp; kn = kernfs_find_and_get_node_by_id(cgrp_dfl_root.kf_root, id); if (!kn) @@ -6056,10 +6056,20 @@ struct cgroup *cgroup_get_from_id(u64 id) cgrp = NULL; rcu_read_unlock(); - kernfs_put(kn); + + if (!cgrp) + goto out; + + spin_lock_irq(&css_set_lock); + root_cgrp = current_cgns_cgroup_from_root(&cgrp_dfl_root); + spin_unlock_irq(&css_set_lock); + if (!cgroup_is_descendant(cgrp, root_cgrp)) { + cgroup_put(cgrp); + cgrp = NULL; + } out: - return cgrp; + return cgrp ?: ERR_PTR(-ENOENT); } EXPORT_SYMBOL_GPL(cgroup_get_from_id); @@ -6635,8 +6645,12 @@ struct cgroup *cgroup_get_from_path(const char *path) { struct kernfs_node *kn; struct cgroup *cgrp = ERR_PTR(-ENOENT); + struct cgroup *root_cgrp; - kn = kernfs_walk_and_get(cgrp_dfl_root.cgrp.kn, path); + spin_lock_irq(&css_set_lock); + root_cgrp = current_cgns_cgroup_from_root(&cgrp_dfl_root); + kn = kernfs_walk_and_get(root_cgrp->kn, path); + spin_unlock_irq(&css_set_lock); if (!kn) goto out; diff --git a/mm/memcontrol.c b/mm/memcontrol.c index b69979c9ced5..86f5ca8c6fa6 100644 --- a/mm/memcontrol.c +++ b/mm/memcontrol.c @@ -5110,8 +5110,8 @@ struct mem_cgroup *mem_cgroup_get_from_ino(unsigned long ino) struct mem_cgroup *memcg; cgrp = cgroup_get_from_id(ino); - if (!cgrp) - return ERR_PTR(-ENOENT); + if (IS_ERR(cgrp)) + return PTR_ERR(cgrp); css = cgroup_get_e_css(cgrp, &memory_cgrp_subsys); if (css) diff --git a/net/netfilter/nft_socket.c b/net/netfilter/nft_socket.c index a7de29137618..49a5348a6a14 100644 --- a/net/netfilter/nft_socket.c +++ b/net/netfilter/nft_socket.c @@ -40,16 +40,17 @@ static noinline bool nft_sock_get_eval_cgroupv2(u32 *dest, struct sock *sk, const struct nft_pktinfo *pkt, u32 level) { struct cgroup *cgrp; + u64 cgid; if (!sk_fullsock(sk)) return false; - cgrp = sock_cgroup_ptr(&sk->sk_cgrp_data); - if (level > cgrp->level) + cgrp = cgroup_ancestor(sock_cgroup_ptr(&sk->sk_cgrp_data), level); + if (!cgrp) return false; - memcpy(dest, &cgrp->ancestor_ids[level], sizeof(u64)); - + cgid = cgroup_id(cgrp); + memcpy(dest, &cgid, sizeof(u64)); return true; } #endif diff --git a/tools/perf/util/bpf_skel/bperf_cgroup.bpf.c b/tools/perf/util/bpf_skel/bperf_cgroup.bpf.c index 292c430768b5..bd6a420acc8f 100644 --- a/tools/perf/util/bpf_skel/bperf_cgroup.bpf.c +++ b/tools/perf/util/bpf_skel/bperf_cgroup.bpf.c @@ -68,7 +68,7 @@ static inline int get_cgroup_v1_idx(__u32 *cgrps, int size) break; // convert cgroup-id to a map index - cgrp_id = BPF_CORE_READ(cgrp, ancestor_ids[i]); + cgrp_id = BPF_CORE_READ(cgrp, ancestors[i], kn, id); elem = bpf_map_lookup_elem(&cgrp_idx, &cgrp_id); if (!elem) continue; |