From: Yafang Shao <laoar.shao-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
To: ast-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org,
daniel-FeC+5ew28dpmcu3hnIyYJQ@public.gmane.org,
john.fastabend-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org,
andrii-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org,
martin.lau-fxUVXftIFDnyG1zEObXtfA@public.gmane.org,
song-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org,
yonghong.song-fxUVXftIFDnyG1zEObXtfA@public.gmane.org,
kpsingh-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org,
sdf-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org,
haoluo-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org,
jolsa-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org,
tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org,
lizefan.x-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org,
hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org,
yosryahmed-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org,
mkoutny-IBi9RG/b67k@public.gmane.org
Cc: cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
bpf-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
Yafang Shao <laoar.shao-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Subject: [RFC PATCH bpf-next 2/8] cgroup: Enable task_under_cgroup_hierarchy() on cgroup1
Date: Fri, 22 Sep 2023 11:28:40 +0000 [thread overview]
Message-ID: <20230922112846.4265-3-laoar.shao@gmail.com> (raw)
In-Reply-To: <20230922112846.4265-1-laoar.shao-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
At present, the task_under_cgroup_hierarchy() function serves the purpose
of determining whether a task resides exclusively within a cgroup2
hierarchy. However, considering the ongoing prevalence of cgroup1 and the
substantial effort and time required to migrate all cgroup1-based
applications to the cgroup2 framework, it becomes beneficial to make a
minor adjustment that expands its functionality to encompass cgroup1 as
well. By implementing this modification, we will gain the capability to
easily confirm a task's cgroup membership within BPF programs. For example,
we can effortlessly verify if a task belongs to a cgroup1 directory, such
as '/sys/fs/cgroup/cpu,cpuacct/kubepods/', or
'/sys/fs/cgroup/cpu,cpuacct/system.slice/'.
Signed-off-by: Yafang Shao <laoar.shao-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
---
include/linux/cgroup-defs.h | 20 ++++++++++++++++++++
include/linux/cgroup.h | 30 ++++++++++++++++++++++++++----
kernel/cgroup/cgroup-internal.h | 20 --------------------
3 files changed, 46 insertions(+), 24 deletions(-)
diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h
index f1b3151ac30b..5795825a04ff 100644
--- a/include/linux/cgroup-defs.h
+++ b/include/linux/cgroup-defs.h
@@ -299,6 +299,26 @@ struct css_set {
struct rcu_head rcu_head;
};
+/*
+ * A cgroup can be associated with multiple css_sets as different tasks may
+ * belong to different cgroups on different hierarchies. In the other
+ * direction, a css_set is naturally associated with multiple cgroups.
+ * This M:N relationship is represented by the following link structure
+ * which exists for each association and allows traversing the associations
+ * from both sides.
+ */
+struct cgrp_cset_link {
+ /* the cgroup and css_set this link associates */
+ struct cgroup *cgrp;
+ struct css_set *cset;
+
+ /* list of cgrp_cset_links anchored at cgrp->cset_links */
+ struct list_head cset_link;
+
+ /* list of cgrp_cset_links anchored at css_set->cgrp_links */
+ struct list_head cgrp_link;
+};
+
struct cgroup_base_stat {
struct task_cputime cputime;
diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index b307013b9c6c..e16cfb98b44c 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -387,8 +387,8 @@ static inline void cgroup_unlock(void)
* The caller can also specify additional allowed conditions via @__c, such
* as locks used during the cgroup_subsys::attach() methods.
*/
-#ifdef CONFIG_PROVE_RCU
extern spinlock_t css_set_lock;
+#ifdef CONFIG_PROVE_RCU
#define task_css_set_check(task, __c) \
rcu_dereference_check((task)->cgroups, \
rcu_read_lock_sched_held() || \
@@ -543,15 +543,37 @@ static inline struct cgroup *cgroup_ancestor(struct cgroup *cgrp,
* @ancestor: possible ancestor of @task's cgroup
*
* Tests whether @task's default cgroup hierarchy is a descendant of @ancestor.
- * It follows all the same rules as cgroup_is_descendant, and only applies
- * to the default hierarchy.
+ * It follows all the same rules as cgroup_is_descendant.
*/
static inline bool task_under_cgroup_hierarchy(struct task_struct *task,
struct cgroup *ancestor)
{
struct css_set *cset = task_css_set(task);
+ struct cgrp_cset_link *link;
+ struct cgroup *cgrp = NULL;
+ bool ret = false;
+
+ if (ancestor->root == &cgrp_dfl_root)
+ return cgroup_is_descendant(cset->dfl_cgrp, ancestor);
+
+ if (cset == &init_css_set)
+ return ancestor == &ancestor->root->cgrp;
+
+ spin_lock_irq(&css_set_lock);
+ list_for_each_entry(link, &cset->cgrp_links, cgrp_link) {
+ struct cgroup *c = link->cgrp;
+
+ if (c->root == ancestor->root) {
+ cgrp = c;
+ break;
+ }
+ }
+ spin_unlock_irq(&css_set_lock);
- return cgroup_is_descendant(cset->dfl_cgrp, ancestor);
+ WARN_ON_ONCE(!cgrp);
+ if (cgroup_is_descendant(cgrp, ancestor))
+ ret = true;
+ return ret;
}
/* no synchronization, the result can only be used as a hint */
diff --git a/kernel/cgroup/cgroup-internal.h b/kernel/cgroup/cgroup-internal.h
index c56071f150f2..620c60c9daa3 100644
--- a/kernel/cgroup/cgroup-internal.h
+++ b/kernel/cgroup/cgroup-internal.h
@@ -83,26 +83,6 @@ struct cgroup_file_ctx {
} procs1;
};
-/*
- * A cgroup can be associated with multiple css_sets as different tasks may
- * belong to different cgroups on different hierarchies. In the other
- * direction, a css_set is naturally associated with multiple cgroups.
- * This M:N relationship is represented by the following link structure
- * which exists for each association and allows traversing the associations
- * from both sides.
- */
-struct cgrp_cset_link {
- /* the cgroup and css_set this link associates */
- struct cgroup *cgrp;
- struct css_set *cset;
-
- /* list of cgrp_cset_links anchored at cgrp->cset_links */
- struct list_head cset_link;
-
- /* list of cgrp_cset_links anchored at css_set->cgrp_links */
- struct list_head cgrp_link;
-};
-
/* used to track tasks and csets during migration */
struct cgroup_taskset {
/* the src and dst cset list running through cset->mg_node */
--
2.30.1 (Apple Git-130)
next prev parent reply other threads:[~2023-09-22 11:28 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-09-22 11:28 [RFC PATCH bpf-next 0/8] bpf, cgroup: Add bpf support for cgroup controller Yafang Shao
[not found] ` <20230922112846.4265-1-laoar.shao-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2023-09-22 11:28 ` [RFC PATCH bpf-next 1/8] bpf: Fix missed rcu read lock in bpf_task_under_cgroup() Yafang Shao
2023-09-22 11:28 ` Yafang Shao [this message]
2023-09-22 11:28 ` [RFC PATCH bpf-next 3/8] cgroup: Add cgroup_get_from_id_within_subsys() Yafang Shao
2023-09-22 11:28 ` [RFC PATCH bpf-next 4/8] bpf: Add new kfuncs support for cgroup controller Yafang Shao
2023-09-22 11:28 ` [RFC PATCH bpf-next 5/8] selftests/bpf: Fix issues in setup_classid_environment() Yafang Shao
2023-09-22 11:28 ` [RFC PATCH bpf-next 6/8] selftests/bpf: Add parallel support for classid Yafang Shao
2023-09-22 11:28 ` [RFC PATCH bpf-next 7/8] selftests/bpf: Add new cgroup helper get_classid_cgroup_id() Yafang Shao
2023-09-22 11:28 ` [RFC PATCH bpf-next 8/8] selftests/bpf: Add selftests for cgroup controller Yafang Shao
2023-09-22 16:52 ` [RFC PATCH bpf-next 0/8] bpf, cgroup: Add bpf support " Tejun Heo
[not found] ` <ZQ3GQmYrYyKAg2uK-NiLfg/pYEd1N0TnZuCh8vA@public.gmane.org>
2023-09-24 6:32 ` Yafang Shao
[not found] ` <CALOAHbA9-BT1daw-KXHtsrN=uRQyt-p6LU=BEpvF2Yk42A_Vxw-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2023-09-25 18:43 ` Tejun Heo
2023-09-25 18:43 ` Tejun Heo
[not found] ` <ZRHU6MfwqRxjBFUH-NiLfg/pYEd1N0TnZuCh8vA@public.gmane.org>
2023-09-26 3:01 ` Yafang Shao
2023-09-26 3:01 ` Yafang Shao
2023-09-26 18:25 ` Tejun Heo
2023-09-27 2:27 ` Yafang Shao
2023-09-25 18:22 ` Kui-Feng Lee
2023-09-25 18:22 ` Kui-Feng Lee
[not found] ` <9e83bda8-ea1b-75b9-c55f-61cf11b4cd83-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2023-09-26 3:08 ` Yafang Shao
2023-09-26 3:08 ` Yafang Shao
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230922112846.4265-3-laoar.shao@gmail.com \
--to=laoar.shao-re5jqeeqqe8avxtiumwx3w@public.gmane.org \
--cc=andrii-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
--cc=ast-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
--cc=bpf-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=cgroups-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=daniel-FeC+5ew28dpmcu3hnIyYJQ@public.gmane.org \
--cc=hannes-druUgvl0LCNAfugRpC6u6w@public.gmane.org \
--cc=haoluo-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
--cc=john.fastabend-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org \
--cc=jolsa-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
--cc=kpsingh-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
--cc=lizefan.x-EC8Uxl6Npydl57MIdRCFDg@public.gmane.org \
--cc=martin.lau-fxUVXftIFDnyG1zEObXtfA@public.gmane.org \
--cc=mkoutny-IBi9RG/b67k@public.gmane.org \
--cc=sdf-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
--cc=song-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
--cc=tj-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
--cc=yonghong.song-fxUVXftIFDnyG1zEObXtfA@public.gmane.org \
--cc=yosryahmed-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox