linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Oleg Nesterov <oleg@redhat.com>
To: Roman Gushchin <guroan@gmail.com>
Cc: Tejun Heo <tj@kernel.org>,
	kernel-team@fb.com, cgroups@vger.kernel.org,
	linux-kernel@vger.kernel.org, Roman Gushchin <guro@fb.com>
Subject: Re: [PATCH v10 4/9] cgroup: cgroup v2 freezer
Date: Wed, 24 Apr 2019 18:02:38 +0200	[thread overview]
Message-ID: <20190424160237.GH16167@redhat.com> (raw)
In-Reply-To: <20190405174708.1010-5-guro@fb.com>

On 04/05, Roman Gushchin wrote:
>
> @@ -5830,6 +5926,12 @@ void cgroup_exit(struct task_struct *tsk)
>  		spin_lock_irq(&css_set_lock);
>  		css_set_move_task(tsk, cset, NULL, false);
>  		cset->nr_tasks--;
> +
> +		if (unlikely(cgroup_task_frozen(tsk)))
> +			cgroup_freezer_frozen_exit(tsk);
> +		else if (unlikely(cgroup_task_freeze(tsk)))
> +			cgroup_update_frozen(task_dfl_cgroup(tsk));
> +

Now that I actually applied these patches, I can't understand how can
cgroup_exit() hit cgroup_task_frozen(current)...

A cgroup_task_frozen() task must never return to user-mode or escape
from get_signal() without leave_frozen() or we have a bug?

I must have missed something, could you explain why the patch below is
wrong?

Oleg.


diff --git a/include/linux/cgroup.h b/include/linux/cgroup.h
index 3e2efd4..24b8757 100644
--- a/include/linux/cgroup.h
+++ b/include/linux/cgroup.h
@@ -889,7 +889,6 @@ void cgroup_update_frozen(struct cgroup *cgrp);
 void cgroup_freeze(struct cgroup *cgrp, bool freeze);
 void cgroup_freezer_migrate_task(struct task_struct *task, struct cgroup *src,
 				 struct cgroup *dst);
-void cgroup_freezer_frozen_exit(struct task_struct *task);
 static inline bool cgroup_task_freeze(struct task_struct *task)
 {
 	bool ret;
diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index 6f09f9b..3d5f76a 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -6008,11 +6008,8 @@ void cgroup_exit(struct task_struct *tsk)
 		css_set_move_task(tsk, cset, NULL, false);
 		cset->nr_tasks--;
 
-		if (unlikely(cgroup_task_frozen(tsk)))
-			cgroup_freezer_frozen_exit(tsk);
-		else if (unlikely(cgroup_task_freeze(tsk)))
+		if (unlikely(cgroup_task_freeze(tsk)))
 			cgroup_update_frozen(task_dfl_cgroup(tsk));
-
 		spin_unlock_irq(&css_set_lock);
 	} else {
 		get_css_set(cset);
diff --git a/kernel/cgroup/freezer.c b/kernel/cgroup/freezer.c
index b5ccc87..68aa506 100644
--- a/kernel/cgroup/freezer.c
+++ b/kernel/cgroup/freezer.c
@@ -249,16 +249,6 @@ void cgroup_freezer_migrate_task(struct task_struct *task,
 	cgroup_freeze_task(task, test_bit(CGRP_FREEZE, &dst->flags));
 }
 
-void cgroup_freezer_frozen_exit(struct task_struct *task)
-{
-	struct cgroup *cgrp = task_dfl_cgroup(task);
-
-	lockdep_assert_held(&css_set_lock);
-
-	cgroup_dec_frozen_cnt(cgrp);
-	cgroup_update_frozen(cgrp);
-}
-
 void cgroup_freeze(struct cgroup *cgrp, bool freeze)
 {
 	struct cgroup_subsys_state *css;


  parent reply	other threads:[~2019-04-24 16:02 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-04-05 17:46 [PATCH v10 0/9] freezer for cgroup v2 Roman Gushchin
2019-04-05 17:47 ` [PATCH v10 1/9] cgroup: rename freezer.c into legacy_freezer.c Roman Gushchin
2019-04-05 17:47 ` [PATCH v10 2/9] cgroup: implement __cgroup_task_count() helper Roman Gushchin
2019-04-05 17:47 ` [PATCH v10 3/9] cgroup: protect cgroup->nr_(dying_)descendants by css_set_lock Roman Gushchin
2019-04-05 17:47 ` [PATCH v10 4/9] cgroup: cgroup v2 freezer Roman Gushchin
2019-04-19 15:19   ` Oleg Nesterov
2019-04-19 16:08     ` Oleg Nesterov
2019-04-19 16:36       ` Roman Gushchin
2019-04-19 16:11     ` Roman Gushchin
2019-04-19 16:26       ` Oleg Nesterov
2019-04-19 16:56         ` Roman Gushchin
2019-04-20 10:58           ` Oleg Nesterov
2019-04-22 22:11             ` Roman Gushchin
2019-04-24 15:46               ` Oleg Nesterov
2019-04-24 22:06                 ` Roman Gushchin
2019-04-26 17:40                   ` Oleg Nesterov
2019-04-24 16:02   ` Oleg Nesterov [this message]
2019-04-24 22:10     ` Roman Gushchin
2019-04-26 17:30       ` Oleg Nesterov
2019-04-05 17:47 ` [PATCH v10 5/9] kselftests: cgroup: don't fail on cg_kill_all() error in cg_destroy() Roman Gushchin
2019-04-05 17:47 ` [PATCH v10 6/9] kselftests: cgroup: add freezer controller self-tests Roman Gushchin
2019-07-16 14:48   ` Naresh Kamboju
2019-07-17  0:49     ` Roman Gushchin
2019-07-17  8:56       ` Naresh Kamboju
2019-07-18 18:19         ` Roman Gushchin
2019-04-05 17:47 ` [PATCH v10 7/9] cgroup: make TRACE_CGROUP_PATH irq-safe Roman Gushchin
2019-04-05 17:47 ` [PATCH v10 8/9] cgroup: add tracing points for cgroup v2 freezer Roman Gushchin
2019-04-05 17:47 ` [PATCH v10 9/9] cgroup: document cgroup v2 freezer interface Roman Gushchin
2019-04-19 18:29 ` [PATCH v10 0/9] freezer for cgroup v2 Tejun Heo

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=20190424160237.GH16167@redhat.com \
    --to=oleg@redhat.com \
    --cc=cgroups@vger.kernel.org \
    --cc=guro@fb.com \
    --cc=guroan@gmail.com \
    --cc=kernel-team@fb.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=tj@kernel.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;
as well as URLs for NNTP newsgroup(s).