Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]


Groups > linux.kernel > #1212674 > unrolled thread

[PATCH v2] cgroup: pids: fix invalid get/put usage

Started byAleksa Sarai <cyphar@cyphar.com>
First post2015-08-25 05:00 +0200
Last post2015-08-25 20:30 +0200
Articles 2 — 2 participants

Back to article view | Back to linux.kernel


Contents

  [PATCH v2] cgroup: pids: fix invalid get/put usage Aleksa Sarai <cyphar@cyphar.com> - 2015-08-25 05:00 +0200
    Re: [PATCH v2] cgroup: pids: fix invalid get/put usage Tejun Heo <tj@kernel.org> - 2015-08-25 20:30 +0200

#1212674 — [PATCH v2] cgroup: pids: fix invalid get/put usage

FromAleksa Sarai <cyphar@cyphar.com>
Date2015-08-25 05:00 +0200
Subject[PATCH v2] cgroup: pids: fix invalid get/put usage
Message-ID<q1cCR-21v-1@gated-at.bofh.it>
Fix incorrect usage of css_get and css_put to put a different css in
pids_{cancel_,}attach() than the one grabbed in pids_can_attach(). This
could lead to quite serious memory leakage (and unsafe operations on the
putted css).

Signed-off-by: Aleksa Sarai <cyphar@cyphar.com>
---
 kernel/cgroup_pids.c | 19 ++++---------------
 1 file changed, 4 insertions(+), 15 deletions(-)

diff --git a/kernel/cgroup_pids.c b/kernel/cgroup_pids.c
index d75488824ae2..056564f8dd85 100644
--- a/kernel/cgroup_pids.c
+++ b/kernel/cgroup_pids.c
@@ -173,11 +173,11 @@ static int pids_can_attach(struct cgroup_subsys_state *css,
 		struct pids_cgroup *old_pids;
 
 		/*
-		 * Grab a ref to each task's css. We don't drop the ref until
-		 * we either fail and hit ->cancel_attach() or succeed and hit
-		 * ->attach().
+		 * We don't need to grab a ref between here and cancel_attach()
+		 * because cgroup_migrate_prepare_src() protects the task's css
+		 * from being freed before the migration completed or failed.
 		 */
-		old_css = task_get_css(task, pids_cgrp_id);
+		old_css = task_css(task, pids_cgrp_id);
 		old_pids = css_pids(old_css);
 
 		pids_charge(pids, 1);
@@ -202,19 +202,9 @@ static void pids_cancel_attach(struct cgroup_subsys_state *css,
 
 		pids_charge(old_pids, 1);
 		pids_uncharge(pids, 1);
-		css_put(old_css);
 	}
 }
 
-static void pids_attach(struct cgroup_subsys_state *css,
-			struct cgroup_taskset *tset)
-{
-	struct task_struct *task;
-
-	cgroup_taskset_for_each(task, tset)
-		css_put(task_css(task, pids_cgrp_id));
-}
-
 static int pids_can_fork(struct task_struct *task, void **priv_p)
 {
 	struct cgroup_subsys_state *css;
@@ -354,7 +344,6 @@ static struct cftype pids_files[] = {
 struct cgroup_subsys pids_cgrp_subsys = {
 	.css_alloc	= pids_css_alloc,
 	.css_free	= pids_css_free,
-	.attach		= pids_attach,
 	.can_attach 	= pids_can_attach,
 	.cancel_attach 	= pids_cancel_attach,
 	.can_fork	= pids_can_fork,
-- 
2.5.0

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [next] | [standalone]


#1213220

FromTejun Heo <tj@kernel.org>
Date2015-08-25 20:30 +0200
Message-ID<q1r8R-68b-3@gated-at.bofh.it>
In reply to#1212674
On Tue, Aug 25, 2015 at 12:50:44PM +1000, Aleksa Sarai wrote:
> Fix incorrect usage of css_get and css_put to put a different css in
> pids_{cancel_,}attach() than the one grabbed in pids_can_attach(). This
> could lead to quite serious memory leakage (and unsafe operations on the
> putted css).
> 
> Signed-off-by: Aleksa Sarai <cyphar@cyphar.com>

Applied to cgroup/for-4.3 w/ minor comment update.

Thanks.

-- 
tejun
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[toc] | [prev] | [standalone]


Back to top | Article view | linux.kernel


csiph-web