linux-kernel - [-mm][PATCH 2/2] Memory rlimit enhance mm_owner_changed callback to deal with exited owner

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Mon, 11 Aug 2008 15:37:43 +0530
From:	Balbir Singh <balbir@...ux.vnet.ibm.com>
To:	linux-mm@...ck.org
Cc:	Sudhir Kumar <skumar@...ux.vnet.ibm.com>,
	YAMAMOTO Takashi <yamamoto@...inux.co.jp>,
	Paul Menage <menage@...gle.com>, lizf@...fujitsu.com,
	linux-kernel@...r.kernel.org, nishimura@....nes.nec.co.jp,
	Pavel Emelianov <xemul@...nvz.org>, hugh@...itas.com,
	Balbir Singh <balbir@...ux.vnet.ibm.com>,
	Andrew Morton <akpm@...ux-foundation.org>,
	KAMEZAWA Hiroyuki <kamezawa.hiroyu@...fujitsu.com>
Subject: [-mm][PATCH 2/2] Memory rlimit enhance mm_owner_changed callback to deal with exited owner



mm_owner_changed callback can also be called with new task set to NULL.
(race between try_to_unuse() and mm->owner exiting). Surprisingly the order
of cgroup arguments being passed was incorrect (proves that we did not
run into mm_owner_changed callback at all).

Signed-off-by: Balbir Singh <balbir@...ux.vnet.ibm.com>
---

 mm/memrlimitcgroup.c |   15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)

diff -puN mm/memrlimitcgroup.c~memrlimit-handle-mm-owner-notification-with-task-null mm/memrlimitcgroup.c
--- linux-2.6.27-rc1/mm/memrlimitcgroup.c~memrlimit-handle-mm-owner-notification-with-task-null	2008-08-05 10:56:56.000000000 +0530
+++ linux-2.6.27-rc1-balbir/mm/memrlimitcgroup.c	2008-08-05 11:24:04.000000000 +0530
@@ -73,6 +73,12 @@ void memrlimit_cgroup_uncharge_as(struct
 {
 	struct memrlimit_cgroup *memrcg;
 
+	/*
+	 * Uncharge happened as a part of the mm_owner_changed callback
+	 */
+	if (!mm->owner)
+		return;
+
 	memrcg = memrlimit_cgroup_from_task(mm->owner);
 	res_counter_uncharge(&memrcg->as_res, (nr_pages << PAGE_SHIFT));
 }
@@ -235,8 +241,8 @@ out:
  * This callback is called with mmap_sem held
  */
 static void memrlimit_cgroup_mm_owner_changed(struct cgroup_subsys *ss,
-						struct cgroup *cgrp,
 						struct cgroup *old_cgrp,
+						struct cgroup *cgrp,
 						struct task_struct *p)
 {
 	struct memrlimit_cgroup *memrcg, *old_memrcg;
@@ -246,7 +252,12 @@ static void memrlimit_cgroup_mm_owner_ch
 	memrcg = memrlimit_cgroup_from_cgrp(cgrp);
 	old_memrcg = memrlimit_cgroup_from_cgrp(old_cgrp);
 
-	if (res_counter_charge(&memrcg->as_res, (mm->total_vm << PAGE_SHIFT)))
+	/*
+	 * If we don't have a new cgroup, we just uncharge from the old one.
+	 * It means that the task is going away
+	 */
+	if (memrcg &&
+	    res_counter_charge(&memrcg->as_res, (mm->total_vm << PAGE_SHIFT)))
 		goto out;
 	res_counter_uncharge(&old_memrcg->as_res, (mm->total_vm << PAGE_SHIFT));
 out:
_

-- 
	Warm Regards,
	Balbir Singh
	Linux Technology Center
	IBM, ISTL
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/