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  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 May 2020 17:03:53 -0700
From:   Davidlohr Bueso <dave@...olabs.net>
To:     akpm@...ux-foundation.org
Cc:     peterz@...radead.org, oleg@...hat.com, paulmck@...nel.org,
        tglx@...utronix.de, linux-kernel@...r.kernel.org,
        dave@...olabs.net, Davidlohr Bueso <dbueso@...e.de>
Subject: [PATCH 2/2] kernel/sys: do not grab tasklist_lock for sys_setpriority(PRIO_PROCESS)

This option does not iterate the tasklist and we already have
an rcu aware stable pointer. Also, the nice value is not serialized
by this lock. Reduce the scope of this lock to just PRIO_PGRP
and PRIO_USER - which need to to set the priorities atomically
to the list of tasks, at least vs fork().

Signed-off-by: Davidlohr Bueso <dbueso@...e.de>
---
 kernel/sys.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)

diff --git a/kernel/sys.c b/kernel/sys.c
index 0b72184f5e3e..f9f87775d6d2 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -214,16 +214,19 @@ SYSCALL_DEFINE3(setpriority, int, which, int, who, int, niceval)
 		niceval = MAX_NICE;
 
 	rcu_read_lock();
-	read_lock(&tasklist_lock);
-	switch (which) {
-	case PRIO_PROCESS:
+
+	if (which == PRIO_PROCESS) {
 		if (who)
 			p = find_task_by_vpid(who);
 		else
 			p = current;
 		if (p)
 			error = set_one_prio(p, niceval, error);
-		break;
+		goto out_rcu;
+	}
+
+	read_lock(&tasklist_lock);
+	switch (which) {
 	case PRIO_PGRP:
 		if (who)
 			pgrp = find_vpid(who);
@@ -253,6 +256,7 @@ SYSCALL_DEFINE3(setpriority, int, which, int, who, int, niceval)
 	}
 out_unlock:
 	read_unlock(&tasklist_lock);
+out_rcu:
 	rcu_read_unlock();
 out:
 	return error;
-- 
2.26.1

Powered by blists - more mailing lists