linux-kernel - [patch] BFS 421: remove global runqueue lock in running tick

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-next>] [day] [month] [year] [list]

Date:	Fri, 8 Jun 2012 21:37:51 +0800
From:	Hillf Danton <dhillf@...il.com>
To:	LKML <linux-kernel@...r.kernel.org>,
	Hillf Danton <dhillf@...il.com>
Subject: [patch] BFS 421: remove global runqueue lock in running tick

We simply deschedule current, with no bother grabbing global lock, since IPI,
if triggered by preempt, could not change our result, like no schedule when
preempt enabled.

	preempt_enable_no_resched();
	if (unlikely(need_resched()))
		goto need_resched;


--- a/kernel/sched/bfs.c	Wed Jun  6 21:02:50 2012
+++ b/kernel/sched/bfs.c	Fri Jun  8 21:17:24 2012
@@ -2784,11 +2784,13 @@ static void task_running_tick(struct rq
 	} else if (rq->rq_time_slice >= RESCHED_US)
 			return;

-	/* p->time_slice < RESCHED_US. We only modify task_struct under grq lock */
+	/*
+	 * With irq disabled, current is descheduled without global lock
+	 * held and IPI cared.
+	 */
 	p = rq->curr;
-	grq_lock();
-	set_tsk_need_resched(p);
-	grq_unlock();
+	if (!test_tsk_need_resched(p))
+		set_tsk_need_resched(p);
 }


--
--
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/