linux-kernel - [PATCH v3 net-next] net: link_watch: prevent starvation when processing linkwatch wq

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]

Message-ID: <1561684399-235123-1-git-send-email-linyunsheng@huawei.com>
Date:   Fri, 28 Jun 2019 09:13:19 +0800
From:   Yunsheng Lin <linyunsheng@...wei.com>
To:     <davem@...emloft.net>
CC:     <hkallweit1@...il.com>, <gregkh@...uxfoundation.org>,
        <tglx@...utronix.de>, <netdev@...r.kernel.org>,
        <linux-kernel@...r.kernel.org>, <pbonzini@...hat.com>,
        <rkrcmar@...hat.com>, <kvm@...r.kernel.org>
Subject: [PATCH v3 net-next] net: link_watch: prevent starvation when processing linkwatch wq

When user has configured a large number of virtual netdev, such
as 4K vlans, the carrier on/off operation of the real netdev
will also cause it's virtual netdev's link state to be processed
in linkwatch. Currently, the processing is done in a work queue,
which may cause rtnl locking starvation problem and worker
starvation problem for other work queue, such as irqfd_inject wq.

This patch releases the cpu when link watch worker has processed
a fixed number of netdev' link watch event, and schedule the
work queue again when there is still link watch event remaining.

Signed-off-by: Yunsheng Lin <linyunsheng@...wei.com>
---
V2: use cond_resched and rtnl_unlock after processing a fixed
    number of events
V3: fall back to v1 and change commit log to reflect that.
---
---
 net/core/link_watch.c | 13 ++++++++++++-
 1 file changed, 12 insertions(+), 1 deletion(-)

diff --git a/net/core/link_watch.c b/net/core/link_watch.c
index 04fdc95..f153e06 100644
--- a/net/core/link_watch.c
+++ b/net/core/link_watch.c
@@ -163,9 +163,16 @@ static void linkwatch_do_dev(struct net_device *dev)
 
 static void __linkwatch_run_queue(int urgent_only)
 {
+#define MAX_DO_DEV_PER_LOOP	100
+
+	int do_dev = MAX_DO_DEV_PER_LOOP;
 	struct net_device *dev;
 	LIST_HEAD(wrk);
 
+	/* Give urgent case more budget */
+	if (urgent_only)
+		do_dev += MAX_DO_DEV_PER_LOOP;
+
 	/*
 	 * Limit the number of linkwatch events to one
 	 * per second so that a runaway driver does not
@@ -184,7 +191,7 @@ static void __linkwatch_run_queue(int urgent_only)
 	spin_lock_irq(&lweventlist_lock);
 	list_splice_init(&lweventlist, &wrk);
 
-	while (!list_empty(&wrk)) {
+	while (!list_empty(&wrk) && do_dev > 0) {
 
 		dev = list_first_entry(&wrk, struct net_device, link_watch_list);
 		list_del_init(&dev->link_watch_list);
@@ -195,9 +202,13 @@ static void __linkwatch_run_queue(int urgent_only)
 		}
 		spin_unlock_irq(&lweventlist_lock);
 		linkwatch_do_dev(dev);
+		do_dev--;
 		spin_lock_irq(&lweventlist_lock);
 	}
 
+	/* Add the remaining work back to lweventlist */
+	list_splice_init(&wrk, &lweventlist);
+
 	if (!list_empty(&lweventlist))
 		linkwatch_schedule_work(0);
 	spin_unlock_irq(&lweventlist_lock);
-- 
2.8.1