netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Stephen Hemminger <stephen@networkplumber.org>
To: davem@davemloft.net
Cc: netdev@vger.kernel.org, Stephen Hemminger <sthemmin@microsoft.com>
Subject: [PATCH net 3/3] netvsc: fix RCU warning from set_multicast
Date: Mon,  5 Jun 2017 14:10:10 -0700	[thread overview]
Message-ID: <20170605211010.9571-4-sthemmin@microsoft.com> (raw)
In-Reply-To: <20170605211010.9571-1-sthemmin@microsoft.com>

do_set_multicast runs in a work queue and therefore is not holding
RCU read lock. This causes:

WARNING: suspicious RCU usage
4.12.0-rc2-net-00284-ge23454766d55-dirty #2 Not tainted
-----------------------------
drivers/net/hyperv/netvsc_drv.c:65 suspicious rcu_dereference_check() usage!

Fix by acquiring rtnl_lock which is ok in the work queue context.

Better fix is to move do_set_mulitcast into rndis_filter.c but
that has more complex impacts, so hold off until net-next.

Signed-off-by: Stephen Hemminger <sthemmin@microsoft.com>
---
 drivers/net/hyperv/netvsc_drv.c | 34 ++++++++++++++++++----------------
 1 file changed, 18 insertions(+), 16 deletions(-)

diff --git a/drivers/net/hyperv/netvsc_drv.c b/drivers/net/hyperv/netvsc_drv.c
index 252e5d52d17e..ba5c4d037b76 100644
--- a/drivers/net/hyperv/netvsc_drv.c
+++ b/drivers/net/hyperv/netvsc_drv.c
@@ -62,24 +62,26 @@ static void do_set_multicast(struct work_struct *w)
 		container_of(w, struct net_device_context, work);
 	struct hv_device *device_obj = ndevctx->device_ctx;
 	struct net_device *ndev = hv_get_drvdata(device_obj);
-	struct netvsc_device *nvdev = rcu_dereference(ndevctx->nvdev);
+	struct netvsc_device *nvdev;
 	struct rndis_device *rdev;
 
-	if (!nvdev)
-		return;
-
-	rdev = nvdev->extension;
-	if (rdev == NULL)
-		return;
-
-	if (ndev->flags & IFF_PROMISC)
-		rndis_filter_set_packet_filter(rdev,
-			NDIS_PACKET_TYPE_PROMISCUOUS);
-	else
-		rndis_filter_set_packet_filter(rdev,
-			NDIS_PACKET_TYPE_BROADCAST |
-			NDIS_PACKET_TYPE_ALL_MULTICAST |
-			NDIS_PACKET_TYPE_DIRECTED);
+	rtnl_lock();
+	nvdev = rtnl_dereference(ndevctx->nvdev);
+	if (nvdev)
+		rdev = nvdev->extension;
+
+		if (rdev) {
+			if (ndev->flags & IFF_PROMISC)
+				rndis_filter_set_packet_filter(rdev,
+							       NDIS_PACKET_TYPE_PROMISCUOUS);
+			else
+				rndis_filter_set_packet_filter(rdev,
+							       NDIS_PACKET_TYPE_BROADCAST |
+							       NDIS_PACKET_TYPE_ALL_MULTICAST |
+							       NDIS_PACKET_TYPE_DIRECTED);
+		}
+	}
+	rtnl_unlock();
 }
 
 static void netvsc_set_multicast_list(struct net_device *net)
-- 
2.11.0

  parent reply	other threads:[~2017-06-05 21:10 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-06-05 21:10 [PATCH net 0/3] netvsc bug fixes Stephen Hemminger
2017-06-05 21:10 ` [PATCH net 1/3] netvsc: fix rcu dereference warning from ethtool Stephen Hemminger
2017-06-05 21:10 ` [PATCH net 2/3] netvsc: fix net poll mode Stephen Hemminger
2017-06-05 21:10 ` Stephen Hemminger [this message]
2017-06-06  5:17   ` [PATCH net 3/3] netvsc: fix RCU warning from set_multicast kbuild test robot
2017-06-06  5:24   ` kbuild test robot
2017-06-06 19:55   ` David Miller

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20170605211010.9571-4-sthemmin@microsoft.com \
    --to=stephen@networkplumber.org \
    --cc=davem@davemloft.net \
    --cc=netdev@vger.kernel.org \
    --cc=sthemmin@microsoft.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).