[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200915085137.c7xfmk5bpjzniclp@soft-test08>
Date: Tue, 15 Sep 2020 08:51:37 +0000
From: Henrik Bjoernlund <henrik.bjoernlund@...rochip.com>
To: Jakub Kicinski <kuba@...nel.org>
CC: <davem@...emloft.net>, <roopa@...dia.com>, <nikolay@...dia.com>,
<jiri@...lanox.com>, <idosch@...lanox.com>,
<linux-kernel@...r.kernel.org>, <netdev@...r.kernel.org>,
<bridge@...ts.linux-foundation.org>,
<UNGLinuxDriver@...rochip.com>,
Horatiu Vultur <horatiu.vultur@...rochip.com>
Subject: Re: [PATCH RFC 4/7] bridge: cfm: Kernel space implementation of CFM.
Thanks for the review. Comments below.
The 09/04/2020 08:02, Jakub Kicinski wrote:
>
> On Fri, 4 Sep 2020 09:15:24 +0000 Henrik Bjoernlund wrote:
> > + rcu_read_lock();
> > + b_port = rcu_dereference(mep->b_port);
> > + if (!b_port)
> > + return NULL;
> > + skb = dev_alloc_skb(CFM_CCM_MAX_FRAME_LENGTH);
> > + if (!skb)
> > + return NULL;
>
> net/bridge/br_cfm.c:171:23: warning: context imbalance in 'ccm_frame_build' - different lock contexts for basic block
I will assure that rcu_read_unlock() is called before any return.
--
/Henrik
Powered by blists - more mailing lists