[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20191120180735.GD5604@kadam>
Date: Wed, 20 Nov 2019 21:07:35 +0300
From: Dan Carpenter <dan.carpenter@...cle.com>
To: David Ahern <dsahern@...il.com>
Cc: "David S. Miller" <davem@...emloft.net>,
Jiri Pirko <jiri@...lanox.com>,
Ido Schimmel <idosch@...lanox.com>,
Petr Machata <petrm@...lanox.com>,
Jakub Kicinski <jakub.kicinski@...ronome.com>,
Nikolay Aleksandrov <nikolay@...ulusnetworks.com>,
Parav Pandit <parav@...lanox.com>,
Roopa Prabhu <roopa@...ulusnetworks.com>,
Johannes Berg <johannes.berg@...el.com>,
Michal Kubecek <mkubecek@...e.cz>, netdev@...r.kernel.org,
linux-rdma@...r.kernel.org, kernel-janitors@...r.kernel.org
Subject: Re: [PATCH net] net: rtnetlink: prevent underflows in do_setvfinfo()
On Wed, Nov 20, 2019 at 09:41:01AM -0700, David Ahern wrote:
> On 11/20/19 5:34 AM, Dan Carpenter wrote:
> > I reported this bug to the linux-rdma mailing list in April and this
> > patch wasn't considered very elegant. I can see how that's true. The
> > developer offered to write a fix which would update all the drivers to
> > use u32 throughout. I reminded him in September that this bug still
> > needs to be fixed.
>
> Since the uapi (ifla_vf_mac, ifla_vf_vlan, ...) all have u32, I agree
> with that comment -- it seems like the ndo functions should be changed
> from 'int vf' to 'u32 vf'.
It's a lot of changes and there is no way to be sure the static checker
isn't missing any bugs.
regards,
dan carpenter
Powered by blists - more mailing lists