linux-kernel - Re: [PATCH] can: vxcan: Fix memleak in vxcan

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

Date:   Thu, 22 Oct 2020 19:34:14 +0200
From:   Oliver Hartkopp <socketcan@...tkopp.net>
To:     Jakub Kicinski <kuba@...nel.org>
Cc:     Dinghao Liu <dinghao.liu@....edu.cn>, kjlu@....edu,
        Wolfgang Grandegger <wg@...ndegger.com>,
        Marc Kleine-Budde <mkl@...gutronix.de>,
        "David S. Miller" <davem@...emloft.net>, linux-can@...r.kernel.org,
        netdev@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] can: vxcan: Fix memleak in vxcan_newlink



On 22.10.20 18:14, Jakub Kicinski wrote:
> On Wed, 21 Oct 2020 13:20:16 +0200 Oliver Hartkopp wrote:
>> On 21.10.20 07:21, Dinghao Liu wrote:
>>> When rtnl_configure_link() fails, peer needs to be
>>> freed just like when register_netdevice() fails.
>>>
>>> Signed-off-by: Dinghao Liu <dinghao.liu@....edu.cn>
>>
>> Acked-by: Oliver Hartkopp <socketcan@...tkopp.net>
>>
>> Btw. as the vxcan.c driver bases on veth.c the same issue can be found
>> there!
>>
>> At this point:
>> https://elixir.bootlin.com/linux/latest/source/drivers/net/veth.c#L1398
>>
>> err_register_dev:
>>           /* nothing to do */
>> err_configure_peer:
>>           unregister_netdevice(peer);
>>           return err; <<<<<<<<<<<<<<<<<<<<<<<
>>
>> err_register_peer:
>>           free_netdev(peer);
>>           return err;
>> }
>>
>> IMO the return must be removed to fall through the next label and free
>> the netdevice too.
>>
>> Would you like so send a patch for veth.c too?
> 
> Ah, this is where Liu Dinghao got the veth suggestion :)
> 
> Does vxcan actually need this patch?
> 
> static void vxcan_setup(struct net_device *dev)
> {
> 	[...]
>          dev->needs_free_netdev  = true;
> 

Oh!

In fact the vxcan.c is really similar to veth.c in these code snippets - 
so I wondered why this never had been seen in veth.c.

Then vxcan.c doesn't need that patch too :-/

Thanks for the heads up!

@Marc: Can you please make sure that it doesn't get into upstream? Tnx!

Best,
Oliver