netdev - Re: [PATCH v4] iproute2: prevent memory leak

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]

Message-ID: <ZVa2Oha4ahHnYw16@renaissance-vector>
Date: Fri, 17 Nov 2023 01:45:51 +0100
From: Andrea Claudi <aclaudi@...hat.com>
To: Stephen Hemminger <stephen@...workplumber.org>
Cc: heminhong <heminhong@...inos.cn>, petrm@...dia.com,
	netdev@...r.kernel.org
Subject: Re: [PATCH v4] iproute2: prevent memory leak

On Thu, Nov 16, 2023 at 03:05:21PM -0800, Stephen Hemminger wrote:
> On Thu, 16 Nov 2023 11:13:08 +0800
> heminhong <heminhong@...inos.cn> wrote:
> 
> > When the return value of rtnl_talk() is not less than 0,
> > 'answer' will be allocated. The 'answer' should be free
> > after using, otherwise it will cause memory leak.
> > 
> > Signed-off-by: heminhong <heminhong@...inos.cn>
> 
> I am skeptical, what is the code path through rtn_talk() that
> returns non zero, and allocates answer.  If so, that should be fixed
> there.
> 
> In current code, the returns are:
> 	- sendmsg() fails
> 	- recvmsg() fails
> 	- truncated message
> 	
> The paths that set answer are returning 0

IMHO the memory leak is in the same functions this is patching.
For example, in ip/link_gre.c:122 we are effectively returning after
having answer allocated correctly by rtnl_talk().

The confusion here stems from the fact we are jumping into the error
path of rtnl_talk() after rtnl_talk() executed fine.