From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <vdavydov.dev@gmail.com>
Date: Fri, 24 Aug 2018 16:07:48 +0300
From: Vladimir Davydov <vdavydov.dev@gmail.com>
Subject: Re: [PATCH 3/3] socket: prevent recvfrom from returning garbage
Message-ID: <20180824130748.duye24rcblum5f4z@esperanza>
References: <cover.1535076888.git.alexander.turenko@tarantool.org>
 <0c53f7379963bb5fb61c97e08d782357bd82f562.1535076888.git.alexander.turenko@tarantool.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <0c53f7379963bb5fb61c97e08d782357bd82f562.1535076888.git.alexander.turenko@tarantool.org>
To: Alexander Turenko <alexander.turenko@tarantool.org>
Cc: tarantool-patches@freelists.org, Yaroslav Dynnikov <yaroslav.dynnikov@tarantool.org>
List-ID: <tarantool-patches.dev.tarantool.org>

On Fri, Aug 24, 2018 at 05:47:39AM +0300, Alexander Turenko wrote:
> In C recvfrom function sets addrlen parameter to zero when called on TCP
> socket (at least on Linux). The src_addr parameter can contain garbage
> in the case, so we should not dereference it.
> 
> Before this commit socket:recvfrom() can return 'from' table with only
> family field (don't sure why, but addr->sa_family often contain PF_INET
> value in my case) or return nil depending on the garbage at the address.
> Now it always return nil.
> ---
>  src/lua/socket.c | 5 +++++
>  1 file changed, 5 insertions(+)

Could you please add a test case?