[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAEP_g=9f4TVsLAyRt9WkOagYO3-kuEO+TVPjGSr0kB94VsCg7A@mail.gmail.com>
Date: Wed, 20 Mar 2013 17:06:56 -0700
From: Jesse Gross <jesse@...ira.com>
To: David Miller <davem@...emloft.net>
Cc: weiyj.lk@...il.com, yongjun_wei@...ndmicro.com.cn,
dev@...nvswitch.org, netdev@...r.kernel.org
Subject: Re: [PATCH -next] openvswitch: fix error return code in ovs_vport_cmd_set()
On Wed, Mar 20, 2013 at 9:22 AM, David Miller <davem@...emloft.net> wrote:
> From: Jesse Gross <jesse@...ira.com>
> Date: Wed, 20 Mar 2013 08:57:53 -0700
>
>> On Wed, Mar 20, 2013 at 5:14 AM, Wei Yongjun <weiyj.lk@...il.com> wrote:
>>> From: Wei Yongjun <yongjun_wei@...ndmicro.com.cn>
>>>
>>> Fix to return a negative error code from the error handling
>>> case instead of 0, as returned elsewhere in this function.
>>>
>>> Signed-off-by: Wei Yongjun <yongjun_wei@...ndmicro.com.cn>
>>
>> This isn't right because at this point in the function the change has
>> already been made, only the notification allocation failed. Therefore
>> we don't want to return an error since the caller will assume that
>> nothing has changed; instead an error is set on the Netlink socket.
>
> Then the way to fix this is to allocate the netlink SKB first, before
> any config changes occur, then pass that SKB into ovs_vport_cmd_build_info.
>
> We can still get errors, for -EMSGSIZE situations, but I'd say that'd
> be an implementation bug that should be logged. If we aren't allocating
> large enough SKBs for the netlink reply, that's really an internal
> error.
That solves the problem here. I'll take care of it, thanks.
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists