netdev - Re:Re: [PATCH net] net: Correct wrong skb_flow

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: <64493b13.1231.1634cd0dd6c.Coremail.gfree.wind@vip.163.com>
Date:   Fri, 11 May 2018 09:29:06 +0800 (CST)
From:   "Gao Feng" <gfree.wind@....163.com>
To:     "Eric Dumazet" <eric.dumazet@...il.com>
Cc:     "davem@...emloft.net" <davem@...emloft.net>,
        "daniel@...earbox.net" <daniel@...earbox.net>,
        "jakub.kicinski@...ronome.com" <jakub.kicinski@...ronome.com>,
        "David Ahern" <dsahern@...il.com>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>
Subject: Re:Re: [PATCH net] net: Correct wrong skb_flow_limit check when
 enable RPS

<div style="line-height:1.7;color:#000000;font-size:14px;font-family:Arial"><pre>At 2018-05-11 08:55:47, "Eric Dumazet" &lt;eric.dumazet@...il.com&gt; wrote:
&gt;
&gt;
&gt;On 05/10/2018 05:18 PM, Gao Feng wrote:
&gt;&gt; At 2018-05-10 21:02:55, "Eric Dumazet" &lt;eric.dumazet@...il.com&gt; wrote:
&gt;&gt;&gt;
&gt;&gt;&gt;
&gt;&gt;&gt; On 05/10/2018 01:28 AM, gfree.wind@....163.com wrote:
&gt;&gt;&gt;&gt; From: Gao Feng &lt;gfree.wind@....163.com&gt;
&gt;&gt;&gt;&gt;
&gt;&gt;&gt;&gt; The skb flow limit is implemented for each CPU independently. In the
&gt;&gt;&gt;&gt; current codes, the function skb_flow_limit gets the softnet_data by
&gt;&gt;&gt;&gt; this_cpu_ptr. But the target cpu of enqueue_to_backlog would be not
&gt;&gt;&gt;&gt; the current cpu when enable RPS. As the result, the skb_flow_limit checks
&gt;&gt;&gt;&gt; the stats of current CPU, while the skb is going to append the queue of
&gt;&gt;&gt;&gt; another CPU. It isn't the expected behavior.
&gt;&gt;&gt;&gt;
&gt;&gt;&gt;&gt; Now pass the softnet_data as a param to softnet_data to make consistent.
&gt;&gt;&gt;&gt;
&gt;&gt;&gt;
&gt;&gt;&gt; Please add a correct Fixes: tag
&gt;&gt; 
&gt;&gt; Thanks Eric.
&gt;&gt; 
&gt;&gt; I have one question about the "Fixes: tag".
&gt;&gt; Most of patches are bug fixes, but when need to add the "Fixes: tag", and when not ?
&gt;&gt; 
&gt;&gt; I'm not clear about it. Could you explain it please?
&gt;&gt; 
&gt;
&gt;For this particular patch, since you have not CC Willem (author of the patch),
&gt;I found very useful that you did a search to find out.
&gt;Once you found which commit added the problem, simply add the Fixes: tag and CC: the author.
&gt;
<div>&gt;Doing so saves us (stable teams, reviewers, maintainers) a lot of time really.</div><div><br /></div><div> Normally I get the "to" list by get_maintainer.pl script, now I would save the stable team ASAP.</div>&gt;
<div>&gt;In my opinion, Fixes: tags should be mandatory when applicable.</div><div><br /></div><div>Thanks your explanations, I get it.</div><div><br /></div><div>Best Regards</div><div>Feng</div><div><br /></div>&gt;
&gt;&gt; Best Regards
&gt;&gt; Feng
&gt;&gt; 
&gt;&gt;&gt;
&gt;&gt;&gt; By doing so, you will likely add a CC: tag to make sure the author of the code
&gt;&gt;&gt; will receive your email and give feed back.
&gt;&gt;&gt;
&gt;&gt;&gt; Thanks !
&gt;&gt;&gt;
</pre></div>