netdev - Re: TC stats / hw offload question

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: <ec4092a6-196d-7eca-04be-0654e762c0b2@solarflare.com>
Date:   Wed, 24 Apr 2019 15:05:05 +0100
From:   Edward Cree <ecree@...arflare.com>
To:     Jamal Hadi Salim <jhs@...atatu.com>,
        netdev <netdev@...r.kernel.org>,
        Pablo Neira Ayuso <pablo@...filter.org>
CC:     Jiri Pirko <jiri@...nulli.us>, Cong Wang <xiyou.wangcong@...il.com>
Subject: Re: TC stats / hw offload question

On 06/02/2019 02:20, Jamal Hadi Salim wrote:
> The classifiers dont mod the packets. The actions do. And they
> maintain stats on the size on "entry" i.e pre-edit.
> Each action keeps its own counters. If you did something like:
>
> tc match using flower blah \
> action vlan push tag ... \
> action redirect to egress of eth0
>
> And you submited a packet of size x bytes,
> then the "match" would record x bytes.
> the "vlan action" would record x bytes.
> the "redirect" would record size x+vlaninfo bytes
> the egress of eth0 would  recorr x+vlaninfo bytes
So, I had this working for a while, by calling tcf_action_stats_update()
 directly on the correct element of tc->exts->actions[], instead of using
 tcf_exts_stats_update().  And this was fine, until I tried to port my
 code to the 5.1 kernel, with Pablo's flow action infrastructure.  On that
 it's not possible, because there is only flow_stats_update(), which takes
 a single bytes value for _all_ the actions in the rule.
Pablo, was your patch series[1] intended to change the semantics of TC
 action bytes-counters?  What are the proper semantics now?

-Ed

[1] http://patchwork.ozlabs.org/cover/1035414/