netdev - Re: [PATCH net v2 2/6] bnxt_en: do not allow wildcard matches for L2 flows

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] [day] [month] [year] [list]

Message-ID: <20180411141928.491dfb6f@cakuba.netronome.com>
Date:   Wed, 11 Apr 2018 14:19:28 -0700
From:   Jakub Kicinski <kubakici@...pl>
To:     Michael Chan <michael.chan@...adcom.com>,
        Andy Gospodarek <andrew.gospodarek@...adcom.com>
Cc:     David Miller <davem@...emloft.net>, Netdev <netdev@...r.kernel.org>
Subject: Re: [PATCH net v2 2/6] bnxt_en: do not allow wildcard matches for
 L2 flows

On Wed, 11 Apr 2018 13:55:11 -0700, Michael Chan wrote:
> On Wed, Apr 11, 2018 at 1:50 PM, Andy Gospodarek wrote:
> > On Wed, Apr 11, 2018 at 01:41:31PM -0700, Michael Chan wrote:  
> > True, but I'm not sure that tc_cls_common_offload is used in all cases.
> > Take red_offload() as one of those.  
> 
> For Flower, we know we have the extack pointer in
> tc_cls_common_offload struct and we can use it to set the netlink
> error message.  The point is that we don't have to modify
> ndo_setup_tc().

Yes, the extack is actually only populated when skip_sw is specified to
avoid warning users who don't care about offloads.

Flower offloads don't go via .ndo_setup_tc but TC block callbacks.  But
one day we will hopefully find a reasonable way to pass extack to qdisc
offloads as well..

FWIW your driver is actually already using extack under the veil of
tc_cls_can_offload_and_chain0() :)