netdev - Re: [patch net-next 10/17] ipv6: fib: Add offload indication to routes

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: <20170719161913.GA10775@splinter>
Date:   Wed, 19 Jul 2017 19:19:13 +0300
From:   Ido Schimmel <idosch@...lanox.com>
To:     David Ahern <dsahern@...il.com>
Cc:     Jiri Pirko <jiri@...nulli.us>, netdev@...r.kernel.org,
        davem@...emloft.net, mlxsw@...lanox.com, roopa@...ulusnetworks.com,
        nikolay@...ulusnetworks.com, kafai@...com,
        hannes@...essinduktion.org, yoshfuji@...ux-ipv6.org,
        edumazet@...gle.com, yanhaishuang@...s.chinamobile.com
Subject: Re: [patch net-next 10/17] ipv6: fib: Add offload indication to
 routes

On Wed, Jul 19, 2017 at 09:53:28AM -0600, David Ahern wrote:
> On 7/19/17 9:49 AM, Ido Schimmel wrote:
> > On Wed, Jul 19, 2017 at 09:27:30AM -0600, David Ahern wrote:
> >> On 7/19/17 1:02 AM, Jiri Pirko wrote:
> >>> Allow user space applications to see which routes are offloaded and
> >>> which aren't by setting the RTNH_F_OFFLOAD flag when dumping them.
> >>>
> >>> To be consistent with IPv4, a multipath route is marked as offloaded if
> >>> one of its nexthops is offloaded. Individual nexthops aren't marked with
> >>> the 'offload' flag.
> >>
> >> It is more user friendly to report the offload per nexthop especially
> >> given the implications. There are already flags per nexthop and those
> >> flags are pushed to userspace so not an API change at all.
> > 
> > I thought about it, but then just decided to be consistent with IPv4.
> 
> And the comment stems from just that. I was looking at IPv4 ECMP routes
> a few days ago and the existence / lack of offload flag was not intuitive.

Understood. I intend to change that.