netdev - Re: [PATCH V7 mlx5-next 00/16] Add support to get xmit slave

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Mon, 4 May 2020 11:39:11 -0400
From:   Jarod Wilson <jarod@...hat.com>
To:     David Ahern <dsahern@...il.com>
Cc:     Maor Gottlieb <maorg@...lanox.com>,
        "David S. Miller" <davem@...emloft.net>, jgg@...lanox.com,
        Doug Ledford <dledford@...hat.com>, j.vosburgh@...il.com,
        vfalico@...il.com, andy@...yhouse.net, kuba@...nel.org,
        jiri@...lanox.com, dsahern@...nel.org, leonro@...lanox.com,
        saeedm@...lanox.com, linux-rdma@...r.kernel.org,
        Netdev <netdev@...r.kernel.org>, alexr@...lanox.com
Subject: Re: [PATCH V7 mlx5-next 00/16] Add support to get xmit slave

On Mon, May 4, 2020 at 10:42 AM David Ahern <dsahern@...il.com> wrote:
>
> On 5/4/20 8:36 AM, Jarod Wilson wrote:
> > At a glance, I'm not sure why all the "get the xmit slave" functions
> > are being passed an skb. None of them should be manipulating the skb,
> > that should all be done in the respective xmit functions after the
> > slave has been returned.
>
> a number of them select a slave based on hash of the packet data -- from
> ethernet data only to L3+L4 hashing.

Ah, right, I see it now for the 3ad/xor case. So it makes sense to
pass it into the ndo, but for example
bond_xmit_activebackup_slave_get() simply does "return
rcu_dereference(bond->curr_active_slave);", so no real need to pass it
skb. The per-mode functions are all static, could drop the unneeded
passing of skb there. Looks like v8 is already merged though, so meh.


-- 
Jarod Wilson
jarod@...hat.com