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: <58F83C89.4070808@gmail.com>
Date:   Wed, 19 Apr 2017 21:43:53 -0700
From:   John Fastabend <john.fastabend@...il.com>
To:     Andy Gospodarek <andy@...yhouse.net>,
        Jesper Dangaard Brouer <brouer@...hat.com>
Cc:     Daniel Borkmann <borkmann@...earbox.net>,
        Alexei Starovoitov <alexei.starovoitov@...il.com>,
        Alexei Starovoitov <ast@...com>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        "xdp-newbies@...r.kernel.org" <xdp-newbies@...r.kernel.org>
Subject: Re: XDP question: best API for returning/setting egress port?

[...]

> JohnF, any test results with this you can share?  Presumably you tested with
> virtio-net, right?
> 

For my patches using the xdp_redirect with virtio-net showed no measurable
difference from running just the straight XDP_TX in the PPS column. However
virtio_net XDP_TX was quite low to start with (and I was mostly just looking
at functionality so I don't recall if there was a CPU usage hit) so I expect on
ixgbe to see a difference. However I have plenty of cpu overhead to consume on
the 10Gbps card so it will likely just be a CPU % cost not a PPS cost. I'll
bother Alex about the 40Gbps cards ;)

Anyways, I'll put it on my list to rerun. But like the xdp generic pieces it
wont be for a couple days until I get around to it. Sorry need to take care of
a few things first.

Thanks,
John

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ