netdev - Re: XDP performance regression due to CONFIG

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]

Date:   Tue, 17 Apr 2018 09:07:01 +0200
From:   Jesper Dangaard Brouer <brouer@...hat.com>
To:     Christoph Hellwig <hch@...radead.org>
Cc:     "xdp-newbies@...r.kernel.org" <xdp-newbies@...r.kernel.org>,
        "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        Christoph Hellwig <hch@....de>,
        David Woodhouse <dwmw2@...radead.org>,
        William Tu <u9012063@...il.com>,
        Björn Töpel <bjorn.topel@...el.com>,
        "Karlsson, Magnus" <magnus.karlsson@...el.com>,
        Alexander Duyck <alexander.duyck@...il.com>,
        Arnaldo Carvalho de Melo <acme@...hat.com>,
        brouer@...hat.com
Subject: Re: XDP performance regression due to CONFIG_RETPOLINE Spectre V2

On Mon, 16 Apr 2018 23:15:50 -0700
Christoph Hellwig <hch@...radead.org> wrote:

> On Mon, Apr 16, 2018 at 11:07:04PM +0200, Jesper Dangaard Brouer wrote:
> > On X86 swiotlb fallback (via get_dma_ops -> get_arch_dma_ops) to use
> > x86_swiotlb_dma_ops, instead of swiotlb_dma_ops.  I also included that
> > in below fix patch.  
> 
> x86_swiotlb_dma_ops should not exist any mor, and x86 now uses
> dma_direct_ops.  Looks like you are applying it to an old kernel :)
> 
> > Performance improved to 8.9 Mpps from approx 6.5Mpps.
> > 
> > (This was without my bulking for net_device->ndo_xdp_xmit, so that
> > number should improve more).  
> 
> What is the number for the otherwise comparable setup without repolines?

Approx 12 Mpps.

You forgot to handle the dma_direct_mapping_error() case, which still
used the retpoline in the above 8.9 Mpps measurement, I fixed it up and
performance increased to 9.6 Mpps.

Notice, in this test there are still two retpoline/indirect-calls
left.  The net_device->ndo_xdp_xmit and the invocation of the XDP BPF
prog.

-- 
Best regards,
  Jesper Dangaard Brouer
  MSc.CS, Principal Kernel Engineer at Red Hat
  LinkedIn: http://www.linkedin.com/in/brouer