netdev - Re: [PATCH net-next v2] net: Add sysctl to toggle early demux for tcp and udp

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: <1489951119.16816.3.camel@edumazet-glaptop3.roam.corp.google.com>
Date:   Sun, 19 Mar 2017 12:18:39 -0700
From:   Eric Dumazet <eric.dumazet@...il.com>
To:     Subash Abhinov Kasiviswanathan <subashab@...eaurora.org>
Cc:     Tom Herbert <tom@...bertland.com>,
        Linux Kernel Network Developers <netdev@...r.kernel.org>,
        Stephen Hemminger <stephen@...workplumber.org>,
        netdev-owner@...r.kernel.org
Subject: Re: [PATCH net-next v2] net: Add sysctl to toggle early demux for
 tcp and udp

On Sat, 2017-03-18 at 20:07 -0600, Subash Abhinov Kasiviswanathan wrote:
> > Less than 1% performance improvement in a benchmark doesn't justify
> > the complexity of the patch. Eric's hypothesis was that an unconnected
> > UDP socket may show issues because of cache misses in look-ups due to
> > so many different sources. This should be fairly easy to benchmark by
> > randomly setting source address in your test (IP any and routing my
> > need to be set appropriately).
> > 
> 
> With different source addresses, a larger increase is seen here
> (633->654Mbps).

Yes, also note the extra cache line might be not noticed if the cpu can
keep in its caches the whole UDP hash table (1 MB on 64bit kernels for
the portion touched in IP early demux)

So the exact slowdown depends on CPU generation / cache sizes.