Follow @Openwall on Twitter for new release announcements and other news
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date: 21 Dec 2016 13:07:22 -0500
From: "George Spelvin" <linux@...encehorizons.net>
To: linux@...encehorizons.net, torvalds@...ux-foundation.org
Cc: ak@...ux.intel.com, davem@...emloft.net, David.Laight@...lab.com,
  djb@...yp.to, ebiggers3@...il.com, eric.dumazet@...il.com,
  hannes@...essinduktion.org, Jason@...c4.com,
  jeanphilippe.aumasson@...il.com, kernel-hardening@...ts.openwall.com,
  linux-crypto@...r.kernel.org, linux-kernel@...r.kernel.org,
  luto@...capital.net, netdev@...r.kernel.org, tom@...bertland.com,
  tytso@....edu, vegard.nossum@...il.com
Subject: Re: HalfSipHash Acceptable Usage

Linus wrote:
>> How much does kernel_fpu_begin()/kernel_fpu_end() cost?
>
> It's now better than it used to be, but it's absolutely disastrous
> still. We're talking easily many hundreds of cycles. Under some loads,
> thousands.

I think I've been thoroughly dissuaded, but just to clarify one thing
that resembles a misunderstanding:

> In contrast, in reality, especially with things like "do it once or
> twice per incoming packet", you'll easily hit the absolute worst
> cases, where not only does it take a few hundred cycles to save the FP
> state, you'll then return to user space in between packets, which
> triggers the slow-path return code and reloads the FP state, which is
> another few hundred cycles plus.

Everything being discussed is per-TCP-connection overhead, *not* per
packet.  (Twice for outgoing connections, because one is to generate
the ephemeral port number.)

I know you know this, but I don't want anyone spectating to be confused
about it.

Powered by blists - more mailing lists

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.