From: "Jason A. Donenfeld" <Jason@zx2c4.com>
To: Andrew Lunn <andrew@lunn.ch>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>,
Jean-Philippe Aumasson <jeanphilippe.aumasson@gmail.com>,
Netdev <netdev@vger.kernel.org>,
LKML <linux-kernel@vger.kernel.org>,
Russell King - ARM Linux <linux@armlinux.org.uk>,
Samuel Neves <sneves@dei.uc.pt>,
Linux Crypto Mailing List <linux-crypto@vger.kernel.org>,
Andrew Lutomirski <luto@kernel.org>,
Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
David Miller <davem@davemloft.net>,
linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH net-next v6 07/23] zinc: ChaCha20 ARM and ARM64 implementations
Date: Wed, 26 Sep 2018 17:25:48 +0200 [thread overview]
Message-ID: <CAHmME9pm0tsW7ou-M811nsoN0pk2Tb49pH2b0Fzzq_ae6zHR4g@mail.gmail.com> (raw)
In-Reply-To: <20180926143614.GL1676@lunn.ch>
On Wed, Sep 26, 2018 at 4:36 PM Andrew Lunn <andrew@lunn.ch> wrote:
> The wireguard interface claims it is GSO capable. This means the
> network stack will pass it big chunks of data and leave it to the
> network interface to perform the segmentation into 1500 byte MTU
> frames on the wire. I've not looked at how wireguard actually handles
> these big chunks. But to get maximum performance, it should try to
> keep them whole, just add a header and/or trailer. Will wireguard pass
> these big chunks of data to the crypto code? Do we now have 64K blocks
> being worked on? Does the latency jump from 4K to 64K? That might be
> new, so the existing state of the tree does not help you here.
No, it only requests GSO superpackets so that it can group the pieces
and encrypt them on the same core. But they're each encrypted
separately (broken up immediately after ndo_start_xmit), and so they
wind up being ~1420 bytes each to encrypt. I spoke about this at
netdev2.2 if you're interested in the architecture; there's a paper:
https://www.wireguard.com/papers/wireguard-netdev22.pdf
https://www.youtube.com/watch?v=54orFwtQ1XY
https://www.wireguard.com/talks/netdev2017-slides.pdf
WARNING: multiple messages have this Message-ID (diff)
From: Jason@zx2c4.com (Jason A. Donenfeld)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH net-next v6 07/23] zinc: ChaCha20 ARM and ARM64 implementations
Date: Wed, 26 Sep 2018 17:25:48 +0200 [thread overview]
Message-ID: <CAHmME9pm0tsW7ou-M811nsoN0pk2Tb49pH2b0Fzzq_ae6zHR4g@mail.gmail.com> (raw)
In-Reply-To: <20180926143614.GL1676@lunn.ch>
On Wed, Sep 26, 2018 at 4:36 PM Andrew Lunn <andrew@lunn.ch> wrote:
> The wireguard interface claims it is GSO capable. This means the
> network stack will pass it big chunks of data and leave it to the
> network interface to perform the segmentation into 1500 byte MTU
> frames on the wire. I've not looked at how wireguard actually handles
> these big chunks. But to get maximum performance, it should try to
> keep them whole, just add a header and/or trailer. Will wireguard pass
> these big chunks of data to the crypto code? Do we now have 64K blocks
> being worked on? Does the latency jump from 4K to 64K? That might be
> new, so the existing state of the tree does not help you here.
No, it only requests GSO superpackets so that it can group the pieces
and encrypt them on the same core. But they're each encrypted
separately (broken up immediately after ndo_start_xmit), and so they
wind up being ~1420 bytes each to encrypt. I spoke about this at
netdev2.2 if you're interested in the architecture; there's a paper:
https://www.wireguard.com/papers/wireguard-netdev22.pdf
https://www.youtube.com/watch?v=54orFwtQ1XY
https://www.wireguard.com/talks/netdev2017-slides.pdf
next prev parent reply other threads:[~2018-09-26 15:25 UTC|newest]
Thread overview: 213+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-09-25 14:55 [PATCH net-next v6 00/23] WireGuard: Secure Network Tunnel Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 01/23] asm: simd context helper API Jason A. Donenfeld
2018-09-28 8:28 ` Ard Biesheuvel
2018-09-28 8:28 ` Ard Biesheuvel
2018-09-28 8:49 ` Ard Biesheuvel
2018-09-28 8:49 ` Ard Biesheuvel
2018-09-28 13:47 ` Jason A. Donenfeld
2018-09-28 13:52 ` Ard Biesheuvel
2018-09-28 13:59 ` Jason A. Donenfeld
2018-09-28 14:00 ` Ard Biesheuvel
2018-09-28 14:01 ` Jason A. Donenfeld
2018-09-30 4:20 ` Joe Perches
2018-09-30 5:35 ` Andy Lutomirski
2018-10-01 1:43 ` Jason A. Donenfeld
2018-10-02 7:18 ` Ard Biesheuvel
2018-09-28 13:45 ` Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 02/23] zinc: introduce minimal cryptography library Jason A. Donenfeld
2018-09-25 18:33 ` Joe Perches
2018-09-25 19:43 ` Jason A. Donenfeld
2018-09-25 20:00 ` Andy Lutomirski
2018-09-25 20:02 ` Jason A. Donenfeld
2018-09-25 20:05 ` Joe Perches
2018-09-25 20:12 ` Jason A. Donenfeld
2018-09-25 20:21 ` Joe Perches
2018-09-25 20:54 ` Jason A. Donenfeld
2018-09-25 21:02 ` Joe Perches
2018-09-25 21:03 ` Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 03/23] zinc: ChaCha20 generic C implementation and selftest Jason A. Donenfeld
2018-09-28 15:40 ` Ard Biesheuvel
2018-09-28 15:40 ` Ard Biesheuvel
2018-09-29 1:53 ` Jason A. Donenfeld
2018-10-02 3:15 ` Herbert Xu
2018-10-02 3:18 ` Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 04/23] zinc: ChaCha20 x86_64 implementation Jason A. Donenfeld
2018-09-28 15:47 ` Ard Biesheuvel
2018-09-28 15:47 ` Ard Biesheuvel
2018-09-29 2:01 ` Jason A. Donenfeld
2018-09-29 7:56 ` Borislav Petkov
2018-09-29 8:00 ` Ard Biesheuvel
2018-09-29 8:11 ` Borislav Petkov
2018-09-29 8:27 ` Abel Vesa
2018-10-02 1:09 ` Jason A. Donenfeld
2018-10-02 1:07 ` Jason A. Donenfeld
2018-10-02 3:18 ` Herbert Xu
2018-10-02 3:20 ` Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 05/23] zinc: import Andy Polyakov's ChaCha20 ARM and ARM64 implementations Jason A. Donenfeld
2018-09-25 14:56 ` Jason A. Donenfeld
2018-09-28 15:49 ` Ard Biesheuvel
2018-09-28 15:49 ` Ard Biesheuvel
2018-09-28 15:51 ` Ard Biesheuvel
2018-09-28 15:51 ` Ard Biesheuvel
2018-09-28 15:51 ` Ard Biesheuvel
2018-09-28 15:57 ` Jason A. Donenfeld
2018-09-28 15:57 ` Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 06/23] zinc: port " Jason A. Donenfeld
2018-09-25 14:56 ` Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 07/23] zinc: " Jason A. Donenfeld
2018-09-25 14:56 ` Jason A. Donenfeld
2018-09-26 8:59 ` Ard Biesheuvel
2018-09-26 8:59 ` Ard Biesheuvel
2018-09-26 8:59 ` Ard Biesheuvel
2018-09-26 13:32 ` Jason A. Donenfeld
2018-09-26 13:32 ` Jason A. Donenfeld
2018-09-26 14:02 ` Ard Biesheuvel
2018-09-26 14:02 ` Ard Biesheuvel
2018-09-26 14:02 ` Ard Biesheuvel
2018-09-26 15:41 ` Jason A. Donenfeld
2018-09-26 15:41 ` Jason A. Donenfeld
2018-09-26 16:54 ` Ard Biesheuvel
2018-09-26 16:54 ` Ard Biesheuvel
2018-09-26 16:54 ` Ard Biesheuvel
2018-09-26 17:07 ` Jason A. Donenfeld
2018-09-26 17:07 ` Jason A. Donenfeld
2018-09-26 17:37 ` Eric Biggers
2018-09-26 17:37 ` Eric Biggers
2018-09-26 17:46 ` Jason A. Donenfeld
2018-09-26 17:46 ` Jason A. Donenfeld
2018-09-26 15:41 ` Ard Biesheuvel
2018-09-26 15:41 ` Ard Biesheuvel
2018-09-26 15:41 ` Ard Biesheuvel
2018-09-26 15:45 ` Jason A. Donenfeld
2018-09-26 15:45 ` Jason A. Donenfeld
2018-09-26 15:49 ` Jason A. Donenfeld
2018-09-26 15:49 ` Jason A. Donenfeld
2018-09-26 15:51 ` Ard Biesheuvel
2018-09-26 15:51 ` Ard Biesheuvel
2018-09-26 15:51 ` Ard Biesheuvel
2018-09-26 15:58 ` Jason A. Donenfeld
2018-09-26 15:58 ` Jason A. Donenfeld
2018-09-27 0:04 ` Jason A. Donenfeld
2018-09-27 0:04 ` Jason A. Donenfeld
2018-09-27 13:26 ` Jason A. Donenfeld
2018-09-27 13:26 ` Jason A. Donenfeld
2018-09-27 15:19 ` Jason A. Donenfeld
2018-09-27 15:19 ` Jason A. Donenfeld
2018-09-27 15:19 ` Jason A. Donenfeld
2018-09-27 16:26 ` Andy Lutomirski
2018-09-27 16:26 ` Andy Lutomirski
2018-09-27 17:06 ` Jason A. Donenfeld
2018-09-27 17:06 ` Jason A. Donenfeld
2018-09-26 16:21 ` Andy Lutomirski
2018-09-26 16:21 ` Andy Lutomirski
2018-09-26 16:21 ` Andy Lutomirski
2018-09-26 17:03 ` Jason A. Donenfeld
2018-09-26 17:03 ` Jason A. Donenfeld
2018-09-26 17:08 ` Ard Biesheuvel
2018-09-26 17:08 ` Ard Biesheuvel
2018-09-26 17:08 ` Ard Biesheuvel
2018-09-26 17:23 ` Andy Lutomirski
2018-09-26 17:23 ` Andy Lutomirski
2018-09-26 14:36 ` Andrew Lunn
2018-09-26 14:36 ` Andrew Lunn
2018-09-26 15:25 ` Jason A. Donenfeld [this message]
2018-09-26 15:25 ` Jason A. Donenfeld
2018-09-28 16:01 ` Ard Biesheuvel
2018-09-28 16:01 ` Ard Biesheuvel
2018-09-28 16:01 ` Ard Biesheuvel
2018-09-29 2:20 ` Jason A. Donenfeld
2018-09-29 2:20 ` Jason A. Donenfeld
2018-09-29 6:16 ` Ard Biesheuvel
2018-09-29 6:16 ` Ard Biesheuvel
2018-09-30 2:33 ` Jason A. Donenfeld
2018-09-30 2:33 ` Jason A. Donenfeld
2018-09-30 2:33 ` Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 08/23] zinc: ChaCha20 MIPS32r2 implementation Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 09/23] zinc: Poly1305 generic C implementations and selftest Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 10/23] zinc: Poly1305 x86_64 implementation Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 11/23] zinc: import Andy Polyakov's Poly1305 ARM and ARM64 implementations Jason A. Donenfeld
2018-09-25 14:56 ` Jason A. Donenfeld
2018-10-03 6:12 ` Eric Biggers
2018-10-03 6:12 ` Eric Biggers
2018-10-03 7:58 ` Ard Biesheuvel
2018-10-03 7:58 ` Ard Biesheuvel
2018-10-03 7:58 ` Ard Biesheuvel
2018-10-03 14:08 ` Jason A. Donenfeld
2018-10-03 14:08 ` Jason A. Donenfeld
2018-10-03 14:45 ` Jason A. Donenfeld
2018-10-03 14:45 ` Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 12/23] zinc: " Jason A. Donenfeld
2018-09-25 14:56 ` Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 13/23] zinc: Poly1305 MIPS32r2 and MIPS64 implementations Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 14/23] zinc: ChaCha20Poly1305 construction and selftest Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 15/23] zinc: BLAKE2s generic C implementation " Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 16/23] zinc: BLAKE2s x86_64 implementation Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 17/23] zinc: Curve25519 generic C implementations and selftest Jason A. Donenfeld
2018-09-25 18:38 ` Joe Perches
2018-09-25 14:56 ` [PATCH net-next v6 18/23] zinc: Curve25519 x86_64 implementation Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 19/23] zinc: Curve25519 ARM implementation Jason A. Donenfeld
2018-09-25 14:56 ` Jason A. Donenfeld
2018-10-02 16:59 ` Ard Biesheuvel
2018-10-02 16:59 ` Ard Biesheuvel
2018-10-02 16:59 ` Ard Biesheuvel
2018-10-02 21:35 ` Richard Weinberger
2018-10-02 21:35 ` Richard Weinberger
2018-10-03 1:03 ` Jason A. Donenfeld
2018-10-03 1:03 ` Jason A. Donenfeld
2018-10-05 15:05 ` D. J. Bernstein
2018-10-05 15:05 ` D. J. Bernstein
2018-10-05 15:16 ` Ard Biesheuvel
2018-10-05 15:16 ` Ard Biesheuvel
2018-10-05 15:16 ` Ard Biesheuvel
2018-10-05 18:40 ` Jason A. Donenfeld
2018-10-05 18:40 ` Jason A. Donenfeld
2018-10-03 3:10 ` Jason A. Donenfeld
2018-10-03 3:10 ` Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 20/23] crypto: port Poly1305 to Zinc Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 21/23] crypto: port ChaCha20 " Jason A. Donenfeld
2018-10-02 3:26 ` Herbert Xu
2018-10-02 3:31 ` Jason A. Donenfeld
2018-10-03 5:56 ` Eric Biggers
2018-10-03 14:01 ` Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 22/23] security/keys: rewrite big_key crypto to use Zinc Jason A. Donenfeld
2018-09-25 14:56 ` [PATCH net-next v6 23/23] net: WireGuard secure network tunnel Jason A. Donenfeld
2018-09-26 16:00 ` Ivan Labáth
2018-09-26 16:04 ` Jason A. Donenfeld
2018-11-05 13:06 ` Ivan Labáth
2018-11-12 23:53 ` Jason A. Donenfeld
2018-11-13 0:10 ` Dave Taht
2018-11-13 0:13 ` Jason A. Donenfeld
2018-09-27 1:15 ` Andrew Lunn
2018-09-27 22:37 ` Jason A. Donenfeld
2018-09-28 1:09 ` Jason A. Donenfeld
2018-09-28 15:01 ` Andrew Lunn
2018-09-28 15:04 ` Jason A. Donenfeld
2018-10-03 11:15 ` Ard Biesheuvel
2018-10-03 11:15 ` Ard Biesheuvel
2018-10-03 14:12 ` Jason A. Donenfeld
2018-10-03 14:13 ` Ard Biesheuvel
2018-10-03 14:25 ` Ard Biesheuvel
2018-10-03 14:28 ` Jason A. Donenfeld
2018-09-27 18:29 ` [PATCH net-next v6 00/23] WireGuard: Secure Network Tunnel Eric Biggers
2018-09-27 21:35 ` Jason A. Donenfeld
2018-09-28 1:17 ` Eric Biggers
2018-09-28 2:35 ` Jason A. Donenfeld
2018-09-28 4:55 ` Eric Biggers
2018-09-28 5:46 ` Jason A. Donenfeld
2018-09-28 7:52 ` Ard Biesheuvel
2018-09-28 13:40 ` Jason A. Donenfeld
2018-10-02 3:39 ` Herbert Xu
2018-10-02 3:45 ` Jason A. Donenfeld
2018-10-02 3:49 ` Herbert Xu
2018-10-02 6:04 ` Ard Biesheuvel
2018-10-02 6:43 ` Richard Weinberger
2018-10-02 12:22 ` Jason A. Donenfeld
2018-10-03 6:49 ` Eric Biggers
2018-10-05 13:13 ` Jason A. Donenfeld
2018-10-05 13:37 ` Richard Weinberger
2018-10-05 13:46 ` Jason A. Donenfeld
2018-10-05 13:53 ` Richard Weinberger
2018-10-05 17:50 ` David Miller
2018-09-28 17:47 ` Ard Biesheuvel
2018-09-29 2:40 ` Jason A. Donenfeld
2018-09-29 5:35 ` Willy Tarreau
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAHmME9pm0tsW7ou-M811nsoN0pk2Tb49pH2b0Fzzq_ae6zHR4g@mail.gmail.com \
--to=jason@zx2c4.com \
--cc=andrew@lunn.ch \
--cc=ard.biesheuvel@linaro.org \
--cc=davem@davemloft.net \
--cc=gregkh@linuxfoundation.org \
--cc=jeanphilippe.aumasson@gmail.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-crypto@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux@armlinux.org.uk \
--cc=luto@kernel.org \
--cc=netdev@vger.kernel.org \
--cc=sneves@dei.uc.pt \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.