From: Ard Biesheuvel <ard.biesheuvel@linaro.org>
To: "Jason A. Donenfeld" <Jason@zx2c4.com>
Cc: Martin Willi <martin@strongswan.org>,
Linux Crypto Mailing List <linux-crypto@vger.kernel.org>
Subject: Re: [PATCH crypto-next v2 1/3] crypto: poly1305 - add new 32 and 64-bit generic versions
Date: Thu, 12 Dec 2019 15:26:22 +0100 [thread overview]
Message-ID: <CAKv+Gu80EVN-_aHPSYUu=0TvFJERBMKFvQS-gce3z_jx=X7www@mail.gmail.com> (raw)
In-Reply-To: <CAHmME9reEXXSmQr+6vPM1cwr+pjvwPwJ5n3UZ0BUSjO2kQQcNg@mail.gmail.com>
On Thu, 12 Dec 2019 at 14:47, Jason A. Donenfeld <Jason@zx2c4.com> wrote:
>
> On Thu, Dec 12, 2019 at 2:08 PM Jason A. Donenfeld <Jason@zx2c4.com> wrote:
> >
> > Hi Martin,
> >
> > On Thu, Dec 12, 2019 at 1:03 PM Martin Willi <martin@strongswan.org> wrote:
> > > Can you provide some numbers to testify that? In my tests, the 32-bit
> > > version gives me exact the same results.
> >
> > On 32-bit, if you only call update() once, then the results are the
> > same. However, as soon as you call it more than once, this new version
> > has increasing gains. Other than that, they should behave pretty much
> > identically.
>
> Oh, you asked for numbers. I just fired up an Armada 370/XP and am
> seeing a 8% increase in performance on calls to the update function.
It would help if we could get some actual numbers. I usually try to
capture the performance delta for a small set of block sizes that are
significant for the use case at hand, e.g., like so [0], and also
include blocksizes that are not 2^n. If the change improves the
general case without causing any significant regressions elsewhere, I
don't think we need to continue this debate.
[0] https://git.kernel.org/pub/scm/linux/kernel/git/herbert/cryptodev-2.6.git/commit/?id=11031c0d7d6e9bca0df233a8acfd6708d2b89470
next prev parent reply other threads:[~2019-12-12 14:26 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-12-11 17:09 [PATCH crypto-next v1] crypto: poly1305 - add new 32 and 64-bit generic versions Jason A. Donenfeld
2019-12-11 19:06 ` Eric Biggers
2019-12-11 22:04 ` Jason A. Donenfeld
2019-12-12 9:30 ` [PATCH crypto-next v2 1/3] " Jason A. Donenfeld
2019-12-12 9:30 ` [PATCH crypto-next v2 2/3] crypto: x86_64/poly1305 - add faster implementations Jason A. Donenfeld
2019-12-12 10:26 ` Jason A. Donenfeld
2019-12-12 15:34 ` Martin Willi
2019-12-12 15:39 ` Jason A. Donenfeld
2019-12-15 17:04 ` Andy Polyakov
2019-12-12 9:30 ` [PATCH crypto-next v2 3/3] crypto: arm/arm64/mips/poly1305 - remove redundant non-reduction from emit Jason A. Donenfeld
2019-12-12 14:59 ` Ard Biesheuvel
2019-12-12 12:03 ` [PATCH crypto-next v2 1/3] crypto: poly1305 - add new 32 and 64-bit generic versions Martin Willi
2019-12-12 13:08 ` Jason A. Donenfeld
2019-12-12 13:46 ` Jason A. Donenfeld
2019-12-12 14:26 ` Ard Biesheuvel [this message]
2019-12-12 14:30 ` Jason A. Donenfeld
2019-12-12 15:30 ` Martin Willi
2019-12-12 15:35 ` Jason A. Donenfeld
2019-12-13 3:28 ` Eric Biggers
2019-12-14 8:56 ` Herbert Xu
2019-12-14 12:21 ` Jason A. Donenfeld
2019-12-14 13:05 ` Jason A. Donenfeld
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='CAKv+Gu80EVN-_aHPSYUu=0TvFJERBMKFvQS-gce3z_jx=X7www@mail.gmail.com' \
--to=ard.biesheuvel@linaro.org \
--cc=Jason@zx2c4.com \
--cc=linux-crypto@vger.kernel.org \
--cc=martin@strongswan.org \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).