From: Eric Biggers <ebiggers@kernel.org>
To: Ondrej Mosnacek <omosnace@redhat.com>
Cc: linux-crypto@vger.kernel.org, x86@kernel.org
Subject: Re: [PATCH 03/10] crypto: x86/aegis128 - eliminate some indirect calls
Date: Tue, 15 Oct 2024 15:43:36 +0000 [thread overview]
Message-ID: <20241015154336.GB2444622@google.com> (raw)
In-Reply-To: <CAFqZXNsoJdJi51CiTxCzsk3Xpt88EeVYDRAzAk8Jgph_DoFKOg@mail.gmail.com>
On Tue, Oct 15, 2024 at 02:41:34PM +0200, Ondrej Mosnacek wrote:
> On Mon, Oct 7, 2024 at 3:33 AM Eric Biggers <ebiggers@kernel.org> wrote:
> >
> > From: Eric Biggers <ebiggers@google.com>
> >
> > Instead of using a struct of function pointers to decide whether to call
> > the encryption or decryption assembly functions, use a conditional
> > branch on a bool. Force-inline the functions to avoid actually
> > generating the branch. This improves performance slightly since
> > indirect calls are slow. Remove the now-unnecessary CFI stubs.
>
> Wouldn't the compiler be able to optimize out the indirect calls
> already if you merely force-inline the functions without the other
> changes? Then again, it's just a few places that grow the if-else, so
> I'm fine with the boolean approach, too.
There's no guarantee that the compiler will actually optimize out the indirect
calls that way.
- Eric
next prev parent reply other threads:[~2024-10-15 15:43 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-10-07 1:24 [PATCH 00/10] AEGIS x86 assembly tuning Eric Biggers
2024-10-07 1:24 ` [PATCH 01/10] crypto: x86/aegis128 - access 32-bit arguments as 32-bit Eric Biggers
2024-10-07 1:24 ` [PATCH 02/10] crypto: x86/aegis128 - remove no-op init and exit functions Eric Biggers
2024-10-07 1:24 ` [PATCH 03/10] crypto: x86/aegis128 - eliminate some indirect calls Eric Biggers
2024-10-15 12:41 ` Ondrej Mosnacek
2024-10-15 15:43 ` Eric Biggers [this message]
2024-10-07 1:24 ` [PATCH 04/10] crypto: x86/aegis128 - don't bother with special code for aligned data Eric Biggers
2024-10-07 1:24 ` [PATCH 05/10] crypto: x86/aegis128 - optimize length block preparation using SSE4.1 Eric Biggers
2024-10-07 1:24 ` [PATCH 06/10] crypto: x86/aegis128 - improve assembly function prototypes Eric Biggers
2024-10-07 1:24 ` [PATCH 07/10] crypto: x86/aegis128 - optimize partial block handling using SSE4.1 Eric Biggers
2024-10-07 1:24 ` [PATCH 08/10] crypto: x86/aegis128 - take advantage of block-aligned len Eric Biggers
2024-10-07 1:24 ` [PATCH 09/10] crypto: x86/aegis128 - remove unneeded FRAME_BEGIN and FRAME_END Eric Biggers
2024-10-07 1:24 ` [PATCH 10/10] crypto: x86/aegis128 - remove unneeded RETs Eric Biggers
2024-10-15 12:48 ` [PATCH 00/10] AEGIS x86 assembly tuning Ondrej Mosnacek
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=20241015154336.GB2444622@google.com \
--to=ebiggers@kernel.org \
--cc=linux-crypto@vger.kernel.org \
--cc=omosnace@redhat.com \
--cc=x86@kernel.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).