Received-SPF: pass (google.com: domain of linux-crypto+bounces-3427-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45d1:ec00::1 as permitted sender) client-ip=2604:1380:45d1:ec00::1;
Message-ID: <C0FA88ECA90F43B1BF9E7849C53440D7@H270>
From: "Stefan Kanthak" <stefan.kanthak@nexgo.de>
To: "Eric Biggers" <ebiggers@kernel.org>,
	<linux-crypto@vger.kernel.org>
Cc: <linux-kernel@vger.kernel.org>,
	<ardb@kernel.org>
References: <20240409124216.9261-1-ebiggers@kernel.org> <20240409124216.9261-2-ebiggers@kernel.org>
In-Reply-To: <20240409124216.9261-2-ebiggers@kernel.org>
Subject: Re: [PATCH 1/2] crypto: x86/sha256-ni - convert to use rounds macros
Date: Tue, 9 Apr 2024 18:52:02 +0200
Organization: Me, myself & IT
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain;
	charset="Windows-1252"
Content-Transfer-Encoding: 7bit

"Eric Biggers" <ebiggers@kernel.org> wrote:

> +.macro do_4rounds i, m0, m1, m2, m3
> +.if \i < 16
> +        movdqu  \i*4(DATA_PTR), MSG
> +        pshufb  SHUF_MASK, MSG
> +        movdqa  MSG, \m0
> +.else
> +        movdqa  \m0, MSG
> +.endif
> +        paddd   \i*4(SHA256CONSTANTS), MSG

To load the round constant independent from and parallel to the previous
instructions which use \m0 I recommend to change the first lines of the
do_4rounds macro as follows (this might save 1+ cycle per macro invocation,
and most obviously 2 lines):

.macro do_4rounds i, m0, m1, m2, m3
.if \i < 16
        movdqu  \i*4(DATA_PTR), \m0
        pshufb  SHUF_MASK, \m0
.endif
        movdqa  \i*4(SHA256CONSTANTS), MSG
        paddd   \m0, MSG
...

regards
Stefan