[PATCH 09/10] crypto: x86/aegis128 - remove unneeded FRAME_BEGIN and FRAME_END

All of lore.kernel.org
 help / color / mirror / Atom feed

From: Eric Biggers <ebiggers@kernel.org>
To: linux-crypto@vger.kernel.org
Cc: x86@kernel.org, Ondrej Mosnacek <omosnace@redhat.com>
Subject: [PATCH 09/10] crypto: x86/aegis128 - remove unneeded FRAME_BEGIN and FRAME_END
Date: Sun,  6 Oct 2024 18:24:29 -0700	[thread overview]
Message-ID: <20241007012430.163606-10-ebiggers@kernel.org> (raw)
In-Reply-To: <20241007012430.163606-1-ebiggers@kernel.org>

From: Eric Biggers <ebiggers@google.com>

Stop using FRAME_BEGIN and FRAME_END in the AEGIS assembly functions,
since all these functions are now leaf functions.  This eliminates some
unnecessary instructions.

Signed-off-by: Eric Biggers <ebiggers@google.com>
---
 arch/x86/crypto/aegis128-aesni-asm.S | 34 ----------------------------
 1 file changed, 34 deletions(-)

diff --git a/arch/x86/crypto/aegis128-aesni-asm.S b/arch/x86/crypto/aegis128-aesni-asm.S
index 583e4515e1f1..e025c6bfadbd 100644
--- a/arch/x86/crypto/aegis128-aesni-asm.S
+++ b/arch/x86/crypto/aegis128-aesni-asm.S
@@ -6,11 +6,10 @@
  * Copyright (C) 2017-2018 Red Hat, Inc. All rights reserved.
  * Copyright 2024 Google LLC
  */
 
 #include <linux/linkage.h>
-#include <asm/frame.h>
 
 #define STATE0	%xmm0
 #define STATE1	%xmm1
 #define STATE2	%xmm2
 #define STATE3	%xmm3
@@ -144,11 +143,10 @@
  */
 SYM_FUNC_START(aegis128_aesni_init)
 	.set STATEP, %rdi
 	.set KEYP, %rsi
 	.set IVP, %rdx
-	FRAME_BEGIN
 
 	/* load IV: */
 	movdqu (IVP), T1
 
 	/* load key: */
@@ -180,12 +178,10 @@ SYM_FUNC_START(aegis128_aesni_init)
 	movdqu STATE0, 0x00(STATEP)
 	movdqu STATE1, 0x10(STATEP)
 	movdqu STATE2, 0x20(STATEP)
 	movdqu STATE3, 0x30(STATEP)
 	movdqu STATE4, 0x40(STATEP)
-
-	FRAME_END
 	RET
 SYM_FUNC_END(aegis128_aesni_init)
 
 /*
  * void aegis128_aesni_ad(struct aegis_state *state, const u8 *data,
@@ -195,11 +191,10 @@ SYM_FUNC_END(aegis128_aesni_init)
  */
 SYM_FUNC_START(aegis128_aesni_ad)
 	.set STATEP, %rdi
 	.set SRC, %rsi
 	.set LEN, %edx
-	FRAME_BEGIN
 
 	test LEN, LEN
 	jz .Lad_out
 
 	/* load the state: */
@@ -249,51 +244,45 @@ SYM_FUNC_START(aegis128_aesni_ad)
 	movdqu STATE0, 0x00(STATEP)
 	movdqu STATE1, 0x10(STATEP)
 	movdqu STATE2, 0x20(STATEP)
 	movdqu STATE3, 0x30(STATEP)
 	movdqu STATE4, 0x40(STATEP)
-	FRAME_END
 	RET
 
 .Lad_out_1:
 	movdqu STATE4, 0x00(STATEP)
 	movdqu STATE0, 0x10(STATEP)
 	movdqu STATE1, 0x20(STATEP)
 	movdqu STATE2, 0x30(STATEP)
 	movdqu STATE3, 0x40(STATEP)
-	FRAME_END
 	RET
 
 .Lad_out_2:
 	movdqu STATE3, 0x00(STATEP)
 	movdqu STATE4, 0x10(STATEP)
 	movdqu STATE0, 0x20(STATEP)
 	movdqu STATE1, 0x30(STATEP)
 	movdqu STATE2, 0x40(STATEP)
-	FRAME_END
 	RET
 
 .Lad_out_3:
 	movdqu STATE2, 0x00(STATEP)
 	movdqu STATE3, 0x10(STATEP)
 	movdqu STATE4, 0x20(STATEP)
 	movdqu STATE0, 0x30(STATEP)
 	movdqu STATE1, 0x40(STATEP)
-	FRAME_END
 	RET
 
 .Lad_out_4:
 	movdqu STATE1, 0x00(STATEP)
 	movdqu STATE2, 0x10(STATEP)
 	movdqu STATE3, 0x20(STATEP)
 	movdqu STATE4, 0x30(STATEP)
 	movdqu STATE0, 0x40(STATEP)
-	FRAME_END
 	RET
 
 .Lad_out:
-	FRAME_END
 	RET
 SYM_FUNC_END(aegis128_aesni_ad)
 
 .macro encrypt_block s0 s1 s2 s3 s4 i
 	movdqu (\i * 0x10)(SRC), MSG
@@ -321,11 +310,10 @@ SYM_FUNC_END(aegis128_aesni_ad)
 SYM_FUNC_START(aegis128_aesni_enc)
 	.set STATEP, %rdi
 	.set SRC, %rsi
 	.set DST, %rdx
 	.set LEN, %ecx
-	FRAME_BEGIN
 
 	/* load the state: */
 	movdqu 0x00(STATEP), STATE0
 	movdqu 0x10(STATEP), STATE1
 	movdqu 0x20(STATEP), STATE2
@@ -349,51 +337,45 @@ SYM_FUNC_START(aegis128_aesni_enc)
 	movdqu STATE4, 0x00(STATEP)
 	movdqu STATE0, 0x10(STATEP)
 	movdqu STATE1, 0x20(STATEP)
 	movdqu STATE2, 0x30(STATEP)
 	movdqu STATE3, 0x40(STATEP)
-	FRAME_END
 	RET
 
 .Lenc_out_1:
 	movdqu STATE3, 0x00(STATEP)
 	movdqu STATE4, 0x10(STATEP)
 	movdqu STATE0, 0x20(STATEP)
 	movdqu STATE1, 0x30(STATEP)
 	movdqu STATE2, 0x40(STATEP)
-	FRAME_END
 	RET
 
 .Lenc_out_2:
 	movdqu STATE2, 0x00(STATEP)
 	movdqu STATE3, 0x10(STATEP)
 	movdqu STATE4, 0x20(STATEP)
 	movdqu STATE0, 0x30(STATEP)
 	movdqu STATE1, 0x40(STATEP)
-	FRAME_END
 	RET
 
 .Lenc_out_3:
 	movdqu STATE1, 0x00(STATEP)
 	movdqu STATE2, 0x10(STATEP)
 	movdqu STATE3, 0x20(STATEP)
 	movdqu STATE4, 0x30(STATEP)
 	movdqu STATE0, 0x40(STATEP)
-	FRAME_END
 	RET
 
 .Lenc_out_4:
 	movdqu STATE0, 0x00(STATEP)
 	movdqu STATE1, 0x10(STATEP)
 	movdqu STATE2, 0x20(STATEP)
 	movdqu STATE3, 0x30(STATEP)
 	movdqu STATE4, 0x40(STATEP)
-	FRAME_END
 	RET
 
 .Lenc_out:
-	FRAME_END
 	RET
 SYM_FUNC_END(aegis128_aesni_enc)
 
 /*
  * void aegis128_aesni_enc_tail(struct aegis_state *state, const u8 *src,
@@ -402,11 +384,10 @@ SYM_FUNC_END(aegis128_aesni_enc)
 SYM_FUNC_START(aegis128_aesni_enc_tail)
 	.set STATEP, %rdi
 	.set SRC, %rsi
 	.set DST, %rdx
 	.set LEN, %ecx	/* {load,store}_partial rely on this being %ecx */
-	FRAME_BEGIN
 
 	/* load the state: */
 	movdqu 0x00(STATEP), STATE0
 	movdqu 0x10(STATEP), STATE1
 	movdqu 0x20(STATEP), STATE2
@@ -434,12 +415,10 @@ SYM_FUNC_START(aegis128_aesni_enc_tail)
 	movdqu STATE4, 0x00(STATEP)
 	movdqu STATE0, 0x10(STATEP)
 	movdqu STATE1, 0x20(STATEP)
 	movdqu STATE2, 0x30(STATEP)
 	movdqu STATE3, 0x40(STATEP)
-
-	FRAME_END
 	RET
 SYM_FUNC_END(aegis128_aesni_enc_tail)
 
 .macro decrypt_block s0 s1 s2 s3 s4 i
 	movdqu (\i * 0x10)(SRC), MSG
@@ -466,11 +445,10 @@ SYM_FUNC_END(aegis128_aesni_enc_tail)
 SYM_FUNC_START(aegis128_aesni_dec)
 	.set STATEP, %rdi
 	.set SRC, %rsi
 	.set DST, %rdx
 	.set LEN, %ecx
-	FRAME_BEGIN
 
 	/* load the state: */
 	movdqu 0x00(STATEP), STATE0
 	movdqu 0x10(STATEP), STATE1
 	movdqu 0x20(STATEP), STATE2
@@ -494,51 +472,45 @@ SYM_FUNC_START(aegis128_aesni_dec)
 	movdqu STATE4, 0x00(STATEP)
 	movdqu STATE0, 0x10(STATEP)
 	movdqu STATE1, 0x20(STATEP)
 	movdqu STATE2, 0x30(STATEP)
 	movdqu STATE3, 0x40(STATEP)
-	FRAME_END
 	RET
 
 .Ldec_out_1:
 	movdqu STATE3, 0x00(STATEP)
 	movdqu STATE4, 0x10(STATEP)
 	movdqu STATE0, 0x20(STATEP)
 	movdqu STATE1, 0x30(STATEP)
 	movdqu STATE2, 0x40(STATEP)
-	FRAME_END
 	RET
 
 .Ldec_out_2:
 	movdqu STATE2, 0x00(STATEP)
 	movdqu STATE3, 0x10(STATEP)
 	movdqu STATE4, 0x20(STATEP)
 	movdqu STATE0, 0x30(STATEP)
 	movdqu STATE1, 0x40(STATEP)
-	FRAME_END
 	RET
 
 .Ldec_out_3:
 	movdqu STATE1, 0x00(STATEP)
 	movdqu STATE2, 0x10(STATEP)
 	movdqu STATE3, 0x20(STATEP)
 	movdqu STATE4, 0x30(STATEP)
 	movdqu STATE0, 0x40(STATEP)
-	FRAME_END
 	RET
 
 .Ldec_out_4:
 	movdqu STATE0, 0x00(STATEP)
 	movdqu STATE1, 0x10(STATEP)
 	movdqu STATE2, 0x20(STATEP)
 	movdqu STATE3, 0x30(STATEP)
 	movdqu STATE4, 0x40(STATEP)
-	FRAME_END
 	RET
 
 .Ldec_out:
-	FRAME_END
 	RET
 SYM_FUNC_END(aegis128_aesni_dec)
 
 /*
  * void aegis128_aesni_dec_tail(struct aegis_state *state, const u8 *src,
@@ -547,11 +519,10 @@ SYM_FUNC_END(aegis128_aesni_dec)
 SYM_FUNC_START(aegis128_aesni_dec_tail)
 	.set STATEP, %rdi
 	.set SRC, %rsi
 	.set DST, %rdx
 	.set LEN, %ecx	/* {load,store}_partial rely on this being %ecx */
-	FRAME_BEGIN
 
 	/* load the state: */
 	movdqu 0x00(STATEP), STATE0
 	movdqu 0x10(STATEP), STATE1
 	movdqu 0x20(STATEP), STATE2
@@ -584,12 +555,10 @@ SYM_FUNC_START(aegis128_aesni_dec_tail)
 	movdqu STATE4, 0x00(STATEP)
 	movdqu STATE0, 0x10(STATEP)
 	movdqu STATE1, 0x20(STATEP)
 	movdqu STATE2, 0x30(STATEP)
 	movdqu STATE3, 0x40(STATEP)
-
-	FRAME_END
 	RET
 SYM_FUNC_END(aegis128_aesni_dec_tail)
 
 /*
  * void aegis128_aesni_final(struct aegis_state *state,
@@ -599,11 +568,10 @@ SYM_FUNC_END(aegis128_aesni_dec_tail)
 SYM_FUNC_START(aegis128_aesni_final)
 	.set STATEP, %rdi
 	.set TAG_XOR, %rsi
 	.set ASSOCLEN, %edx
 	.set CRYPTLEN, %ecx
-	FRAME_BEGIN
 
 	/* load the state: */
 	movdqu 0x00(STATEP), STATE0
 	movdqu 0x10(STATEP), STATE1
 	movdqu 0x20(STATEP), STATE2
@@ -634,9 +602,7 @@ SYM_FUNC_START(aegis128_aesni_final)
 	pxor STATE2, MSG
 	pxor STATE3, MSG
 	pxor STATE4, MSG
 
 	movdqu MSG, (TAG_XOR)
-
-	FRAME_END
 	RET
 SYM_FUNC_END(aegis128_aesni_final)
-- 
2.46.2

next prev parent reply	other threads:[~2024-10-07  1:24 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
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 ` Eric Biggers [this message]
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

find likely ancestor, descendant, or conflicting patches for this message:
( dfblob:583e4515e1f dfblob:e025c6bfadb )
 OR (
bs:"[PATCH 09/10] crypto: x86/aegis128 - remove unneeded FRAME_BEGIN and FRAME_END" )
	(help)

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=20241007012430.163606-10-ebiggers@kernel.org \
    --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 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.