libc/memset: Implement function in assembler
Trace analysis of FVP_Base_AEMv8A model running in
Aarch32 mode with the build options listed below:
TRUSTED_BOARD_BOOT=1 GENERATE_COT=1
ARM_ROTPK_LOCATION=devel_ecdsa KEY_ALG=ecdsa
ROT_KEY=plat/arm/board/common/rotpk/arm_rotprivk_ecdsa.pem
shows that when auth_signature() gets called
71.84% of CPU execution time is spent in memset() function
written in C using single byte write operations,
see lib\libc\memset.c.
This patch replaces C memset() implementation with assembler
version giving the following results:
- for Aarch32 in auth_signature() call memset() CPU time
reduced to 24.84%.
- Number of CPU instructions executed during TF-A
boot stage before start of BL33 in RELEASE builds:
----------------------------------------------
| Arch | C | assembler | % |
----------------------------------------------
| Aarch32 | 2073275460 | 1487400003 | -28.25 |
| Aarch64 | 2056807158 | 1244898303 | -39.47 |
----------------------------------------------
The patch also replaces memset.c with aarch64/memset.S
in plat\nvidia\tegra\platform.mk.
Change-Id: Ifbf085a2f577a25491e2d28446ee95a4ac891597
Signed-off-by: Alexei Fedorov <Alexei.Fedorov@arm.com>
diff --git a/plat/nvidia/tegra/platform.mk b/plat/nvidia/tegra/platform.mk
index a4724e6..5cac46f 100644
--- a/plat/nvidia/tegra/platform.mk
+++ b/plat/nvidia/tegra/platform.mk
@@ -1,5 +1,5 @@
#
-# Copyright (c) 2015-2019, ARM Limited and Contributors. All rights reserved.
+# Copyright (c) 2015-2020, ARM Limited and Contributors. All rights reserved.
# Copyright (c) 2020, NVIDIA Corporation. All rights reserved.
#
# SPDX-License-Identifier: BSD-3-Clause
@@ -69,11 +69,11 @@
# override with necessary libc files for the Tegra platform
override LIBC_SRCS := $(addprefix lib/libc/, \
+ aarch64/memset.S \
aarch64/setjmp.S \
assert.c \
memcpy.c \
memmove.c \
- memset.c \
printf.c \
putchar.c \
strlen.c \