Merge "plat/arm: Introduce and use libc_asm.mk makefile" into integration

lib/libc/aarch32/memset.S

@@ -0,0 +1,74 @@
+/*
+ * Copyright (c) 2020, Arm Limited. All rights reserved.
+ *
+ * SPDX-License-Identifier: BSD-3-Clause
+ */
+
+#include <asm_macros.S>
+
+	.syntax unified
+	.global	memset
+
+/* -----------------------------------------------------------------------
+ * void *memset(void *dst, int val, size_t count)
+ *
+ * Copy the value of 'val' (converted to an unsigned char) into
+ * each of the first 'count' characters of the object pointed to by 'dst'.
+ *
+ * Returns the value of 'dst'.
+ * -----------------------------------------------------------------------
+ */
+func memset
+	mov	r12, r0			/* keep r0 */
+	tst	r0, #3
+	beq	aligned			/* 4-bytes aligned */
+
+	/* Unaligned 'dst' */
+unaligned:
+	subs	r2, r2, #1
+	strbhs	r1, [r12], #1
+	bxls	lr			/* return if 0 */
+	tst	r12, #3
+	bne	unaligned		/* continue while unaligned */
+
+	/* 4-bytes aligned */
+aligned:bfi	r1, r1, #8, #8		/* propagate 'val' */
+	bfi	r1, r1, #16, #16
+
+	mov	r3, r1
+
+	cmp	r2, #16
+	blo	less_16			/* < 16 */
+
+	push	{r4, lr}
+	mov	r4, r1
+	mov	lr, r1
+
+write_32:
+	subs	r2, r2, #32
+	stmiahs	r12!, {r1, r3, r4, lr}
+	stmiahs	r12!, {r1, r3, r4, lr}
+	bhi	write_32		/* write 32 bytes in a loop */
+	popeq	{r4, pc}		/* return if 0 */
+	lsls	r2, r2, #28		/* C = r2[4]; N = r2[3]; Z = r2[3:0] */
+	stmiacs	r12!, {r1, r3, r4, lr}	/* write 16 bytes */
+	popeq	{r4, pc}		/* return if 16 */
+	stmiami	r12!, {r1, r3}		/* write 8 bytes */
+	lsls	r2, r2, #2		/* C = r2[2]; N = r2[1]; Z = r2[1:0] */
+	strcs	r1, [r12], #4		/* write 4 bytes */
+	popeq	{r4, pc}		/* return if 8 or 4 */
+	strhmi	r1, [r12], #2		/* write 2 bytes */
+	lsls	r2, r2, #1		/* N = Z = r2[0] */
+	strbmi	r1, [r12]		/* write 1 byte */
+	pop	{r4, pc}
+
+less_16:lsls	r2, r2, #29		/* C = r2[3]; N = r2[2]; Z = r2[2:0] */
+	stmiacs	r12!, {r1, r3}		/* write 8 bytes */
+	bxeq	lr			/* return if 8 */
+	strmi	r1, [r12], #4		/* write 4 bytes */
+	lsls	r2, r2, #2		/* C = r2[1]; N = Z = r2[0] */
+	strhcs	r1, [r12], #2		/* write 2 bytes */
+	strbmi	r1, [r12]		/* write 1 byte */
+	bx	lr
+
+endfunc memset

lib/libc/aarch64/memset.S

@@ -0,0 +1,64 @@
+/*
+ * Copyright (c) 2020, Arm Limited. All rights reserved.
+ *
+ * SPDX-License-Identifier: BSD-3-Clause
+ */
+
+#include <asm_macros.S>
+
+	.global	memset
+
+/* -----------------------------------------------------------------------
+ * void *memset(void *dst, int val, size_t count)
+ *
+ * Copy the value of 'val' (converted to an unsigned char) into
+ * each of the first 'count' characters of the object pointed to by 'dst'.
+ *
+ * Returns the value of 'dst'.
+ * -----------------------------------------------------------------------
+ */
+func memset
+	cbz	x2, exit		/* exit if 'count' = 0 */
+	mov	x3, x0			/* keep x0 */
+	tst	x0, #7
+	b.eq	aligned			/* 8-bytes aligned */
+
+	/* Unaligned 'dst' */
+unaligned:
+	strb	w1, [x3], #1
+	subs	x2, x2, #1
+	b.eq	exit			/* exit if 0 */
+	tst	x3, #7
+	b.ne	unaligned		/* continue while unaligned */
+
+	/* 8-bytes aligned */
+aligned:cbz	x1, x1_zero
+	bfi	w1, w1, #8, #8		/* propagate 'val' */
+	bfi	w1, w1, #16, #16
+	bfi	x1, x1, #32, #32
+
+x1_zero:ands	x4, x2, #~0x3f
+	b.eq	less_64
+
+write_64:
+	.rept	4
+	stp	x1, x1, [x3], #16	/* write 64 bytes in a loop */
+	.endr
+	subs	x4, x4, #64
+	b.ne	write_64
+less_64:tbz	w2, #5, less_32		/* < 32 bytes */
+	stp	x1, x1, [x3], #16	/* write 32 bytes */
+	stp	x1, x1, [x3], #16
+less_32:tbz	w2, #4, less_16		/* < 16 bytes */
+	stp	x1, x1, [x3], #16	/* write 16 bytes */
+less_16:tbz	w2, #3, less_8		/* < 8 bytes */
+	str	x1, [x3], #8		/* write 8 bytes */
+less_8:	tbz	w2, #2, less_4		/* < 4 bytes */
+	str	w1, [x3], #4		/* write 4 bytes */
+less_4:	tbz	w2, #1, less_2		/* < 2 bytes */
+	strh	w1, [x3], #2		/* write 2 bytes */
+less_2:	tbz	w2, #0, exit
+	strb	w1, [x3]		/* write 1 byte */
+exit:	ret
+
+endfunc	memset

lib/libc/libc_asm.mk

@@ -0,0 +1,38 @@
+#
+# Copyright (c) 2020, Arm Limited. All rights reserved.
+#
+# SPDX-License-Identifier: BSD-3-Clause
+#
+
+LIBC_SRCS	:=	$(addprefix lib/libc/,		\
+			abort.c				\
+			assert.c			\
+			exit.c				\
+			memchr.c			\
+			memcmp.c			\
+			memcpy.c			\
+			memmove.c			\
+			memrchr.c			\
+			printf.c			\
+			putchar.c			\
+			puts.c				\
+			snprintf.c			\
+			strchr.c			\
+			strcmp.c			\
+			strlcpy.c			\
+			strlen.c			\
+			strncmp.c			\
+			strnlen.c			\
+			strrchr.c)
+
+ifeq (${ARCH},aarch64)
+LIBC_SRCS	+=	$(addprefix lib/libc/aarch64/,	\
+			memset.S			\
+			setjmp.S)
+else
+LIBC_SRCS	+=	$(addprefix lib/libc/aarch32/,	\
+			memset.S)
+endif
+
+INCLUDES	+=	-Iinclude/lib/libc		\
+			-Iinclude/lib/libc/$(ARCH)	\

plat/arm/common/arm_common.mk

@@ -121,6 +121,12 @@ endif
 ENABLE_PSCI_STAT		:=	1
 ENABLE_PMF			:=	1
 
+# Override the standard libc with optimised libc_asm
+OVERRIDE_LIBC			:=	1
+ifeq (${OVERRIDE_LIBC},1)
+    include lib/libc/libc_asm.mk
+endif
+
 # On ARM platforms, separate the code and read-only data sections to allow
 # mapping the former as executable and the latter as execute-never.
 SEPARATE_CODE_AND_RODATA	:=	1