Commit e6dc85f1b4

Andrew Kelley <andrew@ziglang.org>
2025-01-17 08:26:18
remove memcpy and memmove from bundled libcs
These are provided instead by compiler_rt. Part of #2879
1 parent c748eb2
Changed files (15)
lib
libc
musl
wasi
libc-top-half
musl
src
src
lib/libc/musl/src/string/aarch64/memcpy.S
@@ -1,186 +0,0 @@
-/*
- * memcpy - copy memory area
- *
- * Copyright (c) 2012-2020, Arm Limited.
- * SPDX-License-Identifier: MIT
- */
-
-/* Assumptions:
- *
- * ARMv8-a, AArch64, unaligned accesses.
- *
- */
-
-#define dstin   x0
-#define src     x1
-#define count   x2
-#define dst     x3
-#define srcend  x4
-#define dstend  x5
-#define A_l     x6
-#define A_lw    w6
-#define A_h     x7
-#define B_l     x8
-#define B_lw    w8
-#define B_h     x9
-#define C_l     x10
-#define C_lw    w10
-#define C_h     x11
-#define D_l     x12
-#define D_h     x13
-#define E_l     x14
-#define E_h     x15
-#define F_l     x16
-#define F_h     x17
-#define G_l     count
-#define G_h     dst
-#define H_l     src
-#define H_h     srcend
-#define tmp1    x14
-
-/* This implementation of memcpy uses unaligned accesses and branchless
-   sequences to keep the code small, simple and improve performance.
-
-   Copies are split into 3 main cases: small copies of up to 32 bytes, medium
-   copies of up to 128 bytes, and large copies.  The overhead of the overlap
-   check is negligible since it is only required for large copies.
-
-   Large copies use a software pipelined loop processing 64 bytes per iteration.
-   The destination pointer is 16-byte aligned to minimize unaligned accesses.
-   The loop tail is handled by always copying 64 bytes from the end.
-*/
-
-.global memcpy
-.type memcpy,%function
-memcpy:
-	add     srcend, src, count
-	add     dstend, dstin, count
-	cmp     count, 128
-	b.hi    .Lcopy_long
-	cmp     count, 32
-	b.hi    .Lcopy32_128
-
-	/* Small copies: 0..32 bytes.  */
-	cmp     count, 16
-	b.lo    .Lcopy16
-	ldp     A_l, A_h, [src]
-	ldp     D_l, D_h, [srcend, -16]
-	stp     A_l, A_h, [dstin]
-	stp     D_l, D_h, [dstend, -16]
-	ret
-
-	/* Copy 8-15 bytes.  */
-.Lcopy16:
-	tbz     count, 3, .Lcopy8
-	ldr     A_l, [src]
-	ldr     A_h, [srcend, -8]
-	str     A_l, [dstin]
-	str     A_h, [dstend, -8]
-	ret
-
-	.p2align 3
-	/* Copy 4-7 bytes.  */
-.Lcopy8:
-	tbz     count, 2, .Lcopy4
-	ldr     A_lw, [src]
-	ldr     B_lw, [srcend, -4]
-	str     A_lw, [dstin]
-	str     B_lw, [dstend, -4]
-	ret
-
-	/* Copy 0..3 bytes using a branchless sequence.  */
-.Lcopy4:
-	cbz     count, .Lcopy0
-	lsr     tmp1, count, 1
-	ldrb    A_lw, [src]
-	ldrb    C_lw, [srcend, -1]
-	ldrb    B_lw, [src, tmp1]
-	strb    A_lw, [dstin]
-	strb    B_lw, [dstin, tmp1]
-	strb    C_lw, [dstend, -1]
-.Lcopy0:
-	ret
-
-	.p2align 4
-	/* Medium copies: 33..128 bytes.  */
-.Lcopy32_128:
-	ldp     A_l, A_h, [src]
-	ldp     B_l, B_h, [src, 16]
-	ldp     C_l, C_h, [srcend, -32]
-	ldp     D_l, D_h, [srcend, -16]
-	cmp     count, 64
-	b.hi    .Lcopy128
-	stp     A_l, A_h, [dstin]
-	stp     B_l, B_h, [dstin, 16]
-	stp     C_l, C_h, [dstend, -32]
-	stp     D_l, D_h, [dstend, -16]
-	ret
-
-	.p2align 4
-	/* Copy 65..128 bytes.  */
-.Lcopy128:
-	ldp     E_l, E_h, [src, 32]
-	ldp     F_l, F_h, [src, 48]
-	cmp     count, 96
-	b.ls    .Lcopy96
-	ldp     G_l, G_h, [srcend, -64]
-	ldp     H_l, H_h, [srcend, -48]
-	stp     G_l, G_h, [dstend, -64]
-	stp     H_l, H_h, [dstend, -48]
-.Lcopy96:
-	stp     A_l, A_h, [dstin]
-	stp     B_l, B_h, [dstin, 16]
-	stp     E_l, E_h, [dstin, 32]
-	stp     F_l, F_h, [dstin, 48]
-	stp     C_l, C_h, [dstend, -32]
-	stp     D_l, D_h, [dstend, -16]
-	ret
-
-	.p2align 4
-	/* Copy more than 128 bytes.  */
-.Lcopy_long:
-
-	/* Copy 16 bytes and then align dst to 16-byte alignment.  */
-
-	ldp     D_l, D_h, [src]
-	and     tmp1, dstin, 15
-	bic     dst, dstin, 15
-	sub     src, src, tmp1
-	add     count, count, tmp1      /* Count is now 16 too large.  */
-	ldp     A_l, A_h, [src, 16]
-	stp     D_l, D_h, [dstin]
-	ldp     B_l, B_h, [src, 32]
-	ldp     C_l, C_h, [src, 48]
-	ldp     D_l, D_h, [src, 64]!
-	subs    count, count, 128 + 16  /* Test and readjust count.  */
-	b.ls    .Lcopy64_from_end
-
-.Lloop64:
-	stp     A_l, A_h, [dst, 16]
-	ldp     A_l, A_h, [src, 16]
-	stp     B_l, B_h, [dst, 32]
-	ldp     B_l, B_h, [src, 32]
-	stp     C_l, C_h, [dst, 48]
-	ldp     C_l, C_h, [src, 48]
-	stp     D_l, D_h, [dst, 64]!
-	ldp     D_l, D_h, [src, 64]!
-	subs    count, count, 64
-	b.hi    .Lloop64
-
-	/* Write the last iteration and copy 64 bytes from the end.  */
-.Lcopy64_from_end:
-	ldp     E_l, E_h, [srcend, -64]
-	stp     A_l, A_h, [dst, 16]
-	ldp     A_l, A_h, [srcend, -48]
-	stp     B_l, B_h, [dst, 32]
-	ldp     B_l, B_h, [srcend, -32]
-	stp     C_l, C_h, [dst, 48]
-	ldp     C_l, C_h, [srcend, -16]
-	stp     D_l, D_h, [dst, 64]
-	stp     E_l, E_h, [dstend, -64]
-	stp     A_l, A_h, [dstend, -48]
-	stp     B_l, B_h, [dstend, -32]
-	stp     C_l, C_h, [dstend, -16]
-	ret
-
-.size memcpy,.-memcpy
lib/libc/musl/src/string/arm/__aeabi_memcpy.s
@@ -1,45 +0,0 @@
-.syntax unified
-
-.global __aeabi_memcpy8
-.global __aeabi_memcpy4
-.global __aeabi_memcpy
-.global __aeabi_memmove8
-.global __aeabi_memmove4
-.global __aeabi_memmove
-
-.type __aeabi_memcpy8,%function
-.type __aeabi_memcpy4,%function
-.type __aeabi_memcpy,%function
-.type __aeabi_memmove8,%function
-.type __aeabi_memmove4,%function
-.type __aeabi_memmove,%function
-
-__aeabi_memmove8:
-__aeabi_memmove4:
-__aeabi_memmove:
-	cmp   r0, r1
-	bls   3f
-	cmp   r2, #0
-	beq   2f
-	adds  r0, r0, r2
-	adds  r2, r1, r2
-1:	subs  r2, r2, #1
-	ldrb  r3, [r2]
-	subs  r0, r0, #1
-	strb  r3, [r0]
-	cmp   r1, r2
-	bne   1b
-2:	bx    lr
-__aeabi_memcpy8:
-__aeabi_memcpy4:
-__aeabi_memcpy:
-3:	cmp   r2, #0
-	beq   2f
-	adds  r2, r1, r2
-1:	ldrb  r3, [r1]
-	adds  r1, r1, #1
-	strb  r3, [r0]
-	adds  r0, r0, #1
-	cmp   r1, r2
-	bne   1b
-2:	bx    lr
lib/libc/musl/src/string/arm/memcpy.S
@@ -1,479 +0,0 @@
-/*
- * Copyright (C) 2008 The Android Open Source Project
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions
- * are met:
- *  * Redistributions of source code must retain the above copyright
- *    notice, this list of conditions and the following disclaimer.
- *  * Redistributions in binary form must reproduce the above copyright
- *    notice, this list of conditions and the following disclaimer in
- *    the documentation and/or other materials provided with the
- *    distribution.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
- * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
- * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
- * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
- * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
- * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
- * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS
- * OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED
- * AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
- * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
- * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
- * SUCH DAMAGE.
- */
-
-
-/*
- * Optimized memcpy() for ARM.
- *
- * note that memcpy() always returns the destination pointer,
- * so we have to preserve R0.
-  */
-
-/*
- * This file has been modified from the original for use in musl libc.
- * The main changes are: addition of .type memcpy,%function to make the
- * code safely callable from thumb mode, adjusting the return
- * instructions to be compatible with pre-thumb ARM cpus, removal of
- * prefetch code that is not compatible with older cpus and support for
- * building as thumb 2 and big-endian.
- */
-
-.syntax unified
-
-.global memcpy
-.type memcpy,%function
-memcpy:
-	/* The stack must always be 64-bits aligned to be compliant with the
-	 * ARM ABI. Since we have to save R0, we might as well save R4
-	 * which we can use for better pipelining of the reads below
-	 */
-	.fnstart
-	.save       {r0, r4, lr}
-	stmfd       sp!, {r0, r4, lr}
-	/* Making room for r5-r11 which will be spilled later */
-	.pad        #28
-	sub         sp, sp, #28
-
-	/* it simplifies things to take care of len<4 early */
-	cmp     r2, #4
-	blo     copy_last_3_and_return
-
-	/* compute the offset to align the source
-	 * offset = (4-(src&3))&3 = -src & 3
-	 */
-	rsb     r3, r1, #0
-	ands    r3, r3, #3
-	beq     src_aligned
-
-	/* align source to 32 bits. We need to insert 2 instructions between
-	 * a ldr[b|h] and str[b|h] because byte and half-word instructions
-	 * stall 2 cycles.
-	 */
-	movs    r12, r3, lsl #31
-	sub     r2, r2, r3              /* we know that r3 <= r2 because r2 >= 4 */
-	ldrbmi r3, [r1], #1
-	ldrbcs r4, [r1], #1
-	ldrbcs r12,[r1], #1
-	strbmi r3, [r0], #1
-	strbcs r4, [r0], #1
-	strbcs r12,[r0], #1
-
-src_aligned:
-
-	/* see if src and dst are aligned together (congruent) */
-	eor     r12, r0, r1
-	tst     r12, #3
-	bne     non_congruent
-
-	/* Use post-incriment mode for stm to spill r5-r11 to reserved stack
-	 * frame. Don't update sp.
-	 */
-	stmea   sp, {r5-r11}
-
-	/* align the destination to a cache-line */
-	rsb     r3, r0, #0
-	ands    r3, r3, #0x1C
-	beq     congruent_aligned32
-	cmp     r3, r2
-	andhi   r3, r2, #0x1C
-
-	/* conditionnaly copies 0 to 7 words (length in r3) */
-	movs    r12, r3, lsl #28
-	ldmcs   r1!, {r4, r5, r6, r7}           /* 16 bytes */
-	ldmmi   r1!, {r8, r9}                   /*  8 bytes */
-	stmcs   r0!, {r4, r5, r6, r7}
-	stmmi   r0!, {r8, r9}
-	tst     r3, #0x4
-	ldrne   r10,[r1], #4                    /*  4 bytes */
-	strne   r10,[r0], #4
-	sub     r2, r2, r3
-
-congruent_aligned32:
-	/*
-	 * here source is aligned to 32 bytes.
-	 */
-
-cached_aligned32:
-	subs    r2, r2, #32
-	blo     less_than_32_left
-
-	/*
-	 * We preload a cache-line up to 64 bytes ahead. On the 926, this will
-	 * stall only until the requested world is fetched, but the linefill
-	 * continues in the the background.
-	 * While the linefill is going, we write our previous cache-line
-	 * into the write-buffer (which should have some free space).
-	 * When the linefill is done, the writebuffer will
-	 * start dumping its content into memory
-	 *
-	 * While all this is going, we then load a full cache line into
-	 * 8 registers, this cache line should be in the cache by now
-	 * (or partly in the cache).
-	 *
-	 * This code should work well regardless of the source/dest alignment.
-	 *
-	 */
-
-	/* Align the preload register to a cache-line because the cpu does
-	 * "critical word first" (the first word requested is loaded first).
-	 */
-	@ bic           r12, r1, #0x1F
-	@ add           r12, r12, #64
-
-1:      ldmia   r1!, { r4-r11 }
-	subs    r2, r2, #32
-
-	/* 
-	 * NOTE: if r12 is more than 64 ahead of r1, the following ldrhi
-	 * for ARM9 preload will not be safely guarded by the preceding subs.
-	 * When it is safely guarded the only possibility to have SIGSEGV here
-	 * is because the caller overstates the length.
-	 */
-	@ ldrhi         r3, [r12], #32      /* cheap ARM9 preload */
-	stmia   r0!, { r4-r11 }
-	bhs     1b
-
-	add     r2, r2, #32
-
-less_than_32_left:
-	/*
-	 * less than 32 bytes left at this point (length in r2)
-	 */
-
-	/* skip all this if there is nothing to do, which should
-	 * be a common case (if not executed the code below takes
-	 * about 16 cycles)
-	 */
-	tst     r2, #0x1F
-	beq     1f
-
-	/* conditionnaly copies 0 to 31 bytes */
-	movs    r12, r2, lsl #28
-	ldmcs   r1!, {r4, r5, r6, r7}           /* 16 bytes */
-	ldmmi   r1!, {r8, r9}                   /*  8 bytes */
-	stmcs   r0!, {r4, r5, r6, r7}
-	stmmi   r0!, {r8, r9}
-	movs    r12, r2, lsl #30
-	ldrcs   r3, [r1], #4                    /*  4 bytes */
-	ldrhmi r4, [r1], #2                     /*  2 bytes */
-	strcs   r3, [r0], #4
-	strhmi r4, [r0], #2
-	tst     r2, #0x1
-	ldrbne r3, [r1]                         /*  last byte  */
-	strbne r3, [r0]
-
-	/* we're done! restore everything and return */
-1:      ldmfd   sp!, {r5-r11}
-	ldmfd   sp!, {r0, r4, lr}
-	bx      lr
-
-	/********************************************************************/
-
-non_congruent:
-	/*
-	 * here source is aligned to 4 bytes
-	 * but destination is not.
-	 *
-	 * in the code below r2 is the number of bytes read
-	 * (the number of bytes written is always smaller, because we have
-	 * partial words in the shift queue)
-	 */
-	cmp     r2, #4
-	blo     copy_last_3_and_return
-
-	/* Use post-incriment mode for stm to spill r5-r11 to reserved stack
-	 * frame. Don't update sp.
-	 */
-	stmea   sp, {r5-r11}
-
-	/* compute shifts needed to align src to dest */
-	rsb     r5, r0, #0
-	and     r5, r5, #3                      /* r5 = # bytes in partial words */
-	mov     r12, r5, lsl #3         /* r12 = right */
-	rsb     lr, r12, #32            /* lr = left  */
-
-	/* read the first word */
-	ldr     r3, [r1], #4
-	sub     r2, r2, #4
-
-	/* write a partial word (0 to 3 bytes), such that destination
-	 * becomes aligned to 32 bits (r5 = nb of words to copy for alignment)
-	 */
-	movs    r5, r5, lsl #31
-
-#if __ARMEB__
-	movmi   r3, r3, ror #24
-	strbmi	r3, [r0], #1
-	movcs   r3, r3, ror #24
-	strbcs	r3, [r0], #1
-	movcs   r3, r3, ror #24
-	strbcs	r3, [r0], #1
-#else
-	strbmi r3, [r0], #1
-	movmi   r3, r3, lsr #8
-	strbcs r3, [r0], #1
-	movcs   r3, r3, lsr #8
-	strbcs r3, [r0], #1
-	movcs   r3, r3, lsr #8
-#endif
-
-	cmp     r2, #4
-	blo     partial_word_tail
-
-#if __ARMEB__
-	mov	r3, r3, lsr r12
-	mov	r3, r3, lsl r12
-#endif
-
-	/* Align destination to 32 bytes (cache line boundary) */
-1:      tst     r0, #0x1c
-	beq     2f
-	ldr     r5, [r1], #4
-	sub     r2, r2, #4
-#if __ARMEB__
-	mov     r4, r5,                 lsr lr
-	orr     r4, r4, r3
-	mov     r3, r5,                 lsl r12
-#else
-	mov     r4, r5,                 lsl lr
-	orr     r4, r4, r3
-	mov     r3, r5,                 lsr r12
-#endif
-	str     r4, [r0], #4
-	cmp     r2, #4
-	bhs     1b
-	blo     partial_word_tail
-
-	/* copy 32 bytes at a time */
-2:      subs    r2, r2, #32
-	blo     less_than_thirtytwo
-
-	/* Use immediate mode for the shifts, because there is an extra cycle
-	 * for register shifts, which could account for up to 50% of
-	 * performance hit.
-	 */
-
-	cmp     r12, #24
-	beq     loop24
-	cmp     r12, #8
-	beq     loop8
-
-loop16:
-	ldr     r12, [r1], #4
-1:      mov     r4, r12
-	ldmia   r1!, {   r5,r6,r7,  r8,r9,r10,r11}
-	subs    r2, r2, #32
-	ldrhs   r12, [r1], #4
-#if __ARMEB__
-	orr     r3, r3, r4, lsr #16
-	mov     r4, r4, lsl #16
-	orr     r4, r4, r5, lsr #16
-	mov     r5, r5, lsl #16
-	orr     r5, r5, r6, lsr #16
-	mov     r6, r6, lsl #16
-	orr     r6, r6, r7, lsr #16
-	mov     r7, r7, lsl #16
-	orr     r7, r7, r8, lsr #16
-	mov     r8, r8, lsl #16
-	orr     r8, r8, r9, lsr #16
-	mov     r9, r9, lsl #16
-	orr     r9, r9, r10, lsr #16
-	mov     r10, r10,               lsl #16
-	orr     r10, r10, r11, lsr #16
-	stmia   r0!, {r3,r4,r5,r6, r7,r8,r9,r10}
-	mov     r3, r11, lsl #16
-#else
-	orr     r3, r3, r4, lsl #16
-	mov     r4, r4, lsr #16
-	orr     r4, r4, r5, lsl #16
-	mov     r5, r5, lsr #16
-	orr     r5, r5, r6, lsl #16
-	mov     r6, r6, lsr #16
-	orr     r6, r6, r7, lsl #16
-	mov     r7, r7, lsr #16
-	orr     r7, r7, r8, lsl #16
-	mov     r8, r8, lsr #16
-	orr     r8, r8, r9, lsl #16
-	mov     r9, r9, lsr #16
-	orr     r9, r9, r10, lsl #16
-	mov     r10, r10,               lsr #16
-	orr     r10, r10, r11, lsl #16
-	stmia   r0!, {r3,r4,r5,r6, r7,r8,r9,r10}
-	mov     r3, r11, lsr #16
-#endif
-	bhs     1b
-	b       less_than_thirtytwo
-
-loop8:
-	ldr     r12, [r1], #4
-1:      mov     r4, r12
-	ldmia   r1!, {   r5,r6,r7,  r8,r9,r10,r11}
-	subs    r2, r2, #32
-	ldrhs   r12, [r1], #4
-#if __ARMEB__
-	orr     r3, r3, r4, lsr #24
-	mov     r4, r4, lsl #8
-	orr     r4, r4, r5, lsr #24
-	mov     r5, r5, lsl #8
-	orr     r5, r5, r6, lsr #24
-	mov     r6, r6,  lsl #8
-	orr     r6, r6, r7, lsr #24
-	mov     r7, r7,  lsl #8
-	orr     r7, r7, r8,             lsr #24
-	mov     r8, r8,  lsl #8
-	orr     r8, r8, r9,             lsr #24
-	mov     r9, r9,  lsl #8
-	orr     r9, r9, r10,    lsr #24
-	mov     r10, r10, lsl #8
-	orr     r10, r10, r11,  lsr #24
-	stmia   r0!, {r3,r4,r5,r6, r7,r8,r9,r10}
-	mov     r3, r11, lsl #8
-#else
-	orr     r3, r3, r4, lsl #24
-	mov     r4, r4, lsr #8
-	orr     r4, r4, r5, lsl #24
-	mov     r5, r5, lsr #8
-	orr     r5, r5, r6, lsl #24
-	mov     r6, r6,  lsr #8
-	orr     r6, r6, r7, lsl #24
-	mov     r7, r7,  lsr #8
-	orr     r7, r7, r8,             lsl #24
-	mov     r8, r8,  lsr #8
-	orr     r8, r8, r9,             lsl #24
-	mov     r9, r9,  lsr #8
-	orr     r9, r9, r10,    lsl #24
-	mov     r10, r10, lsr #8
-	orr     r10, r10, r11,  lsl #24
-	stmia   r0!, {r3,r4,r5,r6, r7,r8,r9,r10}
-	mov     r3, r11, lsr #8
-#endif
-	bhs     1b
-	b       less_than_thirtytwo
-
-loop24:
-	ldr     r12, [r1], #4
-1:      mov     r4, r12
-	ldmia   r1!, {   r5,r6,r7,  r8,r9,r10,r11}
-	subs    r2, r2, #32
-	ldrhs   r12, [r1], #4
-#if __ARMEB__
-	orr     r3, r3, r4, lsr #8
-	mov     r4, r4, lsl #24
-	orr     r4, r4, r5, lsr #8
-	mov     r5, r5, lsl #24
-	orr     r5, r5, r6, lsr #8
-	mov     r6, r6, lsl #24
-	orr     r6, r6, r7, lsr #8
-	mov     r7, r7, lsl #24
-	orr     r7, r7, r8, lsr #8
-	mov     r8, r8, lsl #24
-	orr     r8, r8, r9, lsr #8
-	mov     r9, r9, lsl #24
-	orr     r9, r9, r10, lsr #8
-	mov     r10, r10, lsl #24
-	orr     r10, r10, r11, lsr #8
-	stmia   r0!, {r3,r4,r5,r6, r7,r8,r9,r10}
-	mov     r3, r11, lsl #24
-#else
-	orr     r3, r3, r4, lsl #8
-	mov     r4, r4, lsr #24
-	orr     r4, r4, r5, lsl #8
-	mov     r5, r5, lsr #24
-	orr     r5, r5, r6, lsl #8
-	mov     r6, r6, lsr #24
-	orr     r6, r6, r7, lsl #8
-	mov     r7, r7, lsr #24
-	orr     r7, r7, r8, lsl #8
-	mov     r8, r8, lsr #24
-	orr     r8, r8, r9, lsl #8
-	mov     r9, r9, lsr #24
-	orr     r9, r9, r10, lsl #8
-	mov     r10, r10, lsr #24
-	orr     r10, r10, r11, lsl #8
-	stmia   r0!, {r3,r4,r5,r6, r7,r8,r9,r10}
-	mov     r3, r11, lsr #24
-#endif
-	bhs     1b
-
-less_than_thirtytwo:
-	/* copy the last 0 to 31 bytes of the source */
-	rsb     r12, lr, #32            /* we corrupted r12, recompute it  */
-	add     r2, r2, #32
-	cmp     r2, #4
-	blo     partial_word_tail
-
-1:      ldr     r5, [r1], #4
-	sub     r2, r2, #4
-#if __ARMEB__
-	mov     r4, r5,                 lsr lr
-	orr     r4, r4, r3
-	mov     r3,     r5,                     lsl r12
-#else
-	mov     r4, r5,                 lsl lr
-	orr     r4, r4, r3
-	mov     r3,     r5,                     lsr r12
-#endif
-	str     r4, [r0], #4
-	cmp     r2, #4
-	bhs     1b
-
-partial_word_tail:
-	/* we have a partial word in the input buffer */
-	movs    r5, lr, lsl #(31-3)
-#if __ARMEB__
-	movmi   r3, r3, ror #24
-	strbmi r3, [r0], #1
-	movcs   r3, r3, ror #24
-	strbcs r3, [r0], #1
-	movcs   r3, r3, ror #24
-	strbcs r3, [r0], #1
-#else
-	strbmi r3, [r0], #1
-	movmi   r3, r3, lsr #8
-	strbcs r3, [r0], #1
-	movcs   r3, r3, lsr #8
-	strbcs r3, [r0], #1
-#endif
-
-	/* Refill spilled registers from the stack. Don't update sp. */
-	ldmfd   sp, {r5-r11}
-
-copy_last_3_and_return:
-	movs    r2, r2, lsl #31 /* copy remaining 0, 1, 2 or 3 bytes */
-	ldrbmi r2, [r1], #1
-	ldrbcs r3, [r1], #1
-	ldrbcs r12,[r1]
-	strbmi r2, [r0], #1
-	strbcs r3, [r0], #1
-	strbcs r12,[r0]
-
-	/* we're done! restore sp and spilled registers and return */
-	add     sp,  sp, #28
-	ldmfd   sp!, {r0, r4, lr}
-	bx      lr
-
lib/libc/musl/src/string/i386/memcpy.s
@@ -1,32 +0,0 @@
-.global memcpy
-.global __memcpy_fwd
-.hidden __memcpy_fwd
-.type memcpy,@function
-memcpy:
-__memcpy_fwd:
-	push %esi
-	push %edi
-	mov 12(%esp),%edi
-	mov 16(%esp),%esi
-	mov 20(%esp),%ecx
-	mov %edi,%eax
-	cmp $4,%ecx
-	jc 1f
-	test $3,%edi
-	jz 1f
-2:	movsb
-	dec %ecx
-	test $3,%edi
-	jnz 2b
-1:	mov %ecx,%edx
-	shr $2,%ecx
-	rep
-	movsl
-	and $3,%edx
-	jz 1f
-2:	movsb
-	dec %edx
-	jnz 2b
-1:	pop %edi
-	pop %esi
-	ret
lib/libc/musl/src/string/i386/memmove.s
@@ -1,22 +0,0 @@
-.global memmove
-.type memmove,@function
-memmove:
-	mov 4(%esp),%eax
-	sub 8(%esp),%eax
-	cmp 12(%esp),%eax
-.hidden __memcpy_fwd
-	jae __memcpy_fwd
-	push %esi
-	push %edi
-	mov 12(%esp),%edi
-	mov 16(%esp),%esi
-	mov 20(%esp),%ecx
-	lea -1(%edi,%ecx),%edi
-	lea -1(%esi,%ecx),%esi
-	std
-	rep movsb
-	cld
-	lea 1(%edi),%eax
-	pop %edi
-	pop %esi
-	ret
lib/libc/musl/src/string/x86_64/memcpy.s
@@ -1,25 +0,0 @@
-.global memcpy
-.global __memcpy_fwd
-.hidden __memcpy_fwd
-.type memcpy,@function
-memcpy:
-__memcpy_fwd:
-	mov %rdi,%rax
-	cmp $8,%rdx
-	jc 1f
-	test $7,%edi
-	jz 1f
-2:	movsb
-	dec %rdx
-	test $7,%edi
-	jnz 2b
-1:	mov %rdx,%rcx
-	shr $3,%rcx
-	rep
-	movsq
-	and $7,%edx
-	jz 1f
-2:	movsb
-	dec %edx
-	jnz 2b
-1:	ret
lib/libc/musl/src/string/x86_64/memmove.s
@@ -1,16 +0,0 @@
-.global memmove
-.type memmove,@function
-memmove:
-	mov %rdi,%rax
-	sub %rsi,%rax
-	cmp %rdx,%rax
-.hidden __memcpy_fwd
-	jae __memcpy_fwd
-	mov %rdx,%rcx
-	lea -1(%rdi,%rdx),%rdi
-	lea -1(%rsi,%rdx),%rsi
-	std
-	rep movsb
-	cld
-	lea 1(%rdi),%rax
-	ret
lib/libc/musl/src/string/memcpy.c
@@ -1,124 +0,0 @@
-#include <string.h>
-#include <stdint.h>
-#include <endian.h>
-
-void *memcpy(void *restrict dest, const void *restrict src, size_t n)
-{
-	unsigned char *d = dest;
-	const unsigned char *s = src;
-
-#ifdef __GNUC__
-
-#if __BYTE_ORDER == __LITTLE_ENDIAN
-#define LS >>
-#define RS <<
-#else
-#define LS <<
-#define RS >>
-#endif
-
-	typedef uint32_t __attribute__((__may_alias__)) u32;
-	uint32_t w, x;
-
-	for (; (uintptr_t)s % 4 && n; n--) *d++ = *s++;
-
-	if ((uintptr_t)d % 4 == 0) {
-		for (; n>=16; s+=16, d+=16, n-=16) {
-			*(u32 *)(d+0) = *(u32 *)(s+0);
-			*(u32 *)(d+4) = *(u32 *)(s+4);
-			*(u32 *)(d+8) = *(u32 *)(s+8);
-			*(u32 *)(d+12) = *(u32 *)(s+12);
-		}
-		if (n&8) {
-			*(u32 *)(d+0) = *(u32 *)(s+0);
-			*(u32 *)(d+4) = *(u32 *)(s+4);
-			d += 8; s += 8;
-		}
-		if (n&4) {
-			*(u32 *)(d+0) = *(u32 *)(s+0);
-			d += 4; s += 4;
-		}
-		if (n&2) {
-			*d++ = *s++; *d++ = *s++;
-		}
-		if (n&1) {
-			*d = *s;
-		}
-		return dest;
-	}
-
-	if (n >= 32) switch ((uintptr_t)d % 4) {
-	case 1:
-		w = *(u32 *)s;
-		*d++ = *s++;
-		*d++ = *s++;
-		*d++ = *s++;
-		n -= 3;
-		for (; n>=17; s+=16, d+=16, n-=16) {
-			x = *(u32 *)(s+1);
-			*(u32 *)(d+0) = (w LS 24) | (x RS 8);
-			w = *(u32 *)(s+5);
-			*(u32 *)(d+4) = (x LS 24) | (w RS 8);
-			x = *(u32 *)(s+9);
-			*(u32 *)(d+8) = (w LS 24) | (x RS 8);
-			w = *(u32 *)(s+13);
-			*(u32 *)(d+12) = (x LS 24) | (w RS 8);
-		}
-		break;
-	case 2:
-		w = *(u32 *)s;
-		*d++ = *s++;
-		*d++ = *s++;
-		n -= 2;
-		for (; n>=18; s+=16, d+=16, n-=16) {
-			x = *(u32 *)(s+2);
-			*(u32 *)(d+0) = (w LS 16) | (x RS 16);
-			w = *(u32 *)(s+6);
-			*(u32 *)(d+4) = (x LS 16) | (w RS 16);
-			x = *(u32 *)(s+10);
-			*(u32 *)(d+8) = (w LS 16) | (x RS 16);
-			w = *(u32 *)(s+14);
-			*(u32 *)(d+12) = (x LS 16) | (w RS 16);
-		}
-		break;
-	case 3:
-		w = *(u32 *)s;
-		*d++ = *s++;
-		n -= 1;
-		for (; n>=19; s+=16, d+=16, n-=16) {
-			x = *(u32 *)(s+3);
-			*(u32 *)(d+0) = (w LS 8) | (x RS 24);
-			w = *(u32 *)(s+7);
-			*(u32 *)(d+4) = (x LS 8) | (w RS 24);
-			x = *(u32 *)(s+11);
-			*(u32 *)(d+8) = (w LS 8) | (x RS 24);
-			w = *(u32 *)(s+15);
-			*(u32 *)(d+12) = (x LS 8) | (w RS 24);
-		}
-		break;
-	}
-	if (n&16) {
-		*d++ = *s++; *d++ = *s++; *d++ = *s++; *d++ = *s++;
-		*d++ = *s++; *d++ = *s++; *d++ = *s++; *d++ = *s++;
-		*d++ = *s++; *d++ = *s++; *d++ = *s++; *d++ = *s++;
-		*d++ = *s++; *d++ = *s++; *d++ = *s++; *d++ = *s++;
-	}
-	if (n&8) {
-		*d++ = *s++; *d++ = *s++; *d++ = *s++; *d++ = *s++;
-		*d++ = *s++; *d++ = *s++; *d++ = *s++; *d++ = *s++;
-	}
-	if (n&4) {
-		*d++ = *s++; *d++ = *s++; *d++ = *s++; *d++ = *s++;
-	}
-	if (n&2) {
-		*d++ = *s++; *d++ = *s++;
-	}
-	if (n&1) {
-		*d = *s;
-	}
-	return dest;
-#endif
-
-	for (; n; n--) *d++ = *s++;
-	return dest;
-}
lib/libc/musl/src/string/memmove.c
@@ -1,42 +0,0 @@
-#include <string.h>
-#include <stdint.h>
-
-#ifdef __GNUC__
-typedef __attribute__((__may_alias__)) size_t WT;
-#define WS (sizeof(WT))
-#endif
-
-void *memmove(void *dest, const void *src, size_t n)
-{
-	char *d = dest;
-	const char *s = src;
-
-	if (d==s) return d;
-	if ((uintptr_t)s-(uintptr_t)d-n <= -2*n) return memcpy(d, s, n);
-
-	if (d<s) {
-#ifdef __GNUC__
-		if ((uintptr_t)s % WS == (uintptr_t)d % WS) {
-			while ((uintptr_t)d % WS) {
-				if (!n--) return dest;
-				*d++ = *s++;
-			}
-			for (; n>=WS; n-=WS, d+=WS, s+=WS) *(WT *)d = *(WT *)s;
-		}
-#endif
-		for (; n; n--) *d++ = *s++;
-	} else {
-#ifdef __GNUC__
-		if ((uintptr_t)s % WS == (uintptr_t)d % WS) {
-			while ((uintptr_t)(d+n) % WS) {
-				if (!n--) return dest;
-				d[n] = s[n];
-			}
-			while (n>=WS) n-=WS, *(WT *)(d+n) = *(WT *)(s+n);
-		}
-#endif
-		while (n) n--, d[n] = s[n];
-	}
-
-	return dest;
-}
lib/libc/wasi/libc-top-half/musl/src/string/aarch64/memcpy.S
@@ -1,186 +0,0 @@
-/*
- * memcpy - copy memory area
- *
- * Copyright (c) 2012-2020, Arm Limited.
- * SPDX-License-Identifier: MIT
- */
-
-/* Assumptions:
- *
- * ARMv8-a, AArch64, unaligned accesses.
- *
- */
-
-#define dstin   x0
-#define src     x1
-#define count   x2
-#define dst     x3
-#define srcend  x4
-#define dstend  x5
-#define A_l     x6
-#define A_lw    w6
-#define A_h     x7
-#define B_l     x8
-#define B_lw    w8
-#define B_h     x9
-#define C_l     x10
-#define C_lw    w10
-#define C_h     x11
-#define D_l     x12
-#define D_h     x13
-#define E_l     x14
-#define E_h     x15
-#define F_l     x16
-#define F_h     x17
-#define G_l     count
-#define G_h     dst
-#define H_l     src
-#define H_h     srcend
-#define tmp1    x14
-
-/* This implementation of memcpy uses unaligned accesses and branchless
-   sequences to keep the code small, simple and improve performance.
-
-   Copies are split into 3 main cases: small copies of up to 32 bytes, medium
-   copies of up to 128 bytes, and large copies.  The overhead of the overlap
-   check is negligible since it is only required for large copies.
-
-   Large copies use a software pipelined loop processing 64 bytes per iteration.
-   The destination pointer is 16-byte aligned to minimize unaligned accesses.
-   The loop tail is handled by always copying 64 bytes from the end.
-*/
-
-.global memcpy
-.type memcpy,%function
-memcpy:
-	add     srcend, src, count
-	add     dstend, dstin, count
-	cmp     count, 128
-	b.hi    .Lcopy_long
-	cmp     count, 32
-	b.hi    .Lcopy32_128
-
-	/* Small copies: 0..32 bytes.  */
-	cmp     count, 16
-	b.lo    .Lcopy16
-	ldp     A_l, A_h, [src]
-	ldp     D_l, D_h, [srcend, -16]
-	stp     A_l, A_h, [dstin]
-	stp     D_l, D_h, [dstend, -16]
-	ret
-
-	/* Copy 8-15 bytes.  */
-.Lcopy16:
-	tbz     count, 3, .Lcopy8
-	ldr     A_l, [src]
-	ldr     A_h, [srcend, -8]
-	str     A_l, [dstin]
-	str     A_h, [dstend, -8]
-	ret
-
-	.p2align 3
-	/* Copy 4-7 bytes.  */
-.Lcopy8:
-	tbz     count, 2, .Lcopy4
-	ldr     A_lw, [src]
-	ldr     B_lw, [srcend, -4]
-	str     A_lw, [dstin]
-	str     B_lw, [dstend, -4]
-	ret
-
-	/* Copy 0..3 bytes using a branchless sequence.  */
-.Lcopy4:
-	cbz     count, .Lcopy0
-	lsr     tmp1, count, 1
-	ldrb    A_lw, [src]
-	ldrb    C_lw, [srcend, -1]
-	ldrb    B_lw, [src, tmp1]
-	strb    A_lw, [dstin]
-	strb    B_lw, [dstin, tmp1]
-	strb    C_lw, [dstend, -1]
-.Lcopy0:
-	ret
-
-	.p2align 4
-	/* Medium copies: 33..128 bytes.  */
-.Lcopy32_128:
-	ldp     A_l, A_h, [src]
-	ldp     B_l, B_h, [src, 16]
-	ldp     C_l, C_h, [srcend, -32]
-	ldp     D_l, D_h, [srcend, -16]
-	cmp     count, 64
-	b.hi    .Lcopy128
-	stp     A_l, A_h, [dstin]
-	stp     B_l, B_h, [dstin, 16]
-	stp     C_l, C_h, [dstend, -32]
-	stp     D_l, D_h, [dstend, -16]
-	ret
-
-	.p2align 4
-	/* Copy 65..128 bytes.  */
-.Lcopy128:
-	ldp     E_l, E_h, [src, 32]
-	ldp     F_l, F_h, [src, 48]
-	cmp     count, 96
-	b.ls    .Lcopy96
-	ldp     G_l, G_h, [srcend, -64]
-	ldp     H_l, H_h, [srcend, -48]
-	stp     G_l, G_h, [dstend, -64]
-	stp     H_l, H_h, [dstend, -48]
-.Lcopy96:
-	stp     A_l, A_h, [dstin]
-	stp     B_l, B_h, [dstin, 16]
-	stp     E_l, E_h, [dstin, 32]
-	stp     F_l, F_h, [dstin, 48]
-	stp     C_l, C_h, [dstend, -32]
-	stp     D_l, D_h, [dstend, -16]
-	ret
-
-	.p2align 4
-	/* Copy more than 128 bytes.  */
-.Lcopy_long:
-
-	/* Copy 16 bytes and then align dst to 16-byte alignment.  */
-
-	ldp     D_l, D_h, [src]
-	and     tmp1, dstin, 15
-	bic     dst, dstin, 15
-	sub     src, src, tmp1
-	add     count, count, tmp1      /* Count is now 16 too large.  */
-	ldp     A_l, A_h, [src, 16]
-	stp     D_l, D_h, [dstin]
-	ldp     B_l, B_h, [src, 32]
-	ldp     C_l, C_h, [src, 48]
-	ldp     D_l, D_h, [src, 64]!
-	subs    count, count, 128 + 16  /* Test and readjust count.  */
-	b.ls    .Lcopy64_from_end
-
-.Lloop64:
-	stp     A_l, A_h, [dst, 16]
-	ldp     A_l, A_h, [src, 16]
-	stp     B_l, B_h, [dst, 32]
-	ldp     B_l, B_h, [src, 32]
-	stp     C_l, C_h, [dst, 48]
-	ldp     C_l, C_h, [src, 48]
-	stp     D_l, D_h, [dst, 64]!
-	ldp     D_l, D_h, [src, 64]!
-	subs    count, count, 64
-	b.hi    .Lloop64
-
-	/* Write the last iteration and copy 64 bytes from the end.  */
-.Lcopy64_from_end:
-	ldp     E_l, E_h, [srcend, -64]
-	stp     A_l, A_h, [dst, 16]
-	ldp     A_l, A_h, [srcend, -48]
-	stp     B_l, B_h, [dst, 32]
-	ldp     B_l, B_h, [srcend, -32]
-	stp     C_l, C_h, [dst, 48]
-	ldp     C_l, C_h, [srcend, -16]
-	stp     D_l, D_h, [dst, 64]
-	stp     E_l, E_h, [dstend, -64]
-	stp     A_l, A_h, [dstend, -48]
-	stp     B_l, B_h, [dstend, -32]
-	stp     C_l, C_h, [dstend, -16]
-	ret
-
-.size memcpy,.-memcpy
lib/libc/wasi/libc-top-half/musl/src/string/arm/memcpy.S
@@ -1,479 +0,0 @@
-/*
- * Copyright (C) 2008 The Android Open Source Project
- * All rights reserved.
- *
- * Redistribution and use in source and binary forms, with or without
- * modification, are permitted provided that the following conditions
- * are met:
- *  * Redistributions of source code must retain the above copyright
- *    notice, this list of conditions and the following disclaimer.
- *  * Redistributions in binary form must reproduce the above copyright
- *    notice, this list of conditions and the following disclaimer in
- *    the documentation and/or other materials provided with the
- *    distribution.
- *
- * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
- * "AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
- * LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
- * FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
- * COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
- * INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING,
- * BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS
- * OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED
- * AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
- * OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
- * OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
- * SUCH DAMAGE.
- */
-
-
-/*
- * Optimized memcpy() for ARM.
- *
- * note that memcpy() always returns the destination pointer,
- * so we have to preserve R0.
-  */
-
-/*
- * This file has been modified from the original for use in musl libc.
- * The main changes are: addition of .type memcpy,%function to make the
- * code safely callable from thumb mode, adjusting the return
- * instructions to be compatible with pre-thumb ARM cpus, removal of
- * prefetch code that is not compatible with older cpus and support for
- * building as thumb 2 and big-endian.
- */
-
-.syntax unified
-
-.global memcpy
-.type memcpy,%function
-memcpy:
-	/* The stack must always be 64-bits aligned to be compliant with the
-	 * ARM ABI. Since we have to save R0, we might as well save R4
-	 * which we can use for better pipelining of the reads below
-	 */
-	.fnstart
-	.save       {r0, r4, lr}
-	stmfd       sp!, {r0, r4, lr}
-	/* Making room for r5-r11 which will be spilled later */
-	.pad        #28
-	sub         sp, sp, #28
-
-	/* it simplifies things to take care of len<4 early */
-	cmp     r2, #4
-	blo     copy_last_3_and_return
-
-	/* compute the offset to align the source
-	 * offset = (4-(src&3))&3 = -src & 3
-	 */
-	rsb     r3, r1, #0
-	ands    r3, r3, #3
-	beq     src_aligned
-
-	/* align source to 32 bits. We need to insert 2 instructions between
-	 * a ldr[b|h] and str[b|h] because byte and half-word instructions
-	 * stall 2 cycles.
-	 */
-	movs    r12, r3, lsl #31
-	sub     r2, r2, r3              /* we know that r3 <= r2 because r2 >= 4 */
-	ldrbmi r3, [r1], #1
-	ldrbcs r4, [r1], #1
-	ldrbcs r12,[r1], #1
-	strbmi r3, [r0], #1
-	strbcs r4, [r0], #1
-	strbcs r12,[r0], #1
-
-src_aligned:
-
-	/* see if src and dst are aligned together (congruent) */
-	eor     r12, r0, r1
-	tst     r12, #3
-	bne     non_congruent
-
-	/* Use post-incriment mode for stm to spill r5-r11 to reserved stack
-	 * frame. Don't update sp.
-	 */
-	stmea   sp, {r5-r11}
-
-	/* align the destination to a cache-line */
-	rsb     r3, r0, #0
-	ands    r3, r3, #0x1C
-	beq     congruent_aligned32
-	cmp     r3, r2
-	andhi   r3, r2, #0x1C
-
-	/* conditionnaly copies 0 to 7 words (length in r3) */
-	movs    r12, r3, lsl #28
-	ldmcs   r1!, {r4, r5, r6, r7}           /* 16 bytes */
-	ldmmi   r1!, {r8, r9}                   /*  8 bytes */
-	stmcs   r0!, {r4, r5, r6, r7}
-	stmmi   r0!, {r8, r9}
-	tst     r3, #0x4
-	ldrne   r10,[r1], #4                    /*  4 bytes */
-	strne   r10,[r0], #4
-	sub     r2, r2, r3
-
-congruent_aligned32:
-	/*
-	 * here source is aligned to 32 bytes.
-	 */
-
-cached_aligned32:
-	subs    r2, r2, #32
-	blo     less_than_32_left
-
-	/*
-	 * We preload a cache-line up to 64 bytes ahead. On the 926, this will
-	 * stall only until the requested world is fetched, but the linefill
-	 * continues in the the background.
-	 * While the linefill is going, we write our previous cache-line
-	 * into the write-buffer (which should have some free space).
-	 * When the linefill is done, the writebuffer will
-	 * start dumping its content into memory
-	 *
-	 * While all this is going, we then load a full cache line into
-	 * 8 registers, this cache line should be in the cache by now
-	 * (or partly in the cache).
-	 *
-	 * This code should work well regardless of the source/dest alignment.
-	 *
-	 */
-
-	/* Align the preload register to a cache-line because the cpu does
-	 * "critical word first" (the first word requested is loaded first).
-	 */
-	@ bic           r12, r1, #0x1F
-	@ add           r12, r12, #64
-
-1:      ldmia   r1!, { r4-r11 }
-	subs    r2, r2, #32
-
-	/* 
-	 * NOTE: if r12 is more than 64 ahead of r1, the following ldrhi
-	 * for ARM9 preload will not be safely guarded by the preceding subs.
-	 * When it is safely guarded the only possibility to have SIGSEGV here
-	 * is because the caller overstates the length.
-	 */
-	@ ldrhi         r3, [r12], #32      /* cheap ARM9 preload */
-	stmia   r0!, { r4-r11 }
-	bhs     1b
-
-	add     r2, r2, #32
-
-less_than_32_left:
-	/*
-	 * less than 32 bytes left at this point (length in r2)
-	 */
-
-	/* skip all this if there is nothing to do, which should
-	 * be a common case (if not executed the code below takes
-	 * about 16 cycles)
-	 */
-	tst     r2, #0x1F
-	beq     1f
-
-	/* conditionnaly copies 0 to 31 bytes */
-	movs    r12, r2, lsl #28
-	ldmcs   r1!, {r4, r5, r6, r7}           /* 16 bytes */
-	ldmmi   r1!, {r8, r9}                   /*  8 bytes */
-	stmcs   r0!, {r4, r5, r6, r7}
-	stmmi   r0!, {r8, r9}
-	movs    r12, r2, lsl #30
-	ldrcs   r3, [r1], #4                    /*  4 bytes */
-	ldrhmi r4, [r1], #2                     /*  2 bytes */
-	strcs   r3, [r0], #4
-	strhmi r4, [r0], #2
-	tst     r2, #0x1
-	ldrbne r3, [r1]                         /*  last byte  */
-	strbne r3, [r0]
-
-	/* we're done! restore everything and return */
-1:      ldmfd   sp!, {r5-r11}
-	ldmfd   sp!, {r0, r4, lr}
-	bx      lr
-
-	/********************************************************************/
-
-non_congruent:
-	/*
-	 * here source is aligned to 4 bytes
-	 * but destination is not.
-	 *
-	 * in the code below r2 is the number of bytes read
-	 * (the number of bytes written is always smaller, because we have
-	 * partial words in the shift queue)
-	 */
-	cmp     r2, #4
-	blo     copy_last_3_and_return
-
-	/* Use post-incriment mode for stm to spill r5-r11 to reserved stack
-	 * frame. Don't update sp.
-	 */
-	stmea   sp, {r5-r11}
-
-	/* compute shifts needed to align src to dest */
-	rsb     r5, r0, #0
-	and     r5, r5, #3                      /* r5 = # bytes in partial words */
-	mov     r12, r5, lsl #3         /* r12 = right */
-	rsb     lr, r12, #32            /* lr = left  */
-
-	/* read the first word */
-	ldr     r3, [r1], #4
-	sub     r2, r2, #4
-
-	/* write a partial word (0 to 3 bytes), such that destination
-	 * becomes aligned to 32 bits (r5 = nb of words to copy for alignment)
-	 */
-	movs    r5, r5, lsl #31
-
-#if __ARMEB__
-	movmi   r3, r3, ror #24
-	strbmi	r3, [r0], #1
-	movcs   r3, r3, ror #24
-	strbcs	r3, [r0], #1
-	movcs   r3, r3, ror #24
-	strbcs	r3, [r0], #1
-#else
-	strbmi r3, [r0], #1
-	movmi   r3, r3, lsr #8
-	strbcs r3, [r0], #1
-	movcs   r3, r3, lsr #8
-	strbcs r3, [r0], #1
-	movcs   r3, r3, lsr #8
-#endif
-
-	cmp     r2, #4
-	blo     partial_word_tail
-
-#if __ARMEB__
-	mov	r3, r3, lsr r12
-	mov	r3, r3, lsl r12
-#endif
-
-	/* Align destination to 32 bytes (cache line boundary) */
-1:      tst     r0, #0x1c
-	beq     2f
-	ldr     r5, [r1], #4
-	sub     r2, r2, #4
-#if __ARMEB__
-	mov     r4, r5,                 lsr lr
-	orr     r4, r4, r3
-	mov     r3, r5,                 lsl r12
-#else
-	mov     r4, r5,                 lsl lr
-	orr     r4, r4, r3
-	mov     r3, r5,                 lsr r12
-#endif
-	str     r4, [r0], #4
-	cmp     r2, #4
-	bhs     1b
-	blo     partial_word_tail
-
-	/* copy 32 bytes at a time */
-2:      subs    r2, r2, #32
-	blo     less_than_thirtytwo
-
-	/* Use immediate mode for the shifts, because there is an extra cycle
-	 * for register shifts, which could account for up to 50% of
-	 * performance hit.
-	 */
-
-	cmp     r12, #24
-	beq     loop24
-	cmp     r12, #8
-	beq     loop8
-
-loop16:
-	ldr     r12, [r1], #4
-1:      mov     r4, r12
-	ldmia   r1!, {   r5,r6,r7,  r8,r9,r10,r11}
-	subs    r2, r2, #32
-	ldrhs   r12, [r1], #4
-#if __ARMEB__
-	orr     r3, r3, r4, lsr #16
-	mov     r4, r4, lsl #16
-	orr     r4, r4, r5, lsr #16
-	mov     r5, r5, lsl #16
-	orr     r5, r5, r6, lsr #16
-	mov     r6, r6, lsl #16
-	orr     r6, r6, r7, lsr #16
-	mov     r7, r7, lsl #16
-	orr     r7, r7, r8, lsr #16
-	mov     r8, r8, lsl #16
-	orr     r8, r8, r9, lsr #16
-	mov     r9, r9, lsl #16
-	orr     r9, r9, r10, lsr #16
-	mov     r10, r10,               lsl #16
-	orr     r10, r10, r11, lsr #16
-	stmia   r0!, {r3,r4,r5,r6, r7,r8,r9,r10}
-	mov     r3, r11, lsl #16
-#else
-	orr     r3, r3, r4, lsl #16
-	mov     r4, r4, lsr #16
-	orr     r4, r4, r5, lsl #16
-	mov     r5, r5, lsr #16
-	orr     r5, r5, r6, lsl #16
-	mov     r6, r6, lsr #16
-	orr     r6, r6, r7, lsl #16
-	mov     r7, r7, lsr #16
-	orr     r7, r7, r8, lsl #16
-	mov     r8, r8, lsr #16
-	orr     r8, r8, r9, lsl #16
-	mov     r9, r9, lsr #16
-	orr     r9, r9, r10, lsl #16
-	mov     r10, r10,               lsr #16
-	orr     r10, r10, r11, lsl #16
-	stmia   r0!, {r3,r4,r5,r6, r7,r8,r9,r10}
-	mov     r3, r11, lsr #16
-#endif
-	bhs     1b
-	b       less_than_thirtytwo
-
-loop8:
-	ldr     r12, [r1], #4
-1:      mov     r4, r12
-	ldmia   r1!, {   r5,r6,r7,  r8,r9,r10,r11}
-	subs    r2, r2, #32
-	ldrhs   r12, [r1], #4
-#if __ARMEB__
-	orr     r3, r3, r4, lsr #24
-	mov     r4, r4, lsl #8
-	orr     r4, r4, r5, lsr #24
-	mov     r5, r5, lsl #8
-	orr     r5, r5, r6, lsr #24
-	mov     r6, r6,  lsl #8
-	orr     r6, r6, r7, lsr #24
-	mov     r7, r7,  lsl #8
-	orr     r7, r7, r8,             lsr #24
-	mov     r8, r8,  lsl #8
-	orr     r8, r8, r9,             lsr #24
-	mov     r9, r9,  lsl #8
-	orr     r9, r9, r10,    lsr #24
-	mov     r10, r10, lsl #8
-	orr     r10, r10, r11,  lsr #24
-	stmia   r0!, {r3,r4,r5,r6, r7,r8,r9,r10}
-	mov     r3, r11, lsl #8
-#else
-	orr     r3, r3, r4, lsl #24
-	mov     r4, r4, lsr #8
-	orr     r4, r4, r5, lsl #24
-	mov     r5, r5, lsr #8
-	orr     r5, r5, r6, lsl #24
-	mov     r6, r6,  lsr #8
-	orr     r6, r6, r7, lsl #24
-	mov     r7, r7,  lsr #8
-	orr     r7, r7, r8,             lsl #24
-	mov     r8, r8,  lsr #8
-	orr     r8, r8, r9,             lsl #24
-	mov     r9, r9,  lsr #8
-	orr     r9, r9, r10,    lsl #24
-	mov     r10, r10, lsr #8
-	orr     r10, r10, r11,  lsl #24
-	stmia   r0!, {r3,r4,r5,r6, r7,r8,r9,r10}
-	mov     r3, r11, lsr #8
-#endif
-	bhs     1b
-	b       less_than_thirtytwo
-
-loop24:
-	ldr     r12, [r1], #4
-1:      mov     r4, r12
-	ldmia   r1!, {   r5,r6,r7,  r8,r9,r10,r11}
-	subs    r2, r2, #32
-	ldrhs   r12, [r1], #4
-#if __ARMEB__
-	orr     r3, r3, r4, lsr #8
-	mov     r4, r4, lsl #24
-	orr     r4, r4, r5, lsr #8
-	mov     r5, r5, lsl #24
-	orr     r5, r5, r6, lsr #8
-	mov     r6, r6, lsl #24
-	orr     r6, r6, r7, lsr #8
-	mov     r7, r7, lsl #24
-	orr     r7, r7, r8, lsr #8
-	mov     r8, r8, lsl #24
-	orr     r8, r8, r9, lsr #8
-	mov     r9, r9, lsl #24
-	orr     r9, r9, r10, lsr #8
-	mov     r10, r10, lsl #24
-	orr     r10, r10, r11, lsr #8
-	stmia   r0!, {r3,r4,r5,r6, r7,r8,r9,r10}
-	mov     r3, r11, lsl #24
-#else
-	orr     r3, r3, r4, lsl #8
-	mov     r4, r4, lsr #24
-	orr     r4, r4, r5, lsl #8
-	mov     r5, r5, lsr #24
-	orr     r5, r5, r6, lsl #8
-	mov     r6, r6, lsr #24
-	orr     r6, r6, r7, lsl #8
-	mov     r7, r7, lsr #24
-	orr     r7, r7, r8, lsl #8
-	mov     r8, r8, lsr #24
-	orr     r8, r8, r9, lsl #8
-	mov     r9, r9, lsr #24
-	orr     r9, r9, r10, lsl #8
-	mov     r10, r10, lsr #24
-	orr     r10, r10, r11, lsl #8
-	stmia   r0!, {r3,r4,r5,r6, r7,r8,r9,r10}
-	mov     r3, r11, lsr #24
-#endif
-	bhs     1b
-
-less_than_thirtytwo:
-	/* copy the last 0 to 31 bytes of the source */
-	rsb     r12, lr, #32            /* we corrupted r12, recompute it  */
-	add     r2, r2, #32
-	cmp     r2, #4
-	blo     partial_word_tail
-
-1:      ldr     r5, [r1], #4
-	sub     r2, r2, #4
-#if __ARMEB__
-	mov     r4, r5,                 lsr lr
-	orr     r4, r4, r3
-	mov     r3,     r5,                     lsl r12
-#else
-	mov     r4, r5,                 lsl lr
-	orr     r4, r4, r3
-	mov     r3,     r5,                     lsr r12
-#endif
-	str     r4, [r0], #4
-	cmp     r2, #4
-	bhs     1b
-
-partial_word_tail:
-	/* we have a partial word in the input buffer */
-	movs    r5, lr, lsl #(31-3)
-#if __ARMEB__
-	movmi   r3, r3, ror #24
-	strbmi r3, [r0], #1
-	movcs   r3, r3, ror #24
-	strbcs r3, [r0], #1
-	movcs   r3, r3, ror #24
-	strbcs r3, [r0], #1
-#else
-	strbmi r3, [r0], #1
-	movmi   r3, r3, lsr #8
-	strbcs r3, [r0], #1
-	movcs   r3, r3, lsr #8
-	strbcs r3, [r0], #1
-#endif
-
-	/* Refill spilled registers from the stack. Don't update sp. */
-	ldmfd   sp, {r5-r11}
-
-copy_last_3_and_return:
-	movs    r2, r2, lsl #31 /* copy remaining 0, 1, 2 or 3 bytes */
-	ldrbmi r2, [r1], #1
-	ldrbcs r3, [r1], #1
-	ldrbcs r12,[r1]
-	strbmi r2, [r0], #1
-	strbcs r3, [r0], #1
-	strbcs r12,[r0]
-
-	/* we're done! restore sp and spilled registers and return */
-	add     sp,  sp, #28
-	ldmfd   sp!, {r0, r4, lr}
-	bx      lr
-
lib/libc/wasi/libc-top-half/musl/src/string/memcpy.c
@@ -1,128 +0,0 @@
-#include <string.h>
-#include <stdint.h>
-#include <endian.h>
-
-void *memcpy(void *restrict dest, const void *restrict src, size_t n)
-{
-#if defined(__wasm_bulk_memory__)
-	if (n > BULK_MEMORY_THRESHOLD)
-	  return __builtin_memcpy(dest, src, n);
-#endif
-	unsigned char *d = dest;
-	const unsigned char *s = src;
-
-#ifdef __GNUC__
-
-#if __BYTE_ORDER == __LITTLE_ENDIAN
-#define LS >>
-#define RS <<
-#else
-#define LS <<
-#define RS >>
-#endif
-
-	typedef uint32_t __attribute__((__may_alias__)) u32;
-	uint32_t w, x;
-
-	for (; (uintptr_t)s % 4 && n; n--) *d++ = *s++;
-
-	if ((uintptr_t)d % 4 == 0) {
-		for (; n>=16; s+=16, d+=16, n-=16) {
-			*(u32 *)(d+0) = *(u32 *)(s+0);
-			*(u32 *)(d+4) = *(u32 *)(s+4);
-			*(u32 *)(d+8) = *(u32 *)(s+8);
-			*(u32 *)(d+12) = *(u32 *)(s+12);
-		}
-		if (n&8) {
-			*(u32 *)(d+0) = *(u32 *)(s+0);
-			*(u32 *)(d+4) = *(u32 *)(s+4);
-			d += 8; s += 8;
-		}
-		if (n&4) {
-			*(u32 *)(d+0) = *(u32 *)(s+0);
-			d += 4; s += 4;
-		}
-		if (n&2) {
-			*d++ = *s++; *d++ = *s++;
-		}
-		if (n&1) {
-			*d = *s;
-		}
-		return dest;
-	}
-
-	if (n >= 32) switch ((uintptr_t)d % 4) {
-	case 1:
-		w = *(u32 *)s;
-		*d++ = *s++;
-		*d++ = *s++;
-		*d++ = *s++;
-		n -= 3;
-		for (; n>=17; s+=16, d+=16, n-=16) {
-			x = *(u32 *)(s+1);
-			*(u32 *)(d+0) = (w LS 24) | (x RS 8);
-			w = *(u32 *)(s+5);
-			*(u32 *)(d+4) = (x LS 24) | (w RS 8);
-			x = *(u32 *)(s+9);
-			*(u32 *)(d+8) = (w LS 24) | (x RS 8);
-			w = *(u32 *)(s+13);
-			*(u32 *)(d+12) = (x LS 24) | (w RS 8);
-		}
-		break;
-	case 2:
-		w = *(u32 *)s;
-		*d++ = *s++;
-		*d++ = *s++;
-		n -= 2;
-		for (; n>=18; s+=16, d+=16, n-=16) {
-			x = *(u32 *)(s+2);
-			*(u32 *)(d+0) = (w LS 16) | (x RS 16);
-			w = *(u32 *)(s+6);
-			*(u32 *)(d+4) = (x LS 16) | (w RS 16);
-			x = *(u32 *)(s+10);
-			*(u32 *)(d+8) = (w LS 16) | (x RS 16);
-			w = *(u32 *)(s+14);
-			*(u32 *)(d+12) = (x LS 16) | (w RS 16);
-		}
-		break;
-	case 3:
-		w = *(u32 *)s;
-		*d++ = *s++;
-		n -= 1;
-		for (; n>=19; s+=16, d+=16, n-=16) {
-			x = *(u32 *)(s+3);
-			*(u32 *)(d+0) = (w LS 8) | (x RS 24);
-			w = *(u32 *)(s+7);
-			*(u32 *)(d+4) = (x LS 8) | (w RS 24);
-			x = *(u32 *)(s+11);
-			*(u32 *)(d+8) = (w LS 8) | (x RS 24);
-			w = *(u32 *)(s+15);
-			*(u32 *)(d+12) = (x LS 8) | (w RS 24);
-		}
-		break;
-	}
-	if (n&16) {
-		*d++ = *s++; *d++ = *s++; *d++ = *s++; *d++ = *s++;
-		*d++ = *s++; *d++ = *s++; *d++ = *s++; *d++ = *s++;
-		*d++ = *s++; *d++ = *s++; *d++ = *s++; *d++ = *s++;
-		*d++ = *s++; *d++ = *s++; *d++ = *s++; *d++ = *s++;
-	}
-	if (n&8) {
-		*d++ = *s++; *d++ = *s++; *d++ = *s++; *d++ = *s++;
-		*d++ = *s++; *d++ = *s++; *d++ = *s++; *d++ = *s++;
-	}
-	if (n&4) {
-		*d++ = *s++; *d++ = *s++; *d++ = *s++; *d++ = *s++;
-	}
-	if (n&2) {
-		*d++ = *s++; *d++ = *s++;
-	}
-	if (n&1) {
-		*d = *s;
-	}
-	return dest;
-#endif
-
-	for (; n; n--) *d++ = *s++;
-	return dest;
-}
lib/libc/wasi/libc-top-half/musl/src/string/memmove.c
@@ -1,46 +0,0 @@
-#include <string.h>
-#include <stdint.h>
-
-#ifdef __GNUC__
-typedef __attribute__((__may_alias__)) size_t WT;
-#define WS (sizeof(WT))
-#endif
-
-void *memmove(void *dest, const void *src, size_t n)
-{
-#if defined(__wasm_bulk_memory__)
-	if (n > BULK_MEMORY_THRESHOLD)
-		return __builtin_memmove(dest, src, n);
-#endif
-	char *d = dest;
-	const char *s = src;
-
-	if (d==s) return d;
-	if ((uintptr_t)s-(uintptr_t)d-n <= -2*n) return memcpy(d, s, n);
-
-	if (d<s) {
-#ifdef __GNUC__
-		if ((uintptr_t)s % WS == (uintptr_t)d % WS) {
-			while ((uintptr_t)d % WS) {
-				if (!n--) return dest;
-				*d++ = *s++;
-			}
-			for (; n>=WS; n-=WS, d+=WS, s+=WS) *(WT *)d = *(WT *)s;
-		}
-#endif
-		for (; n; n--) *d++ = *s++;
-	} else {
-#ifdef __GNUC__
-		if ((uintptr_t)s % WS == (uintptr_t)d % WS) {
-			while ((uintptr_t)(d+n) % WS) {
-				if (!n--) return dest;
-				d[n] = s[n];
-			}
-			while (n>=WS) n-=WS, *(WT *)(d+n) = *(WT *)(s+n);
-		}
-#endif
-		while (n) n--, d[n] = s[n];
-	}
-
-	return dest;
-}
src/musl.zig
@@ -1899,25 +1899,18 @@ const src_files = [_][]const u8{
     "musl/src/stdlib/strtol.c",
     "musl/src/stdlib/wcstod.c",
     "musl/src/stdlib/wcstol.c",
-    "musl/src/string/aarch64/memcpy.S",
     "musl/src/string/aarch64/memset.S",
-    "musl/src/string/arm/__aeabi_memcpy.s",
     "musl/src/string/arm/__aeabi_memset.s",
-    "musl/src/string/arm/memcpy.S",
     "musl/src/string/bcmp.c",
     "musl/src/string/bcopy.c",
     "musl/src/string/bzero.c",
     "musl/src/string/explicit_bzero.c",
-    "musl/src/string/i386/memcpy.s",
-    "musl/src/string/i386/memmove.s",
     "musl/src/string/i386/memset.s",
     "musl/src/string/index.c",
     "musl/src/string/memccpy.c",
     "musl/src/string/memchr.c",
     "musl/src/string/memcmp.c",
-    "musl/src/string/memcpy.c",
     "musl/src/string/memmem.c",
-    "musl/src/string/memmove.c",
     "musl/src/string/mempcpy.c",
     "musl/src/string/memrchr.c",
     "musl/src/string/memset.c",
@@ -1981,8 +1974,6 @@ const src_files = [_][]const u8{
     "musl/src/string/wmemcpy.c",
     "musl/src/string/wmemmove.c",
     "musl/src/string/wmemset.c",
-    "musl/src/string/x86_64/memcpy.s",
-    "musl/src/string/x86_64/memmove.s",
     "musl/src/string/x86_64/memset.s",
     "musl/src/temp/mkdtemp.c",
     "musl/src/temp/mkostemp.c",
src/wasi_libc.zig
@@ -694,9 +694,7 @@ const libc_top_half_src_files = [_][]const u8{
     "wasi/libc-top-half/musl/src/string/memccpy.c",
     "wasi/libc-top-half/musl/src/string/memchr.c",
     "wasi/libc-top-half/musl/src/string/memcmp.c",
-    "wasi/libc-top-half/musl/src/string/memcpy.c",
     "wasi/libc-top-half/musl/src/string/memmem.c",
-    "wasi/libc-top-half/musl/src/string/memmove.c",
     "wasi/libc-top-half/musl/src/string/mempcpy.c",
     "wasi/libc-top-half/musl/src/string/memrchr.c",
     "wasi/libc-top-half/musl/src/string/memset.c",