From d6434e1042f3b0a6dfe1b1f615af369486f9b1fa Mon Sep 17 00:00:00 2001
From: Ralf Baechle <ralf@linux-mips.org>
Date: Sat, 9 Oct 1999 00:00:47 +0000
Subject: Merge with 2.3.19.

---
 arch/sh/lib/Makefile            |   14 +
 arch/sh/lib/checksum.c          |  170 ++++++
 arch/sh/lib/csum_partial_copy.c |   75 +++
 arch/sh/lib/delay.c             |   21 +
 arch/sh/lib/memcpy.S            |  131 ++++
 arch/sh/lib/memmove.S           |  422 +++++++++++++
 arch/sh/lib/memset.S            |   72 +++
 arch/sh/lib/old-checksum.c      |   19 +
 arch/sh/lib/wordcopy.S          | 1289 +++++++++++++++++++++++++++++++++++++++
 9 files changed, 2213 insertions(+)
 create mode 100644 arch/sh/lib/Makefile
 create mode 100644 arch/sh/lib/checksum.c
 create mode 100644 arch/sh/lib/csum_partial_copy.c
 create mode 100644 arch/sh/lib/delay.c
 create mode 100644 arch/sh/lib/memcpy.S
 create mode 100644 arch/sh/lib/memmove.S
 create mode 100644 arch/sh/lib/memset.S
 create mode 100644 arch/sh/lib/old-checksum.c
 create mode 100644 arch/sh/lib/wordcopy.S

(limited to 'arch/sh/lib')

diff --git a/arch/sh/lib/Makefile b/arch/sh/lib/Makefile
new file mode 100644
index 000000000..a9010ddb2
--- /dev/null
+++ b/arch/sh/lib/Makefile
@@ -0,0 +1,14 @@
+#
+# Makefile for SuperH-specific library files..
+#
+
+.S.o:
+	$(CC) -D__ASSEMBLY__ $(AFLAGS) -traditional -c $< -o $*.o
+
+L_TARGET = lib.a
+# L_OBJS  = checksum.o old-checksum.o semaphore.o delay.o \
+# 	usercopy.o getuser.o putuser.o
+L_OBJS  = delay.o memcpy.o memset.o memmove.o csum_partial_copy.o \
+	  wordcopy.o checksum.o # usercopy.o getuser.o putuser.o
+
+include $(TOPDIR)/Rules.make
diff --git a/arch/sh/lib/checksum.c b/arch/sh/lib/checksum.c
new file mode 100644
index 000000000..f076bd3c6
--- /dev/null
+++ b/arch/sh/lib/checksum.c
@@ -0,0 +1,170 @@
+/*
+ * Taken from:
+ * arch/alpha/lib/checksum.c
+ *
+ * This file contains network checksum routines that are better done
+ * in an architecture-specific manner due to speed..
+ */
+ 
+#include <linux/string.h>
+
+#include <asm/byteorder.h>
+
+static inline unsigned short from64to16(unsigned long long x)
+{
+	/* add up 32-bit words for 33 bits */
+	x = (x & 0xffffffff) + (x >> 32);
+	/* add up 16-bit and 17-bit words for 17+c bits */
+	x = (x & 0xffff) + (x >> 16);
+	/* add up 16-bit and 2-bit for 16+c bit */
+	x = (x & 0xffff) + (x >> 16);
+	/* add up carry.. */
+	x = (x & 0xffff) + (x >> 16);
+	return x;
+}
+
+/*
+ * computes the checksum of the TCP/UDP pseudo-header
+ * returns a 16-bit checksum, already complemented.
+ */
+unsigned short int csum_tcpudp_magic(unsigned long saddr,
+				   unsigned long daddr,
+				   unsigned short len,
+				   unsigned short proto,
+				   unsigned int sum)
+{
+	return ~from64to16(saddr + daddr + sum +
+		((unsigned long) ntohs(len) << 16) +
+		((unsigned long) proto << 8));
+}
+
+unsigned int csum_tcpudp_nofold(unsigned long saddr,
+				unsigned long daddr,
+				unsigned short len,
+				unsigned short proto,
+				unsigned int sum)
+{
+	unsigned long long result;
+
+	result = (saddr + daddr + sum +
+		  ((unsigned long) ntohs(len) << 16) +
+		  ((unsigned long) proto << 8));
+
+	/* Fold down to 32-bits so we don't loose in the typedef-less 
+	   network stack.  */
+	/* 64 to 33 */
+	result = (result & 0xffffffff) + (result >> 32);
+	/* 33 to 32 */
+	result = (result & 0xffffffff) + (result >> 32);
+	return result;
+}
+
+/*
+ * Do a 64-bit checksum on an arbitrary memory area..
+ *
+ * This isn't a great routine, but it's not _horrible_ either. The
+ * inner loop could be unrolled a bit further, and there are better
+ * ways to do the carry, but this is reasonable.
+ */
+static inline unsigned long do_csum(const unsigned char * buff, int len)
+{
+	int odd, count;
+	unsigned long long result = 0;
+
+	if (len <= 0)
+		goto out;
+	odd = 1 & (unsigned long) buff;
+	if (odd) {
+		result = *buff << 8;
+		len--;
+		buff++;
+	}
+	count = len >> 1;		/* nr of 16-bit words.. */
+	if (count) {
+		if (2 & (unsigned long) buff) {
+			result += *(unsigned short *) buff;
+			count--;
+			len -= 2;
+			buff += 2;
+		}
+		count >>= 1;		/* nr of 32-bit words.. */
+		if (count) {
+			if (4 & (unsigned long) buff) {
+				result += *(unsigned int *) buff;
+				count--;
+				len -= 4;
+				buff += 4;
+			}
+			count >>= 1;	/* nr of 64-bit words.. */
+			if (count) {
+				unsigned long carry = 0;
+				do {
+					unsigned long w = *(unsigned long *) buff;
+					count--;
+					buff += 8;
+					result += carry;
+					result += w;
+					carry = (w > result);
+				} while (count);
+				result += carry;
+				result = (result & 0xffffffff) + (result >> 32);
+			}
+			if (len & 4) {
+				result += *(unsigned int *) buff;
+				buff += 4;
+			}
+		}
+		if (len & 2) {
+			result += *(unsigned short *) buff;
+			buff += 2;
+		}
+	}
+	if (len & 1)
+		result += *buff;
+	result = from64to16(result);
+	if (odd)
+		result = ((result >> 8) & 0xff) | ((result & 0xff) << 8);
+out:
+	return result;
+}
+
+/*
+ *	This is a version of ip_compute_csum() optimized for IP headers,
+ *	which always checksum on 4 octet boundaries.
+ */
+unsigned short ip_fast_csum(unsigned char * iph, unsigned int ihl)
+{
+	return ~do_csum(iph,ihl*4);
+}
+
+/*
+ * computes the checksum of a memory block at buff, length len,
+ * and adds in "sum" (32-bit)
+ *
+ * returns a 32-bit number suitable for feeding into itself
+ * or csum_tcpudp_magic
+ *
+ * this function must be called with even lengths, except
+ * for the last fragment, which may be odd
+ *
+ * it's best to have buff aligned on a 32-bit boundary
+ */
+unsigned int csum_partial(const unsigned char * buff, int len, unsigned int sum)
+{
+	unsigned long long result = do_csum(buff, len);
+
+	/* add in old sum, and carry.. */
+	result += sum;
+	/* 32+c bits -> 32 bits */
+	result = (result & 0xffffffff) + (result >> 32);
+	return result;
+}
+
+/*
+ * this routine is used for miscellaneous IP-like checksums, mainly
+ * in icmp.c
+ */
+unsigned short ip_compute_csum(unsigned char * buff, int len)
+{
+	return ~from64to16(do_csum(buff,len));
+}
diff --git a/arch/sh/lib/csum_partial_copy.c b/arch/sh/lib/csum_partial_copy.c
new file mode 100644
index 000000000..1fb36ab05
--- /dev/null
+++ b/arch/sh/lib/csum_partial_copy.c
@@ -0,0 +1,75 @@
+/*
+ * INET		An implementation of the TCP/IP protocol suite for the LINUX
+ *		operating system.  INET is implemented using the  BSD Socket
+ *		interface as the means of communication with the user level.
+ *
+ *		MIPS specific IP/TCP/UDP checksumming routines
+ *
+ * Authors:	Ralf Baechle, <ralf@waldorf-gmbh.de>
+ *		Lots of code moved from tcp.c and ip.c; see those files
+ *		for more names.
+ *
+ *		This program is free software; you can redistribute it and/or
+ *		modify it under the terms of the GNU General Public License
+ *		as published by the Free Software Foundation; either version
+ *		2 of the License, or (at your option) any later version.
+ *
+ * $Id: csum_partial_copy.c,v 1.2 1998/09/16 13:29:32 ralf Exp $
+ */
+#include <net/checksum.h>
+#include <linux/types.h>
+#include <asm/byteorder.h>
+#include <asm/string.h>
+#include <asm/uaccess.h>
+
+/*
+ * copy while checksumming, otherwise like csum_partial
+ */
+unsigned int csum_partial_copy(const char *src, char *dst, 
+                               int len, unsigned int sum)
+{
+	/*
+	 * It's 2:30 am and I don't feel like doing it real ...
+	 * This is lots slower than the real thing (tm)
+	 */
+	sum = csum_partial(src, len, sum);
+	memcpy(dst, src, len);
+
+	return sum;
+}
+
+/*
+ * Copy from userspace and compute checksum.  If we catch an exception
+ * then zero the rest of the buffer.
+ */
+unsigned int csum_partial_copy_from_user (const char *src, char *dst,
+                                          int len, unsigned int sum,
+                                          int *err_ptr)
+{
+	int missing;
+
+	missing = copy_from_user(dst, src, len);
+	if (missing) {
+		memset(dst + len - missing, 0, missing);
+		*err_ptr = -EFAULT;
+	}
+		
+	return csum_partial(dst, len, sum);
+}
+
+/*
+ * Copy to userspace and compute checksum.
+ */
+unsigned int csum_partial_copy_to_user (const char *src, char *dst,
+                                        int len, unsigned int sum,
+                                        int *err_ptr)
+{
+	sum = csum_partial(src, len, sum);
+
+	if (copy_to_user(dst, src, len)) {
+		*err_ptr = -EFAULT;
+		return sum;
+	}
+
+	return sum;
+}
diff --git a/arch/sh/lib/delay.c b/arch/sh/lib/delay.c
new file mode 100644
index 000000000..cadd7367e
--- /dev/null
+++ b/arch/sh/lib/delay.c
@@ -0,0 +1,21 @@
+/*
+ *	Precise Delay Loops for SuperH
+ *
+ *	Copyright (C) 1999 Niibe Yutaka
+ */
+
+#include <linux/sched.h>
+#include <linux/delay.h>
+
+inline void __const_udelay(unsigned long xloops)
+{
+	xloops *= current_cpu_data.loops_per_sec;
+        __delay(xloops);
+}
+
+#if 0
+void __udelay(unsigned long usecs)
+{
+	__const_udelay(usecs * 0x000010c6);  /* 2**32 / 1000000 */
+}
+#endif
diff --git a/arch/sh/lib/memcpy.S b/arch/sh/lib/memcpy.S
new file mode 100644
index 000000000..3a4aca919
--- /dev/null
+++ b/arch/sh/lib/memcpy.S
@@ -0,0 +1,131 @@
+! Taken from newlib-1.8.0
+
+!
+! Fast SH memcpy
+!
+! by Toshiyasu Morita (tm@netcom.com)
+! hacked by J"orn Rernnecke (amylaar@cygnus.co.uk) ("o for o-umlaut)
+!
+! Entry: r4: destination pointer
+!        r5: source pointer
+!        r6: byte count
+!
+! Exit:  r0: destination pointer
+!        r1-r7: trashed
+!
+! Notes: Usually one wants to do small reads and write a longword, but
+!        unfortunately it is difficult in some cases to concatanate bytes
+!        into a longword on the SH, so this does a longword read and small
+!        writes.
+!
+! This implementation makes two assumptions about how it is called:
+!
+! 1.: If the byte count is nonzero, the address of the last byte to be
+!     copied is unsigned greater than the address of the first byte to
+!     be copied.  This could be easily swapped for a signed comparison,
+!     but the algorithm used needs some comparison.
+!
+! 2.: When there are two or three bytes in the last word of an 11-or-bore
+!     bytes memory chunk to b copied, the rest of the word can be read
+!     without size effects.
+!     This could be easily changed by increasing the minumum size of
+!     a fast memcpy and the amount subtracted from r7 before L_2l_loop be 2,
+!     however, this would cost a few extra cyles on average.
+!
+
+#include <linux/linkage.h>
+ENTRY(memcpy)
+	! Big endian version copies with decreasing addresses.
+	mov	r4,r0
+	add	r6,r0
+	sub	r4,r5
+	mov	#11,r1
+	cmp/hs	r1,r6
+	bf/s	L_small
+	 add	#-1,r5
+	mov	r5,r3
+	add	r0,r3
+	shlr	r3
+	bt/s	L_even
+	 mov	r4,r7
+	mov.b	@(r0,r5),r2
+	add	#-1,r3
+	mov.b	r2,@-r0
+L_even:
+	tst	#1,r0
+	add	#-1,r5
+	bf/s	L_odddst
+	 add	#8,r7
+	tst	#2,r0
+	bt	L_al4dst
+	add	#-1,r3
+	mov.w	@(r0,r5),r1
+	mov.w	r1,@-r0
+L_al4dst:
+	shlr	r3
+	bt	L_al4both
+	mov.w	@(r0,r5),r1
+	swap.w	r1,r1
+	add	#4,r7
+	add	#-4,r5
+	.align	2
+L_2l_loop:
+	mov.l	@(r0,r5),r2
+	xtrct	r2,r1
+	mov.l	r1,@-r0
+	cmp/hs	r7,r0
+	mov.l	@(r0,r5),r1
+	xtrct	r1,r2
+	mov.l	r2,@-r0
+	bt	L_2l_loop
+	bra	L_cleanup
+	 add	#5,r5
+
+	nop ! avoid nop in executed code.
+L_al4both:
+	add	#-2,r5
+	.align	2
+L_al4both_loop:
+	mov.l	@(r0,r5),r1
+	cmp/hs	r7,r0
+	bt/s	L_al4both_loop
+	 mov.l	r1,@-r0
+	bra	L_cleanup
+	 add	#3,r5
+
+	nop ! avoid nop in executed code.
+L_odddst:
+	shlr	r3
+	bt	L_al4src
+	mov.w	@(r0,r5),r1
+	mov.b	r1,@-r0
+	shlr8	r1
+	mov.b	r1,@-r0
+L_al4src:
+	add	#-2,r5
+	.align	2
+L_odd_loop:
+	mov.l	@(r0,r5),r2
+	cmp/hs	r7,r0
+	mov.b	r2,@-r0
+	shlr8	r2
+	mov.w	r2,@-r0
+	shlr16	r2
+	mov.b	r2,@-r0
+	bt	L_odd_loop
+
+	add	#3,r5
+L_cleanup:
+L_small:
+	cmp/eq	r4,r0
+	bt	L_ready
+	add	#1,r4
+	.align	2
+L_cleanup_loop:
+	mov.b	@(r0,r5),r2
+	cmp/eq	r4,r0
+	mov.b	r2,@-r0
+	bf	L_cleanup_loop
+L_ready:
+	rts
+	 nop
diff --git a/arch/sh/lib/memmove.S b/arch/sh/lib/memmove.S
new file mode 100644
index 000000000..e95dc3709
--- /dev/null
+++ b/arch/sh/lib/memmove.S
@@ -0,0 +1,422 @@
+#include <linux/linkage.h>
+ENTRY(memmove)
+	mov.l	r8,@-r15
+	mov.l	r9,@-r15
+	mov.l	r14,@-r15
+	sts.l	pr,@-r15
+	add	#-28,r15
+	mov	r15,r14
+	mov.l	r4,@r14
+	mov.l	r5,@(4,r14)
+	mov.l	r6,@(8,r14)
+	mov.l	@r14,r1
+	mov.l	r1,@(12,r14)
+	mov.l	@(4,r14),r1
+	mov.l	r1,@(16,r14)
+	mov.l	@(12,r14),r1
+	mov.l	@(16,r14),r2
+	sub	r2,r1
+	mov.l	@(8,r14),r2
+	cmp/hs	r2,r1
+	bt	.L54
+	bra	.L2
+	nop
+.L54:
+	mov.l	@(8,r14),r1
+	mov	#15,r2
+	cmp/gt	r2,r1
+	bt	.LF100
+	bra	.L52
+	nop
+.LF100:
+	mov.l	@(12,r14),r2
+	neg	r2,r1
+	mov	#3,r2
+	and	r1,r2
+	mov.l	@(8,r14),r1
+	mov	r1,r9
+	sub	r2,r9
+	mov	r9,r2
+	mov.l	r2,@(8,r14)
+.L4:
+	mov.l	@(12,r14),r2
+	neg	r2,r1
+	mov	#3,r2
+	and	r1,r2
+	mov.l	r2,@(20,r14)
+.L7:
+	mov.l	@(20,r14),r1
+	cmp/pl	r1
+	bt	.L9
+	bra	.L6
+	nop
+	.align 2
+.L9:
+	mov	r14,r2
+	mov	r14,r1
+	add	#24,r1
+	mov.l	@(16,r14),r2
+	mov.b	@r2,r3
+	mov.b	r3,@r1
+	mov.l	@(16,r14),r1
+	mov	r1,r2
+	add	#1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@(20,r14),r1
+	mov	r1,r2
+	add	#-1,r2
+	mov.l	r2,@(20,r14)
+	mov.l	@(12,r14),r1
+	mov	r14,r2
+	mov	r14,r3
+	add	#24,r3
+	mov.b	@r3,r2
+	mov.b	r2,@r1
+	mov.l	@(12,r14),r1
+	mov	r1,r2
+	add	#1,r2
+	mov.l	r2,@(12,r14)
+	bra	.L7
+	nop
+	.align 2
+.L8:
+.L6:
+	bra	.L5
+	nop
+	.align 2
+.L10:
+	bra	.L4
+	nop
+	.align 2
+.L5:
+	nop
+.L11:
+	mov.l	@(16,r14),r1
+	mov	#3,r2
+	and	r1,r2
+	tst	r2,r2
+	bf	.L14
+	mov	r15,r2
+	mov.l	@(12,r14),r1
+	mov.l	@(16,r14),r2
+	mov.l	@(8,r14),r7
+	mov	r7,r3
+	shlr2	r3
+	mov	r1,r4
+	mov	r2,r5
+	mov	r3,r6
+	mov.l	.L46,r8
+	jsr	@r8
+	nop
+	bra	.L15
+	nop
+	.align 2
+.L14:
+	mov	r15,r2
+	mov.l	@(12,r14),r1
+	mov.l	@(16,r14),r2
+	mov.l	@(8,r14),r7
+	mov	r7,r3
+	shlr2	r3
+	mov	r1,r4
+	mov	r2,r5
+	mov	r3,r6
+	mov.l	.L47,r8
+	jsr	@r8
+	nop
+.L15:
+	mov.l	@(8,r14),r1
+	mov	#-4,r2
+	and	r2,r1
+	mov.l	@(16,r14),r2
+	add	r2,r1
+	mov.l	r1,@(16,r14)
+	mov.l	@(8,r14),r1
+	mov	#-4,r2
+	and	r2,r1
+	mov.l	@(12,r14),r2
+	add	r2,r1
+	mov.l	r1,@(12,r14)
+	mov.l	@(8,r14),r1
+	mov	#3,r2
+	and	r1,r2
+	mov.l	r2,@(8,r14)
+.L13:
+.L52:
+	bra	.L3
+	nop
+	.align 2
+.L16:
+	bra	.L11
+	nop
+	.align 2
+.L12:
+.L3:
+	nop
+.L17:
+	mov.l	@(8,r14),r1
+	mov.l	r1,@(20,r14)
+.L20:
+	mov.l	@(20,r14),r1
+	cmp/pl	r1
+	bt	.L22
+	bra	.L19
+	nop
+	.align 2
+.L22:
+	mov	r14,r2
+	mov	r14,r1
+	add	#24,r1
+	mov.l	@(16,r14),r2
+	mov.b	@r2,r3
+	mov.b	r3,@r1
+	mov.l	@(16,r14),r1
+	mov	r1,r2
+	add	#1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@(20,r14),r1
+	mov	r1,r2
+	add	#-1,r2
+	mov.l	r2,@(20,r14)
+	mov.l	@(12,r14),r1
+	mov	r14,r2
+	mov	r14,r3
+	add	#24,r3
+	mov.b	@r3,r2
+	mov.b	r2,@r1
+	mov.l	@(12,r14),r1
+	mov	r1,r2
+	add	#1,r2
+	mov.l	r2,@(12,r14)
+	bra	.L20
+	nop
+	.align 2
+.L21:
+.L19:
+	bra	.L18
+	nop
+	.align 2
+.L23:
+	bra	.L17
+	nop
+	.align 2
+.L18:
+	bra	.L24
+	nop
+	.align 2
+.L2:
+	mov.l	@(16,r14),r1
+	mov.l	@(8,r14),r2
+	add	r2,r1
+	mov.l	r1,@(16,r14)
+	mov.l	@(12,r14),r1
+	mov.l	@(8,r14),r2
+	add	r2,r1
+	mov.l	r1,@(12,r14)
+	mov.l	@(8,r14),r1
+	mov	#15,r2
+	cmp/gt	r2,r1
+	bt	.LF101
+	bra	.L53
+	nop
+.LF101:
+	mov.l	@(12,r14),r1
+	mov	#3,r2
+	and	r1,r2
+	mov.l	@(8,r14),r1
+	mov	r1,r9
+	sub	r2,r9
+	mov	r9,r2
+	mov.l	r2,@(8,r14)
+.L26:
+	mov.l	@(12,r14),r1
+	mov	#3,r2
+	and	r1,r2
+	mov.l	r2,@(20,r14)
+.L29:
+	mov.l	@(20,r14),r1
+	cmp/pl	r1
+	bt	.L31
+	bra	.L28
+	nop
+	.align 2
+.L31:
+	mov.l	@(16,r14),r1
+	mov	r1,r2
+	add	#-1,r2
+	mov.l	r2,@(16,r14)
+	mov	r14,r2
+	mov	r14,r1
+	add	#24,r1
+	mov.l	@(16,r14),r2
+	mov.b	@r2,r3
+	mov.b	r3,@r1
+	mov.l	@(12,r14),r1
+	mov	r1,r2
+	add	#-1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@(20,r14),r1
+	mov	r1,r2
+	add	#-1,r2
+	mov.l	r2,@(20,r14)
+	mov.l	@(12,r14),r1
+	mov	r14,r2
+	mov	r14,r3
+	add	#24,r3
+	mov.b	@r3,r2
+	mov.b	r2,@r1
+	bra	.L29
+	nop
+	.align 2
+.L30:
+.L28:
+	bra	.L27
+	nop
+	.align 2
+.L32:
+	bra	.L26
+	nop
+	.align 2
+.L27:
+	nop
+.L33:
+	mov.l	@(16,r14),r1
+	mov	#3,r2
+	and	r1,r2
+	tst	r2,r2
+	bf	.L36
+	mov	r15,r2
+	mov.l	@(12,r14),r1
+	mov.l	@(16,r14),r2
+	mov.l	@(8,r14),r7
+	mov	r7,r3
+	shlr2	r3
+	mov	r1,r4
+	mov	r2,r5
+	mov	r3,r6
+	mov.l	.L48,r8
+	jsr	@r8
+	nop
+	bra	.L37
+	nop
+	.align 2
+.L36:
+	mov	r15,r2
+	mov.l	@(12,r14),r1
+	mov.l	@(16,r14),r2
+	mov.l	@(8,r14),r7
+	mov	r7,r3
+	shlr2	r3
+	mov	r1,r4
+	mov	r2,r5
+	mov	r3,r6
+	mov.l	.L49,r8
+	jsr	@r8
+	nop
+.L37:
+	mov.l	@(8,r14),r1
+	mov	#-4,r2
+	and	r2,r1
+	mov.l	@(16,r14),r2
+	mov	r2,r9
+	sub	r1,r9
+	mov	r9,r1
+	mov.l	r1,@(16,r14)
+	mov.l	@(8,r14),r1
+	mov	#-4,r2
+	and	r2,r1
+	mov.l	@(12,r14),r2
+	mov	r2,r9
+	sub	r1,r9
+	mov	r9,r1
+	mov.l	r1,@(12,r14)
+	mov.l	@(8,r14),r1
+	mov	#3,r2
+	and	r1,r2
+	mov.l	r2,@(8,r14)
+.L35:
+.L53:
+	bra	.L25
+	nop
+	.align 2
+.L38:
+	bra	.L33
+	nop
+	.align 2
+.L34:
+.L25:
+	nop
+.L39:
+	mov.l	@(8,r14),r1
+	mov.l	r1,@(20,r14)
+.L42:
+	mov.l	@(20,r14),r1
+	cmp/pl	r1
+	bt	.L44
+	bra	.L41
+	nop
+	.align 2
+.L44:
+	mov.l	@(16,r14),r1
+	mov	r1,r2
+	add	#-1,r2
+	mov.l	r2,@(16,r14)
+	mov	r14,r2
+	mov	r14,r1
+	add	#24,r1
+	mov.l	@(16,r14),r2
+	mov.b	@r2,r3
+	mov.b	r3,@r1
+	mov.l	@(12,r14),r1
+	mov	r1,r2
+	add	#-1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@(20,r14),r1
+	mov	r1,r2
+	add	#-1,r2
+	mov.l	r2,@(20,r14)
+	mov.l	@(12,r14),r1
+	mov	r14,r2
+	mov	r14,r3
+	add	#24,r3
+	mov.b	@r3,r2
+	mov.b	r2,@r1
+	bra	.L42
+	nop
+	.align 2
+.L43:
+.L41:
+	bra	.L24
+	nop
+	.align 2
+.L45:
+	bra	.L39
+	nop
+	.align 2
+.L40:
+.L24:
+	mov.l	@r14,r1
+	mov	r1,r0
+	bra	.L1
+	nop
+	.align 2
+.L1:
+	add	#28,r14
+	mov	r14,r15
+	lds.l	@r15+,pr
+	mov.l	@r15+,r14
+	mov.l	@r15+,r9
+	mov.l	@r15+,r8
+	rts	
+	 nop
+.L50:
+	.align 2
+.L46:
+	.long	__wordcopy_fwd_aligned
+.L47:
+	.long	__wordcopy_fwd_dest_aligned
+.L48:
+	.long	__wordcopy_bwd_aligned
+.L49:
+	.long	__wordcopy_bwd_dest_aligned
+.Lfe1:
diff --git a/arch/sh/lib/memset.S b/arch/sh/lib/memset.S
new file mode 100644
index 000000000..c97648a4a
--- /dev/null
+++ b/arch/sh/lib/memset.S
@@ -0,0 +1,72 @@
+! Taken from newlib-1.8.0
+
+!
+! Fast SH memset
+!
+! by Toshiyasu Morita (tm@netcom.com)
+!
+! Entry: r4: destination pointer
+!        r5: fill value
+!        r6: byte count
+!
+! Exit:  r0-r3: trashed
+!
+#include <linux/linkage.h>
+
+ENTRY(memset)
+	mov	r4,r3	! Save return value
+	
+	mov	r6,r0	! Check explicitly for zero
+	cmp/eq	#0,r0
+	bt	L_exit
+
+	mov	#12,r0	! Check for small number of bytes
+	cmp/gt	r6,r0
+	bt	L_store_byte_loop
+
+	neg	r4,r0	! Align destination
+	add	#4,r0
+	and	#3,r0
+	tst	r0,r0
+	bt	L_dup_bytes
+	.balignw 4,0x0009
+L_align_loop:
+	mov.b	r5,@r4
+	add	#-1,r6
+	add	#1,r4
+	dt	r0
+	bf	L_align_loop
+
+L_dup_bytes:	
+	extu.b	r5,r5	! Duplicate bytes across longword
+	swap.b	r5,r0
+	or	r0,r5
+	swap.w	r5,r0
+	or	r0,r5
+
+	mov	r6,r2	! Calculate number of double longwords
+	shlr2	r2
+	shlr	r2
+
+	.balignw 4,0x0009
+L_store_long_loop:
+	mov.l	r5,@r4	! Store double longs to memory
+	dt	r2
+	mov.l	r5,@(4,r4)
+	add	#8,r4
+	bf	L_store_long_loop
+
+	mov	#7,r0
+	and	r0,r6
+	tst	r6,r6
+	bt	L_exit
+	.balignw 4,0x0009
+L_store_byte_loop:
+	mov.b	r5,@r4	! Store bytes to memory
+	add	#1,r4
+	dt	r6
+	bf	L_store_byte_loop
+
+L_exit:
+	rts
+	 mov	r3,r0
diff --git a/arch/sh/lib/old-checksum.c b/arch/sh/lib/old-checksum.c
new file mode 100644
index 000000000..ae3a38043
--- /dev/null
+++ b/arch/sh/lib/old-checksum.c
@@ -0,0 +1,19 @@
+/*
+ * FIXME: old compatibility stuff, will be removed soon.
+ */
+
+#include <net/checksum.h>
+
+unsigned int csum_partial_copy( const char *src, char *dst, int len, int sum)
+{
+	int src_err=0, dst_err=0;
+
+	sum = csum_partial_copy_generic ( src, dst, len, sum, &src_err, &dst_err);
+
+	if (src_err || dst_err)
+		printk("old csum_partial_copy_fromuser(), tell mingo to convert me.\n");
+
+	return sum;
+}
+
+
diff --git a/arch/sh/lib/wordcopy.S b/arch/sh/lib/wordcopy.S
new file mode 100644
index 000000000..c116623d0
--- /dev/null
+++ b/arch/sh/lib/wordcopy.S
@@ -0,0 +1,1289 @@
+#include <linux/linkage.h>
+ENTRY(_wordcopy_fwd_aligned)
+	mov.l	r14,@-r15
+	add	#-20,r15
+	mov	r15,r14
+	mov.l	r4,@r14
+	mov.l	r5,@(4,r14)
+	mov.l	r6,@(8,r14)
+	mov.l	@(8,r14),r2
+	mov	#7,r1
+	and	r2,r1
+	mov	#0,r2
+	mov	#7,r3
+	sub	r2,r1
+	cmp/hi	r3,r1
+	bf	.L29
+	bra	.L2
+	nop
+.L29:
+	mova	.L22,r0
+	add	r1,r1
+	mov.w	@(r0,r1),r1
+	add	r0,r1
+	jmp  @r1
+	nop
+	.align 2
+	.align 2
+.L22:
+	.word	.L15-.L22
+	.word	.L18-.L22
+	.word	.L3-.L22
+	.word	.L5-.L22
+	.word	.L7-.L22
+	.word	.L9-.L22
+	.word	.L11-.L22
+	.word	.L13-.L22
+	.align 2
+.L3:
+	mov.l	@(4,r14),r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-24,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-28,r2
+	mov.l	r2,@r14
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#6,r2
+	mov.l	r2,@(8,r14)
+	bra	.L4
+	nop
+	.align 2
+.L5:
+	mov.l	@(4,r14),r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-20,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-24,r2
+	mov.l	r2,@r14
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#5,r2
+	mov.l	r2,@(8,r14)
+	bra	.L6
+	nop
+	.align 2
+.L7:
+	mov.l	@(4,r14),r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-16,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-20,r2
+	mov.l	r2,@r14
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#4,r2
+	mov.l	r2,@(8,r14)
+	bra	.L8
+	nop
+	.align 2
+.L9:
+	mov.l	@(4,r14),r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-12,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-16,r2
+	mov.l	r2,@r14
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#3,r2
+	mov.l	r2,@(8,r14)
+	bra	.L10
+	nop
+	.align 2
+.L11:
+	mov.l	@(4,r14),r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-8,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-12,r2
+	mov.l	r2,@r14
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#2,r2
+	mov.l	r2,@(8,r14)
+	bra	.L12
+	nop
+	.align 2
+.L13:
+	mov.l	@(4,r14),r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-4,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-8,r2
+	mov.l	r2,@r14
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#1,r2
+	mov.l	r2,@(8,r14)
+	bra	.L14
+	nop
+	.align 2
+.L15:
+	bra	.L16
+	nop
+	bra	.L1
+	nop
+	.align 2
+.L16:
+	mov.l	@(4,r14),r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-4,r2
+	mov.l	r2,@r14
+	bra	.L17
+	nop
+	.align 2
+.L18:
+	mov.l	@(4,r14),r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#4,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#-1,r2
+	mov.l	r2,@(8,r14)
+	bra	.L19
+	nop
+	bra	.L20
+	nop
+	.align 2
+.L19:
+	bra	.L21
+	nop
+	.align 2
+.L23:
+.L2:
+	nop
+.L24:
+.L21:
+	mov.l	@(4,r14),r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@r14,r1
+	mov.l	@(16,r14),r2
+	mov.l	r2,@r1
+.L17:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#4,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@r14,r2
+	mov	r2,r1
+	add	#4,r1
+	mov.l	@(12,r14),r2
+	mov.l	r2,@r1
+.L14:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#8,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@r14,r2
+	mov	r2,r1
+	add	#8,r1
+	mov.l	@(16,r14),r2
+	mov.l	r2,@r1
+.L12:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#12,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@r14,r2
+	mov	r2,r1
+	add	#12,r1
+	mov.l	@(12,r14),r2
+	mov.l	r2,@r1
+.L10:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#16,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@r14,r2
+	mov	r2,r1
+	add	#16,r1
+	mov.l	@(16,r14),r2
+	mov.l	r2,@r1
+.L8:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#20,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@r14,r2
+	mov	r2,r1
+	add	#20,r1
+	mov.l	@(12,r14),r2
+	mov.l	r2,@r1
+.L6:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#24,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@r14,r2
+	mov	r2,r1
+	add	#24,r1
+	mov.l	@(16,r14),r2
+	mov.l	r2,@r1
+.L4:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#28,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@r14,r2
+	mov	r2,r1
+	add	#28,r1
+	mov.l	@(12,r14),r2
+	mov.l	r2,@r1
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#32,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#32,r2
+	mov.l	r2,@r14
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#-8,r2
+	mov.l	r2,@(8,r14)
+.L26:
+	mov.l	@(8,r14),r1
+	tst	r1,r1
+	bf	.L27
+	bra	.L25
+	nop
+	.align 2
+.L27:
+	bra	.L21
+	nop
+	.align 2
+.L25:
+	nop
+.L20:
+	mov.l	@r14,r1
+	mov.l	@(16,r14),r2
+	mov.l	r2,@r1
+.L1:
+	add	#20,r14
+	mov	r14,r15
+	mov.l	@r15+,r14
+	rts	
+	nop
+.Lfe1:
+	.size	 __wordcopy_fwd_aligned,.Lfe1-__wordcopy_fwd_aligned
+	.global	___lshrsi3
+	.global	___ashlsi3
+	.align 2
+	.global	__wordcopy_fwd_dest_aligned
+	.type	 __wordcopy_fwd_dest_aligned,@function
+__wordcopy_fwd_dest_aligned:
+	mov.l	r8,@-r15
+	mov.l	r9,@-r15
+	mov.l	r14,@-r15
+	sts.l	pr,@-r15
+	add	#-40,r15
+	mov	r15,r14
+	mov.l	r4,@r14
+	mov.l	r5,@(4,r14)
+	mov.l	r6,@(8,r14)
+	mov.l	@(4,r14),r1
+	mov	#3,r2
+	and	r1,r2
+	mov	r2,r1
+	mov	r1,r2
+	shll2	r2
+	add	r2,r2
+	mov.l	r2,@(28,r14)
+	mov.l	@(28,r14),r2
+	neg	r2,r1
+	add	#32,r1
+	mov.l	r1,@(32,r14)
+	mov.l	@(4,r14),r1
+	mov	#-4,r2
+	and	r2,r1
+	mov.l	r1,@(4,r14)
+	mov.l	@(8,r14),r2
+	mov	#3,r1
+	and	r2,r1
+	mov	#0,r2
+	mov	#3,r3
+	sub	r2,r1
+	cmp/hi	r3,r1
+	bf	.L53
+	bra	.L31
+	nop
+.L53:
+	mova	.L43,r0
+	add	r1,r1
+	mov.w	@(r0,r1),r1
+	add	r0,r1
+	jmp  @r1
+	nop
+	.align 2
+	.align 2
+.L43:
+	.word	.L36-.L43
+	.word	.L39-.L43
+	.word	.L32-.L43
+	.word	.L34-.L43
+	.align 2
+.L32:
+	mov.l	@(4,r14),r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#4,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(20,r14)
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-4,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-12,r2
+	mov.l	r2,@r14
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#2,r2
+	mov.l	r2,@(8,r14)
+	bra	.L33
+	nop
+	.align 2
+.L34:
+	mov.l	@(4,r14),r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#4,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-8,r2
+	mov.l	r2,@r14
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#1,r2
+	mov.l	r2,@(8,r14)
+	bra	.L35
+	nop
+	.align 2
+.L36:
+	bra	.L37
+	nop
+	bra	.L30
+	nop
+	.align 2
+.L37:
+	mov.l	@(4,r14),r1
+	mov.l	@r1,r2
+	mov.l	r2,@(24,r14)
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#4,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#4,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-4,r2
+	mov.l	r2,@r14
+	bra	.L38
+	nop
+	.align 2
+.L39:
+	mov.l	@(4,r14),r1
+	mov.l	@r1,r2
+	mov.l	r2,@(20,r14)
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#4,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(24,r14)
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#8,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#-1,r2
+	mov.l	r2,@(8,r14)
+	bra	.L40
+	nop
+	bra	.L41
+	nop
+	.align 2
+.L40:
+	bra	.L42
+	nop
+	.align 2
+.L44:
+.L31:
+	nop
+.L45:
+.L42:
+	mov.l	@(4,r14),r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@r14,r8
+	mov.l	.L49,r1
+	mov.l	@(20,r14),r4
+	mov.l	@(28,r14),r5
+	jsr	@r1
+	nop
+	mov	r0,r9
+	mov.l	.L50,r1
+	mov.l	@(24,r14),r4
+	mov.l	@(32,r14),r5
+	jsr	@r1
+	nop
+	mov.l	r0,@(36,r14)
+	mov.l	@(36,r14),r1
+	or	r9,r1
+	mov.l	r1,@r8
+.L38:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#4,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@r14,r1
+	mov	r1,r8
+	add	#4,r8
+	mov.l	.L49,r1
+	mov.l	@(24,r14),r4
+	mov.l	@(28,r14),r5
+	jsr	@r1
+	nop
+	mov	r0,r9
+	mov.l	.L50,r1
+	mov.l	@(12,r14),r4
+	mov.l	@(32,r14),r5
+	jsr	@r1
+	nop
+	mov.l	r0,@(36,r14)
+	mov.l	@(36,r14),r1
+	or	r9,r1
+	mov.l	r1,@r8
+.L35:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#8,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(20,r14)
+	mov.l	@r14,r1
+	mov	r1,r8
+	add	#8,r8
+	mov.l	.L49,r1
+	mov.l	@(12,r14),r4
+	mov.l	@(28,r14),r5
+	jsr	@r1
+	nop
+	mov	r0,r9
+	mov.l	.L50,r1
+	mov.l	@(16,r14),r4
+	mov.l	@(32,r14),r5
+	jsr	@r1
+	nop
+	mov.l	r0,@(36,r14)
+	mov.l	@(36,r14),r1
+	or	r9,r1
+	mov.l	r1,@r8
+.L33:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#12,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(24,r14)
+	mov.l	@r14,r1
+	mov	r1,r8
+	add	#12,r8
+	mov.l	.L49,r1
+	mov.l	@(16,r14),r4
+	mov.l	@(28,r14),r5
+	jsr	@r1
+	nop
+	mov	r0,r9
+	mov.l	.L50,r1
+	mov.l	@(20,r14),r4
+	mov.l	@(32,r14),r5
+	jsr	@r1
+	nop
+	mov.l	r0,@(36,r14)
+	mov.l	@(36,r14),r1
+	or	r9,r1
+	mov.l	r1,@r8
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#16,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#16,r2
+	mov.l	r2,@r14
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#-4,r2
+	mov.l	r2,@(8,r14)
+.L47:
+	mov.l	@(8,r14),r1
+	tst	r1,r1
+	bf	.L48
+	bra	.L46
+	nop
+	.align 2
+.L48:
+	bra	.L42
+	nop
+	.align 2
+.L46:
+	nop
+.L41:
+	mov.l	@r14,r8
+	mov.l	.L49,r1
+	mov.l	@(20,r14),r4
+	mov.l	@(28,r14),r5
+	jsr	@r1
+	nop
+	mov	r0,r9
+	mov.l	.L50,r1
+	mov.l	@(24,r14),r4
+	mov.l	@(32,r14),r5
+	jsr	@r1
+	nop
+	mov.l	r0,@(36,r14)
+	mov.l	@(36,r14),r1
+	or	r9,r1
+	mov.l	r1,@r8
+.L30:
+	add	#40,r14
+	mov	r14,r15
+	lds.l	@r15+,pr
+	mov.l	@r15+,r14
+	mov.l	@r15+,r9
+	mov.l	@r15+,r8
+	rts	
+	nop
+.L51:
+	.align 2
+.L49:
+	.long	___lshrsi3
+.L50:
+	.long	___ashlsi3
+.Lfe2:
+	.size	 __wordcopy_fwd_dest_aligned,.Lfe2-__wordcopy_fwd_dest_aligned
+	.align 2
+	.global	__wordcopy_bwd_aligned
+	.type	 __wordcopy_bwd_aligned,@function
+__wordcopy_bwd_aligned:
+	mov.l	r14,@-r15
+	add	#-20,r15
+	mov	r15,r14
+	mov.l	r4,@r14
+	mov.l	r5,@(4,r14)
+	mov.l	r6,@(8,r14)
+	mov.l	@(8,r14),r2
+	mov	#7,r1
+	and	r2,r1
+	mov	#0,r2
+	mov	#7,r3
+	sub	r2,r1
+	cmp/hi	r3,r1
+	bf	.L82
+	bra	.L55
+	nop
+.L82:
+	mova	.L75,r0
+	add	r1,r1
+	mov.w	@(r0,r1),r1
+	add	r0,r1
+	jmp  @r1
+	nop
+	.align 2
+	.align 2
+.L75:
+	.word	.L68-.L75
+	.word	.L71-.L75
+	.word	.L56-.L75
+	.word	.L58-.L75
+	.word	.L60-.L75
+	.word	.L62-.L75
+	.word	.L64-.L75
+	.word	.L66-.L75
+	.align 2
+.L56:
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-8,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-4,r2
+	mov.l	r2,@r14
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#4,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#6,r2
+	mov.l	r2,@(8,r14)
+	bra	.L57
+	nop
+	.align 2
+.L58:
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-12,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-8,r2
+	mov.l	r2,@r14
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#8,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#5,r2
+	mov.l	r2,@(8,r14)
+	bra	.L59
+	nop
+	.align 2
+.L60:
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-16,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-12,r2
+	mov.l	r2,@r14
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#12,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#4,r2
+	mov.l	r2,@(8,r14)
+	bra	.L61
+	nop
+	.align 2
+.L62:
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-20,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-16,r2
+	mov.l	r2,@r14
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#16,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#3,r2
+	mov.l	r2,@(8,r14)
+	bra	.L63
+	nop
+	.align 2
+.L64:
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-24,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-20,r2
+	mov.l	r2,@r14
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#20,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#2,r2
+	mov.l	r2,@(8,r14)
+	bra	.L65
+	nop
+	.align 2
+.L66:
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-28,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-24,r2
+	mov.l	r2,@r14
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#24,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#1,r2
+	mov.l	r2,@(8,r14)
+	bra	.L67
+	nop
+	.align 2
+.L68:
+	bra	.L69
+	nop
+	bra	.L54
+	nop
+	.align 2
+.L69:
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-32,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-28,r2
+	mov.l	r2,@r14
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#28,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	bra	.L70
+	nop
+	.align 2
+.L71:
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-36,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-32,r2
+	mov.l	r2,@r14
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#32,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#-1,r2
+	mov.l	r2,@(8,r14)
+	bra	.L72
+	nop
+	bra	.L73
+	nop
+	.align 2
+.L72:
+	bra	.L74
+	nop
+	.align 2
+.L76:
+.L55:
+	nop
+.L77:
+.L74:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#28,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@r14,r2
+	mov	r2,r1
+	add	#28,r1
+	mov.l	@(16,r14),r2
+	mov.l	r2,@r1
+.L70:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#24,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@r14,r2
+	mov	r2,r1
+	add	#24,r1
+	mov.l	@(12,r14),r2
+	mov.l	r2,@r1
+.L67:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#20,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@r14,r2
+	mov	r2,r1
+	add	#20,r1
+	mov.l	@(16,r14),r2
+	mov.l	r2,@r1
+.L65:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#16,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@r14,r2
+	mov	r2,r1
+	add	#16,r1
+	mov.l	@(12,r14),r2
+	mov.l	r2,@r1
+.L63:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#12,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@r14,r2
+	mov	r2,r1
+	add	#12,r1
+	mov.l	@(16,r14),r2
+	mov.l	r2,@r1
+.L61:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#8,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@r14,r2
+	mov	r2,r1
+	add	#8,r1
+	mov.l	@(12,r14),r2
+	mov.l	r2,@r1
+.L59:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#4,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@r14,r2
+	mov	r2,r1
+	add	#4,r1
+	mov.l	@(16,r14),r2
+	mov.l	r2,@r1
+.L57:
+	mov.l	@(4,r14),r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@r14,r1
+	mov.l	@(12,r14),r2
+	mov.l	r2,@r1
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-32,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-32,r2
+	mov.l	r2,@r14
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#-8,r2
+	mov.l	r2,@(8,r14)
+.L79:
+	mov.l	@(8,r14),r1
+	tst	r1,r1
+	bf	.L80
+	bra	.L78
+	nop
+	.align 2
+.L80:
+	bra	.L74
+	nop
+	.align 2
+.L78:
+	nop
+.L73:
+	mov.l	@r14,r2
+	mov	r2,r1
+	add	#28,r1
+	mov.l	@(16,r14),r2
+	mov.l	r2,@r1
+.L54:
+	add	#20,r14
+	mov	r14,r15
+	mov.l	@r15+,r14
+	rts	
+	nop
+.Lfe3:
+	.size	 __wordcopy_bwd_aligned,.Lfe3-__wordcopy_bwd_aligned
+	.align 2
+	.global	__wordcopy_bwd_dest_aligned
+	.type	 __wordcopy_bwd_dest_aligned,@function
+__wordcopy_bwd_dest_aligned:
+	mov.l	r8,@-r15
+	mov.l	r9,@-r15
+	mov.l	r14,@-r15
+	sts.l	pr,@-r15
+	add	#-40,r15
+	mov	r15,r14
+	mov.l	r4,@r14
+	mov.l	r5,@(4,r14)
+	mov.l	r6,@(8,r14)
+	mov.l	@(4,r14),r1
+	mov	#3,r2
+	and	r1,r2
+	mov	r2,r1
+	mov	r1,r2
+	shll2	r2
+	add	r2,r2
+	mov.l	r2,@(28,r14)
+	mov.l	@(28,r14),r2
+	neg	r2,r1
+	add	#32,r1
+	mov.l	r1,@(32,r14)
+	mov.l	@(4,r14),r1
+	mov	#-4,r2
+	and	r2,r1
+	mov.l	r1,@(4,r14)
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#4,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@(8,r14),r2
+	mov	#3,r1
+	and	r2,r1
+	mov	#0,r2
+	mov	#3,r3
+	sub	r2,r1
+	cmp/hi	r3,r1
+	bf	.L106
+	bra	.L84
+	nop
+.L106:
+	mova	.L96,r0
+	add	r1,r1
+	mov.w	@(r0,r1),r1
+	add	r0,r1
+	jmp  @r1
+	nop
+	.align 2
+	.align 2
+.L96:
+	.word	.L89-.L96
+	.word	.L92-.L96
+	.word	.L85-.L96
+	.word	.L87-.L96
+	.align 2
+.L85:
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-12,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-4,r2
+	mov.l	r2,@r14
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#8,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(20,r14)
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#4,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#2,r2
+	mov.l	r2,@(8,r14)
+	bra	.L86
+	nop
+	.align 2
+.L87:
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-16,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-8,r2
+	mov.l	r2,@r14
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#12,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(24,r14)
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#8,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(20,r14)
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#1,r2
+	mov.l	r2,@(8,r14)
+	bra	.L88
+	nop
+	.align 2
+.L89:
+	bra	.L90
+	nop
+	bra	.L83
+	nop
+	.align 2
+.L90:
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-20,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-12,r2
+	mov.l	r2,@r14
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#16,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#12,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(24,r14)
+	bra	.L91
+	nop
+	.align 2
+.L92:
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-24,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-16,r2
+	mov.l	r2,@r14
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#20,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#16,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#-1,r2
+	mov.l	r2,@(8,r14)
+	bra	.L93
+	nop
+	bra	.L94
+	nop
+	.align 2
+.L93:
+	bra	.L95
+	nop
+	.align 2
+.L97:
+.L84:
+	nop
+.L98:
+.L95:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#12,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(24,r14)
+	mov.l	@r14,r1
+	mov	r1,r8
+	add	#12,r8
+	mov.l	.L102,r1
+	mov.l	@(12,r14),r4
+	mov.l	@(28,r14),r5
+	jsr	@r1
+	nop
+	mov	r0,r9
+	mov.l	.L103,r1
+	mov.l	@(16,r14),r4
+	mov.l	@(32,r14),r5
+	jsr	@r1
+	nop
+	mov.l	r0,@(36,r14)
+	mov.l	@(36,r14),r1
+	or	r9,r1
+	mov.l	r1,@r8
+.L91:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#8,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(20,r14)
+	mov.l	@r14,r1
+	mov	r1,r8
+	add	#8,r8
+	mov.l	.L102,r1
+	mov.l	@(24,r14),r4
+	mov.l	@(28,r14),r5
+	jsr	@r1
+	nop
+	mov	r0,r9
+	mov.l	.L103,r1
+	mov.l	@(12,r14),r4
+	mov.l	@(32,r14),r5
+	jsr	@r1
+	nop
+	mov.l	r0,@(36,r14)
+	mov.l	@(36,r14),r1
+	or	r9,r1
+	mov.l	r1,@r8
+.L88:
+	mov.l	@(4,r14),r2
+	mov	r2,r1
+	add	#4,r1
+	mov.l	@r1,r2
+	mov.l	r2,@(16,r14)
+	mov.l	@r14,r1
+	mov	r1,r8
+	add	#4,r8
+	mov.l	.L102,r1
+	mov.l	@(20,r14),r4
+	mov.l	@(28,r14),r5
+	jsr	@r1
+	nop
+	mov	r0,r9
+	mov.l	.L103,r1
+	mov.l	@(24,r14),r4
+	mov.l	@(32,r14),r5
+	jsr	@r1
+	nop
+	mov.l	r0,@(36,r14)
+	mov.l	@(36,r14),r1
+	or	r9,r1
+	mov.l	r1,@r8
+.L86:
+	mov.l	@(4,r14),r1
+	mov.l	@r1,r2
+	mov.l	r2,@(12,r14)
+	mov.l	@r14,r8
+	mov.l	.L102,r1
+	mov.l	@(16,r14),r4
+	mov.l	@(28,r14),r5
+	jsr	@r1
+	nop
+	mov	r0,r9
+	mov.l	.L103,r1
+	mov.l	@(20,r14),r4
+	mov.l	@(32,r14),r5
+	jsr	@r1
+	nop
+	mov.l	r0,@(36,r14)
+	mov.l	@(36,r14),r1
+	or	r9,r1
+	mov.l	r1,@r8
+	mov.l	@(4,r14),r1
+	mov	r1,r2
+	add	#-16,r2
+	mov.l	r2,@(4,r14)
+	mov.l	@r14,r1
+	mov	r1,r2
+	add	#-16,r2
+	mov.l	r2,@r14
+	mov.l	@(8,r14),r1
+	mov	r1,r2
+	add	#-4,r2
+	mov.l	r2,@(8,r14)
+.L100:
+	mov.l	@(8,r14),r1
+	tst	r1,r1
+	bf	.L101
+	bra	.L99
+	nop
+	.align 2
+.L101:
+	bra	.L95
+	nop
+	.align 2
+.L99:
+	nop
+.L94:
+	mov.l	@r14,r1
+	mov	r1,r8
+	add	#12,r8
+	mov.l	.L102,r1
+	mov.l	@(12,r14),r4
+	mov.l	@(28,r14),r5
+	jsr	@r1
+	nop
+	mov	r0,r9
+	mov.l	.L103,r1
+	mov.l	@(16,r14),r4
+	mov.l	@(32,r14),r5
+	jsr	@r1
+	nop
+	mov.l	r0,@(36,r14)
+	mov.l	@(36,r14),r1
+	or	r9,r1
+	mov.l	r1,@r8
+.L83:
+	add	#40,r14
+	mov	r14,r15
+	lds.l	@r15+,pr
+	mov.l	@r15+,r14
+	mov.l	@r15+,r9
+	mov.l	@r15+,r8
+	rts	
+	nop
+.L104:
+	.align 2
+.L102:
+	.long	___lshrsi3
+.L103:
+	.long	___ashlsi3
+.Lfe4:
-- 
cgit v1.2.3