Linux-2.6.12-rc2v2.6.12-rc2

Initial git repository build. I'm not bothering with the full history,
even though we have it. We can create a separate "historical" git
archive of that later if we want to, and in the meantime it's about
3.2GB when imported into git - space that would just make the early
git days unnecessarily complicated, when we don't have a lot of good
infrastructure for it.

Let it rip!

9 files changed, 866 insertions, 0 deletions

diff --git a/arch/mips/lib/Makefile b/arch/mips/lib/Makefile
new file mode 100644
index 000000000000..21b92b9dd013
--- /dev/null
+++ b/arch/mips/lib/Makefile
@@ -0,0 +1,10 @@
+#
+# Makefile for MIPS-specific library files..
+#
+
+lib-y	+= csum_partial_copy.o dec_and_lock.o memcpy.o promlib.o \
+	   strlen_user.o strncpy_user.o strnlen_user.o
+
+obj-y	+= iomap.o
+
+EXTRA_AFLAGS := $(CFLAGS)
diff --git a/arch/mips/lib/csum_partial_copy.c b/arch/mips/lib/csum_partial_copy.c
new file mode 100644
index 000000000000..ffed0a6a1c16
--- /dev/null
+++ b/arch/mips/lib/csum_partial_copy.c
@@ -0,0 +1,49 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (C) 1994, 1995 Waldorf Electronics GmbH
+ * Copyright (C) 1998, 1999 Ralf Baechle
+ */
+#include <linux/kernel.h>
+#include <linux/types.h>
+#include <asm/byteorder.h>
+#include <asm/string.h>
+#include <asm/uaccess.h>
+#include <net/checksum.h>
+
+/*
+ * copy while checksumming, otherwise like csum_partial
+ */
+unsigned int csum_partial_copy_nocheck(const unsigned char *src, unsigned 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 unsigned char *src, unsigned char *dst,
+	int len, unsigned int sum, int *err_ptr)
+{
+	int missing;
+
+	might_sleep();
+	missing = copy_from_user(dst, src, len);
+	if (missing) {
+		memset(dst + len - missing, 0, missing);
+		*err_ptr = -EFAULT;
+	}
+
+	return csum_partial(dst, len, sum);
+}
diff --git a/arch/mips/lib/dec_and_lock.c b/arch/mips/lib/dec_and_lock.c
new file mode 100644
index 000000000000..e44e9579bd36
--- /dev/null
+++ b/arch/mips/lib/dec_and_lock.c
@@ -0,0 +1,55 @@
+/*
+ * MIPS version of atomic_dec_and_lock() using cmpxchg
+ *
+ * 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.
+ */
+
+#include <linux/module.h>
+#include <linux/spinlock.h>
+#include <asm/atomic.h>
+#include <asm/system.h>
+
+/*
+ * This is an implementation of the notion of "decrement a
+ * reference count, and return locked if it decremented to zero".
+ *
+ * This implementation can be used on any architecture that
+ * has a cmpxchg, and where atomic->value is an int holding
+ * the value of the atomic (i.e. the high bits aren't used
+ * for a lock or anything like that).
+ *
+ * N.B. ATOMIC_DEC_AND_LOCK gets defined in include/linux/spinlock.h
+ * if spinlocks are empty and thus atomic_dec_and_lock is defined
+ * to be atomic_dec_and_test - in that case we don't need it
+ * defined here as well.
+ */
+
+#ifndef ATOMIC_DEC_AND_LOCK
+int _atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock)
+{
+	int counter;
+	int newcount;
+
+	for (;;) {
+		counter = atomic_read(atomic);
+		newcount = counter - 1;
+		if (!newcount)
+			break;		/* do it the slow way */
+
+		newcount = cmpxchg(&atomic->counter, counter, newcount);
+		if (newcount == counter)
+			return 0;
+	}
+
+	spin_lock(lock);
+	if (atomic_dec_and_test(atomic))
+		return 1;
+	spin_unlock(lock);
+	return 0;
+}
+
+EXPORT_SYMBOL(_atomic_dec_and_lock);
+#endif /* ATOMIC_DEC_AND_LOCK */
diff --git a/arch/mips/lib/iomap.c b/arch/mips/lib/iomap.c
new file mode 100644
index 000000000000..b5d5fa833762
--- /dev/null
+++ b/arch/mips/lib/iomap.c
@@ -0,0 +1,78 @@
+/*
+ *  iomap.c, Memory Mapped I/O routines for MIPS architecture.
+ *
+ *  This code is based on lib/iomap.c, by Linus Torvalds.
+ *
+ *  Copyright (C) 2004-2005  Yoichi Yuasa <yuasa@hh.iij4u.or.jp>
+ *
+ *  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.
+ *
+ *  This program is distributed in the hope that it will be useful,
+ *  but WITHOUT ANY WARRANTY; without even the implied warranty of
+ *  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ *  GNU General Public License for more details.
+ *
+ *  You should have received a copy of the GNU General Public License
+ *  along with this program; if not, write to the Free Software
+ *  Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA  02111-1307  USA
+ */
+#include <linux/ioport.h>
+#include <linux/module.h>
+#include <linux/pci.h>
+
+#include <asm/io.h>
+
+void __iomem *ioport_map(unsigned long port, unsigned int nr)
+{
+	unsigned long end;
+
+	end = port + nr - 1UL;
+	if (ioport_resource.start > port ||
+	    ioport_resource.end < end || port > end)
+		return NULL;
+
+	return (void __iomem *)(mips_io_port_base + port);
+}
+
+void ioport_unmap(void __iomem *addr)
+{
+}
+EXPORT_SYMBOL(ioport_map);
+EXPORT_SYMBOL(ioport_unmap);
+
+void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long maxlen)
+{
+	unsigned long start, len, flags;
+
+	if (dev == NULL)
+		return NULL;
+
+	start = pci_resource_start(dev, bar);
+	len = pci_resource_len(dev, bar);
+	if (!start || !len)
+		return NULL;
+
+	if (maxlen != 0 && len > maxlen)
+		len = maxlen;
+
+	flags = pci_resource_flags(dev, bar);
+	if (flags & IORESOURCE_IO)
+		return ioport_map(start, len);
+	if (flags & IORESOURCE_MEM) {
+		if (flags & IORESOURCE_CACHEABLE)
+			return ioremap_cacheable_cow(start, len);
+		return ioremap_nocache(start, len);
+	}
+
+	return NULL;
+}
+
+void pci_iounmap(struct pci_dev *dev, void __iomem *addr)
+{
+	iounmap(addr);
+}
+EXPORT_SYMBOL(pci_iomap);
+EXPORT_SYMBOL(pci_iounmap);
diff --git a/arch/mips/lib/memcpy.S b/arch/mips/lib/memcpy.S
new file mode 100644
index 000000000000..afa8eae18ff6
--- /dev/null
+++ b/arch/mips/lib/memcpy.S
@@ -0,0 +1,508 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Unified implementation of memcpy, memmove and the __copy_user backend.
+ *
+ * Copyright (C) 1998, 99, 2000, 01, 2002 Ralf Baechle (ralf@gnu.org)
+ * Copyright (C) 1999, 2000, 01, 2002 Silicon Graphics, Inc.
+ * Copyright (C) 2002 Broadcom, Inc.
+ *   memcpy/copy_user author: Mark Vandevoorde
+ *
+ * Mnemonic names for arguments to memcpy/__copy_user
+ */
+#include <linux/config.h>
+#include <asm/asm.h>
+#include <asm/offset.h>
+#include <asm/regdef.h>
+
+#define dst a0
+#define src a1
+#define len a2
+
+/*
+ * Spec
+ *
+ * memcpy copies len bytes from src to dst and sets v0 to dst.
+ * It assumes that
+ *   - src and dst don't overlap
+ *   - src is readable
+ *   - dst is writable
+ * memcpy uses the standard calling convention
+ *
+ * __copy_user copies up to len bytes from src to dst and sets a2 (len) to
+ * the number of uncopied bytes due to an exception caused by a read or write.
+ * __copy_user assumes that src and dst don't overlap, and that the call is
+ * implementing one of the following:
+ *   copy_to_user
+ *     - src is readable  (no exceptions when reading src)
+ *   copy_from_user
+ *     - dst is writable  (no exceptions when writing dst)
+ * __copy_user uses a non-standard calling convention; see
+ * include/asm-mips/uaccess.h
+ *
+ * When an exception happens on a load, the handler must
+ # ensure that all of the destination buffer is overwritten to prevent
+ * leaking information to user mode programs.
+ */
+
+/*
+ * Implementation
+ */
+
+/*
+ * The exception handler for loads requires that:
+ *  1- AT contain the address of the byte just past the end of the source
+ *     of the copy,
+ *  2- src_entry <= src < AT, and
+ *  3- (dst - src) == (dst_entry - src_entry),
+ * The _entry suffix denotes values when __copy_user was called.
+ *
+ * (1) is set up up by uaccess.h and maintained by not writing AT in copy_user
+ * (2) is met by incrementing src by the number of bytes copied
+ * (3) is met by not doing loads between a pair of increments of dst and src
+ *
+ * The exception handlers for stores adjust len (if necessary) and return.
+ * These handlers do not need to overwrite any data.
+ *
+ * For __rmemcpy and memmove an exception is always a kernel bug, therefore
+ * they're not protected.
+ */
+
+#define EXC(inst_reg,addr,handler)		\
+9:	inst_reg, addr;				\
+	.section __ex_table,"a";		\
+	PTR	9b, handler;			\
+	.previous
+
+/*
+ * Only on the 64-bit kernel we can made use of 64-bit registers.
+ */
+#ifdef CONFIG_MIPS64
+#define USE_DOUBLE
+#endif
+
+#ifdef USE_DOUBLE
+
+#define LOAD   ld
+#define LOADL  ldl
+#define LOADR  ldr
+#define STOREL sdl
+#define STORER sdr
+#define STORE  sd
+#define ADD    daddu
+#define SUB    dsubu
+#define SRL    dsrl
+#define SRA    dsra
+#define SLL    dsll
+#define SLLV   dsllv
+#define SRLV   dsrlv
+#define NBYTES 8
+#define LOG_NBYTES 3
+
+/* 
+ * As we are sharing code base with the mips32 tree (which use the o32 ABI
+ * register definitions). We need to redefine the register definitions from
+ * the n64 ABI register naming to the o32 ABI register naming.
+ */
+#undef t0
+#undef t1
+#undef t2
+#undef t3
+#define t0	$8
+#define t1	$9
+#define t2	$10
+#define t3	$11
+#define t4	$12
+#define t5	$13
+#define t6	$14
+#define t7	$15
+	
+#else
+
+#define LOAD   lw
+#define LOADL  lwl
+#define LOADR  lwr
+#define STOREL swl
+#define STORER swr
+#define STORE  sw
+#define ADD    addu
+#define SUB    subu
+#define SRL    srl
+#define SLL    sll
+#define SRA    sra
+#define SLLV   sllv
+#define SRLV   srlv
+#define NBYTES 4
+#define LOG_NBYTES 2
+
+#endif /* USE_DOUBLE */
+
+#ifdef CONFIG_CPU_LITTLE_ENDIAN
+#define LDFIRST LOADR
+#define LDREST  LOADL
+#define STFIRST STORER
+#define STREST  STOREL
+#define SHIFT_DISCARD SLLV
+#else
+#define LDFIRST LOADL
+#define LDREST  LOADR
+#define STFIRST STOREL
+#define STREST  STORER
+#define SHIFT_DISCARD SRLV
+#endif
+
+#define FIRST(unit) ((unit)*NBYTES)
+#define REST(unit)  (FIRST(unit)+NBYTES-1)
+#define UNIT(unit)  FIRST(unit)
+
+#define ADDRMASK (NBYTES-1)
+
+	.text
+	.set	noreorder
+	.set	noat
+
+/*
+ * A combined memcpy/__copy_user
+ * __copy_user sets len to 0 for success; else to an upper bound of
+ * the number of uncopied bytes.
+ * memcpy sets v0 to dst.
+ */
+	.align	5
+LEAF(memcpy)					/* a0=dst a1=src a2=len */
+	move	v0, dst				/* return value */
+__memcpy:
+FEXPORT(__copy_user)
+	/*
+	 * Note: dst & src may be unaligned, len may be 0
+	 * Temps
+	 */
+#define rem t8
+
+	/*
+	 * The "issue break"s below are very approximate.
+	 * Issue delays for dcache fills will perturb the schedule, as will
+	 * load queue full replay traps, etc.
+	 *
+	 * If len < NBYTES use byte operations.
+	 */
+	PREF(	0, 0(src) )
+	PREF(	1, 0(dst) )
+	sltu	t2, len, NBYTES
+	and	t1, dst, ADDRMASK
+	PREF(	0, 1*32(src) )
+	PREF(	1, 1*32(dst) )
+	bnez	t2, copy_bytes_checklen
+	 and	t0, src, ADDRMASK
+	PREF(	0, 2*32(src) )
+	PREF(	1, 2*32(dst) )
+	bnez	t1, dst_unaligned
+	 nop
+	bnez	t0, src_unaligned_dst_aligned
+	/*
+	 * use delay slot for fall-through
+	 * src and dst are aligned; need to compute rem
+	 */
+both_aligned:
+	 SRL	t0, len, LOG_NBYTES+3    # +3 for 8 units/iter
+	beqz	t0, cleanup_both_aligned # len < 8*NBYTES
+	 and	rem, len, (8*NBYTES-1)	 # rem = len % (8*NBYTES)
+	PREF(	0, 3*32(src) )
+	PREF(	1, 3*32(dst) )
+	.align	4
+1:
+EXC(	LOAD	t0, UNIT(0)(src),	l_exc)
+EXC(	LOAD	t1, UNIT(1)(src),	l_exc_copy)
+EXC(	LOAD	t2, UNIT(2)(src),	l_exc_copy)
+EXC(	LOAD	t3, UNIT(3)(src),	l_exc_copy)
+	SUB	len, len, 8*NBYTES
+EXC(	LOAD	t4, UNIT(4)(src),	l_exc_copy)
+EXC(	LOAD	t7, UNIT(5)(src),	l_exc_copy)
+EXC(	STORE	t0, UNIT(0)(dst),	s_exc_p8u)
+EXC(	STORE	t1, UNIT(1)(dst),	s_exc_p7u)
+EXC(	LOAD	t0, UNIT(6)(src),	l_exc_copy)
+EXC(	LOAD	t1, UNIT(7)(src),	l_exc_copy)
+	ADD	src, src, 8*NBYTES
+	ADD	dst, dst, 8*NBYTES
+EXC(	STORE	t2, UNIT(-6)(dst),	s_exc_p6u)
+EXC(	STORE	t3, UNIT(-5)(dst),	s_exc_p5u)
+EXC(	STORE	t4, UNIT(-4)(dst),	s_exc_p4u)
+EXC(	STORE	t7, UNIT(-3)(dst),	s_exc_p3u)
+EXC(	STORE	t0, UNIT(-2)(dst),	s_exc_p2u)
+EXC(	STORE	t1, UNIT(-1)(dst),	s_exc_p1u)
+	PREF(	0, 8*32(src) )
+	PREF(	1, 8*32(dst) )
+	bne	len, rem, 1b
+	 nop
+
+	/*
+	 * len == rem == the number of bytes left to copy < 8*NBYTES
+	 */
+cleanup_both_aligned:
+	beqz	len, done
+	 sltu	t0, len, 4*NBYTES
+	bnez	t0, less_than_4units
+	 and	rem, len, (NBYTES-1)	# rem = len % NBYTES
+	/*
+	 * len >= 4*NBYTES
+	 */
+EXC(	LOAD	t0, UNIT(0)(src),	l_exc)
+EXC(	LOAD	t1, UNIT(1)(src),	l_exc_copy)
+EXC(	LOAD	t2, UNIT(2)(src),	l_exc_copy)
+EXC(	LOAD	t3, UNIT(3)(src),	l_exc_copy)
+	SUB	len, len, 4*NBYTES
+	ADD	src, src, 4*NBYTES
+EXC(	STORE	t0, UNIT(0)(dst),	s_exc_p4u)
+EXC(	STORE	t1, UNIT(1)(dst),	s_exc_p3u)
+EXC(	STORE	t2, UNIT(2)(dst),	s_exc_p2u)
+EXC(	STORE	t3, UNIT(3)(dst),	s_exc_p1u)
+	beqz	len, done
+	 ADD	dst, dst, 4*NBYTES
+less_than_4units:
+	/*
+	 * rem = len % NBYTES
+	 */
+	beq	rem, len, copy_bytes
+	 nop
+1:
+EXC(	LOAD	t0, 0(src),		l_exc)
+	ADD	src, src, NBYTES
+	SUB	len, len, NBYTES
+EXC(	STORE	t0, 0(dst),		s_exc_p1u)
+	bne	rem, len, 1b
+	 ADD	dst, dst, NBYTES
+
+	/*
+	 * src and dst are aligned, need to copy rem bytes (rem < NBYTES)
+	 * A loop would do only a byte at a time with possible branch
+	 * mispredicts.  Can't do an explicit LOAD dst,mask,or,STORE
+	 * because can't assume read-access to dst.  Instead, use
+	 * STREST dst, which doesn't require read access to dst.
+	 *
+	 * This code should perform better than a simple loop on modern,
+	 * wide-issue mips processors because the code has fewer branches and
+	 * more instruction-level parallelism.
+	 */
+#define bits t2
+	beqz	len, done
+	 ADD	t1, dst, len	# t1 is just past last byte of dst
+	li	bits, 8*NBYTES
+	SLL	rem, len, 3	# rem = number of bits to keep
+EXC(	LOAD	t0, 0(src),		l_exc)
+	SUB	bits, bits, rem	# bits = number of bits to discard
+	SHIFT_DISCARD t0, t0, bits
+EXC(	STREST	t0, -1(t1),		s_exc)
+	jr	ra
+	 move	len, zero
+dst_unaligned:
+	/*
+	 * dst is unaligned
+	 * t0 = src & ADDRMASK
+	 * t1 = dst & ADDRMASK; T1 > 0
+	 * len >= NBYTES
+	 *
+	 * Copy enough bytes to align dst
+	 * Set match = (src and dst have same alignment)
+	 */
+#define match rem
+EXC(	LDFIRST	t3, FIRST(0)(src),	l_exc)
+	ADD	t2, zero, NBYTES
+EXC(	LDREST	t3, REST(0)(src),	l_exc_copy)
+	SUB	t2, t2, t1	# t2 = number of bytes copied
+	xor	match, t0, t1
+EXC(	STFIRST t3, FIRST(0)(dst),	s_exc)
+	beq	len, t2, done
+	 SUB	len, len, t2
+	ADD	dst, dst, t2
+	beqz	match, both_aligned
+	 ADD	src, src, t2
+
+src_unaligned_dst_aligned:
+	SRL	t0, len, LOG_NBYTES+2    # +2 for 4 units/iter
+	PREF(	0, 3*32(src) )
+	beqz	t0, cleanup_src_unaligned
+	 and	rem, len, (4*NBYTES-1)   # rem = len % 4*NBYTES
+	PREF(	1, 3*32(dst) )
+1:
+/*
+ * Avoid consecutive LD*'s to the same register since some mips
+ * implementations can't issue them in the same cycle.
+ * It's OK to load FIRST(N+1) before REST(N) because the two addresses
+ * are to the same unit (unless src is aligned, but it's not).
+ */
+EXC(	LDFIRST	t0, FIRST(0)(src),	l_exc)
+EXC(	LDFIRST	t1, FIRST(1)(src),	l_exc_copy)
+	SUB     len, len, 4*NBYTES
+EXC(	LDREST	t0, REST(0)(src),	l_exc_copy)
+EXC(	LDREST	t1, REST(1)(src),	l_exc_copy)
+EXC(	LDFIRST	t2, FIRST(2)(src),	l_exc_copy)
+EXC(	LDFIRST	t3, FIRST(3)(src),	l_exc_copy)
+EXC(	LDREST	t2, REST(2)(src),	l_exc_copy)
+EXC(	LDREST	t3, REST(3)(src),	l_exc_copy)
+	PREF(	0, 9*32(src) )		# 0 is PREF_LOAD  (not streamed)
+	ADD	src, src, 4*NBYTES
+#ifdef CONFIG_CPU_SB1
+	nop				# improves slotting
+#endif
+EXC(	STORE	t0, UNIT(0)(dst),	s_exc_p4u)
+EXC(	STORE	t1, UNIT(1)(dst),	s_exc_p3u)
+EXC(	STORE	t2, UNIT(2)(dst),	s_exc_p2u)
+EXC(	STORE	t3, UNIT(3)(dst),	s_exc_p1u)
+	PREF(	1, 9*32(dst) )     	# 1 is PREF_STORE (not streamed)
+	bne	len, rem, 1b
+	 ADD	dst, dst, 4*NBYTES
+
+cleanup_src_unaligned:
+	beqz	len, done
+	 and	rem, len, NBYTES-1  # rem = len % NBYTES
+	beq	rem, len, copy_bytes
+	 nop
+1:
+EXC(	LDFIRST t0, FIRST(0)(src),	l_exc)
+EXC(	LDREST	t0, REST(0)(src),	l_exc_copy)
+	ADD	src, src, NBYTES
+	SUB	len, len, NBYTES
+EXC(	STORE	t0, 0(dst),		s_exc_p1u)
+	bne	len, rem, 1b
+	 ADD	dst, dst, NBYTES
+
+copy_bytes_checklen:
+	beqz	len, done
+	 nop
+copy_bytes:
+	/* 0 < len < NBYTES  */
+#define COPY_BYTE(N)			\
+EXC(	lb	t0, N(src), l_exc);	\
+	SUB	len, len, 1;		\
+	beqz	len, done;		\
+EXC(	 sb	t0, N(dst), s_exc_p1)
+
+	COPY_BYTE(0)
+	COPY_BYTE(1)
+#ifdef USE_DOUBLE
+	COPY_BYTE(2)
+	COPY_BYTE(3)
+	COPY_BYTE(4)
+	COPY_BYTE(5)
+#endif
+EXC(	lb	t0, NBYTES-2(src), l_exc)
+	SUB	len, len, 1
+	jr	ra
+EXC(	 sb	t0, NBYTES-2(dst), s_exc_p1)
+done:
+	jr	ra
+	 nop
+	END(memcpy)
+
+l_exc_copy:
+	/*
+	 * Copy bytes from src until faulting load address (or until a
+	 * lb faults)
+	 *
+	 * When reached by a faulting LDFIRST/LDREST, THREAD_BUADDR($28)
+	 * may be more than a byte beyond the last address.
+	 * Hence, the lb below may get an exception.
+	 *
+	 * Assumes src < THREAD_BUADDR($28)
+	 */
+	LOAD	t0, TI_TASK($28)
+	 nop
+	LOAD	t0, THREAD_BUADDR(t0)
+1:
+EXC(	lb	t1, 0(src),	l_exc)
+	ADD	src, src, 1
+	sb	t1, 0(dst)	# can't fault -- we're copy_from_user
+	bne	src, t0, 1b
+	 ADD	dst, dst, 1
+l_exc:
+	LOAD	t0, TI_TASK($28)
+	 nop
+	LOAD	t0, THREAD_BUADDR(t0)	# t0 is just past last good address
+	 nop
+	SUB	len, AT, t0		# len number of uncopied bytes
+	/*
+	 * Here's where we rely on src and dst being incremented in tandem,
+	 *   See (3) above.
+	 * dst += (fault addr - src) to put dst at first byte to clear
+	 */
+	ADD	dst, t0			# compute start address in a1
+	SUB	dst, src
+	/*
+	 * Clear len bytes starting at dst.  Can't call __bzero because it
+	 * might modify len.  An inefficient loop for these rare times...
+	 */
+	beqz	len, done
+	 SUB	src, len, 1
+1:	sb	zero, 0(dst)
+	ADD	dst, dst, 1
+	bnez	src, 1b
+	 SUB	src, src, 1
+	jr	ra
+	 nop
+
+
+#define SEXC(n)				\
+s_exc_p ## n ## u:			\
+	jr	ra;			\
+	 ADD	len, len, n*NBYTES
+
+SEXC(8)
+SEXC(7)
+SEXC(6)
+SEXC(5)
+SEXC(4)
+SEXC(3)
+SEXC(2)
+SEXC(1)
+
+s_exc_p1:
+	jr	ra
+	 ADD	len, len, 1
+s_exc:
+	jr	ra
+	 nop
+
+	.align	5
+LEAF(memmove)
+	ADD	t0, a0, a2
+	ADD	t1, a1, a2
+	sltu	t0, a1, t0			# dst + len <= src -> memcpy
+	sltu	t1, a0, t1			# dst >= src + len -> memcpy
+	and	t0, t1
+	beqz	t0, __memcpy
+	 move	v0, a0				/* return value */
+	beqz	a2, r_out
+	END(memmove)
+
+	/* fall through to __rmemcpy */
+LEAF(__rmemcpy)					/* a0=dst a1=src a2=len */
+	 sltu	t0, a1, a0
+	beqz	t0, r_end_bytes_up		# src >= dst
+	 nop
+	ADD	a0, a2				# dst = dst + len
+	ADD	a1, a2				# src = src + len
+
+r_end_bytes:
+	lb	t0, -1(a1)
+	SUB	a2, a2, 0x1
+	sb	t0, -1(a0)
+	SUB	a1, a1, 0x1
+	bnez	a2, r_end_bytes
+	 SUB	a0, a0, 0x1
+
+r_out:
+	jr	ra
+	 move	a2, zero
+
+r_end_bytes_up:
+	lb	t0, (a1)
+	SUB	a2, a2, 0x1
+	sb	t0, (a0)
+	ADD	a1, a1, 0x1
+	bnez	a2, r_end_bytes_up
+	 ADD	a0, a0, 0x1
+
+	jr	ra
+	 move	a2, zero
+	END(__rmemcpy)
diff --git a/arch/mips/lib/promlib.c b/arch/mips/lib/promlib.c
new file mode 100644
index 000000000000..dddfe98b4ded
--- /dev/null
+++ b/arch/mips/lib/promlib.c
@@ -0,0 +1,24 @@
+#include <stdarg.h>
+#include <linux/kernel.h>
+
+extern void prom_putchar(char);
+
+void prom_printf(char *fmt, ...)
+{
+	va_list args;
+	char ppbuf[1024];
+	char *bptr;
+
+	va_start(args, fmt);
+	vsprintf(ppbuf, fmt, args);
+
+	bptr = ppbuf;
+
+	while (*bptr != 0) {
+		if (*bptr == '\n')
+			prom_putchar('\r');
+
+		prom_putchar(*bptr++);
+	}
+	va_end(args);
+}
diff --git a/arch/mips/lib/strlen_user.S b/arch/mips/lib/strlen_user.S
new file mode 100644
index 000000000000..07660e86c99d
--- /dev/null
+++ b/arch/mips/lib/strlen_user.S
@@ -0,0 +1,39 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (c) 1996, 1998, 1999, 2004 by Ralf Baechle
+ * Copyright (c) 1999 Silicon Graphics, Inc.
+ */
+#include <asm/asm.h>
+#include <asm/offset.h>
+#include <asm/regdef.h>
+
+#define EX(insn,reg,addr,handler)			\
+9:	insn	reg, addr;				\
+	.section __ex_table,"a";			\
+	PTR	9b, handler;				\
+	.previous
+
+/*
+ * Return the size of a string (including the ending 0)
+ *
+ * Return 0 for error
+ */
+LEAF(__strlen_user_asm)
+	LONG_L		v0, TI_ADDR_LIMIT($28)	# pointer ok?
+	and		v0, a0
+	bnez		v0, fault
+
+FEXPORT(__strlen_user_nocheck_asm)
+	move		v0, a0
+1:	EX(lb, t0, (v0), fault)
+	PTR_ADDIU	v0, 1
+	bnez		t0, 1b
+	PTR_SUBU	v0, a0
+	jr		ra
+	END(__strlen_user_asm)
+
+fault:	move		v0, zero
+	jr		ra
diff --git a/arch/mips/lib/strncpy_user.S b/arch/mips/lib/strncpy_user.S
new file mode 100644
index 000000000000..14bed17c1648
--- /dev/null
+++ b/arch/mips/lib/strncpy_user.S
@@ -0,0 +1,58 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (c) 1996, 1999 by Ralf Baechle
+ */
+#include <linux/errno.h>
+#include <asm/asm.h>
+#include <asm/offset.h>
+#include <asm/regdef.h>
+
+#define EX(insn,reg,addr,handler)			\
+9:	insn	reg, addr;				\
+	.section __ex_table,"a";			\
+	PTR	9b, handler;				\
+	.previous
+
+/*
+ * Returns: -EFAULT if exception before terminator, N if the entire
+ * buffer filled, else strlen.
+ */
+
+/*
+ * Ugly special case have to check: we might get passed a user space
+ * pointer which wraps into the kernel space.  We don't deal with that.  If
+ * it happens at most some bytes of the exceptions handlers will be copied.
+ */
+
+LEAF(__strncpy_from_user_asm)
+	LONG_L		v0, TI_ADDR_LIMIT($28)	# pointer ok?
+	and		v0, a1
+	bnez		v0, fault
+
+FEXPORT(__strncpy_from_user_nocheck_asm)
+	move		v0, zero
+	move		v1, a1
+	.set		noreorder
+1:	EX(lbu, t0, (v1), fault)
+	PTR_ADDIU	v1, 1
+	beqz		t0, 2f
+	 sb		t0, (a0)
+	PTR_ADDIU	v0, 1
+	bne		v0, a2, 1b
+	 PTR_ADDIU	a0, 1
+	.set		reorder
+2:	PTR_ADDU	t0, a1, v0
+	xor		t0, a1
+	bltz		t0, fault
+	jr		ra			# return n
+	END(__strncpy_from_user_asm)
+
+fault:	li		v0, -EFAULT
+	jr		ra
+
+	.section	__ex_table,"a"
+	PTR		1b, fault
+	.previous
diff --git a/arch/mips/lib/strnlen_user.S b/arch/mips/lib/strnlen_user.S
new file mode 100644
index 000000000000..6e7a8eed4de8
--- /dev/null
+++ b/arch/mips/lib/strnlen_user.S
@@ -0,0 +1,45 @@
+/*
+ * This file is subject to the terms and conditions of the GNU General Public
+ * License.  See the file "COPYING" in the main directory of this archive
+ * for more details.
+ *
+ * Copyright (c) 1996, 1998, 1999, 2004 by Ralf Baechle
+ * Copyright (c) 1999 Silicon Graphics, Inc.
+ */
+#include <asm/asm.h>
+#include <asm/offset.h>
+#include <asm/regdef.h>
+
+#define EX(insn,reg,addr,handler)			\
+9:	insn	reg, addr;				\
+	.section __ex_table,"a";			\
+	PTR	9b, handler;				\
+	.previous
+
+/*
+ * Return the size of a string including the ending NUL character upto a
+ * maximum of a1 or 0 in case of error.
+ *
+ * Note: for performance reasons we deliberately accept that a user may
+ *       make strlen_user and strnlen_user access the first few KSEG0
+ *       bytes.  There's nothing secret there.  On 64-bit accessing beyond
+ *       the maximum is a tad hairier ...
+ */
+LEAF(__strnlen_user_asm)
+	LONG_L		v0, TI_ADDR_LIMIT($28)	# pointer ok?
+	and		v0, a0
+	bnez		v0, fault
+
+FEXPORT(__strnlen_user_nocheck_asm)
+	move		v0, a0
+	PTR_ADDU	a1, a0			# stop pointer
+1:	beq		v0, a1, 1f		# limit reached?
+	EX(lb, t0, (v0), fault)
+	PTR_ADDU	v0, 1
+	bnez		t0, 1b
+1:	PTR_SUBU	v0, a0
+	jr		ra
+	END(__strnlen_user_asm)
+
+fault:	move		v0, zero
+	jr		ra