Bitops:Cleanup ARM bitops

Cleanup of ARM bitops functions.
Introduce the findbits.S which allows for
optimized algo.

Signed-off-by: Nishanth Menon <x0nishan@ti.com>

3 files changed, 273 insertions, 42 deletions

diff --git a/arch/arm/lib/Makefile b/arch/arm/lib/Makefile
index 9656ca81d9..2fddcc67e9 100644
--- a/arch/arm/lib/Makefile
+++ b/arch/arm/lib/Makefile
@@ -8,6 +8,7 @@ obj-y	+= _modsi3.o
 obj-y	+= _udivsi3.o
 obj-y	+= _umodsi3.o
 obj-y	+= _lshrdi3.o
+obj-y	+= findbit.o
 obj-y	+= arm.o
 obj-$(CONFIG_MODULES) += module.o
 extra-$(CONFIG_GENERIC_LINKER_SCRIPT) += u-boot.lds
diff --git a/arch/arm/lib/findbit.S b/arch/arm/lib/findbit.S
new file mode 100644
index 0000000000..8023a7fcc4
--- /dev/null
+++ b/arch/arm/lib/findbit.S
@@ -0,0 +1,181 @@
+/**
+ * @file
+ * @brief common bitops
+ */
+/*
+ * Originally from Linux kernel
+ *  arch/arm/lib/findbit.S
+ *
+ *  Copyright (C) 1995-2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * 16th March 2001 - John Ripley <jripley@sonicblue.com>
+ *   Fixed so that "size" is an exclusive not an inclusive quantity.
+ *   All users of these functions expect exclusive sizes, and may
+ *   also call with zero size.
+ * Reworked by rmk.
+ */
+                .text
+
+/*
+ * Purpose  : Find a 'zero' bit
+ * Prototype: int find_first_zero_bit(void *addr, unsigned int maxbit);
+ */
+	.globl _find_first_zero_bit_le;
+_find_first_zero_bit_le:
+		teq	r1, #0
+		beq	3f
+		mov	r2, #0
+1:		ldrb	r3, [r0, r2, lsr #3]
+		eors	r3, r3, #0xff		@ invert bits
+		bne	.L_found		@ any now set - found zero bit
+		add	r2, r2, #8		@ next bit pointer
+2:		cmp	r2, r1			@ any more?
+		blo	1b
+3:		mov	r0, r1			@ no free bits
+		mov	pc, lr
+
+/*
+ * Purpose  : Find next 'zero' bit
+ * Prototype: int find_next_zero_bit(void *addr, unsigned int maxbit,
+ *					int offset)
+ */
+	.globl _find_next_zero_bit_le;
+_find_next_zero_bit_le:
+		teq	r1, #0
+		beq	3b
+		ands	ip, r2, #7
+		beq	1b			@ If new byte, goto old routine
+		ldrb	r3, [r0, r2, lsr #3]
+		eor	r3, r3, #0xff		@ now looking for a 1 bit
+		movs	r3, r3, lsr ip		@ shift off unused bits
+		bne	.L_found
+		orr	r2, r2, #7		@ if zero, then no bits here
+		add	r2, r2, #1		@ align bit pointer
+		b	2b			@ loop for next bit
+
+/*
+ * Purpose  : Find a 'one' bit
+ * Prototype: int find_first_bit(const unsigned long *addr, unsigned int maxbit)
+ */
+	.globl _find_first_bit_le;
+_find_first_bit_le:
+		teq	r1, #0
+		beq	3f
+		mov	r2, #0
+1:		ldrb	r3, [r0, r2, lsr #3]
+		movs	r3, r3
+		bne	.L_found		@ any now set - found zero bit
+		add	r2, r2, #8		@ next bit pointer
+2:		cmp	r2, r1			@ any more?
+		blo	1b
+3:		mov	r0, r1			@ no free bits
+		mov	pc, lr
+
+/*
+ * Purpose  : Find next 'one' bit
+ * Prototype: int find_next_zero_bit(void *addr, unsigned int maxbit,
+ * 					int offset)
+ */
+	.globl _find_next_bit_le;
+_find_next_bit_le:
+		teq	r1, #0
+		beq	3b
+		ands	ip, r2, #7
+		beq	1b			@ If new byte, goto old routine
+		ldrb	r3, [r0, r2, lsr #3]
+		movs	r3, r3, lsr ip		@ shift off unused bits
+		bne	.L_found
+		orr	r2, r2, #7		@ if zero, then no bits here
+		add	r2, r2, #1		@ align bit pointer
+		b	2b			@ loop for next bit
+
+#ifdef __ARMEB__
+
+	.globl _find_first_zero_bit_be;
+_find_first_zero_bit_be:
+		teq	r1, #0
+		beq	3f
+		mov	r2, #0
+1:		eor	r3, r2, #0x18		@ big endian byte ordering
+		ldrb	r3, [r0, r3, lsr #3]
+		eors	r3, r3, #0xff		@ invert bits
+		bne	.L_found		@ any now set - found zero bit
+		add	r2, r2, #8		@ next bit pointer
+2:		cmp	r2, r1			@ any more?
+		blo	1b
+3:		mov	r0, r1			@ no free bits
+		mov	pc, lr
+
+	.globl _find_next_zero_bit_be;
+_find_next_zero_bit_be:
+		teq	r1, #0
+		beq	3b
+		ands	ip, r2, #7
+		beq	1b			@ If new byte, goto old routine
+		eor	r3, r2, #0x18		@ big endian byte ordering
+		ldrb	r3, [r0, r3, lsr #3]
+		eor	r3, r3, #0xff		@ now looking for a 1 bit
+		movs	r3, r3, lsr ip		@ shift off unused bits
+		bne	.L_found
+		orr	r2, r2, #7		@ if zero, then no bits here
+		add	r2, r2, #1		@ align bit pointer
+		b	2b			@ loop for next bit
+
+	.globl _find_first_bit_be;
+_find_first_bit_be:
+		teq	r1, #0
+		beq	3f
+		mov	r2, #0
+1:		eor	r3, r2, #0x18		@ big endian byte ordering
+		ldrb	r3, [r0, r3, lsr #3]
+		movs	r3, r3
+		bne	.L_found		@ any now set - found zero bit
+		add	r2, r2, #8		@ next bit pointer
+2:		cmp	r2, r1			@ any more?
+		blo	1b
+3:		mov	r0, r1			@ no free bits
+		mov	pc, lr
+
+	.globl _find_next_bit_be;
+_find_next_bit_be:
+		teq	r1, #0
+		beq	3b
+		ands	ip, r2, #7
+		beq	1b			@ If new byte, goto old routine
+		eor	r3, r2, #0x18		@ big endian byte ordering
+		ldrb	r3, [r0, r3, lsr #3]
+		movs	r3, r3, lsr ip		@ shift off unused bits
+		bne	.L_found
+		orr	r2, r2, #7		@ if zero, then no bits here
+		add	r2, r2, #1		@ align bit pointer
+		b	2b			@ loop for next bit
+
+#endif
+
+/*
+ * One or more bits in the LSB of r3 are assumed to be set.
+ */
+.L_found:
+#if __LINUX_ARM_ARCH__ >= 5
+		rsb	r1, r3, #0
+		and	r3, r3, r1
+		clz	r3, r3
+		rsb	r3, r3, #31
+		add	r0, r2, r3
+#else
+		tst	r3, #0x0f
+		addeq	r2, r2, #4
+		movne	r3, r3, lsl #4
+		tst	r3, #0x30
+		addeq	r2, r2, #2
+		movne	r3, r3, lsl #2
+		tst	r3, #0x40
+		addeq	r2, r2, #1
+		mov	r0, r2
+#endif
+		mov	pc, lr
+
diff --git a/include/asm-arm/bitops.h b/include/asm-arm/bitops.h
index ea52add55b..b1d34f8633 100644
--- a/include/asm-arm/bitops.h
+++ b/include/asm-arm/bitops.h
@@ -15,32 +15,28 @@
 #ifndef __ASM_ARM_BITOPS_H
 #define __ASM_ARM_BITOPS_H
 
+#ifndef _LINUX_BITOPS_H
+#error only <linux/bitops.h> can be included directly
+#endif
+
 /*
- * Function prototypes to keep gcc -Wall happy.
+ * Functions equivalent of ops.h
  */
-extern void set_bit(int nr, volatile void * addr);
-
 static inline void __set_bit(int nr, volatile void *addr)
 {
 	((unsigned char *) addr)[nr >> 3] |= (1U << (nr & 7));
 }
 
-extern void clear_bit(int nr, volatile void * addr);
-
 static inline void __clear_bit(int nr, volatile void *addr)
 {
 	((unsigned char *) addr)[nr >> 3] &= ~(1U << (nr & 7));
 }
 
-extern void change_bit(int nr, volatile void * addr);
-
 static inline void __change_bit(int nr, volatile void *addr)
 {
 	((unsigned char *) addr)[nr >> 3] ^= (1U << (nr & 7));
 }
 
-extern int test_and_set_bit(int nr, volatile void * addr);
-
 static inline int __test_and_set_bit(int nr, volatile void *addr)
 {
 	unsigned int mask = 1 << (nr & 7);
@@ -51,8 +47,6 @@ static inline int __test_and_set_bit(int nr, volatile void *addr)
 	return oldval & mask;
 }
 
-extern int test_and_clear_bit(int nr, volatile void * addr);
-
 static inline int __test_and_clear_bit(int nr, volatile void *addr)
 {
 	unsigned int mask = 1 << (nr & 7);
@@ -63,8 +57,6 @@ static inline int __test_and_clear_bit(int nr, volatile void *addr)
 	return oldval & mask;
 }
 
-extern int test_and_change_bit(int nr, volatile void * addr);
-
 static inline int __test_and_change_bit(int nr, volatile void *addr)
 {
 	unsigned int mask = 1 << (nr & 7);
@@ -75,9 +67,6 @@ static inline int __test_and_change_bit(int nr, volatile void *addr)
 	return oldval & mask;
 }
 
-extern int find_first_zero_bit(void * addr, unsigned size);
-extern int find_next_zero_bit(void * addr, int size, int offset);
-
 /*
  * This routine doesn't need to be atomic.
  */
@@ -86,6 +75,48 @@ static inline int test_bit(int nr, const void * addr)
     return ((unsigned char *) addr)[nr >> 3] & (1U << (nr & 7));
 }
 
+#define set_bit(x, y)			__set_bit(x, y)
+#define clear_bit(x, y)			__clear_bit(x, y)
+#define change_bit(x, y)		__change_bit(x, y)
+#define test_and_set_bit(x, y)		__test_and_set_bit(x, y)
+#define test_and_clear_bit(x, y)	__test_and_clear_bit(x, y)
+#define test_and_change_bit(x, y)	__test_and_change_bit(x, y)
+
+#ifndef __ARMEB__
+/*
+ * These are the little endian definitions.
+ */
+extern int _find_first_zero_bit_le(const void *p, unsigned size);
+extern int _find_next_zero_bit_le(const void *p, int size, int offset);
+extern int _find_first_bit_le(const unsigned long *p, unsigned size);
+extern int _find_next_bit_le(const unsigned long *p, int size, int offset);
+#define find_first_zero_bit(p, sz)	_find_first_zero_bit_le(p, sz)
+#define find_next_zero_bit(p, sz, off)	_find_next_zero_bit_le(p, sz, off)
+#define find_first_bit(p, sz)		_find_first_bit_le(p, sz)
+#define find_next_bit(p, sz, off)	_find_next_bit_le(p, sz, off)
+
+#define WORD_BITOFF_TO_LE(x)		((x))
+
+#else		/* ! __ARMEB__ */
+
+/*
+ * These are the big endian definitions.
+ */
+extern int _find_first_zero_bit_be(const void *p, unsigned size);
+extern int _find_next_zero_bit_be(const void *p, int size, int offset);
+extern int _find_first_bit_be(const unsigned long *p, unsigned size);
+extern int _find_next_bit_be(const unsigned long *p, int size, int offset);
+#define find_first_zero_bit(p, sz)	_find_first_zero_bit_be(p, sz)
+#define find_next_zero_bit(p, sz, off)	_find_next_zero_bit_be(p, sz, off)
+#define find_first_bit(p, sz)		_find_first_bit_be(p, sz)
+#define find_next_bit(p, sz, off)	_find_next_bit_be(p, sz, off)
+
+#define WORD_BITOFF_TO_LE(x)		((x) ^ 0x18)
+
+#endif		/* __ARMEB__ */
+
+#if defined(__LINUX_ARM_ARCH__) && (__LINUX_ARM_ARCH__ < 5)
+
 /*
  * ffz = Find First Zero in word. Undefined if no zero exists,
  * so code should check against ~0UL first..
@@ -103,35 +134,53 @@ static inline unsigned long ffz(unsigned long word)
 	if (word & 0x40000000) { k -= 1; }
 	return k;
 }
+#include <asm-generic/bitops/__ffs.h>
+#include <asm-generic/bitops/ffs.h>
+#include <asm-generic/bitops/fls.h>
+#else		/* ! __ARM__USE_GENERIC_FF */
 
-/*
- * ffs: find first bit set. This is defined the same way as
- * the libc and compiler builtin ffs routines, therefore
- * differs in spirit from the above ffz (man ffs).
- */
-
-#define ffs(x) generic_ffs(x)
+static inline int constant_fls(int x)
+{
+	int r = 32;
+
+	if (!x)
+		return 0;
+	if (!(x & 0xffff0000u)) {
+		x <<= 16;
+		r -= 16;
+	}
+	if (!(x & 0xff000000u)) {
+		x <<= 8;
+		r -= 8;
+	}
+	if (!(x & 0xf0000000u)) {
+		x <<= 4;
+		r -= 4;
+	}
+	if (!(x & 0xc0000000u)) {
+		x <<= 2;
+		r -= 2;
+	}
+	if (!(x & 0x80000000u)) {
+		x <<= 1;
+		r -= 1;
+	}
+	return r;
+}
 
 /*
- * hweightN: returns the hamming weight (i.e. the number
- * of bits set) of a N-bit word
+ * On ARMv5 and above those functions can be implemented around
+ * the clz instruction for much better code efficiency.
  */
-
-#define hweight32(x) generic_hweight32(x)
-#define hweight16(x) generic_hweight16(x)
-#define hweight8(x) generic_hweight8(x)
-
-#define ext2_set_bit			test_and_set_bit
-#define ext2_clear_bit			test_and_clear_bit
-#define ext2_test_bit			test_bit
-#define ext2_find_first_zero_bit	find_first_zero_bit
-#define ext2_find_next_zero_bit		find_next_zero_bit
-
-/* Bitmap functions for the minix filesystem. */
-#define minix_test_and_set_bit(nr,addr)	test_and_set_bit(nr,addr)
-#define minix_set_bit(nr,addr)		set_bit(nr,addr)
-#define minix_test_and_clear_bit(nr,addr)	test_and_clear_bit(nr,addr)
-#define minix_test_bit(nr,addr)		test_bit(nr,addr)
-#define minix_find_first_zero_bit(addr,size)	find_first_zero_bit(addr,size)
+#define fls(x) \
+	(__builtin_constant_p(x) ? constant_fls(x) : \
+	({ int __r; asm("clz\t%0, %1" : "=r"(__r) : "r"(x) : "cc"); 32-__r; }))
+#define ffs(x) ({ unsigned long __t = (x); fls(__t &-__t); })
+#define __ffs(x) (ffs(x) - 1)
+#define ffz(x) __ffs(~(x))
+#endif		/* __ARM__USE_GENERIC_FF */
+#include <asm-generic/bitops/fls64.h>
+
+#include <asm-generic/bitops/hweight.h>
 
 #endif /* _ARM_BITOPS_H */