[PARISC] Update bitops from parisc tree

Optimize ext2_find_next_zero_bit. Gives about 25% perf improvement with a
rsync test with ext3.

Signed-off-by: Randolph Chung <tausq@parisc-linux.org>

fix ext3 performance - ext2_find_next_zero() was culprit.
Kudos to jejb for pointing out the the possibility that ext2_test_bit
and ext2_find_next_zero() may in fact not be enumerating bits in
the bitmap because of endianess. Took sparc64 implementation and
adapted it to our tree. I suspect the real problem is ffz() wants
an unsigned long and was getting garbage in the top half of the
unsigned int. Not confirmed but that's what I suspect.

Signed-off-by: Grant Grundler <grundler@parisc-linux.org>

Fix find_next_bit for 32-bit
Make masking consistent for bitops

From: Joel Soete <soete.joel@tiscali.be>
Signed-off-by: Randolph Chung <tausq@parisc-linux.org>

Add back incorrectly removed ext2_find_first_zero_bit definition

Signed-off-by: James Bottomley <jejb@parisc-linux.org>

Fixup bitops.h to use volatile for *_bit() ops

Based on this email thread:
       http://marc.theaimsgroup.com/?t=108826637900003

In a nutshell:
        *_bit() want use of volatile.
        __*_bit() are "relaxed" and don't use spinlock or volatile.

other minor changes:
o replaces hweight64() macro with alias to generic_hweight64() (Joel Soete)
o cleanup ext2* macros so (a) it's obvious what the XOR magic is about
  and (b) one version that works for both 32/64-bit.
o replace 2 uses of CONFIG_64BIT with __LP64__. bitops.h used both.
  I think header files that might go to user space should use
  something userspace will know about (__LP64__).

Signed-off-by: Grant Grundler <grundler@parisc-linux.org>

Move SHIFT_PER_LONG to standard location for BITS_PER_LONG (asm/types.h)
and ditch the second definition of BITS_PER_LONG in bitops.h

Signed-off-by: Grant Grundler <grundler@parisc-linux.org>

Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>

2 files changed, 134 insertions, 158 deletions

diff --git a/include/asm-parisc/bitops.h b/include/asm-parisc/bitops.h
index af7db69..55b98c6 100644
--- a/include/asm-parisc/bitops.h
+++ b/include/asm-parisc/bitops.h
@@ -2,7 +2,7 @@
 #define _PARISC_BITOPS_H
 
 #include <linux/compiler.h>
-#include <asm/spinlock.h>
+#include <asm/types.h>		/* for BITS_PER_LONG/SHIFT_PER_LONG */
 #include <asm/byteorder.h>
 #include <asm/atomic.h>
 
@@ -12,193 +12,157 @@
  * to include/asm-i386/bitops.h or kerneldoc
  */
 
-#ifdef __LP64__
-#   define SHIFT_PER_LONG 6
-#ifndef BITS_PER_LONG
-#   define BITS_PER_LONG 64
-#endif
-#else
-#   define SHIFT_PER_LONG 5
-#ifndef BITS_PER_LONG
-#   define BITS_PER_LONG 32
-#endif
-#endif
-
-#define CHOP_SHIFTCOUNT(x) ((x) & (BITS_PER_LONG - 1))
+#define CHOP_SHIFTCOUNT(x) (((unsigned long) (x)) & (BITS_PER_LONG - 1))
 
 
 #define smp_mb__before_clear_bit()      smp_mb()
 #define smp_mb__after_clear_bit()       smp_mb()
 
-static __inline__ void set_bit(int nr, volatile unsigned long * address)
+/* See http://marc.theaimsgroup.com/?t=108826637900003 for discussion
+ * on use of volatile and __*_bit() (set/clear/change):
+ *	*_bit() want use of volatile.
+ *	__*_bit() are "relaxed" and don't use spinlock or volatile.
+ */
+
+static __inline__ void set_bit(int nr, volatile unsigned long * addr)
 {
-	unsigned long mask;
-	unsigned long *addr = (unsigned long *) address;
+	unsigned long mask = 1UL << CHOP_SHIFTCOUNT(nr);
 	unsigned long flags;
 
 	addr += (nr >> SHIFT_PER_LONG);
-	mask = 1L << CHOP_SHIFTCOUNT(nr);
 	_atomic_spin_lock_irqsave(addr, flags);
 	*addr |= mask;
 	_atomic_spin_unlock_irqrestore(addr, flags);
 }
 
-static __inline__ void __set_bit(int nr, volatile unsigned long * address)
+static __inline__ void __set_bit(unsigned long nr, volatile unsigned long * addr)
 {
-	unsigned long mask;
-	unsigned long *addr = (unsigned long *) address;
+	unsigned long *m = (unsigned long *) addr + (nr >> SHIFT_PER_LONG);
 
-	addr += (nr >> SHIFT_PER_LONG);
-	mask = 1L << CHOP_SHIFTCOUNT(nr);
-	*addr |= mask;
+	*m |= 1UL << CHOP_SHIFTCOUNT(nr);
 }
 
-static __inline__ void clear_bit(int nr, volatile unsigned long * address)
+static __inline__ void clear_bit(int nr, volatile unsigned long * addr)
 {
-	unsigned long mask;
-	unsigned long *addr = (unsigned long *) address;
+	unsigned long mask = ~(1UL << CHOP_SHIFTCOUNT(nr));
 	unsigned long flags;
 
 	addr += (nr >> SHIFT_PER_LONG);
-	mask = 1L << CHOP_SHIFTCOUNT(nr);
 	_atomic_spin_lock_irqsave(addr, flags);
-	*addr &= ~mask;
+	*addr &= mask;
 	_atomic_spin_unlock_irqrestore(addr, flags);
 }
 
-static __inline__ void __clear_bit(unsigned long nr, volatile unsigned long * address)
+static __inline__ void __clear_bit(unsigned long nr, volatile unsigned long * addr)
 {
-	unsigned long mask;
-	unsigned long *addr = (unsigned long *) address;
+	unsigned long *m = (unsigned long *) addr + (nr >> SHIFT_PER_LONG);
 
-	addr += (nr >> SHIFT_PER_LONG);
-	mask = 1L << CHOP_SHIFTCOUNT(nr);
-	*addr &= ~mask;
+	*m &= ~(1UL << CHOP_SHIFTCOUNT(nr));
 }
 
-static __inline__ void change_bit(int nr, volatile unsigned long * address)
+static __inline__ void change_bit(int nr, volatile unsigned long * addr)
 {
-	unsigned long mask;
-	unsigned long *addr = (unsigned long *) address;
+	unsigned long mask = 1UL << CHOP_SHIFTCOUNT(nr);
 	unsigned long flags;
 
 	addr += (nr >> SHIFT_PER_LONG);
-	mask = 1L << CHOP_SHIFTCOUNT(nr);
 	_atomic_spin_lock_irqsave(addr, flags);
 	*addr ^= mask;
 	_atomic_spin_unlock_irqrestore(addr, flags);
 }
 
-static __inline__ void __change_bit(int nr, volatile unsigned long * address)
+static __inline__ void __change_bit(unsigned long nr, volatile unsigned long * addr)
 {
-	unsigned long mask;
-	unsigned long *addr = (unsigned long *) address;
+	unsigned long *m = (unsigned long *) addr + (nr >> SHIFT_PER_LONG);
 
-	addr += (nr >> SHIFT_PER_LONG);
-	mask = 1L << CHOP_SHIFTCOUNT(nr);
-	*addr ^= mask;
+	*m ^= 1UL << CHOP_SHIFTCOUNT(nr);
 }
 
-static __inline__ int test_and_set_bit(int nr, volatile unsigned long * address)
+static __inline__ int test_and_set_bit(int nr, volatile unsigned long * addr)
 {
-	unsigned long mask;
-	unsigned long *addr = (unsigned long *) address;
-	int oldbit;
+	unsigned long mask = 1UL << CHOP_SHIFTCOUNT(nr);
+	unsigned long oldbit;
 	unsigned long flags;
 
 	addr += (nr >> SHIFT_PER_LONG);
-	mask = 1L << CHOP_SHIFTCOUNT(nr);
 	_atomic_spin_lock_irqsave(addr, flags);
-	oldbit = (*addr & mask) ? 1 : 0;
-	*addr |= mask;
+	oldbit = *addr;
+	*addr = oldbit | mask;
 	_atomic_spin_unlock_irqrestore(addr, flags);
 
-	return oldbit;
+	return (oldbit & mask) ? 1 : 0;
 }
 
 static __inline__ int __test_and_set_bit(int nr, volatile unsigned long * address)
 {
-	unsigned long mask;
-	unsigned long *addr = (unsigned long *) address;
-	int oldbit;
+	unsigned long mask = 1UL << CHOP_SHIFTCOUNT(nr);
+	unsigned long oldbit;
+	unsigned long *addr = (unsigned long *)address + (nr >> SHIFT_PER_LONG);
 
-	addr += (nr >> SHIFT_PER_LONG);
-	mask = 1L << CHOP_SHIFTCOUNT(nr);
-	oldbit = (*addr & mask) ? 1 : 0;
-	*addr |= mask;
+	oldbit = *addr;
+	*addr = oldbit | mask;
 
-	return oldbit;
+	return (oldbit & mask) ? 1 : 0;
 }
 
-static __inline__ int test_and_clear_bit(int nr, volatile unsigned long * address)
+static __inline__ int test_and_clear_bit(int nr, volatile unsigned long * addr)
 {
-	unsigned long mask;
-	unsigned long *addr = (unsigned long *) address;
-	int oldbit;
+	unsigned long mask = 1UL << CHOP_SHIFTCOUNT(nr);
+	unsigned long oldbit;
 	unsigned long flags;
 
 	addr += (nr >> SHIFT_PER_LONG);
-	mask = 1L << CHOP_SHIFTCOUNT(nr);
 	_atomic_spin_lock_irqsave(addr, flags);
-	oldbit = (*addr & mask) ? 1 : 0;
-	*addr &= ~mask;
+	oldbit = *addr;
+	*addr = oldbit & ~mask;
 	_atomic_spin_unlock_irqrestore(addr, flags);
 
-	return oldbit;
+	return (oldbit & mask) ? 1 : 0;
 }
 
 static __inline__ int __test_and_clear_bit(int nr, volatile unsigned long * address)
 {
-	unsigned long mask;
-	unsigned long *addr = (unsigned long *) address;
-	int oldbit;
+	unsigned long mask = 1UL << CHOP_SHIFTCOUNT(nr);
+	unsigned long *addr = (unsigned long *)address + (nr >> SHIFT_PER_LONG);
+	unsigned long oldbit;
 
-	addr += (nr >> SHIFT_PER_LONG);
-	mask = 1L << CHOP_SHIFTCOUNT(nr);
-	oldbit = (*addr & mask) ? 1 : 0;
-	*addr &= ~mask;
+	oldbit = *addr;
+	*addr = oldbit & ~mask;
 
-	return oldbit;
+	return (oldbit & mask) ? 1 : 0;
 }
 
-static __inline__ int test_and_change_bit(int nr, volatile unsigned long * address)
+static __inline__ int test_and_change_bit(int nr, volatile unsigned long * addr)
 {
-	unsigned long mask;
-	unsigned long *addr = (unsigned long *) address;
-	int oldbit;
+	unsigned long mask = 1UL << CHOP_SHIFTCOUNT(nr);
+	unsigned long oldbit;
 	unsigned long flags;
 
 	addr += (nr >> SHIFT_PER_LONG);
-	mask = 1L << CHOP_SHIFTCOUNT(nr);
 	_atomic_spin_lock_irqsave(addr, flags);
-	oldbit = (*addr & mask) ? 1 : 0;
-	*addr ^= mask;
+	oldbit = *addr;
+	*addr = oldbit ^ mask;
 	_atomic_spin_unlock_irqrestore(addr, flags);
 
-	return oldbit;
+	return (oldbit & mask) ? 1 : 0;
 }
 
 static __inline__ int __test_and_change_bit(int nr, volatile unsigned long * address)
 {
-	unsigned long mask;
-	unsigned long *addr = (unsigned long *) address;
-	int oldbit;
+	unsigned long mask = 1UL << CHOP_SHIFTCOUNT(nr);
+	unsigned long *addr = (unsigned long *)address + (nr >> SHIFT_PER_LONG);
+	unsigned long oldbit;
 
-	addr += (nr >> SHIFT_PER_LONG);
-	mask = 1L << CHOP_SHIFTCOUNT(nr);
-	oldbit = (*addr & mask) ? 1 : 0;
-	*addr ^= mask;
+	oldbit = *addr;
+	*addr = oldbit ^ mask;
 
-	return oldbit;
+	return (oldbit & mask) ? 1 : 0;
 }
 
 static __inline__ int test_bit(int nr, const volatile unsigned long *address)
 {
-	unsigned long mask;
-	const unsigned long *addr = (const unsigned long *)address;
-	
-	addr += (nr >> SHIFT_PER_LONG);
-	mask = 1L << CHOP_SHIFTCOUNT(nr);
+	unsigned long mask = 1UL << CHOP_SHIFTCOUNT(nr);
+	const unsigned long *addr = (const unsigned long *)address + (nr >> SHIFT_PER_LONG);
 	
 	return !!(*addr & mask);
 }
@@ -229,7 +193,7 @@ static __inline__ unsigned long __ffs(unsigned long x)
 	unsigned long ret;
 
 	__asm__(
-#if BITS_PER_LONG > 32
+#ifdef __LP64__
 		" ldi       63,%1\n"
 		" extrd,u,*<>  %0,63,32,%%r0\n"
 		" extrd,u,*TR  %0,31,32,%0\n"	/* move top 32-bits down */
@@ -304,14 +268,7 @@ static __inline__ int fls(int x)
  * hweightN: returns the hamming weight (i.e. the number
  * of bits set) of a N-bit word
  */
-#define hweight64(x)						\
-({								\
-	unsigned long __x = (x);				\
-	unsigned int __w;					\
-	__w = generic_hweight32((unsigned int) __x);		\
-	__w += generic_hweight32((unsigned int) (__x>>32));	\
-	__w;							\
-})
+#define hweight64(x) generic_hweight64(x)
 #define hweight32(x) generic_hweight32(x)
 #define hweight16(x) generic_hweight16(x)
 #define hweight8(x) generic_hweight8(x)
@@ -324,7 +281,13 @@ static __inline__ int fls(int x)
  */
 static inline int sched_find_first_bit(const unsigned long *b)
 {
-#ifndef __LP64__
+#ifdef __LP64__
+	if (unlikely(b[0]))
+		return __ffs(b[0]);
+	if (unlikely(b[1]))
+		return __ffs(b[1]) + 64;
+	return __ffs(b[2]) + 128;
+#else
 	if (unlikely(b[0]))
 		return __ffs(b[0]);
 	if (unlikely(b[1]))
@@ -334,14 +297,6 @@ static inline int sched_find_first_bit(const unsigned long *b)
 	if (b[3])
 		return __ffs(b[3]) + 96;
 	return __ffs(b[4]) + 128;
-#else
-	if (unlikely(b[0]))
-		return __ffs(b[0]);
-	if (unlikely(((unsigned int)b[1])))
-		return __ffs(b[1]) + 64;
-	if (b[1] >> 32)
-		return __ffs(b[1] >> 32) + 96;
-	return __ffs(b[2]) + 128;
 #endif
 }
 
@@ -391,7 +346,7 @@ found_middle:
 
 static __inline__ unsigned long find_next_bit(const unsigned long *addr, unsigned long size, unsigned long offset)
 {
-	const unsigned long *p = addr + (offset >> 6);
+	const unsigned long *p = addr + (offset >> SHIFT_PER_LONG);
 	unsigned long result = offset & ~(BITS_PER_LONG-1);
 	unsigned long tmp;
 
@@ -445,71 +400,90 @@ found_middle:
  * test_and_{set,clear}_bit guarantee atomicity without
  * disabling interrupts.
  */
-#ifdef __LP64__
-#define ext2_set_bit(nr, addr)		__test_and_set_bit((nr) ^ 0x38, (unsigned long *)addr)
-#define ext2_set_bit_atomic(l,nr,addr)  test_and_set_bit((nr) ^ 0x38, (unsigned long *)addr)
-#define ext2_clear_bit(nr, addr)	__test_and_clear_bit((nr) ^ 0x38, (unsigned long *)addr)
-#define ext2_clear_bit_atomic(l,nr,addr) test_and_clear_bit((nr) ^ 0x38, (unsigned long *)addr)
-#else
-#define ext2_set_bit(nr, addr)		__test_and_set_bit((nr) ^ 0x18, (unsigned long *)addr)
-#define ext2_set_bit_atomic(l,nr,addr)  test_and_set_bit((nr) ^ 0x18, (unsigned long *)addr)
-#define ext2_clear_bit(nr, addr)	__test_and_clear_bit((nr) ^ 0x18, (unsigned long *)addr)
-#define ext2_clear_bit_atomic(l,nr,addr) test_and_clear_bit((nr) ^ 0x18, (unsigned long *)addr)
-#endif
 
-#endif	/* __KERNEL__ */
+/* '3' is bits per byte */
+#define LE_BYTE_ADDR ((sizeof(unsigned long) - 1) << 3)
 
-static __inline__ int ext2_test_bit(int nr, __const__ void * addr)
-{
-	__const__ unsigned char	*ADDR = (__const__ unsigned char *) addr;
+#define ext2_test_bit(nr, addr) \
+			test_bit((nr)	^ LE_BYTE_ADDR, (unsigned long *)addr)
+#define ext2_set_bit(nr, addr)	\
+		__test_and_set_bit((nr) ^ LE_BYTE_ADDR, (unsigned long *)addr)
+#define ext2_clear_bit(nr, addr) \
+		__test_and_clear_bit((nr) ^ LE_BYTE_ADDR, (unsigned long *)addr)
 
-	return (ADDR[nr >> 3] >> (nr & 7)) & 1;
-}
+#define ext2_set_bit_atomic(l,nr,addr) \
+		test_and_set_bit((nr)   ^ LE_BYTE_ADDR, (unsigned long *)addr)
+#define ext2_clear_bit_atomic(l,nr,addr) \
+		test_and_clear_bit( (nr) ^ LE_BYTE_ADDR, (unsigned long *)addr)
+
+#endif	/* __KERNEL__ */
 
-/*
- * This implementation of ext2_find_{first,next}_zero_bit was stolen from
- * Linus' asm-alpha/bitops.h and modified for a big-endian machine.
- */
 
 #define ext2_find_first_zero_bit(addr, size) \
-        ext2_find_next_zero_bit((addr), (size), 0)
+	ext2_find_next_zero_bit((addr), (size), 0)
 
-extern __inline__ unsigned long ext2_find_next_zero_bit(void *addr,
-	unsigned long size, unsigned long offset)
+/* include/linux/byteorder does not support "unsigned long" type */
+static inline unsigned long ext2_swabp(unsigned long * x)
 {
-	unsigned int *p = ((unsigned int *) addr) + (offset >> 5);
-	unsigned int result = offset & ~31UL;
-	unsigned int tmp;
+#ifdef __LP64__
+	return (unsigned long) __swab64p((u64 *) x);
+#else
+	return (unsigned long) __swab32p((u32 *) x);
+#endif
+}
+
+/* include/linux/byteorder doesn't support "unsigned long" type */
+static inline unsigned long ext2_swab(unsigned long y)
+{
+#ifdef __LP64__
+	return (unsigned long) __swab64((u64) y);
+#else
+	return (unsigned long) __swab32((u32) y);
+#endif
+}
+
+static __inline__ unsigned long ext2_find_next_zero_bit(void *addr, unsigned long size, unsigned long offset)
+{
+	unsigned long *p = (unsigned long *) addr + (offset >> SHIFT_PER_LONG);
+	unsigned long result = offset & ~(BITS_PER_LONG - 1);
+	unsigned long tmp;
 
 	if (offset >= size)
 		return size;
 	size -= result;
-	offset &= 31UL;
+	offset &= (BITS_PER_LONG - 1UL);
 	if (offset) {
-		tmp = cpu_to_le32p(p++);
-		tmp |= ~0UL >> (32-offset);
-		if (size < 32)
+		tmp = ext2_swabp(p++);
+		tmp |= (~0UL >> (BITS_PER_LONG - offset));
+		if (size < BITS_PER_LONG)
 			goto found_first;
-		if (tmp != ~0U)
+		if (~tmp)
 			goto found_middle;
-		size -= 32;
-		result += 32;
+		size -= BITS_PER_LONG;
+		result += BITS_PER_LONG;
 	}
-	while (size >= 32) {
-		if ((tmp = cpu_to_le32p(p++)) != ~0U)
-			goto found_middle;
-		result += 32;
-		size -= 32;
+
+	while (size & ~(BITS_PER_LONG - 1)) {
+		if (~(tmp = *(p++)))
+			goto found_middle_swap;
+		result += BITS_PER_LONG;
+		size -= BITS_PER_LONG;
 	}
 	if (!size)
 		return result;
-	tmp = cpu_to_le32p(p);
+	tmp = ext2_swabp(p);
 found_first:
-	tmp |= ~0U << size;
+	tmp |= ~0UL << size;
+	if (tmp == ~0UL)	/* Are any bits zero? */
+		return result + size; /* Nope. Skip ffz */
 found_middle:
 	return result + ffz(tmp);
+
+found_middle_swap:
+	return result + ffz(ext2_swab(tmp));
 }
 
+
 /* Bitmap functions for the minix filesystem.  */
 #define minix_test_and_set_bit(nr,addr) ext2_set_bit(nr,addr)
 #define minix_set_bit(nr,addr) ((void)ext2_set_bit(nr,addr))
diff --git a/include/asm-parisc/types.h b/include/asm-parisc/types.h
index d21b9d0..34fdce3 100644
--- a/include/asm-parisc/types.h
+++ b/include/asm-parisc/types.h
@@ -33,8 +33,10 @@ typedef unsigned long long __u64;
 
 #ifdef __LP64__
 #define BITS_PER_LONG 64
+#define SHIFT_PER_LONG 6
 #else
 #define BITS_PER_LONG 32
+#define SHIFT_PER_LONG 5
 #endif
 
 #ifndef __ASSEMBLY__