1
0
Fork 0
alistair23-linux/include/asm-parisc/bitops.h

234 lines
6.0 KiB
C
Raw Normal View History

#ifndef _PARISC_BITOPS_H
#define _PARISC_BITOPS_H
#include <linux/compiler.h>
[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>
2005-10-21 20:45:22 -06:00
#include <asm/types.h> /* for BITS_PER_LONG/SHIFT_PER_LONG */
#include <asm/byteorder.h>
#include <asm/atomic.h>
/*
* HP-PARISC specific bit operations
* for a detailed description of the functions please refer
* to include/asm-i386/bitops.h or kerneldoc
*/
[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>
2005-10-21 20:45:22 -06:00
#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()
[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>
2005-10-21 20:45:22 -06:00
/* 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)
{
[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>
2005-10-21 20:45:22 -06:00
unsigned long mask = 1UL << CHOP_SHIFTCOUNT(nr);
unsigned long flags;
addr += (nr >> SHIFT_PER_LONG);
_atomic_spin_lock_irqsave(addr, flags);
*addr |= mask;
_atomic_spin_unlock_irqrestore(addr, flags);
}
[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>
2005-10-21 20:45:22 -06:00
static __inline__ void clear_bit(int nr, volatile unsigned long * addr)
{
[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>
2005-10-21 20:45:22 -06:00
unsigned long mask = ~(1UL << CHOP_SHIFTCOUNT(nr));
unsigned long flags;
addr += (nr >> SHIFT_PER_LONG);
_atomic_spin_lock_irqsave(addr, flags);
[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>
2005-10-21 20:45:22 -06:00
*addr &= mask;
_atomic_spin_unlock_irqrestore(addr, flags);
}
[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>
2005-10-21 20:45:22 -06:00
static __inline__ void change_bit(int nr, volatile unsigned long * addr)
{
[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>
2005-10-21 20:45:22 -06:00
unsigned long mask = 1UL << CHOP_SHIFTCOUNT(nr);
unsigned long flags;
addr += (nr >> SHIFT_PER_LONG);
_atomic_spin_lock_irqsave(addr, flags);
*addr ^= mask;
_atomic_spin_unlock_irqrestore(addr, flags);
}
[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>
2005-10-21 20:45:22 -06:00
static __inline__ int test_and_set_bit(int nr, volatile unsigned long * addr)
{
[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>
2005-10-21 20:45:22 -06:00
unsigned long mask = 1UL << CHOP_SHIFTCOUNT(nr);
unsigned long old;
unsigned long flags;
int set;
addr += (nr >> SHIFT_PER_LONG);
_atomic_spin_lock_irqsave(addr, flags);
old = *addr;
set = (old & mask) ? 1 : 0;
if (!set)
*addr = old | mask;
_atomic_spin_unlock_irqrestore(addr, flags);
return set;
}
[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>
2005-10-21 20:45:22 -06:00
static __inline__ int test_and_clear_bit(int nr, volatile unsigned long * addr)
{
[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>
2005-10-21 20:45:22 -06:00
unsigned long mask = 1UL << CHOP_SHIFTCOUNT(nr);
unsigned long old;
unsigned long flags;
int set;
addr += (nr >> SHIFT_PER_LONG);
_atomic_spin_lock_irqsave(addr, flags);
old = *addr;
set = (old & mask) ? 1 : 0;
if (set)
*addr = old & ~mask;
_atomic_spin_unlock_irqrestore(addr, flags);
return set;
}
[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>
2005-10-21 20:45:22 -06:00
static __inline__ int test_and_change_bit(int nr, volatile unsigned long * addr)
{
[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>
2005-10-21 20:45:22 -06:00
unsigned long mask = 1UL << CHOP_SHIFTCOUNT(nr);
unsigned long oldbit;
unsigned long flags;
addr += (nr >> SHIFT_PER_LONG);
_atomic_spin_lock_irqsave(addr, flags);
[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>
2005-10-21 20:45:22 -06:00
oldbit = *addr;
*addr = oldbit ^ mask;
_atomic_spin_unlock_irqrestore(addr, flags);
[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>
2005-10-21 20:45:22 -06:00
return (oldbit & mask) ? 1 : 0;
}
#include <asm-generic/bitops/non-atomic.h>
#ifdef __KERNEL__
/**
* __ffs - find first bit in word. returns 0 to "BITS_PER_LONG-1".
* @word: The word to search
*
* __ffs() return is undefined if no bit is set.
*
* 32-bit fast __ffs by LaMont Jones "lamont At hp com".
* 64-bit enhancement by Grant Grundler "grundler At parisc-linux org".
* (with help from willy/jejb to get the semantics right)
*
* This algorithm avoids branches by making use of nullification.
* One side effect of "extr" instructions is it sets PSW[N] bit.
* How PSW[N] (nullify next insn) gets set is determined by the
* "condition" field (eg "<>" or "TR" below) in the extr* insn.
* Only the 1st and one of either the 2cd or 3rd insn will get executed.
* Each set of 3 insn will get executed in 2 cycles on PA8x00 vs 16 or so
* cycles for each mispredicted branch.
*/
static __inline__ unsigned long __ffs(unsigned long x)
{
unsigned long ret;
__asm__(
#ifdef CONFIG_64BIT
" ldi 63,%1\n"
" extrd,u,*<> %0,63,32,%%r0\n"
" extrd,u,*TR %0,31,32,%0\n" /* move top 32-bits down */
" addi -32,%1,%1\n"
#else
" ldi 31,%1\n"
#endif
" extru,<> %0,31,16,%%r0\n"
" extru,TR %0,15,16,%0\n" /* xxxx0000 -> 0000xxxx */
" addi -16,%1,%1\n"
" extru,<> %0,31,8,%%r0\n"
" extru,TR %0,23,8,%0\n" /* 0000xx00 -> 000000xx */
" addi -8,%1,%1\n"
" extru,<> %0,31,4,%%r0\n"
" extru,TR %0,27,4,%0\n" /* 000000x0 -> 0000000x */
" addi -4,%1,%1\n"
" extru,<> %0,31,2,%%r0\n"
" extru,TR %0,29,2,%0\n" /* 0000000y, 1100b -> 0011b */
" addi -2,%1,%1\n"
" extru,= %0,31,1,%%r0\n" /* check last bit */
" addi -1,%1,%1\n"
: "+r" (x), "=r" (ret) );
return ret;
}
#include <asm-generic/bitops/ffz.h>
/*
* ffs: find first bit set. returns 1 to BITS_PER_LONG or 0 (if none 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).
*/
static __inline__ int ffs(int x)
{
return x ? (__ffs((unsigned long)x) + 1) : 0;
}
/*
* fls: find last (most significant) bit set.
* fls(0) = 0, fls(1) = 1, fls(0x80000000) = 32.
*/
static __inline__ int fls(int x)
{
int ret;
if (!x)
return 0;
__asm__(
" ldi 1,%1\n"
" extru,<> %0,15,16,%%r0\n"
" zdep,TR %0,15,16,%0\n" /* xxxx0000 */
" addi 16,%1,%1\n"
" extru,<> %0,7,8,%%r0\n"
" zdep,TR %0,23,24,%0\n" /* xx000000 */
" addi 8,%1,%1\n"
" extru,<> %0,3,4,%%r0\n"
" zdep,TR %0,27,28,%0\n" /* x0000000 */
" addi 4,%1,%1\n"
" extru,<> %0,1,2,%%r0\n"
" zdep,TR %0,29,30,%0\n" /* y0000000 (y&3 = 0) */
" addi 2,%1,%1\n"
" extru,= %0,0,1,%%r0\n"
" addi 1,%1,%1\n" /* if y & 8, add 1 */
: "+r" (x), "=r" (ret) );
return ret;
}
#include <asm-generic/bitops/fls64.h>
#include <asm-generic/bitops/hweight.h>
#include <asm-generic/bitops/sched.h>
#endif /* __KERNEL__ */
#include <asm-generic/bitops/find.h>
#ifdef __KERNEL__
#include <asm-generic/bitops/ext2-non-atomic.h>
[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>
2005-10-21 20:45:22 -06:00
/* '3' is bits per byte */
#define LE_BYTE_ADDR ((sizeof(unsigned long) - 1) << 3)
[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>
2005-10-21 20:45:22 -06:00
#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__ */
#include <asm-generic/bitops/minix-le.h>
#endif /* _PARISC_BITOPS_H */