Eliminate bad hash multipliers from hash_32() and hash_64()
The "simplified" prime multipliers made very bad hash functions, so get rid
of them. This completes the work of 689de1d6ca
.
To avoid the inefficiency which was the motivation for the "simplified"
multipliers, hash_64() on 32-bit systems is changed to use a different
algorithm. It makes two calls to hash_32() instead.
drivers/media/usb/dvb-usb-v2/af9015.c uses the old GOLDEN_RATIO_PRIME_32
for some horrible reason, so it inherits a copy of the old definition.
Signed-off-by: George Spelvin <linux@sciencehorizons.net>
Cc: Antti Palosaari <crope@iki.fi>
Cc: Mauro Carvalho Chehab <m.chehab@samsung.com>
This commit is contained in:
parent
92d567740f
commit
ef703f49a6
|
@ -398,6 +398,8 @@ error:
|
|||
}
|
||||
|
||||
#define AF9015_EEPROM_SIZE 256
|
||||
/* 2^31 + 2^29 - 2^25 + 2^22 - 2^19 - 2^16 + 1 */
|
||||
#define GOLDEN_RATIO_PRIME_32 0x9e370001UL
|
||||
|
||||
/* hash (and dump) eeprom */
|
||||
static int af9015_eeprom_hash(struct dvb_usb_device *d)
|
||||
|
|
|
@ -3,85 +3,65 @@
|
|||
/* Fast hashing routine for ints, longs and pointers.
|
||||
(C) 2002 Nadia Yvette Chambers, IBM */
|
||||
|
||||
/*
|
||||
* Knuth recommends primes in approximately golden ratio to the maximum
|
||||
* integer representable by a machine word for multiplicative hashing.
|
||||
* Chuck Lever verified the effectiveness of this technique:
|
||||
* http://www.citi.umich.edu/techreports/reports/citi-tr-00-1.pdf
|
||||
*
|
||||
* These primes are chosen to be bit-sparse, that is operations on
|
||||
* them can use shifts and additions instead of multiplications for
|
||||
* machines where multiplications are slow.
|
||||
*/
|
||||
|
||||
#include <asm/types.h>
|
||||
#include <linux/compiler.h>
|
||||
|
||||
/* 2^31 + 2^29 - 2^25 + 2^22 - 2^19 - 2^16 + 1 */
|
||||
#define GOLDEN_RATIO_PRIME_32 0x9e370001UL
|
||||
/* 2^63 + 2^61 - 2^57 + 2^54 - 2^51 - 2^18 + 1 */
|
||||
#define GOLDEN_RATIO_PRIME_64 0x9e37fffffffc0001UL
|
||||
|
||||
/*
|
||||
* The "GOLDEN_RATIO_PRIME" is used in ifs/btrfs/brtfs_inode.h and
|
||||
* fs/inode.c. It's not actually prime any more (the previous primes
|
||||
* were actively bad for hashing), but the name remains.
|
||||
*/
|
||||
#if BITS_PER_LONG == 32
|
||||
#define GOLDEN_RATIO_PRIME GOLDEN_RATIO_PRIME_32
|
||||
#define GOLDEN_RATIO_PRIME GOLDEN_RATIO_32
|
||||
#define hash_long(val, bits) hash_32(val, bits)
|
||||
#elif BITS_PER_LONG == 64
|
||||
#define hash_long(val, bits) hash_64(val, bits)
|
||||
#define GOLDEN_RATIO_PRIME GOLDEN_RATIO_PRIME_64
|
||||
#define GOLDEN_RATIO_PRIME GOLDEN_RATIO_64
|
||||
#else
|
||||
#error Wordsize not 32 or 64
|
||||
#endif
|
||||
|
||||
/*
|
||||
* The above primes are actively bad for hashing, since they are
|
||||
* too sparse. The 32-bit one is mostly ok, the 64-bit one causes
|
||||
* real problems. Besides, the "prime" part is pointless for the
|
||||
* multiplicative hash.
|
||||
* This hash multiplies the input by a large odd number and takes the
|
||||
* high bits. Since multiplication propagates changes to the most
|
||||
* significant end only, it is essential that the high bits of the
|
||||
* product be used for the hash value.
|
||||
*
|
||||
* Chuck Lever verified the effectiveness of this technique:
|
||||
* http://www.citi.umich.edu/techreports/reports/citi-tr-00-1.pdf
|
||||
*
|
||||
* Although a random odd number will do, it turns out that the golden
|
||||
* ratio phi = (sqrt(5)-1)/2, or its negative, has particularly nice
|
||||
* properties.
|
||||
* properties. (See Knuth vol 3, section 6.4, exercise 9.)
|
||||
*
|
||||
* These are the negative, (1 - phi) = (phi^2) = (3 - sqrt(5))/2.
|
||||
* (See Knuth vol 3, section 6.4, exercise 9.)
|
||||
* These are the negative, (1 - phi) = phi**2 = (3 - sqrt(5))/2,
|
||||
* which is very slightly easier to multiply by and makes no
|
||||
* difference to the hash distribution.
|
||||
*/
|
||||
#define GOLDEN_RATIO_32 0x61C88647
|
||||
#define GOLDEN_RATIO_64 0x61C8864680B583EBull
|
||||
|
||||
static __always_inline u32 hash_64(u64 val, unsigned int bits)
|
||||
|
||||
static inline u32 __hash_32(u32 val)
|
||||
{
|
||||
u64 hash = val;
|
||||
|
||||
#if BITS_PER_LONG == 64
|
||||
hash = hash * GOLDEN_RATIO_64;
|
||||
#else
|
||||
/* Sigh, gcc can't optimise this alone like it does for 32 bits. */
|
||||
u64 n = hash;
|
||||
n <<= 18;
|
||||
hash -= n;
|
||||
n <<= 33;
|
||||
hash -= n;
|
||||
n <<= 3;
|
||||
hash += n;
|
||||
n <<= 3;
|
||||
hash -= n;
|
||||
n <<= 4;
|
||||
hash += n;
|
||||
n <<= 2;
|
||||
hash += n;
|
||||
#endif
|
||||
|
||||
/* High bits are more random, so use them. */
|
||||
return (u32)(hash >> (64 - bits));
|
||||
return val * GOLDEN_RATIO_32;
|
||||
}
|
||||
|
||||
static inline u32 hash_32(u32 val, unsigned int bits)
|
||||
{
|
||||
/* On some cpus multiply is faster, on others gcc will do shifts */
|
||||
u32 hash = val * GOLDEN_RATIO_PRIME_32;
|
||||
|
||||
/* High bits are more random, so use them. */
|
||||
return hash >> (32 - bits);
|
||||
return __hash_32(val) >> (32 - bits);
|
||||
}
|
||||
|
||||
static __always_inline u32 hash_64(u64 val, unsigned int bits)
|
||||
{
|
||||
#if BITS_PER_LONG == 64
|
||||
/* 64x64-bit multiply is efficient on all 64-bit processors */
|
||||
return val * GOLDEN_RATIO_64 >> (64 - bits);
|
||||
#else
|
||||
/* Hash 64 bits using only 32x32-bit multiply. */
|
||||
return hash_32((u32)val ^ __hash_32(val >> 32), bits);
|
||||
#endif
|
||||
}
|
||||
|
||||
static inline u32 hash_ptr(const void *ptr, unsigned int bits)
|
||||
|
@ -89,6 +69,7 @@ static inline u32 hash_ptr(const void *ptr, unsigned int bits)
|
|||
return hash_long((unsigned long)ptr, bits);
|
||||
}
|
||||
|
||||
/* This really should be called fold32_ptr; it does no hashing to speak of. */
|
||||
static inline u32 hash32_ptr(const void *ptr)
|
||||
{
|
||||
unsigned long val = (unsigned long)ptr;
|
||||
|
|
Loading…
Reference in New Issue