mirror of
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git
synced 2024-08-24 09:50:04 +00:00
![Amir Goldstein](/assets/img/avatar_default.png)
The comment claims that this helper will try not to loose bits, but for 64bit long it looses the high bits before hashing 64bit long into 32bit int. Use the helper hash_long() to do the right thing for 64bit long. For 32bit long, there is no change. All the callers of end_name_hash() either assign the result to qstr->hash, which is u32 or return the result as an int value (e.g. full_name_hash()). Change the helper return type to int to conform to its users. [ It took me a while to apply this, because my initial reaction to it was - incorrectly - that it could make for slower code. After having looked more at it, I take back all my complaints about the patch, Amir was right and I was mis-reading things or just being stupid. I also don't worry too much about the possible performance impact of this on 64-bit, since most architectures that actually care about performance end up not using this very much (the dcache code is the most performance-critical, but the word-at-a-time case uses its own hashing anyway). So this ends up being mostly used for filesystems that do their own degraded hashing (usually because they want a case-insensitive comparison function). A _tiny_ worry remains, in that not everybody uses DCACHE_WORD_ACCESS, and then this potentially makes things more expensive on 64-bit architectures with slow or lacking multipliers even for the normal case. That said, realistically the only such architecture I can think of is PA-RISC. Nobody really cares about performance on that, it's more of a "look ma, I've got warts^W an odd machine" platform. So the patch is fine, and all my initial worries were just misplaced from not looking at this properly. - Linus ] Signed-off-by: Amir Goldstein <amir73il@gmail.com> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
79 lines
2.7 KiB
C
79 lines
2.7 KiB
C
/* SPDX-License-Identifier: GPL-2.0 */
|
|
#ifndef __LINUX_STRINGHASH_H
|
|
#define __LINUX_STRINGHASH_H
|
|
|
|
#include <linux/compiler.h> /* For __pure */
|
|
#include <linux/types.h> /* For u32, u64 */
|
|
#include <linux/hash.h>
|
|
|
|
/*
|
|
* Routines for hashing strings of bytes to a 32-bit hash value.
|
|
*
|
|
* These hash functions are NOT GUARANTEED STABLE between kernel
|
|
* versions, architectures, or even repeated boots of the same kernel.
|
|
* (E.g. they may depend on boot-time hardware detection or be
|
|
* deliberately randomized.)
|
|
*
|
|
* They are also not intended to be secure against collisions caused by
|
|
* malicious inputs; much slower hash functions are required for that.
|
|
*
|
|
* They are optimized for pathname components, meaning short strings.
|
|
* Even if a majority of files have longer names, the dynamic profile of
|
|
* pathname components skews short due to short directory names.
|
|
* (E.g. /usr/lib/libsesquipedalianism.so.3.141.)
|
|
*/
|
|
|
|
/*
|
|
* Version 1: one byte at a time. Example of use:
|
|
*
|
|
* unsigned long hash = init_name_hash;
|
|
* while (*p)
|
|
* hash = partial_name_hash(tolower(*p++), hash);
|
|
* hash = end_name_hash(hash);
|
|
*
|
|
* Although this is designed for bytes, fs/hfsplus/unicode.c
|
|
* abuses it to hash 16-bit values.
|
|
*/
|
|
|
|
/* Hash courtesy of the R5 hash in reiserfs modulo sign bits */
|
|
#define init_name_hash(salt) (unsigned long)(salt)
|
|
|
|
/* partial hash update function. Assume roughly 4 bits per character */
|
|
static inline unsigned long
|
|
partial_name_hash(unsigned long c, unsigned long prevhash)
|
|
{
|
|
return (prevhash + (c << 4) + (c >> 4)) * 11;
|
|
}
|
|
|
|
/*
|
|
* Finally: cut down the number of bits to a int value (and try to avoid
|
|
* losing bits). This also has the property (wanted by the dcache)
|
|
* that the msbits make a good hash table index.
|
|
*/
|
|
static inline unsigned int end_name_hash(unsigned long hash)
|
|
{
|
|
return hash_long(hash, 32);
|
|
}
|
|
|
|
/*
|
|
* Version 2: One word (32 or 64 bits) at a time.
|
|
* If CONFIG_DCACHE_WORD_ACCESS is defined (meaning <asm/word-at-a-time.h>
|
|
* exists, which describes major Linux platforms like x86 and ARM), then
|
|
* this computes a different hash function much faster.
|
|
*
|
|
* If not set, this falls back to a wrapper around the preceding.
|
|
*/
|
|
extern unsigned int __pure full_name_hash(const void *salt, const char *, unsigned int);
|
|
|
|
/*
|
|
* A hash_len is a u64 with the hash of a string in the low
|
|
* half and the length in the high half.
|
|
*/
|
|
#define hashlen_hash(hashlen) ((u32)(hashlen))
|
|
#define hashlen_len(hashlen) ((u32)((hashlen) >> 32))
|
|
#define hashlen_create(hash, len) ((u64)(len)<<32 | (u32)(hash))
|
|
|
|
/* Return the "hash_len" (hash and length) of a null-terminated string */
|
|
extern u64 __pure hashlen_string(const void *salt, const char *name);
|
|
|
|
#endif /* __LINUX_STRINGHASH_H */
|