word-at-a-time: make the interfaces truly generic
This changes the interfaces in <asm/word-at-a-time.h> to be a bit more
complicated, but a lot more generic.
In particular, it allows us to really do the operations efficiently on
both little-endian and big-endian machines, pretty much regardless of
machine details. For example, if you can rely on a fast population
count instruction on your architecture, this will allow you to make your
optimized <asm/word-at-a-time.h> file with that.
NOTE! The "generic" version in include/asm-generic/word-at-a-time.h is
not truly generic, it actually only works on big-endian. Why? Because
on little-endian the generic algorithms are wasteful, since you can
inevitably do better. The x86 implementation is an example of that.
(The only truly non-generic part of the asm-generic implementation is
the "find_zero()" function, and you could make a little-endian version
of it. And if the Kbuild infrastructure allowed us to pick a particular
header file, that would be lovely)
The <asm/word-at-a-time.h> functions are as follows:
- WORD_AT_A_TIME_CONSTANTS: specific constants that the algorithm
uses.
- has_zero(): take a word, and determine if it has a zero byte in it.
It gets the word, the pointer to the constant pool, and a pointer to
an intermediate "data" field it can set.
This is the "quick-and-dirty" zero tester: it's what is run inside
the hot loops.
- "prep_zero_mask()": take the word, the data that has_zero() produced,
and the constant pool, and generate an *exact* mask of which byte had
the first zero. This is run directly *outside* the loop, and allows
the "has_zero()" function to answer the "is there a zero byte"
question without necessarily getting exactly *which* byte is the
first one to contain a zero.
If you do multiple byte lookups concurrently (eg "hash_name()", which
looks for both NUL and '/' bytes), after you've done the prep_zero_mask()
phase, the result of those can be or'ed together to get the "either
or" case.
- The result from "prep_zero_mask()" can then be fed into "find_zero()"
(to find the byte offset of the first byte that was zero) or into
"zero_bytemask()" (to find the bytemask of the bytes preceding the
zero byte).
The existence of zero_bytemask() is optional, and is not necessary
for the normal string routines. But dentry name hashing needs it, so
if you enable DENTRY_WORD_AT_A_TIME you need to expose it.
This changes the generic strncpy_from_user() function and the dentry
hashing functions to use these modified word-at-a-time interfaces. This
gets us back to the optimized state of the x86 strncpy that we lost in
the previous commit when moving over to the generic version.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-05-27 00:43:17 +07:00
|
|
|
#ifndef _ASM_WORD_AT_A_TIME_H
|
|
|
|
#define _ASM_WORD_AT_A_TIME_H
|
|
|
|
|
|
|
|
#include <linux/kernel.h>
|
2015-04-29 23:48:40 +07:00
|
|
|
#include <asm/byteorder.h>
|
|
|
|
|
|
|
|
#ifdef __BIG_ENDIAN
|
word-at-a-time: make the interfaces truly generic
This changes the interfaces in <asm/word-at-a-time.h> to be a bit more
complicated, but a lot more generic.
In particular, it allows us to really do the operations efficiently on
both little-endian and big-endian machines, pretty much regardless of
machine details. For example, if you can rely on a fast population
count instruction on your architecture, this will allow you to make your
optimized <asm/word-at-a-time.h> file with that.
NOTE! The "generic" version in include/asm-generic/word-at-a-time.h is
not truly generic, it actually only works on big-endian. Why? Because
on little-endian the generic algorithms are wasteful, since you can
inevitably do better. The x86 implementation is an example of that.
(The only truly non-generic part of the asm-generic implementation is
the "find_zero()" function, and you could make a little-endian version
of it. And if the Kbuild infrastructure allowed us to pick a particular
header file, that would be lovely)
The <asm/word-at-a-time.h> functions are as follows:
- WORD_AT_A_TIME_CONSTANTS: specific constants that the algorithm
uses.
- has_zero(): take a word, and determine if it has a zero byte in it.
It gets the word, the pointer to the constant pool, and a pointer to
an intermediate "data" field it can set.
This is the "quick-and-dirty" zero tester: it's what is run inside
the hot loops.
- "prep_zero_mask()": take the word, the data that has_zero() produced,
and the constant pool, and generate an *exact* mask of which byte had
the first zero. This is run directly *outside* the loop, and allows
the "has_zero()" function to answer the "is there a zero byte"
question without necessarily getting exactly *which* byte is the
first one to contain a zero.
If you do multiple byte lookups concurrently (eg "hash_name()", which
looks for both NUL and '/' bytes), after you've done the prep_zero_mask()
phase, the result of those can be or'ed together to get the "either
or" case.
- The result from "prep_zero_mask()" can then be fed into "find_zero()"
(to find the byte offset of the first byte that was zero) or into
"zero_bytemask()" (to find the bytemask of the bytes preceding the
zero byte).
The existence of zero_bytemask() is optional, and is not necessary
for the normal string routines. But dentry name hashing needs it, so
if you enable DENTRY_WORD_AT_A_TIME you need to expose it.
This changes the generic strncpy_from_user() function and the dentry
hashing functions to use these modified word-at-a-time interfaces. This
gets us back to the optimized state of the x86 strncpy that we lost in
the previous commit when moving over to the generic version.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-05-27 00:43:17 +07:00
|
|
|
|
|
|
|
struct word_at_a_time {
|
|
|
|
const unsigned long high_bits, low_bits;
|
|
|
|
};
|
|
|
|
|
|
|
|
#define WORD_AT_A_TIME_CONSTANTS { REPEAT_BYTE(0xfe) + 1, REPEAT_BYTE(0x7f) }
|
|
|
|
|
|
|
|
/* Bit set in the bytes that have a zero */
|
|
|
|
static inline long prep_zero_mask(unsigned long val, unsigned long rhs, const struct word_at_a_time *c)
|
|
|
|
{
|
|
|
|
unsigned long mask = (val & c->low_bits) + c->low_bits;
|
|
|
|
return ~(mask | rhs);
|
|
|
|
}
|
|
|
|
|
|
|
|
#define create_zero_mask(mask) (mask)
|
|
|
|
|
|
|
|
static inline long find_zero(unsigned long mask)
|
|
|
|
{
|
|
|
|
long byte = 0;
|
|
|
|
#ifdef CONFIG_64BIT
|
|
|
|
if (mask >> 32)
|
|
|
|
mask >>= 32;
|
|
|
|
else
|
|
|
|
byte = 4;
|
|
|
|
#endif
|
|
|
|
if (mask >> 16)
|
|
|
|
mask >>= 16;
|
|
|
|
else
|
|
|
|
byte += 2;
|
|
|
|
return (mask >> 8) ? byte : byte + 1;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline bool has_zero(unsigned long val, unsigned long *data, const struct word_at_a_time *c)
|
|
|
|
{
|
|
|
|
unsigned long rhs = val | c->low_bits;
|
|
|
|
*data = rhs;
|
|
|
|
return (val + c->high_bits) & ~rhs;
|
|
|
|
}
|
|
|
|
|
2013-12-13 00:40:22 +07:00
|
|
|
#ifndef zero_bytemask
|
2014-05-01 04:22:19 +07:00
|
|
|
#define zero_bytemask(mask) (~1ul << __fls(mask))
|
word-at-a-time: avoid undefined behaviour in zero_bytemask macro
The asm-generic, big-endian version of zero_bytemask creates a mask of
bytes preceding the first zero-byte by left shifting ~0ul based on the
position of the first zero byte.
Unfortunately, if the first (top) byte is zero, the output of
prep_zero_mask has only the top bit set, resulting in undefined C
behaviour as we shift left by an amount equal to the width of the type.
As it happens, GCC doesn't manage to spot this through the call to fls(),
but the issue remains if architectures choose to implement their shift
instructions differently.
An example would be arch/arm/ (AArch32), where LSL Rd, Rn, #32 results
in Rd == 0x0, whilst on arch/arm64 (AArch64) LSL Xd, Xn, #64 results in
Xd == Xn.
Rather than check explicitly for the problematic shift, this patch adds
an extra shift by 1, replacing fls with __fls. Since zero_bytemask is
never called with a zero argument (has_zero() is used to check the data
first), we don't need to worry about calling __fls(0), which is
undefined.
Cc: <stable@vger.kernel.org>
Cc: Victor Kamensky <victor.kamensky@linaro.org>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-04-23 23:52:52 +07:00
|
|
|
#endif
|
2013-12-13 00:40:22 +07:00
|
|
|
|
2015-04-29 23:48:40 +07:00
|
|
|
#else
|
|
|
|
|
|
|
|
/*
|
|
|
|
* The optimal byte mask counting is probably going to be something
|
|
|
|
* that is architecture-specific. If you have a reliably fast
|
|
|
|
* bit count instruction, that might be better than the multiply
|
|
|
|
* and shift, for example.
|
|
|
|
*/
|
|
|
|
struct word_at_a_time {
|
|
|
|
const unsigned long one_bits, high_bits;
|
|
|
|
};
|
|
|
|
|
|
|
|
#define WORD_AT_A_TIME_CONSTANTS { REPEAT_BYTE(0x01), REPEAT_BYTE(0x80) }
|
|
|
|
|
|
|
|
#ifdef CONFIG_64BIT
|
|
|
|
|
|
|
|
/*
|
|
|
|
* Jan Achrenius on G+: microoptimized version of
|
|
|
|
* the simpler "(mask & ONEBYTES) * ONEBYTES >> 56"
|
|
|
|
* that works for the bytemasks without having to
|
|
|
|
* mask them first.
|
|
|
|
*/
|
|
|
|
static inline long count_masked_bytes(unsigned long mask)
|
|
|
|
{
|
|
|
|
return mask*0x0001020304050608ul >> 56;
|
|
|
|
}
|
|
|
|
|
|
|
|
#else /* 32-bit case */
|
|
|
|
|
|
|
|
/* Carl Chatfield / Jan Achrenius G+ version for 32-bit */
|
|
|
|
static inline long count_masked_bytes(long mask)
|
|
|
|
{
|
|
|
|
/* (000000 0000ff 00ffff ffffff) -> ( 1 1 2 3 ) */
|
|
|
|
long a = (0x0ff0001+mask) >> 23;
|
|
|
|
/* Fix the 1 for 00 case */
|
|
|
|
return a & mask;
|
|
|
|
}
|
|
|
|
|
|
|
|
#endif
|
|
|
|
|
|
|
|
/* Return nonzero if it has a zero */
|
|
|
|
static inline unsigned long has_zero(unsigned long a, unsigned long *bits, const struct word_at_a_time *c)
|
|
|
|
{
|
|
|
|
unsigned long mask = ((a - c->one_bits) & ~a) & c->high_bits;
|
|
|
|
*bits = mask;
|
|
|
|
return mask;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline unsigned long prep_zero_mask(unsigned long a, unsigned long bits, const struct word_at_a_time *c)
|
|
|
|
{
|
|
|
|
return bits;
|
|
|
|
}
|
|
|
|
|
|
|
|
static inline unsigned long create_zero_mask(unsigned long bits)
|
|
|
|
{
|
|
|
|
bits = (bits - 1) & ~bits;
|
|
|
|
return bits >> 7;
|
|
|
|
}
|
|
|
|
|
|
|
|
/* The mask we created is directly usable as a bytemask */
|
|
|
|
#define zero_bytemask(mask) (mask)
|
|
|
|
|
|
|
|
static inline unsigned long find_zero(unsigned long mask)
|
|
|
|
{
|
|
|
|
return count_masked_bytes(mask);
|
|
|
|
}
|
|
|
|
|
|
|
|
#endif /* __BIG_ENDIAN */
|
|
|
|
|
word-at-a-time: make the interfaces truly generic
This changes the interfaces in <asm/word-at-a-time.h> to be a bit more
complicated, but a lot more generic.
In particular, it allows us to really do the operations efficiently on
both little-endian and big-endian machines, pretty much regardless of
machine details. For example, if you can rely on a fast population
count instruction on your architecture, this will allow you to make your
optimized <asm/word-at-a-time.h> file with that.
NOTE! The "generic" version in include/asm-generic/word-at-a-time.h is
not truly generic, it actually only works on big-endian. Why? Because
on little-endian the generic algorithms are wasteful, since you can
inevitably do better. The x86 implementation is an example of that.
(The only truly non-generic part of the asm-generic implementation is
the "find_zero()" function, and you could make a little-endian version
of it. And if the Kbuild infrastructure allowed us to pick a particular
header file, that would be lovely)
The <asm/word-at-a-time.h> functions are as follows:
- WORD_AT_A_TIME_CONSTANTS: specific constants that the algorithm
uses.
- has_zero(): take a word, and determine if it has a zero byte in it.
It gets the word, the pointer to the constant pool, and a pointer to
an intermediate "data" field it can set.
This is the "quick-and-dirty" zero tester: it's what is run inside
the hot loops.
- "prep_zero_mask()": take the word, the data that has_zero() produced,
and the constant pool, and generate an *exact* mask of which byte had
the first zero. This is run directly *outside* the loop, and allows
the "has_zero()" function to answer the "is there a zero byte"
question without necessarily getting exactly *which* byte is the
first one to contain a zero.
If you do multiple byte lookups concurrently (eg "hash_name()", which
looks for both NUL and '/' bytes), after you've done the prep_zero_mask()
phase, the result of those can be or'ed together to get the "either
or" case.
- The result from "prep_zero_mask()" can then be fed into "find_zero()"
(to find the byte offset of the first byte that was zero) or into
"zero_bytemask()" (to find the bytemask of the bytes preceding the
zero byte).
The existence of zero_bytemask() is optional, and is not necessary
for the normal string routines. But dentry name hashing needs it, so
if you enable DENTRY_WORD_AT_A_TIME you need to expose it.
This changes the generic strncpy_from_user() function and the dentry
hashing functions to use these modified word-at-a-time interfaces. This
gets us back to the optimized state of the x86 strncpy that we lost in
the previous commit when moving over to the generic version.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2012-05-27 00:43:17 +07:00
|
|
|
#endif /* _ASM_WORD_AT_A_TIME_H */
|