1
0
mirror of https://git.FreeBSD.org/src.git synced 2024-10-18 02:19:39 +00:00

zcommon: add specialized versions of cityhash4

Specializing cityhash4 on 32-bit architectures can reduce the size
of stack frames as well as instruction count. This is a tiny but
useful optimization, since some callers invoke it frequently.

When specializing into 1/2/3/4-arg versions, the stack usage
(in bytes) on some 32-bit arches are listed as follows:

- x86: 32, 32, 32, 40
- arm-v7a: 20, 20, 28, 36
- riscv: 0, 0, 0, 16
- power: 16, 16, 16, 32
- mipsel: 8, 8, 8, 24

And each actual argument (even if passing 0) contributes evenly
to the number of multiplication instructions generated:

- x86: 9, 12, 15 ,18
- arm-v7a: 6, 8, 10, 12
- riscv / power: 12, 18, 20, 24
- mipsel: 9, 12, 15, 19

On 64-bit architectures, the tendencies are similar. But both stack
sizes and instruction counts are significantly smaller thus negligible.

Reviewed by: Brian Behlendorf <behlendorf1@llnl.gov>
Reviewed-by: Alexander Motin <mav@FreeBSD.org>
Reviewed-by: Tino Reichardt <milky-zfs@mcmilk.de>
Signed-off-by: Shengqi Chen <harry-chen@outlook.com>
Closes #16131
Closes #16483
This commit is contained in:
Shengqi Chen 2024-09-07 21:55:03 +08:00 committed by Brian Behlendorf
parent 1c35206124
commit 0ae4460c61
3 changed files with 56 additions and 2 deletions

View File

@ -32,6 +32,13 @@
extern "C" {
#endif
/*
* Define 1/2/3-argument specialized versions of cityhash4, which can reduce
* instruction count (especially multiplication) on some 32-bit arches.
*/
_SYS_CITYHASH_H uint64_t cityhash1(uint64_t);
_SYS_CITYHASH_H uint64_t cityhash2(uint64_t, uint64_t);
_SYS_CITYHASH_H uint64_t cityhash3(uint64_t, uint64_t, uint64_t);
_SYS_CITYHASH_H uint64_t cityhash4(uint64_t, uint64_t, uint64_t, uint64_t);
#ifdef __cplusplus

View File

@ -154,6 +154,9 @@
<elf-symbol name='avl_update_lt' type='func-type' binding='global-binding' visibility='default-visibility' is-defined='yes'/>
<elf-symbol name='avl_walk' type='func-type' binding='global-binding' visibility='default-visibility' is-defined='yes'/>
<elf-symbol name='bookmark_namecheck' type='func-type' binding='global-binding' visibility='default-visibility' is-defined='yes'/>
<elf-symbol name='cityhash1' type='func-type' binding='global-binding' visibility='default-visibility' is-defined='yes'/>
<elf-symbol name='cityhash2' type='func-type' binding='global-binding' visibility='default-visibility' is-defined='yes'/>
<elf-symbol name='cityhash3' type='func-type' binding='global-binding' visibility='default-visibility' is-defined='yes'/>
<elf-symbol name='cityhash4' type='func-type' binding='global-binding' visibility='default-visibility' is-defined='yes'/>
<elf-symbol name='color_end' type='func-type' binding='global-binding' visibility='default-visibility' is-defined='yes'/>
<elf-symbol name='color_start' type='func-type' binding='global-binding' visibility='default-visibility' is-defined='yes'/>
@ -9241,6 +9244,21 @@
</function-decl>
</abi-instr>
<abi-instr address-size='64' path='module/zcommon/cityhash.c' language='LANG_C99'>
<function-decl name='cityhash1' mangled-name='cityhash1' visibility='default' binding='global' size-in-bits='64' elf-symbol-id='cityhash1'>
<parameter type-id='9c313c2d' name='w'/>
<return type-id='9c313c2d'/>
</function-decl>
<function-decl name='cityhash2' mangled-name='cityhash2' visibility='default' binding='global' size-in-bits='64' elf-symbol-id='cityhash2'>
<parameter type-id='9c313c2d' name='w1'/>
<parameter type-id='9c313c2d' name='w2'/>
<return type-id='9c313c2d'/>
</function-decl>
<function-decl name='cityhash3' mangled-name='cityhash3' visibility='default' binding='global' size-in-bits='64' elf-symbol-id='cityhash3'>
<parameter type-id='9c313c2d' name='w1'/>
<parameter type-id='9c313c2d' name='w2'/>
<parameter type-id='9c313c2d' name='w3'/>
<return type-id='9c313c2d'/>
</function-decl>
<function-decl name='cityhash4' mangled-name='cityhash4' visibility='default' binding='global' size-in-bits='64' elf-symbol-id='cityhash4'>
<parameter type-id='9c313c2d' name='w1'/>
<parameter type-id='9c313c2d' name='w2'/>

View File

@ -49,8 +49,8 @@ cityhash_helper(uint64_t u, uint64_t v, uint64_t mul)
return (b);
}
uint64_t
cityhash4(uint64_t w1, uint64_t w2, uint64_t w3, uint64_t w4)
static inline uint64_t
cityhash_impl(uint64_t w1, uint64_t w2, uint64_t w3, uint64_t w4)
{
uint64_t mul = HASH_K2 + 64;
uint64_t a = w1 * HASH_K1;
@ -59,9 +59,38 @@ cityhash4(uint64_t w1, uint64_t w2, uint64_t w3, uint64_t w4)
uint64_t d = w3 * HASH_K2;
return (cityhash_helper(rotate(a + b, 43) + rotate(c, 30) + d,
a + rotate(b + HASH_K2, 18) + c, mul));
}
/*
* Passing w as the 2nd argument could save one 64-bit multiplication.
*/
uint64_t
cityhash1(uint64_t w)
{
return (cityhash_impl(0, w, 0, 0));
}
uint64_t
cityhash2(uint64_t w1, uint64_t w2)
{
return (cityhash_impl(w1, w2, 0, 0));
}
uint64_t
cityhash3(uint64_t w1, uint64_t w2, uint64_t w3)
{
return (cityhash_impl(w1, w2, w3, 0));
}
uint64_t
cityhash4(uint64_t w1, uint64_t w2, uint64_t w3, uint64_t w4)
{
return (cityhash_impl(w1, w2, w3, w4));
}
#if defined(_KERNEL)
EXPORT_SYMBOL(cityhash1);
EXPORT_SYMBOL(cityhash2);
EXPORT_SYMBOL(cityhash3);
EXPORT_SYMBOL(cityhash4);
#endif