core/ngx_cpuinfo源代码分析
本文主要介绍一下ngx_cpuinfo.
1. core/ngx_cpuinfo.c源文件
源文件内容如下：
/*
* Copyright (C) Igor Sysoev
* Copyright (C) Nginx, Inc.
*/
#include <ngx_config.h>
#include <ngx_core.h>
#if (( __i386__ || __amd64__ ) && ( __GNUC__ || __INTEL_COMPILER ))
static ngx_inline void ngx_cpuid(uint32_t i, uint32_t *buf);
#if ( __i386__ )
static ngx_inline void
ngx_cpuid(uint32_t i, uint32_t *buf)
{
/*
* we could not use %ebx as output parameter if gcc builds PIC,
* and we could not save %ebx on stack, because %esp is used,
* when the -fomit-frame-pointer optimization is specified.
*/
__asm__ (
" mov %%ebx, %%esi; "
" cpuid; "
" mov %%eax, (%1); "
" mov %%ebx, 4(%1); "
" mov %%edx, 8(%1); "
" mov %%ecx, 12(%1); "
" mov %%esi, %%ebx; "
: : "a" (i), "D" (buf) : "ecx", "edx", "esi", "memory" );
}
#else /* __amd64__ */
static ngx_inline void
ngx_cpuid(uint32_t i, uint32_t *buf)
{
uint32_t eax, ebx, ecx, edx;
__asm__ (
"cpuid"
: "=a" (eax), "=b" (ebx), "=c" (ecx), "=d" (edx) : "a" (i) );
buf[0] = eax;
buf[1] = ebx;
buf[2] = edx;
buf[3] = ecx;
}
#endif
/* auto detect the L2 cache line size of modern and widespread CPUs */
void
ngx_cpuinfo(void)
{
u_char *vendor;
uint32_t vbuf[5], cpu[4], model;
vbuf[0] = 0;
vbuf[1] = 0;
vbuf[2] = 0;
vbuf[3] = 0;
vbuf[4] = 0;
ngx_cpuid(0, vbuf);
vendor = (u_char *) &vbuf[1];
if (vbuf[0] == 0) {
return;
}
ngx_cpuid(1, cpu);
if (ngx_strcmp(vendor, "GenuineIntel") == 0) {
switch ((cpu[0] & 0xf00) >> 8) {
/* Pentium */
case 5:
ngx_cacheline_size = 32;
break;
/* Pentium Pro, II, III */
case 6:
ngx_cacheline_size = 32;
model = ((cpu[0] & 0xf0000) >> 8) | (cpu[0] & 0xf0);
if (model >= 0xd0) {
/* Intel Core, Core 2, Atom */
ngx_cacheline_size = 64;
}
break;
/*
* Pentium 4, although its cache line size is 64 bytes,
* it prefetches up to two cache lines during memory read
*/
case 15:
ngx_cacheline_size = 128;
break;
}
} else if (ngx_strcmp(vendor, "AuthenticAMD") == 0) {
ngx_cacheline_size = 64;
}
}
#else
void
ngx_cpuinfo(void)
{
}
#endif2. 相应宏测试
在具体分析源代码之前，我们先来对其中的一些宏定义进行测试(test4.c)：
#include <stdio.h>
#include <stdlib.h>
#ifndef _GNU_SOURCE
#define _GNU_SOURCE /* pread(), pwrite(), gethostname() */
#endif
int main(int argc,char *argv[])
{
#if (__i386__)
printf(" __i386__\n");
#endif
#if (__i386)
printf(" __i386\n");
#endif
#if (__amd64__)
printf(" __amd64__\n");
#endif
#if (__amd64)
printf(" __amd64\n");
#endif
#if (__GNUC__)
printf(" __GNUC__\n");
#endif
#if (__INTEL_COMPILER)
printf(" __INTEL_COMPILER\n");
#endif
return 0;
}编译运行：
root@ubuntu:~/test-src# gcc -o test4 test4.c root@ubuntu:~/test-src# ./test4 __i386__ __i386 __GNUC__
由上我们看到对于我们当前操作系统环境，定义了如下宏：
-
i386
-
__i386
-
GNUC
3. 测试ngx_cpuid()函数
通过上面相应宏测试我们再对相应的ngx_cpuid()函数进行测试：
#include <stdio.h>
#include <stdlib.h>
#include <inttypes.h>
#ifndef _GNU_SOURCE
#define _GNU_SOURCE /* pread(), pwrite(), gethostname() */
#endif
#ifndef ngx_inline
#define ngx_inline inline
#endif
static ngx_inline void ngx_cpuid(uint32_t i, uint32_t *buf)
{
/*
* we could not use %ebx as output parameter if gcc builds PIC,
* and we could not save %ebx on stack, because %esp is used,
* when the -fomit-frame-pointer optimization is specified.
*/
__asm__ (
" mov %%ebx, %%esi; "
" cpuid; "
" mov %%eax, (%1); "
" mov %%ebx, 4(%1); "
" mov %%edx, 8(%1); "
" mov %%ecx, 12(%1); "
" mov %%esi, %%ebx; "
: : "a" (i), "D" (buf) : "ecx", "edx", "esi", "memory" );
}
int main(int argc,char *argv[])
{
u_char *vendor;
uint32_t vbuf[5], cpu[4], model;
vbuf[0] = 0;
vbuf[1] = 0;
vbuf[2] = 0;
vbuf[3] = 0;
vbuf[4] = 0;
ngx_cpuid(0, vbuf);
vendor = (u_char *) &vbuf[1];
if (vbuf[0] == 0) {
return -1;
}
ngx_cpuid(1, cpu);
printf("vbuf[0]: %u\n",vbuf[0]);
printf("vendor: %s\n",vendor);
printf("cpu[0]: %u\n",cpu[0]);
return 0;
}编译运行：
root@ubuntu:~/test-src# gcc -o test5 test5.c root@ubuntu:~/test-src# ./test5 vbuf[0]: 13 vendor: AuthenticAMD cpu[0]: 6360833
3. ngx_cpuinfo()函数
/* auto detect the L2 cache line size of modern and widespread CPUs */
void
ngx_cpuinfo(void)
{
u_char *vendor;
uint32_t vbuf[5], cpu[4], model;
vbuf[0] = 0;
vbuf[1] = 0;
vbuf[2] = 0;
vbuf[3] = 0;
vbuf[4] = 0;
ngx_cpuid(0, vbuf);
vendor = (u_char *) &vbuf[1];
if (vbuf[0] == 0) {
return;
}
ngx_cpuid(1, cpu);
if (ngx_strcmp(vendor, "GenuineIntel") == 0) {
switch ((cpu[0] & 0xf00) >> 8) {
/* Pentium */
case 5:
ngx_cacheline_size = 32;
break;
/* Pentium Pro, II, III */
case 6:
ngx_cacheline_size = 32;
model = ((cpu[0] & 0xf0000) >> 8) | (cpu[0] & 0xf0);
if (model >= 0xd0) {
/* Intel Core, Core 2, Atom */
ngx_cacheline_size = 64;
}
break;
/*
* Pentium 4, although its cache line size is 64 bytes,
* it prefetches up to two cache lines during memory read
*/
case 15:
ngx_cacheline_size = 128;
break;
}
} else if (ngx_strcmp(vendor, "AuthenticAMD") == 0) {
ngx_cacheline_size = 64;
}
}由上我们可知，vendor为AuthenticAMD, 最后求得ngx_cacheline_size为64。
[参看]:
