用于检测 APIC id 的代码为不同的逻辑处理器返回相同的 id

发布于 2024-08-04 11:50:18 字数 1113 浏览 10 评论 0原文

我在基于 Intel Core2 的 Win2k3 机器上运行 NT 服务，我需要迭代所有逻辑 CPU（进程关联中的所有位）。为此，我调用 GetProcessAffinityMask() 来检索系统关联掩码，然后依次将进程切换到每个处理器：

 DWORD systemMask;
 GetProcessAffinityMask( ... &systemMask );
 DWORD processorId = 1;
 while( systemMask != 0 ) {
    SetProcessAffinityMask(... processorId );
    Sleep( 1 ); // to be sure that it shifts to that processor
    systemMask >>= 1;
    processorId <<= 1;
 }

在每次迭代中，我调用此处的代码用于检索当前处理器 APIC id。问题是，对于不同的处理器，它有时会返回相同的 APIC id。根据文档，系统中的每个处理器必须具有相同的 ID。

我尝试对此进行调试 - 检查 Windows 是否确实更改了亲和性：

 while( systemMask != 0 ) {
    SetProcessAffinityMask(... processorId );
    Sleep( 1 ); // to be sure that it shifts to that processor
    DWORD tempAffinity;
    GetProcessAffinityMask( ... &tempAffinity );
    // run APIC id detection code here
    systemMask >>= 1;
    processorId <<= 1;
 }

它返回的亲和性掩码完全符合我的预期，但对于不同的处理器，APIC id 仍然可以相同。

对于这种奇怪的情况有解释吗？

原文

I run my NT service on an Intel Core2 based Win2k3 machine where I need to iterate through all logical CPUs (all bits in process affinity). To do so I call GetProcessAffinityMask() to retrieve the system affinity mask and then switch the process to each processor in turn:

 DWORD systemMask;
 GetProcessAffinityMask( ... &systemMask );
 DWORD processorId = 1;
 while( systemMask != 0 ) {
    SetProcessAffinityMask(... processorId );
    Sleep( 1 ); // to be sure that it shifts to that processor
    systemMask >>= 1;
    processorId <<= 1;
 }

On each iteration I invoke code from here to retrieve the current processor APIC id. The problem is that for different processors it sometimes returns identical APIC ids. According to documentation each processor in the system must have an identical id.

I tried debugging this - checked whether Windows actually changes the affinity:

 while( systemMask != 0 ) {
    SetProcessAffinityMask(... processorId );
    Sleep( 1 ); // to be sure that it shifts to that processor
    DWORD tempAffinity;
    GetProcessAffinityMask( ... &tempAffinity );
    // run APIC id detection code here
    systemMask >>= 1;
    processorId <<= 1;
 }

It returns exactly the affinity mask I expect but APIC ids can still be the same for different processors.

Is there an explanation to this weird situation?

分享到QQ

分享到微博

如果你对这篇内容有疑问，欢迎到本站社区发帖提问参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。

发布评论

需要登录才能够评论，你可以免费注册一个本站的账号。

压抑⊿情绪 2024-08-11 11:50:18

您无法使用 Windows API 调用来确定这一点。

首先，在使用 EAX = 0x1 调用 CPUID 后，Intel CPU 上的 APCID（不确定 AMD 或其他 CPU）最初被编码在 EBX 寄存器（位 24 - 31）又名 EBX[31:24] 的 8 位部分内。

考虑以下带有内联汇编的 C/C++ 代码：

unsigned int cpu_eax;
unsigned int cpu_ebx;
unsigned int cpu_ecx;
unsigned int cpu_edx;
unsigned int apic_id;

__asm
{
  mov    eax, 0x1
  cpuid
  mov     [cpu_eax], eax
  mov     [cpu_ebx], ebx
  mov     [cpu_ecx], ecx
  mov     [cpu_edx], edx
}

apic_id = ( cpu_ebx & 0xFF000000) >> 24;

GetProcessAffinityMask 只是枚举所有可用核心、CPU 内的核心以及每个 CPU 具有超线程功能的核心，并返回这些组合的总数。它没有物理 CPU 与 CPU 内的内核与支持超线程的内核的概念。至少它没有这样报道。

You can not use the Windows API call to determine this.

First, the APCID on Intel CPUs (not sure about AMD or others) is encoded initially within an 8 bit section of the EBX register (bit 24 - 31) aka EBX[31:24] after a call the CPUID with EAX = 0x1.

Consider the following C/C++ code with in-line assembly:

unsigned int cpu_eax;
unsigned int cpu_ebx;
unsigned int cpu_ecx;
unsigned int cpu_edx;
unsigned int apic_id;

__asm
{
  mov    eax, 0x1
  cpuid
  mov     [cpu_eax], eax
  mov     [cpu_ebx], ebx
  mov     [cpu_ecx], ecx
  mov     [cpu_edx], edx
}

apic_id = ( cpu_ebx & 0xFF000000) >> 24;

GetProcessAffinityMask simply enumerates all of the available cores, cores within a CPU, and hyper-thread capable cores per CPU and returns the total of these combinations. It has no concept of physical CPU vs. cores within a CPU vs. hyper thread capable cores. At least it does not report it as such.

回复收藏 0 原文