AuroraMiddleware/glibc
1
0
Fork 0
mirror of https://sourceware.org/git/glibc.git synced 2024-11-22 13:00:06 +00:00
Code Issues Packages Projects Releases Wiki Activity

Properly count number of logical processors on Intel CPUs.

Browse Source 
The meaning of the 25-14 bits in EAX returned from cpuid with EAX = 4
has been changed from "the maximum number of threads sharing the cache"
to "the maximum number of addressable IDs for logical processors sharing
the cache" if cpuid takes EAX = 11.  We need to use results from both
EAX = 4 and EAX = 11 to get the number of threads sharing the cache.

The 25-14 bits in EAX on Core i7 is 15 although the number of logical
processors is 8.  Here is a white paper on this:

http://software.intel.com/en-us/articles/intel-64-architecture-processor-topology-enumeration/

This patch correctly counts number of logical processors on Intel CPUs
with EAX = 11 support on cpuid.  Tested on Dinnington, Core i7 and
Nehalem EX/EP.

It also fixed Pentium Ds workaround since EBX may not have the right
value returned from cpuid with EAX = 1.
This commit is contained in:
H.J. Lu 2009-08-07 09:39:36 -07:00 committed by Ulrich Drepper
parent 77c84aeb81
commit a546baa9cd
2 changed files with 43 additions and 4 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

5
ChangeLog
View File

@ -1,3 +1,8 @@
2009-08-05 H.J. Lu <hongjiu.lu@intel.com>
* sysdeps/x86_64/cacheinfo.c (init_cacheinfo): Properly use
EBX from EAX = 1. Handle EAX = 11.
2009-08-07 Andreas Schwab <schwab@redhat.com>
* Makefile (TAGS): Use separate sed -e expressions to avoid \

42
sysdeps/x86_64/cacheinfo.c
View File

@ -516,13 +516,15 @@ init_cacheinfo (void)
shared = handle_intel (_SC_LEVEL2_CACHE_SIZE, max_cpuid);
}
unsigned int ebx_1;
#ifdef USE_MULTIARCH
eax = __cpu_features.cpuid[COMMON_CPUID_INDEX_1].eax;
ebx = __cpu_features.cpuid[COMMON_CPUID_INDEX_1].ebx;
ebx_1 = __cpu_features.cpuid[COMMON_CPUID_INDEX_1].ebx;
ecx = __cpu_features.cpuid[COMMON_CPUID_INDEX_1].ecx;
edx = __cpu_features.cpuid[COMMON_CPUID_INDEX_1].edx;
#else
__cpuid (1, eax, ebx, ecx, edx);
__cpuid (1, eax, ebx_1, ecx, edx);
#endif
#ifndef DISABLE_PREFERRED_MEMORY_INSTRUCTION
@ -554,14 +556,46 @@ init_cacheinfo (void)
}
while (((eax >> 5) & 0x7) != level);
threads = ((eax >> 14) & 0x3ff) + 1;
threads = (eax >> 14) & 0x3ff;
/* If max_cpuid >= 11, THREADS is the maximum number of
addressable IDs for logical processors sharing the
cache, instead of the maximum number of threads
sharing the cache. */
if (threads && max_cpuid >= 11)
{
/* Find the number of logical processors shipped in
one core and apply count mask. */
i = 0;
while (1)
{
__cpuid_count (11, i++, eax, ebx, ecx, edx);
int shipped = ebx & 0xff;
int type = ecx & 0xff0;
if (shipped == 0 || type == 0)
break;
else if (type == 0x200)
{
int count_mask;
/* Compute count mask. */
asm ("bsr %1, %0"
: "=r" (count_mask) : "g" (threads));
count_mask = ~(-1 << (count_mask + 1));
threads = (shipped - 1) & count_mask;
break;
}
}
}
threads += 1;
}
else
{
intel_bug_no_cache_info:
/* Assume that all logical threads share the highest cache level. */
threads = (ebx >> 16) & 0xff;
threads = (ebx_1 >> 16) & 0xff;
}
/* Cap usage of highest cache level to the number of supported