mirror of
https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git
synced 2025-01-07 14:32:23 +00:00
f949cb7590
The content of zh_TW was too outdated comparing to the original files.
Consequently carry out improvements in order to both keep track of sources
and fix several grammatical mistakes in traditional Chinese.
This is a thorough rewrite of the previous patch:
https://lore.kernel.org/linux-doc/20230807120006.6361-1-src.res.211@gmail.com/
in order to get rid of text damage and merging errors, created based on
linux-next (date: Oct. 9, 2023).
Signed-off-by: Hu Haowen <src.res.211@gmail.com>
Signed-off-by: Jonathan Corbet <corbet@lwn.net>
Link: https://lore.kernel.org/r/20231011051212.17580-1-src.res.211@gmail.com
98 lines
3.7 KiB
ReStructuredText
98 lines
3.7 KiB
ReStructuredText
.. SPDX-License-Identifier: GPL-2.0
|
||
.. include:: ../disclaimer-zh_TW.rst
|
||
|
||
:Original: Documentation/admin-guide/cputopology.rst
|
||
|
||
:翻譯:
|
||
|
||
唐藝舟 Tang Yizhou <tangyeechou@gmail.com>
|
||
|
||
==========================
|
||
如何通過sysfs將CPU拓撲導出
|
||
==========================
|
||
|
||
CPU拓撲信息通過sysfs導出。顯示的項（屬性）和某些架構的/proc/cpuinfo輸出相似。它們位於
|
||
/sys/devices/system/cpu/cpuX/topology/。請閱讀ABI文件：
|
||
Documentation/ABI/stable/sysfs-devices-system-cpu。
|
||
|
||
drivers/base/topology.c是體系結構中性的，它導出了這些屬性。然而，die、cluster、book、
|
||
draw這些層次結構相關的文件僅在體系結構提供了下文描述的宏的條件下被創建。
|
||
|
||
對於支持這個特性的體系結構，它必須在include/asm-XXX/topology.h中定義這些宏中的一部分::
|
||
|
||
#define topology_physical_package_id(cpu)
|
||
#define topology_die_id(cpu)
|
||
#define topology_cluster_id(cpu)
|
||
#define topology_core_id(cpu)
|
||
#define topology_book_id(cpu)
|
||
#define topology_drawer_id(cpu)
|
||
#define topology_sibling_cpumask(cpu)
|
||
#define topology_core_cpumask(cpu)
|
||
#define topology_cluster_cpumask(cpu)
|
||
#define topology_die_cpumask(cpu)
|
||
#define topology_book_cpumask(cpu)
|
||
#define topology_drawer_cpumask(cpu)
|
||
|
||
``**_id macros`` 的類型是int。
|
||
``**_cpumask macros`` 的類型是 ``(const) struct cpumask *`` 。後者和恰當的
|
||
``**_siblings`` sysfs屬性對應（除了topology_sibling_cpumask()，它和thread_siblings
|
||
對應）。
|
||
|
||
爲了在所有體系結構上保持一致，include/linux/topology.h提供了上述所有宏的默認定義，以防
|
||
它們未在include/asm-XXX/topology.h中定義:
|
||
|
||
1) topology_physical_package_id: -1
|
||
2) topology_die_id: -1
|
||
3) topology_cluster_id: -1
|
||
4) topology_core_id: 0
|
||
5) topology_book_id: -1
|
||
6) topology_drawer_id: -1
|
||
7) topology_sibling_cpumask: 僅入參CPU
|
||
8) topology_core_cpumask: 僅入參CPU
|
||
9) topology_cluster_cpumask: 僅入參CPU
|
||
10) topology_die_cpumask: 僅入參CPU
|
||
11) topology_book_cpumask: 僅入參CPU
|
||
12) topology_drawer_cpumask: 僅入參CPU
|
||
|
||
此外，CPU拓撲信息由/sys/devices/system/cpu提供，包含下述文件。輸出對應的內部數據源放在
|
||
方括號（"[]"）中。
|
||
|
||
=========== ==================================================================
|
||
kernel_max: 內核配置允許的最大CPU下標值。[NR_CPUS-1]
|
||
|
||
offline: 由於熱插拔移除或者超過內核允許的CPU上限（上文描述的kernel_max）
|
||
導致未上線的CPU。[~cpu_online_mask + cpus >= NR_CPUS]
|
||
|
||
online: 在線的CPU，可供調度使用。[cpu_online_mask]
|
||
|
||
possible: 已被分配資源的CPU，如果它們CPU實際存在，可以上線。
|
||
[cpu_possible_mask]
|
||
|
||
present: 被系統識別實際存在的CPU。[cpu_present_mask]
|
||
=========== ==================================================================
|
||
|
||
上述輸出的格式和cpulist_parse()兼容[參見 <linux/cpumask.h>]。下面給些例子。
|
||
|
||
在本例中，系統中有64個CPU，但是CPU 32-63超過了kernel_max值，因爲NR_CPUS配置項是32，
|
||
取值範圍被限制爲0..31。此外注意CPU2和4-31未上線，但是可以上線，因爲它們同時存在於
|
||
present和possible::
|
||
|
||
kernel_max: 31
|
||
offline: 2,4-31,32-63
|
||
online: 0-1,3
|
||
possible: 0-31
|
||
present: 0-31
|
||
|
||
在本例中，NR_CPUS配置項是128，但內核啓動時設置possible_cpus=144。系統中有4個CPU，
|
||
CPU2被手動設置下線（也是唯一一個可以上線的CPU）::
|
||
|
||
kernel_max: 127
|
||
offline: 2,4-127,128-143
|
||
online: 0-1,3
|
||
possible: 0-127
|
||
present: 0-3
|
||
|
||
閱讀Documentation/core-api/cpu_hotplug.rst可瞭解開機參數possible_cpus=NUM，同時還
|
||
可以瞭解各種cpumask的信息。
|
||
|