2017-05-14 06:52:01 -06:00
|
|
|
===========================================
|
|
|
|
How CPU topology info is exported via sysfs
|
|
|
|
===========================================
|
[PATCH] Export cpu topology in sysfs
The patch implements cpu topology exportation by sysfs.
Items (attributes) are similar to /proc/cpuinfo.
1) /sys/devices/system/cpu/cpuX/topology/physical_package_id:
represent the physical package id of cpu X;
2) /sys/devices/system/cpu/cpuX/topology/core_id:
represent the cpu core id to cpu X;
3) /sys/devices/system/cpu/cpuX/topology/thread_siblings:
represent the thread siblings to cpu X in the same core;
4) /sys/devices/system/cpu/cpuX/topology/core_siblings:
represent the thread siblings to cpu X in the same physical package;
To implement it in an architecture-neutral way, a new source file,
driver/base/topology.c, is to export the 5 attributes.
If one architecture wants to support this feature, it just needs to
implement 4 defines, typically in file include/asm-XXX/topology.h.
The 4 defines are:
#define topology_physical_package_id(cpu)
#define topology_core_id(cpu)
#define topology_thread_siblings(cpu)
#define topology_core_siblings(cpu)
The type of **_id is int.
The type of siblings is cpumask_t.
To be consistent on all architectures, the 4 attributes should have
deafult values if their values are unavailable. Below is the rule.
1) physical_package_id: If cpu has no physical package id, -1 is the
default value.
2) core_id: If cpu doesn't support multi-core, its core id is 0.
3) thread_siblings: Just include itself, if the cpu doesn't support
HT/multi-thread.
4) core_siblings: Just include itself, if the cpu doesn't support
multi-core and HT/Multi-thread.
So be careful when declaring the 4 defines in include/asm-XXX/topology.h.
If an attribute isn't defined on an architecture, it won't be exported.
Thank Nathan, Greg, Andi, Paul and Venki.
The patch provides defines for i386/x86_64/ia64.
Signed-off-by: Zhang, Yanmin <yanmin.zhang@intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-03 04:04:36 -07:00
|
|
|
|
2009-10-21 21:45:31 -06:00
|
|
|
Export CPU topology info via sysfs. Items (attributes) are similar
|
2019-02-25 23:20:00 -07:00
|
|
|
to /proc/cpuinfo output of some architectures. They reside in
|
|
|
|
/sys/devices/system/cpu/cpuX/topology/:
|
[PATCH] Export cpu topology in sysfs
The patch implements cpu topology exportation by sysfs.
Items (attributes) are similar to /proc/cpuinfo.
1) /sys/devices/system/cpu/cpuX/topology/physical_package_id:
represent the physical package id of cpu X;
2) /sys/devices/system/cpu/cpuX/topology/core_id:
represent the cpu core id to cpu X;
3) /sys/devices/system/cpu/cpuX/topology/thread_siblings:
represent the thread siblings to cpu X in the same core;
4) /sys/devices/system/cpu/cpuX/topology/core_siblings:
represent the thread siblings to cpu X in the same physical package;
To implement it in an architecture-neutral way, a new source file,
driver/base/topology.c, is to export the 5 attributes.
If one architecture wants to support this feature, it just needs to
implement 4 defines, typically in file include/asm-XXX/topology.h.
The 4 defines are:
#define topology_physical_package_id(cpu)
#define topology_core_id(cpu)
#define topology_thread_siblings(cpu)
#define topology_core_siblings(cpu)
The type of **_id is int.
The type of siblings is cpumask_t.
To be consistent on all architectures, the 4 attributes should have
deafult values if their values are unavailable. Below is the rule.
1) physical_package_id: If cpu has no physical package id, -1 is the
default value.
2) core_id: If cpu doesn't support multi-core, its core id is 0.
3) thread_siblings: Just include itself, if the cpu doesn't support
HT/multi-thread.
4) core_siblings: Just include itself, if the cpu doesn't support
multi-core and HT/Multi-thread.
So be careful when declaring the 4 defines in include/asm-XXX/topology.h.
If an attribute isn't defined on an architecture, it won't be exported.
Thank Nathan, Greg, Andi, Paul and Venki.
The patch provides defines for i386/x86_64/ia64.
Signed-off-by: Zhang, Yanmin <yanmin.zhang@intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-03 04:04:36 -07:00
|
|
|
|
2019-02-25 23:20:00 -07:00
|
|
|
physical_package_id:
|
2009-10-21 21:45:31 -06:00
|
|
|
|
|
|
|
physical package id of cpuX. Typically corresponds to a physical
|
|
|
|
socket number, but the actual value is architecture and platform
|
|
|
|
dependent.
|
|
|
|
|
2019-02-25 23:20:00 -07:00
|
|
|
core_id:
|
2009-10-21 21:45:31 -06:00
|
|
|
|
|
|
|
the CPU core ID of cpuX. Typically it is the hardware platform's
|
|
|
|
identifier (rather than the kernel's). The actual value is
|
|
|
|
architecture and platform dependent.
|
|
|
|
|
2019-02-25 23:20:00 -07:00
|
|
|
book_id:
|
2010-08-31 02:28:17 -06:00
|
|
|
|
|
|
|
the book ID of cpuX. Typically it is the hardware platform's
|
|
|
|
identifier (rather than the kernel's). The actual value is
|
|
|
|
architecture and platform dependent.
|
|
|
|
|
2019-02-25 23:20:00 -07:00
|
|
|
drawer_id:
|
2016-05-21 03:10:13 -06:00
|
|
|
|
|
|
|
the drawer ID of cpuX. Typically it is the hardware platform's
|
|
|
|
identifier (rather than the kernel's). The actual value is
|
|
|
|
architecture and platform dependent.
|
|
|
|
|
2019-02-25 23:20:00 -07:00
|
|
|
thread_siblings:
|
2009-10-21 21:45:31 -06:00
|
|
|
|
2013-09-04 04:58:29 -06:00
|
|
|
internal kernel map of cpuX's hardware threads within the same
|
2015-05-26 07:11:29 -06:00
|
|
|
core as cpuX.
|
2009-10-21 21:45:31 -06:00
|
|
|
|
2019-02-25 23:20:00 -07:00
|
|
|
thread_siblings_list:
|
2015-05-26 07:11:29 -06:00
|
|
|
|
|
|
|
human-readable list of cpuX's hardware threads within the same
|
|
|
|
core as cpuX.
|
|
|
|
|
2019-02-25 23:20:00 -07:00
|
|
|
core_siblings:
|
2009-10-21 21:45:31 -06:00
|
|
|
|
|
|
|
internal kernel map of cpuX's hardware threads within the same
|
|
|
|
physical_package_id.
|
[PATCH] Export cpu topology in sysfs
The patch implements cpu topology exportation by sysfs.
Items (attributes) are similar to /proc/cpuinfo.
1) /sys/devices/system/cpu/cpuX/topology/physical_package_id:
represent the physical package id of cpu X;
2) /sys/devices/system/cpu/cpuX/topology/core_id:
represent the cpu core id to cpu X;
3) /sys/devices/system/cpu/cpuX/topology/thread_siblings:
represent the thread siblings to cpu X in the same core;
4) /sys/devices/system/cpu/cpuX/topology/core_siblings:
represent the thread siblings to cpu X in the same physical package;
To implement it in an architecture-neutral way, a new source file,
driver/base/topology.c, is to export the 5 attributes.
If one architecture wants to support this feature, it just needs to
implement 4 defines, typically in file include/asm-XXX/topology.h.
The 4 defines are:
#define topology_physical_package_id(cpu)
#define topology_core_id(cpu)
#define topology_thread_siblings(cpu)
#define topology_core_siblings(cpu)
The type of **_id is int.
The type of siblings is cpumask_t.
To be consistent on all architectures, the 4 attributes should have
deafult values if their values are unavailable. Below is the rule.
1) physical_package_id: If cpu has no physical package id, -1 is the
default value.
2) core_id: If cpu doesn't support multi-core, its core id is 0.
3) thread_siblings: Just include itself, if the cpu doesn't support
HT/multi-thread.
4) core_siblings: Just include itself, if the cpu doesn't support
multi-core and HT/Multi-thread.
So be careful when declaring the 4 defines in include/asm-XXX/topology.h.
If an attribute isn't defined on an architecture, it won't be exported.
Thank Nathan, Greg, Andi, Paul and Venki.
The patch provides defines for i386/x86_64/ia64.
Signed-off-by: Zhang, Yanmin <yanmin.zhang@intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-03 04:04:36 -07:00
|
|
|
|
2019-02-25 23:20:00 -07:00
|
|
|
core_siblings_list:
|
2015-05-26 07:11:29 -06:00
|
|
|
|
|
|
|
human-readable list of cpuX's hardware threads within the same
|
|
|
|
physical_package_id.
|
|
|
|
|
2019-02-25 23:20:00 -07:00
|
|
|
book_siblings:
|
2010-08-31 02:28:17 -06:00
|
|
|
|
|
|
|
internal kernel map of cpuX's hardware threads within the same
|
|
|
|
book_id.
|
|
|
|
|
2019-02-25 23:20:00 -07:00
|
|
|
book_siblings_list:
|
2015-05-26 07:11:29 -06:00
|
|
|
|
|
|
|
human-readable list of cpuX's hardware threads within the same
|
|
|
|
book_id.
|
|
|
|
|
2019-02-25 23:20:00 -07:00
|
|
|
drawer_siblings:
|
2016-05-21 03:10:13 -06:00
|
|
|
|
|
|
|
internal kernel map of cpuX's hardware threads within the same
|
|
|
|
drawer_id.
|
|
|
|
|
2019-02-25 23:20:00 -07:00
|
|
|
drawer_siblings_list:
|
2016-05-21 03:10:13 -06:00
|
|
|
|
|
|
|
human-readable list of cpuX's hardware threads within the same
|
|
|
|
drawer_id.
|
|
|
|
|
2019-02-25 23:20:00 -07:00
|
|
|
Architecture-neutral, drivers/base/topology.c, exports these attributes.
|
|
|
|
However, the book and drawer related sysfs files will only be created if
|
|
|
|
CONFIG_SCHED_BOOK and CONFIG_SCHED_DRAWER are selected, respectively.
|
2016-05-21 03:10:13 -06:00
|
|
|
|
2019-02-25 23:20:00 -07:00
|
|
|
CONFIG_SCHED_BOOK and CONFIG_SCHED_DRAWER are currently only used on s390,
|
|
|
|
where they reflect the cpu and cache hierarchy.
|
[PATCH] Export cpu topology in sysfs
The patch implements cpu topology exportation by sysfs.
Items (attributes) are similar to /proc/cpuinfo.
1) /sys/devices/system/cpu/cpuX/topology/physical_package_id:
represent the physical package id of cpu X;
2) /sys/devices/system/cpu/cpuX/topology/core_id:
represent the cpu core id to cpu X;
3) /sys/devices/system/cpu/cpuX/topology/thread_siblings:
represent the thread siblings to cpu X in the same core;
4) /sys/devices/system/cpu/cpuX/topology/core_siblings:
represent the thread siblings to cpu X in the same physical package;
To implement it in an architecture-neutral way, a new source file,
driver/base/topology.c, is to export the 5 attributes.
If one architecture wants to support this feature, it just needs to
implement 4 defines, typically in file include/asm-XXX/topology.h.
The 4 defines are:
#define topology_physical_package_id(cpu)
#define topology_core_id(cpu)
#define topology_thread_siblings(cpu)
#define topology_core_siblings(cpu)
The type of **_id is int.
The type of siblings is cpumask_t.
To be consistent on all architectures, the 4 attributes should have
deafult values if their values are unavailable. Below is the rule.
1) physical_package_id: If cpu has no physical package id, -1 is the
default value.
2) core_id: If cpu doesn't support multi-core, its core id is 0.
3) thread_siblings: Just include itself, if the cpu doesn't support
HT/multi-thread.
4) core_siblings: Just include itself, if the cpu doesn't support
multi-core and HT/Multi-thread.
So be careful when declaring the 4 defines in include/asm-XXX/topology.h.
If an attribute isn't defined on an architecture, it won't be exported.
Thank Nathan, Greg, Andi, Paul and Venki.
The patch provides defines for i386/x86_64/ia64.
Signed-off-by: Zhang, Yanmin <yanmin.zhang@intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-03 04:04:36 -07:00
|
|
|
|
2008-06-04 22:47:29 -06:00
|
|
|
For an architecture to support this feature, it must define some of
|
2017-05-14 06:52:01 -06:00
|
|
|
these macros in include/asm-XXX/topology.h::
|
|
|
|
|
|
|
|
#define topology_physical_package_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_book_cpumask(cpu)
|
|
|
|
#define topology_drawer_cpumask(cpu)
|
|
|
|
|
|
|
|
The type of ``**_id macros`` is int.
|
|
|
|
The type of ``**_cpumask macros`` is ``(const) struct cpumask *``. The latter
|
|
|
|
correspond with appropriate ``**_siblings`` sysfs attributes (except for
|
2015-05-26 07:11:29 -06:00
|
|
|
topology_sibling_cpumask() which corresponds with thread_siblings).
|
[PATCH] Export cpu topology in sysfs
The patch implements cpu topology exportation by sysfs.
Items (attributes) are similar to /proc/cpuinfo.
1) /sys/devices/system/cpu/cpuX/topology/physical_package_id:
represent the physical package id of cpu X;
2) /sys/devices/system/cpu/cpuX/topology/core_id:
represent the cpu core id to cpu X;
3) /sys/devices/system/cpu/cpuX/topology/thread_siblings:
represent the thread siblings to cpu X in the same core;
4) /sys/devices/system/cpu/cpuX/topology/core_siblings:
represent the thread siblings to cpu X in the same physical package;
To implement it in an architecture-neutral way, a new source file,
driver/base/topology.c, is to export the 5 attributes.
If one architecture wants to support this feature, it just needs to
implement 4 defines, typically in file include/asm-XXX/topology.h.
The 4 defines are:
#define topology_physical_package_id(cpu)
#define topology_core_id(cpu)
#define topology_thread_siblings(cpu)
#define topology_core_siblings(cpu)
The type of **_id is int.
The type of siblings is cpumask_t.
To be consistent on all architectures, the 4 attributes should have
deafult values if their values are unavailable. Below is the rule.
1) physical_package_id: If cpu has no physical package id, -1 is the
default value.
2) core_id: If cpu doesn't support multi-core, its core id is 0.
3) thread_siblings: Just include itself, if the cpu doesn't support
HT/multi-thread.
4) core_siblings: Just include itself, if the cpu doesn't support
multi-core and HT/Multi-thread.
So be careful when declaring the 4 defines in include/asm-XXX/topology.h.
If an attribute isn't defined on an architecture, it won't be exported.
Thank Nathan, Greg, Andi, Paul and Venki.
The patch provides defines for i386/x86_64/ia64.
Signed-off-by: Zhang, Yanmin <yanmin.zhang@intel.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Nick Piggin <nickpiggin@yahoo.com.au>
Cc: Greg KH <greg@kroah.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2006-02-03 04:04:36 -07:00
|
|
|
|
2008-06-04 22:47:29 -06:00
|
|
|
To be consistent on all architectures, include/linux/topology.h
|
|
|
|
provides default definitions for any of the above macros that are
|
|
|
|
not defined by include/asm-XXX/topology.h:
|
2017-05-14 06:52:01 -06:00
|
|
|
|
2019-02-25 23:20:00 -07:00
|
|
|
1) topology_physical_package_id: -1
|
|
|
|
2) topology_core_id: 0
|
|
|
|
3) topology_sibling_cpumask: just the given CPU
|
|
|
|
4) topology_core_cpumask: just the given CPU
|
2008-12-17 15:14:30 -07:00
|
|
|
|
2010-08-31 02:28:17 -06:00
|
|
|
For architectures that don't support books (CONFIG_SCHED_BOOK) there are no
|
|
|
|
default definitions for topology_book_id() and topology_book_cpumask().
|
2017-04-21 05:16:03 -06:00
|
|
|
For architectures that don't support drawers (CONFIG_SCHED_DRAWER) there are
|
2016-05-21 03:10:13 -06:00
|
|
|
no default definitions for topology_drawer_id() and topology_drawer_cpumask().
|
2010-08-31 02:28:17 -06:00
|
|
|
|
2009-10-21 21:45:31 -06:00
|
|
|
Additionally, CPU topology information is provided under
|
2008-12-17 15:14:30 -07:00
|
|
|
/sys/devices/system/cpu and includes these files. The internal
|
|
|
|
source for the output is in brackets ("[]").
|
|
|
|
|
2017-05-14 06:52:01 -06:00
|
|
|
=========== ==========================================================
|
2009-10-21 21:45:31 -06:00
|
|
|
kernel_max: the maximum CPU index allowed by the kernel configuration.
|
2008-12-17 15:14:30 -07:00
|
|
|
[NR_CPUS-1]
|
|
|
|
|
2009-10-21 21:45:31 -06:00
|
|
|
offline: CPUs that are not online because they have been
|
2008-12-17 15:14:30 -07:00
|
|
|
HOTPLUGGED off (see cpu-hotplug.txt) or exceed the limit
|
2009-10-21 21:45:31 -06:00
|
|
|
of CPUs allowed by the kernel configuration (kernel_max
|
2008-12-17 15:14:30 -07:00
|
|
|
above). [~cpu_online_mask + cpus >= NR_CPUS]
|
|
|
|
|
2009-10-21 21:45:31 -06:00
|
|
|
online: CPUs that are online and being scheduled [cpu_online_mask]
|
2008-12-17 15:14:30 -07:00
|
|
|
|
2009-10-21 21:45:31 -06:00
|
|
|
possible: CPUs that have been allocated resources and can be
|
2008-12-17 15:14:30 -07:00
|
|
|
brought online if they are present. [cpu_possible_mask]
|
|
|
|
|
2009-10-21 21:45:31 -06:00
|
|
|
present: CPUs that have been identified as being present in the
|
2008-12-17 15:14:30 -07:00
|
|
|
system. [cpu_present_mask]
|
2017-05-14 06:52:01 -06:00
|
|
|
=========== ==========================================================
|
2008-12-17 15:14:30 -07:00
|
|
|
|
|
|
|
The format for the above output is compatible with cpulist_parse()
|
|
|
|
[see <linux/cpumask.h>]. Some examples follow.
|
|
|
|
|
2009-10-21 21:45:31 -06:00
|
|
|
In this example, there are 64 CPUs in the system but cpus 32-63 exceed
|
2008-12-17 15:14:30 -07:00
|
|
|
the kernel max which is limited to 0..31 by the NR_CPUS config option
|
2009-10-21 21:45:31 -06:00
|
|
|
being 32. Note also that CPUs 2 and 4-31 are not online but could be
|
2017-05-14 06:52:01 -06:00
|
|
|
brought online as they are both present and possible::
|
2008-12-17 15:14:30 -07:00
|
|
|
|
|
|
|
kernel_max: 31
|
|
|
|
offline: 2,4-31,32-63
|
|
|
|
online: 0-1,3
|
|
|
|
possible: 0-31
|
|
|
|
present: 0-31
|
|
|
|
|
|
|
|
In this example, the NR_CPUS config option is 128, but the kernel was
|
2009-10-21 21:45:31 -06:00
|
|
|
started with possible_cpus=144. There are 4 CPUs in the system and cpu2
|
|
|
|
was manually taken offline (and is the only CPU that can be brought
|
2017-05-14 06:52:01 -06:00
|
|
|
online.)::
|
2008-12-17 15:14:30 -07:00
|
|
|
|
|
|
|
kernel_max: 127
|
|
|
|
offline: 2,4-127,128-143
|
|
|
|
online: 0-1,3
|
|
|
|
possible: 0-127
|
|
|
|
present: 0-3
|
|
|
|
|
|
|
|
See cpu-hotplug.txt for the possible_cpus=NUM kernel start parameter
|
2009-10-21 21:45:31 -06:00
|
|
|
as well as more information on the various cpumasks.
|