cachepc-linux

Fork of AMDESE/linux with modifications for CachePC side-channel attack
git clone https://git.sinitax.com/sinitax/cachepc-linux
Log | Files | Refs | README | LICENSE | sfeed.txt

cputopology.rst (3932B)


      1===========================================
      2How CPU topology info is exported via sysfs
      3===========================================
      4
      5CPU topology info is exported via sysfs. Items (attributes) are similar
      6to /proc/cpuinfo output of some architectures. They reside in
      7/sys/devices/system/cpu/cpuX/topology/. Please refer to the ABI file:
      8Documentation/ABI/stable/sysfs-devices-system-cpu.
      9
     10Architecture-neutral, drivers/base/topology.c, exports these attributes.
     11However the die, cluster, book, and drawer hierarchy related sysfs files will
     12only be created if an architecture provides the related macros as described
     13below.
     14
     15For an architecture to support this feature, it must define some of
     16these macros in include/asm-XXX/topology.h::
     17
     18	#define topology_physical_package_id(cpu)
     19	#define topology_die_id(cpu)
     20	#define topology_cluster_id(cpu)
     21	#define topology_core_id(cpu)
     22	#define topology_book_id(cpu)
     23	#define topology_drawer_id(cpu)
     24	#define topology_sibling_cpumask(cpu)
     25	#define topology_core_cpumask(cpu)
     26	#define topology_cluster_cpumask(cpu)
     27	#define topology_die_cpumask(cpu)
     28	#define topology_book_cpumask(cpu)
     29	#define topology_drawer_cpumask(cpu)
     30
     31The type of ``**_id macros`` is int.
     32The type of ``**_cpumask macros`` is ``(const) struct cpumask *``. The latter
     33correspond with appropriate ``**_siblings`` sysfs attributes (except for
     34topology_sibling_cpumask() which corresponds with thread_siblings).
     35
     36To be consistent on all architectures, include/linux/topology.h
     37provides default definitions for any of the above macros that are
     38not defined by include/asm-XXX/topology.h:
     39
     401) topology_physical_package_id: -1
     412) topology_die_id: -1
     423) topology_cluster_id: -1
     434) topology_core_id: 0
     445) topology_book_id: -1
     456) topology_drawer_id: -1
     467) topology_sibling_cpumask: just the given CPU
     478) topology_core_cpumask: just the given CPU
     489) topology_cluster_cpumask: just the given CPU
     4910) topology_die_cpumask: just the given CPU
     5011) topology_book_cpumask:  just the given CPU
     5112) topology_drawer_cpumask: just the given CPU
     52
     53Additionally, CPU topology information is provided under
     54/sys/devices/system/cpu and includes these files.  The internal
     55source for the output is in brackets ("[]").
     56
     57    =========== ==========================================================
     58    kernel_max: the maximum CPU index allowed by the kernel configuration.
     59		[NR_CPUS-1]
     60
     61    offline:	CPUs that are not online because they have been
     62		HOTPLUGGED off or exceed the limit of CPUs allowed by the
     63		kernel configuration (kernel_max above).
     64		[~cpu_online_mask + cpus >= NR_CPUS]
     65
     66    online:	CPUs that are online and being scheduled [cpu_online_mask]
     67
     68    possible:	CPUs that have been allocated resources and can be
     69		brought online if they are present. [cpu_possible_mask]
     70
     71    present:	CPUs that have been identified as being present in the
     72		system. [cpu_present_mask]
     73    =========== ==========================================================
     74
     75The format for the above output is compatible with cpulist_parse()
     76[see <linux/cpumask.h>].  Some examples follow.
     77
     78In this example, there are 64 CPUs in the system but cpus 32-63 exceed
     79the kernel max which is limited to 0..31 by the NR_CPUS config option
     80being 32.  Note also that CPUs 2 and 4-31 are not online but could be
     81brought online as they are both present and possible::
     82
     83     kernel_max: 31
     84        offline: 2,4-31,32-63
     85         online: 0-1,3
     86       possible: 0-31
     87        present: 0-31
     88
     89In this example, the NR_CPUS config option is 128, but the kernel was
     90started with possible_cpus=144.  There are 4 CPUs in the system and cpu2
     91was manually taken offline (and is the only CPU that can be brought
     92online.)::
     93
     94     kernel_max: 127
     95        offline: 2,4-127,128-143
     96         online: 0-1,3
     97       possible: 0-127
     98        present: 0-3
     99
    100See Documentation/core-api/cpu_hotplug.rst for the possible_cpus=NUM
    101kernel start parameter as well as more information on the various cpumasks.