| /kernel/linux/linux-5.10/Documentation/core-api/ |
| D | this_cpu_ops.rst | 8 this_cpu operations are a way of optimizing access to per cpu
11 the cpu permanently stored the beginning of the per cpu area for a
14 this_cpu operations add a per cpu variable offset to the processor
15 specific per cpu base and encode that operation in the instruction
16 operating on the per cpu variable.
24 Read-modify-write operations are of particular interest. Frequently
32 synchronization is not necessary since we are dealing with per cpu
37 Please note that accesses by remote processors to a per cpu area are
66 ------------------------------------
69 per cpu area. It is then possible to simply use the segment override
[all …]
|
| D | local_ops.rst | 29 Local atomic operations are meant to provide fast and highly reentrant per CPU
34 Having fast per CPU atomic counters is interesting in many cases: it does not
40 CPU which owns the data. Therefore, care must taken to make sure that only one
41 CPU writes to the ``local_t`` data. This is done by using per cpu data and
43 however permitted to read ``local_t`` data from any CPU: it will then appear to
44 be written out of order wrt other memory writes by the owner CPU.
54 ``asm-generic/local.h`` in your architecture's ``local.h`` is sufficient.
66 * Variables touched by local ops must be per cpu variables.
67 * *Only* the CPU owner of these variables must write to them.
68 * This CPU can use local ops from any context (process, irq, softirq, nmi, ...)
[all …]
|
| /kernel/linux/linux-5.10/arch/ia64/mm/ |
| D | discontig.c | 1 // SPDX-License-Identifier: GPL-2.0
7 * Copyright (c) 2002 Kimio Suganuma <k-suganuma@da.jp.nec.com>
33 * Track per-node information needed to setup the boot memory allocator, the
34 * per-node areas, and the real VM.
50 * To prevent cache aliasing effects, align per-node structures so that they
55 ((((addr) + 1024*1024-1) & ~(1024*1024-1)) + \
56 (((node)*PERCPU_PAGE_SIZE) & (MAX_NODE_ALIGN_OFFSET - 1)))
59 * build_node_maps - callback to setup mem_data structs for each node
67 * if necessary. Any non-existent pages will simply be part of the virtual
90 * early_nr_cpus_node - return number of cpus on a given node
[all …]
|
| /kernel/linux/linux-5.10/tools/power/x86/x86_energy_perf_policy/ |
| D | x86_energy_perf_policy.8 | 1 .\" This page Copyright (C) 2010 - 2015 Len Brown <len.brown@intel.com>
5 x86_energy_perf_policy \- Manage Energy vs. Performance Policy via x86 Model Specific Registers
10 .RB "scope: \-\-cpu\ cpu-list | \-\-pkg\ pkg-list"
12 .RB "cpu-list, pkg-list: # | #,# | #-# | all"
14 .RB "field: \-\-all | \-\-epb | \-\-hwp-epp | \-\-hwp-min | \-\-hwp-max | \-\-hwp-desired"
16 .RB "other: (\-\-force | \-\-hwp-enable | \-\-turbo-enable) value)"
18 .RB "value: # | default | performance | balance-performance | balance-power | power"
21 displays and updates energy-performance policy settings specific to
23 updates, no matter if the Linux cpufreq sub-system is enabled or not.
27 such as how aggressively the hardware enters and exits CPU idle states (C-states)
[all …]
|
| /kernel/linux/linux-5.10/Documentation/virt/ |
| D | guest-halt-polling.rst | 15 2) The VM-exit cost can be avoided.
25 ("per-cpu guest_halt_poll_ns"), which is adjusted by the algorithm
42 Division factor used to shrink per-cpu guest_halt_poll_ns when
49 Multiplication factor used to grow per-cpu guest_halt_poll_ns
50 when event occurs after per-cpu guest_halt_poll_ns
57 The per-cpu guest_halt_poll_ns eventually reaches zero
59 per-cpu guest_halt_poll_ns when growing. This can
70 to avoid it (per-cpu guest_halt_poll_ns will remain
82 - Care should be taken when setting the guest_halt_poll_ns parameter as a
83 large value has the potential to drive the cpu usage to 100% on a machine
|
| /kernel/linux/linux-5.10/Documentation/trace/ |
| D | events-kmem.rst | 8 - Slab allocation of small objects of unknown type (kmalloc)
9 - Slab allocation of small objects of known type
10 - Page allocation
11 - Per-CPU Allocator Activity
12 - External Fragmentation
40 These events are similar in usage to the kmalloc-related events except that
50 mm_page_alloc_zone_locked page=%p pfn=%lu order=%u migratetype=%d cpu=%d percpu_refill=%d
56 the per-CPU allocator (high performance) or the buddy allocator.
60 amounts of activity imply high activity on the zone->lock. Taking this lock
72 contention on the zone->lru_lock.
[all …]
|
| /kernel/linux/linux-5.10/include/linux/ |
| D | local_lock.h | 1 /* SPDX-License-Identifier: GPL-2.0 */
8 * local_lock_init - Runtime initialize a lock instance
13 * local_lock - Acquire a per CPU local lock
19 * local_lock_irq - Acquire a per CPU local lock and disable interrupts
25 * local_lock_irqsave - Acquire a per CPU local lock, save and disable
34 * local_unlock - Release a per CPU local lock
40 * local_unlock_irq - Release a per CPU local lock and enable interrupts
46 * local_unlock_irqrestore - Release a per CPU local lock and restore
|
| D | srcutree.h | 1 /* SPDX-License-Identifier: GPL-2.0+ */
3 * Sleepable Read-Copy Update mechanism for mutual exclusion,
21 * Per-CPU structure feeding into leaf srcu_node, similar in function
25 /* Read-side state. */
26 unsigned long srcu_lock_count[2]; /* Locks per CPU. */
27 unsigned long srcu_unlock_count[2]; /* Unlocks per CPU. */
29 /* Update-side state. */
40 /* ->srcu_data_have_cbs[]. */
41 int cpu; member
52 /* is > ->srcu_gq_seq. */
[all …]
|
| D | percpu-defs.h | 1 /* SPDX-License-Identifier: GPL-2.0-only */
3 * linux/percpu-defs.h - basic definitions for percpu areas
40 * Base implementations of per-CPU variable declarations and definitions, where
108 * Variant on the per-CPU variable declaration/definition theme used for
109 * ordinary per-CPU variables.
118 * Declaration/definition used for per-CPU variables that must come first in
128 * Declaration/definition used for per-CPU variables that must be cacheline
130 * data corresponds to a particular CPU, inefficiencies due to direct access by
134 * An example of this would be statistical data, where each CPU's set of data
135 * is updated by that CPU alone, but the data from across all CPUs is collated
[all …]
|
| /kernel/linux/linux-5.10/tools/perf/pmu-events/arch/x86/broadwell/ |
| D | bdw-metrics.json | 7 …-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into mi…
10 …nd undersupplies its Backend. SMT version; use when SMT is enabled and measuring per logical CPU.",
14 …-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into mi…
18 …"MetricExpr": "( UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * INT_MISC.RECOVERY_CYCLES ) / (4…
21 …s for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For…
24 …ue to incorrect speculations. SMT version; use when SMT is enabled and measuring per logical CPU.",
25 …"MetricExpr": "( UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (( INT_MISC.RECOVERY_CYCLES_ANY…
28 …-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work…
32 …"MetricExpr": "1 - ( (IDQ_UOPS_NOT_DELIVERED.CORE / (4 * cycles)) + (( UOPS_ISSUED.ANY - UOPS_RETI…
35 …-of-order scheduler dispatches ready uops into their respective execution units; and once complete…
[all …]
|
| /kernel/linux/linux-5.10/arch/x86/kernel/ |
| D | apb_timer.c | 1 // SPDX-License-Identifier: GPL-2.0-only
17 * - timer 0 - NR_CPUs for per cpu timer
18 * - one timer for clocksource
19 * - one timer for watchdog driver.
20 * It is also worth notice that APB timer does not support true one-shot mode,
21 * free-running mode will be used here to emulate one-shot mode.
22 * APB timer can also be used as broadcast timer along with per cpu local APIC
34 #include <linux/cpu.h>
39 #include <asm/intel-mid.h>
60 int cpu; member
[all …]
|
| /kernel/linux/linux-5.10/tools/perf/pmu-events/arch/x86/jaketown/ |
| D | jkt-metrics.json | 7 …-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into mi…
10 …nd undersupplies its Backend. SMT version; use when SMT is enabled and measuring per logical CPU.",
14 …-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into mi…
18 …"MetricExpr": "( UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * INT_MISC.RECOVERY_CYCLES ) / (4…
21 …s for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For…
24 …ue to incorrect speculations. SMT version; use when SMT is enabled and measuring per logical CPU.",
25 …"MetricExpr": "( UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (( INT_MISC.RECOVERY_CYCLES_ANY…
28 …-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work…
32 …"MetricExpr": "1 - ( (IDQ_UOPS_NOT_DELIVERED.CORE / (4 * cycles)) + (( UOPS_ISSUED.ANY - UOPS_RETI…
35 …-of-order scheduler dispatches ready uops into their respective execution units; and once complete…
[all …]
|
| /kernel/linux/linux-5.10/tools/perf/pmu-events/arch/x86/sandybridge/ |
| D | snb-metrics.json | 7 …-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into mi…
10 …nd undersupplies its Backend. SMT version; use when SMT is enabled and measuring per logical CPU.",
14 …-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into mi…
18 …"MetricExpr": "( UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * INT_MISC.RECOVERY_CYCLES ) / (4…
21 …s for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For…
24 …ue to incorrect speculations. SMT version; use when SMT is enabled and measuring per logical CPU.",
25 …"MetricExpr": "( UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (( INT_MISC.RECOVERY_CYCLES_ANY…
28 …-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work…
32 …"MetricExpr": "1 - ( (IDQ_UOPS_NOT_DELIVERED.CORE / (4 * cycles)) + (( UOPS_ISSUED.ANY - UOPS_RETI…
35 …-of-order scheduler dispatches ready uops into their respective execution units; and once complete…
[all …]
|
| /kernel/linux/linux-5.10/tools/perf/pmu-events/arch/x86/haswell/ |
| D | hsw-metrics.json | 7 …-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into mi…
10 …nd undersupplies its Backend. SMT version; use when SMT is enabled and measuring per logical CPU.",
14 …-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into mi…
18 …"MetricExpr": "( UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * INT_MISC.RECOVERY_CYCLES ) / (4…
21 …s for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For…
24 …ue to incorrect speculations. SMT version; use when SMT is enabled and measuring per logical CPU.",
25 …"MetricExpr": "( UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (( INT_MISC.RECOVERY_CYCLES_ANY…
28 …-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work…
32 …"MetricExpr": "1 - ( (IDQ_UOPS_NOT_DELIVERED.CORE / (4 * cycles)) + (( UOPS_ISSUED.ANY - UOPS_RETI…
35 …-of-order scheduler dispatches ready uops into their respective execution units; and once complete…
[all …]
|
| /kernel/linux/linux-5.10/tools/perf/pmu-events/arch/x86/haswellx/ |
| D | hsx-metrics.json | 7 …-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into mi…
10 …nd undersupplies its Backend. SMT version; use when SMT is enabled and measuring per logical CPU.",
14 …-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into mi…
18 …"MetricExpr": "( UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * INT_MISC.RECOVERY_CYCLES ) / (4…
21 …s for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For…
24 …ue to incorrect speculations. SMT version; use when SMT is enabled and measuring per logical CPU.",
25 …"MetricExpr": "( UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (( INT_MISC.RECOVERY_CYCLES_ANY…
28 …-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work…
32 …"MetricExpr": "1 - ( (IDQ_UOPS_NOT_DELIVERED.CORE / (4 * cycles)) + (( UOPS_ISSUED.ANY - UOPS_RETI…
35 …-of-order scheduler dispatches ready uops into their respective execution units; and once complete…
[all …]
|
| /kernel/linux/linux-5.10/Documentation/x86/ |
| D | topology.rst | 1 .. SPDX-License-Identifier: GPL-2.0
11 The architecture-agnostic topology definitions are in
12 Documentation/admin-guide/cputopology.rst. This file holds x86-specific
17 Needless to say, code should use the generic functions - this file is *only*
35 - packages
36 - cores
37 - threads
48 Package-related topology information in the kernel:
50 - cpuinfo_x86.x86_max_cores:
54 - cpuinfo_x86.x86_max_dies:
[all …]
|
| /kernel/linux/linux-5.10/tools/perf/pmu-events/arch/x86/broadwellde/ |
| D | bdwde-metrics.json | 3 "BriefDescription": "Instructions Per Cycle (per logical thread)",
9 "BriefDescription": "Uops Per Instruction",
27 "BriefDescription": "Cycles Per Instruction (threaded)",
33 …"BriefDescription": "Per-thread actual clocks when the logical processor is active. This is called…
39 "BriefDescription": "Total issue-pipeline slots",
51 "BriefDescription": "Instructions Per Cycle (per physical core)",
57 …"BriefDescription": "Instruction-Level-Parallelism (average number of uops executed when there is …
58 …"MetricExpr": "UOPS_EXECUTED.THREAD / ( cpu@uops_executed.core\\,cmask\\=1@ / 2) if #SMT_on else U…
64 …"MetricExpr": "2* ( RS_EVENTS.EMPTY_CYCLES - ICACHE.IFDATA_STALL - ( 14 * ITLB_MISSES.STLB_HIT + …
75 "BriefDescription": "Actual Average Latency for L1 data-cache miss demand loads",
[all …]
|
| /kernel/linux/linux-5.10/drivers/irqchip/ |
| D | irq-armada-370-xp.c | 7 * Gregory CLEMENT <gregory.clement@free-electrons.com>
8 * Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
23 #include <linux/cpu.h>
40 * To CPU 0 To CPU 1
44 * +---------------+ +---------------+
46 * | per-CPU | | per-CPU |
50 * +---------------+ +---------------+
55 * +-------------------+
60 * +-------------------+
71 * The "per-CPU mask/unmask" is modified using the
[all …]
|
| /kernel/linux/linux-5.10/tools/perf/Documentation/ |
| D | perf-stat.txt | 1 perf-stat(1)
5 ----
6 perf-stat - Run a command and gather performance counter statistics
9 --------
11 'perf stat' [-e <EVENT> | --event=EVENT] [-a] <command>
12 'perf stat' [-e <EVENT> | --event=EVENT] [-a] -- <command> [<options>]
13 'perf stat' [-e <EVENT> | --event=EVENT] [-a] record [-o file] -- <command> [<options>]
14 'perf stat' report [-i file]
17 -----------
23 -------
[all …]
|
| /kernel/linux/linux-5.10/scripts/gdb/linux/ |
| D | cpus.py | 4 # per-cpu tools
6 # Copyright (c) Siemens AG, 2011-2013
24 return gdb.selected_thread().num - 1
27 if tid > (0x100000000 - MAX_CPUS - 2):
28 return 0x100000000 - tid - 2
30 return tasks.get_thread_info(tasks.get_task_by_pid(tid))['cpu']
32 raise gdb.GdbError("Sorry, obtaining the current CPU is not yet "
36 def per_cpu(var_ptr, cpu): argument
37 if cpu == -1:
38 cpu = get_current_cpu()
[all …]
|
| /kernel/linux/linux-5.10/tools/perf/pmu-events/arch/x86/ivybridge/ |
| D | ivb-metrics.json | 7 …-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into mi…
10 …nd undersupplies its Backend. SMT version; use when SMT is enabled and measuring per logical CPU.",
14 …-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into mi…
18 …"MetricExpr": "( UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * INT_MISC.RECOVERY_CYCLES ) / (4…
21 …s for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For…
24 …ue to incorrect speculations. SMT version; use when SMT is enabled and measuring per logical CPU.",
25 …"MetricExpr": "( UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (( INT_MISC.RECOVERY_CYCLES_ANY…
28 …-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work…
32 …"MetricExpr": "1 - ( (IDQ_UOPS_NOT_DELIVERED.CORE / (4 * cycles)) + (( UOPS_ISSUED.ANY - UOPS_RETI…
35 …-of-order scheduler dispatches ready uops into their respective execution units; and once complete…
[all …]
|
| /kernel/linux/linux-5.10/tools/perf/pmu-events/arch/x86/ivytown/ |
| D | ivt-metrics.json | 7 …-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into mi…
10 …nd undersupplies its Backend. SMT version; use when SMT is enabled and measuring per logical CPU.",
14 …-lines are fetched from the memory subsystem; parsed into instructions; and lastly decoded into mi…
18 …"MetricExpr": "( UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * INT_MISC.RECOVERY_CYCLES ) / (4…
21 …s for which the issue-pipeline was blocked due to recovery from earlier incorrect speculation. For…
24 …ue to incorrect speculations. SMT version; use when SMT is enabled and measuring per logical CPU.",
25 …"MetricExpr": "( UOPS_ISSUED.ANY - UOPS_RETIRED.RETIRE_SLOTS + 4 * (( INT_MISC.RECOVERY_CYCLES_ANY…
28 …-pipeline was blocked due to recovery from earlier incorrect speculation. For example; wasted work…
32 …"MetricExpr": "1 - ( (IDQ_UOPS_NOT_DELIVERED.CORE / (4 * cycles)) + (( UOPS_ISSUED.ANY - UOPS_RETI…
35 …-of-order scheduler dispatches ready uops into their respective execution units; and once complete…
[all …]
|
| /kernel/linux/linux-5.10/Documentation/cpu-freq/ |
| D | cpu-drivers.rst | 1 .. SPDX-License-Identifier: GPL-2.0
10 - Dominik Brodowski <linux@brodo.de>
11 - Rafael J. Wysocki <rafael.j.wysocki@intel.com>
12 - Viresh Kumar <viresh.kumar@linaro.org>
18 1.2 Per-CPU Initialization
31 So, you just got a brand-new CPU / chipset with datasheets and want to
32 add cpufreq support for this CPU / chipset? Great. Here are some hints
37 ------------------
40 function check whether this kernel runs on the right CPU and the right
46 .name - The name of this driver.
[all …]
|
| /kernel/linux/linux-5.10/kernel/printk/ |
| D | printk_safe.c | 1 // SPDX-License-Identifier: GPL-2.0-or-later
3 * printk_safe.c - Safe printk for printk-deadlock-prone contexts
21 * the strings into a per-CPU buffer. The content of the buffer
26 * per-CPU variable.
28 * The implementation allows to flush the strings also from another CPU.
33 #define SAFE_LOG_BUF_LEN ((1 << CONFIG_PRINTK_SAFE_LOG_BUF_SHIFT) - \
34 sizeof(atomic_t) - \
35 sizeof(atomic_t) - \
58 irq_work_queue(&s->work); in queue_flush_work()
62 * Add a message to per-CPU context-dependent buffer. NMI and printk-safe
[all …]
|
| /kernel/linux/linux-5.10/Documentation/admin-guide/ |
| D | perf-security.rst | 7 --------
19 1. System hardware and software configuration data, for example: a CPU
30 faults, CPU migrations), architectural hardware performance counters
50 -------------------------------
66 independently enabled and disabled on per-thread basis for processes and
100 ---------------------------------
102 Mechanisms of capabilities, privileged capability-dumb files [6]_ and
115 # ls -alhF
116 -rwxr-xr-x 2 root root 11M Oct 19 15:12 perf
118 # ls -alhF
[all …]
|