aboutsummaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
-rw-r--r--Documentation/admin-guide/pm/cpufreq.rst10
-rw-r--r--Documentation/admin-guide/pm/intel_pstate.rst25
2 files changed, 23 insertions, 12 deletions
diff --git a/Documentation/admin-guide/pm/cpufreq.rst b/Documentation/admin-guide/pm/cpufreq.rst
index 7eca9026a9ed..b97ce64d5976 100644
--- a/Documentation/admin-guide/pm/cpufreq.rst
+++ b/Documentation/admin-guide/pm/cpufreq.rst
@@ -396,8 +396,8 @@ RT or deadline scheduling classes, the governor will increase the frequency to
396the allowed maximum (that is, the ``scaling_max_freq`` policy limit). In turn, 396the allowed maximum (that is, the ``scaling_max_freq`` policy limit). In turn,
397if it is invoked by the CFS scheduling class, the governor will use the 397if it is invoked by the CFS scheduling class, the governor will use the
398Per-Entity Load Tracking (PELT) metric for the root control group of the 398Per-Entity Load Tracking (PELT) metric for the root control group of the
399given CPU as the CPU utilization estimate (see the `Per-entity load tracking`_ 399given CPU as the CPU utilization estimate (see the *Per-entity load tracking*
400LWN.net article for a description of the PELT mechanism). Then, the new 400LWN.net article [1]_ for a description of the PELT mechanism). Then, the new
401CPU frequency to apply is computed in accordance with the formula 401CPU frequency to apply is computed in accordance with the formula
402 402
403 f = 1.25 * ``f_0`` * ``util`` / ``max`` 403 f = 1.25 * ``f_0`` * ``util`` / ``max``
@@ -698,4 +698,8 @@ hardware feature (e.g. all Intel ones), even if the
698:c:macro:`CONFIG_X86_ACPI_CPUFREQ_CPB` configuration option is set. 698:c:macro:`CONFIG_X86_ACPI_CPUFREQ_CPB` configuration option is set.
699 699
700 700
701.. _Per-entity load tracking: https://lwn.net/Articles/531853/ 701References
702==========
703
704.. [1] Jonathan Corbet, *Per-entity load tracking*,
705 https://lwn.net/Articles/531853/
diff --git a/Documentation/admin-guide/pm/intel_pstate.rst b/Documentation/admin-guide/pm/intel_pstate.rst
index ec0f7c111f65..6dba90b753d7 100644
--- a/Documentation/admin-guide/pm/intel_pstate.rst
+++ b/Documentation/admin-guide/pm/intel_pstate.rst
@@ -20,11 +20,10 @@ you have not done that yet.]
20 20
21For the processors supported by ``intel_pstate``, the P-state concept is broader 21For the processors supported by ``intel_pstate``, the P-state concept is broader
22than just an operating frequency or an operating performance point (see the 22than just an operating frequency or an operating performance point (see the
23`LinuxCon Europe 2015 presentation by Kristen Accardi <LCEU2015_>`_ for more 23LinuxCon Europe 2015 presentation by Kristen Accardi [1]_ for more
24information about that). For this reason, the representation of P-states used 24information about that). For this reason, the representation of P-states used
25by ``intel_pstate`` internally follows the hardware specification (for details 25by ``intel_pstate`` internally follows the hardware specification (for details
26refer to `Intel® 64 and IA-32 Architectures Software Developer’s Manual 26refer to Intel Software Developer’s Manual [2]_). However, the ``CPUFreq`` core
27Volume 3: System Programming Guide <SDM_>`_). However, the ``CPUFreq`` core
28uses frequencies for identifying operating performance points of CPUs and 27uses frequencies for identifying operating performance points of CPUs and
29frequencies are involved in the user space interface exposed by it, so 28frequencies are involved in the user space interface exposed by it, so
30``intel_pstate`` maps its internal representation of P-states to frequencies too 29``intel_pstate`` maps its internal representation of P-states to frequencies too
@@ -561,9 +560,9 @@ or to pin every task potentially sensitive to them to a specific CPU.]
561 560
562On the majority of systems supported by ``intel_pstate``, the ACPI tables 561On the majority of systems supported by ``intel_pstate``, the ACPI tables
563provided by the platform firmware contain ``_PSS`` objects returning information 562provided by the platform firmware contain ``_PSS`` objects returning information
564that can be used for CPU performance scaling (refer to the `ACPI specification`_ 563that can be used for CPU performance scaling (refer to the ACPI specification
565for details on the ``_PSS`` objects and the format of the information returned 564[3]_ for details on the ``_PSS`` objects and the format of the information
566by them). 565returned by them).
567 566
568The information returned by the ACPI ``_PSS`` objects is used by the 567The information returned by the ACPI ``_PSS`` objects is used by the
569``acpi-cpufreq`` scaling driver. On systems supported by ``intel_pstate`` 568``acpi-cpufreq`` scaling driver. On systems supported by ``intel_pstate``
@@ -728,6 +727,14 @@ P-state is called, the ``ftrace`` filter can be set to to
728 <idle>-0 [000] ..s. 2537.654843: intel_pstate_set_pstate <-intel_pstate_timer_func 727 <idle>-0 [000] ..s. 2537.654843: intel_pstate_set_pstate <-intel_pstate_timer_func
729 728
730 729
731.. _LCEU2015: http://events.linuxfoundation.org/sites/events/files/slides/LinuxConEurope_2015.pdf 730References
732.. _SDM: http://www.intel.com/content/www/us/en/architecture-and-technology/64-ia-32-architectures-software-developer-system-programming-manual-325384.html 731==========
733.. _ACPI specification: http://www.uefi.org/sites/default/files/resources/ACPI_6_1.pdf 732
733.. [1] Kristen Accardi, *Balancing Power and Performance in the Linux Kernel*,
734 http://events.linuxfoundation.org/sites/events/files/slides/LinuxConEurope_2015.pdf
735
736.. [2] *Intel® 64 and IA-32 Architectures Software Developer’s Manual Volume 3: System Programming Guide*,
737 http://www.intel.com/content/www/us/en/architecture-and-technology/64-ia-32-architectures-software-developer-system-programming-manual-325384.html
738
739.. [3] *Advanced Configuration and Power Interface Specification*,
740 https://uefi.org/sites/default/files/resources/ACPI_6_3_final_Jan30.pdf