28 Jun, 2017
1 commit
-
Fixes wrong bits shift operation in the rapl_write_data_raw function, which
might cause overridding bits outside of the mask.For example, writing new TIME_WINDOW1 value can override POWER_LIMIT1.
Signed-off-by: Adam Lessnau
Signed-off-by: Rafael J. Wysocki
14 May, 2017
1 commit
-
In the current code we accidentally return the successful result from
idr_alloc() instead of a negative error pointer. The caller is looking
for an error pointer and so it treats the returned value as a valid
pointer.This one might be a bit serious because if it lets people get around the
kernel's protection for remapping NULL. I'm not sure.Fixes: 75d2364ea0ca (PowerCap: Add class driver)
Signed-off-by: Dan Carpenter
Reviewed-by: Srinivas Pandruvada
Signed-off-by: Rafael J. Wysocki
29 Apr, 2017
1 commit
-
Gemini Lake RAPL support is similar to Goldmont.
Signed-off-by: David E. Box
Acked-by: Pan, Jacob jun
Signed-off-by: Rafael J. Wysocki
01 Dec, 2016
1 commit
-
Add Knights Mill (KNM) to the list of CPUIDs supported by intel_rapl
Signed-off-by: Piotr Luc
Reviewed-by: Dave Hansen
Signed-off-by: Rafael J. Wysocki
29 Nov, 2016
1 commit
-
Commit e1399ba20eee ("powercap / RAPL: handle missing MSRs") added
contraint_to_pl() function to return index into an array. But it
can potentially return -EINVAL if powercap layer sends an out of
range constraint ID. This patch adds sanity check.Unnecessary RAPL domain pointer check is removed since it must be
initialized before calling rapl_unit_xlate().Fixes: e1399ba20eee ("powercap / RAPL: handle missing MSRs")
Reported-by: Odzioba, Lukasz
Reported-by: Koss, Marcin
Signed-off-by: Jacob Pan
Signed-off-by: Rafael J. Wysocki
25 Nov, 2016
5 commits
-
The ability of the CPU hotplug code to stop online/offline at each step
makes it necessary to track the activated CPUs in a package directly,
because outerwise a CPU offline callback can find CPUs which have already
executed the offline callback, but are not yet marked offline in the
topology mask. That could make such a CPU the package leader and in case
that CPU goes fully offline leave the package lead orphaned.Signed-off-by: Thomas Gleixner
Tested-by: Jacob Pan
Signed-off-by: Rafael J. Wysocki -
The whole init/exit code is a duplicate of the cpuhotplug code. So we can
just let the hotplug code do the actual work of setting up and tearing down
the domains.This also restores the package hardware when a package is removed during
hotplug instead of leaving it in the last configured state and only reset
it when the driver is removed.Signed-off-by: Thomas Gleixner
Tested-by: Jacob Pan
Signed-off-by: Rafael J. Wysocki -
Install the callbacks via the state machine as a first step. The init/exit
code is a duplicate of the hotplug code. This is cleaned up in a
consecutive patch.Signed-off-by: Sebastian Andrzej Siewior
Signed-off-by: Thomas Gleixner
Tested-by: Jacob Pan
Signed-off-by: Rafael J. Wysocki -
If rapl_package_register_powercap() fails in rapl_add_package() the
function happily returns 0.Capture the error code and propagate it.
Signed-off-by: Thomas Gleixner
Tested-by: Jacob Pan
Signed-off-by: Rafael J. Wysocki -
The domain data of packages is only updated at init time, but new packages
created by hotplug miss that treatment.Add it there and remove the global update at init time, because it's now
obsolete.The more interesting question is why rapl_update_domain_data() exists at
all as nothing ever uses that data.Signed-off-by: Thomas Gleixner
Tested-by: Jacob Pan
Signed-off-by: Rafael J. Wysocki
08 Sep, 2016
1 commit
-
Merrifield2 is actually Moorefield.
Rename it accordingly and drop tail digit from Merrifield1.
Signed-off-by: Andy Shevchenko
Cc: Dave Hansen
Cc: Linus Torvalds
Cc: Peter Zijlstra
Cc: Thomas Gleixner
Link: http://lkml.kernel.org/r/20160906184254.94440-1-andriy.shevchenko@linux.intel.com
Signed-off-by: Ingo Molnar
25 Jun, 2016
1 commit
-
It's confirmed that RAPL works as expected on Ivy Bridge servers.
Tested against processor: Intel(R) Xeon(R) CPU E5-2697 v2 @2.70GHzSigned-off-by: Xiaolong Wang
Signed-off-by: Rafael J. Wysocki
15 Jun, 2016
3 commits
-
Denverton micro server is Atom-based, but its RAPL interface
is compatible with Core-based CPUs.Signed-off-by: Jacob Pan
Signed-off-by: Rafael J. Wysocki -
Some RAPL MSRs may not exist on some CPUs, we need to continue
the topology detection and enumerate what is available.This patch handles the missing MSRs, then reports to the powercap
layer only the features available.Signed-off-by: Jacob Pan
Signed-off-by: Rafael J. Wysocki
10 Jun, 2016
1 commit
-
Since the RAPL interface is not architectual, its enumeration depends
on poking MSRs instead of using the CPUID method.In KVM guests, the RAPL driver probe will fail and emit the following
message for every CPU: "no valid rapl domains found in package"This patch converts the warning to a debug message only (still return
-ENODEV so that RAPL does not run in KVM guests).Signed-off-by: Jacob Pan
[ rjw: Subject & changelog ]
Signed-off-by: Rafael J. Wysocki
08 Jun, 2016
3 commits
-
SKX uses similar RAPL interface as Broadwell server according to
Jacob Pan.Signed-off-by: Dave Hansen
Cc: Andy Lutomirski
Cc: Borislav Petkov
Cc: Brian Gerst
Cc: Dave Hansen
Cc: Denys Vlasenko
Cc: H. Peter Anvin
Cc: Jacob Pan
Cc: Linus Torvalds
Cc: Peter Zijlstra
Cc: Rafael J. Wysocki
Cc: Thomas Gleixner
Cc: jacob.jun.pan@intel.com
Cc: linux-pm@vger.kernel.org
Link: http://lkml.kernel.org/r/20160603001955.38E1E684@viggo.jf.intel.com
Signed-off-by: Ingo Molnar -
Let's make an effort to group these things by microarchitecture
name. It makes it easier to see if something got missed.Signed-off-by: Dave Hansen
Acked-by: Rafael J. Wysocki
Cc: Andy Lutomirski
Cc: Borislav Petkov
Cc: Brian Gerst
Cc: Dave Hansen
Cc: Denys Vlasenko
Cc: H. Peter Anvin
Cc: Linus Torvalds
Cc: Peter Zijlstra
Cc: Thomas Gleixner
Cc: jacob.jun.pan@intel.com
Cc: linux-pm@vger.kernel.org
Link: http://lkml.kernel.org/r/20160603001937.B53A383A@viggo.jf.intel.com
Signed-off-by: Ingo Molnar -
Use the new macros to remove another large set of open-coded values.
Signed-off-by: Dave Hansen
Acked-by: Rafael J. Wysocki
Cc: Andy Lutomirski
Cc: Borislav Petkov
Cc: Brian Gerst
Cc: Dave Hansen
Cc: Denys Vlasenko
Cc: H. Peter Anvin
Cc: Linus Torvalds
Cc: Peter Zijlstra
Cc: Thomas Gleixner
Cc: jacob.jun.pan@intel.com
Cc: linux-pm@vger.kernel.org
Link: http://lkml.kernel.org/r/20160603001936.F474F9D8@viggo.jf.intel.com
Signed-off-by: Ingo Molnar
17 May, 2016
1 commit
-
Pull power management updates from Rafael Wysocki:
"The majority of changes go into the cpufreq subsystem this time.To me, quite obviously, the biggest ticket item is the new "schedutil"
governor. Interestingly enough, it's the first new cpufreq governor
since the beginning of the git era (except for some out-of-the-tree
ones).There are two main differences between it and the existing governors.
First, it uses the information provided by the scheduler directly for
making its decisions, so it doesn't have to track anything by itself.
Second, it can invoke drivers (supporting that feature) to adjust CPU
performance right away without having to spawn work items to be
executed in process context or similar. Currently, the acpi-cpufreq
driver is the only one supporting that mode of operation, but then it
is used on a large number of systems.The "schedutil" governor as included here is very simple and mostly
regarded as a foundation for future work on the integration of the
scheduler with CPU power management (in fact, there is work in
progress on top of it already). Nevertheless it works and the
preliminary results obtained with it are encouraging.There also is some consolidation of CPU frequency management for ARM
platforms that can add their machine IDs the the new stub dt-platdev
driver now and that will take care of creating the requisite platform
device for cpufreq-dt, so it is not necessary to do that in platform
code any more. Several ARM platforms are switched over to using this
generic mechanism.In addition to that, the intel_pstate driver is now going to respect
CPU frequency limits set by the platform firmware (or a BMC) and
provided via the ACPI _PPC object.The devfreq subsystem is getting a new "passive" governor for SoCs
subsystems that will depend on somebody else to manage their voltage
rails and its support for Samsung Exynos SoCs is consolidated.The rest is support for new hardware (Intel Broxton support in
intel_idle for one example), bug fixes, optimizations and cleanups in
a number of places.Specifics:
- New cpufreq "schedutil" governor (making decisions based on CPU
utilization information provided by the scheduler and capable of
switching CPU frequencies right away if the underlying driver
supports that) and support for fast frequency switching in the
acpi-cpufreq driver (Rafael Wysocki)- Consolidation of CPU frequency management on ARM platforms allowing
them to get rid of some platform-specific boilerplate code if they
are going to use the cpufreq-dt driver (Viresh Kumar, Finley Xiao,
Marc Gonzalez)- Support for ACPI _PPC and CPU frequency limits in the intel_pstate
driver (Srinivas Pandruvada)- Fixes and cleanups in the cpufreq core and generic governor code
(Rafael Wysocki, Sai Gurrappadi)- intel_pstate driver optimizations and cleanups (Rafael Wysocki,
Philippe Longepe, Chen Yu, Joe Perches)- cpufreq powernv driver fixes and cleanups (Akshay Adiga, Shilpasri
Bhat)- cpufreq qoriq driver fixes and cleanups (Jia Hongtao)
- ACPI cpufreq driver cleanups (Viresh Kumar)
- Assorted cpufreq driver updates (Ashwin Chaugule, Geliang Tang,
Javier Martinez Canillas, Paul Gortmaker, Sudeep Holla)- Assorted cpufreq fixes and cleanups (Joe Perches, Arnd Bergmann)
- Fixes and cleanups in the OPP (Operating Performance Points)
framework, mostly related to OPP sharing, and reorganization of
OF-dependent code in it (Viresh Kumar, Arnd Bergmann, Sudeep Holla)- New "passive" governor for devfreq (for SoC subsystems that will
rely on someone else for the management of their power resources)
and consolidation of devfreq support for Exynos platforms, coding
style and typo fixes for devfreq (Chanwoo Choi, MyungJoo Ham)- PM core fixes and cleanups, mostly to make it work better with the
generic power domains (genpd) framework, and updates for that
framework (Ulf Hansson, Thierry Reding, Colin Ian King)- Intel Broxton support for the intel_idle driver (Len Brown)
- cpuidle core optimization and fix (Daniel Lezcano, Dave Gerlach)
- ARM cpuidle cleanups (Jisheng Zhang)
- Intel Kabylake support for the RAPL power capping driver (Jacob
Pan)- AVS (Adaptive Voltage Switching) rockchip-io driver update (Heiko
Stuebner)- Updates for the cpupower tool (Arjun Sreedharan, Colin Ian King,
Mattia Dongili, Thomas Renninger)"* tag 'pm-4.7-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm: (112 commits)
intel_pstate: Clean up get_target_pstate_use_performance()
intel_pstate: Use sample.core_avg_perf in get_avg_pstate()
intel_pstate: Clarify average performance computation
intel_pstate: Avoid unnecessary synchronize_sched() during initialization
cpufreq: schedutil: Make default depend on CONFIG_SMP
cpufreq: powernv: del_timer_sync when global and local pstate are equal
cpufreq: powernv: Move smp_call_function_any() out of irq safe block
intel_pstate: Clean up intel_pstate_get()
cpufreq: schedutil: Make it depend on CONFIG_SMP
cpufreq: governor: Fix handling of special cases in dbs_update()
PM / OPP: Move CONFIG_OF dependent code in a separate file
cpufreq: intel_pstate: Ignore _PPC processing under HWP
cpufreq: arm_big_little: use generic OPP functions for {init, free}_opp_table
PM / OPP: add non-OF versions of dev_pm_opp_{cpumask_, }remove_table
cpufreq: tango: Use generic platdev driver
PM / OPP: pass cpumask by reference
cpufreq: Fix GOV_LIMITS handling for the userspace governor
cpupower: fix potential memory leak
PM / devfreq: style/typo fixes
PM / devfreq: exynos: Add the detailed correlation for Exynos5422 bus
..
28 Apr, 2016
1 commit
-
Skylake processor supports a new set of RAPL registers for controlling
entire SoC instead of just CPU package. This is useful for thermal
and power control when source of power/thermal is not just CPU/GPU.
This change adds a new platform domain (AKA PSys) to the current
power capping Intel RAPL driver.PSys also supports PL1 (long term) and PL2 (short term) control like
package domain. This also follows same MSRs for energy and time
units as package domain.Unlike package domain, PSys support requires more than just processor
level implementation. The other parts in the system need additional
implementation, which OEMs needs to support. So not all Skylake
systems will support PSys.Signed-off-by: Srinivas Pandruvada
Signed-off-by: Peter Zijlstra (Intel)
Acked-by: Rafael J. Wysocki
Cc: Alexander Shishkin
Cc: Arnaldo Carvalho de Melo
Cc: Jiri Olsa
Cc: Peter Zijlstra
Cc: Stephane Eranian
Cc: Thomas Gleixner
Cc: Vince Weaver
Cc: bp@alien8.de
Cc: hpa@zytor.com
Cc: jacob.jun.pan@linux.intel.com
Cc: rjw@rjwysocki.net
Link: http://lkml.kernel.org/r/1460930581-29748-3-git-send-email-srinivas.pandruvada@linux.intel.com
Signed-off-by: Ingo Molnar
27 Apr, 2016
1 commit
-
Kabylake is similar to Skylake in terms of RAPL.
Signed-off-by: Jacob Pan
Signed-off-by: Rafael J. Wysocki
05 Apr, 2016
1 commit
-
Added missing model 0x46.
Tested-and-reported-by: Piotr Maksymiuk
Signed-off-by: Srinivas Pandruvada
Signed-off-by: Rafael J. Wysocki
09 Mar, 2016
3 commits
-
RAPL driver operates on MSRs that are under package/socket
scope instead of core scope. However, the current code does not
keep track of which CPUs are available on each package for MSR
access. Therefore it has to search for an active CPU on a given
package each time.This patch optimizes the package level operations by tracking a
per package lead CPU during initialization and CPU hotplug. The
runtime search for active CPU is avoided.Suggested-by: Thomas Gleixner
Signed-off-by: Jacob Pan
Signed-off-by: Rafael J. Wysocki -
This patch adds to each rapl domain a reference of the package
it belongs to. At runtime, we can then avoid searching the package
data for each access. It simplifies the domain level operations
which depend on package level information.Signed-off-by: Jacob Pan
Signed-off-by: Rafael J. Wysocki -
Reduce remote CPU calls for MSR access by combining read
modify write into one function.Suggested-by: Peter Zijlstra
Signed-off-by: Jacob Pan
Signed-off-by: Rafael J. Wysocki
12 Jan, 2016
2 commits
-
* powercap:
powercap: constify powercap_zone_ops and powercap_zone_constraint_ops structures -
* acpi-soc:
PM / clk: don't leave clocks enabled when driver not bound
i2c: dw: Add APM X-Gene ACPI I2C device support
ACPI / APD: Add APM X-Gene ACPI I2C device support
ACPI / LPSS: change 'does not have' to 'has' in comment
Revert "dmaengine: dw: platform: provide platform data for Intel"
dmaengine: dw: return immediately from IRQ when DMA isn't in use
dmaengine: dw: platform: power on device on shutdown
ACPI / LPSS: override power state for LPSS DMA device
ACPI / LPSS: power on when probe() and otherwise when remove()
ACPI / LPSS: do delay for all LPSS devices when D3->D0
ACPI / LPSS: allow to use specific PM domain during ->probe()
Revert "ACPI / LPSS: allow to use specific PM domain during ->probe()"
device core: add BUS_NOTIFY_DRIVER_NOT_BOUND notification
x86/platform/iosf_mbi: Remove duplicate definitionsConflicts:
drivers/i2c/busses/i2c-designware-platdrv.c
02 Jan, 2016
1 commit
-
The powercap_zone_ops and powercap_zone_constraint_ops structures are never
modified, so declare them as const.Most of the actual changes adjust indentation to accomodate the const
keyword.Done with the help of Coccinelle.
Signed-off-by: Julia Lawall
Signed-off-by: Rafael J. Wysocki
12 Dec, 2015
1 commit
-
Intel RAPL initialized on several systems where the BIOS lock bit (msr
0x610, bit 63) was set. This occured because the return value of
rapl_read_data_raw() was being checked, rather than the value of the variable
passed in, locked.This patch properly implments the rapl_read_data_raw() call to check the
variable locked, and now the Intel RAPL driver outputs the warning:intel_rapl: RAPL package 0 domain package locked by BIOS
and does not initialize for the package.
Signed-off-by: Prarit Bhargava
Acked-by: Jacob Pan
Cc: All applicable
Signed-off-by: Rafael J. Wysocki
09 Dec, 2015
1 commit
-
The read and write opcodes are global for all units on SoC and even across
Intel SoCs. Remove duplication of corresponding constants. At the same time
convert all current users.No functional change.
Acked-by: Thomas Gleixner
Acked-by: Boon Leong Ong
Acked-by: Jacob Pan
Signed-off-by: Andy Shevchenko
Signed-off-by: Rafael J. Wysocki
17 Oct, 2015
1 commit
-
Broxton CPU ID for the RAPL driver.
Signed-off-by: Amy Wiles
[ rjw: Subject and changelog. ]
Signed-off-by: Rafael J. Wysocki
29 Aug, 2015
1 commit
-
Let rapl_unregister_powercap() disable the second power limit
only if it exists.
Intel64 SDM Vol.3 14.9 says that the package domain has it
but neither the power plane domain nor the DRAM domain has it.Signed-off-by: Seiichi Ikarashi
Signed-off-by: Rafael J. Wysocki
01 Aug, 2015
2 commits
-
This patch enabled RAPL to support Broadwell-H
Signed-off-by: Radivoje Jovanovic
Signed-off-by: Rafael J. Wysocki -
This patche enabled RAPL to support Intel Skylake H/S
Signed-off-by: Radivoje Jovanovic
Signed-off-by: Rafael J. Wysocki
21 May, 2015
1 commit
-
This patch enables intel_rapl driver to run on the
next-generation Intel(R) Xeon Phi Microarchitecture
code named "Knights Landing"Signed-off-by: Dasaratharaman Chandramouli
Acked-by: Jacob Pan
Signed-off-by: Rafael J. Wysocki
05 May, 2015
1 commit
-
CPU Floor frequency is set in BIOS for newer Atom
SoCs. This patch handles configuration of floor
frequency for different variants of Atom SoCs
appropriately and ensures configuration of floor
frequency is not done from driver for these
newer Atom SoCs.Since address of the register for configuring
floor frequency might change for different
Atom SoCs, this patch also prevents potential
overwriting of wrong registers.Reviewed-by: Jacob Pan
Signed-off-by: Ajay Thomas
Signed-off-by: Rafael J. Wysocki
16 Apr, 2015
1 commit
-
Signed-off-by: Brian Bian
Acked-by: Jacob Pan
Signed-off-by: Rafael J. Wysocki
11 Apr, 2015
2 commits
-
The RAPL ids are only tested in rapl_init() which is itself an __init
function. For the MODULE_DEVICE_TABLE() file2alias doesn't care about
the section, just about the symbol name. Therefore it's safe to mark the
rapl_ids[] array as __initconst so its memory can be released after
initialization is done.Signed-off-by: Mathias Krause
Acked-by: Jacob Pan
Signed-off-by: Rafael J. Wysocki -
Validated RAPL functions on this platform. Add support to driver.
Signed-off-by: Jacob Pan
Signed-off-by: Rafael J. Wysocki
