25 Mar, 2016
2 commits
-
Pull more power management and ACPI updates from Rafael Wysocki:
"The second batch of power management and ACPI updates for v4.6.Included are fixups on top of the previous PM/ACPI pull request and
other material that didn't make into it but still should go into 4.6.Among other things, there's a fix for an intel_pstate driver issue
uncovered by recent cpufreq changes, a workaround for a boot hang on
Skylake-H related to the handling of deep C-states by the platform and
a PCI/ACPI fix for the handling of IO port resources on non-x86
architectures plus some new device IDs and similar.Specifics:
- Fix for an intel_pstate driver issue related to the handling of MSR
updates uncovered by the recent cpufreq rework (Rafael Wysocki).- cpufreq core cleanups related to starting governors and frequency
synchronization during resume from system suspend and a locking fix
for cpufreq_quick_get() (Rafael Wysocki, Richard Cochran).- acpi-cpufreq and powernv cpufreq driver updates (Jisheng Zhang,
Michael Neuling, Richard Cochran, Shilpasri Bhat).- intel_idle driver update preventing some Skylake-H systems from
hanging during initialization by disabling deep C-states mishandled
by the platform in the problematic configurations (Len Brown).- Intel Xeon Phi Processor x200 support for intel_idle
(Dasaratharaman Chandramouli).- cpuidle menu governor updates to make it always honor PM QoS
latency constraints (and prevent C1 from being used as the fallback
C-state on x86 when they are set below its exit latency) and to
restore the previous behavior to fall back to C1 if the next timer
event is set far enough in the future that was changed in 4.4 which
led to an energy consumption regression (Rik van Riel, Rafael
Wysocki).- New device ID for a future AMD UART controller in the ACPI driver
for AMD SoCs (Wang Hongcheng).- Rockchip rk3399 support for the rockchip-io-domain adaptive voltage
scaling (AVS) driver (David Wu).- ACPI PCI resources management fix for the handling of IO space
resources on architectures where the IO space is memory mapped
(IA64 and ARM64) broken by the introduction of common ACPI
resources parsing for PCI host bridges in 4.4 (Lorenzo Pieralisi).- Fix for the ACPI backend of the generic device properties API to
make it parse non-device (data node only) children of an ACPI
device correctly (Irina Tirdea).- Fixes for the handling of global suspend flags (introduced in 4.4)
during hibernation and resume from it (Lukas Wunner).- Support for obtaining configuration information from Device Trees
in the PM clocks framework (Jon Hunter).- ACPI _DSM helper code and devfreq framework cleanups (Colin Ian
King, Geert Uytterhoeven)"* tag 'pm+acpi-4.6-rc1-2' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm: (23 commits)
PM / AVS: rockchip-io: add io selectors and supplies for rk3399
intel_idle: Support for Intel Xeon Phi Processor x200 Product Family
intel_idle: prevent SKL-H boot failure when C8+C9+C10 enabled
ACPI / PM: Runtime resume devices when waking from hibernate
PM / sleep: Clear pm_suspend_global_flags upon hibernate
cpufreq: governor: Always schedule work on the CPU running update
cpufreq: Always update current frequency before startig governor
cpufreq: Introduce cpufreq_update_current_freq()
cpufreq: Introduce cpufreq_start_governor()
cpufreq: powernv: Add sysfs attributes to show throttle stats
cpufreq: acpi-cpufreq: make Intel/AMD MSR access, io port access static
PCI: ACPI: IA64: fix IO port generic range check
ACPI / util: cast data to u64 before shifting to fix sign extension
cpufreq: powernv: Define per_cpu chip pointer to optimize hot-path
cpuidle: menu: Fall back to polling if next timer event is near
cpufreq: acpi-cpufreq: Clean up hot plug notifier callback
intel_pstate: Do not call wrmsrl_on_cpu() with disabled interrupts
cpufreq: Make cpufreq_quick_get() safe to call
ACPI / property: fix data node parsing in acpi_get_next_subnode()
ACPI / APD: Add device HID for future AMD UART controller
... -
* pm-cpufreq:
cpufreq: governor: Always schedule work on the CPU running update
cpufreq: Always update current frequency before startig governor
cpufreq: Introduce cpufreq_update_current_freq()
cpufreq: Introduce cpufreq_start_governor()
cpufreq: powernv: Add sysfs attributes to show throttle stats
cpufreq: acpi-cpufreq: make Intel/AMD MSR access, io port access static
cpufreq: powernv: Define per_cpu chip pointer to optimize hot-path
cpufreq: acpi-cpufreq: Clean up hot plug notifier callback
intel_pstate: Do not call wrmsrl_on_cpu() with disabled interrupts
cpufreq: Make cpufreq_quick_get() safe to call* pm-cpuidle:
intel_idle: Support for Intel Xeon Phi Processor x200 Product Family
intel_idle: prevent SKL-H boot failure when C8+C9+C10 enabled
cpuidle: menu: Fall back to polling if next timer event is near
cpuidle: menu: use high confidence factors only when considering polling
23 Mar, 2016
6 commits
-
Modify dbs_irq_work() to always schedule the process-context work
on the current CPU which also ran the dbs_update_util_handler()
that the irq_work being handled came from.This causes the entire frequency update handling (involving the
"ondemand" or "conservative" governors) to be carried out by the
CPU whose frequency is to be updated and reduces the overall amount
of inter-CPU noise related to cpufreq.Signed-off-by: Rafael J. Wysocki
-
Make policy->cur match the current frequency returned by the driver's
->get() callback before starting the governor in case they went out of
sync in the meantime and drop the piece of code attempting to
resync policy->cur with the real frequency of the boot CPU from
cpufreq_resume() as it serves no purpose any more (and it's racy and
super-ugly anyway).Signed-off-by: Rafael J. Wysocki
Acked-by: Viresh Kumar -
Move the part of cpufreq_update_policy() that obtains the current
frequency from the driver and updates policy->cur if necessary to
a separate function, cpufreq_get_current_freq().That should not introduce functional changes and subsequent change
set will need it.Signed-off-by: Rafael J. Wysocki
Acked-by: Viresh Kumar -
Starting a governor in cpufreq always follows the same pattern
involving two calls to cpufreq_governor(), one with the event
argument set to CPUFREQ_GOV_START and one with that argument set to
CPUFREQ_GOV_LIMITS.Introduce cpufreq_start_governor() that will carry out those two
operations and make all places where governors are started use it.That slightly modifies the behavior of cpufreq_set_policy() which
now also will go back to the old governor if the second call to
cpufreq_governor() (the one with event equal to CPUFREQ_GOV_LIMITS)
fails, but that really is how it should work in the first place.Also cpufreq_resume() will now pring an error message if the
CPUFREQ_GOV_LIMITS call to cpufreq_governor() fails, but that
makes it follow cpufreq_add_policy_cpu() and cpufreq_offline()
in that respect.Signed-off-by: Rafael J. Wysocki
Acked-by: Viresh Kumar -
Create sysfs attributes to export throttle information in
/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats directory. The
newly added sysfs files are as follows:1)/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/turbo_stat
2)/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/sub-turbo_stat
3)/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/unthrottle
4)/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/powercap
5)/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/overtemp
6)/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/supply_fault
7)/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/overcurrent
8)/sys/devices/system/cpu/cpuX/cpufreq/throttle_stats/occ_resetDetailed explanation of each attribute is added to
Documentation/ABI/testing/sysfs-devices-system-cpuSigned-off-by: Shilpasri G Bhat
Acked-by: Viresh Kumar
Signed-off-by: Rafael J. Wysocki -
These frequency register read/write operations' implementations for the
given processor (Intel/AMD MSR access or I/O port access) are only used
internally in acpi-cpufreq, so make them static.Signed-off-by: Jisheng Zhang
Acked-by: Viresh Kumar
Signed-off-by: Rafael J. Wysocki
22 Mar, 2016
1 commit
-
Commit 96c4726f01cd "cpufreq: powernv: Remove cpu_to_chip_id() from
hot-path" introduced a 'core_to_chip_map' array to cache the chip-ids
of all cores.Replace this with a per-CPU variable that stores the pointer to the
chip-array. This removes the linear lookup and provides a neater and
simpler solution.Signed-off-by: Michael Neuling
Signed-off-by: Shilpasri G Bhat
Acked-by: Viresh Kumar
Signed-off-by: Rafael J. Wysocki
21 Mar, 2016
2 commits
-
Pull ARM SoC driver updates from Arnd Bergmann:
"Driver updates for ARM SoCs, these contain various things that touch
the drivers/ directory but got merged through arm-soc for practical
reasons:- Rockchip rk3368 gains power domain support
- Small updates for the ARM spmi driver
- The Atmel PMC driver saw a larger rework, touching both
arch/arm/mach-at91 and drivers/clk/at91
- All reset controller driver changes alway get merged through
arm-soc, though this time the largest change is the addition of a
MIPS pistachio reset driver
- One bugfix for the NXP (formerly Freescale) i.MX weim bus driver"* tag 'armsoc-drivers' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc: (43 commits)
bus: imx-weim: Take the 'status' property value into account
clk: at91: remove useless includes
clk: at91: pmc: remove useless capacities handling
clk: at91: pmc: drop at91_pmc_base
usb: gadget: atmel: access the PMC using regmap
ARM: at91: remove useless includes and function prototypes
ARM: at91: pm: move idle functions to pm.c
ARM: at91: pm: find and remap the pmc
ARM: at91: pm: simply call at91_pm_init
clk: at91: pmc: move pmc structures to C file
clk: at91: pmc: merge at91_pmc_init in atmel_pmc_probe
clk: at91: remove IRQ handling and use polling
clk: at91: make use of syscon/regmap internally
clk: at91: make use of syscon to share PMC registers in several drivers
hwmon: (scpi) add energy meter support
firmware: arm_scpi: add support for 64-bit sensor values
firmware: arm_scpi: decrease Tx timeout to 20ms
firmware: arm_scpi: fix send_message and sensor_get_value for big-endian
reset: sti: Make reset_control_ops const
reset: zynq: Make reset_control_ops const
... -
Pull ARM SoC platform updates from Arnd Bergmann:
"Newly added support for additional SoCs:
- Axis Artpec-6 SoC family
- Allwinner A83T SoC
- Mediatek MT7623
- NXP i.MX6QP SoC
- ST Microelectronics stm32f469 microcontrollerNew features:
- SMP support for Mediatek mt2701
- Big-endian support for NXP i.MX
- DaVinci now uses the new DMA engine dma_slave_map
- OMAP now uses the new DMA engine dma_slave_map
- earlyprintk support for palmchip uart on mach-tango
- delay timer support for orionOther:
- Exynos PMU driver moved out to drivers/soc/
- Various smaller updates for Renesas, Xilinx, PXA, AT91, OMAP,
uniphier"* tag 'armsoc-soc' of git://git.kernel.org/pub/scm/linux/kernel/git/arm/arm-soc: (83 commits)
ARM: uniphier: rework SMP code to support new System Bus binding
ARM: uniphier: add missing of_node_put()
ARM: at91: avoid defining CONFIG_* symbols in source code
ARM: DRA7: hwmod: Add data for eDMA tpcc, tptc0, tptc1
ARM: imx: Make reset_control_ops const
ARM: imx: Do L2 errata only if the L2 cache isn't enabled
ARM: imx: select ARM_CPU_SUSPEND only for imx6
dmaengine: pxa_dma: fix the maximum requestor line
ARM: alpine: select the Alpine MSI controller driver
ARM: pxa: add the number of DMA requestor lines
dmaengine: mmp-pdma: add number of requestors
dma: mmp_pdma: Add the #dma-requests DT property documentation
ARM: OMAP2+: Add rtc hwmod configuration for ti81xx
ARM: s3c24xx: Avoid warning for inb/outb
ARM: zynq: Move early printk virtual address to vmalloc area
ARM: DRA7: hwmod: Add custom reset handler for PCIeSS
ARM: SAMSUNG: Remove unused register offset definition
ARM: EXYNOS: Cleanup header files inclusion
drivers: soc: samsung: Enable COMPILE_TEST
MAINTAINERS: Add maintainers entry for drivers/soc/samsung
...
20 Mar, 2016
2 commits
-
This driver has two issues. First, it tries to fiddle with the hot
plugged CPU's MSR on the UP_PREPARE event, at a time when the CPU is
not yet online. Second, the driver sets the "boost-disable" bit for a
CPU when going down, but does not clear the bit again if the CPU comes
up again due to DOWN_FAILED.This patch fixes the issues by changing the driver to react to the
ONLINE/DOWN_FAILED events instead of UP_PREPARE. As an added benefit,
the driver also becomes symmetric with respect to the hot plug
mechanism.Signed-off-by: Richard Cochran
Signed-off-by: Rafael J. Wysocki -
After commit a4675fbc4a7a (cpufreq: intel_pstate: Replace timers with
utilization update callbacks) wrmsrl_on_cpu() cannot be called in the
intel_pstate_adjust_busy_pstate() path as that is executed with
disabled interrupts. However, atom_set_pstate() called from there
via intel_pstate_set_pstate() uses wrmsrl_on_cpu() to update the
IA32_PERF_CTL MSR which triggers the WARN_ON_ONCE() in
smp_call_function_single().The reason why wrmsrl_on_cpu() is used by atom_set_pstate() is
because intel_pstate_set_pstate() calling it is also invoked during
the initialization and cleanup of the driver and in those cases it is
not guaranteed to be run on the CPU that is being updated. However,
in the case when intel_pstate_set_pstate() is called by
intel_pstate_adjust_busy_pstate(), wrmsrl() can be used to update
the register safely. Moreover, intel_pstate_set_pstate() already
contains code that only is executed if the function is called by
intel_pstate_adjust_busy_pstate() and there is a special argument
passed to it because of that.To fix the problem at hand, rearrange the code taking the above
observations into account.First, replace the ->set() callback in struct pstate_funcs with a
->get_val() one that will return the value to be written to the
IA32_PERF_CTL MSR without updating the register.Second, split intel_pstate_set_pstate() into two functions,
intel_pstate_update_pstate() to be called by
intel_pstate_adjust_busy_pstate() that will contain all of the
intel_pstate_set_pstate() code which only needs to be executed in
that case and will use wrmsrl() to update the MSR (after obtaining
the value to write to it from the ->get_val() callback), and
intel_pstate_set_min_pstate() to be invoked during the
initialization and cleanup that will set the P-state to the
minimum one and will update the MSR using wrmsrl_on_cpu().Finally, move the code shared between intel_pstate_update_pstate()
and intel_pstate_set_min_pstate() to a new static inline function
intel_pstate_record_pstate() and make them both call it.Of course, that unifies the handling of the IA32_PERF_CTL MSR writes
between Atom and Core.Fixes: a4675fbc4a7a (cpufreq: intel_pstate: Replace timers with utilization update callbacks)
Reported-and-tested-by: Josh Boyer
Signed-off-by: Rafael J. Wysocki
18 Mar, 2016
1 commit
-
The function, cpufreq_quick_get, accesses the global 'cpufreq_driver' and
its fields without taking the associated lock, cpufreq_driver_lock.Without the locking, nothing guarantees that 'cpufreq_driver' remains
consistent during the call. This patch fixes the issue by taking the lock
before accessing the data structure.Signed-off-by: Richard Cochran
Acked-by: Viresh Kumar
Signed-off-by: Rafael J. Wysocki
14 Mar, 2016
1 commit
-
* pm-cpufreq: (94 commits)
intel_pstate: Do not skip samples partially
intel_pstate: Remove freq calculation from intel_pstate_calc_busy()
intel_pstate: Move intel_pstate_calc_busy() into get_target_pstate_use_performance()
intel_pstate: Optimize calculation for max/min_perf_adj
intel_pstate: Remove extra conversions in pid calculation
cpufreq: Move scheduler-related code to the sched directory
Revert "cpufreq: postfix policy directory with the first CPU in related_cpus"
cpufreq: Reduce cpufreq_update_util() overhead a bit
cpufreq: Select IRQ_WORK if CPU_FREQ_GOV_COMMON is set
cpufreq: Remove 'policy->governor_enabled'
cpufreq: Rename __cpufreq_governor() to cpufreq_governor()
cpufreq: Relocate handle_update() to kill its declaration
cpufreq: governor: Drop unnecessary checks from show() and store()
cpufreq: governor: Fix race in dbs_update_util_handler()
cpufreq: governor: Make gov_set_update_util() static
cpufreq: governor: Narrow down the dbs_data_mutex coverage
cpufreq: governor: Make dbs_data_mutex static
cpufreq: governor: Relocate definitions of tuners structures
cpufreq: governor: Move per-CPU data to the common code
cpufreq: governor: Make governor private data per-policy
...
11 Mar, 2016
7 commits
-
If the current value of MPERF or the current value of TSC is the
same as the previous one, respectively, intel_pstate_sample() bails
out early and skips the sample.However, intel_pstate_adjust_busy_pstate() is still called in that
case which is not correct, so modify intel_pstate_sample() to
return a bool value indicating whether or not the sample has been
taken and use it to decide whether or not to call
intel_pstate_adjust_busy_pstate().While at it, remove redundant parentheses from the MPERF/TSC
check in intel_pstate_sample().Signed-off-by: Rafael J. Wysocki
Acked-by: Srinivas Pandruvada -
Use a helper function to compute the average pstate and call it only
where it is needed (only when tracing or in intel_pstate_get).Signed-off-by: Philippe Longepe
Acked-by: Srinivas Pandruvada
Signed-off-by: Rafael J. Wysocki -
The cpu_load algorithm doesn't need to invoke intel_pstate_calc_busy(),
so move that call from intel_pstate_sample() to
get_target_pstate_use_performance().Signed-off-by: Philippe Longepe
Acked-by: Srinivas Pandruvada
Signed-off-by: Rafael J. Wysocki -
mul_fp(int_tofp(A), B) expands to:
((A << FRAC_BITS) * B) >> FRAC_BITS, so the same result can be obtained
via simple multiplication A * B. Apply this observation to
max_perf * limits->max_perf and max_perf * limits->min_perf in
intel_pstate_get_min_max()."Signed-off-by: Philippe Longepe
Acked-by: Srinivas Pandruvada
Signed-off-by: Rafael J. Wysocki -
pid->setpoint and pid->deadband can be initialized in fixed point, so we
can avoid the int_tofp in pid_calc.Signed-off-by: Philippe Longepe
Signed-off-by: Rafael J. Wysocki -
Create cpufreq.c under kernel/sched/ and move the cpufreq code
related to the scheduler to that file and to sched.h.Redefine cpufreq_update_util() as a static inline function to avoid
function calls at its call sites in the scheduler code (as suggested
by Peter Zijlstra).Also move the definition of struct update_util_data and declaration
of cpufreq_set_update_util_data() from include/linux/cpufreq.h to
include/linux/sched.h.Signed-off-by: Rafael J. Wysocki
Acked-by: Peter Zijlstra (Intel)
10 Mar, 2016
1 commit
-
Revert commit 3510fac45492 (cpufreq: postfix policy directory with the
first CPU in related_cpus).Earlier, the policy->kobj was added to the kobject core, before ->init()
callback was called for the cpufreq drivers. Which allowed those drivers
to add or remove, driver dependent, sysfs files/directories to the same
kobj from their ->init() and ->exit() callbacks.That isn't possible anymore after commit 3510fac45492.
Now, there is no other clean alternative that people can adopt.
Its better to revert the earlier commit to allow cpufreq drivers to
create/remove sysfs files from ->init() and ->exit() callbacks.Signed-off-by: Viresh Kumar
Signed-off-by: Rafael J. Wysocki
09 Mar, 2016
17 commits
-
Use the observation that cpufreq_update_util() is only called
by the scheduler with rq->lock held, so the callers of
cpufreq_set_update_util_data() can use synchronize_sched()
instead of synchronize_rcu() to wait for cpufreq_update_util()
to complete. Moreover, if they are updated to do that,
rcu_read_(un)lock() calls in cpufreq_update_util() might be
replaced with rcu_read_(un)lock_sched(), respectively, but
those aren't really necessary, because the scheduler calls
that function from RCU-sched read-side critical sections
already.In addition to that, if cpufreq_set_update_util_data() checks
the func field in the struct update_util_data before setting
the per-CPU pointer to it, the data->func check may be dropped
from cpufreq_update_util() as well.Make the above changes to reduce the overhead from
cpufreq_update_util() in the scheduler paths invoking it
and to make the cleanup after removing its callbacks less
heavy-weight somewhat.Signed-off-by: Rafael J. Wysocki
Acked-by: Viresh Kumar
Acked-by: Peter Zijlstra (Intel) -
Commit 0eb463be3436 (cpufreq: governor: Replace timers with utilization
update callbacks) made CPU_FREQ select IRQ_WORK, but that's not
necessary, as it is sufficient for IRQ_WORK to be selected by
CPU_FREQ_GOV_COMMON, so modify the cpufreq Kconfig to that effect.Signed-off-by: Rafael J. Wysocki
Acked-by: Viresh Kumar -
The entire sequence of events (like INIT/START or STOP/EXIT) for which
cpufreq_governor() is called, is guaranteed to be protected by
policy->rwsem now.The additional checks that were added earlier (as we were forced to drop
policy->rwsem before calling cpufreq_governor() for EXIT event), aren't
required anymore.Over that, they weren't sufficient really. They just take care of
START/STOP events, but not INIT/EXIT and the state machine was never
maintained properly by them.Kill the unnecessary checks and policy->governor_enabled field.
Signed-off-by: Viresh Kumar
Signed-off-by: Rafael J. Wysocki -
The __ at the beginning of the routine aren't really necessary at all.
Rename it to cpufreq_governor() instead.Signed-off-by: Viresh Kumar
Signed-off-by: Rafael J. Wysocki -
handle_update() is declared at the top of the file as its user appear
before its definition. Relocate the routine to get rid of this.Signed-off-by: Viresh Kumar
Signed-off-by: Rafael J. Wysocki -
The show() and store() routines in the cpufreq-governor core don't need
to check if the struct governor_attr they want to use really provides
the callbacks they need as expected (if that's not the case, it means a
bug in the code anyway), so change them to avoid doing that.Also change the error value to -EBUSY, if the governor is getting
removed and we aren't allowed to store any more changes.Signed-off-by: Viresh Kumar
Signed-off-by: Rafael J. Wysocki -
There is a scenario that may lead to undesired results in
dbs_update_util_handler(). Namely, if two CPUs sharing a policy
enter the funtion at the same time, pass the sample delay check
and then one of them is stalled until dbs_work_handler() (queued
up by the other CPU) clears the work counter, it may update the
work counter and queue up another work item prematurely.To prevent that from happening, use the observation that the CPU
queuing up a work item in dbs_update_util_handler() updates the
last sample time. This means that if another CPU was stalling after
passing the sample delay check and now successfully updated the work
counter as a result of the race described above, it will see the new
value of the last sample time which is different from what it used in
the sample delay check before. If that happens, the sample delay
check passed previously is not valid any more, so the CPU should not
continue.Fixes: f17cbb53783c (cpufreq: governor: Avoid atomic operations in hot paths)
Signed-off-by: Rafael J. Wysocki
Acked-by: Viresh Kumar -
The gov_set_update_util() routine is only used internally by the
common governor code and it doesn't need to be exported, so make
it static.No functional changes.
Signed-off-by: Rafael J. Wysocki
Acked-by: Viresh Kumar -
Since cpufreq_governor_dbs() is now always called with policy->rwsem
held, it cannot be executed twice in parallel for the same policy.
Thus it is not necessary to hold dbs_data_mutex around the invocations
of cpufreq_governor_start/stop/limits() from it as those functions
never modify any data that can be shared between different policies.However, cpufreq_governor_dbs() may be executed twice in parallal
for different policies using the same gov->gdbs_data object and
dbs_data_mutex is still necessary to protect that object against
concurrent updates.For this reason, narrow down the dbs_data_mutex locking to
cpufreq_governor_init/exit() where it is needed and rename the
mutex to gov_dbs_data_mutex to reflect its purpose.Signed-off-by: Rafael J. Wysocki
Acked-by: Viresh Kumar -
That mutex is only used by cpufreq_governor_dbs() and it doesn't
need to be exported to modules, so make it static and drop the
export incantation.No functional changes.
Signed-off-by: Rafael J. Wysocki
Acked-by: Viresh Kumar -
Move the definitions of struct od_dbs_tuners and struct cs_dbs_tuners
from the common governor header to the ondemand and conservative
governor code, respectively, as they don't need to be in the common
header any more.No functional changes.
Signed-off-by: Rafael J. Wysocki
Acked-by: Viresh Kumar -
After previous changes there is only one piece of code in the
ondemand governor making references to per-CPU data structures,
but it can be easily modified to avoid doing that, so modify it
accordingly and move the definition of per-CPU data used by the
ondemand and conservative governors to the common code. Next,
change that code to access the per-CPU data structures directly
rather than via a governor callback.This causes the ->get_cpu_cdbs governor callback to become
unnecessary, so drop it along with the macro and function
definitions related to it.Finally, drop the definitions of struct od_cpu_dbs_info_s and
struct cs_cpu_dbs_info_s that aren't necessary any more.Signed-off-by: Rafael J. Wysocki
Acked-by: Viresh Kumar -
Some fields in struct od_cpu_dbs_info_s and struct cs_cpu_dbs_info_s
are only used for a limited set of CPUs. Namely, if a policy is
shared between multiple CPUs, those fields will only be used for one
of them (policy->cpu). This means that they really are per-policy
rather than per-CPU and holding room for them in per-CPU data
structures is generally wasteful. Also moving those fields into
per-policy data structures will allow some significant simplifications
to be made going forward.For this reason, introduce struct cs_policy_dbs_info and
struct od_policy_dbs_info to hold those fields. Define each of the
new structures as an extension of struct policy_dbs_info (such that
struct policy_dbs_info is embedded in each of them) and introduce
new ->alloc and ->free governor callbacks to allocate and free
those structures, respectively, such that ->alloc() will return
a pointer to the struct policy_dbs_info embedded in the allocated
data structure and ->free() will take that pointer as its argument.With that, modify the code accessing the data fields in question
in per-CPU data objects to look for them in the new structures
via the struct policy_dbs_info pointer available to it and drop
them from struct od_cpu_dbs_info_s and struct cs_cpu_dbs_info_s.Signed-off-by: Rafael J. Wysocki
Acked-by: Viresh Kumar -
The ondemand_powersave_bias_init() function used for resetting data
fields related to the powersave bias tunable of the ondemand governor
works by walking all of the online CPUs in the system and updating the
od_cpu_dbs_info_s structures for all of them.However, if governor tunables are per policy, the update should not
touch the CPUs that are not associated with the given dbs_data.Moreover, since the data fields in question are only ever used for
policy->cpu in each policy governed by ondemand, the update can be
limited to those specific CPUs.Rework the code to take the above observations into account.
Signed-off-by: Rafael J. Wysocki
Acked-by: Viresh Kumar -
The ->store() callbacks of some tunable sysfs attributes of the
ondemand and conservative governors trigger immediate updates of
the CPU load information for all CPUs "governed" by the given
dbs_data by walking the cpu_dbs_info structures for all online
CPUs in the system and updating them.This is questionable for two reasons. First, it may lead to a lot of
extra overhead on a system with many CPUs if the given dbs_data is
only associated with a few of them. Second, if governor tunables are
per-policy, the CPUs associated with the other sets of governor
tunables should not be updated.To address this issue, use the observation that in all of the places
in question the update operation may be carried out in the same way
(because all of the tunables involved are now located in struct
dbs_data and readily available to the common code) and make the
code in those places invoke the same (new) helper function that
will carry out the update correctly.That new function always checks the ignore_nice_load tunable value
and updates the CPUs' prev_cpu_nice data fields if that's set, which
wasn't done by the original code in store_io_is_busy(), but it
should have been done in there too.Signed-off-by: Rafael J. Wysocki
Acked-by: Viresh Kumar -
The ->powersave_bias_init_cpu callback in struct od_ops is only used
in one place and that invocation may be replaced with a direct call
to the function pointed to by that callback, so change the code
accordingly and drop the callback.Signed-off-by: Rafael J. Wysocki
Acked-by: Viresh Kumar -
After some previous changes, the ->get_cpu_dbs_info_s governor
callback and the "governor" field in struct dbs_governor (whose
value represents the governor type) are not used any more, so
drop them.Also drop the unused gov_ops field from struct dbs_governor.
No functional changes.
Signed-off-by: Rafael J. Wysocki
Acked-by: Viresh Kumar