diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2022-08-02 11:17:00 -0700 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2022-08-02 11:17:00 -0700 |
commit | a771ea6413c00cf4af0570745f2e27084d7e2376 (patch) | |
tree | 780a2ca2b837d0de3c1ac8b2f0d3e6475f1ee817 | |
parent | 8fa0db3a9b8ef310eb40cf91481141ece084f584 (diff) | |
parent | aa727b7b4b67dba2a6cfebdd98767c455c2b3d69 (diff) | |
download | linux-next-a771ea6413c00cf4af0570745f2e27084d7e2376.tar.gz |
Merge tag 'pm-5.20-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull power management updates from Rafael Wysocki:
"These are mostly minor improvements all over including new CPU IDs for
the Intel RAPL driver, an Energy Model rework to use micro-Watt as the
power unit, cpufreq fixes and cleanus, cpuidle updates, devfreq
updates, documentation cleanups and a new version of the pm-graph
suite of utilities.
Specifics:
- Make cpufreq_show_cpus() more straightforward (Viresh Kumar).
- Drop unnecessary CPU hotplug locking from store() used by cpufreq
sysfs attributes (Viresh Kumar).
- Make the ACPI cpufreq driver support the boost control interface on
Zhaoxin/Centaur processors (Tony W Wang-oc).
- Print a warning message on attempts to free an active cpufreq
policy which should never happen (Viresh Kumar).
- Fix grammar in the Kconfig help text for the loongson2 cpufreq
driver (Randy Dunlap).
- Use cpumask_var_t for an on-stack CPU mask in the ondemand cpufreq
governor (Zhao Liu).
- Add trace points for guest_halt_poll_ns grow/shrink to the haltpoll
cpuidle driver (Eiichi Tsukata).
- Modify intel_idle to treat C1 and C1E as independent idle states on
Sapphire Rapids (Artem Bityutskiy).
- Extend support for wakeirq to callback wrappers used during system
suspend and resume (Ulf Hansson).
- Defer waiting for device probe before loading a hibernation image
till the first actual device access to avoid possible deadlocks
reported by syzbot (Tetsuo Handa).
- Unify device_init_wakeup() for PM_SLEEP and !PM_SLEEP (Bjorn
Helgaas).
- Add Raptor Lake-P to the list of processors supported by the Intel
RAPL driver (George D Sworo).
- Add Alder Lake-N and Raptor Lake-P to the list of processors for
which Power Limit4 is supported in the Intel RAPL driver (Sumeet
Pawnikar).
- Make pm_genpd_remove() check genpd_debugfs_dir against NULL before
attempting to remove it (Hsin-Yi Wang).
- Change the Energy Model code to represent power in micro-Watts and
adjust its users accordingly (Lukasz Luba).
- Add new devfreq driver for Mediatek CCI (Cache Coherent
Interconnect) (Johnson Wang).
- Convert the Samsung Exynos SoC Bus bindings to DT schema of
exynos-bus.c (Krzysztof Kozlowski).
- Address kernel-doc warnings by adding the description for unused
function parameters in devfreq core (Mauro Carvalho Chehab).
- Use NULL to pass a null pointer rather than zero according to the
function propotype in imx-bus.c (Colin Ian King).
- Print error message instead of error interger value in
tegra30-devfreq.c (Dmitry Osipenko).
- Add checks to prevent setting negative frequency QoS limits for
CPUs (Shivnandan Kumar).
- Update the pm-graph suite of utilities to the latest revision 5.9
including multiple improvements (Todd Brandt).
- Drop pme_interrupt reference from the PCI power management
documentation (Mario Limonciello)"
* tag 'pm-5.20-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm: (27 commits)
powercap: RAPL: Add Power Limit4 support for Alder Lake-N and Raptor Lake-P
PM: QoS: Add check to make sure CPU freq is non-negative
PM: hibernate: defer device probing when resuming from hibernation
intel_idle: make SPR C1 and C1E be independent
cpufreq: ondemand: Use cpumask_var_t for on-stack cpu mask
cpufreq: loongson2: fix Kconfig "its" grammar
pm-graph v5.9
cpufreq: Warn users while freeing active policy
cpufreq: scmi: Support the power scale in micro-Watts in SCMI v3.1
firmware: arm_scmi: Get detailed power scale from perf
Documentation: EM: Switch to micro-Watts scale
PM: EM: convert power field to micro-Watts precision and align drivers
PM / devfreq: tegra30: Add error message for devm_devfreq_add_device()
PM / devfreq: imx-bus: use NULL to pass a null pointer rather than zero
PM / devfreq: shut up kernel-doc warnings
dt-bindings: interconnect: samsung,exynos-bus: convert to dtschema
PM / devfreq: mediatek: Introduce MediaTek CCI devfreq driver
dt-bindings: interconnect: Add MediaTek CCI dt-bindings
PM: domains: Ensure genpd_debugfs_dir exists before remove
PM: runtime: Extend support for wakeirq for force_suspend|resume
...
41 files changed, 1498 insertions, 827 deletions
diff --git a/Documentation/devicetree/bindings/devfreq/exynos-bus.txt b/Documentation/devicetree/bindings/devfreq/exynos-bus.txt deleted file mode 100644 index bcaa2c08ac11..000000000000 --- a/Documentation/devicetree/bindings/devfreq/exynos-bus.txt +++ /dev/null @@ -1,488 +0,0 @@ -* Generic Exynos Bus frequency device - -The Samsung Exynos SoC has many buses for data transfer between DRAM -and sub-blocks in SoC. Most Exynos SoCs share the common architecture -for buses. Generally, each bus of Exynos SoC includes a source clock -and a power line, which are able to change the clock frequency -of the bus in runtime. To monitor the usage of each bus in runtime, -the driver uses the PPMU (Platform Performance Monitoring Unit), which -is able to measure the current load of sub-blocks. - -The Exynos SoC includes the various sub-blocks which have the each AXI bus. -The each AXI bus has the owned source clock but, has not the only owned -power line. The power line might be shared among one more sub-blocks. -So, we can divide into two type of device as the role of each sub-block. -There are two type of bus devices as following: -- parent bus device -- passive bus device - -Basically, parent and passive bus device share the same power line. -The parent bus device can only change the voltage of shared power line -and the rest bus devices (passive bus device) depend on the decision of -the parent bus device. If there are three blocks which share the VDD_xxx -power line, Only one block should be parent device and then the rest blocks -should depend on the parent device as passive device. - - VDD_xxx |--- A block (parent) - |--- B block (passive) - |--- C block (passive) - -There are a little different composition among Exynos SoC because each Exynos -SoC has different sub-blocks. Therefore, such difference should be specified -in devicetree file instead of each device driver. In result, this driver -is able to support the bus frequency for all Exynos SoCs. - -Required properties for all bus devices: -- compatible: Should be "samsung,exynos-bus". -- clock-names : the name of clock used by the bus, "bus". -- clocks : phandles for clock specified in "clock-names" property. -- operating-points-v2: the OPP table including frequency/voltage information - to support DVFS (Dynamic Voltage/Frequency Scaling) feature. - -Required properties only for parent bus device: -- vdd-supply: the regulator to provide the buses with the voltage. -- devfreq-events: the devfreq-event device to monitor the current utilization - of buses. - -Required properties only for passive bus device: -- devfreq: the parent bus device. - -Optional properties only for parent bus device: -- exynos,saturation-ratio: the percentage value which is used to calibrate - the performance count against total cycle count. - -Optional properties for the interconnect functionality (QoS frequency -constraints): -- #interconnect-cells: should be 0. -- interconnects: as documented in ../interconnect.txt, describes a path at the - higher level interconnects used by this interconnect provider. - If this interconnect provider is directly linked to a top level interconnect - provider the property contains only one phandle. The provider extends - the interconnect graph by linking its node to a node registered by provider - pointed to by first phandle in the 'interconnects' property. - -- samsung,data-clock-ratio: ratio of the data throughput in B/s to minimum data - clock frequency in Hz, default value is 8 when this property is missing. - -Detailed correlation between sub-blocks and power line according to Exynos SoC: -- In case of Exynos3250, there are two power line as following: - VDD_MIF |--- DMC - - VDD_INT |--- LEFTBUS (parent device) - |--- PERIL - |--- MFC - |--- G3D - |--- RIGHTBUS - |--- PERIR - |--- FSYS - |--- LCD0 - |--- PERIR - |--- ISP - |--- CAM - -- In case of Exynos4210, there is one power line as following: - VDD_INT |--- DMC (parent device) - |--- LEFTBUS - |--- PERIL - |--- MFC(L) - |--- G3D - |--- TV - |--- LCD0 - |--- RIGHTBUS - |--- PERIR - |--- MFC(R) - |--- CAM - |--- FSYS - |--- GPS - |--- LCD0 - |--- LCD1 - -- In case of Exynos4x12, there are two power line as following: - VDD_MIF |--- DMC - - VDD_INT |--- LEFTBUS (parent device) - |--- PERIL - |--- MFC(L) - |--- G3D - |--- TV - |--- IMAGE - |--- RIGHTBUS - |--- PERIR - |--- MFC(R) - |--- CAM - |--- FSYS - |--- GPS - |--- LCD0 - |--- ISP - -- In case of Exynos5422, there are two power line as following: - VDD_MIF |--- DREX 0 (parent device, DRAM EXpress controller) - |--- DREX 1 - - VDD_INT |--- NoC_Core (parent device) - |--- G2D - |--- G3D - |--- DISP1 - |--- NoC_WCORE - |--- GSCL - |--- MSCL - |--- ISP - |--- MFC - |--- GEN - |--- PERIS - |--- PERIC - |--- FSYS - |--- FSYS2 - -- In case of Exynos5433, there is VDD_INT power line as following: - VDD_INT |--- G2D (parent device) - |--- MSCL - |--- GSCL - |--- JPEG - |--- MFC - |--- HEVC - |--- BUS0 - |--- BUS1 - |--- BUS2 - |--- PERIS (Fixed clock rate) - |--- PERIC (Fixed clock rate) - |--- FSYS (Fixed clock rate) - -Example 1: - Show the AXI buses of Exynos3250 SoC. Exynos3250 divides the buses to - power line (regulator). The MIF (Memory Interface) AXI bus is used to - transfer data between DRAM and CPU and uses the VDD_MIF regulator. - - - MIF (Memory Interface) block - : VDD_MIF |--- DMC (Dynamic Memory Controller) - - - INT (Internal) block - : VDD_INT |--- LEFTBUS (parent device) - |--- PERIL - |--- MFC - |--- G3D - |--- RIGHTBUS - |--- FSYS - |--- LCD0 - |--- PERIR - |--- ISP - |--- CAM - - - MIF bus's frequency/voltage table - ----------------------- - |Lv| Freq | Voltage | - ----------------------- - |L1| 50000 |800000 | - |L2| 100000 |800000 | - |L3| 134000 |800000 | - |L4| 200000 |825000 | - |L5| 400000 |875000 | - ----------------------- - - - INT bus's frequency/voltage table - ---------------------------------------------------------- - |Block|LEFTBUS|RIGHTBUS|MCUISP |ISP |PERIL ||VDD_INT | - | name| |LCD0 | | | || | - | | |FSYS | | | || | - | | |MFC | | | || | - ---------------------------------------------------------- - |Mode |*parent|passive |passive|passive|passive|| | - ---------------------------------------------------------- - |Lv |Frequency ||Voltage | - ---------------------------------------------------------- - |L1 |50000 |50000 |50000 |50000 |50000 ||900000 | - |L2 |80000 |80000 |80000 |80000 |80000 ||900000 | - |L3 |100000 |100000 |100000 |100000 |100000 ||1000000 | - |L4 |134000 |134000 |200000 |200000 | ||1000000 | - |L5 |200000 |200000 |400000 |300000 | ||1000000 | - ---------------------------------------------------------- - -Example 2: - The bus of DMC (Dynamic Memory Controller) block in exynos3250.dtsi - is listed below: - - bus_dmc: bus_dmc { - compatible = "samsung,exynos-bus"; - clocks = <&cmu_dmc CLK_DIV_DMC>; - clock-names = "bus"; - operating-points-v2 = <&bus_dmc_opp_table>; - status = "disabled"; - }; - - bus_dmc_opp_table: opp_table1 { - compatible = "operating-points-v2"; - opp-shared; - - opp-50000000 { - opp-hz = /bits/ 64 <50000000>; - opp-microvolt = <800000>; - }; - opp-100000000 { - opp-hz = /bits/ 64 <100000000>; - opp-microvolt = <800000>; - }; - opp-134000000 { - opp-hz = /bits/ 64 <134000000>; - opp-microvolt = <800000>; - }; - opp-200000000 { - opp-hz = /bits/ 64 <200000000>; - opp-microvolt = <825000>; - }; - opp-400000000 { - opp-hz = /bits/ 64 <400000000>; - opp-microvolt = <875000>; - }; - }; - - bus_leftbus: bus_leftbus { - compatible = "samsung,exynos-bus"; - clocks = <&cmu CLK_DIV_GDL>; - clock-names = "bus"; - operating-points-v2 = <&bus_leftbus_opp_table>; - status = "disabled"; - }; - - bus_rightbus: bus_rightbus { - compatible = "samsung,exynos-bus"; - clocks = <&cmu CLK_DIV_GDR>; - clock-names = "bus"; - operating-points-v2 = <&bus_leftbus_opp_table>; - status = "disabled"; - }; - - bus_lcd0: bus_lcd0 { - compatible = "samsung,exynos-bus"; - clocks = <&cmu CLK_DIV_ACLK_160>; - clock-names = "bus"; - operating-points-v2 = <&bus_leftbus_opp_table>; - status = "disabled"; - }; - - bus_fsys: bus_fsys { - compatible = "samsung,exynos-bus"; - clocks = <&cmu CLK_DIV_ACLK_200>; - clock-names = "bus"; - operating-points-v2 = <&bus_leftbus_opp_table>; - status = "disabled"; - }; - - bus_mcuisp: bus_mcuisp { - compatible = "samsung,exynos-bus"; - clocks = <&cmu CLK_DIV_ACLK_400_MCUISP>; - clock-names = "bus"; - operating-points-v2 = <&bus_mcuisp_opp_table>; - status = "disabled"; - }; - - bus_isp: bus_isp { - compatible = "samsung,exynos-bus"; - clocks = <&cmu CLK_DIV_ACLK_266>; - clock-names = "bus"; - operating-points-v2 = <&bus_isp_opp_table>; - status = "disabled"; - }; - - bus_peril: bus_peril { - compatible = "samsung,exynos-bus"; - clocks = <&cmu CLK_DIV_ACLK_100>; - clock-names = "bus"; - operating-points-v2 = <&bus_peril_opp_table>; - status = "disabled"; - }; - - bus_mfc: bus_mfc { - compatible = "samsung,exynos-bus"; - clocks = <&cmu CLK_SCLK_MFC>; - clock-names = "bus"; - operating-points-v2 = <&bus_leftbus_opp_table>; - status = "disabled"; - }; - - bus_leftbus_opp_table: opp_table1 { - compatible = "operating-points-v2"; - opp-shared; - - opp-50000000 { - opp-hz = /bits/ 64 <50000000>; - opp-microvolt = <900000>; - }; - opp-80000000 { - opp-hz = /bits/ 64 <80000000>; - opp-microvolt = <900000>; - }; - opp-100000000 { - opp-hz = /bits/ 64 <100000000>; - opp-microvolt = <1000000>; - }; - opp-134000000 { - opp-hz = /bits/ 64 <134000000>; - opp-microvolt = <1000000>; - }; - opp-200000000 { - opp-hz = /bits/ 64 <200000000>; - opp-microvolt = <1000000>; - }; - }; - - bus_mcuisp_opp_table: opp_table2 { - compatible = "operating-points-v2"; - opp-shared; - - opp-50000000 { - opp-hz = /bits/ 64 <50000000>; - }; - opp-80000000 { - opp-hz = /bits/ 64 <80000000>; - }; - opp-100000000 { - opp-hz = /bits/ 64 <100000000>; - }; - opp-200000000 { - opp-hz = /bits/ 64 <200000000>; - }; - opp-400000000 { - opp-hz = /bits/ 64 <400000000>; - }; - }; - - bus_isp_opp_table: opp_table3 { - compatible = "operating-points-v2"; - opp-shared; - - opp-50000000 { - opp-hz = /bits/ 64 <50000000>; - }; - opp-80000000 { - opp-hz = /bits/ 64 <80000000>; - }; - opp-100000000 { - opp-hz = /bits/ 64 <100000000>; - }; - opp-200000000 { - opp-hz = /bits/ 64 <200000000>; - }; - opp-300000000 { - opp-hz = /bits/ 64 <300000000>; - }; - }; - - bus_peril_opp_table: opp_table4 { - compatible = "operating-points-v2"; - opp-shared; - - opp-50000000 { - opp-hz = /bits/ 64 <50000000>; - }; - opp-80000000 { - opp-hz = /bits/ 64 <80000000>; - }; - opp-100000000 { - opp-hz = /bits/ 64 <100000000>; - }; - }; - - - Usage case to handle the frequency and voltage of bus on runtime - in exynos3250-rinato.dts is listed below: - - &bus_dmc { - devfreq-events = <&ppmu_dmc0_3>, <&ppmu_dmc1_3>; - vdd-supply = <&buck1_reg>; /* VDD_MIF */ - status = "okay"; - }; - - &bus_leftbus { - devfreq-events = <&ppmu_leftbus_3>, <&ppmu_rightbus_3>; - vdd-supply = <&buck3_reg>; - status = "okay"; - }; - - &bus_rightbus { - devfreq = <&bus_leftbus>; - status = "okay"; - }; - - &bus_lcd0 { - devfreq = <&bus_leftbus>; - status = "okay"; - }; - - &bus_fsys { - devfreq = <&bus_leftbus>; - status = "okay"; - }; - - &bus_mcuisp { - devfreq = <&bus_leftbus>; - status = "okay"; - }; - - &bus_isp { - devfreq = <&bus_leftbus>; - status = "okay"; - }; - - &bus_peril { - devfreq = <&bus_leftbus>; - status = "okay"; - }; - - &bus_mfc { - devfreq = <&bus_leftbus>; - status = "okay"; - }; - -Example 3: - An interconnect path "bus_display -- bus_leftbus -- bus_dmc" on - Exynos4412 SoC with video mixer as an interconnect consumer device. - - soc { - bus_dmc: bus_dmc { - compatible = "samsung,exynos-bus"; - clocks = <&clock CLK_DIV_DMC>; - clock-names = "bus"; - operating-points-v2 = <&bus_dmc_opp_table>; - samsung,data-clock-ratio = <4>; - #interconnect-cells = <0>; - }; - - bus_leftbus: bus_leftbus { - compatible = "samsung,exynos-bus"; - clocks = <&clock CLK_DIV_GDL>; - clock-names = "bus"; - operating-points-v2 = <&bus_leftbus_opp_table>; - #interconnect-cells = <0>; - interconnects = <&bus_dmc>; - }; - - bus_display: bus_display { - compatible = "samsung,exynos-bus"; - clocks = <&clock CLK_ACLK160>; - clock-names = "bus"; - operating-points-v2 = <&bus_display_opp_table>; - #interconnect-cells = <0>; - interconnects = <&bus_leftbus &bus_dmc>; - }; - - bus_dmc_opp_table: opp_table1 { - compatible = "operating-points-v2"; - /* ... */ - } - - bus_leftbus_opp_table: opp_table3 { - compatible = "operating-points-v2"; - /* ... */ - }; - - bus_display_opp_table: opp_table4 { - compatible = "operating-points-v2"; - /* .. */ - }; - - &mixer { - compatible = "samsung,exynos4212-mixer"; - interconnects = <&bus_display &bus_dmc>; - /* ... */ - }; - }; diff --git a/Documentation/devicetree/bindings/interconnect/mediatek,cci.yaml b/Documentation/devicetree/bindings/interconnect/mediatek,cci.yaml new file mode 100644 index 000000000000..449c7c988229 --- /dev/null +++ b/Documentation/devicetree/bindings/interconnect/mediatek,cci.yaml @@ -0,0 +1,141 @@ +# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause) +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/interconnect/mediatek,cci.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: MediaTek Cache Coherent Interconnect (CCI) frequency and voltage scaling + +maintainers: + - Jia-Wei Chang <jia-wei.chang@mediatek.com> + - Johnson Wang <johnson.wang@mediatek.com> + +description: | + MediaTek Cache Coherent Interconnect (CCI) is a hardware engine used by + MT8183 and MT8186 SoCs to scale the frequency and adjust the voltage in + hardware. It can also optimize the voltage to reduce the power consumption. + +properties: + compatible: + enum: + - mediatek,mt8183-cci + - mediatek,mt8186-cci + + clocks: + items: + - description: + The multiplexer for clock input of the bus. + - description: + A parent of "bus" clock which is used as an intermediate clock source + when the original clock source (PLL) is under transition and not + stable yet. + + clock-names: + items: + - const: cci + - const: intermediate + + operating-points-v2: true + opp-table: true + + proc-supply: + description: + Phandle of the regulator for CCI that provides the supply voltage. + + sram-supply: + description: + Phandle of the regulator for sram of CCI that provides the supply + voltage. When it is present, the implementation needs to do + "voltage tracking" to step by step scale up/down Vproc and Vsram to fit + SoC specific needs. When absent, the voltage scaling flow is handled by + hardware, hence no software "voltage tracking" is needed. + +required: + - compatible + - clocks + - clock-names + - operating-points-v2 + - proc-supply + +additionalProperties: false + +examples: + - | + #include <dt-bindings/clock/mt8183-clk.h> + cci: cci { + compatible = "mediatek,mt8183-cci"; + clocks = <&mcucfg CLK_MCU_BUS_SEL>, + <&topckgen CLK_TOP_ARMPLL_DIV_PLL1>; + clock-names = "cci", "intermediate"; + operating-points-v2 = <&cci_opp>; + proc-supply = <&mt6358_vproc12_reg>; + }; + + cci_opp: opp-table-cci { + compatible = "operating-points-v2"; + opp-shared; + opp2_00: opp-273000000 { + opp-hz = /bits/ 64 <273000000>; + opp-microvolt = <650000>; + }; + opp2_01: opp-338000000 { + opp-hz = /bits/ 64 <338000000>; + opp-microvolt = <687500>; + }; + opp2_02: opp-403000000 { + opp-hz = /bits/ 64 <403000000>; + opp-microvolt = <718750>; + }; + opp2_03: opp-463000000 { + opp-hz = /bits/ 64 <463000000>; + opp-microvolt = <756250>; + }; + opp2_04: opp-546000000 { + opp-hz = /bits/ 64 <546000000>; + opp-microvolt = <800000>; + }; + opp2_05: opp-624000000 { + opp-hz = /bits/ 64 <624000000>; + opp-microvolt = <818750>; + }; + opp2_06: opp-689000000 { + opp-hz = /bits/ 64 <689000000>; + opp-microvolt = <850000>; + }; + opp2_07: opp-767000000 { + opp-hz = /bits/ 64 <767000000>; + opp-microvolt = <868750>; + }; + opp2_08: opp-845000000 { + opp-hz = /bits/ 64 <845000000>; + opp-microvolt = <893750>; + }; + opp2_09: opp-871000000 { + opp-hz = /bits/ 64 <871000000>; + opp-microvolt = <906250>; + }; + opp2_10: opp-923000000 { + opp-hz = /bits/ 64 <923000000>; + opp-microvolt = <931250>; + }; + opp2_11: opp-962000000 { + opp-hz = /bits/ 64 <962000000>; + opp-microvolt = <943750>; + }; + opp2_12: opp-1027000000 { + opp-hz = /bits/ 64 <1027000000>; + opp-microvolt = <975000>; + }; + opp2_13: opp-1092000000 { + opp-hz = /bits/ 64 <1092000000>; + opp-microvolt = <1000000>; + }; + opp2_14: opp-1144000000 { + opp-hz = /bits/ 64 <1144000000>; + opp-microvolt = <1025000>; + }; + opp2_15: opp-1196000000 { + opp-hz = /bits/ 64 <1196000000>; + opp-microvolt = <1050000>; + }; + }; diff --git a/Documentation/devicetree/bindings/interconnect/samsung,exynos-bus.yaml b/Documentation/devicetree/bindings/interconnect/samsung,exynos-bus.yaml new file mode 100644 index 000000000000..ad9ed596dfef --- /dev/null +++ b/Documentation/devicetree/bindings/interconnect/samsung,exynos-bus.yaml @@ -0,0 +1,290 @@ +# SPDX-License-Identifier: GPL-2.0 +%YAML 1.2 +--- +$id: http://devicetree.org/schemas/interconnect/samsung,exynos-bus.yaml# +$schema: http://devicetree.org/meta-schemas/core.yaml# + +title: Samsung Exynos SoC Bus and Interconnect + +maintainers: + - Chanwoo Choi <cw00.choi@samsung.com> + - Krzysztof Kozlowski <krzk@kernel.org> + +description: | + The Samsung Exynos SoC has many buses for data transfer between DRAM and + sub-blocks in SoC. Most Exynos SoCs share the common architecture for buses. + Generally, each bus of Exynos SoC includes a source clock and a power line, + which are able to change the clock frequency of the bus in runtime. To + monitor the usage of each bus in runtime, the driver uses the PPMU (Platform + Performance Monitoring Unit), which is able to measure the current load of + sub-blocks. + + The Exynos SoC includes the various sub-blocks which have the each AXI bus. + The each AXI bus has the owned source clock but, has not the only owned power + line. The power line might be shared among one more sub-blocks. So, we can + divide into two type of device as the role of each sub-block. There are two + type of bus devices as following:: + - parent bus device + - passive bus device + + Basically, parent and passive bus device share the same power line. The + parent bus device can only change the voltage of shared power line and the + rest bus devices (passive bus device) depend on the decision of the parent + bus device. If there are three blocks which share the VDD_xxx power line, + Only one block should be parent device and then the rest blocks should depend + on the parent device as passive device. + + VDD_xxx |--- A block (parent) + |--- B block (passive) + |--- C block (passive) + + There are a little different composition among Exynos SoC because each Exynos + SoC has different sub-blocks. Therefore, such difference should be specified + in devicetree file instead of each device driver. In result, this driver is + able to support the bus frequency for all Exynos SoCs. + + Detailed correlation between sub-blocks and power line according + to Exynos SoC:: + - In case of Exynos3250, there are two power line as following:: + VDD_MIF |--- DMC (Dynamic Memory Controller) + + VDD_INT |--- LEFTBUS (parent device) + |--- PERIL + |--- MFC + |--- G3D + |--- RIGHTBUS + |--- PERIR + |--- FSYS + |--- LCD0 + |--- PERIR + |--- ISP + |--- CAM + + - MIF bus's frequency/voltage table + ----------------------- + |Lv| Freq | Voltage | + ----------------------- + |L1| 50000 |800000 | + |L2| 100000 |800000 | + |L3| 134000 |800000 | + |L4| 200000 |825000 | + |L5| 400000 |875000 | + ----------------------- + + - INT bus's frequency/voltage table + ---------------------------------------------------------- + |Block|LEFTBUS|RIGHTBUS|MCUISP |ISP |PERIL ||VDD_INT | + | name| |LCD0 | | | || | + | | |FSYS | | | || | + | | |MFC | | | || | + ---------------------------------------------------------- + |Mode |*parent|passive |passive|passive|passive|| | + ---------------------------------------------------------- + |Lv |Frequency ||Voltage | + ---------------------------------------------------------- + |L1 |50000 |50000 |50000 |50000 |50000 ||900000 | + |L2 |80000 |80000 |80000 |80000 |80000 ||900000 | + |L3 |100000 |100000 |100000 |100000 |100000 ||1000000 | + |L4 |134000 |134000 |200000 |200000 | ||1000000 | + |L5 |200000 |200000 |400000 |300000 | ||1000000 | + ---------------------------------------------------------- + + - In case of Exynos4210, there is one power line as following:: + VDD_INT |--- DMC (parent device, Dynamic Memory Controller) + |--- LEFTBUS + |--- PERIL + |--- MFC(L) + |--- G3D + |--- TV + |--- LCD0 + |--- RIGHTBUS + |--- PERIR + |--- MFC(R) + |--- CAM + |--- FSYS + |--- GPS + |--- LCD0 + |--- LCD1 + + - In case of Exynos4x12, there are two power line as following:: + VDD_MIF |--- DMC (Dynamic Memory Controller) + + VDD_INT |--- LEFTBUS (parent device) + |--- PERIL + |--- MFC(L) + |--- G3D + |--- TV + |--- IMAGE + |--- RIGHTBUS + |--- PERIR + |--- MFC(R) + |--- CAM + |--- FSYS + |--- GPS + |--- LCD0 + |--- ISP + + - In case of Exynos5422, there are two power line as following:: + VDD_MIF |--- DREX 0 (parent device, DRAM EXpress controller) + |--- DREX 1 + + VDD_INT |--- NoC_Core (parent device) + |--- G2D + |--- G3D + |--- DISP1 + |--- NoC_WCORE + |--- GSCL + |--- MSCL + |--- ISP + |--- MFC + |--- GEN + |--- PERIS + |--- PERIC + |--- FSYS + |--- FSYS2 + + - In case of Exynos5433, there is VDD_INT power line as following:: + VDD_INT |--- G2D (parent device) + |--- MSCL + |--- GSCL + |--- JPEG + |--- MFC + |--- HEVC + |--- BUS0 + |--- BUS1 + |--- BUS2 + |--- PERIS (Fixed clock rate) + |--- PERIC (Fixed clock rate) + |--- FSYS (Fixed clock rate) + +properties: + compatible: + enum: + - samsung,exynos-bus + + clocks: + maxItems: 1 + + clock-names: + items: + - const: bus + + devfreq: + $ref: /schemas/types.yaml#/definitions/phandle + description: + Parent bus device. Valid and required only for the passive bus devices. + + devfreq-events: + $ref: /schemas/types.yaml#/definitions/phandle-array + minItems: 1 + maxItems: 4 + description: + Devfreq-event device to monitor the current utilization of buses. Valid + and required only for the parent bus devices. + + exynos,saturation-ratio: + $ref: /schemas/types.yaml#/definitions/uint32 + description: + Percentage value which is used to calibrate the performance count against + total cycle count. Valid only for the parent bus devices. + + '#interconnect-cells': + const: 0 + + interconnects: + minItems: 1 + maxItems: 2 + + operating-points-v2: true + + samsung,data-clock-ratio: + $ref: /schemas/types.yaml#/definitions/uint32 + default: 8 + description: + Ratio of the data throughput in B/s to minimum data clock frequency in + Hz. + + vdd-supply: + description: + Main bus power rail. Valid and required only for the parent bus devices. + +required: + - compatible + - clocks + - clock-names + - operating-points-v2 + +additionalProperties: false + +examples: + - | + #include <dt-bindings/clock/exynos3250.h> + + bus-dmc { + compatible = "samsung,exynos-bus"; + clocks = <&cmu_dmc CLK_DIV_DMC>; + clock-names = "bus"; + operating-points-v2 = <&bus_dmc_opp_table>; + devfreq-events = <&ppmu_dmc0_3>, <&ppmu_dmc1_3>; + vdd-supply = <&buck1_reg>; + }; + + ppmu_dmc0: ppmu@106a0000 { + compatible = "samsung,exynos-ppmu"; + reg = <0x106a0000 0x2000>; + events { + ppmu_dmc0_3: ppmu-event3-dmc0 { + event-name = "ppmu-event3-dmc0"; + }; + }; + }; + + bus_leftbus: bus-leftbus { + compatible = "samsung,exynos-bus"; + clocks = <&cmu CLK_DIV_GDL>; + clock-names = "bus"; + operating-points-v2 = <&bus_leftbus_opp_table>; + devfreq-events = <&ppmu_leftbus_3>, <&ppmu_rightbus_3>; + vdd-supply = <&buck3_reg>; + }; + + bus-rightbus { + compatible = "samsung,exynos-bus"; + clocks = <&cmu CLK_DIV_GDR>; + clock-names = "bus"; + operating-points-v2 = <&bus_leftbus_opp_table>; + devfreq = <&bus_leftbus>; + }; + + - | + dmc: bus-dmc { + compatible = "samsung,exynos-bus"; + clocks = <&clock CLK_DIV_DMC>; + clock-names = "bus"; + operating-points-v2 = <&bus_dmc_opp_table>; + samsung,data-clock-ratio = <4>; + #interconnect-cells = <0>; + devfreq-events = <&ppmu_dmc0_3>, <&ppmu_dmc1_3>; + vdd-supply = <&buck1_reg>; + }; + + leftbus: bus-leftbus { + compatible = "samsung,exynos-bus"; + clocks = <&clock CLK_DIV_GDL>; + clock-names = "bus"; + operating-points-v2 = <&bus_leftbus_opp_table>; + interconnects = <&dmc>; + #interconnect-cells = <0>; + devfreq-events = <&ppmu_leftbus_3>, <&ppmu_rightbus_3>; + vdd-supply = <&buck3_reg>; + }; + + display: bus-display { + compatible = "samsung,exynos-bus"; + clocks = <&clock CLK_DIV_ACLK_266>; + clock-names = "bus"; + operating-points-v2 = <&bus_display_opp_table>; + interconnects = <&leftbus &dmc>; + #interconnect-cells = <0>; + devfreq = <&leftbus>; + }; diff --git a/Documentation/power/energy-model.rst b/Documentation/power/energy-model.rst index feb257b7f350..ef341be2882b 100644 --- a/Documentation/power/energy-model.rst +++ b/Documentation/power/energy-model.rst @@ -20,20 +20,20 @@ possible source of information on its own, the EM framework intervenes as an abstraction layer which standardizes the format of power cost tables in the kernel, hence enabling to avoid redundant work. -The power values might be expressed in milli-Watts or in an 'abstract scale'. +The power values might be expressed in micro-Watts or in an 'abstract scale'. Multiple subsystems might use the EM and it is up to the system integrator to check that the requirements for the power value scale types are met. An example can be found in the Energy-Aware Scheduler documentation Documentation/scheduler/sched-energy.rst. For some subsystems like thermal or powercap power values expressed in an 'abstract scale' might cause issues. These subsystems are more interested in estimation of power used in the past, -thus the real milli-Watts might be needed. An example of these requirements can +thus the real micro-Watts might be needed. An example of these requirements can be found in the Intelligent Power Allocation in Documentation/driver-api/thermal/power_allocator.rst. Kernel subsystems might implement automatic detection to check whether EM registered devices have inconsistent scale (based on EM internal flag). Important thing to keep in mind is that when the power values are expressed in -an 'abstract scale' deriving real energy in milli-Joules would not be possible. +an 'abstract scale' deriving real energy in micro-Joules would not be possible. The figure below depicts an example of drivers (Arm-specific here, but the approach is applicable to any architecture) providing power costs to the EM @@ -98,7 +98,7 @@ Drivers are expected to register performance domains into the EM framework by calling the following API:: int em_dev_register_perf_domain(struct device *dev, unsigned int nr_states, - struct em_data_callback *cb, cpumask_t *cpus, bool milliwatts); + struct em_data_callback *cb, cpumask_t *cpus, bool microwatts); Drivers must provide a callback function returning <frequency, power> tuples for each performance state. The callback function provided by the driver is free @@ -106,10 +106,10 @@ to fetch data from any relevant location (DT, firmware, ...), and by any mean deemed necessary. Only for CPU devices, drivers must specify the CPUs of the performance domains using cpumask. For other devices than CPUs the last argument must be set to NULL. -The last argument 'milliwatts' is important to set with correct value. Kernel +The last argument 'microwatts' is important to set with correct value. Kernel subsystems which use EM might rely on this flag to check if all EM devices use the same scale. If there are different scales, these subsystems might decide -to: return warning/error, stop working or panic. +to return warning/error, stop working or panic. See Section 3. for an example of driver implementing this callback, or Section 2.4 for further documentation on this API @@ -137,7 +137,7 @@ The .get_cost() allows to provide the 'cost' values which reflect the efficiency of the CPUs. This would allow to provide EAS information which has different relation than what would be forced by the EM internal formulas calculating 'cost' values. To register an EM for such platform, the -driver must set the flag 'milliwatts' to 0, provide .get_power() callback +driver must set the flag 'microwatts' to 0, provide .get_power() callback and provide .get_cost() callback. The EM framework would handle such platform properly during registration. A flag EM_PERF_DOMAIN_ARTIFICIAL is set for such platform. Special care should be taken by other frameworks which are using EM diff --git a/Documentation/power/pci.rst b/Documentation/power/pci.rst index b04fb18cc4e2..a125544b4cb6 100644 --- a/Documentation/power/pci.rst +++ b/Documentation/power/pci.rst @@ -315,7 +315,7 @@ that these callbacks operate on:: configuration space */ unsigned int pme_support:5; /* Bitmask of states from which PME# can be generated */ - unsigned int pme_interrupt:1;/* Is native PCIe PME signaling used? */ + unsigned int pme_poll:1; /* Poll device's PME status bit */ unsigned int d1_support:1; /* Low power state D1 is supported */ unsigned int d2_support:1; /* Low power state D2 is supported */ unsigned int no_d1d2:1; /* D1 and D2 are forbidden */ diff --git a/MAINTAINERS b/MAINTAINERS index 21b58f9b8b10..c096da63ea63 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -4408,7 +4408,7 @@ L: linux-pm@vger.kernel.org L: linux-samsung-soc@vger.kernel.org S: Maintained T: git git://git.kernel.org/pub/scm/linux/kernel/git/chanwoo/linux.git -F: Documentation/devicetree/bindings/devfreq/exynos-bus.txt +F: Documentation/devicetree/bindings/interconnect/samsung,exynos-bus.yaml F: drivers/devfreq/exynos-bus.c BUSLOGIC SCSI DRIVER @@ -5890,6 +5890,7 @@ L: linux-pm@vger.kernel.org S: Maintained T: git git://git.kernel.org/pub/scm/linux/kernel/git/chanwoo/linux.git F: Documentation/devicetree/bindings/devfreq/ +F: Documentation/devicetree/bindings/interconnect/mediatek,cci.yaml F: drivers/devfreq/ F: include/linux/devfreq.h F: include/trace/events/devfreq.h diff --git a/drivers/base/power/domain.c b/drivers/base/power/domain.c index 739e52cd4aba..55a10e6d4e2a 100644 --- a/drivers/base/power/domain.c +++ b/drivers/base/power/domain.c @@ -222,6 +222,9 @@ static void genpd_debug_remove(struct generic_pm_domain *genpd) { struct dentry *d; + if (!genpd_debugfs_dir) + return; + d = debugfs_lookup(genpd->name, genpd_debugfs_dir); debugfs_remove(d); } diff --git a/drivers/base/power/runtime.c b/drivers/base/power/runtime.c index 949907e2e242..997be3ac20a7 100644 --- a/drivers/base/power/runtime.c +++ b/drivers/base/power/runtime.c @@ -1862,10 +1862,13 @@ int pm_runtime_force_suspend(struct device *dev) callback = RPM_GET_CALLBACK(dev, runtime_suspend); + dev_pm_enable_wake_irq_check(dev, true); ret = callback ? callback(dev) : 0; if (ret) goto err; + dev_pm_enable_wake_irq_complete(dev); + /* * If the device can stay in suspend after the system-wide transition * to the working state that will follow, drop the children counter of @@ -1882,6 +1885,7 @@ int pm_runtime_force_suspend(struct device *dev) return 0; err: + dev_pm_disable_wake_irq_check(dev, true); pm_runtime_enable(dev); return ret; } @@ -1915,9 +1919,11 @@ int pm_runtime_force_resume(struct device *dev) callback = RPM_GET_CALLBACK(dev, runtime_resume); + dev_pm_disable_wake_irq_check(dev, false); ret = callback ? callback(dev) : 0; if (ret) { pm_runtime_set_suspended(dev); + dev_pm_enable_wake_irq_check(dev, false); goto out; } diff --git a/drivers/base/power/wakeup.c b/drivers/base/power/wakeup.c index 11a4ffe91367..e3befa2c1b66 100644 --- a/drivers/base/power/wakeup.c +++ b/drivers/base/power/wakeup.c @@ -501,36 +501,6 @@ void device_set_wakeup_capable(struct device *dev, bool capable) EXPORT_SYMBOL_GPL(device_set_wakeup_capable); /** - * device_init_wakeup - Device wakeup initialization. - * @dev: Device to handle. - * @enable: Whether or not to enable @dev as a wakeup device. - * - * By default, most devices should leave wakeup disabled. The exceptions are - * devices that everyone expects to be wakeup sources: keyboards, power buttons, - * possibly network interfaces, etc. Also, devices that don't generate their - * own wakeup requests but merely forward requests from one bus to another - * (like PCI bridges) should have wakeup enabled by default. - */ -int device_init_wakeup(struct device *dev, bool enable) -{ - int ret = 0; - - if (!dev) - return -EINVAL; - - if (enable) { - device_set_wakeup_capable(dev, true); - ret = device_wakeup_enable(dev); - } else { - device_wakeup_disable(dev); - device_set_wakeup_capable(dev, false); - } - - return ret; -} -EXPORT_SYMBOL_GPL(device_init_wakeup); - -/** * device_set_wakeup_enable - Enable or disable a device to wake up the system. * @dev: Device to handle. * @enable: enable/disable flag diff --git a/drivers/cpufreq/Kconfig b/drivers/cpufreq/Kconfig index c3038cdc6865..2a84fc63371e 100644 --- a/drivers/cpufreq/Kconfig +++ b/drivers/cpufreq/Kconfig @@ -268,7 +268,7 @@ config LOONGSON2_CPUFREQ This option adds a CPUFreq driver for loongson processors which support software configurable cpu frequency. - Loongson2F and it's successors support this feature. + Loongson2F and its successors support this feature. If in doubt, say N. diff --git a/drivers/cpufreq/acpi-cpufreq.c b/drivers/cpufreq/acpi-cpufreq.c index 3d514b82d055..1bb2b90ebb21 100644 --- a/drivers/cpufreq/acpi-cpufreq.c +++ b/drivers/cpufreq/acpi-cpufreq.c @@ -78,6 +78,8 @@ static bool boost_state(unsigned int cpu) switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_INTEL: + case X86_VENDOR_CENTAUR: + case X86_VENDOR_ZHAOXIN: rdmsr_on_cpu(cpu, MSR_IA32_MISC_ENABLE, &lo, &hi); msr = lo | ((u64)hi << 32); return !(msr & MSR_IA32_MISC_ENABLE_TURBO_DISABLE); @@ -97,6 +99,8 @@ static int boost_set_msr(bool enable) switch (boot_cpu_data.x86_vendor) { case X86_VENDOR_INTEL: + case X86_VENDOR_CENTAUR: + case X86_VENDOR_ZHAOXIN: msr_addr = MSR_IA32_MISC_ENABLE; msr_mask = MSR_IA32_MISC_ENABLE_TURBO_DISABLE; break; diff --git a/drivers/cpufreq/cpufreq.c b/drivers/cpufreq/cpufreq.c index 2cad42774164..954eef26685f 100644 --- a/drivers/cpufreq/cpufreq.c +++ b/drivers/cpufreq/cpufreq.c @@ -843,12 +843,14 @@ ssize_t cpufreq_show_cpus(const struct cpumask *mask, char *buf) unsigned int cpu; for_each_cpu(cpu, mask) { - if (i) - i += scnprintf(&buf[i], (PAGE_SIZE - i - 2), " "); - i += scnprintf(&buf[i], (PAGE_SIZE - i - 2), "%u", cpu); + i += scnprintf(&buf[i], (PAGE_SIZE - i - 2), "%u ", cpu); if (i >= (PAGE_SIZE - 5)) break; } + + /* Remove the extra space at the end */ + i--; + i += sprintf(&buf[i], "\n"); return i; } @@ -971,21 +973,10 @@ static ssize_t store(struct kobject *kobj, struct attribute *attr, if (!fattr->store) return -EIO; - /* - * cpus_read_trylock() is used here to work around a circular lock - * dependency problem with respect to the cpufreq_register_driver(). - */ - if (!cpus_read_trylock()) - return -EBUSY; - - if (cpu_online(policy->cpu)) { - down_write(&policy->rwsem); - if (likely(!policy_is_inactive(policy))) - ret = fattr->store(policy, buf, count); - up_write(&policy->rwsem); - } - - cpus_read_unlock(); + down_write(&policy->rwsem); + if (likely(!policy_is_inactive(policy))) + ret = fattr->store(policy, buf, count); + up_write(&policy->rwsem); return ret; } @@ -1282,6 +1273,13 @@ static void cpufreq_policy_free(struct cpufreq_policy *policy) unsigned long flags; int cpu; + /* + * The callers must ensure the policy is inactive by now, to avoid any + * races with show()/store() callbacks. + */ + if (unlikely(!policy_is_inactive(policy))) + pr_warn("%s: Freeing active policy\n", __func__); + /* Remove policy from list */ write_lock_irqsave(&cpufreq_driver_lock, flags); list_del(&policy->policy_list); @@ -1536,8 +1534,6 @@ out_destroy_policy: for_each_cpu(j, policy->real_cpus) remove_cpu_dev_symlink(policy, j, get_cpu_device(j)); - cpumask_clear(policy->cpus); - out_offline_policy: if (cpufreq_driver->offline) cpufreq_driver->offline(policy); @@ -1547,6 +1543,7 @@ out_exit_policy: cpufreq_driver->exit(policy); out_free_policy: + cpumask_clear(policy->cpus); up_write(&policy->rwsem); cpufreq_policy_free(policy); diff --git a/drivers/cpufreq/cpufreq_ondemand.c b/drivers/cpufreq/cpufreq_ondemand.c index e8fbf970ff07..c52d19d67557 100644 --- a/drivers/cpufreq/cpufreq_ondemand.c +++ b/drivers/cpufreq/cpufreq_ondemand.c @@ -416,10 +416,13 @@ static struct dbs_governor od_dbs_gov = { static void od_set_powersave_bias(unsigned int powersave_bias) { unsigned int cpu; - cpumask_t done; + cpumask_var_t done; + + if (!alloc_cpumask_var(&done, GFP_KERNEL)) + return; default_powersave_bias = powersave_bias; - cpumask_clear(&done); + cpumask_clear(done); cpus_read_lock(); for_each_online_cpu(cpu) { @@ -428,7 +431,7 @@ static void od_set_powersave_bias(unsigned int powersave_bias) struct dbs_data *dbs_data; struct od_dbs_tuners *od_tuners; - if (cpumask_test_cpu(cpu, &done)) + if (cpumask_test_cpu(cpu, done)) continue; policy = cpufreq_cpu_get_raw(cpu); @@ -439,13 +442,15 @@ static void od_set_powersave_bias(unsigned int powersave_bias) if (!policy_dbs) continue; - cpumask_or(&done, &done, policy->cpus); + cpumask_or(done, done, policy->cpus); dbs_data = policy_dbs->dbs_data; od_tuners = dbs_data->tuners; od_tuners->powersave_bias = default_powersave_bias; } cpus_read_unlock(); + + free_cpumask_var(done); } void od_register_powersave_bias_handler(unsigned int (*f) diff --git a/drivers/cpufreq/mediatek-cpufreq-hw.c b/drivers/cpufreq/mediatek-cpufreq-hw.c index 813cccbfe934..f0e0a35c7f21 100644 --- a/drivers/cpufreq/mediatek-cpufreq-hw.c +++ b/drivers/cpufreq/mediatek-cpufreq-hw.c @@ -51,7 +51,7 @@ static const u16 cpufreq_mtk_offsets[REG_ARRAY_SIZE] = { }; static int __maybe_unused -mtk_cpufreq_get_cpu_power(struct device *cpu_dev, unsigned long *mW, +mtk_cpufreq_get_cpu_power(struct device *cpu_dev, unsigned long *uW, unsigned long *KHz) { struct mtk_cpufreq_data *data; @@ -71,8 +71,9 @@ mtk_cpufreq_get_cpu_power(struct device *cpu_dev, unsigned long *mW, i--; *KHz = data->table[i].frequency; - *mW = readl_relaxed(data->reg_bases[REG_EM_POWER_TBL] + - i * LUT_ROW_SIZE) / 1000; + /* Provide micro-Watts value to the Energy Model */ + *uW = readl_relaxed(data->reg_bases[REG_EM_POWER_TBL] + + i * LUT_ROW_SIZE); return 0; } diff --git a/drivers/cpufreq/scmi-cpufreq.c b/drivers/cpufreq/scmi-cpufreq.c index 6d2a4cf46db7..513a071845c2 100644 --- a/drivers/cpufreq/scmi-cpufreq.c +++ b/drivers/cpufreq/scmi-cpufreq.c @@ -19,6 +19,7 @@ #include <linux/slab.h> #include <linux/scmi_protocol.h> #include <linux/types.h> +#include <linux/units.h> struct scmi_data { int domain_id; @@ -99,6 +100,7 @@ static int __maybe_unused scmi_get_cpu_power(struct device *cpu_dev, unsigned long *power, unsigned long *KHz) { + enum scmi_power_scale power_scale = perf_ops->power_scale_get(ph); unsigned long Hz; int ret, domain; @@ -112,6 +114,10 @@ scmi_get_cpu_power(struct device *cpu_dev, unsigned long *power, if (ret) return ret; + /* Convert the power to uW if it is mW (ignore bogoW) */ + if (power_scale == SCMI_POWER_MILLIWATTS) + *power *= MICROWATT_PER_MILLIWATT; + /* The EM framework specifies the frequency in KHz. */ *KHz = Hz / 1000; @@ -249,8 +255,9 @@ static int scmi_cpufreq_exit(struct cpufreq_policy *policy) static void scmi_cpufreq_register_em(struct cpufreq_policy *policy) { struct em_data_callback em_cb = EM_DATA_CB(scmi_get_cpu_power); - bool power_scale_mw = perf_ops->power_scale_mw_get(ph); + enum scmi_power_scale power_scale = perf_ops->power_scale_get(ph); struct scmi_data *priv = policy->driver_data; + bool em_power_scale = false; /* * This callback will be called for each policy, but we don't need to @@ -262,9 +269,13 @@ static void scmi_cpufreq_register_em(struct cpufreq_policy *policy) if (!priv->nr_opp) return; + if (power_scale == SCMI_POWER_MILLIWATTS + || power_scale == SCMI_POWER_MICROWATTS) + em_power_scale = true; + em_dev_register_perf_domain(get_cpu_device(policy->cpu), priv->nr_opp, &em_cb, priv->opp_shared_cpus, - power_scale_mw); + em_power_scale); } static struct cpufreq_driver scmi_cpufreq_driver = { diff --git a/drivers/cpuidle/governors/haltpoll.c b/drivers/cpuidle/governors/haltpoll.c index cb2a96eafc02..1dff3a52917d 100644 --- a/drivers/cpuidle/governors/haltpoll.c +++ b/drivers/cpuidle/governors/haltpoll.c @@ -19,6 +19,7 @@ #include <linux/sched.h> #include <linux/module.h> #include <linux/kvm_para.h> +#include <trace/events/power.h> static unsigned int guest_halt_poll_ns __read_mostly = 200000; module_param(guest_halt_poll_ns, uint, 0644); @@ -90,6 +91,7 @@ static void adjust_poll_limit(struct cpuidle_device *dev, u64 block_ns) if (val > guest_halt_poll_ns) val = guest_halt_poll_ns; + trace_guest_halt_poll_ns_grow(val, dev->poll_limit_ns); dev->poll_limit_ns = val; } else if (block_ns > guest_halt_poll_ns && guest_halt_poll_allow_shrink) { @@ -100,6 +102,7 @@ static void adjust_poll_limit(struct cpuidle_device *dev, u64 block_ns) val = 0; else val /= shrink; + trace_guest_halt_poll_ns_shrink(val, dev->poll_limit_ns); dev->poll_limit_ns = val; } } diff --git a/drivers/devfreq/Kconfig b/drivers/devfreq/Kconfig index 87eb2b837e68..9754d8b31621 100644 --- a/drivers/devfreq/Kconfig +++ b/drivers/devfreq/Kconfig @@ -120,6 +120,16 @@ config ARM_TEGRA_DEVFREQ It reads ACTMON counters of memory controllers and adjusts the operating frequencies and voltages with OPP support. +config ARM_MEDIATEK_CCI_DEVFREQ + tristate "MEDIATEK CCI DEVFREQ Driver" + depends on ARM_MEDIATEK_CPUFREQ || COMPILE_TEST + select DEVFREQ_GOV_PASSIVE + help + This adds a devfreq driver for MediaTek Cache Coherent Interconnect + which is shared the same regulators with the cpu cluster. It can track + buck voltages and update a proper CCI frequency. Use the notification + to get the regulator status. + config ARM_RK3399_DMC_DEVFREQ tristate "ARM RK3399 DMC DEVFREQ Driver" depends on (ARCH_ROCKCHIP && HAVE_ARM_SMCCC) || \ diff --git a/drivers/devfreq/Makefile b/drivers/devfreq/Makefile index 0b6be92a25d9..bf40d04928d0 100644 --- a/drivers/devfreq/Makefile +++ b/drivers/devfreq/Makefile @@ -11,6 +11,7 @@ obj-$(CONFIG_DEVFREQ_GOV_PASSIVE) += governor_passive.o obj-$(CONFIG_ARM_EXYNOS_BUS_DEVFREQ) += exynos-bus.o obj-$(CONFIG_ARM_IMX_BUS_DEVFREQ) += imx-bus.o obj-$(CONFIG_ARM_IMX8M_DDRC_DEVFREQ) += imx8m-ddrc.o +obj-$(CONFIG_ARM_MEDIATEK_CCI_DEVFREQ) += mtk-cci-devfreq.o obj-$(CONFIG_ARM_RK3399_DMC_DEVFREQ) += rk3399_dmc.o obj-$(CONFIG_ARM_SUN8I_A33_MBUS_DEVFREQ) += sun8i-a33-mbus.o obj-$(CONFIG_ARM_TEGRA_DEVFREQ) += tegra30-devfreq.o diff --git a/drivers/devfreq/devfreq.c b/drivers/devfreq/devfreq.c index 9602141bb8ec..63347a5ae599 100644 --- a/drivers/devfreq/devfreq.c +++ b/drivers/devfreq/devfreq.c @@ -696,6 +696,8 @@ static int qos_notifier_call(struct devfreq *devfreq) /** * qos_min_notifier_call() - Callback for QoS min_freq changes. * @nb: Should be devfreq->nb_min + * @val: not used + * @ptr: not used */ static int qos_min_notifier_call(struct notifier_block *nb, unsigned long val, void *ptr) @@ -706,6 +708,8 @@ static int qos_min_notifier_call(struct notifier_block *nb, /** * qos_max_notifier_call() - Callback for QoS max_freq changes. * @nb: Should be devfreq->nb_max + * @val: not used + * @ptr: not used */ static int qos_max_notifier_call(struct notifier_block *nb, unsigned long val, void *ptr) diff --git a/drivers/devfreq/imx-bus.c b/drivers/devfreq/imx-bus.c index f3f6e25053ed..f87067fc574d 100644 --- a/drivers/devfreq/imx-bus.c +++ b/drivers/devfreq/imx-bus.c @@ -59,7 +59,7 @@ static int imx_bus_init_icc(struct device *dev) struct imx_bus *priv = dev_get_drvdata(dev); const char *icc_driver_name; - if (!of_get_property(dev->of_node, "#interconnect-cells", 0)) + if (!of_get_property(dev->of_node, "#interconnect-cells", NULL)) return 0; if (!IS_ENABLED(CONFIG_INTERCONNECT_IMX)) { dev_warn(dev, "imx interconnect drivers disabled\n"); diff --git a/drivers/devfreq/mtk-cci-devfreq.c b/drivers/devfreq/mtk-cci-devfreq.c new file mode 100644 index 000000000000..71abb3fbd042 --- /dev/null +++ b/drivers/devfreq/mtk-cci-devfreq.c @@ -0,0 +1,440 @@ +// SPDX-License-Identifier: GPL-2.0-only +/* + * Copyright (C) 2022 MediaTek Inc. + */ + +#include <linux/clk.h> +#include <linux/devfreq.h> +#include <linux/minmax.h> +#include <linux/module.h> +#include <linux/of.h> +#include <linux/of_device.h> +#include <linux/platform_device.h> +#include <linux/pm_opp.h> +#include <linux/regulator/consumer.h> + +struct mtk_ccifreq_platform_data { + int min_volt_shift; + int max_volt_shift; + int proc_max_volt; + int sram_min_volt; + int sram_max_volt; +}; + +struct mtk_ccifreq_drv { + struct device *dev; + struct devfreq *devfreq; + struct regulator *proc_reg; + struct regulator *sram_reg; + struct clk *cci_clk; + struct clk *inter_clk; + int inter_voltage; + unsigned long pre_freq; + /* Avoid race condition for regulators between notify and policy */ + struct mutex reg_lock; + struct notifier_block opp_nb; + const struct mtk_ccifreq_platform_data *soc_data; + int vtrack_max; +}; + +static int mtk_ccifreq_set_voltage(struct mtk_ccifreq_drv *drv, int new_voltage) +{ + const struct mtk_ccifreq_platform_data *soc_data = drv->soc_data; + struct device *dev = drv->dev; + int pre_voltage, pre_vsram, new_vsram, vsram, voltage, ret; + int retry_max = drv->vtrack_max; + + if (!drv->sram_reg) { + ret = regulator_set_voltage(drv->proc_reg, new_voltage, + drv->soc_data->proc_max_volt); + return ret; + } + + pre_voltage = regulator_get_voltage(drv->proc_reg); + if (pre_voltage < 0) { + dev_err(dev, "invalid vproc value: %d\n", pre_voltage); + return pre_voltage; + } + + pre_vsram = regulator_get_voltage(drv->sram_reg); + if (pre_vsram < 0) { + dev_err(dev, "invalid vsram value: %d\n", pre_vsram); + return pre_vsram; + } + + new_vsram = clamp(new_voltage + soc_data->min_volt_shift, + soc_data->sram_min_volt, soc_data->sram_max_volt); + + do { + if (pre_voltage <= new_voltage) { + vsram = clamp(pre_voltage + soc_data->max_volt_shift, + soc_data->sram_min_volt, new_vsram); + ret = regulator_set_voltage(drv->sram_reg, vsram, + soc_data->sram_max_volt); + if (ret) + return ret; + + if (vsram == soc_data->sram_max_volt || + new_vsram == soc_data->sram_min_volt) + voltage = new_voltage; + else + voltage = vsram - soc_data->min_volt_shift; + + ret = regulator_set_voltage(drv->proc_reg, voltage, + soc_data->proc_max_volt); + if (ret) { + regulator_set_voltage(drv->sram_reg, pre_vsram, + soc_data->sram_max_volt); + return ret; + } + } else if (pre_voltage > new_voltage) { + voltage = max(new_voltage, + pre_vsram - soc_data->max_volt_shift); + ret = regulator_set_voltage(drv->proc_reg, voltage, + soc_data->proc_max_volt); + if (ret) + return ret; + + if (voltage == new_voltage) + vsram = new_vsram; + else + vsram = max(new_vsram, + voltage + soc_data->min_volt_shift); + + ret = regulator_set_voltage(drv->sram_reg, vsram, + soc_data->sram_max_volt); + if (ret) { + regulator_set_voltage(drv->proc_reg, pre_voltage, + soc_data->proc_max_volt); + return ret; + } + } + + pre_voltage = voltage; + pre_vsram = vsram; + + if (--retry_max < 0) { + dev_err(dev, + "over loop count, failed to set voltage\n"); + return -EINVAL; + } + } while (voltage != new_voltage || vsram != new_vsram); + + return 0; +} + +static int mtk_ccifreq_target(struct device *dev, unsigned long *freq, + u32 flags) +{ + struct mtk_ccifreq_drv *drv = dev_get_drvdata(dev); + struct clk *cci_pll = clk_get_parent(drv->cci_clk); + struct dev_pm_opp *opp; + unsigned long opp_rate; + int voltage, pre_voltage, inter_voltage, target_voltage, ret; + + if (!drv) + return -EINVAL; + + if (drv->pre_freq == *freq) + return 0; + + inter_voltage = drv->inter_voltage; + + opp_rate = *freq; + opp = devfreq_recommended_opp(dev, &opp_rate, 1); + if (IS_ERR(opp)) { + dev_err(dev, "failed to find opp for freq: %ld\n", opp_rate); + return PTR_ERR(opp); + } + + mutex_lock(&drv->reg_lock); + + voltage = dev_pm_opp_get_voltage(opp); + dev_pm_opp_put(opp); + + pre_voltage = regulator_get_voltage(drv->proc_reg); + if (pre_voltage < 0) { + dev_err(dev, "invalid vproc value: %d\n", pre_voltage); + ret = pre_voltage; + goto out_unlock; + } + + /* scale up: set voltage first then freq. */ + target_voltage = max(inter_voltage, voltage); + if (pre_voltage <= target_voltage) { + ret = mtk_ccifreq_set_voltage(drv, target_voltage); + if (ret) { + dev_err(dev, "failed to scale up voltage\n"); + goto out_restore_voltage; + } + } + + /* switch the cci clock to intermediate clock source. */ + ret = clk_set_parent(drv->cci_clk, drv->inter_clk); + if (ret) { + dev_err(dev, "failed to re-parent cci clock\n"); + goto out_restore_voltage; + } + + /* set the original clock to target rate. */ + ret = clk_set_rate(cci_pll, *freq); + if (ret) { + dev_err(dev, "failed to set cci pll rate: %d\n", ret); + clk_set_parent(drv->cci_clk, cci_pll); + goto out_restore_voltage; + } + + /* switch the cci clock back to the original clock source. */ + ret = clk_set_parent(drv->cci_clk, cci_pll); + if (ret) { + dev_err(dev, "failed to re-parent cci clock\n"); + mtk_ccifreq_set_voltage(drv, inter_voltage); + goto out_unlock; + } + + /* + * If the new voltage is lower than the intermediate voltage or the + * original voltage, scale down to the new voltage. + */ + if (voltage < inter_voltage || voltage < pre_voltage) { + ret = mtk_ccifreq_set_voltage(drv, voltage); + if (ret) { + dev_err(dev, "failed to scale down voltage\n"); + goto out_unlock; + } + } + + drv->pre_freq = *freq; + mutex_unlock(&drv->reg_lock); + + return 0; + +out_restore_voltage: + mtk_ccifreq_set_voltage(drv, pre_voltage); + +out_unlock: + mutex_unlock(&drv->reg_lock); + return ret; +} + +static int mtk_ccifreq_opp_notifier(struct notifier_block *nb, + unsigned long event, void *data) +{ + struct dev_pm_opp *opp = data; + struct mtk_ccifreq_drv *drv; + unsigned long freq, volt; + + drv = container_of(nb, struct mtk_ccifreq_drv, opp_nb); + + if (event == OPP_EVENT_ADJUST_VOLTAGE) { + freq = dev_pm_opp_get_freq(opp); + + mutex_lock(&drv->reg_lock); + /* current opp item is changed */ + if (freq == drv->pre_freq) { + volt = dev_pm_opp_get_voltage(opp); + mtk_ccifreq_set_voltage(drv, volt); + } + mutex_unlock(&drv->reg_lock); + } + + return 0; +} + +static struct devfreq_dev_profile mtk_ccifreq_profile = { + .target = mtk_ccifreq_target, +}; + +static int mtk_ccifreq_probe(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct mtk_ccifreq_drv *drv; + struct devfreq_passive_data *passive_data; + struct dev_pm_opp *opp; + unsigned long rate, opp_volt; + int ret; + + drv = devm_kzalloc(dev, sizeof(*drv), GFP_KERNEL); + if (!drv) + return -ENOMEM; + + drv->dev = dev; + drv->soc_data = (const struct mtk_ccifreq_platform_data *) + of_device_get_match_data(&pdev->dev); + mutex_init(&drv->reg_lock); + platform_set_drvdata(pdev, drv); + + drv->cci_clk = devm_clk_get(dev, "cci"); + if (IS_ERR(drv->cci_clk)) { + ret = PTR_ERR(drv->cci_clk); + return dev_err_probe(dev, ret, "failed to get cci clk\n"); + } + + drv->inter_clk = devm_clk_get(dev, "intermediate"); + if (IS_ERR(drv->inter_clk)) { + ret = PTR_ERR(drv->inter_clk); + return dev_err_probe(dev, ret, + "failed to get intermediate clk\n"); + } + + drv->proc_reg = devm_regulator_get_optional(dev, "proc"); + if (IS_ERR(drv->proc_reg)) { + ret = PTR_ERR(drv->proc_reg); + return dev_err_probe(dev, ret, + "failed to get proc regulator\n"); + } + + ret = regulator_enable(drv->proc_reg); + if (ret) { + dev_err(dev, "failed to enable proc regulator\n"); + return ret; + } + + drv->sram_reg = devm_regulator_get_optional(dev, "sram"); + if (IS_ERR(drv->sram_reg)) + drv->sram_reg = NULL; + else { + ret = regulator_enable(drv->sram_reg); + if (ret) { + dev_err(dev, "failed to enable sram regulator\n"); + goto out_free_resources; + } + } + + /* + * We assume min voltage is 0 and tracking target voltage using + * min_volt_shift for each iteration. + * The retry_max is 3 times of expected iteration count. + */ + drv->vtrack_max = 3 * DIV_ROUND_UP(max(drv->soc_data->sram_max_volt, + drv->soc_data->proc_max_volt), + drv->soc_data->min_volt_shift); + + ret = clk_prepare_enable(drv->cci_clk); + if (ret) + goto out_free_resources; + + ret = dev_pm_opp_of_add_table(dev); + if (ret) { + dev_err(dev, "failed to add opp table: %d\n", ret); + goto out_disable_cci_clk; + } + + rate = clk_get_rate(drv->inter_clk); + opp = dev_pm_opp_find_freq_ceil(dev, &rate); + if (IS_ERR(opp)) { + ret = PTR_ERR(opp); + dev_err(dev, "failed to get intermediate opp: %d\n", ret); + goto out_remove_opp_table; + } + drv->inter_voltage = dev_pm_opp_get_voltage(opp); + dev_pm_opp_put(opp); + + rate = U32_MAX; + opp = dev_pm_opp_find_freq_floor(drv->dev, &rate); + if (IS_ERR(opp)) { + dev_err(dev, "failed to get opp\n"); + ret = PTR_ERR(opp); + goto out_remove_opp_table; + } + + opp_volt = dev_pm_opp_get_voltage(opp); + dev_pm_opp_put(opp); + ret = mtk_ccifreq_set_voltage(drv, opp_volt); + if (ret) { + dev_err(dev, "failed to scale to highest voltage %lu in proc_reg\n", + opp_volt); + goto out_remove_opp_table; + } + + passive_data = devm_kzalloc(dev, sizeof(*passive_data), GFP_KERNEL); + if (!passive_data) { + ret = -ENOMEM; + goto out_remove_opp_table; + } + + passive_data->parent_type = CPUFREQ_PARENT_DEV; + drv->devfreq = devm_devfreq_add_device(dev, &mtk_ccifreq_profile, + DEVFREQ_GOV_PASSIVE, + passive_data); + if (IS_ERR(drv->devfreq)) { + ret = -EPROBE_DEFER; + dev_err(dev, "failed to add devfreq device: %ld\n", + PTR_ERR(drv->devfreq)); + goto out_remove_opp_table; + } + + drv->opp_nb.notifier_call = mtk_ccifreq_opp_notifier; + ret = dev_pm_opp_register_notifier(dev, &drv->opp_nb); + if (ret) { + dev_err(dev, "failed to register opp notifier: %d\n", ret); + goto out_remove_opp_table; + } + return 0; + +out_remove_opp_table: + dev_pm_opp_of_remove_table(dev); + +out_disable_cci_clk: + clk_disable_unprepare(drv->cci_clk); + +out_free_resources: + if (regulator_is_enabled(drv->proc_reg)) + regulator_disable(drv->proc_reg); + if (drv->sram_reg && regulator_is_enabled(drv->sram_reg)) + regulator_disable(drv->sram_reg); + + return ret; +} + +static int mtk_ccifreq_remove(struct platform_device *pdev) +{ + struct device *dev = &pdev->dev; + struct mtk_ccifreq_drv *drv; + + drv = platform_get_drvdata(pdev); + + dev_pm_opp_unregister_notifier(dev, &drv->opp_nb); + dev_pm_opp_of_remove_table(dev); + clk_disable_unprepare(drv->cci_clk); + regulator_disable(drv->proc_reg); + if (drv->sram_reg) + regulator_disable(drv->sram_reg); + + return 0; +} + +static const struct mtk_ccifreq_platform_data mt8183_platform_data = { + .min_volt_shift = 100000, + .max_volt_shift = 200000, + .proc_max_volt = 1150000, +}; + +static const struct mtk_ccifreq_platform_data mt8186_platform_data = { + .min_volt_shift = 100000, + .max_volt_shift = 250000, + .proc_max_volt = 1118750, + .sram_min_volt = 850000, + .sram_max_volt = 1118750, +}; + +static const struct of_device_id mtk_ccifreq_machines[] = { + { .compatible = "mediatek,mt8183-cci", .data = &mt8183_platform_data }, + { .compatible = "mediatek,mt8186-cci", .data = &mt8186_platform_data }, + { }, +}; +MODULE_DEVICE_TABLE(of, mtk_ccifreq_machines); + +static struct platform_driver mtk_ccifreq_platdrv = { + .probe = mtk_ccifreq_probe, + .remove = mtk_ccifreq_remove, + .driver = { + .name = "mtk-ccifreq", + .of_match_table = mtk_ccifreq_machines, + }, +}; +module_platform_driver(mtk_ccifreq_platdrv); + +MODULE_DESCRIPTION("MediaTek CCI devfreq driver"); +MODULE_AUTHOR("Jia-Wei Chang <jia-wei.chang@mediatek.com>"); +MODULE_LICENSE("GPL v2"); diff --git a/drivers/devfreq/tegra30-devfreq.c b/drivers/devfreq/tegra30-devfreq.c index 65ecf17a36f4..585a95fe2bd6 100644 --- a/drivers/devfreq/tegra30-devfreq.c +++ b/drivers/devfreq/tegra30-devfreq.c @@ -922,8 +922,10 @@ static int tegra_devfreq_probe(struct platform_device *pdev) devfreq = devm_devfreq_add_device(&pdev->dev, &tegra_devfreq_profile, "tegra_actmon", NULL); - if (IS_ERR(devfreq)) + if (IS_ERR(devfreq)) { + dev_err(&pdev->dev, "Failed to add device: %pe\n", devfreq); return PTR_ERR(devfreq); + } return 0; } diff --git a/drivers/firmware/arm_scmi/perf.c b/drivers/firmware/arm_scmi/perf.c index 64ea2d2f2875..ecf5c4de851b 100644 --- a/drivers/firmware/arm_scmi/perf.c +++ b/drivers/firmware/arm_scmi/perf.c @@ -140,8 +140,7 @@ struct perf_dom_info { struct scmi_perf_info { u32 version; int num_domains; - bool power_scale_mw; - bool power_scale_uw; + enum scmi_power_scale power_scale; u64 stats_addr; u32 stats_size; struct perf_dom_info *dom_info; @@ -171,9 +170,13 @@ static int scmi_perf_attributes_get(const struct scmi_protocol_handle *ph, u16 flags = le16_to_cpu(attr->flags); pi->num_domains = le16_to_cpu(attr->num_domains); - pi->power_scale_mw = POWER_SCALE_IN_MILLIWATT(flags); + + if (POWER_SCALE_IN_MILLIWATT(flags)) + pi->power_scale = SCMI_POWER_MILLIWATTS; if (PROTOCOL_REV_MAJOR(pi->version) >= 0x3) - pi->power_scale_uw = POWER_SCALE_IN_MICROWATT(flags); + if (POWER_SCALE_IN_MICROWATT(flags)) + pi->power_scale = SCMI_POWER_MICROWATTS; + pi->stats_addr = le32_to_cpu(attr->stats_addr_low) | (u64)le32_to_cpu(attr->stats_addr_high) << 32; pi->stats_size = le32_to_cpu(attr->stats_size); @@ -675,11 +678,12 @@ static bool scmi_fast_switch_possible(const struct scmi_protocol_handle *ph, return dom->fc_info && dom->fc_info[PERF_FC_LEVEL].set_addr; } -static bool scmi_power_scale_mw_get(const struct scmi_protocol_handle *ph) +static enum scmi_power_scale +scmi_power_scale_get(const struct scmi_protocol_handle *ph) { struct scmi_perf_info *pi = ph->get_priv(ph); - return pi->power_scale_mw; + return pi->power_scale; } static const struct scmi_perf_proto_ops perf_proto_ops = { @@ -694,7 +698,7 @@ static const struct scmi_perf_proto_ops perf_proto_ops = { .freq_get = scmi_dvfs_freq_get, .est_power_get = scmi_dvfs_est_power_get, .fast_switch_possible = scmi_fast_switch_possible, - .power_scale_mw_get = scmi_power_scale_mw_get, + .power_scale_get = scmi_power_scale_get, }; static int scmi_perf_set_notify_enabled(const struct scmi_protocol_handle *ph, diff --git a/drivers/idle/intel_idle.c b/drivers/idle/intel_idle.c index 445b19d20b9a..3e101719689a 100644 --- a/drivers/idle/intel_idle.c +++ b/drivers/idle/intel_idle.c @@ -928,16 +928,6 @@ static struct cpuidle_state adl_l_cstates[] __initdata = { .enter = NULL } }; -/* - * On Sapphire Rapids Xeon C1 has to be disabled if C1E is enabled, and vice - * versa. On SPR C1E is enabled only if "C1E promotion" bit is set in - * MSR_IA32_POWER_CTL. But in this case there effectively no C1, because C1 - * requests are promoted to C1E. If the "C1E promotion" bit is cleared, then - * both C1 and C1E requests end up with C1, so there is effectively no C1E. - * - * By default we enable C1 and disable C1E by marking it with - * 'CPUIDLE_FLAG_UNUSABLE'. - */ static struct cpuidle_state spr_cstates[] __initdata = { { .name = "C1", @@ -950,8 +940,7 @@ static struct cpuidle_state spr_cstates[] __initdata = { { .name = "C1E", .desc = "MWAIT 0x01", - .flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE | - CPUIDLE_FLAG_UNUSABLE, + .flags = MWAIT2flg(0x01) | CPUIDLE_FLAG_ALWAYS_ENABLE, .exit_latency = 2, .target_residency = 4, .enter = &intel_idle, @@ -1774,17 +1763,6 @@ static void __init spr_idle_state_table_update(void) { unsigned long long msr; - /* Check if user prefers C1E over C1. */ - if ((preferred_states_mask & BIT(2)) && - !(preferred_states_mask & BIT(1))) { - /* Disable C1 and enable C1E. */ - spr_cstates[0].flags |= CPUIDLE_FLAG_UNUSABLE; - spr_cstates[1].flags &= ~CPUIDLE_FLAG_UNUSABLE; - - /* Enable C1E using the "C1E promotion" bit. */ - c1e_promotion = C1E_PROMOTION_ENABLE; - } - /* * By default, the C6 state assumes the worst-case scenario of package * C6. However, if PC6 is disabled, we update the numbers to match diff --git a/drivers/opp/of.c b/drivers/opp/of.c index 30394929d700..eb89c9a75985 100644 --- a/drivers/opp/of.c +++ b/drivers/opp/of.c @@ -1443,12 +1443,12 @@ EXPORT_SYMBOL_GPL(dev_pm_opp_get_of_node); * It provides the power used by @dev at @kHz if it is the frequency of an * existing OPP, or at the frequency of the first OPP above @kHz otherwise * (see dev_pm_opp_find_freq_ceil()). This function updates @kHz to the ceiled - * frequency and @mW to the associated power. + * frequency and @uW to the associated power. * * Returns 0 on success or a proper -EINVAL value in case of error. */ static int __maybe_unused -_get_dt_power(struct device *dev, unsigned long *mW, unsigned long *kHz) +_get_dt_power(struct device *dev, unsigned long *uW, unsigned long *kHz) { struct dev_pm_opp *opp; unsigned long opp_freq, opp_power; @@ -1465,7 +1465,7 @@ _get_dt_power(struct device *dev, unsigned long *mW, unsigned long *kHz) return -EINVAL; *kHz = opp_freq / 1000; - *mW = opp_power / 1000; + *uW = opp_power; return 0; } @@ -1475,14 +1475,14 @@ _get_dt_power(struct device *dev, unsigned long *mW, unsigned long *kHz) * This computes the power estimated by @dev at @kHz if it is the frequency * of an existing OPP, or at the frequency of the first OPP above @kHz otherwise * (see dev_pm_opp_find_freq_ceil()). This function updates @kHz to the ceiled - * frequency and @mW to the associated power. The power is estimated as + * frequency and @uW to the associated power. The power is estimated as * P = C * V^2 * f with C being the device's capacitance and V and f * respectively the voltage and frequency of the OPP. * * Returns -EINVAL if the power calculation failed because of missing * parameters, 0 otherwise. */ -static int __maybe_unused _get_power(struct device *dev, unsigned long *mW, +static int __maybe_unused _get_power(struct device *dev, unsigned long *uW, unsigned long *kHz) { struct dev_pm_opp *opp; @@ -1512,9 +1512,10 @@ static int __maybe_unused _get_power(struct device *dev, unsigned long *mW, return -EINVAL; tmp = (u64)cap * mV * mV * (Hz / 1000000); - do_div(tmp, 1000000000); + /* Provide power in micro-Watts */ + do_div(tmp, 1000000); - *mW = (unsigned long)tmp; + *uW = (unsigned long)tmp; *kHz = Hz / 1000; return 0; diff --git a/drivers/powercap/dtpm_cpu.c b/drivers/powercap/dtpm_cpu.c index 6a88eb7e9f75..2ff7717530bf 100644 --- a/drivers/powercap/dtpm_cpu.c +++ b/drivers/powercap/dtpm_cpu.c @@ -53,7 +53,7 @@ static u64 set_pd_power_limit(struct dtpm *dtpm, u64 power_limit) for (i = 0; i < pd->nr_perf_states; i++) { - power = pd->table[i].power * MICROWATT_PER_MILLIWATT * nr_cpus; + power = pd->table[i].power * nr_cpus; if (power > power_limit) break; @@ -63,8 +63,7 @@ static u64 set_pd_power_limit(struct dtpm *dtpm, u64 power_limit) freq_qos_update_request(&dtpm_cpu->qos_req, freq); - power_limit = pd->table[i - 1].power * - MICROWATT_PER_MILLIWATT * nr_cpus; + power_limit = pd->table[i - 1].power * nr_cpus; return power_limit; } diff --git a/drivers/powercap/intel_rapl_common.c b/drivers/powercap/intel_rapl_common.c index a9c99d9e8b42..21d624f9f5fb 100644 --- a/drivers/powercap/intel_rapl_common.c +++ b/drivers/powercap/intel_rapl_common.c @@ -1109,6 +1109,7 @@ static const struct x86_cpu_id rapl_ids[] __initconst = { X86_MATCH_INTEL_FAM6_MODEL(ALDERLAKE_L, &rapl_defaults_core), X86_MATCH_INTEL_FAM6_MODEL(ALDERLAKE_N, &rapl_defaults_core), X86_MATCH_INTEL_FAM6_MODEL(RAPTORLAKE, &rapl_defaults_core), + X86_MATCH_INTEL_FAM6_MODEL(RAPTORLAKE_P, &rapl_defaults_core), X86_MATCH_INTEL_FAM6_MODEL(SAPPHIRERAPIDS_X, &rapl_defaults_spr_server), X86_MATCH_INTEL_FAM6_MODEL(LAKEFIELD, &rapl_defaults_core), diff --git a/drivers/powercap/intel_rapl_msr.c b/drivers/powercap/intel_rapl_msr.c index 9d23984d8931..bc6adda58883 100644 --- a/drivers/powercap/intel_rapl_msr.c +++ b/drivers/powercap/intel_rapl_msr.c @@ -140,7 +140,9 @@ static const struct x86_cpu_id pl4_support_ids[] = { { X86_VENDOR_INTEL, 6, INTEL_FAM6_TIGERLAKE_L, X86_FEATURE_ANY }, { X86_VENDOR_INTEL, 6, INTEL_FAM6_ALDERLAKE, X86_FEATURE_ANY }, { X86_VENDOR_INTEL, 6, INTEL_FAM6_ALDERLAKE_L, X86_FEATURE_ANY }, + { X86_VENDOR_INTEL, 6, INTEL_FAM6_ALDERLAKE_N, X86_FEATURE_ANY }, { X86_VENDOR_INTEL, 6, INTEL_FAM6_RAPTORLAKE, X86_FEATURE_ANY }, + { X86_VENDOR_INTEL, 6, INTEL_FAM6_RAPTORLAKE_P, X86_FEATURE_ANY }, {} }; diff --git a/drivers/thermal/cpufreq_cooling.c b/drivers/thermal/cpufreq_cooling.c index b263b0fde03c..cb75f1365615 100644 --- a/drivers/thermal/cpufreq_cooling.c +++ b/drivers/thermal/cpufreq_cooling.c @@ -21,6 +21,7 @@ #include <linux/pm_qos.h> #include <linux/slab.h> #include <linux/thermal.h> +#include <linux/units.h> #include <trace/events/thermal.h> @@ -101,6 +102,7 @@ static unsigned long get_level(struct cpufreq_cooling_device *cpufreq_cdev, static u32 cpu_freq_to_power(struct cpufreq_cooling_device *cpufreq_cdev, u32 freq) { + unsigned long power_mw; int i; for (i = cpufreq_cdev->max_level - 1; i >= 0; i--) { @@ -108,16 +110,23 @@ static u32 cpu_freq_to_power(struct cpufreq_cooling_device *cpufreq_cdev, break; } - return cpufreq_cdev->em->table[i + 1].power; + power_mw = cpufreq_cdev->em->table[i + 1].power; + power_mw /= MICROWATT_PER_MILLIWATT; + + return power_mw; } static u32 cpu_power_to_freq(struct cpufreq_cooling_device *cpufreq_cdev, u32 power) { + unsigned long em_power_mw; int i; for (i = cpufreq_cdev->max_level; i > 0; i--) { - if (power >= cpufreq_cdev->em->table[i].power) + /* Convert EM power to milli-Watts to make safe comparison */ + em_power_mw = cpufreq_cdev->em->table[i].power; + em_power_mw /= MICROWATT_PER_MILLIWATT; + if (power >= em_power_mw) break; } diff --git a/drivers/thermal/devfreq_cooling.c b/drivers/thermal/devfreq_cooling.c index 8c76f9655e57..8d1260f65061 100644 --- a/drivers/thermal/devfreq_cooling.c +++ b/drivers/thermal/devfreq_cooling.c @@ -200,7 +200,11 @@ static int devfreq_cooling_get_requested_power(struct thermal_cooling_device *cd res = dfc->power_ops->get_real_power(df, power, freq, voltage); if (!res) { state = dfc->capped_state; + + /* Convert EM power into milli-Watts first */ dfc->res_util = dfc->em_pd->table[state].power; + dfc->res_util /= MICROWATT_PER_MILLIWATT; + dfc->res_util *= SCALE_ERROR_MITIGATION; if (*power > 1) @@ -218,8 +222,10 @@ static int devfreq_cooling_get_requested_power(struct thermal_cooling_device *cd _normalize_load(&status); - /* Scale power for utilization */ + /* Convert EM power into milli-Watts first */ *power = dfc->em_pd->table[perf_idx].power; + *power /= MICROWATT_PER_MILLIWATT; + /* Scale power for utilization */ *power *= status.busy_time; *power >>= 10; } @@ -244,6 +250,7 @@ static int devfreq_cooling_state2power(struct thermal_cooling_device *cdev, perf_idx = dfc->max_state - state; *power = dfc->em_pd->table[perf_idx].power; + *power /= MICROWATT_PER_MILLIWATT; return 0; } @@ -254,7 +261,7 @@ static int devfreq_cooling_power2state(struct thermal_cooling_device *cdev, struct devfreq_cooling_device *dfc = cdev->devdata; struct devfreq *df = dfc->devfreq; struct devfreq_dev_status status; - unsigned long freq; + unsigned long freq, em_power_mw; s32 est_power; int i; @@ -279,9 +286,13 @@ static int devfreq_cooling_power2state(struct thermal_cooling_device *cdev, * Find the first cooling state that is within the power * budget. The EM power table is sorted ascending. */ - for (i = dfc->max_state; i > 0; i--) - if (est_power >= dfc->em_pd->table[i].power) + for (i = dfc->max_state; i > 0; i--) { + /* Convert EM power to milli-Watts to make safe comparison */ + em_power_mw = dfc->em_pd->table[i].power; + em_power_mw /= MICROWATT_PER_MILLIWATT; + if (est_power >= em_power_mw) break; + } *state = dfc->max_state - i; dfc->capped_state = *state; diff --git a/include/linux/energy_model.h b/include/linux/energy_model.h index 8419bffb4398..b9caa01dfac4 100644 --- a/include/linux/energy_model.h +++ b/include/linux/energy_model.h @@ -62,7 +62,7 @@ struct em_perf_domain { /* * em_perf_domain flags: * - * EM_PERF_DOMAIN_MILLIWATTS: The power values are in milli-Watts or some + * EM_PERF_DOMAIN_MICROWATTS: The power values are in micro-Watts or some * other scale. * * EM_PERF_DOMAIN_SKIP_INEFFICIENCIES: Skip inefficient states when estimating @@ -71,7 +71,7 @@ struct em_perf_domain { * EM_PERF_DOMAIN_ARTIFICIAL: The power values are artificial and might be * created by platform missing real power information */ -#define EM_PERF_DOMAIN_MILLIWATTS BIT(0) +#define EM_PERF_DOMAIN_MICROWATTS BIT(0) #define EM_PERF_DOMAIN_SKIP_INEFFICIENCIES BIT(1) #define EM_PERF_DOMAIN_ARTIFICIAL BIT(2) @@ -79,22 +79,44 @@ struct em_perf_domain { #define em_is_artificial(em) ((em)->flags & EM_PERF_DOMAIN_ARTIFICIAL) #ifdef CONFIG_ENERGY_MODEL -#define EM_MAX_POWER 0xFFFF +/* + * The max power value in micro-Watts. The limit of 64 Watts is set as + * a safety net to not overflow multiplications on 32bit platforms. The + * 32bit value limit for total Perf Domain power implies a limit of + * maximum CPUs in such domain to 64. + */ +#define EM_MAX_POWER (64000000) /* 64 Watts */ + +/* + * To avoid possible energy estimation overflow on 32bit machines add + * limits to number of CPUs in the Perf. Domain. + * We are safe on 64bit machine, thus some big number. + */ +#ifdef CONFIG_64BIT +#define EM_MAX_NUM_CPUS 4096 +#else +#define EM_MAX_NUM_CPUS 16 +#endif /* - * Increase resolution of energy estimation calculations for 64-bit - * architectures. The extra resolution improves decision made by EAS for the - * task placement when two Performance Domains might provide similar energy - * estimation values (w/o better resolution the values could be equal). + * To avoid an overflow on 32bit machines while calculating the energy + * use a different order in the operation. First divide by the 'cpu_scale' + * which would reduce big value stored in the 'cost' field, then multiply by + * the 'sum_util'. This would allow to handle existing platforms, which have + * e.g. power ~1.3 Watt at max freq, so the 'cost' value > 1mln micro-Watts. + * In such scenario, where there are 4 CPUs in the Perf. Domain the 'sum_util' + * could be 4096, then multiplication: 'cost' * 'sum_util' would overflow. + * This reordering of operations has some limitations, we lose small + * precision in the estimation (comparing to 64bit platform w/o reordering). * - * We increase resolution only if we have enough bits to allow this increased - * resolution (i.e. 64-bit). The costs for increasing resolution when 32-bit - * are pretty high and the returns do not justify the increased costs. + * We are safe on 64bit machine. */ #ifdef CONFIG_64BIT -#define em_scale_power(p) ((p) * 1000) +#define em_estimate_energy(cost, sum_util, scale_cpu) \ + (((cost) * (sum_util)) / (scale_cpu)) #else -#define em_scale_power(p) (p) +#define em_estimate_energy(cost, sum_util, scale_cpu) \ + (((cost) / (scale_cpu)) * (sum_util)) #endif struct em_data_callback { @@ -112,7 +134,7 @@ struct em_data_callback { * and frequency. * * In case of CPUs, the power is the one of a single CPU in the domain, - * expressed in milli-Watts or an abstract scale. It is expected to + * expressed in micro-Watts or an abstract scale. It is expected to * fit in the [0, EM_MAX_POWER] range. * * Return 0 on success. @@ -148,7 +170,7 @@ struct em_perf_domain *em_cpu_get(int cpu); struct em_perf_domain *em_pd_get(struct device *dev); int em_dev_register_perf_domain(struct device *dev, unsigned int nr_states, struct em_data_callback *cb, cpumask_t *span, - bool milliwatts); + bool microwatts); void em_dev_unregister_perf_domain(struct device *dev); /** @@ -273,7 +295,7 @@ static inline unsigned long em_cpu_energy(struct em_perf_domain *pd, * pd_nrg = ------------------------ (4) * scale_cpu */ - return ps->cost * sum_util / scale_cpu; + return em_estimate_energy(ps->cost, sum_util, scale_cpu); } /** @@ -297,7 +319,7 @@ struct em_data_callback {}; static inline int em_dev_register_perf_domain(struct device *dev, unsigned int nr_states, struct em_data_callback *cb, cpumask_t *span, - bool milliwatts) + bool microwatts) { return -EINVAL; } diff --git a/include/linux/pm_wakeup.h b/include/linux/pm_wakeup.h index 196a157456aa..77f4849e3418 100644 --- a/include/linux/pm_wakeup.h +++ b/include/linux/pm_wakeup.h @@ -109,7 +109,6 @@ extern struct wakeup_source *wakeup_sources_walk_next(struct wakeup_source *ws); extern int device_wakeup_enable(struct device *dev); extern int device_wakeup_disable(struct device *dev); extern void device_set_wakeup_capable(struct device *dev, bool capable); -extern int device_init_wakeup(struct device *dev, bool val); extern int device_set_wakeup_enable(struct device *dev, bool enable); extern void __pm_stay_awake(struct wakeup_source *ws); extern void pm_stay_awake(struct device *dev); @@ -167,13 +166,6 @@ static inline int device_set_wakeup_enable(struct device *dev, bool enable) return 0; } -static inline int device_init_wakeup(struct device *dev, bool val) -{ - device_set_wakeup_capable(dev, val); - device_set_wakeup_enable(dev, val); - return 0; -} - static inline bool device_may_wakeup(struct device *dev) { return dev->power.can_wakeup && dev->power.should_wakeup; @@ -217,4 +209,27 @@ static inline void pm_wakeup_hard_event(struct device *dev) return pm_wakeup_dev_event(dev, 0, true); } +/** + * device_init_wakeup - Device wakeup initialization. + * @dev: Device to handle. + * @enable: Whether or not to enable @dev as a wakeup device. + * + * By default, most devices should leave wakeup disabled. The exceptions are + * devices that everyone expects to be wakeup sources: keyboards, power buttons, + * possibly network interfaces, etc. Also, devices that don't generate their + * own wakeup requests but merely forward requests from one bus to another + * (like PCI bridges) should have wakeup enabled by default. + */ +static inline int device_init_wakeup(struct device *dev, bool enable) +{ + if (enable) { + device_set_wakeup_capable(dev, true); + return device_wakeup_enable(dev); + } else { + device_wakeup_disable(dev); + device_set_wakeup_capable(dev, false); + return 0; + } +} + #endif /* _LINUX_PM_WAKEUP_H */ diff --git a/include/linux/scmi_protocol.h b/include/linux/scmi_protocol.h index 7f4f9df1b20f..a193884ecf2b 100644 --- a/include/linux/scmi_protocol.h +++ b/include/linux/scmi_protocol.h @@ -60,6 +60,12 @@ struct scmi_clock_info { }; }; +enum scmi_power_scale { + SCMI_POWER_BOGOWATTS, + SCMI_POWER_MILLIWATTS, + SCMI_POWER_MICROWATTS +}; + struct scmi_handle; struct scmi_device; struct scmi_protocol_handle; @@ -135,7 +141,7 @@ struct scmi_perf_proto_ops { unsigned long *rate, unsigned long *power); bool (*fast_switch_possible)(const struct scmi_protocol_handle *ph, struct device *dev); - bool (*power_scale_mw_get)(const struct scmi_protocol_handle *ph); + enum scmi_power_scale (*power_scale_get)(const struct scmi_protocol_handle *ph); }; /** diff --git a/include/trace/events/power.h b/include/trace/events/power.h index af5018aa9517..c708521e4ed5 100644 --- a/include/trace/events/power.h +++ b/include/trace/events/power.h @@ -500,6 +500,35 @@ DEFINE_EVENT(dev_pm_qos_request, dev_pm_qos_remove_request, TP_ARGS(name, type, new_value) ); + +TRACE_EVENT(guest_halt_poll_ns, + + TP_PROTO(bool grow, unsigned int new, unsigned int old), + + TP_ARGS(grow, new, old), + + TP_STRUCT__entry( + __field(bool, grow) + __field(unsigned int, new) + __field(unsigned int, old) + ), + + TP_fast_assign( + __entry->grow = grow; + __entry->new = new; + __entry->old = old; + ), + + TP_printk("halt_poll_ns %u (%s %u)", + __entry->new, + __entry->grow ? "grow" : "shrink", + __entry->old) +); + +#define trace_guest_halt_poll_ns_grow(new, old) \ + trace_guest_halt_poll_ns(true, new, old) +#define trace_guest_halt_poll_ns_shrink(new, old) \ + trace_guest_halt_poll_ns(false, new, old) #endif /* _TRACE_POWER_H */ /* This part must be outside protection */ diff --git a/kernel/power/energy_model.c b/kernel/power/energy_model.c index 6c373f2960e7..f82111837b8d 100644 --- a/kernel/power/energy_model.c +++ b/kernel/power/energy_model.c @@ -145,7 +145,7 @@ static int em_create_perf_table(struct device *dev, struct em_perf_domain *pd, /* * The power returned by active_state() is expected to be - * positive and to fit into 16 bits. + * positive and be in range. */ if (!power || power > EM_MAX_POWER) { dev_err(dev, "EM: invalid power: %lu\n", @@ -170,7 +170,7 @@ static int em_create_perf_table(struct device *dev, struct em_perf_domain *pd, goto free_ps_table; } } else { - power_res = em_scale_power(table[i].power); + power_res = table[i].power; cost = div64_u64(fmax * power_res, table[i].frequency); } @@ -201,9 +201,17 @@ static int em_create_pd(struct device *dev, int nr_states, { struct em_perf_domain *pd; struct device *cpu_dev; - int cpu, ret; + int cpu, ret, num_cpus; if (_is_cpu_device(dev)) { + num_cpus = cpumask_weight(cpus); + + /* Prevent max possible energy calculation to not overflow */ + if (num_cpus > EM_MAX_NUM_CPUS) { + dev_err(dev, "EM: too many CPUs, overflow possible\n"); + return -EINVAL; + } + pd = kzalloc(sizeof(*pd) + cpumask_size(), GFP_KERNEL); if (!pd) return -ENOMEM; @@ -314,13 +322,13 @@ EXPORT_SYMBOL_GPL(em_cpu_get); * @cpus : Pointer to cpumask_t, which in case of a CPU device is * obligatory. It can be taken from i.e. 'policy->cpus'. For other * type of devices this should be set to NULL. - * @milliwatts : Flag indicating that the power values are in milliWatts or + * @microwatts : Flag indicating that the power values are in micro-Watts or * in some other scale. It must be set properly. * * Create Energy Model tables for a performance domain using the callbacks * defined in cb. * - * The @milliwatts is important to set with correct value. Some kernel + * The @microwatts is important to set with correct value. Some kernel * sub-systems might rely on this flag and check if all devices in the EM are * using the same scale. * @@ -331,7 +339,7 @@ EXPORT_SYMBOL_GPL(em_cpu_get); */ int em_dev_register_perf_domain(struct device *dev, unsigned int nr_states, struct em_data_callback *cb, cpumask_t *cpus, - bool milliwatts) + bool microwatts) { unsigned long cap, prev_cap = 0; unsigned long flags = 0; @@ -381,8 +389,8 @@ int em_dev_register_perf_domain(struct device *dev, unsigned int nr_states, } } - if (milliwatts) - flags |= EM_PERF_DOMAIN_MILLIWATTS; + if (microwatts) + flags |= EM_PERF_DOMAIN_MICROWATTS; else if (cb->get_cost) flags |= EM_PERF_DOMAIN_ARTIFICIAL; diff --git a/kernel/power/qos.c b/kernel/power/qos.c index ec7e1e85923e..af51ed6d45ef 100644 --- a/kernel/power/qos.c +++ b/kernel/power/qos.c @@ -531,7 +531,7 @@ int freq_qos_add_request(struct freq_constraints *qos, { int ret; - if (IS_ERR_OR_NULL(qos) || !req) + if (IS_ERR_OR_NULL(qos) || !req || value < 0) return -EINVAL; if (WARN(freq_qos_request_active(req), @@ -563,7 +563,7 @@ EXPORT_SYMBOL_GPL(freq_qos_add_request); */ int freq_qos_update_request(struct freq_qos_request *req, s32 new_value) { - if (!req) + if (!req || new_value < 0) return -EINVAL; if (WARN(!freq_qos_request_active(req), diff --git a/kernel/power/user.c b/kernel/power/user.c index ad241b4ff64c..d43c2aa583b2 100644 --- a/kernel/power/user.c +++ b/kernel/power/user.c @@ -26,6 +26,7 @@ #include "power.h" +static bool need_wait; static struct snapshot_data { struct snapshot_handle handle; @@ -78,7 +79,7 @@ static int snapshot_open(struct inode *inode, struct file *filp) * Resuming. We may need to wait for the image device to * appear. */ - wait_for_device_probe(); + need_wait = true; data->swap = -1; data->mode = O_WRONLY; @@ -168,6 +169,11 @@ static ssize_t snapshot_write(struct file *filp, const char __user *buf, ssize_t res; loff_t pg_offp = *offp & ~PAGE_MASK; + if (need_wait) { + wait_for_device_probe(); + need_wait = false; + } + lock_system_sleep(); data = filp->private_data; @@ -244,6 +250,11 @@ static long snapshot_ioctl(struct file *filp, unsigned int cmd, loff_t size; sector_t offset; + if (need_wait) { + wait_for_device_probe(); + need_wait = false; + } + if (_IOC_TYPE(cmd) != SNAPSHOT_IOC_MAGIC) return -ENOTTY; if (_IOC_NR(cmd) > SNAPSHOT_IOC_MAXNR) diff --git a/tools/power/pm-graph/README b/tools/power/pm-graph/README index da468bd510ca..e6020c0d59ec 100644 --- a/tools/power/pm-graph/README +++ b/tools/power/pm-graph/README @@ -6,7 +6,7 @@ |_| |___/ |_| pm-graph: suspend/resume/boot timing analysis tools - Version: 5.8 + Version: 5.9 Author: Todd Brandt <todd.e.brandt@intel.com> Home Page: https://01.org/pm-graph @@ -97,8 +97,8 @@ (kernel/pre-3.15/enable_trace_events_suspend_resume.patch) (kernel/pre-3.15/enable_trace_events_device_pm_callback.patch) - If you're using a kernel older than 3.15.0, the following - additional kernel parameters are required: + If you're using bootgraph, or sleepgraph with a kernel older than 3.15.0, + the following additional kernel parameters are required: (e.g. in file /etc/default/grub) GRUB_CMDLINE_LINUX_DEFAULT="... initcall_debug log_buf_len=32M ..." diff --git a/tools/power/pm-graph/bootgraph.py b/tools/power/pm-graph/bootgraph.py index 2823cd3122f7..f96f50e0c336 100755 --- a/tools/power/pm-graph/bootgraph.py +++ b/tools/power/pm-graph/bootgraph.py @@ -69,22 +69,24 @@ class SystemValues(aslib.SystemValues): bootloader = 'grub' blexec = [] def __init__(self): - self.hostname = platform.node() + self.kernel, self.hostname = 'unknown', platform.node() self.testtime = datetime.now().strftime('%Y-%m-%d_%H:%M:%S') if os.path.exists('/proc/version'): fp = open('/proc/version', 'r') - val = fp.read().strip() + self.kernel = self.kernelVersion(fp.read().strip()) fp.close() - self.kernel = self.kernelVersion(val) - else: - self.kernel = 'unknown' self.testdir = datetime.now().strftime('boot-%y%m%d-%H%M%S') def kernelVersion(self, msg): - return msg.split()[2] + m = re.match('^[Ll]inux *[Vv]ersion *(?P<v>\S*) .*', msg) + if m: + return m.group('v') + return 'unknown' def checkFtraceKernelVersion(self): - val = tuple(map(int, self.kernel.split('-')[0].split('.'))) - if val >= (4, 10, 0): - return True + m = re.match('^(?P<x>[0-9]*)\.(?P<y>[0-9]*)\.(?P<z>[0-9]*).*', self.kernel) + if m: + val = tuple(map(int, m.groups())) + if val >= (4, 10, 0): + return True return False def kernelParams(self): cmdline = 'initcall_debug log_buf_len=32M' diff --git a/tools/power/pm-graph/config/custom-timeline-functions.cfg b/tools/power/pm-graph/config/custom-timeline-functions.cfg index 962e5768681c..4f80ad7d7275 100644 --- a/tools/power/pm-graph/config/custom-timeline-functions.cfg +++ b/tools/power/pm-graph/config/custom-timeline-functions.cfg @@ -125,7 +125,7 @@ acpi_suspend_begin: suspend_console: acpi_pm_prepare: syscore_suspend: -arch_thaw_secondary_cpus_end: +arch_enable_nonboot_cpus_end: syscore_resume: acpi_pm_finish: resume_console: diff --git a/tools/power/pm-graph/sleepgraph.py b/tools/power/pm-graph/sleepgraph.py index ffd50953a024..33981adcdd68 100755 --- a/tools/power/pm-graph/sleepgraph.py +++ b/tools/power/pm-graph/sleepgraph.py @@ -66,8 +66,13 @@ from threading import Thread from subprocess import call, Popen, PIPE import base64 +debugtiming = False +mystarttime = time.time() def pprint(msg): - print(msg) + if debugtiming: + print('[%09.3f] %s' % (time.time()-mystarttime, msg)) + else: + print(msg) sys.stdout.flush() def ascii(text): @@ -81,13 +86,14 @@ def ascii(text): # store system values and test parameters class SystemValues: title = 'SleepGraph' - version = '5.8' + version = '5.9' ansi = False rs = 0 display = '' gzip = False sync = False wifi = False + netfix = False verbose = False testlog = True dmesglog = True @@ -108,6 +114,7 @@ class SystemValues: cpucount = 0 memtotal = 204800 memfree = 204800 + osversion = '' srgap = 0 cgexp = False testdir = '' @@ -116,6 +123,7 @@ class SystemValues: fpdtpath = '/sys/firmware/acpi/tables/FPDT' epath = '/sys/kernel/debug/tracing/events/power/' pmdpath = '/sys/power/pm_debug_messages' + s0ixpath = '/sys/module/intel_pmc_core/parameters/warn_on_s0ix_failures' acpipath='/sys/module/acpi/parameters/debug_level' traceevents = [ 'suspend_resume', @@ -156,6 +164,7 @@ class SystemValues: ftop = False usetraceevents = False usetracemarkers = True + useftrace = True usekprobes = True usedevsrc = False useprocmon = False @@ -279,10 +288,16 @@ class SystemValues: 'intel_fbdev_set_suspend': {}, } infocmds = [ + [0, 'sysinfo', 'uname', '-a'], + [0, 'cpuinfo', 'head', '-7', '/proc/cpuinfo'], [0, 'kparams', 'cat', '/proc/cmdline'], [0, 'mcelog', 'mcelog'], [0, 'pcidevices', 'lspci', '-tv'], - [0, 'usbdevices', 'lsusb', '-t'], + [0, 'usbdevices', 'lsusb', '-tv'], + [0, 'acpidevices', 'sh', '-c', 'ls -l /sys/bus/acpi/devices/*/physical_node'], + [0, 's0ix_require', 'cat', '/sys/kernel/debug/pmc_core/substate_requirements'], + [0, 's0ix_debug', 'cat', '/sys/kernel/debug/pmc_core/slp_s0_debug_status'], + [1, 's0ix_residency', 'cat', '/sys/kernel/debug/pmc_core/slp_s0_residency_usec'], [1, 'interrupts', 'cat', '/proc/interrupts'], [1, 'wakeups', 'cat', '/sys/kernel/debug/wakeup_sources'], [2, 'gpecounts', 'sh', '-c', 'grep -v invalid /sys/firmware/acpi/interrupts/*'], @@ -358,8 +373,19 @@ class SystemValues: self.outputResult({'error':msg}) sys.exit(1) return False - def usable(self, file): - return (os.path.exists(file) and os.path.getsize(file) > 0) + def usable(self, file, ishtml=False): + if not os.path.exists(file) or os.path.getsize(file) < 1: + return False + if ishtml: + try: + fp = open(file, 'r') + res = fp.read(1000) + fp.close() + except: + return False + if '<html>' not in res: + return False + return True def getExec(self, cmd): try: fp = Popen(['which', cmd], stdout=PIPE, stderr=PIPE).stdout @@ -413,12 +439,16 @@ class SystemValues: r = info['bios-release-date'] if 'bios-release-date' in info else '' self.sysstamp = '# sysinfo | man:%s | plat:%s | cpu:%s | bios:%s | biosdate:%s | numcpu:%d | memsz:%d | memfr:%d' % \ (m, p, c, b, r, self.cpucount, self.memtotal, self.memfree) + if self.osversion: + self.sysstamp += ' | os:%s' % self.osversion def printSystemInfo(self, fatal=False): self.rootCheck(True) out = dmidecode(self.mempath, fatal) if len(out) < 1: return fmt = '%-24s: %s' + if self.osversion: + print(fmt % ('os-version', self.osversion)) for name in sorted(out): print(fmt % (name, out[name])) print(fmt % ('cpucount', ('%d' % self.cpucount))) @@ -426,20 +456,25 @@ class SystemValues: print(fmt % ('memfree', ('%d kB' % self.memfree))) def cpuInfo(self): self.cpucount = 0 - fp = open('/proc/cpuinfo', 'r') - for line in fp: - if re.match('^processor[ \t]*:[ \t]*[0-9]*', line): - self.cpucount += 1 - fp.close() - fp = open('/proc/meminfo', 'r') - for line in fp: - m = re.match('^MemTotal:[ \t]*(?P<sz>[0-9]*) *kB', line) - if m: - self.memtotal = int(m.group('sz')) - m = re.match('^MemFree:[ \t]*(?P<sz>[0-9]*) *kB', line) - if m: - self.memfree = int(m.group('sz')) - fp.close() + if os.path.exists('/proc/cpuinfo'): + with open('/proc/cpuinfo', 'r') as fp: + for line in fp: + if re.match('^processor[ \t]*:[ \t]*[0-9]*', line): + self.cpucount += 1 + if os.path.exists('/proc/meminfo'): + with open('/proc/meminfo', 'r') as fp: + for line in fp: + m = re.match('^MemTotal:[ \t]*(?P<sz>[0-9]*) *kB', line) + if m: + self.memtotal = int(m.group('sz')) + m = re.match('^MemFree:[ \t]*(?P<sz>[0-9]*) *kB', line) + if m: + self.memfree = int(m.group('sz')) + if os.path.exists('/etc/os-release'): + with open('/etc/os-release', 'r') as fp: + for line in fp: + if line.startswith('PRETTY_NAME='): + self.osversion = line[12:].strip().replace('"', '') def initTestOutput(self, name): self.prefix = self.hostname v = open('/proc/version', 'r').read().strip() @@ -698,6 +733,8 @@ class SystemValues: return False return True def fsetVal(self, val, path): + if not self.useftrace: + return False return self.setVal(val, self.tpath+path) def getVal(self, file): res = '' @@ -711,9 +748,11 @@ class SystemValues: pass return res def fgetVal(self, path): + if not self.useftrace: + return '' return self.getVal(self.tpath+path) def cleanupFtrace(self): - if(self.usecallgraph or self.usetraceevents or self.usedevsrc): + if self.useftrace: self.fsetVal('0', 'events/kprobes/enable') self.fsetVal('', 'kprobe_events') self.fsetVal('1024', 'buffer_size_kb') @@ -734,13 +773,14 @@ class SystemValues: return True return False def initFtrace(self, quiet=False): + if not self.useftrace: + return if not quiet: sysvals.printSystemInfo(False) pprint('INITIALIZING FTRACE...') # turn trace off self.fsetVal('0', 'tracing_on') self.cleanupFtrace() - self.testVal(self.pmdpath, 'basic', '1') # set the trace clock to global self.fsetVal('global', 'trace_clock') self.fsetVal('nop', 'current_tracer') @@ -766,6 +806,10 @@ class SystemValues: # set trace type self.fsetVal('function_graph', 'current_tracer') self.fsetVal('', 'set_ftrace_filter') + # temporary hack to fix https://bugzilla.kernel.org/show_bug.cgi?id=212761 + fp = open(self.tpath+'set_ftrace_notrace', 'w') + fp.write('native_queued_spin_lock_slowpath\ndev_driver_string') + fp.close() # set trace format options self.fsetVal('print-parent', 'trace_options') self.fsetVal('funcgraph-abstime', 'trace_options') @@ -846,6 +890,8 @@ class SystemValues: fp.write('# turbostat %s\n' % test['turbo']) if 'wifi' in test: fp.write('# wifi %s\n' % test['wifi']) + if 'netfix' in test: + fp.write('# netfix %s\n' % test['netfix']) if test['error'] or len(testdata) > 1: fp.write('# enter_sleep_error %s\n' % test['error']) return fp @@ -865,6 +911,8 @@ class SystemValues: fp.write('error%s: %s\n' % (n, testdata['error'])) else: fp.write('result%s: pass\n' % n) + if 'mode' in testdata: + fp.write('mode%s: %s\n' % (n, testdata['mode'])) for v in ['suspend', 'resume', 'boot', 'lastinit']: if v in testdata: fp.write('%s%s: %.3f\n' % (v, n, testdata[v])) @@ -901,6 +949,8 @@ class SystemValues: fp.write(text) fp.close() def dlog(self, text): + if not self.dmesgfile: + return self.putlog(self.dmesgfile, '# %s\n' % text) def flog(self, text): self.putlog(self.ftracefile, text) @@ -954,34 +1004,31 @@ class SystemValues: dirname = props[dev].syspath if not dirname or not os.path.exists(dirname): continue - with open(dirname+'/power/async') as fp: - text = fp.read() - props[dev].isasync = False - if 'enabled' in text: + props[dev].isasync = False + if os.path.exists(dirname+'/power/async'): + fp = open(dirname+'/power/async') + if 'enabled' in fp.read(): props[dev].isasync = True + fp.close() fields = os.listdir(dirname) - if 'product' in fields: - with open(dirname+'/product', 'rb') as fp: - props[dev].altname = ascii(fp.read()) - elif 'name' in fields: - with open(dirname+'/name', 'rb') as fp: - props[dev].altname = ascii(fp.read()) - elif 'model' in fields: - with open(dirname+'/model', 'rb') as fp: - props[dev].altname = ascii(fp.read()) - elif 'description' in fields: - with open(dirname+'/description', 'rb') as fp: - props[dev].altname = ascii(fp.read()) - elif 'id' in fields: - with open(dirname+'/id', 'rb') as fp: - props[dev].altname = ascii(fp.read()) - elif 'idVendor' in fields and 'idProduct' in fields: - idv, idp = '', '' - with open(dirname+'/idVendor', 'rb') as fp: - idv = ascii(fp.read()).strip() - with open(dirname+'/idProduct', 'rb') as fp: - idp = ascii(fp.read()).strip() - props[dev].altname = '%s:%s' % (idv, idp) + for file in ['product', 'name', 'model', 'description', 'id', 'idVendor']: + if file not in fields: + continue + try: + with open(os.path.join(dirname, file), 'rb') as fp: + props[dev].altname = ascii(fp.read()) + except: + continue + if file == 'idVendor': + idv, idp = props[dev].altname.strip(), '' + try: + with open(os.path.join(dirname, 'idProduct'), 'rb') as fp: + idp = ascii(fp.read()).strip() + except: + props[dev].altname = '' + break + props[dev].altname = '%s:%s' % (idv, idp) + break if props[dev].altname: out = props[dev].altname.strip().replace('\n', ' ')\ .replace(',', ' ').replace(';', ' ') @@ -1047,7 +1094,7 @@ class SystemValues: self.cmd1[name] = self.dictify(info, delta) elif not debug and delta and name in self.cmd1: before, after = self.cmd1[name], self.dictify(info, delta) - dinfo = ('\t%s\n' % before['@']) if '@' in before else '' + dinfo = ('\t%s\n' % before['@']) if '@' in before and len(before) > 1 else '' prefix = self.commonPrefix(list(before.keys())) for key in sorted(before): if key in after and before[key] != after[key]: @@ -1128,6 +1175,22 @@ class SystemValues: val = valline[idx] out.append('%s=%s' % (key, val)) return '|'.join(out) + def netfixon(self, net='both'): + cmd = self.getExec('netfix') + if not cmd: + return '' + fp = Popen([cmd, '-s', net, 'on'], stdout=PIPE, stderr=PIPE).stdout + out = ascii(fp.read()).strip() + fp.close() + return out + def wifiRepair(self): + out = self.netfixon('wifi') + if not out or 'error' in out.lower(): + return '' + m = re.match('WIFI \S* ONLINE (?P<action>\S*)', out) + if not m: + return 'dead' + return m.group('action') def wifiDetails(self, dev): try: info = open('/sys/class/net/%s/device/uevent' % dev, 'r').read().strip() @@ -1144,12 +1207,12 @@ class SystemValues: except: return '' for line in reversed(w.split('\n')): - m = re.match(' *(?P<dev>.*): (?P<stat>[0-9a-f]*) .*', w.split('\n')[-1]) + m = re.match(' *(?P<dev>.*): (?P<stat>[0-9a-f]*) .*', line) if not m or (dev and dev != m.group('dev')): continue return m.group('dev') return '' - def pollWifi(self, dev, timeout=60): + def pollWifi(self, dev, timeout=10): start = time.time() while (time.time() - start) < timeout: w = self.checkWifi(dev) @@ -1157,6 +1220,11 @@ class SystemValues: return '%s reconnected %.2f' % \ (self.wifiDetails(dev), max(0, time.time() - start)) time.sleep(0.01) + if self.netfix: + res = self.wifiRepair() + if res: + timeout = max(0, time.time() - start) + return '%s %s %d' % (self.wifiDetails(dev), res, timeout) return '%s timeout %d' % (self.wifiDetails(dev), timeout) def errorSummary(self, errinfo, msg): found = False @@ -1283,10 +1351,10 @@ sysvals = SystemValues() switchvalues = ['enable', 'disable', 'on', 'off', 'true', 'false', '1', '0'] switchoff = ['disable', 'off', 'false', '0'] suspendmodename = { - 'freeze': 'Freeze (S0)', - 'standby': 'Standby (S1)', - 'mem': 'Suspend (S3)', - 'disk': 'Hibernate (S4)' + 'standby': 'standby (S1)', + 'freeze': 'freeze (S2idle)', + 'mem': 'suspend (S3)', + 'disk': 'hibernate (S4)' } # Class: DevProps @@ -1376,6 +1444,7 @@ class Data: 'INVALID' : r'(?i).*\bINVALID\b.*', 'CRASH' : r'(?i).*\bCRASHED\b.*', 'TIMEOUT' : r'(?i).*\bTIMEOUT\b.*', + 'ABORT' : r'(?i).*\bABORT\b.*', 'IRQ' : r'.*\bgenirq: .*', 'TASKFAIL': r'.*Freezing of tasks *.*', 'ACPI' : r'.*\bACPI *(?P<b>[A-Za-z]*) *Error[: ].*', @@ -1724,9 +1793,9 @@ class Data: if 'waking' in self.dmesg[lp]: tCnt = self.dmesg[lp]['waking'][0] if self.dmesg[lp]['waking'][1] >= 0.001: - tTry = '-%.0f' % (round(self.dmesg[lp]['waking'][1] * 1000)) + tTry = '%.0f' % (round(self.dmesg[lp]['waking'][1] * 1000)) else: - tTry = '-%.3f' % (self.dmesg[lp]['waking'][1] * 1000) + tTry = '%.3f' % (self.dmesg[lp]['waking'][1] * 1000) text = '%.0f (%s ms waking %d times)' % (tL * 1000, tTry, tCnt) else: text = '%.0f' % (tL * 1000) @@ -2107,6 +2176,30 @@ class Data: # set resume complete to end at end marker if 'resume_complete' in dm: dm['resume_complete']['end'] = time + def initcall_debug_call(self, line, quick=False): + m = re.match('.*(\[ *)(?P<t>[0-9\.]*)(\]) .* (?P<f>.*)\: '+\ + 'PM: *calling .* @ (?P<n>.*), parent: (?P<p>.*)', line) + if not m: + m = re.match('.*(\[ *)(?P<t>[0-9\.]*)(\]) .* (?P<f>.*)\: '+\ + 'calling .* @ (?P<n>.*), parent: (?P<p>.*)', line) + if not m: + m = re.match('.*(\[ *)(?P<t>[0-9\.]*)(\]) calling '+\ + '(?P<f>.*)\+ @ (?P<n>.*), parent: (?P<p>.*)', line) + if m: + return True if quick else m.group('t', 'f', 'n', 'p') + return False if quick else ('', '', '', '') + def initcall_debug_return(self, line, quick=False): + m = re.match('.*(\[ *)(?P<t>[0-9\.]*)(\]) .* (?P<f>.*)\: PM: '+\ + '.* returned (?P<r>[0-9]*) after (?P<dt>[0-9]*) usecs', line) + if not m: + m = re.match('.*(\[ *)(?P<t>[0-9\.]*)(\]) .* (?P<f>.*)\: '+\ + '.* returned (?P<r>[0-9]*) after (?P<dt>[0-9]*) usecs', line) + if not m: + m = re.match('.*(\[ *)(?P<t>[0-9\.]*)(\]) call '+\ + '(?P<f>.*)\+ returned .* after (?P<dt>.*) usecs', line) + if m: + return True if quick else m.group('t', 'f', 'dt') + return False if quick else ('', '', '') def debugPrint(self): for p in self.sortedPhases(): list = self.dmesg[p]['list'] @@ -2880,10 +2973,11 @@ class TestProps: cmdlinefmt = '^# command \| (?P<cmd>.*)' kparamsfmt = '^# kparams \| (?P<kp>.*)' devpropfmt = '# Device Properties: .*' - pinfofmt = '# platform-(?P<val>[a-z,A-Z,0-9]*): (?P<info>.*)' + pinfofmt = '# platform-(?P<val>[a-z,A-Z,0-9,_]*): (?P<info>.*)' tracertypefmt = '# tracer: (?P<t>.*)' firmwarefmt = '# fwsuspend (?P<s>[0-9]*) fwresume (?P<r>[0-9]*)$' procexecfmt = 'ps - (?P<ps>.*)$' + procmultifmt = '@(?P<n>[0-9]*)\|(?P<ps>.*)$' ftrace_line_fmt_fg = \ '^ *(?P<time>[0-9\.]*) *\| *(?P<cpu>[0-9]*)\)'+\ ' *(?P<proc>.*)-(?P<pid>[0-9]*) *\|'+\ @@ -2893,6 +2987,9 @@ class TestProps: '(?P<flags>\S*) *(?P<time>[0-9\.]*): *'+\ '(?P<msg>.*)' machinesuspend = 'machine_suspend\[.*' + multiproclist = dict() + multiproctime = 0.0 + multiproccnt = 0 def __init__(self): self.stamp = '' self.sysinfo = '' @@ -3063,6 +3160,7 @@ class TestRun: self.ttemp = dict() class ProcessMonitor: + maxchars = 512 def __init__(self): self.proclist = dict() self.running = False @@ -3088,19 +3186,23 @@ class ProcessMonitor: if ujiff > 0 or kjiff > 0: running[pid] = ujiff + kjiff process.wait() - out = '' + out = [''] for pid in running: jiffies = running[pid] val = self.proclist[pid] - if out: - out += ',' - out += '%s-%s %d' % (val['name'], pid, jiffies) - return 'ps - '+out + if len(out[-1]) > self.maxchars: + out.append('') + elif len(out[-1]) > 0: + out[-1] += ',' + out[-1] += '%s-%s %d' % (val['name'], pid, jiffies) + if len(out) > 1: + for line in out: + sysvals.fsetVal('ps - @%d|%s' % (len(out), line), 'trace_marker') + else: + sysvals.fsetVal('ps - %s' % out[0], 'trace_marker') def processMonitor(self, tid): while self.running: - out = self.procstat() - if out: - sysvals.fsetVal(out, 'trace_marker') + self.procstat() def start(self): self.thread = Thread(target=self.processMonitor, args=(0,)) self.running = True @@ -3144,7 +3246,6 @@ def doesTraceLogHaveTraceEvents(): # Function: appendIncompleteTraceLog # Description: -# [deprecated for kernel 3.15 or newer] # Adds callgraph data which lacks trace event data. This is only # for timelines generated from 3.15 or older # Arguments: @@ -3246,6 +3347,61 @@ def appendIncompleteTraceLog(testruns): dev['ftrace'] = cg break +# Function: loadTraceLog +# Description: +# load the ftrace file into memory and fix up any ordering issues +# Output: +# TestProps instance and an array of lines in proper order +def loadTraceLog(): + tp, data, lines, trace = TestProps(), dict(), [], [] + tf = sysvals.openlog(sysvals.ftracefile, 'r') + for line in tf: + # remove any latent carriage returns + line = line.replace('\r\n', '') + if tp.stampInfo(line, sysvals): + continue + # ignore all other commented lines + if line[0] == '#': + continue + # ftrace line: parse only valid lines + m = re.match(tp.ftrace_line_fmt, line) + if(not m): + continue + dur = m.group('dur') if tp.cgformat else 'traceevent' + info = (m.group('time'), m.group('proc'), m.group('pid'), + m.group('msg'), dur) + # group the data by timestamp + t = float(info[0]) + if t in data: + data[t].append(info) + else: + data[t] = [info] + # we only care about trace event ordering + if (info[3].startswith('suspend_resume:') or \ + info[3].startswith('tracing_mark_write:')) and t not in trace: + trace.append(t) + tf.close() + for t in sorted(data): + first, last, blk = [], [], data[t] + if len(blk) > 1 and t in trace: + # move certain lines to the start or end of a timestamp block + for i in range(len(blk)): + if 'SUSPEND START' in blk[i][3]: + first.append(i) + elif re.match('.* timekeeping_freeze.*begin', blk[i][3]): + last.append(i) + elif re.match('.* timekeeping_freeze.*end', blk[i][3]): + first.append(i) + elif 'RESUME COMPLETE' in blk[i][3]: + last.append(i) + if len(first) == 1 and len(last) == 0: + blk.insert(0, blk.pop(first[0])) + elif len(last) == 1 and len(first) == 0: + blk.append(blk.pop(last[0])) + for info in blk: + lines.append(info) + return (tp, lines) + # Function: parseTraceLog # Description: # Analyze an ftrace log output file generated from this app during @@ -3271,32 +3427,12 @@ def parseTraceLog(live=False): # extract the callgraph and traceevent data s2idle_enter = hwsus = False - tp = TestProps() testruns, testdata = [], [] testrun, data, limbo = 0, 0, True - tf = sysvals.openlog(sysvals.ftracefile, 'r') phase = 'suspend_prepare' - for line in tf: - # remove any latent carriage returns - line = line.replace('\r\n', '') - if tp.stampInfo(line, sysvals): - continue - # ignore all other commented lines - if line[0] == '#': - continue - # ftrace line: parse only valid lines - m = re.match(tp.ftrace_line_fmt, line) - if(not m): - continue + tp, tf = loadTraceLog() + for m_time, m_proc, m_pid, m_msg, m_param3 in tf: # gather the basic message data from the line - m_time = m.group('time') - m_proc = m.group('proc') - m_pid = m.group('pid') - m_msg = m.group('msg') - if(tp.cgformat): - m_param3 = m.group('dur') - else: - m_param3 = 'traceevent' if(m_time and m_pid and m_msg): t = FTraceLine(m_time, m_msg, m_param3) pid = int(m_pid) @@ -3322,14 +3458,29 @@ def parseTraceLog(live=False): if t.type == 'tracing_mark_write': m = re.match(tp.procexecfmt, t.name) if(m): - proclist = dict() - for ps in m.group('ps').split(','): + parts, msg = 1, m.group('ps') + m = re.match(tp.procmultifmt, msg) + if(m): + parts, msg = int(m.group('n')), m.group('ps') + if tp.multiproccnt == 0: + tp.multiproctime = t.time + tp.multiproclist = dict() + proclist = tp.multiproclist + tp.multiproccnt += 1 + else: + proclist = dict() + tp.multiproccnt = 0 + for ps in msg.split(','): val = ps.split() - if not val: + if not val or len(val) != 2: continue name = val[0].replace('--', '-') proclist[name] = int(val[1]) - data.pstl[t.time] = proclist + if parts == 1: + data.pstl[t.time] = proclist + elif parts == tp.multiproccnt: + data.pstl[tp.multiproctime] = proclist + tp.multiproccnt = 0 continue # find the end of resume if(t.endMarker()): @@ -3545,7 +3696,6 @@ def parseTraceLog(live=False): testrun.ftemp[key].append(FTraceCallGraph(pid, sysvals)) if(res == -1): testrun.ftemp[key][-1].addLine(t) - tf.close() if len(testdata) < 1: sysvals.vprint('WARNING: ftrace start marker is missing') if data and not data.devicegroups: @@ -3667,7 +3817,13 @@ def parseTraceLog(live=False): if p not in data.dmesg: if not terr: ph = p if 'machine' in p else lp - terr = '%s%s failed in %s phase' % (sysvals.suspendmode, tn, ph) + if p == 'suspend_machine': + sm = sysvals.suspendmode + if sm in suspendmodename: + sm = suspendmodename[sm] + terr = 'test%s did not enter %s power mode' % (tn, sm) + else: + terr = '%s%s failed in %s phase' % (sysvals.suspendmode, tn, ph) pprint('TEST%s FAILED: %s' % (tn, terr)) error.append(terr) if data.tSuspended == 0: @@ -3708,9 +3864,7 @@ def parseTraceLog(live=False): # Function: loadKernelLog # Description: -# [deprecated for kernel 3.15.0 or newer] # load the dmesg file into memory and fix up any ordering issues -# The dmesg filename is taken from sysvals # Output: # An array of empty Data objects with only their dmesgtext attributes set def loadKernelLog(): @@ -3736,7 +3890,8 @@ def loadKernelLog(): if(not m): continue msg = m.group("msg") - if(re.match('PM: Syncing filesystems.*', msg)): + if re.match('PM: Syncing filesystems.*', msg) or \ + re.match('PM: suspend entry.*', msg): if(data): testruns.append(data) data = Data(len(testruns)) @@ -3747,11 +3902,17 @@ def loadKernelLog(): if(m): sysvals.stamp['kernel'] = m.group('k') m = re.match('PM: Preparing system for (?P<m>.*) sleep', msg) - if(m): + if not m: + m = re.match('PM: Preparing system for sleep \((?P<m>.*)\)', msg) + if m: sysvals.stamp['mode'] = sysvals.suspendmode = m.group('m') data.dmesgtext.append(line) lf.close() + if sysvals.suspendmode == 's2idle': + sysvals.suspendmode = 'freeze' + elif sysvals.suspendmode == 'deep': + sysvals.suspendmode = 'mem' if data: testruns.append(data) if len(testruns) < 1: @@ -3762,12 +3923,9 @@ def loadKernelLog(): for data in testruns: last = '' for line in data.dmesgtext: - mc = re.match('.*(\[ *)(?P<t>[0-9\.]*)(\]) calling '+\ - '(?P<f>.*)\+ @ .*, parent: .*', line) - mr = re.match('.*(\[ *)(?P<t>[0-9\.]*)(\]) call '+\ - '(?P<f>.*)\+ returned .* after (?P<dt>.*) usecs', last) - if(mc and mr and (mc.group('t') == mr.group('t')) and - (mc.group('f') == mr.group('f'))): + ct, cf, n, p = data.initcall_debug_call(line) + rt, rf, l = data.initcall_debug_return(last) + if ct and rt and ct == rt and cf == rf: i = data.dmesgtext.index(last) j = data.dmesgtext.index(line) data.dmesgtext[i] = line @@ -3777,7 +3935,6 @@ def loadKernelLog(): # Function: parseKernelLog # Description: -# [deprecated for kernel 3.15.0 or newer] # Analyse a dmesg log output file generated from this app during # the execution phase. Create a set of device structures in memory # for subsequent formatting in the html output file @@ -3796,30 +3953,30 @@ def parseKernelLog(data): # dmesg phase match table dm = { - 'suspend_prepare': ['PM: Syncing filesystems.*'], - 'suspend': ['PM: Entering [a-z]* sleep.*', 'Suspending console.*'], - 'suspend_late': ['PM: suspend of devices complete after.*'], - 'suspend_noirq': ['PM: late suspend of devices complete after.*'], - 'suspend_machine': ['PM: noirq suspend of devices complete after.*'], - 'resume_machine': ['ACPI: Low-level resume complete.*'], - 'resume_noirq': ['ACPI: Waking up from system sleep state.*'], - 'resume_early': ['PM: noirq resume of devices complete after.*'], - 'resume': ['PM: early resume of devices complete after.*'], - 'resume_complete': ['PM: resume of devices complete after.*'], + 'suspend_prepare': ['PM: Syncing filesystems.*', 'PM: suspend entry.*'], + 'suspend': ['PM: Entering [a-z]* sleep.*', 'Suspending console.*', + 'PM: Suspending system .*'], + 'suspend_late': ['PM: suspend of devices complete after.*', + 'PM: freeze of devices complete after.*'], + 'suspend_noirq': ['PM: late suspend of devices complete after.*', + 'PM: late freeze of devices complete after.*'], + 'suspend_machine': ['PM: suspend-to-idle', + 'PM: noirq suspend of devices complete after.*', + 'PM: noirq freeze of devices complete after.*'], + 'resume_machine': ['PM: Timekeeping suspended for.*', + 'ACPI: Low-level resume complete.*', + 'ACPI: resume from mwait', + 'Suspended for [0-9\.]* seconds'], + 'resume_noirq': ['PM: resume from suspend-to-idle', + 'ACPI: Waking up from system sleep state.*'], + 'resume_early': ['PM: noirq resume of devices complete after.*', + 'PM: noirq restore of devices complete after.*'], + 'resume': ['PM: early resume of devices complete after.*', + 'PM: early restore of devices complete after.*'], + 'resume_complete': ['PM: resume of devices complete after.*', + 'PM: restore of devices complete after.*'], 'post_resume': ['.*Restarting tasks \.\.\..*'], } - if(sysvals.suspendmode == 'standby'): - dm['resume_machine'] = ['PM: Restoring platform NVS memory'] - elif(sysvals.suspendmode == 'disk'): - dm['suspend_late'] = ['PM: freeze of devices complete after.*'] - dm['suspend_noirq'] = ['PM: late freeze of devices complete after.*'] - dm['suspend_machine'] = ['PM: noirq freeze of devices complete after.*'] - dm['resume_machine'] = ['PM: Restoring platform NVS memory'] - dm['resume_early'] = ['PM: noirq restore of devices complete after.*'] - dm['resume'] = ['PM: early restore of devices complete after.*'] - dm['resume_complete'] = ['PM: restore of devices complete after.*'] - elif(sysvals.suspendmode == 'freeze'): - dm['resume_machine'] = ['ACPI: resume from mwait'] # action table (expected events that occur and show up in dmesg) at = { @@ -3867,12 +4024,13 @@ def parseKernelLog(data): for s in dm[p]: if(re.match(s, msg)): phasechange, phase = True, p + dm[p] = [s] break # hack for determining resume_machine end for freeze if(not sysvals.usetraceevents and sysvals.suspendmode == 'freeze' \ and phase == 'resume_machine' and \ - re.match('calling (?P<f>.*)\+ @ .*, parent: .*', msg)): + data.initcall_debug_call(line, True)): data.setPhase(phase, ktime, False) phase = 'resume_noirq' data.setPhase(phase, ktime, True) @@ -3945,26 +4103,18 @@ def parseKernelLog(data): # -- device callbacks -- if(phase in data.sortedPhases()): # device init call - if(re.match('calling (?P<f>.*)\+ @ .*, parent: .*', msg)): - sm = re.match('calling (?P<f>.*)\+ @ '+\ - '(?P<n>.*), parent: (?P<p>.*)', msg); - f = sm.group('f') - n = sm.group('n') - p = sm.group('p') - if(f and n and p): - data.newAction(phase, f, int(n), p, ktime, -1, '') - # device init return - elif(re.match('call (?P<f>.*)\+ returned .* after '+\ - '(?P<t>.*) usecs', msg)): - sm = re.match('call (?P<f>.*)\+ returned .* after '+\ - '(?P<t>.*) usecs(?P<a>.*)', msg); - f = sm.group('f') - t = sm.group('t') - list = data.dmesg[phase]['list'] - if(f in list): - dev = list[f] - dev['length'] = int(t) - dev['end'] = ktime + t, f, n, p = data.initcall_debug_call(line) + if t and f and n and p: + data.newAction(phase, f, int(n), p, ktime, -1, '') + else: + # device init return + t, f, l = data.initcall_debug_return(line) + if t and f and l: + list = data.dmesg[phase]['list'] + if(f in list): + dev = list[f] + dev['length'] = int(l) + dev['end'] = ktime # if trace events are not available, these are better than nothing if(not sysvals.usetraceevents): @@ -4006,6 +4156,8 @@ def parseKernelLog(data): # fill in any missing phases phasedef = data.phasedef terr, lp = '', 'suspend_prepare' + if lp not in data.dmesg: + doError('dmesg log format has changed, could not find start of suspend') for p in sorted(phasedef, key=lambda k:phasedef[k]['order']): if p not in data.dmesg: if not terr: @@ -5302,7 +5454,7 @@ def executeSuspend(quiet=False): sv.dlog('read dmesg') sv.initdmesg() # start ftrace - if(sv.usecallgraph or sv.usetraceevents): + if sv.useftrace: if not quiet: pprint('START TRACING') sv.dlog('start ftrace tracing') @@ -5334,8 +5486,7 @@ def executeSuspend(quiet=False): sv.dlog('enable RTC wake alarm') sv.rtcWakeAlarmOn() # start of suspend trace marker - if(sv.usecallgraph or sv.usetraceevents): - sv.fsetVal(datetime.now().strftime(sv.tmstart), 'trace_marker') + sv.fsetVal(datetime.now().strftime(sv.tmstart), 'trace_marker') # predelay delay if(count == 1 and sv.predelay > 0): sv.fsetVal('WAIT %d' % sv.predelay, 'trace_marker') @@ -5384,11 +5535,17 @@ def executeSuspend(quiet=False): sv.fsetVal('WAIT END', 'trace_marker') # return from suspend pprint('RESUME COMPLETE') - if(sv.usecallgraph or sv.usetraceevents): - sv.fsetVal(datetime.now().strftime(sv.tmend), 'trace_marker') + sv.fsetVal(datetime.now().strftime(sv.tmend), 'trace_marker') if sv.wifi and wifi: tdata['wifi'] = sv.pollWifi(wifi) sv.dlog('wifi check, %s' % tdata['wifi']) + if sv.netfix: + netfixout = sv.netfixon('wired') + elif sv.netfix: + netfixout = sv.netfixon() + if sv.netfix and netfixout: + tdata['netfix'] = netfixout + sv.dlog('netfix, %s' % tdata['netfix']) if(sv.suspendmode == 'mem' or sv.suspendmode == 'command'): sv.dlog('read the ACPI FPDT') tdata['fw'] = getFPDT(False) @@ -5396,7 +5553,7 @@ def executeSuspend(quiet=False): sv.dlog('run the cmdinfo list after') cmdafter = sv.cmdinfo(False) # stop ftrace - if(sv.usecallgraph or sv.usetraceevents): + if sv.useftrace: if sv.useprocmon: sv.dlog('stop the process monitor') pm.stop() @@ -5407,7 +5564,7 @@ def executeSuspend(quiet=False): sysvals.dlog('EXECUTION TRACE END') sv.getdmesg(testdata) # grab a copy of the ftrace output - if(sv.usecallgraph or sv.usetraceevents): + if sv.useftrace: if not quiet: pprint('CAPTURING TRACE') op = sv.writeDatafileHeader(sv.ftracefile, testdata) @@ -5838,13 +5995,19 @@ def statusCheck(probecheck=False): pprint(' please choose one with -m') # check if ftrace is available - res = sysvals.colorText('NO') - ftgood = sysvals.verifyFtrace() - if(ftgood): - res = 'YES' - elif(sysvals.usecallgraph): - status = 'ftrace is not properly supported' - pprint(' is ftrace supported: %s' % res) + if sysvals.useftrace: + res = sysvals.colorText('NO') + sysvals.useftrace = sysvals.verifyFtrace() + efmt = '"{0}" uses ftrace, and it is not properly supported' + if sysvals.useftrace: + res = 'YES' + elif sysvals.usecallgraph: + status = efmt.format('-f') + elif sysvals.usedevsrc: + status = efmt.format('-dev') + elif sysvals.useprocmon: + status = efmt.format('-proc') + pprint(' is ftrace supported: %s' % res) # check if kprobes are available if sysvals.usekprobes: @@ -5857,8 +6020,8 @@ def statusCheck(probecheck=False): pprint(' are kprobes supported: %s' % res) # what data source are we using - res = 'DMESG' - if(ftgood): + res = 'DMESG (very limited, ftrace is preferred)' + if sysvals.useftrace: sysvals.usetraceevents = True for e in sysvals.traceevents: if not os.path.exists(sysvals.epath+e): @@ -5879,7 +6042,7 @@ def statusCheck(probecheck=False): pprint(' optional commands this tool may use for info:') no = sysvals.colorText('MISSING') yes = sysvals.colorText('FOUND', 32) - for c in ['turbostat', 'mcelog', 'lspci', 'lsusb']: + for c in ['turbostat', 'mcelog', 'lspci', 'lsusb', 'netfix']: if c == 'turbostat': res = yes if sysvals.haveTurbostat() else no else: @@ -5971,7 +6134,7 @@ def processData(live=False, quiet=False): if not sysvals.stamp: pprint('ERROR: data does not include the expected stamp') return (testruns, {'error': 'timeline generation failed'}) - shown = ['bios', 'biosdate', 'cpu', 'host', 'kernel', 'man', 'memfr', + shown = ['os', 'bios', 'biosdate', 'cpu', 'host', 'kernel', 'man', 'memfr', 'memsz', 'mode', 'numcpu', 'plat', 'time', 'wifi'] sysvals.vprint('System Info:') for key in sorted(sysvals.stamp): @@ -6052,6 +6215,8 @@ def runTest(n=0, quiet=False): if sysvals.display: ret = sysvals.displayControl('init') sysvals.dlog('xset display init, ret = %d' % ret) + sysvals.testVal(sysvals.pmdpath, 'basic', '1') + sysvals.testVal(sysvals.s0ixpath, 'basic', 'Y') sysvals.dlog('initialize ftrace') sysvals.initFtrace(quiet) @@ -6145,9 +6310,12 @@ def data_from_html(file, outpath, issues, fulldetail=False): elist[err[0]] += 1 for i in elist: ilist.append('%sx%d' % (i, elist[i]) if elist[i] > 1 else i) - wifi = find_in_html(html, 'Wifi Resume: ', '</td>') - if wifi: - extra['wifi'] = wifi + line = find_in_html(log, '# wifi ', '\n') + if line: + extra['wifi'] = line + line = find_in_html(log, '# netfix ', '\n') + if line: + extra['netfix'] = line low = find_in_html(html, 'freeze time: <b>', ' ms</b>') for lowstr in ['waking', '+']: if not low: @@ -6243,7 +6411,7 @@ def genHtml(subdir, force=False): sysvals.ftracefile = file sysvals.setOutputFile() if (sysvals.dmesgfile or sysvals.ftracefile) and sysvals.htmlfile and \ - (force or not sysvals.usable(sysvals.htmlfile)): + (force or not sysvals.usable(sysvals.htmlfile, True)): pprint('FTRACE: %s' % sysvals.ftracefile) if sysvals.dmesgfile: pprint('DMESG : %s' % sysvals.dmesgfile) @@ -6533,6 +6701,7 @@ def printHelp(): ' -skiphtml Run the test and capture the trace logs, but skip the timeline (default: disabled)\n'\ ' -result fn Export a results table to a text file for parsing.\n'\ ' -wifi If a wifi connection is available, check that it reconnects after resume.\n'\ + ' -netfix Use netfix to reset the network in the event it fails to resume.\n'\ ' [testprep]\n'\ ' -sync Sync the filesystems before starting the test\n'\ ' -rs on/off Enable/disable runtime suspend for all devices, restore all after test\n'\ @@ -6615,6 +6784,8 @@ if __name__ == '__main__': elif(arg == '-v'): pprint("Version %s" % sysvals.version) sys.exit(0) + elif(arg == '-debugtiming'): + debugtiming = True elif(arg == '-x2'): sysvals.execcount = 2 elif(arg == '-x2delay'): @@ -6657,6 +6828,8 @@ if __name__ == '__main__': sysvals.sync = True elif(arg == '-wifi'): sysvals.wifi = True + elif(arg == '-netfix'): + sysvals.netfix = True elif(arg == '-gzip'): sysvals.gzip = True elif(arg == '-info'): @@ -6819,7 +6992,7 @@ if __name__ == '__main__': sysvals.outdir = val sysvals.notestrun = True if(os.path.isdir(val) == False): - doError('%s is not accessible' % val) + doError('%s is not accesible' % val) elif(arg == '-filter'): try: val = next(args) @@ -6942,12 +7115,11 @@ if __name__ == '__main__': time.sleep(sysvals.multitest['delay']) fmt = 'suspend-%y%m%d-%H%M%S' sysvals.testdir = os.path.join(sysvals.outdir, datetime.now().strftime(fmt)) - ret = runTest(i+1, True) + ret = runTest(i+1, not sysvals.verbose) failcnt = 0 if not ret else failcnt + 1 if sysvals.maxfail > 0 and failcnt >= sysvals.maxfail: pprint('Maximum fail count of %d reached, aborting multitest' % (sysvals.maxfail)) break - time.sleep(5) sysvals.resetlog() sysvals.multistat(False, i, finish) if 'time' in sysvals.multitest and datetime.now() >= finish: |