David E. Box [Thu, 3 Jul 2025 02:28:23 +0000 (19:28 -0700)]
platform/x86/intel/pmt: Add PMT Discovery driver
This patch introduces a new driver to enumerate and expose Intel Platform
Monitoring Technology (PMT) capabilities via a simple discovery mechanism.
The PMT Discovery driver parses hardware-provided discovery tables from
Intel Out of Band Management Services Modules (OOBMSM) and extracts feature
information for various providers (such as TPMI, Telemetry, Crash Log,
etc). This unified interface simplifies the process of determining which
manageability and telemetry features are supported by a given platform.
This new feature is described in the Intel Platform Monitoring Technology
3.0 specification, section 6.6 Capability.
Key changes and additions:
New file drivers/platform/x86/intel/pmt/discovery.c:
– Implements the discovery logic to map the discovery resource, read
the feature discovery table, and validate feature parameters.
New file drivers/platform/x86/intel/pmt/features.c:
– Defines feature names, layouts, and associated capability masks.
– Provides a mapping between raw hardware attributes and sysfs
representations for easier integration with user-space tools.
New header include/linux/intel_pmt_features.h:
– Declares constants, masks, and feature identifiers used across the
PMT framework.
Sysfs integration:
– Feature attributes are exposed under /sys/class/intel_pmt.
– Each device is represented by a subfolder within the intel_pmt class,
named using its DBDF (Domain:Bus:Device.Function), e.g.:
features-0000:00:03.1
– Example directory layout for a device:
By exposing PMT feature details through sysfs and integrating with the
existing PMT class, this driver paves the way for more streamlined
integration of PMT-based manageability and telemetry tools.
David E. Box [Thu, 3 Jul 2025 02:28:22 +0000 (19:28 -0700)]
platform/x86/intel/vsec: Add new Discovery feature
Add the PCIe VSEC ID for new Intel Platform Monitoring Technology
Capability Discovery feature. Discovery provides detailed information for
the various Intel VSEC features. Also make the driver a supplier for
TPMI and Telemetry drivers which will use the information.
David E. Box [Thu, 3 Jul 2025 02:28:21 +0000 (19:28 -0700)]
platform/x86/intel/vsec: Skip driverless features
If a feature lacks a corresponding driver and that feature is also a
supplier, registering it would be prevent the consumer driver from probing.
Introduces logic to skip such features during device registration.
David E. Box [Thu, 3 Jul 2025 02:28:20 +0000 (19:28 -0700)]
platform/x86/intel/vsec: Skip absent features during initialization
Some VSEC features depend on the presence of supplier features that may not
always be present. To prevent unnecessary retries and device linking during
initialization, introduce logic to skip attempts to link consumers to
missing suppliers.
David E. Box [Thu, 3 Jul 2025 02:28:19 +0000 (19:28 -0700)]
platform/x86/intel/vsec: Add device links to enforce dependencies
New Intel VSEC features will have dependencies on other features, requiring
certain supplier drivers to be probed before their consumers. To enforce
this dependency ordering, introduce device links using device_link_add(),
ensuring that suppliers are fully registered before consumers are probed.
- Add device link tracking by storing supplier devices and tracking their
state.
- Implement intel_vsec_link_devices() to establish links between suppliers
and consumers based on feature dependencies.
- Add get_consumer_dependencies() to retrieve supplier-consumer
relationships.
- Modify feature registration logic:
* Consumers now check that all required suppliers are registered before
being initialized.
* suppliers_ready() verifies that all required supplier devices are
available.
- Prevent potential null consumer name issue in sysfs:
- Use dev_set_name() when creating auxiliary devices to ensure a
unique, non-null consumer name.
- Update intel_vsec_pci_probe() to loop up to the number of possible
features or when all devices are registered, whichever comes first.
- Introduce VSEC_CAP_UNUSED to prevent sub-features (registered via
exported APIs) from being mistakenly linked.
David E. Box [Thu, 3 Jul 2025 02:28:18 +0000 (19:28 -0700)]
platform/x86/intel/vsec: Create wrapper to walk PCI config space
Combine three PCI config space walkers — intel_vsec_walk_dvsec(),
intel_vsec_walk_vsec(), and intel_vsec_walk_header() — into a new wrapper
function, intel_vsec_feature_walk(). This refactoring simplifies the probe
logic and lays the groundwork for future patches that will loop over these
calls. No functional changes.
David E. Box [Thu, 3 Jul 2025 02:28:17 +0000 (19:28 -0700)]
platform/x86/intel/vsec: Add private data for per-device data
Introduce a new private structure, struct vsec_priv, to hold a pointer to
the platform-specific information. Although the driver didn’t previously
require this per-device data, adding it now lays the groundwork for
upcoming patches that will manage such data. No functional changes.
Derek J. Clark [Wed, 2 Jul 2025 03:38:26 +0000 (20:38 -0700)]
platform/x86: Add Lenovo Other Mode WMI Driver
Adds lenovo-wmi-other driver which provides the Lenovo "Other Mode" WMI
interface that comes on some Lenovo "Gaming Series" hardware. Provides a
firmware-attributes class which enables the use of tunable knobs for SPL,
SPPT, and FPPT.
Reviewed-by: Alok Tiwari <alok.a.tiwari@oracle.com> Reviewed-by: Armin Wolf <W_Armin@gmx.de> Signed-off-by: Derek J. Clark <derekjohn.clark@gmail.com> Link: https://lore.kernel.org/r/20250702033826.1057762-7-derekjohn.clark@gmail.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Derek J. Clark [Wed, 2 Jul 2025 03:38:25 +0000 (20:38 -0700)]
platform/x86: Add Lenovo Gamezone WMI Driver
Adds lenovo-wmi-gamezone driver which provides the Lenovo Gamezone WMI
interface that comes on Lenovo "Gaming Series" hardware. Provides ACPI
platform profiles over WMI.
Derek J. Clark [Wed, 2 Jul 2025 03:38:24 +0000 (20:38 -0700)]
platform/x86: Add Lenovo Capability Data 01 WMI Driver
Adds lenovo-wmi-capdata01 driver which provides the
LENOVO_CAPABILITY_DATA_01 WMI data block that comes on "Other Mode"
enabled hardware. Provides an interface for querying if a given
attribute is supported by the hardware, as well as its default_value,
max_value, min_value, and step increment.
Reviewed-by: Alok Tiwari <alok.a.tiwari@oracle.com> Reviewed-by: Armin Wolf <W_Armin@gmx.de> Signed-off-by: Derek J. Clark <derekjohn.clark@gmail.com> Link: https://lore.kernel.org/r/20250702033826.1057762-5-derekjohn.clark@gmail.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Derek J. Clark [Wed, 2 Jul 2025 03:38:23 +0000 (20:38 -0700)]
platform/x86: Add Lenovo WMI Events Driver
Adds lenovo-wmi-events driver. The events driver is designed as a
general entrypoint for all Lenovo WMI Events. It acts as a notification
chain head that will process event data and pass it on to registered
drivers so they can react to the events.
Currently only the Gamezone interface Thermal Mode Event GUID is
implemented in this driver. It is documented in the Gamezone
documentation.
Suggested-by: Armin Wolf <W_Armin@gmx.de> Reviewed-by: Alok Tiwari <alok.a.tiwari@oracle.com> Reviewed-by: Armin Wolf <W_Armin@gmx.de> Reviewed-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Derek J. Clark <derekjohn.clark@gmail.com> Link: https://lore.kernel.org/r/20250702033826.1057762-4-derekjohn.clark@gmail.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Derek J. Clark [Wed, 2 Jul 2025 03:38:22 +0000 (20:38 -0700)]
platform/x86: Add lenovo-wmi-helpers
Adds lenovo-wmi-helpers, which provides a common wrapper function for
wmidev_evaluate_method that does data validation and error handling.
Reviewed-by: Alok Tiwari <alok.a.tiwari@oracle.com> Reviewed-by: Armin Wolf <W_Armin@gmx.de> Reviewed-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Derek J. Clark <derekjohn.clark@gmail.com> Link: https://lore.kernel.org/r/20250702033826.1057762-3-derekjohn.clark@gmail.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Reviewed-by: Alok Tiwari <alok.a.tiwari@oracle.com> Reviewed-by: Armin Wolf <W_Armin@gmx.de> Reviewed-by: Mario Limonciello <mario.limonciello@amd.com> Signed-off-by: Derek J. Clark <derekjohn.clark@gmail.com> Link: https://lore.kernel.org/r/20250702033826.1057762-2-derekjohn.clark@gmail.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Ilpo Järvinen [Wed, 2 Jul 2025 10:21:23 +0000 (13:21 +0300)]
Merge branch 'fixes' into for-next
Merge fixes back into for-next to be able to take dell_rbu change that
is build on top of fixes material, and to bring lenovo related changes
in sync after the move under lenovo/ subdir in the for-next branch and
diverging changes in the fixes branch.
Kurt Borja [Mon, 30 Jun 2025 17:31:20 +0000 (14:31 -0300)]
platform/x86: think-lmi: Fix kobject cleanup
In tlmi_analyze(), allocated structs with an embedded kobject are freed
in error paths after the they were already initialized.
Fix this by first by avoiding the initialization of kobjects in
tlmi_analyze() and then by correctly cleaning them up in
tlmi_release_attr() using their kset's kobject list.
Fixes: a40cd7ef22fb ("platform/x86: think-lmi: Add WMI interface support on Lenovo platforms") Fixes: 30e78435d3bf ("platform/x86: think-lmi: Split kobject_init() and kobject_add() calls") Cc: stable@vger.kernel.org Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca> Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Kurt Borja <kuurtb@gmail.com> Link: https://lore.kernel.org/r/20250630-lmi-fix-v3-2-ce4f81c9c481@gmail.com Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Avoid entering tlmi_release_attr() in error paths if both ksets are not
yet created.
This is accomplished by initializing them side by side.
Reviewed-by: Mark Pearson <mpearson-lenovo@squebb.ca> Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Cc: stable@vger.kernel.org Signed-off-by: Kurt Borja <kuurtb@gmail.com> Link: https://lore.kernel.org/r/20250630-lmi-fix-v3-1-ce4f81c9c481@gmail.com Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Alok Tiwari [Mon, 30 Jun 2025 10:58:08 +0000 (03:58 -0700)]
platform/mellanox: mlxreg-lc: Fix logic error in power state check
Fixes a logic issue in mlxreg_lc_completion_notify() where the
intention was to check if MLXREG_LC_POWERED flag is not set before
powering on the device.
The original code used "state & ~MLXREG_LC_POWERED" to check for the
absence of the POWERED bit. However this condition evaluates to true
even when other bits are set, leading to potentially incorrect
behavior.
Corrected the logic to explicitly check for the absence of
MLXREG_LC_POWERED using !(state & MLXREG_LC_POWERED).
Fixes: 62f9529b8d5c ("platform/mellanox: mlxreg-lc: Add initial support for Nvidia line card devices") Suggested-by: Vadim Pasternak <vadimp@nvidia.com> Signed-off-by: Alok Tiwari <alok.a.tiwari@oracle.com> Link: https://lore.kernel.org/r/20250630105812.601014-1-alok.a.tiwari@oracle.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Move FCH header to a location accessible by all archs
A new header fch.h was created to store registers used by different AMD
drivers. This header was included by i2c-piix4 in
commit 624b0d5696a8 ("i2c: piix4, x86/platform: Move the SB800 PIIX4 FCH
definitions to <asm/amd/fch.h>"). To prevent compile failures on non-x86
archs i2c-piix4 was set to only compile on x86 by commit 7e173eb82ae9717
("i2c: piix4: Make CONFIG_I2C_PIIX4 dependent on CONFIG_X86").
This was not a good decision because loongarch and mips both actually
support i2c-piix4 and set it enabled in the defconfig.
Move the header to a location accessible by all architectures.
Fixes: 624b0d5696a89 ("i2c: piix4, x86/platform: Move the SB800 PIIX4 FCH definitions to <asm/amd/fch.h>") Suggested-by: Hans de Goede <hansg@kernel.org> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Reviewed-by: Hans de Goede <hansg@kernel.org> Link: https://lore.kernel.org/r/20250610205817.3912944-1-superm1@kernel.org Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Hans de Goede [Mon, 9 Jun 2025 10:46:20 +0000 (12:46 +0200)]
platform/x86: x86-android-tablets: Add ovc-capacity-table info
Add ovc-capacity-table info to the generic battery nodes.
The values come from the ug3105 driver which currently hardcodes these
values. The ug3105 driver will be modified to stop hardcoding this and
instead get the values from device-properties.
Reviewed-by: Andy Shevchenko <andy@kernel.org> Signed-off-by: Hans de Goede <hansg@kernel.org> Link: https://lore.kernel.org/r/20250609104620.25896-2-hansg@kernel.org Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Hans de Goede [Mon, 9 Jun 2025 10:46:19 +0000 (12:46 +0200)]
platform/x86: x86-android-tablets: Add generic_lipo_4v2_battery info
Move the asus_tf103c_battery_node to shared-psy-info.c and rename it to
generic_lipo_4v2_battery_node.
This is a preparation patch for adding ovc-capacity-table info to
the battery nodes.
Reviewed-by: Andy Shevchenko <andy@kernel.org> Signed-off-by: Hans de Goede <hansg@kernel.org> Link: https://lore.kernel.org/r/20250609104620.25896-1-hansg@kernel.org Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Kurt Borja [Thu, 26 Jun 2025 01:17:37 +0000 (22:17 -0300)]
platform/x86: dell-wmi-sysman: Fix class device unregistration
Devices under the firmware_attributes_class do not have unique a dev_t.
Therefore, device_unregister() should be used instead of
device_destroy(), since the latter may match any device with a given
dev_t.
Fixes: e8a60aa7404b ("platform/x86: Introduce support for Systems Management Driver over WMI for Dell Systems") Signed-off-by: Kurt Borja <kuurtb@gmail.com> Link: https://lore.kernel.org/r/20250625-dest-fix-v1-3-3a0f342312bb@gmail.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Kurt Borja [Thu, 26 Jun 2025 01:17:36 +0000 (22:17 -0300)]
platform/x86: think-lmi: Fix class device unregistration
Devices under the firmware_attributes_class do not have unique a dev_t.
Therefore, device_unregister() should be used instead of
device_destroy(), since the latter may match any device with a given
dev_t.
Fixes: a40cd7ef22fb ("platform/x86: think-lmi: Add WMI interface support on Lenovo platforms") Signed-off-by: Kurt Borja <kuurtb@gmail.com> Link: https://lore.kernel.org/r/20250625-dest-fix-v1-2-3a0f342312bb@gmail.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Kurt Borja [Thu, 26 Jun 2025 01:17:35 +0000 (22:17 -0300)]
platform/x86: hp-bioscfg: Fix class device unregistration
Devices under the firmware_attributes_class do not have unique a dev_t.
Therefore, device_unregister() should be used instead of
device_destroy(), since the latter may match any device with a given
dev_t.
Fixes: a34fc329b189 ("platform/x86: hp-bioscfg: bioscfg") Signed-off-by: Kurt Borja <kuurtb@gmail.com> Link: https://lore.kernel.org/r/20250625-dest-fix-v1-1-3a0f342312bb@gmail.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Pratap Nirujogi [Wed, 18 Jun 2025 20:29:48 +0000 (16:29 -0400)]
platform/x86: Update swnode graph for amd isp4
Existing swnode graph format is specific to sensor device
and is causing conflicts when accessing standard property
variables outside the sensor driver.
To address this issue, enhanced swnode graph format with
dedicated nodes for i2c and isp devices, with sensor node
added as child to i2c node. This approach allows to have
standard property variables (ex: 'clock-frequency') with
values applicable for each of the devices (sensor, i2c and
isp).
ACPI device driver_data handle is also initialized with root
camera swnode to access the property variables in the graph
in isp and i2c drivers.
Kurt Borja [Mon, 30 Jun 2025 03:43:12 +0000 (00:43 -0300)]
platform/x86: dell-wmi-sysman: Fix WMI data block retrieval in sysfs callbacks
After retrieving WMI data blocks in sysfs callbacks, check for the
validity of them before dereferencing their content.
Reported-by: Jan Graczyk <jangraczyk@yahoo.ca> Closes: https://lore.kernel.org/r/CAHk-=wgMiSKXf7SvQrfEnxVtmT=QVQPjJdNjfm3aXS7wc=rzTw@mail.gmail.com/ Fixes: e8a60aa7404b ("platform/x86: Introduce support for Systems Management Driver over WMI for Dell Systems") Suggested-by: Linus Torvalds <torvalds@linux-foundation.org> Reviewed-by: Armin Wolf <W_Armin@gmx.de> Signed-off-by: Kurt Borja <kuurtb@gmail.com> Link: https://lore.kernel.org/r/20250630-sysman-fix-v2-1-d185674d0a30@gmail.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Armin Wolf [Thu, 19 Jun 2025 22:14:40 +0000 (00:14 +0200)]
platform/x86: wmi: Update documentation of WCxx/WExx ACPI methods
Inform potential firmware developers that WExx ACPI methods are
always called by the WMI driver core even if the associated WMI event
is not marked as being expensive.
Armin Wolf [Thu, 19 Jun 2025 22:14:39 +0000 (00:14 +0200)]
platform/x86: wmi: Fix WMI event enablement
It turns out that the Windows WMI-ACPI driver always enables/disables
WMI events regardless of whether they are marked as expensive or not.
This finding is further reinforced when reading the documentation of
the WMI_FUNCTION_CONTROL_CALLBACK callback used by Windows drivers
for enabling/disabling WMI devices:
The DpWmiFunctionControl routine enables or disables
notification of events, and enables or disables data
collection for data blocks that the driver registered
as expensive to collect.
Follow this behavior to fix the WMI event used for reporting hotkey
events on the Dell Latitude 5400 and likely many more devices.
Alok Tiwari [Sun, 22 Jun 2025 07:29:11 +0000 (00:29 -0700)]
platform/mellanox: Fix spelling and comment clarity in Mellanox drivers
This commit corrects several minor typographical errors in comments
and error messages across multiple Mellanox platform driver.
Fixed spelling of "thresholds", "region", "platform", "default",
and removed redundant spaces in comment strings and error logs.
These changes are cosmetic and do not affect runtime behavior.
Signed-off-by: Alok Tiwari <alok.a.tiwari@oracle.com> Reviewed-by: Vadim Pasternak <vadimp@nvidia.com> Reviewed-by: David Thompson <davthompson@nvidia.com> Link: https://lore.kernel.org/r/20250622072921.4111552-1-alok.a.tiwari@oracle.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Alok Tiwari [Thu, 19 Jun 2025 06:05:00 +0000 (23:05 -0700)]
platform/mellanox: mlxbf-pmc: Fix duplicate event ID for CACHE_DATA1
same ID (103) was assigned to both GDC_BANK0_G_RSE_PIPE_CACHE_DATA0
and GDC_BANK0_G_RSE_PIPE_CACHE_DATA1. This could lead to incorrect
event mapping.
Updated the ID to 104 to ensure uniqueness.
Fixes: 423c3361855c ("platform/mellanox: mlxbf-pmc: Add support for BlueField-3") Signed-off-by: Alok Tiwari <alok.a.tiwari@oracle.com> Reviewed-by: David Thompson <davthompson@nvidia.com> Link: https://lore.kernel.org/r/20250619060502.3594350-1-alok.a.tiwari@oracle.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
2025 Thinkpads F11 key launch the Intel Unison app on Windows,
which does some sort of smart sharing between laptop and phone.
Map this key event to KEY_LINK_PHONE as the closest thing we have.
This prevents an error message being displayed on key press.
Reported-by: Damjan Georgievski <gdamjan@gmail.com> Closes: https://sourceforge.net/p/ibm-acpi/mailman/message/59189556/ Signed-off-by: Mark Pearson <mpearson-lenovo@squebb.ca> Link: https://lore.kernel.org/r/20250620181119.2519546-1-mpearson-lenovo@squebb.ca
[ij: converted directory to pre-lenovo move as this is fixes material.] Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Paul Menzel [Sun, 22 Jun 2025 08:07:20 +0000 (10:07 +0200)]
platform/x86: dell-lis3lv02d: Add Latitude 5500
Add 0x29 as the accelerometer address for the Dell Latitude 5500 to
lis3lv02d_devices[].
The address was verified as below:
$ cd /sys/bus/pci/drivers/i801_smbus/0000:00:1f.4
$ ls -d i2c-?
i2c-2
$ sudo modprobe i2c-dev
$ sudo i2cdetect 2
WARNING! This program can confuse your I2C bus, cause data loss and worse!
I will probe file /dev/i2c-2.
I will probe address range 0x08-0x77.
Continue? [Y/n] Y
0 1 2 3 4 5 6 7 8 9 a b c d e f
00: 08 -- -- -- -- -- -- --
10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
20: -- -- -- -- -- -- -- -- -- 29 -- -- -- -- -- --
30: 30 -- -- -- -- 35 UU UU -- -- -- -- -- -- -- --
40: -- -- -- -- 44 -- -- -- -- -- -- -- -- -- -- --
50: UU -- 52 -- -- -- -- -- -- -- -- -- -- -- -- --
60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
70: -- -- -- -- -- -- -- --
$ echo lis3lv02d 0x29 | sudo tee /sys/bus/i2c/devices/i2c-2/new_device
lis3lv02d 0x29
$ sudo dmesg
[ 0.000000] Linux version 6.12.32-amd64 (debian-kernel@lists.debian.org) (x86_64-linux-gnu-gcc-14 (Debian 14.2.0-19) 14.2.0, GNU ld (GNU Binutils for Debian) 2.44) #1 SMP PREEMPT_DYNAMIC Debian 6.12.32-1 (2025-06-07)
[…]
[ 0.000000] DMI: Dell Inc. Latitude 5500/0M14W7, BIOS 1.38.0 03/06/2025
[…]
[ 609.063488] i2c_dev: i2c /dev entries driver
[ 639.135020] i2c i2c-2: new_device: Instantiated device lis3lv02d at 0x29
Signed-off-by: Paul Menzel <pmenzel@molgen.mpg.de> Reviewed-by: Hans de Goede <hansg@kernel.org> Link: https://lore.kernel.org/r/20250622080721.4661-1-pmenzel@molgen.mpg.de Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Fix warnings reported by sparse, related to incorrect type:
drivers/platform/mellanox/mlxbf-tmfifo.c:284:38: warning: incorrect type in assignment (different base types)
drivers/platform/mellanox/mlxbf-tmfifo.c:284:38: expected restricted __virtio32 [usertype] len
drivers/platform/mellanox/mlxbf-tmfifo.c:284:38: got unsigned long
Reported-by: kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202404040339.S7CUIgf3-lkp@intel.com/ Fixes: 78034cbece79 ("platform/mellanox: mlxbf-tmfifo: Drop the Rx packet if no more descriptors") Signed-off-by: David Thompson <davthompson@nvidia.com> Link: https://lore.kernel.org/r/20250613214608.2250130-1-davthompson@nvidia.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Ivan Hu [Mon, 16 Jun 2025 07:48:19 +0000 (15:48 +0800)]
platform/x86: portwell-ec: Move watchdog device under correct platform hierarchy
Without explicitly setting a parent for the watchdog device, the device is
registered with a NULL parent. This causes device_add() (called internally
by devm_watchdog_register_device()) to register the device under
/sys/devices/virtual, since no parent is provided. The result is:
DEVPATH=/devices/virtual/watchdog/watchdog0
To fix this, assign &pdev->dev as the parent of the watchdog device before
calling devm_watchdog_register_device(). This ensures the device is
associated with the Portwell EC platform device and placed correctly in
sysfs as:
This aligns the device hierarchy with expectations and avoids misplacement
under the virtual class.
Fixes: 835796753310 ("platform/x86: portwell-ec: Add GPIO and WDT driver for Portwell EC") Signed-off-by: Ivan Hu <ivan.hu@canonical.com> Link: https://lore.kernel.org/r/20250616074819.63547-1-ivan.hu@canonical.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Although the Alienware m16 R1 AMD model supports G-Mode, it actually has
a lower power ceiling than plain "performance" profile, which results in
lower performance.
Reported-by: Cihan Ozakca <cozakca@outlook.com> Cc: stable@vger.kernel.org # 6.15.x Signed-off-by: Kurt Borja <kuurtb@gmail.com> Link: https://lore.kernel.org/r/20250611-m16-rev-v1-1-72d13bad03c9@gmail.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
platform/x86/amd/pmc: Add PCSpecialist Lafite Pro V 14M to 8042 quirks list
Every other s2idle cycle fails to reach hardware sleep when keyboard
wakeup is enabled. This appears to be an EC bug, but the vendor
refuses to fix it.
It was confirmed that turning off i8042 wakeup avoids ths issue
(albeit keyboard wakeup is disabled). Take the lesser of two evils
and add it to the i8042 quirk list.
Reported-by: Raoul <ein4rth@gmail.com> Closes: https://bugzilla.kernel.org/show_bug.cgi?id=220116 Tested-by: Raoul <ein4rth@gmail.com> Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Link: https://lore.kernel.org/r/20250611203341.3733478-1-superm1@kernel.org Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
If the WMI core driver somehow unbinds from PNP0C14:00, the following
will happen upon rebinding:
1. The WMI driver core counts all registered WMI devices with a GUID
of 05901221-D566-11D1-B2F0-00A0C9062910 (count: 1).
2. The new WMI device will be named
"05901221-D566-11D1-B2F0-00A0C9062910-1" because another device
with the same GUID is already registered (on PNP0C14:01).
3. The new WMI device cannot be registered due to a name conflict.
Use a IDA when building the WMI device name to avoid such name
collisions by ensuring that a given WMI device ID is not reused.
Userspace applications using udev for WMI device detection are not
impacted by this change. Additionally userspace applications that do
fully support the existing naming scheme are also not impacted. Only
userspace applications using hardcoded sysfs paths will break.
Introduce a kconfig option for restoring the old naming scheme to
give developers time to fix any compatibility issues.
Armin Wolf [Tue, 10 Jun 2025 05:55:25 +0000 (07:55 +0200)]
Documentation: ABI: Update WMI device paths in ABI docs
The WMI driver core might append an ID to the WMI device name to avoid
name collisions in case multiple WMI devices with the same GUID are
present. Update all sysfs path referring to WMI devices to inform
users about this important detail.
platform/x86: fujitsu: clamp charge_control_end_threshold values to 50
Follow the sysfs ABI documentation that drivers should round written
values to the nearest supported value instead of returning an error.
Tested on a Fujitsu Lifebook U7720.
Signed-off-by: Jelle van der Waa <jvanderwaa@redhat.com> Reviewed-by: Hans de Goede <hansg@kernel.org> Acked-by: Jonathan Woithe <jwoithe@just42.net> Link: https://lore.kernel.org/r/20250609102115.36936-3-jvanderwaa@redhat.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
platform/x86: fujitsu: use unsigned int for kstrtounit
The charge control threshold value ranges from 0-100.
Signed-off-by: Jelle van der Waa <jvanderwaa@redhat.com> Reviewed-by: Hans de Goede <hansg@kernel.org> Acked-by: Jonathan Woithe <jwoithe@just42.net> Link: https://lore.kernel.org/r/20250609102115.36936-2-jvanderwaa@redhat.com
[ij: use reverse xmas-tree order] Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
platform/x86: silicom: remove unnecessary GPIO line direction check
As of commit 92ac7de3175e3 ("gpiolib: don't allow setting values on input
lines"), the GPIO core makes sure values cannot be set on input lines.
Remove the unnecessary check.
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org> Link: https://lore.kernel.org/r/20250612074835.13800-1-brgl@bgdev.pl Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Kees Cook [Thu, 29 May 2025 18:18:37 +0000 (11:18 -0700)]
platform/x86: thinkpad_acpi: Handle KCOV __init vs inline mismatches
When KCOV is enabled all functions get instrumented, unless the
__no_sanitize_coverage attribute is used. To prepare for
__no_sanitize_coverage being applied to __init functions[1], we have
to handle differences in how GCC's inline optimizations get resolved.
For thinkpad_acpi routines, this means forcing two functions to be
inline with __always_inline.
Some Ideapad models support a battery conservation mode which limits the
battery charge threshold for longer battery longevity. This is currently
exposed via a custom conservation_mode attribute in sysfs.
The newly introduced charge_types sysfs attribute is a standardized
replacement for laptops with a fixed end charge threshold. Setting it to
`Long Life` would enable battery conservation mode. The standardized
user space API would allow applications such as UPower to detect laptops
which support this battery longevity mode and set it.
Tested on an Lenovo ideapad U330p.
Signed-off-by: Jelle van der Waa <jvanderwaa@redhat.com> Suggested-By: Hans de Goede <hdegoede@redhat.com> Reviewed-by: Thomas Weißschuh <linux@weissschuh.net> Reviewed-by: Armin Wolf <W_Armin@gmx.de> Link: https://lore.kernel.org/r/20250514201054.381320-1-jvanderwaa@redhat.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Mark Pearson [Tue, 10 Jun 2025 19:28:25 +0000 (15:28 -0400)]
platform/x86: Move Lenovo files into lenovo subdir
Create lenovo subdirectory for holding Lenovo specific drivers.
Signed-off-by: Mark Pearson <mpearson-lenovo@squebb.ca> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Link: https://lore.kernel.org/r/20250610192830.1731454-2-mpearson-lenovo@squebb.ca
[ij: put depends on DMI back, fix trailing empty lines.] Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Mark Pearson [Tue, 10 Jun 2025 19:28:24 +0000 (15:28 -0400)]
platform/x86: thinklmi: improved DMI handling
Fix issues reported by kernel test robot.
- Require DMI for think-lmi.
- Check return from getting serial string
Reported-by: kernel test robot <lkp@intel.com> Closes: https://lore.kernel.org/oe-kbuild-all/202506062319.F0IpDxF6-lkp@intel.com/ Signed-off-by: Mark Pearson <mpearson-lenovo@squebb.ca> Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Link: https://lore.kernel.org/r/20250610192830.1731454-1-mpearson-lenovo@squebb.ca Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
were both added by the commit 378f956e3f93 ("platform/x86: Add Intel
Telemetry Core Driver") in 2016 but have remained unused.
They're each a tiny wrapper that is the only caller through a similarly
named function pointer, and for each function pointer there's a 'def'
empty implementation and a plt implementation.
were all added by the commit 378f956e3f93 ("platform/x86: Add Intel
Telemetry Core Driver") in 2016 but have remained unused.
They're each a tiny wrapper that is the only caller through a similarly
named function pointer, and for each function pointer there's a 'def'
empty implementation and a plt implementation.
Stuart Hayes [Mon, 9 Jun 2025 18:46:58 +0000 (13:46 -0500)]
platform/x86: dell_rbu: Stop overwriting data buffer
The dell_rbu driver will use memset() to clear the data held by each
packet when it is no longer needed (when the driver is unloaded, the
packet size is changed, etc).
The amount of memory that is cleared (before this patch) is the normal
packet size. However, the last packet in the list may be smaller.
Fix this to only clear the memory actually used by each packet, to prevent
it from writing past the end of data buffer.
Because the packet data buffers are allocated with __get_free_pages() (in
page-sized increments), this bug could only result in a buffer being
overwritten when a packet size larger than one page is used. The only user
of the dell_rbu module should be the Dell BIOS update program, which uses
a packet size of 4096, so no issues should be seen without the patch, it
just blocks the possiblity.
Fixes: 6c54c28e69f2 ("[PATCH] dell_rbu: new Dell BIOS update driver") Signed-off-by: Stuart Hayes <stuart.w.hayes@gmail.com> Link: https://lore.kernel.org/r/20250609184659.7210-5-stuart.w.hayes@gmail.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Stuart Hayes [Mon, 9 Jun 2025 18:46:56 +0000 (13:46 -0500)]
platform/x86: dell_rbu: Fix list usage
Pass the correct list head to list_for_each_entry*() when looping through
the packet list.
Without this patch, reading the packet data via sysfs will show the data
incorrectly (because it starts at the wrong packet), and clearing the
packet list will result in a NULL pointer dereference.
Fixes: d19f359fbdc6 ("platform/x86: dell_rbu: don't open code list_for_each_entry*()") Signed-off-by: Stuart Hayes <stuart.w.hayes@gmail.com> Link: https://lore.kernel.org/r/20250609184659.7210-3-stuart.w.hayes@gmail.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
platform/x86/amd: pmf: Simplify error flow in amd_pmf_init_smart_pc()
commit 5b1122fc4995f ("platform/x86/amd/pmf: fix cleanup in
amd_pmf_init_smart_pc()") adjusted the error handling flow to use a ladder
but this isn't actually needed because work is only scheduled in
amd_pmf_start_policy_engine() and with device managed cleanups pointers
for allocations don't need to be freed.
Adjust the error flow to a single call to amd_pmf_deinit_smart_pc() for
the cases that need to clean up.
platform/x86/amd: pmf: Use device managed allocations
If setting up smart PC fails for any reason then this can lead to
a double free when unloading amd-pmf. This is because dev->buf was
freed but never set to NULL and is again freed in amd_pmf_remove().
To avoid subtle allocation bugs in failures leading to a double free
change all allocations into device managed allocations.
Suma Hegde [Tue, 3 Jun 2025 05:58:07 +0000 (05:58 +0000)]
platform/x86/amd/hsmp: Use IS_ENABLED() instead of IS_REACHABLE()
IS_REACHABLE() was required when CONFIG_HWMON was set to m and HSMP to y.
However, commit 69157b00b526 ("platform/x86/amd/hsmp: fix building with
CONFIG_HWMON=m") added a HWMON dependency for HSMP in Kconfig. With
this change, using IS_ENABLED() is sufficient.
Jake Hillion [Thu, 5 Jun 2025 18:09:27 +0000 (19:09 +0100)]
x86/platform/amd: replace down_timeout() with down_interruptible()
Currently hsmp_send_message() uses down_timeout() with a 100ms timeout
to take the semaphore. However __hsmp_send_message(), the content of the
critical section, has a sleep in it. On systems with significantly
delayed scheduling behaviour this may take over 100ms.
Convert this method to down_interruptible(). Leave the error handling
the same as the documentation currently is not specific about what error
is returned.
Previous behaviour: a caller who competes with another caller stuck in
the critical section due to scheduler delays would receive -ETIME.
New behaviour: a caller who competes with another caller stuck in the
critical section due to scheduler delays will complete successfully.
Jake Hillion [Thu, 5 Jun 2025 18:09:26 +0000 (19:09 +0100)]
x86/platform/amd: move final timeout check to after final sleep
__hsmp_send_message sleeps between result read attempts and has a
timeout of 100ms. Under extreme load it's possible for these sleeps to
take a long time, exceeding the 100ms. In this case the current code
does not check the register and fails with ETIMEDOUT.
Refactor the loop to ensure there is at least one read of the register
after a sleep of any duration. This removes instances of ETIMEDOUT with
a single caller, even with a misbehaving scheduler. Tested on AMD
Bergamo machines.
platform/x86/amd: pmc: Clear metrics table at start of cycle
The area of memory that contains the metrics table may contain garbage
when the cycle starts. This normally doesn't matter because the cycle
itself will populate it with valid data, however commit 9f5595d5f03fd
("platform/x86/amd: pmc: Require at least 2.5 seconds between HW sleep
cycles") started to use it during the check() phase. Depending upon
what garbage is in the table it's possible that the system will wait
2.5 seconds for even the first cycle, which will be visible to a user.
To prevent this from happening explicitly clear the table when logging
is started.
Fixes: 9f5595d5f03fd ("platform/x86/amd: pmc: Require at least 2.5 seconds between HW sleep cycles") Signed-off-by: Mario Limonciello <mario.limonciello@amd.com> Link: https://lore.kernel.org/r/20250603132412.3555302-1-superm1@kernel.org Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Dan Carpenter [Fri, 6 Jun 2025 09:04:32 +0000 (12:04 +0300)]
platform/x86/intel: power-domains: Fix error code in tpmi_init()
Return -ENOMEM instead of success if kcalloc() fails.
Fixes: e37be5d85c60 ("platform/x86/intel: power-domains: Add interface to get Linux die ID") Signed-off-by: Dan Carpenter <dan.carpenter@linaro.org> Link: https://lore.kernel.org/r/aEKvIGCt6d8Gcx4S@stanley.mountain Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
platform/x86/intel-uncore-freq: Fail module load when plat_info is NULL
Address a Smatch static checker warning regarding an unchecked
dereference in the function call:
set_cdie_id(i, cluster_info, plat_info)
when plat_info is NULL.
Instead of addressing this one case, in general if plat_info is NULL
then it can cause other issues. For example in a two package system it
will give warning for duplicate sysfs entry as package ID will be always
zero for both packages when creating string for attribute group name.
plat_info is derived from TPMI ID TPMI_BUS_INFO, which is integral to
the core TPMI design. Therefore, it should not be NULL on a production
platform. Consequently, the module should fail to load if plat_info is
NULL.
Reported-by: Dan Carpenter <dan.carpenter@linaro.org> Closes: https://lore.kernel.org/platform-driver-x86/aEKvGCLd1qmX04Tc@stanley.mountain/T/#u Fixes: 8a54e2253e4c ("platform/x86/intel-uncore-freq: Uncore frequency control via TPMI") Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com> Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20250606205300.2384494-1-srinivas.pandruvada@linux.intel.com Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Rong Zhang [Sun, 25 May 2025 20:18:07 +0000 (04:18 +0800)]
platform/x86: ideapad-laptop: use usleep_range() for EC polling
It was reported that ideapad-laptop sometimes causes some recent (since
2024) Lenovo ThinkBook models shut down when:
- suspending/resuming
- closing/opening the lid
- (dis)connecting a charger
- reading/writing some sysfs properties, e.g., fan_mode, touchpad
- pressing down some Fn keys, e.g., Brightness Up/Down (Fn+F5/F6)
- (seldom) loading the kmod
The issue has existed since the launch day of such models, and there
have been some out-of-tree workarounds (see Link:) for the issue. One
disables some functionalities, while another one simply shortens
IDEAPAD_EC_TIMEOUT. The disabled functionalities have read_ec_data() in
their call chains, which calls schedule() between each poll.
It turns out that these models suffer from the indeterminacy of
schedule() because of their low tolerance for being polled too
frequently. Sometimes schedule() returns too soon due to the lack of
ready tasks, causing the margin between two polls to be too short.
In this case, the command is somehow aborted, and too many subsequent
polls (they poll for "nothing!") may eventually break the state machine
in the EC, resulting in a hard shutdown. This explains why shortening
IDEAPAD_EC_TIMEOUT works around the issue - it reduces the total number
of polls sent to the EC.
Even when it doesn't lead to a shutdown, frequent polls may also disturb
the ongoing operation and notably delay (+ 10-20ms) the availability of
EC response. This phenomenon is unlikely to be exclusive to the models
mentioned above, so dropping the schedule() manner should also slightly
improve the responsiveness of various models.
Fix these issues by migrating to usleep_range(150, 300). The interval is
chosen to add some margin to the minimal 50us and considering EC
responses are usually available after 150-2500us based on my test. It
should be enough to fix these issues on all models subject to the EC bug
without introducing latency on other models.
Tested on ThinkBook 14 G7+ ASP and solved both issues. No regression was
introduced in the test on a model without the EC bug (ThinkBook X IMH,
thanks Eric).
Linus Torvalds [Sun, 8 Jun 2025 18:33:00 +0000 (11:33 -0700)]
Merge tag 'timers-cleanups-2025-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull timer cleanup from Thomas Gleixner:
"The delayed from_timer() API cleanup:
The renaming to the timer_*() namespace was delayed due massive
conflicts against Linux-next. Now that everything is upstream finish
the conversion"
* tag 'timers-cleanups-2025-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
treewide, timers: Rename from_timer() to timer_container_of()
Linus Torvalds [Sun, 8 Jun 2025 18:27:20 +0000 (11:27 -0700)]
Merge tag 'x86-urgent-2025-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 fixes from Thomas Gleixner:
"A small set of x86 fixes:
- Cure IO bitmap inconsistencies
A failed fork cleans up all resources of the newly created thread
via exit_thread(). exit_thread() invokes io_bitmap_exit() which
does the IO bitmap cleanups, which unfortunately assume that the
cleanup is related to the current task, which is obviously bogus.
Make it work correctly
- A lockdep fix in the resctrl code removed the clearing of the
command buffer in two places, which keeps stale error messages
around. Bring them back.
- Remove unused trace events"
* tag 'x86-urgent-2025-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
fs/resctrl: Restore the rdt_last_cmd_clear() calls after acquiring rdtgroup_mutex
x86/iopl: Cure TIF_IO_BITMAP inconsistencies
x86/fpu: Remove unused trace events
Zhang Rui [Fri, 30 May 2025 00:09:28 +0000 (08:09 +0800)]
tools/power turbostat: Avoid probing the same perf counters
For the RAPL package energy status counter, Intel and AMD share the same
perf_subsys and perf_name, but with different MSR addresses.
Both rapl_counter_arch_infos[0] and rapl_counter_arch_infos[1] are
introduced to describe this counter for different Vendors.
As a result, the perf counter is probed twice, and causes a failure in
in get_rapl_counters() because expected_read_size and actual_read_size
don't match.
Fix the problem by skipping the already probed counter.
Note, this is not a perfect fix. For example, if different
vendors/platforms use the same MSR value for different purpose, the code
can be fooled when it probes a rapl_counter_arch_infos[] entry that does
not belong to the running Vendor/Platform.
In a long run, better to put rapl_counter_arch_infos[] into the
platform_features so that this becomes Vendor/Platform specific.
Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
Zhang Rui [Sat, 17 May 2025 09:44:50 +0000 (17:44 +0800)]
tools/power turbostat: Allow probing RAPL with platform_features->rapl_msrs cleared
platform_features->rapl_msrs describes the RAPL MSRs supported. While
RAPL Perf counters can be exposed from different kernel backend drivers,
e.g. RAPL MSR I/F driver, or RAPL TPMI I/F driver.
Thus, turbostat should first blindly probe all the available RAPL Perf
counters, and falls back to the RAPL MSR counters if they are listed in
platform_features->rapl_msrs.
With this, platforms that don't have RAPL MSRs can clear the
platform_features->rapl_msrs bits and use RAPL Perf counters only.
Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
probe_rapl_msr() is reused for probing RAPL MSR counters, cstate MSR
counters and MPERF/APERF/SMI MSR counters, thus its name is misleading.
Similar to add_perf_counter(), introduce add_msr_counter() to probe a
counter via MSR. Introduce wrapper function add_rapl_msr_counter() at
the same time to add extra check for Zero return value for specified
RAPL counters.
No functional change intended.
Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
As the only caller of add_cstate_perf_counter_(),
add_cstate_perf_counter() just gives extra debug output on top. There is
no need to keep both functions.
Remove add_cstate_perf_counter_() and move all the logic to
add_cstate_perf_counter().
No functional change.
Signed-off-by: Zhang Rui <rui.zhang@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
commit 05a2f07db888 ("tools/power turbostat: read RAPL counters via
perf") that adds support to read RAPL counters via perf defines the
notion of a RAPL domain_id which is set to physical_core_id on
platforms which support per_core_rapl counters (Eg: AMD processors
Family 17h onwards) and is set to the physical_package_id on all the
other platforms.
However, the physical_core_id is only unique within a package and on
platforms with multiple packages more than one core can have the same
physical_core_id and thus the same domain_id. (For eg, the first cores
of each package have the physical_core_id = 0). This results in all
these cores with the same physical_core_id using the same entry in the
rapl_counter_info_perdomain[]. Since rapl_perf_init() skips the
perf-initialization for cores whose domain_ids have already been
visited, cores that have the same physical_core_id always read the
perf file corresponding to the physical_core_id of the first package
and thus the package-energy is incorrectly reported to be the same
value for different packages.
Note: This issue only arises when RAPL counters are read via perf and
not when they are read via MSRs since in the latter case the MSRs are
read separately on each core.
Fix this issue by associating each CPU with rapl_core_id which is
unique across all the packages in the system.
Fixes: 05a2f07db888 ("tools/power turbostat: read RAPL counters via perf") Signed-off-by: Gautham R. Shenoy <gautham.shenoy@amd.com> Signed-off-by: Len Brown <len.brown@intel.com>
tools/power turbostat: Add Android support for MSR device handling
It uses /dev/msrN device paths on Android instead of /dev/cpu/N/msr,
updates error messages and permission checks to reflect the Android
device path, and wraps platform-specific code with #if defined(ANDROID)
to ensure correct behavior on both Android and non-Android systems.
These changes improve compatibility and usability of turbostat on
Android devices.
Signed-off-by: Kaushlendra Kumar <kaushlendra.kumar@intel.com> Signed-off-by: Len Brown <len.brown@intel.com>
Linus Torvalds [Sun, 8 Jun 2025 18:07:33 +0000 (11:07 -0700)]
Merge tag 'perf-urgent-2025-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 perf fix from Thomas Gleixner:
"A single fix for the x86 performance counters on Intel CPUs:
The MSR offset calculations for fixed performance counters are stored
at the wrong index in the configuration array causing the general
purpose counter MSR offset to be overwritten, so both the general
purpose and the fixed counters offsets are incorrect.
Correct the array index calculation to fix that"
* tag 'perf-urgent-2025-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
perf/x86/intel: Fix incorrect MSR index calculations in intel_pmu_config_acr()
Linus Torvalds [Sun, 8 Jun 2025 18:02:53 +0000 (11:02 -0700)]
Merge tag 'irq-urgent-2025-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull irq fix from Thomas Gleixner:
"A single fix for the PCI/MSI code:
The conversion to per device MSI domains created a MSI domain with
size 1 instead of sizing it to the maximum possible number of MSI
interrupts for the device. This "worked" as the subsequent allocations
resized the domain, but the recent change to move the prepare() call
into the domain creation path broke this works by chance mechanism.
Size the domain properly at creation time"
* tag 'irq-urgent-2025-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
PCI/MSI: Size device MSI domain with the maximum number of vectors
projects / thirdparty / linux.git / log
