Since v2025.11.22:
Initial SoC Slider support
SoC Slider is an SoC-wide power/performance policy setting.
On SoC Slider systems, EPP plays a diminished role.
Whitespace cleanup via: indent -npro -kr -i8 -ts8 -sob -l160 -ss -ncs -cp1
No functional changes
Signed-off-by: Len Brown <len.brown@intel.com>
x86_energy_perf_policy accesses the SoC Slider via standard
user/kernel APIs to the processor_thermal_soc_slider driver.
Machines that support SoC Slider largely use it instead of EPP,
which may continue to exist in a diminished role, or vanish entirely.
Signed-off-by: Len Brown <len.brown@intel.com>
When processor_thermal_soc_slider is loaded, its slider
and offset modparams are visible. Check that the driver
actually registered the profile named "SoC Slider" before
reading or writing these modparams.
n.b. This utility allows writing the Slider and Offset modparams
even if the driver policy is not "balanced". Currently the
processor_thermal_soc_slider consults those modparams
only in "balanced" mode.
Signed-off-by: Len Brown <len.brown@intel.com>
Since v2026.02.14
Display HT siblings in cpu# order.
Add Module-ID column.
Print Core-ID and APIC-ID in hex.
Fix misc bugs.
Signed-off-by: Len Brown <len.brown@intel.com>
On large systems with HT sibling cpu#'s more than 32 apart,
HT siblings were processed and displayed in reverse order.
This was due to how set_thread_siblings() parsed the
sibling-bit-mask.
Update set_thread_siblings to instead parse the sibling-list,
like other cpu lists, and to thus order HT siblings
by ascending CPU number, no matter the size of the system.
Signed-off-by: Len Brown <len.brown@intel.com>
Get the "module_id" from the Linux topology "cluster_id".
If the there is more than one id, show it by default.
Module joins Die etc. in the "topology" group.
Display in hex, as it is usually based mask of the APIC-id
Signed-off-by: Len Brown <len.brown@intel.com>
The core_id is based on a mask of the apic_id.
Print them both in hex, rather than decimal,
to make this relationship visibly clear.
Signed-off-by: Len Brown <len.brown@intel.com>
Make printer helper functions more readable by factoring
out a local 'sep' variable.
Remove the redundant parentheses around sprintf() calls.
Remove an unnecessary cast to "unsigned int" by using the '%08llx' instead
of '%08x'.
No functional changes.
[lenb: fix typos, simplify]
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
When the "--cpu-set" option limits turbostat to run on
a higher numbered HT sibling, it exits upon dividing by zero.
This is because the HT support handles higher numbered siblings
at the same time as lower numbered siblings. But when that lower
number sibling is dis-allowed, the higher numbered sibling is
never processed. The result is a time delta of 0, which results
in a divide by 0 for any of the "per-second" metrics.
Enhance the HT enumeration code to record all siblings (up to SMT4).
Consult this complete HT sibling list to determine when
to process an HT sibling, and when to skip it.
Fixes: a2b4d0f8bf ("tools/power turbostat: Favor cpu# over core#")
Signed-off-by: Len Brown <len.brown@intel.com>
"turbostat --cpu-set 0" appears to hang if cpu0 has an HT sibling.
This is because the initialization code recognizes that it does not
have to open perf files for the HT sibling, but the HT support
in the collection code sees the HT sibling and tries to read
from an uninitialized file descriptor, 0 (standard input).
Access HT siblings only when they are in the allowed set.
Fixes: a2b4d0f8bf ("tools/power turbostat: Favor cpu# over core#")
Signed-off-by: Len Brown <len.brown@intel.com>
Reported-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>
The '-P' short option (shorthand for --no-perf) is not present in the
optstring of the second call to getopt_long_only(). This results in
the "unrecognized option" error when the tool reaches the main parsing
loop.
Add 'P' to the second getopt_long_only() call to ensure it is
consistently recognized.
Fixes: a0e86c90b8 ("tools/power turbostat: Add --no-perf option")
Signed-off-by: David Arcari <darcari@redhat.com>
Signed-off-by: Len Brown <len.brown@intel.com>
turbostat.c:8688: rapl_perf_init: Assertion `next_domain < num_domains' failed.
The initial fix for this regression was incomplete, as it did not
handle multi-package systems with sparse core ids.
Fixes: ef0e60083f ("tools/power turbostat: Fix AMD RAPL regression")
Signed-off-by: Len Brown <len.brown@intel.com>
Highlights:
- asus-wmi:
- Retain battery charge threshold during boot which avoids unsolicited
change to 100%. Return -ENODATA when the limit is not yet known
- Improve screenpad power/brightness handling consistency
- Fix screenpad brightness range
- barco-p50-gpio: Normalize gpio_get return values
- bitland-mifs-wmi: Add driver for Bitland laptops (supports platform
profile,hwmon, kbd backlight, gpu mode, hotkeys, and
fan boost)
- dell_rbu: Fix using uninitialized value in sysfs write function
- dell-wmi-sysman: Respect destination length when constructing enum
strings
- hp-wmi:
- Propagate fan setting apply failures and log an error
- Fix sysfs write vs work handler cancel_delayed_work_sync() deadlock
- Correct keepalive schedule_delayed_work() to mod_delayed_work()
- Fix u8 underflows in GPU delta calculation
- Use mutex to protect fan pwm/mode
- Ignore kbd backlight and FnLock key events that are handled by FW
- Fix fan table parsing (use correct field)
- Add support for Omen 14-fb0xxx, 16-n0xxx, 16-wf1xxx, and Omen MAX
16-ak0xxxx
- input: trackpoint & thinkpad_acpi: Enable doubletap by default and
add sysfs enable/disable
- int3472: Add support for GPIO type 0x02 (IR flood LED)
- intel-speed-select: (updated to v1.26)
- Avoid using current base frequency as maximum
- Fix CPU extended family ID decoding
- Fix exit code
- Improve error reporting
- intel/vsec: Refactor to support ACPI-enumerated PMT endpoints.
- pcengines-apuv2: Attach software node to the gpiochip
- uniwill:
- Refactor hwmon to smaller parts to accomodate HW diversity
- Support USB-C power/performance priority switch through sysfs
- Add another XMG Fusion 15 (L19) DMI vendor
- Enable fine-grained features to device lineup mapping
- wmi: Perform output size check within WMI core to allow simpler WMI
drivers
- acpi_driver -> platform driver conversions (a large number of changes
from Rafael J. Wysocki)
- Miscellaneous cleanups / refactoring / improvements
The following is an automated shortlog grouped by driver:
acer-wireless:
- Convert ACPI driver to a platform one
- Register ACPI notify handler directly
asus-laptop:
- Convert ACPI driver to a platform one
- Register ACPI notify handler directly
asus-wireless:
- Convert ACPI driver to a platform one
- Register ACPI notify handler directly
asus-wmi:
- adjust screenpad power/brightness handling
- do not enforce a battery charge threshold
- fix screenpad brightness range
barco-p50-gpio:
- convert to guard() notation
- normalize return value of gpio_get
bitland-mifs-wmi:
- Add new Bitland MIFS WMI driver
dell/dell-rbtn:
- Convert ACPI driver to a platform one
- Register ACPI notify handler directly
dell_rbu:
- avoid uninit value usage in packet_size_write()
dell-wmi-sysman:
- bound enumeration string aggregation
- Clean up security buffer helpers
- Fix typo in function comment
- Use standard kobj_sysfs_ops
- Use sysfs_emit{_at} in show functions
Documentation: laptops:
- Update documentation for uniwill laptops
Documentation:
- thinkpad-acpi - Document doubletap_enable attribute
eeepc-laptop:
- Convert ACPI driver to a platform one
- Register ACPI notify handler directly
fujitsu:
- Convert backlight driver to a platform one
- Convert laptop driver to a platform one
- Register ACPI notify handlers directly
- Reorder code to avoid forward declarations
fujitsu-tablet:
- Convert ACPI driver to a platform one
hp-wmi:
- add locking for concurrent hwmon access
- add Omen 14-fb0xxx (board 8C58) support
- Add support for Omen 16-n0xxx (8A44)
- Add support for Omen 16-wf1xxx (8C77)
- Add support for OMEN MAX 16-ak0xxx (8D87)
- avoid cancel_delayed_work_sync from work handler
- fix fan table parsing
- fix ignored return values in fan settings
- fix u8 underflow in gpu_delta calculation
- Ignore backlight and FnLock events
- use mod_delayed_work to reset keep-alive timer
input:
- trackpoint - Enable doubletap by default on capable devices
int3472:
- Add support for GPIO type 0x02 (IR flood LED)
- Parameterize LED con_id in registration
- Rename pled to led in LED registration code
- Use local variable for LED struct access
intel/rst:
- Convert ACPI driver to a platform one
intel/smartconnect:
- Convert ACPI driver to a platform one
intel/tpmi:
- Use 32 bit aligned address for debugfs mem write
intel/vsec:
- Decouple add/link helpers from PCI
- Make driver_data info const
- Plumb ACPI PMT discovery tables through vsec
- Refactor base_addr handling
- Return real error codes from registration path
- Switch exported helpers from pci_dev to device
lg-laptop:
- Convert ACPI driver to a platform one
- Drop debug-only ACPI notify handler
nvsw-sn2201:
- Drop unused include
panasonic-laptop:
- Convert ACPI driver to a platform one
- Fix OPTD notifier registration and cleanup
- Make pcc_register_optd_notifier() void
- Register ACPI notify handler directly
- Remove redundant checks from 3 functions
pcengines-apuv2:
- attach software node to the gpiochip
sony-laptop:
- Convert NC driver to a platform one
- Convert PIC driver to a platform one
- Register ACPI notify handler directly
surface: hotplug:
- Correct inclusion for GPIO APIs
surface: surfacepro3_button:
- Convert to a platform driver
- Drop wakeup source on remove
- Register ACPI notify handler
system76:
- Convert ACPI driver to a platform one
- Drop redundant devm_led_classdev_unregister()
- Register ACPI notify handler directly
thinkpad_acpi:
- Add sysfs control for TrackPoint double-tap
- Drop ACPI driver registration
- remove obsolete TODO comment
- use seq_puts() instead of seq_printf()
tools/power/x86/intel-speed-select:
- Avoid current base freq as maximum
- Fix cpu extended family ID decoding
- Fix output when running on unsupported CLX platforms
- Fix some program return value
- Print Version info when Incompatible API version is detected
- v1.26 release
topstar-laptop:
- Convert ACPI driver to a platform one
- Register ACPI notify handler directly
toshiba_acpi:
- Convert ACPI driver to a platform one
- Register ACPI notify handler directly
- Reorder code to avoid forward declaration
toshiba_bluetooth:
- Convert ACPI driver to a platform one
- Register ACPI notify handler directly
toshiba_haps:
- Convert ACPI driver to a platform one
- Register ACPI notify handler directly
uniwill-laptop:
- Apply features across all TUXEDO devices
- Fix signedness bug
- Fix XMG Fusion 15 (L19) entries
- Implement USB-C power priority setting
- Rework hwmon feature defines
wireless-hotkey:
- Convert ACPI driver to a platform one
- Register ACPI notify handler directly
wmi:
- Add wmidev_invoke_procedure()
- Convert drivers to use wmidev_invoke_procedure()
- Extend wmidev_invoke_method() to reject undersized data
- Extend wmidev_query_block() to reject undersized data
- Prepare to reject undersized unmarshalling results
- Replace .no_notify_data with .min_event_size
x86:
- remove unnecessary module_init/exit() functions
Merges:
- Merge branch 'intel-sst' of https://github.com/spandruvada/linux-kernel into review-ilpo-next
- Merge tag 'fixes' into 'for-next'
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQSCSUwRdwTNL2MhaBlZrE9hU+XOMQUCaeYxSwAKCRBZrE9hU+XO
MRzTAQCPUKVOpSY/cjtPXVBn0uJbAo1MSvytv00Kv7dcatvrrwEA9lqmwOfl0kzr
CowEVVCD3om++W9vOsL65hachbAl1QA=
=1qkw
-----END PGP SIGNATURE-----
Merge tag 'platform-drivers-x86-v7.1-1' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86
Pull x86 platform driver updates from Ilpo Järvinen:
"asus-wmi:
- Retain battery charge threshold during boot which avoids
unsolicited change to 100%. Return -ENODATA when the limit
is not yet known
- Improve screenpad power/brightness handling consistency
- Fix screenpad brightness range
barco-p50-gpio:
- Normalize gpio_get return values
bitland-mifs-wmi:
- Add driver for Bitland laptops (supports platform profile,
hwmon, kbd backlight, gpu mode, hotkeys, and fan boost)
dell_rbu:
- Fix using uninitialized value in sysfs write function
dell-wmi-sysman:
- Respect destination length when constructing enum strings
hp-wmi:
- Propagate fan setting apply failures and log an error
- Fix sysfs write vs work handler cancel_delayed_work_sync() deadlock
- Correct keepalive schedule_delayed_work() to mod_delayed_work()
- Fix u8 underflows in GPU delta calculation
- Use mutex to protect fan pwm/mode
- Ignore kbd backlight and FnLock key events that are handled by FW
- Fix fan table parsing (use correct field)
- Add support for Omen 14-fb0xxx, 16-n0xxx, 16-wf1xxx, and
Omen MAX 16-ak0xxxx
input: trackpoint & thinkpad_acpi:
- Enable doubletap by default and add sysfs enable/disable
int3472:
- Add support for GPIO type 0x02 (IR flood LED)
intel-speed-select: (updated to v1.26)
- Avoid using current base frequency as maximum
- Fix CPU extended family ID decoding
- Fix exit code
- Improve error reporting
intel/vsec:
- Refactor to support ACPI-enumerated PMT endpoints.
pcengines-apuv2:
- Attach software node to the gpiochip
uniwill:
- Refactor hwmon to smaller parts to accomodate HW diversity
- Support USB-C power/performance priority switch through sysfs
- Add another XMG Fusion 15 (L19) DMI vendor
- Enable fine-grained features to device lineup mapping
wmi:
- Perform output size check within WMI core to allow simpler WMI
drivers
misc:
- acpi_driver -> platform driver conversions (a large number of
changes from Rafael J. Wysocki)
- cleanups / refactoring / improvements"
* tag 'platform-drivers-x86-v7.1-1' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86: (106 commits)
platform/x86: hp-wmi: Add support for Omen 16-wf1xxx (8C77)
platform/x86: hp-wmi: Add support for Omen 16-n0xxx (8A44)
platform/x86: hp-wmi: Add support for OMEN MAX 16-ak0xxx (8D87)
platform/x86: hp-wmi: fix fan table parsing
platform/x86: hp-wmi: add Omen 14-fb0xxx (board 8C58) support
platform/wmi: Replace .no_notify_data with .min_event_size
platform/wmi: Extend wmidev_query_block() to reject undersized data
platform/wmi: Extend wmidev_invoke_method() to reject undersized data
platform/wmi: Prepare to reject undersized unmarshalling results
platform/wmi: Convert drivers to use wmidev_invoke_procedure()
platform/wmi: Add wmidev_invoke_procedure()
platform/x86: int3472: Add support for GPIO type 0x02 (IR flood LED)
platform/x86: int3472: Parameterize LED con_id in registration
platform/x86: int3472: Rename pled to led in LED registration code
platform/x86: int3472: Use local variable for LED struct access
platform/x86: thinkpad_acpi: remove obsolete TODO comment
platform/x86: dell-wmi-sysman: bound enumeration string aggregation
platform/x86: hp-wmi: Ignore backlight and FnLock events
platform/x86: uniwill-laptop: Fix signedness bug
platform/x86: dell_rbu: avoid uninit value usage in packet_size_write()
...
Add support for reading and writing SOC slider parameters and
platform profile via sysfs in x86_energy_perf_policy.
New command-line options:
--soc-slider-balance <value>
--soc-slider-offset <value>
--platform-profile <name>
These options allow control of the processor thermal SOC
slider balance and offset through the
processor_thermal_soc_slider module, as well as the
platform profile class interface.
When no update flags are set, the tool now also prints
the current SOC slider and platform profile values
alongside existing MSR output.
Signed-off-by: Kaushlendra Kumar <kaushlendra.kumar@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Currently, if perf_l2_init() fails turbostat exits after issuing the
following error (which was encountered on AlderLake):
turbostat: perf_l2_init(cpu0, 0x0, 0xff24) REFS: Invalid argument
This occurs because perf_l2_init() calls err(). However, the code has been
written in such a manner that it is able to perform cleanup and continue.
Therefore, this issue can be addressed by changing the appropriate calls
to err() to warnx().
Additionally, correct the PMU type arguments passed to the warning strings
in the ecore and lcore blocks so the logs accurately reflect the failing
counter type.
Signed-off-by: David Arcari <darcari@redhat.com>
Signed-off-by: Len Brown <len.brown@intel.com>
This version includes the following changes:
- Setting current base frequency as maximum for SST-BF with
kernel QOS changes
- Harmonize extended family decoded with the rest of the kernel
- Minor changes for error codes and messages
Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
When running intel-speed-select on unsupported CLX platforms, it prints
intel-speed-select: Invalid CPU model (85)
: Success
Because this is not a system error and errno is not set.
Replace err() with exit().
Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
When running an old version intel-speed-select tool on newer platforms,
even with "intel-speed-select -v", the tool only complains about
"Incompatible API version", without giving the current version info.
Print Version info whenever Incompatible API version is detected.
Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
When running the "intel-speed-select -h" command, it returns
1. 0 when using a version that is API incompatible.
2. 1 when using a version that is API compatible.
And this is confusing.
Fix the program to return 0 for "-h" parameter, and return 1 whenever
"Incompatible API versions" is detected.
Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
When decode and use CPU extended family ID in intel-speed-select, there
are several potential issues,
1. Mask with 0x0f to get CPU extended family ID is bogus because
CPU extended family ID takes 8 bits (bit 27:20).
2. Use CPU extended family ID fields without checking CPU family ID is
risky. Because Intel SDM says, "The Extended Family ID needs to be
examined only when the Family ID is 0FH."
3. Saving cpu family ID and cpu extended family ID separately doesn't
align with Linux kernel. And it may bring extra complexity when
making family specific changes in the future.
Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
SST-PP level change results in online/offline of CPUs with -o option.
The Linux intel-pstate driver internally stores the current HWP_REQ MSR
value during offline and restores them during online.
It is possible that during SST-PP level change, the new HWP_CAP limits
can be updated. So, when a CPU is online, the HWP_REQ MSR should be
updated to new values based on HWP_CAP values.
This is particularly problematic when either turbo is disabled or the
current HWP_REQ value (stored before online) is less than the base
frequency from the updated HWP_CAP MSR guaranteed value. If the HWP_REQ
MSR is not updated, then the performance will be limited to the value
before perf level change.
Hence the tool updates cpufreq scaling_max_freq to the newer
base_frequency value in this case. This step is not required when HWP
interrupts are enabled, as the perf level change should result in a new
interrupt with HWP_GUARANTEED_PERF_CHANGE_STATUS and the intel_pstate
driver will update to new limits.
But the tool needs to handle the case when HWP interrupts are not
enabled but there is no way for the tool to know that HWP interrupts are
enabled or not. So, it has to still update the scaling_max_freq.
With the QOS changes in the kernel, user space writes to scaling_max_freq
are treated as hard limits. So, when base frequency is increased with
SST-BF enabled, the cpufreq subsystem will still not allow setting to the
SST-BF high priority core frequency. So, the HWP_REQ MSR will still be
capped to the user-set scaling_max_freq after SST-PP level change.
To address this, instead of setting scaling_max_freq to the current HWP_CAP
highest frequency, set it to the maximum integer value to set the QOS limit
as unconstrained. In this case, the actual HWP_REQ maximum frequency will
still be capped to HWP_CAP highest performance by the intel-pstate driver.
So, it will not result in invalid HWP_REQ values.
Signed-off-by: Srinivas Pandruvada <srinivas.pandruvada@linux.intel.com>
Commands that add counters, such as 'turbostat --show C1,C1+'
display merged columns without a delimiter.
This is caused by the bad syntax: '(*printed++ ? delim : "")', shared by
print_name()/print_hex_value()/print_decimal_value()/print_float_value()
Use '((*printed)++ ? delim : "")' to correctly increment the value at *printed.
[lenb: fix code and commit message typo, re-word]
Fixes: 56dbb87850 ("tools/power turbostat: Refactor added column header printing")
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Problem: individual swidle counter names (C1, C1+, C1-, etc.) cannot be
selected via --show/--hide due to two bugs in probe_cpuidle_counts():
1. The function returns immediately when BIC_cpuidle is not enabled,
without checking deferred_add_index.
2. The deferred name check runs against name_buf before the trailing
newline is stripped, so is_deferred_add("C1\n") never matches "C1".
Fix:
1. Relax the early return to pass through when deferred names are
queued.
2. Strip the trailing newline from name_buf before performing deferred
name checks.
3. Check each suffixed variant (C1+, C1, C1-) individually so that
e.g. "--show C1+" enables only the requested metric.
In addition, introduce a helper function to avoid repeating the
condition (readability cleanup).
Fixes: ec4acd3166 ("tools/power turbostat: disable "cpuidle" invocation counters, by default")
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
In the perf thread, core, and package counter loops, an incorrect
'mp->format' variable is used instead of 'pp->format'.
[lenb: edit commit message]
Fixes: 696d15cbd8 ("tools/power turbostat: Refactor floating point printout code")
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Fix the PMT thread code to use print_float_value(),
to be consistent with the PMT core and package code.
[lenb: commit message]
Signed-off-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
turbostat always used the same logic to read the microcode patch level,
which is correct for Intel but not for AMD/Hygon.
While Intel stores the patch level in the upper 32 bits of MSR, AMD
stores it in the lower 32 bits, which causes turbostat to report the
microcode version as 0x0 on AMD/Hygon.
Fix by shifting right by 32 for non-AMD/Hygon, preserving the existing
behavior for Intel and unknown vendors.
Fixes: 3e4048466c ("tools/power turbostat: Add --no-msr option")
Signed-off-by: Serhii Pievniev <spevnev16@gmail.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Linux core_id's are a per-package namespace, not a per-node namespace.
Rename topo.cores_per_node to topo.cores_per_pkg to reflect this.
Eliminate topo.nodes_per_pkg from the sizing for core data structures,
since it has no role except to unnecessarily bloat the allocation.
Validated on multiple Intel platforms (ICX/SPR/SRF/EMR/GNR/CWF) with
various CPU online/offline configurations and SMT enabled/disabled
scenarios.
No functional changes.
[lenb: commit message]
Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
I changed my mind about displaying swidle statistics,
which are "added counters". Recently I reverted the
column headers to 8-columns, but kept print_decimal_value()
padding out to 16-columns for all 64-bit counters.
Simplify by keeping print_decimial_value() at %lld -- which
will often fit into 8-columns, and live with the fact
that it can overflow and shift the other columns,
which continue to tab-delimited.
This is a better compromise than inserting a bunch
of space characters that most users don't like.
Fixes: 1a23ba6a1b ("tools/power turbostat: Print wide names only for RAW 64-bit columns")
Reported-by: Artem Bityutskiy <artem.bityutskiy@linux.intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
When SMT is present and disabled, turbostat may under-size
the thread_data array. This can corrupt results or
cause turbostat to exit with a segmentation fault.
[lenb: commit message]
Fixes: a2b4d0f8bf ("tools/power turbostat: Favor cpu# over core#")
Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
turbostat.c:8688: rapl_perf_init: Assertion `next_domain < num_domains' failed.
Two recent cleanup patches that were not supposed to change anything
broke the core_id code needed for AMD RAPL initialization:
commit 070e92361e ("tools/power turbostat: Enhance HT enumeration")
commit ddf60e38ca ("tools/power turbostat: Simplify global core_id calculation")
Reported-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Len Brown <len.brown@intel.com>
Since release 2025.12.02:
Add L2 statistics columns for recent Intel processors:
L2MRPS = L2 Cache M-References Per Second
L2%hit = L2 Cache Hit %
Sort work and output by cpu# rather than core#
Minor features and fixes.
-----BEGIN PGP SIGNATURE-----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=UG/p
-----END PGP SIGNATURE-----
Merge tag 'turbostat-2026.02.14' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux
Pull turbostat updates from Len Brown:
- Add L2 statistics columns for recent Intel processors:
L2MRPS = L2 Cache M-References Per Second
L2%hit = L2 Cache Hit %
- Sort work and output by cpu# rather than core#
- Minor features and fixes
* tag 'turbostat-2026.02.14' of git://git.kernel.org/pub/scm/linux/kernel/git/lenb/linux: (23 commits)
tools/power turbostat: version 2026.02.14
tools/power turbostat: Fix and document --header_iterations
tools/power turbostat: Use strtoul() for iteration parsing
tools/power turbostat: Favor cpu# over core#
tools/power turbostat: Expunge logical_cpu_id
tools/power turbostat: Enhance HT enumeration
tools/power turbostat: Simplify global core_id calculation
tools/power turbostat: Unify even/odd/average counter referencing
tools/power turbostat: Allocate average counters dynamically
tools/power turbostat: Delete core_data.core_id
tools/power turbostat: Rename physical_core_id to core_id
tools/power turbostat: Cleanup package_id
tools/power turbostat: Cleanup internal use of "base_cpu"
tools/power turbostat: Add L2 cache statistics
tools/power turbostat: Remove redundant newlines from err(3) strings
tools/power turbostat: Allow more use of is_hybrid flag
tools/power turbostat: Rename "LLCkRPS" column to "LLCMRPS"
tools/power turbostat.8: Document the "--force" option
tools/power turbostat: Harden against unexpected values
tools/power turbostat: Dump hypervisor name
...
Since release 2025.12.02:
Add L2 statistics columns for recent Intel processors:
L2MRPS = L2 Cache M-References Per Second
L2%hit = L2 Cache Hit %
Sort work and output by cpu# rather than core#
This commit:
Version number and white space (indent -l160)
No functional change.
Signed-off-by: Len Brown <len.brown@intel.com>
The "header_iterations" option is commonly used to de-clutter
the screen of redundant header label rows in an interactive session:
Eg. every 10 rows:
$ sudo turbostat --header_iterations 10 -S -q -i 1
But --header_iterations was missing from turbostat.8
Also turbostat help advertised the "-N" short option
that did not actually work:
$ turbostat --help
-N, --header_iterations num
print header every num iterations
Repair "-N"
Document "--header_iterations" on turbostat.8
Signed-off-by: Len Brown <len.brown@intel.com>
Replace strtod() with strtoul() and check errno for -n/-N options, since
num_iterations and header_iterations are unsigned long counters. Reject
zero and conversion errors; negative inputs wrap to large positive values
per standard unsigned semantics.
Signed-off-by: Kaushlendra Kumar <kaushlendra.kumar@intel.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Turbostat collects statistics and outputs results in "topology order",
which means it prioritizes the core# over the cpu#.
The strategy is to minimize wakesups to a core -- which is
important when measuring an idle system.
But core order is problematic, because Linux core#'s are physical
(within each package), and thus subject to APIC-id scrambling
that may be done by the hardware or the BIOS.
As a result users may be are faced with rows in a confusing order:
sudo turbostat -q --show topology,Busy%,CPU%c6,UncMHz sleep 1
Core CPU Busy% CPU%c6 UncMHz
- - 1.25 72.18 3400
0 4 7.74 0.00
1 5 1.77 88.59
2 6 0.48 96.73
3 7 0.21 98.34
4 8 0.14 96.85
5 9 0.26 97.55
6 10 0.44 97.24
7 11 0.12 96.18
8 0 5.41 0.31 3400
8 1 0.19
12 2 0.41 0.22
12 3 0.08
32 12 0.04 99.21
33 13 0.25 94.92
Abandon the legacy "core# topology order" in favor of simply
ordering by cpu#, with a special case to handle HT siblings
that may not have adjacent cpu#'s.
sudo ./turbostat -q --show topology,Busy%,CPU%c6,UncMHz sleep 1
1.003001 sec
Core CPU Busy% CPU%c6 UncMHz
- - 1.38 80.55 1600
8 0 10.94 0.00 1600
8 1 0.53
12 2 2.90 0.45
12 3 0.11
0 4 1.96 91.20
1 5 0.97 96.40
2 6 0.24 94.72
3 7 0.31 98.01
4 8 0.20 98.20
5 9 0.62 96.00
6 10 0.06 98.15
7 11 0.12 99.31
32 12 0.04 99.07
33 13 0.27 95.09
The result is that cpu#'s now take precedence over core#'s.
Signed-off-by: Len Brown <len.brown@intel.com>
Record the cpu_id of each CPU HT sibling -- will need this later.
Rename "thread_id" to "ht_id" to disambiguate that the scope
of this id is within a Core -- it is not a global cpu_id.
No functional change.
Signed-off-by: Len Brown <len.brown@intel.com>
Standardize the generation of globally unique core_id's
in a macro, and simplify the related code.
No functional change.
Signed-off-by: Len Brown <len.brown@intel.com>
Update the syntax of accesses to the even and odd counters
to match the average counters.
No functional change.
Signed-off-by: Len Brown <len.brown@intel.com>
The current static definition of average{} is inconsistent with
the dynamically allocated even{} and odd{} counters.
Allocate average{} counters dynamically.
No functional change.
Signed-off-by: Len Brown <len.brown@intel.com>
Delete redundant core_data.core_id.
Use cpus[].core_id as the single copy of the truth.
No functional change.
Signed-off-by: Len Brown <len.brown@intel.com>
The Linux Kernel topology sysfs is flawed.
core_id is not globally unique, but is per-package.
Turbostat works around this when it needs to, with
rapl_core_id = cpus[cpu].core_id;
rapl_core_id += cpus[cpu].package_id * nr_cores_per_package
Otherwise, turbostat handles core_id as subservient to each package.
As there is only one core_id namespace, rename
physical_core_id to simply be core_id.
No functional change.
Signed-off-by: Len Brown <len.brown@intel.com>
The kernel topology sysfs uses the name "physical_package_id"
because it is allowed to be sparse.
Inside Turbostat, that physical package_id namespace is the only
package_id namespace, so re-name it to simply be "package_id"
in cpus[].
Delete the redundant copy of package_id in pkg_data.
Rely instead on the single copy of the truth in cpus[].
No functional change.
Signed-off-by: Len Brown <len.brown@intel.com>
Disambiguate the uses "base_cpu":
master_cpu: lowest permitted cpu#, read global MSRs here
package_data.first_cpu: lowest permitted cpu# in that package
core_data.first_cpu: lowest permitted cpu# in the core
current_cpu: where I'm running now
No functional change.
Signed-off-by: Len Brown <len.brown@intel.com>
version 2026.02.04
Add support for L2 cache statistics: L2MRPS and L2%hit
L2 statistics join the LLC in the "cache" counter group.
While the underlying LLC perf kernel support was architectural,
L2 perf counters are model-specific:
Support Intel Xeon -- Sapphire Rapids and newer.
Support Intel Atom -- Gracemont and newer.
Support Intel Hybrid -- Alder Lake and newer.
Example:
alder-lake-n$ sudo turbostat --quiet --show CPU,Busy%,cache my_workload
CPU Busy% LLCMRPS LLC%hit L2MRPS L2%hit
- 49.82 1210 85.02 2909 31.63
0 99.14 322 88.89 767 32.38
1 0.91 1 32.47 1 18.86
2 0.20 0 40.78 0 23.34
3 99.17 295 81.79 706 31.89
4 0.68 1 58.71 1 15.61
5 99.16 299 85.65 726 31.32
6 0.08 0 45.35 0 31.71
7 99.21 293 83.63 707 30.92
where "my_workload" is a wrapper for a yogini workload
that has 4 fully-busy threads with 2MB working set each.
Note that analogous to the system summary for multiple LLC systems,
the system summary row for the L2 is the aggregate of all CPUS in the
system -- there is no per-cache roll-up.
Signed-off-by: Len Brown <len.brown@intel.com>
The "is_hybrid" is set and used only in !quiet mode.
Make it valid in both quiet and !quiet mode to allow more uses.
Signed-off-by: Len Brown <len.brown@intel.com>
