extern char *optarg and extern int optind, opterr, optopt are
already declared by <getopt.h>, which is included at the top of
the file. Repeating extern declarations inside a function body
is misleading and unnecessary.
Signed-off-by: Kaushlendra Kumar <kaushlendra.kumar@intel.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
The cpupower-info(1) man page only mentions the short form of the
'--perf-bias' option in the synopsys, but the long form is not documented
and its effect is not explained.
cpupower-info.c:
{"perf-bias", optional_argument, NULL, 'b'},
Signed-off-by: Roberto Ricci <io@r-ricci.it>
Link: https://lore.kernel.org/r/20260324223921.14317-5-io@r-ricci.it
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
`cpupower frequency-info` supports the '--boost' option since the program
was first added with commit 7fe2f6399a ("cpupowerutils - cpufrequtils
extended with quite some features"), but the man page lacks it.
'--epp' has been added with commit 5f567afc28 ("cpupower: Add support for
showing energy performance preference") but it has never been added to the
man page.
cpufreq-info.c:
{"boost", no_argument, NULL, 'b'},
...
{"epp", no_argument, NULL, 'z'},
Signed-off-by: Roberto Ricci <io@r-ricci.it>
Link: https://lore.kernel.org/r/20260324223921.14317-4-io@r-ricci.it
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
The cpupower-idle-info(1) man page describes '-f' as the short form of the
'--silent' option and '-e' as the short form of the '--proc' option.
But they are not correct:
$ cpupower idle-info -f
idle-info: invalid option -- 'f'
invalid or unknown argument
$ cpupower idle-info -e
idle-info: invalid option -- 'e'
invalid or unknown argument
The short form of '--silent' is actually '-s' and the short form of
'--proc' is actually '-o':
cpuidle-info.c:
{"silent", no_argument, NULL, 's'},
{"proc", no_argument, NULL, 'o'},
Signed-off-by: Roberto Ricci <io@r-ricci.it>
Link: https://lore.kernel.org/r/20260324223921.14317-2-io@r-ricci.it
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
On modern Intel platforms, the intel_pstate driver is commonly used and
it provides turbo boost control via
/sys/devices/system/cpu/intel_pstate/no_turbo.
However, cpupower doesn't handle this. it
1. shows turbo boost as "active" blindly for Intel platforms
2. controls turbo boost functionality via the generic
/sys/devices/system/cpu/cpufreq/boost sysfs interface only.
Enhance the cpupower tool to ensure the "--boost" command works
seamlessly on Intel platforms with intel_pstate driver running.
Without this patch,
$ echo 1 | sudo tee /sys/devices/system/cpu/intel_pstate/no_turbo
1
$ sudo cpupower frequency-info --boost
analyzing CPU 21:
boost state support:
Supported: yes
Active: yes
$ sudo cpupower set --boost 0
Error setting turbo-boost
$ sudo cpupower set --boost 1
Error setting turbo-boost
With this patch,
$ cat /sys/devices/system/cpu/intel_pstate/no_turbo
0
$ sudo cpupower set --boost 0
$ sudo cpupower frequency-info --boost
analyzing CPU 21:
boost state support:
Supported: yes
Active: no
$ cat /sys/devices/system/cpu/intel_pstate/no_turbo
1
$ sudo cpupower set --boost 1
$ sudo cpupower frequency-info --boost
analyzing CPU 28:
boost state support:
Supported: yes
Active: yes
$ cat /sys/devices/system/cpu/intel_pstate/no_turbo
0
Signed-off-by: Zhang Rui <rui.zhang@intel.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
Extend the systemd service so that it can be used for tuning the Energy
Performance Preference (EPP) as well. Available options can be read from
/sys/devices/system/cpu/cpufreq/policy0/energy_performance_available_preferences.
The desired one can then be set in cpupower-service.conf.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
cpupower currently installs a cpupower.service unit file into unitdir
unconditionally, regardless of whether systemd is used by the host.
Improve the installation procedure by making this systemd step optional:
a 'SYSTEMD' build parameter that defaults to 'true' and can be set to
'false' to disable the installation of systemd's unit file.
Since 'SYSTEMD' defaults to true, the current behavior is kept as the
default.
Link: https://lore.kernel.org/r/20260113132753.1730020-2-joaomarcos.costa@bootlin.com
Signed-off-by: João Marcos Costa <joaomarcos.costa@bootlin.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
The cpuidle sysfs monitor printed the previous sample’s counter
value in cpuidle_stop() instead of the freshly read one. The dprint
line used previous_count[cpu][state] while current_count[cpu][state]
had just been populated. This caused misleading debug output.
Switch the logging to current_count so the post-interval snapshot
matches the displayed value.
Link: https://lore.kernel.org/r/20251202044012.3844790-1-kaushlendra.kumar@intel.com
Signed-off-by: Kaushlendra Kumar <kaushlendra.kumar@intel.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
The cpupower Makefile built and installed libcpupower as a shared
library (libcpupower.so) without passing `STATIC=true`, but did not
build a static version of the library even with `STATIC=true`. (Only the
programs were static). Thus, out-of-tree programs using libcpupower
were unable to link statically against the library without having access
to intermediate object files produced during the build.
This fixes that situation by ensuring that libcpupower.a is built and
installed when `STATIC=true` is specified.
Link: https://lore.kernel.org/r/x7geegquiks3zndiavw2arihdc2rk7e2dx3lk7yxkewqii6zpg@tzjijqxyzwmu
Signed-off-by: Zuo An <zuoan.penguin@gmail.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
Pull Kbuild updates from Nathan Chancellor:
- Extend modules.builtin.modinfo to include module aliases from
MODULE_DEVICE_TABLE for builtin modules so that userspace tools (such
as kmod) can verify that a particular module alias will be handled by
a builtin module
- Bump the minimum version of LLVM for building the kernel to 15.0.0
- Upgrade several userspace API checks in headers_check.pl to errors
- Unify and consolidate CONFIG_WERROR / W=e handling
- Turn assembler and linker warnings into errors with CONFIG_WERROR /
W=e
- Respect CONFIG_WERROR / W=e when building userspace programs
(userprogs)
- Enable -Werror unconditionally when building host programs
(hostprogs)
- Support copy_file_range() and data segment alignment in gen_init_cpio
to improve performance on filesystems that support reflinks such as
btrfs and XFS
- Miscellaneous small changes to scripts and configuration files
* tag 'kbuild-6.18-1' of git://git.kernel.org/pub/scm/linux/kernel/git/kbuild/linux: (47 commits)
modpost: Initialize builtin_modname to stop SIGSEGVs
Documentation: kbuild: note CONFIG_DEBUG_EFI in reproducible builds
kbuild: vmlinux.unstripped should always depend on .vmlinux.export.o
modpost: Create modalias for builtin modules
modpost: Add modname to mod_device_table alias
scsi: Always define blogic_pci_tbl structure
kbuild: extract modules.builtin.modinfo from vmlinux.unstripped
kbuild: keep .modinfo section in vmlinux.unstripped
kbuild: always create intermediate vmlinux.unstripped
s390: vmlinux.lds.S: Reorder sections
KMSAN: Remove tautological checks
objtool: Drop noinstr hack for KCSAN_WEAK_MEMORY
lib/Kconfig.debug: Drop CLANG_VERSION check from DEBUG_INFO_DWARF_TOOLCHAIN_DEFAULT
riscv: Remove ld.lld version checks from many TOOLCHAIN_HAS configs
riscv: Unconditionally use linker relaxation
riscv: Remove version check for LTO_CLANG selects
powerpc: Drop unnecessary initializations in __copy_inst_from_kernel_nofault()
mips: Unconditionally select ARCH_HAS_CURRENT_STACK_POINTER
arm64: Remove tautological LLVM Kconfig conditions
ARM: Clean up definition of ARM_HAS_GROUP_RELOCS
...
Fix incorrect size parameter passed to cpuidle_state_write_file() in
cpuidle_state_disable().
The function was incorrectly using sizeof(disable) which returns the
size of the unsigned int variable (4 bytes) instead of the actual
length of the string stored in the 'value' buffer.
Since 'value' is populated with snprintf() to contain the string
representation of the disable value, we should use the length
returned by snprintf() to get the correct string length for
writing to the sysfs file.
This ensures the correct number of bytes is written to the cpuidle
state disable file in sysfs.
Link: https://lore.kernel.org/r/20250917050820.1785377-1-kaushlendra.kumar@intel.com
Signed-off-by: Kaushlendra Kumar <kaushlendra.kumar@intel.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
Commit bd7c231212 ("pinctrl: meson: Fix typo in device table macro")
is needed in kbuild-next to avoid a build error with a future change.
While at it, address the conflict between commit 41f9049cff ("riscv:
Only allow LTO with CMODEL_MEDANY") and commit 6578a1ff6a ("riscv:
Remove version check for LTO_CLANG selects"), as reported by Stephen
Rothwell [1].
Link: https://lore.kernel.org/20250908134913.68778b7b@canb.auug.org.au/ [1]
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
The cpupower_write_sysfs() function currently returns -1 on
write failure, but the function signature indicates it should
return an unsigned int. Returning -1 from an unsigned function
results in a large positive value rather than indicating
an error condition.
Fix this by returning 0 on failure, which is more appropriate
for an unsigned return type and maintains consistency with typical
success/failure semantics where 0 indicates failure and non-zero
indicates success (bytes written).
Link: https://lore.kernel.org/r/20250828063000.803229-1-kaushlendra.kumar@intel.com
Signed-off-by: Kaushlendra Kumar <kaushlendra.kumar@intel.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
The cpufreq subsystem has a generic sysfs interface for controlling boost
(/sys/devices/system/cpu/cpufreq/boost).
The sysfs interface can be used to enable boost control from the cpupower
command on non-x86 platforms as well. So, allow boost controlling
on non-x86 system if boost sysfs file exists.
The set subcommand enables/disables the boost feature using the following
syntax:
cpupower set --boost 1
cpupower set --boost 0
The --boost option is an alias for --turbo-boost. We provided the neutral
option name because the name "turbo boost" is specific to Intel technology.
The frequency-info subcommand displays the enabled/disabled state of
the boost feature as follows:
boost state support:
Active: yes (or no)
Link: https://lore.kernel.org/r/20250522061122.2149188-3-fj5851bi@fujitsu.com
Signed-off-by: Shinji Nomoto <fj5851bi@fujitsu.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
compile_commands.json can be used with clangd to enable language server
protocol-based assistance. For kernel itself this can be built with
scripts/gen_compile_commands.py, but other projects (e.g., libbpf, or
BPF selftests) can benefit from their own compilation database file,
which can be generated successfully using external tools, like bear [0].
So, instead of adding compile_commands.json to .gitignore in respective
individual projects, let's just ignore it globally anywhere in Linux repo.
While at it, remove exactly such a local .gitignore rule under
tools/power/cpupower.
[0] https://github.com/rizsotto/Bear
Reviewed-by: Nathan Chancellor <nathan@kernel.org>
Suggested-by: Eduard Zingerman <eddyz87@gmail.com>
Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
Reviewed-by: Miguel Ojeda <ojeda@kernel.org>
Link: https://lore.kernel.org/r/20250606214840.3165754-1-andrii@kernel.org
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
After the commit 0014f65e3d ("pm: cpupower: remove hard-coded
topology depth values"), "cpupower monitor" output ceased to print the
CORE and the CPU fields on a multi-socket platform.
The reason for this is that the patch changed the behaviour to break
out of the switch-case after printing the PKG details, while prior to
the patch, the CORE and the CPU details would also get printed since
the "if" condition check would pass for any level whose topology depth
was lesser than that of a package.
Fix this ensuring all the details below a desired topology depth are
printed in the cpupower monitor output.
Link: https://lore.kernel.org/r/20250612122355.19629-3-gautham.shenoy@amd.com
Fixes: 0014f65e3d ("pm: cpupower: remove hard-coded topology depth values")
Signed-off-by: Gautham R. Shenoy <gautham.shenoy@amd.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
In the function mperf_start(), mperf_monitor snapshots the time, tsc
and finally the aperf,mperf MSRs. However, this order of snapshotting
in is reversed in mperf_stop(). As a result, the C0 residency (which
is computed as delta_mperf * 100 / delta_tsc) is under-reported on
CPUs that is 100% busy.
Fix this by snapshotting time, tsc and then aperf,mperf in
mperf_stop() in the same order as in mperf_start().
Link: https://lore.kernel.org/r/20250612122355.19629-2-gautham.shenoy@amd.com
Signed-off-by: Gautham R. Shenoy <gautham.shenoy@amd.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
One of the most typical use cases of the 'cpupower' utility works as
follows: run 'cpupower' at boot with the desired command-line options
and then forget about it.
Add a systemd service (disabled by default) that automates this use
case (for environments where the initialization system is 'systemd'),
by running 'cpupower' at boot with the settings read from a default
configuration file.
The systemd service, the associated support script and the
corresponding default configuration file are derived from what is
provided by the Arch Linux package (under "GPL-2.0-or-later" terms),
modernized and enhanced in various ways (the script has also been
checked with 'shellcheck').
Link: dd2e2a311e
Signed-off-by: Francesco Poli (wintermute) <invernomuto@paranoici.org>
Reviewed-by: John B. Wyatt IV <jwyatt@redhat.com>
Reviewed-by: John B. Wyatt IV <sageofredondo@gmail.com>
Tested-by: John B. Wyatt IV <jwyatt@redhat.com>
Tested-by: John B. Wyatt IV <sageofredondo@gmail.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
library versioning was broken:
libcpupower.so.0.0.1
libcpupower.so -> libcpupower.so.0.0.1
libcpupower.so.1 -> libcpupower.so.0.0.1
and is fixed by this patch to:
libcpupower.so.1.0.1
libcpupower.so -> libcpupower.so.1.0.1
libcpupower.so.1 -> libcpupower.so.1.0.1
Link: https://lore.kernel.org/r/20250307094334.39587-1-trenn@suse.de
Signed-off-by: Thomas Renninger <trenn@suse.de>
Signed-off-by: Shuah Khan <shuah@kernel.org>
This patch is also an issue report. get_cpu_topology will always save
into cpupower_topology a cores size of 0. The code to handle this looks
like it was commented out, and what is commented out is missing a curly
bracket.
https://elixir.bootlin.com/linux/v6.13.5/source/tools/power/cpupower/lib/cpupower.c#L206-L212
Inspiration was taken from psutil to implement this by querying
core_cpu_list. Instead of using a hashmap, I used a sorted array, and
counted the number of valid unique strings. The counting of this takes
place before the qsort for .pkg as the following code says it is
dependent on the order of that sort.
The previous code claimed Intel CPUs are not numbered correctly. I was
not able to reproduce that issue and removed that comment and the code.
This commit was tested with the libcpupower SWIG Python bindings and
performed correctly on 4 different setups. The most notable is the
Framework Intel laptop; a hybrid system of 4 P cores (8 threads) and 8 E
cores (8 threads).
The 4 setups: A 4 core virt-manager VM running Fedora 41 4c/4t (specs not
listed) was tested as a sanity test for VMs. A Lenovo Ryzen 7 Pro 7840HS
8c/16t. A Supermico Intel(R) Xeon(R) Gold 6330 CPU w/ 56c/112t with 2 CPU
sockets. A Framework 12th Gen Intel(R) Core(TM) i5-1240P with hybrid
cores.
CPU(s): 16
On-line CPU(s) list: 0-15
Vendor ID: AuthenticAMD
Model name: AMD Ryzen 7 PRO 7840HS w/ Radeon 780M Graphics
CPU family: 25
Model: 116
Thread(s) per core: 2
Core(s) per socket: 8
Socket(s): 1
Stepping: 1
CPU(s): 112
On-line CPU(s) list: 0-111
Vendor ID: GenuineIntel
BIOS Vendor ID: Intel(R) Corporation
Model name: Intel(R) Xeon(R) Gold 6330 CPU @ 2.00GHz
BIOS Model name: Intel(R) Xeon(R) Gold 6330 CPU @ 2.00GHz CPU @ 2.0GHz
BIOS CPU family: 179
CPU family: 6
Model: 106
Thread(s) per core: 2
Core(s) per socket: 28
Socket(s): 2
Stepping: 6
CPU(s): 16
On-line CPU(s) list: 0-15
Vendor ID: GenuineIntel
Model name: 12th Gen Intel(R) Core(TM) i5-1240P
CPU family: 6
Model: 154
Thread(s) per core: 2
Core(s) per socket: 12
Socket(s): 1
Stepping: 3
Link: https://lore.kernel.org/r/20250305210901.24177-1-jwyatt@redhat.com
Signed-off-by: "John B. Wyatt IV" <jwyatt@redhat.com>
Signed-off-by: "John B. Wyatt IV" <sageofredondo@gmail.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
In the case that we give a invalid command to idle_monitor for
monitoring, the execvp() will fail and thus go to the next line.
As a result, we'll see two differnt monitoring output. For
example, running `cpupower monitor -i 5 invalidcmd` which `invalidcmd`
is not executable.
Link: https://lore.kernel.org/r/20250220163846.2765-1-s921975628@gmail.com
Signed-off-by: Yiwei Lin <s921975628@gmail.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
Commit 'cpupower: Make TSC read per CPU for Mperf monitor' (c2adb1877b)
changes TSC counter reads per cpu, but left time diff global (from start
of all cpus to end of all cpus), thus diff(time) is too large for a
cpu's tsc counting, resulting in far less than acutal TSC_Mhz and thus
`cpupower monitor` showing far less than actual cpu realtime frequency.
/proc/cpuinfo shows frequency:
cat /proc/cpuinfo | egrep -e 'processor' -e 'MHz'
...
processor : 171
cpu MHz : 4108.498
...
before fix (System 100% busy):
| Mperf || Idle_Stats
CPU| C0 | Cx | Freq || POLL | C1 | C2
171| 0.77| 99.23| 2279|| 0.00| 0.00| 0.00
after fix (System 100% busy):
| Mperf || Idle_Stats
CPU| C0 | Cx | Freq || POLL | C1 | C2
171| 0.46| 99.54| 4095|| 0.00| 0.00| 0.00
Fixes: c2adb1877b ("cpupower: Make TSC read per CPU for Mperf monitor")
Signed-off-by: He Rongguang <herongguang@linux.alibaba.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
The is_valid flag should reflect the validity state of both
the XXX_start and XXX_stop functions. But the use of '=' in
XXX_stop overwrites the validity state set by XXX_start. This
commit changes '=' to '|=' in XXX_stop to preserve and combine
the validity state of XXX_start and XXX_stop.
Signed-off-by: wangfushuai <wangfushuai@baidu.com>
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
