powercurve

A Linux power management daemon with built-in fan curves.

What it does

• Power profiles compatible with power-profiles-daemon. Desktop tooling and powerprofilesctl keep working unchanged.
• Configurable fan control through hwmon PWM with curves, hysteresis, and thermal protection.

Fan control is optional. Without a fan config, PowerCurve behaves like ppd.

Desktop only. This is built for desktops and workstations. Most laptops manage fans through embedded controllers that don't expose PWM via hwmon. Fan control won't work there. The power profile side functions fine on any hardware.

Installation

Arch

Available on the AUR:

yay -S powercurve-git

Debian / Ubuntu / Pop!_OS

Download the latest .deb from the releases page:

sudo apt install ./powercurve_*.deb

Both packages provide and conflict with power-profiles-daemon. The stock daemon gets replaced cleanly.

Power profiles

Three profiles, switchable via D-Bus or the CLI:

• Quiet maps to power-saver. Conservative CPU governor, turbo disabled, aggressive PCI and SCSI power management.
• Balanced is the default. Dynamic CPU scaling with turbo enabled, moderate power management.
• Performance runs the CPU at full frequency with turbo and disables PCI runtime power management.

Set a profile:

powercurve profile quiet
powercurve profile balanced
powercurve profile performance

Query the current profile:

powercurve profile

The active profile persists across restarts.

Fan control

Fan control reads /etc/powercurve/fan.toml. The Arch package generates one on install by scanning hwmon devices. Without a config, the daemon manages profiles only.

To inspect your hardware:

powercurve fan-detect

To generate a config (or regenerate after hardware changes):

powercurve fan-detect --generate | sudo tee /etc/powercurve/fan.toml > /dev/null

Critical temperatures get set 15C below the CPU's thermal throttle point.

A minimal config

platform = "nct6775"
critical_cpu_temp = 80
critical_gpu_temp = 75
hysteresis = 3.0

[[curve]]
temp = 30.0
duty = 10

[[curve]]
temp = 50.0
duty = 30

[[curve]]
temp = 70.0
duty = 80

[[curve]]
temp = 75.0
duty = 100

[[channels]]
pwm = "pwm1"
source = "cpu"

[[channels]]
pwm = "pwm2"
source = "all"

platform is the hwmon device with your PWM outputs (fan-detect fills it in). The source on each channel picks the sensor: cpu, gpu, or all (max of the two). Temperatures are Celsius, duties are percentages. The daemon interpolates linearly between points.

Curves are layered: per-channel per-profile, per-channel default, per-profile global, and a shared top-level fallback. First match wins. The example above uses only the shared fallback. That covers most setups.

The examples/ directory has progressively more involved configs: fan-simple.toml, fan-desktop.toml, fan-profiles.toml. Everything else lives in man powercurve: minimum-duty floors, stall detection, passthrough mode, thermal fallback, per-channel per-profile curves, NVIDIA support via NVML, and the full config field reference.

Status and overrides

Check the current state:

powercurve status

This shows the active profile, CPU and GPU temperatures, per-channel PWM duties, RPMs, and the active curve for each channel. Doesn't need root.

For testing, override a channel's duty without editing the config:

powercurve fan pwm3 50      # set pwm3 to 50%
powercurve fan pwm3 clear   # release

Switching profiles clears all overrides.

To find a channel's spin-up floor (the lowest duty where the fan actually spins):

powercurve fan-test pwm1

Use the reported value as min_duty in fan.toml.

Desktop notifications on profile switches and thermal events come from the monitor user service:

systemctl --user enable --now powercurve-monitor

Process watcher

The watcher polls /proc and switches profiles automatically when a process matches a rule. Useful for games and encoding jobs.

Configure it at ~/.config/powercurve/watcher.toml:

[watcher]
poll_interval = 5
default_profile = "balanced"

[[rule]]
name = "gaming"
match_exe = "steam_app_*"
profile = "performance"

[[rule]]
name = "video-encoding"
match_cmd = "ffmpeg.*libx265"
profile = "performance"

match_exe is a glob on the process name. match_cmd is a regex on the full command line. First match wins. When nothing matches, the default_profile is restored if set.

Run as a user service:

systemctl --user enable --now powercurve-watcher

Without a config file, the watcher idles.

Validation and hot reload

Validate without restarting:

powercurve config

This checks curve monotonicity, duty ranges, critical temp bounds, hysteresis and cooldown values, profile names, per-channel profile curves, and whether the referenced hwmon devices exist on the current machine.

Reload after editing:

sudo systemctl reload powercurve

The daemon validates first. If validation fails, the running config is kept and the error is logged.

Building

Requires a stable Rust toolchain:

make
sudo make install
sudo systemctl enable --now com.vintagetechie.PowerCurve

Use at your own risk

This daemon writes directly to hardware PWM registers. A bad fan config can result in inadequate cooling and hardware damage. The built-in thermal protection sets fans to maximum when critical temperatures are reached, but it is a safety net, not a substitute for a correct configuration. Test your curves, verify your temps, and don't deploy a config you haven't validated. See the LICENSE for full terms.

More

man powercurve is the full reference. Every command, every config option, the D-Bus interface, signals, and file locations.

CHANGELOG.md lists what changed in each release. ROLLBACK.md covers switching back to power-profiles-daemon or downgrading to a previous version.

License

GPL-3.0-only

Based on system76-power by System76, also licensed under GPL-3.0-only.