Struct Opts

pub(crate) struct Opts {

    pub(crate) exit_dump_len: u32,
    pub(crate) slice_us: u64,
    pub(crate) slice_us_min: u64,
    pub(crate) slice_us_lag: i64,
    pub(crate) idle_resume_us: i64,
    pub(crate) no_preempt: bool,
    pub(crate) local_pcpu: bool,
    pub(crate) local_kthreads: bool,
    pub(crate) no_wake_sync: bool,
    pub(crate) primary_domain: String,
    pub(crate) disable_l2: bool,
    pub(crate) disable_l3: bool,
    pub(crate) disable_smt: bool,
    pub(crate) disable_numa: bool,
    pub(crate) cpufreq: bool,
    pub(crate) nvcsw_max_thresh: u64,
    pub(crate) stats: Option<f64>,
    pub(crate) monitor: Option<f64>,
    pub(crate) debug: bool,
    pub(crate) verbose: bool,
    pub(crate) version: bool,
    pub(crate) help_stats: bool,
}

scx_bpfland: a vruntime-based sched_ext scheduler that prioritizes interactive workloads.

This scheduler is derived from scx_rustland, but it is fully implemented in BPF. It has a minimal user-space part written in Rust to process command line options, collect metrics and log out scheduling statistics.

The BPF part makes all the scheduling decisions (see src/bpf/main.bpf.c).

Fields

exit_dump_len: u32

Exit debug dump buffer length. 0 indicates default.

slice_us: u64

Maximum scheduling slice duration in microseconds.

slice_us_min: u64

Minimum scheduling slice duration in microseconds.

slice_us_lag: i64

Maximum time slice lag in microseconds.

A positive value can help to enhance the responsiveness of interactive tasks, but it can also make performance more “spikey”.

A negative value can make performance more consistent, but it can also reduce the responsiveness of interactive tasks (by smoothing the effect of the vruntime scheduling and making the task ordering closer to a FIFO).

idle_resume_us: i64

Set CPU idle QoS resume latency in microseconds (-1 = disabled).

Setting a lower latency value makes CPUs less likely to enter deeper idle states, enhancing performance at the cost of higher power consumption. Alternatively, increasing the latency value may reduce performance, but also improve power efficiency.

no_preempt: bool

Disable preemption.

Never allow tasks to be directly dispatched. This can help to increase fairness over responsiveness.

local_pcpu: bool

Enable per-CPU tasks prioritization.

This allows to prioritize per-CPU tasks that usually tend to be de-prioritized (since they can’t be migrated when their only usable CPU is busy). Enabling this option can introduce unfairness and potentially trigger stalls, but it can improve performance of server-type workloads (such as large parallel builds).

local_kthreads: bool

Enable kthreads prioritization (EXPERIMENTAL).

Enabling this can improve system performance, but it may also introduce noticeable interactivity issues or unfairness in scenarios with high kthread activity, such as heavy I/O or network traffic.

Use it only when conducting specific experiments or if you have a clear understanding of its implications.

no_wake_sync: bool

Disable direct dispatch during synchronous wakeups.

Enabling this option can lead to a more uniform load distribution across available cores, potentially improving performance in certain scenarios. However, it may come at the cost of reduced efficiency for pipe-intensive workloads that benefit from tighter producer-consumer coupling.

primary_domain: String

Specifies the initial set of CPUs, represented as a bitmask in hex (e.g., 0xff), that the scheduler will use to dispatch tasks, until the system becomes saturated, at which point tasks may overflow to other available CPUs.

Special values:

• “auto” = automatically detect the CPUs based on the active power profile
• “performance” = automatically detect and prioritize the fastest CPUs
• “powersave” = automatically detect and prioritize the slowest CPUs
• “all” = all CPUs assigned to the primary domain
• “none” = no prioritization, tasks are dispatched on the first CPU available

disable_l2: bool

Disable L2 cache awareness.

disable_l3: bool

Disable L3 cache awareness.

disable_smt: bool

Disable SMT awareness.

disable_numa: bool

Disable NUMA rebalancing.

cpufreq: bool

Enable CPU frequency control (only with schedutil governor).

With this option enabled the CPU frequency will be automatically scaled based on the load.

nvcsw_max_thresh: u64

[DEPRECATED] Maximum threshold of voluntary context switches per second. This is used to classify interactive.

tasks (0 = disable interactive tasks classification).

stats: Option<f64>

Enable stats monitoring with the specified interval.

monitor: Option<f64>

Run in stats monitoring mode with the specified interval. Scheduler is not launched.

debug: bool

Enable BPF debugging via /sys/kernel/tracing/trace_pipe.

verbose: bool

Enable verbose output, including libbpf details.

version: bool

Print scheduler version and exit.

help_stats: bool

Show descriptions for statistics.

Trait Implementations

impl Args for Opts

fn group_id() -> Option<Id>

Report the [ArgGroup::id][crate::ArgGroup::id] for this set of arguments

fn augment_args<'b>(__clap_app: Command) -> Command

Append to [Command] so it can instantiate Self via [FromArgMatches::from_arg_matches_mut]

fn augment_args_for_update<'b>(__clap_app: Command) -> Command

Append to [Command] so it can instantiate self via [FromArgMatches::update_from_arg_matches_mut]

impl CommandFactory for Opts

fn command<'b>() -> Command

Build a [Command] that can instantiate Self.

fn command_for_update<'b>() -> Command

Build a [Command] that can update self.

impl Debug for Opts

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl FromArgMatches for Opts

fn from_arg_matches(__clap_arg_matches: &ArgMatches) -> Result<Self, Error>

Instantiate Self from [ArgMatches], parsing the arguments as needed.

fn from_arg_matches_mut( __clap_arg_matches: &mut ArgMatches, ) -> Result<Self, Error>

Instantiate Self from [ArgMatches], parsing the arguments as needed.

fn update_from_arg_matches( &mut self, __clap_arg_matches: &ArgMatches, ) -> Result<(), Error>

Assign values from ArgMatches to self.

fn update_from_arg_matches_mut( &mut self, __clap_arg_matches: &mut ArgMatches, ) -> Result<(), Error>

Assign values from ArgMatches to self.

impl Parser for Opts

fn parse() -> Self

Parse from std::env::args_os(), [exit][Error::exit] on error.

fn try_parse() -> Result<Self, Error>

Parse from std::env::args_os(), return Err on error.

fn parse_from<I, T>(itr: I) -> Self

where I: IntoIterator<Item = T>, T: Into<OsString> + Clone,

Parse from iterator, [exit][Error::exit] on error.

fn try_parse_from<I, T>(itr: I) -> Result<Self, Error>

where I: IntoIterator<Item = T>, T: Into<OsString> + Clone,

Parse from iterator, return Err on error.

fn update_from<I, T>(&mut self, itr: I)

where I: IntoIterator<Item = T>, T: Into<OsString> + Clone,

Update from iterator, [exit][Error::exit] on error.

fn try_update_from<I, T>(&mut self, itr: I) -> Result<(), Error>

where I: IntoIterator<Item = T>, T: Into<OsString> + Clone,

Update from iterator, return Err on error.