Initial PR for scheduler scx_lunar

Cargo.toml

@@ -26,6 +26,7 @@ members = [
     "scheds/rust/scx_tickless",
     "scheds/experimental/scx_flow",
     "scheds/experimental/scx_rlfifo",
+    "scheds/experimental/scx_lunar",
     "tools/scx_characterize",
     "tools/scxtop",
     "tools/scx_forge_agent",

scheds/experimental/scx_lunar/Cargo.toml

@@ -0,0 +1,22 @@
+[package]
+name = "scx_lunar"
+version = "1.2.0"
+authors = ["Timon Stipkovits <timon2201@gmail.com>"]
+edition = "2021"
+description = "Multi-queue all purpose scheduler with per-CPU/per-LLC dispatch queues"
+license = "GPL-2.0-only"
+
+[dependencies]
+anyhow = "1"
+ctrlc = { version = "3", features = ["termination"] }
+clap = { version = "4", features = ["derive", "env", "unicode", "wrap_help"] }
+scx_utils = { path = "../../../rust/scx_utils", version = "1.1.2" }
+libbpf-rs = "=0.26.2"
+log = "0.4"
+simplelog = "0.12"
+
+[build-dependencies]
+scx_cargo = { path = "../../../rust/scx_cargo", version = "1.1.2" }
+
+[features]
+enable_backtrace = []

scheds/experimental/scx_lunar/LICENSE

@@ -0,0 +1 @@
+../../../LICENSE

scheds/experimental/scx_lunar/README.md

@@ -0,0 +1,81 @@
+
+# Lunar
+
+## Introduction
+
+Scx_lunar is a multipurpose scheduler which was originally invented with the goal to make frametimes in games as smooth as possible
+But then it grew a little and changed to a desktop usage focused scheduler which focuses on IO bound threads.
+
+Which makes the scheduler one of the best when it comes to responsiveness.
+
+This scheduler uses only FIFO queues.
+
+## Explanation
+
+The scheduler works with accounting of duty.
+
+Duty goes from 0 to 1023.
+
+The higher the duty number the more the task hogs cpu power.
+The lower the more it is sleeping or dependent on io.
+
+The duty is calculated from a window of the last 100ms
+
+It is calculated like this.
+
+duty = sleep_time * 1024 /(run_time + sleep_time + 1)
+
+The Tier are calculated as Percent of the 1024 max duty value.
+
+It has 5 tiers. Which are: 
+
+1. LC with duty <= 5%
+2. INTERACTIVE with duty <= 20%
+3. NORMAL with duty <= 40%
+4. BATCH with duty <= 90%
+5. GREEDY with duty <= 100%
+
+All new tasks get thrown into greedy. And start with duty of 1023.
+There is also a min. sample rate of the duty value to be eligible for promotion into higher tiers. 
+
+Each tier also has a slice time of 500us.
+
+When a lower tier task is running at the moment a higher tier gets enqueued then the current task gets kicked and preempted.
+
+## MODES
+
+This scheduler also has 2 modes.
+
+`--mode dsqs_per_llc` 
+
+
+Where the above explained are available for each LLC. So more than one core pull from the same DSQs.
+
+and:
+
+`--mode dsqs_per_cpu`
+
+DEFAULT MODE!
+Where the above explained dsqs are available for each cpu core. So each core has its own queues.
+This mode is used automatically when starting without start parameters.
+
+## Dispatch
+
+For mode `dsqs_per_cpu`
+Each core first tries to run its own queued tasks, then from another core from the same llc and then from core of other llcs.
+From which core the core startes stealing is randomized for better load distribution.
+
+for mode `dsqs_per_llc`
+Each core tries to first to run from the dsqs of the llc from the core. Then it tries to steal from other llcs.
+
+## Testing
+
+There where 2 design goals for this scheduler.
+
+1. That music keeps playing normally when executing the cachyos benchmarker https://github.com/CachyOS/cachyos-benchmarker
+2. To keep frametimes as smooth as possible with as little frametime spikes as possible. 
+
+As far as i have tested. Both modes do accomplish these tasks very well.
+
+The only problem is i couldn't test the functionality with different llcs as i don't have such an cpu by hand.
+The next thing is, that i mostly developed this scheduler with SMT disabled. As i found that SMT off works the best for this ryzen 5800x3d. But you can test both. Your mileage may vary.

scheds/experimental/scx_lunar/build.rs

@@ -0,0 +1,7 @@
+fn main() {
+    scx_cargo::BpfBuilder::new()
+        .unwrap()
+        .enable_skel("src/bpf/main.bpf.c", "bpf")
+        .build()
+        .unwrap();
+}

scheds/experimental/scx_lunar/src/bpf/datatypes.h

@@ -0,0 +1,57 @@
+// SPDX-License-Identifier: GPL-2.0
+//
+// Author: Timon Stipkovits <timon2201@gmail.com>
+//
+// This software may be used and distributed according to the terms of the
+// GNU General Public License version 2.
+
+#ifndef DATATYPES_H
+#define DATATYPES_H
+#include "defines.h"
+
+const volatile u32 nr_llcs = 1;
+const volatile u32 cpu_to_llc[MAX_CPUS] = {};
+const volatile u32 schedulerMode = SCHED_MODE_DSQ_PER_CPU;
+
+extern const int CONFIG_HZ __kconfig;
+
+struct task_ctx
+{
+  u64 current_dsq_type;
+  u64 runtime_avg;
+  u64 current_runtime;
+  u64 blocked_at;
+  u64 runnable_at;
+  s64 duty;
+  u64 run_acc;
+  u64 sleep_acc;
+  u64 last_run_granted_slice;
+  bool first_runtime_avg_sample_taken;
+  u64 started_at;
+  u64 duty_samples;
+};
+
+struct dispatch_ctx
+{
+  u64 current_task_deadline;
+  u64 current_task_dsq_type;
+  u64 last_kick_timestamp;
+};
+
+struct
+{
+  __uint(type, BPF_MAP_TYPE_PERCPU_ARRAY);
+  __uint(max_entries, 1);
+  __type(key, u32);
+  __type(value, struct dispatch_ctx);
+} dispatch_state SEC(".maps");
+
+struct
+{
+  __uint(type, BPF_MAP_TYPE_TASK_STORAGE);
+  __uint(map_flags, BPF_F_NO_PREALLOC);
+  __type(key, int);
+  __type(value, struct task_ctx);
+} task_ctx_stor SEC(".maps");
+
+#endif  // DATATYPES_H

scheds/experimental/scx_lunar/src/bpf/defines.h

@@ -0,0 +1,99 @@
+// SPDX-License-Identifier: GPL-2.0
+//
+// Author: Timon Stipkovits <timon2201@gmail.com>
+//
+// This software may be used and distributed according to the terms of the
+// GNU General Public License version 2.
+
+#ifndef DEFINES_H
+#define DEFINES_H
+
+#define SCHED_MODE_DSQ_PER_LLC 0
+#define SCHED_MODE_DSQ_PER_CPU 1
+
+#define NS_PER_US 1000ULL
+#define NS_PER_MS (1000ULL * NS_PER_US)
+
+#define NS_PER_MS_LL (s64) NS_PER_MS
+
+#define SLICE_LC (500 * NS_PER_US)
+#define SLICE_INTERACTIVE (500 * NS_PER_US)
+#define SLICE_NORMAL (500 * NS_PER_US)
+#define SLICE_BATCH (500 * NS_PER_US)
+#define SLICE_GREEDY (500 * NS_PER_US)
+
+#define RUNTIME_PRIO_BOUNDARY_LC (50 * NS_PER_US)
+#define RUNTIME_PRIO_BOUNDARY_INTERACTIVE (500 * NS_PER_US)
+#define RUNTIME_PRIO_BOUNDARY_NORMAL (2000 * NS_PER_US)
+#define RUNTIME_PRIO_BOUNDARY_BATCH (8000 * NS_PER_US)
+
+#define VLAG_PRIO_BOUNDARY_LC (12 * NS_PER_MS_LL)
+#define VLAG_PRIO_BOUNDARY_INTERACTIVE (8 * NS_PER_MS_LL)
+#define VLAG_PRIO_BOUNDARY_NORMAL (4 * NS_PER_MS_LL)
+#define VLAG_PRIO_BOUNDARY_BATCH (0 * NS_PER_MS_LL)
+
+#define MIN_RUN_BEFORE_PREEMPT 20000LL
+
+#define RUNTIME_THRESH_PERCENT 5LL
+#define VLAG_THRESH_PERCENT 5LL
+
+#define DUTY_WINDOW_NS (100ULL * NS_PER_MS)
+#define TASK_CLASSIFICATION_AGE_NS (50ULL * NS_PER_MS)
+#define DUTY_SAMPLES_NEEDED 10
+#define DUTY_SAMPLES_MAX 1000
+
+#define DUTY_RANGE 1024
+
+#define DUTY_EDGE_LC ((5 * DUTY_RANGE) / 100)
+#define DUTY_EDGE_INTERACTIVE ((20 * DUTY_RANGE) / 100)
+#define DUTY_EDGE_NORMAL ((40 * DUTY_RANGE) / 100)
+#define DUTY_EDGE_BATCH ((90 * DUTY_RANGE) / 100)
+
+#define DUTY_HYST_HIGH ((5 * DUTY_RANGE) / 100)
+#define DUTY_HYST ((3 * DUTY_RANGE) / 100)
+
+#define MAX_RT_PRIO 100
+
+#define MIN_AVG_RUNTIME 50ULL
+#define MAX_RUNTIME_PER_TASK (400 * NS_PER_MS)
+
+#define HISTORIC_TASK_SAMPLES 4
+#define AVG_RUNTIME_OVERRIDE_FACTOR 5
+
+#define DSQ_TYPE_LC 1
+#define DSQ_TYPE_INTERACTIVE 2
+#define DSQ_TYPE_NORMAL 3
+#define DSQ_TYPE_BATCH 4
+#define DSQ_TYPE_GREEDY 5
+#define DSQ_PRIO_QUEUE_AMOUNT 5
+#define DSQ_TYPE_EMPTY 6
+
+#define VLAG_PROMOTE_THRESH 2000000LL
+#define VLAG_DEMOTE_THRESH -2000000LL
+
+#define VLAG_MIN -14000000LL
+#define VLAG_MAX 14000000LL
+#define SLEEP_CREDIT_DIVISOR 1LL
+#define MAX_CREDITABLE_SLEEP 100000000LL
+
+#define DEFAULT_DSQ_LOCAL_ON 0xC000000000000000ULL
+
+#define DSQ_CPU_QUEUE_BASE_LC 1536
+#define DSQ_CPU_QUEUE_BASE_INTERACTIVE 2048
+#define DSQ_CPU_QUEUE_BASE_NORMAL 2560
+#define DSQ_CPU_QUEUE_BASE_BATCH 3072
+#define DSQ_CPU_QUEUE_BASE_GREEDY 3584
+
+#define DSQ_LLC_QUEUE_BASE_LC 4224
+#define DSQ_LLC_QUEUE_BASE_INTERACTIVE 4288
+#define DSQ_LLC_QUEUE_BASE_NORMAL 4352
+#define DSQ_LLC_QUEUE_BASE_BATCH 4416
+#define DSQ_LLC_QUEUE_BASE_GREEDY 4480
+
+#define QUEUE_START DSQ_TYPE_GREEDY
+
+#define AVG_RUNTIME_START (500 * NS_PER_US)
+
+#define MAX_CPUS 512
+
+#endif  // DEFINES_H