feat: P0-P6 kernel scheduler + relibc threading comprehensive implementation

P0-P2: Barrier SMP, sigmask/pthread_kill races, robust mutexes, RT scheduling, POSIX sched API P3: PerCpuSched struct, per-CPU wiring, work stealing, load balancing, initial placement P4: 64-shard futex table, REQUEUE, PI futexes (LOCK_PI/UNLOCK_PI/TRYLOCK_PI), robust futexes, vruntime tracking, min-vruntime SCHED_OTHER selection P5: setpriority/getpriority, pthread_setaffinity_np, pthread_setname_np, pthread_setschedparam (Redox) P6: Cache-affine scheduling (last_cpu + vruntime bonus), NUMA topology kernel hints + numad userspace daemon Stability fixes: make_consistent stores 0 (dead TID fix), cond.rs error propagation, SPIN_COUNT adaptive spinning, Sys::open &str fix, PI futex CAS race, proc.rs lock ordering, barrier destroy Patches: 33 kernel + 58 relibc patches, all tracked in recipes Docs: KERNEL-SCHEDULER-MULTITHREAD-IMPROVEMENT-PLAN.md updated, SCHEDULER-REVIEW-FINAL.md created Architecture: NUMA topology parsing stays userspace (numad daemon), kernel stores lightweight NumaTopology hints
2026-04-30 18:21:48 +01:00
parent 55d00c3a24
commit 34360e1e4f
70 changed files with 15268 additions and 10 deletions
@@ -0,0 +1,56 @@
+diff --git a/src/context/mod.rs b/src/context/mod.rs
+--- a/src/context/mod.rs
+++ b/src/context/mod.rs
+@@ -10,9 +10,9 @@ use core::{num::NonZeroUsize, ops::Deref};
+ 
+ use crate::{
+     context::memory::AddrSpaceWrapper,
+-    cpu_set::LogicalCpuSet,
+    cpu_set::{LogicalCpuId, LogicalCpuSet},
+     memory::{RmmA, RmmArch, TableKind},
+-    percpu::PercpuBlock,
+    percpu::{get_percpu_block, PercpuBlock},
+     sync::{
+         ArcRwLockWriteGuard, CleanLockToken, LockToken, Mutex, MutexGuard, RwLock, RwLockReadGuard,
+         RwLockWriteGuard, L0, L1, L2, L4,
+@@ -118,6 +118,30 @@ pub fn run_contexts(token: LockToken<'_, L0>) -> MutexGuard<'_, L1, RunContextDa
+     RUN_CONTEXTS.lock(token)
+ }
+ 
+fn least_loaded_cpu() -> LogicalCpuId {
+    let current_cpu = crate::cpu_id();
+    let mut best_cpu = current_cpu;
+    let mut best_depth = usize::MAX;
+
+    for raw_id in 0..crate::cpu_count() {
+        let cpu_id = LogicalCpuId::new(raw_id);
+        let Some(percpu) = get_percpu_block(cpu_id) else {
+            continue;
+        };
+
+        percpu.sched.take_lock();
+        let depth = unsafe { percpu.sched.queues().iter().map(|queue| queue.len()).sum() };
+        percpu.sched.release_lock();
+
+        if depth < best_depth {
+            best_depth = depth;
+            best_cpu = cpu_id;
+        }
+    }
+
+    best_cpu
+}
+
+ pub fn init(token: &mut CleanLockToken) {
+     let owner = None; // kmain not owned by any fd
+     let mut context = Context::new(owner).expect("failed to create kmain context");
+@@ -238,6 +262,9 @@ pub fn spawn(
+ 
+     context.kstack = Some(stack);
+     context.userspace = userspace_allowed;
+    let target_cpu = least_loaded_cpu();
+    context.sched_affinity = LogicalCpuSet::empty();
+    context.sched_affinity.atomic_set(target_cpu);
+ 
+     let context_lock = Arc::new(ContextLock::new(context));
+     let context_ref = ContextRef(Arc::clone(&context_lock));