kernel.git/kernel/sched/ext.c, branch linux-rolling-stable

sched_ext: Use WRITE_ONCE() for the write side of scx_enable helper pointer

2026-03-19T15:15:33Z

commit 2fcfe5951eb2e8440fc5e1dd6ea977336ff83a1d upstream. scx_enable() uses double-checked locking to lazily initialize a static kthread_worker pointer. The fast path reads helper locklessly: if (!READ_ONCE(helper)) { // lockless read -- no helper_mutex The write side initializes helper under helper_mutex, but previously used a plain assignment: helper = kthread_run_worker(0, "scx_enable_helper"); ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ plain write -- KCSAN data race with READ_ONCE() above Since READ_ONCE() on the fast path and the plain write on the initialization path access the same variable without a common lock, they constitute a data race. KCSAN requires that all sides of a lock-free access use READ_ONCE()/WRITE_ONCE() consistently. Use a temporary variable to stage the result of kthread_run_worker(), and only WRITE_ONCE() into helper after confirming the pointer is valid. This avoids a window where a concurrent caller on the fast path could observe an ERR pointer via READ_ONCE(helper) before the error check completes. Fixes: b06ccbabe250 ("sched_ext: Fix starvation of scx_enable() under fair-class saturation") Signed-off-by: zhidao su Acked-by: Andrea Righi Signed-off-by: Tejun Heo Signed-off-by: Greg Kroah-Hartman

sched_ext: Fix enqueue_task_scx() truncation of upper enqueue flags

2026-03-19T15:15:26Z

commit 57ccf5ccdc56954f2a91a7f66684fd31c566bde5 upstream. enqueue_task_scx() takes int enq_flags from the sched_class interface. SCX enqueue flags starting at bit 32 (SCX_ENQ_PREEMPT and above) are silently truncated when passed through activate_task(). extra_enq_flags was added as a workaround - storing high bits in rq->scx.extra_enq_flags and OR-ing them back in enqueue_task_scx(). However, the OR target is still the int parameter, so the high bits are lost anyway. The current impact is limited as the only affected flag is SCX_ENQ_PREEMPT which is informational to the BPF scheduler - its loss means the scheduler doesn't know about preemption but doesn't cause incorrect behavior. Fix by renaming the int parameter to core_enq_flags and introducing a u64 enq_flags local that merges both sources. All downstream functions already take u64 enq_flags. Fixes: f0e1a0643a59 ("sched_ext: Implement BPF extensible scheduler class") Cc: stable@vger.kernel.org # v6.12+ Acked-by: Andrea Righi Signed-off-by: Tejun Heo Signed-off-by: Greg Kroah-Hartman

sched_ext: Fix starvation of scx_enable() under fair-class saturation

2026-03-19T15:15:15Z

commit b06ccbabe2506fd70b9167a644978b049150224a upstream. During scx_enable(), the READY -> ENABLED task switching loop changes the calling thread's sched_class from fair to ext. Since fair has higher priority than ext, saturating fair-class workloads can indefinitely starve the enable thread, hanging the system. This was introduced when the enable path switched from preempt_disable() to scx_bypass() which doesn't protect against fair-class starvation. Note that the original preempt_disable() protection wasn't complete either - in partial switch modes, the calling thread could still be starved after preempt_enable() as it may have been switched to ext class. Fix it by offloading the enable body to a dedicated system-wide RT (SCHED_FIFO) kthread which cannot be starved by either fair or ext class tasks. scx_enable() lazily creates the kthread on first use and passes the ops pointer through a struct scx_enable_cmd containing the kthread_work, then synchronously waits for completion. The workfn runs on a different kthread from sch->helper (which runs disable_work), so it can safely flush disable_work on the error path without deadlock. Fixes: 8c2090c504e9 ("sched_ext: Initialize in bypass mode") Cc: stable@vger.kernel.org # v6.12+ Signed-off-by: Tejun Heo Signed-off-by: Greg Kroah-Hartman

sched_ext: Disable preemption between scx_claim_exit() and kicking helper work

2026-03-19T15:15:15Z

commit 83236b2e43dba00bee5b82eb5758816b1a674f6a upstream. scx_claim_exit() atomically sets exit_kind, which prevents scx_error() from triggering further error handling. After claiming exit, the caller must kick the helper kthread work which initiates bypass mode and teardown. If the calling task gets preempted between claiming exit and kicking the helper work, and the BPF scheduler fails to schedule it back (since error handling is now disabled), the helper work is never queued, bypass mode never activates, tasks stop being dispatched, and the system wedges. Disable preemption across scx_claim_exit() and the subsequent work kicking in all callers - scx_disable() and scx_vexit(). Add lockdep_assert_preemption_disabled() to scx_claim_exit() to enforce the requirement. Fixes: f0e1a0643a59 ("sched_ext: Implement BPF extensible scheduler class") Cc: stable@vger.kernel.org # v6.12+ Reviewed-by: Andrea Righi Signed-off-by: Tejun Heo Signed-off-by: Greg Kroah-Hartman

sched_ext: Remove redundant css_put() in scx_cgroup_init()

2026-03-19T15:14:59Z

commit 1336b579f6079fb8520be03624fcd9ba443c930b upstream. The iterator css_for_each_descendant_pre() walks the cgroup hierarchy under cgroup_lock(). It does not increment the reference counts on yielded css structs. According to the cgroup documentation, css_put() should only be used to release a reference obtained via css_get() or css_tryget_online(). Since the iterator does not use either of these to acquire a reference, calling css_put() in the error path of scx_cgroup_init() causes a refcount underflow. Remove the unbalanced css_put() to prevent a potential Use-After-Free (UAF) vulnerability. Fixes: 819513666966 ("sched_ext: Add cgroup support") Cc: stable@vger.kernel.org # v6.12+ Signed-off-by: Cheng-Yang Chou Reviewed-by: Andrea Righi Signed-off-by: Tejun Heo Signed-off-by: Greg Kroah-Hartman

Merge tag 'sched_ext-for-6.19-rc8-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext

2026-02-04T23:11:24Z

Pull sched_ext fix from Tejun Heo: - Fix race where sched_class operations (sched_setscheduler() and friends) could be invoked on dead tasks after sched_ext_dead() already ran, causing invalid SCX task state transitions and NULL pointer dereferences. This was a regression from the cgroup exit ordering fix which moved sched_ext_free() to finish_task_switch(). * tag 'sched_ext-for-6.19-rc8-fixes' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/sched_ext: sched_ext: Short-circuit sched_class operations on dead tasks

sched_ext: Short-circuit sched_class operations on dead tasks

2026-02-04T22:22:11Z

7900aa699c34 ("sched_ext: Fix cgroup exit ordering by moving sched_ext_free() to finish_task_switch()") moved sched_ext_free() to finish_task_switch() and renamed it to sched_ext_dead() to fix cgroup exit ordering issues. However, this created a race window where certain sched_class ops may be invoked on dead tasks leading to failures - e.g. sched_setscheduler() may try to switch a task which finished sched_ext_dead() back into SCX triggering invalid SCX task state transitions. Add task_dead_and_done() which tests whether a task is TASK_DEAD and has completed its final context switch, and use it to short-circuit sched_class operations which may be called on dead tasks. Fixes: 7900aa699c34 ("sched_ext: Fix cgroup exit ordering by moving sched_ext_free() to finish_task_switch()") Reported-by: Andrea Righi Link: http://lkml.kernel.org/r/20260202151341.796959-1-arighi@nvidia.com Reviewed-by: Andrea Righi Signed-off-by: Tejun Heo

sched: Audit MOVE vs balance_callbacks

2026-01-15T20:57:53Z

The {DE,EN}QUEUE_MOVE flag indicates a task is allowed to change priority, which means there could be balance callbacks queued. Therefore audit all MOVE users and make sure they do run balance callbacks before dropping rq-lock. Fixes: 6455ad5346c9 ("sched: Move sched_class::prio_changed() into the change pattern") Signed-off-by: Peter Zijlstra (Intel) Tested-by: Pierre Gondois Tested-by: Juri Lelli Link: https://patch.msgid.link/20260114130528.GB831285@noisy.programming.kicks-ass.net

sched_ext: Use the resched_cpu() to replace resched_curr() in the bypass_lb_node()

2025-12-23T03:51:51Z

For the PREEMPT_RT kernels, the scx_bypass_lb_timerfn() running in the preemptible per-CPU ktimer kthread context, this means that the following scenarios will occur(for x86 platform): cpu1 cpu2 ktimer kthread: ->scx_bypass_lb_timerfn ->bypass_lb_node ->for_each_cpu(cpu, resched_mask) migration/1: by preempt by migration/2: multi_cpu_stop() multi_cpu_stop() ->take_cpu_down() ->__cpu_disable() ->set cpu1 offline ->rq1 = cpu_rq(cpu1) ->resched_curr(rq1) ->smp_send_reschedule(cpu1) ->native_smp_send_reschedule(cpu1) ->if(unlikely(cpu_is_offline(cpu))) { WARN(1, "sched: Unexpected reschedule of offline CPU#%d!\n", cpu); return; } This commit therefore use the resched_cpu() to replace resched_curr() in the bypass_lb_node() to avoid send-ipi to offline CPUs. Signed-off-by: Zqiang Reviewed-by: Andrea Righi Signed-off-by: Tejun Heo

sched_ext: Fix some comments in ext.c

2025-12-19T23:11:22Z

This commit update balance_scx() in the comments to balance_one(). Signed-off-by: Zqiang Reviewed-by: Andrea Righi Signed-off-by: Tejun Heo