Kernel test robot reported that
tools/testing/selftests/kvm/hardware_disable_test was failing due to
commit
704069649b5b ("sched/core: Rework sched_class::wakeup_preempt()
and rq_modified_*()")
It turns out there were two related problems that could lead to a
missed preemption:
- when hitting newidle balance from the idle thread, it would elevate
rb->next_class from &idle_sched_class to &fair_sched_class, causing
later wakeup_preempt() calls to not hit the sched_class_above()
case, and not issue resched_curr().
Notably, this modification pattern should only lower the
next_class, and never raise it. Create two new helper functions to
wrap this.
- when doing schedule_idle(), it was possible to miss (re)setting
rq->next_class to &idle_sched_class, leading to the very same
problem.
Cc: Sean Christopherson <seanjc@google.com>
Fixes: 704069649b5b ("sched/core: Rework sched_class::wakeup_preempt() and rq_modified_*()")
Reported-by: kernel test robot <oliver.sang@intel.com>
Closes: https://lore.kernel.org/oe-lkp/202602122157.4e861298-lkp@intel.com
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://patch.msgid.link/20260218163329.GQ1395416@noisy.programming.kicks-ass.net
/* SCX must consult the BPF scheduler to tell if rq is empty */
if (!rq->nr_running && !scx_enabled()) {
next = prev;
+ rq->next_class = &idle_sched_class;
goto picked;
}
} else if (!preempt && prev_state) {
/* see kick_cpus_irq_workfn() */
smp_store_release(&rq->scx.kick_sync, rq->scx.kick_sync + 1);
- rq->next_class = &ext_sched_class;
+ rq_modified_begin(rq, &ext_sched_class);
rq_unpin_lock(rq, rf);
balance_one(rq, prev);
* If @force_scx is true, always try to pick a SCHED_EXT task,
* regardless of any higher-priority sched classes activity.
*/
- if (!force_scx && sched_class_above(rq->next_class, &ext_sched_class))
+ if (!force_scx && rq_modified_above(rq, &ext_sched_class))
return RETRY_TASK;
keep_prev = rq->scx.flags & SCX_RQ_BAL_KEEP;
t0 = sched_clock_cpu(this_cpu);
__sched_balance_update_blocked_averages(this_rq);
- this_rq->next_class = &fair_sched_class;
+ rq_modified_begin(this_rq, &fair_sched_class);
raw_spin_rq_unlock(this_rq);
for_each_domain(this_cpu, sd) {
pulled_task = 1;
/* If a higher prio class was modified, restart the pick */
- if (sched_class_above(this_rq->next_class, &fair_sched_class))
+ if (rq_modified_above(this_rq, &fair_sched_class))
pulled_task = -1;
out:
#define sched_class_above(_a, _b) ((_a) < (_b))
+static inline void rq_modified_begin(struct rq *rq, const struct sched_class *class)
+{
+ if (sched_class_above(rq->next_class, class))
+ rq->next_class = class;
+}
+
+static inline bool rq_modified_above(struct rq *rq, const struct sched_class *class)
+{
+ return sched_class_above(rq->next_class, class);
+}
+
static inline bool sched_stop_runnable(struct rq *rq)
{
return rq->stop && task_on_rq_queued(rq->stop);