sched/ext: Fix invalid task state transitions on class switch
Subsystems (2)
Andrea Righi
reviewerPATCH 1/1 sched/ext: Fix invalid task state transitions on class switch
When enabling a sched_ext scheduler, we may trigger invalid task state transitions, resulting in warnings like the following (which can be easily reproduced by running the hotplug selftest in a loop):
sched_ext: Invalid task state transition 0 -> 3 for fish[770]
WARNING: CPU: 18 PID: 787 at kernel/sched/ext.c:3862 scx_set_task_state+0x7c/0xc0 ... RIP: 0010:scx_set_task_state+0x7c/0xc0 ... Call Trace: <TASK> scx_enable_task+0x11f/0x2e0 switching_to_scx+0x24/0x110 scx_enable.isra.0+0xd14/0x13d0 bpf_struct_ops_link_create+0x136/0x1a0 __sys_bpf+0x1edd/0x2c30 __x64_sys_bpf+0x21/0x30 do_syscall_64+0xbb/0x370 entry_SYSCALL_64_after_hwframe+0x77/0x7f
This happens because we skip initialization for tasks that are already dead (with their usage counter set to zero), but we don't exclude them during the scheduling class transition phase.
Fix this by also skipping dead tasks during class swiching, preventing invalid task state transitions.
--- kernel/sched/ext.c | 4 ++++ 1 file changed, 4 insertions(+)
kernel/sched/ext.c +4 -0
@@ -5749,6 +5749,9 @@ static int scx_enable(struct sched_ext_ops *ops, struct bpf_link *link)__setscheduler_class (p->policy, p->prio);struct sched_enq_and_set_ctx ctx; +if (!tryget_task_struct (p)) +continue ; +if (old_class != new_class && p->se.sched_delayed)dequeue_task (task_rq (p), p,DEQUEUE_SLEEP |DEQUEUE_DELAYED );
@@ -5761,6 +5764,7 @@ static int scx_enable(struct sched_ext_ops *ops, struct bpf_link *link)sched_enq_and_set_task (&ctx);check_class_changed (task_rq (p), p, old_class, p->prio); +put_task_struct (p); }scx_task_iter_stop (&sti);percpu_up_write (&scx_fork_rwsem);
Signature · 1 line
2.50.1
Quoted 24 lines · Andrea Righi
When enabling a
sched_extscheduler, we may trigger invalid task state transitions, resulting in warnings like the following (which can be easily reproduced by running the hotplug selftest in a loop):
sched_ext: Invalid task state transition 0 -> 3 for fish[770]WARNING: CPU: 18 PID: 787 at kernel/sched/ext.c:3862 scx_set_task_state+0x7c/0xc0 ... RIP: 0010:scx_set_task_state+0x7c/0xc0 ... Call Trace: <TASK> scx_enable_task+0x11f/0x2e0 switching_to_scx+0x24/0x110 scx_enable.isra.0+0xd14/0x13d0 bpf_struct_ops_link_create+0x136/0x1a0 __sys_bpf+0x1edd/0x2c30 __x64_sys_bpf+0x21/0x30 do_syscall_64+0xbb/0x370 entry_SYSCALL_64_after_hwframe+0x77/0x7fThis happens because we skip initialization for tasks that are already dead (with their usage counter set to zero), but we don't exclude them during the scheduling class transition phase.
Quoted 5 lines
Fix this by also skipping dead tasks during class swiching, preventing invalid task state transitions.
Fixes a8532fac7b5d2 ("sched_ext: TASK_DEAD tasks must be switched into SCX on ops_enable")Signed-off-by Andrea Righi <[email protected]>
Applied to sched_ext/for-6.17-fixes.
Thanks.
Signature · 1 line
tejun