sched_ext: Use SCX_TASK_READY test instead of tryget_task_struct() during class switch

ddf7233fcab6 ("sched/ext: Fix invalid task state transitions on class
switch") added tryget_task_struct() test during scx_enable()'s class
switching loop. The reason for the addition was to avoid enabling tasks which
skipped prep in the previous loop due to being dead.

While tryget_task_struct() does work for this purpose as tasks that fail
tryget always will fail it, it's a bit roundabout. A more direct way is
testing whether the task is in READY state. Switch to testing SCX_TASK_READY
directly.

Cc: Andrea Righi <arighi@nvidia.com>
Acked-by: Andrea Righi <arighi@nvidia.com>
Signed-off-by: Tejun Heo <tj@kernel.org>

diff --git a/kernel/sched/ext.c b/kernel/sched/ext.c
index 1cfe4b43d31b0dab78e68f96c13926ed4727c56f..603de983a509cd87607ad532682a10f9d339853f 100644 (file)
--- a/kernel/sched/ext.c
+++ b/kernel/sched/ext.c
@@ -4753,7 +4753,7 @@ static int scx_enable(struct sched_ext_ops *ops, struct bpf_link *link)
                const struct sched_class *new_class =
                        __setscheduler_class(p->policy, p->prio);
 
-               if (!tryget_task_struct(p))
+               if (scx_get_task_state(p) != SCX_TASK_READY)
                        continue;
 
                if (old_class != new_class)
@@ -4763,8 +4763,6 @@ static int scx_enable(struct sched_ext_ops *ops, struct bpf_link *link)
                        p->scx.slice = SCX_SLICE_DFL;
                        p->sched_class = new_class;
                }
-
-               put_task_struct(p);
        }
        scx_task_iter_stop(&sti);
        percpu_up_write(&scx_fork_rwsem);