Remote consumption now reaches the !dsq branch with holding_cpu set,
just like dispatch_to_local_dsq(). Clearing holding_cpu in this branch
also tells dispatch_to_local_dsq() that it lost to a dequeue. The
dsq-present branch races with unlink_dsq_and_switch_rq_lock(), not
dispatch_to_local_dsq().

Update both comments to describe the current transfer paths and race
semantics. No functional change.

Suggested-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Usama Arif <usama.arif@linux.dev>
---
v1 -> v2: (Tejun)
- Name dispatch_to_local_dsq() in the comments and commit description.
- Explain that clearing holding_cpu tells dispatch_to_local_dsq() that it
  lost to a dequeue.
- Rewrap both comments to 80 columns.
---
 kernel/sched/ext/ext.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

diff --git a/kernel/sched/ext/ext.c b/kernel/sched/ext/ext.c
index 911bb6d433ba7..d01f44839da32 100644
--- a/kernel/sched/ext/ext.c
+++ b/kernel/sched/ext/ext.c
@@ -1821,10 +1821,10 @@ void scx_dispatch_dequeue(struct rq *rq, struct task_struct *p)
 			list_del_init(&p->scx.dsq_list.node);
 
 		/*
-		 * When dispatching directly from the BPF scheduler to a local
-		 * DSQ, the task isn't associated with any DSQ but
-		 * @p->scx.holding_cpu may be set under the protection of
-		 * %SCX_OPSS_DISPATCHING.
+		 * When dispatch_to_local_dsq() or remote consumption moves a
+		 * task to a local DSQ, the task isn't associated with any DSQ
+		 * but @p->scx.holding_cpu may be set. Clearing holding_cpu
+		 * tells dispatch_to_local_dsq() that it lost to a dequeue.
 		 */
 		if (p->scx.holding_cpu >= 0)
 			p->scx.holding_cpu = -1;
@@ -1844,10 +1844,10 @@ void scx_dispatch_dequeue(struct rq *rq, struct task_struct *p)
 		scx_task_unlink_from_dsq(p, dsq);
 	} else {
 		/*
-		 * We're racing against dispatch_to_local_dsq() which already
-		 * removed @p from @dsq and set @p->scx.holding_cpu. Clear the
-		 * holding_cpu which tells dispatch_to_local_dsq() that it lost
-		 * the race.
+		 * We're racing against unlink_dsq_and_switch_rq_lock(),
+		 * which already removed @p from @dsq and set
+		 * @p->scx.holding_cpu. Clear holding_cpu to tell
+		 * unlink_dsq_and_switch_rq_lock() that it lost the race.
 		 */
 		WARN_ON_ONCE(!list_empty(&p->scx.dsq_list.node));
 		p->scx.holding_cpu = -1;
-- 
2.53.0-Meta