Skip to contentSkip to conversation

sched_ext: Fix rq lock state in hotplug ops

PATCH Applied Andrea Righi 3 participants 3 messages Apr 28, 2025 – Apr 29, 2025 Scheduler

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 sched_ext: Fix rq lock state in hotplug ops

The ops.cpu_online() and ops.cpu_offline() callbacks incorrectly assume that the rq involved in the operation is locked, which is not the case during hotplug, triggering the following warning:

  WARNING: CPU: 1 PID: 20 at kernel/sched/sched.h:1504 handle_hotplug+0x280/0x340

Fix by not tracking the target rq as locked in the context of ops.cpu_online() and ops.cpu_offline().

Fixes 18853ba782bef ("sched_ext: Track currently locked rq")
---
 kernel/sched/ext.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
kernel/sched/ext.c +2 -2
index ac79067dc87e6..d0f40fd258752 100644
@@ -3477,9 +3477,9 @@ static void handle_hotplug(struct rq *rq, bool online)
 		scx_idle_update_selcpu_topology(&scx_ops);
 
 	if (online && SCX_HAS_OP(cpu_online))
-		SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_online, rq, cpu);
+		SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_online, NULL, cpu);
 	else if (!online && SCX_HAS_OP(cpu_offline))
-		SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_offline, rq, cpu);
+		SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_offline, NULL, cpu);
 	else
 		scx_ops_exit(SCX_ECODE_ACT_RESTART | SCX_ECODE_RSN_HOTPLUG,
 			     "cpu %d going %s, exiting scheduler", cpu,
Signature · 1 line
2.49.0
Tested-by on v1 1/1
#1 lore raw

Hi Andrea,

Thanks for the fix. Now it cleanly passes the hotplug selftest.

Quoted 32 lines · Andrea Righi

The ops.cpu_online() and ops.cpu_offline() callbacks incorrectly assume that the rq involved in the operation is locked, which is not the case during hotplug, triggering the following warning:

   WARNING: CPU: 1 PID: 20 at kernel/sched/sched.h:1504 handle_hotplug+0x280/0x340

Fix by not tracking the target rq as locked in the context of ops.cpu_online() and ops.cpu_offline().

Fixes 18853ba782bef ("sched_ext: Track currently locked rq")
---
  kernel/sched/ext.c | 4 ++--
  1 file changed, 2 insertions(+), 2 deletions(-)
kernel/sched/ext.c +2 -2
index ac79067dc87e6..d0f40fd258752 100644
@@ -3477,9 +3477,9 @@ static void handle_hotplug(struct rq *rq, bool online)
  		scx_idle_update_selcpu_topology(&scx_ops);
 
  	if (online && SCX_HAS_OP(cpu_online))
-		SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_online, rq, cpu);
+		SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_online, NULL, cpu);
  	else if (!online && SCX_HAS_OP(cpu_offline))
-		SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_offline, rq, cpu);
+		SCX_CALL_OP(SCX_KF_UNLOCKED, cpu_offline, NULL, cpu);
  	else
  		scx_ops_exit(SCX_ECODE_ACT_RESTART | SCX_ECODE_RSN_HOTPLUG,
  			     "cpu %d going %s, exiting scheduler", cpu,
Applied
#2 lore raw
In reply to Andrea Righi
Quoted 4 lines · Andrea Righi

The ops.cpu_online() and ops.cpu_offline() callbacks incorrectly assume that the rq involved in the operation is locked, which is not the case during hotplug, triggering the following warning:

Quoted 8 lines
  WARNING: CPU: 1 PID: 20 at kernel/sched/sched.h:1504 handle_hotplug+0x280/0x340

Fix by not tracking the target rq as locked in the context of ops.cpu_online() and ops.cpu_offline().

Fixes 18853ba782bef ("sched_ext: Track currently locked rq")

Applied to sched_ext/for-6.15-fixes.

Thanks.

Signature · 1 line
tejun