(nuttx) 01/05: sched: Remove race condition in sched_unlock

[jiuzhudong]

This is an automated email from the ASF dual-hosted git repository.

jiuzhudong pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/nuttx.git

commit f55e6f63838a7b2381ebeeb4d9a9077f09890e2e
Author: Jukka Laitinen <jukka.laiti...@tii.ae>
AuthorDate: Mon Jul 21 22:46:42 2025 +0300

    sched: Remove race condition in sched_unlock
    
    When exiting schedlock, that task should first take the critical section and
    only after that decrease the lockcount to 0. Otherwise an interrupt might
    cause a re-schedule before the task enters the critical section, which makes
    the following code meaningless.
    
    Signed-off-by: Jukka Laitinen <jukka.laiti...@tii.ae>
---
 sched/sched/sched_unlock.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

diff --git a/sched/sched/sched_unlock.c b/sched/sched/sched_unlock.c
index 4c7d6814542..dabd07cbc80 100644
--- a/sched/sched/sched_unlock.c
+++ b/sched/sched/sched_unlock.c
@@ -69,10 +69,12 @@ void sched_unlock(void)
        * then pre-emption has been re-enabled.
        */
 
-      if (rtcb != NULL && --rtcb->lockcount == 0)
+      if (rtcb != NULL && rtcb->lockcount == 1)
         {
           irqstate_t flags = enter_critical_section_wo_note();
 
+          rtcb->lockcount = 0;
+
           /* Note that we no longer have pre-emption disabled. */
 
           nxsched_critmon_preemption(rtcb, false, return_address(0));
@@ -164,5 +166,9 @@ void sched_unlock(void)
 
           leave_critical_section_wo_note(flags);
         }
+      else
+        {
+          rtcb->lockcount--;
+        }
     }
 }