From fb5127ad6675b1cdbba0ae6e4fca29e13bf6047c Mon Sep 17 00:00:00 2001 From: Rongsong Shen Date: Thu, 30 Apr 2026 05:49:49 +0000 Subject: [PATCH] fix: skip GC in sched_end_session when live_jobs allocation fails (OOM safety) --- src/scheduler/src/scheduler.c | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/src/scheduler/src/scheduler.c b/src/scheduler/src/scheduler.c index 7214ca13..325f1911 100644 --- a/src/scheduler/src/scheduler.c +++ b/src/scheduler/src/scheduler.c @@ -822,6 +822,12 @@ void sched_end_session(xce_scheduler_t* sched) } } + if (live_jobs == NULL) { + /* OOM: cannot safely GC — skip to avoid sweeping nodes still in use */ + atomic_store_explicit(&sched->staging_merge_complete, 1, memory_order_release); + return; + } + /* Mark phase */ gc_clear_marks(); for (int i = 0; i < njobs; i++) {