crimson/osd/pg: fix dangerous lambda captures

In PG::submit_transaction(), osd_op_p was first moved to ReplicatedBackend::submit_transaction(), and later its at_version is captured in another lambda. The latter may be already empty at the time of the capture. Signed-off-by: Xuehan Xu <[email protected]>
cybozu · Nov 12, 2024 · cbd22c7 · cbd22c7
1 parent 1c93566
commit cbd22c7
Showing 1 changed file with 4 additions and 3 deletions.
diff --git a/src/crimson/osd/pg.cc b/src/crimson/osd/pg.cc
@@ -924,8 +924,9 @@ PG::submit_transaction(
   }
 
   epoch_t map_epoch = get_osdmap_epoch();
+  auto at_version = osd_op_p.at_version;
 
-  peering_state.pre_submit_op(obc->obs.oi.soid, log_entries, osd_op_p.at_version);
+  peering_state.pre_submit_op(obc->obs.oi.soid, log_entries, at_version);
   peering_state.update_trim_to();
 
   ceph_assert(!log_entries.empty());
@@ -947,8 +948,8 @@ PG::submit_transaction(
   co_return std::make_tuple(
     std::move(submitted),
     all_completed.then_interruptible(
-      [this, last_complete=peering_state.get_info().last_complete,
-      at_version=osd_op_p.at_version](auto acked) {
+      [this, at_version,
+      last_complete=peering_state.get_info().last_complete](auto acked) {
       for (const auto& peer : acked) {
         peering_state.update_peer_last_complete_ondisk(
           peer.shard, peer.last_complete_ondisk);