Avoid detoasting failure after COMMIT inside a plpgsql FOR loop.

exec_for_query() normally tries to prefetch a few rows at a time
from the query being iterated over, so as to reduce executor
entry/exit overhead.  Unfortunately this is unsafe if we have
COMMIT or ROLLBACK within the loop, because there might be
TOAST references in the data that we prefetched but haven't
yet examined.  Immediately after the COMMIT/ROLLBACK, we have
no snapshots in the session, meaning that VACUUM is at liberty
to remove recently-deleted TOAST rows.

This was originally reported as a case triggering the "no known
snapshots" error in init_toast_snapshot(), but even if you miss
hitting that, you can get "missing toast chunk", as illustrated
by the added isolation test case.

To fix, just disable prefetching in non-atomic contexts.  Maybe
there will be performance complaints prompting us to work harder
later, but it's not clear at the moment that this really costs
much, and I doubt we'd want to back-patch any complicated fix.

In passing, adjust that error message in init_toast_snapshot()
to be a little clearer about the likely cause of the problem.

Patch by me, based on earlier investigation by Konstantin Knizhnik.

Per bug #15990 from Andreas Wicht.  Back-patch to v11 where
intra-procedure COMMIT was added.

Discussion: https://postgr.es/m/15990-eee2ac466b11293d@postgresql.org

Author: tglsfdc

src/backend/access/heap/tuptoaster.c

@@ -2425,8 +2425,21 @@ init_toast_snapshot(Snapshot toast_snapshot)
 {
 	Snapshot	snapshot = GetOldestSnapshot();
 
+	/*
+	 * GetOldestSnapshot returns NULL if the session has no active snapshots.
+	 * We can get that if, for example, a procedure fetches a toasted value
+	 * into a local variable, commits, and then tries to detoast the value.
+	 * Such coding is unsafe, because once we commit there is nothing to
+	 * prevent the toast data from being deleted.  Detoasting *must* happen in
+	 * the same transaction that originally fetched the toast pointer.  Hence,
+	 * rather than trying to band-aid over the problem, throw an error.  (This
+	 * is not very much protection, because in many scenarios the procedure
+	 * would have already created a new transaction snapshot, preventing us
+	 * from detecting the problem.  But it's better than nothing, and for sure
+	 * we shouldn't expend code on masking the problem more.)
+	 */
 	if (snapshot == NULL)
-		elog(ERROR, "no known snapshots");
+		elog(ERROR, "cannot fetch toast data without an active snapshot");
 
 	InitToastSnapshot(*toast_snapshot, snapshot->lsn, snapshot->whenTaken);
 }

src/pl/plpgsql/src/pl_exec.c

@@ -5981,6 +5981,17 @@ exec_for_query(PLpgSQL_execstate *estate, PLpgSQL_stmt_forq *stmt,
 	 */
 	PinPortal(portal);
 
+	/*
+	 * In a non-atomic context, we dare not prefetch, even if it would
+	 * otherwise be safe.  Aside from any semantic hazards that that might
+	 * create, if we prefetch toasted data and then the user commits the
+	 * transaction, the toast references could turn into dangling pointers.
+	 * (Rows we haven't yet fetched from the cursor are safe, because the
+	 * PersistHoldablePortal mechanism handles this scenario.)
+	 */
+	if (!estate->atomic)
+		prefetch_ok = false;
+
 	/*
 	 * Fetch the initial tuple(s).  If prefetching is allowed then we grab a
 	 * few more rows to avoid multiple trips through executor startup

src/test/isolation/expected/plpgsql-toast.out

@@ -192,3 +192,46 @@ pg_advisory_unlock
 t              
 s1: NOTICE:  length(r) = 6002
 step assign5: <... completed>
+
+starting permutation: lock assign6 vacuum unlock
+pg_advisory_unlock_all
+
+               
+pg_advisory_unlock_all
+
+               
+step lock: 
+    SELECT pg_advisory_lock(1);
+
+pg_advisory_lock
+
+               
+step assign6: 
+do $$
+  declare
+    r record;
+  begin
+    insert into test1 values (2, repeat('bar', 3000));
+    insert into test1 values (3, repeat('baz', 4000));
+    for r in select test1.b from test1 loop
+      delete from test1;
+      commit;
+      perform pg_advisory_lock(1);
+      raise notice 'length(r) = %', length(r::text);
+    end loop;
+  end;
+$$;
+ <waiting ...>
+step vacuum: 
+    VACUUM test1;
+
+step unlock: 
+    SELECT pg_advisory_unlock(1);
+
+pg_advisory_unlock
+
+t              
+s1: NOTICE:  length(r) = 6002
+s1: NOTICE:  length(r) = 9002
+s1: NOTICE:  length(r) = 12002
+step assign6: <... completed>

src/test/isolation/specs/plpgsql-toast.spec

@@ -112,6 +112,25 @@ do $$
 $$;
 }
 
+# FOR loop must not hold any fetched-but-not-detoasted values across commit
+step "assign6"
+{
+do $$
+  declare
+    r record;
+  begin
+    insert into test1 values (2, repeat('bar', 3000));
+    insert into test1 values (3, repeat('baz', 4000));
+    for r in select test1.b from test1 loop
+      delete from test1;
+      commit;
+      perform pg_advisory_lock(1);
+      raise notice 'length(r) = %', length(r::text);
+    end loop;
+  end;
+$$;
+}
+
 session "s2"
 setup
 {
@@ -135,3 +154,4 @@ permutation "lock" "assign2" "vacuum" "unlock"
 permutation "lock" "assign3" "vacuum" "unlock"
 permutation "lock" "assign4" "vacuum" "unlock"
 permutation "lock" "assign5" "vacuum" "unlock"
+permutation "lock" "assign6" "vacuum" "unlock"