Avoid testing tuple visibility without buffer lock.

tglsfdc · tglsfdc · commit fdcee9f1fc12 · 2016-10-23T19:14:32.000-04:00
INSERT ... ON CONFLICT (specifically ExecCheckHeapTupleVisible) contains another example of this unsafe coding practice. It is much harder to get a failure out of it than the case fixed in commit 6292c23, because in most scenarios any hint bits that could be set would have already been set earlier in the command. However, Konstantin Knizhnik reported a failure with a custom transaction manager, and it's clearly possible to get a failure via a race condition in async-commit mode. For lack of a reproducible example, no regression test case in this commit. I did some testing with Asserts added to tqual.c's functions, and can say that running "make check-world" exposed these two bugs and no others. The Asserts are messy enough that I've not added them to the code for now. Report: <57EE93C8.8080504@postgrespro.ru> Related-Discussion: <CAO3NbwOycQjt2Oqy2VW-eLTq2M5uGMyHnGm=RNga4mjqcYD7gQ@mail.gmail.com>
diff --git a/src/backend/executor/nodeModifyTable.c b/src/backend/executor/nodeModifyTable.c
@@ -194,6 +194,11 @@ ExecCheckHeapTupleVisible(EState *estate,
 	if (!IsolationUsesXactSnapshot())
 		return;
 
+	/*
+	 * We need buffer pin and lock to call HeapTupleSatisfiesVisibility.
+	 * Caller should be holding pin, but not lock.
+	 */
+	LockBuffer(buffer, BUFFER_LOCK_SHARE);
 	if (!HeapTupleSatisfiesVisibility(tuple, estate->es_snapshot, buffer))
 	{
 		/*
@@ -207,6 +212,7 @@ ExecCheckHeapTupleVisible(EState *estate,
 					(errcode(ERRCODE_T_R_SERIALIZATION_FAILURE),
 			 errmsg("could not serialize access due to concurrent update")));
 	}
+	LockBuffer(buffer, BUFFER_LOCK_UNLOCK);
 }
 
 /*

Original file line number	Diff line number	Diff line change
`@@ -194,6 +194,11 @@ ExecCheckHeapTupleVisible(EState *estate,`
`194`	`194`	`if (!IsolationUsesXactSnapshot())`
`195`	`195`	`return;`
`196`	`196`
	`197`	`+ /*`
	`198`	`+ * We need buffer pin and lock to call HeapTupleSatisfiesVisibility.`
	`199`	`+ * Caller should be holding pin, but not lock.`
	`200`	`+ */`
	`201`	`+ LockBuffer(buffer, BUFFER_LOCK_SHARE);`
`197`	`202`	`if (!HeapTupleSatisfiesVisibility(tuple, estate->es_snapshot, buffer))`
`198`	`203`	`{`
`199`	`204`	`/*`
`@@ -207,6 +212,7 @@ ExecCheckHeapTupleVisible(EState *estate,`
`207`	`212`	`(errcode(ERRCODE_T_R_SERIALIZATION_FAILURE),`
`208`	`213`	`errmsg("could not serialize access due to concurrent update")));`
`209`	`214`	`}`
	`215`	`+ LockBuffer(buffer, BUFFER_LOCK_UNLOCK);`
`210`	`216`	`}`
`211`	`217`
`212`	`218`	`/*`