Commit71932ec

committed

Add comment about memory reordering to PredicateLockTupleRowVersionLink.

Dan Ports, per head-scratching from Simon Riggs and myself.

1 parent5c5f835 commit71932ecCopy full SHA for 71932ec

File tree

-0

lines changed

-0

lines changed

Lines changed: 3 additions & 0 deletions

Original file line number	Diff line number	Diff line change
`@@ -2283,6 +2283,9 @@ PredicateLockTupleRowVersionLink(const Relation relation,`
`2283`	`2283`	`* locks. Even if a serializable transaction starts concurrently,`
`2284`	`2284`	`* we know it can't take any SIREAD locks on the modified tuple`
`2285`	`2285`	`* because the caller is holding the associated buffer page lock.`
	`2286`	`+ * Memory reordering isn't an issue; the memory barrier in the`
	`2287`	`+ * LWLock acquisition guarantees that this read occurs while the`
	`2288`	`+ * buffer page lock is held.`
`2286`	`2289`	`*/`
`2287`	`2290`	`if (!TransactionIdIsValid(PredXact->SxactGlobalXmin))`
`2288`	`2291`	`return;`

Comments

(0)