Commit6827e46

committed

Fix misleading comment in nodeIndexonlyscan.c.

The stated reason for acquiring predicate locks on heap pages hasn'texisted since commitc01262a, so fix the comment. Perhaps in a laterrelease we'll also be able to change the code to use tuple locks.Back-patch all the way.Reviewed-by: Ashwin AgrawalDiscussion:https://postgr.es/m/CAEepm%3D2GK3FVdnt5V3d%2Bh9njWipCv_fNL%3DwjxyUhzsF%3D0PcbNg%40mail.gmail.com

1 parent803cdf2 commit6827e46Copy full SHA for 6827e46

File tree

-5

lines changed

-5

lines changed

Lines changed: 2 additions & 5 deletions

Original file line number	Diff line number	Diff line change
`@@ -236,11 +236,8 @@ IndexOnlyNext(IndexOnlyScanState *node)`
`236`	`236`	`errmsg("lossy distance functions are not supported in index-only scans")));`
`237`	`237`
`238`	`238`	`/*`
`239`		`- * Predicate locks for index-only scans must be acquired at the page`
`240`		`- * level when the heap is not accessed, since tuple-level predicate`
`241`		`- * locks need the tuple's xmin value. If we had to visit the tuple`
`242`		`- * anyway, then we already have the tuple-level lock and can skip the`
`243`		`- * page lock.`
	`239`	`+ * If we didn't access the heap, then we'll need to take a predicate`
	`240`	`+ * lock explicitly, as if we had. For now we do that at page level.`
`244`	`241`	`*/`
`245`	`242`	`if (tuple==NULL)`
`246`	`243`	`PredicateLockPage(scandesc->heapRelation,`

Comments

(0)