NotificationsYou must be signed in to change notification settings
Fork5
Star26

Commitc89d507

committed

Round rowcount estimate for a partial path to an integer.

I'd been wondering why I was sometimes seeing fractional rowcountestimates in parallel-query situations, and this seems to be thereason. (You won't see the fractional parts in EXPLAIN, because itprints rowcounts with %.0f, but they are apparent in the debugger.)A fractional rowcount is not any saner for a partial path than anyother kind of path, and it's equally likely to break cost estimationfor higher paths, so apply clamp_row_est() like we do in other places.

1 parent3a4a33a commitc89d507Copy full SHA for c89d507

File tree

1 file changed

-1

lines changed

src/backend/optimizer/path
- costsize.c

1 file changed

-1

lines changed

`‎src/backend/optimizer/path/costsize.c`

Lines changed: 1 addition & 1 deletion

Original file line number	Diff line number	Diff line change
`@@ -263,7 +263,7 @@ cost_seqscan(Path path, PlannerInfo root,`
`263`	`263`	`* because they'll anticipate receiving more rows than any given copy`
`264`	`264`	`* will actually get.`
`265`	`265`	`*/`
`266`		`-path->rows/=parallel_divisor;`
	`266`	`+path->rows=clamp_row_est(path->rows /parallel_divisor);`
`267`	`267`
`268`	`268`	`/* The CPU cost is divided among all the workers. */`
`269`	`269`	`cpu_run_cost /=parallel_divisor;`

0 commit comments

Comments

(0)

Movatterモバイル変換

Navigation Menu

Search code, repositories, users, issues, pull requests...

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Uh oh!

Commitc89d507

File tree

1 file changed

1 file changed

`‎src/backend/optimizer/path/costsize.c`

0 commit comments