Bug #99999 | EXPLAIN FORMAT=TREE does not show cost/rows for semijoin materialization | ||
---|---|---|---|
Submitted: | 26 Jun 2020 7:58 | Modified: | 29 Jun 2020 10:46 |
Reporter: | Øystein Grøvlen | Email Updates: | |
Status: | Closed | Impact on me: | |
Category: | MySQL Server: Optimizer | Severity: | S3 (Non-critical) |
Version: | 8.0.20 | OS: | Any |
Assigned to: | CPU Architecture: | Any |
[26 Jun 2020 7:58]
Øystein Grøvlen
[26 Jun 2020 12:19]
MySQL Verification Team
Hi Mr. Oystein Grovlen, Thank you for your bug report. This is what I got from the EXPLAINs : -------------------------------------- EXPLAIN -> Nested loop inner join\n -> Table scan on <subquery2>\n -> Materialize with deduplication\n -> Table scan on t11 (cost=1.05 rows=8)\n -> Single-row index lookup on t1 using PRIMARY (a=`<subquery2>`.a) (cost=0.35 rows=1)\n EXPLAIN -> Nested loop inner join (cost=3.85 rows=8)\n -> Table scan on t11 (cost=1.05 rows=8)\n -> Single-row index lookup on t1 using PRIMARY (a=t11.a) (cost=0.26 rows=1)\n -------------------------------------- Hence, I repeated the behaviour that you reported. Verified as reported.
[29 Jun 2020 10:46]
Steinar Gunderson
Fixed as part of WL#14061: EXPLAIN -> Nested loop inner join (cost=2.10 rows=8) -> Table scan on <subquery2> (cost=1.60 rows=8) -> Materialize with deduplication (cost=1.05 rows=8) -> Table scan on t11 (cost=1.05 rows=8) -> Single-row index lookup on t1 using PRIMARY (a=`<subquery2>`.a) (cost=0.35 rows=1)
[29 Jun 2020 13:11]
MySQL Verification Team
Thank you, Steinar.