planner: avoid double scan for index prefix col is (not) null

[xuyifangreeneyes]

What problem does this PR solve?

Issue Number: ref #21145

Problem Summary:

CREATE TABLE t1 (
  id char(1) DEFAULT NULL,
  c1 varchar(255) DEFAULT NULL,
  c2 text DEFAULT NULL,
  KEY idx1 (c1),
  KEY idx2 (c1,c2(5))
) ENGINE=InnoDB DEFAULT CHARSET=utf8mb4 COLLATE=utf8mb4_bin;

select count(1) from t1 where c1 = '0xfff' and c2 is not null;

We want to avoid double scan when using index idx2 for the query.

What is changed and how it works?

1. Introduce a system variable tidb_opt_prefix_index_single_scan. If the switch is on, do the following two optimizations.
2. When checking whether an access condition needs to be reserved to filter conditions, IsNull doesn't need to be reserved even if the index column is prefix column.
3. When deciding path.IsSingleScan, the old way is to check whether ds.schema.Columns is subset of index columns which are not prefix column. While the new way is to check two things:
   a. whether ds.colsRequiringFullLen is subset of index columns which are not prefix column.
   b. whether ds.allConds is covered by the index.

Check List

Tests

• Unit test
• Integration test
• Manual test (add detailed scripts or steps below)
• No code

Side effects

• Performance regression: Consumes more CPU
• Performance regression: Consumes more Memory
• Breaking backward compatibility

Documentation

• Affects user behaviors
• Contains syntax changes
• Contains variable changes
• Contains experimental features
• Changes MySQL compatibility

Release note

Please refer to Release Notes Language Style Guide to write a quality release note.

None

[ti-chi-bot]

[REVIEW NOTIFICATION]

This pull request has been approved by:

• Yisaer
• time-and-fate

To complete the pull request process, please ask the reviewers in the list to review by filling /cc @reviewer in the comment.
After your PR has acquired the required number of LGTMs, you can assign this pull request to the committer in the list by filling /assign @committer in the comment to help you merge this pull request.

The full list of commands accepted by this bot can be found here.

Reviewer can indicate their review by submitting an approval review.
Reviewer can cancel approval by submitting a request changes review.

[xuyifangreeneyes]

/run-mysql-test tidb-test=pr/1999

[xuyifangreeneyes]

/run-mysql-test tidb-test=pr/1999

[xuyifangreeneyes]

/hold

[xuyifangreeneyes]

/unhold

[xuyifangreeneyes]

/merge

[ti-chi-bot]

This pull request has been accepted and is ready to merge.

Commit hash: 42778ff

[xuyifangreeneyes]

/hold

[xuyifangreeneyes]

/unhold

[xuyifangreeneyes]

/merge

[ti-chi-bot]

This pull request has been accepted and is ready to merge.

Commit hash: 5ecc79e

[xuyifangreeneyes]

/run-mysql-test tidb-test=pr/1999

[sre-bot]

TiDB MergeCI notify

🔴 Bad News! New failing [3] after this pr merged.
These new failed integration tests seem to be caused by the current PR, please try to fix these new failed integration tests, thanks!

CI Name	Result	Duration	Compare with Parent commit
idc-jenkins-ci-tidb/integration-ddl-test	🟥 failed 1, success 5, total 6	23 min	New failing
idc-jenkins-ci-tidb/common-test	🟥 failed 1, success 10, total 11	9 min 12 sec	New failing
idc-jenkins-ci-tidb/integration-common-test	🟥 failed 4, success 13, total 17	8 min 52 sec	New failing
idc-jenkins-ci/integration-cdc-test	🟢 all 38 tests passed	25 min	Existing passed
idc-jenkins-ci-tidb/tics-test	🟢 all 1 tests passed	6 min 59 sec	Existing passed
idc-jenkins-ci-tidb/sqllogic-test-1	🟢 all 26 tests passed	4 min 24 sec	Existing passed
idc-jenkins-ci-tidb/sqllogic-test-2	🟢 all 28 tests passed	4 min 5 sec	Existing passed
idc-jenkins-ci-tidb/integration-compatibility-test	🟢 all 1 tests passed	2 min 53 sec	Existing passed
idc-jenkins-ci-tidb/mybatis-test	🟢 all 1 tests passed	2 min 44 sec	Existing passed
idc-jenkins-ci-tidb/plugin-test	🟢 build success, plugin test success	4min	Existing passed

[AilinKid]

just curious, should this be theoretical possible
a(2) != xxx
[-min, xxx(prefix)0000 ), [xx(x+1)(prefix)FFFF, +max]

[xuyifangreeneyes]

It is theoretical possible for some types and collations. But there may be some problems like #21145 (comment) for some string types and collations.

[AilinKid]

Rest LGTM