planner,util/ranger: reduce mem allocs on rebuildRange

[zyguan]

What problem does this PR solve?

Issue Number: ref #56649

Problem Summary: A lot of small objects will be allocated when rebuildRange, which leads to notable runtime overhead. See the following cpu profile of a index-scan workload (nearly half of rebuildRange time is spent on mallocgc).

What changed and how does it work?

Introduce a planctx level allocator for allocating/reusing range releated objects, includes:

• internal point
• small []*point (len <= 4)
• common Range converted from points (with single low/high value)

After this PR, the cpu profile of the same benchmark can be:

The heap alloc metrics are shown as the following:

Here are benchmark results:

workload	threads	baseline	this PR	diff
idxscan	30	44223.97	44990.54	1.7%
idxscan	60	49583.53	50980.52	2.8%
idxscan	120	50929.30	52141.18	2.4%

The idxscan workload executes select id from sbtest1 where k between {N} and {N+10} on sysbench oltp dataset.

Check List

Tests

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

  • I checked and no code files have been changed.

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]

[APPROVALNOTIFIER] This PR is NOT APPROVED

This pull-request has been approved by:
Once this PR has been reviewed and has the lgtm label, please assign rustin170506 for approval. For more information see the Code Review Process.

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

Needs approval from an approver in each of these files:

• OWNERS
• pkg/planner/OWNERS

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment

[tiprow]

Hi @zyguan. Thanks for your PR.

PRs from untrusted users cannot be marked as trusted with /ok-to-test in this repo meaning untrusted PR authors can never trigger tests themselves. Collaborators can still trigger tests on the PR using /test all.

I understand the commands that are listed here.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository.

[codecov]

Codecov Report

Attention: Patch coverage is 80.89172% with 60 lines in your changes missing coverage. Please review.

Project coverage is 73.6063%. Comparing base (cea46f1) to head (880415e).
Report is 74 commits behind head on master.

Additional details and impacted files

@@               Coverage Diff                @@
##             master     #58362        +/-   ##
================================================
+ Coverage   73.2259%   73.6063%   +0.3803%     
================================================
  Files          1675       1675                
  Lines        462254     466298      +4044     
================================================
+ Hits         338490     343225      +4735     
+ Misses       102989     102226       -763     
- Partials      20775      20847        +72     

Flag	Coverage Δ
integration	43.2124% <80.8917%> (?)

Flags with carried forward coverage won't be shown. Click here to find out more.

Components	Coverage Δ
dumpling	52.6910% <ø> (ø)
parser	∅ <ø> (∅)
br	46.0470% <ø> (+0.0090%)	⬆️

[ti-chi-bot]

@zyguan: The following test failed, say /retest to rerun all failed tests or /retest-required to rerun all mandatory failed tests:

Test name	Commit	Details	Required	Rerun command
idc-jenkins-ci-tidb/unit-test	880415e	link	true	/test unit-test

Full PR test history. Your PR dashboard.

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes-sigs/prow repository. I understand the commands that are listed here.

[winoros]

The plan cache is going to be instance level. I think this is not resonable for instance level plan cache?

[zyguan]

The plan cache is going to be instance level. I think this is not resonable for instance level plan cache?

I'm not sure, but the instance-level plan cache also need to adjustCachedPlan -> RebuildPlan4CachedPlan -> rebuildRange, right?

This is a draft PR to show the issue of util/ranger, a possible solution and the improvement can be archived by it. After discussing with @cfzjywxk and @qw4990 , we think the risk of this PR outweighs the benefits, so no plan to move forward.