*: Use strict validation for stale read ts & flashback ts

[MyonKeminta]

What problem does this PR solve?

Issue Number: close #56809

Problem Summary:

What changed and how does it work?

Use strict check for stale read ts & flashback ts to avoid any reading on future ts, which has risk to break the linearizability of async commit and 1PC transactions.

Pending and waiting for tikv/client-go#1484

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]

Skipping CI for Draft Pull Request.
If you want CI signal for your change, please convert it to an actual PR.
You can still manually trigger a test run with /test all

[tiprow]

Hi @MyonKeminta. 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.

[ti-chi-bot]

PR needs rebase.

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.

[cfzjywxk]

LGTM

[ti-chi-bot]

@cfzjywxk: You cannot manually add or delete the reviewing state labels, only I and the tursted members have permission to do so.

In response to removing label named needs-1-more-lgtm.

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 ti-community-infra/tichi repository.

[ti-chi-bot]

In response to a cherrypick label: new pull request created to branch release-8.5: #57372.

[MyonKeminta]

A brief benchmark result on one of our internal stale read workload, with update low resolution ts interval configured by default (2s):

version	read_staleness / s	QPS	.95 lat / ms
baseline	-5	4730.05	5.77
baseline	-1	4690.31	5.77
fix	-5	4658.86	5.88
fix	-1	4709.33	5.77
fix - tso 5ms injection	-1	4723.33	5.77

This can be a proof that when the read staleness is shorter than the update interval of low resolution ts, the performance can still be kept.

[ti-chi-bot]

In response to a cherrypick label: new pull request created to branch release-7.5: #58086.

[ti-chi-bot]

In response to a cherrypick label: new pull request created to branch release-8.1: #58158.

[ti-chi-bot]

In response to a cherrypick label: new pull request created to branch release-6.5: #58409.

[ti-chi-bot]

In response to a cherrypick label: new pull request created to branch release-6.1: #58410.