MDEV-35069 - IMPORT TABLESPACE does not work for tables with vector, although allowed

[svoj]

• The Jira issue number for this PR is: MDEV-35069

Description

Propagate discard/import tablespace request to hlindexes.

Let FLUSH TABLES ... FOR EXPORT open/lock hlindexes, so that InnoDB prepares hlindexes for export.

Moved reset_hlindexes() to external_lock(F_UNLCK), so that hlindexes are available for export until UNLOCK TABLES.

Release Notes

Probably update "Copying Transportable Tablespaces" with information about VECTOR indexes. Their tablespaces have to be copied along with parent table.

How can this PR be tested?

mtr vector_innodb

Basing the PR against the correct MariaDB version

• This is a new feature or a refactoring, and the PR is based against the main branch.
• This is a bug fix, and the PR is based against the earliest maintained branch in which the bug can be reproduced.

PR quality check

• I checked the CODING_STANDARDS.md file and my PR conforms to this where appropriate.
• For any trivial modifications to the PR, I am ok with the reviewer making the changes themselves.

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[vuvova]

Where does InnoDB "prepares hlindexes for export" (or tables, FWIW), REFRESH_FOR_EXPORT isn't passed down to the engine, as far as I can see.

I don't understand "reset_hlindexes" move either, why hlindexes aren't "available for export until UNLOCK TABLES" now?

[svoj]

@vuvova

Where does InnoDB "prepares hlindexes for export" (or tables, FWIW), REFRESH_FOR_EXPORT isn't passed down to the engine, as far as I can see.

ha_innobase::external_lock() / row_quiesce_table_start() / row_quiesce_write_cfg()

I don't understand "reset_hlindexes" move either, why hlindexes aren't "available for export until UNLOCK TABLES" now?

InnoDB creates .cfg file on FTFE and keeps it until UNLOCK TABLES (external_lock(F_UNLCK)). This file has to be copied along with .ibd to destination.

Currently hlindexes are open until handler::ha_reset(), that is end of statement. It means .cfg will be removed before FTFE itself completes.

[vuvova]

ok to push, thanks