Re: [PATCHES] Post-special page storage TDE support - Mailing list pgsql-hackers

From David Christensen
Subject Re: [PATCHES] Post-special page storage TDE support
Date
Msg-id CAOxo6XJXTh5O9oQA9zJgTTLybwWNQ8eq-Fpvwtd-5pJO2tBEog@mail.gmail.com
Whole thread Raw
In response to Re: [PATCHES] Post-special page storage TDE support  (David Christensen <david.christensen@crunchydata.com>)
Responses Re: [PATCHES] Post-special page storage TDE support  (David Christensen <david.christensen@crunchydata.com>)
List pgsql-hackers
So here is a v3 here, incorporating additional bug fixes and some
design revisions.  I have narrowed this down to 3 patches, fixing the
bugs that were leading to the instability of the specific test file so
dropping that as well as removing the useless POC "wasted space".

The following pieces are left:

0001 - adjust the codebase to utilize the "reserved_page_space"
variable for all offsets rather than assuming compile-time constants.
This allows us to effectively allocate a fixed chunk of storage from
the end of the page and have everything still work on this cluster.
0002 - add the Page Feature abstraction. This allows you to utilize
this chunk of storage, as well as query for feature use at the page
level.
0003 - the first page feature, 64-bit encryption (soon to be
renumbered when GCM storage for TDE is introduced, though the two
features are designed to be incompatible).  This includes an
arbitrarily found 64-bit checksum, so we probably will need to write
our own or ensure that we have something license-compatible.

This is rebased and current as-of-today and passes all CI tests, so
should be in a good place to start looking at.

Best,

David

Attachment

pgsql-hackers by date:

Previous
From: Michael Paquier
Date:
Subject: Re: 011_crash_recovery.pl intermittently fails
Next
From: Tom Lane
Date:
Subject: Re: 011_crash_recovery.pl intermittently fails