Re: pgsql: Invent PageIndexTupleOverwrite, and teach BRIN and GiST to use i

2016-09-22 23:11 GMT+05:00 Tom Lane <tgl@sss.pgh.pa.us>:
> Yeah, I explained it in part of the commit message you didn't quote:

Oh.. sorry, my bad. Focused on code and missed whole thing.

>     Note that this patch causes a rather subtle WAL incompatibility: the
>     physical page content change represented by certain WAL records is now
>     different than it was before, because while the tuples have the same
>     itempointer line numbers, the tuples themselves are in different places.
As for GiST, that was my initial suggestion, but tests of Anastasia
Lubennikova showed that in some cases index size is reduced by 3. It's
not possible to get 3 times less pages by compatible WAL replay.

>     I have not bumped the WAL version number because I think it doesn't matter
>     unless you are trying to do bitwise comparisons of original and replayed
>     pages, and in any case we're early in a devel cycle and there will probably
>     be more WAL changes before v10 gets out the door.
I think that's correct assumption. Sorry for time taken.

Best regards, Andrey Borodin.