Re: Compiling PostgreSQL for WIndows with 16kb blocksize

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Yannick Collette <yannickcollette(at)gmail(dot)com>
Cc: pgsql-hackers(at)lists(dot)postgresql(dot)org
Subject: Re: Compiling PostgreSQL for WIndows with 16kb blocksize
Date: 2022-01-19 17:22:55
Message-ID: 543457.1642612975@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Yannick Collette <yannickcollette(at)gmail(dot)com> writes:
> For test purposes I need to compile PostgreSQL 14.1 using a 16kb blocksize.

> CFLAGS="-D WINVER=0x0600 -D _WIN32_WINNT=0x0600" LIBS="-ladvapi32"
> ./configure --host=x86_64-w64-mingw32 --with-blocksize=16
> --with-wal-blocksize=16 --with-openssl --with-libxml
> --prefix=/c/postgresql/pg14/ 2>&1 | tee configure.log

I don't know anything about the Windows-specific details here,
but the --with-blocksize option looks fine, and it works for me:

regression=# show block_size;
block_size
------------
16384
(1 row)

(Worth noting here is that a lot of our regression tests fail
at non-default blocksizes, because plans change, or rows no
longer need toasting, or the like.)

> Unfortunately my blocksize is still 8kb when checking in DB.
> postgres=# show block_size;
> block_size
> ------------
> 8192
> (1 row)

I think you must be connecting to the wrong server.

regards, tom lane

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Andres Freund 2022-01-19 17:42:31 Re: slowest tap tests - split or accelerate?
Previous Message Tom Lane 2022-01-19 17:14:21 Re: slowest tap tests - split or accelerate?