Re: change default default_toast_compression to lz4?

From: Peter Eisentraut <peter(at)eisentraut(dot)org>
To: Michael Paquier <michael(at)paquier(dot)xyz>, Euler Taveira <euler(at)eulerto(dot)com>
Cc: Aleksander Alekseev <aleksander(at)tigerdata(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: change default default_toast_compression to lz4?
Date: 2026-02-24 21:54:52
Message-ID: c755c769-54ed-4abe-8d9f-cf214fdef4a1@eisentraut.org
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 24.02.26 09:01, Michael Paquier wrote:
> diff --git a/meson.build b/meson.build
> index 055e96315d0..13ef9c18477 100644
> --- a/meson.build
> +++ b/meson.build
> @@ -1133,6 +1133,8 @@ if not lz4opt.disabled()
> if lz4.found()
> cdata.set('USE_LZ4', 1)
> cdata.set('HAVE_LIBLZ4', 1)
> + else
> + lz4 = not_found_dep
> endif
>
> Hmm. Isn't this bit something that we should actually backpatch? It
> feels wrong to not enforce lz4 to not_found_dep in this meson path.
> So this looks unrelated to the switch of the default value.

I don't think this change is correct, or at least it's not required, and
you won't find it in the handling of other dependencies.

The point of assigning not_found_dep is to make lz4.found() return
false, but in that particular code location lz4.found() is already false.

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Chao Li 2026-02-24 22:03:14 Re: Fix bug of clearing of waitStart in ProcWakeup()
Previous Message Peter Eisentraut 2026-02-24 21:52:39 Re: change default default_toast_compression to lz4?