Date: Mon, 14 Aug 2023 12:22:24 +0200 (CEST)
From: Fabien COELHO <coelho@cri.ensmp.fr>
To: Thomas Munro <thomas.munro@gmail.com>
cc: Tatsuo Ishii <ishii@sraoss.co.jp>, 
    PostgreSQL Developers <pgsql-hackers@lists.postgresql.org>
Subject: Re: pgbench with libevent?
In-Reply-To: <CA+hUKGJRD3p8rc8rf7fXcQqBQmOShEuwAZybpFC08+Rf7Y1nJw@mail.gmail.com>
Message-ID: <eb36f089-98ca-6277-dbae-d8fb7c6ec585@mines-paristech.fr>
References: <f6a5e724-e2d2-af78-4a6-4547994761b1@minesparis.psl.eu> <9633b9f1-c128-7031-6a9e-1af6d247caa0@minesparis.psl.eu> <CA+hUKGK8St8+hJ1kb1fygEzNwgJunVyBL-+1pg-6W+JJPmOnyA@mail.gmail.com> <20230814.150659.1167547855459810626.t-ishii@sranhm.sra.co.jp>
 <CA+hUKGJRD3p8rc8rf7fXcQqBQmOShEuwAZybpFC08+Rf7Y1nJw@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII; format=flowed
Archived-At: <https://www.postgresql.org/message-id/eb36f089-98ca-6277-dbae-d8fb7c6ec585%40mines-paristech.fr>
Precedence: bulk


>> Interesting. In my understanding this also needs to make Latch
>> frontend-friendly?
>
> It could be refactored to support a different subset of event types --
> maybe just sockets, no latches and obviously no 'postmaster death'.
> But figuring out how to make latches work between threads might also
> be interesting for future projects...
>
> Maybe Fabien has completion-based I/O in mind (not just "readiness").

Pgbench is really a primitive client on top of libpq. ISTM that 
completion-based I/O would require to enhance libpq asynchronous-ity, not 
just expose its underlying fd to allow asynchronous implementations.
Currently pgbench only actuall "waits" for results from the server
and testing PQisBusy to check whether they are there.

> That's something that some of those libraries can do, IIUC.  For
> example, when your thread wakes up, it tells you "your socket read is
> finished, the data is already in your target buffer".

Indeed, libevent has a higher level "buffer" oriented API.

> As opposed to "you can now call recv() without blocking", so you avoid 
> another trip into the kernel.  But that's also something we'll 
> eventually want to figure out in the server.

-- 
Fabien.