Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 28 Sep 2025 03:06:01 +0900
From:      Tomoaki AOKI <junchoon@dec.sakura.ne.jp>
To:        Tomek CEDRO <tomek@cedro.info>
Cc:        Andriy Gapon <avg@freebsd.org>, FreeBSD Current <current@freebsd.org>
Subject:   Re: rc.d/tmp: tmpfs support not entirely complete?
Message-ID:  <20250928030601.8489ed419f642d6f5c4b50b6@dec.sakura.ne.jp>
In-Reply-To: <CAFYkXjkGt5dvsw2cQbTufh9jqavohJQ3FnEtoYWC9syq5cZKuA@mail.gmail.com>
References:  <da28cb1d-9045-4963-bcdc-783512d85313@FreeBSD.org> <CAFYkXjkGt5dvsw2cQbTufh9jqavohJQ3FnEtoYWC9syq5cZKuA@mail.gmail.com>

index | next in thread | previous in thread | raw e-mail

On Sat, 27 Sep 2025 18:02:54 +0200
Tomek CEDRO <tomek@cedro.info> wrote:

> one question here by the way is md faster than tmpfs? i find tmpfs a bit
> slow..
> 
> --
> CeDeROM, SQ7MHZ, http://www.tomek.cedro.info

My understanding is that both are different in their purpose.

md acts as virtual "disk", backed by disk image like *.iso,
and needed to be mounted as the backed filesystem (if the image
is partitioned, mount the partition as the defined filesystem type,
like FAT, UFS, ...).
Not 100% sure, but if filesystems inside md can be updated via md
if mounted with rw.


OTOH, tmpfs itself act as a actual (but volatile) filesystem, mounted as
tmpfs. And once in memory pressures, less actively used part of it is
swapped, if swap partitions or swap files are cofigured (swap backed).

And these differences can affect performance, depending on
memory pressures.

-- 
Tomoaki AOKI    <junchoon@dec.sakura.ne.jp>


home | help

Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20250928030601.8489ed419f642d6f5c4b50b6>