Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 10 Feb 2014 07:46:54 -0800
From:      Peter Grehan <grehan@freebsd.org>
To:        =?ISO-8859-1?Q?Olivier_Cochard-Labb=E9?= <olivier@cochard.me>
Cc:        freebsd-virtualization@freebsd.org
Subject:   Re: Report of my virtual network lab migrated from virtualbox to bhyve
Message-ID:  <52F8F46E.4060704@freebsd.org>
In-Reply-To: <CA%2Bq%2BTcrZZb5o51F4pvLtxKM%2BNvO6SdVEQk_UMLLYSF8JfK6gpg@mail.gmail.com>
References:  <CA%2Bq%2BTcqw7uHLV3=DeZF4=i0hbmECkPP-d5-4ReSQqKCV-JaJ=Q@mail.gmail.com> <52F5363D.8040102@freebsd.org> <CA%2Bq%2BTcrZZb5o51F4pvLtxKM%2BNvO6SdVEQk_UMLLYSF8JfK6gpg@mail.gmail.com>

next in thread | previous in thread | raw e-mail | index | archive | help
Hi Olivier,

>       If you create a sparse file for the bhyve raw disk (e.g. with
>     truncate -s), du will show the actual blocks used rather than the
>     total size.
>
>
> But can I truncate an already existing image disk (downloaded nanobsd
> image as example) ?

  Not without some processing. One way to do it might be a downloaded -> 
qcow2 -> raw using qemu-img, which will create sparse files if possible.

>     There is an em(4) emulation slowly being worked on. It should also
>     be possible to add altq functionality to FreeBSD's virtio net driver.
>
>
> Adding altq(4) to vtnet(4) should be a better idea.

  As bryanv@ mentioned, it's there with a conditional flag.

>       For internal-only networks, there will most likely be a user-space
>     ethernet switch (ala VDE) that bhyve network interfaces can be
>     pointed at.
>
>
> Great: Why not the high-perf netmap VALE switch ?

  That's being worked on, but I think that requires root access.

> Because with a dummy TCP iperf bench between bhyve guests report only
> 254 Mbits/sec.

  Hmmm, I've seen better numbers than that. What's your host h/w, guest 
config and iperf params ?

later,

Peter.



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?52F8F46E.4060704>