All of lore.kernel.org
 help / color / mirror / Atom feed
From: Paul Barker <paul@paulbarker.me.uk>
To: Marlon Smith <marlon.smith10@gmail.com>
Cc: yocto@yoctoproject.org
Subject: Re: Why use Yocto?
Date: Tue, 10 Jun 2014 10:37:37 +0000	[thread overview]
Message-ID: <20140610103736.GC24880@gmail.com> (raw)
In-Reply-To: <1402354370.9385.9.camel@marlon-Z68X-UD3H-B3>

[-- Attachment #1: Type: text/plain, Size: 2213 bytes --]

On Mon, Jun 09, 2014 at 03:52:50PM -0700, Marlon Smith wrote:
> Hi everyone,
> 
> I'm developing a product that will run on a custom i.MX6 board and I'm
> trying to decide whether to use Yocto or Ubuntu (there's a version of
> Ubuntu packaged for the Wandboard that will run on our board).  The
> board will run our own custom app, and we'll modify the Linux kernel to
> support our hardware.
> 
> Ubuntu seems like it would be ready to go - just put it on an SD card,
> boot the board, compile the app and create a new SD card image from the
> result to use for manufacturing.
> 
> Yocto seems like it would be easier to remove unneeded packages from,
> and easier to cross-compile the application for.  This means we could
> have a smaller SD card image in the end.
> 
> What are your thoughts on this?

Philip has already mentioned license compliance in his reply, I'd like to add a
couple of other points:

- In addition to a smaller image, you should have less services running by
  default and so lower power usage.

- It's much easier to do consistent, reproducible image builds which include
  your own packages. Rather than having a series of steps such as installing
  Ubuntu on an SD card, booting, installing required additional packages,
  downloading your source code to the card, building and then installing, you
  just do 'bitbake my-image' and everything you need is encoded in recipe files
  which you can keep under version control. There's less chance for human error
  to creep in.

- You don't need to install the toolchain on the board itself, you can do the
  system build on a separate machine and not pollute the SD card image with the
  history of building your software. It saves you the time of going through and
  removing the things you need to build your software but aren't needed to run
  it, which you'll probably end up doing to reduce the image size.

- You'll have a great community of people doing similar things with the Yocto
  Project. I don't know of a similar community for modifying Ubuntu SD card
  images in this fashion.

Hope this helps,

-- 
Paul Barker

Email: paul@paulbarker.me.uk
http://www.paulbarker.me.uk

[-- Attachment #2: Type: application/pgp-signature, Size: 484 bytes --]

  parent reply	other threads:[~2014-06-10 10:37 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-06-09 22:52 Why use Yocto? Marlon Smith
2014-06-09 23:11 ` Philip Balister
2014-06-10 10:37 ` Paul Barker [this message]
2014-06-10 11:25   ` Christian Ege
2014-06-10 14:39     ` Bob Cochran
2014-06-10 15:42       ` Nicolas Dechesne
2014-06-10 16:50         ` Marlon Smith
2014-06-10 17:16           ` Alex J Lennon
2014-06-17 20:17             ` Insop Song
2014-06-10 14:47 ` Maxim Radugin

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20140610103736.GC24880@gmail.com \
    --to=paul@paulbarker.me.uk \
    --cc=marlon.smith10@gmail.com \
    --cc=yocto@yoctoproject.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.