mailarchive of the ptxdist mailing list
 help / color / mirror / Atom feed
From: Alexander Aring <aar@pengutronix.de>
To: ptxdist@pengutronix.de
Subject: Re: [ptxdist] Generic BSP for x86
Date: Wed, 12 Oct 2016 13:42:31 +0200	[thread overview]
Message-ID: <9415ccfe-31ee-449e-02db-b423727d66fb@pengutronix.de> (raw)
In-Reply-To: <2d419d2c-a908-de90-d78e-7f09b72f4263@init-ka.de>


Hi,

On 10/12/2016 01:31 PM, Koch, Alexander wrote:
> Hi Alex,
> 
>>> Is there anything newer to act as a starting point for creating x86
>>> images with recent PTXdist versions?
>>>
>> The DistroKit [0].
> 
> Thanks for the hint. I had had a quick look at this before, but found
> only the platform definitions for arm7a and rpi, and therefore dropped it.
> 

Yes, this is also enough for testing it on several arm platforms.

>>
>>> Any advice on how to get to an x86 image without creating a new BSP from
>>> scratch is appreciated. I'm not new to PTXdist but have only done ARM
>>> images so far.
>>>
>> I currently have some i586 plattform here (because doing some
>> experiments with a project which has 32 bit support only and I don't
>> fight currently for x86_64 multilib support.) Nevertheless, it's not
>> mainline and x86 32 bit system is somewhat deprecated (in my opinion).
> 
> I know 32bit x86 embedded is dull, but I really need to work that Atom
> Z510 which does not support x86_64. The box does not have a NIC (yet),
> so NFS is currently not an option for me.
> 

If you run it in QEMU then you can emulate an atom which has a NIC. This
is good for develop your BSP/Userspace Softwate/ETC. If it's at some
point that you want the 1:1 behaviour on your atom board -> flash the
image and make a release after some testing.

> So your advice would be to take the DistroKit BSP and create a new
> platform for my x86 target?
> 

I would suggest to copy v7a platform to a atom2510 platform. Then
select this platform and start ptxdist platformconfig -> change
everything to your atom platform.

Then start ptxdist kernelconfig -> change everything to your platform.

Then look into run-nfs/run script and qemu-common source file and change
everything that it emulates a atom.

---

For the switch between QEMU and your atom board, I would disable then
everything what you don't need in the kernelconfig which was necessary
only for doing QEMU testing stuff.

btw:

qemu-i386 -cpu "?"

gives you a list of possible cpu emulations, there is also some:

x86             n270  Intel(R) Atom(TM) CPU N270   @ 1.60GHz

don't know if this fits into your use-case.

For the specific platform changes -> I can't help you there. But I think
there exists now a DistroKit Mailinglist [0], you can send-patches there and
we will review.

- Alex

[0] http://www.pengutronix.de/mailinglists/index_en.html

_______________________________________________
ptxdist mailing list
ptxdist@pengutronix.de

      reply	other threads:[~2016-10-12 11:42 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-10-12  6:38 Koch, Alexander
2016-10-12 10:26 ` Alexander Aring
2016-10-12 11:31   ` Koch, Alexander
2016-10-12 11:42     ` Alexander Aring [this message]

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=9415ccfe-31ee-449e-02db-b423727d66fb@pengutronix.de \
    --to=aar@pengutronix.de \
    --cc=ptxdist@pengutronix.de \
    /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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox