All of lore.kernel.org
 help / color / mirror / Atom feed
From: Trevor Woerner <twoerner@gmail.com>
To: "Iván Castell" <icastell@nayarsystems.com>
Cc: Yocto Project <yocto@yoctoproject.org>
Subject: Re: Issue when integrating different bsp-layers on a single bblayers.conf
Date: Thu, 7 Jun 2018 12:47:04 -0400	[thread overview]
Message-ID: <20180607164704.GA24833@linux-uys3> (raw)
In-Reply-To: <CAALXFYwr=UUkS7dVb3QzHzsoQ3pU56j58CZunkKJkRYYcpMv3g@mail.gmail.com>

Hi Iván,

On Thu 2018-06-07 @ 04:39:11 PM, Iván Castell wrote:
> When setting the "build" directory with a bblayers.conf customized for a
> single platform, each platform builds the image recipe properly.
> 
> However, when I have integrated all bsp-layers in a single bblayers.conf,
> the compilation of some platforms has been broken.

This is a problem that comes up fairly often, and is a bug with the BSP layer
itself. All BSP layers are supposed to be created so that they work well when
intermixed with other BSP layers, but often they don't. See Angstrom:

	http://www.angstrom-distribution.org/

> The specific problem is this: one bsp layer (meta-rockchip +
> meta-rockchip-extra) defines a recipes-graphics/mesa/mesa_%.bbappend with
> this content inside:

I am the maintainer of the non-vendor meta-rockchip layer:

	https://git.yoctoproject.org/cgit/cgit.cgi/meta-rockchip/

The fact you mention this problem and meta-rockchip-extra implies to me that
you're using the vendor-provided meta-rockchip and meta-rockchip-extra:

	https://github.com/rockchip-linux/meta-rockchip
	https://github.com/rockchip-linux/meta-rockchip-extra

The problem is, the vendor is only interested in their own hardware, so to
them there's no reason to ever mix their BSP layer with any other BSP layer.
There's no interest on their part to "play well" with others or to check
whether their layers comply with community guidelines.

As a person who likes OpenEmbedded as a whole, I try my best to make sure any
layer I'm interested in plays well within the entire OpenEmbedded ecosystem.
However, because mali graphics are currently all binary blobs and projects
like Lima and Panfrost aren't yet ready for prime-time, there isn't much
support yet for _accelerated_ graphics in the non-vendor layer.

>     - Can you suggest a fix to solve this issue?

Fix the problem in the layer by using MACHINE-specific OVERRIDES and submit
a pull request to the owners. However, I've noticed there haven't been any
updates to their OE layers in the past 4 months or more, so I can't help
wonder if they've abandoned them.

Out of curiosity, which MACHINE(s) are you interested in?

Best regards,
	Trevor


  parent reply	other threads:[~2018-06-07 16:47 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-06-07 14:39 Issue when integrating different bsp-layers on a single bblayers.conf Iván Castell
2018-06-07 15:09 ` Ulf Samuelsson
2018-06-07 15:27 ` Ulf Samuelsson
2018-06-07 16:47 ` Trevor Woerner [this message]
2018-06-08  7:22   ` Iván Castell
  -- strict thread matches above, loose matches on Subject: below --
2018-06-14 20:50 Peter Kjellerstedt

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=20180607164704.GA24833@linux-uys3 \
    --to=twoerner@gmail.com \
    --cc=icastell@nayarsystems.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.