All of lore.kernel.org
 help / color / mirror / Atom feed
From: Randolph Sapp <rs@ti.com>
To: Denys Dmytriyenko <denis@denix.org>, <sebastian.krahl@lacon.de>
Cc: <meta-ti@lists.yoctoproject.org>
Subject: Re: [meta-ti] core-image-sato scarthgap beaglebone x11 startup error
Date: Tue, 6 May 2025 16:55:20 -0500	[thread overview]
Message-ID: <D9PEV9NDIIDO.AP7Z9507B59N@ti.com> (raw)
In-Reply-To: <20250506162944.GC18383@denix.org>

On Tue May 6, 2025 at 11:29 AM CDT, Denys Dmytriyenko wrote:
> + Randolph
>
> Unlike yocto-bsp, which uses vanilla mainline kernel with unaccelerated 
> framebuffer, meta-ti by default enables accelerated graphics with out-of-tree 
> drivers and user-space GLES libraries - SGX in case of Beaglebone and AM335x.
>
> When meta-ti migrated from 6.1 kernel to 6.6, I believe X11 support was 
> dropped. It should still be possible to fall back to unaccelerated framebuffer 
> and get X11 working, but that's not something that is configured nor tested in 
> meta-ti.

This was not a result of a kernel change, it was due to a Yocto lts version
change. A lot of layers use a common PACKAGECONFIG pattern that configures
packages to use Desktop GL instead of GLES when opengl and x11 distro features
are enabled. Arago took a hard stance that we will disable x11 to avoid
reconfiguring numerous packages for partial acceleration under x11. Users can
try it, but they will need to be careful when configuring things.

> On Tue, May 06, 2025 at 07:57:48AM -0700, Sebastian via lists.yoctoproject.org wrote:
>> I've tested meta-ti for kirkstone the xserver starts up correctly into an X 
>> session.
>> 
>> With meta-ti for scarthgap it fails. As it looks for a /dev/fb device which 
>> is not present.

Can you elaborate on this? Missing framebuffer devices is actually a display
controller related issue. If you have a splash screen then some fb device is
being registered. What device is this for?
 
>> Aside of this, with kirkstone and scarthgap in yocto-bsp and 
>> beaglebone-yocto it works in both cases.
>> 
>> The framebuffer is filled up by the splash screen also in powering down.
>> 
>> What am I missing?
>> 
>> What info to supply?
>> 
>> Regards,
>> 
>> Sebastian



  reply	other threads:[~2025-05-06 21:55 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-05-06 14:57 core-image-sato scarthgap beaglebone x11 startup error Sebastian
2025-05-06 16:29 ` [meta-ti] " Denys Dmytriyenko
2025-05-06 21:55   ` Randolph Sapp [this message]
2025-05-06 22:53     ` Denys Dmytriyenko
2025-05-07  5:40       ` Sebastian
2025-05-08 18:33         ` [meta-ti] " Randolph Sapp
2025-05-08 19:13           ` Denys Dmytriyenko
2025-05-08 20:37             ` Sebastian
2025-05-13 11:05               ` Sebastian
2025-05-21 13:37                 ` Sebastian
2025-05-22 13:30                   ` [meta-ti] " Andrew Davis
2025-05-27  7:59                     ` Sebastian

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=D9PEV9NDIIDO.AP7Z9507B59N@ti.com \
    --to=rs@ti.com \
    --cc=denis@denix.org \
    --cc=meta-ti@lists.yoctoproject.org \
    --cc=sebastian.krahl@lacon.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 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.