All of lore.kernel.org
 help / color / mirror / Atom feed
From: Eric Nelson <eric.nelson@boundarydevices.com>
To: Philip Craig <phil@blackmoth.com.au>
Cc: "meta-freescale@yoctoproject.org" <meta-freescale@yoctoproject.org>
Subject: Re: gpu-viv-mx6q and libEGL.so/libEGL.so.1
Date: Fri, 26 Jul 2013 07:10:25 -0700	[thread overview]
Message-ID: <51F28351.20901@boundarydevices.com> (raw)
In-Reply-To: <CAGaPg5+i2b35cnTcVHUnUwRKGKiUjX9ZUhy6PryAPcw8dRN3Og@mail.gmail.com>

Hi Philip,

On 07/26/2013 03:05 AM, Philip Craig wrote:
> On Fri, Jul 26, 2013 at 7:54 PM, Philip Craig <phil@blackmoth.com.au> wrote:
>> On Fri, Jul 26, 2013 at 10:44 AM, Eric Nelson
>> <eric.nelson@boundarydevices.com> wrote:
>>> Hi all,
>>>
>>> When running the sample apps in /opt/viv_samples/vdk
>>> under strace on a frame-buffer only build, I can see that
>>> something is trying to load libEGL.so (no trailing 1).
>>>
>>> Looking at the recipe, I'm not quite grokking how this is
>>> placed into the image though.
>>>
>>> I can see a symlink being created pointing libEGL.so at
>>> libEGL.so.1 here:
>>>
>>>
>>> https://github.com/Freescale/meta-fsl-arm/blob/master/recipes-graphics/gpu-viv-bin-mx6q/gpu-viv-bin-mx6q.inc#L130
>>>
>>> Somehow or another, the symlink is ending up in the -dev package
>>> though:
>>>
>>> $ find gpu-viv-bin-mx6q/1_3.0.35-4.0.0-r5.0/packages-split/ \
>>>          -name libEGL\*
>>> gpu-viv-bin-mx6q/1_3.0.35-4.0.0-r5.0/packages-split/libegl-mx6-dev/usr/lib/libEGL.so
>>> gpu-viv-bin-mx6q/1_3.0.35-4.0.0-r5.0/packages-split/libegl-mx6/usr/lib/libEGL.so.1
>>>
>>> Can anybody give some guidance on where and how that's happening?
>>
>> See lines 152 and 153 of gpu-viv-bin-mx6q.inc. If something needs
>> libEGL.so then this probably needs to be changed to how libgles and
>> libgles2 are done. It would be nice to document exactly where the
>> dependency is though so that we know when we can remove the hack once
>> it is fixed.
>>
>>> Does anybody know where the missing reference to libEGL.so
>>> might be coming from? This doesn't appear to be a hard-coded
>>> linker dependency from any of the sample apps or from the
>>> other shared libraries.
>>
>> I can't see the dependency from inspection. What's the exact error message?
>
> I see now. The samples uses libGAL.so, and it contains the string
> libEGL.so, as well as a gcoOS_LoadEGLLibrary function.
>

Thanks for finding this.

Since these are binaries, we'll need to defer to Freescale/Vivante
for a proper fix.




  parent reply	other threads:[~2013-07-26 14:10 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-07-26  0:44 gpu-viv-mx6q and libEGL.so/libEGL.so.1 Eric Nelson
2013-07-26  9:54 ` Philip Craig
2013-07-26 10:05   ` Philip Craig
2013-07-26 13:46     ` Otavio Salvador
2013-07-26 14:10     ` Eric Nelson [this message]
2013-07-26 14:06   ` Eric Nelson
2013-07-26 14:15   ` Eric Nelson
2013-07-26 15:31     ` Otavio Salvador

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=51F28351.20901@boundarydevices.com \
    --to=eric.nelson@boundarydevices.com \
    --cc=meta-freescale@yoctoproject.org \
    --cc=phil@blackmoth.com.au \
    /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.