[ptxdist] GTK+ on DirectFB

[ptxdist] GTK+ on DirectFB

[Guillermo Rodriguez Garcia]

Hello all,

I'm trying to test GTK+ on top of DirectFB on the mini2440 (2011.12.0).
I'm CCing the ptxdist ML in case this is not mini2440 specific.

I have built DirectFB with framebuffer support, and tslib and
linuxinput as input drivers. This works fine: I can run most of the
df_xxx examples (e.g. df_input, df_window, df_dok..)

I can also run GTK+ when built for the X.Org target.

However if I build GTK+ for the DirectFB target, the testgtk example
clears the screen, then crashes. Here's the output:

root@mini2440:~ testgtk

** (process:739): WARNING **: Couldn't find file "testgtkrc".
commandline read: testgtk

   ~~~~~~~~~~~~~~~~~~~~~~~~~~| DirectFB 1.4.3
|~~~~~~~~~~~~~~~~~~~~~~~~~~
        (c) 2001-2009  The world wide DirectFB Open Source Community
        (c) 2000-2004  Convergence (integrated media) GmbH
      ----------------------------------------------------------------

(*) DirectFB/Core: Single Application Core. (2012-01-26 09:34)
(*) Direct/Memcpy: Using armasm_memcpy()
(*) Direct/Thread: Started 'VT Switcher' (-1) [CRITICAL OTHER/OTHER
0/0] <8388608>...
(*) Direct/Thread: Started 'VT Flusher' (-1) [DEFAULT OTHER/OTHER 0/0]
<8388608>...
(*) DirectFB/FBDev: Found 's3c2410fb' (ID 0) with frame buffer at
0x33940000, 150k (MMIO 0x00000000, 0k)
(*) Direct/Thread: Started 'tslib Input' (-1) [INPUT OTHER/OTHER 0/0]
<8388608>...
(*) DirectFB/Input: tslib touchscreen 0 0.1 (tslib)
(*) Direct/Thread: Started 'Linux Input' (-1) [INPUT OTHER/OTHER 0/0]
<8388608>...
(*) DirectFB/Input: gpio-keys 0.1 (directfb.org)
(*) Direct/Thread: Started 'Keyboard Input' (-1) [INPUT OTHER/OTHER
0/0] <8388608>...
(*) DirectFB/Input: Keyboard 0.9 (directfb.org)
(*) DirectFB/Graphics: Generic Software Rasterizer 0.6 (directfb.org)
(*) DirectFB/Core/WM: Default 0.3 (directfb.org)

(testgtk:739): Gdk-CRITICAL **: IA__gdk_drawable_set_colormap:
assertion `cmap == NULL || gdk_drawable_get_depth (drawable) == cmd
(*) Direct/Thread: Started 'EventBufferFeed' (-1) [MESSAGING
OTHER/OTHER 0/0] <8388608>...
(!) [  739:    0.000] --> Caught signal 11 (at 0x28, invalid address)
<--
 (!!!)  *** WARNING [still objects in 'Window Pool'] *** [object.c:241
in fusion_object_pool_destroy()]
 (!!!)  *** WARNING [still objects in 'Layer Region Pool'] ***
[object.c:241 in fusion_object_pool_destroy()]
 (!!!)  *** WARNING [still objects in 'Layer Context Pool'] ***
[object.c:241 in fusion_object_pool_destroy()]
 (!!!)  *** WARNING [still objects in 'Surface Pool'] ***
[object.c:241 in fusion_object_pool_destroy()]
Aborted

Can anyone share any experiences on this?

Thanks,

Guillermo Rodriguez Garcia
guille.rodriguez@gmail.com

-- 
ptxdist mailing list
ptxdist@pengutronix.de

Re: [ptxdist] [oselas] GTK+ on DirectFB

[Juergen Beisert]

Hi Guillermo,

Guillermo Rodriguez Garcia wrote:
> [...]
> (testgtk:739): Gdk-CRITICAL **: IA__gdk_drawable_set_colormap:
> assertion `cmap == NULL || gdk_drawable_get_depth (drawable) == cmd
> (*) Direct/Thread: Started 'EventBufferFeed' (-1) [MESSAGING
> OTHER/OTHER 0/0] <8388608>...
> (!) [  739:    0.000] --> Caught signal 11 (at 0x28, invalid address)
> <--
>  (!!!)  *** WARNING [still objects in 'Window Pool'] *** [object.c:241 in fusion_object_pool_destroy()]
>  (!!!)  *** WARNING [still objects in 'Layer Region Pool'] ***
> [object.c:241 in fusion_object_pool_destroy()]
>  (!!!)  *** WARNING [still objects in 'Layer Context Pool'] ***
> [object.c:241 in fusion_object_pool_destroy()]
>  (!!!)  *** WARNING [still objects in 'Surface Pool'] ***
> [object.c:241 in fusion_object_pool_destroy()]
> Aborted
>
> Can anyone share any experiences on this?

Did you re-start with a clean GTK after you switch from Xorg to DirectFB?
Sometimes the package's buildsystem does not re-compile everything after
changing some parameters, so some object files are still compiled for the
previous parameter settings. Just an idea.

Sorry, haven't worked with DirectFB for ages, no other idea.

Regards,
Juergen

-- 
Pengutronix e.K.                              | Juergen Beisert             |
Linux Solutions for Science and Industry      | http://www.pengutronix.de/  |

-- 
ptxdist mailing list
ptxdist@pengutronix.de

Re: [ptxdist] GTK+ on DirectFB

[Robert Schwebel]

Guillermo,

On Thu, Jan 26, 2012 at 11:13:51AM +0100, Guillermo Rodriguez Garcia wrote:
> I'm trying to test GTK+ on top of DirectFB on the mini2440 (2011.12.0).

I cannot help you with the current issue, but note that we have removed
gtk/directfb support from ptxdist mainline in the meantime.

The reason is that nobody has maintained the DFB backend in gtk for
quite a long time, so upstream has removed this backend from gtk
entirely, and we didn't want to be stuck with old gtk versions.

Nevertheless, if you need it, you can stay with the glib/gtk stack from
before the changes. Be prepared to fix the kind of errors you are
currently noticing.

rsc
-- 
Pengutronix e.K.                           |                             |
Industrial Linux Solutions                 | http://www.pengutronix.de/  |
Peiner Str. 6-8, 31137 Hildesheim, Germany | Phone: +49-5121-206917-0    |
Amtsgericht Hildesheim, HRA 2686           | Fax:   +49-5121-206917-5555 |

-- 
ptxdist mailing list
ptxdist@pengutronix.de

Re: [ptxdist] [oselas] GTK+ on DirectFB

[Guillermo Rodriguez Garcia]

Hello Juergen,

2012/1/26 Juergen Beisert <jbe@pengutronix.de>:
> Hi Guillermo,
>
> Guillermo Rodriguez Garcia wrote:
>> [...]
>> (testgtk:739): Gdk-CRITICAL **: IA__gdk_drawable_set_colormap:
>> assertion `cmap == NULL || gdk_drawable_get_depth (drawable) == cmd
>> (*) Direct/Thread: Started 'EventBufferFeed' (-1) [MESSAGING
>> OTHER/OTHER 0/0] <8388608>...
>> (!) [  739:    0.000] --> Caught signal 11 (at 0x28, invalid address)
>> <--
>>  (!!!)  *** WARNING [still objects in 'Window Pool'] *** [object.c:241 in fusion_object_pool_destroy()]
>>  (!!!)  *** WARNING [still objects in 'Layer Region Pool'] ***
>> [object.c:241 in fusion_object_pool_destroy()]
>>  (!!!)  *** WARNING [still objects in 'Layer Context Pool'] ***
>> [object.c:241 in fusion_object_pool_destroy()]
>>  (!!!)  *** WARNING [still objects in 'Surface Pool'] ***
>> [object.c:241 in fusion_object_pool_destroy()]
>> Aborted
>>
>> Can anyone share any experiences on this?
>
> Did you re-start with a clean GTK after you switch from Xorg to DirectFB?
> Sometimes the package's buildsystem does not re-compile everything after
> changing some parameters, so some object files are still compiled for the
> previous parameter settings. Just an idea.

No, I didn't. I assumed ptxdist would take care. I'll give this a
shot, rebuilding everything from scratch.

>
> Sorry, haven't worked with DirectFB for ages, no other idea.

Thanks -- your feedback is welcome.

Guillermo Rodriguez Garcia
guille.rodriguez@gmail.com

-- 
ptxdist mailing list
ptxdist@pengutronix.de

Re: [ptxdist] GTK+ on DirectFB

[Guillermo Rodriguez Garcia]

Hello Robert,

2012/1/26 Robert Schwebel <r.schwebel@pengutronix.de>:
> Guillermo,
>
> On Thu, Jan 26, 2012 at 11:13:51AM +0100, Guillermo Rodriguez Garcia wrote:
>> I'm trying to test GTK+ on top of DirectFB on the mini2440 (2011.12.0).
>
> I cannot help you with the current issue, but note that we have removed
> gtk/directfb support from ptxdist mainline in the meantime.

Thanks for the heads up. I was not aware of this at all.

Guillermo Rodriguez Garcia
guille.rodriguez@gmail.com

-- 
ptxdist mailing list
ptxdist@pengutronix.de

Re: [ptxdist] [oselas] GTK+ on DirectFB

[Juergen Beisert]

Hi Guillermo,

Guillermo Rodriguez Garcia wrote:
> 2012/1/26 Juergen Beisert <jbe@pengutronix.de>:
> > Hi Guillermo,
> >
> > Guillermo Rodriguez Garcia wrote:
> >> [...]
> >> (testgtk:739): Gdk-CRITICAL **: IA__gdk_drawable_set_colormap:
> >> assertion `cmap == NULL || gdk_drawable_get_depth (drawable) == cmd
> >> (*) Direct/Thread: Started 'EventBufferFeed' (-1) [MESSAGING
> >> OTHER/OTHER 0/0] <8388608>...
> >> (!) [  739:    0.000] --> Caught signal 11 (at 0x28, invalid address)
> >> <--
> >>  (!!!)  *** WARNING [still objects in 'Window Pool'] *** [object.c:241
> >> in fusion_object_pool_destroy()] (!!!)  *** WARNING [still objects in
> >> 'Layer Region Pool'] ***
> >> [object.c:241 in fusion_object_pool_destroy()]
> >>  (!!!)  *** WARNING [still objects in 'Layer Context Pool'] ***
> >> [object.c:241 in fusion_object_pool_destroy()]
> >>  (!!!)  *** WARNING [still objects in 'Surface Pool'] ***
> >> [object.c:241 in fusion_object_pool_destroy()]
> >> Aborted
> >>
> >> Can anyone share any experiences on this?
> >
> > Did you re-start with a clean GTK after you switch from Xorg to DirectFB?
> > Sometimes the package's buildsystem does not re-compile everything after
> > changing some parameters, so some object files are still compiled for the
> > previous parameter settings. Just an idea.
>
> No, I didn't. I assumed ptxdist would take care.

PTXdist: yes. It calls again the package's configure script with the new 
settings. But sometimes the package's buildsystem is broken and it does not 
re-compile all objects in accordance to the new configuration.

> I'll give this a shot, rebuilding everything from scratch.

Not everything: a "ptxdist clean gtk" should be enough. This means only the 
gtk package starts from scratch.

Regards,
Juergen

-- 
Pengutronix e.K.                              | Juergen Beisert             |
Linux Solutions for Science and Industry      | http://www.pengutronix.de/  |

-- 
ptxdist mailing list
ptxdist@pengutronix.de

Re: [ptxdist] [oselas] GTK+ on DirectFB

[Guillermo Rodriguez Garcia]

2012/1/26 Juergen Beisert <jbe@pengutronix.de>:
> Hi Guillermo,
>
> Guillermo Rodriguez Garcia wrote:
>> 2012/1/26 Juergen Beisert <jbe@pengutronix.de>:
>> > Hi Guillermo,
>> >
>> > Guillermo Rodriguez Garcia wrote:
>> >> [...]
>> >> (testgtk:739): Gdk-CRITICAL **: IA__gdk_drawable_set_colormap:
>> >> assertion `cmap == NULL || gdk_drawable_get_depth (drawable) == cmd
>> >> (*) Direct/Thread: Started 'EventBufferFeed' (-1) [MESSAGING
>> >> OTHER/OTHER 0/0] <8388608>...
>> >> (!) [  739:    0.000] --> Caught signal 11 (at 0x28, invalid address)
>> >> <--
>> >>  (!!!)  *** WARNING [still objects in 'Window Pool'] *** [object.c:241
>> >> in fusion_object_pool_destroy()] (!!!)  *** WARNING [still objects in
>> >> 'Layer Region Pool'] ***
>> >> [object.c:241 in fusion_object_pool_destroy()]
>> >>  (!!!)  *** WARNING [still objects in 'Layer Context Pool'] ***
>> >> [object.c:241 in fusion_object_pool_destroy()]
>> >>  (!!!)  *** WARNING [still objects in 'Surface Pool'] ***
>> >> [object.c:241 in fusion_object_pool_destroy()]
>> >> Aborted
>> >>
>> >> Can anyone share any experiences on this?
>> >
>> > Did you re-start with a clean GTK after you switch from Xorg to DirectFB?
>> > Sometimes the package's buildsystem does not re-compile everything after
>> > changing some parameters, so some object files are still compiled for the
>> > previous parameter settings. Just an idea.
>>
>> No, I didn't. I assumed ptxdist would take care.
>
> PTXdist: yes. It calls again the package's configure script with the new
> settings. But sometimes the package's buildsystem is broken and it does not
> re-compile all objects in accordance to the new configuration.

I see.

>
>> I'll give this a shot, rebuilding everything from scratch.
>
> Not everything: a "ptxdist clean gtk" should be enough. This means only the
> gtk package starts from scratch.

(Too late :)
Anyway I rebuilt everything from scratch and the problem persists.

If someone has tested this (GTK with DirectFB target), pointers are welcome.

Thanks,
-- 
Guillermo Rodriguez Garcia
guille.rodriguez@gmail.com

-- 
ptxdist mailing list
ptxdist@pengutronix.de