Enlightenment

Try to get more log output with: EINA_LOG_LEVELS="eina_module:5" (I
actually don't remember the number, but 5 is big enough).

On Thu, Feb 28, 2013 at 7:41 PM, Andreas Volz <lists@...> wrote:
> Hello,
>
> I traced my module loader and I get in eina_module_new in line
>
> at DBG("m=%p, file=%s", m, file);
>
> this output:
> m=0x5115eb20, file=eina_module_new
>
> and in eina_module_load(Eina_Module *m)
>
> at DBG("m=%p, handle=%p, file=%s, refs=%d", m, m->handle, m->file,
> m->ref);
>
> m=0x5115eb10, handle=0x50d14f30, file=lib/eina/eina_module.c, refs=4
>
> and at WRN("could not dlopen(\"%s\", RTLD_NOW): %s", m->file,
> dlerror());
>
> could not dlopen("H�/", RTLD_NOW): eina_module_load
>
> For me this sounds as something went really wrong... :-)
>
> I traced it back and I call _ecore_evas_engine_load("android");
>
> and direct in the first line in _ecore_evas_engine_load I do:
>
> DBG("engine (av): %s", engine);
>
> and get: engine (av): ^TY�P0^G�L
>
> Sounds even more strange. Seems I broke something really serious. Any
> hints for a (today) blind one... :-)
>
> regards
>         Andreas
>
> --
> Technical Blog <http://andreasvolz.wordpress.com/>
>
> ------------------------------------------------------------------------------
> Everyone hates slow websites. So do we.
> Make your web apps faster with AppDynamics
> Download AppDynamics Lite for free today:
> http://p.sf.net/sfu/appdyn_d2d_feb
> _______________________________________________
> enlightenment-devel mailing list
> enlightenment-devel@...
> https://lists.sourceforge.net/lists/listinfo/enlightenment-devel



--
Rafael Antognolli
http://antognolli.org/

Am Fri, 1 Mar 2013 09:45:22 -0300 schrieb Rafael Antognolli:

> Try to get more log output with: EINA_LOG_LEVELS="eina_module:5" (I
> actually don't remember the number, but 5 is big enough).

I just uploaded complete logging here:

http://www.pasteall.org/40137

To summarize what I do is:

   ecore_evas_init();
   edje_init();
   ecore_evas_software_android_new();

and

EAPI Ecore_Evas *
ecore_evas_software_android_new(void *android_bitmap_info, int w, int h)
{
   Ecore_Evas *(*new)(void*, int, int);
    
  DBG("+engine (av): ANDROID"); 
   Eina_Module *m = _ecore_evas_engine_load("android");
  DBG("-engine (av): ANDROID");
   
   EINA_SAFETY_ON_NULL_RETURN_VAL(m, NULL);
   
   new = eina_module_symbol_get(m, "ecore_evas_software_android_new_internal");
   EINA_SAFETY_ON_NULL_RETURN_VAL(new, NULL);
   
   return new(android_bitmap_info, w, h);
}

and as I said traces in _ecore_evas_engine_load doesn't even print out "android"
string. This is somehow strange...

> On Thu, Feb 28, 2013 at 7:41 PM, Andreas Volz <lists@...>
> wrote:
> > Hello,
> >
> > I traced my module loader and I get in eina_module_new in line
> >
> > at DBG("m=%p, file=%s", m, file);
> >
> > this output:
> > m=0x5115eb20, file=eina_module_new
> >
> > and in eina_module_load(Eina_Module *m)
> >
> > at DBG("m=%p, handle=%p, file=%s, refs=%d", m, m->handle, m->file,
> > m->ref);
> >
> > m=0x5115eb10, handle=0x50d14f30, file=lib/eina/eina_module.c, refs=4
> >
> > and at WRN("could not dlopen(\"%s\", RTLD_NOW): %s", m->file,
> > dlerror());
> >
> > could not dlopen("H�/", RTLD_NOW): eina_module_load
> >
> > For me this sounds as something went really wrong... :-)
> >
> > I traced it back and I call _ecore_evas_engine_load("android");
> >
> > and direct in the first line in _ecore_evas_engine_load I do:
> >
> > DBG("engine (av): %s", engine);
> >
> > and get: engine (av): ^TY�P0^G�L
> >
> > Sounds even more strange. Seems I broke something really serious.
> > Any hints for a (today) blind one... :-)
> >
> > regards
> >         Andreas
> >
> > --
> > Technical Blog <http://andreasvolz.wordpress.com/>
> >
> > ------------------------------------------------------------------------------
> > Everyone hates slow websites. So do we.
> > Make your web apps faster with AppDynamics
> > Download AppDynamics Lite for free today:
> > http://p.sf.net/sfu/appdyn_d2d_feb
> > _______________________________________________
> > enlightenment-devel mailing list
> > enlightenment-devel@...
> > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel
> 
> 
> 
> --
> Rafael Antognolli
> http://antognolli.org/
> 
> ------------------------------------------------------------------------------
> Everyone hates slow websites. So do we.
> Make your web apps faster with AppDynamics
> Download AppDynamics Lite for free today:
> http://p.sf.net/sfu/appdyn_d2d_feb
> _______________________________________________
> enlightenment-devel mailing list
> enlightenment-devel@...
> https://lists.sourceforge.net/lists/listinfo/enlightenment-devel


-- 
Technical Blog <http://andreasvolz.wordpress.com/>

Am Fri, 1 Mar 2013 21:25:28 +0100 schrieb Andreas Volz:

> Am Fri, 1 Mar 2013 09:45:22 -0300 schrieb Rafael Antognolli:
> 
> > Try to get more log output with: EINA_LOG_LEVELS="eina_module:5" (I
> > actually don't remember the number, but 5 is big enough).
> 
> I just uploaded complete logging here:
> 
> http://www.pasteall.org/40137
> 
> To summarize what I do is:
> 
>    ecore_evas_init();
>    edje_init();
>    ecore_evas_software_android_new();
> 
> and
> 
> EAPI Ecore_Evas *
> ecore_evas_software_android_new(void *android_bitmap_info, int w, int
> h) {
>    Ecore_Evas *(*new)(void*, int, int);
>     
>   DBG("+engine (av): ANDROID"); 
>    Eina_Module *m = _ecore_evas_engine_load("android");
>   DBG("-engine (av): ANDROID");
>    
>    EINA_SAFETY_ON_NULL_RETURN_VAL(m, NULL);
>    
>    new = eina_module_symbol_get(m,
> "ecore_evas_software_android_new_internal");
> EINA_SAFETY_ON_NULL_RETURN_VAL(new, NULL); 
>    return new(android_bitmap_info, w, h);
> }
> 
> and as I said traces in _ecore_evas_engine_load doesn't even print
> out "android" string. This is somehow strange...

It was just eina_log_print_cb_set() which did a wrong va_arg handling
and so all debug prints were broken. I was nearly about to went
crazy. Now everything looks nice. Sometimes if it looks really strange
one has to verify the debug function itself :-)

-- 
Technical Blog <http://andreasvolz.wordpress.com/>

Am Fri, 1 Mar 2013 23:04:33 +0100 schrieb Andreas Volz:

> Am Fri, 1 Mar 2013 21:25:28 +0100 schrieb Andreas Volz:
> 
> > Am Fri, 1 Mar 2013 09:45:22 -0300 schrieb Rafael Antognolli:
> > 
> > > Try to get more log output with:
> > > EINA_LOG_LEVELS="eina_module:5" (I actually don't remember the
> > > number, but 5 is big enough).
> > 
> > I just uploaded complete logging here:
> > 
> > http://www.pasteall.org/40137
> > 
> > To summarize what I do is:
> > 
> >    ecore_evas_init();
> >    edje_init();
> >    ecore_evas_software_android_new();
> > 
> > and
> > 
> > EAPI Ecore_Evas *
> > ecore_evas_software_android_new(void *android_bitmap_info, int w,
> > int h) {
> >    Ecore_Evas *(*new)(void*, int, int);
> >     
> >   DBG("+engine (av): ANDROID"); 
> >    Eina_Module *m = _ecore_evas_engine_load("android");
> >   DBG("-engine (av): ANDROID");
> >    
> >    EINA_SAFETY_ON_NULL_RETURN_VAL(m, NULL);
> >    
> >    new = eina_module_symbol_get(m,
> > "ecore_evas_software_android_new_internal");
> > EINA_SAFETY_ON_NULL_RETURN_VAL(new, NULL); 
> >    return new(android_bitmap_info, w, h);
> > }
> > 
> > and as I said traces in _ecore_evas_engine_load doesn't even print
> > out "android" string. This is somehow strange...
> 
> It was just eina_log_print_cb_set() which did a wrong va_arg handling
> and so all debug prints were broken. I was nearly about to went
> crazy. Now everything looks nice. Sometimes if it looks really strange
> one has to verify the debug function itself :-)

Now result looks much better:

http://www.pasteall.org/40139

But the new engine still crashes while startup and ecore_job shows error
(line 214ff)

Has anyone experienced similar problems?

regards
	Andreas

-- 
Technical Blog <http://andreasvolz.wordpress.com/>