From: Gustavo C. <gl...@pr...> - 2012-01-27 13:52:13
|
Those last commits on e break it completely for people using it with x11_gl. On Thu, Jan 26, 2012 at 4:28 PM, Enlightenment SVN < no-...@en...> wrote: > Log: > E Comp (wayland): Fix a few more instances of > ecore_x_composite_name_window_pixmap_get when building w/ wayland > client support. > > > > Author: devilhorns > Date: 2012-01-26 10:28:15 -0800 (Thu, 26 Jan 2012) > New Revision: 67556 > Trac: http://trac.enlightenment.org/e/changeset/67556 > > Modified: > trunk/e/src/modules/comp/e_mod_comp.c > > Modified: trunk/e/src/modules/comp/e_mod_comp.c > =================================================================== > --- trunk/e/src/modules/comp/e_mod_comp.c 2012-01-26 18:24:00 UTC > (rev 67555) > +++ trunk/e/src/modules/comp/e_mod_comp.c 2012-01-26 18:28:15 UTC > (rev 67556) > @@ -598,6 +598,9 @@ > if ((cw->c->gl) && (_comp_mod->conf->texture_from_pixmap) && > (!cw->shaped) && (!cw->rects)) > { > +#ifdef HAVE_WAYLAND > + DBG("DEBUG - pm now %x\n", e_mod_comp_wl_pixmap_get(cw->win)); > +#endif > DBG("DEBUG - pm now %x\n", > ecore_x_composite_name_window_pixmap_get(cw->win)); > evas_object_image_size_set(cw->obj, cw->pw, cw->ph); > EINA_LIST_FOREACH(cw->obj_mirror, l, o) > @@ -1114,7 +1117,12 @@ > printf("^^^^ redirect2 %x\n", cw->win); > printf(" redr\n"); > ecore_x_composite_redirect_window(cw->win, > ECORE_X_COMPOSITE_UPDATE_MANUAL); > - cw->pixmap = > ecore_x_composite_name_window_pixmap_get(cw->win); > + > +#ifdef HAVE_WAYLAND > + cw->pixmap = e_mod_comp_wl_pixmap_get(cw->win); > +#endif > + if (!cw->pixmap) > + cw->pixmap = > ecore_x_composite_name_window_pixmap_get(cw->win); > if (cw->pixmap) > { > ecore_x_pixmap_geometry_get(cw->pixmap, NULL, > NULL, &(cw->pw), &(cw->ph)); > @@ -2052,38 +2060,42 @@ > cw->dmg_updates = 1; > if ((!cw->redirected) || (!cw->pixmap)) > { > -// we redirect all subwindows anyway > -// ecore_x_composite_redirect_window(cw->win, > ECORE_X_COMPOSITE_UPDATE_MANUAL); > - cw->pixmap = > ecore_x_composite_name_window_pixmap_get(cw->win); > - if (cw->pixmap) > - { > - ecore_x_pixmap_geometry_get(cw->pixmap, NULL, NULL, > &(cw->pw), &(cw->ph)); > - _e_mod_comp_win_ready_timeout_setup(cw); > - } > - else > - { > - cw->pw = 0; > - cw->ph = 0; > - } > - if ((cw->pw <= 0) || (cw->ph <= 0)) > - { > - if (cw->pixmap) > - { > - ecore_x_pixmap_free(cw->pixmap); > - cw->pixmap = 0; > - } > -// cw->show_ready = 0; // hmm maybe not needed? > - } > - cw->redirected = 1; > - DBG(" [0x%x] up resize %ix%i\n", cw->win, cw->pw, cw->ph); > - e_mod_comp_update_resize(cw->up, cw->pw, cw->ph); > - e_mod_comp_update_add(cw->up, 0, 0, cw->pw, cw->ph); > - evas_object_image_size_set(cw->obj, cw->pw, cw->ph); > - EINA_LIST_FOREACH(cw->obj_mirror, l, o) > - { > - evas_object_image_size_set(o, cw->pw, cw->ph); > - } > - ecore_x_e_comp_pixmap_set(cw->win, cw->pixmap); > + // we redirect all subwindows anyway > + // ecore_x_composite_redirect_window(cw->win, > ECORE_X_COMPOSITE_UPDATE_MANUAL); > +#ifdef HAVE_WAYLAND > + cw->pixmap = e_mod_comp_wl_pixmap_get(cw->win); > +#endif > + if (!cw->pixmap) > + cw->pixmap = ecore_x_composite_name_window_pixmap_get(cw->win); > + if (cw->pixmap) > + { > + ecore_x_pixmap_geometry_get(cw->pixmap, NULL, NULL, > &(cw->pw), &(cw->ph)); > + _e_mod_comp_win_ready_timeout_setup(cw); > + } > + else > + { > + cw->pw = 0; > + cw->ph = 0; > + } > + if ((cw->pw <= 0) || (cw->ph <= 0)) > + { > + if (cw->pixmap) > + { > + ecore_x_pixmap_free(cw->pixmap); > + cw->pixmap = 0; > + } > + // cw->show_ready = 0; // hmm maybe not needed? > + } > + cw->redirected = 1; > + DBG(" [0x%x] up resize %ix%i\n", cw->win, cw->pw, cw->ph); > + e_mod_comp_update_resize(cw->up, cw->pw, cw->ph); > + e_mod_comp_update_add(cw->up, 0, 0, cw->pw, cw->ph); > + evas_object_image_size_set(cw->obj, cw->pw, cw->ph); > + EINA_LIST_FOREACH(cw->obj_mirror, l, o) > + { > + evas_object_image_size_set(o, cw->pw, cw->ph); > + } > + ecore_x_e_comp_pixmap_set(cw->win, cw->pixmap); > } > if ((cw->dmg_updates >= 1) && (cw->show_ready)) > { > > > > ------------------------------------------------------------------------------ > Keep Your Developer Skills Current with LearnDevNow! > The most comprehensive online learning library for Microsoft developers > is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, > Metro Style Apps, more. Free future releases when you subscribe now! > http://p.sf.net/sfu/learndevnow-d2d > _______________________________________________ > enlightenment-svn mailing list > enl...@li... > https://lists.sourceforge.net/lists/listinfo/enlightenment-svn > -- Gustavo Lima Chaves Senior Developer ProFUSION embedded systems http://profusion.mobi |
From: Cedric B. <ced...@fr...> - 2012-01-27 14:57:23
|
On Fri, Jan 27, 2012 at 2:51 PM, Gustavo Chaves <gl...@pr...> wrote: > Those last commits on e break it completely for people using it with x11_gl. Actually it just broke all composite, software included. Disabling wayland client doesn't solve the issue either. > On Thu, Jan 26, 2012 at 4:28 PM, Enlightenment SVN < > no-...@en...> wrote: > >> Log: >> E Comp (wayland): Fix a few more instances of >> ecore_x_composite_name_window_pixmap_get when building w/ wayland >> client support. >> >> >> >> Author: devilhorns >> Date: 2012-01-26 10:28:15 -0800 (Thu, 26 Jan 2012) >> New Revision: 67556 >> Trac: http://trac.enlightenment.org/e/changeset/67556 >> >> Modified: >> trunk/e/src/modules/comp/e_mod_comp.c >> >> Modified: trunk/e/src/modules/comp/e_mod_comp.c >> =================================================================== >> --- trunk/e/src/modules/comp/e_mod_comp.c 2012-01-26 18:24:00 UTC >> (rev 67555) >> +++ trunk/e/src/modules/comp/e_mod_comp.c 2012-01-26 18:28:15 UTC >> (rev 67556) >> @@ -598,6 +598,9 @@ >> if ((cw->c->gl) && (_comp_mod->conf->texture_from_pixmap) && >> (!cw->shaped) && (!cw->rects)) >> { >> +#ifdef HAVE_WAYLAND >> + DBG("DEBUG - pm now %x\n", e_mod_comp_wl_pixmap_get(cw->win)); >> +#endif >> DBG("DEBUG - pm now %x\n", >> ecore_x_composite_name_window_pixmap_get(cw->win)); >> evas_object_image_size_set(cw->obj, cw->pw, cw->ph); >> EINA_LIST_FOREACH(cw->obj_mirror, l, o) >> @@ -1114,7 +1117,12 @@ >> printf("^^^^ redirect2 %x\n", cw->win); >> printf(" redr\n"); >> ecore_x_composite_redirect_window(cw->win, >> ECORE_X_COMPOSITE_UPDATE_MANUAL); >> - cw->pixmap = >> ecore_x_composite_name_window_pixmap_get(cw->win); >> + >> +#ifdef HAVE_WAYLAND >> + cw->pixmap = e_mod_comp_wl_pixmap_get(cw->win); >> +#endif >> + if (!cw->pixmap) >> + cw->pixmap = >> ecore_x_composite_name_window_pixmap_get(cw->win); >> if (cw->pixmap) >> { >> ecore_x_pixmap_geometry_get(cw->pixmap, NULL, >> NULL, &(cw->pw), &(cw->ph)); >> @@ -2052,38 +2060,42 @@ >> cw->dmg_updates = 1; >> if ((!cw->redirected) || (!cw->pixmap)) >> { >> -// we redirect all subwindows anyway >> -// ecore_x_composite_redirect_window(cw->win, >> ECORE_X_COMPOSITE_UPDATE_MANUAL); >> - cw->pixmap = >> ecore_x_composite_name_window_pixmap_get(cw->win); >> - if (cw->pixmap) >> - { >> - ecore_x_pixmap_geometry_get(cw->pixmap, NULL, NULL, >> &(cw->pw), &(cw->ph)); >> - _e_mod_comp_win_ready_timeout_setup(cw); >> - } >> - else >> - { >> - cw->pw = 0; >> - cw->ph = 0; >> - } >> - if ((cw->pw <= 0) || (cw->ph <= 0)) >> - { >> - if (cw->pixmap) >> - { >> - ecore_x_pixmap_free(cw->pixmap); >> - cw->pixmap = 0; >> - } >> -// cw->show_ready = 0; // hmm maybe not needed? >> - } >> - cw->redirected = 1; >> - DBG(" [0x%x] up resize %ix%i\n", cw->win, cw->pw, cw->ph); >> - e_mod_comp_update_resize(cw->up, cw->pw, cw->ph); >> - e_mod_comp_update_add(cw->up, 0, 0, cw->pw, cw->ph); >> - evas_object_image_size_set(cw->obj, cw->pw, cw->ph); >> - EINA_LIST_FOREACH(cw->obj_mirror, l, o) >> - { >> - evas_object_image_size_set(o, cw->pw, cw->ph); >> - } >> - ecore_x_e_comp_pixmap_set(cw->win, cw->pixmap); >> + // we redirect all subwindows anyway >> + // ecore_x_composite_redirect_window(cw->win, >> ECORE_X_COMPOSITE_UPDATE_MANUAL); >> +#ifdef HAVE_WAYLAND >> + cw->pixmap = e_mod_comp_wl_pixmap_get(cw->win); >> +#endif >> + if (!cw->pixmap) >> + cw->pixmap = ecore_x_composite_name_window_pixmap_get(cw->win); >> + if (cw->pixmap) >> + { >> + ecore_x_pixmap_geometry_get(cw->pixmap, NULL, NULL, >> &(cw->pw), &(cw->ph)); >> + _e_mod_comp_win_ready_timeout_setup(cw); >> + } >> + else >> + { >> + cw->pw = 0; >> + cw->ph = 0; >> + } >> + if ((cw->pw <= 0) || (cw->ph <= 0)) >> + { >> + if (cw->pixmap) >> + { >> + ecore_x_pixmap_free(cw->pixmap); >> + cw->pixmap = 0; >> + } >> + // cw->show_ready = 0; // hmm maybe not needed? >> + } >> + cw->redirected = 1; >> + DBG(" [0x%x] up resize %ix%i\n", cw->win, cw->pw, cw->ph); >> + e_mod_comp_update_resize(cw->up, cw->pw, cw->ph); >> + e_mod_comp_update_add(cw->up, 0, 0, cw->pw, cw->ph); >> + evas_object_image_size_set(cw->obj, cw->pw, cw->ph); >> + EINA_LIST_FOREACH(cw->obj_mirror, l, o) >> + { >> + evas_object_image_size_set(o, cw->pw, cw->ph); >> + } >> + ecore_x_e_comp_pixmap_set(cw->win, cw->pixmap); >> } >> if ((cw->dmg_updates >= 1) && (cw->show_ready)) >> { >> >> >> >> ------------------------------------------------------------------------------ >> Keep Your Developer Skills Current with LearnDevNow! >> The most comprehensive online learning library for Microsoft developers >> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, >> Metro Style Apps, more. Free future releases when you subscribe now! >> http://p.sf.net/sfu/learndevnow-d2d >> _______________________________________________ >> enlightenment-svn mailing list >> enl...@li... >> https://lists.sourceforge.net/lists/listinfo/enlightenment-svn >> > > > > -- > Gustavo Lima Chaves > Senior Developer > > ProFUSION embedded systems > > http://profusion.mobi > ------------------------------------------------------------------------------ > Try before you buy = See our experts in action! > The most comprehensive online learning library for Microsoft developers > is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, > Metro Style Apps, more. Free future releases when you subscribe now! > http://p.sf.net/sfu/learndevnow-dev2 > _______________________________________________ > enlightenment-devel mailing list > enl...@li... > https://lists.sourceforge.net/lists/listinfo/enlightenment-devel > -- Cedric BAIL |
From: Gustavo S. B. <bar...@pr...> - 2012-01-27 15:08:32
|
On Fri, Jan 27, 2012 at 12:57 PM, Cedric BAIL <ced...@fr...> wrote: > On Fri, Jan 27, 2012 at 2:51 PM, Gustavo Chaves <gl...@pr...> wrote: >> Those last commits on e break it completely for people using it with x11_gl. > > Actually it just broke all composite, software included. Disabling > wayland client doesn't solve the issue either. > Who is nearby devilhorns to spank him up to his horns are gone?! :-) -- Gustavo Sverzut Barbieri http://profusion.mobi embedded systems -------------------------------------- MSN: bar...@gm... Skype: gsbarbieri Mobile: +55 (19) 9225-2202 |
From: Christopher M. <cpm...@co...> - 2012-01-27 15:08:33
|
Fixed in 67568. I just disable the recent changes (for now), so I can rewrite them in such a way that it doesn't break composite for others. dh On 01/27/12 09:57, Cedric BAIL wrote: > On Fri, Jan 27, 2012 at 2:51 PM, Gustavo Chaves<gl...@pr...> wrote: >> Those last commits on e break it completely for people using it with x11_gl. > > Actually it just broke all composite, software included. Disabling > wayland client doesn't solve the issue either. > >> On Thu, Jan 26, 2012 at 4:28 PM, Enlightenment SVN< >> no-...@en...> wrote: >> >>> Log: >>> E Comp (wayland): Fix a few more instances of >>> ecore_x_composite_name_window_pixmap_get when building w/ wayland >>> client support. >>> >>> >>> >>> Author: devilhorns >>> Date: 2012-01-26 10:28:15 -0800 (Thu, 26 Jan 2012) >>> New Revision: 67556 >>> Trac: http://trac.enlightenment.org/e/changeset/67556 >>> >>> Modified: >>> trunk/e/src/modules/comp/e_mod_comp.c >>> >>> Modified: trunk/e/src/modules/comp/e_mod_comp.c >>> =================================================================== >>> --- trunk/e/src/modules/comp/e_mod_comp.c 2012-01-26 18:24:00 UTC >>> (rev 67555) >>> +++ trunk/e/src/modules/comp/e_mod_comp.c 2012-01-26 18:28:15 UTC >>> (rev 67556) >>> @@ -598,6 +598,9 @@ >>> if ((cw->c->gl)&& (_comp_mod->conf->texture_from_pixmap)&& >>> (!cw->shaped)&& (!cw->rects)) >>> { >>> +#ifdef HAVE_WAYLAND >>> + DBG("DEBUG - pm now %x\n", e_mod_comp_wl_pixmap_get(cw->win)); >>> +#endif >>> DBG("DEBUG - pm now %x\n", >>> ecore_x_composite_name_window_pixmap_get(cw->win)); >>> evas_object_image_size_set(cw->obj, cw->pw, cw->ph); >>> EINA_LIST_FOREACH(cw->obj_mirror, l, o) >>> @@ -1114,7 +1117,12 @@ >>> printf("^^^^ redirect2 %x\n", cw->win); >>> printf(" redr\n"); >>> ecore_x_composite_redirect_window(cw->win, >>> ECORE_X_COMPOSITE_UPDATE_MANUAL); >>> - cw->pixmap = >>> ecore_x_composite_name_window_pixmap_get(cw->win); >>> + >>> +#ifdef HAVE_WAYLAND >>> + cw->pixmap = e_mod_comp_wl_pixmap_get(cw->win); >>> +#endif >>> + if (!cw->pixmap) >>> + cw->pixmap = >>> ecore_x_composite_name_window_pixmap_get(cw->win); >>> if (cw->pixmap) >>> { >>> ecore_x_pixmap_geometry_get(cw->pixmap, NULL, >>> NULL,&(cw->pw),&(cw->ph)); >>> @@ -2052,38 +2060,42 @@ >>> cw->dmg_updates = 1; >>> if ((!cw->redirected) || (!cw->pixmap)) >>> { >>> -// we redirect all subwindows anyway >>> -// ecore_x_composite_redirect_window(cw->win, >>> ECORE_X_COMPOSITE_UPDATE_MANUAL); >>> - cw->pixmap = >>> ecore_x_composite_name_window_pixmap_get(cw->win); >>> - if (cw->pixmap) >>> - { >>> - ecore_x_pixmap_geometry_get(cw->pixmap, NULL, NULL, >>> &(cw->pw),&(cw->ph)); >>> - _e_mod_comp_win_ready_timeout_setup(cw); >>> - } >>> - else >>> - { >>> - cw->pw = 0; >>> - cw->ph = 0; >>> - } >>> - if ((cw->pw<= 0) || (cw->ph<= 0)) >>> - { >>> - if (cw->pixmap) >>> - { >>> - ecore_x_pixmap_free(cw->pixmap); >>> - cw->pixmap = 0; >>> - } >>> -// cw->show_ready = 0; // hmm maybe not needed? >>> - } >>> - cw->redirected = 1; >>> - DBG(" [0x%x] up resize %ix%i\n", cw->win, cw->pw, cw->ph); >>> - e_mod_comp_update_resize(cw->up, cw->pw, cw->ph); >>> - e_mod_comp_update_add(cw->up, 0, 0, cw->pw, cw->ph); >>> - evas_object_image_size_set(cw->obj, cw->pw, cw->ph); >>> - EINA_LIST_FOREACH(cw->obj_mirror, l, o) >>> - { >>> - evas_object_image_size_set(o, cw->pw, cw->ph); >>> - } >>> - ecore_x_e_comp_pixmap_set(cw->win, cw->pixmap); >>> + // we redirect all subwindows anyway >>> + // ecore_x_composite_redirect_window(cw->win, >>> ECORE_X_COMPOSITE_UPDATE_MANUAL); >>> +#ifdef HAVE_WAYLAND >>> + cw->pixmap = e_mod_comp_wl_pixmap_get(cw->win); >>> +#endif >>> + if (!cw->pixmap) >>> + cw->pixmap = ecore_x_composite_name_window_pixmap_get(cw->win); >>> + if (cw->pixmap) >>> + { >>> + ecore_x_pixmap_geometry_get(cw->pixmap, NULL, NULL, >>> &(cw->pw),&(cw->ph)); >>> + _e_mod_comp_win_ready_timeout_setup(cw); >>> + } >>> + else >>> + { >>> + cw->pw = 0; >>> + cw->ph = 0; >>> + } >>> + if ((cw->pw<= 0) || (cw->ph<= 0)) >>> + { >>> + if (cw->pixmap) >>> + { >>> + ecore_x_pixmap_free(cw->pixmap); >>> + cw->pixmap = 0; >>> + } >>> + // cw->show_ready = 0; // hmm maybe not needed? >>> + } >>> + cw->redirected = 1; >>> + DBG(" [0x%x] up resize %ix%i\n", cw->win, cw->pw, cw->ph); >>> + e_mod_comp_update_resize(cw->up, cw->pw, cw->ph); >>> + e_mod_comp_update_add(cw->up, 0, 0, cw->pw, cw->ph); >>> + evas_object_image_size_set(cw->obj, cw->pw, cw->ph); >>> + EINA_LIST_FOREACH(cw->obj_mirror, l, o) >>> + { >>> + evas_object_image_size_set(o, cw->pw, cw->ph); >>> + } >>> + ecore_x_e_comp_pixmap_set(cw->win, cw->pixmap); >>> } >>> if ((cw->dmg_updates>= 1)&& (cw->show_ready)) >>> { >>> >>> >>> >>> ------------------------------------------------------------------------------ >>> Keep Your Developer Skills Current with LearnDevNow! >>> The most comprehensive online learning library for Microsoft developers >>> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, >>> Metro Style Apps, more. Free future releases when you subscribe now! >>> http://p.sf.net/sfu/learndevnow-d2d >>> _______________________________________________ >>> enlightenment-svn mailing list >>> enl...@li... >>> https://lists.sourceforge.net/lists/listinfo/enlightenment-svn >>> >> >> >> >> -- >> Gustavo Lima Chaves >> Senior Developer >> >> ProFUSION embedded systems >> >> http://profusion.mobi >> ------------------------------------------------------------------------------ >> Try before you buy = See our experts in action! >> The most comprehensive online learning library for Microsoft developers >> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3, >> Metro Style Apps, more. Free future releases when you subscribe now! >> http://p.sf.net/sfu/learndevnow-dev2 >> _______________________________________________ >> enlightenment-devel mailing list >> enl...@li... >> https://lists.sourceforge.net/lists/listinfo/enlightenment-devel >> > > > |
From: Christopher M. <cpm...@co...> - 2012-01-28 14:14:41
|
For those that were having composite problems with x11_gl because of this, can you provide more detail please ?? I ask because composite worked fine here (software_x11) even when this was enabled. dh On 01/27/12 10:08, Christopher Michael wrote: > Fixed in 67568. > > I just disable the recent changes (for now), so I can rewrite them in > such a way that it doesn't break composite for others. > > dh > > On 01/27/12 09:57, Cedric BAIL wrote: >> On Fri, Jan 27, 2012 at 2:51 PM, Gustavo Chaves<gl...@pr...> >> wrote: >>> Those last commits on e break it completely for people using it with >>> x11_gl. >> >> Actually it just broke all composite, software included. Disabling >> wayland client doesn't solve the issue either. >> >>> On Thu, Jan 26, 2012 at 4:28 PM, Enlightenment SVN< >>> no-...@en...> wrote: >>> >>>> Log: >>>> E Comp (wayland): Fix a few more instances of >>>> ecore_x_composite_name_window_pixmap_get when building w/ wayland >>>> client support. >>>> >>>> >>>> >>>> Author: devilhorns >>>> Date: 2012-01-26 10:28:15 -0800 (Thu, 26 Jan 2012) >>>> New Revision: 67556 >>>> Trac: http://trac.enlightenment.org/e/changeset/67556 >>>> >>>> Modified: >>>> trunk/e/src/modules/comp/e_mod_comp.c >>>> >>>> Modified: trunk/e/src/modules/comp/e_mod_comp.c >>>> =================================================================== >>>> --- trunk/e/src/modules/comp/e_mod_comp.c 2012-01-26 18:24:00 UTC >>>> (rev 67555) >>>> +++ trunk/e/src/modules/comp/e_mod_comp.c 2012-01-26 18:28:15 UTC >>>> (rev 67556) >>>> @@ -598,6 +598,9 @@ >>>> if ((cw->c->gl)&& (_comp_mod->conf->texture_from_pixmap)&& >>>> (!cw->shaped)&& (!cw->rects)) >>>> { >>>> +#ifdef HAVE_WAYLAND >>>> + DBG("DEBUG - pm now %x\n", e_mod_comp_wl_pixmap_get(cw->win)); >>>> +#endif >>>> DBG("DEBUG - pm now %x\n", >>>> ecore_x_composite_name_window_pixmap_get(cw->win)); >>>> evas_object_image_size_set(cw->obj, cw->pw, cw->ph); >>>> EINA_LIST_FOREACH(cw->obj_mirror, l, o) >>>> @@ -1114,7 +1117,12 @@ >>>> printf("^^^^ redirect2 %x\n", cw->win); >>>> printf(" redr\n"); >>>> ecore_x_composite_redirect_window(cw->win, >>>> ECORE_X_COMPOSITE_UPDATE_MANUAL); >>>> - cw->pixmap = >>>> ecore_x_composite_name_window_pixmap_get(cw->win); >>>> + >>>> +#ifdef HAVE_WAYLAND >>>> + cw->pixmap = e_mod_comp_wl_pixmap_get(cw->win); >>>> +#endif >>>> + if (!cw->pixmap) >>>> + cw->pixmap = >>>> ecore_x_composite_name_window_pixmap_get(cw->win); >>>> if (cw->pixmap) >>>> { >>>> ecore_x_pixmap_geometry_get(cw->pixmap, NULL, >>>> NULL,&(cw->pw),&(cw->ph)); >>>> @@ -2052,38 +2060,42 @@ >>>> cw->dmg_updates = 1; >>>> if ((!cw->redirected) || (!cw->pixmap)) >>>> { >>>> -// we redirect all subwindows anyway >>>> -// ecore_x_composite_redirect_window(cw->win, >>>> ECORE_X_COMPOSITE_UPDATE_MANUAL); >>>> - cw->pixmap = >>>> ecore_x_composite_name_window_pixmap_get(cw->win); >>>> - if (cw->pixmap) >>>> - { >>>> - ecore_x_pixmap_geometry_get(cw->pixmap, NULL, NULL, >>>> &(cw->pw),&(cw->ph)); >>>> - _e_mod_comp_win_ready_timeout_setup(cw); >>>> - } >>>> - else >>>> - { >>>> - cw->pw = 0; >>>> - cw->ph = 0; >>>> - } >>>> - if ((cw->pw<= 0) || (cw->ph<= 0)) >>>> - { >>>> - if (cw->pixmap) >>>> - { >>>> - ecore_x_pixmap_free(cw->pixmap); >>>> - cw->pixmap = 0; >>>> - } >>>> -// cw->show_ready = 0; // hmm maybe not needed? >>>> - } >>>> - cw->redirected = 1; >>>> - DBG(" [0x%x] up resize %ix%i\n", cw->win, cw->pw, cw->ph); >>>> - e_mod_comp_update_resize(cw->up, cw->pw, cw->ph); >>>> - e_mod_comp_update_add(cw->up, 0, 0, cw->pw, cw->ph); >>>> - evas_object_image_size_set(cw->obj, cw->pw, cw->ph); >>>> - EINA_LIST_FOREACH(cw->obj_mirror, l, o) >>>> - { >>>> - evas_object_image_size_set(o, cw->pw, cw->ph); >>>> - } >>>> - ecore_x_e_comp_pixmap_set(cw->win, cw->pixmap); >>>> + // we redirect all subwindows anyway >>>> + // ecore_x_composite_redirect_window(cw->win, >>>> ECORE_X_COMPOSITE_UPDATE_MANUAL); >>>> +#ifdef HAVE_WAYLAND >>>> + cw->pixmap = e_mod_comp_wl_pixmap_get(cw->win); >>>> +#endif >>>> + if (!cw->pixmap) >>>> + cw->pixmap = ecore_x_composite_name_window_pixmap_get(cw->win); >>>> + if (cw->pixmap) >>>> + { >>>> + ecore_x_pixmap_geometry_get(cw->pixmap, NULL, NULL, >>>> &(cw->pw),&(cw->ph)); >>>> + _e_mod_comp_win_ready_timeout_setup(cw); >>>> + } >>>> + else >>>> + { >>>> + cw->pw = 0; >>>> + cw->ph = 0; >>>> + } >>>> + if ((cw->pw<= 0) || (cw->ph<= 0)) >>>> + { >>>> + if (cw->pixmap) >>>> + { >>>> + ecore_x_pixmap_free(cw->pixmap); >>>> + cw->pixmap = 0; >>>> + } >>>> + // cw->show_ready = 0; // hmm maybe not needed? >>>> + } >>>> + cw->redirected = 1; >>>> + DBG(" [0x%x] up resize %ix%i\n", cw->win, cw->pw, cw->ph); >>>> + e_mod_comp_update_resize(cw->up, cw->pw, cw->ph); >>>> + e_mod_comp_update_add(cw->up, 0, 0, cw->pw, cw->ph); >>>> + evas_object_image_size_set(cw->obj, cw->pw, cw->ph); >>>> + EINA_LIST_FOREACH(cw->obj_mirror, l, o) >>>> + { >>>> + evas_object_image_size_set(o, cw->pw, cw->ph); >>>> + } >>>> + ecore_x_e_comp_pixmap_set(cw->win, cw->pixmap); >>>> } >>>> if ((cw->dmg_updates>= 1)&& (cw->show_ready)) >>>> { >>>> >>>> >>>> >>> >>> >>> >>> -- >>> Gustavo Lima Chaves >>> Senior Developer >>> >>> ProFUSION embedded systems >>> >>> http://profusion.mobi |
From: Cedric B. <ced...@fr...> - 2012-01-28 15:10:20
|
On Sat, Jan 28, 2012 at 3:14 PM, Christopher Michael <cpm...@co...> wrote: > For those that were having composite problems with x11_gl because of this, > can you provide more detail please ?? I ask because composite worked fine > here (software_x11) even when this was enabled. It wasn't working properly with both gl and software backend. Issue with software was random lag before correctly updating and displaying a window (almost fine with a small number of window). Gl was complete random content, most of the time black but sometime just garbage and sometime right. Hope it help. -- Cedric BAIL |
From: Christopher M. <cpm...@co...> - 2012-01-28 15:12:26
|
On 01/28/12 10:10, Cedric BAIL wrote: > On Sat, Jan 28, 2012 at 3:14 PM, Christopher Michael > <cpm...@co...> wrote: >> For those that were having composite problems with x11_gl because of this, >> can you provide more detail please ?? I ask because composite worked fine >> here (software_x11) even when this was enabled. > > It wasn't working properly with both gl and software backend. Issue > with software was random lag before correctly updating and displaying > a window (almost fine with a small number of window). Gl was complete > random content, most of the time black but sometime just garbage and > sometime right. Hope it help. Hmmm, very interesting. Thanks for the report :) I am in the process now of rebuilding the whole efl stack with various options (xcb, x11, gl, wayland, etc) and running tests to see if I get similar behavior. dh |
From: Christopher M. <cpm...@co...> - 2012-01-28 18:44:16
|
On 01/28/12 10:12, Christopher Michael wrote: > On 01/28/12 10:10, Cedric BAIL wrote: >> On Sat, Jan 28, 2012 at 3:14 PM, Christopher Michael >> <cpm...@co...> wrote: >>> For those that were having composite problems with x11_gl because of this, >>> can you provide more detail please ?? I ask because composite worked fine >>> here (software_x11) even when this was enabled. >> >> It wasn't working properly with both gl and software backend. Issue >> with software was random lag before correctly updating and displaying >> a window (almost fine with a small number of window). Gl was complete >> random content, most of the time black but sometime just garbage and >> sometime right. Hope it help. > > Hmmm, very interesting. Thanks for the report :) I am in the process now > of rebuilding the whole efl stack with various options (xcb, x11, gl, > wayland, etc) and running tests to see if I get similar behavior. > > dh > Cedric, Can you provide your build options for evas, ecore, & E please ? I've tried every variation of the build options, but I don't experience any problems with either software or gl rendering for composite module. Also, what video card & driver are you using ?? Thanks, dh |
From: Cedric B. <ced...@fr...> - 2012-01-28 21:24:20
|
On Sat, Jan 28, 2012 at 7:44 PM, Christopher Michael <cpm...@co...> wrote: > On 01/28/12 10:12, Christopher Michael wrote: >> On 01/28/12 10:10, Cedric BAIL wrote: >>> On Sat, Jan 28, 2012 at 3:14 PM, Christopher Michael >>> <cpm...@co...> wrote: >>>> For those that were having composite problems with x11_gl because of >>>> this, >>>> can you provide more detail please ?? I ask because composite worked >>>> fine >>>> here (software_x11) even when this was enabled. >>> >>> It wasn't working properly with both gl and software backend. Issue >>> with software was random lag before correctly updating and displaying >>> a window (almost fine with a small number of window). Gl was complete >>> random content, most of the time black but sometime just garbage and >>> sometime right. Hope it help. >> >> Hmmm, very interesting. Thanks for the report :) I am in the process now >> of rebuilding the whole efl stack with various options (xcb, x11, gl, >> wayland, etc) and running tests to see if I get similar behavior. > > Can you provide your build options for evas, ecore, & E please ? I've tried > every variation of the build options, but I don't experience any problems > with either software or gl rendering for composite module. Also, what video > card & driver are you using ?? Yes, I can ! So build option, evas : --enable-async-render --enable-software-sdl --enable-pipe-render --enable-gl-xlib ecore and e have nothing special, but eina has : --enable-on-off-threads --disable-magic-debug As for video card, I have an i7 with optimus. I only use Intel i915 driver for kernel part and intel_drv.so for the X.org part in version 2.14.0. I know that it's a really buggy driver, but it was working fine until your change. As a side note, I have noticed very recently that their is still a bug left with compositor in the locking screen. It take a lot of time to have the window content updated when I type the password, that wasn't the case a few days ago. Maybe it's related to your change. -- Cedric BAIL |
From: Christopher M. <cpm...@co...> - 2012-01-28 22:01:46
|
On 01/28/12 16:24, Cedric BAIL wrote: > On Sat, Jan 28, 2012 at 7:44 PM, Christopher Michael > <cpm...@co...> wrote: >> On 01/28/12 10:12, Christopher Michael wrote: >>> On 01/28/12 10:10, Cedric BAIL wrote: >>>> On Sat, Jan 28, 2012 at 3:14 PM, Christopher Michael >>>> <cpm...@co...> wrote: >>>>> For those that were having composite problems with x11_gl because of >>>>> this, >>>>> can you provide more detail please ?? I ask because composite worked >>>>> fine >>>>> here (software_x11) even when this was enabled. >>>> >>>> It wasn't working properly with both gl and software backend. Issue >>>> with software was random lag before correctly updating and displaying >>>> a window (almost fine with a small number of window). Gl was complete >>>> random content, most of the time black but sometime just garbage and >>>> sometime right. Hope it help. >>> >>> Hmmm, very interesting. Thanks for the report :) I am in the process now >>> of rebuilding the whole efl stack with various options (xcb, x11, gl, >>> wayland, etc) and running tests to see if I get similar behavior. >> >> Can you provide your build options for evas, ecore,& E please ? I've tried >> every variation of the build options, but I don't experience any problems >> with either software or gl rendering for composite module. Also, what video >> card& driver are you using ?? > > Yes, I can ! So build option, evas : > --enable-async-render --enable-software-sdl --enable-pipe-render > --enable-gl-xlib > ecore and e have nothing special, but eina has : > --enable-on-off-threads --disable-magic-debug > > As for video card, I have an i7 with optimus. I only use Intel i915 > driver for kernel part and intel_drv.so for the X.org part in version > 2.14.0. I know that it's a really buggy driver, but it was working > fine until your change. As a side note, I have noticed very recently > that their is still a bug left with compositor in the locking screen. > It take a lot of time to have the window content updated when I type > the password, that wasn't the case a few days ago. Maybe it's related > to your change. Thanks :) Well, wrt the bug in the locking screen ... I have no clue there. Nothing I was doing is related to the lock screen.... And as an aside note, all the stuff I was working on is #ifdef'd out for normal users unless you explicitly enable the wayland client support. If you don't, then the code path should follow the same path that WAS there before I started changing things. That's why this is so puzzling... dh |
From: Christopher M. <cpm...@co...> - 2012-01-28 22:49:31
|
On 01/28/12 17:01, Christopher Michael wrote: > On 01/28/12 16:24, Cedric BAIL wrote: >> On Sat, Jan 28, 2012 at 7:44 PM, Christopher Michael >> <cpm...@co...> wrote: >>> On 01/28/12 10:12, Christopher Michael wrote: >>>> On 01/28/12 10:10, Cedric BAIL wrote: >>>>> On Sat, Jan 28, 2012 at 3:14 PM, Christopher Michael >>>>> <cpm...@co...> wrote: >>>>>> For those that were having composite problems with x11_gl because of >>>>>> this, >>>>>> can you provide more detail please ?? I ask because composite worked >>>>>> fine >>>>>> here (software_x11) even when this was enabled. >>>>> >>>>> It wasn't working properly with both gl and software backend. Issue >>>>> with software was random lag before correctly updating and displaying >>>>> a window (almost fine with a small number of window). Gl was complete >>>>> random content, most of the time black but sometime just garbage and >>>>> sometime right. Hope it help. >>>> >>>> Hmmm, very interesting. Thanks for the report :) I am in the process >>>> now >>>> of rebuilding the whole efl stack with various options (xcb, x11, gl, >>>> wayland, etc) and running tests to see if I get similar behavior. >>> >>> Can you provide your build options for evas, ecore,& E please ? I've >>> tried >>> every variation of the build options, but I don't experience any >>> problems >>> with either software or gl rendering for composite module. Also, what >>> video >>> card& driver are you using ?? >> >> Yes, I can ! So build option, evas : >> --enable-async-render --enable-software-sdl --enable-pipe-render >> --enable-gl-xlib >> ecore and e have nothing special, but eina has : >> --enable-on-off-threads --disable-magic-debug >> >> As for video card, I have an i7 with optimus. I only use Intel i915 >> driver for kernel part and intel_drv.so for the X.org part in version >> 2.14.0. I know that it's a really buggy driver, but it was working >> fine until your change. As a side note, I have noticed very recently >> that their is still a bug left with compositor in the locking screen. >> It take a lot of time to have the window content updated when I type >> the password, that wasn't the case a few days ago. Maybe it's related >> to your change. > > Thanks :) > > Well, wrt the bug in the locking screen ... I have no clue there. > Nothing I was doing is related to the lock screen.... And as an aside > note, all the stuff I was working on is #ifdef'd out for normal users > unless you explicitly enable the wayland client support. If you don't, > then the code path should follow the same path that WAS there before I > started changing things. That's why this is so puzzling... > > dh Cedric, Ok, I think I have reduced the lagging issue quite a lot :) so going to do a commit now. Please let me know ASAP if there are still issues with this. I tested it here for software_x11 and gl_x11 without issue. Thanks, dh |