[Mesa-dev] [PATCH v2 8/8] egl: add EGL_platform_device support

[Mathias Fröhlich]

> What is missing for merging this?

I saw the pbuffer swrast crash and proposed to disable them via the
3rd patch that I pushed into my branch for you.
Emil never responded to that proposal directly. In one mail he mentioned
that he wanted to care for swrast to cope with the pbuffers. So I hoped that
he continues the work he mentioned...

For me it would be ok to squash patch #2 and #3 into one and for now
disable pbuffers on swrast from device.

What I don't want to have is mesa crashing on the egl pbuffer on a swrast
device as crashing is not an option.

best

Mathias

> 
> Thanks,
> Marek
> 
> On Wed, Apr 3, 2019 at 12:30 AM Mathias Fröhlich <Mathias.Froehlich at gmx.net>
> wrote:
> 
> > Marek,
> >
> > On Tuesday, 2 April 2019 23:07:50 CEST Marek Olšák wrote:
> > > Do you have a branch with patch 7/8 and 8/8? I'm interested in
> > > EGL_EXT_platform_device on radeonsi.
> >
> > You can take a look at the egl-device-4 branch on
> > https://gitlab.freedesktop.org/frohlich/mesa.git
> > I pushed a rebased and slightly changed wip for you.
> >
> > I have also some more tests in my local piglit as well as separate
> > tests/examples that I initially wrote as an example to make use of the
> > extension.
> > But I have to dig out these first...
> >
> > best
> >
> > Mathias
> >
> > >
> > > Thanks,
> > > Marek
> > >
> > > On Wed, Oct 3, 2018 at 4:36 AM Mathias Fröhlich <
> > Mathias.Froehlich at gmx.net>
> > > wrote:
> > >
> > > > Hi Emil,
> > > >
> > > > Ok, thanks for picking that up.
> > > >
> > > > On Tuesday, 2 October 2018 12:23:30 CEST Emil Velikov wrote:
> > > > > On Thu, 20 Sep 2018 at 15:13, Mathias Fröhlich
> > > > > <Mathias.Froehlich at gmx.net> wrote:
> > > > >
> > > > > >
> > > > > > If I replace the above with
> > > > > >
> > > > > >          EGLint surface_type = 0;
> > > > > >          /* Only advertise pbuffer configs for non swrast devices
> > */
> > > > > >          if (dri2_dpy->image_driver)
> > > > > >             surface_type = EGL_PBUFFER_BIT;
> > > > > >
> > > > > >          dri2_conf = dri2_add_config(dpy,
> > dri2_dpy->driver_configs[i],
> > > > > >                config_count + 1, surface_type, NULL,
> > > > > >                visuals[j].rgba_masks);
> > > > > >
> > > > > > then I can easily prohibit the crash that I mentioned when trying
> > to
> > > > > > create a pbuffer surface on the swrast device.
> > > > > > At least I do no longer get a valid pbuffer config from
> > eglChooseConfig
> > > > > > and without that we cannot reach the crashing
> > > > > > dri2_dpy->image_driver->createNewDrawable
> > > > > > call somewhere from eglCreatePbufferSurface anymore.
> > > > > >
> > > > > > Still getting a surface less context on swrast should work...
> > > > > >
> > > > > Issue is that we do not know how to create a "pbuffer only" swrast.
> > > > >
> > > > > Hence one resolves to hacks like the ones we have in
> > > > platform_surfaceless.
> > > > > Effectively pilling hacks upon hacks - see swrast_loader_extensions
> > > > > and "software path w/o DRM.".
> > > > >
> > > > > That said, I'm working on proper solution but since it will take some
> > > > > time to finish/merge I'll drop this and 7/8 for now.
> > > >
> > > >
> > > > That's fine too.
> > > >
> > > > What I wanted to avoid is that people using the extension correctly
> > fail
> > > > with a
> > > > crash in the application when running on mesa.
> > > > Think about it. Thats really bad from an application writers point of
> > view
> > > > as you do all right with checking extensions and that and then the
> > opengl
> > > > stack suddenly crashes.
> > > >
> > > > best
> > > >
> > > > Mathias
> > > >
> > > >
> > > > _______________________________________________
> > > > mesa-dev mailing list
> > > > mesa-dev at lists.freedesktop.org
> > > > https://lists.freedesktop.org/mailman/listinfo/mesa-dev
> > > >
> > >
> >
> >
> >
> >
> >
>