[vlc-devel] [PATCH 0/3] DRM-GBM mode support
thomas at gllm.fr
Thu Jun 22 18:25:17 CEST 2017
I think we can push the drm_gbm window but fails in the probe if
--drm-device is not set. Therefore, the user has to explicitly set the
drm device and this module won't be loaded otherwise.
On Wed, Jun 21, 2017, at 22:01, Rémi Denis-Courmont wrote:
> Le keskiviikkona 21. kesäkuuta 2017, 13.51.13 EEST Thomas Guillem a écrit
> > > But still GBM will have higher priority than non-windowed video
> > > outputs, rendering them mostly unusable. I am in favor of removing
> > > them (SDL in particular) or making them windowed somehow. But I
> > > don't think that others agree to that, and so that patchset would
> > > break stuff.
> > I would like to kill every non windowed vouts and/or make them depend on
> > a dummy vout_widow module. But maybe this should wait for 4.0 ?
> Leaving aside the actually dummy video output, I considered that. But I
> ´t come up with a way that that would actually work, especially in
> to KMS.
> First, the ASCII art libraries support X11, text mode, etc. But they
> provide no reasonable ways to direct the output to a specific window.
> Then SDL exhibits the same problem. And even if it did not, I can´t see
> functional benefit of SDL in VLC nowadays. Yet my recent request to
> remove it
> was denied.
> Lastly, LinuxFB ought to be superseded by KMS. It could maybe get a TTY
> pseudo-window type and provider. But it would probably become wholly
> as soon as DRM/KMS support is merged (if applicable).
> Postponing the removal of the unwindowed output support from the core
> like the only thing that might make sense to postpone until after
> But I don´t see any benefits to waiting to remove the affected plugins.
> would SDL and ASCII art be more dispensable in 4.0 than in 3.0 ?
> vlc-devel mailing list
> To unsubscribe or modify your subscription options:
More information about the vlc-devel