On Wed, Jan 29, 2014 at 11:52:55AM -0800, Keith Packard wrote: > Hans de Goede <[email protected]> writes: > > > As for the worries of drivers not being prepared to deal with this, > > systemd_logind_take_fd will only succeed for evdev devices, and AFAIK we've > > only 2 drivers for those evdev and synaptics. I'll make sure I've changes > > for those ready to merge as soon as everyone is happy with the server > > bits. > > On a procedural level, one thing we've discussed in the past is whether > we shouldn't just merge the input driver repositories back into the X > server. That would let us more easily mutate the API/ABI as needed to > resolve issues like this.
it wouldn't help much. we have 5 drivers that really matter, evdev, keyboard, mouse, synaptics and wacom. the latter is GPL, so merging it is questionable, and it is also the one that sees the most changes. And while you have one important driver outside the tree, you can't mutate the ABI at random anyway. For several server releases few changes in the drivers have actually required API changes. Merging the drivers back in is a lot of work for currently little benefit. Cheers, Peter > Eric and I were chatting about this as his work on Glamor, along with > Dave's work on xf86-video-modesetting may shortly provide > device-independent output that could be merged into the X server to > reduce the cost of making changes across the system. > > -- > [email protected] _______________________________________________ [email protected]: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
