5 - dnd, figure out large object transfer: through wayland protocol or
6 pass an fd through the compositor to the other client and let them
9 - copy-n-paste, store data in server (only one mime-type available)
10 or do X style (content mime-type negotiation, but data goes away
13 - protocol for setting the cursor image
15 - should we have a mechanism to attach surface to cursor for
16 guaranteed non-laggy drag?
18 - drawing cursors, moving them, cursor themes, attaching surfaces
19 to cursors. How do you change cursors when you mouse over a
20 text field if you don't have subwindows? This is what we do: a
21 client can set a cursor for a surface and wayland will set that
22 on enter and revert to default on leave
24 - Discard buffer, as in "wayland discarded your buffer, it's no
25 longer visible, you can stop updating it now.", reattach, as in "oh
26 hey, I'm about to show your buffer that I threw away, what was it
27 again?". for wayland system compositor vt switcing, for example,
28 to be able to throw away the surfaces in the session we're
29 switching away from. for minimized windows that we don't want live
32 - Consolidate drm buffer upload with a create_buffer request, returns
33 buffer object we can use in surface.attach, cache.upload and
34 input.attach? Will increase object id usage significantly, each
35 buffer swap allocates and throws away a new id. Does consolidate
36 the details of a buffer very nicely though.
38 compositor.create_buffer(new_id, visual, name, stride, width, height)
40 surface.attach(buffer)
42 cache.upload(buffer, x, y, width, height)
44 input.set_cursor(buffer, hotspot_x, hotspot_y)
46 Doesn't increase id usage too much, can keep buffers around.
48 - Move/resize protocol in the style of the dnd protocol: a surface
49 who has a grabbed device can send a request to initiate a
50 resize(top/bottom+rigth/left) or a move. The compositor will then
51 resize or move the window and take into account windows, panels and
52 screen edges and constrain and snap the motion accordingly. As the
53 cursor moves, the compositor sends resize or move (maybe not move
54 events?) events to the app, which responds by attaching a new
55 surface at the new size (optionally, reducing the allocated space
56 to satisfy aspect ratio or resize increments).
58 - Initial placement of surfaces. Guess we can do, 1)
59 surface-relative (menus), 2) pointer-relative (tooltips and
60 right-click menus) or 3) server-decides (all other top-levels).
62 - When a surface is the size of the screen and on top, we can set the
63 scanout buffer to that surface directly. Like compiz unredirect
64 top-level window feature. Except it won't have any protocol state
65 side-effects and the client that owns the surface won't know. We
66 lose control of updates. Should work well for X server root window
67 under wayland. Should be possible for yuv overlays as well.
69 - what about cursors then? maybe use hw cursors if the cursor
70 satisfies hw limitations (64x64, only one cursor), switch to
71 composited cursors if not.
73 - clients needs to allocate the surface to be suitable for
74 scanout, which they can do whenever they go fullscreen.
76 - multihead, screen geometry and crtc layout protocol, hotplug
78 - input device discovery, hotplug
80 - Advertise axes as part of the discovery, use something like
81 "org.wayland.input.x" to identify the axes.
83 - keyboard state, layout events at connect time and when it
84 changes, keyboard leds
90 - synaptics, 3-button emulation, scim
92 - sparse/gcc plugin based idl compiler
96 - xml based description instead?
98 - actually make batch/commit batch up commands
100 - auth; We need to generate a random socket name and advertise that
101 on dbus along with a connection cookie. Something like a method
102 that returns the socket name and a connection cookie. The
103 connection cookie is just another random string that the client
104 must pass to the wayland server to become authenticated. The
105 Wayland server generates the cookie on demand when the dbus method
106 is called and expires it after 5s or so.
108 - or just pass the fd over dbus
110 - drm bo access control, authentication, flink_to
112 - Range protocol may not be sufficient... if a server cycles through
113 2^32 object IDs we don't have a way to handle wrapping. And since
114 we hand out a range of 256 IDs to each new clients, we're just
115 talking about 2^24 clients. That's 31 years with a new client
116 every minute... Maybe just use bigger ranges, then it's feasible
117 to track and garbage collect them when a client dies.
119 - Add protocol to let applications specify the effective/logical
120 surface rectangle, that is, the edge of the window, ignoring drop
121 shadows and other padding. The compositor needs this for snapping
122 and constraining window motion. Also, maybe communicate the opaque
123 region of the window (or just a conservative, simple estimate), to
124 let the compositor reduce overdraw.
126 - multi gpu, needs queue and seqno to wait on in requests
132 - eek, so much X legacy stuff there...
134 - draw window decorations in gtkwindow.c
136 - start from alexl's client-side-windows branch
138 - Details about pointer grabs. wayland doesn't have active grabs,
139 menus will behave subtly different. Under X, clicking a menu
140 open grabs the pointer and clicking outside the window pops down
141 the menu and swallows the click. without active grabs we can't
142 swallow the click. I'm sure there much more...
144 - Port Qt? There's already talk about this on the list.
148 - move most of the code from xf86-video-intel into a Xorg wayland
151 - don't ask KMS for available output and modes, use the info from
152 the wayland server. then stop mooching off of drmmode.c.
154 - map multiple wayland input devices to MPX in Xorg.
156 - rootless; avoid allocating and setting the front buffer, draw
157 window decorations in the X server (!), how to map input?
159 - gnome-shell as a wayland session compositor
161 - runs as a client of the wayland session compositor, uses
162 clutter+egl on wayland
164 - talks to an Xorg server as the compositing and window manager
165 for that server and renders the output to a wayland surface.
166 the Xorg server should be modified to take input from the system
167 compositor through gnome-shell, but not allocate a front buffer.
169 - make gnome-shell itself a nested wayland server and allow native
170 wayland clients to connect and can native wayland windows with
171 the windows from the X server.
173 - qemu as a wayland client; session surface as X case
175 - qemu has too simple acceleration, so a Wayland backend like the
176 SDL/VNC ones it has now is trivial.
178 - paravirt: forward wayland screen info as mmio, expose gem ioctls as mmio
180 - mapping vmem is tricky, should try to only use ioctl (pwrite+pread)
182 - not useful for Windows without a windows paravirt driver.
184 - two approaches: 1) do a toplevel qemu window, or 2) expose a
185 wayland server in the guest that forwards to the host wayland
186 server, ie a "remote" compositor, but with the gem buffers
187 shared. could do a wl_connection directly on mmio memory, with
188 head and tail pointers. use an alloc_head register to indicate
189 desired data to write, if it overwrites tail, block guest. just
190 a socket would be easier.
192 - moblin as a wayland compositor
194 - clutter as a wayland compositors