3 - surface.set_grab_mode(GRAB_OWNER_EVENTS vs GRAB_SURFACE_EVENTS), to
4 make menus work right: click and drag in a menubar grabs the
5 pointer to the menubar (which we need for detecting motion into
6 another menu item), but we need events for the popup menu surface
9 - The message format has to include information about number of fds
10 in the message so we can skip a message correctly. Or we should
11 just give up on trying to recover from unknown messages. We need
12 to make sure you never get a message from an interface you don't
13 know about (using per-client id space and subscribe) or include
14 information on number of fds, so marshalling logic can skip.
16 - generate pointer_focus (and drag focus) on raise/lower, move
17 windows, all kinds of changes in surface stacking.
21 buffer = drm.create_buffer(); /* buffer with stuff in it */
23 cache.upload(buffer, x, y, width, height, int hash)
25 drm.buffer: id, name, stride etc /* event to announce cache buffer */
27 cache.image: hash, buffer, x, y, stride /* event to announce
28 * location in cache */
30 cache.reject: hash /* no upload for you! */
32 cache.retire: buffer /* cache has stopped using buffer, please
33 * reupload whatever you had in that buffer */
37 Root window must send NULL type (to decline drop) or
38 x-wayland/root-something type if the source offers that. But the
39 target deletes the drag_offer object when drag.pointer_focus leaves
42 How do we animate the drag icon back to the drag origin in case of
45 How to handle surfaces from clients that don't know about dnd or
46 don't care? Maybe the dnd object should have a
47 dnd.register_surface() method so clients can opt-in the surfaces
48 that will participate in dnd. Or just assume client is not
49 participating until we receive an accept request.
51 - Selection/copy+paste issues: is it sufficient to only introduce the
52 selection offer when a client receives kb focus? Or maybe it is
53 actually a security feature? Clipboard manager in server for
56 - Pointer image issue:
58 - A direct touch input device (eg touch screen) doesn't have a
59 pointer; indicate that somehow.
61 - Cursor themes, tie in with glyph/image cache.
63 - Discard buffer, as in "wayland discarded your buffer, it's no
64 longer visible, you can stop updating it now.", reattach, as in "oh
65 hey, I'm about to show your buffer that I threw away, what was it
66 again?". for wayland system compositor vt switcing, for example,
67 to be able to throw away the surfaces in the session we're
68 switching away from. for minimized windows that we don't want live
71 - Import relevants bits from EWMH spec, stuff like window title,
72 window class, app groups, icons, lower window, need attention,
73 fullscreen (maybe different types of fullscreen).
75 - Per client id space. Each client has an entire 32 bit id namespace
76 to itself. On the server side, each struct wl_client has an object
77 hash table. Object announcements use a server id space and clients
78 must respond with subscribe request with a client id for the
79 object. Part of wl_proxy_create_for_id():
81 wl_display_subscribe(display, id, new_id, my_version);
85 wl_display_bind(display, id, new_id, my_version);
89 - Maps the global object into the client id space, lets client
90 allocate the id. All ids are allocated by the client this way,
91 which fixes the range protocol problem.
93 - Tells the server that the client is interested in events from
94 the object. Lets the server know that a client participates in a
95 certain protocol (like drag and drop), so the server can account
96 for whether or not the client is expected to reply
98 - Server emits initial object state event(s) in reponse to
99 receiving the subscribe request. Introduces an extra round trip
100 at initialization time, but the server will still announces all
101 objects in one burst and the client can subscribe in a burst as
104 - Separates client resources, since each client will have it's own
105 hash table. It's not longer possible to guess the id of another
106 surface and access it.
108 - Server must track the client id for each client an object is
109 exposed to. In some cases we know this (a surface is always
110 only owned by one client), in other cases it provides a way to
111 track who's interested in the object events. For input device
112 events, we can look up the client name when it receives pointer
113 focus or keyboard focus and cache it in the device.
115 - Server must know which id to send when passing object references
116 in events. We could say that any object we're passing to a
117 client must have a server id, and each client has a server id ->
120 - When a surface is the size of the screen and on top, we can set the
121 scanout buffer to that surface directly. Like compiz unredirect
122 top-level window feature. Except it won't have any protocol state
123 side-effects and the client that owns the surface won't know. We
124 lose control of updates. Should work well for X server root window
125 under wayland. Should be possible for yuv overlays as well.
127 - what about cursors then? maybe use hw cursors if the cursor
128 satisfies hw limitations (64x64, only one cursor), switch to
129 composited cursors if not.
131 - clients needs to allocate the surface to be suitable for
132 scanout, which they can do whenever they go fullscreen.
134 - multihead, screen geometry and crtc layout protocol, hotplug, lcd
137 - a wayland settings protocol to tell clients about themes (icons,
138 cursors, widget themes), fonts details (family, hinting
139 preferences) etc. Just send all settings at connect time, send
140 updates when a setting change. Getting a little close to gconf
141 here, but could be pretty simple:
143 interface "settings":
144 event int_value(string name, int value)
145 event string_value(string name, string value)
147 but maybe it's better to just require that clients get that from
148 somewhere else (gconf/dbus).
150 - input device discovery, hotplug
152 - Advertise axes as part of the discovery, use something like
153 "org.wayland.input.x" to identify the axes.
155 - keyboard state, layout events at connect time and when it
156 changes, keyboard leds
162 - synaptics, 3-button emulation, scim
164 - drm bo access control, authentication, flink_to
166 - Range protocol may not be sufficient... if a server cycles through
167 2^32 object IDs we don't have a way to handle wrapping. And since
168 we hand out a range of 256 IDs to each new clients, we're just
169 talking about 2^24 clients. That's 31 years with a new client
170 every minute... Maybe just use bigger ranges, then it's feasible
171 to track and garbage collect them when a client dies.
173 - Add protocol to let applications specify the effective/logical
174 surface rectangle, that is, the edge of the window, ignoring drop
175 shadows and other padding. The compositor needs this for snapping
176 and constraining window motion. Also, maybe communicate the opaque
177 region of the window (or just a conservative, simple estimate), to
178 let the compositor reduce overdraw.
180 - multi gpu, needs queue and seqno to wait on in requests
186 - draw window decorations in gtkwindow.c
188 - Details about pointer grabs. wayland doesn't have active grabs,
189 menus will behave subtly different. Under X, clicking a menu
190 open grabs the pointer and clicking outside the window pops down
191 the menu and swallows the click. without active grabs we can't
192 swallow the click. I'm sure there much more...
194 - Port Qt? There's already talk about this on the list.
198 - move most of the code from xf86-video-intel into a Xorg wayland
201 - don't ask KMS for available output and modes, use the info from
202 the wayland server. then stop mooching off of drmmode.c.
204 - map multiple wayland input devices to MPX in Xorg.
206 - rootless; avoid allocating and setting the front buffer, draw
207 window decorations in the X server (!), how to map input?
209 - gnome-shell as a wayland session compositor
211 - runs as a client of the wayland session compositor, uses
212 clutter+egl on wayland
214 - talks to an Xorg server as the compositing and window manager
215 for that server and renders the output to a wayland surface.
216 the Xorg server should be modified to take input from the system
217 compositor through gnome-shell, but not allocate a front buffer.
219 - make gnome-shell itself a nested wayland server and allow native
220 wayland clients to connect and can native wayland windows with
221 the windows from the X server.
223 - qemu as a wayland client; session surface as X case
225 - qemu has too simple acceleration, so a Wayland backend like the
226 SDL/VNC ones it has now is trivial.
228 - paravirt: forward wayland screen info as mmio, expose gem ioctls as mmio
230 - mapping vmem is tricky, should try to only use ioctl (pwrite+pread)
232 - not useful for Windows without a windows paravirt driver.
234 - two approaches: 1) do a toplevel qemu window, or 2) expose a
235 wayland server in the guest that forwards to the host wayland
236 server, ie a "remote" compositor, but with the gem buffers
237 shared. could do a wl_connection directly on mmio memory, with
238 head and tail pointers. use an alloc_head register to indicate
239 desired data to write, if it overwrites tail, block guest. just
240 a socket would be easier.
242 - moblin as a wayland compositor
244 - clutter as a wayland compositors