Commit Graph

184 Commits

Author SHA1 Message Date
Geoffrey McRae
0603a55492 [client] x11: prevent possible calibration underflow 2021-08-05 07:11:23 +10:00
Geoffrey McRae
0af558345f [client] wayland: fix build with libdecor after invalidateWindow change 2021-08-05 06:58:43 +10:00
Geoffrey McRae
6e7f39edee [client] app: allow selective full invalidation
Overlays only need to trigger a new frame if they have changed and not
full window invalidation, this change allows for this.
2021-08-05 06:47:36 +10:00
Quantum
51b9cd4e5a [all] copyright: use unicode copyright sign ©
This is done for consistency with the license strings in appstrings.c.
2021-08-04 21:16:35 +10:00
Geoffrey McRae
80c9f7223a [client] wayland: fix failure to build 2021-08-04 07:21:57 +10:00
Geoffrey McRae
c15d0dc672 [client] ds: waitFrame now returns a bool to force rendering if needed
X11 needs to calibrate to get the best possible latency, as such it
needs the scene to render so that the render time of the scene can be
accounted for in the delay calculation.
2021-08-04 06:49:35 +10:00
Geoffrey McRae
0f7fa32d12 [client] x11: move sleep and calibration code outside of the event loop
Sleeping in the x11 event loop is not ideal as it will introduce latency
when processing other events, instead do this in the waitFrame handler.
2021-08-04 06:34:27 +10:00
Geoffrey McRae
6933c278ce [client] x11: cosmetics 2021-08-04 06:14:27 +10:00
Geoffrey McRae
7fc717a839 [client] x11: don't setup XPresent if jitRender is not enabled 2021-08-04 06:12:21 +10:00
Quantum
a25c93b28e [client] wayland: document wlroots viewport handling workaround
Add a comment so we don't forget and remove the "useless" code.
2021-08-03 21:12:25 +10:00
Geoffrey McRae
03ed8b7304 [client] x11: added code to calibrate a delay for presentation
XPresent doesn't give us the time before presentation, but the time just
after. This code calculates and calibrates a delay to sleep for before
signaling the wait event for render when using jitRender
2021-08-03 07:51:03 +10:00
Geoffrey McRae
504bf02855 [client] x11: provide an empty rect region to XPresentPixmap
Providing None informs the comppositor that there is full screen damage,
instead we provide an empty rectangle to prevent this behaviour.
2021-08-03 07:49:59 +10:00
Geoffrey McRae
50f9baedba [client] x11: remove junk code from evaluation 2021-08-03 01:11:46 +10:00
Quantum
8d78a5aa95 [commit] wayland: invalidate window on scale changes
This makes it not require new frames for new scale to be properly applied.
2021-08-03 00:57:04 +10:00
Geoffrey McRae
891ee3e789 [client] x11: make use of the x11 present extension for jitRender
This implementation is a bit dodgy and needs some work but is currently
functional. Consider this feature highly experiemental under X11.
2021-08-02 23:59:26 +10:00
Quantum
f08e2ece93 [client] wayland: implement stopWaitFrame 2021-08-01 19:54:28 +10:00
Quantum
2e4614cbc4 [client] wayland: make waitFrame work when not rendering
Implement skipFrame and do various things to wake waitFrame.
2021-08-01 18:54:55 +10:00
Quantum
77b3d45e0e [client] ds: change signalNextFrame to waitFrame 2021-08-01 17:29:15 +10:00
Quantum
96bcfff28b [client] wayland: use zxdg_output_manager_v1 version 3 if possible
Version 3 does not send xdg_output.done events, instead guaranteeing that
all xdg_output.* events are sent before wl_output.done. This saves us from
doing the work twice.
2021-08-01 13:05:41 +10:00
Quantum
0ad469178a [client] wayland: make it possible to disable fractional scaling
The method used is not guaranteed to work on all Wayland compositors,
so offer a way out. We need to support it anyways in case xdg_output
or wp_viewporter protocols are not available.
2021-08-01 13:05:41 +10:00
Quantum
b3ca872cef [client] wayland: improve fractional scale handling
Currently, we scale the desktop up to the next largest integer, and rely on
the wayland compositor to scale it back down to the correct size.
This is obviously undesirable.

In this commit, we attempt to detect the actual fractional scaling by finding
the current active mode in wl_output, and dividing it by the logical screen
size reported by xdg_output, taking into consideration screen rotation.

We then use wp_viewporter to set the exact buffer and viewport sizes if
fractional scaling is needed.
2021-08-01 13:05:41 +10:00
Quantum
a213ee960a [client] wayland: implement signalNextFrame with frame callbacks 2021-07-31 19:05:21 +10:00
Geoffrey McRae
5e2dd589a1 [client] x11: implement cursor support for the imgui overlay
This commit adds libxcursor to the dependencies for X11.
2021-07-31 18:23:05 +10:00
Geoffrey McRae
3143dc1e84 [client] x11: call app_handleWheelMotion for imgui scroll interaction 2021-07-31 16:12:52 +10:00
Geoffrey McRae
7f6fd02d06 [client] x11: Add event handling for ungrabed mouse press/release
The imgui overlay requires input even if the display is not captured and
operating in raw mode. XInput2 correctly only sends
XI_Press/ReleaseButton events if the device has not been captured, as
such it's safe to handle both raw and non raw buttons events at the same
time.
2021-07-31 15:23:31 +10:00
Quantum
72ccd44681 [client] wayland: pass mouse wheel motion events to imgui 2021-07-31 14:58:35 +10:00
Quantum
6b6b3b724a [client] wayland: reject horizontal scroll events
Currently, we handle horizontal scroll events as if they are vertical scrolls.
This is not correct and we should instead reject them.
2021-07-31 14:57:56 +10:00
Quantum
11a5864969 [client] wayland: implement support for other cursors 2021-07-31 14:56:04 +10:00
Quantum
da28db2ca4 [client] ds: change showPointer to setPointer for more cursors
Currently everything only supports LG_POINTER_NONE and LG_POINTER_SQUARE.
2021-07-31 14:56:04 +10:00
Quantum
134829cbf2 [client] imgui: make graph y-axis configurable
The default of [0, 50] makes sense for FPS/UPS graphs, but does not for
things like the import graph. The latter should not take more than 5 ms
for sure.

This commit allows the min/max y-axis value to be specified when registering
the graph.
2021-07-29 15:54:15 +10:00
Quantum
e7761abf3c [client] wayland: invalidate window on configure
This avoids sending potentially meaningless damage values after a surface
configuration event.
2021-07-25 15:36:44 +10:00
Geoffrey McRae
3905834807 [client] x11: call app_invalidateWindow on expose 2021-07-25 15:30:56 +10:00
arcnmx
3b37898eb2 [all] use cmake FindPkgConfig IMPORTED_TARGETs 2021-07-24 12:35:48 +10:00
Quantum
b26067b0a0 [client] wayland: use new libdecor pkgconfig package name 2021-07-23 15:52:05 +10:00
Quantum
85f34602f4 [client] wayland: do not leak wp_presentation_feedback objects 2021-07-22 13:12:47 +10:00
Quantum
bb60107a3b [client] wayland: dispatch events through libdecor when using it
Currently, we dispatch the events on the wayland display server ourselves.
This is fine when using the cairo backend of libdecor, as it does the same
thign we do, but other backends may require other things to be dispatched.

This commit lets libdecor dispatch events instead through libdecor_get_fd
and libdecor_dispatch, which should hopefully makes things less sketchy.
2021-07-22 12:03:55 +10:00
Quantum
3f7261d7d9 [client] wayland: do not crash when presentation time is unavailable
We don't need the presentation time protocol, so fail gracefully when
it's unavailable.
2021-07-20 14:23:08 +10:00
Quantum
0cbc529640 [client] ds: refactor common EGL swap with damage logic
This commit creates a new utility library, eglutil.h, which contains code
to detect and use EGL_KHR_swap_buffers_with_damage or its EXT equivalent.

This logic used to be duplicated between the X11 and Wayland display servers,
which is not ideal.
2021-07-19 19:35:52 +10:00
Quantum
09893fd728 [client] wayland: display swap to photon latency graph 2021-07-18 16:15:10 +10:00
Quantum
e87bc3a83e [client] wayland: measure presentation time 2021-07-18 16:15:10 +10:00
Quantum
6695ca3f34 [client] ds: remove SDL display server 2021-07-18 10:34:41 +10:00
Geoffrey McRae
03c247a9ff [client] x11: make use of eglSwapBuffersWithDamage if it's available 2021-07-17 21:09:51 +10:00
Quantum
48ae5c69f4 [client] wayland: fix typo in warp usage
The unwanted ! was introduced in 4b99bba200.
This basically caused warp to never be used.
2021-07-15 08:40:36 +10:00
Geoffrey McRae
203ebc73eb [client] x11/opengl: enable double buffering for OpenGL
Drawing to the front buffer directly requires special handling to
prevent seeing the draw progress (avoiding glClear, etc) and as a result
the output is quite bad unless a compositor is running. Also vsync if
enabled will not function without double buffering enabled.

As OpenGL is the legacy fallback, there are no plans to implement clean
front buffer draw support, so just enable double buffering.
2021-07-02 21:45:12 +10:00
Quantum
6419279c3c [client] wayland/clipboard: prefer text/plain over text/html
When both text/plain and text/html are available, we should prefer the plain
text as we can't paste HTML.
2021-06-26 12:49:47 +10:00
Quantum
f3a1eaf557 [client] wayland: fix full screen toggle in capture mode regression
This was fixed in 9db3cd7b and accidentally broke again in 4b99bba2.
2021-06-20 09:27:33 +10:00
Geoffrey McRae
facbb54776 [client] x11: create _NET_WM_BYPASS_COMPOSITOR if it doesn't exist
If a compositor has not yet started or is not running the atom
`_NET_WM_BYPASS_COMPOSITOR` may not have yet been created. As such we
need to create it so that if a compositor is started it will see this
propery and honour it.
2021-06-17 11:16:26 +10:00
Quantum
4b99bba200 [client] wayland: lock confine-related code to avoid race
This should fix the occasional Wayland protocol errors that arise when
the UI thread and the cursor thread race.

Example of error that is fixed:

    zwp_pointer_constraints_v1@11: error 1: a pointer constraint with a wl_pointer of the same wl_seat is already on this surface
2021-06-08 21:04:14 +10:00
Quantum
5c00f73d23 [client] x11: use correct spelling of _NET_WM_BYPASS_COMPOSITOR
The incorrect spelling, NETWM_BYPASS_COMPOSITOR, somehow worked in the
past, but it appears to not work right now. Corrgan on Discord reported
the issue and confirmed that changing the spelling allowed the compositor
to be bypassed and the client to update faster than 60 Hz on his mixed
refresh rate setup.
2021-06-07 11:17:06 +10:00
Quantum
7f5f46c448 [client] wayland: offer all supported clipboard formats
This commit restructures the Wayland clipboard handling for host->VM.

Before, we select one clipboard format and buffers the data for it, to
be presented to spice when needed.

Now, we simply offer all clipboard formats supported, and only when spice
asks for the data do we actually read the wl_data_offer. The wl_data_offer
is kept around until a new offer is presented, the offer invalidated, or
when we lose keyboard focus. This is in accordance with the specification
for wl_data_device::selection, which states that:

> The data_offer is valid until a new data_offer or NULL is received or
> until the client loses keyboard focus. The client must destroy the
> previous selection data_offer, if any, upon receiving this event.

We still buffer the entire clipboard data into memory because we have no
knowledge of the clipboard data size in advance and cannot do incremental
transfers.

Furthermore, if the user performs drag-and-drop on our window, we may have
need to handle multiple wl_data_offer objects at the same time. Therefore,
instead of storing state on the global wlCb object, we instead allocate
memory and store it as user_data on the wl_data_offer. As a result, we also
handle drag-and-drop so that we can free the memory.
2021-06-06 12:20:42 +10:00