Merge pull request #101454 from Riteo/fifo_ftw

Wayland: Handle `fifo_v1` and clean up suspension logic
This commit is contained in:
Thaddeus Crews
2025-04-25 11:36:04 -05:00
3 changed files with 93 additions and 65 deletions

View File

@ -822,8 +822,9 @@ void DisplayServerWayland::show_window(WindowID p_window_id) {
}
#endif
// NOTE: The public window-handling methods might depend on this flag being
// set. Ensure to not make any of these calls before this assignment.
// NOTE: Some public window-handling methods might depend on this flag being
// set. Make sure the method you're calling does not depend on it before this
// assignment.
wd.visible = true;
// Actually try to apply the window's mode now that it's visible.
@ -1349,7 +1350,7 @@ void DisplayServerWayland::window_set_vsync_mode(DisplayServer::VSyncMode p_vsyn
if (rendering_context) {
rendering_context->window_set_vsync_mode(p_window_id, p_vsync_mode);
wd.emulate_vsync = (rendering_context->window_get_vsync_mode(p_window_id) == DisplayServer::VSYNC_ENABLED);
wd.emulate_vsync = (!wayland_thread.is_fifo_available() && rendering_context->window_get_vsync_mode(p_window_id) == DisplayServer::VSYNC_ENABLED);
if (wd.emulate_vsync) {
print_verbose("VSYNC: manually throttling frames using MAILBOX.");
@ -1362,6 +1363,8 @@ void DisplayServerWayland::window_set_vsync_mode(DisplayServer::VSyncMode p_vsyn
if (egl_manager) {
egl_manager->set_use_vsync(p_vsync_mode != DisplayServer::VSYNC_DISABLED);
// NOTE: Mesa's EGL implementation does not seem to make use of fifo_v1 so
// we'll have to always emulate V-Sync.
wd.emulate_vsync = egl_manager->is_using_vsync();
if (wd.emulate_vsync) {
@ -1542,38 +1545,14 @@ bool DisplayServerWayland::color_picker(const Callable &p_callback) {
}
void DisplayServerWayland::try_suspend() {
bool must_emulate = false;
for (KeyValue<DisplayServer::WindowID, WindowData> &pair : windows) {
if (pair.value.emulate_vsync) {
must_emulate = true;
break;
}
}
// Due to various reasons, we manually handle display synchronization by
// waiting for a frame event (request to draw) or, if available, the actual
// window's suspend status. When a window is suspended, we can avoid drawing
// altogether, either because the compositor told us that we don't need to or
// because the pace of the frame events became unreliable.
if (must_emulate) {
bool frame = wayland_thread.wait_frame_suspend_ms(WAYLAND_MAX_FRAME_TIME_US / 1000);
if (!frame) {
suspend_state = SuspendState::TIMEOUT;
}
}
// If we suspended by capability, we'll know with this check. We must do this
// after `wait_frame_suspend_ms` as it progressively dispatches the event queue
// during the "timeout".
if (wayland_thread.is_suspended()) {
suspend_state = SuspendState::CAPABILITY;
}
if (suspend_state == SuspendState::TIMEOUT) {
DEBUG_LOG_WAYLAND("Suspending. Reason: timeout.");
} else if (suspend_state == SuspendState::CAPABILITY) {
DEBUG_LOG_WAYLAND("Suspending. Reason: capability.");
bool frame = wayland_thread.wait_frame_suspend_ms(WAYLAND_MAX_FRAME_TIME_US / 1000);
if (!frame) {
suspend_state = SuspendState::TIMEOUT;
}
}
@ -1724,39 +1703,54 @@ void DisplayServerWayland::process_events() {
wayland_thread.keyboard_echo_keys();
if (suspend_state == SuspendState::NONE) {
// Due to the way legacy suspension works, we have to treat low processor
// usage mode very differently than the regular one.
if (OS::get_singleton()->is_in_low_processor_usage_mode()) {
// NOTE: We must avoid committing a surface if we expect a new frame, as we
// might otherwise commit some inconsistent data (e.g. buffer scale). Note
// that if a new frame is expected it's going to be committed by the renderer
// soon anyways.
if (!RenderingServer::get_singleton()->has_changed()) {
// We _can't_ commit in a different thread (such as in the frame callback
// itself) because we would risk to step on the renderer's feet, which would
// cause subtle but severe issues, such as crashes on setups with explicit
// sync. This isn't normally a problem, as the renderer commits at every
// frame (which is what we need for atomic surface updates anyways), but in
// low processor usage mode that expectation is broken. When it's on, our
// frame rate stops being constant. This also reflects in the frame
// information we use for legacy suspension. In order to avoid issues, let's
// manually commit all surfaces, so that we can get fresh frame data.
wayland_thread.commit_surfaces();
try_suspend();
switch (suspend_state) {
case SuspendState::NONE: {
bool emulate_vsync = false;
for (KeyValue<DisplayServer::WindowID, WindowData> &pair : windows) {
if (pair.value.emulate_vsync) {
emulate_vsync = true;
break;
}
}
} else {
try_suspend();
}
} else {
if (suspend_state == SuspendState::CAPABILITY) {
// If we suspended by capability we can assume that it will be reset when
// the compositor wants us to repaint.
if (!wayland_thread.is_suspended()) {
suspend_state = SuspendState::NONE;
DEBUG_LOG_WAYLAND("Unsuspending from capability.");
if (emulate_vsync) {
// Due to the way legacy suspension works, we have to treat low processor
// usage mode very differently than the regular one.
if (OS::get_singleton()->is_in_low_processor_usage_mode()) {
// NOTE: We must avoid committing a surface if we expect a new frame, as we
// might otherwise commit some inconsistent data (e.g. buffer scale). Note
// that if a new frame is expected it's going to be committed by the renderer
// soon anyways.
if (!RenderingServer::get_singleton()->has_changed()) {
// We _can't_ commit in a different thread (such as in the frame callback
// itself) because we would risk to step on the renderer's feet, which would
// cause subtle but severe issues, such as crashes on setups with explicit
// sync. This isn't normally a problem, as the renderer commits at every
// frame (which is what we need for atomic surface updates anyways), but in
// low processor usage mode that expectation is broken. When it's on, our
// frame rate stops being constant. This also reflects in the frame
// information we use for legacy suspension. In order to avoid issues, let's
// manually commit all surfaces, so that we can get fresh frame data.
wayland_thread.commit_surfaces();
try_suspend();
}
} else {
try_suspend();
}
}
} else if (suspend_state == SuspendState::TIMEOUT) {
if (wayland_thread.is_suspended()) {
suspend_state = SuspendState::CAPABILITY;
}
if (suspend_state == SuspendState::TIMEOUT) {
DEBUG_LOG_WAYLAND("Suspending. Reason: timeout.");
} else if (suspend_state == SuspendState::CAPABILITY) {
DEBUG_LOG_WAYLAND("Suspending. Reason: capability.");
}
} break;
case SuspendState::TIMEOUT: {
// Certain compositors might not report the "suspended" wm_capability flag.
// Because of this we'll wake up at the next frame event, indicating the
// desire for the compositor to let us repaint.
@ -1764,11 +1758,20 @@ void DisplayServerWayland::process_events() {
suspend_state = SuspendState::NONE;
DEBUG_LOG_WAYLAND("Unsuspending from timeout.");
}
}
// Since we're not rendering, nothing is committing the windows'
// surfaces. We have to do it ourselves.
wayland_thread.commit_surfaces();
// Since we're not rendering, nothing is committing the windows'
// surfaces. We have to do it ourselves.
wayland_thread.commit_surfaces();
} break;
case SuspendState::CAPABILITY: {
// If we suspended by capability we can assume that it will be reset when
// the compositor wants us to repaint.
if (!wayland_thread.is_suspended()) {
suspend_state = SuspendState::NONE;
DEBUG_LOG_WAYLAND("Unsuspending from capability.");
}
} break;
}
#ifdef DBUS_ENABLED

View File

@ -60,6 +60,10 @@
#define DEBUG_LOG_WAYLAND_THREAD(...)
#endif
// Since we're never going to use this interface directly, it's not worth
// generating the whole deal.
#define FIFO_INTERFACE_NAME "wp_fifo_manager_v1"
// Read the content pointed by fd into a Vector<uint8_t>.
Vector<uint8_t> WaylandThread::_read_fd(int fd) {
// This is pretty much an arbitrary size.
@ -640,6 +644,10 @@ void WaylandThread::_wl_registry_on_global(void *data, struct wl_registry *wl_re
return;
}
if (strcmp(interface, FIFO_INTERFACE_NAME) == 0) {
registry->wp_fifo_manager_name = name;
}
}
void WaylandThread::_wl_registry_on_global_remove(void *data, struct wl_registry *wl_registry, uint32_t name) {
@ -1028,6 +1036,10 @@ void WaylandThread::_wl_registry_on_global_remove(void *data, struct wl_registry
it = it->next();
}
}
if (name == registry->wp_fifo_manager_name) {
registry->wp_fifo_manager_name = 0;
}
}
void WaylandThread::_wl_surface_on_enter(void *data, struct wl_surface *wl_surface, struct wl_output *wl_output) {
@ -4275,6 +4287,10 @@ Error WaylandThread::init() {
}
#endif // DBUS_ENABLED
if (!registry.wp_fifo_manager_name) {
WARN_PRINT("FIFO protocol not found! Frame pacing will be degraded.");
}
// Wait for seat capabilities.
wl_display_roundtrip(wl_display);
@ -4777,6 +4793,10 @@ bool WaylandThread::window_is_suspended(DisplayServer::WindowID p_window_id) con
return windows[p_window_id].suspended;
}
bool WaylandThread::is_fifo_available() const {
return registry.wp_fifo_manager_name != 0;
}
bool WaylandThread::is_suspended() const {
for (const KeyValue<DisplayServer::WindowID, WindowState> &E : windows) {
if (!E.value.suspended) {

View File

@ -216,6 +216,10 @@ public:
struct zwp_text_input_manager_v3 *wp_text_input_manager = nullptr;
uint32_t wp_text_input_manager_name = 0;
// We're really not meant to use this one directly but we still need to know
// whether it's available.
uint32_t wp_fifo_manager_name = 0;
};
// General Wayland-specific states. Shouldn't be accessed directly.
@ -1068,6 +1072,7 @@ public:
void set_frame();
bool get_reset_frame();
bool wait_frame_suspend_ms(int p_timeout);
bool is_fifo_available() const;
uint64_t window_get_last_frame_time(DisplayServer::WindowID p_window_id) const;
bool window_is_suspended(DisplayServer::WindowID p_window_id) const;