Commit Graph

5125 Commits

Author SHA1 Message Date
Simon Ser c930160286 backend/noop: add missing rollback_render output impl
3c5dbfd97c ("output: make rollback_render mandatory") makes a no-op
output init fail without this function.
2020-06-19 13:07:24 -06:00
Simon Ser 155d57b01d output: fix dangling renderer context after wlr_output_preferred_read_format
attach_render was called without un-setting the current rendering
context afterwards.

Closes: https://github.com/swaywm/wlroots/issues/2164
2020-06-19 11:50:42 -06:00
Simon Ser 3c5dbfd97c output: make rollback_render mandatory
If the output backend provides attach_render, assert it also provides a
way to revert it via rollback_render.
2020-06-19 11:50:42 -06:00
Simon Ser bf93d2e67c output: rename impl->rollback to rollback_render
The output backend API is now mostly state-less thanks to the atomic
hooks (commit and test). There is one exception though: attach_render.
This function makes the rendering context current. However sometimes the
compositor might decide not to render after attach_render (e.g. when
there's nothing new to render to the back buffer). Thus
wlr_output_rollback has been introduced to revert the pending state.

Because the output backend API is mostly state-less, the only thing
wlr_output_impl.rollback needs to do is revert the current rendering
context. Rename the function to rollback_render to make this clear. Add
a check in the common wlr_output code to only call rollback_render when
attach_buffer has been previously called.

On the long term, we'll be able to remove attach_render and
rollback_render together.
2020-06-19 11:50:42 -06:00
Simon Ser 58df3eda9f render/egl: print error name
Allows for easier debugging.
2020-06-18 07:56:05 -06:00
Simon Ser d177abecae surface: ignore viewport src rect on NULL buffer
According to the viewporter protocol:

> If the wl_buffer is NULL, the surface has no content and therefore no size.
2020-06-18 07:55:33 -06:00
Rouven Czerwinski c18c419b56 surface: don't unset width and height in finalize
During surface finalization we may not have received a new buffer,
resetting width and height in this case is wrong since we display the
old buffer in this case.
2020-06-18 09:54:15 +02:00
Tudor Brindus c768309ab4 input/keyboard: send modifiers on first keyboard enter
Will fix Firefox bug
https://bugzilla.mozilla.org/show_bug.cgi?id=1643991.

Fixes swaywm/sway#5462.
2020-06-17 18:54:38 +02:00
Simon Ser 9e68ed2159 viewporter: new protocol implementation
Closes: https://github.com/swaywm/wlroots/issues/633
2020-06-17 09:10:54 -06:00
Simon Ser c618a76540 surface: introduce wlr_surface_get_buffer_source_box
This helper allows compositors to retrieve the buffer source box in
buffer-local coordinates.
2020-06-17 09:10:54 -06:00
Simon Ser eb22ae97b0 surface: add wlr_surface_state.viewport
This field contains the viewport source and destination parameters. It's
intended to be updated by a third-party protocol, for instance
viewporter.
2020-06-17 09:10:54 -06:00
Simon Ser 315bf08733 render: add wlr_render_subtexture_with_matrix
This renders only a subset of the texture, instead of the full texture.
2020-06-17 09:10:54 -06:00
Simon Ser 00ccb89288 util/region: add wlr_region_scale_xy 2020-06-17 09:10:54 -06:00
Simon Ser 9814213a91 box: add wlr_fbox
Same as wlr_box, but for floating-point numbers.
2020-06-17 09:10:54 -06:00
Simon Ser 5118189a2b render/gles2: use .x/.y instead of .s/.t
texcoord is a vector of coordinates, with the first member being the X
axis value and the second member being the Y axis value. It makes more
sense to use the accessors with the same names.
2020-06-17 09:10:54 -06:00
Rouven Czerwinski d3d1bac1c2 render: assert sane values for box functions
Width and height should always be > 0 for render functions which take a
wlr_box.

References https://github.com/swaywm/wlroots/issues/2281
2020-06-16 15:16:02 +02:00
Andri Yngvason cae533cad2 screencopy: Use correct constant for y-inversion 2020-06-13 15:39:49 +02:00
Andri Yngvason a6a5a19356 backend: drm: Fix dmabuf resource leak 2020-06-13 15:39:49 +02:00
Simon Ser 3c13527ead examples/fullscreen-shell: stop advertising linux-dmabuf unconditonally
Remove the wlr_linux_dmabuf_v1_create call. wlr_renderer_init_wl_display
will take care of creating the linux-dmabuf global if the OpenGL
implementation supports it.
2020-06-11 11:50:17 -06:00
Simon Ser 4615ce9099 examples/screencopy-dmabuf: call strncpy with maxlen - 1
The original code wasn't wrong since we were manually writing a null
byte anyway, but this makes GCC happy.

Closes: https://github.com/swaywm/wlroots/issues/2273
2020-06-11 07:34:55 -06:00
Simon Ser 7f9bbaaa17 render/egl: explicit client extension handling
Prior to this commit, wlr_egl_init seemed to assume the extension string
queried via EGL_NO_DISPLAY was a subset of the extension string queried
via an initialized display. This isn't correct.

EGL_EXT_client_extensions [1] defines two types of extensions: client
extensions and display extensions. The set of supported client and
display extensions are disjoint (ie. an extension is either a client or
a display extension, not both). Client extensions are queried via
EGL_NO_DISPLAY, display extensions are queried via an initialized
display.

Rename the variables to make this clear. Remove the misleading comment.
Log both client and display extensions.

[1]: https://www.khronos.org/registry/EGL/extensions/EXT/EGL_EXT_client_extensions.txt
2020-06-10 14:55:29 +02:00
Simon Ser 79e03c7d45 output: document wlr_output_export_dmabuf 2020-06-09 07:12:43 -06:00
Simon Ser a8a4a12c4b buffer: document wlr_buffer_get_dmabuf DMA-BUF lifetime 2020-06-09 07:12:43 -06:00
Andri Yngvason 78d6eed6b3 render: Fix blit_dmabuf() breakage due to API change 2020-06-08 21:37:32 +02:00
Andri Yngvason e0d4f75172 examples: Add screencopy-dmabuf example 2020-06-08 20:49:41 +02:00
Andri Yngvason ea83957ecc screencopy: Implement copying into linux-dmabuf 2020-06-08 20:49:41 +02:00
Andri Yngvason 909b0213b3 protocol: screencopy: Extend to report linux-dmabuf capability
A new version of the protocol is required so that the client can
discover which buffer types are supported by the compositor.
2020-06-08 20:49:41 +02:00
Andri Yngvason b64a8a7f98 render: Add wlr_renderer_blit_dmabuf() 2020-06-08 20:49:41 +02:00
Andri Yngvason 11b598fe33 render: egl: Add utility functions for saving/restoring context 2020-06-08 20:49:41 +02:00
Simon Ser e77c046cf9 backend/drm: fix stack overflow in dealloc_crtc
Call drm_crtc_commit directly instead of calling drm_connector_set_mode.
This restores the previous behaviour where conn_enable was called [1].

[1]: 0c7c562482/backend/drm/drm.c (L1093)

Closes: https://github.com/swaywm/wlroots/issues/2253
2020-06-08 10:13:00 -06:00
Simon Ser 9b85a8b43d contributing: add note about events that destroy objects 2020-06-08 10:12:38 -06:00
Simon Ser fd0a845cb4 contributing: resource destroy handlers need to cleanup pointers 2020-06-08 10:12:38 -06:00
Simon Ser a3ba82885c render: choose DMA-BUF texture target via eglQueryDmaBufModifiersEXT
EGL_EXT_image_dma_buf_import_modifiers tells us whether we should use
GL_TEXTURE_2D or GL_TEXTURE_EXTERNAL_OES. Using the right texture target
can fix some failures and/or improve performance on some drivers.

This does the same as a Weston commit [1].

[1]: https://gitlab.freedesktop.org/wayland/weston/commit/40c519a3e613

Closes: https://github.com/swaywm/wlroots/issues/2173
2020-06-06 08:59:38 +02:00
Tudor Brindus 363bf44a35 util/time: move `timespec_sub` to time utilities 2020-06-06 00:09:19 +02:00
Tudor Brindus c9c31f803e util/time: de-duplicate `timespec_to_msec` 2020-06-06 00:09:19 +02:00
Tudor Brindus dc13bb827d util: fix and move `get_current_time_msec` into a util file
This commit makes `get_current_time_msec` correctly return milliseconds
as opposed to microseconds. It also considers the value of `tv_sec`, so
we don't lose occasionally go back in time by one second. Finally, the
function is moved into `util/time.cc` so that it can be reused elsewhere
without having to consider these pitfalls.
2020-06-06 00:09:19 +02:00
Thomas Hebb dcae6f1431 Allow keyboard and pointer grabs to hook clear_focus()
This is necessary for some grabs, which currently have no way of knowing
when the pointer/keyboard focus has left a surface. For example, without
this, a drag-and-drop grab can erroneously drop into a window that the
cursor is no longer over.

This is the plumbing needed to properly fix swaywm/sway#5220. The
existing fix, swaywm/sway#5222, relies on every grab's `enter()` hook
allowing a `NULL` surface. This is not guaranteed by the API and, in
fact, is not the case for the xdg-shell popup grab and results in a
crash when the cursor leaves a surface and does not immediately enter
another one while a popup is open (#2161).

This fix also adds an assertion to wlr_seat_pointer_notify_enter() that
ensures it's never called with a `NULL` surface. This will make Sway
crash much more until it fixes its usage of the API, so we should land
this at the same time as a fix in Sway (which I haven't posted yet).
2020-06-05 17:20:26 +02:00
Thomas Hebb 8bf9f5bb8b wlr_seat: Clarify wording of non-grab-respecting function comments
I found the previous wording a bit confusing when I first read it.
Reword these comments to explicitly say that the grab-respecting
variants should be used in most cases.

This change has no functional effect.
2020-06-05 17:20:26 +02:00
Thomas Hebb e8c00e918a wlr_seat: Declare functions in the same order for pointer, keyboard, and touch
These three APIs are very similar to one another, but they all had
slightly different function orderings. For consistency, always declare
the non_`notify` functions first, then the `notify` functions, then
`{start,end,has}_grab`.

This change has no functional effect.
2020-06-05 17:20:26 +02:00
Simon Ser 8f28f5b2f8 backend/drm: fix black screens when enabling output
This patch fixes this failure:

    01:57:16.642 [ERROR] [backend/drm/drm.c:360] Failed to page-flip output 'eDP-1': a page-flip is already pending
    01:57:16.684 [ERROR] [backend/drm/drm.c:360] Failed to page-flip output 'eDP-1': a page-flip is already pending
    01:57:16.684 [ERROR] [backend/drm/drm.c:732] Failed to initialize renderer on connector 'eDP-1': initial page-flip failed
    01:57:16.684 [ERROR] [backend/drm/drm.c:805] Failed to initialize renderer for plane
    01:57:16.684 [sway/config/output.c:423] Failed to commit output eDP-1

References: https://github.com/swaywm/sway/issues/5101
2020-06-03 10:27:21 -06:00
Simon Ser e91417ea8d render: stop making EGL context current in wlr_egl_init
This leaves an EGL context current behind. wlr_gles2_renderer_create was
assuming the EGL context was already current because of this (because it
called a GL function right off the bat).
2020-06-02 14:23:24 -06:00
Simon Ser 019fe8bb7e backend/wayland: fix spurious eglSwapBuffers failures
This was introduced in [1]. However after reverting that PR I still
can't reproduce the bug the PR aimed to fix [2].

Since there's no good explanation why we would need to swap buffers
before resizing, let's just revert the PR.

[1]: https://github.com/swaywm/wlroots/pull/1486
[2]: https://github.com/swaywm/wlroots/issues/1371

Closes: https://github.com/swaywm/wlroots/issues/1768
2020-06-02 14:14:08 -06:00
Tudor Brindus 47a30957d1 backend/wayland: scale tablet tool position into [0, 1] range
Fixes #2233.
2020-06-01 22:36:46 +02:00
Simon Ser b03eebf7d4 backend/drm: always perform a CRTC commit in drm_connector_commit
When the mode, status or buffer hasn't changed, drm_connector_commit was
a no-op. Because of this individual changes to the gamma LUT or adaptive
sync status were ignored (if committed without a buffer).

Instead, perform a commit to apply the changes.
2020-06-01 11:45:11 -06:00
Simon Ser 1a2e82e327 backend/drm: drop extra wlr_output_update_enabled call
drm_connector_set_mode already takes care of keeping this up-to-date.
2020-06-01 11:45:11 -06:00
Simon Ser 80467f6ae8 backend/drm: commit/rollback FBs in drm_crtc_commit
We need to perform the FB bookkeeping on all commits, not just on
page-flips.
2020-06-01 11:45:11 -06:00
Simon Ser 70f3a0bb27 backend/drm: make adaptive_sync atomic
Stop using drmModeObjectSetProperty, set the property in the crtc_commit
function instead.
2020-06-01 11:45:11 -06:00
Simon Ser 42126c2022 backend/drm: rollback pending CRTC state on test commit
A test commit doesn't apply the pending state.

The CRTC state will be populated again if the compositor decides to
perform a regular commit afterwards.
2020-06-01 11:45:11 -06:00
Andri Yngvason 034384f5aa export-dmabuf: Keep frame->output = NULL until frame is ready
This fixes a crash that happens when a client requests a frame on a
backend that does not implement export_dmabuf.

An assertion fails with the message:
sway: types/wlr_output.c:777: wlr_output_lock_attach_render: Assertion
`output->attach_render_locks > 0' failed.
2020-06-01 12:50:26 +02:00
Brian Ashworth 32148808ad wlr_keyboard_group: introduce enter and leave
This introduces the enter and leave events for wlr_keyboard_group.

The enter event is emitted when a keyboard is added to the group while a
key is pressed that is not pressed by any other keyboard in the group.
The data is a wl_array of the pressed key codes unique to the keyboard
that should now be considered pressed.

Similarly the leave event is emitted when a keyboard is removed from the
group while at least one key is pressed that is not pressed by any other
keyboard in the group. The data is a wl_array of the pressed key codes
unique to the keyboard that should now be considered released.

The purpose of these events are to allow the compositor to update its
state to avoid corruption. Additionally, for the leave event, the
focused surface may have been notified of a key press for some or all of
the key codes and needs to be notified of a key release to avoid state
corruption.

These were previously emitted as normal key events, but they are not
normal key events. There is no actual key press or release associated
with the events. It's purely for state keeping purposes. Emitting them
as separate events allows the compositor to handle them differently.
Since these are purely for state keeping purposes and are not associated
with an actual key being pressed or released, bindings should not be
triggered as a result of these events.
2020-05-31 17:28:18 -04:00