Update for wlroots 0.18 #1641

Consolatis · 2024-03-19T02:59:57Z

Same story as

Update for wlroots 0.17 #626

This branch is not intended to be merged anytime soon but should allow to collaborate on chasing wlroots master.
As in the previous PR, feel free to push changes into this branch.

I suggest not using force pushes for now to reduce chances of accidentally overwriting something. If some existing commit should be amended just use !fixup or [fixup] or something along that line and refer to the subject of the commit to be amended (the hash may change due to rebases). We can remove that no-force-push restriction once we have a somewhat stable base to work with.

Like in #626 our goal should also be to make this PR as small as possible to reduce rebase conflicts with master. So if we can merge a commit in master that works for 0.17 and 0.18 without introducing any unnecessary complexity we should likely do that.

@ahesford @jlindgren90 @johanmalm

Further information:

breaking changes for 0.18: https://gitlab.freedesktop.org/wlroots/wlroots/-/issues/3760
todo for wlroots 0.18 #1260

Current issues:

parentless xdg popups are not be handled correctly
the commit messages and wlroots.wrap files do not yet contain the correct wlroots hash and might also be out of order, I only did the first step of getting labwc to work with current wlroots master.
https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/4600 seems to implement a immediate scale event for surfaces within scene. We may be able to remove our downstream fix (4238d7f) for that issue, @ahesford could say more about that
listen to wlr_renderer.events.lost to catch GPU resets and do a Reconfigure to reload all compositor side buffers
xdg decorations must not be confirmed directly but only if xdg_surface->initialized == true, also needs storage of the requested deco mode or link from view to the deco object itself which can be used from the xdg commit handler
explicit sync (wlroots MR, sway PR)
icon spec (wlroots MR)

Fixed:

the initial xdg_surface configure flow has been changed, compositors now need to do that on their own
at least when running nested the rounded corners sometimes flicker on mouse movements and are then completely transparent (caused by a wlroots bug, MR pending)
as wlroots removed wlr_output->pending we have to use the wlr_output_state_() variants everywhere. But to reduce unnecessary screen blanking during mode changes we likely want to keep using a shared pending state. Rather than changing everything around to that I added a new src/output-state.c file which provides wrappers for the old wlr_output_set_()functions. This massively reduces rebase conflicts.
we currently crash on ctrl-c in gdb when running nested, ~~needs investigation~~ this is a libwayland bug, fixed via this MR but not yet released
lab_wlr_scene_commit() still contains a comment about handling damage, that was an issue in wlroots 0.17.x but might now have been fixed upstream, needs to be verified
The CI is currently not set up to deal with wlroots compilations, should be easy to fix
WLR_SCENE_DEBUG_DAMAGE=highlight damage tracking is pretty broken, wlroots MR

There might be more issues, only slightly tested.

jlindgren90 · 2024-03-19T17:53:39Z

Thanks for getting this going!

Consolatis · 2024-04-23T21:38:53Z

The damage issues and wayland native windows not be appearing should hopefully be fixed by the last commits. Also rebased on top of master.

kode54 · 2024-04-24T03:02:53Z

Further outstanding issues:

The display refreshes at less than its native refresh rate. More like 20-30fps instead of 60. This doesn't seem to be related to use of adaptive sync on one of the displays.
Atomic modesetting tearing support, which is supposed to be enabled upstream without any changes to the compositor to allow it, other than removing your check for the WLR_DRM_NO_ATOMIC variable, doesn't seem to work. I can try to pull DRM debug logs, but last I checked, it was wlroots committing things it shouldn't be committing during the atomic flips. Hell, AMD doesn't even want it to be committing cursor plane updates to allow tearing.
(added by Consolatis: copied over from a previous post) Steam's Xwayland override redirect popup menus seem to display a single frame of solid white before resolving to Steam's intended surface contents. Not sure if this is a bug in Steam, though.

kode54 · 2024-04-26T03:30:21Z

Logs for the above issues, possibly also relevant to wlroots 0.18 instead of just labwc:

dmesg.log.gz
labwc.log.gz

kode54 · 2024-05-01T02:58:28Z

Confirmed these frame rate issues do not occur with Sway master, so it's down to labwc and this PR being updated to handle whatever the refresh issue is.

Consolatis · 2024-05-10T18:11:51Z

@kode54 added a FPS counter per output via debug log, maybe this helps to shed some light on your half monitor refresh rate observation. I have a slight feeling that this has something to do with the presentation time protocol but I am not sure. At least nested I get the full fps via es2gears_x11 and es2gears_wayland. The FPS counter via debug log shows a reduction of the fps to about screen refresh rate in case of es2gears_x11, for the _wayland case the fps roughly matches between the es2gears output and labwc debug log.

johanmalm · 2024-05-12T18:13:07Z

As discussed the other day, I've been running this PR for a couple of days now and it seems fine.

I've also carried out an independent review by studying the PR and the sway commit-history between 2023-11-21 and 2024-05-12 (i.e. since wlroots 0.17.0 release). See findings below.

wlroots chasing - can be done on master now (I think)

wlroots chasing required on `0.18` PR

There are three that I couldn't see covered in the PR.

wlroots chasing not required by labwc

Things in `0.18 PR` that may need more work

xdg-shell (toplevels and popups)

Ref: cb8d89e

It looks like there is a lot of xdg/layer-shell toplevel/popup stuff that has changed. Found the following commits that I don't think we've not fully covered in PR 0.18. Guess some of them can be done on master now.

XWayland re-stacking

I'm sure you're right on this, but couldn't find anything in sways commit log that gave me confidence that we're approaching it right. How do we gain confidence that this XWayland restacking stuff is right? What changed in wlroots to trigger this amendment?

0f903cc

Aside

As an aside, I spotted the following on sway master and think we could consider these at some point (but not pre-req to 0.18 merge). We could probably start a separate issue with these later, but am including them here for the minute for completeness.

Consolatis · 2024-05-12T18:29:03Z

I'm sure you're right on this, but couldn't find anything in sways commit log that gave me confidence that we're approaching it right. How do we gain confidence that this XWayland restacking stuff is right? What changed in wlroots to trigger this amendment?

wlroots now handles that internally (and in fact asserts that the given surface is not an unamanged one), see https://gitlab.freedesktop.org/wlroots/wlroots/-/merge_requests/4052. That was implemented in wlroots to fix a sway bug because it did not restack unmanged surfaces on top like we do / did.

kode54 · 2024-05-24T09:25:14Z

Frame rate issues seem to be resolved now.

Consolatis · 2024-05-24T10:07:28Z

Rebased on master and disabled the fps counter, can be re-enabled by uncommenting #define FPS_TIMEOUT (2 * 1000) in src/output-state.c.

kode54 · 2024-05-24T11:17:33Z

Closing virt-manager window for a VM now crashes with memory corruption:

#0  __pthread_kill_implementation
    (threadid=<optimized out>, signo=signo@entry=6, no_tid=no_tid@entry=0)
    at pthread_kill.c:44
#1  0x00007604079e4eb3 in __pthread_kill_internal
    (threadid=<optimized out>, signo=6) at pthread_kill.c:78
#2  0x000076040798ca30 in __GI_raise (sig=sig@entry=6)
    at ../sysdeps/posix/raise.c:26
#3  0x00007604079744c3 in __GI_abort () at abort.c:79
#4  0x0000760407975354 in __libc_message_impl
    (fmt=fmt@entry=0x760407afe2ea "%s\n") at ../sysdeps/posix/libc_fatal.c:132
#5  0x00007604079ef085 in malloc_printerr
    (str=str@entry=0x760407afbffe "corrupted double-linked list")
    at malloc.c:5772
#6  0x00007604079efbfc in unlink_chunk
    (p=p@entry=0x5af189f7bcf0, av=0x760407b32ac0 <main_arena>) at malloc.c:1617
#7  0x00007604079f2a9a in _int_malloc
    (av=av@entry=0x760407b32ac0 <main_arena>, bytes=bytes@entry=208)
    at malloc.c:4381
#8  0x00007604079f477e in __libc_calloc
    (n=n@entry=1, elem_size=elem_size@entry=208) at malloc.c:3754
#9  0x0000760408397130 in zalloc (s=208)
    at ../wayland-1.22.0/src/wayland-private.h:234
#10 wl_closure_init
    (message=message@entry=0x760408385fb0 <zwlr_foreign_toplevel_handle_v1_events.lto_priv+144>, size=size@entry=0, num_arrays=num_arrays@entry=0x0, args=args@entry=0x7ffe3d108350) at ../wayland-1.22.0/src/connection.c:572
#11 0x0000760408399abf in wl_closure_marshal
    (message=0x760408385fb0 <zwlr_foreign_toplevel_handle_v1_events.lto_priv+144>, args=0x7ffe3d108350, opcode=6, sender=0x5af189f9c7a0)
    at ../wayland-1.22.0/src/connection.c:608
#12 handle_array
    (resource=resource@entry=0x5af189f9c7a0, opcode=opcode@entry=6, args=args@entry=0x7ffe3d108350, send_func=send_func@entry=0x7604083976f0 <wl_closure_send>)
    at ../wayland-1.22.0/src/wayland-server.c:224
#13 0x0000760408399cf0 in wl_resource_post_event_array
    (resource=resource@entry=0x5af189f9c7a0, opcode=opcode@entry=6, args=args@entry=0x7ffe3d108350) at ../wayland-1.22.0/src/wayland-server.c:244
#14 0x0000760408399ddf in wl_resource_post_event
    (resource=resource@entry=0x5af189f9c7a0, opcode=opcode@entry=6)
    at ../wayland-1.22.0/src/wayland-server.c:259
#15 0x00007604082f3eb8 in zwlr_foreign_toplevel_handle_v1_send_closed
    (resource_=0x5af189f9c7a0)
    at protocol/wlr-foreign-toplevel-management-unstable-v1-protocol.h:490
#16 wlr_foreign_toplevel_handle_v1_destroy (toplevel=0x5af189f9c6c0)
    at ../wlroots-hidpi-xprop-git/types/wlr_foreign_toplevel_management_v1.c:499
#17 wlr_foreign_toplevel_handle_v1_destroy (toplevel=0x5af189f9c6c0)
    at ../wlroots-hidpi-xprop-git/types/wlr_foreign_toplevel_management_v1.c:489
#18 0x00005af17d759615 in view_destroy (view=0x5af189f7be60)
    at ../labwc/src/view.c:2325
#19 0x000076040839900e in wl_signal_emit_mutable
    (signal=signal@entry=0x5af189ef4da8, data=data@entry=0x0)
    at ../wayland-1.22.0/src/wayland-server.c:2241
#20 0x00007604082ec0c8 in destroy_xdg_surface (surface=0x5af189ef4d00)
    at ../wlroots-hidpi-xprop-git/types/xdg_shell/wlr_xdg_surface.c:492
#21 0x00007604082f1e0d in surface_destroy_role_object (surface=0x5af189ef48a0)
    at ../wlroots-hidpi-xprop-git/types/wlr_compositor.c:910
#22 surface_destroy_role_object (surface=0x5af189ef48a0)
    at ../wlroots-hidpi-xprop-git/types/wlr_compositor.c:904
#23 surface_handle_role_resource_destroy
    (listener=0x5af189ef4bf0, data=<optimized out>)
    at ../wlroots-hidpi-xprop-git/types/wlr_compositor.c:891
#24 0x000076040839a99f in wl_priv_signal_final_emit
    (data=<optimized out>, signal=<optimized out>)
    at ../wayland-1.22.0/src/wayland-server.c:2377
#25 destroy_resource (element=0x5af189ef4e50, data=data@entry=0x0, flags=0)
    at ../wayland-1.22.0/src/wayland-server.c:729
#26 0x000076040839c7e9 in wl_resource_destroy (resource=<optimized out>)
    at ../wayland-1.22.0/src/wayland-server.c:749
#27 0x00007604073fc596 in ffi_call_unix64 () at ../src/x86/unix64.S:104
#28 0x00007604073f900e in ffi_call_int
    (cif=cif@entry=0x7ffe3d1087e0, fn=<optimized out>, rvalue=<optimized out>, avalue=<optimized out>, closure=closure@entry=0x0) at ../src/x86/ffi64.c:673
#29 0x00007604073fbbd3 in ffi_call
    (cif=cif@entry=0x7ffe3d1087e0, fn=<optimized out>, rvalue=rvalue@entry=0x0, avalue=avalue@entry=0x7ffe3d1088b0) at ../src/x86/ffi64.c:710
#30 0x0000760408397ab0 in wl_closure_invoke
    (closure=closure@entry=0x5af189ed0500, target=<optimized out>,
    target@entry=0x5af189ef4e50, opcode=opcode@entry=0, data=<optimized out>,
    data@entry=0x5af189e79b40, flags=2)
    at ../wayland-1.22.0/src/connection.c:1025
#31 0x000076040839c198 in wl_client_connection_data
    (fd=<optimized out>, mask=<optimized out>, data=<optimized out>)
    at ../wayland-1.22.0/src/wayland-server.c:438
#32 0x000076040839ab02 in wl_event_loop_dispatch
    (loop=0x5af188923c20, timeout=timeout@entry=-1)
    at ../wayland-1.22.0/src/event-loop.c:1027
#33 0x000076040839b317 in wl_display_run (display=0x5af188923b30)
    at ../wayland-1.22.0/src/wayland-server.c:1493
#34 0x00005af17d7436e3 in main (argc=<optimized out>, argv=<optimized out>)
    at ../labwc/src/main.c:217

droc12345 · 2024-05-24T18:26:42Z

Started running this PR as my daily driver.
No problems so far, seems a tad snappier than the wlroots 17.* version, at least running the vulkan renderer.

Good job.

kode54 · 2024-05-25T06:54:46Z

Updated wlroots, one new commit dealing with direct scanout and color processing, and now the above crash isn't happening any longer?

kode54 · 2024-05-26T05:10:45Z

This apparently fixes the quarter screen issue I was having with Waydroid? Seems to work without requiring me to suspend my displays first.

droc12345 · 2024-06-01T11:21:23Z

I've been running this version for a while, and turned on logging this morning and noticed error messages

00:00:00.182 [ERROR] [types/wlr_layer_shell_v1.c:296] A configure is sent to an uninitialized wlr_layer_surface_v1 0x17d9a80

It doesn't seem to matter which window, every time I open a new app/window, the message appears.
conky, foot, etc trigger it, whatever it is I start, doesn't seem to affect the running.
I've updated everything this morning to check, wlroots, foot, and several apps, still the messages.

Edit to add: seems to come from src/xdg.c

if (xdg_surface->initial_commit) { wlr_log(WLR_DEBUG, "scheduling configure"); //FIXME: confirm potential xdg-deco mode wlr_xdg_surface_schedule_configure(xdg_surface); return; }

Note: I do run with decorations turned completely off

ETA2:
The error messages come from wlr_xdg_surface_schedule_configure, and it's only called in the
one place, but I only see the "scheduling configure" message occasionally, So I'm not sure what
exactly is triggering off the messages.

Consolatis · 2024-06-01T15:03:06Z

The error messages come from wlr_xdg_surface_schedule_configure, and it's only called in the
one place, but I only see the "scheduling configure" message occasionally, So I'm not sure what
exactly is triggering off the messages.

Its caused by the decoration manager:

xdg decorations must not be confirmed directly but only if xdg_surface->initialized == true, also needs storage of the requested deco mode or link from view to the deco object itself which can be used from the xdg commit handler

Edit:
Sorry, your log is actually pointing to a layer surface rather than a xdg surface. The same error should happen for xdg surface for the above reason. For layer surfaces this must be something different. Thanks for pointing it out.

droc12345 · 2024-06-01T15:39:24Z

It does report both depending on whether a layer app or regular app.

Edit to add:
[types/wlr_layer_shell_v1.c:296] A configure is sent to an uninitialized wlr_layer_surface_v1
and
[types/xdg_shell/wlr_xdg_surface.c:169] A configure is scheduled for an uninitialized xdg_surface

kode54 · 2024-06-10T02:53:45Z

Is this worth rebasing without addressing the issues mentioned above? Or is there some pending work already?

Already fixed in its master branch but not yet available in a release. https://gitlab.freedesktop.org/wayland/wayland/-/merge_requests/366

…) (MR 4396)

Consolatis · 2024-06-10T03:19:47Z

Is this worth rebasing without addressing the issues mentioned above?

Yes, rebased on current master.

Or is there some pending work already?

No, didn't yet look into them.

kode54 · 2024-06-10T16:44:44Z

I'll rebuild again tonight, but there was a bug I noticed with Fullscreen toggle, if toggling from a Maximized window. If toggling from Maximized, it wants to resize the window, but fit it within the same position offset as Maximized, and leaves the dock overlay onscreen.

Consolatis force-pushed the chase/0.18 branch from 7d6aa25 to 2abef4a Compare March 19, 2024 11:36

This comment was marked as resolved.

Sign in to view

Consolatis added this to the wlroots 0.18.x milestone Apr 10, 2024

Consolatis force-pushed the chase/0.18 branch from 2abef4a to 00a3168 Compare April 23, 2024 11:58

Consolatis mentioned this pull request Apr 24, 2024

todo for wlroots 0.18 #1260

Closed

3 tasks

Consolatis force-pushed the chase/0.18 branch 2 times, most recently from b63309f to faf33a7 Compare May 9, 2024 11:34

johanmalm mentioned this pull request May 19, 2024

Johan's Meta Issue - to bind a few others #1622

Open

11 tasks

Consolatis force-pushed the chase/0.18 branch from 657c759 to fbd62be Compare May 24, 2024 10:05

Consolatis force-pushed the chase/0.18 branch 3 times, most recently from 0008d36 to 4f3fa22 Compare May 31, 2024 03:35

Consolatis added 19 commits June 10, 2024 05:13

[wip] version dep to 0.18

88bd600

[wip] CI: unbreak wlroots build

1e0acef

[revert later] CI: fix runtime tests due to libwayland bug

efef879

Already fixed in its master branch but not yet available in a release. https://gitlab.freedesktop.org/wayland/wayland/-/merge_requests/366

[wip] chase: output->pending (MR 4202)

8ea1341

[wip] chase: output_layout_req_wl_display (MR 4310)

dfb59bc

[wip] chase: tearing hint (MR #4480)

edfa013

[wip] chase: tablet_pad to tablet (MR #4585)

8dd5460

[wip] chase: input inhibit removal (MR #4440)

6b2e6fb

[wip] chase: relative axis direction (MR #4003)

4c81ad8

[wip] chase: presentation is now handled internally (MR #4482)

6f78d23

[wip] chase: use wayland pointer enums rather than wlr ones (MR #4575)

c99833b

[wip] chase: backend create now requires event_loop arg

1111d92

[dirty] chase: xdg new toplevel (parent-less popups missing) (MR 4345)

8556584

[wip] chase: move xdg destroy signal to toplevel / popup (MR 4345)

b2fb864

[dirty] chase: request initial configure (decoration callback missing…

b59c4ae

…) (MR 4396)

[wip] chase: let wlr_scene track damage again (MR #4253)

da1f2aa

[wip] disable restacking of unmanaged windows (MR #4052)

1e16acb

[revert_later] debug frame timing + add FPS counter

30251f6

[wip] chase: version pkgconfig, headers, and library (MR 4614)

0d2b711

Consolatis force-pushed the chase/0.18 branch from 4f3fa22 to 0d2b711 Compare June 10, 2024 03:18

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Update for wlroots 0.18 #1641

Update for wlroots 0.18 #1641

Consolatis commented Mar 19, 2024 •

edited

jlindgren90 commented Mar 19, 2024

This comment was marked as resolved.

This comment was marked as resolved.

Consolatis commented Apr 23, 2024 •

edited

kode54 commented Apr 24, 2024 •

edited by Consolatis

kode54 commented Apr 26, 2024

kode54 commented May 1, 2024

Consolatis commented May 10, 2024

johanmalm commented May 12, 2024

Consolatis commented May 12, 2024

kode54 commented May 24, 2024

Consolatis commented May 24, 2024

kode54 commented May 24, 2024

droc12345 commented May 24, 2024

kode54 commented May 25, 2024

kode54 commented May 26, 2024

droc12345 commented Jun 1, 2024 •

edited

Consolatis commented Jun 1, 2024 •

edited

droc12345 commented Jun 1, 2024 •

edited

kode54 commented Jun 10, 2024

Consolatis commented Jun 10, 2024

kode54 commented Jun 10, 2024

Update for wlroots 0.18 #1641

Are you sure you want to change the base?

Update for wlroots 0.18 #1641

Conversation

Consolatis commented Mar 19, 2024 • edited

jlindgren90 commented Mar 19, 2024

This comment was marked as resolved.

This comment was marked as resolved.

Consolatis commented Apr 23, 2024 • edited

kode54 commented Apr 24, 2024 • edited by Consolatis

kode54 commented Apr 26, 2024

kode54 commented May 1, 2024

Consolatis commented May 10, 2024

johanmalm commented May 12, 2024

wlroots chasing - can be done on master now (I think)

wlroots chasing required on 0.18 PR

wlroots chasing not required by labwc

Things in 0.18 PR that may need more work

xdg-shell (toplevels and popups)

XWayland re-stacking

Aside

Consolatis commented May 12, 2024

kode54 commented May 24, 2024

Consolatis commented May 24, 2024

kode54 commented May 24, 2024

droc12345 commented May 24, 2024

kode54 commented May 25, 2024

kode54 commented May 26, 2024

droc12345 commented Jun 1, 2024 • edited

Consolatis commented Jun 1, 2024 • edited

droc12345 commented Jun 1, 2024 • edited

kode54 commented Jun 10, 2024

Consolatis commented Jun 10, 2024

kode54 commented Jun 10, 2024

Consolatis commented Mar 19, 2024 •

edited

Consolatis commented Apr 23, 2024 •

edited

kode54 commented Apr 24, 2024 •

edited by Consolatis

wlroots chasing required on `0.18` PR

Things in `0.18 PR` that may need more work

droc12345 commented Jun 1, 2024 •

edited

Consolatis commented Jun 1, 2024 •

edited

droc12345 commented Jun 1, 2024 •

edited