1803016 - [Wayland] Screesharing using xdg-desktop-portal-wlr on Sway crashes latest Nightly: wl_surface@103: error 2: Buffer size (1001x64) is not divisible by scale (2)

Status: RESOLVED FIXED

(Core :: Widget: Gtk, defect, P3)

Description

[Jonny Mako]

Steps to reproduce:

Trying to screenshare (with zoom) in Sway via pipewire and xdg-desktop-portal-wlr crashes Firefox (109.0a1 (2022-11-28) (64-bit)):

To reproduce, On sway version 1.9-dev-dca0bb57 and latest xdg-desktop-portal-wlr/pipewire:

1. Start zoom session on Firefox
2. Hit Share screen
3. Firefox asks for permission to see your screen with "Use operating system settings" selected in the drop down menu. (This is the only option available)
4. Click "Allow"
5. A "Sharing Indicator" window pops up with the following message:
   "You are sharing another application window" and "Stop Sharing" button.
6. Simultaneously, you get an "overlay" that indicates which screen to share.
7. Click the screen you want to share.
8. <crash>

Here is the link to the report:
https://crash-stats.mozilla.org/report/index/e888f55f-832a-48b1-8434-31f680221129

Actual results:

Browser crashed

Expected results:

Browser start sharing screen.

Comment 1

[BugBot [:suhaib / :marco/ :calixte]]

The Bugbug bot thinks this bug should belong to the 'Core::WebRTC' component, and is moving the bug to that component. Please correct in case you think the bot is wrong.

Comment 2

[Jim Mathies [:jimm]]

wl_surface@92: error 2: Buffer size (1001x64) is not divisible by scale (2)
https://crash-stats.mozilla.org/report/index/e888f55f-832a-48b1-8434-31f680221129

An issue with Wayland and capture + painting, or maybe widget related? Bouncing to graphics for triage.

Comment 3

[Darkspirit]

(In reply to Jonny Mako from comment #0)

Sway

(In reply to Jim Mathies [:jimm] from comment #2)

wl_surface@92: error 2: Buffer size (1001x64) is not divisible by scale (2)

That's bug 1795851.

Comment 4

[Jonny Mako]

That bug does not seem to fix the issue in this case. I'm still getting the crash with latest nightly:

https://crash-stats.mozilla.org/report/index/08e7779d-0fe8-4108-bd57-7e1c10221212

Comment 5

[Jonny Mako]

With updated build id: https://crash-stats.mozilla.org/report/index/966e1293-b8e5-44fc-9bd3-0cfac0221212

Comment 6

[Darkspirit]

Thanks for testing!

Comment 7

[BugBot [:suhaib / :marco/ :calixte]]

The bug is linked to a topcrash signature, which matches the following criteria:

• Top 5 desktop browser crashes on Linux on beta
• Top 5 desktop browser crashes on Linux on release

:stransky, could you consider increasing the severity of this top-crash bug?

For more information, please visit auto_nag documentation.

Comment 8

[Jonny Mako]

FYI, here is a discussion on Sway to the related wlroots assert to enforce wayland protocol that might be causing this crash:

https://github.com/swaywm/sway/pull/6844#issuecomment-1353478644

and commit:

https://gitlab.freedesktop.org/wlroots/wlroots/-/commit/18595000f3a21502fd60bf213122859cc348f9af

Comment 9

[Martin Stránský [:stransky] (ni? me)]

Do I understand correctly that this crash is about cursor surfaces? In such case it's a bug in Gtk and not in Firefox itself.

Comment 10

[Jonny Mako]

I've opened an issue with GTK:
https://gitlab.gnome.org/GNOME/gtk/-/issues/5463

Comment 11

[Bob Hood [:bhood]]

Martin, could you please apply a severity to this one?

Comment 12

[BugBot [:suhaib / :marco/ :calixte]]

Based on the topcrash criteria, the crash signature linked to this bug is not a topcrash signature anymore.

For more information, please visit auto_nag documentation.

Comment 13

[Jakob Jarmar]

(In reply to Martin Stránský [:stransky] (ni? me) from comment #9)

Do I understand correctly that this crash is about cursor surfaces? In such case it's a bug in Gtk and not in Firefox itself.

I believe it's the other way around: wlroots allows cursor surfaces to use the protocol incorrectly, but other surfaces trying to do so will trigger the error seen in this bug.

Comment 14

[BugBot [:suhaib / :marco/ :calixte]]

Sorry for removing the keyword earlier but there is a recent change in the ranking, so the bug is again linked to a topcrash signature, which matches the following criterion:

• Top 10 desktop browser crashes on nightly

For more information, please visit auto_nag documentation.

Comment 15

[Martin Stránský [:stransky] (ni? me)]

Attachment: Bug 1803016 [Wayland] Don't commit wl_buffer if buffer scale doesn't match its size r?rmader,emilio

Comment 17

[BugBot [:suhaib / :marco/ :calixte]]

Based on the topcrash criteria, the crash signature linked to this bug is not a topcrash signature anymore.

For more information, please visit auto_nag documentation.

Comment 18

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/1068e0955cfb
[Wayland] Don't commit wl_buffer if buffer scale doesn't match its size r=emilio

Comment 19

[Sandor Molnar[:smolnar]]

https://hg.mozilla.org/mozilla-central/rev/1068e0955cfb

Comment 20

[BugBot [:suhaib / :marco/ :calixte]]

The patch landed in nightly and beta is affected.
:stransky, is this bug important enough to require an uplift?

• If yes, please nominate the patch for beta approval.
• If no, please set status-firefox112 to wontfix.

For more information, please visit auto_nag documentation.

Comment 22

[BugBot [:suhaib / :marco/ :calixte]]

Copying crash signatures from duplicate bugs.

Comment 25

[Jonny Mako]

I'm still seeing this crash in latest nightly: 114.0a1 (2023-04-18) (64-bit)

Comment 26

[Martin Stránský [:stransky] (ni? me)]

(In reply to Jonny Mako from comment #25)

I'm still seeing this crash in latest nightly: 114.0a1 (2023-04-18) (64-bit)

Please file a new bug for it and attach a backtrace of the crash there (from about:crashes).
Thanks.

Comment 30

[Jonny Mako]

(In reply to Martin Stránský [:stransky] (ni? me) from comment #26)

(In reply to Jonny Mako from comment #25)

I'm still seeing this crash in latest nightly: 114.0a1 (2023-04-18) (64-bit)

Please file a new bug for it and attach a backtrace of the crash there (from about:crashes).
Thanks.

I can reliable reproduce this crash, but the crash reporter does not open. How do I force the crash reporter to open?

Comment 31

[Jonny Mako]

(In reply to Jonny Mako from comment #30)

(In reply to Martin Stránský [:stransky] (ni? me) from comment #26)

(In reply to Jonny Mako from comment #25)

I'm still seeing this crash in latest nightly: 114.0a1 (2023-04-18) (64-bit)

Please file a new bug for it and attach a backtrace of the crash there (from about:crashes).
Thanks.

I can reliable reproduce this crash, but the crash reporter does not open. How do I force the crash reporter to open?

Also there are no crashes in Help->More Troubleshooting Information

Comment 32

[Jonny Mako]

Attachment: ff.log

Here is the next best thing I could think of, given I'm unable to get the crash reporter to appear. Attaching wayland debug log.

Comment 33

[Jonny Mako]

If someone can help me fix the issue with crash reporter, I'll open a new report as requested by Martin @stransky.

Comment 34

[Jonny Mako]

Also, the crash happens when display scale factor is anything other than 1. My entry in Sway config is:

output eDP-1 pos 0 0 mode 3840x2160 scale 1.5

Comment 35

[Darkspirit]

Let's continue in bug 1832158 (slightly different crash reason).