1690216 - [XMonad window manager] Developer Edition doesn't render immediately after update

Status: RESOLVED FIXED

(Core :: Graphics, defect, P3)

Description

[18gatenmaker6]

User Agent: Mozilla/5.0 (X11; Linux x86_64; rv:86.0) Gecko/20100101 Firefox/86.0

Steps to reproduce:

(My version is 86.0b3 (64-bit) to be precise)
Start Firefox, go to some webpage, then unfocus from Firefox. Make sure Firefox is not in sight (I switched to another workspace in XMonad). Then go back to Firefox. Don't give any mouse/keyboard inputs afterwards. (render bug will go away after scrolling the page, for example.)

Actual results:

A render bug occurs. I am unable to make a screenshot, but for a variable time Firefox just doesn't render.

Expected results:

No render bugs should have occured, the bugs look weird and broken.

Comment 1

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

Bugbug thinks this bug should belong to this component, but please revert this change in case of error.

Comment 2

[Andrew Osmond [:aosmond] (he/him)]

Would you be able to capture a screen recording of this (Ctrl+Alt+Shift+R to start/stop, saved in $HOME/Videos by default)? Also, could you attach your about:support contents? Thanks!

Comment 3

[18gatenmaker6]

Attachment: about:support

Sure thing, here is my about:support in advance: ```

```

Comment 4

[18gatenmaker6]

I did the recording while watching youtube, because if I use a video player, the render bug occurs longer. In the clip, I scroll a bit to make the screen normal again. link

Comment 5

[Lee Salzman [:lsalzman]]

Attachment: Bug 1690216 - report window damage back to SwCompositor from RenderCompositorSWGL::StarCompositing. r?mattwoodrow

RenderCompositorSWGL knows the true dirty region in which it needs to composite
from StartRemoteDrawingInRegion. However, SwCompositor only knows the dirty rects
that WR reports to it, which may only be a subset of what we actually need to
redraw the window. This modifies StartCompositing/start_compositing so they
can report back to us the unioned dirty rect that we actually need to consider.

Comment 6

[Lee Salzman [:lsalzman]]

Attachment: Bug 1690216 - Report whether to redraw on invalidation to WR. r?mattwoodrow

This requires us to plumb CompositorCapabilities to support the extra field.
This is complicated by the fact that since it is a Rust struct, it has no
default constructor that can pass through to C++ via bindings, so every
one of our RenderCompositors was forced to manually initialize fields. To
get around this brittle footgun, instead the structure is initialized on
the Rust side, and RenderCompositor's are encouraged to only change fields
that actually diverge from the defaults as passed in via pointer.

Finally, we can then do what we need to do, which is just to send the
ForceRedraw message that needs to happen based on what we know about
CompositorCapabilities.

Comment 7

[Lee Salzman [:lsalzman]]

Attachment: Bug 1690216 - Clarify that StartRemoteDrawingInRegion does not actually modify the region. r?mattwoodrow

It caused us substantial confusion investigating this bug under the belief that
StartRemoteDrawingInRegion may have been modifying the dirty region. None of our
existing widget code anymore uses the API in this way, so it makes sense to just
force this dirty region to be const so that we no longer support the assumption
and alleviate confusion in the future about how our widget code actually behaves.

Depends on D106246

Comment 8

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/45b36995cfd6
Report whether to redraw on invalidation to WR. r=mattwoodrow

Comment 9

[Andreea Pavel [:apavel]]

Backed out for build bustages at /rules.mk

Push with failures: https://treeherder.mozilla.org/jobs?repo=autoland&selectedTaskRun=HRC8_5VzSLO1al-W09VFzw.0&resultStatus=testfailed%2Cbusted%2Cexception%2Cretry%2Cusercancel&revision=45b36995cfd6ead7bcb3f0f8a390be6ef257ee04

failure log: https://treeherder.mozilla.org/logviewer?job_id=331002205&repo=autoland&lineNumber=20281

Backout: https://hg.mozilla.org/integration/autoland/rev/51d922846c42c42982ccc55fdc701ee344252101

Comment 10

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/b16a82dcc68b
Report whether to redraw on invalidation to WR. r=mattwoodrow

Comment 11

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/6629b9999267
Clarify that StartRemoteDrawingInRegion does not actually modify the region. r=mattwoodrow

Comment 12

[Natalia Csoregi [:nataliaCs]]

https://hg.mozilla.org/mozilla-central/rev/b16a82dcc68b

Comment 13

[Sandor Molnar[:smolnar]]

https://hg.mozilla.org/mozilla-central/rev/6629b9999267

Comment 14

[Lee Salzman [:lsalzman]]

Can you verify if the patch fixes the bug for you? In my local reproduction, it did seem to do the job for me.

Comment 15

[18gatenmaker6]

You mean building from source or...?

Comment 16

[Lee Salzman [:lsalzman]]

(In reply to 18gatenmaker6 from comment #15)

You mean building from source or...?

It should be rolled into nightly builds by tonight.

Comment 17

[Andrew Osmond [:aosmond] (he/him)]

(In reply to Lee Salzman [:lsalzman] from comment #16)

(In reply to 18gatenmaker6 from comment #15)

You mean building from source or...?

It should be rolled into nightly builds by tonight.

It should be in 20210225092306 nightly already.

Comment 18

[Lee Salzman [:lsalzman]]

Comment on attachment 9205044 [details]
Bug 1690216 - Report whether to redraw on invalidation to WR. r?mattwoodrow

Beta/Release Uplift Approval Request

• User impact if declined: This resolves a Linux usability regression with Software WebRender. Any overlapping windows, workspace switching, etc. will cause the window to fail to repaint properly.
• Is this code covered by automated tests?: Yes
• Has the fix been verified in Nightly?: Yes
• Needs manual test from QE?: No
• If yes, steps to reproduce:
• List of other uplifts needed: None
• Risk to taking this patch: Low
• Why is the change risky/not risky? (and alternatives if risky): Just forces an extra redraw when the window is damaged, but otherwise doesn't make other changes to how things are drawn.
• String changes made/needed:

Comment 19

[18gatenmaker6]

Okay, I will try to reproduce the bug tomorrow on the latest nightly build (right now: https://ftp.mozilla.org/pub/firefox/nightly/2021/02/2021-02-27-09-44-58-mozilla-central/firefox-88.0a1.en-US.linux-x86_64.tar.bz2 )

Comment 20

[18gatenmaker6]

https://imgur.com/a/rHtWVnV

Comment 21

[Ryan VanderMeulen [:RyanVM]]

Comment on attachment 9205044 [details]
Bug 1690216 - Report whether to redraw on invalidation to WR. r?mattwoodrow

This needs a rebased patch for Beta.

Comment 22

[Lee Salzman [:lsalzman]]

Attachment: Bug 1690216 - Report whether to redraw on invalidation to WR (beta 87). r=mattwoodrow

A file just changed name, so a fairly uneventful rebase for 87.

Comment 23

[Julien Cristau [:jcristau]]

Comment on attachment 9205953 [details] [diff] [review]
Bug 1690216 - Report whether to redraw on invalidation to WR (beta 87). r=mattwoodrow

approved for 87.0b5

Comment 24

[Julien Cristau [:jcristau]]

https://hg.mozilla.org/releases/mozilla-beta/rev/c6be492a28c6