1531294 - Performance issues on Sketchfab.com with WebRender on

Status: VERIFIED FIXED

(Core :: Graphics: WebRender, defect, P2)

Description

[Rares Doghi, Desktop QA]

Attachment: 2019-02-28_13h10_15.mp4

[Affected versions]:
Beta 67.0a1

Steps :

1. Launch the Firefox Browser and reach https://sketchfab.com/3d-models/the-old-knight-fb720a9fc67443a6907e849a2915f7b7
2. Wait for the 3D image to load and then move it around, zoom in and out or rotate the image.

Expected Results :

The 3D image should move around and rotate smooth and without issues.

Actual Results :

With WebRender on the 3D image is lagging a lot and drops frames when rotated.

This issue was tested on a Windows 10 with Nvidia GT 730.

Comment 1

[Jeff Muizelaar [:jrmuizel]]

How well does it run for you with WebRender off?

Comment 2

[Jeff Muizelaar [:jrmuizel]]

I also notice a performance difference in fullscreen mode.

Sotaro can you reproduce this? If you can. Do you want to take a look at it?

Comment 3

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

It seems better to uplift Bug 1526213 at first to optimize canvas.

Comment 4

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

(In reply to Sotaro Ikeda [:sotaro] from comment #3)

It seems better to uplift Bug 1526213 at first to optimize canvas.

I confirmed that fix of Bug 1526213 suppressed doc.build_frame() call during the STR in comment 0.

Comment 5

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

I checked the STR of comment 0 with Firefox nightly on my Win10 Intel PC(Intel(R) HD Graphics P530). The performance was very bad compared to non-WebRender(Direct3D 11 (Advanced Layers)). Though on my win10 desktop PC(NVIDIA GeForce GTX 960), I did not saw the performance problem with nightly and with beta.

Comment 6

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

Profile during the STR of comment 0 on my Win10 Intel PC(Intel(R) HD Graphics P530) with nightly.

https://perfht.ml/2tKrwqI

Main thread of content was blocked at rx::Context11::flush() very long time. The blocked function was actually ID3D11DeviceContext::Flush(). And Render thread was also busy. From it, GPU driver seemed very busy and it seemed to cause the problem.

Comment 7

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

I am going to look into the bug a bit more.

Comment 8

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

It seems better to check if WaitForGPUQuery() works for the problem.

Comment 9

[Rares Doghi, Desktop QA]

Hi Jeff, it runs the same at first a bit laggy but after a few seconds it starts moving around a lot smoother as with WebRender ON it's still causing lag and dropped frames, and yes it's even more noticeable in Full Screen.

Comment 10

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

Attachment: wip patch - Use ID3D11Query

Comment 11

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

The following is win64-pgo build by applying attachment 9047666 [details] [diff] [review] to latest m-c.

https://treeherder.mozilla.org/#/jobs?repo=try&revision=fd3cc82e790cc536923915579ed47c9468a77b3d

Comment 12

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

:jrmuizel, can you check if the win64-pgo build in comment 11 mitigate the problem?

Comment 13

[Jeff Muizelaar [:jrmuizel]]

(In reply to Sotaro Ikeda [:sotaro] from comment #12)

:jrmuizel, can you check if the win64-pgo build in comment 11 mitigate the problem?

This build runs much better for me.

Comment 14

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

Thanks for checking! The build run a lot better also for me. When the problem happened, too many gpu requests seemed to freeze D3D api on content main thread and render thread.

Comment 15

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

Hmm, attachment 9047666 [details] [diff] [review] caused a lot of crashes during test.
https://treeherder.mozilla.org/#/jobs?repo=try&revision=51cadae180aac7c10af21d4766c88bf41c5930f8

Comment 16

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

Attachment: temporal patch - Add log of SharedSurface_ANGLEShareHandles count

Comment 17

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

By applying attachment 9048411 [details] [diff] [review], when the problem happened, amount of SharedSurface_ANGLEShareHandles became more than 30.

On my high end desktop Win10 PC, the amount was 8 at most.

Comment 18

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

Attachment: wip patch - Use Fence

Comment 19

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

attachment 9048452 [details] [diff] [review] did not caused test failures. But it regressed glterrain a lot :(

https://treeherder.mozilla.org/#/jobs?repo=try&revision=7f3d601531c4d536c23277c7f670be4ce4d05da8

https://treeherder.mozilla.org/perf.html#/compare?originalProject=try&originalRevision=d8db6f07c1109ecb58ab4abbb21a48857454a581&newProject=try&newRevision=7f3d601531c4d536c23277c7f670be4ce4d05da8&framework=1

Comment 20

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

(In reply to Sotaro Ikeda [:sotaro] from comment #19)

attachment 9048452 [details] [diff] [review] did not caused test failures. But it regressed glterrain a lot :(

On possible workaround is that using Fence only when size of mRecycleTotalPool is large.

Comment 21

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

Attachment: wip patch - Use Fence with throttling

Comment 22

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

When I revert Bug 1526213, the problem seemed to be disappeared. "doc.build_frame()" seemed to work as throttling webgl task. Since Bug 1526213 fix, we could skip doc.build_frame() during canvas empty transaction. It caused to webgl task more rapid and d3d11 api stalled by too many gpu tasks.

Comment 23

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

By Bug 1526213, WebRenderBridgeParent::RecvEmptyTransaction() had a bug. It does not handle a case that resource update is handled by WebRenderTextureHostWrapper. In this case, txn.IsResourceUpdatesEmpty() became true and the function thought there was no resource update and the function returned DidComposite soon to client side. Then it caused over production at SharedSurface_ANGLEShareHandle().

Comment 24

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

Attachment: Bug 1531294 - Fix RecvEmptyTransaction() as to handle WebRenderTextureHostWrapper usage case

Comment 25

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

attachment 9049875 [details] is going to loose the performance gain of glterrain of Bug 1526213, but it is correct behavior. Bug 1526213 made a huge over production of SharedSurface_ANGLEShareHandle(). More than 30 SharedSurface_ANGLEShareHandle instances were used for one WebGL. glterrain score does not show actual valid performance. It should be addressed.

https://treeherder.mozilla.org/perf.html#/compare?originalProject=try&originalRevision=3e5e9cfbb44a88e32208323ef7d9a18a8c177f6b&newProject=try&newRevision=fc3d968b84776c871c012abd69185edfaa943f4f&framework=1

Comment 26

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

Confirmed that attachment 9049875 [details] address the problem of comment 0 on my Win10 Intel laptop.

Comment 27

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/ec39fd695adb
Fix RecvEmptyTransaction() as to handle WebRenderTextureHostWrapper usage case r=mattwoodrow

Comment 28

[Dorel Luca [:dluca]]

Backed out changeset ec39fd695adb (bug 1531294) for webrender failures

Log:
https://treeherder.mozilla.org/logviewer.html#/jobs?job_id=233291875&repo=autoland&lineNumber=12947

Push with failures:
https://treeherder.mozilla.org/#/jobs?repo=autoland&revision=ec39fd695adb0548e5c689d839843fa2e9b3acbd

Backout:
https://hg.mozilla.org/integration/autoland/rev/ede970649f873246c6f23b83924345827fb6d5c3

Comment 29

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

(In reply to Dorel Luca [:dluca] from comment #28)

Backed out changeset ec39fd695adb (bug 1531294) for webrender failures

Log:
https://treeherder.mozilla.org/logviewer.html#/jobs?job_id=233291875&repo=autoland&lineNumber=12947

It is weird, the log said that webrender was not enabled. The change only affect to webrender enabled.

Tests are grouped as Wr2, but it does not mean webrender. The Wr2 is a group of Web platform tests.

Comment 30

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

I wonder if Bug 1531863 is culprit.

Comment 31

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

(In reply to Sotaro Ikeda [:sotaro] from comment #29)

(In reply to Dorel Luca [:dluca] from comment #28)

Backed out changeset ec39fd695adb (bug 1531294) for webrender failures

Log:
https://treeherder.mozilla.org/logviewer.html#/jobs?job_id=233291875&repo=autoland&lineNumber=12947

It is weird, the log said that webrender was not enabled. The change only affect to webrender enabled.

latest auto land already caused same test failures.

https://treeherder.mozilla.org/#/jobs?repo=try&revision=608247da49a0dab39f76fb8c2d69c9984b7ca342

Comment 32

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

And confirmed that the failure did not happen with attachment 9049875 [details].

https://treeherder.mozilla.org/#/jobs?repo=try&revision=269be65b360338f0e7e1181bdbb8ed253e4476f6

Comment 33

[Sotaro Ikeda [:sotaro OOO July/27 - July/31]]

https://hg.mozilla.org/integration/mozilla-inbound/rev/581eed270cffaf9c19285ce97029aa5516123829

Comment 34

[Bogdan Tara[:bogdan_tara | bogdant]]

https://hg.mozilla.org/mozilla-central/rev/581eed270cff

Comment 35

[Rares Doghi, Desktop QA]

Hi this issue is Verified as fixed in Firefox 67.0a1 (2019-03-13)on Windows 10 with Nvidia GT 730.