1416651 - Bad performance on running tresize talos suite

Status: RESOLVED INCOMPLETE

(Core :: Graphics: WebRender, defect, P3)

Description

[Vincent Liu[:vliu]]

See [1] and [2] for the Comparison link in detail. They showed that

tresize opt e10s in Linux x64 opt: 33.71%
tresize opt e10s stylo_disabled Linux x64 opt: 33.38%

tresize opt e10s in Windows 10 x64 opt: 89.89%
tresize opt e10s stylo_disabled in Windows 10 x64 opt: 78.57%

[1]: Comparison link on Linux x64 opt
     https://treeherder.mozilla.org/perf.html#/compare?originalProject=try&originalRevision=226893bbeb3e&newProject=try&newRevision=624d8ef848c2d76eed6a2854e03e3c54950ffd12&framework=1
[2]: Comparison link on Windows 10 x64 opt
     https://treeherder.mozilla.org/perf.html#/compare?originalProject=try&originalRevision=e20b42426413&newProject=try&newRevision=ec5b6c878d2abaf1c98a5cdc1284fd38625f1822&framework=1

Comment 1

[Vincent Liu[:vliu]]

I tried to sent try with geckoProfile enabled for both disable/enable wr. [1] and [2] were the try result.

disable wr:
[1]: https://treeherder.mozilla.org/#/jobs?repo=try&revision=77301c9c3e7f8038e78c838ea5292180db65fde8

enable wr
[2]: https://treeherder.mozilla.org/#/jobs?repo=try&revision=cb38bd99ffc84342caeee0363d2beaa088c74d12

From looked into geckoProfile above, I found the execution time of a sync call PCompositorBridge::Msg_FlushRendering in wr disabled was many times shorter than enabled case. It is worth taking time to see what is going on at this part.

Comment 2

[Vincent Liu[:vliu]]

(In reply to Vincent Liu[:vliu] from comment #1)
> I tried to sent try with geckoProfile enabled for both disable/enable wr.
> [1] and [2] were the try result.
> 
> disable wr:
> [1]:
> https://treeherder.mozilla.org/#/
> jobs?repo=try&revision=77301c9c3e7f8038e78c838ea5292180db65fde8
> 
> enable wr
> [2]:
> https://treeherder.mozilla.org/#/
> jobs?repo=try&revision=cb38bd99ffc84342caeee0363d2beaa088c74d12
> 
> From looked into geckoProfile above, I found the execution time of a sync
> call PCompositorBridge::Msg_FlushRendering in wr disabled was many times
> shorter than enabled case. It is worth taking time to see what is going on
> at this part.

A more look into it, the most time in this Msg_FlushRendering sync call is calling WaitFlushed(). 
This WaitFlushed() is used to waits all pending WebRender task are done. WebRender needs this wait in the all run because it works as multi threaded way. Also, there is no better way to become an async call in the current design.

Comment 3

[Vincent Liu[:vliu]]

I will set this bug to Nobody because it might have above one bottlenecks affecting it. If I found any of them, I will file a new bug and depend to this one.

Comment 4

[Vincent Liu[:vliu]]

Based on Bug 1416630 Comment 1, it still worse than disabling wr. I will still keep this open.


linux64[1]: 
tresize opt e10s: 38.44%
tresize opt e10s stylo_disabled: 36.57%

Windows 10 x64 opt[2]:
tresize opt e10s: 79.39%

[1]: https://treeherder.mozilla.org/perf.html#/compare?originalProject=try&originalRevision=03107377dac6&newProject=try&newRevision=26095b988f06e8cb097118dd9ba051966fe9b195&framework=1
[2]: https://treeherder.mozilla.org/perf.html#/compare?originalProject=try&originalRevision=051d048108f2&newProject=try&newRevision=52ee54a2130bcea167e32308374a6ccd0d5950b4&framework=1

Comment 5

[Vincent Liu[:vliu]]

I will take this bug for better tracking.

Comment 6

[Vincent Liu[:vliu]]

Here to attach a geckoProfie link I run in mac machine for more detailed profiling.

[1]: https://perfht.ml/2Aj5YnC

Comment 7

[Peter Chang[:pchang]]

After discussed with Vincent, this should be related to the longer pipelines in WR, compared to gecko. We could deal with this after solving other critical perf bugs.

Comment 8

[Kartikaya Gupta (email:[e-mail])]

https://treeherder.mozilla.org/perf.html#/graphs?series=autoland,1645986,1,1&series=autoland,1682853,1,1&series=autoland,1650995,1,1&series=autoland,1683335,1,1

Windows and Linux are both worse on tresize with WR enabled than WR disabled. Bumping to P1 for Windows regression.

Comment 9

[Kartikaya Gupta (email:[e-mail])]

The WR update in bug 1466549 helps on both Linux and Windows, but there's still a bit of a regression compared to non-WR.

Comment 10

[Kartikaya Gupta (email:[e-mail])]

This got better on linux64-qr because of bug 1471962, but it's still slightly worse than non-WR on both Linux and Windows.

Comment 11

[Kartikaya Gupta (email:[e-mail])]

bug 1476636/bug 1455422 made this slightly worse on Linux.

Comment 12

[Kartikaya Gupta (email:[e-mail])]

Bug 1477783 helped but not enough. Still worse on both Linux and Windows.

Comment 14

[Maire Reavy [:mreavy]]

We can't release this to the field, but we can let this ride to beta.  However, we want to investigate and understand the bad perf numbers asap.

Comment 15

[Matt Woodrow (:mattwoodrow)]

Once bug 1489189 merges, then we have pretty reasonable results.

8.89 for WR vs 7.6 without.

I've spent some time investigating the test and it seems like we're measuring the right things here.

The remaining difference is pretty small, I don't think we need to block release on this.

Comment 16

[Sotaro Ikeda [:sotaro]]

Updated url
https://treeherder.mozilla.org/perf.html#/graphs?series=autoland,1787201,1,1&series=autoland,1787203,1,1&series=autoland,1787215,1,1&series=autoland,1787214,1,1

Comment 17

[Glenn Watson [:gw]]

There is only WR.