1061014 - Enable e10s Linux opt browser chrome tests

Status: RESOLVED FIXED

(Release Engineering :: General, defect)

Description

[Bill McCloskey [inactive unless it's an emergency] (:billm)]

Attachment: enable-e10s-bc

Once the dependent bugs land, we should be green enough to start running browser-chrome tests on Linux opt.

Comment 1

[Bill McCloskey [inactive unless it's an emergency] (:billm)]

Attachment: enable-e10s-bc v2

RyanVM pointed out that we can enable on ASAN too since it doesn't do window leak detection.

Comment 2

[Jim Mathies [:jimm]]

Comment on attachment 8482052 [details] [diff] [review]
enable-e10s-bc v2

Review of attachment 8482052 [details] [diff] [review]:
-----------------------------------------------------------------

Looking good on holly. I retriggered 5 times for each set. Ryan, how's your comfort level here with turning these on?

https://tbpl.mozilla.org/?tree=Holly

Comment 3

[Jim Mathies [:jimm]]

looks like we can turn on debug as well, lets see how the retriggers come out.

Comment 4

[Ryan VanderMeulen [:RyanVM]]

Looks like it causes bug 945182 to make a not-so-triumphant return ~50% of the time? :(

Comment 5

[Ryan VanderMeulen [:RyanVM]]

Comment on attachment 8482052 [details] [diff] [review]
enable-e10s-bc v2

Review of attachment 8482052 [details] [diff] [review]:
-----------------------------------------------------------------

Looks like debug hits a few leaks here and there, but opt looks good to me.

Comment 6

[Armen [:armenzg]]

Comment on attachment 8482052 [details] [diff] [review]
enable-e10s-bc v2

Review of attachment 8482052 [details] [diff] [review]:
-----------------------------------------------------------------

It looks good to me.
https://tbpl.mozilla.org/?tree=Holly&jobname=e10s

Comment 7

[Jim Mathies [:jimm]]

https://hg.mozilla.org/build/buildbot-configs/rev/82429c54883f

thanks for the help guys!

Comment 8

[Jim Mathies [:jimm]]

Hmm, I may have jumped the gun, bug 1061014 is required for this, and its still on inbound.

Comment 9

[Jim Mathies [:jimm]]

(In reply to Jim Mathies [:jimm] from comment #8)
> Hmm, I may have jumped the gun, bug 1061014 is required for this, and its
> still on inbound.

sorry - bug 1060620

Comment 10

[Armen [:armenzg]]

It is best to back out and wait until Monday when the changes will most likely be merged into all other trunk trees.

Mind backing out if it works for you?

Comment 11

[Ryan VanderMeulen [:RyanVM]]

It's highly likely that inbound will be merged to m-c before the next reconfig.

Comment 12

[Armen [:armenzg]]

(Keeping open until it gets into the masters)

What about trees like fx-team or even people pushing their local checkouts to try?

If Ryan is happy, I'm happy as well.

Comment 13

[Ryan VanderMeulen [:RyanVM]]

m-c is merged back around to the integration trees as soon as all the merges to m-c are completed. As for local trees, I don't see why that'd be any different than with any other tests regardless of timing.

Comment 14

[Jim Mathies [:jimm]]

Did we miss something here, on inbound I see bc red runs, with

https://tbpl.mozilla.org/php/getParsedLog.php?id=47474750&tree=Mozilla-Inbound&full=1

Required config file not set! (use --config-file option)
program finished with exit code 255

Comment 15

[Armen [:armenzg]]

Attachment: e10s.diff

We forgot this.
I'm asking coop for a reconfig for it.

Comment 16

[Armen [:armenzg]]

Comment on attachment 8484954 [details] [diff] [review]
e10s.diff

r=coop
https://hg.mozilla.org/build/buildbot-configs/rev/04cc9fbd7e15

The reconfig will come in 30 mins.

Comment 17

[Chris Cooper [:coop] (he/him)]

Merged to production, and deployed.

Comment 18

[Ryan VanderMeulen [:RyanVM]]

Jobs are running green, thanks!

Comment 19

[Jim Mathies [:jimm]]

Thanks for the help!

Comment 20

[Armen [:armenzg]]

You're welcome!
Thanks coop!