583339 - flip switch to build with sync

Status: RESOLVED FIXED

(Firefox :: General, defect)

Description

[Paul O'Shannessy [:zpao] (not bugmail, email directly)]

Attachment: Patch v0.1

Comment 1

[Paul O'Shannessy [:zpao] (not bugmail, email directly)]

Comment on attachment 461660 [details] [diff] [review]
Patch v0.1

Ted - not sure if you're the one or if Dolske can do it, but you were my best guess.

Comment 2

[Justin Dolske [:Dolske]]

Comment on attachment 461660 [details] [diff] [review]
Patch v0.1

r+, with the (obvious) caveat that this only lands once we've finished our sanity testing tomorrow.

Comment 3

[Philipp von Weitershausen [:philikon]]

The mobile guys are seeing assertions fail in nsURLHelper.cpp when MOZ_SERVICES_SYNC=1 is turned on (bug 580508). Not a hard blocker I guess but would be good to fix before we land.

Comment 4

[Paul O'Shannessy [:zpao] (not bugmail, email directly)]

(In reply to comment #3)
> The mobile guys are seeing assertions fail in nsURLHelper.cpp when
> MOZ_SERVICES_SYNC=1 is turned on (bug 580508). Not a hard blocker I guess but
> would be good to fix before we land.

Ah yes. I had seen that but everything worked so I forgot to file it. Looks like we could poke biesi to review that ASAP.

Comment 5

[Paul O'Shannessy [:zpao] (not bugmail, email directly)]

Landed http://hg.mozilla.org/mozilla-central/rev/029f8bd8d06a

Fingers crossed that it sticks.

Comment 6

[Justin Dolske [:Dolske]]

Sadtimes, backed out (disabled).

The debug builders ("Bd") were orange due to Rlk > 0, looks like we missed this from the try pushes with everything else going on. There are some other oranges that may or may not be related, we'll look in the morning.

Comment 7

[Justin Dolske [:Dolske]]

(Not actually backed out, because hg is broken and the push failed. See bug 584001. We've killed the nightlies while this gets sorted out.)

Comment 8

[:Gavin Sharp [email: [e-mail]]]

Disabled by https://hg.mozilla.org/mozilla-central/rev/4daa2ea5747b

Comment 9

[Justin Dolske [:Dolske]]

Reenabled this morning for in-tree debugging with http://hg.mozilla.org/mozilla-central/rev/f6a4fd52d468

Disabled now, again, in http://hg.mozilla.org/mozilla-central/rev/b35930e655d5, to ensure availability of  a checkpoint (plus some other, unrealated, bugs to land after) for a no-Sync Beta 3.

Comment 10

[Mike Beltzner [:beltzner, not reading bugmail]]

I don't think we can switch Sync on for Beta 3 with the number and size of performance regressions. Can we get a summary of those regressions listed here and bugs filed on fixing them?

Comment 11

[Axel Hecht [:Pike]]

Is there a status update on this? Also, as this bug seems to be *the* sync-on bug, are the dependencies set up correctly?

Comment 12

[Philipp von Weitershausen [:philikon]]

(In reply to comment #11)
> Is there a status update on this?

We're working on plugging the leaks and are investigating the talos regressions.

> Also, as this bug seems to be *the* sync-on bug, are the dependencies set up correctly?

I'm tracking dependencies for the next merge in bug 584481 which in turn is a dependency of this bug.

Comment 13

[Mike Connor [:mconnor]]

Just down to the front-end fix for the Ts regressions.

Comment 14

[Philipp von Weitershausen [:philikon]]

http://hg.mozilla.org/mozilla-central/rev/8b618223ad69