[Sync] Remove now-unneeded initialized_mutex_.

[Sync] Remove now-unneeded initialized_mutex_.

initialized_ is only accessed by the sync thread, so it doesn't need
a mutex.

BUG=
TEST=

Committed: http://src.chromium.org/viewvc/chrome?view=rev&revision=89263

[akalin, 2011-06-09 01:18:36]

+tim for review

Note that this depends on my sync thread/core thread combination CL.

[tim (not reviewing), 2011-06-10 01:53:20]

http://codereview.chromium.org/7003077/diff/3001/chrome/browser/sync/engine/s...
File chrome/browser/sync/engine/syncapi.cc (right):

http://codereview.chromium.org/7003077/diff/3001/chrome/browser/sync/engine/s...
chrome/browser/sync/engine/syncapi.cc:1786: OnInitializationComplete());
So, in the reviews yesterday to make our observers threadsafe, I noted I didn't
understand where the races were because we should always all init here in a
deterministic fashion.  Here, you don't do the CopyObservers trick (which was
added to MarkAndNotify yesterday), which suggests that there is in fact no race
as I thought.

Am I smoking crack?!

[akalin, 2011-06-14 17:10:25]

http://codereview.chromium.org/7003077/diff/3001/chrome/browser/sync/engine/s...
File chrome/browser/sync/engine/syncapi.cc (right):

http://codereview.chromium.org/7003077/diff/3001/chrome/browser/sync/engine/s...
chrome/browser/sync/engine/syncapi.cc:1786: OnInitializationComplete());
On 2011/06/10 01:53:20, timsteele wrote:
> So, in the reviews yesterday to make our observers threadsafe, I noted I
didn't
> understand where the races were because we should always all init here in a
> deterministic fashion.  Here, you don't do the CopyObservers trick (which was
> added to MarkAndNotify yesterday), which suggests that there is in fact no
race
> as I thought.
> 
> Am I smoking crack?!

The races occurred because the observer add/remove was happening on the
core/sync thread and the notifies were happening from threads that
WriteTransactions are on (UI, DB, etc.).  In this case, initialization happens
on the sync thread, and all the observers are on the sync thread.  We don't even
have an AuthHandler thread anymore, do we?

[akalin, 2011-06-14 21:27:23]

http://codereview.chromium.org/7003077/diff/3001/chrome/browser/sync/engine/s...
File chrome/browser/sync/engine/syncapi.cc (right):

http://codereview.chromium.org/7003077/diff/3001/chrome/browser/sync/engine/s...
chrome/browser/sync/engine/syncapi.cc:1786: OnInitializationComplete());
On 2011/06/10 01:53:20, timsteele wrote:
> So, in the reviews yesterday to make our observers threadsafe, I noted I
didn't
> understand where the races were because we should always all init here in a
> deterministic fashion.  Here, you don't do the CopyObservers trick (which was
> added to MarkAndNotify yesterday), which suggests that there is in fact no
race
> as I thought.
> 
> Am I smoking crack?!

Oh, I see what you were saying.  That was just a merge error, there definitely
needs to be a CopyObserver here (the unit tests caught it).

Fixed.

[akalin, 2011-06-15 17:12:45]

Ping!

[tim (not reviewing), 2011-06-15 18:33:06]

On 2011/06/15 17:12:45, akalin wrote:
> Ping!

LGTM

[commit-bot: I haz the power, 2011-06-15 20:17:22]

Can't apply patch for file chrome/browser/sync/engine/syncapi.cc.
While running patch -p1 --forward --force;
patching file chrome/browser/sync/engine/syncapi.cc
Hunk #2 succeeded at 1266 (offset 8 lines).
Hunk #3 succeeded at 1290 (offset 8 lines).
Hunk #4 FAILED at 1395.
Hunk #5 succeeded at 1583 (offset 5 lines).
Hunk #6 succeeded at 1736 (offset 5 lines).
Hunk #7 FAILED at 1781.
Hunk #8 FAILED at 1843.
Hunk #9 succeeded at 1867 (offset -6 lines).
Hunk #10 succeeded at 2553 (offset 3 lines).
Hunk #11 succeeded at 2989 (offset 2 lines).
3 out of 11 hunks FAILED -- saving rejects to file
chrome/browser/sync/engine/syncapi.cc.rej