Replace Atomic32 with webrtc/base/atomicops.h.

Replace Atomic32 with webrtc/base/atomicops.h.

One atomic type is enough.

BUG=
R=kwiberg@webrtc.org, tommi@webrtc.org

[pbos-webrtc, 2015-09-17 12:47:53]

known platform fixes

[pbos-webrtc, 2015-09-17 12:57:58]

more fixes

[pbos-webrtc, 2015-09-17 13:03:47]

fix memory_pool_win

[pbos-webrtc, 2015-09-24 17:35:02]

explicit initialization

[noahric, 2015-09-24 18:01:55]

The CQ bit was checked by noahric@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2015-09-24 18:02:15]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1347793005/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1347793005/80001

[commit-bot: I haz the power, 2015-09-24 18:08:28]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-09-24 18:08:29]

Dry run: Try jobs failed on following builders:
  presubmit on tryserver.webrtc (JOB_FAILED,
http://build.chromium.org/p/tryserver.webrtc/builders/presubmit/builds/936)

[pbos-webrtc, 2015-09-25 09:26:12]

pbos@webrtc.org changed reviewers:
- noahric@chromium.org

[pbos-webrtc, 2015-09-25 14:50:25]

convert to volatile int and use AtomicOps

[pbos-webrtc, 2015-09-25 14:50:47]

The CQ bit was checked by pbos@webrtc.org to run a CQ dry run

[commit-bot: I haz the power, 2015-09-25 14:51:02]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1347793005/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1347793005/100001

[pbos-webrtc, 2015-09-25 14:51:48]

pbos@webrtc.org changed reviewers:
+ tommi@webrtc.org

[commit-bot: I haz the power, 2015-09-25 14:54:08]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-09-25 14:54:09]

Dry run: Try jobs failed on following builders:
  win_x64_rel on tryserver.webrtc (JOB_FAILED,
http://build.chromium.org/p/tryserver.webrtc/builders/win_x64_rel/builds/9618)

[pbos-webrtc, 2015-09-25 14:57:06]

PTAL, I suggest you review it in three steps (the way I converted it):

PS1-4 (Review PS4): Replace {+=,-=,++,--} operators for Atomic32, they are
invalid for use on volatile int (for atomicity). Instead use explicit
AtomicIncrement etc. calls (which we can replace with rtc::AtomicOps::Increment
etc, through scripting). This was done manually after making the operators
private/protected, so make sure the translation looks OK.
PS5: Remove default constructor, makes sure all volatile ints are initialized
once converting.
PS6: Replace .Value() with AcquireLoad, .CompareExchange with ::ReleaseStore or
::CompareAndSwap and .AtomicIncrement/.AtomicDecrement with ::Increment and
::Decrement.

[pbos-webrtc, 2015-09-25 14:57:46]

fix typo again

[pbos-webrtc, 2015-09-25 14:58:36]

The CQ bit was checked by pbos@webrtc.org

[pbos-webrtc, 2015-09-25 14:58:45]

The CQ bit was unchecked by pbos@webrtc.org

[pbos-webrtc, 2015-09-25 14:58:50]

The CQ bit was checked by pbos@webrtc.org to run a CQ dry run

[commit-bot: I haz the power, 2015-09-25 14:58:57]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1347793005/120001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1347793005/120001

[commit-bot: I haz the power, 2015-09-25 14:59:00]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-09-25 14:59:01]

No L-G-T-M from a valid reviewer yet. Only full committers are accepted.
Even if an L-G-T-M may have been provided, it was from a non-committer,
_not_ a full super star committer.
See http://www.chromium.org/getting-involved/become-a-committer
Note that this has nothing to do with OWNERS files.

[pbos-webrtc, 2015-09-25 14:59:36]

The CQ bit was checked by pbos@webrtc.org to run a CQ dry run

[commit-bot: I haz the power, 2015-09-25 14:59:53]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1347793005/120001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1347793005/120001

[commit-bot: I haz the power, 2015-09-25 17:00:09]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2015-09-25 17:00:11]

Dry run: Try jobs failed on following builders:
  android_arm64_rel on tryserver.webrtc (JOB_TIMED_OUT, no build URL)
  android_dbg on tryserver.webrtc (JOB_TIMED_OUT, no build URL)

[kwiberg-webrtc, 2015-11-01 03:05:55]

kwiberg@webrtc.org changed reviewers:
+ kwiberg@webrtc.org

[kwiberg-webrtc, 2015-11-01 03:05:55]

While I sympathize with wanting to reduce the number of types that do the same
thing, I must say that going from a dedicated Atomic32 type to volatile int
seems like a definite step backwards. It's going to be *very* easy to introduce
subtle bugs by accidentally treating an atomic volatile int as just an int.

If changing atomicops.h is an option, I'd suggest doing so first. However, it's
also probably worth first looking into how far away we are from being able to
use the atomics from the C++11 standard library.

(FWIW, I looked through the diff and found nothing else to complain about.)

https://codereview.webrtc.org/1347793005/diff/120001/webrtc/system_wrappers/i...
File webrtc/system_wrappers/interface/atomic32.h (left):

https://codereview.webrtc.org/1347793005/diff/120001/webrtc/system_wrappers/i...
webrtc/system_wrappers/interface/atomic32.h:66: #endif  //
WEBRTC_SYSTEM_WRAPPERS_INTERFACE_ATOMIC32_H_
You make this file empty; shouldn't it be deleted?

[pbos-webrtc, 2015-11-02 11:39:11]

Description was changed from

==========
Replace Atomic32 with webrtc/base/atomicops.h.

One atomic type is enough.

BUG=
R=tommi@webrtc.org
==========

to

==========
Replace Atomic32 with webrtc/base/atomicops.h.

One atomic type is enough.

BUG=
R=kwiberg@webrtc.org, tommi@webrtc.org
==========