Make WARN_UNUSED_RESULT a no-op on gcc

Make WARN_UNUSED_RESULT a no-op on gcc

Because on gcc, cast to void doesn't silence the warning. See
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66425

Also add an RTC_ prefix to the macro instead of only defining it if it
wasn't already defined, to ensure that we always get our own version.

BUG=none

Review-Url: https://codereview.webrtc.org/2797983003
Cr-Commit-Position: refs/heads/master@{#17563}
Committed: https://chromium.googlesource.com/external/webrtc/+/388fe425c783b1994608789842503ec756327913

[kwiberg-webrtc, 2017-04-05 13:12:53]

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

[commit-bot: I haz the power, 2017-04-05 13:12:57]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...

[kwiberg-webrtc, 2017-04-05 13:14:24]

kwiberg@webrtc.org changed reviewers:
+ ossu@webrtc.org, peah@webrtc.org

[peah-webrtc, 2017-04-05 13:29:15]

On 2017/04/05 13:14:25, kwiberg-webrtc wrote:

Nice!!! lgtm, but I guess there should be a bug-number!?

[commit-bot: I haz the power, 2017-04-05 13:36:36]

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

[commit-bot: I haz the power, 2017-04-05 13:36:36]

Dry run: Try jobs failed on following builders:
  android_compile_x86_dbg on master.tryserver.webrtc (JOB_FAILED,
http://build.chromium.org/p/tryserver.webrtc/builders/android_compile_x86_dbg...)

[ossu, 2017-04-05 13:44:08]

The robots! They're so angry!

Seems to clash with an already defined WARN_UNUSED_RESULT on android. Also, why
do we want to silence this? Where is it used?

[kwiberg-webrtc, 2017-04-05 23:39:14]

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

[commit-bot: I haz the power, 2017-04-05 23:39:22]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...

[kwiberg-webrtc, 2017-04-05 23:40:33]

Description was changed from

==========
Make WARN_UNUSED_RESULT a no-op on gcc

Because on gcc, cast to void doesn't silence the warning. See
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66425

BUG=none
==========

to

==========
Make WARN_UNUSED_RESULT a no-op on gcc

Because on gcc, cast to void doesn't silence the warning. See
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66425

Also add an RTC_ prefix to the macro instead of only defining it if it
wasn't already defined, to ensure that we always get our own version.

BUG=none
==========

[kwiberg-webrtc, 2017-04-05 23:41:56]

On 2017/04/05 13:44:08, ossu wrote:
> The robots! They're so angry!
> 
> Seems to clash with an already defined WARN_UNUSED_RESULT on android. Also,
why
> do we want to silence this? Where is it used?

peah@ wanted to discard the return value of one of the SwapQueue methods.

[kwiberg-webrtc, 2017-04-05 23:42:55]

On 2017/04/05 13:29:15, peah-webrtc wrote:
> On 2017/04/05 13:14:25, kwiberg-webrtc wrote:
> 
> Nice!!! lgtm, but I guess there should be a bug-number!?

I figured this was tiny enough that a bug wouldn't be necessary. Feel free to
file one if you like...

[commit-bot: I haz the power, 2017-04-06 00:17:01]

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

[commit-bot: I haz the power, 2017-04-06 00:17:02]

Dry run: This issue passed the CQ dry run.

[kwiberg-webrtc, 2017-04-06 00:29:16]

New patch set posted.

[peah-webrtc, 2017-04-06 06:30:08]

On 2017/04/06 00:29:16, kwiberg-webrtc wrote:
> New patch set posted.

Even better. Still lgtm

[ossu, 2017-04-06 11:21:15]

On 2017/04/05 23:41:56, kwiberg-webrtc wrote:
> On 2017/04/05 13:44:08, ossu wrote:
> > The robots! They're so angry!
> > 
> > Seems to clash with an already defined WARN_UNUSED_RESULT on android. Also,
> why
> > do we want to silence this? Where is it used?
> 
> peah@ wanted to discard the return value of one of the SwapQueue methods.

Although I'm quite wary of that usage, looking at the two methods marked as
WARN_UNUSED_RESULT (Insert and Remove), I do think it makes sense that we have
such a marker that isn't more restrictive on gcc than on clang. Especially with
an eye towards getting eventually getting a [[nodiscard]] that behaves
homogeneously.

lgtm

peah@ how did you want to use it? Seems a bit dangerous to ignore the results of
either Insert or Remove for SwapQueue. (We can talk offline)

[kwiberg-webrtc, 2017-04-06 11:26:58]

The CQ bit was checked by kwiberg@webrtc.org

[commit-bot: I haz the power, 2017-04-06 11:27:04]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.webrtc.org/...

[commit-bot: I haz the power, 2017-04-06 11:32:28]

CQ is committing da patch.

Bot data: {"patchset_id": 20001, "attempt_start_ts": 1491478018379340,
"parent_rev": "177b17e050965c4150d1bd0d9602561b704eca1b", "commit_rev":
"388fe425c783b1994608789842503ec756327913"}

[commit-bot: I haz the power, 2017-04-06 11:32:31]

Description was changed from

==========
Make WARN_UNUSED_RESULT a no-op on gcc

Because on gcc, cast to void doesn't silence the warning. See
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66425

Also add an RTC_ prefix to the macro instead of only defining it if it
wasn't already defined, to ensure that we always get our own version.

BUG=none
==========

to

==========
Make WARN_UNUSED_RESULT a no-op on gcc

Because on gcc, cast to void doesn't silence the warning. See
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66425

Also add an RTC_ prefix to the macro instead of only defining it if it
wasn't already defined, to ensure that we always get our own version.

BUG=none

Review-Url: https://codereview.webrtc.org/2797983003
Cr-Commit-Position: refs/heads/master@{#17563}
Committed:
https://chromium.googlesource.com/external/webrtc/+/388fe425c783b199460878984...
==========

[commit-bot: I haz the power, 2017-04-06 11:32:32]

Committed patchset #2 (id:20001) as
https://chromium.googlesource.com/external/webrtc/+/388fe425c783b199460878984...