Reland: Introduce a V8_NORETURN macro and use it to make GCC 4.9.2 happy again.

Reland: Introduce a V8_NORETURN macro and use it to make GCC 4.9.2 happy again.

Without that, it has a few false positives about out-of-bounds array accesses.
Also makes the clang static-analyzer happy.

Original code review from Sven Panne:
https://codereview.chromium.org/790723002/

CQ_INCLUDE_TRYBOTS=tryserver.v8:v8_linux_arm_dbg,v8_linux_arm64_dbg,v8_mac64_dbg,v8_win_compile_dbg,v8_linux_gcc_rel

Committed: https://crrev.com/9a6c8b2455dd4121f77860b8c6de5f3ebb6ab9b0
Cr-Commit-Position: refs/heads/master@{#31185}

[skomski, 2015-10-08 11:48:16]

karl@skomski.com changed reviewers:
+ bmeurer@chromium.org, jochen@chromium.org

[skomski, 2015-10-08 11:48:16]

Hello again :(

PTAL 

Since switching the checks resolved the check fail on v8 arm release but made it
appear on v8 arm debug I now moved both checks to the bottom which works in both
modes (on my arm server with vanilla gcc 4.8).

[Benedikt Meurer, 2015-10-08 12:55:07]

lgtm

[jochen (gone - plz use gerrit), 2015-10-08 17:05:41]

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

[commit-bot: I haz the power, 2015-10-08 17:05:50]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1393023003/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1393023003/20001

[jochen (gone - plz use gerrit), 2015-10-08 17:24:05]

lgtm

[commit-bot: I haz the power, 2015-10-08 17:45:59]

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

[commit-bot: I haz the power, 2015-10-08 17:45:59]

Dry run: This issue passed the CQ dry run.

[skomski, 2015-10-08 18:51:00]

The CQ bit was checked by karl@skomski.com to run a CQ dry run

[commit-bot: I haz the power, 2015-10-08 18:51:10]

Dry run: CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1393023003/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1393023003/20001

[commit-bot: I haz the power, 2015-10-08 19:00:47]

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

[commit-bot: I haz the power, 2015-10-08 19:00:47]

Dry run: This issue passed the CQ dry run.

[skomski, 2015-10-08 19:01:15]

The CQ bit was checked by karl@skomski.com

[commit-bot: I haz the power, 2015-10-08 19:01:30]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1393023003/20001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1393023003/20001

[commit-bot: I haz the power, 2015-10-08 19:02:35]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-10-08 19:03:36]

Patchset 2 (id:??) landed as
https://crrev.com/9a6c8b2455dd4121f77860b8c6de5f3ebb6ab9b0
Cr-Commit-Position: refs/heads/master@{#31185}

[Reid Kleckner, 2015-10-09 03:42:43]

rnk@chromium.org changed reviewers:
+ rnk@chromium.org

[Reid Kleckner, 2015-10-09 03:42:43]

This broke Clang on Windows:
http://build.chromium.org/p/chromium.fyi/builders/ClangToTWin%28dbg%29/builds...

I guess ::raise is not marked noreturn in the MSVC headers, so clang warns that
the function might return.

[hans, 2015-10-09 16:32:02]

On 2015/10/09 03:42:43, Reid Kleckner wrote:
> This broke Clang on Windows:
>
http://build.chromium.org/p/chromium.fyi/builders/ClangToTWin%28dbg%29/builds...
> 
> I guess ::raise is not marked noreturn in the MSVC headers, so clang warns
that
> the function might return.

I wonder why it's not showing on our release bots. It does repro locally for me
in a release build.

Anyway, I suppose we can just put a FATAL() or something after the raise call.

[Nico, 2015-10-09 16:36:54]

On 2015/10/09 16:32:02, hans wrote:
> On 2015/10/09 03:42:43, Reid Kleckner wrote:
> > This broke Clang on Windows:
> >
>
http://build.chromium.org/p/chromium.fyi/builders/ClangToTWin%28dbg%29/builds...
> > 
> > I guess ::raise is not marked noreturn in the MSVC headers, so clang warns
> that
> > the function might return.
> 
> I wonder why it's not showing on our release bots. It does repro locally for
me
> in a release build.
> 
> Anyway, I suppose we can just put a FATAL() or something after the raise call.

Yes, something like https://codereview.chromium.org/1213623017 (that's in nacl,
but we could do the same in v8)

[skomski, 2015-10-09 16:41:25]

On 2015/10/09 16:36:54, Nico (offline until Fri Oct 9) wrote:
> On 2015/10/09 16:32:02, hans wrote:
> > On 2015/10/09 03:42:43, Reid Kleckner wrote:
> > > This broke Clang on Windows:
> > >
> >
>
http://build.chromium.org/p/chromium.fyi/builders/ClangToTWin%28dbg%29/builds...
> > > 
> > > I guess ::raise is not marked noreturn in the MSVC headers, so clang warns
> > that
> > > the function might return.
> > 
> > I wonder why it's not showing on our release bots. It does repro locally for
> me
> > in a release build.
> > 
> > Anyway, I suppose we can just put a FATAL() or something after the raise
call.
> 
> Yes, something like https://codereview.chromium.org/1213623017 (that's in
nacl,
> but we could do the same in v8)

Already fixed it in https://codereview.chromium.org/1390193003/.