Bug 119266 - Add back c++11 features removed by buildfixes after all ports did the switch
Summary: Add back c++11 features removed by buildfixes after all ports did the switch
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: JavaScriptCore (show other bugs)
Version: 528+ (Nightly build)
Hardware: All All
: P2 Normal
Assignee: Csaba Osztrogonác
URL:
Keywords:
Depends on: 119402
Blocks:
  Show dependency treegraph
 
Reported: 2013-07-30 09:21 PDT by Csaba Osztrogonác
Modified: 2013-09-12 08:27 PDT (History)
16 users (show)

See Also:


Attachments
Patch (12.04 KB, patch)
2013-07-30 10:35 PDT, Csaba Osztrogonác
no flags Details | Formatted Diff | Diff
Patch (12.71 KB, patch)
2013-07-30 10:42 PDT, Csaba Osztrogonác
no flags Details | Formatted Diff | Diff
Patch (12.71 KB, patch)
2013-07-30 11:12 PDT, Csaba Osztrogonác
no flags Details | Formatted Diff | Diff
Patch (14.53 KB, patch)
2013-08-01 10:31 PDT, Csaba Osztrogonác
no flags Details | Formatted Diff | Diff
Patch (6.41 KB, patch)
2013-09-12 04:58 PDT, Csaba Osztrogonác
no flags Details | Formatted Diff | Diff
Patch (5.36 KB, patch)
2013-09-12 05:11 PDT, Csaba Osztrogonác
no flags Details | Formatted Diff | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Csaba Osztrogonác 2013-07-30 09:21:48 PDT
Patch is coming soon.
Comment 1 Csaba Osztrogonác 2013-07-30 10:35:06 PDT
Created attachment 207746 [details]
Patch

WIP patch for EWS bots to check if all ports are ready for c++11 in JSC
Comment 2 Early Warning System Bot 2013-07-30 10:40:05 PDT
Comment on attachment 207746 [details]
Patch

Attachment 207746 [details] did not pass qt-ews (qt):
Output: http://webkit-queues.appspot.com/results/1290444
Comment 3 Early Warning System Bot 2013-07-30 10:40:42 PDT
Comment on attachment 207746 [details]
Patch

Attachment 207746 [details] did not pass qt-wk2-ews (qt-wk2):
Output: http://webkit-queues.appspot.com/results/1295414
Comment 4 EFL EWS Bot 2013-07-30 10:42:08 PDT
Comment on attachment 207746 [details]
Patch

Attachment 207746 [details] did not pass efl-wk2-ews (efl-wk2):
Output: http://webkit-queues.appspot.com/results/1293385
Comment 5 Csaba Osztrogonác 2013-07-30 10:42:42 PDT
Created attachment 207747 [details]
Patch

WIP patch for EWS bots, let's try to enable c++11 for whole QtWebKit
Comment 6 EFL EWS Bot 2013-07-30 10:55:41 PDT
Comment on attachment 207747 [details]
Patch

Attachment 207747 [details] did not pass efl-wk2-ews (efl-wk2):
Output: http://webkit-queues.appspot.com/results/1295420
Comment 7 Csaba Osztrogonác 2013-07-30 11:12:50 PDT
Created attachment 207750 [details]
Patch

one more round after killing the Qt EWS bots :) (clean build worked for me locally)
Comment 8 EFL EWS Bot 2013-07-30 11:20:44 PDT
Comment on attachment 207750 [details]
Patch

Attachment 207750 [details] did not pass efl-wk2-ews (efl-wk2):
Output: http://webkit-queues.appspot.com/results/1298425
Comment 9 Allan Sandfeld Jensen 2013-07-30 16:06:06 PDT
(In reply to comment #7)
> Created an attachment (id=207750) [details]
> Patch
> 
> one more round after killing the Qt EWS bots :) (clean build worked for me locally)

Did anyone actually enable C++11 on Qt?
Comment 10 Csaba Osztrogonác 2013-07-30 23:56:31 PDT
(In reply to comment #9)
> (In reply to comment #7)
> > Created an attachment (id=207750) [details] [details]
> > Patch
> > 
> > one more round after killing the Qt EWS bots :) (clean build worked for me locally)
> 
> Did anyone actually enable C++11 on Qt?

No, I tried to enable it by the uploaded patch. But unfortunately
the EWS bot couldn't handle this situation and needed a clean build.
Comment 11 Csaba Osztrogonác 2013-08-01 10:31:26 PDT
Created attachment 207935 [details]
Patch

updated patch for EWS bots, Qt fix removed (Qt switched to c++11), cmake fixes added to make EFL bots happy
Comment 12 Build Bot 2013-08-11 03:30:08 PDT
Comment on attachment 207935 [details]
Patch

Attachment 207935 [details] did not pass win-ews (win):
Output: http://webkit-queues.appspot.com/results/1359999
Comment 13 Allan Sandfeld Jensen 2013-08-13 01:47:53 PDT
(In reply to comment #11)
> Created an attachment (id=207935) [details]
> Patch
> 
> updated patch for EWS bots, Qt fix removed (Qt switched to c++11), cmake fixes added to make EFL bots happy

Actually I would prefer if we didn't add these back.

We might have enabled C++11, but using nullptr limits WebKit from GCC 4.3+ to GCC 4.6+, and considering how little using nullptr brings, I would prefer we avoid them.
Comment 14 Allan Sandfeld Jensen 2013-08-13 02:29:08 PDT
(In reply to comment #13)
> (In reply to comment #11)
> > Created an attachment (id=207935) [details] [details]
> > Patch
> > 
> > updated patch for EWS bots, Qt fix removed (Qt switched to c++11), cmake fixes added to make EFL bots happy
> 
> Actually I would prefer if we didn't add these back.
> 
> We might have enabled C++11, but using nullptr limits WebKit from GCC 4.3+ to GCC 4.6+, and considering how little using nullptr brings, I would prefer we avoid them.

Btw, the reason support older GCC still makes sense is mostly Red Hat. Since Debian stable updated earlier this year, it is now RHEL6 that restraining gcc on linux. The latest Red Hat Enterprise Linux ships with GCC 4.4.
Comment 15 Csaba Osztrogonác 2013-08-20 03:33:56 PDT
(In reply to comment #13)
> (In reply to comment #11)
> > Created an attachment (id=207935) [details] [details]
> > Patch
> > 
> > updated patch for EWS bots, Qt fix removed (Qt switched to c++11), cmake fixes added to make EFL bots happy
> 
> Actually I would prefer if we didn't add these back.
> 
> We might have enabled C++11, but using nullptr limits WebKit from GCC 4.3+ to GCC 4.6+, and considering how little using nullptr brings, I would prefer we avoid them.

It's all the same for me, but it would be great to reply to the original
webkit-dev message about it. Otherwise I don't think if there is any
buildbot or EWS with GCC 4.4 to ensure nobody adds nullptr again and again.
Comment 16 Csaba Osztrogonác 2013-09-12 04:58:48 PDT
Created attachment 211419 [details]
Patch

updated patch to ToT, Qt and EFL buildsystem hacks aren't needed, both of them builds with c++11
Comment 17 Build Bot 2013-09-12 05:02:11 PDT
Comment on attachment 211419 [details]
Patch

Attachment 211419 [details] did not pass mac-ews (mac):
Output: http://webkit-queues.appspot.com/results/1795131
Comment 18 EFL EWS Bot 2013-09-12 05:02:13 PDT
Comment on attachment 211419 [details]
Patch

Attachment 211419 [details] did not pass efl-ews (efl):
Output: http://webkit-queues.appspot.com/results/1843106
Comment 19 Early Warning System Bot 2013-09-12 05:02:52 PDT
Comment on attachment 211419 [details]
Patch

Attachment 211419 [details] did not pass qt-ews (qt):
Output: http://webkit-queues.appspot.com/results/1857085
Comment 20 EFL EWS Bot 2013-09-12 05:04:05 PDT
Comment on attachment 211419 [details]
Patch

Attachment 211419 [details] did not pass efl-wk2-ews (efl-wk2):
Output: http://webkit-queues.appspot.com/results/1843107
Comment 21 Csaba Osztrogonác 2013-09-12 05:04:11 PDT
I grepped to nullptr and it seems there are zillion everywhere:
-   8 nullptr in JavaScriptCore
-  71 nullptr in WTF
- 383 nullptr in WebCore
-  99 nullptr in WebKit2

So I don't see why shouldn't we add them back.
Comment 22 Early Warning System Bot 2013-09-12 05:04:21 PDT
Comment on attachment 211419 [details]
Patch

Attachment 211419 [details] did not pass qt-wk2-ews (qt-wk2):
Output: http://webkit-queues.appspot.com/results/1818100
Comment 23 Csaba Osztrogonác 2013-09-12 05:11:04 PDT
Created attachment 211421 [details]
Patch
Comment 24 Zan Dobersek 2013-09-12 08:09:04 PDT
If you're taking requests here, feel free to also include the std::nullptr reintroduction in ChildProcessProxy in bug #109162.
Comment 25 Csaba Osztrogonác 2013-09-12 08:27:17 PDT
Comment on attachment 211421 [details]
Patch

Clearing flags on attachment: 211421

Committed r155613: <http://trac.webkit.org/changeset/155613>
Comment 26 Csaba Osztrogonác 2013-09-12 08:27:25 PDT
All reviewed patches have been landed.  Closing bug.