RESOLVED FIXED 156892
[JSC] Fix build break since r199866
https://bugs.webkit.org/show_bug.cgi?id=156892
Summary [JSC] Fix build break since r199866
Joonghun Park
Reported 2016-04-21 22:50:45 PDT
Add using namespace std statement.
Attachments
To check EWS build status (1.99 KB, patch)
2016-04-22 04:02 PDT, Joonghun Park
no flags
Fix a typo ignoring style check warning. cloop build was successful (2.03 KB, patch)
2016-04-22 04:14 PDT, Joonghun Park
no flags
Change commit log to let this patch be reviewed. (2.00 KB, patch)
2016-04-22 04:30 PDT, Joonghun Park
no flags
Follow up to Bejamin's quick fix (1.09 KB, patch)
2016-04-22 15:13 PDT, Joonghun Park
no flags
Joonghun Park
Comment 1 2016-04-21 22:56:21 PDT
Csaba Osztrogonác
Comment 2 2016-04-22 02:05:16 PDT
(In reply to comment #1) > Committed r199868: <http://trac.webkit.org/changeset/199868> And Darin landed one more fix: (I don't konw which build it fixed.) http://trac.webkit.org/changeset/199870 But now the Apple Mac cloop build is broken: /Volumes/Data/slave/elcapitan-cloop-debug/build/Source/JavaScriptCore/runtime/MathCommon.h:61:28: error: expected unqualified-id if (!constant || !std::isnormal(constant)) ^ /Volumes/Data/slave/elcapitan-cloop-debug/build/Source/JavaScriptCore/runtime/MathCommon.h:65:9: error: no member named 'frexp' in namespace 'std'; did you mean simply 'frexp'? if (std::frexp(constant, &exponent) != 0.5) ^~~~~~~~~~ frexp /Volumes/Data/slave/elcapitan-cloop-debug/build/Source/JavaScriptCore/runtime/MathCommon.h:77:25: error: no member named 'ldexp' in namespace 'std'; did you mean simply 'ldexp'? double reciprocal = std::ldexp(1, -exponent); ^~~~~~~~~~ ldexp /Volumes/Data/slave/elcapitan-cloop-debug/build/Source/JavaScriptCore/runtime/MathCommon.h:78:17: error: expected unqualified-id ASSERT(std::isnormal(reciprocal)); ^ Common, nobody watches bots after landing random patches? Why ???? The original patch broke all Linux builds, the buildfixes broke the Apple Mac cloop build.
Joonghun Park
Comment 3 2016-04-22 02:35:26 PDT
(In reply to comment #2) > (In reply to comment #1) > > Committed r199868: <http://trac.webkit.org/changeset/199868> > > And Darin landed one more fix: (I don't konw which build it fixed.) > http://trac.webkit.org/changeset/199870 > > But now the Apple Mac cloop build is broken: > > /Volumes/Data/slave/elcapitan-cloop-debug/build/Source/JavaScriptCore/ > runtime/MathCommon.h:61:28: error: expected unqualified-id > if (!constant || !std::isnormal(constant)) > ^ > /Volumes/Data/slave/elcapitan-cloop-debug/build/Source/JavaScriptCore/ > runtime/MathCommon.h:65:9: error: no member named 'frexp' in namespace > 'std'; did you mean simply 'frexp'? > if (std::frexp(constant, &exponent) != 0.5) > ^~~~~~~~~~ > frexp > > /Volumes/Data/slave/elcapitan-cloop-debug/build/Source/JavaScriptCore/ > runtime/MathCommon.h:77:25: error: no member named 'ldexp' in namespace > 'std'; did you mean simply 'ldexp'? > double reciprocal = std::ldexp(1, -exponent); > ^~~~~~~~~~ > ldexp > > /Volumes/Data/slave/elcapitan-cloop-debug/build/Source/JavaScriptCore/ > runtime/MathCommon.h:78:17: error: expected unqualified-id > ASSERT(std::isnormal(reciprocal)); > ^ > > Common, nobody watches bots after landing random patches? Why ???? > The original patch broke all Linux builds, the buildfixes broke the > Apple Mac cloop build. I'm sorry, Ossy. If it is not fixed soon, let me check it out again. My mac is at home, so I couldn't check mac cloop port in advance.
Joonghun Park
Comment 4 2016-04-22 04:02:39 PDT
Reopening to attach new patch.
Joonghun Park
Comment 5 2016-04-22 04:02:43 PDT
Created attachment 277039 [details] To check EWS build status
Csaba Osztrogonác
Comment 6 2016-04-22 04:11:41 PDT
(In reply to comment #5) > Created attachment 277039 [details] > To check EWS build status FYI: there is no EWS for cloop build.
Joonghun Park
Comment 7 2016-04-22 04:14:36 PDT
Created attachment 277040 [details] Fix a typo ignoring style check warning. cloop build was successful
WebKit Commit Bot
Comment 8 2016-04-22 04:16:37 PDT
Attachment 277040 [details] did not pass style-queue: ERROR: Source/JavaScriptCore/runtime/MathCommon.h:36: Do not use 'using namespace std;'. [build/using_namespace] [4] Total errors found: 1 in 2 files If any of these errors are false positives, please file a bug against check-webkit-style.
Joonghun Park
Comment 9 2016-04-22 04:30:00 PDT
Created attachment 277041 [details] Change commit log to let this patch be reviewed.
WebKit Commit Bot
Comment 10 2016-04-22 04:31:20 PDT
Attachment 277041 [details] did not pass style-queue: ERROR: Source/JavaScriptCore/runtime/MathCommon.h:36: Do not use 'using namespace std;'. [build/using_namespace] [4] Total errors found: 1 in 2 files If any of these errors are false positives, please file a bug against check-webkit-style.
Mark Lam
Comment 11 2016-04-22 09:49:33 PDT
Comment on attachment 277041 [details] Change commit log to let this patch be reviewed. r=me
Darin Adler
Comment 12 2016-04-22 09:50:28 PDT
Comment on attachment 277041 [details] Change commit log to let this patch be reviewed. View in context: https://bugs.webkit.org/attachment.cgi?id=277041&action=review > Source/JavaScriptCore/runtime/MathCommon.h:36 > +using namespace std; I think the correct fix is to add an include so we get the "std::" versions of these rather than doing this.
Mark Lam
Comment 13 2016-04-22 09:51:54 PDT
Comment on attachment 277041 [details] Change commit log to let this patch be reviewed. r- per Darin's comment.
Benjamin Poulain
Comment 14 2016-04-22 15:02:18 PDT
Quick fix: http://trac.webkit.org/changeset/199913 Let's see if the bots like that better.
Joonghun Park
Comment 15 2016-04-22 15:13:42 PDT
Created attachment 277105 [details] Follow up to Bejamin's quick fix
WebKit Commit Bot
Comment 16 2016-04-22 18:11:22 PDT
Comment on attachment 277105 [details] Follow up to Bejamin's quick fix Clearing flags on attachment: 277105 Committed r199943: <http://trac.webkit.org/changeset/199943>
WebKit Commit Bot
Comment 17 2016-04-22 18:11:25 PDT
All reviewed patches have been landed. Closing bug.
Note You need to log in before you can comment on or make changes to this bug.