When building QtWebKit on QNX, we get plenty of compilation failures, most of them related to pthreads in JSC. Even after fixing these compilation failures, we still have a broken webkit in the end because JIT is not working, so it'll have to be disabled. Problem reported by Ritt Konstantin <ritt.ks@gmail.com>. I'll be uploading a patch based on his changes.
Created attachment 99300 [details] patch
Attachment 99300 [details] did not pass style-queue: Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'Source/JavaScriptCore/ChangeLog', u'Source..." exit_code: 1 Source/JavaScriptCore/wtf/StackBounds.cpp:44: "pthread.h" already included at Source/JavaScriptCore/wtf/StackBounds.cpp:28 [build/include] [4] Total errors found: 1 in 5 files If any of these errors are false positives, please file a bug against check-webkit-style.
(In reply to comment #2) > Source/JavaScriptCore/wtf/StackBounds.cpp:44: "pthread.h" already included at Source/JavaScriptCore/wtf/StackBounds.cpp:28 [build/include] [4] > Total errors found: 1 in 5 files This looks like a false positive, as the includes are guarded with #if OS().
Comment on attachment 99300 [details] patch View in context: https://bugs.webkit.org/attachment.cgi?id=99300&action=review > Source/JavaScriptCore/heap/MachineStackMarker.cpp:83 > +#if !OS(QNX) > +/* QNX doesn't support SA_RESTART yet. */ Are you sure you want ENABLE(JSC_MULTIPLE_THREADS)? Why not disable JSC_MULTIPLE_THREADS if you can't support it correctly? > Source/JavaScriptCore/wtf/Platform.h:663 > +#if OS(QNX) > +#define ENABLE_JIT 0 > +#endif Why do you put it in it's own #if? IMHO http://trac.webkit.org/browser/trunk/Source/JavaScriptCore/wtf/Platform.h?rev=90167#L1027 is the correct place for this. > Source/JavaScriptCore/wtf/StackBounds.cpp:49 > +#include <pthread.h> > #include <fcntl.h> > #include <sys/procfs.h> > #include <stdio.h> > #include <errno.h> > +#include <string.h> I agree with the false positive, but please sort the headers.
(In reply to comment #4) > (From update of attachment 99300 [details]) > View in context: https://bugs.webkit.org/attachment.cgi?id=99300&action=review > Thanks for the review, the patch is much more clean now. :-) > > Source/JavaScriptCore/heap/MachineStackMarker.cpp:83 > > +#if !OS(QNX) > > +/* QNX doesn't support SA_RESTART yet. */ > > Are you sure you want ENABLE(JSC_MULTIPLE_THREADS)? > Why not disable JSC_MULTIPLE_THREADS if you can't support it correctly? Perfect, done. > > > Source/JavaScriptCore/wtf/Platform.h:663 > > +#if OS(QNX) > > +#define ENABLE_JIT 0 > > +#endif > > Why do you put it in it's own #if? > IMHO http://trac.webkit.org/browser/trunk/Source/JavaScriptCore/wtf/Platform.h?rev=90167#L1027 is the correct place for this. Done. > > > Source/JavaScriptCore/wtf/StackBounds.cpp:49 > > +#include <pthread.h> > > #include <fcntl.h> > > #include <sys/procfs.h> > > #include <stdio.h> > > #include <errno.h> > > +#include <string.h> > > I agree with the false positive, but please sort the headers. Done.
Created attachment 100552 [details] patch v2
Attachment 100552 [details] did not pass style-queue: Failed to run "['Tools/Scripts/check-webkit-style', '--diff-files', u'Source/JavaScriptCore/ChangeLog', u'Source..." exit_code: 1 Source/JavaScriptCore/wtf/StackBounds.cpp:46: "pthread.h" already included at Source/JavaScriptCore/wtf/StackBounds.cpp:28 [build/include] [4] Total errors found: 1 in 4 files If any of these errors are false positives, please file a bug against check-webkit-style.
Comment on attachment 100552 [details] patch v2 LGTM, but I'm not a reviewer
Comment on attachment 100552 [details] patch v2 Cool, r=me
Comment on attachment 100552 [details] patch v2 Clearing flags on attachment: 100552 Committed r90916: <http://trac.webkit.org/changeset/90916>
All reviewed patches have been landed. Closing bug.