When running a simple WebKit-based tool that generates GObject introspection files (this is part of the build process), the program crashes with a SIGBUS in pageheap_memory. Mike Hommey from Debian has a patch to fix this issue.
Created attachment 54918 [details] patch from Debian Mike Hommey sent this patch to me. I took the liberty of doing the changelogging for him.
Attachment 54918 [details] did not pass style-queue: Failed to run "['WebKitTools/Scripts/check-webkit-style', '--no-squash']" exit_code: 1 JavaScriptCore/wtf/FastMalloc.cpp:2332: pageheap_memory is incorrectly named. Don't use underscores in your identifier names. [readability/naming] [4] Total errors found: 1 in 1 files If any of these errors are false positives, please file a bug against check-webkit-style.
Comment on attachment 54918 [details] patch from Debian This should use AllocAlignementInteger rather than uint64_t.
Created attachment 54938 [details] patch suggested by Darin Here it is!
Attachment 54938 [details] did not pass style-queue: Failed to run "['WebKitTools/Scripts/check-webkit-style', '--no-squash']" exit_code: 1 JavaScriptCore/wtf/FastMalloc.cpp:2332: pageheap_memory is incorrectly named. Don't use underscores in your identifier names. [readability/naming] [4] Total errors found: 1 in 1 files If any of these errors are false positives, please file a bug against check-webkit-style.
Comment on attachment 54938 [details] patch suggested by Darin Rejecting patch 54938 from commit-queue. Found no modified ChangeLogs, cannot create a commit message. All changes require a ChangeLog. See: http://webkit.org/coding/contributing.html
(In reply to comment #6) > (From update of attachment 54938 [details]) > Rejecting patch 54938 from commit-queue. > > Found no modified ChangeLogs, cannot create a commit message. > All changes require a ChangeLog. See: > http://webkit.org/coding/contributing.html This should work now if you cq+ it again. There was a regression in svn-apply that affected e-mailed diffs: https://bugs.webkit.org/show_bug.cgi?id=38507
Landed manually as r58753.
I was observing this problem on Solaris 10 x86_64 as well, and I can confirm r58753 fixes the problem on x86_64.
(In reply to comment #9) > I was observing this problem on Solaris 10 x86_64 as well, and I can confirm > r58753 fixes the problem on x86_64. Great news! =)