As Chandler points out I've been trying to stabilize this test, but he made a good observation that I think will lead to a good solution. This particular part of the test that is failing, however, should be disabled in trunk right now.
On Sep 8, 2012, at 12:28 AM, Richard Smith <[email protected]> wrote: > On Sat, Sep 8, 2012 at 12:18 AM, Ted Kremenek <[email protected]> wrote: > Author: kremenek > Date: Sat Sep 8 02:18:18 2012 > New Revision: 163460 > > URL: http://llvm.org/viewvc/llvm-project?rev=163460&view=rev > Log: > Fix bug in BugReporter::RemoveUneededCalls() where "prunable" > PathDiagnosticEventPieces were *always* pruned. Instead, they > are suppose to only be pruned if the entire call gets pruned. > > Modified: > cfe/trunk/lib/StaticAnalyzer/Core/BugReporter.cpp > cfe/trunk/lib/StaticAnalyzer/Core/BugReporterVisitors.cpp > cfe/trunk/test/Analysis/malloc-plist.c > cfe/trunk/test/Analysis/plist-output-alternate.m > cfe/trunk/test/Analysis/plist-output.m > cfe/trunk/test/Analysis/retain-release.m > > Analysis/retain-release.m has been failing nondeterministically since r163455 > (maybe before). Can you take a look? At r163460, the failure looks like this: > > src/tools/clang/test/Analysis/retain-release.m:13645:16: error: CHECK-NEXT: > is not on the line after the previous match > // CHECK-NEXT: <key>kind</key><string>control</string> > ^ > build/tools/clang/test/Analysis/Output/retain-release.m.tmp.objcpp.plist:11772:2: > note: 'next' match was here > <key>kind</key><string>control</string> > ^ > build/tools/clang/test/Analysis/Output/retain-release.m.tmp.objcpp.plist:11728:8: > note: previous match was here > <dict> > ^ > > Thanks!
_______________________________________________ cfe-commits mailing list [email protected] http://lists.cs.uiuc.edu/mailman/listinfo/cfe-commits
