[PATCH] xfail tests that failed with clang-3.7, gcc4.8.2 and on i386 to get buildbot green

Ying Chen chying at google.com
Tue Mar 24 16:28:12 PDT 2015


================
Comment at: test/functionalities/data-formatter/data-formatter-stl/libstdcpp/vbool/TestDataFormatterStdVBool.py:27
@@ -26,2 +26,3 @@
     @skipIfDarwin
+    @expectedFailurei386 #xfail to get buildbot green, test fails on ubuntu x86_64
     def test_with_dwarf_and_run_command(self):
----------------
vharron wrote:
> sivachandra wrote:
> > This and similar comments attached to expectedFailurei386 below are confusing. Perhaps s/x86_64/i386 ? Even then, seems redundant.
> > 
> > Another note: Tamas has a fix for this approved. He might land it any time soon. So, may be observe the bot after his fix lands and take out the xfail decorator if the test starts passing?
> Suggested replacement comment
> 
> 
> ```
> # failing config: i386 binary running on lldb built for ubuntu 14.04 x86_64
> ```
> 
The test was compiled for i386, but failed when running on x86_64 ubuntu system. It may or may not fail on actual i386 system, i didn't run this test on i386 system.


================
Comment at: test/lang/cpp/this/TestCPPThis.py:25
@@ -24,2 +24,3 @@
     @dwarf_test
+    @expectedFailureClang(bugnumber='19893563', compiler_version=['>=','3.6'])#xfail to get buildbot green, test failed with totclang - clang3.7
     def test_with_dwarf_and_run_command(self):
----------------
vharron wrote:
> sivachandra wrote:
> > Why do we have a bug number only for this? Also, seems like the right tracking bug number is 19893491.
> > 
> > Unrelated: If we should be putting internal bug numbers at all, could we just put http://b/<bugnumber> similar to how apple puts links to rdar bugs?
> Yeah, let's put llvm bugzilla urls here only, try to keep it open.
I agree that we do not include internal bug numbers. I put bug number for this one, only because we need the second parameter. And currently expectedFailureClang doesn't work if the first argument is skipped.
I have another patch to fix this problem.
http://reviews.llvm.org/D8595
Once that one is landed, I will remove bugnumber from here.

http://reviews.llvm.org/D8573

EMAIL PREFERENCES
  http://reviews.llvm.org/settings/panel/emailpreferences/






More information about the llvm-commits mailing list