[LLVMbugs] [Bug 16250] New: Builder llvm-mips-linux "make check" LLVM-Unit::MCJITTest.return_global failure

[bugzilla-daemon at llvm.org]

http://llvm.org/bugs/show_bug.cgi?id=16250

            Bug ID: 16250
           Summary: Builder llvm-mips-linux "make check"
                    LLVM-Unit::MCJITTest.return_global failure
           Product: libraries
           Version: trunk
          Hardware: PC
                OS: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: MC
          Assignee: unassignedbugs at nondot.org
          Reporter: dgilmore at mips.com
                CC: llvmbugs at cs.uiuc.edu
    Classification: Unclassified

Recent changes broke LLVM "make check" for Builder llvm-mips-linux.

The following tests files:

LLVM-Unit::MCJITTest.return_global
LLVM-Unit::MCJITTest.multiple_functions

One issue that slowed down the analysis of the issue was reproducing the
problem in isolation, Note that running just ExecutionEngine portion of
"make check" did not report the problem:

$ /usr/bin/python /home/dgilmore/llvm-pp2/llvm/utils/lit/lit.py -s -v
ExecutionEngine
-- Testing: 115 tests, 6 threads --
Testing: 0 .. 10.. 20.. 30.. 40.. 50.. 60.. 70.. 80.. 90..
Testing Time: 2.56s
  Expected Passes    : 107
  Expected Failures  : 8
$

It wasn't apparent how the test was run.

Adding the following patch:


diff --git a/utils/lit/lit/TestRunner.py b/utils/lit/lit/TestRunner.py
index 8417699..e9b15f8 100644
--- a/utils/lit/lit/TestRunner.py
+++ b/utils/lit/lit/TestRunner.py
@@ -26,6 +26,7 @@ kAvoidDevNull = kIsWindows
 def executeCommand(command, cwd=None, env=None):
     # Close extra file handles on UNIX (on Windows this cannot be done while
     # also redirecting input).
+    print command
     close_fds = not kIsWindows

     p = subprocess.Popen(command, cwd=cwd,

I was able to determine the failing commands:

$
/home/dgilmore/llvm-pp2/bcbug969/unittests/ExecutionEngine/MCJIT/Release+Asserts/MCJITTests
--gtest_filter=MCJITTest.return_global
Note: Google Test filter = MCJITTest.return_global
[==========] Running 1 test from 1 test case.
[----------] Global test environment set-up.
[----------] 1 test from MCJITTest
[ RUN      ] MCJITTest.return_global
'generic' is not a recognized processor for this target (ignoring processor)
'generic' is not a recognized processor for this target (ignoring processor)
'generic' is not a recognized processor for this target (ignoring processor)
'generic' is not a recognized processor for this target (ignoring processor)
'generic' is not a recognized processor for this target (ignoring processor)
0 Segmentation fault
$
/home/dgilmore/llvm-pp2/bcbug969/unittests/ExecutionEngine/MCJIT/Release+Asserts/MCJITTests
--gtest_filter=MCJITTest.multiple_functions
Note: Google Test filter = MCJITTest.multiple_functions
[==========] Running 1 test from 1 test case.
[----------] Global test environment set-up.
[----------] 1 test from MCJITTest
[ RUN      ] MCJITTest.multiple_functions
'generic' is not a recognized processor for this target (ignoring processor)
'generic' is not a recognized processor for this target (ignoring processor)
'generic' is not a recognized processor for this target (ignoring processor)
'generic' is not a recognized processor for this target (ignoring processor)
'generic' is not a recognized processor for this target (ignoring processor)
0 Segmentation fault
$

We are tracking this this down -- last status blurb on this was:

LLVM/MCJIT failures:
- Issue with multiple relocations - relocations are not correctly re-applied;
-- debugging in progress;

-- 
You are receiving this mail because:
You are on the CC list for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-bugs/attachments/20130606/8394623e/attachment.html>