[LLVMbugs] [Bug 19028] New: [powerpc-darwin] Impossible reg-to-reg copy crashing tests

[bugzilla-daemon at llvm.org]

http://llvm.org/bugs/show_bug.cgi?id=19028

            Bug ID: 19028
           Summary: [powerpc-darwin] Impossible reg-to-reg copy crashing
                    tests
           Product: libraries
           Version: trunk
          Hardware: Macintosh
                OS: MacOS X
            Status: NEW
          Severity: normal
          Priority: P
         Component: Backend: PowerPC
          Assignee: unassignedbugs at nondot.org
          Reporter: fang at csl.cornell.edu
                CC: hfinkel at anl.gov, iains-llvm at btconnect.com,
                    llvmbugs at cs.uiuc.edu
    Classification: Unclassified

On powerpc-darwin8, I'm seeing new regressions between r202302 and r202541:


FAIL: LLVM :: CodeGen/Generic/bool-to-double.ll (1410 of 8907)
******************** TEST 'LLVM :: CodeGen/Generic/bool-to-double.ll' FAILED
***
*****************
Script:
--
gtimeout 1m  /Users/fang/local/src/LLVM-svn/gcc40-cmake-build/./bin/llc <
/Volumes/Isolde/sources/LLVM-svn/llvm.git/test/CodeGen/Generic/bool-to-double.ll
--
Exit Code: 134

Command Output (stderr):
--
/Users/fang/local/src/LLVM-svn/llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp:247
2: failed assertion `NewInTy.isInteger() && "Ran out of possibilities!"'
Stack dump:
0.      Program arguments:
/Users/fang/local/src/LLVM-svn/gcc40-cmake-build/./bin/llc 
1.      Running pass 'Function Pass Manager' on module '<stdin>'.
2.      Running pass 'PowerPC DAG->DAG Pattern Instruction Selection' on
function '@test'

--

********************
FAIL: LLVM :: CodeGen/Generic/print-arith-fp.ll (1454 of 8907)
******************** TEST 'LLVM :: CodeGen/Generic/print-arith-fp.ll' FAILED
***
*****************
Script:
--
gtimeout 1m  /Users/fang/local/src/LLVM-svn/gcc40-cmake-build/./bin/llc <
/Volumes/Isolde/sources/LLVM-svn/llvm.git/test/CodeGen/Generic/print-arith-fp.ll
--
Exit Code: 133

Command Output (stderr):
--
Impossible reg-to-reg copy
UNREACHABLE executed at
/Users/fang/local/src/LLVM-svn/llvm/lib/Target/PowerPC/PPCInstrInfo.cpp:672!
Stack dump:
0.      Program arguments:
/Users/fang/local/src/LLVM-svn/gcc40-cmake-build/./bin/llc 
1.      Running pass 'Function Pass Manager' on module '<stdin>'.
2.      Running pass 'Post-RA pseudo instruction expansion pass' on function
'@main'

--

********************
FAIL: LLVM :: CodeGen/Generic/print-arith-int.ll (1455 of 8907)
******************** TEST 'LLVM :: CodeGen/Generic/print-arith-int.ll' FAILED
**
******************
Script:
--
gtimeout 1m  /Users/fang/local/src/LLVM-svn/gcc40-cmake-build/./bin/llc <
/Volumes/Isolde/sources/LLVM-svn/llvm.git/test/CodeGen/Generic/print-arith-int.ll
--
Exit Code: 133

Command Output (stderr):
--
Impossible reg-to-reg copy
UNREACHABLE executed at
/Users/fang/local/src/LLVM-svn/llvm/lib/Target/PowerPC/PPCInstrInfo.cpp:672!
Stack dump:
0.      Program arguments:
/Users/fang/local/src/LLVM-svn/gcc40-cmake-build/./bin/llc 
1.      Running pass 'Function Pass Manager' on module '<stdin>'.
2.      Running pass 'Post-RA pseudo instruction expansion pass' on function
'@main'

--

********************

FAIL: LLVM ::
ExecutionEngine/2003-10-18-PHINode-ConstantExpr-CondCode-Failure.ll (5506 of
8907)
******************** TEST 'LLVM ::
ExecutionEngine/2003-10-18-PHINode-ConstantExpr-CondCode-Failure.ll' FAILED
********************
Script:
--
gtimeout 1m  /Users/fang/local/src/LLVM-svn/gcc40-cmake-build/./bin/lli
/Volumes/Isolde/sources/LLVM-svn/llvm.git/test/ExecutionEngine/2003-10-18-PHINode-ConstantExpr-CondCode-Failure.ll
> /dev/null
--
Exit Code: 133

Command Output (stderr):
--
Impossible reg-to-reg copy
UNREACHABLE executed at
/Users/fang/local/src/LLVM-svn/llvm/lib/Target/PowerPC/PPCInstrInfo.cpp:672!
Stack dump:
0.      Program arguments:
/Users/fang/local/src/LLVM-svn/gcc40-cmake-build/./bin/lli
/Volumes/Isolde/sources/LLVM-svn/llvm.git/test/ExecutionEngine/2003-10-18-PHINode-ConstantExpr-CondCode-Failure.ll 
1.      Running pass 'Post-RA pseudo instruction expansion pass' on function
'@main'

--

********************

That revision range happens to span the adding of CR bit tracking.  Related?
I think Iain is also affected by this on powerpc-darwin9.

-- 
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/20140303/d69243b0/attachment.html>