[llvm] r186046 - Set REQUIRES shell on the test cases for r186044
Quentin Colombet
qcolombet at apple.com
Wed Jul 10 18:42:26 PDT 2013
The ‘not’ not in the path thing is not deterministic (are you still following :)).
I will look closer that particular failure.
But the segmentation fault is deterministic.
I suggest you revert it until you figure out what is going on here.
Here is the debugger backtrace so that you see the line numbers.
* thread #1: tid = 0x2c9cdf, 0x0000000100375d5c llc`llvm::Value::getType(this=0x8b48c8758948f875) const + 12 at Value.h:108, queue = 'com.apple.main-thread, stop reason = EXC_BAD_ACCESS (code=EXC_I386_GPFLT)
frame #0: 0x0000000100375d5c llc`llvm::Value::getType(this=0x8b48c8758948f875) const + 12 at Value.h:108
frame #1: 0x0000000100288925 llc`llvm::GlobalValue::getType(this=0x8b48c8758948f875) const + 21 at GlobalValue.h:109
frame #2: 0x00000001013510b0 llc`llvm::TypeFinder::run(this=0x00007fff5fbfe3f8, M=0x0000000101328ec0, onlyNamed=false) + 128 at TypeFinder.cpp:29
frame #3: 0x00000001011964fc llc`llvm::TypePrinting::incorporateTypes(this=0x00007fff5fbfe3f8, M=0x0000000101328ec0) + 76 at AsmWriter.cpp:163
frame #4: 0x0000000101198471 llc`llvm::WriteAsOperand(Out=0x00007fff5fbfe5d8, V=0x0000000102c06ae0, PrintType=false, Context=0x0000000101328ec0) + 225 at AsmWriter.cpp:1191
frame #5: 0x0000000100a86218 llc`lowerConstant(CV=0x0000000102c06ae0, AP=0x0000000102c169c0) + 1032 at AsmPrinter.cpp:1492
frame #6: 0x0000000100a83cc1 llc`emitGlobalConstantImpl(CV=0x0000000102c06ae0, AP=0x0000000102c169c0) + 977 at AsmPrinter.cpp:1922
frame #7: 0x0000000100a7d58d llc`llvm::AsmPrinter::EmitGlobalConstant(this=0x0000000102c169c0, CV=0x0000000102c06ae0) + 93 at AsmPrinter.cpp:1929
frame #8: 0x0000000100a7cd88 llc`llvm::AsmPrinter::EmitGlobalVariable(this=0x0000000102c169c0, GV=0x0000000102c06bd8) + 2584 at AsmPrinter.cpp:414
frame #9: 0x0000000100a80dde llc`llvm::AsmPrinter::doFinalization(this=0x0000000102c169c0, M=0x0000000102c064f0) + 174 at AsmPrinter.cpp:868
frame #10: 0x000000010132769c llc`llvm::FPPassManager::doFinalization(this=0x0000000102c0ae80, M=0x0000000102c064f0) + 92 at PassManager.cpp:1568
frame #11: 0x0000000101327e50 llc`llvm::MPPassManager::runOnModule(this=0x0000000102c07bd0, M=0x0000000102c064f0) + 1904 at PassManager.cpp:1625
frame #12: 0x00000001013286ee llc`llvm::PassManagerImpl::run(this=0x0000000102c078e0, M=0x0000000102c064f0) + 302 at PassManager.cpp:1703
frame #13: 0x0000000101328981 llc`llvm::PassManager::run(this=0x00007fff5fbff4d8, M=0x0000000102c064f0) + 33 at PassManager.cpp:1738
frame #14: 0x0000000100002c40 llc`compileModule(argv=0x00007fff5fbffbc0, Context=0x0000000102c03ce0) + 6320 at llc.cpp:378
frame #15: 0x0000000100001322 llc`main(argc=1, argv=0x00007fff5fbffbc0) + 226 at llc.cpp:196
frame #16: 0x00007fff8857e60d libdyld.dylib`start + 1
Thanks,
-Quentin
On Jul 10, 2013, at 6:22 PM, Quentin Colombet <qcolombet at apple.com> wrote:
> Also is it expected that llc is doing a segmentation fault?
>
> See nonconst-static-ev.ll.tmp
> In my previous post.
>
> Put here again:
> 0 llc 0x0000000109e566ce llvm::sys::PrintStackTrace(__sFILE*) + 46
> 1 llc 0x0000000109e569db PrintStackTraceSignalHandler(void*) + 27
> 2 llc 0x0000000109e56d58 SignalHandler(int) + 408
> 3 libsystem_platform.dylib 0x00007fff86e5f5ca _sigtramp + 26
> 4 libsystem_platform.dylib 0x00007fff5718ae40 _sigtramp + 3492984976
> 5 llc 0x0000000108cfb925 llvm::GlobalValue::getType() const + 21
> 6 llc 0x0000000109dc40b0 llvm::TypeFinder::run(llvm::Module const&, bool) + 128
> 7 llc 0x0000000109c094fc llvm::TypePrinting::incorporateTypes(llvm::Module const&) + 76
> 8 llc 0x0000000109c0b471 llvm::WriteAsOperand(llvm::raw_ostream&, llvm::Value const*, bool, llvm::Module const*) + 225
> 9 llc 0x00000001094f9218 lowerConstant(llvm::Constant const*, llvm::AsmPrinter&) + 1032
> 10 llc 0x00000001094f6cc1 emitGlobalConstantImpl(llvm::Constant const*, llvm::AsmPrinter&) + 977
> 11 llc 0x00000001094f058d llvm::AsmPrinter::EmitGlobalConstant(llvm::Constant const*) + 93
> 12 llc 0x00000001094efd88 llvm::AsmPrinter::EmitGlobalVariable(llvm::GlobalVariable const*) + 2584
> 13 llc 0x00000001094f3dde llvm::AsmPrinter::doFinalization(llvm::Module&) + 174
> 14 llc 0x0000000109d9a69c llvm::FPPassManager::doFinalization(llvm::Module&) + 92
> 15 llc 0x0000000109d9ae50 llvm::MPPassManager::runOnModule(llvm::Module&) + 1904
> 16 llc 0x0000000109d9b6ee llvm::PassManagerImpl::run(llvm::Module&) + 302
> 17 llc 0x0000000109d9b981 llvm::PassManager::run(llvm::Module&) + 33
> 18 llc 0x0000000108a75c40 compileModule(char**, llvm::LLVMContext&) + 6320
> 19 llc 0x0000000108a74322 main + 226
> 20 libdyld.dylib 0x00007fff8857e60d start + 1
> Stack dump:
>
>
> -Quentin
>
> On Jul 10, 2013, at 6:19 PM, Quentin Colombet <qcolombet at apple.com> wrote:
>
>>
>> On Jul 10, 2013, at 6:11 PM, Hal Finkel <hfinkel at anl.gov> wrote:
>>
>>> ----- Original Message -----
>>>>
>>>> The complete output may be useful too!
>>>>
>>>>
>>>> Apparently the path to ‘not' is not expanded.
>>>
>>> Can you see if you can figure out why? I believe that the relevant code below this in test/lit.cfg :
>> I will try to gather more information.
>>>
>>> # Handle these specially as they are strings searched
>>> # for during testing.
>>> r"\| \bcount\b", r"\| \bnot\b"]:
>>>
>>> Also, FWIW, it seems to work on the buildbots, right?
>> Not on some internal buildbot.
>> which is strange.
>>
>>>
>>> -Hal
>>>
>>>>
>>>>
>>>>
>>>>
>>>> -Quentin
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>
>>>> On Jul 10, 2013, at 5:37 PM, Quentin Colombet < qcolombet at apple.com >
>>>> wrote:
>>>>
>>>>
>>>>
>>>> Hi Hal,
>>>>
>>>>
>>>> This is failing for me on OSX:
>>>>
>>>> -bash: not: command not found
>>>>
>>>>
>>>>
>>>> -Quentin
>>>>
>>>>
>>>> On Jul 10, 2013, at 4:25 PM, Hal Finkel < hfinkel at anl.gov > wrote:
>>>>
>>>>
>>>>
>>>> Author: hfinkel
>>>> Date: Wed Jul 10 18:25:03 2013
>>>> New Revision: 186046
>>>>
>>>> URL: http://llvm.org/viewvc/llvm-project?rev=186046&view=rev
>>>> Log:
>>>> Set REQUIRES shell on the test cases for r186044
>>>>
>>>> Trying to fix the i686-mingw32 build.
>>>>
>>>> Modified:
>>>> llvm/trunk/test/Other/nonconst-static-ev.ll
>>>> llvm/trunk/test/Other/nonconst-static-iv.ll
>>>>
>>>> Modified: llvm/trunk/test/Other/nonconst-static-ev.ll
>>>> URL:
>>>> http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Other/nonconst-static-ev.ll?rev=186046&r1=186045&r2=186046&view=diff
>>>> ==============================================================================
>>>> --- llvm/trunk/test/Other/nonconst-static-ev.ll (original)
>>>> +++ llvm/trunk/test/Other/nonconst-static-ev.ll Wed Jul 10 18:25:03
>>>> 2013
>>>> @@ -1,6 +1,7 @@
>>>> ; RUN: not llc < %s 2> %t
>>>> ; RUN: FileCheck --check-prefix=CHECK-ERRORS < %t %s
>>>> ; XFAIL: hexagon
>>>> +; REQUIRES: shell
>>>>
>>>> @0 = global i8 extractvalue ([1 x i8] select (i1 ptrtoint (i32* @1 to
>>>> i1), [1 x i8] [ i8 1 ], [1 x i8] [ i8 2 ]), 0)
>>>> @1 = external global i32
>>>>
>>>> Modified: llvm/trunk/test/Other/nonconst-static-iv.ll
>>>> URL:
>>>> http://llvm.org/viewvc/llvm-project/llvm/trunk/test/Other/nonconst-static-iv.ll?rev=186046&r1=186045&r2=186046&view=diff
>>>> ==============================================================================
>>>> --- llvm/trunk/test/Other/nonconst-static-iv.ll (original)
>>>> +++ llvm/trunk/test/Other/nonconst-static-iv.ll Wed Jul 10 18:25:03
>>>> 2013
>>>> @@ -1,6 +1,7 @@
>>>> ; RUN: not llc < %s 2> %t
>>>> ; RUN: FileCheck --check-prefix=CHECK-ERRORS < %t %s
>>>> ; XFAIL: hexagon
>>>> +; REQUIRES: shell
>>>>
>>>> @0 = global i8 insertvalue( { i8 } select (i1 ptrtoint (i32* @1 to
>>>> i1), { i8 } { i8 1 }, { i8 } { i8 2 }), i8 0, 0)
>>>> @1 = external global i32
>>>>
>>>>
>>>> _______________________________________________
>>>> llvm-commits mailing list
>>>> llvm-commits at cs.uiuc.edu
>>>> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>>>> _______________________________________________
>>>> llvm-commits mailing list
>>>> llvm-commits at cs.uiuc.edu
>>>> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>>>>
>>>
>>> --
>>> Hal Finkel
>>> Assistant Computational Scientist
>>> Leadership Computing Facility
>>> Argonne National Laboratory
>>
>> _______________________________________________
>> llvm-commits mailing list
>> llvm-commits at cs.uiuc.edu
>> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20130710/a5828ec3/attachment.html>
More information about the llvm-commits
mailing list