[LLVMbugs] [Bug 16255] New: llvm-aarch64-linux buildbot unit test failures

[bugzilla-daemon at llvm.org]

http://llvm.org/bugs/show_bug.cgi?id=16255

            Bug ID: 16255
           Summary: llvm-aarch64-linux buildbot unit test failures
           Product: libraries
           Version: trunk
          Hardware: PC
                OS: Linux
            Status: NEW
          Severity: normal
          Priority: P
         Component: Interprocedural Optimizations
          Assignee: unassignedbugs at nondot.org
          Reporter: kristof.beyls at arm.com
                CC: llvmbugs at cs.uiuc.edu
    Classification: Unclassified

The http://lab.llvm.org:8011/builders/llvm-aarch64-linux buildbot has been set
up recently. This buildbot cross-compiles clang with itself, targeting
aarch64-linux, and then runs the LLVM regression tests with the resulting
binary in an aarch64-linux model.
The following regression tests fail, highlighting that there's an incorrect
code generation when clang/llvm compiles itself targeting AArch64 linux:

    LLVM-Unit :: ADT/Release+Asserts/ADTTests/SmallVectorTest/0.InsertRangeTest
    LLVM-Unit ::
ADT/Release+Asserts/ADTTests/SmallVectorTest/0.InsertRepeatedTest
    LLVM-Unit :: ADT/Release+Asserts/ADTTests/SmallVectorTest/1.InsertRangeTest
    LLVM-Unit ::
ADT/Release+Asserts/ADTTests/SmallVectorTest/1.InsertRepeatedTest
    LLVM-Unit :: ADT/Release+Asserts/ADTTests/SmallVectorTest/2.InsertRangeTest
    LLVM-Unit ::
ADT/Release+Asserts/ADTTests/SmallVectorTest/2.InsertRepeatedTest
    LLVM-Unit :: ADT/Release+Asserts/ADTTests/SmallVectorTest/3.InsertRangeTest
    LLVM-Unit ::
ADT/Release+Asserts/ADTTests/SmallVectorTest/3.InsertRepeatedTest
    LLVM-Unit ::
Support/Release+Asserts/SupportTests/AllocationTests/MappedMemoryTest.AllocAndRelease/3
    LLVM-Unit ::
Support/Release+Asserts/SupportTests/AllocationTests/MappedMemoryTest.AllocAndRelease/4
    LLVM-Unit ::
Support/Release+Asserts/SupportTests/AllocationTests/MappedMemoryTest.DuplicateNear/3
    LLVM-Unit ::
Support/Release+Asserts/SupportTests/AllocationTests/MappedMemoryTest.DuplicateNear/4
    LLVM-Unit ::
Support/Release+Asserts/SupportTests/AllocationTests/MappedMemoryTest.EnabledWrite/3
    LLVM-Unit ::
Support/Release+Asserts/SupportTests/AllocationTests/MappedMemoryTest.EnabledWrite/4
    LLVM-Unit ::
Support/Release+Asserts/SupportTests/AllocationTests/MappedMemoryTest.MultipleAllocAndRelease/3
    LLVM-Unit ::
Support/Release+Asserts/SupportTests/AllocationTests/MappedMemoryTest.MultipleAllocAndRelease/4
    LLVM-Unit ::
Support/Release+Asserts/SupportTests/AllocationTests/MappedMemoryTest.SuccessiveNear/3
    LLVM-Unit ::
Support/Release+Asserts/SupportTests/AllocationTests/MappedMemoryTest.SuccessiveNear/4
    LLVM-Unit ::
Support/Release+Asserts/SupportTests/AllocationTests/MappedMemoryTest.UnalignedNear/3
    LLVM-Unit ::
Support/Release+Asserts/SupportTests/AllocationTests/MappedMemoryTest.UnalignedNear/4
    LLVM-Unit ::
Support/Release+Asserts/SupportTests/AllocationTests/MappedMemoryTest.ZeroNear/3
    LLVM-Unit ::
Support/Release+Asserts/SupportTests/AllocationTests/MappedMemoryTest.ZeroNear/4
    LLVM-Unit ::
Support/Release+Asserts/SupportTests/AllocationTests/MappedMemoryTest.ZeroSizeNear/3
    LLVM-Unit ::
Support/Release+Asserts/SupportTests/AllocationTests/MappedMemoryTest.ZeroSizeNear/4

These test failures are probably caused by the DeadArgumentElimination pass
removing function arguments,
resulting in changes in how the AArch64 backend decides which arguments should
go in registers and which
ones should go on the stack.

Since there doesn't seem to be a backend:AArch64 component in bugzilla (yet),
I'm setting the component
to where DeadArgumentElimination lives: IPO.

-- 
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/20130607/75e032e2/attachment.html>