Broken tests following r160899

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

Broken tests following r160899

Andy Gibbs
Hi,

I'm afraid it looks like commit r160899 has broken two test-cases which are part of the "debuginfo-tests" repository.  I believe I have narrowed it down to the correct commit...

Cheers
Andy



_______________________________________________
LLVM Developers mailing list
[hidden email]         http://llvm.cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
Reply | Threaded
Open this post in threaded view
|

Re: Broken tests following r160899

Andy Gibbs
On Tuesday, July 31, 2012 7:19 PM, Andy Gibbs wrote:
> I'm afraid it looks like commit r160899 has broken two test-cases
> which are part of the "debuginfo-tests" repository.  I believe I
> have narrowed it down to the correct commit...

Erm, but I didn't post the details (obviously too much in a rush to
get home last night! My apologies!)

So here it is:

The tests that fail are sret.cpp and aggregate-indirect-arg.cpp.


Output from r160898 (working) for sret.cpp:
-------------------------------------------

Breakpoint 1 at 0x400867: file llvm/tools/clang/test/debuginfo-tests/sret.cpp, line 61.

Breakpoint 1, B::AInstance (this=0x7fffffffeab8) at llvm/tools/clang/test/debuginfo-tests/sret.cpp:61
61     return a;
$1 = (A &) @0x7fffffffeaa0: {_vptr$A = 0x400ae0, m_int = 12}
A debugging session is active.

 Inferior 1 [process 9573] will be killed.

Quit anyway? (y or n) [answered Y; input not from terminal]



Output from r160899 (broken) for sret.cpp:
------------------------------------------

Breakpoint 1 at 0x400867: file llvm/tools/clang/test/debuginfo-tests/sret.cpp, line 61.

Breakpoint 1, B::AInstance (this=0x7fffffffeab8) at llvm/tools/clang/test/debuginfo-tests/sret.cpp:61
61     return a;
$1 = <optimized out>
A debugging session is active.

 Inferior 1 [process 19005] will be killed.

Quit anyway? (y or n) [answered Y; input not from terminal]



Output from r160898 (working) for aggregate-indirect-arg.cpp:
-------------------------------------------------------------

Breakpoint 1 at 0x4007ec: file llvm/tools/clang/test/debuginfo-tests/aggregate-indirect-arg.cpp, line 22.

Breakpoint 1, A::foo (this=0x7fffffffea80, v=...) at llvm/tools/clang/test/debuginfo-tests/aggregate-indirect-arg.cpp:22
22   void foo(SVal v) { bar(v); }
$1 = (SVal &) @0x7fffffffea70: {Data = 0x0, Kind = 2142}
A debugging session is active.

 Inferior 1 [process 9508] will be killed.

Quit anyway? (y or n) [answered Y; input not from terminal]



Output from r160899 (broken) for aggregate-indirect-arg.cpp:
------------------------------------------------------------

Breakpoint 1 at 0x4007e4 (3 locations)

Breakpoint 1, 0x00000000004007e4 in A::foo(SVal) ()
build/tools/clang/test/debuginfo-tests/Output/aggregate-indirect-arg.cpp.debugger.script:3: Error in sourced command file:
No symbol "v" in current context.



Hope this helps,

Andy



_______________________________________________
LLVM Developers mailing list
[hidden email]         http://llvm.cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev
Reply | Threaded
Open this post in threaded view
|

Re: [cfe-dev] Broken tests following r160899

Eric Christopher-2

On Aug 1, 2012, at 12:51 AM, Andy Gibbs <[hidden email]> wrote:

> Erm, but I didn't post the details (obviously too much in a rush to
> get home last night! My apologies!)

Odd that no one saw that via the bots, but I've reverted that patch for now. It's still a good idea in general, just needs more work.

Thanks for the report!

-eric
_______________________________________________
LLVM Developers mailing list
[hidden email]         http://llvm.cs.uiuc.edu
http://lists.cs.uiuc.edu/mailman/listinfo/llvmdev