[llvm] r194538 - Introduce an AnalysisManager which is like a pass manager but with a lot

[Chandler Carruth]

Thanks for the review!

On Fri, Nov 15, 2013 at 6:21 AM, Aaron Ballman <aaron at aaronballman.com>wrote:

> This is causing crashes in MSVC debug builds due to being an invalid
> iterator operation. Specifically, when I == ResultsList.begin(),
> attempting to decrement walks off the front of the iterator list.
>

Doh, thanks. Fixed in r194857.


>
> Once I correct the issue locally, I get several new failing unit tests:
>

I think these are unrelated -- they aren't failing for me after fixing.


>
> E:\llvm\llvm\unittests\IR\ValueMapTest.cpp(189): error: Value of:
> Data.M->tryacquire()
>   Actual: true
> Expected: false
> Mutex should already be locked.
> E:\llvm\llvm\unittests\IR\ValueMapTest.cpp(193): error: Value of:
> Data.M->tryacquire()
>   Actual: true
> Expected: false
> Mutex should already be locked.
> E:\llvm\llvm\unittests\IR\ValueMapTest.cpp(189): error: Value of:
> Data.M->tryacquire()
>   Actual: true
> Expected: false
> Mutex should already be locked.
> E:\llvm\llvm\unittests\IR\ValueMapTest.cpp(193): error: Value of:
> Data.M->tryacquire()
>   Actual: true
> Expected: false
> Mutex should already be locked.
> E:\llvm\llvm\unittests\IR\ValueMapTest.cpp(189): error: Value of:
> Data.M->tryacquire()
>   Actual: true
> Expected: false
> Mutex should already be locked.
> E:\llvm\llvm\unittests\IR\ValueMapTest.cpp(193): error: Value of:
> Data.M->tryacquire()
>   Actual: true
> Expected: false
> Mutex should already be locked.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20131115/881a7a7f/attachment.html>