[llvm] r268529 - [ARM] Fix Scavenger assert due to underestimated stack size

[Sanjay Patel via llvm-commits]

Those failures were caused by my patches:
http://reviews.llvm.org/rL268550
http://reviews.llvm.org/rL268751

but this version has now passed the danger stage for that bot:
http://reviews.llvm.org/rL268767

If anyone can see how the earlier versions could fail, I'd love to know. :)


On Fri, May 6, 2016 at 11:57 AM, Renato Golin via llvm-commits <
llvm-commits at lists.llvm.org> wrote:

> On 6 May 2016 at 17:56, Zhao, Weiming <weimingz at codeaurora.org> wrote:
> > I tried to build my local LLVM with Memory Sanitizer, but it failed
> because
> > the generated llvmTblGen hit the same use-of-undef-var error.  Do you
> know
> > if there is any other way we can reproduce the error?
>
> To be honest, the MSAN is still failing on other AArch64 tests:
>
> http://lab.llvm.org:8011/builders/sanitizer-x86_64-linux-fast/builds/12575
>
> Failing Tests (8):
>     LLVM :: CodeGen/AArch64/arm64-2012-01-11-ComparisonDAGCrash.ll
>     LLVM :: CodeGen/AArch64/arm64-andCmpBrToTBZ.ll
>     LLVM :: CodeGen/ARM/2009-06-04-MissingLiveIn.ll
>     LLVM :: CodeGen/ARM/2009-09-13-InvalidSubreg.ll
>     LLVM :: CodeGen/ARM/2010-06-21-LdStMultipleBug.ll
>     LLVM :: Transforms/SimplifyCFG/2006-06-12-InfLoop.ll
>     LLVM :: Transforms/SimplifyCFG/2008-07-13-InfLoopMiscompile.ll
>     LLVM :: Transforms/SimplifyCFG/branch-fold.ll
>
> Which are very similar to the ones in the build that had your patch. I
> think you can re-commit your patch as it was, and we should look for
> the source of the uninitialised variables elsewhere.
>
> cheers,
> --renato
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at lists.llvm.org
> http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-commits
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20160506/ef329efc/attachment.html>