[llvm] r198578 - Remove a failing test to get the buildbots back to green.

[Rafael Espíndola]

Why was it failing? Do we have a bug to track getting it fixed?

Deleting a test seems bad practice. We normally revert the full patch
that caused the issue while the problem is investigated.

On 5 January 2014 16:43, Bill Wendling <isanbard at gmail.com> wrote:
> Author: void
> Date: Sun Jan  5 18:43:09 2014
> New Revision: 198578
>
> URL: http://llvm.org/viewvc/llvm-project?rev=198578&view=rev
> Log:
> Remove a failing test to get the buildbots back to green.
>
> Removed:
>     llvm/trunk/test/CodeGen/Generic/builtin-returnaddress-error.ll
>
> Removed: llvm/trunk/test/CodeGen/Generic/builtin-returnaddress-error.ll
> URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/Generic/builtin-returnaddress-error.ll?rev=198577&view=auto
> ==============================================================================
> --- llvm/trunk/test/CodeGen/Generic/builtin-returnaddress-error.ll (original)
> +++ llvm/trunk/test/CodeGen/Generic/builtin-returnaddress-error.ll (removed)
> @@ -1,29 +0,0 @@
> -; RUN: not llc < %s -O0 2> %t1
> -; RUN: FileCheck %s < %t1
> -; XFAIL: ppc64,mips,s390x
> -; PR18363
> -
> -; CHECK: argument to '__builtin_return_address' must be a constant integer
> -
> -define i32* @foo() {
> -entry:
> -  %t1 = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 0, i32 0)
> -  %t2 = extractvalue { i32, i1 } %t1, 0
> -  %t3 = extractvalue { i32, i1 } %t1, 1
> -  br i1 %t3, label %cont, label %trap
> -
> -trap:
> -  call void @llvm.trap()
> -  unreachable
> -
> -cont:
> -  %t5 = call i8* @llvm.returnaddress(i32 %t2)
> -  %t6 = bitcast i8* %t5 to i32*
> -  ret i32* %t6
> -}
> -
> -declare { i32, i1 } @llvm.sadd.with.overflow.i32(i32, i32)
> -
> -declare void @llvm.trap()
> -
> -declare i8* @llvm.returnaddress(i32)
>
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits