[llvm-commits] make check failures

[Chris Lattner]

On Nov 13, 2007, at 10:46 AM, Tanya Lattner wrote:

> There are quite a few make check failures that the nightly testers  
> are reporting.

> On my machine (x86 Mac OsX) I am seeing the following with TOT (svn: )
> test/CFrontend/2007-11-07-CopyAggregateAlign.c
> test/Transforms/GlobalOpt/deadglobal-2.llx

The second was my fault, it is fixed now.  The first also fails for  
me on ppc/darwin, I'm not responsible for that one.  See below.

> From the nightly testers last night:
>
> OldGrawp (x86):
> test/CFrontend/2007-10-30-Volatile.c
> test/CFrontend/2007-11-07-CopyAggregateAlign.c
> test/Transforms/GlobalOpt/deadglobal-2.llx

the last one is fixed, the first one is probably due to evan's cfe  
being out of date.

> sabre-ppc32:
> test/CFrontend/2007-10-15-VoidPtr.c
> test/CFrontend/2007-10-30-Volatile.c
> test/CFrontend/sret2.c
> test/FrontendObjC/2007-10-17-SJLJExceptions.m
> test/FrontendObjC/2007-10-23-GC-WriteBarrier.m
> test/Transforms/GlobalOpt/deadglobal-2.llx

All of these were because I haven't updated my cfe recently.  These  
should pass tonight.

> test/CFrontend/2007-11-07-CopyAggregateAlign.c

This still fails for me.  Duncan, thoughts?  Greping for align 2  
yields just two hits:

         %tmp14 = load i16* %tmp13, align 2              ; <i16>  
[#uses=1]
         store i16 %tmp14, i16* %tmp12, align 2

not 6.

-Chris