[llvm-commits] [llvm-gcc-4.2] r49438 - /llvm-gcc-4.2/trunk/gcc/config/rs6000/rs6000.c

[Dale Johannesen]

This is my doing, not devang's.  I'm not sure what structs with  
nonconstant size ought
to do, but ppc64 doesn't work anyway, so let's worry about it later;  
this patch is fine.

On Apr 9, 2008, at 10:50 AM, Bill Wendling wrote:

> Author: void
> Date: Wed Apr  9 12:50:44 2008
> New Revision: 49438
>
> URL: http://llvm.org/viewvc/llvm-project?rev=49438&view=rev
> Log:
> Temporary fix to get the build working again. Devang, please review.
>
>
> Modified:
>    llvm-gcc-4.2/trunk/gcc/config/rs6000/rs6000.c
>
> Modified: llvm-gcc-4.2/trunk/gcc/config/rs6000/rs6000.c
> URL: http://llvm.org/viewvc/llvm-project/llvm-gcc-4.2/trunk/gcc/config/rs6000/rs6000.c?rev=49438&r1=49437&r2=49438&view=diff
>
> = 
> = 
> = 
> = 
> = 
> = 
> = 
> = 
> ======================================================================
> --- llvm-gcc-4.2/trunk/gcc/config/rs6000/rs6000.c (original)
> +++ llvm-gcc-4.2/trunk/gcc/config/rs6000/rs6000.c Wed Apr  9  
> 12:50:44 2008
> @@ -4906,7 +4906,8 @@
>      means aggregate_value_p is not useful on this target for this  
> purpose.
>      This is a big nasty longterm problem.  For now put things back  
> the
>      way they used to be (wrong, but fewer crashes). */
> -  if (TARGET_64BIT && TREE_INT_CST_LOW(TYPE_SIZE_UNIT(type)) > 8)
> +  if (TARGET_64BIT && TREE_CODE(TYPE_SIZE_UNIT(type)) ==  
> INTEGER_CST &&
> +      TREE_INT_CST_LOW(TYPE_SIZE_UNIT(type)) > 8)
>     return true;
>   /* LLVM LOCAL end strcut return check */
>
>
>
> _______________________________________________
> llvm-commits mailing list
> llvm-commits at cs.uiuc.edu
> http://lists.cs.uiuc.edu/mailman/listinfo/llvm-commits