[llvm-commits] [patch] Use correct alignment when merging zero sized arrays in instruction combine

[Richard Osborne]

On 17/09/12 19:22, Duncan Sands wrote:
> Hi Richard,
>
>> Thanks for your comments. I've updated to patch to use the logic you 
>> suggested.
>
> LGTM modulo the following tiny nit pick:
>
> +; CHECK-NEXT: alloca [0 x i8], align 1024
> +; CHECK-NOT: alloc
>
> Should probably be: ; CHECK-NOT: alloca
>
> Ciao, Duncan.
Good spot. Committed with this fix in r164117. Thanks for the review.

-- 
Richard Osborne | XMOS
http://www.xmos.com