RFA: fix execute/simd-4 for little endian

[Joern Rennecke]

Aldy Hernandez wrote:
> 
> I may be going out on a limb here, but I think this is obviously ok.

If you say so, it is - as you wrote the original test and would know
best if the exact implementation of main has any significance for the
effectiveness of the test.

Thanks, I've checked in the patch.

-- 
--------------------------
SuperH (UK) Ltd.
2410 Aztec West / Almondsbury / BRISTOL / BS32 4QX
T:+44 1454 465658