if we have a large GEP offset on a 32-bit or other target, make
[oota-llvm.git] / test / CodeGen / Generic / vector-casts.ll
2008-08-14 Dan GohmanImprove support for vector casts in LLVM IR and CodeGen.