[llvm] r260625 - AMDGPU: Quick fix for extreme slowness in spill-scavenge-offset.ll test

[Matt Arsenault via llvm-commits]

On 02/11/2016 04:05 PM, Nicolai Haehnle via llvm-commits wrote:
> +; There is something about Tonga that causes this test to spend a lot of time
> +; in the default register allocator.
Could this have something to do with the more complicated set of 
reserved SGPRs?