[llvm] [AMDGPU] Improved Lowering of abs(i16) and -abs(i16) (PR #165626)

[Patrick Simmons via llvm-commits]

linuxrocks123 wrote:

@jmmartinez interestingly, the problem doesn't exist with global isel enabled.  The main branch generates the optimal instruction selection without this PR.

If we are planning to switch to global isel soon, we may not need this.

https://github.com/llvm/llvm-project/pull/165626