| commit | fc8387b8d187909ead65b8e3b02f9701c377a84f | [log] [tgz] |
|---|---|---|
| author | Matt Arsenault <Matthew.Arsenault@amd.com> | Wed Mar 08 00:48:46 2017 +0000 |
| committer | Matt Arsenault <Matthew.Arsenault@amd.com> | Wed Mar 08 00:48:46 2017 +0000 |
| tree | c2fe62b4587dce269575cc78ea0dafb9a5a7e84a | |
| parent | df032da7f1a1df65433f504223f4c4cf7a620332 [diff] |
AMDGPU: Constant fold rcp node When doing arcp optimization with a constant denominator, this was leaving behind rcps with constant inputs. git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@297248 91177308-0d34-0410-b5e6-96231b3b80d8