]> granicus.if.org Git - llvm/commit
Merging r268259:
authorTom Stellard <thomas.stellard@amd.com>
Sat, 4 Jun 2016 03:43:05 +0000 (03:43 +0000)
committerTom Stellard <thomas.stellard@amd.com>
Sat, 4 Jun 2016 03:43:05 +0000 (03:43 +0000)
commit28fe7a0044c321354e5ed830e1b2572f7f3360f2
tree33cbc0bd1a8b40d3a20b6fe219f22ed6c9fd5f2a
parent3dcb576961931fa9a653b42cc467f29044ec06d7
Merging r268259:

------------------------------------------------------------------------
r268259 | nhaehnle | 2016-05-02 10:37:01 -0700 (Mon, 02 May 2016) | 14 lines

AMDGPU: llvm.SI.fs.constant is a source of divergence

Summary:
This intrinsic is used to get flat-shaded fragment shader inputs. Those are
uniform across a primitive, but a fragment shader wave may process pixels from
multiple primitives (as indicated by the prim_mask), and so that's where
divergence can arise.

Reviewers: arsenm, tstellarAMD

Subscribers: arsenm, llvm-commits

Differential Revision: http://reviews.llvm.org/D19747

------------------------------------------------------------------------

git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_38@271769 91177308-0d34-0410-b5e6-96231b3b80d8
lib/Target/AMDGPU/AMDGPUTargetTransformInfo.cpp
test/Analysis/DivergenceAnalysis/AMDGPU/interp-intrinsics.ll [new file with mode: 0644]