From d6b5bd6f44e8091a4c4870f1c52921c25a4f8cca Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Micha=C5=82=20G=C3=B3rny?= <mgorny@gentoo.org>
Date: Sat, 21 Jan 2017 12:35:36 +0100
Subject: [PATCH] Revert "AMDGPU: Fix an interaction between WQM and polygon
 stippling"

https://bugs.gentoo.org/603858
---
 lib/Target/AMDGPU/SIInstructions.td   |  1 -
 lib/Target/AMDGPU/SIWholeQuadMode.cpp |  7 +++++
 test/CodeGen/AMDGPU/wqm.ll            | 49 +++--------------------------------
 3 files changed, 11 insertions(+), 46 deletions(-)

diff --git a/lib/Target/AMDGPU/SIInstructions.td b/lib/Target/AMDGPU/SIInstructions.td
index dde5f2fc6b4..18b7d5d62ef 100644
--- a/lib/Target/AMDGPU/SIInstructions.td
+++ b/lib/Target/AMDGPU/SIInstructions.td
@@ -2029,7 +2029,6 @@ def SI_RETURN : PseudoInstSI <
   let hasSideEffects = 1;
   let SALU = 1;
   let hasNoSchedulingInfo = 1;
-  let DisableWQM = 1;
 }
 
 let Uses = [EXEC], Defs = [EXEC, VCC, M0],
diff --git a/lib/Target/AMDGPU/SIWholeQuadMode.cpp b/lib/Target/AMDGPU/SIWholeQuadMode.cpp
index 1534d582569..b200c153df0 100644
--- a/lib/Target/AMDGPU/SIWholeQuadMode.cpp
+++ b/lib/Target/AMDGPU/SIWholeQuadMode.cpp
@@ -219,6 +219,13 @@ char SIWholeQuadMode::scanInstructions(MachineFunction &MF,
       markInstruction(MI, Flags, Worklist);
       GlobalFlags |= Flags;
     }
+
+    if (WQMOutputs && MBB.succ_empty()) {
+      // This is a prolog shader. Make sure we go back to exact mode at the end.
+      Blocks[&MBB].OutNeeds = StateExact;
+      Worklist.push_back(&MBB);
+      GlobalFlags |= StateExact;
+    }
   }
 
   return GlobalFlags;
diff --git a/test/CodeGen/AMDGPU/wqm.ll b/test/CodeGen/AMDGPU/wqm.ll
index 41e42645788..809a7ba9b82 100644
--- a/test/CodeGen/AMDGPU/wqm.ll
+++ b/test/CodeGen/AMDGPU/wqm.ll
@@ -17,18 +17,17 @@ main_body:
 ;CHECK-LABEL: {{^}}test2:
 ;CHECK-NEXT: ; %main_body
 ;CHECK-NEXT: s_wqm_b64 exec, exec
+;CHECK: image_sample
 ;CHECK-NOT: exec
-define amdgpu_ps void @test2(<8 x i32> inreg %rsrc, <4 x i32> inreg %sampler, float addrspace(1)* inreg %ptr, <4 x i32> %c) {
+;CHECK: _load_dword v0,
+define amdgpu_ps float @test2(<8 x i32> inreg %rsrc, <4 x i32> inreg %sampler, float addrspace(1)* inreg %ptr, <4 x i32> %c) {
 main_body:
   %c.1 = call <4 x float> @llvm.SI.image.sample.v4i32(<4 x i32> %c, <8 x i32> %rsrc, <4 x i32> %sampler, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
   %c.2 = bitcast <4 x float> %c.1 to <4 x i32>
   %c.3 = extractelement <4 x i32> %c.2, i32 0
   %gep = getelementptr float, float addrspace(1)* %ptr, i32 %c.3
   %data = load float, float addrspace(1)* %gep
-
-  call void @llvm.SI.export(i32 15, i32 1, i32 1, i32 0, i32 1, float %data, float undef, float undef, float undef)
-
-  ret void
+  ret float %data
 }
 
 ; ... but disabled for stores (and, in this simple case, not re-enabled).
@@ -415,46 +414,6 @@ entry:
   ret void
 }
 
-; Must return to exact at the end of a non-void returning shader,
-; otherwise the EXEC mask exported by the epilog will be wrong. This is true
-; even if the shader has no kills, because a kill could have happened in a
-; previous shader fragment.
-;
-; CHECK-LABEL: {{^}}test_nonvoid_return:
-; CHECK: s_mov_b64 [[LIVE:s\[[0-9]+:[0-9]+\]]], exec
-; CHECK: s_wqm_b64 exec, exec
-;
-; CHECK: s_and_b64 exec, exec, [[LIVE]]
-; CHECK-NOT: exec
-define amdgpu_ps <4 x float> @test_nonvoid_return() nounwind {
-  %tex = call <4 x float> @llvm.SI.image.sample.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
-  %tex.i = bitcast <4 x float> %tex to <4 x i32>
-  %dtex = call <4 x float> @llvm.SI.image.sample.v4i32(<4 x i32> %tex.i, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
-  ret <4 x float> %dtex
-}
-
-; CHECK-LABEL: {{^}}test_nonvoid_return_unreachable:
-; CHECK: s_mov_b64 [[LIVE:s\[[0-9]+:[0-9]+\]]], exec
-; CHECK: s_wqm_b64 exec, exec
-;
-; CHECK: s_and_b64 exec, exec, [[LIVE]]
-; CHECK-NOT: exec
-define amdgpu_ps <4 x float> @test_nonvoid_return_unreachable(i32 inreg %c) nounwind {
-entry:
-  %tex = call <4 x float> @llvm.SI.image.sample.v4i32(<4 x i32> undef, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
-  %tex.i = bitcast <4 x float> %tex to <4 x i32>
-  %dtex = call <4 x float> @llvm.SI.image.sample.v4i32(<4 x i32> %tex.i, <8 x i32> undef, <4 x i32> undef, i32 15, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0, i32 0)
-
-  %cc = icmp sgt i32 %c, 0
-  br i1 %cc, label %if, label %else
-
-if:
-  store volatile <4 x float> %dtex, <4 x float>* undef
-  unreachable
-
-else:
-  ret <4 x float> %dtex
-}
 
 declare void @llvm.amdgcn.image.store.v4i32(<4 x float>, <4 x i32>, <8 x i32>, i32, i1, i1, i1, i1) #1
 declare void @llvm.amdgcn.buffer.store.f32(float, <4 x i32>, i32, i32, i1, i1) #1
-- 
2.11.0