From aec9240be2bc0bc852167b26fd2e217355ecd745 Mon Sep 17 00:00:00 2001
From: Andrew Trick <atrick@apple.com>
Date: Thu, 26 Apr 2012 21:48:25 +0000
Subject: [PATCH] Fix the SD scheduler to avoid gluing the same node twice.

DAGCombine strangeness may result in multiple loads from the same
offset. They both may try to glue themselves to another load. We could
insist that the redundant loads glue themselves to each other, but the
beter fix is to bail out from bad gluing at the time we detect it.

Fixes rdar://11314175: BuildSchedUnits assert.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@155668 91177308-0d34-0410-b5e6-96231b3b80d8
---
 .../SelectionDAG/ScheduleDAGSDNodes.cpp       |  8 ++--
 test/CodeGen/X86/2012-04-26-sdglue.ll         | 46 +++++++++++++++++++
 2 files changed, 51 insertions(+), 3 deletions(-)
 create mode 100644 test/CodeGen/X86/2012-04-26-sdglue.ll

diff --git a/lib/CodeGen/SelectionDAG/ScheduleDAGSDNodes.cpp b/lib/CodeGen/SelectionDAG/ScheduleDAGSDNodes.cpp
index 69dd813b24e..8ec1ae8620f 100644
--- a/lib/CodeGen/SelectionDAG/ScheduleDAGSDNodes.cpp
+++ b/lib/CodeGen/SelectionDAG/ScheduleDAGSDNodes.cpp
@@ -138,9 +138,11 @@ static void AddGlue(SDNode *N, SDValue Glue, bool AddGlue, SelectionDAG *DAG) {
   // Don't add glue from a node to itself.
   if (GlueDestNode == N) return;
 
-  // Don't add glue to something which already has glue.
-  if (N->getValueType(N->getNumValues() - 1) == MVT::Glue) return;
-
+  // Don't add glue to something that already has it, either as a use or value.
+  if (N->getOperand(N->getNumOperands()-1).getValueType() == MVT::Glue ||
+      N->getValueType(N->getNumValues() - 1) == MVT::Glue) {
+    return;
+  }
   for (unsigned I = 0, E = N->getNumValues(); I != E; ++I)
     VTs.push_back(N->getValueType(I));
 
diff --git a/test/CodeGen/X86/2012-04-26-sdglue.ll b/test/CodeGen/X86/2012-04-26-sdglue.ll
new file mode 100644
index 00000000000..9543587747a
--- /dev/null
+++ b/test/CodeGen/X86/2012-04-26-sdglue.ll
@@ -0,0 +1,46 @@
+; RUN: llc < %s -mtriple=x86_64-apple-darwin -mcpu=core-avx2 -mattr=+avx | FileCheck %s
+; rdar://11314175: SD Scheduler, BuildSchedUnits assert:
+;                  N->getNodeId() == -1 && "Node already inserted!
+
+; It's hard to test for the ISEL condition because CodeGen optimizes
+; away the bugpointed code. Just ensure the basics are still there.
+;CHECK: func:
+;CHECK: vmovups
+;CHECK: vpshufd
+;CHECK: vpshufd
+;CHECK: vmulps
+;CHECK: vmulps
+;CHECK: ret
+
+define void @func() nounwind ssp {
+  %tmp = load <4 x float>* null, align 1
+  %tmp14 = getelementptr <4 x float>* null, i32 2
+  %tmp15 = load <4 x float>* %tmp14, align 1
+  %tmp16 = shufflevector <4 x float> %tmp, <4 x float> <float 0.000000e+00, float undef, float undef, float undef>, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 4, i32 4, i32 4>
+  %tmp17 = call <8 x float> @llvm.x86.avx.vinsertf128.ps.256(<8 x float> %tmp16, <4 x float> undef, i8 1)
+  %tmp18 = bitcast <4 x float> %tmp to <16 x i8>
+  %tmp19 = shufflevector <16 x i8> %tmp18, <16 x i8> undef, <16 x i32> <i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19>
+  %tmp20 = bitcast <16 x i8> %tmp19 to <4 x float>
+  %tmp21 = bitcast <4 x float> %tmp15 to <16 x i8>
+  %tmp22 = shufflevector <16 x i8> undef, <16 x i8> %tmp21, <16 x i32> <i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15, i32 16, i32 17, i32 18, i32 19>
+  %tmp23 = bitcast <16 x i8> %tmp22 to <4 x float>
+  %tmp24 = shufflevector <4 x float> %tmp20, <4 x float> <float 0.000000e+00, float undef, float undef, float undef>, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 4, i32 4, i32 4>
+  %tmp25 = call <8 x float> @llvm.x86.avx.vinsertf128.ps.256(<8 x float> %tmp24, <4 x float> %tmp23, i8 1)
+  %tmp26 = fmul <8 x float> %tmp17, undef
+  %tmp27 = fmul <8 x float> %tmp25, undef
+  %tmp28 = fadd <8 x float> %tmp26, %tmp27
+  %tmp29 = fadd <8 x float> %tmp28, undef
+  %tmp30 = shufflevector <8 x float> %tmp29, <8 x float> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+  %tmp31 = fmul <4 x float> undef, %tmp30
+  %tmp32 = call <8 x float> @llvm.x86.avx.vinsertf128.ps.256(<8 x float> zeroinitializer, <4 x float> %tmp31, i8 1)
+  %tmp33 = fadd <8 x float> undef, %tmp32
+  %tmp34 = call <8 x float> @llvm.x86.avx.hadd.ps.256(<8 x float> %tmp33, <8 x float> undef) nounwind
+  %tmp35 = fsub <8 x float> %tmp34, undef
+  %tmp36 = call <8 x float> @llvm.x86.avx.hadd.ps.256(<8 x float> zeroinitializer, <8 x float> %tmp35) nounwind
+  store <8 x float> %tmp36, <8 x float>* undef, align 32
+  ret void
+}
+
+declare <8 x float> @llvm.x86.avx.vinsertf128.ps.256(<8 x float>, <4 x float>, i8) nounwind readnone
+
+declare <8 x float> @llvm.x86.avx.hadd.ps.256(<8 x float>, <8 x float>) nounwind readnone