[SLPVectorizer] Don't insert an extractelement before a catchswitch

A catchswitch cannot be preceded by another instruction in the same
basic block (other than a PHI node).

Instead, insert the extract element right after the materialization of
the vectorized value.  This isn't optimal but is a reasonable compromise
given the constraints of WinEH.

This fixes PR27163.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@265157 91177308-0d34-0410-b5e6-96231b3b80d8

lib/Transforms/Vectorize/SLPVectorizer.cpp

@@ -2570,11 +2570,18 @@ Value *BoUpSLP::vectorizeTree() {
     Value *Lane = Builder.getInt32(it->Lane);
     // Generate extracts for out-of-tree users.
     // Find the insertion point for the extractelement lane.
-    if (isa<Instruction>(Vec)){
+    if (auto *VecI = dyn_cast<Instruction>(Vec)) {
       if (PHINode *PH = dyn_cast<PHINode>(User)) {
         for (int i = 0, e = PH->getNumIncomingValues(); i != e; ++i) {
           if (PH->getIncomingValue(i) == Scalar) {
-            Builder.SetInsertPoint(PH->getIncomingBlock(i)->getTerminator());
+            TerminatorInst *IncomingTerminator =
+                PH->getIncomingBlock(i)->getTerminator();
+            if (isa<CatchSwitchInst>(IncomingTerminator)) {
+              Builder.SetInsertPoint(VecI->getParent(),
+                                     std::next(VecI->getIterator()));
+            } else {
+              Builder.SetInsertPoint(PH->getIncomingBlock(i)->getTerminator());
+            }
             Value *Ex = Builder.CreateExtractElement(Vec, Lane);
             if (MinBWs.count(ScalarRoot))
               Ex = Builder.CreateSExt(Ex, Scalar->getType());

test/Transforms/SLPVectorizer/X86/pr27163.ll

@@ -0,0 +1,50 @@
+; RUN: opt -slp-vectorizer -S < %s | FileCheck %s
+target datalayout = "e-m:w-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-pc-windows-msvc18.0.0"
+
+%struct.B = type { i64, i64 }
+
+define void @test1(%struct.B* %p) personality i32 (...)* @__CxxFrameHandler3 {
+invoke.cont:
+  %gep1 = getelementptr inbounds %struct.B, %struct.B* %p, i64 0, i32 0
+  %gep2 = getelementptr inbounds %struct.B, %struct.B* %p, i64 0, i32 1
+  %load1 = load i64, i64* %gep1, align 8
+  %load2 = load i64, i64* %gep2, align 8
+  store i64 %load1, i64* %gep1, align 8
+  store i64 %load2, i64* %gep2, align 8
+  invoke void @throw()
+          to label %unreachable unwind label %catch.dispatch
+
+catch.dispatch:                                   ; preds = %invoke.cont
+  %cs = catchswitch within none [label %invoke.cont1] unwind label %ehcleanup
+
+invoke.cont1:                                     ; preds = %catch.dispatch
+  %catch = catchpad within %cs [i8* null, i32 64, i8* null]
+  invoke void @throw() [ "funclet"(token %catch) ]
+          to label %unreachable unwind label %ehcleanup
+
+ehcleanup:                                        ; preds = %invoke.cont1, %catch.dispatch
+  %phi = phi i64 [ %load1, %catch.dispatch ], [ 9, %invoke.cont1 ]
+  %cleanup = cleanuppad within none []
+  call void @release(i64 %phi) [ "funclet"(token %cleanup) ]
+  cleanupret from %cleanup unwind to caller
+
+unreachable:                                      ; preds = %invoke.cont1, %invoke.cont
+  unreachable
+}
+
+
+; CHECK-LABEL: define void @test1(
+; CHECK: %[[gep:.*]] = getelementptr inbounds %struct.B, %struct.B* %p, i64 0, i32 0
+; CHECK: %[[bc:.*]]  = bitcast i64* %[[gep]] to <2 x i64>*
+; CHECK: %[[ld:.*]]  = load <2 x i64>, <2 x i64>* %[[bc]], align 8
+; CHECK: %[[ee:.*]]  = extractelement <2 x i64> %[[ld]], i32 0
+
+; CHECK: %[[phi:.*]] = phi i64 [ %[[ee]], {{.*}} ], [ 9, {{.*}} ]
+; CHECK: call void @release(i64 %[[phi]])
+
+declare i32 @__CxxFrameHandler3(...)
+
+declare void @throw()
+
+declare void @release(i64)