X86ISelLowering: Fix cmov(cmov) special lowering bug

There's a special case in EmitLoweredSelect() that produces an improved
lowering for cmov(cmov) patterns. However this special lowering is
currently broken if the inner cmov has multiple users so this patch
stops using it in this case.

If you wonder why this wasn't fixed by continuing to use the special
lowering and inserting a 2nd PHI for the inner cmov: I believe this
would incur additional copies/register pressure so the special lowering
does not improve upon the normal one anymore in this case.

This fixes http://llvm.org/PR26256 (= rdar://24329747)

llvm-svn: 258729

2 files changed, 51 insertions, 1 deletions

diff --git a/llvm/lib/Target/X86/X86ISelLowering.cpp b/llvm/lib/Target/X86/X86ISelLowering.cpp
index 6d104e9bf72..7bfa5de6632 100644
--- a/llvm/lib/Target/X86/X86ISelLowering.cpp
+++ b/llvm/lib/Target/X86/X86ISelLowering.cpp
@@ -22050,7 +22050,8 @@ X86TargetLowering::EmitLoweredSelect(MachineInstr *MI,
   if (LastCMOV == MI &&
       NextMIIt != BB->end() && NextMIIt->getOpcode() == MI->getOpcode() &&
       NextMIIt->getOperand(2).getReg() == MI->getOperand(2).getReg() &&
-      NextMIIt->getOperand(1).getReg() == MI->getOperand(0).getReg()) {
+      NextMIIt->getOperand(1).getReg() == MI->getOperand(0).getReg() &&
+      NextMIIt->getOperand(1).isKill()) {
     CascadedCMOV = &*NextMIIt;
   }
 
diff --git a/llvm/test/CodeGen/X86/cmovcmov.ll b/llvm/test/CodeGen/X86/cmovcmov.ll
index d3d9748d653..9363d31866d 100644
--- a/llvm/test/CodeGen/X86/cmovcmov.ll
+++ b/llvm/test/CodeGen/X86/cmovcmov.ll
@@ -224,3 +224,52 @@ entry:
 }
 
 attributes #0 = { nounwind }
+
+@g8 = global i8 0
+
+; The following test failed because llvm had a bug where a structure like:
+;
+; %vreg12<def> = CMOV_GR8 %vreg7, %vreg11 ... (lt)
+; %vreg13<def> = CMOV_GR8 %vreg12, %vreg11 ... (gt)
+;
+; was lowered to:
+;
+; The first two cmovs got expanded to:
+; BB#0:
+;   JL_1 BB#9
+; BB#7:
+;   JG_1 BB#9
+; BB#8:
+; BB#9:
+;   vreg12 = phi(vreg7, BB#8, vreg11, BB#0, vreg12, BB#7)
+;   vreg13 = COPY vreg12
+; Which was invalid as %vreg12 is not the same value as %vreg13
+
+; CHECK-LABEL: no_cascade_opt:
+; CMOV-DAG: cmpl %edx, %esi
+; CMOV-DAG: movb $20, %al
+; CMOV-DAG: movb $20, %dl
+; CMOV:   jl [[BB0:.LBB[0-9_]+]]
+; CMOV:   movb %cl, %dl
+; CMOV: [[BB0]]:
+; CMOV:   jg [[BB1:.LBB[0-9_]+]]
+; CMOV:   movb %dl, %al
+; CMOV: [[BB1]]:
+; CMOV:   testl %edi, %edi
+; CMOV:   je [[BB2:.LBB[0-9_]+]]
+; CMOV:   movb %dl, %al
+; CMOV: [[BB2]]:
+; CMOV:   movb %al, g8(%rip)
+; CMOV:   retq
+define void @no_cascade_opt(i32 %v0, i32 %v1, i32 %v2, i32 %v3) {
+entry:
+  %c0 = icmp eq i32 %v0, 0
+  %c1 = icmp slt i32 %v1, %v2
+  %c2 = icmp sgt i32 %v1, %v2
+  %trunc = trunc i32 %v3 to i8
+  %sel0 = select i1 %c1, i8 20, i8 %trunc
+  %sel1 = select i1 %c2, i8 20, i8 %sel0
+  %sel2 = select i1 %c0, i8 %sel1, i8 %sel0
+  store volatile i8 %sel2, i8* @g8
+  ret void
+}