From 31ef15e0442ac13135717179d32f438af5bd6ab1 Mon Sep 17 00:00:00 2001
From: Ahsan Saghir <saghir@ca.ibm.com>
Date: Mon, 31 May 2021 08:52:56 -0500
Subject: [PATCH] Teach peephole optimizer to not emit sub-register defs

Peephole optimizer should not be introducing sub-reg definitions
as they are illegal in machine SSA phase. This patch modifies
the optimizer to not emit sub-register definitions.

Reviewed By: arsenm

Differential Revision: https://reviews.llvm.org/D103408
---
 llvm/lib/CodeGen/PeepholeOptimizer.cpp            | 29 ++++++++++++----
 llvm/test/CodeGen/PowerPC/peephole-subreg-def.mir | 41 +++++++++++++++++++++++
 2 files changed, 63 insertions(+), 7 deletions(-)
 create mode 100644 llvm/test/CodeGen/PowerPC/peephole-subreg-def.mir

diff --git a/llvm/lib/CodeGen/PeepholeOptimizer.cpp b/llvm/lib/CodeGen/PeepholeOptimizer.cpp
index 34ac396..49bdba5 100644
--- a/llvm/lib/CodeGen/PeepholeOptimizer.cpp
+++ b/llvm/lib/CodeGen/PeepholeOptimizer.cpp
@@ -585,15 +585,30 @@ optimizeExtInstr(MachineInstr &MI, MachineBasicBlock &MBB,
         MRI->constrainRegClass(DstReg, DstRC);
       }
 
+      // SubReg defs are illegal in machine SSA phase,
+      // we should not generate SubReg defs.
+      //
+      // For example, for the instructions:
+      //
+      // %1:g8rc_and_g8rc_nox0 = EXTSW %0:g8rc
+      // %3:gprc_and_gprc_nor0 = COPY %0.sub_32:g8rc
+      //
+      // We should generate:
+      //
+      // %1:g8rc_and_g8rc_nox0 = EXTSW %0:g8rc
+      // %6:gprc_and_gprc_nor0 = COPY %1.sub_32:g8rc_and_g8rc_nox0
+      // %3:gprc_and_gprc_nor0 = COPY %6:gprc_and_gprc_nor0
+      //
+      if (UseSrcSubIdx)
+        RC = MRI->getRegClass(UseMI->getOperand(0).getReg());
+
       Register NewVR = MRI->createVirtualRegister(RC);
-      MachineInstr *Copy = BuildMI(*UseMBB, UseMI, UseMI->getDebugLoc(),
-                                   TII->get(TargetOpcode::COPY), NewVR)
+      BuildMI(*UseMBB, UseMI, UseMI->getDebugLoc(),
+              TII->get(TargetOpcode::COPY), NewVR)
         .addReg(DstReg, 0, SubIdx);
-      // SubIdx applies to both SrcReg and DstReg when UseSrcSubIdx is set.
-      if (UseSrcSubIdx) {
-        Copy->getOperand(0).setSubReg(SubIdx);
-        Copy->getOperand(0).setIsUndef();
-      }
+      if (UseSrcSubIdx)
+        UseMO->setSubReg(0);
+
       UseMO->setReg(NewVR);
       ++NumReuse;
       Changed = true;
diff --git a/llvm/test/CodeGen/PowerPC/peephole-subreg-def.mir b/llvm/test/CodeGen/PowerPC/peephole-subreg-def.mir
new file mode 100644
index 0000000..e89e206
--- /dev/null
+++ b/llvm/test/CodeGen/PowerPC/peephole-subreg-def.mir
@@ -0,0 +1,41 @@
+# NOTE: Assertions have been autogenerated by utils/update_mir_test_checks.py
+# RUN: llc -mtriple=powerpc64le -simplify-mir -verify-machineinstrs \
+# RUN:   -run-pass=peephole-opt %s -o - | FileCheck %s
+
+# This tests to make sure that we do not generate subreg def
+# as it is illegal to generate subreg defs in machine SSA phase.
+
+---
+name:            test_peephole_subreg_def
+alignment:       16
+tracksRegLiveness: true
+frameInfo:
+  maxAlignment:    1
+machineFunctionInfo: {}
+body:             |
+  bb.0.entry:
+    liveins: $x3
+
+    ; CHECK-LABEL: name: test_peephole_subreg_def
+    ; CHECK: liveins: $x3
+    ; CHECK: [[COPY:%[0-9]+]]:g8rc_and_g8rc_nox0 = COPY $x3
+    ; CHECK: [[ADDI8_:%[0-9]+]]:g8rc = ADDI8 [[COPY]], 1
+    ; CHECK: [[EXTSW:%[0-9]+]]:g8rc_and_g8rc_nox0 = EXTSW [[ADDI8_]]
+    ; CHECK: [[LI8_:%[0-9]+]]:g8rc = LI8 0
+    ; CHECK: STB8 [[LI8_]], 0, [[EXTSW]]
+    ; CHECK: [[COPY1:%[0-9]+]]:gprc_and_gprc_nor0 = COPY [[EXTSW]].sub_32
+    ; CHECK: [[COPY2:%[0-9]+]]:gprc_and_gprc_nor0 = COPY [[COPY1]]
+    ; CHECK: [[ADDI:%[0-9]+]]:gprc = ADDI killed [[COPY2]], 1
+    ; CHECK: [[EXTSW_32_64_:%[0-9]+]]:g8rc_and_g8rc_nox0 = EXTSW_32_64 killed [[ADDI]]
+    ; CHECK: STB8 [[LI8_]], 0, killed [[EXTSW_32_64_]]
+    %0:g8rc_and_g8rc_nox0 = COPY $x3
+    %1:g8rc = ADDI8 %0, 1
+    %2:g8rc_and_g8rc_nox0 = EXTSW %1
+    %3:g8rc = LI8 0
+    STB8 %3, 0, killed %2
+    %4:gprc_and_gprc_nor0 = COPY %1.sub_32
+    %5:gprc = ADDI killed %4, 1
+    %6:g8rc_and_g8rc_nox0 = EXTSW_32_64 killed %5
+    STB8 %3, 0, killed %6
+
+...
-- 
2.7.4