Reapply "[AArch64] Merge LDRSWpre-LD[U]RSW pair into LDPSWpre"

This reverts commit 0def4e6, applies a
quick fix that disallows merging two pre-indexed loads, and adds MIR
regression tests.

Differential Revision: https://reviews.llvm.org/D152407

llvm/lib/Target/AArch64/AArch64InstrInfo.cpp

@@ -2270,6 +2270,7 @@ bool AArch64InstrInfo::hasUnscaledLdStOffset(unsigned Opc) {
   case AArch64::LDRWpre:
   case AArch64::LDURXi:
   case AArch64::LDRXpre:
+  case AArch64::LDRSWpre:
   case AArch64::LDURSWi:
   case AArch64::LDURHHi:
   case AArch64::LDURBBi:
@@ -2479,6 +2480,7 @@ bool AArch64InstrInfo::isPairableLdStInst(const MachineInstr &MI) {
   case AArch64::LDURXi:
   case AArch64::LDRXpre:
   case AArch64::LDURSWi:
+  case AArch64::LDRSWpre:
     return true;
   }
 }
@@ -2599,7 +2601,8 @@ bool AArch64InstrInfo::isCandidateToMergeOrPair(const MachineInstr &MI) const {
   // Can't merge/pair if the instruction modifies the base register.
   // e.g., ldr x0, [x0]
   // This case will never occur with an FI base.
-  // However, if the instruction is an LDR/STR<S,D,Q,W,X>pre, it can be merged.
+  // However, if the instruction is an LDR<S,D,Q,W,X,SW>pre or
+  // STR<S,D,Q,W,X>pre, it can be merged.
   // For example:
   //   ldr q0, [x11, #32]!
   //   ldr q1, [x11, #16]
@@ -3176,6 +3179,7 @@ int AArch64InstrInfo::getMemScale(unsigned Opc) {
   case AArch64::LDRSpre:
   case AArch64::LDRSWui:
   case AArch64::LDURSWi:
+  case AArch64::LDRSWpre:
   case AArch64::LDRWpre:
   case AArch64::LDRWui:
   case AArch64::LDURWi:
@@ -3231,6 +3235,7 @@ bool AArch64InstrInfo::isPreLd(const MachineInstr &MI) {
     return false;
   case AArch64::LDRWpre:
   case AArch64::LDRXpre:
+  case AArch64::LDRSWpre:
   case AArch64::LDRSpre:
   case AArch64::LDRDpre:
   case AArch64::LDRQpre:

llvm/lib/Target/AArch64/AArch64LoadStoreOptimizer.cpp

@@ -293,6 +293,8 @@ static unsigned getMatchingNonSExtOpcode(unsigned Opc,
     return AArch64::LDRWui;
   case AArch64::LDURSWi:
     return AArch64::LDURWi;
+  case AArch64::LDRSWpre:
+    return AArch64::LDRWpre;
   }
 }
 
@@ -372,6 +374,8 @@ static unsigned getMatchingPairOpcode(unsigned Opc) {
   case AArch64::LDRSWui:
   case AArch64::LDURSWi:
     return AArch64::LDPSWi;
+  case AArch64::LDRSWpre:
+    return AArch64::LDPSWpre;
   }
 }
 
@@ -585,6 +589,8 @@ static bool isPreLdStPairCandidate(MachineInstr &FirstMI, MachineInstr &MI) {
     return (OpcB == AArch64::LDRWui) || (OpcB == AArch64::LDURWi);
   case AArch64::LDRXpre:
     return (OpcB == AArch64::LDRXui) || (OpcB == AArch64::LDURXi);
+  case AArch64::LDRSWpre:
+    return (OpcB == AArch64::LDRSWui) || (OpcB == AArch64::LDURSWi);
   }
 }
 
@@ -1318,6 +1324,10 @@ static bool areCandidatesToMergeOrPair(MachineInstr &FirstMI, MachineInstr &MI,
   if (OpcA == OpcB)
     return !AArch64InstrInfo::isPreLdSt(FirstMI);
 
+  // Two pre ld/st of different opcodes cannot be merged either
+  if (AArch64InstrInfo::isPreLdSt(FirstMI) && AArch64InstrInfo::isPreLdSt(MI))
+    return false;
+
   // Try to match a sign-extended load/store with a zero-extended load/store.
   bool IsValidLdStrOpc, PairIsValidLdStrOpc;
   unsigned NonSExtOpc = getMatchingNonSExtOpcode(OpcA, &IsValidLdStrOpc);
@@ -1340,7 +1350,7 @@ static bool areCandidatesToMergeOrPair(MachineInstr &FirstMI, MachineInstr &MI,
     return false;
 
   // The STR<S,D,Q,W,X>pre - STR<S,D,Q,W,X>ui and
-  // LDR<S,D,Q,W,X>pre-LDR<S,D,Q,W,X>ui
+  // LDR<S,D,Q,W,X,SW>pre-LDR<S,D,Q,W,X,SW>ui
   // are candidate pairs that can be merged.
   if (isPreLdStPairCandidate(FirstMI, MI))
     return true;

llvm/test/CodeGen/AArch64/ldrpre-ldr-merge.mir

@@ -588,7 +588,7 @@ body:             |
 
 
 ---
-name:            21-ldrswpre-ldrswui-no-merge
+name:            21-ldrswpre-ldrswui-merge
 tracksRegLiveness: true
 liveins:
   - { reg: '$x0' }
@@ -599,10 +599,9 @@ machineFunctionInfo:
 body:             |
   bb.0:
     liveins: $x0, $x1, $x2
-    ; CHECK-LABEL: name: 21-ldrswpre-ldrswui-no-merge
+    ; CHECK-LABEL: name: 21-ldrswpre-ldrswui-merge
     ; CHECK: liveins: $x0, $x1, $x2
-    ; CHECK: early-clobber renamable $x1, renamable $x0 = LDRSWpre renamable $x1, 40, implicit $w1 :: (load (s32))
-    ; CHECK: renamable $x2 = LDRSWui renamable $x1, 1 :: (load (s32))
+    ; CHECK: early-clobber $x1, renamable $x0, renamable $x2 = LDPSWpre renamable $x1, 10 :: (load (s32))
     ; CHECK: STPXi renamable $x0, renamable $x2, renamable $x1, 0 :: (store (s64))
     ; CHECK: RET undef $lr
     early-clobber renamable $x1, renamable $x0 = LDRSWpre killed renamable $x1, 40 :: (load (s32))
@@ -614,7 +613,7 @@ body:             |
 
 
 ---
-name:            22-ldrswpre-ldurswi-no-merge
+name:            22-ldrswpre-ldurswi-merge
 tracksRegLiveness: true
 liveins:
   - { reg: '$x0' }
@@ -625,10 +624,9 @@ machineFunctionInfo:
 body:             |
   bb.0:
     liveins: $x0, $x1, $x2
-    ; CHECK-LABEL: name: 22-ldrswpre-ldurswi-no-merge
+    ; CHECK-LABEL: name: 22-ldrswpre-ldurswi-merge
     ; CHECK: liveins: $x0, $x1, $x2
-    ; CHECK: early-clobber renamable $x1, renamable $x0 = LDRSWpre renamable $x1, 40, implicit $w1 :: (load (s32))
-    ; CHECK: renamable $x2 = LDURSWi renamable $x1, 4 :: (load (s32))
+    ; CHECK: early-clobber $x1, renamable $x0, renamable $x2 = LDPSWpre renamable $x1, 10 :: (load (s32))
     ; CHECK: STPXi renamable $x0, renamable $x2, renamable $x1, 0 :: (store (s64))
     ; CHECK: RET undef $lr
     early-clobber renamable $x1, renamable $x0 = LDRSWpre killed renamable $x1, 40 :: (load (s32))
@@ -775,3 +773,61 @@ body:             |
     STRXui killed renamable $x2, renamable $x1, 1 :: (store (s64))
     RET undef $lr
 ...
+
+
+---
+name:            28-ldrswpre-ldrwpre-no-merge
+tracksRegLiveness: true
+liveins:
+  - { reg: '$x11' }
+  - { reg: '$x13' }
+machineFunctionInfo:
+  hasRedZone:      false
+body:             |
+  bb.0:
+    liveins: $x11, $x13
+    ; CHECK-LABEL: name: 28-ldrswpre-ldrwpre-no-merge
+    ; CHECK: liveins: $x11, $x13
+    ; CHECK: early-clobber renamable $x11, dead renamable $x10 = LDRSWpre renamable $x11, 8, implicit $w11 :: (load (s32), align 8)
+    ; CHECK: $x14 = EORXrs renamable $x11, renamable $x13, 0
+    ; CHECK: early-clobber renamable $x11, dead renamable $w12 = LDRWpre renamable $x11, 4, implicit $w11 :: (load (s32))
+    ; CHECK: $x13 = EORXrs renamable $x11, renamable $x13, 0
+    ; CHECK: STPXi renamable $x13, renamable $x14, renamable $x11, 0 :: (store (s64))
+    ; CHECK: RET undef $lr
+    early-clobber renamable $x11, renamable $x10 = LDRSWpre killed renamable $x11, 8 :: (load (s32), align 8)
+    $x14 = EORXrs renamable $x11, renamable $x13, 0
+    early-clobber renamable $x11, renamable $w12 = LDRWpre killed renamable $x11, 4 :: (load (s32))
+    $x13 = EORXrs renamable $x11, killed renamable $x13, 0
+    STRXui killed renamable $x13, renamable $x11, 0 :: (store (s64))
+    STRXui killed renamable $x14, renamable $x11, 1 :: (store (s64))
+    RET undef $lr
+...
+
+
+---
+name:            29-ldrwpre-ldrswpre-no-merge
+tracksRegLiveness: true
+liveins:
+  - { reg: '$x11' }
+  - { reg: '$x13' }
+machineFunctionInfo:
+  hasRedZone:      false
+body:             |
+  bb.0:
+    liveins: $x11, $x13
+    ; CHECK-LABEL: name: 29-ldrwpre-ldrswpre-no-merge
+    ; CHECK: liveins: $x11, $x13
+    ; CHECK: early-clobber renamable $x11, dead renamable $w12 = LDRWpre renamable $x11, 8, implicit $w11 :: (load (s32))
+    ; CHECK: $x14 = EORXrs renamable $x11, renamable $x13, 0
+    ; CHECK: early-clobber renamable $x11, dead renamable $x10 = LDRSWpre renamable $x11, 4, implicit $w11 :: (load (s32), align 8)
+    ; CHECK: $x13 = EORXrs renamable $x11, renamable $x13, 0
+    ; CHECK: STPXi renamable $x13, renamable $x14, renamable $x11, 0 :: (store (s64))
+    ; CHECK: RET undef $lr
+    early-clobber renamable $x11, renamable $w12 = LDRWpre killed renamable $x11, 8 :: (load (s32))
+    $x14 = EORXrs renamable $x11, renamable $x13, 0
+    early-clobber renamable $x11, renamable $x10 = LDRSWpre killed renamable $x11, 4 :: (load (s32), align 8)
+    $x13 = EORXrs renamable $x11, killed renamable $x13, 0
+    STRXui killed renamable $x13, renamable $x11, 0 :: (store (s64))
+    STRXui killed renamable $x14, renamable $x11, 1 :: (store (s64))
+    RET undef $lr
+...