diff --git a/llvm/lib/Target/AMDGPU/AMDGPU.td b/llvm/lib/Target/AMDGPU/AMDGPU.td
--- a/llvm/lib/Target/AMDGPU/AMDGPU.td
+++ b/llvm/lib/Target/AMDGPU/AMDGPU.td
@@ -415,7 +415,7 @@
   "Support DPP (Data Parallel Primitives) extension"
 >;
 
-// DPP8 allows arbitrary cross-lane swizzling withing groups of 8 lanes.
+// DPP8 allows arbitrary cross-lane swizzling within groups of 8 lanes.
 def FeatureDPP8 : SubtargetFeature<"dpp8",
   "HasDPP8",
   "true",
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.h b/llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.h
--- a/llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.h
+++ b/llvm/lib/Target/AMDGPU/AMDGPUInstrInfo.h
@@ -76,8 +76,8 @@
 };
 const ImageDimIntrinsicInfo *getImageDimIntrinsicInfo(unsigned Intr);
 
-const ImageDimIntrinsicInfo *getImageDimInstrinsicByBaseOpcode(unsigned BaseOpcode,
-                                                               unsigned Dim);
+const ImageDimIntrinsicInfo *
+getImageDimIntrinsicByBaseOpcode(unsigned BaseOpcode, unsigned Dim);
 
 } // end AMDGPU namespace
 } // End llvm namespace
diff --git a/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.h b/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.h
--- a/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.h
+++ b/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.h
@@ -109,8 +109,8 @@
                                      Register Den) const;
 
   void legalizeUnsignedDIV_REM64Impl(MachineIRBuilder &B, Register DstDivReg,
-                                     Register DstRemReg, Register Numer,
-                                     Register Denom) const;
+                                     Register DstRemReg, Register Num,
+                                     Register Den) const;
 
   bool legalizeSignedDIV_REM(MachineInstr &MI, MachineRegisterInfo &MRI,
                              MachineIRBuilder &B) const;
diff --git a/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp b/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
--- a/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPULegalizerInfo.cpp
@@ -4408,8 +4408,8 @@
       if (ConstantLod->isZero() || ConstantLod->isNegative()) {
         // Set new opcode to _lz variant of _l, and change the intrinsic ID.
         const AMDGPU::ImageDimIntrinsicInfo *NewImageDimIntr =
-            AMDGPU::getImageDimInstrinsicByBaseOpcode(LZMappingInfo->LZ,
-                                                      Intr->Dim);
+            AMDGPU::getImageDimIntrinsicByBaseOpcode(LZMappingInfo->LZ,
+                                                     Intr->Dim);
 
         // The starting indexes should remain in the same place.
         --CorrectedNumVAddrs;
diff --git a/llvm/lib/Target/AMDGPU/AMDGPULibCalls.cpp b/llvm/lib/Target/AMDGPU/AMDGPULibCalls.cpp
--- a/llvm/lib/Target/AMDGPU/AMDGPULibCalls.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPULibCalls.cpp
@@ -125,7 +125,7 @@
   BasicBlock::iterator getEntryIns(CallInst * UI);
   // Insert an Alloc instruction.
   AllocaInst* insertAlloca(CallInst * UI, IRBuilder<> &B, const char *prefix);
-  // Get a scalar native builtin signle argument FP function
+  // Get a scalar native builtin single argument FP function
   FunctionCallee getNativeFunction(Module *M, const FuncInfo &FInfo);
 
 protected:
diff --git a/llvm/lib/Target/AMDGPU/AMDGPULibFunc.cpp b/llvm/lib/Target/AMDGPU/AMDGPULibFunc.cpp
--- a/llvm/lib/Target/AMDGPU/AMDGPULibFunc.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPULibFunc.cpp
@@ -455,7 +455,8 @@
       break;
     }
 
-    default: llvm_unreachable("Unhandeled param rule");
+    default:
+      llvm_unreachable("Unhandled param rule");
     }
   }
   ++Index;
@@ -747,7 +748,8 @@
   case AMDGPULibFunc::IMG3D:   return "11ocl_image3d";
   case AMDGPULibFunc::SAMPLER: return "11ocl_sampler";
   case AMDGPULibFunc::EVENT:   return "9ocl_event";
-  default: llvm_unreachable("Unhandeled param type");
+  default:
+    llvm_unreachable("Unhandled param type");
   }
   return nullptr;
 }
@@ -761,7 +763,7 @@
 // substitution candidates from the grammar, but are explicitly excluded:
 // 1. <builtin-type> other than vendor extended types ..."
 
-// For the purpose of functions the following productions make sence for the
+// For the purpose of functions the following productions make sense for the
 // substitution:
 //  <type> ::= <builtin-type>
 //    ::= <class-enum-type>
@@ -774,8 +776,8 @@
 // using <class-enum-type> production rule they're not used for substitution
 // because clang consider them as builtin types.
 //
-// DvNN_ type is GCC extension for vectors and is a subject for the substitution.
-
+// DvNN_ type is GCC extension for vectors and is a subject for the
+// substitution.
 
 class ItaniumMangler {
   SmallVector<AMDGPULibFunc::Param, 10> Str; // list of accumulated substitutions
@@ -902,7 +904,7 @@
   case AMDGPULibFunc::EVENT:
     T = StructType::create(C,"ocl_event")->getPointerTo(); break;
   default:
-    llvm_unreachable("Unhandeled param type");
+    llvm_unreachable("Unhandled param type");
     return nullptr;
   }
   if (P.VectorSize > 1)
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUOpenCLEnqueuedBlockLowering.cpp b/llvm/lib/Target/AMDGPU/AMDGPUOpenCLEnqueuedBlockLowering.cpp
--- a/llvm/lib/Target/AMDGPU/AMDGPUOpenCLEnqueuedBlockLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUOpenCLEnqueuedBlockLowering.cpp
@@ -13,7 +13,7 @@
 //
 // In LLVM CodeGen the runtime-handle metadata will be translated to
 // RuntimeHandle metadata in code object. Runtime allocates a global buffer
-// for each kernel with RuntimeHandel metadata and saves the kernel address
+// for each kernel with RuntimeHandle metadata and saves the kernel address
 // required for the AQL packet into the buffer. __enqueue_kernel function
 // in device library knows that the invoke function pointer in the block
 // literal is actually runtime handle and loads the kernel address from it
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp b/llvm/lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp
--- a/llvm/lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp
+++ b/llvm/lib/Target/AMDGPU/AMDGPUPromoteAlloca.cpp
@@ -274,7 +274,7 @@
 
   // We could do a single 64-bit load here, but it's likely that the basic
   // 32-bit and extract sequence is already present, and it is probably easier
-  // to CSE this. The loads should be mergable later anyway.
+  // to CSE this. The loads should be mergeable later anyway.
   Value *GEPXY = Builder.CreateConstInBoundsGEP1_64(I32Ty, CastDispatchPtr, 1);
   LoadInst *LoadXY = Builder.CreateAlignedLoad(I32Ty, GEPXY, Align(4));
 
diff --git a/llvm/lib/Target/AMDGPU/AMDGPUSubtarget.h b/llvm/lib/Target/AMDGPU/AMDGPUSubtarget.h
--- a/llvm/lib/Target/AMDGPU/AMDGPUSubtarget.h
+++ b/llvm/lib/Target/AMDGPU/AMDGPUSubtarget.h
@@ -251,7 +251,7 @@
   uint64_t getExplicitKernArgSize(const Function &F, Align &MaxAlign) const;
   unsigned getKernArgSegmentSize(const Function &F, Align &MaxAlign) const;
 
-  /// \returns Corresponsing DWARF register number mapping flavour for the
+  /// \returns Corresponding DWARF register number mapping flavour for the
   /// \p WavefrontSize.
   AMDGPUDwarfFlavour getAMDGPUDwarfFlavour() const;
 
diff --git a/llvm/lib/Target/AMDGPU/GCNHazardRecognizer.cpp b/llvm/lib/Target/AMDGPU/GCNHazardRecognizer.cpp
--- a/llvm/lib/Target/AMDGPU/GCNHazardRecognizer.cpp
+++ b/llvm/lib/Target/AMDGPU/GCNHazardRecognizer.cpp
@@ -1545,7 +1545,7 @@
 }
 
 int GCNHazardRecognizer::checkMAILdStHazards(MachineInstr *MI) {
-  // On gfx90a+ releveant hazards are checked in checkMAIVALUHazards()
+  // On gfx90a+ relevant hazards are checked in checkMAIVALUHazards()
   if (!ST.hasMAIInsts() || ST.hasGFX90AInsts())
     return 0;
 
diff --git a/llvm/lib/Target/AMDGPU/GCNIterativeScheduler.cpp b/llvm/lib/Target/AMDGPU/GCNIterativeScheduler.cpp
--- a/llvm/lib/Target/AMDGPU/GCNIterativeScheduler.cpp
+++ b/llvm/lib/Target/AMDGPU/GCNIterativeScheduler.cpp
@@ -188,7 +188,7 @@
                printRegion(dbgs(), Rgn.Begin, Rgn.End, Sch.LIS, 2));
     Sch.BaseClass::schedule();
 
-    // Unfortunatelly placeDebugValues incorrectly modifies RegionEnd, restore
+    // Unfortunately placeDebugValues incorrectly modifies RegionEnd, restore
     Sch.RegionEnd = Rgn.End;
     //assert(Rgn.End == Sch.RegionEnd);
     Rgn.Begin = Sch.RegionBegin;
@@ -280,7 +280,7 @@
   return RPTracker.moveMaxPressure();
 }
 
-void GCNIterativeScheduler::enterRegion(MachineBasicBlock *BB, // overriden
+void GCNIterativeScheduler::enterRegion(MachineBasicBlock *BB, // overridden
                                         MachineBasicBlock::iterator Begin,
                                         MachineBasicBlock::iterator End,
                                         unsigned NumRegionInstrs) {
@@ -293,7 +293,7 @@
   }
 }
 
-void GCNIterativeScheduler::schedule() { // overriden
+void GCNIterativeScheduler::schedule() { // overridden
   // do nothing
   LLVM_DEBUG(printLivenessInfo(dbgs(), RegionBegin, RegionEnd, LIS);
              if (!Regions.empty() && Regions.back()->Begin == RegionBegin) {
@@ -304,7 +304,7 @@
              << '\n';);
 }
 
-void GCNIterativeScheduler::finalizeSchedule() { // overriden
+void GCNIterativeScheduler::finalizeSchedule() { // overridden
   if (Regions.empty())
     return;
   switch (Strategy) {
@@ -391,8 +391,8 @@
   // and already interleaved with debug values
   if (!std::is_same<decltype(*Schedule.begin()), MachineInstr*>::value) {
     placeDebugValues();
-    // Unfortunatelly placeDebugValues incorrectly modifies RegionEnd, restore
-    //assert(R.End == RegionEnd);
+    // Unfortunately placeDebugValues incorrectly modifies RegionEnd, restore
+    // assert(R.End == RegionEnd);
     RegionEnd = R.End;
   }
 
diff --git a/llvm/lib/Target/AMDGPU/GCNMinRegStrategy.cpp b/llvm/lib/Target/AMDGPU/GCNMinRegStrategy.cpp
--- a/llvm/lib/Target/AMDGPU/GCNMinRegStrategy.cpp
+++ b/llvm/lib/Target/AMDGPU/GCNMinRegStrategy.cpp
@@ -7,7 +7,7 @@
 //===----------------------------------------------------------------------===//
 ///
 /// \file
-/// This file defines and imlements the class GCNMinRegScheduler, which
+/// This file defines and implements the class GCNMinRegScheduler, which
 /// implements an experimental, simple scheduler whose main goal is to learn
 /// ways about consuming less possible registers for a region.
 ///
diff --git a/llvm/lib/Target/AMDGPU/GCNPreRAOptimizations.cpp b/llvm/lib/Target/AMDGPU/GCNPreRAOptimizations.cpp
--- a/llvm/lib/Target/AMDGPU/GCNPreRAOptimizations.cpp
+++ b/llvm/lib/Target/AMDGPU/GCNPreRAOptimizations.cpp
@@ -7,7 +7,7 @@
 //===----------------------------------------------------------------------===//
 //
 /// \file
-/// This pass combines split register tuple initialization into a single psuedo:
+/// This pass combines split register tuple initialization into a single pseudo:
 ///
 ///   undef %0.sub1:sreg_64 = S_MOV_B32 1
 ///   %0.sub0:sreg_64 = S_MOV_B32 2
diff --git a/llvm/lib/Target/AMDGPU/GCNSchedStrategy.cpp b/llvm/lib/Target/AMDGPU/GCNSchedStrategy.cpp
--- a/llvm/lib/Target/AMDGPU/GCNSchedStrategy.cpp
+++ b/llvm/lib/Target/AMDGPU/GCNSchedStrategy.cpp
@@ -121,7 +121,7 @@
 
   // Register pressure is considered 'CRITICAL' if it is approaching a value
   // that would reduce the wave occupancy for the execution unit.  When
-  // register pressure is 'CRITICAL', increading SGPR and VGPR pressure both
+  // register pressure is 'CRITICAL', increasing SGPR and VGPR pressure both
   // has the same cost, so we don't need to prefer one over the other.
 
   int SGPRDelta = NewSGPRPressure - SGPRCriticalLimit;
diff --git a/llvm/lib/Target/AMDGPU/MCA/AMDGPUCustomBehaviour.h b/llvm/lib/Target/AMDGPU/MCA/AMDGPUCustomBehaviour.h
--- a/llvm/lib/Target/AMDGPU/MCA/AMDGPUCustomBehaviour.h
+++ b/llvm/lib/Target/AMDGPU/MCA/AMDGPUCustomBehaviour.h
@@ -60,7 +60,7 @@
   /// where we setup the InstrWaitCntInfo vector.
   /// The core logic for determining which CNTs an instruction
   /// interacts with is taken from SIInsertWaitcnts::updateEventWaitcntAfter().
-  /// Unfortunately, some of the logic from that function is not avalable to us
+  /// Unfortunately, some of the logic from that function is not available to us
   /// in this scope so we conservatively end up assuming that some
   /// instructions interact with more CNTs than they do in reality.
   void generateWaitCntInfo();
diff --git a/llvm/lib/Target/AMDGPU/MIMGInstructions.td b/llvm/lib/Target/AMDGPU/MIMGInstructions.td
--- a/llvm/lib/Target/AMDGPU/MIMGInstructions.td
+++ b/llvm/lib/Target/AMDGPU/MIMGInstructions.td
@@ -1108,7 +1108,7 @@
   let PrimaryKeyEarlyOut = 1;
 }
 
-def getImageDimInstrinsicByBaseOpcode : SearchIndex {
+def getImageDimIntrinsicByBaseOpcode : SearchIndex {
   let Table = ImageDimIntrinsicTable;
   let Key = ["BaseOpcode", "Dim"];
 }
diff --git a/llvm/lib/Target/AMDGPU/R600ISelLowering.cpp b/llvm/lib/Target/AMDGPU/R600ISelLowering.cpp
--- a/llvm/lib/Target/AMDGPU/R600ISelLowering.cpp
+++ b/llvm/lib/Target/AMDGPU/R600ISelLowering.cpp
@@ -926,7 +926,7 @@
       std::swap(LHS, RHS);
       CC = DAG.getCondCode(CCSwapped);
     } else {
-      // Try inverting the conditon and then swapping the operands
+      // Try inverting the condition and then swapping the operands
       ISD::CondCode CCInv = ISD::getSetCCInverse(CCOpcode, CompareVT);
       CCSwapped = ISD::getSetCCSwappedOperands(CCInv);
       if (isCondCodeLegal(CCSwapped, CompareVT.getSimpleVT())) {
diff --git a/llvm/lib/Target/AMDGPU/R600InstrInfo.h b/llvm/lib/Target/AMDGPU/R600InstrInfo.h
--- a/llvm/lib/Target/AMDGPU/R600InstrInfo.h
+++ b/llvm/lib/Target/AMDGPU/R600InstrInfo.h
@@ -175,7 +175,7 @@
                         int *BytesAdded = nullptr) const override;
 
   unsigned removeBranch(MachineBasicBlock &MBB,
-                        int *BytesRemvoed = nullptr) const override;
+                        int *BytesRemoved = nullptr) const override;
 
   bool isPredicated(const MachineInstr &MI) const override;
 
diff --git a/llvm/lib/Target/AMDGPU/R600Instructions.td b/llvm/lib/Target/AMDGPU/R600Instructions.td
--- a/llvm/lib/Target/AMDGPU/R600Instructions.td
+++ b/llvm/lib/Target/AMDGPU/R600Instructions.td
@@ -1346,7 +1346,7 @@
 
 
 //===----------------------------------------------------------------------===//
-// Regist loads and stores - for indirect addressing
+// Register loads and stores - for indirect addressing
 //===----------------------------------------------------------------------===//
 
 let Namespace = "R600" in {
diff --git a/llvm/lib/Target/AMDGPU/R600MachineScheduler.h b/llvm/lib/Target/AMDGPU/R600MachineScheduler.h
--- a/llvm/lib/Target/AMDGPU/R600MachineScheduler.h
+++ b/llvm/lib/Target/AMDGPU/R600MachineScheduler.h
@@ -63,7 +63,7 @@
 
   int InstKindLimit[IDLast];
 
-  int OccupedSlotsMask;
+  int OccupiedSlotsMask;
 
 public:
   R600SchedStrategy() = default;
diff --git a/llvm/lib/Target/AMDGPU/R600MachineScheduler.cpp b/llvm/lib/Target/AMDGPU/R600MachineScheduler.cpp
--- a/llvm/lib/Target/AMDGPU/R600MachineScheduler.cpp
+++ b/llvm/lib/Target/AMDGPU/R600MachineScheduler.cpp
@@ -29,7 +29,7 @@
   MRI = &DAG->MRI;
   CurInstKind = IDOther;
   CurEmitted = 0;
-  OccupedSlotsMask = 31;
+  OccupiedSlotsMask = 31;
   InstKindLimit[IDAlu] = TII->getMaxAlusPerClause();
   InstKindLimit[IDOther] = 32;
   InstKindLimit[IDFetch] = ST.getTexVTXClauseSize();
@@ -138,7 +138,7 @@
   if (NextInstKind != CurInstKind) {
     LLVM_DEBUG(dbgs() << "Instruction Type Switch\n");
     if (NextInstKind != IDAlu)
-      OccupedSlotsMask |= 31;
+      OccupiedSlotsMask |= 31;
     CurEmitted = 0;
     CurInstKind = NextInstKind;
   }
@@ -339,10 +339,10 @@
 
 void R600SchedStrategy::PrepareNextSlot() {
   LLVM_DEBUG(dbgs() << "New Slot\n");
-  assert (OccupedSlotsMask && "Slot wasn't filled");
-  OccupedSlotsMask = 0;
-//  if (HwGen == AMDGPUSubtarget::NORTHERN_ISLANDS)
-//    OccupedSlotsMask |= 16;
+  assert(OccupiedSlotsMask && "Slot wasn't filled");
+  OccupiedSlotsMask = 0;
+  //  if (HwGen == AMDGPUSubtarget::NORTHERN_ISLANDS)
+  //    OccupiedSlotsMask |= 16;
   InstructionsGroupCandidate.clear();
   LoadAlu();
 }
@@ -400,41 +400,41 @@
 
 SUnit* R600SchedStrategy::pickAlu() {
   while (AvailablesAluCount() || !Pending[IDAlu].empty()) {
-    if (!OccupedSlotsMask) {
+    if (!OccupiedSlotsMask) {
       // Bottom up scheduling : predX must comes first
       if (!AvailableAlus[AluPredX].empty()) {
-        OccupedSlotsMask |= 31;
+        OccupiedSlotsMask |= 31;
         return PopInst(AvailableAlus[AluPredX], false);
       }
       // Flush physical reg copies (RA will discard them)
       if (!AvailableAlus[AluDiscarded].empty()) {
-        OccupedSlotsMask |= 31;
+        OccupiedSlotsMask |= 31;
         return PopInst(AvailableAlus[AluDiscarded], false);
       }
       // If there is a T_XYZW alu available, use it
       if (!AvailableAlus[AluT_XYZW].empty()) {
-        OccupedSlotsMask |= 15;
+        OccupiedSlotsMask |= 15;
         return PopInst(AvailableAlus[AluT_XYZW], false);
       }
     }
-    bool TransSlotOccuped = OccupedSlotsMask & 16;
-    if (!TransSlotOccuped && VLIW5) {
+    bool TransSlotOccupied = OccupiedSlotsMask & 16;
+    if (!TransSlotOccupied && VLIW5) {
       if (!AvailableAlus[AluTrans].empty()) {
-        OccupedSlotsMask |= 16;
+        OccupiedSlotsMask |= 16;
         return PopInst(AvailableAlus[AluTrans], false);
       }
       SUnit *SU = AttemptFillSlot(3, true);
       if (SU) {
-        OccupedSlotsMask |= 16;
+        OccupiedSlotsMask |= 16;
         return SU;
       }
     }
     for (int Chan = 3; Chan > -1; --Chan) {
-      bool isOccupied = OccupedSlotsMask & (1 << Chan);
+      bool isOccupied = OccupiedSlotsMask & (1 << Chan);
       if (!isOccupied) {
         SUnit *SU = AttemptFillSlot(Chan, false);
         if (SU) {
-          OccupedSlotsMask |= (1 << Chan);
+          OccupiedSlotsMask |= (1 << Chan);
           InstructionsGroupCandidate.push_back(SU->getInstr());
           return SU;
         }
diff --git a/llvm/lib/Target/AMDGPU/SIFoldOperands.cpp b/llvm/lib/Target/AMDGPU/SIFoldOperands.cpp
--- a/llvm/lib/Target/AMDGPU/SIFoldOperands.cpp
+++ b/llvm/lib/Target/AMDGPU/SIFoldOperands.cpp
@@ -228,7 +228,7 @@
       MachineOperand &Mod = MI->getOperand(ModIdx);
       unsigned Val = Mod.getImm();
       if (!(Val & SISrcMods::OP_SEL_0) && (Val & SISrcMods::OP_SEL_1)) {
-        // Only apply the following transformation if that operand requries
+        // Only apply the following transformation if that operand requires
         // a packed immediate.
         switch (TII.get(Opcode).OpInfo[OpNo].OperandType) {
         case AMDGPU::OPERAND_REG_IMM_V2FP16:
@@ -688,7 +688,7 @@
 
     // Don't fold into a copy to a physical register with the same class. Doing
     // so would interfere with the register coalescer's logic which would avoid
-    // redundant initalizations.
+    // redundant initializations.
     if (DestReg.isPhysical() && SrcRC->contains(DestReg))
       return;
 
@@ -902,7 +902,7 @@
     tryAddToFoldList(FoldList, UseMI, UseOpIdx, &OpToFold, TII);
 
     // FIXME: We could try to change the instruction from 64-bit to 32-bit
-    // to enable more folding opportunites.  The shrink operands pass
+    // to enable more folding opportunities.  The shrink operands pass
     // already does this.
     return;
   }
diff --git a/llvm/lib/Target/AMDGPU/SIInsertWaitcnts.cpp b/llvm/lib/Target/AMDGPU/SIInsertWaitcnts.cpp
--- a/llvm/lib/Target/AMDGPU/SIInsertWaitcnts.cpp
+++ b/llvm/lib/Target/AMDGPU/SIInsertWaitcnts.cpp
@@ -73,7 +73,7 @@
 
 // Class of object that encapsulates latest instruction counter score
 // associated with the operand.  Used for determining whether
-// s_waitcnt instruction needs to be emited.
+// s_waitcnt instruction needs to be emitted.
 
 #define CNT_MASK(t) (1u << (t))
 
diff --git a/llvm/lib/Target/AMDGPU/SILoadStoreOptimizer.cpp b/llvm/lib/Target/AMDGPU/SILoadStoreOptimizer.cpp
--- a/llvm/lib/Target/AMDGPU/SILoadStoreOptimizer.cpp
+++ b/llvm/lib/Target/AMDGPU/SILoadStoreOptimizer.cpp
@@ -146,7 +146,7 @@
         if (!AddrOp->isReg())
           return false;
 
-        // TODO: We should be able to merge physical reg addreses.
+        // TODO: We should be able to merge physical reg addresses.
         if (AddrOp->getReg().isPhysical())
           return false;
 
@@ -652,7 +652,7 @@
 }
 
 // This function assumes that \p A and \p B have are identical except for
-// size and offset, and they referecne adjacent memory.
+// size and offset, and they reference adjacent memory.
 static MachineMemOperand *combineKnownAdjacentMMOs(MachineFunction &MF,
                                                    const MachineMemOperand *A,
                                                    const MachineMemOperand *B) {
diff --git a/llvm/lib/Target/AMDGPU/SILowerControlFlow.cpp b/llvm/lib/Target/AMDGPU/SILowerControlFlow.cpp
--- a/llvm/lib/Target/AMDGPU/SILowerControlFlow.cpp
+++ b/llvm/lib/Target/AMDGPU/SILowerControlFlow.cpp
@@ -13,7 +13,7 @@
 /// All control flow is handled using predicated instructions and
 /// a predicate stack.  Each Scalar ALU controls the operations of 64 Vector
 /// ALUs.  The Scalar ALU can update the predicate for any of the Vector ALUs
-/// by writting to the 64-bit EXEC register (each bit corresponds to a
+/// by writing to the 64-bit EXEC register (each bit corresponds to a
 /// single vector ALU).  Typically, for predicates, a vector ALU will write
 /// to its bit of the VCC register (like EXEC VCC is 64-bits, one for each
 /// Vector ALU) and then the ScalarALU will AND the VCC register with the
@@ -38,7 +38,8 @@
 /// %vgpr0 = V_ADD_F32 %vgpr0, %vgpr0 // Do the IF block of the branch
 ///
 /// label0:
-/// %sgpr0 = S_OR_SAVEEXEC_B64 %sgpr0  // Restore the exec mask for the Then block
+/// %sgpr0 = S_OR_SAVEEXEC_B64 %sgpr0  // Restore the exec mask for the Then
+///                                    // block
 /// %exec = S_XOR_B64 %sgpr0, %exec    // Update the exec mask
 /// S_BRANCH_EXECZ label1              // Use our branch optimization
 ///                                    // instruction again.
diff --git a/llvm/lib/Target/AMDGPU/SILowerSGPRSpills.cpp b/llvm/lib/Target/AMDGPU/SILowerSGPRSpills.cpp
--- a/llvm/lib/Target/AMDGPU/SILowerSGPRSpills.cpp
+++ b/llvm/lib/Target/AMDGPU/SILowerSGPRSpills.cpp
@@ -368,7 +368,7 @@
     }
 
     // All those frame indices which are dead by now should be removed from the
-    // function frame. Othewise, there is a side effect such as re-mapping of
+    // function frame. Otherwise, there is a side effect such as re-mapping of
     // free frame index ids by the later pass(es) like "stack slot coloring"
     // which in turn could mess-up with the book keeping of "frame index to VGPR
     // lane".
diff --git a/llvm/lib/Target/AMDGPU/SIMachineScheduler.h b/llvm/lib/Target/AMDGPU/SIMachineScheduler.h
--- a/llvm/lib/Target/AMDGPU/SIMachineScheduler.h
+++ b/llvm/lib/Target/AMDGPU/SIMachineScheduler.h
@@ -72,7 +72,7 @@
   // store the live virtual and real registers.
   // We do care only of SGPR32 and VGPR32 and do track only virtual registers.
   // Pressure of additional registers required inside the block.
-  std::vector<unsigned> InternalAdditionnalPressure;
+  std::vector<unsigned> InternalAdditionalPressure;
   // Pressure of input and output registers
   std::vector<unsigned> LiveInPressure;
   std::vector<unsigned> LiveOutPressure;
@@ -153,8 +153,8 @@
 
   // Needs the block to be scheduled inside
   // TODO: find a way to compute it.
-  std::vector<unsigned> &getInternalAdditionnalRegUsage() {
-    return InternalAdditionnalPressure;
+  std::vector<unsigned> &getInternalAdditionalRegUsage() {
+    return InternalAdditionalPressure;
   }
 
   std::set<unsigned> &getInRegs() { return LiveInRegs; }
diff --git a/llvm/lib/Target/AMDGPU/SIMachineScheduler.cpp b/llvm/lib/Target/AMDGPU/SIMachineScheduler.cpp
--- a/llvm/lib/Target/AMDGPU/SIMachineScheduler.cpp
+++ b/llvm/lib/Target/AMDGPU/SIMachineScheduler.cpp
@@ -403,7 +403,7 @@
   }
 
   // TODO: compute InternalAdditionnalPressure.
-  InternalAdditionnalPressure.resize(TopPressure.MaxSetPressure.size());
+  InternalAdditionalPressure.resize(TopPressure.MaxSetPressure.size());
 
   // Check everything is right.
 #ifndef NDEBUG
diff --git a/llvm/lib/Target/AMDGPU/SIModeRegister.cpp b/llvm/lib/Target/AMDGPU/SIModeRegister.cpp
--- a/llvm/lib/Target/AMDGPU/SIModeRegister.cpp
+++ b/llvm/lib/Target/AMDGPU/SIModeRegister.cpp
@@ -225,7 +225,7 @@
   // RequirePending is used to indicate whether we are collecting the initial
   // requirements for the block, and need to defer the first InsertionPoint to
   // Phase 3. It is set to false once we have set FirstInsertionPoint, or when
-  // we discover an explict setreg that means this block doesn't have any
+  // we discover an explicit setreg that means this block doesn't have any
   // initial requirements.
   bool RequirePending = true;
   Status IPChange;
diff --git a/llvm/lib/Target/AMDGPU/SIOptimizeVGPRLiveRange.cpp b/llvm/lib/Target/AMDGPU/SIOptimizeVGPRLiveRange.cpp
--- a/llvm/lib/Target/AMDGPU/SIOptimizeVGPRLiveRange.cpp
+++ b/llvm/lib/Target/AMDGPU/SIOptimizeVGPRLiveRange.cpp
@@ -11,7 +11,7 @@
 /// structures and waterfall loops.
 ///
 /// When we do structurization, we usually transform an if-else into two
-/// sucessive if-then (with a flow block to do predicate inversion). Consider a
+/// successive if-then (with a flow block to do predicate inversion). Consider a
 /// simple case after structurization: A divergent value %a was defined before
 /// if-else and used in both THEN (use in THEN is optional) and ELSE part:
 ///    bb.if:
diff --git a/llvm/lib/Target/AMDGPU/SIPeepholeSDWA.cpp b/llvm/lib/Target/AMDGPU/SIPeepholeSDWA.cpp
--- a/llvm/lib/Target/AMDGPU/SIPeepholeSDWA.cpp
+++ b/llvm/lib/Target/AMDGPU/SIPeepholeSDWA.cpp
@@ -365,7 +365,7 @@
 
       if (Dst &&
           DstUnused->getImm() == AMDGPU::SDWA::DstUnused::UNUSED_PRESERVE) {
-        // This will work if the tied src is acessing WORD_0, and the dst is
+        // This will work if the tied src is accessing WORD_0, and the dst is
         // writing WORD_1. Modifiers don't matter because all the bits that
         // would be impacted are being overwritten by the dst.
         // Any other case will not work.
diff --git a/llvm/lib/Target/AMDGPU/SIPostRABundler.cpp b/llvm/lib/Target/AMDGPU/SIPostRABundler.cpp
--- a/llvm/lib/Target/AMDGPU/SIPostRABundler.cpp
+++ b/llvm/lib/Target/AMDGPU/SIPostRABundler.cpp
@@ -8,7 +8,7 @@
 //
 /// \file
 /// This pass creates bundles of memory instructions to protect adjacent loads
-/// and stores from beeing rescheduled apart from each other post-RA.
+/// and stores from being rescheduled apart from each other post-RA.
 ///
 //===----------------------------------------------------------------------===//
 
diff --git a/llvm/lib/Target/AMDGPU/SIPreEmitPeephole.cpp b/llvm/lib/Target/AMDGPU/SIPreEmitPeephole.cpp
--- a/llvm/lib/Target/AMDGPU/SIPreEmitPeephole.cpp
+++ b/llvm/lib/Target/AMDGPU/SIPreEmitPeephole.cpp
@@ -174,7 +174,7 @@
     MI.setDesc(TII->get(AMDGPU::S_BRANCH));
   } else if (IsVCCZ && MaskValue == 0) {
     // Will always branch
-    // Remove all succesors shadowed by new unconditional branch
+    // Remove all successors shadowed by new unconditional branch
     MachineBasicBlock *Parent = MI.getParent();
     SmallVector<MachineInstr *, 4> ToRemove;
     bool Found = false;
diff --git a/llvm/lib/Target/AMDGPU/SIRegisterInfo.td b/llvm/lib/Target/AMDGPU/SIRegisterInfo.td
--- a/llvm/lib/Target/AMDGPU/SIRegisterInfo.td
+++ b/llvm/lib/Target/AMDGPU/SIRegisterInfo.td
@@ -834,7 +834,7 @@
 
 // This is not a real register. This is just to have a register to add
 // to VReg_1 that does not alias any real register that would
-// introduce inferred register classess.
+// introduce inferred register classes.
 def ARTIFICIAL_VGPR : SIReg <"invalid vgpr", 0> {
   let isArtificial = 1;
 }
diff --git a/llvm/lib/Target/AMDGPU/SIShrinkInstructions.cpp b/llvm/lib/Target/AMDGPU/SIShrinkInstructions.cpp
--- a/llvm/lib/Target/AMDGPU/SIShrinkInstructions.cpp
+++ b/llvm/lib/Target/AMDGPU/SIShrinkInstructions.cpp
@@ -188,7 +188,7 @@
     return;
 
   // eq/ne is special because the imm16 can be treated as signed or unsigned,
-  // and initially selectd to the unsigned versions.
+  // and initially selected to the unsigned versions.
   if (SOPKOpc == AMDGPU::S_CMPK_EQ_U32 || SOPKOpc == AMDGPU::S_CMPK_LG_U32) {
     bool HasUImm;
     if (isKImmOrKUImmOperand(TII, Src1, HasUImm)) {
diff --git a/llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.h b/llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.h
--- a/llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.h
+++ b/llvm/lib/Target/AMDGPU/Utils/AMDGPUBaseInfo.h
@@ -771,7 +771,7 @@
 /// Is this floating-point operand?
 bool isSISrcFPOperand(const MCInstrDesc &Desc, unsigned OpNo);
 
-/// Does this opearnd support only inlinable literals?
+/// Does this operand support only inlinable literals?
 bool isSISrcInlinableOperand(const MCInstrDesc &Desc, unsigned OpNo);
 
 /// Get the size in bits of a register from the register class \p RC.
diff --git a/llvm/lib/Target/AMDGPU/Utils/AMDGPULDSUtils.h b/llvm/lib/Target/AMDGPU/Utils/AMDGPULDSUtils.h
--- a/llvm/lib/Target/AMDGPU/Utils/AMDGPULDSUtils.h
+++ b/llvm/lib/Target/AMDGPU/Utils/AMDGPULDSUtils.h
@@ -48,7 +48,7 @@
 /// as an use within some instruction (either from kernel or from non-kernel).
 bool hasUserInstruction(const GlobalValue *GV);
 
-/// \returns true if an LDS global requres lowering to a module LDS structure
+/// \returns true if an LDS global requires lowering to a module LDS structure
 /// if \p F is not given. If \p F is given it must be a kernel and function
 /// \returns true if an LDS global is directly used from that kernel and it
 /// is safe to replace its uses with a kernel LDS structure member.