[PEI] Add basic opt-remarks support

Add optimization remarks support to the PrologueEpilogueInserter. For
now, emit the stack size as an analysis remark, but more additions wrt
shrink-wrapping may be added.

https://reviews.llvm.org/D35645

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@308556 91177308-0d34-0410-b5e6-96231b3b80d8

lib/CodeGen/PrologEpilogInserter.cpp

@@ -25,6 +25,7 @@
 #include "llvm/CodeGen/MachineInstr.h"
 #include "llvm/CodeGen/MachineLoopInfo.h"
 #include "llvm/CodeGen/MachineModuleInfo.h"
+#include "llvm/CodeGen/MachineOptimizationRemarkEmitter.h"
 #include "llvm/CodeGen/MachineRegisterInfo.h"
 #include "llvm/CodeGen/Passes.h"
 #include "llvm/CodeGen/RegisterScavenging.h"
@@ -98,6 +99,9 @@ private:
   // FrameIndexVirtualScavenging is used.
   bool FrameIndexEliminationScavenging;
 
+  // Emit remarks.
+  MachineOptimizationRemarkEmitter *ORE = nullptr;
+
   void calculateCallFrameInfo(MachineFunction &Fn);
   void calculateSaveRestoreBlocks(MachineFunction &Fn);
   void doSpillCalleeSavedRegs(MachineFunction &MF);
@@ -122,6 +126,7 @@ INITIALIZE_PASS_BEGIN(PEI, DEBUG_TYPE, "Prologue/Epilogue Insertion", false,
 INITIALIZE_PASS_DEPENDENCY(MachineLoopInfo)
 INITIALIZE_PASS_DEPENDENCY(MachineDominatorTree)
 INITIALIZE_PASS_DEPENDENCY(StackProtector)
+INITIALIZE_PASS_DEPENDENCY(MachineOptimizationRemarkEmitterPass)
 INITIALIZE_PASS_END(PEI, DEBUG_TYPE,
                     "Prologue/Epilogue Insertion & Frame Finalization", false,
                     false)
@@ -138,6 +143,7 @@ void PEI::getAnalysisUsage(AnalysisUsage &AU) const {
   AU.addPreserved<MachineLoopInfo>();
   AU.addPreserved<MachineDominatorTree>();
   AU.addRequired<StackProtector>();
+  AU.addRequired<MachineOptimizationRemarkEmitterPass>();
   MachineFunctionPass::getAnalysisUsage(AU);
 }
 
@@ -171,6 +177,7 @@ bool PEI::runOnMachineFunction(MachineFunction &Fn) {
   FrameIndexVirtualScavenging = TRI->requiresFrameIndexScavenging(Fn);
   FrameIndexEliminationScavenging = (RS && !FrameIndexVirtualScavenging) ||
     TRI->requiresFrameIndexReplacementScavenging(Fn);
+  ORE = &getAnalysis<MachineOptimizationRemarkEmitterPass>().getORE();
 
   // Calculate the MaxCallFrameSize and AdjustsStack variables for the
   // function's frame information. Also eliminates call frame pseudo
@@ -938,6 +945,12 @@ void PEI::calculateFrameObjectOffsets(MachineFunction &Fn) {
   int64_t StackSize = Offset - LocalAreaOffset;
   MFI.setStackSize(StackSize);
   NumBytesStackSpace += StackSize;
+
+  MachineOptimizationRemarkAnalysis R(
+      DEBUG_TYPE, "StackSize", Fn.getFunction()->getSubprogram(), &Fn.front());
+  R << ore::NV("NumStackBytes", static_cast<unsigned>(StackSize))
+    << " stack bytes in function";
+  ORE->emit(R);
 }
 
 /// insertPrologEpilogCode - Scan the function for modified callee saved

test/CodeGen/AArch64/prologue-epilogue-remarks.mir

@@ -0,0 +1,57 @@
+# RUN: llc -mtriple=aarch64-unknown-unknown -run-pass=prologepilog -pass-remarks-output=%t -pass-remarks-analysis=prologepilog -o /dev/null %s 2>&1
+# RUN: cat %t | FileCheck %s
+...
+---
+name:            fun0
+stack:
+  - { id: 0, type: default, offset: 0, size: 8, alignment: 4 }
+# CHECK: --- !Analysis
+# CHECK-NEXT: Pass:            prologepilog
+# CHECK-NEXT: Name:            StackSize
+# CHECK-NEXT: Function:        fun0
+# CHECK-NEXT: Args:
+# CHECK-NEXT:   - NumStackBytes:   '16'
+# CHECK-NEXT:   - String:          ' stack bytes in function'
+# CHECK-NEXT: ...
+constants:
+body:             |
+  bb.0:
+    RET_ReallyLR
+
+...
+---
+name:            fun1
+stack:
+  - { id: 0, type: default, offset: 0, size: 19, alignment: 4 }
+# CHECK: --- !Analysis
+# CHECK-NEXT: Pass:            prologepilog
+# CHECK-NEXT: Name:            StackSize
+# CHECK-NEXT: Function:        fun1
+# CHECK-NEXT: Args:
+# CHECK-NEXT:   - NumStackBytes:   '32'
+# CHECK-NEXT:   - String:          ' stack bytes in function'
+# CHECK-NEXT: ...
+constants:
+body:             |
+  bb.0:
+    RET_ReallyLR
+
+...
+---
+name:            fun2
+stack:
+  - { id: 0, type: default, offset: 0, size: 1024, alignment: 4 }
+# --- !Analysis
+# CHECK: Pass:            prologepilog
+# CHECK-NEXT: Name:            StackSize
+# CHECK-NEXT: Function:        fun2
+# CHECK-NEXT: Args:
+# CHECK-NEXT:   - NumStackBytes:   '1040'
+# CHECK-NEXT:   - String:          ' stack bytes in function'
+# CHECK-NEXT: ...
+constants:
+body:             |
+  bb.0:
+    RET_ReallyLR
+
+...

test/CodeGen/X86/O0-pipeline.ll

@@ -42,6 +42,8 @@
 ; CHECK-NEXT:       Fast Register Allocator
 ; CHECK-NEXT:       Bundle Machine CFG Edges
 ; CHECK-NEXT:       X86 FP Stackifier
+; CHECK-NEXT:       Lazy Machine Block Frequency Analysis
+; CHECK-NEXT:       Machine Optimization Remark Emitter
 ; CHECK-NEXT:       Prologue/Epilogue Insertion & Frame Finalization
 ; CHECK-NEXT:       Post-RA pseudo instruction expansion pass
 ; CHECK-NEXT:       X86 pseudo instruction expansion pass

test/CodeGen/X86/prologue-epilogue-remarks.mir

@@ -0,0 +1,58 @@
+# RUN: llc -mtriple=x86_64-unknown-unknown -run-pass=prologepilog -pass-remarks-output=%t -pass-remarks-analysis=prologepilog -o /dev/null %s 2>&1
+# RUN: cat %t | FileCheck %s
+...
+---
+name:            fun0
+stack:
+  - { id: 0, type: default, offset: 0, size: 8, alignment: 4 }
+# --- !Analysis
+# CHECK: Pass:            prologepilog
+# CHECK-NEXT: Name:            StackSize
+# CHECK-NEXT: Function:        fun0
+# CHECK-NEXT: Args:
+# CHECK-NEXT:   - NumStackBytes:   '8'
+# CHECK-NEXT:   - String:          ' stack bytes in function'
+# CHECK-NEXT: ...
+
+constants:
+body:             |
+  bb.0:
+    RETQ
+
+...
+---
+name:            fun1
+stack:
+  - { id: 0, type: default, offset: 0, size: 19, alignment: 4 }
+# --- !Analysis
+# CHECK: Pass:            prologepilog
+# CHECK-NEXT: Name:            StackSize
+# CHECK-NEXT: Function:        fun1
+# CHECK-NEXT: Args:
+# CHECK-NEXT:   - NumStackBytes:   '20'
+# CHECK-NEXT:   - String:          ' stack bytes in function'
+# CHECK-NEXT: ...
+constants:
+body:             |
+  bb.0:
+    RETQ
+
+...
+---
+name:            fun2
+stack:
+  - { id: 0, type: default, offset: 0, size: 1024, alignment: 8 }
+# --- !Analysis
+# CHECK: Pass:            prologepilog
+# CHECK-NEXT: Name:            StackSize
+# CHECK-NEXT: Function:        fun2
+# CHECK-NEXT: Args:
+# CHECK-NEXT:   - NumStackBytes:   '1024'
+# CHECK-NEXT:   - String:          ' stack bytes in function'
+# CHECK-NEXT: ...
+constants:
+body:             |
+  bb.0:
+    RETQ
+
+...