[Hexagon][llvm-objdump] Start a fresh packet at symbol boundaries. (#163466)

Hexagon packets can visually straddle labels when data (e.g. jump
tables) in the text section does not carry end-of-packet bits. In such
cases the next instruction, even at a new symbol, appears to continue
the previous packet.

This patch resets packet state when encountering a new symbol so that
packets at symbol starts are guaranteed to start in their own packet.

Author: quic-areg

llvm/lib/Target/Hexagon/Disassembler/HexagonDisassembler.cpp

@@ -66,6 +66,10 @@ class HexagonDisassembler : public MCDisassembler {
 
   void remapInstruction(MCInst &Instr) const;
 
+  Expected<bool> onSymbolStart(SymbolInfoTy &Symbol, uint64_t &Size,
+                               ArrayRef<uint8_t> Bytes,
+                               uint64_t Address) const override;
+
 private:
   bool makeBundle(ArrayRef<uint8_t> Bytes, uint64_t Address,
                   uint64_t &BytesToSkip, raw_ostream &CS) const;
@@ -567,6 +571,18 @@ DecodeStatus HexagonDisassembler::getSingleInstruction(MCInst &MI, MCInst &MCB,
   return Result;
 }
 
+Expected<bool> HexagonDisassembler::onSymbolStart(SymbolInfoTy &Symbol,
+                                                  uint64_t &Size,
+                                                  ArrayRef<uint8_t> Bytes,
+                                                  uint64_t Address) const {
+  // At the start of a symbol, force a fresh packet by resetting any
+  // in-progress bundle state. This prevents packets from straddling label
+  // boundaries when data (e.g. jump tables) appears in between.
+  Size = 0;
+  resetBundle();
+  return true;
+}
+
 static DecodeStatus DecodeRegisterClass(MCInst &Inst, unsigned RegNo,
                                         ArrayRef<MCPhysReg> Table) {
   if (RegNo < Table.size()) {

llvm/test/tools/llvm-objdump/ELF/Hexagon/packet-reset-on-label.s

@@ -0,0 +1,23 @@
+// RUN: llvm-mc -triple=hexagon -mcpu=hexagonv75 -filetype=obj %s \
+// RUN:   | llvm-objdump -d - \
+// RUN:   | FileCheck %s
+
+foo:
+  { nop }
+  /// a nop without end-of-packet bits set to simulate data that is
+  /// not a proper packet end.
+  .long 0x7f004000
+bar:
+  { nop
+    nop
+  }
+
+// CHECK-LABEL: <foo>:
+// CHECK: { nop }
+// CHECK-NEXT: { nop
+
+/// The instruction starting after <bar> should start in a new packet.
+// CHECK-LABEL: <bar>:
+// CHECK: { nop
+// CHECK-NEXT: nop }
+

llvm/tools/llvm-objdump/llvm-objdump.cpp

@@ -728,11 +728,17 @@ class PrettyPrinter {
     } while (!Comments.empty());
     FOS.flush();
   }
+
+  // Hook invoked when starting to disassemble a symbol at the current position.
+  // Default is no-op.
+  virtual void onSymbolStart() {}
 };
 PrettyPrinter PrettyPrinterInst;
 
 class HexagonPrettyPrinter : public PrettyPrinter {
 public:
+  void onSymbolStart() override { reset(); }
+
   void printLead(ArrayRef<uint8_t> Bytes, uint64_t Address,
                  formatted_raw_ostream &OS) {
     if (LeadingAddr)
@@ -2228,6 +2234,8 @@ disassembleObject(ObjectFile &Obj, const ObjectFile &DbgObj,
         Start += Size;
         break;
       }
+      // Allow targets to reset any per-symbol state.
+      DT->Printer->onSymbolStart();
       formatted_raw_ostream FOS(OS);
       Index = Start;
       if (SectionAddr < StartAddress)