Use llvm for dumping DWARF expressions

[labath]

Summary

It uses the new ability of ABI plugins to vend llvm::MCRegisterInfo
structs (which is what is needed to turn dwarf register numbers into
strings).

[JDevlieghere]

This is great!

[aprantl]

Awesome!

[clayborg]

nice!

include/lldb/Utility/DataExtractor.h

 //===-- DataExtractor.h -----------------------------------------*- C++ -*-===//
 //
 // Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
 // See https://llvm.org/LICENSE.txt for license information.
 // SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
 //
 //===----------------------------------------------------------------------===//
 
 #ifndef LLDB_UTILITY_DATAEXTRACTOR_H
 #define LLDB_UTILITY_DATAEXTRACTOR_H
 
 #include "lldb/lldb-defines.h"
 #include "lldb/lldb-enumerations.h"
 #include "lldb/lldb-forward.h"
 #include "lldb/lldb-types.h"
 #include "llvm/ADT/ArrayRef.h"
+#include "llvm/Support/DataExtractor.h"
 
 #include <cassert>
 #include <stdint.h>
 #include <string.h>
 
 namespace lldb_private {
 class Log;
 class Stream;
#endif
   }
 
   void Checksum(llvm::SmallVectorImpl<uint8_t> &dest, uint64_t max_data = 0);
 
   llvm::ArrayRef<uint8_t> GetData() const {
     return {GetDataStart(), size_t(GetByteSize())};
   }
 
+  llvm::DataExtractor GetAsLLVM() const {
+    return {GetData(), GetByteOrder() == lldb::eByteOrderLittle,
+            uint8_t(GetAddressByteSize())};
+  }
+
 protected:
   // Member variables
   const uint8_t *m_start; ///< A pointer to the first byte of data.
   const uint8_t
       *m_end; ///< A pointer to the byte that is past the end of the data.
   lldb::ByteOrder
       m_byte_order;     ///< The byte order of the data we are extracting from.
   uint32_t m_addr_size; ///< The address size to use when extracting pointers or

lit/SymbolFile/DWARF/debug_loc.s

 # Test debug_loc parsing, including the cases of invalid input. The exact
 # behavior in the invalid cases is not particularly important, but it should be
 # "reasonable".
 
 # REQUIRES: x86
 
 # RUN: llvm-mc -triple=x86_64-pc-linux -filetype=obj %s > %t
 # RUN: %lldb %t -o "image lookup -v -a 0" -o "image lookup -v -a 2" -o exit \
 # RUN:   | FileCheck %s
 
 # CHECK-LABEL: image lookup -v -a 0
-# CHECK: Variable: {{.*}}, name = "x0", type = "int", location = rdi,
+# CHECK: Variable: {{.*}}, name = "x0", type = "int", location = DW_OP_reg5 RDI,
 # CHECK: Variable: {{.*}}, name = "x1", type = "int", location = ,
 # CHECK: Variable: {{.*}}, name = "x2", type = "int", location = ,
 
 # CHECK-LABEL: image lookup -v -a 2
-# CHECK: Variable: {{.*}}, name = "x0", type = "int", location = rax,
+# CHECK: Variable: {{.*}}, name = "x0", type = "int", location = DW_OP_reg0 RAX,
 # CHECK: Variable: {{.*}}, name = "x1", type = "int", location = ,
 # CHECK: Variable: {{.*}}, name = "x2", type = "int", location = ,
 
         .type   f,@function
 f:                                      # @f
 .Lfunc_begin0:
         movl    %edi, %eax
 .Ltmp0:

lit/SymbolFile/DWARF/dwarf5_locations.s

 # This tests that lldb is able to process DW_OP_addrx tags introduced in dwarf5.
 
 # REQUIRES: lld, x86
 
 # RUN: llvm-mc -g -dwarf-version=5 -triple x86_64-unknown-linux-gnu %s -filetype=obj > %t.o
 # RUN: ld.lld -m elf_x86_64 %t.o -o %t 
 # RUN: lldb-test symbols %t | FileCheck %s
 
 # CHECK: Variable{0x7fffffff00000011}, name = "color"
-# CHECK-SAME: location = DW_OP_addrx(0x0)
+# CHECK-SAME: location = DW_OP_addrx 0x0
 
         .text
         .section        .debug_str,"MS",@progbits,1
 .Lstr_offsets_base0:
         .asciz  "color"
 
         .section        .debug_str_offsets,"",@progbits
         .long   .Lstr_offsets_base0

source/Expression/DWARFExpression.cpp

void DWARFExpression::UpdateValue(uint64_t const_value,
   m_data.SetByteOrder(endian::InlHostByteOrder());
   m_data.SetAddressByteSize(addr_byte_size);
 }
 
 void DWARFExpression::DumpLocation(Stream *s, lldb::offset_t offset,
                                    lldb::offset_t length,
                                    lldb::DescriptionLevel level,
                                    ABI *abi) const {
-  if (!m_data.ValidOffsetForDataOfSize(offset, length))
-    return;
-  const lldb::offset_t start_offset = offset;
-  const lldb::offset_t end_offset = offset + length;
-
-  // An operation within a DWARF expression may contain a sub-expression. The
-  // motivating example for this is DW_OP_entry_value. Keep track of where each
-  // each sub-expression ends.
-  std::vector<lldb::offset_t> ends_of_subexprs;
-
-  // "Finish" (i.e. print the closing right-parens) for sub-expressions up to
-  // the specified \p op_offset.
-  auto finish_subexpressions_to = [&](const lldb::offset_t op_offset) {
-    while (!ends_of_subexprs.empty() && op_offset >= ends_of_subexprs.back()) {
-      ends_of_subexprs.pop_back();
-      s->Printf(")");
-      if (!ends_of_subexprs.empty())
-        s->Printf(" ");
-    }
-  };
-
-  while (m_data.ValidOffset(offset) && offset < end_offset) {
-    const lldb::offset_t op_offset = offset;
-    const uint8_t op = m_data.GetU8(&offset);
-    finish_subexpressions_to(op_offset);
-
-    switch (level) {
-    default:
-      break;
-
-    case lldb::eDescriptionLevelBrief:
-      if (op_offset > start_offset)
-        s->PutChar(' ');
-      break;
-
-    case lldb::eDescriptionLevelFull:
-    case lldb::eDescriptionLevelVerbose:
-      if (op_offset > start_offset)
-        s->EOL();
-      s->Indent();
-      if (level == lldb::eDescriptionLevelFull)
-        break;
-      // Fall through for verbose and print offset and DW_OP prefix..
-      s->Printf("0x%8.8" PRIx64 ": %s", op_offset,
-                op >= DW_OP_APPLE_uninit ? "DW_OP_APPLE_" : "DW_OP_");
-      break;
-    }
-
-    switch (op) {
-    case DW_OP_addr:
-      *s << "DW_OP_addr(" << m_data.GetAddress(&offset) << ") ";
-      break; // 0x03 1 address
-    case DW_OP_deref:
-      *s << "DW_OP_deref";
-      break; // 0x06
-    case DW_OP_const1u:
-      s->Printf("DW_OP_const1u(0x%2.2x)", m_data.GetU8(&offset));
-      break; // 0x08 1 1-byte constant
-    case DW_OP_const1s:
-      s->Printf("DW_OP_const1s(0x%2.2x)", m_data.GetU8(&offset));
-      break; // 0x09 1 1-byte constant
-    case DW_OP_const2u:
-      s->Printf("DW_OP_const2u(0x%4.4x)", m_data.GetU16(&offset));
-      break; // 0x0a 1 2-byte constant
-    case DW_OP_const2s:
-      s->Printf("DW_OP_const2s(0x%4.4x)", m_data.GetU16(&offset));
-      break; // 0x0b 1 2-byte constant
-    case DW_OP_const4u:
-      s->Printf("DW_OP_const4u(0x%8.8x)", m_data.GetU32(&offset));
-      break; // 0x0c 1 4-byte constant
-    case DW_OP_const4s:
-      s->Printf("DW_OP_const4s(0x%8.8x)", m_data.GetU32(&offset));
-      break; // 0x0d 1 4-byte constant
-    case DW_OP_const8u:
-      s->Printf("DW_OP_const8u(0x%16.16" PRIx64 ")", m_data.GetU64(&offset));
-      break; // 0x0e 1 8-byte constant
-    case DW_OP_const8s:
-      s->Printf("DW_OP_const8s(0x%16.16" PRIx64 ")", m_data.GetU64(&offset));
-      break; // 0x0f 1 8-byte constant
-    case DW_OP_constu:
-      s->Printf("DW_OP_constu(0x%" PRIx64 ")", m_data.GetULEB128(&offset));
-      break; // 0x10 1 ULEB128 constant
-    case DW_OP_consts:
-      s->Printf("DW_OP_consts(0x%" PRId64 ")", m_data.GetSLEB128(&offset));
-      break; // 0x11 1 SLEB128 constant
-    case DW_OP_dup:
-      s->PutCString("DW_OP_dup");
-      break; // 0x12
-    case DW_OP_drop:
-      s->PutCString("DW_OP_drop");
-      break; // 0x13
-    case DW_OP_over:
-      s->PutCString("DW_OP_over");
-      break; // 0x14
-    case DW_OP_pick:
-      s->Printf("DW_OP_pick(0x%2.2x)", m_data.GetU8(&offset));
-      break; // 0x15 1 1-byte stack index
-    case DW_OP_swap:
-      s->PutCString("DW_OP_swap");
-      break; // 0x16
-    case DW_OP_rot:
-      s->PutCString("DW_OP_rot");
-      break; // 0x17
-    case DW_OP_xderef:
-      s->PutCString("DW_OP_xderef");
-      break; // 0x18
-    case DW_OP_abs:
-      s->PutCString("DW_OP_abs");
-      break; // 0x19
-    case DW_OP_and:
-      s->PutCString("DW_OP_and");
-      break; // 0x1a
-    case DW_OP_div:
-      s->PutCString("DW_OP_div");
-      break; // 0x1b
-    case DW_OP_minus:
-      s->PutCString("DW_OP_minus");
-      break; // 0x1c
-    case DW_OP_mod:
-      s->PutCString("DW_OP_mod");
-      break; // 0x1d
-    case DW_OP_mul:
-      s->PutCString("DW_OP_mul");
-      break; // 0x1e
-    case DW_OP_neg:
-      s->PutCString("DW_OP_neg");
-      break; // 0x1f
-    case DW_OP_not:
-      s->PutCString("DW_OP_not");
-      break; // 0x20
-    case DW_OP_or:
-      s->PutCString("DW_OP_or");
-      break; // 0x21
-    case DW_OP_plus:
-      s->PutCString("DW_OP_plus");
-      break;                // 0x22
-    case DW_OP_plus_uconst: // 0x23 1 ULEB128 addend
-      s->Printf("DW_OP_plus_uconst(0x%" PRIx64 ")",
-                m_data.GetULEB128(&offset));
-      break;
-
-    case DW_OP_shl:
-      s->PutCString("DW_OP_shl");
-      break; // 0x24
-    case DW_OP_shr:
-      s->PutCString("DW_OP_shr");
-      break; // 0x25
-    case DW_OP_shra:
-      s->PutCString("DW_OP_shra");
-      break; // 0x26
-    case DW_OP_xor:
-      s->PutCString("DW_OP_xor");
-      break; // 0x27
-    case DW_OP_skip:
-      s->Printf("DW_OP_skip(0x%4.4x)", m_data.GetU16(&offset));
-      break; // 0x2f 1 signed 2-byte constant
-    case DW_OP_bra:
-      s->Printf("DW_OP_bra(0x%4.4x)", m_data.GetU16(&offset));
-      break; // 0x28 1 signed 2-byte constant
-    case DW_OP_eq:
-      s->PutCString("DW_OP_eq");
-      break; // 0x29
-    case DW_OP_ge:
-      s->PutCString("DW_OP_ge");
-      break; // 0x2a
-    case DW_OP_gt:
-      s->PutCString("DW_OP_gt");
-      break; // 0x2b
-    case DW_OP_le:
-      s->PutCString("DW_OP_le");
-      break; // 0x2c
-    case DW_OP_lt:
-      s->PutCString("DW_OP_lt");
-      break; // 0x2d
-    case DW_OP_ne:
-      s->PutCString("DW_OP_ne");
-      break; // 0x2e
-
-    case DW_OP_lit0:  // 0x30
-    case DW_OP_lit1:  // 0x31
-    case DW_OP_lit2:  // 0x32
-    case DW_OP_lit3:  // 0x33
-    case DW_OP_lit4:  // 0x34
-    case DW_OP_lit5:  // 0x35
-    case DW_OP_lit6:  // 0x36
-    case DW_OP_lit7:  // 0x37
-    case DW_OP_lit8:  // 0x38
-    case DW_OP_lit9:  // 0x39
-    case DW_OP_lit10: // 0x3A
-    case DW_OP_lit11: // 0x3B
-    case DW_OP_lit12: // 0x3C
-    case DW_OP_lit13: // 0x3D
-    case DW_OP_lit14: // 0x3E
-    case DW_OP_lit15: // 0x3F
-    case DW_OP_lit16: // 0x40
-    case DW_OP_lit17: // 0x41
-    case DW_OP_lit18: // 0x42
-    case DW_OP_lit19: // 0x43
-    case DW_OP_lit20: // 0x44
-    case DW_OP_lit21: // 0x45
-    case DW_OP_lit22: // 0x46
-    case DW_OP_lit23: // 0x47
-    case DW_OP_lit24: // 0x48
-    case DW_OP_lit25: // 0x49
-    case DW_OP_lit26: // 0x4A
-    case DW_OP_lit27: // 0x4B
-    case DW_OP_lit28: // 0x4C
-    case DW_OP_lit29: // 0x4D
-    case DW_OP_lit30: // 0x4E
-    case DW_OP_lit31:
-      s->Printf("DW_OP_lit%i", op - DW_OP_lit0);
-      break; // 0x4f
-
-    case DW_OP_reg0:  // 0x50
-    case DW_OP_reg1:  // 0x51
-    case DW_OP_reg2:  // 0x52
-    case DW_OP_reg3:  // 0x53
-    case DW_OP_reg4:  // 0x54
-    case DW_OP_reg5:  // 0x55
-    case DW_OP_reg6:  // 0x56
-    case DW_OP_reg7:  // 0x57
-    case DW_OP_reg8:  // 0x58
-    case DW_OP_reg9:  // 0x59
-    case DW_OP_reg10: // 0x5A
-    case DW_OP_reg11: // 0x5B
-    case DW_OP_reg12: // 0x5C
-    case DW_OP_reg13: // 0x5D
-    case DW_OP_reg14: // 0x5E
-    case DW_OP_reg15: // 0x5F
-    case DW_OP_reg16: // 0x60
-    case DW_OP_reg17: // 0x61
-    case DW_OP_reg18: // 0x62
-    case DW_OP_reg19: // 0x63
-    case DW_OP_reg20: // 0x64
-    case DW_OP_reg21: // 0x65
-    case DW_OP_reg22: // 0x66
-    case DW_OP_reg23: // 0x67
-    case DW_OP_reg24: // 0x68
-    case DW_OP_reg25: // 0x69
-    case DW_OP_reg26: // 0x6A
-    case DW_OP_reg27: // 0x6B
-    case DW_OP_reg28: // 0x6C
-    case DW_OP_reg29: // 0x6D
-    case DW_OP_reg30: // 0x6E
-    case DW_OP_reg31: // 0x6F
-    {
-      uint32_t reg_num = op - DW_OP_reg0;
-      if (abi) {
-        RegisterInfo reg_info;
-        if (abi->GetRegisterInfoByKind(m_reg_kind, reg_num, reg_info)) {
-          if (reg_info.name) {
-            s->PutCString(reg_info.name);
-            break;
-          } else if (reg_info.alt_name) {
-            s->PutCString(reg_info.alt_name);
-            break;
-          }
-        }
-      }
-      s->Printf("DW_OP_reg%u", reg_num);
-      break;
-    } break;
-
-    case DW_OP_breg0:
-    case DW_OP_breg1:
-    case DW_OP_breg2:
-    case DW_OP_breg3:
-    case DW_OP_breg4:
-    case DW_OP_breg5:
-    case DW_OP_breg6:
-    case DW_OP_breg7:
-    case DW_OP_breg8:
-    case DW_OP_breg9:
-    case DW_OP_breg10:
-    case DW_OP_breg11:
-    case DW_OP_breg12:
-    case DW_OP_breg13:
-    case DW_OP_breg14:
-    case DW_OP_breg15:
-    case DW_OP_breg16:
-    case DW_OP_breg17:
-    case DW_OP_breg18:
-    case DW_OP_breg19:
-    case DW_OP_breg20:
-    case DW_OP_breg21:
-    case DW_OP_breg22:
-    case DW_OP_breg23:
-    case DW_OP_breg24:
-    case DW_OP_breg25:
-    case DW_OP_breg26:
-    case DW_OP_breg27:
-    case DW_OP_breg28:
-    case DW_OP_breg29:
-    case DW_OP_breg30:
-    case DW_OP_breg31: {
-      uint32_t reg_num = op - DW_OP_breg0;
-      int64_t reg_offset = m_data.GetSLEB128(&offset);
-      if (abi) {
-        RegisterInfo reg_info;
-        if (abi->GetRegisterInfoByKind(m_reg_kind, reg_num, reg_info)) {
-          if (reg_info.name) {
-            s->Printf("[%s%+" PRIi64 "]", reg_info.name, reg_offset);
-            break;
-          } else if (reg_info.alt_name) {
-            s->Printf("[%s%+" PRIi64 "]", reg_info.alt_name, reg_offset);
-            break;
-          }
-        }
-      }
-      s->Printf("DW_OP_breg%i(0x%" PRIx64 ")", reg_num, reg_offset);
-    } break;
-
-    case DW_OP_regx: // 0x90 1 ULEB128 register
-    {
-      uint32_t reg_num = m_data.GetULEB128(&offset);
-      if (abi) {
-        RegisterInfo reg_info;
-        if (abi->GetRegisterInfoByKind(m_reg_kind, reg_num, reg_info)) {
-          if (reg_info.name) {
-            s->PutCString(reg_info.name);
-            break;
-          } else if (reg_info.alt_name) {
-            s->PutCString(reg_info.alt_name);
-            break;
-          }
-        }
-      }
-      s->Printf("DW_OP_regx(%" PRIu32 ")", reg_num);
-      break;
-    } break;
-    case DW_OP_fbreg: // 0x91 1 SLEB128 offset
-      s->Printf("DW_OP_fbreg(%" PRIi64 ")", m_data.GetSLEB128(&offset));
-      break;
-    case DW_OP_bregx: // 0x92 2 ULEB128 register followed by SLEB128 offset
-    {
-      uint32_t reg_num = m_data.GetULEB128(&offset);
-      int64_t reg_offset = m_data.GetSLEB128(&offset);
-      if (abi) {
-        RegisterInfo reg_info;
-        if (abi->GetRegisterInfoByKind(m_reg_kind, reg_num, reg_info)) {
-          if (reg_info.name) {
-            s->Printf("[%s%+" PRIi64 "]", reg_info.name, reg_offset);
-            break;
-          } else if (reg_info.alt_name) {
-            s->Printf("[%s%+" PRIi64 "]", reg_info.alt_name, reg_offset);
-            break;
-          }
-        }
-      }
-      s->Printf("DW_OP_bregx(reg=%" PRIu32 ",offset=%" PRIi64 ")", reg_num,
-                reg_offset);
-    } break;
-    case DW_OP_piece: // 0x93 1 ULEB128 size of piece addressed
-      s->Printf("DW_OP_piece(0x%" PRIx64 ")", m_data.GetULEB128(&offset));
-      break;
-    case DW_OP_deref_size: // 0x94 1 1-byte size of data retrieved
-      s->Printf("DW_OP_deref_size(0x%2.2x)", m_data.GetU8(&offset));
-      break;
-    case DW_OP_xderef_size: // 0x95 1 1-byte size of data retrieved
-      s->Printf("DW_OP_xderef_size(0x%2.2x)", m_data.GetU8(&offset));
-      break;
-    case DW_OP_nop:
-      s->PutCString("DW_OP_nop");
-      break; // 0x96
-    case DW_OP_push_object_address:
-      s->PutCString("DW_OP_push_object_address");
-      break;          // 0x97 DWARF3
-    case DW_OP_call2: // 0x98 DWARF3 1 2-byte offset of DIE
-      s->Printf("DW_OP_call2(0x%4.4x)", m_data.GetU16(&offset));
-      break;
-    case DW_OP_call4: // 0x99 DWARF3 1 4-byte offset of DIE
-      s->Printf("DW_OP_call4(0x%8.8x)", m_data.GetU32(&offset));
-      break;
-    case DW_OP_call_ref: // 0x9a DWARF3 1 4- or 8-byte offset of DIE
-      s->Printf("DW_OP_call_ref(0x%8.8" PRIx64 ")", m_data.GetAddress(&offset));
-      break;
-    case DW_OP_form_tls_address:
-      s->PutCString("DW_OP_form_tls_address"); // 0x9b
-      break;
-    case DW_OP_GNU_addr_index: // 0xfb
-      s->Printf("DW_OP_GNU_addr_index(0x%" PRIx64 ")",
-                m_data.GetULEB128(&offset));
-      break;
-    case DW_OP_addrx:
-      s->Printf("DW_OP_addrx(0x%" PRIx64 ")",
-                m_data.GetULEB128(&offset));
-      break;
-    case DW_OP_GNU_const_index: // 0xfc
-      s->Printf("DW_OP_GNU_const_index(0x%" PRIx64 ")",
-                m_data.GetULEB128(&offset));
-      break;
-    case DW_OP_GNU_push_tls_address:
-      s->PutCString("DW_OP_GNU_push_tls_address"); // 0xe0
-      break;
-    case DW_OP_APPLE_uninit:
-      s->PutCString("DW_OP_APPLE_uninit"); // 0xF0
-      break;
-    case DW_OP_entry_value: {
-      uint32_t subexpr_len = m_data.GetULEB128(&offset);
-      s->PutCString("DW_OP_entry_value(");
-      ends_of_subexprs.push_back(offset + subexpr_len);
-      break;
-    }
-    }
-  }
-
-  finish_subexpressions_to(end_offset);
+  llvm::DWARFExpression(DataExtractor(m_data, offset, length).GetAsLLVM(),
+                        llvm::dwarf::DWARF_VERSION, m_data.GetAddressByteSize())
+      .print(s->AsRawOstream(), abi ? &abi->GetMCRegisterInfo() : nullptr,
+             nullptr);
 }
 
 void DWARFExpression::SetLocationListSlide(addr_t slide) {
   m_loclist_slide = slide;
 }
 
 int DWARFExpression::GetRegisterKind() { return m_reg_kind; }
 

source/Plugins/Process/gdb-remote/ProcessGDBRemote.cpp

 #include "lldb/Utility/StringExtractorGDBRemote.h"
 
 #include "llvm/ADT/ScopeExit.h"
 #include "llvm/ADT/StringSwitch.h"
 #include "llvm/Support/Threading.h"
 #include "llvm/Support/raw_ostream.h"
 
 #define DEBUGSERVER_BASENAME "debugserver"
-using namespace llvm;
 using namespace lldb;
 using namespace lldb_private;
 using namespace lldb_private::process_gdb_remote;
 
 namespace lldb {
 // Provide a function that can easily dump the packet history if we know a
 // ProcessGDBRemote * value (which we can get from logs or from debugging). We
 // need the function in the lldb namespace so it makes it into the final

source/Symbol/ClangASTContext.cpp

 #include "Plugins/SymbolFile/PDB/PDBASTParser.h"
 
 #include <stdio.h>
 
 #include <mutex>
 
 using namespace lldb;
 using namespace lldb_private;
-using namespace llvm;
 using namespace clang;
+using llvm::StringSwitch;
 
 namespace {
 #ifdef LLDB_CONFIGURATION_DEBUG
 static void VerifyDecl(clang::Decl *decl) {
   assert(decl && "VerifyDecl called with nullptr?");
   decl->getAccess();
 }
 #endif

unittests/Utility/ScalarTest.cpp

 #include "lldb/Utility/DataExtractor.h"
 #include "lldb/Utility/Endian.h"
 #include "lldb/Utility/Scalar.h"
 #include "lldb/Utility/Status.h"
 #include "lldb/Utility/StreamString.h"
 #include "llvm/Testing/Support/Error.h"
 
 using namespace lldb_private;
-using namespace llvm;
+using llvm::APInt;
+using llvm::Failed;
+using llvm::Succeeded;
 
 template <typename T>
 bool checkInequality(T c1, T c2) {
   return (Scalar(c1) != Scalar(c2));
 }
 
 template <typename T>
 bool checkEquality(T c1, T c2) {