Add support for parsing the DW_AT_ranges attributes

This enables the DWARF reader to properly parse DW_AT_ranges attributes
in compilation units and functions. Code covered by a function is now
represented by a vector of ranges instead of a single contiguous range
and DW_AT_ranges entries are used to populate it. All the code and tests
that assumed functions to be contiguous entities has been updated to
reflect the change. DW_AT_ranges attributes found in compilation units
are parsed but no data is generated for them as it is not currently needed.

BUG=754

Change-Id: I310391b525aaba0dd329f1e3187486f2e0c6d442
Reviewed-on: https://chromium-review.googlesource.com/1124721
Reviewed-by: Ted Mielczarek <ted.mielczarek@gmail.com>

4 files changed, 70 insertions, 0 deletions

diff --git a/src/common/dwarf/dwarf2reader.cc b/src/common/dwarf/dwarf2reader.cc
index f2dc4afe..27f3a83e 100644
--- a/src/common/dwarf/dwarf2reader.cc
+++ b/src/common/dwarf/dwarf2reader.cc
@@ -1247,6 +1247,41 @@ void LineInfo::ReadLines() {
   after_header_ = lengthstart + header_.total_length;
 }
 
+RangeListReader::RangeListReader(const uint8_t *buffer, uint64 size,
+                                 ByteReader *reader, RangeListHandler *handler)
+    : buffer_(buffer), size_(size), reader_(reader), handler_(handler) { }
+
+bool RangeListReader::ReadRangeList(uint64 offset) {
+  const uint64 max_address =
+    (reader_->AddressSize() == 4) ? 0xffffffffUL
+                                  : 0xffffffffffffffffULL;
+  const uint64 entry_size = reader_->AddressSize() * 2;
+  bool list_end = false;
+
+  do {
+    if (offset > size_ - entry_size) {
+      return false; // Invalid range detected
+    }
+
+    uint64 start_address = reader_->ReadAddress(buffer_ + offset);
+    uint64 end_address =
+      reader_->ReadAddress(buffer_ + offset + reader_->AddressSize());
+
+    if (start_address == max_address) { // Base address selection
+      handler_->SetBaseAddress(end_address);
+    } else if (start_address == 0 && end_address == 0) { // End-of-list
+      handler_->Finish();
+      list_end = true;
+    } else { // Add a range entry
+      handler_->AddRange(start_address, end_address);
+    }
+
+    offset += entry_size;
+  } while (!list_end);
+
+  return true;
+}
+
 // A DWARF rule for recovering the address or value of a register, or
 // computing the canonical frame address. There is one subclass of this for
 // each '*Rule' member function in CallFrameInfo::Handler.
diff --git a/src/common/dwarf/dwarf2reader.h b/src/common/dwarf/dwarf2reader.h
index 5d2d7f60..cf3ba3cd 100644
--- a/src/common/dwarf/dwarf2reader.h
+++ b/src/common/dwarf/dwarf2reader.h
@@ -187,6 +187,36 @@ class LineInfoHandler {
                        uint32 file_num, uint32 line_num, uint32 column_num) { }
 };
 
+class RangeListHandler {
+ public:
+  RangeListHandler() { }
+
+  virtual ~RangeListHandler() { }
+
+  // Add a range.
+  virtual void AddRange(uint64 begin, uint64 end) { };
+
+  // A new base address must be set for computing the ranges' addresses.
+  virtual void SetBaseAddress(uint64 base_address) { };
+
+  // Finish processing the range list.
+  virtual void Finish() { };
+};
+
+class RangeListReader {
+ public:
+  RangeListReader(const uint8_t *buffer, uint64 size, ByteReader *reader,
+                  RangeListHandler *handler);
+
+  bool ReadRangeList(uint64 offset);
+
+ private:
+  const uint8_t *buffer_;
+  uint64 size_;
+  ByteReader* reader_;
+  RangeListHandler *handler_;
+};
+
 // This class is the main interface between the reader and the
 // client.  The virtual functions inside this get called for
 // interesting events that happen during DWARF2 reading.
diff --git a/src/common/dwarf/functioninfo.cc b/src/common/dwarf/functioninfo.cc
index 55a255ed..a0def62e 100644
--- a/src/common/dwarf/functioninfo.cc
+++ b/src/common/dwarf/functioninfo.cc
@@ -184,6 +184,9 @@ void CUFunctionInfoHandler::ProcessAttributeUnsigned(uint64 offset,
       case DW_AT_decl_file:
         current_function_info_->file = files_->at(data).name;
         break;
+      case DW_AT_ranges:
+        current_function_info->ranges = data;
+        break;
       default:
         break;
     }
diff --git a/src/common/dwarf/functioninfo.h b/src/common/dwarf/functioninfo.h
index 0b08a5fc..9efae6d4 100644
--- a/src/common/dwarf/functioninfo.h
+++ b/src/common/dwarf/functioninfo.h
@@ -58,6 +58,8 @@ struct FunctionInfo {
   uint64 lowpc;
   // End address for this function.
   uint64 highpc;
+  // Ranges offset
+  uint64 ranges;
 };
 
 struct SourceFileInfo {