Use DW_FORM_ref_addr for DIE offset in .debug_names

Message ID 20240210004253.2352981-1-tom@tromey.com
State New
Headers
Series Use DW_FORM_ref_addr for DIE offset in .debug_names |

Checks

Context Check Description
linaro-tcwg-bot/tcwg_gdb_build--master-aarch64 success Testing passed
linaro-tcwg-bot/tcwg_gdb_build--master-arm success Testing passed
linaro-tcwg-bot/tcwg_gdb_check--master-aarch64 success Testing passed
linaro-tcwg-bot/tcwg_gdb_check--master-arm fail Testing failed

Commit Message

Tom Tromey Feb. 10, 2024, 12:42 a.m. UTC
  Today I realized that while the .debug_names writer uses DW_FORM_udata
for the DIE offset, DW_FORM_ref_addr would be more appropriate here.
This patch makes this change.

Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=31361
---
 gdb/dwarf2/index-write.c      | 7 +++++--
 gdb/dwarf2/read-debug-names.c | 4 ++++
 2 files changed, 9 insertions(+), 2 deletions(-)
  

Comments

Alexandra Petlanova Hajkova Feb. 12, 2024, 5:19 p.m. UTC | #1
On Sat, Feb 10, 2024 at 1:43 AM Tom Tromey <tom@tromey.com> wrote:

> Today I realized that while the .debug_names writer uses DW_FORM_udata
> for the DIE offset, DW_FORM_ref_addr would be more appropriate here.
> This patch makes this change.
>
> Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=31361
>
> I can confirm this change does not introduce any regressions for ppc64le,
Fedora Rawhide
  
Tom Tromey Feb. 29, 2024, 11:38 p.m. UTC | #2
>>>>> "Tom" == Tom Tromey <tom@tromey.com> writes:

Tom> Today I realized that while the .debug_names writer uses DW_FORM_udata
Tom> for the DIE offset, DW_FORM_ref_addr would be more appropriate here.
Tom> This patch makes this change.

Tom> Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=31361

I'm checking this in.

Tom
  

Patch

diff --git a/gdb/dwarf2/index-write.c b/gdb/dwarf2/index-write.c
index 0123e7f2435..a114001f4ee 100644
--- a/gdb/dwarf2/index-write.c
+++ b/gdb/dwarf2/index-write.c
@@ -761,7 +761,7 @@  class debug_names
 		   : DW_IDX_type_unit);
 		m_abbrev_table.append_unsigned_leb128 (DW_FORM_udata);
 		m_abbrev_table.append_unsigned_leb128 (DW_IDX_die_offset);
-		m_abbrev_table.append_unsigned_leb128 (DW_FORM_udata);
+		m_abbrev_table.append_unsigned_leb128 (DW_FORM_ref_addr);
 		m_abbrev_table.append_unsigned_leb128 (DW_IDX_GNU_language);
 		m_abbrev_table.append_unsigned_leb128 (DW_FORM_udata);
 		if ((entry->flags & IS_STATIC) != 0)
@@ -796,7 +796,10 @@  class debug_names
 	    gdb_assert (it != m_cu_index_htab.cend ());
 	    m_entry_pool.append_unsigned_leb128 (it->second);
 
-	    m_entry_pool.append_unsigned_leb128 (to_underlying (entry->die_offset));
+	    m_entry_pool.append_uint (dwarf5_offset_size (),
+				      m_dwarf5_byte_order,
+				      to_underlying (entry->die_offset));
+
 	    m_entry_pool.append_unsigned_leb128 (entry->per_cu->dw_lang ());
 
 	    if (parent != nullptr)
diff --git a/gdb/dwarf2/read-debug-names.c b/gdb/dwarf2/read-debug-names.c
index e2563e84fcf..0add8040894 100644
--- a/gdb/dwarf2/read-debug-names.c
+++ b/gdb/dwarf2/read-debug-names.c
@@ -155,6 +155,10 @@  mapped_debug_names_reader::scan_one_entry (const char *name,
 	  ull = read_unsigned_leb128 (abfd, entry, &bytes_read);
 	  entry += bytes_read;
 	  break;
+	case DW_FORM_ref_addr:
+	  ull = read_offset (abfd, entry, offset_size);
+	  entry += offset_size;
+	  break;
 	case DW_FORM_ref4:
 	  ull = read_4_bytes (abfd, entry);
 	  entry += 4;