The offset_size_flag is an DWARF 6 extension described in
https://dwarfstd.org/issues/220708.2.html
While writing tests for eu-dwp some issues with the eu-readelf display
of the .debug_cu_index were found. The offset_size_flag was only used
for the offsets table, not the size (lengths) table. And if the flag
was set only 4 bytes of the offset were read (instead of 8 bytes).
* src/readelf.c (print_cu_index_section): Calculate
lengths_end using off_bytes, use read_8ubyte_unaligned to
read offsets when offset_size_flag is set, calculate prow
using off_bytes, read and print sizes based on off_bytes.
Signed-off-by: Mark Wielaard <[email protected]>
---
src/readelf.c | 29 +++++++++++++++++++----------
1 file changed, 19 insertions(+), 10 deletions(-)
diff --git a/src/readelf.c b/src/readelf.c
index 8d7950d3a6c8..9d3bbbf6f240 100644
--- a/src/readelf.c
+++ b/src/readelf.c
@@ -12290,6 +12290,7 @@ print_cu_index_section (Dwfl_Module *dwflmod
__attribute__ ((unused)),
readp++;
}
+ /* Both offsets and sizes are either 32 or 64 bits (4 or 8 bytes). */
const unsigned char off_bytes = offset_size_flag == 0 ? 4 : 8;
if (unlikely (readp > dataend - 4))
@@ -12325,17 +12326,17 @@ print_cu_index_section (Dwfl_Module *dwflmod
__attribute__ ((unused)),
the hash table (ids of 8 bytes). */
const unsigned char *indices = hash_table + slot_count * 8;
/* Sections used starts after the indices, indices and hash table
- have the same number of slots, indeces are 4 bytes each, */
+ have the same number of slots, indices are 4 bytes each, */
const unsigned char *sections = indices + slot_count * 4;
/* Offset slots for each section follow the one row of sections. */
const unsigned char *offsets = sections + section_count * 4;
/* Size slots for each section follow the offsets (used rows). */
const unsigned char *lengths = (offsets +
unit_count * section_count * off_bytes);
- /* The size table has one row of section_count 4-byte slots per unit,
+ /* The size table has one row of section_count slots per unit,
just like the offset table, so it spans unit_count rows (not one). */
const unsigned char *lengths_end = (lengths +
- unit_count * section_count * 4);
+ unit_count * section_count * off_bytes);
/* Sanity check the above against dataend. */
if ((slot_count > UINT32_MAX / 8)
@@ -12395,7 +12396,7 @@ print_cu_index_section (Dwfl_Module *dwflmod
__attribute__ ((unused)),
else
for (size_t j = 0; j < section_count; j++)
{
- uint64_t off = read_4ubyte_unaligned (dbg, prow + j * 8);
+ uint64_t off = read_8ubyte_unaligned (dbg, prow + j * 8);
fprintf (out, " %6" PRIu64, off);
}
fprintf (out, "\n");
@@ -12433,12 +12434,20 @@ print_cu_index_section (Dwfl_Module *dwflmod
__attribute__ ((unused)),
continue;
}
/* Note row is one based, not zero based. */
- const unsigned char *prow = lengths + (row - 1) * section_count * 4;
- for (size_t j = 0; j < section_count; j++)
- {
- uint32_t off = read_4ubyte_unaligned (dbg, prow + j * 4);
- fprintf (out, " %6" PRIu32, off);
- }
+ const unsigned char *prow = (lengths
+ + (row - 1) * section_count * off_bytes);
+ if (off_bytes == 4)
+ for (size_t j = 0; j < section_count; j++)
+ {
+ uint32_t sz = read_4ubyte_unaligned (dbg, prow + j * 4);
+ fprintf (out, " %6" PRIu32, sz);
+ }
+ else
+ for (size_t j = 0; j < section_count; j++)
+ {
+ uint64_t sz = read_8ubyte_unaligned (dbg, prow + j * 8);
+ fprintf (out, " %6" PRIu64, sz);
+ }
fprintf (out, "\n");
}
}
--
2.54.0
