[PATCH] Fix end of CU calculation in cooked_indexer::index_dies

[PATCH] Fix end of CU calculation in cooked_indexer::index_dies

[Tom Tromey]

cooked_indexer::index_dies incorrect computes the end of the current
CU in the .debug_info.  This isn't readily testable without writing
intentionally corrupt DWARF, but it's apparent through observation: it
is currently based on 'info_ptr', which does not always point to the
start of the CU.  This patch fixes the expression.  Tested on x86-64
Fedora 34.
---
 gdb/dwarf2/read.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/gdb/dwarf2/read.c b/gdb/dwarf2/read.c
index 696277d931a..403f00c3b95 100644
--- a/gdb/dwarf2/read.c
+++ b/gdb/dwarf2/read.c
@@ -18276,7 +18276,9 @@ cooked_indexer::index_dies (cutu_reader *reader,
 			    const cooked_index_entry *parent_entry,
 			    bool fully)
 {
-  const gdb_byte *end_ptr = info_ptr + reader->cu->header.get_length ();
+  const gdb_byte *end_ptr = (reader->buffer
+			     + to_underlying (reader->cu->header.sect_off)
+			     + reader->cu->header.get_length ());
 
   while (info_ptr < end_ptr)
     {
-- 
2.34.1

Re: [PATCH] Fix end of CU calculation in cooked_indexer::index_dies

[Tom Tromey]

>>>>> "Tom" == Tom Tromey <tom@tromey.com> writes:

Tom> cooked_indexer::index_dies incorrect computes the end of the current
Tom> CU in the .debug_info.  This isn't readily testable without writing
Tom> intentionally corrupt DWARF, but it's apparent through observation: it
Tom> is currently based on 'info_ptr', which does not always point to the
Tom> start of the CU.  This patch fixes the expression.  Tested on x86-64
Tom> Fedora 34.

I'm checking this in.

Tom