This is the mail archive of the
gdb-prs@sources.redhat.com
mailing list for the GDB project.
Re: gdb/1658: dwarf2read.c line of code wrongly assumes length is 1 byte, read leb128
- From: Daniel Jacobowitz <drow at false dot org>
- To: nobody at sources dot redhat dot com
- Cc: gdb-prs at sources dot redhat dot com,
- Date: 27 May 2004 03:48:00 -0000
- Subject: Re: gdb/1658: dwarf2read.c line of code wrongly assumes length is 1 byte, read leb128
- Reply-to: Daniel Jacobowitz <drow at false dot org>
The following reply was made to PR gdb/1658; it has been noted by GNATS.
From: Daniel Jacobowitz <drow@false.org>
To: coulter@cup.hp.com
Cc: gdb-gnats@sources.redhat.com
Subject: Re: gdb/1658: dwarf2read.c line of code wrongly assumes length is 1 byte, read leb128
Date: Wed, 26 May 2004 23:44:10 -0400
On Wed, May 26, 2004 at 07:38:54PM -0000, coulter@cup.hp.com wrote:
> In dwarf2read.c, there is a line:
> line_ptr += 1; /* ignore length */
> This improperly presumes that the leb128 length will only
> use 1 byte. This can be violated with inline function calls.
With what compiler? Could you show me readelf output for a line table
which has a greater than 127-byte long extended opcode?
I'm guessing this uses DW_LNE_define_file; GCC doesn't generate that,
which is why the error was never noticed.
--
Daniel Jacobowitz