This is the mail archive of the
gdb-prs@sources.redhat.com
mailing list for the GDB project.
gdb/211: Some error messages in buildsym.c don't work cross-wordsize
- To: gdb-gnats at sources dot redhat dot com
- Subject: gdb/211: Some error messages in buildsym.c don't work cross-wordsize
- From: gustav at virtutech dot se
- Date: 11 Sep 2001 13:06:52 -0000
- Cc: mattias at virtutech dot se
- Reply-To: gustav at virtutech dot se
>Number: 211
>Category: gdb
>Synopsis: Some error messages in buildsym.c don't work cross-wordsize
>Confidential: no
>Severity: non-critical
>Priority: medium
>Responsible: unassigned
>State: open
>Class: sw-bug
>Submitter-Id: net
>Arrival-Date: Tue Sep 11 06:08:01 PDT 2001
>Closed-Date:
>Last-Modified:
>Originator: gustav@virtutech.se
>Release: 5.0
>Organization:
>Environment:
Linux 2.4.6 #1 Fri Jul 13 17:45:33 CEST 2001 i686 unknown
>Description:
The anon_block_end_complaint and innerblock_anon_complaint
messages in buildsym.c use %lx to print a target address,
which fails miserably on a 32-bit host with a 64-bit target.
>How-To-Repeat:
Well, it's a bit tricky, since it's only trigged if you
first repair gdb to read 64-bit binaries (and then read
Sun workshop produced 64-bit binaries' debuginfo),
so I have no good testcase, but reading the code makes
it pretty obvious that it won't work.
>Fix:
Use sprintf_vma or similar to produce the addresses in those
messagse.
>Release-Note:
>Audit-Trail:
>Unformatted: