qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: P J P <ppandit@redhat.com>
To: qemu-devel@nongnu.org
Cc: Gerben van der Lubbe <spoofedexistence@gmail.com>
Subject: [Qemu-devel] [PATCH] Limit memory r/w length to buffer size
Date: Tue, 13 Oct 2015 01:01:09 +0530 (IST)	[thread overview]
Message-ID: <alpine.LFD.2.20.1510130051540.19063@wniryva> (raw)

    Hello,

An OOB r/w access issue was reported by Mr Gerben Lubbe(CC'd here).

The GDB(1) stub protocol supports commands 'm/M' to read & write 'len' bytes 
from/to the stub memory area. In that, the 'len' parameter value supplied by 
the host gdb(1) is not validated against the local buffer size. Which in turn 
could lead to OOB r/w memory access.

Below is a proposed patch to fix this issue.

===
>From 88edb457a66f8ff96209a1603914171eade0658b Mon Sep 17 00:00:00 2001
From: Prasad J Pandit <pjp@fedoraproject.org>
Date: Mon, 12 Oct 2015 22:56:41 +0530
Subject: Limit memory r/w length to buffer size

GDB(1) stub communication protocol supports commands m/M to read
and write 'len' bytes from/to the stub memory area.

    m addr,len        : read 'len' bytes from address 'addr'
    M addr,len:<data> : write 'len' bytes of 'data' to 'addr'

Qemu stub uses automatic buffers of size 'MAX_PACKET_LENGTH=4096'
to process these commands. Limit 'len' parameter value supplied
by the host gdb(1) to the maximum buffer size to avoid any OOB
buffer access.

Reported-by: Gerben van der Lubbe <spoofedexistence@gmail.com>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
---
  gdbstub.c | 2 ++
  1 file changed, 2 insertions(+)

diff --git a/gdbstub.c b/gdbstub.c
index ffe7e6e..39da736 100644
--- a/gdbstub.c
+++ b/gdbstub.c
@@ -956,6 +956,7 @@ static int gdb_handle_packet(GDBState *s, const char *line_buf)
          if (*p == ',')
              p++;
          len = strtoull(p, NULL, 16);
+        len = len > MAX_PACKET_LENGTH ? MAX_PACKET_LENGTH : len;
          if (target_memory_rw_debug(s->g_cpu, addr, mem_buf, len, false) != 0) {
              put_packet (s, "E14");
          } else {
@@ -968,6 +969,7 @@ static int gdb_handle_packet(GDBState *s, const char *line_buf)
          if (*p == ',')
              p++;
          len = strtoull(p, (char **)&p, 16);
+        len = len > MAX_PACKET_LENGTH ? MAX_PACKET_LENGTH : len;
          if (*p == ':')
              p++;
          hextomem(mem_buf, p, len);
-- 
2.4.3
===


Thank you.
--
Prasad J Pandit / Red Hat Product Security Team
47AF CE69 3A90 54AA 9045 1053 DD13 3D32 FE5B 041F

             reply	other threads:[~2015-10-12 19:31 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-10-12 19:31 P J P [this message]
2015-10-13 18:29 ` [Qemu-devel] [PATCH] Limit memory r/w length to buffer size P J P
2015-10-13 18:58 ` Markus Armbruster
2015-10-14  6:27   ` P J P

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=alpine.LFD.2.20.1510130051540.19063@wniryva \
    --to=ppandit@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=spoofedexistence@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).