lkml.org 
[lkml]   [2018]   [Nov]   [12]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
Patch in this message
/
From
Date
Subject[PATCH 3.16 300/366] net: socket: fix potential spectre v1 gadget in socketcall
3.16.61-rc1 review patch.  If anyone has any objections, please let me know.

------------------

From: Jeremy Cline <jcline@redhat.com>

commit c8e8cd579bb4265651df8223730105341e61a2d1 upstream.

'call' is a user-controlled value, so sanitize the array index after the
bounds check to avoid speculating past the bounds of the 'nargs' array.

Found with the help of Smatch:

net/socket.c:2508 __do_sys_socketcall() warn: potential spectre issue
'nargs' [r] (local cap)

Cc: Josh Poimboeuf <jpoimboe@redhat.com>
Signed-off-by: Jeremy Cline <jcline@redhat.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
---
net/socket.c | 2 ++
1 file changed, 2 insertions(+)

--- a/net/socket.c
+++ b/net/socket.c
@@ -89,6 +89,7 @@
#include <linux/magic.h>
#include <linux/slab.h>
#include <linux/xattr.h>
+#include <linux/nospec.h>

#include <asm/uaccess.h>
#include <asm/unistd.h>
@@ -2494,6 +2495,7 @@ SYSCALL_DEFINE2(socketcall, int, call, u

if (call < 1 || call > SYS_SENDMMSG)
return -EINVAL;
+ call = array_index_nospec(call, SYS_SENDMMSG + 1);

len = nargs[call];
if (len > sizeof(a))
\
 
 \ /
  Last update: 2018-11-11 21:28    [W:1.029 / U:0.296 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site