[PATCH] tty: vt_ioctl: fix potential Spectre v1

[PATCH] tty: vt_ioctl: fix potential Spectre v1

[Gustavo A. R. Silva]

vsa.console is indirectly controlled by user-space, hence leading to
a potential exploitation of the Spectre variant 1 vulnerability.

This issue was detected with the help of Smatch:

drivers/tty/vt/vt_ioctl.c:711 vt_ioctl() warn: potential spectre issue
'vc_cons' [r]

Fix this by sanitizing vsa.console before using it to index vc_cons

Notice that given that speculation windows are large, the policy is
to kill the speculation on the first load and not worry if it can be
completed with a dependent load/store [1].

[1] https://marc.info/?l=linux-kernel&m=152449131114778&w=2

Cc: stable@vger.kernel.org
Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>
---
 drivers/tty/vt/vt_ioctl.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/drivers/tty/vt/vt_ioctl.c b/drivers/tty/vt/vt_ioctl.c
index a78ad10..73cdc0d 100644
--- a/drivers/tty/vt/vt_ioctl.c
+++ b/drivers/tty/vt/vt_ioctl.c
@@ -32,6 +32,8 @@
 #include <asm/io.h>
 #include <linux/uaccess.h>
 
+#include <linux/nospec.h>
+
 #include <linux/kbd_kern.h>
 #include <linux/vt_kern.h>
 #include <linux/kbd_diacr.h>
@@ -700,6 +702,8 @@ int vt_ioctl(struct tty_struct *tty,
 		if (vsa.console == 0 || vsa.console > MAX_NR_CONSOLES)
 			ret = -ENXIO;
 		else {
+			vsa.console = array_index_nospec(vsa.console,
+							 MAX_NR_CONSOLES + 1);
 			vsa.console--;
 			console_lock();
 			ret = vc_allocate(vsa.console);
-- 
2.7.4

Re: [PATCH] tty: vt_ioctl: fix potential Spectre v1

[Alan Cox]

On Thu, 16 Aug 2018 15:30:38 -0500
"Gustavo A. R. Silva" <gustavo@embeddedor.com> wrote:

> vsa.console is indirectly controlled by user-space, hence leading to
> a potential exploitation of the Spectre variant 1 vulnerability.
> 
> This issue was detected with the help of Smatch:
> 
> drivers/tty/vt/vt_ioctl.c:711 vt_ioctl() warn: potential spectre issue
> 'vc_cons' [r]
> 
> Fix this by sanitizing vsa.console before using it to index vc_cons
> 
> Notice that given that speculation windows are large, the policy is
> to kill the speculation on the first load and not worry if it can be
> completed with a dependent load/store [1].
> 
> [1] https://marc.info/?l=linux-kernel&m=152449131114778&w=2
> 
> Cc: stable@vger.kernel.org
> Signed-off-by: Gustavo A. R. Silva <gustavo@embeddedor.com>

Reviewed-by: Alan Cox <alan@linux.intel.com>

Alan