Re: 6pack: stack-out-of-bounds in sixpack_receive_buf

[Dmitry Vyukov]

Hello,

While running syzkaller fuzzer I've got the following report:

BUG: KASAN: stack-out-of-bounds in sixpack_receive_buf+0xf8a/0x1450 at
addr ffff880037fbf850
Read of size 1 by task syz-executor/6759
page:ffffea0000dfefc0 count:0 mapcount:0 mapping: (null) index:0x0
flags: 0x1fffc0000000000()
page dumped because: kasan: bad access detected
CPU: 3 PID: 6759 Comm: syz-executor Not tainted 4.8.0-rc3-next-20160825+ #8
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS Bochs 01/01/2011
ffffffff886b6fe0 ffff880037fbf520 ffffffff82db38d9 ffffffff37fbf5b0
fffffbfff10d6dfc ffff880037fbf5b0 ffff880037fbf850 ffff880037fbf850
ffff880037d3f180 dffffc0000000000 ffff880037fbf5a0 ffffffff8180a383
Call Trace:
[<ffffffff8180a3ee>] __asan_report_load1_noabort+0x3e/0x40
mm/kasan/report.c:319
[< inline >] sixpack_decode drivers/net/hamradio/6pack.c:1001
[<ffffffff8425f96a>] sixpack_receive_buf+0xf8a/0x1450
drivers/net/hamradio/6pack.c:462
[<ffffffff8323b368>] tty_ldisc_receive_buf+0x168/0x1b0
drivers/tty/tty_buffer.c:433
[<ffffffff832616de>] paste_selection+0x27e/0x3e0 drivers/tty/vt/selection.c:363
[<ffffffff8327f286>] tioclinux+0x126/0x410 drivers/tty/vt/vt.c:2683
[<ffffffff8325c1ef>] vt_ioctl+0x13ef/0x2910 drivers/tty/vt/vt_ioctl.c:365
[<ffffffff832245cd>] tty_ioctl+0x69d/0x21e0 drivers/tty/tty_io.c:2983
[< inline >] vfs_ioctl fs/ioctl.c:43
[<ffffffff818a1dfc>] do_vfs_ioctl+0x18c/0x1080 fs/ioctl.c:675
[< inline >] SYSC_ioctl fs/ioctl.c:690
[<ffffffff818a2d7f>] SyS_ioctl+0x8f/0xc0 fs/ioctl.c:681
[<ffffffff86e10700>] entry_SYSCALL_64_fastpath+0x23/0xc1
Memory state around the buggy address:
ffff880037fbf700: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
ffff880037fbf780: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
>ffff880037fbf800: 00 00 00 00 00 00 00 00 00 00 f3 f3 f3 f3 00 00
^
ffff880037fbf880: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
ffff880037fbf900: 00 00 00 00 f1 f1 f1 f1 00 f4 f4 f4 f2 f2 f2 f2
==================================================================


It is then followed by similar reports that access subsequent stack bytes.
Unfortunately I can't reproduce it (though, I got 6 similar crashes in
different runs). Looking at code, the following looks suspicious -- we
limit copy by 512 bytes, but use the original count which can be
larger than 512:

static void sixpack_receive_buf(struct tty_struct *tty,
const unsigned char *cp, char *fp, int count)
{
unsigned char buf[512];
....
memcpy(buf, cp, count < sizeof(buf) ? count : sizeof(buf));
....
sixpack_decode(sp, buf, count1);


On commit 0f98f121e1670eaa2a2fbb675e07d6ba7f0e146f of linux-next.

[One Thousand Gnomes]

> different runs). Looking at code, the following looks suspicious -- we
> limit copy by 512 bytes, but use the original count which can be
> larger than 512:
>
> static void sixpack_receive_buf(struct tty_struct *tty,
> const unsigned char *cp, char *fp, int count)
> {
> unsigned char buf[512];
> ....
> memcpy(buf, cp, count < sizeof(buf) ? count : sizeof(buf));
> ....
> sixpack_decode(sp, buf, count1);
>
>
> On commit 0f98f121e1670eaa2a2fbb675e07d6ba7f0e146f of linux-next.

With the sane tty locking we now have I believe the following is safe as
we consume the bytes and move them into the decoded buffer before
returning.

diff --git a/drivers/net/hamradio/6pack.c b/drivers/net/hamradio/6pack.c
index 5a1e985..470b3dc 100644
--- a/drivers/net/hamradio/6pack.c
+++ b/drivers/net/hamradio/6pack.c
@@ -127,7 +127,7 @@ struct sixpack {

#define AX25_6PACK_HEADER_LEN 0

-static void sixpack_decode(struct sixpack *, unsigned char[], int);
+static void sixpack_decode(struct sixpack *, const unsigned char[], int);
static int encode_sixpack(unsigned char *, unsigned char *, int, unsigned char);

/*
@@ -428,7 +428,7 @@ out:

/*
* Handle the 'receiver data ready' interrupt.
- * This function is called by the 'tty_io' module in the kernel when
+ * This function is called by the tty module in the kernel when
* a block of 6pack data has been received, which can now be decapsulated
* and sent on to some IP layer for further processing.
*/
@@ -436,7 +436,6 @@ static void sixpack_receive_buf(struct tty_struct *tty,

const unsigned char *cp, char *fp, int count)
{

struct sixpack *sp;
- unsigned char buf[512];
int count1;

if (!count)
@@ -446,10 +445,7 @@ static void sixpack_receive_buf(struct tty_struct *tty,
if (!sp)
return;

- memcpy(buf, cp, count < sizeof(buf) ? count : sizeof(buf));
-
/* Read the characters out of the buffer */
-
count1 = count;
while (count) {
count--;
@@ -459,7 +455,7 @@ static void sixpack_receive_buf(struct tty_struct *tty,
continue;
}
}
- sixpack_decode(sp, buf, count1);
+ sixpack_decode(sp, cp, count1);

sp_put(sp);
tty_unthrottle(tty);
@@ -992,7 +988,7 @@ static void decode_std_command(struct sixpack *sp, unsigned char cmd)
/* decode a 6pack packet */

static void
-sixpack_decode(struct sixpack *sp, unsigned char *pre_rbuff, int count)
+sixpack_decode(struct sixpack *sp, const unsigned char *pre_rbuff, int count)
{
unsigned char inbyte;
int count1;

[Dmitry Vyukov]

Applied locally for testing. I will notify if I see this bug again.
Thanks!