linux-kernel - [PATCH 08/26] dib0700_core: don't use stack on I2C reads

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <bbcbc1d7e3cebee244e425931a2ad2cbd23bc6c8.1475860773.git.mchehab@s-opensource.com>
Date:   Fri,  7 Oct 2016 14:24:18 -0300
From:   Mauro Carvalho Chehab <mchehab@...pensource.com>
To:     Linux Media Mailing List <linux-media@...r.kernel.org>
Cc:     Mauro Carvalho Chehab <mchehab@...pensource.com>,
        Mauro Carvalho Chehab <mchehab@...radead.org>,
        Andy Lutomirski <luto@...capital.net>,
        Johannes Stezenbach <js@...uxtv.org>,
        Jiri Kosina <jikos@...nel.org>,
        Patrick Boettcher <patrick.boettcher@...teo.de>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Andy Lutomirski <luto@...nel.org>,
        Michael Krufky <mkrufky@...uxtv.org>,
        Mauro Carvalho Chehab <mchehab@...nel.org>,
        Jörg Otte <jrg.otte@...il.com>,
        Hans Verkuil <hans.verkuil@...co.com>,
        Sean Young <sean@...s.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Kees Cook <keescook@...omium.org>,
        Wolfram Sang <wsa-dev@...g-engineering.com>
Subject: [PATCH 08/26] dib0700_core: don't use stack on I2C reads

Be sure that I2C reads won't use stack by passing
a pointer to the state buffer, that we know it was
allocated via kmalloc, instead of relying on the buffer
allocated by an I2C client.

Signed-off-by: Mauro Carvalho Chehab <mchehab@...pensource.com>
---
 drivers/media/usb/dvb-usb/dib0700_core.c | 27 ++++++++++++++++++++++++++-
 1 file changed, 26 insertions(+), 1 deletion(-)

diff --git a/drivers/media/usb/dvb-usb/dib0700_core.c b/drivers/media/usb/dvb-usb/dib0700_core.c
index 515f89dba199..92d5408684ac 100644
--- a/drivers/media/usb/dvb-usb/dib0700_core.c
+++ b/drivers/media/usb/dvb-usb/dib0700_core.c
@@ -213,7 +213,7 @@ static int dib0700_i2c_xfer_new(struct i2c_adapter *adap, struct i2c_msg *msg,
 						 usb_rcvctrlpipe(d->udev, 0),
 						 REQUEST_NEW_I2C_READ,
 						 USB_TYPE_VENDOR | USB_DIR_IN,
-						 value, index, msg[i].buf,
+						 value, index, st->buf,
 						 msg[i].len,
 						 USB_CTRL_GET_TIMEOUT);
 			if (result < 0) {
@@ -221,6 +221,14 @@ static int dib0700_i2c_xfer_new(struct i2c_adapter *adap, struct i2c_msg *msg,
 				break;
 			}
 
+			if (msg[i].len > sizeof(st->buf)) {
+				deb_info("buffer too small to fit %d bytes\n",
+					 msg[i].len);
+				return -EIO;
+			}
+
+			memcpy(msg[i].buf, st->buf, msg[i].len);
+
 			deb_data("<<< ");
 			debug_dump(msg[i].buf, msg[i].len, deb_data);
 
@@ -238,6 +246,13 @@ static int dib0700_i2c_xfer_new(struct i2c_adapter *adap, struct i2c_msg *msg,
 			/* I2C ctrl + FE bus; */
 			st->buf[3] = ((gen_mode << 6) & 0xC0) |
 				 ((bus_mode << 4) & 0x30);
+
+			if (msg[i].len > sizeof(st->buf) - 4) {
+				deb_info("i2c message to big: %d\n",
+					 msg[i].len);
+				return -EIO;
+			}
+
 			/* The Actual i2c payload */
 			memcpy(&st->buf[4], msg[i].buf, msg[i].len);
 
@@ -283,6 +298,11 @@ static int dib0700_i2c_xfer_legacy(struct i2c_adapter *adap,
 		/* fill in the address */
 		st->buf[1] = msg[i].addr << 1;
 		/* fill the buffer */
+		if (msg[i].len > sizeof(st->buf) - 2) {
+			deb_info("i2c xfer to big: %d\n",
+				msg[i].len);
+			return -EIO;
+		}
 		memcpy(&st->buf[2], msg[i].buf, msg[i].len);
 
 		/* write/read request */
@@ -299,6 +319,11 @@ static int dib0700_i2c_xfer_legacy(struct i2c_adapter *adap,
 				break;
 			}
 
+			if (msg[i + 1].len > sizeof(st->buf)) {
+				deb_info("i2c xfer buffer to small for %d\n",
+					msg[i].len);
+				return -EIO;
+			}
 			memcpy(msg[i + 1].buf, st->buf, msg[i + 1].len);
 
 			msg[i+1].len = len;
-- 
2.7.4