Skip to contentSkip to conversation

KMSAN: uninit-value in digitv_rc_query

PATCH In review syzbot 4 participants 5 messages Nov 8, 2019 – Jan 15, 2020 Media (V4L/DVB)

lore
Share & tools
lore

syzbot

author syzbot
#0 lore raw

Hello,

syzbot found the following crash on:

HEAD commit:    1e76a3e5 kmsan: replace __GFP_NO_KMSAN_SHADOW with kmsan_i..
git tree:       https://github.com/google/kmsan.git master
console output: https://syzkaller.appspot.com/x/log.txt?x=16860a63600000
kernel config:  https://syzkaller.appspot.com/x/.config?x=f03c659d0830ab8d
dashboard link: https://syzkaller.appspot.com/bug?extid=6bf9606ee955b646c0e1
compiler:       clang version 9.0.0 (/home/glider/llvm/clang 80fee25776c2fb61e74c1ecb1a523375c2500b69)
syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=12c1101b600000
C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=15db3cfd600000

IMPORTANT: if you fix the bug, please add the following tag to the commit:

Reported-by [email protected]

dvb-usb: schedule remote query interval to 1000 msecs.
dvb-usb: Nebula Electronics uDigiTV DVB-T USB2.0) successfully initialized and connected.
dvb-usb: bulk message failed: -22 (7/0)
dvb-usb: bulk message failed: -22 (7/0)
dvb-usb: bulk message failed: -22 (7/-30591)
dvb-usb: bulk message failed: -22 (7/0)
=====================================================

BUG: KMSAN: uninit-value in legacy_dvb_usb_read_remote_control+0x106/0x790 drivers/media/usb/dvb-usb/dvb-usb-remote.c:123
CPU: 1 PID: 3844 Comm: kworker/1:2 Not tainted 5.3.0-rc7+ #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Workqueue: events legacy_dvb_usb_read_remote_control
Call Trace:
 __dump_stack lib/dump_stack.c:77 [inline]
 dump_stack+0x191/0x1f0 lib/dump_stack.c:113
 kmsan_report+0x13a/0x2b0 mm/kmsan/kmsan_report.c:108
 __msan_warning+0x73/0xe0 mm/kmsan/kmsan_instr.c:250
 digitv_rc_query+0x76a/0x890 drivers/media/usb/dvb-usb/digitv.c:259
 legacy_dvb_usb_read_remote_control+0x106/0x790  
drivers/media/usb/dvb-usb/dvb-usb-remote.c:123
 process_one_work+0x1572/0x1ef0 kernel/workqueue.c:2269
 worker_thread+0x111b/0x2460 kernel/workqueue.c:2415
 kthread+0x4b5/0x4f0 kernel/kthread.c:256
 ret_from_fork+0x35/0x40 arch/x86/entry/entry_64.S:355

Local variable description: ----key@digitv_rc_query Variable was created at:

 digitv_rc_query+0x78/0x890 drivers/media/usb/dvb-usb/digitv.c:234
 legacy_dvb_usb_read_remote_control+0x106/0x790  
drivers/media/usb/dvb-usb/dvb-usb-remote.c:123
=====================================================
Kernel panic - not syncing: panic_on_warn set ...
CPU: 1 PID: 3844 Comm: kworker/1:2 Tainted: G    B             5.3.0-rc7+ #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Workqueue: events legacy_dvb_usb_read_remote_control
Call Trace:
 __dump_stack lib/dump_stack.c:77 [inline]
 dump_stack+0x191/0x1f0 lib/dump_stack.c:113
 panic+0x3c9/0xc1e kernel/panic.c:219
 kmsan_report+0x2a2/0x2b0 mm/kmsan/kmsan_report.c:131
 __msan_warning+0x73/0xe0 mm/kmsan/kmsan_instr.c:250
 digitv_rc_query+0x76a/0x890 drivers/media/usb/dvb-usb/digitv.c:259
 legacy_dvb_usb_read_remote_control+0x106/0x790  
drivers/media/usb/dvb-usb/dvb-usb-remote.c:123
 process_one_work+0x1572/0x1ef0 kernel/workqueue.c:2269
 worker_thread+0x111b/0x2460 kernel/workqueue.c:2415
 kthread+0x4b5/0x4f0 kernel/kthread.c:256
 ret_from_fork+0x35/0x40 arch/x86/entry/entry_64.S:355

Kernel Offset: disabled
Rebooting in 86400 seconds..

---
This bug is generated by a bot. It may contain errors. See https://goo.gl/tpsmEJ for more information about syzbot. syzbot engineers can be reached at [email protected].

syzbot will keep track of this bug report. See:
https://goo.gl/tpsmEJ#status for how to communicate with syzbot. syzbot can test patches for this bug, for details see:
https://goo.gl/tpsmEJ#testing-patches

#1 lore raw

PATCH 1/1 media: dvb: check return value digitv_ctrl_msg

For fixing syzbot "KMSAN: uninit-value in digitv_rc_query"

In scenario testing for syzbot, failure reading from digitv_ctrl_msg() [1].

Eg:

[   91.846657][ T3844] dvb-usb: bulk message failed: -22 (7/0)

digitv_rc_query() always return 0. But in this case a wrong thing happens.

[1]: https://syzkaller.appspot.com/text?tag=CrashLog&x=16860a63600000
[2]: https://groups.google.com/d/msg/syzkaller-bugs/-TXIJAZ0J9Q/T4PEUQoeAQAJ

---
 drivers/media/usb/dvb-usb/digitv.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)
drivers/media/usb/dvb-usb/digitv.c +8 -4
index dd5bb230cec1..61bc8945e6b9 100644
@@ -231,17 +231,21 @@ static struct rc_map_table rc_map_digitv_table[] = {
 static int digitv_rc_query(struct dvb_usb_device *d, u32 *event, int *state)
 {
 	int i;
-	u8 key[5];
+	u8 key[5] = { 0 };
 	u8 b[4] = { 0 };
+	int ret;
 
 	*event = 0;
 	*state = REMOTE_NO_KEY_PRESSED;
 
-	digitv_ctrl_msg(d,USB_READ_REMOTE,0,NULL,0,&key[1],4);
-
+	ret = digitv_ctrl_msg(d, USB_READ_REMOTE, 0, NULL, 0, &key[1], 4);
+	if (ret < 0)
+		return ret;
 	/* Tell the device we've read the remote. Not sure how necessary
 	   this is, but the Nebula SDK does it. */
-	digitv_ctrl_msg(d,USB_WRITE_REMOTE,0,b,4,NULL,0);
+	ret = digitv_ctrl_msg(d, USB_WRITE_REMOTE, 0, b, 4, NULL, 0);
+	if (ret < 0)
+		return ret;
 
 	/* if something is inside the buffer, simulate key press */
 	if (key[1] != 0)
Signature · 1 line
2.20.1
#2 lore raw

Re: [PATCH] media: dvb: check return value digitv_ctrl_msg

Hello,

Quoted 5 lines · Phong Tran

For fixing syzbot "KMSAN: uninit-value in digitv_rc_query"

In scenario testing for syzbot, failure reading from digitv_ctrl_msg() [1].

Quoted 7 lines

Eg:

[   91.846657][ T3844] dvb-usb: bulk message failed: -22 (7/0)

digitv_rc_query() always return 0. But in this case a wrong thing happens.

A fix for this was already merged I'm afraid, see commit eecc70d22ae5 ("media: digitv: don't continue if remote control state can't be read").

Quoted 13 lines

[1]: https://syzkaller.appspot.com/text?tag=CrashLog&x=16860a63600000
[2]: https://groups.google.com/d/msg/syzkaller-bugs/-TXIJAZ0J9Q/T4PEUQoeAQAJ

---
 drivers/media/usb/dvb-usb/digitv.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)
drivers/media/usb/dvb-usb/digitv.c +0 -0
index dd5bb230cec1..61bc8945e6b9 100644

drivers/media/usb/dvb-usb/digitv.c static struct rc_map_table rc_map_digitv_table[] = { · line 234

Quoted 5 lines
drivers/media/usb/dvb-usb/digitv.c +1 -1
@@ -231,17 +231,21 @@ static struct rc_map_table rc_map_digitv_table[] = {
 static int digitv_rc_query(struct dvb_usb_device *d, u32 *event, int *state)
 {
 	int i;
-	u8 key[5];
+	u8 key[5] = { 0 };

The merged commit does not change this line. Why was this changed?

Thanks

Sean

Quoted 22 lines
drivers/media/usb/dvb-usb/digitv.c +7 -3
 	u8 b[4] = { 0 };
+	int ret;
 
 	*event = 0;
 	*state = REMOTE_NO_KEY_PRESSED;
 
-	digitv_ctrl_msg(d,USB_READ_REMOTE,0,NULL,0,&key[1],4);
-
+	ret = digitv_ctrl_msg(d, USB_READ_REMOTE, 0, NULL, 0, &key[1], 4);
+	if (ret < 0)
+		return ret;
 	/* Tell the device we've read the remote. Not sure how necessary
 	   this is, but the Nebula SDK does it. */
-	digitv_ctrl_msg(d,USB_WRITE_REMOTE,0,b,4,NULL,0);
+	ret = digitv_ctrl_msg(d, USB_WRITE_REMOTE, 0, b, 4, NULL, 0);
+	if (ret < 0)
+		return ret;
 
 	/* if something is inside the buffer, simulate key press */
 	if (key[1] != 0)
Signature · 1 line
2.20.1
#3 lore raw

Re: [PATCH] media: dvb: check return value digitv_ctrl_msg

Quoted 27 lines · Sean Young

Hello,

Quoted 5 lines · Phong Tran

For fixing syzbot "KMSAN: uninit-value in digitv_rc_query"

In scenario testing for syzbot, failure reading from digitv_ctrl_msg() [1].

Quoted 7 lines

Eg:

[   91.846657][ T3844] dvb-usb: bulk message failed: -22 (7/0)

digitv_rc_query() always return 0. But in this case a wrong thing happens.

A fix for this was already merged I'm afraid, see commit eecc70d22ae5 ("media: digitv: don't continue if remote control state can't be read").

Quoted 8 lines

[1]: https://syzkaller.appspot.com/text?tag=CrashLog&x=16860a63600000
[2]: https://groups.google.com/d/msg/syzkaller-bugs/-TXIJAZ0J9Q/T4PEUQoeAQAJ

---
 drivers/media/usb/dvb-usb/digitv.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)
Quoted 13 lines

drivers/media/usb/dvb-usb/digitv.c static struct rc_map_table rc_map_digitv_table[] = { · line 234

Quoted 10 lines
drivers/media/usb/dvb-usb/digitv.c +1 -1
index dd5bb230cec1..61bc8945e6b9 100644
@@ -231,17 +231,21 @@ static struct rc_map_table rc_map_digitv_table[] = {
 static int digitv_rc_query(struct dvb_usb_device *d, u32 *event, int *state)
 {
 	int i;
-	u8 key[5];
+	u8 key[5] = { 0 };

The merged commit does not change this line. Why was this changed?

It would fix the problem that key[0] is never initialized... But the correct fix is to make key 4 elements long and delete key[0].

regards,
dan carpenter

#4 lore raw

Re: [PATCH] media: dvb: check return value digitv_ctrl_msg

Quoted 42 lines · Dan Carpenter
Quoted 27 lines · Sean Young

Hello,

Quoted 5 lines · Phong Tran

For fixing syzbot "KMSAN: uninit-value in digitv_rc_query"

In scenario testing for syzbot, failure reading from digitv_ctrl_msg() [1].

Quoted 7 lines

Eg:

[   91.846657][ T3844] dvb-usb: bulk message failed: -22 (7/0)

digitv_rc_query() always return 0. But in this case a wrong thing happens.

A fix for this was already merged I'm afraid, see commit eecc70d22ae5 ("media: digitv: don't continue if remote control state can't be read").

Quoted 8 lines

[1]: https://syzkaller.appspot.com/text?tag=CrashLog&x=16860a63600000
[2]: https://groups.google.com/d/msg/syzkaller-bugs/-TXIJAZ0J9Q/T4PEUQoeAQAJ

---
 drivers/media/usb/dvb-usb/digitv.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)
Quoted 13 lines

drivers/media/usb/dvb-usb/digitv.c static struct rc_map_table rc_map_digitv_table[] = { · line 234

Quoted 10 lines
drivers/media/usb/dvb-usb/digitv.c +1 -1
index dd5bb230cec1..61bc8945e6b9 100644
@@ -231,17 +231,21 @@ static struct rc_map_table rc_map_digitv_table[] = {
 static int digitv_rc_query(struct dvb_usb_device *d, u32 *event, int *state)
 {
 	int i;
-	u8 key[5];
+	u8 key[5] = { 0 };

The merged commit does not change this line. Why was this changed?

Quoted 2 lines

It would fix the problem that key[0] is never initialized... But the correct fix is to make key 4 elements long and delete key[0].

Phong,

Presumably you can fix this? You will have to renumber key[1] to key[0] and key[2] to key[1] etc... Add a fixes tag.

Fixes 774c0de4aed4 ("V4L/DVB (4616): [PATCH] Nebula DigiTV USB RC support")

Otherwise if you want I can send the patch.

regards,
dan carpenter