[PATCH 1/2 net] NFC: nci: Add some bounds checking in nci_hci_cmd_received()

[PATCH 1/2 net] NFC: nci: Add some bounds checking in nci_hci_cmd_received()

[Dan Carpenter]

This is similar to commit 674d9de02aa7 ("NFC: Fix possible memory
corruption when handling SHDLC I-Frame commands").

I'm not totally sure, but I think that commit description may have
overstated the danger.  I was under the impression that this data came
from the firmware?  If you can't trust your networking firmware, then
you're already in trouble.

Anyway, these days we add bounds checking where ever we can and we call
it kernel hardening.  Better safe than sorry.

Fixes: 11f54f228643 ("NFC: nci: Add HCI over NCI protocol support")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
---
 net/nfc/nci/hci.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/net/nfc/nci/hci.c b/net/nfc/nci/hci.c
index ddfc52ac1f9b..c0d323b58e73 100644
--- a/net/nfc/nci/hci.c
+++ b/net/nfc/nci/hci.c
@@ -312,6 +312,10 @@ static void nci_hci_cmd_received(struct nci_dev *ndev, u8 pipe,
 		create_info = (struct nci_hci_create_pipe_resp *)skb->data;
 		dest_gate = create_info->dest_gate;
 		new_pipe = create_info->pipe;
+		if (new_pipe >= NCI_HCI_MAX_PIPES) {
+			status = NCI_HCI_ANY_E_NOK;
+			goto exit;
+		}
 
 		/* Save the new created pipe and bind with local gate,
 		 * the description for skb->data[3] is destination gate id
@@ -336,6 +340,10 @@ static void nci_hci_cmd_received(struct nci_dev *ndev, u8 pipe,
 			goto exit;
 		}
 		delete_info = (struct nci_hci_delete_pipe_noti *)skb->data;
+		if (delete_info->pipe >= NCI_HCI_MAX_PIPES) {
+			status = NCI_HCI_ANY_E_NOK;
+			goto exit;
+		}
 
 		ndev->hci_dev->pipes[delete_info->pipe].gate =
 						NCI_HCI_INVALID_GATE;
-- 
2.17.1

[PATCH 2/2 net] nfc: nci: Potential off by one in ->pipes[] array

[Dan Carpenter]

This is similar to commit e285d5bfb7e9 ("NFC: Fix the number of pipes")
where we changed NFC_HCI_MAX_PIPES from 127 to 128.

As the comment next to the define explains, the pipe identifier is 7
bits long.  The highest possible pipe is 127, but the number of possible
pipes is 128.  As the code is now, then there is potential for an
out of bounds array access:

    net/nfc/nci/hci.c:297 nci_hci_cmd_received() warn: array off by one?
    'ndev->hci_dev->pipes[pipe]' '0-127 == 127'

Fixes: 11f54f228643 ("NFC: nci: Add HCI over NCI protocol support")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
---
 include/net/nfc/nci_core.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/include/net/nfc/nci_core.h b/include/net/nfc/nci_core.h
index 87499b6b35d6..df5c69db68af 100644
--- a/include/net/nfc/nci_core.h
+++ b/include/net/nfc/nci_core.h
@@ -166,7 +166,7 @@ struct nci_conn_info {
  * According to specification 102 622 chapter 4.4 Pipes,
  * the pipe identifier is 7 bits long.
  */
-#define NCI_HCI_MAX_PIPES          127
+#define NCI_HCI_MAX_PIPES          128
 
 struct nci_hci_gate {
 	u8 gate;
-- 
2.17.1

Re: [PATCH 1/2 net] NFC: nci: Add some bounds checking in nci_hci_cmd_received()

[David Miller]

From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Wed, 3 Apr 2019 10:12:48 +0300

> This is similar to commit 674d9de02aa7 ("NFC: Fix possible memory
> corruption when handling SHDLC I-Frame commands").
> 
> I'm not totally sure, but I think that commit description may have
> overstated the danger.  I was under the impression that this data came
> from the firmware?  If you can't trust your networking firmware, then
> you're already in trouble.
> 
> Anyway, these days we add bounds checking where ever we can and we call
> it kernel hardening.  Better safe than sorry.
> 
> Fixes: 11f54f228643 ("NFC: nci: Add HCI over NCI protocol support")
> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>

Applied.

Re: [PATCH 2/2 net] nfc: nci: Potential off by one in ->pipes[] array

[David Miller]

From: Dan Carpenter <dan.carpenter@oracle.com>
Date: Wed, 3 Apr 2019 10:13:51 +0300

> This is similar to commit e285d5bfb7e9 ("NFC: Fix the number of pipes")
> where we changed NFC_HCI_MAX_PIPES from 127 to 128.
> 
> As the comment next to the define explains, the pipe identifier is 7
> bits long.  The highest possible pipe is 127, but the number of possible
> pipes is 128.  As the code is now, then there is potential for an
> out of bounds array access:
> 
>     net/nfc/nci/hci.c:297 nci_hci_cmd_received() warn: array off by one?
>     'ndev->hci_dev->pipes[pipe]' '0-127 == 127'
> 
> Fixes: 11f54f228643 ("NFC: nci: Add HCI over NCI protocol support")
> Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>

Applied.