Skip to contentSkip to conversation

Bluetooth: L2CAP: Fix ERTM re-init and zero pdu_len infinite loop

PATCH Superseded Hyunwoo Kim 3 participants 4 messages Mar 19

Subsystems (2)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 Bluetooth: L2CAP: Fix ERTM re-init and zero pdu_len infinite loop

l2cap_config_req() accepts CONFIG_REQ when the channel is already in BT_CONNECTED state. This causes l2cap_ertm_init() to be called a second time, re-initializing tx_q, srej_q, srej_list, and retrans_list without freeing the previous allocations, and setting chan->sdu to NULL without freeing the existing skb. This leaks all previously allocated ERTM resources.

Additionally, l2cap_parse_conf_req() does not validate the minimum value of remote_mps derived from the RFC max_pdu_size option. A zero value propagates to l2cap_segment_sdu() where pdu_len becomes zero, causing the while loop to never terminate since len is never decremented, exhausting all available memory.

Fix by removing BT_CONNECTED from the allowed states in l2cap_config_req() and adding a pdu_len zero check in l2cap_segment_sdu() as a safeguard.

Fixes 96298f640104 ("Bluetooth: L2CAP: handle l2cap config request during open state")
---
 net/bluetooth/l2cap_core.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
net/bluetooth/l2cap_core.c +4 -2
index ad98db9632fd..a6ce09781083 100644
@@ -2381,6 +2381,9 @@ static int l2cap_segment_sdu(struct l2cap_chan *chan,
 	/* Remote device may have requested smaller PDUs */
 	pdu_len = min_t(size_t, pdu_len, chan->remote_mps);
 
+	if (!pdu_len)
+		return -EINVAL;
+
 	if (len <= pdu_len) {
 		sar = L2CAP_SAR_UNSEGMENTED;
 		sdu_len = 0;
@@ -4267,8 +4270,7 @@ static inline int l2cap_config_req(struct l2cap_conn *conn,
 		return 0;
 	}
 
-	if (chan->state != BT_CONFIG && chan->state != BT_CONNECT2 &&
-	    chan->state != BT_CONNECTED) {
+	if (chan->state != BT_CONFIG && chan->state != BT_CONNECT2) {
 		cmd_reject_invalid_cid(conn, cmd->ident, chan->scid,
 				       chan->dcid);
 		goto unlock;
Signature · 1 line
2.43.0
#1 lore raw

This is automated email and please do not reply to this email!

Dear submitter,

Thank you for submitting the patches to the linux bluetooth mailing list. This is a CI test results with your patch series:
PW Link:https://patchwork.kernel.org/project/bluetooth/list/?series=1069346

---Test result---

Test Summary:

CheckPatch                    PENDING   0.67 seconds
GitLint                       PENDING   0.50 seconds
SubjectPrefix                 PASS      0.05 seconds
BuildKernel                   PASS      26.37 seconds
CheckAllWarning               PASS      28.52 seconds
CheckSparse                   PASS      27.38 seconds
BuildKernel32                 PASS      25.35 seconds
TestRunnerSetup               PASS      564.19 seconds
TestRunner_l2cap-tester       PASS      27.33 seconds
TestRunner_iso-tester         FAIL      29.99 seconds
TestRunner_bnep-tester        PASS      6.30 seconds
TestRunner_mgmt-tester        FAIL      112.94 seconds
TestRunner_rfcomm-tester      PASS      9.29 seconds
TestRunner_sco-tester         FAIL      14.23 seconds
TestRunner_ioctl-tester       PASS      13.73 seconds
TestRunner_mesh-tester        FAIL      11.48 seconds
TestRunner_smp-tester         PASS      8.37 seconds
TestRunner_userchan-tester    PASS      6.66 seconds
IncrementalBuild              PENDING   0.70 seconds

Details
##############################
Test: CheckPatch - PENDING
Desc: Run checkpatch.pl script
Output:

##############################
Test: GitLint - PENDING
Desc: Run gitlint
Output:

##############################
Test: TestRunner_iso-tester - FAIL
Desc: Run iso-tester with test-runner
Output:

BUG: KASAN: slab-use-after-free in le_read_features_complete+0x7e/0x2b0

Total: 141, Passed: 141 (100.0%), Failed: 0, Not Run: 0
##############################
Test: TestRunner_mgmt-tester - FAIL
Desc: Run mgmt-tester with test-runner
Output:
Total: 494, Passed: 489 (99.0%), Failed: 1, Not Run: 4

Failed Test Cases
Read Exp Feature - Success Failed 0.105 seconds
##############################
Test: TestRunner_sco-tester - FAIL
Desc: Run sco-tester with test-runner
Output:
WARNING: possible circular locking dependency detected

BUG: sleeping function called from invalid context at net/core/sock.c:3782

Total: 30, Passed: 30 (100.0%), Failed: 0, Not Run: 0
##############################
Test: TestRunner_mesh-tester - FAIL
Desc: Run mesh-tester with test-runner Output:
Total: 10, Passed: 8 (80.0%), Failed: 2, Not Run: 0

Failed Test Cases

Mesh - Send cancel - 1                               Timed out    1.938 seconds
Mesh - Send cancel - 2                               Timed out    1.995 seconds

##############################
Test: IncrementalBuild - PENDING
Desc: Incremental build with the patches in the series Output:

---
Regards,
Linux Bluetooth

#2 lore raw
In reply to Hyunwoo Kim

Hi Hyunwoo,

Quoted 14 lines · Hyunwoo Kim

l2cap_config_req() accepts CONFIG_REQ when the channel is already in BT_CONNECTED state. This causes l2cap_ertm_init() to be called a second time, re-initializing tx_q, srej_q, srej_list, and retrans_list without freeing the previous allocations, and setting chan->sdu to NULL without freeing the existing skb. This leaks all previously allocated ERTM resources.

Additionally, l2cap_parse_conf_req() does not validate the minimum value of remote_mps derived from the RFC max_pdu_size option. A zero value propagates to l2cap_segment_sdu() where pdu_len becomes zero, causing the while loop to never terminate since len is never decremented, exhausting all available memory.

Quoted 3 lines

Fix by removing BT_CONNECTED from the allowed states in l2cap_config_req() and adding a pdu_len zero check in l2cap_segment_sdu() as a safeguard.

This part is not clerarm it should be allowed to change the MTU while connected.

Quoted 33 lines
Fixes 96298f640104 ("Bluetooth: L2CAP: handle l2cap config request during open state")
---
 net/bluetooth/l2cap_core.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
net/bluetooth/l2cap_core.c +4 -3
index ad98db9632fd..a6ce09781083 100644
@@ -2381,6 +2381,9 @@ static int l2cap_segment_sdu(struct l2cap_chan *chan,
        /* Remote device may have requested smaller PDUs */
        pdu_len = min_t(size_t, pdu_len, chan->remote_mps);
 
+       if (!pdu_len)
+               return -EINVAL;
+
        if (len <= pdu_len) {
                sar = L2CAP_SAR_UNSEGMENTED;
                sdu_len = 0;
@@ -4267,8 +4270,7 @@ static inline int l2cap_config_req(struct l2cap_conn *conn,
                return 0;
        }
 
-       if (chan->state != BT_CONFIG && chan->state != BT_CONNECT2 &&
-           chan->state != BT_CONNECTED) {
+       if (chan->state != BT_CONFIG && chan->state != BT_CONNECT2) {
                cmd_reject_invalid_cid(conn, cmd->ident, chan->scid,
                                       chan->dcid);
                goto unlock;
--

2.43.0

Signature · 1 line
Luiz Augusto von Dentz
#3 lore raw

Hi Hyunwoo,

Quoted 18 lines · Luiz Augusto von Dentz

Hi Hyunwoo,

Quoted 14 lines · Hyunwoo Kim

l2cap_config_req() accepts CONFIG_REQ when the channel is already in BT_CONNECTED state. This causes l2cap_ertm_init() to be called a second time, re-initializing tx_q, srej_q, srej_list, and retrans_list without freeing the previous allocations, and setting chan->sdu to NULL without freeing the existing skb. This leaks all previously allocated ERTM resources.

Additionally, l2cap_parse_conf_req() does not validate the minimum value of remote_mps derived from the RFC max_pdu_size option. A zero value propagates to l2cap_segment_sdu() where pdu_len becomes zero, causing the while loop to never terminate since len is never decremented, exhausting all available memory.

Quoted 5 lines
Quoted 3 lines

Fix by removing BT_CONNECTED from the allowed states in l2cap_config_req() and adding a pdu_len zero check in l2cap_segment_sdu() as a safeguard.

This part is not clerarm it should be allowed to change the MTU while connected.

Looks like AI agree with me (Second point, first point is probably not possible):

https://sashiko.dev/#/patchset/abwTCkavRurEcEGO%40v4bel

Quoted 38 lines
Quoted 32 lines
Fixes 96298f640104 ("Bluetooth: L2CAP: handle l2cap config request during open state")
---
 net/bluetooth/l2cap_core.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
net/bluetooth/l2cap_core.c +4 -3
index ad98db9632fd..a6ce09781083 100644
@@ -2381,6 +2381,9 @@ static int l2cap_segment_sdu(struct l2cap_chan *chan,
        /* Remote device may have requested smaller PDUs */
        pdu_len = min_t(size_t, pdu_len, chan->remote_mps);
 
+       if (!pdu_len)
+               return -EINVAL;
+
        if (len <= pdu_len) {
                sar = L2CAP_SAR_UNSEGMENTED;
                sdu_len = 0;
@@ -4267,8 +4270,7 @@ static inline int l2cap_config_req(struct l2cap_conn *conn,
                return 0;
        }
 
-       if (chan->state != BT_CONFIG && chan->state != BT_CONNECT2 &&
-           chan->state != BT_CONNECTED) {
+       if (chan->state != BT_CONFIG && chan->state != BT_CONNECT2) {
                cmd_reject_invalid_cid(conn, cmd->ident, chan->scid,
                                       chan->dcid);
                goto unlock;
--
Quoted 2 lines

2.43.0

--
Luiz Augusto von Dentz

Signature · 1 line
Luiz Augusto von Dentz