Skip to contentSkip to conversation

5.15.203-rc1 review

PATCH In review Greg Kroah-Hartman 17 participants 597 messages Apr 13 – Jun 2 Page 3 of 5

Subsystems (7)
Stable branch Documentation SCSI HID +3 more subsystems
lore
Share & tools
lore
author maintainer
#277 lore raw

PATCH 266/570 sched: idle: Consolidate the handling of two special cases

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafael J. Wysocki <[email protected]>

[ Upstream commit f4c31b07b136839e0fb3026f8a5b6543e3b14d2f ]

There are two special cases in the idle loop that are handled inconsistently even though they are analogous.

The first one is when a cpuidle driver is absent and the default CPU idle time power management implemented by the architecture code is used. In that case, the scheduler tick is stopped every time before invoking default_idle_call().

The second one is when a cpuidle driver is present, but there is only one idle state in its table. In that case, the scheduler tick is never stopped at all.

Since each of these approaches has its drawbacks, reconcile them with the help of one simple heuristic. Namely, stop the tick if the CPU has been woken up by it in the previous iteration of the idle loop, or let it tick otherwise.

Fixes ed98c3491998 ("sched: idle: Do not stop the tick before cpuidle_idle_call()")
[ rjw: Added Fixes tag, changelog edits ]
---
 kernel/sched/idle.c | 30 +++++++++++++++++++++---------
 1 file changed, 21 insertions(+), 9 deletions(-)
kernel/sched/idle.c +21 -9
index 407835d23eacf..f1c58e2fc3b5c 100644
@@ -158,6 +158,14 @@ static int call_cpuidle(struct cpuidle_driver *drv, struct cpuidle_device *dev,
 	return cpuidle_enter(drv, dev, next_state);
 }
 
+static void idle_call_stop_or_retain_tick(bool stop_tick)
+{
+	if (stop_tick || tick_nohz_tick_stopped())
+		tick_nohz_idle_stop_tick();
+	else
+		tick_nohz_idle_retain_tick();
+}
+
 /**
  * cpuidle_idle_call - the main idle function
  *
@@ -167,7 +175,7 @@ static int call_cpuidle(struct cpuidle_driver *drv, struct cpuidle_device *dev,
  * set, and it returns with polling set.  If it ever stops polling, it
  * must clear the polling bit.
  */
-static void cpuidle_idle_call(void)
+static void cpuidle_idle_call(bool stop_tick)
 {
 	struct cpuidle_device *dev = cpuidle_get_device();
 	struct cpuidle_driver *drv = cpuidle_get_cpu_driver(dev);
@@ -189,7 +197,7 @@ static void cpuidle_idle_call(void)
 	 */
 
 	if (cpuidle_not_available(drv, dev)) {
-		tick_nohz_idle_stop_tick();
+		idle_call_stop_or_retain_tick(stop_tick);
 
 		default_idle_call();
 		goto exit_idle;
@@ -224,17 +232,19 @@ static void cpuidle_idle_call(void)
 		next_state = cpuidle_find_deepest_state(drv, dev, max_latency_ns);
 		call_cpuidle(drv, dev, next_state);
 	} else if (drv->state_count > 1) {
-		bool stop_tick = true;
+		/*
+		 * stop_tick is expected to be true by default by cpuidle
+		 * governors, which allows them to select idle states with
+		 * target residency above the tick period length.
+		 */
+		stop_tick = true;
 
 		/*
 		 * Ask the cpuidle framework to choose a convenient idle state.
 		 */
 		next_state = cpuidle_select(drv, dev, &stop_tick);
 
-		if (stop_tick || tick_nohz_tick_stopped())
-			tick_nohz_idle_stop_tick();
-		else
-			tick_nohz_idle_retain_tick();
+		idle_call_stop_or_retain_tick(stop_tick);
 
 		entered_state = call_cpuidle(drv, dev, next_state);
 		/*
@@ -242,7 +252,7 @@ static void cpuidle_idle_call(void)
 		 */
 		cpuidle_reflect(dev, entered_state);
 	} else {
-		tick_nohz_idle_retain_tick();
+		idle_call_stop_or_retain_tick(stop_tick);
 
 		/*
 		 * If there is only a single idle state (or none), there is
@@ -270,6 +280,7 @@ static void cpuidle_idle_call(void)
 static void do_idle(void)
 {
 	int cpu = smp_processor_id();
+	bool got_tick = false;
 
 	/*
 	 * Check if we need to update blocked load
@@ -312,8 +323,9 @@ static void do_idle(void)
 			tick_nohz_idle_restart_tick();
 			cpu_idle_poll();
 		} else {
-			cpuidle_idle_call();
+			cpuidle_idle_call(got_tick);
 		}
+		got_tick = tick_nohz_idle_got_tick();
 		arch_cpu_idle_exit();
 	}
 
Signature · 1 line
2.51.0
author maintainer
#278 lore raw

PATCH 267/570 PM: runtime: Fix a race condition related to device removal

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Bart Van Assche <[email protected]>

[ Upstream commit 29ab768277617452d88c0607c9299cdc63b6e9ff ]

The following code in pm_runtime_work() may dereference the dev->parent pointer after the parent device has been freed:

	/* Maybe the parent is now able to suspend. */
	if (parent && !parent->power.ignore_children) {
		spin_unlock(&dev->power.lock);

		spin_lock(&parent->power.lock);
		rpm_idle(parent, RPM_ASYNC);
		spin_unlock(&parent->power.lock);

		spin_lock(&dev->power.lock);
	}

Fix this by inserting a flush_work() call in pm_runtime_remove().

Without this patch blktest block/001 triggers the following complaint sporadically:

BUG: KASAN: slab-use-after-free in lock_acquire+0x70/0x160
Read of size 1 at addr ffff88812bef7198 by task kworker/u553:1/3081
Workqueue: pm pm_runtime_work
Call Trace:
 <TASK>
 dump_stack_lvl+0x61/0x80
 print_address_description.constprop.0+0x8b/0x310
 print_report+0xfd/0x1d7
 kasan_report+0xd8/0x1d0
 __kasan_check_byte+0x42/0x60
 lock_acquire.part.0+0x38/0x230
 lock_acquire+0x70/0x160
 _raw_spin_lock+0x36/0x50
 rpm_suspend+0xc6a/0xfe0
 rpm_idle+0x578/0x770
 pm_runtime_work+0xee/0x120
 process_one_work+0xde3/0x1410
 worker_thread+0x5eb/0xfe0
 kthread+0x37b/0x480
 ret_from_fork+0x6cb/0x920
 ret_from_fork_asm+0x11/0x20
 </TASK>

Allocated by task 4314:
 kasan_save_stack+0x2a/0x50
 kasan_save_track+0x18/0x40
 kasan_save_alloc_info+0x3d/0x50
 __kasan_kmalloc+0xa0/0xb0
 __kmalloc_noprof+0x311/0x990
 scsi_alloc_target+0x122/0xb60 [scsi_mod]
 __scsi_scan_target+0x101/0x460 [scsi_mod]
 scsi_scan_channel+0x179/0x1c0 [scsi_mod]
 scsi_scan_host_selected+0x259/0x2d0 [scsi_mod]
 store_scan+0x2d2/0x390 [scsi_mod]
 dev_attr_store+0x43/0x80
 sysfs_kf_write+0xde/0x140
 kernfs_fop_write_iter+0x3ef/0x670
 vfs_write+0x506/0x1470
 ksys_write+0xfd/0x230
 __x64_sys_write+0x76/0xc0
 x64_sys_call+0x213/0x1810
 do_syscall_64+0xee/0xfc0
 entry_SYSCALL_64_after_hwframe+0x4b/0x53

Freed by task 4314:
 kasan_save_stack+0x2a/0x50
 kasan_save_track+0x18/0x40
 kasan_save_free_info+0x3f/0x50
 __kasan_slab_free+0x67/0x80
 kfree+0x225/0x6c0
 scsi_target_dev_release+0x3d/0x60 [scsi_mod]
 device_release+0xa3/0x220
 kobject_cleanup+0x105/0x3a0
 kobject_put+0x72/0xd0
 put_device+0x17/0x20
 scsi_device_dev_release+0xacf/0x12c0 [scsi_mod]
 device_release+0xa3/0x220
 kobject_cleanup+0x105/0x3a0
 kobject_put+0x72/0xd0
 put_device+0x17/0x20
 scsi_device_put+0x7f/0xc0 [scsi_mod]
 sdev_store_delete+0xa5/0x120 [scsi_mod]
 dev_attr_store+0x43/0x80
 sysfs_kf_write+0xde/0x140
 kernfs_fop_write_iter+0x3ef/0x670
 vfs_write+0x506/0x1470
 ksys_write+0xfd/0x230
 __x64_sys_write+0x76/0xc0
 x64_sys_call+0x213/0x1810
---
 drivers/base/power/runtime.c | 1 +
 1 file changed, 1 insertion(+)
drivers/base/power/runtime.c +1 -0
index 7dcf2498965a3..f94d9223ab151 100644
@@ -1774,6 +1774,7 @@ void pm_runtime_reinit(struct device *dev)
 void pm_runtime_remove(struct device *dev)
 {
 	__pm_runtime_disable(dev, false);
+	flush_work(&dev->power.work);
 	pm_runtime_reinit(dev);
 }
 
Signature · 1 line
2.51.0
author maintainer
#279 lore raw

PATCH 268/570 net/smc: Only save the original clcsock callback functions

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wen Gu <[email protected]>

[ Upstream commit 97b9af7a70936e331170c79040cc9bf20071b566 ]

Both listen and fallback process will save the current clcsock callback functions and establish new ones. But if both of them happen, the saved callback functions will be overwritten.

So this patch introduces some helpers to ensure that only save the original callback functions of clcsock.

Fixes 341adeec9ada ("net/smc: Forward wakeup to smc socket waitqueue after fallback")

Stable-dep-of: 6d5e4538364b ("net/smc: fix NULL dereference and UAF in smc_tcp_syn_recv_sock()")

---
 net/smc/af_smc.c    | 55 +++++++++++++++++++++++++++++----------------
 net/smc/smc.h       | 29 ++++++++++++++++++++++++
 net/smc/smc_close.c |  3 ++-
 3 files changed, 67 insertions(+), 20 deletions(-)
net/smc/af_smc.c +36 -19
index 2a642dfbc94a1..5c6759d2e271d 100644
@@ -300,6 +300,7 @@ static struct sock *smc_sock_alloc(struct net *net, struct socket *sock,
 	sk->sk_prot->hash(sk);
 	sk_refcnt_debug_inc(sk);
 	mutex_init(&smc->clcsock_release_lock);
+	smc_init_saved_callbacks(smc);
 
 	return sk;
 }
@@ -696,9 +697,24 @@ static void smc_fback_error_report(struct sock *clcsk)
 	smc_fback_forward_wakeup(smc, clcsk, smc->clcsk_error_report);
 }
 
+static void smc_fback_replace_callbacks(struct smc_sock *smc)
+{
+	struct sock *clcsk = smc->clcsock->sk;
+
+	clcsk->sk_user_data = (void *)((uintptr_t)smc | SK_USER_DATA_NOCOPY);
+
+	smc_clcsock_replace_cb(&clcsk->sk_state_change, smc_fback_state_change,
+			       &smc->clcsk_state_change);
+	smc_clcsock_replace_cb(&clcsk->sk_data_ready, smc_fback_data_ready,
+			       &smc->clcsk_data_ready);
+	smc_clcsock_replace_cb(&clcsk->sk_write_space, smc_fback_write_space,
+			       &smc->clcsk_write_space);
+	smc_clcsock_replace_cb(&clcsk->sk_error_report, smc_fback_error_report,
+			       &smc->clcsk_error_report);
+}
+
 static int smc_switch_to_fallback(struct smc_sock *smc, int reason_code)
 {
-	struct sock *clcsk;
 	int rc = 0;
 
 	mutex_lock(&smc->clcsock_release_lock);
@@ -706,10 +722,7 @@ static int smc_switch_to_fallback(struct smc_sock *smc, int reason_code)
 		rc = -EBADF;
 		goto out;
 	}
-	clcsk = smc->clcsock->sk;
 
-	if (smc->use_fallback)
-		goto out;
 	smc->use_fallback = true;
 	smc->fallback_rsn = reason_code;
 	smc_stat_fallback(smc);
@@ -723,18 +736,7 @@ static int smc_switch_to_fallback(struct smc_sock *smc, int reason_code)
 		 * in smc sk->sk_wq and they should be woken up
 		 * as clcsock's wait queue is woken up.
 		 */
-		smc->clcsk_state_change = clcsk->sk_state_change;
-		smc->clcsk_data_ready = clcsk->sk_data_ready;
-		smc->clcsk_write_space = clcsk->sk_write_space;
-		smc->clcsk_error_report = clcsk->sk_error_report;
-
-		clcsk->sk_state_change = smc_fback_state_change;
-		clcsk->sk_data_ready = smc_fback_data_ready;
-		clcsk->sk_write_space = smc_fback_write_space;
-		clcsk->sk_error_report = smc_fback_error_report;
-
-		smc->clcsock->sk->sk_user_data =
-			(void *)((uintptr_t)smc | SK_USER_DATA_NOCOPY);
+		smc_fback_replace_callbacks(smc);
 	}
 out:
 	mutex_unlock(&smc->clcsock_release_lock);
@@ -1388,6 +1390,19 @@ static int smc_clcsock_accept(struct smc_sock *lsmc, struct smc_sock **new_smc)
 	 * function; switch it back to the original sk_data_ready function
 	 */
 	new_clcsock->sk->sk_data_ready = lsmc->clcsk_data_ready;
+
+	/* if new clcsock has also inherited the fallback-specific callback
+	 * functions, switch them back to the original ones.
+	 */
+	if (lsmc->use_fallback) {
+		if (lsmc->clcsk_state_change)
+			new_clcsock->sk->sk_state_change = lsmc->clcsk_state_change;
+		if (lsmc->clcsk_write_space)
+			new_clcsock->sk->sk_write_space = lsmc->clcsk_write_space;
+		if (lsmc->clcsk_error_report)
+			new_clcsock->sk->sk_error_report = lsmc->clcsk_error_report;
+	}
+
 	(*new_smc)->clcsock = new_clcsock;
 out:
 	return rc;
@@ -2122,10 +2137,10 @@ static int smc_listen(struct socket *sock, int backlog)
 	/* save original sk_data_ready function and establish
 	 * smc-specific sk_data_ready function
 	 */
-	smc->clcsk_data_ready = smc->clcsock->sk->sk_data_ready;
-	smc->clcsock->sk->sk_data_ready = smc_clcsock_data_ready;
 	smc->clcsock->sk->sk_user_data =
 		(void *)((uintptr_t)smc | SK_USER_DATA_NOCOPY);
+	smc_clcsock_replace_cb(&smc->clcsock->sk->sk_data_ready,
+			       smc_clcsock_data_ready, &smc->clcsk_data_ready);
 
 	/* save original ops */
 	smc->ori_af_ops = inet_csk(smc->clcsock->sk)->icsk_af_ops;
@@ -2137,7 +2152,9 @@ static int smc_listen(struct socket *sock, int backlog)
 
 	rc = kernel_listen(smc->clcsock, backlog);
 	if (rc) {
-		smc->clcsock->sk->sk_data_ready = smc->clcsk_data_ready;
+		smc_clcsock_restore_cb(&smc->clcsock->sk->sk_data_ready,
+				       &smc->clcsk_data_ready);
+		smc->clcsock->sk->sk_user_data = NULL;
 		goto out;
 	}
 	sk->sk_max_ack_backlog = backlog;
net/smc/smc.h +29 -0
index 1c00f1bba2cdb..268dc975249f8 100644
@@ -269,12 +269,41 @@ static inline struct smc_sock *smc_sk(const struct sock *sk)
 	return (struct smc_sock *)sk;
 }
 
+static inline void smc_init_saved_callbacks(struct smc_sock *smc)
+{
+	smc->clcsk_state_change	= NULL;
+	smc->clcsk_data_ready	= NULL;
+	smc->clcsk_write_space	= NULL;
+	smc->clcsk_error_report	= NULL;
+}
+
 static inline struct smc_sock *smc_clcsock_user_data(const struct sock *clcsk)
 {
 	return (struct smc_sock *)
 	       ((uintptr_t)clcsk->sk_user_data & ~SK_USER_DATA_NOCOPY);
 }
 
+/* save target_cb in saved_cb, and replace target_cb with new_cb */
+static inline void smc_clcsock_replace_cb(void (**target_cb)(struct sock *),
+					  void (*new_cb)(struct sock *),
+					  void (**saved_cb)(struct sock *))
+{
+	/* only save once */
+	if (!*saved_cb)
+		*saved_cb = *target_cb;
+	*target_cb = new_cb;
+}
+
+/* restore target_cb to saved_cb, and reset saved_cb to NULL */
+static inline void smc_clcsock_restore_cb(void (**target_cb)(struct sock *),
+					  void (**saved_cb)(struct sock *))
+{
+	if (!*saved_cb)
+		return;
+	*target_cb = *saved_cb;
+	*saved_cb = NULL;
+}
+
 extern struct workqueue_struct	*smc_hs_wq;	/* wq for handshake work */
 extern struct workqueue_struct	*smc_close_wq;	/* wq for close work */
 
net/smc/smc_close.c +2 -1
index bcd3ea894555d..42f9a7cf9e671 100644
@@ -212,7 +212,8 @@ int smc_close_active(struct smc_sock *smc)
 		sk->sk_state = SMC_CLOSED;
 		sk->sk_state_change(sk); /* wake up accept */
 		if (smc->clcsock && smc->clcsock->sk) {
-			smc->clcsock->sk->sk_data_ready = smc->clcsk_data_ready;
+			smc_clcsock_restore_cb(&smc->clcsock->sk->sk_data_ready,
+					       &smc->clcsk_data_ready);
 			smc->clcsock->sk->sk_user_data = NULL;
 			rc = kernel_sock_shutdown(smc->clcsock, SHUT_RDWR);
 		}
Signature · 1 line
2.51.0
author maintainer
#280 lore raw

PATCH 269/570 net/smc: Fix slab-out-of-bounds issue in fallback

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wen Gu <[email protected]>

[ Upstream commit 0558226cebee256aa3f8ec0cc5a800a10bf120a6 ]

syzbot reported a slab-out-of-bounds/use-after-free issue, which was caused by accessing an already freed smc sock in fallback-specific callback functions of clcsock.

This patch fixes the issue by restoring fallback-specific callback functions to original ones and resetting clcsock sk_user_data to NULL before freeing smc sock.

Meanwhile, this patch introduces sk_callback_lock to make the access and assignment to sk_user_data mutually exclusive.

Reported-by [email protected]
Fixes 341adeec9ada ("net/smc: Forward wakeup to smc socket waitqueue after fallback")

Stable-dep-of: 6d5e4538364b ("net/smc: fix NULL dereference and UAF in smc_tcp_syn_recv_sock()")

---
 net/smc/af_smc.c    | 80 ++++++++++++++++++++++++++++++++-------------
 net/smc/smc_close.c |  2 ++
 2 files changed, 59 insertions(+), 23 deletions(-)
net/smc/af_smc.c +57 -23
index 5c6759d2e271d..ea1a185327629 100644
@@ -170,11 +170,27 @@ struct proto smc_proto6 = {
 };
 EXPORT_SYMBOL_GPL(smc_proto6);
 
+static void smc_fback_restore_callbacks(struct smc_sock *smc)
+{
+	struct sock *clcsk = smc->clcsock->sk;
+
+	write_lock_bh(&clcsk->sk_callback_lock);
+	clcsk->sk_user_data = NULL;
+
+	smc_clcsock_restore_cb(&clcsk->sk_state_change, &smc->clcsk_state_change);
+	smc_clcsock_restore_cb(&clcsk->sk_data_ready, &smc->clcsk_data_ready);
+	smc_clcsock_restore_cb(&clcsk->sk_write_space, &smc->clcsk_write_space);
+	smc_clcsock_restore_cb(&clcsk->sk_error_report, &smc->clcsk_error_report);
+
+	write_unlock_bh(&clcsk->sk_callback_lock);
+}
+
 static void smc_restore_fallback_changes(struct smc_sock *smc)
 {
 	if (smc->clcsock->file) { /* non-accepted sockets have no file yet */
 		smc->clcsock->file->private_data = smc->sk.sk_socket;
 		smc->clcsock->file = NULL;
+		smc_fback_restore_callbacks(smc);
 	}
 }
 
@@ -659,48 +675,57 @@ static void smc_fback_forward_wakeup(struct smc_sock *smc, struct sock *clcsk,
 
 static void smc_fback_state_change(struct sock *clcsk)
 {
-	struct smc_sock *smc =
-		smc_clcsock_user_data(clcsk);
+	struct smc_sock *smc;
 
-	if (!smc)
-		return;
-	smc_fback_forward_wakeup(smc, clcsk, smc->clcsk_state_change);
+	read_lock_bh(&clcsk->sk_callback_lock);
+	smc = smc_clcsock_user_data(clcsk);
+	if (smc)
+		smc_fback_forward_wakeup(smc, clcsk,
+					 smc->clcsk_state_change);
+	read_unlock_bh(&clcsk->sk_callback_lock);
 }
 
 static void smc_fback_data_ready(struct sock *clcsk)
 {
-	struct smc_sock *smc =
-		smc_clcsock_user_data(clcsk);
+	struct smc_sock *smc;
 
-	if (!smc)
-		return;
-	smc_fback_forward_wakeup(smc, clcsk, smc->clcsk_data_ready);
+	read_lock_bh(&clcsk->sk_callback_lock);
+	smc = smc_clcsock_user_data(clcsk);
+	if (smc)
+		smc_fback_forward_wakeup(smc, clcsk,
+					 smc->clcsk_data_ready);
+	read_unlock_bh(&clcsk->sk_callback_lock);
 }
 
 static void smc_fback_write_space(struct sock *clcsk)
 {
-	struct smc_sock *smc =
-		smc_clcsock_user_data(clcsk);
+	struct smc_sock *smc;
 
-	if (!smc)
-		return;
-	smc_fback_forward_wakeup(smc, clcsk, smc->clcsk_write_space);
+	read_lock_bh(&clcsk->sk_callback_lock);
+	smc = smc_clcsock_user_data(clcsk);
+	if (smc)
+		smc_fback_forward_wakeup(smc, clcsk,
+					 smc->clcsk_write_space);
+	read_unlock_bh(&clcsk->sk_callback_lock);
 }
 
 static void smc_fback_error_report(struct sock *clcsk)
 {
-	struct smc_sock *smc =
-		smc_clcsock_user_data(clcsk);
+	struct smc_sock *smc;
 
-	if (!smc)
-		return;
-	smc_fback_forward_wakeup(smc, clcsk, smc->clcsk_error_report);
+	read_lock_bh(&clcsk->sk_callback_lock);
+	smc = smc_clcsock_user_data(clcsk);
+	if (smc)
+		smc_fback_forward_wakeup(smc, clcsk,
+					 smc->clcsk_error_report);
+	read_unlock_bh(&clcsk->sk_callback_lock);
 }
 
 static void smc_fback_replace_callbacks(struct smc_sock *smc)
 {
 	struct sock *clcsk = smc->clcsock->sk;
 
+	write_lock_bh(&clcsk->sk_callback_lock);
 	clcsk->sk_user_data = (void *)((uintptr_t)smc | SK_USER_DATA_NOCOPY);
 
 	smc_clcsock_replace_cb(&clcsk->sk_state_change, smc_fback_state_change,
@@ -711,6 +736,8 @@ static void smc_fback_replace_callbacks(struct smc_sock *smc)
 			       &smc->clcsk_write_space);
 	smc_clcsock_replace_cb(&clcsk->sk_error_report, smc_fback_error_report,
 			       &smc->clcsk_error_report);
+
+	write_unlock_bh(&clcsk->sk_callback_lock);
 }
 
 static int smc_switch_to_fallback(struct smc_sock *smc, int reason_code)
@@ -2095,17 +2122,20 @@ static void smc_tcp_listen_work(struct work_struct *work)
 
 static void smc_clcsock_data_ready(struct sock *listen_clcsock)
 {
-	struct smc_sock *lsmc =
-		smc_clcsock_user_data(listen_clcsock);
+	struct smc_sock *lsmc;
 
+	read_lock_bh(&listen_clcsock->sk_callback_lock);
+	lsmc = smc_clcsock_user_data(listen_clcsock);
 	if (!lsmc)
-		return;
+		goto out;
 	lsmc->clcsk_data_ready(listen_clcsock);
 	if (lsmc->sk.sk_state == SMC_LISTEN) {
 		sock_hold(&lsmc->sk); /* sock_put in smc_tcp_listen_work() */
 		if (!queue_work(smc_hs_wq, &lsmc->tcp_listen_work))
 			sock_put(&lsmc->sk);
 	}
+out:
+	read_unlock_bh(&listen_clcsock->sk_callback_lock);
 }
 
 static int smc_listen(struct socket *sock, int backlog)
@@ -2137,10 +2167,12 @@ static int smc_listen(struct socket *sock, int backlog)
 	/* save original sk_data_ready function and establish
 	 * smc-specific sk_data_ready function
 	 */
+	write_lock_bh(&smc->clcsock->sk->sk_callback_lock);
 	smc->clcsock->sk->sk_user_data =
 		(void *)((uintptr_t)smc | SK_USER_DATA_NOCOPY);
 	smc_clcsock_replace_cb(&smc->clcsock->sk->sk_data_ready,
 			       smc_clcsock_data_ready, &smc->clcsk_data_ready);
+	write_unlock_bh(&smc->clcsock->sk->sk_callback_lock);
 
 	/* save original ops */
 	smc->ori_af_ops = inet_csk(smc->clcsock->sk)->icsk_af_ops;
@@ -2152,9 +2184,11 @@ static int smc_listen(struct socket *sock, int backlog)
 
 	rc = kernel_listen(smc->clcsock, backlog);
 	if (rc) {
+		write_lock_bh(&smc->clcsock->sk->sk_callback_lock);
 		smc_clcsock_restore_cb(&smc->clcsock->sk->sk_data_ready,
 				       &smc->clcsk_data_ready);
 		smc->clcsock->sk->sk_user_data = NULL;
+		write_unlock_bh(&smc->clcsock->sk->sk_callback_lock);
 		goto out;
 	}
 	sk->sk_max_ack_backlog = backlog;
net/smc/smc_close.c +2 -0
index 42f9a7cf9e671..313ef522dfab4 100644
@@ -212,9 +212,11 @@ int smc_close_active(struct smc_sock *smc)
 		sk->sk_state = SMC_CLOSED;
 		sk->sk_state_change(sk); /* wake up accept */
 		if (smc->clcsock && smc->clcsock->sk) {
+			write_lock_bh(&smc->clcsock->sk->sk_callback_lock);
 			smc_clcsock_restore_cb(&smc->clcsock->sk->sk_data_ready,
 					       &smc->clcsk_data_ready);
 			smc->clcsock->sk->sk_user_data = NULL;
+			write_unlock_bh(&smc->clcsock->sk->sk_callback_lock);
 			rc = kernel_sock_shutdown(smc->clcsock, SHUT_RDWR);
 		}
 		smc_close_cleanup_listen(sk);
Signature · 1 line
2.51.0
author maintainer
#281 lore raw

PATCH 270/570 net/smc: fix NULL dereference and UAF in smc_tcp_syn_recv_sock()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jiayuan Chen <[email protected]>

[ Upstream commit 6d5e4538364b9ceb1ac2941a4deb86650afb3538 ]

Syzkaller reported a panic in smc_tcp_syn_recv_sock() [1].

smc_tcp_syn_recv_sock() is called in the TCP receive path (softirq) via icsk_af_ops->syn_recv_sock on the clcsock (TCP listening socket). It reads sk_user_data to get the smc_sock pointer. However, when the SMC listen socket is being closed concurrently, smc_close_active() sets clcsock->sk_user_data to NULL under sk_callback_lock, and then the smc_sock itself can be freed via sock_put() in smc_release().

This leads to two issues:

1) NULL pointer dereference: sk_user_data is NULL when accessed.
2) Use-after-free: sk_user_data is read as non-NULL, but the smc_sock is freed before its fields (e.g., queued_smc_hs, ori_af_ops) are accessed.

The race window looks like this (the syzkaller crash [1] triggers via the SYN cookie path: tcp_get_cookie_sock() -> smc_tcp_syn_recv_sock(), but the normal tcp_check_req() path has the same race):

CPU A (softirq) CPU B (process ctx)

tcp_v4_rcv()

    TCP_NEW_SYN_RECV:
    sk = req->rsk_listener
    sock_hold(sk)
    /* No lock on listener */
                               smc_close_active():
                                 write_lock_bh(cb_lock)
                                 sk_user_data = NULL
                                 write_unlock_bh(cb_lock)
                                 ...
                                 smc_clcsock_release()
                                 sock_put(smc->sk) x2
                                   -> smc_sock freed!
    tcp_check_req()
      smc_tcp_syn_recv_sock():
        smc = user_data(sk)
          -> NULL or dangling
        smc->queued_smc_hs
          -> crash!

Note that the clcsock and smc_sock are two independent objects with separate refcounts. TCP stack holds a reference on the clcsock, which keeps it alive, but this does NOT prevent the smc_sock from being freed.

Fix this by using RCU and refcount_inc_not_zero() to safely access smc_sock. Since smc_tcp_syn_recv_sock() is called in the TCP three-way handshake path, taking read_lock_bh on sk_callback_lock is too heavy and would not survive a SYN flood attack. Using rcu_read_lock() is much more lightweight.

- Set SOCK_RCU_FREE on the SMC listen socket so that smc_sock freeing is deferred until after the RCU grace period. This guarantees the memory is still valid when accessed inside rcu_read_lock().
- Use rcu_read_lock() to protect reading sk_user_data.
- Use refcount_inc_not_zero(&smc->sk.sk_refcnt) to pin the smc_sock. If the refcount has already reached zero (close path completed), it returns false and we bail out safely.

Note: smc_hs_congested() has a similar lockless read of sk_user_data without rcu_read_lock(), but it only checks for NULL and accesses the global smc_hs_wq, never dereferencing any smc_sock field, so it is not affected.

Reproducer was verified with mdelay injection and smc_run, the issue no longer occurs with this patch applied.

[1] https://syzkaller.appspot.com/bug?extid=827ae2bfb3a3529333e9

---
 net/smc/af_smc.c    | 23 +++++++++++++++++------
 net/smc/smc.h       |  5 +++++
 net/smc/smc_close.c |  2 +-
 3 files changed, 23 insertions(+), 7 deletions(-)
net/smc/af_smc.c +17 -6
index ea1a185327629..5425c46a2e7c7 100644
@@ -81,7 +81,14 @@ static struct sock *smc_tcp_syn_recv_sock(const struct sock *sk,
 	struct smc_sock *smc;
 	struct sock *child;
 
-	smc = smc_clcsock_user_data(sk);
+	rcu_read_lock();
+	smc = smc_clcsock_user_data_rcu(sk);
+	if (!smc || !refcount_inc_not_zero(&smc->sk.sk_refcnt)) {
+		rcu_read_unlock();
+		smc = NULL;
+		goto drop;
+	}
+	rcu_read_unlock();
 
 	if (READ_ONCE(sk->sk_ack_backlog) + atomic_read(&smc->queued_smc_hs) >
 				sk->sk_max_ack_backlog)
@@ -103,11 +110,14 @@ static struct sock *smc_tcp_syn_recv_sock(const struct sock *sk,
 		if (inet_csk(child)->icsk_af_ops == inet_csk(sk)->icsk_af_ops)
 			inet_csk(child)->icsk_af_ops = smc->ori_af_ops;
 	}
+	sock_put(&smc->sk);
 	return child;
 
 drop:
 	dst_release(dst);
 	tcp_listendrop(sk);
+	if (smc)
+		sock_put(&smc->sk);
 	return NULL;
 }
 
@@ -175,7 +185,7 @@ static void smc_fback_restore_callbacks(struct smc_sock *smc)
 	struct sock *clcsk = smc->clcsock->sk;
 
 	write_lock_bh(&clcsk->sk_callback_lock);
-	clcsk->sk_user_data = NULL;
+	rcu_assign_sk_user_data(clcsk, NULL);
 
 	smc_clcsock_restore_cb(&clcsk->sk_state_change, &smc->clcsk_state_change);
 	smc_clcsock_restore_cb(&clcsk->sk_data_ready, &smc->clcsk_data_ready);
@@ -726,7 +736,7 @@ static void smc_fback_replace_callbacks(struct smc_sock *smc)
 	struct sock *clcsk = smc->clcsock->sk;
 
 	write_lock_bh(&clcsk->sk_callback_lock);
-	clcsk->sk_user_data = (void *)((uintptr_t)smc | SK_USER_DATA_NOCOPY);
+	__rcu_assign_sk_user_data_with_flags(clcsk, smc, SK_USER_DATA_NOCOPY);
 
 	smc_clcsock_replace_cb(&clcsk->sk_state_change, smc_fback_state_change,
 			       &smc->clcsk_state_change);
@@ -2168,8 +2178,8 @@ static int smc_listen(struct socket *sock, int backlog)
 	 * smc-specific sk_data_ready function
 	 */
 	write_lock_bh(&smc->clcsock->sk->sk_callback_lock);
-	smc->clcsock->sk->sk_user_data =
-		(void *)((uintptr_t)smc | SK_USER_DATA_NOCOPY);
+	__rcu_assign_sk_user_data_with_flags(smc->clcsock->sk, smc,
+					     SK_USER_DATA_NOCOPY);
 	smc_clcsock_replace_cb(&smc->clcsock->sk->sk_data_ready,
 			       smc_clcsock_data_ready, &smc->clcsk_data_ready);
 	write_unlock_bh(&smc->clcsock->sk->sk_callback_lock);
@@ -2187,10 +2197,11 @@ static int smc_listen(struct socket *sock, int backlog)
 		write_lock_bh(&smc->clcsock->sk->sk_callback_lock);
 		smc_clcsock_restore_cb(&smc->clcsock->sk->sk_data_ready,
 				       &smc->clcsk_data_ready);
-		smc->clcsock->sk->sk_user_data = NULL;
+		rcu_assign_sk_user_data(smc->clcsock->sk, NULL);
 		write_unlock_bh(&smc->clcsock->sk->sk_callback_lock);
 		goto out;
 	}
+	sock_set_flag(sk, SOCK_RCU_FREE);
 	sk->sk_max_ack_backlog = backlog;
 	sk->sk_ack_backlog = 0;
 	sk->sk_state = SMC_LISTEN;
net/smc/smc.h +5 -0
index 268dc975249f8..6455371430a3c 100644
@@ -283,6 +283,11 @@ static inline struct smc_sock *smc_clcsock_user_data(const struct sock *clcsk)
 	       ((uintptr_t)clcsk->sk_user_data & ~SK_USER_DATA_NOCOPY);
 }
 
+static inline struct smc_sock *smc_clcsock_user_data_rcu(const struct sock *clcsk)
+{
+	return (struct smc_sock *)rcu_dereference_sk_user_data(clcsk);
+}
+
 /* save target_cb in saved_cb, and replace target_cb with new_cb */
 static inline void smc_clcsock_replace_cb(void (**target_cb)(struct sock *),
 					  void (*new_cb)(struct sock *),
net/smc/smc_close.c +1 -1
index 313ef522dfab4..e156039ff1e50 100644
@@ -215,7 +215,7 @@ int smc_close_active(struct smc_sock *smc)
 			write_lock_bh(&smc->clcsock->sk->sk_callback_lock);
 			smc_clcsock_restore_cb(&smc->clcsock->sk->sk_data_ready,
 					       &smc->clcsk_data_ready);
-			smc->clcsock->sk->sk_user_data = NULL;
+			rcu_assign_sk_user_data(smc->clcsock->sk, NULL);
 			write_unlock_bh(&smc->clcsock->sk->sk_callback_lock);
 			rc = kernel_sock_shutdown(smc->clcsock, SHUT_RDWR);
 		}
Signature · 1 line
2.51.0
author maintainer
#282 lore raw

PATCH 271/570 net: usb: aqc111: Do not perform PM inside suspend callback

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Nikola Z. Ivanov <[email protected]>

[ Upstream commit 069c8f5aebe4d5224cf62acc7d4b3486091c658a ]

syzbot reports "task hung in rpm_resume"

This is caused by aqc111_suspend calling
the PM variant of its write_cmd routine.

The simplified call trace looks like this:

rpm_suspend()
usb_suspend_both() - here udev->dev.power.runtime_status == RPM_SUSPENDING
aqc111_suspend() - called for the usb device interface
aqc111_write32_cmd()
usb_autopm_get_interface()
pm_runtime_resume_and_get()
rpm_resume() - here we call rpm_resume() on our parent
rpm_resume() - Here we wait for a status change that will never happen.

At this point we block another task which holds rtnl_lock and locks up the whole networking stack.

Fix this by replacing the write_cmd calls with their _nopm variants

---
 drivers/net/usb/aqc111.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
drivers/net/usb/aqc111.c +6 -6
index 00aba7e1d0b95..81093c4fb8194 100644
@@ -1400,14 +1400,14 @@ static int aqc111_suspend(struct usb_interface *intf, pm_message_t message)
 		aqc111_write16_cmd_nopm(dev, AQ_ACCESS_MAC,
 					SFR_MEDIUM_STATUS_MODE, 2, &reg16);
 
-		aqc111_write_cmd(dev, AQ_WOL_CFG, 0, 0,
-				 WOL_CFG_SIZE, &wol_cfg);
-		aqc111_write32_cmd(dev, AQ_PHY_OPS, 0, 0,
-				   &aqc111_data->phy_cfg);
+		aqc111_write_cmd_nopm(dev, AQ_WOL_CFG, 0, 0,
+				      WOL_CFG_SIZE, &wol_cfg);
+		aqc111_write32_cmd_nopm(dev, AQ_PHY_OPS, 0, 0,
+					&aqc111_data->phy_cfg);
 	} else {
 		aqc111_data->phy_cfg |= AQ_LOW_POWER;
-		aqc111_write32_cmd(dev, AQ_PHY_OPS, 0, 0,
-				   &aqc111_data->phy_cfg);
+		aqc111_write32_cmd_nopm(dev, AQ_PHY_OPS, 0, 0,
+					&aqc111_data->phy_cfg);
 
 		/* Disable RX path */
 		aqc111_read16_cmd_nopm(dev, AQ_ACCESS_MAC,
Signature · 1 line
2.51.0
author maintainer
#283 lore raw

PATCH 272/570 igc: fix missing update of skb->tail in igc_xmit_frame()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kohei Enju <[email protected]>

[ Upstream commit 0ffba246652faf4a36aedc66059c2f94e4c83ea5 ]

igc_xmit_frame() misses updating skb->tail when the packet size is shorter than the minimum one.
Use skb_put_padto() in alignment with other Intel Ethernet drivers.

Fixes 0507ef8a0372 ("igc: Add transmit and receive fastpath and interrupt handlers")
---
 drivers/net/ethernet/intel/igc/igc_main.c | 7 ++-----
 1 file changed, 2 insertions(+), 5 deletions(-)
drivers/net/ethernet/intel/igc/igc_main.c +2 -5
index 6a9ad4231b0c2..d2825170c1e1d 100644
@@ -1666,11 +1666,8 @@ static netdev_tx_t igc_xmit_frame(struct sk_buff *skb,
 	/* The minimum packet size with TCTL.PSP set is 17 so pad the skb
 	 * in order to meet this minimum size requirement.
 	 */
-	if (skb->len < 17) {
-		if (skb_padto(skb, 17))
-			return NETDEV_TX_OK;
-		skb->len = 17;
-	}
+	if (skb_put_padto(skb, 17))
+		return NETDEV_TX_OK;
 
 	return igc_xmit_frame_ring(skb, igc_tx_queue_mapping(adapter, skb));
 }
Signature · 1 line
2.51.0
author maintainer
#284 lore raw

PATCH 273/570 wifi: mac80211: fix NULL deref in mesh_matches_local()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Xiang Mei <[email protected]>

[ Upstream commit c73bb9a2d33bf81f6eecaa0f474b6c6dbe9855bd ]

mesh_matches_local() unconditionally dereferences ie->mesh_config to compare mesh configuration parameters. When called from mesh_rx_csa_frame(), the parsed action-frame elements may not contain a Mesh Configuration IE, leaving ie->mesh_config NULL and triggering a kernel NULL pointer dereference.

The other two callers are already safe:
- ieee80211_mesh_rx_bcn_presp() checks !elems->mesh_config before calling mesh_matches_local()
- mesh_plink_get_event() is only reached through

    mesh_process_plink_frame(), which checks !elems->mesh_config, too

mesh_rx_csa_frame() is the only caller that passes raw parsed elements to mesh_matches_local() without guarding mesh_config. An adjacent attacker can exploit this by sending a crafted CSA action frame that includes a valid Mesh ID IE but omits the Mesh Configuration IE, crashing the kernel.

The captured crash log:

Oops: general protection fault, probably for non-canonical address ...
KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]
Workqueue: events_unbound cfg80211_wiphy_work
[...]
Call Trace:
 <TASK>
 ? __pfx_mesh_matches_local (net/mac80211/mesh.c:65)
 ieee80211_mesh_rx_queued_mgmt (net/mac80211/mesh.c:1686)
 [...]
 ieee80211_iface_work (net/mac80211/iface.c:1754 net/mac80211/iface.c:1802)
 [...]
 cfg80211_wiphy_work (net/wireless/core.c:426)
 process_one_work (net/kernel/workqueue.c:3280)
 ? assign_work (net/kernel/workqueue.c:1219)
 worker_thread (net/kernel/workqueue.c:3352)
 ? __pfx_worker_thread (net/kernel/workqueue.c:3385)
 kthread (net/kernel/kthread.c:436)
 [...]
 ret_from_fork_asm (net/arch/x86/entry/entry_64.S:255)
 </TASK>

This patch adds a NULL check for ie->mesh_config at the top of mesh_matches_local() to return false early when the Mesh Configuration IE is absent.

---
 net/mac80211/mesh.c | 3 +++
 1 file changed, 3 insertions(+)
net/mac80211/mesh.c +3 -0
index e75f53f08b611..167b0625b1a17 100644
@@ -75,6 +75,9 @@ bool mesh_matches_local(struct ieee80211_sub_if_data *sdata,
 	 *   - MDA enabled
 	 * - Power management control on fc
 	 */
+	if (!ie->mesh_config)
+		return false;
+
 	if (!(ifmsh->mesh_id_len == ie->mesh_id_len &&
 	     memcmp(ifmsh->mesh_id, ie->mesh_id, ie->mesh_id_len) == 0 &&
 	     (ifmsh->mesh_pp_id == ie->mesh_config->meshconf_psel) &&
Signature · 1 line
2.51.0
author maintainer
#285 lore raw

PATCH 274/570 wifi: wlcore: Return -ENOMEM instead of -EAGAIN if there is not enough headroom

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Guenter Roeck <[email protected]>

[ Upstream commit deb353d9bb009638b7762cae2d0b6e8fdbb41a69 ]

Since upstream commit e75665dd0968 ("wifi: wlcore: ensure skb headroom before skb_push"), wl1271_tx_allocate() and with it wl1271_prepare_tx_frame() returns -EAGAIN if pskb_expand_head() fails. However, in wlcore_tx_work_locked(), a return value of -EAGAIN from wl1271_prepare_tx_frame() is interpreted as the aggregation buffer being full. This causes the code to flush the buffer, put the skb back at the head of the queue, and immediately retry the same skb in a tight while loop.

Because wlcore_tx_work_locked() holds wl->mutex, and the retry happens immediately with GFP_ATOMIC, this will result in an infinite loop and a CPU soft lockup. Return -ENOMEM instead so the packet is dropped and the loop terminates.

The problem was found by an experimental code review agent based on gemini-3.1-pro while reviewing backports into v6.18.y.

Assisted-by Gemini:gemini-3.1-pro
Fixes e75665dd0968 ("wifi: wlcore: ensure skb headroom before skb_push")
Cc Peter Astrand <[email protected]>
---
 drivers/net/wireless/ti/wlcore/tx.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/ti/wlcore/tx.c +1 -1
index e86cc3425e997..ac1411db8e5a8 100644
@@ -213,7 +213,7 @@ static int wl1271_tx_allocate(struct wl1271 *wl, struct wl12xx_vif *wlvif,
 		if (skb_headroom(skb) < (total_len - skb->len) &&
 		    pskb_expand_head(skb, (total_len - skb->len), 0, GFP_ATOMIC)) {
 			wl1271_free_tx_id(wl, id);
-			return -EAGAIN;
+			return -ENOMEM;
 		}
 		desc = skb_push(skb, total_len - skb->len);
 
Signature · 1 line
2.51.0
author maintainer
#286 lore raw

PATCH 275/570 ACPI: processor: Fix previous acpi_processor_errata_piix4() fix

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafael J. Wysocki <[email protected]>

[ Upstream commit bf504b229cb8d534eccbaeaa23eba34c05131e25 ]

After commi f132e089fe89 ("ACPI: processor: Fix NULL-pointer dereference in acpi_processor_errata_piix4()"), device pointers may be dereferenced after dropping references to the device objects pointed to by them, which may cause a use-after-free to occur.

Moreover, debug messages about enabling the errata may be printed if the errata flags corresponding to them are unset.

Address all of these issues by moving message printing to the points in the code where the errata flags are set.

---
 drivers/acpi/acpi_processor.c | 15 ++++++++-------
 1 file changed, 8 insertions(+), 7 deletions(-)
drivers/acpi/acpi_processor.c +8 -7
index 669398045c0fd..07acdaee6ce5c 100644
@@ -96,6 +96,10 @@ static int acpi_processor_errata_piix4(struct pci_dev *dev)
 				     PCI_ANY_ID, PCI_ANY_ID, NULL);
 		if (ide_dev) {
 			errata.piix4.bmisx = pci_resource_start(ide_dev, 4);
+			if (errata.piix4.bmisx)
+				dev_dbg(&ide_dev->dev,
+					"Bus master activity detection (BM-IDE) erratum enabled\n");
+
 			pci_dev_put(ide_dev);
 		}
 
@@ -114,20 +118,17 @@ static int acpi_processor_errata_piix4(struct pci_dev *dev)
 		if (isa_dev) {
 			pci_read_config_byte(isa_dev, 0x76, &value1);
 			pci_read_config_byte(isa_dev, 0x77, &value2);
-			if ((value1 & 0x80) || (value2 & 0x80))
+			if ((value1 & 0x80) || (value2 & 0x80)) {
 				errata.piix4.fdma = 1;
+				dev_dbg(&isa_dev->dev,
+					"Type-F DMA livelock erratum (C3 disabled)\n");
+			}
 			pci_dev_put(isa_dev);
 		}
 
 		break;
 	}
 
-	if (ide_dev)
-		dev_dbg(&ide_dev->dev, "Bus master activity detection (BM-IDE) erratum enabled\n");
-
-	if (isa_dev)
-		dev_dbg(&isa_dev->dev, "Type-F DMA livelock erratum (C3 disabled)\n");
-
 	return 0;
 }
 
Signature · 1 line
2.51.0
author maintainer
#287 lore raw

PATCH 276/570 net: macb: fix uninitialized rx_fs_lock

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Fedor Pchelkin <[email protected]>

[ Upstream commit 34b11cc56e4369bc08b1f4c4a04222d75ed596ce ]

If hardware doesn't support RX Flow Filters, rx_fs_lock spinlock is not initialized leading to the following assertion splat triggerable via set_rxnfc callback.

INFO: trying to register non-static key.
The code is fine but needs lockdep annotation, or maybe you didn't initialize this object before use? turning off the locking correctness validator.

CPU: 1 PID: 949 Comm: syz.0.6 Not tainted 6.1.164+ #113
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.16.1-0-g3208b098f51a-prebuilt.qemu.org 04/01/2014
Call Trace:
 <TASK>
 __dump_stack lib/dump_stack.c:88 [inline]
 dump_stack_lvl+0x8d/0xba lib/dump_stack.c:106
 assign_lock_key kernel/locking/lockdep.c:974 [inline]
 register_lock_class+0x141b/0x17f0 kernel/locking/lockdep.c:1287
 __lock_acquire+0x74f/0x6c40 kernel/locking/lockdep.c:4928
 lock_acquire kernel/locking/lockdep.c:5662 [inline]
 lock_acquire+0x190/0x4b0 kernel/locking/lockdep.c:5627
 __raw_spin_lock_irqsave include/linux/spinlock_api_smp.h:110 [inline]
 _raw_spin_lock_irqsave+0x33/0x50 kernel/locking/spinlock.c:162
 gem_del_flow_filter drivers/net/ethernet/cadence/macb_main.c:3562 [inline]
 gem_set_rxnfc+0x533/0xac0 drivers/net/ethernet/cadence/macb_main.c:3667
 ethtool_set_rxnfc+0x18c/0x280 net/ethtool/ioctl.c:961
 __dev_ethtool net/ethtool/ioctl.c:2956 [inline]
 dev_ethtool+0x229c/0x6290 net/ethtool/ioctl.c:3095
 dev_ioctl+0x637/0x1070 net/core/dev_ioctl.c:510
 sock_do_ioctl+0x20d/0x2c0 net/socket.c:1215
 sock_ioctl+0x577/0x6d0 net/socket.c:1320
 vfs_ioctl fs/ioctl.c:51 [inline]
 __do_sys_ioctl fs/ioctl.c:870 [inline]
 __se_sys_ioctl fs/ioctl.c:856 [inline]
 __x64_sys_ioctl+0x18c/0x210 fs/ioctl.c:856
 do_syscall_x64 arch/x86/entry/common.c:46 [inline]
 do_syscall_64+0x35/0x80 arch/x86/entry/common.c:76
 entry_SYSCALL_64_after_hwframe+0x6e/0xd8

A more straightforward solution would be to always initialize rx_fs_lock, just like rx_fs_list. However, in this case the driver set_rxnfc callback would return with a rather confusing error code, e.g. -EINVAL. So deny set_rxnfc attempts directly if the RX filtering feature is not supported by hardware.

Fixes ae8223de3df5 ("net: macb: Added support for RX filtering")
---
 drivers/net/ethernet/cadence/macb_main.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/net/ethernet/cadence/macb_main.c +3 -0
index d4a4d72460a42..6a3e9082bda8c 100644
@@ -3572,6 +3572,9 @@ static int gem_set_rxnfc(struct net_device *netdev, struct ethtool_rxnfc *cmd)
 	struct macb *bp = netdev_priv(netdev);
 	int ret;
 
+	if (!(netdev->hw_features & NETIF_F_NTUPLE))
+		return -EOPNOTSUPP;
+
 	switch (cmd->cmd) {
 	case ETHTOOL_SRXCLSRLINS:
 		if ((cmd->fs.location >= bp->max_tuples)
Signature · 1 line
2.51.0
author maintainer
#288 lore raw

PATCH 277/570 udp_tunnel: fix NULL deref caused by udp_sock_create6 when CONFIG_IPV6=n

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Xiang Mei <[email protected]>

[ Upstream commit b3a6df291fecf5f8a308953b65ca72b7fc9e015d ]

When CONFIG_IPV6 is disabled, the udp_sock_create6() function returns 0 (success) without actually creating a socket. Callers such as fou_create() then proceed to dereference the uninitialized socket pointer, resulting in a NULL pointer dereference.

The captured NULL deref crash:

  BUG: kernel NULL pointer dereference, address: 0000000000000018
  RIP: 0010:fou_nl_add_doit (net/ipv4/fou_core.c:590 net/ipv4/fou_core.c:764)
  [...]
  Call Trace:
    <TASK>
    genl_family_rcv_msg_doit.constprop.0 (net/netlink/genetlink.c:1114)
    genl_rcv_msg (net/netlink/genetlink.c:1194 net/netlink/genetlink.c:1209)
    [...]
    netlink_rcv_skb (net/netlink/af_netlink.c:2550)
    genl_rcv (net/netlink/genetlink.c:1219)
    netlink_unicast (net/netlink/af_netlink.c:1319 net/netlink/af_netlink.c:1344)
    netlink_sendmsg (net/netlink/af_netlink.c:1894)
    __sock_sendmsg (net/socket.c:727 (discriminator 1) net/socket.c:742 (discriminator 1))
    __sys_sendto (./include/linux/file.h:62 (discriminator 1) ./include/linux/file.h:83 (discriminator 1) net/socket.c:2183 (discriminator 1))
    __x64_sys_sendto (net/socket.c:2213 (discriminator 1) net/socket.c:2209 (discriminator 1) net/socket.c:2209 (discriminator 1))
    do_syscall_64 (arch/x86/entry/syscall_64.c:63 (discriminator 1) arch/x86/entry/syscall_64.c:94 (discriminator 1))
    entry_SYSCALL_64_after_hwframe (net/arch/x86/entry/entry_64.S:130)

This patch makes udp_sock_create6 return -EPFNOSUPPORT instead, so callers correctly take their error paths. There is only one caller of the vulnerable function and only privileged users can trigger it.

Fixes fd384412e199b ("udp_tunnel: Seperate ipv6 functions into its own file.")
---
 include/net/udp_tunnel.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
include/net/udp_tunnel.h +1 -1
index 72394f441dad8..b6af537abdc5a 100644
@@ -47,7 +47,7 @@ int udp_sock_create6(struct net *net, struct udp_port_cfg *cfg,
 static inline int udp_sock_create6(struct net *net, struct udp_port_cfg *cfg,
 				   struct socket **sockp)
 {
-	return 0;
+	return -EPFNOSUPPORT;
 }
 #endif
 
Signature · 1 line
2.51.0
author maintainer
#289 lore raw

PATCH 278/570 net: bonding: fix NULL deref in bond_debug_rlb_hash_show

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Xiang Mei <[email protected]>

[ Upstream commit 605b52497bf89b3b154674deb135da98f916e390 ]

rlb_clear_slave intentionally keeps RLB hash-table entries on the rx_hashtbl_used_head list with slave set to NULL when no replacement slave is available. However, bond_debug_rlb_hash_show visites client_info->slave without checking if it's NULL.

Other used-list iterators in bond_alb.c already handle this NULL-slave state safely:

- rlb_update_client returns early on !client_info->slave
- rlb_req_update_slave_clients, rlb_clear_slave, and rlb_rebalance compare slave values before visiting
- lb_req_update_subnet_clients continues if slave is NULL

The following NULL deref crash can be trigger in
bond_debug_rlb_hash_show:

[    1.289791] BUG: kernel NULL pointer dereference, address: 0000000000000000
[    1.292058] RIP: 0010:bond_debug_rlb_hash_show (drivers/net/bonding/bond_debugfs.c:41)
[    1.293101] RSP: 0018:ffffc900004a7d00 EFLAGS: 00010286
[    1.293333] RAX: 0000000000000000 RBX: ffff888102b48200 RCX: ffff888102b48204
[    1.293631] RDX: ffff888102b48200 RSI: ffffffff839daad5 RDI: ffff888102815078
[    1.293924] RBP: ffff888102815078 R08: ffff888102b4820e R09: 0000000000000000
[    1.294267] R10: 0000000000000000 R11: 0000000000000000 R12: ffff888100f929c0
[    1.294564] R13: ffff888100f92a00 R14: 0000000000000001 R15: ffffc900004a7ed8
[    1.294864] FS:  0000000001395380(0000) GS:ffff888196e75000(0000) knlGS:0000000000000000
[    1.295239] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[    1.295480] CR2: 0000000000000000 CR3: 0000000102adc004 CR4: 0000000000772ef0
[    1.295897] Call Trace:
[    1.296134]  seq_read_iter (fs/seq_file.c:231)
[    1.296341]  seq_read (fs/seq_file.c:164)
[    1.296493]  full_proxy_read (fs/debugfs/file.c:378 (discriminator 1))
[    1.296658]  vfs_read (fs/read_write.c:572)
[    1.296981]  ksys_read (fs/read_write.c:717)
[    1.297132]  do_syscall_64 (arch/x86/entry/syscall_64.c:63 (discriminator 1) arch/x86/entry/syscall_64.c:94 (discriminator 1))
[    1.297325]  entry_SYSCALL_64_after_hwframe (arch/x86/entry/entry_64.S:130)

Add a NULL check and print "(none)" for entries with no assigned slave.

Fixes caafa84251b88 ("bonding: add the debugfs interface to see RLB hash table")
---
 drivers/net/bonding/bond_debugfs.c | 16 +++++++++++-----
 1 file changed, 11 insertions(+), 5 deletions(-)
drivers/net/bonding/bond_debugfs.c +11 -5
index 5940945266489..624bf1f745266 100644
@@ -34,11 +34,17 @@ static int bond_debug_rlb_hash_show(struct seq_file *m, void *v)
 	for (; hash_index != RLB_NULL_INDEX;
 	     hash_index = client_info->used_next) {
 		client_info = &(bond_info->rx_hashtbl[hash_index]);
-		seq_printf(m, "%-15pI4 %-15pI4 %-17pM %s\n",
-			&client_info->ip_src,
-			&client_info->ip_dst,
-			&client_info->mac_dst,
-			client_info->slave->dev->name);
+		if (client_info->slave)
+			seq_printf(m, "%-15pI4 %-15pI4 %-17pM %s\n",
+				   &client_info->ip_src,
+				   &client_info->ip_dst,
+				   &client_info->mac_dst,
+				   client_info->slave->dev->name);
+		else
+			seq_printf(m, "%-15pI4 %-15pI4 %-17pM (none)\n",
+				   &client_info->ip_src,
+				   &client_info->ip_dst,
+				   &client_info->mac_dst);
 	}
 
 	spin_unlock_bh(&bond->mode_lock);
Signature · 1 line
2.51.0
author maintainer
#290 lore raw

PATCH 279/570 nfnetlink_osf: validate individual option lengths in fingerprints

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Weiming Shi <[email protected]>

[ Upstream commit dbdfaae9609629a9569362e3b8f33d0a20fd783c ]

nfnl_osf_add_callback() validates opt_num bounds and string NUL-termination but does not check individual option length fields. A zero-length option causes nf_osf_match_one() to enter the option matching loop even when foptsize sums to zero, which matches packets with no TCP options where ctx->optp is NULL:

 Oops: general protection fault
 KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]
 RIP: 0010:nf_osf_match_one (net/netfilter/nfnetlink_osf.c:98)
 Call Trace:
  nf_osf_match (net/netfilter/nfnetlink_osf.c:227)
  xt_osf_match_packet (net/netfilter/xt_osf.c:32)
  ipt_do_table (net/ipv4/netfilter/ip_tables.c:293)
  nf_hook_slow (net/netfilter/core.c:623)
  ip_local_deliver (net/ipv4/ip_input.c:262)
  ip_rcv (net/ipv4/ip_input.c:573)

Additionally, an MSS option (kind=2) with length < 4 causes out-of-bounds reads when nf_osf_match_one() unconditionally accesses optp[2] and optp[3] for MSS value extraction. While RFC 9293 section 3.2 specifies that the MSS option is always exactly 4 bytes (Kind=2, Length=4), the check uses "< 4" rather than "!= 4" because lengths greater than 4 do not cause memory safety issues -- the buffer is guaranteed to be at least foptsize bytes by the ctx->optsize == foptsize check.

Reject fingerprints where any option has zero length, or where an MSS option has length less than 4, at add time rather than trusting these values in the packet matching hot path.

Fixes 11eeef41d5f6 ("netfilter: passive OS fingerprint xtables match")
---
 net/netfilter/nfnetlink_osf.c | 13 +++++++++++++
 1 file changed, 13 insertions(+)
net/netfilter/nfnetlink_osf.c +13 -0
index 50723ba082890..da9d5d6de98f4 100644
@@ -302,7 +302,9 @@ static int nfnl_osf_add_callback(struct sk_buff *skb,
 {
 	struct nf_osf_user_finger *f;
 	struct nf_osf_finger *kf = NULL, *sf;
+	unsigned int tot_opt_len = 0;
 	int err = 0;
+	int i;
 
 	if (!capable(CAP_NET_ADMIN))
 		return -EPERM;
@@ -318,6 +320,17 @@ static int nfnl_osf_add_callback(struct sk_buff *skb,
 	if (f->opt_num > ARRAY_SIZE(f->opt))
 		return -EINVAL;
 
+	for (i = 0; i < f->opt_num; i++) {
+		if (!f->opt[i].length || f->opt[i].length > MAX_IPOPTLEN)
+			return -EINVAL;
+		if (f->opt[i].kind == OSFOPT_MSS && f->opt[i].length < 4)
+			return -EINVAL;
+
+		tot_opt_len += f->opt[i].length;
+		if (tot_opt_len > MAX_IPOPTLEN)
+			return -EINVAL;
+	}
+
 	if (!memchr(f->genre, 0, MAXGENRELEN) ||
 	    !memchr(f->subtype, 0, MAXGENRELEN) ||
 	    !memchr(f->version, 0, MAXGENRELEN))
Signature · 1 line
2.51.0
author maintainer
#291 lore raw

PATCH 280/570 net: mvpp2: guard flow control update with global_tx_fc in buffer switching

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Muhammad Hammad Ijaz <[email protected]>

[ Upstream commit 8a63baadf08453f66eb582fdb6dd234f72024723 ]

mvpp2_bm_switch_buffers() unconditionally calls mvpp2_bm_pool_update_priv_fc() when switching between per-cpu and shared buffer pool modes. This function programs CM3 flow control registers via mvpp2_cm3_read()/mvpp2_cm3_write(), which dereference priv->cm3_base without any NULL check.

When the CM3 SRAM resource is not present in the device tree (the third reg entry added by commit 60523583b07c ("dts: marvell: add CM3 SRAM memory to cp11x ethernet device tree")), priv->cm3_base remains NULL and priv->global_tx_fc is false. Any operation that triggers mvpp2_bm_switch_buffers(), for example an MTU change that crosses the jumbo frame threshold, will crash:

  Unable to handle kernel NULL pointer dereference at
  virtual address 0000000000000000
  Mem abort info:
    ESR = 0x0000000096000006
    EC = 0x25: DABT (current EL), IL = 32 bits
  pc : readl+0x0/0x18
  lr : mvpp2_cm3_read.isra.0+0x14/0x20
  Call trace:
   readl+0x0/0x18
   mvpp2_bm_pool_update_fc+0x40/0x12c
   mvpp2_bm_pool_update_priv_fc+0x94/0xd8
   mvpp2_bm_switch_buffers.isra.0+0x80/0x1c0
   mvpp2_change_mtu+0x140/0x380
   __dev_set_mtu+0x1c/0x38
   dev_set_mtu_ext+0x78/0x118
   dev_set_mtu+0x48/0xa8
   dev_ifsioc+0x21c/0x43c
   dev_ioctl+0x2d8/0x42c
   sock_ioctl+0x314/0x378

Every other flow control call site in the driver already guards hardware access with either priv->global_tx_fc or port->tx_fc. mvpp2_bm_switch_buffers() is the only place that omits this check.

Add the missing priv->global_tx_fc guard to both the disable and re-enable calls in mvpp2_bm_switch_buffers(), consistent with the rest of the driver.

Fixes 3a616b92a9d1 ("net: mvpp2: Add TX flow control support for jumbo frames")
---
 drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c +2 -2
index 7fa880e62d096..fdfdd55fdb1dc 100644
@@ -5006,7 +5006,7 @@ static int mvpp2_bm_switch_buffers(struct mvpp2 *priv, bool percpu)
 	if (priv->percpu_pools)
 		numbufs = port->nrxqs * 2;
 
-	if (change_percpu)
+	if (change_percpu && priv->global_tx_fc)
 		mvpp2_bm_pool_update_priv_fc(priv, false);
 
 	for (i = 0; i < numbufs; i++)
@@ -5023,7 +5023,7 @@ static int mvpp2_bm_switch_buffers(struct mvpp2 *priv, bool percpu)
 			mvpp2_open(port->dev);
 	}
 
-	if (change_percpu)
+	if (change_percpu && priv->global_tx_fc)
 		mvpp2_bm_pool_update_priv_fc(priv, true);
 
 	return 0;
Signature · 1 line
2.51.0
author maintainer
#292 lore raw

PATCH 281/570 net: dsa: bcm_sf2: fix missing clk_disable_unprepare() in error paths

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Anas Iqbal <[email protected]>

[ Upstream commit b48731849609cbd8c53785a48976850b443153fd ]

Smatch reports:
drivers/net/dsa/bcm_sf2.c:997 bcm_sf2_sw_resume() warn:
'priv->clk' from clk_prepare_enable() not released on lines: 983,990.

The clock enabled by clk_prepare_enable() in bcm_sf2_sw_resume() is not released if bcm_sf2_sw_rst() or bcm_sf2_cfp_resume() fails.

Add the missing clk_disable_unprepare() calls in the error paths to properly release the clock resource.

---
 drivers/net/dsa/bcm_sf2.c | 8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)
drivers/net/dsa/bcm_sf2.c +6 -2
index f259b0add5b2e..6105f4d8faf06 100644
@@ -962,15 +962,19 @@ static int bcm_sf2_sw_resume(struct dsa_switch *ds)
 	ret = bcm_sf2_sw_rst(priv);
 	if (ret) {
 		pr_err("%s: failed to software reset switch\n", __func__);
+		if (!priv->wol_ports_mask)
+			clk_disable_unprepare(priv->clk);
 		return ret;
 	}
 
 	bcm_sf2_crossbar_setup(priv);
 
 	ret = bcm_sf2_cfp_resume(ds);
-	if (ret)
+	if (ret) {
+		if (!priv->wol_ports_mask)
+			clk_disable_unprepare(priv->clk);
 		return ret;
-
+	}
 	if (priv->hw_params.num_gphy == 1)
 		bcm_sf2_gphy_enable_set(ds, true);
 
Signature · 1 line
2.51.0
author maintainer
#293 lore raw

PATCH 282/570 icmp: fix NULL pointer dereference in icmp_tag_validation()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Weiming Shi <[email protected]>

[ Upstream commit 614aefe56af8e13331e50220c936fc0689cf5675 ]

icmp_tag_validation() unconditionally dereferences the result of rcu_dereference(inet_protos[proto]) without checking for NULL. The inet_protos[] array is sparse -- only about 15 of 256 protocol numbers have registered handlers. When ip_no_pmtu_disc is set to 3 (hardened PMTU mode) and the kernel receives an ICMP Fragmentation Needed error with a quoted inner IP header containing an unregistered protocol number, the NULL dereference causes a kernel panic in softirq context.

 Oops: general protection fault, probably for non-canonical address 0xdffffc0000000002: 0000 [#1] SMP KASAN NOPTI
 KASAN: null-ptr-deref in range [0x0000000000000010-0x0000000000000017]
 RIP: 0010:icmp_unreach (net/ipv4/icmp.c:1085 net/ipv4/icmp.c:1143)
 Call Trace:
  <IRQ>
  icmp_rcv (net/ipv4/icmp.c:1527)
  ip_protocol_deliver_rcu (net/ipv4/ip_input.c:207)
  ip_local_deliver_finish (net/ipv4/ip_input.c:242)
  ip_local_deliver (net/ipv4/ip_input.c:262)
  ip_rcv (net/ipv4/ip_input.c:573)
  __netif_receive_skb_one_core (net/core/dev.c:6164)
  process_backlog (net/core/dev.c:6628)
  handle_softirqs (kernel/softirq.c:561)
  </IRQ>

Add a NULL check before accessing icmp_strict_tag_validation. If the protocol has no registered handler, return false since it cannot perform strict tag validation.

Fixes 8ed1dc44d3e9 ("ipv4: introduce hardened ip_no_pmtu_disc mode")
---
 net/ipv4/icmp.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
net/ipv4/icmp.c +3 -1
index 8a70e51654264..0215e2510670a 100644
@@ -845,10 +845,12 @@ static void icmp_socket_deliver(struct sk_buff *skb, u32 info)
 
 static bool icmp_tag_validation(int proto)
 {
+	const struct net_protocol *ipprot;
 	bool ok;
 
 	rcu_read_lock();
-	ok = rcu_dereference(inet_protos[proto])->icmp_strict_tag_validation;
+	ipprot = rcu_dereference(inet_protos[proto]);
+	ok = ipprot ? ipprot->icmp_strict_tag_validation : false;
 	rcu_read_unlock();
 	return ok;
 }
Signature · 1 line
2.51.0
author maintainer
#294 lore raw

PATCH 283/570 hwmon: (pmbus/isl68137) Fix unchecked return value and use sysfs_emit()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sanman Pradhan <[email protected]>

commit 86259558e422b250aa6aa57163a6d759074573f5 upstream.

isl68137_avs_enable_show_page() uses the return value of pmbus_read_byte_data() without checking for errors. If the I2C transaction fails, a negative error code is passed through bitwise operations, producing incorrect output.

Add an error check to propagate the return value if it is negative. Additionally, modernize the callback by replacing sprintf() with sysfs_emit().

Fixes 038a9c3d1e424 ("hwmon: (pmbus/isl68137) Add driver for Intersil ISL68137 PWM Controller")
---
 drivers/hwmon/pmbus/isl68137.c |    7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/hwmon/pmbus/isl68137.c +5 -2
@@ -80,8 +80,11 @@ static ssize_t isl68137_avs_enable_show_
 {
 	int val = pmbus_read_byte_data(client, page, PMBUS_OPERATION);
 
-	return sprintf(buf, "%d\n",
-		       (val & ISL68137_VOUT_AVS) == ISL68137_VOUT_AVS ? 1 : 0);
+	if (val < 0)
+		return val;
+
+	return sysfs_emit(buf, "%d\n",
+			   (val & ISL68137_VOUT_AVS) == ISL68137_VOUT_AVS);
 }
 
 static ssize_t isl68137_avs_enable_store_page(struct i2c_client *client,
author maintainer
#295 lore raw

PATCH 284/570 i2c: fsi: Fix a potential leak in fsi_i2c_probe()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Christophe JAILLET <[email protected]>

commit be627abcc0d5dbd5882873bd85fbc18aa3d189ed upstream.

In the commit in Fixes:, when the code has been updated to use an explicit for loop, instead of for_each_available_child_of_node(), the assumption that a reference to a device_node structure would be released at each iteration has been broken.

Now, an explicit of_node_put() is needed to release the reference.

---
 drivers/i2c/busses/i2c-fsi.c |    1 +
 1 file changed, 1 insertion(+)
drivers/i2c/busses/i2c-fsi.c +1 -0
@@ -728,6 +728,7 @@ static int fsi_i2c_probe(struct device *
 		rc = i2c_add_adapter(&port->adapter);
 		if (rc < 0) {
 			dev_err(dev, "Failed to register adapter: %d\n", rc);
+			of_node_put(np);
 			kfree(port);
 			continue;
 		}
author maintainer
#296 lore raw

PATCH 285/570 mtd: rawnand: serialize lock/unlock against other NAND operations

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kamal Dasu <[email protected]>

[ Upstream commit bab2bc6e850a697a23b9e5f0e21bb8c187615e95 ]

nand_lock() and nand_unlock() call into chip->ops.lock_area/unlock_area without holding the NAND device lock. On controllers that implement SET_FEATURES via multiple low-level PIO commands, these can race with concurrent UBI/UBIFS background erase/write operations that hold the device lock, resulting in cmd_pending conflicts on the NAND controller.

Add nand_get_device()/nand_release_device() around the lock/unlock operations to serialize them against all other NAND controller access.

Fixes 92270086b7e5 ("mtd: rawnand: Add support for manufacturer specific lock/unlock operation")
---
 drivers/mtd/nand/raw/nand_base.c | 14 ++++++++++++--
 1 file changed, 12 insertions(+), 2 deletions(-)
drivers/mtd/nand/raw/nand_base.c +12 -2
index ee8f47feeaf43..60ad72392b9f8 100644
@@ -4680,11 +4680,16 @@ static void nand_shutdown(struct mtd_info *mtd)
 static int nand_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
 {
 	struct nand_chip *chip = mtd_to_nand(mtd);
+	int ret;
 
 	if (!chip->ops.lock_area)
 		return -ENOTSUPP;
 
-	return chip->ops.lock_area(chip, ofs, len);
+	nand_get_device(chip);
+	ret = chip->ops.lock_area(chip, ofs, len);
+	nand_release_device(chip);
+
+	return ret;
 }
 
 /**
@@ -4696,11 +4701,16 @@ static int nand_lock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
 static int nand_unlock(struct mtd_info *mtd, loff_t ofs, uint64_t len)
 {
 	struct nand_chip *chip = mtd_to_nand(mtd);
+	int ret;
 
 	if (!chip->ops.unlock_area)
 		return -ENOTSUPP;
 
-	return chip->ops.unlock_area(chip, ofs, len);
+	nand_get_device(chip);
+	ret = chip->ops.unlock_area(chip, ofs, len);
+	nand_release_device(chip);
+
+	return ret;
 }
 
 /* Set default functions */
Signature · 1 line
2.51.0
author maintainer
#297 lore raw

PATCH 286/570 mtd: rawnand: brcmnand: skip DMA during panic write

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kamal Dasu <[email protected]>

[ Upstream commit da9ba4dcc01e7cf52b7676f0ee9607b8358c2171 ]

When oops_panic_write is set, the driver disables interrupts and switches to PIO polling mode but still falls through into the DMA path. DMA cannot be used reliably in panic context, so make the DMA path an else branch to ensure only PIO is used during panic writes.

Fixes c1ac2dc34b51 ("mtd: rawnand: brcmnand: When oops in progress use pio and interrupt polling")
---
 drivers/mtd/nand/raw/brcmnand/brcmnand.c | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)
drivers/mtd/nand/raw/brcmnand/brcmnand.c +2 -4
index aa89fcfd71ea0..4bf91b00a2183 100644
@@ -2399,14 +2399,12 @@ static int brcmnand_write(struct mtd_info *mtd, struct nand_chip *chip,
 	for (i = 0; i < ctrl->max_oob; i += 4)
 		oob_reg_write(ctrl, i, 0xffffffff);
 
-	if (mtd->oops_panic_write)
+	if (mtd->oops_panic_write) {
 		/* switch to interrupt polling and PIO mode */
 		disable_ctrl_irqs(ctrl);
-
-	if (use_dma(ctrl) && (has_edu(ctrl) || !oob) && flash_dma_buf_ok(buf)) {
+	} else if (use_dma(ctrl) && (has_edu(ctrl) || !oob) && flash_dma_buf_ok(buf)) {
 		if (ctrl->dma_trans(host, addr, (u32 *)buf, oob, mtd->writesize,
 				    CMD_PROGRAM_PAGE))
-
 			ret = -EIO;
 
 		goto out;
Signature · 1 line
2.51.0
author maintainer
#298 lore raw

PATCH 287/570 ksmbd: fix use-after-free of share_conf in compound request

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hyunwoo Kim <[email protected]>

[ Upstream commit c33615f995aee80657b9fdfbc4ee7f49c2bd733d ]

smb2_get_ksmbd_tcon() reuses work->tcon in compound requests without validating tcon->t_state. ksmbd_tree_conn_lookup() checks t_state == TREE_CONNECTED on the initial lookup path, but the compound reuse path bypasses this check entirely.

If a prior command in the compound (SMB2_TREE_DISCONNECT) sets t_state to TREE_DISCONNECTED and frees share_conf via ksmbd_share_config_put(), subsequent commands dereference the freed share_conf through work->tcon->share_conf.

KASAN report:

[    4.144653] ==================================================================
[    4.145059] BUG: KASAN: slab-use-after-free in smb2_write+0xc74/0xe70
[    4.145415] Read of size 4 at addr ffff88810430c194 by task kworker/1:1/44
[    4.145772]
[    4.145867] CPU: 1 UID: 0 PID: 44 Comm: kworker/1:1 Not tainted 7.0.0-rc3+ #60 PREEMPTLAZY
[    4.145871] Hardware name: QEMU Ubuntu 24.04 PC v2 (i440FX + PIIX, arch_caps fix, 1996), BIOS 1.16.3-debian-1.16.3-2 04/01/2014
[    4.145875] Workqueue: ksmbd-io handle_ksmbd_work
[    4.145888] Call Trace:
[    4.145892]  <TASK>
[    4.145894]  dump_stack_lvl+0x64/0x80
[    4.145910]  print_report+0xce/0x660
[    4.145919]  ? __pfx__raw_spin_lock_irqsave+0x10/0x10
[    4.145928]  ? smb2_write+0xc74/0xe70
[    4.145931]  kasan_report+0xce/0x100
[    4.145934]  ? smb2_write+0xc74/0xe70
[    4.145937]  smb2_write+0xc74/0xe70
[    4.145939]  ? __pfx_smb2_write+0x10/0x10
[    4.145942]  ? _raw_spin_unlock+0xe/0x30
[    4.145945]  ? ksmbd_smb2_check_message+0xeb2/0x24c0
[    4.145948]  ? smb2_tree_disconnect+0x31c/0x480
[    4.145951]  handle_ksmbd_work+0x40f/0x1080
[    4.145953]  process_one_work+0x5fa/0xef0
[    4.145962]  ? assign_work+0x122/0x3e0
[    4.145964]  worker_thread+0x54b/0xf70
[    4.145967]  ? __pfx_worker_thread+0x10/0x10
[    4.145970]  kthread+0x346/0x470
[    4.145976]  ? recalc_sigpending+0x19b/0x230
[    4.145980]  ? __pfx_kthread+0x10/0x10
[    4.145984]  ret_from_fork+0x4fb/0x6c0
[    4.145992]  ? __pfx_ret_from_fork+0x10/0x10
[    4.145995]  ? __switch_to+0x36c/0xbe0
[    4.145999]  ? __pfx_kthread+0x10/0x10
[    4.146003]  ret_from_fork_asm+0x1a/0x30
[    4.146013]  </TASK>
[    4.146014]
[    4.149858] Allocated by task 44:
[    4.149953]  kasan_save_stack+0x33/0x60
[    4.150061]  kasan_save_track+0x14/0x30
[    4.150169]  __kasan_kmalloc+0x8f/0xa0
[    4.150274]  ksmbd_share_config_get+0x1dd/0xdd0
[    4.150401]  ksmbd_tree_conn_connect+0x7e/0x600
[    4.150529]  smb2_tree_connect+0x2e6/0x1000
[    4.150645]  handle_ksmbd_work+0x40f/0x1080
[    4.150761]  process_one_work+0x5fa/0xef0
[    4.150873]  worker_thread+0x54b/0xf70
[    4.150978]  kthread+0x346/0x470
[    4.151071]  ret_from_fork+0x4fb/0x6c0
[    4.151176]  ret_from_fork_asm+0x1a/0x30
[    4.151286]
[    4.151332] Freed by task 44:
[    4.151418]  kasan_save_stack+0x33/0x60
[    4.151526]  kasan_save_track+0x14/0x30
[    4.151634]  kasan_save_free_info+0x3b/0x60
[    4.151751]  __kasan_slab_free+0x43/0x70
[    4.151861]  kfree+0x1ca/0x430
[    4.151952]  __ksmbd_tree_conn_disconnect+0xc8/0x190
[    4.152088]  smb2_tree_disconnect+0x1cd/0x480
[    4.152211]  handle_ksmbd_work+0x40f/0x1080
[    4.152326]  process_one_work+0x5fa/0xef0
[    4.152438]  worker_thread+0x54b/0xf70
[    4.152545]  kthread+0x346/0x470
[    4.152638]  ret_from_fork+0x4fb/0x6c0
[    4.152743]  ret_from_fork_asm+0x1a/0x30
[    4.152853]
[    4.152900] The buggy address belongs to the object at ffff88810430c180
[    4.152900]  which belongs to the cache kmalloc-96 of size 96
[    4.153226] The buggy address is located 20 bytes inside of
[    4.153226]  freed 96-byte region [ffff88810430c180, ffff88810430c1e0)
[    4.153549]
[    4.153596] The buggy address belongs to the physical page:
[    4.153750] page: refcount:0 mapcount:0 mapping:0000000000000000 index:0xffff88810430ce80 pfn:0x10430c
[    4.154000] flags: 0x100000000000200(workingset|node=0|zone=2)
[    4.154160] page_type: f5(slab)
[    4.154251] raw: 0100000000000200 ffff888100041280 ffff888100040110 ffff888100040110
[    4.154461] raw: ffff88810430ce80 0000000800200009 00000000f5000000 0000000000000000
[    4.154668] page dumped because: kasan: bad access detected
[    4.154820]
[    4.154866] Memory state around the buggy address:
[    4.155002]  ffff88810430c080: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
[    4.155196]  ffff88810430c100: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
[    4.155391] >ffff88810430c180: fa fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc
[    4.155587]                          ^
[    4.155693]  ffff88810430c200: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
[    4.155891]  ffff88810430c280: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
[    4.156087] ==================================================================

Add the same t_state validation to the compound reuse path, consistent with ksmbd_tree_conn_lookup().

Fixes 5005bcb42191 ("ksmbd: validate session id and tree id in the compound request")
---
 fs/ksmbd/smb2pdu.c | 2 ++
 1 file changed, 2 insertions(+)
fs/ksmbd/smb2pdu.c +2 -0
index 222d3a31f5106..1210a3231d1c4 100644
@@ -116,6 +116,8 @@ int smb2_get_ksmbd_tcon(struct ksmbd_work *work)
 			pr_err("The first operation in the compound does not have tcon\n");
 			return -EINVAL;
 		}
+		if (work->tcon->t_state != TREE_CONNECTED)
+			return -ENOENT;
 		if (tree_id != UINT_MAX && work->tcon->id != tree_id) {
 			pr_err("tree id(%u) is different with id(%u) in first operation\n",
 					tree_id, work->tcon->id);
Signature · 1 line
2.51.0
author maintainer
#299 lore raw

PATCH 288/570 drm/i915/gt: Check set_default_submission() before deferencing

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rahul Bukte <[email protected]>

[ Upstream commit 0162ab3220bac870e43e229e6e3024d1a21c3f26 ]

When the i915 driver firmware binaries are not present, the set_default_submission pointer is not set. This pointer is dereferenced during suspend anyways.

Add a check to make sure it is set before dereferencing.

[   23.289926] PM: suspend entry (deep)
[   23.293558] Filesystems sync: 0.000 seconds
[   23.298010] Freezing user space processes
[   23.302771] Freezing user space processes completed (elapsed 0.000 seconds)
[   23.309766] OOM killer disabled.
[   23.313027] Freezing remaining freezable tasks
[   23.318540] Freezing remaining freezable tasks completed (elapsed 0.001 seconds)
[   23.342038] serial 00:05: disabled
[   23.345719] serial 00:02: disabled
[   23.349342] serial 00:01: disabled
[   23.353782] sd 0:0:0:0: [sda] Synchronizing SCSI cache
[   23.358993] sd 1:0:0:0: [sdb] Synchronizing SCSI cache
[   23.361635] ata1.00: Entering standby power mode
[   23.368863] ata2.00: Entering standby power mode
[   23.445187] BUG: kernel NULL pointer dereference, address: 0000000000000000
[   23.452194] #PF: supervisor instruction fetch in kernel mode
[   23.457896] #PF: error_code(0x0010) - not-present page
[   23.463065] PGD 0 P4D 0
[   23.465640] Oops: Oops: 0010 [#1] SMP NOPTI
[   23.469869] CPU: 8 UID: 0 PID: 211 Comm: kworker/u48:18 Tainted: G S      W           6.19.0-rc4-00020-gf0b9d8eb98df #10 PREEMPT(voluntary)
[   23.482512] Tainted: [S]=CPU_OUT_OF_SPEC, [W]=WARN
[   23.496511] Workqueue: async async_run_entry_fn
[   23.501087] RIP: 0010:0x0
[   23.503755] Code: Unable to access opcode bytes at 0xffffffffffffffd6.
[   23.510324] RSP: 0018:ffffb4a60065fca8 EFLAGS: 00010246
[   23.515592] RAX: 0000000000000000 RBX: ffff9f428290e000 RCX: 000000000000000f
[   23.522765] RDX: 0000000000000000 RSI: 0000000000000282 RDI: ffff9f428290e000
[   23.529937] RBP: ffff9f4282907070 R08: ffff9f4281130428 R09: 00000000ffffffff
[   23.537111] R10: 0000000000000000 R11: 0000000000000001 R12: ffff9f42829070f8
[   23.544284] R13: ffff9f4282906028 R14: ffff9f4282900000 R15: ffff9f4282906b68
[   23.551457] FS:  0000000000000000(0000) GS:ffff9f466b2cf000(0000) knlGS:0000000000000000
[   23.559588] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   23.565365] CR2: ffffffffffffffd6 CR3: 000000031c230001 CR4: 0000000000f70ef0
[   23.572539] PKRU: 55555554
[   23.575281] Call Trace:
[   23.577770]  <TASK>
[   23.579905]  intel_engines_reset_default_submission+0x42/0x60
[   23.585695]  __intel_gt_unset_wedged+0x191/0x200
[   23.590360]  intel_gt_unset_wedged+0x20/0x40
[   23.594675]  gt_sanitize+0x15e/0x170
[   23.598290]  i915_gem_suspend_late+0x6b/0x180
[   23.602692]  i915_drm_suspend_late+0x35/0xf0
[   23.607008]  ? __pfx_pci_pm_suspend_late+0x10/0x10
[   23.611843]  dpm_run_callback+0x78/0x1c0
[   23.615817]  device_suspend_late+0xde/0x2e0
[   23.620037]  async_suspend_late+0x18/0x30
[   23.624082]  async_run_entry_fn+0x25/0xa0
[   23.628129]  process_one_work+0x15b/0x380
[   23.632182]  worker_thread+0x2a5/0x3c0
[   23.635973]  ? __pfx_worker_thread+0x10/0x10
[   23.640279]  kthread+0xf6/0x1f0
[   23.643464]  ? __pfx_kthread+0x10/0x10
[   23.647263]  ? __pfx_kthread+0x10/0x10
[   23.651045]  ret_from_fork+0x131/0x190
[   23.654837]  ? __pfx_kthread+0x10/0x10
[   23.658634]  ret_from_fork_asm+0x1a/0x30
[   23.662597]  </TASK>
[   23.664826] Modules linked in:
[   23.667914] CR2: 0000000000000000
[   23.671271] ------------[ cut here ]------------
(cherry picked from commit daa199abc3d3d1740c9e3a2c3e9216ae5b447cad)
Fixes ff44ad51ebf8 ("drm/i915: Move engine->submit_request selection to a vfunc")
---
 drivers/gpu/drm/i915/gt/intel_engine_cs.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/gt/intel_engine_cs.c +2 -1
index 42cb3ad04d89a..e8e495694c182 100644
@@ -1363,7 +1363,8 @@ void intel_engines_reset_default_submission(struct intel_gt *gt)
 		if (engine->sanitize)
 			engine->sanitize(engine);
 
-		engine->set_default_submission(engine);
+		if (engine->set_default_submission)
+			engine->set_default_submission(engine);
 	}
 }
 
Signature · 1 line
2.51.0
author maintainer
#300 lore raw

PATCH 289/570 lib/bootconfig: check xbc_init_node() return in override path

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Josh Law <[email protected]>

[ Upstream commit bb288d7d869e86d382f35a0e26242c5ccb05ca82 ]

The ':=' override path in xbc_parse_kv() calls xbc_init_node() to re-initialize an existing value node but does not check the return value. If xbc_init_node() fails (data offset out of range), parsing silently continues with stale node data.

Add the missing error check to match the xbc_add_node() call path which already checks for failure.

In practice, a bootconfig using ':=' to override a value near the 32KB data limit could silently retain the old value, meaning a security-relevant boot parameter override (e.g., a trace filter or debug setting) would not take effect as intended.

Fixes e5efaeb8a8f5 ("bootconfig: Support mixing a value and subkeys under a key")
---
 lib/bootconfig.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
lib/bootconfig.c +2 -1
index 0579a93cb163e..9873c6372adca 100644
@@ -646,7 +646,8 @@ static int __init xbc_parse_kv(char **k, char *v, int op)
 		if (op == ':') {
 			unsigned short nidx = child->next;
 
-			xbc_init_node(child, v, XBC_VALUE);
+			if (xbc_init_node(child, v, XBC_VALUE) < 0)
+				return xbc_parse_error("Failed to override value", v);
 			child->next = nidx;	/* keep subkeys */
 			goto array;
 		}
Signature · 1 line
2.51.0
author maintainer
#301 lore raw

PATCH 290/570 tools/bootconfig: fix fd leak in load_xbc_file() on fstat failure

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Josh Law <[email protected]>

[ Upstream commit 3b2c2ab4ceb82af484310c3087541eab00ea288b ]

If fstat() fails after open() succeeds, the function returns without closing the file descriptor. Also preserve errno across close(), since close() may overwrite it before the error is returned.

Fixes 950313ebf79c ("tools: bootconfig: Add bootconfig command")
---
 tools/bootconfig/main.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
tools/bootconfig/main.c +5 -2
index fc922cfdadaa6..78b252ac3817e 100644
@@ -156,8 +156,11 @@ static int load_xbc_file(const char *path, char **buf)
 	if (fd < 0)
 		return -errno;
 	ret = fstat(fd, &stat);
-	if (ret < 0)
-		return -errno;
+	if (ret < 0) {
+		ret = -errno;
+		close(fd);
+		return ret;
+	}
 
 	ret = load_xbc_fd(fd, buf, stat.st_size);
 
Signature · 1 line
2.51.0
author maintainer
#302 lore raw

PATCH 291/570 netfilter: nf_tables: de-constify set commit ops function argument

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Florian Westphal <[email protected]>

commit 256001672153af5786c6ca148114693d7d76d836 upstream.

The set backend using this already has to work around this via ugly cast, don't spread this pattern.

---
 include/net/netfilter/nf_tables.h |    2 +-
 net/netfilter/nft_set_pipapo.c    |    7 +++----
 2 files changed, 4 insertions(+), 5 deletions(-)
include/net/netfilter/nf_tables.h +1 -1
@@ -455,7 +455,7 @@ struct nft_set_ops {
 					       const struct nft_set *set,
 					       const struct nft_set_elem *elem,
 					       unsigned int flags);
-	void				(*commit)(const struct nft_set *set);
+	void				(*commit)(struct nft_set *set);
 	void				(*abort)(const struct nft_set *set);
 	u64				(*privsize)(const struct nlattr * const nla[],
 						    const struct nft_set_desc *desc);
net/netfilter/nft_set_pipapo.c +3 -4
@@ -1587,12 +1587,11 @@ static void nft_pipapo_gc_deactivate(str
 
 /**
  * pipapo_gc() - Drop expired entries from set, destroy start and end elements
- * @_set:	nftables API set representation
+ * @set:	nftables API set representation
  * @m:		Matching data
  */
-static void pipapo_gc(const struct nft_set *_set, struct nft_pipapo_match *m)
+static void pipapo_gc(struct nft_set *set, struct nft_pipapo_match *m)
 {
-	struct nft_set *set = (struct nft_set *) _set;
 	struct nft_pipapo *priv = nft_set_priv(set);
 	struct net *net = read_pnet(&set->net);
 	u64 tstamp = nft_net_tstamp(net);
@@ -1707,7 +1706,7 @@ static void pipapo_reclaim_match(struct
  * We also need to create a new working copy for subsequent insertions and
  * deletions.
  */
-static void nft_pipapo_commit(const struct nft_set *set)
+static void nft_pipapo_commit(struct nft_set *set)
 {
 	struct nft_pipapo *priv = nft_set_priv(set);
 	struct nft_pipapo_match *new_clone, *old;
author maintainer
#303 lore raw

PATCH 292/570 netfilter: nft_set_pipapo: split gc into unlink and reclaim phase

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Florian Westphal <[email protected]>

commit 9df95785d3d8302f7c066050117b04cd3c2048c2 upstream.

Yiming Qian reports Use-after-free in the pipapo set type:
Under a large number of expired elements, commit-time GC can run for a very long time in a non-preemptible context, triggering soft lockup warnings and RCU stall reports (local denial of service).

We must split GC in an unlink and a reclaim phase.

We cannot queue elements for freeing until pointers have been swapped. Expired elements are still exposed to both the packet path and userspace dumpers via the live copy of the data structure.

call_rcu() does not protect us: dump operations or element lookups starting after call_rcu has fired can still observe the free'd element, unless the commit phase has made enough progress to swap the clone and live pointers before any new reader has picked up the old version.

This a similar approach as done recently for the rbtree backend in commit 35f83a75529a ("netfilter: nft_set_rbtree: don't gc elements on insert").

Fixes 3c4287f62044 ("nf_tables: Add set type for arbitrary concatenation of ranges")
---
 include/net/netfilter/nf_tables.h |    5 +++
 net/netfilter/nf_tables_api.c     |    5 ---
 net/netfilter/nft_set_pipapo.c    |   51 ++++++++++++++++++++++++++++++++------
 net/netfilter/nft_set_pipapo.h    |    2 +
 4 files changed, 50 insertions(+), 13 deletions(-)
include/net/netfilter/nf_tables.h +5 -0
@@ -1655,6 +1655,11 @@ struct nft_trans_gc {
 	struct rcu_head		rcu;
 };
 
+static inline int nft_trans_gc_space(const struct nft_trans_gc *trans)
+{
+	return NFT_TRANS_GC_BATCHCOUNT - trans->count;
+}
+
 struct nft_trans_gc *nft_trans_gc_alloc(struct nft_set *set,
 					unsigned int gc_seq, gfp_t gfp);
 void nft_trans_gc_destroy(struct nft_trans_gc *trans);
net/netfilter/nf_tables_api.c +0 -5
@@ -9146,11 +9146,6 @@ static void nft_trans_gc_queue_work(stru
 	schedule_work(&trans_gc_work);
 }
 
-static int nft_trans_gc_space(struct nft_trans_gc *trans)
-{
-	return NFT_TRANS_GC_BATCHCOUNT - trans->count;
-}
-
 struct nft_trans_gc *nft_trans_gc_queue_async(struct nft_trans_gc *gc,
 					      unsigned int gc_seq, gfp_t gfp)
 {
net/netfilter/nft_set_pipapo.c +43 -8

110 lines not on this page · raw message

net/netfilter/nft_set_pipapo.h +2 -0
@@ -165,6 +165,7 @@ struct nft_pipapo_match {
  * @width:	Total bytes to be matched for one packet, including padding
  * @dirty:	Working copy has pending insertions or deletions
  * @last_gc:	Timestamp of last garbage collection run, jiffies
+ * @gc_head:	list of nft_trans_gc to queue up for mem reclaim
  */
 struct nft_pipapo {
 	struct nft_pipapo_match __rcu *match;
@@ -172,6 +173,7 @@ struct nft_pipapo {
 	int width;
 	bool dirty;
 	unsigned long last_gc;
+	struct list_head gc_head;
 };
 
 struct nft_pipapo_elem;
author maintainer
#304 lore raw

PATCH 293/570 xen/privcmd: restrict usage in unprivileged domU

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Juergen Gross <[email protected]>

commit 453b8fb68f3641fea970db88b7d9a153ed2a37e8 upstream.

The Xen privcmd driver allows to issue arbitrary hypercalls from user space processes. This is normally no problem, as access is usually limited to root and the hypervisor will deny any hypercalls affecting other domains.

In case the guest is booted using secure boot, however, the privcmd driver would be enabling a root user process to modify e.g. kernel memory contents, thus breaking the secure boot feature.

The only known case where an unprivileged domU is really needing to use the privcmd driver is the case when it is acting as the device model for another guest. In this case all hypercalls issued via the privcmd driver will target that other guest.

Fortunately the privcmd driver can already be locked down to allow only hypercalls targeting a specific domain, but this mode can be activated from user land only today.

The target domain can be obtained from Xenstore, so when not running in dom0 restrict the privcmd driver to that target domain from the beginning, resolving the potential problem of breaking secure boot.

This is XSA-482

Fixes 1c5de1939c20 ("xen: add privcmd driver")
---
 drivers/xen/privcmd.c |   60 +++++++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 57 insertions(+), 3 deletions(-)
drivers/xen/privcmd.c +57 -3
@@ -10,6 +10,7 @@
 #define pr_fmt(fmt) "xen:" KBUILD_MODNAME ": " fmt
 
 #include <linux/kernel.h>
+#include <linux/kstrtox.h>
 #include <linux/module.h>
 #include <linux/sched.h>
 #include <linux/slab.h>
@@ -24,6 +25,8 @@
 #include <linux/seq_file.h>
 #include <linux/miscdevice.h>
 #include <linux/moduleparam.h>
+#include <linux/notifier.h>
+#include <linux/wait.h>
 
 #include <asm/xen/hypervisor.h>
 #include <asm/xen/hypercall.h>
@@ -37,6 +40,7 @@
 #include <xen/page.h>
 #include <xen/xen-ops.h>
 #include <xen/balloon.h>
+#include <xen/xenbus.h>
 
 #include "privcmd.h"
 
@@ -59,6 +63,11 @@ struct privcmd_data {
 	domid_t domid;
 };
 
+/* DOMID_INVALID implies no restriction */
+static domid_t target_domain = DOMID_INVALID;
+static bool restrict_wait;
+static DECLARE_WAIT_QUEUE_HEAD(restrict_wait_wq);
+
 static int privcmd_vma_range_is_mapped(
                struct vm_area_struct *vma,
                unsigned long addr,
@@ -877,13 +886,16 @@ static long privcmd_ioctl(struct file *f
 
 static int privcmd_open(struct inode *ino, struct file *file)
 {
-	struct privcmd_data *data = kzalloc(sizeof(*data), GFP_KERNEL);
+	struct privcmd_data *data;
+
+	if (wait_event_interruptible(restrict_wait_wq, !restrict_wait) < 0)
+		return -EINTR;
 
+	data = kzalloc(sizeof(*data), GFP_KERNEL);
 	if (!data)
 		return -ENOMEM;
 
-	/* DOMID_INVALID implies no restriction */
-	data->domid = DOMID_INVALID;
+	data->domid = target_domain;
 
 	file->private_data = data;
 	return 0;
@@ -976,6 +988,45 @@ static struct miscdevice privcmd_dev = {
 	.fops = &xen_privcmd_fops,
 };
 
+static int init_restrict(struct notifier_block *notifier,
+			 unsigned long event,
+			 void *data)
+{
+	char *target;
+	unsigned int domid;
+
+	/* Default to an guaranteed unused domain-id. */
+	target_domain = DOMID_IDLE;
+
+	target = xenbus_read(XBT_NIL, "target", "", NULL);
+	if (IS_ERR(target) || kstrtouint(target, 10, &domid)) {
+		pr_err("No target domain found, blocking all hypercalls\n");
+		goto out;
+	}
+
+	target_domain = domid;
+
+ out:
+	if (!IS_ERR(target))
+		kfree(target);
+
+	restrict_wait = false;
+	wake_up_all(&restrict_wait_wq);
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block xenstore_notifier = {
+	.notifier_call = init_restrict,
+};
+
+static void __init restrict_driver(void)
+{
+	restrict_wait = true;
+
+	register_xenstore_notifier(&xenstore_notifier);
+}
+
 static int __init privcmd_init(void)
 {
 	int err;
@@ -983,6 +1034,9 @@ static int __init privcmd_init(void)
 	if (!xen_domain())
 		return -ENODEV;
 
+	if (!xen_initial_domain())
+		restrict_driver();
+
 	err = misc_register(&privcmd_dev);
 	if (err != 0) {
 		pr_err("Could not register Xen privcmd device\n");
author maintainer
#305 lore raw

PATCH 294/570 xen/privcmd: add boot control for restricted usage in domU

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Juergen Gross <[email protected]>

commit 1613462be621ad5103ec338a7b0ca0746ec4e5f1 upstream.

When running in an unprivileged domU under Xen, the privcmd driver is restricted to allow only hypercalls against a target domain, for which the current domU is acting as a device model.

Add a boot parameter "unrestricted" to allow all hypercalls (the hypervisor will still refuse destructive hypercalls affecting other guests).

Make this new parameter effective only in case the domU wasn't started using secure boot, as otherwise hypercalls targeting the domU itself might result in violating the secure boot functionality.

This is achieved by adding another lockdown reason, which can be tested to not being set when applying the "unrestricted" option.

This is part of XSA-482

---
 drivers/xen/privcmd.c    |   13 +++++++++++++
 include/linux/security.h |    1 +
 security/security.c      |    1 +
 3 files changed, 15 insertions(+)
drivers/xen/privcmd.c +13 -0
@@ -26,6 +26,7 @@
 #include <linux/miscdevice.h>
 #include <linux/moduleparam.h>
 #include <linux/notifier.h>
+#include <linux/security.h>
 #include <linux/wait.h>
 
 #include <asm/xen/hypervisor.h>
@@ -59,6 +60,11 @@ module_param_named(dm_op_buf_max_size, p
 MODULE_PARM_DESC(dm_op_buf_max_size,
 		 "Maximum size of a dm_op hypercall buffer");
 
+static bool unrestricted;
+module_param(unrestricted, bool, 0);
+MODULE_PARM_DESC(unrestricted,
+	"Don't restrict hypercalls to target domain if running in a domU");
+
 struct privcmd_data {
 	domid_t domid;
 };
@@ -1022,6 +1028,13 @@ static struct notifier_block xenstore_no
 
 static void __init restrict_driver(void)
 {
+	if (unrestricted) {
+		if (security_locked_down(LOCKDOWN_XEN_USER_ACTIONS))
+			pr_warn("Kernel is locked down, parameter \"unrestricted\" ignored\n");
+		else
+			return;
+	}
+
 	restrict_wait = true;
 
 	register_xenstore_notifier(&xenstore_notifier);
include/linux/security.h +1 -0
@@ -122,6 +122,7 @@ enum lockdown_reason {
 	LOCKDOWN_XMON_WR,
 	LOCKDOWN_BPF_WRITE_USER,
 	LOCKDOWN_DBG_WRITE_KERNEL,
+	LOCKDOWN_XEN_USER_ACTIONS,
 	LOCKDOWN_INTEGRITY_MAX,
 	LOCKDOWN_KCORE,
 	LOCKDOWN_KPROBES,
security/security.c +1 -0
@@ -60,6 +60,7 @@ const char *const lockdown_reasons[LOCKD
 	[LOCKDOWN_XMON_WR] = "xmon write access",
 	[LOCKDOWN_BPF_WRITE_USER] = "use of bpf to write user RAM",
 	[LOCKDOWN_DBG_WRITE_KERNEL] = "use of kgdb/kdb to write kernel RAM",
+	[LOCKDOWN_XEN_USER_ACTIONS] = "Xen guest user action",
 	[LOCKDOWN_INTEGRITY_MAX] = "integrity",
 	[LOCKDOWN_KCORE] = "/proc/kcore access",
 	[LOCKDOWN_KPROBES] = "use of kprobes",
author maintainer
#306 lore raw

PATCH 295/570 sh: platform_early: remove pdev->driver_override check

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Danilo Krummrich <[email protected]>

[ Upstream commit c5f60e3f07b6609562d21efda878e83ce8860728 ]

In commit 507fd01d5333 ("drivers: move the early platform device support to arch/sh") platform_match() was copied over to the sh platform_early code, accidentally including the driver_override check.

This check does not make sense for platform_early, as sysfs is not even available in first place at this point in the boot process, hence remove the check.

Fixes 507fd01d5333 ("drivers: move the early platform device support to arch/sh")
---
 arch/sh/drivers/platform_early.c | 4 ----
 1 file changed, 4 deletions(-)
arch/sh/drivers/platform_early.c +0 -4
index 143747c45206f..48ddbc547bd9a 100644
@@ -26,10 +26,6 @@ static int platform_match(struct device *dev, struct device_driver *drv)
 	struct platform_device *pdev = to_platform_device(dev);
 	struct platform_driver *pdrv = to_platform_driver(drv);
 
-	/* When driver_override is set, only bind to the matching driver */
-	if (pdev->driver_override)
-		return !strcmp(pdev->driver_override, drv->name);
-
 	/* Then try to match against the id table */
 	if (pdrv->id_table)
 		return platform_match_id(pdrv->id_table, pdev) != NULL;
Signature · 1 line
2.51.0
author maintainer
#307 lore raw

PATCH 296/570 bpf: Release module BTF IDR before module unload

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Kumar Kartikeya Dwivedi <[email protected]>

[ Upstream commit 146bd2a87a65aa407bb17fac70d8d583d19aba06 ]

Gregory reported in [0] that the global_map_resize test when run in repeatedly ends up failing during program load. This stems from the fact that BTF reference has not dropped to zero after the previous run's module is unloaded, and the older module's BTF is still discoverable and visible. Later, in libbpf, load_module_btfs() will find the ID for this stale BTF, open its fd, and then it will be used during program load where later steps taking module reference using btf_try_get_module() fail since the underlying module for the BTF is gone.

Logically, once a module is unloaded, it's associated BTF artifacts should become hidden. The BTF object inside the kernel may still remain alive as long its reference counts are alive, but it should no longer be discoverable.

To fix this, let us call btf_free_id() from the MODULE_STATE_GOING case for the module unload to free the BTF associated IDR entry, and disable its discovery once module unload returns to user space. If a race happens during unload, the outcome is non-deterministic anyway. However, user space should be able to rely on the guarantee that once it has synchronously established a successful module unload, no more stale artifacts associated with this module can be obtained subsequently.

Note that we must be careful to not invoke btf_free_id() in btf_put() when btf_is_module() is true now. There could be a window where the module unload drops a non-terminal reference, frees the IDR, but the same ID gets reused and the second unconditional btf_free_id() ends up releasing an unrelated entry.

To avoid a special case for btf_is_module() case, set btf->id to zero to make btf_free_id() idempotent, such that we can unconditionally invoke it from btf_put(), and also from the MODULE_STATE_GOING case. Since zero is an invalid IDR, the idr_remove() should be a noop.

Note that we can be sure that by the time we reach final btf_put() for btf_is_module() case, the btf_free_id() is already done, since the module itself holds the BTF reference, and it will call this function for the BTF before dropping its own reference.

[0]: https://lore.kernel.org/bpf/[email protected]

---
 kernel/bpf/btf.c | 24 ++++++++++++++++++++----
 1 file changed, 20 insertions(+), 4 deletions(-)
kernel/bpf/btf.c +20 -4
index d3eb75bfd9718..5d87df80c4bd7 100644
@@ -1501,7 +1501,16 @@ static void btf_free_id(struct btf *btf)
 	 * of the _bh() version.
 	 */
 	spin_lock_irqsave(&btf_idr_lock, flags);
-	idr_remove(&btf_idr, btf->id);
+	if (btf->id) {
+		idr_remove(&btf_idr, btf->id);
+		/*
+		 * Clear the id here to make this function idempotent, since it will get
+		 * called a couple of times for module BTFs: on module unload, and then
+		 * the final btf_put(). btf_alloc_id() starts IDs with 1, so we can use
+		 * 0 as sentinel value.
+		 */
+		WRITE_ONCE(btf->id, 0);
+	}
 	spin_unlock_irqrestore(&btf_idr_lock, flags);
 }
 
@@ -5890,7 +5899,7 @@ static void bpf_btf_show_fdinfo(struct seq_file *m, struct file *filp)
 {
 	const struct btf *btf = filp->private_data;
 
-	seq_printf(m, "btf_id:\t%u\n", btf->id);
+	seq_printf(m, "btf_id:\t%u\n", READ_ONCE(btf->id));
 }
 #endif
 
@@ -5985,7 +5994,7 @@ int btf_get_info_by_fd(const struct btf *btf,
 	if (copy_from_user(&info, uinfo, info_copy))
 		return -EFAULT;
 
-	info.id = btf->id;
+	info.id = READ_ONCE(btf->id);
 	ubtf = u64_to_user_ptr(info.btf);
 	btf_copy = min_t(u32, btf->data_size, info.btf_size);
 	if (copy_to_user(ubtf, btf->data, btf_copy))
@@ -6048,7 +6057,7 @@ int btf_get_fd_by_id(u32 id)
 
 u32 btf_obj_id(const struct btf *btf)
 {
-	return btf->id;
+	return READ_ONCE(btf->id);
 }
 
 bool btf_is_kernel(const struct btf *btf)
@@ -6185,6 +6194,13 @@ static int btf_module_notify(struct notifier_block *nb, unsigned long op,
 			if (btf_mod->module != module)
 				continue;
 
+			/*
+			 * For modules, we do the freeing of BTF IDR as soon as
+			 * module goes away to disable BTF discovery, since the
+			 * btf_try_get_module() on such BTFs will fail. This may
+			 * be called again on btf_put(), but it's ok to do so.
+			 */
+			btf_free_id(btf_mod->btf);
 			list_del(&btf_mod->list);
 			if (btf_mod->sysfs_attr)
 				sysfs_remove_bin_file(btf_kobj, btf_mod->sysfs_attr);
Signature · 1 line
2.51.0
author maintainer
#308 lore raw

PATCH 297/570 HID: asus: avoid memory leak in asus_report_fixup()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Günther Noack <[email protected]>

[ Upstream commit 2bad24c17742fc88973d6aea526ce1353f5334a3 ]

The asus_report_fixup() function was returning a newly allocated kmemdup()-allocated buffer, but never freeing it. Switch to devm_kzalloc() to ensure the memory is managed and freed automatically when the device is removed.

The caller of report_fixup() does not take ownership of the returned pointer, but it is permitted to return a pointer whose lifetime is at least that of the input buffer.

Also fix a harmless out-of-bounds read by copying only the original descriptor size.

Assisted-by Gemini-CLI:Google Gemini 3
---
 drivers/hid/hid-asus.c | 15 +++++++++++----
 1 file changed, 11 insertions(+), 4 deletions(-)
drivers/hid/hid-asus.c +11 -4
index feec0724328ff..3be17a8b7a293 100644
@@ -1224,14 +1224,21 @@ static __u8 *asus_report_fixup(struct hid_device *hdev, __u8 *rdesc,
 		 */
 		if (*rsize == rsize_orig &&
 			rdesc[offs] == 0x09 && rdesc[offs + 1] == 0x76) {
-			*rsize = rsize_orig + 1;
-			rdesc = kmemdup(rdesc, *rsize, GFP_KERNEL);
-			if (!rdesc)
-				return NULL;
+			__u8 *new_rdesc;
+
+			new_rdesc = devm_kzalloc(&hdev->dev, rsize_orig + 1,
+						 GFP_KERNEL);
+			if (!new_rdesc)
+				return rdesc;
 
 			hid_info(hdev, "Fixing up %s keyb report descriptor\n",
 				drvdata->quirks & QUIRK_T100CHI ?
 				"T100CHI" : "T90CHI");
+
+			memcpy(new_rdesc, rdesc, rsize_orig);
+			*rsize = rsize_orig + 1;
+			rdesc = new_rdesc;
+
 			memmove(rdesc + offs + 4, rdesc + offs + 2, 12);
 			rdesc[offs] = 0x19;
 			rdesc[offs + 1] = 0x00;
Signature · 1 line
2.51.0
author maintainer
#309 lore raw

PATCH 298/570 platform/x86: intel-hid: Add Dell 14 Plus 2-in-1 to dmi_vgbs_allow_list

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Peter Metz <[email protected]>

[ Upstream commit 6b3fa0615cd8432148581de62a52f83847af3d70 ]

The Dell 14 Plus 2-in-1 (model DB04250) requires the VGBS allow list entry to correctly enable the tablet mode switch. Without this, the chassis state is not reported, and the hinge rotation only emits unknown scancodes.

Verified on Dell 14 Plus 2-in-1 DB04250.

---
 drivers/platform/x86/intel/hid.c | 6 ++++++
 1 file changed, 6 insertions(+)
drivers/platform/x86/intel/hid.c +6 -0
index 4d488e985dc55..6331469ee6585 100644
@@ -156,6 +156,12 @@ static const struct dmi_system_id dmi_vgbs_allow_list[] = {
 			DMI_MATCH(DMI_PRODUCT_NAME, "Dell Pro Rugged 12 Tablet RA02260"),
 		},
 	},
+	{
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Dell Inc."),
+			DMI_MATCH(DMI_PRODUCT_NAME, "Dell 14 Plus 2-in-1 DB04250"),
+		},
+	},
 	{ }
 };
 
Signature · 1 line
2.51.0
author maintainer
#310 lore raw

PATCH 299/570 nvme-pci: cap queue creation to used queues

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Keith Busch <[email protected]>

[ Upstream commit 4735b510a00fb2d4ac9e8d21a8c9552cb281f585 ]

If the user reduces the special queue count at runtime and resets the controller, we need to reduce the number of queues and interrupts requested accordingly rather than start with the pre-allocated queue count.

---
 drivers/nvme/host/pci.c | 8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)
drivers/nvme/host/pci.c +7 -1
index 04cccbb05372a..c4a33e9d2c717 100644
@@ -2306,7 +2306,13 @@ static int nvme_setup_io_queues(struct nvme_dev *dev)
 	dev->nr_write_queues = write_queues;
 	dev->nr_poll_queues = poll_queues;
 
-	nr_io_queues = dev->nr_allocated_queues - 1;
+	/*
+	 * The initial number of allocated queue slots may be too large if the
+	 * user reduced the special queue parameters. Cap the value to the
+	 * number we need for this round.
+	 */
+	nr_io_queues = min(nvme_max_io_queues(dev),
+			   dev->nr_allocated_queues - 1);
 	result = nvme_set_queue_count(&dev->ctrl, &nr_io_queues);
 	if (result < 0)
 		return result;
Signature · 1 line
2.51.0
author maintainer
#311 lore raw

PATCH 300/570 platform/x86: intel-hid: Enable 5-button array on ThinkPad X1 Fold 16 Gen 1

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Leif Skunberg <[email protected]>

[ Upstream commit b38d478dad79e61e8a65931021bdfd7a71741212 ]

The Lenovo ThinkPad X1 Fold 16 Gen 1 has physical volume up/down buttons that are handled through the intel-hid 5-button array interface. The firmware does not advertise 5-button array support via HEBC, so the driver relies on a DMI allowlist to enable it.

Add the ThinkPad X1 Fold 16 Gen 1 to the button_array_table so the volume buttons work out of the box.

---
 drivers/platform/x86/intel/hid.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/platform/x86/intel/hid.c +7 -0
index 6331469ee6585..cbc4ec2f8479b 100644
@@ -102,6 +102,13 @@ static const struct dmi_system_id button_array_table[] = {
 			DMI_MATCH(DMI_PRODUCT_FAMILY, "ThinkPad X1 Tablet Gen 2"),
 		},
 	},
+	{
+		.ident = "Lenovo ThinkPad X1 Fold 16 Gen 1",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "LENOVO"),
+			DMI_MATCH(DMI_PRODUCT_FAMILY, "ThinkPad X1 Fold 16 Gen 1"),
+		},
+	},
 	{
 		.ident = "Microsoft Surface Go 3",
 		.matches = {
Signature · 1 line
2.51.0
author maintainer
#312 lore raw

PATCH 301/570 platform/x86: touchscreen_dmi: Add quirk for y-inverted Goodix touchscreen on SUPI S10

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hans de Goede <[email protected]>

[ Upstream commit 7d87ed70fc95482c12edf9493c249b6413be485e ]

The touchscreen on the SUPI S10 tablet reports inverted Y coordinates, causing touch input to be mirrored vertically relative to the display.

Add a quirk to set the "touchscreen-inverted-y" boolean device-property on the touchscreen device, so that the goodix_ts driver will fixup the coordinates.

---
 drivers/platform/x86/touchscreen_dmi.c | 18 ++++++++++++++++++
 1 file changed, 18 insertions(+)
drivers/platform/x86/touchscreen_dmi.c +18 -0
index b0b1f1b201682..50b1ce7f450eb 100644
@@ -402,6 +402,16 @@ static const struct ts_dmi_data gdix1002_00_upside_down_data = {
 	.properties	= gdix1001_upside_down_props,
 };
 
+static const struct property_entry gdix1001_y_inverted_props[] = {
+	PROPERTY_ENTRY_BOOL("touchscreen-inverted-y"),
+	{ }
+};
+
+static const struct ts_dmi_data gdix1001_y_inverted_data = {
+	.acpi_name	= "GDIX1001",
+	.properties	= gdix1001_y_inverted_props,
+};
+
 static const struct property_entry gp_electronic_t701_props[] = {
 	PROPERTY_ENTRY_U32("touchscreen-size-x", 960),
 	PROPERTY_ENTRY_U32("touchscreen-size-y", 640),
@@ -1631,6 +1641,14 @@ const struct dmi_system_id touchscreen_dmi_table[] = {
 			DMI_MATCH(DMI_PRODUCT_SKU, "PN20170413488"),
 		},
 	},
+	{
+		/* SUPI S10 */
+		.driver_data = (void *)&gdix1001_y_inverted_data,
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "SUPI"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "S10"),
+		},
+	},
 	{
 		/* Techbite Arc 11.6 */
 		.driver_data = (void *)&techbite_arc_11_6_data,
Signature · 1 line
2.51.0
author maintainer
#313 lore raw

PATCH 302/570 nvme-pci: ensure were polling a polled queue

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Keith Busch <[email protected]>

[ Upstream commit 166e31d7dbf6aa44829b98aa446bda5c9580f12a ]

A user can change the polled queue count at run time. There's a brief window during a reset where a hipri task may try to poll that queue before the block layer has updated the queue maps, which would race with the now interrupt driven queue and may cause double completions.

---
 drivers/nvme/host/pci.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/nvme/host/pci.c +2 -1
index c4a33e9d2c717..432c21d3a9c4a 100644
@@ -1096,7 +1096,8 @@ static int nvme_poll(struct blk_mq_hw_ctx *hctx)
 	struct nvme_queue *nvmeq = hctx->driver_data;
 	bool found;
 
-	if (!nvme_cqe_pending(nvmeq))
+	if (!test_bit(NVMEQ_POLLED, &nvmeq->flags) ||
+	    !nvme_cqe_pending(nvmeq))
 		return 0;
 
 	spin_lock(&nvmeq->cq_poll_lock);
Signature · 1 line
2.51.0
author maintainer
#314 lore raw

PATCH 303/570 HID: magicmouse: fix battery reporting for Apple Magic Trackpad 2

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Julius Lehmann <[email protected]>

[ Upstream commit 5f3518d77419255f8b12bb23c8ec22acbeb6bc5b ]

Battery reporting does not work for the Apple Magic Trackpad 2 if it is connected via USB. The current hid descriptor fixup code checks for a hid descriptor length of exactly 83 bytes. If the hid descriptor is larger, which is the case for newer apple mice, the fixup is not applied.

This fix checks for hid descriptor sizes greater/equal 83 bytes which applies the fixup for newer devices as well.

---
 drivers/hid/hid-magicmouse.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/hid/hid-magicmouse.c +1 -1
index ec7b4f7b3d8c9..df5809cd4b637 100644
@@ -930,7 +930,7 @@ static __u8 *magicmouse_report_fixup(struct hid_device *hdev, __u8 *rdesc,
 	 */
 	if ((is_usb_magicmouse2(hdev->vendor, hdev->product) ||
 	     is_usb_magictrackpad2(hdev->vendor, hdev->product)) &&
-	    *rsize == 83 && rdesc[46] == 0x84 && rdesc[58] == 0x85) {
+	    *rsize >= 83 && rdesc[46] == 0x84 && rdesc[58] == 0x85) {
 		hid_info(hdev,
 			 "fixing up magicmouse battery report descriptor\n");
 		*rsize = *rsize - 1;
Signature · 1 line
2.51.0
author maintainer
#315 lore raw

PATCH 304/570 HID: magicmouse: avoid memory leak in magicmouse_report_fixup()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Günther Noack <[email protected]>

[ Upstream commit 91e8c6e601bdc1ccdf886479b6513c01c7e51c2c ]

The magicmouse_report_fixup() function was returning a newly kmemdup()-allocated buffer, but never freeing it.

The caller of report_fixup() does not take ownership of the returned pointer, but it *is* permitted to return a sub-portion of the input rdesc, whose lifetime is managed by the caller.

Assisted-by Gemini-CLI:Google Gemini 3
---
 drivers/hid/hid-magicmouse.c | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)
drivers/hid/hid-magicmouse.c +1 -3
index df5809cd4b637..2eda56779b4c4 100644
@@ -934,9 +934,7 @@ static __u8 *magicmouse_report_fixup(struct hid_device *hdev, __u8 *rdesc,
 		hid_info(hdev,
 			 "fixing up magicmouse battery report descriptor\n");
 		*rsize = *rsize - 1;
-		rdesc = kmemdup(rdesc + 1, *rsize, GFP_KERNEL);
-		if (!rdesc)
-			return NULL;
+		rdesc = rdesc + 1;
 
 		rdesc[0] = 0x05;
 		rdesc[1] = 0x01;
Signature · 1 line
2.51.0
author maintainer
#316 lore raw

PATCH 305/570 net: usb: r8152: add TRENDnet TUC-ET2G

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Valentin Spreckels <[email protected]>

[ Upstream commit 15fba71533bcdfaa8eeba69a5a5a2927afdf664a ]

The TRENDnet TUC-ET2G is a RTL8156 based usb ethernet adapter. Add its vendor and product IDs.

---
 drivers/net/usb/r8152.c   | 1 +
 include/linux/usb/r8152.h | 1 +
 2 files changed, 2 insertions(+)
drivers/net/usb/r8152.c +1 -0
index e70f3cb8bad94..59baa673738b6 100644
@@ -9857,6 +9857,7 @@ static const struct usb_device_id rtl8152_table[] = {
 	{ USB_DEVICE(VENDOR_ID_DLINK,   0xb301) },
 	{ USB_DEVICE(VENDOR_ID_DELL,    0xb097) },
 	{ USB_DEVICE(VENDOR_ID_ASUS,    0x1976) },
+	{ USB_DEVICE(VENDOR_ID_TRENDNET, 0xe02b) },
 	{}
 };
 
include/linux/usb/r8152.h +1 -0
index 2ca60828f28bb..1502b2a355f98 100644
@@ -32,6 +32,7 @@
 #define VENDOR_ID_DLINK			0x2001
 #define VENDOR_ID_DELL			0x413c
 #define VENDOR_ID_ASUS			0x0b05
+#define VENDOR_ID_TRENDNET		0x20f4
 
 #if IS_REACHABLE(CONFIG_USB_RTL8152)
 extern u8 rtl8152_get_version(struct usb_interface *intf);
Signature · 1 line
2.51.0
author maintainer
#317 lore raw

PATCH 306/570 HID: mcp2221: cancel last I2C command on read error

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Romain Sioen <[email protected]>

[ Upstream commit e31b556c0ba21f20c298aa61181b96541140b7b9 ]

When an I2C SMBus read operation fails, the MCP2221 internal state machine may not reset correctly, causing subsequent transactions to fail.

By adding a short delay and explicitly cancelling the last command, we ensure the device is ready for the next operation.

Fix an issue where i2cdetect was not able to detect all devices correctly on the bus.

---
 drivers/hid/hid-mcp2221.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/hid/hid-mcp2221.c +2 -0
index 589f13ff0b606..9fb98c8e1ffb7 100644
@@ -319,6 +319,8 @@ static int mcp_i2c_smbus_read(struct mcp2221 *mcp,
 				usleep_range(90, 100);
 				retries++;
 			} else {
+				usleep_range(980, 1000);
+				mcp_cancel_last_cmd(mcp);
 				return ret;
 			}
 		} else {
Signature · 1 line
2.51.0
author maintainer
#318 lore raw

PATCH 307/570 module: Fix kernel panic when a symbol st_shndx is out of bounds

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ihor Solodrai <[email protected]>

[ Upstream commit f9d69d5e7bde2295eb7488a56f094ac8f5383b92 ]

The module loader doesn't check for bounds of the ELF section index in simplify_symbols():

       for (i = 1; i < symsec->sh_size / sizeof(Elf_Sym); i++) {
		const char *name = info->strtab + sym[i].st_name;

		switch (sym[i].st_shndx) {
		case SHN_COMMON:

		[...]

		default:
			/* Divert to percpu allocation if a percpu var. */
			if (sym[i].st_shndx == info->index.pcpu)
				secbase = (unsigned long)mod_percpu(mod);
			else
  /** HERE --> **/		secbase = info->sechdrs[sym[i].st_shndx].sh_addr;
			sym[i].st_value += secbase;
			break;
		}
	}

A symbol with an out-of-bounds st_shndx value, for example 0xffff (known as SHN_XINDEX or SHN_HIRESERVE), may cause a kernel panic:

  BUG: unable to handle page fault for address: ...
  RIP: 0010:simplify_symbols+0x2b2/0x480
  ...
  Kernel panic - not syncing: Fatal exception

This can happen when module ELF is legitimately using SHN_XINDEX or when it is corrupted.

Add a bounds check in simplify_symbols() to validate that st_shndx is within the valid range before using it.

This issue was discovered due to a bug in llvm-objcopy, see relevant discussion for details [1].

[1] https://lore.kernel.org/linux-modules/[email protected]/

---
 kernel/module.c | 7 +++++++
 1 file changed, 7 insertions(+)
kernel/module.c +7 -0
index 2226b591b52e0..07fa34461fa2f 100644
@@ -2347,6 +2347,13 @@ static int simplify_symbols(struct module *mod, const struct load_info *info)
 			break;
 
 		default:
+			if (sym[i].st_shndx >= info->hdr->e_shnum) {
+				pr_err("%s: Symbol %s has an invalid section index %u (max %u)\n",
+				       mod->name, name, sym[i].st_shndx, info->hdr->e_shnum - 1);
+				ret = -ENOEXEC;
+				break;
+			}
+
 			/* Divert to percpu allocation if a percpu var. */
 			if (sym[i].st_shndx == info->index.pcpu)
 				secbase = (unsigned long)mod_percpu(mod);
Signature · 1 line
2.51.0
author maintainer
#319 lore raw

PATCH 308/570 ASoC: fsl_easrc: Fix event generation in fsl_easrc_iec958_set_reg()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mark Brown <[email protected]>

[ Upstream commit 31ddc62c1cd92e51b9db61d7954b85ae2ec224da ]

ALSA controls should return 1 if the value in the control changed but the control put operation fsl_easrc_set_reg() only returns 0 or a negative error code, causing ALSA to not generate any change events. Add a suitable check by using regmap_update_bits_check() with the underlying regmap, this is more clearly and simply correct than trying to verify that one of the generic ops is exactly equivalent to this one.

---
 sound/soc/fsl/fsl_easrc.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
sound/soc/fsl/fsl_easrc.c +6 -3
index c7ff48208d005..95ad4eeec4863 100644
@@ -93,14 +93,17 @@ static int fsl_easrc_set_reg(struct snd_kcontrol *kcontrol,
 	struct snd_soc_component *component = snd_kcontrol_chip(kcontrol);
 	struct soc_mreg_control *mc =
 		(struct soc_mreg_control *)kcontrol->private_value;
+	struct fsl_asrc *easrc = snd_soc_component_get_drvdata(component);
 	unsigned int regval = ucontrol->value.integer.value[0];
+	bool changed;
 	int ret;
 
-	ret = snd_soc_component_write(component, mc->regbase, regval);
-	if (ret < 0)
+	ret = regmap_update_bits_check(easrc->regmap, mc->regbase,
+				       GENMASK(31, 0), regval, &changed);
+	if (ret != 0)
 		return ret;
 
-	return 0;
+	return changed;
 }
 
 #define SOC_SINGLE_REG_RW(xname, xreg) \
Signature · 1 line
2.51.0
author maintainer
#320 lore raw

PATCH 309/570 ASoC: fsl_easrc: Fix event generation in fsl_easrc_iec958_put_bits()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mark Brown <[email protected]>

[ Upstream commit 54a86cf48eaa6d1ab5130d756b718775e81e1748 ]

ALSA controls should return 1 if the value in the control changed but the control put operation fsl_easrc_iec958_put_bits() unconditionally returns 0, causing ALSA to not generate any change events. This is detected by mixer-test with large numbers of messages in the form:

No event generated for Context 3 IEC958 CS5 Context 3 IEC958 CS5.0 orig 5224 read 5225, is_volatile 0

Add a suitable check.

---
 sound/soc/fsl/fsl_easrc.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
sound/soc/fsl/fsl_easrc.c +4 -1
index 95ad4eeec4863..84ef6758cc003 100644
@@ -52,10 +52,13 @@ static int fsl_easrc_iec958_put_bits(struct snd_kcontrol *kcontrol,
 	struct soc_mreg_control *mc =
 		(struct soc_mreg_control *)kcontrol->private_value;
 	unsigned int regval = ucontrol->value.integer.value[0];
+	int ret;
+
+	ret = (easrc_priv->bps_iec958[mc->regbase] != regval);
 
 	easrc_priv->bps_iec958[mc->regbase] = regval;
 
-	return 0;
+	return ret;
 }
 
 static int fsl_easrc_iec958_get_bits(struct snd_kcontrol *kcontrol,
Signature · 1 line
2.51.0
author maintainer
#321 lore raw

PATCH 310/570 dma-buf: Include ioctl.h in UAPI header

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Isaac J. Manjarres <[email protected]>

[ Upstream commit a116bac87118903925108e57781bbfc7a7eea27b ]

include/uapi/linux/dma-buf.h uses several macros from ioctl.h to define its ioctl commands. However, it does not include ioctl.h itself. So, if userspace source code tries to include the dma-buf.h file without including ioctl.h, it can result in build failures.

Therefore, include ioctl.h in the dma-buf UAPI header.

---
 include/uapi/linux/dma-buf.h | 1 +
 1 file changed, 1 insertion(+)
include/uapi/linux/dma-buf.h +1 -0
index b1523cb8ab307..5e2b1949ffd49 100644
@@ -20,6 +20,7 @@
 #ifndef _DMA_BUF_UAPI_H_
 #define _DMA_BUF_UAPI_H_
 
+#include <linux/ioctl.h>
 #include <linux/types.h>
 
 /**
Signature · 1 line
2.51.0
author maintainer
#322 lore raw

PATCH 311/570 ALSA: hda/realtek: Add headset jack quirk for Thinkpad X390

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Uzair Mughal <[email protected]>

[ Upstream commit 542127f6528ca7cc3cf61e1651d6ccb58495f953 ]

The Lenovo ThinkPad X390 (ALC257 codec, subsystem ID 0x17aa2288) does not report headset button press events. Headphone insertion is detected (SW_HEADPHONE_INSERT), but pressing the inline microphone button on a headset produces no input events.

Add a SND_PCI_QUIRK entry that maps this subsystem ID to ALC285_FIXUP_THINKPAD_NO_BASS_SPK_HEADSET_JACK, which enables headset jack button detection through alc_fixup_headset_jack() and ThinkPad ACPI integration. This is the same fixup used by similar ThinkPad models (P1 Gen 3, X1 Extreme Gen 3).

---
 sound/pci/hda/patch_realtek.c | 1 +
 1 file changed, 1 insertion(+)
sound/pci/hda/patch_realtek.c +1 -0
index 72d9ea5171bbd..38fda5dbd75ba 100644
@@ -9608,6 +9608,7 @@ static const struct snd_pci_quirk alc269_fixup_tbl[] = {
 	SND_PCI_QUIRK(0x17aa, 0x224c, "Thinkpad", ALC298_FIXUP_TPT470_DOCK),
 	SND_PCI_QUIRK(0x17aa, 0x224d, "Thinkpad", ALC298_FIXUP_TPT470_DOCK),
 	SND_PCI_QUIRK(0x17aa, 0x225d, "Thinkpad T480", ALC269_FIXUP_LIMIT_INT_MIC_BOOST),
+	SND_PCI_QUIRK(0x17aa, 0x2288, "Thinkpad X390", ALC285_FIXUP_THINKPAD_NO_BASS_SPK_HEADSET_JACK),
 	SND_PCI_QUIRK(0x17aa, 0x2292, "Thinkpad X1 Carbon 7th", ALC285_FIXUP_THINKPAD_HEADSET_JACK),
 	SND_PCI_QUIRK(0x17aa, 0x22be, "Thinkpad X1 Carbon 8th", ALC285_FIXUP_THINKPAD_HEADSET_JACK),
 	SND_PCI_QUIRK(0x17aa, 0x22c1, "Thinkpad P1 Gen 3", ALC285_FIXUP_THINKPAD_NO_BASS_SPK_HEADSET_JACK),
Signature · 1 line
2.51.0
author maintainer
#323 lore raw

PATCH 312/570 xfrm: call xdo_dev_state_delete during state update

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sabrina Dubroca <[email protected]>

[ Upstream commit 7d2fc41f91bc69acb6e01b0fa23cd7d0109a6a23 ]

When we update an SA, we construct a new state and call xdo_dev_state_add, but never insert it. The existing state is updated, then we immediately destroy the new state. Since we haven't added it, we don't go through the standard state delete code, and we're skipping removing it from the device (but xdo_dev_state_free will get called when we destroy the temporary state).

This is similar to commit c5d4d7d83165 ("xfrm: Fix deletion of offloaded SAs on failure.").

Fixes d77e38e612a0 ("xfrm: Add an IPsec hardware offloading API")
---
 net/xfrm/xfrm_state.c | 1 +
 1 file changed, 1 insertion(+)
net/xfrm/xfrm_state.c +1 -0
index 54ae99f69f25f..f7f568bfb93a8 100644
@@ -1770,6 +1770,7 @@ int xfrm_state_update(struct xfrm_state *x)
 
 		err = 0;
 		x->km.state = XFRM_STATE_DEAD;
+		xfrm_dev_state_delete(x);
 		__xfrm_state_put(x);
 	}
 
Signature · 1 line
2.51.0
author maintainer
#324 lore raw

PATCH 313/570 xfrm: Fix the usage of skb->sk

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Steffen Klassert <[email protected]>

[ Upstream commit 1620c88887b16940e00dbe57dd38c74eda9bad9e ]

xfrm assumed to always have a full socket at skb->sk. This is not always true, so fix it by converting to a full socket before it is used.

Stable-dep-of: 0c0eef8ccd24 ("esp: fix skb leak with espintcp and async crypto")

---
 net/ipv4/esp4.c                | 2 +-
 net/ipv6/esp6.c                | 2 +-
 net/ipv6/xfrm6_output.c        | 4 ++--
 net/xfrm/xfrm_interface_core.c | 2 +-
 net/xfrm/xfrm_output.c         | 7 ++++---
 net/xfrm/xfrm_policy.c         | 2 +-
 6 files changed, 10 insertions(+), 9 deletions(-)
net/ipv4/esp4.c +1 -1
index 53170ecb2de08..272b64fd09eed 100644
@@ -278,7 +278,7 @@ static void esp_output_done(struct crypto_async_request *base, int err)
 		    x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP)
 			esp_output_tail_tcp(x, skb);
 		else
-			xfrm_output_resume(skb->sk, skb, err);
+			xfrm_output_resume(skb_to_full_sk(skb), skb, err);
 	}
 }
 
net/ipv6/esp6.c +1 -1
index cbe575ade34d6..57e48dd905a48 100644
@@ -315,7 +315,7 @@ static void esp_output_done(struct crypto_async_request *base, int err)
 		    x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP)
 			esp_output_tail_tcp(x, skb);
 		else
-			xfrm_output_resume(skb->sk, skb, err);
+			xfrm_output_resume(skb_to_full_sk(skb), skb, err);
 	}
 }
 
net/ipv6/xfrm6_output.c +2 -2
index ad07904642cad..ff183bd76c998 100644
@@ -82,14 +82,14 @@ static int __xfrm6_output(struct net *net, struct sock *sk, struct sk_buff *skb)
 
 	toobig = skb->len > mtu && !skb_is_gso(skb);
 
-	if (toobig && xfrm6_local_dontfrag(skb->sk)) {
+	if (toobig && xfrm6_local_dontfrag(sk)) {
 		xfrm6_local_rxpmtu(skb, mtu);
 		kfree_skb(skb);
 		return -EMSGSIZE;
 	} else if (toobig && xfrm6_noneed_fragment(skb)) {
 		skb->ignore_df = 1;
 		goto skip_frag;
-	} else if (!skb->ignore_df && toobig && skb->sk) {
+	} else if (!skb->ignore_df && toobig && sk) {
 		xfrm_local_error(skb, mtu);
 		kfree_skb(skb);
 		return -EMSGSIZE;
net/xfrm/xfrm_interface_core.c +1 -1
index 9bd69887e16d2..4baa7a61df0e6 100644
@@ -369,7 +369,7 @@ xfrmi_xmit2(struct sk_buff *skb, struct net_device *dev, struct flowi *fl)
 	skb_dst_set(skb, dst);
 	skb->dev = tdev;
 
-	err = dst_output(xi->net, skb->sk, skb);
+	err = dst_output(xi->net, skb_to_full_sk(skb), skb);
 	if (net_xmit_eval(err) == 0) {
 		dev_sw_netstats_tx_add(dev, 1, length);
 	} else {
net/xfrm/xfrm_output.c +4 -3
index 29ce7f6f16a09..3e27daceebfe4 100644
@@ -781,7 +781,7 @@ static int xfrm4_tunnel_check_size(struct sk_buff *skb)
 	     !skb_gso_validate_network_len(skb, ip_skb_dst_mtu(skb->sk, skb)))) {
 		skb->protocol = htons(ETH_P_IP);
 
-		if (skb->sk)
+		if (skb->sk && sk_fullsock(skb->sk))
 			xfrm_local_error(skb, mtu);
 		else
 			icmp_send(skb, ICMP_DEST_UNREACH,
@@ -817,6 +817,7 @@ static int xfrm6_tunnel_check_size(struct sk_buff *skb)
 {
 	int mtu, ret = 0;
 	struct dst_entry *dst = skb_dst(skb);
+	struct sock *sk = skb_to_full_sk(skb);
 
 	if (skb->ignore_df)
 		goto out;
@@ -831,9 +832,9 @@ static int xfrm6_tunnel_check_size(struct sk_buff *skb)
 		skb->dev = dst->dev;
 		skb->protocol = htons(ETH_P_IPV6);
 
-		if (xfrm6_local_dontfrag(skb->sk))
+		if (xfrm6_local_dontfrag(sk))
 			ipv6_stub->xfrm6_local_rxpmtu(skb, mtu);
-		else if (skb->sk)
+		else if (sk)
 			xfrm_local_error(skb, mtu);
 		else
 			icmpv6_send(skb, ICMPV6_PKT_TOOBIG, 0, mtu);
net/xfrm/xfrm_policy.c +1 -1
index 16958656b6d43..851029a5383a2 100644
@@ -2856,7 +2856,7 @@ static void xfrm_policy_queue_process(struct timer_list *t)
 		skb_dst_drop(skb);
 		skb_dst_set(skb, dst);
 
-		dst_output(net, skb->sk, skb);
+		dst_output(net, skb_to_full_sk(skb), skb);
 	}
 
 out:
Signature · 1 line
2.51.0
author maintainer
#325 lore raw

PATCH 314/570 esp: fix skb leak with espintcp and async crypto

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sabrina Dubroca <[email protected]>

[ Upstream commit 0c0eef8ccd2413b0a10eb6bbd3442333b1e64dd2 ]

When the TX queue for espintcp is full, esp_output_tail_tcp will return an error and not free the skb, because with synchronous crypto, the common xfrm output code will drop the packet for us.

With async crypto (esp_output_done), we need to drop the skb when esp_output_tail_tcp returns an error.

Fixes e27cca96cd68 ("xfrm: add espintcp (RFC 8229)")
---
 net/ipv4/esp4.c | 9 ++++++---
 net/ipv6/esp6.c | 9 ++++++---
 2 files changed, 12 insertions(+), 6 deletions(-)
net/ipv4/esp4.c +6 -3
index 272b64fd09eed..c69cee3feff00 100644
@@ -275,10 +275,13 @@ static void esp_output_done(struct crypto_async_request *base, int err)
 		xfrm_dev_resume(skb);
 	} else {
 		if (!err &&
-		    x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP)
-			esp_output_tail_tcp(x, skb);
-		else
+		    x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP) {
+			err = esp_output_tail_tcp(x, skb);
+			if (err != -EINPROGRESS)
+				kfree_skb(skb);
+		} else {
 			xfrm_output_resume(skb_to_full_sk(skb), skb, err);
+		}
 	}
 }
 
net/ipv6/esp6.c +6 -3
index 57e48dd905a48..e87f3f8f06818 100644
@@ -312,10 +312,13 @@ static void esp_output_done(struct crypto_async_request *base, int err)
 		xfrm_dev_resume(skb);
 	} else {
 		if (!err &&
-		    x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP)
-			esp_output_tail_tcp(x, skb);
-		else
+		    x->encap && x->encap->encap_type == TCP_ENCAP_ESPINTCP) {
+			err = esp_output_tail_tcp(x, skb);
+			if (err != -EINPROGRESS)
+				kfree_skb(skb);
+		} else {
 			xfrm_output_resume(skb_to_full_sk(skb), skb, err);
+		}
 	}
 }
 
Signature · 1 line
2.51.0
author maintainer
#326 lore raw

PATCH 315/570 af_key: validate families in pfkey_send_migrate()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit eb2d16a7d599dc9d4df391b5e660df9949963786 ]

syzbot was able to trigger a crash in skb_put() [1]

Issue is that pfkey_send_migrate() does not check old/new families, and that set_ipsecrequest() @family argument was truncated, thus possibly overfilling the skb.

Validate families early, do not wait set_ipsecrequest().

[1]

skbuff: skb_over_panic: text:ffffffff8a752120 len:392 put:16 head:ffff88802a4ad040 data:ffff88802a4ad040 tail:0x188 end:0x180 dev:<NULL>
kernel BUG at net/core/skbuff.c:214 !

Call Trace:
 <TASK>
  skb_over_panic net/core/skbuff.c:219 [inline]
  skb_put+0x159/0x210 net/core/skbuff.c:2655
  skb_put_zero include/linux/skbuff.h:2788 [inline]
  set_ipsecrequest net/key/af_key.c:3532 [inline]
  pfkey_send_migrate+0x1270/0x2e50 net/key/af_key.c:3636
  km_migrate+0x155/0x260 net/xfrm/xfrm_state.c:2848
  xfrm_migrate+0x2140/0x2450 net/xfrm/xfrm_policy.c:4705
  xfrm_do_migrate+0x8ff/0xaa0 net/xfrm/xfrm_user.c:3150
Fixes 08de61beab8a ("[PFKEYV2]: Extension for dynamic update of endpoint address(es)")
Reported-by [email protected]
Cc Steffen Klassert <[email protected]>
Cc Herbert Xu <[email protected]>
---
 net/key/af_key.c | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)
net/key/af_key.c +12 -7
index 925fe4f89966b..638333f1b8c7d 100644
@@ -3518,7 +3518,7 @@ static int set_sadb_kmaddress(struct sk_buff *skb, const struct xfrm_kmaddress *
 
 static int set_ipsecrequest(struct sk_buff *skb,
 			    uint8_t proto, uint8_t mode, int level,
-			    uint32_t reqid, uint8_t family,
+			    uint32_t reqid, sa_family_t family,
 			    const xfrm_address_t *src, const xfrm_address_t *dst)
 {
 	struct sadb_x_ipsecrequest *rq;
@@ -3583,12 +3583,17 @@ static int pfkey_send_migrate(const struct xfrm_selector *sel, u8 dir, u8 type,
 
 	/* ipsecrequests */
 	for (i = 0, mp = m; i < num_bundles; i++, mp++) {
-		/* old locator pair */
-		size_pol += sizeof(struct sadb_x_ipsecrequest) +
-			    pfkey_sockaddr_pair_size(mp->old_family);
-		/* new locator pair */
-		size_pol += sizeof(struct sadb_x_ipsecrequest) +
-			    pfkey_sockaddr_pair_size(mp->new_family);
+		int pair_size;
+
+		pair_size = pfkey_sockaddr_pair_size(mp->old_family);
+		if (!pair_size)
+			return -EINVAL;
+		size_pol += sizeof(struct sadb_x_ipsecrequest) + pair_size;
+
+		pair_size = pfkey_sockaddr_pair_size(mp->new_family);
+		if (!pair_size)
+			return -EINVAL;
+		size_pol += sizeof(struct sadb_x_ipsecrequest) + pair_size;
 	}
 
 	size += sizeof(struct sadb_msg) + size_pol;
Signature · 1 line
2.51.0
author maintainer
#327 lore raw

PATCH 316/570 can: statistics: add missing atomic access in hot path

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Oliver Hartkopp <[email protected]>

[ Upstream commit 46eee1661aa9b49966e6c43d07126fe408edda57 ]

Commit 80b5f90158d1 ("can: statistics: use atomic access in hot path") fixed a KCSAN issue in can_receive() but missed to convert the 'matches' variable used in can_rcv_filter().

Fixes 80b5f90158d1 ("can: statistics: use atomic access in hot path")
---
 net/can/af_can.c | 4 ++--
 net/can/af_can.h | 2 +-
 net/can/proc.c   | 3 ++-
 3 files changed, 5 insertions(+), 4 deletions(-)
net/can/af_can.c +2 -2
index edf01b73d2878..85b01dea76dff 100644
@@ -469,7 +469,7 @@ int can_rx_register(struct net *net, struct net_device *dev, canid_t can_id,
 
 	rcv->can_id = can_id;
 	rcv->mask = mask;
-	rcv->matches = 0;
+	atomic_long_set(&rcv->matches, 0);
 	rcv->func = func;
 	rcv->data = data;
 	rcv->ident = ident;
@@ -573,7 +573,7 @@ EXPORT_SYMBOL(can_rx_unregister);
 static inline void deliver(struct sk_buff *skb, struct receiver *rcv)
 {
 	rcv->func(skb, rcv->data);
-	rcv->matches++;
+	atomic_long_inc(&rcv->matches);
 }
 
 static int can_rcv_filter(struct can_dev_rcv_lists *dev_rcv_lists, struct sk_buff *skb)
net/can/af_can.h +1 -1
index 22f3352c77fec..87887014f5628 100644
@@ -52,7 +52,7 @@ struct receiver {
 	struct hlist_node list;
 	canid_t can_id;
 	canid_t mask;
-	unsigned long matches;
+	atomic_long_t matches;
 	void (*func)(struct sk_buff *skb, void *data);
 	void *data;
 	char *ident;
net/can/proc.c +2 -1
index 0533a3c4ff0e1..f81f8a698071e 100644
@@ -196,7 +196,8 @@ static void can_print_rcvlist(struct seq_file *m, struct hlist_head *rx_list,
 			"   %-5s     %03x    %08x  %pK  %pK  %8ld  %s\n";
 
 		seq_printf(m, fmt, DNAME(dev), r->can_id, r->mask,
-				r->func, r->data, r->matches, r->ident);
+			   r->func, r->data, atomic_long_read(&r->matches),
+			   r->ident);
 	}
 }
 
Signature · 1 line
2.51.0
author maintainer
#328 lore raw

PATCH 317/570 Bluetooth: L2CAP: Validate PDU length before reading SDU length in l2cap_ecred_data_rcv()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hyunwoo Kim <[email protected]>

[ Upstream commit c65bd945d1c08c3db756821b6bf9f1c4a77b29c6 ]

l2cap_ecred_data_rcv() reads the SDU length field from skb->data using get_unaligned_le16() without first verifying that skb contains at least L2CAP_SDULEN_SIZE (2) bytes. When skb->len is less than 2, this reads past the valid data in the skb.

The ERTM reassembly path correctly calls pskb_may_pull() before reading the SDU length (l2cap_reassemble_sdu, L2CAP_SAR_START case). Apply the same validation to the Enhanced Credit Based Flow Control data path.

Fixes aac23bf63659 ("Bluetooth: Implement LE L2CAP reassembly")
---
 net/bluetooth/l2cap_core.c | 5 +++++
 1 file changed, 5 insertions(+)
net/bluetooth/l2cap_core.c +5 -0
index 5010c200b2c41..e69c6041437b0 100644
@@ -7650,6 +7650,11 @@ static int l2cap_ecred_data_rcv(struct l2cap_chan *chan, struct sk_buff *skb)
 	if (!chan->sdu) {
 		u16 sdu_len;
 
+		if (!pskb_may_pull(skb, L2CAP_SDULEN_SIZE)) {
+			err = -EINVAL;
+			goto failed;
+		}
+
 		sdu_len = get_unaligned_le16(skb->data);
 		skb_pull(skb, L2CAP_SDULEN_SIZE);
 
Signature · 1 line
2.51.0
author maintainer
#329 lore raw

PATCH 318/570 Bluetooth: SCO: Fix use-after-free in sco_recv_frame() due to missing sock_hold

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hyunwoo Kim <[email protected]>

[ Upstream commit 598dbba9919c5e36c54fe1709b557d64120cb94b ]

sco_recv_frame() reads conn->sk under sco_conn_lock() but immediately releases the lock without holding a reference to the socket. A concurrent close() can free the socket between the lock release and the subsequent sk->sk_state access, resulting in a use-after-free.

Other functions in the same file (sco_sock_timeout(), sco_conn_del()) correctly use sco_sock_hold() to safely hold a reference under the lock.

Fix by using sco_sock_hold() to take a reference before releasing the lock, and adding sock_put() on all exit paths.

Fixes 1da177e4c3f4 ("Linux-2.6.12-rc2")
---
 net/bluetooth/sco.c | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)
net/bluetooth/sco.c +7 -3
index d98648bcc1a85..d0ef74c45914c 100644
@@ -311,7 +311,7 @@ static void sco_recv_frame(struct sco_conn *conn, struct sk_buff *skb)
 	struct sock *sk;
 
 	sco_conn_lock(conn);
-	sk = conn->sk;
+	sk = sco_sock_hold(conn);
 	sco_conn_unlock(conn);
 
 	if (!sk)
@@ -320,11 +320,15 @@ static void sco_recv_frame(struct sco_conn *conn, struct sk_buff *skb)
 	BT_DBG("sk %p len %u", sk, skb->len);
 
 	if (sk->sk_state != BT_CONNECTED)
-		goto drop;
+		goto drop_put;
 
-	if (!sock_queue_rcv_skb(sk, skb))
+	if (!sock_queue_rcv_skb(sk, skb)) {
+		sock_put(sk);
 		return;
+	}
 
+drop_put:
+	sock_put(sk);
 drop:
 	kfree_skb(skb);
 }
Signature · 1 line
2.51.0
author maintainer
#330 lore raw

PATCH 319/570 Bluetooth: hci_ll: Fix firmware leak on error path

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Anas Iqbal <[email protected]>

[ Upstream commit 31148a7be723aa9f2e8fbd62424825ab8d577973 ]

Smatch reports:

drivers/bluetooth/hci_ll.c:587 download_firmware() warn:
'fw' from request_firmware() not released on lines: 544.

In download_firmware(), if request_firmware() succeeds but the returned firmware content is invalid (no data or zero size), the function returns without releasing the firmware, resulting in a resource leak.

Fix this by calling release_firmware() before returning when request_firmware() succeeded but the firmware content is invalid.

Fixes 371805522f87 ("bluetooth: hci_uart: add LL protocol serdev driver support")
---
 drivers/bluetooth/hci_ll.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/bluetooth/hci_ll.c +2 -0
index e4e5b26e2c33b..02c2122b452a8 100644
@@ -541,6 +541,8 @@ static int download_firmware(struct ll_device *lldev)
 	if (err || !fw->data || !fw->size) {
 		bt_dev_err(lldev->hu.hdev, "request_firmware failed(errno %d) for %s",
 			   err, bts_scr_name);
+		if (!err)
+			release_firmware(fw);
 		return -EINVAL;
 	}
 	ptr = (void *)fw->data;
Signature · 1 line
2.51.0
author maintainer
#331 lore raw

PATCH 320/570 Bluetooth: L2CAP: Fix null-ptr-deref on l2cap_sock_ready_cb

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Helen Koike <[email protected]>

[ Upstream commit b6552e0503973daf6f23bd6ed9273ef131ee364f ]

Before using sk pointer, check if it is null.

Fix the following:

 KASAN: null-ptr-deref in range [0x0000000000000260-0x0000000000000267]
 CPU: 0 UID: 0 PID: 5985 Comm: kworker/0:5 Not tainted 7.0.0-rc4-00029-ga989fde763f4 #1 PREEMPT(full)
 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.17.0-9.fc43 06/10/2025
 Workqueue: events l2cap_info_timeout
 RIP: 0010:kasan_byte_accessible+0x12/0x30
 Code: 79 ff ff ff 0f 1f 40 00 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 0f 1f 40 d6 48 c1 ef 03 48 b8 00 00 00 00 00 fc ff df <0f> b6 04 07 3c 08 0f 92 c0 c3 cc cce
 veth0_macvtap: entered promiscuous mode
 RSP: 0018:ffffc90006e0f808 EFLAGS: 00010202
 RAX: dffffc0000000000 RBX: ffffffff89746018 RCX: 0000000080000001
 RDX: 0000000000000000 RSI: ffffffff89746018 RDI: 000000000000004c
 RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000000
 R10: dffffc0000000000 R11: ffffffff8aae3e70 R12: 0000000000000000
 R13: 0000000000000260 R14: 0000000000000260 R15: 0000000000000001
 FS:  0000000000000000(0000) GS:ffff8880983c2000(0000) knlGS:0000000000000000
 CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
 CR2: 00005582615a5008 CR3: 000000007007e000 CR4: 0000000000752ef0
 PKRU: 55555554
 Call Trace:
  <TASK>
  __kasan_check_byte+0x12/0x40
  lock_acquire+0x79/0x2e0
  lock_sock_nested+0x48/0x100
  ? l2cap_sock_ready_cb+0x46/0x160
  l2cap_sock_ready_cb+0x46/0x160
  l2cap_conn_start+0x779/0xff0
  ? __pfx_l2cap_conn_start+0x10/0x10
  ? l2cap_info_timeout+0x60/0xa0
  ? __pfx___mutex_lock+0x10/0x10
  l2cap_info_timeout+0x68/0xa0
  ? process_scheduled_works+0xa8d/0x18c0
  process_scheduled_works+0xb6e/0x18c0
  ? __pfx_process_scheduled_works+0x10/0x10
  ? assign_work+0x3d5/0x5e0
  worker_thread+0xa53/0xfc0
  kthread+0x388/0x470
  ? __pfx_worker_thread+0x10/0x10
  ? __pfx_kthread+0x10/0x10
  ret_from_fork+0x51e/0xb90
  ? __pfx_ret_from_fork+0x10/0x10
 veth1_macvtap: entered promiscuous mode
  ? __switch_to+0xc7d/0x1450
  ? __pfx_kthread+0x10/0x10
  ret_from_fork_asm+0x1a/0x30
  </TASK>
 Modules linked in:
 ---[ end trace 0000000000000000 ]---
 batman_adv: batadv0: Interface activated: batadv_slave_0
 batman_adv: batadv0: Interface activated: batadv_slave_1
 netdevsim netdevsim7 netdevsim0: set [1, 0] type 2 family 0 port 6081 - 0
 netdevsim netdevsim7 netdevsim1: set [1, 0] type 2 family 0 port 6081 - 0
 netdevsim netdevsim7 netdevsim2: set [1, 0] type 2 family 0 port 6081 - 0
 netdevsim netdevsim7 netdevsim3: set [1, 0] type 2 family 0 port 6081 - 0
 RIP: 0010:kasan_byte_accessible+0x12/0x30
 Code: 79 ff ff ff 0f 1f 40 00 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 90 0f 1f 40 d6 48 c1 ef 03 48 b8 00 00 00 00 00 fc ff df <0f> b6 04 07 3c 08 0f 92 c0 c3 cc cce
 ieee80211 phy39: Selected rate control algorithm 'minstrel_ht'
 RSP: 0018:ffffc90006e0f808 EFLAGS: 00010202
 RAX: dffffc0000000000 RBX: ffffffff89746018 RCX: 0000000080000001
 RDX: 0000000000000000 RSI: ffffffff89746018 RDI: 000000000000004c
 RBP: 0000000000000000 R08: 0000000000000001 R09: 0000000000000000
 R10: dffffc0000000000 R11: ffffffff8aae3e70 R12: 0000000000000000
 R13: 0000000000000260 R14: 0000000000000260 R15: 0000000000000001
 FS:  0000000000000000(0000) GS:ffff8880983c2000(0000) knlGS:0000000000000000
 CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
 CR2: 00007f7e16139e9c CR3: 000000000e74e000 CR4: 0000000000752ef0
 PKRU: 55555554
 Kernel panic - not syncing: Fatal exception
Fixes 54a59aa2b562 ("Bluetooth: Add l2cap_chan->ops->ready()")
---
 net/bluetooth/l2cap_sock.c | 3 +++
 1 file changed, 3 insertions(+)
net/bluetooth/l2cap_sock.c +3 -0
index 8e2e6d1a6dd15..faaa5e4525c07 100644
@@ -1649,6 +1649,9 @@ static void l2cap_sock_ready_cb(struct l2cap_chan *chan)
 	struct sock *sk = chan->data;
 	struct sock *parent;
 
+	if (!sk)
+		return;
+
 	lock_sock(sk);
 
 	parent = bt_sk(sk)->parent;
Signature · 1 line
2.51.0
author maintainer
#332 lore raw

PATCH 321/570 pinctrl: mediatek: common: Fix probe failure for devices without EINT

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Luca Leonardo Scorcia <[email protected]>

[ Upstream commit 8f9f64c8f90dca07d3b9f1d7ce5d34ccd246c9dd ]

Some pinctrl devices like mt6397 or mt6392 don't support EINT at all, but the mtk_eint_init function is always called and returns -ENODEV, which then bubbles up and causes probe failure.

To address this only call mtk_eint_init if EINT pins are present.

Tested on Xiaomi Mi Smart Clock x04g (mt6392).

Fixes e46df235b4e6 ("pinctrl: mediatek: refactor EINT related code for all MediaTek pinctrl can fit")
---
 drivers/pinctrl/mediatek/pinctrl-mtk-common.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
drivers/pinctrl/mediatek/pinctrl-mtk-common.c +6 -3
index 334cb85855a93..a3f1ff77e0646 100644
@@ -1110,9 +1110,12 @@ int mtk_pctrl_init(struct platform_device *pdev,
 		goto chip_error;
 	}
 
-	ret = mtk_eint_init(pctl, pdev);
-	if (ret)
-		goto chip_error;
+	/* Only initialize EINT if we have EINT pins */
+	if (data->eint_hw.ap_num > 0) {
+		ret = mtk_eint_init(pctl, pdev);
+		if (ret)
+			goto chip_error;
+	}
 
 	return 0;
 
Signature · 1 line
2.51.0
author maintainer
#333 lore raw

PATCH 322/570 ionic: fix persistent MAC address override on PF

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mohammad Heib <[email protected]>

[ Upstream commit cbcb3cfcdc436d6f91a3d95ecfa9c831abe14aed ]

The use of IONIC_CMD_LIF_SETATTR in the MAC address update path causes the ionic firmware to update the LIF's identity in its persistent state. Since the firmware state is maintained across host warm boots and driver reloads, any MAC change on the Physical Function (PF) becomes "sticky.

This is problematic because it causes ethtool -P to report the user-configured MAC as the permanent factory address, which breaks system management tools that rely on a stable hardware identity.

While Virtual Functions (VFs) need this hardware-level programming to properly handle MAC assignments in guest environments, the PF should maintain standard transient behavior. This patch gates the ionic_program_mac call using is_virtfn so that PF MAC changes remain local to the netdev filters and do not overwrite the firmware's permanent identity block.

Fixes 19058be7c48c ("ionic: VF initial random MAC address if no assigned mac")
---
 drivers/net/ethernet/pensando/ionic/ionic_lif.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)
drivers/net/ethernet/pensando/ionic/ionic_lif.c +11 -6
index cdc3c55fab6ac..0e44e616c6da1 100644
@@ -1763,13 +1763,18 @@ static int ionic_set_mac_address(struct net_device *netdev, void *sa)
 	if (ether_addr_equal(netdev->dev_addr, mac))
 		return 0;
 
-	err = ionic_program_mac(lif, mac);
-	if (err < 0)
-		return err;
+	/* Only program macs for virtual functions to avoid losing the permanent
+	 * Mac across warm reset/reboot.
+	 */
+	if (lif->ionic->pdev->is_virtfn) {
+		err = ionic_program_mac(lif, mac);
+		if (err < 0)
+			return err;
 
-	if (err > 0)
-		netdev_dbg(netdev, "%s: SET and GET ATTR Mac are not equal-due to old FW running\n",
-			   __func__);
+		if (err > 0)
+			netdev_dbg(netdev, "%s: SET and GET ATTR Mac are not equal-due to old FW running\n",
+				   __func__);
+	}
 
 	err = eth_prepare_mac_addr_change(netdev, addr);
 	if (err)
Signature · 1 line
2.51.0
author maintainer
#334 lore raw

PATCH 323/570 nfc: nci: fix circular locking dependency in nci_close_device

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jakub Kicinski <[email protected]>

[ Upstream commit 4527025d440ce84bf56e75ce1df2e84cb8178616 ]

nci_close_device() flushes rx_wq and tx_wq while holding req_lock. This causes a circular locking dependency because nci_rx_work() running on rx_wq can end up taking req_lock too:

nci_rx_work -> nci_rx_data_packet -> nci_data_exchange_complete

    -> __sk_destruct -> rawsock_destruct -> nfc_deactivate_target
    -> nci_deactivate_target -> nci_request -> mutex_lock(&ndev->req_lock)

Move the flush of rx_wq after req_lock has been released. This should safe (I think) because NCI_UP has already been cleared and the transport is closed, so the work will see it and return -ENETDOWN.

NIPA has been hitting this running the nci selftest with a debug kernel on roughly 4% of the runs.

Fixes 6a2968aaf50c ("NFC: basic NCI protocol implementation")
---
 net/nfc/nci/core.c | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)
net/nfc/nci/core.c +6 -4
index ca9543be400c4..992bd13649cb4 100644
@@ -574,8 +574,7 @@ static int nci_close_device(struct nci_dev *ndev)
 	skb_queue_purge(&ndev->rx_q);
 	skb_queue_purge(&ndev->tx_q);
 
-	/* Flush RX and TX wq */
-	flush_workqueue(ndev->rx_wq);
+	/* Flush TX wq, RX wq flush can't be under the lock */
 	flush_workqueue(ndev->tx_wq);
 
 	/* Reset device */
@@ -587,13 +586,13 @@ static int nci_close_device(struct nci_dev *ndev)
 		      msecs_to_jiffies(NCI_RESET_TIMEOUT));
 
 	/* After this point our queues are empty
-	 * and no works are scheduled.
+	 * rx work may be running but will see that NCI_UP was cleared
 	 */
 	ndev->ops->close(ndev);
 
 	clear_bit(NCI_INIT, &ndev->flags);
 
-	/* Flush cmd wq */
+	/* Flush cmd and tx wq */
 	flush_workqueue(ndev->cmd_wq);
 
 	del_timer_sync(&ndev->cmd_timer);
@@ -603,6 +602,9 @@ static int nci_close_device(struct nci_dev *ndev)
 
 	mutex_unlock(&ndev->req_lock);
 
+	/* rx_work may take req_lock via nci_deactivate_target */
+	flush_workqueue(ndev->rx_wq);
+
 	return 0;
 }
 
Signature · 1 line
2.51.0
author maintainer
#335 lore raw

PATCH 324/570 net: openvswitch: Avoid releasing netdev before teardown completes

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Toke Høiland-Jørgensen <[email protected]>

[ Upstream commit 7c770dadfda5cbbde6aa3c4363ed513f1d212bf8 ]

The patch cited in the Fixes tag below changed the teardown code for OVS ports to no longer unconditionally take the RTNL. After this change, the netdev_destroy() callback can proceed immediately to the call_rcu() invocation if the IFF_OVS_DATAPATH flag is already cleared on the netdev.

The ovs_netdev_detach_dev() function clears the flag before completing the unregistration, and if it gets preempted after clearing the flag (as can happen on an -rt kernel), netdev_destroy() can complete and the device can be freed before the unregistration completes. This leads to a splat like:

[  998.393867] Oops: general protection fault, probably for non-canonical address 0xff00000001000239: 0000 [#1] SMP PTI
[  998.393877] CPU: 42 UID: 0 PID: 55177 Comm: ip Kdump: loaded Not tainted 6.12.0-211.1.1.el10_2.x86_64+rt #1 PREEMPT_RT
[  998.393886] Hardware name: Dell Inc. PowerEdge R740/0JMK61, BIOS 2.24.0 03/27/2025
[  998.393889] RIP: 0010:dev_set_promiscuity+0x8d/0xa0
[  998.393901] Code: 00 00 75 d8 48 8b 53 08 48 83 ba b0 02 00 00 00 75 ca 48 83 c4 08 5b c3 cc cc cc cc 48 83 bf 48 09 00 00 00 75 91 48 8b 47 08 <48> 83 b8 b0 02 00 00 00 74 97 eb 81 0f 1f 80 00 00 00 00 90 90 90
[  998.393906] RSP: 0018:ffffce5864a5f6a0 EFLAGS: 00010246
[  998.393912] RAX: ff00000000ffff89 RBX: ffff894d0adf5a05 RCX: 0000000000000000
[  998.393917] RDX: 0000000000000000 RSI: 00000000ffffffff RDI: ffff894d0adf5a05
[  998.393921] RBP: ffff894d19252000 R08: ffff894d19252000 R09: 0000000000000000
[  998.393924] R10: ffff894d19252000 R11: ffff894d192521b8 R12: 0000000000000006
[  998.393927] R13: ffffce5864a5f738 R14: 00000000ffffffe2 R15: 0000000000000000
[  998.393931] FS:  00007fad61971800(0000) GS:ffff894cc0140000(0000) knlGS:0000000000000000
[  998.393936] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  998.393940] CR2: 000055df0a2a6e40 CR3: 000000011c7fe003 CR4: 00000000007726f0
[  998.393944] PKRU: 55555554
[  998.393946] Call Trace:
[  998.393949]  <TASK>
[  998.393952]  ? show_trace_log_lvl+0x1b0/0x2f0
[  998.393961]  ? show_trace_log_lvl+0x1b0/0x2f0
[  998.393975]  ? dp_device_event+0x41/0x80 [openvswitch]
[  998.394009]  ? __die_body.cold+0x8/0x12
[  998.394016]  ? die_addr+0x3c/0x60
[  998.394027]  ? exc_general_protection+0x16d/0x390
[  998.394042]  ? asm_exc_general_protection+0x26/0x30
[  998.394058]  ? dev_set_promiscuity+0x8d/0xa0
[  998.394066]  ? ovs_netdev_detach_dev+0x3a/0x80 [openvswitch]
[  998.394092]  dp_device_event+0x41/0x80 [openvswitch]
[  998.394102]  notifier_call_chain+0x5a/0xd0
[  998.394106]  unregister_netdevice_many_notify+0x51b/0xa60
[  998.394110]  rtnl_dellink+0x169/0x3e0
[  998.394121]  ? rt_mutex_slowlock.constprop.0+0x95/0xd0
[  998.394125]  rtnetlink_rcv_msg+0x142/0x3f0
[  998.394128]  ? avc_has_perm_noaudit+0x69/0xf0
[  998.394130]  ? __pfx_rtnetlink_rcv_msg+0x10/0x10
[  998.394132]  netlink_rcv_skb+0x50/0x100
[  998.394138]  netlink_unicast+0x292/0x3f0
[  998.394141]  netlink_sendmsg+0x21b/0x470
[  998.394145]  ____sys_sendmsg+0x39d/0x3d0
[  998.394149]  ___sys_sendmsg+0x9a/0xe0
[  998.394156]  __sys_sendmsg+0x7a/0xd0
[  998.394160]  do_syscall_64+0x7f/0x170
[  998.394162]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[  998.394165] RIP: 0033:0x7fad61bf4724
[  998.394188] Code: 89 02 b8 ff ff ff ff eb bb 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 00 f3 0f 1e fa 80 3d c5 e9 0c 00 00 74 13 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 54 c3 0f 1f 00 48 83 ec 28 89 54 24 1c 48 89
[  998.394189] RSP: 002b:00007ffd7e2f7cb8 EFLAGS: 00000202 ORIG_RAX: 000000000000002e
[  998.394191] RAX: ffffffffffffffda RBX: 0000000000000001 RCX: 00007fad61bf4724
[  998.394193] RDX: 0000000000000000 RSI: 00007ffd7e2f7d20 RDI: 0000000000000003
[  998.394194] RBP: 00007ffd7e2f7d90 R08: 0000000000000010 R09: 000000000000003f
[  998.394195] R10: 000055df11558010 R11: 0000000000000202 R12: 00007ffd7e2f8380
[  998.394196] R13: 0000000069b233d7 R14: 000055df0a256040 R15: 0000000000000000
[  998.394200]  </TASK>

To fix this, reorder the operations in ovs_netdev_detach_dev() to only clear the flag after completing the other operations, and introduce an smp_wmb() to make the ordering requirement explicit. The smp_wmb() is paired with a full smp_mb() in netdev_destroy() to make sure the call_rcu() invocation does not happen before the unregister operations are visible.

Fixes 549822767630 ("net: openvswitch: Avoid needlessly taking the RTNL on vport destroy")
---
 net/openvswitch/vport-netdev.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)
net/openvswitch/vport-netdev.c +8 -1
index 3beec619283a9..10e30a85d5c5c 100644
@@ -145,11 +145,15 @@ static void vport_netdev_free(struct rcu_head *rcu)
 void ovs_netdev_detach_dev(struct vport *vport)
 {
 	ASSERT_RTNL();
-	vport->dev->priv_flags &= ~IFF_OVS_DATAPATH;
 	netdev_rx_handler_unregister(vport->dev);
 	netdev_upper_dev_unlink(vport->dev,
 				netdev_master_upper_dev_get(vport->dev));
 	dev_set_promiscuity(vport->dev, -1);
+
+	/* paired with smp_mb() in netdev_destroy() */
+	smp_wmb();
+
+	vport->dev->priv_flags &= ~IFF_OVS_DATAPATH;
 }
 
 static void netdev_destroy(struct vport *vport)
@@ -168,6 +172,9 @@ static void netdev_destroy(struct vport *vport)
 		rtnl_unlock();
 	}
 
+	/* paired with smp_wmb() in ovs_netdev_detach_dev() */
+	smp_mb();
+
 	call_rcu(&vport->rcu, vport_netdev_free);
 }
 
Signature · 1 line
2.51.0
author maintainer
#336 lore raw

PATCH 325/570 openvswitch: validate MPLS set/set_masked payload length

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yang Yang <[email protected]>

[ Upstream commit 546b68ac893595877ffbd7751e5c55fd1c43ede6 ]

validate_set() accepted OVS_KEY_ATTR_MPLS as variable-sized payload for SET/SET_MASKED actions. In action handling, OVS expects fixed-size MPLS key data (struct ovs_key_mpls).

Use the already normalized key_len (masked case included) and reject non-matching MPLS action key sizes.

Reject invalid MPLS action payload lengths early.

---
 net/openvswitch/flow_netlink.c | 2 ++
 1 file changed, 2 insertions(+)
net/openvswitch/flow_netlink.c +2 -0
index a7a9e4df3f600..1b2941e9c6d14 100644
@@ -2908,6 +2908,8 @@ static int validate_set(const struct nlattr *a,
 	case OVS_KEY_ATTR_MPLS:
 		if (!eth_p_mpls(eth_type))
 			return -EINVAL;
+		if (key_len != sizeof(struct ovs_key_mpls))
+			return -EINVAL;
 		break;
 
 	case OVS_KEY_ATTR_SCTP:
Signature · 1 line
2.51.0
author maintainer
#337 lore raw

PATCH 326/570 net/smc: fix double-free of smc_spd_priv when tee() duplicates splice pipe buffer

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Qi Tang <[email protected]>

[ Upstream commit 24dd586bb4cbba1889a50abe74143817a095c1c9 ]

smc_rx_splice() allocates one smc_spd_priv per pipe_buffer and stores the pointer in pipe_buffer.private. The pipe_buf_operations for these buffers used .get = generic_pipe_buf_get, which only increments the page reference count when tee(2) duplicates a pipe buffer. The smc_spd_priv pointer itself was not handled, so after tee() both the original and the cloned pipe_buffer share the same smc_spd_priv *.

When both pipes are subsequently released, smc_rx_pipe_buf_release() is called twice against the same object:

  1st call: kfree(priv)  sock_put(sk)  smc_rx_update_cons()  [correct]
  2nd call: kfree(priv)  sock_put(sk)  smc_rx_update_cons()  [UAF]

KASAN reports a slab-use-after-free in smc_rx_pipe_buf_release(), which then escalates to a NULL-pointer dereference and kernel panic via smc_rx_update_consumer() when it chases the freed priv->smc pointer:

  BUG: KASAN: slab-use-after-free in smc_rx_pipe_buf_release+0x78/0x2a0
  Read of size 8 at addr ffff888004a45740 by task smc_splice_tee_/74
  Call Trace:
   <TASK>
   dump_stack_lvl+0x53/0x70
   print_report+0xce/0x650
   kasan_report+0xc6/0x100
   smc_rx_pipe_buf_release+0x78/0x2a0
   free_pipe_info+0xd4/0x130
   pipe_release+0x142/0x160
   __fput+0x1c6/0x490
   __x64_sys_close+0x4f/0x90
   do_syscall_64+0xa6/0x1a0
   entry_SYSCALL_64_after_hwframe+0x77/0x7f
   </TASK>

  BUG: kernel NULL pointer dereference, address: 0000000000000020
  RIP: 0010:smc_rx_update_consumer+0x8d/0x350
  Call Trace:
   <TASK>
   smc_rx_pipe_buf_release+0x121/0x2a0
   free_pipe_info+0xd4/0x130
   pipe_release+0x142/0x160
   __fput+0x1c6/0x490
   __x64_sys_close+0x4f/0x90
   do_syscall_64+0xa6/0x1a0
   entry_SYSCALL_64_after_hwframe+0x77/0x7f
   </TASK>
  Kernel panic - not syncing: Fatal exception

Beyond the memory-safety problem, duplicating an SMC splice buffer is semantically questionable: smc_rx_update_cons() would advance the consumer cursor twice for the same data, corrupting receive-window accounting. A refcount on smc_spd_priv could fix the double-free, but the cursor-accounting issue would still need to be addressed separately.

The .get callback is invoked by both tee(2) and splice_pipe_to_pipe() for partial transfers; both will now return -EFAULT. Users who need to duplicate SMC socket data must use a copy-based read path.

Fixes 9014db202cb7 ("smc: add support for splice()")
---
 net/smc/smc_rx.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)
net/smc/smc_rx.c +8 -1
index 81cf611eae750..3468c58bbeebe 100644
@@ -130,9 +130,16 @@ static void smc_rx_pipe_buf_release(struct pipe_inode_info *pipe,
 	sock_put(sk);
 }
 
+static bool smc_rx_pipe_buf_get(struct pipe_inode_info *pipe,
+				struct pipe_buffer *buf)
+{
+	/* smc_spd_priv in buf->private is not shareable; disallow cloning. */
+	return false;
+}
+
 static const struct pipe_buf_operations smc_pipe_ops = {
 	.release = smc_rx_pipe_buf_release,
-	.get = generic_pipe_buf_get
+	.get	 = smc_rx_pipe_buf_get,
 };
 
 static void smc_rx_spd_release(struct splice_pipe_desc *spd,
Signature · 1 line
2.51.0
author maintainer
#338 lore raw

PATCH 327/570 rtnetlink: count IFLA_INFO_SLAVE_KIND in if_nlmsg_size

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sabrina Dubroca <[email protected]>

[ Upstream commit ee00a12593ffb69db4dd1a1c00ecb0253376874a ]

rtnl_link_get_slave_info_data_size counts IFLA_INFO_SLAVE_DATA, but rtnl_link_slave_info_fill adds both IFLA_INFO_SLAVE_DATA and IFLA_INFO_SLAVE_KIND.

---
 net/core/rtnetlink.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
net/core/rtnetlink.c +6 -3
index 674f33bae66e2..89c22b66886d4 100644
@@ -532,11 +532,14 @@ static size_t rtnl_link_get_slave_info_data_size(const struct net_device *dev)
 		goto out;
 
 	ops = master_dev->rtnl_link_ops;
-	if (!ops || !ops->get_slave_size)
+	if (!ops)
+		goto out;
+	size += nla_total_size(strlen(ops->kind) + 1);  /* IFLA_INFO_SLAVE_KIND */
+	if (!ops->get_slave_size)
 		goto out;
 	/* IFLA_INFO_SLAVE_DATA + nested data */
-	size = nla_total_size(sizeof(struct nlattr)) +
-	       ops->get_slave_size(master_dev, dev);
+	size += nla_total_size(sizeof(struct nlattr)) +
+		ops->get_slave_size(master_dev, dev);
 
 out:
 	rcu_read_unlock();
Signature · 1 line
2.51.0
author maintainer
#339 lore raw

PATCH 328/570 platform/olpc: olpc-xo175-ec: Fix overflow error message to print inlen

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Alok Tiwari <[email protected]>

[ Upstream commit 2061f7b042f88d372cca79615f8425f3564c0b40 ]

The command length check validates inlen (> 5), but the error message incorrectly printed resp_len. Print inlen so the log reflects the actual command length.

---
 drivers/platform/olpc/olpc-xo175-ec.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/platform/olpc/olpc-xo175-ec.c +1 -1
index 0d46706afd2da..d8a9a215a14cd 100644
@@ -482,7 +482,7 @@ static int olpc_xo175_ec_cmd(u8 cmd, u8 *inbuf, size_t inlen, u8 *resp,
 	dev_dbg(dev, "CMD %x, %zd bytes expected\n", cmd, resp_len);
 
 	if (inlen > 5) {
-		dev_err(dev, "command len %zd too big!\n", resp_len);
+		dev_err(dev, "command len %zd too big!\n", inlen);
 		return -EOVERFLOW;
 	}
 
Signature · 1 line
2.51.0
author maintainer
#340 lore raw

PATCH 329/570 net: fix fanout UAF in packet_release() via NETDEV_UP race

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yochai Eisenrich <[email protected]>

[ Upstream commit 42156f93d123436f2a27c468f18c966b7e5db796 ]

packet_release() has a race window where NETDEV_UP can re-register a socket into a fanout group's arr[] array. The re-registration is not cleaned up by fanout_release(), leaving a dangling pointer in the fanout array.
packet_release() does NOT zero po->num in its bind_lock section. After releasing bind_lock, po->num is still non-zero and po->ifindex still matches the bound device. A concurrent packet_notifier(NETDEV_UP) that already found the socket in sklist can re-register the hook. For fanout sockets, this re-registration calls __fanout_link(sk, po) which adds the socket back into f->arr[] and increments f->num_members, but does NOT increment f->sk_ref.

The fix sets po->num to zero in packet_release while bind_lock is held to prevent NETDEV_UP from linking, preventing the race window.

This bug was found following an additional audit with Claude Code based on CVE-2025-38617.

---
 net/packet/af_packet.c | 1 +
 1 file changed, 1 insertion(+)
net/packet/af_packet.c +1 -0
index 6e7c94fa02bd9..d1ad069271f8b 100644
@@ -3147,6 +3147,7 @@ static int packet_release(struct socket *sock)
 
 	spin_lock(&po->bind_lock);
 	unregister_prot_hook(sk, false);
+	WRITE_ONCE(po->num, 0);
 	packet_cached_dev_reset(po);
 
 	if (po->prot_hook.dev) {
Signature · 1 line
2.51.0
author maintainer
#341 lore raw

PATCH 330/570 net: enetc: fix the output issue of ethtool --show-ring

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Wei Fang <[email protected]>

[ Upstream commit 70b439bf06f6a12e491f827fa81a9887a11501f9 ]

Currently, enetc_get_ringparam() only provides rx_pending and tx_pending, but 'ethtool --show-ring' no longer displays these fields. Because the ringparam retrieval path has moved to the new netlink interface, where rings_fill_reply() emits the *x_pending only if the *x_max_pending values are non-zero. So rx_max_pending and tx_max_pending to are added to enetc_get_ringparam() to fix the issue.

Note that the maximum tx/rx ring size of hardware is 64K, but we haven't added set_ringparam() to make the ring size configurable. To avoid users mistakenly believing that the ring size can be increased, so set the *x_max_pending to priv->*x_bd_count.

Fixes e4a1717b677c ("ethtool: provide ring sizes with RINGS_GET request")
---
 drivers/net/ethernet/freescale/enetc/enetc_ethtool.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/net/ethernet/freescale/enetc/enetc_ethtool.c +2 -0
index d62c188c87480..89234613ef806 100644
@@ -566,6 +566,8 @@ static void enetc_get_ringparam(struct net_device *ndev,
 {
 	struct enetc_ndev_priv *priv = netdev_priv(ndev);
 
+	ring->rx_max_pending = priv->rx_bd_count;
+	ring->tx_max_pending = priv->tx_bd_count;
 	ring->rx_pending = priv->rx_bd_count;
 	ring->tx_pending = priv->tx_bd_count;
 
Signature · 1 line
2.51.0
author maintainer
#342 lore raw

PATCH 331/570 dma-mapping: add missing `inline` for `dma_free_attrs`

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Miguel Ojeda <[email protected]>

[ Upstream commit 2cdaff22ed26f1e619aa2b43f27bb84f2c6ef8f8 ]

Under an UML build for an upcoming series [1], I got -Wstatic-in-inline for dma_free_attrs:

BINDGEN rust/bindings/bindings_generated.rs - due to target missing
In file included from rust/helpers/helpers.c:59:
rust/helpers/dma.c:17:2: warning: static function 'dma_free_attrs' is used in an inline function with external linkage [-Wstatic-in-inline]
17 | dma_free_attrs(dev, size, cpu_addr, dma_handle, attrs);
| ^
rust/helpers/dma.c:12:1: note: use 'static' to give inline function 'rust_helper_dma_free_attrs' internal linkage
12 | __rust_helper void rust_helper_dma_free_attrs(struct device *dev, size_t size,
| ^
| static

The issue is that dma_free_attrs was not marked inline when it was introduced alongside the rest of the stubs.

Thus mark it.

---
 include/linux/dma-mapping.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
include/linux/dma-mapping.h +2 -2
index d7b91f82b0dce..96f4d63390aea 100644
@@ -218,8 +218,8 @@ static inline void *dma_alloc_attrs(struct device *dev, size_t size,
 {
 	return NULL;
 }
-static void dma_free_attrs(struct device *dev, size_t size, void *cpu_addr,
-		dma_addr_t dma_handle, unsigned long attrs)
+static inline void dma_free_attrs(struct device *dev, size_t size,
+		void *cpu_addr, dma_addr_t dma_handle, unsigned long attrs)
 {
 }
 static inline void *dmam_alloc_attrs(struct device *dev, size_t size,
Signature · 1 line
2.51.0
author maintainer
#343 lore raw

PATCH 332/570 Bluetooth: L2CAP: Fix ERTM re-init and zero pdu_len infinite loop

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hyunwoo Kim <[email protected]>

[ Upstream commit 25f420a0d4cfd61d3d23ec4b9c56d9f443d91377 ]

l2cap_config_req() processes CONFIG_REQ for channels in BT_CONNECTED state to support L2CAP reconfiguration (e.g. MTU changes). However, since both CONF_INPUT_DONE and CONF_OUTPUT_DONE are already set from the initial configuration, the reconfiguration path falls through to l2cap_ertm_init(), which re-initializes tx_q, srej_q, srej_list, and retrans_list without freeing the previous allocations and sets 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 the double-init by skipping l2cap_ertm_init() and l2cap_chan_ready() when the channel is already in BT_CONNECTED state, while still allowing the reconfiguration parameters to be updated through l2cap_parse_conf_req(). Also add 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 | 19 ++++++++++++-------
 1 file changed, 12 insertions(+), 7 deletions(-)
net/bluetooth/l2cap_core.c +12 -7
index e69c6041437b0..ef71a43ab702d 100644
@@ -2541,6 +2541,9 @@ 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;
@@ -4515,14 +4518,16 @@ static inline int l2cap_config_req(struct l2cap_conn *conn,
 	if (test_bit(CONF_INPUT_DONE, &chan->conf_state)) {
 		set_default_fcs(chan);
 
-		if (chan->mode == L2CAP_MODE_ERTM ||
-		    chan->mode == L2CAP_MODE_STREAMING)
-			err = l2cap_ertm_init(chan);
+		if (chan->state != BT_CONNECTED) {
+			if (chan->mode == L2CAP_MODE_ERTM ||
+			    chan->mode == L2CAP_MODE_STREAMING)
+				err = l2cap_ertm_init(chan);
 
-		if (err < 0)
-			l2cap_send_disconn_req(chan, -err);
-		else
-			l2cap_chan_ready(chan);
+			if (err < 0)
+				l2cap_send_disconn_req(chan, -err);
+			else
+				l2cap_chan_ready(chan);
+		}
 
 		goto unlock;
 	}
Signature · 1 line
2.51.0
author maintainer
#344 lore raw

PATCH 333/570 Bluetooth: btusb: clamp SCO altsetting table indices

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pengpeng Hou <[email protected]>

[ Upstream commit 129fa608b6ad08b8ab7178eeb2ec272c993aaccc ]

btusb_work() maps the number of active SCO links to USB alternate settings through a three-entry lookup table when CVSD traffic uses transparent voice settings. The lookup currently indexes alts[] with data->sco_num - 1 without first constraining sco_num to the number of available table entries.

While the table only defines alternate settings for up to three SCO links, data->sco_num comes from hci_conn_num() and is used directly. Cap the lookup to the last table entry before indexing it so the driver keeps selecting the highest supported alternate setting without reading past alts[].

Fixes baac6276c0a9 ("Bluetooth: btusb: handle mSBC audio over USB Endpoints")
---
 drivers/bluetooth/btusb.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/bluetooth/btusb.c +4 -1
index a79fd106fad7a..12bcc07e2e502 100644
@@ -1845,8 +1845,11 @@ static void btusb_work(struct work_struct *work)
 		if (data->air_mode == HCI_NOTIFY_ENABLE_SCO_CVSD) {
 			if (hdev->voice_setting & 0x0020) {
 				static const int alts[3] = { 2, 4, 5 };
+				unsigned int sco_idx;
 
-				new_alts = alts[data->sco_num - 1];
+				sco_idx = min_t(unsigned int, data->sco_num - 1,
+						ARRAY_SIZE(alts) - 1);
+				new_alts = alts[sco_idx];
 			} else {
 				new_alts = data->sco_num;
 			}
Signature · 1 line
2.51.0
author maintainer
#345 lore raw

PATCH 334/570 netfilter: nfnetlink_log: fix uninitialized padding leak in NFULA_PAYLOAD

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Weiming Shi <[email protected]>

[ Upstream commit 52025ebaa29f4eb4ed8bf92ce83a68f24ab7fdf7 ]

__build_packet_message() manually constructs the NFULA_PAYLOAD netlink attribute using skb_put() and skb_copy_bits(), bypassing the standard nla_reserve()/nla_put() helpers. While nla_total_size(data_len) bytes are allocated (including NLA alignment padding), only data_len bytes of actual packet data are copied. The trailing nla_padlen(data_len) bytes (1-3 when data_len is not 4-byte aligned) are never initialized, leaking stale heap contents to userspace via the NFLOG netlink socket.

Replace the manual attribute construction with nla_reserve(), which handles the tailroom check, header setup, and padding zeroing via __nla_reserve(). The subsequent skb_copy_bits() fills in the payload data on top of the properly initialized attribute.

Fixes df6fb868d611 ("[NETFILTER]: nfnetlink: convert to generic netlink attribute functions")
---
 net/netfilter/nfnetlink_log.c | 8 ++------
 1 file changed, 2 insertions(+), 6 deletions(-)
net/netfilter/nfnetlink_log.c +2 -6
index 09fe6cf358ec7..6484d08223dbc 100644
@@ -636,15 +636,11 @@ __build_packet_message(struct nfnl_log_net *log,
 
 	if (data_len) {
 		struct nlattr *nla;
-		int size = nla_attr_size(data_len);
 
-		if (skb_tailroom(inst->skb) < nla_total_size(data_len))
+		nla = nla_reserve(inst->skb, NFULA_PAYLOAD, data_len);
+		if (!nla)
 			goto nla_put_failure;
 
-		nla = skb_put(inst->skb, nla_total_size(data_len));
-		nla->nla_type = NFULA_PAYLOAD;
-		nla->nla_len = size;
-
 		if (skb_copy_bits(skb, 0, nla_data(nla), data_len))
 			BUG();
 	}
Signature · 1 line
2.51.0
author maintainer
#346 lore raw

PATCH 335/570 netfilter: ip6t_rt: reject oversized addrnr in rt_mt6_check()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ren Wei <[email protected]>

[ Upstream commit 9d3f027327c2fa265f7f85ead41294792c3296ed ]

Reject rt match rules whose addrnr exceeds IP6T_RT_HOPS.

rt_mt6() expects addrnr to stay within the bounds of rtinfo->addrs[]. Validate addrnr during rule installation so malformed rules are rejected before the match logic can use an out-of-range value.

---
 net/ipv6/netfilter/ip6t_rt.c | 4 ++++
 1 file changed, 4 insertions(+)
net/ipv6/netfilter/ip6t_rt.c +4 -0
index 4ad8b2032f1f9..5561bd9cea818 100644
@@ -157,6 +157,10 @@ static int rt_mt6_check(const struct xt_mtchk_param *par)
 		pr_debug("unknown flags %X\n", rtinfo->invflags);
 		return -EINVAL;
 	}
+	if (rtinfo->addrnr > IP6T_RT_HOPS) {
+		pr_debug("too many addresses specified\n");
+		return -EINVAL;
+	}
 	if ((rtinfo->flags & (IP6T_RT_RES | IP6T_RT_FST_MASK)) &&
 	    (!(rtinfo->flags & IP6T_RT_TYP) ||
 	     (rtinfo->rt_type != 0) ||
Signature · 1 line
2.51.0
author maintainer
#347 lore raw

PATCH 336/570 netfilter: nf_conntrack_expect: skip expectations in other netns via proc

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Pablo Neira Ayuso <[email protected]>

[ Upstream commit 3db5647984de03d9cae0dcddb509b058351f0ee4 ]

Skip expectations that do not reside in this netns.

Similar to e77e6ff502ea ("netfilter: conntrack: do not dump other netns's conntrack entries via proc").

Fixes 9b03f38d0487 ("netfilter: netns nf_conntrack: per-netns expectations")
---
 net/netfilter/nf_conntrack_expect.c | 4 ++++
 1 file changed, 4 insertions(+)
net/netfilter/nf_conntrack_expect.c +4 -0
index 6d056ebba57c6..10d4dfbdde226 100644
@@ -627,11 +627,15 @@ static int exp_seq_show(struct seq_file *s, void *v)
 {
 	struct nf_conntrack_expect *expect;
 	struct nf_conntrack_helper *helper;
+	struct net *net = seq_file_net(s);
 	struct hlist_node *n = v;
 	char *delim = "";
 
 	expect = hlist_entry(n, struct nf_conntrack_expect, hnode);
 
+	if (!net_eq(nf_ct_exp_net(expect), net))
+		return 0;
+
 	if (expect->timeout.function)
 		seq_printf(s, "%ld ", timer_pending(&expect->timeout)
 			   ? (long)(expect->timeout.expires - jiffies)/HZ : 0);
Signature · 1 line
2.51.0
#348 lore raw

Re: [PATCH 5.15 336/570] netfilter: nf_conntrack_expect: skip expectations in other netns via proc

Hi,

Quoted 4 lines · Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

Quoted 9 lines

From: Pablo Neira Ayuso <[email protected]>

[ Upstream commit 3db5647984de03d9cae0dcddb509b058351f0ee4 ]

Skip expectations that do not reside in this netns.

Similar to e77e6ff502ea ("netfilter: conntrack: do not dump other netns's conntrack entries via proc").

AI assisted review spotted a probable issue: I have gone through the analysis and the summary is:

I think this fix relies on commit: 02a3231b6d82 ("netfilter: nf_conntrack_expect: store netns and zone in expectation")

This references commit explicitly states:
" This patch is required by the follow up fix not to dump expectations that do not belong
to this netns." which is this patch.

Also part of patch series 4 and 5: https://lore.kernel.org/all/[email protected]/

Given that we haven't taken 02a3231b6d82 ("netfilter: nf_conntrack_expect: store netns and zone in expectation") to 5.15.y should we drop this ?

Why ? Without it, the 5.15 backport still uses master-conntrack-derived context instead of expectation-owned stored netns/zone state

i.e Upstream has:

possible_net_t net;

static inline struct net *nf_ct_exp_net(struct nf_conntrack_expect *exp)

{
      return read_pnet(&exp->net);
}

Downstream has:

static inline struct net *nf_ct_exp_net(struct nf_conntrack_expect *exp)

{
      return nf_ct_net(exp->master);
}

I don't know the internals of this fully, but looks like we might not want to take this fix without 02a3231b6d82 ("netfilter: nf_conntrack_expect: store netns and zone in expectation")

Thanks,
Harshit

Quoted 28 lines
Fixes 9b03f38d0487 ("netfilter: netns nf_conntrack: per-netns expectations")
---
  net/netfilter/nf_conntrack_expect.c | 4 ++++
  1 file changed, 4 insertions(+)
net/netfilter/nf_conntrack_expect.c +4 -0
index 6d056ebba57c6..10d4dfbdde226 100644
@@ -627,11 +627,15 @@ static int exp_seq_show(struct seq_file *s, void *v)
  {
  	struct nf_conntrack_expect *expect;
  	struct nf_conntrack_helper *helper;
+	struct net *net = seq_file_net(s);
  	struct hlist_node *n = v;
  	char *delim = "";
 
  	expect = hlist_entry(n, struct nf_conntrack_expect, hnode);
 
+	if (!net_eq(nf_ct_exp_net(expect), net))
+		return 0;
+
  	if (expect->timeout.function)
  		seq_printf(s, "%ld ", timer_pending(&expect->timeout)
  			   ? (long)(expect->timeout.expires - jiffies)/HZ : 0);
#349 lore raw

Re: [PATCH 5.15 336/570] netfilter: nf_conntrack_expect: skip expectations in other netns via proc

Quoted 50 lines · Harshit Mogalapalli

Hi,

Quoted 4 lines · Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

Quoted 9 lines

From: Pablo Neira Ayuso <[email protected]>

[ Upstream commit 3db5647984de03d9cae0dcddb509b058351f0ee4 ]

Skip expectations that do not reside in this netns.

Similar to e77e6ff502ea ("netfilter: conntrack: do not dump other netns's conntrack entries via proc").

AI assisted review spotted a probable issue: I have gone through the analysis and the summary is:

I think this fix relies on commit: 02a3231b6d82 ("netfilter:
nf_conntrack_expect: store netns and zone in expectation")

This references commit explicitly states:
" This patch is required by the follow up fix not to dump expectations that do not belong
to this netns." which is this patch.

Also part of patch series 4 and 5:
https://lore.kernel.org/all/[email protected]/

Given that we haven't taken 02a3231b6d82 ("netfilter: nf_conntrack_expect:
store netns and zone in expectation") to 5.15.y should we drop this ?

Why ? Without it, the 5.15 backport still uses master-conntrack-derived context instead of expectation-owned stored netns/zone state

i.e Upstream has:

possible_net_t net;

static inline struct net *nf_ct_exp_net(struct nf_conntrack_expect *exp)

{
      return read_pnet(&exp->net);
}

Downstream has:

static inline struct net *nf_ct_exp_net(struct nf_conntrack_expect *exp)

Quoted 8 lines
{
      return nf_ct_net(exp->master);
}

I don't know the internals of this fully, but looks like we might not want to take this fix without 02a3231b6d82 ("netfilter: nf_conntrack_expect:
store netns and zone in expectation")

Yes:

02a3231b6d82 ("netfilter: nf_conntrack_expect: store netns and zone in expectation")

is good to have as a Stable-Dep.

Quoted 35 lines

Thanks,
Harshit

Quoted 26 lines
Fixes 9b03f38d0487 ("netfilter: netns nf_conntrack: per-netns expectations")
---
  net/netfilter/nf_conntrack_expect.c | 4 ++++
  1 file changed, 4 insertions(+)
net/netfilter/nf_conntrack_expect.c +4 -0
index 6d056ebba57c6..10d4dfbdde226 100644
@@ -627,11 +627,15 @@ static int exp_seq_show(struct seq_file *s, void *v)
  {
  	struct nf_conntrack_expect *expect;
  	struct nf_conntrack_helper *helper;
+	struct net *net = seq_file_net(s);
  	struct hlist_node *n = v;
  	char *delim = "";
  	expect = hlist_entry(n, struct nf_conntrack_expect, hnode);
+	if (!net_eq(nf_ct_exp_net(expect), net))
+		return 0;
+
  	if (expect->timeout.function)
  		seq_printf(s, "%ld ", timer_pending(&expect->timeout)
  			   ? (long)(expect->timeout.expires - jiffies)/HZ : 0);
maintainer
#350 lore raw

Re: [PATCH 5.15 336/570] netfilter: nf_conntrack_expect: skip expectations in other netns via proc

Quoted 2 lines

02a3231b6d82 ("netfilter: nf_conntrack_expect: store netns and zone in expectation") is good to have as a Stable-Dep.

Dropped from the 5.15 and 5.10 queues until 02a3231b6d82 is properly backported, thanks.

author maintainer
#351 lore raw

PATCH 337/570 netfilter: nf_conntrack_sip: fix use of uninitialized rtp_addr in process_sdp

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Weiming Shi <[email protected]>

[ Upstream commit 6a2b724460cb67caed500c508c2ae5cf012e4db4 ]

process_sdp() declares union nf_inet_addr rtp_addr on the stack and passes it to the nf_nat_sip sdp_session hook after walking the SDP media descriptions. However rtp_addr is only initialized inside the media loop when a recognized media type with a non-zero port is found.

If the SDP body contains no m= lines, only inactive media sections (m=audio 0 ...) or only unrecognized media types, rtp_addr is never assigned. Despite that, the function still calls hooks->sdp_session() with &rtp_addr, causing nf_nat_sdp_session() to format the stale stack value as an IP address and rewrite the SDP session owner and connection lines with it.

With CONFIG_INIT_STACK_ALL_ZERO (default on most distributions) this results in the session-level o= and c= addresses being rewritten to 0.0.0.0 for inactive SDP sessions. Without stack auto-init the rewritten address is whatever happened to be on the stack.

Fix this by pre-initializing rtp_addr from the session-level connection address (caddr) when available, and tracking via a have_rtp_addr flag whether any valid address was established. Skip the sdp_session hook entirely when no valid address exists.

Fixes 4ab9e64e5e3c ("[NETFILTER]: nf_nat_sip: split up SDP mangling")
---
 net/netfilter/nf_conntrack_sip.c | 14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)
net/netfilter/nf_conntrack_sip.c +10 -4
index 5db17768ec2ad..dcb0a5e592775 100644
@@ -1040,6 +1040,7 @@ static int process_sdp(struct sk_buff *skb, unsigned int protoff,
 	unsigned int port;
 	const struct sdp_media_type *t;
 	int ret = NF_ACCEPT;
+	bool have_rtp_addr = false;
 
 	hooks = rcu_dereference(nf_nat_sip_hooks);
 
@@ -1056,8 +1057,11 @@ static int process_sdp(struct sk_buff *skb, unsigned int protoff,
 	caddr_len = 0;
 	if (ct_sip_parse_sdp_addr(ct, *dptr, sdpoff, *datalen,
 				  SDP_HDR_CONNECTION, SDP_HDR_MEDIA,
-				  &matchoff, &matchlen, &caddr) > 0)
+				  &matchoff, &matchlen, &caddr) > 0) {
 		caddr_len = matchlen;
+		memcpy(&rtp_addr, &caddr, sizeof(rtp_addr));
+		have_rtp_addr = true;
+	}
 
 	mediaoff = sdpoff;
 	for (i = 0; i < ARRAY_SIZE(sdp_media_types); ) {
@@ -1091,9 +1095,11 @@ static int process_sdp(struct sk_buff *skb, unsigned int protoff,
 					  &matchoff, &matchlen, &maddr) > 0) {
 			maddr_len = matchlen;
 			memcpy(&rtp_addr, &maddr, sizeof(rtp_addr));
-		} else if (caddr_len)
+			have_rtp_addr = true;
+		} else if (caddr_len) {
 			memcpy(&rtp_addr, &caddr, sizeof(rtp_addr));
-		else {
+			have_rtp_addr = true;
+		} else {
 			nf_ct_helper_log(skb, ct, "cannot parse SDP message");
 			return NF_DROP;
 		}
@@ -1125,7 +1131,7 @@ static int process_sdp(struct sk_buff *skb, unsigned int protoff,
 
 	/* Update session connection and owner addresses */
 	hooks = rcu_dereference(nf_nat_sip_hooks);
-	if (hooks && ct->status & IPS_NAT_MASK)
+	if (hooks && ct->status & IPS_NAT_MASK && have_rtp_addr)
 		ret = hooks->sdp_session(skb, protoff, dataoff,
 					 dptr, datalen, sdpoff,
 					 &rtp_addr);
Signature · 1 line
2.51.0
author maintainer
#352 lore raw

PATCH 338/570 netlink: introduce NLA_POLICY_MAX_BE

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Florian Westphal <[email protected]>

[ Upstream commit 08724ef69907214ce622344fe4945412e38368f0 ]

netlink allows to specify allowed ranges for integer types. Unfortunately, nfnetlink passes integers in big endian, so the existing NLA_POLICY_MAX() cannot be used.

At the moment, nfnetlink users, such as nf_tables, need to resort to programmatic checking via helpers such as nft_parse_u32_check().

This is both cumbersome and error prone. This adds NLA_POLICY_MAX_BE which adds range check support for BE16, BE32 and BE64 integers.

Stable-dep-of: 8f15b5071b45 ("netfilter: ctnetlink: use netlink policy range checks")

---
 include/net/netlink.h |  9 +++++++++
 lib/nlattr.c          | 31 +++++++++++++++++++++++++++----
 2 files changed, 36 insertions(+), 4 deletions(-)
include/net/netlink.h +9 -0
index 7a2a9d3144ba6..6eb4593983319 100644
@@ -325,6 +325,7 @@ struct nla_policy {
 		struct netlink_range_validation_signed *range_signed;
 		struct {
 			s16 min, max;
+			u8 network_byte_order:1;
 		};
 		int (*validate)(const struct nlattr *attr,
 				struct netlink_ext_ack *extack);
@@ -418,6 +419,14 @@ struct nla_policy {
 	.type = NLA_ENSURE_INT_OR_BINARY_TYPE(tp),	\
 	.validation_type = NLA_VALIDATE_MAX,		\
 	.max = _max,					\
+	.network_byte_order = 0,			\
+}
+
+#define NLA_POLICY_MAX_BE(tp, _max) {			\
+	.type = NLA_ENSURE_UINT_TYPE(tp),		\
+	.validation_type = NLA_VALIDATE_MAX,		\
+	.max = _max,					\
+	.network_byte_order = 1,			\
 }
 
 #define NLA_POLICY_MASK(tp, _mask) {			\
lib/nlattr.c +27 -4
index 73635bdb00620..a0c80fbf71895 100644
@@ -160,6 +160,31 @@ void nla_get_range_unsigned(const struct nla_policy *pt,
 	}
 }
 
+static u64 nla_get_attr_bo(const struct nla_policy *pt,
+			   const struct nlattr *nla)
+{
+	switch (pt->type) {
+	case NLA_U16:
+		if (pt->network_byte_order)
+			return ntohs(nla_get_be16(nla));
+
+		return nla_get_u16(nla);
+	case NLA_U32:
+		if (pt->network_byte_order)
+			return ntohl(nla_get_be32(nla));
+
+		return nla_get_u32(nla);
+	case NLA_U64:
+		if (pt->network_byte_order)
+			return be64_to_cpu(nla_get_be64(nla));
+
+		return nla_get_u64(nla);
+	}
+
+	WARN_ON_ONCE(1);
+	return 0;
+}
+
 static int nla_validate_range_unsigned(const struct nla_policy *pt,
 				       const struct nlattr *nla,
 				       struct netlink_ext_ack *extack,
@@ -173,12 +198,10 @@ static int nla_validate_range_unsigned(const struct nla_policy *pt,
 		value = nla_get_u8(nla);
 		break;
 	case NLA_U16:
-		value = nla_get_u16(nla);
-		break;
 	case NLA_U32:
-		value = nla_get_u32(nla);
-		break;
 	case NLA_U64:
+		value = nla_get_attr_bo(pt, nla);
+		break;
 	case NLA_MSECS:
 		value = nla_get_u64(nla);
 		break;
Signature · 1 line
2.51.0
author maintainer
#353 lore raw

PATCH 339/570 netfilter: nft_payload: reject out-of-range attributes via policy

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Florian Westphal <[email protected]>

[ Upstream commit e7af210e6dd0de633d3f4850383310cf57473bc8 ]

Now that nla_policy allows range checks for bigendian data make use of this to reject such attributes. At this time, reject happens later from the init or select_ops callbacks, but its prone to errors.

In the future, new attributes can be handled via NLA_POLICY_MAX_BE and exiting ones can be converted one by one.

Stable-dep-of: 8f15b5071b45 ("netfilter: ctnetlink: use netlink policy range checks")

---
 net/netfilter/nft_payload.c | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
net/netfilter/nft_payload.c +3 -3
index 49a1cf53064fe..cb6ee00bbc09b 100644
@@ -176,10 +176,10 @@ static const struct nla_policy nft_payload_policy[NFTA_PAYLOAD_MAX + 1] = {
 	[NFTA_PAYLOAD_SREG]		= { .type = NLA_U32 },
 	[NFTA_PAYLOAD_DREG]		= { .type = NLA_U32 },
 	[NFTA_PAYLOAD_BASE]		= { .type = NLA_U32 },
-	[NFTA_PAYLOAD_OFFSET]		= { .type = NLA_U32 },
-	[NFTA_PAYLOAD_LEN]		= { .type = NLA_U32 },
+	[NFTA_PAYLOAD_OFFSET]		= NLA_POLICY_MAX_BE(NLA_U32, 255),
+	[NFTA_PAYLOAD_LEN]		= NLA_POLICY_MAX_BE(NLA_U32, 255),
 	[NFTA_PAYLOAD_CSUM_TYPE]	= { .type = NLA_U32 },
-	[NFTA_PAYLOAD_CSUM_OFFSET]	= { .type = NLA_U32 },
+	[NFTA_PAYLOAD_CSUM_OFFSET]	= NLA_POLICY_MAX_BE(NLA_U32, 255),
 	[NFTA_PAYLOAD_CSUM_FLAGS]	= { .type = NLA_U32 },
 };
 
Signature · 1 line
2.51.0
author maintainer
#354 lore raw

PATCH 340/570 netlink: hide validation union fields from kdoc

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jakub Kicinski <[email protected]>

[ Upstream commit 7354c9024f2835f6122ed9612e21ab379df050f9 ]

Mark the validation fields as private, users shouldn't set them directly and they are too complicated to explain in a more succinct way (there's already a long explanation in the comment above).

The strict_start_type field is set directly and has a dedicated comment so move that above the "private" section.

Stable-dep-of: 8f15b5071b45 ("netfilter: ctnetlink: use netlink policy range checks")

---
 include/net/netlink.h | 31 ++++++++++++++++++-------------
 1 file changed, 18 insertions(+), 13 deletions(-)
include/net/netlink.h +18 -13
index 6eb4593983319..8c67db47556e2 100644
@@ -317,19 +317,10 @@ struct nla_policy {
 	u8		validation_type;
 	u16		len;
 	union {
-		const u32 bitfield32_valid;
-		const u32 mask;
-		const char *reject_message;
-		const struct nla_policy *nested_policy;
-		struct netlink_range_validation *range;
-		struct netlink_range_validation_signed *range_signed;
-		struct {
-			s16 min, max;
-			u8 network_byte_order:1;
-		};
-		int (*validate)(const struct nlattr *attr,
-				struct netlink_ext_ack *extack);
-		/* This entry is special, and used for the attribute at index 0
+		/**
+		 * @strict_start_type: first attribute to validate strictly
+		 *
+		 * This entry is special, and used for the attribute at index 0
 		 * only, and specifies special data about the policy, namely it
 		 * specifies the "boundary type" where strict length validation
 		 * starts for any attribute types >= this value, also, strict
@@ -348,6 +339,20 @@ struct nla_policy {
 		 * was added to enforce strict validation from thereon.
 		 */
 		u16 strict_start_type;
+
+		/* private: use NLA_POLICY_*() to set */
+		const u32 bitfield32_valid;
+		const u32 mask;
+		const char *reject_message;
+		const struct nla_policy *nested_policy;
+		struct netlink_range_validation *range;
+		struct netlink_range_validation_signed *range_signed;
+		struct {
+			s16 min, max;
+			u8 network_byte_order:1;
+		};
+		int (*validate)(const struct nlattr *attr,
+				struct netlink_ext_ack *extack);
 	};
 };
 
Signature · 1 line
2.51.0
author maintainer
#355 lore raw

PATCH 341/570 netlink: introduce bigendian integer types

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Florian Westphal <[email protected]>

[ Upstream commit ecaf75ffd5f5db320d8b1da0198eef5a5ce64a3f ]

Jakub reported that the addition of the "network_byte_order" member in struct nla_policy increases size of 32bit platforms.

Instead of scraping the bit from elsewhere Johannes suggested to add explicit NLA_BE types instead, so do this here.

NLA_POLICY_MAX_BE() macro is removed again, there is no need for it: NLA_POLICY_MAX(NLA_BE.., ..) will do the right thing.

NLA_BE64 can be added later.

Stable-dep-of: 8f15b5071b45 ("netfilter: ctnetlink: use netlink policy range checks")

---
 include/net/netlink.h       | 19 +++++++++--------
 lib/nlattr.c                | 41 ++++++++++++++-----------------------
 net/netfilter/nft_payload.c |  6 +++---
 3 files changed, 28 insertions(+), 38 deletions(-)
include/net/netlink.h +10 -9
index 8c67db47556e2..356b805f50151 100644
@@ -181,6 +181,8 @@ enum {
 	NLA_S64,
 	NLA_BITFIELD32,
 	NLA_REJECT,
+	NLA_BE16,
+	NLA_BE32,
 	__NLA_TYPE_MAX,
 };
 
@@ -231,6 +233,7 @@ enum nla_policy_validation {
  *    NLA_U32, NLA_U64,
  *    NLA_S8, NLA_S16,
  *    NLA_S32, NLA_S64,
+ *    NLA_BE16, NLA_BE32,
  *    NLA_MSECS            Leaving the length field zero will verify the
  *                         given type fits, using it verifies minimum length
  *                         just like "All other"
@@ -261,6 +264,8 @@ enum nla_policy_validation {
  *    NLA_U16,
  *    NLA_U32,
  *    NLA_U64,
+ *    NLA_BE16,
+ *    NLA_BE32,
  *    NLA_S8,
  *    NLA_S16,
  *    NLA_S32,
@@ -349,7 +354,6 @@ struct nla_policy {
 		struct netlink_range_validation_signed *range_signed;
 		struct {
 			s16 min, max;
-			u8 network_byte_order:1;
 		};
 		int (*validate)(const struct nlattr *attr,
 				struct netlink_ext_ack *extack);
@@ -374,6 +378,8 @@ struct nla_policy {
 	(tp == NLA_U8 || tp == NLA_U16 || tp == NLA_U32 || tp == NLA_U64)
 #define __NLA_IS_SINT_TYPE(tp)						\
 	(tp == NLA_S8 || tp == NLA_S16 || tp == NLA_S32 || tp == NLA_S64)
+#define __NLA_IS_BEINT_TYPE(tp)						\
+	(tp == NLA_BE16 || tp == NLA_BE32)
 
 #define __NLA_ENSURE(condition) BUILD_BUG_ON_ZERO(!(condition))
 #define NLA_ENSURE_UINT_TYPE(tp)			\
@@ -387,6 +393,7 @@ struct nla_policy {
 #define NLA_ENSURE_INT_OR_BINARY_TYPE(tp)		\
 	(__NLA_ENSURE(__NLA_IS_UINT_TYPE(tp) ||		\
 		      __NLA_IS_SINT_TYPE(tp) ||		\
+		      __NLA_IS_BEINT_TYPE(tp) ||	\
 		      tp == NLA_MSECS ||		\
 		      tp == NLA_BINARY) + tp)
 #define NLA_ENSURE_NO_VALIDATION_PTR(tp)		\
@@ -394,6 +401,8 @@ struct nla_policy {
 		      tp != NLA_REJECT &&		\
 		      tp != NLA_NESTED &&		\
 		      tp != NLA_NESTED_ARRAY) + tp)
+#define NLA_ENSURE_BEINT_TYPE(tp)			\
+	(__NLA_ENSURE(__NLA_IS_BEINT_TYPE(tp)) + tp)
 
 #define NLA_POLICY_RANGE(tp, _min, _max) {		\
 	.type = NLA_ENSURE_INT_OR_BINARY_TYPE(tp),	\
@@ -424,14 +433,6 @@ struct nla_policy {
 	.type = NLA_ENSURE_INT_OR_BINARY_TYPE(tp),	\
 	.validation_type = NLA_VALIDATE_MAX,		\
 	.max = _max,					\
-	.network_byte_order = 0,			\
-}
-
-#define NLA_POLICY_MAX_BE(tp, _max) {			\
-	.type = NLA_ENSURE_UINT_TYPE(tp),		\
-	.validation_type = NLA_VALIDATE_MAX,		\
-	.max = _max,					\
-	.network_byte_order = 1,			\
 }
 
 #define NLA_POLICY_MASK(tp, _mask) {			\
lib/nlattr.c +15 -26

77 lines not on this page · raw message

net/netfilter/nft_payload.c +3 -3
index cb6ee00bbc09b..dbafb964bd706 100644
@@ -176,10 +176,10 @@ static const struct nla_policy nft_payload_policy[NFTA_PAYLOAD_MAX + 1] = {
 	[NFTA_PAYLOAD_SREG]		= { .type = NLA_U32 },
 	[NFTA_PAYLOAD_DREG]		= { .type = NLA_U32 },
 	[NFTA_PAYLOAD_BASE]		= { .type = NLA_U32 },
-	[NFTA_PAYLOAD_OFFSET]		= NLA_POLICY_MAX_BE(NLA_U32, 255),
-	[NFTA_PAYLOAD_LEN]		= NLA_POLICY_MAX_BE(NLA_U32, 255),
+	[NFTA_PAYLOAD_OFFSET]		= NLA_POLICY_MAX(NLA_BE32, 255),
+	[NFTA_PAYLOAD_LEN]		= NLA_POLICY_MAX(NLA_BE32, 255),
 	[NFTA_PAYLOAD_CSUM_TYPE]	= { .type = NLA_U32 },
-	[NFTA_PAYLOAD_CSUM_OFFSET]	= NLA_POLICY_MAX_BE(NLA_U32, 255),
+	[NFTA_PAYLOAD_CSUM_OFFSET]	= NLA_POLICY_MAX(NLA_BE32, 255),
 	[NFTA_PAYLOAD_CSUM_FLAGS]	= { .type = NLA_U32 },
 };
 
Signature · 1 line
2.51.0
author maintainer
#356 lore raw

PATCH 342/570 netlink: allow be16 and be32 types in all uint policy checks

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Florian Westphal <[email protected]>

[ Upstream commit 5fac9b7c16c50c6c7699517f582b56e3743f453a ]

__NLA_IS_BEINT_TYPE(tp) isn't useful. NLA_BE16/32 are identical to NLA_U16/32, the only difference is that it tells the netlink validation functions that byteorder conversion might be needed before comparing the value to the policy min/max ones.

After this change all policy macros that can be used with UINT types, such as NLA_POLICY_MASK() can also be used with NLA_BE16/32.

This will be used to validate nf_tables flag attributes which are in bigendian byte order.

Stable-dep-of: 8f15b5071b45 ("netfilter: ctnetlink: use netlink policy range checks")

---
 include/net/netlink.h | 10 +++-------
 lib/nlattr.c          |  6 ++++++
 2 files changed, 9 insertions(+), 7 deletions(-)
include/net/netlink.h +3 -7
index 356b805f50151..8536301842c97 100644
@@ -374,12 +374,11 @@ struct nla_policy {
 #define NLA_POLICY_BITFIELD32(valid) \
 	{ .type = NLA_BITFIELD32, .bitfield32_valid = valid }
 
-#define __NLA_IS_UINT_TYPE(tp)						\
-	(tp == NLA_U8 || tp == NLA_U16 || tp == NLA_U32 || tp == NLA_U64)
+#define __NLA_IS_UINT_TYPE(tp)					\
+	(tp == NLA_U8 || tp == NLA_U16 || tp == NLA_U32 ||	\
+	 tp == NLA_U64 || tp == NLA_BE16 || tp == NLA_BE32)
 #define __NLA_IS_SINT_TYPE(tp)						\
 	(tp == NLA_S8 || tp == NLA_S16 || tp == NLA_S32 || tp == NLA_S64)
-#define __NLA_IS_BEINT_TYPE(tp)						\
-	(tp == NLA_BE16 || tp == NLA_BE32)
 
 #define __NLA_ENSURE(condition) BUILD_BUG_ON_ZERO(!(condition))
 #define NLA_ENSURE_UINT_TYPE(tp)			\
@@ -393,7 +392,6 @@ struct nla_policy {
 #define NLA_ENSURE_INT_OR_BINARY_TYPE(tp)		\
 	(__NLA_ENSURE(__NLA_IS_UINT_TYPE(tp) ||		\
 		      __NLA_IS_SINT_TYPE(tp) ||		\
-		      __NLA_IS_BEINT_TYPE(tp) ||	\
 		      tp == NLA_MSECS ||		\
 		      tp == NLA_BINARY) + tp)
 #define NLA_ENSURE_NO_VALIDATION_PTR(tp)		\
@@ -401,8 +399,6 @@ struct nla_policy {
 		      tp != NLA_REJECT &&		\
 		      tp != NLA_NESTED &&		\
 		      tp != NLA_NESTED_ARRAY) + tp)
-#define NLA_ENSURE_BEINT_TYPE(tp)			\
-	(__NLA_ENSURE(__NLA_IS_BEINT_TYPE(tp)) + tp)
 
 #define NLA_POLICY_RANGE(tp, _min, _max) {		\
 	.type = NLA_ENSURE_INT_OR_BINARY_TYPE(tp),	\
lib/nlattr.c +6 -0
index dffd60e4065fd..a98e7191adaf6 100644
@@ -355,6 +355,12 @@ static int nla_validate_mask(const struct nla_policy *pt,
 	case NLA_U64:
 		value = nla_get_u64(nla);
 		break;
+	case NLA_BE16:
+		value = ntohs(nla_get_be16(nla));
+		break;
+	case NLA_BE32:
+		value = ntohl(nla_get_be32(nla));
+		break;
 	default:
 		return -EINVAL;
 	}
Signature · 1 line
2.51.0
author maintainer
#357 lore raw

PATCH 343/570 netfilter: ctnetlink: use netlink policy range checks

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: David Carlier <[email protected]>

[ Upstream commit 8f15b5071b4548b0aafc03b366eb45c9c6566704 ]

Replace manual range and mask validations with netlink policy annotations in ctnetlink code paths, so that the netlink core rejects invalid values early and can generate extack errors.

- CTA_PROTOINFO_TCP_STATE: reject values > TCP_CONNTRACK_SYN_SENT2 at policy level, removing the manual >= TCP_CONNTRACK_MAX check.
- CTA_PROTOINFO_TCP_WSCALE_ORIGINAL/REPLY: reject values > TCP_MAX_WSCALE (14). The normal TCP option parsing path already clamps to this value, but the ctnetlink path accepted 0-255, causing undefined behavior when used as a u32 shift count.
- CTA_FILTER_ORIG_FLAGS/REPLY_FLAGS: use NLA_POLICY_MASK with CTA_FILTER_F_ALL, removing the manual mask checks.
- CTA_EXPECT_FLAGS: use NLA_POLICY_MASK with NF_CT_EXPECT_MASK, adding a new mask define grouping all valid expect flags.

Extracted from a broader nf-next patch by Florian Westphal, scoped to ctnetlink for the fixes tree.

Fixes c8e2078cfe41 ("[NETFILTER]: ctnetlink: add support for internal tcp connection tracking flags handling")
---
 .../uapi/linux/netfilter/nf_conntrack_common.h   |  4 ++++
 net/netfilter/nf_conntrack_netlink.c             | 16 +++++-----------
 net/netfilter/nf_conntrack_proto_tcp.c           | 10 +++-------
 3 files changed, 12 insertions(+), 18 deletions(-)
include/uapi/linux/netfilter/nf_conntrack_common.h +4 -0
index 26071021e986f..56b6b60a814f5 100644
@@ -159,5 +159,9 @@ enum ip_conntrack_expect_events {
 #define NF_CT_EXPECT_INACTIVE		0x2
 #define NF_CT_EXPECT_USERSPACE		0x4
 
+#ifdef __KERNEL__
+#define NF_CT_EXPECT_MASK	(NF_CT_EXPECT_PERMANENT | NF_CT_EXPECT_INACTIVE | \
+				 NF_CT_EXPECT_USERSPACE)
+#endif
 
 #endif /* _UAPI_NF_CONNTRACK_COMMON_H */
net/netfilter/nf_conntrack_netlink.c +5 -11
index 055bff0a04da9..5087ab9b137f2 100644
@@ -872,8 +872,8 @@ struct ctnetlink_filter {
 };
 
 static const struct nla_policy cta_filter_nla_policy[CTA_FILTER_MAX + 1] = {
-	[CTA_FILTER_ORIG_FLAGS]		= { .type = NLA_U32 },
-	[CTA_FILTER_REPLY_FLAGS]	= { .type = NLA_U32 },
+	[CTA_FILTER_ORIG_FLAGS]		= NLA_POLICY_MASK(NLA_U32, CTA_FILTER_F_ALL),
+	[CTA_FILTER_REPLY_FLAGS]	= NLA_POLICY_MASK(NLA_U32, CTA_FILTER_F_ALL),
 };
 
 static int ctnetlink_parse_filter(const struct nlattr *attr,
@@ -887,17 +887,11 @@ static int ctnetlink_parse_filter(const struct nlattr *attr,
 	if (ret)
 		return ret;
 
-	if (tb[CTA_FILTER_ORIG_FLAGS]) {
+	if (tb[CTA_FILTER_ORIG_FLAGS])
 		filter->orig_flags = nla_get_u32(tb[CTA_FILTER_ORIG_FLAGS]);
-		if (filter->orig_flags & ~CTA_FILTER_F_ALL)
-			return -EOPNOTSUPP;
-	}
 
-	if (tb[CTA_FILTER_REPLY_FLAGS]) {
+	if (tb[CTA_FILTER_REPLY_FLAGS])
 		filter->reply_flags = nla_get_u32(tb[CTA_FILTER_REPLY_FLAGS]);
-		if (filter->reply_flags & ~CTA_FILTER_F_ALL)
-			return -EOPNOTSUPP;
-	}
 
 	return 0;
 }
@@ -2642,7 +2636,7 @@ static const struct nla_policy exp_nla_policy[CTA_EXPECT_MAX+1] = {
 	[CTA_EXPECT_HELP_NAME]	= { .type = NLA_NUL_STRING,
 				    .len = NF_CT_HELPER_NAME_LEN - 1 },
 	[CTA_EXPECT_ZONE]	= { .type = NLA_U16 },
-	[CTA_EXPECT_FLAGS]	= { .type = NLA_U32 },
+	[CTA_EXPECT_FLAGS]	= NLA_POLICY_MASK(NLA_BE32, NF_CT_EXPECT_MASK),
 	[CTA_EXPECT_CLASS]	= { .type = NLA_U32 },
 	[CTA_EXPECT_NAT]	= { .type = NLA_NESTED },
 	[CTA_EXPECT_FN]		= { .type = NLA_NUL_STRING },
net/netfilter/nf_conntrack_proto_tcp.c +3 -7
index f33e6aea7f4da..10bd7f604ebbc 100644
@@ -1325,9 +1325,9 @@ static int tcp_to_nlattr(struct sk_buff *skb, struct nlattr *nla,
 }
 
 static const struct nla_policy tcp_nla_policy[CTA_PROTOINFO_TCP_MAX+1] = {
-	[CTA_PROTOINFO_TCP_STATE]	    = { .type = NLA_U8 },
-	[CTA_PROTOINFO_TCP_WSCALE_ORIGINAL] = { .type = NLA_U8 },
-	[CTA_PROTOINFO_TCP_WSCALE_REPLY]    = { .type = NLA_U8 },
+	[CTA_PROTOINFO_TCP_STATE]	    = NLA_POLICY_MAX(NLA_U8, TCP_CONNTRACK_SYN_SENT2),
+	[CTA_PROTOINFO_TCP_WSCALE_ORIGINAL] = NLA_POLICY_MAX(NLA_U8, TCP_MAX_WSCALE),
+	[CTA_PROTOINFO_TCP_WSCALE_REPLY]    = NLA_POLICY_MAX(NLA_U8, TCP_MAX_WSCALE),
 	[CTA_PROTOINFO_TCP_FLAGS_ORIGINAL]  = { .len = sizeof(struct nf_ct_tcp_flags) },
 	[CTA_PROTOINFO_TCP_FLAGS_REPLY]	    = { .len = sizeof(struct nf_ct_tcp_flags) },
 };
@@ -1354,10 +1354,6 @@ static int nlattr_to_tcp(struct nlattr *cda[], struct nf_conn *ct)
 	if (err < 0)
 		return err;
 
-	if (tb[CTA_PROTOINFO_TCP_STATE] &&
-	    nla_get_u8(tb[CTA_PROTOINFO_TCP_STATE]) >= TCP_CONNTRACK_MAX)
-		return -EINVAL;
-
 	spin_lock_bh(&ct->lock);
 	if (tb[CTA_PROTOINFO_TCP_STATE])
 		ct->proto.tcp.state = nla_get_u8(tb[CTA_PROTOINFO_TCP_STATE]);
Signature · 1 line
2.51.0
author maintainer
#358 lore raw

PATCH 344/570 net: macb: use the current queue number for stats

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Paolo Valerio <[email protected]>

[ Upstream commit 72d96e4e24bbefdcfbc68bdb9341a05d8f5cb6e5 ]

There's a potential mismatch between the memory reserved for statistics and the amount of memory written.

gem_get_sset_count() correctly computes the number of stats based on the active queues, whereas gem_get_ethtool_stats() indiscriminately copies data using the maximum number of queues, and in the case the number of active queues is less than MACB_MAX_QUEUES, this results in a OOB write as observed in the KASAN splat.

==================================================================

BUG: KASAN: vmalloc-out-of-bounds in gem_get_ethtool_stats+0x54/0x78
  [macb]
Write of size 760 at addr ffff80008080b000 by task ethtool/1027

CPU: [...]

Tainted: [E]=UNSIGNED_MODULE
Hardware name: raspberrypi rpi/rpi, BIOS 2025.10 10/01/2025
Call trace:
 show_stack+0x20/0x38 (C)
 dump_stack_lvl+0x80/0xf8
 print_report+0x384/0x5e0
 kasan_report+0xa0/0xf0
 kasan_check_range+0xe8/0x190
 __asan_memcpy+0x54/0x98
 gem_get_ethtool_stats+0x54/0x78 [macb
   926c13f3af83b0c6fe64badb21ec87d5e93fcf65]
 dev_ethtool+0x1220/0x38c0
 dev_ioctl+0x4ac/0xca8
 sock_do_ioctl+0x170/0x1d8
 sock_ioctl+0x484/0x5d8
 __arm64_sys_ioctl+0x12c/0x1b8
 invoke_syscall+0xd4/0x258
 el0_svc_common.constprop.0+0xb4/0x240
 do_el0_svc+0x48/0x68
 el0_svc+0x40/0xf8
 el0t_64_sync_handler+0xa0/0xe8
 el0t_64_sync+0x1b0/0x1b8

The buggy address belongs to a 1-page vmalloc region starting at
  0xffff80008080b000 allocated at dev_ethtool+0x11f0/0x38c0
The buggy address belongs to the physical page:
page: refcount:1 mapcount:0 mapping:0000000000000000
  index:0xffff00000a333000 pfn:0xa333
flags: 0x7fffc000000000(node=0|zone=0|lastcpupid=0x1ffff)
raw: 007fffc000000000 0000000000000000 dead000000000122 0000000000000000
raw: ffff00000a333000 0000000000000000 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected

Memory state around the buggy address:
 ffff80008080b080: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
 ffff80008080b100: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
>ffff80008080b180: 00 00 00 00 00 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8
                                  ^
 ffff80008080b200: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8
 ffff80008080b280: f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8 f8
==================================================================

Fix it by making sure the copied size only considers the active number of queues.

---
 drivers/net/ethernet/cadence/macb_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/cadence/macb_main.c +1 -1
index 6a3e9082bda8c..a0beb6eb505da 100644
@@ -2977,7 +2977,7 @@ static void gem_get_ethtool_stats(struct net_device *dev,
 	spin_lock_irq(&bp->stats_lock);
 	gem_update_stats(bp);
 	memcpy(data, &bp->ethtool_stats, sizeof(u64)
-			* (GEM_STATS_LEN + QUEUE_STATS_LEN * MACB_MAX_QUEUES));
+			* (GEM_STATS_LEN + QUEUE_STATS_LEN * bp->num_queues));
 	spin_unlock_irq(&bp->stats_lock);
 }
 
Signature · 1 line
2.51.0
author maintainer
#359 lore raw

PATCH 345/570 regmap: Synchronize cache for the page selector

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Andy Shevchenko <[email protected]>

[ Upstream commit 09e70e4f119ff650d24c96161fd2f62ac7e424b0 ]

If the selector register is represented in each page, its value according to the debugfs is stale because it gets synchronized only after the real page switch happens. Hence the regmap cache initialisation from the HW inherits outdated data in the selector register.

Synchronize cache for the page selector just in time.

Before (offset followed by hexdump, the first byte is selector):

    // Real registers
    18: 05 ff 00 00 ff 0f 00 00 f0 00 00 00
    ...
    // Virtual (per port)
    40: 05 ff 00 00 e0 e0 00 00 00 00 00 1f
    50: 00 ff 00 00 e0 e0 00 00 00 00 00 1f
    60: 01 ff 00 00 ff ff 00 00 00 00 00 00
    70: 02 ff 00 00 cf f3 00 00 00 00 00 0c
    80: 03 ff 00 00 00 00 00 00 00 00 00 ff
    90: 04 ff 00 00 ff 0f 00 00 f0 00 00 00

After:

    // Real registers
    18: 05 ff 00 00 ff 0f 00 00 f0 00 00 00
    ...
    // Virtual (per port)
    40: 00 ff 00 00 e0 e0 00 00 00 00 00 1f
    50: 01 ff 00 00 e0 e0 00 00 00 00 00 1f
    60: 02 ff 00 00 ff ff 00 00 00 00 00 00
    70: 03 ff 00 00 cf f3 00 00 00 00 00 0c
    80: 04 ff 00 00 00 00 00 00 00 00 00 ff
    90: 05 ff 00 00 ff 0f 00 00 f0 00 00 00
Fixes 6863ca622759 ("regmap: Add support for register indirect addressing.")
---
 drivers/base/regmap/regmap.c | 30 ++++++++++++++++++++++++++----
 1 file changed, 26 insertions(+), 4 deletions(-)
drivers/base/regmap/regmap.c +26 -4
index e1380b08685f4..b1cae7db6318a 100644
@@ -1629,6 +1629,7 @@ static int _regmap_select_page(struct regmap *map, unsigned int *reg,
 			       unsigned int val_num)
 {
 	void *orig_work_buf;
+	unsigned int selector_reg;
 	unsigned int win_offset;
 	unsigned int win_page;
 	bool page_chg;
@@ -1647,10 +1648,31 @@ static int _regmap_select_page(struct regmap *map, unsigned int *reg,
 			return -EINVAL;
 	}
 
-	/* It is possible to have selector register inside data window.
-	   In that case, selector register is located on every page and
-	   it needs no page switching, when accessed alone. */
+	/*
+	 * Calculate the address of the selector register in the corresponding
+	 * data window if it is located on every page.
+	 */
+	page_chg = in_range(range->selector_reg, range->window_start, range->window_len);
+	if (page_chg)
+		selector_reg = range->range_min + win_page * range->window_len +
+			       range->selector_reg - range->window_start;
+
+	/*
+	 * It is possible to have selector register inside data window.
+	 * In that case, selector register is located on every page and it
+	 * needs no page switching, when accessed alone.
+	 *
+	 * Nevertheless we should synchronize the cache values for it.
+	 * This can't be properly achieved if the selector register is
+	 * the first and the only one to be read inside the data window.
+	 * That's why we update it in that case as well.
+	 *
+	 * However, we specifically avoid updating it for the default page,
+	 * when it's overlapped with the real data window, to prevent from
+	 * infinite looping.
+	 */
 	if (val_num > 1 ||
+	    (page_chg && selector_reg != range->selector_reg) ||
 	    range->window_start + win_offset != range->selector_reg) {
 		/* Use separate work_buf during page switching */
 		orig_work_buf = map->work_buf;
@@ -1659,7 +1681,7 @@ static int _regmap_select_page(struct regmap *map, unsigned int *reg,
 		ret = _regmap_update_bits(map, range->selector_reg,
 					  range->selector_mask,
 					  win_page << range->selector_shift,
-					  &page_chg, false);
+					  NULL, false);
 
 		map->work_buf = orig_work_buf;
 
Signature · 1 line
2.53.0
author maintainer
#360 lore raw

PATCH 346/570 RDMA/rw: Fall back to direct SGE on MR pool exhaustion

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Chuck Lever <[email protected]>

[ Upstream commit 00da250c21b074ea9494c375d0117b69e5b1d0a4 ]

When IOMMU passthrough mode is active, ib_dma_map_sgtable_attrs() produces no coalescing: each scatterlist page maps 1:1 to a DMA entry, so sgt.nents equals the raw page count. A 1 MB transfer yields 256 DMA entries. If that count exceeds the device's max_sgl_rd threshold (an optimization hint from mlx5 firmware), rdma_rw_io_needs_mr() steers the operation into the MR registration path. Each such operation consumes one or more MRs from a pool sized at max_rdma_ctxs -- roughly one MR per concurrent context. Under write-intensive workloads that issue many concurrent RDMA READs, the pool is rapidly exhausted, ib_mr_pool_get() returns NULL, and rdma_rw_init_one_mr() returns -EAGAIN. Upper layer protocols treat this as a fatal DMA mapping failure and tear down the connection.

The max_sgl_rd check is a performance optimization, not a correctness requirement: the device can handle large SGE counts via direct posting, just less efficiently than with MR registration. When the MR pool cannot satisfy a request, falling back to the direct SGE (map_wrs) path avoids the connection reset while preserving the MR optimization for the common case where pool resources are available.

Add a fallback in rdma_rw_ctx_init() so that -EAGAIN from rdma_rw_init_mr_wrs() triggers direct SGE posting instead of propagating the error. iWARP devices, which mandate MR registration for RDMA READs, and force_mr debug mode continue to treat -EAGAIN as terminal.

Fixes 00bd1439f464 ("RDMA/rw: Support threshold for registration vs scattering to local pages")
---
 drivers/infiniband/core/rw.c | 21 ++++++++++++++++++---
 1 file changed, 18 insertions(+), 3 deletions(-)
drivers/infiniband/core/rw.c +18 -3
index 3b6cfa6362e04..bfa91d2ff956c 100644
@@ -341,14 +341,29 @@ int rdma_rw_ctx_init(struct rdma_rw_ctx *ctx, struct ib_qp *qp, u32 port_num,
 	if (rdma_rw_io_needs_mr(qp->device, port_num, dir, sg_cnt)) {
 		ret = rdma_rw_init_mr_wrs(ctx, qp, port_num, sg, sg_cnt,
 				sg_offset, remote_addr, rkey, dir);
-	} else if (sg_cnt > 1) {
+		/*
+		 * If MR init succeeded or failed for a reason other
+		 * than pool exhaustion, that result is final.
+		 *
+		 * Pool exhaustion (-EAGAIN) from the max_sgl_rd
+		 * optimization is recoverable: fall back to
+		 * direct SGE posting. iWARP and force_mr require
+		 * MRs unconditionally, so -EAGAIN is terminal.
+		 */
+		if (ret != -EAGAIN ||
+		    rdma_protocol_iwarp(qp->device, port_num) ||
+		    unlikely(rdma_rw_force_mr))
+			goto out;
+	}
+
+	if (sg_cnt > 1)
 		ret = rdma_rw_init_map_wrs(ctx, qp, sg, sg_cnt, sg_offset,
 				remote_addr, rkey, dir);
-	} else {
+	else
 		ret = rdma_rw_init_single_wr(ctx, qp, sg, sg_offset,
 				remote_addr, rkey, dir);
-	}
 
+out:
 	if (ret < 0)
 		goto out_unmap_sg;
 	return ret;
Signature · 1 line
2.53.0
author maintainer
#361 lore raw

PATCH 347/570 RDMA/irdma: Update ibqp state to error if QP is already in error state

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Tatyana Nikolova <[email protected]>

[ Upstream commit 8c1f19a2225cf37b3f8ab0b5a8a5322291cda620 ]

In irdma_modify_qp() update ibqp state to error if the irdma QP is already in error state, otherwise the ibqp state which is visible to the consumer app remains stale.

Fixes b48c24c2d710 ("RDMA/irdma: Implement device supported verb APIs")
---
 drivers/infiniband/hw/irdma/verbs.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/infiniband/hw/irdma/verbs.c +2 -0
index 40960f0803fbc..f1924e84d2568 100644
@@ -1353,6 +1353,7 @@ int irdma_modify_qp_roce(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 		case IB_QPS_ERR:
 		case IB_QPS_RESET:
 			if (iwqp->iwarp_state == IRDMA_QP_STATE_ERROR) {
+				iwqp->ibqp_state = attr->qp_state;
 				spin_unlock_irqrestore(&iwqp->lock, flags);
 				if (udata) {
 					if (ib_copy_from_udata(&ureq, udata,
@@ -1549,6 +1550,7 @@ int irdma_modify_qp(struct ib_qp *ibqp, struct ib_qp_attr *attr, int attr_mask,
 		case IB_QPS_ERR:
 		case IB_QPS_RESET:
 			if (iwqp->iwarp_state == IRDMA_QP_STATE_ERROR) {
+				iwqp->ibqp_state = attr->qp_state;
 				spin_unlock_irqrestore(&iwqp->lock, flags);
 				if (udata) {
 					if (ib_copy_from_udata(&ureq, udata,
Signature · 1 line
2.53.0
author maintainer
#362 lore raw

PATCH 348/570 RDMA/irdma: Remove a NOP wait_event() in irdma_modify_qp_roce()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Tatyana Nikolova <[email protected]>

[ Upstream commit 5e8f0239731a83753473b7aa91bda67bbdff5053 ]

Remove a NOP wait_event() in irdma_modify_qp_roce() which is relevant for iWARP and likely a copy and paste artifact for RoCEv2. The wait event is for sending a reset on a TCP connection, after the reset has been requested in irdma_modify_qp(), which occurs only in iWarp mode.

Fixes b48c24c2d710 ("RDMA/irdma: Implement device supported verb APIs")
---
 drivers/infiniband/hw/irdma/verbs.c | 2 --
 1 file changed, 2 deletions(-)
drivers/infiniband/hw/irdma/verbs.c +0 -2
index f1924e84d2568..b7186bb6a7bbc 100644
@@ -1275,8 +1275,6 @@ int irdma_modify_qp_roce(struct ib_qp *ibqp, struct ib_qp_attr *attr,
 			roce_info->rd_en = true;
 	}
 
-	wait_event(iwqp->mod_qp_waitq, !atomic_read(&iwqp->hw_mod_qp_pend));
-
 	ibdev_dbg(&iwdev->ibdev,
 		  "VERBS: caller: %pS qp_id=%d to_ibqpstate=%d ibqpstate=%d irdma_qpstate=%d attr_mask=0x%x\n",
 		  __builtin_return_address(0), ibqp->qp_num, attr->qp_state,
Signature · 1 line
2.53.0
author maintainer
#363 lore raw

PATCH 349/570 RDMA/irdma: Clean up unnecessary dereference of event->cm_node

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ivan Barrera <[email protected]>

[ Upstream commit b415399c9a024d574b65479636f0d4eb625b9abd ]

The cm_node is available and the usage of cm_node and event->cm_node seems arbitrary. Clean up unnecessary dereference of event->cm_node.

Fixes 146b9756f14c ("RDMA/irdma: Add connection manager")
---
 drivers/infiniband/hw/irdma/cm.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
drivers/infiniband/hw/irdma/cm.c +6 -6
index d2c6a1bcf1de9..8a99a040917f7 100644
@@ -4171,21 +4171,21 @@ static void irdma_cm_event_handler(struct work_struct *work)
 		irdma_cm_event_reset(event);
 		break;
 	case IRDMA_CM_EVENT_CONNECTED:
-		if (!event->cm_node->cm_id ||
-		    event->cm_node->state != IRDMA_CM_STATE_OFFLOADED)
+		if (!cm_node->cm_id ||
+		    cm_node->state != IRDMA_CM_STATE_OFFLOADED)
 			break;
 		irdma_cm_event_connected(event);
 		break;
 	case IRDMA_CM_EVENT_MPA_REJECT:
-		if (!event->cm_node->cm_id ||
+		if (!cm_node->cm_id ||
 		    cm_node->state == IRDMA_CM_STATE_OFFLOADED)
 			break;
 		irdma_send_cm_event(cm_node, cm_node->cm_id,
 				    IW_CM_EVENT_CONNECT_REPLY, -ECONNREFUSED);
 		break;
 	case IRDMA_CM_EVENT_ABORTED:
-		if (!event->cm_node->cm_id ||
-		    event->cm_node->state == IRDMA_CM_STATE_OFFLOADED)
+		if (!cm_node->cm_id ||
+		    cm_node->state == IRDMA_CM_STATE_OFFLOADED)
 			break;
 		irdma_event_connect_error(event);
 		break;
@@ -4195,7 +4195,7 @@ static void irdma_cm_event_handler(struct work_struct *work)
 		break;
 	}
 
-	irdma_rem_ref_cm_node(event->cm_node);
+	irdma_rem_ref_cm_node(cm_node);
 	kfree(event);
 }
 
Signature · 1 line
2.53.0
author maintainer
#364 lore raw

PATCH 350/570 RDMA/irdma: Remove reset check from irdma_modify_qp_to_err()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Tatyana Nikolova <[email protected]>

[ Upstream commit c45c6ebd693b944f1ffe429fdfb6cc1674c237be ]

During reset, irdma_modify_qp() to error should be called to disconnect the QP. Without this fix, if not preceded by irdma_modify_qp() to error, the API call irdma_destroy_qp() gets stuck waiting for the QP refcount to go to zero, because the cm_node associated with this QP isn't disconnected.

Fixes 915cc7ac0f8e ("RDMA/irdma: Add miscellaneous utility definitions")
---
 drivers/infiniband/hw/irdma/utils.c | 2 --
 1 file changed, 2 deletions(-)
drivers/infiniband/hw/irdma/utils.c +0 -2
index 235515e8bf9b7..5a0672345cab1 100644
@@ -2512,8 +2512,6 @@ void irdma_modify_qp_to_err(struct irdma_sc_qp *sc_qp)
 	struct irdma_qp *qp = sc_qp->qp_uk.back_qp;
 	struct ib_qp_attr attr;
 
-	if (qp->iwdev->rf->reset)
-		return;
 	attr.qp_state = IB_QPS_ERR;
 
 	if (rdma_protocol_roce(qp->ibqp.device, 1))
Signature · 1 line
2.53.0
author maintainer
#365 lore raw

PATCH 351/570 RDMA/irdma: Fix deadlock during netdev reset with active connections

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Anil Samal <[email protected]>

[ Upstream commit 6f52370970ac07d352a7af4089e55e0e6425f827 ]

Resolve deadlock that occurs when user executes netdev reset while RDMA applications (e.g., rping) are active. The netdev reset causes ice driver to remove irdma auxiliary driver, triggering device_delete and subsequent client removal. During client removal, uverbs_client waits for QP reference count to reach zero while cma_client holds the final reference, creating circular dependency and indefinite wait in iWARP mode. Skip QP reference count wait during device reset to prevent deadlock.

Fixes c8f304d75f6c ("RDMA/irdma: Prevent QP use after free")
---
 drivers/infiniband/hw/irdma/verbs.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/infiniband/hw/irdma/verbs.c +2 -1
index b7186bb6a7bbc..986acd446c651 100644
@@ -518,7 +518,8 @@ static int irdma_destroy_qp(struct ib_qp *ibqp, struct ib_udata *udata)
 	}
 
 	irdma_qp_rem_ref(&iwqp->ibqp);
-	wait_for_completion(&iwqp->free_qp);
+	if (!iwdev->rf->reset)
+		wait_for_completion(&iwqp->free_qp);
 	irdma_free_lsmm_rsrc(iwqp);
 	irdma_cqp_qp_destroy_cmd(&iwdev->rf->sc_dev, &iwqp->sc_qp);
 
Signature · 1 line
2.53.0
author maintainer
#366 lore raw

PATCH 352/570 RDMA/irdma: Return EINVAL for invalid arp index error

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Tatyana Nikolova <[email protected]>

[ Upstream commit 7221f581eefa79ead06e171044f393fb7ee22f87 ]

When rdma_connect() fails due to an invalid arp index, user space rdma core reports ENOMEM which is confusing. Modify irdma_make_cm_node() to return the correct error code.

Fixes 146b9756f14c ("RDMA/irdma: Add connection manager")
---
 drivers/infiniband/hw/irdma/cm.c | 17 ++++++++++-------
 1 file changed, 10 insertions(+), 7 deletions(-)
drivers/infiniband/hw/irdma/cm.c +10 -7
index 8a99a040917f7..8d671bd64f37a 100644
@@ -2197,11 +2197,12 @@ irdma_make_cm_node(struct irdma_cm_core *cm_core, struct irdma_device *iwdev,
 	int oldarpindex;
 	int arpindex;
 	struct net_device *netdev = iwdev->netdev;
+	int ret;
 
 	/* create an hte and cm_node for this instance */
 	cm_node = kzalloc(sizeof(*cm_node), GFP_ATOMIC);
 	if (!cm_node)
-		return NULL;
+		return ERR_PTR(-ENOMEM);
 
 	/* set our node specific transport info */
 	cm_node->ipv4 = cm_info->ipv4;
@@ -2296,8 +2297,10 @@ irdma_make_cm_node(struct irdma_cm_core *cm_core, struct irdma_device *iwdev,
 			arpindex = -EINVAL;
 	}
 
-	if (arpindex < 0)
+	if (arpindex < 0) {
+		ret = -EINVAL;
 		goto err;
+	}
 
 	ether_addr_copy(cm_node->rem_mac,
 			iwdev->rf->arp_table[arpindex].mac_addr);
@@ -2308,7 +2311,7 @@ irdma_make_cm_node(struct irdma_cm_core *cm_core, struct irdma_device *iwdev,
 err:
 	kfree(cm_node);
 
-	return NULL;
+	return ERR_PTR(ret);
 }
 
 static void irdma_destroy_connection(struct irdma_cm_node *cm_node)
@@ -2969,8 +2972,8 @@ static int irdma_create_cm_node(struct irdma_cm_core *cm_core,
 
 	/* create a CM connection node */
 	cm_node = irdma_make_cm_node(cm_core, iwdev, cm_info, NULL);
-	if (!cm_node)
-		return -ENOMEM;
+	if (IS_ERR(cm_node))
+		return PTR_ERR(cm_node);
 
 	/* set our node side to client (active) side */
 	cm_node->tcp_cntxt.client = 1;
@@ -3167,9 +3170,9 @@ void irdma_receive_ilq(struct irdma_sc_vsi *vsi, struct irdma_puda_buf *rbuf)
 		cm_info.cm_id = listener->cm_id;
 		cm_node = irdma_make_cm_node(cm_core, iwdev, &cm_info,
 					     listener);
-		if (!cm_node) {
+		if (IS_ERR(cm_node)) {
 			ibdev_dbg(&cm_core->iwdev->ibdev,
-				  "CM: allocate node failed\n");
+				  "CM: allocate node failed ret=%ld\n", PTR_ERR(cm_node));
 			refcount_dec(&listener->refcnt);
 			return;
 		}
Signature · 1 line
2.53.0
author maintainer
#367 lore raw

PATCH 353/570 scsi: scsi_transport_sas: Fix the maximum channel scanning issue

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yihang Li <[email protected]>

[ Upstream commit d71afa9deb4d413232ba16d693f7d43b321931b4 ]

After commit 37c4e72b0651 ("scsi: Fix sas_user_scan() to handle wildcard and multi-channel scans"), if the device supports multiple channels (0 to shost->max_channel), user_scan() invokes updated sas_user_scan() to perform the scan behavior for a specific transfer. However, when the user specifies shost->max_channel, it will return -EINVAL, which is not expected.

Fix and support specifying the scan shost->max_channel for scanning.

Fixes 37c4e72b0651 ("scsi: Fix sas_user_scan() to handle wildcard and multi-channel scans")
---
 drivers/scsi/scsi_transport_sas.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/scsi_transport_sas.c +1 -1
index 87c5ed56e47bd..ecb2e8aed93bb 100644
@@ -1732,7 +1732,7 @@ static int sas_user_scan(struct Scsi_Host *shost, uint channel,
 		break;
 
 	default:
-		if (channel < shost->max_channel) {
+		if (channel <= shost->max_channel) {
 			res = scsi_scan_host_selected(shost, channel, id, lun,
 						      SCSI_SCAN_MANUAL);
 		} else {
Signature · 1 line
2.53.0
author maintainer
#368 lore raw

PATCH 354/570 x86/efi: efi_unmap_boot_services: fix calculation of ranges_to_free size

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mike Rapoport (Microsoft) <[email protected]>

[ Upstream commit 217c0a5c177a3d4f7c8497950cbf5c36756e8bbb ]

ranges_to_free array should have enough room to store the entire EFI memmap plus an extra element for NULL entry.
The calculation of this array size wrongly adds 1 to the overall size instead of adding 1 to the number of elements.

Add parentheses to properly size the array.

Fixes a4b0bf6a40f3 ("x86/efi: defer freeing of boot services memory")
---
 arch/x86/platform/efi/quirks.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/platform/efi/quirks.c +1 -1
index e3b00f05a2532..b0d0376940ba8 100644
@@ -416,7 +416,7 @@ void __init efi_unmap_boot_services(void)
 	if (efi_enabled(EFI_DBG))
 		return;
 
-	sz = sizeof(*ranges_to_free) * efi.memmap.nr_map + 1;
+	sz = sizeof(*ranges_to_free) * (efi.memmap.nr_map + 1);
 	ranges_to_free = kzalloc(sz, GFP_KERNEL);
 	if (!ranges_to_free) {
 		pr_err("Failed to allocate storage for freeable EFI regions\n");
Signature · 1 line
2.53.0
author maintainer
#369 lore raw

PATCH 355/570 drm/i915/gmbus: fix spurious timeout on 512-byte burst reads

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Samasth Norway Ananda <[email protected]>

[ Upstream commit 08441f10f4dc09fdeb64529953ac308abc79dd38 ]

When reading exactly 512 bytes with burst read enabled, the extra_byte_added path breaks out of the inner do-while without decrementing len. The outer while(len) then re-enters and gmbus_wait() times out since all data has been delivered. Decrement len before the break so the outer loop terminates correctly.

Fixes d5dc0f43f268 ("drm/i915/gmbus: Enable burst read")
(cherry picked from commit 4ab0f09ee73fc853d00466682635f67c531f909c)
---
 drivers/gpu/drm/i915/display/intel_gmbus.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/gpu/drm/i915/display/intel_gmbus.c +3 -1
index ceb1bf8a8c3c2..01b046578cd15 100644
@@ -432,8 +432,10 @@ gmbus_xfer_read_chunk(struct drm_i915_private *dev_priv,
 
 		val = intel_de_read_fw(dev_priv, GMBUS3);
 		do {
-			if (extra_byte_added && len == 1)
+			if (extra_byte_added && len == 1) {
+				len--;
 				break;
+			}
 
 			*buf++ = val & 0xff;
 			val >>= 8;
Signature · 1 line
2.53.0
author maintainer
#370 lore raw

PATCH 356/570 ASoC: Intel: catpt: Fix the device initialization

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Cezary Rojewski <[email protected]>

[ Upstream commit 5a184f1cb43a8e035251c635f5c47da5dc3e3049 ]

The DMA mask shall be coerced before any buffer allocations for the device are done. At the same time explain why DMA mask of 31 bits is used in the first place.

Cc Andy Shevchenko <[email protected]>
Fixes 7a10b66a5df9 ("ASoC: Intel: catpt: Device driver lifecycle")
---
 sound/soc/intel/catpt/device.c | 10 +++++++++-
 sound/soc/intel/catpt/dsp.c    |  3 ---
 2 files changed, 9 insertions(+), 4 deletions(-)
sound/soc/intel/catpt/device.c +9 -1
index 85a34e37316d0..68b01bdef2bfe 100644
@@ -275,7 +275,15 @@ static int catpt_acpi_probe(struct platform_device *pdev)
 	if (IS_ERR(cdev->pci_ba))
 		return PTR_ERR(cdev->pci_ba);
 
-	/* alloc buffer for storing DRAM context during dx transitions */
+	/*
+	 * As per design HOST is responsible for preserving firmware's runtime
+	 * context during D0 -> D3 -> D0 transitions.  Addresses used for DMA
+	 * to/from HOST memory shall be outside the reserved range of 0xFFFxxxxx.
+	 */
+	ret = dma_coerce_mask_and_coherent(cdev->dev, DMA_BIT_MASK(31));
+	if (ret)
+		return ret;
+
 	cdev->dxbuf_vaddr = dmam_alloc_coherent(dev, catpt_dram_size(cdev),
 						&cdev->dxbuf_paddr, GFP_KERNEL);
 	if (!cdev->dxbuf_vaddr)
sound/soc/intel/catpt/dsp.c +0 -3
index 346bec0003066..3cde6b7ae9237 100644
@@ -125,9 +125,6 @@ int catpt_dmac_probe(struct catpt_dev *cdev)
 	dmac->dev = cdev->dev;
 	dmac->irq = cdev->irq;
 
-	ret = dma_coerce_mask_and_coherent(cdev->dev, DMA_BIT_MASK(31));
-	if (ret)
-		return ret;
 	/*
 	 * Caller is responsible for putting device in D0 to allow
 	 * for I/O and memory access before probing DW.
Signature · 1 line
2.53.0
author maintainer
#371 lore raw

PATCH 357/570 ACPICA: include/acpi/acpixf.h: Fix indentation

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hans de Goede <[email protected]>

[ Upstream commit 7a9d74e7e403cb2e60d4d00c05f2f3ab2a33d0c3 ]

A bunch of the functions declared in include/acpi/acpixf.h have their name aligned a space after the '(' of e.g. the ACPI_EXTERNAL_RETURN_STATUS(acpi_status line above rather then being directly aligned after the '('.

This breaks applying patches generated from the ACPICA upstream git, remove the extra space before the function-names and all the arguments to fix this.

Stable-dep-of: f6484cadbcaf ("ACPI: EC: clean up handlers on probe failure in acpi_ec_setup()")

---
 include/acpi/acpixf.h | 120 +++++++++++++++++++++---------------------
 1 file changed, 60 insertions(+), 60 deletions(-)
include/acpi/acpixf.h +60 -60
index 9d45a6001bc00..2cf4c90730d6b 100644
@@ -587,82 +587,82 @@ ACPI_EXTERNAL_RETURN_STATUS(acpi_status
 			    acpi_install_initialization_handler
 			    (acpi_init_handler handler, u32 function))
 ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
-				 acpi_install_sci_handler(acpi_sci_handler
-							  address,
-							  void *context))
-ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
-				 acpi_remove_sci_handler(acpi_sci_handler
-							 address))
+				acpi_install_sci_handler(acpi_sci_handler
+							 address,
+							 void *context))
 ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
-				 acpi_install_global_event_handler
-				 (acpi_gbl_event_handler handler,
-				  void *context))
+				acpi_remove_sci_handler(acpi_sci_handler
+							address))
 ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
-				 acpi_install_fixed_event_handler(u32
-								  acpi_event,
-								  acpi_event_handler
-								  handler,
-								  void
-								  *context))
+				acpi_install_global_event_handler
+				(acpi_gbl_event_handler handler,
+				 void *context))
 ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
-				 acpi_remove_fixed_event_handler(u32 acpi_event,
+				acpi_install_fixed_event_handler(u32
+								 acpi_event,
 								 acpi_event_handler
-								 handler))
-ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
-				 acpi_install_gpe_handler(acpi_handle
-							  gpe_device,
-							  u32 gpe_number,
-							  u32 type,
-							  acpi_gpe_handler
-							  address,
-							  void *context))
+								 handler,
+								 void
+								 *context))
 ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
-				 acpi_install_gpe_raw_handler(acpi_handle
-							      gpe_device,
-							      u32 gpe_number,
-							      u32 type,
-							      acpi_gpe_handler
-							      address,
-							      void *context))
+				acpi_remove_fixed_event_handler(u32 acpi_event,
+								acpi_event_handler
+								handler))
 ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
-				 acpi_remove_gpe_handler(acpi_handle gpe_device,
+				acpi_install_gpe_handler(acpi_handle
+							 gpe_device,
 							 u32 gpe_number,
+							 u32 type,
 							 acpi_gpe_handler
-							 address))
-ACPI_EXTERNAL_RETURN_STATUS(acpi_status
-			     acpi_install_notify_handler(acpi_handle device,
-							 u32 handler_type,
-							 acpi_notify_handler
-							 handler,
+							 address,
 							 void *context))
+ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
+				acpi_install_gpe_raw_handler(acpi_handle
+							     gpe_device,
+							     u32 gpe_number,
+							     u32 type,
+							     acpi_gpe_handler
+							     address,
+							     void *context))
+ACPI_HW_DEPENDENT_RETURN_STATUS(acpi_status
+				acpi_remove_gpe_handler(acpi_handle gpe_device,
+							u32 gpe_number,
+							acpi_gpe_handler
+							address))
 ACPI_EXTERNAL_RETURN_STATUS(acpi_status
-			     acpi_remove_notify_handler(acpi_handle device,
+			    acpi_install_notify_handler(acpi_handle device,
 							u32 handler_type,
 							acpi_notify_handler
-							handler))
-ACPI_EXTERNAL_RETURN_STATUS(acpi_status
-			     acpi_install_address_space_handler(acpi_handle
-								device,
-								acpi_adr_space_type
-								space_id,
-								acpi_adr_space_handler
-								handler,
-								acpi_adr_space_setup
-								setup,
-								void *context))
-ACPI_EXTERNAL_RETURN_STATUS(acpi_status
-			     acpi_remove_address_space_handler(acpi_handle
+							handler,
+							void *context))
+ACPI_EXTERNAL_RETURN_STATUS(acpi_status
+			    acpi_remove_notify_handler(acpi_handle device,
+						       u32 handler_type,
+						       acpi_notify_handler
+						       handler))
+ACPI_EXTERNAL_RETURN_STATUS(acpi_status
+			    acpi_install_address_space_handler(acpi_handle
 							       device,
 							       acpi_adr_space_type
 							       space_id,
 							       acpi_adr_space_handler
-							       handler))
-ACPI_EXTERNAL_RETURN_STATUS(acpi_status
-			     acpi_install_exception_handler
-			     (acpi_exception_handler handler))
-ACPI_EXTERNAL_RETURN_STATUS(acpi_status
-			     acpi_install_interface_handler
-			     (acpi_interface_handler handler))
+							       handler,
+							       acpi_adr_space_setup
+							       setup,
+							       void *context))
+ACPI_EXTERNAL_RETURN_STATUS(acpi_status
+			    acpi_remove_address_space_handler(acpi_handle
+							      device,
+							      acpi_adr_space_type
+							      space_id,
+							      acpi_adr_space_handler
+							      handler))
+ACPI_EXTERNAL_RETURN_STATUS(acpi_status
+			    acpi_install_exception_handler
+			    (acpi_exception_handler handler))
+ACPI_EXTERNAL_RETURN_STATUS(acpi_status
+			    acpi_install_interface_handler
+			    (acpi_interface_handler handler))
 
 /*
  * Global Lock interfaces
Signature · 1 line
2.53.0
author maintainer
#372 lore raw

PATCH 358/570 ACPICA: Allow address_space_handler Install and _REG execution as 2 separate steps

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hans de Goede <[email protected]>

[ Upstream commit 54c516aeb8b39eeae6450b7d8076d381568dca46 ]

ACPI-2.0 says that the EC op_region handler must be available immediately (like the standard default op_region handlers):

Quoting from the ACPI spec version 6.3: "6.5.4 _REG (Region) ...
2. OSPM must make Embedded Controller operation regions, accessed via the Embedded Controllers described in ECDT, available before executing any control method. These operation regions may become inaccessible after OSPM runs _REG(EmbeddedControl, 0)."

So the OS must probe the ECDT described EC and install the OpRegion handler before calling acpi_enable_subsystem() and acpi_initialize_objects().

This is a problem because calling acpi_install_address_space_handler() does not just install the op_region handler, it also runs the EC's _REG method. This _REG method may rely on initialization done by the _INI methods of one of the PCI / _SB root devices.

For the other early/default op_region handlers the op_region handler install and the _REG execution is split into 2 separate steps:
1. acpi_ev_install_region_handlers(), called early from acpi_load_tables()
2. acpi_ev_initialize_op_regions(), called from acpi_initialize_objects()

To fix the EC op_region issue, add 2 bew functions:
1. acpi_install_address_space_handler_no_reg()
2. acpi_execute_reg_methods()
to allow doing things in 2 steps for other op_region handlers, like the EC handler, too.

Note that the comment describing acpi_ev_install_region_handlers() even has an alinea describing this problem. Using the new methods allows users to avoid this problem.

Stable-dep-of: f6484cadbcaf ("ACPI: EC: clean up handlers on probe failure in acpi_ec_setup()")

---
 drivers/acpi/acpica/evxfregn.c | 92 +++++++++++++++++++++++++++++++---
 include/acpi/acpixf.h          | 10 ++++
 2 files changed, 95 insertions(+), 7 deletions(-)
drivers/acpi/acpica/evxfregn.c +85 -7
index 7672d70da850d..282cd4e96e122 100644
@@ -20,13 +20,14 @@ ACPI_MODULE_NAME("evxfregn")
 
 /*******************************************************************************
  *
- * FUNCTION:    acpi_install_address_space_handler
+ * FUNCTION:    acpi_install_address_space_handler_internal
  *
  * PARAMETERS:  device          - Handle for the device
  *              space_id        - The address space ID
  *              handler         - Address of the handler
  *              setup           - Address of the setup function
  *              context         - Value passed to the handler on each access
+ *              Run_reg         - Run _REG methods for this address space?
  *
  * RETURN:      Status
  *
@@ -37,13 +38,16 @@ ACPI_MODULE_NAME("evxfregn")
  * are executed here, and these methods can only be safely executed after
  * the default handlers have been installed and the hardware has been
  * initialized (via acpi_enable_subsystem.)
+ * To avoid this problem pass FALSE for Run_Reg and later on call
+ * acpi_execute_reg_methods() to execute _REG.
  *
  ******************************************************************************/
-acpi_status
-acpi_install_address_space_handler(acpi_handle device,
-				   acpi_adr_space_type space_id,
-				   acpi_adr_space_handler handler,
-				   acpi_adr_space_setup setup, void *context)
+static acpi_status
+acpi_install_address_space_handler_internal(acpi_handle device,
+					    acpi_adr_space_type space_id,
+					    acpi_adr_space_handler handler,
+					    acpi_adr_space_setup setup,
+					    void *context, u8 run_reg)
 {
 	struct acpi_namespace_node *node;
 	acpi_status status;
@@ -80,14 +84,40 @@ acpi_install_address_space_handler(acpi_handle device,
 
 	/* Run all _REG methods for this address space */
 
-	acpi_ev_execute_reg_methods(node, space_id, ACPI_REG_CONNECT);
+	if (run_reg) {
+		acpi_ev_execute_reg_methods(node, space_id, ACPI_REG_CONNECT);
+	}
 
 unlock_and_exit:
 	(void)acpi_ut_release_mutex(ACPI_MTX_NAMESPACE);
 	return_ACPI_STATUS(status);
 }
 
+acpi_status
+acpi_install_address_space_handler(acpi_handle device,
+				   acpi_adr_space_type space_id,
+				   acpi_adr_space_handler handler,
+				   acpi_adr_space_setup setup, void *context)
+{
+	return acpi_install_address_space_handler_internal(device, space_id,
+							   handler, setup,
+							   context, TRUE);
+}
+
 ACPI_EXPORT_SYMBOL(acpi_install_address_space_handler)
+acpi_status
+acpi_install_address_space_handler_no_reg(acpi_handle device,
+					  acpi_adr_space_type space_id,
+					  acpi_adr_space_handler handler,
+					  acpi_adr_space_setup setup,
+					  void *context)
+{
+	return acpi_install_address_space_handler_internal(device, space_id,
+							   handler, setup,
+							   context, FALSE);
+}
+
+ACPI_EXPORT_SYMBOL(acpi_install_address_space_handler_no_reg)
 
 /*******************************************************************************
  *
@@ -226,3 +256,51 @@ acpi_remove_address_space_handler(acpi_handle device,
 }
 
 ACPI_EXPORT_SYMBOL(acpi_remove_address_space_handler)
+/*******************************************************************************
+ *
+ * FUNCTION:    acpi_execute_reg_methods
+ *
+ * PARAMETERS:  device          - Handle for the device
+ *              space_id        - The address space ID
+ *
+ * RETURN:      Status
+ *
+ * DESCRIPTION: Execute _REG for all op_regions of a given space_id.
+ *
+ ******************************************************************************/
+acpi_status
+acpi_execute_reg_methods(acpi_handle device, acpi_adr_space_type space_id)
+{
+	struct acpi_namespace_node *node;
+	acpi_status status;
+
+	ACPI_FUNCTION_TRACE(acpi_execute_reg_methods);
+
+	/* Parameter validation */
+
+	if (!device) {
+		return_ACPI_STATUS(AE_BAD_PARAMETER);
+	}
+
+	status = acpi_ut_acquire_mutex(ACPI_MTX_NAMESPACE);
+	if (ACPI_FAILURE(status)) {
+		return_ACPI_STATUS(status);
+	}
+
+	/* Convert and validate the device handle */
+
+	node = acpi_ns_validate_handle(device);
+	if (node) {
+
+		/* Run all _REG methods for this address space */
+
+		acpi_ev_execute_reg_methods(node, space_id, ACPI_REG_CONNECT);
+	} else {
+		status = AE_BAD_PARAMETER;
+	}
+
+	(void)acpi_ut_release_mutex(ACPI_MTX_NAMESPACE);
+	return_ACPI_STATUS(status);
+}
+
+ACPI_EXPORT_SYMBOL(acpi_execute_reg_methods)
include/acpi/acpixf.h +10 -0
index 2cf4c90730d6b..27ff6bf066ef4 100644
@@ -650,6 +650,16 @@ ACPI_EXTERNAL_RETURN_STATUS(acpi_status
 							       acpi_adr_space_setup
 							       setup,
 							       void *context))
+ACPI_EXTERNAL_RETURN_STATUS(acpi_status
+			    acpi_install_address_space_handler_no_reg
+			    (acpi_handle device, acpi_adr_space_type space_id,
+			     acpi_adr_space_handler handler,
+			     acpi_adr_space_setup setup,
+			     void *context))
+ACPI_EXTERNAL_RETURN_STATUS(acpi_status
+			    acpi_execute_reg_methods(acpi_handle device,
+						     acpi_adr_space_type
+						     space_id))
 ACPI_EXTERNAL_RETURN_STATUS(acpi_status
 			    acpi_remove_address_space_handler(acpi_handle
 							      device,
Signature · 1 line
2.53.0
author maintainer
#373 lore raw

PATCH 359/570 ACPI: EC: Fix EC address space handler unregistration

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hans de Goede <[email protected]>

[ Upstream commit a5072078dbfaa9d70130805766dfa34bbb7bf2a7 ]

When an ECDT table is present the EC address space handler gets registered on the root node. So to unregister it properly the unregister call also must be done on the root node.

Store the ACPI handle used for the acpi_install_address_space_handler() call and use te same handle for the acpi_remove_address_space_handler() call.

Stable-dep-of: f6484cadbcaf ("ACPI: EC: clean up handlers on probe failure in acpi_ec_setup()")

---
 drivers/acpi/ec.c       | 4 +++-
 drivers/acpi/internal.h | 1 +
 2 files changed, 4 insertions(+), 1 deletion(-)
drivers/acpi/ec.c +3 -1
index ddc5b3a3d9b38..4d38a00dbf50a 100644
@@ -1513,6 +1513,7 @@ static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device)
 			return -ENODEV;
 		}
 		set_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags);
+		ec->address_space_handler_holder = ec->handle;
 	}
 
 	if (!device)
@@ -1564,7 +1565,8 @@ static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device)
 static void ec_remove_handlers(struct acpi_ec *ec)
 {
 	if (test_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags)) {
-		if (ACPI_FAILURE(acpi_remove_address_space_handler(ec->handle,
+		if (ACPI_FAILURE(acpi_remove_address_space_handler(
+					ec->address_space_handler_holder,
 					ACPI_ADR_SPACE_EC, &acpi_ec_space_handler)))
 			pr_err("failed to remove space handler\n");
 		clear_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags);
drivers/acpi/internal.h +1 -0
index 54b2be94d23dc..33cf3f38f8530 100644
@@ -168,6 +168,7 @@ static inline void acpi_early_processor_osc(void) {}
    -------------------------------------------------------------------------- */
 struct acpi_ec {
 	acpi_handle handle;
+	acpi_handle address_space_handler_holder;
 	int gpe;
 	int irq;
 	unsigned long command_addr;
Signature · 1 line
2.53.0
author maintainer
#374 lore raw

PATCH 360/570 ACPI: EC: Fix ECDT probe ordering issues

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Hans de Goede <[email protected]>

[ Upstream commit ab4620f58d38206687b9f99d9d2cc1d5a2640985 ]

ACPI-2.0 says that the EC OpRegion handler must be available immediately (like the standard default OpRegion handlers):

Quoting from the ACPI spec version 6.3: "6.5.4 _REG (Region) ...
2. OSPM must make Embedded Controller operation regions, accessed via the Embedded Controllers described in ECDT, available before executing any control method. These operation regions may become inaccessible after OSPM runs _REG(EmbeddedControl, 0)."

So acpi_bus_init() calls acpi_ec_ecdt_probe(), which calls acpi_install_address_space_handler() to install the EC's OpRegion handler, early on.

This not only installs the OpRegion handler, but also calls the EC's _REG method. The _REG method call is a problem because it may rely on initialization done by the _INI methods of one of the PCI / _SB root devs, see for example: https://bugzilla.kernel.org/show_bug.cgi?id=214899 .

Generally speaking _REG methods are executed when the ACPI-device they are part of has a driver bound to it. Where as _INI methods must be executed at table load time (according to the spec). The problem here is that the early acpi_install_address_space_handler() call causes the _REG handler to run too early.

To allow fixing this the ACPICA code now allows to split the OpRegion handler installation and the executing of _REG into 2 separate steps.

This commit uses this ACPICA functionality to fix the EC probe ordering by delaying the executing of _REG for ECDT described ECs till the matching EC device in the DSDT gets parsed and acpi_ec_add() for it gets called. This moves the calling of _REG for the EC on devices with an ECDT to the same point in time where it is called on devices without an ECDT table.

Stable-dep-of: f6484cadbcaf ("ACPI: EC: clean up handlers on probe failure in acpi_ec_setup()")

---
 drivers/acpi/ec.c | 28 ++++++++++++++++++----------
 1 file changed, 18 insertions(+), 10 deletions(-)
drivers/acpi/ec.c +18 -10
index 4d38a00dbf50a..bbc0cfb8fc81b 100644
@@ -96,6 +96,7 @@ enum {
 	EC_FLAGS_QUERY_GUARDING,	/* Guard for SCI_EVT check */
 	EC_FLAGS_EVENT_HANDLER_INSTALLED,	/* Event handler installed */
 	EC_FLAGS_EC_HANDLER_INSTALLED,	/* OpReg handler installed */
+	EC_FLAGS_EC_REG_CALLED,		/* OpReg ACPI _REG method called */
 	EC_FLAGS_QUERY_METHODS_INSTALLED, /* _Qxx handlers installed */
 	EC_FLAGS_STARTED,		/* Driver is started */
 	EC_FLAGS_STOPPED,		/* Driver is stopped */
@@ -1484,6 +1485,7 @@ static bool install_gpio_irq_event_handler(struct acpi_ec *ec)
  * ec_install_handlers - Install service callbacks and register query methods.
  * @ec: Target EC.
  * @device: ACPI device object corresponding to @ec.
+ * @call_reg: If _REG should be called to notify OpRegion availability
  *
  * Install a handler for the EC address space type unless it has been installed
  * already.  If @device is not NULL, also look for EC query methods in the
@@ -1496,7 +1498,8 @@ static bool install_gpio_irq_event_handler(struct acpi_ec *ec)
  * -EPROBE_DEFER if GPIO IRQ acquisition needs to be deferred,
  * or 0 (success) otherwise.
  */
-static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device)
+static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device,
+			       bool call_reg)
 {
 	acpi_status status;
 
@@ -1504,10 +1507,10 @@ static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device)
 
 	if (!test_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags)) {
 		acpi_ec_enter_noirq(ec);
-		status = acpi_install_address_space_handler(ec->handle,
-							    ACPI_ADR_SPACE_EC,
-							    &acpi_ec_space_handler,
-							    NULL, ec);
+		status = acpi_install_address_space_handler_no_reg(ec->handle,
+								   ACPI_ADR_SPACE_EC,
+								   &acpi_ec_space_handler,
+								   NULL, ec);
 		if (ACPI_FAILURE(status)) {
 			acpi_ec_stop(ec, false);
 			return -ENODEV;
@@ -1516,6 +1519,11 @@ static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device)
 		ec->address_space_handler_holder = ec->handle;
 	}
 
+	if (call_reg && !test_bit(EC_FLAGS_EC_REG_CALLED, &ec->flags)) {
+		acpi_execute_reg_methods(ec->handle, ACPI_ADR_SPACE_EC);
+		set_bit(EC_FLAGS_EC_REG_CALLED, &ec->flags);
+	}
+
 	if (!device)
 		return 0;
 
@@ -1602,11 +1610,11 @@ static void ec_remove_handlers(struct acpi_ec *ec)
 	}
 }
 
-static int acpi_ec_setup(struct acpi_ec *ec, struct acpi_device *device)
+static int acpi_ec_setup(struct acpi_ec *ec, struct acpi_device *device, bool call_reg)
 {
 	int ret;
 
-	ret = ec_install_handlers(ec, device);
+	ret = ec_install_handlers(ec, device, call_reg);
 	if (ret)
 		return ret;
 
@@ -1668,7 +1676,7 @@ static int acpi_ec_add(struct acpi_device *device)
 		}
 	}
 
-	ret = acpi_ec_setup(ec, device);
+	ret = acpi_ec_setup(ec, device, true);
 	if (ret)
 		goto err;
 
@@ -1788,7 +1796,7 @@ void __init acpi_ec_dsdt_probe(void)
 	 * At this point, the GPE is not fully initialized, so do not to
 	 * handle the events.
 	 */
-	ret = acpi_ec_setup(ec, NULL);
+	ret = acpi_ec_setup(ec, NULL, true);
 	if (ret) {
 		acpi_ec_free(ec);
 		return;
@@ -1952,7 +1960,7 @@ void __init acpi_ec_ecdt_probe(void)
 	 * At this point, the namespace is not initialized, so do not find
 	 * the namespace objects, or handle the events.
 	 */
-	ret = acpi_ec_setup(ec, NULL);
+	ret = acpi_ec_setup(ec, NULL, false);
 	if (ret) {
 		acpi_ec_free(ec);
 		goto out;
Signature · 1 line
2.53.0
author maintainer
#375 lore raw

PATCH 361/570 ACPI: EC: Install address space handler at the namespace root

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Rafael J. Wysocki <[email protected]>

[ Upstream commit 60fa6ae6e6d09e377fce6f8d9b6f6a4d88769f63 ]

It is reported that _DSM evaluation fails in ucsi_acpi_dsm() on Lenovo IdeaPad Pro 5 due to a missing address space handler for the EC address space:

ACPI Error: No handler for Region [ECSI] (000000007b8176ee) [EmbeddedControl] (20230628/evregion-130)

This happens because if there is no ECDT, the EC driver only registers the EC address space handler for operation regions defined in the EC device scope of the ACPI namespace while the operation region being accessed by the _DSM in question is located beyond that scope.

To address this, modify the ACPI EC driver to install the EC address space handler at the root of the ACPI namespace for the first EC that can be found regardless of whether or not an ECDT is present.

Note that this change is consistent with some examples in the ACPI specification in which EC operation regions located outside the EC device scope are used (for example, see Section 9.17.15 in ACPI 6.5), so the current behavior of the EC driver is arguably questionable.

Stable-dep-of: f6484cadbcaf ("ACPI: EC: clean up handlers on probe failure in acpi_ec_setup()")

---
 drivers/acpi/ec.c       | 25 ++++++++++++++++---------
 drivers/acpi/internal.h |  1 -
 2 files changed, 16 insertions(+), 10 deletions(-)
drivers/acpi/ec.c +16 -9
index bbc0cfb8fc81b..8b54e8fe4a808 100644
@@ -1501,13 +1501,14 @@ static bool install_gpio_irq_event_handler(struct acpi_ec *ec)
 static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device,
 			       bool call_reg)
 {
+	acpi_handle scope_handle = ec == first_ec ? ACPI_ROOT_OBJECT : ec->handle;
 	acpi_status status;
 
 	acpi_ec_start(ec, false);
 
 	if (!test_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags)) {
 		acpi_ec_enter_noirq(ec);
-		status = acpi_install_address_space_handler_no_reg(ec->handle,
+		status = acpi_install_address_space_handler_no_reg(scope_handle,
 								   ACPI_ADR_SPACE_EC,
 								   &acpi_ec_space_handler,
 								   NULL, ec);
@@ -1516,11 +1517,10 @@ static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device,
 			return -ENODEV;
 		}
 		set_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags);
-		ec->address_space_handler_holder = ec->handle;
 	}
 
 	if (call_reg && !test_bit(EC_FLAGS_EC_REG_CALLED, &ec->flags)) {
-		acpi_execute_reg_methods(ec->handle, ACPI_ADR_SPACE_EC);
+		acpi_execute_reg_methods(scope_handle, ACPI_ADR_SPACE_EC);
 		set_bit(EC_FLAGS_EC_REG_CALLED, &ec->flags);
 	}
 
@@ -1572,10 +1572,13 @@ static int ec_install_handlers(struct acpi_ec *ec, struct acpi_device *device,
 
 static void ec_remove_handlers(struct acpi_ec *ec)
 {
+	acpi_handle scope_handle = ec == first_ec ? ACPI_ROOT_OBJECT : ec->handle;
+
 	if (test_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags)) {
 		if (ACPI_FAILURE(acpi_remove_address_space_handler(
-					ec->address_space_handler_holder,
-					ACPI_ADR_SPACE_EC, &acpi_ec_space_handler)))
+						scope_handle,
+						ACPI_ADR_SPACE_EC,
+						&acpi_ec_space_handler)))
 			pr_err("failed to remove space handler\n");
 		clear_bit(EC_FLAGS_EC_HANDLER_INSTALLED, &ec->flags);
 	}
@@ -1614,14 +1617,18 @@ static int acpi_ec_setup(struct acpi_ec *ec, struct acpi_device *device, bool ca
 {
 	int ret;
 
-	ret = ec_install_handlers(ec, device, call_reg);
-	if (ret)
-		return ret;
-
 	/* First EC capable of handling transactions */
 	if (!first_ec)
 		first_ec = ec;
 
+	ret = ec_install_handlers(ec, device, call_reg);
+	if (ret) {
+		if (ec == first_ec)
+			first_ec = NULL;
+
+		return ret;
+	}
+
 	pr_info("EC_CMD/EC_SC=0x%lx, EC_DATA=0x%lx\n", ec->command_addr,
 		ec->data_addr);
 
drivers/acpi/internal.h +0 -1
index 33cf3f38f8530..54b2be94d23dc 100644
@@ -168,7 +168,6 @@ static inline void acpi_early_processor_osc(void) {}
    -------------------------------------------------------------------------- */
 struct acpi_ec {
 	acpi_handle handle;
-	acpi_handle address_space_handler_holder;
 	int gpe;
 	int irq;
 	unsigned long command_addr;
Signature · 1 line
2.53.0
author maintainer
#376 lore raw

PATCH 362/570 ACPI: EC: clean up handlers on probe failure in acpi_ec_setup()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Weiming Shi <[email protected]>

[ Upstream commit f6484cadbcaf26b5844b51bd7307a663dda48ef6 ]

When ec_install_handlers() returns -EPROBE_DEFER on reduced-hardware platforms, it has already started the EC and installed the address space handler with the struct acpi_ec pointer as handler context. However, acpi_ec_setup() propagates the error without any cleanup.

The caller acpi_ec_add() then frees the struct acpi_ec for non-boot instances, leaving a dangling handler context in ACPICA.

Any subsequent AML evaluation that accesses an EC OpRegion field dispatches into acpi_ec_space_handler() with the freed pointer, causing a use-after-free:

 BUG: KASAN: slab-use-after-free in mutex_lock (kernel/locking/mutex.c:289)
 Write of size 8 at addr ffff88800721de38 by task init/1
 Call Trace:
  <TASK>
  mutex_lock (kernel/locking/mutex.c:289)
  acpi_ec_space_handler (drivers/acpi/ec.c:1362)
  acpi_ev_address_space_dispatch (drivers/acpi/acpica/evregion.c:293)
  acpi_ex_access_region (drivers/acpi/acpica/exfldio.c:246)
  acpi_ex_field_datum_io (drivers/acpi/acpica/exfldio.c:509)
  acpi_ex_extract_from_field (drivers/acpi/acpica/exfldio.c:700)
  acpi_ex_read_data_from_field (drivers/acpi/acpica/exfield.c:327)
  acpi_ex_resolve_node_to_value (drivers/acpi/acpica/exresolv.c:392)
  </TASK>

 Allocated by task 1:
  acpi_ec_alloc (drivers/acpi/ec.c:1424)
  acpi_ec_add (drivers/acpi/ec.c:1692)

 Freed by task 1:
  kfree (mm/slub.c:6876)
  acpi_ec_add (drivers/acpi/ec.c:1751)

The bug triggers on reduced-hardware EC platforms (ec->gpe < 0) when the GPIO IRQ provider defers probing. Once the stale handler exists, any unprivileged sysfs read that causes AML to touch an EC OpRegion (battery, thermal, backlight) exercises the dangling pointer.

Fix this by calling ec_remove_handlers() in the error path of acpi_ec_setup() before clearing first_ec. ec_remove_handlers() checks each EC_FLAGS_* bit before acting, so it is safe to call regardless of how far ec_install_handlers() progressed:

-ENODEV (handler not installed): only calls acpi_ec_stop() -EPROBE_DEFER (handler installed): removes handler, stops EC

Fixes 03e9a0e05739 ("ACPI: EC: Consolidate event handler installation code")
---
 drivers/acpi/ec.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/acpi/ec.c +2 -0
index 8b54e8fe4a808..25de4e6b4c2ed 100644
@@ -1623,6 +1623,8 @@ static int acpi_ec_setup(struct acpi_ec *ec, struct acpi_device *device, bool ca
 
 	ret = ec_install_handlers(ec, device, call_reg);
 	if (ret) {
+		ec_remove_handlers(ec);
+
 		if (ec == first_ec)
 			first_ec = NULL;
 
Signature · 1 line
2.53.0
author maintainer
#377 lore raw

PATCH 363/570 hwmon: (adm1177) fix sysfs ABI violation and current unit conversion

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Sanman Pradhan <[email protected]>

[ Upstream commit bf08749a6abb6d1959bfdc0edc32c640df407558 ]

The adm1177 driver exposes the current alert threshold through hwmon_curr_max_alarm. This violates the hwmon sysfs ABI, where *_alarm attributes are read-only status flags and writable thresholds must use currN_max.

The driver also stores the threshold internally in microamps, while currN_max is defined in milliamps. Convert the threshold accordingly on both the read and write paths.

Widen the cached threshold and related calculations to 64 bits so that small shunt resistor values do not cause truncation or overflow. Also use 64-bit arithmetic for the mA/uA conversions, clamp writes to the range the hardware can represent, and propagate failures from adm1177_write_alert_thr() instead of silently ignoring them.

Update the hwmon documentation to reflect the attribute rename and the correct units returned by the driver.

Fixes 09b08ac9e8d5 ("hwmon: (adm1177) Add ADM1177 Hot Swap Controller and Digital Power Monitor driver")
---
 Documentation/hwmon/adm1177.rst |  8 ++---
 drivers/hwmon/adm1177.c         | 54 +++++++++++++++++++--------------
 2 files changed, 35 insertions(+), 27 deletions(-)
Documentation/hwmon/adm1177.rst +4 -4
index 1c85a2af92bf7..375f6d6e03a7d 100644
@@ -27,10 +27,10 @@ for details.
 Sysfs entries
 -------------
 
-The following attributes are supported. Current maxim attribute
+The following attributes are supported. Current maximum attribute
 is read-write, all other attributes are read-only.
 
-in0_input		Measured voltage in microvolts.
+in0_input		Measured voltage in millivolts.
 
-curr1_input		Measured current in microamperes.
-curr1_max_alarm		Overcurrent alarm in microamperes.
+curr1_input		Measured current in milliamperes.
+curr1_max		Overcurrent shutdown threshold in milliamperes.
drivers/hwmon/adm1177.c +31 -23

111 lines not on this page · raw message

Signature · 1 line
2.53.0
author maintainer
#378 lore raw

PATCH 364/570 sysctl: fix uninitialized variable in proc_do_large_bitmap

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Marc Buerg <[email protected]>

[ Upstream commit f63a9df7e3f9f842945d292a19d9938924f066f9 ]

proc_do_large_bitmap() does not initialize variable c, which is expected to be set to a trailing character by proc_get_long().

However, proc_get_long() only sets c when the input buffer contains a trailing character after the parsed value.

If c is not initialized it may happen to contain a '-'. If this is the case proc_do_large_bitmap() expects to be able to parse a second part of the input buffer. If there is no second part an unjustified -EINVAL will be returned.

Initialize c to 0 to prevent returning -EINVAL on valid input.

Fixes 9f977fb7ae9d ("sysctl: add proc_do_large_bitmap")
---
 kernel/sysctl.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/sysctl.c +1 -1
index eaf9dd6a2f12f..ac16c3084c96c 100644
@@ -1528,7 +1528,7 @@ int proc_do_large_bitmap(struct ctl_table *table, int write,
 	unsigned long bitmap_len = table->maxlen;
 	unsigned long *bitmap = *(unsigned long **) table->data;
 	unsigned long *tmp_bitmap = NULL;
-	char tr_a[] = { '-', ',', '\n' }, tr_b[] = { ',', '\n', 0 }, c;
+	char tr_a[] = { '-', ',', '\n' }, tr_b[] = { ',', '\n', 0 }, c = 0;
 
 	if (!bitmap || !bitmap_len || !left || (*ppos && !write)) {
 		*lenp = 0;
Signature · 1 line
2.53.0
author maintainer
#379 lore raw

PATCH 365/570 ASoC: adau1372: Fix unchecked clk_prepare_enable() return value

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jihed Chaibi <[email protected]>

[ Upstream commit 326fe8104a4020d30080d37ac8b6b43893cdebca ]

adau1372_set_power() calls clk_prepare_enable() but discards the return value. If the clock enable fails, the driver proceeds to access registers on unpowered hardware, potentially causing silent corruption.

Make adau1372_set_power() return int and propagate the error from clk_prepare_enable(). Update adau1372_set_bias_level() to return the error directly for the STANDBY and OFF cases.

Fixes 6cd4c6459e47 ("ASoC: Add ADAU1372 audio CODEC support")
---
 sound/soc/codecs/adau1372.c | 17 ++++++++++-------
 1 file changed, 10 insertions(+), 7 deletions(-)
sound/soc/codecs/adau1372.c +10 -7
index 6811a8b3866dd..0997d5f3e4039 100644
@@ -781,15 +781,18 @@ static void adau1372_enable_pll(struct adau1372 *adau1372)
 		dev_err(adau1372->dev, "Failed to lock PLL\n");
 }
 
-static void adau1372_set_power(struct adau1372 *adau1372, bool enable)
+static int adau1372_set_power(struct adau1372 *adau1372, bool enable)
 {
 	if (adau1372->enabled == enable)
-		return;
+		return 0;
 
 	if (enable) {
 		unsigned int clk_ctrl = ADAU1372_CLK_CTRL_MCLK_EN;
+		int ret;
 
-		clk_prepare_enable(adau1372->mclk);
+		ret = clk_prepare_enable(adau1372->mclk);
+		if (ret)
+			return ret;
 		if (adau1372->pd_gpio)
 			gpiod_set_value(adau1372->pd_gpio, 0);
 
@@ -828,6 +831,8 @@ static void adau1372_set_power(struct adau1372 *adau1372, bool enable)
 	}
 
 	adau1372->enabled = enable;
+
+	return 0;
 }
 
 static int adau1372_set_bias_level(struct snd_soc_component *component,
@@ -841,11 +846,9 @@ static int adau1372_set_bias_level(struct snd_soc_component *component,
 	case SND_SOC_BIAS_PREPARE:
 		break;
 	case SND_SOC_BIAS_STANDBY:
-		adau1372_set_power(adau1372, true);
-		break;
+		return adau1372_set_power(adau1372, true);
 	case SND_SOC_BIAS_OFF:
-		adau1372_set_power(adau1372, false);
-		break;
+		return adau1372_set_power(adau1372, false);
 	}
 
 	return 0;
Signature · 1 line
2.53.0
author maintainer
#380 lore raw

PATCH 366/570 ASoC: adau1372: Fix clock leak on PLL lock failure

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jihed Chaibi <[email protected]>

[ Upstream commit bfe6a264effcb6fe99ad7ceaf9e8c7439fc9555b ]

adau1372_enable_pll() was a void function that logged a dev_err() on PLL lock timeout but did not propagate the error. As a result, adau1372_set_power() would continue with adau1372->enabled set to true despite the PLL being unlocked, and the mclk left enabled with no corresponding disable on the error path.

Convert adau1372_enable_pll() to return int, using -ETIMEDOUT on lock timeout and propagating regmap errors directly. In adau1372_set_power(), check the return value and unwind in reverse order: restore regcache to cache-only mode, reassert GPIO power-down, and disable the clock before returning the error.

Fixes 6cd4c6459e47 ("ASoC: Add ADAU1372 audio CODEC support")
---
 sound/soc/codecs/adau1372.c | 17 ++++++++++++++---
 1 file changed, 14 insertions(+), 3 deletions(-)
sound/soc/codecs/adau1372.c +14 -3
index 0997d5f3e4039..e20145812f12b 100644
@@ -761,7 +761,7 @@ static int adau1372_startup(struct snd_pcm_substream *substream, struct snd_soc_
 	return 0;
 }
 
-static void adau1372_enable_pll(struct adau1372 *adau1372)
+static int adau1372_enable_pll(struct adau1372 *adau1372)
 {
 	unsigned int val, timeout = 0;
 	int ret;
@@ -777,8 +777,12 @@ static void adau1372_enable_pll(struct adau1372 *adau1372)
 		timeout++;
 	} while (!(val & 1) && timeout < 3);
 
-	if (ret < 0 || !(val & 1))
+	if (ret < 0 || !(val & 1)) {
 		dev_err(adau1372->dev, "Failed to lock PLL\n");
+		return ret < 0 ? ret : -ETIMEDOUT;
+	}
+
+	return 0;
 }
 
 static int adau1372_set_power(struct adau1372 *adau1372, bool enable)
@@ -806,7 +810,14 @@ static int adau1372_set_power(struct adau1372 *adau1372, bool enable)
 		 * accessed.
 		 */
 		if (adau1372->use_pll) {
-			adau1372_enable_pll(adau1372);
+			ret = adau1372_enable_pll(adau1372);
+			if (ret) {
+				regcache_cache_only(adau1372->regmap, true);
+				if (adau1372->pd_gpio)
+					gpiod_set_value(adau1372->pd_gpio, 1);
+				clk_disable_unprepare(adau1372->mclk);
+				return ret;
+			}
 			clk_ctrl |= ADAU1372_CLK_CTRL_CLKSRC;
 		}
 
Signature · 1 line
2.53.0
author maintainer
#381 lore raw

PATCH 367/570 spi: spi-fsl-lpspi: fix teardown order issue (UAF)

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Marc Kleine-Budde <[email protected]>

[ Upstream commit b341c1176f2e001b3adf0b47154fc31589f7410e ]

There is a teardown order issue in the driver. The SPI controller is registered using devm_spi_register_controller(), which delays unregistration of the SPI controller until after the fsl_lpspi_remove() function returns.

As the fsl_lpspi_remove() function synchronously tears down the DMA channels, a running SPI transfer triggers the following NULL pointer dereference due to use after free:

| fsl_lpspi 42550000.spi: I/O Error in DMA RX
| Unable to handle kernel NULL pointer dereference at virtual address 0000000000000000 [...]

| Call trace:
|  fsl_lpspi_dma_transfer+0x260/0x340 [spi_fsl_lpspi]
|  fsl_lpspi_transfer_one+0x198/0x448 [spi_fsl_lpspi]
|  spi_transfer_one_message+0x49c/0x7c8
|  __spi_pump_transfer_message+0x120/0x420
|  __spi_sync+0x2c4/0x520
|  spi_sync+0x34/0x60
|  spidev_message+0x20c/0x378 [spidev]
|  spidev_ioctl+0x398/0x750 [spidev]

[...]

Switch from devm_spi_register_controller() to spi_register_controller() in fsl_lpspi_probe() and add the corresponding spi_unregister_controller() in fsl_lpspi_remove().

---
 drivers/spi/spi-fsl-lpspi.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/spi/spi-fsl-lpspi.c +2 -1
index 319cd96bd201b..540815ae49e78 100644
@@ -914,7 +914,7 @@ static int fsl_lpspi_probe(struct platform_device *pdev)
 		enable_irq(irq);
 	}
 
-	ret = devm_spi_register_controller(&pdev->dev, controller);
+	ret = spi_register_controller(controller);
 	if (ret < 0) {
 		dev_err_probe(&pdev->dev, ret, "spi_register_controller error\n");
 		goto free_dma;
@@ -943,6 +943,7 @@ static int fsl_lpspi_remove(struct platform_device *pdev)
 	struct fsl_lpspi_data *fsl_lpspi =
 				spi_controller_get_devdata(controller);
 
+	spi_unregister_controller(controller);
 	fsl_lpspi_dma_exit(controller);
 
 	pm_runtime_dont_use_autosuspend(fsl_lpspi->dev);
Signature · 1 line
2.53.0
author maintainer
#382 lore raw

PATCH 368/570 s390/syscalls: Add spectre boundary for syscall dispatch table

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Greg Kroah-Hartman <[email protected]>

commit 48b8814e25d073dd84daf990a879a820bad2bcbd upstream.

The s390 syscall number is directly controlled by userspace, but does not have an array_index_nospec() boundary to prevent access past the syscall function pointer tables.

Cc Heiko Carstens <[email protected]>
Cc Vasily Gorbik <[email protected]>
Cc Alexander Gordeev <[email protected]>
Cc Christian Borntraeger <[email protected]>
Cc Sven Schnelle <[email protected]>
Cc Arnd Bergmann <[email protected]>
Fixes 56e62a737028 ("s390: convert to generic entry")
Assisted-by gkh_clanker_2000
---
 arch/s390/kernel/syscall.c |    2 ++
 1 file changed, 2 insertions(+)
arch/s390/kernel/syscall.c +2 -0
@@ -13,6 +13,7 @@
  */
 
 #include <linux/errno.h>
+#include <linux/nospec.h>
 #include <linux/sched.h>
 #include <linux/mm.h>
 #include <linux/fs.h>
@@ -141,6 +142,7 @@ static void do_syscall(struct pt_regs *r
 	if (likely(nr >= NR_syscalls))
 		goto out;
 	do {
+		nr = array_index_nospec(nr, NR_syscalls);
 		regs->gprs[2] = current->thread.sys_call_table[nr](regs);
 	} while (test_and_clear_pt_regs_flag(regs, PIF_EXECVE_PGSTE_RESTART));
 out:
author maintainer
#383 lore raw

PATCH 369/570 s390/barrier: Make array_index_mask_nospec() __always_inline

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vasily Gorbik <[email protected]>

commit c5c0a268b38adffbb2e70e6957017537ff54c157 upstream.

Mark array_index_mask_nospec() as __always_inline to guarantee the mitigation is emitted inline regardless of compiler inlining decisions.

Fixes e2dd833389cc ("s390: add optimized array_index_mask_nospec")
---
 arch/s390/include/asm/barrier.h |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
arch/s390/include/asm/barrier.h +2 -2
@@ -56,8 +56,8 @@ do {									\
  * @size: number of elements in array
  */
 #define array_index_mask_nospec array_index_mask_nospec
-static inline unsigned long array_index_mask_nospec(unsigned long index,
-						    unsigned long size)
+static __always_inline unsigned long array_index_mask_nospec(unsigned long index,
+							     unsigned long size)
 {
 	unsigned long mask;
 
author maintainer
#384 lore raw

PATCH 370/570 can: gw: fix OOB heap access in cgw_csum_crc8_rel()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ali Norouzi <[email protected]>

commit b9c310d72783cc2f30d103eed83920a5a29c671a upstream.

cgw_csum_crc8_rel() correctly computes bounds-safe indices via calc_idx():

    int from = calc_idx(crc8->from_idx, cf->len);
    int to   = calc_idx(crc8->to_idx,   cf->len);
    int res  = calc_idx(crc8->result_idx, cf->len);

    if (from < 0 || to < 0 || res < 0)
        return;

However, the loop and the result write then use the raw s8 fields directly instead of the computed variables:

    for (i = crc8->from_idx; ...)        /* BUG: raw negative index */
    cf->data[crc8->result_idx] = ...;    /* BUG: raw negative index */

With from_idx = to_idx = result_idx = -64 on a 64-byte CAN FD frame, calc_idx(-64, 64) = 0 so the guard passes, but the loop iterates with i = -64, reading cf->data[-64], and the write goes to cf->data[-64]. This write might end up to 56 (7.0-rc) or 40 (<= 6.19) bytes before the start of the canfd_frame on the heap.

The companion function cgw_csum_xor_rel() uses from/to/res correctly throughout; fix cgw_csum_crc8_rel() to match.

Confirmed with KASAN on linux-7.0-rc2:

  BUG: KASAN: slab-out-of-bounds in cgw_csum_crc8_rel+0x515/0x5b0
  Read of size 1 at addr ffff8880076619c8 by task poc_cgw_oob/62

To configure the can-gw crc8 checksums CAP_NET_ADMIN is needed.

---
 net/can/gw.c |    6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)
net/can/gw.c +3 -3
@@ -374,10 +374,10 @@ static void cgw_csum_crc8_rel(struct can
 		return;
 
 	if (from <= to) {
-		for (i = crc8->from_idx; i <= crc8->to_idx; i++)
+		for (i = from; i <= to; i++)
 			crc = crc8->crctab[crc ^ cf->data[i]];
 	} else {
-		for (i = crc8->from_idx; i >= crc8->to_idx; i--)
+		for (i = from; i >= to; i--)
 			crc = crc8->crctab[crc ^ cf->data[i]];
 	}
 
@@ -396,7 +396,7 @@ static void cgw_csum_crc8_rel(struct can
 		break;
 	}
 
-	cf->data[crc8->result_idx] = crc ^ crc8->final_xor_val;
+	cf->data[res] = crc ^ crc8->final_xor_val;
 }
 
 static void cgw_csum_crc8_pos(struct canfd_frame *cf,
author maintainer
#385 lore raw

PATCH 371/570 cpufreq: conservative: Reset requested_freq on limits change

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Viresh Kumar <[email protected]>

commit 6a28fb8cb28b9eb39a392e531d938a889eacafc5 upstream.

A recently reported issue highlighted that the cached requested_freq is not guaranteed to stay in sync with policy->cur. If the platform changes the actual CPU frequency after the governor sets one (e.g. due to platform-specific frequency scaling) and a re-sync occurs later, policy->cur may diverge from requested_freq.

This can lead to incorrect behavior in the conservative governor. For example, the governor may assume the CPU is already running at the maximum frequency and skip further increases even though there is still headroom.

Avoid this by resetting the cached requested_freq to policy->cur on detecting a change in policy limits.

---
 drivers/cpufreq/cpufreq_conservative.c |   12 ++++++++++++
 drivers/cpufreq/cpufreq_governor.c     |    3 +++
 drivers/cpufreq/cpufreq_governor.h     |    1 +
 3 files changed, 16 insertions(+)
drivers/cpufreq/cpufreq_conservative.c +12 -0
@@ -311,6 +311,17 @@ static void cs_start(struct cpufreq_poli
 	dbs_info->requested_freq = policy->cur;
 }
 
+static void cs_limits(struct cpufreq_policy *policy)
+{
+	struct cs_policy_dbs_info *dbs_info = to_dbs_info(policy->governor_data);
+
+	/*
+	 * The limits have changed, so may have the current frequency. Reset
+	 * requested_freq to avoid any unintended outcomes due to the mismatch.
+	 */
+	dbs_info->requested_freq = policy->cur;
+}
+
 static struct dbs_governor cs_governor = {
 	.gov = CPUFREQ_DBS_GOVERNOR_INITIALIZER("conservative"),
 	.kobj_type = { .default_attrs = cs_attributes },
@@ -320,6 +331,7 @@ static struct dbs_governor cs_governor =
 	.init = cs_init,
 	.exit = cs_exit,
 	.start = cs_start,
+	.limits = cs_limits,
 };
 
 #define CPU_FREQ_GOV_CONSERVATIVE	(cs_governor.gov)
drivers/cpufreq/cpufreq_governor.c +3 -0
@@ -561,6 +561,7 @@ EXPORT_SYMBOL_GPL(cpufreq_dbs_governor_s
 
 void cpufreq_dbs_governor_limits(struct cpufreq_policy *policy)
 {
+	struct dbs_governor *gov = dbs_governor_of(policy);
 	struct policy_dbs_info *policy_dbs;
 
 	/* Protect gov->gdbs_data against cpufreq_dbs_governor_exit() */
@@ -572,6 +573,8 @@ void cpufreq_dbs_governor_limits(struct
 	mutex_lock(&policy_dbs->update_mutex);
 	cpufreq_policy_apply_limits(policy);
 	gov_update_sample_delay(policy_dbs, 0);
+	if (gov->limits)
+		gov->limits(policy);
 	mutex_unlock(&policy_dbs->update_mutex);
 
 out:
drivers/cpufreq/cpufreq_governor.h +1 -0
@@ -140,6 +140,7 @@ struct dbs_governor {
 	int (*init)(struct dbs_data *dbs_data);
 	void (*exit)(struct dbs_data *dbs_data);
 	void (*start)(struct cpufreq_policy *policy);
+	void (*limits)(struct cpufreq_policy *policy);
 };
 
 static inline struct dbs_governor *dbs_governor_of(struct cpufreq_policy *policy)
author maintainer
#386 lore raw

PATCH 372/570 media: mc, v4l2: serialize REINIT and REQBUFS with req_queue_mutex

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yuchan Nam <[email protected]>

commit bef4f4a88b73e4cc550d25f665b8a9952af22773 upstream.

MEDIA_REQUEST_IOC_REINIT can run concurrently with VIDIOC_REQBUFS(0) queue teardown paths. This can race request object cleanup against vb2 queue cancellation and lead to use-after-free reports.

We already serialize request queueing against STREAMON/OFF with req_queue_mutex. Extend that serialization to REQBUFS, and also take the same mutex in media_request_ioctl_reinit() so REINIT is in the same exclusion domain.

This keeps request cleanup and queue cancellation from running in parallel for request-capable devices.

Fixes 6093d3002eab ("media: vb2: keep a reference to the request until dqbuf")
---
 drivers/media/mc/mc-request.c        |    5 +++++
 drivers/media/v4l2-core/v4l2-ioctl.c |    5 +++--
 2 files changed, 8 insertions(+), 2 deletions(-)
drivers/media/mc/mc-request.c +5 -0
@@ -190,6 +190,8 @@ static long media_request_ioctl_reinit(s
 	struct media_device *mdev = req->mdev;
 	unsigned long flags;
 
+	mutex_lock(&mdev->req_queue_mutex);
+
 	spin_lock_irqsave(&req->lock, flags);
 	if (req->state != MEDIA_REQUEST_STATE_IDLE &&
 	    req->state != MEDIA_REQUEST_STATE_COMPLETE) {
@@ -197,6 +199,7 @@ static long media_request_ioctl_reinit(s
 			"request: %s not in idle or complete state, cannot reinit\n",
 			req->debug_str);
 		spin_unlock_irqrestore(&req->lock, flags);
+		mutex_unlock(&mdev->req_queue_mutex);
 		return -EBUSY;
 	}
 	if (req->access_count) {
@@ -204,6 +207,7 @@ static long media_request_ioctl_reinit(s
 			"request: %s is being accessed, cannot reinit\n",
 			req->debug_str);
 		spin_unlock_irqrestore(&req->lock, flags);
+		mutex_unlock(&mdev->req_queue_mutex);
 		return -EBUSY;
 	}
 	req->state = MEDIA_REQUEST_STATE_CLEANING;
@@ -214,6 +218,7 @@ static long media_request_ioctl_reinit(s
 	spin_lock_irqsave(&req->lock, flags);
 	req->state = MEDIA_REQUEST_STATE_IDLE;
 	spin_unlock_irqrestore(&req->lock, flags);
+	mutex_unlock(&mdev->req_queue_mutex);
 
 	return 0;
 }
drivers/media/v4l2-core/v4l2-ioctl.c +3 -2
@@ -2926,13 +2926,14 @@ static long __video_do_ioctl(struct file
 		vfh = file->private_data;
 
 	/*
-	 * We need to serialize streamon/off with queueing new requests.
+	 * We need to serialize streamon/off/reqbufs with queueing new requests.
 	 * These ioctls may trigger the cancellation of a streaming
 	 * operation, and that should not be mixed with queueing a new
 	 * request at the same time.
 	 */
 	if (v4l2_device_supports_requests(vfd->v4l2_dev) &&
-	    (cmd == VIDIOC_STREAMON || cmd == VIDIOC_STREAMOFF)) {
+	    (cmd == VIDIOC_STREAMON || cmd == VIDIOC_STREAMOFF ||
+	     cmd == VIDIOC_REQBUFS)) {
 		req_queue_lock = &vfd->v4l2_dev->mdev->req_queue_mutex;
 
 		if (mutex_lock_interruptible(req_queue_lock))
author maintainer
#387 lore raw

PATCH 373/570 virtio_net: Fix UAF on dst_ops when IFF_XMIT_DST_RELEASE is cleared and napi_tx is false

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: xietangxin <[email protected]>

commit ba8bda9a0896746053aa97ac6c3e08168729172c upstream.

A UAF issue occurs when the virtio_net driver is configured with napi_tx=N and the device's IFF_XMIT_DST_RELEASE flag is cleared
(e.g., during the configuration of tc route filter rules).

When IFF_XMIT_DST_RELEASE is removed from the net_device, the network stack expects the driver to hold the reference to skb->dst until the packet is fully transmitted and freed. In virtio_net with napi_tx=N, skbs may remain in the virtio transmit ring for an extended period.

If the network namespace is destroyed while these skbs are still pending, the corresponding dst_ops structure has freed. When a subsequent packet is transmitted, free_old_xmit() is triggered to clean up old skbs. It then calls dst_release() on the skb associated with the stale dst_entry. Since the dst_ops (referenced by the dst_entry) has already been freed, a UAF kernel paging request occurs.

fix it by adds skb_dst_drop(skb) in start_xmit to explicitly release the dst reference before the skb is queued in virtio_net.

Call Trace:
 Unable to handle kernel paging request at virtual address ffff80007e150000
 CPU: 2 UID: 0 PID: 6236 Comm: ping Kdump: loaded Not tainted 7.0.0-rc1+ #6 PREEMPT
  ...
  percpu_counter_add_batch+0x3c/0x158 lib/percpu_counter.c:98 (P)
  dst_release+0xe0/0x110  net/core/dst.c:177
  skb_release_head_state+0xe8/0x108 net/core/skbuff.c:1177
  sk_skb_reason_drop+0x54/0x2d8 net/core/skbuff.c:1255
  dev_kfree_skb_any_reason+0x64/0x78 net/core/dev.c:3469
  napi_consume_skb+0x1c4/0x3a0 net/core/skbuff.c:1527
  __free_old_xmit+0x164/0x230  drivers/net/virtio_net.c:611 [virtio_net]
  free_old_xmit drivers/net/virtio_net.c:1081 [virtio_net]
  start_xmit+0x7c/0x530 drivers/net/virtio_net.c:3329 [virtio_net]
  ...

Reproduction Steps:
NETDEV="enp3s0"

config_qdisc_route_filter() {
tc qdisc del dev $NETDEV root
tc qdisc add dev $NETDEV root handle 1: prio tc filter add dev $NETDEV parent 1:0 \
protocol ip prio 100 route to 100 flowid 1:1
ip route add 192.168.1.100/32 dev $NETDEV realm 100
}

test_ns() {
ip netns add testns
ip link set $NETDEV netns testns
ip netns exec testns ifconfig $NETDEV 10.0.32.46/24 ip netns exec testns ping -c 1 10.0.32.1
ip netns del testns
}

config_qdisc_route_filter

test_ns
sleep 2
test_ns

Fixes f2fc6a54585a ("[NETNS][IPV6] route6 - move ip6_dst_ops inside the network namespace")
Fixes 0287587884b1 ("net: better IFF_XMIT_DST_RELEASE support")
---
 drivers/net/virtio_net.c |    1 +
 1 file changed, 1 insertion(+)
drivers/net/virtio_net.c +1 -0
@@ -1831,6 +1831,7 @@ static netdev_tx_t start_xmit(struct sk_
 	/* Don't wait up for transmitted skbs to be freed. */
 	if (!use_napi) {
 		skb_orphan(skb);
+		skb_dst_drop(skb);
 		nf_reset_ct(skb);
 	}
 
author maintainer
#388 lore raw

PATCH 374/570 erofs: add GFP_NOIO in the bio completion if needed

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jiucheng Xu <[email protected]>

commit c23df30915f83e7257c8625b690a1cece94142a0 upstream.

The bio completion path in the process context (e.g. dm-verity) will directly call into decompression rather than trigger another workqueue context for minimal scheduling latencies, which can then call vm_map_ram() with GFP_KERNEL.

Due to insufficient memory, vm_map_ram() may generate memory swapping I/O, which can cause submit_bio_wait to deadlock in some scenarios.

Trimmed down the call stack, as follows:

f2fs_submit_read_io
submit_bio //bio_list is initialized.
mmc_blk_mq_recovery
z_erofs_endio
vm_map_ram
__pte_alloc_kernel
__alloc_pages_direct_reclaim
shrink_folio_list
__swap_writepage
submit_bio_wait //bio_list is non-NULL, hang!!!

Use memalloc_noio_{save,restore}() to wrap up this path.

---
 fs/erofs/zdata.c |    3 +++
 1 file changed, 3 insertions(+)
fs/erofs/zdata.c +3 -0
@@ -782,6 +782,7 @@ static void z_erofs_decompress_kickoff(s
 				       bool sync, int bios)
 {
 	struct erofs_sb_info *const sbi = EROFS_SB(io->sb);
+	int gfp_flag;
 
 	/* wake up the caller thread for sync decompression */
 	if (sync) {
@@ -802,7 +803,9 @@ static void z_erofs_decompress_kickoff(s
 		sbi->opt.readahead_sync_decompress = true;
 		return;
 	}
+	gfp_flag = memalloc_noio_save();
 	z_erofs_decompressqueue_work(&io->u.work);
+	memalloc_noio_restore(gfp_flag);
 }
 
 static bool z_erofs_page_is_invalidated(struct page *page)
author maintainer
#389 lore raw

PATCH 375/570 alarmtimer: Fix argument order in alarm_timer_forward()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Zhan Xusheng <[email protected]>

commit 5d16467ae56343b9205caedf85e3a131e0914ad8 upstream.

alarm_timer_forward() passes arguments to alarm_forward() in the wrong order:

  alarm_forward(alarm, timr->it_interval, now);

However, alarm_forward() is defined as:

  u64 alarm_forward(struct alarm *alarm, ktime_t now, ktime_t interval);

and uses the second argument as the current time:

  delta = ktime_sub(now, alarm->node.expires);

Passing the interval as "now" results in incorrect delta computation, which can lead to missed expirations or incorrect overrun accounting.

This issue has been present since the introduction of alarm_timer_forward().

Fix this by swapping the arguments.

---
 kernel/time/alarmtimer.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
kernel/time/alarmtimer.c +1 -1
@@ -609,7 +609,7 @@ static s64 alarm_timer_forward(struct k_
 {
 	struct alarm *alarm = &timr->it.alarm.alarmtimer;
 
-	return alarm_forward(alarm, timr->it_interval, now);
+	return alarm_forward(alarm, now, timr->it_interval);
 }
 
 /**
author maintainer
#390 lore raw

PATCH 376/570 scsi: ibmvfc: Fix OOB access in ibmvfc_discover_targets_done()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Tyllis Xu <[email protected]>

commit 61d099ac4a7a8fb11ebdb6e2ec8d77f38e77362f upstream.

A malicious or compromised VIO server can return a num_written value in the discover targets MAD response that exceeds max_targets. This value is stored directly in vhost->num_targets without validation, and is then used as the loop bound in ibmvfc_alloc_targets() to index into disc_buf[], which is only allocated for max_targets entries. Indices at or beyond max_targets access kernel memory outside the DMA-coherent allocation. The out-of-bounds data is subsequently embedded in Implicit Logout and PLOGI MADs that are sent back to the VIO server, leaking kernel memory.

Fix by clamping num_written to max_targets before storing it.

---
 drivers/scsi/ibmvscsi/ibmvfc.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/scsi/ibmvscsi/ibmvfc.c +2 -1
@@ -4926,7 +4926,8 @@ static void ibmvfc_discover_targets_done
 	switch (mad_status) {
 	case IBMVFC_MAD_SUCCESS:
 		ibmvfc_dbg(vhost, "Discover Targets succeeded\n");
-		vhost->num_targets = be32_to_cpu(rsp->num_written);
+		vhost->num_targets = min_t(u32, be32_to_cpu(rsp->num_written),
+					   max_targets);
 		ibmvfc_set_host_action(vhost, IBMVFC_HOST_ACTION_ALLOC_TGTS);
 		break;
 	case IBMVFC_MAD_FAILED:
author maintainer
#391 lore raw

PATCH 377/570 scsi: ses: Handle positive SCSI error from ses_recv_diag()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Greg Kroah-Hartman <[email protected]>

commit 7a9f448d44127217fabc4065c5ba070d4e0b5d37 upstream.

ses_recv_diag() can return a positive value, which also means that an error happened, so do not only test for negative values.

---
 drivers/scsi/ses.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/scsi/ses.c +1 -1
@@ -184,7 +184,7 @@ static unsigned char *ses_get_page2_desc
 	unsigned char *type_ptr = ses_dev->page1_types;
 	unsigned char *desc_ptr = ses_dev->page2 + 8;
 
-	if (ses_recv_diag(sdev, 2, ses_dev->page2, ses_dev->page2_len) < 0)
+	if (ses_recv_diag(sdev, 2, ses_dev->page2, ses_dev->page2_len))
 		return NULL;
 
 	for (i = 0; i < ses_dev->page1_num_types; i++, type_ptr += 4) {
author maintainer
#392 lore raw

PATCH 378/570 jbd2: gracefully abort on checkpointing state corruptions

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Milos Nikic <[email protected]>

commit bac3190a8e79beff6ed221975e0c9b1b5f2a21da upstream.

This patch targets two internal state machine invariants in checkpoint.c residing inside functions that natively return integer error codes.

- In jbd2_cleanup_journal_tail(): A blocknr of 0 indicates a severely corrupted journal superblock. Replaced the J_ASSERT with a WARN_ON_ONCE and a graceful journal abort, returning -EFSCORRUPTED.

- In jbd2_log_do_checkpoint(): Replaced the J_ASSERT_BH checking for an unexpected buffer_jwrite state. If the warning triggers, we explicitly drop the just-taken get_bh() reference and call __flush_batch() to safely clean up any previously queued buffers in the j_chkpt_bhs array, preventing a memory leak before returning -EFSCORRUPTED.

---
 fs/jbd2/checkpoint.c |   15 +++++++++++++--
 1 file changed, 13 insertions(+), 2 deletions(-)
fs/jbd2/checkpoint.c +13 -2
@@ -279,7 +279,15 @@ restart:
 			 */
 			BUFFER_TRACE(bh, "queue");
 			get_bh(bh);
-			J_ASSERT_BH(bh, !buffer_jwrite(bh));
+			if (WARN_ON_ONCE(buffer_jwrite(bh))) {
+				put_bh(bh); /* drop the ref we just took */
+				spin_unlock(&journal->j_list_lock);
+				/* Clean up any previously batched buffers */
+				if (batch_count)
+					__flush_batch(journal, &batch_count);
+				jbd2_journal_abort(journal, -EFSCORRUPTED);
+				return -EFSCORRUPTED;
+			}
 			journal->j_chkpt_bhs[batch_count++] = bh;
 			transaction->t_chp_stats.cs_written++;
 			transaction->t_checkpoint_list = jh->b_cpnext;
@@ -337,7 +345,10 @@ int jbd2_cleanup_journal_tail(journal_t
 
 	if (!jbd2_journal_get_log_tail(journal, &first_tid, &blocknr))
 		return 1;
-	J_ASSERT(blocknr != 0);
+	if (WARN_ON_ONCE(blocknr == 0)) {
+		jbd2_journal_abort(journal, -EFSCORRUPTED);
+		return -EFSCORRUPTED;
+	}
 
 	/*
 	 * We need to make sure that any blocks that were recently written out
author maintainer
#393 lore raw

PATCH 379/570 xfs: stop reclaim before pushing AIL during unmount

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yuto Ohnuki <[email protected]>

commit 4f24a767e3d64a5f58c595b5c29b6063a201f1e3 upstream.

The unmount sequence in xfs_unmount_flush_inodes() pushed the AIL while background reclaim and inodegc are still running. This is broken independently of any use-after-free issues - background reclaim and inodegc should not be running while the AIL is being pushed during unmount, as inodegc can dirty and insert inodes into the AIL during the flush, and background reclaim can race to abort and free dirty inodes.

Reorder xfs_unmount_flush_inodes() to stop inodegc and cancel background reclaim before pushing the AIL. Stop inodegc before cancelling m_reclaim_work because the inodegc worker can re-queue m_reclaim_work via xfs_inodegc_set_reclaimable.

---
 fs/xfs/xfs_mount.c |    7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)
fs/xfs/xfs_mount.c +4 -3
@@ -535,8 +535,9 @@ xfs_check_summary_counts(
  * have been retrying in the background.  This will prevent never-ending
  * retries in AIL pushing from hanging the unmount.
  *
- * Finally, we can push the AIL to clean all the remaining dirty objects, then
- * reclaim the remaining inodes that are still in memory at this point in time.
+ * Stop inodegc and background reclaim before pushing the AIL so that they
+ * are not running while the AIL is being flushed. Then push the AIL to
+ * clean all the remaining dirty objects and reclaim the remaining inodes.
  */
 static void
 xfs_unmount_flush_inodes(
@@ -548,9 +549,9 @@ xfs_unmount_flush_inodes(
 
 	set_bit(XFS_OPSTATE_UNMOUNTING, &mp->m_opstate);
 
-	xfs_ail_push_all_sync(mp->m_ail);
 	xfs_inodegc_stop(mp);
 	cancel_delayed_work_sync(&mp->m_reclaim_work);
+	xfs_ail_push_all_sync(mp->m_ail);
 	xfs_reclaim_inodes(mp);
 	xfs_health_unmount(mp);
 }
author maintainer
#394 lore raw

PATCH 380/570 ext4: convert inline data to extents when truncate exceeds inline size

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Deepanshu Kartikey <[email protected]>

commit ed9356a30e59c7cc3198e7fc46cfedf3767b9b17 upstream.

Add a check in ext4_setattr() to convert files from inline data storage to extent-based storage when truncate() grows the file size beyond the inline capacity. This prevents the filesystem from entering an inconsistent state where the inline data flag is set but the file size exceeds what can be stored inline.

Without this fix, the following sequence causes a kernel BUG_ON():

1. Mount filesystem with inode that has inline flag set and small size
2. truncate(file, 50MB) - grows size but inline flag remains set
3. sendfile() attempts to write data
4. ext4_write_inline_data() hits BUG_ON(write_size > inline_capacity)

The crash occurs because ext4_write_inline_data() expects inline storage to accommodate the write, but the actual inline capacity (~60 bytes for i_block + ~96 bytes for xattrs) is far smaller than the file size and write request.

The fix checks if the new size from setattr exceeds the inode's actual inline capacity (EXT4_I(inode)->i_inline_size) and converts the file to extent-based storage before proceeding with the size change.

This addresses the root cause by ensuring the inline data flag and file size remain consistent during truncate operations.

---
 fs/ext4/inode.c |   12 ++++++++++++
 1 file changed, 12 insertions(+)
fs/ext4/inode.c +12 -0
@@ -5507,6 +5507,18 @@ int ext4_setattr(struct user_namespace *
 		if (IS_I_VERSION(inode) && attr->ia_size != inode->i_size)
 			inode_inc_iversion(inode);
 
+		/*
+		 * If file has inline data but new size exceeds inline capacity,
+		 * convert to extent-based storage first to prevent inconsistent
+		 * state (inline flag set but size exceeds inline capacity).
+		 */
+		if (ext4_has_inline_data(inode) &&
+		    attr->ia_size > EXT4_I(inode)->i_inline_size) {
+			error = ext4_convert_inline_data(inode);
+			if (error)
+				goto err_out;
+		}
+
 		if (shrink) {
 			if (ext4_should_order_data(inode)) {
 				error = ext4_begin_ordered_truncate(inode,
author maintainer
#395 lore raw

PATCH 381/570 ext4: make recently_deleted() properly work with lazy itable initialization

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jan Kara <[email protected]>

commit bd060afa7cc3e0ad30afa9ecc544a78638498555 upstream.

recently_deleted() checks whether inode has been used in the near past. However this can give false positive result when inode table is not initialized yet and we are in fact comparing to random garbage (or stale itable block of a filesystem before mkfs). Ultimately this results in uninitialized inodes being skipped during inode allocation and possibly they are never initialized and thus e2fsck complains. Verify if the inode has been initialized before checking for dtime.

---
 fs/ext4/ialloc.c |    6 ++++++
 1 file changed, 6 insertions(+)
fs/ext4/ialloc.c +6 -0
@@ -688,6 +688,12 @@ static int recently_deleted(struct super
 	if (unlikely(!gdp))
 		return 0;
 
+	/* Inode was never used in this filesystem? */
+	if (ext4_has_group_desc_csum(sb) &&
+	    (gdp->bg_flags & cpu_to_le16(EXT4_BG_INODE_UNINIT) ||
+	     ino >= EXT4_INODES_PER_GROUP(sb) - ext4_itable_unused_count(sb, gdp)))
+		return 0;
+
 	bh = sb_find_get_block(sb, ext4_inode_table(sb, gdp) +
 		       (ino / inodes_per_block));
 	if (!bh || !buffer_uptodate(bh))
author maintainer
#396 lore raw

PATCH 382/570 ext4: avoid allocate block from corrupted group in ext4_mb_find_by_goal()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Ye Bin <[email protected]>

commit 46066e3a06647c5b186cc6334409722622d05c44 upstream.

There's issue as follows:
...
EXT4-fs (mmcblk0p1): Delayed block allocation failed for inode 206 at logical offset 0 with max blocks 1 with error 117 EXT4-fs (mmcblk0p1): This should not happen!! Data will be lost

EXT4-fs (mmcblk0p1): Delayed block allocation failed for inode 206 at logical offset 0 with max blocks 1 with error 117 EXT4-fs (mmcblk0p1): This should not happen!! Data will be lost

EXT4-fs (mmcblk0p1): Delayed block allocation failed for inode 206 at logical offset 0 with max blocks 1 with error 117 EXT4-fs (mmcblk0p1): This should not happen!! Data will be lost

EXT4-fs (mmcblk0p1): Delayed block allocation failed for inode 206 at logical offset 0 with max blocks 1 with error 117 EXT4-fs (mmcblk0p1): This should not happen!! Data will be lost

EXT4-fs (mmcblk0p1): Delayed block allocation failed for inode 2243 at logical offset 0 with max blocks 1 with error 117 EXT4-fs (mmcblk0p1): This should not happen!! Data will be lost

EXT4-fs (mmcblk0p1): Delayed block allocation failed for inode 2239 at logical offset 0 with max blocks 1 with error 117 EXT4-fs (mmcblk0p1): This should not happen!! Data will be lost

EXT4-fs (mmcblk0p1): error count since last fsck: 1
EXT4-fs (mmcblk0p1): initial error at time 1765597433: ext4_mb_generate_buddy:760 EXT4-fs (mmcblk0p1): last error at time 1765597433: ext4_mb_generate_buddy:760
...

According to the log analysis, blocks are always requested from the corrupted block group. This may happen as follows:
ext4_mb_find_by_goal
ext4_mb_load_buddy
ext4_mb_load_buddy_gfp
ext4_mb_init_cache
ext4_read_block_bitmap_nowait
ext4_wait_block_bitmap
ext4_validate_block_bitmap
if (!grp || EXT4_MB_GRP_BBITMAP_CORRUPT(grp))
return -EFSCORRUPTED; // There's no logs.
if (err)
return err; // Will return error
ext4_lock_group(ac->ac_sb, group);
if (unlikely(EXT4_MB_GRP_BBITMAP_CORRUPT(e4b->bd_info))) // Unreachable
goto out;

After commit 9008a58e5dce ("ext4: make the bitmap read routines return real error codes") merged, Commit 163a203ddb36 ("ext4: mark block group as corrupt on block bitmap error") is no real solution for allocating blocks from corrupted block groups. This is because if 'EXT4_MB_GRP_BBITMAP_CORRUPT(e4b->bd_info)' is true, then 'ext4_mb_load_buddy()' may return an error. This means that the block allocation will fail.
Therefore, check block group if corrupted when ext4_mb_load_buddy() returns error.

Fixes 163a203ddb36 ("ext4: mark block group as corrupt on block bitmap error")
Fixes 9008a58e5dce ("ext4: make the bitmap read routines return real error codes")
---
 fs/ext4/mballoc.c |    6 +++++-
 1 file changed, 5 insertions(+), 1 deletion(-)
fs/ext4/mballoc.c +5 -1
@@ -2280,8 +2280,12 @@ int ext4_mb_find_by_goal(struct ext4_all
 		return 0;
 
 	err = ext4_mb_load_buddy(ac->ac_sb, group, e4b);
-	if (err)
+	if (err) {
+		if (EXT4_MB_GRP_BBITMAP_CORRUPT(e4b->bd_info) &&
+		    !(ac->ac_flags & EXT4_MB_HINT_GOAL_ONLY))
+			return 0;
 		return err;
+	}
 
 	ext4_lock_group(ac->ac_sb, group);
 	if (unlikely(EXT4_MB_GRP_BBITMAP_CORRUPT(e4b->bd_info)))
author maintainer
#397 lore raw

PATCH 383/570 ext4: reject mount if bigalloc with s_first_data_block != 0

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Helen Koike <[email protected]>

commit 3822743dc20386d9897e999dbb990befa3a5b3f8 upstream.

bigalloc with s_first_data_block != 0 is not supported, reject mounting it.

---
 fs/ext4/super.c |    7 +++++++
 1 file changed, 7 insertions(+)
fs/ext4/super.c +7 -0
@@ -3214,6 +3214,13 @@ int ext4_feature_set_ok(struct super_blo
 			 "extents feature\n");
 		return 0;
 	}
+	if (ext4_has_feature_bigalloc(sb) &&
+	    le32_to_cpu(EXT4_SB(sb)->s_es->s_first_data_block)) {
+		ext4_msg(sb, KERN_WARNING,
+			 "bad geometry: bigalloc file system with non-zero "
+			 "first_data_block\n");
+		return 0;
+	}
 
 #if !IS_ENABLED(CONFIG_QUOTA) || !IS_ENABLED(CONFIG_QFMT_V2)
 	if (!readonly && (ext4_has_feature_quota(sb) ||
author maintainer
#398 lore raw

PATCH 384/570 ext4: fix iloc.bh leak in ext4_fc_replay_inode() error paths

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Baokun Li <[email protected]>

commit ec0a7500d8eace5b4f305fa0c594dd148f0e8d29 upstream.

During code review, Joseph found that ext4_fc_replay_inode() calls ext4_get_fc_inode_loc() to get the inode location, which holds a reference to iloc.bh that must be released via brelse().

However, several error paths jump to the 'out' label without releasing iloc.bh:

- ext4_handle_dirty_metadata() failure
- sync_dirty_buffer() failure
- ext4_mark_inode_used() failure
- ext4_iget() failure

Fix this by introducing an 'out_brelse' label placed just before the existing 'out' label to ensure iloc.bh is always released.

Additionally, make ext4_fc_replay_inode() propagate errors properly instead of always returning 0.

---
 fs/ext4/fast_commit.c |   13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)
fs/ext4/fast_commit.c +8 -5
@@ -1582,19 +1582,21 @@ static int ext4_fc_replay_inode(struct s
 	/* Immediately update the inode on disk. */
 	ret = ext4_handle_dirty_metadata(NULL, NULL, iloc.bh);
 	if (ret)
-		goto out;
+		goto out_brelse;
 	ret = sync_dirty_buffer(iloc.bh);
 	if (ret)
-		goto out;
+		goto out_brelse;
 	ret = ext4_mark_inode_used(sb, ino);
 	if (ret)
-		goto out;
+		goto out_brelse;
 
 	/* Given that we just wrote the inode on disk, this SHOULD succeed. */
 	inode = ext4_iget(sb, ino, EXT4_IGET_NORMAL);
 	if (IS_ERR(inode)) {
 		ext4_debug("Inode not found.");
-		return -EFSCORRUPTED;
+		inode = NULL;
+		ret = -EFSCORRUPTED;
+		goto out_brelse;
 	}
 
 	/*
@@ -1610,13 +1612,14 @@ static int ext4_fc_replay_inode(struct s
 	ext4_inode_csum_set(inode, ext4_raw_inode(&iloc), EXT4_I(inode));
 	ret = ext4_handle_dirty_metadata(NULL, NULL, iloc.bh);
 	sync_dirty_buffer(iloc.bh);
+out_brelse:
 	brelse(iloc.bh);
 out:
 	iput(inode);
 	if (!ret)
 		blkdev_issue_flush(sb->s_bdev);
 
-	return 0;
+	return ret;
 }
 
 /*
author maintainer
#399 lore raw

PATCH 385/570 ext4: always drain queued discard work in ext4_mb_release()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Theodore Ts'o <[email protected]>

commit 9ee29d20aab228adfb02ca93f87fb53c56c2f3af upstream.

While reviewing recent ext4 patch[1], Sashiko raised the following concern[2]:

Quoted 5 lines

If the filesystem is initially mounted with the discard option, deleting files will populate sbi->s_discard_list and queue s_discard_work. If it is then remounted with nodiscard, the EXT4_MOUNT_DISCARD flag is cleared, but the pending s_discard_work is neither cancelled nor flushed.

[1] https://lore.kernel.org/r/[email protected]/
[2] https://sashiko.dev/#/patchset/20260319094545.19291-1-qiang.zhang%40linux.dev

The concern was valid, but it had nothing to do with the patch[1]. One of the problems with Sashiko in its current (early) form is that it will detect pre-existing issues and report it as a problem with the patch that it is reviewing.

In practice, it would be hard to hit deliberately (unless you are a malicious syzkaller fuzzer), since it would involve mounting the file system with -o discard, and then deleting a large number of files, remounting the file system with -o nodiscard, and then immediately unmounting the file system before the queued discard work has a change to drain on its own.

Fix it because it's a real bug, and to avoid Sashiko from raising this concern when analyzing future patches to mballoc.c.

Fixes 55cdd0af2bc5 ("ext4: get discard out of jbd2 commit kthread contex")
---
 fs/ext4/mballoc.c |   12 +++++-------
 1 file changed, 5 insertions(+), 7 deletions(-)
fs/ext4/mballoc.c +5 -7
@@ -3628,13 +3628,11 @@ int ext4_mb_release(struct super_block *
 	struct kmem_cache *cachep = get_groupinfo_cache(sb->s_blocksize_bits);
 	int count;
 
-	if (test_opt(sb, DISCARD)) {
-		/*
-		 * wait the discard work to drain all of ext4_free_data
-		 */
-		flush_work(&sbi->s_discard_work);
-		WARN_ON_ONCE(!list_empty(&sbi->s_discard_list));
-	}
+	/*
+	 * wait the discard work to drain all of ext4_free_data
+	 */
+	flush_work(&sbi->s_discard_work);
+	WARN_ON_ONCE(!list_empty(&sbi->s_discard_list));
 
 	if (sbi->s_group_info) {
 		for (i = 0; i < ngroups; i++) {
author maintainer
#400 lore raw

PATCH 386/570 dmaengine: idxd: Fix not releasing workqueue on .release()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vinicius Costa Gomes <[email protected]>

[ Upstream commit 3d33de353b1ff9023d5ec73b9becf80ea87af695 ]

The workqueue associated with an DSA/IAA device is not released when the object is freed.

---
 drivers/dma/idxd/sysfs.c | 1 +
 1 file changed, 1 insertion(+)
drivers/dma/idxd/sysfs.c +1 -0
index 489a9d8850764..ee208dfdd0cb5 100644
@@ -1271,6 +1271,7 @@ static void idxd_conf_device_release(struct device *dev)
 {
 	struct idxd_device *idxd = confdev_to_idxd(dev);
 
+	destroy_workqueue(idxd->wq);
 	kfree(idxd->groups);
 	kfree(idxd->wqs);
 	kfree(idxd->engines);
Signature · 1 line
2.53.0
#401 lore raw

Re: [PATCH 5.15 386/570] dmaengine: idxd: Fix not releasing workqueue on .release()

Hi,

Quoted 26 lines · Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vinicius Costa Gomes <[email protected]>

[ Upstream commit 3d33de353b1ff9023d5ec73b9becf80ea87af695 ]

The workqueue associated with an DSA/IAA device is not released when the object is freed.

---
  drivers/dma/idxd/sysfs.c | 1 +
  1 file changed, 1 insertion(+)
drivers/dma/idxd/sysfs.c +0 -0
index 489a9d8850764..ee208dfdd0cb5 100644

drivers/dma/idxd/sysfs.c static void idxd_conf_device_release(struct device *dev) · line 1274

Quoted 4 lines
drivers/dma/idxd/sysfs.c +1 -0
@@ -1271,6 +1271,7 @@ static void idxd_conf_device_release(struct device *dev)
  {
  	struct idxd_device *idxd = confdev_to_idxd(dev);
 
+	destroy_workqueue(idxd->wq);

I have run an AI assisted backport review and it spotted an issue: I have taken a look and the issues goes like:

in 5.15.y code base:

the remove() function is something like this:

  get_device(idxd_confdev(idxd));
  device_unregister(idxd_confdev(idxd));
  ...
  destroy_workqueue(idxd->wq);
  ...
  put_device(idxd_confdev(idxd));

put_device() --> release() call and this patch tries to add destroy_workqueue(idxd->wq); so that's destroying it twice.

idxd_remove()
-> destroy_workqueue(idxd->wq) // first destroy
-> put_device(idxd_confdev(idxd))

      -> idxd_conf_device_release()
          -> destroy_workqueue(idxd->wq) // second destroy

This changed in ustream during the refactoring commit and followup fixes of commit: a409e919ca32 ("dmaengine: idxd: Refactor remove call with idxd_cleanup() helper"), so upstream (6.15-rc1 origin, it is backported to 6.6-rc1) doesn't have this problem. looks like we will have same problem in 6.1.y as well. I just checked, it has the same problem

Thanks,
Harshit

Quoted 3 lines
drivers/dma/idxd/sysfs.c +0 -0
  	kfree(idxd->groups);
  	kfree(idxd->wqs);
  	kfree(idxd->engines);
maintainer
#402 lore raw

Re: [PATCH 5.15 386/570] dmaengine: idxd: Fix not releasing workqueue on .release()

Quoted 5 lines

idxd_remove()
-> destroy_workqueue(idxd->wq) // first destroy
-> put_device(idxd_confdev(idxd))

       -> idxd_conf_device_release()
           -> destroy_workqueue(idxd->wq) // second destroy

Makes sense. I'll drop it from 5.15. Thanks!

--
Thanks,
Sasha

author maintainer
#403 lore raw

PATCH 387/570 phy: ti: j721e-wiz: Fix device node reference leak in wiz_get_lane_phy_types()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Felix Gu <[email protected]>

[ Upstream commit 584b457f4166293bdfa50f930228e9fb91a38392 ]

The serdes device_node is obtained using of_get_child_by_name(), which increments the reference count. However, it is never put, leading to a reference leak.

Add the missing of_node_put() calls to ensure the reference count is properly balanced.

Fixes 7ae14cf581f2 ("phy: ti: j721e-wiz: Implement DisplayPort mode to the wiz driver")
---
 drivers/phy/ti/phy-j721e-wiz.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/phy/ti/phy-j721e-wiz.c +2 -0
index 8963fbf7aa73b..a3908a579115c 100644
@@ -1116,6 +1116,7 @@ static int wiz_get_lane_phy_types(struct device *dev, struct wiz *wiz)
 			dev_err(dev,
 				"%s: Reading \"reg\" from \"%s\" failed: %d\n",
 				__func__, subnode->name, ret);
+			of_node_put(serdes);
 			return ret;
 		}
 		of_property_read_u32(subnode, "cdns,num-lanes", &num_lanes);
@@ -1128,6 +1129,7 @@ static int wiz_get_lane_phy_types(struct device *dev, struct wiz *wiz)
 			wiz->lane_phy_type[i] = phy_type;
 	}
 
+	of_node_put(serdes);
 	return 0;
 }
 
Signature · 1 line
2.53.0
author maintainer
#404 lore raw

PATCH 388/570 dmaengine: xilinx: xilinx_dma: Fix dma_device directions

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Marek Vasut <[email protected]>

[ Upstream commit e9cc95397bb7da13fe8a5b53a2f23cfaf9018ade ]

Unlike chan->direction , struct dma_device .directions field is a bitfield. Turn chan->direction into a bitfield to make it compatible with struct dma_device .directions .

Fixes 7e01511443c3 ("dmaengine: xilinx_dma: Set dma_device directions")
---
 drivers/dma/xilinx/xilinx_dma.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/dma/xilinx/xilinx_dma.c +1 -1
index ba5850ca39ddd..e2adc33631ca3 100644
@@ -2862,7 +2862,7 @@ static int xilinx_dma_chan_probe(struct xilinx_dma_device *xdev,
 		return -EINVAL;
 	}
 
-	xdev->common.directions |= chan->direction;
+	xdev->common.directions |= BIT(chan->direction);
 
 	/* Request the interrupt */
 	chan->irq = irq_of_parse_and_map(node, chan->tdest);
Signature · 1 line
2.53.0
author maintainer
#405 lore raw

PATCH 389/570 dmaengine: xilinx: xilinx_dma: Fix residue calculation for cyclic DMA

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Marek Vasut <[email protected]>

[ Upstream commit f61d145999d61948a23cd436ebbfa4c3b9ab8987 ]

The cyclic DMA calculation is currently entirely broken and reports residue only for the first segment. The problem is twofold.

First, when the first descriptor finishes, it is moved from active_list to done_list, but it is never returned back into the active_list. The xilinx_dma_tx_status() expects the descriptor to be in the active_list to report any meaningful residue information, which never happens after the first descriptor finishes. Fix this up in xilinx_dma_start_transfer() and if the descriptor is cyclic, lift it from done_list and place it back into active_list list.

Second, the segment .status fields of the descriptor remain dirty. Once the DMA did one pass on the descriptor, the .status fields are populated with data by the DMA, but the .status fields are not cleared before reuse during the next cyclic DMA round. The xilinx_dma_get_residue() recognizes that as if the descriptor was complete and had 0 residue, which is bogus. Reinitialize the status field before placing the descriptor back into the active_list.

Fixes c0bba3a99f07 ("dmaengine: vdma: Add Support for Xilinx AXI Direct Memory Access Engine")
---
 drivers/dma/xilinx/xilinx_dma.c | 23 ++++++++++++++++++++++-
 1 file changed, 22 insertions(+), 1 deletion(-)
drivers/dma/xilinx/xilinx_dma.c +22 -1
index e2adc33631ca3..948093a47d9f9 100644
@@ -1513,8 +1513,29 @@ static void xilinx_dma_start_transfer(struct xilinx_dma_chan *chan)
 	if (chan->err)
 		return;
 
-	if (list_empty(&chan->pending_list))
+	if (list_empty(&chan->pending_list)) {
+		if (chan->cyclic) {
+			struct xilinx_dma_tx_descriptor *desc;
+			struct list_head *entry;
+
+			desc = list_last_entry(&chan->done_list,
+					       struct xilinx_dma_tx_descriptor, node);
+			list_for_each(entry, &desc->segments) {
+				struct xilinx_axidma_tx_segment *axidma_seg;
+				struct xilinx_axidma_desc_hw *axidma_hw;
+				axidma_seg = list_entry(entry,
+							struct xilinx_axidma_tx_segment,
+							node);
+				axidma_hw = &axidma_seg->hw;
+				axidma_hw->status = 0;
+			}
+
+			list_splice_tail_init(&chan->done_list, &chan->active_list);
+			chan->desc_pendingcount = 0;
+			chan->idle = false;
+		}
 		return;
+	}
 
 	if (!chan->idle)
 		return;
Signature · 1 line
2.53.0
author maintainer
#406 lore raw

PATCH 390/570 dmaengine: xilinx: xilinx_dma: Fix unmasked residue subtraction

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Marek Vasut <[email protected]>

[ Upstream commit c7d812e33f3e8ca0fa9eeabf71d1c7bc3acedc09 ]

The segment .control and .status fields both contain top bits which are not part of the buffer size, the buffer size is located only in the bottom max_buffer_len bits. To avoid interference from those top bits, mask out the size using max_buffer_len first, and only then subtract the values.

Fixes a575d0b4e663 ("dmaengine: xilinx_dma: Introduce xilinx_dma_get_residue")
---
 drivers/dma/xilinx/xilinx_dma.c | 12 ++++++------
 1 file changed, 6 insertions(+), 6 deletions(-)
drivers/dma/xilinx/xilinx_dma.c +6 -6
index 948093a47d9f9..00971b867ade2 100644
@@ -964,16 +964,16 @@ static u32 xilinx_dma_get_residue(struct xilinx_dma_chan *chan,
 					      struct xilinx_cdma_tx_segment,
 					      node);
 			cdma_hw = &cdma_seg->hw;
-			residue += (cdma_hw->control - cdma_hw->status) &
-				   chan->xdev->max_buffer_len;
+			residue += (cdma_hw->control & chan->xdev->max_buffer_len) -
+			           (cdma_hw->status & chan->xdev->max_buffer_len);
 		} else if (chan->xdev->dma_config->dmatype ==
 			   XDMA_TYPE_AXIDMA) {
 			axidma_seg = list_entry(entry,
 						struct xilinx_axidma_tx_segment,
 						node);
 			axidma_hw = &axidma_seg->hw;
-			residue += (axidma_hw->control - axidma_hw->status) &
-				   chan->xdev->max_buffer_len;
+			residue += (axidma_hw->control & chan->xdev->max_buffer_len) -
+			           (axidma_hw->status & chan->xdev->max_buffer_len);
 		} else {
 			aximcdma_seg =
 				list_entry(entry,
@@ -981,8 +981,8 @@ static u32 xilinx_dma_get_residue(struct xilinx_dma_chan *chan,
 					   node);
 			aximcdma_hw = &aximcdma_seg->hw;
 			residue +=
-				(aximcdma_hw->control - aximcdma_hw->status) &
-				chan->xdev->max_buffer_len;
+				(aximcdma_hw->control & chan->xdev->max_buffer_len) -
+				(aximcdma_hw->status & chan->xdev->max_buffer_len);
 		}
 	}
 
Signature · 1 line
2.53.0
author maintainer
#407 lore raw

PATCH 391/570 btrfs: fix super block offset in error message in btrfs_validate_super()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Mark Harmstone <[email protected]>

[ Upstream commit b52fe51f724385b3ed81e37e510a4a33107e8161 ]

Fix the superblock offset mismatch error message in btrfs_validate_super(): we changed it so that it considers all the superblocks, but the message still assumes we're only looking at the first one.

The change from %u to %llu is because we're changing from a constant to a u64.

Fixes 069ec957c35e ("btrfs: Refactor btrfs_check_super_valid")
---
 fs/btrfs/disk-io.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
fs/btrfs/disk-io.c +2 -2
index 136902f27e441..41cc27ba4355a 100644
@@ -2657,8 +2657,8 @@ int btrfs_validate_super(struct btrfs_fs_info *fs_info,
 
 	if (mirror_num >= 0 &&
 	    btrfs_super_bytenr(sb) != btrfs_sb_offset(mirror_num)) {
-		btrfs_err(fs_info, "super offset mismatch %llu != %u",
-			  btrfs_super_bytenr(sb), BTRFS_SUPER_INFO_OFFSET);
+		btrfs_err(fs_info, "super offset mismatch %llu != %llu",
+			  btrfs_super_bytenr(sb), btrfs_sb_offset(mirror_num));
 		ret = -EINVAL;
 	}
 
Signature · 1 line
2.53.0
author maintainer
#408 lore raw

PATCH 392/570 btrfs: fix lost error when running device stats on multiple devices fs

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Filipe Manana <[email protected]>

[ Upstream commit 1c37d896b12dfd0d4c96e310b0033c6676933917 ]

Whenever we get an error updating the device stats item for a device in btrfs_run_dev_stats() we allow the loop to go to the next device, and if updating the stats item for the next device succeeds, we end up losing the error we had from the previous device.

Fix this by breaking out of the loop once we get an error and make sure it's returned to the caller. Since we are in the transaction commit path (and in the critical section actually), returning the error will result in a transaction abort.

Fixes 733f4fbbc108 ("Btrfs: read device stats on mount, write modified ones during commit")
---
 fs/btrfs/volumes.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)
fs/btrfs/volumes.c +3 -2
index 839ee01827b26..9ab226814cfde 100644
@@ -8016,8 +8016,9 @@ int btrfs_run_dev_stats(struct btrfs_trans_handle *trans)
 		smp_rmb();
 
 		ret = update_dev_stat_item(trans, device);
-		if (!ret)
-			atomic_sub(stats_cnt, &device->dev_stats_ccnt);
+		if (ret)
+			break;
+		atomic_sub(stats_cnt, &device->dev_stats_ccnt);
 	}
 	mutex_unlock(&fs_devices->device_list_mutex);
 
Signature · 1 line
2.53.0
author maintainer
#409 lore raw

PATCH 393/570 dmaengine: idxd: Remove usage of the deprecated ida_simple_xx() API

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Christophe JAILLET <[email protected]>

[ Upstream commit 1075ee66a8c19bfa375b19c236fd6a22a867f138 ]

ida_alloc() and ida_free() should be preferred to the deprecated ida_simple_get() and ida_simple_remove().

This is less verbose.

Note that the upper limit of ida_simple_get() is exclusive, but the one of ida_alloc_range() is inclusive. Sothis change allows one more device.

MINORMASK is ((1U << MINORBITS) - 1), so allowing MINORMASK as a maximum value makes sense. It is also consistent with other "ida_.*MINORMASK" and "ida_*MINOR()" usages.

Stable-dep-of: c311f5e92484 ("dmaengine: idxd: Fix freeing the allocated ida too late")

---
 drivers/dma/idxd/cdev.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/dma/idxd/cdev.c +2 -2
index 033df43db0cec..8172c3f1f782e 100644
@@ -46,7 +46,7 @@ static void idxd_cdev_dev_release(struct device *dev)
 	struct idxd_wq *wq = idxd_cdev->wq;
 
 	cdev_ctx = &ictx[wq->idxd->data->type];
-	ida_simple_remove(&cdev_ctx->minor_ida, idxd_cdev->minor);
+	ida_free(&cdev_ctx->minor_ida, idxd_cdev->minor);
 	kfree(idxd_cdev);
 }
 
@@ -260,7 +260,7 @@ int idxd_wq_add_cdev(struct idxd_wq *wq)
 	cdev = &idxd_cdev->cdev;
 	dev = cdev_dev(idxd_cdev);
 	cdev_ctx = &ictx[wq->idxd->data->type];
-	minor = ida_simple_get(&cdev_ctx->minor_ida, 0, MINORMASK, GFP_KERNEL);
+	minor = ida_alloc_max(&cdev_ctx->minor_ida, MINORMASK, GFP_KERNEL);
 	if (minor < 0) {
 		kfree(idxd_cdev);
 		return minor;
Signature · 1 line
2.53.0
author maintainer
#410 lore raw

PATCH 394/570 dmaengine: idxd: Fix freeing the allocated ida too late

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Vinicius Costa Gomes <[email protected]>

[ Upstream commit c311f5e9248471a950f0a524c2fd736414d98900 ]

It can happen that when the cdev .release() is called, the driver already called ida_destroy(). Move ida_free() to the _del() path.

We see with DEBUG_KOBJECT_RELEASE enabled and forcing an early PCI unbind.

---
 drivers/dma/idxd/cdev.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/dma/idxd/cdev.c +4 -4
index 8172c3f1f782e..005eef4df2162 100644
@@ -42,11 +42,7 @@ struct idxd_user_context {
 static void idxd_cdev_dev_release(struct device *dev)
 {
 	struct idxd_cdev *idxd_cdev = dev_to_cdev(dev);
-	struct idxd_cdev_context *cdev_ctx;
-	struct idxd_wq *wq = idxd_cdev->wq;
 
-	cdev_ctx = &ictx[wq->idxd->data->type];
-	ida_free(&cdev_ctx->minor_ida, idxd_cdev->minor);
 	kfree(idxd_cdev);
 }
 
@@ -295,11 +291,15 @@ int idxd_wq_add_cdev(struct idxd_wq *wq)
 
 void idxd_wq_del_cdev(struct idxd_wq *wq)
 {
+	struct idxd_cdev_context *cdev_ctx;
 	struct idxd_cdev *idxd_cdev;
 
 	idxd_cdev = wq->idxd_cdev;
 	wq->idxd_cdev = NULL;
 	cdev_device_del(&idxd_cdev->cdev, cdev_dev(idxd_cdev));
+
+	cdev_ctx = &ictx[wq->idxd->data->type];
+	ida_free(&cdev_ctx->minor_ida, idxd_cdev->minor);
 	put_device(cdev_dev(idxd_cdev));
 }
 
Signature · 1 line
2.53.0
author maintainer
#411 lore raw

PATCH 395/570 dmaengine: xilinx_dma: Program interrupt delay timeout

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Radhey Shyam Pandey <[email protected]>

[ Upstream commit 84b798fedf3fa8f0ab0c096593ba817abc454fe5 ]

Program IRQDelay for AXI DMA. The interrupt timeout mechanism causes the DMA engine to generate an interrupt after the delay time period has expired. It enables dmaengine to respond in real-time even though interrupt coalescing is configured. It also remove the placeholder for delay interrupt and merge it with frame completion interrupt. Since by default interrupt delay timeout is disabled this feature addition has no functional impact on VDMA, MCDMA and CDMA IP's.

Stable-dep-of: a17ce4bc6f4f ("dmaengine: xilinx_dma: Fix reset related timeout with two-channel AXIDMA")

---
 drivers/dma/xilinx/xilinx_dma.c | 20 +++++++++++---------
 1 file changed, 11 insertions(+), 9 deletions(-)
drivers/dma/xilinx/xilinx_dma.c +11 -9
index 00971b867ade2..f1162f7c9a52c 100644
@@ -174,8 +174,10 @@
 #define XILINX_DMA_MAX_TRANS_LEN_MAX	23
 #define XILINX_DMA_V2_MAX_TRANS_LEN_MAX	26
 #define XILINX_DMA_CR_COALESCE_MAX	GENMASK(23, 16)
+#define XILINX_DMA_CR_DELAY_MAX		GENMASK(31, 24)
 #define XILINX_DMA_CR_CYCLIC_BD_EN_MASK	BIT(4)
 #define XILINX_DMA_CR_COALESCE_SHIFT	16
+#define XILINX_DMA_CR_DELAY_SHIFT	24
 #define XILINX_DMA_BD_SOP		BIT(27)
 #define XILINX_DMA_BD_EOP		BIT(26)
 #define XILINX_DMA_COALESCE_MAX		255
@@ -411,6 +413,7 @@ struct xilinx_dma_tx_descriptor {
  * @stop_transfer: Differentiate b/w DMA IP's quiesce
  * @tdest: TDEST value for mcdma
  * @has_vflip: S2MM vertical flip
+ * @irq_delay: Interrupt delay timeout
  */
 struct xilinx_dma_chan {
 	struct xilinx_dma_device *xdev;
@@ -449,6 +452,7 @@ struct xilinx_dma_chan {
 	int (*stop_transfer)(struct xilinx_dma_chan *chan);
 	u16 tdest;
 	bool has_vflip;
+	u8 irq_delay;
 };
 
 /**
@@ -1559,6 +1563,9 @@ static void xilinx_dma_start_transfer(struct xilinx_dma_chan *chan)
 	if (chan->has_sg)
 		xilinx_write(chan, XILINX_DMA_REG_CURDESC,
 			     head_desc->async_tx.phys);
+	reg  &= ~XILINX_DMA_CR_DELAY_MAX;
+	reg  |= chan->irq_delay << XILINX_DMA_CR_DELAY_SHIFT;
+	dma_ctrl_write(chan, XILINX_DMA_REG_DMACR, reg);
 
 	xilinx_dma_start(chan);
 
@@ -1886,15 +1893,8 @@ static irqreturn_t xilinx_dma_irq_handler(int irq, void *data)
 		}
 	}
 
-	if (status & XILINX_DMA_DMASR_DLY_CNT_IRQ) {
-		/*
-		 * Device takes too long to do the transfer when user requires
-		 * responsiveness.
-		 */
-		dev_dbg(chan->dev, "Inter-packet latency too long\n");
-	}
-
-	if (status & XILINX_DMA_DMASR_FRM_CNT_IRQ) {
+	if (status & (XILINX_DMA_DMASR_FRM_CNT_IRQ |
+		      XILINX_DMA_DMASR_DLY_CNT_IRQ)) {
 		spin_lock(&chan->lock);
 		xilinx_dma_complete_descriptor(chan);
 		chan->idle = true;
@@ -2818,6 +2818,8 @@ static int xilinx_dma_chan_probe(struct xilinx_dma_device *xdev,
 	/* Retrieve the channel properties from the device tree */
 	has_dre = of_property_read_bool(node, "xlnx,include-dre");
 
+	of_property_read_u8(node, "xlnx,irq-delay", &chan->irq_delay);
+
 	chan->genlock = of_property_read_bool(node, "xlnx,genlock-mode");
 
 	err = of_property_read_u32(node, "xlnx,datawidth", &value);
Signature · 1 line
2.53.0
author maintainer
#412 lore raw

PATCH 396/570 dmaengine: xilinx_dma: Fix reset related timeout with two-channel AXIDMA

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Tomi Valkeinen <[email protected]>

[ Upstream commit a17ce4bc6f4f9acf77ba416c36791a15602e53aa ]

A single AXIDMA controller can have one or two channels. When it has two channels, the reset for both are tied together: resetting one channel resets the other as well. This creates a problem where resetting one channel will reset the registers for both channels, including clearing interrupt enable bits for the other channel, which can then lead to timeouts as the driver is waiting for an interrupt which never comes.

The driver currently has a probe-time work around for this: when a channel is created, the driver also resets and enables the interrupts. With two channels the reset for the second channel will clear the interrupt enables for the first one. The work around in the driver is just to manually enable the interrupts again in xilinx_dma_alloc_chan_resources().

This workaround only addresses the probe-time issue. When channels are reset at runtime (e.g., in xilinx_dma_terminate_all() or during error recovery), there's no corresponding mechanism to restore the other channel's interrupt enables. This leads to one channel having its interrupts disabled while the driver expects them to work, causing timeouts and DMA failures.

A proper fix is a complicated matter, as we should not reset the other channel when it's operating normally. So, perhaps, there should be some kind of synchronization for a common reset, which is not trivial to implement. To add to the complexity, the driver also supports other DMA types, like VDMA, CDMA and MCDMA, which don't have a shared reset.

However, when the two-channel AXIDMA is used in the (assumably) normal use case, providing DMA for a single memory-to-memory device, the common reset is a bit smaller issue: when something bad happens on one channel, or when one channel is terminated, the assumption is that we also want to terminate the other channel. And thus resetting both at the same time is "ok".

With that line of thinking we can implement a bit better work around than just the current probe time work around: let's enable the AXIDMA interrupts at xilinx_dma_start_transfer() instead. This ensures interrupts are enabled whenever a transfer starts, regardless of any prior resets that may have cleared them.

This approach is also more logical: enable interrupts only when needed for a transfer, rather than at resource allocation time, and, I think, all the other DMA types should also use this model, but I'm reluctant to do such changes as I cannot test them.

The reset function still enables interrupts even though it's not needed for AXIDMA anymore, but it's common code for all DMA types (VDMA, CDMA, MCDMA), so leave it unchanged to avoid affecting other variants.

Fixes c0bba3a99f07 ("dmaengine: vdma: Add Support for Xilinx AXI Direct Memory Access Engine")
---
 drivers/dma/xilinx/xilinx_dma.c | 9 +--------
 1 file changed, 1 insertion(+), 8 deletions(-)
drivers/dma/xilinx/xilinx_dma.c +1 -8
index f1162f7c9a52c..7a596eaba4660 100644
@@ -1190,14 +1190,6 @@ static int xilinx_dma_alloc_chan_resources(struct dma_chan *dchan)
 
 	dma_cookie_init(dchan);
 
-	if (chan->xdev->dma_config->dmatype == XDMA_TYPE_AXIDMA) {
-		/* For AXI DMA resetting once channel will reset the
-		 * other channel as well so enable the interrupts here.
-		 */
-		dma_ctrl_set(chan, XILINX_DMA_REG_DMACR,
-			      XILINX_DMA_DMAXR_ALL_IRQ_MASK);
-	}
-
 	if ((chan->xdev->dma_config->dmatype == XDMA_TYPE_CDMA) && chan->has_sg)
 		dma_ctrl_set(chan, XILINX_DMA_REG_DMACR,
 			     XILINX_CDMA_CR_SGMODE);
@@ -1565,6 +1557,7 @@ static void xilinx_dma_start_transfer(struct xilinx_dma_chan *chan)
 			     head_desc->async_tx.phys);
 	reg  &= ~XILINX_DMA_CR_DELAY_MAX;
 	reg  |= chan->irq_delay << XILINX_DMA_CR_DELAY_SHIFT;
+	reg |= XILINX_DMA_DMAXR_ALL_IRQ_MASK;
 	dma_ctrl_write(chan, XILINX_DMA_REG_DMACR, reg);
 
 	xilinx_dma_start(chan);
Signature · 1 line
2.53.0
author maintainer
#413 lore raw

PATCH 397/570 futex: Clear stale exiting pointer in futex_lock_pi() retry path

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Davidlohr Bueso <[email protected]>

commit 210d36d892de5195e6766c45519dfb1e65f3eb83 upstream.

Fuzzying/stressing futexes triggered:

    WARNING: kernel/futex/core.c:825 at wait_for_owner_exiting+0x7a/0x80, CPU#11: futex_lock_pi_s/524

When futex_lock_pi_atomic() sees the owner is exiting, it returns -EBUSY and stores a refcounted task pointer in 'exiting'.

After wait_for_owner_exiting() consumes that reference, the local pointer is never reset to nil. Upon a retry, if futex_lock_pi_atomic() returns a different error, the bogus pointer is passed to wait_for_owner_exiting().

CPU0 CPU1 CPU2
futex_lock_pi(uaddr)
// acquires the PI futex
exit()

    futex_cleanup_begin()
      futex_state = EXITING;
			     futex_lock_pi(uaddr)
			       futex_lock_pi_atomic()
				 attach_to_pi_owner()
				   // observes EXITING
				   *exiting = owner;  // takes ref
				   return -EBUSY
			       wait_for_owner_exiting(-EBUSY, owner)
				 put_task_struct();   // drops ref
			       // exiting still points to owner
			       goto retry;
			       futex_lock_pi_atomic()
				 lock_pi_update_atomic()
				   cmpxchg(uaddr)
					*uaddr ^= WAITERS // whatever
				   // value changed
				 return -EAGAIN;
			       wait_for_owner_exiting(-EAGAIN, exiting) // stale
				 WARN_ON_ONCE(exiting)

Fix this by resetting upon retry, essentially aligning it with requeue_pi.

---
 kernel/futex/core.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
kernel/futex/core.c +2 -1
@@ -3029,9 +3029,9 @@ static int futex_lock_pi(u32 __user *uad
 			 ktime_t *time, int trylock)
 {
 	struct hrtimer_sleeper timeout, *to;
-	struct task_struct *exiting = NULL;
 	struct rt_mutex_waiter rt_waiter;
 	struct futex_hash_bucket *hb;
+	struct task_struct *exiting;
 	struct futex_q q = futex_q_init;
 	int res, ret;
 
@@ -3044,6 +3044,7 @@ static int futex_lock_pi(u32 __user *uad
 	to = futex_setup_timer(time, &timeout, flags, 0);
 
 retry:
+	exiting = NULL;
 	ret = get_futex_key(uaddr, flags & FLAGS_SHARED, &q.key, FUTEX_WRITE);
 	if (unlikely(ret != 0))
 		goto out;
author maintainer
#414 lore raw

PATCH 398/570 HID: wacom: fix out-of-bounds read in wacom_intuos_bt_irq

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Benoît Sevens <[email protected]>

[ Upstream commit 2f1763f62909ccb6386ac50350fa0abbf5bb16a9 ]

The wacom_intuos_bt_irq() function processes Bluetooth HID reports without sufficient bounds checking. A maliciously crafted short report can trigger an out-of-bounds read when copying data into the wacom structure.

Specifically, report 0x03 requires at least 22 bytes to safely read the processed data and battery status, while report 0x04 (which falls through to 0x03) requires 32 bytes.

Add explicit length checks for these report IDs and log a warning if a short report is received.

---
 drivers/hid/wacom_wac.c | 10 ++++++++++
 1 file changed, 10 insertions(+)
drivers/hid/wacom_wac.c +10 -0
index 0ab473f372ad0..0cc979d99b3d8 100644
@@ -1258,10 +1258,20 @@ static int wacom_intuos_bt_irq(struct wacom_wac *wacom, size_t len)
 
 	switch (data[0]) {
 	case 0x04:
+		if (len < 32) {
+			dev_warn(wacom->pen_input->dev.parent,
+				 "Report 0x04 too short: %zu bytes\n", len);
+			break;
+		}
 		wacom_intuos_bt_process_data(wacom, data + i);
 		i += 10;
 		fallthrough;
 	case 0x03:
+		if (i == 1 && len < 22) {
+			dev_warn(wacom->pen_input->dev.parent,
+				 "Report 0x03 too short: %zu bytes\n", len);
+			break;
+		}
 		wacom_intuos_bt_process_data(wacom, data + i);
 		i += 10;
 		wacom_intuos_bt_process_data(wacom, data + i);
Signature · 1 line
2.53.0
author maintainer
#415 lore raw

PATCH 399/570 atm: lec: fix use-after-free in sock_def_readable()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Deepanshu Kartikey <[email protected]>

[ Upstream commit 922814879542c2e397b0e9641fd36b8202a8e555 ]

A race condition exists between lec_atm_close() setting priv->lecd to NULL and concurrent access to priv->lecd in send_to_lecd(), lec_handle_bridge(), and lec_atm_send(). When the socket is freed via RCU while another thread is still using it, a use-after-free occurs in sock_def_readable() when accessing the socket's wait queue.

The root cause is that lec_atm_close() clears priv->lecd without any synchronization, while callers dereference priv->lecd without any protection against concurrent teardown.

Fix this by converting priv->lecd to an RCU-protected pointer:
- Mark priv->lecd as __rcu in lec.h
- Use rcu_assign_pointer() in lec_atm_close() and lecd_attach() for safe pointer assignment
- Use rcu_access_pointer() for NULL checks that do not dereference the pointer in lec_start_xmit(), lec_push(), send_to_lecd() and lecd_attach()
- Use rcu_read_lock/rcu_dereference/rcu_read_unlock in send_to_lecd(), lec_handle_bridge() and lec_atm_send() to safely access lecd
- Use rcu_assign_pointer() followed by synchronize_rcu() in lec_atm_close() to ensure all readers have completed before proceeding. This is safe since lec_atm_close() is called from vcc_release() which holds lock_sock(), a sleeping lock.
- Remove the manual sk_receive_queue drain from lec_atm_close() since vcc_destroy_socket() already drains it after lec_atm_close() returns.

v2: Switch from spinlock + sock_hold/put approach to RCU to properly
fix the race. The v1 spinlock approach had two issues pointed out by Eric Dumazet:
1. priv->lecd was still accessed directly after releasing the lock instead of using a local copy.
2. The spinlock did not prevent packets being queued after lec_atm_close() drains sk_receive_queue since timer and workqueue paths bypass netif_stop_queue().

Note: Syzbot patch testing was attempted but the test VM terminated
unexpectedly with "Connection to localhost closed by remote host", likely due to a QEMU AHCI emulation issue unrelated to this fix. Compile testing with "make W=1 net/atm/lec.o" passes cleanly.

---
 net/atm/lec.c |   72 +++++++++++++++++++++++++++++++++++++---------------------
 net/atm/lec.h |    2 -
 2 files changed, 48 insertions(+), 26 deletions(-)
net/atm/lec.c +47 -25
@@ -142,6 +142,7 @@ static void lec_handle_bridge(struct sk_
 		struct sock *sk;
 		struct sk_buff *skb2;
 		struct atmlec_msg *mesg;
+		struct atm_vcc *vcc;
 
 		skb2 = alloc_skb(sizeof(struct atmlec_msg), GFP_ATOMIC);
 		if (skb2 == NULL)
@@ -154,10 +155,18 @@ static void lec_handle_bridge(struct sk_
 					/* 0x01 is topology change */
 
 		priv = netdev_priv(dev);
-		atm_force_charge(priv->lecd, skb2->truesize);
-		sk = sk_atm(priv->lecd);
-		skb_queue_tail(&sk->sk_receive_queue, skb2);
-		sk->sk_data_ready(sk);
+
+		rcu_read_lock();
+		vcc = rcu_dereference(priv->lecd);
+		if (vcc) {
+			atm_force_charge(vcc, skb2->truesize);
+			sk = sk_atm(vcc);
+			skb_queue_tail(&sk->sk_receive_queue, skb2);
+			sk->sk_data_ready(sk);
+		} else {
+			dev_kfree_skb(skb2);
+		}
+		rcu_read_unlock();
 	}
 }
 #endif /* IS_ENABLED(CONFIG_BRIDGE) */
@@ -216,7 +225,7 @@ static netdev_tx_t lec_start_xmit(struct
 	int is_rdesc;
 
 	pr_debug("called\n");
-	if (!priv->lecd) {
+	if (!rcu_access_pointer(priv->lecd)) {
 		pr_info("%s:No lecd attached\n", dev->name);
 		dev->stats.tx_errors++;
 		netif_stop_queue(dev);
@@ -443,6 +452,7 @@ static int lec_atm_send(struct atm_vcc *
 			/* hit from bridge table, send LE_ARP_RESPONSE */
 			struct sk_buff *skb2;
 			struct sock *sk;
+			struct atm_vcc *vcc;
 
 			pr_debug("%s: entry found, responding to zeppelin\n",
 				 dev->name);
@@ -451,10 +461,18 @@ static int lec_atm_send(struct atm_vcc *
 				break;
 			skb2->len = sizeof(struct atmlec_msg);
 			skb_copy_to_linear_data(skb2, mesg, sizeof(*mesg));
-			atm_force_charge(priv->lecd, skb2->truesize);
-			sk = sk_atm(priv->lecd);
-			skb_queue_tail(&sk->sk_receive_queue, skb2);
-			sk->sk_data_ready(sk);
+
+			rcu_read_lock();
+			vcc = rcu_dereference(priv->lecd);
+			if (vcc) {
+				atm_force_charge(vcc, skb2->truesize);
+				sk = sk_atm(vcc);
+				skb_queue_tail(&sk->sk_receive_queue, skb2);
+				sk->sk_data_ready(sk);
+			} else {
+				dev_kfree_skb(skb2);
+			}
+			rcu_read_unlock();
 		}
 	}
 #endif /* IS_ENABLED(CONFIG_BRIDGE) */
@@ -470,23 +488,16 @@ static int lec_atm_send(struct atm_vcc *
 
 static void lec_atm_close(struct atm_vcc *vcc)
 {
-	struct sk_buff *skb;
 	struct net_device *dev = (struct net_device *)vcc->proto_data;
 	struct lec_priv *priv = netdev_priv(dev);
 
-	priv->lecd = NULL;
+	rcu_assign_pointer(priv->lecd, NULL);
+	synchronize_rcu();
 	/* Do something needful? */
 
 	netif_stop_queue(dev);
 	lec_arp_destroy(priv);
 
-	if (skb_peek(&sk_atm(vcc)->sk_receive_queue))
-		pr_info("%s closing with messages pending\n", dev->name);
-	while ((skb = skb_dequeue(&sk_atm(vcc)->sk_receive_queue))) {
-		atm_return(vcc, skb->truesize);
-		dev_kfree_skb(skb);
-	}
-
 	pr_info("%s: Shut down!\n", dev->name);
 	module_put(THIS_MODULE);
 }
@@ -512,12 +523,14 @@ send_to_lecd(struct lec_priv *priv, atml
 	     const unsigned char *mac_addr, const unsigned char *atm_addr,
 	     struct sk_buff *data)
 {
+	struct atm_vcc *vcc;
 	struct sock *sk;
 	struct sk_buff *skb;
 	struct atmlec_msg *mesg;
 
-	if (!priv || !priv->lecd)
+	if (!priv || !rcu_access_pointer(priv->lecd))
 		return -1;
+
 	skb = alloc_skb(sizeof(struct atmlec_msg), GFP_ATOMIC);
 	if (!skb)
 		return -1;
@@ -534,18 +547,27 @@ send_to_lecd(struct lec_priv *priv, atml
 	if (atm_addr)
 		memcpy(&mesg->content.normal.atm_addr, atm_addr, ATM_ESA_LEN);
 
-	atm_force_charge(priv->lecd, skb->truesize);
-	sk = sk_atm(priv->lecd);
+	rcu_read_lock();
+	vcc = rcu_dereference(priv->lecd);
+	if (!vcc) {
+		rcu_read_unlock();
+		kfree_skb(skb);
+		return -1;
+	}
+
+	atm_force_charge(vcc, skb->truesize);
+	sk = sk_atm(vcc);
 	skb_queue_tail(&sk->sk_receive_queue, skb);
 	sk->sk_data_ready(sk);
 
 	if (data != NULL) {
 		pr_debug("about to send %d bytes of data\n", data->len);
-		atm_force_charge(priv->lecd, data->truesize);
+		atm_force_charge(vcc, data->truesize);
 		skb_queue_tail(&sk->sk_receive_queue, data);
 		sk->sk_data_ready(sk);
 	}
 
+	rcu_read_unlock();
 	return 0;
 }
 
@@ -620,7 +642,7 @@ static void lec_push(struct atm_vcc *vcc
 
 		atm_return(vcc, skb->truesize);
 		if (*(__be16 *) skb->data == htons(priv->lecid) ||
-		    !priv->lecd || !(dev->flags & IFF_UP)) {
+		    !rcu_access_pointer(priv->lecd) || !(dev->flags & IFF_UP)) {
 			/*
 			 * Probably looping back, or if lecd is missing,
 			 * lecd has gone down
@@ -755,12 +777,12 @@ static int lecd_attach(struct atm_vcc *v
 		priv = netdev_priv(dev_lec[i]);
 	} else {
 		priv = netdev_priv(dev_lec[i]);
-		if (priv->lecd)
+		if (rcu_access_pointer(priv->lecd))
 			return -EADDRINUSE;
 	}
 	lec_arp_init(priv);
 	priv->itfnum = i;	/* LANE2 addition */
-	priv->lecd = vcc;
+	rcu_assign_pointer(priv->lecd, vcc);
 	vcc->dev = &lecatm_dev;
 	vcc_insert_socket(sk_atm(vcc));
 
net/atm/lec.h +1 -1
@@ -91,7 +91,7 @@ struct lec_priv {
 						 */
 	spinlock_t lec_arp_lock;
 	struct atm_vcc *mcast_vcc;		/* Default Multicast Send VCC */
-	struct atm_vcc *lecd;
+	struct atm_vcc __rcu *lecd;
 	struct delayed_work lec_arp_work;	/* C10 */
 	unsigned int maximum_unknown_frame_count;
 						/*
author maintainer
#416 lore raw

PATCH 400/570 btrfs: dont take device_list_mutex when querying zone info

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Johannes Thumshirn <[email protected]>

[ Upstream commit 77603ab10429fe713a03345553ca8dbbfb1d91c6 ]

Shin'ichiro reported sporadic hangs when running generic/013 in our CI system. When enabling lockdep, there is a lockdep splat when calling btrfs_get_dev_zone_info_all_devices() in the mount path that can be triggered by i.e. generic/013:

======================================================
WARNING: possible circular locking dependency detected
7.0.0-rc1+ #355 Not tainted
------------------------------------------------------
mount/1043 is trying to acquire lock:
ffff8881020b5470 (&vblk->vdev_mutex){+.+.}-{4:4}, at: virtblk_report_zones+0xda/0x430

but task is already holding lock:
ffff888102a738e0 (&fs_devs->device_list_mutex){+.+.}-{4:4}, at: btrfs_get_dev_zone_info_all_devices+0x45/0x90

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #4 (&fs_devs->device_list_mutex){+.+.}-{4:4}:

	 __mutex_lock+0xa3/0x1360
	 btrfs_create_pending_block_groups+0x1f4/0x9d0
	 __btrfs_end_transaction+0x3e/0x2e0
	 btrfs_zoned_reserve_data_reloc_bg+0x2f8/0x390
	 open_ctree+0x1934/0x23db
	 btrfs_get_tree.cold+0x105/0x26c
	 vfs_get_tree+0x28/0xb0
	 __do_sys_fsconfig+0x324/0x680
	 do_syscall_64+0x92/0x4f0
	 entry_SYSCALL_64_after_hwframe+0x76/0x7e

-> #3 (btrfs_trans_num_extwriters){++++}-{0:0}:

	 join_transaction+0xc2/0x5c0
	 start_transaction+0x17c/0xbc0
	 btrfs_zoned_reserve_data_reloc_bg+0x2b4/0x390
	 open_ctree+0x1934/0x23db
	 btrfs_get_tree.cold+0x105/0x26c
	 vfs_get_tree+0x28/0xb0
	 __do_sys_fsconfig+0x324/0x680
	 do_syscall_64+0x92/0x4f0
	 entry_SYSCALL_64_after_hwframe+0x76/0x7e

-> #2 (btrfs_trans_num_writers){++++}-{0:0}:

	 lock_release+0x163/0x4b0
	 __btrfs_end_transaction+0x1c7/0x2e0
	 btrfs_dirty_inode+0x6f/0xd0
	 touch_atime+0xe5/0x2c0
	 btrfs_file_mmap_prepare+0x65/0x90
	 __mmap_region+0x4b9/0xf00
	 mmap_region+0xf7/0x120
	 do_mmap+0x43d/0x610
	 vm_mmap_pgoff+0xd6/0x190
	 ksys_mmap_pgoff+0x7e/0xc0
	 do_syscall_64+0x92/0x4f0
	 entry_SYSCALL_64_after_hwframe+0x76/0x7e

-> #1 (&mm->mmap_lock){++++}-{4:4}:

	 __might_fault+0x68/0xa0
	 _copy_to_user+0x22/0x70
	 blkdev_copy_zone_to_user+0x22/0x40
	 virtblk_report_zones+0x282/0x430
	 blkdev_report_zones_ioctl+0xfd/0x130
	 blkdev_ioctl+0x20f/0x2c0
	 __x64_sys_ioctl+0x86/0xd0
	 do_syscall_64+0x92/0x4f0
	 entry_SYSCALL_64_after_hwframe+0x76/0x7e

-> #0 (&vblk->vdev_mutex){+.+.}-{4:4}:

	 __lock_acquire+0x1522/0x2680
	 lock_acquire+0xd5/0x2f0
	 __mutex_lock+0xa3/0x1360
	 virtblk_report_zones+0xda/0x430
	 blkdev_report_zones_cached+0x162/0x190
	 btrfs_get_dev_zones+0xdc/0x2e0
	 btrfs_get_dev_zone_info+0x219/0xe80
	 btrfs_get_dev_zone_info_all_devices+0x62/0x90
	 open_ctree+0x1200/0x23db
	 btrfs_get_tree.cold+0x105/0x26c
	 vfs_get_tree+0x28/0xb0
	 __do_sys_fsconfig+0x324/0x680
	 do_syscall_64+0x92/0x4f0
	 entry_SYSCALL_64_after_hwframe+0x76/0x7e

other info that might help us debug this:

  Chain exists of:
    &vblk->vdev_mutex --> btrfs_trans_num_extwriters --> &fs_devs->device_list_mutex

   Possible unsafe locking scenario:
	 CPU0                    CPU1
	 ----                    ----
    lock(&fs_devs->device_list_mutex);
				 lock(btrfs_trans_num_extwriters);
				 lock(&fs_devs->device_list_mutex);
    lock(&vblk->vdev_mutex);

*** DEADLOCK ***

3 locks held by mount/1043:
#0: ffff88811063e878 (&fc->uapi_mutex){+.+.}-{4:4}, at: __do_sys_fsconfig+0x2ae/0x680
#1: ffff88810cb9f0e8 (&type->s_umount_key#31/1){+.+.}-{4:4}, at: alloc_super+0xc0/0x3e0
#2: ffff888102a738e0 (&fs_devs->device_list_mutex){+.+.}-{4:4}, at: btrfs_get_dev_zone_info_all_devices+0x45/0x90

  stack backtrace:
  CPU: 2 UID: 0 PID: 1043 Comm: mount Not tainted 7.0.0-rc1+ #355 PREEMPT(full)
  Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.17.0-9.fc43 06/10/2025
  Call Trace:
   <TASK>
   dump_stack_lvl+0x5b/0x80
   print_circular_bug.cold+0x18d/0x1d8
   check_noncircular+0x10d/0x130
   __lock_acquire+0x1522/0x2680
   ? vmap_small_pages_range_noflush+0x3ef/0x820
   lock_acquire+0xd5/0x2f0
   ? virtblk_report_zones+0xda/0x430
   ? lock_is_held_type+0xcd/0x130
   __mutex_lock+0xa3/0x1360
   ? virtblk_report_zones+0xda/0x430
   ? virtblk_report_zones+0xda/0x430
   ? __pfx_copy_zone_info_cb+0x10/0x10
   ? virtblk_report_zones+0xda/0x430
   virtblk_report_zones+0xda/0x430
   ? __pfx_copy_zone_info_cb+0x10/0x10
   blkdev_report_zones_cached+0x162/0x190
   ? __pfx_copy_zone_info_cb+0x10/0x10
   btrfs_get_dev_zones+0xdc/0x2e0
   btrfs_get_dev_zone_info+0x219/0xe80
   btrfs_get_dev_zone_info_all_devices+0x62/0x90
   open_ctree+0x1200/0x23db
   btrfs_get_tree.cold+0x105/0x26c
   ? rcu_is_watching+0x18/0x50
   vfs_get_tree+0x28/0xb0
   __do_sys_fsconfig+0x324/0x680
   do_syscall_64+0x92/0x4f0
   entry_SYSCALL_64_after_hwframe+0x76/0x7e
  RIP: 0033:0x7f615e27a40e
  RSP: 002b:00007fff11b18fb8 EFLAGS: 00000246 ORIG_RAX: 00000000000001af
  RAX: ffffffffffffffda RBX: 000055572e92ab10 RCX: 00007f615e27a40e
  RDX: 0000000000000000 RSI: 0000000000000006 RDI: 0000000000000003
  RBP: 00007fff11b19100 R08: 0000000000000000 R09: 0000000000000000
  R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000
  R13: 000055572e92bc40 R14: 00007f615e3faa60 R15: 000055572e92bd08
   </TASK>

Don't hold the device_list_mutex while calling into btrfs_get_dev_zone_info() in btrfs_get_dev_zone_info_all_devices() to mitigate the issue. This is safe, as no other thread can touch the device list at the moment of execution.

---
 fs/btrfs/zoned.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
fs/btrfs/zoned.c +4 -2
index 8c858f31bdbc0..bff56f87b426a 100644
@@ -313,7 +313,10 @@ int btrfs_get_dev_zone_info_all_devices(struct btrfs_fs_info *fs_info)
 	if (!btrfs_fs_incompat(fs_info, ZONED))
 		return 0;
 
-	mutex_lock(&fs_devices->device_list_mutex);
+	/*
+	 * No need to take the device_list mutex here, we're still in the mount
+	 * path and devices cannot be added to or removed from the list yet.
+	 */
 	list_for_each_entry(device, &fs_devices->devices, dev_list) {
 		/* We can skip reading of zone info for missing devices */
 		if (!device->bdev)
@@ -323,7 +326,6 @@ int btrfs_get_dev_zone_info_all_devices(struct btrfs_fs_info *fs_info)
 		if (ret)
 			break;
 	}
-	mutex_unlock(&fs_devices->device_list_mutex);
 
 	return ret;
 }
Signature · 1 line
2.53.0
author maintainer
#417 lore raw

PATCH 401/570 objtool: Fix Clang jump table detection

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Josh Poimboeuf <[email protected]>

[ Upstream commit 4e5019216402ad0b4a84cff457b662d26803f103 ]

With Clang, there can be a conditional forward jump between the load of the jump table address and the indirect branch.

Fixes the following warning:

vmlinux.o: warning: objtool: ___bpf_prog_run+0x1c5: sibling call from callable instruction with modified stack frame

---
 tools/objtool/check.c | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)
tools/objtool/check.c +2 -3
index d2366ec61edc4..72a7e49dec276 100644
@@ -1759,12 +1759,11 @@ static void mark_func_jump_tables(struct objtool_file *file,
 			last = insn;
 
 		/*
-		 * Store back-pointers for unconditional forward jumps such
+		 * Store back-pointers for forward jumps such
 		 * that find_jump_table() can back-track using those and
 		 * avoid some potentially confusing code.
 		 */
-		if (insn->type == INSN_JUMP_UNCONDITIONAL && insn->jump_dest &&
-		    insn->offset > last->offset &&
+		if (insn->jump_dest &&
 		    insn->jump_dest->offset > insn->offset &&
 		    !insn->jump_dest->first_jump_src) {
 
Signature · 1 line
2.53.0
author maintainer
#418 lore raw

PATCH 402/570 HID: multitouch: Check to ensure report responses match the request

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Lee Jones <[email protected]>

[ Upstream commit e716edafedad4952fe3a4a273d2e039a84e8681a ]

It is possible for a malicious (or clumsy) device to respond to a specific report's feature request using a completely different report ID. This can cause confusion in the HID core resulting in nasty side-effects such as OOB writes.

Add a check to ensure that the report ID in the response, matches the one that was requested. If it doesn't, omit reporting the raw event and return early.

---
 drivers/hid/hid-multitouch.c | 7 +++++++
 1 file changed, 7 insertions(+)
drivers/hid/hid-multitouch.c +7 -0
index 30769b37aabe7..7a092a2a1bf00 100644
@@ -472,12 +472,19 @@ static void mt_get_feature(struct hid_device *hdev, struct hid_report *report)
 		dev_warn(&hdev->dev, "failed to fetch feature %d\n",
 			 report->id);
 	} else {
+		/* The report ID in the request and the response should match */
+		if (report->id != buf[0]) {
+			hid_err(hdev, "Returned feature report did not match the request\n");
+			goto free;
+		}
+
 		ret = hid_report_raw_event(hdev, HID_FEATURE_REPORT, buf,
 					   size, 0);
 		if (ret)
 			dev_warn(&hdev->dev, "failed to report feature\n");
 	}
 
+free:
 	kfree(buf);
 }
 
Signature · 1 line
2.53.0
author maintainer
#419 lore raw

PATCH 403/570 btrfs: reject root items with drop_progress and zero drop_level

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: ZhengYuan Huang <[email protected]>

[ Upstream commit b17b79ff896305fd74980a5f72afec370ee88ca4 ]

[BUG]
When recovering relocation at mount time, merge_reloc_root() and btrfs_drop_snapshot() both use BUG_ON(level == 0) to guard against an impossible state: a non-zero drop_progress combined with a zero drop_level in a root_item, which can be triggered:

------------[ cut here ]------------
kernel BUG at fs/btrfs/relocation.c:1545!
Oops: invalid opcode: 0000 [#1] SMP KASAN NOPTI
CPU: 1 UID: 0 PID: 283 ... Tainted: 6.18.0+ #16 PREEMPT(voluntary)
Tainted: [O]=OOT_MODULE, [E]=UNSIGNED_MODULE
Hardware name: QEMU Ubuntu 24.04 PC v2, BIOS 1.16.3-debian-1.16.3-2
RIP: 0010:merge_reloc_root+0x1266/0x1650 fs/btrfs/relocation.c:1545
Code: ffff0000 00004589 d7e9acfa ffffe8a1 79bafebe 02000000
Call Trace:
 merge_reloc_roots+0x295/0x890 fs/btrfs/relocation.c:1861
 btrfs_recover_relocation+0xd6e/0x11d0 fs/btrfs/relocation.c:4195
 btrfs_start_pre_rw_mount+0xa4d/0x1810 fs/btrfs/disk-io.c:3130
 open_ctree+0x5824/0x5fe0 fs/btrfs/disk-io.c:3640
 btrfs_fill_super fs/btrfs/super.c:987 [inline]
 btrfs_get_tree_super fs/btrfs/super.c:1951 [inline]
 btrfs_get_tree_subvol fs/btrfs/super.c:2094 [inline]
 btrfs_get_tree+0x111c/0x2190 fs/btrfs/super.c:2128
 vfs_get_tree+0x9a/0x370 fs/super.c:1758
 fc_mount fs/namespace.c:1199 [inline]
 do_new_mount_fc fs/namespace.c:3642 [inline]
 do_new_mount fs/namespace.c:3718 [inline]
 path_mount+0x5b8/0x1ea0 fs/namespace.c:4028
 do_mount fs/namespace.c:4041 [inline]
 __do_sys_mount fs/namespace.c:4229 [inline]
 __se_sys_mount fs/namespace.c:4206 [inline]
 __x64_sys_mount+0x282/0x320 fs/namespace.c:4206
 ...
RIP: 0033:0x7f969c9a8fde
Code: 0f1f4000 48c7c2b0 fffffff7 d8648902 b8ffffff ffc3660f
---[ end trace 0000000000000000 ]---

The bug is reproducible on 7.0.0-rc2-next-20260310 with our dynamic metadata fuzzing tool that corrupts btrfs metadata at runtime.

[CAUSE]
A non-zero drop_progress.objectid means an interrupted btrfs_drop_snapshot() left a resume point on disk, and in that case drop_level must be greater than 0 because the checkpoint is only saved at internal node levels.

Although this invariant is enforced when the kernel writes the root item, it is not validated when the root item is read back from disk. That allows on-disk corruption to provide an invalid state with drop_progress.objectid != 0 and drop_level == 0.

When relocation recovery later processes such a root item, merge_reloc_root() reads drop_level and hits BUG_ON(level == 0). The same invalid metadata can also trigger the corresponding BUG_ON() in btrfs_drop_snapshot().

[FIX]
Fix this by validating the root_item invariant in tree-checker when reading root items from disk: if drop_progress.objectid is non-zero, drop_level must also be non-zero. Reject such malformed metadata with -EUCLEAN before it reaches merge_reloc_root() or btrfs_drop_snapshot() and triggers the BUG_ON.

After the fix, the same corruption is correctly rejected by tree-checker and the BUG_ON is no longer triggered.

---
 fs/btrfs/tree-checker.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)
fs/btrfs/tree-checker.c +17 -0
index b0afa47032104..e370ad75072c7 100644
@@ -1187,6 +1187,23 @@ static int check_root_item(struct extent_buffer *leaf, struct btrfs_key *key,
 			    btrfs_root_drop_level(&ri), BTRFS_MAX_LEVEL - 1);
 		return -EUCLEAN;
 	}
+	/*
+	 * If drop_progress.objectid is non-zero, a btrfs_drop_snapshot() was
+	 * interrupted and the resume point was recorded in drop_progress and
+	 * drop_level.  In that case drop_level must be >= 1: level 0 is the
+	 * leaf level and drop_snapshot never saves a checkpoint there (it
+	 * only records checkpoints at internal node levels in DROP_REFERENCE
+	 * stage).  A zero drop_level combined with a non-zero drop_progress
+	 * objectid indicates on-disk corruption and would cause a BUG_ON in
+	 * merge_reloc_root() and btrfs_drop_snapshot() at mount time.
+	 */
+	if (unlikely(btrfs_disk_key_objectid(&ri.drop_progress) != 0 &&
+		     btrfs_root_drop_level(&ri) == 0)) {
+		generic_err(leaf, slot,
+			    "invalid root drop_level 0 with non-zero drop_progress objectid %llu",
+			    btrfs_disk_key_objectid(&ri.drop_progress));
+		return -EUCLEAN;
+	}
 
 	/* Flags check */
 	if (unlikely(btrfs_root_flags(&ri) & ~valid_root_flags)) {
Signature · 1 line
2.53.0
author maintainer
#420 lore raw

PATCH 404/570 dt-bindings: auxdisplay: ht16k33: Use unevaluatedProperties to fix common property warning

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Frank Li <[email protected]>

[ Upstream commit 398c0c8bbc8f5a9d2f43863275a427a9d3720b6f ]

Change additionalProperties to unevaluatedProperties because it refs to /schemas/input/matrix-keymap.yaml.

Fix below CHECK_DTBS warnings:
arch/arm/boot/dts/nxp/imx/imx6dl-victgo.dtb: keypad@70 (holtek,ht16k33): 'keypad,num-columns', 'keypad,num-rows' do not match any of the regexes: '^pinctrl-[0-9]+$'
from schema $id: http://devicetree.org/schemas/auxdisplay/holtek,ht16k33.yaml#

Fixes f12b457c6b25c ("dt-bindings: auxdisplay: ht16k33: Convert to json-schema")
---
 .../devicetree/bindings/auxdisplay/holtek,ht16k33.yaml          | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
Documentation/devicetree/bindings/auxdisplay/holtek,ht16k33.yaml +1 -1
index 64ffff4600260..3ee00bcfcf827 100644
@@ -46,7 +46,7 @@ required:
   - reg
   - refresh-rate-hz
 
-additionalProperties: false
+unevaluatedProperties: false
 
 examples:
   - |
Signature · 1 line
2.53.0
author maintainer
#421 lore raw

PATCH 405/570 crypto: af-alg - fix NULL pointer dereference in scatterwalk

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Norbert Szetei <[email protected]>

[ Upstream commit 62397b493e14107ae82d8b80938f293d95425bcb ]

The AF_ALG interface fails to unmark the end of a Scatter/Gather List (SGL) when chaining a new af_alg_tsgl structure. If a sendmsg() fills an SGL exactly to MAX_SGL_ENTS, the last entry is marked as the end. A subsequent sendmsg() allocates a new SGL and chains it, but fails to clear the end marker on the previous SGL's last data entry.

This causes the crypto scatterwalk to hit a premature end, returning NULL on sg_next() and leading to a kernel panic during dereference.

Fix this by explicitly unmarking the end of the previous SGL when performing sg_chain() in af_alg_alloc_tsgl().

Fixes 8ff590903d5f ("crypto: algif_skcipher - User-space interface for skcipher operations")
---
 crypto/af_alg.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
crypto/af_alg.c +3 -1
index 658d5c3c88b7b..631ee6a220d5b 100644
@@ -515,8 +515,10 @@ static int af_alg_alloc_tsgl(struct sock *sk)
 		sg_init_table(sgl->sg, MAX_SGL_ENTS + 1);
 		sgl->cur = 0;
 
-		if (sg)
+		if (sg) {
+			sg_unmark_end(sg + MAX_SGL_ENTS - 1);
 			sg_chain(sg, MAX_SGL_ENTS + 1, sgl->sg);
+		}
 
 		list_add_tail(&sgl->list, &ctx->tsgl_list);
 	}
Signature · 1 line
2.53.0
author maintainer
#422 lore raw

PATCH 406/570 net: qrtr: replace qrtr_tx_flow radix_tree with xarray to fix memory leak

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Jiayuan Chen <[email protected]>

[ Upstream commit 2428083101f6883f979cceffa76cd8440751ffe6 ]

__radix_tree_create() allocates and links intermediate nodes into the tree one by one. If a subsequent allocation fails, the already-linked nodes remain in the tree with no corresponding leaf entry. These orphaned internal nodes are never reclaimed because radix_tree_for_each_slot() only visits slots containing leaf values.

The radix_tree API is deprecated in favor of xarray. As suggested by Matthew Wilcox, migrate qrtr_tx_flow from radix_tree to xarray instead of fixing the radix_tree itself [1]. xarray properly handles cleanup of internal nodes — xa_destroy() frees all internal xarray nodes when the qrtr_node is released, preventing the leak.

[1] https://lore.kernel.org/all/[email protected]/T/

---
 net/qrtr/af_qrtr.c | 31 +++++++++++++------------------
 1 file changed, 13 insertions(+), 18 deletions(-)
net/qrtr/af_qrtr.c +13 -18
index c7a8260fa6ddb..431fd1f2b80c1 100644
@@ -116,7 +116,7 @@ static DEFINE_XARRAY_ALLOC(qrtr_ports);
  * @ep: endpoint
  * @ref: reference count for node
  * @nid: node id
- * @qrtr_tx_flow: tree of qrtr_tx_flow, keyed by node << 32 | port
+ * @qrtr_tx_flow: xarray of qrtr_tx_flow, keyed by node << 32 | port
  * @qrtr_tx_lock: lock for qrtr_tx_flow inserts
  * @rx_queue: receive queue
  * @item: list item for broadcast list
@@ -127,7 +127,7 @@ struct qrtr_node {
 	struct kref ref;
 	unsigned int nid;
 
-	struct radix_tree_root qrtr_tx_flow;
+	struct xarray qrtr_tx_flow;
 	struct mutex qrtr_tx_lock; /* for qrtr_tx_flow */
 
 	struct sk_buff_head rx_queue;
@@ -170,6 +170,7 @@ static void __qrtr_node_release(struct kref *kref)
 	struct qrtr_tx_flow *flow;
 	unsigned long flags;
 	void __rcu **slot;
+	unsigned long index;
 
 	spin_lock_irqsave(&qrtr_nodes_lock, flags);
 	/* If the node is a bridge for other nodes, there are possibly
@@ -187,11 +188,9 @@ static void __qrtr_node_release(struct kref *kref)
 	skb_queue_purge(&node->rx_queue);
 
 	/* Free tx flow counters */
-	radix_tree_for_each_slot(slot, &node->qrtr_tx_flow, &iter, 0) {
-		flow = *slot;
-		radix_tree_iter_delete(&node->qrtr_tx_flow, &iter, slot);
+	xa_for_each(&node->qrtr_tx_flow, index, flow)
 		kfree(flow);
-	}
+	xa_destroy(&node->qrtr_tx_flow);
 	kfree(node);
 }
 
@@ -226,9 +225,7 @@ static void qrtr_tx_resume(struct qrtr_node *node, struct sk_buff *skb)
 
 	key = remote_node << 32 | remote_port;
 
-	rcu_read_lock();
-	flow = radix_tree_lookup(&node->qrtr_tx_flow, key);
-	rcu_read_unlock();
+	flow = xa_load(&node->qrtr_tx_flow, key);
 	if (flow) {
 		spin_lock(&flow->resume_tx.lock);
 		flow->pending = 0;
@@ -267,12 +264,13 @@ static int qrtr_tx_wait(struct qrtr_node *node, int dest_node, int dest_port,
 		return 0;
 
 	mutex_lock(&node->qrtr_tx_lock);
-	flow = radix_tree_lookup(&node->qrtr_tx_flow, key);
+	flow = xa_load(&node->qrtr_tx_flow, key);
 	if (!flow) {
 		flow = kzalloc(sizeof(*flow), GFP_KERNEL);
 		if (flow) {
 			init_waitqueue_head(&flow->resume_tx);
-			if (radix_tree_insert(&node->qrtr_tx_flow, key, flow)) {
+			if (xa_err(xa_store(&node->qrtr_tx_flow, key, flow,
+					    GFP_KERNEL))) {
 				kfree(flow);
 				flow = NULL;
 			}
@@ -324,9 +322,7 @@ static void qrtr_tx_flow_failed(struct qrtr_node *node, int dest_node,
 	unsigned long key = (u64)dest_node << 32 | dest_port;
 	struct qrtr_tx_flow *flow;
 
-	rcu_read_lock();
-	flow = radix_tree_lookup(&node->qrtr_tx_flow, key);
-	rcu_read_unlock();
+	flow = xa_load(&node->qrtr_tx_flow, key);
 	if (flow) {
 		spin_lock_irq(&flow->resume_tx.lock);
 		flow->tx_failed = 1;
@@ -594,7 +590,7 @@ int qrtr_endpoint_register(struct qrtr_endpoint *ep, unsigned int nid)
 	node->nid = QRTR_EP_NID_AUTO;
 	node->ep = ep;
 
-	INIT_RADIX_TREE(&node->qrtr_tx_flow, GFP_KERNEL);
+	xa_init(&node->qrtr_tx_flow);
 	mutex_init(&node->qrtr_tx_lock);
 
 	qrtr_node_assign(node, nid);
@@ -622,6 +618,7 @@ void qrtr_endpoint_unregister(struct qrtr_endpoint *ep)
 	struct qrtr_tx_flow *flow;
 	struct sk_buff *skb;
 	unsigned long flags;
+	unsigned long index;
 	void __rcu **slot;
 
 	mutex_lock(&node->ep_lock);
@@ -644,10 +641,8 @@ void qrtr_endpoint_unregister(struct qrtr_endpoint *ep)
 
 	/* Wake up any transmitters waiting for resume-tx from the node */
 	mutex_lock(&node->qrtr_tx_lock);
-	radix_tree_for_each_slot(slot, &node->qrtr_tx_flow, &iter, 0) {
-		flow = *slot;
+	xa_for_each(&node->qrtr_tx_flow, index, flow)
 		wake_up_interruptible_all(&flow->resume_tx);
-	}
 	mutex_unlock(&node->qrtr_tx_lock);
 
 	qrtr_node_release(node);
Signature · 1 line
2.53.0
author maintainer
#423 lore raw

PATCH 407/570 net: ipv6: ndisc: fix ndisc_ra_useropt to initialize nduseropt_padX fields to zero to prevent an info-leak

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yochai Eisenrich <[email protected]>

[ Upstream commit ae05340ccaa9d347fe85415609e075545bec589f ]

When processing Router Advertisements with user options the kernel builds an RTM_NEWNDUSEROPT netlink message. The nduseroptmsg struct has three padding fields that are never zeroed and can leak kernel data

The fix is simple, just zeroes the padding fields.

Fixes 31910575a9de ("[IPv6]: Export userland ND options through netlink (RDNSS support)")
---
 net/ipv6/ndisc.c | 3 +++
 1 file changed, 3 insertions(+)
net/ipv6/ndisc.c +3 -0
index 1821c1aa97ad4..74e82982ecd08 100644
@@ -1155,6 +1155,9 @@ static void ndisc_ra_useropt(struct sk_buff *ra, struct nd_opt_hdr *opt)
 	ndmsg->nduseropt_icmp_type = icmp6h->icmp6_type;
 	ndmsg->nduseropt_icmp_code = icmp6h->icmp6_code;
 	ndmsg->nduseropt_opts_len = opt->nd_opt_len << 3;
+	ndmsg->nduseropt_pad1 = 0;
+	ndmsg->nduseropt_pad2 = 0;
+	ndmsg->nduseropt_pad3 = 0;
 
 	memcpy(ndmsg + 1, opt, opt->nd_opt_len << 3);
 
Signature · 1 line
2.53.0
author maintainer
#424 lore raw

PATCH 408/570 tg3: Fix race for querying speed/duplex

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Thomas Bogendoerfer <[email protected]>

[ Upstream commit bb417456c7814d1493d98b7dd9c040bf3ce3b4ed ]

When driver signals carrier up via netif_carrier_on() its internal link_up state isn't updated immediately. This leads to inconsistent speed/duplex in /proc/net/bonding/bondX where the speed and duplex is shown as unknown while ethtool shows correct values. Fix this by using netif_carrier_ok() for link checking in get_ksettings function.

Fixes 84421b99cedc ("tg3: Update link_up flag for phylib devices")
---
 drivers/net/ethernet/broadcom/tg3.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/broadcom/tg3.c +1 -1
index bd3b56c7aab8d..e18e58f8258e6 100644
@@ -12223,7 +12223,7 @@ static int tg3_get_link_ksettings(struct net_device *dev,
 	ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.advertising,
 						advertising);
 
-	if (netif_running(dev) && tp->link_up) {
+	if (netif_running(dev) && netif_carrier_ok(dev)) {
 		cmd->base.speed = tp->link_config.active_speed;
 		cmd->base.duplex = tp->link_config.active_duplex;
 		ethtool_convert_legacy_u32_to_link_mode(
Signature · 1 line
2.53.0
author maintainer
#425 lore raw

PATCH 409/570 ipv6: icmp: clear skb2->cb[] in ip6_err_gen_icmpv6_unreach()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit 86ab3e55673a7a49a841838776f1ab18d23a67b5 ]

Sashiko AI-review observed:

In ip6_err_gen_icmpv6_unreach(), the skb is an outer IPv4 ICMP error packet where its cb contains an IPv4 inet_skb_parm. When skb is cloned into skb2 and passed to icmp6_send(), it uses IP6CB(skb2).

IP6CB interprets the IPv4 inet_skb_parm as an inet6_skb_parm. The cipso offset in inet_skb_parm.opt directly overlaps with dsthao in inet6_skb_parm at offset 18.

If an attacker sends a forged ICMPv4 error with a CIPSO IP option, dsthao would be a non-zero offset. Inside icmp6_send(), mip6_addr_swap() is called and uses ipv6_find_tlv(skb, opt->dsthao, IPV6_TLV_HAO).

This would scan the inner, attacker-controlled IPv6 packet starting at that offset, potentially returning a fake TLV without checking if the remaining packet length can hold the full 18-byte struct ipv6_destopt_hao.

Could mip6_addr_swap() then perform a 16-byte swap that extends past the end of the packet data into skb_shared_info?

Should the cb array also be cleared in ip6_err_gen_icmpv6_unreach() and ip6ip6_err() to prevent this?

This patch implements the first suggestion.

I am not sure if ip6ip6_err() needs to be changed. A separate patch would be better anyway.

---
 net/ipv6/icmp.c | 3 +++
 1 file changed, 3 insertions(+)
net/ipv6/icmp.c +3 -0
index 8601c76f3cc93..6f053874de741 100644
@@ -674,6 +674,9 @@ int ip6_err_gen_icmpv6_unreach(struct sk_buff *skb, int nhs, int type,
 	if (!skb2)
 		return 1;
 
+	/* Remove debris left by IPv4 stack. */
+	memset(IP6CB(skb2), 0, sizeof(*IP6CB(skb2)));
+
 	skb_dst_drop(skb2);
 	skb_pull(skb2, nhs);
 	skb_reset_network_header(skb2);
Signature · 1 line
2.53.0
author maintainer
#426 lore raw

PATCH 410/570 ip6_tunnel: clear skb2->cb[] in ip4ip6_err()

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <[email protected]>

[ Upstream commit 2edfa31769a4add828a7e604b21cb82aaaa05925 ]

Oskar Kjos reported the following problem.

ip4ip6_err() calls icmp_send() on a cloned skb whose cb[] was written by the IPv6 receive path as struct inet6_skb_parm. icmp_send() passes IPCB(skb2) to __ip_options_echo(), which interprets that cb[] region as struct inet_skb_parm (IPv4). The layouts differ: inet6_skb_parm.nhoff at offset 14 overlaps inet_skb_parm.opt.rr, producing a non-zero rr value. __ip_options_echo() then reads optlen from attacker-controlled packet data at sptr[rr+1] and copies that many bytes into dopt->__data, a fixed 40-byte stack buffer (IP_OPTIONS_DATA_FIXED_SIZE).

To fix this we clear skb2->cb[], as suggested by Oskar Kjos.

Also add minimal IPv4 header validation (version == 4, ihl >= 5).

Fixes c4d3efafcc93 ("[IPV6] IP6TUNNEL: Add support to IPv4 over IPv6 tunnel.")
---
 net/ipv6/ip6_tunnel.c | 5 +++++
 1 file changed, 5 insertions(+)
net/ipv6/ip6_tunnel.c +5 -0
index 7c1b5d01f8203..53930c28b6946 100644
@@ -603,11 +603,16 @@ ip4ip6_err(struct sk_buff *skb, struct inet6_skb_parm *opt,
 	if (!skb2)
 		return 0;
 
+	/* Remove debris left by IPv6 stack. */
+	memset(IPCB(skb2), 0, sizeof(*IPCB(skb2)));
+
 	skb_dst_drop(skb2);
 
 	skb_pull(skb2, offset);
 	skb_reset_network_header(skb2);
 	eiph = ip_hdr(skb2);
+	if (eiph->version != 4 || eiph->ihl < 5)
+		goto out;
 
 	/* Try to guess incoming interface */
 	rt = ip_route_output_ports(dev_net(skb->dev), &fl4, NULL, eiph->saddr,
Signature · 1 line
2.53.0
author maintainer
#427 lore raw

PATCH 411/570 bridge: br_nd_send: linearize skb before parsing ND options

In reply to Greg Kroah-Hartman

5.15-stable review patch. If anyone has any objections, please let me know.

------------------

From: Yang Yang <[email protected]>

[ Upstream commit a01aee7cafc575bb82f5529e8734e7052f9b16ea ]

br_nd_send() parses neighbour discovery options from ns->opt[] and assumes that these options are in the linear part of request.

Its callers only guarantee that the ICMPv6 header and target address are available, so the option area can still be non-linear. Parsing ns->opt[] in that case can access data past the linear buffer.

Linearize request before option parsing and derive ns from the linear network header.

---
 net/bridge/br_arp_nd_proxy.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)
net/bridge/br_arp_nd_proxy.c +7 -5
index 3db1def4437b3..a44d14b94865e 100644
@@ -248,12 +248,12 @@ struct nd_msg *br_is_nd_neigh_msg(struct sk_buff *skb, struct nd_msg *msg)
 
 static void br_nd_send(struct net_bridge *br, struct net_bridge_port *p,
 		       struct sk_buff *request, struct neighbour *n,
-		       __be16 vlan_proto, u16 vlan_tci, struct nd_msg *ns)
+		       __be16 vlan_proto, u16 vlan_tci)
 {
 	struct net_device *dev = request->dev;
 	struct net_bridge_vlan_group *vg;
+	struct nd_msg *na, *ns;
 	struct sk_buff *reply;
-	struct nd_msg *na;
 	struct ipv6hdr *pip6;
 	int na_olen = 8; /* opt hdr + ETH_ALEN for target */
 	int ns_olen;
@@ -261,7 +261,7 @@ static void br_nd_send(struct net_bridge *br, struct net_bridge_port *p,
 	u8 *daddr;
 	u16 pvid;
 
-	if (!dev)
+	if (!dev || skb_linearize(request))
 		return;
 
 	len = LL_RESERVED_SPACE(dev) + sizeof(struct ipv6hdr) +
@@ -278,6 +278,8 @@ static void br_nd_send(struct net_bridge *br, struct net_bridge_port *p,
 	skb_set_mac_header(reply, 0);
 
 	daddr = eth_hdr(request)->h_source;
+	ns = (struct nd_msg *)(skb_network_header(request) +
+			       sizeof(struct ipv6hdr));
 
 	/* Do we need option processing ? */
 	ns_olen = request->len - (skb_network_offset(request) +
@@ -465,9 +467,9 @@ void br_do_suppress_nd(struct sk_buff *skb, struct net_bridge *br,
 				if (vid != 0)
 					br_nd_send(br, p, skb, n,
 						   skb->vlan_proto,
-						   skb_vlan_tag_get(skb), msg);
+						   skb_vlan_tag_get(skb));
 				else
-					br_nd_send(br, p, skb, n, 0, 0, msg);
+					br_nd_send(br, p, skb, n, 0, 0);
 				replied = true;
 			}
 
Signature · 1 line
2.53.0