5.10.181-rc1 review

From: Matthew Howell <[email protected]>

[ Upstream commit 14ee78d5932afeb710c8305196a676a715bfdea8 ]

Add support for Sealevel 7xxxC serial cards.

This patch:
* Adds IDs to recognize 7xxxC cards from Sealevel Systems.
* Updates exar_pci_probe() to set nr_ports to last two bytes of primary dev ID for these cards.

Stable-dep-of: 95d698869b40 ("serial: 8250_exar: Add support for USR298x PCI Modems")

---
 drivers/tty/serial/8250/8250_exar.c | 14 ++++++++++++++
 1 file changed, 14 insertions(+)

index 3d82902bfe5b9..2c9f721091125 100644
@@ -43,6 +43,12 @@
 #define PCI_DEVICE_ID_EXAR_XR17V4358		0x4358
 #define PCI_DEVICE_ID_EXAR_XR17V8358		0x8358
 
+#define PCI_DEVICE_ID_SEALEVEL_710xC		0x1001
+#define PCI_DEVICE_ID_SEALEVEL_720xC		0x1002
+#define PCI_DEVICE_ID_SEALEVEL_740xC		0x1004
+#define PCI_DEVICE_ID_SEALEVEL_780xC		0x1008
+#define PCI_DEVICE_ID_SEALEVEL_716xC		0x1010
+
 #define UART_EXAR_INT0		0x80
 #define UART_EXAR_8XMODE	0x88	/* 8X sampling rate select */
 #define UART_EXAR_SLEEP		0x8b	/* Sleep mode */

@@ -600,6 +606,8 @@ exar_pci_probe(struct pci_dev *pcidev, const struct pci_device_id *ent)
 		nr_ports = BIT(((pcidev->device & 0x38) >> 3) - 1);
 	else if (board->num_ports)
 		nr_ports = board->num_ports;
+	else if (pcidev->vendor == PCI_VENDOR_ID_SEALEVEL)
+		nr_ports = pcidev->device & 0xff;
 	else
 		nr_ports = pcidev->device & 0x0f;
 

@@ -826,6 +834,12 @@ static const struct pci_device_id exar_pci_tbl[] = {
 	EXAR_DEVICE(COMMTECH, 4224PCI335, pbn_fastcom335_4),
 	EXAR_DEVICE(COMMTECH, 2324PCI335, pbn_fastcom335_4),
 	EXAR_DEVICE(COMMTECH, 2328PCI335, pbn_fastcom335_8),
+
+	EXAR_DEVICE(SEALEVEL, 710xC, pbn_exar_XR17V35x),
+	EXAR_DEVICE(SEALEVEL, 720xC, pbn_exar_XR17V35x),
+	EXAR_DEVICE(SEALEVEL, 740xC, pbn_exar_XR17V35x),
+	EXAR_DEVICE(SEALEVEL, 780xC, pbn_exar_XR17V35x),
+	EXAR_DEVICE(SEALEVEL, 716xC, pbn_exar_XR17V35x),
 	{ 0, }
 };
 MODULE_DEVICE_TABLE(pci, exar_pci_tbl);

2.39.2

From: Andrew Davis <[email protected]>

[ Upstream commit 95d698869b404772cc8b72560df71548491c10bc ]

Possibly the last PCI controller-based (i.e. not a soft/winmodem) dial-up modem one can still buy.

Looks to have a stock XR17C154 PCI UART chip for communication, but for some reason when provisioning the PCI IDs they swapped the vendor and subvendor IDs. Otherwise this card would have worked out of the box.

Searching online, some folks seem to not have this issue and others do, so it is possible only some batches of cards have this error.

Create a new macro to handle the switched IDs and add support here.

---
 drivers/tty/serial/8250/8250_exar.c | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)

index 2c9f721091125..5c2adf14049b7 100644
@@ -40,9 +40,13 @@
 #define PCI_DEVICE_ID_COMMTECH_4224PCIE		0x0020
 #define PCI_DEVICE_ID_COMMTECH_4228PCIE		0x0021
 #define PCI_DEVICE_ID_COMMTECH_4222PCIE		0x0022
+
 #define PCI_DEVICE_ID_EXAR_XR17V4358		0x4358
 #define PCI_DEVICE_ID_EXAR_XR17V8358		0x8358
 
+#define PCI_SUBDEVICE_ID_USR_2980		0x0128
+#define PCI_SUBDEVICE_ID_USR_2981		0x0129
+
 #define PCI_DEVICE_ID_SEALEVEL_710xC		0x1001
 #define PCI_DEVICE_ID_SEALEVEL_720xC		0x1002
 #define PCI_DEVICE_ID_SEALEVEL_740xC		0x1004

@@ -791,6 +795,15 @@ static const struct exar8250_board pbn_exar_XR17V8358 = {
 		(kernel_ulong_t)&bd			\
 	}
 
+#define USR_DEVICE(devid, sdevid, bd) {			\
+	PCI_DEVICE_SUB(					\
+		PCI_VENDOR_ID_USR,			\
+		PCI_DEVICE_ID_EXAR_##devid,		\
+		PCI_VENDOR_ID_EXAR,			\
+		PCI_SUBDEVICE_ID_USR_##sdevid), 0, 0,	\
+		(kernel_ulong_t)&bd			\
+	}
+
 static const struct pci_device_id exar_pci_tbl[] = {
 	EXAR_DEVICE(ACCESSIO, COM_2S, pbn_exar_XR17C15x),
 	EXAR_DEVICE(ACCESSIO, COM_4S, pbn_exar_XR17C15x),

@@ -815,6 +828,10 @@ static const struct pci_device_id exar_pci_tbl[] = {
 
 	IBM_DEVICE(XR17C152, SATURN_SERIAL_ONE_PORT, pbn_exar_ibm_saturn),
 
+	/* USRobotics USR298x-OEM PCI Modems */
+	USR_DEVICE(XR17C152, 2980, pbn_exar_XR17C15x),
+	USR_DEVICE(XR17C152, 2981, pbn_exar_XR17C15x),
+
 	/* Exar Corp. XR17C15[248] Dual/Quad/Octal UART */
 	EXAR_DEVICE(EXAR, XR17C152, pbn_exar_XR17C15x),
 	EXAR_DEVICE(EXAR, XR17C154, pbn_exar_XR17C15x),

2.39.2

From: Heiko Carstens <[email protected]>

[ Upstream commit d3e2ff5436d6ee38b572ba5c01dc7994769bec54 ]

Stable-dep-of: 2862a2fdfae8 ("s390/qdio: fix do_sqbs() inline assembly constraint")

---
 drivers/s390/cio/qdio.h      | 25 ++++++++-------
 drivers/s390/cio/qdio_main.c | 62 +++++++++++++++++++-----------------
 2 files changed, 46 insertions(+), 41 deletions(-)

index cd2df4ff8e0ef..854a21e1d3b7b 100644
@@ -88,15 +88,15 @@ enum qdio_irq_states {
 static inline int do_sqbs(u64 token, unsigned char state, int queue,
 			  int *start, int *count)
 {
-	register unsigned long _ccq asm ("0") = *count;
-	register unsigned long _token asm ("1") = token;
 	unsigned long _queuestart = ((unsigned long)queue << 32) | *start;
+	unsigned long _ccq = *count;
 
 	asm volatile(
-		"	.insn	rsy,0xeb000000008A,%1,0,0(%2)"
-		: "+d" (_ccq), "+d" (_queuestart)
-		: "d" ((unsigned long)state), "d" (_token)
-		: "memory", "cc");
+		"	lgr	1,%[token]\n"
+		"	.insn	rsy,0xeb000000008a,%[qs],%[ccq],0(%[state])"
+		: [ccq] "+&d" (_ccq), [qs] "+&d" (_queuestart)
+		: [state] "d" ((unsigned long)state), [token] "d" (token)
+		: "memory", "cc", "1");
 	*count = _ccq & 0xff;
 	*start = _queuestart & 0xff;
 

@@ -106,16 +106,17 @@ static inline int do_sqbs(u64 token, unsigned char state, int queue,
 static inline int do_eqbs(u64 token, unsigned char *state, int queue,
 			  int *start, int *count, int ack)
 {
-	register unsigned long _ccq asm ("0") = *count;
-	register unsigned long _token asm ("1") = token;
 	unsigned long _queuestart = ((unsigned long)queue << 32) | *start;
 	unsigned long _state = (unsigned long)ack << 63;
+	unsigned long _ccq = *count;
 
 	asm volatile(
-		"	.insn	rrf,0xB99c0000,%1,%2,0,0"
-		: "+d" (_ccq), "+d" (_queuestart), "+d" (_state)
-		: "d" (_token)
-		: "memory", "cc");
+		"	lgr	1,%[token]\n"
+		"	.insn	rrf,0xb99c0000,%[qs],%[state],%[ccq],0"
+		: [ccq] "+&d" (_ccq), [qs] "+&d" (_queuestart),
+		  [state] "+&d" (_state)
+		: [token] "d" (token)
+		: "memory", "cc", "1");
 	*count = _ccq & 0xff;
 	*start = _queuestart & 0xff;
 	*state = _state & 0xff;

2.39.2

From: Heiko Carstens <[email protected]>

[ Upstream commit 2862a2fdfae875888e3c1c3634e3422e01d98147 ]

Use "a" constraint instead of "d" constraint to pass the state parameter to the do_sqbs() inline assembly. This prevents that general purpose register zero is used for the state parameter.

If the compiler would select general purpose register zero this would be problematic for the used instruction in rsy format: the register used for the state parameter is a base register. If the base register is general purpose register zero the contents of the register are unexpectedly ignored when the instruction is executed.

This only applies to z/VM guests using QIOASSIST with dedicated (pass through) QDIO-based devices such as FCP [zfcp driver] as well as real OSA or HiperSockets [qeth driver].

A possible symptom for this case using zfcp is the following repeating kernel message pattern:

zfcp <devbusid>: A QDIO problem occurred
zfcp <devbusid>: A QDIO problem occurred
zfcp <devbusid>: qdio: ZFCP on SC <sc> using AI:1 QEBSM:1 PRI:1 TDD:1 SIGA: W zfcp <devbusid>: A QDIO problem occurred
zfcp <devbusid>: A QDIO problem occurred

Each of the qdio problem message can be accompanied by the following entries for the affected subchannel <sc> in /sys/kernel/debug/s390dbf/qdio_error/hex_ascii for zfcp or qeth:

<sc> ccq: 69....
<sc> SQBS ERROR.

---
 drivers/s390/cio/qdio.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

index 854a21e1d3b7b..919d106141664 100644
@@ -95,7 +95,7 @@ static inline int do_sqbs(u64 token, unsigned char state, int queue,
 		"	lgr	1,%[token]\n"
 		"	.insn	rsy,0xeb000000008a,%[qs],%[ccq],0(%[state])"
 		: [ccq] "+&d" (_ccq), [qs] "+&d" (_queuestart)
-		: [state] "d" ((unsigned long)state), [token] "d" (token)
+		: [state] "a" ((unsigned long)state), [token] "d" (token)
 		: "memory", "cc", "1");
 	*count = _ccq & 0xff;
 	*start = _queuestart & 0xff;

2.39.2

From: Gregory Oakes <[email protected]>

commit 4eda19cc8a29cde3580ed73bf11dc73b4e757697 upstream.

The watchdog countdown is supposed to begin when the device file is opened. Instead, it would begin countdown upon the first write to or close of the device file. Now, the ping operation is called within the start operation which ensures the countdown begins. From experimenation, it does not appear possible to do this with a single write including both the start bit and the trigger bit. So, it is done as two distinct writes.

---
 drivers/watchdog/sp5100_tco.c |    4 ++++
 1 file changed, 4 insertions(+)

@@ -104,6 +104,10 @@ static int tco_timer_start(struct watchd
 	val |= SP5100_WDT_START_STOP_BIT;
 	writel(val, SP5100_WDT_CONTROL(tco->tcobase));
 
+	/* This must be a distinct write. */
+	val |= SP5100_WDT_TRIGGER_BIT;
+	writel(val, SP5100_WDT_CONTROL(tco->tcobase));
+
 	return 0;
 }
 

From: Olivier Moysan <[email protected]>

commit ee2aacb6f3a901a95b1dd68964b69c92cdbbf213 upstream.

Replace sai2a-2 node name by sai2a-sleep-2, to avoid name duplication.

---
 arch/arm/boot/dts/stm32mp15-pinctrl.dtsi |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -1102,7 +1102,7 @@
 		};
 	};
 
-	sai2a_sleep_pins_c: sai2a-2 {
+	sai2a_sleep_pins_c: sai2a-sleep-2 {
 		pins {
 			pinmux = <STM32_PINMUX('D', 13, ANALOG)>, /* SAI2_SCK_A */
 				 <STM32_PINMUX('D', 11, ANALOG)>, /* SAI2_SD_A */

From: Greg Thelen <[email protected]>

5.10 stable commit 2b00b2a0e642 ("writeback, cgroup: fix null-ptr-deref write in bdi_split_work_to_wbs") is a backport of upstream 6.3 commit 1ba1199ec574.

In the 5.10 stable commit backport percpu_ref_exit() is called twice:
first in cgwb_release_workfn() and then in cgwb_free_rcu(). The 2nd call is benign as percpu_ref_exit() internally detects there's nothing to do.

This fixes an non-upstream issue that only applies to 5.10.y.

---
 mm/backing-dev.c |    1 -
 1 file changed, 1 deletion(-)

@@ -404,7 +404,6 @@ static void cgwb_release_workfn(struct w
 	blkcg_unpin_online(blkcg);
 
 	fprop_local_destroy_percpu(&wb->memcg_completions);
-	percpu_ref_exit(&wb->refcnt);
 	wb_exit(wb);
 	call_rcu(&wb->rcu, cgwb_free_rcu);
 }

From: wenxu <[email protected]>

[ Upstream commit fa6d639930ee5cd3f932cc314f3407f07a06582d ]

This one is prepare for the next patch.

---
 include/net/sch_generic.h |    5 -----
 net/sched/act_mirred.c    |   21 +++++++++++++++------
 2 files changed, 15 insertions(+), 11 deletions(-)

@@ -1320,11 +1320,6 @@ void mini_qdisc_pair_init(struct mini_Qd
 void mini_qdisc_pair_block_init(struct mini_Qdisc_pair *miniqp,
 				struct tcf_block *block);
 
-static inline int skb_tc_reinsert(struct sk_buff *skb, struct tcf_result *res)
-{
-	return res->ingress ? netif_receive_skb(skb) : dev_queue_xmit(skb);
-}
-
 /* Make sure qdisc is no longer in SCHED state. */
 static inline void qdisc_synchronize(const struct Qdisc *q)
 {

@@ -206,6 +206,18 @@ release_idr:
 	return err;
 }
 
+static int tcf_mirred_forward(bool want_ingress, struct sk_buff *skb)
+{
+	int err;
+
+	if (!want_ingress)
+		err = dev_queue_xmit(skb);
+	else
+		err = netif_receive_skb(skb);
+
+	return err;
+}
+
 static int tcf_mirred_act(struct sk_buff *skb, const struct tc_action *a,
 			  struct tcf_result *res)
 {

@@ -295,18 +307,15 @@ static int tcf_mirred_act(struct sk_buff
 		/* let's the caller reinsert the packet, if possible */
 		if (use_reinsert) {
 			res->ingress = want_ingress;
-			if (skb_tc_reinsert(skb, res))
+			err = tcf_mirred_forward(res->ingress, skb);
+			if (err)
 				tcf_action_inc_overlimit_qstats(&m->common);
 			__this_cpu_dec(mirred_rec_level);
 			return TC_ACT_CONSUMED;
 		}
 	}
 
-	if (!want_ingress)
-		err = dev_queue_xmit(skb2);
-	else
-		err = netif_receive_skb(skb2);
-
+	err = tcf_mirred_forward(want_ingress, skb2);
 	if (err) {
 out:
 		tcf_action_inc_overlimit_qstats(&m->common);

From: Davide Caratti <[email protected]>

[ Upstream commit 78dcdffe0418ac8f3f057f26fe71ccf4d8ed851f ]

with commit e2ca070f89ec ("net: sched: protect against stack overflow in TC act_mirred"), act_mirred protected itself against excessive stack growth using per_cpu counter of nested calls to tcf_mirred_act(), and capping it to MIRRED_RECURSION_LIMIT. However, such protection does not detect recursion/loops in case the packet is enqueued to the backlog (for example, when the mirred target device has RPS or skb timestamping enabled). Change the wording from "recursion" to "nesting" to make it more clear to readers.

Stable-dep-of: ca22da2fbd69 ("act_mirred: use the backlog for nested calls to mirred ingress")

---
 net/sched/act_mirred.c |   16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)

@@ -28,8 +28,8 @@
 static LIST_HEAD(mirred_list);
 static DEFINE_SPINLOCK(mirred_list_lock);
 
-#define MIRRED_RECURSION_LIMIT    4
-static DEFINE_PER_CPU(unsigned int, mirred_rec_level);
+#define MIRRED_NEST_LIMIT    4
+static DEFINE_PER_CPU(unsigned int, mirred_nest_level);
 
 static bool tcf_mirred_is_act_redirect(int action)
 {

@@ -225,7 +225,7 @@ static int tcf_mirred_act(struct sk_buff
 	struct sk_buff *skb2 = skb;
 	bool m_mac_header_xmit;
 	struct net_device *dev;
-	unsigned int rec_level;
+	unsigned int nest_level;
 	int retval, err = 0;
 	bool use_reinsert;
 	bool want_ingress;

@@ -236,11 +236,11 @@ static int tcf_mirred_act(struct sk_buff
 	int mac_len;
 	bool at_nh;
 
-	rec_level = __this_cpu_inc_return(mirred_rec_level);
-	if (unlikely(rec_level > MIRRED_RECURSION_LIMIT)) {
+	nest_level = __this_cpu_inc_return(mirred_nest_level);
+	if (unlikely(nest_level > MIRRED_NEST_LIMIT)) {
 		net_warn_ratelimited("Packet exceeded mirred recursion limit on dev %s\n",
 				     netdev_name(skb->dev));
-		__this_cpu_dec(mirred_rec_level);
+		__this_cpu_dec(mirred_nest_level);
 		return TC_ACT_SHOT;
 	}
 

@@ -310,7 +310,7 @@ static int tcf_mirred_act(struct sk_buff
 			err = tcf_mirred_forward(res->ingress, skb);
 			if (err)
 				tcf_action_inc_overlimit_qstats(&m->common);
-			__this_cpu_dec(mirred_rec_level);
+			__this_cpu_dec(mirred_nest_level);
 			return TC_ACT_CONSUMED;
 		}
 	}

@@ -322,7 +322,7 @@ out:
 		if (tcf_mirred_is_act_redirect(m_eaction))
 			retval = TC_ACT_SHOT;
 	}
-	__this_cpu_dec(mirred_rec_level);
+	__this_cpu_dec(mirred_nest_level);
 
 	return retval;
 }

From: Davide Caratti <[email protected]>

[ Upstream commit ca22da2fbd693b54dc8e3b7b54ccc9f7e9ba3640 ]

William reports kernel soft-lockups on some OVS topologies when TC mirred egress->ingress action is hit by local TCP traffic [1].
The same can also be reproduced with SCTP (thanks Xin for verifying), when client and server reach themselves through mirred egress to ingress, and one of the two peers sends a "heartbeat" packet (from within a timer).

Enqueueing to backlog proved to fix this soft lockup; however, as Cong noticed [2], we should preserve - when possible - the current mirred behavior that counts as "overlimits" any eventual packet drop subsequent to the mirred forwarding action [3]. A compromise solution might use the backlog only when tcf_mirred_act() has a nest level greater than one:
change tcf_mirred_forward() accordingly.

Also, add a kselftest that can reproduce the lockup and verifies TC mirred ability to account for further packet drops after TC mirred egress->ingress (when the nest level is 1).

[1] https://lore.kernel.org/netdev/33dc43f587ec1388ba456b4915c75f02a8aae226.1663945716.git.dcaratti@redhat.com/
[2] https://lore.kernel.org/netdev/Y0w%[email protected]/
[3] such behavior is not guaranteed: for example, if RPS or skb RX
timestamping is enabled on the mirred target device, the kernel
can defer receiving the skb and return NET_RX_SUCCESS inside
tcf_mirred_forward().

---
 net/sched/act_mirred.c                               |    7 ++
 tools/testing/selftests/net/forwarding/tc_actions.sh |   48 ++++++++++++++++++-
 2 files changed, 54 insertions(+), 1 deletion(-)

@@ -206,12 +206,19 @@ release_idr:
 	return err;
 }
 
+static bool is_mirred_nested(void)
+{
+	return unlikely(__this_cpu_read(mirred_nest_level) > 1);
+}
+
 static int tcf_mirred_forward(bool want_ingress, struct sk_buff *skb)
 {
 	int err;
 
 	if (!want_ingress)
 		err = dev_queue_xmit(skb);
+	else if (is_mirred_nested())
+		err = netif_rx(skb);
 	else
 		err = netif_receive_skb(skb);
 

From: Christophe Leroy <[email protected]>

(backported from upstream 8a5299a1278eadf1e08a598a5345c376206f171e)

For different reasons, fsl-spi driver performs bits_per_word modifications for different reasons:
- On CPU mode, to minimise amount of interrupts
- On CPM/QE mode to work around controller byte order

For CPU mode that's done in fsl_spi_prepare_message() while for CPM mode that's done in fsl_spi_setup_transfer().

Reunify all of it in fsl_spi_prepare_message(), and catch impossible cases early through master's bits_per_word_mask instead of returning EINVAL later.

---
 drivers/spi/spi-fsl-spi.c |   50 +++++++++++++++++++++-------------------------
 1 file changed, 23 insertions(+), 27 deletions(-)

@@ -203,26 +203,6 @@ static int mspi_apply_cpu_mode_quirks(st
 	return bits_per_word;
 }
 
-static int mspi_apply_qe_mode_quirks(struct spi_mpc8xxx_cs *cs,
-				struct spi_device *spi,
-				int bits_per_word)
-{
-	/* CPM/QE uses Little Endian for words > 8
-	 * so transform 16 and 32 bits words into 8 bits
-	 * Unfortnatly that doesn't work for LSB so
-	 * reject these for now */
-	/* Note: 32 bits word, LSB works iff
-	 * tfcr/rfcr is set to CPMFCR_GBL */
-	if (spi->mode & SPI_LSB_FIRST &&
-	    bits_per_word > 8)
-		return -EINVAL;
-	if (bits_per_word <= 8)
-		return bits_per_word;
-	if (bits_per_word == 16 || bits_per_word == 32)
-		return 8; /* pretend its 8 bits */
-	return -EINVAL;
-}
-
 static int fsl_spi_setup_transfer(struct spi_device *spi,
 					struct spi_transfer *t)
 {

@@ -250,9 +230,6 @@ static int fsl_spi_setup_transfer(struct
 		bits_per_word = mspi_apply_cpu_mode_quirks(cs, spi,
 							   mpc8xxx_spi,
 							   bits_per_word);
-	else
-		bits_per_word = mspi_apply_qe_mode_quirks(cs, spi,
-							  bits_per_word);
 
 	if (bits_per_word < 0)
 		return bits_per_word;

@@ -370,14 +347,27 @@ static int fsl_spi_do_one_msg(struct spi
 	 * In CPU mode, optimize large byte transfers to use larger
 	 * bits_per_word values to reduce number of interrupts taken.
 	 */
-	if (!(mpc8xxx_spi->flags & SPI_CPM_MODE)) {
-		list_for_each_entry(t, &m->transfers, transfer_list) {
+	list_for_each_entry(t, &m->transfers, transfer_list) {
+		if (!(mpc8xxx_spi->flags & SPI_CPM_MODE)) {
 			if (t->len < 256 || t->bits_per_word != 8)
 				continue;
 			if ((t->len & 3) == 0)
 				t->bits_per_word = 32;
 			else if ((t->len & 1) == 0)
 				t->bits_per_word = 16;
+		} else {
+			/*
+			 * CPM/QE uses Little Endian for words > 8
+			 * so transform 16 and 32 bits words into 8 bits
+			 * Unfortnatly that doesn't work for LSB so
+			 * reject these for now
+			 * Note: 32 bits word, LSB works iff
+			 * tfcr/rfcr is set to CPMFCR_GBL
+			 */
+			if (m->spi->mode & SPI_LSB_FIRST && t->bits_per_word > 8)
+				return -EINVAL;
+			if (t->bits_per_word == 16 || t->bits_per_word == 32)
+				t->bits_per_word = 8; /* pretend its 8 bits */
 		}
 	}
 

@@ -635,8 +625,14 @@ static struct spi_master *fsl_spi_probe(
 	if (mpc8xxx_spi->type == TYPE_GRLIB)
 		fsl_spi_grlib_probe(dev);
 
-	master->bits_per_word_mask =
-		(SPI_BPW_RANGE_MASK(4, 16) | SPI_BPW_MASK(32)) &
+	if (mpc8xxx_spi->flags & SPI_CPM_MODE)
+		master->bits_per_word_mask =
+			(SPI_BPW_RANGE_MASK(4, 8) | SPI_BPW_MASK(16) | SPI_BPW_MASK(32));
+	else
+		master->bits_per_word_mask =
+			(SPI_BPW_RANGE_MASK(4, 16) | SPI_BPW_MASK(32));
+
+	master->bits_per_word_mask &=
 		SPI_BPW_RANGE_MASK(1, mpc8xxx_spi->max_bits_per_word);
 
 	if (mpc8xxx_spi->flags & SPI_QE_CPU_MODE)

From: Christophe Leroy <[email protected]>

(cherry picked from upstream fc96ec826bced75cc6b9c07a4ac44bbf651337ab)

On CPM, the RISC core is a lot more efficiant when doing transfers in 16-bits chunks than in 8-bits chunks, but unfortunately the words need to be byte swapped as seen in a previous commit.

So, for large tranfers with an even size, allocate a temporary tx buffer and byte-swap data before and after transfer.

This change allows setting higher speed for transfer. For instance on an MPC 8xx (CPM1 comms RISC processor), the documentation tells that transfer in byte mode at 1 kbit/s uses 0.200% of CPM load at 25 MHz while a word transfer at the same speed uses 0.032% of CPM load. This means the speed can be 6 times higher in word mode for the same CPM load.

For the time being, only do it on CPM1 as there must be a trade-off between the CPM load reduction and the CPU load required to byte swap the data.

---
 drivers/spi/spi-fsl-cpm.c |   23 +++++++++++++++++++++++
 drivers/spi/spi-fsl-spi.c |    3 +++
 2 files changed, 26 insertions(+)

@@ -21,6 +21,7 @@
 #include <linux/spi/spi.h>
 #include <linux/types.h>
 #include <linux/platform_device.h>
+#include <linux/byteorder/generic.h>
 
 #include "spi-fsl-cpm.h"
 #include "spi-fsl-lib.h"

@@ -120,6 +121,21 @@ int fsl_spi_cpm_bufs(struct mpc8xxx_spi
 		mspi->rx_dma = mspi->dma_dummy_rx;
 		mspi->map_rx_dma = 0;
 	}
+	if (t->bits_per_word == 16 && t->tx_buf) {
+		const u16 *src = t->tx_buf;
+		u16 *dst;
+		int i;
+
+		dst = kmalloc(t->len, GFP_KERNEL);
+		if (!dst)
+			return -ENOMEM;
+
+		for (i = 0; i < t->len >> 1; i++)
+			dst[i] = cpu_to_le16p(src + i);
+
+		mspi->tx = dst;
+		mspi->map_tx_dma = 1;
+	}
 
 	if (mspi->map_tx_dma) {
 		void *nonconst_tx = (void *)mspi->tx; /* shut up gcc */

@@ -173,6 +189,13 @@ void fsl_spi_cpm_bufs_complete(struct mp
 	if (mspi->map_rx_dma)
 		dma_unmap_single(dev, mspi->rx_dma, t->len, DMA_FROM_DEVICE);
 	mspi->xfer_in_progress = NULL;
+
+	if (t->bits_per_word == 16 && t->rx_buf) {
+		int i;
+
+		for (i = 0; i < t->len; i += 2)
+			le16_to_cpus(t->rx_buf + i);
+	}
 }
 EXPORT_SYMBOL_GPL(fsl_spi_cpm_bufs_complete);
 

@@ -368,6 +368,9 @@ static int fsl_spi_do_one_msg(struct spi
 				return -EINVAL;
 			if (t->bits_per_word == 16 || t->bits_per_word == 32)
 				t->bits_per_word = 8; /* pretend its 8 bits */
+			if (t->bits_per_word == 8 && t->len >= 256 &&
+			    (mpc8xxx_spi->flags & SPI_CPM1))
+				t->bits_per_word = 16;
 		}
 	}
 

From: Roberto Sassu <[email protected]>

commit de3004c874e740304cc4f4a83d6200acb511bbda upstream.

In preparation for removing security_old_inode_init_security(), switch to security_inode_init_security().

Extend the existing ocfs2_initxattrs() to take the ocfs2_security_xattr_info structure from fs_info, and populate the name/value/len triple with the first xattr provided by LSMs.

As fs_info was not used before, ocfs2_initxattrs() can now handle the case of replicating the behavior of security_old_inode_init_security(), i.e. just obtaining the xattr, in addition to setting all xattrs provided by LSMs.

Supporting multiple xattrs is not currently supported where security_old_inode_init_security() was called (mknod, symlink), as it requires non-trivial changes that can be done at a later time. Like for reiserfs, even if EVM is invoked, it will not provide an xattr (if it is not the first to set it, its xattr will be discarded; if it is the first, it does not have xattrs to calculate the HMAC on).

Finally, since security_inode_init_security(), unlike security_old_inode_init_security(), returns zero instead of -EOPNOTSUPP if no xattrs were provided by LSMs or if inodes are private, additionally check in ocfs2_init_security_get() if the xattr name is set.

If not, act as if security_old_inode_init_security() returned -EOPNOTSUPP, and set si->enable to zero to notify to the functions following ocfs2_init_security_get() that no xattrs are available.

---
 fs/ocfs2/namei.c |    2 ++
 fs/ocfs2/xattr.c |   30 ++++++++++++++++++++++++++----
 2 files changed, 28 insertions(+), 4 deletions(-)

@@ -242,6 +242,7 @@ static int ocfs2_mknod(struct inode *dir
 	int want_meta = 0;
 	int xattr_credits = 0;
 	struct ocfs2_security_xattr_info si = {
+		.name = NULL,
 		.enable = 1,
 	};
 	int did_quota_inode = 0;

@@ -1801,6 +1802,7 @@ static int ocfs2_symlink(struct inode *d
 	int want_clusters = 0;
 	int xattr_credits = 0;
 	struct ocfs2_security_xattr_info si = {
+		.name = NULL,
 		.enable = 1,
 	};
 	int did_quota = 0, did_quota_inode = 0;

@@ -7260,9 +7260,21 @@ static int ocfs2_xattr_security_set(cons
 static int ocfs2_initxattrs(struct inode *inode, const struct xattr *xattr_array,
 		     void *fs_info)
 {
+	struct ocfs2_security_xattr_info *si = fs_info;
 	const struct xattr *xattr;
 	int err = 0;
 
+	if (si) {
+		si->value = kmemdup(xattr_array->value, xattr_array->value_len,
+				    GFP_KERNEL);
+		if (!si->value)
+			return -ENOMEM;
+
+		si->name = xattr_array->name;
+		si->value_len = xattr_array->value_len;
+		return 0;
+	}
+
 	for (xattr = xattr_array; xattr->name != NULL; xattr++) {
 		err = ocfs2_xattr_set(inode, OCFS2_XATTR_INDEX_SECURITY,
 				      xattr->name, xattr->value,

@@ -7278,13 +7290,23 @@ int ocfs2_init_security_get(struct inode
 			    const struct qstr *qstr,
 			    struct ocfs2_security_xattr_info *si)
 {
+	int ret;
+
 	/* check whether ocfs2 support feature xattr */
 	if (!ocfs2_supports_xattr(OCFS2_SB(dir->i_sb)))
 		return -EOPNOTSUPP;
-	if (si)
-		return security_old_inode_init_security(inode, dir, qstr,
-							&si->name, &si->value,
-							&si->value_len);
+	if (si) {
+		ret = security_inode_init_security(inode, dir, qstr,
+						   &ocfs2_initxattrs, si);
+		/*
+		 * security_inode_init_security() does not return -EOPNOTSUPP,
+		 * we have to check the xattr ourselves.
+		 */
+		if (!ret && !si->name)
+			si->enable = 0;
+
+		return ret;
+	}
 
 	return security_inode_init_security(inode, dir, qstr,
 					    &ocfs2_initxattrs, NULL);

From: Adam Stylinski <[email protected]>

commit 7843380d07bbeffd3ce6504e73cf61f840ae76ca upstream.

This quirk is necessary for surround and other DSP effects to work with the onboard ca0132 based audio chipset for the EVGA X299 dark mainboard.

---
 sound/pci/hda/patch_ca0132.c |    1 +
 1 file changed, 1 insertion(+)

@@ -1272,6 +1272,7 @@ static const struct snd_pci_quirk ca0132
 	SND_PCI_QUIRK(0x1458, 0xA026, "Gigabyte G1.Sniper Z97", QUIRK_R3DI),
 	SND_PCI_QUIRK(0x1458, 0xA036, "Gigabyte GA-Z170X-Gaming 7", QUIRK_R3DI),
 	SND_PCI_QUIRK(0x3842, 0x1038, "EVGA X99 Classified", QUIRK_R3DI),
+	SND_PCI_QUIRK(0x3842, 0x104b, "EVGA X299 Dark", QUIRK_R3DI),
 	SND_PCI_QUIRK(0x3842, 0x1055, "EVGA Z390 DARK", QUIRK_R3DI),
 	SND_PCI_QUIRK(0x1102, 0x0013, "Recon3D", QUIRK_R3D),
 	SND_PCI_QUIRK(0x1102, 0x0018, "Recon3D", QUIRK_R3D),

From: Takashi Iwai <[email protected]>

commit 81302b1c7c997e8a56c1c2fc63a296ebeb0cd2d0 upstream.

It's reported that the recording started right after the driver probe doesn't work properly, and it turned out that this is related with the codec auto-suspend. Namely, after the probe phase, the usage count goes zero, and the auto-suspend is programmed, but the codec is kept still active until the auto-suspend expiration. When an application (e.g. alsactl) updates the mixer values at this moment, the values are cached but not actually written. Then, starting arecord thereafter also results in the silence because of the missing unmute.

The root cause is the handling of "lazy update" mode; when a mixer value is updated *after* the suspend, it should update only the cache and exits. At the resume, the cached value is written to the device, in turn. The problem is that the current code misinterprets the state of auto-suspend as if it were already suspended.

Although we can add the check of the actual device state after pm_runtime_get_if_in_use() for catching the missing state, this won't suffice; the second call of regmap_update_bits_check() will skip writing the register because the cache has been already updated by the first call. So we'd need fixes in two different places.

OTOH, a simpler fix is to replace pm_runtime_get_if_in_use() with pm_runtime_get_if_active() (with ign_usage_count=true). This change implies that the driver takes the pm refcount if the device is still in ACTIVE state and continues the processing. A small caveat is that this will leave the auto-suspend timer. But, since the timer callback itself checks the device state and aborts gracefully when it's active, this won't be any substantial problem.

Long story short: we address the missing register-write problem just by replacing the pm_runtime_*() call in snd_hda_keep_power_up().

---
 sound/hda/hdac_device.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -611,7 +611,7 @@ EXPORT_SYMBOL_GPL(snd_hdac_power_up_pm);
 int snd_hdac_keep_power_up(struct hdac_device *codec)
 {
 	if (!atomic_inc_not_zero(&codec->in_pm)) {
-		int ret = pm_runtime_get_if_in_use(&codec->dev);
+		int ret = pm_runtime_get_if_active(&codec->dev, true);
 		if (!ret)
 			return -1;
 		if (ret < 0)

From: Bin Li <[email protected]>

commit 4ca110cab46561cd74a2acd9b447435acb4bec5f upstream.

Lenovo M70/M90 Gen4 are equipped with ALC897, and they need ALC897_FIXUP_HEADSET_MIC_PIN quirk to make its headset mic work. The previous quirk for M70/M90 is for Gen3.

---
 sound/pci/hda/patch_realtek.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -11187,6 +11187,8 @@ static const struct snd_pci_quirk alc662
 	SND_PCI_QUIRK(0x17aa, 0x32cb, "Lenovo ThinkCentre M70", ALC897_FIXUP_HEADSET_MIC_PIN),
 	SND_PCI_QUIRK(0x17aa, 0x32cf, "Lenovo ThinkCentre M950", ALC897_FIXUP_HEADSET_MIC_PIN),
 	SND_PCI_QUIRK(0x17aa, 0x32f7, "Lenovo ThinkCentre M90", ALC897_FIXUP_HEADSET_MIC_PIN),
+	SND_PCI_QUIRK(0x17aa, 0x3321, "Lenovo ThinkCentre M70 Gen4", ALC897_FIXUP_HEADSET_MIC_PIN),
+	SND_PCI_QUIRK(0x17aa, 0x331b, "Lenovo ThinkCentre M90 Gen4", ALC897_FIXUP_HEADSET_MIC_PIN),
 	SND_PCI_QUIRK(0x17aa, 0x3742, "Lenovo TianYi510Pro-14IOB", ALC897_FIXUP_HEADSET_MIC_PIN2),
 	SND_PCI_QUIRK(0x17aa, 0x38af, "Lenovo Ideapad Y550P", ALC662_FIXUP_IDEAPAD),
 	SND_PCI_QUIRK(0x17aa, 0x3a0d, "Lenovo Ideapad Y550", ALC662_FIXUP_IDEAPAD),

From: Tudor Ambarus <[email protected]>

commit 7e01c7f7046efc2c7c192c3619db43292b98e997 upstream.

Currently in cdc_ncm_check_tx_max(), if dwNtbOutMaxSize is lower than the calculated "min" value, but greater than zero, the logic sets tx_max to dwNtbOutMaxSize. This is then used to allocate a new SKB in cdc_ncm_fill_tx_frame() where all the data is handled.

For small values of dwNtbOutMaxSize the memory allocated during alloc_skb(dwNtbOutMaxSize, GFP_ATOMIC) will have the same size, due to how size is aligned at alloc time:

	size = SKB_DATA_ALIGN(size);
        size += SKB_DATA_ALIGN(sizeof(struct skb_shared_info));

Thus we hit the same bug that we tried to squash with
commit 2be6d4d16a084 ("net: cdc_ncm: Allow for dwNtbOutMaxSize to be unset or zero")

Low values of dwNtbOutMaxSize do not cause an issue presently because at alloc_skb() time more memory (512b) is allocated than required for the SKB headers alone (320b), leaving some space (512b - 320b = 192b) for CDC data (172b).

However, if more elements (for example 3 x u64 = [24b]) were added to one of the SKB header structs, say 'struct skb_shared_info', increasing its original size (320b [320b aligned]) to something larger (344b [384b aligned]), then suddenly the CDC data (172b) no longer fits in the spare SKB data area (512b - 384b = 128b).

Consequently the SKB bounds checking semantics fails and panics:

skbuff: skb_over_panic: text:ffffffff831f755b len:184 put:172 head:ffff88811f1c6c00 data:ffff88811f1c6c00 tail:0xb8 end:0x80 dev:<NULL>

------------[ cut here ]------------
kernel BUG at net/core/skbuff.c:113!
invalid opcode: 0000 [#1] PREEMPT SMP KASAN
CPU: 0 PID: 57 Comm: kworker/0:2 Not tainted 5.15.106-syzkaller-00249-g19c0ed55a470 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 04/14/2023
Workqueue: mld mld_ifc_work
RIP: 0010:skb_panic net/core/skbuff.c:113 [inline]
RIP: 0010:skb_over_panic+0x14c/0x150 net/core/skbuff.c:118
[snip]
Call Trace:
 <TASK>
 skb_put+0x151/0x210 net/core/skbuff.c:2047
 skb_put_zero include/linux/skbuff.h:2422 [inline]
 cdc_ncm_ndp16 drivers/net/usb/cdc_ncm.c:1131 [inline]
 cdc_ncm_fill_tx_frame+0x11ab/0x3da0 drivers/net/usb/cdc_ncm.c:1308
 cdc_ncm_tx_fixup+0xa3/0x100

Deal with too low values of dwNtbOutMaxSize, clamp it in the range [USB_CDC_NCM_NTB_MIN_OUT_SIZE, CDC_NCM_NTB_MAX_SIZE_TX]. We ensure enough data space is allocated to handle CDC data by making sure dwNtbOutMaxSize is not smaller than USB_CDC_NCM_NTB_MIN_OUT_SIZE.

---
 drivers/net/usb/cdc_ncm.c |   24 +++++++++++++++---------
 1 file changed, 15 insertions(+), 9 deletions(-)

@@ -180,9 +180,12 @@ static u32 cdc_ncm_check_tx_max(struct u
 	else
 		min = ctx->max_datagram_size + ctx->max_ndp_size + sizeof(struct usb_cdc_ncm_nth32);
 
-	max = min_t(u32, CDC_NCM_NTB_MAX_SIZE_TX, le32_to_cpu(ctx->ncm_parm.dwNtbOutMaxSize));
-	if (max == 0)
+	if (le32_to_cpu(ctx->ncm_parm.dwNtbOutMaxSize) == 0)
 		max = CDC_NCM_NTB_MAX_SIZE_TX; /* dwNtbOutMaxSize not set */
+	else
+		max = clamp_t(u32, le32_to_cpu(ctx->ncm_parm.dwNtbOutMaxSize),
+			      USB_CDC_NCM_NTB_MIN_OUT_SIZE,
+			      CDC_NCM_NTB_MAX_SIZE_TX);
 
 	/* some devices set dwNtbOutMaxSize too low for the above default */
 	min = min(min, max);

@@ -1230,6 +1233,9 @@ cdc_ncm_fill_tx_frame(struct usbnet *dev
 			 * further.
 			 */
 			if (skb_out == NULL) {
+				/* If even the smallest allocation fails, abort. */
+				if (ctx->tx_curr_size == USB_CDC_NCM_NTB_MIN_OUT_SIZE)
+					goto alloc_failed;
 				ctx->tx_low_mem_max_cnt = min(ctx->tx_low_mem_max_cnt + 1,
 							      (unsigned)CDC_NCM_LOW_MEM_MAX_CNT);
 				ctx->tx_low_mem_val = ctx->tx_low_mem_max_cnt;

@@ -1248,13 +1254,8 @@ cdc_ncm_fill_tx_frame(struct usbnet *dev
 			skb_out = alloc_skb(ctx->tx_curr_size, GFP_ATOMIC);
 
 			/* No allocation possible so we will abort */
-			if (skb_out == NULL) {
-				if (skb != NULL) {
-					dev_kfree_skb_any(skb);
-					dev->net->stats.tx_dropped++;
-				}
-				goto exit_no_skb;
-			}
+			if (!skb_out)
+				goto alloc_failed;
 			ctx->tx_low_mem_val--;
 		}
 		if (ctx->is_ndp16) {

@@ -1447,6 +1448,11 @@ cdc_ncm_fill_tx_frame(struct usbnet *dev
 
 	return skb_out;
 
+alloc_failed:
+	if (skb) {
+		dev_kfree_skb_any(skb);
+		dev->net->stats.tx_dropped++;
+	}
 exit_no_skb:
 	/* Start timer, if there is a remaining non-empty skb */
 	if (ctx->tx_curr_skb != NULL && n > 0)

From: Finn Thain <[email protected]>

commit b845b574f86dcb6a70dfa698aa87a237b0878d2a upstream.

On 68030/020, an instruction such as, moveml %a2-%a3/%a5,%sp@- may cause a stack page fault during instruction execution (i.e. not at an instruction boundary) and produce a format 0xB exception frame.

In this situation, the value of USP will be unreliable. If a signal is to be delivered following the exception, this USP value is used to calculate the location for a signal frame. This can result in a corrupted user stack.

The corruption was detected in dash (actually in glibc) where it showed up as an intermittent "stack smashing detected" message and crash following signal delivery for SIGCHLD.

It was hard to reproduce that failure because delivery of the signal raced with the page fault and because the kernel places an unpredictable gap of up to 7 bytes between the USP and the signal frame.

A format 0xB exception frame can be produced by a bus error or an address error. The 68030 Users Manual says that address errors occur immediately upon detection during instruction prefetch. The instruction pipeline allows prefetch to overlap with other instructions, which means an address error can arise during the execution of a different instruction. So it seems likely that this patch may help in the address error case also.

---
 arch/m68k/kernel/signal.c |   14 ++++++++++----
 1 file changed, 10 insertions(+), 4 deletions(-)

@@ -882,11 +882,17 @@ static inline int rt_setup_ucontext(stru
 }
 
 static inline void __user *
-get_sigframe(struct ksignal *ksig, size_t frame_size)
+get_sigframe(struct ksignal *ksig, struct pt_regs *tregs, size_t frame_size)
 {
 	unsigned long usp = sigsp(rdusp(), ksig);
+	unsigned long gap = 0;
 
-	return (void __user *)((usp - frame_size) & -8UL);
+	if (CPU_IS_020_OR_030 && tregs->format == 0xb) {
+		/* USP is unreliable so use worst-case value */
+		gap = 256;
+	}
+
+	return (void __user *)((usp - gap - frame_size) & -8UL);
 }
 
 static int setup_frame(struct ksignal *ksig, sigset_t *set,

@@ -904,7 +910,7 @@ static int setup_frame(struct ksignal *k
 		return -EFAULT;
 	}
 
-	frame = get_sigframe(ksig, sizeof(*frame) + fsize);
+	frame = get_sigframe(ksig, tregs, sizeof(*frame) + fsize);
 
 	if (fsize)
 		err |= copy_to_user (frame + 1, regs + 1, fsize);

@@ -976,7 +982,7 @@ static int setup_rt_frame(struct ksignal
 		return -EFAULT;
 	}
 
-	frame = get_sigframe(ksig, sizeof(*frame));
+	frame = get_sigframe(ksig, tregs, sizeof(*frame));
 
 	if (fsize)
 		err |= copy_to_user (&frame->uc.uc_extra, regs + 1, fsize);

From: Helge Deller <[email protected]>

commit 6888ff04e37d01295620a73f3f7efbc79f6ef152 upstream.

The kernel kgdb break instructions should only be handled when running in kernel context.

---
 arch/parisc/kernel/traps.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -305,8 +305,8 @@ static void handle_break(struct pt_regs
 #endif
 
 #ifdef CONFIG_KGDB
-	if (unlikely(iir == PARISC_KGDB_COMPILED_BREAK_INSN ||
-		iir == PARISC_KGDB_BREAK_INSN)) {
+	if (unlikely((iir == PARISC_KGDB_COMPILED_BREAK_INSN ||
+		iir == PARISC_KGDB_BREAK_INSN)) && !user_mode(regs)) {
 		kgdb_handle_exception(9, SIGTRAP, 0, regs);
 		return;
 	}

From: Helge Deller <[email protected]>

commit 2028315cf59bb899a5ac7e87dc48ecb8fac7ac24 upstream.

In case a machine can't power-off itself on system shutdown, allow the user to reboot it by pressing the RETURN key.

---
 arch/parisc/kernel/process.c |   11 ++++++++---
 1 file changed, 8 insertions(+), 3 deletions(-)

@@ -123,13 +123,18 @@ void machine_power_off(void)
 	/* It seems we have no way to power the system off via
 	 * software. The user has to press the button himself. */
 
-	printk(KERN_EMERG "System shut down completed.\n"
-	       "Please power this system off now.");
+	printk("Power off or press RETURN to reboot.\n");
 
 	/* prevent soft lockup/stalled CPU messages for endless loop. */
 	rcu_sysrq_start();
 	lockup_detector_soft_poweroff();
-	for (;;);
+	while (1) {
+		/* reboot if user presses RETURN key */
+		if (pdc_iodc_getc() == 13) {
+			printk("Rebooting...\n");
+			machine_restart(NULL);
+		}
+	}
 }
 
 void (*pm_power_off)(void);

From: Zev Weiss <[email protected]>

commit 0a1bb16e0fe6650c3841e611de374bfd5578ad70 upstream.

This driver's debugfs files have had a read operation since commit 2a9e27408e12 ("gpio: mockup: rework debugfs interface"), but were still being created with write-only mode bits. Update them to indicate that the files can also be read.

---
 drivers/gpio/gpio-mockup.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -370,7 +370,7 @@ static void gpio_mockup_debugfs_setup(st
 		priv->offset = i;
 		priv->desc = &gc->gpiodev->descs[i];
 
-		debugfs_create_file(name, 0200, chip->dbg_dir, priv,
+		debugfs_create_file(name, 0600, chip->dbg_dir, priv,
 				    &gpio_mockup_debugfs_ops);
 	}
 }

From: Josef Bacik <[email protected]>

commit 597441b3436a43011f31ce71dc0a6c0bf5ce958a upstream.

Our CI system caught a lockdep splat:

======================================================
WARNING: possible circular locking dependency detected
6.3.0-rc7+ #1167 Not tainted
------------------------------------------------------
kswapd0/46 is trying to acquire lock:
ffff8c6543abd650 (sb_internal#2){++++}-{0:0}, at: btrfs_commit_inode_delayed_inode+0x5f/0x120

but task is already holding lock:
ffffffffabe61b40 (fs_reclaim){+.+.}-{0:0}, at: balance_pgdat+0x4aa/0x7a0

which lock already depends on the new lock.

the existing dependency chain (in reverse order) is:

-> #1 (fs_reclaim){+.+.}-{0:0}:

	 fs_reclaim_acquire+0xa5/0xe0
	 kmem_cache_alloc+0x31/0x2c0
	 alloc_extent_state+0x1d/0xd0
	 __clear_extent_bit+0x2e0/0x4f0
	 try_release_extent_mapping+0x216/0x280
	 btrfs_release_folio+0x2e/0x90
	 invalidate_inode_pages2_range+0x397/0x470
	 btrfs_cleanup_dirty_bgs+0x9e/0x210
	 btrfs_cleanup_one_transaction+0x22/0x760
	 btrfs_commit_transaction+0x3b7/0x13a0
	 create_subvol+0x59b/0x970
	 btrfs_mksubvol+0x435/0x4f0
	 __btrfs_ioctl_snap_create+0x11e/0x1b0
	 btrfs_ioctl_snap_create_v2+0xbf/0x140
	 btrfs_ioctl+0xa45/0x28f0
	 __x64_sys_ioctl+0x88/0xc0
	 do_syscall_64+0x38/0x90
	 entry_SYSCALL_64_after_hwframe+0x72/0xdc

-> #0 (sb_internal#2){++++}-{0:0}:

	 __lock_acquire+0x1435/0x21a0
	 lock_acquire+0xc2/0x2b0
	 start_transaction+0x401/0x730
	 btrfs_commit_inode_delayed_inode+0x5f/0x120
	 btrfs_evict_inode+0x292/0x3d0
	 evict+0xcc/0x1d0
	 inode_lru_isolate+0x14d/0x1e0
	 __list_lru_walk_one+0xbe/0x1c0
	 list_lru_walk_one+0x58/0x80
	 prune_icache_sb+0x39/0x60
	 super_cache_scan+0x161/0x1f0
	 do_shrink_slab+0x163/0x340
	 shrink_slab+0x1d3/0x290
	 shrink_node+0x300/0x720
	 balance_pgdat+0x35c/0x7a0
	 kswapd+0x205/0x410
	 kthread+0xf0/0x120
	 ret_from_fork+0x29/0x50

other info that might help us debug this:

   Possible unsafe locking scenario:

	 CPU0                    CPU1
	 ----                    ----
    lock(fs_reclaim);
				 lock(sb_internal#2);
				 lock(fs_reclaim);
    lock(sb_internal#2);

*** DEADLOCK ***

3 locks held by kswapd0/46:
#0: ffffffffabe61b40 (fs_reclaim){+.+.}-{0:0}, at: balance_pgdat+0x4aa/0x7a0 #1: ffffffffabe50270 (shrinker_rwsem){++++}-{3:3}, at: shrink_slab+0x113/0x290 #2: ffff8c6543abd0e0 (&type->s_umount_key#44){++++}-{3:3}, at: super_cache_scan+0x38/0x1f0

  stack backtrace:
  CPU: 0 PID: 46 Comm: kswapd0 Not tainted 6.3.0-rc7+ #1167
  Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.13.0-2.fc32 04/01/2014
  Call Trace:
   <TASK>
   dump_stack_lvl+0x58/0x90
   check_noncircular+0xd6/0x100
   ? save_trace+0x3f/0x310
   ? add_lock_to_list+0x97/0x120
   __lock_acquire+0x1435/0x21a0
   lock_acquire+0xc2/0x2b0
   ? btrfs_commit_inode_delayed_inode+0x5f/0x120
   start_transaction+0x401/0x730
   ? btrfs_commit_inode_delayed_inode+0x5f/0x120
   btrfs_commit_inode_delayed_inode+0x5f/0x120
   btrfs_evict_inode+0x292/0x3d0
   ? lock_release+0x134/0x270
   ? __pfx_wake_bit_function+0x10/0x10
   evict+0xcc/0x1d0
   inode_lru_isolate+0x14d/0x1e0
   __list_lru_walk_one+0xbe/0x1c0
   ? __pfx_inode_lru_isolate+0x10/0x10
   ? __pfx_inode_lru_isolate+0x10/0x10
   list_lru_walk_one+0x58/0x80
   prune_icache_sb+0x39/0x60
   super_cache_scan+0x161/0x1f0
   do_shrink_slab+0x163/0x340
   shrink_slab+0x1d3/0x290
   shrink_node+0x300/0x720
   balance_pgdat+0x35c/0x7a0
   kswapd+0x205/0x410
   ? __pfx_autoremove_wake_function+0x10/0x10
   ? __pfx_kswapd+0x10/0x10
   kthread+0xf0/0x120
   ? __pfx_kthread+0x10/0x10
   ret_from_fork+0x29/0x50
   </TASK>

This happens because when we abort the transaction in the transaction commit path we call invalidate_inode_pages2_range on our block group cache inodes (if we have space cache v1) and any delalloc inodes we may have. The plain invalidate_inode_pages2_range() call passes through GFP_KERNEL, which makes sense in most cases, but not here. Wrap these two invalidate callees with memalloc_nofs_save/memalloc_nofs_restore to make sure we don't end up with the fs reclaim dependency under the transaction dependency.

---
 fs/btrfs/disk-io.c |    9 +++++++++
 1 file changed, 9 insertions(+)

@@ -4535,7 +4535,11 @@ static void btrfs_destroy_delalloc_inode
 		 */
 		inode = igrab(&btrfs_inode->vfs_inode);
 		if (inode) {
+			unsigned int nofs_flag;
+
+			nofs_flag = memalloc_nofs_save();
 			invalidate_inode_pages2(inode->i_mapping);
+			memalloc_nofs_restore(nofs_flag);
 			iput(inode);
 		}
 		spin_lock(&root->delalloc_lock);

@@ -4640,7 +4644,12 @@ static void btrfs_cleanup_bg_io(struct b
 
 	inode = cache->io_ctl.inode;
 	if (inode) {
+		unsigned int nofs_flag;
+
+		nofs_flag = memalloc_nofs_save();
 		invalidate_inode_pages2(inode->i_mapping);
+		memalloc_nofs_restore(nofs_flag);
+
 		BTRFS_I(inode)->generation = 0;
 		cache->io_ctl.inode = NULL;
 		iput(inode);

From: Frank Li <[email protected]>

commit 50a1726b148ff30778cb8a6cf3736130b07c93fd upstream.

In cdns3-gadget.c, 'cdns,on-chip-buff-size' was read using device_property_read_u16(). It resulted in 0 if a 32bit value was used in dts. This commit fixes the dt binding doc to declare it as u16.

---
 Documentation/devicetree/bindings/usb/cdns,usb3.yaml |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -59,7 +59,7 @@ properties:
     description:
       size of memory intended as internal memory for endpoints
       buffers expressed in KB
-    $ref: /schemas/types.yaml#/definitions/uint32
+    $ref: /schemas/types.yaml#/definitions/uint16
 
   cdns,phyrst-a-enable:
     description: Enable resetting of PHY if Rx fail is detected

From: Dave Hansen <[email protected]>

commit ce0b15d11ad837fbacc5356941712218e38a0a83 upstream.

The INVLPG instruction is used to invalidate TLB entries for a specified virtual address. When PCIDs are enabled, INVLPG is supposed to invalidate TLB entries for the specified address for both the current PCID *and* Global entries. (Note: Only kernel mappings set Global=1.)

Unfortunately, some INVLPG implementations can leave Global translations unflushed when PCIDs are enabled.

As a workaround, never enable PCIDs on affected processors.

I expect there to eventually be microcode mitigations to replace this software workaround. However, the exact version numbers where that will happen are not known today. Once the version numbers are set in stone, the processor list can be tweaked to only disable PCIDs on affected processors with affected microcode.

Note: if anyone wants a quick fix that doesn't require patching, just stick 'nopcid' on your kernel command-line.

---
 arch/x86/include/asm/intel-family.h |    5 +++++
 arch/x86/mm/init.c                  |   25 +++++++++++++++++++++++++
 2 files changed, 30 insertions(+)

@@ -98,6 +98,11 @@
 #define	INTEL_FAM6_LAKEFIELD		0x8A
 #define INTEL_FAM6_ALDERLAKE		0x97
 #define INTEL_FAM6_ALDERLAKE_L		0x9A
+#define INTEL_FAM6_ALDERLAKE_N		0xBE
+
+#define INTEL_FAM6_RAPTORLAKE		0xB7
+#define INTEL_FAM6_RAPTORLAKE_P		0xBA
+#define INTEL_FAM6_RAPTORLAKE_S		0xBF
 
 /* "Small Core" Processors (Atom) */
 

@@ -9,6 +9,7 @@
 #include <linux/sched/task.h>
 
 #include <asm/set_memory.h>
+#include <asm/cpu_device_id.h>
 #include <asm/e820/api.h>
 #include <asm/init.h>
 #include <asm/page.h>

@@ -254,6 +255,24 @@ static void __init probe_page_size_mask(
 	}
 }
 
+#define INTEL_MATCH(_model) { .vendor  = X86_VENDOR_INTEL,	\
+			      .family  = 6,			\
+			      .model = _model,			\
+			    }
+/*
+ * INVLPG may not properly flush Global entries
+ * on these CPUs when PCIDs are enabled.
+ */
+static const struct x86_cpu_id invlpg_miss_ids[] = {
+	INTEL_MATCH(INTEL_FAM6_ALDERLAKE   ),
+	INTEL_MATCH(INTEL_FAM6_ALDERLAKE_L ),
+	INTEL_MATCH(INTEL_FAM6_ALDERLAKE_N ),
+	INTEL_MATCH(INTEL_FAM6_RAPTORLAKE  ),
+	INTEL_MATCH(INTEL_FAM6_RAPTORLAKE_P),
+	INTEL_MATCH(INTEL_FAM6_RAPTORLAKE_S),
+	{}
+};
+
 static void setup_pcid(void)
 {
 	if (!IS_ENABLED(CONFIG_X86_64))

@@ -262,6 +281,12 @@ static void setup_pcid(void)
 	if (!boot_cpu_has(X86_FEATURE_PCID))
 		return;
 
+	if (x86_match_cpu(invlpg_miss_ids)) {
+		pr_info("Incomplete global flushes, disabling PCID");
+		setup_clear_cpu_cap(X86_FEATURE_PCID);
+		return;
+	}
+
 	if (boot_cpu_has(X86_FEATURE_PGE)) {
 		/*
 		 * This can't be cr4_set_bits_and_update_boot() -- the

From: Hardik Garg <[email protected]>

Partially backport v6.3 commit 11f75a01448f ("selftests/memfd: add tests for MFD_NOEXEC_SEAL MFD_EXEC") to fix an unknown type name build error. In some systems, the __u64 typedef is not present due to differences in system headers, causing compilation errors like this one:

fuse_test.c:64:8: error: unknown type name '__u64'
64 | static __u64 mfd_assert_get_seals(int fd)

This header includes the __u64 typedef which increases the likelihood of successful compilation on a wider variety of systems.

---
 tools/testing/selftests/memfd/fuse_test.c |    1 +
 1 file changed, 1 insertion(+)

@@ -22,6 +22,7 @@
 #include <linux/falloc.h>
 #include <linux/fcntl.h>
 #include <linux/memfd.h>
+#include <linux/types.h>
 #include <sched.h>
 #include <stdio.h>
 #include <stdlib.h>

From: Helge Deller <[email protected]>

commit 61e150fb310729c98227a5edf6e4a3619edc3702 upstream.

Since at least kernel 6.1, flush_dcache_page() is called with IRQs disabled, e.g. from aio_complete().

But the current implementation for flush_dcache_page() on parisc unintentionally re-enables IRQs, which may lead to deadlocks.

Fix it by using xa_lock_irqsave() and xa_unlock_irqrestore() for the flush_dcache_mmap_*lock() macros instead.

---
 arch/parisc/include/asm/cacheflush.h |    5 +++++
 arch/parisc/kernel/cache.c           |    5 +++--
 2 files changed, 8 insertions(+), 2 deletions(-)

@@ -57,6 +57,11 @@ extern void flush_dcache_page(struct pag
 
 #define flush_dcache_mmap_lock(mapping)		xa_lock_irq(&mapping->i_pages)
 #define flush_dcache_mmap_unlock(mapping)	xa_unlock_irq(&mapping->i_pages)
+#define flush_dcache_mmap_lock_irqsave(mapping, flags)		\
+		xa_lock_irqsave(&mapping->i_pages, flags)
+#define flush_dcache_mmap_unlock_irqrestore(mapping, flags)	\
+		xa_unlock_irqrestore(&mapping->i_pages, flags)
+
 
 #define flush_icache_page(vma,page)	do { 		\
 	flush_kernel_dcache_page(page);			\

@@ -327,6 +327,7 @@ void flush_dcache_page(struct page *page
 	struct vm_area_struct *mpnt;
 	unsigned long offset;
 	unsigned long addr, old_addr = 0;
+	unsigned long flags;
 	pgoff_t pgoff;
 
 	if (mapping && !mapping_mapped(mapping)) {

@@ -346,7 +347,7 @@ void flush_dcache_page(struct page *page
 	 * declared as MAP_PRIVATE or MAP_SHARED), so we only need
 	 * to flush one address here for them all to become coherent */
 
-	flush_dcache_mmap_lock(mapping);
+	flush_dcache_mmap_lock_irqsave(mapping, flags);
 	vma_interval_tree_foreach(mpnt, &mapping->i_mmap, pgoff, pgoff) {
 		offset = (pgoff - mpnt->vm_pgoff) << PAGE_SHIFT;
 		addr = mpnt->vm_start + offset;

@@ -369,7 +370,7 @@ void flush_dcache_page(struct page *page
 			old_addr = addr;
 		}
 	}
-	flush_dcache_mmap_unlock(mapping);
+	flush_dcache_mmap_unlock_irqrestore(mapping, flags);
 }
 EXPORT_SYMBOL(flush_dcache_page);
 

From: Zhang Rui <[email protected]>

commit edc0a2b5957652f4685ef3516f519f84807087db upstream.

Traditionally, all CPUs in a system have identical numbers of SMT siblings. That changes with hybrid processors where some logical CPUs have a sibling and others have none.

Today, the CPU boot code sets the global variable smp_num_siblings when every CPU thread is brought up. The last thread to boot will overwrite it with the number of siblings of *that* thread. That last thread to boot will "win". If the thread is a Pcore, smp_num_siblings == 2. If it is an Ecore, smp_num_siblings == 1.

smp_num_siblings describes if the *system* supports SMT. It should specify the maximum number of SMT threads among all cores.

Ensure that smp_num_siblings represents the system-wide maximum number of siblings by always increasing its value. Never allow it to decrease.

On MeteorLake-P platform, this fixes a problem that the Ecore CPUs are not updated in any cpu sibling map because the system is treated as an UP system when probing Ecore CPUs.

Below shows part of the CPU topology information before and after the fix, for both Pcore and Ecore CPU (cpu0 is Pcore, cpu 12 is Ecore).
...
-/sys/devices/system/cpu/cpu0/topology/package_cpus:000fff -/sys/devices/system/cpu/cpu0/topology/package_cpus_list:0-11 +/sys/devices/system/cpu/cpu0/topology/package_cpus:3fffff +/sys/devices/system/cpu/cpu0/topology/package_cpus_list:0-21
...
-/sys/devices/system/cpu/cpu12/topology/package_cpus:001000 -/sys/devices/system/cpu/cpu12/topology/package_cpus_list:12 +/sys/devices/system/cpu/cpu12/topology/package_cpus:3fffff +/sys/devices/system/cpu/cpu12/topology/package_cpus_list:0-21

Notice that the "before" 'package_cpus_list' has only one CPU. This means that userspace tools like lscpu will see a little laptop like an 11-socket system:

-Core(s) per socket:  1
-Socket(s):           11
+Core(s) per socket:  16
+Socket(s):           1

This is also expected to make the scheduler do rather wonky things too.

[ dhansen: remove CPUID detail from changelog, add end user effects ]

---
 arch/x86/kernel/cpu/topology.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

@@ -79,7 +79,7 @@ int detect_extended_topology_early(struc
 	 * initial apic id, which also represents 32-bit extended x2apic id.
 	 */
 	c->initial_apicid = edx;
-	smp_num_siblings = LEVEL_MAX_SIBLINGS(ebx);
+	smp_num_siblings = max_t(int, smp_num_siblings, LEVEL_MAX_SIBLINGS(ebx));
 #endif
 	return 0;
 }

@@ -109,7 +109,8 @@ int detect_extended_topology(struct cpui
 	 */
 	cpuid_count(leaf, SMT_LEVEL, &eax, &ebx, &ecx, &edx);
 	c->initial_apicid = edx;
-	core_level_siblings = smp_num_siblings = LEVEL_MAX_SIBLINGS(ebx);
+	core_level_siblings = LEVEL_MAX_SIBLINGS(ebx);
+	smp_num_siblings = max_t(int, smp_num_siblings, LEVEL_MAX_SIBLINGS(ebx));
 	core_plus_mask_width = ht_mask_width = BITS_SHIFT_NEXT_LEVEL(eax);
 	die_level_siblings = LEVEL_MAX_SIBLINGS(ebx);
 	pkg_mask_width = die_plus_mask_width = BITS_SHIFT_NEXT_LEVEL(eax);

From: Tetsuo Handa <[email protected]>

commit eb799279fb1f9c63c520fe8c1c41cb9154252db6 upstream.

syzbot is reporting a lockdep warning in fill_pool() because the allocation from debugobjects is using GFP_ATOMIC, which is (__GFP_HIGH | __GFP_KSWAPD_RECLAIM) and therefore tries to wake up kswapd, which acquires kswapd_wait::lock.

Since fill_pool() might be called with arbitrary locks held, fill_pool() should not assume that acquiring kswapd_wait::lock is safe.

Use __GFP_HIGH instead and remove __GFP_NORETRY as it is pointless for !__GFP_DIRECT_RECLAIM allocation.

---
 lib/debugobjects.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -129,7 +129,7 @@ static const char *obj_states[ODEBUG_STA
 
 static void fill_pool(void)
 {
-	gfp_t gfp = GFP_ATOMIC | __GFP_NORETRY | __GFP_NOWARN;
+	gfp_t gfp = __GFP_HIGH | __GFP_NOWARN;
 	struct debug_obj *obj;
 	unsigned long flags;
 

From: Alan Stern <[email protected]>

commit ed9de4ed39875706607fb08118a58344ae6c5f42 upstream.

The syzbot fuzzer detected a problem in the udlfb driver, caused by an endpoint not having the expected type:

usb 1-1: Read EDID byte 0 failed: -71
usb 1-1: Unable to get valid EDID from device/display

------------[ cut here ]------------
usb 1-1: BOGUS urb xfer, pipe 3 != type 1
WARNING: CPU: 0 PID: 9 at drivers/usb/core/urb.c:504 usb_submit_urb+0xed6/0x1880
drivers/usb/core/urb.c:504
Modules linked in:
CPU: 0 PID: 9 Comm: kworker/0:1 Not tainted
6.4.0-rc1-syzkaller-00016-ga4422ff22142 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google
04/28/2023
Workqueue: usb_hub_wq hub_event
RIP: 0010:usb_submit_urb+0xed6/0x1880 drivers/usb/core/urb.c:504
...
Call Trace:
 <TASK>
 dlfb_submit_urb+0x92/0x180 drivers/video/fbdev/udlfb.c:1980
 dlfb_set_video_mode+0x21f0/0x2950 drivers/video/fbdev/udlfb.c:315
 dlfb_ops_set_par+0x2a7/0x8d0 drivers/video/fbdev/udlfb.c:1111
 dlfb_usb_probe+0x149a/0x2710 drivers/video/fbdev/udlfb.c:1743

The current approach for this issue failed to catch the problem because it only checks for the existence of a bulk-OUT endpoint; it doesn't check whether this endpoint is the one that the driver will actually use.

We can fix the problem by instead checking that the endpoint used by the driver does exist and is bulk-OUT.

---
 drivers/video/fbdev/udlfb.c |   13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

@@ -27,6 +27,8 @@
 #include <video/udlfb.h>
 #include "edid.h"
 
+#define OUT_EP_NUM	1	/* The endpoint number we will use */
+
 static const struct fb_fix_screeninfo dlfb_fix = {
 	.id =           "udlfb",
 	.type =         FB_TYPE_PACKED_PIXELS,

@@ -1651,7 +1653,7 @@ static int dlfb_usb_probe(struct usb_int
 	struct fb_info *info;
 	int retval;
 	struct usb_device *usbdev = interface_to_usbdev(intf);
-	struct usb_endpoint_descriptor *out;
+	static u8 out_ep[] = {OUT_EP_NUM + USB_DIR_OUT, 0};
 
 	/* usb initialization */
 	dlfb = kzalloc(sizeof(*dlfb), GFP_KERNEL);

@@ -1665,9 +1667,9 @@ static int dlfb_usb_probe(struct usb_int
 	dlfb->udev = usb_get_dev(usbdev);
 	usb_set_intfdata(intf, dlfb);
 
-	retval = usb_find_common_endpoints(intf->cur_altsetting, NULL, &out, NULL, NULL);
-	if (retval) {
-		dev_err(&intf->dev, "Device should have at lease 1 bulk endpoint!\n");
+	if (!usb_check_bulk_endpoints(intf, out_ep)) {
+		dev_err(&intf->dev, "Invalid DisplayLink device!\n");
+		retval = -EINVAL;
 		goto error;
 	}
 

@@ -1926,7 +1928,8 @@ retry:
 		}
 
 		/* urb->transfer_buffer_length set to actual before submit */
-		usb_fill_bulk_urb(urb, dlfb->udev, usb_sndbulkpipe(dlfb->udev, 1),
+		usb_fill_bulk_urb(urb, dlfb->udev,
+			usb_sndbulkpipe(dlfb->udev, OUT_EP_NUM),
 			buf, size, dlfb_urb_completion, unode);
 		urb->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
 

From: Taehee Yoo <[email protected]>

commit ae9b15fbe63447bc1d3bba3769f409d17ca6fdf6 upstream.

When the virtual interface's feature is updated, it synchronizes the updated feature for its own lower interface.
This propagation logic should be worked as the iteration, not recursively. But it works recursively due to the netdev notification unexpectedly. This problem occurs when it disables LRO only for the team and bonding interface type.

       team0
         |
  +------+------+-----+-----+
  |      |      |     |     |
team1  team2  team3  ...  team200

If team0's LRO feature is updated, it generates the NETDEV_FEAT_CHANGE event to its own lower interfaces(team1 ~ team200).
It is worked by netdev_sync_lower_features().
So, the NETDEV_FEAT_CHANGE notification logic of each lower interface work iteratively.
But generated NETDEV_FEAT_CHANGE event is also sent to the upper interface too.
upper interface(team0) generates the NETDEV_FEAT_CHANGE event for its own lower interfaces again.
lower and upper interfaces receive this event and generate this event again and again.
So, the stack overflow occurs.

But it is not the infinite loop issue.
Because the netdev_sync_lower_features() updates features before generating the NETDEV_FEAT_CHANGE event.
Already synchronized lower interfaces skip notification logic. So, it is just the problem that iteration logic is changed to the recursive unexpectedly due to the notification mechanism.

Reproducer:

ip link add team0 type team
ethtool -K team0 lro on
for i in {1..200}
do
ip link add team$i master team0 type team
ethtool -K team$i lro on
done

ethtool -K team0 lro off

In order to fix it, the notifier_ctx member of bonding/team is introduced.

---
 drivers/net/bonding/bond_main.c |    8 +++++++-
 drivers/net/team/team.c         |    7 ++++++-
 include/linux/if_team.h         |    1 +
 include/net/bonding.h           |    1 +
 4 files changed, 15 insertions(+), 2 deletions(-)

@@ -3537,7 +3537,11 @@ static int bond_slave_netdev_event(unsig
 		unblock_netpoll_tx();
 		break;
 	case NETDEV_FEAT_CHANGE:
-		bond_compute_features(bond);
+		if (!bond->notifier_ctx) {
+			bond->notifier_ctx = true;
+			bond_compute_features(bond);
+			bond->notifier_ctx = false;
+		}
 		break;
 	case NETDEV_RESEND_IGMP:
 		/* Propagate to master device */

@@ -5360,6 +5364,8 @@ static int bond_init(struct net_device *
 	if (!bond->wq)
 		return -ENOMEM;
 
+	bond->notifier_ctx = false;
+
 	spin_lock_init(&bond->stats_lock);
 	netdev_lockdep_set_classes(bond_dev);
 

@@ -1624,6 +1624,7 @@ static int team_init(struct net_device *
 
 	team->dev = dev;
 	team_set_no_mode(team);
+	team->notifier_ctx = false;
 
 	team->pcpu_stats = netdev_alloc_pcpu_stats(struct team_pcpu_stats);
 	if (!team->pcpu_stats)

@@ -3016,7 +3017,11 @@ static int team_device_event(struct noti
 		team_del_slave(port->team->dev, dev);
 		break;
 	case NETDEV_FEAT_CHANGE:
-		team_compute_features(port->team);
+		if (!port->team->notifier_ctx) {
+			port->team->notifier_ctx = true;
+			team_compute_features(port->team);
+			port->team->notifier_ctx = false;
+		}
 		break;
 	case NETDEV_PRECHANGEMTU:
 		/* Forbid to change mtu of underlaying device */

@@ -208,6 +208,7 @@ struct team {
 	bool queue_override_enabled;
 	struct list_head *qom_lists; /* array of queue override mapping lists */
 	bool port_mtu_change_allowed;
+	bool notifier_ctx;
 	struct {
 		unsigned int count;
 		unsigned int interval; /* in ms */

@@ -216,6 +216,7 @@ struct bonding {
 	struct   bond_up_slave __rcu *usable_slaves;
 	struct   bond_up_slave __rcu *all_slaves;
 	bool     force_primary;
+	bool     notifier_ctx;
 	s32      slave_cnt; /* never change this value outside the attach/detach wrappers */
 	int     (*recv_probe)(const struct sk_buff *, struct bonding *,
 			      struct slave *);

From: Kuniyuki Iwashima <[email protected]>

commit ad42a35bdfc6d3c0fc4cb4027d7b2757ce665665 upstream.

syzbot reported [0] a null-ptr-deref in sk_get_rmem0() while using IPPROTO_UDPLITE (0x88):

14:25:52 executing program 1:
r0 = socket$inet6(0xa, 0x80002, 0x88)

We had a similar report [1] for probably sk_memory_allocated_add() in __sk_mem_raise_allocated(), and commit c915fe13cbaa ("udplite: fix NULL pointer dereference") fixed it by setting .memory_allocated for udplite_prot and udplitev6_prot.

To fix the variant, we need to set either .sysctl_wmem_offset or .sysctl_rmem.

Now UDP and UDPLITE share the same value for .memory_allocated, so we use the same .sysctl_wmem_offset for UDP and UDPLITE.

[0]:

general protection fault, probably for non-canonical address 0xdffffc0000000000: 0000 [#1] PREEMPT SMP KASAN
KASAN: null-ptr-deref in range [0x0000000000000000-0x0000000000000007]
CPU: 0 PID: 6829 Comm: syz-executor.1 Not tainted 6.4.0-rc2-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 04/28/2023
RIP: 0010:sk_get_rmem0 include/net/sock.h:2907 [inline]
RIP: 0010:__sk_mem_raise_allocated+0x806/0x17a0 net/core/sock.c:3006
Code: c1 ea 03 80 3c 02 00 0f 85 23 0f 00 00 48 8b 44 24 08 48 8b 98 38 01 00 00 48 b8 00 00 00 00 00 fc ff df 48 89 da 48 c1 ea 03 <0f> b6 14 02 48 89 d8 83 e0 07 83 c0 03 38 d0 0f 8d 6f 0a 00 00 8b
RSP: 0018:ffffc90005d7f450 EFLAGS: 00010246
RAX: dffffc0000000000 RBX: 0000000000000000 RCX: ffffc90004d92000
RDX: 0000000000000000 RSI: ffffffff88066482 RDI: ffffffff8e2ccbb8
RBP: ffff8880173f7000 R08: 0000000000000005 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000030000
R13: 0000000000000001 R14: 0000000000000340 R15: 0000000000000001
FS:  0000000000000000(0000) GS:ffff8880b9800000(0063) knlGS:00000000f7f1cb40
CS:  0010 DS: 002b ES: 002b CR0: 0000000080050033
CR2: 000000002e82f000 CR3: 0000000034ff0000 CR4: 00000000003506f0
Call Trace:
 <TASK>
 __sk_mem_schedule+0x6c/0xe0 net/core/sock.c:3077
 udp_rmem_schedule net/ipv4/udp.c:1539 [inline]
 __udp_enqueue_schedule_skb+0x776/0xb30 net/ipv4/udp.c:1581
 __udpv6_queue_rcv_skb net/ipv6/udp.c:666 [inline]
 udpv6_queue_rcv_one_skb+0xc39/0x16c0 net/ipv6/udp.c:775
 udpv6_queue_rcv_skb+0x194/0xa10 net/ipv6/udp.c:793
 __udp6_lib_mcast_deliver net/ipv6/udp.c:906 [inline]
 __udp6_lib_rcv+0x1bda/0x2bd0 net/ipv6/udp.c:1013
 ip6_protocol_deliver_rcu+0x2e7/0x1250 net/ipv6/ip6_input.c:437
 ip6_input_finish+0x150/0x2f0 net/ipv6/ip6_input.c:482
 NF_HOOK include/linux/netfilter.h:303 [inline]
 NF_HOOK include/linux/netfilter.h:297 [inline]
 ip6_input+0xa0/0xd0 net/ipv6/ip6_input.c:491
 ip6_mc_input+0x40b/0xf50 net/ipv6/ip6_input.c:585
 dst_input include/net/dst.h:468 [inline]
 ip6_rcv_finish net/ipv6/ip6_input.c:79 [inline]
 NF_HOOK include/linux/netfilter.h:303 [inline]
 NF_HOOK include/linux/netfilter.h:297 [inline]
 ipv6_rcv+0x250/0x380 net/ipv6/ip6_input.c:309
 __netif_receive_skb_one_core+0x114/0x180 net/core/dev.c:5491
 __netif_receive_skb+0x1f/0x1c0 net/core/dev.c:5605
 netif_receive_skb_internal net/core/dev.c:5691 [inline]
 netif_receive_skb+0x133/0x7a0 net/core/dev.c:5750
 tun_rx_batched+0x4b3/0x7a0 drivers/net/tun.c:1553
 tun_get_user+0x2452/0x39c0 drivers/net/tun.c:1989
 tun_chr_write_iter+0xdf/0x200 drivers/net/tun.c:2035
 call_write_iter include/linux/fs.h:1868 [inline]
 new_sync_write fs/read_write.c:491 [inline]
 vfs_write+0x945/0xd50 fs/read_write.c:584
 ksys_write+0x12b/0x250 fs/read_write.c:637
 do_syscall_32_irqs_on arch/x86/entry/common.c:112 [inline]
 __do_fast_syscall_32+0x65/0xf0 arch/x86/entry/common.c:178
 do_fast_syscall_32+0x33/0x70 arch/x86/entry/common.c:203
 entry_SYSENTER_compat_after_hwframe+0x70/0x82
RIP: 0023:0xf7f21579
Code: b8 01 10 06 03 74 b4 01 10 07 03 74 b0 01 10 08 03 74 d8 01 00 00 00 00 00 00 00 00 00 00 00 00 00 51 52 55 89 e5 0f 34 cd 80 <5d> 5a 59 c3 90 90 90 90 8d b4 26 00 00 00 00 8d b4 26 00 00 00 00
RSP: 002b:00000000f7f1c590 EFLAGS: 00000282 ORIG_RAX: 0000000000000004
RAX: ffffffffffffffda RBX: 00000000000000c8 RCX: 0000000020000040
RDX: 0000000000000083 RSI: 00000000f734e000 RDI: 0000000000000000
RBP: 0000000000000000 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000296 R12: 0000000000000000
R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
 </TASK>
Modules linked in:

---
 net/ipv4/udplite.c |    2 ++
 net/ipv6/udplite.c |    2 ++
 2 files changed, 4 insertions(+)

@@ -62,6 +62,8 @@ struct proto 	udplite_prot = {
 	.get_port	   = udp_v4_get_port,
 	.memory_allocated  = &udp_memory_allocated,
 	.sysctl_mem	   = sysctl_udp_mem,
+	.sysctl_wmem_offset = offsetof(struct net, ipv4.sysctl_udp_wmem_min),
+	.sysctl_rmem_offset = offsetof(struct net, ipv4.sysctl_udp_rmem_min),
 	.obj_size	   = sizeof(struct udp_sock),
 	.h.udp_table	   = &udplite_table,
 };

@@ -57,6 +57,8 @@ struct proto udplitev6_prot = {
 	.get_port	   = udp_v6_get_port,
 	.memory_allocated  = &udp_memory_allocated,
 	.sysctl_mem	   = sysctl_udp_mem,
+	.sysctl_wmem_offset = offsetof(struct net, ipv4.sysctl_udp_wmem_min),
+	.sysctl_rmem_offset = offsetof(struct net, ipv4.sysctl_udp_rmem_min),
 	.obj_size	   = sizeof(struct udp6_sock),
 	.h.udp_table	   = &udplite_table,
 };

From: Alan Stern <[email protected]>

commit 13890626501ffda22b18213ddaf7930473da5792 upstream.

Many of the older USB drivers in the Linux USB stack were written based simply on a vendor's device specification. They use the endpoint information in the spec and assume these endpoints will always be present, with the properties listed, in any device matching the given vendor and product IDs.

While that may have been true back then, with spoofing and fuzzing it is not true any more. More and more we are finding that those old drivers need to perform at least a minimum of checking before they try to use any endpoint other than ep0.

To make this checking as simple as possible, we now add a couple of utility routines to the USB core. usb_check_bulk_endpoints() and usb_check_int_endpoints() take an interface pointer together with a list of endpoint addresses (numbers and directions). They check that the interface's current alternate setting includes endpoints with those addresses and that each of these endpoints has the right type:
bulk or interrupt, respectively.

Although we already have usb_find_common_endpoints() and related routines meant for a similar purpose, they are not well suited for this kind of checking. Those routines find endpoints of various kinds, but only one (either the first or the last) of each kind, and they don't verify that the endpoints' addresses agree with what the caller expects.

In theory the new routines could be more general: They could take a particular altsetting as their argument instead of always using the interface's current altsetting. In practice I think this won't matter too much; multiple altsettings tend to be used for transferring media (audio or visual) over isochronous endpoints, not bulk or interrupt. Drivers for such devices will generally require more sophisticated checking than these simplistic routines provide.

---
 drivers/usb/core/usb.c |   76 +++++++++++++++++++++++++++++++++++++++++++++++++
 include/linux/usb.h    |    5 +++
 2 files changed, 81 insertions(+)

@@ -208,6 +208,82 @@ int usb_find_common_endpoints_reverse(st
 EXPORT_SYMBOL_GPL(usb_find_common_endpoints_reverse);
 
 /**
+ * usb_find_endpoint() - Given an endpoint address, search for the endpoint's
+ * usb_host_endpoint structure in an interface's current altsetting.
+ * @intf: the interface whose current altsetting should be searched
+ * @ep_addr: the endpoint address (number and direction) to find
+ *
+ * Search the altsetting's list of endpoints for one with the specified address.
+ *
+ * Return: Pointer to the usb_host_endpoint if found, %NULL otherwise.
+ */
+static const struct usb_host_endpoint *usb_find_endpoint(
+		const struct usb_interface *intf, unsigned int ep_addr)
+{
+	int n;
+	const struct usb_host_endpoint *ep;
+
+	n = intf->cur_altsetting->desc.bNumEndpoints;
+	ep = intf->cur_altsetting->endpoint;
+	for (; n > 0; (--n, ++ep)) {
+		if (ep->desc.bEndpointAddress == ep_addr)
+			return ep;
+	}
+	return NULL;
+}
+
+/**
+ * usb_check_bulk_endpoints - Check whether an interface's current altsetting
+ * contains a set of bulk endpoints with the given addresses.
+ * @intf: the interface whose current altsetting should be searched
+ * @ep_addrs: 0-terminated array of the endpoint addresses (number and
+ * direction) to look for
+ *
+ * Search for endpoints with the specified addresses and check their types.
+ *
+ * Return: %true if all the endpoints are found and are bulk, %false otherwise.
+ */
+bool usb_check_bulk_endpoints(
+		const struct usb_interface *intf, const u8 *ep_addrs)
+{
+	const struct usb_host_endpoint *ep;
+
+	for (; *ep_addrs; ++ep_addrs) {
+		ep = usb_find_endpoint(intf, *ep_addrs);
+		if (!ep || !usb_endpoint_xfer_bulk(&ep->desc))
+			return false;
+	}
+	return true;
+}
+EXPORT_SYMBOL_GPL(usb_check_bulk_endpoints);
+
+/**
+ * usb_check_int_endpoints - Check whether an interface's current altsetting
+ * contains a set of interrupt endpoints with the given addresses.
+ * @intf: the interface whose current altsetting should be searched
+ * @ep_addrs: 0-terminated array of the endpoint addresses (number and
+ * direction) to look for
+ *
+ * Search for endpoints with the specified addresses and check their types.
+ *
+ * Return: %true if all the endpoints are found and are interrupt,
+ * %false otherwise.
+ */
+bool usb_check_int_endpoints(
+		const struct usb_interface *intf, const u8 *ep_addrs)
+{
+	const struct usb_host_endpoint *ep;
+
+	for (; *ep_addrs; ++ep_addrs) {
+		ep = usb_find_endpoint(intf, *ep_addrs);
+		if (!ep || !usb_endpoint_xfer_int(&ep->desc))
+			return false;
+	}
+	return true;
+}
+EXPORT_SYMBOL_GPL(usb_check_int_endpoints);
+
+/**
  * usb_find_alt_setting() - Given a configuration, find the alternate setting
  * for the given interface.
  * @config: the configuration to search (not necessarily the current config).

@@ -279,6 +279,11 @@ void usb_put_intf(struct usb_interface *
 #define USB_MAXINTERFACES	32
 #define USB_MAXIADS		(USB_MAXINTERFACES/2)
 
+bool usb_check_bulk_endpoints(
+		const struct usb_interface *intf, const u8 *ep_addrs);
+bool usb_check_int_endpoints(
+		const struct usb_interface *intf, const u8 *ep_addrs);
+
 /*
  * USB Resume Timer: Every Host controller driver should drive the resume
  * signalling on the bus for the amount of time defined by this macro.

From: Alan Stern <[email protected]>

commit df05a9b05e466a46725564528b277d0c570d0104 upstream.

The syzbot fuzzer was able to provoke a WARNING from the sisusbvga driver:

------------[ cut here ]------------
usb 1-1: BOGUS urb xfer, pipe 3 != type 1
WARNING: CPU: 1 PID: 26 at drivers/usb/core/urb.c:504 usb_submit_urb+0xed6/0x1880 drivers/usb/core/urb.c:504
Modules linked in:
CPU: 1 PID: 26 Comm: kworker/1:1 Not tainted 6.2.0-rc5-syzkaller-00199-g5af6ce704936 #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/12/2023
Workqueue: usb_hub_wq hub_event
RIP: 0010:usb_submit_urb+0xed6/0x1880 drivers/usb/core/urb.c:504
Code: 7c 24 18 e8 6c 50 80 fb 48 8b 7c 24 18 e8 62 1a 01 ff 41 89 d8 44 89 e1 4c 89 ea 48 89 c6 48 c7 c7 60 b1 fa 8a e8 84 b0 be 03 <0f> 0b e9 58 f8 ff ff e8 3e 50 80 fb 48 81 c5 c0 05 00 00 e9 84 f7
RSP: 0018:ffffc90000a1ed18 EFLAGS: 00010282
RAX: 0000000000000000 RBX: 0000000000000001 RCX: 0000000000000000
RDX: ffff888012783a80 RSI: ffffffff816680ec RDI: fffff52000143d95
RBP: ffff888079020000 R08: 0000000000000005 R09: 0000000000000000
R10: 0000000080000000 R11: 0000000000000000 R12: 0000000000000003
R13: ffff888017d33370 R14: 0000000000000003 R15: ffff888021213600
FS:  0000000000000000(0000) GS:ffff8880b9900000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00005592753a60b0 CR3: 0000000022899000 CR4: 00000000003506e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <TASK>
 sisusb_bulkout_msg drivers/usb/misc/sisusbvga/sisusbvga.c:224 [inline]
 sisusb_send_bulk_msg.constprop.0+0x904/0x1230 drivers/usb/misc/sisusbvga/sisusbvga.c:379
 sisusb_send_bridge_packet drivers/usb/misc/sisusbvga/sisusbvga.c:567 [inline]
 sisusb_do_init_gfxdevice drivers/usb/misc/sisusbvga/sisusbvga.c:2077 [inline]
 sisusb_init_gfxdevice+0x87b/0x4000 drivers/usb/misc/sisusbvga/sisusbvga.c:2177
 sisusb_probe+0x9cd/0xbe2 drivers/usb/misc/sisusbvga/sisusbvga.c:2869

...

The problem was caused by the fact that the driver does not check whether the endpoints it uses are actually present and have the appropriate types. This can be fixed by adding a simple check of the endpoints.

---
 drivers/usb/misc/sisusbvga/sisusb.c |   14 ++++++++++++++
 1 file changed, 14 insertions(+)

@@ -3014,6 +3014,20 @@ static int sisusb_probe(struct usb_inter
 	struct usb_device *dev = interface_to_usbdev(intf);
 	struct sisusb_usb_data *sisusb;
 	int retval = 0, i;
+	static const u8 ep_addresses[] = {
+		SISUSB_EP_GFX_IN | USB_DIR_IN,
+		SISUSB_EP_GFX_OUT | USB_DIR_OUT,
+		SISUSB_EP_GFX_BULK_OUT | USB_DIR_OUT,
+		SISUSB_EP_GFX_LBULK_OUT | USB_DIR_OUT,
+		SISUSB_EP_BRIDGE_IN | USB_DIR_IN,
+		SISUSB_EP_BRIDGE_OUT | USB_DIR_OUT,
+		0};
+
+	/* Are the expected endpoints present? */
+	if (!usb_check_bulk_endpoints(intf, ep_addresses)) {
+		dev_err(&intf->dev, "Invalid USB2VGA device\n");
+		return -EINVAL;
+	}
 
 	dev_info(&dev->dev, "USB2VGA dongle found at address %d\n",
 			dev->devnum);

From: Alan Stern <[email protected]>

commit 76e31045ba030e94e72105c01b2e98f543d175ac upstream.

The syzbot fuzzer was able to provoke a WARNING from the radio-shark2 driver:

------------[ cut here ]------------
usb 1-1: BOGUS urb xfer, pipe 1 != type 3
WARNING: CPU: 0 PID: 3271 at drivers/usb/core/urb.c:504 usb_submit_urb+0xed2/0x1880 drivers/usb/core/urb.c:504
Modules linked in:
CPU: 0 PID: 3271 Comm: kworker/0:3 Not tainted 6.1.0-rc4-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 10/26/2022
Workqueue: usb_hub_wq hub_event
RIP: 0010:usb_submit_urb+0xed2/0x1880 drivers/usb/core/urb.c:504
Code: 7c 24 18 e8 00 36 ea fb 48 8b 7c 24 18 e8 36 1c 02 ff 41 89 d8 44 89 e1 4c 89 ea 48 89 c6 48 c7 c7 a0 b6 90 8a e8 9a 29 b8 03 <0f> 0b e9 58 f8 ff ff e8 d2 35 ea fb 48 81 c5 c0 05 00 00 e9 84 f7
RSP: 0018:ffffc90003876dd0 EFLAGS: 00010282
RAX: 0000000000000000 RBX: 0000000000000003 RCX: 0000000000000000
RDX: ffff8880750b0040 RSI: ffffffff816152b8 RDI: fffff5200070edac
RBP: ffff8880172d81e0 R08: 0000000000000005 R09: 0000000000000000
R10: 0000000080000000 R11: 0000000000000000 R12: 0000000000000001
R13: ffff8880285c5040 R14: 0000000000000002 R15: ffff888017158200
FS:  0000000000000000(0000) GS:ffff8880b9a00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00007ffe03235b90 CR3: 000000000bc8e000 CR4: 00000000003506f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 <TASK>
 usb_start_wait_urb+0x101/0x4b0 drivers/usb/core/message.c:58
 usb_bulk_msg+0x226/0x550 drivers/usb/core/message.c:387
 shark_write_reg+0x1ff/0x2e0 drivers/media/radio/radio-shark2.c:88

...

The problem was caused by the fact that the driver does not check whether the endpoints it uses are actually present and have the appropriate types. This can be fixed by adding a simple check of these endpoints (and similarly for the radio-shark driver).

---
 drivers/media/radio/radio-shark.c  |   10 ++++++++++
 drivers/media/radio/radio-shark2.c |   10 ++++++++++
 2 files changed, 20 insertions(+)

@@ -316,6 +316,16 @@ static int usb_shark_probe(struct usb_in
 {
 	struct shark_device *shark;
 	int retval = -ENOMEM;
+	static const u8 ep_addresses[] = {
+		SHARK_IN_EP | USB_DIR_IN,
+		SHARK_OUT_EP | USB_DIR_OUT,
+		0};
+
+	/* Are the expected endpoints present? */
+	if (!usb_check_int_endpoints(intf, ep_addresses)) {
+		dev_err(&intf->dev, "Invalid radioSHARK device\n");
+		return -EINVAL;
+	}
 
 	shark = kzalloc(sizeof(struct shark_device), GFP_KERNEL);
 	if (!shark)

@@ -282,6 +282,16 @@ static int usb_shark_probe(struct usb_in
 {
 	struct shark_device *shark;
 	int retval = -ENOMEM;
+	static const u8 ep_addresses[] = {
+		SHARK_IN_EP | USB_DIR_IN,
+		SHARK_OUT_EP | USB_DIR_OUT,
+		0};
+
+	/* Are the expected endpoints present? */
+	if (!usb_check_int_endpoints(intf, ep_addresses)) {
+		dev_err(&intf->dev, "Invalid radioSHARK2 device\n");
+		return -EINVAL;
+	}
 
 	shark = kzalloc(sizeof(struct shark_device), GFP_KERNEL);
 	if (!shark)

From: Pratyush Yadav <[email protected]>

commit 8a02fb71d7192ff1a9a47c9d937624966c6e09af upstream.

Commit 50749f2dd685 ("tcp/udp: Fix memleaks of sk and zerocopy skbs with TX timestamp.") added a call to skb_orphan_frags_rx() to fix leaks with zerocopy skbs. But it ended up adding a leak of its own. When skb_orphan_frags_rx() fails, the function just returns, leaking the skb it just cloned. Free it before returning.

This bug was discovered and resolved using Coverity Static Analysis Security Testing (SAST) by Synopsys, Inc.

---
 net/core/skbuff.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

@@ -4751,8 +4751,10 @@ void __skb_tstamp_tx(struct sk_buff *ori
 	} else {
 		skb = skb_clone(orig_skb, GFP_ATOMIC);
 
-		if (skb_orphan_frags_rx(skb, GFP_ATOMIC))
+		if (skb_orphan_frags_rx(skb, GFP_ATOMIC)) {
+			kfree_skb(skb);
 			return;
+		}
 	}
 	if (!skb)
 		return;

From: Po-Hsu Lin <[email protected]>

commit d226b1df361988f885c298737d6019c863a25f26 upstream.

In the end of the test, there will be an error message induced by the ip netns del ns1 command in cleanup()

Tests passed: 201
Tests failed: 0
Cannot remove namespace file "/run/netns/ns1": No such file or directory

This can even be reproduced with just ./fib_tests.sh -h as we're calling cleanup() on exit.

Redirect the error message to /dev/null to mute it.

V2: Update commit message and fixes tag.
V3: resubmit due to missing netdev ML in V2

---
 tools/testing/selftests/net/fib_tests.sh |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -68,7 +68,7 @@ setup()
 cleanup()
 {
 	$IP link del dev dummy0 &> /dev/null
-	ip netns del ns1
+	ip netns del ns1 &> /dev/null
 	ip netns del ns2 &> /dev/null
 }
 

From: Sunil Goutham <[email protected]>

commit de678ca38861f2eb58814048076dcf95ed1b5bf9 upstream.

HW adds segment size to the payload length
in the IPv6 header. Fix payload length to
just TCP header length instead of 'TCP header
size + IPv6 header size'.

---
 drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c |    4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

@@ -526,9 +526,7 @@ static void otx2_sqe_add_ext(struct otx2
 				htons(ext->lso_sb - skb_network_offset(skb));
 		} else if (skb_shinfo(skb)->gso_type & SKB_GSO_TCPV6) {
 			ext->lso_format = pfvf->hw.lso_tsov6_idx;
-
-			ipv6_hdr(skb)->payload_len =
-				htons(ext->lso_sb - skb_network_offset(skb));
+			ipv6_hdr(skb)->payload_len = htons(tcp_hdrlen(skb));
 		} else if (skb_shinfo(skb)->gso_type & SKB_GSO_UDP_L4) {
 			__be16 l3_proto = vlan_get_protocol(skb);
 			struct udphdr *udph = udp_hdr(skb);

From: Will Deacon <[email protected]>

commit 0613d8ca9ab382caabe9ed2dceb429e9781e443f upstream.

A narrow load from a 64-bit context field results in a 64-bit load followed potentially by a 64-bit right-shift and then a bitwise AND operation to extract the relevant data.

In the case of a 32-bit access, an immediate mask of 0xffffffff is used to construct a 64-bit BPP_AND operation which then sign-extends the mask value and effectively acts as a glorified no-op. For example:

0: 61 10 00 00 00 00 00 00 r0 = *(u32 *)(r1 + 0)

results in the following code generation for a 64-bit field:

	ldr	x7, [x7]	// 64-bit load
	mov	x10, #0xffffffffffffffff
	and	x7, x7, x10

Fix the mask generation so that narrow loads always perform a 32-bit AND operation:

	ldr	x7, [x7]	// 64-bit load
	mov	w10, #0xffffffff
	and	w7, w7, w10

---
 kernel/bpf/verifier.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -11146,7 +11146,7 @@ static int convert_ctx_accesses(struct b
 					insn_buf[cnt++] = BPF_ALU64_IMM(BPF_RSH,
 									insn->dst_reg,
 									shift);
-				insn_buf[cnt++] = BPF_ALU64_IMM(BPF_AND, insn->dst_reg,
+				insn_buf[cnt++] = BPF_ALU32_IMM(BPF_AND, insn->dst_reg,
 								(1ULL << size * 8) - 1);
 			}
 		}

From: Gavrilov Ilia <[email protected]>

commit 878ecb0897f4737a4c9401f3523fd49589025671 upstream.

optlen is fetched without checking whether there is more than one byte to parse. It can lead to out-of-bounds access.

Found by InfoTeCS on behalf of Linux Verification Center (linuxtesting.org) with SVACE.

---
 net/ipv6/exthdrs_core.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -143,6 +143,8 @@ int ipv6_find_tlv(const struct sk_buff *
 			optlen = 1;
 			break;
 		default:
+			if (len < 2)
+				goto bad;
 			optlen = nh[offset + 1] + 2;
 			if (optlen > len)
 				goto bad;

From: Hans de Goede <[email protected]>

commit e4484643991e0f6b89060092563f0dbab9450cbb upstream.

When a battery's status changes from charging to full then the charging-blink-full-solid trigger tries to change the LED from blinking to solid/on.

As is documented in include/linux/leds.h to deactivate blinking / to make the LED solid a LED_OFF must be send:

"""
* Deactivate blinking again when the brightness is set to LED_OFF
* via the brightness_set() callback.
"""

led_set_brighness() calls with a brightness value other then 0 / LED_OFF merely change the brightness of the LED in its on state while it is blinking.

So power_supply_update_bat_leds() must first send a LED_OFF event before the LED_FULL to disable blinking.

---
 drivers/power/supply/power_supply_leds.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

@@ -34,8 +34,9 @@ static void power_supply_update_bat_leds
 		led_trigger_event(psy->charging_full_trig, LED_FULL);
 		led_trigger_event(psy->charging_trig, LED_OFF);
 		led_trigger_event(psy->full_trig, LED_FULL);
-		led_trigger_event(psy->charging_blink_full_solid_trig,
-			LED_FULL);
+		/* Going from blink to LED on requires a LED_OFF event to stop blink */
+		led_trigger_event(psy->charging_blink_full_solid_trig, LED_OFF);
+		led_trigger_event(psy->charging_blink_full_solid_trig, LED_FULL);
 		break;
 	case POWER_SUPPLY_STATUS_CHARGING:
 		led_trigger_event(psy->charging_full_trig, LED_FULL);

From: Hans de Goede <[email protected]>

commit 5c34c0aef185dcd10881847b9ebf20046aa77cb4 upstream.

bq27xxx_battery_update() assumes / requires that it is only run once, not multiple times at the same time. But there are 3 possible callers:

1. bq27xxx_battery_poll() delayed_work item handler
2. bq27xxx_battery_irq_handler_thread() I2C IRQ handler
3. bq27xxx_battery_setup()

And there is no protection against these racing with each other, fix this race condition by making all callers take di->lock:

- Rename bq27xxx_battery_update() to bq27xxx_battery_update_unlocked()

- Add new bq27xxx_battery_update() which takes di->lock and then calls bq27xxx_battery_update_unlocked()

- Make stale cache check code in bq27xxx_battery_get_property(), which already takes di->lock directly to check the jiffies, call bq27xxx_battery_update_unlocked() instead of messing with the delayed_work item

- Make bq27xxx_battery_update_unlocked() mod the delayed-work item so that the next poll is delayed to poll_interval milliseconds after the last update independent of the source of the update

---
 drivers/power/supply/bq27xxx_battery.c |   21 +++++++++++++--------
 1 file changed, 13 insertions(+), 8 deletions(-)

@@ -1681,7 +1681,7 @@ static int bq27xxx_battery_read_health(s
 	return POWER_SUPPLY_HEALTH_GOOD;
 }
 
-void bq27xxx_battery_update(struct bq27xxx_device_info *di)
+static void bq27xxx_battery_update_unlocked(struct bq27xxx_device_info *di)
 {
 	struct bq27xxx_reg_cache cache = {0, };
 	bool has_ci_flag = di->opts & BQ27XXX_O_HAS_CI;

@@ -1732,6 +1732,16 @@ void bq27xxx_battery_update(struct bq27x
 		di->cache = cache;
 
 	di->last_update = jiffies;
+
+	if (poll_interval > 0)
+		mod_delayed_work(system_wq, &di->work, poll_interval * HZ);
+}
+
+void bq27xxx_battery_update(struct bq27xxx_device_info *di)
+{
+	mutex_lock(&di->lock);
+	bq27xxx_battery_update_unlocked(di);
+	mutex_unlock(&di->lock);
 }
 EXPORT_SYMBOL_GPL(bq27xxx_battery_update);
 

@@ -1742,9 +1752,6 @@ static void bq27xxx_battery_poll(struct
 				     work.work);
 
 	bq27xxx_battery_update(di);
-
-	if (poll_interval > 0)
-		schedule_delayed_work(&di->work, poll_interval * HZ);
 }
 
 /*

@@ -1919,10 +1926,8 @@ static int bq27xxx_battery_get_property(
 	struct bq27xxx_device_info *di = power_supply_get_drvdata(psy);
 
 	mutex_lock(&di->lock);
-	if (time_is_before_jiffies(di->last_update + 5 * HZ)) {
-		cancel_delayed_work_sync(&di->work);
-		bq27xxx_battery_poll(&di->work.work);
-	}
+	if (time_is_before_jiffies(di->last_update + 5 * HZ))
+		bq27xxx_battery_update_unlocked(di);
 	mutex_unlock(&di->lock);
 
 	if (psp != POWER_SUPPLY_PROP_PRESENT && di->cache.flags < 0)

From: Hans de Goede <[email protected]>

commit 444ff00734f3878cd54ddd1ed5e2e6dbea9326d5 upstream.

devm_request_threaded_irq() requested IRQs are only free-ed after the driver's remove function has ran. So the IRQ could trigger and call bq27xxx_battery_update() after bq27xxx_battery_teardown() has already run.

Switch to explicitly free-ing the IRQ in bq27xxx_battery_i2c_remove() to fix this.

---
 drivers/power/supply/bq27xxx_battery_i2c.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -179,7 +179,7 @@ static int bq27xxx_battery_i2c_probe(str
 	i2c_set_clientdata(client, di);
 
 	if (client->irq) {
-		ret = devm_request_threaded_irq(&client->dev, client->irq,
+		ret = request_threaded_irq(client->irq,
 				NULL, bq27xxx_battery_irq_handler_thread,
 				IRQF_ONESHOT,
 				di->name, di);

@@ -209,6 +209,7 @@ static int bq27xxx_battery_i2c_remove(st
 {
 	struct bq27xxx_device_info *di = i2c_get_clientdata(client);
 
+	free_irq(client->irq, di);
 	bq27xxx_battery_teardown(di);
 
 	mutex_lock(&battery_mutex);

From: Hans de Goede <[email protected]>

commit c00bc80462afc7963f449d7f21d896d2f629cacc upstream.

Before this patch bq27xxx_battery_teardown() was setting poll_interval = 0 to avoid bq27xxx_battery_update() requeuing the delayed_work item.

There are 2 problems with this:

1. If the driver is unbound through sysfs, rather then the module being rmmod-ed, this changes poll_interval unexpectedly

2. This is racy, after it being set poll_interval could be changed before bq27xxx_battery_update() checks it through /sys/module/bq27xxx_battery/parameters/poll_interval

Fix this by added a removed attribute to struct bq27xxx_device_info and using that instead of setting poll_interval to 0.

There also is another poll_interval related race on remove(), writing /sys/module/bq27xxx_battery/parameters/poll_interval will requeue the delayed_work item for all devices on the bq27xxx_battery_devices list and the device being removed was only removed from that list after cancelling the delayed_work item.

Fix this by moving the removal from the bq27xxx_battery_devices list to before cancelling the delayed_work item.

---
 drivers/power/supply/bq27xxx_battery.c |   22 +++++++++-------------
 include/linux/power/bq27xxx_battery.h  |    1 +
 2 files changed, 10 insertions(+), 13 deletions(-)

@@ -1733,7 +1733,7 @@ static void bq27xxx_battery_update_unloc
 
 	di->last_update = jiffies;
 
-	if (poll_interval > 0)
+	if (!di->removed && poll_interval > 0)
 		mod_delayed_work(system_wq, &di->work, poll_interval * HZ);
 }
 

@@ -2063,22 +2063,18 @@ EXPORT_SYMBOL_GPL(bq27xxx_battery_setup)
 
 void bq27xxx_battery_teardown(struct bq27xxx_device_info *di)
 {
-	/*
-	 * power_supply_unregister call bq27xxx_battery_get_property which
-	 * call bq27xxx_battery_poll.
-	 * Make sure that bq27xxx_battery_poll will not call
-	 * schedule_delayed_work again after unregister (which cause OOPS).
-	 */
-	poll_interval = 0;
-
-	cancel_delayed_work_sync(&di->work);
-
-	power_supply_unregister(di->bat);
-
 	mutex_lock(&bq27xxx_list_lock);
 	list_del(&di->list);
 	mutex_unlock(&bq27xxx_list_lock);
 
+	/* Set removed to avoid bq27xxx_battery_update() re-queuing the work */
+	mutex_lock(&di->lock);
+	di->removed = true;
+	mutex_unlock(&di->lock);
+
+	cancel_delayed_work_sync(&di->work);
+
+	power_supply_unregister(di->bat);
 	mutex_destroy(&di->lock);
 }
 EXPORT_SYMBOL_GPL(bq27xxx_battery_teardown);

@@ -67,6 +67,7 @@ struct bq27xxx_device_info {
 	struct bq27xxx_access_methods bus;
 	struct bq27xxx_reg_cache cache;
 	int charge_design_full;
+	bool removed;
 	unsigned long last_update;
 	struct delayed_work work;
 	struct power_supply *bat;

From: Daisuke Nojiri <[email protected]>

commit b2f2a3c9800208b0db2c2e34b05323757117faa2 upstream.

CHARGE_INHIBITED bit position of the ChargerStatus register is actually 0 not 1. This patch corrects it.

---
 drivers/power/supply/sbs-charger.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -25,7 +25,7 @@
 #define SBS_CHARGER_REG_STATUS			0x13
 #define SBS_CHARGER_REG_ALARM_WARNING		0x16
 
-#define SBS_CHARGER_STATUS_CHARGE_INHIBITED	BIT(1)
+#define SBS_CHARGER_STATUS_CHARGE_INHIBITED	BIT(0)
 #define SBS_CHARGER_STATUS_RES_COLD		BIT(9)
 #define SBS_CHARGER_STATUS_RES_HOT		BIT(10)
 #define SBS_CHARGER_STATUS_BATTERY_PRESENT	BIT(14)

From: Hao Ge <[email protected]>

commit f15afbd34d8fadbd375f1212e97837e32bc170cc upstream.

Shifting signed 32-bit value by 31 bits is undefined, so changing significant bit to unsigned. It was spotted by UBSAN.

So let's just fix this by using the BIT() helper for all SB_* flags.

Message-Id: <[email protected]>

---
 include/linux/fs.h |   42 +++++++++++++++++++++---------------------
 1 file changed, 21 insertions(+), 21 deletions(-)

@@ -1348,29 +1348,29 @@ extern int send_sigurg(struct fown_struc
  * sb->s_flags.  Note that these mirror the equivalent MS_* flags where
  * represented in both.
  */
-#define SB_RDONLY	 1	/* Mount read-only */
-#define SB_NOSUID	 2	/* Ignore suid and sgid bits */
-#define SB_NODEV	 4	/* Disallow access to device special files */
-#define SB_NOEXEC	 8	/* Disallow program execution */
-#define SB_SYNCHRONOUS	16	/* Writes are synced at once */
-#define SB_MANDLOCK	64	/* Allow mandatory locks on an FS */
-#define SB_DIRSYNC	128	/* Directory modifications are synchronous */
-#define SB_NOATIME	1024	/* Do not update access times. */
-#define SB_NODIRATIME	2048	/* Do not update directory access times */
-#define SB_SILENT	32768
-#define SB_POSIXACL	(1<<16)	/* VFS does not apply the umask */
-#define SB_INLINECRYPT	(1<<17)	/* Use blk-crypto for encrypted files */
-#define SB_KERNMOUNT	(1<<22) /* this is a kern_mount call */
-#define SB_I_VERSION	(1<<23) /* Update inode I_version field */
-#define SB_LAZYTIME	(1<<25) /* Update the on-disk [acm]times lazily */
+#define SB_RDONLY       BIT(0)	/* Mount read-only */
+#define SB_NOSUID       BIT(1)	/* Ignore suid and sgid bits */
+#define SB_NODEV        BIT(2)	/* Disallow access to device special files */
+#define SB_NOEXEC       BIT(3)	/* Disallow program execution */
+#define SB_SYNCHRONOUS  BIT(4)	/* Writes are synced at once */
+#define SB_MANDLOCK     BIT(6)	/* Allow mandatory locks on an FS */
+#define SB_DIRSYNC      BIT(7)	/* Directory modifications are synchronous */
+#define SB_NOATIME      BIT(10)	/* Do not update access times. */
+#define SB_NODIRATIME   BIT(11)	/* Do not update directory access times */
+#define SB_SILENT       BIT(15)
+#define SB_POSIXACL     BIT(16)	/* VFS does not apply the umask */
+#define SB_INLINECRYPT  BIT(17)	/* Use blk-crypto for encrypted files */
+#define SB_KERNMOUNT    BIT(22)	/* this is a kern_mount call */
+#define SB_I_VERSION    BIT(23)	/* Update inode I_version field */
+#define SB_LAZYTIME     BIT(25)	/* Update the on-disk [acm]times lazily */
 
 /* These sb flags are internal to the kernel */
-#define SB_SUBMOUNT     (1<<26)
-#define SB_FORCE    	(1<<27)
-#define SB_NOSEC	(1<<28)
-#define SB_BORN		(1<<29)
-#define SB_ACTIVE	(1<<30)
-#define SB_NOUSER	(1<<31)
+#define SB_SUBMOUNT     BIT(26)
+#define SB_FORCE        BIT(27)
+#define SB_NOSEC        BIT(28)
+#define SB_BORN         BIT(29)
+#define SB_ACTIVE       BIT(30)
+#define SB_NOUSER       BIT(31)
 
 /* These flags relate to encoding and casefolding */
 #define SB_ENC_STRICT_MODE_FL	(1 << 0)

From: Dan Carpenter <[email protected]>

commit f67bc15e526bb9920683ad6c1891ff9e08981335 upstream.

This code generates a Smatch warning:

drivers/hwtracing/coresight/coresight-tmc-etr.c:947 tmc_etr_buf_insert_barrier_packet()
error: uninitialized symbol 'bufp'.

The problem is that if tmc_sg_table_get_data() returns -EINVAL, then when we test if "len < CORESIGHT_BARRIER_PKT_SIZE", the negative "len" value is type promoted to a high unsigned long value which is greater than CORESIGHT_BARRIER_PKT_SIZE. Fix this bug by adding an explicit check for error codes.

---
 drivers/hwtracing/coresight/coresight-tmc-etr.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -926,7 +926,7 @@ tmc_etr_buf_insert_barrier_packet(struct
 
 	len = tmc_etr_buf_get_data(etr_buf, offset,
 				   CORESIGHT_BARRIER_PKT_SIZE, &bufp);
-	if (WARN_ON(len < CORESIGHT_BARRIER_PKT_SIZE))
+	if (WARN_ON(len < 0 || len < CORESIGHT_BARRIER_PKT_SIZE))
 		return -EINVAL;
 	coresight_insert_barrier_packet(bufp);
 	return offset + CORESIGHT_BARRIER_PKT_SIZE;

From: Dan Carpenter <[email protected]>

commit 8fafac202d18230bb9926bda48e563fd2cce2a4f upstream.

In the pvcalls_new_active_socket() function, most error paths call pvcalls_back_release_active(fedata->dev, fedata, map) which calls sock_release() on "sock". The bug is that the caller also frees sock.

Fix this by making every error path in pvcalls_new_active_socket() release the sock, and don't free it in the caller.

---
 drivers/xen/pvcalls-back.c |    9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)

@@ -321,8 +321,10 @@ static struct sock_mapping *pvcalls_new_
 	void *page;
 
 	map = kzalloc(sizeof(*map), GFP_KERNEL);
-	if (map == NULL)
+	if (map == NULL) {
+		sock_release(sock);
 		return NULL;
+	}
 
 	map->fedata = fedata;
 	map->sock = sock;

@@ -414,10 +416,8 @@ static int pvcalls_back_connect(struct x
 					req->u.connect.ref,
 					req->u.connect.evtchn,
 					sock);
-	if (!map) {
+	if (!map)
 		ret = -EFAULT;
-		sock_release(sock);
-	}
 
 out:
 	rsp = RING_GET_RESPONSE(&fedata->ring, fedata->ring.rsp_prod_pvt++);

@@ -558,7 +558,6 @@ static void __pvcalls_back_accept(struct
 					sock);
 	if (!map) {
 		ret = -EFAULT;
-		sock_release(sock);
 		goto out_error;
 	}
 

From: Vernon Lovejoy <[email protected]>

commit 2e4be0d011f21593c6b316806779ba1eba2cd7e0 upstream.

The commit e335bb51cc15 ("x86/unwind: Ensure stack pointer is aligned") tried to align the stack pointer in show_trace_log_lvl(), otherwise the "stack < stack_info.end" check can't guarantee that the last read does not go past the end of the stack.

However, we have the same problem with the initial value of the stack pointer, it can also be unaligned. So without this patch this trivial kernel module

	#include <linux/module.h>

	static int init(void)
	{
		asm volatile("sub    $0x4,%rsp");
		dump_stack();
		asm volatile("add    $0x4,%rsp");

		return -EAGAIN;
	}

	module_init(init);
	MODULE_LICENSE("GPL");

crashes the kernel.

---
 arch/x86/kernel/dumpstack.c |    7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

@@ -195,7 +195,6 @@ void show_trace_log_lvl(struct task_stru
 	printk("%sCall Trace:\n", log_lvl);
 
 	unwind_start(&state, task, regs, stack);
-	stack = stack ? : get_stack_pointer(task, regs);
 	regs = unwind_get_entry_regs(&state, &partial);
 
 	/*

@@ -214,9 +213,13 @@ void show_trace_log_lvl(struct task_stru
 	 * - hardirq stack
 	 * - entry stack
 	 */
-	for ( ; stack; stack = PTR_ALIGN(stack_info.next_sp, sizeof(long))) {
+	for (stack = stack ?: get_stack_pointer(task, regs);
+	     stack;
+	     stack = stack_info.next_sp) {
 		const char *stack_name;
 
+		stack = PTR_ALIGN(stack, sizeof(long));
+
 		if (get_stack_info(stack, task, &stack_info, &visit_mask)) {
 			/*
 			 * We weren't on a valid stack.  It's possible that

From: Cezary Rojewski <[email protected]>

commit 95109657471311601b98e71f03d0244f48dc61bb upstream.

Constant 'C4_CHANNEL' does not exist on the firmware side. Value 0xC is reserved for 'C7_1' instead.

---
 include/uapi/sound/skl-tplg-interface.h |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -66,7 +66,8 @@ enum skl_ch_cfg {
 	SKL_CH_CFG_DUAL_MONO = 9,
 	SKL_CH_CFG_I2S_DUAL_STEREO_0 = 10,
 	SKL_CH_CFG_I2S_DUAL_STEREO_1 = 11,
-	SKL_CH_CFG_4_CHANNEL = 12,
+	SKL_CH_CFG_7_1 = 12,
+	SKL_CH_CFG_4_CHANNEL = SKL_CH_CFG_7_1,
 	SKL_CH_CFG_INVALID
 };
 

From: Christophe JAILLET <[email protected]>

commit 5b17a4971d3b2a073f4078dd65331efbe35baa2d upstream.

If an error occures after calling nv_mgmt_acquire_sema(), it should be undone with a corresponding nv_mgmt_release_sema() call.

Add it in the error handling path of the probe as already done in the remove function.

---
 drivers/net/ethernet/nvidia/forcedeth.c |    1 +
 1 file changed, 1 insertion(+)

@@ -6138,6 +6138,7 @@ static int nv_probe(struct pci_dev *pci_
 	return 0;
 
 out_error:
+	nv_mgmt_release_sema(dev);
 	if (phystate_orig)
 		writel(phystate|NVREG_ADAPTCTL_RUNNING, base + NvRegAdapterControl);
 out_freering:

From: Jakub Kicinski <[email protected]>

commit afbed3f74830163f9559579dee382cac3cff82da upstream.

NAPI gets called with budget of 0 from netpoll, which has interrupts disabled. We should try to free some space on Tx rings and nothing else.

Specifically do not try to handle XDP TX or try to refill Rx buffers - we can't use the page pool from IRQ context. Don't check if IRQs moved, either, that makes no sense in netpoll. Netpoll calls _all_ the rings from whatever CPU it happens to be invoked on.

In general do as little as possible, the work quickly adds up when there's tens of rings to poll.

The immediate stack trace I was seeing is:

    __do_softirq+0xd1/0x2c0
    __local_bh_enable_ip+0xc7/0x120
    </IRQ>
    <TASK>
    page_pool_put_defragged_page+0x267/0x320
    mlx5e_free_xdpsq_desc+0x99/0xd0
    mlx5e_poll_xdpsq_cq+0x138/0x3b0
    mlx5e_napi_poll+0xc3/0x8b0
    netpoll_poll_dev+0xce/0x150

AFAIU page pool takes a BH lock, releases it and since BH is now enabled tries to run softirqs.

---
 drivers/net/ethernet/mellanox/mlx5/core/en_txrx.c |   16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)

@@ -137,20 +137,22 @@ int mlx5e_napi_poll(struct napi_struct *
 	for (i = 0; i < c->num_tc; i++)
 		busy |= mlx5e_poll_tx_cq(&c->sq[i].cq, budget);
 
+	/* budget=0 means we may be in IRQ context, do as little as possible */
+	if (unlikely(!budget))
+		goto out;
+
 	busy |= mlx5e_poll_xdpsq_cq(&c->xdpsq.cq);
 
 	if (c->xdp)
 		busy |= mlx5e_poll_xdpsq_cq(&c->rq_xdpsq.cq);
 
-	if (likely(budget)) { /* budget=0 means: don't poll rx rings */
-		if (xsk_open)
-			work_done = mlx5e_poll_rx_cq(&xskrq->cq, budget);
+	if (xsk_open)
+		work_done = mlx5e_poll_rx_cq(&xskrq->cq, budget);
 
-		if (likely(budget - work_done))
-			work_done += mlx5e_poll_rx_cq(&rq->cq, budget - work_done);
+	if (likely(budget - work_done))
+		work_done += mlx5e_poll_rx_cq(&rq->cq, budget - work_done);
 
-		busy |= work_done == budget;
-	}
+	busy |= work_done == budget;
 
 	mlx5e_poll_ico_cq(&c->icosq.cq);
 	if (mlx5e_poll_ico_cq(&c->async_icosq.cq))

From: Erez Shitrit <[email protected]>

commit 1e5daf5565b61a96e570865091589afc9156e3d3 upstream.

When calculating crc for hash index we use the function crc32 that calculates for little-endian (LE) arch.
Then we convert it to network endianness using htonl(), but it's wrong to do the conversion in BE archs since the crc32 value is already LE.

The solution is to switch the bytes from the crc result for all types of arc.

---
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_ste.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -112,7 +112,8 @@ static u32 dr_ste_crc32_calc(const void
 {
 	u32 crc = crc32(0, input_data, length);
 
-	return (__force u32)htonl(crc);
+	return (__force u32)((crc >> 24) & 0xff) | ((crc << 8) & 0xff0000) |
+			    ((crc >> 8) & 0xff00) | ((crc << 24) & 0xff000000);
 }
 
 u32 mlx5dr_ste_calc_hash_index(u8 *hw_ste_p, struct mlx5dr_ste_htbl *htbl)

From: Roi Dayan <[email protected]>

commit a65735148e0328f80c0f72f9f8d2f609bfcf4aff upstream.

Fix spacing for the error and also the correct error code pointer.

---
 drivers/net/ethernet/mellanox/mlx5/core/main.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -887,7 +887,7 @@ static int mlx5_init_once(struct mlx5_co
 
 	dev->dm = mlx5_dm_create(dev);
 	if (IS_ERR(dev->dm))
-		mlx5_core_warn(dev, "Failed to init device memory%d\n", err);
+		mlx5_core_warn(dev, "Failed to init device memory %ld\n", PTR_ERR(dev->dm));
 
 	dev->tracer = mlx5_fw_tracer_create(dev);
 	dev->hv_vhca = mlx5_hv_vhca_create(dev);

From: Shay Drory <[email protected]>

commit af87194352cad882d787d06fb7efa714acd95427 upstream.

In case devcom allocation is failed, mlx5 is always freeing the priv. However, this priv might have been allocated by a different thread, and freeing it might lead to use-after-free bugs.
Fix it by freeing the priv only in case it was allocated by the running thread.

---
 drivers/net/ethernet/mellanox/mlx5/core/lib/devcom.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -110,7 +110,8 @@ struct mlx5_devcom *mlx5_devcom_register
 	priv->devs[idx] = dev;
 	devcom = mlx5_devcom_alloc(priv, idx);
 	if (!devcom) {
-		kfree(priv);
+		if (new_priv)
+			kfree(priv);
 		return ERR_PTR(-ENOMEM);
 	}
 

From: Hugo Villeneuve <[email protected]>

commit f161cea5a20f3aeeb637a88ad1705fc2720b4d58 upstream.

While testing the ethernet interface on a Variscite symphony carrier board using an imx8mn SOM with an onboard ADIN1300 PHY (EC hardware configuration), the ethernet PHY is not detected.

The ADIN1300 datasheet indicate that the "Management interface active (t4)" state is reached at most 5ms after the reset signal is deasserted.

The device tree in Variscite custom git repository uses the following property:

    phy-reset-post-delay = <20>;

Add a new MDIO property 'reset-deassert-us' of 20ms to have the same delay inside the ethphy node. Adding this property fixes the problem with the PHY detection.

Note that this SOM can also have an Atheros AR8033 PHY. In this case, a 1ms deassert delay is sufficient. Add a comment to that effect.

---
 arch/arm64/boot/dts/freescale/imx8mn-var-som.dtsi |    8 +++++++-
 1 file changed, 7 insertions(+), 1 deletion(-)

@@ -98,11 +98,17 @@
 		#address-cells = <1>;
 		#size-cells = <0>;
 
-		ethphy: ethernet-phy@4 {
+		ethphy: ethernet-phy@4 { /* AR8033 or ADIN1300 */
 			compatible = "ethernet-phy-ieee802.3-c22";
 			reg = <4>;
 			reset-gpios = <&gpio1 9 GPIO_ACTIVE_LOW>;
 			reset-assert-us = <10000>;
+			/*
+			 * Deassert delay:
+			 * ADIN1300 requires 5ms.
+			 * AR8033   requires 1ms.
+			 */
+			reset-deassert-us = <20000>;
 		};
 	};
 };

From: Christophe JAILLET <[email protected]>

commit 640bf95b2c7c2981fb471acdafbd3e0458f8390d upstream.

Should tc589_config() fail, some resources need to be released as already done in the remove function.

---
 drivers/net/ethernet/3com/3c589_cs.c |   11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

@@ -195,6 +195,7 @@ static int tc589_probe(struct pcmcia_dev
 {
 	struct el3_private *lp;
 	struct net_device *dev;
+	int ret;
 
 	dev_dbg(&link->dev, "3c589_attach()\n");
 

@@ -218,7 +219,15 @@ static int tc589_probe(struct pcmcia_dev
 
 	dev->ethtool_ops = &netdev_ethtool_ops;
 
-	return tc589_config(link);
+	ret = tc589_config(link);
+	if (ret)
+		goto err_free_netdev;
+
+	return 0;
+
+err_free_netdev:
+	free_netdev(dev);
+	return ret;
 }
 
 static void tc589_detach(struct pcmcia_device *link)

From: David Epping <[email protected]>

commit 57fb54ab9f6945e204740b696bd4cee61ee04e5e upstream.

The mscc driver implements support for VSC8502, so its ID should be in the MODULE_DEVICE_TABLE for automatic loading.

---
 drivers/net/phy/mscc/mscc_main.c |    1 +
 1 file changed, 1 insertion(+)

@@ -2563,6 +2563,7 @@ static struct phy_driver vsc85xx_driver[
 module_phy_driver(vsc85xx_driver);
 
 static struct mdio_device_id __maybe_unused vsc85xx_tbl[] = {
+	{ PHY_ID_VSC8502, 0xfffffff0, },
 	{ PHY_ID_VSC8504, 0xfffffff0, },
 	{ PHY_ID_VSC8514, 0xfffffff0, },
 	{ PHY_ID_VSC8530, 0xfffffff0, },

Build results:

	total: 162 pass: 162 fail: 0
Qemu test results:
	total: 485 pass: 485 fail: 0

Guenter

Results from Linaro’s test farm.
No regressions on arm64, arm, x86_64, and i386.

## Build
* kernel: 5.10.181-rc1
* git: https://gitlab.com/Linaro/lkft/mirrors/stable/linux-stable-rc
* git branch: linux-5.10.y
* git commit: 80ae453d08c191989cbf98440279674059eca336
* git describe: v5.10.180-212-g80ae453d08c1
* test details:
https://qa-reports.linaro.org/lkft/linux-stable-rc-linux-5.10.y/build/v5.10.180-212-g80ae453d08c1

## Test Regressions (compared to v5.10.180)

## Metric Regressions (compared to v5.10.180)

## Test Fixes (compared to v5.10.180)

## Metric Fixes (compared to v5.10.180)

## Test result summary
total: 133331, pass: 110831, fail: 3975, skip: 18332, xfail: 193

## Build Summary
* arc: 5 total, 5 passed, 0 failed
* arm: 117 total, 116 passed, 1 failed
* arm64: 45 total, 43 passed, 2 failed
* i386: 35 total, 33 passed, 2 failed
* mips: 27 total, 26 passed, 1 failed
* parisc: 8 total, 8 passed, 0 failed
* powerpc: 26 total, 20 passed, 6 failed
* riscv: 12 total, 11 passed, 1 failed
* s390: 12 total, 12 passed, 0 failed
* sh: 14 total, 12 passed, 2 failed
* sparc: 8 total, 8 passed, 0 failed
* x86_64: 38 total, 36 passed, 2 failed

## Test suites summary
* boot
* fwts
* igt-gpu-tools
* kselftest-android
* kselftest-arm64
* kselftest-breakpoints
* kselftest-capabilities
* kselftest-cgroup
* kselftest-clone3
* kselftest-core
* kselftest-cpu-hotplug
* kselftest-cpufreq
* kselftest-drivers-dma-buf
* kselftest-efivarfs
* kselftest-exec
* kselftest-filesystems
* kselftest-filesystems-binderfs
* kselftest-firmware
* kselftest-fpu
* kselftest-ftrace
* kselftest-futex
* kselftest-gpio
* kselftest-intel_pstate
* kselftest-ipc
* kselftest-ir
* kselftest-kcmp
* kselftest-kexec
* kselftest-kvm
* kselftest-lib
* kselftest-livepatch
* kselftest-membarrier
* kselftest-memfd
* kselftest-memory-hotplug
* kselftest-mincore
* kselftest-mount
* kselftest-mqueue
* kselftest-net
* kselftest-net-forwarding
* kselftest-net-mptcp
* kselftest-netfilter
* kselftest-nsfs
* kselftest-openat2
* kselftest-pid_namespace
* kselftest-pidfd
* kselftest-proc
* kselftest-pstore
* kselftest-ptrace
* kselftest-rseq
* kselftest-rtc
* kselftest-sigaltstack
* kselftest-size
* kselftest-tc-testing
* kselftest-timens
* kselftest-timers
* kselftest-tmpfs
* kselftest-tpm2
* kselftest-user
* kselftest-user_events
* kselftest-vDSO
* kselftest-watchdog
* kselftest-x86
* kselftest-zram
* kunit
* kvm-unit-tests
* libgpiod
* libhugetlbfs
* log-parser-boot
* log-parser-test
* ltp-cap_bounds
* ltp-commands
* ltp-containers
* ltp-controllers
* ltp-cpuhotplug
* ltp-crypto
* ltp-cve
* ltp-dio
* ltp-fcntl-locktests
* ltp-filecaps
* ltp-fs
* ltp-fs_bind
* ltp-fs_perms_simple
* ltp-fsx
* ltp-hugetlb
* ltp-io
* ltp-ipc
* ltp-math
* ltp-mm
* ltp-nptl
* ltp-pty
* ltp-sched
* ltp-securebits
* ltp-smoke
* ltp-syscalls
* ltp-tracing
* network-basic-tests
* perf
* rcutorture
* v4l2-compliance
* vdso

--
Linaro LKFT
https://lkft.linaro.org

All tests passing for Tegra ...

Test results for stable-v5.10:

    11 builds:	11 pass, 0 fail
    28 boots:	28 pass, 0 fail
    75 tests:	75 pass, 0 fail

Linux version: 5.10.181-rc1-g80ae453d08c1
Boards tested: tegra124-jetson-tk1, tegra186-p2771-0000,

                tegra194-p2972-0000, tegra194-p3509-0000+p3668-0000,
                tegra20-ventana, tegra210-p2371-2180,
                tegra210-p3450-0000, tegra30-cardhu-a04

Jon

Hello Greg,

CIP configurations built and booted with Linux 5.10.181-rc1 (80ae453d08c1):
https://gitlab.com/cip-project/cip-testing/linux-stable-rc-ci/-/pipelines/881425018
https://gitlab.com/cip-project/cip-testing/linux-stable-rc-ci/-/commits/linux-5.10.y

Kind regards, Chris

Compiled and booted on my x86_64 and ARM64 test systems. No errors or regressions.

Thanks.

On ARCH_BRCMSTB using 32-bit and 64-bit ARM kernels, build tested on BMIPS_GENERIC:

Florian