5.10.77-rc1 review

From: Oliver Neukum <[email protected]>

commit 397430b50a363d8b7bdda00522123f82df6adc5e upstream.

maxpacket of 0 makes no sense and oopses as we need to divide by it. Give up.

V2: fixed typo in log and stylistic issues

---
 drivers/net/usb/usbnet.c |    4 ++++
 1 file changed, 4 insertions(+)

@@ -1755,6 +1755,10 @@ usbnet_probe (struct usb_interface *udev
 	if (!dev->rx_urb_size)
 		dev->rx_urb_size = dev->hard_mtu;
 	dev->maxpacket = usb_maxpacket (dev->udev, dev->out, 1);
+	if (dev->maxpacket == 0) {
+		/* that is a broken device */
+		goto out4;
+	}
 
 	/* let userspace know we have a random address */
 	if (ether_addr_equal(net->dev_addr, node_id))

From: Rafał Miłecki <[email protected]>

commit 6dba4bdfd7a30e77b848a45404b224588bf989e5 upstream.

This reverts commit a49d784d5a8272d0f63c448fe8dc69e589db006e.

The updated binding was wrong / invalid and has been reverted. There isn't any upstream kernel DTS using it and Broadcom isn't known to use it neither. There is close to zero chance this will cause regression for anyone.

Actually in-kernel bcm5301x.dtsi still uses the old good binding and so it's broken since the driver update. This revert fixes it.

---
 drivers/pinctrl/bcm/pinctrl-ns.c |   29 ++++++++++-------------------
 1 file changed, 10 insertions(+), 19 deletions(-)

@@ -5,7 +5,6 @@
 
 #include <linux/err.h>
 #include <linux/io.h>
-#include <linux/mfd/syscon.h>
 #include <linux/module.h>
 #include <linux/of.h>
 #include <linux/of_device.h>

@@ -13,7 +12,6 @@
 #include <linux/pinctrl/pinctrl.h>
 #include <linux/pinctrl/pinmux.h>
 #include <linux/platform_device.h>
-#include <linux/regmap.h>
 #include <linux/slab.h>
 
 #define FLAG_BCM4708		BIT(1)

@@ -24,8 +22,7 @@ struct ns_pinctrl {
 	struct device *dev;
 	unsigned int chipset_flag;
 	struct pinctrl_dev *pctldev;
-	struct regmap *regmap;
-	u32 offset;
+	void __iomem *base;
 
 	struct pinctrl_desc pctldesc;
 	struct ns_pinctrl_group *groups;

@@ -232,9 +229,9 @@ static int ns_pinctrl_set_mux(struct pin
 		unset |= BIT(pin_number);
 	}
 
-	regmap_read(ns_pinctrl->regmap, ns_pinctrl->offset, &tmp);
+	tmp = readl(ns_pinctrl->base);
 	tmp &= ~unset;
-	regmap_write(ns_pinctrl->regmap, ns_pinctrl->offset, tmp);
+	writel(tmp, ns_pinctrl->base);
 
 	return 0;
 }

@@ -266,13 +263,13 @@ static const struct of_device_id ns_pinc
 static int ns_pinctrl_probe(struct platform_device *pdev)
 {
 	struct device *dev = &pdev->dev;
-	struct device_node *np = dev->of_node;
 	const struct of_device_id *of_id;
 	struct ns_pinctrl *ns_pinctrl;
 	struct pinctrl_desc *pctldesc;
 	struct pinctrl_pin_desc *pin;
 	struct ns_pinctrl_group *group;
 	struct ns_pinctrl_function *function;
+	struct resource *res;
 	int i;
 
 	ns_pinctrl = devm_kzalloc(dev, sizeof(*ns_pinctrl), GFP_KERNEL);

@@ -290,18 +287,12 @@ static int ns_pinctrl_probe(struct platf
 		return -EINVAL;
 	ns_pinctrl->chipset_flag = (uintptr_t)of_id->data;
 
-	ns_pinctrl->regmap = syscon_node_to_regmap(of_get_parent(np));
-	if (IS_ERR(ns_pinctrl->regmap)) {
-		int err = PTR_ERR(ns_pinctrl->regmap);
-
-		dev_err(dev, "Failed to map pinctrl regs: %d\n", err);
-
-		return err;
-	}
-
-	if (of_property_read_u32(np, "offset", &ns_pinctrl->offset)) {
-		dev_err(dev, "Failed to get register offset\n");
-		return -ENOENT;
+	res = platform_get_resource_byname(pdev, IORESOURCE_MEM,
+					   "cru_gpio_control");
+	ns_pinctrl->base = devm_ioremap_resource(dev, res);
+	if (IS_ERR(ns_pinctrl->base)) {
+		dev_err(dev, "Failed to map pinctrl regs\n");
+		return PTR_ERR(ns_pinctrl->base);
 	}
 
 	memcpy(pctldesc, &ns_pinctrl_desc, sizeof(*pctldesc));

From: Wang Hai <[email protected]>

commit 6f7c88691191e6c52ef2543d6f1da8d360b27a24 upstream.

Return error code if usb_maxpacket() returns 0 in usbnet_probe()

---
 drivers/net/usb/usbnet.c |    1 +
 1 file changed, 1 insertion(+)

@@ -1757,6 +1757,7 @@ usbnet_probe (struct usb_interface *udev
 	dev->maxpacket = usb_maxpacket (dev->udev, dev->out, 1);
 	if (dev->maxpacket == 0) {
 		/* that is a broken device */
+		status = -ENODEV;
 		goto out4;
 	}
 

From: Sachi King <[email protected]>

commit 4e5a04be88fe335ad5331f4f8c17f4ebd357e065 upstream.

Some systems such as the Microsoft Surface Laptop 4 leave interrupts enabled and configured for use in sleep states on boot, which cause unexpected behaviour such as spurious wakes and failed resumes in s2idle states.

As interrupts should not be enabled until they are claimed and explicitly enabled, disabling any interrupts mistakenly left enabled by firmware should be safe.

---
 drivers/pinctrl/pinctrl-amd.c |   31 +++++++++++++++++++++++++++++++
 1 file changed, 31 insertions(+)

@@ -764,6 +764,34 @@ static const struct pinconf_ops amd_pinc
 	.pin_config_group_set = amd_pinconf_group_set,
 };
 
+static void amd_gpio_irq_init(struct amd_gpio *gpio_dev)
+{
+	struct pinctrl_desc *desc = gpio_dev->pctrl->desc;
+	unsigned long flags;
+	u32 pin_reg, mask;
+	int i;
+
+	mask = BIT(WAKE_CNTRL_OFF_S0I3) | BIT(WAKE_CNTRL_OFF_S3) |
+		BIT(INTERRUPT_MASK_OFF) | BIT(INTERRUPT_ENABLE_OFF) |
+		BIT(WAKE_CNTRL_OFF_S4);
+
+	for (i = 0; i < desc->npins; i++) {
+		int pin = desc->pins[i].number;
+		const struct pin_desc *pd = pin_desc_get(gpio_dev->pctrl, pin);
+
+		if (!pd)
+			continue;
+
+		raw_spin_lock_irqsave(&gpio_dev->lock, flags);
+
+		pin_reg = readl(gpio_dev->base + i * 4);
+		pin_reg &= ~mask;
+		writel(pin_reg, gpio_dev->base + i * 4);
+
+		raw_spin_unlock_irqrestore(&gpio_dev->lock, flags);
+	}
+}
+
 #ifdef CONFIG_PM_SLEEP
 static bool amd_gpio_should_save(struct amd_gpio *gpio_dev, unsigned int pin)
 {

@@ -901,6 +929,9 @@ static int amd_gpio_probe(struct platfor
 		return PTR_ERR(gpio_dev->pctrl);
 	}
 
+	/* Disable and mask interrupts */
+	amd_gpio_irq_init(gpio_dev);
+
 	girq = &gpio_dev->gc.irq;
 	girq->chip = &amd_gpio_irqchip;
 	/* This will let us handle the parent IRQ in the driver */

From: Zheyu Ma <[email protected]>

commit a0023bb9dd9bc439d44604eeec62426a990054cd upstream.

mv_init_host() propagates the value returned by mv_chip_id() which in turn gets propagated by mv_pci_init_one() and hits local_pci_probe().

During the process of driver probing, the probe function should return < 0 for failure, otherwise, the kernel will treat value > 0 as success.

Since this is a bug rather than a recoverable runtime error we should use dev_alert() instead of dev_err().

---
 drivers/ata/sata_mv.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -3892,8 +3892,8 @@ static int mv_chip_id(struct ata_host *h
 		break;
 
 	default:
-		dev_err(host->dev, "BUG: invalid board index %u\n", board_idx);
-		return 1;
+		dev_alert(host->dev, "BUG: invalid board index %u\n", board_idx);
+		return -EINVAL;
 	}
 
 	hpriv->hp_flags = hp_flags;

From: Pavel Skripkin <[email protected]>

commit 10eff1f5788b6ffac212c254e2f3666219576889 upstream.

This reverts commit ab609f25d19858513919369ff3d9a63c02cd9e2e.

This patch is correct in the sense that we _should_ call device_put() in case of device_register() failure, but the problem in this code is more vast.

We need to set bus->state to UNMDIOBUS_REGISTERED before calling device_register() to correctly release the device in mdiobus_free(). This patch prevents us from doing it, since in case of device_register() failure put_device() will be called 2 times and it will cause UAF or something else.

Also, Reported-by: tag in revered commit was wrong, since syzbot reported different leak in same function.

---
 drivers/net/phy/mdio_bus.c |    1 -
 1 file changed, 1 deletion(-)

@@ -544,7 +544,6 @@ int __mdiobus_register(struct mii_bus *b
 	err = device_register(&bus->dev);
 	if (err) {
 		pr_err("mii_bus %s failed to register\n", bus->id);
-		put_device(&bus->dev);
 		return -EINVAL;
 	}
 

From: Max VA <[email protected]>

commit fa40d9734a57bcbfa79a280189799f76c88f7bb0 upstream.

The function tipc_crypto_key_rcv is used to parse MSG_CRYPTO messages to receive keys from other nodes in the cluster in order to decrypt any further messages from them.
This patch verifies that any supplied sizes in the message body are valid for the received message.

---
 net/tipc/crypto.c |   32 +++++++++++++++++++++-----------
 1 file changed, 21 insertions(+), 11 deletions(-)

@@ -2278,43 +2278,53 @@ static bool tipc_crypto_key_rcv(struct t
 	u16 key_gen = msg_key_gen(hdr);
 	u16 size = msg_data_sz(hdr);
 	u8 *data = msg_data(hdr);
+	unsigned int keylen;
+
+	/* Verify whether the size can exist in the packet */
+	if (unlikely(size < sizeof(struct tipc_aead_key) + TIPC_AEAD_KEYLEN_MIN)) {
+		pr_debug("%s: message data size is too small\n", rx->name);
+		goto exit;
+	}
+
+	keylen = ntohl(*((__be32 *)(data + TIPC_AEAD_ALG_NAME)));
+
+	/* Verify the supplied size values */
+	if (unlikely(size != keylen + sizeof(struct tipc_aead_key) ||
+		     keylen > TIPC_AEAD_KEY_SIZE_MAX)) {
+		pr_debug("%s: invalid MSG_CRYPTO key size\n", rx->name);
+		goto exit;
+	}
 
 	spin_lock(&rx->lock);
 	if (unlikely(rx->skey || (key_gen == rx->key_gen && rx->key.keys))) {
 		pr_err("%s: key existed <%p>, gen %d vs %d\n", rx->name,
 		       rx->skey, key_gen, rx->key_gen);
-		goto exit;
+		goto exit_unlock;
 	}
 
 	/* Allocate memory for the key */
 	skey = kmalloc(size, GFP_ATOMIC);
 	if (unlikely(!skey)) {
 		pr_err("%s: unable to allocate memory for skey\n", rx->name);
-		goto exit;
+		goto exit_unlock;
 	}
 
 	/* Copy key from msg data */
-	skey->keylen = ntohl(*((__be32 *)(data + TIPC_AEAD_ALG_NAME)));
+	skey->keylen = keylen;
 	memcpy(skey->alg_name, data, TIPC_AEAD_ALG_NAME);
 	memcpy(skey->key, data + TIPC_AEAD_ALG_NAME + sizeof(__be32),
 	       skey->keylen);
 
-	/* Sanity check */
-	if (unlikely(size != tipc_aead_key_size(skey))) {
-		kfree(skey);
-		skey = NULL;
-		goto exit;
-	}
-
 	rx->key_gen = key_gen;
 	rx->skey_mode = msg_key_mode(hdr);
 	rx->skey = skey;
 	rx->nokey = 0;
 	mb(); /* for nokey flag */
 
-exit:
+exit_unlock:
 	spin_unlock(&rx->lock);
 
+exit:
 	/* Schedule the key attaching on this crypto */
 	if (likely(skey && queue_delayed_work(tx->wq, &rx->work, 0)))
 		return true;

From: Krzysztof Kozlowski <[email protected]>

commit 2195f2062e4cc93870da8e71c318ef98a1c51cef upstream.

During probing, the driver tries to get a list (mask) of supported command types in port100_get_command_type_mask() function. The value is u64 and 0 is treated as invalid mask (no commands supported). The function however returns also -ERRNO as u64 which will be interpret as valid command mask.

Return 0 on every error case of port100_get_command_type_mask(), so the probing will stop.

---
 drivers/nfc/port100.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -1003,11 +1003,11 @@ static u64 port100_get_command_type_mask
 
 	skb = port100_alloc_skb(dev, 0);
 	if (!skb)
-		return -ENOMEM;
+		return 0;
 
 	resp = port100_send_cmd_sync(dev, PORT100_CMD_GET_COMMAND_TYPE, skb);
 	if (IS_ERR(resp))
-		return PTR_ERR(resp);
+		return 0;
 
 	if (resp->len < 8)
 		mask = 0;

From: Daniel Jordan <[email protected]>

commit da353fac65fede6b8b4cfe207f0d9408e3121105 upstream.

sk->sk_err appears to expect a positive value, a convention that ktls doesn't always follow and that leads to memory corruption in other code. For instance,

    [kworker]
    tls_encrypt_done(..., err=<negative error from crypto request>)
      tls_err_abort(.., err)
        sk->sk_err = err;

    [task]
    splice_from_pipe_feed
      ...
        tls_sw_do_sendpage
          if (sk->sk_err) {
            ret = -sk->sk_err;  // ret is positive

    splice_from_pipe_feed (continued)
      ret = actor(...)  // ret is still positive and interpreted as bytes
                        // written, resulting in underflow of buf->len and
                        // sd->len, leading to huge buf->offset and bogus
                        // addresses computed in later calls to actor()

Fix all tls_err_abort() callers to pass a negative error code consistently and centralize the error-prone sign flip there, throwing in a warning to catch future misuse and uninlining the function so it really does only warn once.

---
 include/net/tls.h |    9 ++-------
 net/tls/tls_sw.c  |   17 +++++++++++++----
 2 files changed, 15 insertions(+), 11 deletions(-)

@@ -359,6 +359,7 @@ int tls_sk_query(struct sock *sk, int op
 		int __user *optlen);
 int tls_sk_attach(struct sock *sk, int optname, char __user *optval,
 		  unsigned int optlen);
+void tls_err_abort(struct sock *sk, int err);
 
 int tls_set_sw_offload(struct sock *sk, struct tls_context *ctx, int tx);
 void tls_sw_strparser_arm(struct sock *sk, struct tls_context *ctx);

@@ -467,12 +468,6 @@ static inline bool tls_is_sk_tx_device_o
 #endif
 }
 
-static inline void tls_err_abort(struct sock *sk, int err)
-{
-	sk->sk_err = err;
-	sk->sk_error_report(sk);
-}
-
 static inline bool tls_bigint_increment(unsigned char *seq, int len)
 {
 	int i;

@@ -513,7 +508,7 @@ static inline void tls_advance_record_sn
 					 struct cipher_context *ctx)
 {
 	if (tls_bigint_increment(ctx->rec_seq, prot->rec_seq_size))
-		tls_err_abort(sk, EBADMSG);
+		tls_err_abort(sk, -EBADMSG);
 
 	if (prot->version != TLS_1_3_VERSION)
 		tls_bigint_increment(ctx->iv + TLS_CIPHER_AES_GCM_128_SALT_SIZE,

@@ -35,6 +35,7 @@
  * SOFTWARE.
  */
 
+#include <linux/bug.h>
 #include <linux/sched/signal.h>
 #include <linux/module.h>
 #include <linux/splice.h>

@@ -43,6 +44,14 @@
 #include <net/strparser.h>
 #include <net/tls.h>
 
+noinline void tls_err_abort(struct sock *sk, int err)
+{
+	WARN_ON_ONCE(err >= 0);
+	/* sk->sk_err should contain a positive error code. */
+	sk->sk_err = -err;
+	sk->sk_error_report(sk);
+}
+
 static int __skb_nsg(struct sk_buff *skb, int offset, int len,
                      unsigned int recursion_level)
 {

@@ -419,7 +428,7 @@ int tls_tx_records(struct sock *sk, int
 
 tx_err:
 	if (rc < 0 && rc != -EAGAIN)
-		tls_err_abort(sk, EBADMSG);
+		tls_err_abort(sk, -EBADMSG);
 
 	return rc;
 }

@@ -764,7 +773,7 @@ static int tls_push_record(struct sock *
 			       msg_pl->sg.size + prot->tail_size, i);
 	if (rc < 0) {
 		if (rc != -EINPROGRESS) {
-			tls_err_abort(sk, EBADMSG);
+			tls_err_abort(sk, -EBADMSG);
 			if (split) {
 				tls_ctx->pending_open_record_frags = true;
 				tls_merge_open_record(sk, rec, tmp, orig_end);

@@ -1828,7 +1837,7 @@ int tls_sw_recvmsg(struct sock *sk,
 		err = decrypt_skb_update(sk, skb, &msg->msg_iter,
 					 &chunk, &zc, async_capable);
 		if (err < 0 && err != -EINPROGRESS) {
-			tls_err_abort(sk, EBADMSG);
+			tls_err_abort(sk, -EBADMSG);
 			goto recv_end;
 		}
 

@@ -2008,7 +2017,7 @@ ssize_t tls_sw_splice_read(struct socket
 		}
 
 		if (err < 0) {
-			tls_err_abort(sk, EBADMSG);
+			tls_err_abort(sk, -EBADMSG);
 			goto splice_read_end;
 		}
 		ctx->decrypted = 1;

From: Nick Desaulniers <[email protected]>

commit e6a0c958bdf9b2e1b57501fc9433a461f0a6aadd upstream.

A kernel built with CONFIG_THUMB2_KERNEL=y and using clang as the assembler could generate non-naturally-aligned v7wbi_tlb_fns which results in a boot failure. The original commit adding the macro missed the .align directive on this data.

---
 arch/arm/mm/proc-macros.S |    1 +
 1 file changed, 1 insertion(+)

@@ -340,6 +340,7 @@ ENTRY(\name\()_cache_fns)
 
 .macro define_tlb_functions name:req, flags_up:req, flags_smp
 	.type	\name\()_tlb_fns, #object
+	.align 2
 ENTRY(\name\()_tlb_fns)
 	.long	\name\()_flush_user_tlb_range
 	.long	\name\()_flush_kern_tlb_range

From: Lexi Shao <[email protected]>

commit df909df0770779f1a5560c2bb641a2809655ef28 upstream.

ARM: kasan: Fix __get_user_check failure with kasan

In macro __get_user_check defined in arch/arm/include/asm/uaccess.h, error code is store in register int __e(r0). When kasan is enabled, assigning value to kernel address might trigger kasan check, which unexpectedly overwrites r0 and causes undefined behavior on arm kasan images.

One example is failure in do_futex and results in process soft lockup. Log:

watchdog: BUG: soft lockup - CPU#0 stuck for 62946ms! [rs:main

Q:Reg:1151]
...
(__asan_store4) from (futex_wait_setup+0xf8/0x2b4) (futex_wait_setup) from (futex_wait+0x138/0x394) (futex_wait) from (do_futex+0x164/0xe40) (do_futex) from (sys_futex_time32+0x178/0x230) (sys_futex_time32) from (ret_fast_syscall+0x0/0x50)

The soft lockup happens in function futex_wait_setup. The reason is function get_futex_value_locked always return EINVAL, thus pc jump back to retry label and causes looping.

This line in function get_futex_value_locked

	ret = __get_user(*dest, from);
is expanded to
	*dest = (typeof(*(p))) __r2; ,

in macro __get_user_check. Writing to pointer dest triggers kasan check and overwrites the return value of __get_user_x function. The assembly code of get_futex_value_locked in kernel/futex.c:
...
c01f6dc8: eb0b020e bl c04b7608 <__get_user_4>
// "x = (typeof(*(p))) __r2;" triggers kasan check and r0 is overwritten

c01f6dCc:       e1a00007        mov     r0, r7
c01f6dd0:       e1a05002        mov     r5, r2
c01f6dd4:       eb04f1e6        bl      c0333574 <__asan_store4>
c01f6dd8:       e5875000        str     r5, [r7]

// save ret value of __get_user(*dest, from), which is dest address now
c01f6ddc: e1a05000 mov r5, r0
...
// checking return value of __get_user failed
c01f6e00: e3550000 cmp r5, #0
...
c01f6e0c: 01a00005 moveq r0, r5
// assign return value to EINVAL
c01f6e10: 13e0000d mvnne r0, #13

Return value is the destination address of get_user thus certainly non-zero, so get_futex_value_locked always return EINVAL.

Fix it by using a tmp vairable to store the error code before the assignment. This fix has no effects to non-kasan images thanks to compiler optimization. It only affects cases that overwrite r0 due to kasan check.

This should fix bug discussed in Link:
[1] https://lore.kernel.org/linux-arm-kernel/[email protected]/

---
 arch/arm/include/asm/uaccess.h |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

@@ -200,6 +200,7 @@ extern int __get_user_64t_4(void *);
 		register unsigned long __l asm("r1") = __limit;		\
 		register int __e asm("r0");				\
 		unsigned int __ua_flags = uaccess_save_and_enable();	\
+		int __tmp_e;						\
 		switch (sizeof(*(__p))) {				\
 		case 1:							\
 			if (sizeof((x)) >= 8)				\

@@ -227,9 +228,10 @@ extern int __get_user_64t_4(void *);
 			break;						\
 		default: __e = __get_user_bad(); break;			\
 		}							\
+		__tmp_e = __e;						\
 		uaccess_restore(__ua_flags);				\
 		x = (typeof(*(p))) __r2;				\
-		__e;							\
+		__tmp_e;						\
 	})
 
 #define get_user(x, p)							\

From: Theodore Ts'o <[email protected]>

commit 61bb4a1c417e5b95d9edb4f887f131de32e419cb upstream.

After commit 618f003199c6 ("ext4: fix memory leak in ext4_fill_super"), after the file system is remounted read-only, there is a race where the kmmpd thread can exit, causing sbi->s_mmp_tsk to point at freed memory, which the call to ext4_stop_mmpd() can trip over.

Fix this by only allowing kmmpd() to exit when it is stopped via ext4_stop_mmpd().

Bug-Report-Link: <[email protected]>

---
 fs/ext4/mmp.c   |   31 +++++++++++++++----------------
 fs/ext4/super.c |    6 +++++-
 2 files changed, 20 insertions(+), 17 deletions(-)

@@ -156,7 +156,12 @@ static int kmmpd(void *data)
 	memcpy(mmp->mmp_nodename, init_utsname()->nodename,
 	       sizeof(mmp->mmp_nodename));
 
-	while (!kthread_should_stop()) {
+	while (!kthread_should_stop() && !sb_rdonly(sb)) {
+		if (!ext4_has_feature_mmp(sb)) {
+			ext4_warning(sb, "kmmpd being stopped since MMP feature"
+				     " has been disabled.");
+			goto wait_to_exit;
+		}
 		if (++seq > EXT4_MMP_SEQ_MAX)
 			seq = 1;
 

@@ -177,16 +182,6 @@ static int kmmpd(void *data)
 			failed_writes++;
 		}
 
-		if (!(le32_to_cpu(es->s_feature_incompat) &
-		    EXT4_FEATURE_INCOMPAT_MMP)) {
-			ext4_warning(sb, "kmmpd being stopped since MMP feature"
-				     " has been disabled.");
-			goto exit_thread;
-		}
-
-		if (sb_rdonly(sb))
-			break;
-
 		diff = jiffies - last_update_time;
 		if (diff < mmp_update_interval * HZ)
 			schedule_timeout_interruptible(mmp_update_interval *

@@ -207,7 +202,7 @@ static int kmmpd(void *data)
 				ext4_error_err(sb, -retval,
 					       "error reading MMP data: %d",
 					       retval);
-				goto exit_thread;
+				goto wait_to_exit;
 			}
 
 			mmp_check = (struct mmp_struct *)(bh_check->b_data);

@@ -221,7 +216,7 @@ static int kmmpd(void *data)
 				ext4_error_err(sb, EBUSY, "abort");
 				put_bh(bh_check);
 				retval = -EBUSY;
-				goto exit_thread;
+				goto wait_to_exit;
 			}
 			put_bh(bh_check);
 		}

@@ -244,7 +239,13 @@ static int kmmpd(void *data)
 
 	retval = write_mmp_block(sb, bh);
 
-exit_thread:
+wait_to_exit:
+	while (!kthread_should_stop()) {
+		set_current_state(TASK_INTERRUPTIBLE);
+		if (!kthread_should_stop())
+			schedule();
+	}
+	set_current_state(TASK_RUNNING);
 	return retval;
 }
 

@@ -391,5 +392,3 @@ failed:
 	brelse(bh);
 	return 1;
 }
-
-

@@ -5932,7 +5932,6 @@ static int ext4_remount(struct super_blo
 				 */
 				ext4_mark_recovery_complete(sb, es);
 			}
-			ext4_stop_mmpd(sbi);
 		} else {
 			/* Make sure we can mount this feature set readwrite */
 			if (ext4_has_feature_readonly(sb) ||

@@ -6046,6 +6045,9 @@ static int ext4_remount(struct super_blo
 	if (!test_opt(sb, BLOCK_VALIDITY) && sbi->s_system_blks)
 		ext4_release_system_zone(sb);
 
+	if (!ext4_has_feature_mmp(sb) || sb_rdonly(sb))
+		ext4_stop_mmpd(sbi);
+
 	/*
 	 * Some options can be enabled by ext4 and/or by VFS mount flag
 	 * either way we need to make sure it matches in both *flags and

@@ -6078,6 +6080,8 @@ restore_opts:
 	for (i = 0; i < EXT4_MAXQUOTAS; i++)
 		kfree(to_free[i]);
 #endif
+	if (!ext4_has_feature_mmp(sb) || sb_rdonly(sb))
+		ext4_stop_mmpd(sbi);
 	kfree(orig_data);
 	return err;
 }

From: Wenbin Mei <[email protected]>

commit 92b18252b91de567cd875f2e84722b10ab34ee28 upstream.

While mmc0 enter suspend state, we need halt CQE to send legacy cmd(flush cache) and disable cqe, for resume back, we enable CQE and not clear HALT state.
In this case MediaTek mmc host controller will keep the value for HALT state after CQE disable/enable flow, so the next CQE transfer after resume will be timeout due to CQE is in HALT state, the log as below:
<4>.(4)[318:kworker/4:1H]mmc0: cqhci: timeout for tag 2 <4>.(4)[318:kworker/4:1H]mmc0: cqhci: ============ CQHCI REGISTER DUMP ===========

<4>.(4)[318:kworker/4:1H]mmc0: cqhci: Caps:      0x100020b6 | Version:  0x00000510
<4>.(4)[318:kworker/4:1H]mmc0: cqhci: Config:    0x00001103 | Control:  0x00000001
<4>.(4)[318:kworker/4:1H]mmc0: cqhci: Int stat:  0x00000000 | Int enab: 0x00000006
<4>.(4)[318:kworker/4:1H]mmc0: cqhci: Int sig:   0x00000006 | Int Coal: 0x00000000
<4>.(4)[318:kworker/4:1H]mmc0: cqhci: TDL base:  0xfd05f000 | TDL up32: 0x00000000
<4>.(4)[318:kworker/4:1H]mmc0: cqhci: Doorbell:  0x8000203c | TCN:      0x00000000
<4>.(4)[318:kworker/4:1H]mmc0: cqhci: Dev queue: 0x00000000 | Dev Pend: 0x00000000
<4>.(4)[318:kworker/4:1H]mmc0: cqhci: Task clr:  0x00000000 | SSC1:     0x00001000
<4>.(4)[318:kworker/4:1H]mmc0: cqhci: SSC2:      0x00000001 | DCMD rsp: 0x00000000
<4>.(4)[318:kworker/4:1H]mmc0: cqhci: RED mask:  0xfdf9a080 | TERRI:    0x00000000
<4>.(4)[318:kworker/4:1H]mmc0: cqhci: Resp idx:  0x00000000 | Resp arg: 0x00000000
<4>.(4)[318:kworker/4:1H]mmc0: cqhci: CRNQP:     0x00000000 | CRNQDUN:  0x00000000
<4>.(4)[318:kworker/4:1H]mmc0: cqhci: CRNQIS:    0x00000000 | CRNQIE:   0x00000000

This change check HALT state after CQE enable, if CQE is in HALT state, we will clear it.

---
 drivers/mmc/host/cqhci.c |    3 +++
 1 file changed, 3 insertions(+)

@@ -273,6 +273,9 @@ static void __cqhci_enable(struct cqhci_
 
 	cqhci_writel(cq_host, cqcfg, CQHCI_CFG);
 
+	if (cqhci_readl(cq_host, CQHCI_CTL) & CQHCI_HALT)
+		cqhci_writel(cq_host, 0, CQHCI_CTL);
+
 	mmc->cqe_on = true;
 
 	if (cq_host->ops->enable)

From: Johan Hovold <[email protected]>

commit 8c8171929116cc23f74743d99251eedadf62341a upstream.

USB control-message timeouts are specified in milliseconds and should specifically not vary with CONFIG_HZ.

---
 drivers/mmc/host/vub300.c |   18 +++++++++---------
 1 file changed, 9 insertions(+), 9 deletions(-)

@@ -576,7 +576,7 @@ static void check_vub300_port_status(str
 				GET_SYSTEM_PORT_STATUS,
 				USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
 				0x0000, 0x0000, &vub300->system_port_status,
-				sizeof(vub300->system_port_status), HZ);
+				sizeof(vub300->system_port_status), 1000);
 	if (sizeof(vub300->system_port_status) == retval)
 		new_system_port_status(vub300);
 }

@@ -1241,7 +1241,7 @@ static void __download_offload_pseudocod
 						SET_INTERRUPT_PSEUDOCODE,
 						USB_DIR_OUT | USB_TYPE_VENDOR |
 						USB_RECIP_DEVICE, 0x0000, 0x0000,
-						xfer_buffer, xfer_length, HZ);
+						xfer_buffer, xfer_length, 1000);
 			kfree(xfer_buffer);
 			if (retval < 0)
 				goto copy_error_message;

@@ -1284,7 +1284,7 @@ static void __download_offload_pseudocod
 						SET_TRANSFER_PSEUDOCODE,
 						USB_DIR_OUT | USB_TYPE_VENDOR |
 						USB_RECIP_DEVICE, 0x0000, 0x0000,
-						xfer_buffer, xfer_length, HZ);
+						xfer_buffer, xfer_length, 1000);
 			kfree(xfer_buffer);
 			if (retval < 0)
 				goto copy_error_message;

@@ -1991,7 +1991,7 @@ static void __set_clock_speed(struct vub
 		usb_control_msg(vub300->udev, usb_sndctrlpipe(vub300->udev, 0),
 				SET_CLOCK_SPEED,
 				USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
-				0x00, 0x00, buf, buf_array_size, HZ);
+				0x00, 0x00, buf, buf_array_size, 1000);
 	if (retval != 8) {
 		dev_err(&vub300->udev->dev, "SET_CLOCK_SPEED"
 			" %dkHz failed with retval=%d\n", kHzClock, retval);

@@ -2013,14 +2013,14 @@ static void vub300_mmc_set_ios(struct mm
 		usb_control_msg(vub300->udev, usb_sndctrlpipe(vub300->udev, 0),
 				SET_SD_POWER,
 				USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
-				0x0000, 0x0000, NULL, 0, HZ);
+				0x0000, 0x0000, NULL, 0, 1000);
 		/* must wait for the VUB300 u-proc to boot up */
 		msleep(600);
 	} else if ((ios->power_mode == MMC_POWER_UP) && !vub300->card_powered) {
 		usb_control_msg(vub300->udev, usb_sndctrlpipe(vub300->udev, 0),
 				SET_SD_POWER,
 				USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
-				0x0001, 0x0000, NULL, 0, HZ);
+				0x0001, 0x0000, NULL, 0, 1000);
 		msleep(600);
 		vub300->card_powered = 1;
 	} else if (ios->power_mode == MMC_POWER_ON) {

@@ -2275,14 +2275,14 @@ static int vub300_probe(struct usb_inter
 				GET_HC_INF0,
 				USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
 				0x0000, 0x0000, &vub300->hc_info,
-				sizeof(vub300->hc_info), HZ);
+				sizeof(vub300->hc_info), 1000);
 	if (retval < 0)
 		goto error5;
 	retval =
 		usb_control_msg(vub300->udev, usb_sndctrlpipe(vub300->udev, 0),
 				SET_ROM_WAIT_STATES,
 				USB_DIR_OUT | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
-				firmware_rom_wait_states, 0x0000, NULL, 0, HZ);
+				firmware_rom_wait_states, 0x0000, NULL, 0, 1000);
 	if (retval < 0)
 		goto error5;
 	dev_info(&vub300->udev->dev,

@@ -2297,7 +2297,7 @@ static int vub300_probe(struct usb_inter
 				GET_SYSTEM_PORT_STATUS,
 				USB_DIR_IN | USB_TYPE_VENDOR | USB_RECIP_DEVICE,
 				0x0000, 0x0000, &vub300->system_port_status,
-				sizeof(vub300->system_port_status), HZ);
+				sizeof(vub300->system_port_status), 1000);
 	if (retval < 0) {
 		goto error4;
 	} else if (sizeof(vub300->system_port_status) == retval) {

From: Wenbin Mei <[email protected]>

commit e8a1ff65927080278e6826f797b7c197fb2611a6 upstream.

We must enable clock before cqhci init, because crypto needs read information from CQHCI registers, otherwise, it will hang in MediaTek mmc host controller.

---
 drivers/mmc/host/mtk-sd.c |   38 +++++++++++++++++++-------------------
 1 file changed, 19 insertions(+), 19 deletions(-)

@@ -2503,6 +2503,25 @@ static int msdc_drv_probe(struct platfor
 		host->dma_mask = DMA_BIT_MASK(32);
 	mmc_dev(mmc)->dma_mask = &host->dma_mask;
 
+	host->timeout_clks = 3 * 1048576;
+	host->dma.gpd = dma_alloc_coherent(&pdev->dev,
+				2 * sizeof(struct mt_gpdma_desc),
+				&host->dma.gpd_addr, GFP_KERNEL);
+	host->dma.bd = dma_alloc_coherent(&pdev->dev,
+				MAX_BD_NUM * sizeof(struct mt_bdma_desc),
+				&host->dma.bd_addr, GFP_KERNEL);
+	if (!host->dma.gpd || !host->dma.bd) {
+		ret = -ENOMEM;
+		goto release_mem;
+	}
+	msdc_init_gpd_bd(host, &host->dma);
+	INIT_DELAYED_WORK(&host->req_timeout, msdc_request_timeout);
+	spin_lock_init(&host->lock);
+
+	platform_set_drvdata(pdev, mmc);
+	msdc_ungate_clock(host);
+	msdc_init_hw(host);
+
 	if (mmc->caps2 & MMC_CAP2_CQE) {
 		host->cq_host = devm_kzalloc(mmc->parent,
 					     sizeof(*host->cq_host),

@@ -2523,25 +2542,6 @@ static int msdc_drv_probe(struct platfor
 		mmc->max_seg_size = 64 * 1024;
 	}
 
-	host->timeout_clks = 3 * 1048576;
-	host->dma.gpd = dma_alloc_coherent(&pdev->dev,
-				2 * sizeof(struct mt_gpdma_desc),
-				&host->dma.gpd_addr, GFP_KERNEL);
-	host->dma.bd = dma_alloc_coherent(&pdev->dev,
-				MAX_BD_NUM * sizeof(struct mt_bdma_desc),
-				&host->dma.bd_addr, GFP_KERNEL);
-	if (!host->dma.gpd || !host->dma.bd) {
-		ret = -ENOMEM;
-		goto release_mem;
-	}
-	msdc_init_gpd_bd(host, &host->dma);
-	INIT_DELAYED_WORK(&host->req_timeout, msdc_request_timeout);
-	spin_lock_init(&host->lock);
-
-	platform_set_drvdata(pdev, mmc);
-	msdc_ungate_clock(host);
-	msdc_init_hw(host);
-
 	ret = devm_request_irq(&pdev->dev, host->irq, msdc_irq,
 			       IRQF_TRIGGER_NONE, pdev->name, host);
 	if (ret)

From: Shawn Guo <[email protected]>

commit 4217d07b9fb328751f877d3bd9550122014860a2 upstream.

On Thundercomm TurboX CM2290, the eMMC OCR reports vdd = 23 (3.5 ~ 3.6 V), which is being treated as an invalid value by sdhci_set_power_noreg(). And thus eMMC is totally broken on the platform.

[    1.436599] ------------[ cut here ]------------
[    1.436606] mmc0: Invalid vdd 0x17
[    1.436640] WARNING: CPU: 2 PID: 69 at drivers/mmc/host/sdhci.c:2048 sdhci_set_power_noreg+0x168/0x2b4
[    1.436655] Modules linked in:
[    1.436662] CPU: 2 PID: 69 Comm: kworker/u8:1 Tainted: G        W         5.15.0-rc1+ #137
[    1.436669] Hardware name: Thundercomm TurboX CM2290 (DT)
[    1.436674] Workqueue: events_unbound async_run_entry_fn
[    1.436685] pstate: 60000005 (nZCv daif -PAN -UAO -TCO -DIT -SSBS BTYPE=--)
[    1.436692] pc : sdhci_set_power_noreg+0x168/0x2b4
[    1.436698] lr : sdhci_set_power_noreg+0x168/0x2b4
[    1.436703] sp : ffff800010803a60
[    1.436705] x29: ffff800010803a60 x28: ffff6a9102465f00 x27: ffff6a9101720a70
[    1.436715] x26: ffff6a91014de1c0 x25: ffff6a91014de010 x24: ffff6a91016af280
[    1.436724] x23: ffffaf7b1b276640 x22: 0000000000000000 x21: ffff6a9101720000
[    1.436733] x20: ffff6a9101720370 x19: ffff6a9101720580 x18: 0000000000000020
[    1.436743] x17: 0000000000000000 x16: 0000000000000004 x15: ffffffffffffffff
[    1.436751] x14: 0000000000000000 x13: 00000000fffffffd x12: ffffaf7b1b84b0bc
[    1.436760] x11: ffffaf7b1b720d10 x10: 000000000000000a x9 : ffff800010803a60
[    1.436769] x8 : 000000000000000a x7 : 000000000000000f x6 : 00000000fffff159
[    1.436778] x5 : 0000000000000000 x4 : 0000000000000000 x3 : 00000000ffffffff
[    1.436787] x2 : 0000000000000000 x1 : 0000000000000000 x0 : ffff6a9101718d80
[    1.436797] Call trace:
[    1.436800]  sdhci_set_power_noreg+0x168/0x2b4
[    1.436805]  sdhci_set_ios+0xa0/0x7fc
[    1.436811]  mmc_power_up.part.0+0xc4/0x164
[    1.436818]  mmc_start_host+0xa0/0xb0
[    1.436824]  mmc_add_host+0x60/0x90
[    1.436830]  __sdhci_add_host+0x174/0x330
[    1.436836]  sdhci_msm_probe+0x7c0/0x920
[    1.436842]  platform_probe+0x68/0xe0
[    1.436850]  really_probe.part.0+0x9c/0x31c
[    1.436857]  __driver_probe_device+0x98/0x144
[    1.436863]  driver_probe_device+0xc8/0x15c
[    1.436869]  __device_attach_driver+0xb4/0x120
[    1.436875]  bus_for_each_drv+0x78/0xd0
[    1.436881]  __device_attach_async_helper+0xac/0xd0
[    1.436888]  async_run_entry_fn+0x34/0x110
[    1.436895]  process_one_work+0x1d0/0x354
[    1.436903]  worker_thread+0x13c/0x470
[    1.436910]  kthread+0x150/0x160
[    1.436915]  ret_from_fork+0x10/0x20
[    1.436923] ---[ end trace fcfac44cb045c3a8 ]---

Fix the issue by mapping MMC_VDD_35_36 (and MMC_VDD_34_35) to SDHCI_POWER_330 as well.

---
 drivers/mmc/host/sdhci.c |    6 ++++++
 1 file changed, 6 insertions(+)

@@ -2043,6 +2043,12 @@ void sdhci_set_power_noreg(struct sdhci_
 			break;
 		case MMC_VDD_32_33:
 		case MMC_VDD_33_34:
+		/*
+		 * 3.4 ~ 3.6V are valid only for those platforms where it's
+		 * known that the voltage range is supported by hardware.
+		 */
+		case MMC_VDD_34_35:
+		case MMC_VDD_35_36:
 			pwr = SDHCI_POWER_330;
 			break;
 		default:

From: Sagi Grimberg <[email protected]>

commit 25e1f67eda4a19c91dc05c84d6d413c53efb447b upstream.

We should not access request members after the last send, even to determine if indeed it was the last data payload send. The reason is that a completion could have arrived and trigger a new execution of the request which overridden these members. This was fixed by commit 825619b09ad3 ("nvme-tcp: fix possible use-after-completion").

Commit e371af033c56 broke that assumption again to address cases where multiple r2t pdus are sent per request. To fix it, we need to record the request data_sent and data_len and after the payload network send we reference these counters to determine weather we should advance the request iterator.

---
 drivers/nvme/host/tcp.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

@@ -910,12 +910,14 @@ static void nvme_tcp_fail_request(struct
 static int nvme_tcp_try_send_data(struct nvme_tcp_request *req)
 {
 	struct nvme_tcp_queue *queue = req->queue;
+	int req_data_len = req->data_len;
 
 	while (true) {
 		struct page *page = nvme_tcp_req_cur_page(req);
 		size_t offset = nvme_tcp_req_cur_offset(req);
 		size_t len = nvme_tcp_req_cur_length(req);
 		bool last = nvme_tcp_pdu_last_send(req, len);
+		int req_data_sent = req->data_sent;
 		int ret, flags = MSG_DONTWAIT;
 
 		if (last && !queue->data_digest && !nvme_tcp_queue_more(queue))

@@ -942,7 +944,7 @@ static int nvme_tcp_try_send_data(struct
 		 * in the request where we don't want to modify it as we may
 		 * compete with the RX path completing the request.
 		 */
-		if (req->data_sent + ret < req->data_len)
+		if (req_data_sent + ret < req_data_len)
 			nvme_tcp_advance_req(req, ret);
 
 		/* fully successful last send in current PDU */

From: Gautham Ananthakrishna <[email protected]>

commit 6f1b228529ae49b0f85ab89bcdb6c365df401558 upstream.

Encountered a race between ocfs2_test_bg_bit_allocatable() and jbd2_journal_put_journal_head() resulting in the below vmcore.

PID: 106879 TASK: ffff880244ba9c00 CPU: 2 COMMAND: "loop3"

  Call trace:
    panic
    oops_end
    no_context
    __bad_area_nosemaphore
    bad_area_nosemaphore
    __do_page_fault
    do_page_fault
    page_fault
      [exception RIP: ocfs2_block_group_find_clear_bits+316]
    ocfs2_block_group_find_clear_bits [ocfs2]
    ocfs2_cluster_group_search [ocfs2]
    ocfs2_search_chain [ocfs2]
    ocfs2_claim_suballoc_bits [ocfs2]
    __ocfs2_claim_clusters [ocfs2]
    ocfs2_claim_clusters [ocfs2]
    ocfs2_local_alloc_slide_window [ocfs2]
    ocfs2_reserve_local_alloc_bits [ocfs2]
    ocfs2_reserve_clusters_with_limit [ocfs2]
    ocfs2_reserve_clusters [ocfs2]
    ocfs2_lock_refcount_allocators [ocfs2]
    ocfs2_make_clusters_writable [ocfs2]
    ocfs2_replace_cow [ocfs2]
    ocfs2_refcount_cow [ocfs2]
    ocfs2_file_write_iter [ocfs2]
    lo_rw_aio
    loop_queue_work
    kthread_worker_fn
    kthread
    ret_from_fork

When ocfs2_test_bg_bit_allocatable() called bh2jh(bg_bh), the bg_bh->b_private NULL as jbd2_journal_put_journal_head() raced and released the jounal head from the buffer head. Needed to take bit lock for the bit 'BH_JournalHead' to fix this race.

---
 fs/ocfs2/suballoc.c |   22 +++++++++++++---------
 1 file changed, 13 insertions(+), 9 deletions(-)

@@ -1253,7 +1253,7 @@ static int ocfs2_test_bg_bit_allocatable
 {
 	struct ocfs2_group_desc *bg = (struct ocfs2_group_desc *) bg_bh->b_data;
 	struct journal_head *jh;
-	int ret;
+	int ret = 1;
 
 	if (ocfs2_test_bit(nr, (unsigned long *)bg->bg_bitmap))
 		return 0;

@@ -1261,14 +1261,18 @@ static int ocfs2_test_bg_bit_allocatable
 	if (!buffer_jbd(bg_bh))
 		return 1;
 
-	jh = bh2jh(bg_bh);
-	spin_lock(&jh->b_state_lock);
-	bg = (struct ocfs2_group_desc *) jh->b_committed_data;
-	if (bg)
-		ret = !ocfs2_test_bit(nr, (unsigned long *)bg->bg_bitmap);
-	else
-		ret = 1;
-	spin_unlock(&jh->b_state_lock);
+	jbd_lock_bh_journal_head(bg_bh);
+	if (buffer_jbd(bg_bh)) {
+		jh = bh2jh(bg_bh);
+		spin_lock(&jh->b_state_lock);
+		bg = (struct ocfs2_group_desc *) jh->b_committed_data;
+		if (bg)
+			ret = !ocfs2_test_bit(nr, (unsigned long *)bg->bg_bitmap);
+		else
+			ret = 1;
+		spin_unlock(&jh->b_state_lock);
+	}
+	jbd_unlock_bh_journal_head(bg_bh);
 
 	return ret;
 }

From: Johannes Berg <[email protected]>

commit a2083eeb119fb9307258baea9b7c243ca9a2e0b6 upstream.

The SSID pointer is pointing to RCU protected data, so we need to have it under rcu_read_lock() for the entire use. Fix this.

---
 net/wireless/scan.c |    7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

@@ -418,14 +418,17 @@ cfg80211_add_nontrans_list(struct cfg802
 	}
 	ssid_len = ssid[1];
 	ssid = ssid + 2;
-	rcu_read_unlock();
 
 	/* check if nontrans_bss is in the list */
 	list_for_each_entry(bss, &trans_bss->nontrans_list, nontrans_list) {
-		if (is_bss(bss, nontrans_bss->bssid, ssid, ssid_len))
+		if (is_bss(bss, nontrans_bss->bssid, ssid, ssid_len)) {
+			rcu_read_unlock();
 			return 0;
+		}
 	}
 
+	rcu_read_unlock();
+
 	/* add to the list */
 	list_add_tail(&nontrans_bss->nontrans_list, &trans_bss->nontrans_list);
 	return 0;

From: Johannes Berg <[email protected]>

commit 09b1d5dc6ce1c9151777f6c4e128a59457704c97 upstream.

The management registrations locking was broken, the list was locked for each wdev, but cfg80211_mgmt_registrations_update() iterated it without holding all the correct spinlocks, causing list corruption.

Rather than trying to fix it with fine-grained locking, just move the lock to the wiphy/rdev (still need the list on each wdev), we already need to hold the wdev lock to change it, so there's no contention on the lock in any case. This trivially fixes the bug since we hold one wdev's lock already, and now will hold the lock that protects all lists.

---
 include/net/cfg80211.h |    2 --
 net/wireless/core.c    |    2 +-
 net/wireless/core.h    |    2 ++
 net/wireless/mlme.c    |   26 ++++++++++++++------------
 4 files changed, 17 insertions(+), 15 deletions(-)

@@ -5202,7 +5202,6 @@ struct cfg80211_cqm_config;
  *	netdev and may otherwise be used by driver read-only, will be update
  *	by cfg80211 on change_interface
  * @mgmt_registrations: list of registrations for management frames
- * @mgmt_registrations_lock: lock for the list
  * @mgmt_registrations_need_update: mgmt registrations were updated,
  *	need to propagate the update to the driver
  * @mtx: mutex used to lock data in this struct, may be used by drivers

@@ -5249,7 +5248,6 @@ struct wireless_dev {
 	u32 identifier;
 
 	struct list_head mgmt_registrations;
-	spinlock_t mgmt_registrations_lock;
 	u8 mgmt_registrations_need_update:1;
 
 	struct mutex mtx;

@@ -501,6 +501,7 @@ use_default_name:
 	INIT_WORK(&rdev->propagate_cac_done_wk, cfg80211_propagate_cac_done_wk);
 	INIT_WORK(&rdev->mgmt_registrations_update_wk,
 		  cfg80211_mgmt_registrations_update_wk);
+	spin_lock_init(&rdev->mgmt_registrations_lock);
 
 #ifdef CONFIG_CFG80211_DEFAULT_PS
 	rdev->wiphy.flags |= WIPHY_FLAG_PS_ON_BY_DEFAULT;

@@ -1256,7 +1257,6 @@ void cfg80211_init_wdev(struct wireless_
 	INIT_LIST_HEAD(&wdev->event_list);
 	spin_lock_init(&wdev->event_lock);
 	INIT_LIST_HEAD(&wdev->mgmt_registrations);
-	spin_lock_init(&wdev->mgmt_registrations_lock);
 	INIT_LIST_HEAD(&wdev->pmsr_list);
 	spin_lock_init(&wdev->pmsr_lock);
 	INIT_WORK(&wdev->pmsr_free_wk, cfg80211_pmsr_free_wk);

@@ -101,6 +101,8 @@ struct cfg80211_registered_device {
 	struct work_struct propagate_cac_done_wk;
 
 	struct work_struct mgmt_registrations_update_wk;
+	/* lock for all wdev lists */
+	spinlock_t mgmt_registrations_lock;
 
 	/* must be last because of the way we do wiphy_priv(),
 	 * and it should at least be aligned to NETDEV_ALIGN */

@@ -448,9 +448,9 @@ static void cfg80211_mgmt_registrations_
 
 	ASSERT_RTNL();
 
-	spin_lock_bh(&wdev->mgmt_registrations_lock);
+	spin_lock_bh(&rdev->mgmt_registrations_lock);
 	if (!wdev->mgmt_registrations_need_update) {
-		spin_unlock_bh(&wdev->mgmt_registrations_lock);
+		spin_unlock_bh(&rdev->mgmt_registrations_lock);
 		return;
 	}
 

@@ -475,7 +475,7 @@ static void cfg80211_mgmt_registrations_
 	rcu_read_unlock();
 
 	wdev->mgmt_registrations_need_update = 0;
-	spin_unlock_bh(&wdev->mgmt_registrations_lock);
+	spin_unlock_bh(&rdev->mgmt_registrations_lock);
 
 	rdev_update_mgmt_frame_registrations(rdev, wdev, &upd);
 }

@@ -499,6 +499,7 @@ int cfg80211_mlme_register_mgmt(struct w
 				int match_len, bool multicast_rx,
 				struct netlink_ext_ack *extack)
 {
+	struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy);
 	struct cfg80211_mgmt_registration *reg, *nreg;
 	int err = 0;
 	u16 mgmt_type;

@@ -544,7 +545,7 @@ int cfg80211_mlme_register_mgmt(struct w
 	if (!nreg)
 		return -ENOMEM;
 
-	spin_lock_bh(&wdev->mgmt_registrations_lock);
+	spin_lock_bh(&rdev->mgmt_registrations_lock);
 
 	list_for_each_entry(reg, &wdev->mgmt_registrations, list) {
 		int mlen = min(match_len, reg->match_len);

@@ -579,7 +580,7 @@ int cfg80211_mlme_register_mgmt(struct w
 		list_add(&nreg->list, &wdev->mgmt_registrations);
 	}
 	wdev->mgmt_registrations_need_update = 1;
-	spin_unlock_bh(&wdev->mgmt_registrations_lock);
+	spin_unlock_bh(&rdev->mgmt_registrations_lock);
 
 	cfg80211_mgmt_registrations_update(wdev);
 

@@ -587,7 +588,7 @@ int cfg80211_mlme_register_mgmt(struct w
 
  out:
 	kfree(nreg);
-	spin_unlock_bh(&wdev->mgmt_registrations_lock);
+	spin_unlock_bh(&rdev->mgmt_registrations_lock);
 
 	return err;
 }

@@ -598,7 +599,7 @@ void cfg80211_mlme_unregister_socket(str
 	struct cfg80211_registered_device *rdev = wiphy_to_rdev(wiphy);
 	struct cfg80211_mgmt_registration *reg, *tmp;
 
-	spin_lock_bh(&wdev->mgmt_registrations_lock);
+	spin_lock_bh(&rdev->mgmt_registrations_lock);
 
 	list_for_each_entry_safe(reg, tmp, &wdev->mgmt_registrations, list) {
 		if (reg->nlportid != nlportid)

@@ -611,7 +612,7 @@ void cfg80211_mlme_unregister_socket(str
 		schedule_work(&rdev->mgmt_registrations_update_wk);
 	}
 
-	spin_unlock_bh(&wdev->mgmt_registrations_lock);
+	spin_unlock_bh(&rdev->mgmt_registrations_lock);
 
 	if (nlportid && rdev->crit_proto_nlportid == nlportid) {
 		rdev->crit_proto_nlportid = 0;

@@ -624,15 +625,16 @@ void cfg80211_mlme_unregister_socket(str
 
 void cfg80211_mlme_purge_registrations(struct wireless_dev *wdev)
 {
+	struct cfg80211_registered_device *rdev = wiphy_to_rdev(wdev->wiphy);
 	struct cfg80211_mgmt_registration *reg, *tmp;
 
-	spin_lock_bh(&wdev->mgmt_registrations_lock);
+	spin_lock_bh(&rdev->mgmt_registrations_lock);
 	list_for_each_entry_safe(reg, tmp, &wdev->mgmt_registrations, list) {
 		list_del(&reg->list);
 		kfree(reg);
 	}
 	wdev->mgmt_registrations_need_update = 1;
-	spin_unlock_bh(&wdev->mgmt_registrations_lock);
+	spin_unlock_bh(&rdev->mgmt_registrations_lock);
 
 	cfg80211_mgmt_registrations_update(wdev);
 }

@@ -780,7 +782,7 @@ bool cfg80211_rx_mgmt_khz(struct wireles
 	data = buf + ieee80211_hdrlen(mgmt->frame_control);
 	data_len = len - ieee80211_hdrlen(mgmt->frame_control);
 
-	spin_lock_bh(&wdev->mgmt_registrations_lock);
+	spin_lock_bh(&rdev->mgmt_registrations_lock);
 
 	list_for_each_entry(reg, &wdev->mgmt_registrations, list) {
 		if (reg->frame_type != ftype)

@@ -804,7 +806,7 @@ bool cfg80211_rx_mgmt_khz(struct wireles
 		break;
 	}
 
-	spin_unlock_bh(&wdev->mgmt_registrations_lock);
+	spin_unlock_bh(&rdev->mgmt_registrations_lock);
 
 	trace_cfg80211_return_bool(result);
 	return result;

From: Arnd Bergmann <[email protected]>

commit eaf6cc7165c9c5aa3c2f9faa03a98598123d0afb upstream.

Both the decompressor code and the kasan logic try to override the memcpy() and memmove() definitions, which leading to a clash in a KASAN-enabled kernel with XZ decompression:

arch/arm/boot/compressed/decompress.c:50:9: error: 'memmove' macro redefined [-Werror,-Wmacro-redefined]

 #define memmove memmove
        ^
arch/arm/include/asm/string.h:59:9: note: previous definition is here
 #define memmove(dst, src, len) __memmove(dst, src, len)
        ^
arch/arm/boot/compressed/decompress.c:51:9: error: 'memcpy' macro redefined [-Werror,-Wmacro-redefined]
 #define memcpy memcpy
        ^
arch/arm/include/asm/string.h:58:9: note: previous definition is here
 #define memcpy(dst, src, len) __memcpy(dst, src, len)
        ^

Here we want the set of functions from the decompressor, so undefine the other macros before the override.

---
 arch/arm/boot/compressed/decompress.c |    3 +++
 1 file changed, 3 insertions(+)

@@ -47,7 +47,10 @@ extern char * strchrnul(const char *, in
 #endif
 
 #ifdef CONFIG_KERNEL_XZ
+/* Prevent KASAN override of string helpers in decompressor */
+#undef memmove
 #define memmove memmove
+#undef memcpy
 #define memcpy memcpy
 #include "../../../../lib/decompress_unxz.c"
 #endif

From: Jaehoon Chung <[email protected]>

commit 697542bceae51f7620af333b065dd09d213629fb upstream.

Even though there are candiates value if can't find best value, it's returned -EIO. It's not proper behavior.
If there is not best value, use a first candiate value to work eMMC.

---
 drivers/mmc/host/dw_mmc-exynos.c |   14 ++++++++++++++
 1 file changed, 14 insertions(+)

@@ -464,6 +464,18 @@ static s8 dw_mci_exynos_get_best_clksmpl
 		}
 	}
 
+	/*
+	 * If there is no cadiates value, then it needs to return -EIO.
+	 * If there are candiates values and don't find bset clk sample value,
+	 * then use a first candiates clock sample value.
+	 */
+	for (i = 0; i < iter; i++) {
+		__c = ror8(candiates, i);
+		if ((__c & 0x1) == 0x1) {
+			loc = i;
+			goto out;
+		}
+	}
 out:
 	return loc;
 }

@@ -494,6 +506,8 @@ static int dw_mci_exynos_execute_tuning(
 		priv->tuned_sample = found;
 	} else {
 		ret = -EIO;
+		dev_warn(&mmc->class_dev,
+			"There is no candiates value about clksmpl!\n");
 	}
 
 	return ret;

From: Haibo Chen <[email protected]>

commit 9af372dc70e9fdcbb70939dac75365e7b88580b4 upstream.

To reset standard tuning circuit completely, after clear ESDHC_MIX_CTRL_EXE_TUNE, also need to clear bit buffer_read_ready, this operation will finally clear the USDHC IP internal logic flag execute_tuning_with_clr_buf, make sure the following normal data transfer will not be impacted by standard tuning logic used before.

Find this issue when do quick SD card insert/remove stress test. During standard tuning prodedure, if remove SD card, USDHC standard tuning logic can't clear the internal flag execute_tuning_with_clr_buf. Next time when insert SD card, all data related commands can't get any data related interrupts, include data transfer complete interrupt, data timeout interrupt, data CRC interrupt, data end bit interrupt. Always trigger software timeout issue. Even reset the USDHC through bits in register SYS_CTRL (0x2C, bit28 reset tuning, bit26 reset data, bit 25 reset command, bit 24 reset all) can't recover this. From the user's point of view, USDHC stuck, SD can't be recognized any more.

---
 drivers/mmc/host/sdhci-esdhc-imx.c |   16 ++++++++++++++++
 1 file changed, 16 insertions(+)

@@ -1157,6 +1157,7 @@ static void esdhc_reset_tuning(struct sd
 	struct sdhci_pltfm_host *pltfm_host = sdhci_priv(host);
 	struct pltfm_imx_data *imx_data = sdhci_pltfm_priv(pltfm_host);
 	u32 ctrl;
+	int ret;
 
 	/* Reset the tuning circuit */
 	if (esdhc_is_usdhc(imx_data)) {

@@ -1169,7 +1170,22 @@ static void esdhc_reset_tuning(struct sd
 		} else if (imx_data->socdata->flags & ESDHC_FLAG_STD_TUNING) {
 			ctrl = readl(host->ioaddr + SDHCI_AUTO_CMD_STATUS);
 			ctrl &= ~ESDHC_MIX_CTRL_SMPCLK_SEL;
+			ctrl &= ~ESDHC_MIX_CTRL_EXE_TUNE;
 			writel(ctrl, host->ioaddr + SDHCI_AUTO_CMD_STATUS);
+			/* Make sure ESDHC_MIX_CTRL_EXE_TUNE cleared */
+			ret = readl_poll_timeout(host->ioaddr + SDHCI_AUTO_CMD_STATUS,
+				ctrl, !(ctrl & ESDHC_MIX_CTRL_EXE_TUNE), 1, 50);
+			if (ret == -ETIMEDOUT)
+				dev_warn(mmc_dev(host->mmc),
+				 "Warning! clear execute tuning bit failed\n");
+			/*
+			 * SDHCI_INT_DATA_AVAIL is W1C bit, set this bit will clear the
+			 * usdhc IP internal logic flag execute_tuning_with_clr_buf, which
+			 * will finally make sure the normal data transfer logic correct.
+			 */
+			ctrl = readl(host->ioaddr + SDHCI_INT_STATUS);
+			ctrl |= SDHCI_INT_DATA_AVAIL;
+			writel(ctrl, host->ioaddr + SDHCI_INT_STATUS);
 		}
 	}
 }

From: Arnd Bergmann <[email protected]>

commit 44cc6412e66b2b84544eaf2e14cf1764301e2a80 upstream.

When frame pointers are used instead of the ARM unwinder, and the kernel is built using clang with an external assembler and CONFIG_XIP_KERNEL, every file produces two warnings like:

arm-linux-gnueabi-ld: warning: orphan section .ARM.extab' from net/mac802154/util.o' being placed in section `.ARM.extab'
arm-linux-gnueabi-ld: warning: orphan section .ARM.exidx' from net/mac802154/util.o' being placed in section `.ARM.exidx'

The same fix was already merged for the normal (non-XIP)

linker script, with a longer description.

---
 arch/arm/kernel/vmlinux-xip.lds.S |    4 ++++
 1 file changed, 4 insertions(+)

@@ -40,6 +40,10 @@ SECTIONS
 		ARM_DISCARD
 		*(.alt.smp.init)
 		*(.pv_table)
+#ifndef CONFIG_ARM_UNWIND
+		*(.ARM.exidx) *(.ARM.exidx.*)
+		*(.ARM.extab) *(.ARM.extab.*)
+#endif
 	}
 
 	. = XIP_VIRT_ADDR(CONFIG_XIP_PHYS_ADDR);

From: Arnd Bergmann <[email protected]>

commit 48ccc8edf5b90622cdc4f8878e0042ab5883e2ca upstream.

In randconfig builds, we sometimes come across this warning:

arm-linux-gnueabi-ld: XIP start address may cause MPU programming issues

While this is helpful for actual systems to figure out why it fails, the warning does not provide any benefit for build testing, so guard it in a check for CONFIG_COMPILE_TEST, which is usually set on randconfig builds.

---
 arch/arm/kernel/vmlinux-xip.lds.S |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -176,7 +176,7 @@ ASSERT((__arch_info_end - __arch_info_be
 ASSERT((_end - __bss_start) >= 12288, ".bss too small for CONFIG_XIP_DEFLATED_DATA")
 #endif
 
-#ifdef CONFIG_ARM_MPU
+#if defined(CONFIG_ARM_MPU) && !defined(CONFIG_COMPILE_TEST)
 /*
  * Due to PMSAv7 restriction on base address and size we have to
  * enforce minimal alignment restrictions. It was seen that weaker

From: Arnd Bergmann <[email protected]>

commit 1f323127cab086e4fd618981b1e5edc396eaf0f4 upstream.

With extra warnings enabled, gcc complains about this function definition:

arch/arm/probes/kprobes/core.c: In function 'arch_init_kprobes':
arch/arm/probes/kprobes/core.c:465:12: warning: old-style function definition [-Wold-style-definition]
465 | int __init arch_init_kprobes()

---
 arch/arm/probes/kprobes/core.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -462,7 +462,7 @@ static struct undef_hook kprobes_arm_bre
 
 #endif /* !CONFIG_THUMB2_KERNEL */
 
-int __init arch_init_kprobes()
+int __init arch_init_kprobes(void)
 {
 	arm_probes_decode_init();
 #ifdef CONFIG_THUMB2_KERNEL

From: Naveen N. Rao <[email protected]>

commit 8bbc9d822421d9ac8ff9ed26a3713c9afc69d6c8 upstream.

Only ignore the operation if dividing by 1.

---
 arch/powerpc/net/bpf_jit_comp64.c |   10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

@@ -415,8 +415,14 @@ static int bpf_jit_build_body(struct bpf
 		case BPF_ALU64 | BPF_DIV | BPF_K: /* dst /= imm */
 			if (imm == 0)
 				return -EINVAL;
-			else if (imm == 1)
-				goto bpf_alu32_trunc;
+			if (imm == 1) {
+				if (BPF_OP(code) == BPF_DIV) {
+					goto bpf_alu32_trunc;
+				} else {
+					EMIT(PPC_RAW_LI(dst_reg, 0));
+					break;
+				}
+			}
 
 			PPC_LI32(b2p[TMP_REG_1], imm);
 			switch (BPF_CLASS(code)) {

From: Pavel Begunkov <[email protected]>

commit 792bb6eb862333658bf1bd2260133f0507e2da8d upstream.

[   97.866748] a.out/2890 is trying to acquire lock:
[   97.867829] ffff8881046763e8 (&ctx->uring_lock){+.+.}-{3:3}, at:
io_wq_submit_work+0x155/0x240
[   97.869735]
[   97.869735] but task is already holding lock:
[   97.871033] ffff88810dfe0be8 (&ctx->uring_lock){+.+.}-{3:3}, at:
__x64_sys_io_uring_enter+0x3f0/0x5b0
[   97.873074]
[   97.873074] other info that might help us debug this:
[   97.874520]  Possible unsafe locking scenario:
[   97.874520]
[   97.875845]        CPU0
[   97.876440]        ----
[   97.877048]   lock(&ctx->uring_lock);
[   97.877961]   lock(&ctx->uring_lock);
[   97.878881]
[   97.878881]  *** DEADLOCK ***
[   97.878881]
[   97.880341]  May be due to missing lock nesting notation
[   97.880341]
[   97.881952] 1 lock held by a.out/2890:
[   97.882873]  #0: ffff88810dfe0be8 (&ctx->uring_lock){+.+.}-{3:3}, at:
__x64_sys_io_uring_enter+0x3f0/0x5b0
[   97.885108]
[   97.885108] stack backtrace:
[   97.890457] Call Trace:
[   97.891121]  dump_stack+0xac/0xe3
[   97.891972]  __lock_acquire+0xab6/0x13a0
[   97.892940]  lock_acquire+0x2c3/0x390
[   97.894894]  __mutex_lock+0xae/0x9f0
[   97.901101]  io_wq_submit_work+0x155/0x240
[   97.902112]  io_wq_cancel_cb+0x162/0x490
[   97.904126]  io_async_find_and_cancel+0x3b/0x140
[   97.905247]  io_issue_sqe+0x86d/0x13e0
[   97.909122]  __io_queue_sqe+0x10b/0x550
[   97.913971]  io_queue_sqe+0x235/0x470
[   97.914894]  io_submit_sqes+0xcce/0xf10
[   97.917872]  __x64_sys_io_uring_enter+0x3fb/0x5b0
[   97.921424]  do_syscall_64+0x2d/0x40
[   97.922329]  entry_SYSCALL_64_after_hwframe+0x44/0xa9

While holding uring_lock, e.g. from inline execution, async cancel request may attempt cancellations through io_wq_submit_work, which may try to grab a lock. Delay it to task_work, so we do it from a clean context and don't have to worry about locking.

[Lee: The first hunk solves a different (double free) issue in v5.10.
Only the first hunk of the original patch is relevant to v5.10 AND the first hunk of the original patch is only relevant to v5.10]

---
 fs/io_uring.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -2075,7 +2075,9 @@ static void io_req_task_cancel(struct ca
 	struct io_kiocb *req = container_of(cb, struct io_kiocb, task_work);
 	struct io_ring_ctx *ctx = req->ctx;
 
+	mutex_lock(&ctx->uring_lock);
 	__io_req_task_cancel(req, -ECANCELED);
+	mutex_unlock(&ctx->uring_lock);
 	percpu_ref_put(&ctx->refs);
 }
 

From: Rongwei Wang <[email protected]>

commit 74c42e1baacf206338b1dd6b6199ac964512b5bb upstream.

Currently collapse_file does not explicitly check PG_writeback, instead, page_has_private and try_to_release_page are used to filter writeback pages. This does not work for xfs with blocksize equal to or larger than pagesize, because in such case xfs has no page->private.

This makes collapse_file bail out early for writeback page. Otherwise, xfs end_page_writeback will panic as follows.

page:fffffe00201bcc80 refcount:0 mapcount:0 mapping:ffff0003f88c86a8 index:0x0 pfn:0x84ef32 aops:xfs_address_space_operations [xfs] ino:30000b7 dentry name:"libtest.so"
flags: 0x57fffe0000008027(locked|referenced|uptodate|active|writeback)

  raw: 57fffe0000008027 ffff80001b48bc28 ffff80001b48bc28 ffff0003f88c86a8
  raw: 0000000000000000 0000000000000000 00000000ffffffff ffff0000c3e9a000
  page dumped because: VM_BUG_ON_PAGE(((unsigned int) page_ref_count(page) + 127u <= 127u))
  page->mem_cgroup:ffff0000c3e9a000
  ------------[ cut here ]------------
  kernel BUG at include/linux/mm.h:1212!
  Internal error: Oops - BUG: 0 [#1] SMP
  Modules linked in:
  BUG: Bad page state in process khugepaged  pfn:84ef32
   xfs(E)
  page:fffffe00201bcc80 refcount:0 mapcount:0 mapping:0 index:0x0 pfn:0x84ef32
   libcrc32c(E) rfkill(E) aes_ce_blk(E) crypto_simd(E) ...
  CPU: 25 PID: 0 Comm: swapper/25 Kdump: loaded Tainted: ...
  pstate: 60400005 (nZCv daif +PAN -UAO -TCO BTYPE=--)
  Call trace:
    end_page_writeback+0x1c0/0x214
    iomap_finish_page_writeback+0x13c/0x204
    iomap_finish_ioend+0xe8/0x19c
    iomap_writepage_end_bio+0x38/0x50
    bio_endio+0x168/0x1ec
    blk_update_request+0x278/0x3f0
    blk_mq_end_request+0x34/0x15c
    virtblk_request_done+0x38/0x74 [virtio_blk]
    blk_done_softirq+0xc4/0x110
    __do_softirq+0x128/0x38c
    __irq_exit_rcu+0x118/0x150
    irq_exit+0x1c/0x30
    __handle_domain_irq+0x8c/0xf0
    gic_handle_irq+0x84/0x108
    el1_irq+0xcc/0x180
    arch_cpu_idle+0x18/0x40
    default_idle_call+0x4c/0x1a0
    cpuidle_idle_call+0x168/0x1e0
    do_idle+0xb4/0x104
    cpu_startup_entry+0x30/0x9c
    secondary_start_kernel+0x104/0x180
  Code: d4210000 b0006161 910c8021 94013f4d (d4210000)
  ---[ end trace 4a88c6a074082f8c ]---
  Kernel panic - not syncing: Oops - BUG: Fatal exception in interrupt

---
 mm/khugepaged.c |    7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

@@ -1758,6 +1758,10 @@ static void collapse_file(struct mm_stru
 				filemap_flush(mapping);
 				result = SCAN_FAIL;
 				goto xa_unlocked;
+			} else if (PageWriteback(page)) {
+				xas_unlock_irq(&xas);
+				result = SCAN_FAIL;
+				goto xa_unlocked;
 			} else if (trylock_page(page)) {
 				get_page(page);
 				xas_unlock_irq(&xas);

@@ -1793,7 +1797,8 @@ static void collapse_file(struct mm_stru
 			goto out_unlock;
 		}
 
-		if (!is_shmem && PageDirty(page)) {
+		if (!is_shmem && (PageDirty(page) ||
+				  PageWriteback(page))) {
 			/*
 			 * khugepaged only works on read-only fd, so this
 			 * page is dirty because it hasn't been flushed

From: Johan Hovold <[email protected]>

commit db6c3c064f5d55fa9969f33eafca3cdbefbb3541 upstream.

Add the missing endpoint max-packet sanity check to probe() to avoid division by zero in lan78xx_tx_bh() in case a malicious device has broken descriptors (or when doing descriptor fuzz testing).

Note that USB core will reject URBs submitted for endpoints with zero wMaxPacketSize but that drivers doing packet-size calculations still need to handle this (cf. commit 2548288b4fb0 ("USB: Fix: Don't skip endpoint descriptors with maxpacket=0")).

---
 drivers/net/usb/lan78xx.c |    6 ++++++
 1 file changed, 6 insertions(+)

@@ -3745,6 +3745,12 @@ static int lan78xx_probe(struct usb_inte
 
 	dev->maxpacket = usb_maxpacket(dev->udev, dev->pipe_out, 1);
 
+	/* Reject broken descriptors. */
+	if (dev->maxpacket == 0) {
+		ret = -ENODEV;
+		goto out4;
+	}
+
 	/* driver requires remote-wakeup capability during autosuspend. */
 	intf->needs_remote_wakeup = 1;
 

From: Robin Murphy <[email protected]>

commit 295cf156231ca3f9e3a66bde7fab5e09c41835e0 upstream.

Al reminds us that the usercopy API must only return complete failure if absolutely nothing could be copied. Currently, if userspace does something silly like giving us an unaligned pointer to Device memory, or a size which overruns MTE tag bounds, we may fail to honour that requirement when faulting on a multi-byte access even though a smaller access could have succeeded.

Add a mitigation to the fixup routines to fall back to a single-byte copy if we faulted on a larger access before anything has been written to the destination, to guarantee making *some* forward progress. We needn't be too concerned about the overall performance since this should only occur when callers are doing something a bit dodgy in the first place. Particularly broken userspace might still be able to trick generic_perform_write() into an infinite loop by targeting write() at an mmap() of some read-only device register where the fault-in load succeeds but any store synchronously aborts such that copy_to_user() is genuinely unable to make progress, but, well, don't do that...

---
 arch/arm64/lib/copy_from_user.S |   13 ++++++++++---
 arch/arm64/lib/copy_in_user.S   |   21 ++++++++++++++-------
 arch/arm64/lib/copy_to_user.S   |   14 +++++++++++---
 3 files changed, 35 insertions(+), 13 deletions(-)

@@ -29,7 +29,7 @@
 	.endm
 
 	.macro ldrh1 reg, ptr, val
-	uao_user_alternative 9998f, ldrh, ldtrh, \reg, \ptr, \val
+	uao_user_alternative 9997f, ldrh, ldtrh, \reg, \ptr, \val
 	.endm
 
 	.macro strh1 reg, ptr, val

@@ -37,7 +37,7 @@
 	.endm
 
 	.macro ldr1 reg, ptr, val
-	uao_user_alternative 9998f, ldr, ldtr, \reg, \ptr, \val
+	uao_user_alternative 9997f, ldr, ldtr, \reg, \ptr, \val
 	.endm
 
 	.macro str1 reg, ptr, val

@@ -45,7 +45,7 @@
 	.endm
 
 	.macro ldp1 reg1, reg2, ptr, val
-	uao_ldp 9998f, \reg1, \reg2, \ptr, \val
+	uao_ldp 9997f, \reg1, \reg2, \ptr, \val
 	.endm
 
 	.macro stp1 reg1, reg2, ptr, val

@@ -53,8 +53,10 @@
 	.endm
 
 end	.req	x5
+srcin	.req	x15
 SYM_FUNC_START(__arch_copy_from_user)
 	add	end, x0, x2
+	mov	srcin, x1
 #include "copy_template.S"
 	mov	x0, #0				// Nothing to copy
 	ret

@@ -63,6 +65,11 @@ EXPORT_SYMBOL(__arch_copy_from_user)
 
 	.section .fixup,"ax"
 	.align	2
+9997:	cmp	dst, dstin
+	b.ne	9998f
+	// Before being absolutely sure we couldn't copy anything, try harder
+USER(9998f, ldtrb tmp1w, [srcin])
+	strb	tmp1w, [dst], #1
 9998:	sub	x0, end, dst			// bytes not copied
 	ret
 	.previous

@@ -30,33 +30,34 @@
 	.endm
 
 	.macro ldrh1 reg, ptr, val
-	uao_user_alternative 9998f, ldrh, ldtrh, \reg, \ptr, \val
+	uao_user_alternative 9997f, ldrh, ldtrh, \reg, \ptr, \val
 	.endm
 
 	.macro strh1 reg, ptr, val
-	uao_user_alternative 9998f, strh, sttrh, \reg, \ptr, \val
+	uao_user_alternative 9997f, strh, sttrh, \reg, \ptr, \val
 	.endm
 
 	.macro ldr1 reg, ptr, val
-	uao_user_alternative 9998f, ldr, ldtr, \reg, \ptr, \val
+	uao_user_alternative 9997f, ldr, ldtr, \reg, \ptr, \val
 	.endm
 
 	.macro str1 reg, ptr, val
-	uao_user_alternative 9998f, str, sttr, \reg, \ptr, \val
+	uao_user_alternative 9997f, str, sttr, \reg, \ptr, \val
 	.endm
 
 	.macro ldp1 reg1, reg2, ptr, val
-	uao_ldp 9998f, \reg1, \reg2, \ptr, \val
+	uao_ldp 9997f, \reg1, \reg2, \ptr, \val
 	.endm
 
 	.macro stp1 reg1, reg2, ptr, val
-	uao_stp 9998f, \reg1, \reg2, \ptr, \val
+	uao_stp 9997f, \reg1, \reg2, \ptr, \val
 	.endm
 
 end	.req	x5
-
+srcin	.req	x15
 SYM_FUNC_START(__arch_copy_in_user)
 	add	end, x0, x2
+	mov	srcin, x1
 #include "copy_template.S"
 	mov	x0, #0
 	ret

@@ -65,6 +66,12 @@ EXPORT_SYMBOL(__arch_copy_in_user)
 
 	.section .fixup,"ax"
 	.align	2
+9997:	cmp	dst, dstin
+	b.ne	9998f
+	// Before being absolutely sure we couldn't copy anything, try harder
+USER(9998f, ldtrb tmp1w, [srcin])
+USER(9998f, sttrb tmp1w, [dst])
+	add	dst, dst, #1
 9998:	sub	x0, end, dst			// bytes not copied
 	ret
 	.previous

@@ -32,7 +32,7 @@
 	.endm
 
 	.macro strh1 reg, ptr, val
-	uao_user_alternative 9998f, strh, sttrh, \reg, \ptr, \val
+	uao_user_alternative 9997f, strh, sttrh, \reg, \ptr, \val
 	.endm
 
 	.macro ldr1 reg, ptr, val

@@ -40,7 +40,7 @@
 	.endm
 
 	.macro str1 reg, ptr, val
-	uao_user_alternative 9998f, str, sttr, \reg, \ptr, \val
+	uao_user_alternative 9997f, str, sttr, \reg, \ptr, \val
 	.endm
 
 	.macro ldp1 reg1, reg2, ptr, val

@@ -48,12 +48,14 @@
 	.endm
 
 	.macro stp1 reg1, reg2, ptr, val
-	uao_stp 9998f, \reg1, \reg2, \ptr, \val
+	uao_stp 9997f, \reg1, \reg2, \ptr, \val
 	.endm
 
 end	.req	x5
+srcin	.req	x15
 SYM_FUNC_START(__arch_copy_to_user)
 	add	end, x0, x2
+	mov	srcin, x1
 #include "copy_template.S"
 	mov	x0, #0
 	ret

@@ -62,6 +64,12 @@ EXPORT_SYMBOL(__arch_copy_to_user)
 
 	.section .fixup,"ax"
 	.align	2
+9997:	cmp	dst, dstin
+	b.ne	9998f
+	// Before being absolutely sure we couldn't copy anything, try harder
+	ldrb	tmp1w, [srcin]
+USER(9998f, sttrb tmp1w, [dst])
+	add	dst, dst, #1
 9998:	sub	x0, end, dst			// bytes not copied
 	ret
 	.previous

From: Varun Prakash <[email protected]>

commit d89b9f3bbb58e9e378881209756b0723694f22ff upstream.

ddgst is of type __le32, &req->ddgst + req->offset increases &req->ddgst by 4 * req->offset, fix this by type casting &req->ddgst to u8 *.

---
 drivers/nvme/host/tcp.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -1040,7 +1040,7 @@ static int nvme_tcp_try_send_ddgst(struc
 	int ret;
 	struct msghdr msg = { .msg_flags = MSG_DONTWAIT };
 	struct kvec iov = {
-		.iov_base = &req->ddgst + req->offset,
+		.iov_base = (u8 *)&req->ddgst + req->offset,
 		.iov_len = NVME_TCP_DIGEST_LENGTH - req->offset
 	};
 

From: Varun Prakash <[email protected]>

commit ce7723e9cdae4eb3030da082876580f4b2dc0861 upstream.

With commit db5ad6b7f8cd ("nvme-tcp: try to send request in queue_rq context") r2t and response PDU can get processed while send function is executing.

Current data digest send code uses req->offset after kernel_sendmsg(), this creates a race condition where req->offset gets reset before it is used in send function.

This can happen in two cases -
1. Target sends r2t PDU which resets req->offset.
2. Target send response PDU which completes the req and then req is used for a new command, nvme_tcp_setup_cmd_pdu() resets req->offset.

Fix this by storing req->offset in a local variable and using this local variable after kernel_sendmsg().

---
 drivers/nvme/host/tcp.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -1037,6 +1037,7 @@ static int nvme_tcp_try_send_data_pdu(st
 static int nvme_tcp_try_send_ddgst(struct nvme_tcp_request *req)
 {
 	struct nvme_tcp_queue *queue = req->queue;
+	size_t offset = req->offset;
 	int ret;
 	struct msghdr msg = { .msg_flags = MSG_DONTWAIT };
 	struct kvec iov = {

@@ -1053,7 +1054,7 @@ static int nvme_tcp_try_send_ddgst(struc
 	if (unlikely(ret <= 0))
 		return ret;
 
-	if (req->offset + ret == NVME_TCP_DIGEST_LENGTH) {
+	if (offset + ret == NVME_TCP_DIGEST_LENGTH) {
 		nvme_tcp_done_send_req(queue);
 		return 1;
 	}

From: Rakesh Babu <[email protected]>

commit e77bcdd1f639809950c45234b08647ac6d3ffe7b upstream.

Currently, we are using a fixed buffer size of length 2048 to display rsrc_alloc output. As a result a maximum of 2048 characters of rsrc_alloc output is displayed, which may lead sometimes to display only partial output. This patch fixes this dependency on max limit of buffer size and displays all PF VF entries.

Each column of the debugfs entry "rsrc_alloc" uses a fixed width of 12 characters to print the list of LFs of each block for a PF/VF. If the length of list of LFs of a block exceeds this fixed width then the list gets truncated and displays only a part of the list. This patch fixes this by using the maximum possible length of list of LFs among all blocks of all PFs and VFs entries as the width size.

---
 drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c |  138 ++++++++++++----
 1 file changed, 106 insertions(+), 32 deletions(-)

@@ -139,18 +139,85 @@ static const struct file_operations rvu_
 
 static void print_nix_qsize(struct seq_file *filp, struct rvu_pfvf *pfvf);
 
+static void get_lf_str_list(struct rvu_block block, int pcifunc,
+			    char *lfs)
+{
+	int lf = 0, seq = 0, len = 0, prev_lf = block.lf.max;
+
+	for_each_set_bit(lf, block.lf.bmap, block.lf.max) {
+		if (lf >= block.lf.max)
+			break;
+
+		if (block.fn_map[lf] != pcifunc)
+			continue;
+
+		if (lf == prev_lf + 1) {
+			prev_lf = lf;
+			seq = 1;
+			continue;
+		}
+
+		if (seq)
+			len += sprintf(lfs + len, "-%d,%d", prev_lf, lf);
+		else
+			len += (len ? sprintf(lfs + len, ",%d", lf) :
+				      sprintf(lfs + len, "%d", lf));
+
+		prev_lf = lf;
+		seq = 0;
+	}
+
+	if (seq)
+		len += sprintf(lfs + len, "-%d", prev_lf);
+
+	lfs[len] = '\0';
+}
+
+static int get_max_column_width(struct rvu *rvu)
+{
+	int index, pf, vf, lf_str_size = 12, buf_size = 256;
+	struct rvu_block block;
+	u16 pcifunc;
+	char *buf;
+
+	buf = kzalloc(buf_size, GFP_KERNEL);
+	if (!buf)
+		return -ENOMEM;
+
+	for (pf = 0; pf < rvu->hw->total_pfs; pf++) {
+		for (vf = 0; vf <= rvu->hw->total_vfs; vf++) {
+			pcifunc = pf << 10 | vf;
+			if (!pcifunc)
+				continue;
+
+			for (index = 0; index < BLK_COUNT; index++) {
+				block = rvu->hw->block[index];
+				if (!strlen(block.name))
+					continue;
+
+				get_lf_str_list(block, pcifunc, buf);
+				if (lf_str_size <= strlen(buf))
+					lf_str_size = strlen(buf) + 1;
+			}
+		}
+	}
+
+	kfree(buf);
+	return lf_str_size;
+}
+
 /* Dumps current provisioning status of all RVU block LFs */
 static ssize_t rvu_dbg_rsrc_attach_status(struct file *filp,
 					  char __user *buffer,
 					  size_t count, loff_t *ppos)
 {
-	int index, off = 0, flag = 0, go_back = 0, len = 0;
+	int index, off = 0, flag = 0, len = 0, i = 0;
 	struct rvu *rvu = filp->private_data;
-	int lf, pf, vf, pcifunc;
+	int bytes_not_copied = 0;
 	struct rvu_block block;
-	int bytes_not_copied;
-	int lf_str_size = 12;
+	int pf, vf, pcifunc;
 	int buf_size = 2048;
+	int lf_str_size;
 	char *lfs;
 	char *buf;
 

@@ -162,6 +229,9 @@ static ssize_t rvu_dbg_rsrc_attach_statu
 	if (!buf)
 		return -ENOSPC;
 
+	/* Get the maximum width of a column */
+	lf_str_size = get_max_column_width(rvu);
+
 	lfs = kzalloc(lf_str_size, GFP_KERNEL);
 	if (!lfs) {
 		kfree(buf);

@@ -175,65 +245,69 @@ static ssize_t rvu_dbg_rsrc_attach_statu
 					 "%-*s", lf_str_size,
 					 rvu->hw->block[index].name);
 		}
+
 	off += scnprintf(&buf[off], buf_size - 1 - off, "\n");
+	bytes_not_copied = copy_to_user(buffer + (i * off), buf, off);
+	if (bytes_not_copied)
+		goto out;
+
+	i++;
+	*ppos += off;
 	for (pf = 0; pf < rvu->hw->total_pfs; pf++) {
 		for (vf = 0; vf <= rvu->hw->total_vfs; vf++) {
+			off = 0;
+			flag = 0;
 			pcifunc = pf << 10 | vf;
 			if (!pcifunc)
 				continue;
 
 			if (vf) {
 				sprintf(lfs, "PF%d:VF%d", pf, vf - 1);
-				go_back = scnprintf(&buf[off],
-						    buf_size - 1 - off,
-						    "%-*s", lf_str_size, lfs);
+				off = scnprintf(&buf[off],
+						buf_size - 1 - off,
+						"%-*s", lf_str_size, lfs);
 			} else {
 				sprintf(lfs, "PF%d", pf);
-				go_back = scnprintf(&buf[off],
-						    buf_size - 1 - off,
-						    "%-*s", lf_str_size, lfs);
+				off = scnprintf(&buf[off],
+						buf_size - 1 - off,
+						"%-*s", lf_str_size, lfs);
 			}
 
-			off += go_back;
-			for (index = 0; index < BLKTYPE_MAX; index++) {
+			for (index = 0; index < BLK_COUNT; index++) {
 				block = rvu->hw->block[index];
 				if (!strlen(block.name))
 					continue;
 				len = 0;
 				lfs[len] = '\0';
-				for (lf = 0; lf < block.lf.max; lf++) {
-					if (block.fn_map[lf] != pcifunc)
-						continue;
+				get_lf_str_list(block, pcifunc, lfs);
+				if (strlen(lfs))
 					flag = 1;
-					len += sprintf(&lfs[len], "%d,", lf);
-				}
 
-				if (flag)
-					len--;
-				lfs[len] = '\0';
 				off += scnprintf(&buf[off], buf_size - 1 - off,
 						 "%-*s", lf_str_size, lfs);
-				if (!strlen(lfs))
-					go_back += lf_str_size;
 			}
-			if (!flag)
-				off -= go_back;
-			else
-				flag = 0;
-			off--;
-			off +=	scnprintf(&buf[off], buf_size - 1 - off, "\n");
+			if (flag) {
+				off +=	scnprintf(&buf[off],
+						  buf_size - 1 - off, "\n");
+				bytes_not_copied = copy_to_user(buffer +
+								(i * off),
+								buf, off);
+				if (bytes_not_copied)
+					goto out;
+
+				i++;
+				*ppos += off;
+			}
 		}
 	}
 
-	bytes_not_copied = copy_to_user(buffer, buf, off);
+out:
 	kfree(lfs);
 	kfree(buf);
-
 	if (bytes_not_copied)
 		return -EFAULT;
 
-	*ppos = off;
-	return off;
+	return *ppos;
 }
 
 RVU_DEBUG_FOPS(rsrc_status, rsrc_attach_status, NULL);

This is the start of the stable review cycle for the 5.10.77 release. There are 77 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed, 03 Nov 2021 08:24:20 +0000. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

	https://www.kernel.org/pub/linux/kernel/v5.x/stable-review/patch-5.10.77-rc1.gz

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-5.10.y
and the diffstat can be found below.

thanks,

greg k-h

-------------
Pseudo-Shortlog of commits:

Greg Kroah-Hartman <[email protected]>

    Linux 5.10.77-rc1

Song Liu <[email protected]>

    perf script: Check session->header.env.arch before using it

Alexandre Ghiti <[email protected]>
riscv: Fix asan-stack clang build

Chen Lu <[email protected]>
riscv: fix misalgned trap vector base address

Chanho Park <[email protected]>
scsi: ufs: ufs-exynos: Correct timeout value setting registers

Halil Pasic <[email protected]>
KVM: s390: preserve deliverable_mask in __airqs_kick_single_vcpu

Halil Pasic <[email protected]>
KVM: s390: clear kicked_mask before sleeping again

Alexey Denisov <[email protected]>
lan743x: fix endianness when accessing descriptors

Xin Long <[email protected]>
sctp: add vtag check in sctp_sf_ootb

Xin Long <[email protected]>
sctp: add vtag check in sctp_sf_do_8_5_1_E_sa

Xin Long <[email protected]>
sctp: add vtag check in sctp_sf_violation

Xin Long <[email protected]>
sctp: fix the processing for COOKIE_ECHO chunk

Xin Long <[email protected]>
sctp: fix the processing for INIT_ACK chunk

Xin Long <[email protected]>
sctp: use init_tag from inithdr for ABORT chunk

Andrew Lunn <[email protected]>
phy: phy_ethtool_ksettings_set: Lock the PHY while changing settings

Andrew Lunn <[email protected]>
phy: phy_start_aneg: Add an unlocked version

Andrew Lunn <[email protected]>
phy: phy_ethtool_ksettings_set: Move after phy_start_aneg

Andrew Lunn <[email protected]>
phy: phy_ethtool_ksettings_get: Lock the phy for consistency

Daniel Jordan <[email protected]>
net/tls: Fix flipped sign in async_wait.err assignment

Trevor Woerner <[email protected]>
net: nxp: lpc_eth.c: avoid hang when bringing interface down

Yuiko Oshino <[email protected]>
net: ethernet: microchip: lan743x: Fix dma allocation failure by using dma_set_mask_and_coherent

Yuiko Oshino <[email protected]>
net: ethernet: microchip: lan743x: Fix driver crash when lan743x_pm_resume fails

Ido Schimmel <[email protected]>
mlxsw: pci: Recycle received packet upon allocation failure

Guenter Roeck <[email protected]>
nios2: Make NIOS2_DTB_SOURCE_BOOL depend on !COMPILE_TEST

Jonas Gorski <[email protected]>
gpio: xgs-iproc: fix parsing of ngpios property

Mark Zhang <[email protected]>
RDMA/sa_query: Use strscpy_pad instead of memcpy to copy a string

Michael Chan <[email protected]>
net: Prevent infinite while loop in skb_tx_hash()

Janusz Dziedzic <[email protected]>
cfg80211: correct bridge/4addr mode check

Xin Long <[email protected]>
net-sysfs: initialize uid and gid before calling net_ns_get_ownership

Pavel Skripkin <[email protected]>
net: batman-adv: fix error handling

Yang Yingliang <[email protected]>
regmap: Fix possible double-free in regcache_rbtree_exit()

Jim Quinlan <[email protected]>
reset: brcmstb-rescal: fix incorrect polarity of status bit

Clément Bœsch <[email protected]>
arm64: dts: allwinner: h5: NanoPI Neo 2: Fix ethernet node

Patrisious Haddad <[email protected]>
RDMA/mlx5: Set user priority for DCT

Rakesh Babu <[email protected]>
octeontx2-af: Display all enabled PF VF rsrc_alloc entries.

Varun Prakash <[email protected]>

    nvme-tcp: fix possible req->offset corruption

Varun Prakash <[email protected]>
nvme-tcp: fix data digest pointer calculation

Varun Prakash <[email protected]>
nvmet-tcp: fix data digest pointer calculation

Mike Marciniszyn <[email protected]>
IB/hfi1: Fix abba locking issue with sc_disable()

Mike Marciniszyn <[email protected]>
IB/qib: Protect from buffer overflow in struct qib_user_sdma_pkt fields

Xu Kuohai <[email protected]>
bpf: Fix error usage of map_fd and fdget() in generic_map_update_batch()

Toke Høiland-Jørgensen <[email protected]>
bpf: Fix potential race in tail call compatibility check

Liu Jian <[email protected]>
tcp_bpf: Fix one concurrency problem in the tcp_bpf_send_verdict function

Björn Töpel <[email protected]>
riscv, bpf: Fix potential NULL dereference

Quanyang Wang <[email protected]>
cgroup: Fix memory leak caused by missing cgroup_bpf_offline

Thelford Williams <[email protected]>
drm/amdgpu: fix out of bounds write

Christian König <[email protected]>
drm/ttm: fix memleak in ttm_transfered_destroy

Rongwei Wang <[email protected]>
mm, thp: bail out early in collapse_file for writeback page

Johan Hovold <[email protected]>
net: lan78xx: fix division by zero in send path

Johannes Berg <[email protected]>
cfg80211: fix management registrations locking

Johannes Berg <[email protected]>

    cfg80211: scan: fix RCU in cfg80211_add_nontrans_list()

Sagi Grimberg <[email protected]>
nvme-tcp: fix H2CData PDU send accounting (again)

Gautham Ananthakrishna <[email protected]>
ocfs2: fix race between searching chunks and release journal_head from buffer_head

Haibo Chen <[email protected]>
mmc: sdhci-esdhc-imx: clear the buffer_read_ready to reset standard tuning circuit

Shawn Guo <[email protected]>
mmc: sdhci: Map more voltage level to SDHCI_POWER_330

Jaehoon Chung <[email protected]>
mmc: dw_mmc: exynos: fix the finding clock sample value

Wenbin Mei <[email protected]>
mmc: mediatek: Move cqhci init behind ungate clock

Wenbin Mei <[email protected]>
mmc: cqhci: clear HALT state after CQE enable

Johan Hovold <[email protected]>
mmc: vub300: fix control-message timeouts

Daniel Jordan <[email protected]>
net/tls: Fix flipped sign in tls_err_abort() calls

Pavel Skripkin <[email protected]>
Revert "net: mdiobus: Fix memory leak in __mdiobus_register"

Krzysztof Kozlowski <[email protected]>
nfc: port100: fix using -ERRNO as command type mask

Max VA <[email protected]>
tipc: fix size validations for the MSG_CRYPTO type

Zheyu Ma <[email protected]>
ata: sata_mv: Fix the error handling of mv_chip_id()

Sachi King <[email protected]>
pinctrl: amd: disable and mask interrupts on probe

Rafał Miłecki <[email protected]>
Revert "pinctrl: bcm: ns: support updated DT binding as syscon subnode"

Wang Hai <[email protected]>
usbnet: fix error return code in usbnet_probe()

Oliver Neukum <[email protected]>
usbnet: sanity check for maxpacket

Theodore Ts'o <[email protected]>
ext4: fix possible UAF when remounting r/o a mmp-protected file system

Robin Murphy <[email protected]>
arm64: Avoid premature usercopy failure

Naveen N. Rao <[email protected]>

    powerpc/bpf: Fix BPF_MOD when imm == 1

Pavel Begunkov <[email protected]>
io_uring: don't take uring_lock during iowq cancel

Arnd Bergmann <[email protected]>
ARM: 9141/1: only warn about XIP address when not compile testing

Arnd Bergmann <[email protected]>
ARM: 9139/1: kprobes: fix arch_init_kprobes() prototype

Arnd Bergmann <[email protected]>
ARM: 9138/1: fix link warning with XIP + frame-pointer

Arnd Bergmann <[email protected]>
ARM: 9134/1: remove duplicate memcpy() definition

Nick Desaulniers <[email protected]>
ARM: 9133/1: mm: proc-macros: ensure *_tlb_fns are 4B aligned

Lexi Shao <[email protected]>
ARM: 9132/1: Fix __get_user_check failure with ARM KASAN images

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

Diffstat:

 Makefile                                           |   4 +-
 arch/arm/boot/compressed/decompress.c              |   3 +
 arch/arm/include/asm/uaccess.h                     |   4 +-
 arch/arm/kernel/vmlinux-xip.lds.S                  |   6 +-
 arch/arm/mm/proc-macros.S                          |   1 +
 arch/arm/probes/kprobes/core.c                     |   2 +-
 .../boot/dts/allwinner/sun50i-h5-nanopi-neo2.dts   |   2 +-
 arch/arm64/lib/copy_from_user.S                    |  13 +-
 arch/arm64/lib/copy_in_user.S                      |  21 ++--
 arch/arm64/lib/copy_to_user.S                      |  14 ++-
 arch/nios2/platform/Kconfig.platform               |   1 +
 arch/powerpc/net/bpf_jit_comp64.c                  |  10 +-
 arch/riscv/Kconfig                                 |   6 +
 arch/riscv/include/asm/kasan.h                     |   3 +-
 arch/riscv/kernel/head.S                           |   1 +
 arch/riscv/mm/kasan_init.c                         |   3 +
 arch/riscv/net/bpf_jit_core.c                      |   3 +-
 arch/s390/kvm/interrupt.c                          |   5 +-
 arch/s390/kvm/kvm-s390.c                           |   1 +
 drivers/ata/sata_mv.c                              |   4 +-
 drivers/base/regmap/regcache-rbtree.c              |   7 +-
 drivers/gpio/gpio-xgs-iproc.c                      |   2 +-
 .../drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c  |   2 +-
 drivers/gpu/drm/ttm/ttm_bo_util.c                  |   1 +
 drivers/infiniband/core/sa_query.c                 |   5 +-
 drivers/infiniband/hw/hfi1/pio.c                   |   9 +-
 drivers/infiniband/hw/mlx5/qp.c                    |   2 +
 drivers/infiniband/hw/qib/qib_user_sdma.c          |  33 +++--
 drivers/mmc/host/cqhci.c                           |   3 +
 drivers/mmc/host/dw_mmc-exynos.c                   |  14 +++
 drivers/mmc/host/mtk-sd.c                          |  38 +++---
 drivers/mmc/host/sdhci-esdhc-imx.c                 |  16 +++
 drivers/mmc/host/sdhci.c                           |   6 +
 drivers/mmc/host/vub300.c                          |  18 +--
 .../ethernet/marvell/octeontx2/af/rvu_debugfs.c    | 138 +++++++++++++++-----
 drivers/net/ethernet/mellanox/mlxsw/pci.c          |  25 ++--
 drivers/net/ethernet/microchip/lan743x_main.c      |  88 ++++++++-----
 drivers/net/ethernet/microchip/lan743x_main.h      |  20 +--
 drivers/net/ethernet/nxp/lpc_eth.c                 |   5 +-
 drivers/net/phy/mdio_bus.c                         |   1 -
 drivers/net/phy/phy.c                              | 140 ++++++++++++---------
 drivers/net/usb/lan78xx.c                          |   6 +
 drivers/net/usb/usbnet.c                           |   5 +
 drivers/nfc/port100.c                              |   4 +-
 drivers/nvme/host/tcp.c                            |   9 +-
 drivers/nvme/target/tcp.c                          |   2 +-
 drivers/pinctrl/bcm/pinctrl-ns.c                   |  29 ++---
 drivers/pinctrl/pinctrl-amd.c                      |  31 +++++
 drivers/reset/reset-brcmstb-rescal.c               |   2 +-
 drivers/scsi/ufs/ufs-exynos.c                      |   6 +-
 fs/ext4/mmp.c                                      |  31 +++--
 fs/ext4/super.c                                    |   6 +-
 fs/io_uring.c                                      |   2 +
 fs/ocfs2/suballoc.c                                |  22 ++--
 include/linux/bpf.h                                |   7 +-
 include/net/cfg80211.h                             |   2 -
 include/net/tls.h                                  |   9 +-
 kernel/bpf/arraymap.c                              |   1 +
 kernel/bpf/core.c                                  |  20 +--
 kernel/bpf/syscall.c                               |  11 +-
 kernel/cgroup/cgroup.c                             |   4 +-
 mm/khugepaged.c                                    |   7 +-
 net/batman-adv/bridge_loop_avoidance.c             |   8 +-
 net/batman-adv/main.c                              |  56 ++++++---
 net/batman-adv/network-coding.c                    |   4 +-
 net/batman-adv/translation-table.c                 |   4 +-
 net/core/dev.c                                     |   6 +
 net/core/net-sysfs.c                               |   4 +-
 net/ipv4/tcp_bpf.c                                 |  12 ++
 net/sctp/sm_statefuns.c                            |  67 +++++-----
 net/tipc/crypto.c                                  |  32 +++--
 net/tls/tls_sw.c                                   |  19 ++-
 net/wireless/core.c                                |   2 +-
 net/wireless/core.h                                |   2 +
 net/wireless/mlme.c                                |  26 ++--
 net/wireless/scan.c                                |   7 +-
 net/wireless/util.c                                |  14 +--
 tools/perf/builtin-script.c                        |  12 +-
 78 files changed, 772 insertions(+), 399 deletions(-)

From: Thelford Williams <[email protected]>

commit 5afa7898ab7a0ec9c28556a91df714bf3c2f725e upstream.

Size can be any value and is user controlled resulting in overwriting the 40 byte array wr_buf with an arbitrary length of data from buf.

---
 drivers/gpu/drm/amd/display/amdgpu_dm/amdgpu_dm_debugfs.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -264,7 +264,7 @@ static ssize_t dp_link_settings_write(st
 	if (!wr_buf)
 		return -ENOSPC;
 
-	if (parse_write_buffer_into_params(wr_buf, size,
+	if (parse_write_buffer_into_params(wr_buf, wr_buf_size,
 					   (long *)param, buf,
 					   max_param_num,
 					   &param_nums)) {

From: Clément Bœsch <[email protected]>

commit 0764e365dacd0b8f75c1736f9236be280649bd18 upstream.

RX and TX delay are provided by ethernet PHY. Reflect that in ethernet node.

---
 arch/arm64/boot/dts/allwinner/sun50i-h5-nanopi-neo2.dts |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -75,7 +75,7 @@
 	pinctrl-0 = <&emac_rgmii_pins>;
 	phy-supply = <&reg_gmac_3v3>;
 	phy-handle = <&ext_rgmii_phy>;
-	phy-mode = "rgmii";
+	phy-mode = "rgmii-id";
 	status = "okay";
 };
 

From: Jim Quinlan <[email protected]>

commit f33eb7f29c16ba78db3221ee02346fd832274cdd upstream.

The readl_poll_timeout() should complete when the status bit is a 1, not 0.

---
 drivers/reset/reset-brcmstb-rescal.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -38,7 +38,7 @@ static int brcm_rescal_reset_set(struct
 	}
 
 	ret = readl_poll_timeout(base + BRCM_RESCAL_STATUS, reg,
-				 !(reg & BRCM_RESCAL_STATUS_BIT), 100, 1000);
+				 (reg & BRCM_RESCAL_STATUS_BIT), 100, 1000);
 	if (ret) {
 		dev_err(data->dev, "time out on SATA/PCIe rescal\n");
 		return ret;

From: Pavel Skripkin <[email protected]>

commit 6f68cd634856f8ca93bafd623ba5357e0f648c68 upstream.

Syzbot reported ODEBUG warning in batadv_nc_mesh_free(). The problem was in wrong error handling in batadv_mesh_init().

Before this patch batadv_mesh_init() was calling batadv_mesh_free() in case of any batadv_*_init() calls failure. This approach may work well, when there is some kind of indicator, which can tell which parts of batadv are initialized; but there isn't any.

All written above lead to cleaning up uninitialized fields. Even if we hide ODEBUG warning by initializing bat_priv->nc.work, syzbot was able to hit GPF in batadv_nc_purge_paths(), because hash pointer in still NULL. [1]

To fix these bugs we can unwind batadv_*_init() calls one by one. It is good approach for 2 reasons: 1) It fixes bugs on error handling path 2) It improves the performance, since we won't call unneeded batadv_*_free() functions.

So, this patch makes all batadv_*_init() clean up all allocated memory before returning with an error to no call correspoing batadv_*_free() and open-codes batadv_mesh_free() with proper order to avoid touching uninitialized fields.

---
 net/batman-adv/bridge_loop_avoidance.c |    8 +++-
 net/batman-adv/main.c                  |   56 +++++++++++++++++++++++----------
 net/batman-adv/network-coding.c        |    4 +-
 net/batman-adv/translation-table.c     |    4 +-
 4 files changed, 52 insertions(+), 20 deletions(-)

@@ -1561,10 +1561,14 @@ int batadv_bla_init(struct batadv_priv *
 		return 0;
 
 	bat_priv->bla.claim_hash = batadv_hash_new(128);
-	bat_priv->bla.backbone_hash = batadv_hash_new(32);
+	if (!bat_priv->bla.claim_hash)
+		return -ENOMEM;
 
-	if (!bat_priv->bla.claim_hash || !bat_priv->bla.backbone_hash)
+	bat_priv->bla.backbone_hash = batadv_hash_new(32);
+	if (!bat_priv->bla.backbone_hash) {
+		batadv_hash_destroy(bat_priv->bla.claim_hash);
 		return -ENOMEM;
+	}
 
 	batadv_hash_set_lock_class(bat_priv->bla.claim_hash,
 				   &batadv_claim_hash_lock_class_key);

@@ -155,8 +155,10 @@ int batadv_nc_mesh_init(struct batadv_pr
 				   &batadv_nc_coding_hash_lock_class_key);
 
 	bat_priv->nc.decoding_hash = batadv_hash_new(128);
-	if (!bat_priv->nc.decoding_hash)
+	if (!bat_priv->nc.decoding_hash) {
+		batadv_hash_destroy(bat_priv->nc.coding_hash);
 		goto err;
+	}
 
 	batadv_hash_set_lock_class(bat_priv->nc.decoding_hash,
 				   &batadv_nc_decoding_hash_lock_class_key);

@@ -4405,8 +4405,10 @@ int batadv_tt_init(struct batadv_priv *b
 		return ret;
 
 	ret = batadv_tt_global_init(bat_priv);
-	if (ret < 0)
+	if (ret < 0) {
+		batadv_tt_local_table_free(bat_priv);
 		return ret;
+	}
 
 	batadv_tvlv_handler_register(bat_priv, batadv_tt_tvlv_ogm_handler_v1,
 				     batadv_tt_tvlv_unicast_handler_v1,

From: Xin Long <[email protected]>

commit f7a1e76d0f608961cc2fc681f867a834f2746bce upstream.

Currently in net_ns_get_ownership() it may not be able to set uid or gid if make_kuid or make_kgid returns an invalid value, and an uninit-value issue can be triggered by this.

This patch is to fix it by initializing the uid and gid before calling net_ns_get_ownership(), as it does in kobject_get_ownership()

---
 net/core/net-sysfs.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -1957,9 +1957,9 @@ int netdev_register_kobject(struct net_d
 int netdev_change_owner(struct net_device *ndev, const struct net *net_old,
 			const struct net *net_new)
 {
+	kuid_t old_uid = GLOBAL_ROOT_UID, new_uid = GLOBAL_ROOT_UID;
+	kgid_t old_gid = GLOBAL_ROOT_GID, new_gid = GLOBAL_ROOT_GID;
 	struct device *dev = &ndev->dev;
-	kuid_t old_uid, new_uid;
-	kgid_t old_gid, new_gid;
 	int error;
 
 	net_ns_get_ownership(net_old, &old_uid, &old_gid);

From: Quanyang Wang <[email protected]>

commit 04f8ef5643bcd8bcde25dfdebef998aea480b2ba upstream.

When enabling CONFIG_CGROUP_BPF, kmemleak can be observed by running the command as below:

$mount -t cgroup -o none,name=foo cgroup cgroup/ $umount cgroup/

unreferenced object 0xc3585c40 (size 64):
comm "mount", pid 425, jiffies 4294959825 (age 31.990s) hex dump (first 32 bytes):

    01 00 00 80 84 8c 28 c0 00 00 00 00 00 00 00 00  ......(.........
    00 00 00 00 00 00 00 00 6c 43 a0 c3 00 00 00 00  ........lC......
  backtrace:
    [<e95a2f9e>] cgroup_bpf_inherit+0x44/0x24c
    [<1f03679c>] cgroup_setup_root+0x174/0x37c
    [<ed4b0ac5>] cgroup1_get_tree+0x2c0/0x4a0
    [<f85b12fd>] vfs_get_tree+0x24/0x108
    [<f55aec5c>] path_mount+0x384/0x988
    [<e2d5e9cd>] do_mount+0x64/0x9c
    [<208c9cfe>] sys_mount+0xfc/0x1f4
    [<06dd06e0>] ret_fast_syscall+0x0/0x48
    [<a8308cb3>] 0xbeb4daa8

This is because that since the commit 2b0d3d3e4fcf ("percpu_ref: reduce memory footprint of percpu_ref in fast path") root_cgrp->bpf.refcnt.data is allocated by the function percpu_ref_init in cgroup_bpf_inherit which is called by cgroup_setup_root when mounting, but not freed along with root_cgrp when umounting. Adding cgroup_bpf_offline which calls percpu_ref_kill to cgroup_kill_sb can free root_cgrp->bpf.refcnt.data in umount path.

This patch also fixes the commit 4bfc0bb2c60e ("bpf: decouple the lifetime of cgroup_bpf from cgroup itself"). A cgroup_bpf_offline is needed to do a cleanup that frees the resources which are allocated by cgroup_bpf_inherit in cgroup_setup_root.

And inside cgroup_bpf_offline, cgroup_get() is at the beginning and cgroup_put is at the end of cgroup_bpf_release which is called by cgroup_bpf_offline. So cgroup_bpf_offline can keep the balance of cgroup's refcount.

---
 kernel/cgroup/cgroup.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

@@ -2147,8 +2147,10 @@ static void cgroup_kill_sb(struct super_
 	 * And don't kill the default root.
 	 */
 	if (list_empty(&root->cgrp.self.children) && root != &cgrp_dfl_root &&
-	    !percpu_ref_is_dying(&root->cgrp.self.refcnt))
+	    !percpu_ref_is_dying(&root->cgrp.self.refcnt)) {
+		cgroup_bpf_offline(&root->cgrp);
 		percpu_ref_kill(&root->cgrp.self.refcnt);
+	}
 	cgroup_put(&root->cgrp);
 	kernfs_kill_sb(sb);
 }

From: Janusz Dziedzic <[email protected]>

commit 689a0a9f505f7bffdefe6f17fddb41c8ab6344f6 upstream.

Without the patch we fail:

$ sudo brctl addbr br0
$ sudo brctl addif br0 wlp1s0
$ sudo iw wlp1s0 set 4addr on
command failed: Device or resource busy (-16)

Last command failed but iface was already in 4addr mode.

---
 net/wireless/util.c |   14 +++++++-------
 1 file changed, 7 insertions(+), 7 deletions(-)

@@ -1028,14 +1028,14 @@ int cfg80211_change_iface(struct cfg8021
 	    !(rdev->wiphy.interface_modes & (1 << ntype)))
 		return -EOPNOTSUPP;
 
-	/* if it's part of a bridge, reject changing type to station/ibss */
-	if (netif_is_bridge_port(dev) &&
-	    (ntype == NL80211_IFTYPE_ADHOC ||
-	     ntype == NL80211_IFTYPE_STATION ||
-	     ntype == NL80211_IFTYPE_P2P_CLIENT))
-		return -EBUSY;
-
 	if (ntype != otype) {
+		/* if it's part of a bridge, reject changing type to station/ibss */
+		if (netif_is_bridge_port(dev) &&
+		    (ntype == NL80211_IFTYPE_ADHOC ||
+		     ntype == NL80211_IFTYPE_STATION ||
+		     ntype == NL80211_IFTYPE_P2P_CLIENT))
+			return -EBUSY;
+
 		dev->ieee80211_ptr->use_4addr = false;
 		dev->ieee80211_ptr->mesh_id_up_len = 0;
 		wdev_lock(dev->ieee80211_ptr);

From: Mark Zhang <[email protected]>

commit 64733956ebba7cc629856f4a6ee35a52bc9c023f upstream.

When copying the device name, the length of the data memcpy copied exceeds the length of the source buffer, which cause the KASAN issue below. Use strscpy_pad() instead.

 BUG: KASAN: slab-out-of-bounds in ib_nl_set_path_rec_attrs+0x136/0x320 [ib_core]
 Read of size 64 at addr ffff88811a10f5e0 by task rping/140263
 CPU: 3 PID: 140263 Comm: rping Not tainted 5.15.0-rc1+ #1
 Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014
 Call Trace:
  dump_stack_lvl+0x57/0x7d
  print_address_description.constprop.0+0x1d/0xa0
  kasan_report+0xcb/0x110
  kasan_check_range+0x13d/0x180
  memcpy+0x20/0x60
  ib_nl_set_path_rec_attrs+0x136/0x320 [ib_core]
  ib_nl_make_request+0x1c6/0x380 [ib_core]
  send_mad+0x20a/0x220 [ib_core]
  ib_sa_path_rec_get+0x3e3/0x800 [ib_core]
  cma_query_ib_route+0x29b/0x390 [rdma_cm]
  rdma_resolve_route+0x308/0x3e0 [rdma_cm]
  ucma_resolve_route+0xe1/0x150 [rdma_ucm]
  ucma_write+0x17b/0x1f0 [rdma_ucm]
  vfs_write+0x142/0x4d0
  ksys_write+0x133/0x160
  do_syscall_64+0x43/0x90
  entry_SYSCALL_64_after_hwframe+0x44/0xae
 RIP: 0033:0x7f26499aa90f
 Code: 89 54 24 18 48 89 74 24 10 89 7c 24 08 e8 29 fd ff ff 48 8b 54 24 18 48 8b 74 24 10 41 89 c0 8b 7c 24 08 b8 01 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 31 44 89 c7 48 89 44 24 08 e8 5c fd ff ff 48
 RSP: 002b:00007f26495f2dc0 EFLAGS: 00000293 ORIG_RAX: 0000000000000001
 RAX: ffffffffffffffda RBX: 00000000000007d0 RCX: 00007f26499aa90f
 RDX: 0000000000000010 RSI: 00007f26495f2e00 RDI: 0000000000000003
 RBP: 00005632a8315440 R08: 0000000000000000 R09: 0000000000000001
 R10: 0000000000000000 R11: 0000000000000293 R12: 00007f26495f2e00
 R13: 00005632a83154e0 R14: 00005632a8315440 R15: 00005632a830a810

 Allocated by task 131419:
  kasan_save_stack+0x1b/0x40
  __kasan_kmalloc+0x7c/0x90
  proc_self_get_link+0x8b/0x100
  pick_link+0x4f1/0x5c0
  step_into+0x2eb/0x3d0
  walk_component+0xc8/0x2c0
  link_path_walk+0x3b8/0x580
  path_openat+0x101/0x230
  do_filp_open+0x12e/0x240
  do_sys_openat2+0x115/0x280
  __x64_sys_openat+0xce/0x140
  do_syscall_64+0x43/0x90
  entry_SYSCALL_64_after_hwframe+0x44/0xae

---
 drivers/infiniband/core/sa_query.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

@@ -760,8 +760,9 @@ static void ib_nl_set_path_rec_attrs(str
 
 	/* Construct the family header first */
 	header = skb_put(skb, NLMSG_ALIGN(sizeof(*header)));
-	memcpy(header->device_name, dev_name(&query->port->agent->device->dev),
-	       LS_DEVICE_NAME_MAX);
+	strscpy_pad(header->device_name,
+		    dev_name(&query->port->agent->device->dev),
+		    LS_DEVICE_NAME_MAX);
 	header->port_num = query->port->port_num;
 
 	if ((comp_mask & IB_SA_PATH_REC_REVERSIBLE) &&

From: Michael Chan <[email protected]>

commit 0c57eeecc559ca6bc18b8c4e2808bc78dbe769b0 upstream.

Drivers call netdev_set_num_tc() and then netdev_set_tc_queue() to set the queue count and offset for each TC. So the queue count and offset for the TCs may be zero for a short period after dev->num_tc has been set. If a TX packet is being transmitted at this time in the code path netdev_pick_tx() -> skb_tx_hash(), skb_tx_hash() may see nonzero dev->num_tc but zero qcount for the TC. The while loop that keeps looping while hash >= qcount will not end.

Fix it by checking the TC's qcount to be nonzero before using it.

---
 net/core/dev.c |    6 ++++++
 1 file changed, 6 insertions(+)

@@ -3171,6 +3171,12 @@ static u16 skb_tx_hash(const struct net_
 
 		qoffset = sb_dev->tc_to_txq[tc].offset;
 		qcount = sb_dev->tc_to_txq[tc].count;
+		if (unlikely(!qcount)) {
+			net_warn_ratelimited("%s: invalid qcount, qoffset %u for tc %u\n",
+					     sb_dev->name, qoffset, tc);
+			qoffset = 0;
+			qcount = dev->real_num_tx_queues;
+		}
 	}
 
 	if (skb_rx_queue_recorded(skb)) {

From: Jonas Gorski <[email protected]>

commit 85fe6415c146d5d42ce300c12f1ecf4d4af47d40 upstream.

of_property_read_u32 returns 0 on success, not true, so we need to invert the check to actually take over the provided ngpio value.

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

@@ -224,7 +224,7 @@ static int iproc_gpio_probe(struct platf
 	}
 
 	chip->gc.label = dev_name(dev);
-	if (of_property_read_u32(dn, "ngpios", &num_gpios))
+	if (!of_property_read_u32(dn, "ngpios", &num_gpios))
 		chip->gc.ngpio = num_gpios;
 
 	irq = platform_get_irq(pdev, 0);

From: Patrisious Haddad <[email protected]>

commit 1ab52ac1e9bc9391f592c9fa8340a6e3e9c36286 upstream.

Currently, the driver doesn't set the PCP-based priority for DCT, hence DCT response packets are transmitted without user priority.

Fix it by setting user provided priority in the eth_prio field in the DCT context, which in turn sets the value in the transmitted packet.

---
 drivers/infiniband/hw/mlx5/qp.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -4216,6 +4216,8 @@ static int mlx5_ib_modify_dct(struct ib_
 		MLX5_SET(dctc, dctc, mtu, attr->path_mtu);
 		MLX5_SET(dctc, dctc, my_addr_index, attr->ah_attr.grh.sgid_index);
 		MLX5_SET(dctc, dctc, hop_limit, attr->ah_attr.grh.hop_limit);
+		if (attr->ah_attr.type == RDMA_AH_ATTR_TYPE_ROCE)
+			MLX5_SET(dctc, dctc, eth_prio, attr->ah_attr.sl & 0x7);
 
 		err = mlx5_core_create_dct(dev, &qp->dct.mdct, qp->dct.in,
 					   MLX5_ST_SZ_BYTES(create_dct_in), out,

From: Yang Yingliang <[email protected]>

commit 55e6d8037805b3400096d621091dfbf713f97e83 upstream.

In regcache_rbtree_insert_to_block(), when 'present' realloc failed, the 'blk' which is supposed to assign to 'rbnode->block' will be freed, so 'rbnode->block' points a freed memory, in the error handling path of regcache_rbtree_init(), 'rbnode->block' will be freed again in regcache_rbtree_exit(), KASAN will report double-free as follows:

BUG: KASAN: double-free or invalid-free in kfree+0xce/0x390
Call Trace:
 slab_free_freelist_hook+0x10d/0x240
 kfree+0xce/0x390
 regcache_rbtree_exit+0x15d/0x1a0
 regcache_rbtree_init+0x224/0x2c0
 regcache_init+0x88d/0x1310
 __regmap_init+0x3151/0x4a80
 __devm_regmap_init+0x7d/0x100
 madera_spi_probe+0x10f/0x333 [madera_spi]
 spi_probe+0x183/0x210
 really_probe+0x285/0xc30

To fix this, moving up the assignment of rbnode->block to immediately after the reallocation has succeeded so that the data structure stays valid even if the second reallocation fails.

---
 drivers/base/regmap/regcache-rbtree.c |    7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

@@ -281,14 +281,14 @@ static int regcache_rbtree_insert_to_blo
 	if (!blk)
 		return -ENOMEM;
 
+	rbnode->block = blk;
+
 	if (BITS_TO_LONGS(blklen) > BITS_TO_LONGS(rbnode->blklen)) {
 		present = krealloc(rbnode->cache_present,
 				   BITS_TO_LONGS(blklen) * sizeof(*present),
 				   GFP_KERNEL);
-		if (!present) {
-			kfree(blk);
+		if (!present)
 			return -ENOMEM;
-		}
 
 		memset(present + BITS_TO_LONGS(rbnode->blklen), 0,
 		       (BITS_TO_LONGS(blklen) - BITS_TO_LONGS(rbnode->blklen))

@@ -305,7 +305,6 @@ static int regcache_rbtree_insert_to_blo
 	}
 
 	/* update the rbnode block, its size and the base register */
-	rbnode->block = blk;
 	rbnode->blklen = blklen;
 	rbnode->base_reg = base_reg;
 	rbnode->cache_present = present;

From: Guenter Roeck <[email protected]>

commit 4a089e95b4d6bb625044d47aed0c442a8f7bd093 upstream.

nios2:allmodconfig builds fail with

make[1]: *** No rule to make target 'arch/nios2/boot/dts/""',
needed by 'arch/nios2/boot/dts/built-in.a'. Stop.
make: [Makefile:1868: arch/nios2/boot/dts] Error 2 (ignored)

This is seen with compile tests since those enable NIOS2_DTB_SOURCE_BOOL, which in turn enables NIOS2_DTB_SOURCE. This causes the build error because the default value for NIOS2_DTB_SOURCE is an empty string. Disable NIOS2_DTB_SOURCE_BOOL for compile tests to avoid the error.

---
 arch/nios2/platform/Kconfig.platform |    1 +
 1 file changed, 1 insertion(+)

@@ -37,6 +37,7 @@ config NIOS2_DTB_PHYS_ADDR
 
 config NIOS2_DTB_SOURCE_BOOL
 	bool "Compile and link device tree into kernel image"
+	depends on !COMPILE_TEST
 	help
 	  This allows you to specify a dts (device tree source) file
 	  which will be compiled and linked into the kernel image.

From: Ido Schimmel <[email protected]>

commit 759635760a804b0d8ad0cc677b650f1544cae22f upstream.

When the driver fails to allocate a new Rx buffer, it passes an empty Rx descriptor (contains zero address and size) to the device and marks it as invalid by setting the skb pointer in the descriptor's metadata to NULL.

After processing enough Rx descriptors, the driver will try to process the invalid descriptor, but will return immediately seeing that the skb pointer is NULL. Since the driver no longer passes new Rx descriptors to the device, the Rx queue will eventually become full and the device will start to drop packets.

Fix this by recycling the received packet if allocation of the new packet failed. This means that allocation is no longer performed at the end of the Rx routine, but at the start, before tearing down the DMA mapping of the received packet.

Remove the comment about the descriptor being zeroed as it is no longer correct. This is OK because we either use the descriptor as-is (when recycling) or overwrite its address and size fields with that of the newly allocated Rx buffer.

The issue was discovered when a process ("perf") consumed too much memory and put the system under memory pressure. It can be reproduced by injecting slab allocation failures [1]. After the fix, the Rx queue no longer comes to a halt.

[1]

 # echo 10 > /sys/kernel/debug/failslab/times
 # echo 1000 > /sys/kernel/debug/failslab/interval
 # echo 100 > /sys/kernel/debug/failslab/probability

FAULT_INJECTION: forcing a failure.
name failslab, interval 1000, probability 100, space 0, times 8 [...]

 Call Trace:
  <IRQ>
  dump_stack_lvl+0x34/0x44
  should_fail.cold+0x32/0x37
  should_failslab+0x5/0x10
  kmem_cache_alloc_node+0x23/0x190
  __alloc_skb+0x1f9/0x280
  __netdev_alloc_skb+0x3a/0x150
  mlxsw_pci_rdq_skb_alloc+0x24/0x90
  mlxsw_pci_cq_tasklet+0x3dc/0x1200
  tasklet_action_common.constprop.0+0x9f/0x100
  __do_softirq+0xb5/0x252
  irq_exit_rcu+0x7a/0xa0
  common_interrupt+0x83/0xa0
  </IRQ>
  asm_common_interrupt+0x1e/0x40
 RIP: 0010:cpuidle_enter_state+0xc8/0x340
 [...]
 mlxsw_spectrum2 0000:06:00.0: Failed to alloc skb for RDQ

---
 drivers/net/ethernet/mellanox/mlxsw/pci.c |   25 ++++++++++++-------------
 1 file changed, 12 insertions(+), 13 deletions(-)

@@ -353,13 +353,10 @@ static int mlxsw_pci_rdq_skb_alloc(struc
 	struct sk_buff *skb;
 	int err;
 
-	elem_info->u.rdq.skb = NULL;
 	skb = netdev_alloc_skb_ip_align(NULL, buf_len);
 	if (!skb)
 		return -ENOMEM;
 
-	/* Assume that wqe was previously zeroed. */
-
 	err = mlxsw_pci_wqe_frag_map(mlxsw_pci, wqe, 0, skb->data,
 				     buf_len, DMA_FROM_DEVICE);
 	if (err)

@@ -548,21 +545,26 @@ static void mlxsw_pci_cqe_rdq_handle(str
 	struct pci_dev *pdev = mlxsw_pci->pdev;
 	struct mlxsw_pci_queue_elem_info *elem_info;
 	struct mlxsw_rx_info rx_info = {};
-	char *wqe;
+	char wqe[MLXSW_PCI_WQE_SIZE];
 	struct sk_buff *skb;
 	u16 byte_count;
 	int err;
 
 	elem_info = mlxsw_pci_queue_elem_info_consumer_get(q);
-	skb = elem_info->u.sdq.skb;
-	if (!skb)
-		return;
-	wqe = elem_info->elem;
-	mlxsw_pci_wqe_frag_unmap(mlxsw_pci, wqe, 0, DMA_FROM_DEVICE);
+	skb = elem_info->u.rdq.skb;
+	memcpy(wqe, elem_info->elem, MLXSW_PCI_WQE_SIZE);
 
 	if (q->consumer_counter++ != consumer_counter_limit)
 		dev_dbg_ratelimited(&pdev->dev, "Consumer counter does not match limit in RDQ\n");
 
+	err = mlxsw_pci_rdq_skb_alloc(mlxsw_pci, elem_info);
+	if (err) {
+		dev_err_ratelimited(&pdev->dev, "Failed to alloc skb for RDQ\n");
+		goto out;
+	}
+
+	mlxsw_pci_wqe_frag_unmap(mlxsw_pci, wqe, 0, DMA_FROM_DEVICE);
+
 	if (mlxsw_pci_cqe_lag_get(cqe_v, cqe)) {
 		rx_info.is_lag = true;
 		rx_info.u.lag_id = mlxsw_pci_cqe_lag_id_get(cqe_v, cqe);

@@ -594,10 +596,7 @@ static void mlxsw_pci_cqe_rdq_handle(str
 	skb_put(skb, byte_count);
 	mlxsw_core_skb_receive(mlxsw_pci->core, skb, &rx_info);
 
-	memset(wqe, 0, q->elem_size);
-	err = mlxsw_pci_rdq_skb_alloc(mlxsw_pci, elem_info);
-	if (err)
-		dev_dbg_ratelimited(&pdev->dev, "Failed to alloc skb for RDQ\n");
+out:
 	/* Everything is set up, ring doorbell to pass elem to HW */
 	q->producer_counter++;
 	mlxsw_pci_queue_doorbell_producer_ring(mlxsw_pci, q);

From: Yuiko Oshino <[email protected]>

commit d6423d2ec39cce2bfca418c81ef51792891576bc upstream.

The driver needs to clean up and return when the initialization fails on resume.

---
 drivers/net/ethernet/microchip/lan743x_main.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -3066,6 +3066,8 @@ static int lan743x_pm_resume(struct devi
 	if (ret) {
 		netif_err(adapter, probe, adapter->netdev,
 			  "lan743x_hardware_init returned %d\n", ret);
+		lan743x_pci_cleanup(adapter);
+		return ret;
 	}
 
 	/* open netdev when netdev is at running state while resume.

From: Yuiko Oshino <[email protected]>

commit 95a359c9553342d36d408d35331ff0bfce75272f upstream.

The dma failure was reported in the raspberry pi github (issue #4117).
https://github.com/raspberrypi/linux/issues/4117
The use of dma_set_mask_and_coherent fixes the issue.
Tested on 32/64-bit raspberry pi CM4 and 64-bit ubuntu x86 PC with EVB-LAN7430.

---
 drivers/net/ethernet/microchip/lan743x_main.c |   20 ++++++++++++++++++++
 1 file changed, 20 insertions(+)

@@ -1770,6 +1770,16 @@ static int lan743x_tx_ring_init(struct l
 		ret = -EINVAL;
 		goto cleanup;
 	}
+	if (dma_set_mask_and_coherent(&tx->adapter->pdev->dev,
+				      DMA_BIT_MASK(64))) {
+		if (dma_set_mask_and_coherent(&tx->adapter->pdev->dev,
+					      DMA_BIT_MASK(32))) {
+			dev_warn(&tx->adapter->pdev->dev,
+				 "lan743x_: No suitable DMA available\n");
+			ret = -ENOMEM;
+			goto cleanup;
+		}
+	}
 	ring_allocation_size = ALIGN(tx->ring_size *
 				     sizeof(struct lan743x_tx_descriptor),
 				     PAGE_SIZE);

@@ -2318,6 +2328,16 @@ static int lan743x_rx_ring_init(struct l
 		ret = -EINVAL;
 		goto cleanup;
 	}
+	if (dma_set_mask_and_coherent(&rx->adapter->pdev->dev,
+				      DMA_BIT_MASK(64))) {
+		if (dma_set_mask_and_coherent(&rx->adapter->pdev->dev,
+					      DMA_BIT_MASK(32))) {
+			dev_warn(&rx->adapter->pdev->dev,
+				 "lan743x_: No suitable DMA available\n");
+			ret = -ENOMEM;
+			goto cleanup;
+		}
+	}
 	ring_allocation_size = ALIGN(rx->ring_size *
 				     sizeof(struct lan743x_rx_descriptor),
 				     PAGE_SIZE);

From: Trevor Woerner <[email protected]>

commit ace19b992436a257d9a793672e57abc28fe83e2e upstream.

A hard hang is observed whenever the ethernet interface is brought down. If the PHY is stopped before the LPC core block is reset, the SoC will hang. Comparing lpc_eth_close() and lpc_eth_open() I re-arranged the ordering of the functions calls in lpc_eth_close() to reset the hardware before stopping the PHY.

---
 drivers/net/ethernet/nxp/lpc_eth.c |    5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

@@ -1015,9 +1015,6 @@ static int lpc_eth_close(struct net_devi
 	napi_disable(&pldat->napi);
 	netif_stop_queue(ndev);
 
-	if (ndev->phydev)
-		phy_stop(ndev->phydev);
-
 	spin_lock_irqsave(&pldat->lock, flags);
 	__lpc_eth_reset(pldat);
 	netif_carrier_off(ndev);

@@ -1025,6 +1022,8 @@ static int lpc_eth_close(struct net_devi
 	writel(0, LPC_ENET_MAC2(pldat->net_base));
 	spin_unlock_irqrestore(&pldat->lock, flags);
 
+	if (ndev->phydev)
+		phy_stop(ndev->phydev);
 	clk_disable_unprepare(pldat->clk);
 
 	return 0;

From: Andrew Lunn <[email protected]>

commit 64cd92d5e8180c2ded3fdea76862de6f596ae2c9 upstream.

This allows it to make use of a helper which assume the PHY is already locked.

---
 drivers/net/phy/phy.c |  106 +++++++++++++++++++++++++-------------------------
 1 file changed, 53 insertions(+), 53 deletions(-)

@@ -260,59 +260,6 @@ static void phy_sanitize_settings(struct
 	}
 }
 
-int phy_ethtool_ksettings_set(struct phy_device *phydev,
-			      const struct ethtool_link_ksettings *cmd)
-{
-	__ETHTOOL_DECLARE_LINK_MODE_MASK(advertising);
-	u8 autoneg = cmd->base.autoneg;
-	u8 duplex = cmd->base.duplex;
-	u32 speed = cmd->base.speed;
-
-	if (cmd->base.phy_address != phydev->mdio.addr)
-		return -EINVAL;
-
-	linkmode_copy(advertising, cmd->link_modes.advertising);
-
-	/* We make sure that we don't pass unsupported values in to the PHY */
-	linkmode_and(advertising, advertising, phydev->supported);
-
-	/* Verify the settings we care about. */
-	if (autoneg != AUTONEG_ENABLE && autoneg != AUTONEG_DISABLE)
-		return -EINVAL;
-
-	if (autoneg == AUTONEG_ENABLE && linkmode_empty(advertising))
-		return -EINVAL;
-
-	if (autoneg == AUTONEG_DISABLE &&
-	    ((speed != SPEED_1000 &&
-	      speed != SPEED_100 &&
-	      speed != SPEED_10) ||
-	     (duplex != DUPLEX_HALF &&
-	      duplex != DUPLEX_FULL)))
-		return -EINVAL;
-
-	phydev->autoneg = autoneg;
-
-	if (autoneg == AUTONEG_DISABLE) {
-		phydev->speed = speed;
-		phydev->duplex = duplex;
-	}
-
-	linkmode_copy(phydev->advertising, advertising);
-
-	linkmode_mod_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
-			 phydev->advertising, autoneg == AUTONEG_ENABLE);
-
-	phydev->master_slave_set = cmd->base.master_slave_cfg;
-	phydev->mdix_ctrl = cmd->base.eth_tp_mdix_ctrl;
-
-	/* Restart the PHY */
-	phy_start_aneg(phydev);
-
-	return 0;
-}
-EXPORT_SYMBOL(phy_ethtool_ksettings_set);
-
 void phy_ethtool_ksettings_get(struct phy_device *phydev,
 			       struct ethtool_link_ksettings *cmd)
 {

@@ -818,6 +765,59 @@ static int phy_poll_aneg_done(struct phy
 	return ret < 0 ? ret : 0;
 }
 
+int phy_ethtool_ksettings_set(struct phy_device *phydev,
+			      const struct ethtool_link_ksettings *cmd)
+{
+	__ETHTOOL_DECLARE_LINK_MODE_MASK(advertising);
+	u8 autoneg = cmd->base.autoneg;
+	u8 duplex = cmd->base.duplex;
+	u32 speed = cmd->base.speed;
+
+	if (cmd->base.phy_address != phydev->mdio.addr)
+		return -EINVAL;
+
+	linkmode_copy(advertising, cmd->link_modes.advertising);
+
+	/* We make sure that we don't pass unsupported values in to the PHY */
+	linkmode_and(advertising, advertising, phydev->supported);
+
+	/* Verify the settings we care about. */
+	if (autoneg != AUTONEG_ENABLE && autoneg != AUTONEG_DISABLE)
+		return -EINVAL;
+
+	if (autoneg == AUTONEG_ENABLE && linkmode_empty(advertising))
+		return -EINVAL;
+
+	if (autoneg == AUTONEG_DISABLE &&
+	    ((speed != SPEED_1000 &&
+	      speed != SPEED_100 &&
+	      speed != SPEED_10) ||
+	     (duplex != DUPLEX_HALF &&
+	      duplex != DUPLEX_FULL)))
+		return -EINVAL;
+
+	phydev->autoneg = autoneg;
+
+	if (autoneg == AUTONEG_DISABLE) {
+		phydev->speed = speed;
+		phydev->duplex = duplex;
+	}
+
+	linkmode_copy(phydev->advertising, advertising);
+
+	linkmode_mod_bit(ETHTOOL_LINK_MODE_Autoneg_BIT,
+			 phydev->advertising, autoneg == AUTONEG_ENABLE);
+
+	phydev->master_slave_set = cmd->base.master_slave_cfg;
+	phydev->mdix_ctrl = cmd->base.eth_tp_mdix_ctrl;
+
+	/* Restart the PHY */
+	phy_start_aneg(phydev);
+
+	return 0;
+}
+EXPORT_SYMBOL(phy_ethtool_ksettings_set);
+
 /**
  * phy_speed_down - set speed to lowest speed supported by both link partners
  * @phydev: the phy_device struct

From: Björn Töpel <[email protected]>

commit 27de809a3d83a6199664479ebb19712533d6fd9b upstream.

The bpf_jit_binary_free() function requires a non-NULL argument. When the RISC-V BPF JIT fails to converge in NR_JIT_ITERATIONS steps, jit_data->header will be NULL, which triggers a NULL dereference. Avoid this by checking the argument, prior calling the function.

---
 arch/riscv/net/bpf_jit_core.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -125,7 +125,8 @@ struct bpf_prog *bpf_int_jit_compile(str
 
 	if (i == NR_JIT_ITERATIONS) {
 		pr_err("bpf-jit: image did not converge in <%d passes!\n", i);
-		bpf_jit_binary_free(jit_data->header);
+		if (jit_data->header)
+			bpf_jit_binary_free(jit_data->header);
 		prog = orig_prog;
 		goto out_offset;
 	}

From: Andrew Lunn <[email protected]>

commit c10a485c3de5ccbf1fff65a382cebcb2730c6b06 upstream.

The PHY structure should be locked while copying information out if it, otherwise there is no guarantee of self consistency. Without the lock the PHY state machine could be updating the structure.

---
 drivers/net/phy/phy.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -316,6 +316,7 @@ EXPORT_SYMBOL(phy_ethtool_ksettings_set)
 void phy_ethtool_ksettings_get(struct phy_device *phydev,
 			       struct ethtool_link_ksettings *cmd)
 {
+	mutex_lock(&phydev->lock);
 	linkmode_copy(cmd->link_modes.supported, phydev->supported);
 	linkmode_copy(cmd->link_modes.advertising, phydev->advertising);
 	linkmode_copy(cmd->link_modes.lp_advertising, phydev->lp_advertising);

@@ -334,6 +335,7 @@ void phy_ethtool_ksettings_get(struct ph
 	cmd->base.autoneg = phydev->autoneg;
 	cmd->base.eth_tp_mdix_ctrl = phydev->mdix_ctrl;
 	cmd->base.eth_tp_mdix = phydev->mdix;
+	mutex_unlock(&phydev->lock);
 }
 EXPORT_SYMBOL(phy_ethtool_ksettings_get);
 

From: Andrew Lunn <[email protected]>

commit 707293a56f95f8e7e0cfae008010c7933fb68973 upstream.

Split phy_start_aneg into a wrapper which takes the PHY lock, and a helper doing the real work. This will be needed when phy_ethtook_ksettings_set takes the lock.

---
 drivers/net/phy/phy.c |   30 ++++++++++++++++++++++++------
 1 file changed, 24 insertions(+), 6 deletions(-)

@@ -716,7 +716,7 @@ static int phy_check_link_status(struct
 }
 
 /**
- * phy_start_aneg - start auto-negotiation for this PHY device
+ * _phy_start_aneg - start auto-negotiation for this PHY device
  * @phydev: the phy_device struct
  *
  * Description: Sanitizes the settings (if we're not autonegotiating

@@ -724,25 +724,43 @@ static int phy_check_link_status(struct
  *   If the PHYCONTROL Layer is operating, we change the state to
  *   reflect the beginning of Auto-negotiation or forcing.
  */
-int phy_start_aneg(struct phy_device *phydev)
+static int _phy_start_aneg(struct phy_device *phydev)
 {
 	int err;
 
+	lockdep_assert_held(&phydev->lock);
+
 	if (!phydev->drv)
 		return -EIO;
 
-	mutex_lock(&phydev->lock);
-
 	if (AUTONEG_DISABLE == phydev->autoneg)
 		phy_sanitize_settings(phydev);
 
 	err = phy_config_aneg(phydev);
 	if (err < 0)
-		goto out_unlock;
+		return err;
 
 	if (phy_is_started(phydev))
 		err = phy_check_link_status(phydev);
-out_unlock:
+
+	return err;
+}
+
+/**
+ * phy_start_aneg - start auto-negotiation for this PHY device
+ * @phydev: the phy_device struct
+ *
+ * Description: Sanitizes the settings (if we're not autonegotiating
+ *   them), and then calls the driver's config_aneg function.
+ *   If the PHYCONTROL Layer is operating, we change the state to
+ *   reflect the beginning of Auto-negotiation or forcing.
+ */
+int phy_start_aneg(struct phy_device *phydev)
+{
+	int err;
+
+	mutex_lock(&phydev->lock);
+	err = _phy_start_aneg(phydev);
 	mutex_unlock(&phydev->lock);
 
 	return err;