5.9.13-rc1 review

This is the start of the stable review cycle for the 5.9.13 release. There are 46 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 Tue, 08 Dec 2020 11:15:42 +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-5.9.13-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-5.9.y
and the diffstat can be found below.

thanks,

greg k-h

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

Greg Kroah-Hartman <[email protected]>

    Linux 5.9.13-rc1

Chris Wilson <[email protected]>
drm/i915/gt: Fixup tgl mocs for PTE tracking

Eric Sandeen <[email protected]>
uapi: fix statx attribute value overlap for DAX & MOUNT_ROOT

Vasily Averin <[email protected]>
tracing: Remove WARN_ON in start_thread()

Minchan Kim <[email protected]>
tracing: Fix alignment of static buffer

Linus Walleij <[email protected]>
Input: atmel_mxt_ts - fix lost interrupts

Po-Hsu Lin <[email protected]>
Input: i8042 - add ByteSpeed touchpad to noloop table

Sanjay Govind <[email protected]>
Input: xpad - support Ardwiino Controllers

Hector Martin <[email protected]>
ALSA: usb-audio: US16x08: fix value count for level meters

Randy Dunlap <[email protected]>
net: mlx5e: fix fs_tcp.c build when IPV6 is not enabled

Eran Ben Elisha <[email protected]>
net/mlx5: Fix wrong address reclaim when command interface is down

Yevgeny Kliteynik <[email protected]>
net/mlx5: DR, Proper handling of unsupported Connect-X6DX SW steering

Davide Caratti <[email protected]>
net/sched: act_mpls: ensure LSE is pullable before reading it

Davide Caratti <[email protected]>
net: openvswitch: ensure LSE is pullable before reading it

Davide Caratti <[email protected]>
net: skbuff: ensure LSE is pullable before decrementing the MPLS ttl

Wang Hai <[email protected]>
net: mvpp2: Fix error return code in mvpp2_open()

Dan Carpenter <[email protected]>
chelsio/chtls: fix a double free in chtls_setkey()

Zhang Changzhong <[email protected]>
vxlan: fix error return code in __vxlan_dev_create()

Zhang Changzhong <[email protected]>
net: pasemi: fix error return code in pasemi_mac_open()

Zhang Changzhong <[email protected]>
cxgb3: fix error return code in t3_sge_alloc_qset()

Dan Carpenter <[email protected]>
net/x25: prevent a couple of overflows

Yangbo Lu <[email protected]>
dpaa_eth: copy timestamp fields to new skb in A-050385 workaround

Antoine Tenart <[email protected]>

    net: ip6_gre: set dev->hard_header_len when using header_ops

Eric Dumazet <[email protected]>
geneve: pull IP header before ECN decapsulation

Toke Høiland-Jørgensen <[email protected]>
inet_ecn: Fix endianness of checksum update when setting ECT(1)

Hoang Le <[email protected]>
tipc: fix incompatible mtu of transmission

Thomas Falcon <[email protected]>
ibmvnic: Fix TX completion error handling

Thomas Falcon <[email protected]>
ibmvnic: Ensure that SCRQ entry reads are correctly ordered

Vinay Kumar Yadav <[email protected]>
chelsio/chtls: fix panic during unload reload chtls

Krzysztof Kozlowski <[email protected]>
dt-bindings: net: correct interrupt flags in examples

Guillaume Nault <[email protected]>
ipv4: Fix tos mask in inet_rtm_getroute()

Antoine Tenart <[email protected]>

    netfilter: bridge: reset skb->pkt_type after NF_INET_POST_ROUTING traversal

Eyal Birger <[email protected]>
net/packet: fix packet receive on L3 devices without visible hard header

Paolo Abeni <[email protected]>
mptcp: fix NULL ptr dereference on bad MPJ

Stefano Garzarella <[email protected]>
vsock/virtio: discard packets only when socket is really closed

Yves-Alexis Perez <[email protected]>
usbnet: ipheth: fix connectivity with iOS 14

Jens Axboe <[email protected]>
tun: honor IOCB_NOWAIT flag

Alexander Duyck <[email protected]>
tcp: Set INET_ECN_xmit configuration in tcp_reinit_congestion_control

Willem de Bruijn <[email protected]>
sock: set sk_err to ee_errno on dequeue from errq

Anmol Karn <[email protected]>
rose: Fix Null pointer dereference in rose_send_frame()

Maxim Mikityanskiy <[email protected]>
net/tls: Protect from calling tls_dev_del for TLS RX twice

Vadim Fedorenko <[email protected]>
net/tls: missing received data after fast remote close

Eelco Chaudron <[email protected]>
net: openvswitch: fix TTL decrement action netlink message format

Julian Wiedmann <[email protected]>
net/af_iucv: set correct sk_protocol for child sockets

Wang Hai <[email protected]>
ipv6: addrlabel: fix possible memory leak in ip6addrlbl_net_init

Parav Pandit <[email protected]>
devlink: Make sure devlink instance and port are in same net namespace

Parav Pandit <[email protected]>
devlink: Hold rtnl lock while reading netdev attributes

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

Diffstat:

 .../devicetree/bindings/net/can/tcan4x5x.txt       |  2 +-
 .../devicetree/bindings/net/nfc/nxp-nci.txt        |  2 +-
 .../devicetree/bindings/net/nfc/pn544.txt          |  2 +-
 Makefile                                           |  4 +-
 drivers/crypto/chelsio/chtls/chtls_cm.c            |  1 +
 drivers/crypto/chelsio/chtls/chtls_hw.c            |  1 +
 drivers/gpu/drm/i915/gt/intel_mocs.c               |  5 +-
 drivers/input/joystick/xpad.c                      |  2 +
 drivers/input/serio/i8042-x86ia64io.h              |  4 ++
 drivers/input/touchscreen/atmel_mxt_ts.c           |  4 +-
 drivers/net/ethernet/chelsio/cxgb3/sge.c           |  1 +
 drivers/net/ethernet/freescale/dpaa/dpaa_eth.c     | 10 ++-
 drivers/net/ethernet/ibm/ibmvnic.c                 | 22 ++++++-
 drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c    |  1 +
 .../ethernet/mellanox/mlx5/core/en_accel/fs_tcp.c  |  2 +
 .../net/ethernet/mellanox/mlx5/core/pagealloc.c    | 21 +++++-
 .../ethernet/mellanox/mlx5/core/steering/dr_cmd.c  |  1 +
 .../mellanox/mlx5/core/steering/dr_domain.c        |  5 ++
 .../mellanox/mlx5/core/steering/dr_types.h         |  1 +
 drivers/net/ethernet/pasemi/pasemi_mac.c           |  8 ++-
 drivers/net/geneve.c                               | 20 ++++--
 drivers/net/tun.c                                  | 14 +++-
 drivers/net/usb/ipheth.c                           |  2 +-
 drivers/net/vxlan.c                                |  4 +-
 include/linux/mlx5/mlx5_ifc.h                      |  9 ++-
 include/linux/netdevice.h                          |  5 ++
 include/net/inet_ecn.h                             |  2 +-
 include/net/tls.h                                  |  6 ++
 include/uapi/linux/openvswitch.h                   |  2 +
 include/uapi/linux/stat.h                          |  9 ++-
 kernel/trace/trace.c                               |  2 +-
 kernel/trace/trace_hwlat.c                         |  2 +-
 net/bridge/br_netfilter_hooks.c                    |  7 +-
 net/core/devlink.c                                 |  7 +-
 net/core/skbuff.c                                  |  5 +-
 net/ipv4/route.c                                   |  7 +-
 net/ipv4/tcp_cong.c                                |  5 ++
 net/ipv6/addrlabel.c                               | 26 +++++---
 net/ipv6/ip6_gre.c                                 | 16 ++++-
 net/iucv/af_iucv.c                                 |  4 +-
 net/mptcp/subflow.c                                |  5 +-
 net/openvswitch/actions.c                          | 10 +--
 net/openvswitch/flow_netlink.c                     | 74 ++++++++++++++++------
 net/packet/af_packet.c                             | 38 ++++++-----
 net/rose/rose_loopback.c                           | 17 +++--
 net/sched/act_mpls.c                               |  3 +
 net/tipc/node.c                                    |  2 +
 net/tls/tls_device.c                               |  5 +-
 net/tls/tls_sw.c                                   |  6 ++
 net/vmw_vsock/virtio_transport_common.c            |  8 ++-
 net/x25/af_x25.c                                   |  6 +-
 sound/usb/mixer_us16x08.c                          |  2 +-
 52 files changed, 321 insertions(+), 108 deletions(-)

From: Parav Pandit <[email protected]>

[ Upstream commit b187c9b4178b87954dbc94e78a7094715794714f ]

A netdevice of a devlink port can be moved to different net namespace than its parent devlink instance.
This scenario occurs when devlink reload is not used.

When netdevice is undergoing migration to net namespace, its ifindex and name may change.

In such use case, devlink port query may read stale netdev attributes.

Fix it by reading them under rtnl lock.

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

@@ -616,6 +616,8 @@ static int devlink_nl_port_fill(struct s
 	if (nla_put_u32(msg, DEVLINK_ATTR_PORT_INDEX, devlink_port->index))
 		goto nla_put_failure;
 
+	/* Hold rtnl lock while accessing port's netdev attributes. */
+	rtnl_lock();
 	spin_lock_bh(&devlink_port->type_lock);
 	if (nla_put_u16(msg, DEVLINK_ATTR_PORT_TYPE, devlink_port->type))
 		goto nla_put_failure_type_locked;

@@ -642,6 +644,7 @@ static int devlink_nl_port_fill(struct s
 			goto nla_put_failure_type_locked;
 	}
 	spin_unlock_bh(&devlink_port->type_lock);
+	rtnl_unlock();
 	if (devlink_nl_port_attrs_put(msg, devlink_port))
 		goto nla_put_failure;
 	if (devlink_nl_port_function_attrs_put(msg, devlink_port, extack))

@@ -652,6 +655,7 @@ static int devlink_nl_port_fill(struct s
 
 nla_put_failure_type_locked:
 	spin_unlock_bh(&devlink_port->type_lock);
+	rtnl_unlock();
 nla_put_failure:
 	genlmsg_cancel(msg, hdr);
 	return -EMSGSIZE;

From: Parav Pandit <[email protected]>

[ Upstream commit a7b43649507dae4e55ff0087cad4e4dd1c6d5b99 ]

When devlink reload operation is not used, netdev of an Ethernet port may be present in different net namespace than the net namespace of the devlink instance.

Ensure that both the devlink instance and devlink port netdev are located in same net namespace.

---
 net/core/devlink.c |    3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

@@ -626,9 +626,10 @@ static int devlink_nl_port_fill(struct s
 			devlink_port->desired_type))
 		goto nla_put_failure_type_locked;
 	if (devlink_port->type == DEVLINK_PORT_TYPE_ETH) {
+		struct net *net = devlink_net(devlink_port->devlink);
 		struct net_device *netdev = devlink_port->type_dev;
 
-		if (netdev &&
+		if (netdev && net_eq(net, dev_net(netdev)) &&
 		    (nla_put_u32(msg, DEVLINK_ATTR_PORT_NETDEV_IFINDEX,
 				 netdev->ifindex) ||
 		     nla_put_string(msg, DEVLINK_ATTR_PORT_NETDEV_NAME,

From: Wang Hai <[email protected]>

[ Upstream commit e255e11e66da8281e337e4e352956e8a4999fca4 ]

kmemleak report a memory leak as follows:

unreferenced object 0xffff8880059c6a00 (size 64):
comm "ip", pid 23696, jiffies 4296590183 (age 1755.384s) hex dump (first 32 bytes):

    20 01 00 10 00 00 00 00 00 00 00 00 00 00 00 00   ...............
    1c 00 00 00 00 00 00 00 00 00 00 00 07 00 00 00  ................
  backtrace:
    [<00000000aa4e7a87>] ip6addrlbl_add+0x90/0xbb0
    [<0000000070b8d7f1>] ip6addrlbl_net_init+0x109/0x170
    [<000000006a9ca9d4>] ops_init+0xa8/0x3c0
    [<000000002da57bf2>] setup_net+0x2de/0x7e0
    [<000000004e52d573>] copy_net_ns+0x27d/0x530
    [<00000000b07ae2b4>] create_new_namespaces+0x382/0xa30
    [<000000003b76d36f>] unshare_nsproxy_namespaces+0xa1/0x1d0
    [<0000000030653721>] ksys_unshare+0x3a4/0x780
    [<0000000007e82e40>] __x64_sys_unshare+0x2d/0x40
    [<0000000031a10c08>] do_syscall_64+0x33/0x40
    [<0000000099df30e7>] entry_SYSCALL_64_after_hwframe+0x44/0xa9

We should free all rules when we catch an error in ip6addrlbl_net_init(). otherwise a memory leak will occur.

---
 net/ipv6/addrlabel.c |   26 +++++++++++++++++---------
 1 file changed, 17 insertions(+), 9 deletions(-)

@@ -306,7 +306,9 @@ static int ip6addrlbl_del(struct net *ne
 /* add default label */
 static int __net_init ip6addrlbl_net_init(struct net *net)
 {
-	int err = 0;
+	struct ip6addrlbl_entry *p = NULL;
+	struct hlist_node *n;
+	int err;
 	int i;
 
 	ADDRLABEL(KERN_DEBUG "%s\n", __func__);

@@ -315,14 +317,20 @@ static int __net_init ip6addrlbl_net_ini
 	INIT_HLIST_HEAD(&net->ipv6.ip6addrlbl_table.head);
 
 	for (i = 0; i < ARRAY_SIZE(ip6addrlbl_init_table); i++) {
-		int ret = ip6addrlbl_add(net,
-					 ip6addrlbl_init_table[i].prefix,
-					 ip6addrlbl_init_table[i].prefixlen,
-					 0,
-					 ip6addrlbl_init_table[i].label, 0);
-		/* XXX: should we free all rules when we catch an error? */
-		if (ret && (!err || err != -ENOMEM))
-			err = ret;
+		err = ip6addrlbl_add(net,
+				     ip6addrlbl_init_table[i].prefix,
+				     ip6addrlbl_init_table[i].prefixlen,
+				     0,
+				     ip6addrlbl_init_table[i].label, 0);
+		if (err)
+			goto err_ip6addrlbl_add;
+	}
+	return 0;
+
+err_ip6addrlbl_add:
+	hlist_for_each_entry_safe(p, n, &net->ipv6.ip6addrlbl_table.head, list) {
+		hlist_del_rcu(&p->list);
+		kfree_rcu(p, rcu);
 	}
 	return err;
 }

From: Julian Wiedmann <[email protected]>

[ Upstream commit c5dab0941fcdc9664eb0ec0d4d51433216d91336 ]

Child sockets erroneously inherit their parent's sk_type (ie. SOCK_*), instead of the PF_IUCV protocol that the parent was created with in iucv_sock_create().

We're currently not using sk->sk_protocol ourselves, so this shouldn't have much impact (except eg. getting the output in skb_dump() right).

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

@@ -1645,7 +1645,7 @@ static int iucv_callback_connreq(struct
 	}
 
 	/* Create the new socket */
-	nsk = iucv_sock_alloc(NULL, sk->sk_type, GFP_ATOMIC, 0);
+	nsk = iucv_sock_alloc(NULL, sk->sk_protocol, GFP_ATOMIC, 0);
 	if (!nsk) {
 		err = pr_iucv->path_sever(path, user_data);
 		iucv_path_free(path);

@@ -1851,7 +1851,7 @@ static int afiucv_hs_callback_syn(struct
 		goto out;
 	}
 
-	nsk = iucv_sock_alloc(NULL, sk->sk_type, GFP_ATOMIC, 0);
+	nsk = iucv_sock_alloc(NULL, sk->sk_protocol, GFP_ATOMIC, 0);
 	bh_lock_sock(sk);
 	if ((sk->sk_state != IUCV_LISTEN) ||
 	    sk_acceptq_is_full(sk) ||

From: Eelco Chaudron <[email protected]>

[ Upstream commit 69929d4c49e182f8526d42c43b37b460d562d3a0 ]

Currently, the openvswitch module is not accepting the correctly formated netlink message for the TTL decrement action. For both setting and getting the dec_ttl action, the actions should be nested in the OVS_DEC_TTL_ATTR_ACTION attribute as mentioned in the openvswitch.h uapi.

When the original patch was sent, it was tested with a private OVS userspace implementation. This implementation was unfortunately not upstreamed and reviewed, hence an erroneous version of this patch was sent out.

Leaving the patch as-is would cause problems as the kernel module could interpret additional attributes as actions and vice-versa, due to the actions not being encapsulated/nested within the actual attribute, but being concatinated after it.

---
 include/uapi/linux/openvswitch.h |    2 +
 net/openvswitch/actions.c        |    7 +--
 net/openvswitch/flow_netlink.c   |   74 ++++++++++++++++++++++++++++-----------
 3 files changed, 60 insertions(+), 23 deletions(-)

@@ -1058,4 +1058,6 @@ enum ovs_dec_ttl_attr {
 	__OVS_DEC_TTL_ATTR_MAX
 };
 
+#define OVS_DEC_TTL_ATTR_MAX (__OVS_DEC_TTL_ATTR_MAX - 1)
+
 #endif /* _LINUX_OPENVSWITCH_H */

@@ -970,14 +970,13 @@ static int dec_ttl_exception_handler(str
 {
 	/* The first action is always 'OVS_DEC_TTL_ATTR_ARG'. */
 	struct nlattr *dec_ttl_arg = nla_data(attr);
-	int rem = nla_len(attr);
 
 	if (nla_len(dec_ttl_arg)) {
-		struct nlattr *actions = nla_next(dec_ttl_arg, &rem);
+		struct nlattr *actions = nla_data(dec_ttl_arg);
 
 		if (actions)
-			return clone_execute(dp, skb, key, 0, actions, rem,
-					     last, false);
+			return clone_execute(dp, skb, key, 0, nla_data(actions),
+					     nla_len(actions), last, false);
 	}
 	consume_skb(skb);
 	return 0;

From: Vadim Fedorenko <[email protected]>

[ Upstream commit 20ffc7adf53a5fd3d19751fbff7895bcca66686e ]

In case when tcp socket received FIN after some data and the parser haven't started before reading data caller will receive an empty buffer. This behavior differs from plain TCP socket and leads to special treating in user-space.
The flow that triggers the race is simple. Server sends small amount of data right after the connection is configured to use TLS and closes the connection. In this case receiver sees TLS Handshake data, configures TLS socket right after Change Cipher Spec record. While the configuration is in process, TCP socket receives small Application Data record, Encrypted Alert record and FIN packet. So the TCP socket changes sk_shutdown to RCV_SHUTDOWN and sk_flag with SK_DONE bit set. The received data is not parsed upon arrival and is never sent to user-space.

Patch unpauses parser directly if we have unparsed data in tcp receive queue.

---
 net/tls/tls_sw.c |    6 ++++++
 1 file changed, 6 insertions(+)

@@ -1295,6 +1295,12 @@ static struct sk_buff *tls_wait_data(str
 			return NULL;
 		}
 
+		if (!skb_queue_empty(&sk->sk_receive_queue)) {
+			__strp_unpause(&ctx->strp);
+			if (ctx->recv_pkt)
+				return ctx->recv_pkt;
+		}
+
 		if (sk->sk_shutdown & RCV_SHUTDOWN)
 			return NULL;
 

From: Maxim Mikityanskiy <[email protected]>

[ Upstream commit 025cc2fb6a4e84e9a0552c0017dcd1c24b7ac7da ]

tls_device_offload_cleanup_rx doesn't clear tls_ctx->netdev after calling tls_dev_del if TLX TX offload is also enabled. Clearing tls_ctx->netdev gets postponed until tls_device_gc_task. It leaves a time frame when tls_device_down may get called and call tls_dev_del for RX one extra time, confusing the driver, which may lead to a crash.

This patch corrects this racy behavior by adding a flag to prevent tls_device_down from calling tls_dev_del the second time.

---
 include/net/tls.h    |    6 ++++++
 net/tls/tls_device.c |    5 ++++-
 2 files changed, 10 insertions(+), 1 deletion(-)

@@ -199,6 +199,12 @@ enum tls_context_flags {
 	 * to be atomic.
 	 */
 	TLS_TX_SYNC_SCHED = 1,
+	/* tls_dev_del was called for the RX side, device state was released,
+	 * but tls_ctx->netdev might still be kept, because TX-side driver
+	 * resources might not be released yet. Used to prevent the second
+	 * tls_dev_del call in tls_device_down if it happens simultaneously.
+	 */
+	TLS_RX_DEV_CLOSED = 2,
 };
 
 struct cipher_context {

@@ -1262,6 +1262,8 @@ void tls_device_offload_cleanup_rx(struc
 	if (tls_ctx->tx_conf != TLS_HW) {
 		dev_put(netdev);
 		tls_ctx->netdev = NULL;
+	} else {
+		set_bit(TLS_RX_DEV_CLOSED, &tls_ctx->flags);
 	}
 out:
 	up_read(&device_offload_lock);

@@ -1291,7 +1293,8 @@ static int tls_device_down(struct net_de
 		if (ctx->tx_conf == TLS_HW)
 			netdev->tlsdev_ops->tls_dev_del(netdev, ctx,
 							TLS_OFFLOAD_CTX_DIR_TX);
-		if (ctx->rx_conf == TLS_HW)
+		if (ctx->rx_conf == TLS_HW &&
+		    !test_bit(TLS_RX_DEV_CLOSED, &ctx->flags))
 			netdev->tlsdev_ops->tls_dev_del(netdev, ctx,
 							TLS_OFFLOAD_CTX_DIR_RX);
 		WRITE_ONCE(ctx->netdev, NULL);

From: Anmol Karn <[email protected]>

[ Upstream commit 3b3fd068c56e3fbea30090859216a368398e39bf ]

rose_send_frame() dereferences neigh->dev when called from rose_transmit_clear_request(), and the first occurrence of the neigh is in rose_loopback_timer() as rose_loopback_neigh, and it is initialized in rose_add_loopback_neigh() as NULL. i.e when rose_loopback_neigh used in rose_loopback_timer() its ->dev was still NULL and rose_loopback_timer() was calling rose_rx_call_request() without checking for NULL.

- net/rose/rose_link.c
This bug seems to get triggered in this line:

rose_call = (ax25_address *)neigh->dev->dev_addr;

Fix it by adding NULL checking for rose_loopback_neigh->dev in rose_loopback_timer().

---
 net/rose/rose_loopback.c |   17 +++++++++++++----
 1 file changed, 13 insertions(+), 4 deletions(-)

@@ -96,10 +96,19 @@ static void rose_loopback_timer(struct t
 		}
 
 		if (frametype == ROSE_CALL_REQUEST) {
-			if ((dev = rose_dev_get(dest)) != NULL) {
-				if (rose_rx_call_request(skb, dev, rose_loopback_neigh, lci_o) == 0)
-					kfree_skb(skb);
-			} else {
+			if (!rose_loopback_neigh->dev) {
+				kfree_skb(skb);
+				continue;
+			}
+
+			dev = rose_dev_get(dest);
+			if (!dev) {
+				kfree_skb(skb);
+				continue;
+			}
+
+			if (rose_rx_call_request(skb, dev, rose_loopback_neigh, lci_o) == 0) {
+				dev_put(dev);
 				kfree_skb(skb);
 			}
 		} else {

From: Willem de Bruijn <[email protected]>

[ Upstream commit 985f7337421a811cb354ca93882f943c8335a6f5 ]

When setting sk_err, set it to ee_errno, not ee_origin.

Commit f5f99309fa74 ("sock: do not set sk_err in sock_dequeue_err_skb") disabled updating sk_err on errq dequeue, which is correct for most error types (origins):

- sk->sk_err = err;

Commit 38b257938ac6 ("sock: reset sk_err when the error queue is empty") reenabled the behavior for IMCP origins, which do require it:

+ if (icmp_next)
+ sk->sk_err = SKB_EXT_ERR(skb_next)->ee.ee_origin;

But read from ee_errno.

---
 net/core/skbuff.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -4555,7 +4555,7 @@ struct sk_buff *sock_dequeue_err_skb(str
 	if (skb && (skb_next = skb_peek(q))) {
 		icmp_next = is_icmp_err_skb(skb_next);
 		if (icmp_next)
-			sk->sk_err = SKB_EXT_ERR(skb_next)->ee.ee_origin;
+			sk->sk_err = SKB_EXT_ERR(skb_next)->ee.ee_errno;
 	}
 	spin_unlock_irqrestore(&q->lock, flags);
 

From: Alexander Duyck <[email protected]>

[ Upstream commit 55472017a4219ca965a957584affdb17549ae4a4 ]

When setting congestion control via a BPF program it is seen that the SYN/ACK for packets within a given flow will not include the ECT0 flag. A bit of simple printk debugging shows that when this is configured without BPF we will see the value INET_ECN_xmit value initialized in tcp_assign_congestion_control however when we configure this via BPF the socket is in the closed state and as such it isn't configured, and I do not see it being initialized when we transition the socket into the listen state. The result of this is that the ECT0 bit is configured based on whatever the default state is for the socket.

Any easy way to reproduce this is to monitor the following with tcpdump:
tools/testing/selftests/bpf/test_progs -t bpf_tcp_ca

Without this patch the SYN/ACK will follow whatever the default is. If dctcp all SYN/ACK packets will have the ECT0 bit set, and if it is not then ECT0 will be cleared on all SYN/ACK packets. With this patch applied the SYN/ACK bit matches the value seen on the other packets in the given stream.

---
 net/ipv4/tcp_cong.c |    5 +++++
 1 file changed, 5 insertions(+)

@@ -197,6 +197,11 @@ static void tcp_reinit_congestion_contro
 	icsk->icsk_ca_setsockopt = 1;
 	memset(icsk->icsk_ca_priv, 0, sizeof(icsk->icsk_ca_priv));
 
+	if (ca->flags & TCP_CONG_NEEDS_ECN)
+		INET_ECN_xmit(sk);
+	else
+		INET_ECN_dontxmit(sk);
+
 	if (!((1 << sk->sk_state) & (TCPF_CLOSE | TCPF_LISTEN)))
 		tcp_init_congestion_control(sk);
 }

From: Jens Axboe <[email protected]>

[ Upstream commit 5aac0390a63b8718237a61dd0d24a29201d1c94a ]

tun only checks the file O_NONBLOCK flag, but it should also be checking the iocb IOCB_NOWAIT flag. Any fops using ->read/write_iter() should check both, otherwise it breaks users that correctly expect O_NONBLOCK semantics if IOCB_NOWAIT is set.

---
 drivers/net/tun.c |   14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

@@ -1979,12 +1979,15 @@ static ssize_t tun_chr_write_iter(struct
 	struct tun_file *tfile = file->private_data;
 	struct tun_struct *tun = tun_get(tfile);
 	ssize_t result;
+	int noblock = 0;
 
 	if (!tun)
 		return -EBADFD;
 
-	result = tun_get_user(tun, tfile, NULL, from,
-			      file->f_flags & O_NONBLOCK, false);
+	if ((file->f_flags & O_NONBLOCK) || (iocb->ki_flags & IOCB_NOWAIT))
+		noblock = 1;
+
+	result = tun_get_user(tun, tfile, NULL, from, noblock, false);
 
 	tun_put(tun);
 	return result;

@@ -2203,10 +2206,15 @@ static ssize_t tun_chr_read_iter(struct
 	struct tun_file *tfile = file->private_data;
 	struct tun_struct *tun = tun_get(tfile);
 	ssize_t len = iov_iter_count(to), ret;
+	int noblock = 0;
 
 	if (!tun)
 		return -EBADFD;
-	ret = tun_do_read(tun, tfile, to, file->f_flags & O_NONBLOCK, NULL);
+
+	if ((file->f_flags & O_NONBLOCK) || (iocb->ki_flags & IOCB_NOWAIT))
+		noblock = 1;
+
+	ret = tun_do_read(tun, tfile, to, noblock, NULL);
 	ret = min_t(ssize_t, ret, len);
 	if (ret > 0)
 		iocb->ki_pos = ret;

From: Yves-Alexis Perez <[email protected]>

[ Upstream commit f33d9e2b48a34e1558b67a473a1fc1d6e793f93c ]

Starting with iOS 14 released in September 2020, connectivity using the personal hotspot USB tethering function of iOS devices is broken.

Communication between the host and the device (for example ICMP traffic or DNS resolution using the DNS service running in the device itself) works fine, but communication to endpoints further away doesn't work.

Investigation on the matter shows that no UDP and ICMP traffic from the tethered host is reaching the Internet at all. For TCP traffic there are exchanges between tethered host and server but packets are modified in transit leading to impossible communication.

After some trials Matti Vuorela discovered that reducing the URB buffer size by two bytes restored the previous behavior. While a better solution might exist to fix the issue, since the protocol is not publicly documented and considering the small size of the fix, let's do that.

---
 drivers/net/usb/ipheth.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -59,7 +59,7 @@
 #define IPHETH_USBINTF_SUBCLASS 253
 #define IPHETH_USBINTF_PROTO    1
 
-#define IPHETH_BUF_SIZE         1516
+#define IPHETH_BUF_SIZE         1514
 #define IPHETH_IP_ALIGN		2	/* padding at front of URB */
 #define IPHETH_TX_TIMEOUT       (5 * HZ)
 

From: Stefano Garzarella <[email protected]>

[ Upstream commit 3fe356d58efae54dade9ec94ea7c919ed20cf4db ]

Starting from commit 8692cefc433f ("virtio_vsock: Fix race condition in virtio_transport_recv_pkt"), we discard packets in virtio_transport_recv_pkt() if the socket has been released.

When the socket is connected, we schedule a delayed work to wait the RST packet from the other peer, also if SHUTDOWN_MASK is set in sk->sk_shutdown.
This is done to complete the virtio-vsock shutdown algorithm, releasing the port assigned to the socket definitively only when the other peer has consumed all the packets.

If we discard the RST packet received, the socket will be closed only when the VSOCK_CLOSE_TIMEOUT is reached.

Sergio discovered the issue while running ab(1) HTTP benchmark using libkrun [1] and observing a latency increase with that commit.

To avoid this issue, we discard packet only if the socket is really closed (SOCK_DONE flag is set).
We also set SOCK_DONE in virtio_transport_release() when we don't need to wait any packets from the other peer (we didn't schedule the delayed work). In this case we remove the socket from the vsock lists, releasing the port assigned.

[1] https://github.com/containers/libkrun

---
 net/vmw_vsock/virtio_transport_common.c |    8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

@@ -841,8 +841,10 @@ void virtio_transport_release(struct vso
 		virtio_transport_free_pkt(pkt);
 	}
 
-	if (remove_sock)
+	if (remove_sock) {
+		sock_set_flag(sk, SOCK_DONE);
 		vsock_remove_sock(vsk);
+	}
 }
 EXPORT_SYMBOL_GPL(virtio_transport_release);
 

@@ -1132,8 +1134,8 @@ void virtio_transport_recv_pkt(struct vi
 
 	lock_sock(sk);
 
-	/* Check if sk has been released before lock_sock */
-	if (sk->sk_shutdown == SHUTDOWN_MASK) {
+	/* Check if sk has been closed before lock_sock */
+	if (sock_flag(sk, SOCK_DONE)) {
 		(void)virtio_transport_reset_no_sock(t, pkt);
 		release_sock(sk);
 		sock_put(sk);

From: Paolo Abeni <[email protected]>

[ Upstream commit d3ab78858f1451351221061a1c365495df196500 ]

If an msk listener receives an MPJ carrying an invalid token, it will zero the request socket msk entry. That should later cause fallback and subflow reset - as per RFC - at subflow_syn_recv_sock() time due to failing hmac validation.

Since commit 4cf8b7e48a09 ("subflow: introduce and use mptcp_can_accept_new_subflow()"), we unconditionally dereference
- in mptcp_can_accept_new_subflow - the subflow request msk before performing hmac validation. In the above scenario we hit a NULL ptr dereference.

Address the issue doing the hmac validation earlier.

---
 net/mptcp/subflow.c |    5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

@@ -542,9 +542,8 @@ create_msk:
 			fallback = true;
 	} else if (subflow_req->mp_join) {
 		mptcp_get_options(skb, &mp_opt);
-		if (!mp_opt.mp_join ||
-		    !mptcp_can_accept_new_subflow(subflow_req->msk) ||
-		    !subflow_hmac_valid(req, &mp_opt)) {
+		if (!mp_opt.mp_join || !subflow_hmac_valid(req, &mp_opt) ||
+		    !mptcp_can_accept_new_subflow(subflow_req->msk)) {
 			SUBFLOW_REQ_INC_STATS(req, MPTCP_MIB_JOINACKMAC);
 			fallback = true;
 		}

From: Eyal Birger <[email protected]>

[ Upstream commit d549699048b4b5c22dd710455bcdb76966e55aa3 ]

In the patchset merged by commit b9fcf0a0d826
("Merge branch 'support-AF_PACKET-for-layer-3-devices'") L3 devices which did not have header_ops were given one for the purpose of protocol parsing on af_packet transmit path.

That change made af_packet receive path regard these devices as having a visible L3 header and therefore aligned incoming skb->data to point to the skb's mac_header. Some devices, such as ipip, xfrmi, and others, do not reset their mac_header prior to ingress and therefore their incoming packets became malformed.

Ideally these devices would reset their mac headers, or af_packet would be able to rely on dev->hard_header_len being 0 for such cases, but it seems this is not the case.

Fix by changing af_packet RX ll visibility criteria to include the existence of a '.create()' header operation, which is used when creating a device hard header - via dev_hard_header() - by upper layers, and does not exist in these L3 devices.

As this predicate may be useful in other situations, add it as a common dev_has_header() helper in netdevice.h.

---
 include/linux/netdevice.h |    5 +++++
 net/packet/af_packet.c    |   38 +++++++++++++++++++++-----------------
 2 files changed, 26 insertions(+), 17 deletions(-)

@@ -3103,6 +3103,11 @@ static inline bool dev_validate_header(c
 	return false;
 }
 
+static inline bool dev_has_header(const struct net_device *dev)
+{
+	return dev->header_ops && dev->header_ops->create;
+}
+
 typedef int gifconf_func_t(struct net_device * dev, char __user * bufptr,
 			   int len, int size);
 int register_gifconf(unsigned int family, gifconf_func_t *gifconf);

@@ -93,38 +93,42 @@
 
 /*
    Assumptions:
-   - if device has no dev->hard_header routine, it adds and removes ll header
-     inside itself. In this case ll header is invisible outside of device,
-     but higher levels still should reserve dev->hard_header_len.
-     Some devices are enough clever to reallocate skb, when header
-     will not fit to reserved space (tunnel), another ones are silly
-     (PPP).
+   - If the device has no dev->header_ops->create, there is no LL header
+     visible above the device. In this case, its hard_header_len should be 0.
+     The device may prepend its own header internally. In this case, its
+     needed_headroom should be set to the space needed for it to add its
+     internal header.
+     For example, a WiFi driver pretending to be an Ethernet driver should
+     set its hard_header_len to be the Ethernet header length, and set its
+     needed_headroom to be (the real WiFi header length - the fake Ethernet
+     header length).
    - packet socket receives packets with pulled ll header,
      so that SOCK_RAW should push it back.
 
 On receive:
 -----------
 
-Incoming, dev->hard_header!=NULL
+Incoming, dev_has_header(dev) == true
    mac_header -> ll header
    data       -> data
 
-Outgoing, dev->hard_header!=NULL
+Outgoing, dev_has_header(dev) == true
    mac_header -> ll header
    data       -> ll header
 
-Incoming, dev->hard_header==NULL
-   mac_header -> UNKNOWN position. It is very likely, that it points to ll
-		 header.  PPP makes it, that is wrong, because introduce
-		 assymetry between rx and tx paths.
+Incoming, dev_has_header(dev) == false
+   mac_header -> data
+     However drivers often make it point to the ll header.
+     This is incorrect because the ll header should be invisible to us.
    data       -> data
 
-Outgoing, dev->hard_header==NULL
-   mac_header -> data. ll header is still not built!
+Outgoing, dev_has_header(dev) == false
+   mac_header -> data. ll header is invisible to us.
    data       -> data
 
 Resume
-  If dev->hard_header==NULL we are unlikely to restore sensible ll header.
+  If dev_has_header(dev) == false we are unable to restore the ll header,
+    because it is invisible to us.
 
 
 On transmit:

@@ -2066,7 +2070,7 @@ static int packet_rcv(struct sk_buff *sk
 
 	skb->dev = dev;
 
-	if (dev->header_ops) {
+	if (dev_has_header(dev)) {
 		/* The device has an explicit notion of ll header,
 		 * exported to higher levels.
 		 *

@@ -2195,7 +2199,7 @@ static int tpacket_rcv(struct sk_buff *s
 	if (!net_eq(dev_net(dev), sock_net(sk)))
 		goto drop;
 
-	if (dev->header_ops) {
+	if (dev_has_header(dev)) {
 		if (sk->sk_type != SOCK_DGRAM)
 			skb_push(skb, skb->data - skb_mac_header(skb));
 		else if (skb->pkt_type == PACKET_OUTGOING) {

From: Antoine Tenart <[email protected]>

[ Upstream commit 44f64f23bae2f0fad25503bc7ab86cd08d04cd47 ]

Netfilter changes PACKET_OTHERHOST to PACKET_HOST before invoking the hooks as, while it's an expected value for a bridge, routing expects PACKET_HOST. The change is undone later on after hook traversal. This can be seen with pairs of functions updating skb>pkt_type and then reverting it to its original value:

For hook NF_INET_PRE_ROUTING:
setup_pre_routing / br_nf_pre_routing_finish

For hook NF_INET_FORWARD:
br_nf_forward_ip / br_nf_forward_finish

But the third case where netfilter does this, for hook NF_INET_POST_ROUTING, the packet type is changed in br_nf_post_routing but never reverted. A comment says:

/* We assume any code from br_dev_queue_push_xmit onwards doesn't care
* about the value of skb->pkt_type. */

But when having a tunnel (say vxlan) attached to a bridge we have the following call trace:

br_nf_pre_routing
br_nf_pre_routing_ipv6
br_nf_pre_routing_finish
br_nf_forward_ip
br_nf_forward_finish
br_nf_post_routing <- pkt_type is updated to PACKET_HOST
br_nf_dev_queue_xmit <- but not reverted to its original value
vxlan_xmit
vxlan_xmit_one
skb_tunnel_check_pmtu <- a check on pkt_type is performed

In this specific case, this creates issues such as when an ICMPv6 PTB should be sent back. When CONFIG_BRIDGE_NETFILTER is enabled, the PTB isn't sent (as skb_tunnel_check_pmtu checks if pkt_type is PACKET_HOST and returns early).

If the comment is right and no one cares about the value of skb->pkt_type after br_dev_queue_push_xmit (which isn't true), resetting it to its original value should be safe.

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

@@ -735,6 +735,11 @@ static int br_nf_dev_queue_xmit(struct n
 	mtu_reserved = nf_bridge_mtu_reduction(skb);
 	mtu = skb->dev->mtu;
 
+	if (nf_bridge->pkt_otherhost) {
+		skb->pkt_type = PACKET_OTHERHOST;
+		nf_bridge->pkt_otherhost = false;
+	}
+
 	if (nf_bridge->frag_max_size && nf_bridge->frag_max_size < mtu)
 		mtu = nf_bridge->frag_max_size;
 

@@ -835,8 +840,6 @@ static unsigned int br_nf_post_routing(v
 	else
 		return NF_ACCEPT;
 
-	/* We assume any code from br_dev_queue_push_xmit onwards doesn't care
-	 * about the value of skb->pkt_type. */
 	if (skb->pkt_type == PACKET_OTHERHOST) {
 		skb->pkt_type = PACKET_HOST;
 		nf_bridge->pkt_otherhost = true;

From: Guillaume Nault <[email protected]>

[ Upstream commit 1ebf179037cb46c19da3a9c1e2ca16e7a754b75e ]

When inet_rtm_getroute() was converted to use the RCU variants of ip_route_input() and ip_route_output_key(), the TOS parameters stopped being masked with IPTOS_RT_MASK before doing the route lookup.

As a result, "ip route get" can return a different route than what would be used when sending real packets.

For example:

    $ ip route add 192.0.2.11/32 dev eth0
    $ ip route add unreachable 192.0.2.11/32 tos 2
    $ ip route get 192.0.2.11 tos 2
    RTNETLINK answers: No route to host

But, packets with TOS 2 (ECT(0) if interpreted as an ECN bit) would actually be routed using the first route:

    $ ping -c 1 -Q 2 192.0.2.11
    PING 192.0.2.11 (192.0.2.11) 56(84) bytes of data.
    64 bytes from 192.0.2.11: icmp_seq=1 ttl=64 time=0.173 ms

    --- 192.0.2.11 ping statistics ---
    1 packets transmitted, 1 received, 0% packet loss, time 0ms
    rtt min/avg/max/mdev = 0.173/0.173/0.173/0.000 ms

This patch re-applies IPTOS_RT_MASK in inet_rtm_getroute(), to return results consistent with real route lookups.

---
 net/ipv4/route.c |    7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

@@ -3221,7 +3221,7 @@ static int inet_rtm_getroute(struct sk_b
 
 	fl4.daddr = dst;
 	fl4.saddr = src;
-	fl4.flowi4_tos = rtm->rtm_tos;
+	fl4.flowi4_tos = rtm->rtm_tos & IPTOS_RT_MASK;
 	fl4.flowi4_oif = tb[RTA_OIF] ? nla_get_u32(tb[RTA_OIF]) : 0;
 	fl4.flowi4_mark = mark;
 	fl4.flowi4_uid = uid;

@@ -3245,8 +3245,9 @@ static int inet_rtm_getroute(struct sk_b
 		fl4.flowi4_iif = iif; /* for rt_fill_info */
 		skb->dev	= dev;
 		skb->mark	= mark;
-		err = ip_route_input_rcu(skb, dst, src, rtm->rtm_tos,
-					 dev, &res);
+		err = ip_route_input_rcu(skb, dst, src,
+					 rtm->rtm_tos & IPTOS_RT_MASK, dev,
+					 &res);
 
 		rt = skb_rtable(skb);
 		if (err == 0 && rt->dst.error)

From: Krzysztof Kozlowski <[email protected]>

[ Upstream commit 4d521943f76bd0d1e68ea5e02df7aadd30b2838a ]

GPIO_ACTIVE_x flags are not correct in the context of interrupt flags. These are simple defines so they could be used in DTS but they will not have the same meaning:
1. GPIO_ACTIVE_HIGH = 0 = IRQ_TYPE_NONE
2. GPIO_ACTIVE_LOW = 1 = IRQ_TYPE_EDGE_RISING

Correct the interrupt flags, assuming the author of the code wanted same logical behavior behind the name "ACTIVE_xxx", this is:
ACTIVE_LOW => IRQ_TYPE_LEVEL_LOW
ACTIVE_HIGH => IRQ_TYPE_LEVEL_HIGH

---
 Documentation/devicetree/bindings/net/can/tcan4x5x.txt |    2 +-
 Documentation/devicetree/bindings/net/nfc/nxp-nci.txt  |    2 +-
 Documentation/devicetree/bindings/net/nfc/pn544.txt    |    2 +-
 3 files changed, 3 insertions(+), 3 deletions(-)

@@ -33,7 +33,7 @@ tcan4x5x: tcan4x5x@0 {
 		spi-max-frequency = <10000000>;
 		bosch,mram-cfg = <0x0 0 0 32 0 0 1 1>;
 		interrupt-parent = <&gpio1>;
-		interrupts = <14 GPIO_ACTIVE_LOW>;
+		interrupts = <14 IRQ_TYPE_LEVEL_LOW>;
 		device-state-gpios = <&gpio3 21 GPIO_ACTIVE_HIGH>;
 		device-wake-gpios = <&gpio1 15 GPIO_ACTIVE_HIGH>;
 		reset-gpios = <&gpio1 27 GPIO_ACTIVE_HIGH>;

@@ -25,7 +25,7 @@ Example (for ARM-based BeagleBone with N
 		clock-frequency = <100000>;
 
 		interrupt-parent = <&gpio1>;
-		interrupts = <29 GPIO_ACTIVE_HIGH>;
+		interrupts = <29 IRQ_TYPE_LEVEL_HIGH>;
 
 		enable-gpios = <&gpio0 30 GPIO_ACTIVE_HIGH>;
 		firmware-gpios = <&gpio0 31 GPIO_ACTIVE_HIGH>;

@@ -25,7 +25,7 @@ Example (for ARM-based BeagleBone with P
 		clock-frequency = <400000>;
 
 		interrupt-parent = <&gpio1>;
-		interrupts = <17 GPIO_ACTIVE_HIGH>;
+		interrupts = <17 IRQ_TYPE_LEVEL_HIGH>;
 
 		enable-gpios = <&gpio3 21 GPIO_ACTIVE_HIGH>;
 		firmware-gpios = <&gpio3 19 GPIO_ACTIVE_HIGH>;

From: Vinay Kumar Yadav <[email protected]>

[ Upstream commit e3d5e971d2f83d8ddd4b91a50cea4517fb488383 ]

there is kernel panic in inet_twsk_free() while chtls module unload when socket is in TIME_WAIT state because sk_prot_creator was not preserved on connection socket.

---
 drivers/crypto/chelsio/chtls/chtls_cm.c |    1 +
 1 file changed, 1 insertion(+)

@@ -1206,6 +1206,7 @@ static struct sock *chtls_recv_sock(stru
 	sk_setup_caps(newsk, dst);
 	ctx = tls_get_ctx(lsk);
 	newsk->sk_destruct = ctx->sk_destruct;
+	newsk->sk_prot_creator = lsk->sk_prot_creator;
 	csk->sk = newsk;
 	csk->passive_reap_next = oreq;
 	csk->tx_chan = cxgb4_port_chan(ndev);

From: Thomas Falcon <[email protected]>

[ Upstream commit b71ec952234610b4f90ef17a2fdcb124d5320070 ]

Ensure that received Subordinate Command-Response Queue (SCRQ) entries are properly read in order by the driver. These queues are used in the ibmvnic device to process RX buffer and TX completion descriptors. dma_rmb barriers have been added after checking for a pending descriptor to ensure the correct descriptor entry is checked and after reading the SCRQ descriptor to ensure the entire descriptor is read before processing.

---
 drivers/net/ethernet/ibm/ibmvnic.c |   18 ++++++++++++++++++
 1 file changed, 18 insertions(+)

@@ -2409,6 +2409,12 @@ restart_poll:
 
 		if (!pending_scrq(adapter, adapter->rx_scrq[scrq_num]))
 			break;
+		/* The queue entry at the current index is peeked at above
+		 * to determine that there is a valid descriptor awaiting
+		 * processing. We want to be sure that the current slot
+		 * holds a valid descriptor before reading its contents.
+		 */
+		dma_rmb();
 		next = ibmvnic_next_scrq(adapter, adapter->rx_scrq[scrq_num]);
 		rx_buff =
 		    (struct ibmvnic_rx_buff *)be64_to_cpu(next->

@@ -3107,6 +3113,13 @@ restart_loop:
 		unsigned int pool = scrq->pool_index;
 		int num_entries = 0;
 
+		/* The queue entry at the current index is peeked at above
+		 * to determine that there is a valid descriptor awaiting
+		 * processing. We want to be sure that the current slot
+		 * holds a valid descriptor before reading its contents.
+		 */
+		dma_rmb();
+
 		next = ibmvnic_next_scrq(adapter, scrq);
 		for (i = 0; i < next->tx_comp.num_comps; i++) {
 			if (next->tx_comp.rcs[i]) {

@@ -3507,6 +3520,11 @@ static union sub_crq *ibmvnic_next_scrq(
 	}
 	spin_unlock_irqrestore(&scrq->lock, flags);
 
+	/* Ensure that the entire buffer descriptor has been
+	 * loaded before reading its contents
+	 */
+	dma_rmb();
+
 	return entry;
 }
 

From: Thomas Falcon <[email protected]>

[ Upstream commit ba246c175116e2e8fa4fdfa5f8e958e086a9a818 ]

TX completions received with an error return code are not being processed properly. When an error code is seen, do not proceed to the next completion before cleaning up the existing entry's data structures.

---
 drivers/net/ethernet/ibm/ibmvnic.c |    4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

@@ -3122,11 +3122,9 @@ restart_loop:
 
 		next = ibmvnic_next_scrq(adapter, scrq);
 		for (i = 0; i < next->tx_comp.num_comps; i++) {
-			if (next->tx_comp.rcs[i]) {
+			if (next->tx_comp.rcs[i])
 				dev_err(dev, "tx error %x\n",
 					next->tx_comp.rcs[i]);
-				continue;
-			}
 			index = be32_to_cpu(next->tx_comp.correlators[i]);
 			if (index & IBMVNIC_TSO_POOL_MASK) {
 				tx_pool = &adapter->tso_pool[pool];

From: Hoang Le <[email protected]>

[ Upstream commit 0643334902fcdc770e2d9555811200213339a3f6 ]

In commit 682cd3cf946b6
("tipc: confgiure and apply UDP bearer MTU on running links"), we introduced a function to change UDP bearer MTU and applied this new value across existing per-link. However, we did not apply this new MTU value at node level. This lead to packet dropped at link level if its size is greater than new MTU value.

To fix this issue, we also apply this new MTU value for node level.

---
 net/tipc/node.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -2171,6 +2171,8 @@ void tipc_node_apply_property(struct net
 			else if (prop == TIPC_NLA_PROP_MTU)
 				tipc_link_set_mtu(e->link, b->mtu);
 		}
+		/* Update MTU for node link entry */
+		e->mtu = tipc_link_mss(e->link);
 		tipc_node_write_unlock(n);
 		tipc_bearer_xmit(net, bearer_id, &xmitq, &e->maddr, NULL);
 	}

From: "Toke H�iland-J�rgensen" <[email protected]>

[ Upstream commit 2867e1eac61016f59b3d730e3f7aa488e186e917 ]

When adding support for propagating ECT(1) marking in IP headers it seems I suffered from endianness-confusion in the checksum update calculation: In fact the ECN field is in the *lower* bits of the first 16-bit word of the IP header when calculating in network byte order. This means that the addition performed to update the checksum field was wrong; let's fix that.

---
 include/net/inet_ecn.h |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -107,7 +107,7 @@ static inline int IP_ECN_set_ect1(struct
 	if ((iph->tos & INET_ECN_MASK) != INET_ECN_ECT_0)
 		return 0;
 
-	check += (__force u16)htons(0x100);
+	check += (__force u16)htons(0x1);
 
 	iph->check = (__force __sum16)(check + (check>=0xFFFF));
 	iph->tos ^= INET_ECN_MASK;

From: Eric Dumazet <[email protected]>

[ Upstream commit 4179b00c04d18ea7013f68d578d80f3c9d13150a ]

IP_ECN_decapsulate() and IP6_ECN_decapsulate() assume IP header is already pulled.

geneve does not ensure this yet.

Fixing this generically in IP_ECN_decapsulate() and IP6_ECN_decapsulate() is not possible, since callers pass a pointer that might be freed by pskb_may_pull()

syzbot reported :

BUG: KMSAN: uninit-value in __INET_ECN_decapsulate include/net/inet_ecn.h:238 [inline]
BUG: KMSAN: uninit-value in INET_ECN_decapsulate+0x345/0x1db0 include/net/inet_ecn.h:260
CPU: 1 PID: 8941 Comm: syz-executor.0 Not tainted 5.10.0-rc4-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
 <IRQ>
 __dump_stack lib/dump_stack.c:77 [inline]
 dump_stack+0x21c/0x280 lib/dump_stack.c:118
 kmsan_report+0xf7/0x1e0 mm/kmsan/kmsan_report.c:118
 __msan_warning+0x5f/0xa0 mm/kmsan/kmsan_instr.c:197
 __INET_ECN_decapsulate include/net/inet_ecn.h:238 [inline]
 INET_ECN_decapsulate+0x345/0x1db0 include/net/inet_ecn.h:260
 geneve_rx+0x2103/0x2980 include/net/inet_ecn.h:306
 geneve_udp_encap_recv+0x105c/0x1340 drivers/net/geneve.c:377
 udp_queue_rcv_one_skb+0x193a/0x1af0 net/ipv4/udp.c:2093
 udp_queue_rcv_skb+0x282/0x1050 net/ipv4/udp.c:2167
 udp_unicast_rcv_skb net/ipv4/udp.c:2325 [inline]
 __udp4_lib_rcv+0x399d/0x5880 net/ipv4/udp.c:2394
 udp_rcv+0x5c/0x70 net/ipv4/udp.c:2564
 ip_protocol_deliver_rcu+0x572/0xc50 net/ipv4/ip_input.c:204
 ip_local_deliver_finish net/ipv4/ip_input.c:231 [inline]
 NF_HOOK include/linux/netfilter.h:301 [inline]
 ip_local_deliver+0x583/0x8d0 net/ipv4/ip_input.c:252
 dst_input include/net/dst.h:449 [inline]
 ip_rcv_finish net/ipv4/ip_input.c:428 [inline]
 NF_HOOK include/linux/netfilter.h:301 [inline]
 ip_rcv+0x5c3/0x840 net/ipv4/ip_input.c:539
 __netif_receive_skb_one_core net/core/dev.c:5315 [inline]
 __netif_receive_skb+0x1ec/0x640 net/core/dev.c:5429
 process_backlog+0x523/0xc10 net/core/dev.c:6319
 napi_poll+0x420/0x1010 net/core/dev.c:6763
 net_rx_action+0x35c/0xd40 net/core/dev.c:6833
 __do_softirq+0x1a9/0x6fa kernel/softirq.c:298
 asm_call_irq_on_stack+0xf/0x20
 </IRQ>
 __run_on_irqstack arch/x86/include/asm/irq_stack.h:26 [inline]
 run_on_irqstack_cond arch/x86/include/asm/irq_stack.h:77 [inline]
 do_softirq_own_stack+0x6e/0x90 arch/x86/kernel/irq_64.c:77
 do_softirq kernel/softirq.c:343 [inline]
 __local_bh_enable_ip+0x184/0x1d0 kernel/softirq.c:195
 local_bh_enable+0x36/0x40 include/linux/bottom_half.h:32
 rcu_read_unlock_bh include/linux/rcupdate.h:730 [inline]
 __dev_queue_xmit+0x3a9b/0x4520 net/core/dev.c:4167
 dev_queue_xmit+0x4b/0x60 net/core/dev.c:4173
 packet_snd net/packet/af_packet.c:2992 [inline]
 packet_sendmsg+0x86f9/0x99d0 net/packet/af_packet.c:3017
 sock_sendmsg_nosec net/socket.c:651 [inline]
 sock_sendmsg net/socket.c:671 [inline]
 __sys_sendto+0x9dc/0xc80 net/socket.c:1992
 __do_sys_sendto net/socket.c:2004 [inline]
 __se_sys_sendto+0x107/0x130 net/socket.c:2000
 __x64_sys_sendto+0x6e/0x90 net/socket.c:2000
 do_syscall_64+0x9f/0x140 arch/x86/entry/common.c:48
 entry_SYSCALL_64_after_hwframe+0x44/0xa9

---
 drivers/net/geneve.c |   20 ++++++++++++++++----
 1 file changed, 16 insertions(+), 4 deletions(-)

@@ -258,11 +258,21 @@ static void geneve_rx(struct geneve_dev
 		skb_dst_set(skb, &tun_dst->dst);
 
 	/* Ignore packet loops (and multicast echo) */
-	if (ether_addr_equal(eth_hdr(skb)->h_source, geneve->dev->dev_addr)) {
-		geneve->dev->stats.rx_errors++;
-		goto drop;
-	}
+	if (ether_addr_equal(eth_hdr(skb)->h_source, geneve->dev->dev_addr))
+		goto rx_error;
 
+	switch (skb_protocol(skb, true)) {
+	case htons(ETH_P_IP):
+		if (pskb_may_pull(skb, sizeof(struct iphdr)))
+			goto rx_error;
+		break;
+	case htons(ETH_P_IPV6):
+		if (pskb_may_pull(skb, sizeof(struct ipv6hdr)))
+			goto rx_error;
+		break;
+	default:
+		goto rx_error;
+	}
 	oiph = skb_network_header(skb);
 	skb_reset_network_header(skb);
 

@@ -303,6 +313,8 @@ static void geneve_rx(struct geneve_dev
 		u64_stats_update_end(&stats->syncp);
 	}
 	return;
+rx_error:
+	geneve->dev->stats.rx_errors++;
 drop:
 	/* Consume bad packet */
 	kfree_skb(skb);

From: Antoine Tenart <[email protected]>

[ Upstream commit 832ba596494b2c9eac7760259eff2d8b7dcad0ee ]

syzkaller managed to crash the kernel using an NBMA ip6gre interface. I could reproduce it creating an NBMA ip6gre interface and forwarding traffic to it:

skbuff: skb_under_panic: text:ffffffff8250e927 len:148 put:44 head:ffff8c03c7a33

  ------------[ cut here ]------------
  kernel BUG at net/core/skbuff.c:109!
  Call Trace:
  skb_push+0x10/0x10
  ip6gre_header+0x47/0x1b0
  neigh_connected_output+0xae/0xf0

ip6gre tunnel provides its own header_ops->create, and sets it conditionally when initializing the tunnel in NBMA mode. When header_ops->create is used, dev->hard_header_len should reflect the length of the header created. Otherwise, when not used, dev->needed_headroom should be used.

---
 net/ipv6/ip6_gre.c |   16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)

@@ -1122,8 +1122,13 @@ static void ip6gre_tnl_link_config_route
 			return;
 
 		if (rt->dst.dev) {
-			dev->needed_headroom = rt->dst.dev->hard_header_len +
-					       t_hlen;
+			unsigned short dst_len = rt->dst.dev->hard_header_len +
+						 t_hlen;
+
+			if (t->dev->header_ops)
+				dev->hard_header_len = dst_len;
+			else
+				dev->needed_headroom = dst_len;
 
 			if (set_mtu) {
 				dev->mtu = rt->dst.dev->mtu - t_hlen;

@@ -1148,7 +1153,12 @@ static int ip6gre_calc_hlen(struct ip6_t
 	tunnel->hlen = tunnel->tun_hlen + tunnel->encap_hlen;
 
 	t_hlen = tunnel->hlen + sizeof(struct ipv6hdr);
-	tunnel->dev->needed_headroom = LL_MAX_HEADER + t_hlen;
+
+	if (tunnel->dev->header_ops)
+		tunnel->dev->hard_header_len = LL_MAX_HEADER + t_hlen;
+	else
+		tunnel->dev->needed_headroom = LL_MAX_HEADER + t_hlen;
+
 	return t_hlen;
 }
 

From: Yangbo Lu <[email protected]>

[ Upstream commit 07500a6085806d97039ebcba8d9b8b29129f0106 ]

The timestamp fields should be copied to new skb too in A-050385 workaround for later TX timestamping handling.

---
 drivers/net/ethernet/freescale/dpaa/dpaa_eth.c |   10 +++++++++-
 1 file changed, 9 insertions(+), 1 deletion(-)

@@ -2120,6 +2120,15 @@ workaround:
 	skb_copy_header(new_skb, skb);
 	new_skb->dev = skb->dev;
 
+	/* Copy relevant timestamp info from the old skb to the new */
+	if (priv->tx_tstamp) {
+		skb_shinfo(new_skb)->tx_flags = skb_shinfo(skb)->tx_flags;
+		skb_shinfo(new_skb)->hwtstamps = skb_shinfo(skb)->hwtstamps;
+		skb_shinfo(new_skb)->tskey = skb_shinfo(skb)->tskey;
+		if (skb->sk)
+			skb_set_owner_w(new_skb, skb->sk);
+	}
+
 	/* We move the headroom when we align it so we have to reset the
 	 * network and transport header offsets relative to the new data
 	 * pointer. The checksum offload relies on these offsets.

@@ -2127,7 +2136,6 @@ workaround:
 	skb_set_network_header(new_skb, skb_network_offset(skb));
 	skb_set_transport_header(new_skb, skb_transport_offset(skb));
 
-	/* TODO: does timestamping need the result in the old skb? */
 	dev_kfree_skb(skb);
 	*s = new_skb;
 

From: Dan Carpenter <[email protected]>

[ Upstream commit 6ee50c8e262a0f0693dad264c3c99e30e6442a56 ]

The .x25_addr[] address comes from the user and is not necessarily NUL terminated. This leads to a couple problems. The first problem is that the strlen() in x25_bind() can read beyond the end of the buffer.

The second problem is more subtle and could result in memory corruption. The call tree is:
x25_connect()
--> x25_write_internal()

      --> x25_addr_aton()

The .x25_addr[] buffers are copied to the "addresses" buffer from x25_write_internal() so it will lead to stack corruption.

Verify that the strings are NUL terminated and return -EINVAL if they are not.

---
 net/x25/af_x25.c |    6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)

@@ -681,7 +681,8 @@ static int x25_bind(struct socket *sock,
 	int len, i, rc = 0;
 
 	if (addr_len != sizeof(struct sockaddr_x25) ||
-	    addr->sx25_family != AF_X25) {
+	    addr->sx25_family != AF_X25 ||
+	    strnlen(addr->sx25_addr.x25_addr, X25_ADDR_LEN) == X25_ADDR_LEN) {
 		rc = -EINVAL;
 		goto out;
 	}

@@ -775,7 +776,8 @@ static int x25_connect(struct socket *so
 
 	rc = -EINVAL;
 	if (addr_len != sizeof(struct sockaddr_x25) ||
-	    addr->sx25_family != AF_X25)
+	    addr->sx25_family != AF_X25 ||
+	    strnlen(addr->sx25_addr.x25_addr, X25_ADDR_LEN) == X25_ADDR_LEN)
 		goto out;
 
 	rc = -ENETUNREACH;

From: Zhang Changzhong <[email protected]>

[ Upstream commit ff9924897f8bfed82e61894b373ab9d2dfea5b10 ]

Fix to return a negative error code from the error handling case instead of 0, as done elsewhere in this function.

---
 drivers/net/ethernet/chelsio/cxgb3/sge.c |    1 +
 1 file changed, 1 insertion(+)

@@ -3176,6 +3176,7 @@ int t3_sge_alloc_qset(struct adapter *ad
 			  GFP_KERNEL | __GFP_COMP);
 	if (!avail) {
 		CH_ALERT(adapter, "free list queue 0 initialization failed\n");
+		ret = -ENOMEM;
 		goto err;
 	}
 	if (avail < q->fl[0].size)

From: Zhang Changzhong <[email protected]>

[ Upstream commit aba84871bd4f52c4dfcf3ad5d4501a6c9d2de90e ]

Fix to return a negative error code from the error handling case instead of 0, as done elsewhere in this function.

---
 drivers/net/ethernet/pasemi/pasemi_mac.c |    8 ++++++--
 1 file changed, 6 insertions(+), 2 deletions(-)

@@ -1078,16 +1078,20 @@ static int pasemi_mac_open(struct net_de
 
 	mac->tx = pasemi_mac_setup_tx_resources(dev);
 
-	if (!mac->tx)
+	if (!mac->tx) {
+		ret = -ENOMEM;
 		goto out_tx_ring;
+	}
 
 	/* We might already have allocated rings in case mtu was changed
 	 * before interface was brought up.
 	 */
 	if (dev->mtu > 1500 && !mac->num_cs) {
 		pasemi_mac_setup_csrings(mac);
-		if (!mac->num_cs)
+		if (!mac->num_cs) {
+			ret = -ENOMEM;
 			goto out_tx_ring;
+		}
 	}
 
 	/* Zero out rmon counters */

From: Zhang Changzhong <[email protected]>

[ Upstream commit 832e09798c261cf58de3a68cfcc6556408c16a5a ]

Fix to return a negative error code from the error handling case instead of 0, as done elsewhere in this function.

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

@@ -3881,8 +3881,10 @@ static int __vxlan_dev_create(struct net
 
 	if (dst->remote_ifindex) {
 		remote_dev = __dev_get_by_index(net, dst->remote_ifindex);
-		if (!remote_dev)
+		if (!remote_dev) {
+			err = -ENODEV;
 			goto errout;
+		}
 
 		err = netdev_upper_dev_link(remote_dev, dev, extack);
 		if (err)

From: Dan Carpenter <[email protected]>

[ Upstream commit 391119fb5c5c4bdb4d57c7ffeb5e8d18560783d1 ]

The "skb" is freed by the transmit code in cxgb4_ofld_send() and we shouldn't use it again. But in the current code, if we hit an error later on in the function then the clean up code will call kfree_skb(skb) and so it causes a double free.

Set the "skb" to NULL and that makes the kfree_skb() a no-op.

---
 drivers/crypto/chelsio/chtls/chtls_hw.c |    1 +
 1 file changed, 1 insertion(+)

@@ -391,6 +391,7 @@ int chtls_setkey(struct chtls_sock *csk,
 	csk->wr_unacked += DIV_ROUND_UP(len, 16);
 	enqueue_wr(csk, skb);
 	cxgb4_ofld_send(csk->egress_dev, skb);
+	skb = NULL;
 
 	chtls_set_scmd(csk);
 	/* Clear quiesce for Rx key */

From: Wang Hai <[email protected]>

[ Upstream commit 82a10dc7f0960735f40e8d7d3bee56934291600f ]

Fix to return negative error code -ENOENT from invalid configuration error handling case instead of 0, as done elsewhere in this function.

---
 drivers/net/ethernet/marvell/mvpp2/mvpp2_main.c |    1 +
 1 file changed, 1 insertion(+)

@@ -4249,6 +4249,7 @@ static int mvpp2_open(struct net_device
 	if (!valid) {
 		netdev_err(port->dev,
 			   "invalid configuration: no dt or link IRQ");
+		err = -ENOENT;
 		goto err_free_irq;
 	}
 

From: Davide Caratti <[email protected]>

[ Upstream commit 13de4ed9e3a9ccbe54d05f7d5c773f69ecaf6c64 ]

skb_mpls_dec_ttl() reads the LSE without ensuring that it is contained in the skb "linear" area. Fix this calling pskb_may_pull() before reading the current ttl.

Found by code inspection.

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

@@ -5725,6 +5725,9 @@ int skb_mpls_dec_ttl(struct sk_buff *skb
 	if (unlikely(!eth_p_mpls(skb->protocol)))
 		return -EINVAL;
 
+	if (!pskb_may_pull(skb, skb_network_offset(skb) + MPLS_HLEN))
+		return -ENOMEM;
+
 	lse = be32_to_cpu(mpls_hdr(skb)->label_stack_entry);
 	ttl = (lse & MPLS_LS_TTL_MASK) >> MPLS_LS_TTL_SHIFT;
 	if (!--ttl)

From: Davide Caratti <[email protected]>

[ Upstream commit 43c13605bad44b8abbc9776d6e63f62ccb7a47d6 ]

when openvswitch is configured to mangle the LSE, the current value is read from the packet dereferencing 4 bytes at mpls_hdr(): ensure that the label is contained in the skb "linear" area.

Found by code inspection.

---
 net/openvswitch/actions.c |    3 +++
 1 file changed, 3 insertions(+)

@@ -200,6 +200,9 @@ static int set_mpls(struct sk_buff *skb,
 	__be32 lse;
 	int err;
 
+	if (!pskb_may_pull(skb, skb_network_offset(skb) + MPLS_HLEN))
+		return -ENOMEM;
+
 	stack = mpls_hdr(skb);
 	lse = OVS_MASKED(stack->label_stack_entry, *mpls_lse, *mask);
 	err = skb_mpls_update_lse(skb, lse);

From: Davide Caratti <[email protected]>

[ Upstream commit 9608fa653059c3f72faab0c148ac8773c46e7314 ]

when 'act_mpls' is used to mangle the LSE, the current value is read from the packet dereferencing 4 bytes at mpls_hdr(): ensure that the label is contained in the skb "linear" area.

Found by code inspection.

v2:
- use MPLS_HLEN instead of sizeof(new_lse), thanks to Jakub Kicinski

---
 net/sched/act_mpls.c |    3 +++
 1 file changed, 3 insertions(+)

@@ -88,6 +88,9 @@ static int tcf_mpls_act(struct sk_buff *
 			goto drop;
 		break;
 	case TCA_MPLS_ACT_MODIFY:
+		if (!pskb_may_pull(skb,
+				   skb_network_offset(skb) + MPLS_HLEN))
+			goto drop;
 		new_lse = tcf_mpls_get_lse(mpls_hdr(skb), p, false);
 		if (skb_mpls_update_lse(skb, new_lse))
 			goto drop;

From: Yevgeny Kliteynik <[email protected]>

[ Upstream commit d421e466c2373095f165ddd25cbabd6c5b077928 ]

STEs format for Connect-X5 and Connect-X6DX different. Currently, on Connext-X6DX the SW steering would break at some point when building STEs w/o giving a proper error message. Fix this by checking the STE format of the current device when initializing domain: add mlx5_ifc definitions for Connect-X6DX SW steering, read FW capability to get the current format version, and check this version when domain is being created.

---
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_cmd.c    |    1 +
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_domain.c |    5 +++++
 drivers/net/ethernet/mellanox/mlx5/core/steering/dr_types.h  |    1 +
 include/linux/mlx5/mlx5_ifc.h                                |    9 ++++++++-
 4 files changed, 15 insertions(+), 1 deletion(-)

@@ -92,6 +92,7 @@ int mlx5dr_cmd_query_device(struct mlx5_
 	caps->eswitch_manager	= MLX5_CAP_GEN(mdev, eswitch_manager);
 	caps->gvmi		= MLX5_CAP_GEN(mdev, vhca_id);
 	caps->flex_protocols	= MLX5_CAP_GEN(mdev, flex_parser_protocols);
+	caps->sw_format_ver	= MLX5_CAP_GEN(mdev, steering_format_version);
 
 	if (mlx5dr_matcher_supp_flex_parser_icmp_v4(caps)) {
 		caps->flex_parser_id_icmp_dw0 = MLX5_CAP_GEN(mdev, flex_parser_id_icmp_dw0);

@@ -223,6 +223,11 @@ static int dr_domain_caps_init(struct ml
 	if (ret)
 		return ret;
 
+	if (dmn->info.caps.sw_format_ver != MLX5_STEERING_FORMAT_CONNECTX_5) {
+		mlx5dr_err(dmn, "SW steering is not supported on this device\n");
+		return -EOPNOTSUPP;
+	}
+
 	ret = dr_domain_query_fdb_caps(mdev, dmn);
 	if (ret)
 		return ret;

@@ -621,6 +621,7 @@ struct mlx5dr_cmd_caps {
 	u8 max_ft_level;
 	u16 roce_min_src_udp;
 	u8 num_esw_ports;
+	u8 sw_format_ver;
 	bool eswitch_manager;
 	bool rx_sw_owner;
 	bool tx_sw_owner;

@@ -1222,6 +1222,11 @@ enum mlx5_fc_bulk_alloc_bitmask {
 
 #define MLX5_FC_BULK_NUM_FCS(fc_enum) (MLX5_FC_BULK_SIZE_FACTOR * (fc_enum))
 
+enum {
+	MLX5_STEERING_FORMAT_CONNECTX_5   = 0,
+	MLX5_STEERING_FORMAT_CONNECTX_6DX = 1,
+};
+
 struct mlx5_ifc_cmd_hca_cap_bits {
 	u8         reserved_at_0[0x30];
 	u8         vhca_id[0x10];

@@ -1519,7 +1524,9 @@ struct mlx5_ifc_cmd_hca_cap_bits {
 
 	u8         general_obj_types[0x40];
 
-	u8         reserved_at_440[0x20];
+	u8         reserved_at_440[0x4];
+	u8         steering_format_version[0x4];
+	u8         create_qp_start_hint[0x18];
 
 	u8         reserved_at_460[0x3];
 	u8         log_max_uctx[0x5];

From: Eran Ben Elisha <[email protected]>

[ Upstream commit 1d2bb5ad89f47d8ce8aedc70ef85059ab3870292 ]

When command interface is down, driver to reclaim all 4K page chucks that were hold by the Firmeware. Fix a bug for 64K page size systems, where driver repeatedly released only the first chunk of the page.

Define helper function to fill 4K chunks for a given Firmware pages. Iterate over all unreleased Firmware pages and call the hepler per each.

---
 drivers/net/ethernet/mellanox/mlx5/core/pagealloc.c |   21 ++++++++++++++++++--
 1 file changed, 19 insertions(+), 2 deletions(-)

@@ -422,6 +422,24 @@ static void release_all_pages(struct mlx
 		      npages, ec_function, func_id);
 }
 
+static u32 fwp_fill_manage_pages_out(struct fw_page *fwp, u32 *out, u32 index,
+				     u32 npages)
+{
+	u32 pages_set = 0;
+	unsigned int n;
+
+	for_each_clear_bit(n, &fwp->bitmask, MLX5_NUM_4K_IN_PAGE) {
+		MLX5_ARRAY_SET64(manage_pages_out, out, pas, index + pages_set,
+				 fwp->addr + (n * MLX5_ADAPTER_PAGE_SIZE));
+		pages_set++;
+
+		if (!--npages)
+			break;
+	}
+
+	return pages_set;
+}
+
 static int reclaim_pages_cmd(struct mlx5_core_dev *dev,
 			     u32 *in, int in_size, u32 *out, int out_size)
 {

@@ -448,8 +466,7 @@ static int reclaim_pages_cmd(struct mlx5
 		fwp = rb_entry(p, struct fw_page, rb_node);
 		p = rb_next(p);
 
-		MLX5_ARRAY_SET64(manage_pages_out, out, pas, i, fwp->addr);
-		i++;
+		i += fwp_fill_manage_pages_out(fwp, out, i, npages - i);
 	}
 
 	MLX5_SET(manage_pages_out, out, output_num_entries, i);

From: Randy Dunlap <[email protected]>

[ Upstream commit 8a78a440108e55ddd845b0ef46df575248667520 ]

Fix build when CONFIG_IPV6 is not enabled by making a function be built conditionally.

Fixes these build errors and warnings:

../drivers/net/ethernet/mellanox/mlx5/core/en_accel/fs_tcp.c: In function 'accel_fs_tcp_set_ipv6_flow':
../include/net/sock.h:380:34: error: 'struct sock_common' has no member named 'skc_v6_daddr'; did you mean 'skc_daddr'?
380 | #define sk_v6_daddr __sk_common.skc_v6_daddr

      |                                  ^~~~~~~~~~~~
../drivers/net/ethernet/mellanox/mlx5/core/en_accel/fs_tcp.c:55:14: note: in expansion of macro 'sk_v6_daddr'
   55 |         &sk->sk_v6_daddr, 16);
      |              ^~~~~~~~~~~

At top level:
../drivers/net/ethernet/mellanox/mlx5/core/en_accel/fs_tcp.c:47:13: warning: 'accel_fs_tcp_set_ipv6_flow' defined but not used [-Wunused-function]
47 | static void accel_fs_tcp_set_ipv6_flow(struct mlx5_flow_spec *spec, struct sock *sk)

---
 drivers/net/ethernet/mellanox/mlx5/core/en_accel/fs_tcp.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -44,6 +44,7 @@ static void accel_fs_tcp_set_ipv4_flow(s
 			 outer_headers.dst_ipv4_dst_ipv6.ipv4_layout.ipv4);
 }
 
+#if IS_ENABLED(CONFIG_IPV6)
 static void accel_fs_tcp_set_ipv6_flow(struct mlx5_flow_spec *spec, struct sock *sk)
 {
 	MLX5_SET_TO_ONES(fte_match_param, spec->match_criteria, outer_headers.ip_protocol);

@@ -63,6 +64,7 @@ static void accel_fs_tcp_set_ipv6_flow(s
 			    outer_headers.dst_ipv4_dst_ipv6.ipv6_layout.ipv6),
 	       0xff, 16);
 }
+#endif
 
 void mlx5e_accel_fs_del_sk(struct mlx5_flow_handle *rule)
 {

From: Hector Martin <[email protected]>

commit 402d5840b0d40a2a26c8651165d29b534abb6d36 upstream.

The level meter control returns 34 integers of info. This fixes:

snd-usb-audio 3-1:1.0: control 2:0:0:Level Meter:0: access overflow

---
 sound/usb/mixer_us16x08.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

@@ -607,7 +607,7 @@ static int snd_us16x08_eq_put(struct snd
 static int snd_us16x08_meter_info(struct snd_kcontrol *kcontrol,
 	struct snd_ctl_elem_info *uinfo)
 {
-	uinfo->count = 1;
+	uinfo->count = 34;
 	uinfo->type = SNDRV_CTL_ELEM_TYPE_INTEGER;
 	uinfo->value.integer.max = 0x7FFF;
 	uinfo->value.integer.min = 0;

From: Sanjay Govind <[email protected]>

commit 2aab1561439032be2e98811dd0ddbeb5b2ae4c61 upstream.

This commit adds support for Ardwiino Controllers

---
 drivers/input/joystick/xpad.c |    2 ++
 1 file changed, 2 insertions(+)

@@ -241,6 +241,7 @@ static const struct xpad_device {
 	{ 0x1038, 0x1430, "SteelSeries Stratus Duo", 0, XTYPE_XBOX360 },
 	{ 0x1038, 0x1431, "SteelSeries Stratus Duo", 0, XTYPE_XBOX360 },
 	{ 0x11c9, 0x55f0, "Nacon GC-100XF", 0, XTYPE_XBOX360 },
+	{ 0x1209, 0x2882, "Ardwiino Controller", 0, XTYPE_XBOX360 },
 	{ 0x12ab, 0x0004, "Honey Bee Xbox360 dancepad", MAP_DPAD_TO_BUTTONS, XTYPE_XBOX360 },
 	{ 0x12ab, 0x0301, "PDP AFTERGLOW AX.1", 0, XTYPE_XBOX360 },
 	{ 0x12ab, 0x0303, "Mortal Kombat Klassic FightStick", MAP_TRIGGERS_TO_BUTTONS, XTYPE_XBOX360 },

@@ -418,6 +419,7 @@ static const struct usb_device_id xpad_t
 	XPAD_XBOXONE_VENDOR(0x0f0d),		/* Hori Controllers */
 	XPAD_XBOX360_VENDOR(0x1038),		/* SteelSeries Controllers */
 	XPAD_XBOX360_VENDOR(0x11c9),		/* Nacon GC100XF */
+	XPAD_XBOX360_VENDOR(0x1209),		/* Ardwiino Controllers */
 	XPAD_XBOX360_VENDOR(0x12ab),		/* X-Box 360 dance pads */
 	XPAD_XBOX360_VENDOR(0x1430),		/* RedOctane X-Box 360 controllers */
 	XPAD_XBOX360_VENDOR(0x146b),		/* BigBen Interactive Controllers */

From: Po-Hsu Lin <[email protected]>

commit a48491c65b513e5cdc3e7a886a4db915f848a5f5 upstream.

It looks like the C15B laptop got another vendor: ByteSpeed LLC.

Avoid AUX loopback on this touchpad as well, thus input subsystem will be able to recognize a Synaptics touchpad in the AUX port.

---
 drivers/input/serio/i8042-x86ia64io.h |    4 ++++
 1 file changed, 4 insertions(+)

@@ -219,6 +219,10 @@ static const struct dmi_system_id __init
 			DMI_MATCH(DMI_SYS_VENDOR, "PEGATRON CORPORATION"),
 			DMI_MATCH(DMI_PRODUCT_NAME, "C15B"),
 		},
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "ByteSpeed LLC"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "ByteSpeed Laptop C15B"),
+		},
 	},
 	{ }
 };

From: Linus Walleij <[email protected]>

commit 8c3b55a299c325830a987de21dab6a89ecb71164 upstream.

After commit 74d905d2d38a devices requiring the workaround for edge triggered interrupts stopped working.

The hardware needs the quirk to be used before even proceeding to check if the quirk is needed because mxt_acquire_irq() is called before mxt_check_retrigen() is called and at this point pending IRQs need to be checked, and if the workaround is not active, all interrupts will be lost from this point.

Solve this by switching the calls around.

---
 drivers/input/touchscreen/atmel_mxt_ts.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

@@ -2183,11 +2183,11 @@ static int mxt_initialize(struct mxt_dat
 		msleep(MXT_FW_RESET_TIME);
 	}
 
-	error = mxt_acquire_irq(data);
+	error = mxt_check_retrigen(data);
 	if (error)
 		return error;
 
-	error = mxt_check_retrigen(data);
+	error = mxt_acquire_irq(data);
 	if (error)
 		return error;
 

From: Minchan Kim <[email protected]>

commit 8fa655a3a0013a0c2a2aada6f39a93ee6fc25549 upstream.

With 5.9 kernel on ARM64, I found ftrace_dump output was broken but it had no problem with normal output "cat /sys/kernel/debug/tracing/trace".

With investigation, it seems coping the data into temporal buffer seems to break the align binary printf expects if the static buffer is not aligned with 4-byte. IIUC, get_arg in bstr_printf expects that args has already right align to be decoded and seq_buf_bprintf says ``the arguments are saved in a 32bit word array that is defined by the format string constraints``. So if we don't keep the align under copy to temporal buffer, the output will be broken by shifting some bytes.

This patch fixes it.

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

@@ -3516,7 +3516,7 @@ __find_next_entry(struct trace_iterator
 }
 
 #define STATIC_TEMP_BUF_SIZE	128
-static char static_temp_buf[STATIC_TEMP_BUF_SIZE];
+static char static_temp_buf[STATIC_TEMP_BUF_SIZE] __aligned(4);
 
 /* Find the next real entry, without updating the iterator itself */
 struct trace_entry *trace_find_next_entry(struct trace_iterator *iter,

From: Vasily Averin <[email protected]>

commit 310e3a4b5a4fc718a72201c1e4cf5c64ac6f5442 upstream.

This patch reverts commit 978defee11a5 ("tracing: Do a WARN_ON()
if start_thread() in hwlat is called when thread exists")

.start hook can be legally called several times if according tracer is stopped

screen window 1
[root@localhost ~]# echo 1 > /sys/kernel/tracing/events/kmem/kfree/enable [root@localhost ~]# echo 1 > /sys/kernel/tracing/options/pause-on-trace [root@localhost ~]# less -F /sys/kernel/tracing/trace

screen window 2
[root@localhost ~]# cat /sys/kernel/debug/tracing/tracing_on
0
[root@localhost ~]# echo hwlat > /sys/kernel/debug/tracing/current_tracer [root@localhost ~]# echo 1 > /sys/kernel/debug/tracing/tracing_on [root@localhost ~]# cat /sys/kernel/debug/tracing/tracing_on
0
[root@localhost ~]# echo 2 > /sys/kernel/debug/tracing/tracing_on

triggers warning in dmesg:

WARNING: CPU: 3 PID: 1403 at kernel/trace/trace_hwlat.c:371 hwlat_tracer_start+0xc9/0xd0

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

@@ -368,7 +368,7 @@ static int start_kthread(struct trace_ar
 	struct task_struct *kthread;
 	int next_cpu;
 
-	if (WARN_ON(hwlat_kthread))
+	if (hwlat_kthread)
 		return 0;
 
 	/* Just pick the first CPU on first iteration */

From: Eric Sandeen <[email protected]>

commit 72d1249e2ffdbc344e465031ec5335fa3489d62e upstream.

STATX_ATTR_MOUNT_ROOT and STATX_ATTR_DAX got merged with the same value, so one of them needs fixing. Move STATX_ATTR_DAX.

While we're in here, clarify the value-matching scheme for some of the attributes, and explain why the value for DAX does not match.

---
 include/uapi/linux/stat.h |    9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

@@ -171,9 +171,12 @@ struct statx {
  * be of use to ordinary userspace programs such as GUIs or ls rather than
  * specialised tools.
  *
- * Note that the flags marked [I] correspond to generic FS_IOC_FLAGS
+ * Note that the flags marked [I] correspond to the FS_IOC_SETFLAGS flags
  * semantically.  Where possible, the numerical value is picked to correspond
- * also.
+ * also.  Note that the DAX attribute indicates that the file is in the CPU
+ * direct access state.  It does not correspond to the per-inode flag that
+ * some filesystems support.
+ *
  */
 #define STATX_ATTR_COMPRESSED		0x00000004 /* [I] File is compressed by the fs */
 #define STATX_ATTR_IMMUTABLE		0x00000010 /* [I] File is marked immutable */

@@ -183,7 +186,7 @@ struct statx {
 #define STATX_ATTR_AUTOMOUNT		0x00001000 /* Dir: Automount trigger */
 #define STATX_ATTR_MOUNT_ROOT		0x00002000 /* Root of a mount */
 #define STATX_ATTR_VERITY		0x00100000 /* [I] Verity protected file */
-#define STATX_ATTR_DAX			0x00002000 /* [I] File is DAX */
+#define STATX_ATTR_DAX			0x00200000 /* File is currently in DAX state */
 
 
 #endif /* _UAPI_LINUX_STAT_H */

From: Chris Wilson <[email protected]>

commit be33805c65297611971003d72e7f9235e23ec84d upstream.

Forcing mocs:1 [used for our winsys follows-pte mode] to be cached caused display glitches. Though it is documented as deprecated (and so likely behaves as uncached) use the follow-pte bit and force it out of L3 cache.

Testcase: igt/kms_frontbuffer_tracking
Testcase: igt/kms_big_fb

---
 drivers/gpu/drm/i915/gt/intel_mocs.c |    5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

@@ -243,8 +243,9 @@ static const struct drm_i915_mocs_entry
 	 * only, __init_mocs_table() take care to program unused index with
 	 * this entry.
 	 */
-	MOCS_ENTRY(1, LE_3_WB | LE_TC_1_LLC | LE_LRUM(3),
-		   L3_3_WB),
+	MOCS_ENTRY(I915_MOCS_PTE,
+		   LE_0_PAGETABLE | LE_TC_0_PAGETABLE,
+		   L3_1_UC),
 	GEN11_MOCS_ENTRIES,
 
 	/* Implicitly enable L1 - HDC:L1 + L3 + LLC */

hello,

Compiled and booted 5.9.13-rc1+. No typical regression or regressions.

software engineer
rajagiri school of engineering and technology - autonomous

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

NOTE:
Following two warnings noticed.
WARNING: bad unlock balance detected! - mkfs.ext4/426 is trying to release lock (rcu_read_lock)
https://lore.kernel.org/stable/CA+G9fYs=nR-d0n8kV4=OWD+v=GR2ufOEWU9S4oG1_fZRxhGouQ@mail.gmail.com/

sched: core.c:7270 Illegal context switch in RCU-bh read-side critical section! __alloc_pages_nodemask
https://lore.kernel.org/stable/CA+G9fYvhJTwQkGyH7HQzSsDBHT7pm5ziA9VTkRhE_bDSQp3JYg@mail.gmail.com/

Summary
------------------------------------------------------------------------

kernel: 5.9.13-rc1
git repo: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git
git branch: linux-5.9.y
git commit: 1372e1af58d410676db7917cc3484ca22d471623
git describe: v5.9.12-47-g1372e1af58d4
Test details: https://qa-reports.linaro.org/lkft/linux-stable-rc-linux-5.9.y/build/v5.9.12-47-g1372e1af58d4

No regressions (compared to build v5.9.12)

No fixes (compared to build v5.9.12)

Ran 53360 total tests in the following environments and test suites.

Environments
--------------
- arc
- arm
- arm64
- dragonboard-410c
- hi6220-hikey
- i386
- juno-r2
- juno-r2-compat
- juno-r2-kasan
- mips
- nxp-ls2088
- parisc
- powerpc
- qemu-arm-clang
- qemu-arm64-clang
- qemu-arm64-kasan
- qemu-i386-clang
- qemu-x86_64-clang
- qemu-x86_64-kasan
- qemu-x86_64-kcsan
- qemu_arm
- qemu_arm64
- qemu_arm64-compat
- qemu_i386
- qemu_x86_64
- qemu_x86_64-compat
- riscv
- s390
- sh
- sparc
- x15
- x86
- x86-kasan

Test Suites
-----------
* build
* linux-log-parser
* install-android-platform-tools-r2600
* libhugetlbfs
* ltp-cap_bounds-tests
* ltp-containers-tests
* ltp-cpuhotplug-tests
* ltp-crypto-tests
* ltp-cve-tests
* ltp-dio-tests
* ltp-hugetlb-tests
* ltp-io-tests
* ltp-ipc-tests
* ltp-mm-tests
* ltp-nptl-tests
* ltp-pty-tests
* ltp-sched-tests
* ltp-securebits-tests
* ltp-syscalls-tests
* ltp-tracing-tests
* perf
* v4l2-compliance
* kvm-unit-tests
* ltp-commands-tests
* ltp-fcntl-locktests-tests
* ltp-filecaps-tests
* ltp-fs-tests
* ltp-fs_bind-tests
* ltp-fs_perms_simple-tests
* ltp-fsx-tests
* ltp-math-tests
* network-basic-tests
* kselftest
* ltp-controllers-tests
* ltp-open-posix-tests
* kunit
* kselftest-vsyscall-mode-native
* kselftest-vsyscall-mode-none

Linaro LKFT
https://lkft.linaro.org

Thanks for testing all of these and letting me know.

greg k-h

All tests passing for Tegra ...

Test results for stable-v5.9:

    15 builds:	15 pass, 0 fail
    26 boots:	26 pass, 0 fail
    64 tests:	64 pass, 0 fail

Linux version: 5.9.13-rc1-g1372e1af58d4
Boards tested: tegra124-jetson-tk1, tegra186-p2771-0000,

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

Jon

Thanks for testing all of these and letting me know.

greg k-h

Build results:

	total: 154 pass: 154 fail: 0
Qemu test results:
	total: 426 pass: 426 fail: 0

Guenter

Thanks for testing them all and letting me know.

greg k-h