The patch below does not apply to the 5.10-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable(a)vger.kernel.org>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
>From ae2778a64724f77fd6cad674461a045fb3307df7 Mon Sep 17 00:00:00 2001
From: Xiaoliang Yang <xiaoliang.yang_1(a)nxp.com>
Date: Thu, 23 Dec 2021 15:22:11 +0800
Subject: [PATCH] net: dsa: tag_ocelot: use traffic class to map priority on
injected header
For Ocelot switches, the CPU injected frames have an injection header
where it can specify the QoS class of the packet and the DSA tag, now it
uses the SKB priority to set that. If a traffic class to priority
mapping is configured on the netdevice (with mqprio for example ...), it
won't be considered for CPU injected headers. This patch make the QoS
class aligned to the priority to traffic class mapping if it exists.
Fixes: 8dce89aa5f32 ("net: dsa: ocelot: add tagger for Ocelot/Felix switches")
Signed-off-by: Xiaoliang Yang <xiaoliang.yang_1(a)nxp.com>
Signed-off-by: Marouen Ghodhbane <marouen.ghodhbane(a)nxp.com>
Link: https://lore.kernel.org/r/20211223072211.33130-1-xiaoliang.yang_1@nxp.com
Signed-off-by: Jakub Kicinski <kuba(a)kernel.org>
diff --git a/net/dsa/tag_ocelot.c b/net/dsa/tag_ocelot.c
index de1c849a0a70..4ed74d509d6a 100644
--- a/net/dsa/tag_ocelot.c
+++ b/net/dsa/tag_ocelot.c
@@ -47,9 +47,13 @@ static void ocelot_xmit_common(struct sk_buff *skb, struct net_device *netdev,
void *injection;
__be32 *prefix;
u32 rew_op = 0;
+ u64 qos_class;
ocelot_xmit_get_vlan_info(skb, dp, &vlan_tci, &tag_type);
+ qos_class = netdev_get_num_tc(netdev) ?
+ netdev_get_prio_tc_map(netdev, skb->priority) : skb->priority;
+
injection = skb_push(skb, OCELOT_TAG_LEN);
prefix = skb_push(skb, OCELOT_SHORT_PREFIX_LEN);
@@ -57,7 +61,7 @@ static void ocelot_xmit_common(struct sk_buff *skb, struct net_device *netdev,
memset(injection, 0, OCELOT_TAG_LEN);
ocelot_ifh_set_bypass(injection, 1);
ocelot_ifh_set_src(injection, ds->num_ports);
- ocelot_ifh_set_qos_class(injection, skb->priority);
+ ocelot_ifh_set_qos_class(injection, qos_class);
ocelot_ifh_set_vlan_tci(injection, vlan_tci);
ocelot_ifh_set_tag_type(injection, tag_type);
The patch below does not apply to the 5.15-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable(a)vger.kernel.org>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
>From ae2778a64724f77fd6cad674461a045fb3307df7 Mon Sep 17 00:00:00 2001
From: Xiaoliang Yang <xiaoliang.yang_1(a)nxp.com>
Date: Thu, 23 Dec 2021 15:22:11 +0800
Subject: [PATCH] net: dsa: tag_ocelot: use traffic class to map priority on
injected header
For Ocelot switches, the CPU injected frames have an injection header
where it can specify the QoS class of the packet and the DSA tag, now it
uses the SKB priority to set that. If a traffic class to priority
mapping is configured on the netdevice (with mqprio for example ...), it
won't be considered for CPU injected headers. This patch make the QoS
class aligned to the priority to traffic class mapping if it exists.
Fixes: 8dce89aa5f32 ("net: dsa: ocelot: add tagger for Ocelot/Felix switches")
Signed-off-by: Xiaoliang Yang <xiaoliang.yang_1(a)nxp.com>
Signed-off-by: Marouen Ghodhbane <marouen.ghodhbane(a)nxp.com>
Link: https://lore.kernel.org/r/20211223072211.33130-1-xiaoliang.yang_1@nxp.com
Signed-off-by: Jakub Kicinski <kuba(a)kernel.org>
diff --git a/net/dsa/tag_ocelot.c b/net/dsa/tag_ocelot.c
index de1c849a0a70..4ed74d509d6a 100644
--- a/net/dsa/tag_ocelot.c
+++ b/net/dsa/tag_ocelot.c
@@ -47,9 +47,13 @@ static void ocelot_xmit_common(struct sk_buff *skb, struct net_device *netdev,
void *injection;
__be32 *prefix;
u32 rew_op = 0;
+ u64 qos_class;
ocelot_xmit_get_vlan_info(skb, dp, &vlan_tci, &tag_type);
+ qos_class = netdev_get_num_tc(netdev) ?
+ netdev_get_prio_tc_map(netdev, skb->priority) : skb->priority;
+
injection = skb_push(skb, OCELOT_TAG_LEN);
prefix = skb_push(skb, OCELOT_SHORT_PREFIX_LEN);
@@ -57,7 +61,7 @@ static void ocelot_xmit_common(struct sk_buff *skb, struct net_device *netdev,
memset(injection, 0, OCELOT_TAG_LEN);
ocelot_ifh_set_bypass(injection, 1);
ocelot_ifh_set_src(injection, ds->num_ports);
- ocelot_ifh_set_qos_class(injection, skb->priority);
+ ocelot_ifh_set_qos_class(injection, qos_class);
ocelot_ifh_set_vlan_tci(injection, vlan_tci);
ocelot_ifh_set_tag_type(injection, tag_type);
The patch below does not apply to the 5.15-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable(a)vger.kernel.org>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
>From 635d448a1cce4b4ebee52b351052c70434fa90ea Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb(a)nvidia.com>
Date: Tue, 14 Dec 2021 19:24:35 +0200
Subject: [PATCH] net: openvswitch: Fix matching zone id for invalid conns
arriving from tc
Zone id is not restored if we passed ct and ct rejected the connection,
as there is no ct info on the skb.
Save the zone from tc skb cb to tc skb extension and pass it on to
ovs, use that info to restore the zone id for invalid connections.
Fixes: d29334c15d33 ("net/sched: act_api: fix miss set post_ct for ovs after do conntrack in act_ct")
Signed-off-by: Paul Blakey <paulb(a)nvidia.com>
Signed-off-by: Jakub Kicinski <kuba(a)kernel.org>
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index 2ecf8cfd2223..4507d77d6941 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -286,6 +286,7 @@ struct nf_bridge_info {
struct tc_skb_ext {
__u32 chain;
__u16 mru;
+ __u16 zone;
bool post_ct;
};
#endif
diff --git a/net/openvswitch/flow.c b/net/openvswitch/flow.c
index 9713035b89e3..6d262d9aa10e 100644
--- a/net/openvswitch/flow.c
+++ b/net/openvswitch/flow.c
@@ -34,6 +34,7 @@
#include <net/mpls.h>
#include <net/ndisc.h>
#include <net/nsh.h>
+#include <net/netfilter/nf_conntrack_zones.h>
#include "conntrack.h"
#include "datapath.h"
@@ -860,6 +861,7 @@ int ovs_flow_key_extract(const struct ip_tunnel_info *tun_info,
#endif
bool post_ct = false;
int res, err;
+ u16 zone = 0;
/* Extract metadata from packet. */
if (tun_info) {
@@ -898,6 +900,7 @@ int ovs_flow_key_extract(const struct ip_tunnel_info *tun_info,
key->recirc_id = tc_ext ? tc_ext->chain : 0;
OVS_CB(skb)->mru = tc_ext ? tc_ext->mru : 0;
post_ct = tc_ext ? tc_ext->post_ct : false;
+ zone = post_ct ? tc_ext->zone : 0;
} else {
key->recirc_id = 0;
}
@@ -906,8 +909,11 @@ int ovs_flow_key_extract(const struct ip_tunnel_info *tun_info,
#endif
err = key_extract(skb, key);
- if (!err)
+ if (!err) {
ovs_ct_fill_key(skb, key, post_ct); /* Must be after key_extract(). */
+ if (post_ct && !skb_get_nfct(skb))
+ key->ct_zone = zone;
+ }
return err;
}
diff --git a/net/sched/cls_api.c b/net/sched/cls_api.c
index ff8a9383bf1c..35c74bdde848 100644
--- a/net/sched/cls_api.c
+++ b/net/sched/cls_api.c
@@ -1625,6 +1625,7 @@ int tcf_classify(struct sk_buff *skb,
ext->chain = last_executed_chain;
ext->mru = cb->mru;
ext->post_ct = cb->post_ct;
+ ext->zone = cb->zone;
}
return ret;
The patch below does not apply to the 5.15-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable(a)vger.kernel.org>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
>From 3849595866166b23bf6a0cb9ff87e06423167f67 Mon Sep 17 00:00:00 2001
From: Paul Blakey <paulb(a)nvidia.com>
Date: Tue, 14 Dec 2021 19:24:34 +0200
Subject: [PATCH] net/sched: flow_dissector: Fix matching on zone id for
invalid conns
If ct rejects a flow, it removes the conntrack info from the skb.
act_ct sets the post_ct variable so the dissector will see this case
as an +tracked +invalid state, but the zone id is lost with the
conntrack info.
To restore the zone id on such cases, set the last executed zone,
via the tc control block, when passing ct, and read it back in the
dissector if there is no ct info on the skb (invalid connection).
Fixes: 7baf2429a1a9 ("net/sched: cls_flower add CT_FLAGS_INVALID flag support")
Signed-off-by: Paul Blakey <paulb(a)nvidia.com>
Signed-off-by: Jakub Kicinski <kuba(a)kernel.org>
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index c8cb7e697d47..2ecf8cfd2223 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -1380,7 +1380,7 @@ skb_flow_dissect_ct(const struct sk_buff *skb,
struct flow_dissector *flow_dissector,
void *target_container,
u16 *ctinfo_map, size_t mapsize,
- bool post_ct);
+ bool post_ct, u16 zone);
void
skb_flow_dissect_tunnel_info(const struct sk_buff *skb,
struct flow_dissector *flow_dissector,
diff --git a/include/net/pkt_sched.h b/include/net/pkt_sched.h
index 05f18e81f3e8..9e71691c491b 100644
--- a/include/net/pkt_sched.h
+++ b/include/net/pkt_sched.h
@@ -198,6 +198,7 @@ struct tc_skb_cb {
u16 mru;
bool post_ct;
+ u16 zone; /* Only valid if post_ct = true */
};
static inline struct tc_skb_cb *tc_skb_cb(const struct sk_buff *skb)
diff --git a/net/core/flow_dissector.c b/net/core/flow_dissector.c
index 3255f57f5131..1b094c481f1d 100644
--- a/net/core/flow_dissector.c
+++ b/net/core/flow_dissector.c
@@ -238,7 +238,7 @@ void
skb_flow_dissect_ct(const struct sk_buff *skb,
struct flow_dissector *flow_dissector,
void *target_container, u16 *ctinfo_map,
- size_t mapsize, bool post_ct)
+ size_t mapsize, bool post_ct, u16 zone)
{
#if IS_ENABLED(CONFIG_NF_CONNTRACK)
struct flow_dissector_key_ct *key;
@@ -260,6 +260,7 @@ skb_flow_dissect_ct(const struct sk_buff *skb,
if (!ct) {
key->ct_state = TCA_FLOWER_KEY_CT_FLAGS_TRACKED |
TCA_FLOWER_KEY_CT_FLAGS_INVALID;
+ key->ct_zone = zone;
return;
}
diff --git a/net/sched/act_ct.c b/net/sched/act_ct.c
index 98e248b9c0b1..ab3591408419 100644
--- a/net/sched/act_ct.c
+++ b/net/sched/act_ct.c
@@ -1049,6 +1049,7 @@ static int tcf_ct_act(struct sk_buff *skb, const struct tc_action *a,
skb_push_rcsum(skb, nh_ofs);
tc_skb_cb(skb)->post_ct = true;
+ tc_skb_cb(skb)->zone = p->zone;
out_clear:
if (defrag)
qdisc_skb_cb(skb)->pkt_len = skb->len;
diff --git a/net/sched/cls_flower.c b/net/sched/cls_flower.c
index 9782b93db1b3..ef54ed395874 100644
--- a/net/sched/cls_flower.c
+++ b/net/sched/cls_flower.c
@@ -311,6 +311,7 @@ static int fl_classify(struct sk_buff *skb, const struct tcf_proto *tp,
{
struct cls_fl_head *head = rcu_dereference_bh(tp->root);
bool post_ct = tc_skb_cb(skb)->post_ct;
+ u16 zone = tc_skb_cb(skb)->zone;
struct fl_flow_key skb_key;
struct fl_flow_mask *mask;
struct cls_fl_filter *f;
@@ -328,7 +329,7 @@ static int fl_classify(struct sk_buff *skb, const struct tcf_proto *tp,
skb_flow_dissect_ct(skb, &mask->dissector, &skb_key,
fl_ct_info_to_flower_map,
ARRAY_SIZE(fl_ct_info_to_flower_map),
- post_ct);
+ post_ct, zone);
skb_flow_dissect_hash(skb, &mask->dissector, &skb_key);
skb_flow_dissect(skb, &mask->dissector, &skb_key,
FLOW_DISSECTOR_F_STOP_BEFORE_ENCAP);
commit 890d5b40908bfd1a79be018d2d297cf9df60f4ee upstream.
When listening for notifications through netlink of a new interface being
registered, sporadically, it is possible for the MAC to be read as zero.
The zero MAC address lasts a short period of time and then switches to a
valid random MAC address.
This causes problems for netd in Android, which assumes that the interface
is malfunctioning and will not use it.
In the good case we get this log:
InterfaceController::getCfg() ifName usb0
hwAddr 92:a8:f0:73:79:5b ipv4Addr 0.0.0.0 flags 0x1002
In the error case we get these logs:
InterfaceController::getCfg() ifName usb0
hwAddr 00:00:00:00:00:00 ipv4Addr 0.0.0.0 flags 0x1002
netd : interfaceGetCfg("usb0")
netd : interfaceSetCfg() -> ServiceSpecificException
(99, "[Cannot assign requested address] : ioctl() failed")
The reason for the issue is the order in which the interface is setup,
it is first registered through register_netdev() and after the MAC
address is set.
Fixed by first setting the MAC address of the net_device and after that
calling register_netdev().
Cc: stable(a)vger.kernel.org
Signed-off-by: Marian Postevca <posteuca(a)mutex.one>
Fixes: bcd4a1c40bee885e ("usb: gadget: u_ether: construct with default values and add setters/getters")
---
drivers/usb/gadget/function/u_ether.c | 15 +++++----------
1 file changed, 5 insertions(+), 10 deletions(-)
diff --git a/drivers/usb/gadget/function/u_ether.c b/drivers/usb/gadget/function/u_ether.c
index 38a35f57b22c0..f59c20457e658 100644
--- a/drivers/usb/gadget/function/u_ether.c
+++ b/drivers/usb/gadget/function/u_ether.c
@@ -864,19 +864,23 @@ int gether_register_netdev(struct net_device *net)
{
struct eth_dev *dev;
struct usb_gadget *g;
- struct sockaddr sa;
int status;
if (!net->dev.parent)
return -EINVAL;
dev = netdev_priv(net);
g = dev->gadget;
+
+ memcpy(net->dev_addr, dev->dev_mac, ETH_ALEN);
+ net->addr_assign_type = NET_ADDR_RANDOM;
+
status = register_netdev(net);
if (status < 0) {
dev_dbg(&g->dev, "register_netdev failed, %d\n", status);
return status;
} else {
INFO(dev, "HOST MAC %pM\n", dev->host_mac);
+ INFO(dev, "MAC %pM\n", dev->dev_mac);
/* two kinds of host-initiated state changes:
* - iff DATA transfer is active, carrier is "on"
@@ -884,15 +888,6 @@ int gether_register_netdev(struct net_device *net)
*/
netif_carrier_off(net);
}
- sa.sa_family = net->type;
- memcpy(sa.sa_data, dev->dev_mac, ETH_ALEN);
- rtnl_lock();
- status = dev_set_mac_address(net, &sa);
- rtnl_unlock();
- if (status)
- pr_warn("cannot set self ethernet address: %d\n", status);
- else
- INFO(dev, "MAC %pM\n", dev->dev_mac);
return status;
}
--
2.32.0
From: Christian Brauner <christian.brauner(a)ubuntu.com>
Ceph always inherits the SGID bit if it is set on the parent inode,
while the generic inode_init_owner does not do this in a few cases where
it can create a possible security problem (cf. [1]).
Update ceph to strip the SGID bit just as inode_init_owner would.
This bug was detected by the mapped mount testsuite in [3]. The
testsuite tests all core VFS functionality and semantics with and
without mapped mounts. That is to say it functions as a generic VFS
testsuite in addition to a mapped mount testsuite. While working on
mapped mount support for ceph, SIGD inheritance was the only failing
test for ceph after the port.
The same bug was detected by the mapped mount testsuite in XFS in
January 2021 (cf. [2]).
[1]: commit 0fa3ecd87848 ("Fix up non-directory creation in SGID directories")
[2]: commit 01ea173e103e ("xfs: fix up non-directory creation in SGID directories")
[3]: https://git.kernel.org/fs/xfs/xfstests-dev.git
Cc: stable(a)vger.kernel.org (adapted to v5.10)
Signed-off-by: Christian Brauner <christian.brauner(a)ubuntu.com>
Reviewed-by: Jeff Layton <jlayton(a)kernel.org>
Signed-off-by: Ilya Dryomov <idryomov(a)gmail.com>
---
fs/ceph/file.c | 18 +++++++++++++++---
1 file changed, 15 insertions(+), 3 deletions(-)
diff --git a/fs/ceph/file.c b/fs/ceph/file.c
index 8e6855e7ed83..8ed881fd7440 100644
--- a/fs/ceph/file.c
+++ b/fs/ceph/file.c
@@ -603,13 +603,25 @@ static int ceph_finish_async_create(struct inode *dir, struct dentry *dentry,
in.cap.realm = cpu_to_le64(ci->i_snap_realm->ino);
in.cap.flags = CEPH_CAP_FLAG_AUTH;
in.ctime = in.mtime = in.atime = iinfo.btime;
- in.mode = cpu_to_le32((u32)mode);
in.truncate_seq = cpu_to_le32(1);
in.truncate_size = cpu_to_le64(-1ULL);
in.xattr_version = cpu_to_le64(1);
in.uid = cpu_to_le32(from_kuid(&init_user_ns, current_fsuid()));
- in.gid = cpu_to_le32(from_kgid(&init_user_ns, dir->i_mode & S_ISGID ?
- dir->i_gid : current_fsgid()));
+ if (dir->i_mode & S_ISGID) {
+ in.gid = cpu_to_le32(from_kgid(&init_user_ns, dir->i_gid));
+
+ /* Directories always inherit the setgid bit. */
+ if (S_ISDIR(mode))
+ mode |= S_ISGID;
+ else if ((mode & (S_ISGID | S_IXGRP)) == (S_ISGID | S_IXGRP) &&
+ !in_group_p(dir->i_gid) &&
+ !capable_wrt_inode_uidgid(dir, CAP_FSETID))
+ mode &= ~S_ISGID;
+ } else {
+ in.gid = cpu_to_le32(from_kgid(&init_user_ns, current_fsgid()));
+ }
+ in.mode = cpu_to_le32((u32)mode);
+
in.nlink = cpu_to_le32(1);
in.max_size = cpu_to_le64(lo->stripe_unit);
--
2.30.2