pm24.git/net, branch v6.5-rc1

Merge tag 'net-6.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

2023-07-05T22:44:45Z

Pull networking fixes from Jakub Kicinski: "Including fixes from bluetooth, bpf and wireguard. Current release - regressions: - nvme-tcp: fix comma-related oops after sendpage changes Current release - new code bugs: - ptp: make max_phase_adjustment sysfs device attribute invisible when not supported Previous releases - regressions: - sctp: fix potential deadlock on &net->sctp.addr_wq_lock - mptcp: - ensure subflow is unhashed before cleaning the backlog - do not rely on implicit state check in mptcp_listen() Previous releases - always broken: - net: fix net_dev_start_xmit trace event vs skb_transport_offset() - Bluetooth: - fix use-bdaddr-property quirk - L2CAP: fix multiple UaFs - ISO: use hci_sync for setting CIG parameters - hci_event: fix Set CIG Parameters error status handling - hci_event: fix parsing of CIS Established Event - MGMT: fix marking SCAN_RSP as not connectable - wireguard: queuing: use saner cpu selection wrapping - sched: act_ipt: various bug fixes for iptables <> TC interactions - sched: act_pedit: add size check for TCA_PEDIT_PARMS_EX - dsa: fixes for receiving PTP packets with 8021q and sja1105 tagging - eth: sfc: fix null-deref in devlink port without MAE access - eth: ibmvnic: do not reset dql stats on NON_FATAL err Misc: - xsk: honor SO_BINDTODEVICE on bind" * tag 'net-6.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net: (70 commits) nfp: clean mc addresses in application firmware when closing port selftests: mptcp: pm_nl_ctl: fix 32-bit support selftests: mptcp: depend on SYN_COOKIES selftests: mptcp: userspace_pm: report errors with 'remove' tests selftests: mptcp: userspace_pm: use correct server port selftests: mptcp: sockopt: return error if wrong mark selftests: mptcp: sockopt: use 'iptables-legacy' if available selftests: mptcp: connect: fail if nft supposed to work mptcp: do not rely on implicit state check in mptcp_listen() mptcp: ensure subflow is unhashed before cleaning the backlog s390/qeth: Fix vipa deletion octeontx-af: fix hardware timestamp configuration net: dsa: sja1105: always enable the send_meta options net: dsa: tag_sja1105: fix MAC DA patching from meta frames net: Replace strlcpy with strscpy pptp: Fix fib lookup calls. mlxsw: spectrum_router: Fix an IS_ERR() vs NULL check net/sched: act_pedit: Add size check for TCA_PEDIT_PARMS_EX xsk: Honor SO_BINDTODEVICE on bind ptp: Make max_phase_adjustment sysfs device attribute invisible when not supported ...

Merge tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf

2023-07-05T17:57:13Z

Daniel Borkmann says: ==================== pull-request: bpf 2023-07-05 We've added 2 non-merge commits during the last 1 day(s) which contain a total of 3 files changed, 16 insertions(+), 4 deletions(-). The main changes are: 1) Fix BTF to warn but not returning an error for a NULL BTF to still be able to load modules under CONFIG_DEBUG_INFO_BTF, from SeongJae Park. 2) Fix xsk sockets to honor SO_BINDTODEVICE in bind(), from Ilya Maximets. * tag 'for-netdev' of https://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf: xsk: Honor SO_BINDTODEVICE on bind bpf, btf: Warn but return no error for NULL btf from __register_btf_kfunc_id_set() ==================== Link: https://lore.kernel.org/r/20230705171716.6494-1-daniel@iogearbox.net Signed-off-by: Jakub Kicinski

mptcp: do not rely on implicit state check in mptcp_listen()

2023-07-05T09:51:14Z

Since the blamed commit, closing the first subflow resets the first subflow socket state to SS_UNCONNECTED. The current mptcp listen implementation relies only on such state to prevent touching not-fully-disconnected sockets. Incoming mptcp fastclose (or paired endpoint removal) unconditionally closes the first subflow. All the above allows an incoming fastclose followed by a listen() call to successfully race with a blocking recvmsg(), potentially causing the latter to hit a divide by zero bug in cleanup_rbuf/__tcp_select_window(). Address the issue explicitly checking the msk socket state in mptcp_listen(). An alternative solution would be moving the first subflow socket state update into mptcp_disconnect(), but in the long term the first subflow socket should be removed: better avoid relaying on it for internal consistency check. Fixes: b29fcfb54cd7 ("mptcp: full disconnect implementation") Cc: stable@vger.kernel.org Reported-by: Christoph Paasch Closes: https://github.com/multipath-tcp/mptcp_net-next/issues/414 Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts Signed-off-by: Matthieu Baerts Signed-off-by: David S. Miller

mptcp: ensure subflow is unhashed before cleaning the backlog

2023-07-05T09:51:14Z

While tacking care of the mptcp-level listener I unintentionally moved the subflow level unhash after the subflow listener backlog cleanup. That could cause some nasty race and makes the code harder to read. Address the issue restoring the proper order of operations. Fixes: 57fc0f1ceaa4 ("mptcp: ensure listener is unhashed before updating the sk status") Cc: stable@vger.kernel.org Signed-off-by: Paolo Abeni Reviewed-by: Matthieu Baerts Signed-off-by: Matthieu Baerts Signed-off-by: David S. Miller

net: dsa: sja1105: always enable the send_meta options

2023-07-04T18:42:27Z

incl_srcpt has the limitation, mentioned in commit b4638af8885a ("net: dsa: sja1105: always enable the INCL_SRCPT option"), that frames with a MAC DA of 01:80:c2:xx:yy:zz will be received as 01:80:c2:00:00:zz unless PTP RX timestamping is enabled. The incl_srcpt option was initially unconditionally enabled, then that changed with commit 42824463d38d ("net: dsa: sja1105: Limit use of incl_srcpt to bridge+vlan mode"), then again with b4638af8885a ("net: dsa: sja1105: always enable the INCL_SRCPT option"). Bottom line is that it now needs to be always enabled, otherwise the driver does not have a reliable source of information regarding source_port and switch_id for link-local traffic (tag_8021q VLANs may be imprecise since now they identify an entire bridging domain when ports are not standalone). If we accept that PTP RX timestamping (and therefore, meta frame generation) is always enabled in hardware, then that limitation could be avoided and packets with any MAC DA can be properly received, because meta frames do contain the original bytes from the MAC DA of their associated link-local packet. This change enables meta frame generation unconditionally, which also has the nice side effects of simplifying the switch control path (a switch reset is no longer required on hwtstamping settings change) and the tagger data path (it no longer needs to be informed whether to expect meta frames or not - it always does). Fixes: 227d07a07ef1 ("net: dsa: sja1105: Add support for traffic through standalone ports") Signed-off-by: Vladimir Oltean Reviewed-by: Simon Horman Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller

net: dsa: tag_sja1105: fix MAC DA patching from meta frames

2023-07-04T18:42:27Z

The SJA1105 manual says that at offset 4 into the meta frame payload we have "MAC destination byte 2" and at offset 5 we have "MAC destination byte 1". These are counted from the LSB, so byte 1 is h_dest[ETH_HLEN-2] aka h_dest[4] and byte 2 is h_dest[ETH_HLEN-3] aka h_dest[3]. The sja1105_meta_unpack() function decodes these the other way around, so a frame with MAC DA 01:80:c2:11:22:33 is received by the network stack as having 01:80:c2:22:11:33. Fixes: e53e18a6fe4d ("net: dsa: sja1105: Receive and decode meta frames") Signed-off-by: Vladimir Oltean Reviewed-by: Simon Horman Reviewed-by: Florian Fainelli Signed-off-by: David S. Miller

net/sched: act_pedit: Add size check for TCA_PEDIT_PARMS_EX

2023-07-04T08:31:38Z

The attribute TCA_PEDIT_PARMS_EX is not be included in pedit_policy and one malicious user could fake a TCA_PEDIT_PARMS_EX whose length is smaller than the intended sizeof(struct tc_pedit). Hence, the dereference in tcf_pedit_init() could access dirty heap data. static int tcf_pedit_init(...) { // ... pattr = tb[TCA_PEDIT_PARMS]; // TCA_PEDIT_PARMS is included if (!pattr) pattr = tb[TCA_PEDIT_PARMS_EX]; // but this is not // ... parm = nla_data(pattr); index = parm->index; // parm is able to be smaller than 4 bytes // and this dereference gets dirty skb_buff // data created in netlink_sendmsg } This commit adds TCA_PEDIT_PARMS_EX length in pedit_policy which avoid the above case, just like the TCA_PEDIT_PARMS. Fixes: 71d0ed7079df ("net/act_pedit: Support using offset relative to the conventional network headers") Signed-off-by: Lin Ma Reviewed-by: Pedro Tammela Link: https://lore.kernel.org/r/20230703110842.590282-1-linma@zju.edu.cn Signed-off-by: Paolo Abeni

xsk: Honor SO_BINDTODEVICE on bind

2023-07-04T08:19:48Z

Initial creation of an AF_XDP socket requires CAP_NET_RAW capability. A privileged process might create the socket and pass it to a non-privileged process for later use. However, that process will be able to bind the socket to any network interface. Even though it will not be able to receive any traffic without modification of the BPF map, the situation is not ideal. Sockets already have a mechanism that can be used to restrict what interface they can be attached to. That is SO_BINDTODEVICE. To change the SO_BINDTODEVICE binding the process will need CAP_NET_RAW. Make xsk_bind() honor the SO_BINDTODEVICE in order to allow safer workflow when non-privileged process is using AF_XDP. The intended workflow is following: 1. First process creates a bare socket with socket(AF_XDP, ...). 2. First process loads the XSK program to the interface. 3. First process adds the socket fd to a BPF map. 4. First process ties socket fd to a particular interface using SO_BINDTODEVICE. 5. First process sends socket fd to a second process. 6. Second process allocates UMEM. 7. Second process binds socket to the interface with bind(...). 8. Second process sends/receives the traffic. All the steps above are possible today if the first process is privileged and the second one has sufficient RLIMIT_MEMLOCK and no capabilities. However, the second process will be able to bind the socket to any interface it wants on step 7 and send traffic from it. With the proposed change, the second process will be able to bind the socket only to a specific interface chosen by the first process at step 4. Fixes: 965a99098443 ("xsk: add support for bind for Rx") Signed-off-by: Ilya Maximets Signed-off-by: Daniel Borkmann Acked-by: Magnus Karlsson Acked-by: John Fastabend Acked-by: Jason Wang Link: https://lore.kernel.org/bpf/20230703175329.3259672-1-i.maximets@ovn.org

Merge tag 'leds-next-6.5' of git://git.kernel.org/pub/scm/linux/kernel/git/lee/leds

2023-07-03T18:26:05Z

Pull LED updates from Lee Jones: "New Drivers: - Add support for Intel Cherry Trail Whiskey Cove PMIC LEDs - Add support for Awinic AW20036/AW20054/AW20072 LEDs New Device Support: - Add support for PMI632 LPG to QCom LPG - Add support for PMI8998 to QCom Flash - Add support for MT6331, WLEDs and MT6332 to Mediatek MT6323 PMIC New Functionality: - Implement the LP55xx Charge Pump - Add support for suspend / resume to Intel Cherry Trail Whiskey Cove PMIC - Add support for breathing mode to Intel Cherry Trail Whiskey Cove PMIC - Enable per-pin resolution Pinctrl in LEDs GPIO Fix-ups: - Allow thread to sleep by switching from spinlock to mutex - Add lots of Device Tree bindings / support - Adapt relationships / dependencies driven by Kconfig - Switch I2C drivers from .probe_new() to .probe() - Remove superfluous / duplicate code - Replace strlcpy() with strscpy() for efficiency and overflow prevention - Staticify various functions - Trivial: Fixing coding style - Simplify / reduce code Bug Fixes: - Prevent NETDEV_LED_MODE_LINKUP from being cleared on rename - Repair race between led_set_brightness(LED_{OFF,FULL}) - Fix Oops relating to sleeping in critical sections - Clear LED_INIT_DEFAULT_TRIGGER flag when clearing the current trigger - Do not leak resources in error handling paths - Fix unsigned comparison which can never be negative - Provide missing NULL terminating entries in tables - Fix misnaming issues" * tag 'leds-next-6.5' of git://git.kernel.org/pub/scm/linux/kernel/git/lee/leds: (53 commits) leds: leds-mt6323: Adjust return/parameter types in wled get/set callbacks leds: sgm3140: Add richtek,rt5033-led compatible dt-bindings: leds: sgm3140: Document richtek,rt5033 compatible dt-bindings: backlight: kinetic,ktz8866: Add missing type for "current-num-sinks" dt-bindings: leds: Drop unneeded quotes leds: Fix config reference for AW200xx driver leds: leds-mt6323: Add support for WLEDs and MT6332 leds: leds-mt6323: Add support for MT6331 leds leds: leds-mt6323: Open code and drop MT6323_CAL_HW_DUTY macro leds: leds-mt6323: Drop MT6323_ prefix from macros and defines leds: leds-mt6323: Specify registers and specs in platform data dt-bindings: leds: leds-mt6323: Document mt6332 compatible dt-bindings: leds: leds-mt6323: Document mt6331 compatible leds: simatic-ipc-leds-gpio: Introduce more Kconfig switches leds: simatic-ipc-leds-gpio: Split up into multiple drivers leds: simatic-ipc-leds-gpio: Move two extra gpio pins into another table leds: simatic-ipc-leds-gpio: Add terminating entries to gpio tables leds: flash: leds-qcom-flash: Fix an unsigned comparison which can never be negative leds: cht-wcove: Remove unneeded semicolon leds: cht-wcove: Fix an unsigned comparison which can never be negative ...

tcp: annotate data races in __tcp_oow_rate_limited()

2023-07-03T08:25:02Z

request sockets are lockless, __tcp_oow_rate_limited() could be called on the same object from different cpus. This is harmless. Add READ_ONCE()/WRITE_ONCE() annotations to avoid a KCSAN report. Fixes: 4ce7e93cb3fe ("tcp: rate limit ACK sent by SYN_RECV request sockets") Signed-off-by: Eric Dumazet Signed-off-by: David S. Miller