Package "linux-bluefield-headers-5.4.0-1095"

  linux-bluefield (5.4.0-1090.97) focal; urgency=medium

  * focal/linux-bluefield: 5.4.0-1090.97 -proposed tracker (LP: #2072264)

  [ Ubuntu: 5.4.0-192.212 ]

  * focal/linux: 5.4.0-192.212 -proposed tracker (LP: #2072305)
  * Focal update: v5.4.278 upstream stable release (LP: #2071668)
    - x86/tsc: Trust initial offset in architectural TSC-adjust MSRs
    - speakup: Fix sizeof() vs ARRAY_SIZE() bug
    - ring-buffer: Fix a race between readers and resize checks
    - net: smc91x: Fix m68k kernel compilation for ColdFire CPU
    - nilfs2: fix unexpected freezing of nilfs_segctor_sync()
    - nilfs2: fix potential hang in nilfs_detach_log_writer()
    - wifi: cfg80211: fix the order of arguments for trace events of the tx_rx_evt
      class
    - net: usb: qmi_wwan: add Telit FN920C04 compositions
    - drm/amd/display: Set color_mgmt_changed to true on unsuspend
    - ASoC: rt5645: Fix the electric noise due to the CBJ contacts floating
    - ASoC: dt-bindings: rt5645: add cbj sleeve gpio property
    - ASoC: da7219-aad: fix usage of device_get_named_child_node()
    - drm/amdkfd: Flush the process wq before creating a kfd_process
    - nvme: find numa distance only if controller has valid numa id
    - openpromfs: finish conversion to the new mount API
    - crypto: bcm - Fix pointer arithmetic
    - firmware: raspberrypi: Use correct device for DMA mappings
    - ecryptfs: Fix buffer size for tag 66 packet
    - nilfs2: fix out-of-range warning
    - parisc: add missing export of __cmpxchg_u8()
    - crypto: ccp - drop platform ifdef checks
    - s390/cio: fix tracepoint subchannel type field
    - jffs2: prevent xattr node from overflowing the eraseblock
    - null_blk: Fix missing mutex_destroy() at module removal
    - md: fix resync softlockup when bitmap size is less than array size
    - wifi: ath10k: poll service ready message before failing
    - x86/boot: Ignore relocations in .notes sections in walk_relocs() too
    - qed: avoid truncating work queue length
    - scsi: ufs: qcom: Perform read back after writing reset bit
    - scsi: ufs: cdns-pltfrm: Perform read back after writing HCLKDIV
    - scsi: ufs: core: Perform read back after disabling interrupts
    - scsi: ufs: core: Perform read back after disabling UIC_COMMAND_COMPL
    - irqchip/alpine-msi: Fix off-by-one in allocation error path
    - ACPI: disable -Wstringop-truncation
    - cpufreq: Reorganize checks in cpufreq_offline()
    - cpufreq: Split cpufreq_offline()
    - cpufreq: Rearrange locking in cpufreq_remove_dev()
    - cpufreq: exit() callback is optional
    - scsi: libsas: Fix the failure of adding phy with zero-address to port
    - scsi: hpsa: Fix allocation size for Scsi_Host private data
    - x86/purgatory: Switch to the position-independent small code model
    - wifi: ath10k: Fix an error code problem in
      ath10k_dbg_sta_write_peer_debug_trigger()
    - wifi: ath10k: populate board data for WCN3990
    - tcp: minor optimization in tcp_add_backlog()
    - tcp: fix a signed-integer-overflow bug in tcp_add_backlog()
    - tcp: avoid premature drops in tcp_add_backlog()
    - macintosh/via-macii: Fix "BUG: sleeping function called from invalid
      context"
    - wifi: carl9170: add a proper sanity check for endpoints
    - wifi: ar5523: enable proper endpoint verification
    - sh: kprobes: Merge arch_copy_kprobe() into arch_prepare_kprobe()
    - Revert "sh: Handle calling csum_partial with misaligned data"
    - HID: intel-ish-hid: ipc: Add check for pci_alloc_irq_vectors
    - scsi: bfa: Ensure the copied buf is NUL terminated
    - scsi: qedf: Ensure the copied buf is NUL terminated
    - wifi: mwl8k: initialize cmd->addr[] properly
    - usb: aqc111: stop lying about skb->truesize
    - net: usb: sr9700: stop lying about skb->truesize
    - m68k: Fix spinlock race in kernel thread creation
    - m68k: mac: Fix reboot hang on Mac IIci
    - net: ethernet: cortina: Locking fixes
    - af_unix: Fix data races in unix_release_sock/unix_stream_sendmsg
    - net: usb: smsc95xx: stop lying about skb->truesize
    - net: openvswitch: fix overwriting ct original tuple for ICMPv6
    - ipv6: sr: add missing seg6_local_exit
    - ipv6: sr: fix incorrect unregister order
    - ipv6: sr: fix invalid unregister error path
    - drm/amd/display: Fix potential index out of bounds in color transformation
      function
    - mtd: rawnand: hynix: fixed typo
    - fbdev: shmobile: fix snprintf truncation
    - drm/mediatek: Add 0 size check to mtk_drm_gem_obj
    - powerpc/fsl-soc: hide unused const variable
    - fbdev: sisfb: hide unused variables
    - media: ngene: Add dvb_ca_en50221_init return value check
    - media: radio-shark2: Avoid led_names truncations
    - platform/x86: wmi: Make two functions static
    - fbdev: sh7760fb: allow modular build
    - drm/arm/malidp: fix a possible null pointer dereference
    - ASoC: tracing: Export SND_SOC_DAPM_DIR_OUT to its value
    - drm/panel: simple: Add missing Innolux G121X1-L03 format, flags, connector
    - RDMA/hns: Use complete parentheses in macros
    - x86/insn: Fix PUSH instruction in x86 instruction decoder opcode map
    - ext4: avoid excessive credit estimate in ext4_tmpfile()
    - sunrpc: removed redundant procp check
    - SUNRPC: Fix gss_free_in_token_pages()
    - selftests/kcmp: Make the test output consistent and clear
    - selftests/kcmp: remove unused open mode
    - RDMA/IPoIB: Fix format truncation compilation errors
    - netrom: fix possible dead-lock in nr_rt_ioctl()
    - af_packet: do not call packet_read_pending() from tpacket_destruct_skb()
    - sched/topology: Don't set SD_BALANCE_WAKE on cpuset domain relax
    - sched/fair: Allow disabling sched_balance_newidle with
      sched_relax_domain_level
    - greybus: lights: check return of get_channel_from_mode
    - soundwire: cadence/intel: simplify PDI/port mapping
    - soundwire: intel: don't filter out PDI0/1
    - soundwire: cadence_master: improve PDI

  linux-bluefield (5.4.0-1089.96) focal; urgency=medium

  * focal/linux-bluefield: 5.4.0-1089.96 -proposed tracker (LP: #2072067)

  [ Ubuntu: 5.4.0-190.210 ]

  * focal/linux: 5.4.0-190.210 -proposed tracker (LP: #2072108)
  * CVE-2024-36016
    - tty: n_gsm: fix possible out-of-bounds in gsm0_receive()
  * CVE-2022-48655
    - firmware: arm_scmi: Harden accesses to the reset domains
  * CVE-2024-26907
    - RDMA/mlx5: Fix fortify source warning while accessing Eth segment
  * CVE-2024-26585
    - tls: fix race between tx work scheduling and socket close
  * CVE-2024-26584
    - net: tls: handle backlogging of crypto requests
  * CVE-2024-26583
    - net/tls: Replace TLS_RX_SYNC_RUNNING with RCU
    - net/tls: Fix use-after-free after the TLS device goes down and up
    - tls: splice_read: fix record type check
    - tls splice: remove inappropriate flags checking for MSG_PEEK
    - tls: splice_read: fix accessing pre-processed records
    - tls: Fix context leak on tls_device_down
    - net/tls: Check for errors in tls_device_init
    - net/tls: Remove the context from the list in tls_device_down
    - net/tls: pass context to tls_device_decrypted()
    - net/tls: Perform immediate device ctx cleanup when possible
    - net/tls: Multi-threaded calls to TX tls_dev_del
    - net: tls: avoid discarding data on record close
    - tls: rx: don't store the record type in socket context
    - tls: rx: don't store the decryption status in socket context
    - tls: rx: don't issue wake ups when data is decrypted
    - tls: rx: refactor decrypt_skb_update()
    - tls: hw: rx: use return value of tls_device_decrypted() to carry status
    - tls: rx: drop unnecessary arguments from tls_setup_from_iter()
    - tls: rx: don't report text length from the bowels of decrypt
    - tls: rx: wrap decryption arguments in a structure
    - tls: rx: factor out writing ContentType to cmsg
    - tls: rx: don't track the async count
    - tls: rx: assume crypto always calls our callback
    - tls: rx: use async as an in-out argument
    - tls: decrement decrypt_pending if no async completion will be called
    - net: tls: fix async vs NIC crypto offload
    - tls: rx: simplify async wait
    - tls: extract context alloc/initialization out of tls_set_sw_offload
    - net: tls: factor out tls_*crypt_async_wait()
    - tls: fix race between async notify and socket close

 -- Bartlomiej Zolnierkiewicz <email address hidden> Wed, 10 Jul 2024 09:37:57 +0200

  linux-bluefield (5.4.0-1088.95) focal; urgency=medium

  * focal/linux-bluefield: 5.4.0-1088.95 -proposed tracker (LP: #2068404)

  [ Ubuntu: 5.4.0-189.209 ]

  * focal/linux: 5.4.0-189.209 -proposed tracker (LP: #2068454)
  * Focal update: v5.4.275 upstream stable release (LP: #2067865)
    - batman-adv: Avoid infinite loop trying to resize local TT
    - Bluetooth: Fix memory leak in hci_req_sync_complete()
    - nouveau: fix function cast warning
    - net: openvswitch: fix unwanted error log on timeout policy probing
    - u64_stats: fix u64_stats_init() for lockdep when used repeatedly in one file
    - geneve: fix header validation in geneve[6]_xmit_skb
    - ipv6: fib: hide unused 'pn' variable
    - ipv4/route: avoid unused-but-set-variable warning
    - ipv6: fix race condition between ipv6_get_ifaddr and ipv6_del_addr
    - net/mlx5: Properly link new fs rules into the tree
    - net: ena: Fix potential sign extension issue
    - btrfs: qgroup: correctly model root qgroup rsv in convert
    - drm/client: Fully protect modes[] with dev->mode_config.mutex
    - vhost: Add smp_rmb() in vhost_vq_avail_empty()
    - selftests: timers: Fix abs() warning in posix_timers test
    - x86/apic: Force native_apic_mem_read() to use the MOV instruction
    - btrfs: record delayed inode root in transaction
    - selftests/ftrace: Limit length in subsystem-enable tests
    - kprobes: Fix possible use-after-free issue on kprobe registration
    - Revert "tracing/trigger: Fix to return error if failed to alloc snapshot"
    - netfilter: nf_tables: Fix potential data-race in __nft_expr_type_get()
    - tun: limit printing rate when illegal packet received by tun dev
    - RDMA/rxe: Fix the problem "mutex_destroy missing"
    - RDMA/mlx5: Fix port number for counter query in multi-port configuration
    - drm: nv04: Fix out of bounds access
    - clk: Remove prepare_lock hold assertion in __clk_release()
    - clk: Mark 'all_lists' as const
    - clk: remove extra empty line
    - clk: Print an info line before disabling unused clocks
    - clk: Initialize struct clk_core kref earlier
    - clk: Get runtime PM before walking tree during disable_unused
    - x86/cpufeatures: Fix dependencies for GFNI, VAES, and VPCLMULQDQ
    - comedi: vmk80xx: fix incomplete endpoint checking
    - serial/pmac_zilog: Remove flawed mitigation for rx irq flood
    - USB: serial: option: add Fibocom FM135-GL variants
    - USB: serial: option: add support for Fibocom FM650/FG650
    - USB: serial: option: add Lonsung U8300/U9300 product
    - USB: serial: option: support Quectel EM060K sub-models
    - USB: serial: option: add Rolling RW101-GL and RW135-GL support
    - USB: serial: option: add Telit FN920C04 rmnet compositions
    - usb: dwc2: host: Fix dereference issue in DDMA completion flow.
    - speakup: Avoid crash on very long word
    - fs: sysfs: Fix reference leak in sysfs_break_active_protection()
    - nouveau: fix instmem race condition around ptr stores
    - nilfs2: fix OOB in nilfs_set_de_type
    - KVM: async_pf: Cleanup kvm_setup_async_pf()
    - arm64: dts: rockchip: fix alphabetical ordering RK3399 puma
    - arm64: dts: rockchip: enable internal pull-up on PCIE_WAKE# for RK3399 Puma
    - arm64: dts: mediatek: mt7622: fix IR nodename
    - arm64: dts: mediatek: mt7622: fix ethernet controller "compatible"
    - arm64: dts: mediatek: mt7622: drop "reset-names" from thermal block
    - arm64: dts: mt2712: add ethernet device node
    - arm64: dts: mediatek: mt2712: fix validation errors
    - ARC: [plat-hsdk]: Remove misplaced interrupt-cells property
    - vxlan: drop packets from invalid src-address
    - mlxsw: core: Unregister EMAD trap using FORWARD action
    - NFC: trf7970a: disable all regulators on removal
    - net: usb: ax88179_178a: stop lying about skb->truesize
    - net: gtp: Fix Use-After-Free in gtp_dellink
    - ipvs: Fix checksumming on GSO of SCTP packets
    - net: openvswitch: Fix Use-After-Free in ovs_ct_exit
    - mlxsw: spectrum_acl_tcam: Fix race during rehash delayed work
    - mlxsw: spectrum_acl_tcam: Fix possible use-after-free during activity update
    - mlxsw: spectrum_acl_tcam: Fix possible use-after-free during rehash
    - mlxsw: spectrum_acl_tcam: Rate limit error message
    - mlxsw: spectrum_acl_tcam: Fix memory leak during rehash
    - mlxsw: spectrum_acl_tcam: Fix warning during rehash
    - mlxsw: spectrum_acl_tcam: Fix incorrect list API usage
    - mlxsw: spectrum_acl_tcam: Fix memory leak when canceling rehash work
    - i40e: Do not use WQ_MEM_RECLAIM flag for workqueue
    - iavf: Fix TC config comparison with existing adapter TC config
    - af_unix: Suppress false-positive lockdep splat for spin_lock() in
      __unix_gc().
    - serial: core: Provide port lock wrappers
    - serial: mxs-auart: add spinlock around changing cts state
    - Revert "crypto: api - Disallow identical driver names"
    - net/mlx5e: Fix a race in command alloc flow
    - tracing: Show size of requested perf buffer
    - tracing: Increase PERF_MAX_TRACE_SIZE to handle Sentinel1 and docker
      together
    - Bluetooth: Fix type of len in {l2cap,sco}_sock_getsockopt_old()
    - Bluetooth: btusb: Add Realtek RTL8852BE support ID 0x0bda:0x4853
    - btrfs: fix information leak in btrfs_ioctl_logical_to_ino()
    - arm64: dts: rockchip: enable internal pull-up for Q7_THRM# on RK3399 Puma
    - drm/amdgpu: Fix leak when GPU memory allocation fails
    - irqchip/gic-v3-its: Prevent double free on error
    - ethernet: Add helper for assigning packet type when dest address does not
      match device address
    - net: b44: set pause params only when interface is up
    - stackdepot: respect __GFP_NOLOCKDEP allocation flag
    - mtd: diskonchip: work around ubsan link failure
    - tcp: Clean up kernel listener's reqsk in inet_twsk_purge()
    - tcp: Fix NEW_SYN_RECV handling in inet_twsk_purge()
    - dmaengine: owl: fix register access functions
    - idma64: Don't try

  linux-bluefield (5.4.0-1087.94) focal; urgency=medium

  * focal/linux-bluefield: 5.4.0-1087.94 -proposed tracker (LP: #2068250)

  [ Ubuntu: 5.4.0-187.207 ]

  * focal/linux: 5.4.0-187.207 -proposed tracker (LP: #2068291)
  * CVE-2024-26925
    - netfilter: Cleanup nft_net->module_list from nf_tables_exit_net()
    - netfilter: nf_tables: release batch on table validation from abort path
    - netfilter: nf_tables: release mutex after nft_gc_seq_end from abort path
  * CVE-2024-26643
    - netfilter: nf_tables: mark set as dead when unbinding anonymous set with
      timeout
  * CVE-2024-2201
    - x86/cpufeatures: Add new word for scattered features
    - x86/cpufeatures: Add CPUID_LNX_5 to track recently added Linux-defined word
    - x86/bugs: Change commas to semicolons in 'spectre_v2' sysfs file
    - x86/bhi: Add support for clearing branch history at syscall entry
    - x86/bhi: Define SPEC_CTRL_BHI_DIS_S
    - x86/bhi: Enumerate Branch History Injection (BHI) bug
    - x86/bhi: Add BHI mitigation knob
    - x86/bhi: Mitigate KVM by default
    - [Config] updateconfigs for CONFIG_BHI_{AUTO|ON|OFF}
    - x86/bugs: Fix BHI documentation
    - x86/bugs: Cache the value of MSR_IA32_ARCH_CAPABILITIES
    - x86/bugs: Rename various 'ia32_cap' variables to 'x86_arch_cap_msr'
    - x86/bugs: Fix BHI handling of RRSBA
    - x86/bugs: Clarify that syscall hardening isn't a BHI mitigation
    - x86/bugs: Fix BHI retpoline check

 -- Bartlomiej Zolnierkiewicz <email address hidden> Wed, 12 Jun 2024 15:52:23 +0200

  linux-bluefield (5.4.0-1086.93) focal; urgency=medium

  * focal/linux-bluefield: 5.4.0-1086.93 -proposed tracker (LP: #2063770)

  * mlxbf-gige: Vitesse PHY stuck in a bad state during reboot test
    (LP: #2064163)
    - SAUCE: mlxbf-gige: OOB PHY stuck in a bad state during reboot test

  * mlxbf-gige: autonegotiation fails to complete on BF2 (LP: #2062384)
    - SAUCE: mlxbf-gige: autonegotiation fails to complete on BF2

  [ Ubuntu: 5.4.0-186.206 ]

  * focal/linux: 5.4.0-186.206 -proposed tracker (LP: #2063812)
  * Mount CIFS fails with Permission denied (LP: #2061986)
    - cifs: fix ntlmssp auth when there is no key exchange
  * USB stick can't be detected (LP: #2040948)
    - usb: Disable USB3 LPM at shutdown
  * CVE-2024-26733
    - net: dev: Convert sa_data to flexible array in struct sockaddr
    - arp: Prevent overflow in arp_req_get().
    - stddef: Introduce DECLARE_FLEX_ARRAY() helper
  * CVE-2024-26712
    - powerpc/kasan: Fix addr error caused by page alignment
  * CVE-2023-52530
    - wifi: mac80211: fix potential key use-after-free
  * CVE-2021-47063
    - drm: bridge/panel: Cleanup connector on bridge detach
  * [Ubuntu 22.04.4/linux-image-6.5.0-26-generic] Kernel output "UBSAN: array-
    index-out-of-bounds in /build/linux-hwe-6.5-34pCLi/linux-
    hwe-6.5-6.5.0/drivers/net/hyperv/netvsc.c:1445:41" multiple times,
    especially during boot. (LP: #2058477)
    - hv: hyperv.h: Replace one-element array with flexible-array member
  * CVE-2024-26614
    - tcp: make sure init the accept_queue's spinlocks once
    - ipv6: init the accept_queue's spinlocks in inet6_create
  * Focal update: v5.4.271 upstream stable release (LP: #2060216)
    - netlink: Fix kernel-infoleak-after-free in __skb_datagram_iter
    - net: ip_tunnel: prevent perpetual headroom growth
    - tun: Fix xdp_rxq_info's queue_index when detaching
    - ipv6: fix potential "struct net" leak in inet6_rtm_getaddr()
    - lan78xx: enable auto speed configuration for LAN7850 if no EEPROM is
      detected
    - net: usb: dm9601: fix wrong return value in dm9601_mdio_read
    - Bluetooth: Avoid potential use-after-free in hci_error_reset
    - Bluetooth: hci_event: Fix handling of HCI_EV_IO_CAPA_REQUEST
    - Bluetooth: Enforce validation on max value of connection interval
    - netfilter: nf_tables: allow NFPROTO_INET in nft_(match/target)_validate()
    - rtnetlink: fix error logic of IFLA_BRIDGE_FLAGS writing back
    - efi/capsule-loader: fix incorrect allocation size
    - power: supply: bq27xxx-i2c: Do not free non existing IRQ
    - ALSA: Drop leftover snd-rtctimer stuff from Makefile
    - afs: Fix endless loop in directory parsing
    - gtp: fix use-after-free and null-ptr-deref in gtp_newlink()
    - wifi: nl80211: reject iftype change with mesh ID change
    - btrfs: dev-replace: properly validate device names
    - dmaengine: fsl-qdma: fix SoC may hang on 16 byte unaligned read
    - dmaengine: fsl-qdma: init irq after reg initialization
    - mmc: core: Fix eMMC initialization with 1-bit bus connection
    - x86/cpu/intel: Detect TME keyid bits before setting MTRR mask registers
    - cachefiles: fix memory leak in cachefiles_add_cache()
    - fs,hugetlb: fix NULL pointer dereference in hugetlbs_fill_super
    - gpio: 74x164: Enable output pins after registers are reset
    - Linux 5.4.271
  * Focal update: v5.4.270 upstream stable release (LP: #2060019)
    - KVM: arm64: vgic-its: Test for valid IRQ in its_sync_lpi_pending_table()
    - KVM: arm64: vgic-its: Test for valid IRQ in MOVALL handler
    - net/sched: Retire CBQ qdisc
    - [Config] updateconfigs for NET_SCH_CBQ
    - net/sched: Retire ATM qdisc
    - [Config] updateconfigs for NET_SCH_ATM
    - net/sched: Retire dsmark qdisc
    - [Config] updateconfigs for NET_SCH_DSMARK
    - sched/rt: sysctl_sched_rr_timeslice show default timeslice after reset
    - memcg: add refcnt for pcpu stock to avoid UAF problem in drain_all_stock()
    - nilfs2: replace WARN_ONs for invalid DAT metadata block requests
    - userfaultfd: fix mmap_changing checking in mfill_atomic_hugetlb
    - sched/rt: Fix sysctl_sched_rr_timeslice intial value
    - sched/rt: Disallow writing invalid values to sched_rt_period_us
    - scsi: target: core: Add TMF to tmr_list handling
    - dmaengine: shdma: increase size of 'dev_id'
    - dmaengine: fsl-qdma: increase size of 'irq_name'
    - wifi: cfg80211: fix missing interfaces when dumping
    - wifi: mac80211: fix race condition on enabling fast-xmit
    - fbdev: savage: Error out if pixclock equals zero
    - fbdev: sis: Error out if pixclock equals zero
    - ahci: asm1166: correct count of reported ports
    - ahci: add 43-bit DMA address quirk for ASMedia ASM1061 controllers
    - ext4: avoid allocating blocks from corrupted group in
      ext4_mb_try_best_found()
    - ext4: avoid allocating blocks from corrupted group in ext4_mb_find_by_goal()
    - regulator: pwm-regulator: Add validity checks in continuous .get_voltage
    - nvmet-tcp: fix nvme tcp ida memory leak
    - ASoC: sunxi: sun4i-spdif: Add support for Allwinner H616
    - netfilter: conntrack: check SCTP_CID_SHUTDOWN_ACK for vtag setting in
      sctp_new
    - nvmet-fc: abort command when there is no binding
    - hwmon: (coretemp) Enlarge per package core count limit
    - scsi: lpfc: Use unsigned type for num_sge
    - firewire: core: send bus reset promptly on gap count error
    - virtio-blk: Ensure no requests in virtqueues before deleting vqs.
    - s390/qeth: Fix potential loss of L3-IP@ in case of network issues
    - pmdomain: renesas: r8a77980-sysc: CR7 must be always on
    - tcp: factor out __tcp_close() helper
    - tcp: return EPOLLOUT from tcp_poll only when notsent_bytes is half the limit
    - tcp: add annotations around sk->sk_shutdown accesses
    - pinctrl: pinctrl-rockchip: Fix a bunch of kerneldoc misdemeanours
    - pinctrl: rockchip: Fix refcount leak in rockchip_pinctrl_parse_groups
    - spi: mt7621: Fix an error mes