This bug was fixed in the package linux - 5.4.0-110.124
---------------
linux (5.4.0-110.124) focal; urgency=medium
* focal/linux: 5.4.0-110.124 -proposed tracker (LP: #1969053)
* net/mlx5e: Fix page DMA map/unmap attributes (LP: #1967292)
- net/mlx5e: Fix page DMA map/unmap attributes
* xfs: Fix deadlock between AGI and AGF when target_ip exists in xfs_rename()
(LP: #1966803)
- xfs: Fix deadlock between AGI and AGF when target_ip exists in
xfs_rename()
* LRMv6: add multi-architecture support (LP: #1968774)
- [Packaging] resync dkms-build{,--nvidia-N}
* xfrm interface cannot be changed anymore (LP: #1968591)
- xfrm: fix the if_id check in changelink
* Use kernel-testing repo from launchpad for ADT tests (LP: #1968016)
- [Debian] Use kernel-testing repo from launchpad
* vmx_ldtr_test in ubuntu_kvm_unit_tests failed (FAIL: Expected 0 for L1 LDTR
selector (got 50)) (LP: #1956315)
- KVM: nVMX: Set LDTR to its architecturally defined value on nested VM-Exit
* [SRU][Regression] Revert "PM: ACPI: reboot: Use S5 for reboot" which causes
Bus Fatal Error when rebooting system with BCM5720 NIC (LP: #1917471)
- Revert "PM: ACPI: reboot: Use S5 for reboot"
* Focal update: v5.4.181 upstream stable release (LP: #1967582)
- Makefile.extrawarn: Move -Wunaligned-access to W=1
- HID:Add support for UGTABLET WP5540
- Revert "svm: Add warning message for AVIC IPI invalid target"
- serial: parisc: GSC: fix build when IOSAPIC is not set
- parisc: Drop __init from map_pages declaration
- parisc: Fix data TLB miss in sba_unmap_sg
- parisc: Fix sglist access in ccio-dma.c
- btrfs: send: in case of IO error log it
- platform/x86: ISST: Fix possible circular locking dependency detected
- selftests: rtc: Increase test timeout so that all tests run
- net: ieee802154: at86rf230: Stop leaking skb's
- selftests/zram: Skip max_comp_streams interface on newer kernel
- selftests/zram01.sh: Fix compression ratio calculation
- selftests/zram: Adapt the situation that /dev/zram0 is being used
- ax25: improve the incomplete fix to avoid UAF and NPD bugs
- vfs: make freeze_super abort when sync_filesystem returns error
- quota: make dquot_quota_sync return errors from ->sync_fs
- nvme: fix a possible use-after-free in controller reset during load
- nvme-tcp: fix possible use-after-free in transport error_recovery work
- nvme-rdma: fix possible use-after-free in transport error_recovery work
- drm/amdgpu: fix logic inversion in check
- Revert "module, async: async_synchronize_full() on module init iff async
is
used"
- ftrace: add ftrace_init_nop()
- module/ftrace: handle patchable-function-entry
- arm64: module: rework special section handling
- arm64: module/ftrace: intialize PLT at load time
- iwlwifi: fix use-after-free
- drm/radeon: Fix backlight control on iMac 12,1
- ext4: check for out-of-order index extents in ext4_valid_extent_entries()
- ext4: check for inconsistent extents between index and leaf block
- ext4: prevent partial update of the extent blocks
- taskstats: Cleanup the use of task->exit_code
- dmaengine: at_xdmac: Start transfer for cyclic channels in issue_pending
- vsock: remove vsock from connected table when connect is interrupted by a
signal
- mmc: block: fix read single on recovery logic
- iwlwifi: pcie: fix locking when "HW not ready"
- iwlwifi: pcie: gen2: fix locking when "HW not ready"
- netfilter: nft_synproxy: unregister hooks on init error path
- net: dsa: lan9303: fix reset on probe
- net: ieee802154: ca8210: Fix lifs/sifs periods
- ping: fix the dif and sdif check in ping_lookup
- bonding: force carrier update when releasing slave
- drop_monitor: fix data-race in dropmon_net_event / trace_napi_poll_hit
- bonding: fix data-races around agg_select_timer
- libsubcmd: Fix use-after-free for realloc(..., 0)
- ALSA: hda: Fix regression on forced probe mask option
- ALSA: hda: Fix missing codec probe on Shenker Dock 15
- ASoC: ops: Fix stereo change notifications in snd_soc_put_volsw()
- ASoC: ops: Fix stereo change notifications in snd_soc_put_volsw_range()
- powerpc/lib/sstep: fix 'ptesync' build error
- mtd: rawnand: gpmi: don't leak PM reference in error path
- block/wbt: fix negative inflight counter when remove scsi device
- NFS: LOOKUP_DIRECTORY is also ok with symlinks
- NFS: Do not report writeback errors in nfs_getattr()
- mtd: rawnand: qcom: Fix clock sequencing in qcom_nandc_probe()
- mtd: rawnand: brcmnand: Fixed incorrect sub-page ECC status
- scsi: lpfc: Fix pt2pt NVMe PRLI reject LOGO loop
- EDAC: Fix calculation of returned address and next offset in
edac_align_ptr()
- net: sched: limit TC_ACT_REPEAT loops
- dmaengine: sh: rcar-dmac: Check for error num after setting mask
- copy_process(): Move fd_install() out of sighand->siglock critical section
- i2c: brcmstb: fix support for DSL and CM variants
- Drivers: hv: vmbus: Fix memory leak in vmbus_add_channel_kobj
- KVM: x86/pmu: Use AMD64_RAW_EVENT_MASK for PERF_TYPE_RAW
- ARM: OMAP2+: hwmod: Add of_node_put() before break
- ARM: OMAP2+: adjust the location of put_device() call in omapdss_init_of
- irqchip/sifive-plic: Add missing thead,c900-plic match string
- netfilter: conntrack: don't refresh sctp entries in closed state
- arm64: dts: meson-gx: add ATF BL32 reserved-memory region
- arm64: dts: meson-g12: add ATF BL32 reserved-memory region
- arm64: dts: meson-g12: drop BL32 region from SEI510/SEI610
- kconfig: let 'shell' return enough output for deep path names
- ata: libata-core: Disable TRIM on M88V29
- drm/rockchip: dw_hdmi: Do not leave clock enabled in error case
- tracing: Fix tp_printk option related with tp_printk_stop_on_boot
- net: usb: qmi_wwan: Add support for Dell DW5829e
- net: macb: Align the dma and coherent dma masks
- kconfig: fix failing to generate auto.conf
- Linux 5.4.181
* Focal update: v5.4.180 upstream stable release (LP: #1966118)
- integrity: check the return value of audit_log_start()
- ima: Remove ima_policy file before directory
- ima: Allow template selection with ima_template[_fmt]= after ima_hash=
- ima: Do not print policy rule with inactive LSM labels
- mmc: sdhci-of-esdhc: Check for error num after setting mask
- net: phy: marvell: Fix RGMII Tx/Rx delays setting in 88e1121-compatible
PHYs
- net: phy: marvell: Fix MDI-x polarity setting in 88e1118-compatible PHYs
- NFS: Fix initialisation of nfs_client cl_flags field
- NFSD: Clamp WRITE offsets
- NFSD: Fix offset type in I/O trace points
- NFSv4 only print the label when its queried
- nfs: nfs4clinet: check the return value of kstrdup()
- NFSv4.1: Fix uninitialised variable in devicenotify
- NFSv4 remove zero number of fs_locations entries error check
- NFSv4 expose nfs_parse_server_name function
- drm: panel-orientation-quirks: Add quirk for the 1Netbook OneXPlayer
- net: sched: Clarify error message when qdisc kind is unknown
- scsi: target: iscsi: Make sure the np under each tpg is unique
- scsi: qedf: Fix refcount issue when LOGO is received during TMF
- scsi: myrs: Fix crash in error case
- PM: hibernate: Remove register_nosave_region_late()
- usb: dwc2: gadget: don't try to disable ep0 in dwc2_hsotg_suspend
- net: stmmac: dwmac-sun8i: use return val of readl_poll_timeout()
- KVM: nVMX: eVMCS: Filter out VM_EXIT_SAVE_VMX_PREEMPTION_TIMER
- riscv: fix build with binutils 2.38
- ARM: dts: imx23-evk: Remove MX23_PAD_SSP1_DETECT from hog group
- ARM: socfpga: fix missing RESET_CONTROLLER
- nvme-tcp: fix bogus request completion when failing to send AER
- ACPI/IORT: Check node revision for PMCG resources
- PM: s2idle: ACPI: Fix wakeup interrupts handling
- net: bridge: fix stale eth hdr pointer in br_dev_xmit
- perf probe: Fix ppc64 'perf probe add events failed' case
- ARM: dts: meson: Fix the UART compatible strings
- staging: fbtft: Fix error path in fbtft_driver_module_init()
- ARM: dts: imx6qdl-udoo: Properly describe the SD card detect
- usb: f_fs: Fix use-after-free for epfile
- misc: fastrpc: avoid double fput() on failed usercopy
- ixgbevf: Require large buffers for build_skb on 82599VF
- bonding: pair enable_port with slave_arr_updates
- ipmr,ip6mr: acquire RTNL before calling ip[6]mr_free_table() on failure
path
- nfp: flower: fix ida_idx not being released
- net: do not keep the dst cache when uncloning an skb dst and its metadata
- net: fix a memleak when uncloning an skb dst and its metadata
- veth: fix races around rq->rx_notify_masked
- net: mdio: aspeed: Add missing MODULE_DEVICE_TABLE
- tipc: rate limit warning for received illegal binding update
- net: amd-xgbe: disable interrupts during pci removal
- vt_ioctl: fix array_index_nospec in vt_setactivate
- vt_ioctl: add array_index_nospec to VT_ACTIVATE
- n_tty: wake up poll(POLLRDNORM) on receiving data
- eeprom: ee1004: limit i2c reads to I2C_SMBUS_BLOCK_MAX
- net: usb: ax88179_178a: Fix out-of-bounds accesses in RX fixup
- usb: ulpi: Move of_node_put to ulpi_dev_release
- usb: ulpi: Call of_node_put correctly
- usb: dwc3: gadget: Prevent core from processing stale TRBs
- usb: gadget: udc: renesas_usb3: Fix host to USB_ROLE_NONE transition
- USB: gadget: validate interface OS descriptor requests
- usb: gadget: rndis: check size of RNDIS_MSG_SET command
- usb: gadget: f_uac2: Define specific wTerminalType
- USB: serial: ftdi_sio: add support for Brainboxes US-159/235/320
- USB: serial: option: add ZTE MF286D modem
- USB: serial: ch341: add support for GW Instek USB2.0-Serial devices
- USB: serial: cp210x: add NCR Retail IO box id
- USB: serial: cp210x: add CPI Bulk Coin Recycler id
- seccomp: Invalidate seccomp mode to catch death failures
- hwmon: (dell-smm) Speed up setting of fan speed
- scsi: lpfc: Remove NVMe support if kernel has NVME_FC disabled
- perf: Fix list corruption in perf_cgroup_switch()
- Linux 5.4.180
* Focal update: v5.4.179 upstream stable release (LP: #1965591)
- moxart: fix potential use-after-free on remove path
- Linux 5.4.179
* CVE-2020-27820
- drm/nouveau: Add a dedicated mutex for the clients list
- drm/nouveau: clean up all clients on device removal
* CVE-2022-1016
- netfilter: nf_tables: initialize registers in nft_do_chain()
* CVE-2022-27223
- USB: gadget: validate endpoint index for xilinx udc
* CVE-2022-26490
- nfc: st21nfca: Fix potential buffer overflows in EVT_TRANSACTION
* CVE-2021-26401
- x86/speculation: Use generic retpoline by default on AMD
- x86/speculation: Update link to AMD speculation whitepaper
- x86/speculation: Warn about Spectre v2 LFENCE mitigation
- x86/speculation: Warn about eIBRS + LFENCE + Unprivileged eBPF + SMT
* CVE-2022-0001
- x86/speculation: Include unprivileged eBPF status in Spectre v2 mitigation
reporting
-- Stefan Bader <stefan.ba...@canonical.com> Thu, 14 Apr 2022 14:19:29
+0200
** Changed in: linux (Ubuntu Focal)
Status: Fix Committed => Fix Released
** CVE added: https://cve.mitre.org/cgi-bin/cvename.cgi?name=2020-27820
** CVE added: https://cve.mitre.org/cgi-bin/cvename.cgi?name=2021-26401
** CVE added: https://cve.mitre.org/cgi-bin/cvename.cgi?name=2022-0001
** CVE added: https://cve.mitre.org/cgi-bin/cvename.cgi?name=2022-1016
** CVE added: https://cve.mitre.org/cgi-bin/cvename.cgi?name=2022-26490
** CVE added: https://cve.mitre.org/cgi-bin/cvename.cgi?name=2022-27223
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux in Ubuntu.
https://bugs.launchpad.net/bugs/1967292
Title:
net/mlx5e: Fix page DMA map/unmap attributes
Status in linux package in Ubuntu:
Fix Released
Status in linux source package in Focal:
Fix Released
Status in linux source package in Impish:
Fix Released
Status in linux source package in Jammy:
Fix Released
Bug description:
SRU Justification
[Impact]
Upstream commit 0b7cfa4082fbf550595bc0e40f05614bd83bf0cd is a fix to
the Mellanox mlx5 driver that we would like to have you pick up. It
fixes a bug that was introduced in the 4.7 kernel, so it should be
applicable to pretty much every Linux distribution.
The bug causes a Linux panic when a Mellanox CX-4 VF is present in the
VM (for AccelNet) and swiotlb=force used on the kernel boot line.
swiotlb=force would not normally be used, which is why we haven’t seen
the problem reported previously, but it cropped up during some
internal testing. We are trying to head off a potential problem in a
future scenario for Confidential VMs.
Primarily we want to ensure that any CVMs images have this fix, as we don’t
want CVM images to exist in the wild that would fail in the future when CVMs
with AccelNet are offered.
Business Impact
This scenario could also exist in non-Azure instances.
[Test Case]
Microsoft tested.
[Where problems could appear]
Weird DMA and data corruption issues might crop up.
[Other Info]
SF: #00328938
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux/+bug/1967292/+subscriptions
--
Mailing list: https://launchpad.net/~kernel-packages
Post to : kernel-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kernel-packages
More help : https://help.launchpad.net/ListHelp
