5.12.10-rc1 review
Subsystems (6)
Greg Kroah-Hartman
maintainercover 5.12.10-rc1 review
This is the start of the stable review cycle for the 5.12.10 release. There are 161 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 Thu, 10 Jun 2021 17:59:18 +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-5.12.10-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-5.12.y
and the diffstat can be found below.
thanks,
greg k-h
-------------
Pseudo-Shortlog of commits:
Greg Kroah-Hartman <[email protected]>
Linux 5.12.10-rc1
David Ahern <[email protected]>
neighbour: allow NUD_NOARP entries to be forced GCed
Roger Pau Monne <[email protected]>
xen-netback: take a reference to the RX task thread
Pablo Neira Ayuso <[email protected]>
netfilter: nf_tables: missing error reporting for not selected expressions
Jiashuo Liang <[email protected]>
x86/fault: Don't send SIGSEGV twice on SEGV_PKUERR
Roja Rani Yarubandi <[email protected]>
i2c: qcom-geni: Suspend and resume the bus during SYSTEM_SLEEP_PM ops
Maciej Falkowski <[email protected]>
ARM: OMAP1: isp1301-omap: Add missing gpiod_add_lookup_table function
Zenghui Yu <[email protected]>
KVM: arm64: Resolve all pending PC updates before immediate exit
Marc Zyngier <[email protected]>
KVM: arm64: Commit pending PC adjustemnts before returning to userspace
Vitaly Kuznetsov <[email protected]>
x86/kvm: Disable all PV features on crash
Vitaly Kuznetsov <[email protected]>
x86/kvm: Disable kvmclock on all CPUs on shutdown
Vitaly Kuznetsov <[email protected]>
x86/kvm: Teardown PV features on boot CPU as well
Sean Christopherson <[email protected]>
KVM: SVM: Truncate GPR value for DR and CR accesses in !64-bit mode
Dmitry Baryshkov <[email protected]>
drm/msm/dpu: always use mdp device to scale bandwidth
Mina Almasry <[email protected]>
mm, hugetlb: fix simple resv_huge_pages underflow on UFFDIO_COPY
Filipe Manana <[email protected]>
btrfs: fix deadlock when cloning inline extents and low on available space
Josef Bacik <[email protected]>
btrfs: abort in rename_exchange if we fail to insert the second ref
Josef Bacik <[email protected]>
btrfs: fixup error handling in fixup_inode_link_counts
Josef Bacik <[email protected]>
btrfs: check error value from btrfs_update_inode in tree log
Filipe Manana <[email protected]>
btrfs: fix fsync failure and transaction abort after writes to prealloc extents
Josef Bacik <[email protected]>
btrfs: return errors from btrfs_del_csums in cleanup_ref_head
Josef Bacik <[email protected]>
btrfs: fix error handling in btrfs_del_csums
Josef Bacik <[email protected]>
btrfs: mark ordered extent and inode with error if we fail to finish
Naveen N. Rao <[email protected]>
powerpc/kprobes: Fix validation of prefixed instructions across page boundary
Borislav Petkov <[email protected]>
x86/thermal: Fix LVT thermal setup for SMI delivery mode
Thomas Gleixner <[email protected]>
x86/apic: Mark _all_ legacy interrupts when IO/APIC is missing
Nirmoy Das <[email protected]>
drm/amdgpu: make sure we unpin the UVD BO
Luben Tuikov <[email protected]>
drm/amdgpu: Don't query CE and UE errors
Krzysztof Kozlowski <[email protected]>
nfc: fix NULL ptr dereference in llcp_sock_getname() after failed connect
Nicholas Piggin <[email protected]>
KVM: PPC: Book3S HV: Save host FSCR in the P7/8 path
Pu Wen <[email protected]>
x86/sev: Check SME/SEV support in CPUID first
Borislav Petkov <[email protected]>
dmaengine: idxd: Use cpu_feature_enabled()
Thomas Gleixner <[email protected]>
x86/cpufeatures: Force disable X86_FEATURE_ENQCMD and remove update_pasid()
James Smart <[email protected]>
scsi: lpfc: Fix failure to transmit ABTS on FC link
Ding Hui <[email protected]>
mm/page_alloc: fix counting of free pages after take off from buddy
Gerald Schaefer <[email protected]>
mm/debug_vm_pgtable: fix alignment for pmd/pud_advanced_tests()
Junxiao Bi <[email protected]>
ocfs2: fix data corruption by fallocate
Mark Rutland <[email protected]>
pid: take a reference when initializing cad_pid
Marco Elver <[email protected]>
kfence: use TASK_IDLE when awaiting allocation
Marco Elver <[email protected]>
kfence: maximize allocation wait timeout duration
Thomas Bogendoerfer <[email protected]>
Revert "MIPS: make userspace mapping young by default"
Phil Elwell <[email protected]>
usb: dwc2: Fix build in periphal-only mode
Ritesh Harjani <[email protected]>
ext4: fix accessing uninit percpu counter variable with fast_commit
Phillip Potter <[email protected]>
ext4: fix memory leak in ext4_mb_init_backend on error path.
Harshad Shirwadkar <[email protected]>
ext4: fix fast commit alignment issues
Ye Bin <[email protected]>
ext4: fix bug on in ext4_es_cache_extent as ext4_split_extent_at failed
Alexey Makhalov <[email protected]>
ext4: fix memory leak in ext4_fill_super
Marek Vasut <[email protected]>
ARM: dts: imx6q-dhcom: Add PU,VDD1P1,VDD2P5 regulators
Michal Vokáč <[email protected]>
ARM: dts: imx6dl-yapp4: Fix RGMII connection to QCA8334 switch
Hui Wang <[email protected]>
ALSA: hda: update the power_state during the direct-complete
Carlos M <[email protected]>
ALSA: hda: Fix for mute key LED for HP Pavilion 15-CK0xx
Takashi Iwai <[email protected]>
ALSA: timer: Fix master timer notification
Bob Peterson <[email protected]>
gfs2: fix scheduling while atomic bug in glocks
Ahelenia Ziemiańska <[email protected]>
HID: multitouch: require Finger field to mark Win8 reports as MT
Johan Hovold <[email protected]>
HID: magicmouse: fix NULL-deref on disconnect
Johnny Chuang <[email protected]>
HID: i2c-hid: Skip ELAN power-on command after reset
Pavel Skripkin <[email protected]>
net: caif: fix memory leak in cfusbl_device_notify
Pavel Skripkin <[email protected]>
net: caif: fix memory leak in caif_device_notify
Pavel Skripkin <[email protected]>
net: caif: add proper error handling
Pavel Skripkin <[email protected]>
net: caif: added cfserl_release function
Jason A. Donenfeld <[email protected]>
wireguard: allowedips: free empty intermediate nodes when removing single node
Jason A. Donenfeld <[email protected]>
wireguard: allowedips: allocate nodes in kmem_cache
Jason A. Donenfeld <[email protected]>
wireguard: allowedips: remove nodes in O(1)
Jason A. Donenfeld <[email protected]>
wireguard: allowedips: initialize list head in selftest
Jason A. Donenfeld <[email protected]>
wireguard: selftests: make sure rp_filter is disabled on vethc
Jason A. Donenfeld <[email protected]>
wireguard: selftests: remove old conntrack kconfig value
Jason A. Donenfeld <[email protected]>
wireguard: use synchronize_net rather than synchronize_rcu
Jason A. Donenfeld <[email protected]>
wireguard: peer: allocate in kmem_cache
Jason A. Donenfeld <[email protected]>
wireguard: do not use -O3
Lin Ma <[email protected]>
Bluetooth: use correct lock to prevent UAF of hdev object
Lin Ma <[email protected]>
Bluetooth: fix the erroneous flush_work() order
James Zhu <[email protected]>
drm/amdgpu/jpeg3: add cancel_delayed_work_sync before power gate
James Zhu <[email protected]>
drm/amdgpu/jpeg2.5: add cancel_delayed_work_sync before power gate
James Zhu <[email protected]>
drm/amdgpu/vcn3: add cancel_delayed_work_sync before power gate
Pavel Begunkov <[email protected]>
io_uring: fix ltout double free on completion race
Jens Axboe <[email protected]>
io_uring: wrap io_kiocb reference count manipulation in helpers
Pavel Begunkov <[email protected]>
io_uring: use better types for cflags
Pavel Begunkov <[email protected]>
io_uring: fix link timeout refs
Simon Ser <[email protected]>
amdgpu: fix GEM obj leak in amdgpu_display_user_framebuffer_create
Ilya Dryomov <[email protected]>
libceph: don't set global_id until we get an auth ticket
Jisheng Zhang <[email protected]>
riscv: vdso: fix and clean-up Makefile
Johan Hovold <[email protected]>
serial: stm32: fix threaded interrupt handling
Hoang Le <[email protected]>
tipc: fix unique bearer names sanity check
Hoang Le <[email protected]>
tipc: add extack messages for bearer/media failure
Jerome Brunet <[email protected]>
arm64: meson: select COMMON_CLK
Tony Lindgren <[email protected]>
bus: ti-sysc: Fix flakey idling of uarts and stop using swsup_sidle_act
Geert Uytterhoeven <[email protected]>
ARM: dts: imx: emcon-avari: Fix nxp,pca8574 #gpio-cells
Fabio Estevam <[email protected]>
ARM: dts: imx7d-pico: Fix the 'tuning-step' property
Fabio Estevam <[email protected]>
ARM: dts: imx7d-meerkat96: Fix the 'tuning-step' property
Michael Walle <[email protected]>
arm64: dts: freescale: sl28: var1: fix RGMII clock and voltage
Michael Walle <[email protected]>
arm64: dts: freescale: sl28: var4: fix RGMII clock and voltage
Lucas Stach <[email protected]>
arm64: dts: zii-ultra: fix 12V_MAIN voltage
Lucas Stach <[email protected]>
arm64: dts: zii-ultra: remove second GEN_3V3 regulator instance
Michael Walle <[email protected]>
arm64: dts: ls1028a: fix memory node
Tony Lindgren <[email protected]>
bus: ti-sysc: Fix am335x resume hang for usb otg module
Jens Wiklander <[email protected]>
optee: use export_uuid() to copy client UUID
Vignesh Raghavendra <[email protected]>
arm64: dts: ti: j7200-main: Mark Main NAVSS as dma-coherent
Magnus Karlsson <[email protected]>
ixgbe: add correct exception tracing for XDP
Magnus Karlsson <[email protected]>
ixgbe: optimize for XDP_REDIRECT in xsk path
Magnus Karlsson <[email protected]>
ice: add correct exception tracing for XDP
Magnus Karlsson <[email protected]>
ice: optimize for XDP_REDIRECT in xsk path
Magnus Karlsson <[email protected]>
i40e: add correct exception tracing for XDP
Magnus Karlsson <[email protected]>
i40e: optimize for XDP_REDIRECT in xsk path
Rahul Lakkireddy <[email protected]>
cxgb4: avoid link re-train during TC-MQPRIO configuration
Yunjian Wang <[email protected]>
sch_htb: fix refcount leak in htb_parent_to_leaf_offload
Roja Rani Yarubandi <[email protected]>
i2c: qcom-geni: Add shutdown callback for i2c
Dave Ertman <[email protected]>
ice: Allow all LLDP packets from PF to Tx
Paul Greenwalt <[email protected]>
ice: report supported and advertised autoneg using PHY capabilities
Haiyue Wang <[email protected]>
ice: handle the VF VSI rebuild failure
Brett Creeley <[email protected]>
ice: Fix VFR issues for AVF drivers that expect ATQLEN cleared
Brett Creeley <[email protected]>
ice: Fix allowing VF to request more/less queues via virtchnl
Coco Li <[email protected]>
ipv6: Fix KASAN: slab-out-of-bounds Read in fib6_nh_flush_exceptions
Rahul Lakkireddy <[email protected]>
cxgb4: fix regression with HASH tc prio value update
Maciej Fijalkowski <[email protected]>
ice: track AF_XDP ZC enabled queues in bitmap
Magnus Karlsson <[email protected]>
ixgbevf: add correct exception tracing for XDP
Magnus Karlsson <[email protected]>
igb: add correct exception tracing for XDP
Kurt Kanzenbach <[email protected]>
igb: Fix XDP with PTP enabled
Wei Yongjun <[email protected]>
ieee802154: fix error return code in ieee802154_llsec_getparams()
Zhen Lei <[email protected]>
ieee802154: fix error return code in ieee802154_add_iface()
Daniel Borkmann <[email protected]>
bpf, lockdown, audit: Fix buggy SELinux lockdown permission checks
Zhihao Cheng <[email protected]>
drm/i915/selftests: Fix return value check in live_breadcrumbs_smoketest()
Pablo Neira Ayuso <[email protected]>
netfilter: nfnetlink_cthelper: hit EBUSY on updates if size mismatches
Pablo Neira Ayuso <[email protected]>
netfilter: nft_ct: skip expectations for confirmed conntrack
Max Gurtovoy <[email protected]>
nvmet: fix freeing unallocated p2pmem
Yevgeny Kliteynik <[email protected]>
net/mlx5: DR, Create multi-destination flow table with level less than 64
Roi Dayan <[email protected]>
net/mlx5e: Fix adding encap rules to slow path
Roi Dayan <[email protected]>
net/mlx5e: Check for needed capability for cvlan matching
Moshe Shemesh <[email protected]>
net/mlx5: Check firmware sync reset requested is set before trying to abort it
Aya Levin <[email protected]>
net/mlx5e: Fix incompatible casting
Maxim Mikityanskiy <[email protected]>
net/tls: Fix use-after-free after the TLS device goes down and up
Maxim Mikityanskiy <[email protected]>
net/tls: Replace TLS_RX_SYNC_RUNNING with RCU
Alexander Aring <[email protected]>
net: sock: fix in-kernel mark setting
Vladimir Oltean <[email protected]>
net: dsa: tag_8021q: fix the VLAN IDs used for encoding sub-VLANs
Li Huafei <[email protected]>
perf probe: Fix NULL pointer dereference in convert_variable_location()
Erik Kaneda <[email protected]>
ACPICA: Clean up context mutex during object deletion
Sagi Grimberg <[email protected]>
nvme-rdma: fix in-casule data send for chained sgls
Paolo Abeni <[email protected]>
mptcp: do not reset MP_CAPABLE subflow on mapping errors
Paolo Abeni <[email protected]>
mptcp: always parse mptcp options for MPC reqsk
Paolo Abeni <[email protected]>
mptcp: fix sk_forward_memory corruption on retransmission
Ariel Levkovich <[email protected]>
net/sched: act_ct: Fix ct template allocation for zone 0
Paul Blakey <[email protected]>
net/sched: act_ct: Offload connections with commit action
Parav Pandit <[email protected]>
devlink: Correct VIRTUAL port to not have phys_port attributes
Javier Martinez Canillas <[email protected]>
kbuild: Quote OBJCOPY var to avoid a pahole call break the build
Arnd Bergmann <[email protected]>
HID: i2c-hid: fix format string mismatch
Basavaraj Natikar <[email protected]>
HID: amd_sfh: Fix memory leak in amd_sfh_work
Zhen Lei <[email protected]>
HID: pidff: fix error return code in hid_pidff_init()
Tom Rix <[email protected]>
HID: logitech-hidpp: initialize level variable
Julian Anastasov <[email protected]>
ipvs: ignore IP_VS_SVC_F_HASHED flag when adding service
Max Gurtovoy <[email protected]>
vfio/platform: fix module_put call in error flow
Wei Yongjun <[email protected]>
samples: vfio-mdev: fix error handing in mdpy_fb_probe()
Randy Dunlap <[email protected]>
vfio/pci: zap_vma_ptes() needs MMU
Zhen Lei <[email protected]>
vfio/pci: Fix error return code in vfio_ecap_init()
Rasmus Villemoes <[email protected]>
efi: cper: fix snprintf() use in cper_dimm_err_location()
Dan Carpenter <[email protected]>
efi/libstub: prevent read overflow in find_file_option()
Heiner Kallweit <[email protected]>
efi: Allow EFI_MEMORY_XP and EFI_MEMORY_RO both to be cleared
Changbin Du <[email protected]>
efi/fdt: fix panic when no valid fdt found
Florian Westphal <[email protected]>
netfilter: conntrack: unregister ipv4 sockopts on error unwind
Grant Peltier <[email protected]>
hwmon: (pmbus/isl68137) remove READ_TEMPERATURE_3 for RAA228228
Armin Wolf <[email protected]>
hwmon: (dell-smm-hwmon) Fix index values
Lorenzo Bianconi <[email protected]>
mt76: mt76x0e: fix device hang during suspend/resume
Lorenzo Bianconi <[email protected]>
mt76: mt7921: fix possible AOOB issue in mt7921_mcu_tx_rate_report
Lorenzo Bianconi <[email protected]>
mt76: mt7921: add rcu section in mt7921_mcu_tx_rate_report
-------------
Diffstat:
Makefile | 4 +- arch/arm/boot/dts/imx6dl-yapp4-common.dtsi | 6 +- arch/arm/boot/dts/imx6q-dhcom-som.dtsi | 12 ++ arch/arm/boot/dts/imx6qdl-emcon-avari.dtsi | 2 +- arch/arm/boot/dts/imx7d-meerkat96.dts | 2 +- arch/arm/boot/dts/imx7d-pico.dtsi | 2 +- arch/arm/mach-omap1/board-h2.c | 4 +- arch/arm64/Kconfig.platforms | 1 + .../freescale/fsl-ls1028a-kontron-sl28-var1.dts | 3 +- .../freescale/fsl-ls1028a-kontron-sl28-var4.dts | 5 +- arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi | 4 +- .../boot/dts/freescale/imx8mq-zii-ultra-rmb3.dts | 10 +- .../arm64/boot/dts/freescale/imx8mq-zii-ultra.dtsi | 23 +-- arch/arm64/boot/dts/ti/k3-j7200-main.dtsi | 2 + arch/arm64/include/asm/kvm_asm.h | 1 + arch/arm64/kvm/arm.c | 20 ++- arch/arm64/kvm/hyp/exception.c | 4 +- arch/arm64/kvm/hyp/nvhe/hyp-main.c | 8 + arch/mips/mm/cache.c | 30 ++-- arch/powerpc/kernel/kprobes.c | 4 +- arch/powerpc/kvm/book3s_hv.c | 1 - arch/powerpc/kvm/book3s_hv_rmhandlers.S | 7 + arch/riscv/kernel/vdso/Makefile | 4 +- arch/x86/include/asm/apic.h | 1 + arch/x86/include/asm/disabled-features.h | 7 +- arch/x86/include/asm/fpu/api.h | 6 +- arch/x86/include/asm/fpu/internal.h | 7 - arch/x86/include/asm/kvm_para.h | 10 +- arch/x86/include/asm/thermal.h | 4 +- arch/x86/kernel/apic/apic.c | 1 + arch/x86/kernel/apic/vector.c | 20 +++ arch/x86/kernel/fpu/xstate.c | 57 ------- arch/x86/kernel/kvm.c | 92 +++++++--- arch/x86/kernel/kvmclock.c | 26 +-- arch/x86/kernel/setup.c | 9 + arch/x86/kvm/svm/svm.c | 8 +- arch/x86/mm/fault.c | 4 +- arch/x86/mm/mem_encrypt_identity.c | 11 +- drivers/acpi/acpica/utdelete.c | 8 + drivers/bus/ti-sysc.c | 57 ++++++- drivers/dma/idxd/init.c | 4 +- drivers/firmware/efi/cper.c | 4 +- drivers/firmware/efi/fdtparams.c | 3 + drivers/firmware/efi/libstub/file.c | 2 +- drivers/firmware/efi/memattr.c | 5 - drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c | 16 -- drivers/gpu/drm/amd/amdgpu/amdgpu_display.c | 1 + drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c | 4 +- drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c | 4 +- drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c | 1 + drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c | 5 +- drivers/gpu/drm/i915/selftests/i915_request.c | 4 +- drivers/gpu/drm/msm/disp/dpu1/dpu_kms.c | 3 +- drivers/gpu/drm/msm/disp/dpu1/dpu_mdss.c | 51 +----- drivers/hid/amd-sfh-hid/amd_sfh_client.c | 1 + drivers/hid/hid-logitech-hidpp.c | 1 + drivers/hid/hid-magicmouse.c | 2 +- drivers/hid/hid-multitouch.c | 10 +- drivers/hid/i2c-hid/i2c-hid-core.c | 13 +- drivers/hid/usbhid/hid-pidff.c | 1 + drivers/hwmon/dell-smm-hwmon.c | 4 +- drivers/hwmon/pmbus/isl68137.c | 4 +- drivers/i2c/busses/i2c-qcom-geni.c | 21 ++- drivers/net/ethernet/chelsio/cxgb4/cxgb4.h | 2 - drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c | 4 +- .../net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c | 14 +- .../net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c | 9 +- drivers/net/ethernet/chelsio/cxgb4/sge.c | 6 + drivers/net/ethernet/intel/i40e/i40e_txrx.c | 7 +- drivers/net/ethernet/intel/i40e/i40e_xsk.c | 15 +- drivers/net/ethernet/intel/ice/ice.h | 8 +- drivers/net/ethernet/intel/ice/ice_ethtool.c | 51 +----- drivers/net/ethernet/intel/ice/ice_hw_autogen.h | 1 + drivers/net/ethernet/intel/ice/ice_lib.c | 12 ++ drivers/net/ethernet/intel/ice/ice_txrx.c | 17 +- drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c | 19 ++- drivers/net/ethernet/intel/ice/ice_xsk.c | 19 ++- drivers/net/ethernet/intel/igb/igb.h | 2 +- drivers/net/ethernet/intel/igb/igb_main.c | 55 +++--- drivers/net/ethernet/intel/igb/igb_ptp.c | 23 ++- drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 16 +- drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c | 21 ++- drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c | 3 + .../net/ethernet/mellanox/mlx5/core/en_ethtool.c | 5 +- drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 9 + .../ethernet/mellanox/mlx5/core/eswitch_offloads.c | 3 +- drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c | 3 + .../ethernet/mellanox/mlx5/core/lib/fs_chains.c | 2 +- .../ethernet/mellanox/mlx5/core/lib/fs_chains.h | 5 + .../ethernet/mellanox/mlx5/core/steering/dr_fw.c | 3 +- drivers/net/wireguard/Makefile | 3 +- drivers/net/wireguard/allowedips.c | 189 +++++++++++---------- drivers/net/wireguard/allowedips.h | 14 +- drivers/net/wireguard/main.c | 17 +- drivers/net/wireguard/peer.c | 27 ++- drivers/net/wireguard/peer.h | 3 + drivers/net/wireguard/selftest/allowedips.c | 165 +++++++++--------- drivers/net/wireguard/socket.c | 2 +- drivers/net/wireless/mediatek/mt76/mt76x0/pci.c | 81 ++++++++- drivers/net/wireless/mediatek/mt76/mt7921/mcu.c | 24 ++- drivers/net/xen-netback/interface.c | 6 + drivers/nvme/host/rdma.c | 5 +- drivers/nvme/target/core.c | 33 ++-- drivers/scsi/lpfc/lpfc_sli.c | 4 +- drivers/tee/optee/call.c | 6 +- drivers/tee/optee/optee_msg.h | 6 +- drivers/thermal/intel/therm_throt.c | 15 +- drivers/tty/serial/stm32-usart.c | 22 +-- drivers/usb/dwc2/core_intr.c | 4 + drivers/vfio/pci/Kconfig | 1 + drivers/vfio/pci/vfio_pci_config.c | 2 +- drivers/vfio/platform/vfio_platform_common.c | 2 +- fs/btrfs/extent-tree.c | 2 +- fs/btrfs/file-item.c | 108 +++++++++--- fs/btrfs/inode.c | 19 ++- fs/btrfs/reflink.c | 38 +++-- fs/btrfs/tree-log.c | 21 ++- fs/ext4/extents.c | 43 ++--- fs/ext4/fast_commit.c | 170 +++++++++--------- fs/ext4/fast_commit.h | 19 --- fs/ext4/ialloc.c | 6 +- fs/ext4/mballoc.c | 2 +- fs/ext4/super.c | 11 +- fs/gfs2/glock.c | 2 + fs/io_uring.c | 66 ++++--- fs/ocfs2/file.c | 55 +++++- include/linux/mlx5/mlx5_ifc.h | 2 + include/linux/pgtable.h | 8 + include/linux/platform_data/ti-sysc.h | 1 + include/net/caif/caif_dev.h | 2 +- include/net/caif/cfcnfg.h | 2 +- include/net/caif/cfserl.h | 1 + include/net/tls.h | 10 +- init/main.c | 2 +- kernel/bpf/helpers.c | 7 +- kernel/trace/bpf_trace.c | 32 ++-- mm/debug_vm_pgtable.c | 4 +- mm/hugetlb.c | 14 +- mm/kfence/core.c | 12 +- mm/memory.c | 4 + mm/page_alloc.c | 2 + net/bluetooth/hci_core.c | 7 +- net/bluetooth/hci_sock.c | 4 +- net/caif/caif_dev.c | 13 +- net/caif/caif_usb.c | 14 +- net/caif/cfcnfg.c | 16 +- net/caif/cfserl.c | 5 + net/ceph/auth.c | 36 ++-- net/core/devlink.c | 4 +- net/core/neighbour.c | 1 + net/core/sock.c | 16 +- net/dsa/tag_8021q.c | 2 +- net/ieee802154/nl-mac.c | 4 +- net/ieee802154/nl-phy.c | 4 +- net/ipv6/route.c | 8 +- net/mptcp/protocol.c | 16 +- net/mptcp/subflow.c | 76 ++++----- net/netfilter/ipvs/ip_vs_ctl.c | 2 +- net/netfilter/nf_conntrack_proto.c | 2 +- net/netfilter/nf_tables_api.c | 4 +- net/netfilter/nfnetlink_cthelper.c | 8 +- net/netfilter/nft_ct.c | 2 +- net/nfc/llcp_sock.c | 2 + net/sched/act_ct.c | 10 +- net/sched/sch_htb.c | 8 +- net/tipc/bearer.c | 94 +++++++--- net/tls/tls_device.c | 60 +++++-- net/tls/tls_device_fallback.c | 7 + net/tls/tls_main.c | 1 + samples/vfio-mdev/mdpy-fb.c | 13 +- scripts/Makefile.modfinal | 2 +- scripts/link-vmlinux.sh | 2 +- sound/core/timer.c | 3 +- sound/pci/hda/hda_codec.c | 5 + sound/pci/hda/patch_realtek.c | 1 + tools/perf/util/dwarf-aux.c | 8 +- tools/perf/util/probe-finder.c | 3 + tools/testing/selftests/wireguard/netns.sh | 1 + .../testing/selftests/wireguard/qemu/kernel.config | 1 - 179 files changed, 1677 insertions(+), 1069 deletions(-)
Greg Kroah-Hartman
maintainerPATCH 1/161 mt76: mt7921: add rcu section in mt7921_mcu_tx_rate_report
From: Lorenzo Bianconi <[email protected]>
[ Upstream commit 481fc927c8289919cc0be58666fcd1b7da187a0c ]
Introduce rcu section in mt7921_mcu_tx_rate_report before dereferencing wcid pointer otherwise loockdep will report the following issue:
[ 115.245740] =============================
[ 115.245754] WARNING: suspicious RCU usage
[ 115.245771] 5.10.20 #0 Not tainted
[ 115.245784] -----------------------------
[ 115.245816] other info that might help us debug this:
[ 115.245830] rcu_scheduler_active = 2, debug_locks = 1
[ 115.245845] 3 locks held by kworker/u4:1/20:
[ 115.245858] #0: ffffff80065ab138 ((wq_completion)phy0){+.+.}-{0:0}, at: process_one_work+0x1f8/0x6b8
[ 115.245948] #1: ffffffc01198bdd8 ((work_completion)(&(&dev->mphy.mac_work)->work)){+.+.}-{0:0}, at: process_one_8
[ 115.246027] #2: ffffff8006543ce8 (&dev->mutex#2){+.+.}-{3:3}, at: mt7921_mac_work+0x60/0x2b0 [mt7921e]
[ 115.246125]
[ 115.246125] stack backtrace:
[ 115.246142] CPU: 1 PID: 20 Comm: kworker/u4:1 Not tainted 5.10.20 #0
[ 115.246152] Hardware name: MediaTek MT7622 RFB1 board (DT)
[ 115.246168] Workqueue: phy0 mt7921_mac_work [mt7921e]
[ 115.246188] Call trace:
[ 115.246201] dump_backtrace+0x0/0x1a8
[ 115.246213] show_stack+0x14/0x30
[ 115.246228] dump_stack+0xec/0x134
[ 115.246240] lockdep_rcu_suspicious+0xcc/0xdc
[ 115.246255] mt7921_get_wtbl_info+0x2a4/0x310 [mt7921e]
[ 115.246269] mt7921_mac_work+0x284/0x2b0 [mt7921e]
[ 115.246281] process_one_work+0x2a0/0x6b8
[ 115.246293] worker_thread+0x40/0x440
[ 115.246305] kthread+0x144/0x148
[ 115.246317] ret_from_fork+0x10/0x18--- drivers/net/wireless/mediatek/mt76/mt7921/mcu.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/net/wireless/mediatek/mt76/mt7921/mcu.c +6 -1
@@ -404,9 +404,12 @@ mt7921_mcu_tx_rate_report(struct mt7921_dev *dev, struct sk_buff *skb,if (wlan_idx >=MT76_N_WCIDS )return ; + +rcu_read_lock (); + wcid =rcu_dereference (dev->mt76.wcid[wlan_idx]);if (!wcid) -return ; + goto out; msta = container_of(wcid, struct mt7921_sta, wcid); stats = &msta->stats;
@@ -414,6 +417,8 @@ mt7921_mcu_tx_rate_report(struct mt7921_dev *dev, struct sk_buff *skb,/* current rate */ mt7921_mcu_tx_rate_parse (mphy, &peer, &rate, curr); stats->tx_rate = rate; +out: +rcu_read_unlock (); } static void
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 2/161 mt76: mt7921: fix possible AOOB issue in mt7921_mcu_tx_rate_report
In reply to Greg Kroah-HartmanFrom: Lorenzo Bianconi <[email protected]>
[ Upstream commit d874e6c06952382897d35bf4094193cd44ae91bd ]
Fix possible array out of bound access in mt7921_mcu_tx_rate_report. Remove unnecessary varibable in mt7921_mcu_tx_rate_report
--- drivers/net/wireless/mediatek/mt76/mt7921/mcu.c | 17 ++++++++++------- 1 file changed, 10 insertions(+), 7 deletions(-)
drivers/net/wireless/mediatek/mt76/mt7921/mcu.c +10 -7
@@ -391,20 +391,22 @@ static void mt7921_mcu_tx_rate_report(struct mt7921_dev *dev, struct sk_buff *skb, u16 wlan_idx) { - struct mt7921_mcu_wlan_info_event *wtbl_info = - (struct mt7921_mcu_wlan_info_event *)(skb->data); -struct rate_info rate = {}; -u8 curr_idx = wtbl_info->rate_info.rate_idx; -u16 curr =le16_to_cpu (wtbl_info->rate_info.rate[curr_idx]); -struct mt7921_mcu_peer_cap peer = wtbl_info->peer_cap; + struct mt7921_mcu_wlan_info_event *wtbl_info;struct mt76_phy *mphy = &dev->mphy;struct mt7921_sta_stats *stats; +struct rate_info rate = {};struct mt7921_sta *msta;struct mt76_wcid *wcid; +u8 idx;if (wlan_idx >=MT76_N_WCIDS )return ; + wtbl_info = (struct mt7921_mcu_wlan_info_event *)skb->data; + idx = wtbl_info->rate_info.rate_idx; +if (idx >=ARRAY_SIZE (wtbl_info->rate_info.rate)) +return ; +rcu_read_lock (); wcid =rcu_dereference (dev->mt76.wcid[wlan_idx]);
@@ -415,7 +417,8 @@ mt7921_mcu_tx_rate_report(struct mt7921_dev *dev, struct sk_buff *skb, stats = &msta->stats;/* current rate */ -mt7921_mcu_tx_rate_parse (mphy, &peer, &rate, curr); +mt7921_mcu_tx_rate_parse (mphy, &wtbl_info->peer_cap, &rate, +le16_to_cpu (wtbl_info->rate_info.rate[idx])); stats->tx_rate = rate; out:rcu_read_unlock ();
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 3/161 mt76: mt76x0e: fix device hang during suspend/resume
In reply to Greg Kroah-HartmanFrom: Lorenzo Bianconi <[email protected]>
[ Upstream commit 509559c35bcd23d5a046624b225cb3e99a9f1481 ]
Similar to usb device, re-initialize mt76x0e device after resume in order to fix mt7630e hang during suspend/resume
--- .../net/wireless/mediatek/mt76/mt76x0/pci.c | 81 ++++++++++++++++++- 1 file changed, 77 insertions(+), 4 deletions(-)
drivers/net/wireless/mediatek/mt76/mt76x0/pci.c +77 -4
@@ -87,7 +87,7 @@ static const struct ieee80211_ops mt76x0e_ops = { .reconfig_complete = mt76x02_reconfig_complete, }; -static int mt76x0e_register_device (struct mt76x02_dev *dev) +static int mt76x0e_init_hardware (struct mt76x02_dev *dev,bool resume) {int err;
@@ -100,9 +100,11 @@ static int mt76x0e_register_device(struct mt76x02_dev *dev)if (err <0 )return err; - err =mt76x02_dma_init (dev); -if (err <0 ) -return err; +if (!resume) { + err =mt76x02_dma_init (dev); +if (err <0 ) +return err; + } err =mt76x0_init_hardware (dev);if (err <0 )
@@ -123,6 +125,17 @@ static int mt76x0e_register_device(struct mt76x02_dev *dev)mt76_clear (dev,0x110 ,BIT (9 ));mt76_set (dev,MT_MAX_LEN_CFG ,BIT (13 )); +return 0 ; +} + +static int mt76x0e_register_device (struct mt76x02_dev *dev) +{ +int err; + + err =mt76x0e_init_hardware (dev, false); +if (err <0 ) +return err; + err =mt76x0_register_device (dev);if (err <0 )return err;
@@ -167,6 +180,8 @@ mt76x0e_probe(struct pci_dev *pdev, const struct pci_device_id *id)if (ret)return ret; +mt76_pci_disable_aspm (pdev); + mdev =mt76_alloc_device (&pdev->dev,sizeof (*dev), &mt76x0e_ops, &drv_ops);if (!mdev)
@@ -220,6 +235,60 @@ mt76x0e_remove(struct pci_dev *pdev)mt76_free_device (mdev); } +#ifdef CONFIG_PM +static int mt76x0e_suspend (struct pci_dev *pdev,pm_message_t state) +{ +struct mt76_dev *mdev =pci_get_drvdata (pdev); + struct mt76x02_dev *dev = container_of(mdev, struct mt76x02_dev, mt76); +int i; + +mt76_worker_disable (&mdev->tx_worker); +for (i =0 ; i <ARRAY_SIZE (mdev->phy.q_tx); i++) +mt76_queue_tx_cleanup (dev, mdev->phy.q_tx[i], true); +for (i =0 ; i <ARRAY_SIZE (mdev->q_mcu); i++) +mt76_queue_tx_cleanup (dev, mdev->q_mcu[i], true); +napi_disable (&mdev->tx_napi); + +mt76_for_each_q_rx (mdev, i) +napi_disable (&mdev->napi[i]); + +mt76x02_dma_disable (dev); +mt76x02_mcu_cleanup (dev); +mt76x0_chip_onoff (dev, false, false); + +pci_enable_wake (pdev,pci_choose_state (pdev, state), true); +pci_save_state (pdev); + +return pci_set_power_state (pdev,pci_choose_state (pdev, state)); +} + +static int mt76x0e_resume (struct pci_dev *pdev) +{ +struct mt76_dev *mdev =pci_get_drvdata (pdev); + struct mt76x02_dev *dev = container_of(mdev, struct mt76x02_dev, mt76); +int err, i; + + err =pci_set_power_state (pdev,PCI_D0 ); +if (err) +return err; + +pci_restore_state (pdev); + +mt76_worker_enable (&mdev->tx_worker); + +mt76_for_each_q_rx (mdev, i) { +mt76_queue_rx_reset (dev, i); +napi_enable (&mdev->napi[i]); +napi_schedule (&mdev->napi[i]); + } + +napi_enable (&mdev->tx_napi); +napi_schedule (&mdev->tx_napi); + +return mt76x0e_init_hardware (dev, true); +} +#endif /* CONFIG_PM */ +static const struct pci_device_id mt76x0e_device_table[] = { {PCI_DEVICE (0x14c3 ,0x7610 ) }, {PCI_DEVICE (0x14c3 ,0x7630 ) },
@@ -237,6 +306,10 @@ static struct pci_driver mt76x0e_driver = { .id_table = mt76x0e_device_table, .probe = mt76x0e_probe, .remove = mt76x0e_remove, +#ifdef CONFIG_PM + .suspend = mt76x0e_suspend, + .resume = mt76x0e_resume, +#endif /* CONFIG_PM */ };module_pci_driver (mt76x0e_driver);
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 4/161 hwmon: (dell-smm-hwmon) Fix index values
In reply to Greg Kroah-HartmanFrom: Armin Wolf <[email protected]>
[ Upstream commit 35d470b5fbc9f82feb77b56bb0d5d0b5cd73e9da ]
When support for up to 10 temp sensors and for disabling automatic BIOS fan control was added, noone updated the index values used for disallowing fan support and fan type calls.
Fix those values.
--- drivers/hwmon/dell-smm-hwmon.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/hwmon/dell-smm-hwmon.c +2 -2
@@ -838,10 +838,10 @@ static struct attribute *i8k_attrs[] = {static umode_t i8k_is_visible (struct kobject *kobj,struct attribute *attr,int index) { -if (disallow_fan_support && index >=8 ) +if (disallow_fan_support && index >=20 )return 0 ;if (disallow_fan_type_call && - (index ==9 || index ==12 || index ==15 )) + (index ==21 || index ==25 || index ==28 ))return 0 ;if (index >=0 && index <=1 && !(i8k_hwmon_flags &I8K_HWMON_HAVE_TEMP1 ))
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 5/161 hwmon: (pmbus/isl68137) remove READ_TEMPERATURE_3 for RAA228228
In reply to Greg Kroah-HartmanFrom: Grant Peltier <[email protected]>
[ Upstream commit 2a29db088c7ae7121801a0d7a60740ed2d18c4f3 ]
The initial version of the RAA228228 datasheet claimed that the device supported READ_TEMPERATURE_3 but not READ_TEMPERATURE_1. It has since been discovered that the datasheet was incorrect. The RAA228228 does support READ_TEMPERATURE_1 but does not support READ_TEMPERATURE_3.
--- drivers/hwmon/pmbus/isl68137.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/hwmon/pmbus/isl68137.c +2 -2
@@ -244,8 +244,8 @@ static int isl68137_probe(struct i2c_client *client) info->read_word_data = raa_dmpvr2_read_word_data;break ;case raa_dmpvr2_2rail_nontc: - info->func[0 ] &= ~PMBUS_HAVE_TEMP ; - info->func[1 ] &= ~PMBUS_HAVE_TEMP ; + info->func[0 ] &= ~PMBUS_HAVE_TEMP3 ; + info->func[1 ] &= ~PMBUS_HAVE_TEMP3 ; fallthrough;case raa_dmpvr2_2rail: info->pages =2 ;
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 6/161 netfilter: conntrack: unregister ipv4 sockopts on error unwind
In reply to Greg Kroah-HartmanFrom: Florian Westphal <[email protected]>
[ Upstream commit 22cbdbcfb61acc78d5fc21ebb13ccc0d7e29f793 ]
When ipv6 sockopt register fails, the ipv4 one needs to be removed.
--- net/netfilter/nf_conntrack_proto.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/netfilter/nf_conntrack_proto.c +1 -1
@@ -660,7 +660,7 @@ int nf_conntrack_proto_init(void)#if IS_ENABLED (CONFIG_IPV6 ) cleanup_sockopt: -nf_unregister_sockopt (&so_getorigdst6); +nf_unregister_sockopt (&so_getorigdst);#endif return ret; }
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 7/161 efi/fdt: fix panic when no valid fdt found
In reply to Greg Kroah-HartmanFrom: Changbin Du <[email protected]>
[ Upstream commit 668a84c1bfb2b3fd5a10847825a854d63fac7baa ]
setup_arch() would invoke efi_init()->efi_get_fdt_params(). If no valid fdt found then initial_boot_params will be null. So we should stop further fdt processing here. I encountered this issue on risc-v.
--- drivers/firmware/efi/fdtparams.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/firmware/efi/fdtparams.c +3 -0
@@ -98,6 +98,9 @@ u64 __init efi_get_fdt_params(struct efi_memory_map_data *mm)BUILD_BUG_ON (ARRAY_SIZE (target) !=ARRAY_SIZE (name));BUILD_BUG_ON (ARRAY_SIZE (target) !=ARRAY_SIZE (dt_params[0 ].params)); +if (!fdt) +return 0 ; +for (i =0 ; i <ARRAY_SIZE (dt_params); i++) { node =fdt_path_offset (fdt, dt_params[i].path);if (node <0 )
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 8/161 efi: Allow EFI_MEMORY_XP and EFI_MEMORY_RO both to be cleared
In reply to Greg Kroah-HartmanFrom: Heiner Kallweit <[email protected]>
[ Upstream commit 45add3cc99feaaf57d4b6f01d52d532c16a1caee ]
UEFI spec 2.9, p.108, table 4-1 lists the scenario that both attributes are cleared with the description "No memory access protection is possible for Entry". So we can have valid entries where both attributes are cleared, so remove the check.
--- drivers/firmware/efi/memattr.c | 5 ----- 1 file changed, 5 deletions(-)
drivers/firmware/efi/memattr.c +0 -5
@@ -67,11 +67,6 @@ static bool entry_is_valid(const efi_memory_desc_t *in, efi_memory_desc_t *out)return false; } -if (!(in->attribute & (EFI_MEMORY_RO |EFI_MEMORY_XP ))) { -pr_warn ("Entry attributes invalid: RO and XP bits both cleared\n" ); -return false; - } -if (PAGE_SIZE >EFI_PAGE_SIZE && (!PAGE_ALIGNED (in->phys_addr) || !PAGE_ALIGNED (in->num_pages <<EFI_PAGE_SHIFT ))) {
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 9/161 efi/libstub: prevent read overflow in find_file_option()
In reply to Greg Kroah-HartmanFrom: Dan Carpenter <[email protected]>
[ Upstream commit c4039b29fe9637e1135912813f830994af4c867f ]
If the buffer has slashes up to the end then this will read past the end of the array. I don't anticipate that this is an issue for many people in real life, but it's the right thing to do and it makes static checkers happy.
--- drivers/firmware/efi/libstub/file.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/firmware/efi/libstub/file.c +1 -1
@@ -103,7 +103,7 @@ static int find_file_option(const efi_char16_t *cmdline, int cmdline_len,return 0 ;/* Skip any leading slashes */ -while (cmdline[i] ==L'/' || cmdline[i] ==L'\\' ) +while (i < cmdline_len && (cmdline[i] ==L'/' || cmdline[i] ==L'\\' )) i++;while (--result_len >0 && i < cmdline_len) {
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 10/161 efi: cper: fix snprintf() use in cper_dimm_err_location()
In reply to Greg Kroah-HartmanFrom: Rasmus Villemoes <[email protected]>
[ Upstream commit 942859d969de7f6f7f2659a79237a758b42782da ]
snprintf() should be given the full buffer size, not one less. And it guarantees nul-termination, so doing it manually afterwards is pointless.
It's even potentially harmful (though probably not in practice because CPER_REC_LEN is 256), due to the "return how much would have been written had the buffer been big enough" semantics. I.e., if the bank and/or device strings are long enough that the "DIMM location ..." output gets truncated, writing to msg[n] is a buffer overflow.
--- drivers/firmware/efi/cper.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
drivers/firmware/efi/cper.c +1 -3
@@ -276,8 +276,7 @@ static int cper_dimm_err_location(struct cper_mem_err_compact *mem, char *msg)if (!msg || !(mem->validation_bits &CPER_MEM_VALID_MODULE_HANDLE ))return 0 ; - n =0 ; - len =CPER_REC_LEN -1 ; + len =CPER_REC_LEN ;dmi_memdev_name (mem->mem_dev_handle, &bank, &device);if (bank && device) n =snprintf (msg, len,"DIMM location: %s %s " , bank, device);
@@ -286,7 +285,6 @@ static int cper_dimm_err_location(struct cper_mem_err_compact *mem, char *msg)"DIMM location: not present. DMI handle: 0x%.4x " , mem->mem_dev_handle); - msg[n] ='\0' ;return n; }
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 11/161 vfio/pci: Fix error return code in vfio_ecap_init()
In reply to Greg Kroah-HartmanFrom: Zhen Lei <[email protected]>
[ Upstream commit d1ce2c79156d3baf0830990ab06d296477b93c26 ]
The error code returned from vfio_ext_cap_len() is stored in 'len', not in 'ret'.
Message-Id: <[email protected]>
--- drivers/vfio/pci/vfio_pci_config.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/vfio/pci/vfio_pci_config.c +1 -1
@@ -1581,7 +1581,7 @@ static int vfio_ecap_init(struct vfio_pci_device *vdev)if (len ==0xFF ) { len =vfio_ext_cap_len (vdev, ecap, epos);if (len <0 ) -return ret; +return len; } }
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 12/161 vfio/pci: zap_vma_ptes() needs MMU
In reply to Greg Kroah-HartmanFrom: Randy Dunlap <[email protected]>
[ Upstream commit 2a55ca37350171d9b43d561528f23d4130097255 ]
zap_vma_ptes() is only available when CONFIG_MMU is set/enabled. Without CONFIG_MMU, vfio_pci.o has build errors, so make VFIO_PCI depend on MMU.
riscv64-linux-ld: drivers/vfio/pci/vfio_pci.o: in function `vfio_pci_mmap_open':
vfio_pci.c:(.text+0x1ec): undefined reference to `zap_vma_ptes'
riscv64-linux-ld: drivers/vfio/pci/vfio_pci.o: in function `.L0 ':
vfio_pci.c:(.text+0x165c): undefined reference to `zap_vma_ptes'
Message-Id: <[email protected]>
--- drivers/vfio/pci/Kconfig | 1 + 1 file changed, 1 insertion(+)
drivers/vfio/pci/Kconfig +1 -0
@@ -2,6 +2,7 @@ config VFIO_PCI tristate "VFIO support for PCI devices" depends on VFIO && PCI && EVENTFD + depends on MMU select VFIO_VIRQFD select IRQ_BYPASS_MANAGER help
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 13/161 samples: vfio-mdev: fix error handing in mdpy_fb_probe()
In reply to Greg Kroah-HartmanFrom: Wei Yongjun <[email protected]>
[ Upstream commit 752774ce7793a1f8baa55aae31f3b4caac49cbe4 ]
Fix to return a negative error code from the framebuffer_alloc() error handling case instead of 0, also release regions in some error handing cases.
Message-Id: <[email protected]>
--- samples/vfio-mdev/mdpy-fb.c | 13 +++++++++---- 1 file changed, 9 insertions(+), 4 deletions(-)
samples/vfio-mdev/mdpy-fb.c +9 -4
@@ -117,22 +117,27 @@ static int mdpy_fb_probe(struct pci_dev *pdev,if (format !=DRM_FORMAT_XRGB8888 ) {pci_err (pdev,"format mismatch (0x%x != 0x%x)\n" , format,DRM_FORMAT_XRGB8888 ); -return -EINVAL ; + ret = -EINVAL ; + goto err_release_regions; }if (width <100 || width >10000 ) {pci_err (pdev,"width (%d) out of range\n" , width); -return -EINVAL ; + ret = -EINVAL ; + goto err_release_regions; }if (height <100 || height >10000 ) {pci_err (pdev,"height (%d) out of range\n" , height); -return -EINVAL ; + ret = -EINVAL ; + goto err_release_regions; }pci_info (pdev,"mdpy found: %dx%d framebuffer\n" , width, height); info =framebuffer_alloc (sizeof (struct mdpy_fb_par ), &pdev->dev); -if (!info) +if (!info) { + ret = -ENOMEM ; goto err_release_regions; + }pci_set_drvdata (pdev, info); par = info->par;
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 14/161 vfio/platform: fix module_put call in error flow
In reply to Greg Kroah-HartmanFrom: Max Gurtovoy <[email protected]>
[ Upstream commit dc51ff91cf2d1e9a2d941da483602f71d4a51472 ]
The ->parent_module is the one that use in try_module_get. It should also be the one the we use in module_put during vfio_platform_open().
Message-Id: <[email protected]>
--- drivers/vfio/platform/vfio_platform_common.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/vfio/platform/vfio_platform_common.c +1 -1
@@ -289,7 +289,7 @@ err_irq:vfio_platform_regions_cleanup (vdev); err_reg:mutex_unlock (&driver_lock); -module_put (THIS_MODULE ); +module_put (vdev->parent_module);return ret; }
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 15/161 ipvs: ignore IP_VS_SVC_F_HASHED flag when adding service
In reply to Greg Kroah-HartmanFrom: Julian Anastasov <[email protected]>
[ Upstream commit 56e4ee82e850026d71223262c07df7d6af3bd872 ]
syzbot reported memory leak [1] when adding service with HASHED flag. We should ignore this flag both from sockopt and netlink provided data, otherwise the service is not hashed and not visible while releasing resources.
[1]
BUG: memory leak
unreferenced object 0xffff888115227800 (size 512):
comm "syz-executor263", pid 8658, jiffies 4294951882 (age 12.560s)
hex dump (first 32 bytes):
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
backtrace:
[<ffffffff83977188>] kmalloc include/linux/slab.h:556 [inline]
[<ffffffff83977188>] kzalloc include/linux/slab.h:686 [inline]
[<ffffffff83977188>] ip_vs_add_service+0x598/0x7c0 net/netfilter/ipvs/ip_vs_ctl.c:1343
[<ffffffff8397d770>] do_ip_vs_set_ctl+0x810/0xa40 net/netfilter/ipvs/ip_vs_ctl.c:2570
[<ffffffff838449a8>] nf_setsockopt+0x68/0xa0 net/netfilter/nf_sockopt.c:101
[<ffffffff839ae4e9>] ip_setsockopt+0x259/0x1ff0 net/ipv4/ip_sockglue.c:1435
[<ffffffff839fa03c>] raw_setsockopt+0x18c/0x1b0 net/ipv4/raw.c:857
[<ffffffff83691f20>] __sys_setsockopt+0x1b0/0x360 net/socket.c:2117
[<ffffffff836920f2>] __do_sys_setsockopt net/socket.c:2128 [inline]
[<ffffffff836920f2>] __se_sys_setsockopt net/socket.c:2125 [inline]
[<ffffffff836920f2>] __x64_sys_setsockopt+0x22/0x30 net/socket.c:2125
[<ffffffff84350efa>] do_syscall_64+0x3a/0xb0 arch/x86/entry/common.c:47
[<ffffffff84400068>] entry_SYSCALL_64_after_hwframe+0x44/0xae--- net/netfilter/ipvs/ip_vs_ctl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/netfilter/ipvs/ip_vs_ctl.c +1 -1
@@ -1367,7 +1367,7 @@ ip_vs_add_service(struct netns_ipvs *ipvs, struct ip_vs_service_user_kern *u,ip_vs_addr_copy (svc->af, &svc->addr, &u->addr); svc->port = u->port; svc->fwmark = u->fwmark; - svc->flags = u->flags; + svc->flags = u->flags & ~IP_VS_SVC_F_HASHED ; svc->timeout = u->timeout *HZ ; svc->netmask = u->netmask; svc->ipvs = ipvs;
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 16/161 HID: logitech-hidpp: initialize level variable
In reply to Greg Kroah-HartmanFrom: Tom Rix <[email protected]>
[ Upstream commit 81c8bf9170477d453b24a6bc3300d201d641e645 ]
Static analysis reports this representative problem
hid-logitech-hidpp.c:1356:23: warning: Assigned value is
garbage or undefined
hidpp->battery.level = level;
^ ~~~~~In some cases, 'level' is never set in hidpp20_battery_map_status_voltage() Since level is not available on all hw, initialize level to unknown.
--- drivers/hid/hid-logitech-hidpp.c | 1 + 1 file changed, 1 insertion(+)
drivers/hid/hid-logitech-hidpp.c +1 -0
@@ -1262,6 +1262,7 @@ static int hidpp20_battery_map_status_voltage(u8 data[3], int *voltage,int status;long flags = (long ) data[2 ]; + *level =POWER_SUPPLY_CAPACITY_LEVEL_UNKNOWN ;if (flags &0x80 )switch (flags &0x07 ) {
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 17/161 HID: pidff: fix error return code in hid_pidff_init()
In reply to Greg Kroah-HartmanFrom: Zhen Lei <[email protected]>
[ Upstream commit 3dd653c077efda8152f4dd395359617d577a54cd ]
Fix to return a negative error code from the error handling case instead of 0, as done elsewhere in this function.
--- drivers/hid/usbhid/hid-pidff.c | 1 + 1 file changed, 1 insertion(+)
drivers/hid/usbhid/hid-pidff.c +1 -0
@@ -1292,6 +1292,7 @@ int hid_pidff_init(struct hid_device *hid)if (pidff->pool[PID_DEVICE_MANAGED_POOL ].value && pidff->pool[PID_DEVICE_MANAGED_POOL ].value[0 ] ==0 ) { + error = -EPERM ;hid_notice (hid,"device does not support device managed pool\n" ); goto fail;
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 18/161 HID: amd_sfh: Fix memory leak in amd_sfh_work
In reply to Greg Kroah-HartmanFrom: Basavaraj Natikar <[email protected]>
[ Upstream commit 5ad755fd2b326aa2bc8910b0eb351ee6aece21b1 ]
Kmemleak tool detected a memory leak in the amd_sfh driver.
====================
unreferenced object 0xffff88810228ada0 (size 32):
comm "insmod", pid 3968, jiffies 4295056001 (age 775.792s) hex dump (first 32 bytes):
00 20 73 1f 81 88 ff ff 00 01 00 00 00 00 ad de . s.............
22 01 00 00 00 00 ad de 01 00 02 00 00 00 00 00 "...............
backtrace:
[<000000007b4c8799>] kmem_cache_alloc_trace+0x163/0x4f0
[<0000000005326893>] amd_sfh_get_report+0xa4/0x1d0 [amd_sfh]
[<000000002a9e5ec4>] amdtp_hid_request+0x62/0x80 [amd_sfh]
[<00000000b8a95807>] sensor_hub_get_feature+0x145/0x270 [hid_sensor_hub]
[<00000000fda054ee>] hid_sensor_parse_common_attributes+0x215/0x460 [hid_sensor_iio_common]
[<0000000021279ecf>] hid_accel_3d_probe+0xff/0x4a0 [hid_sensor_accel_3d]
[<00000000915760ce>] platform_probe+0x6a/0xd0
[<0000000060258a1f>] really_probe+0x192/0x620
[<00000000fa812f2d>] driver_probe_device+0x14a/0x1d0
[<000000005e79f7fd>] __device_attach_driver+0xbd/0x110
[<0000000070d15018>] bus_for_each_drv+0xfd/0x160
[<0000000013a3c312>] __device_attach+0x18b/0x220
[<000000008c7b4afc>] device_initial_probe+0x13/0x20
[<00000000e6e99665>] bus_probe_device+0xfe/0x120
[<00000000833fa90b>] device_add+0x6a6/0xe00
[<00000000fa901078>] platform_device_add+0x180/0x380
====================The fix is to freeing request_list entry once the processed entry is removed from the request_list.
--- drivers/hid/amd-sfh-hid/amd_sfh_client.c | 1 + 1 file changed, 1 insertion(+)
drivers/hid/amd-sfh-hid/amd_sfh_client.c +1 -0
@@ -88,6 +88,7 @@ static void amd_sfh_work(struct work_struct *work) sensor_index = req_node->sensor_idx; report_id = req_node->report_id; node_type = req_node->report_type; +kfree (req_node);if (node_type ==HID_FEATURE_REPORT ) { report_size = get_feature_report(sensor_index, report_id,
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 19/161 HID: i2c-hid: fix format string mismatch
In reply to Greg Kroah-HartmanFrom: Arnd Bergmann <[email protected]>
[ Upstream commit dc5f9f55502e13ba05731d5046a14620aa2ff456 ]
clang doesn't like printing a 32-bit integer using %hX format string:
drivers/hid/i2c-hid/i2c-hid-core.c:994:18: error: format specifies type 'unsigned short' but the argument has type '__u32' (aka 'unsigned int') [-Werror,-Wformat]
client->name, hid->vendor, hid->product);
^~~~~~~~~~~
drivers/hid/i2c-hid/i2c-hid-core.c:994:31: error: format specifies type 'unsigned short' but the argument has type '__u32' (aka 'unsigned int') [-Werror,-Wformat]
client->name, hid->vendor, hid->product);
^~~~~~~~~~~~Use an explicit cast to truncate it to the low 16 bits instead.
--- drivers/hid/i2c-hid/i2c-hid-core.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/hid/i2c-hid/i2c-hid-core.c +2 -2
@@ -990,8 +990,8 @@ int i2c_hid_core_probe(struct i2c_client *client, struct i2chid_ops *ops, hid->vendor =le16_to_cpu (ihid->hdesc.wVendorID); hid->product =le16_to_cpu (ihid->hdesc.wProductID); -snprintf (hid->name,sizeof (hid->name),"%s %04hX:%04hX" , - client->name, hid->vendor, hid->product); +snprintf (hid->name,sizeof (hid->name),"%s %04X:%04X" , + client->name, (u16 )hid->vendor, (u16 )hid->product);strlcpy (hid->phys,dev_name (&client->dev),sizeof (hid->phys)); ihid->quirks =i2c_hid_lookup_quirk (hid->vendor, hid->product);
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 20/161 kbuild: Quote OBJCOPY var to avoid a pahole call break the build
In reply to Greg Kroah-HartmanFrom: Javier Martinez Canillas <[email protected]>
[ Upstream commit ff2e6efda0d5c51b33e2bcc0b0b981ac0a0ef214 ]
The ccache tool can be used to speed up cross-compilation, by calling the compiler and binutils through ccache. For example, following should work:
$ export ARCH=arm64 CROSS_COMPILE="ccache aarch64-linux-gnu-"
$ make M=drivers/gpu/drm/rockchip/
but pahole fails to extract the BTF info from DWARF, breaking the build:
CC [M] drivers/gpu/drm/rockchip//rockchipdrm.mod.o LD [M] drivers/gpu/drm/rockchip//rockchipdrm.ko BTF [M] drivers/gpu/drm/rockchip//rockchipdrm.ko
aarch64-linux-gnu-objcopy: invalid option -- 'J'
Usage: aarch64-linux-gnu-objcopy [option(s)] in-file [out-file]
Copies a binary file, possibly transforming it in the process
...
make[1]: *** [scripts/Makefile.modpost:156: __modpost] Error 2
make: *** [Makefile:1866: modules] Error 2
this fails because OBJCOPY is set to "ccache aarch64-linux-gnu-copy" and later pahole is executed with the following command line:
LLVM_OBJCOPY=$(OBJCOPY) $(PAHOLE) -J --btf_base vmlinux $@
which gets expanded to:
LLVM_OBJCOPY=ccache aarch64-linux-gnu-objcopy pahole -J ...
instead of:
LLVM_OBJCOPY="ccache aarch64-linux-gnu-objcopy" pahole -J ...
--- scripts/Makefile.modfinal | 2 +- scripts/link-vmlinux.sh | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
scripts/Makefile.modfinal +1 -1
@@ -59,7 +59,7 @@ quiet_cmd_ld_ko_o = LD [M] $@ quiet_cmd_btf_ko = BTF [M] $@ cmd_btf_ko = \ if [ -f vmlinux ]; then \ - LLVM_OBJCOPY=$(OBJCOPY) $(PAHOLE) -J --btf_base vmlinux $@; \ + LLVM_OBJCOPY="$(OBJCOPY)" $(PAHOLE) -J --btf_base vmlinux $@; \ else \ printf "Skipping BTF generation for %s due to unavailability of vmlinux\n" $@ 1>&2; \ fi;
scripts/link-vmlinux.sh +1 -1
@@ -228,7 +228,7 @@ gen_btf() vmlinux_link ${1} info "BTF" ${2} - LLVM_OBJCOPY=${OBJCOPY} ${PAHOLE} -J ${1} + LLVM_OBJCOPY="${OBJCOPY}" ${PAHOLE} -J ${1} # Create ${2} which contains just .BTF section but no symbols. Add # SHF_ALLOC because .BTF will be part of the vmlinux image. --strip-all
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 21/161 devlink: Correct VIRTUAL port to not have phys_port attributes
In reply to Greg Kroah-HartmanFrom: Parav Pandit <[email protected]>
[ Upstream commit b28d8f0c25a9b0355116cace5f53ea52bd4020c8 ]
Physical port name, port number attributes do not belong to virtual port flavour. When VF or SF virtual ports are registered they incorrectly append "np0" string in the netdevice name of the VF/SF.
Before this fix, VF netdevice name were ens2f0np0v0, ens2f0np0v1 for VF 0 and 1 respectively.
After the fix, they are ens2f0v0, ens2f0v1.
With this fix, reading /sys/class/net/ens2f0v0/phys_port_name returns -EOPNOTSUPP.
Also devlink port show example for 2 VFs on one PF to ensure that any physical port attributes are not exposed.
$ devlink port show pci/0000:06:00.0/65535: type eth netdev ens2f0np0 flavour physical port 0 splittable false pci/0000:06:00.3/196608: type eth netdev ens2f0v0 flavour virtual splittable false pci/0000:06:00.4/262144: type eth netdev ens2f0v1 flavour virtual splittable false
This change introduces a netdevice name change on systemd/udev version 245 and higher which honors phys_port_name sysfs file for generation of netdevice name.
This also aligns to phys_port_name usage which is limited to switchdev ports as described in [1].
--- net/core/devlink.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
net/core/devlink.c +2 -2
@@ -705,7 +705,6 @@ static int devlink_nl_port_attrs_put(struct sk_buff *msg,case DEVLINK_PORT_FLAVOUR_PHYSICAL :case DEVLINK_PORT_FLAVOUR_CPU :case DEVLINK_PORT_FLAVOUR_DSA : -case DEVLINK_PORT_FLAVOUR_VIRTUAL :if (nla_put_u32 (msg,DEVLINK_ATTR_PORT_NUMBER , attrs->phys.port_number))return -EMSGSIZE ;
@@ -8629,7 +8628,6 @@ static int __devlink_port_phys_port_name_get(struct devlink_port *devlink_port,switch (attrs->flavour) {case DEVLINK_PORT_FLAVOUR_PHYSICAL : -case DEVLINK_PORT_FLAVOUR_VIRTUAL :if (!attrs->split) n =snprintf (name, len,"p%u" , attrs->phys.port_number);else
@@ -8670,6 +8668,8 @@ static int __devlink_port_phys_port_name_get(struct devlink_port *devlink_port, n =snprintf (name, len,"pf%usf%u" , attrs->pci_sf.pf, attrs->pci_sf.sf);break ; +case DEVLINK_PORT_FLAVOUR_VIRTUAL : +return -EOPNOTSUPP ; }if (n >= len)
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 22/161 net/sched: act_ct: Offload connections with commit action
In reply to Greg Kroah-HartmanFrom: Paul Blakey <[email protected]>
[ Upstream commit 0cc254e5aa37cf05f65bcdcdc0ac5c58010feb33 ]
Currently established connections are not offloaded if the filter has a "ct commit" action. This behavior will not offload connections of the following scenario:
$ tc_filter add dev $DEV ingress protocol ip prio 1 flower \ ct_state -trk \ action ct commit action goto chain 1
$ tc_filter add dev $DEV ingress protocol ip chain 1 prio 1 flower \ action mirred egress redirect dev $DEV2
$ tc_filter add dev $DEV2 ingress protocol ip prio 1 flower \ action ct commit action goto chain 1
$ tc_filter add dev $DEV2 ingress protocol ip prio 1 chain 1 flower \ ct_state +trk+est \ action mirred egress redirect dev $DEV
Offload established connections, regardless of the commit flag.
--- net/sched/act_ct.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-)
net/sched/act_ct.c +4 -3
@@ -984,7 +984,7 @@ static int tcf_ct_act(struct sk_buff *skb, const struct tc_action *a,*/ cached =tcf_ct_skb_nfct_cached (net, skb, p->zone, force);if (!cached) { -if (!commit &&tcf_ct_flow_table_lookup (p, skb, family)) { +if (tcf_ct_flow_table_lookup (p, skb, family)) { skip_add = true; goto do_nat; }
@@ -1024,10 +1024,11 @@ do_nat:* even if the connection is already confirmed. */ nf_conntrack_confirm (skb); - }else if (!skip_add) { -tcf_ct_flow_table_process_conn (p->ct_ft, ct, ctinfo); } +if (!skip_add) +tcf_ct_flow_table_process_conn (p->ct_ft, ct, ctinfo); + out_push:skb_push_rcsum (skb, nh_ofs);
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 23/161 net/sched: act_ct: Fix ct template allocation for zone 0
In reply to Greg Kroah-HartmanFrom: Ariel Levkovich <[email protected]>
[ Upstream commit fb91702b743dec78d6507c53a2dec8a8883f509d ]
Fix current behavior of skipping template allocation in case the ct action is in zone 0.
Skipping the allocation may cause the datapath ct code to ignore the entire ct action with all its attributes (commit, nat) in case the ct action in zone 0 was preceded by a ct clear action.
The ct clear action sets the ct_state to untracked and resets the skb->_nfct pointer. Under these conditions and without an allocated ct template, the skb->_nfct pointer will remain NULL which will cause the tc ct action handler to exit without handling commit and nat actions, if such exist.
For example, the following rule in OVS dp:
recirc_id(0x2),ct_state(+new-est-rel-rpl+trk),ct_label(0/0x1), \
in_port(eth0),actions:ct_clear,ct(commit,nat(src=10.11.0.12)), \
recirc(0x37a)
Will result in act_ct skipping the commit and nat actions in zone 0.
The change removes the skipping of template allocation for zone 0 and treats it the same as any other zone.
--- net/sched/act_ct.c | 3 --- 1 file changed, 3 deletions(-)
net/sched/act_ct.c +0 -3
@@ -1205,9 +1205,6 @@ static int tcf_ct_fill_params(struct net *net, sizeof(p->zone)); } -if (p->zone ==NF_CT_DEFAULT_ZONE_ID ) -return 0 ; -nf_ct_zone_init (&zone, p->zone,NF_CT_DEFAULT_ZONE_DIR ,0 ); tmpl =nf_ct_tmpl_alloc (net, &zone,GFP_KERNEL );if (!tmpl) {
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 24/161 mptcp: fix sk_forward_memory corruption on retransmission
In reply to Greg Kroah-HartmanFrom: Paolo Abeni <[email protected]>
[ Upstream commit b5941f066b4ca331db225a976dae1d6ca8cf0ae3 ]
MPTCP sk_forward_memory handling is a bit special, as such field is protected by the msk socket spin_lock, instead of the plain socket lock.
Currently we have a code path updating such field without handling the relevant lock:
__mptcp_retrans() -> __mptcp_clean_una_wakeup()
Several helpers in __mptcp_clean_una_wakeup() will update sk_forward_alloc, possibly causing such field corruption, as reported by Matthieu.
Address the issue providing and using a new variant of blamed function which explicitly acquires the msk spin lock.
--- net/mptcp/protocol.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-)
net/mptcp/protocol.c +15 -1
@@ -937,6 +937,10 @@ static void __mptcp_update_wmem(struct sock *sk) {struct mptcp_sock *msk =mptcp_sk (sk); +#ifdef CONFIG_LOCKDEP +WARN_ON_ONCE (!lockdep_is_held (&sk->sk_lock.slock)); +#endif +if (!msk->wmem_reserved)return ;
@@ -1075,10 +1079,20 @@ out:static void __mptcp_clean_una_wakeup (struct sock *sk) { +#ifdef CONFIG_LOCKDEP +WARN_ON_ONCE (!lockdep_is_held (&sk->sk_lock.slock)); +#endif __mptcp_clean_una (sk);mptcp_write_space (sk); } +static void mptcp_clean_una_wakeup (struct sock *sk) +{ +mptcp_data_lock (sk); +__mptcp_clean_una_wakeup (sk); +mptcp_data_unlock (sk); +} +static void mptcp_enter_memory_pressure (struct sock *sk) {struct mptcp_subflow_context *subflow;
@@ -2288,7 +2302,7 @@ static void __mptcp_retrans(struct sock *sk)struct sock *ssk;int ret; -__mptcp_clean_una_wakeup (sk); +mptcp_clean_una_wakeup (sk); dfrag =mptcp_rtx_head (sk);if (!dfrag) {if (mptcp_data_fin_enabled (msk)) {
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 25/161 mptcp: always parse mptcp options for MPC reqsk
In reply to Greg Kroah-HartmanFrom: Paolo Abeni <[email protected]>
[ Upstream commit 06f9a435b3aa12f4de6da91f11fdce8ce7b46205 ]
In subflow_syn_recv_sock() we currently skip options parsing for OoO packet, given that such packets may not carry the relevant MPC option.
If the peer generates an MPC+data TSO packet and some of the early segments are lost or get reorder, we server will ignore the peer key, causing transient, unexpected fallback to TCP.
The solution is always parsing the incoming MPTCP options, and do the fallback only for in-order packets. This actually cleans the existing code a bit.
--- net/mptcp/subflow.c | 17 ++++++++--------- 1 file changed, 8 insertions(+), 9 deletions(-)
net/mptcp/subflow.c +8 -9
@@ -608,21 +608,20 @@ static struct sock *subflow_syn_recv_sock(const struct sock *sk,/* if the sk is MP_CAPABLE, we try to fetch the client key */ if (subflow_req->mp_capable) { -if (TCP_SKB_CB (skb)->seq != subflow_req->ssn_offset +1 ) { -/* here we can receive and accept an in-window, -* out-of-order pkt, which will not carry the MP_CAPABLE -* opt even on mptcp enabled paths -*/ - goto create_msk; - } - +/* we can receive and accept an in-window, out-of-order pkt, +* which may not carry the MP_CAPABLE opt even on mptcp enabled +* paths: always try to extract the peer key, and fallback +* for packets missing it. +* Even OoO DSS packets coming legitly after dropped or +* reordered MPC will cause fallback, but we don't have other +* options. +*/ mptcp_get_options (skb, &mp_opt);if (!mp_opt.mp_capable) { fallback = true; goto create_child; } -create_msk: new_msk =mptcp_sk_clone (listener->conn, &mp_opt, req);if (!new_msk) fallback = true;
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 26/161 mptcp: do not reset MP_CAPABLE subflow on mapping errors
In reply to Greg Kroah-HartmanFrom: Paolo Abeni <[email protected]>
[ Upstream commit dea2b1ea9c705c5ba351a9174403fd83dbb68fc3 ]
When some mapping related errors occurs we close the main MPC subflow with a RST. We should instead fallback gracefully to TCP, and do the reset only for MPJ subflows.
--- net/mptcp/subflow.c | 59 +++++++++++++++++++++++---------------------- 1 file changed, 30 insertions(+), 29 deletions(-)
net/mptcp/subflow.c +30 -29
@@ -984,22 +984,11 @@ static bool subflow_check_data_avail(struct sock *ssk)u64 old_ack; status =get_mapping_status (ssk, msk); -pr_debug ("msk=%p ssk=%p status=%d" , msk, ssk, status); -if (status ==MAPPING_INVALID ) { - ssk->sk_err =EBADMSG ; - goto fatal; - } -if (status ==MAPPING_DUMMY ) { -__mptcp_do_fallback (msk); - skb =skb_peek (&ssk->sk_receive_queue); - subflow->map_valid =1 ; - subflow->map_seq =READ_ONCE (msk->ack_seq); - subflow->map_data_len = skb->len; - subflow->map_subflow_seq =tcp_sk (ssk)->copied_seq - - subflow->ssn_offset; - subflow->data_avail =MPTCP_SUBFLOW_DATA_AVAIL ; -return true; - } +if (unlikely (status ==MAPPING_INVALID )) + goto fallback; + +if (unlikely (status ==MAPPING_DUMMY )) + goto fallback;if (status !=MAPPING_OK ) goto no_data;
@@ -1012,10 +1001,8 @@ static bool subflow_check_data_avail(struct sock *ssk)* MP_CAPABLE-based mapping */ if (unlikely (!READ_ONCE (msk->can_ack))) { -if (!subflow->mpc_map) { - ssk->sk_err =EBADMSG ; - goto fatal; - } +if (!subflow->mpc_map) + goto fallback;WRITE_ONCE (msk->remote_key, subflow->remote_key);WRITE_ONCE (msk->ack_seq, subflow->map_seq);WRITE_ONCE (msk->can_ack, true);
@@ -1043,15 +1030,29 @@ static bool subflow_check_data_avail(struct sock *ssk) no_data:subflow_sched_work_if_closed (msk, ssk);return false; -fatal: -/* fatal protocol error, close the socket */ -/* This barrier is coupled with smp_rmb() in tcp_poll() */ -smp_wmb (); - ssk->sk_error_report (ssk); -tcp_set_state (ssk,TCP_CLOSE ); -tcp_send_active_reset (ssk,GFP_ATOMIC ); - subflow->data_avail =0 ; -return false; + +fallback: +/* RFC 8684 section 3.7. */ +if (subflow->mp_join || subflow->fully_established) { +/* fatal protocol error, close the socket. +* subflow_error_report() will introduce the appropriate barriers +*/ + ssk->sk_err =EBADMSG ; + ssk->sk_error_report (ssk); +tcp_set_state (ssk,TCP_CLOSE ); +tcp_send_active_reset (ssk,GFP_ATOMIC ); + subflow->data_avail =0 ; +return false; + } + +__mptcp_do_fallback (msk); + skb =skb_peek (&ssk->sk_receive_queue); + subflow->map_valid =1 ; + subflow->map_seq =READ_ONCE (msk->ack_seq); + subflow->map_data_len = skb->len; + subflow->map_subflow_seq =tcp_sk (ssk)->copied_seq - subflow->ssn_offset; + subflow->data_avail =MPTCP_SUBFLOW_DATA_AVAIL ; +return true; }bool mptcp_subflow_data_available (struct sock *sk)
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 27/161 nvme-rdma: fix in-casule data send for chained sgls
In reply to Greg Kroah-HartmanFrom: Sagi Grimberg <[email protected]>
[ Upstream commit 12b2aaadb6d5ef77434e8db21f469f46fe2d392e ]
We have only 2 inline sg entries and we allow 4 sg entries for the send wr sge. Larger sgls entries will be chained. However when we build in-capsule send wr sge, we iterate without taking into account that the sgl may be chained and still fit in-capsule (which can happen if the sgl is bigger than 2, but lower-equal to 4).
Fix in-capsule data mapping to correctly iterate chained sgls.
--- drivers/nvme/host/rdma.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/nvme/host/rdma.c +3 -2
@@ -1319,16 +1319,17 @@ static int nvme_rdma_map_sg_inline(struct nvme_rdma_queue *queue, int count) {struct nvme_sgl_desc *sg = &c->common.dptr.sgl; -struct scatterlist *sgl = req->data_sgl.sg_table.sgl;struct ib_sge *sge = &req->sge[1 ]; +struct scatterlist *sgl;u32 len =0 ;int i; -for (i =0 ; i < count; i++, sgl++, sge++) { +for_each_sg (req->data_sgl.sg_table.sgl, sgl, count, i) { sge->addr =sg_dma_address (sgl); sge->length =sg_dma_len (sgl); sge->lkey = queue->device->pd->local_dma_lkey; len += sge->length; + sge++; } sg->addr =cpu_to_le64 (queue->ctrl->ctrl.icdoff);
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 28/161 ACPICA: Clean up context mutex during object deletion
In reply to Greg Kroah-HartmanFrom: Erik Kaneda <[email protected]>
[ Upstream commit e4dfe108371214500ee10c2cf19268f53acaa803 ]
ACPICA commit bc43c878fd4ff27ba75b1d111b97ee90d4a82707
--- drivers/acpi/acpica/utdelete.c | 8 ++++++++ 1 file changed, 8 insertions(+)
drivers/acpi/acpica/utdelete.c +8 -0
@@ -285,6 +285,14 @@ static void acpi_ut_delete_internal_obj(union acpi_operand_object *object) }break ; +case ACPI_TYPE_LOCAL_ADDRESS_HANDLER : + +ACPI_DEBUG_PRINT ((ACPI_DB_ALLOCATIONS , +"***** Address handler %p\n" , object)); + +acpi_os_delete_mutex (object->address_space.context_mutex); +break ; +default :break ;
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 29/161 perf probe: Fix NULL pointer dereference in convert_variable_location()
In reply to Greg Kroah-HartmanFrom: Li Huafei <[email protected]>
[ Upstream commit 3cb17cce1e76ccc5499915a4d7e095a1ad6bf7ff ]
If we just check whether the variable can be converted, 'tvar' should be a null pointer. However, the null pointer check is missing in the 'Constant value' execution path.
The following cases can trigger this problem:
$ cat test.c #include <stdio.h>
void main(void)
{
int a;
const int b = 1;
asm volatile("mov %1, %0" : "=r"(a): "i"(b));
printf("a: %d\n", a);
}
$ gcc test.c -o test -O -g
$ sudo ./perf probe -x ./test -L "main"
<main@/home/lhf/test.c:0>
0 void main(void)
{
2 int a;
const int b = 1;
asm volatile("mov %1, %0" : "=r"(a): "i"(b));
6 printf("a: %d\n", a);
}
$ sudo ./perf probe -x ./test -V "main:6"
Segmentation faultThe check on 'tvar' is added. If 'tavr' is a null pointer, we return 0 to indicate that the variable can be converted. Now, we can successfully show the variables that can be accessed.
$ sudo ./perf probe -x ./test -V "main:6" Available variables at main:6 @<main+13> char* __fmt int a int b
However, the variable 'b' cannot be tracked.
$ sudo ./perf probe -x ./test -D "main:6 b" Failed to find the location of the 'b' variable at this address. Perhaps it has been optimized out. Use -V with the --range option to show 'b' location range. Error: Failed to add events.
This is because __die_find_variable_cb() did not successfully match variable 'b', which has the DW_AT_const_value attribute instead of DW_AT_location. We added support for DW_AT_const_value in __die_find_variable_cb(). With this modification, we can successfully track the variable 'b'.
$ sudo ./perf probe -x ./test -D "main:6 b" p:probe_test/main_L6 /home/lhf/test:0x1156 b=\1:s32
http://lore.kernel.org/lkml/[email protected]
--- tools/perf/util/dwarf-aux.c | 8 ++++++-- tools/perf/util/probe-finder.c | 3 +++ 2 files changed, 9 insertions(+), 2 deletions(-)
tools/perf/util/dwarf-aux.c +6 -2
@@ -975,9 +975,13 @@ static int __die_find_variable_cb(Dwarf_Die *die_mem, void *data)if ((tag == DW_TAG_formal_parameter || tag == DW_TAG_variable) &&die_compare_name (die_mem, fvp->name) && -/* Does the DIE have location information or external instance? */ +/* +* Does the DIE have location information or const value +* or external instance? +*/ (dwarf_attr (die_mem, DW_AT_external, &attr) || -dwarf_attr (die_mem, DW_AT_location, &attr))) +dwarf_attr (die_mem, DW_AT_location, &attr) || +dwarf_attr (die_mem, DW_AT_const_value, &attr)))return DIE_FIND_CB_END ;if (dwarf_haspc (die_mem, fvp->addr))return DIE_FIND_CB_CONTINUE ;
tools/perf/util/probe-finder.c +3 -0
@@ -190,6 +190,9 @@ static int convert_variable_location(Dwarf_Die *vr_die, Dwarf_Addr addr, immediate_value_is_supported()) {Dwarf_Sword snum; +if (!tvar) +return 0 ; +dwarf_formsdata (&attr, &snum); ret =asprintf (&tvar->value,"\\%ld" , (long )snum);
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 30/161 net: dsa: tag_8021q: fix the VLAN IDs used for encoding sub-VLANs
In reply to Greg Kroah-HartmanFrom: Vladimir Oltean <[email protected]>
[ Upstream commit 4ef8d857b5f494e62bce9085031563fda35f9563 ]
When using sub-VLANs in the range of 1-7, the resulting value from:
rx_vid = dsa_8021q_rx_vid_subvlan(ds, port, subvlan);
is wrong according to the description from tag_8021q.c:
| 11 | 10 | 9 | 8 | 7 | 6 | 5 | 4 | 3 | 2 | 1 | 0 | +-----------+-----+-----------------+-----------+-----------------------+ | DIR | SVL | SWITCH_ID | SUBVLAN | PORT | +-----------+-----+-----------------+-----------+-----------------------+
For example, when ds->index == 0, port == 3 and subvlan == 1, dsa_8021q_rx_vid_subvlan() returns 1027, same as it returns for subvlan == 0, but it should have returned 1043.
This is because the low portion of the subvlan bits are not masked properly when writing into the 12-bit VLAN value. They are masked into bits 4:3, but they should be masked into bits 5:4.
--- net/dsa/tag_8021q.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/dsa/tag_8021q.c +1 -1
@@ -64,7 +64,7 @@#define DSA_8021Q_SUBVLAN_HI_SHIFT 9#define DSA_8021Q_SUBVLAN_HI_MASK GENMASK(9, 9)#define DSA_8021Q_SUBVLAN_LO_SHIFT 4 -#define DSA_8021Q_SUBVLAN_LO_MASK GENMASK(4, 3) +#define DSA_8021Q_SUBVLAN_LO_MASK GENMASK(5, 4)#define DSA_8021Q_SUBVLAN_HI (x) (((x) & GENMASK(2, 2)) >> 2)#define DSA_8021Q_SUBVLAN_LO (x) ((x) & GENMASK(1, 0))#define DSA_8021Q_SUBVLAN (x) \
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 31/161 net: sock: fix in-kernel mark setting
In reply to Greg Kroah-HartmanFrom: Alexander Aring <[email protected]>
[ Upstream commit dd9082f4a9f94280fbbece641bf8fc0a25f71f7a ]
This patch fixes the in-kernel mark setting by doing an additional sk_dst_reset() which was introduced by commit 50254256f382 ("sock: Reset dst when changing sk_mark via setsockopt"). The code is now shared to avoid any further suprises when changing the socket mark value.
--- net/core/sock.c | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-)
net/core/sock.c +12 -4
@@ -815,10 +815,18 @@ void sock_set_rcvbuf(struct sock *sk, int val) }EXPORT_SYMBOL (sock_set_rcvbuf); +static void __sock_set_mark (struct sock *sk,u32 val) +{ +if (val != sk->sk_mark) { + sk->sk_mark = val; +sk_dst_reset (sk); + } +} +void sock_set_mark (struct sock *sk,u32 val) {lock_sock (sk); - sk->sk_mark = val; +__sock_set_mark (sk, val);release_sock (sk); }EXPORT_SYMBOL (sock_set_mark);
@@ -1126,10 +1134,10 @@ set_sndbuf:case SO_MARK :if (!ns_capable (sock_net (sk)->user_ns,CAP_NET_ADMIN )) { ret = -EPERM ; - }else if (val != sk->sk_mark) { - sk->sk_mark = val; -sk_dst_reset (sk); +break ; } + +__sock_set_mark (sk, val);break ;case SO_RXQ_OVFL :
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 32/161 net/tls: Replace TLS_RX_SYNC_RUNNING with RCU
In reply to Greg Kroah-HartmanFrom: Maxim Mikityanskiy <[email protected]>
[ Upstream commit 05fc8b6cbd4f979a6f25759c4a17dd5f657f7ecd ]
RCU synchronization is guaranteed to finish in finite time, unlike a busy loop that polls a flag. This patch is a preparation for the bugfix in the next patch, where the same synchronize_net() call will also be used to sync with the TX datapath.
--- include/net/tls.h | 1 - net/tls/tls_device.c | 10 +++------- 2 files changed, 3 insertions(+), 8 deletions(-)
include/net/tls.h +0 -1
@@ -193,7 +193,6 @@ struct tls_offload_context_tx { (sizeof(struct tls_offload_context_tx) + TLS_DRIVER_STATE_SIZE_TX)enum tls_context_flags { -TLS_RX_SYNC_RUNNING =0 ,/* Unlike RX where resync is driven entirely by the core in TX only * the driver knows when things went out of sync, so we need the flag * to be atomic.
net/tls/tls_device.c +3 -7
@@ -680,15 +680,13 @@ static void tls_device_resync_rx(struct tls_context *tls_ctx,struct tls_offload_context_rx *rx_ctx =tls_offload_ctx_rx (tls_ctx);struct net_device *netdev; -if (WARN_ON (test_and_set_bit (TLS_RX_SYNC_RUNNING , &tls_ctx->flags))) -return ; -trace_tls_device_rx_resync_send (sk, seq, rcd_sn, rx_ctx->resync_type); +rcu_read_lock (); netdev =READ_ONCE (tls_ctx->netdev);if (netdev) netdev->tlsdev_ops->tls_dev_resync (netdev, sk, seq, rcd_sn,TLS_OFFLOAD_CTX_DIR_RX ); -clear_bit_unlock (TLS_RX_SYNC_RUNNING , &tls_ctx->flags); +rcu_read_unlock ();TLS_INC_STATS (sock_net (sk),LINUX_MIB_TLSRXDEVICERESYNC ); }
@@ -1300,9 +1298,7 @@ static int tls_device_down(struct net_device *netdev) netdev->tlsdev_ops->tls_dev_del (netdev, ctx,TLS_OFFLOAD_CTX_DIR_RX );WRITE_ONCE (ctx->netdev,NULL ); -smp_mb__before_atomic ();/* pairs with test_and_set_bit() */ -while (test_bit (TLS_RX_SYNC_RUNNING , &ctx->flags)) -usleep_range (10 ,200 ); +synchronize_net ();dev_put (netdev);list_del_init (&ctx->list);
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 33/161 net/tls: Fix use-after-free after the TLS device goes down and up
In reply to Greg Kroah-HartmanFrom: Maxim Mikityanskiy <[email protected]>
[ Upstream commit c55dcdd435aa6c6ad6ccac0a4c636d010ee367a4 ]
When a netdev with active TLS offload goes down, tls_device_down is called to stop the offload and tear down the TLS context. However, the socket stays alive, and it still points to the TLS context, which is now deallocated. If a netdev goes up, while the connection is still active, and the data flow resumes after a number of TCP retransmissions, it will lead to a use-after-free of the TLS context.
This commit addresses this bug by keeping the context alive until its normal destruction, and implements the necessary fallbacks, so that the connection can resume in software (non-offloaded) kTLS mode.
On the TX side tls_sw_fallback is used to encrypt all packets. The RX side already has all the necessary fallbacks, because receiving non-decrypted packets is supported. The thing needed on the RX side is to block resync requests, which are normally produced after receiving non-decrypted packets.
The necessary synchronization is implemented for a graceful teardown:
first the fallbacks are deployed, then the driver resources are released (it used to be possible to have a tls_dev_resync after tls_dev_del).
A new flag called TLS_RX_DEV_DEGRADED is added to indicate the fallback mode. It's used to skip the RX resync logic completely, as it becomes useless, and some objects may be released (for example, resync_async, which is allocated and freed by the driver).
--- include/net/tls.h | 9 ++++++ net/tls/tls_device.c | 52 +++++++++++++++++++++++++++++++---- net/tls/tls_device_fallback.c | 7 +++++ net/tls/tls_main.c | 1 + 4 files changed, 64 insertions(+), 5 deletions(-)
include/net/tls.h +9 -0
@@ -193,6 +193,11 @@ struct tls_offload_context_tx { (sizeof(struct tls_offload_context_tx) + TLS_DRIVER_STATE_SIZE_TX)enum tls_context_flags { +/* tls_device_down was called after the netdev went down, device state +* was released, and kTLS works in software, even though rx_conf is +* still TLS_HW (needed for transition). +*/ +TLS_RX_DEV_DEGRADED =0 ,/* Unlike RX where resync is driven entirely by the core in TX only * the driver knows when things went out of sync, so we need the flag * to be atomic.
@@ -265,6 +270,7 @@ struct tls_context {/* cache cold stuff */ struct proto *sk_proto; +struct sock *sk;void (*sk_destruct)(struct sock *sk);
@@ -447,6 +453,9 @@ static inline u16 tls_user_config(struct tls_context *ctx, bool tx)struct sk_buff *tls_validate_xmit_skb (struct sock *sk,struct net_device *dev,struct sk_buff *skb); +struct sk_buff * +tls_validate_xmit_skb_sw (struct sock *sk,struct net_device *dev, +struct sk_buff *skb);static inline bool tls_is_sk_tx_device_offloaded (struct sock *sk) {
net/tls/tls_device.c +47 -5
84 lines not on this page · raw message
net/tls/tls_device_fallback.c +7 -0
@@ -431,6 +431,13 @@ struct sk_buff *tls_validate_xmit_skb(struct sock *sk, }EXPORT_SYMBOL_GPL (tls_validate_xmit_skb); +struct sk_buff *tls_validate_xmit_skb_sw (struct sock *sk, +struct net_device *dev, +struct sk_buff *skb) +{ +return tls_sw_fallback (sk, skb); +} +struct sk_buff *tls_encrypt_skb (struct sk_buff *skb) {return tls_sw_fallback (skb->sk, skb);
net/tls/tls_main.c +1 -0
@@ -636,6 +636,7 @@ struct tls_context *tls_ctx_create(struct sock *sk)mutex_init (&ctx->tx_lock);rcu_assign_pointer (icsk->icsk_ulp_data, ctx); ctx->sk_proto =READ_ONCE (sk->sk_prot); + ctx->sk = sk;return ctx; }
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 34/161 net/mlx5e: Fix incompatible casting
In reply to Greg Kroah-HartmanFrom: Aya Levin <[email protected]>
[ Upstream commit d8ec92005f806dfa7524e9171eca707c0bb1267e ]
Device supports setting of a single fec mode at a time, enforce this by bitmap_weight == 1. Input from fec command is in u32, avoid cast to unsigned long and use bitmap_from_arr32 to populate bitmap safely.
--- drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/en_ethtool.c +3 -2
@@ -1632,12 +1632,13 @@ static int mlx5e_set_fecparam(struct net_device *netdev, {struct mlx5e_priv *priv =netdev_priv (netdev);struct mlx5_core_dev *mdev = priv->mdev; +unsigned long fec_bitmap;u16 fec_policy =0 ;int mode;int err; -if (bitmap_weight ((unsigned long *)&fecparam->fec, -ETHTOOL_FEC_LLRS_BIT +1 ) >1 ) +bitmap_from_arr32 (&fec_bitmap, &fecparam->fec,sizeof (fecparam->fec) *BITS_PER_BYTE ); +if (bitmap_weight (&fec_bitmap,ETHTOOL_FEC_LLRS_BIT +1 ) >1 )return -EOPNOTSUPP ;for (mode =0 ; mode <ARRAY_SIZE (pplm_fec_2_ethtool); mode++) {
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 35/161 net/mlx5: Check firmware sync reset requested is set before trying to abort it
In reply to Greg Kroah-HartmanFrom: Moshe Shemesh <[email protected]>
[ Upstream commit 5940e64281c09976ce2b560244217e610bf9d029 ]
In case driver sent NACK to firmware on sync reset request, it will get sync reset abort event while it didn't set sync reset requested mode. Thus, on abort sync reset event handler, driver should check reset requested is set before trying to stop sync reset poll.
--- drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/net/ethernet/mellanox/mlx5/core/fw_reset.c +3 -0
@@ -354,6 +354,9 @@ static void mlx5_sync_reset_abort_event(struct work_struct *work) reset_abort_work);struct mlx5_core_dev *dev = fw_reset->dev; +if (!test_bit (MLX5_FW_RESET_FLAGS_RESET_REQUESTED , &fw_reset->reset_flags)) +return ; +mlx5_sync_reset_clear_reset_requested (dev, true);mlx5_core_warn (dev,"PCI Sync FW Update Reset Aborted.\n" ); }
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 36/161 net/mlx5e: Check for needed capability for cvlan matching
In reply to Greg Kroah-HartmanFrom: Roi Dayan <[email protected]>
[ Upstream commit afe93f71b5d3cdae7209213ec8ef25210b837b93 ]
If not supported show an error and return instead of trying to offload to the hardware and fail.
--- drivers/net/ethernet/mellanox/mlx5/core/en_tc.c | 9 +++++++++ 1 file changed, 9 insertions(+)
drivers/net/ethernet/mellanox/mlx5/core/en_tc.c +9 -0
@@ -1964,11 +1964,13 @@ static int __parse_cls_flower(struct mlx5e_priv *priv, misc_parameters);struct flow_rule *rule =flow_cls_offload_flow_rule (f);struct flow_dissector *dissector = rule->match.dissector; +enum fs_flow_table_type fs_type;u16 addr_type =0 ;u8 ip_proto =0 ;u8 *match_level;int err; + fs_type =mlx5e_is_eswitch_flow (flow) ?FS_FT_FDB :FS_FT_NIC_RX ; match_level = outer_match_level; if (dissector->used_keys &
@@ -2093,6 +2095,13 @@ static int __parse_cls_flower(struct mlx5e_priv *priv,if (match.mask->vlan_id || match.mask->vlan_priority || match.mask->vlan_tpid) { +if (!MLX5_CAP_FLOWTABLE_TYPE (priv->mdev, ft_field_support.outer_second_vid, + fs_type)) { +NL_SET_ERR_MSG_MOD (extack, +"Matching on CVLAN is not supported" ); +return -EOPNOTSUPP ; + } +if (match.key->vlan_tpid ==htons (ETH_P_8021AD )) {MLX5_SET (fte_match_set_misc, misc_c, outer_second_svlan_tag,1 );
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 37/161 net/mlx5e: Fix adding encap rules to slow path
In reply to Greg Kroah-HartmanFrom: Roi Dayan <[email protected]>
[ Upstream commit 2a2c84facd4af661d71be6e81fd9d490ac7fdc53 ]
On some devices the ignore flow level cap is not supported and we shouldn't use it. Setting the dest ft with mlx5_chains_get_tc_end_ft() already gives the correct end ft if ignore flow level cap is supported or not.
--- drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c | 3 ++- drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c | 2 +- drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.h | 5 +++++ 3 files changed, 8 insertions(+), 2 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/eswitch_offloads.c +2 -1
@@ -349,7 +349,8 @@ esw_setup_slow_path_dest(struct mlx5_flow_destination *dest, struct mlx5_fs_chains *chains, int i) { - flow_act->flags |=FLOW_ACT_IGNORE_FLOW_LEVEL ; + if (mlx5_chains_ignore_flow_level_supported(chains)) + flow_act->flags |= FLOW_ACT_IGNORE_FLOW_LEVEL; dest[i].type =MLX5_FLOW_DESTINATION_TYPE_FLOW_TABLE ; dest[i].ft =mlx5_chains_get_tc_end_ft (chains); }
drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.c +1 -1
@@ -111,7 +111,7 @@ bool mlx5_chains_prios_supported(struct mlx5_fs_chains *chains)return chains->flags &MLX5_CHAINS_AND_PRIOS_SUPPORTED ; } -static bool mlx5_chains_ignore_flow_level_supported (struct mlx5_fs_chains *chains) +bool mlx5_chains_ignore_flow_level_supported (struct mlx5_fs_chains *chains) {return chains->flags &MLX5_CHAINS_IGNORE_FLOW_LEVEL_SUPPORTED ; }
drivers/net/ethernet/mellanox/mlx5/core/lib/fs_chains.h +5 -0
@@ -27,6 +27,7 @@ struct mlx5_chains_attr {bool mlx5_chains_prios_supported (struct mlx5_fs_chains *chains); +bool mlx5_chains_ignore_flow_level_supported (struct mlx5_fs_chains *chains);bool mlx5_chains_backwards_supported (struct mlx5_fs_chains *chains); u32
@@ -72,6 +73,10 @@ mlx5_chains_set_end_ft(struct mlx5_fs_chains *chains,#else /* CONFIG_MLX5_CLS_ACT */ +static inline bool +mlx5_chains_ignore_flow_level_supported (struct mlx5_fs_chains *chains) +{return false; } +static inline struct mlx5_flow_table *mlx5_chains_get_table (struct mlx5_fs_chains *chains,u32 chain,u32 prio,u32 level) {return ERR_PTR (-EOPNOTSUPP ); }
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 38/161 net/mlx5: DR, Create multi-destination flow table with level less than 64
In reply to Greg Kroah-HartmanFrom: Yevgeny Kliteynik <[email protected]>
[ Upstream commit 216214c64a8c1cb9078c2c0aec7bb4a2f8e75397 ]
Flow table that contains flow pointing to multiple flow tables or multiple TIRs must have a level lower than 64. In our case it applies to muli- destination flow table.
Fix the level of the created table to comply with HW Spec definitions, and still make sure that its level lower than SW-owned tables, so that it would be possible to point from the multi-destination FW table to SW tables.
--- drivers/net/ethernet/mellanox/mlx5/core/steering/dr_fw.c | 3 ++- include/linux/mlx5/mlx5_ifc.h | 2 ++ 2 files changed, 4 insertions(+), 1 deletion(-)
drivers/net/ethernet/mellanox/mlx5/core/steering/dr_fw.c +2 -1
@@ -112,7 +112,8 @@ int mlx5dr_fw_create_md_tbl(struct mlx5dr_domain *dmn,int ret; ft_attr.table_type =MLX5_FLOW_TABLE_TYPE_FDB ; - ft_attr.level = dmn->info.caps.max_ft_level -2 ; + ft_attr.level =min_t (int, dmn->info.caps.max_ft_level -2 , +MLX5_FT_MAX_MULTIPATH_LEVEL ); ft_attr.reformat_en = reformat_req; ft_attr.decap_en = reformat_req;
include/linux/mlx5/mlx5_ifc.h +2 -0
@@ -1260,6 +1260,8 @@ enum mlx5_fc_bulk_alloc_bitmask {#define MLX5_FC_BULK_NUM_FCS (fc_enum) (MLX5_FC_BULK_SIZE_FACTOR * (fc_enum)) +#define MLX5_FT_MAX_MULTIPATH_LEVEL 63 +enum {MLX5_STEERING_FORMAT_CONNECTX_5 =0 ,MLX5_STEERING_FORMAT_CONNECTX_6DX =1 ,
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 39/161 nvmet: fix freeing unallocated p2pmem
In reply to Greg Kroah-HartmanFrom: Max Gurtovoy <[email protected]>
[ Upstream commit bcd9a0797d73eeff659582f23277e7ab6e5f18f3 ]
In case p2p device was found but the p2p pool is empty, the nvme target is still trying to free the sgl from the p2p pool instead of the regular sgl pool and causing a crash (BUG() is called). Instead, assign the p2p_dev for the request only if it was allocated from p2p pool.
This is the crash that was caused:
[Sun May 30 19:13:53 2021] ------------[ cut here ]------------ [Sun May 30 19:13:53 2021] kernel BUG at lib/genalloc.c:518! [Sun May 30 19:13:53 2021] invalid opcode: 0000 [#1] SMP PTI ... [Sun May 30 19:13:53 2021] kernel BUG at lib/genalloc.c:518! ... [Sun May 30 19:13:53 2021] RIP: 0010:gen_pool_free_owner+0xa8/0xb0 ... [Sun May 30 19:13:53 2021] Call Trace: [Sun May 30 19:13:53 2021] ------------[ cut here ]------------ [Sun May 30 19:13:53 2021] pci_free_p2pmem+0x2b/0x70 [Sun May 30 19:13:53 2021] pci_p2pmem_free_sgl+0x4f/0x80 [Sun May 30 19:13:53 2021] nvmet_req_free_sgls+0x1e/0x80 [nvmet] [Sun May 30 19:13:53 2021] kernel BUG at lib/genalloc.c:518! [Sun May 30 19:13:53 2021] nvmet_rdma_release_rsp+0x4e/0x1f0 [nvmet_rdma] [Sun May 30 19:13:53 2021] nvmet_rdma_send_done+0x1c/0x60 [nvmet_rdma]
--- drivers/nvme/target/core.c | 33 ++++++++++++++++----------------- 1 file changed, 16 insertions(+), 17 deletions(-)
drivers/nvme/target/core.c +16 -17
@@ -999,19 +999,23 @@ static unsigned int nvmet_data_transfer_len(struct nvmet_req *req)return req->transfer_len - req->metadata_len; } -static int nvmet_req_alloc_p2pmem_sgls (struct nvmet_req *req) +static int nvmet_req_alloc_p2pmem_sgls (struct pci_dev *p2p_dev, +struct nvmet_req *req) { - req->sg =pci_p2pmem_alloc_sgl (req->p2p_dev, &req->sg_cnt, + req->sg =pci_p2pmem_alloc_sgl (p2p_dev, &req->sg_cnt,nvmet_data_transfer_len (req));if (!req->sg) goto out_err;if (req->metadata_len) { - req->metadata_sg =pci_p2pmem_alloc_sgl (req->p2p_dev, + req->metadata_sg =pci_p2pmem_alloc_sgl (p2p_dev, &req->metadata_sg_cnt, req->metadata_len);if (!req->metadata_sg) goto out_free_sg; } + + req->p2p_dev = p2p_dev; +return 0 ; out_free_sg:pci_p2pmem_free_sgl (req->p2p_dev, req->sg);
@@ -1019,25 +1023,19 @@ out_err:return -ENOMEM ; } -static bool nvmet_req_find_p2p_dev (struct nvmet_req *req) +static struct pci_dev *nvmet_req_find_p2p_dev (struct nvmet_req *req) { -if (!IS_ENABLED (CONFIG_PCI_P2PDMA )) -return false; - -if (req->sq->ctrl && req->sq->qid && req->ns) { - req->p2p_dev =radix_tree_lookup (&req->sq->ctrl->p2p_ns_map, - req->ns->nsid); -if (req->p2p_dev) -return true; - } - - req->p2p_dev =NULL ; -return false; +if (!IS_ENABLED (CONFIG_PCI_P2PDMA ) || + !req->sq->ctrl || !req->sq->qid || !req->ns) +return NULL ; +return radix_tree_lookup (&req->sq->ctrl->p2p_ns_map, req->ns->nsid); }int nvmet_req_alloc_sgls (struct nvmet_req *req) { -if (nvmet_req_find_p2p_dev (req) && !nvmet_req_alloc_p2pmem_sgls (req)) +struct pci_dev *p2p_dev =nvmet_req_find_p2p_dev (req); + +if (p2p_dev && !nvmet_req_alloc_p2pmem_sgls (p2p_dev, req))return 0 ; req->sg = sgl_alloc(nvmet_data_transfer_len(req), GFP_KERNEL,
@@ -1066,6 +1064,7 @@ void nvmet_req_free_sgls(struct nvmet_req *req)pci_p2pmem_free_sgl (req->p2p_dev, req->sg);if (req->metadata_sg)pci_p2pmem_free_sgl (req->p2p_dev, req->metadata_sg); + req->p2p_dev =NULL ; }else {sgl_free (req->sg);if (req->metadata_sg)
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 40/161 netfilter: nft_ct: skip expectations for confirmed conntrack
In reply to Greg Kroah-HartmanFrom: Pablo Neira Ayuso <[email protected]>
[ Upstream commit 1710eb913bdcda3917f44d383c32de6bdabfc836 ]
nft_ct_expect_obj_eval() calls nf_ct_ext_add() for a confirmed conntrack entry. However, nf_ct_ext_add() can only be called for !nf_ct_is_confirmed().
[ 1825.349056] WARNING: CPU: 0 PID: 1279 at net/netfilter/nf_conntrack_extend.c:48 nf_ct_xt_add+0x18e/0x1a0 [nf_conntrack] [ 1825.351391] RIP: 0010:nf_ct_ext_add+0x18e/0x1a0 [nf_conntrack] [ 1825.351493] Code: 41 5c 41 5d 41 5e 41 5f c3 41 bc 0a 00 00 00 e9 15 ff ff ff ba 09 00 00 00 31 f6 4c 89 ff e8 69 6c 3d e9 eb 96 45 31 ed eb cd <0f> 0b e9 b1 fe ff ff e8 86 79 14 e9 eb bf 0f 1f 40 00 0f 1f 44 00 [ 1825.351721] RSP: 0018:ffffc90002e1f1e8 EFLAGS: 00010202 [ 1825.351790] RAX: 000000000000000e RBX: ffff88814f5783c0 RCX: ffffffffc0e4f887 [ 1825.351881] RDX: dffffc0000000000 RSI: 0000000000000008 RDI: ffff88814f578440 [ 1825.351971] RBP: 0000000000000000 R08: 0000000000000000 R09: ffff88814f578447 [ 1825.352060] R10: ffffed1029eaf088 R11: 0000000000000001 R12: ffff88814f578440 [ 1825.352150] R13: ffff8882053f3a00 R14: 0000000000000000 R15: 0000000000000a20 [ 1825.352240] FS: 00007f992261c900(0000) GS:ffff889faec00000(0000) knlGS:0000000000000000 [ 1825.352343] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 1825.352417] CR2: 000056070a4d1158 CR3: 000000015efe0000 CR4: 0000000000350ee0 [ 1825.352508] Call Trace: [ 1825.352544] nf_ct_helper_ext_add+0x10/0x60 [nf_conntrack] [ 1825.352641] nft_ct_expect_obj_eval+0x1b8/0x1e0 [nft_ct] [ 1825.352716] nft_do_chain+0x232/0x850 [nf_tables]
Add the ct helper extension only for unconfirmed conntrack. Skip rule evaluation if the ct helper extension does not exist. Thus, you can only create expectations from the first packet.
It should be possible to remove this limitation by adding a new action to attach a generic ct helper to the first packet. Then, use this ct helper extension from follow up packets to create the ct expectation.
While at it, add a missing check to skip the template conntrack too and remove check for IPCT_UNTRACK which is implicit to !ct.
--- net/netfilter/nft_ct.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
net/netfilter/nft_ct.c +1 -1
@@ -1216,7 +1216,7 @@ static void nft_ct_expect_obj_eval(struct nft_object *obj,struct nf_conn *ct; ct =nf_ct_get (pkt->skb, &ctinfo); -if (!ct || ctinfo ==IP_CT_UNTRACKED ) { +if (!ct ||nf_ct_is_confirmed (ct) ||nf_ct_is_template (ct)) { regs->verdict.code =NFT_BREAK ;return ; }
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 41/161 netfilter: nfnetlink_cthelper: hit EBUSY on updates if size mismatches
In reply to Greg Kroah-HartmanFrom: Pablo Neira Ayuso <[email protected]>
[ Upstream commit 8971ee8b087750a23f3cd4dc55bff2d0303fd267 ]
The private helper data size cannot be updated. However, updates that contain NFCTH_PRIV_DATA_LEN might bogusly hit EBUSY even if the size is the same.
--- net/netfilter/nfnetlink_cthelper.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
net/netfilter/nfnetlink_cthelper.c +6 -2
@@ -380,10 +380,14 @@ static int nfnl_cthelper_update(const struct nlattr * const tb[], struct nf_conntrack_helper *helper) { + u32 size;int ret; -if (tb[NFCTH_PRIV_DATA_LEN ]) -return -EBUSY ; +if (tb[NFCTH_PRIV_DATA_LEN ]) { + size =ntohl (nla_get_be32 (tb[NFCTH_PRIV_DATA_LEN ])); +if (size != helper->data_len) +return -EBUSY ; + }if (tb[NFCTH_POLICY ]) { ret =nfnl_cthelper_update_policy (helper, tb[NFCTH_POLICY ]);
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 42/161 drm/i915/selftests: Fix return value check in live_breadcrumbs_smoketest()
In reply to Greg Kroah-HartmanFrom: Zhihao Cheng <[email protected]>
[ Upstream commit 10c1f0cbcea93beec5d3bdc02b1a3b577b4985e7 ]
In case of error, the function live_context() returns ERR_PTR() and never returns NULL. The NULL test in the return value check should be replaced with IS_ERR().
--- drivers/gpu/drm/i915/selftests/i915_request.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/i915/selftests/i915_request.c +2 -2
@@ -1392,8 +1392,8 @@ static int live_breadcrumbs_smoketest(void *arg)for (n =0 ; n < smoke[0 ].ncontexts; n++) { smoke[0 ].contexts[n] =live_context (i915, file); -if (!smoke[0 ].contexts[n]) { - ret = -ENOMEM ; +if (IS_ERR (smoke[0 ].contexts[n])) { + ret =PTR_ERR (smoke[0 ].contexts[n]); goto out_contexts; } }
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 43/161 bpf, lockdown, audit: Fix buggy SELinux lockdown permission checks
In reply to Greg Kroah-HartmanFrom: Daniel Borkmann <[email protected]>
[ Upstream commit ff40e51043af63715ab413995ff46996ecf9583f ]
Commit 59438b46471a ("security,lockdown,selinux: implement SELinux lockdown") added an implementation of the locked_down LSM hook to SELinux, with the aim to restrict which domains are allowed to perform operations that would breach lockdown. This is indirectly also getting audit subsystem involved to report events. The latter is problematic, as reported by Ondrej and Serhei, since it can bring down the whole system via audit:
1) The audit events that are triggered due to calls to security_locked_down() can OOM kill a machine, see below details [0].
2) It also seems to be causing a deadlock via avc_has_perm()/slow_avc_audit()
when trying to wake up kauditd, for example, when using trace_sched_switch()
tracepoint, see details in [1]. Triggering this was not via some hypothetical
corner case, but with existing tools like runqlat & runqslower from bcc, for
example, which make use of this tracepoint. Rough call sequence goes like:
rq_lock(rq) -> -------------------------+
trace_sched_switch() -> |
bpf_prog_xyz() -> +-> deadlock
selinux_lockdown() -> |
audit_log_end() -> |
wake_up_interruptible() -> |
try_to_wake_up() -> |
rq_lock(rq) --------------+What's worse is that the intention of 59438b46471a to further restrict lockdown settings for specific applications in respect to the global lockdown policy is completely broken for BPF. The SELinux policy rule for the current lockdown check looks something like this:
allow <who> <who> : lockdown { <reason> };
However, this doesn't match with the 'current' task where the security_locked_down() is executed, example: httpd does a syscall. There is a tracing program attached to the syscall which triggers a BPF program to run, which ends up doing a bpf_probe_read_kernel{,_str}() helper call. The selinux_lockdown() hook does the permission check against 'current', that is, httpd in this example. httpd has literally zero relation to this tracing program, and it would be nonsensical having to write an SELinux policy rule against httpd to let the tracing helper pass. The policy in this case needs to be against the entity that is installing the BPF program. For example, if bpftrace would generate a histogram of syscall counts by user space application:
bpftrace -e 'tracepoint:raw_syscalls:sys_enter { @[comm] = count(); }'
bpftrace would then go and generate a BPF program from this internally. One way of doing it [for the sake of the example] could be to call bpf_get_current_task() helper and then access current->comm via one of bpf_probe_read_kernel{,_str}() helpers. So the program itself has nothing to do with httpd or any other random app doing a syscall here. The BPF program _explicitly initiated_ the lockdown check. The allow/deny policy belongs in the context of bpftrace: meaning, you want to grant bpftrace access to use these helpers, but other tracers on the system like my_random_tracer _not_.
Therefore fix all three issues at the same time by taking a completely different approach for the security_locked_down() hook, that is, move the check into the program verification phase where we actually retrieve the BPF func proto. This also reliably gets the task (current) that is trying to install the BPF tracing program, e.g. bpftrace/bcc/perf/systemtap/etc, and it also fixes the OOM since we're moving this out of the BPF helper's fast-path which can be called several millions of times per second.
The check is then also in line with other security_locked_down() hooks in the system where the enforcement is performed at open/load time, for example, open_kcore() for /proc/kcore access or module_sig_check() for module signatures just to pick few random ones. What's out of scope in the fix as well as in other security_locked_down() hook locations /outside/ of BPF subsystem is that if the lockdown policy changes on the fly there is no retrospective action. This requires a different discussion, potentially complex infrastructure, and it's also not clear whether this can be solved generically. Either way, it is out of scope for a suitable stable fix which this one is targeting. Note that the breakage is specifically on 59438b46471a where it started to rely on 'current' as UAPI behavior, and _not_ earlier infrastructure such as 9d1f8be5cf42 ("bpf:
Restrict bpf when kernel lockdown is in confidentiality mode").
[0] https://bugzilla.redhat.com/show_bug.cgi?id=1955585, Jakub Hrozek says:
I starting seeing this with F-34. When I run a container that is traced with BPF to record the syscalls it is doing, auditd is flooded with messages like:
type=AVC msg=audit(1619784520.593:282387): avc: denied { confidentiality }
for pid=476 comm="auditd" lockdown_reason="use of bpf to read kernel RAM"
scontext=system_u:system_r:auditd_t:s0 tcontext=system_u:system_r:auditd_t:s0
tclass=lockdown permissive=0
This seems to be leading to auditd running out of space in the backlog buffer and eventually OOMs the machine.
[...]
auditd running at 99% CPU presumably processing all the messages, eventually I get:
Apr 30 12:20:42 fedora kernel: audit: backlog limit exceeded Apr 30 12:20:42 fedora kernel: audit: backlog limit exceeded Apr 30 12:20:42 fedora kernel: audit: audit_backlog=2152579 > audit_backlog_limit=64 Apr 30 12:20:42 fedora kernel: audit: audit_backlog=2152626 > audit_backlog_limit=64 Apr 30 12:20:42 fedora kernel: audit: audit_backlog=2152694 > audit_backlog_limit=64 Apr 30 12:20:42 fedora kernel: audit: audit_lost=6878426 audit_rate_limit=0 audit_backlog_limit=64 Apr 30 12:20:45 fedora kernel: oci-seccomp-bpf invoked oom-killer: gfp_mask=0x100cca(GFP_HIGHUSER_MOVABLE), order=0, oom_score_adj=-1000 Apr 30 12:20:45 fedora kernel: CPU: 0 PID: 13284 Comm: oci-seccomp-bpf Not tainted 5.11.12-300.fc34.x86_64 #1 Apr 30 12:20:45 fedora kernel: Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.13.0-2.fc32 04/01/2014 [...]
[1] https://lore.kernel.org/linux-audit/CANYvDQN7H5tVp47fbYcRasv4XF07eUbsDwT_eDCHXJUj43J7jQ@mail.gmail.com/,
Serhei Makarov says:
Upstream kernel 5.11.0-rc7 and later was found to deadlock during a bpf_probe_read_compat() call within a sched_switch tracepoint. The problem is reproducible with the reg_alloc3 testcase from SystemTap's BPF backend testsuite on x86_64 as well as the runqlat, runqslower tools from bcc on ppc64le. Example stack trace:
[...]
[ 730.868702] stack backtrace: [ 730.869590] CPU: 1 PID: 701 Comm: in:imjournal Not tainted, 5.12.0-0.rc2.20210309git144c79ef3353.166.fc35.x86_64 #1 [ 730.871605] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.13.0-2.fc32 04/01/2014 [ 730.873278] Call Trace: [ 730.873770] dump_stack+0x7f/0xa1 [ 730.874433] check_noncircular+0xdf/0x100 [ 730.875232] __lock_acquire+0x1202/0x1e10 [ 730.876031] ? __lock_acquire+0xfc0/0x1e10 [ 730.876844] lock_acquire+0xc2/0x3a0 [ 730.877551] ? __wake_up_common_lock+0x52/0x90 [ 730.878434] ? lock_acquire+0xc2/0x3a0 [ 730.879186] ? lock_is_held_type+0xa7/0x120 [ 730.880044] ? skb_queue_tail+0x1b/0x50 [ 730.880800] _raw_spin_lock_irqsave+0x4d/0x90 [ 730.881656] ? __wake_up_common_lock+0x52/0x90 [ 730.882532] __wake_up_common_lock+0x52/0x90 [ 730.883375] audit_log_end+0x5b/0x100 [ 730.884104] slow_avc_audit+0x69/0x90 [ 730.884836] avc_has_perm+0x8b/0xb0 [ 730.885532] selinux_lockdown+0xa5/0xd0 [ 730.886297] security_locked_down+0x20/0x40 [ 730.887133] bpf_probe_read_compat+0x66/0xd0 [ 730.887983] bpf_prog_250599c5469ac7b5+0x10f/0x820 [ 730.888917] trace_call_bpf+0xe9/0x240 [ 730.889672] perf_trace_run_bpf_submit+0x4d/0xc0 [ 730.890579] perf_trace_sched_switch+0x142/0x180 [ 730.891485] ? __schedule+0x6d8/0xb20 [ 730.892209] __schedule+0x6d8/0xb20 [ 730.892899] schedule+0x5b/0xc0 [ 730.893522] exit_to_user_mode_prepare+0x11d/0x240 [ 730.894457] syscall_exit_to_user_mode+0x27/0x70 [ 730.895361] entry_SYSCALL_64_after_hwframe+0x44/0xae [...]
--- kernel/bpf/helpers.c | 7 +++++-- kernel/trace/bpf_trace.c | 32 ++++++++++++-------------------- 2 files changed, 17 insertions(+), 22 deletions(-)
kernel/bpf/helpers.c +5 -2
@@ -14,6 +14,7 @@#include <linux/jiffies.h> #include <linux/pid_namespace.h> #include <linux/proc_ns.h> +#include <linux/security.h> #include "../../lib/kstrtox.h"
@@ -741,11 +742,13 @@ bpf_base_func_proto(enum bpf_func_id func_id)case BPF_FUNC_probe_read_user:return &bpf_probe_read_user_proto;case BPF_FUNC_probe_read_kernel: -return &bpf_probe_read_kernel_proto; +return security_locked_down (LOCKDOWN_BPF_READ ) <0 ? +NULL : &bpf_probe_read_kernel_proto;case BPF_FUNC_probe_read_user_str:return &bpf_probe_read_user_str_proto;case BPF_FUNC_probe_read_kernel_str: -return &bpf_probe_read_kernel_str_proto; +return security_locked_down (LOCKDOWN_BPF_READ ) <0 ? +NULL : &bpf_probe_read_kernel_str_proto;case BPF_FUNC_snprintf_btf:return &bpf_snprintf_btf_proto;default :
kernel/trace/bpf_trace.c +12 -20
@@ -215,16 +215,11 @@ const struct bpf_func_proto bpf_probe_read_user_str_proto = { static __always_inline intbpf_probe_read_kernel_common (void *dst,u32 size,const void *unsafe_ptr) { -int ret =security_locked_down (LOCKDOWN_BPF_READ ); +int ret; -if (unlikely (ret <0 )) - goto fail; ret =copy_from_kernel_nofault (dst, unsafe_ptr, size);if (unlikely (ret <0 )) - goto fail; -return ret; -fail: -memset (dst,0 , size); +memset (dst,0 , size);return ret; }
@@ -246,10 +241,7 @@ const struct bpf_func_proto bpf_probe_read_kernel_proto = { static __always_inline intbpf_probe_read_kernel_str_common (void *dst,u32 size,const void *unsafe_ptr) { -int ret =security_locked_down (LOCKDOWN_BPF_READ ); - -if (unlikely (ret <0 )) - goto fail; +int ret;/* * The strncpy_from_kernel_nofault() call will likely not fill the
@@ -262,11 +254,7 @@ bpf_probe_read_kernel_str_common(void *dst, u32 size, const void *unsafe_ptr)*/ ret =strncpy_from_kernel_nofault (dst, unsafe_ptr, size);if (unlikely (ret <0 )) - goto fail; - -return ret; -fail: -memset (dst,0 , size); +memset (dst,0 , size);return ret; }
@@ -1322,16 +1310,20 @@ bpf_tracing_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)case BPF_FUNC_probe_read_user:return &bpf_probe_read_user_proto;case BPF_FUNC_probe_read_kernel: -return &bpf_probe_read_kernel_proto; +return security_locked_down (LOCKDOWN_BPF_READ ) <0 ? +NULL : &bpf_probe_read_kernel_proto;case BPF_FUNC_probe_read_user_str:return &bpf_probe_read_user_str_proto;case BPF_FUNC_probe_read_kernel_str: -return &bpf_probe_read_kernel_str_proto; +return security_locked_down (LOCKDOWN_BPF_READ ) <0 ? +NULL : &bpf_probe_read_kernel_str_proto;#ifdef CONFIG_ARCH_HAS_NON_OVERLAPPING_ADDRESS_SPACE case BPF_FUNC_probe_read: -return &bpf_probe_read_compat_proto; +return security_locked_down (LOCKDOWN_BPF_READ ) <0 ? +NULL : &bpf_probe_read_compat_proto;case BPF_FUNC_probe_read_str: -return &bpf_probe_read_compat_str_proto; +return security_locked_down (LOCKDOWN_BPF_READ ) <0 ? +NULL : &bpf_probe_read_compat_str_proto;#endif #ifdef CONFIG_CGROUPS case BPF_FUNC_get_current_cgroup_id:
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 44/161 ieee802154: fix error return code in ieee802154_add_iface()
In reply to Greg Kroah-HartmanFrom: Zhen Lei <[email protected]>
[ Upstream commit 79c6b8ed30e54b401c873dbad2511f2a1c525fd5 ]
Fix to return a negative error code from the error handling case instead of 0, as done elsewhere in this function.
--- net/ieee802154/nl-phy.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
net/ieee802154/nl-phy.c +3 -1
@@ -241,8 +241,10 @@ int ieee802154_add_iface(struct sk_buff *skb, struct genl_info *info) }if (nla_put_string (msg,IEEE802154_ATTR_PHY_NAME ,wpan_phy_name (phy)) || -nla_put_string (msg,IEEE802154_ATTR_DEV_NAME , dev->name)) +nla_put_string (msg,IEEE802154_ATTR_DEV_NAME , dev->name)) { + rc = -EMSGSIZE ; goto nla_put_failure; + }dev_put (dev);wpan_phy_put (phy);
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 45/161 ieee802154: fix error return code in ieee802154_llsec_getparams()
In reply to Greg Kroah-HartmanFrom: Wei Yongjun <[email protected]>
[ Upstream commit 373e864cf52403b0974c2f23ca8faf9104234555 ]
Fix to return negative error code -ENOBUFS from the error handling case instead of 0, as done elsewhere in this function.
--- net/ieee802154/nl-mac.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
net/ieee802154/nl-mac.c +3 -1
@@ -680,8 +680,10 @@ int ieee802154_llsec_getparams(struct sk_buff *skb, struct genl_info *info)nla_put_u8 (msg,IEEE802154_ATTR_LLSEC_SECLEVEL , params.out_level) ||nla_put_u32 (msg,IEEE802154_ATTR_LLSEC_FRAME_COUNTER ,be32_to_cpu (params.frame_counter)) || - ieee802154_llsec_fill_key_id(msg, ¶ms.out_key)) + ieee802154_llsec_fill_key_id(msg, ¶ms.out_key)) { + rc = -ENOBUFS; goto out_free; + }dev_put (dev);
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 46/161 igb: Fix XDP with PTP enabled
In reply to Greg Kroah-HartmanFrom: Kurt Kanzenbach <[email protected]>
[ Upstream commit 5379260852b013902abbca691926b3ac1cac36d5 ]
When using native XDP with the igb driver, the XDP frame data doesn't point to the beginning of the packet. It's off by 16 bytes. Everything works as expected with XDP skb mode.
Actually these 16 bytes are used to store the packet timestamps. Therefore, pull the timestamp before executing any XDP operations and adjust all other code accordingly. The igc driver does it like that as well.
Tested with Intel i210 card and AF_XDP sockets.
--- drivers/net/ethernet/intel/igb/igb.h | 2 +- drivers/net/ethernet/intel/igb/igb_main.c | 45 +++++++++++++---------- drivers/net/ethernet/intel/igb/igb_ptp.c | 23 +++++------- 3 files changed, 37 insertions(+), 33 deletions(-)
drivers/net/ethernet/intel/igb/igb.h +1 -1
@@ -749,7 +749,7 @@ void igb_ptp_rx_hang(struct igb_adapter *adapter);void igb_ptp_tx_hang (struct igb_adapter *adapter);void igb_ptp_rx_rgtstamp (struct igb_q_vector *q_vector,struct sk_buff *skb);int igb_ptp_rx_pktstamp (struct igb_q_vector *q_vector,void *va, -struct sk_buff *skb); +ktime_t *timestamp);int igb_ptp_set_ts_config (struct net_device *netdev,struct ifreq *ifr);int igb_ptp_get_ts_config (struct net_device *netdev,struct ifreq *ifr);void igb_set_flag_queue_pairs (struct igb_adapter *,const u32 );
drivers/net/ethernet/intel/igb/igb_main.c +26 -19
94 lines not on this page · raw message
drivers/net/ethernet/intel/igb/igb_ptp.c +10 -13
@@ -856,30 +856,28 @@ static void igb_ptp_tx_hwtstamp(struct igb_adapter *adapter)dev_kfree_skb_any (skb); } -#define IGB_RET_PTP_DISABLED 1 -#define IGB_RET_PTP_INVALID 2 -/** * igb_ptp_rx_pktstamp - retrieve Rx per packet timestamp * @q_vector: Pointer to interrupt specific structure * @va: Pointer to address containing Rx buffer -* @skb: Buffer containing timestamp and packet +* @timestamp: Pointer where timestamp will be stored * * This function is meant to retrieve a timestamp from the first buffer of an * incoming frame. The value is stored in little endian format starting on * byte 8 * -* Returns: 0 if success, nonzero if failure +* Returns: The timestamp header length or 0 if not available **/ int igb_ptp_rx_pktstamp (struct igb_q_vector *q_vector,void *va, -struct sk_buff *skb) +ktime_t *timestamp) {struct igb_adapter *adapter = q_vector->adapter; +struct skb_shared_hwtstamps ts;__le64 *regval = (__le64 *)va;int adjust =0 ;if (!(adapter->ptp_flags &IGB_PTP_ENABLED )) -return IGB_RET_PTP_DISABLED ; +return 0 ;/* The timestamp is recorded in little endian format. * DWORD: 0 1 2 3
@@ -888,10 +886,9 @@ int igb_ptp_rx_pktstamp(struct igb_q_vector *q_vector, void *va,/* check reserved dwords are zero, be/le doesn't matter for zero */ if (regval[0 ]) -return IGB_RET_PTP_INVALID ; +return 0 ; -igb_ptp_systim_to_hwtstamp (adapter,skb_hwtstamps (skb), -le64_to_cpu (regval[1 ])); +igb_ptp_systim_to_hwtstamp (adapter, &ts,le64_to_cpu (regval[1 ]));/* adjust timestamp for the RX latency based on link speed */ if (adapter->hw.mac.type == e1000_i210) {
@@ -907,10 +904,10 @@ int igb_ptp_rx_pktstamp(struct igb_q_vector *q_vector, void *va,break ; } } -skb_hwtstamps (skb)->hwtstamp = -ktime_sub_ns (skb_hwtstamps (skb)->hwtstamp, adjust); -return 0 ; + *timestamp =ktime_sub_ns (ts.hwtstamp, adjust); + +return IGB_TS_HDR_LEN ; }/**
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 47/161 igb: add correct exception tracing for XDP
In reply to Greg Kroah-HartmanFrom: Magnus Karlsson <[email protected]>
[ Upstream commit 74431c40b9c5fa673fff83ec157a76a69efd5c72 ]
Add missing exception tracing to XDP when a number of different errors can occur. The support was only partial. Several errors where not logged which would confuse the user quite a lot not knowing where and why the packets disappeared.
--- drivers/net/ethernet/intel/igb/igb_main.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-)
drivers/net/ethernet/intel/igb/igb_main.c +6 -4
@@ -8395,18 +8395,20 @@ static struct sk_buff *igb_run_xdp(struct igb_adapter *adapter,break ;case XDP_TX : result =igb_xdp_xmit_back (adapter, xdp); +if (result ==IGB_XDP_CONSUMED ) + goto out_failure;break ;case XDP_REDIRECT : err =xdp_do_redirect (adapter->netdev, xdp, xdp_prog); -if (!err) - result =IGB_XDP_REDIR ; -else - result =IGB_XDP_CONSUMED ; +if (err) + goto out_failure; + result =IGB_XDP_REDIR ;break ;default :bpf_warn_invalid_xdp_action (act); fallthrough;case XDP_ABORTED : +out_failure:trace_xdp_exception (rx_ring->netdev, xdp_prog, act); fallthrough;case XDP_DROP :
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 48/161 ixgbevf: add correct exception tracing for XDP
In reply to Greg Kroah-HartmanFrom: Magnus Karlsson <[email protected]>
[ Upstream commit faae81420d162551b6ef2d804aafc00f4cd68e0e ]
Add missing exception tracing to XDP when a number of different errors can occur. The support was only partial. Several errors where not logged which would confuse the user quite a lot not knowing where and why the packets disappeared.
--- drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c +3 -0
@@ -1067,11 +1067,14 @@ static struct sk_buff *ixgbevf_run_xdp(struct ixgbevf_adapter *adapter,case XDP_TX : xdp_ring = adapter->xdp_ring[rx_ring->queue_index]; result =ixgbevf_xmit_xdp_ring (xdp_ring, xdp); +if (result ==IXGBEVF_XDP_CONSUMED ) + goto out_failure;break ;default :bpf_warn_invalid_xdp_action (act); fallthrough;case XDP_ABORTED : +out_failure:trace_xdp_exception (rx_ring->netdev, xdp_prog, act); fallthrough;/* handle aborts by dropping packet */ case XDP_DROP :
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 49/161 ice: track AF_XDP ZC enabled queues in bitmap
In reply to Greg Kroah-HartmanFrom: Maciej Fijalkowski <[email protected]>
[ Upstream commit e102db780e1c14f10c70dafa7684af22a745b51d ]
Commit c7a219048e45 ("ice: Remove xsk_buff_pool from VSI structure") silently introduced a regression and broke the Tx side of AF_XDP in copy mode. xsk_pool on ice_ring is set only based on the existence of the XDP prog on the VSI which in turn picks ice_clean_tx_irq_zc to be executed. That is not something that should happen for copy mode as it should use the regular data path ice_clean_tx_irq.
This results in a following splat when xdpsock is run in txonly or l2fwd scenarios in copy mode:
<snip>
[ 106.050195] BUG: kernel NULL pointer dereference, address: 0000000000000030 [ 106.057269] #PF: supervisor read access in kernel mode [ 106.062493] #PF: error_code(0x0000) - not-present page [ 106.067709] PGD 0 P4D 0 [ 106.070293] Oops: 0000 [#1] PREEMPT SMP NOPTI [ 106.074721] CPU: 61 PID: 0 Comm: swapper/61 Not tainted 5.12.0-rc2+ #45 [ 106.081436] Hardware name: Intel Corporation S2600WFT/S2600WFT, BIOS SE5C620.86B.02.01.0008.031920191559 03/19/2019 [ 106.092027] RIP: 0010:xp_raw_get_dma+0x36/0x50 [ 106.096551] Code: 74 14 48 b8 ff ff ff ff ff ff 00 00 48 21 f0 48 c1 ee 30 48 01 c6 48 8b 87 90 00 00 00 48 89 f2 81 e6 ff 0f 00 00 48 c1 ea 0c <48> 8b 04 d0 48 83 e0 fe 48 01 f0 c3 66 66 2e 0f 1f 84 00 00 00 00 [ 106.115588] RSP: 0018:ffffc9000d694e50 EFLAGS: 00010206 [ 106.120893] RAX: 0000000000000000 RBX: ffff88984b8c8a00 RCX: ffff889852581800 [ 106.128137] RDX: 0000000000000006 RSI: 0000000000000000 RDI: ffff88984cd8b800 [ 106.135383] RBP: ffff888123b50001 R08: ffff889896800000 R09: 0000000000000800 [ 106.142628] R10: 0000000000000000 R11: ffffffff826060c0 R12: 00000000000000ff [ 106.149872] R13: 0000000000000000 R14: 0000000000000040 R15: ffff888123b50018 [ 106.157117] FS: 0000000000000000(0000) GS:ffff8897e0f40000(0000) knlGS:0000000000000000 [ 106.165332] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 106.171163] CR2: 0000000000000030 CR3: 000000000560a004 CR4: 00000000007706e0 [ 106.178408] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [ 106.185653] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [ 106.192898] PKRU: 55555554 [ 106.195653] Call Trace: [ 106.198143] <IRQ> [ 106.200196] ice_clean_tx_irq_zc+0x183/0x2a0 [ice] [ 106.205087] ice_napi_poll+0x3e/0x590 [ice] [ 106.209356] __napi_poll+0x2a/0x160 [ 106.212911] net_rx_action+0xd6/0x200 [ 106.216634] __do_softirq+0xbf/0x29b [ 106.220274] irq_exit_rcu+0x88/0xc0 [ 106.223819] common_interrupt+0x7b/0xa0 [ 106.227719] </IRQ> [ 106.229857] asm_common_interrupt+0x1e/0x40
</snip>
Fix this by introducing the bitmap of queues that are zero-copy enabled, where each bit, corresponding to a queue id that xsk pool is being configured on, will be set/cleared within ice_xsk_pool_{en,dis}able and checked within ice_xsk_pool(). The latter is a function used for deciding which napi poll routine is executed.
Idea is being taken from our other drivers such as i40e and ixgbe.
--- drivers/net/ethernet/intel/ice/ice.h | 8 +++++--- drivers/net/ethernet/intel/ice/ice_lib.c | 10 ++++++++++ drivers/net/ethernet/intel/ice/ice_xsk.c | 3 +++ 3 files changed, 18 insertions(+), 3 deletions(-)
drivers/net/ethernet/intel/ice/ice.h +5 -3
@@ -325,6 +325,7 @@ struct ice_vsi {struct ice_tc_cfg tc_cfg;struct bpf_prog *xdp_prog;struct ice_ring **xdp_rings;/* XDP ring array */ +unsigned long *af_xdp_zc_qps;/* tracks AF_XDP ZC enabled qps */ u16 num_xdp_txq;/* Used XDP queues */ u8 xdp_mapping_mode;/* ICE_MAP_MODE_[CONTIG|SCATTER] */
@@ -534,15 +535,16 @@ static inline void ice_set_ring_xdp(struct ice_ring *ring)*/ static inline struct xsk_buff_pool *ice_xsk_pool (struct ice_ring *ring) { +struct ice_vsi *vsi = ring->vsi;u16 qid = ring->q_index;if (ice_ring_is_xdp (ring)) - qid -= ring->vsi->num_xdp_txq; + qid -= vsi->num_xdp_txq; -if (!ice_is_xdp_ena_vsi (ring->vsi)) +if (!ice_is_xdp_ena_vsi (vsi) || !test_bit (qid, vsi->af_xdp_zc_qps))return NULL ; -return xsk_get_pool_from_qid (ring->vsi->netdev, qid); +return xsk_get_pool_from_qid (vsi->netdev, qid); }/**
drivers/net/ethernet/intel/ice/ice_lib.c +10 -0
@@ -105,8 +105,14 @@ static int ice_vsi_alloc_arrays(struct ice_vsi *vsi)if (!vsi->q_vectors) goto err_vectors; + vsi->af_xdp_zc_qps =bitmap_zalloc (max_t (int, vsi->alloc_txq, vsi->alloc_rxq),GFP_KERNEL ); +if (!vsi->af_xdp_zc_qps) + goto err_zc_qps; +return 0 ; +err_zc_qps: +devm_kfree (dev, vsi->q_vectors); err_vectors:devm_kfree (dev, vsi->rxq_map); err_rxq_map:
@@ -286,6 +292,10 @@ static void ice_vsi_free_arrays(struct ice_vsi *vsi) dev =ice_pf_to_dev (pf); +if (vsi->af_xdp_zc_qps) { +bitmap_free (vsi->af_xdp_zc_qps); + vsi->af_xdp_zc_qps =NULL ; + }/* free the ring and vector containers */ if (vsi->q_vectors) {devm_kfree (dev, vsi->q_vectors);
drivers/net/ethernet/intel/ice/ice_xsk.c +3 -0
@@ -273,6 +273,7 @@ static int ice_xsk_pool_disable(struct ice_vsi *vsi, u16 qid)if (!pool)return -EINVAL ; +clear_bit (qid, vsi->af_xdp_zc_qps);xsk_pool_dma_unmap (pool,ICE_RX_DMA_ATTR );return 0 ;
@@ -303,6 +304,8 @@ ice_xsk_pool_enable(struct ice_vsi *vsi, struct xsk_buff_pool *pool, u16 qid)if (err)return err; +set_bit (qid, vsi->af_xdp_zc_qps); +return 0 ; }
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 50/161 cxgb4: fix regression with HASH tc prio value update
In reply to Greg Kroah-HartmanFrom: Rahul Lakkireddy <[email protected]>
[ Upstream commit a27fb314cba8cb84cd6456a4699c3330a83c326d ]
commit db43b30cd89c ("cxgb4: add ethtool n-tuple filter deletion") has moved searching for next highest priority HASH filter rule to cxgb4_flow_rule_destroy(), which searches the rhashtable before the the rule is removed from it and hence always finds at least 1 entry. Fix by removing the rule from rhashtable first before calling cxgb4_flow_rule_destroy() and hence avoid fetching stale info.
--- .../net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c | 14 +++++--------- 1 file changed, 5 insertions(+), 9 deletions(-)
drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_flower.c +5 -9
@@ -997,20 +997,16 @@ int cxgb4_tc_flower_destroy(struct net_device *dev,if (!ch_flower)return -ENOENT ; +rhashtable_remove_fast (&adap->flower_tbl, &ch_flower->node, + adap->flower_ht_params); + ret =cxgb4_flow_rule_destroy (dev, ch_flower->fs.tc_prio, &ch_flower->fs, ch_flower->filter_id);if (ret) - goto err; +netdev_err (dev,"Flow rule destroy failed for tid: %u, ret: %d" , + ch_flower->filter_id, ret); - ret =rhashtable_remove_fast (&adap->flower_tbl, &ch_flower->node, - adap->flower_ht_params); -if (ret) { -netdev_err (dev,"Flow remove from rhashtable failed" ); - goto err; - }kfree_rcu (ch_flower, rcu); - -err:return ret; }
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 51/161 ipv6: Fix KASAN: slab-out-of-bounds Read in fib6_nh_flush_exceptions
In reply to Greg Kroah-HartmanFrom: Coco Li <[email protected]>
[ Upstream commit 821bbf79fe46a8b1d18aa456e8ed0a3c208c3754 ]
Reported by syzbot:
HEAD commit: 90c911ad Merge tag 'fixes' of git://git.kernel.org/pub/scm.. git tree: git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master dashboard link: https://syzkaller.appspot.com/bug?extid=123aa35098fd3c000eb7 compiler: Debian clang version 11.0.1-2
==================================================================
BUG: KASAN: slab-out-of-bounds in fib6_nh_get_excptn_bucket net/ipv6/route.c:1604 [inline]
BUG: KASAN: slab-out-of-bounds in fib6_nh_flush_exceptions+0xbd/0x360 net/ipv6/route.c:1732
Read of size 8 at addr ffff8880145c78f8 by task syz-executor.4/17760
CPU: 0 PID: 17760 Comm: syz-executor.4 Not tainted 5.12.0-rc8-syzkaller #0
Call Trace:
<IRQ>
__dump_stack lib/dump_stack.c:79 [inline]
dump_stack+0x202/0x31e lib/dump_stack.c:120
print_address_description+0x5f/0x3b0 mm/kasan/report.c:232
__kasan_report mm/kasan/report.c:399 [inline]
kasan_report+0x15c/0x200 mm/kasan/report.c:416
fib6_nh_get_excptn_bucket net/ipv6/route.c:1604 [inline]
fib6_nh_flush_exceptions+0xbd/0x360 net/ipv6/route.c:1732
fib6_nh_release+0x9a/0x430 net/ipv6/route.c:3536
fib6_info_destroy_rcu+0xcb/0x1c0 net/ipv6/ip6_fib.c:174
rcu_do_batch kernel/rcu/tree.c:2559 [inline]
rcu_core+0x8f6/0x1450 kernel/rcu/tree.c:2794
__do_softirq+0x372/0x7a6 kernel/softirq.c:345
invoke_softirq kernel/softirq.c:221 [inline]
__irq_exit_rcu+0x22c/0x260 kernel/softirq.c:422
irq_exit_rcu+0x5/0x20 kernel/softirq.c:434
sysvec_apic_timer_interrupt+0x91/0xb0 arch/x86/kernel/apic/apic.c:1100
</IRQ>
asm_sysvec_apic_timer_interrupt+0x12/0x20 arch/x86/include/asm/idtentry.h:632
RIP: 0010:lock_acquire+0x1f6/0x720 kernel/locking/lockdep.c:5515
Code: f6 84 24 a1 00 00 00 02 0f 85 8d 02 00 00 f7 c3 00 02 00 00 49 bd 00 00 00 00 00 fc ff df 74 01 fb 48 c7 44 24 40 0e 36 e0 45 <4b> c7 44 3d 00 00 00 00 00 4b c7 44 3d 09 00 00 00 00 43 c7 44 3d
RSP: 0018:ffffc90009e06560 EFLAGS: 00000206
RAX: 1ffff920013c0cc0 RBX: 0000000000000246 RCX: dffffc0000000000
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
RBP: ffffc90009e066e0 R08: dffffc0000000000 R09: fffffbfff1f992b1
R10: fffffbfff1f992b1 R11: 0000000000000000 R12: 0000000000000000
R13: dffffc0000000000 R14: 0000000000000000 R15: 1ffff920013c0cb4
rcu_lock_acquire+0x2a/0x30 include/linux/rcupdate.h:267
rcu_read_lock include/linux/rcupdate.h:656 [inline]
ext4_get_group_info+0xea/0x340 fs/ext4/ext4.h:3231
ext4_mb_prefetch+0x123/0x5d0 fs/ext4/mballoc.c:2212
ext4_mb_regular_allocator+0x8a5/0x28f0 fs/ext4/mballoc.c:2379
ext4_mb_new_blocks+0xc6e/0x24f0 fs/ext4/mballoc.c:4982
ext4_ext_map_blocks+0x2be3/0x7210 fs/ext4/extents.c:4238
ext4_map_blocks+0xab3/0x1cb0 fs/ext4/inode.c:638
ext4_getblk+0x187/0x6c0 fs/ext4/inode.c:848
ext4_bread+0x2a/0x1c0 fs/ext4/inode.c:900
ext4_append+0x1a4/0x360 fs/ext4/namei.c:67
ext4_init_new_dir+0x337/0xa10 fs/ext4/namei.c:2768
ext4_mkdir+0x4b8/0xc00 fs/ext4/namei.c:2814
vfs_mkdir+0x45b/0x640 fs/namei.c:3819
ovl_do_mkdir fs/overlayfs/overlayfs.h:161 [inline]
ovl_mkdir_real+0x53/0x1a0 fs/overlayfs/dir.c:146
ovl_create_real+0x280/0x490 fs/overlayfs/dir.c:193
ovl_workdir_create+0x425/0x600 fs/overlayfs/super.c:788
ovl_make_workdir+0xed/0x1140 fs/overlayfs/super.c:1355
ovl_get_workdir fs/overlayfs/super.c:1492 [inline]
ovl_fill_super+0x39ee/0x5370 fs/overlayfs/super.c:2035
mount_nodev+0x52/0xe0 fs/super.c:1413
legacy_get_tree+0xea/0x180 fs/fs_context.c:592
vfs_get_tree+0x86/0x270 fs/super.c:1497
do_new_mount fs/namespace.c:2903 [inline]
path_mount+0x196f/0x2be0 fs/namespace.c:3233
do_mount fs/namespace.c:3246 [inline]
__do_sys_mount fs/namespace.c:3454 [inline]
__se_sys_mount+0x2f9/0x3b0 fs/namespace.c:3431
do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46
entry_SYSCALL_64_after_hwframe+0x44/0xae
RIP: 0033:0x4665f9
Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 bc ff ff ff f7 d8 64 89 01 48
RSP: 002b:00007f68f2b87188 EFLAGS: 00000246 ORIG_RAX: 00000000000000a5
RAX: ffffffffffffffda RBX: 000000000056bf60 RCX: 00000000004665f9
RDX: 00000000200000c0 RSI: 0000000020000000 RDI: 000000000040000a
RBP: 00000000004bfbb9 R08: 0000000020000100 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 000000000056bf60
R13: 00007ffe19002dff R14: 00007f68f2b87300 R15: 0000000000022000
Allocated by task 17768:
kasan_save_stack mm/kasan/common.c:38 [inline]
kasan_set_track mm/kasan/common.c:46 [inline]
set_alloc_info mm/kasan/common.c:427 [inline]
____kasan_kmalloc+0xc2/0xf0 mm/kasan/common.c:506
kasan_kmalloc include/linux/kasan.h:233 [inline]
__kmalloc+0xb4/0x380 mm/slub.c:4055
kmalloc include/linux/slab.h:559 [inline]
kzalloc include/linux/slab.h:684 [inline]
fib6_info_alloc+0x2c/0xd0 net/ipv6/ip6_fib.c:154
ip6_route_info_create+0x55d/0x1a10 net/ipv6/route.c:3638
ip6_route_add+0x22/0x120 net/ipv6/route.c:3728
inet6_rtm_newroute+0x2cd/0x2260 net/ipv6/route.c:5352
rtnetlink_rcv_msg+0xb34/0xe70 net/core/rtnetlink.c:5553
netlink_rcv_skb+0x1f0/0x460 net/netlink/af_netlink.c:2502
netlink_unicast_kernel net/netlink/af_netlink.c:1312 [inline]
netlink_unicast+0x7de/0x9b0 net/netlink/af_netlink.c:1338
netlink_sendmsg+0xaa6/0xe90 net/netlink/af_netlink.c:1927
sock_sendmsg_nosec net/socket.c:654 [inline]
sock_sendmsg net/socket.c:674 [inline]
____sys_sendmsg+0x5a2/0x900 net/socket.c:2350
___sys_sendmsg net/socket.c:2404 [inline]
__sys_sendmsg+0x319/0x400 net/socket.c:2433
do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46
entry_SYSCALL_64_after_hwframe+0x44/0xae
Last potentially related work creation:
kasan_save_stack+0x27/0x50 mm/kasan/common.c:38
kasan_record_aux_stack+0xee/0x120 mm/kasan/generic.c:345
__call_rcu kernel/rcu/tree.c:3039 [inline]
call_rcu+0x1b1/0xa30 kernel/rcu/tree.c:3114
fib6_info_release include/net/ip6_fib.h:337 [inline]
ip6_route_info_create+0x10c4/0x1a10 net/ipv6/route.c:3718
ip6_route_add+0x22/0x120 net/ipv6/route.c:3728
inet6_rtm_newroute+0x2cd/0x2260 net/ipv6/route.c:5352
rtnetlink_rcv_msg+0xb34/0xe70 net/core/rtnetlink.c:5553
netlink_rcv_skb+0x1f0/0x460 net/netlink/af_netlink.c:2502
netlink_unicast_kernel net/netlink/af_netlink.c:1312 [inline]
netlink_unicast+0x7de/0x9b0 net/netlink/af_netlink.c:1338
netlink_sendmsg+0xaa6/0xe90 net/netlink/af_netlink.c:1927
sock_sendmsg_nosec net/socket.c:654 [inline]
sock_sendmsg net/socket.c:674 [inline]
____sys_sendmsg+0x5a2/0x900 net/socket.c:2350
___sys_sendmsg net/socket.c:2404 [inline]
__sys_sendmsg+0x319/0x400 net/socket.c:2433
do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46
entry_SYSCALL_64_after_hwframe+0x44/0xae
Second to last potentially related work creation:
kasan_save_stack+0x27/0x50 mm/kasan/common.c:38
kasan_record_aux_stack+0xee/0x120 mm/kasan/generic.c:345
insert_work+0x54/0x400 kernel/workqueue.c:1331
__queue_work+0x981/0xcc0 kernel/workqueue.c:1497
queue_work_on+0x111/0x200 kernel/workqueue.c:1524
queue_work include/linux/workqueue.h:507 [inline]
call_usermodehelper_exec+0x283/0x470 kernel/umh.c:433
kobject_uevent_env+0x1349/0x1730 lib/kobject_uevent.c:617
kvm_uevent_notify_change+0x309/0x3b0 arch/x86/kvm/../../../virt/kvm/kvm_main.c:4809
kvm_destroy_vm arch/x86/kvm/../../../virt/kvm/kvm_main.c:877 [inline]
kvm_put_kvm+0x9c/0xd10 arch/x86/kvm/../../../virt/kvm/kvm_main.c:920
kvm_vcpu_release+0x53/0x60 arch/x86/kvm/../../../virt/kvm/kvm_main.c:3120
__fput+0x352/0x7b0 fs/file_table.c:280
task_work_run+0x146/0x1c0 kernel/task_work.c:140
tracehook_notify_resume include/linux/tracehook.h:189 [inline]
exit_to_user_mode_loop kernel/entry/common.c:174 [inline]
exit_to_user_mode_prepare+0x10b/0x1e0 kernel/entry/common.c:208
__syscall_exit_to_user_mode_work kernel/entry/common.c:290 [inline]
syscall_exit_to_user_mode+0x26/0x70 kernel/entry/common.c:301
entry_SYSCALL_64_after_hwframe+0x44/0xae
The buggy address belongs to the object at ffff8880145c7800
which belongs to the cache kmalloc-192 of size 192
The buggy address is located 56 bytes to the right of
192-byte region [ffff8880145c7800, ffff8880145c78c0)
The buggy address belongs to the page:
page:ffffea00005171c0 refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x145c7
flags: 0xfff00000000200(slab)
raw: 00fff00000000200 ffffea00006474c0 0000000200000002 ffff888010c41a00
raw: 0000000000000000 0000000080100010 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected
Memory state around the buggy address:
ffff8880145c7780: fb fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc
ffff8880145c7800: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
>ffff8880145c7880: 00 00 00 00 fc fc fc fc fc fc fc fc fc fc fc fc
^
ffff8880145c7900: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
ffff8880145c7980: fb fb fb fb fb fb fb fb fc fc fc fc fc fc fc fc
==================================================================In the ip6_route_info_create function, in the case that the nh pointer is not NULL, the fib6_nh in fib6_info has not been allocated. Therefore, when trying to free fib6_info in this error case using fib6_info_release, the function will call fib6_info_destroy_rcu, which it will access fib6_nh_release(f6i->fib6_nh);
However, f6i->fib6_nh doesn't have any refcount yet given the lack of allocation causing the reported memory issue above.
Therefore, releasing the empty pointer directly instead would be the solution.
--- net/ipv6/route.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
net/ipv6/route.c +6 -2
@@ -3676,11 +3676,11 @@ static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg,if (nh) {if (rt->fib6_src.plen) {NL_SET_ERR_MSG (extack,"Nexthops can not be used with source routing" ); - goto out; + goto out_free; }if (!nexthop_get (nh)) {NL_SET_ERR_MSG (extack,"Nexthop has been deleted" ); - goto out; + goto out_free; } rt->nh = nh; fib6_nh =nexthop_fib6_nh (rt->nh);
@@ -3717,6 +3717,10 @@ static struct fib6_info *ip6_route_info_create(struct fib6_config *cfg, out:fib6_info_release (rt);return ERR_PTR (err); +out_free: +ip_fib_metrics_put (rt->fib6_metrics); +kfree (rt); +return ERR_PTR (err); } int ip6_route_add(struct fib6_config *cfg, gfp_t gfp_flags,
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 52/161 ice: Fix allowing VF to request more/less queues via virtchnl
In reply to Greg Kroah-HartmanFrom: Brett Creeley <[email protected]>
[ Upstream commit f0457690af56673cb0c47af6e25430389a149225 ]
Commit 12bb018c538c ("ice: Refactor VF reset") caused a regression that removes the ability for a VF to request a different amount of queues via VIRTCHNL_OP_REQUEST_QUEUES. This prevents VF drivers to either increase or decrease the number of queue pairs they are allocated. Fix this by using the variable vf->num_req_qs when determining the vf->num_vf_qs during VF VSI creation.
--- drivers/net/ethernet/intel/ice/ice_lib.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/net/ethernet/intel/ice/ice_lib.c +2 -0
@@ -198,6 +198,8 @@ static void ice_vsi_set_num_qs(struct ice_vsi *vsi, u16 vf_id)break ;case ICE_VSI_VF : vf = &pf->vf[vsi->vf_id]; +if (vf->num_req_qs) + vf->num_vf_qs = vf->num_req_qs; vsi->alloc_txq = vf->num_vf_qs; vsi->alloc_rxq = vf->num_vf_qs;/* pf->num_msix_per_vf includes (VF miscellaneous vector +
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 53/161 ice: Fix VFR issues for AVF drivers that expect ATQLEN cleared
In reply to Greg Kroah-HartmanFrom: Brett Creeley <[email protected]>
[ Upstream commit 8679f07a9922068b9b6be81b632f52cac45d1b91 ]
Some AVF drivers expect the VF_MBX_ATQLEN register to be cleared for any type of VFR/VFLR. Fix this by clearing the VF_MBX_ATQLEN register at the same time as VF_MBX_ARQLEN.
--- drivers/net/ethernet/intel/ice/ice_hw_autogen.h | 1 + drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c | 12 +++++++----- 2 files changed, 8 insertions(+), 5 deletions(-)
drivers/net/ethernet/intel/ice/ice_hw_autogen.h +1 -0
@@ -31,6 +31,7 @@#define PF_FW_ATQLEN_ATQOVFL_M BIT(29)#define PF_FW_ATQLEN_ATQCRIT_M BIT(30)#define VF_MBX_ARQLEN (_VF) (0x0022BC00 + ((_VF) * 4)) +#define VF_MBX_ATQLEN (_VF) (0x0022A800 + ((_VF) * 4))#define PF_FW_ATQLEN_ATQENABLE_M BIT(31)#define PF_FW_ATQT 0x00080400#define PF_MBX_ARQBAH 0x0022E400
drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c +7 -5
@@ -435,13 +435,15 @@ static void ice_trigger_vf_reset(struct ice_vf *vf, bool is_vflr, bool is_pfr)*/ clear_bit (ICE_VF_STATE_INIT , vf->vf_states); -/* VF_MBX_ARQLEN is cleared by PFR, so the driver needs to clear it -* in the case of VFR. If this is done for PFR, it can mess up VF -* resets because the VF driver may already have started cleanup -* by the time we get here. +/* VF_MBX_ARQLEN and VF_MBX_ATQLEN are cleared by PFR, so the driver +* needs to clear them in the case of VFR/VFLR. If this is done for +* PFR, it can mess up VF resets because the VF driver may already +* have started cleanup by the time we get here. */ -if (!is_pfr) +if (!is_pfr) {wr32 (hw,VF_MBX_ARQLEN (vf->vf_id),0 ); +wr32 (hw,VF_MBX_ATQLEN (vf->vf_id),0 ); + }/* In the case of a VFLR, the HW has already reset the VF and we * just need to clean up, so don't hit the VFRTRIG register.
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 54/161 ice: handle the VF VSI rebuild failure
In reply to Greg Kroah-HartmanFrom: Haiyue Wang <[email protected]>
[ Upstream commit c7ee6ce1cf60b7fcdbdd2354d377d00bae3fa2d2 ]
VSI rebuild can be failed for LAN queue config, then the VF's VSI will be NULL, the VF reset should be stopped with the VF entering into the disable state.
--- drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/net/ethernet/intel/ice/ice_virtchnl_pf.c +6 -1
@@ -1377,7 +1377,12 @@ bool ice_reset_vf(struct ice_vf *vf, bool is_vflr) }ice_vf_pre_vsi_rebuild (vf); -ice_vf_rebuild_vsi_with_release (vf); + +if (ice_vf_rebuild_vsi_with_release (vf)) { +dev_err (dev,"Failed to release and setup the VF%u's VSI\n" , vf->vf_id); +return false; + } +ice_vf_post_vsi_rebuild (vf);return true;
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 55/161 ice: report supported and advertised autoneg using PHY capabilities
In reply to Greg Kroah-HartmanFrom: Paul Greenwalt <[email protected]>
[ Upstream commit 5cd349c349d6ec52862e550d3576893d35ab8ac2 ]
Ethtool incorrectly reported supported and advertised auto-negotiation settings for a backplane PHY image which did not support auto-negotiation. This can occur when using media or PHY type for reporting ethtool supported and advertised auto-negotiation settings.
Remove setting supported and advertised auto-negotiation settings based on PHY type in ice_phy_type_to_ethtool(), and MAC type in ice_get_link_ksettings().
Ethtool supported and advertised auto-negotiation settings should be based on the PHY image using the AQ command get PHY capabilities with media. Add setting supported and advertised auto-negotiation settings based get PHY capabilities with media in ice_get_link_ksettings().
--- drivers/net/ethernet/intel/ice/ice_ethtool.c | 51 +++----------------- 1 file changed, 6 insertions(+), 45 deletions(-)
drivers/net/ethernet/intel/ice/ice_ethtool.c +6 -45
@@ -1797,49 +1797,6 @@ ice_phy_type_to_ethtool(struct net_device *netdev,ice_ethtool_advertise_link_mode (ICE_AQ_LINK_SPEED_100GB , 100000baseKR4_Full); } - -/* Autoneg PHY types */ -if (phy_types_low &ICE_PHY_TYPE_LOW_100BASE_TX || - phy_types_low &ICE_PHY_TYPE_LOW_1000BASE_T || - phy_types_low &ICE_PHY_TYPE_LOW_1000BASE_KX || - phy_types_low &ICE_PHY_TYPE_LOW_2500BASE_T || - phy_types_low &ICE_PHY_TYPE_LOW_2500BASE_KX || - phy_types_low &ICE_PHY_TYPE_LOW_5GBASE_T || - phy_types_low &ICE_PHY_TYPE_LOW_5GBASE_KR || - phy_types_low &ICE_PHY_TYPE_LOW_10GBASE_T || - phy_types_low &ICE_PHY_TYPE_LOW_10GBASE_KR_CR1 || - phy_types_low &ICE_PHY_TYPE_LOW_25GBASE_T || - phy_types_low &ICE_PHY_TYPE_LOW_25GBASE_CR || - phy_types_low &ICE_PHY_TYPE_LOW_25GBASE_CR_S || - phy_types_low &ICE_PHY_TYPE_LOW_25GBASE_CR1 || - phy_types_low &ICE_PHY_TYPE_LOW_25GBASE_KR || - phy_types_low &ICE_PHY_TYPE_LOW_25GBASE_KR_S || - phy_types_low &ICE_PHY_TYPE_LOW_25GBASE_KR1 || - phy_types_low &ICE_PHY_TYPE_LOW_40GBASE_CR4 || - phy_types_low &ICE_PHY_TYPE_LOW_40GBASE_KR4 ) { -ethtool_link_ksettings_add_link_mode (ks, supported, - Autoneg); -ethtool_link_ksettings_add_link_mode (ks, advertising, - Autoneg); - } -if (phy_types_low &ICE_PHY_TYPE_LOW_50GBASE_CR2 || - phy_types_low &ICE_PHY_TYPE_LOW_50GBASE_KR2 || - phy_types_low &ICE_PHY_TYPE_LOW_50GBASE_CP || - phy_types_low &ICE_PHY_TYPE_LOW_50GBASE_KR_PAM4 ) { -ethtool_link_ksettings_add_link_mode (ks, supported, - Autoneg); -ethtool_link_ksettings_add_link_mode (ks, advertising, - Autoneg); - } -if (phy_types_low &ICE_PHY_TYPE_LOW_100GBASE_CR4 || - phy_types_low &ICE_PHY_TYPE_LOW_100GBASE_KR4 || - phy_types_low &ICE_PHY_TYPE_LOW_100GBASE_KR_PAM4 || - phy_types_low &ICE_PHY_TYPE_LOW_100GBASE_CP2 ) { -ethtool_link_ksettings_add_link_mode (ks, supported, - Autoneg); -ethtool_link_ksettings_add_link_mode (ks, advertising, - Autoneg); - } }#define TEST_SET_BITS_TIMEOUT 50
@@ -1996,9 +1953,7 @@ ice_get_link_ksettings(struct net_device *netdev, ks->base.port =PORT_TP ;break ;case ICE_MEDIA_BACKPLANE : -ethtool_link_ksettings_add_link_mode (ks, supported, Autoneg);ethtool_link_ksettings_add_link_mode (ks, supported, Backplane); -ethtool_link_ksettings_add_link_mode (ks, advertising, Autoneg);ethtool_link_ksettings_add_link_mode (ks, advertising, Backplane); ks->base.port =PORT_NONE ;
@@ -2073,6 +2028,12 @@ ice_get_link_ksettings(struct net_device *netdev,if (caps->link_fec_options &ICE_AQC_PHY_FEC_25G_RS_CLAUSE91_EN )ethtool_link_ksettings_add_link_mode (ks, supported,FEC_RS ); +/* Set supported and advertised autoneg */ +if (ice_is_phy_caps_an_enabled (caps)) { +ethtool_link_ksettings_add_link_mode (ks, supported, Autoneg); +ethtool_link_ksettings_add_link_mode (ks, advertising, Autoneg); + } + done:kfree (caps);return err;
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 56/161 ice: Allow all LLDP packets from PF to Tx
In reply to Greg Kroah-HartmanFrom: Dave Ertman <[email protected]>
[ Upstream commit f9f83202b7263ac371d616d6894a2c9ed79158ef ]
Currently in the ice driver, the check whether to allow a LLDP packet to egress the interface from the PF_VSI is being based on the SKB's priority field. It checks to see if the packets priority is equal to TC_PRIO_CONTROL. Injected LLDP packets do not always meet this condition.
SCAPY defaults to a sk_buff->protocol value of ETH_P_ALL (0x0003) and does not set the priority field. There will be other injection methods (even ones used by end users) that will not correctly configure the socket so that SKB fields are correctly populated.
Then ethernet header has to have to correct value for the protocol though.
Add a check to also allow packets whose ethhdr->h_proto matches ETH_P_LLDP (0x88CC).
--- drivers/net/ethernet/intel/ice/ice_txrx.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/net/ethernet/intel/ice/ice_txrx.c +4 -1
@@ -2331,6 +2331,7 @@ ice_xmit_frame_ring(struct sk_buff *skb, struct ice_ring *tx_ring)struct ice_tx_offload_params offload = {0 };struct ice_vsi *vsi = tx_ring->vsi;struct ice_tx_buf *first; +struct ethhdr *eth;unsigned int count;int tso, csum;
@@ -2377,7 +2378,9 @@ ice_xmit_frame_ring(struct sk_buff *skb, struct ice_ring *tx_ring) goto out_drop;/* allow CONTROL frames egress from main VSI if FW LLDP disabled */ -if (unlikely (skb->priority ==TC_PRIO_CONTROL && + eth = (struct ethhdr *)skb_mac_header (skb); +if (unlikely ((skb->priority ==TC_PRIO_CONTROL || + eth->h_proto ==htons (ETH_P_LLDP )) && vsi->type ==ICE_VSI_PF && vsi->port_info->qos_cfg.is_sw_lldp)) offload.cd_qw1 |= (u64)(ICE_TX_DESC_DTYPE_CTX |
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 57/161 i2c: qcom-geni: Add shutdown callback for i2c
In reply to Greg Kroah-HartmanFrom: Roja Rani Yarubandi <[email protected]>
[ Upstream commit 9f78c607600ce4f2a952560de26534715236f612 ]
If the hardware is still accessing memory after SMMU translation is disabled (as part of smmu shutdown callback), then the IOVAs (I/O virtual address) which it was using will go on the bus as the physical addresses which will result in unknown crashes like NoC/interconnect errors.
So, implement shutdown callback for i2c driver to suspend the bus during system "reboot" or "shutdown".
--- drivers/i2c/busses/i2c-qcom-geni.c | 9 +++++++++ 1 file changed, 9 insertions(+)
drivers/i2c/busses/i2c-qcom-geni.c +9 -0
@@ -650,6 +650,14 @@ static int geni_i2c_remove(struct platform_device *pdev)return 0 ; } +static void geni_i2c_shutdown (struct platform_device *pdev) +{ +struct geni_i2c_dev *gi2c =platform_get_drvdata (pdev); + +/* Make client i2c transfers start failing */ +i2c_mark_adapter_suspended (&gi2c->adap); +} +static int __maybe_unused geni_i2c_runtime_suspend (struct device *dev) {int ret;
@@ -714,6 +722,7 @@ MODULE_DEVICE_TABLE(of, geni_i2c_dt_match);static struct platform_driver geni_i2c_driver = { .probe = geni_i2c_probe, .remove = geni_i2c_remove, + .shutdown = geni_i2c_shutdown, .driver = { .name ="geni_i2c" , .pm = &geni_i2c_pm_ops,
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 58/161 sch_htb: fix refcount leak in htb_parent_to_leaf_offload
In reply to Greg Kroah-HartmanFrom: Yunjian Wang <[email protected]>
[ Upstream commit 944d671d5faa0d78980a3da5c0f04960ef1ad893 ]
The commit ae81feb7338c ("sch_htb: fix null pointer dereference on a null new_q") fixes a NULL pointer dereference bug, but it is not correct.
Because htb_graft_helper properly handles the case when new_q is NULL, and after the previous patch by skipping this call which creates an inconsistency : dev_queue->qdisc will still point to the old qdisc, but cl->parent->leaf.q will point to the new one (which will be noop_qdisc, because new_q was NULL). The code is based on an assumption that these two pointers are the same, so it can lead to refcount leaks.
The correct fix is to add a NULL pointer check to protect qdisc_refcount_inc inside htb_parent_to_leaf_offload.
--- net/sched/sch_htb.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
net/sched/sch_htb.c +4 -4
@@ -1488,7 +1488,8 @@ static void htb_parent_to_leaf_offload(struct Qdisc *sch,struct Qdisc *old_q;/* One ref for cl->leaf.q, the other for dev_queue->qdisc. */ -qdisc_refcount_inc (new_q); +if (new_q) +qdisc_refcount_inc (new_q); old_q =htb_graft_helper (dev_queue, new_q);WARN_ON (!(old_q->flags &TCQ_F_BUILTIN )); }
@@ -1675,10 +1676,9 @@ static int htb_delete(struct Qdisc *sch, unsigned long arg, cl->parent->common.classid, NULL);if (q->offload) { -if (new_q) { +if (new_q)htb_set_lockdep_class_child (new_q); -htb_parent_to_leaf_offload (sch, dev_queue, new_q); - } +htb_parent_to_leaf_offload (sch, dev_queue, new_q); } }
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 59/161 cxgb4: avoid link re-train during TC-MQPRIO configuration
In reply to Greg Kroah-HartmanFrom: Rahul Lakkireddy <[email protected]>
[ Upstream commit 3822d0670c9d4342794d73e0d0e615322b40438e ]
When configuring TC-MQPRIO offload, only turn off netdev carrier and don't bring physical link down in hardware. Otherwise, when the physical link is brought up again after configuration, it gets re-trained and stalls ongoing traffic.
Also, when firmware is no longer accessible or crashed, avoid sending FLOWC and waiting for reply that will never come.
Fix following hung_task_timeout_secs trace seen in these cases.
INFO: task tc:20807 blocked for more than 122 seconds.
Tainted: G S 5.13.0-rc3+ #122
"echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
task:tc state:D stack:14768 pid:20807 ppid: 19366 flags:0x00000000
Call Trace:
__schedule+0x27b/0x6a0
schedule+0x37/0xa0
schedule_preempt_disabled+0x5/0x10
__mutex_lock.isra.14+0x2a0/0x4a0
? netlink_lookup+0x120/0x1a0
? rtnl_fill_ifinfo+0x10f0/0x10f0
__netlink_dump_start+0x70/0x250
rtnetlink_rcv_msg+0x28b/0x380
? rtnl_fill_ifinfo+0x10f0/0x10f0
? rtnl_calcit.isra.42+0x120/0x120
netlink_rcv_skb+0x4b/0xf0
netlink_unicast+0x1a0/0x280
netlink_sendmsg+0x216/0x440
sock_sendmsg+0x56/0x60
__sys_sendto+0xe9/0x150
? handle_mm_fault+0x6d/0x1b0
? do_user_addr_fault+0x1c5/0x620
__x64_sys_sendto+0x1f/0x30
do_syscall_64+0x3c/0x80
entry_SYSCALL_64_after_hwframe+0x44/0xae
RIP: 0033:0x7f7f73218321
RSP: 002b:00007ffd19626208 EFLAGS: 00000246 ORIG_RAX: 000000000000002c
RAX: ffffffffffffffda RBX: 000055b7c0a8b240 RCX: 00007f7f73218321
RDX: 0000000000000028 RSI: 00007ffd19626210 RDI: 0000000000000003
RBP: 000055b7c08680ff R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 000055b7c085f5f6
R13: 000055b7c085f60a R14: 00007ffd19636470 R15: 00007ffd196262a0--- drivers/net/ethernet/chelsio/cxgb4/cxgb4.h | 2 -- drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c | 4 ++-- drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c | 9 ++++++--- drivers/net/ethernet/chelsio/cxgb4/sge.c | 6 ++++++ 4 files changed, 14 insertions(+), 7 deletions(-)
drivers/net/ethernet/chelsio/cxgb4/cxgb4.h +0 -2
@@ -2177,8 +2177,6 @@ int cxgb4_update_mac_filt(struct port_info *pi, unsigned int viid, bool persistent, u8 *smt_idx);int cxgb4_get_msix_idx_from_bmap (struct adapter *adap);void cxgb4_free_msix_idx_in_bmap (struct adapter *adap,u32 msix_idx); -int cxgb_open (struct net_device *dev); -int cxgb_close (struct net_device *dev);void cxgb4_enable_rx (struct adapter *adap,struct sge_rspq *q);void cxgb4_quiesce_rx (struct sge_rspq *q);int cxgb4_port_mirror_alloc (struct net_device *dev);
drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c +2 -2
@@ -2834,7 +2834,7 @@ static void cxgb_down(struct adapter *adapter)/* * net_device operations */ -int cxgb_open (struct net_device *dev) +static int cxgb_open (struct net_device *dev) {struct port_info *pi =netdev_priv (dev);struct adapter *adapter = pi->adapter;
@@ -2882,7 +2882,7 @@ out_unlock:return err; } -int cxgb_close (struct net_device *dev) +static int cxgb_close (struct net_device *dev) {struct port_info *pi =netdev_priv (dev);struct adapter *adapter = pi->adapter;
drivers/net/ethernet/chelsio/cxgb4/cxgb4_tc_mqprio.c +6 -3
@@ -589,7 +589,8 @@ int cxgb4_setup_tc_mqprio(struct net_device *dev,* down before configuring tc params. */ if (netif_running (dev)) { -cxgb_close (dev); +netif_tx_stop_all_queues (dev); +netif_carrier_off (dev); needs_bring_up = true; }
@@ -615,8 +616,10 @@ int cxgb4_setup_tc_mqprio(struct net_device *dev, } out: -if (needs_bring_up) -cxgb_open (dev); +if (needs_bring_up) { +netif_tx_start_all_queues (dev); +netif_carrier_on (dev); + }mutex_unlock (&adap->tc_mqprio->mqprio_mutex);return ret;
drivers/net/ethernet/chelsio/cxgb4/sge.c +6 -0
@@ -2556,6 +2556,12 @@ int cxgb4_ethofld_send_flowc(struct net_device *dev, u32 eotid, u32 tc)if (!eosw_txq)return -ENOMEM ; +if (!(adap->flags &CXGB4_FW_OK )) { +/* Don't stall caller when access to FW is lost */ +complete (&eosw_txq->completion); +return -EIO ; + } + skb =alloc_skb (len,GFP_KERNEL );if (!skb)return -ENOMEM ;
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 60/161 i40e: optimize for XDP_REDIRECT in xsk path
In reply to Greg Kroah-HartmanFrom: Magnus Karlsson <[email protected]>
[ Upstream commit 346497c78d15cdd5bdc3b642a895009359e5457f ]
Optimize i40e_run_xdp_zc() for the XDP program verdict being XDP_REDIRECT in the xsk zero-copy path. This path is only used when having AF_XDP zero-copy on and in that case most packets will be directed to user space. This provides a little over 100k extra packets in throughput on my server when running l2fwd in xdpsock.
--- drivers/net/ethernet/intel/i40e/i40e_xsk.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-)
drivers/net/ethernet/intel/i40e/i40e_xsk.c +7 -4
@@ -160,6 +160,13 @@ static int i40e_run_xdp_zc(struct i40e_ring *rx_ring, struct xdp_buff *xdp) xdp_prog =READ_ONCE (rx_ring->xdp_prog); act =bpf_prog_run_xdp (xdp_prog, xdp); +if (likely (act ==XDP_REDIRECT )) { + err =xdp_do_redirect (rx_ring->netdev, xdp, xdp_prog); + result = !err ?I40E_XDP_REDIR :I40E_XDP_CONSUMED ; +rcu_read_unlock (); +return result; + } +switch (act) {case XDP_PASS :break ;
@@ -167,10 +174,6 @@ static int i40e_run_xdp_zc(struct i40e_ring *rx_ring, struct xdp_buff *xdp) xdp_ring = rx_ring->vsi->xdp_rings[rx_ring->queue_index]; result =i40e_xmit_xdp_tx_ring (xdp, xdp_ring);break ; -case XDP_REDIRECT : - err =xdp_do_redirect (rx_ring->netdev, xdp, xdp_prog); - result = !err ?I40E_XDP_REDIR :I40E_XDP_CONSUMED ; -break ;default :bpf_warn_invalid_xdp_action (act); fallthrough;
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 61/161 i40e: add correct exception tracing for XDP
In reply to Greg Kroah-HartmanFrom: Magnus Karlsson <[email protected]>
[ Upstream commit f6c10b48f8c8da44adaff730d8e700b6272add2b ]
Add missing exception tracing to XDP when a number of different errors can occur. The support was only partial. Several errors where not logged which would confuse the user quite a lot not knowing where and why the packets disappeared.
--- drivers/net/ethernet/intel/i40e/i40e_txrx.c | 7 ++++++- drivers/net/ethernet/intel/i40e/i40e_xsk.c | 8 ++++++-- 2 files changed, 12 insertions(+), 3 deletions(-)
drivers/net/ethernet/intel/i40e/i40e_txrx.c +6 -1
@@ -2313,15 +2313,20 @@ static int i40e_run_xdp(struct i40e_ring *rx_ring, struct xdp_buff *xdp)case XDP_TX : xdp_ring = rx_ring->vsi->xdp_rings[rx_ring->queue_index]; result =i40e_xmit_xdp_tx_ring (xdp, xdp_ring); +if (result ==I40E_XDP_CONSUMED ) + goto out_failure;break ;case XDP_REDIRECT : err =xdp_do_redirect (rx_ring->netdev, xdp, xdp_prog); - result = !err ?I40E_XDP_REDIR :I40E_XDP_CONSUMED ; +if (err) + goto out_failure; + result =I40E_XDP_REDIR ;break ;default :bpf_warn_invalid_xdp_action (act); fallthrough;case XDP_ABORTED : +out_failure:trace_xdp_exception (rx_ring->netdev, xdp_prog, act); fallthrough;/* handle aborts by dropping packet */ case XDP_DROP :
drivers/net/ethernet/intel/i40e/i40e_xsk.c +6 -2
@@ -162,9 +162,10 @@ static int i40e_run_xdp_zc(struct i40e_ring *rx_ring, struct xdp_buff *xdp)if (likely (act ==XDP_REDIRECT )) { err =xdp_do_redirect (rx_ring->netdev, xdp, xdp_prog); - result = !err ?I40E_XDP_REDIR :I40E_XDP_CONSUMED ; +if (err) + goto out_failure;rcu_read_unlock (); -return result; +return I40E_XDP_REDIR ; }switch (act) {
@@ -173,11 +174,14 @@ static int i40e_run_xdp_zc(struct i40e_ring *rx_ring, struct xdp_buff *xdp)case XDP_TX : xdp_ring = rx_ring->vsi->xdp_rings[rx_ring->queue_index]; result =i40e_xmit_xdp_tx_ring (xdp, xdp_ring); +if (result ==I40E_XDP_CONSUMED ) + goto out_failure;break ;default :bpf_warn_invalid_xdp_action (act); fallthrough;case XDP_ABORTED : +out_failure:trace_xdp_exception (rx_ring->netdev, xdp_prog, act); fallthrough;/* handle aborts by dropping packet */ case XDP_DROP :
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 62/161 ice: optimize for XDP_REDIRECT in xsk path
In reply to Greg Kroah-HartmanFrom: Magnus Karlsson <[email protected]>
[ Upstream commit bb52073645a618ab4d93c8d932fb8faf114c55bc ]
Optimize ice_run_xdp_zc() for the XDP program verdict being XDP_REDIRECT in the xsk zero-copy path. This path is only used when having AF_XDP zero-copy on and in that case most packets will be directed to user space. This provides a little over 100k extra packets in throughput on my server when running l2fwd in xdpsock.
--- drivers/net/ethernet/intel/ice/ice_xsk.c | 12 ++++++++---- 1 file changed, 8 insertions(+), 4 deletions(-)
drivers/net/ethernet/intel/ice/ice_xsk.c +8 -4
@@ -476,6 +476,14 @@ ice_run_xdp_zc(struct ice_ring *rx_ring, struct xdp_buff *xdp) xdp_prog =READ_ONCE (rx_ring->xdp_prog); act =bpf_prog_run_xdp (xdp_prog, xdp); + +if (likely (act ==XDP_REDIRECT )) { + err =xdp_do_redirect (rx_ring->netdev, xdp, xdp_prog); + result = !err ?ICE_XDP_REDIR :ICE_XDP_CONSUMED ; +rcu_read_unlock (); +return result; + } +switch (act) {case XDP_PASS :break ;
@@ -483,10 +491,6 @@ ice_run_xdp_zc(struct ice_ring *rx_ring, struct xdp_buff *xdp) xdp_ring = rx_ring->vsi->xdp_rings[rx_ring->q_index]; result =ice_xmit_xdp_buff (xdp, xdp_ring);break ; -case XDP_REDIRECT : - err =xdp_do_redirect (rx_ring->netdev, xdp, xdp_prog); - result = !err ?ICE_XDP_REDIR :ICE_XDP_CONSUMED ; -break ;default :bpf_warn_invalid_xdp_action (act); fallthrough;
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 63/161 ice: add correct exception tracing for XDP
In reply to Greg Kroah-HartmanFrom: Magnus Karlsson <[email protected]>
[ Upstream commit 89d65df024c59988291f643b4e45d1528c51aef9 ]
Add missing exception tracing to XDP when a number of different errors can occur. The support was only partial. Several errors where not logged which would confuse the user quite a lot not knowing where and why the packets disappeared.
--- drivers/net/ethernet/intel/ice/ice_txrx.c | 12 +++++++++--- drivers/net/ethernet/intel/ice/ice_xsk.c | 8 ++++++-- 2 files changed, 15 insertions(+), 5 deletions(-)
drivers/net/ethernet/intel/ice/ice_txrx.c +9 -3
@@ -523,7 +523,7 @@ ice_run_xdp(struct ice_ring *rx_ring, struct xdp_buff *xdp, struct bpf_prog *xdp_prog) {struct ice_ring *xdp_ring; -int err; +int err, result;u32 act; act =bpf_prog_run_xdp (xdp_prog, xdp);
@@ -532,14 +532,20 @@ ice_run_xdp(struct ice_ring *rx_ring, struct xdp_buff *xdp,return ICE_XDP_PASS ;case XDP_TX : xdp_ring = rx_ring->vsi->xdp_rings[smp_processor_id ()]; -return ice_xmit_xdp_buff (xdp, xdp_ring); + result =ice_xmit_xdp_buff (xdp, xdp_ring); +if (result ==ICE_XDP_CONSUMED ) + goto out_failure; +return result;case XDP_REDIRECT : err =xdp_do_redirect (rx_ring->netdev, xdp, xdp_prog); -return !err ?ICE_XDP_REDIR :ICE_XDP_CONSUMED ; +if (err) + goto out_failure; +return ICE_XDP_REDIR ;default :bpf_warn_invalid_xdp_action (act); fallthrough;case XDP_ABORTED : +out_failure:trace_xdp_exception (rx_ring->netdev, xdp_prog, act); fallthrough;case XDP_DROP :
drivers/net/ethernet/intel/ice/ice_xsk.c +6 -2
@@ -479,9 +479,10 @@ ice_run_xdp_zc(struct ice_ring *rx_ring, struct xdp_buff *xdp)if (likely (act ==XDP_REDIRECT )) { err =xdp_do_redirect (rx_ring->netdev, xdp, xdp_prog); - result = !err ?ICE_XDP_REDIR :ICE_XDP_CONSUMED ; +if (err) + goto out_failure;rcu_read_unlock (); -return result; +return ICE_XDP_REDIR ; }switch (act) {
@@ -490,11 +491,14 @@ ice_run_xdp_zc(struct ice_ring *rx_ring, struct xdp_buff *xdp)case XDP_TX : xdp_ring = rx_ring->vsi->xdp_rings[rx_ring->q_index]; result =ice_xmit_xdp_buff (xdp, xdp_ring); +if (result ==ICE_XDP_CONSUMED ) + goto out_failure;break ;default :bpf_warn_invalid_xdp_action (act); fallthrough;case XDP_ABORTED : +out_failure:trace_xdp_exception (rx_ring->netdev, xdp_prog, act); fallthrough;case XDP_DROP :
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 64/161 ixgbe: optimize for XDP_REDIRECT in xsk path
In reply to Greg Kroah-HartmanFrom: Magnus Karlsson <[email protected]>
[ Upstream commit 7d52fe2eaddfa3d7255d43c3e89ebf2748b7ea7a ]
Optimize ixgbe_run_xdp_zc() for the XDP program verdict being XDP_REDIRECT in the xsk zero-copy path. This path is only used when having AF_XDP zero-copy on and in that case most packets will be directed to user space. This provides a little under 100k extra packets in throughput on my server when running l2fwd in xdpsock.
--- drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c | 11 +++++++---- 1 file changed, 7 insertions(+), 4 deletions(-)
drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c +7 -4
@@ -104,6 +104,13 @@ static int ixgbe_run_xdp_zc(struct ixgbe_adapter *adapter, xdp_prog =READ_ONCE (rx_ring->xdp_prog); act =bpf_prog_run_xdp (xdp_prog, xdp); +if (likely (act ==XDP_REDIRECT )) { + err =xdp_do_redirect (rx_ring->netdev, xdp, xdp_prog); + result = !err ?IXGBE_XDP_REDIR :IXGBE_XDP_CONSUMED ; +rcu_read_unlock (); +return result; + } +switch (act) {case XDP_PASS :break ;
@@ -115,10 +122,6 @@ static int ixgbe_run_xdp_zc(struct ixgbe_adapter *adapter, } result =ixgbe_xmit_xdp_ring (adapter, xdpf);break ; -case XDP_REDIRECT : - err =xdp_do_redirect (rx_ring->netdev, xdp, xdp_prog); - result = !err ?IXGBE_XDP_REDIR :IXGBE_XDP_CONSUMED ; -break ;default :bpf_warn_invalid_xdp_action (act); fallthrough;
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 65/161 ixgbe: add correct exception tracing for XDP
In reply to Greg Kroah-HartmanFrom: Magnus Karlsson <[email protected]>
[ Upstream commit 8281356b1cab1cccc71412eb4cf28b99d6bb2c19 ]
Add missing exception tracing to XDP when a number of different errors can occur. The support was only partial. Several errors where not logged which would confuse the user quite a lot not knowing where and why the packets disappeared.
--- drivers/net/ethernet/intel/ixgbe/ixgbe_main.c | 16 ++++++++-------- drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c | 14 ++++++++------ 2 files changed, 16 insertions(+), 14 deletions(-)
drivers/net/ethernet/intel/ixgbe/ixgbe_main.c +8 -8
@@ -2213,23 +2213,23 @@ static struct sk_buff *ixgbe_run_xdp(struct ixgbe_adapter *adapter,break ;case XDP_TX : xdpf =xdp_convert_buff_to_frame (xdp); -if (unlikely (!xdpf)) { - result =IXGBE_XDP_CONSUMED ; -break ; - } +if (unlikely (!xdpf)) + goto out_failure; result =ixgbe_xmit_xdp_ring (adapter, xdpf); +if (result ==IXGBE_XDP_CONSUMED ) + goto out_failure;break ;case XDP_REDIRECT : err =xdp_do_redirect (adapter->netdev, xdp, xdp_prog); -if (!err) - result =IXGBE_XDP_REDIR ; -else - result =IXGBE_XDP_CONSUMED ; +if (err) + goto out_failure; + result =IXGBE_XDP_REDIR ;break ;default :bpf_warn_invalid_xdp_action (act); fallthrough;case XDP_ABORTED : +out_failure:trace_xdp_exception (rx_ring->netdev, xdp_prog, act); fallthrough;/* handle aborts by dropping packet */ case XDP_DROP :
drivers/net/ethernet/intel/ixgbe/ixgbe_xsk.c +8 -6
@@ -106,9 +106,10 @@ static int ixgbe_run_xdp_zc(struct ixgbe_adapter *adapter,if (likely (act ==XDP_REDIRECT )) { err =xdp_do_redirect (rx_ring->netdev, xdp, xdp_prog); - result = !err ?IXGBE_XDP_REDIR :IXGBE_XDP_CONSUMED ; +if (err) + goto out_failure;rcu_read_unlock (); -return result; +return IXGBE_XDP_REDIR ; }switch (act) {
@@ -116,16 +117,17 @@ static int ixgbe_run_xdp_zc(struct ixgbe_adapter *adapter,break ;case XDP_TX : xdpf =xdp_convert_buff_to_frame (xdp); -if (unlikely (!xdpf)) { - result =IXGBE_XDP_CONSUMED ; -break ; - } +if (unlikely (!xdpf)) + goto out_failure; result =ixgbe_xmit_xdp_ring (adapter, xdpf); +if (result ==IXGBE_XDP_CONSUMED ) + goto out_failure;break ;default :bpf_warn_invalid_xdp_action (act); fallthrough;case XDP_ABORTED : +out_failure:trace_xdp_exception (rx_ring->netdev, xdp_prog, act); fallthrough;/* handle aborts by dropping packet */ case XDP_DROP :
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 66/161 arm64: dts: ti: j7200-main: Mark Main NAVSS as dma-coherent
In reply to Greg Kroah-HartmanFrom: Vignesh Raghavendra <[email protected]>
[ Upstream commit 52ae30f55a2a40cff549fac95de82f25403bd387 ]
Traffic through main NAVSS interconnect is coherent wrt ARM caches on J7200 SoC. Add missing dma-coherent property to main_navss node.
Also add dma-ranges to be consistent with mcu_navss node and with AM65/J721e main_navss and mcu_navss nodes.
--- arch/arm64/boot/dts/ti/k3-j7200-main.dtsi | 2 ++ 1 file changed, 2 insertions(+)
arch/arm64/boot/dts/ti/k3-j7200-main.dtsi +2 -0
@@ -85,6 +85,8 @@ #size-cells = <2>; ranges = <0x00 0x30000000 0x00 0x30000000 0x00 0x0c400000>; ti,sci-dev-id = <199>; + dma-coherent; + dma-ranges; main_navss_intr: interrupt-controller1 { compatible = "ti,sci-intr";
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 67/161 optee: use export_uuid() to copy client UUID
In reply to Greg Kroah-HartmanFrom: Jens Wiklander <[email protected]>
[ Upstream commit 673c7aa2436bfc857b92417f3e590a297c586dde ]
Prior to this patch optee_open_session() was making assumptions about the internal format of uuid_t by casting a memory location in a parameter struct to uuid_t *. Fix this using export_uuid() to get a well defined binary representation and also add an octets field in struct optee_msg_param in order to avoid casting.
--- drivers/tee/optee/call.c | 6 ++++-- drivers/tee/optee/optee_msg.h | 6 ++++-- 2 files changed, 8 insertions(+), 4 deletions(-)
drivers/tee/optee/call.c +4 -2
@@ -216,6 +216,7 @@ int optee_open_session(struct tee_context *ctx,struct optee_msg_arg *msg_arg;phys_addr_t msg_parg;struct optee_session *sess =NULL ; +uuid_t client_uuid;/* +2 for the meta parameters added below */ shm =get_msg_arg (ctx, arg->num_params +2 , &msg_arg, &msg_parg);
@@ -236,10 +237,11 @@ int optee_open_session(struct tee_context *ctx,memcpy (&msg_arg->params[0 ].u.value, arg->uuid,sizeof (arg->uuid)); msg_arg->params[1 ].u.value.c = arg->clnt_login; - rc =tee_session_calc_client_uuid ((uuid_t *)&msg_arg->params[1 ].u.value, - arg->clnt_login, arg->clnt_uuid); + rc =tee_session_calc_client_uuid (&client_uuid, arg->clnt_login, + arg->clnt_uuid);if (rc) goto out; +export_uuid (msg_arg->params[1 ].u.octets, &client_uuid); rc =optee_to_msg_param (msg_arg->params +2 , arg->num_params, param);if (rc)
drivers/tee/optee/optee_msg.h +4 -2
@@ -9,7 +9,7 @@#include <linux/types.h> /* -* This file defines the OP-TEE message protocol used to communicate +* This file defines the OP-TEE message protocol (ABI) used to communicate * with an instance of OP-TEE running in secure world. * * This file is divided into two sections.
@@ -144,9 +144,10 @@ struct optee_msg_param_value {* @tmem: parameter by temporary memory reference * @rmem: parameter by registered memory reference * @value: parameter by opaque value +* @octets: parameter by octet string * * @attr & OPTEE_MSG_ATTR_TYPE_MASK indicates if tmem, rmem or value is used in -* the union. OPTEE_MSG_ATTR_TYPE_VALUE_* indicates value, +* the union. OPTEE_MSG_ATTR_TYPE_VALUE_* indicates value or octets, * OPTEE_MSG_ATTR_TYPE_TMEM_* indicates @tmem and * OPTEE_MSG_ATTR_TYPE_RMEM_* indicates @rmem, * OPTEE_MSG_ATTR_TYPE_NONE indicates that none of the members are used.
@@ -157,6 +158,7 @@ struct optee_msg_param {struct optee_msg_param_tmem tmem;struct optee_msg_param_rmem rmem;struct optee_msg_param_value value; +u8 octets[24 ]; } u; };
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 68/161 bus: ti-sysc: Fix am335x resume hang for usb otg module
In reply to Greg Kroah-HartmanFrom: Tony Lindgren <[email protected]>
[ Upstream commit 4d7b324e231366ea772ab10df46be31273ca39af ]
On am335x, suspend and resume only works once, and the system hangs if suspend is attempted again. However, turns out suspend and resume works fine multiple times if the USB OTG driver for musb controller is loaded.
The issue is caused my the interconnect target module losing context during suspend, and it needs a restore on resume to be reconfigure again as debugged earlier by Dave Gerlach <[email protected]>.
There are also other modules that need a restore on resume, like gpmc as noted by Dave. So let's add a common way to restore an interconnect target module based on a quirk flag. For now, let's enable the quirk for am335x otg only to fix the suspend and resume issue.
As gpmc is not causing hangs based on tests with BeagleBone, let's patch gpmc separately. For gpmc, we also need a hardware reset done before restore according to Dave.
To reinit the modules, we decouple system suspend from PM runtime. We replace calls to pm_runtime_force_suspend() and pm_runtime_force_resume() with direct calls to internal functions and rely on the driver internal state. There no point trying to handle complex system suspend and resume quirks via PM runtime.
This is issue should have already been noticed with commit 1819ef2e2d12 ("bus: ti-sysc: Use swsup quirks also for am335x musb") when quirk handling was added for am335x otg for swsup. But the issue went unnoticed as having musb driver loaded hides the issue, and suspend and resume works once without the driver loaded.
--- drivers/bus/ti-sysc.c | 53 +++++++++++++++++++++++++-- include/linux/platform_data/ti-sysc.h | 1 + 2 files changed, 51 insertions(+), 3 deletions(-)
drivers/bus/ti-sysc.c +50 -3
@@ -1334,6 +1334,34 @@ static int __maybe_unused sysc_runtime_resume(struct device *dev)return error; } +static int sysc_reinit_module (struct sysc *ddata,bool leave_enabled) +{ +struct device *dev = ddata->dev; +int error; + +/* Disable target module if it is enabled */ +if (ddata->enabled) { + error =sysc_runtime_suspend (dev); +if (error) +dev_warn (dev,"reinit suspend failed: %i\n" , error); + } + +/* Enable target module */ + error =sysc_runtime_resume (dev); +if (error) +dev_warn (dev,"reinit resume failed: %i\n" , error); + +if (leave_enabled) +return error; + +/* Disable target module if no leave_enabled was set */ + error =sysc_runtime_suspend (dev); +if (error) +dev_warn (dev,"reinit suspend failed: %i\n" , error); + +return error; +} +static int __maybe_unused sysc_noirq_suspend (struct device *dev) {struct sysc *ddata;
@@ -1344,12 +1372,18 @@ static int __maybe_unused sysc_noirq_suspend(struct device *dev) (SYSC_QUIRK_LEGACY_IDLE | SYSC_QUIRK_NO_IDLE))return 0 ; -return pm_runtime_force_suspend (dev); +if (!ddata->enabled) +return 0 ; + + ddata->needs_resume =1 ; + +return sysc_runtime_suspend (dev); }static int __maybe_unused sysc_noirq_resume (struct device *dev) {struct sysc *ddata; +int error =0 ; ddata =dev_get_drvdata (dev);
@@ -1357,7 +1391,19 @@ static int __maybe_unused sysc_noirq_resume(struct device *dev) (SYSC_QUIRK_LEGACY_IDLE | SYSC_QUIRK_NO_IDLE))return 0 ; -return pm_runtime_force_resume (dev); +if (ddata->cfg.quirks &SYSC_QUIRK_REINIT_ON_RESUME ) { + error =sysc_reinit_module (ddata, ddata->needs_resume); +if (error) +dev_warn (dev,"noirq_resume failed: %i\n" , error); + }else if (ddata->needs_resume) { + error =sysc_runtime_resume (dev); +if (error) +dev_warn (dev,"noirq_resume failed: %i\n" , error); + } + + ddata->needs_resume =0 ; + +return error; }static const struct dev_pm_ops sysc_pm_ops = {
@@ -1466,7 +1512,8 @@ static const struct sysc_revision_quirk sysc_revision_quirks[] = {SYSC_QUIRK ("usb_otg_hs" ,0 ,0x400 ,0x404 ,0x408 ,0x00000050 ,0xffffffff ,SYSC_QUIRK_SWSUP_SIDLE |SYSC_QUIRK_SWSUP_MSTANDBY ),SYSC_QUIRK ("usb_otg_hs" ,0 ,0 ,0x10 , -ENODEV ,0x4ea2080d ,0xffffffff , -SYSC_QUIRK_SWSUP_SIDLE |SYSC_QUIRK_SWSUP_MSTANDBY ), +SYSC_QUIRK_SWSUP_SIDLE |SYSC_QUIRK_SWSUP_MSTANDBY | +SYSC_QUIRK_REINIT_ON_RESUME ),SYSC_QUIRK ("wdt" ,0 ,0 ,0x10 ,0x14 ,0x502a0500 ,0xfffff0f0 , SYSC_MODULE_QUIRK_WDT),/* PRUSS on am3, am4 and am5 */
include/linux/platform_data/ti-sysc.h +1 -0
@@ -50,6 +50,7 @@ struct sysc_regbits {s8 emufree_shift; }; +#define SYSC_QUIRK_REINIT_ON_RESUME BIT(27)#define SYSC_QUIRK_GPMC_DEBUG BIT(26)#define SYSC_MODULE_QUIRK_ENA_RESETDONE BIT(25)#define SYSC_MODULE_QUIRK_PRUSS BIT(24)
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 69/161 arm64: dts: ls1028a: fix memory node
In reply to Greg Kroah-HartmanFrom: Michael Walle <[email protected]>
[ Upstream commit dabea675faf16e8682aa478ff3ce65dd775620bc ]
While enabling EDAC support for the LS1028A it was discovered that the memory node has a wrong endianness setting as well as a wrong interrupt assignment. Fix both.
This was tested on a sl28 board. To force ECC errors, you can use the error injection supported by the controller in hardware (with CONFIG_EDAC_DEBUG enabled):
# enable error injection $ echo 0x100 > /sys/devices/system/edac/mc/mc0/inject_ctrl # flip lowest bit of the data $ echo 0x1 > /sys/devices/system/edac/mc/mc0/inject_data_lo
--- arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm64/boot/dts/freescale/fsl-ls1028a.dtsi +2 -2
@@ -201,8 +201,8 @@ ddr: memory-controller@1080000 { compatible = "fsl,qoriq-memory-controller"; reg = <0x0 0x1080000 0x0 0x1000>; - interrupts = <GIC_SPI 144 IRQ_TYPE_LEVEL_HIGH>; - big-endian; + interrupts = <GIC_SPI 17 IRQ_TYPE_LEVEL_HIGH>; + little-endian; }; dcfg: syscon@1e00000 {
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 70/161 arm64: dts: zii-ultra: remove second GEN_3V3 regulator instance
In reply to Greg Kroah-HartmanFrom: Lucas Stach <[email protected]>
[ Upstream commit e98d98028989e023e0cbff539dc616c4e5036839 ]
When adding the sound support a second instance of the GEN_3V3 regulator was added by accident. Remove it and point the consumers to the first instance.
--- .../dts/freescale/imx8mq-zii-ultra-rmb3.dts | 10 +++++----- .../boot/dts/freescale/imx8mq-zii-ultra.dtsi | 19 +++++-------------- 2 files changed, 10 insertions(+), 19 deletions(-)
arch/arm64/boot/dts/freescale/imx8mq-zii-ultra-rmb3.dts +5 -5
@@ -88,11 +88,11 @@ pinctrl-0 = <&pinctrl_codec2>; reg = <0x18>; #sound-dai-cells = <0>; - HPVDD-supply = <®_3p3v>; - SPRVDD-supply = <®_3p3v>; - SPLVDD-supply = <®_3p3v>; - AVDD-supply = <®_3p3v>; - IOVDD-supply = <®_3p3v>; + HPVDD-supply = <®_gen_3p3>; + SPRVDD-supply = <®_gen_3p3>; + SPLVDD-supply = <®_gen_3p3>; + AVDD-supply = <®_gen_3p3>; + IOVDD-supply = <®_gen_3p3>; DVDD-supply = <&vgen4_reg>; reset-gpios = <&gpio3 4 GPIO_ACTIVE_HIGH>; };
arch/arm64/boot/dts/freescale/imx8mq-zii-ultra.dtsi +5 -14
@@ -77,15 +77,6 @@ regulator-always-on; }; - reg_3p3v: regulator-3p3v { - compatible = "regulator-fixed"; - vin-supply = <®_3p3_main>; - regulator-name = "GEN_3V3"; - regulator-min-microvolt = <3300000>; - regulator-max-microvolt = <3300000>; - regulator-always-on; - }; - reg_usdhc2_vmmc: regulator-vsd-3v3 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_reg_usdhc2>;
@@ -415,11 +406,11 @@ pinctrl-0 = <&pinctrl_codec1>; reg = <0x18>; #sound-dai-cells = <0>; - HPVDD-supply = <®_3p3v>; - SPRVDD-supply = <®_3p3v>; - SPLVDD-supply = <®_3p3v>; - AVDD-supply = <®_3p3v>; - IOVDD-supply = <®_3p3v>; + HPVDD-supply = <®_gen_3p3>; + SPRVDD-supply = <®_gen_3p3>; + SPLVDD-supply = <®_gen_3p3>; + AVDD-supply = <®_gen_3p3>; + IOVDD-supply = <®_gen_3p3>; DVDD-supply = <&vgen4_reg>; reset-gpios = <&gpio3 3 GPIO_ACTIVE_LOW>; };
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 71/161 arm64: dts: zii-ultra: fix 12V_MAIN voltage
In reply to Greg Kroah-HartmanFrom: Lucas Stach <[email protected]>
[ Upstream commit ac0cbf9d13dccfd09bebc2f8f5697b6d3ffe27c4 ]
As this is a fixed regulator on the board there was no harm in the wrong voltage being specified, apart from a confusing reporting to userspace.
--- arch/arm64/boot/dts/freescale/imx8mq-zii-ultra.dtsi | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
arch/arm64/boot/dts/freescale/imx8mq-zii-ultra.dtsi +2 -2
@@ -45,8 +45,8 @@ reg_12p0_main: regulator-12p0-main { compatible = "regulator-fixed"; regulator-name = "12V_MAIN"; - regulator-min-microvolt = <5000000>; - regulator-max-microvolt = <5000000>; + regulator-min-microvolt = <12000000>; + regulator-max-microvolt = <12000000>; regulator-always-on; };
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 72/161 arm64: dts: freescale: sl28: var4: fix RGMII clock and voltage
In reply to Greg Kroah-HartmanFrom: Michael Walle <[email protected]>
[ Upstream commit 25201269c6ec3e9398426962ccdd55428261f7d0 ]
During hardware validation it was noticed that the clock isn't continuously enabled when there is no link. This is because the 125MHz clock is derived from the internal PLL which seems to go into some kind of power-down mode every once in a while. The LS1028A expects a contiuous clock. Thus enable the PLL all the time.
Also, the RGMII pad voltage is wrong. It was configured to 2.5V (that is the VDDH regulator). The correct voltage is 1.8V, i.e. the VDDIO regulator.
This fix is for the freescale/fsl-ls1028a-kontron-sl28-var4.dts.
--- .../boot/dts/freescale/fsl-ls1028a-kontron-sl28-var4.dts | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
arch/arm64/boot/dts/freescale/fsl-ls1028a-kontron-sl28-var4.dts +2 -3
@@ -31,11 +31,10 @@ reg = <0x4>; eee-broken-1000t; eee-broken-100tx; - qca,clk-out-frequency = <125000000>; qca,clk-out-strength = <AR803X_STRENGTH_FULL>; - - vddio-supply = <&vddh>; + qca,keep-pll-enabled; + vddio-supply = <&vddio>; vddio: vddio-regulator { regulator-name = "VDDIO";
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 73/161 arm64: dts: freescale: sl28: var1: fix RGMII clock and voltage
In reply to Greg Kroah-HartmanFrom: Michael Walle <[email protected]>
[ Upstream commit 52387bb9a4a75b88887383cb91d3995ae6f4044a ]
During hardware validation it was noticed that the clock isn't continuously enabled when there is no link. This is because the 125MHz clock is derived from the internal PLL which seems to go into some kind of power-down mode every once in a while. The LS1028A expects a contiuous clock. Thus enable the PLL all the time.
Also, the RGMII pad voltage is wrong, it was configured to 2.5V (that is the VDDH regulator). The correct voltage is 1.8V, i.e. the VDDIO regulator.
This fix is for the freescale/fsl-ls1028a-kontron-sl28-var1.dts.
--- .../arm64/boot/dts/freescale/fsl-ls1028a-kontron-sl28-var1.dts | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
arch/arm64/boot/dts/freescale/fsl-ls1028a-kontron-sl28-var1.dts +2 -1
@@ -46,7 +46,8 @@ eee-broken-100tx; qca,clk-out-frequency = <125000000>; qca,clk-out-strength = <AR803X_STRENGTH_FULL>; - vddio-supply = <&vddh>; + qca,keep-pll-enabled; + vddio-supply = <&vddio>; vddio: vddio-regulator { regulator-name = "VDDIO";
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 74/161 ARM: dts: imx7d-meerkat96: Fix the tuning-step property
In reply to Greg Kroah-HartmanFrom: Fabio Estevam <[email protected]>
[ Upstream commit 7c8f0338cdacc90fdf6468adafa8e27952987f00 ]
According to Documentation/devicetree/bindings/mmc/fsl-imx-esdhc.yaml, the correct name of the property is 'fsl,tuning-step'.
Fix it accordingly.
--- arch/arm/boot/dts/imx7d-meerkat96.dts | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/boot/dts/imx7d-meerkat96.dts +1 -1
@@ -193,7 +193,7 @@ pinctrl-names = "default"; pinctrl-0 = <&pinctrl_usdhc1>; keep-power-in-suspend; - tuning-step = <2>; + fsl,tuning-step = <2>; vmmc-supply = <®_3p3v>; no-1-8-v; broken-cd;
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 75/161 ARM: dts: imx7d-pico: Fix the tuning-step property
In reply to Greg Kroah-HartmanFrom: Fabio Estevam <[email protected]>
[ Upstream commit 0e2fa4959c4f44815ce33e46e4054eeb0f346053 ]
According to Documentation/devicetree/bindings/mmc/fsl-imx-esdhc.yaml, the correct name of the property is 'fsl,tuning-step'.
Fix it accordingly.
--- arch/arm/boot/dts/imx7d-pico.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/boot/dts/imx7d-pico.dtsi +1 -1
@@ -351,7 +351,7 @@ pinctrl-2 = <&pinctrl_usdhc1_200mhz>; cd-gpios = <&gpio5 0 GPIO_ACTIVE_LOW>; bus-width = <4>; - tuning-step = <2>; + fsl,tuning-step = <2>; vmmc-supply = <®_3p3v>; wakeup-source; no-1-8-v;
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 76/161 ARM: dts: imx: emcon-avari: Fix nxp,pca8574 #gpio-cells
In reply to Greg Kroah-HartmanFrom: Geert Uytterhoeven <[email protected]>
[ Upstream commit b73eb6b3b91ff7d76cff5f8c7ab92fe0c51e3829 ]
According to the DT bindings, #gpio-cells must be two.
--- arch/arm/boot/dts/imx6qdl-emcon-avari.dtsi | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/boot/dts/imx6qdl-emcon-avari.dtsi +1 -1
@@ -126,7 +126,7 @@ compatible = "nxp,pca8574"; reg = <0x3a>; gpio-controller; - #gpio-cells = <1>; + #gpio-cells = <2>; }; };
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 77/161 bus: ti-sysc: Fix flakey idling of uarts and stop using swsup_sidle_act
In reply to Greg Kroah-HartmanFrom: Tony Lindgren <[email protected]>
[ Upstream commit c8692ad416dcc420ce1b403596a425c8f4c2720b ]
Looks like the swsup_sidle_act quirk handling is unreliable for serial ports. The serial ports just eventually stop idling until woken up and re-idled again. As the serial port not idling blocks any deeper SoC idle states, it's adds an annoying random flakeyness for power management.
Let's just switch to swsup_sidle quirk instead like we already do for omap3 uarts. This means we manually idle the port instead of trying to use the hardware autoidle features when not in use.
For more details on why the serial ports have been using swsup_idle_act, see commit 66dde54e978a ("ARM: OMAP2+: hwmod-data: UART IP needs software control to manage sidle modes"). It seems that the swsup_idle_act quirk handling is not enough though, and for example the TI Android kernel changed to using swsup_sidle with commit 77c34c84e1e0 ("OMAP4: HWMOD:
UART1: disable smart-idle.").
--- drivers/bus/ti-sysc.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/bus/ti-sysc.c +2 -2
@@ -1454,9 +1454,9 @@ static const struct sysc_revision_quirk sysc_revision_quirks[] = { SYSC_QUIRK_SWSUP_SIDLE | SYSC_QUIRK_LEGACY_IDLE),/* Uarts on omap4 and later */ SYSC_QUIRK ("uart" ,0 ,0x50 ,0x54 ,0x58 ,0x50411e03 ,0xffff00ff , -SYSC_QUIRK_SWSUP_SIDLE_ACT |SYSC_QUIRK_LEGACY_IDLE ), +SYSC_QUIRK_SWSUP_SIDLE |SYSC_QUIRK_LEGACY_IDLE ),SYSC_QUIRK ("uart" ,0 ,0x50 ,0x54 ,0x58 ,0x47422e03 ,0xffffffff , -SYSC_QUIRK_SWSUP_SIDLE_ACT |SYSC_QUIRK_LEGACY_IDLE ), +SYSC_QUIRK_SWSUP_SIDLE |SYSC_QUIRK_LEGACY_IDLE ),/* Quirks that need to be set based on the module address */ SYSC_QUIRK("mcpdm", 0x40132000, 0, 0x10, -ENODEV, 0x50000800, 0xffffffff,
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 78/161 arm64: meson: select COMMON_CLK
In reply to Greg Kroah-HartmanFrom: Jerome Brunet <[email protected]>
[ Upstream commit 4cce442ffe5448ef572adc8b3abe7001b398e709 ]
This fix the recent removal of clock drivers selection.
While it is not necessary to select the clock drivers themselves, we need to select a proper implementation of the clock API, which for the meson, is CCF
--- arch/arm64/Kconfig.platforms | 1 + 1 file changed, 1 insertion(+)
arch/arm64/Kconfig.platforms +1 -0
@@ -168,6 +168,7 @@ config ARCH_MEDIATEK config ARCH_MESON bool "Amlogic Platforms" + select COMMON_CLK select MESON_IRQ_GPIO help This enables support for the arm64 based Amlogic SoCs
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 79/161 tipc: add extack messages for bearer/media failure
In reply to Greg Kroah-HartmanFrom: Hoang Le <[email protected]>
[ Upstream commit b83e214b2e04204f1fc674574362061492c37245 ]
Add extack error messages for -EINVAL errors when enabling bearer, getting/setting properties for a media/bearer
--- net/tipc/bearer.c | 50 +++++++++++++++++++++++++++++++++++++---------- 1 file changed, 40 insertions(+), 10 deletions(-)
net/tipc/bearer.c +40 -10
@@ -243,7 +243,8 @@ void tipc_bearer_remove_dest(struct net *net, u32 bearer_id, u32 dest)*/ static int tipc_enable_bearer (struct net *net,const char *name,u32 disc_domain,u32 prio, -struct nlattr *attr[]) +struct nlattr *attr[], +struct netlink_ext_ack *extack) {struct tipc_net *tn =tipc_net (net);struct tipc_bearer_names b_names;
@@ -257,17 +258,20 @@ static int tipc_enable_bearer(struct net *net, const char *name,if (!bearer_name_validate (name, &b_names)) { errstr ="illegal name" ; +NL_SET_ERR_MSG (extack,"Illegal name" ); goto rejected; }if (prio >TIPC_MAX_LINK_PRI && prio !=TIPC_MEDIA_LINK_PRI ) { errstr ="illegal priority" ; +NL_SET_ERR_MSG (extack,"Illegal priority" ); goto rejected; } m =tipc_media_find (b_names.media_name);if (!m) { errstr ="media not registered" ; +NL_SET_ERR_MSG (extack,"Media not registered" ); goto rejected; }
@@ -281,6 +285,7 @@ static int tipc_enable_bearer(struct net *net, const char *name,break ;if (!strcmp (name, b->name)) { errstr ="already enabled" ; +NL_SET_ERR_MSG (extack,"Already enabled" ); goto rejected; } bearer_id++;
@@ -292,6 +297,7 @@ static int tipc_enable_bearer(struct net *net, const char *name, name, prio);if (prio ==TIPC_MIN_LINK_PRI ) { errstr ="cannot adjust to lower" ; +NL_SET_ERR_MSG (extack,"Cannot adjust to lower" ); goto rejected; }pr_warn ("Bearer <%s>: trying with adjusted priority\n" , name);
@@ -302,6 +308,7 @@ static int tipc_enable_bearer(struct net *net, const char *name,if (bearer_id >=MAX_BEARERS ) { errstr ="max 3 bearers permitted" ; +NL_SET_ERR_MSG (extack,"Max 3 bearers permitted" ); goto rejected; }
@@ -315,6 +322,7 @@ static int tipc_enable_bearer(struct net *net, const char *name,if (res) {kfree (b); errstr ="failed to enable media" ; +NL_SET_ERR_MSG (extack,"Failed to enable media" ); goto rejected; }
@@ -331,6 +339,7 @@ static int tipc_enable_bearer(struct net *net, const char *name,if (res) {bearer_disable (net, b); errstr ="failed to create discoverer" ; +NL_SET_ERR_MSG (extack,"Failed to create discoverer" ); goto rejected; }
@@ -909,6 +918,7 @@ int tipc_nl_bearer_get(struct sk_buff *skb, struct genl_info *info) bearer =tipc_bearer_find (net, name);if (!bearer) { err = -EINVAL ; +NL_SET_ERR_MSG (info->extack,"Bearer not found" ); goto err_out; }
@@ -948,8 +958,10 @@ int __tipc_nl_bearer_disable(struct sk_buff *skb, struct genl_info *info) name =nla_data (attrs[TIPC_NLA_BEARER_NAME ]); bearer =tipc_bearer_find (net, name); -if (!bearer) +if (!bearer) { +NL_SET_ERR_MSG (info->extack,"Bearer not found" );return -EINVAL ; + }bearer_disable (net, bearer);
@@ -1007,7 +1019,8 @@ int __tipc_nl_bearer_enable(struct sk_buff *skb, struct genl_info *info) prio =nla_get_u32 (props[TIPC_NLA_PROP_PRIO ]); } -return tipc_enable_bearer (net, bearer, domain, prio, attrs); +return tipc_enable_bearer (net, bearer, domain, prio, attrs, + info->extack); }int tipc_nl_bearer_enable (struct sk_buff *skb,struct genl_info *info)
@@ -1046,6 +1059,7 @@ int tipc_nl_bearer_add(struct sk_buff *skb, struct genl_info *info) b =tipc_bearer_find (net, name);if (!b) {rtnl_unlock (); +NL_SET_ERR_MSG (info->extack,"Bearer not found" );return -EINVAL ; }
@@ -1086,8 +1100,10 @@ int __tipc_nl_bearer_set(struct sk_buff *skb, struct genl_info *info) name =nla_data (attrs[TIPC_NLA_BEARER_NAME ]); b =tipc_bearer_find (net, name); -if (!b) +if (!b) { +NL_SET_ERR_MSG (info->extack,"Bearer not found" );return -EINVAL ; + }if (attrs[TIPC_NLA_BEARER_PROP ]) {struct nlattr *props[TIPC_NLA_PROP_MAX +1 ];
@@ -1106,12 +1122,18 @@ int __tipc_nl_bearer_set(struct sk_buff *skb, struct genl_info *info)if (props[TIPC_NLA_PROP_WIN ]) b->max_win =nla_get_u32 (props[TIPC_NLA_PROP_WIN ]);if (props[TIPC_NLA_PROP_MTU ]) { -if (b->media->type_id !=TIPC_MEDIA_TYPE_UDP ) +if (b->media->type_id !=TIPC_MEDIA_TYPE_UDP ) { +NL_SET_ERR_MSG (info->extack, +"MTU property is unsupported" );return -EINVAL ; + }#ifdef CONFIG_TIPC_MEDIA_UDP if (tipc_udp_mtu_bad (nla_get_u32 - (props[TIPC_NLA_PROP_MTU ]))) + (props[TIPC_NLA_PROP_MTU ]))) { +NL_SET_ERR_MSG (info->extack, +"MTU value is out-of-range" );return -EINVAL ; + } b->mtu =nla_get_u32 (props[TIPC_NLA_PROP_MTU ]);tipc_node_apply_property (net, b,TIPC_NLA_PROP_MTU );#endif
@@ -1239,6 +1261,7 @@ int tipc_nl_media_get(struct sk_buff *skb, struct genl_info *info)rtnl_lock (); media =tipc_media_find (name);if (!media) { +NL_SET_ERR_MSG (info->extack,"Media not found" ); err = -EINVAL ; goto err_out; }
@@ -1275,9 +1298,10 @@ int __tipc_nl_media_set(struct sk_buff *skb, struct genl_info *info) name =nla_data (attrs[TIPC_NLA_MEDIA_NAME ]); m =tipc_media_find (name); -if (!m) +if (!m) { +NL_SET_ERR_MSG (info->extack,"Media not found" );return -EINVAL ; - + }if (attrs[TIPC_NLA_MEDIA_PROP ]) {struct nlattr *props[TIPC_NLA_PROP_MAX +1 ];
@@ -1293,12 +1317,18 @@ int __tipc_nl_media_set(struct sk_buff *skb, struct genl_info *info)if (props[TIPC_NLA_PROP_WIN ]) m->max_win =nla_get_u32 (props[TIPC_NLA_PROP_WIN ]);if (props[TIPC_NLA_PROP_MTU ]) { -if (m->type_id !=TIPC_MEDIA_TYPE_UDP ) +if (m->type_id !=TIPC_MEDIA_TYPE_UDP ) { +NL_SET_ERR_MSG (info->extack, +"MTU property is unsupported" );return -EINVAL ; + }#ifdef CONFIG_TIPC_MEDIA_UDP if (tipc_udp_mtu_bad (nla_get_u32 - (props[TIPC_NLA_PROP_MTU ]))) + (props[TIPC_NLA_PROP_MTU ]))) { +NL_SET_ERR_MSG (info->extack, +"MTU value is out-of-range" );return -EINVAL ; + } m->mtu =nla_get_u32 (props[TIPC_NLA_PROP_MTU ]);#endif }
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 80/161 tipc: fix unique bearer names sanity check
In reply to Greg Kroah-HartmanFrom: Hoang Le <[email protected]>
[ Upstream commit f20a46c3044c3f75232b3d0e2d09af9b25efaf45 ]
When enabling a bearer by name, we don't sanity check its name with higher slot in bearer list. This may have the effect that the name of an already enabled bearer bypasses the check.
To fix the above issue, we just perform an extra checking with all existing bearers.
--- net/tipc/bearer.c | 46 +++++++++++++++++++++++++++------------------- 1 file changed, 27 insertions(+), 19 deletions(-)
net/tipc/bearer.c +27 -19
@@ -255,6 +255,7 @@ static int tipc_enable_bearer(struct net *net, const char *name,int bearer_id =0 ;int res = -EINVAL ;char *errstr ="" ; +u32 i;if (!bearer_name_validate (name, &b_names)) { errstr ="illegal name" ;
@@ -279,31 +280,38 @@ static int tipc_enable_bearer(struct net *net, const char *name, prio = m->priority;/* Check new bearer vs existing ones and find free bearer id if any */ -while (bearer_id <MAX_BEARERS ) { - b =rtnl_dereference (tn->bearer_list[bearer_id]); -if (!b) -break ; + bearer_id =MAX_BEARERS ; + i =MAX_BEARERS ; +while (i-- !=0 ) { + b =rtnl_dereference (tn->bearer_list[i]); +if (!b) { + bearer_id = i; +continue ; + }if (!strcmp (name, b->name)) { errstr ="already enabled" ;NL_SET_ERR_MSG (extack,"Already enabled" ); goto rejected; } - bearer_id++; -if (b->priority != prio) -continue ; -if (++with_this_prio <=2 ) -continue ; -pr_warn ("Bearer <%s>: already 2 bearers with priority %u\n" , - name, prio); -if (prio ==TIPC_MIN_LINK_PRI ) { - errstr ="cannot adjust to lower" ; -NL_SET_ERR_MSG (extack,"Cannot adjust to lower" ); - goto rejected; + +if (b->priority == prio && + (++with_this_prio >2 )) { +pr_warn ("Bearer <%s>: already 2 bearers with priority %u\n" , + name, prio); + +if (prio ==TIPC_MIN_LINK_PRI ) { + errstr ="cannot adjust to lower" ; +NL_SET_ERR_MSG (extack,"Cannot adjust to lower" ); + goto rejected; + } + +pr_warn ("Bearer <%s>: trying with adjusted priority\n" , + name); + prio--; + bearer_id =MAX_BEARERS ; + i =MAX_BEARERS ; + with_this_prio =1 ; } -pr_warn ("Bearer <%s>: trying with adjusted priority\n" , name); - prio--; - bearer_id =0 ; - with_this_prio =1 ; }if (bearer_id >=MAX_BEARERS ) {
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 81/161 serial: stm32: fix threaded interrupt handling
In reply to Greg Kroah-HartmanFrom: Johan Hovold <[email protected]>
[ Upstream commit e359b4411c2836cf87c8776682d1b594635570de ]
When DMA is enabled the receive handler runs in a threaded handler, but the primary handler up until very recently neither disabled interrupts in the device or used IRQF_ONESHOT. This would lead to a deadlock if an interrupt comes in while the threaded receive handler is running under the port lock.
Commit ad7676812437 ("serial: stm32: fix a deadlock condition with wakeup event") claimed to fix an unrelated deadlock, but unfortunately also disabled interrupts in the threaded handler. While this prevents the deadlock mentioned in the previous paragraph it also defeats the purpose of using a threaded handler in the first place.
Fix this by making the interrupt one-shot and not disabling interrupts in the threaded handler.
Note that (receive) DMA must not be used for a console port as the threaded handler could be interrupted while holding the port lock, something which could lead to a deadlock in case an interrupt handler ends up calling printk.
--- drivers/tty/serial/stm32-usart.c | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-)
drivers/tty/serial/stm32-usart.c +12 -10
@@ -214,14 +214,11 @@ static void stm32_usart_receive_chars(struct uart_port *port, bool threaded)struct tty_port *tport = &port->state->port;struct stm32_port *stm32_port =to_stm32_port (port);const struct stm32_usart_offsets *ofs = &stm32_port->info->ofs; -unsigned long c, flags; +unsigned long c;u32 sr;char flag; -if (threaded) -spin_lock_irqsave (&port->lock, flags); -else -spin_lock (&port->lock); +spin_lock (&port->lock);while (stm32_usart_pending_rx (port, &sr, &stm32_port->last_res, threaded)) {
@@ -278,10 +275,7 @@ static void stm32_usart_receive_chars(struct uart_port *port, bool threaded)uart_insert_char (port, sr,USART_SR_ORE , c, flag); } -if (threaded) -spin_unlock_irqrestore (&port->lock, flags); -else -spin_unlock (&port->lock); +spin_unlock (&port->lock);tty_flip_buffer_push (tport); }
@@ -654,7 +648,8 @@ static int stm32_usart_startup(struct uart_port *port) ret =request_threaded_irq (port->irq, stm32_usart_interrupt, stm32_usart_threaded_interrupt, -IRQF_NO_SUSPEND , name, port); +IRQF_ONESHOT |IRQF_NO_SUSPEND , + name, port);if (ret)return ret;
@@ -1136,6 +1131,13 @@ static int stm32_usart_of_dma_rx_probe(struct stm32_port *stm32port,struct dma_async_tx_descriptor *desc =NULL ;int ret; +/* +* Using DMA and threaded handler for the console could lead to +* deadlocks. +*/ +if (uart_console (port)) +return -ENODEV ; +/* Request DMA RX channel */ stm32port->rx_ch =dma_request_slave_channel (dev,"rx" );if (!stm32port->rx_ch) {
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 82/161 riscv: vdso: fix and clean-up Makefile
In reply to Greg Kroah-HartmanFrom: Jisheng Zhang <[email protected]>
[ Upstream commit 772d7891e8b3b0baae7bb88a294d61fd07ba6d15 ]
Running "make" on an already compiled kernel tree will rebuild the kernel even without any modifications:
CALL linux/scripts/checksyscalls.sh CALL linux/scripts/atomic/check-atomics.sh CHK include/generated/compile.h SO2S arch/riscv/kernel/vdso/vdso-syms.S AS arch/riscv/kernel/vdso/vdso-syms.o AR arch/riscv/kernel/vdso/built-in.a AR arch/riscv/kernel/built-in.a AR arch/riscv/built-in.a GEN .version CHK include/generated/compile.h UPD include/generated/compile.h CC init/version.o AR init/built-in.a LD vmlinux.o
The reason is "Any target that utilizes if_changed must be listed in $(targets), otherwise the command line check will fail, and the target will always be built" as explained by Documentation/kbuild/makefiles.rst
Fix this build bug by adding vdso-syms.S to $(targets)
At the same time, there are two trivial clean up modifications:
- the vdso-dummy.o is not needed any more after so remove it.
- vdso.lds is a generated file, so it should be prefixed with $(obj)/ instead of $(src)/
--- arch/riscv/kernel/vdso/Makefile | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
arch/riscv/kernel/vdso/Makefile +2 -2
@@ -23,7 +23,7 @@ ifneq ($(c-gettimeofday-y),) endif # Build rules -targets := $(obj-vdso) vdso.so vdso.so.dbg vdso.lds vdso-dummy.o +targets := $(obj-vdso) vdso.so vdso.so.dbg vdso.lds vdso-syms.S obj-vdso := $(addprefix $(obj)/, $(obj-vdso)) obj-y += vdso.o vdso-syms.o
@@ -41,7 +41,7 @@ KASAN_SANITIZE := n $(obj)/vdso.o: $(obj)/vdso.so # link rule for the .so file, .lds has to be first -$(obj)/vdso.so.dbg: $(src)/vdso.lds $(obj-vdso) FORCE +$(obj)/vdso.so.dbg: $(obj)/vdso.lds $(obj-vdso) FORCE $(call if_changed,vdsold) LDFLAGS_vdso.so.dbg = -shared -s -soname=linux-vdso.so.1 \ --build-id=sha1 --hash-style=both --eh-frame-hdr
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 83/161 libceph: dont set global_id until we get an auth ticket
In reply to Greg Kroah-HartmanFrom: Ilya Dryomov <[email protected]>
[ Upstream commit 61ca49a9105faefa003b37542cebad8722f8ae22 ]
With the introduction of enforcing mode, setting global_id as soon as we get it in the first MAuth reply will result in EACCES if the connection is reset before we get the second MAuth reply containing an auth ticket -- because on retry we would attempt to reclaim that global_id with no auth ticket at hand.
Neither ceph_auth_client nor ceph_mon_client depend on global_id being set ealy, so just delay the setting until we get and process the second MAuth reply. While at it, complain if the monitor sends a zero global_id or changes our global_id as the session is likely to fail after that.
--- net/ceph/auth.c | 36 +++++++++++++++++++++++------------- 1 file changed, 23 insertions(+), 13 deletions(-)
net/ceph/auth.c +23 -13
@@ -36,6 +36,20 @@ static int init_protocol(struct ceph_auth_client *ac, int proto) } } +static void set_global_id (struct ceph_auth_client *ac,u64 global_id) +{ +dout ("%s global_id %llu\n" , __func__, global_id); + +if (!global_id) +pr_err ("got zero global_id\n" ); + +if (ac->global_id && global_id != ac->global_id) +pr_err ("global_id changed from %llu to %llu\n" , ac->global_id, + global_id); + + ac->global_id = global_id; +} +/* * setup, teardown. */
@@ -222,11 +236,6 @@ int ceph_handle_auth_reply(struct ceph_auth_client *ac, payload_end = payload + payload_len; -if (global_id && ac->global_id != global_id) { -dout (" set global_id %lld -> %lld\n" , ac->global_id, global_id); - ac->global_id = global_id; - } -if (ac->negotiating) {/* server does not support our protocols? */ if (!protocol && result <0 ) {
@@ -253,11 +262,16 @@ int ceph_handle_auth_reply(struct ceph_auth_client *ac, ret = ac->ops->handle_reply (ac, result, payload, payload_end,NULL ,NULL ,NULL ,NULL ); -if (ret == -EAGAIN ) +if (ret == -EAGAIN ) { ret =build_request (ac, true, reply_buf, reply_len); -else if (ret) + goto out; + }else if (ret) {pr_err ("auth protocol '%s' mauth authentication failed: %d\n" ,ceph_auth_proto_name (ac->protocol), result); + goto out; + } + +set_global_id (ac, global_id); out:mutex_unlock (&ac->mutex);
@@ -484,15 +498,11 @@ int ceph_auth_handle_reply_done(struct ceph_auth_client *ac,int ret;mutex_lock (&ac->mutex); -if (global_id && ac->global_id != global_id) { -dout ("%s global_id %llu -> %llu\n" , __func__, ac->global_id, - global_id); - ac->global_id = global_id; - } - ret = ac->ops->handle_reply (ac,0 , reply, reply + reply_len, session_key, session_key_len, con_secret, con_secret_len); +if (!ret) +set_global_id (ac, global_id);mutex_unlock (&ac->mutex);return ret; }
Signature · 1 line
2.30.2
Re: [PATCH 5.12 083/161] libceph: dont set global_id until we get an auth ticket
Quoted 89 lines · Greg Kroah-Hartman
From: Ilya Dryomov <[email protected]>
[ Upstream commit
61ca49a9105faefa003b37542cebad8722f8ae22]With the introduction of enforcing mode, setting
global_idas soon as we get it in the first MAuth reply will result in EACCES if the connection is reset before we get the second MAuth reply containing an auth ticket -- because on retry we would attempt to reclaim thatglobal_idwith no auth ticket at hand.Neither
ceph_auth_clientnorceph_mon_clientdepend onglobal_idbeing set ealy, so just delay the setting until we get and process the second MAuth reply. While at it, complain if the monitor sends a zeroglobal_idor changes ourglobal_idas the session is likely to fail after that.Cc [email protected] # needs backporting for < 5.11Signed-off-by Ilya Dryomov <[email protected]>Reviewed-by Sage Weil <[email protected]>Signed-off-by Sasha Levin <[email protected]>--- net/ceph/auth.c | 36 +++++++++++++++++++++++------------- 1 file changed, 23 insertions(+), 13 deletions(-)net/ceph/auth.c +21
-11@@ -36,6 +36,20 @@ static int init_protocol(struct ceph_auth_client *ac, int proto) } } +static void set_global_id (struct ceph_auth_client *ac,u64 global_id) +{ +dout ("%s global_id %llu\n" , __func__, global_id); + +if (!global_id) +pr_err ("got zero global_id\n" ); + +if (ac->global_id && global_id != ac->global_id) +pr_err ("global_id changed from %llu to %llu\n" , ac->global_id, + global_id); + + ac->global_id = global_id; +} +/* * setup, teardown. */ @@ -222,11 +236,6 @@ int ceph_handle_auth_reply(struct ceph_auth_client *ac, payload_end = payload + payload_len; -if (global_id && ac->global_id != global_id) { -dout (" set global_id %lld -> %lld\n" , ac->global_id, global_id); - ac->global_id = global_id; - } -if (ac->negotiating) {/* server does not support our protocols? */ if (!protocol && result <0 ) {@@ -253,11 +262,16 @@ int ceph_handle_auth_reply(struct ceph_auth_client *ac, ret = ac->ops->handle_reply (ac, result, payload, payload_end,NULL ,NULL ,NULL ,NULL ); -if (ret == -EAGAIN ) +if (ret == -EAGAIN ) { ret =build_request (ac, true, reply_buf, reply_len); -else if (ret) + goto out; + }else if (ret) {pr_err ("auth protocol '%s' mauth authentication failed: %d\n" ,ceph_auth_proto_name (ac->protocol), result); + goto out; + } + +set_global_id (ac, global_id); out:mutex_unlock (&ac->mutex);@@ -484,15 +498,11 @@ int ceph_auth_handle_reply_done(struct ceph_auth_client *ac,int ret;mutex_lock (&ac->mutex); -if (global_id && ac->global_id != global_id) { -dout ("%s global_id %llu -> %llu\n" , __func__, ac->global_id, - global_id); - ac->global_id = global_id;
net/ceph/auth.c int ceph_auth_handle_reply_done(struct ceph_auth_client *ac, · line 508
Quoted 10 lines
net/ceph/auth.c +2
-2- } - ret = ac->ops->handle_reply (ac,0 , reply, reply + reply_len, session_key, session_key_len, con_secret, con_secret_len); +if (!ret) +set_global_id (ac, global_id);mutex_unlock (&ac->mutex);return ret; }
Hi Greg,
I asked Sasha to drop this patch earlier today.
Thanks,
Ilya
Sasha Levin
maintainerRe: [PATCH 5.12 083/161] libceph: dont set global_id until we get an auth ticket
Quoted 104 lines · Ilya Dryomov
Quoted 89 lines · Greg Kroah-Hartman
From: Ilya Dryomov <[email protected]>
[ Upstream commit
61ca49a9105faefa003b37542cebad8722f8ae22]With the introduction of enforcing mode, setting
global_idas soon as we get it in the first MAuth reply will result in EACCES if the connection is reset before we get the second MAuth reply containing an auth ticket -- because on retry we would attempt to reclaim thatglobal_idwith no auth ticket at hand.Neither
ceph_auth_clientnorceph_mon_clientdepend onglobal_idbeing set ealy, so just delay the setting until we get and process the second MAuth reply. While at it, complain if the monitor sends a zeroglobal_idor changes ourglobal_idas the session is likely to fail after that.Cc [email protected] # needs backporting for < 5.11Signed-off-by Ilya Dryomov <[email protected]>Reviewed-by Sage Weil <[email protected]>Signed-off-by Sasha Levin <[email protected]>--- net/ceph/auth.c | 36 +++++++++++++++++++++++------------- 1 file changed, 23 insertions(+), 13 deletions(-)net/ceph/auth.c +21
-11@@ -36,6 +36,20 @@ static int init_protocol(struct ceph_auth_client *ac, int proto) } } +static void set_global_id (struct ceph_auth_client *ac,u64 global_id) +{ +dout ("%s global_id %llu\n" , __func__, global_id); + +if (!global_id) +pr_err ("got zero global_id\n" ); + +if (ac->global_id && global_id != ac->global_id) +pr_err ("global_id changed from %llu to %llu\n" , ac->global_id, + global_id); + + ac->global_id = global_id; +} +/* * setup, teardown. */ @@ -222,11 +236,6 @@ int ceph_handle_auth_reply(struct ceph_auth_client *ac, payload_end = payload + payload_len; -if (global_id && ac->global_id != global_id) { -dout (" set global_id %lld -> %lld\n" , ac->global_id, global_id); - ac->global_id = global_id; - } -if (ac->negotiating) {/* server does not support our protocols? */ if (!protocol && result <0 ) {@@ -253,11 +262,16 @@ int ceph_handle_auth_reply(struct ceph_auth_client *ac, ret = ac->ops->handle_reply (ac, result, payload, payload_end,NULL ,NULL ,NULL ,NULL ); -if (ret == -EAGAIN ) +if (ret == -EAGAIN ) { ret =build_request (ac, true, reply_buf, reply_len); -else if (ret) + goto out; + }else if (ret) {pr_err ("auth protocol '%s' mauth authentication failed: %d\n" ,ceph_auth_proto_name (ac->protocol), result); + goto out; + } + +set_global_id (ac, global_id); out:mutex_unlock (&ac->mutex);@@ -484,15 +498,11 @@ int ceph_auth_handle_reply_done(struct ceph_auth_client *ac,int ret;mutex_lock (&ac->mutex); -if (global_id && ac->global_id != global_id) { -dout ("%s global_id %llu -> %llu\n" , __func__, ac->global_id, - global_id); - ac->global_id = global_id;net/ceph/auth.c int ceph_auth_handle_reply_done(struct ceph_auth_client *ac, · line 508
Quoted 10 lines
net/ceph/auth.c +2
-2- } - ret = ac->ops->handle_reply (ac,0 , reply, reply + reply_len, session_key, session_key_len, con_secret, con_secret_len); +if (!ret) +set_global_id (ac, global_id);mutex_unlock (&ac->mutex);return ret; }Hi Greg,
Quoted 1 line
I asked Sasha to drop this patch earlier today.
I've dropped it now, but I think I'm missing your previous request. Was it as a reply to the added-to mail? I just want to make sure I'm not missing your mails.
Signature · 2 lines
Thanks, Sasha
Re: [PATCH 5.12 083/161] libceph: dont set global_id until we get an auth ticket
Quoted 106 lines · Sasha Levin
Quoted 104 lines · Ilya Dryomov
Quoted 89 lines · Greg Kroah-Hartman
From: Ilya Dryomov <[email protected]>
[ Upstream commit
61ca49a9105faefa003b37542cebad8722f8ae22]With the introduction of enforcing mode, setting
global_idas soon as we get it in the first MAuth reply will result in EACCES if the connection is reset before we get the second MAuth reply containing an auth ticket -- because on retry we would attempt to reclaim thatglobal_idwith no auth ticket at hand.Neither
ceph_auth_clientnorceph_mon_clientdepend onglobal_idbeing set ealy, so just delay the setting until we get and process the second MAuth reply. While at it, complain if the monitor sends a zeroglobal_idor changes ourglobal_idas the session is likely to fail after that.Cc [email protected] # needs backporting for < 5.11Signed-off-by Ilya Dryomov <[email protected]>Reviewed-by Sage Weil <[email protected]>Signed-off-by Sasha Levin <[email protected]>--- net/ceph/auth.c | 36 +++++++++++++++++++++++------------- 1 file changed, 23 insertions(+), 13 deletions(-)net/ceph/auth.c +21
-11@@ -36,6 +36,20 @@ static int init_protocol(struct ceph_auth_client *ac, int proto) } } +static void set_global_id (struct ceph_auth_client *ac,u64 global_id) +{ +dout ("%s global_id %llu\n" , __func__, global_id); + +if (!global_id) +pr_err ("got zero global_id\n" ); + +if (ac->global_id && global_id != ac->global_id) +pr_err ("global_id changed from %llu to %llu\n" , ac->global_id, + global_id); + + ac->global_id = global_id; +} +/* * setup, teardown. */ @@ -222,11 +236,6 @@ int ceph_handle_auth_reply(struct ceph_auth_client *ac, payload_end = payload + payload_len; -if (global_id && ac->global_id != global_id) { -dout (" set global_id %lld -> %lld\n" , ac->global_id, global_id); - ac->global_id = global_id; - } -if (ac->negotiating) {/* server does not support our protocols? */ if (!protocol && result <0 ) {@@ -253,11 +262,16 @@ int ceph_handle_auth_reply(struct ceph_auth_client *ac, ret = ac->ops->handle_reply (ac, result, payload, payload_end,NULL ,NULL ,NULL ,NULL ); -if (ret == -EAGAIN ) +if (ret == -EAGAIN ) { ret =build_request (ac, true, reply_buf, reply_len); -else if (ret) + goto out; + }else if (ret) {pr_err ("auth protocol '%s' mauth authentication failed: %d\n" ,ceph_auth_proto_name (ac->protocol), result); + goto out; + } + +set_global_id (ac, global_id); out:mutex_unlock (&ac->mutex);@@ -484,15 +498,11 @@ int ceph_auth_handle_reply_done(struct ceph_auth_client *ac,int ret;mutex_lock (&ac->mutex); -if (global_id && ac->global_id != global_id) { -dout ("%s global_id %llu -> %llu\n" , __func__, ac->global_id, - global_id); - ac->global_id = global_id;net/ceph/auth.c int ceph_auth_handle_reply_done(struct ceph_auth_client *ac, · line 508
Quoted 10 lines
net/ceph/auth.c +2
-2- } - ret = ac->ops->handle_reply (ac,0 , reply, reply + reply_len, session_key, session_key_len, con_secret, con_secret_len); +if (!ret) +set_global_id (ac, global_id);mutex_unlock (&ac->mutex);return ret; }Hi Greg,
Quoted 5 lines
Quoted 1 line
I asked Sasha to drop this patch earlier today.
I've dropped it now, but I think I'm missing your previous request. Was it as a reply to the added-to mail? I just want to make sure I'm not missing your mails.
Yes, but it looks like it didn't make it to stable-commits mailing list either. Weird...
MIME-Version: 1.0
Date: Tue, 8 Jun 2021 11:13:08 +0200
References: <[email protected]>
In-Reply-To: <[email protected]>
Message-ID:<CAOi1vP9Ubs1Cu6sW43H-=dVXSzkFZBycfR_Af4b3vJ9mihkAzA@mail.gmail.com>
Subject: Re: Patch "libceph: don't set global_id until we get an
auth ticket" has been added to the 5.12-stable tree
From: Ilya Dryomov <[email protected]>
To: Sasha Levin <[email protected]>
Cc: [email protected]
Content-Type: text/plain; charset="UTF-8"
Thanks,
Ilya
Greg Kroah-Hartman
maintainerPATCH 84/161 amdgpu: fix GEM obj leak in amdgpu_display_user_framebuffer_create
In reply to Greg Kroah-HartmanFrom: Simon Ser <[email protected]>
[ Upstream commit e0c16eb4b3610298a74ae5504c7f6939b12be991 ]
This error code-path is missing a drm_gem_object_put call. Other error code-paths are fine.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_display.c | 1 + 1 file changed, 1 insertion(+)
drivers/gpu/drm/amd/amdgpu/amdgpu_display.c +1 -0
@@ -944,6 +944,7 @@ amdgpu_display_user_framebuffer_create(struct drm_device *dev, domains =amdgpu_display_supported_domains (drm_to_adev (dev), bo->flags);if (obj->import_attach && !(domains &AMDGPU_GEM_DOMAIN_GTT )) {drm_dbg_kms (dev,"Cannot create framebuffer from imported dma_buf\n" ); +drm_gem_object_put (obj);return ERR_PTR (-EINVAL ); }
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 85/161 io_uring: fix link timeout refs
In reply to Greg Kroah-HartmanFrom: Pavel Begunkov <[email protected]>
[ Upstream commit a298232ee6b9a1d5d732aa497ff8be0d45b5bd82 ]
WARNING: CPU: 0 PID: 10242 at lib/refcount.c:28 refcount_warn_saturate+0x15b/0x1a0 lib/refcount.c:28 RIP: 0010:refcount_warn_saturate+0x15b/0x1a0 lib/refcount.c:28 Call Trace: __refcount_sub_and_test include/linux/refcount.h:283 [inline] __refcount_dec_and_test include/linux/refcount.h:315 [inline] refcount_dec_and_test include/linux/refcount.h:333 [inline] io_put_req fs/io_uring.c:2140 [inline] io_queue_linked_timeout fs/io_uring.c:6300 [inline] __io_queue_sqe+0xbef/0xec0 fs/io_uring.c:6354 io_submit_sqe fs/io_uring.c:6534 [inline] io_submit_sqes+0x2bbd/0x7c50 fs/io_uring.c:6660 __do_sys_io_uring_enter fs/io_uring.c:9240 [inline] __se_sys_io_uring_enter+0x256/0x1d60 fs/io_uring.c:9182
io_link_timeout_fn() should put only one reference of the linked timeout request, however in case of racing with the master request's completion first io_req_complete() puts one and then io_put_req_deferred() is called.
--- fs/io_uring.c | 1 + 1 file changed, 1 insertion(+)
fs/io_uring.c +1 -0
@@ -6272,6 +6272,7 @@ static enum hrtimer_restart io_link_timeout_fn(struct hrtimer *timer)if (prev) {io_async_find_and_cancel (ctx, req, prev->user_data, -ETIME );io_put_req_deferred (prev,1 ); +io_put_req_deferred (req,1 ); }else {io_req_complete_post (req, -ETIME ,0 );io_put_req_deferred (req,1 );
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 86/161 io_uring: use better types for cflags
In reply to Greg Kroah-HartmanFrom: Pavel Begunkov <[email protected]>
[ Upstream commit 8c3f9cd1603d0e4af6c50ebc6d974ab7bdd03cf4 ]
__io_cqring_fill_event() takes cflags as long to squeeze it into u32 in an CQE, awhile all users pass int or unsigned. Replace it with unsigned int and store it as u32 in struct io_completion to match CQE.
--- fs/io_uring.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
fs/io_uring.c +3 -2
@@ -653,7 +653,7 @@ struct io_unlink {struct io_completion {struct file *file;struct list_head list; -int cflags; +u32 cflags; };struct io_async_connect {
@@ -1476,7 +1476,8 @@ static bool io_cqring_overflow_flush(struct io_ring_ctx *ctx, bool force,return ret; } -static void __io_cqring_fill_event (struct io_kiocb *req,long res,long cflags) +static void __io_cqring_fill_event (struct io_kiocb *req,long res, +unsigned int cflags) {struct io_ring_ctx *ctx = req->ctx;struct io_uring_cqe *cqe;
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 87/161 io_uring: wrap io_kiocb reference count manipulation in helpers
In reply to Greg Kroah-HartmanFrom: Jens Axboe <[email protected]>
[ Upstream commit de9b4ccad750f216616730b74ed2be16c80892a4 ]
No functional changes in this patch, just in preparation for handling the references a bit more efficiently.
--- fs/io_uring.c | 55 +++++++++++++++++++++++++++++++++++++-------------- 1 file changed, 40 insertions(+), 15 deletions(-)
fs/io_uring.c +40 -15
@@ -1476,6 +1476,31 @@ static bool io_cqring_overflow_flush(struct io_ring_ctx *ctx, bool force,return ret; } +static inline bool req_ref_inc_not_zero (struct io_kiocb *req) +{ +return refcount_inc_not_zero (&req->refs); +} + +static inline bool req_ref_sub_and_test (struct io_kiocb *req,int refs) +{ +return refcount_sub_and_test (refs, &req->refs); +} + +static inline bool req_ref_put_and_test (struct io_kiocb *req) +{ +return refcount_dec_and_test (&req->refs); +} + +static inline void req_ref_put (struct io_kiocb *req) +{ +refcount_dec (&req->refs); +} + +static inline void req_ref_get (struct io_kiocb *req) +{ +refcount_inc (&req->refs); +} +static void __io_cqring_fill_event (struct io_kiocb *req,long res,unsigned int cflags) {
@@ -1512,7 +1537,7 @@ static void __io_cqring_fill_event(struct io_kiocb *req, long res,io_clean_op (req); req->result = res; req->compl.cflags = cflags; -refcount_inc (&req->refs); +req_ref_get (req);list_add_tail (&req->compl.list, &ctx->cq_overflow_list); } }
@@ -1534,7 +1559,7 @@ static void io_req_complete_post(struct io_kiocb *req, long res,* If we're the last reference to this request, add to our locked * free_list cache. */ -if (refcount_dec_and_test (&req->refs)) { +if (req_ref_put_and_test (req)) {struct io_comp_state *cs = &ctx->submit_state.comp;if (req->flags & (REQ_F_LINK |REQ_F_HARDLINK )) {
@@ -2113,7 +2138,7 @@ static void io_submit_flush_completions(struct io_comp_state *cs, req = cs->reqs[i];/* submission and completion refs */ -if (refcount_sub_and_test (2 , &req->refs)) +if (req_ref_sub_and_test (req,2 ))io_req_free_batch (&rb, req, &ctx->submit_state); }
@@ -2129,7 +2154,7 @@ static struct io_kiocb *io_put_req_find_next(struct io_kiocb *req) {struct io_kiocb *nxt =NULL ; -if (refcount_dec_and_test (&req->refs)) { +if (req_ref_put_and_test (req)) { nxt =io_req_find_next (req);__io_free_req (req); }
@@ -2138,7 +2163,7 @@ static struct io_kiocb *io_put_req_find_next(struct io_kiocb *req)static void io_put_req (struct io_kiocb *req) { -if (refcount_dec_and_test (&req->refs)) +if (req_ref_put_and_test (req))io_free_req (req); }
@@ -2161,14 +2186,14 @@ static void io_free_req_deferred(struct io_kiocb *req)static inline void io_put_req_deferred (struct io_kiocb *req,int refs) { -if (refcount_sub_and_test (refs, &req->refs)) +if (req_ref_sub_and_test (req, refs))io_free_req_deferred (req); }static void io_double_put_req (struct io_kiocb *req) {/* drop both submit and complete references */ -if (refcount_sub_and_test (2 , &req->refs)) +if (req_ref_sub_and_test (req,2 ))io_free_req (req); }
@@ -2254,7 +2279,7 @@ static void io_iopoll_complete(struct io_ring_ctx *ctx, unsigned int *nr_events,__io_cqring_fill_event (req, req->result, cflags); (*nr_events)++; -if (refcount_dec_and_test (&req->refs)) +if (req_ref_put_and_test (req))io_req_free_batch (&rb, req, &ctx->submit_state); }
@@ -2496,7 +2521,7 @@ static bool io_rw_reissue(struct io_kiocb *req)lockdep_assert_held (&req->ctx->uring_lock);if (io_resubmit_prep (req)) { -refcount_inc (&req->refs); +req_ref_get (req);io_queue_async_work (req);return true; }
@@ -3209,7 +3234,7 @@ static int io_async_buf_func(struct wait_queue_entry *wait, unsigned mode,list_del_init (&wait->entry);/* submit ref gets dropped, acquire a new one */ -refcount_inc (&req->refs); +req_ref_get (req);io_req_task_queue (req);return 1 ; }
@@ -4954,7 +4979,7 @@ static void io_poll_remove_double(struct io_kiocb *req)spin_lock (&head->lock);list_del_init (&poll->wait.entry);if (poll->wait.private) -refcount_dec (&req->refs); +req_ref_put (req); poll->head =NULL ;spin_unlock (&head->lock); }
@@ -5020,7 +5045,7 @@ static int io_poll_double_wake(struct wait_queue_entry *wait, unsigned mode, poll->wait.func (&poll->wait, mode, sync, key); } } -refcount_dec (&req->refs); +req_ref_put (req);return 1 ; }
@@ -5063,7 +5088,7 @@ static void __io_queue_proc(struct io_poll_iocb *poll, struct io_poll_table *pt,return ; }io_init_poll_iocb (poll, poll_one->events, io_poll_double_wake); -refcount_inc (&req->refs); +req_ref_get (req); poll->wait.private = req; *poll_ptr = poll; }
@@ -6212,7 +6237,7 @@ static void io_wq_submit_work(struct io_wq_work *work)/* avoid locking problems by failing it from a clean context */ if (ret) {/* io-wq is going to take one down */ -refcount_inc (&req->refs); +req_ref_get (req);io_req_task_queue_fail (req, ret); } }
@@ -6264,7 +6289,7 @@ static enum hrtimer_restart io_link_timeout_fn(struct hrtimer *timer)* We don't expect the list to be empty, that will only happen if we * race with the completion of the linked work. */ -if (prev &&refcount_inc_not_zero (&prev->refs)) +if (prev &&req_ref_inc_not_zero (prev))io_remove_next_linked (prev);else prev =NULL ;
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 88/161 io_uring: fix ltout double free on completion race
In reply to Greg Kroah-HartmanFrom: Pavel Begunkov <[email protected]>
[ Upstream commit 447c19f3b5074409c794b350b10306e1da1ef4ba ]
Always remove linked timeout on io_link_timeout_fn() from the master request link list, otherwise we may get use-after-free when first io_link_timeout_fn() puts linked timeout in the fail path, and then will be found and put on master's free.
--- fs/io_uring.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-)
fs/io_uring.c +4 -3
@@ -6289,10 +6289,11 @@ static enum hrtimer_restart io_link_timeout_fn(struct hrtimer *timer)* We don't expect the list to be empty, that will only happen if we * race with the completion of the linked work. */ -if (prev &&req_ref_inc_not_zero (prev)) +if (prev) {io_remove_next_linked (prev); -else - prev =NULL ; +if (!req_ref_inc_not_zero (prev)) + prev =NULL ; + }spin_unlock_irqrestore (&ctx->completion_lock, flags);if (prev) {
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 89/161 drm/amdgpu/vcn3: add cancel_delayed_work_sync before power gate
In reply to Greg Kroah-HartmanFrom: James Zhu <[email protected]>
[ Upstream commit 4a62542ae064e3b645d6bbf2295a6c05136956c6 ]
Add cancel_delayed_work_sync before set power gating state to avoid race condition issue when power gating.
--- drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/gpu/drm/amd/amdgpu/vcn_v3_0.c +2 -3
@@ -367,15 +367,14 @@ done:static int vcn_v3_0_hw_fini (void *handle) {struct amdgpu_device *adev = (struct amdgpu_device *)handle; -struct amdgpu_ring *ring;int i; +cancel_delayed_work_sync (&adev->vcn.idle_work); +for (i =0 ; i < adev->vcn.num_vcn_inst; ++i) {if (adev->vcn.harvest_config & (1 << i))continue ; - ring = &adev->vcn.inst[i].ring_dec; -if (!amdgpu_sriov_vf (adev)) { if ((adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG) || (adev->vcn.cur_state != AMD_PG_STATE_GATE &&
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 90/161 drm/amdgpu/jpeg2.5: add cancel_delayed_work_sync before power gate
In reply to Greg Kroah-HartmanFrom: James Zhu <[email protected]>
[ Upstream commit 23f10a571da5eaa63b7845d16e2f49837e841ab9 ]
Add cancel_delayed_work_sync before set power gating state to avoid race condition issue when power gating.
--- drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/amd/amdgpu/jpeg_v2_5.c +2 -2
@@ -187,14 +187,14 @@ static int jpeg_v2_5_hw_init(void *handle)static int jpeg_v2_5_hw_fini (void *handle) {struct amdgpu_device *adev = (struct amdgpu_device *)handle; -struct amdgpu_ring *ring;int i; +cancel_delayed_work_sync (&adev->vcn.idle_work); +for (i =0 ; i < adev->jpeg.num_jpeg_inst; ++i) {if (adev->jpeg.harvest_config & (1 << i))continue ; - ring = &adev->jpeg.inst[i].ring_dec;if (adev->jpeg.cur_state !=AMD_PG_STATE_GATE &&RREG32_SOC15 (JPEG , i, mmUVD_JRBC_STATUS))jpeg_v2_5_set_powergating_state (adev,AMD_PG_STATE_GATE );
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 91/161 drm/amdgpu/jpeg3: add cancel_delayed_work_sync before power gate
In reply to Greg Kroah-HartmanFrom: James Zhu <[email protected]>
[ Upstream commit 20ebbfd22f8115a1e4f60d3d289f66be4d47f1ec ]
Add cancel_delayed_work_sync before set power gating state to avoid race condition issue when power gating.
--- drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/amd/amdgpu/jpeg_v3_0.c +2 -2
@@ -159,9 +159,9 @@ static int jpeg_v3_0_hw_init(void *handle)static int jpeg_v3_0_hw_fini (void *handle) {struct amdgpu_device *adev = (struct amdgpu_device *)handle; -struct amdgpu_ring *ring; - ring = &adev->jpeg.inst->ring_dec; +cancel_delayed_work_sync (&adev->vcn.idle_work); +if (adev->jpeg.cur_state !=AMD_PG_STATE_GATE &&RREG32_SOC15 (JPEG ,0 , mmUVD_JRBC_STATUS))jpeg_v3_0_set_powergating_state (adev,AMD_PG_STATE_GATE );
Signature · 1 line
2.30.2
Greg Kroah-Hartman
maintainerPATCH 92/161 Bluetooth: fix the erroneous flush_work() order
In reply to Greg Kroah-HartmanFrom: Lin Ma <[email protected]>
commit 6a137caec23aeb9e036cdfd8a46dd8a366460e5d upstream.
In the cleanup routine for failed initialization of HCI device, the flush_work(&hdev->rx_work) need to be finished before the flush_work(&hdev->cmd_work). Otherwise, the hci_rx_work() can possibly invoke new cmd_work and cause a bug, like double free, in late processings.
This was assigned CVE-2021-3564.
This patch reorder the flush_work() to fix this bug.
--- net/bluetooth/hci_core.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
net/bluetooth/hci_core.c +6 -1
@@ -1608,8 +1608,13 @@ setup_failed: }else {/* Init failed, cleanup */ flush_work (&hdev->tx_work); -flush_work (&hdev->cmd_work); + +/* Since hci_rx_work() is possible to awake new cmd_work +* it should be flushed first to avoid unexpected call of +* hci_cmd_work() +*/ flush_work (&hdev->rx_work); +flush_work (&hdev->cmd_work);skb_queue_purge (&hdev->cmd_q);skb_queue_purge (&hdev->rx_q);
Greg Kroah-Hartman
maintainerPATCH 93/161 Bluetooth: use correct lock to prevent UAF of hdev object
In reply to Greg Kroah-HartmanFrom: Lin Ma <[email protected]>
commit e305509e678b3a4af2b3cfd410f409f7cdaabb52 upstream.
The hci_sock_dev_event() function will cleanup the hdev object for sockets even if this object may still be in used within the hci_sock_bound_ioctl() function, result in UAF vulnerability.
This patch replace the BH context lock to serialize these affairs and prevent the race condition.
--- net/bluetooth/hci_sock.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
net/bluetooth/hci_sock.c +2 -2
@@ -762,7 +762,7 @@ void hci_sock_dev_event(struct hci_dev */* Detach sockets from device */ read_lock (&hci_sk_list.lock);sk_for_each (sk, &hci_sk_list.head) { -bh_lock_sock_nested (sk); +lock_sock (sk);if (hci_pi (sk)->hdev == hdev) {hci_pi (sk)->hdev =NULL ; sk->sk_err =EPIPE ;
@@ -771,7 +771,7 @@ void hci_sock_dev_event(struct hci_dev *hci_dev_put (hdev); } -bh_unlock_sock (sk); +release_sock (sk); }read_unlock (&hci_sk_list.lock); }
Greg Kroah-Hartman
maintainerPATCH 94/161 wireguard: do not use -O3
In reply to Greg Kroah-HartmanFrom: Jason A. Donenfeld <[email protected]>
commit cc5060ca0285efe2728bced399a1955a7ce808b2 upstream.
Apparently, various versions of gcc have O3-related miscompiles. Looking at the difference between -O2 and -O3 for gcc 11 doesn't indicate miscompiles, but the difference also doesn't seem so significant for performance that it's worth risking.
--- drivers/net/wireguard/Makefile | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
drivers/net/wireguard/Makefile +1 -2
@@ -1,5 +1,4 @@ -ccflags-y := -O3 -ccflags-y += -D'pr_fmt(fmt)=KBUILD_MODNAME ": " fmt' +ccflags-y := -D'pr_fmt(fmt)=KBUILD_MODNAME ": " fmt' ccflags-$(CONFIG_WIREGUARD_DEBUG) += -DDEBUG wireguard-y := main.o wireguard-y += noise.o
Greg Kroah-Hartman
maintainerPATCH 95/161 wireguard: peer: allocate in kmem_cache
In reply to Greg Kroah-HartmanFrom: Jason A. Donenfeld <[email protected]>
commit a4e9f8e3287c9eb6bf70df982870980dd3341863 upstream.
With deployments having upwards of 600k peers now, this somewhat heavy structure could benefit from more fine-grained allocations. Specifically, instead of using a 2048-byte slab for a 1544-byte object, we can now use 1544-byte objects directly, thus saving almost 25% per-peer, or with 600k peers, that's a savings of 303 MiB. This also makes wireguard's memory usage more transparent in tools like slabtop and /proc/slabinfo.
--- drivers/net/wireguard/main.c | 7 +++++++ drivers/net/wireguard/peer.c | 21 +++++++++++++++++---- drivers/net/wireguard/peer.h | 3 +++ 3 files changed, 27 insertions(+), 4 deletions(-)
drivers/net/wireguard/main.c +7 -0
@@ -28,6 +28,10 @@ static int __init mod_init(void)#endif wg_noise_init (); + ret =wg_peer_init (); +if (ret <0 ) + goto err_peer; + ret =wg_device_init ();if (ret <0 ) goto err_device;
@@ -44,6 +48,8 @@ static int __init mod_init(void) err_netlink:wg_device_uninit (); err_device: +wg_peer_uninit (); +err_peer:return ret; }
@@ -51,6 +57,7 @@ static void __exit mod_exit(void) {wg_genetlink_uninit ();wg_device_uninit (); +wg_peer_uninit (); }module_init (mod_init);
drivers/net/wireguard/peer.c +17 -4
@@ -15,6 +15,7 @@#include <linux/rcupdate.h> #include <linux/list.h> +static struct kmem_cache *peer_cache;static atomic64_t peer_counter =ATOMIC64_INIT (0 ); struct wg_peer *wg_peer_create(struct wg_device *wg,
@@ -29,10 +30,10 @@ struct wg_peer *wg_peer_create(struct wgif (wg->num_peers >=MAX_PEERS_PER_DEVICE )return ERR_PTR (ret); - peer =kzalloc (sizeof (*peer),GFP_KERNEL ); + peer =kmem_cache_zalloc (peer_cache,GFP_KERNEL );if (unlikely (!peer))return ERR_PTR (ret); -if (dst_cache_init (&peer->endpoint_cache,GFP_KERNEL )) +if (unlikely (dst_cache_init (&peer->endpoint_cache,GFP_KERNEL ))) goto err; peer->device = wg;
@@ -64,7 +65,7 @@ struct wg_peer *wg_peer_create(struct wgreturn peer; err: -kfree (peer); +kmem_cache_free (peer_cache, peer);return ERR_PTR (ret); }
@@ -193,7 +194,8 @@ static void rcu_release(struct rcu_head/* The final zeroing takes care of clearing any remaining handshake key * material and other potentially sensitive information. */ -kfree_sensitive (peer); +memzero_explicit (peer,sizeof (*peer)); +kmem_cache_free (peer_cache, peer); }static void kref_release (struct kref *refcount)
@@ -225,3 +227,14 @@ void wg_peer_put(struct wg_peer *peer)return ;kref_put (&peer->refcount, kref_release); } + +int __init wg_peer_init (void ) +{ + peer_cache =KMEM_CACHE (wg_peer,0 ); +return peer_cache ?0 : -ENOMEM ; +} + +void wg_peer_uninit (void ) +{ +kmem_cache_destroy (peer_cache); +}
drivers/net/wireguard/peer.h +3 -0
@@ -80,4 +80,7 @@ void wg_peer_put(struct wg_peer *peer);void wg_peer_remove (struct wg_peer *peer);void wg_peer_remove_all (struct wg_device *wg); +int wg_peer_init (void ); +void wg_peer_uninit (void ); +#endif /* _WG_PEER_H */
Greg Kroah-Hartman
maintainerPATCH 96/161 wireguard: use synchronize_net rather than synchronize_rcu
In reply to Greg Kroah-HartmanFrom: Jason A. Donenfeld <[email protected]>
commit 24b70eeeb4f46c09487f8155239ebfb1f875774a upstream.
Many of the synchronization points are sometimes called under the rtnl lock, which means we should use synchronize_net rather than synchronize_rcu. Under the hood, this expands to using the expedited flavor of function in the event that rtnl is held, in order to not stall other concurrent changes.
This fixes some very, very long delays when removing multiple peers at once, which would cause some operations to take several minutes.
--- drivers/net/wireguard/peer.c | 6 +++--- drivers/net/wireguard/socket.c | 2 +- 2 files changed, 4 insertions(+), 4 deletions(-)
drivers/net/wireguard/peer.c +3 -3
@@ -89,7 +89,7 @@ static void peer_make_dead(struct wg_pee/* Mark as dead, so that we don't allow jumping contexts after. */ WRITE_ONCE (peer->is_dead, true); -/* The caller must now synchronize_rcu() for this to take effect. */ +/* The caller must now synchronize_net() for this to take effect. */ }static void peer_remove_after_dead (struct wg_peer *peer)
@@ -161,7 +161,7 @@ void wg_peer_remove(struct wg_peer *peerlockdep_assert_held (&peer->device->device_update_lock);peer_make_dead (peer); -synchronize_rcu (); +synchronize_net ();peer_remove_after_dead (peer); }
@@ -179,7 +179,7 @@ void wg_peer_remove_all(struct wg_devicepeer_make_dead (peer);list_add_tail (&peer->peer_list, &dead_peers); } -synchronize_rcu (); +synchronize_net ();list_for_each_entry_safe (peer, temp, &dead_peers, peer_list)peer_remove_after_dead (peer); }
drivers/net/wireguard/socket.c +1 -1
@@ -430,7 +430,7 @@ void wg_socket_reinit(struct wg_device *if (new4) wg->incoming_port =ntohs (inet_sk (new4)->inet_sport);mutex_unlock (&wg->socket_update_lock); -synchronize_rcu (); +synchronize_net ();sock_free (old4);sock_free (old6); }
Greg Kroah-Hartman
maintainerPATCH 97/161 wireguard: selftests: remove old conntrack kconfig value
In reply to Greg Kroah-HartmanFrom: Jason A. Donenfeld <[email protected]>
commit acf2492b51c9a3c4dfb947f4d3477a86d315150f upstream.
On recent kernels, this config symbol is no longer used.
--- tools/testing/selftests/wireguard/qemu/kernel.config | 1 - 1 file changed, 1 deletion(-)
tools/testing/selftests/wireguard/qemu/kernel.config +0 -1
@@ -19,7 +19,6 @@ CONFIG_NETFILTER_XTABLES=y CONFIG_NETFILTER_XT_NAT=y CONFIG_NETFILTER_XT_MATCH_LENGTH=y CONFIG_NETFILTER_XT_MARK=y -CONFIG_NF_CONNTRACK_IPV4=y CONFIG_NF_NAT_IPV4=y CONFIG_IP_NF_IPTABLES=y CONFIG_IP_NF_FILTER=y
Greg Kroah-Hartman
maintainerPATCH 98/161 wireguard: selftests: make sure rp_filter is disabled on vethc
In reply to Greg Kroah-HartmanFrom: Jason A. Donenfeld <[email protected]>
commit f8873d11d4121aad35024f9379e431e0c83abead upstream.
Some distros may enable strict rp_filter by default, which will prevent vethc from receiving the packets with an unrouteable reverse path address.
--- tools/testing/selftests/wireguard/netns.sh | 1 + 1 file changed, 1 insertion(+)
tools/testing/selftests/wireguard/netns.sh +1 -0
@@ -363,6 +363,7 @@ ip1 -6 rule add table main suppress_pref ip1 -4 route add default dev wg0 table 51820 ip1 -4 rule add not fwmark 51820 table 51820 ip1 -4 rule add table main suppress_prefixlength 0 +n1 bash -c 'printf 0 > /proc/sys/net/ipv4/conf/vethc/rp_filter' # Flood the pings instead of sending just one, to trigger routing table reference counting bugs. n1 ping -W 1 -c 100 -f 192.168.99.7 n1 ping -W 1 -c 100 -f abab::1111
Greg Kroah-Hartman
maintainerPATCH 99/161 wireguard: allowedips: initialize list head in selftest
In reply to Greg Kroah-HartmanFrom: Jason A. Donenfeld <[email protected]>
commit 46cfe8eee285cde465b420637507884551f5d7ca upstream.
The randomized trie tests weren't initializing the dummy peer list head, resulting in a NULL pointer dereference when used. Fix this by initializing it in the randomized trie test, just like we do for the static unit test.
While we're at it, all of the other strings like this have the word "self-test", so add it to the missing place here.
--- drivers/net/wireguard/selftest/allowedips.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/wireguard/selftest/allowedips.c +2 -1
@@ -296,6 +296,7 @@ static __init bool randomized_test(void) goto free; }kref_init (&peers[i]->refcount); +INIT_LIST_HEAD (&peers[i]->allowedips_list); }mutex_lock (&mutex);
@@ -333,7 +334,7 @@ static __init bool randomized_test(void)if (wg_allowedips_insert_v4 (&t, (struct in_addr *)mutated, cidr, peer, &mutex) <0 ) { -pr_err ("allowedips random malloc: FAIL\n" ); +pr_err ("allowedips random self-test malloc: FAIL\n" ); goto free_locked; } if (horrible_allowedips_insert_v4(&h,
Greg Kroah-Hartman
maintainerPATCH 100/161 wireguard: allowedips: remove nodes in O(1)
In reply to Greg Kroah-HartmanFrom: Jason A. Donenfeld <[email protected]>
commit f634f418c227c912e7ea95a3299efdc9b10e4022 upstream.
Previously, deleting peers would require traversing the entire trie in order to rebalance nodes and safely free them. This meant that removing 1000 peers from a trie with a half million nodes would take an extremely long time, during which we're holding the rtnl lock. Large-scale users were reporting 200ms latencies added to the networking stack as a whole every time their userspace software would queue up significant removals. That's a serious situation.
This commit fixes that by maintaining a double pointer to the parent's bit pointer for each node, and then using the already existing node list belonging to each peer to go directly to the node, fix up its pointers, and free it with RCU. This means removal is O(1) instead of O(n), and we don't use gobs of stack.
The removal algorithm has the same downside as the code that it fixes:
it won't collapse needlessly long runs of fillers. We can enhance that in the future if it ever becomes a problem. This commit documents that limitation with a TODO comment in code, a small but meaningful improvement over the prior situation.
Currently the biggest flaw, which the next commit addresses, is that because this increases the node size on 64-bit machines from 60 bytes to 68 bytes. 60 rounds up to 64, but 68 rounds up to 128. So we wind up using twice as much memory per node, because of power-of-two allocations, which is a big bummer. We'll need to figure something out there.
--- drivers/net/wireguard/allowedips.c | 130 +++++++++++++++---------------------- drivers/net/wireguard/allowedips.h | 9 -- 2 files changed, 56 insertions(+), 83 deletions(-)
drivers/net/wireguard/allowedips.c +53 -77
@@ -66,60 +66,6 @@ static void root_remove_peer_lists(struc } } -static void walk_remove_by_peer (struct allowedips_node __rcu **top, -struct wg_peer *peer,struct mutex *lock) -{ -#define REF (p) rcu_access_pointer(p) -#define DEREF (p) rcu_dereference_protected(*(p), lockdep_is_held(lock)) -#define PUSH (p) ({ \ - WARN_ON(IS_ENABLED(DEBUG) && len >= 128); \ - stack[len++] = p; \ - }) - -struct allowedips_node __rcu **stack[128 ], **nptr; -struct allowedips_node *node, *prev; -unsigned int len; - -if (unlikely (!peer || !REF (*top))) -return ; - -for (prev =NULL , len =0 ,PUSH (top); len >0 ; prev = node) { - nptr = stack[len -1 ]; - node =DEREF (nptr); -if (!node) { - --len; -continue ; - } -if (!prev ||REF (prev->bit[0 ]) == node || -REF (prev->bit[1 ]) == node) { -if (REF (node->bit[0 ])) -PUSH (&node->bit[0 ]); -else if (REF (node->bit[1 ])) -PUSH (&node->bit[1 ]); - }else if (REF (node->bit[0 ]) == prev) { -if (REF (node->bit[1 ])) -PUSH (&node->bit[1 ]); - }else { -if (rcu_dereference_protected (node->peer, -lockdep_is_held (lock)) == peer) { -RCU_INIT_POINTER (node->peer,NULL ); -list_del_init (&node->peer_list); -if (!node->bit[0 ] || !node->bit[1 ]) { -rcu_assign_pointer (*nptr,DEREF ( - &node->bit[!REF (node->bit[0 ])])); -kfree_rcu (node, rcu); - node =DEREF (nptr); - } - } - --len; - } - } - -#undef REF -#undef DEREF -#undef PUSH -} -static unsigned int fls128 (u64 a,u64 b) {return a ?fls64 (a) +64U :fls64 (b);
@@ -224,6 +170,7 @@ static int add(struct allowedips_node __RCU_INIT_POINTER (node->peer, peer);list_add_tail (&node->peer_list, &peer->allowedips_list);copy_and_assign_cidr (node, key, cidr, bits); +rcu_assign_pointer (node->parent_bit, trie);rcu_assign_pointer (*trie, node);return 0 ; }
@@ -243,9 +190,9 @@ static int add(struct allowedips_node __if (!node) { down =rcu_dereference_protected (*trie,lockdep_is_held (lock)); }else { - down =rcu_dereference_protected (CHOOSE_NODE (node, key), -lockdep_is_held (lock)); + down =rcu_dereference_protected (CHOOSE_NODE (node, key),lockdep_is_held (lock));if (!down) { +rcu_assign_pointer (newnode->parent_bit, &CHOOSE_NODE (node, key));rcu_assign_pointer (CHOOSE_NODE (node, key), newnode);return 0 ; }
@@ -254,29 +201,37 @@ static int add(struct allowedips_node __ parent = node;if (newnode->cidr == cidr) { +rcu_assign_pointer (down->parent_bit, &CHOOSE_NODE (newnode, down->bits));rcu_assign_pointer (CHOOSE_NODE (newnode, down->bits), down); -if (!parent) +if (!parent) { +rcu_assign_pointer (newnode->parent_bit, trie);rcu_assign_pointer (*trie, newnode); -else -rcu_assign_pointer (CHOOSE_NODE (parent, newnode->bits), - newnode); - }else { - node =kzalloc (sizeof (*node),GFP_KERNEL ); -if (unlikely (!node)) { -list_del (&newnode->peer_list); -kfree (newnode); -return -ENOMEM ; + }else { +rcu_assign_pointer (newnode->parent_bit, &CHOOSE_NODE (parent, newnode->bits)); +rcu_assign_pointer (CHOOSE_NODE (parent, newnode->bits), newnode); } -INIT_LIST_HEAD (&node->peer_list); -copy_and_assign_cidr (node, newnode->bits, cidr, bits); +return 0 ; + } + + node =kzalloc (sizeof (*node),GFP_KERNEL ); +if (unlikely (!node)) { +list_del (&newnode->peer_list); +kfree (newnode); +return -ENOMEM ; + } +INIT_LIST_HEAD (&node->peer_list); +copy_and_assign_cidr (node, newnode->bits, cidr, bits); -rcu_assign_pointer (CHOOSE_NODE (node, down->bits), down); -rcu_assign_pointer (CHOOSE_NODE (node, newnode->bits), newnode); -if (!parent) -rcu_assign_pointer (*trie, node); -else -rcu_assign_pointer (CHOOSE_NODE (parent, node->bits), - node); +rcu_assign_pointer (down->parent_bit, &CHOOSE_NODE (node, down->bits)); +rcu_assign_pointer (CHOOSE_NODE (node, down->bits), down); +rcu_assign_pointer (newnode->parent_bit, &CHOOSE_NODE (node, newnode->bits)); +rcu_assign_pointer (CHOOSE_NODE (node, newnode->bits), newnode); +if (!parent) { +rcu_assign_pointer (node->parent_bit, trie); +rcu_assign_pointer (*trie, node); + }else { +rcu_assign_pointer (node->parent_bit, &CHOOSE_NODE (parent, node->bits)); +rcu_assign_pointer (CHOOSE_NODE (parent, node->bits), node); }return 0 ; }
@@ -335,9 +290,30 @@ int wg_allowedips_insert_v6(struct allowvoid wg_allowedips_remove_by_peer (struct allowedips *table,struct wg_peer *peer,struct mutex *lock) { +struct allowedips_node *node, *child, *tmp; + +if (list_empty (&peer->allowedips_list)) +return ; ++table->seq; -walk_remove_by_peer (&table->root4, peer, lock); -walk_remove_by_peer (&table->root6, peer, lock); +list_for_each_entry_safe (node, tmp, &peer->allowedips_list, peer_list) { +list_del_init (&node->peer_list); +RCU_INIT_POINTER (node->peer,NULL ); +if (node->bit[0 ] && node->bit[1 ]) +continue ; + child =rcu_dereference_protected ( + node->bit[!rcu_access_pointer (node->bit[0 ])], +lockdep_is_held (lock)); +if (child) + child->parent_bit = node->parent_bit; + *rcu_dereference_protected (node->parent_bit,lockdep_is_held (lock)) = child; +kfree_rcu (node, rcu); + +/* TODO: Note that we currently don't walk up and down in order to +* free any potential filler nodes. This means that this function +* doesn't free up as much as it could, which could be revisited +* at some point. +*/ + } }int wg_allowedips_read_node (struct allowedips_node *node,u8 ip[16 ],u8 *cidr)
drivers/net/wireguard/allowedips.h +3 -6
@@ -15,14 +15,11 @@ struct wg_peer;struct allowedips_node {struct wg_peer __rcu *peer;struct allowedips_node __rcu *bit[2 ]; -/* While it may seem scandalous that we waste space for v4, -* we're alloc'ing to the nearest power of 2 anyway, so this -* doesn't actually make a difference. -*/ -u8 bits[16 ]__aligned(__alignof(u64)) ;u8 cidr, bit_at_a, bit_at_b, bitlen; +u8 bits[16 ]__aligned(__alignof(u64)) ; -/* Keep rarely used list at bottom to be beyond cache line. */ +/* Keep rarely used members at bottom to be beyond cache line. */ +struct allowedips_node *__rcu *parent_bit;/* XXX: this puts us at 68->128 bytes instead of 60->64 bytes!! */ union {struct list_head peer_list;struct rcu_head rcu;
Greg Kroah-Hartman
maintainerPATCH 101/161 wireguard: allowedips: allocate nodes in kmem_cache
In reply to Greg Kroah-HartmanFrom: Jason A. Donenfeld <[email protected]>
commit dc680de28ca849dfe589dc15ac56d22505f0ef11 upstream.
The previous commit moved from O(n) to O(1) for removal, but in the process introduced an additional pointer member to a struct that increased the size from 60 to 68 bytes, putting nodes in the 128-byte slab. With deployed systems having as many as 2 million nodes, this represents a significant doubling in memory usage (128 MiB -> 256 MiB). Fix this by using our own kmem_cache, that's sized exactly right. This also makes wireguard's memory usage more transparent in tools like slabtop and /proc/slabinfo.
--- drivers/net/wireguard/allowedips.c | 31 +++++++++++++++++++++++++------ drivers/net/wireguard/allowedips.h | 5 ++++- drivers/net/wireguard/main.c | 10 +++++++++- 3 files changed, 38 insertions(+), 8 deletions(-)
drivers/net/wireguard/allowedips.c +25 -6
@@ -6,6 +6,8 @@#include "allowedips.h" #include "peer.h" +static struct kmem_cache *node_cache; +static void swap_endian (u8 *dst,const u8 *src,u8 bits) {if (bits ==32 ) {
@@ -40,6 +42,11 @@ static void push_rcu(struct allowedips_n } } +static void node_free_rcu (struct rcu_head *rcu) +{ + kmem_cache_free(node_cache, container_of(rcu, struct allowedips_node, rcu)); +} +static void root_free_rcu (struct rcu_head *rcu) {struct allowedips_node *node, *stack[128 ] = {
@@ -49,7 +56,7 @@ static void root_free_rcu(struct rcu_heawhile (len >0 && (node = stack[--len])) {push_rcu (stack, node->bit[0 ], &len);push_rcu (stack, node->bit[1 ], &len); -kfree (node); +kmem_cache_free (node_cache, node); } }
@@ -164,7 +171,7 @@ static int add(struct allowedips_node __return -EINVAL ;if (!rcu_access_pointer (*trie)) { - node =kzalloc (sizeof (*node),GFP_KERNEL ); + node =kmem_cache_zalloc (node_cache,GFP_KERNEL );if (unlikely (!node))return -ENOMEM ;RCU_INIT_POINTER (node->peer, peer);
@@ -180,7 +187,7 @@ static int add(struct allowedips_node __return 0 ; } - newnode =kzalloc (sizeof (*newnode),GFP_KERNEL ); + newnode =kmem_cache_zalloc (node_cache,GFP_KERNEL );if (unlikely (!newnode))return -ENOMEM ;RCU_INIT_POINTER (newnode->peer, peer);
@@ -213,10 +220,10 @@ static int add(struct allowedips_node __return 0 ; } - node =kzalloc (sizeof (*node),GFP_KERNEL ); + node =kmem_cache_zalloc (node_cache,GFP_KERNEL );if (unlikely (!node)) {list_del (&newnode->peer_list); -kfree (newnode); +kmem_cache_free (node_cache, newnode);return -ENOMEM ; }INIT_LIST_HEAD (&node->peer_list);
@@ -306,7 +313,7 @@ void wg_allowedips_remove_by_peer(structif (child) child->parent_bit = node->parent_bit; *rcu_dereference_protected (node->parent_bit,lockdep_is_held (lock)) = child; -kfree_rcu (node, rcu); +call_rcu (&node->rcu, node_free_rcu);/* TODO: Note that we currently don't walk up and down in order to * free any potential filler nodes. This means that this function
@@ -350,4 +357,16 @@ struct wg_peer *wg_allowedips_lookup_srcreturn NULL ; } +int __init wg_allowedips_slab_init (void ) +{ + node_cache =KMEM_CACHE (allowedips_node,0 ); +return node_cache ?0 : -ENOMEM ; +} + +void wg_allowedips_slab_uninit (void ) +{ +rcu_barrier (); +kmem_cache_destroy (node_cache); +} +#include "selftest/allowedips.c"
drivers/net/wireguard/allowedips.h +4 -1
@@ -19,7 +19,7 @@ struct allowedips_node {u8 bits[16 ]__aligned(__alignof(u64)) ;/* Keep rarely used members at bottom to be beyond cache line. */ -struct allowedips_node *__rcu *parent_bit;/* XXX: this puts us at 68->128 bytes instead of 60->64 bytes!! */ +struct allowedips_node *__rcu *parent_bit;union {struct list_head peer_list;struct rcu_head rcu;
@@ -53,4 +53,7 @@ struct wg_peer *wg_allowedips_lookup_srcbool wg_allowedips_selftest (void );#endif +int wg_allowedips_slab_init (void ); +void wg_allowedips_slab_uninit (void ); +#endif /* _WG_ALLOWEDIPS_H */
drivers/net/wireguard/main.c +9 -1
@@ -21,10 +21,15 @@ static int __init mod_init(void) {int ret; + ret =wg_allowedips_slab_init (); +if (ret <0 ) + goto err_allowedips; +#ifdef DEBUG + ret = -ENOTRECOVERABLE ;if (!wg_allowedips_selftest () || !wg_packet_counter_selftest () || !wg_ratelimiter_selftest ()) -return -ENOTRECOVERABLE ; + goto err_peer;#endif wg_noise_init ();
@@ -50,6 +55,8 @@ err_netlink: err_device:wg_peer_uninit (); err_peer: +wg_allowedips_slab_uninit (); +err_allowedips:return ret; }
@@ -58,6 +65,7 @@ static void __exit mod_exit(void)wg_genetlink_uninit ();wg_device_uninit ();wg_peer_uninit (); +wg_allowedips_slab_uninit (); }module_init (mod_init);
Greg Kroah-Hartman
maintainerPATCH 102/161 wireguard: allowedips: free empty intermediate nodes when removing single node
In reply to Greg Kroah-HartmanFrom: Jason A. Donenfeld <[email protected]>
commit bf7b042dc62a31f66d3a41dd4dfc7806f267b307 upstream.
When removing single nodes, it's possible that that node's parent is an empty intermediate node, in which case, it too should be removed. Otherwise the trie fills up and never is fully emptied, leading to gradual memory leaks over time for tries that are modified often. There was originally code to do this, but was removed during refactoring in 2016 and never reworked. Now that we have proper parent pointers from the previous commits, we can implement this properly.
In order to reduce branching and expensive comparisons, we want to keep the double pointer for parent assignment (which lets us easily chain up to the root), but we still need to actually get the parent's base address. So encode the bit number into the last two bits of the pointer, and pack and unpack it as needed. This is a little bit clumsy but is the fastest and less memory wasteful of the compromises. Note that we align the root struct here to a minimum of 4, because it's embedded into a larger struct, and we're relying on having the bottom two bits for our flag, which would only be 16-bit aligned on m68k.
The existing macro-based helpers were a bit unwieldy for adding the bit packing to, so this commit replaces them with safer and clearer ordinary functions.
We add a test to the randomized/fuzzer part of the selftests, to free the randomized tries by-peer, refuzz it, and repeat, until it's supposed to be empty, and then then see if that actually resulted in the whole thing being emptied. That combined with kmemcheck should hopefully make sure this commit is doing what it should. Along the way this resulted in various other cleanups of the tests and fixes for recent graphviz.
--- drivers/net/wireguard/allowedips.c | 102 ++++++++++------- drivers/net/wireguard/allowedips.h | 4 drivers/net/wireguard/selftest/allowedips.c | 162 +++++++++++++--------------- 3 files changed, 137 insertions(+), 131 deletions(-)
drivers/net/wireguard/allowedips.c +58 -44
@@ -30,8 +30,11 @@ static void copy_and_assign_cidr(struct node->bitlen = bits;memcpy (node->bits, src, bits /8U ); } -#define CHOOSE_NODE (parent, key) \ - parent->bit[(key[parent->bit_at_a] >> parent->bit_at_b) & 1] + +static inline u8 choose (struct allowedips_node *node,const u8 *key) +{ +return (key[node->bit_at_a] >> node->bit_at_b) &1 ; +}static void push_rcu (struct allowedips_node **stack,struct allowedips_node __rcu *p,unsigned int *len)
@@ -112,7 +115,7 @@ static struct allowedips_node *find_node found = node;if (node->cidr == bits)break ; - node =rcu_dereference_bh (CHOOSE_NODE (node, key)); + node =rcu_dereference_bh (node->bit[choose (node, key)]); }return found; }
@@ -144,8 +147,7 @@ static bool node_placement(struct allowe u8 cidr, u8 bits, struct allowedips_node **rnode, struct mutex *lock) { - struct allowedips_node *node = rcu_dereference_protected(trie, -lockdep_is_held (lock)); + struct allowedips_node *node = rcu_dereference_protected(trie, lockdep_is_held(lock));struct allowedips_node *parent =NULL ;bool exact = false;
@@ -155,13 +157,24 @@ static bool node_placement(struct allowe exact = true;break ; } - node =rcu_dereference_protected (CHOOSE_NODE (parent, key), -lockdep_is_held (lock)); + node =rcu_dereference_protected (parent->bit[choose (parent, key)],lockdep_is_held (lock)); } *rnode = parent;return exact; } +static inline void connect_node (struct allowedips_node **parent,u8 bit,struct allowedips_node *node) +{ + node->parent_bit_packed = (unsigned long )parent | bit; +rcu_assign_pointer (*parent, node); +} + +static inline void choose_and_connect_node (struct allowedips_node *parent,struct allowedips_node *node) +{ +u8 bit =choose (parent, node->bits); +connect_node (&parent->bit[bit], bit, node); +} +static int add (struct allowedips_node __rcu **trie,u8 bits,const u8 *key,u8 cidr,struct wg_peer *peer,struct mutex *lock) {
@@ -177,8 +190,7 @@ static int add(struct allowedips_node __RCU_INIT_POINTER (node->peer, peer);list_add_tail (&node->peer_list, &peer->allowedips_list);copy_and_assign_cidr (node, key, cidr, bits); -rcu_assign_pointer (node->parent_bit, trie); -rcu_assign_pointer (*trie, node); +connect_node (trie,2 , node);return 0 ; }if (node_placement (*trie, key, cidr, bits, &node, lock)) {
@@ -197,10 +209,10 @@ static int add(struct allowedips_node __if (!node) { down =rcu_dereference_protected (*trie,lockdep_is_held (lock)); }else { - down =rcu_dereference_protected (CHOOSE_NODE (node, key),lockdep_is_held (lock)); +const u8 bit =choose (node, key); + down =rcu_dereference_protected (node->bit[bit],lockdep_is_held (lock));if (!down) { -rcu_assign_pointer (newnode->parent_bit, &CHOOSE_NODE (node, key)); -rcu_assign_pointer (CHOOSE_NODE (node, key), newnode); +connect_node (&node->bit[bit], bit, newnode);return 0 ; } }
@@ -208,15 +220,11 @@ static int add(struct allowedips_node __ parent = node;if (newnode->cidr == cidr) { -rcu_assign_pointer (down->parent_bit, &CHOOSE_NODE (newnode, down->bits)); -rcu_assign_pointer (CHOOSE_NODE (newnode, down->bits), down); -if (!parent) { -rcu_assign_pointer (newnode->parent_bit, trie); -rcu_assign_pointer (*trie, newnode); - }else { -rcu_assign_pointer (newnode->parent_bit, &CHOOSE_NODE (parent, newnode->bits)); -rcu_assign_pointer (CHOOSE_NODE (parent, newnode->bits), newnode); - } +choose_and_connect_node (newnode, down); +if (!parent) +connect_node (trie,2 , newnode); +else +choose_and_connect_node (parent, newnode);return 0 ; }
@@ -229,17 +237,12 @@ static int add(struct allowedips_node __INIT_LIST_HEAD (&node->peer_list);copy_and_assign_cidr (node, newnode->bits, cidr, bits); -rcu_assign_pointer (down->parent_bit, &CHOOSE_NODE (node, down->bits)); -rcu_assign_pointer (CHOOSE_NODE (node, down->bits), down); -rcu_assign_pointer (newnode->parent_bit, &CHOOSE_NODE (node, newnode->bits)); -rcu_assign_pointer (CHOOSE_NODE (node, newnode->bits), newnode); -if (!parent) { -rcu_assign_pointer (node->parent_bit, trie); -rcu_assign_pointer (*trie, node); - }else { -rcu_assign_pointer (node->parent_bit, &CHOOSE_NODE (parent, node->bits)); -rcu_assign_pointer (CHOOSE_NODE (parent, node->bits), node); - } +choose_and_connect_node (node, down); +choose_and_connect_node (node, newnode); +if (!parent) +connect_node (trie,2 , node); +else +choose_and_connect_node (parent, node);return 0 ; }
@@ -297,7 +300,8 @@ int wg_allowedips_insert_v6(struct allowvoid wg_allowedips_remove_by_peer (struct allowedips *table,struct wg_peer *peer,struct mutex *lock) { -struct allowedips_node *node, *child, *tmp; +struct allowedips_node *node, *child, **parent_bit, *parent, *tmp; +bool free_parent;if (list_empty (&peer->allowedips_list))return ;
@@ -307,19 +311,29 @@ void wg_allowedips_remove_by_peer(structRCU_INIT_POINTER (node->peer,NULL );if (node->bit[0 ] && node->bit[1 ])continue ; - child =rcu_dereference_protected ( - node->bit[!rcu_access_pointer (node->bit[0 ])], -lockdep_is_held (lock)); + child =rcu_dereference_protected (node->bit[!rcu_access_pointer (node->bit[0 ])], +lockdep_is_held (lock));if (child) - child->parent_bit = node->parent_bit; - *rcu_dereference_protected (node->parent_bit,lockdep_is_held (lock)) = child; + child->parent_bit_packed = node->parent_bit_packed; + parent_bit = (struct allowedips_node **)(node->parent_bit_packed & ~3UL ); + *parent_bit = child; + parent = (void *)parent_bit - + offsetof(struct allowedips_node, bit[node->parent_bit_packed & 1]); + free_parent = !rcu_access_pointer (node->bit[0 ]) && + !rcu_access_pointer (node->bit[1 ]) && + (node->parent_bit_packed &3 ) <=1 && + !rcu_access_pointer (parent->peer); +if (free_parent) + child =rcu_dereference_protected ( + parent->bit[!(node->parent_bit_packed &1 )], +lockdep_is_held (lock));call_rcu (&node->rcu, node_free_rcu); - -/* TODO: Note that we currently don't walk up and down in order to -* free any potential filler nodes. This means that this function -* doesn't free up as much as it could, which could be revisited -* at some point. -*/ +if (!free_parent) +continue ; +if (child) + child->parent_bit_packed = parent->parent_bit_packed; + *(struct allowedips_node **)(parent->parent_bit_packed & ~3UL ) = child; +call_rcu (&parent->rcu, node_free_rcu); } }
drivers/net/wireguard/allowedips.h +2 -2
@@ -19,7 +19,7 @@ struct allowedips_node {u8 bits[16 ]__aligned(__alignof(u64)) ;/* Keep rarely used members at bottom to be beyond cache line. */ -struct allowedips_node *__rcu *parent_bit; +unsigned long parent_bit_packed;union {struct list_head peer_list;struct rcu_head rcu;
@@ -30,7 +30,7 @@ struct allowedips {struct allowedips_node __rcu *root4;struct allowedips_node __rcu *root6;u64 seq; -}; +}__aligned(4) ;/* We pack the lower 2 bits of &root, but m68k only gives 16-bit alignment. */ void wg_allowedips_init (struct allowedips *table);void wg_allowedips_free (struct allowedips *table,struct mutex *mutex);
drivers/net/wireguard/selftest/allowedips.c +77 -85
261 lines not on this page · raw message
Greg Kroah-Hartman
maintainerPATCH 103/161 net: caif: added cfserl_release function
In reply to Greg Kroah-HartmanFrom: Pavel Skripkin <[email protected]>
commit bce130e7f392ddde8cfcb09927808ebd5f9c8669 upstream.
Added cfserl_release() function.
--- include/net/caif/cfserl.h | 1 + net/caif/cfserl.c | 5 +++++ 2 files changed, 6 insertions(+)
include/net/caif/cfserl.h +1 -0
@@ -9,4 +9,5 @@#include <net/caif/caif_layer.h> struct cflayer *cfserl_create (int instance,bool use_stx); +void cfserl_release (struct cflayer *layer);#endif
net/caif/cfserl.c +5 -0
@@ -31,6 +31,11 @@ static int cfserl_transmit(struct cflayestatic void cfserl_ctrlcmd (struct cflayer *layr,enum caif_ctrlcmd ctrl,int phyid); +void cfserl_release (struct cflayer *layer) +{ +kfree (layer); +} +struct cflayer *cfserl_create (int instance,bool use_stx) {struct cfserl *this =kzalloc (sizeof (struct cfserl ),GFP_ATOMIC );
Greg Kroah-Hartman
maintainerPATCH 104/161 net: caif: add proper error handling
In reply to Greg Kroah-HartmanFrom: Pavel Skripkin <[email protected]>
commit a2805dca5107d5603f4bbc027e81e20d93476e96 upstream.
caif_enroll_dev() can fail in some cases. Ingnoring these cases can lead to memory leak due to not assigning link_support pointer to anywhere.
--- include/net/caif/caif_dev.h | 2 +- include/net/caif/cfcnfg.h | 2 +- net/caif/caif_dev.c | 8 +++++--- net/caif/cfcnfg.c | 16 +++++++++++----- 4 files changed, 18 insertions(+), 10 deletions(-)
include/net/caif/caif_dev.h +1 -1
@@ -119,7 +119,7 @@ void caif_free_client(struct cflayer *ad* The link_support layer is used to add any Link Layer specific * framing. */ -void caif_enroll_dev(struct net_device *dev, struct caif_dev_common *caifdev, +int caif_enroll_dev(struct net_device *dev, struct caif_dev_common *caifdev, struct cflayer *link_support, int head_room, struct cflayer **layer, int (**rcv_func)( struct sk_buff *, struct net_device *,
include/net/caif/cfcnfg.h +1 -1
@@ -62,7 +62,7 @@ void cfcnfg_remove(struct cfcnfg *cfg);* @fcs: Specify if checksum is used in CAIF Framing Layer. * @head_room: Head space needed by link specific protocol. */ -void +int cfcnfg_add_phy_layer(struct cfcnfg *cnfg, struct net_device *dev, struct cflayer *phy_layer, enum cfcnfg_phy_preference pref,
net/caif/caif_dev.c +5 -3
@@ -308,7 +308,7 @@ static void dev_flowctrl(struct net_devicaifd_put (caifd); } -void caif_enroll_dev(struct net_device *dev, struct caif_dev_common *caifdev, +int caif_enroll_dev(struct net_device *dev, struct caif_dev_common *caifdev, struct cflayer *link_support, int head_room, struct cflayer **layer, int (**rcv_func)(struct sk_buff *, struct net_device *,
@@ -319,11 +319,12 @@ void caif_enroll_dev(struct net_device *enum cfcnfg_phy_preference pref;struct cfcnfg *cfg =get_cfcnfg (dev_net (dev));struct caif_device_entry_list *caifdevs; +int res; caifdevs =caif_device_list (dev_net (dev)); caifd =caif_device_alloc (dev);if (!caifd) -return ; +return -ENOMEM ; *layer = &caifd->layer;spin_lock_init (&caifd->flow_lock);
@@ -344,7 +345,7 @@ void caif_enroll_dev(struct net_device *strlcpy (caifd->layer.name, dev->name,sizeof (caifd->layer.name)); caifd->layer.transmit = transmit; - cfcnfg_add_phy_layer(cfg, + res = cfcnfg_add_phy_layer(cfg, dev, &caifd->layer, pref,
@@ -354,6 +355,7 @@ void caif_enroll_dev(struct net_device *mutex_unlock (&caifdevs->lock);if (rcv_func) *rcv_func = receive; +return res; }EXPORT_SYMBOL (caif_enroll_dev);
net/caif/cfcnfg.c +11 -5
@@ -450,7 +450,7 @@ unlock:rcu_read_unlock (); } -void +int cfcnfg_add_phy_layer(struct cfcnfg *cnfg, struct net_device *dev, struct cflayer *phy_layer, enum cfcnfg_phy_preference pref,
@@ -459,7 +459,7 @@ cfcnfg_add_phy_layer(struct cfcnfg *cnfg {struct cflayer *frml;struct cfcnfg_phyinfo *phyinfo =NULL ; -int i; +int i, res =0 ;u8 phyid;mutex_lock (&cnfg->lock);
@@ -473,12 +473,15 @@ cfcnfg_add_phy_layer(struct cfcnfg *cnfg goto got_phyid; }pr_warn ("Too many CAIF Link Layers (max 6)\n" ); + res = -EEXIST ; goto out; got_phyid: phyinfo =kzalloc (sizeof (struct cfcnfg_phyinfo ),GFP_ATOMIC ); -if (!phyinfo) +if (!phyinfo) { + res = -ENOMEM ; goto out_err; + } phy_layer->id = phyid; phyinfo->pref = pref;
@@ -492,8 +495,10 @@ got_phyid: frml =cffrml_create (phyid, fcs); -if (!frml) +if (!frml) { + res = -ENOMEM ; goto out_err; + } phyinfo->frm_layer = frml;layer_set_up (frml, cnfg->mux);
@@ -511,11 +516,12 @@ got_phyid:list_add_rcu (&phyinfo->node, &cnfg->phys); out:mutex_unlock (&cnfg->lock); -return ; +return res; out_err:kfree (phyinfo);mutex_unlock (&cnfg->lock); +return res; }EXPORT_SYMBOL (cfcnfg_add_phy_layer);
Greg Kroah-Hartman
maintainerPATCH 105/161 net: caif: fix memory leak in caif_device_notify
In reply to Greg Kroah-HartmanFrom: Pavel Skripkin <[email protected]>
commit b53558a950a89824938e9811eddfc8efcd94e1bb upstream.
In case of caif_enroll_dev() fail, allocated link_support won't be assigned to the corresponding structure. So simply free allocated pointer in case of error
--- net/caif/caif_dev.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
net/caif/caif_dev.c +4 -1
@@ -370,6 +370,7 @@ static int caif_device_notify(struct notstruct cflayer *layer, *link_support;int head_room =0 ;struct caif_device_entry_list *caifdevs; +int res; cfg =get_cfcnfg (dev_net (dev)); caifdevs =caif_device_list (dev_net (dev));
@@ -395,8 +396,10 @@ static int caif_device_notify(struct notbreak ; } } -caif_enroll_dev (dev, caifdev, link_support, head_room, + res =caif_enroll_dev (dev, caifdev, link_support, head_room, &layer,NULL ); +if (res) +cfserl_release (link_support); caifdev->flowctrl = dev_flowctrl;break ;
Greg Kroah-Hartman
maintainerPATCH 106/161 net: caif: fix memory leak in cfusbl_device_notify
In reply to Greg Kroah-HartmanFrom: Pavel Skripkin <[email protected]>
commit 7f5d86669fa4d485523ddb1d212e0a2d90bd62bb upstream.
In case of caif_enroll_dev() fail, allocated link_support won't be assigned to the corresponding structure. So simply free allocated pointer in case of error.
--- net/caif/caif_usb.c | 14 +++++++++++++- 1 file changed, 13 insertions(+), 1 deletion(-)
net/caif/caif_usb.c +13 -1
@@ -115,6 +115,11 @@ static struct cflayer *cfusbl_create(intreturn (struct cflayer *) this; } +static void cfusbl_release (struct cflayer *layer) +{ +kfree (layer); +} +static struct packet_type caif_usb_type__read_mostly = { .type =cpu_to_be16 (ETH_P_802_EX1 ), };
@@ -127,6 +132,7 @@ static int cfusbl_device_notify(struct nstruct cflayer *layer, *link_support;struct usbnet *usbnet;struct usb_device *usbdev; +int res;/* Check whether we have a NCM device, and find its VID/PID. */ if (!(dev->dev.parent && dev->dev.parent->driver &&
@@ -169,8 +175,11 @@ static int cfusbl_device_notify(struct nif (dev->num_tx_queues >1 )pr_warn ("USB device uses more than one tx queue\n" ); -caif_enroll_dev (dev, &common, link_support,CFUSB_MAX_HEADLEN , + res =caif_enroll_dev (dev, &common, link_support,CFUSB_MAX_HEADLEN , &layer, &caif_usb_type.func); +if (res) + goto err; +if (!pack_added)dev_add_pack (&caif_usb_type); pack_added = true;
@@ -178,6 +187,9 @@ static int cfusbl_device_notify(struct nstrlcpy (layer->name, dev->name,sizeof (layer->name));return 0 ; +err: +cfusbl_release (link_support); +return res; }static struct notifier_block caif_device_notifier = {
Greg Kroah-Hartman
maintainerPATCH 107/161 HID: i2c-hid: Skip ELAN power-on command after reset
In reply to Greg Kroah-HartmanFrom: Johnny Chuang <[email protected]>
commit ca66a6770bd9d6d99e469debd1c7363ac455daf9 upstream.
For ELAN touchscreen, we found our boot code of IC was not flexible enough to receive and handle this command.
Once the FW main code of our controller is crashed for some reason, the controller could not be enumerated successfully to be recognized by the system host. therefore, it lost touch functionality.
Add quirk for skip send power-on command after reset.
It will impact to ELAN touchscreen and touchpad on HID over I2C projects.
--- drivers/hid/i2c-hid/i2c-hid-core.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-)
drivers/hid/i2c-hid/i2c-hid-core.c +8 -1
@@ -45,6 +45,7 @@#define I2C_HID_QUIRK_BOGUS_IRQ BIT(4)#define I2C_HID_QUIRK_RESET_ON_RESUME BIT(5)#define I2C_HID_QUIRK_BAD_INPUT_SIZE BIT(6) +#define I2C_HID_QUIRK_NO_WAKEUP_AFTER_RESET BIT(7)/* flags */
@@ -178,6 +179,11 @@ static const struct i2c_hid_quirks { I2C_HID_QUIRK_RESET_ON_RESUME }, {USB_VENDOR_ID_ITE ,I2C_DEVICE_ID_ITE_LENOVO_LEGION_Y720 , I2C_HID_QUIRK_BAD_INPUT_SIZE }, +/* +* Sending the wakeup after reset actually break ELAN touchscreen controller +*/ + { USB_VENDOR_ID_ELAN, HID_ANY_ID, + I2C_HID_QUIRK_NO_WAKEUP_AFTER_RESET }, { 0, 0 } };
@@ -461,7 +467,8 @@ static int i2c_hid_hwreset(struct i2c_cl }/* At least some SIS devices need this after reset */ - ret =i2c_hid_set_power (client,I2C_HID_PWR_ON ); +if (!(ihid->quirks &I2C_HID_QUIRK_NO_WAKEUP_AFTER_RESET )) + ret =i2c_hid_set_power (client,I2C_HID_PWR_ON ); out_unlock:mutex_unlock (&ihid->reset_lock);
Greg Kroah-Hartman
maintainerPATCH 108/161 HID: magicmouse: fix NULL-deref on disconnect
In reply to Greg Kroah-HartmanFrom: Johan Hovold <[email protected]>
commit 4b4f6cecca446abcb686c6e6c451d4f1ec1a7497 upstream.
Commit 9d7b18668956 ("HID: magicmouse: add support for Apple Magic Trackpad 2") added a sanity check for an Apple trackpad but returned success instead of -ENODEV when the check failed. This means that the remove callback will dereference the never-initialised driver data pointer when the driver is later unbound (e.g. on USB disconnect).
--- drivers/hid/hid-magicmouse.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/hid/hid-magicmouse.c +1 -1
@@ -597,7 +597,7 @@ static int magicmouse_probe(struct hid_dif (id->vendor ==USB_VENDOR_ID_APPLE && id->product ==USB_DEVICE_ID_APPLE_MAGICTRACKPAD2 && hdev->type !=HID_TYPE_USBMOUSE ) -return 0 ; +return -ENODEV ; msc =devm_kzalloc (&hdev->dev,sizeof (*msc),GFP_KERNEL );if (msc ==NULL ) {
Greg Kroah-Hartman
maintainerPATCH 109/161 HID: multitouch: require Finger field to mark Win8 reports as MT
In reply to Greg Kroah-HartmanFrom: Ahelenia Ziemiańska <[email protected]>
commit a2353e3b26012ff43bcdf81d37a3eaddd7ecdbf3 upstream.
This effectively changes collection_is_mt from
contact ID in report->field
to
(device is Win8 => collection is finger) && contact ID in report->field
Some devices erroneously report Pen for fingers, and Win8 stylus-on-touchscreen devices report contact ID, but mark the accompanying touchscreen device's collection correctly
--- drivers/hid/hid-multitouch.c | 10 +++++++--- 1 file changed, 7 insertions(+), 3 deletions(-)
drivers/hid/hid-multitouch.c +7 -3
@@ -604,9 +604,13 @@ static struct mt_report_data *mt_allocatif (!(HID_MAIN_ITEM_VARIABLE & field->flags))continue ; -for (n =0 ; n < field->report_count; n++) { -if (field->usage[n].hid ==HID_DG_CONTACTID ) - rdata->is_mt_collection = true; +if (field->logical ==HID_DG_FINGER || td->hdev->group !=HID_GROUP_MULTITOUCH_WIN_8 ) { +for (n =0 ; n < field->report_count; n++) { +if (field->usage[n].hid ==HID_DG_CONTACTID ) { + rdata->is_mt_collection = true; +break ; + } + } } }
Greg Kroah-Hartman
maintainerPATCH 110/161 gfs2: fix scheduling while atomic bug in glocks
In reply to Greg Kroah-HartmanFrom: Bob Peterson <[email protected]>
commit 20265d9a67e40eafd39a8884658ca2e36f05985d upstream.
Before this patch, in the unlikely event that gfs2_glock_dq encountered a withdraw, it would do a wait_on_bit to wait for its journal to be recovered, but it never released the glock's spin_lock, which caused a scheduling-while-atomic error.
This patch unlocks the lockref spin_lock before waiting for recovery.
--- fs/gfs2/glock.c | 2 ++ 1 file changed, 2 insertions(+)
fs/gfs2/glock.c +2 -0
@@ -1465,9 +1465,11 @@ void gfs2_glock_dq(struct gfs2_holder *g glock_blocked_by_withdraw(gl) && gh->gh_gl != sdp->sd_jinode_gl) { sdp->sd_glock_dqs_held++; +spin_unlock (&gl->gl_lockref.lock);might_sleep ();wait_on_bit (&sdp->sd_flags,SDF_WITHDRAW_RECOVERY ,TASK_UNINTERRUPTIBLE ); +spin_lock (&gl->gl_lockref.lock); }if (gh->gh_flags &GL_NOCACHE )handle_callback (gl,LM_ST_UNLOCKED ,0 , false);
Greg Kroah-Hartman
maintainerPATCH 111/161 ALSA: timer: Fix master timer notification
In reply to Greg Kroah-HartmanFrom: Takashi Iwai <[email protected]>
commit 9c1fe96bded935369f8340c2ac2e9e189f697d5d upstream.
snd_timer_notify1() calls the notification to each slave for a master event, but it passes a wrong event number. It should be +10 offset, corresponding to SNDRV_TIMER_EVENT_MXXX, but it's incorrectly with +100 offset. Casually this was spotted by UBSAN check via syzkaller.
--- sound/core/timer.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
sound/core/timer.c +2 -1
@@ -520,9 +520,10 @@ static void snd_timer_notify1(struct sndreturn ;if (timer->hw.flags &SNDRV_TIMER_HW_SLAVE )return ; + event +=10 ;/* convert to SNDRV_TIMER_EVENT_MXXX */ list_for_each_entry (ts, &ti->slave_active_head, active_list)if (ts->ccallback) - ts->ccallback (ts, event +100 , &tstamp, resolution); + ts->ccallback (ts, event, &tstamp, resolution); }/* start/continue a master timer */
Greg Kroah-Hartman
maintainerPATCH 112/161 ALSA: hda: Fix for mute key LED for HP Pavilion 15-CK0xx
In reply to Greg Kroah-HartmanFrom: Carlos M <[email protected]>
commit 901be145a46eb79879367d853194346a549e623d upstream.
For the HP Pavilion 15-CK0xx, with audio subsystem ID 0x103c:0x841c, adding a line in patch_realtek.c to apply the ALC269_FIXUP_HP_MUTE_LED_MIC3 fix activates the mute key LED.
--- sound/pci/hda/patch_realtek.c | 1 + 1 file changed, 1 insertion(+)
sound/pci/hda/patch_realtek.c +1 -0
@@ -8289,6 +8289,7 @@ static const struct snd_pci_quirk alc269SND_PCI_QUIRK (0x103c ,0x82bf ,"HP G3 mini" ,ALC221_FIXUP_HP_MIC_NO_PRESENCE ),SND_PCI_QUIRK (0x103c ,0x82c0 ,"HP G3 mini premium" ,ALC221_FIXUP_HP_MIC_NO_PRESENCE ),SND_PCI_QUIRK (0x103c ,0x83b9 ,"HP Spectre x360" ,ALC269_FIXUP_HP_MUTE_LED_MIC3 ), +SND_PCI_QUIRK (0x103c ,0x841c ,"HP Pavilion 15-CK0xx" ,ALC269_FIXUP_HP_MUTE_LED_MIC3 ),SND_PCI_QUIRK (0x103c ,0x8497 ,"HP Envy x360" ,ALC269_FIXUP_HP_MUTE_LED_MIC3 ),SND_PCI_QUIRK (0x103c ,0x84da ,"HP OMEN dc0019-ur" ,ALC295_FIXUP_HP_OMEN ), SND_PCI_QUIRK(0x103c, 0x84e7, "HP Pavilion 15", ALC269_FIXUP_HP_MUTE_LED_MIC3),
Greg Kroah-Hartman
maintainerPATCH 113/161 ALSA: hda: update the power_state during the direct-complete
In reply to Greg Kroah-HartmanFrom: Hui Wang <[email protected]>
commit b8b90c17602689eeaa5b219d104bbc215d1225cc upstream.
The patch_realtek.c needs to check if the power_state.event equals PM_EVENT_SUSPEND, after using the direct-complete, the suspend() and resume() will be skipped if the codec is already rt_suspended, in this case, the patch_realtek.c will always get PM_EVENT_ON even the system is really resumed from S3.
We could set power_state to PMSG_SUSPEND in the prepare(), if other PM functions are called before complete(), those functions will override power_state; if no other PM functions are called before complete(), we could know the suspend() and resume() are skipped since only S3 pm functions could be skipped by direct-complete, in this case set power_state to PMSG_RESUME in the complete(). This could guarantee the first time of calling hda_codec_runtime_resume() after complete() has the correct power_state.
--- sound/pci/hda/hda_codec.c | 5 +++++ 1 file changed, 5 insertions(+)
sound/pci/hda/hda_codec.c +5 -0
@@ -2973,6 +2973,7 @@ static int hda_codec_runtime_resume(stru#ifdef CONFIG_PM_SLEEP static int hda_codec_pm_prepare (struct device *dev) { + dev->power.power_state =PMSG_SUSPEND ;return pm_runtime_suspended (dev); }
@@ -2980,6 +2981,10 @@ static void hda_codec_pm_complete(struct {struct hda_codec *codec =dev_to_hda_codec (dev); +/* If no other pm-functions are called between prepare() and complete() */ +if (dev->power.power_state.event ==PM_EVENT_SUSPEND ) + dev->power.power_state =PMSG_RESUME ; +if (pm_runtime_suspended (dev) && (codec->jackpoll_interval ||hda_codec_need_resume (codec) || codec->forced_resume))pm_request_resume (dev);
Greg Kroah-Hartman
maintainerPATCH 114/161 ARM: dts: imx6dl-yapp4: Fix RGMII connection to QCA8334 switch
In reply to Greg Kroah-HartmanFrom: Michal Vokáč <[email protected]>
commit 0e4a4a08cd78efcaddbc2e4c5ed86b5a5cb8a15e upstream.
The FEC does not have a PHY so it should not have a phy-handle. It is connected to the switch at RGMII level so we need a fixed-link sub-node on both ends.
This was not a problem until the qca8k.c driver was converted to PHYLINK by commit b3591c2a3661 ("net: dsa: qca8k: Switch to PHYLINK instead of PHYLIB"). That commit revealed the FEC configuration was not correct.
--- arch/arm/boot/dts/imx6dl-yapp4-common.dtsi | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
arch/arm/boot/dts/imx6dl-yapp4-common.dtsi +5 -1
@@ -105,9 +105,13 @@ phy-reset-gpios = <&gpio1 25 GPIO_ACTIVE_LOW>; phy-reset-duration = <20>; phy-supply = <&sw2_reg>; - phy-handle = <ðphy0>; status = "okay"; + fixed-link { + speed = <1000>; + full-duplex; + }; + mdio { #address-cells = <1>; #size-cells = <0>;
Greg Kroah-Hartman
maintainerPATCH 115/161 ARM: dts: imx6q-dhcom: Add PU,VDD1P1,VDD2P5 regulators
In reply to Greg Kroah-HartmanFrom: Marek Vasut <[email protected]>
commit 8967b27a6c1c19251989c7ab33c058d16e4a5f53 upstream.
Per schematic, both PU and SOC regulator are supplied from LTC3676 SW1 via VDDSOC_IN rail, add the PU input. Both VDD1P1, VDD2P5 are supplied from LTC3676 SW2 via VDDHIGH_IN rail, add both inputs.
While no instability or problems are currently observed, the regulators should be fully described in DT and that description should fully match the hardware, else this might lead to unforseen issues later. Fix this.
--- arch/arm/boot/dts/imx6q-dhcom-som.dtsi | 12 ++++++++++++ 1 file changed, 12 insertions(+)
arch/arm/boot/dts/imx6q-dhcom-som.dtsi +12 -0
@@ -406,6 +406,18 @@ vin-supply = <&sw1_reg>; }; +®_pu { + vin-supply = <&sw1_reg>; +}; + +®_vdd1p1 { + vin-supply = <&sw2_reg>; +}; + +®_vdd2p5 { + vin-supply = <&sw2_reg>; +}; + &uart1 { pinctrl-names = "default"; pinctrl-0 = <&pinctrl_uart1>;
Greg Kroah-Hartman
maintainerPATCH 116/161 ext4: fix memory leak in ext4_fill_super
In reply to Greg Kroah-HartmanFrom: Alexey Makhalov <[email protected]>
commit afd09b617db3786b6ef3dc43e28fe728cfea84df upstream.
Buffer head references must be released before calling kill_bdev(); otherwise the buffer head (and its page referenced by b_data) will not be freed by kill_bdev, and subsequently that bh will be leaked.
If blocksizes differ, sb_set_blocksize() will kill current buffers and page cache by using kill_bdev(). And then super block will be reread again but using correct blocksize this time. sb_set_blocksize() didn't fully free superblock page and buffer head, and being busy, they were not freed and instead leaked.
This can easily be reproduced by calling an infinite loop of:
systemctl start <ext4_on_lvm>.mount, and
systemctl stop <ext4_on_lvm>.mount
... since systemd creates a cgroup for each slice which it mounts, and the bh leak get amplified by a dying memory cgroup that also never gets freed, and memory consumption is much more easily noticed.
--- fs/ext4/super.c | 11 +++++++++-- 1 file changed, 9 insertions(+), 2 deletions(-)
fs/ext4/super.c +9 -2
@@ -4449,14 +4449,20 @@ static int ext4_fill_super(struct super_ }if (sb->s_blocksize != blocksize) { +/* +* bh must be released before kill_bdev(), otherwise +* it won't be freed and its page also. kill_bdev() +* is called by sb_set_blocksize(). +*/ +brelse (bh);/* Validate the filesystem blocksize */ if (!sb_set_blocksize (sb, blocksize)) {ext4_msg (sb,KERN_ERR ,"bad block size %d" , blocksize); + bh =NULL ; goto failed_mount; } -brelse (bh); logical_sb_block = sb_block *EXT4_MIN_BLOCK_SIZE ; offset =do_div (logical_sb_block, blocksize); bh =ext4_sb_bread_unmovable (sb, logical_sb_block);
@@ -5176,8 +5182,9 @@ failed_mount:kfree (get_qf_name (sb, sbi, i));#endif fscrypt_free_dummy_policy (&sbi->s_dummy_enc_policy); -ext4_blkdev_remove (sbi); +/* ext4_blkdev_remove() calls kill_bdev(), release bh before it. */ brelse (bh); +ext4_blkdev_remove (sbi); out_fail: sb->s_fs_info =NULL ;kfree (sbi->s_blockgroup_lock);
Greg Kroah-Hartman
maintainerPATCH 117/161 ext4: fix bug on in ext4_es_cache_extent as ext4_split_extent_at failed
In reply to Greg Kroah-HartmanFrom: Ye Bin <[email protected]>
commit 082cd4ec240b8734a82a89ffb890216ac98fec68 upstream.
We got follow bug_on when run fsstress with injecting IO fault:
[130747.323114] kernel BUG at fs/ext4/extents_status.c:762! [130747.323117] Internal error: Oops - BUG: 0 [#1] SMP ...... [130747.334329] Call trace: [130747.334553] ext4_es_cache_extent+0x150/0x168 [ext4] [130747.334975] ext4_cache_extents+0x64/0xe8 [ext4] [130747.335368] ext4_find_extent+0x300/0x330 [ext4] [130747.335759] ext4_ext_map_blocks+0x74/0x1178 [ext4] [130747.336179] ext4_map_blocks+0x2f4/0x5f0 [ext4] [130747.336567] ext4_mpage_readpages+0x4a8/0x7a8 [ext4] [130747.336995] ext4_readpage+0x54/0x100 [ext4] [130747.337359] generic_file_buffered_read+0x410/0xae8 [130747.337767] generic_file_read_iter+0x114/0x190 [130747.338152] ext4_file_read_iter+0x5c/0x140 [ext4] [130747.338556] __vfs_read+0x11c/0x188 [130747.338851] vfs_read+0x94/0x150 [130747.339110] ksys_read+0x74/0xf0
This patch's modification is according to Jan Kara's suggestion in:
https://patchwork.ozlabs.org/project/linux-ext4/patch/[email protected]/
"I see. Now I understand your patch. Honestly, seeing how fragile is trying to fix extent tree after split has failed in the middle, I would probably go even further and make sure we fix the tree properly in case of ENOSPC and EDQUOT (those are easily user triggerable). Anything else indicates a HW problem or fs corruption so I'd rather leave the extent tree as is and don't try to fix it (which also means we will not create overlapping extents)."
--- fs/ext4/extents.c | 43 +++++++++++++++++++++++-------------------- 1 file changed, 23 insertions(+), 20 deletions(-)
fs/ext4/extents.c +23 -20
@@ -3206,7 +3206,10 @@ static int ext4_split_extent_at(handle_text4_ext_mark_unwritten (ex2); err =ext4_ext_insert_extent (handle, inode, ppath, &newex, flags); -if (err == -ENOSPC && (EXT4_EXT_MAY_ZEROOUT & split_flag)) { +if (err != -ENOSPC && err != -EDQUOT ) + goto out; + +if (EXT4_EXT_MAY_ZEROOUT & split_flag) {if (split_flag & (EXT4_EXT_DATA_VALID1 |EXT4_EXT_DATA_VALID2 )) {if (split_flag &EXT4_EXT_DATA_VALID1 ) { err =ext4_ext_zeroout (inode, ex2);
@@ -3232,25 +3235,22 @@ static int ext4_split_extent_at(handle_t ext4_ext_pblock(&orig_ex)); } -if (err) - goto fix_extent_len; -/* update the extent length and mark as initialized */ - ex->ee_len =cpu_to_le16 (ee_len); -ext4_ext_try_to_merge (handle, inode, path, ex); - err =ext4_ext_dirty (handle, inode, path + path->p_depth); -if (err) - goto fix_extent_len; - -/* update extent status tree */ - err =ext4_zeroout_es (inode, &zero_ex); - - goto out; - }else if (err) - goto fix_extent_len; - -out: -ext4_ext_show_leaf (inode, path); -return err; +if (!err) { +/* update the extent length and mark as initialized */ + ex->ee_len =cpu_to_le16 (ee_len); +ext4_ext_try_to_merge (handle, inode, path, ex); + err =ext4_ext_dirty (handle, inode, path + path->p_depth); +if (!err) +/* update extent status tree */ + err =ext4_zeroout_es (inode, &zero_ex); +/* If we failed at this point, we don't know in which +* state the extent tree exactly is so don't try to fix +* length of the original extent as it may do even more +* damage. +*/ + goto out; + } + } fix_extent_len: ex->ee_len = orig_ex.ee_len;
@@ -3260,6 +3260,9 @@ fix_extent_len:*/ ext4_ext_dirty (handle, inode, path + path->p_depth);return err; +out: +ext4_ext_show_leaf (inode, path); +return err; }/*
Greg Kroah-Hartman
maintainerPATCH 118/161 ext4: fix fast commit alignment issues
In reply to Greg Kroah-HartmanFrom: Harshad Shirwadkar <[email protected]>
commit a7ba36bc94f20b6c77f16364b9a23f582ea8faac upstream.
Fast commit recovery data on disk may not be aligned. So, when the recovery code reads it, this patch makes sure that fast commit info found on-disk is first memcpy-ed into an aligned variable before accessing it. As a consequence of it, we also remove some macros that could resulted in unaligned accesses.
--- fs/ext4/fast_commit.c | 170 ++++++++++++++++++++++++++------------------------ fs/ext4/fast_commit.h | 19 ----- 2 files changed, 90 insertions(+), 99 deletions(-)
fs/ext4/fast_commit.c +90 -80
@@ -1288,28 +1288,29 @@ struct dentry_info_args { };static inline void tl_to_darg (struct dentry_info_args *darg, -struct ext4_fc_tl *tl) +struct ext4_fc_tl *tl,u8 *val) { -struct ext4_fc_dentry_info *fcd; +struct ext4_fc_dentry_info fcd; - fcd = (struct ext4_fc_dentry_info *)ext4_fc_tag_val (tl); +memcpy (&fcd, val,sizeof (fcd)); - darg->parent_ino =le32_to_cpu (fcd->fc_parent_ino); - darg->ino =le32_to_cpu (fcd->fc_ino); - darg->dname = fcd->fc_dname; - darg->dname_len =ext4_fc_tag_len (tl) - -sizeof (struct ext4_fc_dentry_info ); + darg->parent_ino =le32_to_cpu (fcd.fc_parent_ino); + darg->ino =le32_to_cpu (fcd.fc_ino); + darg->dname = val + offsetof(struct ext4_fc_dentry_info , fc_dname); + darg->dname_len =le16_to_cpu (tl->fc_len) - +sizeof (struct ext4_fc_dentry_info ); }/* Unlink replay function */ -static int ext4_fc_replay_unlink (struct super_block *sb,struct ext4_fc_tl *tl) +static int ext4_fc_replay_unlink (struct super_block *sb,struct ext4_fc_tl *tl, +u8 *val) {struct inode *inode, *old_parent;struct qstr entry;struct dentry_info_args darg;int ret =0 ; -tl_to_darg (&darg, tl); +tl_to_darg (&darg, tl, val);trace_ext4_fc_replay (sb,EXT4_FC_TAG_UNLINK , darg.ino, darg.parent_ino, darg.dname_len);
@@ -1399,13 +1400,14 @@ out: }/* Link replay function */ -static int ext4_fc_replay_link (struct super_block *sb,struct ext4_fc_tl *tl) +static int ext4_fc_replay_link (struct super_block *sb,struct ext4_fc_tl *tl, +u8 *val) {struct inode *inode;struct dentry_info_args darg;int ret =0 ; -tl_to_darg (&darg, tl); +tl_to_darg (&darg, tl, val);trace_ext4_fc_replay (sb,EXT4_FC_TAG_LINK , darg.ino, darg.parent_ino, darg.dname_len);
@@ -1450,9 +1452,10 @@ static int ext4_fc_record_modified_inode/* * Inode replay function */ -static int ext4_fc_replay_inode (struct super_block *sb,struct ext4_fc_tl *tl) +static int ext4_fc_replay_inode (struct super_block *sb,struct ext4_fc_tl *tl, +u8 *val) { -struct ext4_fc_inode *fc_inode; +struct ext4_fc_inode fc_inode;struct ext4_inode *raw_inode;struct ext4_inode *raw_fc_inode;struct inode *inode =NULL ;
@@ -1460,9 +1463,9 @@ static int ext4_fc_replay_inode(struct sint inode_len, ino, ret, tag =le16_to_cpu (tl->fc_tag);struct ext4_extent_header *eh; - fc_inode = (struct ext4_fc_inode *)ext4_fc_tag_val (tl); +memcpy (&fc_inode, val,sizeof (fc_inode)); - ino =le32_to_cpu (fc_inode->fc_ino); + ino =le32_to_cpu (fc_inode.fc_ino);trace_ext4_fc_replay (sb, tag, ino,0 ,0 ); inode =ext4_iget (sb, ino,EXT4_IGET_NORMAL );
@@ -1474,12 +1477,13 @@ static int ext4_fc_replay_inode(struct sext4_fc_record_modified_inode (sb, ino); - raw_fc_inode = (struct ext4_inode *)fc_inode->fc_raw_inode; + raw_fc_inode = (struct ext4_inode *) + (val + offsetof(struct ext4_fc_inode , fc_raw_inode)); ret =ext4_get_fc_inode_loc (sb, ino, &iloc);if (ret) goto out; - inode_len =ext4_fc_tag_len (tl) -sizeof (struct ext4_fc_inode ); + inode_len =le16_to_cpu (tl->fc_len) -sizeof (struct ext4_fc_inode ); raw_inode =ext4_raw_inode (&iloc);memcpy (raw_inode, raw_fc_inode, offsetof(struct ext4_inode , i_block));
@@ -1547,14 +1551,15 @@ out:* inode for which we are trying to create a dentry here, should already have * been replayed before we start here. */ -static int ext4_fc_replay_create (struct super_block *sb,struct ext4_fc_tl *tl) +static int ext4_fc_replay_create (struct super_block *sb,struct ext4_fc_tl *tl, +u8 *val) {int ret =0 ;struct inode *inode =NULL ;struct inode *dir =NULL ;struct dentry_info_args darg; -tl_to_darg (&darg, tl); +tl_to_darg (&darg, tl, val);trace_ext4_fc_replay (sb,EXT4_FC_TAG_CREAT , darg.ino, darg.parent_ino, darg.dname_len);
@@ -1633,9 +1638,9 @@ static int ext4_fc_record_regions(struct/* Replay add range tag */ static int ext4_fc_replay_add_range (struct super_block *sb, -struct ext4_fc_tl *tl) +struct ext4_fc_tl *tl,u8 *val) { -struct ext4_fc_add_range *fc_add_ex; +struct ext4_fc_add_range fc_add_ex;struct ext4_extent newex, *ex;struct inode *inode;ext4_lblk_t start, cur;
@@ -1645,15 +1650,14 @@ static int ext4_fc_replay_add_range(strustruct ext4_ext_path *path =NULL ;int ret; - fc_add_ex = (struct ext4_fc_add_range *)ext4_fc_tag_val (tl); - ex = (struct ext4_extent *)&fc_add_ex->fc_ex; +memcpy (&fc_add_ex, val,sizeof (fc_add_ex)); + ex = (struct ext4_extent *)&fc_add_ex.fc_ex;trace_ext4_fc_replay (sb,EXT4_FC_TAG_ADD_RANGE , -le32_to_cpu (fc_add_ex->fc_ino),le32_to_cpu (ex->ee_block), +le32_to_cpu (fc_add_ex.fc_ino),le32_to_cpu (ex->ee_block),ext4_ext_get_actual_len (ex)); - inode =ext4_iget (sb,le32_to_cpu (fc_add_ex->fc_ino), -EXT4_IGET_NORMAL ); + inode =ext4_iget (sb,le32_to_cpu (fc_add_ex.fc_ino),EXT4_IGET_NORMAL );if (IS_ERR (inode)) {jbd_debug (1 ,"Inode not found." );return 0 ;
@@ -1762,32 +1766,33 @@ next:/* Replay DEL_RANGE tag */ static int -ext4_fc_replay_del_range (struct super_block *sb,struct ext4_fc_tl *tl) +ext4_fc_replay_del_range (struct super_block *sb,struct ext4_fc_tl *tl, +u8 *val) {struct inode *inode; -struct ext4_fc_del_range *lrange; +struct ext4_fc_del_range lrange;struct ext4_map_blocks map;ext4_lblk_t cur, remaining;int ret; - lrange = (struct ext4_fc_del_range *)ext4_fc_tag_val (tl); - cur =le32_to_cpu (lrange->fc_lblk); - remaining =le32_to_cpu (lrange->fc_len); +memcpy (&lrange, val,sizeof (lrange)); + cur =le32_to_cpu (lrange.fc_lblk); + remaining =le32_to_cpu (lrange.fc_len);trace_ext4_fc_replay (sb,EXT4_FC_TAG_DEL_RANGE , -le32_to_cpu (lrange->fc_ino), cur, remaining); +le32_to_cpu (lrange.fc_ino), cur, remaining); - inode =ext4_iget (sb,le32_to_cpu (lrange->fc_ino),EXT4_IGET_NORMAL ); + inode =ext4_iget (sb,le32_to_cpu (lrange.fc_ino),EXT4_IGET_NORMAL );if (IS_ERR (inode)) { -jbd_debug (1 ,"Inode %d not found" ,le32_to_cpu (lrange->fc_ino)); +jbd_debug (1 ,"Inode %d not found" ,le32_to_cpu (lrange.fc_ino));return 0 ; } ret =ext4_fc_record_modified_inode (sb, inode->i_ino);jbd_debug (1 ,"DEL_RANGE, inode %ld, lblk %d, len %d\n" , - inode->i_ino,le32_to_cpu (lrange->fc_lblk), -le32_to_cpu (lrange->fc_len)); + inode->i_ino,le32_to_cpu (lrange.fc_lblk), +le32_to_cpu (lrange.fc_len));while (remaining >0 ) { map.m_lblk = cur; map.m_len = remaining;
@@ -1808,8 +1813,8 @@ ext4_fc_replay_del_range(struct super_bl } ret =ext4_punch_hole (inode, -le32_to_cpu (lrange->fc_lblk) << sb->s_blocksize_bits, -le32_to_cpu (lrange->fc_len) << sb->s_blocksize_bits); +le32_to_cpu (lrange.fc_lblk) << sb->s_blocksize_bits, +le32_to_cpu (lrange.fc_len) << sb->s_blocksize_bits);if (ret)jbd_debug (1 ,"ext4_punch_hole returned %d" , ret); ext4_ext_replay_shrink_inode(inode,
@@ -1925,11 +1930,11 @@ static int ext4_fc_replay_scan(journal_tstruct ext4_sb_info *sbi =EXT4_SB (sb);struct ext4_fc_replay_state *state;int ret =JBD2_FC_REPLAY_CONTINUE ; -struct ext4_fc_add_range *ext; -struct ext4_fc_tl *tl; -struct ext4_fc_tail *tail; -__u8 *start, *end; -struct ext4_fc_head *head; +struct ext4_fc_add_range ext; +struct ext4_fc_tl tl; +struct ext4_fc_tail tail; +__u8 *start, *end, *cur, *val; +struct ext4_fc_head head;struct ext4_extent *ex; state = &sbi->s_fc_replay_state;
@@ -1956,15 +1961,17 @@ static int ext4_fc_replay_scan(journal_t } state->fc_replay_expected_off++; -fc_for_each_tl (start, end, tl) { +for (cur = start; cur < end; cur = cur +sizeof (tl) +le16_to_cpu (tl.fc_len)) { +memcpy (&tl, cur,sizeof (tl)); + val = cur +sizeof (tl);jbd_debug (3 ,"Scan phase, tag:%s, blk %lld\n" , -tag2str (le16_to_cpu (tl->fc_tag)), bh->b_blocknr); -switch (le16_to_cpu (tl->fc_tag)) { +tag2str (le16_to_cpu (tl.fc_tag)), bh->b_blocknr); +switch (le16_to_cpu (tl.fc_tag)) {case EXT4_FC_TAG_ADD_RANGE : - ext = (struct ext4_fc_add_range *)ext4_fc_tag_val (tl); - ex = (struct ext4_extent *)&ext->fc_ex; +memcpy (&ext, val,sizeof (ext)); + ex = (struct ext4_extent *)&ext.fc_ex; ret =ext4_fc_record_regions (sb, -le32_to_cpu (ext->fc_ino), +le32_to_cpu (ext.fc_ino),le32_to_cpu (ex->ee_block),ext4_ext_pblock (ex),ext4_ext_get_actual_len (ex));if (ret <0 )
@@ -1978,18 +1985,18 @@ static int ext4_fc_replay_scan(journal_tcase EXT4_FC_TAG_INODE :case EXT4_FC_TAG_PAD : state->fc_cur_tag++; - state->fc_crc =ext4_chksum (sbi, state->fc_crc, tl, -sizeof (*tl) +ext4_fc_tag_len (tl)); + state->fc_crc =ext4_chksum (sbi, state->fc_crc, cur, +sizeof (tl) +le16_to_cpu (tl.fc_len));break ;case EXT4_FC_TAG_TAIL : state->fc_cur_tag++; - tail = (struct ext4_fc_tail *)ext4_fc_tag_val (tl); - state->fc_crc =ext4_chksum (sbi, state->fc_crc, tl, -sizeof (*tl) + +memcpy (&tail, val,sizeof (tail)); + state->fc_crc =ext4_chksum (sbi, state->fc_crc, cur, +sizeof (tl) + offsetof(struct ext4_fc_tail , fc_crc)); -if (le32_to_cpu (tail->fc_tid) == expected_tid && -le32_to_cpu (tail->fc_crc) == state->fc_crc) { +if (le32_to_cpu (tail.fc_tid) == expected_tid && +le32_to_cpu (tail.fc_crc) == state->fc_crc) { state->fc_replay_num_tags = state->fc_cur_tag; state->fc_regions_valid = state->fc_regions_used;
@@ -2000,19 +2007,19 @@ static int ext4_fc_replay_scan(journal_t state->fc_crc =0 ;break ;case EXT4_FC_TAG_HEAD : - head = (struct ext4_fc_head *)ext4_fc_tag_val (tl); -if (le32_to_cpu (head->fc_features) & +memcpy (&head, val,sizeof (head)); +if (le32_to_cpu (head.fc_features) & ~EXT4_FC_SUPPORTED_FEATURES ) { ret = -EOPNOTSUPP ;break ; } -if (le32_to_cpu (head->fc_tid) != expected_tid) { +if (le32_to_cpu (head.fc_tid) != expected_tid) { ret =JBD2_FC_REPLAY_STOP ;break ; } state->fc_cur_tag++; - state->fc_crc =ext4_chksum (sbi, state->fc_crc, tl, -sizeof (*tl) +ext4_fc_tag_len (tl)); + state->fc_crc =ext4_chksum (sbi, state->fc_crc, cur, +sizeof (tl) +le16_to_cpu (tl.fc_len));break ;default : ret = state->fc_replay_num_tags ?
@@ -2036,11 +2043,11 @@ static int ext4_fc_replay(journal_t *jou {struct super_block *sb = journal->j_private;struct ext4_sb_info *sbi =EXT4_SB (sb); -struct ext4_fc_tl *tl; -__u8 *start, *end; +struct ext4_fc_tl tl; +__u8 *start, *end, *cur, *val;int ret =JBD2_FC_REPLAY_CONTINUE ;struct ext4_fc_replay_state *state = &sbi->s_fc_replay_state; -struct ext4_fc_tail *tail; +struct ext4_fc_tail tail;if (pass ==PASS_SCAN ) { state->fc_current_pass =PASS_SCAN ;
@@ -2067,49 +2074,52 @@ static int ext4_fc_replay(journal_t *jou start = (u8 *)bh->b_data; end = (__u8 *)bh->b_data + journal->j_blocksize -1 ; -fc_for_each_tl (start, end, tl) { +for (cur = start; cur < end; cur = cur +sizeof (tl) +le16_to_cpu (tl.fc_len)) { +memcpy (&tl, cur,sizeof (tl)); + val = cur +sizeof (tl); +if (state->fc_replay_num_tags ==0 ) { ret =JBD2_FC_REPLAY_STOP ;ext4_fc_set_bitmaps_and_counters (sb);break ; }jbd_debug (3 ,"Replay phase, tag:%s\n" , -tag2str (le16_to_cpu (tl->fc_tag))); +tag2str (le16_to_cpu (tl.fc_tag))); state->fc_replay_num_tags--; -switch (le16_to_cpu (tl->fc_tag)) { +switch (le16_to_cpu (tl.fc_tag)) {case EXT4_FC_TAG_LINK : - ret =ext4_fc_replay_link (sb, tl); + ret =ext4_fc_replay_link (sb, &tl, val);break ;case EXT4_FC_TAG_UNLINK : - ret =ext4_fc_replay_unlink (sb, tl); + ret =ext4_fc_replay_unlink (sb, &tl, val);break ;case EXT4_FC_TAG_ADD_RANGE : - ret =ext4_fc_replay_add_range (sb, tl); + ret =ext4_fc_replay_add_range (sb, &tl, val);break ;case EXT4_FC_TAG_CREAT : - ret =ext4_fc_replay_create (sb, tl); + ret =ext4_fc_replay_create (sb, &tl, val);break ;case EXT4_FC_TAG_DEL_RANGE : - ret =ext4_fc_replay_del_range (sb, tl); + ret =ext4_fc_replay_del_range (sb, &tl, val);break ;case EXT4_FC_TAG_INODE : - ret =ext4_fc_replay_inode (sb, tl); + ret =ext4_fc_replay_inode (sb, &tl, val);break ;case EXT4_FC_TAG_PAD :trace_ext4_fc_replay (sb,EXT4_FC_TAG_PAD ,0 , -ext4_fc_tag_len (tl),0 ); +le16_to_cpu (tl.fc_len),0 );break ;case EXT4_FC_TAG_TAIL :trace_ext4_fc_replay (sb,EXT4_FC_TAG_TAIL ,0 , -ext4_fc_tag_len (tl),0 ); - tail = (struct ext4_fc_tail *)ext4_fc_tag_val (tl); -WARN_ON (le32_to_cpu (tail->fc_tid) != expected_tid); +le16_to_cpu (tl.fc_len),0 ); +memcpy (&tail, val,sizeof (tail)); +WARN_ON (le32_to_cpu (tail.fc_tid) != expected_tid);break ;case EXT4_FC_TAG_HEAD :break ;default : -trace_ext4_fc_replay (sb,le16_to_cpu (tl->fc_tag),0 , -ext4_fc_tag_len (tl),0 ); +trace_ext4_fc_replay (sb,le16_to_cpu (tl.fc_tag),0 , +le16_to_cpu (tl.fc_len),0 ); ret = -ECANCELED ;break ; }
fs/ext4/fast_commit.h +0 -19
@@ -153,13 +153,6 @@ struct ext4_fc_replay_state {#define region_last (__region) (((__region)->lblk) + ((__region)->len) - 1)#endif -#define fc_for_each_tl (__start, __end, __tl) \ - for (tl = (struct ext4_fc_tl *)(__start); \ - (__u8 *)tl < (__u8 *)(__end); \ - tl = (struct ext4_fc_tl *)((__u8 *)tl + \ - sizeof(struct ext4_fc_tl) + \ - + le16_to_cpu(tl->fc_len))) -static inline const char *tag2str (__u16 tag) {switch (tag) {
@@ -186,16 +179,4 @@ static inline const char *tag2str(__u16 } } -/* Get length of a particular tlv */ -static inline int ext4_fc_tag_len (struct ext4_fc_tl *tl) -{ -return le16_to_cpu (tl->fc_len); -} - -/* Get a pointer to "value" of a tlv */ -static inline __u8 *ext4_fc_tag_val (struct ext4_fc_tl *tl) -{ -return (__u8 *)tl +sizeof (*tl); -} -#endif /* __FAST_COMMIT_H__ */
Greg Kroah-Hartman
maintainerPATCH 119/161 ext4: fix memory leak in ext4_mb_init_backend on error path.
In reply to Greg Kroah-HartmanFrom: Phillip Potter <[email protected]>
commit a8867f4e3809050571c98de7a2d465aff5e4daf5 upstream.
Fix a memory leak discovered by syzbot when a file system is corrupted with an illegally large s_log_groups_per_flex.
--- fs/ext4/mballoc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/ext4/mballoc.c +1 -1
@@ -2715,7 +2715,7 @@ static int ext4_mb_init_backend(struct s*/ if (sbi->s_es->s_log_groups_per_flex >=32 ) {ext4_msg (sb,KERN_ERR ,"too many log groups per flexible block group" ); - goto err_freesgi; + goto err_freebuddy; } sbi->s_mb_prefetch =min_t (uint,1 << sbi->s_es->s_log_groups_per_flex,BLK_MAX_SEGMENT_SIZE >> (sb->s_blocksize_bits -9 ));
Greg Kroah-Hartman
maintainerPATCH 120/161 ext4: fix accessing uninit percpu counter variable with fast_commit
In reply to Greg Kroah-HartmanFrom: Ritesh Harjani <[email protected]>
commit b45f189a19b38e01676628db79cd3eeb1333516e upstream.
When running generic/527 with fast_commit configuration, the following issue is seen on Power. With fast_commit, during ext4_fc_replay() (which can be called from ext4_fill_super()), if inode eviction happens then it can access an uninitialized percpu counter variable.
This patch adds the check before accessing the counters in ext4_free_inode() path.
[ 321.165371] run fstests generic/527 at 2021-04-29 08:38:43
[ 323.027786] EXT4-fs (dm-0): mounted filesystem with ordered data mode. Opts: block_validity. Quota mode: none.
[ 323.618772] BUG: Unable to handle kernel data access on read at 0x1fbd80000
[ 323.619767] Faulting instruction address: 0xc000000000bae78c
cpu 0x1: Vector: 300 (Data Access) at [c000000010706ef0]
pc: c000000000bae78c: percpu_counter_add_batch+0x3c/0x100
lr: c0000000006d0bb0: ext4_free_inode+0x780/0xb90
pid = 5593, comm = mount
ext4_free_inode+0x780/0xb90
ext4_evict_inode+0xa8c/0xc60
evict+0xfc/0x1e0
ext4_fc_replay+0xc50/0x20f0
do_one_pass+0xfe0/0x1350
jbd2_journal_recover+0x184/0x2e0
jbd2_journal_load+0x1c0/0x4a0
ext4_fill_super+0x2458/0x4200
mount_bdev+0x1dc/0x290
ext4_mount+0x28/0x40
legacy_get_tree+0x4c/0xa0
vfs_get_tree+0x4c/0x120
path_mount+0xcf8/0xd70
do_mount+0x80/0xd0
sys_mount+0x3fc/0x490
system_call_exception+0x384/0x3d0
system_call_common+0xec/0x278--- fs/ext4/ialloc.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
fs/ext4/ialloc.c +4 -2
@@ -322,14 +322,16 @@ void ext4_free_inode(handle_t *handle, sif (is_directory) { count =ext4_used_dirs_count (sb, gdp) -1 ;ext4_used_dirs_set (sb, gdp, count); -percpu_counter_dec (&sbi->s_dirs_counter); +if (percpu_counter_initialized (&sbi->s_dirs_counter)) +percpu_counter_dec (&sbi->s_dirs_counter); }ext4_inode_bitmap_csum_set (sb, block_group, gdp, bitmap_bh,EXT4_INODES_PER_GROUP (sb) /8 );ext4_group_desc_csum_set (sb, block_group, gdp);ext4_unlock_group (sb, block_group); -percpu_counter_inc (&sbi->s_freeinodes_counter); +if (percpu_counter_initialized (&sbi->s_freeinodes_counter)) +percpu_counter_inc (&sbi->s_freeinodes_counter);if (sbi->s_log_groups_per_flex) {struct flex_groups *fg;
Greg Kroah-Hartman
maintainerPATCH 121/161 usb: dwc2: Fix build in periphal-only mode
In reply to Greg Kroah-HartmanFrom: Phil Elwell <[email protected]>
In branches to which 24d209dba5a3 ("usb: dwc2: Fix hibernation between host and device modes.") has been back-ported, the bus_suspended member of struct dwc2_hsotg is only present in builds that support host-mode. To avoid having to pull in several more non-Fix commits in order to get it to compile, wrap the usage of the member in a macro conditional.
--- drivers/usb/dwc2/core_intr.c | 4 ++++ 1 file changed, 4 insertions(+)
drivers/usb/dwc2/core_intr.c +4 -0
@@ -707,7 +707,11 @@ static inline void dwc_handle_gpwrdn_disdwc2_writel (hsotg, gpwrdn_tmp,GPWRDN ); hsotg->hibernated =0 ; + +#if IS_ENABLED (CONFIG_USB_DWC2_HOST ) || \ +IS_ENABLED (CONFIG_USB_DWC2_DUAL_ROLE ) hsotg->bus_suspended =0 ; +#endif if (gpwrdn &GPWRDN_IDSTS ) { hsotg->op_state =OTG_STATE_B_PERIPHERAL ;
Greg Kroah-Hartman
maintainerPATCH 122/161 Revert "MIPS: make userspace mapping young by default"
In reply to Greg Kroah-HartmanFrom: Thomas Bogendoerfer <[email protected]>
commit 50c25ee97cf6ab011542167ab590c17012cea4ed upstream.
This reverts commit f685a533a7fab35c5d069dcd663f59c8e4171a75.
The MIPS cache flush logic needs to know whether the mapping was already established to decide how to flush caches. This is done by checking the valid bit in the PTE. The commit above breaks this logic by setting the valid in the PTE in new mappings, which causes kernel crashes.
--- arch/mips/mm/cache.c | 30 ++++++++++++++---------------- include/linux/pgtable.h | 8 ++++++++ mm/memory.c | 4 ++++ 3 files changed, 26 insertions(+), 16 deletions(-)
arch/mips/mm/cache.c +14 -16
@@ -157,31 +157,29 @@ unsigned long _page_cachable_default;EXPORT_SYMBOL (_page_cachable_default);#define PM (p) __pgprot(_page_cachable_default | (p)) -#define PVA (p) PM(_PAGE_VALID | _PAGE_ACCESSED | (p))static inline void setup_protection_map (void ) { protection_map[0 ] =PM (_PAGE_PRESENT | _PAGE_NO_EXEC | _PAGE_NO_READ); - protection_map[1 ] =PVA (_PAGE_PRESENT | _PAGE_NO_EXEC); - protection_map[2 ] =PVA (_PAGE_PRESENT | _PAGE_NO_EXEC | _PAGE_NO_READ); - protection_map[3 ] =PVA (_PAGE_PRESENT | _PAGE_NO_EXEC); - protection_map[4 ] =PVA (_PAGE_PRESENT); - protection_map[5 ] =PVA (_PAGE_PRESENT); - protection_map[6 ] =PVA (_PAGE_PRESENT); - protection_map[7 ] =PVA (_PAGE_PRESENT); + protection_map[1 ] =PM (_PAGE_PRESENT | _PAGE_NO_EXEC); + protection_map[2 ] =PM (_PAGE_PRESENT | _PAGE_NO_EXEC | _PAGE_NO_READ); + protection_map[3 ] =PM (_PAGE_PRESENT | _PAGE_NO_EXEC); + protection_map[4 ] =PM (_PAGE_PRESENT); + protection_map[5 ] =PM (_PAGE_PRESENT); + protection_map[6 ] =PM (_PAGE_PRESENT); + protection_map[7 ] =PM (_PAGE_PRESENT); protection_map[8 ] =PM (_PAGE_PRESENT | _PAGE_NO_EXEC | _PAGE_NO_READ); - protection_map[9 ] =PVA (_PAGE_PRESENT | _PAGE_NO_EXEC); - protection_map[10 ] =PVA (_PAGE_PRESENT | _PAGE_NO_EXEC | _PAGE_WRITE | + protection_map[9 ] =PM (_PAGE_PRESENT | _PAGE_NO_EXEC); + protection_map[10 ] =PM (_PAGE_PRESENT | _PAGE_NO_EXEC | _PAGE_WRITE | _PAGE_NO_READ); - protection_map[11 ] =PVA (_PAGE_PRESENT | _PAGE_NO_EXEC | _PAGE_WRITE); - protection_map[12 ] =PVA (_PAGE_PRESENT); - protection_map[13 ] =PVA (_PAGE_PRESENT); - protection_map[14 ] =PVA (_PAGE_PRESENT); - protection_map[15 ] =PVA (_PAGE_PRESENT); + protection_map[11 ] =PM (_PAGE_PRESENT | _PAGE_NO_EXEC | _PAGE_WRITE); + protection_map[12 ] =PM (_PAGE_PRESENT); + protection_map[13 ] =PM (_PAGE_PRESENT); + protection_map[14 ] =PM (_PAGE_PRESENT | _PAGE_WRITE); + protection_map[15 ] =PM (_PAGE_PRESENT | _PAGE_WRITE); } -#undef _PVA#undef PMvoid cpu_cache_init (void )
include/linux/pgtable.h +8 -0
@@ -432,6 +432,14 @@ static inline void ptep_set_wrprotect(st* To be differentiate with macro pte_mkyoung, this macro is used on platforms * where software maintains page access bit. */ +#ifndef pte_sw_mkyoung +static inline pte_t pte_sw_mkyoung (pte_t pte) +{ +return pte; +} +#define pte_sw_mkyoung pte_sw_mkyoung +#endif +#ifndef pte_savedwrite#define pte_savedwrite pte_write#endif
mm/memory.c +4 -0
@@ -2896,6 +2896,7 @@ static vm_fault_t wp_page_copy(struct vm }flush_cache_page (vma, vmf->address,pte_pfn (vmf->orig_pte)); entry =mk_pte (new_page, vma->vm_page_prot); + entry =pte_sw_mkyoung (entry); entry =maybe_mkwrite (pte_mkdirty (entry), vma);/*
@@ -3561,6 +3562,7 @@ static vm_fault_t do_anonymous_page(stru__SetPageUptodate (page); entry =mk_pte (page, vma->vm_page_prot); + entry =pte_sw_mkyoung (entry);if (vma->vm_flags &VM_WRITE ) entry =pte_mkwrite (pte_mkdirty (entry));
@@ -3745,6 +3747,8 @@ void do_set_pte(struct vm_fault *vmf, stif (prefault &&arch_wants_old_prefaulted_pte ()) entry =pte_mkold (entry); +else + entry =pte_sw_mkyoung (entry);if (write) entry =maybe_mkwrite (pte_mkdirty (entry), vma);
Greg Kroah-Hartman
maintainerPATCH 123/161 kfence: maximize allocation wait timeout duration
In reply to Greg Kroah-HartmanFrom: Marco Elver <[email protected]>
commit 37c9284f6932b915043717703d6496dfd59c85f5 upstream.
The allocation wait timeout was initially added because of warnings due to CONFIG_DETECT_HUNG_TASK=y [1]. While the 1 sec timeout is sufficient to resolve the warnings (given the hung task timeout must be 1 sec or larger) it may cause unnecessary wake-ups if the system is idle:
https://lkml.kernel.org/r/CADYN=9J0DQhizAGB0-jz4HOBBh+05kMBXb4c0cXMS7Qi5NAJiw@mail.gmail.com
Fix it by computing the timeout duration in terms of the current sysctl_hung_task_timeout_secs value.
--- mm/kfence/core.c | 12 +++++++++++- 1 file changed, 11 insertions(+), 1 deletion(-)
mm/kfence/core.c +11 -1
@@ -20,6 +20,7 @@#include <linux/moduleparam.h> #include <linux/random.h> #include <linux/rcupdate.h> +#include <linux/sched/sysctl.h> #include <linux/seq_file.h> #include <linux/slab.h> #include <linux/spinlock.h>
@@ -620,7 +621,16 @@ static void toggle_allocation_gate(struc/* Enable static key, and await allocation to happen. */ static_branch_enable (&kfence_allocation_key); -wait_event_timeout (allocation_wait,atomic_read (&kfence_allocation_gate),HZ ); +if (sysctl_hung_task_timeout_secs) { +/* +* During low activity with no allocations we might wait a +* while; let's avoid the hung task warning. +*/ +wait_event_timeout (allocation_wait,atomic_read (&kfence_allocation_gate), + sysctl_hung_task_timeout_secs *HZ /2 ); + }else { +wait_event (allocation_wait,atomic_read (&kfence_allocation_gate)); + }/* Disable static key and reset timer. */ static_branch_disable (&kfence_allocation_key);
Greg Kroah-Hartman
maintainerPATCH 124/161 kfence: use TASK_IDLE when awaiting allocation
In reply to Greg Kroah-HartmanFrom: Marco Elver <[email protected]>
commit 8fd0e995cc7b6a7a8a40bc03d52a2cd445beeff4 upstream.
Since wait_event() uses TASK_UNINTERRUPTIBLE by default, waiting for an allocation counts towards load. However, for KFENCE, this does not make any sense, since there is no busy work we're awaiting.
Instead, use TASK_IDLE via wait_event_idle() to not count towards load.
--- mm/kfence/core.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
mm/kfence/core.c +3 -3
@@ -626,10 +626,10 @@ static void toggle_allocation_gate(struc* During low activity with no allocations we might wait a * while; let's avoid the hung task warning. */ -wait_event_timeout (allocation_wait,atomic_read (&kfence_allocation_gate), - sysctl_hung_task_timeout_secs *HZ /2 ); +wait_event_idle_timeout (allocation_wait,atomic_read (&kfence_allocation_gate), + sysctl_hung_task_timeout_secs *HZ /2 ); }else { -wait_event (allocation_wait,atomic_read (&kfence_allocation_gate)); +wait_event_idle (allocation_wait,atomic_read (&kfence_allocation_gate)); }/* Disable static key and reset timer. */
Greg Kroah-Hartman
maintainerPATCH 125/161 pid: take a reference when initializing `cad_pid`
In reply to Greg Kroah-HartmanFrom: Mark Rutland <[email protected]>
commit 0711f0d7050b9e07c44bc159bbc64ac0a1022c7f upstream.
During boot, kernel_init_freeable() initializes cad_pid to the init task's struct pid. Later on, we may change cad_pid via a sysctl, and when this happens proc_do_cad_pid() will increment the refcount on the new pid via get_pid(), and will decrement the refcount on the old pid via put_pid(). As we never called get_pid() when we initialized cad_pid, we decrement a reference we never incremented, can therefore free the init task's struct pid early. As there can be dangling references to the struct pid, we can later encounter a use-after-free (e.g. when delivering signals).
This was spotted when fuzzing v5.13-rc3 with Syzkaller, but seems to have been around since the conversion of cad_pid to struct pid in commit 9ec52099e4b8 ("[PATCH] replace cad_pid by a struct pid") from the pre-KASAN stone age of v2.6.19.
Fix this by getting a reference to the init task's struct pid when we assign it to cad_pid.
Full KASAN splat below.
==================================================================
BUG: KASAN: use-after-free in ns_of_pid include/linux/pid.h:153 [inline]
BUG: KASAN: use-after-free in task_active_pid_ns+0xc0/0xc8 kernel/pid.c:509
Read of size 4 at addr ffff23794dda0004 by task syz-executor.0/273
CPU: 1 PID: 273 Comm: syz-executor.0 Not tainted 5.12.0-00001-g9aef892b2d15 #1
Hardware name: linux,dummy-virt (DT)
Call trace:
ns_of_pid include/linux/pid.h:153 [inline]
task_active_pid_ns+0xc0/0xc8 kernel/pid.c:509
do_notify_parent+0x308/0xe60 kernel/signal.c:1950
exit_notify kernel/exit.c:682 [inline]
do_exit+0x2334/0x2bd0 kernel/exit.c:845
do_group_exit+0x108/0x2c8 kernel/exit.c:922
get_signal+0x4e4/0x2a88 kernel/signal.c:2781
do_signal arch/arm64/kernel/signal.c:882 [inline]
do_notify_resume+0x300/0x970 arch/arm64/kernel/signal.c:936
work_pending+0xc/0x2dc
Allocated by task 0:
slab_post_alloc_hook+0x50/0x5c0 mm/slab.h:516
slab_alloc_node mm/slub.c:2907 [inline]
slab_alloc mm/slub.c:2915 [inline]
kmem_cache_alloc+0x1f4/0x4c0 mm/slub.c:2920
alloc_pid+0xdc/0xc00 kernel/pid.c:180
copy_process+0x2794/0x5e18 kernel/fork.c:2129
kernel_clone+0x194/0x13c8 kernel/fork.c:2500
kernel_thread+0xd4/0x110 kernel/fork.c:2552
rest_init+0x44/0x4a0 init/main.c:687
arch_call_rest_init+0x1c/0x28
start_kernel+0x520/0x554 init/main.c:1064
0x0
Freed by task 270:
slab_free_hook mm/slub.c:1562 [inline]
slab_free_freelist_hook+0x98/0x260 mm/slub.c:1600
slab_free mm/slub.c:3161 [inline]
kmem_cache_free+0x224/0x8e0 mm/slub.c:3177
put_pid.part.4+0xe0/0x1a8 kernel/pid.c:114
put_pid+0x30/0x48 kernel/pid.c:109
proc_do_cad_pid+0x190/0x1b0 kernel/sysctl.c:1401
proc_sys_call_handler+0x338/0x4b0 fs/proc/proc_sysctl.c:591
proc_sys_write+0x34/0x48 fs/proc/proc_sysctl.c:617
call_write_iter include/linux/fs.h:1977 [inline]
new_sync_write+0x3ac/0x510 fs/read_write.c:518
vfs_write fs/read_write.c:605 [inline]
vfs_write+0x9c4/0x1018 fs/read_write.c:585
ksys_write+0x124/0x240 fs/read_write.c:658
__do_sys_write fs/read_write.c:670 [inline]
__se_sys_write fs/read_write.c:667 [inline]
__arm64_sys_write+0x78/0xb0 fs/read_write.c:667
__invoke_syscall arch/arm64/kernel/syscall.c:37 [inline]
invoke_syscall arch/arm64/kernel/syscall.c:49 [inline]
el0_svc_common.constprop.1+0x16c/0x388 arch/arm64/kernel/syscall.c:129
do_el0_svc+0xf8/0x150 arch/arm64/kernel/syscall.c:168
el0_svc+0x28/0x38 arch/arm64/kernel/entry-common.c:416
el0_sync_handler+0x134/0x180 arch/arm64/kernel/entry-common.c:432
el0_sync+0x154/0x180 arch/arm64/kernel/entry.S:701
The buggy address belongs to the object at ffff23794dda0000
which belongs to the cache pid of size 224
The buggy address is located 4 bytes inside of
224-byte region [ffff23794dda0000, ffff23794dda00e0)
The buggy address belongs to the page:
page:(____ptrval____) refcount:1 mapcount:0 mapping:0000000000000000 index:0x0 pfn:0x4dda0
head:(____ptrval____) order:1 compound_mapcount:0
flags: 0x3fffc0000010200(slab|head)
raw: 03fffc0000010200 dead000000000100 dead000000000122 ffff23794d40d080
raw: 0000000000000000 0000000000190019 00000001ffffffff 0000000000000000
page dumped because: kasan: bad access detected
Memory state around the buggy address:
ffff23794dd9ff00: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
ffff23794dd9ff80: fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc fc
>ffff23794dda0000: fa fb fb fb fb fb fb fb fb fb fb fb fb fb fb fb
^
ffff23794dda0080: fb fb fb fb fb fb fb fb fb fb fb fb fc fc fc fc
ffff23794dda0100: fc fc fc fc fc fc fc fc 00 00 00 00 00 00 00 00
==================================================================--- init/main.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
init/main.c +1 -1
@@ -1514,7 +1514,7 @@ static noinline void __init kernel_init_*/ set_mems_allowed (node_states[N_MEMORY ]); - cad_pid =task_pid (current); + cad_pid =get_pid (task_pid (current));smp_prepare_cpus (setup_max_cpus);
Greg Kroah-Hartman
maintainerPATCH 126/161 ocfs2: fix data corruption by fallocate
In reply to Greg Kroah-HartmanFrom: Junxiao Bi <[email protected]>
commit 6bba4471f0cc1296fe3c2089b9e52442d3074b2e upstream.
When fallocate punches holes out of inode size, if original isize is in the middle of last cluster, then the part from isize to the end of the cluster will be zeroed with buffer write, at that time isize is not yet updated to match the new size, if writeback is kicked in, it will invoke ocfs2_writepage()->block_write_full_page() where the pages out of inode size will be dropped. That will cause file corruption. Fix this by zero out eof blocks when extending the inode size.
Running the following command with qemu-image 4.2.1 can get a corrupted coverted image file easily.
qemu-img convert -p -t none -T none -f qcow2 $qcow_image \
-O qcow2 -o compat=1.1 $qcow_image.convThe usage of fallocate in qemu is like this, it first punches holes out of inode size, then extend the inode size.
fallocate(11, FALLOC_FL_KEEP_SIZE|FALLOC_FL_PUNCH_HOLE, 2276196352, 65536) = 0
fallocate(11, 0, 2276196352, 65536) = 0v1: https://www.spinics.net/lists/linux-fsdevel/msg193999.html
v2: https://lore.kernel.org/linux-fsdevel/[email protected]/T/
--- fs/ocfs2/file.c | 55 ++++++++++++++++++++++++++++++++++++++++++++++++++----- 1 file changed, 50 insertions(+), 5 deletions(-)
fs/ocfs2/file.c +50 -5
@@ -1858,6 +1858,45 @@ out: }/* +* zero out partial blocks of one cluster. +* +* start: file offset where zero starts, will be made upper block aligned. +* len: it will be trimmed to the end of current cluster if "start + len" +* is bigger than it. +*/ +static int ocfs2_zeroout_partial_cluster (struct inode *inode, +u64 start,u64 len) +{ +int ret; +u64 start_block, end_block, nr_blocks; +u64 p_block, offset; +u32 cluster, p_cluster, nr_clusters; +struct super_block *sb = inode->i_sb; +u64 end =ocfs2_align_bytes_to_clusters (sb, start); + +if (start + len < end) + end = start + len; + + start_block =ocfs2_blocks_for_bytes (sb, start); + end_block =ocfs2_blocks_for_bytes (sb, end); + nr_blocks = end_block - start_block; +if (!nr_blocks) +return 0 ; + + cluster =ocfs2_bytes_to_clusters (sb, start); + ret =ocfs2_get_clusters (inode, cluster, &p_cluster, + &nr_clusters,NULL ); +if (ret) +return ret; +if (!p_cluster) +return 0 ; + + offset = start_block -ocfs2_clusters_to_blocks (sb, cluster); + p_block =ocfs2_clusters_to_blocks (sb, p_cluster) + offset; +return sb_issue_zeroout (sb, p_block, nr_blocks,GFP_NOFS ); +} + +/* * Parts of this function taken from xfs_change_file_space() */ static int __ocfs2_change_file_space(struct file *file, struct inode *inode,
@@ -1867,7 +1906,7 @@ static int __ocfs2_change_file_space(str {int ret;s64 llen; -loff_t size; +loff_t size, orig_isize;struct ocfs2_super *osb =OCFS2_SB (inode->i_sb);struct buffer_head *di_bh =NULL ;handle_t *handle;
@@ -1898,6 +1937,7 @@ static int __ocfs2_change_file_space(str goto out_inode_unlock; } + orig_isize =i_size_read (inode);switch (sr->l_whence) {case 0 :/*SEEK_SET*/ break ;
@@ -1905,7 +1945,7 @@ static int __ocfs2_change_file_space(str sr->l_start += f_pos;break ;case 2 :/*SEEK_END*/ - sr->l_start +=i_size_read (inode); + sr->l_start += orig_isize;break ;default : ret = -EINVAL ;
@@ -1959,6 +1999,14 @@ static int __ocfs2_change_file_space(strdefault : ret = -EINVAL ; } + +/* zeroout eof blocks in the cluster. */ +if (!ret && change_size && orig_isize < size) { + ret =ocfs2_zeroout_partial_cluster (inode, orig_isize, + size - orig_isize); +if (!ret) +i_size_write (inode, size); + }up_write (&OCFS2_I (inode)->ip_alloc_sem);if (ret) {mlog_errno (ret);
@@ -1975,9 +2023,6 @@ static int __ocfs2_change_file_space(str goto out_inode_unlock; } -if (change_size &&i_size_read (inode) < size) -i_size_write (inode, size); - inode->i_ctime = inode->i_mtime =current_time (inode); ret =ocfs2_mark_inode_dirty (handle, inode, di_bh);if (ret <0 )
Greg Kroah-Hartman
maintainerPATCH 127/161 mm/debug_vm_pgtable: fix alignment for pmd/pud_advanced_tests()
In reply to Greg Kroah-HartmanFrom: Gerald Schaefer <[email protected]>
commit 04f7ce3f07ce39b1a3ca03a56b238a53acc52cfd upstream.
In pmd/pud_advanced_tests(), the vaddr is aligned up to the next pmd/pud entry, and so it does not match the given pmdp/pudp and (aligned down) pfn any more.
For s390, this results in memory corruption, because the IDTE instruction used e.g. in xxx_get_and_clear() will take the vaddr for some calculations, in combination with the given pmdp. It will then end up with a wrong table origin, ending on ...ff8, and some of those wrongly set low-order bits will also select a wrong pagetable level for the index addition. IDTE could therefore invalidate (or 0x20) something outside of the page tables, depending on the wrongly picked index, which in turn depends on the random vaddr.
As result, we sometimes see "BUG task_struct (Not tainted): Padding overwritten" on s390, where one 0x5a padding value got overwritten with 0x7a.
Fix this by aligning down, similar to how the pmd/pud_aligned pfns are calculated.
--- mm/debug_vm_pgtable.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
mm/debug_vm_pgtable.c +2 -2
@@ -192,7 +192,7 @@ static void __init pmd_advanced_tests(stpr_debug ("Validating PMD advanced\n" );/* Align the address wrt HPAGE_PMD_SIZE */ - vaddr = (vaddr &HPAGE_PMD_MASK ) +HPAGE_PMD_SIZE ; + vaddr &=HPAGE_PMD_MASK ;pgtable_trans_huge_deposit (mm, pmdp, pgtable);
@@ -330,7 +330,7 @@ static void __init pud_advanced_tests(stpr_debug ("Validating PUD advanced\n" );/* Align the address wrt HPAGE_PUD_SIZE */ - vaddr = (vaddr &HPAGE_PUD_MASK ) +HPAGE_PUD_SIZE ; + vaddr &=HPAGE_PUD_MASK ;set_pud_at (mm, vaddr, pudp, pud);pudp_set_wrprotect (mm, vaddr, pudp);
Greg Kroah-Hartman
maintainerPATCH 128/161 mm/page_alloc: fix counting of free pages after take off from buddy
In reply to Greg Kroah-HartmanFrom: Ding Hui <[email protected]>
commit bac9c6fa1f929213bbd0ac9cdf21e8e2f0916828 upstream.
Recently we found that there is a lot MemFree left in /proc/meminfo after do a lot of pages soft offline, it's not quite correct.
Before Oscar's rework of soft offline for free pages [1], if we soft offline free pages, these pages are left in buddy with HWPoison flag, and NR_FREE_PAGES is not updated immediately. So the difference between NR_FREE_PAGES and real number of available free pages is also even big at the beginning.
However, with the workload running, when we catch HWPoison page in any alloc functions subsequently, we will remove it from buddy, meanwhile update the NR_FREE_PAGES and try again, so the NR_FREE_PAGES will get more and more closer to the real number of available free pages. (regardless of unpoison_memory())
Now, for offline free pages, after a successful call take_page_off_buddy(), the page is no longer belong to buddy allocator, and will not be used any more, but we missed accounting NR_FREE_PAGES in this situation, and there is no chance to be updated later.
Do update in take_page_off_buddy() like rmqueue() does, but avoid double counting if some one already set_migratetype_isolate() on the page.
[1]: commit 06be6ff3d2ec ("mm,hwpoison: rework soft offline for free pages")
--- mm/page_alloc.c | 2 ++ 1 file changed, 2 insertions(+)
mm/page_alloc.c +2 -0
@@ -8951,6 +8951,8 @@ bool take_page_off_buddy(struct page *padel_page_from_free_list (page_head, zone, page_order);break_down_buddy_pages (zone, page_head, page,0 , page_order, migratetype); +if (!is_migrate_isolate (migratetype)) +__mod_zone_freepage_state (zone,-1 , migratetype); ret = true;break ; }
Greg Kroah-Hartman
maintainerPATCH 129/161 scsi: lpfc: Fix failure to transmit ABTS on FC link
In reply to Greg Kroah-HartmanFrom: James Smart <[email protected]>
commit 696770e72f2b42b92ea0a4a98087fb2ba376417a upstream.
The abort_cmd_ia flag in an abort wqe describes whether an ABTS basic link service should be transmitted on the FC link or not. Code added in lpfc_sli4_issue_abort_iotag() set the abort_cmd_ia flag incorrectly, surpressing ABTS transmission.
A previous LPFC change to build an abort wqe inverted prior logic that determined whether an ABTS was to be issued on the FC link.
Revert this logic to its proper state.
--- drivers/scsi/lpfc/lpfc_sli.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-)
drivers/scsi/lpfc/lpfc_sli.c +1 -3
@@ -20591,10 +20591,8 @@ lpfc_sli4_issue_abort_iotag(struct lpfc_ abtswqe = &abtsiocb->wqe;memset (abtswqe,0 ,sizeof (*abtswqe)); -if (lpfc_is_link_up (phba)) +if (!lpfc_is_link_up (phba))bf_set (abort_cmd_ia, &abtswqe->abort_cmd,1 ); -else -bf_set (abort_cmd_ia, &abtswqe->abort_cmd,0 );bf_set (abort_cmd_criteria, &abtswqe->abort_cmd,T_XRI_TAG ); abtswqe->abort_cmd.rsrvd5 =0 ; abtswqe->abort_cmd.wqe_com.abort_tag = xritag;
Greg Kroah-Hartman
maintainerPATCH 130/161 x86/cpufeatures: Force disable X86_FEATURE_ENQCMD and remove update_pasid()
In reply to Greg Kroah-HartmanFrom: Thomas Gleixner <[email protected]>
commit 9bfecd05833918526cc7357d55e393393440c5fa upstream.
While digesting the XSAVE-related horrors which got introduced with the supervisor/user split, the recent addition of ENQCMD-related functionality got on the radar and turned out to be similarly broken.
update_pasid(), which is only required when X86_FEATURE_ENQCMD is available, is invoked from two places:
1) From switch_to() for the incoming task
2) Via a SMP function call from the IOMMU/SMV code
#1 is half-ways correct as it hacks around the brokenness of get_xsave_addr()
by enforcing the state to be 'present', but all the conditionals in that code are completely pointless for that.
Also the invocation is just useless overhead because at that point it's guaranteed that TIF_NEED_FPU_LOAD is set on the incoming task and all of this can be handled at return to user space.
#2 is broken beyond repair. The comment in the code claims that it is safe
to invoke this in an IPI, but that's just wishful thinking.
FPU state of a running task is protected by fregs_lock() which is nothing else than a local_bh_disable(). As BH-disabled regions run usually with interrupts enabled the IPI can hit a code section which modifies FPU state and there is absolutely no guarantee that any of the assumptions which are made for the IPI case is true.
Also the IPI is sent to all CPUs in mm_cpumask(mm), but the IPI is invoked with a NULL pointer argument, so it can hit a completely unrelated task and unconditionally force an update for nothing. Worse, it can hit a kernel thread which operates on a user space address space and set a random PASID for it.
The offending commit does not cleanly revert, but it's sufficient to force disable X86_FEATURE_ENQCMD and to remove the broken update_pasid() code to make this dysfunctional all over the place. Anything more complex would require more surgery and none of the related functions outside of the x86 core code are blatantly wrong, so removing those would be overkill.
As nothing enables the PASID bit in the IA32_XSS MSR yet, which is required to make this actually work, this cannot result in a regression except for related out of tree train-wrecks, but they are broken already today.
--- arch/x86/include/asm/disabled-features.h | 7 +-- arch/x86/include/asm/fpu/api.h | 6 --- arch/x86/include/asm/fpu/internal.h | 7 --- arch/x86/kernel/fpu/xstate.c | 57 ------------------------------- 4 files changed, 3 insertions(+), 74 deletions(-)
arch/x86/include/asm/disabled-features.h +2 -5
@@ -56,11 +56,8 @@# define DISABLE_PTI (1 << (X86_FEATURE_PTI & 31))#endif -#ifdef CONFIG_IOMMU_SUPPORT -# define DISABLE_ENQCMD 0 -#else -# define DISABLE_ENQCMD (1 << (X86_FEATURE_ENQCMD & 31)) -#endif +/* Force disable because it's broken beyond repair */ +#define DISABLE_ENQCMD (1 << (X86_FEATURE_ENQCMD & 31))#ifdef CONFIG_X86_SGX # define DISABLE_SGX 0
arch/x86/include/asm/fpu/api.h +1 -5
@@ -106,10 +106,6 @@ extern int cpu_has_xfeatures(u64 xfeatur*/ #define PASID_DISABLED 0 -#ifdef CONFIG_IOMMU_SUPPORT -/* Update current's PASID MSR/state by mm's PASID. */ -void update_pasid (void ); -#else static inline void update_pasid (void ) { } -#endif +#endif /* _ASM_X86_FPU_API_H */
arch/x86/include/asm/fpu/internal.h +0 -7
@@ -584,13 +584,6 @@ static inline void switch_fpu_finish(str pkru_val = pk->pkru; }__write_pkru (pkru_val); - -/* -* Expensive PASID MSR write will be avoided in update_pasid() because -* TIF_NEED_FPU_LOAD was set. And the PASID state won't be updated -* unless it's different from mm->pasid to reduce overhead. -*/ -update_pasid (); }#endif /* _ASM_X86_FPU_INTERNAL_H */
arch/x86/kernel/fpu/xstate.c +0 -57
60 lines not on this page · raw message
Greg Kroah-Hartman
maintainerPATCH 131/161 dmaengine: idxd: Use cpu_feature_enabled()
In reply to Greg Kroah-HartmanFrom: Borislav Petkov <[email protected]>
commit 74b2fc882d380d8fafc2a26f01d401c2a7beeadb upstream.
When testing x86 feature bits, use cpu_feature_enabled() so that build-disabled features can remain off, regardless of what CPUID says.
--- drivers/dma/idxd/init.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/dma/idxd/init.c +2 -2
@@ -675,12 +675,12 @@ static int __init idxd_init_module(void)* If the CPU does not support MOVDIR64B or ENQCMDS, there's no point in * enumerating the device. We can not utilize it. */ -if (!boot_cpu_has (X86_FEATURE_MOVDIR64B )) { +if (!cpu_feature_enabled (X86_FEATURE_MOVDIR64B )) {pr_warn ("idxd driver failed to load without MOVDIR64B.\n" );return -ENODEV ; } -if (!boot_cpu_has (X86_FEATURE_ENQCMD )) +if (!cpu_feature_enabled (X86_FEATURE_ENQCMD ))pr_warn ("Platform does not have ENQCMD(S) support.\n" );else support_enqcmd = true;
Greg Kroah-Hartman
maintainerPATCH 132/161 x86/sev: Check SME/SEV support in CPUID first
In reply to Greg Kroah-HartmanFrom: Pu Wen <[email protected]>
commit 009767dbf42ac0dbe3cf48c1ee224f6b778aa85a upstream.
The first two bits of the CPUID leaf 0x8000001F EAX indicate whether SEV or SME is supported, respectively. It's better to check whether SEV or SME is actually supported before accessing the MSR_AMD64_SEV to check whether SEV or SME is enabled.
This is both a bare-metal issue and a guest/VM issue. Since the first generation Hygon Dhyana CPU doesn't support the MSR_AMD64_SEV, reading that MSR results in a #GP - either directly from hardware in the bare-metal case or via the hypervisor (because the RDMSR is actually intercepted) in the guest/VM case, resulting in a failed boot. And since this is very early in the boot phase, rdmsrl_safe()/native_read_msr_safe() can't be used.
So check the CPUID bits first, before accessing the MSR.
[ tlendacky: Expand and improve commit message. ] [ bp: Massage commit message. ]
--- arch/x86/mm/mem_encrypt_identity.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-)
arch/x86/mm/mem_encrypt_identity.c +6 -5
@@ -504,10 +504,6 @@ void __init sme_enable(struct boot_param#define AMD_SME_BIT BIT(0)#define AMD_SEV_BIT BIT(1) -/* Check the SEV MSR whether SEV or SME is enabled */ - sev_status =__rdmsr (MSR_AMD64_SEV ); - feature_mask = (sev_status &MSR_AMD64_SEV_ENABLED ) ?AMD_SEV_BIT :AMD_SME_BIT ; -/* * Check for the SME/SEV feature: * CPUID Fn8000_001F[EAX]
@@ -519,11 +515,16 @@ void __init sme_enable(struct boot_param eax =0x8000001f ; ecx =0 ;native_cpuid (&eax, &ebx, &ecx, &edx); -if (!(eax & feature_mask)) +/* Check whether SEV or SME is supported */ +if (!(eax & (AMD_SEV_BIT |AMD_SME_BIT )))return ; me_mask =1UL << (ebx &0x3f ); +/* Check the SEV MSR whether SEV or SME is enabled */ + sev_status =__rdmsr (MSR_AMD64_SEV ); + feature_mask = (sev_status &MSR_AMD64_SEV_ENABLED ) ?AMD_SEV_BIT :AMD_SME_BIT ; +/* Check if memory encryption is enabled */ if (feature_mask ==AMD_SME_BIT ) {/*
Greg Kroah-Hartman
maintainerPATCH 133/161 KVM: PPC: Book3S HV: Save host FSCR in the P7/8 path
In reply to Greg Kroah-HartmanFrom: Nicholas Piggin <[email protected]>
commit 1438709e6328925ef496dafd467dbd0353137434 upstream.
Similar to commit 25edcc50d76c ("KVM: PPC: Book3S HV: Save and restore FSCR in the P9 path"), ensure the P7/8 path saves and restores the host FSCR. The logic explained in that patch actually applies there to the old path well: a context switch can be made before kvmppc_vcpu_run_hv restores the host FSCR and returns.
Now both the p9 and the p7/8 paths now save and restore their FSCR, it no longer needs to be restored at the end of kvmppc_vcpu_run_hv
--- arch/powerpc/kvm/book3s_hv.c | 1 - arch/powerpc/kvm/book3s_hv_rmhandlers.S | 7 +++++++ 2 files changed, 7 insertions(+), 1 deletion(-)
arch/powerpc/kvm/book3s_hv.c +0 -1
@@ -4418,7 +4418,6 @@ static int kvmppc_vcpu_run_hv(struct kvmmtspr (SPRN_EBBRR , ebb_regs[1 ]);mtspr (SPRN_BESCR , ebb_regs[2 ]);mtspr (SPRN_TAR , user_tar); -mtspr (SPRN_FSCR , current->thread.fscr); }mtspr (SPRN_VRSAVE , user_vrsave);
arch/powerpc/kvm/book3s_hv_rmhandlers.S +7 -0
@@ -59,6 +59,7 @@ END_FTR_SECTION_IFCLR(CPU_FTR_ARCH_300) #define STACK_SLOT_UAMOR (SFS-88) #define STACK_SLOT_DAWR1 (SFS-96) #define STACK_SLOT_DAWRX1 (SFS-104) +#define STACK_SLOT_FSCR (SFS-112) /* the following is used by the P9 short path */ #define STACK_SLOT_NVGPRS (SFS-152) /* 18 gprs */
@@ -686,6 +687,8 @@ BEGIN_FTR_SECTION std r6, STACK_SLOT_DAWR0(r1) std r7, STACK_SLOT_DAWRX0(r1) std r8, STACK_SLOT_IAMR(r1) + mfspr r5, SPRN_FSCR + std r5, STACK_SLOT_FSCR(r1) END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S) BEGIN_FTR_SECTION mfspr r6, SPRN_DAWR1
@@ -1663,6 +1666,10 @@ FTR_SECTION_ELSE ld r7, STACK_SLOT_HFSCR(r1) mtspr SPRN_HFSCR, r7 ALT_FTR_SECTION_END_IFCLR(CPU_FTR_ARCH_300) +BEGIN_FTR_SECTION + ld r5, STACK_SLOT_FSCR(r1) + mtspr SPRN_FSCR, r5 +END_FTR_SECTION_IFSET(CPU_FTR_ARCH_207S) /* * Restore various registers to 0, where non-zero values * set by the guest could disrupt the host.
Greg Kroah-Hartman
maintainerPATCH 134/161 nfc: fix NULL ptr dereference in llcp_sock_getname() after failed connect
In reply to Greg Kroah-HartmanFrom: Krzysztof Kozlowski <[email protected]>
commit 4ac06a1e013cf5fdd963317ffd3b968560f33bba upstream.
It's possible to trigger NULL pointer dereference by local unprivileged user, when calling getsockname() after failed bind() (e.g. the bind fails because LLCP_SAP_MAX used as SAP):
BUG: kernel NULL pointer dereference, address: 0000000000000000 CPU: 1 PID: 426 Comm: llcp_sock_getna Not tainted 5.13.0-rc2-next-20210521+ #9 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.14.0-1 04/01/2014 Call Trace: llcp_sock_getname+0xb1/0xe0 __sys_getpeername+0x95/0xc0 ? lockdep_hardirqs_on_prepare+0xd5/0x180 ? syscall_enter_from_user_mode+0x1c/0x40 __x64_sys_getpeername+0x11/0x20 do_syscall_64+0x36/0x70 entry_SYSCALL_64_after_hwframe+0x44/0xae
This can be reproduced with Syzkaller C repro (bind followed by getpeername):
https://syzkaller.appspot.com/x/repro.c?x=14def446e00000
--- net/nfc/llcp_sock.c | 2 ++ 1 file changed, 2 insertions(+)
net/nfc/llcp_sock.c +2 -0
@@ -110,6 +110,7 @@ static int llcp_sock_bind(struct socketif (!llcp_sock->service_name) {nfc_llcp_local_put (llcp_sock->local); llcp_sock->local =NULL ; + llcp_sock->dev =NULL ; ret = -ENOMEM ; goto put_dev; }
@@ -119,6 +120,7 @@ static int llcp_sock_bind(struct socket llcp_sock->local =NULL ;kfree (llcp_sock->service_name); llcp_sock->service_name =NULL ; + llcp_sock->dev =NULL ; ret = -EADDRINUSE ; goto put_dev; }
Greg Kroah-Hartman
maintainerPATCH 135/161 drm/amdgpu: Dont query CE and UE errors
In reply to Greg Kroah-HartmanFrom: Luben Tuikov <[email protected]>
commit dce3d8e1d070900e0feeb06787a319ff9379212c upstream.
On QUERY2 IOCTL don't query counts of correctable and uncorrectable errors, since when RAS is enabled and supported on Vega20 server boards, this takes insurmountably long time, in O(n^3), which slows the system down to the point of it being unusable when we have GUI up.
--- drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c | 16 ---------------- 1 file changed, 16 deletions(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.c +0 -16
@@ -337,7 +337,6 @@ static int amdgpu_ctx_query2(struct amdg {struct amdgpu_ctx *ctx;struct amdgpu_ctx_mgr *mgr; -unsigned long ras_counter;if (!fpriv)return -EINVAL ;
@@ -362,21 +361,6 @@ static int amdgpu_ctx_query2(struct amdgif (atomic_read (&ctx->guilty)) out->state.flags |=AMDGPU_CTX_QUERY2_FLAGS_GUILTY ; -/*query ue count*/ - ras_counter =amdgpu_ras_query_error_count (adev, false); -/*ras counter is monotonic increasing*/ -if (ras_counter != ctx->ras_counter_ue) { - out->state.flags |=AMDGPU_CTX_QUERY2_FLAGS_RAS_UE ; - ctx->ras_counter_ue = ras_counter; - } - -/*query ce count*/ - ras_counter =amdgpu_ras_query_error_count (adev, true); -if (ras_counter != ctx->ras_counter_ce) { - out->state.flags |=AMDGPU_CTX_QUERY2_FLAGS_RAS_CE ; - ctx->ras_counter_ce = ras_counter; - } -mutex_unlock (&mgr->lock);return 0 ; }
Greg Kroah-Hartman
maintainerPATCH 136/161 drm/amdgpu: make sure we unpin the UVD BO
In reply to Greg Kroah-HartmanFrom: Nirmoy Das <[email protected]>
commit 07438603a07e52f1c6aa731842bd298d2725b7be upstream.
Releasing pinned BOs is illegal now. UVD 6 was missing from:
commit 2f40801dc553 ("drm/amdgpu: make sure we unpin the UVD BO")
--- drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c | 1 + 1 file changed, 1 insertion(+)
drivers/gpu/drm/amd/amdgpu/uvd_v6_0.c +1 -0
@@ -357,6 +357,7 @@ static int uvd_v6_0_enc_ring_test_ib(str error:dma_fence_put (fence); +amdgpu_bo_unpin (bo);amdgpu_bo_unreserve (bo);amdgpu_bo_unref (&bo);return r;
Greg Kroah-Hartman
maintainerPATCH 137/161 x86/apic: Mark _all_ legacy interrupts when IO/APIC is missing
In reply to Greg Kroah-HartmanFrom: Thomas Gleixner <[email protected]>
commit 7d65f9e80646c595e8c853640a9d0768a33e204c upstream.
PIC interrupts do not support affinity setting and they can end up on any online CPU. Therefore, it's required to mark the associated vectors as system-wide reserved. Otherwise, the corresponding irq descriptors are copied to the secondary CPUs but the vectors are not marked as assigned or reserved. This works correctly for the IO/APIC case.
When the IO/APIC is disabled via config, kernel command line or lack of enumeration then all legacy interrupts are routed through the PIC, but nothing marks them as system-wide reserved vectors.
As a consequence, a subsequent allocation on a secondary CPU can result in allocating one of these vectors, which triggers the BUG() in apic_update_vector() because the interrupt descriptor slot is not empty.
Imran tried to work around that by marking those interrupts as allocated when a CPU comes online. But that's wrong in case that the IO/APIC is available and one of the legacy interrupts, e.g. IRQ0, has been switched to PIC mode because then marking them as allocated will fail as they are already marked as system vectors.
Stay consistent and update the legacy vectors after attempting IO/APIC initialization and mark them as system vectors in case that no IO/APIC is available.
--- arch/x86/include/asm/apic.h | 1 + arch/x86/kernel/apic/apic.c | 1 + arch/x86/kernel/apic/vector.c | 20 ++++++++++++++++++++ 3 files changed, 22 insertions(+)
arch/x86/include/asm/apic.h +1 -0
@@ -174,6 +174,7 @@ static inline int apic_is_clustered_box(extern int setup_APIC_eilvt (u8 lvt_off,u8 vector,u8 msg_type,u8 mask);extern void lapic_assign_system_vectors (void );extern void lapic_assign_legacy_vector (unsigned int isairq,bool replace); +extern void lapic_update_legacy_vectors (void );extern void lapic_online (void );extern void lapic_offline (void );extern bool apic_needs_pit (void );
arch/x86/kernel/apic/apic.c +1 -0
@@ -2604,6 +2604,7 @@ static void __init apic_bsp_setup(bool uend_local_APIC_setup ();irq_remap_enable_fault_handling ();setup_IO_APIC (); +lapic_update_legacy_vectors (); }#ifdef CONFIG_UP_LATE_INIT
arch/x86/kernel/apic/vector.c +20 -0
@@ -730,6 +730,26 @@ void lapic_assign_legacy_vector(unsignedirq_matrix_assign_system (vector_matrix,ISA_IRQ_VECTOR (irq), replace); } +void __init lapic_update_legacy_vectors (void ) +{ +unsigned int i; + +if (IS_ENABLED (CONFIG_X86_IO_APIC ) && nr_ioapics >0 ) +return ; + +/* +* If the IO/APIC is disabled via config, kernel command line or +* lack of enumeration then all legacy interrupts are routed +* through the PIC. Make sure that they are marked as legacy +* vectors. PIC_CASCADE_IRQ has already been marked in +* lapic_assign_system_vectors(). +*/ +for (i =0 ; i <nr_legacy_irqs (); i++) { +if (i !=PIC_CASCADE_IR ) +lapic_assign_legacy_vector (i, true); + } +} +void __init lapic_assign_system_vectors (void ) {unsigned int i, vector =0 ;
Greg Kroah-Hartman
maintainerPATCH 138/161 x86/thermal: Fix LVT thermal setup for SMI delivery mode
In reply to Greg Kroah-HartmanFrom: Borislav Petkov <[email protected]>
commit 9a90ed065a155d13db0d0ffeaad5cc54e51c90c6 upstream.
There are machines out there with added value crap^WBIOS which provide an SMI handler for the local APIC thermal sensor interrupt. Out of reset, the BSP on those machines has something like 0x200 in that APIC register (timestamps left in because this whole issue is timing sensitive):
[ 0.033858] read lvtthmr: 0x330, val: 0x200
which means:
- bit 16 - the interrupt mask bit is clear and thus that interrupt is enabled
- bits [10:8] have 010b which means SMI delivery mode.
Now, later during boot, when the kernel programs the local APIC, it soft-disables it temporarily through the spurious vector register:
setup_local_APIC:
...
/* * If this comes from kexec/kcrash the APIC might be enabled in * SPIV. Soft disable it before doing further initialization. */ value = apic_read(APIC_SPIV); value &= ~APIC_SPIV_APIC_ENABLED; apic_write(APIC_SPIV, value);
which means (from the SDM):
"10.4.7.2 Local APIC State After It Has Been Software Disabled
...
* The mask bits for all the LVT entries are set. Attempts to reset these bits will be ignored."
And this happens too:
[ 0.124111] APIC: Switch to symmetric I/O mode setup [ 0.124117] lvtthmr 0x200 before write 0xf to APIC 0xf0 [ 0.124118] lvtthmr 0x10200 after write 0xf to APIC 0xf0
This results in CPU 0 soft lockups depending on the placement in time when the APIC soft-disable happens. Those soft lockups are not 100% reproducible and the reason for that can only be speculated as no one tells you what SMM does. Likely, it confuses the SMM code that the APIC is disabled and the thermal interrupt doesn't doesn't fire at all, leading to CPU 0 stuck in SMM forever...
Now, before
4f432e8bb15b ("x86/mce: Get rid of mcheck_intel_therm_init()")
due to how the APIC_LVTTHMR was read before APIC initialization in mcheck_intel_therm_init(), it would read the value with the mask bit 16 clear and then intel_init_thermal() would replicate it onto the APs and all would be peachy - the thermal interrupt would remain enabled.
But that commit moved that reading to a later moment in intel_init_thermal(), resulting in reading APIC_LVTTHMR on the BSP too late and with its interrupt mask bit set.
Thus, revert back to the old behavior of reading the thermal LVT register before the APIC gets initialized.
--- arch/x86/include/asm/thermal.h | 4 +++- arch/x86/kernel/setup.c | 9 +++++++++ drivers/thermal/intel/therm_throt.c | 15 +++++++++++---- 3 files changed, 23 insertions(+), 5 deletions(-)
arch/x86/include/asm/thermal.h +3 -1
@@ -3,11 +3,13 @@#define _ASM_X86_THERMAL_H#ifdef CONFIG_X86_THERMAL_VECTOR +void therm_lvt_init (void );void intel_init_thermal (struct cpuinfo_x86 *c);bool x86_thermal_enabled (void );void intel_thermal_interrupt (void );#else -static inline void intel_init_thermal (struct cpuinfo_x86 *c) { } +static inline void therm_lvt_init (void ) { } +static inline void intel_init_thermal (struct cpuinfo_x86 *c) { }#endif #endif /* _ASM_X86_THERMAL_H */
arch/x86/kernel/setup.c +9 -0
@@ -44,6 +44,7 @@#include <asm/pci-direct.h> #include <asm/prom.h> #include <asm/proto.h> +#include <asm/thermal.h> #include <asm/unwind.h> #include <asm/vsyscall.h> #include <linux/vmalloc.h>
@@ -1220,6 +1221,14 @@ void __init setup_arch(char **cmdline_p) x86_init.timers.wallclock_init (); +/* +* This needs to run before setup_local_APIC() which soft-disables the +* local APIC temporarily and that masks the thermal LVT interrupt, +* leading to softlockups on machines which have configured SMI +* interrupt delivery. +*/ +therm_lvt_init (); +mcheck_init ();register_refined_jiffies (CLOCK_TICK_RATE );
drivers/thermal/intel/therm_throt.c +11 -4
@@ -621,6 +621,17 @@ bool x86_thermal_enabled(void)return atomic_read (&therm_throt_en); } +void __init therm_lvt_init (void ) +{ +/* +* This function is only called on boot CPU. Save the init thermal +* LVT value on BSP and use that value to restore APs' thermal LVT +* entry BIOS programmed later +*/ +if (intel_thermal_supported (&boot_cpu_data)) + lvtthmr_init =apic_read (APIC_LVTTHMR ); +} +void intel_init_thermal (struct cpuinfo_x86 *c) {unsigned int cpu =smp_processor_id ();
@@ -630,10 +641,6 @@ void intel_init_thermal(struct cpuinfo_xif (!intel_thermal_supported (c))return ; -/* On the BSP? */ -if (c == &boot_cpu_data) - lvtthmr_init =apic_read (APIC_LVTTHMR ); -/* * First check if its enabled already, in which case there might * be some SMM goo which handles it, so we can't even put a handler
Greg Kroah-Hartman
maintainerPATCH 139/161 powerpc/kprobes: Fix validation of prefixed instructions across page boundary
In reply to Greg Kroah-HartmanFrom: Naveen N. Rao <[email protected]>
commit 82123a3d1d5a306fdf50c968a474cc60fe43a80f upstream.
When checking if the probed instruction is the suffix of a prefixed instruction, we access the instruction at the previous word. If the probed instruction is the very first word of a module, we can end up trying to access an invalid page.
Fix this by skipping the check for all instructions at the beginning of a page. Prefixed instructions cannot cross a 64-byte boundary and as such, we don't expect to encounter a suffix as the very first word in a page for kernel text. Even if there are prefixed instructions crossing a page boundary (from a module, for instance), the instruction will be illegal, so preventing probing on the suffix of such prefix instructions isn't worthwhile.
--- arch/powerpc/kernel/kprobes.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
arch/powerpc/kernel/kprobes.c +2 -2
@@ -108,7 +108,6 @@ int arch_prepare_kprobe(struct kprobe *pint ret =0 ;struct kprobe *prev;struct ppc_inst insn =ppc_inst_read ((struct ppc_inst *)p->addr); -struct ppc_inst prefix =ppc_inst_read ((struct ppc_inst *)(p->addr -1 ));if ((unsigned long )p->addr &0x03 ) {printk ("Attempt to register kprobe at an unaligned address\n" );
@@ -116,7 +115,8 @@ int arch_prepare_kprobe(struct kprobe *p }else if (IS_MTMSRD (insn) ||IS_RFID (insn) ||IS_RFI (insn)) {printk ("Cannot register a kprobe on rfi/rfid or mtmsr[d]\n" ); ret = -EINVAL ; - }else if (ppc_inst_prefixed (prefix)) { + }else if ((unsigned long )p->addr & ~PAGE_MASK && +ppc_inst_prefixed (ppc_inst_read ((struct ppc_inst *)(p->addr -1 )))) {printk ("Cannot register a kprobe on the second word of prefixed instruction\n" ); ret = -EINVAL ; }
Greg Kroah-Hartman
maintainerPATCH 140/161 btrfs: mark ordered extent and inode with error if we fail to finish
In reply to Greg Kroah-HartmanFrom: Josef Bacik <[email protected]>
commit d61bec08b904cf171835db98168f82bc338e92e4 upstream.
While doing error injection testing I saw that sometimes we'd get an abort that wouldn't stop the current transaction commit from completing. This abort was coming from finish ordered IO, but at this point in the transaction commit we should have gotten an error and stopped.
It turns out the abort came from finish ordered io while trying to write out the free space cache. It occurred to me that any failure inside of finish_ordered_io isn't actually raised to the person doing the writing, so we could have any number of failures in this path and think the ordered extent completed successfully and the inode was fine.
Fix this by marking the ordered extent with BTRFS_ORDERED_IOERR, and marking the mapping of the inode with mapping_set_error, so any callers that simply call fdatawait will also get the error.
With this we're seeing the IO error on the free space inode when we fail to do the finish_ordered_io.
--- fs/btrfs/inode.c | 12 ++++++++++++ 1 file changed, 12 insertions(+)
fs/btrfs/inode.c +12 -0
@@ -3011,6 +3011,18 @@ out:if (ret || truncated) {u64 unwritten_start = start; +/* +* If we failed to finish this ordered extent for any reason we +* need to make sure BTRFS_ORDERED_IOERR is set on the ordered +* extent, and mark the inode with the error if it wasn't +* already set. Any error during writeback would have already +* set the mapping error, so we need to set it if we're the ones +* marking this ordered extent as failed. +*/ +if (ret && !test_and_set_bit (BTRFS_ORDERED_IOERR , + &ordered_extent->flags)) +mapping_set_error (ordered_extent->inode->i_mapping, -EIO ); +if (truncated) unwritten_start += logical_len;clear_extent_uptodate (io_tree, unwritten_start, end,NULL );
Greg Kroah-Hartman
maintainerPATCH 141/161 btrfs: fix error handling in btrfs_del_csums
In reply to Greg Kroah-HartmanFrom: Josef Bacik <[email protected]>
commit b86652be7c83f70bf406bed18ecf55adb9bfb91b upstream.
Error injection stress would sometimes fail with checksums on disk that did not have a corresponding extent. This occurred because the pattern in btrfs_del_csums was
while (1) {
ret = btrfs_search_slot();
if (ret < 0)
break;
}
ret = 0;
out:
btrfs_free_path(path);
return ret;If we got an error from btrfs_search_slot we'd clear the error because we were breaking instead of goto out. Instead of using goto out, simply handle the cases where we may leave a random value in ret, and get rid of the
ret = 0;
out:
pattern and simply allow break to have the proper error reporting. With this fix we properly abort the transaction and do not commit thinking we successfully deleted the csum.
--- fs/btrfs/file-item.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-)
fs/btrfs/file-item.c +5 -5
@@ -787,7 +787,7 @@ int btrfs_del_csums(struct btrfs_trans_hu64 end_byte = bytenr + len;u64 csum_end;struct extent_buffer *leaf; -int ret; +int ret =0 ;const u32 csum_size = fs_info->csum_size;u32 blocksize_bits = fs_info->sectorsize_bits;
@@ -805,6 +805,7 @@ int btrfs_del_csums(struct btrfs_trans_h ret =btrfs_search_slot (trans, root, &key, path,-1 ,1 );if (ret >0 ) { + ret =0 ;if (path->slots[0 ] ==0 )break ; path->slots[0 ]--;
@@ -861,7 +862,7 @@ int btrfs_del_csums(struct btrfs_trans_h ret =btrfs_del_items (trans, root, path, path->slots[0 ], del_nr);if (ret) - goto out; +break ;if (key.offset == bytenr)break ; }else if (key.offset < bytenr && csum_end > end_byte) {
@@ -905,8 +906,9 @@ int btrfs_del_csums(struct btrfs_trans_h ret =btrfs_split_item (trans, root, path, &key, offset);if (ret && ret != -EAGAIN ) {btrfs_abort_transaction (trans, ret); - goto out; +break ; } + ret =0 ; key.offset = end_byte -1 ; }else {
@@ -916,8 +918,6 @@ int btrfs_del_csums(struct btrfs_trans_h }btrfs_release_path (path); } - ret =0 ; -out:btrfs_free_path (path);return ret; }
Greg Kroah-Hartman
maintainerPATCH 142/161 btrfs: return errors from btrfs_del_csums in cleanup_ref_head
In reply to Greg Kroah-HartmanFrom: Josef Bacik <[email protected]>
commit 856bd270dc4db209c779ce1e9555c7641ffbc88e upstream.
We are unconditionally returning 0 in cleanup_ref_head, despite the fact that btrfs_del_csums could fail. We need to return the error so the transaction gets aborted properly, fix this by returning ret from btrfs_del_csums in cleanup_ref_head.
--- fs/btrfs/extent-tree.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/btrfs/extent-tree.c +1 -1
@@ -1868,7 +1868,7 @@ static int cleanup_ref_head(struct btrfstrace_run_delayed_ref_head (fs_info, head,0 );btrfs_delayed_ref_unlock (head);btrfs_put_delayed_ref_head (head); -return 0 ; +return ret; } static struct btrfs_delayed_ref_head *btrfs_obtain_ref_head(
Greg Kroah-Hartman
maintainerPATCH 143/161 btrfs: fix fsync failure and transaction abort after writes to prealloc extents
In reply to Greg Kroah-HartmanFrom: Filipe Manana <[email protected]>
commit ea7036de0d36c4e6c9508f68789e9567d514333a upstream.
When doing a series of partial writes to different ranges of preallocated extents with transaction commits and fsyncs in between, we can end up with a checksum items in a log tree. This causes an fsync to fail with -EIO and abort the transaction, turning the filesystem to RO mode, when syncing the log.
For this to happen, we need to have a full fsync of a file following one or more fast fsyncs.
The following example reproduces the problem and explains how it happens:
$ mkfs.btrfs -f /dev/sdc $ mount /dev/sdc /mnt
# Create our test file with 2 preallocated extents. Leave a 1M hole
# between them to ensure that we get two file extent items that will
# never be merged into a single one. The extents are contiguous on disk,
# which will later result in the checksums for their data to be merged
# into a single checksum item in the csums btree.
#
$ xfs_io -f \
-c "falloc 0 1M" \
-c "falloc 3M 3M" \
/mnt/foobar
# Now write to the second extent and leave only 1M of it as unwritten,
# which corresponds to the file range [4M, 5M[.
#
# Then fsync the file to flush delalloc and to clear full sync flag from
# the inode, so that a future fsync will use the fast code path.
#
# After the writeback triggered by the fsync we have 3 file extent items
# that point to the second extent we previously allocated:
#
# 1) One file extent item of type BTRFS_FILE_EXTENT_REG that covers the
# file range [3M, 4M[
#
# 2) One file extent item of type BTRFS_FILE_EXTENT_PREALLOC that covers
# the file range [4M, 5M[
#
# 3) One file extent item of type BTRFS_FILE_EXTENT_REG that covers the
# file range [5M, 6M[
#
# All these file extent items have a generation of 6, which is the ID of
# the transaction where they were created. The split of the original file
# extent item is done at btrfs_mark_extent_written() when ordered extents
# complete for the file ranges [3M, 4M[ and [5M, 6M[.
#
$ xfs_io -c "pwrite -S 0xab 3M 1M" \
-c "pwrite -S 0xef 5M 1M" \
-c "fsync" \
/mnt/foobar# Commit the current transaction. This wipes out the log tree created by
# the previous fsync. sync
# Now write to the unwritten range of the second extent we allocated,
# corresponding to the file range [4M, 5M[, and fsync the file, which
# triggers the fast fsync code path.
#
# The fast fsync code path sees that there is a new extent map covering
# the file range [4M, 5M[ and therefore it will log a checksum item
# covering the range [1M, 2M[ of the second extent we allocated.
#
# Also, after the fsync finishes we no longer have the 3 file extent
# items that pointed to 3 sections of the second extent we allocated.
# Instead we end up with a single file extent item pointing to the whole
# extent, with a type of BTRFS_FILE_EXTENT_REG and a generation of 7 (the
# current transaction ID). This is due to the file extent item merging we
# do when completing ordered extents into ranges that point to unwritten
# (preallocated) extents. This merging is done at
# btrfs_mark_extent_written().
#
$ xfs_io -c "pwrite -S 0xcd 4M 1M" \
-c "fsync" \
/mnt/foobar
# Now do some write to our file outside the range of the second extent
# that we allocated with fallocate() and truncate the file size from 6M
# down to 5M.
#
# The truncate operation sets the full sync runtime flag on the inode,
# forcing the next fsync to use the slow code path. It also changes the
# length of the second file extent item so that it represents the file
# range [3M, 5M[ and not the range [3M, 6M[ anymore.
#
# Finally fsync the file. Since this is a fsync that triggers the slow
# code path, it will remove all items associated to the inode from the
# log tree and then it will scan for file extent items in the
# fs/subvolume tree that have a generation matching the current
# transaction ID, which is 7. This means it will log 2 file extent
# items:
#
# 1) One for the first extent we allocated, covering the file range
# [0, 1M[
#
# 2) Another for the first 2M of the second extent we allocated,
# covering the file range [3M, 5M[
#
# When logging the first file extent item we log a single checksum item
# that has all the checksums for the entire extent.
#
# When logging the second file extent item, we also lookup for the
# checksums that are associated with the range [0, 2M[ of the second
# extent we allocated (file range [3M, 5M[), and then we log them with
# btrfs_csum_file_blocks(). However that results in ending up with a log
# that has two checksum items with ranges that overlap:
#
# 1) One for the range [1M, 2M[ of the second extent we allocated,
# corresponding to the file range [4M, 5M[, which we logged in the
# previous fsync that used the fast code path;
#
# 2) One for the ranges [0, 1M[ and [0, 2M[ of the first and second
# extents, respectively, corresponding to the files ranges [0, 1M[
# and [3M, 5M[. This one was added during this last fsync that uses
# the slow code path and overlaps with the previous one logged by
# the previous fast fsync.
#
# This happens because when logging the checksums for the second
# extent, we notice they start at an offset that matches the end of the
# checksums item that we logged for the first extent, and because both
# extents are contiguous on disk, btrfs_csum_file_blocks() decides to
# extend that existing checksums item and append the checksums for the
# second extent to this item. The end result is we end up with two
# checksum items in the log tree that have overlapping ranges, as
# listed before, resulting in the fsync to fail with -EIO and aborting
# the transaction, turning the filesystem into RO mode.
#
$ xfs_io -c "pwrite -S 0xff 0 1M" \
-c "truncate 5M" \
-c "fsync" \
/mnt/foobar
fsync: Input/output errorAfter running the example, dmesg/syslog shows the tree checker complained about the checksum items with overlapping ranges and we aborted the transaction:
$ dmesg (...) [756289.557487] BTRFS critical (device sdc): corrupt leaf: root=18446744073709551610 block=30720000 slot=5, csum end range (16777216) goes beyond the start range (15728640) of the next csum item [756289.560583] BTRFS info (device sdc): leaf 30720000 gen 7 total ptrs 7 free space 11677 owner 18446744073709551610 [756289.562435] BTRFS info (device sdc): refs 2 lock_owner 0 current 2303929 [756289.563654] item 0 key (257 1 0) itemoff 16123 itemsize 160 [756289.564649] inode generation 6 size 5242880 mode 100600 [756289.565636] item 1 key (257 12 256) itemoff 16107 itemsize 16 [756289.566694] item 2 key (257 108 0) itemoff 16054 itemsize 53 [756289.567725] extent data disk bytenr 13631488 nr 1048576 [756289.568697] extent data offset 0 nr 1048576 ram 1048576 [756289.569689] item 3 key (257 108 1048576) itemoff 16001 itemsize 53 [756289.570682] extent data disk bytenr 0 nr 0 [756289.571363] extent data offset 0 nr 2097152 ram 2097152 [756289.572213] item 4 key (257 108 3145728) itemoff 15948 itemsize 53 [756289.573246] extent data disk bytenr 14680064 nr 3145728 [756289.574121] extent data offset 0 nr 2097152 ram 3145728 [756289.574993] item 5 key (18446744073709551606 128 13631488) itemoff 12876 itemsize 3072 [756289.576113] item 6 key (18446744073709551606 128 15728640) itemoff 11852 itemsize 1024 [756289.577286] BTRFS error (device sdc): block=30720000 write time tree block corruption detected [756289.578644] ------------[ cut here ]------------ [756289.579376] WARNING: CPU: 0 PID: 2303929 at fs/btrfs/disk-io.c:465 csum_one_extent_buffer+0xed/0x100 [btrfs] [756289.580857] Modules linked in: btrfs dm_zero dm_dust loop dm_snapshot (...) [756289.591534] CPU: 0 PID: 2303929 Comm: xfs_io Tainted: G W 5.12.0-rc8-btrfs-next-87 #1 [756289.592580] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.14.0-0-g155821a1990b-prebuilt.qemu.org 04/01/2014 [756289.594161] RIP: 0010:csum_one_extent_buffer+0xed/0x100 [btrfs] [756289.595122] Code: 5d c3 e8 76 60 (...) [756289.597509] RSP: 0018:ffffb51b416cb898 EFLAGS: 00010282 [756289.598142] RAX: 0000000000000000 RBX: fffff02b8a365bc0 RCX: 0000000000000000 [756289.598970] RDX: 0000000000000000 RSI: ffffffffa9112421 RDI: 00000000ffffffff [756289.599798] RBP: ffffa06500880000 R08: 0000000000000000 R09: 0000000000000000 [756289.600619] R10: 0000000000000000 R11: 0000000000000001 R12: 0000000000000000 [756289.601456] R13: ffffa0652b1d8980 R14: ffffa06500880000 R15: 0000000000000000 [756289.602278] FS: 00007f08b23c9800(0000) GS:ffffa0682be00000(0000) knlGS:0000000000000000 [756289.603217] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [756289.603892] CR2: 00005652f32d0138 CR3: 000000025d616003 CR4: 0000000000370ef0 [756289.604725] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 [756289.605563] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 [756289.606400] Call Trace: [756289.606704] btree_csum_one_bio+0x244/0x2b0 [btrfs] [756289.607313] btrfs_submit_metadata_bio+0xb7/0x100 [btrfs] [756289.608040] submit_one_bio+0x61/0x70 [btrfs] [756289.608587] btree_write_cache_pages+0x587/0x610 [btrfs] [756289.609258] ? free_debug_processing+0x1d5/0x240 [756289.609812] ? __module_address+0x28/0xf0 [756289.610298] ? lock_acquire+0x1a0/0x3e0 [756289.610754] ? lock_acquired+0x19f/0x430 [756289.611220] ? lock_acquire+0x1a0/0x3e0 [756289.611675] do_writepages+0x43/0xf0 [756289.612101] ? __filemap_fdatawrite_range+0xa4/0x100 [756289.612800] __filemap_fdatawrite_range+0xc5/0x100 [756289.613393] btrfs_write_marked_extents+0x68/0x160 [btrfs] [756289.614085] btrfs_sync_log+0x21c/0xf20 [btrfs] [756289.614661] ? finish_wait+0x90/0x90 [756289.615096] ? __mutex_unlock_slowpath+0x45/0x2a0 [756289.615661] ? btrfs_log_inode_parent+0x3c9/0xdc0 [btrfs] [756289.616338] ? lock_acquire+0x1a0/0x3e0 [756289.616801] ? lock_acquired+0x19f/0x430 [756289.617284] ? lock_acquire+0x1a0/0x3e0 [756289.617750] ? lock_release+0x214/0x470 [756289.618221] ? lock_acquired+0x19f/0x430 [756289.618704] ? dput+0x20/0x4a0 [756289.619079] ? dput+0x20/0x4a0 [756289.619452] ? lockref_put_or_lock+0x9/0x30 [756289.619969] ? lock_release+0x214/0x470 [756289.620445] ? lock_release+0x214/0x470 [756289.620924] ? lock_release+0x214/0x470 [756289.621415] btrfs_sync_file+0x46a/0x5b0 [btrfs] [756289.621982] do_fsync+0x38/0x70 [756289.622395] __x64_sys_fsync+0x10/0x20 [756289.622907] do_syscall_64+0x33/0x80 [756289.623438] entry_SYSCALL_64_after_hwframe+0x44/0xae [756289.624063] RIP: 0033:0x7f08b27fbb7b [756289.624588] Code: 0f 05 48 3d 00 (...) [756289.626760] RSP: 002b:00007ffe2583f940 EFLAGS: 00000293 ORIG_RAX: 000000000000004a [756289.627639] RAX: ffffffffffffffda RBX: 00005652f32cd0f0 RCX: 00007f08b27fbb7b [756289.628464] RDX: 00005652f32cbca0 RSI: 00005652f32cd110 RDI: 0000000000000003 [756289.629323] RBP: 00005652f32cd110 R08: 0000000000000000 R09: 00007f08b28c4be0 [756289.630172] R10: fffffffffffff39a R11: 0000000000000293 R12: 0000000000000001 [756289.631007] R13: 00005652f32cd0f0 R14: 0000000000000001 R15: 00005652f32cc480 [756289.631819] irq event stamp: 0 [756289.632188] hardirqs last enabled at (0): [<0000000000000000>] 0x0 [756289.632911] hardirqs last disabled at (0): [<ffffffffa7e97c29>] copy_process+0x879/0x1cc0 [756289.633893] softirqs last enabled at (0): [<ffffffffa7e97c29>] copy_process+0x879/0x1cc0 [756289.634871] softirqs last disabled at (0): [<0000000000000000>] 0x0 [756289.635606] ---[ end trace 0a039fdc16ff3fef ]--- [756289.636179] BTRFS: error (device sdc) in btrfs_sync_log:3136: errno=-5 IO failure [756289.637082] BTRFS info (device sdc): forced readonly
Having checksum items covering ranges that overlap is dangerous as in some cases it can lead to having extent ranges for which we miss checksums after log replay or getting the wrong checksum item. There were some fixes in the past for bugs that resulted in this problem, and were explained and fixed by the following commits:
27b9a8122ff71a ("Btrfs: fix csum tree corruption, duplicate and outdated checksums") b84b8390d6009c ("Btrfs: fix file read corruption after extent cloning and fsync") 40e046acbd2f36 ("Btrfs: fix missing data checksums after replaying a log tree") e289f03ea79bbc ("btrfs: fix corrupt log due to concurrent fsync of inodes with shared extents")
Fix the issue by making btrfs_csum_file_blocks() taking into account the start offset of the next checksum item when it decides to extend an existing checksum item, so that it never extends the checksum to end at a range that goes beyond the start range of the next checksum item.
When we can not access the next checksum item without releasing the path, simply drop the optimization of extending the previous checksum item and fallback to inserting a new checksum item - this happens rarely and the optimization is not significant enough for a log tree in order to justify the extra complexity, as it would only save a few bytes (the size of a struct btrfs_item) of leaf space.
This behaviour is only needed when inserting into a log tree because for the regular checksums tree we never have a case where we try to insert a range of checksums that overlap with a range that was previously inserted.
A test case for fstests will follow soon.
--- fs/btrfs/file-item.c | 98 +++++++++++++++++++++++++++++++++++++++------------ 1 file changed, 76 insertions(+), 22 deletions(-)
fs/btrfs/file-item.c +76 -22
@@ -922,6 +922,37 @@ int btrfs_del_csums(struct btrfs_trans_hreturn ret; } +static int find_next_csum_offset (struct btrfs_root *root, +struct btrfs_path *path, +u64 *next_offset) +{ +const u32 nritems =btrfs_header_nritems (path->nodes[0 ]); +struct btrfs_key found_key; +int slot = path->slots[0 ] +1 ; +int ret; + +if (nritems ==0 || slot >= nritems) { + ret =btrfs_next_leaf (root, path); +if (ret <0 ) { +return ret; + }else if (ret >0 ) { + *next_offset = (u64)-1 ; +return 0 ; + } + slot = path->slots[0 ]; + } + +btrfs_item_key_to_cpu (path->nodes[0 ], &found_key, slot); + +if (found_key.objectid !=BTRFS_EXTENT_CSUM_OBJECTID || + found_key.type !=BTRFS_EXTENT_CSUM_KEY ) + *next_offset = (u64)-1 ; +else + *next_offset = found_key.offset; + +return 0 ; +} +int btrfs_csum_file_blocks (struct btrfs_trans_handle *trans,struct btrfs_root *root,struct btrfs_ordered_sum *sums)
@@ -937,7 +968,6 @@ int btrfs_csum_file_blocks(struct btrfs_u64 total_bytes =0 ;u64 csum_offset;u64 bytenr; -u32 nritems;u32 ins_size;int index =0 ;int found_next;
@@ -980,26 +1010,10 @@ again: goto insert; } }else { -int slot = path->slots[0 ] +1 ; -/* we didn't find a csum item, insert one */ - nritems =btrfs_header_nritems (path->nodes[0 ]); -if (!nritems || (path->slots[0 ] >= nritems -1 )) { - ret =btrfs_next_leaf (root, path); -if (ret <0 ) { - goto out; - }else if (ret >0 ) { - found_next =1 ; - goto insert; - } - slot = path->slots[0 ]; - } -btrfs_item_key_to_cpu (path->nodes[0 ], &found_key, slot); -if (found_key.objectid !=BTRFS_EXTENT_CSUM_OBJECTID || - found_key.type !=BTRFS_EXTENT_CSUM_KEY ) { - found_next =1 ; - goto insert; - } - next_offset = found_key.offset; +/* We didn't find a csum item, insert one. */ + ret =find_next_csum_offset (root, path, &next_offset); +if (ret <0 ) + goto out; found_next =1 ; goto insert; }
@@ -1055,8 +1069,48 @@ extend_csum: tmp = sums->len - total_bytes; tmp >>= fs_info->sectorsize_bits;WARN_ON (tmp <1 ); + extend_nr =max_t (int,1 , tmp); + +/* +* A log tree can already have checksum items with a subset of +* the checksums we are trying to log. This can happen after +* doing a sequence of partial writes into prealloc extents and +* fsyncs in between, with a full fsync logging a larger subrange +* of an extent for which a previous fast fsync logged a smaller +* subrange. And this happens in particular due to merging file +* extent items when we complete an ordered extent for a range +* covered by a prealloc extent - this is done at +* btrfs_mark_extent_written(). +* +* So if we try to extend the previous checksum item, which has +* a range that ends at the start of the range we want to insert, +* make sure we don't extend beyond the start offset of the next +* checksum item. If we are at the last item in the leaf, then +* forget the optimization of extending and add a new checksum +* item - it is not worth the complexity of releasing the path, +* getting the first key for the next leaf, repeat the btree +* search, etc, because log trees are temporary anyway and it +* would only save a few bytes of leaf space. +*/ +if (root->root_key.objectid ==BTRFS_TREE_LOG_OBJECTID ) { +if (path->slots[0 ] +1 >= +btrfs_header_nritems (path->nodes[0 ])) { + ret =find_next_csum_offset (root, path, &next_offset); +if (ret <0 ) + goto out; + found_next =1 ; + goto insert; + } + + ret =find_next_csum_offset (root, path, &next_offset); +if (ret <0 ) + goto out; + + tmp = (next_offset - bytenr) >> fs_info->sectorsize_bits; +if (tmp <=INT_MAX ) + extend_nr =min_t (int, extend_nr, tmp); + } - extend_nr =max_t (int,1 , (int )tmp); diff = (csum_offset + extend_nr) * csum_size; diff =min (diff,MAX_CSUM_ITEMS (fs_info, csum_size) * csum_size);
Greg Kroah-Hartman
maintainerPATCH 144/161 btrfs: check error value from btrfs_update_inode in tree log
In reply to Greg Kroah-HartmanFrom: Josef Bacik <[email protected]>
commit f96d44743a44e3332f75d23d2075bb8270900e1d upstream.
Error injection testing uncovered a case where we ended up with invalid link counts on an inode. This happened because we failed to notice an error when updating the inode while replaying the tree log, and committed the transaction with an invalid file system.
Fix this by checking the return value of btrfs_update_inode. This resolved the link count errors I was seeing, and we already properly handle passing up the error values in these paths.
--- fs/btrfs/tree-log.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
fs/btrfs/tree-log.c +6 -2
@@ -1574,7 +1574,9 @@ static noinline int add_inode_ref(structif (ret) goto out; -btrfs_update_inode (trans, root,BTRFS_I (inode)); + ret =btrfs_update_inode (trans, root,BTRFS_I (inode)); +if (ret) + goto out; } ref_ptr = (unsigned long )(ref_ptr + ref_struct_size) + namelen;
@@ -1749,7 +1751,9 @@ static noinline int fixup_inode_link_couif (nlink != inode->i_nlink) {set_nlink (inode, nlink); -btrfs_update_inode (trans, root,BTRFS_I (inode)); + ret =btrfs_update_inode (trans, root,BTRFS_I (inode)); +if (ret) + goto out; }BTRFS_I (inode)->index_cnt = (u64)-1 ;
Greg Kroah-Hartman
maintainerPATCH 145/161 btrfs: fixup error handling in fixup_inode_link_counts
In reply to Greg Kroah-HartmanFrom: Josef Bacik <[email protected]>
commit 011b28acf940eb61c000059dd9e2cfcbf52ed96b upstream.
This function has the following pattern
while (1) {
ret = whatever();
if (ret)
goto out;
}
ret = 0
out:
return ret;However several places in this while loop we simply break; when there's a problem, thus clearing the return value, and in one case we do a return -EIO, and leak the memory for the path.
Fix this by re-arranging the loop to deal with ret == 1 coming from btrfs_search_slot, and then simply delete the
ret = 0;
out:
bit so everybody can break if there is an error, which will allow for proper error handling to occur.
--- fs/btrfs/tree-log.c | 13 +++++++------ 1 file changed, 7 insertions(+), 6 deletions(-)
fs/btrfs/tree-log.c +7 -6
@@ -1791,6 +1791,7 @@ static noinline int fixup_inode_link_coubreak ;if (ret ==1 ) { + ret =0 ;if (path->slots[0 ] ==0 )break ; path->slots[0 ]--;
@@ -1803,17 +1804,19 @@ static noinline int fixup_inode_link_cou ret =btrfs_del_item (trans, root, path);if (ret) - goto out; +break ;btrfs_release_path (path); inode =read_one_inode (root, key.offset); -if (!inode) -return -EIO ; +if (!inode) { + ret = -EIO ; +break ; + } ret =fixup_inode_link_count (trans, root, inode);iput (inode);if (ret) - goto out; +break ;/* * fixup on a directory may create new entries,
@@ -1822,8 +1825,6 @@ static noinline int fixup_inode_link_cou*/ key.offset = (u64)-1 ; } - ret =0 ; -out:btrfs_release_path (path);return ret; }
Greg Kroah-Hartman
maintainerPATCH 146/161 btrfs: abort in rename_exchange if we fail to insert the second ref
In reply to Greg Kroah-HartmanFrom: Josef Bacik <[email protected]>
commit dc09ef3562726cd520c8338c1640872a60187af5 upstream.
Error injection stress uncovered a problem where we'd leave a dangling inode ref if we failed during a rename_exchange. This happens because we insert the inode ref for one side of the rename, and then for the other side. If this second inode ref insert fails we'll leave the first one dangling and leave a corrupt file system behind. Fix this by aborting if we did the insert for the first inode ref.
--- fs/btrfs/inode.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
fs/btrfs/inode.c +6 -1
@@ -9088,6 +9088,7 @@ static int btrfs_rename_exchange(structint ret2;bool root_log_pinned = false;bool dest_log_pinned = false; +bool need_abort = false;/* we only allow rename subvolume link between subvolumes */ if (old_ino !=BTRFS_FIRST_FREE_OBJECTID && root != dest)
@@ -9144,6 +9145,7 @@ static int btrfs_rename_exchange(struct old_idx);if (ret) goto out_fail; + need_abort = true; }/* And now for the dest. */
@@ -9159,8 +9161,11 @@ static int btrfs_rename_exchange(struct new_ino,btrfs_ino (BTRFS_I (old_dir)), new_idx); -if (ret) +if (ret) { +if (need_abort) +btrfs_abort_transaction (trans, ret); goto out_fail; + } }/* Update inode version and ctime/mtime. */